diff --git a/CREDITS b/CREDITS
index b50db17..abe05a0 100644
--- a/CREDITS
+++ b/CREDITS
@@ -369,10 +369,10 @@
 D: sun4 port, Sparc hacker
 
 N: Hugh Blemings
-E: hugh@misc.nu
-W: http://misc.nu/hugh/
-D: Author and maintainer of the Keyspan USB to Serial drivers
-S: Po Box 234
+E: hugh@blemings.org
+W: http://blemings.org/hugh
+D: Original author of the Keyspan USB to serial drivers, random PowerPC hacker
+S: PO Box 234
 S: Belconnen ACT 2616
 S: Australia
 
diff --git a/Documentation/ABI/testing/sysfs-class-uwb_rc b/Documentation/ABI/testing/sysfs-class-uwb_rc
index a0d18db..6a5fd07 100644
--- a/Documentation/ABI/testing/sysfs-class-uwb_rc
+++ b/Documentation/ABI/testing/sysfs-class-uwb_rc
@@ -32,14 +32,16 @@
 Description:
                 Write:
 
-                <channel> [<bpst offset>]
+                <channel>
 
-                to start beaconing on a specific channel, or stop
-                beaconing if <channel> is -1.  Valid channels depends
-                on the radio controller's supported band groups.
+                to force a specific channel to be used when beaconing,
+                or, if <channel> is -1, to prohibit beaconing.  If
+                <channel> is 0, then the default channel selection
+                algorithm will be used.  Valid channels depends on the
+                radio controller's supported band groups.
 
-                <bpst offset> may be used to try and join a specific
-                beacon group if more than one was found during a scan.
+                Reading returns the currently active channel, or -1 if
+                the radio controller is not beaconing.
 
 What:           /sys/class/uwb_rc/uwbN/scan
 Date:           July 2008
diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt
index 94bbc27..9d620c1 100644
--- a/Documentation/cpu-hotplug.txt
+++ b/Documentation/cpu-hotplug.txt
@@ -50,16 +50,17 @@
   			cpu_possible_map = cpu_present_map + additional_cpus
 
 (*) Option valid only for following architectures
-- x86_64, ia64
+- ia64
 
-ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT
-to determine the number of potentially hot-pluggable cpus. The implementation
-should only rely on this to count the # of cpus, but *MUST* not rely on the
-apicid values in those tables for disabled apics. In the event BIOS doesn't
-mark such hot-pluggable cpus as disabled entries, one could use this
-parameter "additional_cpus=x" to represent those cpus in the cpu_possible_map.
+ia64 uses the number of disabled local apics in ACPI tables MADT to
+determine the number of potentially hot-pluggable cpus. The implementation
+should only rely on this to count the # of cpus, but *MUST* not rely
+on the apicid values in those tables for disabled apics. In the event
+BIOS doesn't mark such hot-pluggable cpus as disabled entries, one could
+use this parameter "additional_cpus=x" to represent those cpus in the
+cpu_possible_map.
 
-possible_cpus=n		[s390 only] use this to set hotpluggable cpus.
+possible_cpus=n		[s390,x86_64] use this to set hotpluggable cpus.
 			This option sets possible_cpus bits in
 			cpu_possible_map. Thus keeping the numbers of bits set
 			constant even if the machine gets rebooted.
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index dc7c681..df18d87 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -310,15 +310,6 @@
 
 ---------------------------
 
-What: ide-scsi (BLK_DEV_IDESCSI)
-When: 2.6.29
-Why:  The 2.6 kernel supports direct writing to ide CD drives, which
-      eliminates the need for ide-scsi. The new method is more
-      efficient in every way.
-Who:  FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
-
----------------------------
-
 What:	i2c_attach_client(), i2c_detach_client(), i2c_driver->detach_client()
 When:	2.6.29 (ideally) or 2.6.30 (more likely)
 Why:	Deprecated by the new (standard) device driver binding model. Use
diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 23d2f44..ccec553 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -394,7 +394,6 @@
 	unsigned long (*get_unmapped_area)(struct file *, unsigned long,
 			unsigned long, unsigned long, unsigned long);
 	int (*check_flags)(int);
-	int (*dir_notify)(struct file *, unsigned long);
 };
 
 locking rules:
@@ -424,7 +423,6 @@
 sendpage:		no
 get_unmapped_area:	no
 check_flags:		no
-dir_notify:		no
 
 ->llseek() locking has moved from llseek to the individual llseek
 implementations.  If your fs is not using generic_file_llseek, you
diff --git a/Documentation/filesystems/devpts.txt b/Documentation/filesystems/devpts.txt
new file mode 100644
index 0000000..68dffd8
--- /dev/null
+++ b/Documentation/filesystems/devpts.txt
@@ -0,0 +1,132 @@
+
+To support containers, we now allow multiple instances of devpts filesystem,
+such that indices of ptys allocated in one instance are independent of indices
+allocated in other instances of devpts.
+
+To preserve backward compatibility, this support for multiple instances is
+enabled only if:
+
+	- CONFIG_DEVPTS_MULTIPLE_INSTANCES=y, and
+	- '-o newinstance' mount option is specified while mounting devpts
+
+IOW, devpts now supports both single-instance and multi-instance semantics.
+
+If CONFIG_DEVPTS_MULTIPLE_INSTANCES=n, there is no change in behavior and
+this referred to as the "legacy" mode. In this mode, the new mount options
+(-o newinstance and -o ptmxmode) will be ignored with a 'bogus option' message
+on console.
+
+If CONFIG_DEVPTS_MULTIPLE_INSTANCES=y and devpts is mounted without the
+'newinstance' option (as in current start-up scripts) the new mount binds
+to the initial kernel mount of devpts. This mode is referred to as the
+'single-instance' mode and the current, single-instance semantics are
+preserved, i.e PTYs are common across the system.
+
+The only difference between this single-instance mode and the legacy mode
+is the presence of new, '/dev/pts/ptmx' node with permissions 0000, which
+can safely be ignored.
+
+If CONFIG_DEVPTS_MULTIPLE_INSTANCES=y and 'newinstance' option is specified,
+the mount is considered to be in the multi-instance mode and a new instance
+of the devpts fs is created. Any ptys created in this instance are independent
+of ptys in other instances of devpts. Like in the single-instance mode, the
+/dev/pts/ptmx node is present. To effectively use the multi-instance mode,
+open of /dev/ptmx must be a redirected to '/dev/pts/ptmx' using a symlink or
+bind-mount.
+
+Eg: A container startup script could do the following:
+
+	$ chmod 0666 /dev/pts/ptmx
+	$ rm /dev/ptmx
+	$ ln -s pts/ptmx /dev/ptmx
+	$ ns_exec -cm /bin/bash
+
+	# We are now in new container
+
+	$ umount /dev/pts
+	$ mount -t devpts -o newinstance lxcpts /dev/pts
+	$ sshd -p 1234
+
+where 'ns_exec -cm /bin/bash' calls clone() with CLONE_NEWNS flag and execs
+/bin/bash in the child process.  A pty created by the sshd is not visible in
+the original mount of /dev/pts.
+
+User-space changes
+------------------
+
+In multi-instance mode (i.e '-o newinstance' mount option is specified at least
+once), following user-space issues should be noted.
+
+1. If -o newinstance mount option is never used, /dev/pts/ptmx can be ignored
+   and no change is needed to system-startup scripts.
+
+2. To effectively use multi-instance mode (i.e -o newinstance is specified)
+   administrators or startup scripts should "redirect" open of /dev/ptmx to
+   /dev/pts/ptmx using either a bind mount or symlink.
+
+	$ mount -t devpts -o newinstance devpts /dev/pts
+
+   followed by either
+
+	$ rm /dev/ptmx
+	$ ln -s pts/ptmx /dev/ptmx
+	$ chmod 666 /dev/pts/ptmx
+   or
+	$ mount -o bind /dev/pts/ptmx /dev/ptmx
+
+3. The '/dev/ptmx -> pts/ptmx' symlink is the preferred method since it
+   enables better error-reporting and treats both single-instance and
+   multi-instance mounts similarly.
+
+   But this method requires that system-startup scripts set the mode of
+   /dev/pts/ptmx correctly (default mode is 0000). The scripts can set the
+   mode by, either
+
+   	- adding ptmxmode mount option to devpts entry in /etc/fstab, or
+	- using 'chmod 0666 /dev/pts/ptmx'
+
+4. If multi-instance mode mount is needed for containers, but the system
+   startup scripts have not yet been updated, container-startup scripts
+   should bind mount /dev/ptmx to /dev/pts/ptmx to avoid breaking single-
+   instance mounts.
+
+   Or, in general, container-startup scripts should use:
+
+	mount -t devpts -o newinstance -o ptmxmode=0666 devpts /dev/pts
+	if [ ! -L /dev/ptmx ]; then
+		mount -o bind /dev/pts/ptmx /dev/ptmx
+	fi
+
+   When all devpts mounts are multi-instance, /dev/ptmx can permanently be
+   a symlink to pts/ptmx and the bind mount can be ignored.
+
+5. A multi-instance mount that is not accompanied by the /dev/ptmx to
+   /dev/pts/ptmx redirection would result in an unusable/unreachable pty.
+
+	mount -t devpts -o newinstance lxcpts /dev/pts
+
+   immediately followed by:
+
+	open("/dev/ptmx")
+
+    would create a pty, say /dev/pts/7, in the initial kernel mount.
+    But /dev/pts/7 would be invisible in the new mount.
+
+6. The permissions for /dev/pts/ptmx node should be specified when mounting
+   /dev/pts, using the '-o ptmxmode=%o' mount option (default is 0000).
+
+	mount -t devpts -o newinstance -o ptmxmode=0644 devpts /dev/pts
+
+   The permissions can be later be changed as usual with 'chmod'.
+
+	chmod 666 /dev/pts/ptmx
+
+7. A mount of devpts without the 'newinstance' option results in binding to
+   initial kernel mount.  This behavior while preserving legacy semantics,
+   does not provide strict isolation in a container environment. i.e by
+   mounting devpts without the 'newinstance' option, a container could
+   get visibility into the 'host' or root container's devpts.
+   
+   To workaround this and have strict isolation, all mounts of devpts,
+   including the mount in the root container, should use the newinstance
+   option.
diff --git a/Documentation/filesystems/files.txt b/Documentation/filesystems/files.txt
index bb0142f..ac2facc 100644
--- a/Documentation/filesystems/files.txt
+++ b/Documentation/filesystems/files.txt
@@ -76,13 +76,13 @@
 5. Handling of the file structures is special. Since the look-up
    of the fd (fget()/fget_light()) are lock-free, it is possible
    that look-up may race with the last put() operation on the
-   file structure. This is avoided using atomic_inc_not_zero()
+   file structure. This is avoided using atomic_long_inc_not_zero()
    on ->f_count :
 
 	rcu_read_lock();
 	file = fcheck_files(files, fd);
 	if (file) {
-		if (atomic_inc_not_zero(&file->f_count))
+		if (atomic_long_inc_not_zero(&file->f_count))
 			*fput_needed = 1;
 		else
 		/* Didn't get the reference, someone's freed */
@@ -92,7 +92,7 @@
 	....
 	return file;
 
-   atomic_inc_not_zero() detects if refcounts is already zero or
+   atomic_long_inc_not_zero() detects if refcounts is already zero or
    goes to zero during increment. If it does, we fail
    fget()/fget_light().
 
diff --git a/Documentation/filesystems/ubifs.txt b/Documentation/filesystems/ubifs.txt
index dd84ea3..84da2a4 100644
--- a/Documentation/filesystems/ubifs.txt
+++ b/Documentation/filesystems/ubifs.txt
@@ -95,6 +95,9 @@
 			of this option is that corruption of the contents
 			of a file can go unnoticed.
 chk_data_crc (*)	do not skip checking CRCs on data nodes
+compr=none              override default compressor and set it to "none"
+compr=lzo               override default compressor and set it to "lzo"
+compr=zlib              override default compressor and set it to "zlib"
 
 
 Quick usage instructions
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 5579bda..ef19afa 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -733,7 +733,6 @@
 	ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
 	unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
 	int (*check_flags)(int);
-	int (*dir_notify)(struct file *filp, unsigned long arg);
 	int (*flock) (struct file *, int, struct file_lock *);
 	ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, size_t, unsigned int);
 	ssize_t (*splice_read)(struct file *, struct pipe_inode_info *, size_t, unsigned int);
@@ -800,8 +799,6 @@
 
   check_flags: called by the fcntl(2) system call for F_SETFL command
 
-  dir_notify: called by the fcntl(2) system call for F_NOTIFY command
-
   flock: called by the flock(2) system call
 
   splice_write: called by the VFS to splice data from a pipe to a file. This
@@ -931,7 +928,7 @@
   d_lookup: look up a dentry given its parent and path name component
 	It looks up the child of that given name from the dcache
 	hash table. If it is found, the reference count is incremented
-	and the dentry is returned. The caller must use d_put()
+	and the dentry is returned. The caller must use dput()
 	to free the dentry when it finishes using it.
 
 For further information on dentry locking, please refer to the document
diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt
index b880ce5..8246991 100644
--- a/Documentation/ioctl/ioctl-number.txt
+++ b/Documentation/ioctl/ioctl-number.txt
@@ -97,6 +97,7 @@
 					<http://linux01.gwdg.de/~alatham/ppdd.html>
 'M'	all	linux/soundcard.h
 'N'	00-1F	drivers/usb/scanner.h
+'O'     00-02   include/mtd/ubi-user.h UBI
 'P'	all	linux/soundcard.h
 'Q'	all	linux/soundcard.h
 'R'	00-1F	linux/random.h
@@ -142,6 +143,9 @@
 'n'	00-7F	linux/ncp_fs.h
 'n'	E0-FF	video/matrox.h          matroxfb
 'o'	00-1F	fs/ocfs2/ocfs2_fs.h	OCFS2
+'o'     00-03   include/mtd/ubi-user.h  conflict! (OCFS2 and UBI overlaps)
+'o'     40-41   include/mtd/ubi-user.h  UBI
+'o'     01-A1   include/linux/dvb/*.h DVB
 'p'	00-0F	linux/phantom.h		conflict! (OpenHaptics needs this)
 'p'	00-3F	linux/mc146818rtc.h	conflict!
 'p'	40-7F	linux/nvram.h
diff --git a/Documentation/kbuild/00-INDEX b/Documentation/kbuild/00-INDEX
index 1146442..e8d2b6d 100644
--- a/Documentation/kbuild/00-INDEX
+++ b/Documentation/kbuild/00-INDEX
@@ -1,5 +1,9 @@
 00-INDEX
-    	- this file: info on the kernel build process
+	- this file: info on the kernel build process
+kbuild.txt
+	- developer information on kbuild
+kconfig.txt
+	- usage help for make *config
 kconfig-language.txt
 	- specification of Config Language, the language in Kconfig files
 makefiles.txt
diff --git a/Documentation/kbuild/kbuild.txt b/Documentation/kbuild/kbuild.txt
new file mode 100644
index 0000000..5177184
--- /dev/null
+++ b/Documentation/kbuild/kbuild.txt
@@ -0,0 +1,126 @@
+Environment variables
+
+KCPPFLAGS
+--------------------------------------------------
+Additional options to pass when preprocessing. The preprocessing options
+will be used in all cases where kbuild do preprocessing including
+building C files and assembler files.
+
+KAFLAGS
+--------------------------------------------------
+Additional options to the assembler.
+
+KCFLAGS
+--------------------------------------------------
+Additional options to the C compiler.
+
+KBUILD_VERBOSE
+--------------------------------------------------
+Set the kbuild verbosity. Can be assinged same values as "V=...".
+See make help for the full list.
+Setting "V=..." takes precedence over KBUILD_VERBOSE.
+
+KBUILD_EXTMOD
+--------------------------------------------------
+Set the directory to look for the kernel source when building external
+modules.
+The directory can be specified in several ways:
+1) Use "M=..." on the command line
+2) Environmnet variable KBUILD_EXTMOD
+3) Environmnet variable SUBDIRS
+The possibilities are listed in the order they take precedence.
+Using "M=..." will always override the others.
+
+KBUILD_OUTPUT
+--------------------------------------------------
+Specify the output directory when building the kernel.
+The output directory can also be specificed using "O=...".
+Setting "O=..." takes precedence over KBUILD_OUTPUT
+
+ARCH
+--------------------------------------------------
+Set ARCH to the architecture to be built.
+In most cases the name of the architecture is the same as the
+directory name found in the arch/ directory.
+But some architectures suach as x86 and sparc has aliases.
+x86: i386 for 32 bit, x86_64 for 64 bit
+sparc: sparc for 32 bit, sparc64 for 64 bit
+
+CROSS_COMPILE
+--------------------------------------------------
+Specify an optional fixed part of the binutils filename.
+CROSS_COMPILE can be a part of the filename or the full path.
+
+CROSS_COMPILE is also used for ccache is some setups.
+
+CF
+--------------------------------------------------
+Additional options for sparse.
+CF is often used on the command-line like this:
+
+    make CF=-Wbitwise C=2
+
+INSTALL_PATH
+--------------------------------------------------
+INSTALL_PATH specifies where to place the updated kernel and system map
+images. Default is /boot, but you can set it to other values
+
+
+MODLIB
+--------------------------------------------------
+Specify where to install modules.
+The default value is:
+
+     $(INSTALL_MOD_PATH)/lib/modules/$(KERNELRELEASE)
+
+The value can be overridden in which case the default value is ignored.
+
+INSTALL_MOD_PATH
+--------------------------------------------------
+INSTALL_MOD_PATH specifies a prefix to MODLIB for module directory
+relocations required by build roots.  This is not defined in the
+makefile but the argument can be passed to make if needed.
+
+INSTALL_MOD_STRIP
+--------------------------------------------------
+INSTALL_MOD_STRIP, if defined, will cause modules to be
+stripped after they are installed.  If INSTALL_MOD_STRIP is '1', then
+the default option --strip-debug will be used.  Otherwise,
+INSTALL_MOD_STRIP will used as the options to the strip command.
+
+INSTALL_FW_PATH
+--------------------------------------------------
+INSTALL_FW_PATH specify where to install the firmware blobs.
+The default value is:
+
+    $(INSTALL_MOD_PATH)/lib/firmware
+
+The value can be overridden in which case the default value is ignored.
+
+INSTALL_HDR_PATH
+--------------------------------------------------
+INSTALL_HDR_PATH specify where to install user space headers when
+executing "make headers_*".
+The default value is:
+
+    $(objtree)/usr
+
+$(objtree) is the directory where output files are saved.
+The output directory is often set using "O=..." on the commandline.
+
+The value can be overridden in which case the default value is ignored.
+
+KBUILD_MODPOST_WARN
+--------------------------------------------------
+KBUILD_MODPOST_WARN can be set to avoid error out in case of undefined
+symbols in the final module linking stage.
+
+KBUILD_MODPOST_FINAL
+--------------------------------------------------
+KBUILD_MODPOST_NOFINAL can be set to skip the final link of modules.
+This is solely usefull to speed up test compiles.
+
+KBUILD_EXTRA_SYMBOLS
+--------------------------------------------------
+For modules use symbols from another modules.
+See more details in modules.txt.
diff --git a/Documentation/kbuild/kconfig.txt b/Documentation/kbuild/kconfig.txt
new file mode 100644
index 0000000..26a7c0a
--- /dev/null
+++ b/Documentation/kbuild/kconfig.txt
@@ -0,0 +1,188 @@
+This file contains some assistance for using "make *config".
+
+Use "make help" to list all of the possible configuration targets.
+
+The xconfig ('qconf') and menuconfig ('mconf') programs also
+have embedded help text.  Be sure to check it for navigation,
+search, and other general help text.
+
+======================================================================
+General
+--------------------------------------------------
+
+New kernel releases often introduce new config symbols.  Often more
+important, new kernel releases may rename config symbols.  When
+this happens, using a previously working .config file and running
+"make oldconfig" won't necessarily produce a working new kernel
+for you, so you may find that you need to see what NEW kernel
+symbols have been introduced.
+
+To see a list of new config symbols when using "make oldconfig", use
+
+	cp user/some/old.config .config
+	yes "" | make oldconfig >conf.new
+
+and the config program will list as (NEW) any new symbols that have
+unknown values.  Of course, the .config file is also updated with
+new (default) values, so you can use:
+
+	grep "(NEW)" conf.new
+
+to see the new config symbols or you can 'diff' the previous and
+new .config files to see the differences:
+
+	diff .config.old .config | less
+
+(Yes, we need something better here.)
+
+
+======================================================================
+menuconfig
+--------------------------------------------------
+
+SEARCHING for CONFIG symbols
+
+Searching in menuconfig:
+
+	The Search function searches for kernel configuration symbol
+	names, so you have to know something close to what you are
+	looking for.
+
+	Example:
+		/hotplug
+		This lists all config symbols that contain "hotplug",
+		e.g., HOTPLUG, HOTPLUG_CPU, MEMORY_HOTPLUG.
+
+	For search help, enter / followed TAB-TAB-TAB (to highlight
+	<Help>) and Enter.  This will tell you that you can also use
+	regular expressions (regexes) in the search string, so if you
+	are not interested in MEMORY_HOTPLUG, you could try
+
+		/^hotplug
+
+
+______________________________________________________________________
+Color Themes for 'menuconfig'
+
+It is possible to select different color themes using the variable
+MENUCONFIG_COLOR.  To select a theme use:
+
+	make MENUCONFIG_COLOR=<theme> menuconfig
+
+Available themes are:
+  mono       => selects colors suitable for monochrome displays
+  blackbg    => selects a color scheme with black background
+  classic    => theme with blue background. The classic look
+  bluetitle  => a LCD friendly version of classic. (default)
+
+______________________________________________________________________
+Environment variables in 'menuconfig'
+
+KCONFIG_ALLCONFIG
+--------------------------------------------------
+(partially based on lkml email from/by Rob Landley, re: miniconfig)
+--------------------------------------------------
+The allyesconfig/allmodconfig/allnoconfig/randconfig variants can
+also use the environment variable KCONFIG_ALLCONFIG as a flag or a
+filename that contains config symbols that the user requires to be
+set to a specific value.  If KCONFIG_ALLCONFIG is used without a
+filename, "make *config" checks for a file named
+"all{yes/mod/no/random}.config" (corresponding to the *config command
+that was used) for symbol values that are to be forced.  If this file
+is not found, it checks for a file named "all.config" to contain forced
+values.
+
+This enables you to create "miniature" config (miniconfig) or custom
+config files containing just the config symbols that you are interested
+in.  Then the kernel config system generates the full .config file,
+including dependencies of your miniconfig file, based on the miniconfig
+file.
+
+This 'KCONFIG_ALLCONFIG' file is a config file which contains
+(usually a subset of all) preset config symbols.  These variable
+settings are still subject to normal dependency checks.
+
+Examples:
+	KCONFIG_ALLCONFIG=custom-notebook.config make allnoconfig
+or
+	KCONFIG_ALLCONFIG=mini.config make allnoconfig
+or
+	make KCONFIG_ALLCONFIG=mini.config allnoconfig
+
+These examples will disable most options (allnoconfig) but enable or
+disable the options that are explicitly listed in the specified
+mini-config files.
+
+KCONFIG_NOSILENTUPDATE
+--------------------------------------------------
+If this variable has a non-blank value, it prevents silent kernel
+config udpates (requires explicit updates).
+
+KCONFIG_CONFIG
+--------------------------------------------------
+This environment variable can be used to specify a default kernel config
+file name to override the default name of ".config".
+
+KCONFIG_OVERWRITECONFIG
+--------------------------------------------------
+If you set KCONFIG_OVERWRITECONFIG in the environment, Kconfig will not
+break symlinks when .config is a symlink to somewhere else.
+
+KCONFIG_NOTIMESTAMP
+--------------------------------------------------
+If this environment variable exists and is non-null, the timestamp line
+in generated .config files is omitted.
+
+KCONFIG_AUTOCONFIG
+--------------------------------------------------
+This environment variable can be set to specify the path & name of the
+"auto.conf" file.  Its default value is "include/config/auto.conf".
+
+KCONFIG_AUTOHEADER
+--------------------------------------------------
+This environment variable can be set to specify the path & name of the
+"autoconf.h" (header) file.  Its default value is "include/linux/autoconf.h".
+
+______________________________________________________________________
+menuconfig User Interface Options
+----------------------------------------------------------------------
+MENUCONFIG_MODE
+--------------------------------------------------
+This mode shows all sub-menus in one large tree.
+
+Example:
+	MENUCONFIG_MODE=single_menu make menuconfig
+
+======================================================================
+xconfig
+--------------------------------------------------
+
+Searching in xconfig:
+
+	The Search function searches for kernel configuration symbol
+	names, so you have to know something close to what you are
+	looking for.
+
+	Example:
+		Ctrl-F hotplug
+	or
+		Menu: File, Search, hotplug
+
+	lists all config symbol entries that contain "hotplug" in
+	the symbol name.  In this Search dialog, you may change the
+	config setting for any of the entries that are not grayed out.
+	You can also enter a different search string without having
+	to return to the main menu.
+
+
+======================================================================
+gconfig
+--------------------------------------------------
+
+Searching in gconfig:
+
+	None (gconfig isn't maintained as well as xconfig or menuconfig);
+	however, gconfig does have a few more viewing choices than
+	xconfig does.
+
+###
diff --git a/Documentation/usb/wusb-cbaf b/Documentation/usb/wusb-cbaf
index 2e78b70..426ddaa 100644
--- a/Documentation/usb/wusb-cbaf
+++ b/Documentation/usb/wusb-cbaf
@@ -80,12 +80,6 @@
     start)
         for dev in ${2:-$hdevs}
           do
-          uwb_rc=$(readlink -f $dev/uwb_rc)
-          if cat $uwb_rc/beacon | grep -q -- "-1"
-              then
-              echo 13 0 > $uwb_rc/beacon
-              echo I: started beaconing on ch 13 on $(basename $uwb_rc) >&2
-          fi
           echo $host_CHID > $dev/wusb_chid
           echo I: started host $(basename $dev) >&2
         done
@@ -95,9 +89,6 @@
           do
           echo 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > $dev/wusb_chid
           echo I: stopped host $(basename $dev) >&2
-          uwb_rc=$(readlink -f $dev/uwb_rc)
-          echo -1 | cat > $uwb_rc/beacon
-          echo I: stopped beaconing on $(basename $uwb_rc) >&2
         done
         ;;
     set-chid)
diff --git a/MAINTAINERS b/MAINTAINERS
index ceb32ee..befacf0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2049,6 +2049,12 @@
 W:	http://artax.karlin.mff.cuni.cz/~mikulas/vyplody/hpfs/index-e.cgi
 S:	Maintained
 
+HSO	3G Modem Driver (hso.c)
+P:	Denis Joseph Barrow
+M:	d.barow@option.com
+W:	http://www.pharscape.org
+S:	Maintained
+
 HTCPEN TOUCHSCREEN DRIVER
 P:	Pau Oliva Fora
 M:	pof@eslack.org
@@ -2146,11 +2152,6 @@
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
-IDE-SCSI DRIVER
-L:	linux-ide@vger.kernel.org
-L:	linux-scsi@vger.kernel.org
-S:	Orphan
-
 IDLE-I7300
 P:	Andy Henroid
 M:	andrew.d.henroid@intel.com
@@ -2541,8 +2542,6 @@
 S:	Supported
 
 KERNEL VIRTUAL MACHINE For Itanium (KVM/IA64)
-P:	Anthony Xu
-M:	anthony.xu@intel.com
 P:	Xiantao Zhang
 M:	xiantao.zhang@intel.com
 L:	kvm-ia64@vger.kernel.org
@@ -2635,13 +2634,13 @@
 S:	Maintained
 
 LINUX FOR POWERPC (32-BIT AND 64-BIT)
-P:	Paul Mackerras
-M:	paulus@samba.org
 P:	Benjamin Herrenschmidt
 M:	benh@kernel.crashing.org
+P:	Paul Mackerras
+M:	paulus@samba.org
 W:	http://www.penguinppc.org/
 L:	linuxppc-dev@ozlabs.org
-T:	git kernel.org:/pub/scm/linux/kernel/git/paulus/powerpc.git
+T:	git kernel.org:/pub/scm/linux/kernel/git/benh/powerpc.git
 S:	Supported
 
 LINUX FOR POWER MACINTOSH
diff --git a/Makefile b/Makefile
index d13a969..f900666 100644
--- a/Makefile
+++ b/Makefile
@@ -321,7 +321,8 @@
 PERL		= perl
 CHECK		= sparse
 
-CHECKFLAGS     := -D__linux__ -Dlinux -D__STDC__ -Dunix -D__unix__ -Wbitwise $(CF)
+CHECKFLAGS     := -D__linux__ -Dlinux -D__STDC__ -Dunix -D__unix__ \
+		  -Wbitwise -Wno-return-void $(CF)
 MODFLAGS	= -DMODULE
 CFLAGS_MODULE   = $(MODFLAGS)
 AFLAGS_MODULE   = $(MODFLAGS)
diff --git a/README b/README
index 159912c..90a0765 100644
--- a/README
+++ b/README
@@ -52,11 +52,11 @@
 
  - The Documentation/DocBook/ subdirectory contains several guides for
    kernel developers and users.  These guides can be rendered in a
-   number of formats:  PostScript (.ps), PDF, and HTML, among others.
-   After installation, "make psdocs", "make pdfdocs", or "make htmldocs"
-   will render the documentation in the requested format.
+   number of formats:  PostScript (.ps), PDF, HTML, & man-pages, among others.
+   After installation, "make psdocs", "make pdfdocs", "make htmldocs",
+   or "make mandocs" will render the documentation in the requested format.
 
-INSTALLING the kernel:
+INSTALLING the kernel source:
 
  - If you install the full sources, put the kernel tarball in a
    directory where you have permissions (eg. your home directory) and
@@ -187,14 +187,9 @@
 	"make randconfig"  Create a ./.config file by setting symbol
 			   values to random values.
 
-   The allyesconfig/allmodconfig/allnoconfig/randconfig variants can
-   also use the environment variable KCONFIG_ALLCONFIG to specify a
-   filename that contains config options that the user requires to be
-   set to a specific value.  If KCONFIG_ALLCONFIG=filename is not used,
-   "make *config" checks for a file named "all{yes/mod/no/random}.config"
-   for symbol values that are to be forced.  If this file is not found,
-   it checks for a file named "all.config" to contain forced values.
-   
+   You can find more information on using the Linux kernel config tools
+   in Documentation/kbuild/make-configs.txt.
+
 	NOTES on "make config":
 	- having unnecessary drivers will make the kernel bigger, and can
 	  under some circumstances lead to problems: probing for a
@@ -231,6 +226,19 @@
  - If you configured any of the parts of the kernel as `modules', you
    will also have to do "make modules_install".
 
+ - Verbose kernel compile/build output:
+
+   Normally the kernel build system runs in a fairly quiet mode (but not
+   totally silent).  However, sometimes you or other kernel developers need
+   to see compile, link, or other commands exactly as they are executed.
+   For this, use "verbose" build mode.  This is done by inserting
+   "V=1" in the "make" command.  E.g.:
+
+	make V=1 all
+
+   To have the build system also tell the reason for the rebuild of each
+   target, use "V=2".  The default is "V=0".
+
  - Keep a backup kernel handy in case something goes wrong.  This is 
    especially true for the development releases, since each new release
    contains new code which has not been debugged.  Make sure you keep a
diff --git a/arch/alpha/include/asm/smp.h b/arch/alpha/include/asm/smp.h
index 544c69a..547e909 100644
--- a/arch/alpha/include/asm/smp.h
+++ b/arch/alpha/include/asm/smp.h
@@ -45,7 +45,6 @@
 #define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 extern int smp_num_cpus;
-#define cpu_possible_map	cpu_present_map
 
 extern void arch_send_call_function_single_ipi(int cpu);
 extern void arch_send_call_function_ipi(cpumask_t mask);
diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index ac706c1..b469775 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -8,7 +8,7 @@
 
 obj-y    := entry.o traps.o process.o init_task.o osf_sys.o irq.o \
 	    irq_alpha.o signal.o setup.o ptrace.o time.o \
-	    alpha_ksyms.o systbls.o err_common.o io.o
+	    alpha_ksyms.o systbls.o err_common.o io.o binfmt_loader.o
 
 obj-$(CONFIG_VGA_HOSE)	+= console.o
 obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/alpha/kernel/binfmt_loader.c b/arch/alpha/kernel/binfmt_loader.c
new file mode 100644
index 0000000..4a0af90
--- /dev/null
+++ b/arch/alpha/kernel/binfmt_loader.c
@@ -0,0 +1,51 @@
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mm_types.h>
+#include <linux/binfmts.h>
+#include <linux/a.out.h>
+
+static int load_binary(struct linux_binprm *bprm, struct pt_regs *regs)
+{
+	struct exec *eh = (struct exec *)bprm->buf;
+	unsigned long loader;
+	struct file *file;
+	int retval;
+
+	if (eh->fh.f_magic != 0x183 || (eh->fh.f_flags & 0x3000) != 0x3000)
+		return -ENOEXEC;
+
+	if (bprm->loader)
+		return -ENOEXEC;
+
+	allow_write_access(bprm->file);
+	fput(bprm->file);
+	bprm->file = NULL;
+
+	loader = bprm->vma->vm_end - sizeof(void *);
+
+	file = open_exec("/sbin/loader");
+	retval = PTR_ERR(file);
+	if (IS_ERR(file))
+		return retval;
+
+	/* Remember if the application is TASO.  */
+	bprm->taso = eh->ah.entry < 0x100000000UL;
+
+	bprm->file = file;
+	bprm->loader = loader;
+	retval = prepare_binprm(bprm);
+	if (retval < 0)
+		return retval;
+	return search_binary_handler(bprm,regs);
+}
+
+static struct linux_binfmt loader_format = {
+	.load_binary	= load_binary,
+};
+
+static int __init init_loader_binfmt(void)
+{
+	return register_binfmt(&loader_format);
+}
+arch_initcall(init_loader_binfmt);
diff --git a/arch/alpha/kernel/init_task.c b/arch/alpha/kernel/init_task.c
index 1f76218..c2938e5 100644
--- a/arch/alpha/kernel/init_task.c
+++ b/arch/alpha/kernel/init_task.c
@@ -8,7 +8,6 @@
 #include <asm/uaccess.h>
 
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c
index c626a82..d0f1620 100644
--- a/arch/alpha/kernel/irq.c
+++ b/arch/alpha/kernel/irq.c
@@ -55,7 +55,7 @@
 	last_cpu = cpu;
 
 	irq_desc[irq].affinity = cpumask_of_cpu(cpu);
-	irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu));
+	irq_desc[irq].chip->set_affinity(irq, cpumask_of(cpu));
 	return 0;
 }
 #endif /* CONFIG_SMP */
diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c
index 351407e..f238370 100644
--- a/arch/alpha/kernel/process.c
+++ b/arch/alpha/kernel/process.c
@@ -94,6 +94,7 @@
 		flags |= 0x00040000UL; /* "remain halted" */
 		*pflags = flags;
 		cpu_clear(cpuid, cpu_present_map);
+		cpu_clear(cpuid, cpu_possible_map);
 		halt();
 	}
 #endif
@@ -120,6 +121,7 @@
 #ifdef CONFIG_SMP
 	/* Wait for the secondaries to halt. */
 	cpu_clear(boot_cpuid, cpu_present_map);
+	cpu_clear(boot_cpuid, cpu_possible_map);
 	while (cpus_weight(cpu_present_map))
 		barrier();
 #endif
diff --git a/arch/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c
index cf7da10..d953e51 100644
--- a/arch/alpha/kernel/smp.c
+++ b/arch/alpha/kernel/smp.c
@@ -70,11 +70,6 @@
 /* Set to a secondary's cpuid when it comes online.  */
 static int smp_secondary_alive __devinitdata = 0;
 
-/* Which cpus ids came online.  */
-cpumask_t cpu_online_map;
-
-EXPORT_SYMBOL(cpu_online_map);
-
 int smp_num_probed;		/* Internal processor count */
 int smp_num_cpus = 1;		/* Number that came online.  */
 EXPORT_SYMBOL(smp_num_cpus);
@@ -440,6 +435,7 @@
 				((char *)cpubase + i*hwrpb->processor_size);
 			if ((cpu->flags & 0x1cc) == 0x1cc) {
 				smp_num_probed++;
+				cpu_set(i, cpu_possible_map);
 				cpu_set(i, cpu_present_map);
 				cpu->pal_revision = boot_cpu_palrev;
 			}
@@ -473,6 +469,7 @@
 
 	/* Nothing to do on a UP box, or when told not to.  */
 	if (smp_num_probed == 1 || max_cpus == 0) {
+		cpu_possible_map = cpumask_of_cpu(boot_cpuid);
 		cpu_present_map = cpumask_of_cpu(boot_cpuid);
 		printk(KERN_INFO "SMP mode deactivated.\n");
 		return;
diff --git a/arch/alpha/kernel/sys_dp264.c b/arch/alpha/kernel/sys_dp264.c
index c71b0fd..ab44c16 100644
--- a/arch/alpha/kernel/sys_dp264.c
+++ b/arch/alpha/kernel/sys_dp264.c
@@ -177,19 +177,19 @@
 }
 
 static void
-dp264_set_affinity(unsigned int irq, cpumask_t affinity)
+dp264_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&dp264_irq_lock);
-	cpu_set_irq_affinity(irq, affinity);
+	cpu_set_irq_affinity(irq, *affinity);
 	tsunami_update_irq_hw(cached_irq_mask);
 	spin_unlock(&dp264_irq_lock);
 }
 
 static void
-clipper_set_affinity(unsigned int irq, cpumask_t affinity)
+clipper_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&dp264_irq_lock);
-	cpu_set_irq_affinity(irq - 16, affinity);
+	cpu_set_irq_affinity(irq - 16, *affinity);
 	tsunami_update_irq_hw(cached_irq_mask);
 	spin_unlock(&dp264_irq_lock);
 }
diff --git a/arch/alpha/kernel/sys_titan.c b/arch/alpha/kernel/sys_titan.c
index 52c91cc..27f840a 100644
--- a/arch/alpha/kernel/sys_titan.c
+++ b/arch/alpha/kernel/sys_titan.c
@@ -158,10 +158,10 @@
 }
 
 static void
-titan_set_irq_affinity(unsigned int irq, cpumask_t affinity)
+titan_set_irq_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&titan_irq_lock);
-	titan_cpu_set_irq_affinity(irq - 16, affinity);
+	titan_cpu_set_irq_affinity(irq - 16, *affinity);
 	titan_update_irq_hw(titan_cached_irq_mask);
 	spin_unlock(&titan_irq_lock);
 }
diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 7fc9860..c6884ba 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -109,11 +109,11 @@
 }
 
 #ifdef CONFIG_SMP
-static void gic_set_cpu(unsigned int irq, cpumask_t mask_val)
+static void gic_set_cpu(unsigned int irq, const struct cpumask *mask_val)
 {
 	void __iomem *reg = gic_dist_base(irq) + GIC_DIST_TARGET + (gic_irq(irq) & ~3);
 	unsigned int shift = (irq % 4) * 8;
-	unsigned int cpu = first_cpu(mask_val);
+	unsigned int cpu = cpumask_first(mask_val);
 	u32 val;
 
 	spin_lock(&irq_controller_lock);
diff --git a/arch/arm/kernel/init_task.c b/arch/arm/kernel/init_task.c
index 0bbf806..e859af3 100644
--- a/arch/arm/kernel/init_task.c
+++ b/arch/arm/kernel/init_task.c
@@ -12,7 +12,6 @@
 
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 2f3eb79..7141cee 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -174,7 +174,7 @@
 	pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->cpu, cpu);
 
 	spin_lock_irq(&desc->lock);
-	desc->chip->set_affinity(irq, cpumask_of_cpu(cpu));
+	desc->chip->set_affinity(irq, cpumask_of(cpu));
 	spin_unlock_irq(&desc->lock);
 }
 
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 019237d..55fa7ff 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -34,16 +34,6 @@
 #include <asm/ptrace.h>
 
 /*
- * bitmask of present and online CPUs.
- * The present bitmask indicates that the CPU is physically present.
- * The online bitmask indicates that the CPU is up and running.
- */
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
-
-/*
  * as from 2.5, kernels no longer have an init_tasks structure
  * so we need some other way of telling a new secondary core
  * where to place its SVC stack
diff --git a/arch/arm/mach-at91/at91rm9200_time.c b/arch/arm/mach-at91/at91rm9200_time.c
index d140eae..1ff1bda 100644
--- a/arch/arm/mach-at91/at91rm9200_time.c
+++ b/arch/arm/mach-at91/at91rm9200_time.c
@@ -178,7 +178,6 @@
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
 	.rating		= 150,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= clkevt32k_next_event,
 	.set_mode	= clkevt32k_mode,
 };
@@ -206,7 +205,7 @@
 	clkevt.mult = div_sc(AT91_SLOW_CLOCK, NSEC_PER_SEC, clkevt.shift);
 	clkevt.max_delta_ns = clockevent_delta2ns(AT91_ST_ALMV, &clkevt);
 	clkevt.min_delta_ns = clockevent_delta2ns(2, &clkevt) + 1;
-	clkevt.cpumask = cpumask_of_cpu(0);
+	clkevt.cpumask = cpumask_of(0);
 	clockevents_register_device(&clkevt);
 
 	/* register clocksource */
diff --git a/arch/arm/mach-at91/at91sam926x_time.c b/arch/arm/mach-at91/at91sam926x_time.c
index 122fd77..b63e1d5 100644
--- a/arch/arm/mach-at91/at91sam926x_time.c
+++ b/arch/arm/mach-at91/at91sam926x_time.c
@@ -91,7 +91,6 @@
 	.features	= CLOCK_EVT_FEAT_PERIODIC,
 	.shift		= 32,
 	.rating		= 100,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_mode	= pit_clkevt_mode,
 };
 
@@ -173,6 +172,7 @@
 
 	/* Set up and register clockevents */
 	pit_clkevt.mult = div_sc(pit_rate, NSEC_PER_SEC, pit_clkevt.shift);
+	pit_clkevt.cpumask = cpumask_of(0);
 	clockevents_register_device(&pit_clkevt);
 }
 
diff --git a/arch/arm/mach-davinci/time.c b/arch/arm/mach-davinci/time.c
index 3b9a296..f8bcd29 100644
--- a/arch/arm/mach-davinci/time.c
+++ b/arch/arm/mach-davinci/time.c
@@ -322,7 +322,7 @@
 	clockevent_davinci.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_davinci);
 
-	clockevent_davinci.cpumask = cpumask_of_cpu(0);
+	clockevent_davinci.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_davinci);
 }
 
diff --git a/arch/arm/mach-imx/time.c b/arch/arm/mach-imx/time.c
index a11765f..aff0ebc 100644
--- a/arch/arm/mach-imx/time.c
+++ b/arch/arm/mach-imx/time.c
@@ -184,7 +184,7 @@
 	clockevent_imx.min_delta_ns =
 		clockevent_delta2ns(0xf, &clockevent_imx);
 
-	clockevent_imx.cpumask = cpumask_of_cpu(0);
+	clockevent_imx.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&clockevent_imx);
 
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 7766f46..f4656d2 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -487,7 +487,7 @@
 		clockevent_delta2ns(0xfffffffe, &clockevent_ixp4xx);
 	clockevent_ixp4xx.min_delta_ns =
 		clockevent_delta2ns(0xf, &clockevent_ixp4xx);
-	clockevent_ixp4xx.cpumask = cpumask_of_cpu(0);
+	clockevent_ixp4xx.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&clockevent_ixp4xx);
 	return 0;
diff --git a/arch/arm/mach-msm/timer.c b/arch/arm/mach-msm/timer.c
index 345a14c..444d9c0 100644
--- a/arch/arm/mach-msm/timer.c
+++ b/arch/arm/mach-msm/timer.c
@@ -182,7 +182,7 @@
 			clockevent_delta2ns(0xf0000000 >> clock->shift, ce);
 		/* 4 gets rounded down to 3 */
 		ce->min_delta_ns = clockevent_delta2ns(4, ce);
-		ce->cpumask = cpumask_of_cpu(0);
+		ce->cpumask = cpumask_of(0);
 
 		cs->mult = clocksource_hz2mult(clock->freq, cs->shift);
 		res = clocksource_register(cs);
diff --git a/arch/arm/mach-ns9xxx/time-ns9360.c b/arch/arm/mach-ns9xxx/time-ns9360.c
index a63424d..41df697 100644
--- a/arch/arm/mach-ns9xxx/time-ns9360.c
+++ b/arch/arm/mach-ns9xxx/time-ns9360.c
@@ -173,7 +173,7 @@
 	ns9360_clockevent_device.min_delta_ns =
 		clockevent_delta2ns(1, &ns9360_clockevent_device);
 
-	ns9360_clockevent_device.cpumask = cpumask_of_cpu(0);
+	ns9360_clockevent_device.cpumask = cpumask_of(0);
 	clockevents_register_device(&ns9360_clockevent_device);
 
 	setup_irq(IRQ_NS9360_TIMER0 + TIMER_CLOCKEVENT,
diff --git a/arch/arm/mach-omap1/time.c b/arch/arm/mach-omap1/time.c
index 2cf7e32..495a32c 100644
--- a/arch/arm/mach-omap1/time.c
+++ b/arch/arm/mach-omap1/time.c
@@ -173,7 +173,7 @@
 	clockevent_mpu_timer1.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_mpu_timer1);
 
-	clockevent_mpu_timer1.cpumask = cpumask_of_cpu(0);
+	clockevent_mpu_timer1.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_mpu_timer1);
 }
 
diff --git a/arch/arm/mach-omap1/timer32k.c b/arch/arm/mach-omap1/timer32k.c
index 705367e..fd3f739 100644
--- a/arch/arm/mach-omap1/timer32k.c
+++ b/arch/arm/mach-omap1/timer32k.c
@@ -187,7 +187,7 @@
 	clockevent_32k_timer.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_32k_timer);
 
-	clockevent_32k_timer.cpumask = cpumask_of_cpu(0);
+	clockevent_32k_timer.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_32k_timer);
 }
 
diff --git a/arch/arm/mach-omap2/timer-gp.c b/arch/arm/mach-omap2/timer-gp.c
index 589393b..ae60363 100644
--- a/arch/arm/mach-omap2/timer-gp.c
+++ b/arch/arm/mach-omap2/timer-gp.c
@@ -120,7 +120,7 @@
 	clockevent_gpt.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_gpt);
 
-	clockevent_gpt.cpumask = cpumask_of_cpu(0);
+	clockevent_gpt.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_gpt);
 }
 
diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c
index 0016241..95656a7 100644
--- a/arch/arm/mach-pxa/time.c
+++ b/arch/arm/mach-pxa/time.c
@@ -122,7 +122,6 @@
 	.features	= CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
 	.rating		= 200,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= pxa_osmr0_set_next_event,
 	.set_mode	= pxa_osmr0_set_mode,
 };
@@ -163,6 +162,7 @@
 		clockevent_delta2ns(0x7fffffff, &ckevt_pxa_osmr0);
 	ckevt_pxa_osmr0.min_delta_ns =
 		clockevent_delta2ns(MIN_OSCR_DELTA * 2, &ckevt_pxa_osmr0) + 1;
+	ckevt_pxa_osmr0.cpumask = cpumask_of(0);
 
 	cksrc_pxa_oscr0.mult =
 		clocksource_hz2mult(clock_tick_rate, cksrc_pxa_oscr0.shift);
diff --git a/arch/arm/mach-realview/core.c b/arch/arm/mach-realview/core.c
index 5f1d559..bd2aa4f 100644
--- a/arch/arm/mach-realview/core.c
+++ b/arch/arm/mach-realview/core.c
@@ -624,7 +624,7 @@
 	.set_mode	= timer_set_mode,
 	.set_next_event	= timer_set_next_event,
 	.rating		= 300,
-	.cpumask	= CPU_MASK_ALL,
+	.cpumask	= cpu_all_mask,
 };
 
 static void __init realview_clockevents_init(unsigned int timer_irq)
diff --git a/arch/arm/mach-realview/localtimer.c b/arch/arm/mach-realview/localtimer.c
index 9019ef2..67d6d9c 100644
--- a/arch/arm/mach-realview/localtimer.c
+++ b/arch/arm/mach-realview/localtimer.c
@@ -154,7 +154,7 @@
 	clk->set_mode		= local_timer_set_mode;
 	clk->set_next_event	= local_timer_set_next_event;
 	clk->irq		= IRQ_LOCALTIMER;
-	clk->cpumask		= cpumask_of_cpu(cpu);
+	clk->cpumask		= cpumask_of(cpu);
 	clk->shift		= 20;
 	clk->mult		= div_sc(mpcore_timer_rate, NSEC_PER_SEC, clk->shift);
 	clk->max_delta_ns	= clockevent_delta2ns(0xffffffff, clk);
@@ -193,7 +193,7 @@
 	clk->rating		= 200;
 	clk->set_mode		= dummy_timer_set_mode;
 	clk->broadcast		= smp_timer_broadcast;
-	clk->cpumask		= cpumask_of_cpu(cpu);
+	clk->cpumask		= cpumask_of(cpu);
 
 	clockevents_register_device(clk);
 }
diff --git a/arch/arm/mach-sa1100/time.c b/arch/arm/mach-sa1100/time.c
index 8c5e727..711c029 100644
--- a/arch/arm/mach-sa1100/time.c
+++ b/arch/arm/mach-sa1100/time.c
@@ -73,7 +73,6 @@
 	.features	= CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
 	.rating		= 200,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= sa1100_osmr0_set_next_event,
 	.set_mode	= sa1100_osmr0_set_mode,
 };
@@ -110,6 +109,7 @@
 		clockevent_delta2ns(0x7fffffff, &ckevt_sa1100_osmr0);
 	ckevt_sa1100_osmr0.min_delta_ns =
 		clockevent_delta2ns(MIN_OSCR_DELTA * 2, &ckevt_sa1100_osmr0) + 1;
+	ckevt_sa1100_osmr0.cpumask = cpumask_of(0);
 
 	cksrc_sa1100_oscr.mult =
 		clocksource_hz2mult(CLOCK_TICK_RATE, cksrc_sa1100_oscr.shift);
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index df25aa1..1c43494 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -1005,7 +1005,7 @@
 	timer0_clockevent.min_delta_ns =
 		clockevent_delta2ns(0xf, &timer0_clockevent);
 
-	timer0_clockevent.cpumask = cpumask_of_cpu(0);
+	timer0_clockevent.cpumask = cpumask_of(0);
 	clockevents_register_device(&timer0_clockevent);
 }
 
diff --git a/arch/arm/oprofile/op_model_mpcore.c b/arch/arm/oprofile/op_model_mpcore.c
index 4de366e..6d6bd58 100644
--- a/arch/arm/oprofile/op_model_mpcore.c
+++ b/arch/arm/oprofile/op_model_mpcore.c
@@ -260,10 +260,10 @@
 static void em_route_irq(int irq, unsigned int cpu)
 {
 	struct irq_desc *desc = irq_desc + irq;
-	cpumask_t mask = cpumask_of_cpu(cpu);
+	const struct cpumask *mask = cpumask_of(cpu);
 
 	spin_lock_irq(&desc->lock);
-	desc->affinity = mask;
+	desc->affinity = *mask;
 	desc->chip->set_affinity(irq, mask);
 	spin_unlock_irq(&desc->lock);
 }
diff --git a/arch/arm/plat-mxc/time.c b/arch/arm/plat-mxc/time.c
index fd28f51..758a129 100644
--- a/arch/arm/plat-mxc/time.c
+++ b/arch/arm/plat-mxc/time.c
@@ -190,7 +190,7 @@
 	clockevent_mxc.min_delta_ns =
 			clockevent_delta2ns(0xff, &clockevent_mxc);
 
-	clockevent_mxc.cpumask = cpumask_of_cpu(0);
+	clockevent_mxc.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&clockevent_mxc);
 
diff --git a/arch/arm/plat-orion/time.c b/arch/arm/plat-orion/time.c
index 544d6b3..6fa2923 100644
--- a/arch/arm/plat-orion/time.c
+++ b/arch/arm/plat-orion/time.c
@@ -149,7 +149,6 @@
 	.features	= CLOCK_EVT_FEAT_ONESHOT | CLOCK_EVT_FEAT_PERIODIC,
 	.shift		= 32,
 	.rating		= 300,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= orion_clkevt_next_event,
 	.set_mode	= orion_clkevt_mode,
 };
@@ -199,5 +198,6 @@
 	orion_clkevt.mult = div_sc(tclk, NSEC_PER_SEC, orion_clkevt.shift);
 	orion_clkevt.max_delta_ns = clockevent_delta2ns(0xfffffffe, &orion_clkevt);
 	orion_clkevt.min_delta_ns = clockevent_delta2ns(1, &orion_clkevt);
+	orion_clkevt.cpumask = cpumask_of(0);
 	clockevents_register_device(&orion_clkevt);
 }
diff --git a/arch/avr32/kernel/init_task.c b/arch/avr32/kernel/init_task.c
index 4405846..993d56ee 100644
--- a/arch/avr32/kernel/init_task.c
+++ b/arch/avr32/kernel/init_task.c
@@ -13,7 +13,6 @@
 
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
index 283481d..0ff46bf 100644
--- a/arch/avr32/kernel/time.c
+++ b/arch/avr32/kernel/time.c
@@ -106,7 +106,6 @@
 	.features	= CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 16,
 	.rating		= 50,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= comparator_next_event,
 	.set_mode	= comparator_mode,
 };
@@ -134,6 +133,7 @@
 	comparator.mult = div_sc(counter_hz, NSEC_PER_SEC, comparator.shift);
 	comparator.max_delta_ns = clockevent_delta2ns((u32)~0, &comparator);
 	comparator.min_delta_ns = clockevent_delta2ns(50, &comparator) + 1;
+	comparator.cpumask = cpumask_of(0);
 
 	sysreg_write(COMPARE, 0);
 	timer_irqaction.dev_id = &comparator;
diff --git a/arch/blackfin/kernel/init_task.c b/arch/blackfin/kernel/init_task.c
index 6bdba7b..2c228c0 100644
--- a/arch/blackfin/kernel/init_task.c
+++ b/arch/blackfin/kernel/init_task.c
@@ -33,7 +33,6 @@
 #include <linux/mqueue.h>
 #include <linux/fs.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 
diff --git a/arch/blackfin/kernel/time-ts.c b/arch/blackfin/kernel/time-ts.c
index e887efc..0ed2bad 100644
--- a/arch/blackfin/kernel/time-ts.c
+++ b/arch/blackfin/kernel/time-ts.c
@@ -162,7 +162,6 @@
 	.name		= "bfin_core_timer",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event = bfin_timer_set_next_event,
 	.set_mode	= bfin_timer_set_mode,
 };
@@ -193,6 +192,7 @@
 	clockevent_bfin.mult = div_sc(timer_clk, NSEC_PER_SEC, clockevent_bfin.shift);
 	clockevent_bfin.max_delta_ns = clockevent_delta2ns(-1, &clockevent_bfin);
 	clockevent_bfin.min_delta_ns = clockevent_delta2ns(100, &clockevent_bfin);
+	clockevent_bfin.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_bfin);
 
 	return 0;
diff --git a/arch/cris/arch-v32/kernel/irq.c b/arch/cris/arch-v32/kernel/irq.c
index 173c141..295131f 100644
--- a/arch/cris/arch-v32/kernel/irq.c
+++ b/arch/cris/arch-v32/kernel/irq.c
@@ -325,11 +325,11 @@
 {
 }
 
-void set_affinity_crisv32_irq(unsigned int irq, cpumask_t dest)
+void set_affinity_crisv32_irq(unsigned int irq, const struct cpumask *dest)
 {
 	unsigned long flags;
 	spin_lock_irqsave(&irq_lock, flags);
-	irq_allocations[irq - FIRST_IRQ].mask = dest;
+	irq_allocations[irq - FIRST_IRQ].mask = *dest;
 	spin_unlock_irqrestore(&irq_lock, flags);
 }
 
diff --git a/arch/cris/arch-v32/kernel/smp.c b/arch/cris/arch-v32/kernel/smp.c
index 52e16c6..9dac173 100644
--- a/arch/cris/arch-v32/kernel/smp.c
+++ b/arch/cris/arch-v32/kernel/smp.c
@@ -29,11 +29,7 @@
 spinlock_t cris_atomic_locks[] = { [0 ... LOCK_COUNT - 1] = SPIN_LOCK_UNLOCKED};
 
 /* CPU masks */
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
 cpumask_t phys_cpu_present_map = CPU_MASK_NONE;
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
 EXPORT_SYMBOL(phys_cpu_present_map);
 
 /* Variables used during SMP boot */
diff --git a/arch/cris/include/asm/smp.h b/arch/cris/include/asm/smp.h
index dba33ab..c615a06 100644
--- a/arch/cris/include/asm/smp.h
+++ b/arch/cris/include/asm/smp.h
@@ -4,7 +4,6 @@
 #include <linux/cpumask.h>
 
 extern cpumask_t phys_cpu_present_map;
-extern cpumask_t cpu_possible_map;
 
 #define raw_smp_processor_id() (current_thread_info()->cpu)
 
diff --git a/arch/cris/kernel/process.c b/arch/cris/kernel/process.c
index 5933656..60816e8 100644
--- a/arch/cris/kernel/process.c
+++ b/arch/cris/kernel/process.c
@@ -37,7 +37,6 @@
  * setup.
  */
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/frv/kernel/init_task.c b/arch/frv/kernel/init_task.c
index e219881..29429a8 100644
--- a/arch/frv/kernel/init_task.c
+++ b/arch/frv/kernel/init_task.c
@@ -10,7 +10,6 @@
 #include <asm/pgtable.h>
 
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/h8300/kernel/init_task.c b/arch/h8300/kernel/init_task.c
index 93a4899..cb5dc55 100644
--- a/arch/h8300/kernel/init_task.c
+++ b/arch/h8300/kernel/init_task.c
@@ -12,7 +12,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/ia64/hp/sim/hpsim_irq.c b/arch/ia64/hp/sim/hpsim_irq.c
index c2f58ff..cc0a318 100644
--- a/arch/ia64/hp/sim/hpsim_irq.c
+++ b/arch/ia64/hp/sim/hpsim_irq.c
@@ -22,7 +22,7 @@
 }
 
 static void
-hpsim_set_affinity_noop (unsigned int a, cpumask_t b)
+hpsim_set_affinity_noop(unsigned int a, const struct cpumask *b)
 {
 }
 
diff --git a/arch/ia64/include/asm/kvm.h b/arch/ia64/include/asm/kvm.h
index f38472a..68aa6da 100644
--- a/arch/ia64/include/asm/kvm.h
+++ b/arch/ia64/include/asm/kvm.h
@@ -166,8 +166,6 @@
 };
 
 struct kvm_regs {
-	char *saved_guest;
-	char *saved_stack;
 	struct saved_vpd vpd;
 	/*Arch-regs*/
 	int mp_state;
@@ -200,6 +198,10 @@
 	unsigned long fp_psr;       /*used for lazy float register */
 	unsigned long saved_gp;
 	/*for phycial  emulation */
+
+	union context saved_guest;
+
+	unsigned long reserved[64];	/* for future use */
 };
 
 struct kvm_sregs {
diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index c60d324..0560f3f 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -23,17 +23,6 @@
 #ifndef __ASM_KVM_HOST_H
 #define __ASM_KVM_HOST_H
 
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/kvm.h>
-#include <linux/kvm_para.h>
-#include <linux/kvm_types.h>
-
-#include <asm/pal.h>
-#include <asm/sal.h>
-
-#define KVM_MAX_VCPUS 4
 #define KVM_MEMORY_SLOTS 32
 /* memory slots that does not exposed to userspace */
 #define KVM_PRIVATE_MEM_SLOTS 4
@@ -50,70 +39,132 @@
 #define EXIT_REASON_EXTERNAL_INTERRUPT	6
 #define EXIT_REASON_IPI			7
 #define EXIT_REASON_PTC_G		8
+#define EXIT_REASON_DEBUG		20
 
 /*Define vmm address space and vm data space.*/
-#define KVM_VMM_SIZE (16UL<<20)
+#define KVM_VMM_SIZE (__IA64_UL_CONST(16)<<20)
 #define KVM_VMM_SHIFT 24
-#define KVM_VMM_BASE 0xD000000000000000UL
-#define VMM_SIZE (8UL<<20)
+#define KVM_VMM_BASE 0xD000000000000000
+#define VMM_SIZE (__IA64_UL_CONST(8)<<20)
 
 /*
  * Define vm_buffer, used by PAL Services, base address.
- * Note: vmbuffer is in the VMM-BLOCK, the size must be < 8M
+ * Note: vm_buffer is in the VMM-BLOCK, the size must be < 8M
  */
 #define KVM_VM_BUFFER_BASE (KVM_VMM_BASE + VMM_SIZE)
-#define KVM_VM_BUFFER_SIZE (8UL<<20)
+#define KVM_VM_BUFFER_SIZE (__IA64_UL_CONST(8)<<20)
 
-/*Define Virtual machine data layout.*/
-#define KVM_VM_DATA_SHIFT  24
-#define KVM_VM_DATA_SIZE (1UL << KVM_VM_DATA_SHIFT)
-#define KVM_VM_DATA_BASE (KVM_VMM_BASE + KVM_VMM_SIZE)
+/*
+ * kvm guest's data area looks as follow:
+ *
+ *            +----------------------+	-------	KVM_VM_DATA_SIZE
+ *	      |	    vcpu[n]'s data   |	 |     ___________________KVM_STK_OFFSET
+ *     	      |			     |	 |    /			  |
+ *     	      |	       ..........    |	 |   /vcpu's struct&stack |
+ *     	      |	       ..........    |	 |  /---------------------|---- 0
+ *	      |	    vcpu[5]'s data   |	 | /	   vpd		  |
+ *	      |	    vcpu[4]'s data   |	 |/-----------------------|
+ *	      |	    vcpu[3]'s data   |	 /	   vtlb		  |
+ *	      |	    vcpu[2]'s data   |	/|------------------------|
+ *	      |	    vcpu[1]'s data   |/  |	   vhpt		  |
+ *	      |	    vcpu[0]'s data   |____________________________|
+ *            +----------------------+	 |
+ *	      |	   memory dirty log  |	 |
+ *            +----------------------+	 |
+ *	      |	   vm's data struct  |	 |
+ *            +----------------------+	 |
+ *	      |			     |	 |
+ *	      |			     |	 |
+ *	      |			     |	 |
+ *	      |			     |	 |
+ *	      |			     |	 |
+ *	      |			     |	 |
+ *	      |			     |	 |
+ *	      |	  vm's p2m table  |	 |
+ *	      |			     |	 |
+ *            |			     |	 |
+ *	      |			     |	 |  |
+ * vm's data->|			     |   |  |
+ *	      +----------------------+ ------- 0
+ * To support large memory, needs to increase the size of p2m.
+ * To support more vcpus, needs to ensure it has enough space to
+ * hold vcpus' data.
+ */
 
+#define KVM_VM_DATA_SHIFT	26
+#define KVM_VM_DATA_SIZE	(__IA64_UL_CONST(1) << KVM_VM_DATA_SHIFT)
+#define KVM_VM_DATA_BASE	(KVM_VMM_BASE + KVM_VM_DATA_SIZE)
 
-#define KVM_P2M_BASE    KVM_VM_DATA_BASE
-#define KVM_P2M_OFS     0
-#define KVM_P2M_SIZE    (8UL << 20)
+#define KVM_P2M_BASE		KVM_VM_DATA_BASE
+#define KVM_P2M_SIZE		(__IA64_UL_CONST(24) << 20)
 
-#define KVM_VHPT_BASE   (KVM_P2M_BASE + KVM_P2M_SIZE)
-#define KVM_VHPT_OFS    KVM_P2M_SIZE
-#define KVM_VHPT_BLOCK_SIZE   (2UL << 20)
-#define VHPT_SHIFT      18
-#define VHPT_SIZE       (1UL << VHPT_SHIFT)
-#define VHPT_NUM_ENTRIES (1<<(VHPT_SHIFT-5))
+#define VHPT_SHIFT		16
+#define VHPT_SIZE		(__IA64_UL_CONST(1) << VHPT_SHIFT)
+#define VHPT_NUM_ENTRIES	(__IA64_UL_CONST(1) << (VHPT_SHIFT-5))
 
-#define KVM_VTLB_BASE   (KVM_VHPT_BASE+KVM_VHPT_BLOCK_SIZE)
-#define KVM_VTLB_OFS    (KVM_VHPT_OFS+KVM_VHPT_BLOCK_SIZE)
-#define KVM_VTLB_BLOCK_SIZE   (1UL<<20)
-#define VTLB_SHIFT      17
-#define VTLB_SIZE       (1UL<<VTLB_SHIFT)
-#define VTLB_NUM_ENTRIES (1<<(VTLB_SHIFT-5))
+#define VTLB_SHIFT		16
+#define VTLB_SIZE		(__IA64_UL_CONST(1) << VTLB_SHIFT)
+#define VTLB_NUM_ENTRIES	(1UL << (VHPT_SHIFT-5))
 
-#define KVM_VPD_BASE   (KVM_VTLB_BASE+KVM_VTLB_BLOCK_SIZE)
-#define KVM_VPD_OFS    (KVM_VTLB_OFS+KVM_VTLB_BLOCK_SIZE)
-#define KVM_VPD_BLOCK_SIZE   (2UL<<20)
-#define VPD_SHIFT       16
-#define VPD_SIZE        (1UL<<VPD_SHIFT)
+#define VPD_SHIFT		16
+#define VPD_SIZE		(__IA64_UL_CONST(1) << VPD_SHIFT)
 
-#define KVM_VCPU_BASE   (KVM_VPD_BASE+KVM_VPD_BLOCK_SIZE)
-#define KVM_VCPU_OFS    (KVM_VPD_OFS+KVM_VPD_BLOCK_SIZE)
-#define KVM_VCPU_BLOCK_SIZE   (2UL<<20)
-#define VCPU_SHIFT 18
-#define VCPU_SIZE (1UL<<VCPU_SHIFT)
-#define MAX_VCPU_NUM KVM_VCPU_BLOCK_SIZE/VCPU_SIZE
+#define VCPU_STRUCT_SHIFT	16
+#define VCPU_STRUCT_SIZE	(__IA64_UL_CONST(1) << VCPU_STRUCT_SHIFT)
 
-#define KVM_VM_BASE     (KVM_VCPU_BASE+KVM_VCPU_BLOCK_SIZE)
-#define KVM_VM_OFS      (KVM_VCPU_OFS+KVM_VCPU_BLOCK_SIZE)
-#define KVM_VM_BLOCK_SIZE     (1UL<<19)
+#define KVM_STK_OFFSET		VCPU_STRUCT_SIZE
 
-#define KVM_MEM_DIRTY_LOG_BASE (KVM_VM_BASE+KVM_VM_BLOCK_SIZE)
-#define KVM_MEM_DIRTY_LOG_OFS  (KVM_VM_OFS+KVM_VM_BLOCK_SIZE)
-#define KVM_MEM_DIRTY_LOG_SIZE (1UL<<19)
+#define KVM_VM_STRUCT_SHIFT	19
+#define KVM_VM_STRUCT_SIZE	(__IA64_UL_CONST(1) << KVM_VM_STRUCT_SHIFT)
 
-/* Get vpd, vhpt, tlb, vcpu, base*/
-#define VPD_ADDR(n) (KVM_VPD_BASE+n*VPD_SIZE)
-#define VHPT_ADDR(n) (KVM_VHPT_BASE+n*VHPT_SIZE)
-#define VTLB_ADDR(n) (KVM_VTLB_BASE+n*VTLB_SIZE)
-#define VCPU_ADDR(n) (KVM_VCPU_BASE+n*VCPU_SIZE)
+#define KVM_MEM_DIRY_LOG_SHIFT	19
+#define KVM_MEM_DIRTY_LOG_SIZE (__IA64_UL_CONST(1) << KVM_MEM_DIRY_LOG_SHIFT)
+
+#ifndef __ASSEMBLY__
+
+/*Define the max vcpus and memory for Guests.*/
+#define KVM_MAX_VCPUS	(KVM_VM_DATA_SIZE - KVM_P2M_SIZE - KVM_VM_STRUCT_SIZE -\
+			KVM_MEM_DIRTY_LOG_SIZE) / sizeof(struct kvm_vcpu_data)
+#define KVM_MAX_MEM_SIZE (KVM_P2M_SIZE >> 3 << PAGE_SHIFT)
+
+#define VMM_LOG_LEN 256
+
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/kvm.h>
+#include <linux/kvm_para.h>
+#include <linux/kvm_types.h>
+
+#include <asm/pal.h>
+#include <asm/sal.h>
+#include <asm/page.h>
+
+struct kvm_vcpu_data {
+	char vcpu_vhpt[VHPT_SIZE];
+	char vcpu_vtlb[VTLB_SIZE];
+	char vcpu_vpd[VPD_SIZE];
+	char vcpu_struct[VCPU_STRUCT_SIZE];
+};
+
+struct kvm_vm_data {
+	char kvm_p2m[KVM_P2M_SIZE];
+	char kvm_vm_struct[KVM_VM_STRUCT_SIZE];
+	char kvm_mem_dirty_log[KVM_MEM_DIRTY_LOG_SIZE];
+	struct kvm_vcpu_data vcpu_data[KVM_MAX_VCPUS];
+};
+
+#define VCPU_BASE(n)	KVM_VM_DATA_BASE + \
+				offsetof(struct kvm_vm_data, vcpu_data[n])
+#define VM_BASE		KVM_VM_DATA_BASE + \
+				offsetof(struct kvm_vm_data, kvm_vm_struct)
+#define KVM_MEM_DIRTY_LOG_BASE	KVM_VM_DATA_BASE + \
+				offsetof(struct kvm_vm_data, kvm_mem_dirty_log)
+
+#define VHPT_BASE(n) (VCPU_BASE(n) + offsetof(struct kvm_vcpu_data, vcpu_vhpt))
+#define VTLB_BASE(n) (VCPU_BASE(n) + offsetof(struct kvm_vcpu_data, vcpu_vtlb))
+#define VPD_BASE(n)  (VCPU_BASE(n) + offsetof(struct kvm_vcpu_data, vcpu_vpd))
+#define VCPU_STRUCT_BASE(n)	(VCPU_BASE(n) + \
+				offsetof(struct kvm_vcpu_data, vcpu_struct))
 
 /*IO section definitions*/
 #define IOREQ_READ      1
@@ -389,6 +440,7 @@
 
 	unsigned long opcode;
 	unsigned long cause;
+	char log_buf[VMM_LOG_LEN];
 	union context host;
 	union context guest;
 };
@@ -403,14 +455,13 @@
 };
 
 struct kvm_arch {
+	spinlock_t dirty_log_lock;
+
 	unsigned long	vm_base;
 	unsigned long	metaphysical_rr0;
 	unsigned long	metaphysical_rr4;
 	unsigned long	vmm_init_rr;
-	unsigned long	vhpt_base;
-	unsigned long	vtlb_base;
-	unsigned long 	vpd_base;
-	spinlock_t dirty_log_lock;
+
 	struct kvm_ioapic *vioapic;
 	struct kvm_vm_stat stat;
 	struct kvm_sal_data rdv_sal_data;
@@ -512,7 +563,7 @@
 
 static inline struct kvm_pt_regs *vcpu_regs(struct kvm_vcpu *v)
 {
-	return (struct kvm_pt_regs *) ((unsigned long) v + IA64_STK_OFFSET) - 1;
+	return (struct kvm_pt_regs *) ((unsigned long) v + KVM_STK_OFFSET) - 1;
 }
 
 typedef int kvm_vmm_entry(void);
@@ -531,5 +582,6 @@
 void kvm_sal_emul(struct kvm_vcpu *vcpu);
 
 static inline void kvm_inject_nmi(struct kvm_vcpu *vcpu) {}
+#endif /* __ASSEMBLY__*/
 
 #endif
diff --git a/arch/ia64/include/asm/smp.h b/arch/ia64/include/asm/smp.h
index 12d96e0..21c4023 100644
--- a/arch/ia64/include/asm/smp.h
+++ b/arch/ia64/include/asm/smp.h
@@ -57,7 +57,6 @@
 
 extern char no_int_routing __devinitdata;
 
-extern cpumask_t cpu_online_map;
 extern cpumask_t cpu_core_map[NR_CPUS];
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 extern int smp_num_siblings;
diff --git a/arch/ia64/include/asm/topology.h b/arch/ia64/include/asm/topology.h
index 35bcb64..a3cc9f6 100644
--- a/arch/ia64/include/asm/topology.h
+++ b/arch/ia64/include/asm/topology.h
@@ -55,7 +55,6 @@
 void build_cpu_to_node_map(void);
 
 #define SD_CPU_INIT (struct sched_domain) {		\
-	.span			= CPU_MASK_NONE,	\
 	.parent			= NULL,			\
 	.child			= NULL,			\
 	.groups			= NULL,			\
@@ -80,7 +79,6 @@
 
 /* sched_domains SD_NODE_INIT for IA64 NUMA machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
-	.span			= CPU_MASK_NONE,	\
 	.parent			= NULL,			\
 	.child			= NULL,			\
 	.groups			= NULL,			\
diff --git a/arch/ia64/kernel/init_task.c b/arch/ia64/kernel/init_task.c
index 9d7e1c6..5b0e830 100644
--- a/arch/ia64/kernel/init_task.c
+++ b/arch/ia64/kernel/init_task.c
@@ -17,7 +17,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index 5c4674a..c8adecd 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -330,25 +330,25 @@
 
 
 static void
-iosapic_set_affinity (unsigned int irq, cpumask_t mask)
+iosapic_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 #ifdef CONFIG_SMP
 	u32 high32, low32;
-	int dest, rte_index;
+	int cpu, dest, rte_index;
 	int redir = (irq & IA64_IRQ_REDIRECTED) ? 1 : 0;
 	struct iosapic_rte_info *rte;
 	struct iosapic *iosapic;
 
 	irq &= (~IA64_IRQ_REDIRECTED);
 
-	cpus_and(mask, mask, cpu_online_map);
-	if (cpus_empty(mask))
+	cpu = cpumask_first_and(cpu_online_mask, mask);
+	if (cpu >= nr_cpu_ids)
 		return;
 
-	if (irq_prepare_move(irq, first_cpu(mask)))
+	if (irq_prepare_move(irq, cpu))
 		return;
 
-	dest = cpu_physical_id(first_cpu(mask));
+	dest = cpu_physical_id(cpu);
 
 	if (!iosapic_intr_info[irq].count)
 		return;			/* not an IOSAPIC interrupt */
diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c
index 7fd18f5..0b6db53 100644
--- a/arch/ia64/kernel/irq.c
+++ b/arch/ia64/kernel/irq.c
@@ -133,7 +133,6 @@
  */
 static void migrate_irqs(void)
 {
-	cpumask_t	mask;
 	irq_desc_t *desc;
 	int 		irq, new_cpu;
 
@@ -152,15 +151,14 @@
 		if (desc->status == IRQ_PER_CPU)
 			continue;
 
-		cpus_and(mask, irq_desc[irq].affinity, cpu_online_map);
-		if (any_online_cpu(mask) == NR_CPUS) {
+		if (cpumask_any_and(&irq_desc[irq].affinity, cpu_online_mask)
+		    >= nr_cpu_ids) {
 			/*
 			 * Save it for phase 2 processing
 			 */
 			vectors_in_migration[irq] = irq;
 
 			new_cpu = any_online_cpu(cpu_online_map);
-			mask = cpumask_of_cpu(new_cpu);
 
 			/*
 			 * Al three are essential, currently WARN_ON.. maybe panic?
@@ -168,7 +166,8 @@
 			if (desc->chip && desc->chip->disable &&
 				desc->chip->enable && desc->chip->set_affinity) {
 				desc->chip->disable(irq);
-				desc->chip->set_affinity(irq, mask);
+				desc->chip->set_affinity(irq,
+							 cpumask_of(new_cpu));
 				desc->chip->enable(irq);
 			} else {
 				WARN_ON((!(desc->chip) || !(desc->chip->disable) ||
diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index 702a09c..8903393 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -49,11 +49,12 @@
 static struct irq_chip	ia64_msi_chip;
 
 #ifdef CONFIG_SMP
-static void ia64_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
+static void ia64_set_msi_irq_affinity(unsigned int irq,
+				      const cpumask_t *cpu_mask)
 {
 	struct msi_msg msg;
 	u32 addr, data;
-	int cpu = first_cpu(cpu_mask);
+	int cpu = first_cpu(*cpu_mask);
 
 	if (!cpu_online(cpu))
 		return;
@@ -166,12 +167,11 @@
 
 #ifdef CONFIG_DMAR
 #ifdef CONFIG_SMP
-static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg = irq_cfg + irq;
 	struct msi_msg msg;
-	int cpu = first_cpu(mask);
-
+	int cpu = cpumask_first(mask);
 
 	if (!cpu_online(cpu))
 		return;
@@ -187,7 +187,7 @@
 	msg.address_lo |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu));
 
 	dmar_msi_write(irq, &msg);
-	irq_desc[irq].affinity = mask;
+	irq_desc[irq].affinity = *mask;
 }
 #endif /* CONFIG_SMP */
 
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 1dcbb85..1146399 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -131,12 +131,6 @@
  */
 DEFINE_PER_CPU(int, cpu_state);
 
-/* Bitmasks of currently online, and possible CPUs */
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_possible_map);
-
 cpumask_t cpu_core_map[NR_CPUS] __cacheline_aligned;
 EXPORT_SYMBOL(cpu_core_map);
 DEFINE_PER_CPU_SHARED_ALIGNED(cpumask_t, cpu_sibling_map);
@@ -688,7 +682,7 @@
 {
 	int new_cpei_cpu;
 	irq_desc_t *desc = NULL;
-	cpumask_t 	mask;
+	const struct cpumask *mask;
 	int 		retval = 0;
 
 	/*
@@ -701,7 +695,7 @@
 			 * Now re-target the CPEI to a different processor
 			 */
 			new_cpei_cpu = any_online_cpu(cpu_online_map);
-			mask = cpumask_of_cpu(new_cpei_cpu);
+			mask = cpumask_of(new_cpei_cpu);
 			set_cpei_target_cpu(new_cpei_cpu);
 			desc = irq_desc + ia64_cpe_irq;
 			/*
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 65c10a4..f0ebb34 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -93,13 +93,14 @@
 	now = ia64_get_itc();
 
 	delta_stime = cycle_to_cputime(pi->ac_stime + (now - pi->ac_stamp));
-	account_system_time(prev, 0, delta_stime);
-	account_system_time_scaled(prev, delta_stime);
+	if (idle_task(smp_processor_id()) != prev)
+		account_system_time(prev, 0, delta_stime, delta_stime);
+	else
+		account_idle_time(delta_stime);
 
 	if (pi->ac_utime) {
 		delta_utime = cycle_to_cputime(pi->ac_utime);
-		account_user_time(prev, delta_utime);
-		account_user_time_scaled(prev, delta_utime);
+		account_user_time(prev, delta_utime, delta_utime);
 	}
 
 	pi->ac_stamp = ni->ac_stamp = now;
@@ -122,8 +123,10 @@
 	now = ia64_get_itc();
 
 	delta_stime = cycle_to_cputime(ti->ac_stime + (now - ti->ac_stamp));
-	account_system_time(tsk, 0, delta_stime);
-	account_system_time_scaled(tsk, delta_stime);
+	if (irq_count() || idle_task(smp_processor_id()) != tsk)
+		account_system_time(tsk, 0, delta_stime, delta_stime);
+	else
+		account_idle_time(delta_stime);
 	ti->ac_stime = 0;
 
 	ti->ac_stamp = now;
@@ -143,8 +146,7 @@
 
 	if (ti->ac_utime) {
 		delta_utime = cycle_to_cputime(ti->ac_utime);
-		account_user_time(p, delta_utime);
-		account_user_time_scaled(p, delta_utime);
+		account_user_time(p, delta_utime, delta_utime);
 		ti->ac_utime = 0;
 	}
 }
diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c
index c75b914..a8d61a3 100644
--- a/arch/ia64/kernel/topology.c
+++ b/arch/ia64/kernel/topology.c
@@ -219,7 +219,7 @@
 	cpumask_t shared_cpu_map;
 
 	cpus_and(shared_cpu_map, this_leaf->shared_cpu_map, cpu_online_map);
-	len = cpumask_scnprintf(buf, NR_CPUS+1, shared_cpu_map);
+	len = cpumask_scnprintf(buf, NR_CPUS+1, &shared_cpu_map);
 	len += sprintf(buf+len, "\n");
 	return len;
 }
diff --git a/arch/ia64/kvm/Makefile b/arch/ia64/kvm/Makefile
index 92cef66..76464dc 100644
--- a/arch/ia64/kvm/Makefile
+++ b/arch/ia64/kvm/Makefile
@@ -60,7 +60,7 @@
 
 CFLAGS_vcpu.o += -mfixed-range=f2-f5,f12-f127
 kvm-intel-objs = vmm.o vmm_ivt.o trampoline.o vcpu.o optvfault.o mmio.o \
-	vtlb.o process.o
+	vtlb.o process.o kvm_lib.o
 #Add link memcpy and memset to avoid possible structure assignment error
 kvm-intel-objs += memcpy.o memset.o
 obj-$(CONFIG_KVM_INTEL) += kvm-intel.o
diff --git a/arch/ia64/kvm/asm-offsets.c b/arch/ia64/kvm/asm-offsets.c
index 4e3dc13..0c3564a 100644
--- a/arch/ia64/kvm/asm-offsets.c
+++ b/arch/ia64/kvm/asm-offsets.c
@@ -24,19 +24,10 @@
 
 #include <linux/autoconf.h>
 #include <linux/kvm_host.h>
+#include <linux/kbuild.h>
 
 #include "vcpu.h"
 
-#define task_struct kvm_vcpu
-
-#define DEFINE(sym, val) \
-	asm volatile("\n->" #sym " (%0) " #val : : "i" (val))
-
-#define BLANK() asm volatile("\n->" : :)
-
-#define OFFSET(_sym, _str, _mem) \
-    DEFINE(_sym, offsetof(_str, _mem));
-
 void foo(void)
 {
 	DEFINE(VMM_TASK_SIZE, sizeof(struct kvm_vcpu));
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index af1464f..0f5ebd9 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -180,7 +180,6 @@
 
 	switch (ext) {
 	case KVM_CAP_IRQCHIP:
-	case KVM_CAP_USER_MEMORY:
 	case KVM_CAP_MP_STATE:
 
 		r = 1;
@@ -439,7 +438,6 @@
 		expires = div64_u64(itc_diff, cyc_per_usec);
 		kt = ktime_set(0, 1000 * expires);
 
-		down_read(&vcpu->kvm->slots_lock);
 		vcpu->arch.ht_active = 1;
 		hrtimer_start(p_ht, kt, HRTIMER_MODE_ABS);
 
@@ -452,7 +450,6 @@
 			if (vcpu->arch.mp_state == KVM_MP_STATE_HALTED)
 				vcpu->arch.mp_state =
 					KVM_MP_STATE_RUNNABLE;
-		up_read(&vcpu->kvm->slots_lock);
 
 		if (vcpu->arch.mp_state != KVM_MP_STATE_RUNNABLE)
 			return -EINTR;
@@ -476,6 +473,13 @@
 	return 1;
 }
 
+static int handle_vcpu_debug(struct kvm_vcpu *vcpu,
+				struct kvm_run *kvm_run)
+{
+	printk("VMM: %s", vcpu->arch.log_buf);
+	return 1;
+}
+
 static int (*kvm_vti_exit_handlers[])(struct kvm_vcpu *vcpu,
 		struct kvm_run *kvm_run) = {
 	[EXIT_REASON_VM_PANIC]              = handle_vm_error,
@@ -487,6 +491,7 @@
 	[EXIT_REASON_EXTERNAL_INTERRUPT]    = handle_external_interrupt,
 	[EXIT_REASON_IPI]		    = handle_ipi,
 	[EXIT_REASON_PTC_G]		    = handle_global_purge,
+	[EXIT_REASON_DEBUG]		    = handle_vcpu_debug,
 
 };
 
@@ -698,27 +703,24 @@
 	return r;
 }
 
-/*
- * Allocate 16M memory for every vm to hold its specific data.
- * Its memory map is defined in kvm_host.h.
- */
 static struct kvm *kvm_alloc_kvm(void)
 {
 
 	struct kvm *kvm;
 	uint64_t  vm_base;
 
+	BUG_ON(sizeof(struct kvm) > KVM_VM_STRUCT_SIZE);
+
 	vm_base = __get_free_pages(GFP_KERNEL, get_order(KVM_VM_DATA_SIZE));
 
 	if (!vm_base)
 		return ERR_PTR(-ENOMEM);
-	printk(KERN_DEBUG"kvm: VM data's base Address:0x%lx\n", vm_base);
 
-	/* Zero all pages before use! */
 	memset((void *)vm_base, 0, KVM_VM_DATA_SIZE);
-
-	kvm = (struct kvm *)(vm_base + KVM_VM_OFS);
+	kvm = (struct kvm *)(vm_base +
+			offsetof(struct kvm_vm_data, kvm_vm_struct));
 	kvm->arch.vm_base = vm_base;
+	printk(KERN_DEBUG"kvm: vm's data area:0x%lx\n", vm_base);
 
 	return kvm;
 }
@@ -760,21 +762,12 @@
 
 static void kvm_init_vm(struct kvm *kvm)
 {
-	long vm_base;
-
 	BUG_ON(!kvm);
 
 	kvm->arch.metaphysical_rr0 = GUEST_PHYSICAL_RR0;
 	kvm->arch.metaphysical_rr4 = GUEST_PHYSICAL_RR4;
 	kvm->arch.vmm_init_rr = VMM_INIT_RR;
 
-	vm_base = kvm->arch.vm_base;
-	if (vm_base) {
-		kvm->arch.vhpt_base = vm_base + KVM_VHPT_OFS;
-		kvm->arch.vtlb_base = vm_base + KVM_VTLB_OFS;
-		kvm->arch.vpd_base  = vm_base + KVM_VPD_OFS;
-	}
-
 	/*
 	 *Fill P2M entries for MMIO/IO ranges
 	 */
@@ -838,9 +831,8 @@
 
 int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 {
-	int i;
 	struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);
-	int r;
+	int i;
 
 	vcpu_load(vcpu);
 
@@ -857,18 +849,7 @@
 
 	vpd->vpr = regs->vpd.vpr;
 
-	r = -EFAULT;
-	r = copy_from_user(&vcpu->arch.guest, regs->saved_guest,
-						sizeof(union context));
-	if (r)
-		goto out;
-	r = copy_from_user(vcpu + 1, regs->saved_stack +
-			sizeof(struct kvm_vcpu),
-			IA64_STK_OFFSET - sizeof(struct kvm_vcpu));
-	if (r)
-		goto out;
-	vcpu->arch.exit_data =
-		((struct kvm_vcpu *)(regs->saved_stack))->arch.exit_data;
+	memcpy(&vcpu->arch.guest, &regs->saved_guest, sizeof(union context));
 
 	RESTORE_REGS(mp_state);
 	RESTORE_REGS(vmm_rr);
@@ -902,9 +883,8 @@
 	set_bit(KVM_REQ_RESUME, &vcpu->requests);
 
 	vcpu_put(vcpu);
-	r = 0;
-out:
-	return r;
+
+	return 0;
 }
 
 long kvm_arch_vm_ioctl(struct file *filp,
@@ -1166,10 +1146,11 @@
 		/*Set entry address for first run.*/
 		regs->cr_iip = PALE_RESET_ENTRY;
 
-		/*Initilize itc offset for vcpus*/
+		/*Initialize itc offset for vcpus*/
 		itc_offset = 0UL - ia64_getreg(_IA64_REG_AR_ITC);
-		for (i = 0; i < MAX_VCPU_NUM; i++) {
-			v = (struct kvm_vcpu *)((char *)vcpu + VCPU_SIZE * i);
+		for (i = 0; i < KVM_MAX_VCPUS; i++) {
+			v = (struct kvm_vcpu *)((char *)vcpu +
+					sizeof(struct kvm_vcpu_data) * i);
 			v->arch.itc_offset = itc_offset;
 			v->arch.last_itc = 0;
 		}
@@ -1183,7 +1164,7 @@
 	vcpu->arch.apic->vcpu = vcpu;
 
 	p_ctx->gr[1] = 0;
-	p_ctx->gr[12] = (unsigned long)((char *)vmm_vcpu + IA64_STK_OFFSET);
+	p_ctx->gr[12] = (unsigned long)((char *)vmm_vcpu + KVM_STK_OFFSET);
 	p_ctx->gr[13] = (unsigned long)vmm_vcpu;
 	p_ctx->psr = 0x1008522000UL;
 	p_ctx->ar[40] = FPSR_DEFAULT; /*fpsr*/
@@ -1218,12 +1199,12 @@
 	vcpu->arch.hlt_timer.function = hlt_timer_fn;
 
 	vcpu->arch.last_run_cpu = -1;
-	vcpu->arch.vpd = (struct vpd *)VPD_ADDR(vcpu->vcpu_id);
+	vcpu->arch.vpd = (struct vpd *)VPD_BASE(vcpu->vcpu_id);
 	vcpu->arch.vsa_base = kvm_vsa_base;
 	vcpu->arch.__gp = kvm_vmm_gp;
 	vcpu->arch.dirty_log_lock_pa = __pa(&kvm->arch.dirty_log_lock);
-	vcpu->arch.vhpt.hash = (struct thash_data *)VHPT_ADDR(vcpu->vcpu_id);
-	vcpu->arch.vtlb.hash = (struct thash_data *)VTLB_ADDR(vcpu->vcpu_id);
+	vcpu->arch.vhpt.hash = (struct thash_data *)VHPT_BASE(vcpu->vcpu_id);
+	vcpu->arch.vtlb.hash = (struct thash_data *)VTLB_BASE(vcpu->vcpu_id);
 	init_ptce_info(vcpu);
 
 	r = 0;
@@ -1273,12 +1254,22 @@
 	int r;
 	int cpu;
 
+	BUG_ON(sizeof(struct kvm_vcpu) > VCPU_STRUCT_SIZE/2);
+
+	r = -EINVAL;
+	if (id >= KVM_MAX_VCPUS) {
+		printk(KERN_ERR"kvm: Can't configure vcpus > %ld",
+				KVM_MAX_VCPUS);
+		goto fail;
+	}
+
 	r = -ENOMEM;
 	if (!vm_base) {
 		printk(KERN_ERR"kvm: Create vcpu[%d] error!\n", id);
 		goto fail;
 	}
-	vcpu = (struct kvm_vcpu *)(vm_base + KVM_VCPU_OFS + VCPU_SIZE * id);
+	vcpu = (struct kvm_vcpu *)(vm_base + offsetof(struct kvm_vm_data,
+					vcpu_data[id].vcpu_struct));
 	vcpu->kvm = kvm;
 
 	cpu = get_cpu();
@@ -1374,9 +1365,9 @@
 
 int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 {
-	int i;
-	int r;
 	struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);
+	int i;
+
 	vcpu_load(vcpu);
 
 	for (i = 0; i < 16; i++) {
@@ -1391,14 +1382,8 @@
 	regs->vpd.vpsr = vpd->vpsr;
 	regs->vpd.vpr = vpd->vpr;
 
-	r = -EFAULT;
-	r = copy_to_user(regs->saved_guest, &vcpu->arch.guest,
-					sizeof(union context));
-	if (r)
-		goto out;
-	r = copy_to_user(regs->saved_stack, (void *)vcpu, IA64_STK_OFFSET);
-	if (r)
-		goto out;
+	memcpy(&regs->saved_guest, &vcpu->arch.guest, sizeof(union context));
+
 	SAVE_REGS(mp_state);
 	SAVE_REGS(vmm_rr);
 	memcpy(regs->itrs, vcpu->arch.itrs, sizeof(struct thash_data) * NITRS);
@@ -1426,10 +1411,9 @@
 	SAVE_REGS(metaphysical_saved_rr4);
 	SAVE_REGS(fp_psr);
 	SAVE_REGS(saved_gp);
+
 	vcpu_put(vcpu);
-	r = 0;
-out:
-	return r;
+	return 0;
 }
 
 void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
@@ -1457,6 +1441,9 @@
 	struct kvm_memory_slot *memslot = &kvm->memslots[mem->slot];
 	unsigned long base_gfn = memslot->base_gfn;
 
+	if (base_gfn + npages > (KVM_MAX_MEM_SIZE >> PAGE_SHIFT))
+		return -ENOMEM;
+
 	for (i = 0; i < npages; i++) {
 		pfn = gfn_to_pfn(kvm, base_gfn + i);
 		if (!kvm_is_mmio_pfn(pfn)) {
@@ -1631,8 +1618,8 @@
 	struct kvm_memory_slot *memslot;
 	int r, i;
 	long n, base;
-	unsigned long *dirty_bitmap = (unsigned long *)((void *)kvm - KVM_VM_OFS
-					+ KVM_MEM_DIRTY_LOG_OFS);
+	unsigned long *dirty_bitmap = (unsigned long *)(kvm->arch.vm_base +
+			offsetof(struct kvm_vm_data, kvm_mem_dirty_log));
 
 	r = -EINVAL;
 	if (log->slot >= KVM_MEMORY_SLOTS)
diff --git a/arch/ia64/kvm/kvm_lib.c b/arch/ia64/kvm/kvm_lib.c
new file mode 100644
index 0000000..a85cb61
--- /dev/null
+++ b/arch/ia64/kvm/kvm_lib.c
@@ -0,0 +1,15 @@
+/*
+ * kvm_lib.c: Compile some libraries for kvm-intel module.
+ *
+ *	Just include kernel's library, and disable symbols export.
+ * 	Copyright (C) 2008, Intel Corporation.
+ *  	Xiantao Zhang  (xiantao.zhang@intel.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#undef CONFIG_MODULES
+#include "../../../lib/vsprintf.c"
+#include "../../../lib/ctype.c"
diff --git a/arch/ia64/kvm/kvm_minstate.h b/arch/ia64/kvm/kvm_minstate.h
index 2cc41d1..b2bcaa2 100644
--- a/arch/ia64/kvm/kvm_minstate.h
+++ b/arch/ia64/kvm/kvm_minstate.h
@@ -24,6 +24,8 @@
 #include <asm/asmmacro.h>
 #include <asm/types.h>
 #include <asm/kregs.h>
+#include <asm/kvm_host.h>
+
 #include "asm-offsets.h"
 
 #define KVM_MINSTATE_START_SAVE_MIN	     					\
@@ -33,7 +35,7 @@
 	addl r22 = VMM_RBS_OFFSET,r1;            /* compute base of RBS */	\
 	;;									\
 	lfetch.fault.excl.nt1 [r22];						\
-	addl r1 = IA64_STK_OFFSET-VMM_PT_REGS_SIZE,r1;  /* compute base of memory stack */  \
+	addl r1 = KVM_STK_OFFSET-VMM_PT_REGS_SIZE, r1;  \
 	mov r23 = ar.bspstore;			/* save ar.bspstore */          \
 	;;									\
 	mov ar.bspstore = r22;				/* switch to kernel RBS */\
diff --git a/arch/ia64/kvm/misc.h b/arch/ia64/kvm/misc.h
index e585c46..dd979e0 100644
--- a/arch/ia64/kvm/misc.h
+++ b/arch/ia64/kvm/misc.h
@@ -27,7 +27,8 @@
  */
 static inline uint64_t *kvm_host_get_pmt(struct kvm *kvm)
 {
-	return (uint64_t *)(kvm->arch.vm_base + KVM_P2M_OFS);
+	return (uint64_t *)(kvm->arch.vm_base +
+				offsetof(struct kvm_vm_data, kvm_p2m));
 }
 
 static inline void kvm_set_pmt_entry(struct kvm *kvm, gfn_t gfn,
diff --git a/arch/ia64/kvm/mmio.c b/arch/ia64/kvm/mmio.c
index 7f1a858..21f63ff 100644
--- a/arch/ia64/kvm/mmio.c
+++ b/arch/ia64/kvm/mmio.c
@@ -66,31 +66,25 @@
 
 	switch (addr) {
 	case PIB_OFST_INTA:
-		/*panic_domain(NULL, "Undefined write on PIB INTA\n");*/
-		panic_vm(v);
+		panic_vm(v, "Undefined write on PIB INTA\n");
 		break;
 	case PIB_OFST_XTP:
 		if (length == 1) {
 			vlsapic_write_xtp(v, val);
 		} else {
-			/*panic_domain(NULL,
-			"Undefined write on PIB XTP\n");*/
-			panic_vm(v);
+			panic_vm(v, "Undefined write on PIB XTP\n");
 		}
 		break;
 	default:
 		if (PIB_LOW_HALF(addr)) {
-			/*lower half */
+			/*Lower half */
 			if (length != 8)
-				/*panic_domain(NULL,
-				"Can't LHF write with size %ld!\n",
-				length);*/
-				panic_vm(v);
+				panic_vm(v, "Can't LHF write with size %ld!\n",
+						length);
 			else
 				vlsapic_write_ipi(v, addr, val);
-		} else {   /*	upper half
-				printk("IPI-UHF write %lx\n",addr);*/
-			panic_vm(v);
+		} else {   /*Upper half */
+			panic_vm(v, "IPI-UHF write %lx\n", addr);
 		}
 		break;
 	}
@@ -108,22 +102,18 @@
 		if (length == 1) /* 1 byte load */
 			; /* There is no i8259, there is no INTA access*/
 		else
-			/*panic_domain(NULL,"Undefined read on PIB INTA\n"); */
-			panic_vm(v);
+			panic_vm(v, "Undefined read on PIB INTA\n");
 
 		break;
 	case PIB_OFST_XTP:
 		if (length == 1) {
 			result = VLSAPIC_XTP(v);
-			/* printk("read xtp %lx\n", result); */
 		} else {
-			/*panic_domain(NULL,
-			"Undefined read on PIB XTP\n");*/
-			panic_vm(v);
+			panic_vm(v, "Undefined read on PIB XTP\n");
 		}
 		break;
 	default:
-		panic_vm(v);
+		panic_vm(v, "Undefined addr access for lsapic!\n");
 		break;
 	}
 	return result;
@@ -162,7 +152,7 @@
 			/* it's necessary to ensure zero extending */
 			*dest = p->u.ioreq.data & (~0UL >> (64-(s*8)));
 	} else
-		panic_vm(vcpu);
+		panic_vm(vcpu, "Unhandled mmio access returned!\n");
 out:
 	local_irq_restore(psr);
 	return ;
@@ -324,7 +314,9 @@
 		return;
 	} else {
 		inst_type = -1;
-		panic_vm(vcpu);
+		panic_vm(vcpu, "Unsupported MMIO access instruction! \
+				Bunld[0]=0x%lx, Bundle[1]=0x%lx\n",
+				bundle.i64[0], bundle.i64[1]);
 	}
 
 	size = 1 << size;
@@ -335,7 +327,7 @@
 		if (inst_type == SL_INTEGER)
 			vcpu_set_gr(vcpu, inst.M1.r1, data, 0);
 		else
-			panic_vm(vcpu);
+			panic_vm(vcpu, "Unsupported instruction type!\n");
 
 	}
 	vcpu_increment_iip(vcpu);
diff --git a/arch/ia64/kvm/process.c b/arch/ia64/kvm/process.c
index 8008173..552d077 100644
--- a/arch/ia64/kvm/process.c
+++ b/arch/ia64/kvm/process.c
@@ -527,7 +527,8 @@
 	vector = vec2off[vec];
 
 	if (!(vpsr & IA64_PSR_IC) && (vector != IA64_DATA_NESTED_TLB_VECTOR)) {
-		panic_vm(vcpu);
+		panic_vm(vcpu, "Interruption with vector :0x%lx occurs "
+						"with psr.ic = 0\n", vector);
 		return;
 	}
 
@@ -586,7 +587,7 @@
 		vcpu_set_gr(vcpu, 10, p->u.pal_data.ret.v1, 0);
 		vcpu_set_gr(vcpu, 11, p->u.pal_data.ret.v2, 0);
 	} else
-		panic_vm(vcpu);
+		panic_vm(vcpu, "Mis-set for exit reason!\n");
 }
 
 static void set_sal_call_data(struct kvm_vcpu *vcpu)
@@ -614,7 +615,7 @@
 		vcpu_set_gr(vcpu, 10, p->u.sal_data.ret.r10, 0);
 		vcpu_set_gr(vcpu, 11, p->u.sal_data.ret.r11, 0);
 	} else
-		panic_vm(vcpu);
+		panic_vm(vcpu, "Mis-set for exit reason!\n");
 }
 
 void  kvm_ia64_handle_break(unsigned long ifa, struct kvm_pt_regs *regs,
@@ -680,7 +681,7 @@
 	vpsr = VCPU(vcpu, vpsr);
 	isr = vpsr & IA64_PSR_RI;
 	if (!(vpsr & IA64_PSR_IC))
-		panic_vm(vcpu);
+		panic_vm(vcpu, "Trying to inject one IRQ with psr.ic=0\n");
 	reflect_interruption(0, isr, 0, 12, regs); /* EXT IRQ */
 }
 
@@ -941,8 +942,20 @@
 	ia64_set_pta(vcpu->arch.vhpt.pta.val);
 }
 
+static void vmm_sanity_check(struct kvm_vcpu *vcpu)
+{
+	struct exit_ctl_data *p = &vcpu->arch.exit_data;
+
+	if (!vmm_sanity && p->exit_reason != EXIT_REASON_DEBUG) {
+		panic_vm(vcpu, "Failed to do vmm sanity check,"
+			"it maybe caused by crashed vmm!!\n\n");
+	}
+}
+
 static void kvm_do_resume_op(struct kvm_vcpu *vcpu)
 {
+	vmm_sanity_check(vcpu); /*Guarantee vcpu runing on healthy vmm!*/
+
 	if (test_and_clear_bit(KVM_REQ_RESUME, &vcpu->requests)) {
 		vcpu_do_resume(vcpu);
 		return;
@@ -968,3 +981,11 @@
 						1, 0, 0, 0, 0, 0);
 	kvm_do_resume_op(vcpu);
 }
+
+void vmm_panic_handler(u64 vec)
+{
+	struct kvm_vcpu *vcpu = current_vcpu;
+	vmm_sanity = 0;
+	panic_vm(vcpu, "Unexpected interruption occurs in VMM, vector:0x%lx\n",
+			vec2off[vec]);
+}
diff --git a/arch/ia64/kvm/vcpu.c b/arch/ia64/kvm/vcpu.c
index e44027c..ecd526b 100644
--- a/arch/ia64/kvm/vcpu.c
+++ b/arch/ia64/kvm/vcpu.c
@@ -816,8 +816,9 @@
 	unsigned long vitv = VCPU(vcpu, itv);
 
 	if (vcpu->vcpu_id == 0) {
-		for (i = 0; i < MAX_VCPU_NUM; i++) {
-			v = (struct kvm_vcpu *)((char *)vcpu + VCPU_SIZE * i);
+		for (i = 0; i < KVM_MAX_VCPUS; i++) {
+			v = (struct kvm_vcpu *)((char *)vcpu +
+					sizeof(struct kvm_vcpu_data) * i);
 			VMX(v, itc_offset) = itc_offset;
 			VMX(v, last_itc) = 0;
 		}
@@ -1650,7 +1651,8 @@
 	 * Otherwise panic
 	 */
 	if (val & (IA64_PSR_PK | IA64_PSR_IS | IA64_PSR_VM))
-		panic_vm(vcpu);
+		panic_vm(vcpu, "Only support guests with vpsr.pk =0 \
+				& vpsr.is=0\n");
 
 	/*
 	 * For those IA64_PSR bits: id/da/dd/ss/ed/ia
@@ -2103,7 +2105,7 @@
 
 	if (is_physical_mode(vcpu)) {
 		if (vcpu->arch.mode_flags & GUEST_PHY_EMUL)
-			panic_vm(vcpu);
+			panic_vm(vcpu, "Machine Status conflicts!\n");
 
 		ia64_set_rr((VRN0 << VRN_SHIFT), vcpu->arch.metaphysical_rr0);
 		ia64_dv_serialize_data();
@@ -2152,10 +2154,70 @@
 	return 0;
 }
 
-void panic_vm(struct kvm_vcpu *v)
+static void kvm_show_registers(struct kvm_pt_regs *regs)
 {
-	struct exit_ctl_data *p = &v->arch.exit_data;
+	unsigned long ip = regs->cr_iip + ia64_psr(regs)->ri;
 
+	struct kvm_vcpu *vcpu = current_vcpu;
+	if (vcpu != NULL)
+		printk("vcpu 0x%p vcpu %d\n",
+		       vcpu, vcpu->vcpu_id);
+
+	printk("psr : %016lx ifs : %016lx ip  : [<%016lx>]\n",
+	       regs->cr_ipsr, regs->cr_ifs, ip);
+
+	printk("unat: %016lx pfs : %016lx rsc : %016lx\n",
+	       regs->ar_unat, regs->ar_pfs, regs->ar_rsc);
+	printk("rnat: %016lx bspstore: %016lx pr  : %016lx\n",
+	       regs->ar_rnat, regs->ar_bspstore, regs->pr);
+	printk("ldrs: %016lx ccv : %016lx fpsr: %016lx\n",
+	       regs->loadrs, regs->ar_ccv, regs->ar_fpsr);
+	printk("csd : %016lx ssd : %016lx\n", regs->ar_csd, regs->ar_ssd);
+	printk("b0  : %016lx b6  : %016lx b7  : %016lx\n", regs->b0,
+							regs->b6, regs->b7);
+	printk("f6  : %05lx%016lx f7  : %05lx%016lx\n",
+	       regs->f6.u.bits[1], regs->f6.u.bits[0],
+	       regs->f7.u.bits[1], regs->f7.u.bits[0]);
+	printk("f8  : %05lx%016lx f9  : %05lx%016lx\n",
+	       regs->f8.u.bits[1], regs->f8.u.bits[0],
+	       regs->f9.u.bits[1], regs->f9.u.bits[0]);
+	printk("f10 : %05lx%016lx f11 : %05lx%016lx\n",
+	       regs->f10.u.bits[1], regs->f10.u.bits[0],
+	       regs->f11.u.bits[1], regs->f11.u.bits[0]);
+
+	printk("r1  : %016lx r2  : %016lx r3  : %016lx\n", regs->r1,
+							regs->r2, regs->r3);
+	printk("r8  : %016lx r9  : %016lx r10 : %016lx\n", regs->r8,
+							regs->r9, regs->r10);
+	printk("r11 : %016lx r12 : %016lx r13 : %016lx\n", regs->r11,
+							regs->r12, regs->r13);
+	printk("r14 : %016lx r15 : %016lx r16 : %016lx\n", regs->r14,
+							regs->r15, regs->r16);
+	printk("r17 : %016lx r18 : %016lx r19 : %016lx\n", regs->r17,
+							regs->r18, regs->r19);
+	printk("r20 : %016lx r21 : %016lx r22 : %016lx\n", regs->r20,
+							regs->r21, regs->r22);
+	printk("r23 : %016lx r24 : %016lx r25 : %016lx\n", regs->r23,
+							regs->r24, regs->r25);
+	printk("r26 : %016lx r27 : %016lx r28 : %016lx\n", regs->r26,
+							regs->r27, regs->r28);
+	printk("r29 : %016lx r30 : %016lx r31 : %016lx\n", regs->r29,
+							regs->r30, regs->r31);
+
+}
+
+void panic_vm(struct kvm_vcpu *v, const char *fmt, ...)
+{
+	va_list args;
+	char buf[256];
+
+	struct kvm_pt_regs *regs = vcpu_regs(v);
+	struct exit_ctl_data *p = &v->arch.exit_data;
+	va_start(args, fmt);
+	vsnprintf(buf, sizeof(buf), fmt, args);
+	va_end(args);
+	printk(buf);
+	kvm_show_registers(regs);
 	p->exit_reason = EXIT_REASON_VM_PANIC;
 	vmm_transition(v);
 	/*Never to return*/
diff --git a/arch/ia64/kvm/vcpu.h b/arch/ia64/kvm/vcpu.h
index e9b2a4e..b2f12a5 100644
--- a/arch/ia64/kvm/vcpu.h
+++ b/arch/ia64/kvm/vcpu.h
@@ -737,9 +737,12 @@
 void kvm_init_vhpt(struct kvm_vcpu *v);
 void thash_init(struct thash_cb *hcb, u64 sz);
 
-void panic_vm(struct kvm_vcpu *v);
+void panic_vm(struct kvm_vcpu *v, const char *fmt, ...);
 
 extern u64 ia64_call_vsa(u64 proc, u64 arg1, u64 arg2, u64 arg3,
 		u64 arg4, u64 arg5, u64 arg6, u64 arg7);
+
+extern long vmm_sanity;
+
 #endif
 #endif	/* __VCPU_H__ */
diff --git a/arch/ia64/kvm/vmm.c b/arch/ia64/kvm/vmm.c
index 2275bf4..9eee5c0 100644
--- a/arch/ia64/kvm/vmm.c
+++ b/arch/ia64/kvm/vmm.c
@@ -20,6 +20,7 @@
  */
 
 
+#include<linux/kernel.h>
 #include<linux/module.h>
 #include<asm/fpswa.h>
 
@@ -31,6 +32,8 @@
 extern char kvm_ia64_ivt;
 extern fpswa_interface_t *vmm_fpswa_interface;
 
+long vmm_sanity = 1;
+
 struct kvm_vmm_info vmm_info = {
 	.module	     = THIS_MODULE,
 	.vmm_entry   = vmm_entry,
@@ -62,5 +65,31 @@
 {
 	_vmm_raw_spin_unlock(lock);
 }
+
+static void vcpu_debug_exit(struct kvm_vcpu *vcpu)
+{
+	struct exit_ctl_data *p = &vcpu->arch.exit_data;
+	long psr;
+
+	local_irq_save(psr);
+	p->exit_reason = EXIT_REASON_DEBUG;
+	vmm_transition(vcpu);
+	local_irq_restore(psr);
+}
+
+asmlinkage int printk(const char *fmt, ...)
+{
+	struct kvm_vcpu *vcpu = current_vcpu;
+	va_list args;
+	int r;
+
+	memset(vcpu->arch.log_buf, 0, VMM_LOG_LEN);
+	va_start(args, fmt);
+	r = vsnprintf(vcpu->arch.log_buf, VMM_LOG_LEN, fmt, args);
+	va_end(args);
+	vcpu_debug_exit(vcpu);
+	return r;
+}
+
 module_init(kvm_vmm_init)
 module_exit(kvm_vmm_exit)
diff --git a/arch/ia64/kvm/vmm_ivt.S b/arch/ia64/kvm/vmm_ivt.S
index c1d7251..3ef1a01 100644
--- a/arch/ia64/kvm/vmm_ivt.S
+++ b/arch/ia64/kvm/vmm_ivt.S
@@ -1,5 +1,5 @@
 /*
- * /ia64/kvm_ivt.S
+ * arch/ia64/kvm/vmm_ivt.S
  *
  * Copyright (C) 1998-2001, 2003 Hewlett-Packard Co
  *      Stephane Eranian <eranian@hpl.hp.com>
@@ -70,32 +70,39 @@
 # define PSR_DEFAULT_BITS   0
 #endif
 
-
 #define KVM_FAULT(n)    \
-    kvm_fault_##n:;          \
-    mov r19=n;;          \
-    br.sptk.many kvm_fault_##n;         \
-    ;;                  \
-
+	kvm_fault_##n:;          \
+	mov r19=n;;          \
+	br.sptk.many kvm_vmm_panic;         \
+	;;                  \
 
 #define KVM_REFLECT(n)    \
-    mov r31=pr;           \
-    mov r19=n;       /* prepare to save predicates */ \
-    mov r29=cr.ipsr;      \
-    ;;      \
-    tbit.z p6,p7=r29,IA64_PSR_VM_BIT;       \
-(p7)br.sptk.many kvm_dispatch_reflection;        \
-    br.sptk.many kvm_panic;      \
+	mov r31=pr;           \
+	mov r19=n;       /* prepare to save predicates */ \
+	mov r29=cr.ipsr;      \
+	;;      \
+	tbit.z p6,p7=r29,IA64_PSR_VM_BIT;       \
+(p7)	br.sptk.many kvm_dispatch_reflection;        \
+	br.sptk.many kvm_vmm_panic;      \
 
-
-GLOBAL_ENTRY(kvm_panic)
-    br.sptk.many kvm_panic
-    ;;
-END(kvm_panic)
-
-
-
-
+GLOBAL_ENTRY(kvm_vmm_panic)
+	KVM_SAVE_MIN_WITH_COVER_R19
+	alloc r14=ar.pfs,0,0,1,0
+	mov out0=r15
+	adds r3=8,r2                // set up second base pointer
+	;;
+	ssm psr.ic
+	;;
+	srlz.i    // guarantee that interruption collection is on
+	;;
+	//(p15) ssm psr.i               // restore psr.i
+	addl r14=@gprel(ia64_leave_hypervisor),gp
+	;;
+	KVM_SAVE_REST
+	mov rp=r14
+	;;
+	br.call.sptk.many b6=vmm_panic_handler;
+END(kvm_vmm_panic)
 
     .section .text.ivt,"ax"
 
@@ -105,308 +112,307 @@
 ///////////////////////////////////////////////////////////////
 // 0x0000 Entry 0 (size 64 bundles) VHPT Translation (8,20,47)
 ENTRY(kvm_vhpt_miss)
-    KVM_FAULT(0)
+	KVM_FAULT(0)
 END(kvm_vhpt_miss)
 
-
     .org kvm_ia64_ivt+0x400
 ////////////////////////////////////////////////////////////////
 // 0x0400 Entry 1 (size 64 bundles) ITLB (21)
 ENTRY(kvm_itlb_miss)
-    mov r31 = pr
-    mov r29=cr.ipsr;
-    ;;
-    tbit.z p6,p7=r29,IA64_PSR_VM_BIT;
-    (p6) br.sptk kvm_alt_itlb_miss
-    mov r19 = 1
-    br.sptk kvm_itlb_miss_dispatch
-    KVM_FAULT(1);
+	mov r31 = pr
+	mov r29=cr.ipsr;
+	;;
+	tbit.z p6,p7=r29,IA64_PSR_VM_BIT;
+(p6)	br.sptk kvm_alt_itlb_miss
+	mov r19 = 1
+	br.sptk kvm_itlb_miss_dispatch
+	KVM_FAULT(1);
 END(kvm_itlb_miss)
 
     .org kvm_ia64_ivt+0x0800
 //////////////////////////////////////////////////////////////////
 // 0x0800 Entry 2 (size 64 bundles) DTLB (9,48)
 ENTRY(kvm_dtlb_miss)
-    mov r31 = pr
-    mov r29=cr.ipsr;
-    ;;
-    tbit.z p6,p7=r29,IA64_PSR_VM_BIT;
-(p6)br.sptk kvm_alt_dtlb_miss
-    br.sptk kvm_dtlb_miss_dispatch
+	mov r31 = pr
+	mov r29=cr.ipsr;
+	;;
+	tbit.z p6,p7=r29,IA64_PSR_VM_BIT;
+(p6)	br.sptk kvm_alt_dtlb_miss
+	br.sptk kvm_dtlb_miss_dispatch
 END(kvm_dtlb_miss)
 
      .org kvm_ia64_ivt+0x0c00
 ////////////////////////////////////////////////////////////////////
 // 0x0c00 Entry 3 (size 64 bundles) Alt ITLB (19)
 ENTRY(kvm_alt_itlb_miss)
-    mov r16=cr.ifa    // get address that caused the TLB miss
-    ;;
-    movl r17=PAGE_KERNEL
-    mov r24=cr.ipsr
-    movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
-    ;;
-    and r19=r19,r16     // clear ed, reserved bits, and PTE control bits
-    ;;
-    or r19=r17,r19      // insert PTE control bits into r19
-    ;;
-    movl r20=IA64_GRANULE_SHIFT<<2
-    ;;
-    mov cr.itir=r20
-    ;;
-    itc.i r19		// insert the TLB entry
-    mov pr=r31,-1
-    rfi
+	mov r16=cr.ifa    // get address that caused the TLB miss
+	;;
+	movl r17=PAGE_KERNEL
+	mov r24=cr.ipsr
+	movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
+	;;
+	and r19=r19,r16     // clear ed, reserved bits, and PTE control bits
+	;;
+	or r19=r17,r19      // insert PTE control bits into r19
+	;;
+	movl r20=IA64_GRANULE_SHIFT<<2
+	;;
+	mov cr.itir=r20
+	;;
+	itc.i r19		// insert the TLB entry
+	mov pr=r31,-1
+	rfi
 END(kvm_alt_itlb_miss)
 
     .org kvm_ia64_ivt+0x1000
 /////////////////////////////////////////////////////////////////////
 // 0x1000 Entry 4 (size 64 bundles) Alt DTLB (7,46)
 ENTRY(kvm_alt_dtlb_miss)
-    mov r16=cr.ifa		// get address that caused the TLB miss
-    ;;
-    movl r17=PAGE_KERNEL
-    movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
-    mov r24=cr.ipsr
-    ;;
-    and r19=r19,r16     // clear ed, reserved bits, and PTE control bits
-    ;;
-    or r19=r19,r17	// insert PTE control bits into r19
-    ;;
-    movl r20=IA64_GRANULE_SHIFT<<2
-    ;;
-    mov cr.itir=r20
-    ;;
-    itc.d r19		// insert the TLB entry
-    mov pr=r31,-1
-    rfi
+	mov r16=cr.ifa		// get address that caused the TLB miss
+	;;
+	movl r17=PAGE_KERNEL
+	movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
+	mov r24=cr.ipsr
+	;;
+	and r19=r19,r16     // clear ed, reserved bits, and PTE control bits
+	;;
+	or r19=r19,r17	// insert PTE control bits into r19
+	;;
+	movl r20=IA64_GRANULE_SHIFT<<2
+	;;
+	mov cr.itir=r20
+	;;
+	itc.d r19		// insert the TLB entry
+	mov pr=r31,-1
+	rfi
 END(kvm_alt_dtlb_miss)
 
     .org kvm_ia64_ivt+0x1400
 //////////////////////////////////////////////////////////////////////
 // 0x1400 Entry 5 (size 64 bundles) Data nested TLB (6,45)
 ENTRY(kvm_nested_dtlb_miss)
-    KVM_FAULT(5)
+	KVM_FAULT(5)
 END(kvm_nested_dtlb_miss)
 
     .org kvm_ia64_ivt+0x1800
 /////////////////////////////////////////////////////////////////////
 // 0x1800 Entry 6 (size 64 bundles) Instruction Key Miss (24)
 ENTRY(kvm_ikey_miss)
-    KVM_REFLECT(6)
+	KVM_REFLECT(6)
 END(kvm_ikey_miss)
 
     .org kvm_ia64_ivt+0x1c00
 /////////////////////////////////////////////////////////////////////
 // 0x1c00 Entry 7 (size 64 bundles) Data Key Miss (12,51)
 ENTRY(kvm_dkey_miss)
-    KVM_REFLECT(7)
+	KVM_REFLECT(7)
 END(kvm_dkey_miss)
 
     .org kvm_ia64_ivt+0x2000
 ////////////////////////////////////////////////////////////////////
 // 0x2000 Entry 8 (size 64 bundles) Dirty-bit (54)
 ENTRY(kvm_dirty_bit)
-    KVM_REFLECT(8)
+	KVM_REFLECT(8)
 END(kvm_dirty_bit)
 
     .org kvm_ia64_ivt+0x2400
 ////////////////////////////////////////////////////////////////////
 // 0x2400 Entry 9 (size 64 bundles) Instruction Access-bit (27)
 ENTRY(kvm_iaccess_bit)
-    KVM_REFLECT(9)
+	KVM_REFLECT(9)
 END(kvm_iaccess_bit)
 
     .org kvm_ia64_ivt+0x2800
 ///////////////////////////////////////////////////////////////////
 // 0x2800 Entry 10 (size 64 bundles) Data Access-bit (15,55)
 ENTRY(kvm_daccess_bit)
-    KVM_REFLECT(10)
+	KVM_REFLECT(10)
 END(kvm_daccess_bit)
 
     .org kvm_ia64_ivt+0x2c00
 /////////////////////////////////////////////////////////////////
 // 0x2c00 Entry 11 (size 64 bundles) Break instruction (33)
 ENTRY(kvm_break_fault)
-    mov r31=pr
-    mov r19=11
-    mov r29=cr.ipsr
-    ;;
-    KVM_SAVE_MIN_WITH_COVER_R19
-    ;;
-    alloc r14=ar.pfs,0,0,4,0 // now it's safe (must be first in insn group!)
-    mov out0=cr.ifa
-    mov out2=cr.isr     // FIXME: pity to make this slow access twice
-    mov out3=cr.iim     // FIXME: pity to make this slow access twice
-    adds r3=8,r2                // set up second base pointer
-    ;;
-    ssm psr.ic
-    ;;
-    srlz.i                  // guarantee that interruption collection is on
-    ;;
-    //(p15)ssm psr.i               // restore psr.i
-    addl r14=@gprel(ia64_leave_hypervisor),gp
-    ;;
-    KVM_SAVE_REST
-    mov rp=r14
-    ;;
-    adds out1=16,sp
-    br.call.sptk.many b6=kvm_ia64_handle_break
-    ;;
+	mov r31=pr
+	mov r19=11
+	mov r29=cr.ipsr
+	;;
+	KVM_SAVE_MIN_WITH_COVER_R19
+	;;
+	alloc r14=ar.pfs,0,0,4,0 //(must be first in insn group!)
+	mov out0=cr.ifa
+	mov out2=cr.isr     // FIXME: pity to make this slow access twice
+	mov out3=cr.iim     // FIXME: pity to make this slow access twice
+	adds r3=8,r2                // set up second base pointer
+	;;
+	ssm psr.ic
+	;;
+	srlz.i         // guarantee that interruption collection is on
+	;;
+	//(p15)ssm psr.i               // restore psr.i
+	addl r14=@gprel(ia64_leave_hypervisor),gp
+	;;
+	KVM_SAVE_REST
+	mov rp=r14
+	;;
+	adds out1=16,sp
+	br.call.sptk.many b6=kvm_ia64_handle_break
+	;;
 END(kvm_break_fault)
 
     .org kvm_ia64_ivt+0x3000
 /////////////////////////////////////////////////////////////////
 // 0x3000 Entry 12 (size 64 bundles) External Interrupt (4)
 ENTRY(kvm_interrupt)
-    mov r31=pr		// prepare to save predicates
-    mov r19=12
-    mov r29=cr.ipsr
-    ;;
-    tbit.z p6,p7=r29,IA64_PSR_VM_BIT
-    tbit.z p0,p15=r29,IA64_PSR_I_BIT
-    ;;
-(p7) br.sptk kvm_dispatch_interrupt
-    ;;
-    mov r27=ar.rsc		/* M */
-    mov r20=r1			/* A */
-    mov r25=ar.unat		/* M */
-    mov r26=ar.pfs		/* I */
-    mov r28=cr.iip		/* M */
-    cover			/* B (or nothing) */
-    ;;
-    mov r1=sp
-    ;;
-    invala			/* M */
-    mov r30=cr.ifs
-    ;;
-    addl r1=-VMM_PT_REGS_SIZE,r1
-    ;;
-    adds r17=2*L1_CACHE_BYTES,r1	/* really: biggest cache-line size */
-    adds r16=PT(CR_IPSR),r1
-    ;;
-    lfetch.fault.excl.nt1 [r17],L1_CACHE_BYTES
-    st8 [r16]=r29			/* save cr.ipsr */
-    ;;
-    lfetch.fault.excl.nt1 [r17]
-    mov r29=b0
-    ;;
-    adds r16=PT(R8),r1  	/* initialize first base pointer */
-    adds r17=PT(R9),r1  	/* initialize second base pointer */
-    mov r18=r0      		/* make sure r18 isn't NaT */
-    ;;
+	mov r31=pr		// prepare to save predicates
+	mov r19=12
+	mov r29=cr.ipsr
+	;;
+	tbit.z p6,p7=r29,IA64_PSR_VM_BIT
+	tbit.z p0,p15=r29,IA64_PSR_I_BIT
+	;;
+(p7)	br.sptk kvm_dispatch_interrupt
+	;;
+	mov r27=ar.rsc		/* M */
+	mov r20=r1			/* A */
+	mov r25=ar.unat		/* M */
+	mov r26=ar.pfs		/* I */
+	mov r28=cr.iip		/* M */
+	cover			/* B (or nothing) */
+	;;
+	mov r1=sp
+	;;
+	invala			/* M */
+	mov r30=cr.ifs
+	;;
+	addl r1=-VMM_PT_REGS_SIZE,r1
+	;;
+	adds r17=2*L1_CACHE_BYTES,r1	/* really: biggest cache-line size */
+	adds r16=PT(CR_IPSR),r1
+	;;
+	lfetch.fault.excl.nt1 [r17],L1_CACHE_BYTES
+	st8 [r16]=r29			/* save cr.ipsr */
+	;;
+	lfetch.fault.excl.nt1 [r17]
+	mov r29=b0
+	;;
+	adds r16=PT(R8),r1  	/* initialize first base pointer */
+	adds r17=PT(R9),r1  	/* initialize second base pointer */
+	mov r18=r0      		/* make sure r18 isn't NaT */
+	;;
 .mem.offset 0,0; st8.spill [r16]=r8,16
 .mem.offset 8,0; st8.spill [r17]=r9,16
         ;;
 .mem.offset 0,0; st8.spill [r16]=r10,24
 .mem.offset 8,0; st8.spill [r17]=r11,24
         ;;
-    st8 [r16]=r28,16		/* save cr.iip */
-    st8 [r17]=r30,16		/* save cr.ifs */
-    mov r8=ar.fpsr		/* M */
-    mov r9=ar.csd
-    mov r10=ar.ssd
-    movl r11=FPSR_DEFAULT	/* L-unit */
-    ;;
-    st8 [r16]=r25,16		/* save ar.unat */
-    st8 [r17]=r26,16		/* save ar.pfs */
-    shl r18=r18,16		/* compute ar.rsc to be used for "loadrs" */
-    ;;
-    st8 [r16]=r27,16		/* save ar.rsc */
-    adds r17=16,r17		/* skip over ar_rnat field */
-    ;;
-    st8 [r17]=r31,16		/* save predicates */
-    adds r16=16,r16		/* skip over ar_bspstore field */
-    ;;
-    st8 [r16]=r29,16		/* save b0 */
-    st8 [r17]=r18,16		/* save ar.rsc value for "loadrs" */
-    ;;
+	st8 [r16]=r28,16		/* save cr.iip */
+	st8 [r17]=r30,16		/* save cr.ifs */
+	mov r8=ar.fpsr		/* M */
+	mov r9=ar.csd
+	mov r10=ar.ssd
+	movl r11=FPSR_DEFAULT	/* L-unit */
+	;;
+	st8 [r16]=r25,16		/* save ar.unat */
+	st8 [r17]=r26,16		/* save ar.pfs */
+	shl r18=r18,16		/* compute ar.rsc to be used for "loadrs" */
+	;;
+	st8 [r16]=r27,16		/* save ar.rsc */
+	adds r17=16,r17		/* skip over ar_rnat field */
+	;;
+	st8 [r17]=r31,16		/* save predicates */
+	adds r16=16,r16		/* skip over ar_bspstore field */
+	;;
+	st8 [r16]=r29,16		/* save b0 */
+	st8 [r17]=r18,16		/* save ar.rsc value for "loadrs" */
+	;;
 .mem.offset 0,0; st8.spill [r16]=r20,16    /* save original r1 */
 .mem.offset 8,0; st8.spill [r17]=r12,16
-    adds r12=-16,r1
-    /* switch to kernel memory stack (with 16 bytes of scratch) */
-    ;;
+	adds r12=-16,r1
+	/* switch to kernel memory stack (with 16 bytes of scratch) */
+	;;
 .mem.offset 0,0; st8.spill [r16]=r13,16
 .mem.offset 8,0; st8.spill [r17]=r8,16 /* save ar.fpsr */
-    ;;
+	;;
 .mem.offset 0,0; st8.spill [r16]=r15,16
 .mem.offset 8,0; st8.spill [r17]=r14,16
-    dep r14=-1,r0,60,4
-    ;;
+	dep r14=-1,r0,60,4
+	;;
 .mem.offset 0,0; st8.spill [r16]=r2,16
 .mem.offset 8,0; st8.spill [r17]=r3,16
-    adds r2=VMM_PT_REGS_R16_OFFSET,r1
-    adds r14 = VMM_VCPU_GP_OFFSET,r13
-    ;;
-    mov r8=ar.ccv
-    ld8 r14 = [r14]
-    ;;
-    mov r1=r14       /* establish kernel global pointer */
-    ;;                                          \
-    bsw.1
-    ;;
-    alloc r14=ar.pfs,0,0,1,0	// must be first in an insn group
-    mov out0=r13
-    ;;
-    ssm psr.ic
-    ;;
-    srlz.i
-    ;;
-    //(p15) ssm psr.i
-    adds r3=8,r2		// set up second base pointer for SAVE_REST
-    srlz.i			// ensure everybody knows psr.ic is back on
-    ;;
+	adds r2=VMM_PT_REGS_R16_OFFSET,r1
+	adds r14 = VMM_VCPU_GP_OFFSET,r13
+	;;
+	mov r8=ar.ccv
+	ld8 r14 = [r14]
+	;;
+	mov r1=r14       /* establish kernel global pointer */
+	;;                                          \
+	bsw.1
+	;;
+	alloc r14=ar.pfs,0,0,1,0	// must be first in an insn group
+	mov out0=r13
+	;;
+	ssm psr.ic
+	;;
+	srlz.i
+	;;
+	//(p15) ssm psr.i
+	adds r3=8,r2		// set up second base pointer for SAVE_REST
+	srlz.i			// ensure everybody knows psr.ic is back on
+	;;
 .mem.offset 0,0; st8.spill [r2]=r16,16
 .mem.offset 8,0; st8.spill [r3]=r17,16
-    ;;
+	;;
 .mem.offset 0,0; st8.spill [r2]=r18,16
 .mem.offset 8,0; st8.spill [r3]=r19,16
-    ;;
+	;;
 .mem.offset 0,0; st8.spill [r2]=r20,16
 .mem.offset 8,0; st8.spill [r3]=r21,16
-    mov r18=b6
-    ;;
+	mov r18=b6
+	;;
 .mem.offset 0,0; st8.spill [r2]=r22,16
 .mem.offset 8,0; st8.spill [r3]=r23,16
-    mov r19=b7
-    ;;
+	mov r19=b7
+	;;
 .mem.offset 0,0; st8.spill [r2]=r24,16
 .mem.offset 8,0; st8.spill [r3]=r25,16
-    ;;
+	;;
 .mem.offset 0,0; st8.spill [r2]=r26,16
 .mem.offset 8,0; st8.spill [r3]=r27,16
-    ;;
+	;;
 .mem.offset 0,0; st8.spill [r2]=r28,16
 .mem.offset 8,0; st8.spill [r3]=r29,16
-    ;;
+	;;
 .mem.offset 0,0; st8.spill [r2]=r30,16
 .mem.offset 8,0; st8.spill [r3]=r31,32
-    ;;
-    mov ar.fpsr=r11       /* M-unit */
-    st8 [r2]=r8,8         /* ar.ccv */
-    adds r24=PT(B6)-PT(F7),r3
-    ;;
-    stf.spill [r2]=f6,32
-    stf.spill [r3]=f7,32
-    ;;
-    stf.spill [r2]=f8,32
-    stf.spill [r3]=f9,32
-    ;;
-    stf.spill [r2]=f10
-    stf.spill [r3]=f11
-    adds r25=PT(B7)-PT(F11),r3
-    ;;
-    st8 [r24]=r18,16       /* b6 */
-    st8 [r25]=r19,16       /* b7 */
-    ;;
-    st8 [r24]=r9           /* ar.csd */
-    st8 [r25]=r10          /* ar.ssd */
-    ;;
-    srlz.d		// make sure we see the effect of cr.ivr
-    addl r14=@gprel(ia64_leave_nested),gp
-    ;;
-    mov rp=r14
-    br.call.sptk.many b6=kvm_ia64_handle_irq
-    ;;
+	;;
+	mov ar.fpsr=r11       /* M-unit */
+	st8 [r2]=r8,8         /* ar.ccv */
+	adds r24=PT(B6)-PT(F7),r3
+	;;
+	stf.spill [r2]=f6,32
+	stf.spill [r3]=f7,32
+	;;
+	stf.spill [r2]=f8,32
+	stf.spill [r3]=f9,32
+	;;
+	stf.spill [r2]=f10
+	stf.spill [r3]=f11
+	adds r25=PT(B7)-PT(F11),r3
+	;;
+	st8 [r24]=r18,16       /* b6 */
+	st8 [r25]=r19,16       /* b7 */
+	;;
+	st8 [r24]=r9           /* ar.csd */
+	st8 [r25]=r10          /* ar.ssd */
+	;;
+	srlz.d		// make sure we see the effect of cr.ivr
+	addl r14=@gprel(ia64_leave_nested),gp
+	;;
+	mov rp=r14
+	br.call.sptk.many b6=kvm_ia64_handle_irq
+	;;
 END(kvm_interrupt)
 
     .global kvm_dispatch_vexirq
@@ -414,387 +420,385 @@
 //////////////////////////////////////////////////////////////////////
 // 0x3400 Entry 13 (size 64 bundles) Reserved
 ENTRY(kvm_virtual_exirq)
-    mov r31=pr
-    mov r19=13
-    mov r30 =r0
-    ;;
+	mov r31=pr
+	mov r19=13
+	mov r30 =r0
+	;;
 kvm_dispatch_vexirq:
-    cmp.eq p6,p0 = 1,r30
-    ;;
-(p6)add r29 = VMM_VCPU_SAVED_GP_OFFSET,r21
-    ;;
-(p6)ld8 r1 = [r29]
-    ;;
-    KVM_SAVE_MIN_WITH_COVER_R19
-    alloc r14=ar.pfs,0,0,1,0
-    mov out0=r13
+	cmp.eq p6,p0 = 1,r30
+	;;
+(p6)	add r29 = VMM_VCPU_SAVED_GP_OFFSET,r21
+	;;
+(p6)	ld8 r1 = [r29]
+	;;
+	KVM_SAVE_MIN_WITH_COVER_R19
+	alloc r14=ar.pfs,0,0,1,0
+	mov out0=r13
 
-    ssm psr.ic
-    ;;
-    srlz.i                  // guarantee that interruption collection is on
-    ;;
-    //(p15) ssm psr.i               // restore psr.i
-    adds r3=8,r2                // set up second base pointer
-    ;;
-    KVM_SAVE_REST
-    addl r14=@gprel(ia64_leave_hypervisor),gp
-    ;;
-    mov rp=r14
-    br.call.sptk.many b6=kvm_vexirq
+	ssm psr.ic
+	;;
+	srlz.i // guarantee that interruption collection is on
+	;;
+	//(p15) ssm psr.i               // restore psr.i
+	adds r3=8,r2                // set up second base pointer
+	;;
+	KVM_SAVE_REST
+	addl r14=@gprel(ia64_leave_hypervisor),gp
+	;;
+	mov rp=r14
+	br.call.sptk.many b6=kvm_vexirq
 END(kvm_virtual_exirq)
 
     .org kvm_ia64_ivt+0x3800
 /////////////////////////////////////////////////////////////////////
 // 0x3800 Entry 14 (size 64 bundles) Reserved
-    KVM_FAULT(14)
-    // this code segment is from 2.6.16.13
-
+	KVM_FAULT(14)
+	// this code segment is from 2.6.16.13
 
     .org kvm_ia64_ivt+0x3c00
 ///////////////////////////////////////////////////////////////////////
 // 0x3c00 Entry 15 (size 64 bundles) Reserved
-    KVM_FAULT(15)
-
+	KVM_FAULT(15)
 
     .org kvm_ia64_ivt+0x4000
 ///////////////////////////////////////////////////////////////////////
 // 0x4000 Entry 16 (size 64 bundles) Reserved
-    KVM_FAULT(16)
+	KVM_FAULT(16)
 
     .org kvm_ia64_ivt+0x4400
 //////////////////////////////////////////////////////////////////////
 // 0x4400 Entry 17 (size 64 bundles) Reserved
-    KVM_FAULT(17)
+	KVM_FAULT(17)
 
     .org kvm_ia64_ivt+0x4800
 //////////////////////////////////////////////////////////////////////
 // 0x4800 Entry 18 (size 64 bundles) Reserved
-    KVM_FAULT(18)
+	KVM_FAULT(18)
 
     .org kvm_ia64_ivt+0x4c00
 //////////////////////////////////////////////////////////////////////
 // 0x4c00 Entry 19 (size 64 bundles) Reserved
-    KVM_FAULT(19)
+	KVM_FAULT(19)
 
     .org kvm_ia64_ivt+0x5000
 //////////////////////////////////////////////////////////////////////
 // 0x5000 Entry 20 (size 16 bundles) Page Not Present
 ENTRY(kvm_page_not_present)
-    KVM_REFLECT(20)
+	KVM_REFLECT(20)
 END(kvm_page_not_present)
 
     .org kvm_ia64_ivt+0x5100
 ///////////////////////////////////////////////////////////////////////
 // 0x5100 Entry 21 (size 16 bundles) Key Permission vector
 ENTRY(kvm_key_permission)
-    KVM_REFLECT(21)
+	KVM_REFLECT(21)
 END(kvm_key_permission)
 
     .org kvm_ia64_ivt+0x5200
 //////////////////////////////////////////////////////////////////////
 // 0x5200 Entry 22 (size 16 bundles) Instruction Access Rights (26)
 ENTRY(kvm_iaccess_rights)
-    KVM_REFLECT(22)
+	KVM_REFLECT(22)
 END(kvm_iaccess_rights)
 
     .org kvm_ia64_ivt+0x5300
 //////////////////////////////////////////////////////////////////////
 // 0x5300 Entry 23 (size 16 bundles) Data Access Rights (14,53)
 ENTRY(kvm_daccess_rights)
-    KVM_REFLECT(23)
+	KVM_REFLECT(23)
 END(kvm_daccess_rights)
 
     .org kvm_ia64_ivt+0x5400
 /////////////////////////////////////////////////////////////////////
 // 0x5400 Entry 24 (size 16 bundles) General Exception (5,32,34,36,38,39)
 ENTRY(kvm_general_exception)
-   KVM_REFLECT(24)
-   KVM_FAULT(24)
+	KVM_REFLECT(24)
+	KVM_FAULT(24)
 END(kvm_general_exception)
 
     .org kvm_ia64_ivt+0x5500
 //////////////////////////////////////////////////////////////////////
 // 0x5500 Entry 25 (size 16 bundles) Disabled FP-Register (35)
 ENTRY(kvm_disabled_fp_reg)
-    KVM_REFLECT(25)
+	KVM_REFLECT(25)
 END(kvm_disabled_fp_reg)
 
     .org kvm_ia64_ivt+0x5600
 ////////////////////////////////////////////////////////////////////
 // 0x5600 Entry 26 (size 16 bundles) Nat Consumption (11,23,37,50)
 ENTRY(kvm_nat_consumption)
-    KVM_REFLECT(26)
+	KVM_REFLECT(26)
 END(kvm_nat_consumption)
 
     .org kvm_ia64_ivt+0x5700
 /////////////////////////////////////////////////////////////////////
 // 0x5700 Entry 27 (size 16 bundles) Speculation (40)
 ENTRY(kvm_speculation_vector)
-    KVM_REFLECT(27)
+	KVM_REFLECT(27)
 END(kvm_speculation_vector)
 
     .org kvm_ia64_ivt+0x5800
 /////////////////////////////////////////////////////////////////////
 // 0x5800 Entry 28 (size 16 bundles) Reserved
-    KVM_FAULT(28)
+	KVM_FAULT(28)
 
     .org kvm_ia64_ivt+0x5900
 ///////////////////////////////////////////////////////////////////
 // 0x5900 Entry 29 (size 16 bundles) Debug (16,28,56)
 ENTRY(kvm_debug_vector)
-    KVM_FAULT(29)
+	KVM_FAULT(29)
 END(kvm_debug_vector)
 
     .org kvm_ia64_ivt+0x5a00
 ///////////////////////////////////////////////////////////////
 // 0x5a00 Entry 30 (size 16 bundles) Unaligned Reference (57)
 ENTRY(kvm_unaligned_access)
-    KVM_REFLECT(30)
+	KVM_REFLECT(30)
 END(kvm_unaligned_access)
 
     .org kvm_ia64_ivt+0x5b00
 //////////////////////////////////////////////////////////////////////
 // 0x5b00 Entry 31 (size 16 bundles) Unsupported Data Reference (57)
 ENTRY(kvm_unsupported_data_reference)
-    KVM_REFLECT(31)
+	KVM_REFLECT(31)
 END(kvm_unsupported_data_reference)
 
     .org kvm_ia64_ivt+0x5c00
 ////////////////////////////////////////////////////////////////////
 // 0x5c00 Entry 32 (size 16 bundles) Floating Point FAULT (65)
 ENTRY(kvm_floating_point_fault)
-    KVM_REFLECT(32)
+	KVM_REFLECT(32)
 END(kvm_floating_point_fault)
 
     .org kvm_ia64_ivt+0x5d00
 /////////////////////////////////////////////////////////////////////
 // 0x5d00 Entry 33 (size 16 bundles) Floating Point Trap (66)
 ENTRY(kvm_floating_point_trap)
-    KVM_REFLECT(33)
+	KVM_REFLECT(33)
 END(kvm_floating_point_trap)
 
     .org kvm_ia64_ivt+0x5e00
 //////////////////////////////////////////////////////////////////////
 // 0x5e00 Entry 34 (size 16 bundles) Lower Privilege Transfer Trap (66)
 ENTRY(kvm_lower_privilege_trap)
-    KVM_REFLECT(34)
+	KVM_REFLECT(34)
 END(kvm_lower_privilege_trap)
 
     .org kvm_ia64_ivt+0x5f00
 //////////////////////////////////////////////////////////////////////
 // 0x5f00 Entry 35 (size 16 bundles) Taken Branch Trap (68)
 ENTRY(kvm_taken_branch_trap)
-    KVM_REFLECT(35)
+	KVM_REFLECT(35)
 END(kvm_taken_branch_trap)
 
     .org kvm_ia64_ivt+0x6000
 ////////////////////////////////////////////////////////////////////
 // 0x6000 Entry 36 (size 16 bundles) Single Step Trap (69)
 ENTRY(kvm_single_step_trap)
-    KVM_REFLECT(36)
+	KVM_REFLECT(36)
 END(kvm_single_step_trap)
     .global kvm_virtualization_fault_back
     .org kvm_ia64_ivt+0x6100
 /////////////////////////////////////////////////////////////////////
 // 0x6100 Entry 37 (size 16 bundles) Virtualization Fault
 ENTRY(kvm_virtualization_fault)
-    mov r31=pr
-    adds r16 = VMM_VCPU_SAVED_GP_OFFSET,r21
-    ;;
-    st8 [r16] = r1
-    adds r17 = VMM_VCPU_GP_OFFSET, r21
-    ;;
-    ld8 r1 = [r17]
-    cmp.eq p6,p0=EVENT_MOV_FROM_AR,r24
-    cmp.eq p7,p0=EVENT_MOV_FROM_RR,r24
-    cmp.eq p8,p0=EVENT_MOV_TO_RR,r24
-    cmp.eq p9,p0=EVENT_RSM,r24
-    cmp.eq p10,p0=EVENT_SSM,r24
-    cmp.eq p11,p0=EVENT_MOV_TO_PSR,r24
-    cmp.eq p12,p0=EVENT_THASH,r24
-    (p6) br.dptk.many kvm_asm_mov_from_ar
-    (p7) br.dptk.many kvm_asm_mov_from_rr
-    (p8) br.dptk.many kvm_asm_mov_to_rr
-    (p9) br.dptk.many kvm_asm_rsm
-    (p10) br.dptk.many kvm_asm_ssm
-    (p11) br.dptk.many kvm_asm_mov_to_psr
-    (p12) br.dptk.many kvm_asm_thash
-    ;;
+	mov r31=pr
+	adds r16 = VMM_VCPU_SAVED_GP_OFFSET,r21
+	;;
+	st8 [r16] = r1
+	adds r17 = VMM_VCPU_GP_OFFSET, r21
+	;;
+	ld8 r1 = [r17]
+	cmp.eq p6,p0=EVENT_MOV_FROM_AR,r24
+	cmp.eq p7,p0=EVENT_MOV_FROM_RR,r24
+	cmp.eq p8,p0=EVENT_MOV_TO_RR,r24
+	cmp.eq p9,p0=EVENT_RSM,r24
+	cmp.eq p10,p0=EVENT_SSM,r24
+	cmp.eq p11,p0=EVENT_MOV_TO_PSR,r24
+	cmp.eq p12,p0=EVENT_THASH,r24
+(p6)	br.dptk.many kvm_asm_mov_from_ar
+(p7)	br.dptk.many kvm_asm_mov_from_rr
+(p8)	br.dptk.many kvm_asm_mov_to_rr
+(p9)	br.dptk.many kvm_asm_rsm
+(p10)	br.dptk.many kvm_asm_ssm
+(p11)	br.dptk.many kvm_asm_mov_to_psr
+(p12)	br.dptk.many kvm_asm_thash
+	;;
 kvm_virtualization_fault_back:
-    adds r16 = VMM_VCPU_SAVED_GP_OFFSET,r21
-    ;;
-    ld8 r1 = [r16]
-    ;;
-    mov r19=37
-    adds r16 = VMM_VCPU_CAUSE_OFFSET,r21
-    adds r17 = VMM_VCPU_OPCODE_OFFSET,r21
-    ;;
-    st8 [r16] = r24
-    st8 [r17] = r25
-    ;;
-    cmp.ne p6,p0=EVENT_RFI, r24
-    (p6) br.sptk kvm_dispatch_virtualization_fault
-    ;;
-    adds r18=VMM_VPD_BASE_OFFSET,r21
-    ;;
-    ld8 r18=[r18]
-    ;;
-    adds r18=VMM_VPD_VIFS_OFFSET,r18
-    ;;
-    ld8 r18=[r18]
-    ;;
-    tbit.z p6,p0=r18,63
-    (p6) br.sptk kvm_dispatch_virtualization_fault
-    ;;
-    //if vifs.v=1 desert current register frame
-    alloc r18=ar.pfs,0,0,0,0
-    br.sptk kvm_dispatch_virtualization_fault
+	adds r16 = VMM_VCPU_SAVED_GP_OFFSET,r21
+	;;
+	ld8 r1 = [r16]
+	;;
+	mov r19=37
+	adds r16 = VMM_VCPU_CAUSE_OFFSET,r21
+	adds r17 = VMM_VCPU_OPCODE_OFFSET,r21
+	;;
+	st8 [r16] = r24
+	st8 [r17] = r25
+	;;
+	cmp.ne p6,p0=EVENT_RFI, r24
+(p6)	br.sptk kvm_dispatch_virtualization_fault
+	;;
+	adds r18=VMM_VPD_BASE_OFFSET,r21
+	;;
+	ld8 r18=[r18]
+	;;
+	adds r18=VMM_VPD_VIFS_OFFSET,r18
+	;;
+	ld8 r18=[r18]
+	;;
+	tbit.z p6,p0=r18,63
+(p6)	br.sptk kvm_dispatch_virtualization_fault
+	;;
+//if vifs.v=1 desert current register frame
+	alloc r18=ar.pfs,0,0,0,0
+	br.sptk kvm_dispatch_virtualization_fault
 END(kvm_virtualization_fault)
 
     .org kvm_ia64_ivt+0x6200
 //////////////////////////////////////////////////////////////
 // 0x6200 Entry 38 (size 16 bundles) Reserved
-    KVM_FAULT(38)
+	KVM_FAULT(38)
 
     .org kvm_ia64_ivt+0x6300
 /////////////////////////////////////////////////////////////////
 // 0x6300 Entry 39 (size 16 bundles) Reserved
-    KVM_FAULT(39)
+	KVM_FAULT(39)
 
     .org kvm_ia64_ivt+0x6400
 /////////////////////////////////////////////////////////////////
 // 0x6400 Entry 40 (size 16 bundles) Reserved
-    KVM_FAULT(40)
+	KVM_FAULT(40)
 
     .org kvm_ia64_ivt+0x6500
 //////////////////////////////////////////////////////////////////
 // 0x6500 Entry 41 (size 16 bundles) Reserved
-    KVM_FAULT(41)
+	KVM_FAULT(41)
 
     .org kvm_ia64_ivt+0x6600
 //////////////////////////////////////////////////////////////////
 // 0x6600 Entry 42 (size 16 bundles) Reserved
-    KVM_FAULT(42)
+	KVM_FAULT(42)
 
     .org kvm_ia64_ivt+0x6700
 //////////////////////////////////////////////////////////////////
 // 0x6700 Entry 43 (size 16 bundles) Reserved
-    KVM_FAULT(43)
+	KVM_FAULT(43)
 
     .org kvm_ia64_ivt+0x6800
 //////////////////////////////////////////////////////////////////
 // 0x6800 Entry 44 (size 16 bundles) Reserved
-    KVM_FAULT(44)
+	KVM_FAULT(44)
 
     .org kvm_ia64_ivt+0x6900
 ///////////////////////////////////////////////////////////////////
 // 0x6900 Entry 45 (size 16 bundles) IA-32 Exeception
 //(17,18,29,41,42,43,44,58,60,61,62,72,73,75,76,77)
 ENTRY(kvm_ia32_exception)
-    KVM_FAULT(45)
+	KVM_FAULT(45)
 END(kvm_ia32_exception)
 
     .org kvm_ia64_ivt+0x6a00
 ////////////////////////////////////////////////////////////////////
 // 0x6a00 Entry 46 (size 16 bundles) IA-32 Intercept  (30,31,59,70,71)
 ENTRY(kvm_ia32_intercept)
-    KVM_FAULT(47)
+	KVM_FAULT(47)
 END(kvm_ia32_intercept)
 
     .org kvm_ia64_ivt+0x6c00
 /////////////////////////////////////////////////////////////////////
 // 0x6c00 Entry 48 (size 16 bundles) Reserved
-    KVM_FAULT(48)
+	KVM_FAULT(48)
 
     .org kvm_ia64_ivt+0x6d00
 //////////////////////////////////////////////////////////////////////
 // 0x6d00 Entry 49 (size 16 bundles) Reserved
-    KVM_FAULT(49)
+	KVM_FAULT(49)
 
     .org kvm_ia64_ivt+0x6e00
 //////////////////////////////////////////////////////////////////////
 // 0x6e00 Entry 50 (size 16 bundles) Reserved
-    KVM_FAULT(50)
+	KVM_FAULT(50)
 
     .org kvm_ia64_ivt+0x6f00
 /////////////////////////////////////////////////////////////////////
 // 0x6f00 Entry 51 (size 16 bundles) Reserved
-    KVM_FAULT(52)
+	KVM_FAULT(52)
 
     .org kvm_ia64_ivt+0x7100
 ////////////////////////////////////////////////////////////////////
 // 0x7100 Entry 53 (size 16 bundles) Reserved
-    KVM_FAULT(53)
+	KVM_FAULT(53)
 
     .org kvm_ia64_ivt+0x7200
 /////////////////////////////////////////////////////////////////////
 // 0x7200 Entry 54 (size 16 bundles) Reserved
-    KVM_FAULT(54)
+	KVM_FAULT(54)
 
     .org kvm_ia64_ivt+0x7300
 ////////////////////////////////////////////////////////////////////
 // 0x7300 Entry 55 (size 16 bundles) Reserved
-    KVM_FAULT(55)
+	KVM_FAULT(55)
 
     .org kvm_ia64_ivt+0x7400
 ////////////////////////////////////////////////////////////////////
 // 0x7400 Entry 56 (size 16 bundles) Reserved
-    KVM_FAULT(56)
+	KVM_FAULT(56)
 
     .org kvm_ia64_ivt+0x7500
 /////////////////////////////////////////////////////////////////////
 // 0x7500 Entry 57 (size 16 bundles) Reserved
-    KVM_FAULT(57)
+	KVM_FAULT(57)
 
     .org kvm_ia64_ivt+0x7600
 /////////////////////////////////////////////////////////////////////
 // 0x7600 Entry 58 (size 16 bundles) Reserved
-    KVM_FAULT(58)
+	KVM_FAULT(58)
 
     .org kvm_ia64_ivt+0x7700
 ////////////////////////////////////////////////////////////////////
 // 0x7700 Entry 59 (size 16 bundles) Reserved
-    KVM_FAULT(59)
+	KVM_FAULT(59)
 
     .org kvm_ia64_ivt+0x7800
 ////////////////////////////////////////////////////////////////////
 // 0x7800 Entry 60 (size 16 bundles) Reserved
-    KVM_FAULT(60)
+	KVM_FAULT(60)
 
     .org kvm_ia64_ivt+0x7900
 /////////////////////////////////////////////////////////////////////
 // 0x7900 Entry 61 (size 16 bundles) Reserved
-    KVM_FAULT(61)
+	KVM_FAULT(61)
 
     .org kvm_ia64_ivt+0x7a00
 /////////////////////////////////////////////////////////////////////
 // 0x7a00 Entry 62 (size 16 bundles) Reserved
-    KVM_FAULT(62)
+	KVM_FAULT(62)
 
     .org kvm_ia64_ivt+0x7b00
 /////////////////////////////////////////////////////////////////////
 // 0x7b00 Entry 63 (size 16 bundles) Reserved
-    KVM_FAULT(63)
+	KVM_FAULT(63)
 
     .org kvm_ia64_ivt+0x7c00
 ////////////////////////////////////////////////////////////////////
 // 0x7c00 Entry 64 (size 16 bundles) Reserved
-    KVM_FAULT(64)
+	KVM_FAULT(64)
 
     .org kvm_ia64_ivt+0x7d00
 /////////////////////////////////////////////////////////////////////
 // 0x7d00 Entry 65 (size 16 bundles) Reserved
-    KVM_FAULT(65)
+	KVM_FAULT(65)
 
     .org kvm_ia64_ivt+0x7e00
 /////////////////////////////////////////////////////////////////////
 // 0x7e00 Entry 66 (size 16 bundles) Reserved
-    KVM_FAULT(66)
+	KVM_FAULT(66)
 
     .org kvm_ia64_ivt+0x7f00
 ////////////////////////////////////////////////////////////////////
 // 0x7f00 Entry 67 (size 16 bundles) Reserved
-    KVM_FAULT(67)
+	KVM_FAULT(67)
 
     .org kvm_ia64_ivt+0x8000
 // There is no particular reason for this code to be here, other than that
@@ -804,132 +808,128 @@
 
 
 ENTRY(kvm_dtlb_miss_dispatch)
-    mov r19 = 2
-    KVM_SAVE_MIN_WITH_COVER_R19
-    alloc r14=ar.pfs,0,0,3,0
-    mov out0=cr.ifa
-    mov out1=r15
-    adds r3=8,r2                // set up second base pointer
-    ;;
-    ssm psr.ic
-    ;;
-    srlz.i                  // guarantee that interruption collection is on
-    ;;
-    //(p15) ssm psr.i               // restore psr.i
-    addl r14=@gprel(ia64_leave_hypervisor_prepare),gp
-    ;;
-    KVM_SAVE_REST
-    KVM_SAVE_EXTRA
-    mov rp=r14
-    ;;
-    adds out2=16,r12
-    br.call.sptk.many b6=kvm_page_fault
+	mov r19 = 2
+	KVM_SAVE_MIN_WITH_COVER_R19
+	alloc r14=ar.pfs,0,0,3,0
+	mov out0=cr.ifa
+	mov out1=r15
+	adds r3=8,r2                // set up second base pointer
+	;;
+	ssm psr.ic
+	;;
+	srlz.i     // guarantee that interruption collection is on
+	;;
+	//(p15) ssm psr.i               // restore psr.i
+	addl r14=@gprel(ia64_leave_hypervisor_prepare),gp
+	;;
+	KVM_SAVE_REST
+	KVM_SAVE_EXTRA
+	mov rp=r14
+	;;
+	adds out2=16,r12
+	br.call.sptk.many b6=kvm_page_fault
 END(kvm_dtlb_miss_dispatch)
 
 ENTRY(kvm_itlb_miss_dispatch)
 
-    KVM_SAVE_MIN_WITH_COVER_R19
-    alloc r14=ar.pfs,0,0,3,0
-    mov out0=cr.ifa
-    mov out1=r15
-    adds r3=8,r2                // set up second base pointer
-    ;;
-    ssm psr.ic
-    ;;
-    srlz.i                  // guarantee that interruption collection is on
-    ;;
-    //(p15) ssm psr.i               // restore psr.i
-    addl r14=@gprel(ia64_leave_hypervisor),gp
-    ;;
-    KVM_SAVE_REST
-    mov rp=r14
-    ;;
-    adds out2=16,r12
-    br.call.sptk.many b6=kvm_page_fault
+	KVM_SAVE_MIN_WITH_COVER_R19
+	alloc r14=ar.pfs,0,0,3,0
+	mov out0=cr.ifa
+	mov out1=r15
+	adds r3=8,r2                // set up second base pointer
+	;;
+	ssm psr.ic
+	;;
+	srlz.i   // guarantee that interruption collection is on
+	;;
+	//(p15) ssm psr.i               // restore psr.i
+	addl r14=@gprel(ia64_leave_hypervisor),gp
+	;;
+	KVM_SAVE_REST
+	mov rp=r14
+	;;
+	adds out2=16,r12
+	br.call.sptk.many b6=kvm_page_fault
 END(kvm_itlb_miss_dispatch)
 
 ENTRY(kvm_dispatch_reflection)
-    /*
-     * Input:
-     *  psr.ic: off
-     *  r19:    intr type (offset into ivt, see ia64_int.h)
-     *  r31:    contains saved predicates (pr)
-     */
-    KVM_SAVE_MIN_WITH_COVER_R19
-    alloc r14=ar.pfs,0,0,5,0
-    mov out0=cr.ifa
-    mov out1=cr.isr
-    mov out2=cr.iim
-    mov out3=r15
-    adds r3=8,r2                // set up second base pointer
-    ;;
-    ssm psr.ic
-    ;;
-    srlz.i                  // guarantee that interruption collection is on
-    ;;
-    //(p15) ssm psr.i               // restore psr.i
-    addl r14=@gprel(ia64_leave_hypervisor),gp
-    ;;
-    KVM_SAVE_REST
-    mov rp=r14
-    ;;
-    adds out4=16,r12
-    br.call.sptk.many b6=reflect_interruption
+/*
+ * Input:
+ *  psr.ic: off
+ *  r19:    intr type (offset into ivt, see ia64_int.h)
+ *  r31:    contains saved predicates (pr)
+ */
+	KVM_SAVE_MIN_WITH_COVER_R19
+	alloc r14=ar.pfs,0,0,5,0
+	mov out0=cr.ifa
+	mov out1=cr.isr
+	mov out2=cr.iim
+	mov out3=r15
+	adds r3=8,r2                // set up second base pointer
+	;;
+	ssm psr.ic
+	;;
+	srlz.i   // guarantee that interruption collection is on
+	;;
+	//(p15) ssm psr.i               // restore psr.i
+	addl r14=@gprel(ia64_leave_hypervisor),gp
+	;;
+	KVM_SAVE_REST
+	mov rp=r14
+	;;
+	adds out4=16,r12
+	br.call.sptk.many b6=reflect_interruption
 END(kvm_dispatch_reflection)
 
 ENTRY(kvm_dispatch_virtualization_fault)
-    adds r16 = VMM_VCPU_CAUSE_OFFSET,r21
-    adds r17 = VMM_VCPU_OPCODE_OFFSET,r21
-    ;;
-    st8 [r16] = r24
-    st8 [r17] = r25
-    ;;
-    KVM_SAVE_MIN_WITH_COVER_R19
-    ;;
-    alloc r14=ar.pfs,0,0,2,0 // now it's safe (must be first in insn group!)
-    mov out0=r13        //vcpu
-    adds r3=8,r2                // set up second base pointer
-    ;;
-    ssm psr.ic
-    ;;
-    srlz.i                  // guarantee that interruption collection is on
-    ;;
-    //(p15) ssm psr.i               // restore psr.i
-    addl r14=@gprel(ia64_leave_hypervisor_prepare),gp
-    ;;
-    KVM_SAVE_REST
-    KVM_SAVE_EXTRA
-    mov rp=r14
-    ;;
-    adds out1=16,sp         //regs
-    br.call.sptk.many b6=kvm_emulate
+	adds r16 = VMM_VCPU_CAUSE_OFFSET,r21
+	adds r17 = VMM_VCPU_OPCODE_OFFSET,r21
+	;;
+	st8 [r16] = r24
+	st8 [r17] = r25
+	;;
+	KVM_SAVE_MIN_WITH_COVER_R19
+	;;
+	alloc r14=ar.pfs,0,0,2,0 // (must be first in insn group!)
+	mov out0=r13        //vcpu
+	adds r3=8,r2                // set up second base pointer
+	;;
+	ssm psr.ic
+	;;
+	srlz.i    // guarantee that interruption collection is on
+	;;
+	//(p15) ssm psr.i               // restore psr.i
+	addl r14=@gprel(ia64_leave_hypervisor_prepare),gp
+	;;
+	KVM_SAVE_REST
+	KVM_SAVE_EXTRA
+	mov rp=r14
+	;;
+	adds out1=16,sp         //regs
+	br.call.sptk.many b6=kvm_emulate
 END(kvm_dispatch_virtualization_fault)
 
 
 ENTRY(kvm_dispatch_interrupt)
-    KVM_SAVE_MIN_WITH_COVER_R19	// uses r31; defines r2 and r3
-    ;;
-    alloc r14=ar.pfs,0,0,1,0 // must be first in an insn group
-    //mov out0=cr.ivr		// pass cr.ivr as first arg
-    adds r3=8,r2		// set up second base pointer for SAVE_REST
-    ;;
-    ssm psr.ic
-    ;;
-    srlz.i
-    ;;
-    //(p15) ssm psr.i
-    addl r14=@gprel(ia64_leave_hypervisor),gp
-    ;;
-    KVM_SAVE_REST
-    mov rp=r14
-    ;;
-    mov out0=r13		// pass pointer to pt_regs as second arg
-    br.call.sptk.many b6=kvm_ia64_handle_irq
+	KVM_SAVE_MIN_WITH_COVER_R19	// uses r31; defines r2 and r3
+	;;
+	alloc r14=ar.pfs,0,0,1,0 // must be first in an insn group
+	adds r3=8,r2		// set up second base pointer for SAVE_REST
+	;;
+	ssm psr.ic
+	;;
+	srlz.i
+	;;
+	//(p15) ssm psr.i
+	addl r14=@gprel(ia64_leave_hypervisor),gp
+	;;
+	KVM_SAVE_REST
+	mov rp=r14
+	;;
+	mov out0=r13		// pass pointer to pt_regs as second arg
+	br.call.sptk.many b6=kvm_ia64_handle_irq
 END(kvm_dispatch_interrupt)
 
-
-
-
 GLOBAL_ENTRY(ia64_leave_nested)
 	rsm psr.i
 	;;
@@ -1008,7 +1008,7 @@
 	;;
 	ldf.fill f11=[r2]
 //	mov r18=r13
-//    mov r21=r13
+//	mov r21=r13
 	adds r16=PT(CR_IPSR)+16,r12
 	adds r17=PT(CR_IIP)+16,r12
 	;;
@@ -1058,138 +1058,135 @@
 	rfi
 END(ia64_leave_nested)
 
-
-
 GLOBAL_ENTRY(ia64_leave_hypervisor_prepare)
-    /*
-     * work.need_resched etc. mustn't get changed
-     *by this CPU before it returns to
-    ;;
-     * user- or fsys-mode, hence we disable interrupts early on:
-     */
-    adds r2 = PT(R4)+16,r12
-    adds r3 = PT(R5)+16,r12
-    adds r8 = PT(EML_UNAT)+16,r12
-    ;;
-    ld8 r8 = [r8]
-    ;;
-    mov ar.unat=r8
-    ;;
-    ld8.fill r4=[r2],16    //load r4
-    ld8.fill r5=[r3],16    //load r5
-    ;;
-    ld8.fill r6=[r2]    //load r6
-    ld8.fill r7=[r3]    //load r7
-    ;;
+/*
+ * work.need_resched etc. mustn't get changed
+ *by this CPU before it returns to
+ * user- or fsys-mode, hence we disable interrupts early on:
+ */
+	adds r2 = PT(R4)+16,r12
+	adds r3 = PT(R5)+16,r12
+	adds r8 = PT(EML_UNAT)+16,r12
+	;;
+	ld8 r8 = [r8]
+	;;
+	mov ar.unat=r8
+	;;
+	ld8.fill r4=[r2],16    //load r4
+	ld8.fill r5=[r3],16    //load r5
+	;;
+	ld8.fill r6=[r2]    //load r6
+	ld8.fill r7=[r3]    //load r7
+	;;
 END(ia64_leave_hypervisor_prepare)
 //fall through
 GLOBAL_ENTRY(ia64_leave_hypervisor)
-    rsm psr.i
-    ;;
-    br.call.sptk.many b0=leave_hypervisor_tail
-    ;;
-    adds r20=PT(PR)+16,r12
-    adds r8=PT(EML_UNAT)+16,r12
-    ;;
-    ld8 r8=[r8]
-    ;;
-    mov ar.unat=r8
-    ;;
-    lfetch [r20],PT(CR_IPSR)-PT(PR)
-    adds r2 = PT(B6)+16,r12
-    adds r3 = PT(B7)+16,r12
-    ;;
-    lfetch [r20]
-    ;;
-    ld8 r24=[r2],16        /* B6 */
-    ld8 r25=[r3],16        /* B7 */
-    ;;
-    ld8 r26=[r2],16        /* ar_csd */
-    ld8 r27=[r3],16        /* ar_ssd */
-    mov b6 = r24
-    ;;
-    ld8.fill r8=[r2],16
-    ld8.fill r9=[r3],16
-    mov b7 = r25
-    ;;
-    mov ar.csd = r26
-    mov ar.ssd = r27
-    ;;
-    ld8.fill r10=[r2],PT(R15)-PT(R10)
-    ld8.fill r11=[r3],PT(R14)-PT(R11)
-    ;;
-    ld8.fill r15=[r2],PT(R16)-PT(R15)
-    ld8.fill r14=[r3],PT(R17)-PT(R14)
-    ;;
-    ld8.fill r16=[r2],16
-    ld8.fill r17=[r3],16
-    ;;
-    ld8.fill r18=[r2],16
-    ld8.fill r19=[r3],16
-    ;;
-    ld8.fill r20=[r2],16
-    ld8.fill r21=[r3],16
-    ;;
-    ld8.fill r22=[r2],16
-    ld8.fill r23=[r3],16
-    ;;
-    ld8.fill r24=[r2],16
-    ld8.fill r25=[r3],16
-    ;;
-    ld8.fill r26=[r2],16
-    ld8.fill r27=[r3],16
-    ;;
-    ld8.fill r28=[r2],16
-    ld8.fill r29=[r3],16
-    ;;
-    ld8.fill r30=[r2],PT(F6)-PT(R30)
-    ld8.fill r31=[r3],PT(F7)-PT(R31)
-    ;;
-    rsm psr.i | psr.ic
-    // initiate turning off of interrupt and interruption collection
-    invala          // invalidate ALAT
-    ;;
-    srlz.i          // ensure interruption collection is off
-    ;;
-    bsw.0
-    ;;
-    adds r16 = PT(CR_IPSR)+16,r12
-    adds r17 = PT(CR_IIP)+16,r12
-    mov r21=r13		// get current
-    ;;
-    ld8 r31=[r16],16    // load cr.ipsr
-    ld8 r30=[r17],16    // load cr.iip
-    ;;
-    ld8 r29=[r16],16    // load cr.ifs
-    ld8 r28=[r17],16    // load ar.unat
-    ;;
-    ld8 r27=[r16],16    // load ar.pfs
-    ld8 r26=[r17],16    // load ar.rsc
-    ;;
-    ld8 r25=[r16],16    // load ar.rnat
-    ld8 r24=[r17],16    // load ar.bspstore
-    ;;
-    ld8 r23=[r16],16    // load predicates
-    ld8 r22=[r17],16    // load b0
-    ;;
-    ld8 r20=[r16],16    // load ar.rsc value for "loadrs"
-    ld8.fill r1=[r17],16    //load r1
-    ;;
-    ld8.fill r12=[r16],16    //load r12
-    ld8.fill r13=[r17],PT(R2)-PT(R13)    //load r13
-    ;;
-    ld8 r19=[r16],PT(R3)-PT(AR_FPSR)    //load ar_fpsr
-    ld8.fill r2=[r17],PT(AR_CCV)-PT(R2)    //load r2
-    ;;
-    ld8.fill r3=[r16]	//load r3
-    ld8 r18=[r17]	//load ar_ccv
-    ;;
-    mov ar.fpsr=r19
-    mov ar.ccv=r18
-    shr.u r18=r20,16
-    ;;
+	rsm psr.i
+	;;
+	br.call.sptk.many b0=leave_hypervisor_tail
+	;;
+	adds r20=PT(PR)+16,r12
+	adds r8=PT(EML_UNAT)+16,r12
+	;;
+	ld8 r8=[r8]
+	;;
+	mov ar.unat=r8
+	;;
+	lfetch [r20],PT(CR_IPSR)-PT(PR)
+	adds r2 = PT(B6)+16,r12
+	adds r3 = PT(B7)+16,r12
+	;;
+	lfetch [r20]
+	;;
+	ld8 r24=[r2],16        /* B6 */
+	ld8 r25=[r3],16        /* B7 */
+	;;
+	ld8 r26=[r2],16        /* ar_csd */
+	ld8 r27=[r3],16        /* ar_ssd */
+	mov b6 = r24
+	;;
+	ld8.fill r8=[r2],16
+	ld8.fill r9=[r3],16
+	mov b7 = r25
+	;;
+	mov ar.csd = r26
+	mov ar.ssd = r27
+	;;
+	ld8.fill r10=[r2],PT(R15)-PT(R10)
+	ld8.fill r11=[r3],PT(R14)-PT(R11)
+	;;
+	ld8.fill r15=[r2],PT(R16)-PT(R15)
+	ld8.fill r14=[r3],PT(R17)-PT(R14)
+	;;
+	ld8.fill r16=[r2],16
+	ld8.fill r17=[r3],16
+	;;
+	ld8.fill r18=[r2],16
+	ld8.fill r19=[r3],16
+	;;
+	ld8.fill r20=[r2],16
+	ld8.fill r21=[r3],16
+	;;
+	ld8.fill r22=[r2],16
+	ld8.fill r23=[r3],16
+	;;
+	ld8.fill r24=[r2],16
+	ld8.fill r25=[r3],16
+	;;
+	ld8.fill r26=[r2],16
+	ld8.fill r27=[r3],16
+	;;
+	ld8.fill r28=[r2],16
+	ld8.fill r29=[r3],16
+	;;
+	ld8.fill r30=[r2],PT(F6)-PT(R30)
+	ld8.fill r31=[r3],PT(F7)-PT(R31)
+	;;
+	rsm psr.i | psr.ic
+	// initiate turning off of interrupt and interruption collection
+	invala          // invalidate ALAT
+	;;
+	srlz.i          // ensure interruption collection is off
+	;;
+	bsw.0
+	;;
+	adds r16 = PT(CR_IPSR)+16,r12
+	adds r17 = PT(CR_IIP)+16,r12
+	mov r21=r13		// get current
+	;;
+	ld8 r31=[r16],16    // load cr.ipsr
+	ld8 r30=[r17],16    // load cr.iip
+	;;
+	ld8 r29=[r16],16    // load cr.ifs
+	ld8 r28=[r17],16    // load ar.unat
+	;;
+	ld8 r27=[r16],16    // load ar.pfs
+	ld8 r26=[r17],16    // load ar.rsc
+	;;
+	ld8 r25=[r16],16    // load ar.rnat
+	ld8 r24=[r17],16    // load ar.bspstore
+	;;
+	ld8 r23=[r16],16    // load predicates
+	ld8 r22=[r17],16    // load b0
+	;;
+	ld8 r20=[r16],16    // load ar.rsc value for "loadrs"
+	ld8.fill r1=[r17],16    //load r1
+	;;
+	ld8.fill r12=[r16],16    //load r12
+	ld8.fill r13=[r17],PT(R2)-PT(R13)    //load r13
+	;;
+	ld8 r19=[r16],PT(R3)-PT(AR_FPSR)    //load ar_fpsr
+	ld8.fill r2=[r17],PT(AR_CCV)-PT(R2)    //load r2
+	;;
+	ld8.fill r3=[r16]	//load r3
+	ld8 r18=[r17]	//load ar_ccv
+	;;
+	mov ar.fpsr=r19
+	mov ar.ccv=r18
+	shr.u r18=r20,16
+	;;
 kvm_rbs_switch:
-    mov r19=96
+	mov r19=96
 
 kvm_dont_preserve_current_frame:
 /*
@@ -1201,76 +1198,76 @@
 #   define pReturn	p7
 #   define Nregs	14
 
-    alloc loc0=ar.pfs,2,Nregs-2,2,0
-    shr.u loc1=r18,9		// RNaTslots <= floor(dirtySize / (64*8))
-    sub r19=r19,r18		// r19 = (physStackedSize + 8) - dirtySize
-    ;;
-    mov ar.rsc=r20		// load ar.rsc to be used for "loadrs"
-    shladd in0=loc1,3,r19
-    mov in1=0
-    ;;
-    TEXT_ALIGN(32)
+	alloc loc0=ar.pfs,2,Nregs-2,2,0
+	shr.u loc1=r18,9	// RNaTslots <= floor(dirtySize / (64*8))
+	sub r19=r19,r18		// r19 = (physStackedSize + 8) - dirtySize
+	;;
+	mov ar.rsc=r20		// load ar.rsc to be used for "loadrs"
+	shladd in0=loc1,3,r19
+	mov in1=0
+	;;
+	TEXT_ALIGN(32)
 kvm_rse_clear_invalid:
-    alloc loc0=ar.pfs,2,Nregs-2,2,0
-    cmp.lt pRecurse,p0=Nregs*8,in0
-    // if more than Nregs regs left to clear, (re)curse
-    add out0=-Nregs*8,in0
-    add out1=1,in1		// increment recursion count
-    mov loc1=0
-    mov loc2=0
-    ;;
-    mov loc3=0
-    mov loc4=0
-    mov loc5=0
-    mov loc6=0
-    mov loc7=0
+	alloc loc0=ar.pfs,2,Nregs-2,2,0
+	cmp.lt pRecurse,p0=Nregs*8,in0
+	// if more than Nregs regs left to clear, (re)curse
+	add out0=-Nregs*8,in0
+	add out1=1,in1		// increment recursion count
+	mov loc1=0
+	mov loc2=0
+	;;
+	mov loc3=0
+	mov loc4=0
+	mov loc5=0
+	mov loc6=0
+	mov loc7=0
 (pRecurse) br.call.dptk.few b0=kvm_rse_clear_invalid
-    ;;
-    mov loc8=0
-    mov loc9=0
-    cmp.ne pReturn,p0=r0,in1
-    // if recursion count != 0, we need to do a br.ret
-    mov loc10=0
-    mov loc11=0
+	;;
+	mov loc8=0
+	mov loc9=0
+	cmp.ne pReturn,p0=r0,in1
+	// if recursion count != 0, we need to do a br.ret
+	mov loc10=0
+	mov loc11=0
 (pReturn) br.ret.dptk.many b0
 
 #	undef pRecurse
 #	undef pReturn
 
 // loadrs has already been shifted
-    alloc r16=ar.pfs,0,0,0,0    // drop current register frame
-    ;;
-    loadrs
-    ;;
-    mov ar.bspstore=r24
-    ;;
-    mov ar.unat=r28
-    mov ar.rnat=r25
-    mov ar.rsc=r26
-    ;;
-    mov cr.ipsr=r31
-    mov cr.iip=r30
-    mov cr.ifs=r29
-    mov ar.pfs=r27
-    adds r18=VMM_VPD_BASE_OFFSET,r21
-    ;;
-    ld8 r18=[r18]   //vpd
-    adds r17=VMM_VCPU_ISR_OFFSET,r21
-    ;;
-    ld8 r17=[r17]
-    adds r19=VMM_VPD_VPSR_OFFSET,r18
-    ;;
-    ld8 r19=[r19]        //vpsr
-    mov r25=r18
-    adds r16= VMM_VCPU_GP_OFFSET,r21
-    ;;
-    ld8 r16= [r16] // Put gp in r24
-    movl r24=@gprel(ia64_vmm_entry)  // calculate return address
-    ;;
-    add  r24=r24,r16
-    ;;
-    br.sptk.many  kvm_vps_sync_write       // call the service
-    ;;
+	alloc r16=ar.pfs,0,0,0,0    // drop current register frame
+	;;
+	loadrs
+	;;
+	mov ar.bspstore=r24
+	;;
+	mov ar.unat=r28
+	mov ar.rnat=r25
+	mov ar.rsc=r26
+	;;
+	mov cr.ipsr=r31
+	mov cr.iip=r30
+	mov cr.ifs=r29
+	mov ar.pfs=r27
+	adds r18=VMM_VPD_BASE_OFFSET,r21
+	;;
+	ld8 r18=[r18]   //vpd
+	adds r17=VMM_VCPU_ISR_OFFSET,r21
+	;;
+	ld8 r17=[r17]
+	adds r19=VMM_VPD_VPSR_OFFSET,r18
+	;;
+	ld8 r19=[r19]        //vpsr
+	mov r25=r18
+	adds r16= VMM_VCPU_GP_OFFSET,r21
+	;;
+	ld8 r16= [r16] // Put gp in r24
+	movl r24=@gprel(ia64_vmm_entry)  // calculate return address
+	;;
+	add  r24=r24,r16
+	;;
+	br.sptk.many  kvm_vps_sync_write       // call the service
+	;;
 END(ia64_leave_hypervisor)
 // fall through
 GLOBAL_ENTRY(ia64_vmm_entry)
@@ -1283,16 +1280,14 @@
  *  r22:b0
  *  r23:predicate
  */
-    mov r24=r22
-    mov r25=r18
-    tbit.nz p1,p2 = r19,IA64_PSR_IC_BIT        // p1=vpsr.ic
-    (p1) br.cond.sptk.few kvm_vps_resume_normal
-    (p2) br.cond.sptk.many kvm_vps_resume_handler
-    ;;
+	mov r24=r22
+	mov r25=r18
+	tbit.nz p1,p2 = r19,IA64_PSR_IC_BIT        // p1=vpsr.ic
+(p1) 	br.cond.sptk.few kvm_vps_resume_normal
+(p2)	br.cond.sptk.many kvm_vps_resume_handler
+	;;
 END(ia64_vmm_entry)
 
-
-
 /*
  * extern u64 ia64_call_vsa(u64 proc, u64 arg1, u64 arg2,
  *                  u64 arg3, u64 arg4, u64 arg5,
@@ -1310,88 +1305,88 @@
 entry   =   loc3
 hostret =   r24
 
-    alloc   pfssave=ar.pfs,4,4,0,0
-    mov rpsave=rp
-    adds entry=VMM_VCPU_VSA_BASE_OFFSET, r13
-    ;;
-    ld8 entry=[entry]
-1:  mov hostret=ip
-    mov r25=in1         // copy arguments
-    mov r26=in2
-    mov r27=in3
-    mov psrsave=psr
-    ;;
-    tbit.nz p6,p0=psrsave,14    // IA64_PSR_I
-    tbit.nz p7,p0=psrsave,13    // IA64_PSR_IC
-    ;;
-    add hostret=2f-1b,hostret   // calculate return address
-    add entry=entry,in0
-    ;;
-    rsm psr.i | psr.ic
-    ;;
-    srlz.i
-    mov b6=entry
-    br.cond.sptk b6         // call the service
+	alloc   pfssave=ar.pfs,4,4,0,0
+	mov rpsave=rp
+	adds entry=VMM_VCPU_VSA_BASE_OFFSET, r13
+	;;
+	ld8 entry=[entry]
+1:	mov hostret=ip
+	mov r25=in1         // copy arguments
+	mov r26=in2
+	mov r27=in3
+	mov psrsave=psr
+	;;
+	tbit.nz p6,p0=psrsave,14    // IA64_PSR_I
+	tbit.nz p7,p0=psrsave,13    // IA64_PSR_IC
+	;;
+	add hostret=2f-1b,hostret   // calculate return address
+	add entry=entry,in0
+	;;
+	rsm psr.i | psr.ic
+	;;
+	srlz.i
+	mov b6=entry
+	br.cond.sptk b6         // call the service
 2:
-    // Architectural sequence for enabling interrupts if necessary
+// Architectural sequence for enabling interrupts if necessary
 (p7)    ssm psr.ic
-    ;;
+	;;
 (p7)    srlz.i
-    ;;
+	;;
 //(p6)    ssm psr.i
-    ;;
-    mov rp=rpsave
-    mov ar.pfs=pfssave
-    mov r8=r31
-    ;;
-    srlz.d
-    br.ret.sptk rp
+	;;
+	mov rp=rpsave
+	mov ar.pfs=pfssave
+	mov r8=r31
+	;;
+	srlz.d
+	br.ret.sptk rp
 
 END(ia64_call_vsa)
 
 #define  INIT_BSPSTORE  ((4<<30)-(12<<20)-0x100)
 
 GLOBAL_ENTRY(vmm_reset_entry)
-    //set up ipsr, iip, vpd.vpsr, dcr
-    // For IPSR: it/dt/rt=1, i/ic=1, si=1, vm/bn=1
-    // For DCR: all bits 0
-    bsw.0
-    ;;
-    mov r21 =r13
-    adds r14=-VMM_PT_REGS_SIZE, r12
-    ;;
-    movl r6=0x501008826000      // IPSR dt/rt/it:1;i/ic:1, si:1, vm/bn:1
-    movl r10=0x8000000000000000
-    adds r16=PT(CR_IIP), r14
-    adds r20=PT(R1), r14
-    ;;
-    rsm psr.ic | psr.i
-    ;;
-    srlz.i
-    ;;
-    mov ar.rsc = 0
-    ;;
-    flushrs
-    ;;
-    mov ar.bspstore = 0
-    // clear BSPSTORE
-    ;;
-    mov cr.ipsr=r6
-    mov cr.ifs=r10
-    ld8 r4 = [r16] // Set init iip for first run.
-    ld8 r1 = [r20]
-    ;;
-    mov cr.iip=r4
-    adds r16=VMM_VPD_BASE_OFFSET,r13
-    ;;
-    ld8 r18=[r16]
-    ;;
-    adds r19=VMM_VPD_VPSR_OFFSET,r18
-    ;;
-    ld8 r19=[r19]
-    mov r17=r0
-    mov r22=r0
-    mov r23=r0
-    br.cond.sptk ia64_vmm_entry
-    br.ret.sptk  b0
+	//set up ipsr, iip, vpd.vpsr, dcr
+	// For IPSR: it/dt/rt=1, i/ic=1, si=1, vm/bn=1
+	// For DCR: all bits 0
+	bsw.0
+	;;
+	mov r21 =r13
+	adds r14=-VMM_PT_REGS_SIZE, r12
+	;;
+	movl r6=0x501008826000      // IPSR dt/rt/it:1;i/ic:1, si:1, vm/bn:1
+	movl r10=0x8000000000000000
+	adds r16=PT(CR_IIP), r14
+	adds r20=PT(R1), r14
+	;;
+	rsm psr.ic | psr.i
+	;;
+	srlz.i
+	;;
+	mov ar.rsc = 0
+	;;
+	flushrs
+	;;
+	mov ar.bspstore = 0
+	// clear BSPSTORE
+	;;
+	mov cr.ipsr=r6
+	mov cr.ifs=r10
+	ld8 r4 = [r16] // Set init iip for first run.
+	ld8 r1 = [r20]
+	;;
+	mov cr.iip=r4
+	adds r16=VMM_VPD_BASE_OFFSET,r13
+	;;
+	ld8 r18=[r16]
+	;;
+	adds r19=VMM_VPD_VPSR_OFFSET,r18
+	;;
+	ld8 r19=[r19]
+	mov r17=r0
+	mov r22=r0
+	mov r23=r0
+	br.cond.sptk ia64_vmm_entry
+	br.ret.sptk  b0
 END(vmm_reset_entry)
diff --git a/arch/ia64/kvm/vtlb.c b/arch/ia64/kvm/vtlb.c
index e22b933..6b6307a 100644
--- a/arch/ia64/kvm/vtlb.c
+++ b/arch/ia64/kvm/vtlb.c
@@ -183,8 +183,8 @@
 	u64 i, dirty_pages = 1;
 	u64 base_gfn = (pte&_PAGE_PPN_MASK) >> PAGE_SHIFT;
 	spinlock_t *lock = __kvm_va(v->arch.dirty_log_lock_pa);
-	void *dirty_bitmap = (void *)v - (KVM_VCPU_OFS + v->vcpu_id * VCPU_SIZE)
-						+ KVM_MEM_DIRTY_LOG_OFS;
+	void *dirty_bitmap = (void *)KVM_MEM_DIRTY_LOG_BASE;
+
 	dirty_pages <<= ps <= PAGE_SHIFT ? 0 : ps - PAGE_SHIFT;
 
 	vmm_spin_lock(lock);
diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c
index 0c66dbd..66fd705 100644
--- a/arch/ia64/sn/kernel/irq.c
+++ b/arch/ia64/sn/kernel/irq.c
@@ -227,14 +227,14 @@
 	return new_irq_info;
 }
 
-static void sn_set_affinity_irq(unsigned int irq, cpumask_t mask)
+static void sn_set_affinity_irq(unsigned int irq, const struct cpumask *mask)
 {
 	struct sn_irq_info *sn_irq_info, *sn_irq_info_safe;
 	nasid_t nasid;
 	int slice;
 
-	nasid = cpuid_to_nasid(first_cpu(mask));
-	slice = cpuid_to_slice(first_cpu(mask));
+	nasid = cpuid_to_nasid(cpumask_first(mask));
+	slice = cpuid_to_slice(cpumask_first(mask));
 
 	list_for_each_entry_safe(sn_irq_info, sn_irq_info_safe,
 				 sn_irq_lh[irq], list)
diff --git a/arch/ia64/sn/kernel/msi_sn.c b/arch/ia64/sn/kernel/msi_sn.c
index 83f190f..ca553b0 100644
--- a/arch/ia64/sn/kernel/msi_sn.c
+++ b/arch/ia64/sn/kernel/msi_sn.c
@@ -151,7 +151,8 @@
 }
 
 #ifdef CONFIG_SMP
-static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
+static void sn_set_msi_irq_affinity(unsigned int irq,
+				    const struct cpumask *cpu_mask)
 {
 	struct msi_msg msg;
 	int slice;
@@ -164,7 +165,7 @@
 	struct sn_pcibus_provider *provider;
 	unsigned int cpu;
 
-	cpu = first_cpu(cpu_mask);
+	cpu = cpumask_first(cpu_mask);
 	sn_irq_info = sn_msi_info[irq].sn_irq_info;
 	if (sn_irq_info == NULL || sn_irq_info->irq_int_bit >= 0)
 		return;
@@ -204,7 +205,7 @@
 	msg.address_lo = (u32)(bus_addr & 0x00000000ffffffff);
 
 	write_msi_msg(irq, &msg);
-	irq_desc[irq].affinity = cpu_mask;
+	irq_desc[irq].affinity = *cpu_mask;
 }
 #endif /* CONFIG_SMP */
 
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index 29047d5..cabba33 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -10,6 +10,7 @@
 	default y
 	select HAVE_IDE
 	select HAVE_OPROFILE
+	select INIT_ALL_POSSIBLE
 
 config SBUS
 	bool
diff --git a/arch/m32r/kernel/init_task.c b/arch/m32r/kernel/init_task.c
index 0d658db..016885c 100644
--- a/arch/m32r/kernel/init_task.c
+++ b/arch/m32r/kernel/init_task.c
@@ -11,7 +11,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/m32r/kernel/smpboot.c b/arch/m32r/kernel/smpboot.c
index 39cb6da..0f06b37 100644
--- a/arch/m32r/kernel/smpboot.c
+++ b/arch/m32r/kernel/smpboot.c
@@ -73,17 +73,11 @@
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map;
 
-/* Bitmask of currently online CPUs */
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
-
 cpumask_t cpu_bootout_map;
 cpumask_t cpu_bootin_map;
 static cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
 EXPORT_SYMBOL(cpu_callout_map);
-cpumask_t cpu_possible_map = CPU_MASK_ALL;
-EXPORT_SYMBOL(cpu_possible_map);
 
 /* Per CPU bogomips and other parameters */
 struct cpuinfo_m32r cpu_data[NR_CPUS] __cacheline_aligned;
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index 836fb66..c825bde 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -280,7 +280,6 @@
 
 config MMU_MOTOROLA
 	bool
-	depends on MMU && !MMU_SUN3
 
 config MMU_SUN3
 	bool
diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c
index 3042c2b..632ce01 100644
--- a/arch/m68k/kernel/process.c
+++ b/arch/m68k/kernel/process.c
@@ -40,7 +40,6 @@
  * alignment requirements and potentially different initial
  * setup.
  */
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/m68knommu/kernel/init_task.c b/arch/m68knommu/kernel/init_task.c
index 344c01a..fe282de 100644
--- a/arch/m68knommu/kernel/init_task.c
+++ b/arch/m68knommu/kernel/init_task.c
@@ -12,7 +12,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/m68knommu/platform/coldfire/pit.c b/arch/m68knommu/platform/coldfire/pit.c
index c5b9167..2a12e7f 100644
--- a/arch/m68knommu/platform/coldfire/pit.c
+++ b/arch/m68knommu/platform/coldfire/pit.c
@@ -156,7 +156,7 @@
 {
 	u32 imr;
 
-	cf_pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	cf_pit_clockevent.cpumask = cpumask_of(smp_processor_id());
 	cf_pit_clockevent.mult = div_sc(FREQ, NSEC_PER_SEC, 32);
 	cf_pit_clockevent.max_delta_ns =
 		clockevent_delta2ns(0xFFFF, &cf_pit_clockevent);
diff --git a/arch/mips/include/asm/irq.h b/arch/mips/include/asm/irq.h
index a58f0ee..abc62aa 100644
--- a/arch/mips/include/asm/irq.h
+++ b/arch/mips/include/asm/irq.h
@@ -49,7 +49,8 @@
 #ifdef CONFIG_MIPS_MT_SMTC_IRQAFF
 #include <linux/cpumask.h>
 
-extern void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity);
+extern void plat_set_irq_affinity(unsigned int irq,
+				  const struct cpumask *affinity);
 extern void smtc_forward_irq(unsigned int irq);
 
 /*
diff --git a/arch/mips/include/asm/mach-ip27/topology.h b/arch/mips/include/asm/mach-ip27/topology.h
index 7785bec..1fb959f 100644
--- a/arch/mips/include/asm/mach-ip27/topology.h
+++ b/arch/mips/include/asm/mach-ip27/topology.h
@@ -37,7 +37,6 @@
 
 /* sched_domains SD_NODE_INIT for SGI IP27 machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
-	.span			= CPU_MASK_NONE,	\
 	.parent			= NULL,			\
 	.child			= NULL,			\
 	.groups			= NULL,			\
diff --git a/arch/mips/include/asm/smp.h b/arch/mips/include/asm/smp.h
index 0ff5b52..86557b5 100644
--- a/arch/mips/include/asm/smp.h
+++ b/arch/mips/include/asm/smp.h
@@ -38,9 +38,6 @@
 #define SMP_RESCHEDULE_YOURSELF	0x1	/* XXX braindead */
 #define SMP_CALL_FUNCTION	0x2
 
-extern cpumask_t phys_cpu_present_map;
-#define cpu_possible_map	phys_cpu_present_map
-
 extern void asmlinkage smp_bootstrap(void);
 
 /*
diff --git a/arch/mips/jazz/irq.c b/arch/mips/jazz/irq.c
index d7f8a78..03965cb 100644
--- a/arch/mips/jazz/irq.c
+++ b/arch/mips/jazz/irq.c
@@ -146,7 +146,7 @@
 
 	BUG_ON(HZ != 100);
 
-	cd->cpumask             = cpumask_of_cpu(cpu);
+	cd->cpumask             = cpumask_of(cpu);
 	clockevents_register_device(cd);
 	action->dev_id = cd;
 	setup_irq(JAZZ_TIMER_IRQ, action);
diff --git a/arch/mips/kernel/cevt-bcm1480.c b/arch/mips/kernel/cevt-bcm1480.c
index 0a57f86..b820661 100644
--- a/arch/mips/kernel/cevt-bcm1480.c
+++ b/arch/mips/kernel/cevt-bcm1480.c
@@ -126,7 +126,7 @@
 	cd->min_delta_ns	= clockevent_delta2ns(2, cd);
 	cd->rating		= 200;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= sibyte_next_event;
 	cd->set_mode		= sibyte_set_mode;
 	clockevents_register_device(cd);
@@ -148,6 +148,6 @@
 	action->name	= name;
 	action->dev_id	= cd;
 
-	irq_set_affinity(irq, cpumask_of_cpu(cpu));
+	irq_set_affinity(irq, cpumask_of(cpu));
 	setup_irq(irq, action);
 }
diff --git a/arch/mips/kernel/cevt-ds1287.c b/arch/mips/kernel/cevt-ds1287.c
index df4acb6..1ada45e 100644
--- a/arch/mips/kernel/cevt-ds1287.c
+++ b/arch/mips/kernel/cevt-ds1287.c
@@ -88,7 +88,6 @@
 static struct clock_event_device ds1287_clockevent = {
 	.name		= "ds1287",
 	.features	= CLOCK_EVT_FEAT_PERIODIC,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= ds1287_set_next_event,
 	.set_mode	= ds1287_set_mode,
 	.event_handler	= ds1287_event_handler,
@@ -122,6 +121,7 @@
 	clockevent_set_clock(cd, 32768);
 	cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
 	cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+	cd->cpumask = cpumask_of(0);
 
 	clockevents_register_device(&ds1287_clockevent);
 
diff --git a/arch/mips/kernel/cevt-gt641xx.c b/arch/mips/kernel/cevt-gt641xx.c
index 6e2f585..e9b787f 100644
--- a/arch/mips/kernel/cevt-gt641xx.c
+++ b/arch/mips/kernel/cevt-gt641xx.c
@@ -96,7 +96,6 @@
 static struct clock_event_device gt641xx_timer0_clockevent = {
 	.name		= "gt641xx-timer0",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
-	.cpumask	= CPU_MASK_CPU0,
 	.irq		= GT641XX_TIMER0_IRQ,
 	.set_next_event	= gt641xx_timer0_set_next_event,
 	.set_mode	= gt641xx_timer0_set_mode,
@@ -132,6 +131,7 @@
 	clockevent_set_clock(cd, gt641xx_base_clock);
 	cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
 	cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+	cd->cpumask = cpumask_of(0);
 
 	clockevents_register_device(&gt641xx_timer0_clockevent);
 
diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
index 4a4c59f..e1ec83b 100644
--- a/arch/mips/kernel/cevt-r4k.c
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -195,7 +195,7 @@
 
 	cd->rating		= 300;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= mips_next_event;
 	cd->set_mode		= mips_set_clock_mode;
 	cd->event_handler	= mips_event_handler;
diff --git a/arch/mips/kernel/cevt-sb1250.c b/arch/mips/kernel/cevt-sb1250.c
index 63ac3ad..a2eebaa 100644
--- a/arch/mips/kernel/cevt-sb1250.c
+++ b/arch/mips/kernel/cevt-sb1250.c
@@ -125,7 +125,7 @@
 	cd->min_delta_ns	= clockevent_delta2ns(2, cd);
 	cd->rating		= 200;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= sibyte_next_event;
 	cd->set_mode		= sibyte_set_mode;
 	clockevents_register_device(cd);
@@ -147,6 +147,6 @@
 	action->name	= name;
 	action->dev_id	= cd;
 
-	irq_set_affinity(irq, cpumask_of_cpu(cpu));
+	irq_set_affinity(irq, cpumask_of(cpu));
 	setup_irq(irq, action);
 }
diff --git a/arch/mips/kernel/cevt-smtc.c b/arch/mips/kernel/cevt-smtc.c
index 5162fe4..6d45e24 100644
--- a/arch/mips/kernel/cevt-smtc.c
+++ b/arch/mips/kernel/cevt-smtc.c
@@ -292,7 +292,7 @@
 
 	cd->rating		= 300;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= mips_next_event;
 	cd->set_mode		= mips_set_clock_mode;
 	cd->event_handler	= mips_event_handler;
diff --git a/arch/mips/kernel/cevt-txx9.c b/arch/mips/kernel/cevt-txx9.c
index b5fc4eb..eccf7d6 100644
--- a/arch/mips/kernel/cevt-txx9.c
+++ b/arch/mips/kernel/cevt-txx9.c
@@ -112,7 +112,6 @@
 	.name		= "TXx9",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.rating		= 200,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_mode	= txx9tmr_set_mode,
 	.set_next_event	= txx9tmr_set_next_event,
 };
@@ -150,6 +149,7 @@
 		clockevent_delta2ns(0xffffffff >> (32 - TXX9_TIMER_BITS), cd);
 	cd->min_delta_ns = clockevent_delta2ns(0xf, cd);
 	cd->irq = irq;
+	cd->cpumask = cpumask_of(0),
 	clockevents_register_device(cd);
 	setup_irq(irq, &txx9tmr_irq);
 	printk(KERN_INFO "TXx9: clockevent device at 0x%lx, irq %d\n",
diff --git a/arch/mips/kernel/i8253.c b/arch/mips/kernel/i8253.c
index b6ac551..f4d1878 100644
--- a/arch/mips/kernel/i8253.c
+++ b/arch/mips/kernel/i8253.c
@@ -115,7 +115,7 @@
 	 * Start pit with the boot cpu mask and make it global after the
 	 * IO_APIC has been initialized.
 	 */
-	cd->cpumask = cpumask_of_cpu(cpu);
+	cd->cpumask = cpumask_of(cpu);
 	clockevent_set_clock(cd, CLOCK_TICK_RATE);
 	cd->max_delta_ns = clockevent_delta2ns(0x7FFF, cd);
 	cd->min_delta_ns = clockevent_delta2ns(0xF, cd);
diff --git a/arch/mips/kernel/init_task.c b/arch/mips/kernel/init_task.c
index d72487a..149cd91 100644
--- a/arch/mips/kernel/init_task.c
+++ b/arch/mips/kernel/init_task.c
@@ -9,7 +9,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/mips/kernel/irq-gic.c b/arch/mips/kernel/irq-gic.c
index f0a4bb1..494a49a 100644
--- a/arch/mips/kernel/irq-gic.c
+++ b/arch/mips/kernel/irq-gic.c
@@ -155,7 +155,7 @@
 
 static DEFINE_SPINLOCK(gic_lock);
 
-static void gic_set_affinity(unsigned int irq, cpumask_t cpumask)
+static void gic_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
 	cpumask_t	tmp = CPU_MASK_NONE;
 	unsigned long	flags;
@@ -164,7 +164,7 @@
 	pr_debug(KERN_DEBUG "%s called\n", __func__);
 	irq -= _irqbase;
 
-	cpus_and(tmp, cpumask, cpu_online_map);
+	cpumask_and(&tmp, cpumask, cpu_online_mask);
 	if (cpus_empty(tmp))
 		return;
 
@@ -187,7 +187,7 @@
 		set_bit(irq, pcpu_masks[first_cpu(tmp)].pcpu_mask);
 
 	}
-	irq_desc[irq].affinity = cpumask;
+	irq_desc[irq].affinity = *cpumask;
 	spin_unlock_irqrestore(&gic_lock, flags);
 
 }
diff --git a/arch/mips/kernel/smp-cmp.c b/arch/mips/kernel/smp-cmp.c
index ca476c4..f27beca 100644
--- a/arch/mips/kernel/smp-cmp.c
+++ b/arch/mips/kernel/smp-cmp.c
@@ -51,10 +51,10 @@
 	int len;
 
 	cpus_clear(cpu_allow_map);
-	if (cpulist_parse(str, cpu_allow_map) == 0) {
+	if (cpulist_parse(str, &cpu_allow_map) == 0) {
 		cpu_set(0, cpu_allow_map);
 		cpus_and(cpu_possible_map, cpu_possible_map, cpu_allow_map);
-		len = cpulist_scnprintf(buf, sizeof(buf)-1, cpu_possible_map);
+		len = cpulist_scnprintf(buf, sizeof(buf)-1, &cpu_possible_map);
 		buf[len] = '\0';
 		pr_debug("Allowable CPUs: %s\n", buf);
 		return 1;
@@ -226,7 +226,7 @@
 
 	for (i = 1; i < NR_CPUS; i++) {
 		if (amon_cpu_avail(i)) {
-			cpu_set(i, phys_cpu_present_map);
+			cpu_set(i, cpu_possible_map);
 			__cpu_number_map[i]	= ++ncpu;
 			__cpu_logical_map[ncpu]	= i;
 		}
diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c
index 87a1816..6f7ee5a 100644
--- a/arch/mips/kernel/smp-mt.c
+++ b/arch/mips/kernel/smp-mt.c
@@ -70,7 +70,7 @@
 		write_vpe_c0_vpeconf0(tmp);
 
 		/* Record this as available CPU */
-		cpu_set(tc, phys_cpu_present_map);
+		cpu_set(tc, cpu_possible_map);
 		__cpu_number_map[tc]	= ++ncpu;
 		__cpu_logical_map[ncpu]	= tc;
 	}
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 8bf88faf..3da9470 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -44,15 +44,10 @@
 #include <asm/mipsmtregs.h>
 #endif /* CONFIG_MIPS_MT_SMTC */
 
-cpumask_t phys_cpu_present_map;		/* Bitmask of available CPUs */
 volatile cpumask_t cpu_callin_map;	/* Bitmask of started secondaries */
-cpumask_t cpu_online_map;		/* Bitmask of currently online CPUs */
 int __cpu_number_map[NR_CPUS];		/* Map physical to logical */
 int __cpu_logical_map[NR_CPUS];		/* Map logical to physical */
 
-EXPORT_SYMBOL(phys_cpu_present_map);
-EXPORT_SYMBOL(cpu_online_map);
-
 extern void cpu_idle(void);
 
 /* Number of TCs (or siblings in Intel speak) per CPU core */
@@ -195,7 +190,7 @@
 /* preload SMP state for boot cpu */
 void __devinit smp_prepare_boot_cpu(void)
 {
-	cpu_set(0, phys_cpu_present_map);
+	cpu_set(0, cpu_possible_map);
 	cpu_set(0, cpu_online_map);
 	cpu_set(0, cpu_callin_map);
 }
diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c
index 897fb2b..b6cca01 100644
--- a/arch/mips/kernel/smtc.c
+++ b/arch/mips/kernel/smtc.c
@@ -290,7 +290,7 @@
  * possibly leave some TCs/VPEs as "slave" processors.
  *
  * Use c0_MVPConf0 to find out how many TCs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
+ * cpu_possible_map and the logical/physical mappings.
  */
 
 int __init smtc_build_cpu_map(int start_cpu_slot)
@@ -304,7 +304,7 @@
 	 */
 	ntcs = ((read_c0_mvpconf0() & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT) + 1;
 	for (i=start_cpu_slot; i<NR_CPUS && i<ntcs; i++) {
-		cpu_set(i, phys_cpu_present_map);
+		cpu_set(i, cpu_possible_map);
 		__cpu_number_map[i] = i;
 		__cpu_logical_map[i] = i;
 	}
@@ -521,7 +521,7 @@
 	 * Pull any physically present but unused TCs out of circulation.
 	 */
 	while (tc < (((val & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT) + 1)) {
-		cpu_clear(tc, phys_cpu_present_map);
+		cpu_clear(tc, cpu_possible_map);
 		cpu_clear(tc, cpu_present_map);
 		tc++;
 	}
diff --git a/arch/mips/mti-malta/malta-smtc.c b/arch/mips/mti-malta/malta-smtc.c
index f84a46a..aabd727 100644
--- a/arch/mips/mti-malta/malta-smtc.c
+++ b/arch/mips/mti-malta/malta-smtc.c
@@ -114,9 +114,9 @@
  */
 
 
-void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity)
+void plat_set_irq_affinity(unsigned int irq, const struct cpumask *affinity)
 {
-	cpumask_t tmask = affinity;
+	cpumask_t tmask = *affinity;
 	int cpu = 0;
 	void smtc_set_irq_affinity(unsigned int irq, cpumask_t aff);
 
@@ -139,7 +139,7 @@
 	 * be made to forward to an offline "CPU".
 	 */
 
-	for_each_cpu_mask(cpu, affinity) {
+	for_each_cpu(cpu, affinity) {
 		if ((cpu_data[cpu].vpe_id != 0) || !cpu_online(cpu))
 			cpu_clear(cpu, tmask);
 	}
diff --git a/arch/mips/nxp/pnx8550/common/time.c b/arch/mips/nxp/pnx8550/common/time.c
index 62f495b..cf293b2 100644
--- a/arch/mips/nxp/pnx8550/common/time.c
+++ b/arch/mips/nxp/pnx8550/common/time.c
@@ -102,6 +102,7 @@
 	unsigned int p;
 	unsigned int pow2p;
 
+	pnx8xxx_clockevent.cpumask = cpu_none_mask;
 	clockevents_register_device(&pnx8xxx_clockevent);
 	clocksource_register(&pnx_clocksource);
 
diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c
index 3a7df64..f78c29b 100644
--- a/arch/mips/pmc-sierra/yosemite/smp.c
+++ b/arch/mips/pmc-sierra/yosemite/smp.c
@@ -141,7 +141,7 @@
 }
 
 /*
- * Detect available CPUs, populate phys_cpu_present_map before smp_init
+ * Detect available CPUs, populate cpu_possible_map before smp_init
  *
  * We don't want to start the secondary CPU yet nor do we have a nice probing
  * feature in PMON so we just assume presence of the secondary core.
@@ -150,10 +150,10 @@
 {
 	int i;
 
-	cpus_clear(phys_cpu_present_map);
+	cpus_clear(cpu_possible_map);
 
 	for (i = 0; i < 2; i++) {
-		cpu_set(i, phys_cpu_present_map);
+		cpu_set(i, cpu_possible_map);
 		__cpu_number_map[i]	= i;
 		__cpu_logical_map[i]	= i;
 	}
diff --git a/arch/mips/sgi-ip27/ip27-smp.c b/arch/mips/sgi-ip27/ip27-smp.c
index ba5cdeb..5b47d6b 100644
--- a/arch/mips/sgi-ip27/ip27-smp.c
+++ b/arch/mips/sgi-ip27/ip27-smp.c
@@ -76,7 +76,7 @@
 			/* Only let it join in if it's marked enabled */
 			if ((acpu->cpu_info.flags & KLINFO_ENABLE) &&
 			    (tot_cpus_found != NR_CPUS)) {
-				cpu_set(cpuid, phys_cpu_present_map);
+				cpu_set(cpuid, cpu_possible_map);
 				alloc_cpupda(cpuid, tot_cpus_found);
 				cpus_found++;
 				tot_cpus_found++;
diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c
index 1327c27..f024057 100644
--- a/arch/mips/sgi-ip27/ip27-timer.c
+++ b/arch/mips/sgi-ip27/ip27-timer.c
@@ -134,7 +134,7 @@
 	cd->min_delta_ns        = clockevent_delta2ns(0x300, cd);
 	cd->rating		= 200;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= rt_next_event;
 	cd->set_mode		= rt_set_mode;
 	clockevents_register_device(cd);
diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c
index a35818e..12b465d 100644
--- a/arch/mips/sibyte/bcm1480/irq.c
+++ b/arch/mips/sibyte/bcm1480/irq.c
@@ -50,7 +50,7 @@
 static void disable_bcm1480_irq(unsigned int irq);
 static void ack_bcm1480_irq(unsigned int irq);
 #ifdef CONFIG_SMP
-static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask);
+static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask);
 #endif
 
 #ifdef CONFIG_PCI
@@ -109,7 +109,7 @@
 }
 
 #ifdef CONFIG_SMP
-static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask)
+static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	int i = 0, old_cpu, cpu, int_on, k;
 	u64 cur_ints;
@@ -117,11 +117,11 @@
 	unsigned long flags;
 	unsigned int irq_dirty;
 
-	if (cpus_weight(mask) != 1) {
+	if (cpumask_weight(mask) != 1) {
 		printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq);
 		return;
 	}
-	i = first_cpu(mask);
+	i = cpumask_first(mask);
 
 	/* Convert logical CPU to physical CPU */
 	cpu = cpu_logical_map(i);
diff --git a/arch/mips/sibyte/bcm1480/smp.c b/arch/mips/sibyte/bcm1480/smp.c
index bd9eeb4..dddfda8 100644
--- a/arch/mips/sibyte/bcm1480/smp.c
+++ b/arch/mips/sibyte/bcm1480/smp.c
@@ -136,7 +136,7 @@
 
 /*
  * Use CFE to find out how many CPUs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
+ * cpu_possible_map and the logical/physical mappings.
  * XXXKW will the boot CPU ever not be physical 0?
  *
  * Common setup before any secondaries are started
@@ -145,14 +145,14 @@
 {
 	int i, num;
 
-	cpus_clear(phys_cpu_present_map);
-	cpu_set(0, phys_cpu_present_map);
+	cpus_clear(cpu_possible_map);
+	cpu_set(0, cpu_possible_map);
 	__cpu_number_map[0] = 0;
 	__cpu_logical_map[0] = 0;
 
 	for (i = 1, num = 0; i < NR_CPUS; i++) {
 		if (cfe_cpu_stop(i) == 0) {
-			cpu_set(i, phys_cpu_present_map);
+			cpu_set(i, cpu_possible_map);
 			__cpu_number_map[i] = ++num;
 			__cpu_logical_map[num] = i;
 		}
diff --git a/arch/mips/sibyte/sb1250/irq.c b/arch/mips/sibyte/sb1250/irq.c
index a515848..808ac29 100644
--- a/arch/mips/sibyte/sb1250/irq.c
+++ b/arch/mips/sibyte/sb1250/irq.c
@@ -50,7 +50,7 @@
 static void disable_sb1250_irq(unsigned int irq);
 static void ack_sb1250_irq(unsigned int irq);
 #ifdef CONFIG_SMP
-static void sb1250_set_affinity(unsigned int irq, cpumask_t mask);
+static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask);
 #endif
 
 #ifdef CONFIG_SIBYTE_HAS_LDT
@@ -103,16 +103,16 @@
 }
 
 #ifdef CONFIG_SMP
-static void sb1250_set_affinity(unsigned int irq, cpumask_t mask)
+static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	int i = 0, old_cpu, cpu, int_on;
 	u64 cur_ints;
 	struct irq_desc *desc = irq_desc + irq;
 	unsigned long flags;
 
-	i = first_cpu(mask);
+	i = cpumask_first(mask);
 
-	if (cpus_weight(mask) > 1) {
+	if (cpumask_weight(mask) > 1) {
 		printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq);
 		return;
 	}
diff --git a/arch/mips/sibyte/sb1250/smp.c b/arch/mips/sibyte/sb1250/smp.c
index 0734b93..5950a288 100644
--- a/arch/mips/sibyte/sb1250/smp.c
+++ b/arch/mips/sibyte/sb1250/smp.c
@@ -124,7 +124,7 @@
 
 /*
  * Use CFE to find out how many CPUs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
+ * cpu_possible_map and the logical/physical mappings.
  * XXXKW will the boot CPU ever not be physical 0?
  *
  * Common setup before any secondaries are started
@@ -133,14 +133,14 @@
 {
 	int i, num;
 
-	cpus_clear(phys_cpu_present_map);
-	cpu_set(0, phys_cpu_present_map);
+	cpus_clear(cpu_possible_map);
+	cpu_set(0, cpu_possible_map);
 	__cpu_number_map[0] = 0;
 	__cpu_logical_map[0] = 0;
 
 	for (i = 1, num = 0; i < NR_CPUS; i++) {
 		if (cfe_cpu_stop(i) == 0) {
-			cpu_set(i, phys_cpu_present_map);
+			cpu_set(i, cpu_possible_map);
 			__cpu_number_map[i] = ++num;
 			__cpu_logical_map[num] = i;
 		}
diff --git a/arch/mips/sni/time.c b/arch/mips/sni/time.c
index 796e3ce..69f5f88 100644
--- a/arch/mips/sni/time.c
+++ b/arch/mips/sni/time.c
@@ -80,7 +80,7 @@
 	struct irqaction *action = &a20r_irqaction;
 	unsigned int cpu = smp_processor_id();
 
-	cd->cpumask             = cpumask_of_cpu(cpu);
+	cd->cpumask             = cpumask_of(cpu);
 	clockevents_register_device(cd);
 	action->dev_id = cd;
 	setup_irq(SNI_A20R_IRQ_TIMER, &a20r_irqaction);
diff --git a/arch/mn10300/kernel/init_task.c b/arch/mn10300/kernel/init_task.c
index af16f6e..5ac3566 100644
--- a/arch/mn10300/kernel/init_task.c
+++ b/arch/mn10300/kernel/init_task.c
@@ -18,7 +18,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 644a70b..aacf11d 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -11,6 +11,7 @@
 	select HAVE_OPROFILE
 	select RTC_CLASS
 	select RTC_DRV_PARISC
+	select INIT_ALL_POSSIBLE
 	help
 	  The PA-RISC microprocessor is designed by Hewlett-Packard and used
 	  in many of their workstations & servers (HP9000 700 and 800 series,
diff --git a/arch/parisc/kernel/init_task.c b/arch/parisc/kernel/init_task.c
index f5941c0..1e25a45 100644
--- a/arch/parisc/kernel/init_task.c
+++ b/arch/parisc/kernel/init_task.c
@@ -34,7 +34,6 @@
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c
index 23ef950..4cea935 100644
--- a/arch/parisc/kernel/irq.c
+++ b/arch/parisc/kernel/irq.c
@@ -131,12 +131,12 @@
 	return 0;
 }
 
-static void cpu_set_affinity_irq(unsigned int irq, cpumask_t dest)
+static void cpu_set_affinity_irq(unsigned int irq, const struct cpumask *dest)
 {
-	if (cpu_check_affinity(irq, &dest))
+	if (cpu_check_affinity(irq, dest))
 		return;
 
-	irq_desc[irq].affinity = dest;
+	irq_desc[irq].affinity = *dest;
 }
 #endif
 
diff --git a/arch/parisc/kernel/smp.c b/arch/parisc/kernel/smp.c
index d47f397..80bc000 100644
--- a/arch/parisc/kernel/smp.c
+++ b/arch/parisc/kernel/smp.c
@@ -67,21 +67,6 @@
 
 static int parisc_max_cpus __read_mostly = 1;
 
-/* online cpus are ones that we've managed to bring up completely
- * possible cpus are all valid cpu 
- * present cpus are all detected cpu
- *
- * On startup we bring up the "possible" cpus. Since we discover
- * CPUs later, we add them as hotplug, so the possible cpu mask is
- * empty in the beginning.
- */
-
-cpumask_t cpu_online_map   __read_mostly = CPU_MASK_NONE;	/* Bitmap of online CPUs */
-cpumask_t cpu_possible_map __read_mostly = CPU_MASK_ALL;	/* Bitmap of Present CPUs */
-
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(cpu_possible_map);
-
 DEFINE_PER_CPU(spinlock_t, ipi_lock) = SPIN_LOCK_UNLOCKED;
 
 enum ipi_message_type {
diff --git a/arch/powerpc/include/asm/disassemble.h b/arch/powerpc/include/asm/disassemble.h
new file mode 100644
index 0000000..9b198d1
--- /dev/null
+++ b/arch/powerpc/include/asm/disassemble.h
@@ -0,0 +1,80 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#ifndef __ASM_PPC_DISASSEMBLE_H__
+#define __ASM_PPC_DISASSEMBLE_H__
+
+#include <linux/types.h>
+
+static inline unsigned int get_op(u32 inst)
+{
+	return inst >> 26;
+}
+
+static inline unsigned int get_xop(u32 inst)
+{
+	return (inst >> 1) & 0x3ff;
+}
+
+static inline unsigned int get_sprn(u32 inst)
+{
+	return ((inst >> 16) & 0x1f) | ((inst >> 6) & 0x3e0);
+}
+
+static inline unsigned int get_dcrn(u32 inst)
+{
+	return ((inst >> 16) & 0x1f) | ((inst >> 6) & 0x3e0);
+}
+
+static inline unsigned int get_rt(u32 inst)
+{
+	return (inst >> 21) & 0x1f;
+}
+
+static inline unsigned int get_rs(u32 inst)
+{
+	return (inst >> 21) & 0x1f;
+}
+
+static inline unsigned int get_ra(u32 inst)
+{
+	return (inst >> 16) & 0x1f;
+}
+
+static inline unsigned int get_rb(u32 inst)
+{
+	return (inst >> 11) & 0x1f;
+}
+
+static inline unsigned int get_rc(u32 inst)
+{
+	return inst & 0x1;
+}
+
+static inline unsigned int get_ws(u32 inst)
+{
+	return (inst >> 11) & 0x1f;
+}
+
+static inline unsigned int get_d(u32 inst)
+{
+	return inst & 0xffff;
+}
+
+#endif /* __ASM_PPC_DISASSEMBLE_H__ */
diff --git a/arch/powerpc/include/asm/kvm_44x.h b/arch/powerpc/include/asm/kvm_44x.h
new file mode 100644
index 0000000..f49031b
--- /dev/null
+++ b/arch/powerpc/include/asm/kvm_44x.h
@@ -0,0 +1,61 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#ifndef __ASM_44X_H__
+#define __ASM_44X_H__
+
+#include <linux/kvm_host.h>
+
+#define PPC44x_TLB_SIZE 64
+
+/* If the guest is expecting it, this can be as large as we like; we'd just
+ * need to find some way of advertising it. */
+#define KVM44x_GUEST_TLB_SIZE 64
+
+struct kvmppc_44x_shadow_ref {
+	struct page *page;
+	u16 gtlb_index;
+	u8 writeable;
+	u8 tid;
+};
+
+struct kvmppc_vcpu_44x {
+	/* Unmodified copy of the guest's TLB. */
+	struct kvmppc_44x_tlbe guest_tlb[KVM44x_GUEST_TLB_SIZE];
+
+	/* References to guest pages in the hardware TLB. */
+	struct kvmppc_44x_shadow_ref shadow_refs[PPC44x_TLB_SIZE];
+
+	/* State of the shadow TLB at guest context switch time. */
+	struct kvmppc_44x_tlbe shadow_tlb[PPC44x_TLB_SIZE];
+	u8 shadow_tlb_mod[PPC44x_TLB_SIZE];
+
+	struct kvm_vcpu vcpu;
+};
+
+static inline struct kvmppc_vcpu_44x *to_44x(struct kvm_vcpu *vcpu)
+{
+	return container_of(vcpu, struct kvmppc_vcpu_44x, vcpu);
+}
+
+void kvmppc_set_pid(struct kvm_vcpu *vcpu, u32 new_pid);
+void kvmppc_44x_tlb_put(struct kvm_vcpu *vcpu);
+void kvmppc_44x_tlb_load(struct kvm_vcpu *vcpu);
+
+#endif /* __ASM_44X_H__ */
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index 34b52b7..c1e436f 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -64,27 +64,58 @@
 	u32 halt_wakeup;
 };
 
-struct tlbe {
+struct kvmppc_44x_tlbe {
 	u32 tid; /* Only the low 8 bits are used. */
 	u32 word0;
 	u32 word1;
 	u32 word2;
 };
 
+enum kvm_exit_types {
+	MMIO_EXITS,
+	DCR_EXITS,
+	SIGNAL_EXITS,
+	ITLB_REAL_MISS_EXITS,
+	ITLB_VIRT_MISS_EXITS,
+	DTLB_REAL_MISS_EXITS,
+	DTLB_VIRT_MISS_EXITS,
+	SYSCALL_EXITS,
+	ISI_EXITS,
+	DSI_EXITS,
+	EMULATED_INST_EXITS,
+	EMULATED_MTMSRWE_EXITS,
+	EMULATED_WRTEE_EXITS,
+	EMULATED_MTSPR_EXITS,
+	EMULATED_MFSPR_EXITS,
+	EMULATED_MTMSR_EXITS,
+	EMULATED_MFMSR_EXITS,
+	EMULATED_TLBSX_EXITS,
+	EMULATED_TLBWE_EXITS,
+	EMULATED_RFI_EXITS,
+	DEC_EXITS,
+	EXT_INTR_EXITS,
+	HALT_WAKEUP,
+	USR_PR_INST,
+	FP_UNAVAIL,
+	DEBUG_EXITS,
+	TIMEINGUEST,
+	__NUMBER_OF_KVM_EXIT_TYPES
+};
+
+/* allow access to big endian 32bit upper/lower parts and 64bit var */
+struct kvmppc_exit_timing {
+	union {
+		u64 tv64;
+		struct {
+			u32 tbu, tbl;
+		} tv32;
+	};
+};
+
 struct kvm_arch {
 };
 
 struct kvm_vcpu_arch {
-	/* Unmodified copy of the guest's TLB. */
-	struct tlbe guest_tlb[PPC44x_TLB_SIZE];
-	/* TLB that's actually used when the guest is running. */
-	struct tlbe shadow_tlb[PPC44x_TLB_SIZE];
-	/* Pages which are referenced in the shadow TLB. */
-	struct page *shadow_pages[PPC44x_TLB_SIZE];
-
-	/* Track which TLB entries we've modified in the current exit. */
-	u8 shadow_tlb_mod[PPC44x_TLB_SIZE];
-
 	u32 host_stack;
 	u32 host_pid;
 	u32 host_dbcr0;
@@ -94,32 +125,32 @@
 	u32 host_msr;
 
 	u64 fpr[32];
-	u32 gpr[32];
+	ulong gpr[32];
 
-	u32 pc;
+	ulong pc;
 	u32 cr;
-	u32 ctr;
-	u32 lr;
-	u32 xer;
+	ulong ctr;
+	ulong lr;
+	ulong xer;
 
-	u32 msr;
+	ulong msr;
 	u32 mmucr;
-	u32 sprg0;
-	u32 sprg1;
-	u32 sprg2;
-	u32 sprg3;
-	u32 sprg4;
-	u32 sprg5;
-	u32 sprg6;
-	u32 sprg7;
-	u32 srr0;
-	u32 srr1;
-	u32 csrr0;
-	u32 csrr1;
-	u32 dsrr0;
-	u32 dsrr1;
-	u32 dear;
-	u32 esr;
+	ulong sprg0;
+	ulong sprg1;
+	ulong sprg2;
+	ulong sprg3;
+	ulong sprg4;
+	ulong sprg5;
+	ulong sprg6;
+	ulong sprg7;
+	ulong srr0;
+	ulong srr1;
+	ulong csrr0;
+	ulong csrr1;
+	ulong dsrr0;
+	ulong dsrr1;
+	ulong dear;
+	ulong esr;
 	u32 dec;
 	u32 decar;
 	u32 tbl;
@@ -127,7 +158,7 @@
 	u32 tcr;
 	u32 tsr;
 	u32 ivor[16];
-	u32 ivpr;
+	ulong ivpr;
 	u32 pir;
 
 	u32 shadow_pid;
@@ -140,9 +171,22 @@
 	u32 dbcr0;
 	u32 dbcr1;
 
+#ifdef CONFIG_KVM_EXIT_TIMING
+	struct kvmppc_exit_timing timing_exit;
+	struct kvmppc_exit_timing timing_last_enter;
+	u32 last_exit_type;
+	u32 timing_count_type[__NUMBER_OF_KVM_EXIT_TYPES];
+	u64 timing_sum_duration[__NUMBER_OF_KVM_EXIT_TYPES];
+	u64 timing_sum_quad_duration[__NUMBER_OF_KVM_EXIT_TYPES];
+	u64 timing_min_duration[__NUMBER_OF_KVM_EXIT_TYPES];
+	u64 timing_max_duration[__NUMBER_OF_KVM_EXIT_TYPES];
+	u64 timing_last_exit;
+	struct dentry *debugfs_exit_timing;
+#endif
+
 	u32 last_inst;
-	u32 fault_dear;
-	u32 fault_esr;
+	ulong fault_dear;
+	ulong fault_esr;
 	gpa_t paddr_accessed;
 
 	u8 io_gpr; /* GPR used as IO source/target */
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index bb62ad8..36d2a50 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -29,11 +29,6 @@
 #include <linux/kvm_types.h>
 #include <linux/kvm_host.h>
 
-struct kvm_tlb {
-	struct tlbe guest_tlb[PPC44x_TLB_SIZE];
-	struct tlbe shadow_tlb[PPC44x_TLB_SIZE];
-};
-
 enum emulation_result {
 	EMULATE_DONE,         /* no further processing */
 	EMULATE_DO_MMIO,      /* kvm_run filled with MMIO request */
@@ -41,9 +36,6 @@
 	EMULATE_FAIL,         /* can't emulate this instruction */
 };
 
-extern const unsigned char exception_priority[];
-extern const unsigned char priority_exception[];
-
 extern int __kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu);
 extern char kvmppc_handlers_start[];
 extern unsigned long kvmppc_handler_len;
@@ -58,51 +50,44 @@
 extern int kvmppc_emulate_instruction(struct kvm_run *run,
                                       struct kvm_vcpu *vcpu);
 extern int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu);
+extern void kvmppc_emulate_dec(struct kvm_vcpu *vcpu);
 
-extern void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gfn_t gfn,
-                           u64 asid, u32 flags);
-extern void kvmppc_mmu_invalidate(struct kvm_vcpu *vcpu, gva_t eaddr,
-                                  gva_t eend, u32 asid);
+extern void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gpa_t gpaddr,
+                           u64 asid, u32 flags, u32 max_bytes,
+                           unsigned int gtlb_idx);
 extern void kvmppc_mmu_priv_switch(struct kvm_vcpu *vcpu, int usermode);
 extern void kvmppc_mmu_switch_pid(struct kvm_vcpu *vcpu, u32 pid);
 
-/* XXX Book E specific */
-extern void kvmppc_tlbe_set_modified(struct kvm_vcpu *vcpu, unsigned int i);
+/* Core-specific hooks */
 
-extern void kvmppc_check_and_deliver_interrupts(struct kvm_vcpu *vcpu);
+extern struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm,
+                                                unsigned int id);
+extern void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu);
+extern int kvmppc_core_vcpu_setup(struct kvm_vcpu *vcpu);
+extern int kvmppc_core_check_processor_compat(void);
+extern int kvmppc_core_vcpu_translate(struct kvm_vcpu *vcpu,
+                                      struct kvm_translation *tr);
 
-static inline void kvmppc_queue_exception(struct kvm_vcpu *vcpu, int exception)
-{
-	unsigned int priority = exception_priority[exception];
-	set_bit(priority, &vcpu->arch.pending_exceptions);
-}
+extern void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
+extern void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu);
 
-static inline void kvmppc_clear_exception(struct kvm_vcpu *vcpu, int exception)
-{
-	unsigned int priority = exception_priority[exception];
-	clear_bit(priority, &vcpu->arch.pending_exceptions);
-}
+extern void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu);
+extern void kvmppc_core_load_host_debugstate(struct kvm_vcpu *vcpu);
 
-/* Helper function for "full" MSR writes. No need to call this if only EE is
- * changing. */
-static inline void kvmppc_set_msr(struct kvm_vcpu *vcpu, u32 new_msr)
-{
-	if ((new_msr & MSR_PR) != (vcpu->arch.msr & MSR_PR))
-		kvmppc_mmu_priv_switch(vcpu, new_msr & MSR_PR);
+extern void kvmppc_core_deliver_interrupts(struct kvm_vcpu *vcpu);
+extern int kvmppc_core_pending_dec(struct kvm_vcpu *vcpu);
+extern void kvmppc_core_queue_program(struct kvm_vcpu *vcpu);
+extern void kvmppc_core_queue_dec(struct kvm_vcpu *vcpu);
+extern void kvmppc_core_queue_external(struct kvm_vcpu *vcpu,
+                                       struct kvm_interrupt *irq);
 
-	vcpu->arch.msr = new_msr;
+extern int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
+                                  unsigned int op, int *advance);
+extern int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, int rs);
+extern int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, int rt);
 
-	if (vcpu->arch.msr & MSR_WE)
-		kvm_vcpu_block(vcpu);
-}
-
-static inline void kvmppc_set_pid(struct kvm_vcpu *vcpu, u32 new_pid)
-{
-	if (vcpu->arch.pid != new_pid) {
-		vcpu->arch.pid = new_pid;
-		vcpu->arch.swap_pid = 1;
-	}
-}
+extern int kvmppc_booke_init(void);
+extern void kvmppc_booke_exit(void);
 
 extern void kvmppc_core_destroy_mmu(struct kvm_vcpu *vcpu);
 
diff --git a/arch/powerpc/include/asm/mmu-44x.h b/arch/powerpc/include/asm/mmu-44x.h
index 8a97cfb..27cc6fd 100644
--- a/arch/powerpc/include/asm/mmu-44x.h
+++ b/arch/powerpc/include/asm/mmu-44x.h
@@ -56,6 +56,7 @@
 #ifndef __ASSEMBLY__
 
 extern unsigned int tlb_44x_hwater;
+extern unsigned int tlb_44x_index;
 
 typedef struct {
 	unsigned int	id;
diff --git a/arch/powerpc/include/asm/topology.h b/arch/powerpc/include/asm/topology.h
index c32da6f..373fca3 100644
--- a/arch/powerpc/include/asm/topology.h
+++ b/arch/powerpc/include/asm/topology.h
@@ -48,7 +48,6 @@
 
 /* sched_domains SD_NODE_INIT for PPC64 machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
-	.span			= CPU_MASK_NONE,	\
 	.parent			= NULL,			\
 	.child			= NULL,			\
 	.groups			= NULL,			\
diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index 661d07d..9937fe4 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -23,9 +23,6 @@
 #include <linux/mm.h>
 #include <linux/suspend.h>
 #include <linux/hrtimer.h>
-#ifdef CONFIG_KVM
-#include <linux/kvm_host.h>
-#endif
 #ifdef CONFIG_PPC64
 #include <linux/time.h>
 #include <linux/hardirq.h>
@@ -51,6 +48,9 @@
 #ifdef CONFIG_PPC_ISERIES
 #include <asm/iseries/alpaca.h>
 #endif
+#ifdef CONFIG_KVM
+#include <asm/kvm_44x.h>
+#endif
 
 #if defined(CONFIG_BOOKE) || defined(CONFIG_40x)
 #include "head_booke.h"
@@ -357,12 +357,10 @@
 	DEFINE(PTE_SIZE, sizeof(pte_t));
 
 #ifdef CONFIG_KVM
-	DEFINE(TLBE_BYTES, sizeof(struct tlbe));
+	DEFINE(TLBE_BYTES, sizeof(struct kvmppc_44x_tlbe));
 
 	DEFINE(VCPU_HOST_STACK, offsetof(struct kvm_vcpu, arch.host_stack));
 	DEFINE(VCPU_HOST_PID, offsetof(struct kvm_vcpu, arch.host_pid));
-	DEFINE(VCPU_SHADOW_TLB, offsetof(struct kvm_vcpu, arch.shadow_tlb));
-	DEFINE(VCPU_SHADOW_MOD, offsetof(struct kvm_vcpu, arch.shadow_tlb_mod));
 	DEFINE(VCPU_GPRS, offsetof(struct kvm_vcpu, arch.gpr));
 	DEFINE(VCPU_LR, offsetof(struct kvm_vcpu, arch.lr));
 	DEFINE(VCPU_CR, offsetof(struct kvm_vcpu, arch.cr));
@@ -385,5 +383,16 @@
 	DEFINE(PTE_T_LOG2, PTE_T_LOG2);
 #endif
 
+#ifdef CONFIG_KVM_EXIT_TIMING
+	DEFINE(VCPU_TIMING_EXIT_TBU, offsetof(struct kvm_vcpu,
+						arch.timing_exit.tv32.tbu));
+	DEFINE(VCPU_TIMING_EXIT_TBL, offsetof(struct kvm_vcpu,
+						arch.timing_exit.tv32.tbl));
+	DEFINE(VCPU_TIMING_LAST_ENTER_TBU, offsetof(struct kvm_vcpu,
+					arch.timing_last_enter.tv32.tbu));
+	DEFINE(VCPU_TIMING_LAST_ENTER_TBL, offsetof(struct kvm_vcpu,
+					arch.timing_last_enter.tv32.tbl));
+#endif
+
 	return 0;
 }
diff --git a/arch/powerpc/kernel/init_task.c b/arch/powerpc/kernel/init_task.c
index 4c85b8d..688b329 100644
--- a/arch/powerpc/kernel/init_task.c
+++ b/arch/powerpc/kernel/init_task.c
@@ -7,7 +7,6 @@
 #include <linux/mqueue.h>
 #include <asm/uaccess.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index ac222d0..23b8b5e 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -237,7 +237,7 @@
 			mask = map;
 		}
 		if (irq_desc[irq].chip->set_affinity)
-			irq_desc[irq].chip->set_affinity(irq, mask);
+			irq_desc[irq].chip->set_affinity(irq, &mask);
 		else if (irq_desc[irq].action && !(warned++))
 			printk("Cannot set affinity for irq %i\n", irq);
 	}
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 51b201d..fb7049c 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -33,6 +33,7 @@
 #include <linux/mqueue.h>
 #include <linux/hardirq.h>
 #include <linux/utsname.h>
+#include <linux/kernel_stat.h>
 
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 8ac3f72..65484b2 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -59,13 +59,9 @@
 
 struct thread_info *secondary_ti;
 
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-cpumask_t cpu_online_map = CPU_MASK_NONE;
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map) = CPU_MASK_NONE;
 DEFINE_PER_CPU(cpumask_t, cpu_core_map) = CPU_MASK_NONE;
 
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(cpu_possible_map);
 EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
 EXPORT_PER_CPU_SYMBOL(cpu_core_map);
 
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index e1f3a51..c956403 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -256,8 +256,10 @@
 		delta += sys_time;
 		get_paca()->system_time = 0;
 	}
-	account_system_time(tsk, 0, delta);
-	account_system_time_scaled(tsk, deltascaled);
+	if (in_irq() || idle_task(smp_processor_id()) != tsk)
+		account_system_time(tsk, 0, delta, deltascaled);
+	else
+		account_idle_time(delta);
 	per_cpu(cputime_last_delta, smp_processor_id()) = delta;
 	per_cpu(cputime_scaled_last_delta, smp_processor_id()) = deltascaled;
 	local_irq_restore(flags);
@@ -275,10 +277,8 @@
 
 	utime = get_paca()->user_time;
 	get_paca()->user_time = 0;
-	account_user_time(tsk, utime);
-
 	utimescaled = cputime_to_scaled(utime);
-	account_user_time_scaled(tsk, utimescaled);
+	account_user_time(tsk, utime, utimescaled);
 }
 
 /*
@@ -338,8 +338,12 @@
 	tb = mftb();
 	purr = mfspr(SPRN_PURR);
 	stolen = (tb - pme->tb) - (purr - pme->purr);
-	if (stolen > 0)
-		account_steal_time(current, stolen);
+	if (stolen > 0) {
+		if (idle_task(smp_processor_id()) != current)
+			account_steal_time(stolen);
+		else
+			account_idle_time(stolen);
+	}
 	pme->tb = tb;
 	pme->purr = purr;
 }
@@ -844,7 +848,7 @@
 	struct clock_event_device *dec = &per_cpu(decrementers, cpu).event;
 
 	*dec = decrementer_clockevent;
-	dec->cpumask = cpumask_of_cpu(cpu);
+	dec->cpumask = cpumask_of(cpu);
 
 	printk(KERN_DEBUG "clockevent: %s mult[%lx] shift[%d] cpu[%d]\n",
 	       dec->name, dec->mult, dec->shift, cpu);
diff --git a/arch/powerpc/kvm/44x.c b/arch/powerpc/kvm/44x.c
new file mode 100644
index 0000000..a66bec5
--- /dev/null
+++ b/arch/powerpc/kvm/44x.c
@@ -0,0 +1,228 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#include <linux/kvm_host.h>
+#include <linux/err.h>
+
+#include <asm/reg.h>
+#include <asm/cputable.h>
+#include <asm/tlbflush.h>
+#include <asm/kvm_44x.h>
+#include <asm/kvm_ppc.h>
+
+#include "44x_tlb.h"
+
+/* Note: clearing MSR[DE] just means that the debug interrupt will not be
+ * delivered *immediately*. Instead, it simply sets the appropriate DBSR bits.
+ * If those DBSR bits are still set when MSR[DE] is re-enabled, the interrupt
+ * will be delivered as an "imprecise debug event" (which is indicated by
+ * DBSR[IDE].
+ */
+static void kvm44x_disable_debug_interrupts(void)
+{
+	mtmsr(mfmsr() & ~MSR_DE);
+}
+
+void kvmppc_core_load_host_debugstate(struct kvm_vcpu *vcpu)
+{
+	kvm44x_disable_debug_interrupts();
+
+	mtspr(SPRN_IAC1, vcpu->arch.host_iac[0]);
+	mtspr(SPRN_IAC2, vcpu->arch.host_iac[1]);
+	mtspr(SPRN_IAC3, vcpu->arch.host_iac[2]);
+	mtspr(SPRN_IAC4, vcpu->arch.host_iac[3]);
+	mtspr(SPRN_DBCR1, vcpu->arch.host_dbcr1);
+	mtspr(SPRN_DBCR2, vcpu->arch.host_dbcr2);
+	mtspr(SPRN_DBCR0, vcpu->arch.host_dbcr0);
+	mtmsr(vcpu->arch.host_msr);
+}
+
+void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu)
+{
+	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
+	u32 dbcr0 = 0;
+
+	vcpu->arch.host_msr = mfmsr();
+	kvm44x_disable_debug_interrupts();
+
+	/* Save host debug register state. */
+	vcpu->arch.host_iac[0] = mfspr(SPRN_IAC1);
+	vcpu->arch.host_iac[1] = mfspr(SPRN_IAC2);
+	vcpu->arch.host_iac[2] = mfspr(SPRN_IAC3);
+	vcpu->arch.host_iac[3] = mfspr(SPRN_IAC4);
+	vcpu->arch.host_dbcr0 = mfspr(SPRN_DBCR0);
+	vcpu->arch.host_dbcr1 = mfspr(SPRN_DBCR1);
+	vcpu->arch.host_dbcr2 = mfspr(SPRN_DBCR2);
+
+	/* set registers up for guest */
+
+	if (dbg->bp[0]) {
+		mtspr(SPRN_IAC1, dbg->bp[0]);
+		dbcr0 |= DBCR0_IAC1 | DBCR0_IDM;
+	}
+	if (dbg->bp[1]) {
+		mtspr(SPRN_IAC2, dbg->bp[1]);
+		dbcr0 |= DBCR0_IAC2 | DBCR0_IDM;
+	}
+	if (dbg->bp[2]) {
+		mtspr(SPRN_IAC3, dbg->bp[2]);
+		dbcr0 |= DBCR0_IAC3 | DBCR0_IDM;
+	}
+	if (dbg->bp[3]) {
+		mtspr(SPRN_IAC4, dbg->bp[3]);
+		dbcr0 |= DBCR0_IAC4 | DBCR0_IDM;
+	}
+
+	mtspr(SPRN_DBCR0, dbcr0);
+	mtspr(SPRN_DBCR1, 0);
+	mtspr(SPRN_DBCR2, 0);
+}
+
+void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	kvmppc_44x_tlb_load(vcpu);
+}
+
+void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	kvmppc_44x_tlb_put(vcpu);
+}
+
+int kvmppc_core_check_processor_compat(void)
+{
+	int r;
+
+	if (strcmp(cur_cpu_spec->platform, "ppc440") == 0)
+		r = 0;
+	else
+		r = -ENOTSUPP;
+
+	return r;
+}
+
+int kvmppc_core_vcpu_setup(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	struct kvmppc_44x_tlbe *tlbe = &vcpu_44x->guest_tlb[0];
+	int i;
+
+	tlbe->tid = 0;
+	tlbe->word0 = PPC44x_TLB_16M | PPC44x_TLB_VALID;
+	tlbe->word1 = 0;
+	tlbe->word2 = PPC44x_TLB_SX | PPC44x_TLB_SW | PPC44x_TLB_SR;
+
+	tlbe++;
+	tlbe->tid = 0;
+	tlbe->word0 = 0xef600000 | PPC44x_TLB_4K | PPC44x_TLB_VALID;
+	tlbe->word1 = 0xef600000;
+	tlbe->word2 = PPC44x_TLB_SX | PPC44x_TLB_SW | PPC44x_TLB_SR
+	              | PPC44x_TLB_I | PPC44x_TLB_G;
+
+	/* Since the guest can directly access the timebase, it must know the
+	 * real timebase frequency. Accordingly, it must see the state of
+	 * CCR1[TCS]. */
+	vcpu->arch.ccr1 = mfspr(SPRN_CCR1);
+
+	for (i = 0; i < ARRAY_SIZE(vcpu_44x->shadow_refs); i++)
+		vcpu_44x->shadow_refs[i].gtlb_index = -1;
+
+	return 0;
+}
+
+/* 'linear_address' is actually an encoding of AS|PID|EADDR . */
+int kvmppc_core_vcpu_translate(struct kvm_vcpu *vcpu,
+                               struct kvm_translation *tr)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	struct kvmppc_44x_tlbe *gtlbe;
+	int index;
+	gva_t eaddr;
+	u8 pid;
+	u8 as;
+
+	eaddr = tr->linear_address;
+	pid = (tr->linear_address >> 32) & 0xff;
+	as = (tr->linear_address >> 40) & 0x1;
+
+	index = kvmppc_44x_tlb_index(vcpu, eaddr, pid, as);
+	if (index == -1) {
+		tr->valid = 0;
+		return 0;
+	}
+
+	gtlbe = &vcpu_44x->guest_tlb[index];
+
+	tr->physical_address = tlb_xlate(gtlbe, eaddr);
+	/* XXX what does "writeable" and "usermode" even mean? */
+	tr->valid = 1;
+
+	return 0;
+}
+
+struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x;
+	struct kvm_vcpu *vcpu;
+	int err;
+
+	vcpu_44x = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
+	if (!vcpu_44x) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	vcpu = &vcpu_44x->vcpu;
+	err = kvm_vcpu_init(vcpu, kvm, id);
+	if (err)
+		goto free_vcpu;
+
+	return vcpu;
+
+free_vcpu:
+	kmem_cache_free(kvm_vcpu_cache, vcpu_44x);
+out:
+	return ERR_PTR(err);
+}
+
+void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+
+	kvm_vcpu_uninit(vcpu);
+	kmem_cache_free(kvm_vcpu_cache, vcpu_44x);
+}
+
+static int kvmppc_44x_init(void)
+{
+	int r;
+
+	r = kvmppc_booke_init();
+	if (r)
+		return r;
+
+	return kvm_init(NULL, sizeof(struct kvmppc_vcpu_44x), THIS_MODULE);
+}
+
+static void kvmppc_44x_exit(void)
+{
+	kvmppc_booke_exit();
+}
+
+module_init(kvmppc_44x_init);
+module_exit(kvmppc_44x_exit);
diff --git a/arch/powerpc/kvm/44x_emulate.c b/arch/powerpc/kvm/44x_emulate.c
new file mode 100644
index 0000000..82489a7
--- /dev/null
+++ b/arch/powerpc/kvm/44x_emulate.c
@@ -0,0 +1,371 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#include <asm/kvm_ppc.h>
+#include <asm/dcr.h>
+#include <asm/dcr-regs.h>
+#include <asm/disassemble.h>
+#include <asm/kvm_44x.h>
+#include "timing.h"
+
+#include "booke.h"
+#include "44x_tlb.h"
+
+#define OP_RFI      19
+
+#define XOP_RFI     50
+#define XOP_MFMSR   83
+#define XOP_WRTEE   131
+#define XOP_MTMSR   146
+#define XOP_WRTEEI  163
+#define XOP_MFDCR   323
+#define XOP_MTDCR   451
+#define XOP_TLBSX   914
+#define XOP_ICCCI   966
+#define XOP_TLBWE   978
+
+static void kvmppc_emul_rfi(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.pc = vcpu->arch.srr0;
+	kvmppc_set_msr(vcpu, vcpu->arch.srr1);
+}
+
+int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
+                           unsigned int inst, int *advance)
+{
+	int emulated = EMULATE_DONE;
+	int dcrn;
+	int ra;
+	int rb;
+	int rc;
+	int rs;
+	int rt;
+	int ws;
+
+	switch (get_op(inst)) {
+	case OP_RFI:
+		switch (get_xop(inst)) {
+		case XOP_RFI:
+			kvmppc_emul_rfi(vcpu);
+			kvmppc_set_exit_type(vcpu, EMULATED_RFI_EXITS);
+			*advance = 0;
+			break;
+
+		default:
+			emulated = EMULATE_FAIL;
+			break;
+		}
+		break;
+
+	case 31:
+		switch (get_xop(inst)) {
+
+		case XOP_MFMSR:
+			rt = get_rt(inst);
+			vcpu->arch.gpr[rt] = vcpu->arch.msr;
+			kvmppc_set_exit_type(vcpu, EMULATED_MFMSR_EXITS);
+			break;
+
+		case XOP_MTMSR:
+			rs = get_rs(inst);
+			kvmppc_set_exit_type(vcpu, EMULATED_MTMSR_EXITS);
+			kvmppc_set_msr(vcpu, vcpu->arch.gpr[rs]);
+			break;
+
+		case XOP_WRTEE:
+			rs = get_rs(inst);
+			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
+							 | (vcpu->arch.gpr[rs] & MSR_EE);
+			kvmppc_set_exit_type(vcpu, EMULATED_WRTEE_EXITS);
+			break;
+
+		case XOP_WRTEEI:
+			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
+							 | (inst & MSR_EE);
+			kvmppc_set_exit_type(vcpu, EMULATED_WRTEE_EXITS);
+			break;
+
+		case XOP_MFDCR:
+			dcrn = get_dcrn(inst);
+			rt = get_rt(inst);
+
+			/* The guest may access CPR0 registers to determine the timebase
+			 * frequency, and it must know the real host frequency because it
+			 * can directly access the timebase registers.
+			 *
+			 * It would be possible to emulate those accesses in userspace,
+			 * but userspace can really only figure out the end frequency.
+			 * We could decompose that into the factors that compute it, but
+			 * that's tricky math, and it's easier to just report the real
+			 * CPR0 values.
+			 */
+			switch (dcrn) {
+			case DCRN_CPR0_CONFIG_ADDR:
+				vcpu->arch.gpr[rt] = vcpu->arch.cpr0_cfgaddr;
+				break;
+			case DCRN_CPR0_CONFIG_DATA:
+				local_irq_disable();
+				mtdcr(DCRN_CPR0_CONFIG_ADDR,
+					  vcpu->arch.cpr0_cfgaddr);
+				vcpu->arch.gpr[rt] = mfdcr(DCRN_CPR0_CONFIG_DATA);
+				local_irq_enable();
+				break;
+			default:
+				run->dcr.dcrn = dcrn;
+				run->dcr.data =  0;
+				run->dcr.is_write = 0;
+				vcpu->arch.io_gpr = rt;
+				vcpu->arch.dcr_needed = 1;
+				kvmppc_account_exit(vcpu, DCR_EXITS);
+				emulated = EMULATE_DO_DCR;
+			}
+
+			break;
+
+		case XOP_MTDCR:
+			dcrn = get_dcrn(inst);
+			rs = get_rs(inst);
+
+			/* emulate some access in kernel */
+			switch (dcrn) {
+			case DCRN_CPR0_CONFIG_ADDR:
+				vcpu->arch.cpr0_cfgaddr = vcpu->arch.gpr[rs];
+				break;
+			default:
+				run->dcr.dcrn = dcrn;
+				run->dcr.data = vcpu->arch.gpr[rs];
+				run->dcr.is_write = 1;
+				vcpu->arch.dcr_needed = 1;
+				kvmppc_account_exit(vcpu, DCR_EXITS);
+				emulated = EMULATE_DO_DCR;
+			}
+
+			break;
+
+		case XOP_TLBWE:
+			ra = get_ra(inst);
+			rs = get_rs(inst);
+			ws = get_ws(inst);
+			emulated = kvmppc_44x_emul_tlbwe(vcpu, ra, rs, ws);
+			break;
+
+		case XOP_TLBSX:
+			rt = get_rt(inst);
+			ra = get_ra(inst);
+			rb = get_rb(inst);
+			rc = get_rc(inst);
+			emulated = kvmppc_44x_emul_tlbsx(vcpu, rt, ra, rb, rc);
+			break;
+
+		case XOP_ICCCI:
+			break;
+
+		default:
+			emulated = EMULATE_FAIL;
+		}
+
+		break;
+
+	default:
+		emulated = EMULATE_FAIL;
+	}
+
+	return emulated;
+}
+
+int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, int rs)
+{
+	switch (sprn) {
+	case SPRN_MMUCR:
+		vcpu->arch.mmucr = vcpu->arch.gpr[rs]; break;
+	case SPRN_PID:
+		kvmppc_set_pid(vcpu, vcpu->arch.gpr[rs]); break;
+	case SPRN_CCR0:
+		vcpu->arch.ccr0 = vcpu->arch.gpr[rs]; break;
+	case SPRN_CCR1:
+		vcpu->arch.ccr1 = vcpu->arch.gpr[rs]; break;
+	case SPRN_DEAR:
+		vcpu->arch.dear = vcpu->arch.gpr[rs]; break;
+	case SPRN_ESR:
+		vcpu->arch.esr = vcpu->arch.gpr[rs]; break;
+	case SPRN_DBCR0:
+		vcpu->arch.dbcr0 = vcpu->arch.gpr[rs]; break;
+	case SPRN_DBCR1:
+		vcpu->arch.dbcr1 = vcpu->arch.gpr[rs]; break;
+	case SPRN_TSR:
+		vcpu->arch.tsr &= ~vcpu->arch.gpr[rs]; break;
+	case SPRN_TCR:
+		vcpu->arch.tcr = vcpu->arch.gpr[rs];
+		kvmppc_emulate_dec(vcpu);
+		break;
+
+	/* Note: SPRG4-7 are user-readable. These values are
+	 * loaded into the real SPRGs when resuming the
+	 * guest. */
+	case SPRN_SPRG4:
+		vcpu->arch.sprg4 = vcpu->arch.gpr[rs]; break;
+	case SPRN_SPRG5:
+		vcpu->arch.sprg5 = vcpu->arch.gpr[rs]; break;
+	case SPRN_SPRG6:
+		vcpu->arch.sprg6 = vcpu->arch.gpr[rs]; break;
+	case SPRN_SPRG7:
+		vcpu->arch.sprg7 = vcpu->arch.gpr[rs]; break;
+
+	case SPRN_IVPR:
+		vcpu->arch.ivpr = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR0:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_CRITICAL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR1:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_MACHINE_CHECK] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR2:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DATA_STORAGE] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR3:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_INST_STORAGE] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR4:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_EXTERNAL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR5:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_ALIGNMENT] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR6:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_PROGRAM] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR7:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_FP_UNAVAIL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR8:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_SYSCALL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR9:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_AP_UNAVAIL] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR10:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DECREMENTER] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR11:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_FIT] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR12:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_WATCHDOG] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR13:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DTLB_MISS] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR14:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_ITLB_MISS] = vcpu->arch.gpr[rs];
+		break;
+	case SPRN_IVOR15:
+		vcpu->arch.ivor[BOOKE_IRQPRIO_DEBUG] = vcpu->arch.gpr[rs];
+		break;
+
+	default:
+		return EMULATE_FAIL;
+	}
+
+	kvmppc_set_exit_type(vcpu, EMULATED_MTSPR_EXITS);
+	return EMULATE_DONE;
+}
+
+int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, int rt)
+{
+	switch (sprn) {
+	/* 440 */
+	case SPRN_MMUCR:
+		vcpu->arch.gpr[rt] = vcpu->arch.mmucr; break;
+	case SPRN_CCR0:
+		vcpu->arch.gpr[rt] = vcpu->arch.ccr0; break;
+	case SPRN_CCR1:
+		vcpu->arch.gpr[rt] = vcpu->arch.ccr1; break;
+
+	/* Book E */
+	case SPRN_PID:
+		vcpu->arch.gpr[rt] = vcpu->arch.pid; break;
+	case SPRN_IVPR:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivpr; break;
+	case SPRN_DEAR:
+		vcpu->arch.gpr[rt] = vcpu->arch.dear; break;
+	case SPRN_ESR:
+		vcpu->arch.gpr[rt] = vcpu->arch.esr; break;
+	case SPRN_DBCR0:
+		vcpu->arch.gpr[rt] = vcpu->arch.dbcr0; break;
+	case SPRN_DBCR1:
+		vcpu->arch.gpr[rt] = vcpu->arch.dbcr1; break;
+
+	case SPRN_IVOR0:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_CRITICAL];
+		break;
+	case SPRN_IVOR1:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_MACHINE_CHECK];
+		break;
+	case SPRN_IVOR2:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DATA_STORAGE];
+		break;
+	case SPRN_IVOR3:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_INST_STORAGE];
+		break;
+	case SPRN_IVOR4:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_EXTERNAL];
+		break;
+	case SPRN_IVOR5:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_ALIGNMENT];
+		break;
+	case SPRN_IVOR6:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_PROGRAM];
+		break;
+	case SPRN_IVOR7:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_FP_UNAVAIL];
+		break;
+	case SPRN_IVOR8:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_SYSCALL];
+		break;
+	case SPRN_IVOR9:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_AP_UNAVAIL];
+		break;
+	case SPRN_IVOR10:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DECREMENTER];
+		break;
+	case SPRN_IVOR11:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_FIT];
+		break;
+	case SPRN_IVOR12:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_WATCHDOG];
+		break;
+	case SPRN_IVOR13:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DTLB_MISS];
+		break;
+	case SPRN_IVOR14:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_ITLB_MISS];
+		break;
+	case SPRN_IVOR15:
+		vcpu->arch.gpr[rt] = vcpu->arch.ivor[BOOKE_IRQPRIO_DEBUG];
+		break;
+
+	default:
+		return EMULATE_FAIL;
+	}
+
+	kvmppc_set_exit_type(vcpu, EMULATED_MFSPR_EXITS);
+	return EMULATE_DONE;
+}
+
diff --git a/arch/powerpc/kvm/44x_tlb.c b/arch/powerpc/kvm/44x_tlb.c
index ad72c6f..9a34b8e 100644
--- a/arch/powerpc/kvm/44x_tlb.c
+++ b/arch/powerpc/kvm/44x_tlb.c
@@ -22,20 +22,103 @@
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
 #include <linux/highmem.h>
+
+#include <asm/tlbflush.h>
 #include <asm/mmu-44x.h>
 #include <asm/kvm_ppc.h>
+#include <asm/kvm_44x.h>
+#include "timing.h"
 
 #include "44x_tlb.h"
 
+#ifndef PPC44x_TLBE_SIZE
+#define PPC44x_TLBE_SIZE	PPC44x_TLB_4K
+#endif
+
+#define PAGE_SIZE_4K (1<<12)
+#define PAGE_MASK_4K (~(PAGE_SIZE_4K - 1))
+
+#define PPC44x_TLB_UATTR_MASK \
+	(PPC44x_TLB_U0|PPC44x_TLB_U1|PPC44x_TLB_U2|PPC44x_TLB_U3)
 #define PPC44x_TLB_USER_PERM_MASK (PPC44x_TLB_UX|PPC44x_TLB_UR|PPC44x_TLB_UW)
 #define PPC44x_TLB_SUPER_PERM_MASK (PPC44x_TLB_SX|PPC44x_TLB_SR|PPC44x_TLB_SW)
 
-static unsigned int kvmppc_tlb_44x_pos;
+#ifdef DEBUG
+void kvmppc_dump_tlbs(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_44x_tlbe *tlbe;
+	int i;
+
+	printk("vcpu %d TLB dump:\n", vcpu->vcpu_id);
+	printk("| %2s | %3s | %8s | %8s | %8s |\n",
+			"nr", "tid", "word0", "word1", "word2");
+
+	for (i = 0; i < ARRAY_SIZE(vcpu_44x->guest_tlb); i++) {
+		tlbe = &vcpu_44x->guest_tlb[i];
+		if (tlbe->word0 & PPC44x_TLB_VALID)
+			printk(" G%2d |  %02X | %08X | %08X | %08X |\n",
+			       i, tlbe->tid, tlbe->word0, tlbe->word1,
+			       tlbe->word2);
+	}
+}
+#endif
+
+static inline void kvmppc_44x_tlbie(unsigned int index)
+{
+	/* 0 <= index < 64, so the V bit is clear and we can use the index as
+	 * word0. */
+	asm volatile(
+		"tlbwe %[index], %[index], 0\n"
+	:
+	: [index] "r"(index)
+	);
+}
+
+static inline void kvmppc_44x_tlbre(unsigned int index,
+                                    struct kvmppc_44x_tlbe *tlbe)
+{
+	asm volatile(
+		"tlbre %[word0], %[index], 0\n"
+		"mfspr %[tid], %[sprn_mmucr]\n"
+		"andi. %[tid], %[tid], 0xff\n"
+		"tlbre %[word1], %[index], 1\n"
+		"tlbre %[word2], %[index], 2\n"
+		: [word0] "=r"(tlbe->word0),
+		  [word1] "=r"(tlbe->word1),
+		  [word2] "=r"(tlbe->word2),
+		  [tid]   "=r"(tlbe->tid)
+		: [index] "r"(index),
+		  [sprn_mmucr] "i"(SPRN_MMUCR)
+		: "cc"
+	);
+}
+
+static inline void kvmppc_44x_tlbwe(unsigned int index,
+                                    struct kvmppc_44x_tlbe *stlbe)
+{
+	unsigned long tmp;
+
+	asm volatile(
+		"mfspr %[tmp], %[sprn_mmucr]\n"
+		"rlwimi %[tmp], %[tid], 0, 0xff\n"
+		"mtspr %[sprn_mmucr], %[tmp]\n"
+		"tlbwe %[word0], %[index], 0\n"
+		"tlbwe %[word1], %[index], 1\n"
+		"tlbwe %[word2], %[index], 2\n"
+		: [tmp]   "=&r"(tmp)
+		: [word0] "r"(stlbe->word0),
+		  [word1] "r"(stlbe->word1),
+		  [word2] "r"(stlbe->word2),
+		  [tid]   "r"(stlbe->tid),
+		  [index] "r"(index),
+		  [sprn_mmucr] "i"(SPRN_MMUCR)
+	);
+}
 
 static u32 kvmppc_44x_tlb_shadow_attrib(u32 attrib, int usermode)
 {
-	/* Mask off reserved bits. */
-	attrib &= PPC44x_TLB_PERM_MASK|PPC44x_TLB_ATTR_MASK;
+	/* We only care about the guest's permission and user bits. */
+	attrib &= PPC44x_TLB_PERM_MASK|PPC44x_TLB_UATTR_MASK;
 
 	if (!usermode) {
 		/* Guest is in supervisor mode, so we need to translate guest
@@ -47,18 +130,60 @@
 	/* Make sure host can always access this memory. */
 	attrib |= PPC44x_TLB_SX|PPC44x_TLB_SR|PPC44x_TLB_SW;
 
+	/* WIMGE = 0b00100 */
+	attrib |= PPC44x_TLB_M;
+
 	return attrib;
 }
 
+/* Load shadow TLB back into hardware. */
+void kvmppc_44x_tlb_load(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	int i;
+
+	for (i = 0; i <= tlb_44x_hwater; i++) {
+		struct kvmppc_44x_tlbe *stlbe = &vcpu_44x->shadow_tlb[i];
+
+		if (get_tlb_v(stlbe) && get_tlb_ts(stlbe))
+			kvmppc_44x_tlbwe(i, stlbe);
+	}
+}
+
+static void kvmppc_44x_tlbe_set_modified(struct kvmppc_vcpu_44x *vcpu_44x,
+                                         unsigned int i)
+{
+	vcpu_44x->shadow_tlb_mod[i] = 1;
+}
+
+/* Save hardware TLB to the vcpu, and invalidate all guest mappings. */
+void kvmppc_44x_tlb_put(struct kvm_vcpu *vcpu)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	int i;
+
+	for (i = 0; i <= tlb_44x_hwater; i++) {
+		struct kvmppc_44x_tlbe *stlbe = &vcpu_44x->shadow_tlb[i];
+
+		if (vcpu_44x->shadow_tlb_mod[i])
+			kvmppc_44x_tlbre(i, stlbe);
+
+		if (get_tlb_v(stlbe) && get_tlb_ts(stlbe))
+			kvmppc_44x_tlbie(i);
+	}
+}
+
+
 /* Search the guest TLB for a matching entry. */
 int kvmppc_44x_tlb_index(struct kvm_vcpu *vcpu, gva_t eaddr, unsigned int pid,
                          unsigned int as)
 {
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
 	int i;
 
 	/* XXX Replace loop with fancy data structures. */
-	for (i = 0; i < PPC44x_TLB_SIZE; i++) {
-		struct tlbe *tlbe = &vcpu->arch.guest_tlb[i];
+	for (i = 0; i < ARRAY_SIZE(vcpu_44x->guest_tlb); i++) {
+		struct kvmppc_44x_tlbe *tlbe = &vcpu_44x->guest_tlb[i];
 		unsigned int tid;
 
 		if (eaddr < get_tlb_eaddr(tlbe))
@@ -83,78 +208,89 @@
 	return -1;
 }
 
-struct tlbe *kvmppc_44x_itlb_search(struct kvm_vcpu *vcpu, gva_t eaddr)
+int kvmppc_44x_itlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
 {
 	unsigned int as = !!(vcpu->arch.msr & MSR_IS);
-	unsigned int index;
 
-	index = kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as);
-	if (index == -1)
-		return NULL;
-	return &vcpu->arch.guest_tlb[index];
+	return kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as);
 }
 
-struct tlbe *kvmppc_44x_dtlb_search(struct kvm_vcpu *vcpu, gva_t eaddr)
+int kvmppc_44x_dtlb_index(struct kvm_vcpu *vcpu, gva_t eaddr)
 {
 	unsigned int as = !!(vcpu->arch.msr & MSR_DS);
-	unsigned int index;
 
-	index = kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as);
-	if (index == -1)
-		return NULL;
-	return &vcpu->arch.guest_tlb[index];
+	return kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as);
 }
 
-static int kvmppc_44x_tlbe_is_writable(struct tlbe *tlbe)
+static void kvmppc_44x_shadow_release(struct kvmppc_vcpu_44x *vcpu_44x,
+                                      unsigned int stlb_index)
 {
-	return tlbe->word2 & (PPC44x_TLB_SW|PPC44x_TLB_UW);
-}
+	struct kvmppc_44x_shadow_ref *ref = &vcpu_44x->shadow_refs[stlb_index];
 
-static void kvmppc_44x_shadow_release(struct kvm_vcpu *vcpu,
-                                      unsigned int index)
-{
-	struct tlbe *stlbe = &vcpu->arch.shadow_tlb[index];
-	struct page *page = vcpu->arch.shadow_pages[index];
+	if (!ref->page)
+		return;
 
-	if (get_tlb_v(stlbe)) {
-		if (kvmppc_44x_tlbe_is_writable(stlbe))
-			kvm_release_page_dirty(page);
-		else
-			kvm_release_page_clean(page);
-	}
+	/* Discard from the TLB. */
+	/* Note: we could actually invalidate a host mapping, if the host overwrote
+	 * this TLB entry since we inserted a guest mapping. */
+	kvmppc_44x_tlbie(stlb_index);
+
+	/* Now release the page. */
+	if (ref->writeable)
+		kvm_release_page_dirty(ref->page);
+	else
+		kvm_release_page_clean(ref->page);
+
+	ref->page = NULL;
+
+	/* XXX set tlb_44x_index to stlb_index? */
+
+	KVMTRACE_1D(STLB_INVAL, &vcpu_44x->vcpu, stlb_index, handler);
 }
 
 void kvmppc_core_destroy_mmu(struct kvm_vcpu *vcpu)
 {
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
 	int i;
 
 	for (i = 0; i <= tlb_44x_hwater; i++)
-		kvmppc_44x_shadow_release(vcpu, i);
+		kvmppc_44x_shadow_release(vcpu_44x, i);
 }
 
-void kvmppc_tlbe_set_modified(struct kvm_vcpu *vcpu, unsigned int i)
+/**
+ * kvmppc_mmu_map -- create a host mapping for guest memory
+ *
+ * If the guest wanted a larger page than the host supports, only the first
+ * host page is mapped here and the rest are demand faulted.
+ *
+ * If the guest wanted a smaller page than the host page size, we map only the
+ * guest-size page (i.e. not a full host page mapping).
+ *
+ * Caller must ensure that the specified guest TLB entry is safe to insert into
+ * the shadow TLB.
+ */
+void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gpa_t gpaddr, u64 asid,
+                    u32 flags, u32 max_bytes, unsigned int gtlb_index)
 {
-    vcpu->arch.shadow_tlb_mod[i] = 1;
-}
-
-/* Caller must ensure that the specified guest TLB entry is safe to insert into
- * the shadow TLB. */
-void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gfn_t gfn, u64 asid,
-                    u32 flags)
-{
+	struct kvmppc_44x_tlbe stlbe;
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	struct kvmppc_44x_shadow_ref *ref;
 	struct page *new_page;
-	struct tlbe *stlbe;
 	hpa_t hpaddr;
+	gfn_t gfn;
 	unsigned int victim;
 
-	/* Future optimization: don't overwrite the TLB entry containing the
-	 * current PC (or stack?). */
-	victim = kvmppc_tlb_44x_pos++;
-	if (kvmppc_tlb_44x_pos > tlb_44x_hwater)
-		kvmppc_tlb_44x_pos = 0;
-	stlbe = &vcpu->arch.shadow_tlb[victim];
+	/* Select TLB entry to clobber. Indirectly guard against races with the TLB
+	 * miss handler by disabling interrupts. */
+	local_irq_disable();
+	victim = ++tlb_44x_index;
+	if (victim > tlb_44x_hwater)
+		victim = 0;
+	tlb_44x_index = victim;
+	local_irq_enable();
 
 	/* Get reference to new page. */
+	gfn = gpaddr >> PAGE_SHIFT;
 	new_page = gfn_to_page(vcpu->kvm, gfn);
 	if (is_error_page(new_page)) {
 		printk(KERN_ERR "Couldn't get guest page for gfn %lx!\n", gfn);
@@ -163,10 +299,8 @@
 	}
 	hpaddr = page_to_phys(new_page);
 
-	/* Drop reference to old page. */
-	kvmppc_44x_shadow_release(vcpu, victim);
-
-	vcpu->arch.shadow_pages[victim] = new_page;
+	/* Invalidate any previous shadow mappings. */
+	kvmppc_44x_shadow_release(vcpu_44x, victim);
 
 	/* XXX Make sure (va, size) doesn't overlap any other
 	 * entries. 440x6 user manual says the result would be
@@ -174,78 +308,193 @@
 
 	/* XXX what about AS? */
 
-	stlbe->tid = !(asid & 0xff);
-
 	/* Force TS=1 for all guest mappings. */
-	/* For now we hardcode 4KB mappings, but it will be important to
-	 * use host large pages in the future. */
-	stlbe->word0 = (gvaddr & PAGE_MASK) | PPC44x_TLB_VALID | PPC44x_TLB_TS
-	               | PPC44x_TLB_4K;
-	stlbe->word1 = (hpaddr & 0xfffffc00) | ((hpaddr >> 32) & 0xf);
-	stlbe->word2 = kvmppc_44x_tlb_shadow_attrib(flags,
-	                                            vcpu->arch.msr & MSR_PR);
-	kvmppc_tlbe_set_modified(vcpu, victim);
+	stlbe.word0 = PPC44x_TLB_VALID | PPC44x_TLB_TS;
 
-	KVMTRACE_5D(STLB_WRITE, vcpu, victim,
-			stlbe->tid, stlbe->word0, stlbe->word1, stlbe->word2,
-			handler);
+	if (max_bytes >= PAGE_SIZE) {
+		/* Guest mapping is larger than or equal to host page size. We can use
+		 * a "native" host mapping. */
+		stlbe.word0 |= (gvaddr & PAGE_MASK) | PPC44x_TLBE_SIZE;
+	} else {
+		/* Guest mapping is smaller than host page size. We must restrict the
+		 * size of the mapping to be at most the smaller of the two, but for
+		 * simplicity we fall back to a 4K mapping (this is probably what the
+		 * guest is using anyways). */
+		stlbe.word0 |= (gvaddr & PAGE_MASK_4K) | PPC44x_TLB_4K;
+
+		/* 'hpaddr' is a host page, which is larger than the mapping we're
+		 * inserting here. To compensate, we must add the in-page offset to the
+		 * sub-page. */
+		hpaddr |= gpaddr & (PAGE_MASK ^ PAGE_MASK_4K);
+	}
+
+	stlbe.word1 = (hpaddr & 0xfffffc00) | ((hpaddr >> 32) & 0xf);
+	stlbe.word2 = kvmppc_44x_tlb_shadow_attrib(flags,
+	                                            vcpu->arch.msr & MSR_PR);
+	stlbe.tid = !(asid & 0xff);
+
+	/* Keep track of the reference so we can properly release it later. */
+	ref = &vcpu_44x->shadow_refs[victim];
+	ref->page = new_page;
+	ref->gtlb_index = gtlb_index;
+	ref->writeable = !!(stlbe.word2 & PPC44x_TLB_UW);
+	ref->tid = stlbe.tid;
+
+	/* Insert shadow mapping into hardware TLB. */
+	kvmppc_44x_tlbe_set_modified(vcpu_44x, victim);
+	kvmppc_44x_tlbwe(victim, &stlbe);
+	KVMTRACE_5D(STLB_WRITE, vcpu, victim, stlbe.tid, stlbe.word0, stlbe.word1,
+	            stlbe.word2, handler);
 }
 
-void kvmppc_mmu_invalidate(struct kvm_vcpu *vcpu, gva_t eaddr,
-                           gva_t eend, u32 asid)
+/* For a particular guest TLB entry, invalidate the corresponding host TLB
+ * mappings and release the host pages. */
+static void kvmppc_44x_invalidate(struct kvm_vcpu *vcpu,
+                                  unsigned int gtlb_index)
 {
-	unsigned int pid = !(asid & 0xff);
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
 	int i;
 
-	/* XXX Replace loop with fancy data structures. */
-	for (i = 0; i <= tlb_44x_hwater; i++) {
-		struct tlbe *stlbe = &vcpu->arch.shadow_tlb[i];
-		unsigned int tid;
-
-		if (!get_tlb_v(stlbe))
-			continue;
-
-		if (eend < get_tlb_eaddr(stlbe))
-			continue;
-
-		if (eaddr > get_tlb_end(stlbe))
-			continue;
-
-		tid = get_tlb_tid(stlbe);
-		if (tid && (tid != pid))
-			continue;
-
-		kvmppc_44x_shadow_release(vcpu, i);
-		stlbe->word0 = 0;
-		kvmppc_tlbe_set_modified(vcpu, i);
-		KVMTRACE_5D(STLB_INVAL, vcpu, i,
-				stlbe->tid, stlbe->word0, stlbe->word1,
-				stlbe->word2, handler);
+	for (i = 0; i < ARRAY_SIZE(vcpu_44x->shadow_refs); i++) {
+		struct kvmppc_44x_shadow_ref *ref = &vcpu_44x->shadow_refs[i];
+		if (ref->gtlb_index == gtlb_index)
+			kvmppc_44x_shadow_release(vcpu_44x, i);
 	}
 }
 
-/* Invalidate all mappings on the privilege switch after PID has been changed.
- * The guest always runs with PID=1, so we must clear the entire TLB when
- * switching address spaces. */
 void kvmppc_mmu_priv_switch(struct kvm_vcpu *vcpu, int usermode)
 {
+	vcpu->arch.shadow_pid = !usermode;
+}
+
+void kvmppc_set_pid(struct kvm_vcpu *vcpu, u32 new_pid)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
 	int i;
 
-	if (vcpu->arch.swap_pid) {
-		/* XXX Replace loop with fancy data structures. */
-		for (i = 0; i <= tlb_44x_hwater; i++) {
-			struct tlbe *stlbe = &vcpu->arch.shadow_tlb[i];
+	if (unlikely(vcpu->arch.pid == new_pid))
+		return;
 
-			/* Future optimization: clear only userspace mappings. */
-			kvmppc_44x_shadow_release(vcpu, i);
-			stlbe->word0 = 0;
-			kvmppc_tlbe_set_modified(vcpu, i);
-			KVMTRACE_5D(STLB_INVAL, vcpu, i,
-			            stlbe->tid, stlbe->word0, stlbe->word1,
-			            stlbe->word2, handler);
-		}
-		vcpu->arch.swap_pid = 0;
+	vcpu->arch.pid = new_pid;
+
+	/* Guest userspace runs with TID=0 mappings and PID=0, to make sure it
+	 * can't access guest kernel mappings (TID=1). When we switch to a new
+	 * guest PID, which will also use host PID=0, we must discard the old guest
+	 * userspace mappings. */
+	for (i = 0; i < ARRAY_SIZE(vcpu_44x->shadow_refs); i++) {
+		struct kvmppc_44x_shadow_ref *ref = &vcpu_44x->shadow_refs[i];
+
+		if (ref->tid == 0)
+			kvmppc_44x_shadow_release(vcpu_44x, i);
+	}
+}
+
+static int tlbe_is_host_safe(const struct kvm_vcpu *vcpu,
+                             const struct kvmppc_44x_tlbe *tlbe)
+{
+	gpa_t gpa;
+
+	if (!get_tlb_v(tlbe))
+		return 0;
+
+	/* Does it match current guest AS? */
+	/* XXX what about IS != DS? */
+	if (get_tlb_ts(tlbe) != !!(vcpu->arch.msr & MSR_IS))
+		return 0;
+
+	gpa = get_tlb_raddr(tlbe);
+	if (!gfn_to_memslot(vcpu->kvm, gpa >> PAGE_SHIFT))
+		/* Mapping is not for RAM. */
+		return 0;
+
+	return 1;
+}
+
+int kvmppc_44x_emul_tlbwe(struct kvm_vcpu *vcpu, u8 ra, u8 rs, u8 ws)
+{
+	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+	struct kvmppc_44x_tlbe *tlbe;
+	unsigned int gtlb_index;
+
+	gtlb_index = vcpu->arch.gpr[ra];
+	if (gtlb_index > KVM44x_GUEST_TLB_SIZE) {
+		printk("%s: index %d\n", __func__, gtlb_index);
+		kvmppc_dump_vcpu(vcpu);
+		return EMULATE_FAIL;
 	}
 
-	vcpu->arch.shadow_pid = !usermode;
+	tlbe = &vcpu_44x->guest_tlb[gtlb_index];
+
+	/* Invalidate shadow mappings for the about-to-be-clobbered TLB entry. */
+	if (tlbe->word0 & PPC44x_TLB_VALID)
+		kvmppc_44x_invalidate(vcpu, gtlb_index);
+
+	switch (ws) {
+	case PPC44x_TLB_PAGEID:
+		tlbe->tid = get_mmucr_stid(vcpu);
+		tlbe->word0 = vcpu->arch.gpr[rs];
+		break;
+
+	case PPC44x_TLB_XLAT:
+		tlbe->word1 = vcpu->arch.gpr[rs];
+		break;
+
+	case PPC44x_TLB_ATTRIB:
+		tlbe->word2 = vcpu->arch.gpr[rs];
+		break;
+
+	default:
+		return EMULATE_FAIL;
+	}
+
+	if (tlbe_is_host_safe(vcpu, tlbe)) {
+		u64 asid;
+		gva_t eaddr;
+		gpa_t gpaddr;
+		u32 flags;
+		u32 bytes;
+
+		eaddr = get_tlb_eaddr(tlbe);
+		gpaddr = get_tlb_raddr(tlbe);
+
+		/* Use the advertised page size to mask effective and real addrs. */
+		bytes = get_tlb_bytes(tlbe);
+		eaddr &= ~(bytes - 1);
+		gpaddr &= ~(bytes - 1);
+
+		asid = (tlbe->word0 & PPC44x_TLB_TS) | tlbe->tid;
+		flags = tlbe->word2 & 0xffff;
+
+		kvmppc_mmu_map(vcpu, eaddr, gpaddr, asid, flags, bytes, gtlb_index);
+	}
+
+	KVMTRACE_5D(GTLB_WRITE, vcpu, gtlb_index, tlbe->tid, tlbe->word0,
+	            tlbe->word1, tlbe->word2, handler);
+
+	kvmppc_set_exit_type(vcpu, EMULATED_TLBWE_EXITS);
+	return EMULATE_DONE;
+}
+
+int kvmppc_44x_emul_tlbsx(struct kvm_vcpu *vcpu, u8 rt, u8 ra, u8 rb, u8 rc)
+{
+	u32 ea;
+	int gtlb_index;
+	unsigned int as = get_mmucr_sts(vcpu);
+	unsigned int pid = get_mmucr_stid(vcpu);
+
+	ea = vcpu->arch.gpr[rb];
+	if (ra)
+		ea += vcpu->arch.gpr[ra];
+
+	gtlb_index = kvmppc_44x_tlb_index(vcpu, ea, pid, as);
+	if (rc) {
+		if (gtlb_index < 0)
+			vcpu->arch.cr &= ~0x20000000;
+		else
+			vcpu->arch.cr |= 0x20000000;
+	}
+	vcpu->arch.gpr[rt] = gtlb_index;
+
+	kvmppc_set_exit_type(vcpu, EMULATED_TLBSX_EXITS);
+	return EMULATE_DONE;
 }
diff --git a/arch/powerpc/kvm/44x_tlb.h b/arch/powerpc/kvm/44x_tlb.h
index 2ccd46b..772191f 100644
--- a/arch/powerpc/kvm/44x_tlb.h
+++ b/arch/powerpc/kvm/44x_tlb.h
@@ -25,48 +25,52 @@
 
 extern int kvmppc_44x_tlb_index(struct kvm_vcpu *vcpu, gva_t eaddr,
                                 unsigned int pid, unsigned int as);
-extern struct tlbe *kvmppc_44x_dtlb_search(struct kvm_vcpu *vcpu, gva_t eaddr);
-extern struct tlbe *kvmppc_44x_itlb_search(struct kvm_vcpu *vcpu, gva_t eaddr);
+extern int kvmppc_44x_dtlb_index(struct kvm_vcpu *vcpu, gva_t eaddr);
+extern int kvmppc_44x_itlb_index(struct kvm_vcpu *vcpu, gva_t eaddr);
+
+extern int kvmppc_44x_emul_tlbsx(struct kvm_vcpu *vcpu, u8 rt, u8 ra, u8 rb,
+                                 u8 rc);
+extern int kvmppc_44x_emul_tlbwe(struct kvm_vcpu *vcpu, u8 ra, u8 rs, u8 ws);
 
 /* TLB helper functions */
-static inline unsigned int get_tlb_size(const struct tlbe *tlbe)
+static inline unsigned int get_tlb_size(const struct kvmppc_44x_tlbe *tlbe)
 {
 	return (tlbe->word0 >> 4) & 0xf;
 }
 
-static inline gva_t get_tlb_eaddr(const struct tlbe *tlbe)
+static inline gva_t get_tlb_eaddr(const struct kvmppc_44x_tlbe *tlbe)
 {
 	return tlbe->word0 & 0xfffffc00;
 }
 
-static inline gva_t get_tlb_bytes(const struct tlbe *tlbe)
+static inline gva_t get_tlb_bytes(const struct kvmppc_44x_tlbe *tlbe)
 {
 	unsigned int pgsize = get_tlb_size(tlbe);
 	return 1 << 10 << (pgsize << 1);
 }
 
-static inline gva_t get_tlb_end(const struct tlbe *tlbe)
+static inline gva_t get_tlb_end(const struct kvmppc_44x_tlbe *tlbe)
 {
 	return get_tlb_eaddr(tlbe) + get_tlb_bytes(tlbe) - 1;
 }
 
-static inline u64 get_tlb_raddr(const struct tlbe *tlbe)
+static inline u64 get_tlb_raddr(const struct kvmppc_44x_tlbe *tlbe)
 {
 	u64 word1 = tlbe->word1;
 	return ((word1 & 0xf) << 32) | (word1 & 0xfffffc00);
 }
 
-static inline unsigned int get_tlb_tid(const struct tlbe *tlbe)
+static inline unsigned int get_tlb_tid(const struct kvmppc_44x_tlbe *tlbe)
 {
 	return tlbe->tid & 0xff;
 }
 
-static inline unsigned int get_tlb_ts(const struct tlbe *tlbe)
+static inline unsigned int get_tlb_ts(const struct kvmppc_44x_tlbe *tlbe)
 {
 	return (tlbe->word0 >> 8) & 0x1;
 }
 
-static inline unsigned int get_tlb_v(const struct tlbe *tlbe)
+static inline unsigned int get_tlb_v(const struct kvmppc_44x_tlbe *tlbe)
 {
 	return (tlbe->word0 >> 9) & 0x1;
 }
@@ -81,7 +85,7 @@
 	return (vcpu->arch.mmucr >> 16) & 0x1;
 }
 
-static inline gpa_t tlb_xlate(struct tlbe *tlbe, gva_t eaddr)
+static inline gpa_t tlb_xlate(struct kvmppc_44x_tlbe *tlbe, gva_t eaddr)
 {
 	unsigned int pgmask = get_tlb_bytes(tlbe) - 1;
 
diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
index 53aaa66..6dbdc48 100644
--- a/arch/powerpc/kvm/Kconfig
+++ b/arch/powerpc/kvm/Kconfig
@@ -15,27 +15,33 @@
 if VIRTUALIZATION
 
 config KVM
-	bool "Kernel-based Virtual Machine (KVM) support"
-	depends on 44x && EXPERIMENTAL
+	bool
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
-	# We can only run on Book E hosts so far
-	select KVM_BOOKE_HOST
+
+config KVM_440
+	bool "KVM support for PowerPC 440 processors"
+	depends on EXPERIMENTAL && 44x
+	select KVM
 	---help---
-	  Support hosting virtualized guest machines. You will also
-	  need to select one or more of the processor modules below.
+	  Support running unmodified 440 guest kernels in virtual machines on
+	  440 host processors.
 
 	  This module provides access to the hardware capabilities through
 	  a character device node named /dev/kvm.
 
 	  If unsure, say N.
 
-config KVM_BOOKE_HOST
-	bool "KVM host support for Book E PowerPC processors"
-	depends on KVM && 44x
+config KVM_EXIT_TIMING
+	bool "Detailed exit timing"
+	depends on KVM
 	---help---
-	  Provides host support for KVM on Book E PowerPC processors. Currently
-	  this works on 440 processors only.
+	  Calculate elapsed time for every exit/enter cycle. A per-vcpu
+	  report is available in debugfs kvm/vm#_vcpu#_timing.
+	  The overhead is relatively small, however it is not recommended for
+	  production environments.
+
+	  If unsure, say N.
 
 config KVM_TRACE
 	bool "KVM trace support"
diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
index 2a5d439..df7ba59 100644
--- a/arch/powerpc/kvm/Makefile
+++ b/arch/powerpc/kvm/Makefile
@@ -8,10 +8,16 @@
 
 common-objs-$(CONFIG_KVM_TRACE)  += $(addprefix ../../../virt/kvm/, kvm_trace.o)
 
-kvm-objs := $(common-objs-y) powerpc.o emulate.o booke_guest.o
+kvm-objs := $(common-objs-y) powerpc.o emulate.o
+obj-$(CONFIG_KVM_EXIT_TIMING) += timing.o
 obj-$(CONFIG_KVM) += kvm.o
 
 AFLAGS_booke_interrupts.o := -I$(obj)
 
-kvm-booke-host-objs := booke_host.o booke_interrupts.o 44x_tlb.o
-obj-$(CONFIG_KVM_BOOKE_HOST) += kvm-booke-host.o
+kvm-440-objs := \
+	booke.o \
+	booke_interrupts.o \
+	44x.o \
+	44x_tlb.o \
+	44x_emulate.o
+obj-$(CONFIG_KVM_440) += kvm-440.o
diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c
new file mode 100644
index 0000000..35485dd
--- /dev/null
+++ b/arch/powerpc/kvm/booke.c
@@ -0,0 +1,565 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2007
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ *          Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
+ */
+
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/kvm_host.h>
+#include <linux/module.h>
+#include <linux/vmalloc.h>
+#include <linux/fs.h>
+
+#include <asm/cputable.h>
+#include <asm/uaccess.h>
+#include <asm/kvm_ppc.h>
+#include "timing.h"
+#include <asm/cacheflush.h>
+#include <asm/kvm_44x.h>
+
+#include "booke.h"
+#include "44x_tlb.h"
+
+unsigned long kvmppc_booke_handlers;
+
+#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM
+#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
+
+struct kvm_stats_debugfs_item debugfs_entries[] = {
+	{ "mmio",       VCPU_STAT(mmio_exits) },
+	{ "dcr",        VCPU_STAT(dcr_exits) },
+	{ "sig",        VCPU_STAT(signal_exits) },
+	{ "itlb_r",     VCPU_STAT(itlb_real_miss_exits) },
+	{ "itlb_v",     VCPU_STAT(itlb_virt_miss_exits) },
+	{ "dtlb_r",     VCPU_STAT(dtlb_real_miss_exits) },
+	{ "dtlb_v",     VCPU_STAT(dtlb_virt_miss_exits) },
+	{ "sysc",       VCPU_STAT(syscall_exits) },
+	{ "isi",        VCPU_STAT(isi_exits) },
+	{ "dsi",        VCPU_STAT(dsi_exits) },
+	{ "inst_emu",   VCPU_STAT(emulated_inst_exits) },
+	{ "dec",        VCPU_STAT(dec_exits) },
+	{ "ext_intr",   VCPU_STAT(ext_intr_exits) },
+	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
+	{ NULL }
+};
+
+/* TODO: use vcpu_printf() */
+void kvmppc_dump_vcpu(struct kvm_vcpu *vcpu)
+{
+	int i;
+
+	printk("pc:   %08lx msr:  %08lx\n", vcpu->arch.pc, vcpu->arch.msr);
+	printk("lr:   %08lx ctr:  %08lx\n", vcpu->arch.lr, vcpu->arch.ctr);
+	printk("srr0: %08lx srr1: %08lx\n", vcpu->arch.srr0, vcpu->arch.srr1);
+
+	printk("exceptions: %08lx\n", vcpu->arch.pending_exceptions);
+
+	for (i = 0; i < 32; i += 4) {
+		printk("gpr%02d: %08lx %08lx %08lx %08lx\n", i,
+		       vcpu->arch.gpr[i],
+		       vcpu->arch.gpr[i+1],
+		       vcpu->arch.gpr[i+2],
+		       vcpu->arch.gpr[i+3]);
+	}
+}
+
+static void kvmppc_booke_queue_irqprio(struct kvm_vcpu *vcpu,
+                                       unsigned int priority)
+{
+	set_bit(priority, &vcpu->arch.pending_exceptions);
+}
+
+void kvmppc_core_queue_program(struct kvm_vcpu *vcpu)
+{
+	kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_PROGRAM);
+}
+
+void kvmppc_core_queue_dec(struct kvm_vcpu *vcpu)
+{
+	kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_DECREMENTER);
+}
+
+int kvmppc_core_pending_dec(struct kvm_vcpu *vcpu)
+{
+	return test_bit(BOOKE_IRQPRIO_DECREMENTER, &vcpu->arch.pending_exceptions);
+}
+
+void kvmppc_core_queue_external(struct kvm_vcpu *vcpu,
+                                struct kvm_interrupt *irq)
+{
+	kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_EXTERNAL);
+}
+
+/* Deliver the interrupt of the corresponding priority, if possible. */
+static int kvmppc_booke_irqprio_deliver(struct kvm_vcpu *vcpu,
+                                        unsigned int priority)
+{
+	int allowed = 0;
+	ulong msr_mask;
+
+	switch (priority) {
+	case BOOKE_IRQPRIO_PROGRAM:
+	case BOOKE_IRQPRIO_DTLB_MISS:
+	case BOOKE_IRQPRIO_ITLB_MISS:
+	case BOOKE_IRQPRIO_SYSCALL:
+	case BOOKE_IRQPRIO_DATA_STORAGE:
+	case BOOKE_IRQPRIO_INST_STORAGE:
+	case BOOKE_IRQPRIO_FP_UNAVAIL:
+	case BOOKE_IRQPRIO_AP_UNAVAIL:
+	case BOOKE_IRQPRIO_ALIGNMENT:
+		allowed = 1;
+		msr_mask = MSR_CE|MSR_ME|MSR_DE;
+		break;
+	case BOOKE_IRQPRIO_CRITICAL:
+	case BOOKE_IRQPRIO_WATCHDOG:
+		allowed = vcpu->arch.msr & MSR_CE;
+		msr_mask = MSR_ME;
+		break;
+	case BOOKE_IRQPRIO_MACHINE_CHECK:
+		allowed = vcpu->arch.msr & MSR_ME;
+		msr_mask = 0;
+		break;
+	case BOOKE_IRQPRIO_EXTERNAL:
+	case BOOKE_IRQPRIO_DECREMENTER:
+	case BOOKE_IRQPRIO_FIT:
+		allowed = vcpu->arch.msr & MSR_EE;
+		msr_mask = MSR_CE|MSR_ME|MSR_DE;
+		break;
+	case BOOKE_IRQPRIO_DEBUG:
+		allowed = vcpu->arch.msr & MSR_DE;
+		msr_mask = MSR_ME;
+		break;
+	}
+
+	if (allowed) {
+		vcpu->arch.srr0 = vcpu->arch.pc;
+		vcpu->arch.srr1 = vcpu->arch.msr;
+		vcpu->arch.pc = vcpu->arch.ivpr | vcpu->arch.ivor[priority];
+		kvmppc_set_msr(vcpu, vcpu->arch.msr & msr_mask);
+
+		clear_bit(priority, &vcpu->arch.pending_exceptions);
+	}
+
+	return allowed;
+}
+
+/* Check pending exceptions and deliver one, if possible. */
+void kvmppc_core_deliver_interrupts(struct kvm_vcpu *vcpu)
+{
+	unsigned long *pending = &vcpu->arch.pending_exceptions;
+	unsigned int priority;
+
+	priority = __ffs(*pending);
+	while (priority <= BOOKE_MAX_INTERRUPT) {
+		if (kvmppc_booke_irqprio_deliver(vcpu, priority))
+			break;
+
+		priority = find_next_bit(pending,
+		                         BITS_PER_BYTE * sizeof(*pending),
+		                         priority + 1);
+	}
+}
+
+/**
+ * kvmppc_handle_exit
+ *
+ * Return value is in the form (errcode<<2 | RESUME_FLAG_HOST | RESUME_FLAG_NV)
+ */
+int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
+                       unsigned int exit_nr)
+{
+	enum emulation_result er;
+	int r = RESUME_HOST;
+
+	/* update before a new last_exit_type is rewritten */
+	kvmppc_update_timing_stats(vcpu);
+
+	local_irq_enable();
+
+	run->exit_reason = KVM_EXIT_UNKNOWN;
+	run->ready_for_interrupt_injection = 1;
+
+	switch (exit_nr) {
+	case BOOKE_INTERRUPT_MACHINE_CHECK:
+		printk("MACHINE CHECK: %lx\n", mfspr(SPRN_MCSR));
+		kvmppc_dump_vcpu(vcpu);
+		r = RESUME_HOST;
+		break;
+
+	case BOOKE_INTERRUPT_EXTERNAL:
+		kvmppc_account_exit(vcpu, EXT_INTR_EXITS);
+		if (need_resched())
+			cond_resched();
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_DECREMENTER:
+		/* Since we switched IVPR back to the host's value, the host
+		 * handled this interrupt the moment we enabled interrupts.
+		 * Now we just offer it a chance to reschedule the guest. */
+		kvmppc_account_exit(vcpu, DEC_EXITS);
+		if (need_resched())
+			cond_resched();
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_PROGRAM:
+		if (vcpu->arch.msr & MSR_PR) {
+			/* Program traps generated by user-level software must be handled
+			 * by the guest kernel. */
+			vcpu->arch.esr = vcpu->arch.fault_esr;
+			kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_PROGRAM);
+			r = RESUME_GUEST;
+			kvmppc_account_exit(vcpu, USR_PR_INST);
+			break;
+		}
+
+		er = kvmppc_emulate_instruction(run, vcpu);
+		switch (er) {
+		case EMULATE_DONE:
+			/* don't overwrite subtypes, just account kvm_stats */
+			kvmppc_account_exit_stat(vcpu, EMULATED_INST_EXITS);
+			/* Future optimization: only reload non-volatiles if
+			 * they were actually modified by emulation. */
+			r = RESUME_GUEST_NV;
+			break;
+		case EMULATE_DO_DCR:
+			run->exit_reason = KVM_EXIT_DCR;
+			r = RESUME_HOST;
+			break;
+		case EMULATE_FAIL:
+			/* XXX Deliver Program interrupt to guest. */
+			printk(KERN_CRIT "%s: emulation at %lx failed (%08x)\n",
+			       __func__, vcpu->arch.pc, vcpu->arch.last_inst);
+			/* For debugging, encode the failing instruction and
+			 * report it to userspace. */
+			run->hw.hardware_exit_reason = ~0ULL << 32;
+			run->hw.hardware_exit_reason |= vcpu->arch.last_inst;
+			r = RESUME_HOST;
+			break;
+		default:
+			BUG();
+		}
+		break;
+
+	case BOOKE_INTERRUPT_FP_UNAVAIL:
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_FP_UNAVAIL);
+		kvmppc_account_exit(vcpu, FP_UNAVAIL);
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_DATA_STORAGE:
+		vcpu->arch.dear = vcpu->arch.fault_dear;
+		vcpu->arch.esr = vcpu->arch.fault_esr;
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_DATA_STORAGE);
+		kvmppc_account_exit(vcpu, DSI_EXITS);
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_INST_STORAGE:
+		vcpu->arch.esr = vcpu->arch.fault_esr;
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_INST_STORAGE);
+		kvmppc_account_exit(vcpu, ISI_EXITS);
+		r = RESUME_GUEST;
+		break;
+
+	case BOOKE_INTERRUPT_SYSCALL:
+		kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_SYSCALL);
+		kvmppc_account_exit(vcpu, SYSCALL_EXITS);
+		r = RESUME_GUEST;
+		break;
+
+	/* XXX move to a 440-specific file. */
+	case BOOKE_INTERRUPT_DTLB_MISS: {
+		struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+		struct kvmppc_44x_tlbe *gtlbe;
+		unsigned long eaddr = vcpu->arch.fault_dear;
+		int gtlb_index;
+		gfn_t gfn;
+
+		/* Check the guest TLB. */
+		gtlb_index = kvmppc_44x_dtlb_index(vcpu, eaddr);
+		if (gtlb_index < 0) {
+			/* The guest didn't have a mapping for it. */
+			kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_DTLB_MISS);
+			vcpu->arch.dear = vcpu->arch.fault_dear;
+			vcpu->arch.esr = vcpu->arch.fault_esr;
+			kvmppc_account_exit(vcpu, DTLB_REAL_MISS_EXITS);
+			r = RESUME_GUEST;
+			break;
+		}
+
+		gtlbe = &vcpu_44x->guest_tlb[gtlb_index];
+		vcpu->arch.paddr_accessed = tlb_xlate(gtlbe, eaddr);
+		gfn = vcpu->arch.paddr_accessed >> PAGE_SHIFT;
+
+		if (kvm_is_visible_gfn(vcpu->kvm, gfn)) {
+			/* The guest TLB had a mapping, but the shadow TLB
+			 * didn't, and it is RAM. This could be because:
+			 * a) the entry is mapping the host kernel, or
+			 * b) the guest used a large mapping which we're faking
+			 * Either way, we need to satisfy the fault without
+			 * invoking the guest. */
+			kvmppc_mmu_map(vcpu, eaddr, vcpu->arch.paddr_accessed, gtlbe->tid,
+			               gtlbe->word2, get_tlb_bytes(gtlbe), gtlb_index);
+			kvmppc_account_exit(vcpu, DTLB_VIRT_MISS_EXITS);
+			r = RESUME_GUEST;
+		} else {
+			/* Guest has mapped and accessed a page which is not
+			 * actually RAM. */
+			r = kvmppc_emulate_mmio(run, vcpu);
+			kvmppc_account_exit(vcpu, MMIO_EXITS);
+		}
+
+		break;
+	}
+
+	/* XXX move to a 440-specific file. */
+	case BOOKE_INTERRUPT_ITLB_MISS: {
+		struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
+		struct kvmppc_44x_tlbe *gtlbe;
+		unsigned long eaddr = vcpu->arch.pc;
+		gpa_t gpaddr;
+		gfn_t gfn;
+		int gtlb_index;
+
+		r = RESUME_GUEST;
+
+		/* Check the guest TLB. */
+		gtlb_index = kvmppc_44x_itlb_index(vcpu, eaddr);
+		if (gtlb_index < 0) {
+			/* The guest didn't have a mapping for it. */
+			kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_ITLB_MISS);
+			kvmppc_account_exit(vcpu, ITLB_REAL_MISS_EXITS);
+			break;
+		}
+
+		kvmppc_account_exit(vcpu, ITLB_VIRT_MISS_EXITS);
+
+		gtlbe = &vcpu_44x->guest_tlb[gtlb_index];
+		gpaddr = tlb_xlate(gtlbe, eaddr);
+		gfn = gpaddr >> PAGE_SHIFT;
+
+		if (kvm_is_visible_gfn(vcpu->kvm, gfn)) {
+			/* The guest TLB had a mapping, but the shadow TLB
+			 * didn't. This could be because:
+			 * a) the entry is mapping the host kernel, or
+			 * b) the guest used a large mapping which we're faking
+			 * Either way, we need to satisfy the fault without
+			 * invoking the guest. */
+			kvmppc_mmu_map(vcpu, eaddr, gpaddr, gtlbe->tid,
+			               gtlbe->word2, get_tlb_bytes(gtlbe), gtlb_index);
+		} else {
+			/* Guest mapped and leaped at non-RAM! */
+			kvmppc_booke_queue_irqprio(vcpu, BOOKE_IRQPRIO_MACHINE_CHECK);
+		}
+
+		break;
+	}
+
+	case BOOKE_INTERRUPT_DEBUG: {
+		u32 dbsr;
+
+		vcpu->arch.pc = mfspr(SPRN_CSRR0);
+
+		/* clear IAC events in DBSR register */
+		dbsr = mfspr(SPRN_DBSR);
+		dbsr &= DBSR_IAC1 | DBSR_IAC2 | DBSR_IAC3 | DBSR_IAC4;
+		mtspr(SPRN_DBSR, dbsr);
+
+		run->exit_reason = KVM_EXIT_DEBUG;
+		kvmppc_account_exit(vcpu, DEBUG_EXITS);
+		r = RESUME_HOST;
+		break;
+	}
+
+	default:
+		printk(KERN_EMERG "exit_nr %d\n", exit_nr);
+		BUG();
+	}
+
+	local_irq_disable();
+
+	kvmppc_core_deliver_interrupts(vcpu);
+
+	if (!(r & RESUME_HOST)) {
+		/* To avoid clobbering exit_reason, only check for signals if
+		 * we aren't already exiting to userspace for some other
+		 * reason. */
+		if (signal_pending(current)) {
+			run->exit_reason = KVM_EXIT_INTR;
+			r = (-EINTR << 2) | RESUME_HOST | (r & RESUME_FLAG_NV);
+			kvmppc_account_exit(vcpu, SIGNAL_EXITS);
+		}
+	}
+
+	return r;
+}
+
+/* Initial guest state: 16MB mapping 0 -> 0, PC = 0, MSR = 0, R1 = 16MB */
+int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.pc = 0;
+	vcpu->arch.msr = 0;
+	vcpu->arch.gpr[1] = (16<<20) - 8; /* -8 for the callee-save LR slot */
+
+	vcpu->arch.shadow_pid = 1;
+
+	/* Eye-catching number so we know if the guest takes an interrupt
+	 * before it's programmed its own IVPR. */
+	vcpu->arch.ivpr = 0x55550000;
+
+	kvmppc_init_timing_stats(vcpu);
+
+	return kvmppc_core_vcpu_setup(vcpu);
+}
+
+int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
+{
+	int i;
+
+	regs->pc = vcpu->arch.pc;
+	regs->cr = vcpu->arch.cr;
+	regs->ctr = vcpu->arch.ctr;
+	regs->lr = vcpu->arch.lr;
+	regs->xer = vcpu->arch.xer;
+	regs->msr = vcpu->arch.msr;
+	regs->srr0 = vcpu->arch.srr0;
+	regs->srr1 = vcpu->arch.srr1;
+	regs->pid = vcpu->arch.pid;
+	regs->sprg0 = vcpu->arch.sprg0;
+	regs->sprg1 = vcpu->arch.sprg1;
+	regs->sprg2 = vcpu->arch.sprg2;
+	regs->sprg3 = vcpu->arch.sprg3;
+	regs->sprg5 = vcpu->arch.sprg4;
+	regs->sprg6 = vcpu->arch.sprg5;
+	regs->sprg7 = vcpu->arch.sprg6;
+
+	for (i = 0; i < ARRAY_SIZE(regs->gpr); i++)
+		regs->gpr[i] = vcpu->arch.gpr[i];
+
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
+{
+	int i;
+
+	vcpu->arch.pc = regs->pc;
+	vcpu->arch.cr = regs->cr;
+	vcpu->arch.ctr = regs->ctr;
+	vcpu->arch.lr = regs->lr;
+	vcpu->arch.xer = regs->xer;
+	kvmppc_set_msr(vcpu, regs->msr);
+	vcpu->arch.srr0 = regs->srr0;
+	vcpu->arch.srr1 = regs->srr1;
+	vcpu->arch.sprg0 = regs->sprg0;
+	vcpu->arch.sprg1 = regs->sprg1;
+	vcpu->arch.sprg2 = regs->sprg2;
+	vcpu->arch.sprg3 = regs->sprg3;
+	vcpu->arch.sprg5 = regs->sprg4;
+	vcpu->arch.sprg6 = regs->sprg5;
+	vcpu->arch.sprg7 = regs->sprg6;
+
+	for (i = 0; i < ARRAY_SIZE(vcpu->arch.gpr); i++)
+		vcpu->arch.gpr[i] = regs->gpr[i];
+
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
+                                  struct kvm_sregs *sregs)
+{
+	return -ENOTSUPP;
+}
+
+int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
+                                  struct kvm_sregs *sregs)
+{
+	return -ENOTSUPP;
+}
+
+int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
+{
+	return -ENOTSUPP;
+}
+
+int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
+{
+	return -ENOTSUPP;
+}
+
+int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
+                                  struct kvm_translation *tr)
+{
+	return kvmppc_core_vcpu_translate(vcpu, tr);
+}
+
+int kvmppc_booke_init(void)
+{
+	unsigned long ivor[16];
+	unsigned long max_ivor = 0;
+	int i;
+
+	/* We install our own exception handlers by hijacking IVPR. IVPR must
+	 * be 16-bit aligned, so we need a 64KB allocation. */
+	kvmppc_booke_handlers = __get_free_pages(GFP_KERNEL | __GFP_ZERO,
+	                                         VCPU_SIZE_ORDER);
+	if (!kvmppc_booke_handlers)
+		return -ENOMEM;
+
+	/* XXX make sure our handlers are smaller than Linux's */
+
+	/* Copy our interrupt handlers to match host IVORs. That way we don't
+	 * have to swap the IVORs on every guest/host transition. */
+	ivor[0] = mfspr(SPRN_IVOR0);
+	ivor[1] = mfspr(SPRN_IVOR1);
+	ivor[2] = mfspr(SPRN_IVOR2);
+	ivor[3] = mfspr(SPRN_IVOR3);
+	ivor[4] = mfspr(SPRN_IVOR4);
+	ivor[5] = mfspr(SPRN_IVOR5);
+	ivor[6] = mfspr(SPRN_IVOR6);
+	ivor[7] = mfspr(SPRN_IVOR7);
+	ivor[8] = mfspr(SPRN_IVOR8);
+	ivor[9] = mfspr(SPRN_IVOR9);
+	ivor[10] = mfspr(SPRN_IVOR10);
+	ivor[11] = mfspr(SPRN_IVOR11);
+	ivor[12] = mfspr(SPRN_IVOR12);
+	ivor[13] = mfspr(SPRN_IVOR13);
+	ivor[14] = mfspr(SPRN_IVOR14);
+	ivor[15] = mfspr(SPRN_IVOR15);
+
+	for (i = 0; i < 16; i++) {
+		if (ivor[i] > max_ivor)
+			max_ivor = ivor[i];
+
+		memcpy((void *)kvmppc_booke_handlers + ivor[i],
+		       kvmppc_handlers_start + i * kvmppc_handler_len,
+		       kvmppc_handler_len);
+	}
+	flush_icache_range(kvmppc_booke_handlers,
+	                   kvmppc_booke_handlers + max_ivor + kvmppc_handler_len);
+
+	return 0;
+}
+
+void __exit kvmppc_booke_exit(void)
+{
+	free_pages(kvmppc_booke_handlers, VCPU_SIZE_ORDER);
+	kvm_exit();
+}
diff --git a/arch/powerpc/kvm/booke.h b/arch/powerpc/kvm/booke.h
new file mode 100644
index 0000000..cf7c94c
--- /dev/null
+++ b/arch/powerpc/kvm/booke.h
@@ -0,0 +1,60 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#ifndef __KVM_BOOKE_H__
+#define __KVM_BOOKE_H__
+
+#include <linux/types.h>
+#include <linux/kvm_host.h>
+#include "timing.h"
+
+/* interrupt priortity ordering */
+#define BOOKE_IRQPRIO_DATA_STORAGE 0
+#define BOOKE_IRQPRIO_INST_STORAGE 1
+#define BOOKE_IRQPRIO_ALIGNMENT 2
+#define BOOKE_IRQPRIO_PROGRAM 3
+#define BOOKE_IRQPRIO_FP_UNAVAIL 4
+#define BOOKE_IRQPRIO_SYSCALL 5
+#define BOOKE_IRQPRIO_AP_UNAVAIL 6
+#define BOOKE_IRQPRIO_DTLB_MISS 7
+#define BOOKE_IRQPRIO_ITLB_MISS 8
+#define BOOKE_IRQPRIO_MACHINE_CHECK 9
+#define BOOKE_IRQPRIO_DEBUG 10
+#define BOOKE_IRQPRIO_CRITICAL 11
+#define BOOKE_IRQPRIO_WATCHDOG 12
+#define BOOKE_IRQPRIO_EXTERNAL 13
+#define BOOKE_IRQPRIO_FIT 14
+#define BOOKE_IRQPRIO_DECREMENTER 15
+
+/* Helper function for "full" MSR writes. No need to call this if only EE is
+ * changing. */
+static inline void kvmppc_set_msr(struct kvm_vcpu *vcpu, u32 new_msr)
+{
+	if ((new_msr & MSR_PR) != (vcpu->arch.msr & MSR_PR))
+		kvmppc_mmu_priv_switch(vcpu, new_msr & MSR_PR);
+
+	vcpu->arch.msr = new_msr;
+
+	if (vcpu->arch.msr & MSR_WE) {
+		kvm_vcpu_block(vcpu);
+		kvmppc_set_exit_type(vcpu, EMULATED_MTMSRWE_EXITS);
+	};
+}
+
+#endif /* __KVM_BOOKE_H__ */
diff --git a/arch/powerpc/kvm/booke_guest.c b/arch/powerpc/kvm/booke_guest.c
deleted file mode 100644
index 7b2591e..0000000
--- a/arch/powerpc/kvm/booke_guest.c
+++ /dev/null
@@ -1,605 +0,0 @@
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License, version 2, as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
- *
- * Copyright IBM Corp. 2007
- *
- * Authors: Hollis Blanchard <hollisb@us.ibm.com>
- *          Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
- */
-
-#include <linux/errno.h>
-#include <linux/err.h>
-#include <linux/kvm_host.h>
-#include <linux/module.h>
-#include <linux/vmalloc.h>
-#include <linux/fs.h>
-#include <asm/cputable.h>
-#include <asm/uaccess.h>
-#include <asm/kvm_ppc.h>
-
-#include "44x_tlb.h"
-
-#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM
-#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
-
-struct kvm_stats_debugfs_item debugfs_entries[] = {
-	{ "exits",      VCPU_STAT(sum_exits) },
-	{ "mmio",       VCPU_STAT(mmio_exits) },
-	{ "dcr",        VCPU_STAT(dcr_exits) },
-	{ "sig",        VCPU_STAT(signal_exits) },
-	{ "light",      VCPU_STAT(light_exits) },
-	{ "itlb_r",     VCPU_STAT(itlb_real_miss_exits) },
-	{ "itlb_v",     VCPU_STAT(itlb_virt_miss_exits) },
-	{ "dtlb_r",     VCPU_STAT(dtlb_real_miss_exits) },
-	{ "dtlb_v",     VCPU_STAT(dtlb_virt_miss_exits) },
-	{ "sysc",       VCPU_STAT(syscall_exits) },
-	{ "isi",        VCPU_STAT(isi_exits) },
-	{ "dsi",        VCPU_STAT(dsi_exits) },
-	{ "inst_emu",   VCPU_STAT(emulated_inst_exits) },
-	{ "dec",        VCPU_STAT(dec_exits) },
-	{ "ext_intr",   VCPU_STAT(ext_intr_exits) },
-	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
-	{ NULL }
-};
-
-static const u32 interrupt_msr_mask[16] = {
-	[BOOKE_INTERRUPT_CRITICAL]      = MSR_ME,
-	[BOOKE_INTERRUPT_MACHINE_CHECK] = 0,
-	[BOOKE_INTERRUPT_DATA_STORAGE]  = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_INST_STORAGE]  = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_EXTERNAL]      = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_ALIGNMENT]     = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_PROGRAM]       = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_FP_UNAVAIL]    = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_SYSCALL]       = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_AP_UNAVAIL]    = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_DECREMENTER]   = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_FIT]           = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_WATCHDOG]      = MSR_ME,
-	[BOOKE_INTERRUPT_DTLB_MISS]     = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_ITLB_MISS]     = MSR_CE|MSR_ME|MSR_DE,
-	[BOOKE_INTERRUPT_DEBUG]         = MSR_ME,
-};
-
-const unsigned char exception_priority[] = {
-	[BOOKE_INTERRUPT_DATA_STORAGE] = 0,
-	[BOOKE_INTERRUPT_INST_STORAGE] = 1,
-	[BOOKE_INTERRUPT_ALIGNMENT] = 2,
-	[BOOKE_INTERRUPT_PROGRAM] = 3,
-	[BOOKE_INTERRUPT_FP_UNAVAIL] = 4,
-	[BOOKE_INTERRUPT_SYSCALL] = 5,
-	[BOOKE_INTERRUPT_AP_UNAVAIL] = 6,
-	[BOOKE_INTERRUPT_DTLB_MISS] = 7,
-	[BOOKE_INTERRUPT_ITLB_MISS] = 8,
-	[BOOKE_INTERRUPT_MACHINE_CHECK] = 9,
-	[BOOKE_INTERRUPT_DEBUG] = 10,
-	[BOOKE_INTERRUPT_CRITICAL] = 11,
-	[BOOKE_INTERRUPT_WATCHDOG] = 12,
-	[BOOKE_INTERRUPT_EXTERNAL] = 13,
-	[BOOKE_INTERRUPT_FIT] = 14,
-	[BOOKE_INTERRUPT_DECREMENTER] = 15,
-};
-
-const unsigned char priority_exception[] = {
-	BOOKE_INTERRUPT_DATA_STORAGE,
-	BOOKE_INTERRUPT_INST_STORAGE,
-	BOOKE_INTERRUPT_ALIGNMENT,
-	BOOKE_INTERRUPT_PROGRAM,
-	BOOKE_INTERRUPT_FP_UNAVAIL,
-	BOOKE_INTERRUPT_SYSCALL,
-	BOOKE_INTERRUPT_AP_UNAVAIL,
-	BOOKE_INTERRUPT_DTLB_MISS,
-	BOOKE_INTERRUPT_ITLB_MISS,
-	BOOKE_INTERRUPT_MACHINE_CHECK,
-	BOOKE_INTERRUPT_DEBUG,
-	BOOKE_INTERRUPT_CRITICAL,
-	BOOKE_INTERRUPT_WATCHDOG,
-	BOOKE_INTERRUPT_EXTERNAL,
-	BOOKE_INTERRUPT_FIT,
-	BOOKE_INTERRUPT_DECREMENTER,
-};
-
-
-void kvmppc_dump_tlbs(struct kvm_vcpu *vcpu)
-{
-	struct tlbe *tlbe;
-	int i;
-
-	printk("vcpu %d TLB dump:\n", vcpu->vcpu_id);
-	printk("| %2s | %3s | %8s | %8s | %8s |\n",
-			"nr", "tid", "word0", "word1", "word2");
-
-	for (i = 0; i < PPC44x_TLB_SIZE; i++) {
-		tlbe = &vcpu->arch.guest_tlb[i];
-		if (tlbe->word0 & PPC44x_TLB_VALID)
-			printk(" G%2d |  %02X | %08X | %08X | %08X |\n",
-			       i, tlbe->tid, tlbe->word0, tlbe->word1,
-			       tlbe->word2);
-	}
-
-	for (i = 0; i < PPC44x_TLB_SIZE; i++) {
-		tlbe = &vcpu->arch.shadow_tlb[i];
-		if (tlbe->word0 & PPC44x_TLB_VALID)
-			printk(" S%2d | %02X | %08X | %08X | %08X |\n",
-			       i, tlbe->tid, tlbe->word0, tlbe->word1,
-			       tlbe->word2);
-	}
-}
-
-/* TODO: use vcpu_printf() */
-void kvmppc_dump_vcpu(struct kvm_vcpu *vcpu)
-{
-	int i;
-
-	printk("pc:   %08x msr:  %08x\n", vcpu->arch.pc, vcpu->arch.msr);
-	printk("lr:   %08x ctr:  %08x\n", vcpu->arch.lr, vcpu->arch.ctr);
-	printk("srr0: %08x srr1: %08x\n", vcpu->arch.srr0, vcpu->arch.srr1);
-
-	printk("exceptions: %08lx\n", vcpu->arch.pending_exceptions);
-
-	for (i = 0; i < 32; i += 4) {
-		printk("gpr%02d: %08x %08x %08x %08x\n", i,
-		       vcpu->arch.gpr[i],
-		       vcpu->arch.gpr[i+1],
-		       vcpu->arch.gpr[i+2],
-		       vcpu->arch.gpr[i+3]);
-	}
-}
-
-/* Check if we are ready to deliver the interrupt */
-static int kvmppc_can_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt)
-{
-	int r;
-
-	switch (interrupt) {
-	case BOOKE_INTERRUPT_CRITICAL:
-		r = vcpu->arch.msr & MSR_CE;
-		break;
-	case BOOKE_INTERRUPT_MACHINE_CHECK:
-		r = vcpu->arch.msr & MSR_ME;
-		break;
-	case BOOKE_INTERRUPT_EXTERNAL:
-		r = vcpu->arch.msr & MSR_EE;
-		break;
-	case BOOKE_INTERRUPT_DECREMENTER:
-		r = vcpu->arch.msr & MSR_EE;
-		break;
-	case BOOKE_INTERRUPT_FIT:
-		r = vcpu->arch.msr & MSR_EE;
-		break;
-	case BOOKE_INTERRUPT_WATCHDOG:
-		r = vcpu->arch.msr & MSR_CE;
-		break;
-	case BOOKE_INTERRUPT_DEBUG:
-		r = vcpu->arch.msr & MSR_DE;
-		break;
-	default:
-		r = 1;
-	}
-
-	return r;
-}
-
-static void kvmppc_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt)
-{
-	switch (interrupt) {
-	case BOOKE_INTERRUPT_DECREMENTER:
-		vcpu->arch.tsr |= TSR_DIS;
-		break;
-	}
-
-	vcpu->arch.srr0 = vcpu->arch.pc;
-	vcpu->arch.srr1 = vcpu->arch.msr;
-	vcpu->arch.pc = vcpu->arch.ivpr | vcpu->arch.ivor[interrupt];
-	kvmppc_set_msr(vcpu, vcpu->arch.msr & interrupt_msr_mask[interrupt]);
-}
-
-/* Check pending exceptions and deliver one, if possible. */
-void kvmppc_check_and_deliver_interrupts(struct kvm_vcpu *vcpu)
-{
-	unsigned long *pending = &vcpu->arch.pending_exceptions;
-	unsigned int exception;
-	unsigned int priority;
-
-	priority = find_first_bit(pending, BITS_PER_BYTE * sizeof(*pending));
-	while (priority <= BOOKE_MAX_INTERRUPT) {
-		exception = priority_exception[priority];
-		if (kvmppc_can_deliver_interrupt(vcpu, exception)) {
-			kvmppc_clear_exception(vcpu, exception);
-			kvmppc_deliver_interrupt(vcpu, exception);
-			break;
-		}
-
-		priority = find_next_bit(pending,
-		                         BITS_PER_BYTE * sizeof(*pending),
-		                         priority + 1);
-	}
-}
-
-/**
- * kvmppc_handle_exit
- *
- * Return value is in the form (errcode<<2 | RESUME_FLAG_HOST | RESUME_FLAG_NV)
- */
-int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
-                       unsigned int exit_nr)
-{
-	enum emulation_result er;
-	int r = RESUME_HOST;
-
-	local_irq_enable();
-
-	run->exit_reason = KVM_EXIT_UNKNOWN;
-	run->ready_for_interrupt_injection = 1;
-
-	switch (exit_nr) {
-	case BOOKE_INTERRUPT_MACHINE_CHECK:
-		printk("MACHINE CHECK: %lx\n", mfspr(SPRN_MCSR));
-		kvmppc_dump_vcpu(vcpu);
-		r = RESUME_HOST;
-		break;
-
-	case BOOKE_INTERRUPT_EXTERNAL:
-	case BOOKE_INTERRUPT_DECREMENTER:
-		/* Since we switched IVPR back to the host's value, the host
-		 * handled this interrupt the moment we enabled interrupts.
-		 * Now we just offer it a chance to reschedule the guest. */
-
-		/* XXX At this point the TLB still holds our shadow TLB, so if
-		 * we do reschedule the host will fault over it. Perhaps we
-		 * should politely restore the host's entries to minimize
-		 * misses before ceding control. */
-		if (need_resched())
-			cond_resched();
-		if (exit_nr == BOOKE_INTERRUPT_DECREMENTER)
-			vcpu->stat.dec_exits++;
-		else
-			vcpu->stat.ext_intr_exits++;
-		r = RESUME_GUEST;
-		break;
-
-	case BOOKE_INTERRUPT_PROGRAM:
-		if (vcpu->arch.msr & MSR_PR) {
-			/* Program traps generated by user-level software must be handled
-			 * by the guest kernel. */
-			vcpu->arch.esr = vcpu->arch.fault_esr;
-			kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM);
-			r = RESUME_GUEST;
-			break;
-		}
-
-		er = kvmppc_emulate_instruction(run, vcpu);
-		switch (er) {
-		case EMULATE_DONE:
-			/* Future optimization: only reload non-volatiles if
-			 * they were actually modified by emulation. */
-			vcpu->stat.emulated_inst_exits++;
-			r = RESUME_GUEST_NV;
-			break;
-		case EMULATE_DO_DCR:
-			run->exit_reason = KVM_EXIT_DCR;
-			r = RESUME_HOST;
-			break;
-		case EMULATE_FAIL:
-			/* XXX Deliver Program interrupt to guest. */
-			printk(KERN_CRIT "%s: emulation at %x failed (%08x)\n",
-			       __func__, vcpu->arch.pc, vcpu->arch.last_inst);
-			/* For debugging, encode the failing instruction and
-			 * report it to userspace. */
-			run->hw.hardware_exit_reason = ~0ULL << 32;
-			run->hw.hardware_exit_reason |= vcpu->arch.last_inst;
-			r = RESUME_HOST;
-			break;
-		default:
-			BUG();
-		}
-		break;
-
-	case BOOKE_INTERRUPT_FP_UNAVAIL:
-		kvmppc_queue_exception(vcpu, exit_nr);
-		r = RESUME_GUEST;
-		break;
-
-	case BOOKE_INTERRUPT_DATA_STORAGE:
-		vcpu->arch.dear = vcpu->arch.fault_dear;
-		vcpu->arch.esr = vcpu->arch.fault_esr;
-		kvmppc_queue_exception(vcpu, exit_nr);
-		vcpu->stat.dsi_exits++;
-		r = RESUME_GUEST;
-		break;
-
-	case BOOKE_INTERRUPT_INST_STORAGE:
-		vcpu->arch.esr = vcpu->arch.fault_esr;
-		kvmppc_queue_exception(vcpu, exit_nr);
-		vcpu->stat.isi_exits++;
-		r = RESUME_GUEST;
-		break;
-
-	case BOOKE_INTERRUPT_SYSCALL:
-		kvmppc_queue_exception(vcpu, exit_nr);
-		vcpu->stat.syscall_exits++;
-		r = RESUME_GUEST;
-		break;
-
-	case BOOKE_INTERRUPT_DTLB_MISS: {
-		struct tlbe *gtlbe;
-		unsigned long eaddr = vcpu->arch.fault_dear;
-		gfn_t gfn;
-
-		/* Check the guest TLB. */
-		gtlbe = kvmppc_44x_dtlb_search(vcpu, eaddr);
-		if (!gtlbe) {
-			/* The guest didn't have a mapping for it. */
-			kvmppc_queue_exception(vcpu, exit_nr);
-			vcpu->arch.dear = vcpu->arch.fault_dear;
-			vcpu->arch.esr = vcpu->arch.fault_esr;
-			vcpu->stat.dtlb_real_miss_exits++;
-			r = RESUME_GUEST;
-			break;
-		}
-
-		vcpu->arch.paddr_accessed = tlb_xlate(gtlbe, eaddr);
-		gfn = vcpu->arch.paddr_accessed >> PAGE_SHIFT;
-
-		if (kvm_is_visible_gfn(vcpu->kvm, gfn)) {
-			/* The guest TLB had a mapping, but the shadow TLB
-			 * didn't, and it is RAM. This could be because:
-			 * a) the entry is mapping the host kernel, or
-			 * b) the guest used a large mapping which we're faking
-			 * Either way, we need to satisfy the fault without
-			 * invoking the guest. */
-			kvmppc_mmu_map(vcpu, eaddr, gfn, gtlbe->tid,
-			               gtlbe->word2);
-			vcpu->stat.dtlb_virt_miss_exits++;
-			r = RESUME_GUEST;
-		} else {
-			/* Guest has mapped and accessed a page which is not
-			 * actually RAM. */
-			r = kvmppc_emulate_mmio(run, vcpu);
-		}
-
-		break;
-	}
-
-	case BOOKE_INTERRUPT_ITLB_MISS: {
-		struct tlbe *gtlbe;
-		unsigned long eaddr = vcpu->arch.pc;
-		gfn_t gfn;
-
-		r = RESUME_GUEST;
-
-		/* Check the guest TLB. */
-		gtlbe = kvmppc_44x_itlb_search(vcpu, eaddr);
-		if (!gtlbe) {
-			/* The guest didn't have a mapping for it. */
-			kvmppc_queue_exception(vcpu, exit_nr);
-			vcpu->stat.itlb_real_miss_exits++;
-			break;
-		}
-
-		vcpu->stat.itlb_virt_miss_exits++;
-
-		gfn = tlb_xlate(gtlbe, eaddr) >> PAGE_SHIFT;
-
-		if (kvm_is_visible_gfn(vcpu->kvm, gfn)) {
-			/* The guest TLB had a mapping, but the shadow TLB
-			 * didn't. This could be because:
-			 * a) the entry is mapping the host kernel, or
-			 * b) the guest used a large mapping which we're faking
-			 * Either way, we need to satisfy the fault without
-			 * invoking the guest. */
-			kvmppc_mmu_map(vcpu, eaddr, gfn, gtlbe->tid,
-			               gtlbe->word2);
-		} else {
-			/* Guest mapped and leaped at non-RAM! */
-			kvmppc_queue_exception(vcpu,
-			                       BOOKE_INTERRUPT_MACHINE_CHECK);
-		}
-
-		break;
-	}
-
-	case BOOKE_INTERRUPT_DEBUG: {
-		u32 dbsr;
-
-		vcpu->arch.pc = mfspr(SPRN_CSRR0);
-
-		/* clear IAC events in DBSR register */
-		dbsr = mfspr(SPRN_DBSR);
-		dbsr &= DBSR_IAC1 | DBSR_IAC2 | DBSR_IAC3 | DBSR_IAC4;
-		mtspr(SPRN_DBSR, dbsr);
-
-		run->exit_reason = KVM_EXIT_DEBUG;
-		r = RESUME_HOST;
-		break;
-	}
-
-	default:
-		printk(KERN_EMERG "exit_nr %d\n", exit_nr);
-		BUG();
-	}
-
-	local_irq_disable();
-
-	kvmppc_check_and_deliver_interrupts(vcpu);
-
-	/* Do some exit accounting. */
-	vcpu->stat.sum_exits++;
-	if (!(r & RESUME_HOST)) {
-		/* To avoid clobbering exit_reason, only check for signals if
-		 * we aren't already exiting to userspace for some other
-		 * reason. */
-		if (signal_pending(current)) {
-			run->exit_reason = KVM_EXIT_INTR;
-			r = (-EINTR << 2) | RESUME_HOST | (r & RESUME_FLAG_NV);
-
-			vcpu->stat.signal_exits++;
-		} else {
-			vcpu->stat.light_exits++;
-		}
-	} else {
-		switch (run->exit_reason) {
-		case KVM_EXIT_MMIO:
-			vcpu->stat.mmio_exits++;
-			break;
-		case KVM_EXIT_DCR:
-			vcpu->stat.dcr_exits++;
-			break;
-		case KVM_EXIT_INTR:
-			vcpu->stat.signal_exits++;
-			break;
-		}
-	}
-
-	return r;
-}
-
-/* Initial guest state: 16MB mapping 0 -> 0, PC = 0, MSR = 0, R1 = 16MB */
-int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
-{
-	struct tlbe *tlbe = &vcpu->arch.guest_tlb[0];
-
-	tlbe->tid = 0;
-	tlbe->word0 = PPC44x_TLB_16M | PPC44x_TLB_VALID;
-	tlbe->word1 = 0;
-	tlbe->word2 = PPC44x_TLB_SX | PPC44x_TLB_SW | PPC44x_TLB_SR;
-
-	tlbe++;
-	tlbe->tid = 0;
-	tlbe->word0 = 0xef600000 | PPC44x_TLB_4K | PPC44x_TLB_VALID;
-	tlbe->word1 = 0xef600000;
-	tlbe->word2 = PPC44x_TLB_SX | PPC44x_TLB_SW | PPC44x_TLB_SR
-	              | PPC44x_TLB_I | PPC44x_TLB_G;
-
-	vcpu->arch.pc = 0;
-	vcpu->arch.msr = 0;
-	vcpu->arch.gpr[1] = (16<<20) - 8; /* -8 for the callee-save LR slot */
-
-	vcpu->arch.shadow_pid = 1;
-
-	/* Eye-catching number so we know if the guest takes an interrupt
-	 * before it's programmed its own IVPR. */
-	vcpu->arch.ivpr = 0x55550000;
-
-	/* Since the guest can directly access the timebase, it must know the
-	 * real timebase frequency. Accordingly, it must see the state of
-	 * CCR1[TCS]. */
-	vcpu->arch.ccr1 = mfspr(SPRN_CCR1);
-
-	return 0;
-}
-
-int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
-{
-	int i;
-
-	regs->pc = vcpu->arch.pc;
-	regs->cr = vcpu->arch.cr;
-	regs->ctr = vcpu->arch.ctr;
-	regs->lr = vcpu->arch.lr;
-	regs->xer = vcpu->arch.xer;
-	regs->msr = vcpu->arch.msr;
-	regs->srr0 = vcpu->arch.srr0;
-	regs->srr1 = vcpu->arch.srr1;
-	regs->pid = vcpu->arch.pid;
-	regs->sprg0 = vcpu->arch.sprg0;
-	regs->sprg1 = vcpu->arch.sprg1;
-	regs->sprg2 = vcpu->arch.sprg2;
-	regs->sprg3 = vcpu->arch.sprg3;
-	regs->sprg5 = vcpu->arch.sprg4;
-	regs->sprg6 = vcpu->arch.sprg5;
-	regs->sprg7 = vcpu->arch.sprg6;
-
-	for (i = 0; i < ARRAY_SIZE(regs->gpr); i++)
-		regs->gpr[i] = vcpu->arch.gpr[i];
-
-	return 0;
-}
-
-int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
-{
-	int i;
-
-	vcpu->arch.pc = regs->pc;
-	vcpu->arch.cr = regs->cr;
-	vcpu->arch.ctr = regs->ctr;
-	vcpu->arch.lr = regs->lr;
-	vcpu->arch.xer = regs->xer;
-	vcpu->arch.msr = regs->msr;
-	vcpu->arch.srr0 = regs->srr0;
-	vcpu->arch.srr1 = regs->srr1;
-	vcpu->arch.sprg0 = regs->sprg0;
-	vcpu->arch.sprg1 = regs->sprg1;
-	vcpu->arch.sprg2 = regs->sprg2;
-	vcpu->arch.sprg3 = regs->sprg3;
-	vcpu->arch.sprg5 = regs->sprg4;
-	vcpu->arch.sprg6 = regs->sprg5;
-	vcpu->arch.sprg7 = regs->sprg6;
-
-	for (i = 0; i < ARRAY_SIZE(vcpu->arch.gpr); i++)
-		vcpu->arch.gpr[i] = regs->gpr[i];
-
-	return 0;
-}
-
-int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
-                                  struct kvm_sregs *sregs)
-{
-	return -ENOTSUPP;
-}
-
-int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
-                                  struct kvm_sregs *sregs)
-{
-	return -ENOTSUPP;
-}
-
-int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
-{
-	return -ENOTSUPP;
-}
-
-int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
-{
-	return -ENOTSUPP;
-}
-
-/* 'linear_address' is actually an encoding of AS|PID|EADDR . */
-int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
-                                  struct kvm_translation *tr)
-{
-	struct tlbe *gtlbe;
-	int index;
-	gva_t eaddr;
-	u8 pid;
-	u8 as;
-
-	eaddr = tr->linear_address;
-	pid = (tr->linear_address >> 32) & 0xff;
-	as = (tr->linear_address >> 40) & 0x1;
-
-	index = kvmppc_44x_tlb_index(vcpu, eaddr, pid, as);
-	if (index == -1) {
-		tr->valid = 0;
-		return 0;
-	}
-
-	gtlbe = &vcpu->arch.guest_tlb[index];
-
-	tr->physical_address = tlb_xlate(gtlbe, eaddr);
-	/* XXX what does "writeable" and "usermode" even mean? */
-	tr->valid = 1;
-
-	return 0;
-}
diff --git a/arch/powerpc/kvm/booke_host.c b/arch/powerpc/kvm/booke_host.c
deleted file mode 100644
index b480341..0000000
--- a/arch/powerpc/kvm/booke_host.c
+++ /dev/null
@@ -1,83 +0,0 @@
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License, version 2, as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
- *
- * Copyright IBM Corp. 2008
- *
- * Authors: Hollis Blanchard <hollisb@us.ibm.com>
- */
-
-#include <linux/errno.h>
-#include <linux/kvm_host.h>
-#include <linux/module.h>
-#include <asm/cacheflush.h>
-#include <asm/kvm_ppc.h>
-
-unsigned long kvmppc_booke_handlers;
-
-static int kvmppc_booke_init(void)
-{
-	unsigned long ivor[16];
-	unsigned long max_ivor = 0;
-	int i;
-
-	/* We install our own exception handlers by hijacking IVPR. IVPR must
-	 * be 16-bit aligned, so we need a 64KB allocation. */
-	kvmppc_booke_handlers = __get_free_pages(GFP_KERNEL | __GFP_ZERO,
-	                                         VCPU_SIZE_ORDER);
-	if (!kvmppc_booke_handlers)
-		return -ENOMEM;
-
-	/* XXX make sure our handlers are smaller than Linux's */
-
-	/* Copy our interrupt handlers to match host IVORs. That way we don't
-	 * have to swap the IVORs on every guest/host transition. */
-	ivor[0] = mfspr(SPRN_IVOR0);
-	ivor[1] = mfspr(SPRN_IVOR1);
-	ivor[2] = mfspr(SPRN_IVOR2);
-	ivor[3] = mfspr(SPRN_IVOR3);
-	ivor[4] = mfspr(SPRN_IVOR4);
-	ivor[5] = mfspr(SPRN_IVOR5);
-	ivor[6] = mfspr(SPRN_IVOR6);
-	ivor[7] = mfspr(SPRN_IVOR7);
-	ivor[8] = mfspr(SPRN_IVOR8);
-	ivor[9] = mfspr(SPRN_IVOR9);
-	ivor[10] = mfspr(SPRN_IVOR10);
-	ivor[11] = mfspr(SPRN_IVOR11);
-	ivor[12] = mfspr(SPRN_IVOR12);
-	ivor[13] = mfspr(SPRN_IVOR13);
-	ivor[14] = mfspr(SPRN_IVOR14);
-	ivor[15] = mfspr(SPRN_IVOR15);
-
-	for (i = 0; i < 16; i++) {
-		if (ivor[i] > max_ivor)
-			max_ivor = ivor[i];
-
-		memcpy((void *)kvmppc_booke_handlers + ivor[i],
-		       kvmppc_handlers_start + i * kvmppc_handler_len,
-		       kvmppc_handler_len);
-	}
-	flush_icache_range(kvmppc_booke_handlers,
-	                   kvmppc_booke_handlers + max_ivor + kvmppc_handler_len);
-
-	return kvm_init(NULL, sizeof(struct kvm_vcpu), THIS_MODULE);
-}
-
-static void __exit kvmppc_booke_exit(void)
-{
-	free_pages(kvmppc_booke_handlers, VCPU_SIZE_ORDER);
-	kvm_exit();
-}
-
-module_init(kvmppc_booke_init)
-module_exit(kvmppc_booke_exit)
diff --git a/arch/powerpc/kvm/booke_interrupts.S b/arch/powerpc/kvm/booke_interrupts.S
index 95e165b..084ebcd 100644
--- a/arch/powerpc/kvm/booke_interrupts.S
+++ b/arch/powerpc/kvm/booke_interrupts.S
@@ -107,6 +107,18 @@
 	li	r6, 1
 	slw	r6, r6, r5
 
+#ifdef CONFIG_KVM_EXIT_TIMING
+	/* save exit time */
+1:
+	mfspr	r7, SPRN_TBRU
+	mfspr	r8, SPRN_TBRL
+	mfspr	r9, SPRN_TBRU
+	cmpw	r9, r7
+	bne	1b
+	stw	r8, VCPU_TIMING_EXIT_TBL(r4)
+	stw	r9, VCPU_TIMING_EXIT_TBU(r4)
+#endif
+
 	/* Save the faulting instruction and all GPRs for emulation. */
 	andi.	r7, r6, NEED_INST_MASK
 	beq	..skip_inst_copy
@@ -335,54 +347,6 @@
 	lwz	r3, VCPU_SHADOW_PID(r4)
 	mtspr	SPRN_PID, r3
 
-	/* Prevent all asynchronous TLB updates. */
-	mfmsr	r5
-	lis	r6, (MSR_EE|MSR_CE|MSR_ME|MSR_DE)@h
-	ori	r6, r6, (MSR_EE|MSR_CE|MSR_ME|MSR_DE)@l
-	andc	r6, r5, r6
-	mtmsr	r6
-
-	/* Load the guest mappings, leaving the host's "pinned" kernel mappings
-	 * in place. */
-	mfspr	r10, SPRN_MMUCR			/* Save host MMUCR. */
-	li	r5, PPC44x_TLB_SIZE
-	lis	r5, tlb_44x_hwater@ha
-	lwz	r5, tlb_44x_hwater@l(r5)
-	mtctr	r5
-	addi	r9, r4, VCPU_SHADOW_TLB
-	addi	r5, r4, VCPU_SHADOW_MOD
-	li	r3, 0
-1:
-	lbzx	r7, r3, r5
-	cmpwi	r7, 0
-	beq	3f
-
-	/* Load guest entry. */
-	mulli	r11, r3, TLBE_BYTES
-	add	r11, r11, r9
-	lwz	r7, 0(r11)
-	mtspr	SPRN_MMUCR, r7
-	lwz	r7, 4(r11)
-	tlbwe	r7, r3, PPC44x_TLB_PAGEID
-	lwz	r7, 8(r11)
-	tlbwe	r7, r3, PPC44x_TLB_XLAT
-	lwz	r7, 12(r11)
-	tlbwe	r7, r3, PPC44x_TLB_ATTRIB
-3:
-	addi	r3, r3, 1                       /* Increment index. */
-	bdnz	1b
-
-	mtspr	SPRN_MMUCR, r10			/* Restore host MMUCR. */
-
-	/* Clear bitmap of modified TLB entries */
-	li	r5, PPC44x_TLB_SIZE>>2
-	mtctr	r5
-	addi	r5, r4, VCPU_SHADOW_MOD - 4
-	li	r6, 0
-1:
-	stwu	r6, 4(r5)
-	bdnz	1b
-
 	iccci	0, 0 /* XXX hack */
 
 	/* Load some guest volatiles. */
@@ -423,6 +387,18 @@
 	lwz	r3, VCPU_SPRG7(r4)
 	mtspr	SPRN_SPRG7, r3
 
+#ifdef CONFIG_KVM_EXIT_TIMING
+	/* save enter time */
+1:
+	mfspr	r6, SPRN_TBRU
+	mfspr	r7, SPRN_TBRL
+	mfspr	r8, SPRN_TBRU
+	cmpw	r8, r6
+	bne	1b
+	stw	r7, VCPU_TIMING_LAST_ENTER_TBL(r4)
+	stw	r8, VCPU_TIMING_LAST_ENTER_TBU(r4)
+#endif
+
 	/* Finish loading guest volatiles and jump to guest. */
 	lwz	r3, VCPU_CTR(r4)
 	mtctr	r3
diff --git a/arch/powerpc/kvm/emulate.c b/arch/powerpc/kvm/emulate.c
index 0fce4fb..d1d38da 100644
--- a/arch/powerpc/kvm/emulate.c
+++ b/arch/powerpc/kvm/emulate.c
@@ -23,161 +23,14 @@
 #include <linux/string.h>
 #include <linux/kvm_host.h>
 
-#include <asm/dcr.h>
-#include <asm/dcr-regs.h>
+#include <asm/reg.h>
 #include <asm/time.h>
 #include <asm/byteorder.h>
 #include <asm/kvm_ppc.h>
+#include <asm/disassemble.h>
+#include "timing.h"
 
-#include "44x_tlb.h"
-
-/* Instruction decoding */
-static inline unsigned int get_op(u32 inst)
-{
-	return inst >> 26;
-}
-
-static inline unsigned int get_xop(u32 inst)
-{
-	return (inst >> 1) & 0x3ff;
-}
-
-static inline unsigned int get_sprn(u32 inst)
-{
-	return ((inst >> 16) & 0x1f) | ((inst >> 6) & 0x3e0);
-}
-
-static inline unsigned int get_dcrn(u32 inst)
-{
-	return ((inst >> 16) & 0x1f) | ((inst >> 6) & 0x3e0);
-}
-
-static inline unsigned int get_rt(u32 inst)
-{
-	return (inst >> 21) & 0x1f;
-}
-
-static inline unsigned int get_rs(u32 inst)
-{
-	return (inst >> 21) & 0x1f;
-}
-
-static inline unsigned int get_ra(u32 inst)
-{
-	return (inst >> 16) & 0x1f;
-}
-
-static inline unsigned int get_rb(u32 inst)
-{
-	return (inst >> 11) & 0x1f;
-}
-
-static inline unsigned int get_rc(u32 inst)
-{
-	return inst & 0x1;
-}
-
-static inline unsigned int get_ws(u32 inst)
-{
-	return (inst >> 11) & 0x1f;
-}
-
-static inline unsigned int get_d(u32 inst)
-{
-	return inst & 0xffff;
-}
-
-static int tlbe_is_host_safe(const struct kvm_vcpu *vcpu,
-                             const struct tlbe *tlbe)
-{
-	gpa_t gpa;
-
-	if (!get_tlb_v(tlbe))
-		return 0;
-
-	/* Does it match current guest AS? */
-	/* XXX what about IS != DS? */
-	if (get_tlb_ts(tlbe) != !!(vcpu->arch.msr & MSR_IS))
-		return 0;
-
-	gpa = get_tlb_raddr(tlbe);
-	if (!gfn_to_memslot(vcpu->kvm, gpa >> PAGE_SHIFT))
-		/* Mapping is not for RAM. */
-		return 0;
-
-	return 1;
-}
-
-static int kvmppc_emul_tlbwe(struct kvm_vcpu *vcpu, u32 inst)
-{
-	u64 eaddr;
-	u64 raddr;
-	u64 asid;
-	u32 flags;
-	struct tlbe *tlbe;
-	unsigned int ra;
-	unsigned int rs;
-	unsigned int ws;
-	unsigned int index;
-
-	ra = get_ra(inst);
-	rs = get_rs(inst);
-	ws = get_ws(inst);
-
-	index = vcpu->arch.gpr[ra];
-	if (index > PPC44x_TLB_SIZE) {
-		printk("%s: index %d\n", __func__, index);
-		kvmppc_dump_vcpu(vcpu);
-		return EMULATE_FAIL;
-	}
-
-	tlbe = &vcpu->arch.guest_tlb[index];
-
-	/* Invalidate shadow mappings for the about-to-be-clobbered TLBE. */
-	if (tlbe->word0 & PPC44x_TLB_VALID) {
-		eaddr = get_tlb_eaddr(tlbe);
-		asid = (tlbe->word0 & PPC44x_TLB_TS) | tlbe->tid;
-		kvmppc_mmu_invalidate(vcpu, eaddr, get_tlb_end(tlbe), asid);
-	}
-
-	switch (ws) {
-	case PPC44x_TLB_PAGEID:
-		tlbe->tid = vcpu->arch.mmucr & 0xff;
-		tlbe->word0 = vcpu->arch.gpr[rs];
-		break;
-
-	case PPC44x_TLB_XLAT:
-		tlbe->word1 = vcpu->arch.gpr[rs];
-		break;
-
-	case PPC44x_TLB_ATTRIB:
-		tlbe->word2 = vcpu->arch.gpr[rs];
-		break;
-
-	default:
-		return EMULATE_FAIL;
-	}
-
-	if (tlbe_is_host_safe(vcpu, tlbe)) {
-		eaddr = get_tlb_eaddr(tlbe);
-		raddr = get_tlb_raddr(tlbe);
-		asid = (tlbe->word0 & PPC44x_TLB_TS) | tlbe->tid;
-		flags = tlbe->word2 & 0xffff;
-
-		/* Create a 4KB mapping on the host. If the guest wanted a
-		 * large page, only the first 4KB is mapped here and the rest
-		 * are mapped on the fly. */
-		kvmppc_mmu_map(vcpu, eaddr, raddr >> PAGE_SHIFT, asid, flags);
-	}
-
-	KVMTRACE_5D(GTLB_WRITE, vcpu, index,
-			tlbe->tid, tlbe->word0, tlbe->word1, tlbe->word2,
-			handler);
-
-	return EMULATE_DONE;
-}
-
-static void kvmppc_emulate_dec(struct kvm_vcpu *vcpu)
+void kvmppc_emulate_dec(struct kvm_vcpu *vcpu)
 {
 	if (vcpu->arch.tcr & TCR_DIE) {
 		/* The decrementer ticks at the same rate as the timebase, so
@@ -193,12 +46,6 @@
 	}
 }
 
-static void kvmppc_emul_rfi(struct kvm_vcpu *vcpu)
-{
-	vcpu->arch.pc = vcpu->arch.srr0;
-	kvmppc_set_msr(vcpu, vcpu->arch.srr1);
-}
-
 /* XXX to do:
  * lhax
  * lhaux
@@ -213,40 +60,30 @@
  *
  * XXX is_bigendian should depend on MMU mapping or MSR[LE]
  */
+/* XXX Should probably auto-generate instruction decoding for a particular core
+ * from opcode tables in the future. */
 int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 {
 	u32 inst = vcpu->arch.last_inst;
 	u32 ea;
 	int ra;
 	int rb;
-	int rc;
 	int rs;
 	int rt;
 	int sprn;
-	int dcrn;
 	enum emulation_result emulated = EMULATE_DONE;
 	int advance = 1;
 
+	/* this default type might be overwritten by subcategories */
+	kvmppc_set_exit_type(vcpu, EMULATED_INST_EXITS);
+
 	switch (get_op(inst)) {
-	case 3:                                                 /* trap */
-		printk("trap!\n");
-		kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM);
+	case 3:                                             /* trap */
+		vcpu->arch.esr |= ESR_PTR;
+		kvmppc_core_queue_program(vcpu);
 		advance = 0;
 		break;
 
-	case 19:
-		switch (get_xop(inst)) {
-		case 50:                                        /* rfi */
-			kvmppc_emul_rfi(vcpu);
-			advance = 0;
-			break;
-
-		default:
-			emulated = EMULATE_FAIL;
-			break;
-		}
-		break;
-
 	case 31:
 		switch (get_xop(inst)) {
 
@@ -255,27 +92,11 @@
 			emulated = kvmppc_handle_load(run, vcpu, rt, 4, 1);
 			break;
 
-		case 83:                                        /* mfmsr */
-			rt = get_rt(inst);
-			vcpu->arch.gpr[rt] = vcpu->arch.msr;
-			break;
-
 		case 87:                                        /* lbzx */
 			rt = get_rt(inst);
 			emulated = kvmppc_handle_load(run, vcpu, rt, 1, 1);
 			break;
 
-		case 131:                                       /* wrtee */
-			rs = get_rs(inst);
-			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
-			                 | (vcpu->arch.gpr[rs] & MSR_EE);
-			break;
-
-		case 146:                                       /* mtmsr */
-			rs = get_rs(inst);
-			kvmppc_set_msr(vcpu, vcpu->arch.gpr[rs]);
-			break;
-
 		case 151:                                       /* stwx */
 			rs = get_rs(inst);
 			emulated = kvmppc_handle_store(run, vcpu,
@@ -283,11 +104,6 @@
 			                               4, 1);
 			break;
 
-		case 163:                                       /* wrteei */
-			vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE)
-			                 | (inst & MSR_EE);
-			break;
-
 		case 215:                                       /* stbx */
 			rs = get_rs(inst);
 			emulated = kvmppc_handle_store(run, vcpu,
@@ -328,42 +144,6 @@
 			vcpu->arch.gpr[ra] = ea;
 			break;
 
-		case 323:                                       /* mfdcr */
-			dcrn = get_dcrn(inst);
-			rt = get_rt(inst);
-
-			/* The guest may access CPR0 registers to determine the timebase
-			 * frequency, and it must know the real host frequency because it
-			 * can directly access the timebase registers.
-			 *
-			 * It would be possible to emulate those accesses in userspace,
-			 * but userspace can really only figure out the end frequency.
-			 * We could decompose that into the factors that compute it, but
-			 * that's tricky math, and it's easier to just report the real
-			 * CPR0 values.
-			 */
-			switch (dcrn) {
-			case DCRN_CPR0_CONFIG_ADDR:
-				vcpu->arch.gpr[rt] = vcpu->arch.cpr0_cfgaddr;
-				break;
-			case DCRN_CPR0_CONFIG_DATA:
-				local_irq_disable();
-				mtdcr(DCRN_CPR0_CONFIG_ADDR,
-				      vcpu->arch.cpr0_cfgaddr);
-				vcpu->arch.gpr[rt] = mfdcr(DCRN_CPR0_CONFIG_DATA);
-				local_irq_enable();
-				break;
-			default:
-				run->dcr.dcrn = dcrn;
-				run->dcr.data =  0;
-				run->dcr.is_write = 0;
-				vcpu->arch.io_gpr = rt;
-				vcpu->arch.dcr_needed = 1;
-				emulated = EMULATE_DO_DCR;
-			}
-
-			break;
-
 		case 339:                                       /* mfspr */
 			sprn = get_sprn(inst);
 			rt = get_rt(inst);
@@ -373,26 +153,8 @@
 				vcpu->arch.gpr[rt] = vcpu->arch.srr0; break;
 			case SPRN_SRR1:
 				vcpu->arch.gpr[rt] = vcpu->arch.srr1; break;
-			case SPRN_MMUCR:
-				vcpu->arch.gpr[rt] = vcpu->arch.mmucr; break;
-			case SPRN_PID:
-				vcpu->arch.gpr[rt] = vcpu->arch.pid; break;
-			case SPRN_IVPR:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivpr; break;
-			case SPRN_CCR0:
-				vcpu->arch.gpr[rt] = vcpu->arch.ccr0; break;
-			case SPRN_CCR1:
-				vcpu->arch.gpr[rt] = vcpu->arch.ccr1; break;
 			case SPRN_PVR:
 				vcpu->arch.gpr[rt] = vcpu->arch.pvr; break;
-			case SPRN_DEAR:
-				vcpu->arch.gpr[rt] = vcpu->arch.dear; break;
-			case SPRN_ESR:
-				vcpu->arch.gpr[rt] = vcpu->arch.esr; break;
-			case SPRN_DBCR0:
-				vcpu->arch.gpr[rt] = vcpu->arch.dbcr0; break;
-			case SPRN_DBCR1:
-				vcpu->arch.gpr[rt] = vcpu->arch.dbcr1; break;
 
 			/* Note: mftb and TBRL/TBWL are user-accessible, so
 			 * the guest can always access the real TB anyways.
@@ -413,42 +175,12 @@
 			/* Note: SPRG4-7 are user-readable, so we don't get
 			 * a trap. */
 
-			case SPRN_IVOR0:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[0]; break;
-			case SPRN_IVOR1:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[1]; break;
-			case SPRN_IVOR2:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[2]; break;
-			case SPRN_IVOR3:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[3]; break;
-			case SPRN_IVOR4:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[4]; break;
-			case SPRN_IVOR5:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[5]; break;
-			case SPRN_IVOR6:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[6]; break;
-			case SPRN_IVOR7:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[7]; break;
-			case SPRN_IVOR8:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[8]; break;
-			case SPRN_IVOR9:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[9]; break;
-			case SPRN_IVOR10:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[10]; break;
-			case SPRN_IVOR11:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[11]; break;
-			case SPRN_IVOR12:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[12]; break;
-			case SPRN_IVOR13:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[13]; break;
-			case SPRN_IVOR14:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[14]; break;
-			case SPRN_IVOR15:
-				vcpu->arch.gpr[rt] = vcpu->arch.ivor[15]; break;
-
 			default:
-				printk("mfspr: unknown spr %x\n", sprn);
-				vcpu->arch.gpr[rt] = 0;
+				emulated = kvmppc_core_emulate_mfspr(vcpu, sprn, rt);
+				if (emulated == EMULATE_FAIL) {
+					printk("mfspr: unknown spr %x\n", sprn);
+					vcpu->arch.gpr[rt] = 0;
+				}
 				break;
 			}
 			break;
@@ -478,25 +210,6 @@
 			vcpu->arch.gpr[ra] = ea;
 			break;
 
-		case 451:                                       /* mtdcr */
-			dcrn = get_dcrn(inst);
-			rs = get_rs(inst);
-
-			/* emulate some access in kernel */
-			switch (dcrn) {
-			case DCRN_CPR0_CONFIG_ADDR:
-				vcpu->arch.cpr0_cfgaddr = vcpu->arch.gpr[rs];
-				break;
-			default:
-				run->dcr.dcrn = dcrn;
-				run->dcr.data = vcpu->arch.gpr[rs];
-				run->dcr.is_write = 1;
-				vcpu->arch.dcr_needed = 1;
-				emulated = EMULATE_DO_DCR;
-			}
-
-			break;
-
 		case 467:                                       /* mtspr */
 			sprn = get_sprn(inst);
 			rs = get_rs(inst);
@@ -505,22 +218,6 @@
 				vcpu->arch.srr0 = vcpu->arch.gpr[rs]; break;
 			case SPRN_SRR1:
 				vcpu->arch.srr1 = vcpu->arch.gpr[rs]; break;
-			case SPRN_MMUCR:
-				vcpu->arch.mmucr = vcpu->arch.gpr[rs]; break;
-			case SPRN_PID:
-				kvmppc_set_pid(vcpu, vcpu->arch.gpr[rs]); break;
-			case SPRN_CCR0:
-				vcpu->arch.ccr0 = vcpu->arch.gpr[rs]; break;
-			case SPRN_CCR1:
-				vcpu->arch.ccr1 = vcpu->arch.gpr[rs]; break;
-			case SPRN_DEAR:
-				vcpu->arch.dear = vcpu->arch.gpr[rs]; break;
-			case SPRN_ESR:
-				vcpu->arch.esr = vcpu->arch.gpr[rs]; break;
-			case SPRN_DBCR0:
-				vcpu->arch.dbcr0 = vcpu->arch.gpr[rs]; break;
-			case SPRN_DBCR1:
-				vcpu->arch.dbcr1 = vcpu->arch.gpr[rs]; break;
 
 			/* XXX We need to context-switch the timebase for
 			 * watchdog and FIT. */
@@ -532,14 +229,6 @@
 				kvmppc_emulate_dec(vcpu);
 				break;
 
-			case SPRN_TSR:
-				vcpu->arch.tsr &= ~vcpu->arch.gpr[rs]; break;
-
-			case SPRN_TCR:
-				vcpu->arch.tcr = vcpu->arch.gpr[rs];
-				kvmppc_emulate_dec(vcpu);
-				break;
-
 			case SPRN_SPRG0:
 				vcpu->arch.sprg0 = vcpu->arch.gpr[rs]; break;
 			case SPRN_SPRG1:
@@ -549,56 +238,10 @@
 			case SPRN_SPRG3:
 				vcpu->arch.sprg3 = vcpu->arch.gpr[rs]; break;
 
-			/* Note: SPRG4-7 are user-readable. These values are
-			 * loaded into the real SPRGs when resuming the
-			 * guest. */
-			case SPRN_SPRG4:
-				vcpu->arch.sprg4 = vcpu->arch.gpr[rs]; break;
-			case SPRN_SPRG5:
-				vcpu->arch.sprg5 = vcpu->arch.gpr[rs]; break;
-			case SPRN_SPRG6:
-				vcpu->arch.sprg6 = vcpu->arch.gpr[rs]; break;
-			case SPRN_SPRG7:
-				vcpu->arch.sprg7 = vcpu->arch.gpr[rs]; break;
-
-			case SPRN_IVPR:
-				vcpu->arch.ivpr = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR0:
-				vcpu->arch.ivor[0] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR1:
-				vcpu->arch.ivor[1] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR2:
-				vcpu->arch.ivor[2] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR3:
-				vcpu->arch.ivor[3] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR4:
-				vcpu->arch.ivor[4] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR5:
-				vcpu->arch.ivor[5] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR6:
-				vcpu->arch.ivor[6] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR7:
-				vcpu->arch.ivor[7] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR8:
-				vcpu->arch.ivor[8] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR9:
-				vcpu->arch.ivor[9] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR10:
-				vcpu->arch.ivor[10] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR11:
-				vcpu->arch.ivor[11] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR12:
-				vcpu->arch.ivor[12] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR13:
-				vcpu->arch.ivor[13] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR14:
-				vcpu->arch.ivor[14] = vcpu->arch.gpr[rs]; break;
-			case SPRN_IVOR15:
-				vcpu->arch.ivor[15] = vcpu->arch.gpr[rs]; break;
-
 			default:
-				printk("mtspr: unknown spr %x\n", sprn);
-				emulated = EMULATE_FAIL;
+				emulated = kvmppc_core_emulate_mtspr(vcpu, sprn, rs);
+				if (emulated == EMULATE_FAIL)
+					printk("mtspr: unknown spr %x\n", sprn);
 				break;
 			}
 			break;
@@ -629,36 +272,6 @@
 			                               4, 0);
 			break;
 
-		case 978:                                       /* tlbwe */
-			emulated = kvmppc_emul_tlbwe(vcpu, inst);
-			break;
-
-		case 914:       {                               /* tlbsx */
-			int index;
-			unsigned int as = get_mmucr_sts(vcpu);
-			unsigned int pid = get_mmucr_stid(vcpu);
-
-			rt = get_rt(inst);
-			ra = get_ra(inst);
-			rb = get_rb(inst);
-			rc = get_rc(inst);
-
-			ea = vcpu->arch.gpr[rb];
-			if (ra)
-				ea += vcpu->arch.gpr[ra];
-
-			index = kvmppc_44x_tlb_index(vcpu, ea, pid, as);
-			if (rc) {
-				if (index < 0)
-					vcpu->arch.cr &= ~0x20000000;
-				else
-					vcpu->arch.cr |= 0x20000000;
-			}
-			vcpu->arch.gpr[rt] = index;
-
-			}
-			break;
-
 		case 790:                                       /* lhbrx */
 			rt = get_rt(inst);
 			emulated = kvmppc_handle_load(run, vcpu, rt, 2, 0);
@@ -674,14 +287,9 @@
 			                               2, 0);
 			break;
 
-		case 966:                                       /* iccci */
-			break;
-
 		default:
-			printk("unknown: op %d xop %d\n", get_op(inst),
-				get_xop(inst));
+			/* Attempt core-specific emulation below. */
 			emulated = EMULATE_FAIL;
-			break;
 		}
 		break;
 
@@ -764,12 +372,19 @@
 		break;
 
 	default:
-		printk("unknown op %d\n", get_op(inst));
 		emulated = EMULATE_FAIL;
-		break;
 	}
 
-	KVMTRACE_3D(PPC_INSTR, vcpu, inst, vcpu->arch.pc, emulated, entryexit);
+	if (emulated == EMULATE_FAIL) {
+		emulated = kvmppc_core_emulate_op(run, vcpu, inst, &advance);
+		if (emulated == EMULATE_FAIL) {
+			advance = 0;
+			printk(KERN_ERR "Couldn't emulate instruction 0x%08x "
+			       "(op %d xop %d)\n", inst, get_op(inst), get_xop(inst));
+		}
+	}
+
+	KVMTRACE_3D(PPC_INSTR, vcpu, inst, (int)vcpu->arch.pc, emulated, entryexit);
 
 	if (advance)
 		vcpu->arch.pc += 4; /* Advance past emulated instruction. */
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 8bef0ef..2822c8c 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -28,9 +28,9 @@
 #include <asm/uaccess.h>
 #include <asm/kvm_ppc.h>
 #include <asm/tlbflush.h>
+#include "timing.h"
 #include "../mm/mmu_decl.h"
 
-
 gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
 {
 	return gfn;
@@ -99,14 +99,7 @@
 
 void kvm_arch_check_processor_compat(void *rtn)
 {
-	int r;
-
-	if (strcmp(cur_cpu_spec->platform, "ppc440") == 0)
-		r = 0;
-	else
-		r = -ENOTSUPP;
-
-	*(int *)rtn = r;
+	*(int *)rtn = kvmppc_core_check_processor_compat();
 }
 
 struct kvm *kvm_arch_create_vm(void)
@@ -144,9 +137,6 @@
 	int r;
 
 	switch (ext) {
-	case KVM_CAP_USER_MEMORY:
-		r = 1;
-		break;
 	case KVM_CAP_COALESCED_MMIO:
 		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
 		break;
@@ -179,30 +169,15 @@
 struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
 {
 	struct kvm_vcpu *vcpu;
-	int err;
-
-	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
-	if (!vcpu) {
-		err = -ENOMEM;
-		goto out;
-	}
-
-	err = kvm_vcpu_init(vcpu, kvm, id);
-	if (err)
-		goto free_vcpu;
-
+	vcpu = kvmppc_core_vcpu_create(kvm, id);
+	kvmppc_create_vcpu_debugfs(vcpu, id);
 	return vcpu;
-
-free_vcpu:
-	kmem_cache_free(kvm_vcpu_cache, vcpu);
-out:
-	return ERR_PTR(err);
 }
 
 void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
 {
-	kvm_vcpu_uninit(vcpu);
-	kmem_cache_free(kvm_vcpu_cache, vcpu);
+	kvmppc_remove_vcpu_debugfs(vcpu);
+	kvmppc_core_vcpu_free(vcpu);
 }
 
 void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
@@ -212,16 +187,14 @@
 
 int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 {
-	unsigned int priority = exception_priority[BOOKE_INTERRUPT_DECREMENTER];
-
-	return test_bit(priority, &vcpu->arch.pending_exceptions);
+	return kvmppc_core_pending_dec(vcpu);
 }
 
 static void kvmppc_decrementer_func(unsigned long data)
 {
 	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data;
 
-	kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_DECREMENTER);
+	kvmppc_core_queue_dec(vcpu);
 
 	if (waitqueue_active(&vcpu->wq)) {
 		wake_up_interruptible(&vcpu->wq);
@@ -242,96 +215,25 @@
 	kvmppc_core_destroy_mmu(vcpu);
 }
 
-/* Note: clearing MSR[DE] just means that the debug interrupt will not be
- * delivered *immediately*. Instead, it simply sets the appropriate DBSR bits.
- * If those DBSR bits are still set when MSR[DE] is re-enabled, the interrupt
- * will be delivered as an "imprecise debug event" (which is indicated by
- * DBSR[IDE].
- */
-static void kvmppc_disable_debug_interrupts(void)
-{
-	mtmsr(mfmsr() & ~MSR_DE);
-}
-
-static void kvmppc_restore_host_debug_state(struct kvm_vcpu *vcpu)
-{
-	kvmppc_disable_debug_interrupts();
-
-	mtspr(SPRN_IAC1, vcpu->arch.host_iac[0]);
-	mtspr(SPRN_IAC2, vcpu->arch.host_iac[1]);
-	mtspr(SPRN_IAC3, vcpu->arch.host_iac[2]);
-	mtspr(SPRN_IAC4, vcpu->arch.host_iac[3]);
-	mtspr(SPRN_DBCR1, vcpu->arch.host_dbcr1);
-	mtspr(SPRN_DBCR2, vcpu->arch.host_dbcr2);
-	mtspr(SPRN_DBCR0, vcpu->arch.host_dbcr0);
-	mtmsr(vcpu->arch.host_msr);
-}
-
-static void kvmppc_load_guest_debug_registers(struct kvm_vcpu *vcpu)
-{
-	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
-	u32 dbcr0 = 0;
-
-	vcpu->arch.host_msr = mfmsr();
-	kvmppc_disable_debug_interrupts();
-
-	/* Save host debug register state. */
-	vcpu->arch.host_iac[0] = mfspr(SPRN_IAC1);
-	vcpu->arch.host_iac[1] = mfspr(SPRN_IAC2);
-	vcpu->arch.host_iac[2] = mfspr(SPRN_IAC3);
-	vcpu->arch.host_iac[3] = mfspr(SPRN_IAC4);
-	vcpu->arch.host_dbcr0 = mfspr(SPRN_DBCR0);
-	vcpu->arch.host_dbcr1 = mfspr(SPRN_DBCR1);
-	vcpu->arch.host_dbcr2 = mfspr(SPRN_DBCR2);
-
-	/* set registers up for guest */
-
-	if (dbg->bp[0]) {
-		mtspr(SPRN_IAC1, dbg->bp[0]);
-		dbcr0 |= DBCR0_IAC1 | DBCR0_IDM;
-	}
-	if (dbg->bp[1]) {
-		mtspr(SPRN_IAC2, dbg->bp[1]);
-		dbcr0 |= DBCR0_IAC2 | DBCR0_IDM;
-	}
-	if (dbg->bp[2]) {
-		mtspr(SPRN_IAC3, dbg->bp[2]);
-		dbcr0 |= DBCR0_IAC3 | DBCR0_IDM;
-	}
-	if (dbg->bp[3]) {
-		mtspr(SPRN_IAC4, dbg->bp[3]);
-		dbcr0 |= DBCR0_IAC4 | DBCR0_IDM;
-	}
-
-	mtspr(SPRN_DBCR0, dbcr0);
-	mtspr(SPRN_DBCR1, 0);
-	mtspr(SPRN_DBCR2, 0);
-}
-
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
-	int i;
-
 	if (vcpu->guest_debug.enabled)
-		kvmppc_load_guest_debug_registers(vcpu);
+		kvmppc_core_load_guest_debugstate(vcpu);
 
-	/* Mark every guest entry in the shadow TLB entry modified, so that they
-	 * will all be reloaded on the next vcpu run (instead of being
-	 * demand-faulted). */
-	for (i = 0; i <= tlb_44x_hwater; i++)
-		kvmppc_tlbe_set_modified(vcpu, i);
+	kvmppc_core_vcpu_load(vcpu, cpu);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
 	if (vcpu->guest_debug.enabled)
-		kvmppc_restore_host_debug_state(vcpu);
+		kvmppc_core_load_host_debugstate(vcpu);
 
 	/* Don't leave guest TLB entries resident when being de-scheduled. */
 	/* XXX It would be nice to differentiate between heavyweight exit and
 	 * sched_out here, since we could avoid the TLB flush for heavyweight
 	 * exits. */
 	_tlbil_all();
+	kvmppc_core_vcpu_put(vcpu);
 }
 
 int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
@@ -355,14 +257,14 @@
 static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
 {
-	u32 *gpr = &vcpu->arch.gpr[vcpu->arch.io_gpr];
+	ulong *gpr = &vcpu->arch.gpr[vcpu->arch.io_gpr];
 	*gpr = run->dcr.data;
 }
 
 static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                       struct kvm_run *run)
 {
-	u32 *gpr = &vcpu->arch.gpr[vcpu->arch.io_gpr];
+	ulong *gpr = &vcpu->arch.gpr[vcpu->arch.io_gpr];
 
 	if (run->mmio.len > sizeof(*gpr)) {
 		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
@@ -460,7 +362,7 @@
 		vcpu->arch.dcr_needed = 0;
 	}
 
-	kvmppc_check_and_deliver_interrupts(vcpu);
+	kvmppc_core_deliver_interrupts(vcpu);
 
 	local_irq_disable();
 	kvm_guest_enter();
@@ -478,7 +380,7 @@
 
 int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
 {
-	kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_EXTERNAL);
+	kvmppc_core_queue_external(vcpu, irq);
 
 	if (waitqueue_active(&vcpu->wq)) {
 		wake_up_interruptible(&vcpu->wq);
diff --git a/arch/powerpc/kvm/timing.c b/arch/powerpc/kvm/timing.c
new file mode 100644
index 0000000..47ee603
--- /dev/null
+++ b/arch/powerpc/kvm/timing.c
@@ -0,0 +1,239 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ *          Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
+ */
+
+#include <linux/kvm_host.h>
+#include <linux/fs.h>
+#include <linux/seq_file.h>
+#include <linux/debugfs.h>
+#include <linux/uaccess.h>
+
+#include <asm/time.h>
+#include <asm-generic/div64.h>
+
+#include "timing.h"
+
+void kvmppc_init_timing_stats(struct kvm_vcpu *vcpu)
+{
+	int i;
+
+	/* pause guest execution to avoid concurrent updates */
+	local_irq_disable();
+	mutex_lock(&vcpu->mutex);
+
+	vcpu->arch.last_exit_type = 0xDEAD;
+	for (i = 0; i < __NUMBER_OF_KVM_EXIT_TYPES; i++) {
+		vcpu->arch.timing_count_type[i] = 0;
+		vcpu->arch.timing_max_duration[i] = 0;
+		vcpu->arch.timing_min_duration[i] = 0xFFFFFFFF;
+		vcpu->arch.timing_sum_duration[i] = 0;
+		vcpu->arch.timing_sum_quad_duration[i] = 0;
+	}
+	vcpu->arch.timing_last_exit = 0;
+	vcpu->arch.timing_exit.tv64 = 0;
+	vcpu->arch.timing_last_enter.tv64 = 0;
+
+	mutex_unlock(&vcpu->mutex);
+	local_irq_enable();
+}
+
+static void add_exit_timing(struct kvm_vcpu *vcpu, u64 duration, int type)
+{
+	u64 old;
+
+	do_div(duration, tb_ticks_per_usec);
+	if (unlikely(duration > 0xFFFFFFFF)) {
+		printk(KERN_ERR"%s - duration too big -> overflow"
+			" duration %lld type %d exit #%d\n",
+			__func__, duration, type,
+			vcpu->arch.timing_count_type[type]);
+		return;
+	}
+
+	vcpu->arch.timing_count_type[type]++;
+
+	/* sum */
+	old = vcpu->arch.timing_sum_duration[type];
+	vcpu->arch.timing_sum_duration[type] += duration;
+	if (unlikely(old > vcpu->arch.timing_sum_duration[type])) {
+		printk(KERN_ERR"%s - wrap adding sum of durations"
+			" old %lld new %lld type %d exit # of type %d\n",
+			__func__, old, vcpu->arch.timing_sum_duration[type],
+			type, vcpu->arch.timing_count_type[type]);
+	}
+
+	/* square sum */
+	old = vcpu->arch.timing_sum_quad_duration[type];
+	vcpu->arch.timing_sum_quad_duration[type] += (duration*duration);
+	if (unlikely(old > vcpu->arch.timing_sum_quad_duration[type])) {
+		printk(KERN_ERR"%s - wrap adding sum of squared durations"
+			" old %lld new %lld type %d exit # of type %d\n",
+			__func__, old,
+			vcpu->arch.timing_sum_quad_duration[type],
+			type, vcpu->arch.timing_count_type[type]);
+	}
+
+	/* set min/max */
+	if (unlikely(duration < vcpu->arch.timing_min_duration[type]))
+		vcpu->arch.timing_min_duration[type] = duration;
+	if (unlikely(duration > vcpu->arch.timing_max_duration[type]))
+		vcpu->arch.timing_max_duration[type] = duration;
+}
+
+void kvmppc_update_timing_stats(struct kvm_vcpu *vcpu)
+{
+	u64 exit = vcpu->arch.timing_last_exit;
+	u64 enter = vcpu->arch.timing_last_enter.tv64;
+
+	/* save exit time, used next exit when the reenter time is known */
+	vcpu->arch.timing_last_exit = vcpu->arch.timing_exit.tv64;
+
+	if (unlikely(vcpu->arch.last_exit_type == 0xDEAD || exit == 0))
+		return; /* skip incomplete cycle (e.g. after reset) */
+
+	/* update statistics for average and standard deviation */
+	add_exit_timing(vcpu, (enter - exit), vcpu->arch.last_exit_type);
+	/* enter -> timing_last_exit is time spent in guest - log this too */
+	add_exit_timing(vcpu, (vcpu->arch.timing_last_exit - enter),
+			TIMEINGUEST);
+}
+
+static const char *kvm_exit_names[__NUMBER_OF_KVM_EXIT_TYPES] = {
+	[MMIO_EXITS] =              "MMIO",
+	[DCR_EXITS] =               "DCR",
+	[SIGNAL_EXITS] =            "SIGNAL",
+	[ITLB_REAL_MISS_EXITS] =    "ITLBREAL",
+	[ITLB_VIRT_MISS_EXITS] =    "ITLBVIRT",
+	[DTLB_REAL_MISS_EXITS] =    "DTLBREAL",
+	[DTLB_VIRT_MISS_EXITS] =    "DTLBVIRT",
+	[SYSCALL_EXITS] =           "SYSCALL",
+	[ISI_EXITS] =               "ISI",
+	[DSI_EXITS] =               "DSI",
+	[EMULATED_INST_EXITS] =     "EMULINST",
+	[EMULATED_MTMSRWE_EXITS] =  "EMUL_WAIT",
+	[EMULATED_WRTEE_EXITS] =    "EMUL_WRTEE",
+	[EMULATED_MTSPR_EXITS] =    "EMUL_MTSPR",
+	[EMULATED_MFSPR_EXITS] =    "EMUL_MFSPR",
+	[EMULATED_MTMSR_EXITS] =    "EMUL_MTMSR",
+	[EMULATED_MFMSR_EXITS] =    "EMUL_MFMSR",
+	[EMULATED_TLBSX_EXITS] =    "EMUL_TLBSX",
+	[EMULATED_TLBWE_EXITS] =    "EMUL_TLBWE",
+	[EMULATED_RFI_EXITS] =      "EMUL_RFI",
+	[DEC_EXITS] =               "DEC",
+	[EXT_INTR_EXITS] =          "EXTINT",
+	[HALT_WAKEUP] =             "HALT",
+	[USR_PR_INST] =             "USR_PR_INST",
+	[FP_UNAVAIL] =              "FP_UNAVAIL",
+	[DEBUG_EXITS] =             "DEBUG",
+	[TIMEINGUEST] =             "TIMEINGUEST"
+};
+
+static int kvmppc_exit_timing_show(struct seq_file *m, void *private)
+{
+	struct kvm_vcpu *vcpu = m->private;
+	int i;
+
+	seq_printf(m, "%s", "type	count	min	max	sum	sum_squared\n");
+
+	for (i = 0; i < __NUMBER_OF_KVM_EXIT_TYPES; i++) {
+		seq_printf(m, "%12s	%10d	%10lld	%10lld	%20lld	%20lld\n",
+			kvm_exit_names[i],
+			vcpu->arch.timing_count_type[i],
+			vcpu->arch.timing_min_duration[i],
+			vcpu->arch.timing_max_duration[i],
+			vcpu->arch.timing_sum_duration[i],
+			vcpu->arch.timing_sum_quad_duration[i]);
+	}
+	return 0;
+}
+
+/* Write 'c' to clear the timing statistics. */
+static ssize_t kvmppc_exit_timing_write(struct file *file,
+				       const char __user *user_buf,
+				       size_t count, loff_t *ppos)
+{
+	int err = -EINVAL;
+	char c;
+
+	if (count > 1) {
+		goto done;
+	}
+
+	if (get_user(c, user_buf)) {
+		err = -EFAULT;
+		goto done;
+	}
+
+	if (c == 'c') {
+		struct seq_file *seqf = (struct seq_file *)file->private_data;
+		struct kvm_vcpu *vcpu = seqf->private;
+		/* Write does not affect our buffers previously generated with
+		 * show. seq_file is locked here to prevent races of init with
+		 * a show call */
+		mutex_lock(&seqf->lock);
+		kvmppc_init_timing_stats(vcpu);
+		mutex_unlock(&seqf->lock);
+		err = count;
+	}
+
+done:
+	return err;
+}
+
+static int kvmppc_exit_timing_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, kvmppc_exit_timing_show, inode->i_private);
+}
+
+static struct file_operations kvmppc_exit_timing_fops = {
+	.owner   = THIS_MODULE,
+	.open    = kvmppc_exit_timing_open,
+	.read    = seq_read,
+	.write   = kvmppc_exit_timing_write,
+	.llseek  = seq_lseek,
+	.release = single_release,
+};
+
+void kvmppc_create_vcpu_debugfs(struct kvm_vcpu *vcpu, unsigned int id)
+{
+	static char dbg_fname[50];
+	struct dentry *debugfs_file;
+
+	snprintf(dbg_fname, sizeof(dbg_fname), "vm%u_vcpu%u_timing",
+		 current->pid, id);
+	debugfs_file = debugfs_create_file(dbg_fname, 0666,
+					kvm_debugfs_dir, vcpu,
+					&kvmppc_exit_timing_fops);
+
+	if (!debugfs_file) {
+		printk(KERN_ERR"%s: error creating debugfs file %s\n",
+			__func__, dbg_fname);
+		return;
+	}
+
+	vcpu->arch.debugfs_exit_timing = debugfs_file;
+}
+
+void kvmppc_remove_vcpu_debugfs(struct kvm_vcpu *vcpu)
+{
+	if (vcpu->arch.debugfs_exit_timing) {
+		debugfs_remove(vcpu->arch.debugfs_exit_timing);
+		vcpu->arch.debugfs_exit_timing = NULL;
+	}
+}
diff --git a/arch/powerpc/kvm/timing.h b/arch/powerpc/kvm/timing.h
new file mode 100644
index 0000000..bb13b1f
--- /dev/null
+++ b/arch/powerpc/kvm/timing.h
@@ -0,0 +1,102 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
+ */
+
+#ifndef __POWERPC_KVM_EXITTIMING_H__
+#define __POWERPC_KVM_EXITTIMING_H__
+
+#include <linux/kvm_host.h>
+#include <asm/kvm_host.h>
+
+#ifdef CONFIG_KVM_EXIT_TIMING
+void kvmppc_init_timing_stats(struct kvm_vcpu *vcpu);
+void kvmppc_update_timing_stats(struct kvm_vcpu *vcpu);
+void kvmppc_create_vcpu_debugfs(struct kvm_vcpu *vcpu, unsigned int id);
+void kvmppc_remove_vcpu_debugfs(struct kvm_vcpu *vcpu);
+
+static inline void kvmppc_set_exit_type(struct kvm_vcpu *vcpu, int type)
+{
+	vcpu->arch.last_exit_type = type;
+}
+
+#else
+/* if exit timing is not configured there is no need to build the c file */
+static inline void kvmppc_init_timing_stats(struct kvm_vcpu *vcpu) {}
+static inline void kvmppc_update_timing_stats(struct kvm_vcpu *vcpu) {}
+static inline void kvmppc_create_vcpu_debugfs(struct kvm_vcpu *vcpu,
+						unsigned int id) {}
+static inline void kvmppc_remove_vcpu_debugfs(struct kvm_vcpu *vcpu) {}
+static inline void kvmppc_set_exit_type(struct kvm_vcpu *vcpu, int type) {}
+#endif /* CONFIG_KVM_EXIT_TIMING */
+
+/* account the exit in kvm_stats */
+static inline void kvmppc_account_exit_stat(struct kvm_vcpu *vcpu, int type)
+{
+	/* type has to be known at build time for optimization */
+	BUILD_BUG_ON(__builtin_constant_p(type));
+	switch (type) {
+	case EXT_INTR_EXITS:
+		vcpu->stat.ext_intr_exits++;
+		break;
+	case DEC_EXITS:
+		vcpu->stat.dec_exits++;
+		break;
+	case EMULATED_INST_EXITS:
+		vcpu->stat.emulated_inst_exits++;
+		break;
+	case DCR_EXITS:
+		vcpu->stat.dcr_exits++;
+		break;
+	case DSI_EXITS:
+		vcpu->stat.dsi_exits++;
+		break;
+	case ISI_EXITS:
+		vcpu->stat.isi_exits++;
+		break;
+	case SYSCALL_EXITS:
+		vcpu->stat.syscall_exits++;
+		break;
+	case DTLB_REAL_MISS_EXITS:
+		vcpu->stat.dtlb_real_miss_exits++;
+		break;
+	case DTLB_VIRT_MISS_EXITS:
+		vcpu->stat.dtlb_virt_miss_exits++;
+		break;
+	case MMIO_EXITS:
+		vcpu->stat.mmio_exits++;
+		break;
+	case ITLB_REAL_MISS_EXITS:
+		vcpu->stat.itlb_real_miss_exits++;
+		break;
+	case ITLB_VIRT_MISS_EXITS:
+		vcpu->stat.itlb_virt_miss_exits++;
+		break;
+	case SIGNAL_EXITS:
+		vcpu->stat.signal_exits++;
+		break;
+	}
+}
+
+/* wrapper to set exit time and account for it in kvm_stats */
+static inline void kvmppc_account_exit(struct kvm_vcpu *vcpu, int type)
+{
+	kvmppc_set_exit_type(vcpu, type);
+	kvmppc_account_exit_stat(vcpu, type);
+}
+
+#endif /* __POWERPC_KVM_EXITTIMING_H__ */
diff --git a/arch/powerpc/oprofile/cell/spu_task_sync.c b/arch/powerpc/oprofile/cell/spu_task_sync.c
index 2949126..6b793ae 100644
--- a/arch/powerpc/oprofile/cell/spu_task_sync.c
+++ b/arch/powerpc/oprofile/cell/spu_task_sync.c
@@ -297,7 +297,7 @@
 {
 	unsigned long cookie;
 
-	if (path->dentry->d_cookie)
+	if (path->dentry->d_flags & DCACHE_COOKIE)
 		return (unsigned long)path->dentry;
 	get_dcookie(path, &cookie);
 	return cookie;
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index f7a6902..84e058f 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -332,7 +332,7 @@
 	lpar_xirr_info_set((0xff << 24) | irq);
 }
 
-static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
+static void xics_set_affinity(unsigned int virq, const struct cpumask *cpumask)
 {
 	unsigned int irq;
 	int status;
@@ -870,7 +870,7 @@
 
 		/* Reset affinity to all cpus */
 		irq_desc[virq].affinity = CPU_MASK_ALL;
-		desc->chip->set_affinity(virq, CPU_MASK_ALL);
+		desc->chip->set_affinity(virq, cpu_all_mask);
 unlock:
 		spin_unlock_irqrestore(&desc->lock, flags);
 	}
diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index c82babb..3e0d89d 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -806,7 +806,7 @@
 
 #endif /* CONFIG_SMP */
 
-void mpic_set_affinity(unsigned int irq, cpumask_t cpumask)
+void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
 	struct mpic *mpic = mpic_from_irq(irq);
 	unsigned int src = mpic_irq_to_hw(irq);
@@ -818,7 +818,7 @@
 	} else {
 		cpumask_t tmp;
 
-		cpus_and(tmp, cpumask, cpu_online_map);
+		cpumask_and(&tmp, cpumask, cpu_online_mask);
 
 		mpic_irq_write(src, MPIC_INFO(IRQ_DESTINATION),
 			       mpic_physmask(cpus_addr(tmp)[0]));
diff --git a/arch/powerpc/sysdev/mpic.h b/arch/powerpc/sysdev/mpic.h
index 6209c62..3cef2af 100644
--- a/arch/powerpc/sysdev/mpic.h
+++ b/arch/powerpc/sysdev/mpic.h
@@ -36,6 +36,6 @@
 
 extern int mpic_set_irq_type(unsigned int virq, unsigned int flow_type);
 extern void mpic_set_vector(unsigned int virq, unsigned int vector);
-extern void mpic_set_affinity(unsigned int irq, cpumask_t cpumask);
+extern void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask);
 
 #endif /* _POWERPC_SYSDEV_MPIC_H */
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 8152fef..19577ae 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -83,6 +83,7 @@
 	select HAVE_KRETPROBES
 	select HAVE_KVM if 64BIT
 	select HAVE_ARCH_TRACEHOOK
+	select INIT_ALL_POSSIBLE
 
 source "init/Kconfig"
 
diff --git a/arch/s390/include/asm/cpu.h b/arch/s390/include/asm/cpu.h
index e5a6a9b..d60a2ee 100644
--- a/arch/s390/include/asm/cpu.h
+++ b/arch/s390/include/asm/cpu.h
@@ -14,7 +14,6 @@
 
 struct s390_idle_data {
 	spinlock_t lock;
-	unsigned int in_idle;
 	unsigned long long idle_count;
 	unsigned long long idle_enter;
 	unsigned long long idle_time;
@@ -22,12 +21,12 @@
 
 DECLARE_PER_CPU(struct s390_idle_data, s390_idle);
 
-void s390_idle_leave(void);
+void vtime_start_cpu(void);
 
 static inline void s390_idle_check(void)
 {
-	if ((&__get_cpu_var(s390_idle))->in_idle)
-		s390_idle_leave();
+	if ((&__get_cpu_var(s390_idle))->idle_enter != 0ULL)
+		vtime_start_cpu();
 }
 
 #endif /* _ASM_S390_CPU_H_ */
diff --git a/arch/s390/include/asm/cputime.h b/arch/s390/include/asm/cputime.h
index 133ce05..5217264 100644
--- a/arch/s390/include/asm/cputime.h
+++ b/arch/s390/include/asm/cputime.h
@@ -11,7 +11,7 @@
 
 #include <asm/div64.h>
 
-/* We want to use micro-second resolution. */
+/* We want to use full resolution of the CPU timer: 2**-12 micro-seconds. */
 
 typedef unsigned long long cputime_t;
 typedef unsigned long long cputime64_t;
@@ -53,9 +53,9 @@
 #define cputime_ge(__a, __b)		((__a) >= (__b))
 #define cputime_lt(__a, __b)		((__a) <  (__b))
 #define cputime_le(__a, __b)		((__a) <= (__b))
-#define cputime_to_jiffies(__ct)	(__div((__ct), 1000000 / HZ))
+#define cputime_to_jiffies(__ct)	(__div((__ct), 4096000000ULL / HZ))
 #define cputime_to_scaled(__ct)		(__ct)
-#define jiffies_to_cputime(__hz)	((cputime_t)(__hz) * (1000000 / HZ))
+#define jiffies_to_cputime(__hz)	((cputime_t)(__hz) * (4096000000ULL / HZ))
 
 #define cputime64_zero			(0ULL)
 #define cputime64_add(__a, __b)		((__a) + (__b))
@@ -64,7 +64,7 @@
 static inline u64
 cputime64_to_jiffies64(cputime64_t cputime)
 {
-	do_div(cputime, 1000000 / HZ);
+	do_div(cputime, 4096000000ULL / HZ);
 	return cputime;
 }
 
@@ -74,13 +74,13 @@
 static inline unsigned int
 cputime_to_msecs(const cputime_t cputime)
 {
-	return __div(cputime, 1000);
+	return __div(cputime, 4096000);
 }
 
 static inline cputime_t
 msecs_to_cputime(const unsigned int m)
 {
-	return (cputime_t) m * 1000;
+	return (cputime_t) m * 4096000;
 }
 
 /*
@@ -89,13 +89,13 @@
 static inline unsigned int
 cputime_to_secs(const cputime_t cputime)
 {
-	return __div(cputime, 1000000);
+	return __div(cputime, 2048000000) >> 1;
 }
 
 static inline cputime_t
 secs_to_cputime(const unsigned int s)
 {
-	return (cputime_t) s * 1000000;
+	return (cputime_t) s * 4096000000ULL;
 }
 
 /*
@@ -104,7 +104,7 @@
 static inline cputime_t
 timespec_to_cputime(const struct timespec *value)
 {
-        return value->tv_nsec / 1000 + (u64) value->tv_sec * 1000000;
+	return value->tv_nsec * 4096 / 1000 + (u64) value->tv_sec * 4096000000ULL;
 }
 
 static inline void
@@ -114,12 +114,12 @@
 	register_pair rp;
 
 	rp.pair = cputime >> 1;
-	asm ("dr %0,%1" : "+d" (rp) : "d" (1000000 >> 1));
-	value->tv_nsec = rp.subreg.even * 1000;
+	asm ("dr %0,%1" : "+d" (rp) : "d" (2048000000UL));
+	value->tv_nsec = rp.subreg.even * 1000 / 4096;
 	value->tv_sec = rp.subreg.odd;
 #else
-	value->tv_nsec = (cputime % 1000000) * 1000;
-	value->tv_sec = cputime / 1000000;
+	value->tv_nsec = (cputime % 4096000000ULL) * 1000 / 4096;
+	value->tv_sec = cputime / 4096000000ULL;
 #endif
 }
 
@@ -131,7 +131,7 @@
 static inline cputime_t
 timeval_to_cputime(const struct timeval *value)
 {
-        return value->tv_usec + (u64) value->tv_sec * 1000000;
+	return value->tv_usec * 4096 + (u64) value->tv_sec * 4096000000ULL;
 }
 
 static inline void
@@ -141,12 +141,12 @@
 	register_pair rp;
 
 	rp.pair = cputime >> 1;
-	asm ("dr %0,%1" : "+d" (rp) : "d" (1000000 >> 1));
-	value->tv_usec = rp.subreg.even;
+	asm ("dr %0,%1" : "+d" (rp) : "d" (2048000000UL));
+	value->tv_usec = rp.subreg.even / 4096;
 	value->tv_sec = rp.subreg.odd;
 #else
-	value->tv_usec = cputime % 1000000;
-	value->tv_sec = cputime / 1000000;
+	value->tv_usec = cputime % 4096000000ULL;
+	value->tv_sec = cputime / 4096000000ULL;
 #endif
 }
 
@@ -156,13 +156,13 @@
 static inline clock_t
 cputime_to_clock_t(cputime_t cputime)
 {
-	return __div(cputime, 1000000 / USER_HZ);
+	return __div(cputime, 4096000000ULL / USER_HZ);
 }
 
 static inline cputime_t
 clock_t_to_cputime(unsigned long x)
 {
-	return (cputime_t) x * (1000000 / USER_HZ);
+	return (cputime_t) x * (4096000000ULL / USER_HZ);
 }
 
 /*
@@ -171,7 +171,7 @@
 static inline clock_t
 cputime64_to_clock_t(cputime64_t cputime)
 {
-       return __div(cputime, 1000000 / USER_HZ);
+       return __div(cputime, 4096000000ULL / USER_HZ);
 }
 
 #endif /* _S390_CPUTIME_H */
diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h
index 0bc51d5..ffdef5f 100644
--- a/arch/s390/include/asm/lowcore.h
+++ b/arch/s390/include/asm/lowcore.h
@@ -67,11 +67,11 @@
 #define __LC_SYNC_ENTER_TIMER		0x248
 #define __LC_ASYNC_ENTER_TIMER		0x250
 #define __LC_EXIT_TIMER			0x258
-#define __LC_LAST_UPDATE_TIMER		0x260
-#define __LC_USER_TIMER			0x268
-#define __LC_SYSTEM_TIMER		0x270
-#define __LC_LAST_UPDATE_CLOCK		0x278
-#define __LC_STEAL_CLOCK		0x280
+#define __LC_USER_TIMER			0x260
+#define __LC_SYSTEM_TIMER		0x268
+#define __LC_STEAL_TIMER		0x270
+#define __LC_LAST_UPDATE_TIMER		0x278
+#define __LC_LAST_UPDATE_CLOCK		0x280
 #define __LC_RETURN_MCCK_PSW            0x288
 #define __LC_KERNEL_STACK               0xC40
 #define __LC_THREAD_INFO		0xC44
@@ -89,11 +89,11 @@
 #define __LC_SYNC_ENTER_TIMER		0x250
 #define __LC_ASYNC_ENTER_TIMER		0x258
 #define __LC_EXIT_TIMER			0x260
-#define __LC_LAST_UPDATE_TIMER		0x268
-#define __LC_USER_TIMER			0x270
-#define __LC_SYSTEM_TIMER		0x278
-#define __LC_LAST_UPDATE_CLOCK		0x280
-#define __LC_STEAL_CLOCK		0x288
+#define __LC_USER_TIMER			0x268
+#define __LC_SYSTEM_TIMER		0x270
+#define __LC_STEAL_TIMER		0x278
+#define __LC_LAST_UPDATE_TIMER		0x280
+#define __LC_LAST_UPDATE_CLOCK		0x288
 #define __LC_RETURN_MCCK_PSW            0x290
 #define __LC_KERNEL_STACK               0xD40
 #define __LC_THREAD_INFO		0xD48
@@ -106,8 +106,10 @@
 #define __LC_IPLDEV                     0xDB8
 #define __LC_CURRENT			0xDD8
 #define __LC_INT_CLOCK			0xDE8
+#define __LC_VDSO_PER_CPU		0xE38
 #endif /* __s390x__ */
 
+#define __LC_PASTE			0xE40
 
 #define __LC_PANIC_MAGIC		0xE00
 #ifndef __s390x__
@@ -252,11 +254,11 @@
 	__u64        sync_enter_timer;         /* 0x248 */
 	__u64        async_enter_timer;        /* 0x250 */
 	__u64        exit_timer;               /* 0x258 */
-	__u64        last_update_timer;        /* 0x260 */
-	__u64        user_timer;               /* 0x268 */
-	__u64        system_timer;             /* 0x270 */
-	__u64        last_update_clock;        /* 0x278 */
-	__u64        steal_clock;              /* 0x280 */
+	__u64	     user_timer;	       /* 0x260 */
+	__u64	     system_timer;	       /* 0x268 */
+	__u64	     steal_timer;	       /* 0x270 */
+	__u64	     last_update_timer;        /* 0x278 */
+	__u64	     last_update_clock;        /* 0x280 */
         psw_t        return_mcck_psw;          /* 0x288 */
 	__u8         pad8[0xc00-0x290];        /* 0x290 */
 
@@ -343,11 +345,11 @@
 	__u64        sync_enter_timer;         /* 0x250 */
 	__u64        async_enter_timer;        /* 0x258 */
 	__u64        exit_timer;               /* 0x260 */
-	__u64        last_update_timer;        /* 0x268 */
-	__u64        user_timer;               /* 0x270 */
-	__u64        system_timer;             /* 0x278 */
-	__u64        last_update_clock;        /* 0x280 */
-	__u64        steal_clock;              /* 0x288 */
+	__u64	     user_timer;	       /* 0x268 */
+	__u64	     system_timer;	       /* 0x270 */
+	__u64	     steal_timer;	       /* 0x278 */
+	__u64	     last_update_timer;        /* 0x280 */
+	__u64	     last_update_clock;        /* 0x288 */
         psw_t        return_mcck_psw;          /* 0x290 */
         __u8         pad8[0xc00-0x2a0];        /* 0x2a0 */
         /* System info area */
@@ -381,7 +383,12 @@
         /* whether the kernel died with panic() or not */
         __u32        panic_magic;              /* 0xe00 */
 
-	__u8         pad13[0x11b8-0xe04];      /* 0xe04 */
+	/* Per cpu primary space access list */
+	__u8	     pad_0xe04[0xe3c-0xe04];   /* 0xe04 */
+	__u32	     vdso_per_cpu_data;	       /* 0xe3c */
+	__u32	     paste[16];		       /* 0xe40 */
+
+	__u8	     pad13[0x11b8-0xe80];      /* 0xe80 */
 
 	/* 64 bit extparam used for pfault, diag 250 etc  */
 	__u64        ext_params2;               /* 0x11B8 */
diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
index 024ef42..3a8b26e 100644
--- a/arch/s390/include/asm/system.h
+++ b/arch/s390/include/asm/system.h
@@ -99,7 +99,7 @@
 	prev = __switch_to(prev,next);					     \
 } while (0)
 
-extern void account_vtime(struct task_struct *);
+extern void account_vtime(struct task_struct *, struct task_struct *);
 extern void account_tick_vtime(struct task_struct *);
 extern void account_system_vtime(struct task_struct *);
 
@@ -121,7 +121,7 @@
 
 #define finish_arch_switch(prev) do {					     \
 	set_fs(current->thread.mm_segment);				     \
-	account_vtime(prev);						     \
+	account_vtime(prev, current);					     \
 } while (0)
 
 #define nop() asm volatile("nop")
diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h
index c1eaf96..c544aa5 100644
--- a/arch/s390/include/asm/thread_info.h
+++ b/arch/s390/include/asm/thread_info.h
@@ -47,6 +47,8 @@
 	unsigned int		cpu;		/* current CPU */
 	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 	struct restart_block	restart_block;
+	__u64			user_timer;
+	__u64			system_timer;
 };
 
 /*
diff --git a/arch/s390/include/asm/timer.h b/arch/s390/include/asm/timer.h
index 61705d60..e4bcab7 100644
--- a/arch/s390/include/asm/timer.h
+++ b/arch/s390/include/asm/timer.h
@@ -23,20 +23,18 @@
 	__u64 expires;
 	__u64 interval;
 
-	spinlock_t lock;
-	unsigned long magic;
-
 	void (*function)(unsigned long);
 	unsigned long data;
 };
 
-/* the offset value will wrap after ca. 71 years */
+/* the vtimer value will wrap after ca. 71 years */
 struct vtimer_queue {
 	struct list_head list;
 	spinlock_t lock;
-	__u64 to_expire;	  /* current event expire time */
-	__u64 offset;		  /* list offset to zero */
-	__u64 idle;		  /* temp var for idle */
+	__u64 timer;		/* last programmed timer */
+	__u64 elapsed;		/* elapsed time of timer expire values */
+	__u64 idle;		/* temp var for idle */
+	int do_spt;		/* =1: reprogram cpu timer in idle */
 };
 
 extern void init_virt_timer(struct vtimer_list *timer);
@@ -48,8 +46,8 @@
 extern void init_cpu_vtimer(void);
 extern void vtime_init(void);
 
-extern void vtime_start_cpu_timer(void);
-extern void vtime_stop_cpu_timer(void);
+extern void vtime_stop_cpu(void);
+extern void vtime_start_leave(void);
 
 #endif /* __KERNEL__ */
 
diff --git a/arch/s390/include/asm/vdso.h b/arch/s390/include/asm/vdso.h
index a44f4fe..7bdd7c8 100644
--- a/arch/s390/include/asm/vdso.h
+++ b/arch/s390/include/asm/vdso.h
@@ -12,9 +12,9 @@
 #ifndef __ASSEMBLY__
 
 /*
- * Note about this structure:
+ * Note about the vdso_data and vdso_per_cpu_data structures:
  *
- * NEVER USE THIS IN USERSPACE CODE DIRECTLY. The layout of this
+ * NEVER USE THEM IN USERSPACE CODE DIRECTLY. The layout of the
  * structure is supposed to be known only to the function in the vdso
  * itself and may change without notice.
  */
@@ -28,10 +28,21 @@
 	__u64 wtom_clock_nsec;		/*				0x28 */
 	__u32 tz_minuteswest;		/* Minutes west of Greenwich	0x30 */
 	__u32 tz_dsttime;		/* Type of dst correction	0x34 */
+	__u32 ectg_available;
+};
+
+struct vdso_per_cpu_data {
+	__u64 ectg_timer_base;
+	__u64 ectg_user_time;
 };
 
 extern struct vdso_data *vdso_data;
 
+#ifdef CONFIG_64BIT
+int vdso_alloc_per_cpu(int cpu, struct _lowcore *lowcore);
+void vdso_free_per_cpu(int cpu, struct _lowcore *lowcore);
+#endif
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __KERNEL__ */
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index e641f60..67a6001 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -48,6 +48,11 @@
 	DEFINE(__VDSO_WTOM_SEC, offsetof(struct vdso_data, wtom_clock_sec));
 	DEFINE(__VDSO_WTOM_NSEC, offsetof(struct vdso_data, wtom_clock_nsec));
 	DEFINE(__VDSO_TIMEZONE, offsetof(struct vdso_data, tz_minuteswest));
+	DEFINE(__VDSO_ECTG_OK, offsetof(struct vdso_data, ectg_available));
+	DEFINE(__VDSO_ECTG_BASE,
+	       offsetof(struct vdso_per_cpu_data, ectg_timer_base));
+	DEFINE(__VDSO_ECTG_USER,
+	       offsetof(struct vdso_per_cpu_data, ectg_user_time));
 	/* constants used by the vdso */
 	DEFINE(CLOCK_REALTIME, CLOCK_REALTIME);
 	DEFINE(CLOCK_MONOTONIC, CLOCK_MONOTONIC);
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 55de521..1268aa2 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -583,8 +583,8 @@
 
 	.globl io_int_handler
 io_int_handler:
-	stpt	__LC_ASYNC_ENTER_TIMER
 	stck	__LC_INT_CLOCK
+	stpt	__LC_ASYNC_ENTER_TIMER
 	SAVE_ALL_BASE __LC_SAVE_AREA+16
 	SAVE_ALL_ASYNC __LC_IO_OLD_PSW,__LC_SAVE_AREA+16
 	CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+16
@@ -723,8 +723,8 @@
 
 	.globl	ext_int_handler
 ext_int_handler:
-	stpt	__LC_ASYNC_ENTER_TIMER
 	stck	__LC_INT_CLOCK
+	stpt	__LC_ASYNC_ENTER_TIMER
 	SAVE_ALL_BASE __LC_SAVE_AREA+16
 	SAVE_ALL_ASYNC __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16
 	CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16
@@ -750,6 +750,7 @@
 
 	.globl mcck_int_handler
 mcck_int_handler:
+	stck	__LC_INT_CLOCK
 	spt	__LC_CPU_TIMER_SAVE_AREA	# revalidate cpu timer
 	lm	%r0,%r15,__LC_GPREGS_SAVE_AREA	# revalidate gprs
 	SAVE_ALL_BASE __LC_SAVE_AREA+32
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index 16bb4fd..c6fbde1 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -177,8 +177,11 @@
 	.if !\sync
 	ni	\psworg+1,0xfd		# clear wait state bit
 	.endif
-	lmg	%r0,%r15,SP_R0(%r15)	# load gprs 0-15 of user
+	lg	%r14,__LC_VDSO_PER_CPU
+	lmg	%r0,%r13,SP_R0(%r15)	# load gprs 0-13 of user
 	stpt	__LC_EXIT_TIMER
+	mvc	__VDSO_ECTG_BASE(16,%r14),__LC_EXIT_TIMER
+	lmg	%r14,%r15,SP_R14(%r15)	# load grps 14-15 of user
 	lpswe	\psworg			# back to caller
 	.endm
 
@@ -559,8 +562,8 @@
  */
 	.globl io_int_handler
 io_int_handler:
-	stpt	__LC_ASYNC_ENTER_TIMER
 	stck	__LC_INT_CLOCK
+	stpt	__LC_ASYNC_ENTER_TIMER
 	SAVE_ALL_BASE __LC_SAVE_AREA+32
 	SAVE_ALL_ASYNC __LC_IO_OLD_PSW,__LC_SAVE_AREA+32
 	CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+32
@@ -721,8 +724,8 @@
  */
 	.globl	ext_int_handler
 ext_int_handler:
-	stpt	__LC_ASYNC_ENTER_TIMER
 	stck	__LC_INT_CLOCK
+	stpt	__LC_ASYNC_ENTER_TIMER
 	SAVE_ALL_BASE __LC_SAVE_AREA+32
 	SAVE_ALL_ASYNC __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32
 	CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32
@@ -746,6 +749,7 @@
  */
 	.globl mcck_int_handler
 mcck_int_handler:
+	stck	__LC_INT_CLOCK
 	la	%r1,4095		# revalidate r1
 	spt	__LC_CPU_TIMER_SAVE_AREA-4095(%r1)	# revalidate cpu timer
 	lmg	%r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)# revalidate gprs
@@ -979,23 +983,23 @@
 
 cleanup_sysc_leave:
 	clc	8(8,%r12),BASED(cleanup_sysc_leave_insn)
-	je	2f
-	mvc	__LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER
+	je	3f
 	clc	8(8,%r12),BASED(cleanup_sysc_leave_insn+8)
-	je	2f
-	mvc	__LC_RETURN_PSW(16),SP_PSW(%r15)
+	jhe	0f
+	mvc	__LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER
+0:	mvc	__LC_RETURN_PSW(16),SP_PSW(%r15)
 	cghi	%r12,__LC_MCK_OLD_PSW
-	jne	0f
+	jne	1f
 	mvc	__LC_SAVE_AREA+64(32),SP_R12(%r15)
-	j	1f
-0:	mvc	__LC_SAVE_AREA+32(32),SP_R12(%r15)
-1:	lmg	%r0,%r11,SP_R0(%r15)
+	j	2f
+1:	mvc	__LC_SAVE_AREA+32(32),SP_R12(%r15)
+2:	lmg	%r0,%r11,SP_R0(%r15)
 	lg	%r15,SP_R15(%r15)
-2:	la	%r12,__LC_RETURN_PSW
+3:	la	%r12,__LC_RETURN_PSW
 	br	%r14
 cleanup_sysc_leave_insn:
 	.quad	sysc_done - 4
-	.quad	sysc_done - 8
+	.quad	sysc_done - 16
 
 cleanup_io_return:
 	mvc	__LC_RETURN_PSW(8),0(%r12)
@@ -1005,23 +1009,23 @@
 
 cleanup_io_leave:
 	clc	8(8,%r12),BASED(cleanup_io_leave_insn)
-	je	2f
-	mvc	__LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER
+	je	3f
 	clc	8(8,%r12),BASED(cleanup_io_leave_insn+8)
-	je	2f
-	mvc	__LC_RETURN_PSW(16),SP_PSW(%r15)
+	jhe	0f
+	mvc	__LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER
+0:	mvc	__LC_RETURN_PSW(16),SP_PSW(%r15)
 	cghi	%r12,__LC_MCK_OLD_PSW
-	jne	0f
+	jne	1f
 	mvc	__LC_SAVE_AREA+64(32),SP_R12(%r15)
-	j	1f
-0:	mvc	__LC_SAVE_AREA+32(32),SP_R12(%r15)
-1:	lmg	%r0,%r11,SP_R0(%r15)
+	j	2f
+1:	mvc	__LC_SAVE_AREA+32(32),SP_R12(%r15)
+2:	lmg	%r0,%r11,SP_R0(%r15)
 	lg	%r15,SP_R15(%r15)
-2:	la	%r12,__LC_RETURN_PSW
+3:	la	%r12,__LC_RETURN_PSW
 	br	%r14
 cleanup_io_leave_insn:
 	.quad	io_done - 4
-	.quad	io_done - 8
+	.quad	io_done - 16
 
 /*
  * Integer constants
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 3ccd36b..f9f70aa 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -87,6 +87,8 @@
 	lg	%r12,.Lparmaddr-.LPG1(%r13)	# pointer to parameter area
 					# move IPL device to lowcore
 	mvc	__LC_IPLDEV(4),IPL_DEVICE+4-PARMAREA(%r12)
+	lghi	%r0,__LC_PASTE
+	stg	%r0,__LC_VDSO_PER_CPU
 #
 # Setup stack
 #
diff --git a/arch/s390/kernel/init_task.c b/arch/s390/kernel/init_task.c
index e807168..7db95c0 100644
--- a/arch/s390/kernel/init_task.c
+++ b/arch/s390/kernel/init_task.c
@@ -16,7 +16,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 04f8c67..b6110bd 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -38,6 +38,7 @@
 #include <linux/utsname.h>
 #include <linux/tick.h>
 #include <linux/elfcore.h>
+#include <linux/kernel_stat.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/system.h>
@@ -45,7 +46,6 @@
 #include <asm/processor.h>
 #include <asm/irq.h>
 #include <asm/timer.h>
-#include <asm/cpu.h>
 #include "entry.h"
 
 asmlinkage void ret_from_fork(void) asm ("ret_from_fork");
@@ -75,36 +75,6 @@
 	return sf->gprs[8];
 }
 
-DEFINE_PER_CPU(struct s390_idle_data, s390_idle) = {
-	.lock = __SPIN_LOCK_UNLOCKED(s390_idle.lock)
-};
-
-static int s390_idle_enter(void)
-{
-	struct s390_idle_data *idle;
-
-	idle = &__get_cpu_var(s390_idle);
-	spin_lock(&idle->lock);
-	idle->idle_count++;
-	idle->in_idle = 1;
-	idle->idle_enter = get_clock();
-	spin_unlock(&idle->lock);
-	vtime_stop_cpu_timer();
-	return NOTIFY_OK;
-}
-
-void s390_idle_leave(void)
-{
-	struct s390_idle_data *idle;
-
-	vtime_start_cpu_timer();
-	idle = &__get_cpu_var(s390_idle);
-	spin_lock(&idle->lock);
-	idle->idle_time += get_clock() - idle->idle_enter;
-	idle->in_idle = 0;
-	spin_unlock(&idle->lock);
-}
-
 extern void s390_handle_mcck(void);
 /*
  * The idle loop on a S390...
@@ -117,10 +87,6 @@
 		local_irq_enable();
 		return;
 	}
-	if (s390_idle_enter() == NOTIFY_BAD) {
-		local_irq_enable();
-		return;
-	}
 #ifdef CONFIG_HOTPLUG_CPU
 	if (cpu_is_offline(smp_processor_id())) {
 		preempt_enable_no_resched();
@@ -130,7 +96,6 @@
 	local_mcck_disable();
 	if (test_thread_flag(TIF_MCCK_PENDING)) {
 		local_mcck_enable();
-		s390_idle_leave();
 		local_irq_enable();
 		s390_handle_mcck();
 		return;
@@ -138,9 +103,9 @@
 	trace_hardirqs_on();
 	/* Don't trace preempt off for idle. */
 	stop_critical_timings();
-	/* Wait for external, I/O or machine check interrupt. */
-	__load_psw_mask(psw_kernel_bits | PSW_MASK_WAIT |
-			PSW_MASK_IO | PSW_MASK_EXT);
+	/* Stop virtual timer and halt the cpu. */
+	vtime_stop_cpu();
+	/* Reenable preemption tracer. */
 	start_critical_timings();
 }
 
diff --git a/arch/s390/kernel/s390_ext.c b/arch/s390/kernel/s390_ext.c
index e019b41..a0d2d55 100644
--- a/arch/s390/kernel/s390_ext.c
+++ b/arch/s390/kernel/s390_ext.c
@@ -119,8 +119,8 @@
 	struct pt_regs *old_regs;
 
 	old_regs = set_irq_regs(regs);
-	irq_enter();
 	s390_idle_check();
+	irq_enter();
 	if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator)
 		/* Serve timer interrupts first. */
 		clock_comparator_work();
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index b7a1efd..d825f49 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -427,6 +427,8 @@
 		/* enable extended save area */
 		__ctl_set_bit(14, 29);
 	}
+#else
+	lc->vdso_per_cpu_data = (unsigned long) &lc->paste[0];
 #endif
 	set_prefix((u32)(unsigned long) lc);
 }
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 6fc7854..9c0ccb5 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -47,6 +47,7 @@
 #include <asm/lowcore.h>
 #include <asm/sclp.h>
 #include <asm/cpu.h>
+#include <asm/vdso.h>
 #include "entry.h"
 
 /*
@@ -55,12 +56,6 @@
 struct _lowcore *lowcore_ptr[NR_CPUS];
 EXPORT_SYMBOL(lowcore_ptr);
 
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
-
-cpumask_t cpu_possible_map = CPU_MASK_ALL;
-EXPORT_SYMBOL(cpu_possible_map);
-
 static struct task_struct *current_set[NR_CPUS];
 
 static u8 smp_cpu_type;
@@ -506,6 +501,9 @@
 			goto out;
 		lowcore->extended_save_area_addr = (u32) save_area;
 	}
+#else
+	if (vdso_alloc_per_cpu(cpu, lowcore))
+		goto out;
 #endif
 	lowcore_ptr[cpu] = lowcore;
 	return 0;
@@ -528,6 +526,8 @@
 #ifndef CONFIG_64BIT
 	if (MACHINE_HAS_IEEE)
 		free_page((unsigned long) lowcore->extended_save_area_addr);
+#else
+	vdso_free_per_cpu(cpu, lowcore);
 #endif
 	free_page(lowcore->panic_stack - PAGE_SIZE);
 	free_pages(lowcore->async_stack - ASYNC_SIZE, ASYNC_ORDER);
@@ -670,6 +670,7 @@
 	lowcore = (void *) __get_free_pages(GFP_KERNEL | GFP_DMA, lc_order);
 	panic_stack = __get_free_page(GFP_KERNEL);
 	async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER);
+	BUG_ON(!lowcore || !panic_stack || !async_stack);
 #ifndef CONFIG_64BIT
 	if (MACHINE_HAS_IEEE)
 		save_area = get_zeroed_page(GFP_KERNEL);
@@ -683,6 +684,8 @@
 #ifndef CONFIG_64BIT
 	if (MACHINE_HAS_IEEE)
 		lowcore->extended_save_area_addr = (u32) save_area;
+#else
+	BUG_ON(vdso_alloc_per_cpu(smp_processor_id(), lowcore));
 #endif
 	set_prefix((u32)(unsigned long) lowcore);
 	local_mcck_enable();
@@ -851,9 +854,11 @@
 	unsigned long long idle_count;
 
 	idle = &per_cpu(s390_idle, dev->id);
-	spin_lock_irq(&idle->lock);
+	spin_lock(&idle->lock);
 	idle_count = idle->idle_count;
-	spin_unlock_irq(&idle->lock);
+	if (idle->idle_enter)
+		idle_count++;
+	spin_unlock(&idle->lock);
 	return sprintf(buf, "%llu\n", idle_count);
 }
 static SYSDEV_ATTR(idle_count, 0444, show_idle_count, NULL);
@@ -862,18 +867,17 @@
 				struct sysdev_attribute *attr, char *buf)
 {
 	struct s390_idle_data *idle;
-	unsigned long long new_time;
+	unsigned long long now, idle_time, idle_enter;
 
 	idle = &per_cpu(s390_idle, dev->id);
-	spin_lock_irq(&idle->lock);
-	if (idle->in_idle) {
-		new_time = get_clock();
-		idle->idle_time += new_time - idle->idle_enter;
-		idle->idle_enter = new_time;
-	}
-	new_time = idle->idle_time;
-	spin_unlock_irq(&idle->lock);
-	return sprintf(buf, "%llu\n", new_time >> 12);
+	spin_lock(&idle->lock);
+	now = get_clock();
+	idle_time = idle->idle_time;
+	idle_enter = idle->idle_enter;
+	if (idle_enter != 0ULL && idle_enter < now)
+		idle_time += now - idle_enter;
+	spin_unlock(&idle->lock);
+	return sprintf(buf, "%llu\n", idle_time >> 12);
 }
 static SYSDEV_ATTR(idle_time_us, 0444, show_idle_time, NULL);
 
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 5be981a..d649600 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -160,7 +160,7 @@
 	cd->min_delta_ns	= 1;
 	cd->max_delta_ns	= LONG_MAX;
 	cd->rating		= 400;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= s390_next_event;
 	cd->set_mode		= s390_set_mode;
 
diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c
index 10a6cce..25a6a82 100644
--- a/arch/s390/kernel/vdso.c
+++ b/arch/s390/kernel/vdso.c
@@ -31,9 +31,6 @@
 #include <asm/sections.h>
 #include <asm/vdso.h>
 
-/* Max supported size for symbol names */
-#define MAX_SYMNAME	64
-
 #if defined(CONFIG_32BIT) || defined(CONFIG_COMPAT)
 extern char vdso32_start, vdso32_end;
 static void *vdso32_kbase = &vdso32_start;
@@ -71,6 +68,119 @@
 struct vdso_data *vdso_data = &vdso_data_store.data;
 
 /*
+ * Setup vdso data page.
+ */
+static void vdso_init_data(struct vdso_data *vd)
+{
+	unsigned int facility_list;
+
+	facility_list = stfl();
+	vd->ectg_available = switch_amode && (facility_list & 1);
+}
+
+#ifdef CONFIG_64BIT
+/*
+ * Setup per cpu vdso data page.
+ */
+static void vdso_init_per_cpu_data(int cpu, struct vdso_per_cpu_data *vpcd)
+{
+}
+
+/*
+ * Allocate/free per cpu vdso data.
+ */
+#ifdef CONFIG_64BIT
+#define SEGMENT_ORDER	2
+#else
+#define SEGMENT_ORDER	1
+#endif
+
+int vdso_alloc_per_cpu(int cpu, struct _lowcore *lowcore)
+{
+	unsigned long segment_table, page_table, page_frame;
+	u32 *psal, *aste;
+	int i;
+
+	lowcore->vdso_per_cpu_data = __LC_PASTE;
+
+	if (!switch_amode || !vdso_enabled)
+		return 0;
+
+	segment_table = __get_free_pages(GFP_KERNEL, SEGMENT_ORDER);
+	page_table = get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	page_frame = get_zeroed_page(GFP_KERNEL);
+	if (!segment_table || !page_table || !page_frame)
+		goto out;
+
+	clear_table((unsigned long *) segment_table, _SEGMENT_ENTRY_EMPTY,
+		    PAGE_SIZE << SEGMENT_ORDER);
+	clear_table((unsigned long *) page_table, _PAGE_TYPE_EMPTY,
+		    256*sizeof(unsigned long));
+
+	*(unsigned long *) segment_table = _SEGMENT_ENTRY + page_table;
+	*(unsigned long *) page_table = _PAGE_RO + page_frame;
+
+	psal = (u32 *) (page_table + 256*sizeof(unsigned long));
+	aste = psal + 32;
+
+	for (i = 4; i < 32; i += 4)
+		psal[i] = 0x80000000;
+
+	lowcore->paste[4] = (u32)(addr_t) psal;
+	psal[0] = 0x20000000;
+	psal[2] = (u32)(addr_t) aste;
+	*(unsigned long *) (aste + 2) = segment_table +
+		_ASCE_TABLE_LENGTH + _ASCE_USER_BITS + _ASCE_TYPE_SEGMENT;
+	aste[4] = (u32)(addr_t) psal;
+	lowcore->vdso_per_cpu_data = page_frame;
+
+	vdso_init_per_cpu_data(cpu, (struct vdso_per_cpu_data *) page_frame);
+	return 0;
+
+out:
+	free_page(page_frame);
+	free_page(page_table);
+	free_pages(segment_table, SEGMENT_ORDER);
+	return -ENOMEM;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+void vdso_free_per_cpu(int cpu, struct _lowcore *lowcore)
+{
+	unsigned long segment_table, page_table, page_frame;
+	u32 *psal, *aste;
+
+	if (!switch_amode || !vdso_enabled)
+		return;
+
+	psal = (u32 *)(addr_t) lowcore->paste[4];
+	aste = (u32 *)(addr_t) psal[2];
+	segment_table = *(unsigned long *)(aste + 2) & PAGE_MASK;
+	page_table = *(unsigned long *) segment_table;
+	page_frame = *(unsigned long *) page_table;
+
+	free_page(page_frame);
+	free_page(page_table);
+	free_pages(segment_table, SEGMENT_ORDER);
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static void __vdso_init_cr5(void *dummy)
+{
+	unsigned long cr5;
+
+	cr5 = offsetof(struct _lowcore, paste);
+	__ctl_load(cr5, 5, 5);
+}
+
+static void vdso_init_cr5(void)
+{
+	if (switch_amode && vdso_enabled)
+		on_each_cpu(__vdso_init_cr5, NULL, 1);
+}
+#endif /* CONFIG_64BIT */
+
+/*
  * This is called from binfmt_elf, we create the special vma for the
  * vDSO and insert it into the mm struct tree
  */
@@ -172,6 +282,9 @@
 {
 	int i;
 
+	if (!vdso_enabled)
+		return 0;
+	vdso_init_data(vdso_data);
 #if defined(CONFIG_32BIT) || defined(CONFIG_COMPAT)
 	/* Calculate the size of the 32 bit vDSO */
 	vdso32_pages = ((&vdso32_end - &vdso32_start
@@ -208,6 +321,10 @@
 	}
 	vdso64_pagelist[vdso64_pages - 1] = virt_to_page(vdso_data);
 	vdso64_pagelist[vdso64_pages] = NULL;
+#ifndef CONFIG_SMP
+	BUG_ON(vdso_alloc_per_cpu(0, S390_lowcore));
+#endif
+	vdso_init_cr5();
 #endif /* CONFIG_64BIT */
 
 	get_page(virt_to_page(vdso_data));
diff --git a/arch/s390/kernel/vdso64/clock_getres.S b/arch/s390/kernel/vdso64/clock_getres.S
index 488e31a..9ce8caa 100644
--- a/arch/s390/kernel/vdso64/clock_getres.S
+++ b/arch/s390/kernel/vdso64/clock_getres.S
@@ -22,7 +22,12 @@
 	cghi	%r2,CLOCK_REALTIME
 	je	0f
 	cghi	%r2,CLOCK_MONOTONIC
+	je	0f
+	cghi	%r2,-2		/* CLOCK_THREAD_CPUTIME_ID for this thread */
 	jne	2f
+	larl	%r5,_vdso_data
+	icm	%r0,15,__LC_ECTG_OK(%r5)
+	jz	2f
 0:	ltgr	%r3,%r3
 	jz	1f				/* res == NULL */
 	larl	%r1,3f
diff --git a/arch/s390/kernel/vdso64/clock_gettime.S b/arch/s390/kernel/vdso64/clock_gettime.S
index 738a410..79dbfee 100644
--- a/arch/s390/kernel/vdso64/clock_gettime.S
+++ b/arch/s390/kernel/vdso64/clock_gettime.S
@@ -22,8 +22,10 @@
 	larl	%r5,_vdso_data
 	cghi	%r2,CLOCK_REALTIME
 	je	4f
+	cghi	%r2,-2		/* CLOCK_THREAD_CPUTIME_ID for this thread */
+	je	9f
 	cghi	%r2,CLOCK_MONOTONIC
-	jne	9f
+	jne	12f
 
 	/* CLOCK_MONOTONIC */
 	ltgr	%r3,%r3
@@ -42,7 +44,7 @@
 	alg	%r0,__VDSO_WTOM_SEC(%r5)
 	clg	%r4,__VDSO_UPD_COUNT(%r5)	/* check update counter */
 	jne	0b
-	larl	%r5,10f
+	larl	%r5,13f
 1:	clg	%r1,0(%r5)
 	jl	2f
 	slg	%r1,0(%r5)
@@ -68,7 +70,7 @@
 	lg	%r0,__VDSO_XTIME_SEC(%r5)
 	clg	%r4,__VDSO_UPD_COUNT(%r5)	/* check update counter */
 	jne	5b
-	larl	%r5,10f
+	larl	%r5,13f
 6:	clg	%r1,0(%r5)
 	jl	7f
 	slg	%r1,0(%r5)
@@ -79,11 +81,38 @@
 8:	lghi	%r2,0
 	br	%r14
 
+	/* CLOCK_THREAD_CPUTIME_ID for this thread */
+9:	icm	%r0,15,__VDSO_ECTG_OK(%r5)
+	jz	12f
+	ear	%r2,%a4
+	llilh	%r4,0x0100
+	sar	%a4,%r4
+	lghi	%r4,0
+	sacf	512				/* Magic ectg instruction */
+	.insn	ssf,0xc80100000000,__VDSO_ECTG_BASE(4),__VDSO_ECTG_USER(4),4
+	sacf	0
+	sar	%a4,%r2
+	algr	%r1,%r0				/* r1 = cputime as TOD value */
+	mghi	%r1,1000			/* convert to nanoseconds */
+	srlg	%r1,%r1,12			/* r1 = cputime in nanosec */
+	lgr	%r4,%r1
+	larl	%r5,13f
+	srlg	%r1,%r1,9			/* divide by 1000000000 */
+	mlg	%r0,8(%r5)
+	srlg	%r0,%r0,11			/* r0 = tv_sec */
+	stg	%r0,0(%r3)
+	msg	%r0,0(%r5)			/* calculate tv_nsec */
+	slgr	%r4,%r0				/* r4 = tv_nsec */
+	stg	%r4,8(%r3)
+	lghi	%r2,0
+	br	%r14
+
 	/* Fallback to system call */
-9:	lghi	%r1,__NR_clock_gettime
+12:	lghi	%r1,__NR_clock_gettime
 	svc	0
 	br	%r14
 
-10:	.quad	1000000000
+13:	.quad	1000000000
+14:	.quad	19342813113834067
 	.cfi_endproc
 	.size	__kernel_clock_gettime,.-__kernel_clock_gettime
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index 75a6e62..2fb36e4 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -23,94 +23,24 @@
 #include <asm/s390_ext.h>
 #include <asm/timer.h>
 #include <asm/irq_regs.h>
+#include <asm/cpu.h>
 
 static ext_int_info_t ext_int_info_timer;
+
 static DEFINE_PER_CPU(struct vtimer_queue, virt_cpu_timer);
 
-/*
- * Update process times based on virtual cpu times stored by entry.S
- * to the lowcore fields user_timer, system_timer & steal_clock.
- */
-void account_process_tick(struct task_struct *tsk, int user_tick)
+DEFINE_PER_CPU(struct s390_idle_data, s390_idle) = {
+	.lock = __SPIN_LOCK_UNLOCKED(s390_idle.lock)
+};
+
+static inline __u64 get_vtimer(void)
 {
-	cputime_t cputime;
-	__u64 timer, clock;
-	int rcu_user_flag;
-
-	timer = S390_lowcore.last_update_timer;
-	clock = S390_lowcore.last_update_clock;
-	asm volatile ("  STPT %0\n"    /* Store current cpu timer value */
-		      "  STCK %1"      /* Store current tod clock value */
-		      : "=m" (S390_lowcore.last_update_timer),
-		        "=m" (S390_lowcore.last_update_clock) );
-	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
-	S390_lowcore.steal_clock += S390_lowcore.last_update_clock - clock;
-
-	cputime = S390_lowcore.user_timer >> 12;
-	rcu_user_flag = cputime != 0;
-	S390_lowcore.user_timer -= cputime << 12;
-	S390_lowcore.steal_clock -= cputime << 12;
-	account_user_time(tsk, cputime);
-
-	cputime =  S390_lowcore.system_timer >> 12;
-	S390_lowcore.system_timer -= cputime << 12;
-	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, HARDIRQ_OFFSET, cputime);
-
-	cputime = S390_lowcore.steal_clock;
-	if ((__s64) cputime > 0) {
-		cputime >>= 12;
-		S390_lowcore.steal_clock -= cputime << 12;
-		account_steal_time(tsk, cputime);
-	}
-}
-
-/*
- * Update process times based on virtual cpu times stored by entry.S
- * to the lowcore fields user_timer, system_timer & steal_clock.
- */
-void account_vtime(struct task_struct *tsk)
-{
-	cputime_t cputime;
 	__u64 timer;
 
-	timer = S390_lowcore.last_update_timer;
-	asm volatile ("  STPT %0"    /* Store current cpu timer value */
-		      : "=m" (S390_lowcore.last_update_timer) );
-	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
-
-	cputime = S390_lowcore.user_timer >> 12;
-	S390_lowcore.user_timer -= cputime << 12;
-	S390_lowcore.steal_clock -= cputime << 12;
-	account_user_time(tsk, cputime);
-
-	cputime =  S390_lowcore.system_timer >> 12;
-	S390_lowcore.system_timer -= cputime << 12;
-	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, 0, cputime);
+	asm volatile("STPT %0" : "=m" (timer));
+	return timer;
 }
 
-/*
- * Update process times based on virtual cpu times stored by entry.S
- * to the lowcore fields user_timer, system_timer & steal_clock.
- */
-void account_system_vtime(struct task_struct *tsk)
-{
-	cputime_t cputime;
-	__u64 timer;
-
-	timer = S390_lowcore.last_update_timer;
-	asm volatile ("  STPT %0"    /* Store current cpu timer value */
-		      : "=m" (S390_lowcore.last_update_timer) );
-	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
-
-	cputime =  S390_lowcore.system_timer >> 12;
-	S390_lowcore.system_timer -= cputime << 12;
-	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, 0, cputime);
-}
-EXPORT_SYMBOL_GPL(account_system_vtime);
-
 static inline void set_vtimer(__u64 expires)
 {
 	__u64 timer;
@@ -120,56 +50,192 @@
 		      : "=m" (timer) : "m" (expires) );
 	S390_lowcore.system_timer += S390_lowcore.last_update_timer - timer;
 	S390_lowcore.last_update_timer = expires;
-
-	/* store expire time for this CPU timer */
-	__get_cpu_var(virt_cpu_timer).to_expire = expires;
 }
 
-void vtime_start_cpu_timer(void)
+/*
+ * Update process times based on virtual cpu times stored by entry.S
+ * to the lowcore fields user_timer, system_timer & steal_clock.
+ */
+static void do_account_vtime(struct task_struct *tsk, int hardirq_offset)
 {
-	struct vtimer_queue *vt_list;
+	struct thread_info *ti = task_thread_info(tsk);
+	__u64 timer, clock, user, system, steal;
 
-	vt_list = &__get_cpu_var(virt_cpu_timer);
+	timer = S390_lowcore.last_update_timer;
+	clock = S390_lowcore.last_update_clock;
+	asm volatile ("  STPT %0\n"    /* Store current cpu timer value */
+		      "  STCK %1"      /* Store current tod clock value */
+		      : "=m" (S390_lowcore.last_update_timer),
+		        "=m" (S390_lowcore.last_update_clock) );
+	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
+	S390_lowcore.steal_timer += S390_lowcore.last_update_clock - clock;
 
-	/* CPU timer interrupt is pending, don't reprogramm it */
-	if (vt_list->idle & 1LL<<63)
-		return;
+	user = S390_lowcore.user_timer - ti->user_timer;
+	S390_lowcore.steal_timer -= user;
+	ti->user_timer = S390_lowcore.user_timer;
+	account_user_time(tsk, user, user);
 
-	if (!list_empty(&vt_list->list))
-		set_vtimer(vt_list->idle);
+	system = S390_lowcore.system_timer - ti->system_timer;
+	S390_lowcore.steal_timer -= system;
+	ti->system_timer = S390_lowcore.system_timer;
+	account_system_time(tsk, hardirq_offset, system, system);
+
+	steal = S390_lowcore.steal_timer;
+	if ((s64) steal > 0) {
+		S390_lowcore.steal_timer = 0;
+		account_steal_time(steal);
+	}
 }
 
-void vtime_stop_cpu_timer(void)
+void account_vtime(struct task_struct *prev, struct task_struct *next)
 {
-	struct vtimer_queue *vt_list;
+	struct thread_info *ti;
 
-	vt_list = &__get_cpu_var(virt_cpu_timer);
+	do_account_vtime(prev, 0);
+	ti = task_thread_info(prev);
+	ti->user_timer = S390_lowcore.user_timer;
+	ti->system_timer = S390_lowcore.system_timer;
+	ti = task_thread_info(next);
+	S390_lowcore.user_timer = ti->user_timer;
+	S390_lowcore.system_timer = ti->system_timer;
+}
 
-	/* nothing to do */
-	if (list_empty(&vt_list->list)) {
-		vt_list->idle = VTIMER_MAX_SLICE;
-		goto fire;
+void account_process_tick(struct task_struct *tsk, int user_tick)
+{
+	do_account_vtime(tsk, HARDIRQ_OFFSET);
+}
+
+/*
+ * Update process times based on virtual cpu times stored by entry.S
+ * to the lowcore fields user_timer, system_timer & steal_clock.
+ */
+void account_system_vtime(struct task_struct *tsk)
+{
+	struct thread_info *ti = task_thread_info(tsk);
+	__u64 timer, system;
+
+	timer = S390_lowcore.last_update_timer;
+	S390_lowcore.last_update_timer = get_vtimer();
+	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
+
+	system = S390_lowcore.system_timer - ti->system_timer;
+	S390_lowcore.steal_timer -= system;
+	ti->system_timer = S390_lowcore.system_timer;
+	account_system_time(tsk, 0, system, system);
+}
+EXPORT_SYMBOL_GPL(account_system_vtime);
+
+void vtime_start_cpu(void)
+{
+	struct s390_idle_data *idle = &__get_cpu_var(s390_idle);
+	struct vtimer_queue *vq = &__get_cpu_var(virt_cpu_timer);
+	__u64 idle_time, expires;
+
+	/* Account time spent with enabled wait psw loaded as idle time. */
+	idle_time = S390_lowcore.int_clock - idle->idle_enter;
+	account_idle_time(idle_time);
+	S390_lowcore.last_update_clock = S390_lowcore.int_clock;
+
+	/* Account system time spent going idle. */
+	S390_lowcore.system_timer += S390_lowcore.last_update_timer - vq->idle;
+	S390_lowcore.last_update_timer = S390_lowcore.async_enter_timer;
+
+	/* Restart vtime CPU timer */
+	if (vq->do_spt) {
+		/* Program old expire value but first save progress. */
+		expires = vq->idle - S390_lowcore.async_enter_timer;
+		expires += get_vtimer();
+		set_vtimer(expires);
+	} else {
+		/* Don't account the CPU timer delta while the cpu was idle. */
+		vq->elapsed -= vq->idle - S390_lowcore.async_enter_timer;
 	}
 
-	/* store the actual expire value */
-	asm volatile ("STPT %0" : "=m" (vt_list->idle));
+	spin_lock(&idle->lock);
+	idle->idle_time += idle_time;
+	idle->idle_enter = 0ULL;
+	idle->idle_count++;
+	spin_unlock(&idle->lock);
+}
 
-	/*
-	 * If the CPU timer is negative we don't reprogramm
-	 * it because we will get instantly an interrupt.
-	 */
-	if (vt_list->idle & 1LL<<63)
-		return;
+void vtime_stop_cpu(void)
+{
+	struct s390_idle_data *idle = &__get_cpu_var(s390_idle);
+	struct vtimer_queue *vq = &__get_cpu_var(virt_cpu_timer);
+	psw_t psw;
 
-	vt_list->offset += vt_list->to_expire - vt_list->idle;
+	/* Wait for external, I/O or machine check interrupt. */
+	psw.mask = psw_kernel_bits | PSW_MASK_WAIT | PSW_MASK_IO | PSW_MASK_EXT;
 
-	/*
-	 * We cannot halt the CPU timer, we just write a value that
-	 * nearly never expires (only after 71 years) and re-write
-	 * the stored expire value if we continue the timer
-	 */
- fire:
-	set_vtimer(VTIMER_MAX_SLICE);
+	/* Check if the CPU timer needs to be reprogrammed. */
+	if (vq->do_spt) {
+		__u64 vmax = VTIMER_MAX_SLICE;
+		/*
+		 * The inline assembly is equivalent to
+		 *	vq->idle = get_cpu_timer();
+		 *	set_cpu_timer(VTIMER_MAX_SLICE);
+		 *	idle->idle_enter = get_clock();
+		 *	__load_psw_mask(psw_kernel_bits | PSW_MASK_WAIT |
+		 *			   PSW_MASK_IO | PSW_MASK_EXT);
+		 * The difference is that the inline assembly makes sure that
+		 * the last three instruction are stpt, stck and lpsw in that
+		 * order. This is done to increase the precision.
+		 */
+		asm volatile(
+#ifndef CONFIG_64BIT
+			"	basr	1,0\n"
+			"0:	ahi	1,1f-0b\n"
+			"	st	1,4(%2)\n"
+#else /* CONFIG_64BIT */
+			"	larl	1,1f\n"
+			"	stg	1,8(%2)\n"
+#endif /* CONFIG_64BIT */
+			"	stpt	0(%4)\n"
+			"	spt	0(%5)\n"
+			"	stck	0(%3)\n"
+#ifndef CONFIG_64BIT
+			"	lpsw	0(%2)\n"
+#else /* CONFIG_64BIT */
+			"	lpswe	0(%2)\n"
+#endif /* CONFIG_64BIT */
+			"1:"
+			: "=m" (idle->idle_enter), "=m" (vq->idle)
+			: "a" (&psw), "a" (&idle->idle_enter),
+			  "a" (&vq->idle), "a" (&vmax), "m" (vmax), "m" (psw)
+			: "memory", "cc", "1");
+	} else {
+		/*
+		 * The inline assembly is equivalent to
+		 *	vq->idle = get_cpu_timer();
+		 *	idle->idle_enter = get_clock();
+		 *	__load_psw_mask(psw_kernel_bits | PSW_MASK_WAIT |
+		 *			   PSW_MASK_IO | PSW_MASK_EXT);
+		 * The difference is that the inline assembly makes sure that
+		 * the last three instruction are stpt, stck and lpsw in that
+		 * order. This is done to increase the precision.
+		 */
+		asm volatile(
+#ifndef CONFIG_64BIT
+			"	basr	1,0\n"
+			"0:	ahi	1,1f-0b\n"
+			"	st	1,4(%2)\n"
+#else /* CONFIG_64BIT */
+			"	larl	1,1f\n"
+			"	stg	1,8(%2)\n"
+#endif /* CONFIG_64BIT */
+			"	stpt	0(%4)\n"
+			"	stck	0(%3)\n"
+#ifndef CONFIG_64BIT
+			"	lpsw	0(%2)\n"
+#else /* CONFIG_64BIT */
+			"	lpswe	0(%2)\n"
+#endif /* CONFIG_64BIT */
+			"1:"
+			: "=m" (idle->idle_enter), "=m" (vq->idle)
+			: "a" (&psw), "a" (&idle->idle_enter),
+			  "a" (&vq->idle), "m" (psw)
+			: "memory", "cc", "1");
+	}
 }
 
 /*
@@ -195,30 +261,23 @@
  */
 static void do_callbacks(struct list_head *cb_list)
 {
-	struct vtimer_queue *vt_list;
+	struct vtimer_queue *vq;
 	struct vtimer_list *event, *tmp;
-	void (*fn)(unsigned long);
-	unsigned long data;
 
 	if (list_empty(cb_list))
 		return;
 
-	vt_list = &__get_cpu_var(virt_cpu_timer);
+	vq = &__get_cpu_var(virt_cpu_timer);
 
 	list_for_each_entry_safe(event, tmp, cb_list, entry) {
-		fn = event->function;
-		data = event->data;
-		fn(data);
-
-		if (!event->interval)
-			/* delete one shot timer */
-			list_del_init(&event->entry);
-		else {
-			/* move interval timer back to list */
-			spin_lock(&vt_list->lock);
-			list_del_init(&event->entry);
-			list_add_sorted(event, &vt_list->list);
-			spin_unlock(&vt_list->lock);
+		list_del_init(&event->entry);
+		(event->function)(event->data);
+		if (event->interval) {
+			/* Recharge interval timer */
+			event->expires = event->interval + vq->elapsed;
+			spin_lock(&vq->lock);
+			list_add_sorted(event, &vq->list);
+			spin_unlock(&vq->lock);
 		}
 	}
 }
@@ -228,64 +287,57 @@
  */
 static void do_cpu_timer_interrupt(__u16 error_code)
 {
-	__u64 next, delta;
-	struct vtimer_queue *vt_list;
+	struct vtimer_queue *vq;
 	struct vtimer_list *event, *tmp;
-	struct list_head *ptr;
-	/* the callback queue */
-	struct list_head cb_list;
+	struct list_head cb_list;	/* the callback queue */
+	__u64 elapsed, next;
 
 	INIT_LIST_HEAD(&cb_list);
-	vt_list = &__get_cpu_var(virt_cpu_timer);
+	vq = &__get_cpu_var(virt_cpu_timer);
 
 	/* walk timer list, fire all expired events */
-	spin_lock(&vt_list->lock);
+	spin_lock(&vq->lock);
 
-	if (vt_list->to_expire < VTIMER_MAX_SLICE)
-		vt_list->offset += vt_list->to_expire;
-
-	list_for_each_entry_safe(event, tmp, &vt_list->list, entry) {
-		if (event->expires > vt_list->offset)
-			/* found first unexpired event, leave */
-			break;
-
-		/* re-charge interval timer, we have to add the offset */
-		if (event->interval)
-			event->expires = event->interval + vt_list->offset;
-
-		/* move expired timer to the callback queue */
-		list_move_tail(&event->entry, &cb_list);
+	elapsed = vq->elapsed + (vq->timer - S390_lowcore.async_enter_timer);
+	BUG_ON((s64) elapsed < 0);
+	vq->elapsed = 0;
+	list_for_each_entry_safe(event, tmp, &vq->list, entry) {
+		if (event->expires < elapsed)
+			/* move expired timer to the callback queue */
+			list_move_tail(&event->entry, &cb_list);
+		else
+			event->expires -= elapsed;
 	}
-	spin_unlock(&vt_list->lock);
+	spin_unlock(&vq->lock);
+
+	vq->do_spt = list_empty(&cb_list);
 	do_callbacks(&cb_list);
 
 	/* next event is first in list */
-	spin_lock(&vt_list->lock);
-	if (!list_empty(&vt_list->list)) {
-		ptr = vt_list->list.next;
-		event = list_entry(ptr, struct vtimer_list, entry);
-		next = event->expires - vt_list->offset;
-
-		/* add the expired time from this interrupt handler
-		 * and the callback functions
-		 */
-		asm volatile ("STPT %0" : "=m" (delta));
-		delta = 0xffffffffffffffffLL - delta + 1;
-		vt_list->offset += delta;
-		next -= delta;
-	} else {
-		vt_list->offset = 0;
-		next = VTIMER_MAX_SLICE;
-	}
-	spin_unlock(&vt_list->lock);
-	set_vtimer(next);
+	next = VTIMER_MAX_SLICE;
+	spin_lock(&vq->lock);
+	if (!list_empty(&vq->list)) {
+		event = list_first_entry(&vq->list, struct vtimer_list, entry);
+		next = event->expires;
+	} else
+		vq->do_spt = 0;
+	spin_unlock(&vq->lock);
+	/*
+	 * To improve precision add the time spent by the
+	 * interrupt handler to the elapsed time.
+	 * Note: CPU timer counts down and we got an interrupt,
+	 *	 the current content is negative
+	 */
+	elapsed = S390_lowcore.async_enter_timer - get_vtimer();
+	set_vtimer(next - elapsed);
+	vq->timer = next - elapsed;
+	vq->elapsed = elapsed;
 }
 
 void init_virt_timer(struct vtimer_list *timer)
 {
 	timer->function = NULL;
 	INIT_LIST_HEAD(&timer->entry);
-	spin_lock_init(&timer->lock);
 }
 EXPORT_SYMBOL(init_virt_timer);
 
@@ -299,44 +351,40 @@
  */
 static void internal_add_vtimer(struct vtimer_list *timer)
 {
+	struct vtimer_queue *vq;
 	unsigned long flags;
-	__u64 done;
-	struct vtimer_list *event;
-	struct vtimer_queue *vt_list;
+	__u64 left, expires;
 
-	vt_list = &per_cpu(virt_cpu_timer, timer->cpu);
-	spin_lock_irqsave(&vt_list->lock, flags);
+	vq = &per_cpu(virt_cpu_timer, timer->cpu);
+	spin_lock_irqsave(&vq->lock, flags);
 
 	BUG_ON(timer->cpu != smp_processor_id());
 
-	/* if list is empty we only have to set the timer */
-	if (list_empty(&vt_list->list)) {
-		/* reset the offset, this may happen if the last timer was
-		 * just deleted by mod_virt_timer and the interrupt
-		 * didn't happen until here
-		 */
-		vt_list->offset = 0;
-		goto fire;
+	if (list_empty(&vq->list)) {
+		/* First timer on this cpu, just program it. */
+		list_add(&timer->entry, &vq->list);
+		set_vtimer(timer->expires);
+		vq->timer = timer->expires;
+		vq->elapsed = 0;
+	} else {
+		/* Check progress of old timers. */
+		expires = timer->expires;
+		left = get_vtimer();
+		if (likely((s64) expires < (s64) left)) {
+			/* The new timer expires before the current timer. */
+			set_vtimer(expires);
+			vq->elapsed += vq->timer - left;
+			vq->timer = expires;
+		} else {
+			vq->elapsed += vq->timer - left;
+			vq->timer = left;
+		}
+		/* Insert new timer into per cpu list. */
+		timer->expires += vq->elapsed;
+		list_add_sorted(timer, &vq->list);
 	}
 
-	/* save progress */
-	asm volatile ("STPT %0" : "=m" (done));
-
-	/* calculate completed work */
-	done = vt_list->to_expire - done + vt_list->offset;
-	vt_list->offset = 0;
-
-	list_for_each_entry(event, &vt_list->list, entry)
-		event->expires -= done;
-
- fire:
-	list_add_sorted(timer, &vt_list->list);
-
-	/* get first element, which is the next vtimer slice */
-	event = list_entry(vt_list->list.next, struct vtimer_list, entry);
-
-	set_vtimer(event->expires);
-	spin_unlock_irqrestore(&vt_list->lock, flags);
+	spin_unlock_irqrestore(&vq->lock, flags);
 	/* release CPU acquired in prepare_vtimer or mod_virt_timer() */
 	put_cpu();
 }
@@ -381,14 +429,15 @@
  * If we change a pending timer the function must be called on the CPU
  * where the timer is running on, e.g. by smp_call_function_single()
  *
- * The original mod_timer adds the timer if it is not pending. For compatibility
- * we do the same. The timer will be added on the current CPU as a oneshot timer.
+ * The original mod_timer adds the timer if it is not pending. For
+ * compatibility we do the same. The timer will be added on the current
+ * CPU as a oneshot timer.
  *
  * returns whether it has modified a pending timer (1) or not (0)
  */
 int mod_virt_timer(struct vtimer_list *timer, __u64 expires)
 {
-	struct vtimer_queue *vt_list;
+	struct vtimer_queue *vq;
 	unsigned long flags;
 	int cpu;
 
@@ -404,17 +453,17 @@
 		return 1;
 
 	cpu = get_cpu();
-	vt_list = &per_cpu(virt_cpu_timer, cpu);
+	vq = &per_cpu(virt_cpu_timer, cpu);
 
 	/* check if we run on the right CPU */
 	BUG_ON(timer->cpu != cpu);
 
 	/* disable interrupts before test if timer is pending */
-	spin_lock_irqsave(&vt_list->lock, flags);
+	spin_lock_irqsave(&vq->lock, flags);
 
 	/* if timer isn't pending add it on the current CPU */
 	if (!vtimer_pending(timer)) {
-		spin_unlock_irqrestore(&vt_list->lock, flags);
+		spin_unlock_irqrestore(&vq->lock, flags);
 		/* we do not activate an interval timer with mod_virt_timer */
 		timer->interval = 0;
 		timer->expires = expires;
@@ -431,7 +480,7 @@
 		timer->interval = expires;
 
 	/* the timer can't expire anymore so we can release the lock */
-	spin_unlock_irqrestore(&vt_list->lock, flags);
+	spin_unlock_irqrestore(&vq->lock, flags);
 	internal_add_vtimer(timer);
 	return 1;
 }
@@ -445,25 +494,19 @@
 int del_virt_timer(struct vtimer_list *timer)
 {
 	unsigned long flags;
-	struct vtimer_queue *vt_list;
+	struct vtimer_queue *vq;
 
 	/* check if timer is pending */
 	if (!vtimer_pending(timer))
 		return 0;
 
-	vt_list = &per_cpu(virt_cpu_timer, timer->cpu);
-	spin_lock_irqsave(&vt_list->lock, flags);
+	vq = &per_cpu(virt_cpu_timer, timer->cpu);
+	spin_lock_irqsave(&vq->lock, flags);
 
 	/* we don't interrupt a running timer, just let it expire! */
 	list_del_init(&timer->entry);
 
-	/* last timer removed */
-	if (list_empty(&vt_list->list)) {
-		vt_list->to_expire = 0;
-		vt_list->offset = 0;
-	}
-
-	spin_unlock_irqrestore(&vt_list->lock, flags);
+	spin_unlock_irqrestore(&vq->lock, flags);
 	return 1;
 }
 EXPORT_SYMBOL(del_virt_timer);
@@ -473,24 +516,19 @@
  */
 void init_cpu_vtimer(void)
 {
-	struct vtimer_queue *vt_list;
+	struct vtimer_queue *vq;
 
 	/* kick the virtual timer */
-	S390_lowcore.exit_timer = VTIMER_MAX_SLICE;
-	S390_lowcore.last_update_timer = VTIMER_MAX_SLICE;
-	asm volatile ("SPT %0" : : "m" (S390_lowcore.last_update_timer));
 	asm volatile ("STCK %0" : "=m" (S390_lowcore.last_update_clock));
+	asm volatile ("STPT %0" : "=m" (S390_lowcore.last_update_timer));
+
+	/* initialize per cpu vtimer structure */
+	vq = &__get_cpu_var(virt_cpu_timer);
+	INIT_LIST_HEAD(&vq->list);
+	spin_lock_init(&vq->lock);
 
 	/* enable cpu timer interrupts */
 	__ctl_set_bit(0,10);
-
-	vt_list = &__get_cpu_var(virt_cpu_timer);
-	INIT_LIST_HEAD(&vt_list->list);
-	spin_lock_init(&vt_list->lock);
-	vt_list->to_expire = 0;
-	vt_list->offset = 0;
-	vt_list->idle = 0;
-
 }
 
 void __init vtime_init(void)
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 8b00eb2..be84971 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -113,8 +113,6 @@
 int kvm_dev_ioctl_check_extension(long ext)
 {
 	switch (ext) {
-	case KVM_CAP_USER_MEMORY:
-		return 1;
 	default:
 		return 0;
 	}
@@ -185,8 +183,6 @@
 	debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
 	VM_EVENT(kvm, 3, "%s", "vm created");
 
-	try_module_get(THIS_MODULE);
-
 	return kvm;
 out_nodbf:
 	free_page((unsigned long)(kvm->arch.sca));
@@ -196,13 +192,33 @@
 	return ERR_PTR(rc);
 }
 
+void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
+{
+	VCPU_EVENT(vcpu, 3, "%s", "free cpu");
+	free_page((unsigned long)(vcpu->arch.sie_block));
+	kvm_vcpu_uninit(vcpu);
+	kfree(vcpu);
+}
+
+static void kvm_free_vcpus(struct kvm *kvm)
+{
+	unsigned int i;
+
+	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+		if (kvm->vcpus[i]) {
+			kvm_arch_vcpu_destroy(kvm->vcpus[i]);
+			kvm->vcpus[i] = NULL;
+		}
+	}
+}
+
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
-	debug_unregister(kvm->arch.dbf);
+	kvm_free_vcpus(kvm);
 	kvm_free_physmem(kvm);
 	free_page((unsigned long)(kvm->arch.sca));
+	debug_unregister(kvm->arch.dbf);
 	kfree(kvm);
-	module_put(THIS_MODULE);
 }
 
 /* Section: vcpu related */
@@ -213,8 +229,7 @@
 
 void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 {
-	/* kvm common code refers to this, but does'nt call it */
-	BUG();
+	/* Nothing todo */
 }
 
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
@@ -308,8 +323,6 @@
 	VM_EVENT(kvm, 3, "create cpu %d at %p, sie block at %p", id, vcpu,
 		 vcpu->arch.sie_block);
 
-	try_module_get(THIS_MODULE);
-
 	return vcpu;
 out_free_cpu:
 	kfree(vcpu);
@@ -317,14 +330,6 @@
 	return ERR_PTR(rc);
 }
 
-void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
-{
-	VCPU_EVENT(vcpu, 3, "%s", "destroy cpu");
-	free_page((unsigned long)(vcpu->arch.sie_block));
-	kfree(vcpu);
-	module_put(THIS_MODULE);
-}
-
 int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 {
 	/* kvm common code refers to this, but never calls it */
diff --git a/arch/sh/include/asm/smp.h b/arch/sh/include/asm/smp.h
index 85b660c..c24e9c6 100644
--- a/arch/sh/include/asm/smp.h
+++ b/arch/sh/include/asm/smp.h
@@ -31,7 +31,7 @@
 };
 
 void smp_message_recv(unsigned int msg);
-void smp_timer_broadcast(cpumask_t mask);
+void smp_timer_broadcast(const struct cpumask *mask);
 
 void local_timer_interrupt(void);
 void local_timer_setup(unsigned int cpu);
diff --git a/arch/sh/include/asm/topology.h b/arch/sh/include/asm/topology.h
index 95f0085..279d9cc 100644
--- a/arch/sh/include/asm/topology.h
+++ b/arch/sh/include/asm/topology.h
@@ -5,7 +5,6 @@
 
 /* sched_domains SD_NODE_INIT for sh machines */
 #define SD_NODE_INIT (struct sched_domain) {		\
-	.span			= CPU_MASK_NONE,	\
 	.parent			= NULL,			\
 	.child			= NULL,			\
 	.groups			= NULL,			\
diff --git a/arch/sh/kernel/init_task.c b/arch/sh/kernel/init_task.c
index b151a25..80c35ff 100644
--- a/arch/sh/kernel/init_task.c
+++ b/arch/sh/kernel/init_task.c
@@ -7,7 +7,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct pt_regs fake_swapper_regs;
diff --git a/arch/sh/kernel/smp.c b/arch/sh/kernel/smp.c
index 3c5ad16..8f40274 100644
--- a/arch/sh/kernel/smp.c
+++ b/arch/sh/kernel/smp.c
@@ -31,12 +31,6 @@
 int __cpu_number_map[NR_CPUS];		/* Map physical to logical */
 int __cpu_logical_map[NR_CPUS];		/* Map logical to physical */
 
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
-
-cpumask_t cpu_online_map;
-EXPORT_SYMBOL(cpu_online_map);
-
 static inline void __init smp_store_cpu_info(unsigned int cpu)
 {
 	struct sh_cpuinfo *c = cpu_data + cpu;
@@ -190,11 +184,11 @@
 	plat_send_ipi(cpu, SMP_MSG_FUNCTION_SINGLE);
 }
 
-void smp_timer_broadcast(cpumask_t mask)
+void smp_timer_broadcast(const struct cpumask *mask)
 {
 	int cpu;
 
-	for_each_cpu_mask(cpu, mask)
+	for_each_cpu(cpu, mask)
 		plat_send_ipi(cpu, SMP_MSG_TIMER);
 }
 
diff --git a/arch/sh/kernel/timers/timer-broadcast.c b/arch/sh/kernel/timers/timer-broadcast.c
index c231763..96e8eae 100644
--- a/arch/sh/kernel/timers/timer-broadcast.c
+++ b/arch/sh/kernel/timers/timer-broadcast.c
@@ -51,7 +51,7 @@
 	clk->mult		= 1;
 	clk->set_mode		= dummy_timer_set_mode;
 	clk->broadcast		= smp_timer_broadcast;
-	clk->cpumask		= cpumask_of_cpu(cpu);
+	clk->cpumask		= cpumask_of(cpu);
 
 	clockevents_register_device(clk);
 }
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index 3c61ddd..0db3f95 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -263,7 +263,7 @@
 	tmu0_clockevent.min_delta_ns =
 			clockevent_delta2ns(1, &tmu0_clockevent);
 
-	tmu0_clockevent.cpumask = cpumask_of_cpu(0);
+	tmu0_clockevent.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&tmu0_clockevent);
 
diff --git a/arch/sparc/include/asm/smp_32.h b/arch/sparc/include/asm/smp_32.h
index a8180e5..8408d9d 100644
--- a/arch/sparc/include/asm/smp_32.h
+++ b/arch/sparc/include/asm/smp_32.h
@@ -29,8 +29,6 @@
  */
 
 extern unsigned char boot_cpu_id;
-extern cpumask_t phys_cpu_present_map;
-#define cpu_possible_map phys_cpu_present_map
 
 typedef void (*smpfunc_t)(unsigned long, unsigned long, unsigned long,
 		       unsigned long, unsigned long);
diff --git a/arch/sparc/kernel/init_task.c b/arch/sparc/kernel/init_task.c
index 62126e4..f28cb82 100644
--- a/arch/sparc/kernel/init_task.c
+++ b/arch/sparc/kernel/init_task.c
@@ -8,7 +8,6 @@
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c
index a3ea2bc..cab8e028 100644
--- a/arch/sparc/kernel/irq_64.c
+++ b/arch/sparc/kernel/irq_64.c
@@ -312,7 +312,8 @@
 	}
 }
 
-static void sun4u_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4u_set_affinity(unsigned int virt_irq,
+			       const struct cpumask *mask)
 {
 	sun4u_irq_enable(virt_irq);
 }
@@ -362,7 +363,8 @@
 		       ino, err);
 }
 
-static void sun4v_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4v_set_affinity(unsigned int virt_irq,
+			       const struct cpumask *mask)
 {
 	unsigned int ino = virt_irq_table[virt_irq].dev_ino;
 	unsigned long cpuid = irq_choose_cpu(virt_irq);
@@ -429,7 +431,8 @@
 		       dev_handle, dev_ino, err);
 }
 
-static void sun4v_virt_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4v_virt_set_affinity(unsigned int virt_irq,
+				    const struct cpumask *mask)
 {
 	unsigned long cpuid, dev_handle, dev_ino;
 	int err;
@@ -851,7 +854,7 @@
 		    !(irq_desc[irq].status & IRQ_PER_CPU)) {
 			if (irq_desc[irq].chip->set_affinity)
 				irq_desc[irq].chip->set_affinity(irq,
-					irq_desc[irq].affinity);
+					&irq_desc[irq].affinity);
 		}
 		spin_unlock_irqrestore(&irq_desc[irq].lock, flags);
 	}
diff --git a/arch/sparc/kernel/of_device_64.c b/arch/sparc/kernel/of_device_64.c
index 46e231f..322046c 100644
--- a/arch/sparc/kernel/of_device_64.c
+++ b/arch/sparc/kernel/of_device_64.c
@@ -780,7 +780,7 @@
 	if (nid != -1) {
 		cpumask_t numa_mask = node_to_cpumask(nid);
 
-		irq_set_affinity(irq, numa_mask);
+		irq_set_affinity(irq, &numa_mask);
 	}
 
 	return irq;
diff --git a/arch/sparc/kernel/pci_msi.c b/arch/sparc/kernel/pci_msi.c
index 2e680f3..0d0cd81 100644
--- a/arch/sparc/kernel/pci_msi.c
+++ b/arch/sparc/kernel/pci_msi.c
@@ -288,7 +288,7 @@
 	if (nid != -1) {
 		cpumask_t numa_mask = node_to_cpumask(nid);
 
-		irq_set_affinity(irq, numa_mask);
+		irq_set_affinity(irq, &numa_mask);
 	}
 	err = request_irq(irq, sparc64_msiq_interrupt, 0,
 			  "MSIQ",
diff --git a/arch/sparc/kernel/smp_32.c b/arch/sparc/kernel/smp_32.c
index e396c1f..1e5ac4e2 100644
--- a/arch/sparc/kernel/smp_32.c
+++ b/arch/sparc/kernel/smp_32.c
@@ -39,8 +39,6 @@
 unsigned char boot_cpu_id = 0;
 unsigned char boot_cpu_id4 = 0; /* boot_cpu_id << 2 */
 
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-cpumask_t phys_cpu_present_map = CPU_MASK_NONE;
 cpumask_t smp_commenced_mask = CPU_MASK_NONE;
 
 /* The only guaranteed locking primitive available on all Sparc
@@ -334,7 +332,7 @@
 	instance = 0;
 	while (!cpu_find_by_instance(instance, NULL, &mid)) {
 		if (mid < NR_CPUS) {
-			cpu_set(mid, phys_cpu_present_map);
+			cpu_set(mid, cpu_possible_map);
 			cpu_set(mid, cpu_present_map);
 		}
 		instance++;
@@ -354,7 +352,7 @@
 
 	current_thread_info()->cpu = cpuid;
 	cpu_set(cpuid, cpu_online_map);
-	cpu_set(cpuid, phys_cpu_present_map);
+	cpu_set(cpuid, cpu_possible_map);
 }
 
 int __cpuinit __cpu_up(unsigned int cpu)
diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c
index bfe99d8..4632979 100644
--- a/arch/sparc/kernel/smp_64.c
+++ b/arch/sparc/kernel/smp_64.c
@@ -49,14 +49,10 @@
 
 int sparc64_multi_core __read_mostly;
 
-cpumask_t cpu_possible_map __read_mostly = CPU_MASK_NONE;
-cpumask_t cpu_online_map __read_mostly = CPU_MASK_NONE;
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map) = CPU_MASK_NONE;
 cpumask_t cpu_core_map[NR_CPUS] __read_mostly =
 	{ [0 ... NR_CPUS-1] = CPU_MASK_NONE };
 
-EXPORT_SYMBOL(cpu_possible_map);
-EXPORT_SYMBOL(cpu_online_map);
 EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
 EXPORT_SYMBOL(cpu_core_map);
 
diff --git a/arch/sparc/kernel/sparc_ksyms_32.c b/arch/sparc/kernel/sparc_ksyms_32.c
index a4d45fc..e1e9763 100644
--- a/arch/sparc/kernel/sparc_ksyms_32.c
+++ b/arch/sparc/kernel/sparc_ksyms_32.c
@@ -112,10 +112,6 @@
 #ifdef CONFIG_SMP
 /* IRQ implementation. */
 EXPORT_SYMBOL(synchronize_irq);
-
-/* CPU online map and active count. */
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(phys_cpu_present_map);
 #endif
 
 EXPORT_SYMBOL(__udelay);
diff --git a/arch/sparc/kernel/time_64.c b/arch/sparc/kernel/time_64.c
index 141da37..9df8f09 100644
--- a/arch/sparc/kernel/time_64.c
+++ b/arch/sparc/kernel/time_64.c
@@ -763,7 +763,7 @@
 	sevt = &__get_cpu_var(sparc64_events);
 
 	memcpy(sevt, &sparc64_clockevent, sizeof(*sevt));
-	sevt->cpumask = cpumask_of_cpu(smp_processor_id());
+	sevt->cpumask = cpumask_of(smp_processor_id());
 
 	clockevents_register_device(sevt);
 }
diff --git a/arch/um/kernel/init_task.c b/arch/um/kernel/init_task.c
index 910eda8..806d381 100644
--- a/arch/um/kernel/init_task.c
+++ b/arch/um/kernel/init_task.c
@@ -10,7 +10,6 @@
 #include "linux/mqueue.h"
 #include "asm/uaccess.h"
 
-static struct fs_struct init_fs = INIT_FS;
 struct mm_struct init_mm = INIT_MM(init_mm);
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
diff --git a/arch/um/kernel/smp.c b/arch/um/kernel/smp.c
index 0457721..98351c7 100644
--- a/arch/um/kernel/smp.c
+++ b/arch/um/kernel/smp.c
@@ -25,13 +25,6 @@
 #include "irq_user.h"
 #include "os.h"
 
-/* CPU online map, set by smp_boot_cpus */
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-
-EXPORT_SYMBOL(cpu_online_map);
-EXPORT_SYMBOL(cpu_possible_map);
-
 /* Per CPU bogomips and other parameters
  * The only piece used here is the ipi pipe, which is set before SMP is
  * started and never changed.
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index 47f04f4..b13a87a 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -50,7 +50,7 @@
 static struct clock_event_device itimer_clockevent = {
 	.name		= "itimer",
 	.rating		= 250,
-	.cpumask	= CPU_MASK_ALL,
+	.cpumask	= cpu_all_mask,
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.set_mode	= itimer_set_mode,
 	.set_next_event = itimer_next_event,
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 0f44add..249d1e0 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -601,19 +601,20 @@
 
 config MAXSMP
 	bool "Configure Maximum number of SMP Processors and NUMA Nodes"
-	depends on X86_64 && SMP && BROKEN
+	depends on X86_64 && SMP && DEBUG_KERNEL && EXPERIMENTAL
+	select CPUMASK_OFFSTACK
 	default n
 	help
 	  Configure maximum number of CPUS and NUMA Nodes for this architecture.
 	  If unsure, say N.
 
 config NR_CPUS
-	int "Maximum number of CPUs (2-512)" if !MAXSMP
-	range 2 512
-	depends on SMP
+	int "Maximum number of CPUs" if SMP && !MAXSMP
+	range 2 512 if SMP && !MAXSMP
+	default "1" if !SMP
 	default "4096" if MAXSMP
-	default "32" if X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000
-	default "8"
+	default "32" if SMP && (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000)
+	default "8" if SMP
 	help
 	  This allows you to specify the maximum number of CPUs which this
 	  kernel will support.  The maximum supported value is 512 and the
diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index b195f85..9dabd00 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -24,15 +24,14 @@
 #include <asm/ucontext.h>
 #include <asm/uaccess.h>
 #include <asm/i387.h>
-#include <asm/ia32.h>
 #include <asm/ptrace.h>
 #include <asm/ia32_unistd.h>
 #include <asm/user32.h>
 #include <asm/sigcontext32.h>
 #include <asm/proto.h>
 #include <asm/vdso.h>
-
 #include <asm/sigframe.h>
+#include <asm/sys_ia32.h>
 
 #define DEBUG_SIG 0
 
diff --git a/arch/x86/ia32/ipc32.c b/arch/x86/ia32/ipc32.c
index d21991c..29cdcd0 100644
--- a/arch/x86/ia32/ipc32.c
+++ b/arch/x86/ia32/ipc32.c
@@ -8,6 +8,7 @@
 #include <linux/shm.h>
 #include <linux/ipc.h>
 #include <linux/compat.h>
+#include <asm/sys_ia32.h>
 
 asmlinkage long sys32_ipc(u32 call, int first, int second, int third,
 			  compat_uptr_t ptr, u32 fifth)
diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
index 2e09dcd..6c0d7f6 100644
--- a/arch/x86/ia32/sys_ia32.c
+++ b/arch/x86/ia32/sys_ia32.c
@@ -44,8 +44,8 @@
 #include <asm/types.h>
 #include <asm/uaccess.h>
 #include <asm/atomic.h>
-#include <asm/ia32.h>
 #include <asm/vgtod.h>
+#include <asm/sys_ia32.h>
 
 #define AA(__x)		((unsigned long)(__x))
 
diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index 25caa07..ab1d51a 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -54,7 +54,6 @@
 extern int is_vsmp_box(void);
 extern void xapic_wait_icr_idle(void);
 extern u32 safe_xapic_wait_icr_idle(void);
-extern u64 xapic_icr_read(void);
 extern void xapic_icr_write(u32, u32);
 extern int setup_profiling_timer(unsigned int);
 
@@ -93,7 +92,7 @@
 }
 
 #ifndef CONFIG_X86_32
-extern int x2apic, x2apic_preenabled;
+extern int x2apic;
 extern void check_x2apic(void);
 extern void enable_x2apic(void);
 extern void enable_IR_x2apic(void);
diff --git a/arch/x86/include/asm/bigsmp/apic.h b/arch/x86/include/asm/bigsmp/apic.h
index ce547f2..d8dd9f5 100644
--- a/arch/x86/include/asm/bigsmp/apic.h
+++ b/arch/x86/include/asm/bigsmp/apic.h
@@ -9,12 +9,12 @@
 	return (1);
 }
 
-static inline cpumask_t target_cpus(void)
+static inline const cpumask_t *target_cpus(void)
 {
 #ifdef CONFIG_SMP
-        return cpu_online_map;
+	return &cpu_online_map;
 #else
-        return cpumask_of_cpu(0);
+	return &cpumask_of_cpu(0);
 #endif
 }
 
@@ -79,7 +79,7 @@
 
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
-	if (mps_cpu < NR_CPUS)
+	if (mps_cpu < nr_cpu_ids)
 		return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu);
 
 	return BAD_APICID;
@@ -94,7 +94,7 @@
 /* Mapping from cpu number to logical apicid */
 static inline int cpu_to_logical_apicid(int cpu)
 {
-	if (cpu >= NR_CPUS)
+	if (cpu >= nr_cpu_ids)
 		return BAD_APICID;
 	return cpu_physical_id(cpu);
 }
@@ -119,16 +119,34 @@
 }
 
 /* As we are using single CPU as destination, pick only one CPU here */
-static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
+static inline unsigned int cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
 	int cpu;
 	int apicid;	
 
-	cpu = first_cpu(cpumask);
+	cpu = first_cpu(*cpumask);
 	apicid = cpu_to_logical_apicid(cpu);
 	return apicid;
 }
 
+static inline unsigned int cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return cpu_to_logical_apicid(cpu);
+
+	return BAD_APICID;
+}
+
 static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
 {
 	return cpuid_apic >> index_msb;
diff --git a/arch/x86/include/asm/bigsmp/ipi.h b/arch/x86/include/asm/bigsmp/ipi.h
index 9404c53..27fcd01b 100644
--- a/arch/x86/include/asm/bigsmp/ipi.h
+++ b/arch/x86/include/asm/bigsmp/ipi.h
@@ -1,25 +1,22 @@
 #ifndef __ASM_MACH_IPI_H
 #define __ASM_MACH_IPI_H
 
-void send_IPI_mask_sequence(cpumask_t mask, int vector);
+void send_IPI_mask_sequence(const struct cpumask *mask, int vector);
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector);
 
-static inline void send_IPI_mask(cpumask_t mask, int vector)
+static inline void send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	send_IPI_mask_sequence(mask, vector);
 }
 
 static inline void send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(smp_processor_id(), mask);
-
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, vector);
+	send_IPI_mask_allbutself(cpu_online_mask, vector);
 }
 
 static inline void send_IPI_all(int vector)
 {
-	send_IPI_mask(cpu_online_map, vector);
+	send_IPI_mask(cpu_online_mask, vector);
 }
 
 #endif /* __ASM_MACH_IPI_H */
diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h
index e6b82b17..dc27705 100644
--- a/arch/x86/include/asm/desc.h
+++ b/arch/x86/include/asm/desc.h
@@ -320,16 +320,14 @@
 	_set_gate(n, GATE_INTERRUPT, addr, 0, 0, __KERNEL_CS);
 }
 
-#define SYS_VECTOR_FREE		0
-#define SYS_VECTOR_ALLOCED	1
-
 extern int first_system_vector;
-extern char system_vectors[];
+/* used_vectors is BITMAP for irq is not managed by percpu vector_irq */
+extern unsigned long used_vectors[];
 
 static inline void alloc_system_vector(int vector)
 {
-	if (system_vectors[vector] == SYS_VECTOR_FREE) {
-		system_vectors[vector] = SYS_VECTOR_ALLOCED;
+	if (!test_bit(vector, used_vectors)) {
+		set_bit(vector, used_vectors);
 		if (first_system_vector > vector)
 			first_system_vector = vector;
 	} else
diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h
index a2e545c..ca5ffb2 100644
--- a/arch/x86/include/asm/efi.h
+++ b/arch/x86/include/asm/efi.h
@@ -90,6 +90,7 @@
 
 #endif /* CONFIG_X86_32 */
 
+extern int add_efi_memmap;
 extern void efi_reserve_early(void);
 extern void efi_call_phys_prelog(void);
 extern void efi_call_phys_epilog(void);
diff --git a/arch/x86/include/asm/es7000/apic.h b/arch/x86/include/asm/es7000/apic.h
index e24ef87..51ac123 100644
--- a/arch/x86/include/asm/es7000/apic.h
+++ b/arch/x86/include/asm/es7000/apic.h
@@ -9,14 +9,14 @@
 	        return (1);
 }
 
-static inline cpumask_t target_cpus_cluster(void)
+static inline const cpumask_t *target_cpus_cluster(void)
 {
-	return CPU_MASK_ALL;
+	return &CPU_MASK_ALL;
 }
 
-static inline cpumask_t target_cpus(void)
+static inline const cpumask_t *target_cpus(void)
 {
-	return cpumask_of_cpu(smp_processor_id());
+	return &cpumask_of_cpu(smp_processor_id());
 }
 
 #define APIC_DFR_VALUE_CLUSTER		(APIC_DFR_CLUSTER)
@@ -80,9 +80,10 @@
 static inline void setup_apic_routing(void)
 {
 	int apic = per_cpu(x86_bios_cpu_apicid, smp_processor_id());
-	printk("Enabling APIC mode:  %s.  Using %d I/O APICs, target cpus %lx\n",
+	printk("Enabling APIC mode:  %s. Using %d I/O APICs, target cpus %lx\n",
 		(apic_version[apic] == 0x14) ?
-		"Physical Cluster" : "Logical Cluster", nr_ioapics, cpus_addr(target_cpus())[0]);
+			"Physical Cluster" : "Logical Cluster",
+			nr_ioapics, cpus_addr(*target_cpus())[0]);
 }
 
 static inline int multi_timer_check(int apic, int irq)
@@ -100,7 +101,7 @@
 {
 	if (!mps_cpu)
 		return boot_cpu_physical_apicid;
-	else if (mps_cpu < NR_CPUS)
+	else if (mps_cpu < nr_cpu_ids)
 		return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu);
 	else
 		return BAD_APICID;
@@ -120,9 +121,9 @@
 static inline int cpu_to_logical_apicid(int cpu)
 {
 #ifdef CONFIG_SMP
-       if (cpu >= NR_CPUS)
-	       return BAD_APICID;
-       return (int)cpu_2_logical_apicid[cpu];
+	if (cpu >= nr_cpu_ids)
+		return BAD_APICID;
+	return (int)cpu_2_logical_apicid[cpu];
 #else
 	return logical_smp_processor_id();
 #endif
@@ -146,14 +147,15 @@
 	return (1);
 }
 
-static inline unsigned int cpu_mask_to_apicid_cluster(cpumask_t cpumask)
+static inline unsigned int
+cpu_mask_to_apicid_cluster(const struct cpumask *cpumask)
 {
 	int num_bits_set;
 	int cpus_found = 0;
 	int cpu;
 	int apicid;
 
-	num_bits_set = cpus_weight(cpumask);
+	num_bits_set = cpumask_weight(cpumask);
 	/* Return id to all */
 	if (num_bits_set == NR_CPUS)
 		return 0xFF;
@@ -161,10 +163,10 @@
 	 * The cpus in the mask must all be on the apic cluster.  If are not
 	 * on the same apicid cluster return default value of TARGET_CPUS.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = cpumask_first(cpumask);
 	apicid = cpu_to_logical_apicid(cpu);
 	while (cpus_found < num_bits_set) {
-		if (cpu_isset(cpu, cpumask)) {
+		if (cpumask_test_cpu(cpu, cpumask)) {
 			int new_apicid = cpu_to_logical_apicid(cpu);
 			if (apicid_cluster(apicid) !=
 					apicid_cluster(new_apicid)){
@@ -179,14 +181,14 @@
 	return apicid;
 }
 
-static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
+static inline unsigned int cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
 	int num_bits_set;
 	int cpus_found = 0;
 	int cpu;
 	int apicid;
 
-	num_bits_set = cpus_weight(cpumask);
+	num_bits_set = cpus_weight(*cpumask);
 	/* Return id to all */
 	if (num_bits_set == NR_CPUS)
 		return cpu_to_logical_apicid(0);
@@ -194,10 +196,10 @@
 	 * The cpus in the mask must all be on the apic cluster.  If are not
 	 * on the same apicid cluster return default value of TARGET_CPUS.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = first_cpu(*cpumask);
 	apicid = cpu_to_logical_apicid(cpu);
 	while (cpus_found < num_bits_set) {
-		if (cpu_isset(cpu, cpumask)) {
+		if (cpu_isset(cpu, *cpumask)) {
 			int new_apicid = cpu_to_logical_apicid(cpu);
 			if (apicid_cluster(apicid) !=
 					apicid_cluster(new_apicid)){
@@ -212,6 +214,50 @@
 	return apicid;
 }
 
+
+static inline unsigned int cpu_mask_to_apicid_and(const struct cpumask *inmask,
+						  const struct cpumask *andmask)
+{
+	int num_bits_set;
+	int cpus_found = 0;
+	int cpu;
+	int apicid = cpu_to_logical_apicid(0);
+	cpumask_var_t cpumask;
+
+	if (!alloc_cpumask_var(&cpumask, GFP_ATOMIC))
+		return apicid;
+
+	cpumask_and(cpumask, inmask, andmask);
+	cpumask_and(cpumask, cpumask, cpu_online_mask);
+
+	num_bits_set = cpumask_weight(cpumask);
+	/* Return id to all */
+	if (num_bits_set == NR_CPUS)
+		goto exit;
+	/*
+	 * The cpus in the mask must all be on the apic cluster.  If are not
+	 * on the same apicid cluster return default value of TARGET_CPUS.
+	 */
+	cpu = cpumask_first(cpumask);
+	apicid = cpu_to_logical_apicid(cpu);
+	while (cpus_found < num_bits_set) {
+		if (cpumask_test_cpu(cpu, cpumask)) {
+			int new_apicid = cpu_to_logical_apicid(cpu);
+			if (apicid_cluster(apicid) !=
+					apicid_cluster(new_apicid)){
+				printk ("%s: Not a valid mask!\n", __func__);
+				return cpu_to_logical_apicid(0);
+			}
+			apicid = new_apicid;
+			cpus_found++;
+		}
+		cpu++;
+	}
+exit:
+	free_cpumask_var(cpumask);
+	return apicid;
+}
+
 static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
 {
 	return cpuid_apic >> index_msb;
diff --git a/arch/x86/include/asm/es7000/ipi.h b/arch/x86/include/asm/es7000/ipi.h
index 632a955..7e8ed24 100644
--- a/arch/x86/include/asm/es7000/ipi.h
+++ b/arch/x86/include/asm/es7000/ipi.h
@@ -1,24 +1,22 @@
 #ifndef __ASM_ES7000_IPI_H
 #define __ASM_ES7000_IPI_H
 
-void send_IPI_mask_sequence(cpumask_t mask, int vector);
+void send_IPI_mask_sequence(const struct cpumask *mask, int vector);
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector);
 
-static inline void send_IPI_mask(cpumask_t mask, int vector)
+static inline void send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	send_IPI_mask_sequence(mask, vector);
 }
 
 static inline void send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(smp_processor_id(), mask);
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, vector);
+	send_IPI_mask_allbutself(cpu_online_mask, vector);
 }
 
 static inline void send_IPI_all(int vector)
 {
-	send_IPI_mask(cpu_online_map, vector);
+	send_IPI_mask(cpu_online_mask, vector);
 }
 
 #endif /* __ASM_ES7000_IPI_H */
diff --git a/arch/x86/include/asm/genapic_32.h b/arch/x86/include/asm/genapic_32.h
index 0ac17d3..746f37a 100644
--- a/arch/x86/include/asm/genapic_32.h
+++ b/arch/x86/include/asm/genapic_32.h
@@ -24,7 +24,7 @@
 	int (*probe)(void);
 
 	int (*apic_id_registered)(void);
-	cpumask_t (*target_cpus)(void);
+	const struct cpumask *(*target_cpus)(void);
 	int int_delivery_mode;
 	int int_dest_mode;
 	int ESR_DISABLE;
@@ -57,12 +57,16 @@
 
 	unsigned (*get_apic_id)(unsigned long x);
 	unsigned long apic_id_mask;
-	unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask);
-	cpumask_t (*vector_allocation_domain)(int cpu);
+	unsigned int (*cpu_mask_to_apicid)(const struct cpumask *cpumask);
+	unsigned int (*cpu_mask_to_apicid_and)(const struct cpumask *cpumask,
+					       const struct cpumask *andmask);
+	void (*vector_allocation_domain)(int cpu, struct cpumask *retmask);
 
 #ifdef CONFIG_SMP
 	/* ipi */
-	void (*send_IPI_mask)(cpumask_t mask, int vector);
+	void (*send_IPI_mask)(const struct cpumask *mask, int vector);
+	void (*send_IPI_mask_allbutself)(const struct cpumask *mask,
+					 int vector);
 	void (*send_IPI_allbutself)(int vector);
 	void (*send_IPI_all)(int vector);
 #endif
@@ -114,6 +118,7 @@
 	APICFUNC(get_apic_id)				\
 	.apic_id_mask = APIC_ID_MASK,			\
 	APICFUNC(cpu_mask_to_apicid)			\
+	APICFUNC(cpu_mask_to_apicid_and)		\
 	APICFUNC(vector_allocation_domain)		\
 	APICFUNC(acpi_madt_oem_check)			\
 	IPIFUNC(send_IPI_mask)				\
diff --git a/arch/x86/include/asm/genapic_64.h b/arch/x86/include/asm/genapic_64.h
index 2cae011..adf32fb 100644
--- a/arch/x86/include/asm/genapic_64.h
+++ b/arch/x86/include/asm/genapic_64.h
@@ -1,6 +1,8 @@
 #ifndef _ASM_X86_GENAPIC_64_H
 #define _ASM_X86_GENAPIC_64_H
 
+#include <linux/cpumask.h>
+
 /*
  * Copyright 2004 James Cleverdon, IBM.
  * Subject to the GNU Public License, v.2
@@ -18,16 +20,20 @@
 	u32 int_delivery_mode;
 	u32 int_dest_mode;
 	int (*apic_id_registered)(void);
-	cpumask_t (*target_cpus)(void);
-	cpumask_t (*vector_allocation_domain)(int cpu);
+	const struct cpumask *(*target_cpus)(void);
+	void (*vector_allocation_domain)(int cpu, struct cpumask *retmask);
 	void (*init_apic_ldr)(void);
 	/* ipi */
-	void (*send_IPI_mask)(cpumask_t mask, int vector);
+	void (*send_IPI_mask)(const struct cpumask *mask, int vector);
+	void (*send_IPI_mask_allbutself)(const struct cpumask *mask,
+					 int vector);
 	void (*send_IPI_allbutself)(int vector);
 	void (*send_IPI_all)(int vector);
 	void (*send_IPI_self)(int vector);
 	/* */
-	unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask);
+	unsigned int (*cpu_mask_to_apicid)(const struct cpumask *cpumask);
+	unsigned int (*cpu_mask_to_apicid_and)(const struct cpumask *cpumask,
+					       const struct cpumask *andmask);
 	unsigned int (*phys_pkg_id)(int index_msb);
 	unsigned int (*get_apic_id)(unsigned long x);
 	unsigned long (*set_apic_id)(unsigned int id);
diff --git a/arch/x86/include/asm/ipi.h b/arch/x86/include/asm/ipi.h
index f89dffb..c745a30 100644
--- a/arch/x86/include/asm/ipi.h
+++ b/arch/x86/include/asm/ipi.h
@@ -117,7 +117,8 @@
 	native_apic_mem_write(APIC_ICR, cfg);
 }
 
-static inline void send_IPI_mask_sequence(cpumask_t mask, int vector)
+static inline void send_IPI_mask_sequence(const struct cpumask *mask,
+					  int vector)
 {
 	unsigned long flags;
 	unsigned long query_cpu;
@@ -128,11 +129,29 @@
 	 * - mbligh
 	 */
 	local_irq_save(flags);
-	for_each_cpu_mask_nr(query_cpu, mask) {
+	for_each_cpu(query_cpu, mask) {
 		__send_IPI_dest_field(per_cpu(x86_cpu_to_apicid, query_cpu),
 				      vector, APIC_DEST_PHYSICAL);
 	}
 	local_irq_restore(flags);
 }
 
+static inline void send_IPI_mask_allbutself(const struct cpumask *mask,
+					    int vector)
+{
+	unsigned long flags;
+	unsigned int query_cpu;
+	unsigned int this_cpu = smp_processor_id();
+
+	/* See Hack comment above */
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask)
+		if (query_cpu != this_cpu)
+			__send_IPI_dest_field(
+				per_cpu(x86_cpu_to_apicid, query_cpu),
+				vector, APIC_DEST_PHYSICAL);
+	local_irq_restore(flags);
+}
+
 #endif /* _ASM_X86_IPI_H */
diff --git a/arch/x86/include/asm/irq.h b/arch/x86/include/asm/irq.h
index 28e409f..592688e 100644
--- a/arch/x86/include/asm/irq.h
+++ b/arch/x86/include/asm/irq.h
@@ -33,7 +33,7 @@
 
 #ifdef CONFIG_HOTPLUG_CPU
 #include <linux/cpumask.h>
-extern void fixup_irqs(cpumask_t map);
+extern void fixup_irqs(void);
 #endif
 
 extern unsigned int do_IRQ(struct pt_regs *regs);
@@ -42,5 +42,6 @@
 
 /* Interrupt vector management */
 extern DECLARE_BITMAP(used_vectors, NR_VECTORS);
+extern int vector_used_by_percpu_irq(unsigned int vector);
 
 #endif /* _ASM_X86_IRQ_H */
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 8346be8..97215a4 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -21,6 +21,7 @@
 
 #include <asm/pvclock-abi.h>
 #include <asm/desc.h>
+#include <asm/mtrr.h>
 
 #define KVM_MAX_VCPUS 16
 #define KVM_MEMORY_SLOTS 32
@@ -86,6 +87,7 @@
 #define KVM_MIN_FREE_MMU_PAGES 5
 #define KVM_REFILL_PAGES 25
 #define KVM_MAX_CPUID_ENTRIES 40
+#define KVM_NR_FIXED_MTRR_REGION 88
 #define KVM_NR_VAR_MTRR 8
 
 extern spinlock_t kvm_lock;
@@ -180,6 +182,8 @@
 	struct list_head link;
 	struct hlist_node hash_link;
 
+	struct list_head oos_link;
+
 	/*
 	 * The following two entries are used to key the shadow page in the
 	 * hash table.
@@ -190,13 +194,16 @@
 	u64 *spt;
 	/* hold the gfn of each spte inside spt */
 	gfn_t *gfns;
-	unsigned long slot_bitmap; /* One bit set per slot which has memory
-				    * in this shadow page.
-				    */
+	/*
+	 * One bit set per slot which has memory
+	 * in this shadow page.
+	 */
+	DECLARE_BITMAP(slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS);
 	int multimapped;         /* More than one parent_pte? */
 	int root_count;          /* Currently serving as active root */
 	bool unsync;
-	bool unsync_children;
+	bool global;
+	unsigned int unsync_children;
 	union {
 		u64 *parent_pte;               /* !multimapped */
 		struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */
@@ -327,8 +334,10 @@
 
 	bool nmi_pending;
 	bool nmi_injected;
+	bool nmi_window_open;
 
-	u64 mtrr[0x100];
+	struct mtrr_state_type mtrr_state;
+	u32 pat;
 };
 
 struct kvm_mem_alias {
@@ -350,11 +359,13 @@
 	 */
 	struct list_head active_mmu_pages;
 	struct list_head assigned_dev_head;
+	struct list_head oos_global_pages;
 	struct dmar_domain *intel_iommu_domain;
 	struct kvm_pic *vpic;
 	struct kvm_ioapic *vioapic;
 	struct kvm_pit *vpit;
 	struct hlist_head irq_ack_notifier_list;
+	int vapics_in_nmi_mode;
 
 	int round_robin_prev_vcpu;
 	unsigned int tss_addr;
@@ -378,6 +389,7 @@
 	u32 mmu_recycled;
 	u32 mmu_cache_miss;
 	u32 mmu_unsync;
+	u32 mmu_unsync_global;
 	u32 remote_tlb_flush;
 	u32 lpages;
 };
@@ -397,6 +409,7 @@
 	u32 halt_exits;
 	u32 halt_wakeup;
 	u32 request_irq_exits;
+	u32 request_nmi_exits;
 	u32 irq_exits;
 	u32 host_state_reload;
 	u32 efer_reload;
@@ -405,6 +418,7 @@
 	u32 insn_emulation_fail;
 	u32 hypercalls;
 	u32 irq_injections;
+	u32 nmi_injections;
 };
 
 struct descriptor_table {
@@ -477,6 +491,7 @@
 
 	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
 	int (*get_tdp_level)(void);
+	int (*get_mt_mask_shift)(void);
 };
 
 extern struct kvm_x86_ops *kvm_x86_ops;
@@ -490,7 +505,7 @@
 void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte);
 void kvm_mmu_set_base_ptes(u64 base_pte);
 void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
-		u64 dirty_mask, u64 nx_mask, u64 x_mask);
+		u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 mt_mask);
 
 int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
 void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
@@ -587,12 +602,14 @@
 
 void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu);
 void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
-		       const u8 *new, int bytes);
+		       const u8 *new, int bytes,
+		       bool guest_initiated);
 int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva);
 void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
 int kvm_mmu_load(struct kvm_vcpu *vcpu);
 void kvm_mmu_unload(struct kvm_vcpu *vcpu);
 void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu);
+void kvm_mmu_sync_global(struct kvm_vcpu *vcpu);
 
 int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);
 
@@ -607,6 +624,8 @@
 int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3);
 int complete_pio(struct kvm_vcpu *vcpu);
 
+struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn);
+
 static inline struct kvm_mmu_page *page_header(hpa_t shadow_page)
 {
 	struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT);
@@ -702,18 +721,6 @@
 	kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
 }
 
-#define ASM_VMX_VMCLEAR_RAX       ".byte 0x66, 0x0f, 0xc7, 0x30"
-#define ASM_VMX_VMLAUNCH          ".byte 0x0f, 0x01, 0xc2"
-#define ASM_VMX_VMRESUME          ".byte 0x0f, 0x01, 0xc3"
-#define ASM_VMX_VMPTRLD_RAX       ".byte 0x0f, 0xc7, 0x30"
-#define ASM_VMX_VMREAD_RDX_RAX    ".byte 0x0f, 0x78, 0xd0"
-#define ASM_VMX_VMWRITE_RAX_RDX   ".byte 0x0f, 0x79, 0xd0"
-#define ASM_VMX_VMWRITE_RSP_RDX   ".byte 0x0f, 0x79, 0xd4"
-#define ASM_VMX_VMXOFF            ".byte 0x0f, 0x01, 0xc4"
-#define ASM_VMX_VMXON_RAX         ".byte 0xf3, 0x0f, 0xc7, 0x30"
-#define ASM_VMX_INVEPT		  ".byte 0x66, 0x0f, 0x38, 0x80, 0x08"
-#define ASM_VMX_INVVPID		  ".byte 0x66, 0x0f, 0x38, 0x81, 0x08"
-
 #define MSR_IA32_TIME_STAMP_COUNTER		0x010
 
 #define TSS_IOPB_BASE_OFFSET 0x66
diff --git a/arch/x86/include/asm/kvm_x86_emulate.h b/arch/x86/include/asm/kvm_x86_emulate.h
index 25179a2..6a15973 100644
--- a/arch/x86/include/asm/kvm_x86_emulate.h
+++ b/arch/x86/include/asm/kvm_x86_emulate.h
@@ -123,6 +123,7 @@
 	u8 ad_bytes;
 	u8 rex_prefix;
 	struct operand src;
+	struct operand src2;
 	struct operand dst;
 	bool has_seg_override;
 	u8 seg_override;
@@ -146,22 +147,18 @@
 	/* Register state before/after emulation. */
 	struct kvm_vcpu *vcpu;
 
-	/* Linear faulting address (if emulating a page-faulting instruction) */
 	unsigned long eflags;
-
 	/* Emulated execution mode, represented by an X86EMUL_MODE value. */
 	int mode;
-
 	u32 cs_base;
 
 	/* decode cache */
-
 	struct decode_cache decode;
 };
 
 /* Repeat String Operation Prefix */
-#define REPE_PREFIX  1
-#define REPNE_PREFIX    2
+#define REPE_PREFIX	1
+#define REPNE_PREFIX	2
 
 /* Execution mode, passed to the emulator. */
 #define X86EMUL_MODE_REAL     0	/* Real mode.             */
@@ -170,7 +167,7 @@
 #define X86EMUL_MODE_PROT64   8	/* 64-bit (long) mode.    */
 
 /* Host execution mode. */
-#if defined(__i386__)
+#if defined(CONFIG_X86_32)
 #define X86EMUL_MODE_HOST X86EMUL_MODE_PROT32
 #elif defined(CONFIG_X86_64)
 #define X86EMUL_MODE_HOST X86EMUL_MODE_PROT64
diff --git a/arch/x86/include/asm/mach-default/mach_apic.h b/arch/x86/include/asm/mach-default/mach_apic.h
index 6cb3a46..cc09cbb 100644
--- a/arch/x86/include/asm/mach-default/mach_apic.h
+++ b/arch/x86/include/asm/mach-default/mach_apic.h
@@ -8,12 +8,12 @@
 
 #define APIC_DFR_VALUE	(APIC_DFR_FLAT)
 
-static inline cpumask_t target_cpus(void)
+static inline const struct cpumask *target_cpus(void)
 { 
 #ifdef CONFIG_SMP
-	return cpu_online_map;
+	return cpu_online_mask;
 #else
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 #endif
 } 
 
@@ -28,6 +28,7 @@
 #define apic_id_registered (genapic->apic_id_registered)
 #define init_apic_ldr (genapic->init_apic_ldr)
 #define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid)
+#define cpu_mask_to_apicid_and (genapic->cpu_mask_to_apicid_and)
 #define phys_pkg_id	(genapic->phys_pkg_id)
 #define vector_allocation_domain    (genapic->vector_allocation_domain)
 #define read_apic_id()  (GET_APIC_ID(apic_read(APIC_ID)))
@@ -61,9 +62,19 @@
 	return physid_isset(read_apic_id(), phys_cpu_present_map);
 }
 
-static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
+static inline unsigned int cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
-	return cpus_addr(cpumask)[0];
+	return cpumask_bits(cpumask)[0];
+}
+
+static inline unsigned int cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	unsigned long mask1 = cpumask_bits(cpumask)[0];
+	unsigned long mask2 = cpumask_bits(andmask)[0];
+	unsigned long mask3 = cpumask_bits(cpu_online_mask)[0];
+
+	return (unsigned int)(mask1 & mask2 & mask3);
 }
 
 static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
@@ -88,7 +99,7 @@
 #endif
 }
 
-static inline cpumask_t vector_allocation_domain(int cpu)
+static inline void vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
         /* Careful. Some cpus do not strictly honor the set of cpus
          * specified in the interrupt destination when using lowest
@@ -98,8 +109,7 @@
          * deliver interrupts to the wrong hyperthread when only one
          * hyperthread was specified in the interrupt desitination.
          */
-        cpumask_t domain = { { [0] = APIC_ALL_CPUS, } };
-        return domain;
+	*retmask = (cpumask_t) { { [0] = APIC_ALL_CPUS } };
 }
 #endif
 
@@ -131,7 +141,7 @@
 
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
-	if (mps_cpu < NR_CPUS && cpu_present(mps_cpu))
+	if (mps_cpu < nr_cpu_ids && cpu_present(mps_cpu))
 		return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu);
 	else
 		return BAD_APICID;
diff --git a/arch/x86/include/asm/mach-default/mach_ipi.h b/arch/x86/include/asm/mach-default/mach_ipi.h
index fabca01..191312d 100644
--- a/arch/x86/include/asm/mach-default/mach_ipi.h
+++ b/arch/x86/include/asm/mach-default/mach_ipi.h
@@ -4,7 +4,8 @@
 /* Avoid include hell */
 #define NMI_VECTOR 0x02
 
-void send_IPI_mask_bitmask(cpumask_t mask, int vector);
+void send_IPI_mask_bitmask(const struct cpumask *mask, int vector);
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector);
 void __send_IPI_shortcut(unsigned int shortcut, int vector);
 
 extern int no_broadcast;
@@ -12,28 +13,27 @@
 #ifdef CONFIG_X86_64
 #include <asm/genapic.h>
 #define send_IPI_mask (genapic->send_IPI_mask)
+#define send_IPI_mask_allbutself (genapic->send_IPI_mask_allbutself)
 #else
-static inline void send_IPI_mask(cpumask_t mask, int vector)
+static inline void send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	send_IPI_mask_bitmask(mask, vector);
 }
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector);
 #endif
 
 static inline void __local_send_IPI_allbutself(int vector)
 {
-	if (no_broadcast || vector == NMI_VECTOR) {
-		cpumask_t mask = cpu_online_map;
-
-		cpu_clear(smp_processor_id(), mask);
-		send_IPI_mask(mask, vector);
-	} else
+	if (no_broadcast || vector == NMI_VECTOR)
+		send_IPI_mask_allbutself(cpu_online_mask, vector);
+	else
 		__send_IPI_shortcut(APIC_DEST_ALLBUT, vector);
 }
 
 static inline void __local_send_IPI_all(int vector)
 {
 	if (no_broadcast || vector == NMI_VECTOR)
-		send_IPI_mask(cpu_online_map, vector);
+		send_IPI_mask(cpu_online_mask, vector);
 	else
 		__send_IPI_shortcut(APIC_DEST_ALLINC, vector);
 }
diff --git a/arch/x86/include/asm/mach-generic/mach_apic.h b/arch/x86/include/asm/mach-generic/mach_apic.h
index e430f47..48553e9 100644
--- a/arch/x86/include/asm/mach-generic/mach_apic.h
+++ b/arch/x86/include/asm/mach-generic/mach_apic.h
@@ -24,6 +24,7 @@
 #define check_phys_apicid_present (genapic->check_phys_apicid_present)
 #define check_apicid_used (genapic->check_apicid_used)
 #define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid)
+#define cpu_mask_to_apicid_and (genapic->cpu_mask_to_apicid_and)
 #define vector_allocation_domain (genapic->vector_allocation_domain)
 #define enable_apic_mode (genapic->enable_apic_mode)
 #define phys_pkg_id (genapic->phys_pkg_id)
diff --git a/arch/x86/include/asm/mpspec.h b/arch/x86/include/asm/mpspec.h
index 91885c2..62d14ce 100644
--- a/arch/x86/include/asm/mpspec.h
+++ b/arch/x86/include/asm/mpspec.h
@@ -6,13 +6,13 @@
 #include <asm/mpspec_def.h>
 
 extern int apic_version[MAX_APICS];
+extern int pic_mode;
 
 #ifdef CONFIG_X86_32
 #include <mach_mpspec.h>
 
 extern unsigned int def_to_bigsmp;
 extern u8 apicid_2_node[];
-extern int pic_mode;
 
 #ifdef CONFIG_X86_NUMAQ
 extern int mp_bus_id_to_node[MAX_MP_BUSSES];
diff --git a/arch/x86/include/asm/mtrr.h b/arch/x86/include/asm/mtrr.h
index 7c1e425..cb988aa 100644
--- a/arch/x86/include/asm/mtrr.h
+++ b/arch/x86/include/asm/mtrr.h
@@ -57,6 +57,31 @@
 };
 #endif /* !__i386__ */
 
+struct mtrr_var_range {
+	u32 base_lo;
+	u32 base_hi;
+	u32 mask_lo;
+	u32 mask_hi;
+};
+
+/* In the Intel processor's MTRR interface, the MTRR type is always held in
+   an 8 bit field: */
+typedef u8 mtrr_type;
+
+#define MTRR_NUM_FIXED_RANGES 88
+#define MTRR_MAX_VAR_RANGES 256
+
+struct mtrr_state_type {
+	struct mtrr_var_range var_ranges[MTRR_MAX_VAR_RANGES];
+	mtrr_type fixed_ranges[MTRR_NUM_FIXED_RANGES];
+	unsigned char enabled;
+	unsigned char have_fixed;
+	mtrr_type def_type;
+};
+
+#define MTRRphysBase_MSR(reg) (0x200 + 2 * (reg))
+#define MTRRphysMask_MSR(reg) (0x200 + 2 * (reg) + 1)
+
 /*  These are the various ioctls  */
 #define MTRRIOC_ADD_ENTRY        _IOW(MTRR_IOCTL_BASE,  0, struct mtrr_sentry)
 #define MTRRIOC_SET_ENTRY        _IOW(MTRR_IOCTL_BASE,  1, struct mtrr_sentry)
diff --git a/arch/x86/include/asm/numaq/apic.h b/arch/x86/include/asm/numaq/apic.h
index 0bf2a06..c80f00d 100644
--- a/arch/x86/include/asm/numaq/apic.h
+++ b/arch/x86/include/asm/numaq/apic.h
@@ -7,9 +7,9 @@
 
 #define APIC_DFR_VALUE	(APIC_DFR_CLUSTER)
 
-static inline cpumask_t target_cpus(void)
+static inline const cpumask_t *target_cpus(void)
 {
-	return CPU_MASK_ALL;
+	return &CPU_MASK_ALL;
 }
 
 #define NO_BALANCE_IRQ (1)
@@ -122,7 +122,13 @@
  * We use physical apicids here, not logical, so just return the default
  * physical broadcast to stop people from breaking us
  */
-static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
+static inline unsigned int cpu_mask_to_apicid(const cpumask_t *cpumask)
+{
+	return (int) 0xF;
+}
+
+static inline unsigned int cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
 {
 	return (int) 0xF;
 }
diff --git a/arch/x86/include/asm/numaq/ipi.h b/arch/x86/include/asm/numaq/ipi.h
index 935588d2..a8374c6 100644
--- a/arch/x86/include/asm/numaq/ipi.h
+++ b/arch/x86/include/asm/numaq/ipi.h
@@ -1,25 +1,22 @@
 #ifndef __ASM_NUMAQ_IPI_H
 #define __ASM_NUMAQ_IPI_H
 
-void send_IPI_mask_sequence(cpumask_t, int vector);
+void send_IPI_mask_sequence(const struct cpumask *mask, int vector);
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector);
 
-static inline void send_IPI_mask(cpumask_t mask, int vector)
+static inline void send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	send_IPI_mask_sequence(mask, vector);
 }
 
 static inline void send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(smp_processor_id(), mask);
-
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, vector);
+	send_IPI_mask_allbutself(cpu_online_mask, vector);
 }
 
 static inline void send_IPI_all(int vector)
 {
-	send_IPI_mask(cpu_online_map, vector);
+	send_IPI_mask(cpu_online_mask, vector);
 }
 
 #endif /* __ASM_NUMAQ_IPI_H */
diff --git a/arch/x86/pci/pci.h b/arch/x86/include/asm/pci_x86.h
similarity index 88%
rename from arch/x86/pci/pci.h
rename to arch/x86/include/asm/pci_x86.h
index 1959018..e60fd3e 100644
--- a/arch/x86/pci/pci.h
+++ b/arch/x86/include/asm/pci_x86.h
@@ -57,7 +57,8 @@
 struct irq_info {
 	u8 bus, devfn;			/* Bus, device and function */
 	struct {
-		u8 link;		/* IRQ line ID, chipset dependent, 0=not routed */
+		u8 link;		/* IRQ line ID, chipset dependent,
+					   0 = not routed */
 		u16 bitmap;		/* Available IRQs */
 	} __attribute__((packed)) irq[4];
 	u8 slot;			/* Slot number, 0=onboard */
@@ -69,11 +70,13 @@
 	u16 version;			/* PIRQ_VERSION */
 	u16 size;			/* Table size in bytes */
 	u8 rtr_bus, rtr_devfn;		/* Where the interrupt router lies */
-	u16 exclusive_irqs;		/* IRQs devoted exclusively to PCI usage */
-	u16 rtr_vendor, rtr_device;	/* Vendor and device ID of interrupt router */
+	u16 exclusive_irqs;		/* IRQs devoted exclusively to
+					   PCI usage */
+	u16 rtr_vendor, rtr_device;	/* Vendor and device ID of
+					   interrupt router */
 	u32 miniport_data;		/* Crap */
 	u8 rfu[11];
-	u8 checksum;			/* Modulo 256 checksum must give zero */
+	u8 checksum;			/* Modulo 256 checksum must give 0 */
 	struct irq_info slots[0];
 } __attribute__((packed));
 
@@ -148,15 +151,15 @@
 
 static inline void mmio_config_writeb(void __iomem *pos, u8 val)
 {
-	asm volatile("movb %%al,(%1)" :: "a" (val), "r" (pos) : "memory");
+	asm volatile("movb %%al,(%1)" : : "a" (val), "r" (pos) : "memory");
 }
 
 static inline void mmio_config_writew(void __iomem *pos, u16 val)
 {
-	asm volatile("movw %%ax,(%1)" :: "a" (val), "r" (pos) : "memory");
+	asm volatile("movw %%ax,(%1)" : : "a" (val), "r" (pos) : "memory");
 }
 
 static inline void mmio_config_writel(void __iomem *pos, u32 val)
 {
-	asm volatile("movl %%eax,(%1)" :: "a" (val), "r" (pos) : "memory");
+	asm volatile("movl %%eax,(%1)" : : "a" (val), "r" (pos) : "memory");
 }
diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
index d12811c..830b9fc 100644
--- a/arch/x86/include/asm/smp.h
+++ b/arch/x86/include/asm/smp.h
@@ -60,7 +60,7 @@
 	void (*cpu_die)(unsigned int cpu);
 	void (*play_dead)(void);
 
-	void (*send_call_func_ipi)(cpumask_t mask);
+	void (*send_call_func_ipi)(const struct cpumask *mask);
 	void (*send_call_func_single_ipi)(int cpu);
 };
 
@@ -125,7 +125,7 @@
 
 static inline void arch_send_call_function_ipi(cpumask_t mask)
 {
-	smp_ops.send_call_func_ipi(mask);
+	smp_ops.send_call_func_ipi(&mask);
 }
 
 void cpu_disable_common(void);
@@ -138,7 +138,7 @@
 void native_play_dead(void);
 void play_dead_common(void);
 
-void native_send_call_func_ipi(cpumask_t mask);
+void native_send_call_func_ipi(const struct cpumask *mask);
 void native_send_call_func_single_ipi(int cpu);
 
 extern void prefill_possible_map(void);
diff --git a/arch/x86/include/asm/summit/apic.h b/arch/x86/include/asm/summit/apic.h
index 9b3070f..99327d1 100644
--- a/arch/x86/include/asm/summit/apic.h
+++ b/arch/x86/include/asm/summit/apic.h
@@ -14,13 +14,13 @@
 
 #define APIC_DFR_VALUE	(APIC_DFR_CLUSTER)
 
-static inline cpumask_t target_cpus(void)
+static inline const cpumask_t *target_cpus(void)
 {
 	/* CPU_MASK_ALL (0xff) has undefined behaviour with
 	 * dest_LowestPrio mode logical clustered apic interrupt routing
 	 * Just start on cpu 0.  IRQ balancing will spread load
 	 */
-	return cpumask_of_cpu(0);
+	return &cpumask_of_cpu(0);
 }
 
 #define INT_DELIVERY_MODE (dest_LowestPrio)
@@ -137,14 +137,14 @@
 {
 }
 
-static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
+static inline unsigned int cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
 	int num_bits_set;
 	int cpus_found = 0;
 	int cpu;
 	int apicid;
 
-	num_bits_set = cpus_weight(cpumask);
+	num_bits_set = cpus_weight(*cpumask);
 	/* Return id to all */
 	if (num_bits_set == NR_CPUS)
 		return (int) 0xFF;
@@ -152,10 +152,10 @@
 	 * The cpus in the mask must all be on the apic cluster.  If are not
 	 * on the same apicid cluster return default value of TARGET_CPUS.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = first_cpu(*cpumask);
 	apicid = cpu_to_logical_apicid(cpu);
 	while (cpus_found < num_bits_set) {
-		if (cpu_isset(cpu, cpumask)) {
+		if (cpu_isset(cpu, *cpumask)) {
 			int new_apicid = cpu_to_logical_apicid(cpu);
 			if (apicid_cluster(apicid) !=
 					apicid_cluster(new_apicid)){
@@ -170,6 +170,49 @@
 	return apicid;
 }
 
+static inline unsigned int cpu_mask_to_apicid_and(const struct cpumask *inmask,
+						  const struct cpumask *andmask)
+{
+	int num_bits_set;
+	int cpus_found = 0;
+	int cpu;
+	int apicid = 0xFF;
+	cpumask_var_t cpumask;
+
+	if (!alloc_cpumask_var(&cpumask, GFP_ATOMIC))
+		return (int) 0xFF;
+
+	cpumask_and(cpumask, inmask, andmask);
+	cpumask_and(cpumask, cpumask, cpu_online_mask);
+
+	num_bits_set = cpumask_weight(cpumask);
+	/* Return id to all */
+	if (num_bits_set == nr_cpu_ids)
+		goto exit;
+	/*
+	 * The cpus in the mask must all be on the apic cluster.  If are not
+	 * on the same apicid cluster return default value of TARGET_CPUS.
+	 */
+	cpu = cpumask_first(cpumask);
+	apicid = cpu_to_logical_apicid(cpu);
+	while (cpus_found < num_bits_set) {
+		if (cpumask_test_cpu(cpu, cpumask)) {
+			int new_apicid = cpu_to_logical_apicid(cpu);
+			if (apicid_cluster(apicid) !=
+					apicid_cluster(new_apicid)){
+				printk ("%s: Not a valid mask!\n", __func__);
+				return 0xFF;
+			}
+			apicid = apicid | new_apicid;
+			cpus_found++;
+		}
+		cpu++;
+	}
+exit:
+	free_cpumask_var(cpumask);
+	return apicid;
+}
+
 /* cpuid returns the value latched in the HW at reset, not the APIC ID
  * register's value.  For any box whose BIOS changes APIC IDs, like
  * clustered APIC systems, we must use hard_smp_processor_id.
diff --git a/arch/x86/include/asm/summit/ipi.h b/arch/x86/include/asm/summit/ipi.h
index 53bd1e7..a8a2c24 100644
--- a/arch/x86/include/asm/summit/ipi.h
+++ b/arch/x86/include/asm/summit/ipi.h
@@ -1,9 +1,10 @@
 #ifndef __ASM_SUMMIT_IPI_H
 #define __ASM_SUMMIT_IPI_H
 
-void send_IPI_mask_sequence(cpumask_t mask, int vector);
+void send_IPI_mask_sequence(const cpumask_t *mask, int vector);
+void send_IPI_mask_allbutself(const cpumask_t *mask, int vector);
 
-static inline void send_IPI_mask(cpumask_t mask, int vector)
+static inline void send_IPI_mask(const cpumask_t *mask, int vector)
 {
 	send_IPI_mask_sequence(mask, vector);
 }
@@ -14,12 +15,12 @@
 	cpu_clear(smp_processor_id(), mask);
 
 	if (!cpus_empty(mask))
-		send_IPI_mask(mask, vector);
+		send_IPI_mask(&mask, vector);
 }
 
 static inline void send_IPI_all(int vector)
 {
-	send_IPI_mask(cpu_online_map, vector);
+	send_IPI_mask(&cpu_online_map, vector);
 }
 
 #endif /* __ASM_SUMMIT_IPI_H */
diff --git a/arch/x86/kvm/svm.h b/arch/x86/include/asm/svm.h
similarity index 100%
rename from arch/x86/kvm/svm.h
rename to arch/x86/include/asm/svm.h
diff --git a/arch/x86/include/asm/sys_ia32.h b/arch/x86/include/asm/sys_ia32.h
new file mode 100644
index 0000000..ffb08be
--- /dev/null
+++ b/arch/x86/include/asm/sys_ia32.h
@@ -0,0 +1,101 @@
+/*
+ * sys_ia32.h - Linux ia32 syscall interfaces
+ *
+ * Copyright (c) 2008 Jaswinder Singh Rajput
+ *
+ * This file is released under the GPLv2.
+ * See the file COPYING for more details.
+ */
+
+#ifndef _ASM_X86_SYS_IA32_H
+#define _ASM_X86_SYS_IA32_H
+
+#include <linux/compiler.h>
+#include <linux/linkage.h>
+#include <linux/types.h>
+#include <linux/signal.h>
+#include <asm/compat.h>
+#include <asm/ia32.h>
+
+/* ia32/sys_ia32.c */
+asmlinkage long sys32_truncate64(char __user *, unsigned long, unsigned long);
+asmlinkage long sys32_ftruncate64(unsigned int, unsigned long, unsigned long);
+
+asmlinkage long sys32_stat64(char __user *, struct stat64 __user *);
+asmlinkage long sys32_lstat64(char __user *, struct stat64 __user *);
+asmlinkage long sys32_fstat64(unsigned int, struct stat64 __user *);
+asmlinkage long sys32_fstatat(unsigned int, char __user *,
+			      struct stat64 __user *, int);
+struct mmap_arg_struct;
+asmlinkage long sys32_mmap(struct mmap_arg_struct __user *);
+asmlinkage long sys32_mprotect(unsigned long, size_t, unsigned long);
+
+asmlinkage long sys32_pipe(int __user *);
+struct sigaction32;
+struct old_sigaction32;
+asmlinkage long sys32_rt_sigaction(int, struct sigaction32 __user *,
+				   struct sigaction32 __user *, unsigned int);
+asmlinkage long sys32_sigaction(int, struct old_sigaction32 __user *,
+				struct old_sigaction32 __user *);
+asmlinkage long sys32_rt_sigprocmask(int, compat_sigset_t __user *,
+				     compat_sigset_t __user *, unsigned int);
+asmlinkage long sys32_alarm(unsigned int);
+
+struct sel_arg_struct;
+asmlinkage long sys32_old_select(struct sel_arg_struct __user *);
+asmlinkage long sys32_waitpid(compat_pid_t, unsigned int *, int);
+asmlinkage long sys32_sysfs(int, u32, u32);
+
+asmlinkage long sys32_sched_rr_get_interval(compat_pid_t,
+					    struct compat_timespec __user *);
+asmlinkage long sys32_rt_sigpending(compat_sigset_t __user *, compat_size_t);
+asmlinkage long sys32_rt_sigqueueinfo(int, int, compat_siginfo_t __user *);
+
+#ifdef CONFIG_SYSCTL_SYSCALL
+struct sysctl_ia32;
+asmlinkage long sys32_sysctl(struct sysctl_ia32 __user *);
+#endif
+
+asmlinkage long sys32_pread(unsigned int, char __user *, u32, u32, u32);
+asmlinkage long sys32_pwrite(unsigned int, char __user *, u32, u32, u32);
+
+asmlinkage long sys32_personality(unsigned long);
+asmlinkage long sys32_sendfile(int, int, compat_off_t __user *, s32);
+
+asmlinkage long sys32_mmap2(unsigned long, unsigned long, unsigned long,
+			    unsigned long, unsigned long, unsigned long);
+
+struct oldold_utsname;
+struct old_utsname;
+asmlinkage long sys32_olduname(struct oldold_utsname __user *);
+long sys32_uname(struct old_utsname __user *);
+
+long sys32_ustat(unsigned, struct ustat32 __user *);
+
+asmlinkage long sys32_execve(char __user *, compat_uptr_t __user *,
+			     compat_uptr_t __user *, struct pt_regs *);
+asmlinkage long sys32_clone(unsigned int, unsigned int, struct pt_regs *);
+
+long sys32_lseek(unsigned int, int, unsigned int);
+long sys32_kill(int, int);
+long sys32_fadvise64_64(int, __u32, __u32, __u32, __u32, int);
+long sys32_vm86_warning(void);
+long sys32_lookup_dcookie(u32, u32, char __user *, size_t);
+
+asmlinkage ssize_t sys32_readahead(int, unsigned, unsigned, size_t);
+asmlinkage long sys32_sync_file_range(int, unsigned, unsigned,
+				      unsigned, unsigned, int);
+asmlinkage long sys32_fadvise64(int, unsigned, unsigned, size_t, int);
+asmlinkage long sys32_fallocate(int, int, unsigned,
+				unsigned, unsigned, unsigned);
+
+/* ia32/ia32_signal.c */
+asmlinkage long sys32_sigsuspend(int, int, old_sigset_t);
+asmlinkage long sys32_sigaltstack(const stack_ia32_t __user *,
+				  stack_ia32_t __user *, struct pt_regs *);
+asmlinkage long sys32_sigreturn(struct pt_regs *);
+asmlinkage long sys32_rt_sigreturn(struct pt_regs *);
+
+/* ia32/ipc32.c */
+asmlinkage long sys32_ipc(u32, int, int, int, compat_uptr_t, u32);
+#endif /* _ASM_X86_SYS_IA32_H */
diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h
index ff386ff..79e31e9 100644
--- a/arch/x86/include/asm/topology.h
+++ b/arch/x86/include/asm/topology.h
@@ -226,6 +226,8 @@
 #define topology_core_id(cpu)			(cpu_data(cpu).cpu_core_id)
 #define topology_core_siblings(cpu)		(per_cpu(cpu_core_map, cpu))
 #define topology_thread_siblings(cpu)		(per_cpu(cpu_sibling_map, cpu))
+#define topology_core_cpumask(cpu)		(&per_cpu(cpu_core_map, cpu))
+#define topology_thread_cpumask(cpu)		(&per_cpu(cpu_sibling_map, cpu))
 
 /* indicates that pointers to the topology cpumask_t maps are valid */
 #define arch_provides_topology_pointers		yes
diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h
index e236325..50423c7 100644
--- a/arch/x86/include/asm/uv/uv_bau.h
+++ b/arch/x86/include/asm/uv/uv_bau.h
@@ -133,61 +133,61 @@
  * see table 4.2.3.0.1 in broacast_assist spec.
  */
 struct bau_msg_header {
-	int dest_subnodeid:6;	/* must be zero */
+	unsigned int dest_subnodeid:6;	/* must be zero */
 	/* bits 5:0 */
-	int base_dest_nodeid:15; /* nasid>>1 (pnode) of first bit in node_map */
-	/* bits 20:6 */
-	int command:8;		/* message type */
+	unsigned int base_dest_nodeid:15; /* nasid>>1 (pnode) of */
+	/* bits 20:6 */			  /* first bit in node_map */
+	unsigned int command:8;	/* message type */
 	/* bits 28:21 */
 				/* 0x38: SN3net EndPoint Message */
-	int rsvd_1:3;		/* must be zero */
+	unsigned int rsvd_1:3;	/* must be zero */
 	/* bits 31:29 */
 				/* int will align on 32 bits */
-	int rsvd_2:9;		/* must be zero */
+	unsigned int rsvd_2:9;	/* must be zero */
 	/* bits 40:32 */
 				/* Suppl_A is 56-41 */
-	int payload_2a:8;	/* becomes byte 16 of msg */
+	unsigned int payload_2a:8;/* becomes byte 16 of msg */
 	/* bits 48:41 */	/* not currently using */
-	int payload_2b:8;	/* becomes byte 17 of msg */
+	unsigned int payload_2b:8;/* becomes byte 17 of msg */
 	/* bits 56:49 */	/* not currently using */
 				/* Address field (96:57) is never used as an
 				   address (these are address bits 42:3) */
-	int rsvd_3:1;		/* must be zero */
+	unsigned int rsvd_3:1;	/* must be zero */
 	/* bit 57 */
 				/* address bits 27:4 are payload */
 				/* these 24 bits become bytes 12-14 of msg */
-	int replied_to:1;	/* sent as 0 by the source to byte 12 */
+	unsigned int replied_to:1;/* sent as 0 by the source to byte 12 */
 	/* bit 58 */
 
-	int payload_1a:5;	/* not currently used */
+	unsigned int payload_1a:5;/* not currently used */
 	/* bits 63:59 */
-	int payload_1b:8;	/* not currently used */
+	unsigned int payload_1b:8;/* not currently used */
 	/* bits 71:64 */
-	int payload_1c:8;	/* not currently used */
+	unsigned int payload_1c:8;/* not currently used */
 	/* bits 79:72 */
-	int payload_1d:2;	/* not currently used */
+	unsigned int payload_1d:2;/* not currently used */
 	/* bits 81:80 */
 
-	int rsvd_4:7;		/* must be zero */
+	unsigned int rsvd_4:7;	/* must be zero */
 	/* bits 88:82 */
-	int sw_ack_flag:1;	/* software acknowledge flag */
+	unsigned int sw_ack_flag:1;/* software acknowledge flag */
 	/* bit 89 */
 				/* INTD trasactions at destination are to
 				   wait for software acknowledge */
-	int rsvd_5:6;		/* must be zero */
+	unsigned int rsvd_5:6;	/* must be zero */
 	/* bits 95:90 */
-	int rsvd_6:5;		/* must be zero */
+	unsigned int rsvd_6:5;	/* must be zero */
 	/* bits 100:96 */
-	int int_both:1;		/* if 1, interrupt both sockets on the blade */
+	unsigned int int_both:1;/* if 1, interrupt both sockets on the blade */
 	/* bit 101*/
-	int fairness:3;		/* usually zero */
+	unsigned int fairness:3;/* usually zero */
 	/* bits 104:102 */
-	int multilevel:1;	/* multi-level multicast format */
+	unsigned int multilevel:1;	/* multi-level multicast format */
 	/* bit 105 */
 				/* 0 for TLB: endpoint multi-unicast messages */
-	int chaining:1;		/* next descriptor is part of this activation*/
+	unsigned int chaining:1;/* next descriptor is part of this activation*/
 	/* bit 106 */
-	int rsvd_7:21;		/* must be zero */
+	unsigned int rsvd_7:21;	/* must be zero */
 	/* bits 127:107 */
 };
 
diff --git a/arch/x86/include/asm/virtext.h b/arch/x86/include/asm/virtext.h
new file mode 100644
index 0000000..5936362
--- /dev/null
+++ b/arch/x86/include/asm/virtext.h
@@ -0,0 +1,132 @@
+/* CPU virtualization extensions handling
+ *
+ * This should carry the code for handling CPU virtualization extensions
+ * that needs to live in the kernel core.
+ *
+ * Author: Eduardo Habkost <ehabkost@redhat.com>
+ *
+ * Copyright (C) 2008, Red Hat Inc.
+ *
+ * Contains code from KVM, Copyright (C) 2006 Qumranet, Inc.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ */
+#ifndef _ASM_X86_VIRTEX_H
+#define _ASM_X86_VIRTEX_H
+
+#include <asm/processor.h>
+#include <asm/system.h>
+
+#include <asm/vmx.h>
+#include <asm/svm.h>
+
+/*
+ * VMX functions:
+ */
+
+static inline int cpu_has_vmx(void)
+{
+	unsigned long ecx = cpuid_ecx(1);
+	return test_bit(5, &ecx); /* CPUID.1:ECX.VMX[bit 5] -> VT */
+}
+
+
+/** Disable VMX on the current CPU
+ *
+ * vmxoff causes a undefined-opcode exception if vmxon was not run
+ * on the CPU previously. Only call this function if you know VMX
+ * is enabled.
+ */
+static inline void cpu_vmxoff(void)
+{
+	asm volatile (ASM_VMX_VMXOFF : : : "cc");
+	write_cr4(read_cr4() & ~X86_CR4_VMXE);
+}
+
+static inline int cpu_vmx_enabled(void)
+{
+	return read_cr4() & X86_CR4_VMXE;
+}
+
+/** Disable VMX if it is enabled on the current CPU
+ *
+ * You shouldn't call this if cpu_has_vmx() returns 0.
+ */
+static inline void __cpu_emergency_vmxoff(void)
+{
+	if (cpu_vmx_enabled())
+		cpu_vmxoff();
+}
+
+/** Disable VMX if it is supported and enabled on the current CPU
+ */
+static inline void cpu_emergency_vmxoff(void)
+{
+	if (cpu_has_vmx())
+		__cpu_emergency_vmxoff();
+}
+
+
+
+
+/*
+ * SVM functions:
+ */
+
+/** Check if the CPU has SVM support
+ *
+ * You can use the 'msg' arg to get a message describing the problem,
+ * if the function returns zero. Simply pass NULL if you are not interested
+ * on the messages; gcc should take care of not generating code for
+ * the messages on this case.
+ */
+static inline int cpu_has_svm(const char **msg)
+{
+	uint32_t eax, ebx, ecx, edx;
+
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) {
+		if (msg)
+			*msg = "not amd";
+		return 0;
+	}
+
+	cpuid(0x80000000, &eax, &ebx, &ecx, &edx);
+	if (eax < SVM_CPUID_FUNC) {
+		if (msg)
+			*msg = "can't execute cpuid_8000000a";
+		return 0;
+	}
+
+	cpuid(0x80000001, &eax, &ebx, &ecx, &edx);
+	if (!(ecx & (1 << SVM_CPUID_FEATURE_SHIFT))) {
+		if (msg)
+			*msg = "svm not available";
+		return 0;
+	}
+	return 1;
+}
+
+
+/** Disable SVM on the current CPU
+ *
+ * You should call this only if cpu_has_svm() returned true.
+ */
+static inline void cpu_svm_disable(void)
+{
+	uint64_t efer;
+
+	wrmsrl(MSR_VM_HSAVE_PA, 0);
+	rdmsrl(MSR_EFER, efer);
+	wrmsrl(MSR_EFER, efer & ~MSR_EFER_SVME_MASK);
+}
+
+/** Makes sure SVM is disabled, if it is supported on the CPU
+ */
+static inline void cpu_emergency_svm_disable(void)
+{
+	if (cpu_has_svm(NULL))
+		cpu_svm_disable();
+}
+
+#endif /* _ASM_X86_VIRTEX_H */
diff --git a/arch/x86/kvm/vmx.h b/arch/x86/include/asm/vmx.h
similarity index 92%
rename from arch/x86/kvm/vmx.h
rename to arch/x86/include/asm/vmx.h
index ec5edc3..d0238e6 100644
--- a/arch/x86/kvm/vmx.h
+++ b/arch/x86/include/asm/vmx.h
@@ -63,10 +63,13 @@
 
 #define VM_EXIT_HOST_ADDR_SPACE_SIZE            0x00000200
 #define VM_EXIT_ACK_INTR_ON_EXIT                0x00008000
+#define VM_EXIT_SAVE_IA32_PAT			0x00040000
+#define VM_EXIT_LOAD_IA32_PAT			0x00080000
 
 #define VM_ENTRY_IA32E_MODE                     0x00000200
 #define VM_ENTRY_SMM                            0x00000400
 #define VM_ENTRY_DEACT_DUAL_MONITOR             0x00000800
+#define VM_ENTRY_LOAD_IA32_PAT			0x00004000
 
 /* VMCS Encodings */
 enum vmcs_field {
@@ -112,6 +115,8 @@
 	VMCS_LINK_POINTER_HIGH          = 0x00002801,
 	GUEST_IA32_DEBUGCTL             = 0x00002802,
 	GUEST_IA32_DEBUGCTL_HIGH        = 0x00002803,
+	GUEST_IA32_PAT			= 0x00002804,
+	GUEST_IA32_PAT_HIGH		= 0x00002805,
 	GUEST_PDPTR0                    = 0x0000280a,
 	GUEST_PDPTR0_HIGH               = 0x0000280b,
 	GUEST_PDPTR1                    = 0x0000280c,
@@ -120,6 +125,8 @@
 	GUEST_PDPTR2_HIGH               = 0x0000280f,
 	GUEST_PDPTR3                    = 0x00002810,
 	GUEST_PDPTR3_HIGH               = 0x00002811,
+	HOST_IA32_PAT			= 0x00002c00,
+	HOST_IA32_PAT_HIGH		= 0x00002c01,
 	PIN_BASED_VM_EXEC_CONTROL       = 0x00004000,
 	CPU_BASED_VM_EXEC_CONTROL       = 0x00004002,
 	EXCEPTION_BITMAP                = 0x00004004,
@@ -331,8 +338,9 @@
 
 #define AR_RESERVD_MASK 0xfffe0f00
 
-#define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT	9
-#define IDENTITY_PAGETABLE_PRIVATE_MEMSLOT	10
+#define TSS_PRIVATE_MEMSLOT			(KVM_MEMORY_SLOTS + 0)
+#define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT	(KVM_MEMORY_SLOTS + 1)
+#define IDENTITY_PAGETABLE_PRIVATE_MEMSLOT	(KVM_MEMORY_SLOTS + 2)
 
 #define VMX_NR_VPIDS				(1 << 16)
 #define VMX_VPID_EXTENT_SINGLE_CONTEXT		1
@@ -356,4 +364,19 @@
 
 #define VMX_EPT_IDENTITY_PAGETABLE_ADDR		0xfffbc000ul
 
+
+#define ASM_VMX_VMCLEAR_RAX       ".byte 0x66, 0x0f, 0xc7, 0x30"
+#define ASM_VMX_VMLAUNCH          ".byte 0x0f, 0x01, 0xc2"
+#define ASM_VMX_VMRESUME          ".byte 0x0f, 0x01, 0xc3"
+#define ASM_VMX_VMPTRLD_RAX       ".byte 0x0f, 0xc7, 0x30"
+#define ASM_VMX_VMREAD_RDX_RAX    ".byte 0x0f, 0x78, 0xd0"
+#define ASM_VMX_VMWRITE_RAX_RDX   ".byte 0x0f, 0x79, 0xd0"
+#define ASM_VMX_VMWRITE_RSP_RDX   ".byte 0x0f, 0x79, 0xd4"
+#define ASM_VMX_VMXOFF            ".byte 0x0f, 0x01, 0xc4"
+#define ASM_VMX_VMXON_RAX         ".byte 0xf3, 0x0f, 0xc7, 0x30"
+#define ASM_VMX_INVEPT		  ".byte 0x66, 0x0f, 0x38, 0x80, 0x08"
+#define ASM_VMX_INVVPID		  ".byte 0x66, 0x0f, 0x38, 0x81, 0x08"
+
+
+
 #endif
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 2e2da71..658e29e 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1296,7 +1296,7 @@
  * we don't need to preallocate the protection domains anymore.
  * For now we have to.
  */
-void prealloc_protection_domains(void)
+static void prealloc_protection_domains(void)
 {
 	struct pci_dev *dev = NULL;
 	struct dma_ops_domain *dma_dom;
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index c625800..fb85e8d 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -243,7 +243,7 @@
 }
 
 /* Function to enable the hardware */
-void __init iommu_enable(struct amd_iommu *iommu)
+static void __init iommu_enable(struct amd_iommu *iommu)
 {
 	printk(KERN_INFO "AMD IOMMU: Enabling IOMMU "
 	       "at %02x:%02x.%x cap 0x%hx\n",
@@ -256,7 +256,7 @@
 }
 
 /* Function to enable IOMMU event logging and event interrupts */
-void __init iommu_enable_event_logging(struct amd_iommu *iommu)
+static void __init iommu_enable_event_logging(struct amd_iommu *iommu)
 {
 	iommu_feature_enable(iommu, CONTROL_EVT_LOG_EN);
 	iommu_feature_enable(iommu, CONTROL_EVT_INT_EN);
diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c
index b5229af..d652515 100644
--- a/arch/x86/kernel/apic.c
+++ b/arch/x86/kernel/apic.c
@@ -98,8 +98,8 @@
 #ifdef HAVE_X2APIC
 int x2apic;
 /* x2apic enabled before OS handover */
-int x2apic_preenabled;
-int disable_x2apic;
+static int x2apic_preenabled;
+static int disable_x2apic;
 static __init int setup_nox2apic(char *str)
 {
 	disable_x2apic = 1;
@@ -119,8 +119,6 @@
 
 int first_system_vector = 0xfe;
 
-char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE};
-
 /*
  * Debug level, exported for io_apic.c
  */
@@ -142,7 +140,7 @@
 			    struct clock_event_device *evt);
 static void lapic_timer_setup(enum clock_event_mode mode,
 			      struct clock_event_device *evt);
-static void lapic_timer_broadcast(cpumask_t mask);
+static void lapic_timer_broadcast(const cpumask_t *mask);
 static void apic_pm_activate(void);
 
 /*
@@ -228,7 +226,7 @@
 	apic_write(APIC_ICR, low);
 }
 
-u64 xapic_icr_read(void)
+static u64 xapic_icr_read(void)
 {
 	u32 icr1, icr2;
 
@@ -268,7 +266,7 @@
 	wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low);
 }
 
-u64 x2apic_icr_read(void)
+static u64 x2apic_icr_read(void)
 {
 	unsigned long val;
 
@@ -455,7 +453,7 @@
 /*
  * Local APIC timer broadcast function
  */
-static void lapic_timer_broadcast(cpumask_t mask)
+static void lapic_timer_broadcast(const cpumask_t *mask)
 {
 #ifdef CONFIG_SMP
 	send_IPI_mask(mask, LOCAL_TIMER_VECTOR);
@@ -471,7 +469,7 @@
 	struct clock_event_device *levt = &__get_cpu_var(lapic_events);
 
 	memcpy(levt, &lapic_clockevent, sizeof(*levt));
-	levt->cpumask = cpumask_of_cpu(smp_processor_id());
+	levt->cpumask = cpumask_of(smp_processor_id());
 
 	clockevents_register_device(levt);
 }
@@ -1807,28 +1805,32 @@
 void __cpuinit generic_processor_info(int apicid, int version)
 {
 	int cpu;
-	cpumask_t tmp_map;
 
 	/*
 	 * Validate version
 	 */
 	if (version == 0x0) {
 		pr_warning("BIOS bug, APIC version is 0 for CPU#%d! "
-			"fixing up to 0x10. (tell your hw vendor)\n",
-			version);
+			   "fixing up to 0x10. (tell your hw vendor)\n",
+				version);
 		version = 0x10;
 	}
 	apic_version[apicid] = version;
 
-	if (num_processors >= NR_CPUS) {
-		pr_warning("WARNING: NR_CPUS limit of %i reached."
-			"  Processor ignored.\n", NR_CPUS);
+	if (num_processors >= nr_cpu_ids) {
+		int max = nr_cpu_ids;
+		int thiscpu = max + disabled_cpus;
+
+		pr_warning(
+			"ACPI: NR_CPUS/possible_cpus limit of %i reached."
+			"  Processor %d/0x%x ignored.\n", max, thiscpu, apicid);
+
+		disabled_cpus++;
 		return;
 	}
 
 	num_processors++;
-	cpus_complement(tmp_map, cpu_present_map);
-	cpu = first_cpu(tmp_map);
+	cpu = cpumask_next_zero(-1, cpu_present_mask);
 
 	physid_set(apicid, phys_cpu_present_map);
 	if (apicid == boot_cpu_physical_apicid) {
@@ -1878,8 +1880,8 @@
 	}
 #endif
 
-	cpu_set(cpu, cpu_possible_map);
-	cpu_set(cpu, cpu_present_map);
+	set_cpu_possible(cpu, true);
+	set_cpu_present(cpu, true);
 }
 
 #ifdef CONFIG_X86_64
@@ -2081,7 +2083,7 @@
 	bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid);
 	bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		/* are we being called early in kernel startup? */
 		if (bios_cpu_apicid) {
 			id = bios_cpu_apicid[i];
diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c
index 2a0a2a3..f638827 100644
--- a/arch/x86/kernel/bios_uv.c
+++ b/arch/x86/kernel/bios_uv.c
@@ -25,7 +25,7 @@
 #include <asm/uv/bios.h>
 #include <asm/uv/uv_hub.h>
 
-struct uv_systab uv_systab;
+static struct uv_systab uv_systab;
 
 s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5)
 {
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 68b5d86..c6ecda6 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -534,12 +534,29 @@
 	per_cpu(cpuid4_info, cpu) = NULL;
 }
 
+static void get_cpu_leaves(void *_retval)
+{
+	int j, *retval = _retval, cpu = smp_processor_id();
+
+	/* Do cpuid and store the results */
+	for (j = 0; j < num_cache_leaves; j++) {
+		struct _cpuid4_info *this_leaf;
+		this_leaf = CPUID4_INFO_IDX(cpu, j);
+		*retval = cpuid4_cache_lookup(j, this_leaf);
+		if (unlikely(*retval < 0)) {
+			int i;
+
+			for (i = 0; i < j; i++)
+				cache_remove_shared_cpu_map(cpu, i);
+			break;
+		}
+		cache_shared_cpu_map_setup(cpu, j);
+	}
+}
+
 static int __cpuinit detect_cache_attributes(unsigned int cpu)
 {
-	struct _cpuid4_info	*this_leaf;
-	unsigned long		j;
 	int			retval;
-	cpumask_t		oldmask;
 
 	if (num_cache_leaves == 0)
 		return -ENOENT;
@@ -549,27 +566,7 @@
 	if (per_cpu(cpuid4_info, cpu) == NULL)
 		return -ENOMEM;
 
-	oldmask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-	if (retval)
-		goto out;
-
-	/* Do cpuid and store the results */
-	for (j = 0; j < num_cache_leaves; j++) {
-		this_leaf = CPUID4_INFO_IDX(cpu, j);
-		retval = cpuid4_cache_lookup(j, this_leaf);
-		if (unlikely(retval < 0)) {
-			int i;
-
-			for (i = 0; i < j; i++)
-				cache_remove_shared_cpu_map(cpu, i);
-			break;
-		}
-		cache_shared_cpu_map_setup(cpu, j);
-	}
-	set_cpus_allowed_ptr(current, &oldmask);
-
-out:
+	smp_call_function_single(cpu, get_cpu_leaves, &retval, true);
 	if (retval) {
 		kfree(per_cpu(cpuid4_info, cpu));
 		per_cpu(cpuid4_info, cpu) = NULL;
@@ -626,8 +623,8 @@
 		cpumask_t *mask = &this_leaf->shared_cpu_map;
 
 		n = type?
-			cpulist_scnprintf(buf, len-2, *mask):
-			cpumask_scnprintf(buf, len-2, *mask);
+			cpulist_scnprintf(buf, len-2, mask) :
+			cpumask_scnprintf(buf, len-2, mask);
 		buf[n++] = '\n';
 		buf[n] = '\0';
 	}
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index 748c8f9..a5a5e05 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -83,34 +83,41 @@
  * CPU Initialization
  */
 
+struct thresh_restart {
+	struct threshold_block *b;
+	int reset;
+	u16 old_limit;
+};
+
 /* must be called with correct cpu affinity */
-static void threshold_restart_bank(struct threshold_block *b,
-				   int reset, u16 old_limit)
+static long threshold_restart_bank(void *_tr)
 {
+	struct thresh_restart *tr = _tr;
 	u32 mci_misc_hi, mci_misc_lo;
 
-	rdmsr(b->address, mci_misc_lo, mci_misc_hi);
+	rdmsr(tr->b->address, mci_misc_lo, mci_misc_hi);
 
-	if (b->threshold_limit < (mci_misc_hi & THRESHOLD_MAX))
-		reset = 1;	/* limit cannot be lower than err count */
+	if (tr->b->threshold_limit < (mci_misc_hi & THRESHOLD_MAX))
+		tr->reset = 1;	/* limit cannot be lower than err count */
 
-	if (reset) {		/* reset err count and overflow bit */
+	if (tr->reset) {		/* reset err count and overflow bit */
 		mci_misc_hi =
 		    (mci_misc_hi & ~(MASK_ERR_COUNT_HI | MASK_OVERFLOW_HI)) |
-		    (THRESHOLD_MAX - b->threshold_limit);
-	} else if (old_limit) {	/* change limit w/o reset */
+		    (THRESHOLD_MAX - tr->b->threshold_limit);
+	} else if (tr->old_limit) {	/* change limit w/o reset */
 		int new_count = (mci_misc_hi & THRESHOLD_MAX) +
-		    (old_limit - b->threshold_limit);
+		    (tr->old_limit - tr->b->threshold_limit);
 		mci_misc_hi = (mci_misc_hi & ~MASK_ERR_COUNT_HI) |
 		    (new_count & THRESHOLD_MAX);
 	}
 
-	b->interrupt_enable ?
+	tr->b->interrupt_enable ?
 	    (mci_misc_hi = (mci_misc_hi & ~MASK_INT_TYPE_HI) | INT_TYPE_APIC) :
 	    (mci_misc_hi &= ~MASK_INT_TYPE_HI);
 
 	mci_misc_hi |= MASK_COUNT_EN_HI;
-	wrmsr(b->address, mci_misc_lo, mci_misc_hi);
+	wrmsr(tr->b->address, mci_misc_lo, mci_misc_hi);
+	return 0;
 }
 
 /* cpu init entry point, called from mce.c with preempt off */
@@ -120,6 +127,7 @@
 	unsigned int cpu = smp_processor_id();
 	u8 lvt_off;
 	u32 low = 0, high = 0, address = 0;
+	struct thresh_restart tr;
 
 	for (bank = 0; bank < NR_BANKS; ++bank) {
 		for (block = 0; block < NR_BLOCKS; ++block) {
@@ -162,7 +170,10 @@
 			wrmsr(address, low, high);
 
 			threshold_defaults.address = address;
-			threshold_restart_bank(&threshold_defaults, 0, 0);
+			tr.b = &threshold_defaults;
+			tr.reset = 0;
+			tr.old_limit = 0;
+			threshold_restart_bank(&tr);
 		}
 	}
 }
@@ -251,20 +262,6 @@
 	ssize_t(*store) (struct threshold_block *, const char *, size_t count);
 };
 
-static void affinity_set(unsigned int cpu, cpumask_t *oldmask,
-					   cpumask_t *newmask)
-{
-	*oldmask = current->cpus_allowed;
-	cpus_clear(*newmask);
-	cpu_set(cpu, *newmask);
-	set_cpus_allowed_ptr(current, newmask);
-}
-
-static void affinity_restore(const cpumask_t *oldmask)
-{
-	set_cpus_allowed_ptr(current, oldmask);
-}
-
 #define SHOW_FIELDS(name)                                           \
 static ssize_t show_ ## name(struct threshold_block * b, char *buf) \
 {                                                                   \
@@ -277,15 +274,16 @@
 				      const char *buf, size_t count)
 {
 	char *end;
-	cpumask_t oldmask, newmask;
+	struct thresh_restart tr;
 	unsigned long new = simple_strtoul(buf, &end, 0);
 	if (end == buf)
 		return -EINVAL;
 	b->interrupt_enable = !!new;
 
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 0, 0);
-	affinity_restore(&oldmask);
+	tr.b = b;
+	tr.reset = 0;
+	tr.old_limit = 0;
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 
 	return end - buf;
 }
@@ -294,8 +292,7 @@
 				     const char *buf, size_t count)
 {
 	char *end;
-	cpumask_t oldmask, newmask;
-	u16 old;
+	struct thresh_restart tr;
 	unsigned long new = simple_strtoul(buf, &end, 0);
 	if (end == buf)
 		return -EINVAL;
@@ -303,34 +300,36 @@
 		new = THRESHOLD_MAX;
 	if (new < 1)
 		new = 1;
-	old = b->threshold_limit;
+	tr.old_limit = b->threshold_limit;
 	b->threshold_limit = new;
+	tr.b = b;
+	tr.reset = 0;
 
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 0, old);
-	affinity_restore(&oldmask);
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 
 	return end - buf;
 }
 
+static long local_error_count(void *_b)
+{
+	struct threshold_block *b = _b;
+	u32 low, high;
+
+	rdmsr(b->address, low, high);
+	return (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit);
+}
+
 static ssize_t show_error_count(struct threshold_block *b, char *buf)
 {
-	u32 high, low;
-	cpumask_t oldmask, newmask;
-	affinity_set(b->cpu, &oldmask, &newmask);
-	rdmsr(b->address, low, high);
-	affinity_restore(&oldmask);
-	return sprintf(buf, "%x\n",
-		       (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit));
+	return sprintf(buf, "%lx\n", work_on_cpu(b->cpu, local_error_count, b));
 }
 
 static ssize_t store_error_count(struct threshold_block *b,
 				 const char *buf, size_t count)
 {
-	cpumask_t oldmask, newmask;
-	affinity_set(b->cpu, &oldmask, &newmask);
-	threshold_restart_bank(b, 1, 0);
-	affinity_restore(&oldmask);
+	struct thresh_restart tr = { .b = b, .reset = 1, .old_limit = 0 };
+
+	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
 	return 1;
 }
 
@@ -463,12 +462,19 @@
 	return err;
 }
 
+static long local_allocate_threshold_blocks(void *_bank)
+{
+	unsigned int *bank = _bank;
+
+	return allocate_threshold_blocks(smp_processor_id(), *bank, 0,
+					 MSR_IA32_MC0_MISC + *bank * 4);
+}
+
 /* symlinks sibling shared banks to first core.  first core owns dir/files. */
 static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 {
 	int i, err = 0;
 	struct threshold_bank *b = NULL;
-	cpumask_t oldmask, newmask;
 	char name[32];
 
 	sprintf(name, "threshold_bank%i", bank);
@@ -519,11 +525,7 @@
 
 	per_cpu(threshold_banks, cpu)[bank] = b;
 
-	affinity_set(cpu, &oldmask, &newmask);
-	err = allocate_threshold_blocks(cpu, bank, 0,
-					MSR_IA32_MC0_MISC + bank * 4);
-	affinity_restore(&oldmask);
-
+	err = work_on_cpu(cpu, local_allocate_threshold_blocks, &bank);
 	if (err)
 		goto out_free;
 
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 4e8d77f..b59ddcc 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -14,14 +14,6 @@
 #include <asm/pat.h>
 #include "mtrr.h"
 
-struct mtrr_state {
-	struct mtrr_var_range var_ranges[MAX_VAR_RANGES];
-	mtrr_type fixed_ranges[NUM_FIXED_RANGES];
-	unsigned char enabled;
-	unsigned char have_fixed;
-	mtrr_type def_type;
-};
-
 struct fixed_range_block {
 	int base_msr; /* start address of an MTRR block */
 	int ranges;   /* number of MTRRs in this block  */
@@ -35,10 +27,12 @@
 };
 
 static unsigned long smp_changes_mask;
-static struct mtrr_state mtrr_state = {};
 static int mtrr_state_set;
 u64 mtrr_tom2;
 
+struct mtrr_state_type mtrr_state = {};
+EXPORT_SYMBOL_GPL(mtrr_state);
+
 #undef MODULE_PARAM_PREFIX
 #define MODULE_PARAM_PREFIX "mtrr."
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 1159e26..d259e5d 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -49,7 +49,7 @@
 
 u32 num_var_ranges = 0;
 
-unsigned int mtrr_usage_table[MAX_VAR_RANGES];
+unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 static DEFINE_MUTEX(mtrr_mutex);
 
 u64 size_or_mask, size_and_mask;
@@ -574,7 +574,7 @@
 	unsigned long	lsize;
 };
 
-static struct mtrr_value mtrr_state[MAX_VAR_RANGES];
+static struct mtrr_value mtrr_state[MTRR_MAX_VAR_RANGES];
 
 static int mtrr_save(struct sys_device * sysdev, pm_message_t state)
 {
@@ -824,16 +824,14 @@
 
 static int __init disable_mtrr_cleanup_setup(char *str)
 {
-	if (enable_mtrr_cleanup != -1)
-		enable_mtrr_cleanup = 0;
+	enable_mtrr_cleanup = 0;
 	return 0;
 }
 early_param("disable_mtrr_cleanup", disable_mtrr_cleanup_setup);
 
 static int __init enable_mtrr_cleanup_setup(char *str)
 {
-	if (enable_mtrr_cleanup != -1)
-		enable_mtrr_cleanup = 1;
+	enable_mtrr_cleanup = 1;
 	return 0;
 }
 early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup);
diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index 2dc4ec6..ffd6040 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -8,11 +8,6 @@
 #define MTRRcap_MSR     0x0fe
 #define MTRRdefType_MSR 0x2ff
 
-#define MTRRphysBase_MSR(reg) (0x200 + 2 * (reg))
-#define MTRRphysMask_MSR(reg) (0x200 + 2 * (reg) + 1)
-
-#define NUM_FIXED_RANGES 88
-#define MAX_VAR_RANGES 256
 #define MTRRfix64K_00000_MSR 0x250
 #define MTRRfix16K_80000_MSR 0x258
 #define MTRRfix16K_A0000_MSR 0x259
@@ -29,11 +24,7 @@
 #define MTRR_CHANGE_MASK_VARIABLE  0x02
 #define MTRR_CHANGE_MASK_DEFTYPE   0x04
 
-/* In the Intel processor's MTRR interface, the MTRR type is always held in
-   an 8 bit field: */
-typedef u8 mtrr_type;
-
-extern unsigned int mtrr_usage_table[MAX_VAR_RANGES];
+extern unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 
 struct mtrr_ops {
 	u32	vendor;
@@ -70,13 +61,6 @@
 	u32 ccr3;
 };
 
-struct mtrr_var_range {
-	u32 base_lo;
-	u32 base_hi;
-	u32 mask_lo;
-	u32 mask_hi;
-};
-
 void set_mtrr_done(struct set_mtrr_context *ctxt);
 void set_mtrr_cache_disable(struct set_mtrr_context *ctxt);
 void set_mtrr_prepare_save(struct set_mtrr_context *ctxt);
diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c
index 72cefd1..85d28d5 100644
--- a/arch/x86/kernel/cpuid.c
+++ b/arch/x86/kernel/cpuid.c
@@ -39,10 +39,10 @@
 #include <linux/device.h>
 #include <linux/cpu.h>
 #include <linux/notifier.h>
+#include <linux/uaccess.h>
 
 #include <asm/processor.h>
 #include <asm/msr.h>
-#include <asm/uaccess.h>
 #include <asm/system.h>
 
 static struct class *cpuid_class;
@@ -82,7 +82,7 @@
 }
 
 static ssize_t cpuid_read(struct file *file, char __user *buf,
-			  size_t count, loff_t * ppos)
+			  size_t count, loff_t *ppos)
 {
 	char __user *tmp = buf;
 	struct cpuid_regs cmd;
@@ -117,7 +117,7 @@
 	unsigned int cpu;
 	struct cpuinfo_x86 *c;
 	int ret = 0;
-	
+
 	lock_kernel();
 
 	cpu = iminor(file->f_path.dentry->d_inode);
diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index d84a852..c689d19 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -26,6 +26,7 @@
 #include <linux/kdebug.h>
 #include <asm/smp.h>
 #include <asm/reboot.h>
+#include <asm/virtext.h>
 
 #include <mach_ipi.h>
 
@@ -49,6 +50,15 @@
 #endif
 	crash_save_cpu(regs, cpu);
 
+	/* Disable VMX or SVM if needed.
+	 *
+	 * We need to disable virtualization on all CPUs.
+	 * Having VMX or SVM enabled on any CPU may break rebooting
+	 * after the kdump kernel has finished its task.
+	 */
+	cpu_emergency_vmxoff();
+	cpu_emergency_svm_disable();
+
 	disable_local_APIC();
 }
 
@@ -80,6 +90,14 @@
 	local_irq_disable();
 
 	kdump_nmi_shootdown_cpus();
+
+	/* Booting kdump kernel with VMX or SVM enabled won't work,
+	 * because (among other limitations) we can't disable paging
+	 * with the virt flags.
+	 */
+	cpu_emergency_vmxoff();
+	cpu_emergency_svm_disable();
+
 	lapic_shutdown();
 #if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index 23b138e..504ad19 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -886,7 +886,7 @@
 	va_list ap;
 
 	va_start(ap, fmt);
-	n = vscnprintf(buf, 512, fmt, ap);
+	n = vscnprintf(buf, sizeof(buf), fmt, ap);
 	early_console->write(early_console, buf, n);
 	va_end(ap);
 }
diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c
index c026279..3418548 100644
--- a/arch/x86/kernel/genapic_flat_64.c
+++ b/arch/x86/kernel/genapic_flat_64.c
@@ -30,12 +30,12 @@
 	return 1;
 }
 
-static cpumask_t flat_target_cpus(void)
+static const struct cpumask *flat_target_cpus(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 
-static cpumask_t flat_vector_allocation_domain(int cpu)
+static void flat_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -45,8 +45,8 @@
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	cpumask_t domain = { { [0] = APIC_ALL_CPUS, } };
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
 }
 
 /*
@@ -69,9 +69,8 @@
 	apic_write(APIC_LDR, val);
 }
 
-static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
+static inline void _flat_send_IPI_mask(unsigned long mask, int vector)
 {
-	unsigned long mask = cpus_addr(cpumask)[0];
 	unsigned long flags;
 
 	local_irq_save(flags);
@@ -79,20 +78,41 @@
 	local_irq_restore(flags);
 }
 
+static void flat_send_IPI_mask(const struct cpumask *cpumask, int vector)
+{
+	unsigned long mask = cpumask_bits(cpumask)[0];
+
+	_flat_send_IPI_mask(mask, vector);
+}
+
+static void flat_send_IPI_mask_allbutself(const struct cpumask *cpumask,
+					  int vector)
+{
+	unsigned long mask = cpumask_bits(cpumask)[0];
+	int cpu = smp_processor_id();
+
+	if (cpu < BITS_PER_LONG)
+		clear_bit(cpu, &mask);
+	_flat_send_IPI_mask(mask, vector);
+}
+
 static void flat_send_IPI_allbutself(int vector)
 {
+	int cpu = smp_processor_id();
 #ifdef	CONFIG_HOTPLUG_CPU
 	int hotplug = 1;
 #else
 	int hotplug = 0;
 #endif
 	if (hotplug || vector == NMI_VECTOR) {
-		cpumask_t allbutme = cpu_online_map;
+		if (!cpumask_equal(cpu_online_mask, cpumask_of(cpu))) {
+			unsigned long mask = cpumask_bits(cpu_online_mask)[0];
 
-		cpu_clear(smp_processor_id(), allbutme);
+			if (cpu < BITS_PER_LONG)
+				clear_bit(cpu, &mask);
 
-		if (!cpus_empty(allbutme))
-			flat_send_IPI_mask(allbutme, vector);
+			_flat_send_IPI_mask(mask, vector);
+		}
 	} else if (num_online_cpus() > 1) {
 		__send_IPI_shortcut(APIC_DEST_ALLBUT, vector,APIC_DEST_LOGICAL);
 	}
@@ -101,7 +121,7 @@
 static void flat_send_IPI_all(int vector)
 {
 	if (vector == NMI_VECTOR)
-		flat_send_IPI_mask(cpu_online_map, vector);
+		flat_send_IPI_mask(cpu_online_mask, vector);
 	else
 		__send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
 }
@@ -135,9 +155,18 @@
 	return physid_isset(read_xapic_id(), phys_cpu_present_map);
 }
 
-static unsigned int flat_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int flat_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
-	return cpus_addr(cpumask)[0] & APIC_ALL_CPUS;
+	return cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
+}
+
+static unsigned int flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						const struct cpumask *andmask)
+{
+	unsigned long mask1 = cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
+	unsigned long mask2 = cpumask_bits(andmask)[0] & APIC_ALL_CPUS;
+
+	return mask1 & mask2;
 }
 
 static unsigned int phys_pkg_id(int index_msb)
@@ -157,8 +186,10 @@
 	.send_IPI_all = flat_send_IPI_all,
 	.send_IPI_allbutself = flat_send_IPI_allbutself,
 	.send_IPI_mask = flat_send_IPI_mask,
+	.send_IPI_mask_allbutself = flat_send_IPI_mask_allbutself,
 	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = flat_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = flat_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
@@ -188,35 +219,39 @@
 	return 0;
 }
 
-static cpumask_t physflat_target_cpus(void)
+static const struct cpumask *physflat_target_cpus(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 
-static cpumask_t physflat_vector_allocation_domain(int cpu)
+static void physflat_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	return cpumask_of_cpu(cpu);
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
-static void physflat_send_IPI_mask(cpumask_t cpumask, int vector)
+static void physflat_send_IPI_mask(const struct cpumask *cpumask, int vector)
 {
 	send_IPI_mask_sequence(cpumask, vector);
 }
 
+static void physflat_send_IPI_mask_allbutself(const struct cpumask *cpumask,
+					      int vector)
+{
+	send_IPI_mask_allbutself(cpumask, vector);
+}
+
 static void physflat_send_IPI_allbutself(int vector)
 {
-	cpumask_t allbutme = cpu_online_map;
-
-	cpu_clear(smp_processor_id(), allbutme);
-	physflat_send_IPI_mask(allbutme, vector);
+	send_IPI_mask_allbutself(cpu_online_mask, vector);
 }
 
 static void physflat_send_IPI_all(int vector)
 {
-	physflat_send_IPI_mask(cpu_online_map, vector);
+	physflat_send_IPI_mask(cpu_online_mask, vector);
 }
 
-static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int physflat_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -224,13 +259,31 @@
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = cpumask_first(cpumask);
 	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int
+physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+				const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 struct genapic apic_physflat =  {
 	.name = "physical flat",
 	.acpi_madt_oem_check = physflat_acpi_madt_oem_check,
@@ -243,8 +296,10 @@
 	.send_IPI_all = physflat_send_IPI_all,
 	.send_IPI_allbutself = physflat_send_IPI_allbutself,
 	.send_IPI_mask = physflat_send_IPI_mask,
+	.send_IPI_mask_allbutself = physflat_send_IPI_mask_allbutself,
 	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = physflat_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = physflat_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c
index f6a2c8e..6ce497c 100644
--- a/arch/x86/kernel/genx2apic_cluster.c
+++ b/arch/x86/kernel/genx2apic_cluster.c
@@ -22,19 +22,18 @@
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t x2apic_target_cpus(void)
+static const struct cpumask *x2apic_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
 /*
  * for now each logical cpu is in its own vector allocation domain.
  */
-static cpumask_t x2apic_vector_allocation_domain(int cpu)
+static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
@@ -56,32 +55,53 @@
  * at once. We have 16 cpu's in a cluster. This will minimize IPI register
  * writes.
  */
-static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned long query_cpu;
 
 	local_irq_save(flags);
-	for_each_cpu_mask(query_cpu, mask) {
-		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_logical_apicid, query_cpu),
-				       vector, APIC_DEST_LOGICAL);
-	}
+	for_each_cpu(query_cpu, mask)
+		__x2apic_send_IPI_dest(
+			per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+			vector, APIC_DEST_LOGICAL);
+	local_irq_restore(flags);
+}
+
+static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask,
+					    int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+				vector, APIC_DEST_LOGICAL);
 	local_irq_restore(flags);
 }
 
 static void x2apic_send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	cpu_clear(smp_processor_id(), mask);
-
-	if (!cpus_empty(mask))
-		x2apic_send_IPI_mask(mask, vector);
+	local_irq_save(flags);
+	for_each_online_cpu(query_cpu)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+				vector, APIC_DEST_LOGICAL);
+	local_irq_restore(flags);
 }
 
 static void x2apic_send_IPI_all(int vector)
 {
-	x2apic_send_IPI_mask(cpu_online_map, vector);
+	x2apic_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int x2apic_apic_id_registered(void)
@@ -89,21 +109,38 @@
 	return 1;
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
 	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	cpu = cpumask_first(cpumask);
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_logical_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_logical_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -150,8 +187,10 @@
 	.send_IPI_all = x2apic_send_IPI_all,
 	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
 	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself,
 	.send_IPI_self = x2apic_send_IPI_self,
 	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c
index d042211..21bcc0e 100644
--- a/arch/x86/kernel/genx2apic_phys.c
+++ b/arch/x86/kernel/genx2apic_phys.c
@@ -29,16 +29,15 @@
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t x2apic_target_cpus(void)
+static const struct cpumask *x2apic_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
-static cpumask_t x2apic_vector_allocation_domain(int cpu)
+static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
@@ -54,32 +53,54 @@
 	x2apic_icr_write(cfg, apicid);
 }
 
-static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned long query_cpu;
 
 	local_irq_save(flags);
-	for_each_cpu_mask(query_cpu, mask) {
+	for_each_cpu(query_cpu, mask) {
 		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu),
 				       vector, APIC_DEST_PHYSICAL);
 	}
 	local_irq_restore(flags);
 }
 
+static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask,
+					    int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask) {
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_apicid, query_cpu),
+				vector, APIC_DEST_PHYSICAL);
+	}
+	local_irq_restore(flags);
+}
+
 static void x2apic_send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
+	unsigned long flags;
+	unsigned long query_cpu;
+	unsigned long this_cpu = smp_processor_id();
 
-	cpu_clear(smp_processor_id(), mask);
-
-	if (!cpus_empty(mask))
-		x2apic_send_IPI_mask(mask, vector);
+	local_irq_save(flags);
+	for_each_online_cpu(query_cpu)
+		if (query_cpu != this_cpu)
+			__x2apic_send_IPI_dest(
+				per_cpu(x86_cpu_to_apicid, query_cpu),
+				vector, APIC_DEST_PHYSICAL);
+	local_irq_restore(flags);
 }
 
 static void x2apic_send_IPI_all(int vector)
 {
-	x2apic_send_IPI_mask(cpu_online_map, vector);
+	x2apic_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int x2apic_apic_id_registered(void)
@@ -87,7 +108,7 @@
 	return 1;
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -95,13 +116,30 @@
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
-	if ((unsigned)cpu < NR_CPUS)
+	cpu = cpumask_first(cpumask);
+	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+						  const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -123,12 +161,12 @@
 	return current_cpu_data.initial_apicid >> index_msb;
 }
 
-void x2apic_send_IPI_self(int vector)
+static void x2apic_send_IPI_self(int vector)
 {
 	apic_write(APIC_SELF_IPI, vector);
 }
 
-void init_x2apic_ldr(void)
+static void init_x2apic_ldr(void)
 {
 	return;
 }
@@ -145,8 +183,10 @@
 	.send_IPI_all = x2apic_send_IPI_all,
 	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
 	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself,
 	.send_IPI_self = x2apic_send_IPI_self,
 	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
index dece172..b193e08 100644
--- a/arch/x86/kernel/genx2apic_uv_x.c
+++ b/arch/x86/kernel/genx2apic_uv_x.c
@@ -79,16 +79,15 @@
 
 /* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
 
-static cpumask_t uv_target_cpus(void)
+static const struct cpumask *uv_target_cpus(void)
 {
-	return cpumask_of_cpu(0);
+	return cpumask_of(0);
 }
 
-static cpumask_t uv_vector_allocation_domain(int cpu)
+static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask)
 {
-	cpumask_t domain = CPU_MASK_NONE;
-	cpu_set(cpu, domain);
-	return domain;
+	cpumask_clear(retmask);
+	cpumask_set_cpu(cpu, retmask);
 }
 
 int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
@@ -127,28 +126,37 @@
 	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
 }
 
-static void uv_send_IPI_mask(cpumask_t mask, int vector)
+static void uv_send_IPI_mask(const struct cpumask *mask, int vector)
 {
 	unsigned int cpu;
 
-	for_each_possible_cpu(cpu)
-		if (cpu_isset(cpu, mask))
+	for_each_cpu(cpu, mask)
+		uv_send_IPI_one(cpu, vector);
+}
+
+static void uv_send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+	unsigned int cpu;
+	unsigned int this_cpu = smp_processor_id();
+
+	for_each_cpu(cpu, mask)
+		if (cpu != this_cpu)
 			uv_send_IPI_one(cpu, vector);
 }
 
 static void uv_send_IPI_allbutself(int vector)
 {
-	cpumask_t mask = cpu_online_map;
+	unsigned int cpu;
+	unsigned int this_cpu = smp_processor_id();
 
-	cpu_clear(smp_processor_id(), mask);
-
-	if (!cpus_empty(mask))
-		uv_send_IPI_mask(mask, vector);
+	for_each_online_cpu(cpu)
+		if (cpu != this_cpu)
+			uv_send_IPI_one(cpu, vector);
 }
 
 static void uv_send_IPI_all(int vector)
 {
-	uv_send_IPI_mask(cpu_online_map, vector);
+	uv_send_IPI_mask(cpu_online_mask, vector);
 }
 
 static int uv_apic_id_registered(void)
@@ -160,7 +168,7 @@
 {
 }
 
-static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
+static unsigned int uv_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
 	int cpu;
 
@@ -168,13 +176,30 @@
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
 	 * May as well be the first.
 	 */
-	cpu = first_cpu(cpumask);
+	cpu = cpumask_first(cpumask);
 	if ((unsigned)cpu < nr_cpu_ids)
 		return per_cpu(x86_cpu_to_apicid, cpu);
 	else
 		return BAD_APICID;
 }
 
+static unsigned int uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+					      const struct cpumask *andmask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	for_each_cpu_and(cpu, cpumask, andmask)
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	if (cpu < nr_cpu_ids)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	return BAD_APICID;
+}
+
 static unsigned int get_apic_id(unsigned long x)
 {
 	unsigned int id;
@@ -222,8 +247,10 @@
 	.send_IPI_all = uv_send_IPI_all,
 	.send_IPI_allbutself = uv_send_IPI_allbutself,
 	.send_IPI_mask = uv_send_IPI_mask,
+	.send_IPI_mask_allbutself = uv_send_IPI_mask_allbutself,
 	.send_IPI_self = uv_send_IPI_self,
 	.cpu_mask_to_apicid = uv_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and = uv_cpu_mask_to_apicid_and,
 	.phys_pkg_id = phys_pkg_id,
 	.get_apic_id = get_apic_id,
 	.set_apic_id = set_apic_id,
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 388e05a..b9a4d8c 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -27,7 +27,7 @@
 #include <asm/trampoline.h>
 
 /* boot cpu pda */
-static struct x8664_pda _boot_cpu_pda __read_mostly;
+static struct x8664_pda _boot_cpu_pda;
 
 #ifdef CONFIG_SMP
 /*
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 845ea09..cd759ad 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -248,7 +248,7 @@
 	 * Start hpet with the boot cpu mask and make it
 	 * global after the IO_APIC has been initialized.
 	 */
-	hpet_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	hpet_clockevent.cpumask = cpumask_of(smp_processor_id());
 	clockevents_register_device(&hpet_clockevent);
 	global_clock_event = &hpet_clockevent;
 	printk(KERN_DEBUG "hpet clockevent registered\n");
@@ -303,7 +303,7 @@
 			struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
 			hpet_setup_msi_irq(hdev->irq);
 			disable_irq(hdev->irq);
-			irq_set_affinity(hdev->irq, cpumask_of_cpu(hdev->cpu));
+			irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu));
 			enable_irq(hdev->irq);
 		}
 		break;
@@ -451,7 +451,7 @@
 		return -1;
 
 	disable_irq(dev->irq);
-	irq_set_affinity(dev->irq, cpumask_of_cpu(dev->cpu));
+	irq_set_affinity(dev->irq, cpumask_of(dev->cpu));
 	enable_irq(dev->irq);
 
 	printk(KERN_DEBUG "hpet: %s irq %d for MSI\n",
@@ -502,7 +502,7 @@
 	/* 5 usec minimum reprogramming delta. */
 	evt->min_delta_ns = 5000;
 
-	evt->cpumask = cpumask_of_cpu(hdev->cpu);
+	evt->cpumask = cpumask_of(hdev->cpu);
 	clockevents_register_device(evt);
 }
 
diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c
index c1b5e3e..10f92fb 100644
--- a/arch/x86/kernel/i8253.c
+++ b/arch/x86/kernel/i8253.c
@@ -114,7 +114,7 @@
 	 * Start pit with the boot cpu mask and make it global after the
 	 * IO_APIC has been initialized.
 	 */
-	pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	pit_clockevent.cpumask = cpumask_of(smp_processor_id());
 	pit_clockevent.mult = div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC,
 				     pit_clockevent.shift);
 	pit_clockevent.max_delta_ns =
diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index d399180..df3bf26 100644
--- a/arch/x86/kernel/init_task.c
+++ b/arch/x86/kernel/init_task.c
@@ -10,7 +10,6 @@
 #include <asm/pgtable.h>
 #include <asm/desc.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index f6ea94b..6991172 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -136,8 +136,8 @@
 
 struct irq_cfg {
 	struct irq_pin_list *irq_2_pin;
-	cpumask_t domain;
-	cpumask_t old_domain;
+	cpumask_var_t domain;
+	cpumask_var_t old_domain;
 	unsigned move_cleanup_count;
 	u8 vector;
 	u8 move_in_progress : 1;
@@ -152,25 +152,25 @@
 #else
 static struct irq_cfg irq_cfgx[NR_IRQS] = {
 #endif
-	[0]  = { .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR,  },
-	[1]  = { .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR,  },
-	[2]  = { .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR,  },
-	[3]  = { .domain = CPU_MASK_ALL, .vector = IRQ3_VECTOR,  },
-	[4]  = { .domain = CPU_MASK_ALL, .vector = IRQ4_VECTOR,  },
-	[5]  = { .domain = CPU_MASK_ALL, .vector = IRQ5_VECTOR,  },
-	[6]  = { .domain = CPU_MASK_ALL, .vector = IRQ6_VECTOR,  },
-	[7]  = { .domain = CPU_MASK_ALL, .vector = IRQ7_VECTOR,  },
-	[8]  = { .domain = CPU_MASK_ALL, .vector = IRQ8_VECTOR,  },
-	[9]  = { .domain = CPU_MASK_ALL, .vector = IRQ9_VECTOR,  },
-	[10] = { .domain = CPU_MASK_ALL, .vector = IRQ10_VECTOR, },
-	[11] = { .domain = CPU_MASK_ALL, .vector = IRQ11_VECTOR, },
-	[12] = { .domain = CPU_MASK_ALL, .vector = IRQ12_VECTOR, },
-	[13] = { .domain = CPU_MASK_ALL, .vector = IRQ13_VECTOR, },
-	[14] = { .domain = CPU_MASK_ALL, .vector = IRQ14_VECTOR, },
-	[15] = { .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, },
+	[0]  = { .vector = IRQ0_VECTOR,  },
+	[1]  = { .vector = IRQ1_VECTOR,  },
+	[2]  = { .vector = IRQ2_VECTOR,  },
+	[3]  = { .vector = IRQ3_VECTOR,  },
+	[4]  = { .vector = IRQ4_VECTOR,  },
+	[5]  = { .vector = IRQ5_VECTOR,  },
+	[6]  = { .vector = IRQ6_VECTOR,  },
+	[7]  = { .vector = IRQ7_VECTOR,  },
+	[8]  = { .vector = IRQ8_VECTOR,  },
+	[9]  = { .vector = IRQ9_VECTOR,  },
+	[10] = { .vector = IRQ10_VECTOR, },
+	[11] = { .vector = IRQ11_VECTOR, },
+	[12] = { .vector = IRQ12_VECTOR, },
+	[13] = { .vector = IRQ13_VECTOR, },
+	[14] = { .vector = IRQ14_VECTOR, },
+	[15] = { .vector = IRQ15_VECTOR, },
 };
 
-void __init arch_early_irq_init(void)
+int __init arch_early_irq_init(void)
 {
 	struct irq_cfg *cfg;
 	struct irq_desc *desc;
@@ -183,7 +183,13 @@
 	for (i = 0; i < count; i++) {
 		desc = irq_to_desc(i);
 		desc->chip_data = &cfg[i];
+		alloc_bootmem_cpumask_var(&cfg[i].domain);
+		alloc_bootmem_cpumask_var(&cfg[i].old_domain);
+		if (i < NR_IRQS_LEGACY)
+			cpumask_setall(cfg[i].domain);
 	}
+
+	return 0;
 }
 
 #ifdef CONFIG_SPARSE_IRQ
@@ -207,12 +213,26 @@
 	node = cpu_to_node(cpu);
 
 	cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node);
+	if (cfg) {
+		/* FIXME: needs alloc_cpumask_var_node() */
+		if (!alloc_cpumask_var(&cfg->domain, GFP_ATOMIC)) {
+			kfree(cfg);
+			cfg = NULL;
+		} else if (!alloc_cpumask_var(&cfg->old_domain, GFP_ATOMIC)) {
+			free_cpumask_var(cfg->domain);
+			kfree(cfg);
+			cfg = NULL;
+		} else {
+			cpumask_clear(cfg->domain);
+			cpumask_clear(cfg->old_domain);
+		}
+	}
 	printk(KERN_DEBUG "  alloc irq_cfg on cpu %d node %d\n", cpu, node);
 
 	return cfg;
 }
 
-void arch_init_chip_data(struct irq_desc *desc, int cpu)
+int arch_init_chip_data(struct irq_desc *desc, int cpu)
 {
 	struct irq_cfg *cfg;
 
@@ -224,6 +244,8 @@
 			BUG_ON(1);
 		}
 	}
+
+	return 0;
 }
 
 #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
@@ -329,13 +351,14 @@
 	}
 }
 
-static void set_extra_move_desc(struct irq_desc *desc, cpumask_t mask)
+static void
+set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg = desc->chip_data;
 
 	if (!cfg->move_in_progress) {
 		/* it means that domain is not changed */
-		if (!cpus_intersects(desc->affinity, mask))
+		if (!cpumask_intersects(&desc->affinity, mask))
 			cfg->move_desc_pending = 1;
 	}
 }
@@ -350,7 +373,8 @@
 #endif
 
 #ifndef CONFIG_NUMA_MIGRATE_IRQ_DESC
-static inline void set_extra_move_desc(struct irq_desc *desc, cpumask_t mask)
+static inline void
+set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 }
 #endif
@@ -481,6 +505,26 @@
 }
 
 #ifdef CONFIG_SMP
+static void send_cleanup_vector(struct irq_cfg *cfg)
+{
+	cpumask_var_t cleanup_mask;
+
+	if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) {
+		unsigned int i;
+		cfg->move_cleanup_count = 0;
+		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
+			cfg->move_cleanup_count++;
+		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
+			send_IPI_mask(cpumask_of(i), IRQ_MOVE_CLEANUP_VECTOR);
+	} else {
+		cpumask_and(cleanup_mask, cfg->old_domain, cpu_online_mask);
+		cfg->move_cleanup_count = cpumask_weight(cleanup_mask);
+		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
+		free_cpumask_var(cleanup_mask);
+	}
+	cfg->move_in_progress = 0;
+}
+
 static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg)
 {
 	int apic, pin;
@@ -516,41 +560,55 @@
 	}
 }
 
-static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask);
+static int
+assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask);
 
-static void set_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mask)
+/*
+ * Either sets desc->affinity to a valid value, and returns cpu_mask_to_apicid
+ * of that, or returns BAD_APICID and leaves desc->affinity untouched.
+ */
+static unsigned int
+set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
-	unsigned long flags;
-	unsigned int dest;
-	cpumask_t tmp;
 	unsigned int irq;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
-		return;
+	if (!cpumask_intersects(mask, cpu_online_mask))
+		return BAD_APICID;
 
 	irq = desc->irq;
 	cfg = desc->chip_data;
 	if (assign_irq_vector(irq, cfg, mask))
-		return;
+		return BAD_APICID;
 
+	cpumask_and(&desc->affinity, cfg->domain, mask);
 	set_extra_move_desc(desc, mask);
+	return cpu_mask_to_apicid_and(&desc->affinity, cpu_online_mask);
+}
 
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
-	/*
-	 * Only the high 8 bits are valid.
-	 */
-	dest = SET_APIC_LOGICAL_ID(dest);
+static void
+set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
+{
+	struct irq_cfg *cfg;
+	unsigned long flags;
+	unsigned int dest;
+	unsigned int irq;
+
+	irq = desc->irq;
+	cfg = desc->chip_data;
 
 	spin_lock_irqsave(&ioapic_lock, flags);
-	__target_IO_APIC_irq(irq, dest, cfg);
-	desc->affinity = mask;
+	dest = set_desc_affinity(desc, mask);
+	if (dest != BAD_APICID) {
+		/* Only the high 8 bits are valid. */
+		dest = SET_APIC_LOGICAL_ID(dest);
+		__target_IO_APIC_irq(irq, dest, cfg);
+	}
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+static void
+set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc;
 
@@ -648,7 +706,7 @@
 }
 
 #ifdef CONFIG_X86_64
-void io_apic_sync(struct irq_pin_list *entry)
+static void io_apic_sync(struct irq_pin_list *entry)
 {
 	/*
 	 * Synchronize the IO-APIC and the CPU by doing
@@ -1218,7 +1276,8 @@
 	spin_unlock(&vector_lock);
 }
 
-static int __assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
+static int
+__assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 {
 	/*
 	 * NOTE! The local APIC isn't very good at handling
@@ -1233,49 +1292,49 @@
 	 */
 	static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0;
 	unsigned int old_vector;
-	int cpu;
+	int cpu, err;
+	cpumask_var_t tmp_mask;
 
 	if ((cfg->move_in_progress) || cfg->move_cleanup_count)
 		return -EBUSY;
 
-	/* Only try and allocate irqs on cpus that are present */
-	cpus_and(mask, mask, cpu_online_map);
+	if (!alloc_cpumask_var(&tmp_mask, GFP_ATOMIC))
+		return -ENOMEM;
 
 	old_vector = cfg->vector;
 	if (old_vector) {
-		cpumask_t tmp;
-		cpus_and(tmp, cfg->domain, mask);
-		if (!cpus_empty(tmp))
+		cpumask_and(tmp_mask, mask, cpu_online_mask);
+		cpumask_and(tmp_mask, cfg->domain, tmp_mask);
+		if (!cpumask_empty(tmp_mask)) {
+			free_cpumask_var(tmp_mask);
 			return 0;
+		}
 	}
 
-	for_each_cpu_mask_nr(cpu, mask) {
-		cpumask_t domain, new_mask;
+	/* Only try and allocate irqs on cpus that are present */
+	err = -ENOSPC;
+	for_each_cpu_and(cpu, mask, cpu_online_mask) {
 		int new_cpu;
 		int vector, offset;
 
-		domain = vector_allocation_domain(cpu);
-		cpus_and(new_mask, domain, cpu_online_map);
+		vector_allocation_domain(cpu, tmp_mask);
 
 		vector = current_vector;
 		offset = current_offset;
 next:
 		vector += 8;
 		if (vector >= first_system_vector) {
-			/* If we run out of vectors on large boxen, must share them. */
+			/* If out of vectors on large boxen, must share them. */
 			offset = (offset + 1) % 8;
 			vector = FIRST_DEVICE_VECTOR + offset;
 		}
 		if (unlikely(current_vector == vector))
 			continue;
-#ifdef CONFIG_X86_64
-		if (vector == IA32_SYSCALL_VECTOR)
+
+		if (test_bit(vector, used_vectors))
 			goto next;
-#else
-		if (vector == SYSCALL_VECTOR)
-			goto next;
-#endif
-		for_each_cpu_mask_nr(new_cpu, new_mask)
+
+		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask)
 			if (per_cpu(vector_irq, new_cpu)[vector] != -1)
 				goto next;
 		/* Found one! */
@@ -1283,18 +1342,21 @@
 		current_offset = offset;
 		if (old_vector) {
 			cfg->move_in_progress = 1;
-			cfg->old_domain = cfg->domain;
+			cpumask_copy(cfg->old_domain, cfg->domain);
 		}
-		for_each_cpu_mask_nr(new_cpu, new_mask)
+		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask)
 			per_cpu(vector_irq, new_cpu)[vector] = irq;
 		cfg->vector = vector;
-		cfg->domain = domain;
-		return 0;
+		cpumask_copy(cfg->domain, tmp_mask);
+		err = 0;
+		break;
 	}
-	return -ENOSPC;
+	free_cpumask_var(tmp_mask);
+	return err;
 }
 
-static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
+static int
+assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 {
 	int err;
 	unsigned long flags;
@@ -1307,23 +1369,20 @@
 
 static void __clear_irq_vector(int irq, struct irq_cfg *cfg)
 {
-	cpumask_t mask;
 	int cpu, vector;
 
 	BUG_ON(!cfg->vector);
 
 	vector = cfg->vector;
-	cpus_and(mask, cfg->domain, cpu_online_map);
-	for_each_cpu_mask_nr(cpu, mask)
+	for_each_cpu_and(cpu, cfg->domain, cpu_online_mask)
 		per_cpu(vector_irq, cpu)[vector] = -1;
 
 	cfg->vector = 0;
-	cpus_clear(cfg->domain);
+	cpumask_clear(cfg->domain);
 
 	if (likely(!cfg->move_in_progress))
 		return;
-	cpus_and(mask, cfg->old_domain, cpu_online_map);
-	for_each_cpu_mask_nr(cpu, mask) {
+	for_each_cpu_and(cpu, cfg->old_domain, cpu_online_mask) {
 		for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS;
 								vector++) {
 			if (per_cpu(vector_irq, cpu)[vector] != irq)
@@ -1345,10 +1404,8 @@
 
 	/* Mark the inuse vectors */
 	for_each_irq_desc(irq, desc) {
-		if (!desc)
-			continue;
 		cfg = desc->chip_data;
-		if (!cpu_isset(cpu, cfg->domain))
+		if (!cpumask_test_cpu(cpu, cfg->domain))
 			continue;
 		vector = cfg->vector;
 		per_cpu(vector_irq, cpu)[vector] = irq;
@@ -1360,7 +1417,7 @@
 			continue;
 
 		cfg = irq_cfg(irq);
-		if (!cpu_isset(cpu, cfg->domain))
+		if (!cpumask_test_cpu(cpu, cfg->domain))
 			per_cpu(vector_irq, cpu)[vector] = -1;
 	}
 }
@@ -1496,18 +1553,17 @@
 {
 	struct irq_cfg *cfg;
 	struct IO_APIC_route_entry entry;
-	cpumask_t mask;
+	unsigned int dest;
 
 	if (!IO_APIC_IRQ(irq))
 		return;
 
 	cfg = desc->chip_data;
 
-	mask = TARGET_CPUS;
-	if (assign_irq_vector(irq, cfg, mask))
+	if (assign_irq_vector(irq, cfg, TARGET_CPUS))
 		return;
 
-	cpus_and(mask, cfg->domain, mask);
+	dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 	apic_printk(APIC_VERBOSE,KERN_DEBUG
 		    "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> "
@@ -1517,8 +1573,7 @@
 
 
 	if (setup_ioapic_entry(mp_ioapics[apic].mp_apicid, irq, &entry,
-			       cpu_mask_to_apicid(mask), trigger, polarity,
-			       cfg->vector)) {
+			       dest, trigger, polarity, cfg->vector)) {
 		printk("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
 		       mp_ioapics[apic].mp_apicid, pin);
 		__clear_irq_vector(irq, cfg);
@@ -1730,8 +1785,6 @@
 	for_each_irq_desc(irq, desc) {
 		struct irq_pin_list *entry;
 
-		if (!desc)
-			continue;
 		cfg = desc->chip_data;
 		entry = cfg->irq_2_pin;
 		if (!entry)
@@ -2240,7 +2293,7 @@
 	unsigned long flags;
 
 	spin_lock_irqsave(&vector_lock, flags);
-	send_IPI_mask(cpumask_of_cpu(first_cpu(cfg->domain)), cfg->vector);
+	send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector);
 	spin_unlock_irqrestore(&vector_lock, flags);
 
 	return 1;
@@ -2289,18 +2342,17 @@
  * as simple as edge triggered migration and we can do the irq migration
  * with a simple atomic update to IO-APIC RTE.
  */
-static void migrate_ioapic_irq_desc(struct irq_desc *desc, cpumask_t mask)
+static void
+migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
-	cpumask_t tmp, cleanup_mask;
 	struct irte irte;
 	int modify_ioapic_rte;
 	unsigned int dest;
 	unsigned long flags;
 	unsigned int irq;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
 	irq = desc->irq;
@@ -2313,8 +2365,7 @@
 
 	set_extra_move_desc(desc, mask);
 
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
+	dest = cpu_mask_to_apicid_and(cfg->domain, mask);
 
 	modify_ioapic_rte = desc->status & IRQ_LEVEL;
 	if (modify_ioapic_rte) {
@@ -2331,14 +2382,10 @@
 	 */
 	modify_irte(irq, &irte);
 
-	if (cfg->move_in_progress) {
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
+	if (cfg->move_in_progress)
+		send_cleanup_vector(cfg);
 
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 
 static int migrate_irq_remapped_level_desc(struct irq_desc *desc)
@@ -2360,11 +2407,11 @@
 	}
 
 	/* everthing is clear. we have right of way */
-	migrate_ioapic_irq_desc(desc, desc->pending_mask);
+	migrate_ioapic_irq_desc(desc, &desc->pending_mask);
 
 	ret = 0;
 	desc->status &= ~IRQ_MOVE_PENDING;
-	cpus_clear(desc->pending_mask);
+	cpumask_clear(&desc->pending_mask);
 
 unmask:
 	unmask_IO_APIC_irq_desc(desc);
@@ -2378,9 +2425,6 @@
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		if (!desc)
-			continue;
-
 		if (desc->status & IRQ_MOVE_PENDING) {
 			unsigned long flags;
 
@@ -2392,7 +2436,7 @@
 				continue;
 			}
 
-			desc->chip->set_affinity(irq, desc->pending_mask);
+			desc->chip->set_affinity(irq, &desc->pending_mask);
 			spin_unlock_irqrestore(&desc->lock, flags);
 		}
 	}
@@ -2401,18 +2445,20 @@
 /*
  * Migrates the IRQ destination in the process context.
  */
-static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, cpumask_t mask)
+static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
+					    const struct cpumask *mask)
 {
 	if (desc->status & IRQ_LEVEL) {
 		desc->status |= IRQ_MOVE_PENDING;
-		desc->pending_mask = mask;
+		cpumask_copy(&desc->pending_mask, mask);
 		migrate_irq_remapped_level_desc(desc);
 		return;
 	}
 
 	migrate_ioapic_irq_desc(desc, mask);
 }
-static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+static void set_ir_ioapic_affinity_irq(unsigned int irq,
+				       const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
@@ -2447,7 +2493,7 @@
 		if (!cfg->move_cleanup_count)
 			goto unlock;
 
-		if ((vector == cfg->vector) && cpu_isset(me, cfg->domain))
+		if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
 			goto unlock;
 
 		__get_cpu_var(vector_irq)[vector] = -1;
@@ -2484,20 +2530,14 @@
 
 	vector = ~get_irq_regs()->orig_ax;
 	me = smp_processor_id();
-	if ((vector == cfg->vector) && cpu_isset(me, cfg->domain)) {
-		cpumask_t cleanup_mask;
-
 #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
 		*descp = desc = move_irq_desc(desc, me);
 		/* get the new one */
 		cfg = desc->chip_data;
 #endif
 
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
+	if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
+		send_cleanup_vector(cfg);
 }
 #else
 static inline void irq_complete_move(struct irq_desc **descp) {}
@@ -2670,9 +2710,6 @@
 	 * 0x80, because int 0x80 is hm, kind of importantish. ;)
 	 */
 	for_each_irq_desc(irq, desc) {
-		if (!desc)
-			continue;
-
 		cfg = desc->chip_data;
 		if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) {
 			/*
@@ -3222,16 +3259,13 @@
 	struct irq_cfg *cfg;
 	int err;
 	unsigned dest;
-	cpumask_t tmp;
 
 	cfg = irq_cfg(irq);
-	tmp = TARGET_CPUS;
-	err = assign_irq_vector(irq, cfg, tmp);
+	err = assign_irq_vector(irq, cfg, TARGET_CPUS);
 	if (err)
 		return err;
 
-	cpus_and(tmp, cfg->domain, tmp);
-	dest = cpu_mask_to_apicid(tmp);
+	dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 #ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
@@ -3285,26 +3319,18 @@
 }
 
 #ifdef CONFIG_SMP
-static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	read_msi_msg_desc(desc, &msg);
 
@@ -3314,37 +3340,27 @@
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	write_msi_msg_desc(desc, &msg);
-	desc->affinity = mask;
 }
 #ifdef CONFIG_INTR_REMAP
 /*
  * Migrate the MSI irq to another cpumask. This migration is
  * done in the process context using interrupt-remapping hardware.
  */
-static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void
+ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg;
+	struct irq_cfg *cfg = desc->chip_data;
 	unsigned int dest;
-	cpumask_t tmp, cleanup_mask;
 	struct irte irte;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
-		return;
-
 	if (get_irte(irq, &irte))
 		return;
 
-	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
-
 	irte.vector = cfg->vector;
 	irte.dest_id = IRTE_DEST(dest);
 
@@ -3358,14 +3374,8 @@
 	 * at the new destination. So, time to cleanup the previous
 	 * vector allocation.
 	 */
-	if (cfg->move_in_progress) {
-		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
-		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
-		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
-		cfg->move_in_progress = 0;
-	}
-
-	desc->affinity = mask;
+	if (cfg->move_in_progress)
+		send_cleanup_vector(cfg);
 }
 
 #endif
@@ -3556,26 +3566,18 @@
 
 #ifdef CONFIG_DMAR
 #ifdef CONFIG_SMP
-static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	dmar_msi_read(irq, &msg);
 
@@ -3585,7 +3587,6 @@
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	dmar_msi_write(irq, &msg);
-	desc->affinity = mask;
 }
 
 #endif /* CONFIG_SMP */
@@ -3619,26 +3620,18 @@
 #ifdef CONFIG_HPET_TIMER
 
 #ifdef CONFIG_SMP
-static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	hpet_msi_read(irq, &msg);
 
@@ -3648,7 +3641,6 @@
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
 	hpet_msi_write(irq, &msg);
-	desc->affinity = mask;
 }
 
 #endif /* CONFIG_SMP */
@@ -3703,28 +3695,19 @@
 	write_ht_irq_msg(irq, &msg);
 }
 
-static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	struct irq_cfg *cfg;
 	unsigned int dest;
-	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	dest = set_desc_affinity(desc, mask);
+	if (dest == BAD_APICID)
 		return;
 
 	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
-		return;
-
-	set_extra_move_desc(desc, mask);
-
-	cpus_and(tmp, cfg->domain, mask);
-	dest = cpu_mask_to_apicid(tmp);
 
 	target_ht_irq(irq, dest, cfg->vector);
-	desc->affinity = mask;
 }
 
 #endif
@@ -3744,17 +3727,14 @@
 {
 	struct irq_cfg *cfg;
 	int err;
-	cpumask_t tmp;
 
 	cfg = irq_cfg(irq);
-	tmp = TARGET_CPUS;
-	err = assign_irq_vector(irq, cfg, tmp);
+	err = assign_irq_vector(irq, cfg, TARGET_CPUS);
 	if (!err) {
 		struct ht_irq_msg msg;
 		unsigned dest;
 
-		cpus_and(tmp, cfg->domain, tmp);
-		dest = cpu_mask_to_apicid(tmp);
+		dest = cpu_mask_to_apicid_and(cfg->domain, TARGET_CPUS);
 
 		msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest);
 
@@ -3790,7 +3770,7 @@
 int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 		       unsigned long mmr_offset)
 {
-	const cpumask_t *eligible_cpu = get_cpu_mask(cpu);
+	const struct cpumask *eligible_cpu = cpumask_of(cpu);
 	struct irq_cfg *cfg;
 	int mmr_pnode;
 	unsigned long mmr_value;
@@ -3800,7 +3780,7 @@
 
 	cfg = irq_cfg(irq);
 
-	err = assign_irq_vector(irq, cfg, *eligible_cpu);
+	err = assign_irq_vector(irq, cfg, eligible_cpu);
 	if (err != 0)
 		return err;
 
@@ -3819,7 +3799,7 @@
 	entry->polarity = 0;
 	entry->trigger = 0;
 	entry->mask = 0;
-	entry->dest = cpu_mask_to_apicid(*eligible_cpu);
+	entry->dest = cpu_mask_to_apicid(eligible_cpu);
 
 	mmr_pnode = uv_blade_to_pnode(mmr_blade);
 	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
@@ -4030,7 +4010,7 @@
 	int pin, ioapic, irq, irq_entry;
 	struct irq_desc *desc;
 	struct irq_cfg *cfg;
-	cpumask_t mask;
+	const struct cpumask *mask;
 
 	if (skip_ioapic_setup == 1)
 		return;
@@ -4061,7 +4041,7 @@
 			 */
 			if (desc->status &
 			    (IRQ_NO_BALANCING | IRQ_AFFINITY_SET))
-				mask = desc->affinity;
+				mask = &desc->affinity;
 			else
 				mask = TARGET_CPUS;
 
diff --git a/arch/x86/kernel/ipi.c b/arch/x86/kernel/ipi.c
index f1c688e..285bbf8 100644
--- a/arch/x86/kernel/ipi.c
+++ b/arch/x86/kernel/ipi.c
@@ -116,18 +116,18 @@
 /*
  * This is only used on smaller machines.
  */
-void send_IPI_mask_bitmask(cpumask_t cpumask, int vector)
+void send_IPI_mask_bitmask(const struct cpumask *cpumask, int vector)
 {
-	unsigned long mask = cpus_addr(cpumask)[0];
+	unsigned long mask = cpumask_bits(cpumask)[0];
 	unsigned long flags;
 
 	local_irq_save(flags);
-	WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]);
+	WARN_ON(mask & ~cpumask_bits(cpu_online_mask)[0]);
 	__send_IPI_dest_field(mask, vector);
 	local_irq_restore(flags);
 }
 
-void send_IPI_mask_sequence(cpumask_t mask, int vector)
+void send_IPI_mask_sequence(const struct cpumask *mask, int vector)
 {
 	unsigned long flags;
 	unsigned int query_cpu;
@@ -139,12 +139,24 @@
 	 */
 
 	local_irq_save(flags);
-	for_each_possible_cpu(query_cpu) {
-		if (cpu_isset(query_cpu, mask)) {
+	for_each_cpu(query_cpu, mask)
+		__send_IPI_dest_field(cpu_to_logical_apicid(query_cpu), vector);
+	local_irq_restore(flags);
+}
+
+void send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+	unsigned long flags;
+	unsigned int query_cpu;
+	unsigned int this_cpu = smp_processor_id();
+
+	/* See Hack comment above */
+
+	local_irq_save(flags);
+	for_each_cpu(query_cpu, mask)
+		if (query_cpu != this_cpu)
 			__send_IPI_dest_field(cpu_to_logical_apicid(query_cpu),
 					      vector);
-		}
-	}
 	local_irq_restore(flags);
 }
 
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 3f1d9d1..bce53e1 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -9,6 +9,7 @@
 #include <asm/apic.h>
 #include <asm/io_apic.h>
 #include <asm/smp.h>
+#include <asm/irq.h>
 
 atomic_t irq_err_count;
 
@@ -190,3 +191,5 @@
 #endif
 	return sum;
 }
+
+EXPORT_SYMBOL_GPL(vector_used_by_percpu_irq);
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 119fc9c..9dc5588 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -233,27 +233,28 @@
 #ifdef CONFIG_HOTPLUG_CPU
 #include <mach_apic.h>
 
-void fixup_irqs(cpumask_t map)
+/* A cpu has been removed from cpu_online_mask.  Reset irq affinities. */
+void fixup_irqs(void)
 {
 	unsigned int irq;
 	static int warned;
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		cpumask_t mask;
+		const struct cpumask *affinity;
 
 		if (!desc)
 			continue;
 		if (irq == 2)
 			continue;
 
-		cpus_and(mask, desc->affinity, map);
-		if (any_online_cpu(mask) == NR_CPUS) {
+		affinity = &desc->affinity;
+		if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
 			printk("Breaking affinity for irq %i\n", irq);
-			mask = map;
+			affinity = cpu_all_mask;
 		}
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, affinity);
 		else if (desc->action && !(warned++))
 			printk("Cannot set affinity for irq %i\n", irq);
 	}
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index a174a21..6383d50 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -80,16 +80,17 @@
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-void fixup_irqs(cpumask_t map)
+/* A cpu has been removed from cpu_online_mask.  Reset irq affinities. */
+void fixup_irqs(void)
 {
 	unsigned int irq;
 	static int warned;
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		cpumask_t mask;
 		int break_affinity = 0;
 		int set_affinity = 1;
+		const struct cpumask *affinity;
 
 		if (!desc)
 			continue;
@@ -99,23 +100,23 @@
 		/* interrupt's are disabled at this point */
 		spin_lock(&desc->lock);
 
+		affinity = &desc->affinity;
 		if (!irq_has_action(irq) ||
-		    cpus_equal(desc->affinity, map)) {
+		    cpumask_equal(affinity, cpu_online_mask)) {
 			spin_unlock(&desc->lock);
 			continue;
 		}
 
-		cpus_and(mask, desc->affinity, map);
-		if (cpus_empty(mask)) {
+		if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
 			break_affinity = 1;
-			mask = map;
+			affinity = cpu_all_mask;
 		}
 
 		if (desc->chip->mask)
 			desc->chip->mask(irq);
 
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, affinity);
 		else if (!(warned++))
 			set_affinity = 0;
 
diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c
index 203384e..8472329 100644
--- a/arch/x86/kernel/irqinit_32.c
+++ b/arch/x86/kernel/irqinit_32.c
@@ -110,6 +110,18 @@
 	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
 };
 
+int vector_used_by_percpu_irq(unsigned int vector)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		if (per_cpu(vector_irq, cpu)[vector] != -1)
+			return 1;
+	}
+
+	return 0;
+}
+
 /* Overridden in paravirt.c */
 void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ")));
 
@@ -146,10 +158,12 @@
 	alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
 
 	/* IPI for single call function */
-	set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, call_function_single_interrupt);
+	alloc_intr_gate(CALL_FUNCTION_SINGLE_VECTOR,
+				 call_function_single_interrupt);
 
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
+	set_bit(IRQ_MOVE_CLEANUP_VECTOR, used_vectors);
 #endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 6190e6ef..31ebfe3 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -69,6 +69,18 @@
 	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
 };
 
+int vector_used_by_percpu_irq(unsigned int vector)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		if (per_cpu(vector_irq, cpu)[vector] != -1)
+			return 1;
+	}
+
+	return 0;
+}
+
 void __init init_ISA_irqs(void)
 {
 	int i;
@@ -121,6 +133,7 @@
 
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
+	set_bit(IRQ_MOVE_CLEANUP_VECTOR, used_vectors);
 #endif
 }
 
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index e169ae9..652fce6d 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -89,17 +89,17 @@
  */
 static unsigned long kvm_get_tsc_khz(void)
 {
-	return preset_lpj;
+	struct pvclock_vcpu_time_info *src;
+	src = &per_cpu(hv_clock, 0);
+	return pvclock_tsc_khz(src);
 }
 
 static void kvm_get_preset_lpj(void)
 {
-	struct pvclock_vcpu_time_info *src;
 	unsigned long khz;
 	u64 lpj;
 
-	src = &per_cpu(hv_clock, 0);
-	khz = pvclock_tsc_khz(src);
+	khz = kvm_get_tsc_khz();
 
 	lpj = ((u64)khz * 1000);
 	do_div(lpj, HZ);
@@ -194,5 +194,7 @@
 #endif
 		kvm_get_preset_lpj();
 		clocksource_register(&kvm_clock);
+		pv_info.paravirt_enabled = 1;
+		pv_info.name = "KVM";
 	}
 }
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index eee32b4..71f1d99 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -12,8 +12,8 @@
 #include <linux/mm.h>
 #include <linux/smp.h>
 #include <linux/vmalloc.h>
+#include <linux/uaccess.h>
 
-#include <asm/uaccess.h>
 #include <asm/system.h>
 #include <asm/ldt.h>
 #include <asm/desc.h>
@@ -93,7 +93,7 @@
 	if (err < 0)
 		return err;
 
-	for(i = 0; i < old->size; i++)
+	for (i = 0; i < old->size; i++)
 		write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE);
 	return 0;
 }
diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c
index 3b59951..c12314c 100644
--- a/arch/x86/kernel/mfgpt_32.c
+++ b/arch/x86/kernel/mfgpt_32.c
@@ -287,7 +287,7 @@
 	.set_mode = mfgpt_set_mode,
 	.set_next_event = mfgpt_next_event,
 	.rating = 250,
-	.cpumask = CPU_MASK_ALL,
+	.cpumask = cpu_all_mask,
 	.shift = 32
 };
 
diff --git a/arch/x86/kernel/mmconf-fam10h_64.c b/arch/x86/kernel/mmconf-fam10h_64.c
index efc2f36..666e43d 100644
--- a/arch/x86/kernel/mmconf-fam10h_64.c
+++ b/arch/x86/kernel/mmconf-fam10h_64.c
@@ -13,8 +13,7 @@
 #include <asm/msr.h>
 #include <asm/acpi.h>
 #include <asm/mmconfig.h>
-
-#include "../pci/pci.h"
+#include <asm/pci_x86.h>
 
 struct pci_hostbridge_probe {
 	u32 bus;
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 45e3b69..c5c5b8d 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -16,14 +16,14 @@
 #include <linux/bitops.h>
 #include <linux/acpi.h>
 #include <linux/module.h>
+#include <linux/smp.h>
+#include <linux/acpi.h>
 
-#include <asm/smp.h>
 #include <asm/mtrr.h>
 #include <asm/mpspec.h>
 #include <asm/pgalloc.h>
 #include <asm/io_apic.h>
 #include <asm/proto.h>
-#include <asm/acpi.h>
 #include <asm/bios_ebda.h>
 #include <asm/e820.h>
 #include <asm/trampoline.h>
@@ -95,8 +95,8 @@
 #endif
 
 	if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA) - 1) == 0) {
-		 set_bit(m->mpc_busid, mp_bus_not_pci);
-#if defined(CONFIG_EISA) || defined (CONFIG_MCA)
+		set_bit(m->mpc_busid, mp_bus_not_pci);
+#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
 #endif
 	} else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI) - 1) == 0) {
@@ -104,7 +104,7 @@
 			x86_quirks->mpc_oem_pci_bus(m);
 
 		clear_bit(m->mpc_busid, mp_bus_not_pci);
-#if defined(CONFIG_EISA) || defined (CONFIG_MCA)
+#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
 	} else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA) - 1) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 8bd1bf9..45a09cc 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -26,11 +26,10 @@
 #include <linux/kernel_stat.h>
 #include <linux/kdebug.h>
 #include <linux/smp.h>
+#include <linux/nmi.h>
 
 #include <asm/i8259.h>
 #include <asm/io_apic.h>
-#include <asm/smp.h>
-#include <asm/nmi.h>
 #include <asm/proto.h>
 #include <asm/timer.h>
 
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index a35eaa3..00c2bcd 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -52,7 +52,7 @@
  * to trigger bugs with some popular PCI cards, in particular 3ware (but
  * has been also also seen with Qlogic at least).
  */
-int iommu_fullflush = 1;
+static int iommu_fullflush = 1;
 
 /* Allocation bitmap for the remapping area: */
 static DEFINE_SPINLOCK(iommu_bitmap_lock);
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 61f718d..bf088c6 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -12,6 +12,8 @@
 #include <asm/proto.h>
 #include <asm/reboot_fixups.h>
 #include <asm/reboot.h>
+#include <asm/pci_x86.h>
+#include <asm/virtext.h>
 
 #ifdef CONFIG_X86_32
 # include <linux/dmi.h>
@@ -23,7 +25,6 @@
 
 #include <mach_ipi.h>
 
-
 /*
  * Power off function, if any
  */
@@ -39,6 +40,12 @@
 static int reboot_cpu = -1;
 #endif
 
+/* This is set if we need to go through the 'emergency' path.
+ * When machine_emergency_restart() is called, we may be on
+ * an inconsistent state and won't be able to do a clean cleanup
+ */
+static int reboot_emergency;
+
 /* This is set by the PCI code if either type 1 or type 2 PCI is detected */
 bool port_cf9_safe = false;
 
@@ -368,6 +375,48 @@
 	}
 }
 
+static void vmxoff_nmi(int cpu, struct die_args *args)
+{
+	cpu_emergency_vmxoff();
+}
+
+/* Use NMIs as IPIs to tell all CPUs to disable virtualization
+ */
+static void emergency_vmx_disable_all(void)
+{
+	/* Just make sure we won't change CPUs while doing this */
+	local_irq_disable();
+
+	/* We need to disable VMX on all CPUs before rebooting, otherwise
+	 * we risk hanging up the machine, because the CPU ignore INIT
+	 * signals when VMX is enabled.
+	 *
+	 * We can't take any locks and we may be on an inconsistent
+	 * state, so we use NMIs as IPIs to tell the other CPUs to disable
+	 * VMX and halt.
+	 *
+	 * For safety, we will avoid running the nmi_shootdown_cpus()
+	 * stuff unnecessarily, but we don't have a way to check
+	 * if other CPUs have VMX enabled. So we will call it only if the
+	 * CPU we are running on has VMX enabled.
+	 *
+	 * We will miss cases where VMX is not enabled on all CPUs. This
+	 * shouldn't do much harm because KVM always enable VMX on all
+	 * CPUs anyway. But we can miss it on the small window where KVM
+	 * is still enabling VMX.
+	 */
+	if (cpu_has_vmx() && cpu_vmx_enabled()) {
+		/* Disable VMX on this CPU.
+		 */
+		cpu_vmxoff();
+
+		/* Halt and disable VMX on the other CPUs */
+		nmi_shootdown_cpus(vmxoff_nmi);
+
+	}
+}
+
+
 void __attribute__((weak)) mach_reboot_fixups(void)
 {
 }
@@ -376,6 +425,9 @@
 {
 	int i;
 
+	if (reboot_emergency)
+		emergency_vmx_disable_all();
+
 	/* Tell the BIOS if we want cold or warm reboot */
 	*((unsigned short *)__va(0x472)) = reboot_mode;
 
@@ -482,13 +534,19 @@
 #endif
 }
 
+static void __machine_emergency_restart(int emergency)
+{
+	reboot_emergency = emergency;
+	machine_ops.emergency_restart();
+}
+
 static void native_machine_restart(char *__unused)
 {
 	printk("machine restart\n");
 
 	if (!reboot_force)
 		machine_shutdown();
-	machine_emergency_restart();
+	__machine_emergency_restart(0);
 }
 
 static void native_machine_halt(void)
@@ -532,7 +590,7 @@
 
 void machine_emergency_restart(void)
 {
-	machine_ops.emergency_restart();
+	__machine_emergency_restart(1);
 }
 
 void machine_restart(char *cmd)
@@ -592,10 +650,7 @@
 
 static void smp_send_nmi_allbutself(void)
 {
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(safe_smp_processor_id(), mask);
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, NMI_VECTOR);
+	send_IPI_allbutself(NMI_VECTOR);
 }
 
 static struct notifier_block crash_nmi_nb = {
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index ae0c0d3..0b63b08 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -152,6 +152,11 @@
 	old_size = PERCPU_ENOUGH_ROOM;
 	align = max_t(unsigned long, PAGE_SIZE, align);
 	size = roundup(old_size, align);
+
+	printk(KERN_INFO
+		"NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n",
+		NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids);
+
 	printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
 			  size);
 
@@ -168,24 +173,24 @@
 			       "cpu %d has no node %d or node-local memory\n",
 				cpu, node);
 			if (ptr)
-				printk(KERN_DEBUG "per cpu data for cpu%d at %016lx\n",
+				printk(KERN_DEBUG
+					"per cpu data for cpu%d at %016lx\n",
 					 cpu, __pa(ptr));
 		}
 		else {
 			ptr = __alloc_bootmem_node(NODE_DATA(node), size, align,
 							__pa(MAX_DMA_ADDRESS));
 			if (ptr)
-				printk(KERN_DEBUG "per cpu data for cpu%d on node%d at %016lx\n",
-					 cpu, node, __pa(ptr));
+				printk(KERN_DEBUG
+					"per cpu data for cpu%d on node%d "
+					"at %016lx\n",
+					cpu, node, __pa(ptr));
 		}
 #endif
 		per_cpu_offset(cpu) = ptr - __per_cpu_start;
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 	}
 
-	printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d, nr_node_ids %d\n",
-		NR_CPUS, nr_cpu_ids, nr_node_ids);
-
 	/* Setup percpu data maps */
 	setup_per_cpu_maps();
 
@@ -282,7 +287,7 @@
 	else
 		cpu_clear(cpu, *mask);
 
-	cpulist_scnprintf(buf, sizeof(buf), *mask);
+	cpulist_scnprintf(buf, sizeof(buf), mask);
 	printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n",
 		enable? "numa_add_cpu":"numa_remove_cpu", cpu, node, buf);
  }
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 7e558db..beea2649a 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -118,22 +118,22 @@
 		WARN_ON(1);
 		return;
 	}
-	send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
+	send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR);
 }
 
 void native_send_call_func_single_ipi(int cpu)
 {
-	send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNCTION_SINGLE_VECTOR);
+	send_IPI_mask(cpumask_of(cpu), CALL_FUNCTION_SINGLE_VECTOR);
 }
 
-void native_send_call_func_ipi(cpumask_t mask)
+void native_send_call_func_ipi(const struct cpumask *mask)
 {
 	cpumask_t allbutself;
 
 	allbutself = cpu_online_map;
 	cpu_clear(smp_processor_id(), allbutself);
 
-	if (cpus_equal(mask, allbutself) &&
+	if (cpus_equal(*mask, allbutself) &&
 	    cpus_equal(cpu_online_map, cpu_callout_map))
 		send_IPI_allbutself(CALL_FUNCTION_VECTOR);
 	else
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index f8500c9..31869bf 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -102,14 +102,8 @@
 /* Last level cache ID of each logical CPU */
 DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
 
-/* bitmap of online cpus */
-cpumask_t cpu_online_map __read_mostly;
-EXPORT_SYMBOL(cpu_online_map);
-
 cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
 
 /* representing HT siblings of each logical CPU */
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map);
@@ -1260,6 +1254,15 @@
 	check_nmi_watchdog();
 }
 
+static int __initdata setup_possible_cpus = -1;
+static int __init _setup_possible_cpus(char *str)
+{
+	get_option(&str, &setup_possible_cpus);
+	return 0;
+}
+early_param("possible_cpus", _setup_possible_cpus);
+
+
 /*
  * cpu_possible_map should be static, it cannot change as cpu's
  * are onlined, or offlined. The reason is per-cpu data-structures
@@ -1272,7 +1275,7 @@
  *
  * Three ways to find out the number of additional hotplug CPUs:
  * - If the BIOS specified disabled CPUs in ACPI/mptables use that.
- * - The user can overwrite it with additional_cpus=NUM
+ * - The user can overwrite it with possible_cpus=NUM
  * - Otherwise don't reserve additional CPUs.
  * We do this because additional CPUs waste a lot of memory.
  * -AK
@@ -1285,9 +1288,17 @@
 	if (!num_processors)
 		num_processors = 1;
 
-	possible = num_processors + disabled_cpus;
-	if (possible > NR_CPUS)
-		possible = NR_CPUS;
+	if (setup_possible_cpus == -1)
+		possible = num_processors + disabled_cpus;
+	else
+		possible = setup_possible_cpus;
+
+	if (possible > CONFIG_NR_CPUS) {
+		printk(KERN_WARNING
+			"%d Processors exceeds NR_CPUS limit of %d\n",
+			possible, CONFIG_NR_CPUS);
+		possible = CONFIG_NR_CPUS;
+	}
 
 	printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n",
 		possible, max_t(int, possible - num_processors, 0));
@@ -1352,7 +1363,7 @@
 	lock_vector_lock();
 	remove_cpu_from_maps(cpu);
 	unlock_vector_lock();
-	fixup_irqs(cpu_online_map);
+	fixup_irqs();
 }
 
 int native_cpu_disable(void)
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index 8da059f..ce50546 100644
--- a/arch/x86/kernel/tlb_32.c
+++ b/arch/x86/kernel/tlb_32.c
@@ -163,7 +163,7 @@
 	 * We have to send the IPI only to
 	 * CPUs affected.
 	 */
-	send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR);
+	send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR);
 
 	while (!cpus_empty(flush_cpumask))
 		/* nothing. lockup detection does not belong here */
diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c
index 29887d7..f8be6f1 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -191,7 +191,7 @@
 	 * We have to send the IPI only to
 	 * CPUs affected.
 	 */
-	send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR_START + sender);
+	send_IPI_mask(&cpumask, INVALIDATE_TLB_VECTOR_START + sender);
 
 	while (!cpus_empty(f->flush_cpumask))
 		cpu_relax();
diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c
index 6a00e5f..f885023 100644
--- a/arch/x86/kernel/tlb_uv.c
+++ b/arch/x86/kernel/tlb_uv.c
@@ -582,7 +582,6 @@
 static struct bau_control * __init uv_table_bases_init(int blade, int node)
 {
 	int i;
-	int *ip;
 	struct bau_msg_status *msp;
 	struct bau_control *bau_tabp;
 
@@ -599,13 +598,6 @@
 		bau_cpubits_clear(&msp->seen_by, (int)
 				  uv_blade_nr_possible_cpus(blade));
 
-	bau_tabp->watching =
-	    kmalloc_node(sizeof(int) * DEST_NUM_RESOURCES, GFP_KERNEL, node);
-	BUG_ON(!bau_tabp->watching);
-
-	for (i = 0, ip = bau_tabp->watching; i < DEST_Q_SIZE; i++, ip++)
-		*ip = 0;
-
 	uv_bau_table_bases[blade] = bau_tabp;
 
 	return bau_tabp;
@@ -628,7 +620,6 @@
 		bcp->bau_msg_head	= bau_tablesp->va_queue_first;
 		bcp->va_queue_first	= bau_tablesp->va_queue_first;
 		bcp->va_queue_last	= bau_tablesp->va_queue_last;
-		bcp->watching		= bau_tablesp->watching;
 		bcp->msg_statuses	= bau_tablesp->msg_statuses;
 		bcp->descriptor_base	= adp;
 	}
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 141907a..ce6650e 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -72,9 +72,6 @@
 
 #include "cpu/mcheck/mce.h"
 
-DECLARE_BITMAP(used_vectors, NR_VECTORS);
-EXPORT_SYMBOL_GPL(used_vectors);
-
 asmlinkage int system_call(void);
 
 /* Do we ignore FPU interrupts ? */
@@ -89,6 +86,9 @@
 	__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
 #endif
 
+DECLARE_BITMAP(used_vectors, NR_VECTORS);
+EXPORT_SYMBOL_GPL(used_vectors);
+
 static int ignore_nmis;
 
 static inline void conditional_sti(struct pt_regs *regs)
@@ -292,8 +292,10 @@
 	tsk->thread.error_code = error_code;
 	tsk->thread.trap_no = 8;
 
-	/* This is always a kernel trap and never fixable (and thus must
-	   never return). */
+	/*
+	 * This is always a kernel trap and never fixable (and thus must
+	 * never return).
+	 */
 	for (;;)
 		die(str, regs, error_code);
 }
@@ -520,9 +522,11 @@
 }
 
 #ifdef CONFIG_X86_64
-/* Help handler running on IST stack to switch back to user stack
-   for scheduling or signal handling. The actual stack switch is done in
-   entry.S */
+/*
+ * Help handler running on IST stack to switch back to user stack
+ * for scheduling or signal handling. The actual stack switch is done in
+ * entry.S
+ */
 asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 {
 	struct pt_regs *regs = eregs;
@@ -532,8 +536,10 @@
 	/* Exception from user space */
 	else if (user_mode(eregs))
 		regs = task_pt_regs(current);
-	/* Exception from kernel and interrupts are enabled. Move to
-	   kernel process stack. */
+	/*
+	 * Exception from kernel and interrupts are enabled. Move to
+	 * kernel process stack.
+	 */
 	else if (eregs->flags & X86_EFLAGS_IF)
 		regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs));
 	if (eregs != regs)
@@ -685,12 +691,7 @@
 	cwd = get_fpu_cwd(task);
 	swd = get_fpu_swd(task);
 
-	err = swd & ~cwd & 0x3f;
-
-#ifdef CONFIG_X86_32
-	if (!err)
-		return;
-#endif
+	err = swd & ~cwd;
 
 	if (err & 0x001) {	/* Invalid op */
 		/*
@@ -708,7 +709,11 @@
 	} else if (err & 0x020) { /* Precision */
 		info.si_code = FPE_FLTRES;
 	} else {
-		info.si_code = __SI_FAULT|SI_KERNEL; /* WTF? */
+		/*
+		 * If we're using IRQ 13, or supposedly even some trap 16
+		 * implementations, it's possible we get a spurious trap...
+		 */
+		return;		/* Spurious trap, no error */
 	}
 	force_sig_info(SIGFPE, &info, task);
 }
@@ -941,9 +946,7 @@
 
 void __init trap_init(void)
 {
-#ifdef CONFIG_X86_32
 	int i;
-#endif
 
 #ifdef CONFIG_EISA
 	void __iomem *p = early_ioremap(0x0FFFD9, 4);
@@ -1000,11 +1003,15 @@
 	}
 
 	set_system_trap_gate(SYSCALL_VECTOR, &system_call);
+#endif
 
 	/* Reserve all the builtin and the syscall vector: */
 	for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++)
 		set_bit(i, used_vectors);
 
+#ifdef CONFIG_X86_64
+	set_bit(IA32_SYSCALL_VECTOR, used_vectors);
+#else
 	set_bit(SYSCALL_VECTOR, used_vectors);
 #endif
 	/*
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
index 254ee07..c4c1f9e 100644
--- a/arch/x86/kernel/vmiclock_32.c
+++ b/arch/x86/kernel/vmiclock_32.c
@@ -226,7 +226,7 @@
 	/* Upper bound is clockevent's use of ulong for cycle deltas. */
 	evt->max_delta_ns = clockevent_delta2ns(ULONG_MAX, evt);
 	evt->min_delta_ns = clockevent_delta2ns(1, evt);
-	evt->cpumask = cpumask_of_cpu(cpu);
+	evt->cpumask = cpumask_of(cpu);
 
 	printk(KERN_WARNING "vmi: registering clock event %s. mult=%lu shift=%u\n",
 	       evt->name, evt->mult, evt->shift);
diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c
index 15c3e69..2b54fe0 100644
--- a/arch/x86/kernel/xsave.c
+++ b/arch/x86/kernel/xsave.c
@@ -159,7 +159,7 @@
  * Restore the extended state if present. Otherwise, restore the FP/SSE
  * state.
  */
-int restore_user_xstate(void __user *buf)
+static int restore_user_xstate(void __user *buf)
 {
 	struct _fpx_sw_bytes fx_sw_user;
 	u64 mask;
diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c
index 59ebd37..e665d1c 100644
--- a/arch/x86/kvm/i8254.c
+++ b/arch/x86/kvm/i8254.c
@@ -603,10 +603,29 @@
 
 static void __inject_pit_timer_intr(struct kvm *kvm)
 {
+	struct kvm_vcpu *vcpu;
+	int i;
+
 	mutex_lock(&kvm->lock);
 	kvm_set_irq(kvm, kvm->arch.vpit->irq_source_id, 0, 1);
 	kvm_set_irq(kvm, kvm->arch.vpit->irq_source_id, 0, 0);
 	mutex_unlock(&kvm->lock);
+
+	/*
+	 * Provides NMI watchdog support via Virtual Wire mode.
+	 * The route is: PIT -> PIC -> LVT0 in NMI mode.
+	 *
+	 * Note: Our Virtual Wire implementation is simplified, only
+	 * propagating PIT interrupts to all VCPUs when they have set
+	 * LVT0 to NMI delivery. Other PIC interrupts are just sent to
+	 * VCPU0, and only if its LVT0 is in EXTINT mode.
+	 */
+	if (kvm->arch.vapics_in_nmi_mode > 0)
+		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+			vcpu = kvm->vcpus[i];
+			if (vcpu)
+				kvm_apic_nmi_wd_deliver(vcpu);
+		}
 }
 
 void kvm_inject_pit_timer_irqs(struct kvm_vcpu *vcpu)
diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c
index 17e41e1..179dcb0 100644
--- a/arch/x86/kvm/i8259.c
+++ b/arch/x86/kvm/i8259.c
@@ -26,10 +26,40 @@
  *   Port from Qemu.
  */
 #include <linux/mm.h>
+#include <linux/bitops.h>
 #include "irq.h"
 
 #include <linux/kvm_host.h>
 
+static void pic_lock(struct kvm_pic *s)
+{
+	spin_lock(&s->lock);
+}
+
+static void pic_unlock(struct kvm_pic *s)
+{
+	struct kvm *kvm = s->kvm;
+	unsigned acks = s->pending_acks;
+	bool wakeup = s->wakeup_needed;
+	struct kvm_vcpu *vcpu;
+
+	s->pending_acks = 0;
+	s->wakeup_needed = false;
+
+	spin_unlock(&s->lock);
+
+	while (acks) {
+		kvm_notify_acked_irq(kvm, __ffs(acks));
+		acks &= acks - 1;
+	}
+
+	if (wakeup) {
+		vcpu = s->kvm->vcpus[0];
+		if (vcpu)
+			kvm_vcpu_kick(vcpu);
+	}
+}
+
 static void pic_clear_isr(struct kvm_kpic_state *s, int irq)
 {
 	s->isr &= ~(1 << irq);
@@ -136,17 +166,21 @@
 
 void kvm_pic_update_irq(struct kvm_pic *s)
 {
+	pic_lock(s);
 	pic_update_irq(s);
+	pic_unlock(s);
 }
 
 void kvm_pic_set_irq(void *opaque, int irq, int level)
 {
 	struct kvm_pic *s = opaque;
 
+	pic_lock(s);
 	if (irq >= 0 && irq < PIC_NUM_PINS) {
 		pic_set_irq1(&s->pics[irq >> 3], irq & 7, level);
 		pic_update_irq(s);
 	}
+	pic_unlock(s);
 }
 
 /*
@@ -172,6 +206,7 @@
 	int irq, irq2, intno;
 	struct kvm_pic *s = pic_irqchip(kvm);
 
+	pic_lock(s);
 	irq = pic_get_irq(&s->pics[0]);
 	if (irq >= 0) {
 		pic_intack(&s->pics[0], irq);
@@ -196,6 +231,7 @@
 		intno = s->pics[0].irq_base + irq;
 	}
 	pic_update_irq(s);
+	pic_unlock(s);
 	kvm_notify_acked_irq(kvm, irq);
 
 	return intno;
@@ -203,7 +239,7 @@
 
 void kvm_pic_reset(struct kvm_kpic_state *s)
 {
-	int irq, irqbase;
+	int irq, irqbase, n;
 	struct kvm *kvm = s->pics_state->irq_request_opaque;
 	struct kvm_vcpu *vcpu0 = kvm->vcpus[0];
 
@@ -214,8 +250,10 @@
 
 	for (irq = 0; irq < PIC_NUM_PINS/2; irq++) {
 		if (vcpu0 && kvm_apic_accept_pic_intr(vcpu0))
-			if (s->irr & (1 << irq) || s->isr & (1 << irq))
-				kvm_notify_acked_irq(kvm, irq+irqbase);
+			if (s->irr & (1 << irq) || s->isr & (1 << irq)) {
+				n = irq + irqbase;
+				s->pics_state->pending_acks |= 1 << n;
+			}
 	}
 	s->last_irr = 0;
 	s->irr = 0;
@@ -406,6 +444,7 @@
 			printk(KERN_ERR "PIC: non byte write\n");
 		return;
 	}
+	pic_lock(s);
 	switch (addr) {
 	case 0x20:
 	case 0x21:
@@ -418,6 +457,7 @@
 		elcr_ioport_write(&s->pics[addr & 1], addr, data);
 		break;
 	}
+	pic_unlock(s);
 }
 
 static void picdev_read(struct kvm_io_device *this,
@@ -431,6 +471,7 @@
 			printk(KERN_ERR "PIC: non byte read\n");
 		return;
 	}
+	pic_lock(s);
 	switch (addr) {
 	case 0x20:
 	case 0x21:
@@ -444,6 +485,7 @@
 		break;
 	}
 	*(unsigned char *)val = data;
+	pic_unlock(s);
 }
 
 /*
@@ -459,7 +501,7 @@
 	s->output = level;
 	if (vcpu && level && (s->pics[0].isr_ack & (1 << irq))) {
 		s->pics[0].isr_ack &= ~(1 << irq);
-		kvm_vcpu_kick(vcpu);
+		s->wakeup_needed = true;
 	}
 }
 
@@ -469,6 +511,8 @@
 	s = kzalloc(sizeof(struct kvm_pic), GFP_KERNEL);
 	if (!s)
 		return NULL;
+	spin_lock_init(&s->lock);
+	s->kvm = kvm;
 	s->pics[0].elcr_mask = 0xf8;
 	s->pics[1].elcr_mask = 0xde;
 	s->irq_request = pic_irq_request;
diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
index f17c8f5..2bf32a0 100644
--- a/arch/x86/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -25,6 +25,7 @@
 #include <linux/mm_types.h>
 #include <linux/hrtimer.h>
 #include <linux/kvm_host.h>
+#include <linux/spinlock.h>
 
 #include "iodev.h"
 #include "ioapic.h"
@@ -59,6 +60,10 @@
 };
 
 struct kvm_pic {
+	spinlock_t lock;
+	bool wakeup_needed;
+	unsigned pending_acks;
+	struct kvm *kvm;
 	struct kvm_kpic_state pics[2]; /* 0 is master pic, 1 is slave pic */
 	irq_request_func *irq_request;
 	void *irq_request_opaque;
@@ -87,6 +92,7 @@
 void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
 void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
 void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu);
+void kvm_apic_nmi_wd_deliver(struct kvm_vcpu *vcpu);
 void __kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
 void __kvm_migrate_pit_timer(struct kvm_vcpu *vcpu);
 void __kvm_migrate_timers(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/kvm/kvm_svm.h b/arch/x86/kvm/kvm_svm.h
index 65ef0fc..8e5ee99 100644
--- a/arch/x86/kvm/kvm_svm.h
+++ b/arch/x86/kvm/kvm_svm.h
@@ -7,7 +7,7 @@
 #include <linux/kvm_host.h>
 #include <asm/msr.h>
 
-#include "svm.h"
+#include <asm/svm.h>
 
 static const u32 host_save_user_msrs[] = {
 #ifdef CONFIG_X86_64
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 0fc3cab..afac68c 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -130,6 +130,11 @@
 	return apic_get_reg(apic, APIC_LVTT) & APIC_LVT_TIMER_PERIODIC;
 }
 
+static inline int apic_lvt_nmi_mode(u32 lvt_val)
+{
+	return (lvt_val & (APIC_MODE_MASK | APIC_LVT_MASKED)) == APIC_DM_NMI;
+}
+
 static unsigned int apic_lvt_mask[APIC_LVT_NUM] = {
 	LVT_MASK | APIC_LVT_TIMER_PERIODIC,	/* LVTT */
 	LVT_MASK | APIC_MODE_MASK,	/* LVTTHMR */
@@ -354,6 +359,7 @@
 
 	case APIC_DM_NMI:
 		kvm_inject_nmi(vcpu);
+		kvm_vcpu_kick(vcpu);
 		break;
 
 	case APIC_DM_INIT:
@@ -380,6 +386,14 @@
 		}
 		break;
 
+	case APIC_DM_EXTINT:
+		/*
+		 * Should only be called by kvm_apic_local_deliver() with LVT0,
+		 * before NMI watchdog was enabled. Already handled by
+		 * kvm_apic_accept_pic_intr().
+		 */
+		break;
+
 	default:
 		printk(KERN_ERR "TODO: unsupported delivery mode %x\n",
 		       delivery_mode);
@@ -663,6 +677,20 @@
 					apic->timer.period)));
 }
 
+static void apic_manage_nmi_watchdog(struct kvm_lapic *apic, u32 lvt0_val)
+{
+	int nmi_wd_enabled = apic_lvt_nmi_mode(apic_get_reg(apic, APIC_LVT0));
+
+	if (apic_lvt_nmi_mode(lvt0_val)) {
+		if (!nmi_wd_enabled) {
+			apic_debug("Receive NMI setting on APIC_LVT0 "
+				   "for cpu %d\n", apic->vcpu->vcpu_id);
+			apic->vcpu->kvm->arch.vapics_in_nmi_mode++;
+		}
+	} else if (nmi_wd_enabled)
+		apic->vcpu->kvm->arch.vapics_in_nmi_mode--;
+}
+
 static void apic_mmio_write(struct kvm_io_device *this,
 			    gpa_t address, int len, const void *data)
 {
@@ -743,10 +771,11 @@
 		apic_set_reg(apic, APIC_ICR2, val & 0xff000000);
 		break;
 
+	case APIC_LVT0:
+		apic_manage_nmi_watchdog(apic, val);
 	case APIC_LVTT:
 	case APIC_LVTTHMR:
 	case APIC_LVTPC:
-	case APIC_LVT0:
 	case APIC_LVT1:
 	case APIC_LVTERR:
 		/* TODO: Check vector */
@@ -961,12 +990,26 @@
 	return 0;
 }
 
-static int __inject_apic_timer_irq(struct kvm_lapic *apic)
+static int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type)
 {
-	int vector;
+	u32 reg = apic_get_reg(apic, lvt_type);
+	int vector, mode, trig_mode;
 
-	vector = apic_lvt_vector(apic, APIC_LVTT);
-	return __apic_accept_irq(apic, APIC_DM_FIXED, vector, 1, 0);
+	if (apic_hw_enabled(apic) && !(reg & APIC_LVT_MASKED)) {
+		vector = reg & APIC_VECTOR_MASK;
+		mode = reg & APIC_MODE_MASK;
+		trig_mode = reg & APIC_LVT_LEVEL_TRIGGER;
+		return __apic_accept_irq(apic, mode, vector, 1, trig_mode);
+	}
+	return 0;
+}
+
+void kvm_apic_nmi_wd_deliver(struct kvm_vcpu *vcpu)
+{
+	struct kvm_lapic *apic = vcpu->arch.apic;
+
+	if (apic)
+		kvm_apic_local_deliver(apic, APIC_LVT0);
 }
 
 static enum hrtimer_restart apic_timer_fn(struct hrtimer *data)
@@ -1061,9 +1104,8 @@
 {
 	struct kvm_lapic *apic = vcpu->arch.apic;
 
-	if (apic && apic_lvt_enabled(apic, APIC_LVTT) &&
-		atomic_read(&apic->timer.pending) > 0) {
-		if (__inject_apic_timer_irq(apic))
+	if (apic && atomic_read(&apic->timer.pending) > 0) {
+		if (kvm_apic_local_deliver(apic, APIC_LVTT))
 			atomic_dec(&apic->timer.pending);
 	}
 }
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 410ddbc..83f11c7 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -17,7 +17,6 @@
  *
  */
 
-#include "vmx.h"
 #include "mmu.h"
 
 #include <linux/kvm_host.h>
@@ -33,6 +32,7 @@
 #include <asm/page.h>
 #include <asm/cmpxchg.h>
 #include <asm/io.h>
+#include <asm/vmx.h>
 
 /*
  * When setting this variable to true it enables Two-Dimensional-Paging
@@ -168,6 +168,7 @@
 static u64 __read_mostly shadow_user_mask;
 static u64 __read_mostly shadow_accessed_mask;
 static u64 __read_mostly shadow_dirty_mask;
+static u64 __read_mostly shadow_mt_mask;
 
 void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte)
 {
@@ -183,13 +184,14 @@
 EXPORT_SYMBOL_GPL(kvm_mmu_set_base_ptes);
 
 void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
-		u64 dirty_mask, u64 nx_mask, u64 x_mask)
+		u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 mt_mask)
 {
 	shadow_user_mask = user_mask;
 	shadow_accessed_mask = accessed_mask;
 	shadow_dirty_mask = dirty_mask;
 	shadow_nx_mask = nx_mask;
 	shadow_x_mask = x_mask;
+	shadow_mt_mask = mt_mask;
 }
 EXPORT_SYMBOL_GPL(kvm_mmu_set_mask_ptes);
 
@@ -384,7 +386,9 @@
 {
 	int *write_count;
 
-	write_count = slot_largepage_idx(gfn, gfn_to_memslot(kvm, gfn));
+	gfn = unalias_gfn(kvm, gfn);
+	write_count = slot_largepage_idx(gfn,
+					 gfn_to_memslot_unaliased(kvm, gfn));
 	*write_count += 1;
 }
 
@@ -392,16 +396,20 @@
 {
 	int *write_count;
 
-	write_count = slot_largepage_idx(gfn, gfn_to_memslot(kvm, gfn));
+	gfn = unalias_gfn(kvm, gfn);
+	write_count = slot_largepage_idx(gfn,
+					 gfn_to_memslot_unaliased(kvm, gfn));
 	*write_count -= 1;
 	WARN_ON(*write_count < 0);
 }
 
 static int has_wrprotected_page(struct kvm *kvm, gfn_t gfn)
 {
-	struct kvm_memory_slot *slot = gfn_to_memslot(kvm, gfn);
+	struct kvm_memory_slot *slot;
 	int *largepage_idx;
 
+	gfn = unalias_gfn(kvm, gfn);
+	slot = gfn_to_memslot_unaliased(kvm, gfn);
 	if (slot) {
 		largepage_idx = slot_largepage_idx(gfn, slot);
 		return *largepage_idx;
@@ -613,7 +621,7 @@
 	return NULL;
 }
 
-static void rmap_write_protect(struct kvm *kvm, u64 gfn)
+static int rmap_write_protect(struct kvm *kvm, u64 gfn)
 {
 	unsigned long *rmapp;
 	u64 *spte;
@@ -659,8 +667,7 @@
 		spte = rmap_next(kvm, rmapp, spte);
 	}
 
-	if (write_protected)
-		kvm_flush_remote_tlbs(kvm);
+	return write_protected;
 }
 
 static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
@@ -786,9 +793,11 @@
 	sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
 	set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
 	list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages);
+	INIT_LIST_HEAD(&sp->oos_link);
 	ASSERT(is_empty_shadow_page(sp->spt));
-	sp->slot_bitmap = 0;
+	bitmap_zero(sp->slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS);
 	sp->multimapped = 0;
+	sp->global = 1;
 	sp->parent_pte = parent_pte;
 	--vcpu->kvm->arch.n_free_mmu_pages;
 	return sp;
@@ -900,8 +909,9 @@
 	struct kvm_mmu_page *sp = page_header(__pa(spte));
 
 	index = spte - sp->spt;
-	__set_bit(index, sp->unsync_child_bitmap);
-	sp->unsync_children = 1;
+	if (!__test_and_set_bit(index, sp->unsync_child_bitmap))
+		sp->unsync_children++;
+	WARN_ON(!sp->unsync_children);
 }
 
 static void kvm_mmu_update_parents_unsync(struct kvm_mmu_page *sp)
@@ -928,7 +938,6 @@
 
 static int unsync_walk_fn(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 {
-	sp->unsync_children = 1;
 	kvm_mmu_update_parents_unsync(sp);
 	return 1;
 }
@@ -959,38 +968,66 @@
 {
 }
 
+#define KVM_PAGE_ARRAY_NR 16
+
+struct kvm_mmu_pages {
+	struct mmu_page_and_offset {
+		struct kvm_mmu_page *sp;
+		unsigned int idx;
+	} page[KVM_PAGE_ARRAY_NR];
+	unsigned int nr;
+};
+
 #define for_each_unsync_children(bitmap, idx)		\
 	for (idx = find_first_bit(bitmap, 512);		\
 	     idx < 512;					\
 	     idx = find_next_bit(bitmap, 512, idx+1))
 
-static int mmu_unsync_walk(struct kvm_mmu_page *sp,
-			   struct kvm_unsync_walk *walker)
+int mmu_pages_add(struct kvm_mmu_pages *pvec, struct kvm_mmu_page *sp,
+		   int idx)
 {
-	int i, ret;
+	int i;
 
-	if (!sp->unsync_children)
-		return 0;
+	if (sp->unsync)
+		for (i=0; i < pvec->nr; i++)
+			if (pvec->page[i].sp == sp)
+				return 0;
+
+	pvec->page[pvec->nr].sp = sp;
+	pvec->page[pvec->nr].idx = idx;
+	pvec->nr++;
+	return (pvec->nr == KVM_PAGE_ARRAY_NR);
+}
+
+static int __mmu_unsync_walk(struct kvm_mmu_page *sp,
+			   struct kvm_mmu_pages *pvec)
+{
+	int i, ret, nr_unsync_leaf = 0;
 
 	for_each_unsync_children(sp->unsync_child_bitmap, i) {
 		u64 ent = sp->spt[i];
 
-		if (is_shadow_present_pte(ent)) {
+		if (is_shadow_present_pte(ent) && !is_large_pte(ent)) {
 			struct kvm_mmu_page *child;
 			child = page_header(ent & PT64_BASE_ADDR_MASK);
 
 			if (child->unsync_children) {
-				ret = mmu_unsync_walk(child, walker);
-				if (ret)
+				if (mmu_pages_add(pvec, child, i))
+					return -ENOSPC;
+
+				ret = __mmu_unsync_walk(child, pvec);
+				if (!ret)
+					__clear_bit(i, sp->unsync_child_bitmap);
+				else if (ret > 0)
+					nr_unsync_leaf += ret;
+				else
 					return ret;
-				__clear_bit(i, sp->unsync_child_bitmap);
 			}
 
 			if (child->unsync) {
-				ret = walker->entry(child, walker);
-				__clear_bit(i, sp->unsync_child_bitmap);
-				if (ret)
-					return ret;
+				nr_unsync_leaf++;
+				if (mmu_pages_add(pvec, child, i))
+					return -ENOSPC;
 			}
 		}
 	}
@@ -998,7 +1035,17 @@
 	if (find_first_bit(sp->unsync_child_bitmap, 512) == 512)
 		sp->unsync_children = 0;
 
-	return 0;
+	return nr_unsync_leaf;
+}
+
+static int mmu_unsync_walk(struct kvm_mmu_page *sp,
+			   struct kvm_mmu_pages *pvec)
+{
+	if (!sp->unsync_children)
+		return 0;
+
+	mmu_pages_add(pvec, sp, 0);
+	return __mmu_unsync_walk(sp, pvec);
 }
 
 static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn)
@@ -1021,10 +1068,18 @@
 	return NULL;
 }
 
+static void kvm_unlink_unsync_global(struct kvm *kvm, struct kvm_mmu_page *sp)
+{
+	list_del(&sp->oos_link);
+	--kvm->stat.mmu_unsync_global;
+}
+
 static void kvm_unlink_unsync_page(struct kvm *kvm, struct kvm_mmu_page *sp)
 {
 	WARN_ON(!sp->unsync);
 	sp->unsync = 0;
+	if (sp->global)
+		kvm_unlink_unsync_global(kvm, sp);
 	--kvm->stat.mmu_unsync;
 }
 
@@ -1037,7 +1092,8 @@
 		return 1;
 	}
 
-	rmap_write_protect(vcpu->kvm, sp->gfn);
+	if (rmap_write_protect(vcpu->kvm, sp->gfn))
+		kvm_flush_remote_tlbs(vcpu->kvm);
 	kvm_unlink_unsync_page(vcpu->kvm, sp);
 	if (vcpu->arch.mmu.sync_page(vcpu, sp)) {
 		kvm_mmu_zap_page(vcpu->kvm, sp);
@@ -1048,30 +1104,89 @@
 	return 0;
 }
 
-struct sync_walker {
-	struct kvm_vcpu *vcpu;
-	struct kvm_unsync_walk walker;
+struct mmu_page_path {
+	struct kvm_mmu_page *parent[PT64_ROOT_LEVEL-1];
+	unsigned int idx[PT64_ROOT_LEVEL-1];
 };
 
-static int mmu_sync_fn(struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk)
-{
-	struct sync_walker *sync_walk = container_of(walk, struct sync_walker,
-						     walker);
-	struct kvm_vcpu *vcpu = sync_walk->vcpu;
+#define for_each_sp(pvec, sp, parents, i)			\
+		for (i = mmu_pages_next(&pvec, &parents, -1),	\
+			sp = pvec.page[i].sp;			\
+			i < pvec.nr && ({ sp = pvec.page[i].sp; 1;});	\
+			i = mmu_pages_next(&pvec, &parents, i))
 
-	kvm_sync_page(vcpu, sp);
-	return (need_resched() || spin_needbreak(&vcpu->kvm->mmu_lock));
+int mmu_pages_next(struct kvm_mmu_pages *pvec, struct mmu_page_path *parents,
+		   int i)
+{
+	int n;
+
+	for (n = i+1; n < pvec->nr; n++) {
+		struct kvm_mmu_page *sp = pvec->page[n].sp;
+
+		if (sp->role.level == PT_PAGE_TABLE_LEVEL) {
+			parents->idx[0] = pvec->page[n].idx;
+			return n;
+		}
+
+		parents->parent[sp->role.level-2] = sp;
+		parents->idx[sp->role.level-1] = pvec->page[n].idx;
+	}
+
+	return n;
 }
 
-static void mmu_sync_children(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
+void mmu_pages_clear_parents(struct mmu_page_path *parents)
 {
-	struct sync_walker walker = {
-		.walker = { .entry = mmu_sync_fn, },
-		.vcpu = vcpu,
-	};
+	struct kvm_mmu_page *sp;
+	unsigned int level = 0;
 
-	while (mmu_unsync_walk(sp, &walker.walker))
+	do {
+		unsigned int idx = parents->idx[level];
+
+		sp = parents->parent[level];
+		if (!sp)
+			return;
+
+		--sp->unsync_children;
+		WARN_ON((int)sp->unsync_children < 0);
+		__clear_bit(idx, sp->unsync_child_bitmap);
+		level++;
+	} while (level < PT64_ROOT_LEVEL-1 && !sp->unsync_children);
+}
+
+static void kvm_mmu_pages_init(struct kvm_mmu_page *parent,
+			       struct mmu_page_path *parents,
+			       struct kvm_mmu_pages *pvec)
+{
+	parents->parent[parent->role.level-1] = NULL;
+	pvec->nr = 0;
+}
+
+static void mmu_sync_children(struct kvm_vcpu *vcpu,
+			      struct kvm_mmu_page *parent)
+{
+	int i;
+	struct kvm_mmu_page *sp;
+	struct mmu_page_path parents;
+	struct kvm_mmu_pages pages;
+
+	kvm_mmu_pages_init(parent, &parents, &pages);
+	while (mmu_unsync_walk(parent, &pages)) {
+		int protected = 0;
+
+		for_each_sp(pages, sp, parents, i)
+			protected |= rmap_write_protect(vcpu->kvm, sp->gfn);
+
+		if (protected)
+			kvm_flush_remote_tlbs(vcpu->kvm);
+
+		for_each_sp(pages, sp, parents, i) {
+			kvm_sync_page(vcpu, sp);
+			mmu_pages_clear_parents(&parents);
+		}
 		cond_resched_lock(&vcpu->kvm->mmu_lock);
+		kvm_mmu_pages_init(parent, &parents, &pages);
+	}
 }
 
 static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
@@ -1129,7 +1244,8 @@
 	sp->role = role;
 	hlist_add_head(&sp->hash_link, bucket);
 	if (!metaphysical) {
-		rmap_write_protect(vcpu->kvm, gfn);
+		if (rmap_write_protect(vcpu->kvm, gfn))
+			kvm_flush_remote_tlbs(vcpu->kvm);
 		account_shadowed(vcpu->kvm, gfn);
 	}
 	if (shadow_trap_nonpresent_pte != shadow_notrap_nonpresent_pte)
@@ -1153,6 +1269,8 @@
 	if (level == PT32E_ROOT_LEVEL) {
 		shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3];
 		shadow_addr &= PT64_BASE_ADDR_MASK;
+		if (!shadow_addr)
+			return 1;
 		--level;
 	}
 
@@ -1237,33 +1355,29 @@
 	}
 }
 
-struct zap_walker {
-	struct kvm_unsync_walk walker;
-	struct kvm *kvm;
-	int zapped;
-};
-
-static int mmu_zap_fn(struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk)
+static int mmu_zap_unsync_children(struct kvm *kvm,
+				   struct kvm_mmu_page *parent)
 {
-	struct zap_walker *zap_walk = container_of(walk, struct zap_walker,
-						     walker);
-	kvm_mmu_zap_page(zap_walk->kvm, sp);
-	zap_walk->zapped = 1;
-	return 0;
-}
+	int i, zapped = 0;
+	struct mmu_page_path parents;
+	struct kvm_mmu_pages pages;
 
-static int mmu_zap_unsync_children(struct kvm *kvm, struct kvm_mmu_page *sp)
-{
-	struct zap_walker walker = {
-		.walker = { .entry = mmu_zap_fn, },
-		.kvm = kvm,
-		.zapped = 0,
-	};
-
-	if (sp->role.level == PT_PAGE_TABLE_LEVEL)
+	if (parent->role.level == PT_PAGE_TABLE_LEVEL)
 		return 0;
-	mmu_unsync_walk(sp, &walker.walker);
-	return walker.zapped;
+
+	kvm_mmu_pages_init(parent, &parents, &pages);
+	while (mmu_unsync_walk(parent, &pages)) {
+		struct kvm_mmu_page *sp;
+
+		for_each_sp(pages, sp, parents, i) {
+			kvm_mmu_zap_page(kvm, sp);
+			mmu_pages_clear_parents(&parents);
+		}
+		zapped += pages.nr;
+		kvm_mmu_pages_init(parent, &parents, &pages);
+	}
+
+	return zapped;
 }
 
 static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp)
@@ -1362,7 +1476,7 @@
 	int slot = memslot_id(kvm, gfn_to_memslot(kvm, gfn));
 	struct kvm_mmu_page *sp = page_header(__pa(pte));
 
-	__set_bit(slot, &sp->slot_bitmap);
+	__set_bit(slot, sp->slot_bitmap);
 }
 
 static void mmu_convert_notrap(struct kvm_mmu_page *sp)
@@ -1393,6 +1507,110 @@
 	return page;
 }
 
+/*
+ * The function is based on mtrr_type_lookup() in
+ * arch/x86/kernel/cpu/mtrr/generic.c
+ */
+static int get_mtrr_type(struct mtrr_state_type *mtrr_state,
+			 u64 start, u64 end)
+{
+	int i;
+	u64 base, mask;
+	u8 prev_match, curr_match;
+	int num_var_ranges = KVM_NR_VAR_MTRR;
+
+	if (!mtrr_state->enabled)
+		return 0xFF;
+
+	/* Make end inclusive end, instead of exclusive */
+	end--;
+
+	/* Look in fixed ranges. Just return the type as per start */
+	if (mtrr_state->have_fixed && (start < 0x100000)) {
+		int idx;
+
+		if (start < 0x80000) {
+			idx = 0;
+			idx += (start >> 16);
+			return mtrr_state->fixed_ranges[idx];
+		} else if (start < 0xC0000) {
+			idx = 1 * 8;
+			idx += ((start - 0x80000) >> 14);
+			return mtrr_state->fixed_ranges[idx];
+		} else if (start < 0x1000000) {
+			idx = 3 * 8;
+			idx += ((start - 0xC0000) >> 12);
+			return mtrr_state->fixed_ranges[idx];
+		}
+	}
+
+	/*
+	 * Look in variable ranges
+	 * Look of multiple ranges matching this address and pick type
+	 * as per MTRR precedence
+	 */
+	if (!(mtrr_state->enabled & 2))
+		return mtrr_state->def_type;
+
+	prev_match = 0xFF;
+	for (i = 0; i < num_var_ranges; ++i) {
+		unsigned short start_state, end_state;
+
+		if (!(mtrr_state->var_ranges[i].mask_lo & (1 << 11)))
+			continue;
+
+		base = (((u64)mtrr_state->var_ranges[i].base_hi) << 32) +
+		       (mtrr_state->var_ranges[i].base_lo & PAGE_MASK);
+		mask = (((u64)mtrr_state->var_ranges[i].mask_hi) << 32) +
+		       (mtrr_state->var_ranges[i].mask_lo & PAGE_MASK);
+
+		start_state = ((start & mask) == (base & mask));
+		end_state = ((end & mask) == (base & mask));
+		if (start_state != end_state)
+			return 0xFE;
+
+		if ((start & mask) != (base & mask))
+			continue;
+
+		curr_match = mtrr_state->var_ranges[i].base_lo & 0xff;
+		if (prev_match == 0xFF) {
+			prev_match = curr_match;
+			continue;
+		}
+
+		if (prev_match == MTRR_TYPE_UNCACHABLE ||
+		    curr_match == MTRR_TYPE_UNCACHABLE)
+			return MTRR_TYPE_UNCACHABLE;
+
+		if ((prev_match == MTRR_TYPE_WRBACK &&
+		     curr_match == MTRR_TYPE_WRTHROUGH) ||
+		    (prev_match == MTRR_TYPE_WRTHROUGH &&
+		     curr_match == MTRR_TYPE_WRBACK)) {
+			prev_match = MTRR_TYPE_WRTHROUGH;
+			curr_match = MTRR_TYPE_WRTHROUGH;
+		}
+
+		if (prev_match != curr_match)
+			return MTRR_TYPE_UNCACHABLE;
+	}
+
+	if (prev_match != 0xFF)
+		return prev_match;
+
+	return mtrr_state->def_type;
+}
+
+static u8 get_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn)
+{
+	u8 mtrr;
+
+	mtrr = get_mtrr_type(&vcpu->arch.mtrr_state, gfn << PAGE_SHIFT,
+			     (gfn << PAGE_SHIFT) + PAGE_SIZE);
+	if (mtrr == 0xfe || mtrr == 0xff)
+		mtrr = MTRR_TYPE_WRBACK;
+	return mtrr;
+}
+
 static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 {
 	unsigned index;
@@ -1409,9 +1627,15 @@
 		if (s->role.word != sp->role.word)
 			return 1;
 	}
-	kvm_mmu_mark_parents_unsync(vcpu, sp);
 	++vcpu->kvm->stat.mmu_unsync;
 	sp->unsync = 1;
+
+	if (sp->global) {
+		list_add(&sp->oos_link, &vcpu->kvm->arch.oos_global_pages);
+		++vcpu->kvm->stat.mmu_unsync_global;
+	} else
+		kvm_mmu_mark_parents_unsync(vcpu, sp);
+
 	mmu_convert_notrap(sp);
 	return 0;
 }
@@ -1437,11 +1661,24 @@
 static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
 		    unsigned pte_access, int user_fault,
 		    int write_fault, int dirty, int largepage,
-		    gfn_t gfn, pfn_t pfn, bool speculative,
+		    int global, gfn_t gfn, pfn_t pfn, bool speculative,
 		    bool can_unsync)
 {
 	u64 spte;
 	int ret = 0;
+	u64 mt_mask = shadow_mt_mask;
+	struct kvm_mmu_page *sp = page_header(__pa(shadow_pte));
+
+	if (!(vcpu->arch.cr4 & X86_CR4_PGE))
+		global = 0;
+	if (!global && sp->global) {
+		sp->global = 0;
+		if (sp->unsync) {
+			kvm_unlink_unsync_global(vcpu->kvm, sp);
+			kvm_mmu_mark_parents_unsync(vcpu, sp);
+		}
+	}
+
 	/*
 	 * We don't set the accessed bit, since we sometimes want to see
 	 * whether the guest actually used the pte (in order to detect
@@ -1460,6 +1697,11 @@
 		spte |= shadow_user_mask;
 	if (largepage)
 		spte |= PT_PAGE_SIZE_MASK;
+	if (mt_mask) {
+		mt_mask = get_memory_type(vcpu, gfn) <<
+			  kvm_x86_ops->get_mt_mask_shift();
+		spte |= mt_mask;
+	}
 
 	spte |= (u64)pfn << PAGE_SHIFT;
 
@@ -1474,6 +1716,15 @@
 
 		spte |= PT_WRITABLE_MASK;
 
+		/*
+		 * Optimization: for pte sync, if spte was writable the hash
+		 * lookup is unnecessary (and expensive). Write protection
+		 * is responsibility of mmu_get_page / kvm_sync_page.
+		 * Same reasoning can be applied to dirty page accounting.
+		 */
+		if (!can_unsync && is_writeble_pte(*shadow_pte))
+			goto set_pte;
+
 		if (mmu_need_write_protect(vcpu, gfn, can_unsync)) {
 			pgprintk("%s: found shadow page for %lx, marking ro\n",
 				 __func__, gfn);
@@ -1495,8 +1746,8 @@
 static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
 			 unsigned pt_access, unsigned pte_access,
 			 int user_fault, int write_fault, int dirty,
-			 int *ptwrite, int largepage, gfn_t gfn,
-			 pfn_t pfn, bool speculative)
+			 int *ptwrite, int largepage, int global,
+			 gfn_t gfn, pfn_t pfn, bool speculative)
 {
 	int was_rmapped = 0;
 	int was_writeble = is_writeble_pte(*shadow_pte);
@@ -1529,7 +1780,7 @@
 		}
 	}
 	if (set_spte(vcpu, shadow_pte, pte_access, user_fault, write_fault,
-		      dirty, largepage, gfn, pfn, speculative, true)) {
+		      dirty, largepage, global, gfn, pfn, speculative, true)) {
 		if (write_fault)
 			*ptwrite = 1;
 		kvm_x86_ops->tlb_flush(vcpu);
@@ -1586,7 +1837,7 @@
 	    || (walk->largepage && level == PT_DIRECTORY_LEVEL)) {
 		mmu_set_spte(vcpu, sptep, ACC_ALL, ACC_ALL,
 			     0, walk->write, 1, &walk->pt_write,
-			     walk->largepage, gfn, walk->pfn, false);
+			     walk->largepage, 0, gfn, walk->pfn, false);
 		++vcpu->stat.pf_fixed;
 		return 1;
 	}
@@ -1773,6 +2024,15 @@
 	}
 }
 
+static void mmu_sync_global(struct kvm_vcpu *vcpu)
+{
+	struct kvm *kvm = vcpu->kvm;
+	struct kvm_mmu_page *sp, *n;
+
+	list_for_each_entry_safe(sp, n, &kvm->arch.oos_global_pages, oos_link)
+		kvm_sync_page(vcpu, sp);
+}
+
 void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu)
 {
 	spin_lock(&vcpu->kvm->mmu_lock);
@@ -1780,6 +2040,13 @@
 	spin_unlock(&vcpu->kvm->mmu_lock);
 }
 
+void kvm_mmu_sync_global(struct kvm_vcpu *vcpu)
+{
+	spin_lock(&vcpu->kvm->mmu_lock);
+	mmu_sync_global(vcpu);
+	spin_unlock(&vcpu->kvm->mmu_lock);
+}
+
 static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr)
 {
 	return vaddr;
@@ -2178,7 +2445,8 @@
 }
 
 void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
-		       const u8 *new, int bytes)
+		       const u8 *new, int bytes,
+		       bool guest_initiated)
 {
 	gfn_t gfn = gpa >> PAGE_SHIFT;
 	struct kvm_mmu_page *sp;
@@ -2204,15 +2472,17 @@
 	kvm_mmu_free_some_pages(vcpu);
 	++vcpu->kvm->stat.mmu_pte_write;
 	kvm_mmu_audit(vcpu, "pre pte write");
-	if (gfn == vcpu->arch.last_pt_write_gfn
-	    && !last_updated_pte_accessed(vcpu)) {
-		++vcpu->arch.last_pt_write_count;
-		if (vcpu->arch.last_pt_write_count >= 3)
-			flooded = 1;
-	} else {
-		vcpu->arch.last_pt_write_gfn = gfn;
-		vcpu->arch.last_pt_write_count = 1;
-		vcpu->arch.last_pte_updated = NULL;
+	if (guest_initiated) {
+		if (gfn == vcpu->arch.last_pt_write_gfn
+		    && !last_updated_pte_accessed(vcpu)) {
+			++vcpu->arch.last_pt_write_count;
+			if (vcpu->arch.last_pt_write_count >= 3)
+				flooded = 1;
+		} else {
+			vcpu->arch.last_pt_write_gfn = gfn;
+			vcpu->arch.last_pt_write_count = 1;
+			vcpu->arch.last_pte_updated = NULL;
+		}
 	}
 	index = kvm_page_table_hashfn(gfn);
 	bucket = &vcpu->kvm->arch.mmu_page_hash[index];
@@ -2352,9 +2622,7 @@
 
 void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva)
 {
-	spin_lock(&vcpu->kvm->mmu_lock);
 	vcpu->arch.mmu.invlpg(vcpu, gva);
-	spin_unlock(&vcpu->kvm->mmu_lock);
 	kvm_mmu_flush_tlb(vcpu);
 	++vcpu->stat.invlpg;
 }
@@ -2451,7 +2719,7 @@
 		int i;
 		u64 *pt;
 
-		if (!test_bit(slot, &sp->slot_bitmap))
+		if (!test_bit(slot, sp->slot_bitmap))
 			continue;
 
 		pt = sp->spt;
@@ -2860,8 +3128,8 @@
 		if (sp->role.metaphysical)
 			continue;
 
-		slot = gfn_to_memslot(vcpu->kvm, sp->gfn);
 		gfn = unalias_gfn(vcpu->kvm, sp->gfn);
+		slot = gfn_to_memslot_unaliased(vcpu->kvm, sp->gfn);
 		rmapp = &slot->rmap[gfn - slot->base_gfn];
 		if (*rmapp)
 			printk(KERN_ERR "%s: (%s) shadow page has writable"
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 84eee43..9fd78b6 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -82,6 +82,7 @@
 	int *ptwrite;
 	pfn_t pfn;
 	u64 *sptep;
+	gpa_t pte_gpa;
 };
 
 static gfn_t gpte_to_gfn(pt_element_t gpte)
@@ -222,7 +223,7 @@
 		if (ret)
 			goto walk;
 		pte |= PT_DIRTY_MASK;
-		kvm_mmu_pte_write(vcpu, pte_gpa, (u8 *)&pte, sizeof(pte));
+		kvm_mmu_pte_write(vcpu, pte_gpa, (u8 *)&pte, sizeof(pte), 0);
 		walker->ptes[walker->level - 1] = pte;
 	}
 
@@ -274,7 +275,8 @@
 		return;
 	kvm_get_pfn(pfn);
 	mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
-		     gpte & PT_DIRTY_MASK, NULL, largepage, gpte_to_gfn(gpte),
+		     gpte & PT_DIRTY_MASK, NULL, largepage,
+		     gpte & PT_GLOBAL_MASK, gpte_to_gfn(gpte),
 		     pfn, true);
 }
 
@@ -301,8 +303,9 @@
 		mmu_set_spte(vcpu, sptep, access, gw->pte_access & access,
 			     sw->user_fault, sw->write_fault,
 			     gw->ptes[gw->level-1] & PT_DIRTY_MASK,
-			     sw->ptwrite, sw->largepage, gw->gfn, sw->pfn,
-			     false);
+			     sw->ptwrite, sw->largepage,
+			     gw->ptes[gw->level-1] & PT_GLOBAL_MASK,
+			     gw->gfn, sw->pfn, false);
 		sw->sptep = sptep;
 		return 1;
 	}
@@ -466,10 +469,22 @@
 				      struct kvm_vcpu *vcpu, u64 addr,
 				      u64 *sptep, int level)
 {
+	struct shadow_walker *sw =
+		container_of(_sw, struct shadow_walker, walker);
 
-	if (level == PT_PAGE_TABLE_LEVEL) {
-		if (is_shadow_present_pte(*sptep))
+	/* FIXME: properly handle invlpg on large guest pages */
+	if (level == PT_PAGE_TABLE_LEVEL ||
+	    ((level == PT_DIRECTORY_LEVEL) && is_large_pte(*sptep))) {
+		struct kvm_mmu_page *sp = page_header(__pa(sptep));
+
+		sw->pte_gpa = (sp->gfn << PAGE_SHIFT);
+		sw->pte_gpa += (sptep - sp->spt) * sizeof(pt_element_t);
+
+		if (is_shadow_present_pte(*sptep)) {
 			rmap_remove(vcpu->kvm, sptep);
+			if (is_large_pte(*sptep))
+				--vcpu->kvm->stat.lpages;
+		}
 		set_shadow_pte(sptep, shadow_trap_nonpresent_pte);
 		return 1;
 	}
@@ -480,11 +495,26 @@
 
 static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva)
 {
+	pt_element_t gpte;
 	struct shadow_walker walker = {
 		.walker = { .entry = FNAME(shadow_invlpg_entry), },
+		.pte_gpa = -1,
 	};
 
+	spin_lock(&vcpu->kvm->mmu_lock);
 	walk_shadow(&walker.walker, vcpu, gva);
+	spin_unlock(&vcpu->kvm->mmu_lock);
+	if (walker.pte_gpa == -1)
+		return;
+	if (kvm_read_guest_atomic(vcpu->kvm, walker.pte_gpa, &gpte,
+				  sizeof(pt_element_t)))
+		return;
+	if (is_present_pte(gpte) && (gpte & PT_ACCESSED_MASK)) {
+		if (mmu_topup_memory_caches(vcpu))
+			return;
+		kvm_mmu_pte_write(vcpu, walker.pte_gpa, (const u8 *)&gpte,
+				  sizeof(pt_element_t), 0);
+	}
 }
 
 static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr)
@@ -580,7 +610,7 @@
 		nr_present++;
 		pte_access = sp->role.access & FNAME(gpte_access)(vcpu, gpte);
 		set_spte(vcpu, &sp->spt[i], pte_access, 0, 0,
-			 is_dirty_pte(gpte), 0, gfn,
+			 is_dirty_pte(gpte), 0, gpte & PT_GLOBAL_MASK, gfn,
 			 spte_to_pfn(sp->spt[i]), true, false);
 	}
 
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 9c4ce65..1452851 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -28,6 +28,8 @@
 
 #include <asm/desc.h>
 
+#include <asm/virtext.h>
+
 #define __ex(x) __kvm_handle_fault_on_reboot(x)
 
 MODULE_AUTHOR("Qumranet");
@@ -245,34 +247,19 @@
 
 static int has_svm(void)
 {
-	uint32_t eax, ebx, ecx, edx;
+	const char *msg;
 
-	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) {
-		printk(KERN_INFO "has_svm: not amd\n");
+	if (!cpu_has_svm(&msg)) {
+		printk(KERN_INFO "has_svn: %s\n", msg);
 		return 0;
 	}
 
-	cpuid(0x80000000, &eax, &ebx, &ecx, &edx);
-	if (eax < SVM_CPUID_FUNC) {
-		printk(KERN_INFO "has_svm: can't execute cpuid_8000000a\n");
-		return 0;
-	}
-
-	cpuid(0x80000001, &eax, &ebx, &ecx, &edx);
-	if (!(ecx & (1 << SVM_CPUID_FEATURE_SHIFT))) {
-		printk(KERN_DEBUG "has_svm: svm not available\n");
-		return 0;
-	}
 	return 1;
 }
 
 static void svm_hardware_disable(void *garbage)
 {
-	uint64_t efer;
-
-	wrmsrl(MSR_VM_HSAVE_PA, 0);
-	rdmsrl(MSR_EFER, efer);
-	wrmsrl(MSR_EFER, efer & ~MSR_EFER_SVME_MASK);
+	cpu_svm_disable();
 }
 
 static void svm_hardware_enable(void *garbage)
@@ -772,6 +759,22 @@
 	var->l = (s->attrib >> SVM_SELECTOR_L_SHIFT) & 1;
 	var->db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1;
 	var->g = (s->attrib >> SVM_SELECTOR_G_SHIFT) & 1;
+
+	/*
+	 * SVM always stores 0 for the 'G' bit in the CS selector in
+	 * the VMCB on a VMEXIT. This hurts cross-vendor migration:
+	 * Intel's VMENTRY has a check on the 'G' bit.
+	 */
+	if (seg == VCPU_SREG_CS)
+		var->g = s->limit > 0xfffff;
+
+	/*
+	 * Work around a bug where the busy flag in the tr selector
+	 * isn't exposed
+	 */
+	if (seg == VCPU_SREG_TR)
+		var->type |= 0x2;
+
 	var->unusable = !var->present;
 }
 
@@ -1099,6 +1102,7 @@
 	rep = (io_info & SVM_IOIO_REP_MASK) != 0;
 	down = (svm->vmcb->save.rflags & X86_EFLAGS_DF) != 0;
 
+	skip_emulated_instruction(&svm->vcpu);
 	return kvm_emulate_pio(&svm->vcpu, kvm_run, in, size, port);
 }
 
@@ -1912,6 +1916,11 @@
 #endif
 }
 
+static int svm_get_mt_mask_shift(void)
+{
+	return 0;
+}
+
 static struct kvm_x86_ops svm_x86_ops = {
 	.cpu_has_kvm_support = has_svm,
 	.disabled_by_bios = is_disabled,
@@ -1967,6 +1976,7 @@
 
 	.set_tss_addr = svm_set_tss_addr,
 	.get_tdp_level = get_npt_level,
+	.get_mt_mask_shift = svm_get_mt_mask_shift,
 };
 
 static int __init svm_init(void)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index a4018b0..6259d74 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -16,7 +16,6 @@
  */
 
 #include "irq.h"
-#include "vmx.h"
 #include "mmu.h"
 
 #include <linux/kvm_host.h>
@@ -31,6 +30,8 @@
 
 #include <asm/io.h>
 #include <asm/desc.h>
+#include <asm/vmx.h>
+#include <asm/virtext.h>
 
 #define __ex(x) __kvm_handle_fault_on_reboot(x)
 
@@ -90,6 +91,11 @@
 	} rmode;
 	int vpid;
 	bool emulation_required;
+
+	/* Support for vnmi-less CPUs */
+	int soft_vnmi_blocked;
+	ktime_t entry_time;
+	s64 vnmi_blocked_time;
 };
 
 static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu)
@@ -122,7 +128,7 @@
 	u32 vmentry_ctrl;
 } vmcs_config;
 
-struct vmx_capability {
+static struct vmx_capability {
 	u32 ept;
 	u32 vpid;
 } vmx_capability;
@@ -957,6 +963,13 @@
 		pr_unimpl(vcpu, "unimplemented perfctr wrmsr: 0x%x data 0x%llx\n", msr_index, data);
 
 		break;
+	case MSR_IA32_CR_PAT:
+		if (vmcs_config.vmentry_ctrl & VM_ENTRY_LOAD_IA32_PAT) {
+			vmcs_write64(GUEST_IA32_PAT, data);
+			vcpu->arch.pat = data;
+			break;
+		}
+		/* Otherwise falls through to kvm_set_msr_common */
 	default:
 		vmx_load_host_state(vmx);
 		msr = find_msr_entry(vmx, msr_index);
@@ -1032,8 +1045,7 @@
 
 static __init int cpu_has_kvm_support(void)
 {
-	unsigned long ecx = cpuid_ecx(1);
-	return test_bit(5, &ecx); /* CPUID.1:ECX.VMX[bit 5] -> VT */
+	return cpu_has_vmx();
 }
 
 static __init int vmx_disabled_by_bios(void)
@@ -1079,11 +1091,20 @@
 		__vcpu_clear(vmx);
 }
 
+
+/* Just like cpu_vmxoff(), but with the __kvm_handle_fault_on_reboot()
+ * tricks.
+ */
+static void kvm_cpu_vmxoff(void)
+{
+	asm volatile (__ex(ASM_VMX_VMXOFF) : : : "cc");
+	write_cr4(read_cr4() & ~X86_CR4_VMXE);
+}
+
 static void hardware_disable(void *garbage)
 {
 	vmclear_local_vcpus();
-	asm volatile (__ex(ASM_VMX_VMXOFF) : : : "cc");
-	write_cr4(read_cr4() & ~X86_CR4_VMXE);
+	kvm_cpu_vmxoff();
 }
 
 static __init int adjust_vmx_controls(u32 ctl_min, u32 ctl_opt,
@@ -1176,12 +1197,13 @@
 #ifdef CONFIG_X86_64
 	min |= VM_EXIT_HOST_ADDR_SPACE_SIZE;
 #endif
-	opt = 0;
+	opt = VM_EXIT_SAVE_IA32_PAT | VM_EXIT_LOAD_IA32_PAT;
 	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_EXIT_CTLS,
 				&_vmexit_control) < 0)
 		return -EIO;
 
-	min = opt = 0;
+	min = 0;
+	opt = VM_ENTRY_LOAD_IA32_PAT;
 	if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_ENTRY_CTLS,
 				&_vmentry_control) < 0)
 		return -EIO;
@@ -2087,8 +2109,9 @@
  */
 static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
 {
-	u32 host_sysenter_cs;
+	u32 host_sysenter_cs, msr_low, msr_high;
 	u32 junk;
+	u64 host_pat;
 	unsigned long a;
 	struct descriptor_table dt;
 	int i;
@@ -2176,6 +2199,20 @@
 	rdmsrl(MSR_IA32_SYSENTER_EIP, a);
 	vmcs_writel(HOST_IA32_SYSENTER_EIP, a);   /* 22.2.3 */
 
+	if (vmcs_config.vmexit_ctrl & VM_EXIT_LOAD_IA32_PAT) {
+		rdmsr(MSR_IA32_CR_PAT, msr_low, msr_high);
+		host_pat = msr_low | ((u64) msr_high << 32);
+		vmcs_write64(HOST_IA32_PAT, host_pat);
+	}
+	if (vmcs_config.vmentry_ctrl & VM_ENTRY_LOAD_IA32_PAT) {
+		rdmsr(MSR_IA32_CR_PAT, msr_low, msr_high);
+		host_pat = msr_low | ((u64) msr_high << 32);
+		/* Write the default value follow host pat */
+		vmcs_write64(GUEST_IA32_PAT, host_pat);
+		/* Keep arch.pat sync with GUEST_IA32_PAT */
+		vmx->vcpu.arch.pat = host_pat;
+	}
+
 	for (i = 0; i < NR_VMX_MSR; ++i) {
 		u32 index = vmx_msr_index[i];
 		u32 data_low, data_high;
@@ -2230,6 +2267,8 @@
 
 	vmx->vcpu.arch.rmode.active = 0;
 
+	vmx->soft_vnmi_blocked = 0;
+
 	vmx->vcpu.arch.regs[VCPU_REGS_RDX] = get_rdx_init_val();
 	kvm_set_cr8(&vmx->vcpu, 0);
 	msr = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
@@ -2335,6 +2374,29 @@
 	return ret;
 }
 
+static void enable_irq_window(struct kvm_vcpu *vcpu)
+{
+	u32 cpu_based_vm_exec_control;
+
+	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
+	cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING;
+	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
+}
+
+static void enable_nmi_window(struct kvm_vcpu *vcpu)
+{
+	u32 cpu_based_vm_exec_control;
+
+	if (!cpu_has_virtual_nmis()) {
+		enable_irq_window(vcpu);
+		return;
+	}
+
+	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
+	cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_NMI_PENDING;
+	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
+}
+
 static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
@@ -2358,10 +2420,54 @@
 
 static void vmx_inject_nmi(struct kvm_vcpu *vcpu)
 {
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (!cpu_has_virtual_nmis()) {
+		/*
+		 * Tracking the NMI-blocked state in software is built upon
+		 * finding the next open IRQ window. This, in turn, depends on
+		 * well-behaving guests: They have to keep IRQs disabled at
+		 * least as long as the NMI handler runs. Otherwise we may
+		 * cause NMI nesting, maybe breaking the guest. But as this is
+		 * highly unlikely, we can live with the residual risk.
+		 */
+		vmx->soft_vnmi_blocked = 1;
+		vmx->vnmi_blocked_time = 0;
+	}
+
+	++vcpu->stat.nmi_injections;
+	if (vcpu->arch.rmode.active) {
+		vmx->rmode.irq.pending = true;
+		vmx->rmode.irq.vector = NMI_VECTOR;
+		vmx->rmode.irq.rip = kvm_rip_read(vcpu);
+		vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
+			     NMI_VECTOR | INTR_TYPE_SOFT_INTR |
+			     INTR_INFO_VALID_MASK);
+		vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1);
+		kvm_rip_write(vcpu, vmx->rmode.irq.rip - 1);
+		return;
+	}
 	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
 			INTR_TYPE_NMI_INTR | INTR_INFO_VALID_MASK | NMI_VECTOR);
 }
 
+static void vmx_update_window_states(struct kvm_vcpu *vcpu)
+{
+	u32 guest_intr = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
+
+	vcpu->arch.nmi_window_open =
+		!(guest_intr & (GUEST_INTR_STATE_STI |
+				GUEST_INTR_STATE_MOV_SS |
+				GUEST_INTR_STATE_NMI));
+	if (!cpu_has_virtual_nmis() && to_vmx(vcpu)->soft_vnmi_blocked)
+		vcpu->arch.nmi_window_open = 0;
+
+	vcpu->arch.interrupt_window_open =
+		((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) &&
+		 !(guest_intr & (GUEST_INTR_STATE_STI |
+				 GUEST_INTR_STATE_MOV_SS)));
+}
+
 static void kvm_do_inject_irq(struct kvm_vcpu *vcpu)
 {
 	int word_index = __ffs(vcpu->arch.irq_summary);
@@ -2374,40 +2480,49 @@
 	kvm_queue_interrupt(vcpu, irq);
 }
 
-
 static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 				       struct kvm_run *kvm_run)
 {
-	u32 cpu_based_vm_exec_control;
+	vmx_update_window_states(vcpu);
 
-	vcpu->arch.interrupt_window_open =
-		((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) &&
-		 (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0);
+	if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) {
+		if (vcpu->arch.interrupt.pending) {
+			enable_nmi_window(vcpu);
+		} else if (vcpu->arch.nmi_window_open) {
+			vcpu->arch.nmi_pending = false;
+			vcpu->arch.nmi_injected = true;
+		} else {
+			enable_nmi_window(vcpu);
+			return;
+		}
+	}
+	if (vcpu->arch.nmi_injected) {
+		vmx_inject_nmi(vcpu);
+		if (vcpu->arch.nmi_pending)
+			enable_nmi_window(vcpu);
+		else if (vcpu->arch.irq_summary
+			 || kvm_run->request_interrupt_window)
+			enable_irq_window(vcpu);
+		return;
+	}
 
-	if (vcpu->arch.interrupt_window_open &&
-	    vcpu->arch.irq_summary && !vcpu->arch.interrupt.pending)
-		kvm_do_inject_irq(vcpu);
+	if (vcpu->arch.interrupt_window_open) {
+		if (vcpu->arch.irq_summary && !vcpu->arch.interrupt.pending)
+			kvm_do_inject_irq(vcpu);
 
-	if (vcpu->arch.interrupt_window_open && vcpu->arch.interrupt.pending)
-		vmx_inject_irq(vcpu, vcpu->arch.interrupt.nr);
-
-	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
+		if (vcpu->arch.interrupt.pending)
+			vmx_inject_irq(vcpu, vcpu->arch.interrupt.nr);
+	}
 	if (!vcpu->arch.interrupt_window_open &&
 	    (vcpu->arch.irq_summary || kvm_run->request_interrupt_window))
-		/*
-		 * Interrupts blocked.  Wait for unblock.
-		 */
-		cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING;
-	else
-		cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_INTR_PENDING;
-	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
+		enable_irq_window(vcpu);
 }
 
 static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
 {
 	int ret;
 	struct kvm_userspace_memory_region tss_mem = {
-		.slot = 8,
+		.slot = TSS_PRIVATE_MEMSLOT,
 		.guest_phys_addr = addr,
 		.memory_size = PAGE_SIZE * 3,
 		.flags = 0,
@@ -2492,7 +2607,7 @@
 		set_bit(irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
 	}
 
-	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) /* nmi */
+	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == INTR_TYPE_NMI_INTR)
 		return 1;  /* already handled by vmx_vcpu_run() */
 
 	if (is_no_device(intr_info)) {
@@ -2581,6 +2696,7 @@
 	rep = (exit_qualification & 32) != 0;
 	port = exit_qualification >> 16;
 
+	skip_emulated_instruction(vcpu);
 	return kvm_emulate_pio(vcpu, kvm_run, in, size, port);
 }
 
@@ -2767,6 +2883,7 @@
 	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
 
 	KVMTRACE_0D(PEND_INTR, vcpu, handler);
+	++vcpu->stat.irq_window_exits;
 
 	/*
 	 * If the user space waits to inject interrupts, exit as soon as
@@ -2775,7 +2892,6 @@
 	if (kvm_run->request_interrupt_window &&
 	    !vcpu->arch.irq_summary) {
 		kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
-		++vcpu->stat.irq_window_exits;
 		return 0;
 	}
 	return 1;
@@ -2832,6 +2948,7 @@
 
 static int handle_task_switch(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	unsigned long exit_qualification;
 	u16 tss_selector;
 	int reason;
@@ -2839,6 +2956,15 @@
 	exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
 
 	reason = (u32)exit_qualification >> 30;
+	if (reason == TASK_SWITCH_GATE && vmx->vcpu.arch.nmi_injected &&
+	    (vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK) &&
+	    (vmx->idt_vectoring_info & VECTORING_INFO_TYPE_MASK)
+	    == INTR_TYPE_NMI_INTR) {
+		vcpu->arch.nmi_injected = false;
+		if (cpu_has_virtual_nmis())
+			vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO,
+				      GUEST_INTR_STATE_NMI);
+	}
 	tss_selector = exit_qualification;
 
 	return kvm_task_switch(vcpu, tss_selector, reason);
@@ -2927,16 +3053,12 @@
 	while (!guest_state_valid(vcpu)) {
 		err = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
 
-		switch (err) {
-			case EMULATE_DONE:
-				break;
-			case EMULATE_DO_MMIO:
-				kvm_report_emulation_failure(vcpu, "mmio");
-				/* TODO: Handle MMIO */
-				return;
-			default:
-				kvm_report_emulation_failure(vcpu, "emulation failure");
-				return;
+		if (err == EMULATE_DO_MMIO)
+			break;
+
+		if (err != EMULATE_DONE) {
+			kvm_report_emulation_failure(vcpu, "emulation failure");
+			return;
 		}
 
 		if (signal_pending(current))
@@ -2948,8 +3070,10 @@
 	local_irq_disable();
 	preempt_disable();
 
-	/* Guest state should be valid now, no more emulation should be needed */
-	vmx->emulation_required = 0;
+	/* Guest state should be valid now except if we need to
+	 * emulate an MMIO */
+	if (guest_state_valid(vcpu))
+		vmx->emulation_required = 0;
 }
 
 /*
@@ -2996,6 +3120,11 @@
 	KVMTRACE_3D(VMEXIT, vcpu, exit_reason, (u32)kvm_rip_read(vcpu),
 		    (u32)((u64)kvm_rip_read(vcpu) >> 32), entryexit);
 
+	/* If we need to emulate an MMIO from handle_invalid_guest_state
+	 * we just return 0 */
+	if (vmx->emulation_required && emulate_invalid_guest_state)
+		return 0;
+
 	/* Access CR3 don't cause VMExit in paging mode, so we need
 	 * to sync with guest real CR3. */
 	if (vm_need_ept() && is_paging(vcpu)) {
@@ -3012,9 +3141,32 @@
 
 	if ((vectoring_info & VECTORING_INFO_VALID_MASK) &&
 			(exit_reason != EXIT_REASON_EXCEPTION_NMI &&
-			exit_reason != EXIT_REASON_EPT_VIOLATION))
-		printk(KERN_WARNING "%s: unexpected, valid vectoring info and "
-		       "exit reason is 0x%x\n", __func__, exit_reason);
+			exit_reason != EXIT_REASON_EPT_VIOLATION &&
+			exit_reason != EXIT_REASON_TASK_SWITCH))
+		printk(KERN_WARNING "%s: unexpected, valid vectoring info "
+		       "(0x%x) and exit reason is 0x%x\n",
+		       __func__, vectoring_info, exit_reason);
+
+	if (unlikely(!cpu_has_virtual_nmis() && vmx->soft_vnmi_blocked)) {
+		if (vcpu->arch.interrupt_window_open) {
+			vmx->soft_vnmi_blocked = 0;
+			vcpu->arch.nmi_window_open = 1;
+		} else if (vmx->vnmi_blocked_time > 1000000000LL &&
+			   vcpu->arch.nmi_pending) {
+			/*
+			 * This CPU don't support us in finding the end of an
+			 * NMI-blocked window if the guest runs with IRQs
+			 * disabled. So we pull the trigger after 1 s of
+			 * futile waiting, but inform the user about this.
+			 */
+			printk(KERN_WARNING "%s: Breaking out of NMI-blocked "
+			       "state on VCPU %d after 1 s timeout\n",
+			       __func__, vcpu->vcpu_id);
+			vmx->soft_vnmi_blocked = 0;
+			vmx->vcpu.arch.nmi_window_open = 1;
+		}
+	}
+
 	if (exit_reason < kvm_vmx_max_exit_handlers
 	    && kvm_vmx_exit_handlers[exit_reason])
 		return kvm_vmx_exit_handlers[exit_reason](vcpu, kvm_run);
@@ -3042,51 +3194,6 @@
 	vmcs_write32(TPR_THRESHOLD, (max_irr > tpr) ? tpr >> 4 : max_irr >> 4);
 }
 
-static void enable_irq_window(struct kvm_vcpu *vcpu)
-{
-	u32 cpu_based_vm_exec_control;
-
-	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
-	cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING;
-	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
-}
-
-static void enable_nmi_window(struct kvm_vcpu *vcpu)
-{
-	u32 cpu_based_vm_exec_control;
-
-	if (!cpu_has_virtual_nmis())
-		return;
-
-	cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
-	cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_NMI_PENDING;
-	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
-}
-
-static int vmx_nmi_enabled(struct kvm_vcpu *vcpu)
-{
-	u32 guest_intr = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
-	return !(guest_intr & (GUEST_INTR_STATE_NMI |
-			       GUEST_INTR_STATE_MOV_SS |
-			       GUEST_INTR_STATE_STI));
-}
-
-static int vmx_irq_enabled(struct kvm_vcpu *vcpu)
-{
-	u32 guest_intr = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
-	return (!(guest_intr & (GUEST_INTR_STATE_MOV_SS |
-			       GUEST_INTR_STATE_STI)) &&
-		(vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF));
-}
-
-static void enable_intr_window(struct kvm_vcpu *vcpu)
-{
-	if (vcpu->arch.nmi_pending)
-		enable_nmi_window(vcpu);
-	else if (kvm_cpu_has_interrupt(vcpu))
-		enable_irq_window(vcpu);
-}
-
 static void vmx_complete_interrupts(struct vcpu_vmx *vmx)
 {
 	u32 exit_intr_info;
@@ -3109,7 +3216,9 @@
 		if (unblock_nmi && vector != DF_VECTOR)
 			vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO,
 				      GUEST_INTR_STATE_NMI);
-	}
+	} else if (unlikely(vmx->soft_vnmi_blocked))
+		vmx->vnmi_blocked_time +=
+			ktime_to_ns(ktime_sub(ktime_get(), vmx->entry_time));
 
 	idt_vectoring_info = vmx->idt_vectoring_info;
 	idtv_info_valid = idt_vectoring_info & VECTORING_INFO_VALID_MASK;
@@ -3147,26 +3256,29 @@
 {
 	update_tpr_threshold(vcpu);
 
-	if (cpu_has_virtual_nmis()) {
-		if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) {
-			if (vcpu->arch.interrupt.pending) {
-				enable_nmi_window(vcpu);
-			} else if (vmx_nmi_enabled(vcpu)) {
-				vcpu->arch.nmi_pending = false;
-				vcpu->arch.nmi_injected = true;
-			} else {
-				enable_intr_window(vcpu);
-				return;
-			}
-		}
-		if (vcpu->arch.nmi_injected) {
-			vmx_inject_nmi(vcpu);
-			enable_intr_window(vcpu);
+	vmx_update_window_states(vcpu);
+
+	if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) {
+		if (vcpu->arch.interrupt.pending) {
+			enable_nmi_window(vcpu);
+		} else if (vcpu->arch.nmi_window_open) {
+			vcpu->arch.nmi_pending = false;
+			vcpu->arch.nmi_injected = true;
+		} else {
+			enable_nmi_window(vcpu);
 			return;
 		}
 	}
+	if (vcpu->arch.nmi_injected) {
+		vmx_inject_nmi(vcpu);
+		if (vcpu->arch.nmi_pending)
+			enable_nmi_window(vcpu);
+		else if (kvm_cpu_has_interrupt(vcpu))
+			enable_irq_window(vcpu);
+		return;
+	}
 	if (!vcpu->arch.interrupt.pending && kvm_cpu_has_interrupt(vcpu)) {
-		if (vmx_irq_enabled(vcpu))
+		if (vcpu->arch.interrupt_window_open)
 			kvm_queue_interrupt(vcpu, kvm_cpu_get_interrupt(vcpu));
 		else
 			enable_irq_window(vcpu);
@@ -3174,6 +3286,8 @@
 	if (vcpu->arch.interrupt.pending) {
 		vmx_inject_irq(vcpu, vcpu->arch.interrupt.nr);
 		kvm_timer_intr_post(vcpu, vcpu->arch.interrupt.nr);
+		if (kvm_cpu_has_interrupt(vcpu))
+			enable_irq_window(vcpu);
 	}
 }
 
@@ -3213,6 +3327,10 @@
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	u32 intr_info;
 
+	/* Record the guest's net vcpu time for enforced NMI injections. */
+	if (unlikely(!cpu_has_virtual_nmis() && vmx->soft_vnmi_blocked))
+		vmx->entry_time = ktime_get();
+
 	/* Handle invalid guest state instead of entering VMX */
 	if (vmx->emulation_required && emulate_invalid_guest_state) {
 		handle_invalid_guest_state(vcpu, kvm_run);
@@ -3327,9 +3445,7 @@
 	if (vmx->rmode.irq.pending)
 		fixup_rmode_irq(vmx);
 
-	vcpu->arch.interrupt_window_open =
-		(vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) &
-		 (GUEST_INTR_STATE_STI | GUEST_INTR_STATE_MOV_SS)) == 0;
+	vmx_update_window_states(vcpu);
 
 	asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS));
 	vmx->launched = 1;
@@ -3337,7 +3453,7 @@
 	intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
 
 	/* We need to handle NMIs before interrupts are enabled */
-	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200 &&
+	if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == INTR_TYPE_NMI_INTR &&
 	    (intr_info & INTR_INFO_VALID_MASK)) {
 		KVMTRACE_0D(NMI, vcpu, handler);
 		asm("int $2");
@@ -3455,6 +3571,11 @@
 	return VMX_EPT_DEFAULT_GAW + 1;
 }
 
+static int vmx_get_mt_mask_shift(void)
+{
+	return VMX_EPT_MT_EPTE_SHIFT;
+}
+
 static struct kvm_x86_ops vmx_x86_ops = {
 	.cpu_has_kvm_support = cpu_has_kvm_support,
 	.disabled_by_bios = vmx_disabled_by_bios,
@@ -3510,6 +3631,7 @@
 
 	.set_tss_addr = vmx_set_tss_addr,
 	.get_tdp_level = get_ept_level,
+	.get_mt_mask_shift = vmx_get_mt_mask_shift,
 };
 
 static int __init vmx_init(void)
@@ -3566,10 +3688,10 @@
 		bypass_guest_pf = 0;
 		kvm_mmu_set_base_ptes(VMX_EPT_READABLE_MASK |
 			VMX_EPT_WRITABLE_MASK |
-			VMX_EPT_DEFAULT_MT << VMX_EPT_MT_EPTE_SHIFT |
 			VMX_EPT_IGMT_BIT);
 		kvm_mmu_set_mask_ptes(0ull, 0ull, 0ull, 0ull,
-				VMX_EPT_EXECUTABLE_MASK);
+				VMX_EPT_EXECUTABLE_MASK,
+				VMX_EPT_DEFAULT_MT << VMX_EPT_MT_EPTE_SHIFT);
 		kvm_enable_tdp();
 	} else
 		kvm_disable_tdp();
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index f1f8ff2..0e6aa81 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -39,6 +39,7 @@
 #include <asm/uaccess.h>
 #include <asm/msr.h>
 #include <asm/desc.h>
+#include <asm/mtrr.h>
 
 #define MAX_IO_MSRS 256
 #define CR0_RESERVED_BITS						\
@@ -86,6 +87,7 @@
 	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
 	{ "hypercalls", VCPU_STAT(hypercalls) },
 	{ "request_irq", VCPU_STAT(request_irq_exits) },
+	{ "request_nmi", VCPU_STAT(request_nmi_exits) },
 	{ "irq_exits", VCPU_STAT(irq_exits) },
 	{ "host_state_reload", VCPU_STAT(host_state_reload) },
 	{ "efer_reload", VCPU_STAT(efer_reload) },
@@ -93,6 +95,7 @@
 	{ "insn_emulation", VCPU_STAT(insn_emulation) },
 	{ "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) },
 	{ "irq_injections", VCPU_STAT(irq_injections) },
+	{ "nmi_injections", VCPU_STAT(nmi_injections) },
 	{ "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) },
 	{ "mmu_pte_write", VM_STAT(mmu_pte_write) },
 	{ "mmu_pte_updated", VM_STAT(mmu_pte_updated) },
@@ -101,6 +104,7 @@
 	{ "mmu_recycled", VM_STAT(mmu_recycled) },
 	{ "mmu_cache_miss", VM_STAT(mmu_cache_miss) },
 	{ "mmu_unsync", VM_STAT(mmu_unsync) },
+	{ "mmu_unsync_global", VM_STAT(mmu_unsync_global) },
 	{ "remote_tlb_flush", VM_STAT(remote_tlb_flush) },
 	{ "largepages", VM_STAT(lpages) },
 	{ NULL }
@@ -312,6 +316,7 @@
 	kvm_x86_ops->set_cr0(vcpu, cr0);
 	vcpu->arch.cr0 = cr0;
 
+	kvm_mmu_sync_global(vcpu);
 	kvm_mmu_reset_context(vcpu);
 	return;
 }
@@ -355,6 +360,7 @@
 	}
 	kvm_x86_ops->set_cr4(vcpu, cr4);
 	vcpu->arch.cr4 = cr4;
+	kvm_mmu_sync_global(vcpu);
 	kvm_mmu_reset_context(vcpu);
 }
 EXPORT_SYMBOL_GPL(kvm_set_cr4);
@@ -449,7 +455,7 @@
 	MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR,
 #endif
 	MSR_IA32_TIME_STAMP_COUNTER, MSR_KVM_SYSTEM_TIME, MSR_KVM_WALL_CLOCK,
-	MSR_IA32_PERF_STATUS,
+	MSR_IA32_PERF_STATUS, MSR_IA32_CR_PAT
 };
 
 static unsigned num_msrs_to_save;
@@ -648,10 +654,38 @@
 
 static int set_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 {
+	u64 *p = (u64 *)&vcpu->arch.mtrr_state.fixed_ranges;
+
 	if (!msr_mtrr_valid(msr))
 		return 1;
 
-	vcpu->arch.mtrr[msr - 0x200] = data;
+	if (msr == MSR_MTRRdefType) {
+		vcpu->arch.mtrr_state.def_type = data;
+		vcpu->arch.mtrr_state.enabled = (data & 0xc00) >> 10;
+	} else if (msr == MSR_MTRRfix64K_00000)
+		p[0] = data;
+	else if (msr == MSR_MTRRfix16K_80000 || msr == MSR_MTRRfix16K_A0000)
+		p[1 + msr - MSR_MTRRfix16K_80000] = data;
+	else if (msr >= MSR_MTRRfix4K_C0000 && msr <= MSR_MTRRfix4K_F8000)
+		p[3 + msr - MSR_MTRRfix4K_C0000] = data;
+	else if (msr == MSR_IA32_CR_PAT)
+		vcpu->arch.pat = data;
+	else {	/* Variable MTRRs */
+		int idx, is_mtrr_mask;
+		u64 *pt;
+
+		idx = (msr - 0x200) / 2;
+		is_mtrr_mask = msr - 0x200 - 2 * idx;
+		if (!is_mtrr_mask)
+			pt =
+			  (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].base_lo;
+		else
+			pt =
+			  (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].mask_lo;
+		*pt = data;
+	}
+
+	kvm_mmu_reset_context(vcpu);
 	return 0;
 }
 
@@ -747,10 +781,37 @@
 
 static int get_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 {
+	u64 *p = (u64 *)&vcpu->arch.mtrr_state.fixed_ranges;
+
 	if (!msr_mtrr_valid(msr))
 		return 1;
 
-	*pdata = vcpu->arch.mtrr[msr - 0x200];
+	if (msr == MSR_MTRRdefType)
+		*pdata = vcpu->arch.mtrr_state.def_type +
+			 (vcpu->arch.mtrr_state.enabled << 10);
+	else if (msr == MSR_MTRRfix64K_00000)
+		*pdata = p[0];
+	else if (msr == MSR_MTRRfix16K_80000 || msr == MSR_MTRRfix16K_A0000)
+		*pdata = p[1 + msr - MSR_MTRRfix16K_80000];
+	else if (msr >= MSR_MTRRfix4K_C0000 && msr <= MSR_MTRRfix4K_F8000)
+		*pdata = p[3 + msr - MSR_MTRRfix4K_C0000];
+	else if (msr == MSR_IA32_CR_PAT)
+		*pdata = vcpu->arch.pat;
+	else {	/* Variable MTRRs */
+		int idx, is_mtrr_mask;
+		u64 *pt;
+
+		idx = (msr - 0x200) / 2;
+		is_mtrr_mask = msr - 0x200 - 2 * idx;
+		if (!is_mtrr_mask)
+			pt =
+			  (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].base_lo;
+		else
+			pt =
+			  (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].mask_lo;
+		*pdata = *pt;
+	}
+
 	return 0;
 }
 
@@ -903,7 +964,6 @@
 	case KVM_CAP_IRQCHIP:
 	case KVM_CAP_HLT:
 	case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
-	case KVM_CAP_USER_MEMORY:
 	case KVM_CAP_SET_TSS_ADDR:
 	case KVM_CAP_EXT_CPUID:
 	case KVM_CAP_CLOCKSOURCE:
@@ -1188,6 +1248,7 @@
 		int t, times = entry->eax & 0xff;
 
 		entry->flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
+		entry->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
 		for (t = 1; t < times && *nent < maxnent; ++t) {
 			do_cpuid_1_ent(&entry[t], function, 0);
 			entry[t].flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
@@ -1218,7 +1279,7 @@
 		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
 		/* read more entries until level_type is zero */
 		for (i = 1; *nent < maxnent; ++i) {
-			level_type = entry[i - 1].ecx & 0xff;
+			level_type = entry[i - 1].ecx & 0xff00;
 			if (!level_type)
 				break;
 			do_cpuid_1_ent(&entry[i], function, i);
@@ -1318,6 +1379,15 @@
 	return 0;
 }
 
+static int kvm_vcpu_ioctl_nmi(struct kvm_vcpu *vcpu)
+{
+	vcpu_load(vcpu);
+	kvm_inject_nmi(vcpu);
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
 static int vcpu_ioctl_tpr_access_reporting(struct kvm_vcpu *vcpu,
 					   struct kvm_tpr_access_ctl *tac)
 {
@@ -1377,6 +1447,13 @@
 		r = 0;
 		break;
 	}
+	case KVM_NMI: {
+		r = kvm_vcpu_ioctl_nmi(vcpu);
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
 	case KVM_SET_CPUID: {
 		struct kvm_cpuid __user *cpuid_arg = argp;
 		struct kvm_cpuid cpuid;
@@ -1968,7 +2045,7 @@
 	ret = kvm_write_guest(vcpu->kvm, gpa, val, bytes);
 	if (ret < 0)
 		return 0;
-	kvm_mmu_pte_write(vcpu, gpa, val, bytes);
+	kvm_mmu_pte_write(vcpu, gpa, val, bytes, 1);
 	return 1;
 }
 
@@ -2404,8 +2481,6 @@
 	val = kvm_register_read(vcpu, VCPU_REGS_RAX);
 	memcpy(vcpu->arch.pio_data, &val, 4);
 
-	kvm_x86_ops->skip_emulated_instruction(vcpu);
-
 	pio_dev = vcpu_find_pio_dev(vcpu, port, size, !in);
 	if (pio_dev) {
 		kernel_pio(pio_dev, vcpu, vcpu->arch.pio_data);
@@ -2541,7 +2616,7 @@
 	kvm_mmu_set_nonpresent_ptes(0ull, 0ull);
 	kvm_mmu_set_base_ptes(PT_PRESENT_MASK);
 	kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK,
-			PT_DIRTY_MASK, PT64_NX_MASK, 0);
+			PT_DIRTY_MASK, PT64_NX_MASK, 0, 0);
 	return 0;
 
 out:
@@ -2729,7 +2804,7 @@
 
 	e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT;
 	/* when no next entry is found, the current entry[i] is reselected */
-	for (j = i + 1; j == i; j = (j + 1) % nent) {
+	for (j = i + 1; ; j = (j + 1) % nent) {
 		struct kvm_cpuid_entry2 *ej = &vcpu->arch.cpuid_entries[j];
 		if (ej->function == e->function) {
 			ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
@@ -2973,7 +3048,7 @@
 		pr_debug("vcpu %d received sipi with vector # %x\n",
 			 vcpu->vcpu_id, vcpu->arch.sipi_vector);
 		kvm_lapic_reset(vcpu);
-		r = kvm_x86_ops->vcpu_reset(vcpu);
+		r = kvm_arch_vcpu_reset(vcpu);
 		if (r)
 			return r;
 		vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
@@ -3275,9 +3350,9 @@
 	kvm_desct->padding = 0;
 }
 
-static void get_segment_descritptor_dtable(struct kvm_vcpu *vcpu,
-					   u16 selector,
-					   struct descriptor_table *dtable)
+static void get_segment_descriptor_dtable(struct kvm_vcpu *vcpu,
+					  u16 selector,
+					  struct descriptor_table *dtable)
 {
 	if (selector & 1 << 2) {
 		struct kvm_segment kvm_seg;
@@ -3302,7 +3377,7 @@
 	struct descriptor_table dtable;
 	u16 index = selector >> 3;
 
-	get_segment_descritptor_dtable(vcpu, selector, &dtable);
+	get_segment_descriptor_dtable(vcpu, selector, &dtable);
 
 	if (dtable.limit < index * 8 + 7) {
 		kvm_queue_exception_e(vcpu, GP_VECTOR, selector & 0xfffc);
@@ -3321,7 +3396,7 @@
 	struct descriptor_table dtable;
 	u16 index = selector >> 3;
 
-	get_segment_descritptor_dtable(vcpu, selector, &dtable);
+	get_segment_descriptor_dtable(vcpu, selector, &dtable);
 
 	if (dtable.limit < index * 8 + 7)
 		return 1;
@@ -3900,6 +3975,7 @@
 	/* We do fxsave: this must be aligned. */
 	BUG_ON((unsigned long)&vcpu->arch.host_fx_image & 0xF);
 
+	vcpu->arch.mtrr_state.have_fixed = 1;
 	vcpu_load(vcpu);
 	r = kvm_arch_vcpu_reset(vcpu);
 	if (r == 0)
@@ -3925,6 +4001,9 @@
 
 int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu)
 {
+	vcpu->arch.nmi_pending = false;
+	vcpu->arch.nmi_injected = false;
+
 	return kvm_x86_ops->vcpu_reset(vcpu);
 }
 
@@ -4012,6 +4091,7 @@
 		return ERR_PTR(-ENOMEM);
 
 	INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
+	INIT_LIST_HEAD(&kvm->arch.oos_global_pages);
 	INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
 
 	/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */
@@ -4048,8 +4128,8 @@
 
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
-	kvm_iommu_unmap_guest(kvm);
 	kvm_free_all_assigned_devices(kvm);
+	kvm_iommu_unmap_guest(kvm);
 	kvm_free_pit(kvm);
 	kfree(kvm->arch.vpic);
 	kfree(kvm->arch.vioapic);
@@ -4127,7 +4207,8 @@
 int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 {
 	return vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE
-	       || vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED;
+	       || vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED
+	       || vcpu->arch.nmi_pending;
 }
 
 static void vcpu_kick_intr(void *info)
diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c
index ea05117..d174db7 100644
--- a/arch/x86/kvm/x86_emulate.c
+++ b/arch/x86/kvm/x86_emulate.c
@@ -58,6 +58,7 @@
 #define SrcMem32    (4<<4)	/* Memory operand (32-bit). */
 #define SrcImm      (5<<4)	/* Immediate operand. */
 #define SrcImmByte  (6<<4)	/* 8-bit sign-extended immediate operand. */
+#define SrcOne      (7<<4)	/* Implied '1' */
 #define SrcMask     (7<<4)
 /* Generic ModRM decode. */
 #define ModRM       (1<<7)
@@ -70,17 +71,23 @@
 #define Group       (1<<14)     /* Bits 3:5 of modrm byte extend opcode */
 #define GroupDual   (1<<15)     /* Alternate decoding of mod == 3 */
 #define GroupMask   0xff        /* Group number stored in bits 0:7 */
+/* Source 2 operand type */
+#define Src2None    (0<<29)
+#define Src2CL      (1<<29)
+#define Src2ImmByte (2<<29)
+#define Src2One     (3<<29)
+#define Src2Mask    (7<<29)
 
 enum {
 	Group1_80, Group1_81, Group1_82, Group1_83,
 	Group1A, Group3_Byte, Group3, Group4, Group5, Group7,
 };
 
-static u16 opcode_table[256] = {
+static u32 opcode_table[256] = {
 	/* 0x00 - 0x07 */
 	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
 	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
-	0, 0, 0, 0,
+	ByteOp | DstAcc | SrcImm, DstAcc | SrcImm, 0, 0,
 	/* 0x08 - 0x0F */
 	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM,
 	ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM,
@@ -195,7 +202,7 @@
 	ImplicitOps, ImplicitOps, Group | Group4, Group | Group5,
 };
 
-static u16 twobyte_table[256] = {
+static u32 twobyte_table[256] = {
 	/* 0x00 - 0x0F */
 	0, Group | GroupDual | Group7, 0, 0, 0, 0, ImplicitOps, 0,
 	ImplicitOps, ImplicitOps, 0, 0, 0, ImplicitOps | ModRM, 0, 0,
@@ -230,9 +237,14 @@
 	/* 0x90 - 0x9F */
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 	/* 0xA0 - 0xA7 */
-	0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, 0, 0,
+	0, 0, 0, DstMem | SrcReg | ModRM | BitOp,
+	DstMem | SrcReg | Src2ImmByte | ModRM,
+	DstMem | SrcReg | Src2CL | ModRM, 0, 0,
 	/* 0xA8 - 0xAF */
-	0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, ModRM, 0,
+	0, 0, 0, DstMem | SrcReg | ModRM | BitOp,
+	DstMem | SrcReg | Src2ImmByte | ModRM,
+	DstMem | SrcReg | Src2CL | ModRM,
+	ModRM, 0,
 	/* 0xB0 - 0xB7 */
 	ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, 0,
 	    DstMem | SrcReg | ModRM | BitOp,
@@ -253,7 +265,7 @@
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
 };
 
-static u16 group_table[] = {
+static u32 group_table[] = {
 	[Group1_80*8] =
 	ByteOp | DstMem | SrcImm | ModRM, ByteOp | DstMem | SrcImm | ModRM,
 	ByteOp | DstMem | SrcImm | ModRM, ByteOp | DstMem | SrcImm | ModRM,
@@ -297,9 +309,9 @@
 	SrcMem16 | ModRM | Mov, SrcMem | ModRM | ByteOp,
 };
 
-static u16 group2_table[] = {
+static u32 group2_table[] = {
 	[Group7*8] =
-	SrcNone | ModRM, 0, 0, 0,
+	SrcNone | ModRM, 0, 0, SrcNone | ModRM,
 	SrcNone | ModRM | DstMem | Mov, 0,
 	SrcMem16 | ModRM | Mov, 0,
 };
@@ -359,49 +371,48 @@
 	"andl %"_msk",%"_LO32 _tmp"; "		\
 	"orl  %"_LO32 _tmp",%"_sav"; "
 
+#ifdef CONFIG_X86_64
+#define ON64(x) x
+#else
+#define ON64(x)
+#endif
+
+#define ____emulate_2op(_op, _src, _dst, _eflags, _x, _y, _suffix)	\
+	do {								\
+		__asm__ __volatile__ (					\
+			_PRE_EFLAGS("0", "4", "2")			\
+			_op _suffix " %"_x"3,%1; "			\
+			_POST_EFLAGS("0", "4", "2")			\
+			: "=m" (_eflags), "=m" ((_dst).val),		\
+			  "=&r" (_tmp)					\
+			: _y ((_src).val), "i" (EFLAGS_MASK));		\
+	} while (0)
+
+
 /* Raw emulation: instruction has two explicit operands. */
 #define __emulate_2op_nobyte(_op,_src,_dst,_eflags,_wx,_wy,_lx,_ly,_qx,_qy) \
-	do { 								    \
-		unsigned long _tmp;					    \
-									    \
-		switch ((_dst).bytes) {					    \
-		case 2:							    \
-			__asm__ __volatile__ (				    \
-				_PRE_EFLAGS("0", "4", "2")		    \
-				_op"w %"_wx"3,%1; "			    \
-				_POST_EFLAGS("0", "4", "2")		    \
-				: "=m" (_eflags), "=m" ((_dst).val),        \
-				  "=&r" (_tmp)				    \
-				: _wy ((_src).val), "i" (EFLAGS_MASK));     \
-			break;						    \
-		case 4:							    \
-			__asm__ __volatile__ (				    \
-				_PRE_EFLAGS("0", "4", "2")		    \
-				_op"l %"_lx"3,%1; "			    \
-				_POST_EFLAGS("0", "4", "2")		    \
-				: "=m" (_eflags), "=m" ((_dst).val),	    \
-				  "=&r" (_tmp)				    \
-				: _ly ((_src).val), "i" (EFLAGS_MASK));     \
-			break;						    \
-		case 8:							    \
-			__emulate_2op_8byte(_op, _src, _dst,		    \
-					    _eflags, _qx, _qy);		    \
-			break;						    \
-		}							    \
+	do {								\
+		unsigned long _tmp;					\
+									\
+		switch ((_dst).bytes) {					\
+		case 2:							\
+			____emulate_2op(_op,_src,_dst,_eflags,_wx,_wy,"w"); \
+			break;						\
+		case 4:							\
+			____emulate_2op(_op,_src,_dst,_eflags,_lx,_ly,"l"); \
+			break;						\
+		case 8:							\
+			ON64(____emulate_2op(_op,_src,_dst,_eflags,_qx,_qy,"q")); \
+			break;						\
+		}							\
 	} while (0)
 
 #define __emulate_2op(_op,_src,_dst,_eflags,_bx,_by,_wx,_wy,_lx,_ly,_qx,_qy) \
 	do {								     \
-		unsigned long __tmp;					     \
+		unsigned long _tmp;					     \
 		switch ((_dst).bytes) {				             \
 		case 1:							     \
-			__asm__ __volatile__ (				     \
-				_PRE_EFLAGS("0", "4", "2")		     \
-				_op"b %"_bx"3,%1; "			     \
-				_POST_EFLAGS("0", "4", "2")		     \
-				: "=m" (_eflags), "=m" ((_dst).val),	     \
-				  "=&r" (__tmp)				     \
-				: _by ((_src).val), "i" (EFLAGS_MASK));      \
+			____emulate_2op(_op,_src,_dst,_eflags,_bx,_by,"b");  \
 			break;						     \
 		default:						     \
 			__emulate_2op_nobyte(_op, _src, _dst, _eflags,	     \
@@ -425,72 +436,69 @@
 	__emulate_2op_nobyte(_op, _src, _dst, _eflags,			\
 			     "w", "r", _LO32, "r", "", "r")
 
-/* Instruction has only one explicit operand (no source operand). */
-#define emulate_1op(_op, _dst, _eflags)                                    \
+/* Instruction has three operands and one operand is stored in ECX register */
+#define __emulate_2op_cl(_op, _cl, _src, _dst, _eflags, _suffix, _type) 	\
+	do {									\
+		unsigned long _tmp;						\
+		_type _clv  = (_cl).val;  					\
+		_type _srcv = (_src).val;    					\
+		_type _dstv = (_dst).val;					\
+										\
+		__asm__ __volatile__ (						\
+			_PRE_EFLAGS("0", "5", "2")				\
+			_op _suffix " %4,%1 \n"					\
+			_POST_EFLAGS("0", "5", "2")				\
+			: "=m" (_eflags), "+r" (_dstv), "=&r" (_tmp)		\
+			: "c" (_clv) , "r" (_srcv), "i" (EFLAGS_MASK)		\
+			); 							\
+										\
+		(_cl).val  = (unsigned long) _clv;				\
+		(_src).val = (unsigned long) _srcv;				\
+		(_dst).val = (unsigned long) _dstv;				\
+	} while (0)
+
+#define emulate_2op_cl(_op, _cl, _src, _dst, _eflags)				\
+	do {									\
+		switch ((_dst).bytes) {						\
+		case 2:								\
+			__emulate_2op_cl(_op, _cl, _src, _dst, _eflags,  	\
+						"w", unsigned short);         	\
+			break;							\
+		case 4: 							\
+			__emulate_2op_cl(_op, _cl, _src, _dst, _eflags,  	\
+						"l", unsigned int);           	\
+			break;							\
+		case 8:								\
+			ON64(__emulate_2op_cl(_op, _cl, _src, _dst, _eflags,	\
+						"q", unsigned long));  		\
+			break;							\
+		}								\
+	} while (0)
+
+#define __emulate_1op(_op, _dst, _eflags, _suffix)			\
 	do {								\
 		unsigned long _tmp;					\
 									\
+		__asm__ __volatile__ (					\
+			_PRE_EFLAGS("0", "3", "2")			\
+			_op _suffix " %1; "				\
+			_POST_EFLAGS("0", "3", "2")			\
+			: "=m" (_eflags), "+m" ((_dst).val),		\
+			  "=&r" (_tmp)					\
+			: "i" (EFLAGS_MASK));				\
+	} while (0)
+
+/* Instruction has only one explicit operand (no source operand). */
+#define emulate_1op(_op, _dst, _eflags)                                    \
+	do {								\
 		switch ((_dst).bytes) {				        \
-		case 1:							\
-			__asm__ __volatile__ (				\
-				_PRE_EFLAGS("0", "3", "2")		\
-				_op"b %1; "				\
-				_POST_EFLAGS("0", "3", "2")		\
-				: "=m" (_eflags), "=m" ((_dst).val),	\
-				  "=&r" (_tmp)				\
-				: "i" (EFLAGS_MASK));			\
-			break;						\
-		case 2:							\
-			__asm__ __volatile__ (				\
-				_PRE_EFLAGS("0", "3", "2")		\
-				_op"w %1; "				\
-				_POST_EFLAGS("0", "3", "2")		\
-				: "=m" (_eflags), "=m" ((_dst).val),	\
-				  "=&r" (_tmp)				\
-				: "i" (EFLAGS_MASK));			\
-			break;						\
-		case 4:							\
-			__asm__ __volatile__ (				\
-				_PRE_EFLAGS("0", "3", "2")		\
-				_op"l %1; "				\
-				_POST_EFLAGS("0", "3", "2")		\
-				: "=m" (_eflags), "=m" ((_dst).val),	\
-				  "=&r" (_tmp)				\
-				: "i" (EFLAGS_MASK));			\
-			break;						\
-		case 8:							\
-			__emulate_1op_8byte(_op, _dst, _eflags);	\
-			break;						\
+		case 1:	__emulate_1op(_op, _dst, _eflags, "b"); break;	\
+		case 2:	__emulate_1op(_op, _dst, _eflags, "w"); break;	\
+		case 4:	__emulate_1op(_op, _dst, _eflags, "l"); break;	\
+		case 8:	ON64(__emulate_1op(_op, _dst, _eflags, "q")); break; \
 		}							\
 	} while (0)
 
-/* Emulate an instruction with quadword operands (x86/64 only). */
-#if defined(CONFIG_X86_64)
-#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy)           \
-	do {								  \
-		__asm__ __volatile__ (					  \
-			_PRE_EFLAGS("0", "4", "2")			  \
-			_op"q %"_qx"3,%1; "				  \
-			_POST_EFLAGS("0", "4", "2")			  \
-			: "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \
-			: _qy ((_src).val), "i" (EFLAGS_MASK));		\
-	} while (0)
-
-#define __emulate_1op_8byte(_op, _dst, _eflags)                           \
-	do {								  \
-		__asm__ __volatile__ (					  \
-			_PRE_EFLAGS("0", "3", "2")			  \
-			_op"q %1; "					  \
-			_POST_EFLAGS("0", "3", "2")			  \
-			: "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \
-			: "i" (EFLAGS_MASK));				  \
-	} while (0)
-
-#elif defined(__i386__)
-#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy)
-#define __emulate_1op_8byte(_op, _dst, _eflags)
-#endif				/* __i386__ */
-
 /* Fetch next part of the instruction being emulated. */
 #define insn_fetch(_type, _size, _eip)                                  \
 ({	unsigned long _x;						\
@@ -1041,6 +1049,33 @@
 		c->src.bytes = 1;
 		c->src.val = insn_fetch(s8, 1, c->eip);
 		break;
+	case SrcOne:
+		c->src.bytes = 1;
+		c->src.val = 1;
+		break;
+	}
+
+	/*
+	 * Decode and fetch the second source operand: register, memory
+	 * or immediate.
+	 */
+	switch (c->d & Src2Mask) {
+	case Src2None:
+		break;
+	case Src2CL:
+		c->src2.bytes = 1;
+		c->src2.val = c->regs[VCPU_REGS_RCX] & 0x8;
+		break;
+	case Src2ImmByte:
+		c->src2.type = OP_IMM;
+		c->src2.ptr = (unsigned long *)c->eip;
+		c->src2.bytes = 1;
+		c->src2.val = insn_fetch(u8, 1, c->eip);
+		break;
+	case Src2One:
+		c->src2.bytes = 1;
+		c->src2.val = 1;
+		break;
 	}
 
 	/* Decode and fetch the destination operand: register or memory. */
@@ -1100,20 +1135,33 @@
 					       c->regs[VCPU_REGS_RSP]);
 }
 
+static int emulate_pop(struct x86_emulate_ctxt *ctxt,
+		       struct x86_emulate_ops *ops)
+{
+	struct decode_cache *c = &ctxt->decode;
+	int rc;
+
+	rc = ops->read_emulated(register_address(c, ss_base(ctxt),
+						 c->regs[VCPU_REGS_RSP]),
+				&c->src.val, c->src.bytes, ctxt->vcpu);
+	if (rc != 0)
+		return rc;
+
+	register_address_increment(c, &c->regs[VCPU_REGS_RSP], c->src.bytes);
+	return rc;
+}
+
 static inline int emulate_grp1a(struct x86_emulate_ctxt *ctxt,
 				struct x86_emulate_ops *ops)
 {
 	struct decode_cache *c = &ctxt->decode;
 	int rc;
 
-	rc = ops->read_std(register_address(c, ss_base(ctxt),
-					    c->regs[VCPU_REGS_RSP]),
-			   &c->dst.val, c->dst.bytes, ctxt->vcpu);
+	c->src.bytes = c->dst.bytes;
+	rc = emulate_pop(ctxt, ops);
 	if (rc != 0)
 		return rc;
-
-	register_address_increment(c, &c->regs[VCPU_REGS_RSP], c->dst.bytes);
-
+	c->dst.val = c->src.val;
 	return 0;
 }
 
@@ -1415,24 +1463,15 @@
 		emulate_1op("dec", c->dst, ctxt->eflags);
 		break;
 	case 0x50 ... 0x57:  /* push reg */
-		c->dst.type  = OP_MEM;
-		c->dst.bytes = c->op_bytes;
-		c->dst.val = c->src.val;
-		register_address_increment(c, &c->regs[VCPU_REGS_RSP],
-					   -c->op_bytes);
-		c->dst.ptr = (void *) register_address(
-			c, ss_base(ctxt), c->regs[VCPU_REGS_RSP]);
+		emulate_push(ctxt);
 		break;
 	case 0x58 ... 0x5f: /* pop reg */
 	pop_instruction:
-		if ((rc = ops->read_std(register_address(c, ss_base(ctxt),
-			c->regs[VCPU_REGS_RSP]), c->dst.ptr,
-			c->op_bytes, ctxt->vcpu)) != 0)
+		c->src.bytes = c->op_bytes;
+		rc = emulate_pop(ctxt, ops);
+		if (rc != 0)
 			goto done;
-
-		register_address_increment(c, &c->regs[VCPU_REGS_RSP],
-					   c->op_bytes);
-		c->dst.type = OP_NONE;	/* Disable writeback. */
+		c->dst.val = c->src.val;
 		break;
 	case 0x63:		/* movsxd */
 		if (ctxt->mode != X86EMUL_MODE_PROT64)
@@ -1591,7 +1630,9 @@
 		emulate_push(ctxt);
 		break;
 	case 0x9d: /* popf */
+		c->dst.type = OP_REG;
 		c->dst.ptr = (unsigned long *) &ctxt->eflags;
+		c->dst.bytes = c->op_bytes;
 		goto pop_instruction;
 	case 0xa0 ... 0xa1:	/* mov */
 		c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX];
@@ -1689,7 +1730,9 @@
 		emulate_grp2(ctxt);
 		break;
 	case 0xc3: /* ret */
+		c->dst.type = OP_REG;
 		c->dst.ptr = &c->eip;
+		c->dst.bytes = c->op_bytes;
 		goto pop_instruction;
 	case 0xc6 ... 0xc7:	/* mov (sole member of Grp11) */
 	mov:
@@ -1778,7 +1821,7 @@
 			c->eip = saved_eip;
 			goto cannot_emulate;
 		}
-		return 0;
+		break;
 	case 0xf4:              /* hlt */
 		ctxt->vcpu->arch.halt_request = 1;
 		break;
@@ -1999,12 +2042,20 @@
 		c->src.val &= (c->dst.bytes << 3) - 1;
 		emulate_2op_SrcV_nobyte("bt", c->src, c->dst, ctxt->eflags);
 		break;
+	case 0xa4: /* shld imm8, r, r/m */
+	case 0xa5: /* shld cl, r, r/m */
+		emulate_2op_cl("shld", c->src2, c->src, c->dst, ctxt->eflags);
+		break;
 	case 0xab:
 	      bts:		/* bts */
 		/* only subword offset */
 		c->src.val &= (c->dst.bytes << 3) - 1;
 		emulate_2op_SrcV_nobyte("bts", c->src, c->dst, ctxt->eflags);
 		break;
+	case 0xac: /* shrd imm8, r, r/m */
+	case 0xad: /* shrd cl, r, r/m */
+		emulate_2op_cl("shrd", c->src2, c->src, c->dst, ctxt->eflags);
+		break;
 	case 0xae:              /* clflush */
 		break;
 	case 0xb0 ... 0xb1:	/* cmpxchg */
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index 50a7792..a7ed208 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -738,7 +738,7 @@
 
 	/* We can't set cpumask in the initializer: damn C limitations!  Set it
 	 * here and register our timer device. */
-	lguest_clockevent.cpumask = cpumask_of_cpu(0);
+	lguest_clockevent.cpumask = cpumask_of(0);
 	clockevents_register_device(&lguest_clockevent);
 
 	/* Finally, we unblock the timer interrupt. */
diff --git a/arch/x86/mach-default/setup.c b/arch/x86/mach-default/setup.c
index 37b9ae4..df167f2 100644
--- a/arch/x86/mach-default/setup.c
+++ b/arch/x86/mach-default/setup.c
@@ -133,29 +133,28 @@
  **/
 void mca_nmi_hook(void)
 {
-	/* If I recall correctly, there's a whole bunch of other things that
+	/*
+	 * If I recall correctly, there's a whole bunch of other things that
 	 * we can do to check for NMI problems, but that's all I know about
 	 * at the moment.
 	 */
-
-	printk("NMI generated from unknown source!\n");
+	pr_warning("NMI generated from unknown source!\n");
 }
 #endif
 
 static __init int no_ipi_broadcast(char *str)
 {
 	get_option(&str, &no_broadcast);
-	printk ("Using %s mode\n", no_broadcast ? "No IPI Broadcast" :
-											"IPI Broadcast");
+	pr_info("Using %s mode\n",
+		no_broadcast ? "No IPI Broadcast" : "IPI Broadcast");
 	return 1;
 }
-
 __setup("no_ipi_broadcast=", no_ipi_broadcast);
 
 static int __init print_ipi_mode(void)
 {
-	printk ("Using IPI %s mode\n", no_broadcast ? "No-Shortcut" :
-											"Shortcut");
+	pr_info("Using IPI %s mode\n",
+		no_broadcast ? "No-Shortcut" : "Shortcut");
 	return 0;
 }
 
diff --git a/arch/x86/mach-generic/bigsmp.c b/arch/x86/mach-generic/bigsmp.c
index 3624a36..bc4c784 100644
--- a/arch/x86/mach-generic/bigsmp.c
+++ b/arch/x86/mach-generic/bigsmp.c
@@ -42,9 +42,10 @@
 	 { }
 };
 
-static cpumask_t vector_allocation_domain(int cpu)
+static void vector_allocation_domain(int cpu, cpumask_t *retmask)
 {
-        return cpumask_of_cpu(cpu);
+	cpus_clear(*retmask);
+	cpu_set(cpu, *retmask);
 }
 
 static int probe_bigsmp(void)
diff --git a/arch/x86/mach-generic/es7000.c b/arch/x86/mach-generic/es7000.c
index 7b4e6d0..4ba5cca 100644
--- a/arch/x86/mach-generic/es7000.c
+++ b/arch/x86/mach-generic/es7000.c
@@ -87,7 +87,7 @@
 }
 #endif
 
-static cpumask_t vector_allocation_domain(int cpu)
+static void vector_allocation_domain(int cpu, cpumask_t *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -97,8 +97,7 @@
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	cpumask_t domain = { { [0] = APIC_ALL_CPUS, } };
-	return domain;
+	*retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } };
 }
 
 struct genapic __initdata_refok apic_es7000 = APIC_INIT("es7000", probe_es7000);
diff --git a/arch/x86/mach-generic/numaq.c b/arch/x86/mach-generic/numaq.c
index 71a309b..511d794 100644
--- a/arch/x86/mach-generic/numaq.c
+++ b/arch/x86/mach-generic/numaq.c
@@ -38,7 +38,7 @@
 	return 0;
 }
 
-static cpumask_t vector_allocation_domain(int cpu)
+static void vector_allocation_domain(int cpu, cpumask_t *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -48,8 +48,7 @@
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	cpumask_t domain = { { [0] = APIC_ALL_CPUS, } };
-	return domain;
+	*retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } };
 }
 
 struct genapic apic_numaq = APIC_INIT("NUMAQ", probe_numaq);
diff --git a/arch/x86/mach-generic/summit.c b/arch/x86/mach-generic/summit.c
index 2c6d234..2821ffc 100644
--- a/arch/x86/mach-generic/summit.c
+++ b/arch/x86/mach-generic/summit.c
@@ -24,7 +24,7 @@
 	return 0;
 }
 
-static cpumask_t vector_allocation_domain(int cpu)
+static void vector_allocation_domain(int cpu, cpumask_t *retmask)
 {
 	/* Careful. Some cpus do not strictly honor the set of cpus
 	 * specified in the interrupt destination when using lowest
@@ -34,8 +34,7 @@
 	 * deliver interrupts to the wrong hyperthread when only one
 	 * hyperthread was specified in the interrupt desitination.
 	 */
-	cpumask_t domain = { { [0] = APIC_ALL_CPUS, } };
-	return domain;
+	*retmask = (cpumask_t){ { [0] = APIC_ALL_CPUS, } };
 }
 
 struct genapic apic_summit = APIC_INIT("summit", probe_summit);
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 5214500..a5bc054 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -63,11 +63,6 @@
 /* Used for the invalidate map that's also checked in the spinlock */
 static volatile unsigned long smp_invalidate_needed;
 
-/* Bitmask of currently online CPUs - used by setup.c for
-   /proc/cpuinfo, visible externally but still physical */
-cpumask_t cpu_online_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_online_map);
-
 /* Bitmask of CPUs present in the system - exported by i386_syms.c, used
  * by scheduler but indexed physically */
 cpumask_t phys_cpu_present_map = CPU_MASK_NONE;
@@ -218,8 +213,6 @@
 /* This is for the new dynamic CPU boot code */
 cpumask_t cpu_callin_map = CPU_MASK_NONE;
 cpumask_t cpu_callout_map = CPU_MASK_NONE;
-cpumask_t cpu_possible_map = CPU_MASK_NONE;
-EXPORT_SYMBOL(cpu_possible_map);
 
 /* The per processor IRQ masks (these are usually kept in sync) */
 static __u16 vic_irq_mask[NR_CPUS] __cacheline_aligned;
@@ -679,7 +672,7 @@
 
 	/* loop over all the extended VIC CPUs and boot them.  The
 	 * Quad CPUs must be bootstrapped by their extended VIC cpu */
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		if (i == boot_cpu_id || !cpu_isset(i, phys_cpu_present_map))
 			continue;
 		do_boot_cpu(i);
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 8655b5b..f99a6c6 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -435,8 +435,12 @@
 #endif /* !CONFIG_NUMA */
 
 #else
-# define permanent_kmaps_init(pgd_base)		do { } while (0)
-# define set_highmem_pages_init()	do { } while (0)
+static inline void permanent_kmaps_init(pgd_t *pgd_base)
+{
+}
+static inline void set_highmem_pages_init(void)
+{
+}
 #endif /* CONFIG_HIGHMEM */
 
 void __init native_pagetable_setup_start(pgd_t *base)
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index cebcbf1..71a14f8 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -278,7 +278,7 @@
 	int rr, i;
 
 	rr = first_node(node_online_map);
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		if (early_cpu_to_node(i) != NUMA_NO_NODE)
 			continue;
 		numa_set_node(i, rr);
@@ -549,7 +549,7 @@
 	memnodemap[0] = 0;
 	node_set_online(0);
 	node_set(0, node_possible_map);
-	for (i = 0; i < NR_CPUS; i++)
+	for (i = 0; i < nr_cpu_ids; i++)
 		numa_set_node(i, 0);
 	e820_register_active_regions(0, start_pfn, last_pfn);
 	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, last_pfn << PAGE_SHIFT);
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index 51c0a2f..09737c8 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -382,7 +382,7 @@
 		if (!node_online(i))
 			setup_node_bootmem(i, nodes[i].start, nodes[i].end);
 
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		int node = early_cpu_to_node(i);
 
 		if (node == NUMA_NO_NODE)
diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 1d88d2b..9e5752f 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -4,7 +4,7 @@
 #include <linux/irq.h>
 #include <linux/dmi.h>
 #include <asm/numa.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 struct pci_root_info {
 	char *name;
diff --git a/arch/x86/pci/amd_bus.c b/arch/x86/pci/amd_bus.c
index 22e0576..9bb0982 100644
--- a/arch/x86/pci/amd_bus.c
+++ b/arch/x86/pci/amd_bus.c
@@ -2,7 +2,7 @@
 #include <linux/pci.h>
 #include <linux/topology.h>
 #include <linux/cpu.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 #ifdef CONFIG_X86_64
 #include <asm/pci-direct.h>
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index bb1a01f..62ddb73 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -14,8 +14,7 @@
 #include <asm/segment.h>
 #include <asm/io.h>
 #include <asm/smp.h>
-
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 |
 				PCI_PROBE_MMCONF;
diff --git a/arch/x86/pci/direct.c b/arch/x86/pci/direct.c
index 9a5af6c..bd13c3e 100644
--- a/arch/x86/pci/direct.c
+++ b/arch/x86/pci/direct.c
@@ -5,7 +5,7 @@
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/dmi.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /*
  * Functions for accessing PCI base (first 256 bytes) and extended
diff --git a/arch/x86/pci/early.c b/arch/x86/pci/early.c
index 86631cc..f6adf2c 100644
--- a/arch/x86/pci/early.c
+++ b/arch/x86/pci/early.c
@@ -2,7 +2,7 @@
 #include <linux/pci.h>
 #include <asm/pci-direct.h>
 #include <asm/io.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /* Direct PCI access. This is used for PCI accesses in early boot before
    the PCI subsystem works. */
diff --git a/arch/x86/pci/fixup.c b/arch/x86/pci/fixup.c
index 2051dc9..7d388d5 100644
--- a/arch/x86/pci/fixup.c
+++ b/arch/x86/pci/fixup.c
@@ -6,8 +6,7 @@
 #include <linux/dmi.h>
 #include <linux/pci.h>
 #include <linux/init.h>
-#include "pci.h"
-
+#include <asm/pci_x86.h>
 
 static void __devinit pci_fixup_i450nx(struct pci_dev *d)
 {
diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c
index 844df0c..e51bf2c 100644
--- a/arch/x86/pci/i386.c
+++ b/arch/x86/pci/i386.c
@@ -34,8 +34,8 @@
 
 #include <asm/pat.h>
 #include <asm/e820.h>
+#include <asm/pci_x86.h>
 
-#include "pci.h"
 
 static int
 skip_isa_ioresource_align(struct pci_dev *dev) {
diff --git a/arch/x86/pci/init.c b/arch/x86/pci/init.c
index d6c950f..bec3b04 100644
--- a/arch/x86/pci/init.c
+++ b/arch/x86/pci/init.c
@@ -1,6 +1,6 @@
 #include <linux/pci.h>
 #include <linux/init.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /* arch_initcall has too random ordering, so call the initializers
    in the right sequence from here. */
diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c
index bf69dbe..373b9af 100644
--- a/arch/x86/pci/irq.c
+++ b/arch/x86/pci/irq.c
@@ -16,8 +16,7 @@
 #include <asm/io_apic.h>
 #include <linux/irq.h>
 #include <linux/acpi.h>
-
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 #define PIRQ_SIGNATURE	(('$' << 0) + ('P' << 8) + ('I' << 16) + ('R' << 24))
 #define PIRQ_VERSION 0x0100
diff --git a/arch/x86/pci/legacy.c b/arch/x86/pci/legacy.c
index b722dd4..f1065b1 100644
--- a/arch/x86/pci/legacy.c
+++ b/arch/x86/pci/legacy.c
@@ -3,7 +3,7 @@
  */
 #include <linux/init.h>
 #include <linux/pci.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /*
  * Discover remaining PCI buses in case there are peer host bridges.
diff --git a/arch/x86/pci/mmconfig-shared.c b/arch/x86/pci/mmconfig-shared.c
index 654a223..89bf924 100644
--- a/arch/x86/pci/mmconfig-shared.c
+++ b/arch/x86/pci/mmconfig-shared.c
@@ -15,8 +15,7 @@
 #include <linux/acpi.h>
 #include <linux/bitmap.h>
 #include <asm/e820.h>
-
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /* aperture is up to 256MB but BIOS may reserve less */
 #define MMCONFIG_APER_MIN	(2 * 1024*1024)
diff --git a/arch/x86/pci/mmconfig_32.c b/arch/x86/pci/mmconfig_32.c
index f3c761d..8b2d561 100644
--- a/arch/x86/pci/mmconfig_32.c
+++ b/arch/x86/pci/mmconfig_32.c
@@ -13,7 +13,7 @@
 #include <linux/init.h>
 #include <linux/acpi.h>
 #include <asm/e820.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /* Assume systems with more busses have correct MCFG */
 #define mmcfg_virt_addr ((void __iomem *) fix_to_virt(FIX_PCIE_MCFG))
diff --git a/arch/x86/pci/mmconfig_64.c b/arch/x86/pci/mmconfig_64.c
index a199416..30007ff 100644
--- a/arch/x86/pci/mmconfig_64.c
+++ b/arch/x86/pci/mmconfig_64.c
@@ -10,8 +10,7 @@
 #include <linux/acpi.h>
 #include <linux/bitmap.h>
 #include <asm/e820.h>
-
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /* Static virtual mapping of the MMCONFIG aperture */
 struct mmcfg_virt {
diff --git a/arch/x86/pci/numaq_32.c b/arch/x86/pci/numaq_32.c
index 1177845..2089354 100644
--- a/arch/x86/pci/numaq_32.c
+++ b/arch/x86/pci/numaq_32.c
@@ -7,7 +7,7 @@
 #include <linux/nodemask.h>
 #include <mach_apic.h>
 #include <asm/mpspec.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 #define XQUAD_PORTIO_BASE 0xfe400000
 #define XQUAD_PORTIO_QUAD 0x40000  /* 256k per quad. */
diff --git a/arch/x86/pci/olpc.c b/arch/x86/pci/olpc.c
index e11e9e8..b889d82 100644
--- a/arch/x86/pci/olpc.c
+++ b/arch/x86/pci/olpc.c
@@ -29,7 +29,7 @@
 #include <linux/init.h>
 #include <asm/olpc.h>
 #include <asm/geode.h>
-#include "pci.h"
+#include <asm/pci_x86.h>
 
 /*
  * In the tables below, the first two line (8 longwords) are the
diff --git a/arch/x86/pci/pcbios.c b/arch/x86/pci/pcbios.c
index 37472fc..b82cae9 100644
--- a/arch/x86/pci/pcbios.c
+++ b/arch/x86/pci/pcbios.c
@@ -6,9 +6,8 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/uaccess.h>
-#include "pci.h"
-#include "pci-functions.h"
-
+#include <asm/pci_x86.h>
+#include <asm/mach-default/pci-functions.h>
 
 /* BIOS32 signature: "_32_" */
 #define BIOS32_SIGNATURE	(('_' << 0) + ('3' << 8) + ('2' << 16) + ('_' << 24))
diff --git a/arch/x86/pci/visws.c b/arch/x86/pci/visws.c
index 42f4cb1..16d0c0e 100644
--- a/arch/x86/pci/visws.c
+++ b/arch/x86/pci/visws.c
@@ -9,11 +9,10 @@
 #include <linux/init.h>
 
 #include <asm/setup.h>
+#include <asm/pci_x86.h>
 #include <asm/visws/cobalt.h>
 #include <asm/visws/lithium.h>
 
-#include "pci.h"
-
 static int pci_visws_enable_irq(struct pci_dev *dev) { return 0; }
 static void pci_visws_disable_irq(struct pci_dev *dev) { }
 
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 773d68d..503c240 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -1082,7 +1082,7 @@
 
 static void xen_drop_mm_ref(struct mm_struct *mm)
 {
-	cpumask_t mask;
+	cpumask_var_t mask;
 	unsigned cpu;
 
 	if (current->active_mm == mm) {
@@ -1094,7 +1094,16 @@
 	}
 
 	/* Get the "official" set of cpus referring to our pagetable. */
-	mask = mm->cpu_vm_mask;
+	if (!alloc_cpumask_var(&mask, GFP_ATOMIC)) {
+		for_each_online_cpu(cpu) {
+			if (!cpumask_test_cpu(cpu, &mm->cpu_vm_mask)
+			    && per_cpu(xen_current_cr3, cpu) != __pa(mm->pgd))
+				continue;
+			smp_call_function_single(cpu, drop_other_mm_ref, mm, 1);
+		}
+		return;
+	}
+	cpumask_copy(mask, &mm->cpu_vm_mask);
 
 	/* It's possible that a vcpu may have a stale reference to our
 	   cr3, because its in lazy mode, and it hasn't yet flushed
@@ -1103,11 +1112,12 @@
 	   if needed. */
 	for_each_online_cpu(cpu) {
 		if (per_cpu(xen_current_cr3, cpu) == __pa(mm->pgd))
-			cpu_set(cpu, mask);
+			cpumask_set_cpu(cpu, mask);
 	}
 
-	if (!cpus_empty(mask))
-		smp_call_function_mask(mask, drop_other_mm_ref, mm, 1);
+	if (!cpumask_empty(mask))
+		smp_call_function_many(mask, drop_other_mm_ref, mm, 1);
+	free_cpumask_var(mask);
 }
 #else
 static void xen_drop_mm_ref(struct mm_struct *mm)
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index acd9b67..c44e206 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -33,7 +33,7 @@
 #include "xen-ops.h"
 #include "mmu.h"
 
-cpumask_t xen_cpu_initialized_map;
+cpumask_var_t xen_cpu_initialized_map;
 
 static DEFINE_PER_CPU(int, resched_irq);
 static DEFINE_PER_CPU(int, callfunc_irq);
@@ -158,7 +158,7 @@
 {
 	int i, rc;
 
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		rc = HYPERVISOR_vcpu_op(VCPUOP_is_up, i, NULL);
 		if (rc >= 0) {
 			num_processors++;
@@ -192,11 +192,14 @@
 	if (xen_smp_intr_init(0))
 		BUG();
 
-	xen_cpu_initialized_map = cpumask_of_cpu(0);
+	if (!alloc_cpumask_var(&xen_cpu_initialized_map, GFP_KERNEL))
+		panic("could not allocate xen_cpu_initialized_map\n");
+
+	cpumask_copy(xen_cpu_initialized_map, cpumask_of(0));
 
 	/* Restrict the possible_map according to max_cpus. */
 	while ((num_possible_cpus() > 1) && (num_possible_cpus() > max_cpus)) {
-		for (cpu = NR_CPUS - 1; !cpu_possible(cpu); cpu--)
+		for (cpu = nr_cpu_ids - 1; !cpu_possible(cpu); cpu--)
 			continue;
 		cpu_clear(cpu, cpu_possible_map);
 	}
@@ -221,7 +224,7 @@
 	struct vcpu_guest_context *ctxt;
 	struct desc_struct *gdt;
 
-	if (cpu_test_and_set(cpu, xen_cpu_initialized_map))
+	if (cpumask_test_and_set_cpu(cpu, xen_cpu_initialized_map))
 		return 0;
 
 	ctxt = kzalloc(sizeof(*ctxt), GFP_KERNEL);
@@ -408,24 +411,23 @@
 	xen_send_IPI_one(cpu, XEN_RESCHEDULE_VECTOR);
 }
 
-static void xen_send_IPI_mask(cpumask_t mask, enum ipi_vector vector)
+static void xen_send_IPI_mask(const struct cpumask *mask,
+			      enum ipi_vector vector)
 {
 	unsigned cpu;
 
-	cpus_and(mask, mask, cpu_online_map);
-
-	for_each_cpu_mask_nr(cpu, mask)
+	for_each_cpu_and(cpu, mask, cpu_online_mask)
 		xen_send_IPI_one(cpu, vector);
 }
 
-static void xen_smp_send_call_function_ipi(cpumask_t mask)
+static void xen_smp_send_call_function_ipi(const struct cpumask *mask)
 {
 	int cpu;
 
 	xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
 
 	/* Make sure other vcpus get a chance to run if they need to. */
-	for_each_cpu_mask_nr(cpu, mask) {
+	for_each_cpu(cpu, mask) {
 		if (xen_vcpu_stolen(cpu)) {
 			HYPERVISOR_sched_op(SCHEDOP_yield, 0);
 			break;
@@ -435,7 +437,8 @@
 
 static void xen_smp_send_call_function_single_ipi(int cpu)
 {
-	xen_send_IPI_mask(cpumask_of_cpu(cpu), XEN_CALL_FUNCTION_SINGLE_VECTOR);
+	xen_send_IPI_mask(cpumask_of(cpu),
+			  XEN_CALL_FUNCTION_SINGLE_VECTOR);
 }
 
 static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id)
diff --git a/arch/x86/xen/suspend.c b/arch/x86/xen/suspend.c
index 2a234db..212ffe0 100644
--- a/arch/x86/xen/suspend.c
+++ b/arch/x86/xen/suspend.c
@@ -35,7 +35,8 @@
 			pfn_to_mfn(xen_start_info->console.domU.mfn);
 	} else {
 #ifdef CONFIG_SMP
-		xen_cpu_initialized_map = cpu_online_map;
+		BUG_ON(xen_cpu_initialized_map == NULL);
+		cpumask_copy(xen_cpu_initialized_map, cpu_online_mask);
 #endif
 		xen_vcpu_restore();
 	}
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index c9f7cda..14f2406 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -132,8 +132,7 @@
 	*snap = state;
 
 	/* Add the appropriate number of ticks of stolen time,
-	   including any left-overs from last time.  Passing NULL to
-	   account_steal_time accounts the time as stolen. */
+	   including any left-overs from last time. */
 	stolen = runnable + offline + __get_cpu_var(residual_stolen);
 
 	if (stolen < 0)
@@ -141,11 +140,10 @@
 
 	ticks = iter_div_u64_rem(stolen, NS_PER_TICK, &stolen);
 	__get_cpu_var(residual_stolen) = stolen;
-	account_steal_time(NULL, ticks);
+	account_steal_ticks(ticks);
 
 	/* Add the appropriate number of ticks of blocked time,
-	   including any left-overs from last time.  Passing idle to
-	   account_steal_time accounts the time as idle/wait. */
+	   including any left-overs from last time. */
 	blocked += __get_cpu_var(residual_blocked);
 
 	if (blocked < 0)
@@ -153,7 +151,7 @@
 
 	ticks = iter_div_u64_rem(blocked, NS_PER_TICK, &blocked);
 	__get_cpu_var(residual_blocked) = blocked;
-	account_steal_time(idle_task(smp_processor_id()), ticks);
+	account_idle_ticks(ticks);
 }
 
 /*
@@ -437,7 +435,7 @@
 	evt = &per_cpu(xen_clock_events, cpu);
 	memcpy(evt, xen_clockevent, sizeof(*evt));
 
-	evt->cpumask = cpumask_of_cpu(cpu);
+	evt->cpumask = cpumask_of(cpu);
 	evt->irq = irq;
 
 	setup_runstate_info(cpu);
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index 9e1afae..c1f8faf 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -58,7 +58,7 @@
 __cpuinit void xen_init_lock_cpu(int cpu);
 void xen_uninit_lock_cpu(int cpu);
 
-extern cpumask_t xen_cpu_initialized_map;
+extern cpumask_var_t xen_cpu_initialized_map;
 #else
 static inline void xen_smp_init(void) {}
 #endif
diff --git a/arch/xtensa/kernel/init_task.c b/arch/xtensa/kernel/init_task.c
index 3df469d..e07f5c9 100644
--- a/arch/xtensa/kernel/init_task.c
+++ b/arch/xtensa/kernel/init_task.c
@@ -21,7 +21,6 @@
 
 #include <asm/uaccess.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 64f5d54..4259072 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -109,7 +109,7 @@
  */
 static ssize_t print_cpus_map(char *buf, cpumask_t *map)
 {
-	int n = cpulist_scnprintf(buf, PAGE_SIZE-2, *map);
+	int n = cpulist_scnprintf(buf, PAGE_SIZE-2, map);
 
 	buf[n++] = '\n';
 	buf[n] = '\0';
diff --git a/drivers/base/node.c b/drivers/base/node.c
index f520709..91636cd 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -30,8 +30,8 @@
 	BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1));
 
 	len = type?
-		cpulist_scnprintf(buf, PAGE_SIZE-2, *mask):
-		cpumask_scnprintf(buf, PAGE_SIZE-2, *mask);
+		cpulist_scnprintf(buf, PAGE_SIZE-2, mask) :
+		cpumask_scnprintf(buf, PAGE_SIZE-2, mask);
  	buf[len++] = '\n';
  	buf[len] = '\0';
 	return len;
diff --git a/drivers/base/topology.c b/drivers/base/topology.c
index 199cd97..a8bc1cb 100644
--- a/drivers/base/topology.c
+++ b/drivers/base/topology.c
@@ -49,8 +49,8 @@
 
 	if (len > 1) {
 		n = type?
-			cpulist_scnprintf(buf, len-2, *mask):
-			cpumask_scnprintf(buf, len-2, *mask);
+			cpulist_scnprintf(buf, len-2, mask) :
+			cpumask_scnprintf(buf, len-2, mask);
 		buf[n++] = '\n';
 		buf[n] = '\0';
 	}
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index c602b54..16970431 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -190,7 +190,7 @@
 
 config ESPSERIAL
 	tristate "Hayes ESP serial port support"
-	depends on SERIAL_NONSTANDARD && ISA && ISA_DMA_API
+	depends on SERIAL_NONSTANDARD && ISA && ISA_DMA_API && BROKEN
 	help
 	  This is a driver which supports Hayes ESP serial ports.  Both single
 	  port cards and multiport cards are supported.  Make sure to read
@@ -443,6 +443,17 @@
 	  All modern Linux systems use the Unix98 ptys.  Say Y unless
 	  you're on an embedded system and want to conserve memory.
 
+config DEVPTS_MULTIPLE_INSTANCES
+	bool "Support multiple instances of devpts"
+	depends on UNIX98_PTYS
+	default n
+	---help---
+	  Enable support for multiple instances of devpts filesystem.
+	  If you want to have isolated PTY namespaces (eg: in containers),
+	  say Y here.  Otherwise, say N. If enabled, each mount of devpts
+	  filesystem with the '-o newinstance' option will create an
+	  independent PTY namespace.
+
 config LEGACY_PTYS
 	bool "Legacy (BSD) PTY support"
 	default y
diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c
index b97aebd..4e0cfde 100644
--- a/drivers/char/amiserial.c
+++ b/drivers/char/amiserial.c
@@ -170,7 +170,7 @@
  */
 static void rs_stop(struct tty_struct *tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 	unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "rs_stop"))
@@ -190,7 +190,7 @@
 
 static void rs_start(struct tty_struct *tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 	unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "rs_start"))
@@ -861,7 +861,7 @@
 
 static void rs_flush_chars(struct tty_struct *tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 	unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "rs_flush_chars"))
@@ -934,7 +934,7 @@
 
 static int rs_write_room(struct tty_struct *tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 
 	if (serial_paranoia_check(info, tty->name, "rs_write_room"))
 		return 0;
@@ -943,7 +943,7 @@
 
 static int rs_chars_in_buffer(struct tty_struct *tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 
 	if (serial_paranoia_check(info, tty->name, "rs_chars_in_buffer"))
 		return 0;
@@ -952,7 +952,7 @@
 
 static void rs_flush_buffer(struct tty_struct *tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 	unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "rs_flush_buffer"))
@@ -969,7 +969,7 @@
  */
 static void rs_send_xchar(struct tty_struct *tty, char ch)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
         unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "rs_send_char"))
@@ -1004,7 +1004,7 @@
  */
 static void rs_throttle(struct tty_struct * tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 	unsigned long flags;
 #ifdef SERIAL_DEBUG_THROTTLE
 	char	buf[64];
@@ -1029,7 +1029,7 @@
 
 static void rs_unthrottle(struct tty_struct * tty)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 	unsigned long flags;
 #ifdef SERIAL_DEBUG_THROTTLE
 	char	buf[64];
@@ -1194,7 +1194,7 @@
 
 static int rs_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct async_struct * info = (struct async_struct *)tty->driver_data;
+	struct async_struct * info = tty->driver_data;
 	unsigned char control, status;
 	unsigned long flags;
 
@@ -1217,7 +1217,7 @@
 static int rs_tiocmset(struct tty_struct *tty, struct file *file,
 		       unsigned int set, unsigned int clear)
 {
-	struct async_struct * info = (struct async_struct *)tty->driver_data;
+	struct async_struct * info = tty->driver_data;
 	unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "rs_ioctl"))
@@ -1244,7 +1244,7 @@
  */
 static int rs_break(struct tty_struct *tty, int break_state)
 {
-	struct async_struct * info = (struct async_struct *)tty->driver_data;
+	struct async_struct * info = tty->driver_data;
 	unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "rs_break"))
@@ -1264,7 +1264,7 @@
 static int rs_ioctl(struct tty_struct *tty, struct file * file,
 		    unsigned int cmd, unsigned long arg)
 {
-	struct async_struct * info = (struct async_struct *)tty->driver_data;
+	struct async_struct * info = tty->driver_data;
 	struct async_icount cprev, cnow;	/* kernel counter temps */
 	struct serial_icounter_struct icount;
 	void __user *argp = (void __user *)arg;
@@ -1368,7 +1368,7 @@
 
 static void rs_set_termios(struct tty_struct *tty, struct ktermios *old_termios)
 {
-	struct async_struct *info = (struct async_struct *)tty->driver_data;
+	struct async_struct *info = tty->driver_data;
 	unsigned long flags;
 	unsigned int cflag = tty->termios->c_cflag;
 
@@ -1428,7 +1428,7 @@
  */
 static void rs_close(struct tty_struct *tty, struct file * filp)
 {
-	struct async_struct * info = (struct async_struct *)tty->driver_data;
+	struct async_struct * info = tty->driver_data;
 	struct serial_state *state;
 	unsigned long flags;
 
@@ -1523,7 +1523,7 @@
  */
 static void rs_wait_until_sent(struct tty_struct *tty, int timeout)
 {
-	struct async_struct * info = (struct async_struct *)tty->driver_data;
+	struct async_struct * info = tty->driver_data;
 	unsigned long orig_jiffies, char_time;
 	int lsr;
 
@@ -1587,7 +1587,7 @@
  */
 static void rs_hangup(struct tty_struct *tty)
 {
-	struct async_struct * info = (struct async_struct *)tty->driver_data;
+	struct async_struct * info = tty->driver_data;
 	struct serial_state *state = info->state;
 
 	if (serial_paranoia_check(info, tty->name, "rs_hangup"))
diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c
index 5e5b1dc..6a59f72 100644
--- a/drivers/char/cyclades.c
+++ b/drivers/char/cyclades.c
@@ -5010,7 +5010,7 @@
 		if (nchan == 0) {
 			dev_err(&pdev->dev, "Cyclom-Y PCI host card with no "
 					"Serial-Modules\n");
-			return -EIO;
+			goto err_unmap;
 		}
 	} else if (device_id == PCI_DEVICE_ID_CYCLOM_Z_Hi) {
 		struct RUNTIME_9060 __iomem *ctl_addr;
diff --git a/drivers/char/epca.c b/drivers/char/epca.c
index cf2461d..39ad820 100644
--- a/drivers/char/epca.c
+++ b/drivers/char/epca.c
@@ -69,7 +69,9 @@
 
 /*
  * The ISA boards do window flipping into the same spaces so its only sane with
- * a single lock. It's still pretty efficient.
+ * a single lock. It's still pretty efficient. This lock guards the hardware
+ * and the tty_port lock guards the kernel side stuff like use counts. Take
+ * this lock inside the port lock if you must take both.
  */
 static DEFINE_SPINLOCK(epca_lock);
 
@@ -156,14 +158,12 @@
 static void pc_sched_event(struct channel *, int);
 static void epca_error(int, char *);
 static void pc_close(struct tty_struct *, struct file *);
-static void shutdown(struct channel *);
+static void shutdown(struct channel *, struct tty_struct *tty);
 static void pc_hangup(struct tty_struct *);
 static int pc_write_room(struct tty_struct *);
 static int pc_chars_in_buffer(struct tty_struct *);
 static void pc_flush_buffer(struct tty_struct *);
 static void pc_flush_chars(struct tty_struct *);
-static int block_til_ready(struct tty_struct *, struct file *,
-			struct channel *);
 static int pc_open(struct tty_struct *, struct file *);
 static void post_fep_init(unsigned int crd);
 static void epcapoll(unsigned long);
@@ -173,7 +173,7 @@
 static unsigned termios2digi_i(struct channel *ch, unsigned);
 static unsigned termios2digi_c(struct channel *ch, unsigned);
 static void epcaparam(struct tty_struct *, struct channel *);
-static void receive_data(struct channel *);
+static void receive_data(struct channel *, struct tty_struct *tty);
 static int pc_ioctl(struct tty_struct *, struct file *,
 			unsigned int, unsigned long);
 static int info_ioctl(struct tty_struct *, struct file *,
@@ -392,7 +392,7 @@
 	 * through tty->driver_data this should catch it.
 	 */
 	if (tty) {
-		struct channel *ch = (struct channel *)tty->driver_data;
+		struct channel *ch = tty->driver_data;
 		if (ch >= &digi_channels[0] && ch < &digi_channels[nbdevs]) {
 			if (ch->magic == EPCA_MAGIC)
 				return ch;
@@ -419,76 +419,34 @@
 static void pc_close(struct tty_struct *tty, struct file *filp)
 {
 	struct channel *ch;
-	unsigned long flags;
+	struct tty_port *port;
 	/*
 	 * verifyChannel returns the channel from the tty struct if it is
 	 * valid. This serves as a sanity check.
 	 */
 	ch = verifyChannel(tty);
-	if (ch != NULL) {
-		spin_lock_irqsave(&epca_lock, flags);
-		if (tty_hung_up_p(filp)) {
-			spin_unlock_irqrestore(&epca_lock, flags);
-			return;
-		}
-		if (ch->port.count-- > 1)  {
-			/* Begin channel is open more than once */
-			/*
-			 * Return without doing anything. Someone might still
-			 * be using the channel.
-			 */
-			spin_unlock_irqrestore(&epca_lock, flags);
-			return;
-		}
-		/* Port open only once go ahead with shutdown & reset */
-		BUG_ON(ch->port.count < 0);
+	if (ch == NULL)
+		return;
+	port = &ch->port;
 
-		/*
-		 * Let the rest of the driver know the channel is being closed.
-		 * This becomes important if an open is attempted before close
-		 * is finished.
-		 */
-		ch->port.flags |= ASYNC_CLOSING;
-		tty->closing = 1;
+	if (tty_port_close_start(port, tty, filp) == 0)
+		return;
 
-		spin_unlock_irqrestore(&epca_lock, flags);
+	pc_flush_buffer(tty);
+	shutdown(ch, tty);
 
-		if (ch->port.flags & ASYNC_INITIALIZED)  {
-			/* Setup an event to indicate when the
-			   transmit buffer empties */
-			setup_empty_event(tty, ch);
-			/* 30 seconds timeout */
-			tty_wait_until_sent(tty, 3000);
-		}
-		pc_flush_buffer(tty);
-
-		tty_ldisc_flush(tty);
-		shutdown(ch);
-
-		spin_lock_irqsave(&epca_lock, flags);
-		tty->closing = 0;
-		ch->event = 0;
-		ch->port.tty = NULL;
-		spin_unlock_irqrestore(&epca_lock, flags);
-
-		if (ch->port.blocked_open) {
-			if (ch->close_delay)
-				msleep_interruptible(jiffies_to_msecs(ch->close_delay));
-			wake_up_interruptible(&ch->port.open_wait);
-		}
-		ch->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_INITIALIZED |
-					ASYNC_CLOSING);
-		wake_up_interruptible(&ch->port.close_wait);
-	}
+	tty_port_close_end(port, tty);
+	ch->event = 0;	/* FIXME: review ch->event locking */
+	tty_port_tty_set(port, NULL);
 }
 
-static void shutdown(struct channel *ch)
+static void shutdown(struct channel *ch, struct tty_struct *tty)
 {
 	unsigned long flags;
-	struct tty_struct *tty;
 	struct board_chan __iomem *bc;
+	struct tty_port *port = &ch->port;
 
-	if (!(ch->port.flags & ASYNC_INITIALIZED))
+	if (!(port->flags & ASYNC_INITIALIZED))
 		return;
 
 	spin_lock_irqsave(&epca_lock, flags);
@@ -503,7 +461,6 @@
 	 */
 	if (bc)
 		writeb(0, &bc->idata);
-	tty = ch->port.tty;
 
 	/* If we're a modem control device and HUPCL is on, drop RTS & DTR. */
 	if (tty->termios->c_cflag & HUPCL)  {
@@ -517,32 +474,26 @@
 	 * will have to reinitialized. Set a flag to indicate this.
 	 */
 	/* Prevent future Digi programmed interrupts from coming active */
-	ch->port.flags &= ~ASYNC_INITIALIZED;
+	port->flags &= ~ASYNC_INITIALIZED;
 	spin_unlock_irqrestore(&epca_lock, flags);
 }
 
 static void pc_hangup(struct tty_struct *tty)
 {
 	struct channel *ch;
+
 	/*
 	 * verifyChannel returns the channel from the tty struct if it is
 	 * valid. This serves as a sanity check.
 	 */
 	ch = verifyChannel(tty);
 	if (ch != NULL) {
-		unsigned long flags;
-
 		pc_flush_buffer(tty);
 		tty_ldisc_flush(tty);
-		shutdown(ch);
+		shutdown(ch, tty);
 
-		spin_lock_irqsave(&epca_lock, flags);
-		ch->port.tty   = NULL;
-		ch->event = 0;
-		ch->port.count = 0;
-		ch->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_INITIALIZED);
-		spin_unlock_irqrestore(&epca_lock, flags);
-		wake_up_interruptible(&ch->port.open_wait);
+		ch->event = 0;	/* FIXME: review locking of ch->event */
+		tty_port_hangup(&ch->port);
 	}
 }
 
@@ -786,100 +737,22 @@
 	}
 }
 
-static int block_til_ready(struct tty_struct *tty,
-				struct file *filp, struct channel *ch)
+static int epca_carrier_raised(struct tty_port *port)
 {
-	DECLARE_WAITQUEUE(wait, current);
-	int retval, do_clocal = 0;
-	unsigned long flags;
-
-	if (tty_hung_up_p(filp)) {
-		if (ch->port.flags & ASYNC_HUP_NOTIFY)
-			retval = -EAGAIN;
-		else
-			retval = -ERESTARTSYS;
-		return retval;
-	}
-
-	/*
-	 * If the device is in the middle of being closed, then block until
-	 * it's done, and then try again.
-	 */
-	if (ch->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&ch->port.close_wait);
-
-		if (ch->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		else
-			return -ERESTARTSYS;
-	}
-
-	if (filp->f_flags & O_NONBLOCK)  {
-		/*
-		 * If non-blocking mode is set, then make the check up front
-		 * and then exit.
-		 */
-		ch->port.flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-	if (tty->termios->c_cflag & CLOCAL)
-		do_clocal = 1;
-	/* Block waiting for the carrier detect and the line to become free */
-
-	retval = 0;
-	add_wait_queue(&ch->port.open_wait, &wait);
-
-	spin_lock_irqsave(&epca_lock, flags);
-	/* We dec count so that pc_close will know when to free things */
-	if (!tty_hung_up_p(filp))
-		ch->port.count--;
-	ch->port.blocked_open++;
-	while (1) {
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) ||
-				!(ch->port.flags & ASYNC_INITIALIZED)) {
-			if (ch->port.flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		if (!(ch->port.flags & ASYNC_CLOSING) &&
-			  (do_clocal || (ch->imodem & ch->dcd)))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-		spin_unlock_irqrestore(&epca_lock, flags);
-		/*
-		 * Allow someone else to be scheduled. We will occasionally go
-		 * through this loop until one of the above conditions change.
-		 * The below schedule call will allow other processes to enter
-		 * and prevent this loop from hogging the cpu.
-		 */
-		schedule();
-		spin_lock_irqsave(&epca_lock, flags);
-	}
-
-	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&ch->port.open_wait, &wait);
-	if (!tty_hung_up_p(filp))
-		ch->port.count++;
-	ch->port.blocked_open--;
-
-	spin_unlock_irqrestore(&epca_lock, flags);
-
-	if (retval)
-		return retval;
-
-	ch->port.flags |= ASYNC_NORMAL_ACTIVE;
+	struct channel *ch = container_of(port, struct channel, port);
+	if (ch->imodem & ch->dcd)
+		return 1;
 	return 0;
 }
 
+static void epca_raise_dtr_rts(struct tty_port *port)
+{
+}
+
 static int pc_open(struct tty_struct *tty, struct file *filp)
 {
 	struct channel *ch;
+	struct tty_port *port;
 	unsigned long flags;
 	int line, retval, boardnum;
 	struct board_chan __iomem *bc;
@@ -890,6 +763,7 @@
 		return -ENODEV;
 
 	ch = &digi_channels[line];
+	port = &ch->port;
 	boardnum = ch->boardnum;
 
 	/* Check status of board configured in system.  */
@@ -926,22 +800,24 @@
 		return -ENODEV;
 	}
 
-	spin_lock_irqsave(&epca_lock, flags);
+	spin_lock_irqsave(&port->lock, flags);
 	/*
 	 * Every time a channel is opened, increment a counter. This is
 	 * necessary because we do not wish to flush and shutdown the channel
 	 * until the last app holding the channel open, closes it.
 	 */
-	ch->port.count++;
+	port->count++;
 	/*
 	 * Set a kernel structures pointer to our local channel structure. This
 	 * way we can get to it when passed only a tty struct.
 	 */
 	tty->driver_data = ch;
+	port->tty = tty;
 	/*
 	 * If this is the first time the channel has been opened, initialize
 	 * the tty->termios struct otherwise let pc_close handle it.
 	 */
+	spin_lock(&epca_lock);
 	globalwinon(ch);
 	ch->statusflags = 0;
 
@@ -956,31 +832,33 @@
 	writew(head, &bc->rout);
 
 	/* Set the channels associated tty structure */
-	ch->port.tty = tty;
 
 	/*
 	 * The below routine generally sets up parity, baud, flow control
 	 * issues, etc.... It effect both control flags and input flags.
 	 */
 	epcaparam(tty, ch);
-	ch->port.flags |= ASYNC_INITIALIZED;
 	memoff(ch);
-	spin_unlock_irqrestore(&epca_lock, flags);
+	spin_unlock(&epca_lock);
+	port->flags |= ASYNC_INITIALIZED;
+	spin_unlock_irqrestore(&port->lock, flags);
 
-	retval = block_til_ready(tty, filp, ch);
+	retval = tty_port_block_til_ready(port, tty, filp);
 	if (retval)
 		return retval;
 	/*
 	 * Set this again in case a hangup set it to zero while this open() was
 	 * waiting for the line...
 	 */
-	spin_lock_irqsave(&epca_lock, flags);
-	ch->port.tty = tty;
+	spin_lock_irqsave(&port->lock, flags);
+	port->tty = tty;
+	spin_lock(&epca_lock);
 	globalwinon(ch);
 	/* Enable Digi Data events */
 	writeb(1, &bc->idata);
 	memoff(ch);
-	spin_unlock_irqrestore(&epca_lock, flags);
+	spin_unlock(&epca_lock);
+	spin_unlock_irqrestore(&port->lock, flags);
 	return 0;
 }
 
@@ -1016,8 +894,11 @@
 		}
 		ch = card_ptr[crd];
 		for (count = 0; count < bd->numports; count++, ch++) {
-			if (ch && ch->port.tty)
-				tty_hangup(ch->port.tty);
+			struct tty_struct *tty = tty_port_tty_get(&ch->port);
+			if (tty) {
+				tty_hangup(tty);
+				tty_kref_put(tty);
+			}
 		}
 	}
 	pci_unregister_driver(&epca_driver);
@@ -1042,6 +923,11 @@
 	.break_ctl = pc_send_break
 };
 
+static const struct tty_port_operations epca_port_ops = {
+	.carrier_raised = epca_carrier_raised,
+	.raise_dtr_rts = epca_raise_dtr_rts,
+};
+
 static int info_open(struct tty_struct *tty, struct file *filp)
 {
 	return 0;
@@ -1377,6 +1263,7 @@
 		u16 tseg, rseg;
 
 		tty_port_init(&ch->port);
+		ch->port.ops = &epca_port_ops;
 		ch->brdchan = bc;
 		ch->mailbox = gd;
 		INIT_WORK(&ch->tqueue, do_softint);
@@ -1428,7 +1315,7 @@
 		ch->boardnum   = crd;
 		ch->channelnum = i;
 		ch->magic      = EPCA_MAGIC;
-		ch->port.tty        = NULL;
+		tty_port_tty_set(&ch->port, NULL);
 
 		if (shrinkmem) {
 			fepcmd(ch, SETBUFFER, 32, 0, 0, 0);
@@ -1510,7 +1397,7 @@
 		ch->fepstartca = 0;
 		ch->fepstopca = 0;
 
-		ch->close_delay = 50;
+		ch->port.close_delay = 50;
 
 		spin_unlock_irqrestore(&epca_lock, flags);
 	}
@@ -1622,15 +1509,16 @@
 		if (bc == NULL)
 			goto next;
 
+		tty = tty_port_tty_get(&ch->port);
 		if (event & DATA_IND)  { /* Begin DATA_IND */
-			receive_data(ch);
+			receive_data(ch, tty);
 			assertgwinon(ch);
 		} /* End DATA_IND */
 		/* else *//* Fix for DCD transition missed bug */
 		if (event & MODEMCHG_IND) {
 			/* A modem signal change has been indicated */
 			ch->imodem = mstat;
-			if (ch->port.flags & ASYNC_CHECK_CD) {
+			if (test_bit(ASYNC_CHECK_CD, &ch->port.flags)) {
 				/* We are now receiving dcd */
 				if (mstat & ch->dcd)
 					wake_up_interruptible(&ch->port.open_wait);
@@ -1638,7 +1526,6 @@
 					pc_sched_event(ch, EPCA_EVENT_HANGUP);
 			}
 		}
-		tty = ch->port.tty;
 		if (tty) {
 			if (event & BREAK_IND) {
 				/* A break has been indicated */
@@ -1658,6 +1545,7 @@
 					tty_wakeup(tty);
 				}
 			}
+			tty_kref_put(tty);
 		}
 next:
 		globalwinon(ch);
@@ -1877,9 +1765,9 @@
 		 * that the driver will wait on carrier detect.
 		 */
 		if (ts->c_cflag & CLOCAL)
-			ch->port.flags &= ~ASYNC_CHECK_CD;
+			clear_bit(ASYNC_CHECK_CD, &ch->port.flags);
 		else
-			ch->port.flags |= ASYNC_CHECK_CD;
+			set_bit(ASYNC_CHECK_CD, &ch->port.flags);
 		mval = ch->m_dtr | ch->m_rts;
 	} /* End CBAUD not detected */
 	iflag = termios2digi_i(ch, ts->c_iflag);
@@ -1952,11 +1840,10 @@
 }
 
 /* Caller holds lock */
-static void receive_data(struct channel *ch)
+static void receive_data(struct channel *ch, struct tty_struct *tty)
 {
 	unchar *rptr;
 	struct ktermios *ts = NULL;
-	struct tty_struct *tty;
 	struct board_chan __iomem *bc;
 	int dataToRead, wrapgap, bytesAvailable;
 	unsigned int tail, head;
@@ -1969,7 +1856,6 @@
 	globalwinon(ch);
 	if (ch->statusflags & RXSTOPPED)
 		return;
-	tty = ch->port.tty;
 	if (tty)
 		ts = tty->termios;
 	bc = ch->brdchan;
@@ -2029,7 +1915,7 @@
 	globalwinon(ch);
 	writew(tail, &bc->rout);
 	/* Must be called with global data */
-	tty_schedule_flip(ch->port.tty);
+	tty_schedule_flip(tty);
 }
 
 static int info_ioctl(struct tty_struct *tty, struct file *file,
@@ -2097,7 +1983,7 @@
 
 static int pc_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct channel *ch = (struct channel *) tty->driver_data;
+	struct channel *ch = tty->driver_data;
 	struct board_chan __iomem *bc;
 	unsigned int mstat, mflag = 0;
 	unsigned long flags;
@@ -2131,7 +2017,7 @@
 static int pc_tiocmset(struct tty_struct *tty, struct file *file,
 		       unsigned int set, unsigned int clear)
 {
-	struct channel *ch = (struct channel *) tty->driver_data;
+	struct channel *ch = tty->driver_data;
 	unsigned long flags;
 
 	if (!ch)
@@ -2178,7 +2064,7 @@
 	unsigned int mflag, mstat;
 	unsigned char startc, stopc;
 	struct board_chan __iomem *bc;
-	struct channel *ch = (struct channel *) tty->driver_data;
+	struct channel *ch = tty->driver_data;
 	void __user *argp = (void __user *)arg;
 
 	if (ch)
@@ -2352,15 +2238,16 @@
 	struct channel *ch = container_of(work, struct channel, tqueue);
 	/* Called in response to a modem change event */
 	if (ch && ch->magic == EPCA_MAGIC) {
-		struct tty_struct *tty = ch->port.tty;
+		struct tty_struct *tty = tty_port_tty_get(&ch->port);;
 
 		if (tty && tty->driver_data) {
 			if (test_and_clear_bit(EPCA_EVENT_HANGUP, &ch->event)) {
 				tty_hangup(tty);
 				wake_up_interruptible(&ch->port.open_wait);
-				ch->port.flags &= ~ASYNC_NORMAL_ACTIVE;
+				clear_bit(ASYNC_NORMAL_ACTIVE, &ch->port.flags);
 			}
 		}
+		tty_kref_put(tty);
 	}
 }
 
@@ -2473,7 +2360,7 @@
 
 static int pc_send_break(struct tty_struct *tty, int msec)
 {
-	struct channel *ch = (struct channel *) tty->driver_data;
+	struct channel *ch = tty->driver_data;
 	unsigned long flags;
 
 	if (msec == -1)
diff --git a/drivers/char/esp.c b/drivers/char/esp.c
index 7f077c0..45ec263 100644
--- a/drivers/char/esp.c
+++ b/drivers/char/esp.c
@@ -2054,6 +2054,15 @@
 	wake_up_interruptible(&info->port.open_wait);
 }
 
+static int esp_carrier_raised(struct tty_port *port)
+{
+	struct esp_struct *info = container_of(port, struct esp_struct, port);
+	serial_out(info, UART_ESI_CMD1, ESI_GET_UART_STAT);
+	if (serial_in(info, UART_ESI_STAT2) & UART_MSR_DCD)
+		return 1;
+	return 0;
+}
+
 /*
  * ------------------------------------------------------------
  * esp_open() and friends
@@ -2066,17 +2075,19 @@
 	int		retval;
 	int		do_clocal = 0;
 	unsigned long	flags;
+	int		cd;
+	struct tty_port *port = &info->port;
 
 	/*
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
 	if (tty_hung_up_p(filp) ||
-	    (info->port.flags & ASYNC_CLOSING)) {
-		if (info->port.flags & ASYNC_CLOSING)
-			interruptible_sleep_on(&info->port.close_wait);
+	    (port->flags & ASYNC_CLOSING)) {
+		if (port->flags & ASYNC_CLOSING)
+			interruptible_sleep_on(&port->close_wait);
 #ifdef SERIAL_DO_RESTART
-		if (info->port.flags & ASYNC_HUP_NOTIFY)
+		if (port->flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -2091,7 +2102,7 @@
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -2101,20 +2112,20 @@
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * rs_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 #ifdef SERIAL_DEBUG_OPEN
 	printk(KERN_DEBUG "block_til_ready before block: ttys%d, count = %d\n",
-	       info->line, info->port.count);
+	       info->line, port->count);
 #endif
 	spin_lock_irqsave(&info->lock, flags);
 	if (!tty_hung_up_p(filp))
-		info->port.count--;
-	info->port.blocked_open++;
+		port->count--;
+	port->blocked_open++;
 	while (1) {
 		if ((tty->termios->c_cflag & CBAUD)) {
 			unsigned int scratch;
@@ -2129,9 +2140,9 @@
 		}
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(info->port.flags & ASYNC_INITIALIZED)) {
+		    !(port->flags & ASYNC_INITIALIZED)) {
 #ifdef SERIAL_DO_RESTART
-			if (info->port.flags & ASYNC_HUP_NOTIFY)
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
@@ -2141,11 +2152,9 @@
 			break;
 		}
 
-		serial_out(info, UART_ESI_CMD1, ESI_GET_UART_STAT);
-		if (serial_in(info, UART_ESI_STAT2) & UART_MSR_DCD)
-			do_clocal = 1;
+		cd = tty_port_carrier_raised(port);
 
-		if (!(info->port.flags & ASYNC_CLOSING) &&
+		if (!(port->flags & ASYNC_CLOSING) &&
 		    (do_clocal))
 			break;
 		if (signal_pending(current)) {
@@ -2154,25 +2163,25 @@
 		}
 #ifdef SERIAL_DEBUG_OPEN
 		printk(KERN_DEBUG "block_til_ready blocking: ttys%d, count = %d\n",
-		       info->line, info->port.count);
+		       info->line, port->count);
 #endif
 		spin_unlock_irqrestore(&info->lock, flags);
 		schedule();
 		spin_lock_irqsave(&info->lock, flags);
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	if (!tty_hung_up_p(filp))
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 	spin_unlock_irqrestore(&info->lock, flags);
 #ifdef SERIAL_DEBUG_OPEN
 	printk(KERN_DEBUG "block_til_ready after blocking: ttys%d, count = %d\n",
-	       info->line, info->port.count);
+	       info->line, port->count);
 #endif
 	if (retval)
 		return retval;
-	info->port.flags |= ASYNC_NORMAL_ACTIVE;
+	port->flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }
 
@@ -2329,6 +2338,10 @@
 	.tiocmset = esp_tiocmset,
 };
 
+static const struct tty_port_operations esp_port_ops = {
+	.esp_carrier_raised,
+};
+
 /*
  * The serial driver boot-time initialization code!
  */
@@ -2415,6 +2428,8 @@
 	offset = 0;
 
 	do {
+		tty_port_init(&info->port);
+		info->port.ops = &esp_port_ops;
 		info->io_port = esp[i] + offset;
 		info->irq = irq[i];
 		info->line = (i * 8) + (offset / 8);
@@ -2437,8 +2452,6 @@
 		info->config.flow_off = flow_off;
 		info->config.pio_threshold = pio_threshold;
 		info->next_port = ports;
-		init_waitqueue_head(&info->port.open_wait);
-		init_waitqueue_head(&info->port.close_wait);
 		init_waitqueue_head(&info->delta_msr_wait);
 		init_waitqueue_head(&info->break_wait);
 		ports = info;
diff --git a/drivers/char/generic_serial.c b/drivers/char/generic_serial.c
index c6090f8..9e4e569 100644
--- a/drivers/char/generic_serial.c
+++ b/drivers/char/generic_serial.c
@@ -376,7 +376,8 @@
 
 void gs_hangup(struct tty_struct *tty)
 {
-	struct gs_port   *port;
+	struct gs_port *port;
+	unsigned long flags;
 
 	func_enter ();
 
@@ -386,9 +387,11 @@
 		return;
 
 	gs_shutdown_port (port);
+	spin_lock_irqsave(&port->port.lock, flags);
 	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|GS_ACTIVE);
 	port->port.tty = NULL;
 	port->port.count = 0;
+	spin_unlock_irqrestore(&port->port.lock, flags);
 
 	wake_up_interruptible(&port->port.open_wait);
 	func_exit ();
@@ -397,7 +400,8 @@
 
 int gs_block_til_ready(void *port_, struct file * filp)
 {
-	struct gs_port *port = port_;
+	struct gs_port *gp = port_;
+	struct tty_port *port = &gp->port;
 	DECLARE_WAITQUEUE(wait, current);
 	int    retval;
 	int    do_clocal = 0;
@@ -409,16 +413,16 @@
 
 	if (!port) return 0;
 
-	tty = port->port.tty;
+	tty = port->tty;
 
 	gs_dprintk (GS_DEBUG_BTR, "Entering gs_block_till_ready.\n"); 
 	/*
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
-	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&port->port.close_wait);
-		if (port->port.flags & ASYNC_HUP_NOTIFY)
+	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&port->close_wait);
+		if (port->flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -432,7 +436,7 @@
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -444,34 +448,34 @@
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, port->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * rs_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
 
-	add_wait_queue(&port->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	gs_dprintk (GS_DEBUG_BTR, "after add waitq.\n"); 
-	spin_lock_irqsave(&port->driver_lock, flags);
+	spin_lock_irqsave(&port->lock, flags);
 	if (!tty_hung_up_p(filp)) {
-		port->port.count--;
+		port->count--;
 	}
-	spin_unlock_irqrestore(&port->driver_lock, flags);
-	port->port.blocked_open++;
+	port->blocked_open++;
+	spin_unlock_irqrestore(&port->lock, flags);
 	while (1) {
-		CD = port->rd->get_CD (port);
+		CD = tty_port_carrier_raised(port);
 		gs_dprintk (GS_DEBUG_BTR, "CD is now %d.\n", CD);
 		set_current_state (TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
+		    !(port->flags & ASYNC_INITIALIZED)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
+		if (!(port->flags & ASYNC_CLOSING) &&
 		    (do_clocal || CD))
 			break;
 		gs_dprintk (GS_DEBUG_BTR, "signal_pending is now: %d (%lx)\n", 
@@ -483,19 +487,20 @@
 		schedule();
 	}
 	gs_dprintk (GS_DEBUG_BTR, "Got out of the loop. (%d)\n",
-		    port->port.blocked_open);
+		    port->blocked_open);
 	set_current_state (TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
+	
+	spin_lock_irqsave(&port->lock, flags);
 	if (!tty_hung_up_p(filp)) {
-		port->port.count++;
+		port->count++;
 	}
-	port->port.blocked_open--;
-	if (retval)
-		return retval;
-
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
+	port->blocked_open--;
+	if (retval == 0)
+        	port->flags |= ASYNC_NORMAL_ACTIVE;
+	spin_unlock_irqrestore(&port->lock, flags);
 	func_exit ();
-	return 0;
+	return retval;
 }			 
 
 
@@ -506,7 +511,7 @@
 	
 	func_enter ();
 
-	port = (struct gs_port *) tty->driver_data;
+	port = tty->driver_data;
 
 	if (!port) return;
 
@@ -516,10 +521,10 @@
 		port->port.tty = tty;
 	}
 
-	spin_lock_irqsave(&port->driver_lock, flags);
+	spin_lock_irqsave(&port->port.lock, flags);
 
 	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&port->driver_lock, flags);
+		spin_unlock_irqrestore(&port->port.lock, flags);
 		if (port->rd->hungup)
 			port->rd->hungup (port);
 		func_exit ();
@@ -538,7 +543,7 @@
 
 	if (port->port.count) {
 		gs_dprintk(GS_DEBUG_CLOSE, "gs_close port %p: count: %d\n", port, port->port.count);
-		spin_unlock_irqrestore(&port->driver_lock, flags);
+		spin_unlock_irqrestore(&port->port.lock, flags);
 		func_exit ();
 		return;
 	}
@@ -559,8 +564,10 @@
 	 * line status register.
 	 */
 
+	spin_lock_irqsave(&port->driver_lock, flags);
 	port->rd->disable_rx_interrupts (port);
 	spin_unlock_irqrestore(&port->driver_lock, flags);
+	spin_unlock_irqrestore(&port->port.lock, flags);
 
 	/* close has no way of returning "EINTR", so discard return value */
 	if (port->closing_wait != ASYNC_CLOSING_WAIT_NONE)
@@ -573,20 +580,25 @@
 	tty_ldisc_flush(tty);
 	tty->closing = 0;
 
+	spin_lock_irqsave(&port->driver_lock, flags);
 	port->event = 0;
 	port->rd->close (port);
 	port->rd->shutdown_port (port);
+	spin_unlock_irqrestore(&port->driver_lock, flags);
+
+	spin_lock_irqsave(&port->port.lock, flags);
 	port->port.tty = NULL;
 
 	if (port->port.blocked_open) {
 		if (port->close_delay) {
-			spin_unlock_irqrestore(&port->driver_lock, flags);
+			spin_unlock_irqrestore(&port->port.lock, flags);
 			msleep_interruptible(jiffies_to_msecs(port->close_delay));
-			spin_lock_irqsave(&port->driver_lock, flags);
+			spin_lock_irqsave(&port->port.lock, flags);
 		}
 		wake_up_interruptible(&port->port.open_wait);
 	}
 	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING | ASYNC_INITIALIZED);
+	spin_unlock_irqrestore(&port->port.lock, flags);
 	wake_up_interruptible(&port->port.close_wait);
 
 	func_exit ();
diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c
index 0587b66..5a8a4c2 100644
--- a/drivers/char/hvc_console.c
+++ b/drivers/char/hvc_console.c
@@ -529,7 +529,7 @@
 	tty = tty_kref_get(hp->tty);
 	spin_unlock_irqrestore(&hp->lock, hvc_flags);
 
-	tty_do_resize(tty, tty, &ws);
+	tty_do_resize(tty, &ws);
 	tty_kref_put(tty);
 }
 
diff --git a/drivers/char/hvsi.c b/drivers/char/hvsi.c
index af05528..406f874 100644
--- a/drivers/char/hvsi.c
+++ b/drivers/char/hvsi.c
@@ -997,14 +997,14 @@
 
 static int hvsi_write_room(struct tty_struct *tty)
 {
-	struct hvsi_struct *hp = (struct hvsi_struct *)tty->driver_data;
+	struct hvsi_struct *hp = tty->driver_data;
 
 	return N_OUTBUF - hp->n_outbuf;
 }
 
 static int hvsi_chars_in_buffer(struct tty_struct *tty)
 {
-	struct hvsi_struct *hp = (struct hvsi_struct *)tty->driver_data;
+	struct hvsi_struct *hp = tty->driver_data;
 
 	return hp->n_outbuf;
 }
@@ -1070,7 +1070,7 @@
  */
 static void hvsi_throttle(struct tty_struct *tty)
 {
-	struct hvsi_struct *hp = (struct hvsi_struct *)tty->driver_data;
+	struct hvsi_struct *hp = tty->driver_data;
 
 	pr_debug("%s\n", __func__);
 
@@ -1079,7 +1079,7 @@
 
 static void hvsi_unthrottle(struct tty_struct *tty)
 {
-	struct hvsi_struct *hp = (struct hvsi_struct *)tty->driver_data;
+	struct hvsi_struct *hp = tty->driver_data;
 	unsigned long flags;
 	int shouldflip = 0;
 
@@ -1100,7 +1100,7 @@
 
 static int hvsi_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct hvsi_struct *hp = (struct hvsi_struct *)tty->driver_data;
+	struct hvsi_struct *hp = tty->driver_data;
 
 	hvsi_get_mctrl(hp);
 	return hp->mctrl;
@@ -1109,7 +1109,7 @@
 static int hvsi_tiocmset(struct tty_struct *tty, struct file *file,
 		unsigned int set, unsigned int clear)
 {
-	struct hvsi_struct *hp = (struct hvsi_struct *)tty->driver_data;
+	struct hvsi_struct *hp = tty->driver_data;
 	unsigned long flags;
 	uint16_t new_mctrl;
 
diff --git a/drivers/char/i8k.c b/drivers/char/i8k.c
index b60d425..fc8cf7a 100644
--- a/drivers/char/i8k.c
+++ b/drivers/char/i8k.c
@@ -485,7 +485,21 @@
 			DMI_MATCH(DMI_PRODUCT_NAME, "MP061"),
 		},
 	},
-	{ }
+	{
+		.ident = "Dell Precision",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Precision"),
+		},
+	},
+	{
+		.ident = "Dell Vostro",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Vostro"),
+		},
+	},
+        { }
 };
 
 /*
diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index 04e4549..24aa6e8 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -328,11 +328,13 @@
 }
 
 /* card->lock MUST NOT be held */
-static inline void raise_dtr_rts(struct isi_port *port)
+
+static void isicom_raise_dtr_rts(struct tty_port *port)
 {
-	struct isi_board *card = port->card;
+	struct isi_port *ip = container_of(port, struct isi_port, port);
+	struct isi_board *card = ip->card;
 	unsigned long base = card->base;
-	u16 channel = port->channel;
+	u16 channel = ip->channel;
 
 	if (!lock_card(card))
 		return;
@@ -340,7 +342,7 @@
 	outw(0x8000 | (channel << card->shift_count) | 0x02, base);
 	outw(0x0f04, base);
 	InterruptTheCard(base);
-	port->status |= (ISI_DTR | ISI_RTS);
+	ip->status |= (ISI_DTR | ISI_RTS);
 	unlock_card(card);
 }
 
@@ -830,80 +832,10 @@
 	return 0;
 }
 
-static int block_til_ready(struct tty_struct *tty, struct file *filp,
-	struct isi_port *port)
+static int isicom_carrier_raised(struct tty_port *port)
 {
-	struct isi_board *card = port->card;
-	int do_clocal = 0, retval;
-	unsigned long flags;
-	DECLARE_WAITQUEUE(wait, current);
-
-	/* block if port is in the process of being closed */
-
-	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
-		pr_dbg("block_til_ready: close in progress.\n");
-		interruptible_sleep_on(&port->port.close_wait);
-		if (port->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		else
-			return -ERESTARTSYS;
-	}
-
-	/* if non-blocking mode is set ... */
-
-	if ((filp->f_flags & O_NONBLOCK) ||
-			(tty->flags & (1 << TTY_IO_ERROR))) {
-		pr_dbg("block_til_ready: non-block mode.\n");
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (C_CLOCAL(tty))
-		do_clocal = 1;
-
-	/* block waiting for DCD to be asserted, and while
-						callout dev is busy */
-	retval = 0;
-	add_wait_queue(&port->port.open_wait, &wait);
-
-	spin_lock_irqsave(&card->card_lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->port.count--;
-	port->port.blocked_open++;
-	spin_unlock_irqrestore(&card->card_lock, flags);
-
-	while (1) {
-		raise_dtr_rts(port);
-
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) || !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
-				(do_clocal || (port->status & ISI_DCD))) {
-			break;
-		}
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-		schedule();
-	}
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
-	spin_lock_irqsave(&card->card_lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->port.count++;
-	port->port.blocked_open--;
-	spin_unlock_irqrestore(&card->card_lock, flags);
-	if (retval)
-		return retval;
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
-	return 0;
+	struct isi_port *ip = container_of(port, struct isi_port, port);
+	return (ip->status & ISI_DCD)?1 : 0;
 }
 
 static int isicom_open(struct tty_struct *tty, struct file *filp)
@@ -932,12 +864,13 @@
 
 	isicom_setup_board(card);
 
+	/* FIXME: locking on port.count etc */
 	port->port.count++;
 	tty->driver_data = port;
 	tty_port_tty_set(&port->port, tty);
 	error = isicom_setup_port(tty);
 	if (error == 0)
-		error = block_til_ready(tty, filp, port);
+		error = tty_port_block_til_ready(&port->port, tty, filp);
 	return error;
 }
 
@@ -1012,76 +945,30 @@
 
 static void isicom_close(struct tty_struct *tty, struct file *filp)
 {
-	struct isi_port *port = tty->driver_data;
+	struct isi_port *ip = tty->driver_data;
+	struct tty_port *port = &ip->port;
 	struct isi_board *card;
 	unsigned long flags;
 
-	if (!port)
-		return;
-	card = port->card;
-	if (isicom_paranoia_check(port, tty->name, "isicom_close"))
+	BUG_ON(!ip);
+
+	card = ip->card;
+	if (isicom_paranoia_check(ip, tty->name, "isicom_close"))
 		return;
 
-	pr_dbg("Close start!!!.\n");
-
-	spin_lock_irqsave(&card->card_lock, flags);
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&card->card_lock, flags);
-		return;
-	}
-
-	if (tty->count == 1 && port->port.count != 1) {
-		printk(KERN_WARNING "ISICOM:(0x%lx) isicom_close: bad port "
-			"count tty->count = 1 port count = %d.\n",
-			card->base, port->port.count);
-		port->port.count = 1;
-	}
-	if (--port->port.count < 0) {
-		printk(KERN_WARNING "ISICOM:(0x%lx) isicom_close: bad port "
-			"count for channel%d = %d", card->base, port->channel,
-			port->port.count);
-		port->port.count = 0;
-	}
-
-	if (port->port.count) {
-		spin_unlock_irqrestore(&card->card_lock, flags);
-		return;
-	}
-	port->port.flags |= ASYNC_CLOSING;
-	tty->closing = 1;
-	spin_unlock_irqrestore(&card->card_lock, flags);
-
-	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, port->port.closing_wait);
 	/* indicate to the card that no more data can be received
 	   on this port */
 	spin_lock_irqsave(&card->card_lock, flags);
-	if (port->port.flags & ASYNC_INITIALIZED) {
-		card->port_status &= ~(1 << port->channel);
+	if (port->flags & ASYNC_INITIALIZED) {
+		card->port_status &= ~(1 << ip->channel);
 		outw(card->port_status, card->base + 0x02);
 	}
-	isicom_shutdown_port(port);
+	isicom_shutdown_port(ip);
 	spin_unlock_irqrestore(&card->card_lock, flags);
 
 	isicom_flush_buffer(tty);
-	tty_ldisc_flush(tty);
-
-	spin_lock_irqsave(&card->card_lock, flags);
-	tty->closing = 0;
-
-	if (port->port.blocked_open) {
-		spin_unlock_irqrestore(&card->card_lock, flags);
-		if (port->port.close_delay) {
-			pr_dbg("scheduling until time out.\n");
-			msleep_interruptible(
-				jiffies_to_msecs(port->port.close_delay));
-		}
-		spin_lock_irqsave(&card->card_lock, flags);
-		wake_up_interruptible(&port->port.open_wait);
-	}
-	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
-	wake_up_interruptible(&port->port.close_wait);
-	spin_unlock_irqrestore(&card->card_lock, flags);
+	
+	tty_port_close_end(port, tty);
 }
 
 /* write et all */
@@ -1420,10 +1307,7 @@
 	isicom_shutdown_port(port);
 	spin_unlock_irqrestore(&port->card->card_lock, flags);
 
-	port->port.count = 0;
-	port->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	tty_port_tty_set(&port->port, NULL);
-	wake_up_interruptible(&port->port.open_wait);
+	tty_port_hangup(&port->port);
 }
 
 
@@ -1452,6 +1336,11 @@
 	.break_ctl		= isicom_send_break,
 };
 
+static const struct tty_port_operations isicom_port_ops = {
+	.carrier_raised		= isicom_carrier_raised,
+	.raise_dtr_rts		= isicom_raise_dtr_rts,
+};
+
 static int __devinit reset_card(struct pci_dev *pdev,
 	const unsigned int card, unsigned int *signature)
 {
@@ -1794,6 +1683,7 @@
 		spin_lock_init(&isi_card[idx].card_lock);
 		for (channel = 0; channel < 16; channel++, port++) {
 			tty_port_init(&port->port);
+			port->port.ops = &isicom_port_ops;
 			port->magic = ISICOM_MAGIC;
 			port->card = &isi_card[idx];
 			port->channel = channel;
diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 4b10770..5c3dc6b 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -151,7 +151,7 @@
 static char	*stli_serialname = "ttyE";
 
 static struct tty_driver	*stli_serial;
-
+static const struct tty_port_operations stli_port_ops;
 
 #define	STLI_TXBUFSIZE		4096
 
@@ -626,8 +626,6 @@
 static int	stli_initopen(struct tty_struct *tty, struct stlibrd *brdp, struct stliport *portp);
 static int	stli_rawopen(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait);
 static int	stli_rawclose(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait);
-static int	stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
-				struct stliport *portp, struct file *filp);
 static int	stli_setport(struct tty_struct *tty);
 static int	stli_cmdwait(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback);
 static void	stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback);
@@ -769,7 +767,7 @@
 			break;
 	}
 	if (i == ARRAY_SIZE(stli_brdstr)) {
-		printk("STALLION: unknown board name, %s?\n", argp[0]);
+		printk(KERN_WARNING "istallion: unknown board name, %s?\n", argp[0]);
 		return 0;
 	}
 
@@ -787,6 +785,7 @@
 {
 	struct stlibrd *brdp;
 	struct stliport *portp;
+	struct tty_port *port;
 	unsigned int minordev, brdnr, portnr;
 	int rc;
 
@@ -808,30 +807,19 @@
 		return -ENODEV;
 	if (portp->devnr < 1)
 		return -ENODEV;
-
-
-/*
- *	Check if this port is in the middle of closing. If so then wait
- *	until it is closed then return error status based on flag settings.
- *	The sleep here does not need interrupt protection since the wakeup
- *	for it is done with the same context.
- */
-	if (portp->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->port.close_wait);
-		if (portp->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		return -ERESTARTSYS;
-	}
+	port = &portp->port;
 
 /*
  *	On the first open of the device setup the port hardware, and
  *	initialize the per port data structure. Since initializing the port
  *	requires several commands to the board we will need to wait for any
  *	other open that is already initializing the port.
+ *
+ *	Review - locking
  */
-	tty_port_tty_set(&portp->port, tty);
+	tty_port_tty_set(port, tty);
 	tty->driver_data = portp;
-	portp->port.count++;
+	port->count++;
 
 	wait_event_interruptible(portp->raw_wait,
 			!test_bit(ST_INITIALIZING, &portp->state));
@@ -841,7 +829,8 @@
 	if ((portp->port.flags & ASYNC_INITIALIZED) == 0) {
 		set_bit(ST_INITIALIZING, &portp->state);
 		if ((rc = stli_initopen(tty, brdp, portp)) >= 0) {
-			portp->port.flags |= ASYNC_INITIALIZED;
+			/* Locking */
+			port->flags |= ASYNC_INITIALIZED;
 			clear_bit(TTY_IO_ERROR, &tty->flags);
 		}
 		clear_bit(ST_INITIALIZING, &portp->state);
@@ -849,31 +838,7 @@
 		if (rc < 0)
 			return rc;
 	}
-
-/*
- *	Check if this port is in the middle of closing. If so then wait
- *	until it is closed then return error status, based on flag settings.
- *	The sleep here does not need interrupt protection since the wakeup
- *	for it is done with the same context.
- */
-	if (portp->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->port.close_wait);
-		if (portp->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		return -ERESTARTSYS;
-	}
-
-/*
- *	Based on type of open being done check if it can overlap with any
- *	previous opens still in effect. If we are a normal serial device
- *	then also we might have to wait for carrier.
- */
-	if (!(filp->f_flags & O_NONBLOCK)) {
-		if ((rc = stli_waitcarrier(tty, brdp, portp, filp)) != 0)
-			return rc;
-	}
-	portp->port.flags |= ASYNC_NORMAL_ACTIVE;
-	return 0;
+	return tty_port_block_til_ready(&portp->port, tty, filp);
 }
 
 /*****************************************************************************/
@@ -882,25 +847,16 @@
 {
 	struct stlibrd *brdp;
 	struct stliport *portp;
+	struct tty_port *port;
 	unsigned long flags;
 
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
+	port = &portp->port;
 
-	spin_lock_irqsave(&stli_lock, flags);
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&stli_lock, flags);
+	if (tty_port_close_start(port, tty, filp) == 0)
 		return;
-	}
-	if ((tty->count == 1) && (portp->port.count != 1))
-		portp->port.count = 1;
-	if (portp->port.count-- > 1) {
-		spin_unlock_irqrestore(&stli_lock, flags);
-		return;
-	}
-
-	portp->port.flags |= ASYNC_CLOSING;
 
 /*
  *	May want to wait for data to drain before closing. The BUSY flag
@@ -908,15 +864,19 @@
  *	updated by messages from the slave - indicating when all chars
  *	really have drained.
  */
+ 	spin_lock_irqsave(&stli_lock, flags);
 	if (tty == stli_txcooktty)
 		stli_flushchars(tty);
-	tty->closing = 1;
 	spin_unlock_irqrestore(&stli_lock, flags);
 
+	/* We end up doing this twice for the moment. This needs looking at
+	   eventually. Note we still use portp->closing_wait as a result */
 	if (portp->closing_wait != ASYNC_CLOSING_WAIT_NONE)
 		tty_wait_until_sent(tty, portp->closing_wait);
 
-	portp->port.flags &= ~ASYNC_INITIALIZED;
+	/* FIXME: port locking here needs attending to */
+	port->flags &= ~ASYNC_INITIALIZED;
+
 	brdp = stli_brds[portp->brdnr];
 	stli_rawclose(brdp, portp, 0, 0);
 	if (tty->termios->c_cflag & HUPCL) {
@@ -934,17 +894,8 @@
 	set_bit(ST_DOFLUSHRX, &portp->state);
 	stli_flushbuffer(tty);
 
-	tty->closing = 0;
-	tty_port_tty_set(&portp->port, NULL);
-
-	if (portp->openwaitcnt) {
-		if (portp->close_delay)
-			msleep_interruptible(jiffies_to_msecs(portp->close_delay));
-		wake_up_interruptible(&portp->port.open_wait);
-	}
-
-	portp->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&portp->port.close_wait);
+	tty_port_close_end(port, tty);
+	tty_port_tty_set(port, NULL);
 }
 
 /*****************************************************************************/
@@ -1183,62 +1134,23 @@
 
 /*****************************************************************************/
 
-/*
- *	Possibly need to wait for carrier (DCD signal) to come high. Say
- *	maybe because if we are clocal then we don't need to wait...
- */
-
-static int stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
-				struct stliport *portp, struct file *filp)
+static int stli_carrier_raised(struct tty_port *port)
 {
-	unsigned long flags;
-	int rc, doclocal;
-
-	rc = 0;
-	doclocal = 0;
-
-	if (tty->termios->c_cflag & CLOCAL)
-		doclocal++;
-
-	spin_lock_irqsave(&stli_lock, flags);
-	portp->openwaitcnt++;
-	if (! tty_hung_up_p(filp))
-		portp->port.count--;
-	spin_unlock_irqrestore(&stli_lock, flags);
-
-	for (;;) {
-		stli_mkasysigs(&portp->asig, 1, 1);
-		if ((rc = stli_cmdwait(brdp, portp, A_SETSIGNALS,
-		    &portp->asig, sizeof(asysigs_t), 0)) < 0)
-			break;
-		if (tty_hung_up_p(filp) ||
-		    ((portp->port.flags & ASYNC_INITIALIZED) == 0)) {
-			if (portp->port.flags & ASYNC_HUP_NOTIFY)
-				rc = -EBUSY;
-			else
-				rc = -ERESTARTSYS;
-			break;
-		}
-		if (((portp->port.flags & ASYNC_CLOSING) == 0) &&
-		    (doclocal || (portp->sigs & TIOCM_CD))) {
-			break;
-		}
-		if (signal_pending(current)) {
-			rc = -ERESTARTSYS;
-			break;
-		}
-		interruptible_sleep_on(&portp->port.open_wait);
-	}
-
-	spin_lock_irqsave(&stli_lock, flags);
-	if (! tty_hung_up_p(filp))
-		portp->port.count++;
-	portp->openwaitcnt--;
-	spin_unlock_irqrestore(&stli_lock, flags);
-
-	return rc;
+	struct stliport *portp = container_of(port, struct stliport, port);
+	return (portp->sigs & TIOCM_CD) ? 1 : 0;
 }
 
+static void stli_raise_dtr_rts(struct tty_port *port)
+{
+	struct stliport *portp = container_of(port, struct stliport, port);
+	struct stlibrd *brdp = stli_brds[portp->brdnr];
+	stli_mkasysigs(&portp->asig, 1, 1);
+	if (stli_cmdwait(brdp, portp, A_SETSIGNALS, &portp->asig,
+		sizeof(asysigs_t), 0) < 0)
+			printk(KERN_WARNING "istallion: dtr raise failed.\n");
+}
+
+
 /*****************************************************************************/
 
 /*
@@ -1550,7 +1462,7 @@
 	sio.irq = 0;
 	sio.flags = portp->port.flags;
 	sio.baud_base = portp->baud_base;
-	sio.close_delay = portp->close_delay;
+	sio.close_delay = portp->port.close_delay;
 	sio.closing_wait = portp->closing_wait;
 	sio.custom_divisor = portp->custom_divisor;
 	sio.xmit_fifo_size = 0;
@@ -1582,7 +1494,7 @@
 		return -EFAULT;
 	if (!capable(CAP_SYS_ADMIN)) {
 		if ((sio.baud_base != portp->baud_base) ||
-		    (sio.close_delay != portp->close_delay) ||
+		    (sio.close_delay != portp->port.close_delay) ||
 		    ((sio.flags & ~ASYNC_USR_MASK) !=
 		    (portp->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
@@ -1591,7 +1503,7 @@
 	portp->port.flags = (portp->port.flags & ~ASYNC_USR_MASK) |
 		(sio.flags & ASYNC_USR_MASK);
 	portp->baud_base = sio.baud_base;
-	portp->close_delay = sio.close_delay;
+	portp->port.close_delay = sio.close_delay;
 	portp->closing_wait = sio.closing_wait;
 	portp->custom_divisor = sio.custom_divisor;
 
@@ -1821,6 +1733,7 @@
 {
 	struct stliport *portp;
 	struct stlibrd *brdp;
+	struct tty_port *port;
 	unsigned long flags;
 
 	portp = tty->driver_data;
@@ -1831,8 +1744,11 @@
 	brdp = stli_brds[portp->brdnr];
 	if (brdp == NULL)
 		return;
+	port = &portp->port;
 
-	portp->port.flags &= ~ASYNC_INITIALIZED;
+	spin_lock_irqsave(&port->lock, flags);
+	port->flags &= ~ASYNC_INITIALIZED;
+	spin_unlock_irqrestore(&port->lock, flags);
 
 	if (!test_bit(ST_CLOSING, &portp->state))
 		stli_rawclose(brdp, portp, 0, 0);
@@ -1853,12 +1769,9 @@
 	clear_bit(ST_TXBUSY, &portp->state);
 	clear_bit(ST_RXSTOP, &portp->state);
 	set_bit(TTY_IO_ERROR, &tty->flags);
-	tty_port_tty_set(&portp->port, NULL);
-	portp->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	portp->port.count = 0;
 	spin_unlock_irqrestore(&stli_lock, flags);
 
-	wake_up_interruptible(&portp->port.open_wait);
+	tty_port_hangup(port);
 }
 
 /*****************************************************************************/
@@ -2132,7 +2045,7 @@
 	unsigned char __iomem *bits;
 
 	if (test_bit(ST_CMDING, &portp->state)) {
-		printk(KERN_ERR "STALLION: command already busy, cmd=%x!\n",
+		printk(KERN_ERR "istallion: command already busy, cmd=%x!\n",
 				(int) cmd);
 		return;
 	}
@@ -2692,16 +2605,17 @@
 	for (i = 0, panelnr = 0, panelport = 0; (i < brdp->nrports); i++) {
 		portp = kzalloc(sizeof(struct stliport), GFP_KERNEL);
 		if (!portp) {
-			printk("STALLION: failed to allocate port structure\n");
+			printk(KERN_WARNING "istallion: failed to allocate port structure\n");
 			continue;
 		}
 		tty_port_init(&portp->port);
+		portp->port.ops = &stli_port_ops;
 		portp->magic = STLI_PORTMAGIC;
 		portp->portnr = i;
 		portp->brdnr = brdp->brdnr;
 		portp->panelnr = panelnr;
 		portp->baud_base = STL_BAUDBASE;
-		portp->close_delay = STL_CLOSEDELAY;
+		portp->port.close_delay = STL_CLOSEDELAY;
 		portp->closing_wait = 30 * HZ;
 		init_waitqueue_head(&portp->port.open_wait);
 		init_waitqueue_head(&portp->port.close_wait);
@@ -2758,7 +2672,7 @@
 	unsigned char val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2832,7 +2746,7 @@
 	unsigned char	val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2884,7 +2798,7 @@
 	unsigned char val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2929,7 +2843,7 @@
 	unsigned char	val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), board=%d\n",
 				(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2994,7 +2908,7 @@
 	void __iomem *ptr;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 				(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -3060,7 +2974,7 @@
 	unsigned char val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -3499,7 +3413,7 @@
 #endif
 
 	if (nrdevs < (brdp->nrports + 1)) {
-		printk(KERN_ERR "STALLION: slave failed to allocate memory for "
+		printk(KERN_ERR "istallion: slave failed to allocate memory for "
 				"all devices, devices=%d\n", nrdevs);
 		brdp->nrports = nrdevs - 1;
 	}
@@ -3509,13 +3423,13 @@
 	brdp->bitsize = (nrdevs + 7) / 8;
 	memoff = readl(&hdrp->memp);
 	if (memoff > brdp->memsize) {
-		printk(KERN_ERR "STALLION: corrupted shared memory region?\n");
+		printk(KERN_ERR "istallion: corrupted shared memory region?\n");
 		rc = -EIO;
 		goto stli_donestartup;
 	}
 	memp = (cdkmem_t __iomem *) EBRDGETMEMPTR(brdp, memoff);
 	if (readw(&memp->dtype) != TYP_ASYNCTRL) {
-		printk(KERN_ERR "STALLION: no slave control device found\n");
+		printk(KERN_ERR "istallion: no slave control device found\n");
 		goto stli_donestartup;
 	}
 	memp++;
@@ -3600,7 +3514,7 @@
 		retval = stli_initonb(brdp);
 		break;
 	default:
-		printk(KERN_ERR "STALLION: board=%d is unknown board "
+		printk(KERN_ERR "istallion: board=%d is unknown board "
 				"type=%d\n", brdp->brdnr, brdp->brdtype);
 		retval = -ENODEV;
 	}
@@ -3609,7 +3523,7 @@
 		return retval;
 
 	stli_initports(brdp);
-	printk(KERN_INFO "STALLION: %s found, board=%d io=%x mem=%x "
+	printk(KERN_INFO "istallion: %s found, board=%d io=%x mem=%x "
 		"nrpanels=%d nrports=%d\n", stli_brdnames[brdp->brdtype],
 		brdp->brdnr, brdp->iobase, (int) brdp->memaddr,
 		brdp->nrpanels, brdp->nrports);
@@ -3703,7 +3617,7 @@
 	if (! foundit) {
 		brdp->memaddr = 0;
 		brdp->membase = NULL;
-		printk(KERN_ERR "STALLION: failed to probe shared memory "
+		printk(KERN_ERR "istallion: failed to probe shared memory "
 				"region for %s in EISA slot=%d\n",
 			stli_brdnames[brdp->brdtype], (brdp->iobase >> 12));
 		return -ENODEV;
@@ -3848,7 +3762,7 @@
 	mutex_lock(&stli_brdslock);
 	brdnr = stli_getbrdnr();
 	if (brdnr < 0) {
-		printk(KERN_INFO "STALLION: too many boards found, "
+		printk(KERN_INFO "istallion: too many boards found, "
 			"maximum supported %d\n", STL_MAXBRDS);
 		mutex_unlock(&stli_brdslock);
 		retval = -EIO;
@@ -3920,7 +3834,7 @@
 
 	brdp = kzalloc(sizeof(struct stlibrd), GFP_KERNEL);
 	if (!brdp) {
-		printk(KERN_ERR "STALLION: failed to allocate memory "
+		printk(KERN_ERR "istallion: failed to allocate memory "
 				"(size=%Zd)\n", sizeof(struct stlibrd));
 		return NULL;
 	}
@@ -4518,6 +4432,11 @@
 	.tiocmset = stli_tiocmset,
 };
 
+static const struct tty_port_operations stli_port_ops = {
+	.carrier_raised = stli_carrier_raised,
+	.raise_dtr_rts = stli_raise_dtr_rts,
+};
+
 /*****************************************************************************/
 /*
  *	Loadable module initialization stuff.
@@ -4554,7 +4473,7 @@
 
 	stli_txcookbuf = kmalloc(STLI_TXBUFSIZE, GFP_KERNEL);
 	if (!stli_txcookbuf) {
-		printk(KERN_ERR "STALLION: failed to allocate memory "
+		printk(KERN_ERR "istallion: failed to allocate memory "
 				"(size=%d)\n", STLI_TXBUFSIZE);
 		retval = -ENOMEM;
 		goto err;
@@ -4579,7 +4498,7 @@
 
 	retval = tty_register_driver(stli_serial);
 	if (retval) {
-		printk(KERN_ERR "STALLION: failed to register serial driver\n");
+		printk(KERN_ERR "istallion: failed to register serial driver\n");
 		goto err_ttyput;
 	}
 
@@ -4593,7 +4512,7 @@
  */
 	retval = register_chrdev(STL_SIOMEMMAJOR, "staliomem", &stli_fsiomem);
 	if (retval) {
-		printk(KERN_ERR "STALLION: failed to register serial memory "
+		printk(KERN_ERR "istallion: failed to register serial memory "
 				"device\n");
 		goto err_deinit;
 	}
diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c
index 12d327a..8b0da97 100644
--- a/drivers/char/moxa.c
+++ b/drivers/char/moxa.c
@@ -206,6 +206,7 @@
 static void moxa_set_tty_param(struct tty_struct *, struct ktermios *);
 static void moxa_setup_empty_event(struct tty_struct *);
 static void moxa_shut_down(struct tty_struct *);
+static int moxa_carrier_raised(struct tty_port *);
 /*
  * moxa board interface functions:
  */
@@ -405,6 +406,10 @@
 	.tiocmset = moxa_tiocmset,
 };
 
+static const struct tty_port_operations moxa_port_ops = {
+	.carrier_raised = moxa_carrier_raised,
+};
+
 static struct tty_driver *moxaDriver;
 static DEFINE_TIMER(moxaTimer, moxa_poll, 0, 0);
 static DEFINE_SPINLOCK(moxa_lock);
@@ -826,6 +831,7 @@
 
 	for (i = 0, p = brd->ports; i < MAX_PORTS_PER_BOARD; i++, p++) {
 		tty_port_init(&p->port);
+		p->port.ops = &moxa_port_ops;
 		p->type = PORT_16550A;
 		p->cflag = B9600 | CS8 | CREAD | CLOCAL | HUPCL;
 	}
@@ -1115,15 +1121,27 @@
 	tty_port_tty_set(&ch->port, NULL);
 }
 
+static int moxa_carrier_raised(struct tty_port *port)
+{
+	struct moxa_port *ch = container_of(port, struct moxa_port, port);
+	int dcd;
+
+	spin_lock_bh(&moxa_lock);
+	dcd = ch->DCDState;
+	spin_unlock_bh(&moxa_lock);
+	return dcd;
+}
+
 static int moxa_block_till_ready(struct tty_struct *tty, struct file *filp,
 			    struct moxa_port *ch)
 {
+	struct tty_port *port = &ch->port;
 	DEFINE_WAIT(wait);
 	int retval = 0;
 	u8 dcd;
 
 	while (1) {
-		prepare_to_wait(&ch->port.open_wait, &wait, TASK_INTERRUPTIBLE);
+		prepare_to_wait(&port->open_wait, &wait, TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp)) {
 #ifdef SERIAL_DO_RESTART
 			retval = -ERESTARTSYS;
@@ -1132,9 +1150,7 @@
 #endif
 			break;
 		}
-		spin_lock_bh(&moxa_lock);
-		dcd = ch->DCDState;
-		spin_unlock_bh(&moxa_lock);
+		dcd = tty_port_carrier_raised(port);
 		if (dcd)
 			break;
 
@@ -1144,7 +1160,7 @@
 		}
 		schedule();
 	}
-	finish_wait(&ch->port.open_wait, &wait);
+	finish_wait(&port->open_wait, &wait);
 
 	return retval;
 }
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 0477669..402c9f2 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -541,74 +541,21 @@
 	return status;
 }
 
-static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
-		struct mxser_port *port)
+static int mxser_carrier_raised(struct tty_port *port)
 {
-	DECLARE_WAITQUEUE(wait, current);
-	int retval;
-	int do_clocal = 0;
+	struct mxser_port *mp = container_of(port, struct mxser_port, port);
+	return (inb(mp->ioaddr + UART_MSR) & UART_MSR_DCD)?1:0;
+}
+
+static void mxser_raise_dtr_rts(struct tty_port *port)
+{
+	struct mxser_port *mp = container_of(port, struct mxser_port, port);
 	unsigned long flags;
 
-	/*
-	 * If non-blocking mode is set, or the port is not enabled,
-	 * then make the check up front and then exit.
-	 */
-	if ((filp->f_flags & O_NONBLOCK) ||
-			test_bit(TTY_IO_ERROR, &tty->flags)) {
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (tty->termios->c_cflag & CLOCAL)
-		do_clocal = 1;
-
-	/*
-	 * Block waiting for the carrier detect and the line to become
-	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, port->port.count is dropped by one, so that
-	 * mxser_close() knows when to free things.  We restore it upon
-	 * exit, either normal or abnormal.
-	 */
-	retval = 0;
-	add_wait_queue(&port->port.open_wait, &wait);
-
-	spin_lock_irqsave(&port->slock, flags);
-	if (!tty_hung_up_p(filp))
-		port->port.count--;
-	spin_unlock_irqrestore(&port->slock, flags);
-	port->port.blocked_open++;
-	while (1) {
-		spin_lock_irqsave(&port->slock, flags);
-		outb(inb(port->ioaddr + UART_MCR) |
-			UART_MCR_DTR | UART_MCR_RTS, port->ioaddr + UART_MCR);
-		spin_unlock_irqrestore(&port->slock, flags);
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) || !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
-				(do_clocal ||
-				(inb(port->ioaddr + UART_MSR) & UART_MSR_DCD)))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-		schedule();
-	}
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
-	if (!tty_hung_up_p(filp))
-		port->port.count++;
-	port->port.blocked_open--;
-	if (retval)
-		return retval;
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
-	return 0;
+	spin_lock_irqsave(&mp->slock, flags);
+	outb(inb(mp->ioaddr + UART_MCR) |
+		UART_MCR_DTR | UART_MCR_RTS, mp->ioaddr + UART_MCR);
+	spin_unlock_irqrestore(&mp->slock, flags);
 }
 
 static int mxser_set_baud(struct tty_struct *tty, long newspd)
@@ -1087,14 +1034,14 @@
 	/*
 	 * Start up serial port
 	 */
-	spin_lock_irqsave(&info->slock, flags);
+	spin_lock_irqsave(&info->port.lock, flags);
 	info->port.count++;
-	spin_unlock_irqrestore(&info->slock, flags);
+	spin_unlock_irqrestore(&info->port.lock, flags);
 	retval = mxser_startup(tty);
 	if (retval)
 		return retval;
 
-	retval = mxser_block_til_ready(tty, filp, info);
+	retval = tty_port_block_til_ready(&info->port, tty, filp);
 	if (retval)
 		return retval;
 
@@ -1133,58 +1080,27 @@
 static void mxser_close(struct tty_struct *tty, struct file *filp)
 {
 	struct mxser_port *info = tty->driver_data;
+	struct tty_port *port = &info->port;
 
 	unsigned long timeout;
-	unsigned long flags;
 
 	if (tty->index == MXSER_PORTS)
 		return;
 	if (!info)
 		return;
 
-	spin_lock_irqsave(&info->slock, flags);
+	if (tty_port_close_start(port, tty, filp) == 0)
+		return;
 
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&info->slock, flags);
-		return;
-	}
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * Uh, oh.  tty->count is 1, which means that the tty
-		 * structure will be freed.  Info->port.count should always
-		 * be one in these conditions.  If it's greater than
-		 * one, we've got real problems, since it means the
-		 * serial port won't be shutdown.
-		 */
-		printk(KERN_ERR "mxser_close: bad serial port count; "
-			"tty->count is 1, info->port.count is %d\n", info->port.count);
-		info->port.count = 1;
-	}
-	if (--info->port.count < 0) {
-		printk(KERN_ERR "mxser_close: bad serial port count for "
-			"ttys%d: %d\n", tty->index, info->port.count);
-		info->port.count = 0;
-	}
-	if (info->port.count) {
-		spin_unlock_irqrestore(&info->slock, flags);
-		return;
-	}
-	info->port.flags |= ASYNC_CLOSING;
-	spin_unlock_irqrestore(&info->slock, flags);
 	/*
 	 * Save the termios structure, since this port may have
 	 * separate termios for callout and dialin.
+	 *
+	 * FIXME: Can this go ?
 	 */
 	if (info->port.flags & ASYNC_NORMAL_ACTIVE)
 		info->normal_termios = *tty->termios;
 	/*
-	 * Now we wait for the transmit buffer to clear; and we notify
-	 * the line discipline to only process XON/XOFF characters.
-	 */
-	tty->closing = 1;
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, info->port.closing_wait);
-	/*
 	 * At this point we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts, and tell the
 	 * interrupt driver to stop checking the data ready bit in the
@@ -1209,19 +1125,12 @@
 		}
 	}
 	mxser_shutdown(tty);
-
 	mxser_flush_buffer(tty);
-	tty_ldisc_flush(tty);
 
-	tty->closing = 0;
-	tty_port_tty_set(&info->port, NULL);
-	if (info->port.blocked_open) {
-		if (info->port.close_delay)
-			schedule_timeout_interruptible(info->port.close_delay);
-		wake_up_interruptible(&info->port.open_wait);
-	}
-
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
+	/* Right now the tty_port set is done outside of the close_end helper
+	   as we don't yet have everyone using refcounts */	
+	tty_port_close_end(port, tty);
+	tty_port_tty_set(port, NULL);
 }
 
 static int mxser_write(struct tty_struct *tty, const unsigned char *buf, int count)
@@ -2146,10 +2055,7 @@
 
 	mxser_flush_buffer(tty);
 	mxser_shutdown(tty);
-	info->port.count = 0;
-	info->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	tty_port_tty_set(&info->port, NULL);
-	wake_up_interruptible(&info->port.open_wait);
+	tty_port_hangup(&info->port);
 }
 
 /*
@@ -2449,6 +2355,11 @@
 	.tiocmset = mxser_tiocmset,
 };
 
+struct tty_port_operations mxser_port_ops = {
+	.carrier_raised = mxser_carrier_raised,
+	.raise_dtr_rts = mxser_raise_dtr_rts,
+};
+
 /*
  * The MOXA Smartio/Industio serial driver boot-time initialization code!
  */
@@ -2482,6 +2393,7 @@
 	for (i = 0; i < brd->info->nports; i++) {
 		info = &brd->ports[i];
 		tty_port_init(&info->port);
+		info->port.ops = &mxser_port_ops;
 		info->board = brd;
 		info->stop_rx = 0;
 		info->ldisc_stop_rx = 0;
diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c
index 4a8215a..d2e93e3 100644
--- a/drivers/char/n_r3964.c
+++ b/drivers/char/n_r3964.c
@@ -1003,7 +1003,7 @@
 
 static void r3964_close(struct tty_struct *tty)
 {
-	struct r3964_info *pInfo = (struct r3964_info *)tty->disc_data;
+	struct r3964_info *pInfo = tty->disc_data;
 	struct r3964_client_info *pClient, *pNext;
 	struct r3964_message *pMsg;
 	struct r3964_block_header *pHeader, *pNextHeader;
@@ -1058,7 +1058,7 @@
 static ssize_t r3964_read(struct tty_struct *tty, struct file *file,
 			  unsigned char __user * buf, size_t nr)
 {
-	struct r3964_info *pInfo = (struct r3964_info *)tty->disc_data;
+	struct r3964_info *pInfo = tty->disc_data;
 	struct r3964_client_info *pClient;
 	struct r3964_message *pMsg;
 	struct r3964_client_message theMsg;
@@ -1113,7 +1113,7 @@
 static ssize_t r3964_write(struct tty_struct *tty, struct file *file,
 			   const unsigned char *data, size_t count)
 {
-	struct r3964_info *pInfo = (struct r3964_info *)tty->disc_data;
+	struct r3964_info *pInfo = tty->disc_data;
 	struct r3964_block_header *pHeader;
 	struct r3964_client_info *pClient;
 	unsigned char *new_data;
@@ -1182,7 +1182,7 @@
 static int r3964_ioctl(struct tty_struct *tty, struct file *file,
 		unsigned int cmd, unsigned long arg)
 {
-	struct r3964_info *pInfo = (struct r3964_info *)tty->disc_data;
+	struct r3964_info *pInfo = tty->disc_data;
 	if (pInfo == NULL)
 		return -EINVAL;
 	switch (cmd) {
@@ -1216,7 +1216,7 @@
 static unsigned int r3964_poll(struct tty_struct *tty, struct file *file,
 			struct poll_table_struct *wait)
 {
-	struct r3964_info *pInfo = (struct r3964_info *)tty->disc_data;
+	struct r3964_info *pInfo = tty->disc_data;
 	struct r3964_client_info *pClient;
 	struct r3964_message *pMsg = NULL;
 	unsigned long flags;
@@ -1241,7 +1241,7 @@
 static void r3964_receive_buf(struct tty_struct *tty, const unsigned char *cp,
 			char *fp, int count)
 {
-	struct r3964_info *pInfo = (struct r3964_info *)tty->disc_data;
+	struct r3964_info *pInfo = tty->disc_data;
 	const unsigned char *p;
 	char *f, flags = 0;
 	int i;
diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c
index efbfe961..f6f0e4e 100644
--- a/drivers/char/n_tty.c
+++ b/drivers/char/n_tty.c
@@ -47,8 +47,8 @@
 #include <linux/bitops.h>
 #include <linux/audit.h>
 #include <linux/file.h>
+#include <linux/uaccess.h>
 
-#include <asm/uaccess.h>
 #include <asm/system.h>
 
 /* number of characters left in xmit buffer before select has we have room */
@@ -62,6 +62,17 @@
 #define TTY_THRESHOLD_THROTTLE		128 /* now based on remaining room */
 #define TTY_THRESHOLD_UNTHROTTLE 	128
 
+/*
+ * Special byte codes used in the echo buffer to represent operations
+ * or special handling of characters.  Bytes in the echo buffer that
+ * are not part of such special blocks are treated as normal character
+ * codes.
+ */
+#define ECHO_OP_START 0xff
+#define ECHO_OP_MOVE_BACK_COL 0x80
+#define ECHO_OP_SET_CANON_COL 0x81
+#define ECHO_OP_ERASE_TAB 0x82
+
 static inline unsigned char *alloc_buf(void)
 {
 	gfp_t prio = in_interrupt() ? GFP_ATOMIC : GFP_KERNEL;
@@ -169,6 +180,7 @@
  *
  *	Locking: tty_read_lock for read fields.
  */
+
 static void reset_buffer_flags(struct tty_struct *tty)
 {
 	unsigned long flags;
@@ -176,6 +188,11 @@
 	spin_lock_irqsave(&tty->read_lock, flags);
 	tty->read_head = tty->read_tail = tty->read_cnt = 0;
 	spin_unlock_irqrestore(&tty->read_lock, flags);
+
+	mutex_lock(&tty->echo_lock);
+	tty->echo_pos = tty->echo_cnt = tty->echo_overrun = 0;
+	mutex_unlock(&tty->echo_lock);
+
 	tty->canon_head = tty->canon_data = tty->erasing = 0;
 	memset(&tty->read_flags, 0, sizeof tty->read_flags);
 	n_tty_set_room(tty);
@@ -266,89 +283,118 @@
 }
 
 /**
- *	opost			-	output post processor
+ *	do_output_char			-	output one character
+ *	@c: character (or partial unicode symbol)
+ *	@tty: terminal device
+ *	@space: space available in tty driver write buffer
+ *
+ *	This is a helper function that handles one output character
+ *	(including special characters like TAB, CR, LF, etc.),
+ *	putting the results in the tty driver's write buffer.
+ *
+ *	Note that Linux currently ignores TABDLY, CRDLY, VTDLY, FFDLY
+ *	and NLDLY.  They simply aren't relevant in the world today.
+ *	If you ever need them, add them here.
+ *
+ *	Returns the number of bytes of buffer space used or -1 if
+ *	no space left.
+ *
+ *	Locking: should be called under the output_lock to protect
+ *		 the column state and space left in the buffer
+ */
+
+static int do_output_char(unsigned char c, struct tty_struct *tty, int space)
+{
+	int	spaces;
+
+	if (!space)
+		return -1;
+
+	switch (c) {
+	case '\n':
+		if (O_ONLRET(tty))
+			tty->column = 0;
+		if (O_ONLCR(tty)) {
+			if (space < 2)
+				return -1;
+			tty->canon_column = tty->column = 0;
+			tty_put_char(tty, '\r');
+			tty_put_char(tty, c);
+			return 2;
+		}
+		tty->canon_column = tty->column;
+		break;
+	case '\r':
+		if (O_ONOCR(tty) && tty->column == 0)
+			return 0;
+		if (O_OCRNL(tty)) {
+			c = '\n';
+			if (O_ONLRET(tty))
+				tty->canon_column = tty->column = 0;
+			break;
+		}
+		tty->canon_column = tty->column = 0;
+		break;
+	case '\t':
+		spaces = 8 - (tty->column & 7);
+		if (O_TABDLY(tty) == XTABS) {
+			if (space < spaces)
+				return -1;
+			tty->column += spaces;
+			tty->ops->write(tty, "        ", spaces);
+			return spaces;
+		}
+		tty->column += spaces;
+		break;
+	case '\b':
+		if (tty->column > 0)
+			tty->column--;
+		break;
+	default:
+		if (!iscntrl(c)) {
+			if (O_OLCUC(tty))
+				c = toupper(c);
+			if (!is_continuation(c, tty))
+				tty->column++;
+		}
+		break;
+	}
+
+	tty_put_char(tty, c);
+	return 1;
+}
+
+/**
+ *	process_output			-	output post processor
  *	@c: character (or partial unicode symbol)
  *	@tty: terminal device
  *
  *	Perform OPOST processing.  Returns -1 when the output device is
- *	full and the character must be retried. Note that Linux currently
- *	ignores TABDLY, CRDLY, VTDLY, FFDLY and NLDLY. They simply aren't
- *	relevant in the world today. If you ever need them, add them here.
+ *	full and the character must be retried.
  *
- *	Called from both the receive and transmit sides and can be called
- *	re-entrantly. Relies on lock_kernel() for tty->column state.
+ *	Locking: output_lock to protect column state and space left
+ *		 (also, this is called from n_tty_write under the
+ *		  tty layer write lock)
  */
 
-static int opost(unsigned char c, struct tty_struct *tty)
+static int process_output(unsigned char c, struct tty_struct *tty)
 {
-	int	space, spaces;
+	int	space, retval;
+
+	mutex_lock(&tty->output_lock);
 
 	space = tty_write_room(tty);
-	if (!space)
-		return -1;
+	retval = do_output_char(c, tty, space);
 
-	lock_kernel();
-	if (O_OPOST(tty)) {
-		switch (c) {
-		case '\n':
-			if (O_ONLRET(tty))
-				tty->column = 0;
-			if (O_ONLCR(tty)) {
-				if (space < 2) {
-					unlock_kernel();
-					return -1;
-				}
-				tty_put_char(tty, '\r');
-				tty->column = 0;
-			}
-			tty->canon_column = tty->column;
-			break;
-		case '\r':
-			if (O_ONOCR(tty) && tty->column == 0) {
-				unlock_kernel();
-				return 0;
-			}
-			if (O_OCRNL(tty)) {
-				c = '\n';
-				if (O_ONLRET(tty))
-					tty->canon_column = tty->column = 0;
-				break;
-			}
-			tty->canon_column = tty->column = 0;
-			break;
-		case '\t':
-			spaces = 8 - (tty->column & 7);
-			if (O_TABDLY(tty) == XTABS) {
-				if (space < spaces) {
-					unlock_kernel();
-					return -1;
-				}
-				tty->column += spaces;
-				tty->ops->write(tty, "        ", spaces);
-				unlock_kernel();
-				return 0;
-			}
-			tty->column += spaces;
-			break;
-		case '\b':
-			if (tty->column > 0)
-				tty->column--;
-			break;
-		default:
-			if (O_OLCUC(tty))
-				c = toupper(c);
-			if (!iscntrl(c) && !is_continuation(c, tty))
-				tty->column++;
-			break;
-		}
-	}
-	tty_put_char(tty, c);
-	unlock_kernel();
-	return 0;
+	mutex_unlock(&tty->output_lock);
+	if (retval < 0)
+		return -1;
+	else
+		return 0;
 }
 
 /**
- *	opost_block		-	block postprocess
+ *	process_output_block		-	block post processor
  *	@tty: terminal device
  *	@inbuf: user buffer
  *	@nr: number of bytes
@@ -358,26 +404,32 @@
  *	the simple cases normally found and helps to generate blocks of
  *	symbols for the console driver and thus improve performance.
  *
- *	Called from n_tty_write under the tty layer write lock. Relies
- *	on lock_kernel for the tty->column state.
+ *	Locking: output_lock to protect column state and space left
+ *		 (also, this is called from n_tty_write under the
+ *		  tty layer write lock)
  */
 
-static ssize_t opost_block(struct tty_struct *tty,
-		       const unsigned char *buf, unsigned int nr)
+static ssize_t process_output_block(struct tty_struct *tty,
+				    const unsigned char *buf, unsigned int nr)
 {
 	int	space;
 	int 	i;
 	const unsigned char *cp;
 
+	mutex_lock(&tty->output_lock);
+
 	space = tty_write_room(tty);
-	if (!space)
+	if (!space) {
+		mutex_unlock(&tty->output_lock);
 		return 0;
+	}
 	if (nr > space)
 		nr = space;
 
-	lock_kernel();
 	for (i = 0, cp = buf; i < nr; i++, cp++) {
-		switch (*cp) {
+		unsigned char c = *cp;
+
+		switch (c) {
 		case '\n':
 			if (O_ONLRET(tty))
 				tty->column = 0;
@@ -399,54 +451,403 @@
 				tty->column--;
 			break;
 		default:
-			if (O_OLCUC(tty))
-				goto break_out;
-			if (!iscntrl(*cp))
-				tty->column++;
+			if (!iscntrl(c)) {
+				if (O_OLCUC(tty))
+					goto break_out;
+				if (!is_continuation(c, tty))
+					tty->column++;
+			}
 			break;
 		}
 	}
 break_out:
-	if (tty->ops->flush_chars)
-		tty->ops->flush_chars(tty);
 	i = tty->ops->write(tty, buf, i);
-	unlock_kernel();
+
+	mutex_unlock(&tty->output_lock);
 	return i;
 }
 
+/**
+ *	process_echoes	-	write pending echo characters
+ *	@tty: terminal device
+ *
+ *	Write previously buffered echo (and other ldisc-generated)
+ *	characters to the tty.
+ *
+ *	Characters generated by the ldisc (including echoes) need to
+ *	be buffered because the driver's write buffer can fill during
+ *	heavy program output.  Echoing straight to the driver will
+ *	often fail under these conditions, causing lost characters and
+ *	resulting mismatches of ldisc state information.
+ *
+ *	Since the ldisc state must represent the characters actually sent
+ *	to the driver at the time of the write, operations like certain
+ *	changes in column state are also saved in the buffer and executed
+ *	here.
+ *
+ *	A circular fifo buffer is used so that the most recent characters
+ *	are prioritized.  Also, when control characters are echoed with a
+ *	prefixed "^", the pair is treated atomically and thus not separated.
+ *
+ *	Locking: output_lock to protect column state and space left,
+ *		 echo_lock to protect the echo buffer
+ */
+
+static void process_echoes(struct tty_struct *tty)
+{
+	int	space, nr;
+	unsigned char c;
+	unsigned char *cp, *buf_end;
+
+	if (!tty->echo_cnt)
+		return;
+
+	mutex_lock(&tty->output_lock);
+	mutex_lock(&tty->echo_lock);
+
+	space = tty_write_room(tty);
+
+	buf_end = tty->echo_buf + N_TTY_BUF_SIZE;
+	cp = tty->echo_buf + tty->echo_pos;
+	nr = tty->echo_cnt;
+	while (nr > 0) {
+		c = *cp;
+		if (c == ECHO_OP_START) {
+			unsigned char op;
+			unsigned char *opp;
+			int no_space_left = 0;
+
+			/*
+			 * If the buffer byte is the start of a multi-byte
+			 * operation, get the next byte, which is either the
+			 * op code or a control character value.
+			 */
+			opp = cp + 1;
+			if (opp == buf_end)
+				opp -= N_TTY_BUF_SIZE;
+			op = *opp;
+
+			switch (op) {
+				unsigned int num_chars, num_bs;
+
+			case ECHO_OP_ERASE_TAB:
+				if (++opp == buf_end)
+					opp -= N_TTY_BUF_SIZE;
+				num_chars = *opp;
+
+				/*
+				 * Determine how many columns to go back
+				 * in order to erase the tab.
+				 * This depends on the number of columns
+				 * used by other characters within the tab
+				 * area.  If this (modulo 8) count is from
+				 * the start of input rather than from a
+				 * previous tab, we offset by canon column.
+				 * Otherwise, tab spacing is normal.
+				 */
+				if (!(num_chars & 0x80))
+					num_chars += tty->canon_column;
+				num_bs = 8 - (num_chars & 7);
+
+				if (num_bs > space) {
+					no_space_left = 1;
+					break;
+				}
+				space -= num_bs;
+				while (num_bs--) {
+					tty_put_char(tty, '\b');
+					if (tty->column > 0)
+						tty->column--;
+				}
+				cp += 3;
+				nr -= 3;
+				break;
+
+			case ECHO_OP_SET_CANON_COL:
+				tty->canon_column = tty->column;
+				cp += 2;
+				nr -= 2;
+				break;
+
+			case ECHO_OP_MOVE_BACK_COL:
+				if (tty->column > 0)
+					tty->column--;
+				cp += 2;
+				nr -= 2;
+				break;
+
+			case ECHO_OP_START:
+				/* This is an escaped echo op start code */
+				if (!space) {
+					no_space_left = 1;
+					break;
+				}
+				tty_put_char(tty, ECHO_OP_START);
+				tty->column++;
+				space--;
+				cp += 2;
+				nr -= 2;
+				break;
+
+			default:
+				if (iscntrl(op)) {
+					if (L_ECHOCTL(tty)) {
+						/*
+						 * Ensure there is enough space
+						 * for the whole ctrl pair.
+						 */
+						if (space < 2) {
+							no_space_left = 1;
+							break;
+						}
+						tty_put_char(tty, '^');
+						tty_put_char(tty, op ^ 0100);
+						tty->column += 2;
+						space -= 2;
+					} else {
+						if (!space) {
+							no_space_left = 1;
+							break;
+						}
+						tty_put_char(tty, op);
+						space--;
+					}
+				}
+				/*
+				 * If above falls through, this was an
+				 * undefined op.
+				 */
+				cp += 2;
+				nr -= 2;
+			}
+
+			if (no_space_left)
+				break;
+		} else {
+			int retval;
+
+			retval = do_output_char(c, tty, space);
+			if (retval < 0)
+				break;
+			space -= retval;
+			cp += 1;
+			nr -= 1;
+		}
+
+		/* When end of circular buffer reached, wrap around */
+		if (cp >= buf_end)
+			cp -= N_TTY_BUF_SIZE;
+	}
+
+	if (nr == 0) {
+		tty->echo_pos = 0;
+		tty->echo_cnt = 0;
+		tty->echo_overrun = 0;
+	} else {
+		int num_processed = tty->echo_cnt - nr;
+		tty->echo_pos += num_processed;
+		tty->echo_pos &= N_TTY_BUF_SIZE - 1;
+		tty->echo_cnt = nr;
+		if (num_processed > 0)
+			tty->echo_overrun = 0;
+	}
+
+	mutex_unlock(&tty->echo_lock);
+	mutex_unlock(&tty->output_lock);
+
+	if (tty->ops->flush_chars)
+		tty->ops->flush_chars(tty);
+}
 
 /**
- *	echo_char	-	echo characters
+ *	add_echo_byte	-	add a byte to the echo buffer
+ *	@c: unicode byte to echo
+ *	@tty: terminal device
+ *
+ *	Add a character or operation byte to the echo buffer.
+ *
+ *	Should be called under the echo lock to protect the echo buffer.
+ */
+
+static void add_echo_byte(unsigned char c, struct tty_struct *tty)
+{
+	int	new_byte_pos;
+
+	if (tty->echo_cnt == N_TTY_BUF_SIZE) {
+		/* Circular buffer is already at capacity */
+		new_byte_pos = tty->echo_pos;
+
+		/*
+		 * Since the buffer start position needs to be advanced,
+		 * be sure to step by a whole operation byte group.
+		 */
+		if (tty->echo_buf[tty->echo_pos] == ECHO_OP_START) {
+			if (tty->echo_buf[(tty->echo_pos + 1) &
+					  (N_TTY_BUF_SIZE - 1)] ==
+						ECHO_OP_ERASE_TAB) {
+				tty->echo_pos += 3;
+				tty->echo_cnt -= 2;
+			} else {
+				tty->echo_pos += 2;
+				tty->echo_cnt -= 1;
+			}
+		} else {
+			tty->echo_pos++;
+		}
+		tty->echo_pos &= N_TTY_BUF_SIZE - 1;
+
+		tty->echo_overrun = 1;
+	} else {
+		new_byte_pos = tty->echo_pos + tty->echo_cnt;
+		new_byte_pos &= N_TTY_BUF_SIZE - 1;
+		tty->echo_cnt++;
+	}
+
+	tty->echo_buf[new_byte_pos] = c;
+}
+
+/**
+ *	echo_move_back_col	-	add operation to move back a column
+ *	@tty: terminal device
+ *
+ *	Add an operation to the echo buffer to move back one column.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_move_back_col(struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	add_echo_byte(ECHO_OP_START, tty);
+	add_echo_byte(ECHO_OP_MOVE_BACK_COL, tty);
+
+	mutex_unlock(&tty->echo_lock);
+}
+
+/**
+ *	echo_set_canon_col	-	add operation to set the canon column
+ *	@tty: terminal device
+ *
+ *	Add an operation to the echo buffer to set the canon column
+ *	to the current column.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_set_canon_col(struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	add_echo_byte(ECHO_OP_START, tty);
+	add_echo_byte(ECHO_OP_SET_CANON_COL, tty);
+
+	mutex_unlock(&tty->echo_lock);
+}
+
+/**
+ *	echo_erase_tab	-	add operation to erase a tab
+ *	@num_chars: number of character columns already used
+ *	@after_tab: true if num_chars starts after a previous tab
+ *	@tty: terminal device
+ *
+ *	Add an operation to the echo buffer to erase a tab.
+ *
+ *	Called by the eraser function, which knows how many character
+ *	columns have been used since either a previous tab or the start
+ *	of input.  This information will be used later, along with
+ *	canon column (if applicable), to go back the correct number
+ *	of columns.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_erase_tab(unsigned int num_chars, int after_tab,
+			   struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	add_echo_byte(ECHO_OP_START, tty);
+	add_echo_byte(ECHO_OP_ERASE_TAB, tty);
+
+	/* We only need to know this modulo 8 (tab spacing) */
+	num_chars &= 7;
+
+	/* Set the high bit as a flag if num_chars is after a previous tab */
+	if (after_tab)
+		num_chars |= 0x80;
+
+	add_echo_byte(num_chars, tty);
+
+	mutex_unlock(&tty->echo_lock);
+}
+
+/**
+ *	echo_char_raw	-	echo a character raw
  *	@c: unicode byte to echo
  *	@tty: terminal device
  *
  *	Echo user input back onto the screen. This must be called only when
  *	L_ECHO(tty) is true. Called from the driver receive_buf path.
  *
- *	Relies on BKL for tty column locking
+ *	This variant does not treat control characters specially.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_char_raw(unsigned char c, struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	if (c == ECHO_OP_START) {
+		add_echo_byte(ECHO_OP_START, tty);
+		add_echo_byte(ECHO_OP_START, tty);
+	} else {
+		add_echo_byte(c, tty);
+	}
+
+	mutex_unlock(&tty->echo_lock);
+}
+
+/**
+ *	echo_char	-	echo a character
+ *	@c: unicode byte to echo
+ *	@tty: terminal device
+ *
+ *	Echo user input back onto the screen. This must be called only when
+ *	L_ECHO(tty) is true. Called from the driver receive_buf path.
+ *
+ *	This variant tags control characters to be possibly echoed as
+ *	as "^X" (where X is the letter representing the control char).
+ *
+ *	Locking: echo_lock to protect the echo buffer
  */
 
 static void echo_char(unsigned char c, struct tty_struct *tty)
 {
-	if (L_ECHOCTL(tty) && iscntrl(c) && c != '\t') {
-		tty_put_char(tty, '^');
-		tty_put_char(tty, c ^ 0100);
-		tty->column += 2;
-	} else
-		opost(c, tty);
+	mutex_lock(&tty->echo_lock);
+
+	if (c == ECHO_OP_START) {
+		add_echo_byte(ECHO_OP_START, tty);
+		add_echo_byte(ECHO_OP_START, tty);
+	} else {
+		if (iscntrl(c) && c != '\t')
+			add_echo_byte(ECHO_OP_START, tty);
+		add_echo_byte(c, tty);
+	}
+
+	mutex_unlock(&tty->echo_lock);
 }
 
 /**
- *	finsh_erasing		-	complete erase
+ *	finish_erasing		-	complete erase
  *	@tty: tty doing the erase
- *
- *	Relies on BKL for tty column locking
  */
+
 static inline void finish_erasing(struct tty_struct *tty)
 {
 	if (tty->erasing) {
-		tty_put_char(tty, '/');
-		tty->column++;
+		echo_char_raw('/', tty);
 		tty->erasing = 0;
 	}
 }
@@ -460,7 +861,7 @@
  *	present in the stream from the driver layer. Handles the complexities
  *	of UTF-8 multibyte symbols.
  *
- *	Locking: read_lock for tty buffers, BKL for column/erasing state
+ *	Locking: read_lock for tty buffers
  */
 
 static void eraser(unsigned char c, struct tty_struct *tty)
@@ -471,7 +872,7 @@
 
 	/* FIXME: locking needed ? */
 	if (tty->read_head == tty->canon_head) {
-		/* opost('\a', tty); */		/* what do you think? */
+		/* process_output('\a', tty); */ /* what do you think? */
 		return;
 	}
 	if (c == ERASE_CHAR(tty))
@@ -497,7 +898,7 @@
 			echo_char(KILL_CHAR(tty), tty);
 			/* Add a newline if ECHOK is on and ECHOKE is off. */
 			if (L_ECHOK(tty))
-				opost('\n', tty);
+				echo_char_raw('\n', tty);
 			return;
 		}
 		kill_type = KILL;
@@ -533,67 +934,61 @@
 		if (L_ECHO(tty)) {
 			if (L_ECHOPRT(tty)) {
 				if (!tty->erasing) {
-					tty_put_char(tty, '\\');
-					tty->column++;
+					echo_char_raw('\\', tty);
 					tty->erasing = 1;
 				}
 				/* if cnt > 1, output a multi-byte character */
 				echo_char(c, tty);
 				while (--cnt > 0) {
 					head = (head+1) & (N_TTY_BUF_SIZE-1);
-					tty_put_char(tty, tty->read_buf[head]);
+					echo_char_raw(tty->read_buf[head], tty);
+					echo_move_back_col(tty);
 				}
 			} else if (kill_type == ERASE && !L_ECHOE(tty)) {
 				echo_char(ERASE_CHAR(tty), tty);
 			} else if (c == '\t') {
-				unsigned int col = tty->canon_column;
-				unsigned long tail = tty->canon_head;
+				unsigned int num_chars = 0;
+				int after_tab = 0;
+				unsigned long tail = tty->read_head;
 
-				/* Find the column of the last char. */
-				while (tail != tty->read_head) {
+				/*
+				 * Count the columns used for characters
+				 * since the start of input or after a
+				 * previous tab.
+				 * This info is used to go back the correct
+				 * number of columns.
+				 */
+				while (tail != tty->canon_head) {
+					tail = (tail-1) & (N_TTY_BUF_SIZE-1);
 					c = tty->read_buf[tail];
-					if (c == '\t')
-						col = (col | 7) + 1;
-					else if (iscntrl(c)) {
+					if (c == '\t') {
+						after_tab = 1;
+						break;
+					} else if (iscntrl(c)) {
 						if (L_ECHOCTL(tty))
-							col += 2;
-					} else if (!is_continuation(c, tty))
-						col++;
-					tail = (tail+1) & (N_TTY_BUF_SIZE-1);
+							num_chars += 2;
+					} else if (!is_continuation(c, tty)) {
+						num_chars++;
+					}
 				}
-
-				/* should never happen */
-				if (tty->column > 0x80000000)
-					tty->column = 0;
-
-				/* Now backup to that column. */
-				while (tty->column > col) {
-					/* Can't use opost here. */
-					tty_put_char(tty, '\b');
-					if (tty->column > 0)
-						tty->column--;
-				}
+				echo_erase_tab(num_chars, after_tab, tty);
 			} else {
 				if (iscntrl(c) && L_ECHOCTL(tty)) {
-					tty_put_char(tty, '\b');
-					tty_put_char(tty, ' ');
-					tty_put_char(tty, '\b');
-					if (tty->column > 0)
-						tty->column--;
+					echo_char_raw('\b', tty);
+					echo_char_raw(' ', tty);
+					echo_char_raw('\b', tty);
 				}
 				if (!iscntrl(c) || L_ECHOCTL(tty)) {
-					tty_put_char(tty, '\b');
-					tty_put_char(tty, ' ');
-					tty_put_char(tty, '\b');
-					if (tty->column > 0)
-						tty->column--;
+					echo_char_raw('\b', tty);
+					echo_char_raw(' ', tty);
+					echo_char_raw('\b', tty);
 				}
 			}
 		}
 		if (kill_type == ERASE)
 			break;
 	}
-	if (tty->read_head == tty->canon_head)
+	if (tty->read_head == tty->canon_head && L_ECHO(tty))
 		finish_erasing(tty);
 }
 
@@ -712,6 +1107,7 @@
 static inline void n_tty_receive_char(struct tty_struct *tty, unsigned char c)
 {
 	unsigned long flags;
+	int parmrk;
 
 	if (tty->raw) {
 		put_tty_queue(c, tty);
@@ -721,18 +1117,21 @@
 	if (I_ISTRIP(tty))
 		c &= 0x7f;
 	if (I_IUCLC(tty) && L_IEXTEN(tty))
-		c=tolower(c);
+		c = tolower(c);
 
 	if (tty->stopped && !tty->flow_stopped && I_IXON(tty) &&
-	    ((I_IXANY(tty) && c != START_CHAR(tty) && c != STOP_CHAR(tty)) ||
-	     c == INTR_CHAR(tty) || c == QUIT_CHAR(tty) || c == SUSP_CHAR(tty)))
+	    I_IXANY(tty) && c != START_CHAR(tty) && c != STOP_CHAR(tty) &&
+	    c != INTR_CHAR(tty) && c != QUIT_CHAR(tty) && c != SUSP_CHAR(tty)) {
 		start_tty(tty);
+		process_echoes(tty);
+	}
 
 	if (tty->closing) {
 		if (I_IXON(tty)) {
-			if (c == START_CHAR(tty))
+			if (c == START_CHAR(tty)) {
 				start_tty(tty);
-			else if (c == STOP_CHAR(tty))
+				process_echoes(tty);
+			} else if (c == STOP_CHAR(tty))
 				stop_tty(tty);
 		}
 		return;
@@ -745,19 +1144,23 @@
 	 * up.
 	 */
 	if (!test_bit(c, tty->process_char_map) || tty->lnext) {
-		finish_erasing(tty);
 		tty->lnext = 0;
+		parmrk = (c == (unsigned char) '\377' && I_PARMRK(tty)) ? 1 : 0;
+		if (tty->read_cnt >= (N_TTY_BUF_SIZE - parmrk - 1)) {
+			/* beep if no space */
+			if (L_ECHO(tty))
+				process_output('\a', tty);
+			return;
+		}
 		if (L_ECHO(tty)) {
-			if (tty->read_cnt >= N_TTY_BUF_SIZE-1) {
-				tty_put_char(tty, '\a'); /* beep if no space */
-				return;
-			}
+			finish_erasing(tty);
 			/* Record the column of first canon char. */
 			if (tty->canon_head == tty->read_head)
-				tty->canon_column = tty->column;
+				echo_set_canon_col(tty);
 			echo_char(c, tty);
+			process_echoes(tty);
 		}
-		if (I_PARMRK(tty) && c == (unsigned char) '\377')
+		if (parmrk)
 			put_tty_queue(c, tty);
 		put_tty_queue(c, tty);
 		return;
@@ -766,6 +1169,7 @@
 	if (I_IXON(tty)) {
 		if (c == START_CHAR(tty)) {
 			start_tty(tty);
+			process_echoes(tty);
 			return;
 		}
 		if (c == STOP_CHAR(tty)) {
@@ -786,7 +1190,6 @@
 		if (c == SUSP_CHAR(tty)) {
 send_signal:
 			/*
-			 * Echo character, and then send the signal.
 			 * Note that we do not use isig() here because we want
 			 * the order to be:
 			 * 1) flush, 2) echo, 3) signal
@@ -795,8 +1198,12 @@
 				n_tty_flush_buffer(tty);
 				tty_driver_flush_buffer(tty);
 			}
-			if (L_ECHO(tty))
+			if (I_IXON(tty))
+				start_tty(tty);
+			if (L_ECHO(tty)) {
 				echo_char(c, tty);
+				process_echoes(tty);
+			}
 			if (tty->pgrp)
 				kill_pgrp(tty->pgrp, signal, 1);
 			return;
@@ -815,6 +1222,7 @@
 		if (c == ERASE_CHAR(tty) || c == KILL_CHAR(tty) ||
 		    (c == WERASE_CHAR(tty) && L_IEXTEN(tty))) {
 			eraser(c, tty);
+			process_echoes(tty);
 			return;
 		}
 		if (c == LNEXT_CHAR(tty) && L_IEXTEN(tty)) {
@@ -822,8 +1230,9 @@
 			if (L_ECHO(tty)) {
 				finish_erasing(tty);
 				if (L_ECHOCTL(tty)) {
-					tty_put_char(tty, '^');
-					tty_put_char(tty, '\b');
+					echo_char_raw('^', tty);
+					echo_char_raw('\b', tty);
+					process_echoes(tty);
 				}
 			}
 			return;
@@ -834,22 +1243,29 @@
 
 			finish_erasing(tty);
 			echo_char(c, tty);
-			opost('\n', tty);
+			echo_char_raw('\n', tty);
 			while (tail != tty->read_head) {
 				echo_char(tty->read_buf[tail], tty);
 				tail = (tail+1) & (N_TTY_BUF_SIZE-1);
 			}
+			process_echoes(tty);
 			return;
 		}
 		if (c == '\n') {
+			if (tty->read_cnt >= N_TTY_BUF_SIZE) {
+				if (L_ECHO(tty))
+					process_output('\a', tty);
+				return;
+			}
 			if (L_ECHO(tty) || L_ECHONL(tty)) {
-				if (tty->read_cnt >= N_TTY_BUF_SIZE-1)
-					tty_put_char(tty, '\a');
-				opost('\n', tty);
+				echo_char_raw('\n', tty);
+				process_echoes(tty);
 			}
 			goto handle_newline;
 		}
 		if (c == EOF_CHAR(tty)) {
+			if (tty->read_cnt >= N_TTY_BUF_SIZE)
+				return;
 			if (tty->canon_head != tty->read_head)
 				set_bit(TTY_PUSH, &tty->flags);
 			c = __DISABLED_CHAR;
@@ -857,22 +1273,28 @@
 		}
 		if ((c == EOL_CHAR(tty)) ||
 		    (c == EOL2_CHAR(tty) && L_IEXTEN(tty))) {
+			parmrk = (c == (unsigned char) '\377' && I_PARMRK(tty))
+				 ? 1 : 0;
+			if (tty->read_cnt >= (N_TTY_BUF_SIZE - parmrk)) {
+				if (L_ECHO(tty))
+					process_output('\a', tty);
+				return;
+			}
 			/*
 			 * XXX are EOL_CHAR and EOL2_CHAR echoed?!?
 			 */
 			if (L_ECHO(tty)) {
-				if (tty->read_cnt >= N_TTY_BUF_SIZE-1)
-					tty_put_char(tty, '\a');
 				/* Record the column of first canon char. */
 				if (tty->canon_head == tty->read_head)
-					tty->canon_column = tty->column;
+					echo_set_canon_col(tty);
 				echo_char(c, tty);
+				process_echoes(tty);
 			}
 			/*
 			 * XXX does PARMRK doubling happen for
 			 * EOL_CHAR and EOL2_CHAR?
 			 */
-			if (I_PARMRK(tty) && c == (unsigned char) '\377')
+			if (parmrk)
 				put_tty_queue(c, tty);
 
 handle_newline:
@@ -889,23 +1311,27 @@
 		}
 	}
 
-	finish_erasing(tty);
+	parmrk = (c == (unsigned char) '\377' && I_PARMRK(tty)) ? 1 : 0;
+	if (tty->read_cnt >= (N_TTY_BUF_SIZE - parmrk - 1)) {
+		/* beep if no space */
+		if (L_ECHO(tty))
+			process_output('\a', tty);
+		return;
+	}
 	if (L_ECHO(tty)) {
-		if (tty->read_cnt >= N_TTY_BUF_SIZE-1) {
-			tty_put_char(tty, '\a'); /* beep if no space */
-			return;
-		}
+		finish_erasing(tty);
 		if (c == '\n')
-			opost('\n', tty);
+			echo_char_raw('\n', tty);
 		else {
 			/* Record the column of first canon char. */
 			if (tty->canon_head == tty->read_head)
-				tty->canon_column = tty->column;
+				echo_set_canon_col(tty);
 			echo_char(c, tty);
 		}
+		process_echoes(tty);
 	}
 
-	if (I_PARMRK(tty) && c == (unsigned char) '\377')
+	if (parmrk)
 		put_tty_queue(c, tty);
 
 	put_tty_queue(c, tty);
@@ -923,10 +1349,11 @@
 
 static void n_tty_write_wakeup(struct tty_struct *tty)
 {
-	if (tty->fasync) {
-		set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
+	/* Write out any echoed characters that are still pending */
+	process_echoes(tty);
+
+	if (tty->fasync && test_and_clear_bit(TTY_DO_WRITE_WAKEUP, &tty->flags))
 		kill_fasync(&tty->fasync, SIGIO, POLL_OUT);
-	}
 }
 
 /**
@@ -1134,6 +1561,10 @@
 		free_buf(tty->read_buf);
 		tty->read_buf = NULL;
 	}
+	if (tty->echo_buf) {
+		free_buf(tty->echo_buf);
+		tty->echo_buf = NULL;
+	}
 }
 
 /**
@@ -1151,13 +1582,19 @@
 	if (!tty)
 		return -EINVAL;
 
-	/* This one is ugly. Currently a malloc failure here can panic */
+	/* These are ugly. Currently a malloc failure here can panic */
 	if (!tty->read_buf) {
 		tty->read_buf = alloc_buf();
 		if (!tty->read_buf)
 			return -ENOMEM;
 	}
+	if (!tty->echo_buf) {
+		tty->echo_buf = alloc_buf();
+		if (!tty->echo_buf)
+			return -ENOMEM;
+	}
 	memset(tty->read_buf, 0, N_TTY_BUF_SIZE);
+	memset(tty->echo_buf, 0, N_TTY_BUF_SIZE);
 	reset_buffer_flags(tty);
 	tty->column = 0;
 	n_tty_set_termios(tty, NULL);
@@ -1487,16 +1924,23 @@
  *	@buf: userspace buffer pointer
  *	@nr: size of I/O
  *
- *	Write function of the terminal device. This is serialized with
+ *	Write function of the terminal device.  This is serialized with
  *	respect to other write callers but not to termios changes, reads
- *	and other such events. We must be careful with N_TTY as the receive
- *	code will echo characters, thus calling driver write methods.
+ *	and other such events.  Since the receive code will echo characters,
+ *	thus calling driver write methods, the output_lock is used in
+ *	the output processing functions called here as well as in the
+ *	echo processing function to protect the column state and space
+ *	left in the buffer.
  *
  *	This code must be sure never to sleep through a hangup.
+ *
+ *	Locking: output_lock to protect column state and space left
+ *		 (note that the process_output*() functions take this
+ *		  lock themselves)
  */
 
 static ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
-			  const unsigned char *buf, size_t nr)
+			   const unsigned char *buf, size_t nr)
 {
 	const unsigned char *b = buf;
 	DECLARE_WAITQUEUE(wait, current);
@@ -1510,6 +1954,9 @@
 			return retval;
 	}
 
+	/* Write out any echoed characters that are still pending */
+	process_echoes(tty);
+
 	add_wait_queue(&tty->write_wait, &wait);
 	while (1) {
 		set_current_state(TASK_INTERRUPTIBLE);
@@ -1523,7 +1970,7 @@
 		}
 		if (O_OPOST(tty) && !(test_bit(TTY_HW_COOK_OUT, &tty->flags))) {
 			while (nr > 0) {
-				ssize_t num = opost_block(tty, b, nr);
+				ssize_t num = process_output_block(tty, b, nr);
 				if (num < 0) {
 					if (num == -EAGAIN)
 						break;
@@ -1535,7 +1982,7 @@
 				if (nr == 0)
 					break;
 				c = *b;
-				if (opost(c, tty) < 0)
+				if (process_output(c, tty) < 0)
 					break;
 				b++; nr--;
 			}
@@ -1565,6 +2012,8 @@
 break_out:
 	__set_current_state(TASK_RUNNING);
 	remove_wait_queue(&tty->write_wait, &wait);
+	if (b - buf != nr && tty->fasync)
+		set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
 	return (b - buf) ? b - buf : retval;
 }
 
@@ -1663,4 +2112,3 @@
 	.receive_buf     = n_tty_receive_buf,
 	.write_wakeup    = n_tty_write_wakeup
 };
-
diff --git a/drivers/char/nozomi.c b/drivers/char/nozomi.c
index 9a34a19..d6102b6 100644
--- a/drivers/char/nozomi.c
+++ b/drivers/char/nozomi.c
@@ -353,6 +353,7 @@
 
 /* This holds all information that is needed regarding a port */
 struct port {
+	struct tty_port port;
 	u8 update_flow_control;
 	struct ctrl_ul ctrl_ul;
 	struct ctrl_dl ctrl_dl;
@@ -365,8 +366,6 @@
 	u8 toggle_ul;
 	u16 token_dl;
 
-	struct tty_struct *tty;
-	int tty_open_count;
 	/* mutex to ensure one access patch to this port */
 	struct mutex tty_sem;
 	wait_queue_head_t tty_wait;
@@ -788,14 +787,14 @@
  * Return 1 - send buffer to card and ack.
  * Return 0 - don't ack, don't send buffer to card.
  */
-static int send_data(enum port_type index, const struct nozomi *dc)
+static int send_data(enum port_type index, struct nozomi *dc)
 {
 	u32 size = 0;
-	const struct port *port = &dc->port[index];
+	struct port *port = &dc->port[index];
 	const u8 toggle = port->toggle_ul;
 	void __iomem *addr = port->ul_addr[toggle];
 	const u32 ul_size = port->ul_size[toggle];
-	struct tty_struct *tty = port->tty;
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
 
 	/* Get data from tty and place in buf for now */
 	size = __kfifo_get(port->fifo_ul, dc->send_buf,
@@ -803,6 +802,7 @@
 
 	if (size == 0) {
 		DBG4("No more data to send, disable link:");
+		tty_kref_put(tty);
 		return 0;
 	}
 
@@ -815,6 +815,7 @@
 	if (tty)
 		tty_wakeup(tty);
 
+	tty_kref_put(tty);
 	return 1;
 }
 
@@ -826,7 +827,7 @@
 	u32 offset = 4;
 	struct port *port = &dc->port[index];
 	void __iomem *addr = port->dl_addr[port->toggle_dl];
-	struct tty_struct *tty = port->tty;
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
 	int i;
 
 	if (unlikely(!tty)) {
@@ -870,7 +871,7 @@
 	}
 
 	set_bit(index, &dc->flip);
-
+	tty_kref_put(tty);
 	return 1;
 }
 
@@ -1276,9 +1277,15 @@
 
 exit_handler:
 	spin_unlock(&dc->spin_mutex);
-	for (a = 0; a < NOZOMI_MAX_PORTS; a++)
-		if (test_and_clear_bit(a, &dc->flip))
-			tty_flip_buffer_push(dc->port[a].tty);
+	for (a = 0; a < NOZOMI_MAX_PORTS; a++) {
+		struct tty_struct *tty;
+		if (test_and_clear_bit(a, &dc->flip)) {
+			tty = tty_port_tty_get(&dc->port[a].port);
+			if (tty)
+				tty_flip_buffer_push(tty);
+			tty_kref_put(tty);
+		}
+	}
 	return IRQ_HANDLED;
 none:
 	spin_unlock(&dc->spin_mutex);
@@ -1453,12 +1460,10 @@
 
 	for (i = 0; i < MAX_PORT; i++) {
 		mutex_init(&dc->port[i].tty_sem);
-		dc->port[i].tty_open_count = 0;
-		dc->port[i].tty = NULL;
+		tty_port_init(&dc->port[i].port);
 		tty_register_device(ntty_driver, dc->index_start + i,
 							&pdev->dev);
 	}
-
 	return 0;
 
 err_free_sbuf:
@@ -1482,14 +1487,16 @@
 
 	flush_scheduled_work();
 
-	for (i = 0; i < MAX_PORT; ++i)
-		if (dc->port[i].tty && \
-				list_empty(&dc->port[i].tty->hangup_work.entry))
-			tty_hangup(dc->port[i].tty);
-
+	for (i = 0; i < MAX_PORT; ++i) {
+		struct tty_struct *tty = tty_port_tty_get(&dc->port[i].port);
+		if (tty && list_empty(&tty->hangup_work.entry))
+			tty_hangup(tty);
+		tty_kref_put(tty);
+	}
+	/* Racy below - surely should wait for scheduled work to be done or
+	   complete off a hangup method ? */
 	while (dc->open_ttys)
 		msleep(1);
-
 	for (i = dc->index_start; i < dc->index_start + MAX_PORT; ++i)
 		tty_unregister_device(ntty_driver, i);
 }
@@ -1579,23 +1586,22 @@
 	if (mutex_lock_interruptible(&port->tty_sem))
 		return -ERESTARTSYS;
 
-	port->tty_open_count++;
+	port->port.count++;
 	dc->open_ttys++;
 
 	/* Enable interrupt downlink for channel */
-	if (port->tty_open_count == 1) {
+	if (port->port.count == 1) {
+		/* FIXME: is this needed now ? */
 		tty->low_latency = 1;
 		tty->driver_data = port;
-		port->tty = tty;
+		tty_port_tty_set(&port->port, tty);
 		DBG1("open: %d", port->token_dl);
 		spin_lock_irqsave(&dc->spin_mutex, flags);
 		dc->last_ier = dc->last_ier | port->token_dl;
 		writew(dc->last_ier, dc->reg_ier);
 		spin_unlock_irqrestore(&dc->spin_mutex, flags);
 	}
-
 	mutex_unlock(&port->tty_sem);
-
 	return 0;
 }
 
@@ -1606,31 +1612,30 @@
 static void ntty_close(struct tty_struct *tty, struct file *file)
 {
 	struct nozomi *dc = get_dc_by_tty(tty);
-	struct port *port = tty->driver_data;
+	struct port *nport = tty->driver_data;
+	struct tty_port *port = &nport->port;
 	unsigned long flags;
 
-	if (!dc || !port)
+	if (!dc || !nport)
 		return;
 
-	if (mutex_lock_interruptible(&port->tty_sem))
-		return;
+	/* Users cannot interrupt a close */
+	mutex_lock(&nport->tty_sem);
 
-	if (!port->tty_open_count)
-		goto exit;
+	WARN_ON(!port->count);
 
 	dc->open_ttys--;
-	port->tty_open_count--;
+	port->count--;
+	tty_port_tty_set(port, NULL);
 
-	if (port->tty_open_count == 0) {
-		DBG1("close: %d", port->token_dl);
+	if (port->count == 0) {
+		DBG1("close: %d", nport->token_dl);
 		spin_lock_irqsave(&dc->spin_mutex, flags);
-		dc->last_ier &= ~(port->token_dl);
+		dc->last_ier &= ~(nport->token_dl);
 		writew(dc->last_ier, dc->reg_ier);
 		spin_unlock_irqrestore(&dc->spin_mutex, flags);
 	}
-
-exit:
-	mutex_unlock(&port->tty_sem);
+	mutex_unlock(&nport->tty_sem);
 }
 
 /*
@@ -1660,7 +1665,7 @@
 		return -EAGAIN;
 	}
 
-	if (unlikely(!port->tty_open_count)) {
+	if (unlikely(!port->port.count)) {
 		DBG1(" ");
 		goto exit;
 	}
@@ -1710,7 +1715,7 @@
 	if (!mutex_trylock(&port->tty_sem))
 		return 0;
 
-	if (!port->tty_open_count)
+	if (!port->port.count)
 		goto exit;
 
 	room = port->fifo_ul->size - __kfifo_len(port->fifo_ul);
@@ -1866,7 +1871,7 @@
 		goto exit_in_buffer;
 	}
 
-	if (unlikely(!port->tty_open_count)) {
+	if (unlikely(!port->port.count)) {
 		dev_err(&dc->pdev->dev, "No tty open?\n");
 		rval = -ENODEV;
 		goto exit_in_buffer;
diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 4d64a02..dc073e1 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -138,20 +138,15 @@
  */
 
 typedef struct _mgslpc_info {
+	struct tty_port		port;
 	void *if_ptr;	/* General purpose pointer (used by SPPP) */
 	int			magic;
-	int			flags;
-	int			count;		/* count of opens */
 	int			line;
-	unsigned short		close_delay;
-	unsigned short		closing_wait;	/* time to wait before closing */
 
 	struct mgsl_icount	icount;
 
-	struct tty_struct 	*tty;
 	int			timeout;
 	int			x_char;		/* xon/xoff character */
-	int			blocked_open;	/* # of blocked opens */
 	unsigned char		read_status_mask;
 	unsigned char		ignore_status_mask;
 
@@ -170,9 +165,6 @@
 	int            rx_buf_count;   /* total number of rx buffers */
 	int            rx_frame_count; /* number of full rx buffers */
 
-	wait_queue_head_t	open_wait;
-	wait_queue_head_t	close_wait;
-
 	wait_queue_head_t	status_event_wait_q;
 	wait_queue_head_t	event_wait_q;
 	struct timer_list	tx_timer;	/* HDLC transmit timeout timer */
@@ -375,7 +367,7 @@
 static void rx_start(MGSLPC_INFO *info);
 static void rx_stop(MGSLPC_INFO *info);
 
-static void tx_start(MGSLPC_INFO *info);
+static void tx_start(MGSLPC_INFO *info, struct tty_struct *tty);
 static void tx_stop(MGSLPC_INFO *info);
 static void tx_set_idle(MGSLPC_INFO *info);
 
@@ -389,7 +381,8 @@
 
 static void tx_timeout(unsigned long context);
 
-static int ioctl_common(MGSLPC_INFO *info, unsigned int cmd, unsigned long arg);
+static int carrier_raised(struct tty_port *port);
+static void raise_dtr_rts(struct tty_port *port);
 
 #if SYNCLINK_GENERIC_HDLC
 #define dev_to_port(D) (dev_to_hdlc(D)->priv)
@@ -410,7 +403,7 @@
 static void mgslpc_add_device(MGSLPC_INFO *info);
 static void mgslpc_remove_device(MGSLPC_INFO *info);
 
-static bool rx_get_frame(MGSLPC_INFO *info);
+static bool rx_get_frame(MGSLPC_INFO *info, struct tty_struct *tty);
 static void rx_reset_buffers(MGSLPC_INFO *info);
 static int  rx_alloc_buffers(MGSLPC_INFO *info);
 static void rx_free_buffers(MGSLPC_INFO *info);
@@ -421,7 +414,7 @@
  * Bottom half interrupt handlers
  */
 static void bh_handler(struct work_struct *work);
-static void bh_transmit(MGSLPC_INFO *info);
+static void bh_transmit(MGSLPC_INFO *info, struct tty_struct *tty);
 static void bh_status(MGSLPC_INFO *info);
 
 /*
@@ -432,10 +425,10 @@
 		    unsigned int set, unsigned int clear);
 static int get_stats(MGSLPC_INFO *info, struct mgsl_icount __user *user_icount);
 static int get_params(MGSLPC_INFO *info, MGSL_PARAMS __user *user_params);
-static int set_params(MGSLPC_INFO *info, MGSL_PARAMS __user *new_params);
+static int set_params(MGSLPC_INFO *info, MGSL_PARAMS __user *new_params, struct tty_struct *tty);
 static int get_txidle(MGSLPC_INFO *info, int __user *idle_mode);
 static int set_txidle(MGSLPC_INFO *info, int idle_mode);
-static int set_txenable(MGSLPC_INFO *info, int enable);
+static int set_txenable(MGSLPC_INFO *info, int enable, struct tty_struct *tty);
 static int tx_abort(MGSLPC_INFO *info);
 static int set_rxenable(MGSLPC_INFO *info, int enable);
 static int wait_events(MGSLPC_INFO *info, int __user *mask);
@@ -474,7 +467,7 @@
 /* number of characters left in xmit buffer before we ask for more */
 #define WAKEUP_CHARS 256
 
-static void mgslpc_change_params(MGSLPC_INFO *info);
+static void mgslpc_change_params(MGSLPC_INFO *info, struct tty_struct *tty);
 static void mgslpc_wait_until_sent(struct tty_struct *tty, int timeout);
 
 /* PCMCIA prototypes */
@@ -517,6 +510,11 @@
 	}
 }
 
+static const struct tty_port_operations mgslpc_port_ops = {
+	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts
+};
+
 static int mgslpc_probe(struct pcmcia_device *link)
 {
     MGSLPC_INFO *info;
@@ -532,12 +530,12 @@
     }
 
     info->magic = MGSLPC_MAGIC;
+    tty_port_init(&info->port);
+    info->port.ops = &mgslpc_port_ops;
     INIT_WORK(&info->task, bh_handler);
     info->max_frame_size = 4096;
-    info->close_delay = 5*HZ/10;
-    info->closing_wait = 30*HZ;
-    init_waitqueue_head(&info->open_wait);
-    init_waitqueue_head(&info->close_wait);
+    info->port.close_delay = 5*HZ/10;
+    info->port.closing_wait = 30*HZ;
     init_waitqueue_head(&info->status_event_wait_q);
     init_waitqueue_head(&info->event_wait_q);
     spin_lock_init(&info->lock);
@@ -784,7 +782,7 @@
 
 	spin_lock_irqsave(&info->lock,flags);
 	if (!info->tx_enabled)
-	 	tx_start(info);
+	 	tx_start(info, tty);
 	spin_unlock_irqrestore(&info->lock,flags);
 }
 
@@ -823,6 +821,7 @@
 static void bh_handler(struct work_struct *work)
 {
 	MGSLPC_INFO *info = container_of(work, MGSLPC_INFO, task);
+	struct tty_struct *tty;
 	int action;
 
 	if (!info)
@@ -833,6 +832,7 @@
 			__FILE__,__LINE__,info->device_name);
 
 	info->bh_running = true;
+	tty = tty_port_tty_get(&info->port);
 
 	while((action = bh_action(info)) != 0) {
 
@@ -844,10 +844,10 @@
 		switch (action) {
 
 		case BH_RECEIVE:
-			while(rx_get_frame(info));
+			while(rx_get_frame(info, tty));
 			break;
 		case BH_TRANSMIT:
-			bh_transmit(info);
+			bh_transmit(info, tty);
 			break;
 		case BH_STATUS:
 			bh_status(info);
@@ -859,14 +859,14 @@
 		}
 	}
 
+	tty_kref_put(tty);
 	if (debug_level >= DEBUG_LEVEL_BH)
 		printk( "%s(%d):bh_handler(%s) exit\n",
 			__FILE__,__LINE__,info->device_name);
 }
 
-static void bh_transmit(MGSLPC_INFO *info)
+static void bh_transmit(MGSLPC_INFO *info, struct tty_struct *tty)
 {
-	struct tty_struct *tty = info->tty;
 	if (debug_level >= DEBUG_LEVEL_BH)
 		printk("bh_transmit() entry on %s\n", info->device_name);
 
@@ -945,12 +945,11 @@
 	issue_command(info, CHA, CMD_RXFIFO);
 }
 
-static void rx_ready_async(MGSLPC_INFO *info, int tcd)
+static void rx_ready_async(MGSLPC_INFO *info, int tcd, struct tty_struct *tty)
 {
 	unsigned char data, status, flag;
 	int fifo_count;
 	int work = 0;
- 	struct tty_struct *tty = info->tty;
  	struct mgsl_icount *icount = &info->icount;
 
 	if (tcd) {
@@ -1013,7 +1012,7 @@
 }
 
 
-static void tx_done(MGSLPC_INFO *info)
+static void tx_done(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	if (!info->tx_active)
 		return;
@@ -1042,7 +1041,7 @@
 	else
 #endif
 	{
-		if (info->tty->stopped || info->tty->hw_stopped) {
+		if (tty->stopped || tty->hw_stopped) {
 			tx_stop(info);
 			return;
 		}
@@ -1050,7 +1049,7 @@
 	}
 }
 
-static void tx_ready(MGSLPC_INFO *info)
+static void tx_ready(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	unsigned char fifo_count = 32;
 	int c;
@@ -1062,7 +1061,7 @@
 		if (!info->tx_active)
 			return;
 	} else {
-		if (info->tty->stopped || info->tty->hw_stopped) {
+		if (tty->stopped || tty->hw_stopped) {
 			tx_stop(info);
 			return;
 		}
@@ -1099,7 +1098,7 @@
 	}
 }
 
-static void cts_change(MGSLPC_INFO *info)
+static void cts_change(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	get_signals(info);
 	if ((info->cts_chkcount)++ >= IO_PIN_SHUTDOWN_LIMIT)
@@ -1112,14 +1111,14 @@
 	wake_up_interruptible(&info->status_event_wait_q);
 	wake_up_interruptible(&info->event_wait_q);
 
-	if (info->flags & ASYNC_CTS_FLOW) {
-		if (info->tty->hw_stopped) {
+	if (info->port.flags & ASYNC_CTS_FLOW) {
+		if (tty->hw_stopped) {
 			if (info->serial_signals & SerialSignal_CTS) {
 				if (debug_level >= DEBUG_LEVEL_ISR)
 					printk("CTS tx start...");
-				if (info->tty)
-					info->tty->hw_stopped = 0;
-				tx_start(info);
+				if (tty)
+					tty->hw_stopped = 0;
+				tx_start(info, tty);
 				info->pending_bh |= BH_TRANSMIT;
 				return;
 			}
@@ -1127,8 +1126,8 @@
 			if (!(info->serial_signals & SerialSignal_CTS)) {
 				if (debug_level >= DEBUG_LEVEL_ISR)
 					printk("CTS tx stop...");
-				if (info->tty)
-					info->tty->hw_stopped = 1;
+				if (tty)
+					tty->hw_stopped = 1;
 				tx_stop(info);
 			}
 		}
@@ -1136,7 +1135,7 @@
 	info->pending_bh |= BH_STATUS;
 }
 
-static void dcd_change(MGSLPC_INFO *info)
+static void dcd_change(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	get_signals(info);
 	if ((info->dcd_chkcount)++ >= IO_PIN_SHUTDOWN_LIMIT)
@@ -1158,17 +1157,17 @@
 	wake_up_interruptible(&info->status_event_wait_q);
 	wake_up_interruptible(&info->event_wait_q);
 
-	if (info->flags & ASYNC_CHECK_CD) {
+	if (info->port.flags & ASYNC_CHECK_CD) {
 		if (debug_level >= DEBUG_LEVEL_ISR)
 			printk("%s CD now %s...", info->device_name,
 			       (info->serial_signals & SerialSignal_DCD) ? "on" : "off");
 		if (info->serial_signals & SerialSignal_DCD)
-			wake_up_interruptible(&info->open_wait);
+			wake_up_interruptible(&info->port.open_wait);
 		else {
 			if (debug_level >= DEBUG_LEVEL_ISR)
 				printk("doing serial hangup...");
-			if (info->tty)
-				tty_hangup(info->tty);
+			if (tty)
+				tty_hangup(tty);
 		}
 	}
 	info->pending_bh |= BH_STATUS;
@@ -1214,6 +1213,7 @@
 static irqreturn_t mgslpc_isr(int dummy, void *dev_id)
 {
 	MGSLPC_INFO *info = dev_id;
+	struct tty_struct *tty;
 	unsigned short isr;
 	unsigned char gis, pis;
 	int count=0;
@@ -1224,6 +1224,8 @@
 	if (!(info->p_dev->_locked))
 		return IRQ_HANDLED;
 
+	tty = tty_port_tty_get(&info->port);
+
 	spin_lock(&info->lock);
 
 	while ((gis = read_reg(info, CHA + GIS))) {
@@ -1239,9 +1241,9 @@
 		if (gis & (BIT1 + BIT0)) {
 			isr = read_reg16(info, CHB + ISR);
 			if (isr & IRQ_DCD)
-				dcd_change(info);
+				dcd_change(info, tty);
 			if (isr & IRQ_CTS)
-				cts_change(info);
+				cts_change(info, tty);
 		}
 		if (gis & (BIT3 + BIT2))
 		{
@@ -1258,8 +1260,8 @@
 			}
 			if (isr & IRQ_BREAK_ON) {
 				info->icount.brk++;
-				if (info->flags & ASYNC_SAK)
-					do_SAK(info->tty);
+				if (info->port.flags & ASYNC_SAK)
+					do_SAK(tty);
 			}
 			if (isr & IRQ_RXTIME) {
 				issue_command(info, CHA, CMD_RXFIFO_READ);
@@ -1268,7 +1270,7 @@
 				if (info->params.mode == MGSL_MODE_HDLC)
 					rx_ready_hdlc(info, isr & IRQ_RXEOM);
 				else
-					rx_ready_async(info, isr & IRQ_RXEOM);
+					rx_ready_async(info, isr & IRQ_RXEOM, tty);
 			}
 
 			/* transmit IRQs */
@@ -1277,14 +1279,14 @@
 					info->icount.txabort++;
 				else
 					info->icount.txunder++;
-				tx_done(info);
+				tx_done(info, tty);
 			}
 			else if (isr & IRQ_ALLSENT) {
 				info->icount.txok++;
-				tx_done(info);
+				tx_done(info, tty);
 			}
 			else if (isr & IRQ_TXFIFO)
-				tx_ready(info);
+				tx_ready(info, tty);
 		}
 		if (gis & BIT7) {
 			pis = read_reg(info, CHA + PIS);
@@ -1308,6 +1310,7 @@
 	}
 
 	spin_unlock(&info->lock);
+	tty_kref_put(tty);
 
 	if (debug_level >= DEBUG_LEVEL_ISR)
 		printk("%s(%d):mgslpc_isr(%d)exit.\n",
@@ -1318,14 +1321,14 @@
 
 /* Initialize and start device.
  */
-static int startup(MGSLPC_INFO * info)
+static int startup(MGSLPC_INFO * info, struct tty_struct *tty)
 {
 	int retval = 0;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):startup(%s)\n",__FILE__,__LINE__,info->device_name);
 
-	if (info->flags & ASYNC_INITIALIZED)
+	if (info->port.flags & ASYNC_INITIALIZED)
 		return 0;
 
 	if (!info->tx_buf) {
@@ -1352,30 +1355,30 @@
 		retval = adapter_test(info);
 
 	if ( retval ) {
-  		if (capable(CAP_SYS_ADMIN) && info->tty)
-			set_bit(TTY_IO_ERROR, &info->tty->flags);
+  		if (capable(CAP_SYS_ADMIN) && tty)
+			set_bit(TTY_IO_ERROR, &tty->flags);
 		release_resources(info);
   		return retval;
   	}
 
 	/* program hardware for current parameters */
-	mgslpc_change_params(info);
+	mgslpc_change_params(info, tty);
 
-	if (info->tty)
-		clear_bit(TTY_IO_ERROR, &info->tty->flags);
+	if (tty)
+		clear_bit(TTY_IO_ERROR, &tty->flags);
 
-	info->flags |= ASYNC_INITIALIZED;
+	info->port.flags |= ASYNC_INITIALIZED;
 
 	return 0;
 }
 
 /* Called by mgslpc_close() and mgslpc_hangup() to shutdown hardware
  */
-static void shutdown(MGSLPC_INFO * info)
+static void shutdown(MGSLPC_INFO * info, struct tty_struct *tty)
 {
 	unsigned long flags;
 
-	if (!(info->flags & ASYNC_INITIALIZED))
+	if (!(info->port.flags & ASYNC_INITIALIZED))
 		return;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1402,7 +1405,7 @@
 	/* TODO:disable interrupts instead of reset to preserve signal states */
 	reset_device(info);
 
- 	if (!info->tty || info->tty->termios->c_cflag & HUPCL) {
+ 	if (!tty || tty->termios->c_cflag & HUPCL) {
  		info->serial_signals &= ~(SerialSignal_DTR + SerialSignal_RTS);
 		set_signals(info);
 	}
@@ -1411,13 +1414,13 @@
 
 	release_resources(info);
 
-	if (info->tty)
-		set_bit(TTY_IO_ERROR, &info->tty->flags);
+	if (tty)
+		set_bit(TTY_IO_ERROR, &tty->flags);
 
-	info->flags &= ~ASYNC_INITIALIZED;
+	info->port.flags &= ~ASYNC_INITIALIZED;
 }
 
-static void mgslpc_program_hw(MGSLPC_INFO *info)
+static void mgslpc_program_hw(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	unsigned long flags;
 
@@ -1443,7 +1446,7 @@
 	port_irq_enable(info, (unsigned char) PVR_DSR | PVR_RI);
 	get_signals(info);
 
-	if (info->netcount || info->tty->termios->c_cflag & CREAD)
+	if (info->netcount || (tty && (tty->termios->c_cflag & CREAD)))
 		rx_start(info);
 
 	spin_unlock_irqrestore(&info->lock,flags);
@@ -1451,19 +1454,19 @@
 
 /* Reconfigure adapter based on new parameters
  */
-static void mgslpc_change_params(MGSLPC_INFO *info)
+static void mgslpc_change_params(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	unsigned cflag;
 	int bits_per_char;
 
-	if (!info->tty || !info->tty->termios)
+	if (!tty || !tty->termios)
 		return;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgslpc_change_params(%s)\n",
 			 __FILE__,__LINE__, info->device_name );
 
-	cflag = info->tty->termios->c_cflag;
+	cflag = tty->termios->c_cflag;
 
 	/* if B0 rate (hangup) specified then negate DTR and RTS */
 	/* otherwise assert DTR and RTS */
@@ -1510,7 +1513,7 @@
 	 * current data rate.
 	 */
 	if (info->params.data_rate <= 460800) {
-		info->params.data_rate = tty_get_baud_rate(info->tty);
+		info->params.data_rate = tty_get_baud_rate(tty);
 	}
 
 	if ( info->params.data_rate ) {
@@ -1520,24 +1523,24 @@
 	info->timeout += HZ/50;		/* Add .02 seconds of slop */
 
 	if (cflag & CRTSCTS)
-		info->flags |= ASYNC_CTS_FLOW;
+		info->port.flags |= ASYNC_CTS_FLOW;
 	else
-		info->flags &= ~ASYNC_CTS_FLOW;
+		info->port.flags &= ~ASYNC_CTS_FLOW;
 
 	if (cflag & CLOCAL)
-		info->flags &= ~ASYNC_CHECK_CD;
+		info->port.flags &= ~ASYNC_CHECK_CD;
 	else
-		info->flags |= ASYNC_CHECK_CD;
+		info->port.flags |= ASYNC_CHECK_CD;
 
 	/* process tty input control flags */
 
 	info->read_status_mask = 0;
-	if (I_INPCK(info->tty))
+	if (I_INPCK(tty))
 		info->read_status_mask |= BIT7 | BIT6;
-	if (I_IGNPAR(info->tty))
+	if (I_IGNPAR(tty))
 		info->ignore_status_mask |= BIT7 | BIT6;
 
-	mgslpc_program_hw(info);
+	mgslpc_program_hw(info, tty);
 }
 
 /* Add a character to the transmit buffer
@@ -1597,7 +1600,7 @@
 
 	spin_lock_irqsave(&info->lock,flags);
 	if (!info->tx_active)
-	 	tx_start(info);
+	 	tx_start(info, tty);
 	spin_unlock_irqrestore(&info->lock,flags);
 }
 
@@ -1659,7 +1662,7 @@
  	if (info->tx_count && !tty->stopped && !tty->hw_stopped) {
 		spin_lock_irqsave(&info->lock,flags);
 		if (!info->tx_active)
-		 	tx_start(info);
+		 	tx_start(info, tty);
 		spin_unlock_irqrestore(&info->lock,flags);
  	}
 cleanup:
@@ -1764,7 +1767,7 @@
 	if (ch) {
 		spin_lock_irqsave(&info->lock,flags);
 		if (!info->tx_enabled)
-		 	tx_start(info);
+		 	tx_start(info, tty);
 		spin_unlock_irqrestore(&info->lock,flags);
 	}
 }
@@ -1862,7 +1865,7 @@
  *
  * Returns:	0 if success, otherwise error code
  */
-static int set_params(MGSLPC_INFO * info, MGSL_PARAMS __user *new_params)
+static int set_params(MGSLPC_INFO * info, MGSL_PARAMS __user *new_params, struct tty_struct *tty)
 {
  	unsigned long flags;
 	MGSL_PARAMS tmp_params;
@@ -1883,7 +1886,7 @@
 	memcpy(&info->params,&tmp_params,sizeof(MGSL_PARAMS));
 	spin_unlock_irqrestore(&info->lock,flags);
 
- 	mgslpc_change_params(info);
+ 	mgslpc_change_params(info, tty);
 
 	return 0;
 }
@@ -1944,7 +1947,7 @@
 	return 0;
 }
 
-static int set_txenable(MGSLPC_INFO * info, int enable)
+static int set_txenable(MGSLPC_INFO * info, int enable, struct tty_struct *tty)
 {
  	unsigned long flags;
 
@@ -1954,7 +1957,7 @@
 	spin_lock_irqsave(&info->lock,flags);
 	if (enable) {
 		if (!info->tx_enabled)
-			tx_start(info);
+			tx_start(info, tty);
 	} else {
 		if (info->tx_enabled)
 			tx_stop(info);
@@ -2263,6 +2266,11 @@
 			unsigned int cmd, unsigned long arg)
 {
 	MGSLPC_INFO * info = (MGSLPC_INFO *)tty->driver_data;
+	int error;
+	struct mgsl_icount cnow;	/* kernel counter temps */
+	struct serial_icounter_struct __user *p_cuser;	/* user space */
+	void __user *argp = (void __user *)arg;
+	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgslpc_ioctl %s cmd=%08X\n", __FILE__,__LINE__,
@@ -2277,22 +2285,11 @@
 		    return -EIO;
 	}
 
-	return ioctl_common(info, cmd, arg);
-}
-
-static int ioctl_common(MGSLPC_INFO *info, unsigned int cmd, unsigned long arg)
-{
-	int error;
-	struct mgsl_icount cnow;	/* kernel counter temps */
-	struct serial_icounter_struct __user *p_cuser;	/* user space */
-	void __user *argp = (void __user *)arg;
-	unsigned long flags;
-
 	switch (cmd) {
 	case MGSL_IOCGPARAMS:
 		return get_params(info, argp);
 	case MGSL_IOCSPARAMS:
-		return set_params(info, argp);
+		return set_params(info, argp, tty);
 	case MGSL_IOCGTXIDLE:
 		return get_txidle(info, argp);
 	case MGSL_IOCSTXIDLE:
@@ -2302,7 +2299,7 @@
 	case MGSL_IOCSIF:
 		return set_interface(info,(int)arg);
 	case MGSL_IOCTXENABLE:
-		return set_txenable(info,(int)arg);
+		return set_txenable(info,(int)arg, tty);
 	case MGSL_IOCRXENABLE:
 		return set_rxenable(info,(int)arg);
 	case MGSL_IOCTXABORT:
@@ -2369,7 +2366,7 @@
 		== RELEVANT_IFLAG(old_termios->c_iflag)))
 	  return;
 
-	mgslpc_change_params(info);
+	mgslpc_change_params(info, tty);
 
 	/* Handle transition to B0 status */
 	if (old_termios->c_cflag & CBAUD &&
@@ -2404,81 +2401,34 @@
 static void mgslpc_close(struct tty_struct *tty, struct file * filp)
 {
 	MGSLPC_INFO * info = (MGSLPC_INFO *)tty->driver_data;
+	struct tty_port *port = &info->port;
 
 	if (mgslpc_paranoia_check(info, tty->name, "mgslpc_close"))
 		return;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgslpc_close(%s) entry, count=%d\n",
-			 __FILE__,__LINE__, info->device_name, info->count);
+			 __FILE__,__LINE__, info->device_name, port->count);
 
-	if (!info->count)
-		return;
+	WARN_ON(!port->count);
 
-	if (tty_hung_up_p(filp))
+	if (tty_port_close_start(port, tty, filp) == 0)
 		goto cleanup;
 
-	if ((tty->count == 1) && (info->count != 1)) {
-		/*
-		 * tty->count is 1 and the tty structure will be freed.
-		 * info->count should be one in this case.
-		 * if it's not, correct it so that the port is shutdown.
-		 */
-		printk("mgslpc_close: bad refcount; tty->count is 1, "
-		       "info->count is %d\n", info->count);
-		info->count = 1;
-	}
-
-	info->count--;
-
-	/* if at least one open remaining, leave hardware active */
-	if (info->count)
-		goto cleanup;
-
-	info->flags |= ASYNC_CLOSING;
-
-	/* set tty->closing to notify line discipline to
-	 * only process XON/XOFF characters. Only the N_TTY
-	 * discipline appears to use this (ppp does not).
-	 */
-	tty->closing = 1;
-
-	/* wait for transmit data to clear all layers */
-
-	if (info->closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		if (debug_level >= DEBUG_LEVEL_INFO)
-			printk("%s(%d):mgslpc_close(%s) calling tty_wait_until_sent\n",
-				 __FILE__,__LINE__, info->device_name );
-		tty_wait_until_sent(tty, info->closing_wait);
-	}
-
- 	if (info->flags & ASYNC_INITIALIZED)
+ 	if (port->flags & ASYNC_INITIALIZED)
  		mgslpc_wait_until_sent(tty, info->timeout);
 
 	mgslpc_flush_buffer(tty);
 
 	tty_ldisc_flush(tty);
-
-	shutdown(info);
-
-	tty->closing = 0;
-	info->tty = NULL;
-
-	if (info->blocked_open) {
-		if (info->close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->close_delay));
-		}
-		wake_up_interruptible(&info->open_wait);
-	}
-
-	info->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-
-	wake_up_interruptible(&info->close_wait);
-
+	shutdown(info, tty);
+	
+	tty_port_close_end(port, tty);
+	tty_port_tty_set(port, NULL);
 cleanup:
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgslpc_close(%s) exit, count=%d\n", __FILE__,__LINE__,
-			tty->driver->name, info->count);
+			tty->driver->name, port->count);
 }
 
 /* Wait until the transmitter is empty.
@@ -2498,7 +2448,7 @@
 	if (mgslpc_paranoia_check(info, tty->name, "mgslpc_wait_until_sent"))
 		return;
 
-	if (!(info->flags & ASYNC_INITIALIZED))
+	if (!(info->port.flags & ASYNC_INITIALIZED))
 		goto exit;
 
 	orig_jiffies = jiffies;
@@ -2559,120 +2509,40 @@
 		return;
 
 	mgslpc_flush_buffer(tty);
-	shutdown(info);
-
-	info->count = 0;
-	info->flags &= ~ASYNC_NORMAL_ACTIVE;
-	info->tty = NULL;
-
-	wake_up_interruptible(&info->open_wait);
+	shutdown(info, tty);
+	tty_port_hangup(&info->port);
 }
 
-/* Block the current process until the specified port
- * is ready to be opened.
- */
-static int block_til_ready(struct tty_struct *tty, struct file *filp,
-			   MGSLPC_INFO *info)
+static int carrier_raised(struct tty_port *port)
 {
-	DECLARE_WAITQUEUE(wait, current);
-	int		retval;
-	bool		do_clocal = false;
-	bool		extra_count = false;
-	unsigned long	flags;
+	MGSLPC_INFO *info = container_of(port, MGSLPC_INFO, port);
+	unsigned long flags;
 
-	if (debug_level >= DEBUG_LEVEL_INFO)
-		printk("%s(%d):block_til_ready on %s\n",
-			 __FILE__,__LINE__, tty->driver->name );
+	spin_lock_irqsave(&info->lock,flags);
+ 	get_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
 
-	if (filp->f_flags & O_NONBLOCK || tty->flags & (1 << TTY_IO_ERROR)){
-		/* nonblock mode is set or port is not enabled */
-		/* just verify that callout device is not active */
-		info->flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (tty->termios->c_cflag & CLOCAL)
-		do_clocal = true;
-
-	/* Wait for carrier detect and the line to become
-	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->count is dropped by one, so that
-	 * mgslpc_close() knows when to free things.  We restore it upon
-	 * exit, either normal or abnormal.
-	 */
-
-	retval = 0;
-	add_wait_queue(&info->open_wait, &wait);
-
-	if (debug_level >= DEBUG_LEVEL_INFO)
-		printk("%s(%d):block_til_ready before block on %s count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->count );
-
-	spin_lock_irqsave(&info->lock, flags);
-	if (!tty_hung_up_p(filp)) {
-		extra_count = true;
-		info->count--;
-	}
-	spin_unlock_irqrestore(&info->lock, flags);
-	info->blocked_open++;
-
-	while (1) {
-		if ((tty->termios->c_cflag & CBAUD)) {
-			spin_lock_irqsave(&info->lock,flags);
-			info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
-		 	set_signals(info);
-			spin_unlock_irqrestore(&info->lock,flags);
-		}
-
-		set_current_state(TASK_INTERRUPTIBLE);
-
-		if (tty_hung_up_p(filp) || !(info->flags & ASYNC_INITIALIZED)){
-			retval = (info->flags & ASYNC_HUP_NOTIFY) ?
-					-EAGAIN : -ERESTARTSYS;
-			break;
-		}
-
-		spin_lock_irqsave(&info->lock,flags);
-	 	get_signals(info);
-		spin_unlock_irqrestore(&info->lock,flags);
-
- 		if (!(info->flags & ASYNC_CLOSING) &&
- 		    (do_clocal || (info->serial_signals & SerialSignal_DCD)) ) {
- 			break;
-		}
-
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-
-		if (debug_level >= DEBUG_LEVEL_INFO)
-			printk("%s(%d):block_til_ready blocking on %s count=%d\n",
-				 __FILE__,__LINE__, tty->driver->name, info->count );
-
-		schedule();
-	}
-
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->open_wait, &wait);
-
-	if (extra_count)
-		info->count++;
-	info->blocked_open--;
-
-	if (debug_level >= DEBUG_LEVEL_INFO)
-		printk("%s(%d):block_til_ready after blocking on %s count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->count );
-
-	if (!retval)
-		info->flags |= ASYNC_NORMAL_ACTIVE;
-
-	return retval;
+	if (info->serial_signals & SerialSignal_DCD)
+		return 1;
+	return 0;
 }
 
+static void raise_dtr_rts(struct tty_port *port)
+{
+	MGSLPC_INFO *info = container_of(port, MGSLPC_INFO, port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&info->lock,flags);
+	info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
+	set_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+}
+
+
 static int mgslpc_open(struct tty_struct *tty, struct file * filp)
 {
 	MGSLPC_INFO	*info;
+	struct tty_port *port;
 	int 			retval, line;
 	unsigned long flags;
 
@@ -2691,23 +2561,24 @@
 	if (mgslpc_paranoia_check(info, tty->name, "mgslpc_open"))
 		return -ENODEV;
 
+	port = &info->port;
 	tty->driver_data = info;
-	info->tty = tty;
+	tty_port_tty_set(port, tty);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgslpc_open(%s), old ref count = %d\n",
-			 __FILE__,__LINE__,tty->driver->name, info->count);
+			 __FILE__,__LINE__,tty->driver->name, port->count);
 
 	/* If port is closing, signal caller to try again */
-	if (tty_hung_up_p(filp) || info->flags & ASYNC_CLOSING){
-		if (info->flags & ASYNC_CLOSING)
-			interruptible_sleep_on(&info->close_wait);
-		retval = ((info->flags & ASYNC_HUP_NOTIFY) ?
+	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING){
+		if (port->flags & ASYNC_CLOSING)
+			interruptible_sleep_on(&port->close_wait);
+		retval = ((port->flags & ASYNC_HUP_NOTIFY) ?
 			-EAGAIN : -ERESTARTSYS);
 		goto cleanup;
 	}
 
-	info->tty->low_latency = (info->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
+	tty->low_latency = (port->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
 
 	spin_lock_irqsave(&info->netlock, flags);
 	if (info->netcount) {
@@ -2715,17 +2586,19 @@
 		spin_unlock_irqrestore(&info->netlock, flags);
 		goto cleanup;
 	}
-	info->count++;
+	spin_lock(&port->lock);
+	port->count++;
+	spin_unlock(&port->lock);
 	spin_unlock_irqrestore(&info->netlock, flags);
 
-	if (info->count == 1) {
+	if (port->count == 1) {
 		/* 1st open on this device, init hardware */
-		retval = startup(info);
+		retval = startup(info, tty);
 		if (retval < 0)
 			goto cleanup;
 	}
 
-	retval = block_til_ready(tty, filp, info);
+	retval = tty_port_block_til_ready(&info->port, tty, filp);
 	if (retval) {
 		if (debug_level >= DEBUG_LEVEL_INFO)
 			printk("%s(%d):block_til_ready(%s) returned %d\n",
@@ -2739,13 +2612,6 @@
 	retval = 0;
 
 cleanup:
-	if (retval) {
-		if (tty->count == 1)
-			info->tty = NULL; /* tty layer will release tty struct */
-		if(info->count)
-			info->count--;
-	}
-
 	return retval;
 }
 
@@ -3500,7 +3366,7 @@
 	info->rx_enabled = true;
 }
 
-static void tx_start(MGSLPC_INFO *info)
+static void tx_start(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	if (debug_level >= DEBUG_LEVEL_ISR)
 		printk("%s(%d):tx_start(%s)\n",
@@ -3524,11 +3390,11 @@
 		if (info->params.mode == MGSL_MODE_ASYNC) {
 			if (!info->tx_active) {
 				info->tx_active = true;
-				tx_ready(info);
+				tx_ready(info, tty);
 			}
 		} else {
 			info->tx_active = true;
-			tx_ready(info);
+			tx_ready(info, tty);
 			mod_timer(&info->tx_timer, jiffies +
 					msecs_to_jiffies(5000));
 		}
@@ -3849,13 +3715,12 @@
  *
  * Returns true if frame returned, otherwise false
  */
-static bool rx_get_frame(MGSLPC_INFO *info)
+static bool rx_get_frame(MGSLPC_INFO *info, struct tty_struct *tty)
 {
 	unsigned short status;
 	RXBUF *buf;
 	unsigned int framesize = 0;
 	unsigned long flags;
-	struct tty_struct *tty = info->tty;
 	bool return_frame = false;
 
 	if (info->rx_frame_count == 0)
@@ -4075,7 +3940,11 @@
 		hdlcdev_tx_done(info);
 	else
 #endif
-		bh_transmit(info);
+	{
+		struct tty_struct *tty = tty_port_tty_get(&info->port);
+		bh_transmit(info, tty);
+		tty_kref_put(tty);
+	}
 }
 
 #if SYNCLINK_GENERIC_HDLC
@@ -4094,11 +3963,12 @@
 			  unsigned short parity)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
+	struct tty_struct *tty;
 	unsigned char  new_encoding;
 	unsigned short new_crctype;
 
 	/* return error if TTY interface open */
-	if (info->count)
+	if (info->port.count)
 		return -EBUSY;
 
 	switch (encoding)
@@ -4123,8 +3993,11 @@
 	info->params.crc_type = new_crctype;
 
 	/* if network interface up, reprogram hardware */
-	if (info->netcount)
-		mgslpc_program_hw(info);
+	if (info->netcount) {
+		tty = tty_port_tty_get(&info->port);
+		mgslpc_program_hw(info, tty);
+		tty_kref_put(tty);
+	}
 
 	return 0;
 }
@@ -4165,8 +4038,11 @@
 
 	/* start hardware transmitter if necessary */
 	spin_lock_irqsave(&info->lock,flags);
-	if (!info->tx_active)
-	 	tx_start(info);
+	if (!info->tx_active) {
+		struct tty_struct *tty = tty_port_tty_get(&info->port);
+	 	tx_start(info, tty);
+	 	tty_kref_put(tty);
+	}
 	spin_unlock_irqrestore(&info->lock,flags);
 
 	return 0;
@@ -4183,6 +4059,7 @@
 static int hdlcdev_open(struct net_device *dev)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
+	struct tty_struct *tty;
 	int rc;
 	unsigned long flags;
 
@@ -4195,7 +4072,7 @@
 
 	/* arbitrate between network and tty opens */
 	spin_lock_irqsave(&info->netlock, flags);
-	if (info->count != 0 || info->netcount != 0) {
+	if (info->port.count != 0 || info->netcount != 0) {
 		printk(KERN_WARNING "%s: hdlc_open returning busy\n", dev->name);
 		spin_unlock_irqrestore(&info->netlock, flags);
 		return -EBUSY;
@@ -4203,17 +4080,19 @@
 	info->netcount=1;
 	spin_unlock_irqrestore(&info->netlock, flags);
 
+	tty = tty_port_tty_get(&info->port);
 	/* claim resources and init adapter */
-	if ((rc = startup(info)) != 0) {
+	if ((rc = startup(info, tty)) != 0) {
+		tty_kref_put(tty);
 		spin_lock_irqsave(&info->netlock, flags);
 		info->netcount=0;
 		spin_unlock_irqrestore(&info->netlock, flags);
 		return rc;
 	}
-
 	/* assert DTR and RTS, apply hardware settings */
 	info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
-	mgslpc_program_hw(info);
+	mgslpc_program_hw(info, tty);
+	tty_kref_put(tty);
 
 	/* enable network layer transmit */
 	dev->trans_start = jiffies;
@@ -4241,6 +4120,7 @@
 static int hdlcdev_close(struct net_device *dev)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
+	struct tty_struct *tty = tty_port_tty_get(&info->port);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -4249,8 +4129,8 @@
 	netif_stop_queue(dev);
 
 	/* shutdown adapter and release resources */
-	shutdown(info);
-
+	shutdown(info, tty);
+	tty_kref_put(tty);
 	hdlc_close(dev);
 
 	spin_lock_irqsave(&info->netlock, flags);
@@ -4281,7 +4161,7 @@
 		printk("%s:hdlcdev_ioctl(%s)\n",__FILE__,dev->name);
 
 	/* return error if TTY interface open */
-	if (info->count)
+	if (info->port.count)
 		return -EBUSY;
 
 	if (cmd != SIOCWANDEV)
@@ -4354,8 +4234,11 @@
 			info->params.clock_speed = 0;
 
 		/* if network interface up, reprogram hardware */
-		if (info->netcount)
-			mgslpc_program_hw(info);
+		if (info->netcount) {
+			struct tty_struct *tty = tty_port_tty_get(&info->port);
+			mgslpc_program_hw(info, tty);
+			tty_kref_put(tty);
+		}
 		return 0;
 
 	default:
diff --git a/drivers/char/pty.c b/drivers/char/pty.c
index 6d45827..112a6ba 100644
--- a/drivers/char/pty.c
+++ b/drivers/char/pty.c
@@ -5,8 +5,6 @@
  *
  *  Added support for a Unix98-style ptmx device.
  *    -- C. Scott Ananian <cananian@alumni.princeton.edu>, 14-Jan-1998
- *  Added TTY_DO_WRITE_WAKEUP to enable n_tty to send POLL_OUT to
- *      waiting writers -- Sapan Bhatia <sapan@corewars.org>
  *
  *  When reading this code see also fs/devpts. In particular note that the
  *  driver_data field is used by the devpts side as a binding to the devpts
@@ -217,7 +215,6 @@
 
 	clear_bit(TTY_OTHER_CLOSED, &tty->link->flags);
 	set_bit(TTY_THROTTLED, &tty->flags);
-	set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
 	retval = 0;
 out:
 	return retval;
@@ -230,6 +227,55 @@
 	tty->termios->c_cflag |= (CS8 | CREAD);
 }
 
+/**
+ *	pty_do_resize		-	resize event
+ *	@tty: tty being resized
+ *	@real_tty: real tty (not the same as tty if using a pty/tty pair)
+ *	@rows: rows (character)
+ *	@cols: cols (character)
+ *
+ *	Update the termios variables and send the neccessary signals to
+ *	peform a terminal resize correctly
+ */
+
+int pty_resize(struct tty_struct *tty,  struct winsize *ws)
+{
+	struct pid *pgrp, *rpgrp;
+	unsigned long flags;
+	struct tty_struct *pty = tty->link;
+
+	/* For a PTY we need to lock the tty side */
+	mutex_lock(&tty->termios_mutex);
+	if (!memcmp(ws, &tty->winsize, sizeof(*ws)))
+		goto done;
+
+	/* Get the PID values and reference them so we can
+	   avoid holding the tty ctrl lock while sending signals.
+	   We need to lock these individually however. */
+
+	spin_lock_irqsave(&tty->ctrl_lock, flags);
+	pgrp = get_pid(tty->pgrp);
+	spin_unlock_irqrestore(&tty->ctrl_lock, flags);
+
+	spin_lock_irqsave(&pty->ctrl_lock, flags);
+	rpgrp = get_pid(pty->pgrp);
+	spin_unlock_irqrestore(&pty->ctrl_lock, flags);
+
+	if (pgrp)
+		kill_pgrp(pgrp, SIGWINCH, 1);
+	if (rpgrp != pgrp && rpgrp)
+		kill_pgrp(rpgrp, SIGWINCH, 1);
+
+	put_pid(pgrp);
+	put_pid(rpgrp);
+
+	tty->winsize = *ws;
+	pty->winsize = *ws;	/* Never used so will go away soon */
+done:
+	mutex_unlock(&tty->termios_mutex);
+	return 0;
+}
+
 static int pty_install(struct tty_driver *driver, struct tty_struct *tty)
 {
 	struct tty_struct *o_tty;
@@ -290,6 +336,7 @@
 	.chars_in_buffer = pty_chars_in_buffer,
 	.unthrottle = pty_unthrottle,
 	.set_termios = pty_set_termios,
+	.resize = pty_resize
 };
 
 /* Traditional BSD devices */
@@ -319,6 +366,7 @@
 	.unthrottle = pty_unthrottle,
 	.set_termios = pty_set_termios,
 	.ioctl = pty_bsd_ioctl,
+	.resize = pty_resize
 };
 
 static void __init legacy_pty_init(void)
@@ -561,7 +609,8 @@
 	.unthrottle = pty_unthrottle,
 	.set_termios = pty_set_termios,
 	.ioctl = pty_unix98_ioctl,
-	.shutdown = pty_unix98_shutdown
+	.shutdown = pty_unix98_shutdown,
+	.resize = pty_resize
 };
 
 static const struct tty_operations pty_unix98_ops = {
diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c
index a8f68a3..2e8a6ee 100644
--- a/drivers/char/rio/rio_linux.c
+++ b/drivers/char/rio/rio_linux.c
@@ -173,7 +173,7 @@
 static void rio_enable_tx_interrupts(void *ptr);
 static void rio_disable_rx_interrupts(void *ptr);
 static void rio_enable_rx_interrupts(void *ptr);
-static int rio_get_CD(void *ptr);
+static int rio_carrier_raised(struct tty_port *port);
 static void rio_shutdown_port(void *ptr);
 static int rio_set_real_termios(void *ptr);
 static void rio_hungup(void *ptr);
@@ -224,7 +224,6 @@
 	rio_enable_tx_interrupts,
 	rio_disable_rx_interrupts,
 	rio_enable_rx_interrupts,
-	rio_get_CD,
 	rio_shutdown_port,
 	rio_set_real_termios,
 	rio_chars_in_buffer,
@@ -476,9 +475,9 @@
 
 
 /* Jeez. Isn't this simple?  */
-static int rio_get_CD(void *ptr)
+static int rio_carrier_raised(struct tty_port *port)
 {
-	struct Port *PortP = ptr;
+	struct Port *PortP = container_of(port, struct Port, gs.port);
 	int rv;
 
 	func_enter();
@@ -797,16 +796,9 @@
 	return 1;
 }
 
-
-static void *ckmalloc(int size)
-{
-	void *p;
-
-	p = kzalloc(size, GFP_KERNEL);
-	return p;
-}
-
-
+static const struct tty_port_operations rio_port_ops = {
+	.carrier_raised = rio_carrier_raised,
+};
 
 static int rio_init_datastructures(void)
 {
@@ -826,33 +818,30 @@
 #define TMIO_SZ sizeof(struct termios *)
 	rio_dprintk(RIO_DEBUG_INIT, "getting : %Zd %Zd %Zd %Zd %Zd bytes\n", RI_SZ, RIO_HOSTS * HOST_SZ, RIO_PORTS * PORT_SZ, RIO_PORTS * TMIO_SZ, RIO_PORTS * TMIO_SZ);
 
-	if (!(p = ckmalloc(RI_SZ)))
+	if (!(p = kzalloc(RI_SZ, GFP_KERNEL)))
 		goto free0;
-	if (!(p->RIOHosts = ckmalloc(RIO_HOSTS * HOST_SZ)))
+	if (!(p->RIOHosts = kzalloc(RIO_HOSTS * HOST_SZ, GFP_KERNEL)))
 		goto free1;
-	if (!(p->RIOPortp = ckmalloc(RIO_PORTS * PORT_SZ)))
+	if (!(p->RIOPortp = kzalloc(RIO_PORTS * PORT_SZ, GFP_KERNEL)))
 		goto free2;
 	p->RIOConf = RIOConf;
 	rio_dprintk(RIO_DEBUG_INIT, "Got : %p %p %p\n", p, p->RIOHosts, p->RIOPortp);
 
 #if 1
 	for (i = 0; i < RIO_PORTS; i++) {
-		port = p->RIOPortp[i] = ckmalloc(sizeof(struct Port));
+		port = p->RIOPortp[i] = kzalloc(sizeof(struct Port), GFP_KERNEL);
 		if (!port) {
 			goto free6;
 		}
 		rio_dprintk(RIO_DEBUG_INIT, "initing port %d (%d)\n", i, port->Mapped);
+		tty_port_init(&port->gs.port);
+		port->gs.port.ops = &rio_port_ops;
 		port->PortNum = i;
 		port->gs.magic = RIO_MAGIC;
 		port->gs.close_delay = HZ / 2;
 		port->gs.closing_wait = 30 * HZ;
 		port->gs.rd = &rio_real_driver;
 		spin_lock_init(&port->portSem);
-		/*
-		 * Initializing wait queue
-		 */
-		init_waitqueue_head(&port->gs.port.open_wait);
-		init_waitqueue_head(&port->gs.port.close_wait);
 	}
 #else
 	/* We could postpone initializing them to when they are configured. */
diff --git a/drivers/char/riscom8.c b/drivers/char/riscom8.c
index 2c6c8f3..9af8d74 100644
--- a/drivers/char/riscom8.c
+++ b/drivers/char/riscom8.c
@@ -857,98 +857,21 @@
 		rc_shutdown_board(bp);
 }
 
-static int block_til_ready(struct tty_struct *tty, struct file *filp,
-			   struct riscom_port *port)
+static int carrier_raised(struct tty_port *port)
 {
-	DECLARE_WAITQUEUE(wait, current);
-	struct riscom_board *bp = port_Board(port);
-	int    retval;
-	int    do_clocal = 0;
-	int    CD;
+	struct riscom_port *p = container_of(port, struct riscom_port, port);
+	struct riscom_board *bp = port_Board(p);
 	unsigned long flags;
-
-	/*
-	 * If the device is in the middle of being closed, then block
-	 * until it's done, and then try again.
-	 */
-	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&port->port.close_wait);
-		if (port->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		else
-			return -ERESTARTSYS;
-	}
-
-	/*
-	 * If non-blocking mode is set, or the port is not enabled,
-	 * then make the check up front and then exit.
-	 */
-	if ((filp->f_flags & O_NONBLOCK) ||
-	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (C_CLOCAL(tty))
-		do_clocal = 1;
-
-	/*
-	 * Block waiting for the carrier detect and the line to become
-	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->count is dropped by one, so that
-	 * rs_close() knows when to free things.  We restore it upon
-	 * exit, either normal or abnormal.
-	 */
-	retval = 0;
-	add_wait_queue(&port->port.open_wait, &wait);
-
+	int CD;
+	
 	spin_lock_irqsave(&riscom_lock, flags);
-
-	if (!tty_hung_up_p(filp))
-		port->port.count--;
-
+	rc_out(bp, CD180_CAR, port_No(p));
+	CD = rc_in(bp, CD180_MSVR) & MSVR_CD;
+	rc_out(bp, CD180_MSVR, MSVR_RTS);
+	bp->DTR &= ~(1u << port_No(p));
+	rc_out(bp, RC_DTR, bp->DTR);
 	spin_unlock_irqrestore(&riscom_lock, flags);
-
-	port->port.blocked_open++;
-	while (1) {
-		spin_lock_irqsave(&riscom_lock, flags);
-
-		rc_out(bp, CD180_CAR, port_No(port));
-		CD = rc_in(bp, CD180_MSVR) & MSVR_CD;
-		rc_out(bp, CD180_MSVR, MSVR_RTS);
-		bp->DTR &= ~(1u << port_No(port));
-		rc_out(bp, RC_DTR, bp->DTR);
-
-		spin_unlock_irqrestore(&riscom_lock, flags);
-
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) ||
-		    !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
-		    (do_clocal || CD))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-		schedule();
-	}
-	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
-	if (!tty_hung_up_p(filp))
-		port->port.count++;
-	port->port.blocked_open--;
-	if (retval)
-		return retval;
-
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
-	return 0;
+	return CD;
 }
 
 static int rc_open(struct tty_struct *tty, struct file *filp)
@@ -977,13 +900,13 @@
 
 	error = rc_setup_port(bp, port);
 	if (error == 0)
-		error = block_til_ready(tty, filp, port);
+		error = tty_port_block_til_ready(&port->port, tty, filp);
 	return error;
 }
 
 static void rc_flush_buffer(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	unsigned long flags;
 
 	if (rc_paranoia_check(port, tty->name, "rc_flush_buffer"))
@@ -998,7 +921,7 @@
 
 static void rc_close(struct tty_struct *tty, struct file *filp)
 {
-	struct riscom_port *port = (struct riscom_port *) tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
 	unsigned long flags;
 	unsigned long timeout;
@@ -1006,40 +929,19 @@
 	if (!port || rc_paranoia_check(port, tty->name, "close"))
 		return;
 
-	spin_lock_irqsave(&riscom_lock, flags);
-
-	if (tty_hung_up_p(filp))
-		goto out;
-
 	bp = port_Board(port);
-	if ((tty->count == 1) && (port->port.count != 1))  {
-		printk(KERN_INFO "rc%d: rc_close: bad port count;"
-		       " tty->count is 1, port count is %d\n",
-		       board_No(bp), port->port.count);
-		port->port.count = 1;
-	}
-	if (--port->port.count < 0)  {
-		printk(KERN_INFO "rc%d: rc_close: bad port count "
-				 "for tty%d: %d\n",
-		       board_No(bp), port_No(port), port->port.count);
-		port->port.count = 0;
-	}
-	if (port->port.count)
-		goto out;
-	port->port.flags |= ASYNC_CLOSING;
-	/*
-	 * Now we wait for the transmit buffer to clear; and we notify
-	 * the line discipline to only process XON/XOFF characters.
-	 */
-	tty->closing = 1;
-	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, port->port.closing_wait);
+	
+	if (tty_port_close_start(&port->port, tty, filp) == 0)
+		return;
+	
 	/*
 	 * At this point we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts, and tell the
 	 * interrupt driver to stop checking the data ready bit in the
 	 * line status register.
 	 */
+
+	spin_lock_irqsave(&riscom_lock, flags);
 	port->IER &= ~IER_RXD;
 	if (port->port.flags & ASYNC_INITIALIZED) {
 		port->IER &= ~IER_TXRDY;
@@ -1053,33 +955,24 @@
 		 */
 		timeout = jiffies + HZ;
 		while (port->IER & IER_TXEMPTY) {
+			spin_unlock_irqrestore(&riscom_lock, flags);
 			msleep_interruptible(jiffies_to_msecs(port->timeout));
+			spin_lock_irqsave(&riscom_lock, flags);
 			if (time_after(jiffies, timeout))
 				break;
 		}
 	}
 	rc_shutdown_port(tty, bp, port);
 	rc_flush_buffer(tty);
-	tty_ldisc_flush(tty);
-
-	tty->closing = 0;
-	port->port.tty = NULL;
-	if (port->port.blocked_open) {
-		if (port->port.close_delay)
-			msleep_interruptible(jiffies_to_msecs(port->port.close_delay));
-		wake_up_interruptible(&port->port.open_wait);
-	}
-	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&port->port.close_wait);
-
-out:
 	spin_unlock_irqrestore(&riscom_lock, flags);
+
+	tty_port_close_end(&port->port, tty);
 }
 
 static int rc_write(struct tty_struct *tty,
 		    const unsigned char *buf, int count)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
 	int c, total = 0;
 	unsigned long flags;
@@ -1122,7 +1015,7 @@
 
 static int rc_put_char(struct tty_struct *tty, unsigned char ch)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	unsigned long flags;
 	int ret = 0;
 
@@ -1146,7 +1039,7 @@
 
 static void rc_flush_chars(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	unsigned long flags;
 
 	if (rc_paranoia_check(port, tty->name, "rc_flush_chars"))
@@ -1166,7 +1059,7 @@
 
 static int rc_write_room(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	int	ret;
 
 	if (rc_paranoia_check(port, tty->name, "rc_write_room"))
@@ -1180,7 +1073,7 @@
 
 static int rc_chars_in_buffer(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 
 	if (rc_paranoia_check(port, tty->name, "rc_chars_in_buffer"))
 		return 0;
@@ -1190,7 +1083,7 @@
 
 static int rc_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
 	unsigned char status;
 	unsigned int result;
@@ -1220,7 +1113,7 @@
 static int rc_tiocmset(struct tty_struct *tty, struct file *file,
 		       unsigned int set, unsigned int clear)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	unsigned long flags;
 	struct riscom_board *bp;
 
@@ -1252,7 +1145,7 @@
 
 static int rc_send_break(struct tty_struct *tty, int length)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp = port_Board(port);
 	unsigned long flags;
 
@@ -1345,7 +1238,7 @@
 static int rc_ioctl(struct tty_struct *tty, struct file *filp,
 		    unsigned int cmd, unsigned long arg)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	void __user *argp = (void __user *)arg;
 	int retval;
 
@@ -1371,7 +1264,7 @@
 
 static void rc_throttle(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
 	unsigned long flags;
 
@@ -1393,7 +1286,7 @@
 
 static void rc_unthrottle(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
 	unsigned long flags;
 
@@ -1415,7 +1308,7 @@
 
 static void rc_stop(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
 	unsigned long flags;
 
@@ -1433,7 +1326,7 @@
 
 static void rc_start(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
 	unsigned long flags;
 
@@ -1454,8 +1347,9 @@
 
 static void rc_hangup(struct tty_struct *tty)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	struct riscom_board *bp;
+	unsigned long flags;
 
 	if (rc_paranoia_check(port, tty->name, "rc_hangup"))
 		return;
@@ -1463,16 +1357,18 @@
 	bp = port_Board(port);
 
 	rc_shutdown_port(tty, bp, port);
+	spin_lock_irqsave(&port->port.lock, flags);
 	port->port.count = 0;
 	port->port.flags &= ~ASYNC_NORMAL_ACTIVE;
 	port->port.tty = NULL;
 	wake_up_interruptible(&port->port.open_wait);
+	spin_unlock_irqrestore(&port->port.lock, flags);
 }
 
 static void rc_set_termios(struct tty_struct *tty,
 					struct ktermios *old_termios)
 {
-	struct riscom_port *port = (struct riscom_port *)tty->driver_data;
+	struct riscom_port *port = tty->driver_data;
 	unsigned long flags;
 
 	if (rc_paranoia_check(port, tty->name, "rc_set_termios"))
@@ -1510,6 +1406,11 @@
 	.break_ctl = rc_send_break,
 };
 
+static const struct tty_port_operations riscom_port_ops = {
+	.carrier_raised = carrier_raised,
+};
+
+
 static int __init rc_init_drivers(void)
 {
 	int error;
@@ -1541,6 +1442,7 @@
 	memset(rc_port, 0, sizeof(rc_port));
 	for (i = 0; i < RC_NPORT * RC_NBOARD; i++)  {
 		tty_port_init(&rc_port[i].port);
+		rc_port[i].port.ops = &riscom_port_ops;
 		rc_port[i].magic = RISCOM8_MAGIC;
 	}
 	return 0;
diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c
index 584d791..f59fc5c 100644
--- a/drivers/char/rocket.c
+++ b/drivers/char/rocket.c
@@ -135,6 +135,7 @@
 static int is_PCI[NUM_BOARDS];
 static rocketModel_t rocketModel[NUM_BOARDS];
 static int max_board;
+static const struct tty_port_operations rocket_port_ops;
 
 /*
  * The following arrays define the interrupt bits corresponding to each AIOP.
@@ -435,15 +436,15 @@
 #endif
 	if (!info)
 		return;
-	if (!info->port.tty) {
-		printk(KERN_WARNING "rp: WARNING %s called with "
-				"info->port.tty==NULL\n", __func__);
+	tty = tty_port_tty_get(&info->port);
+
+	if (tty == NULL) {
+		printk(KERN_WARNING "rp: WARNING %s called with tty==NULL\n", __func__);
 		clear_bit((info->aiop * 8) + info->chan, (void *) &xmit_flags[info->board]);
 		return;
 	}
 
 	spin_lock_irqsave(&info->slock, flags);
-	tty = info->port.tty;
 	info->xmit_fifo_room = TXFIFO_SIZE - sGetTxCnt(cp);
 
 	/*  Loop sending data to FIFO until done or FIFO full */
@@ -477,6 +478,7 @@
 	}
 
 	spin_unlock_irqrestore(&info->slock, flags);
+	tty_kref_put(tty);
 
 #ifdef ROCKET_DEBUG_INTR
 	printk(KERN_DEBUG "(%d,%d,%d,%d)...\n", info->xmit_cnt, info->xmit_head,
@@ -498,18 +500,18 @@
 	if (!info)
 		return;
 
-	if ((info->flags & ROCKET_INITIALIZED) == 0) {
+	if ((info->port.flags & ASYNC_INITIALIZED) == 0) {
 		printk(KERN_WARNING "rp: WARNING: rp_handle_port called with "
 				"info->flags & NOT_INIT\n");
 		return;
 	}
-	if (!info->port.tty) {
+	tty = tty_port_tty_get(&info->port);
+	if (!tty) {
 		printk(KERN_WARNING "rp: WARNING: rp_handle_port called with "
-				"info->port.tty==NULL\n");
+				"tty==NULL\n");
 		return;
 	}
 	cp = &info->channel;
-	tty = info->port.tty;
 
 	IntMask = sGetChanIntID(cp) & info->intmask;
 #ifdef ROCKET_DEBUG_INTR
@@ -541,6 +543,7 @@
 		printk(KERN_INFO "DSR change...\n");
 	}
 #endif
+	tty_kref_put(tty);
 }
 
 /*
@@ -649,9 +652,8 @@
 	info->board = board;
 	info->aiop = aiop;
 	info->chan = chan;
-	info->port.closing_wait = 3000;
-	info->port.close_delay = 50;
-	init_waitqueue_head(&info->port.open_wait);
+	tty_port_init(&info->port);
+	info->port.ops = &rocket_port_ops;
 	init_completion(&info->close_wait);
 	info->flags &= ~ROCKET_MODE_MASK;
 	switch (pc104[board][line]) {
@@ -710,7 +712,7 @@
  *  Configures a rocketport port according to its termio settings.  Called from 
  *  user mode into the driver (exception handler).  *info CD manipulation is spinlock protected.
  */
-static void configure_r_port(struct r_port *info,
+static void configure_r_port(struct tty_struct *tty, struct r_port *info,
 			     struct ktermios *old_termios)
 {
 	unsigned cflag;
@@ -718,7 +720,7 @@
 	unsigned rocketMode;
 	int bits, baud, divisor;
 	CHANNEL_t *cp;
-	struct ktermios *t = info->port.tty->termios;
+	struct ktermios *t = tty->termios;
 
 	cp = &info->channel;
 	cflag = t->c_cflag;
@@ -751,7 +753,7 @@
 	}
 
 	/* baud rate */
-	baud = tty_get_baud_rate(info->port.tty);
+	baud = tty_get_baud_rate(tty);
 	if (!baud)
 		baud = 9600;
 	divisor = ((rp_baud_base[info->board] + (baud >> 1)) / baud) - 1;
@@ -769,7 +771,7 @@
 	sSetBaud(cp, divisor);
 
 	/* FIXME: Should really back compute a baud rate from the divisor */
-	tty_encode_baud_rate(info->port.tty, baud, baud);
+	tty_encode_baud_rate(tty, baud, baud);
 
 	if (cflag & CRTSCTS) {
 		info->intmask |= DELTA_CTS;
@@ -794,15 +796,15 @@
 	 * Handle software flow control in the board
 	 */
 #ifdef ROCKET_SOFT_FLOW
-	if (I_IXON(info->port.tty)) {
+	if (I_IXON(tty)) {
 		sEnTxSoftFlowCtl(cp);
-		if (I_IXANY(info->port.tty)) {
+		if (I_IXANY(tty)) {
 			sEnIXANY(cp);
 		} else {
 			sDisIXANY(cp);
 		}
-		sSetTxXONChar(cp, START_CHAR(info->port.tty));
-		sSetTxXOFFChar(cp, STOP_CHAR(info->port.tty));
+		sSetTxXONChar(cp, START_CHAR(tty));
+		sSetTxXOFFChar(cp, STOP_CHAR(tty));
 	} else {
 		sDisTxSoftFlowCtl(cp);
 		sDisIXANY(cp);
@@ -814,24 +816,24 @@
 	 * Set up ignore/read mask words
 	 */
 	info->read_status_mask = STMRCVROVRH | 0xFF;
-	if (I_INPCK(info->port.tty))
+	if (I_INPCK(tty))
 		info->read_status_mask |= STMFRAMEH | STMPARITYH;
-	if (I_BRKINT(info->port.tty) || I_PARMRK(info->port.tty))
+	if (I_BRKINT(tty) || I_PARMRK(tty))
 		info->read_status_mask |= STMBREAKH;
 
 	/*
 	 * Characters to ignore
 	 */
 	info->ignore_status_mask = 0;
-	if (I_IGNPAR(info->port.tty))
+	if (I_IGNPAR(tty))
 		info->ignore_status_mask |= STMFRAMEH | STMPARITYH;
-	if (I_IGNBRK(info->port.tty)) {
+	if (I_IGNBRK(tty)) {
 		info->ignore_status_mask |= STMBREAKH;
 		/*
 		 * If we're ignoring parity and break indicators,
 		 * ignore overruns too.  (For real raw support).
 		 */
-		if (I_IGNPAR(info->port.tty))
+		if (I_IGNPAR(tty))
 			info->ignore_status_mask |= STMRCVROVRH;
 	}
 
@@ -864,106 +866,17 @@
 	}
 }
 
-/*  info->port.count is considered critical, protected by spinlocks.  */
-static int block_til_ready(struct tty_struct *tty, struct file *filp,
-			   struct r_port *info)
+static int carrier_raised(struct tty_port *port)
 {
-	DECLARE_WAITQUEUE(wait, current);
-	int retval;
-	int do_clocal = 0, extra_count = 0;
-	unsigned long flags;
+	struct r_port *info = container_of(port, struct r_port, port);
+	return (sGetChanStatusLo(&info->channel) & CD_ACT) ? 1 : 0;
+}
 
-	/*
-	 * If the device is in the middle of being closed, then block
-	 * until it's done, and then try again.
-	 */
-	if (tty_hung_up_p(filp))
-		return ((info->flags & ROCKET_HUP_NOTIFY) ? -EAGAIN : -ERESTARTSYS);
-	if (info->flags & ROCKET_CLOSING) {
-		if (wait_for_completion_interruptible(&info->close_wait))
-			return -ERESTARTSYS;
-		return ((info->flags & ROCKET_HUP_NOTIFY) ? -EAGAIN : -ERESTARTSYS);
-	}
-
-	/*
-	 * If non-blocking mode is set, or the port is not enabled,
-	 * then make the check up front and then exit.
-	 */
-	if ((filp->f_flags & O_NONBLOCK) || (tty->flags & (1 << TTY_IO_ERROR))) {
-		info->flags |= ROCKET_NORMAL_ACTIVE;
-		return 0;
-	}
-	if (tty->termios->c_cflag & CLOCAL)
-		do_clocal = 1;
-
-	/*
-	 * Block waiting for the carrier detect and the line to become free.  While we are in
-	 * this loop, info->port.count is dropped by one, so that rp_close() knows when to free things.
-         * We restore it upon exit, either normal or abnormal.
-	 */
-	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
-#ifdef ROCKET_DEBUG_OPEN
-	printk(KERN_INFO "block_til_ready before block: ttyR%d, count = %d\n", info->line, info->port.count);
-#endif
-	spin_lock_irqsave(&info->slock, flags);
-
-#ifdef ROCKET_DISABLE_SIMUSAGE
-	info->flags |= ROCKET_NORMAL_ACTIVE;
-#else
-	if (!tty_hung_up_p(filp)) {
-		extra_count = 1;
-		info->port.count--;
-	}
-#endif
-	info->port.blocked_open++;
-
-	spin_unlock_irqrestore(&info->slock, flags);
-
-	while (1) {
-		if (tty->termios->c_cflag & CBAUD) {
-			sSetDTR(&info->channel);
-			sSetRTS(&info->channel);
-		}
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) || !(info->flags & ROCKET_INITIALIZED)) {
-			if (info->flags & ROCKET_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		if (!(info->flags & ROCKET_CLOSING) && (do_clocal || (sGetChanStatusLo(&info->channel) & CD_ACT)))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-#ifdef ROCKET_DEBUG_OPEN
-		printk(KERN_INFO "block_til_ready blocking: ttyR%d, count = %d, flags=0x%0x\n",
-		     info->line, info->port.count, info->flags);
-#endif
-		schedule();	/*  Don't hold spinlock here, will hang PC */
-	}
-	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
-
-	spin_lock_irqsave(&info->slock, flags);
-
-	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
-
-	spin_unlock_irqrestore(&info->slock, flags);
-
-#ifdef ROCKET_DEBUG_OPEN
-	printk(KERN_INFO "block_til_ready after blocking: ttyR%d, count = %d\n",
-	       info->line, info->port.count);
-#endif
-	if (retval)
-		return retval;
-	info->flags |= ROCKET_NORMAL_ACTIVE;
-	return 0;
+static void raise_dtr_rts(struct tty_port *port)
+{
+	struct r_port *info = container_of(port, struct r_port, port);
+	sSetDTR(&info->channel);
+	sSetRTS(&info->channel);
 }
 
 /*
@@ -973,24 +886,26 @@
 static int rp_open(struct tty_struct *tty, struct file *filp)
 {
 	struct r_port *info;
+	struct tty_port *port;
 	int line = 0, retval;
 	CHANNEL_t *cp;
 	unsigned long page;
 
 	line = tty->index;
-	if ((line < 0) || (line >= MAX_RP_PORTS) || ((info = rp_table[line]) == NULL))
+	if (line < 0 || line >= MAX_RP_PORTS || ((info = rp_table[line]) == NULL))
 		return -ENXIO;
-
+	port = &info->port;
+	
 	page = __get_free_page(GFP_KERNEL);
 	if (!page)
 		return -ENOMEM;
 
-	if (info->flags & ROCKET_CLOSING) {
+	if (port->flags & ASYNC_CLOSING) {
 		retval = wait_for_completion_interruptible(&info->close_wait);
 		free_page(page);
 		if (retval)
 			return retval;
-		return ((info->flags & ROCKET_HUP_NOTIFY) ? -EAGAIN : -ERESTARTSYS);
+		return ((port->flags & ASYNC_HUP_NOTIFY) ? -EAGAIN : -ERESTARTSYS);
 	}
 
 	/*
@@ -1002,9 +917,9 @@
 		info->xmit_buf = (unsigned char *) page;
 
 	tty->driver_data = info;
-	info->port.tty = tty;
+	tty_port_tty_set(port, tty);
 
-	if (info->port.count++ == 0) {
+	if (port->count++ == 0) {
 		atomic_inc(&rp_num_ports_open);
 
 #ifdef ROCKET_DEBUG_OPEN
@@ -1019,7 +934,7 @@
 	/*
 	 * Info->count is now 1; so it's safe to sleep now.
 	 */
-	if ((info->flags & ROCKET_INITIALIZED) == 0) {
+	if (!test_bit(ASYNC_INITIALIZED, &port->flags)) {
 		cp = &info->channel;
 		sSetRxTrigger(cp, TRIG_1);
 		if (sGetChanStatus(cp) & CD_ACT)
@@ -1043,21 +958,21 @@
 		sEnRxFIFO(cp);
 		sEnTransmit(cp);
 
-		info->flags |= ROCKET_INITIALIZED;
+		set_bit(ASYNC_INITIALIZED, &info->port.flags);
 
 		/*
 		 * Set up the tty->alt_speed kludge
 		 */
 		if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_HI)
-			info->port.tty->alt_speed = 57600;
+			tty->alt_speed = 57600;
 		if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_VHI)
-			info->port.tty->alt_speed = 115200;
+			tty->alt_speed = 115200;
 		if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_SHI)
-			info->port.tty->alt_speed = 230400;
+			tty->alt_speed = 230400;
 		if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_WARP)
-			info->port.tty->alt_speed = 460800;
+			tty->alt_speed = 460800;
 
-		configure_r_port(info, NULL);
+		configure_r_port(tty, info, NULL);
 		if (tty->termios->c_cflag & CBAUD) {
 			sSetDTR(cp);
 			sSetRTS(cp);
@@ -1066,7 +981,7 @@
 	/*  Starts (or resets) the maint polling loop */
 	mod_timer(&rocket_timer, jiffies + POLL_PERIOD);
 
-	retval = block_til_ready(tty, filp, info);
+	retval = tty_port_block_til_ready(port, tty, filp);
 	if (retval) {
 #ifdef ROCKET_DEBUG_OPEN
 		printk(KERN_INFO "rp_open returning after block_til_ready with %d\n", retval);
@@ -1081,8 +996,8 @@
  */
 static void rp_close(struct tty_struct *tty, struct file *filp)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
-	unsigned long flags;
+	struct r_port *info = tty->driver_data;
+	struct tty_port *port = &info->port;
 	int timeout;
 	CHANNEL_t *cp;
 	
@@ -1093,53 +1008,10 @@
 	printk(KERN_INFO "rp_close ttyR%d, count = %d\n", info->line, info->port.count);
 #endif
 
-	if (tty_hung_up_p(filp))
+	if (tty_port_close_start(port, tty, filp) == 0)
 		return;
-	spin_lock_irqsave(&info->slock, flags);
-
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * Uh, oh.  tty->count is 1, which means that the tty
-		 * structure will be freed.  Info->count should always
-		 * be one in these conditions.  If it's greater than
-		 * one, we've got real problems, since it means the
-		 * serial port won't be shutdown.
-		 */
-		printk(KERN_WARNING "rp_close: bad serial port count; "
-			"tty->count is 1, info->port.count is %d\n", info->port.count);
-		info->port.count = 1;
-	}
-	if (--info->port.count < 0) {
-		printk(KERN_WARNING "rp_close: bad serial port count for "
-				"ttyR%d: %d\n", info->line, info->port.count);
-		info->port.count = 0;
-	}
-	if (info->port.count) {
-		spin_unlock_irqrestore(&info->slock, flags);
-		return;
-	}
-	info->flags |= ROCKET_CLOSING;
-	spin_unlock_irqrestore(&info->slock, flags);
 
 	cp = &info->channel;
-
-	/*
-	 * Notify the line discpline to only process XON/XOFF characters
-	 */
-	tty->closing = 1;
-
-	/*
-	 * If transmission was throttled by the application request,
-	 * just flush the xmit buffer.
-	 */
-	if (tty->flow_stopped)
-		rp_flush_buffer(tty);
-
-	/*
-	 * Wait for the transmit buffer to clear
-	 */
-	if (info->port.closing_wait != ROCKET_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, info->port.closing_wait);
 	/*
 	 * Before we drop DTR, make sure the UART transmitter
 	 * has completely drained; this is especially
@@ -1168,19 +1040,24 @@
 
 	clear_bit((info->aiop * 8) + info->chan, (void *) &xmit_flags[info->board]);
 
-	if (info->port.blocked_open) {
-		if (info->port.close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
+	/* We can't yet use tty_port_close_end as the buffer handling in this
+	   driver is a bit different to the usual */
+
+	if (port->blocked_open) {
+		if (port->close_delay) {
+			msleep_interruptible(jiffies_to_msecs(port->close_delay));
 		}
-		wake_up_interruptible(&info->port.open_wait);
+		wake_up_interruptible(&port->open_wait);
 	} else {
 		if (info->xmit_buf) {
 			free_page((unsigned long) info->xmit_buf);
 			info->xmit_buf = NULL;
 		}
 	}
-	info->flags &= ~(ROCKET_INITIALIZED | ROCKET_CLOSING | ROCKET_NORMAL_ACTIVE);
+	info->port.flags &= ~(ASYNC_INITIALIZED | ASYNC_CLOSING | ASYNC_NORMAL_ACTIVE);
 	tty->closing = 0;
+	tty_port_tty_set(port, NULL);
+	wake_up_interruptible(&port->close_wait);
 	complete_all(&info->close_wait);
 	atomic_dec(&rp_num_ports_open);
 
@@ -1195,7 +1072,7 @@
 static void rp_set_termios(struct tty_struct *tty,
 			   struct ktermios *old_termios)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 	unsigned cflag;
 
@@ -1213,7 +1090,7 @@
 	/* Or CMSPAR */
 	tty->termios->c_cflag &= ~CMSPAR;
 
-	configure_r_port(info, old_termios);
+	configure_r_port(tty, info, old_termios);
 
 	cp = &info->channel;
 
@@ -1238,7 +1115,7 @@
 
 static int rp_break(struct tty_struct *tty, int break_state)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	unsigned long flags;
 
 	if (rocket_paranoia_check(info, "rp_break"))
@@ -1284,7 +1161,7 @@
  */
 static int rp_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct r_port *info = (struct r_port *)tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	unsigned int control, result, ChanStatus;
 
 	ChanStatus = sGetChanStatusLo(&info->channel);
@@ -1305,7 +1182,7 @@
 static int rp_tiocmset(struct tty_struct *tty, struct file *file,
 		    unsigned int set, unsigned int clear)
 {
-	struct r_port *info = (struct r_port *)tty->driver_data;
+	struct r_port *info = tty->driver_data;
 
 	if (set & TIOCM_RTS)
 		info->channel.TxControl[3] |= SET_RTS;
@@ -1338,7 +1215,8 @@
 	return 0;
 }
 
-static int set_config(struct r_port *info, struct rocket_config __user *new_info)
+static int set_config(struct tty_struct *tty, struct r_port *info,
+					struct rocket_config __user *new_info)
 {
 	struct rocket_config new_serial;
 
@@ -1350,7 +1228,7 @@
 		if ((new_serial.flags & ~ROCKET_USR_MASK) != (info->flags & ~ROCKET_USR_MASK))
 			return -EPERM;
 		info->flags = ((info->flags & ~ROCKET_USR_MASK) | (new_serial.flags & ROCKET_USR_MASK));
-		configure_r_port(info, NULL);
+		configure_r_port(tty, info, NULL);
 		return 0;
 	}
 
@@ -1359,15 +1237,15 @@
 	info->port.closing_wait = new_serial.closing_wait;
 
 	if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_HI)
-		info->port.tty->alt_speed = 57600;
+		tty->alt_speed = 57600;
 	if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_VHI)
-		info->port.tty->alt_speed = 115200;
+		tty->alt_speed = 115200;
 	if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_SHI)
-		info->port.tty->alt_speed = 230400;
+		tty->alt_speed = 230400;
 	if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_WARP)
-		info->port.tty->alt_speed = 460800;
+		tty->alt_speed = 460800;
 
-	configure_r_port(info, NULL);
+	configure_r_port(tty, info, NULL);
 	return 0;
 }
 
@@ -1434,7 +1312,7 @@
 static int rp_ioctl(struct tty_struct *tty, struct file *file,
 		    unsigned int cmd, unsigned long arg)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	void __user *argp = (void __user *)arg;
 	int ret = 0;
 
@@ -1452,7 +1330,7 @@
 		ret = get_config(info, argp);
 		break;
 	case RCKP_SET_CONFIG:
-		ret = set_config(info, argp);
+		ret = set_config(tty, info, argp);
 		break;
 	case RCKP_GET_PORTS:
 		ret = get_ports(info, argp);
@@ -1472,7 +1350,7 @@
 
 static void rp_send_xchar(struct tty_struct *tty, char ch)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 
 	if (rocket_paranoia_check(info, "rp_send_xchar"))
@@ -1487,7 +1365,7 @@
 
 static void rp_throttle(struct tty_struct *tty)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 
 #ifdef ROCKET_DEBUG_THROTTLE
@@ -1507,7 +1385,7 @@
 
 static void rp_unthrottle(struct tty_struct *tty)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 #ifdef ROCKET_DEBUG_THROTTLE
 	printk(KERN_INFO "unthrottle %s: %d....\n", tty->name,
@@ -1534,7 +1412,7 @@
  */
 static void rp_stop(struct tty_struct *tty)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 
 #ifdef ROCKET_DEBUG_FLOW
 	printk(KERN_INFO "stop %s: %d %d....\n", tty->name,
@@ -1550,7 +1428,7 @@
 
 static void rp_start(struct tty_struct *tty)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 
 #ifdef ROCKET_DEBUG_FLOW
 	printk(KERN_INFO "start %s: %d %d....\n", tty->name,
@@ -1570,7 +1448,7 @@
  */
 static void rp_wait_until_sent(struct tty_struct *tty, int timeout)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 	unsigned long orig_jiffies;
 	int check_time, exit_time;
@@ -1627,7 +1505,7 @@
 static void rp_hangup(struct tty_struct *tty)
 {
 	CHANNEL_t *cp;
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 
 	if (rocket_paranoia_check(info, "rp_hangup"))
 		return;
@@ -1636,15 +1514,13 @@
 	printk(KERN_INFO "rp_hangup of ttyR%d...\n", info->line);
 #endif
 	rp_flush_buffer(tty);
-	if (info->flags & ROCKET_CLOSING)
+	if (info->port.flags & ASYNC_CLOSING)
 		return;
 	if (info->port.count)
 		atomic_dec(&rp_num_ports_open);
 	clear_bit((info->aiop * 8) + info->chan, (void *) &xmit_flags[info->board]);
 
-	info->port.count = 0;
-	info->flags &= ~ROCKET_NORMAL_ACTIVE;
-	info->port.tty = NULL;
+	tty_port_hangup(&info->port);
 
 	cp = &info->channel;
 	sDisRxFIFO(cp);
@@ -1653,7 +1529,7 @@
 	sDisCTSFlowCtl(cp);
 	sDisTxSoftFlowCtl(cp);
 	sClrTxXOFF(cp);
-	info->flags &= ~ROCKET_INITIALIZED;
+	info->port.flags &= ~ASYNC_INITIALIZED;
 
 	wake_up_interruptible(&info->port.open_wait);
 }
@@ -1667,7 +1543,7 @@
  */
 static int rp_put_char(struct tty_struct *tty, unsigned char ch)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 	unsigned long flags;
 
@@ -1714,7 +1590,7 @@
 static int rp_write(struct tty_struct *tty,
 		    const unsigned char *buf, int count)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 	const unsigned char *b;
 	int c, retval = 0;
@@ -1764,7 +1640,8 @@
 
 	/*  Write remaining data into the port's xmit_buf */
 	while (1) {
-		if (!info->port.tty)		/* Seemingly obligatory check... */
+		/* Hung up ? */
+		if (!test_bit(ASYNC_NORMAL_ACTIVE, &info->port.flags))
 			goto end;
 		c = min(count, XMIT_BUF_SIZE - info->xmit_cnt - 1);
 		c = min(c, XMIT_BUF_SIZE - info->xmit_head);
@@ -1806,7 +1683,7 @@
  */
 static int rp_write_room(struct tty_struct *tty)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	int ret;
 
 	if (rocket_paranoia_check(info, "rp_write_room"))
@@ -1827,7 +1704,7 @@
  */
 static int rp_chars_in_buffer(struct tty_struct *tty)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 
 	if (rocket_paranoia_check(info, "rp_chars_in_buffer"))
@@ -1848,7 +1725,7 @@
  */
 static void rp_flush_buffer(struct tty_struct *tty)
 {
-	struct r_port *info = (struct r_port *) tty->driver_data;
+	struct r_port *info = tty->driver_data;
 	CHANNEL_t *cp;
 	unsigned long flags;
 
@@ -2371,6 +2248,11 @@
 	.tiocmset = rp_tiocmset,
 };
 
+static const struct tty_port_operations rocket_port_ops = {
+	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
+};
+
 /*
  * The module "startup" routine; it's run when the module is loaded.
  */
diff --git a/drivers/char/rocket.h b/drivers/char/rocket.h
index a8b0919..ec863f3 100644
--- a/drivers/char/rocket.h
+++ b/drivers/char/rocket.h
@@ -39,7 +39,7 @@
 /*
  * Rocketport flags
  */
-#define ROCKET_CALLOUT_NOHUP    0x00000001
+/*#define ROCKET_CALLOUT_NOHUP    0x00000001 */
 #define ROCKET_FORCE_CD		0x00000002
 #define ROCKET_HUP_NOTIFY	0x00000004
 #define ROCKET_SPLIT_TERMIOS	0x00000008
diff --git a/drivers/char/rocket_int.h b/drivers/char/rocket_int.h
index 21f3ff5..67e0f1e 100644
--- a/drivers/char/rocket_int.h
+++ b/drivers/char/rocket_int.h
@@ -1162,11 +1162,6 @@
 /* number of characters left in xmit buffer before we ask for more */
 #define WAKEUP_CHARS 256
 
-/* Internal flags used only by the rocketport driver */
-#define ROCKET_INITIALIZED	0x80000000	/* Port is active */
-#define ROCKET_CLOSING		0x40000000	/* Serial port is closing */
-#define ROCKET_NORMAL_ACTIVE	0x20000000	/* Normal port is active */
-
 /*
  * Assigned major numbers for the Comtrol Rocketport
  */
diff --git a/drivers/char/selection.c b/drivers/char/selection.c
index 2978a49..f29fbe9 100644
--- a/drivers/char/selection.c
+++ b/drivers/char/selection.c
@@ -306,7 +306,7 @@
  */
 int paste_selection(struct tty_struct *tty)
 {
-	struct vc_data *vc = (struct vc_data *)tty->driver_data;
+	struct vc_data *vc = tty->driver_data;
 	int	pasted = 0;
 	unsigned int count;
 	struct  tty_ldisc *ld;
diff --git a/drivers/char/ser_a2232.c b/drivers/char/ser_a2232.c
index 7b0c352..33872a2 100644
--- a/drivers/char/ser_a2232.c
+++ b/drivers/char/ser_a2232.c
@@ -122,7 +122,7 @@
 static void a2232_enable_tx_interrupts(void *ptr);
 static void a2232_disable_rx_interrupts(void *ptr);
 static void a2232_enable_rx_interrupts(void *ptr);
-static int  a2232_get_CD(void *ptr);
+static int  a2232_carrier_raised(struct tty_port *port);
 static void a2232_shutdown_port(void *ptr);
 static int  a2232_set_real_termios(void *ptr);
 static int  a2232_chars_in_buffer(void *ptr);
@@ -148,7 +148,6 @@
         a2232_enable_tx_interrupts,
         a2232_disable_rx_interrupts,
         a2232_enable_rx_interrupts,
-        a2232_get_CD,
         a2232_shutdown_port,
         a2232_set_real_termios,
         a2232_chars_in_buffer,
@@ -260,9 +259,10 @@
 	port->disable_rx = 0;
 }
 
-static int  a2232_get_CD(void *ptr)
+static int  a2232_carrier_raised(struct tty_port *port)
 {
-	return ((struct a2232_port *) ptr)->cd_status;
+	struct a2232_port *ap = container_of(port, struct a2232_port, gs.port);
+	return ap->cd_status;
 }
 
 static void a2232_shutdown_port(void *ptr)
@@ -460,14 +460,14 @@
    if switched on. So the only thing we can do at this
    layer here is not taking any characters out of the
    A2232 buffer any more. */
-	struct a2232_port *port = (struct a2232_port *) tty->driver_data;
+	struct a2232_port *port = tty->driver_data;
 	port->throttle_input = -1;
 }
 
 static void a2232_unthrottle(struct tty_struct *tty)
 {
 /* Unthrottle: dual to "throttle()" above. */
-	struct a2232_port *port = (struct a2232_port *) tty->driver_data;
+	struct a2232_port *port = tty->driver_data;
 	port->throttle_input = 0;
 }
 
@@ -638,6 +638,10 @@
 	return IRQ_HANDLED;
 }
 
+static const struct tty_port_operations a2232_port_ops = {
+	.carrier_raised = a2232_carrier_raised,
+};
+
 static void a2232_init_portstructs(void)
 {
 	struct a2232_port *port;
@@ -645,6 +649,8 @@
 
 	for (i = 0; i < MAX_A2232_BOARDS*NUMLINES; i++) {
 		port = a2232_ports + i;
+		tty_port_init(&port->gs.port);
+		port->gs.port.ops = &a2232_port_ops;
 		port->which_a2232 = i/NUMLINES;
 		port->which_port_on_a2232 = i%NUMLINES;
 		port->disable_rx = port->throttle_input = port->cd_status = 0;
@@ -652,11 +658,6 @@
 		port->gs.close_delay = HZ/2;
 		port->gs.closing_wait = 30 * HZ;
 		port->gs.rd = &a2232_real_driver;
-#ifdef NEW_WRITE_LOCKING
-		mutex_init(&(port->gs.port_write_mutex));
-#endif
-		init_waitqueue_head(&port->gs.port.open_wait);
-		init_waitqueue_head(&port->gs.port.close_wait);
 	}
 }
 
diff --git a/drivers/char/serial167.c b/drivers/char/serial167.c
index a8f15e6..f1f24f0 100644
--- a/drivers/char/serial167.c
+++ b/drivers/char/serial167.c
@@ -315,7 +315,7 @@
 
 static void cy_stop(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	volatile unsigned char *base_addr = (unsigned char *)BASE_ADDR;
 	int channel;
 	unsigned long flags;
@@ -337,7 +337,7 @@
 
 static void cy_start(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	volatile unsigned char *base_addr = (unsigned char *)BASE_ADDR;
 	int channel;
 	unsigned long flags;
@@ -1062,7 +1062,7 @@
 
 static int cy_put_char(struct tty_struct *tty, unsigned char ch)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	unsigned long flags;
 
 #ifdef SERIAL_DEBUG_IO
@@ -1090,7 +1090,7 @@
 
 static void cy_flush_chars(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	unsigned long flags;
 	volatile unsigned char *base_addr = (u_char *) BASE_ADDR;
 	int channel;
@@ -1122,7 +1122,7 @@
  */
 static int cy_write(struct tty_struct *tty, const unsigned char *buf, int count)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	unsigned long flags;
 	int c, total = 0;
 
@@ -1166,7 +1166,7 @@
 
 static int cy_write_room(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	int ret;
 
 #ifdef SERIAL_DEBUG_IO
@@ -1183,7 +1183,7 @@
 
 static int cy_chars_in_buffer(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 
 #ifdef SERIAL_DEBUG_IO
 	printk("cy_chars_in_buffer %s %d\n", tty->name, info->xmit_cnt);	/* */
@@ -1197,7 +1197,7 @@
 
 static void cy_flush_buffer(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	unsigned long flags;
 
 #ifdef SERIAL_DEBUG_IO
@@ -1218,7 +1218,7 @@
  */
 static void cy_throttle(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	unsigned long flags;
 	volatile unsigned char *base_addr = (u_char *) BASE_ADDR;
 	int channel;
@@ -1250,7 +1250,7 @@
 
 static void cy_unthrottle(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	unsigned long flags;
 	volatile unsigned char *base_addr = (u_char *) BASE_ADDR;
 	int channel;
@@ -1345,7 +1345,7 @@
 
 static int cy_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	int channel;
 	volatile unsigned char *base_addr = (u_char *) BASE_ADDR;
 	unsigned long flags;
@@ -1369,7 +1369,7 @@
 cy_tiocmset(struct tty_struct *tty, struct file *file,
 	    unsigned int set, unsigned int clear)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	int channel;
 	volatile unsigned char *base_addr = (u_char *) BASE_ADDR;
 	unsigned long flags;
@@ -1532,7 +1532,7 @@
 	 unsigned int cmd, unsigned long arg)
 {
 	unsigned long val;
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 	int ret_val = 0;
 	void __user *argp = (void __user *)arg;
 
@@ -1607,7 +1607,7 @@
 
 static void cy_set_termios(struct tty_struct *tty, struct ktermios *old_termios)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 
 #ifdef SERIAL_DEBUG_OTHER
 	printk("cy_set_termios %s\n", tty->name);
@@ -1631,7 +1631,7 @@
 
 static void cy_close(struct tty_struct *tty, struct file *filp)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 
 /* CP('C'); */
 #ifdef SERIAL_DEBUG_OTHER
@@ -1698,7 +1698,7 @@
  */
 void cy_hangup(struct tty_struct *tty)
 {
-	struct cyclades_port *info = (struct cyclades_port *)tty->driver_data;
+	struct cyclades_port *info = tty->driver_data;
 
 #ifdef SERIAL_DEBUG_OTHER
 	printk("cy_hangup %s\n", tty->name);	/* */
diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c
index a16b94f..3c67c3d 100644
--- a/drivers/char/specialix.c
+++ b/drivers/char/specialix.c
@@ -1450,7 +1450,7 @@
 
 static void sx_flush_buffer(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	unsigned long flags;
 	struct specialix_board  *bp;
 
@@ -1472,7 +1472,7 @@
 
 static void sx_close(struct tty_struct *tty, struct file *filp)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	unsigned long flags;
 	unsigned long timeout;
@@ -1585,7 +1585,7 @@
 static int sx_write(struct tty_struct *tty,
 					const unsigned char *buf, int count)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	int c, total = 0;
 	unsigned long flags;
@@ -1637,7 +1637,7 @@
 
 static int sx_put_char(struct tty_struct *tty, unsigned char ch)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	unsigned long flags;
 	struct specialix_board  *bp;
 
@@ -1676,7 +1676,7 @@
 
 static void sx_flush_chars(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	unsigned long flags;
 	struct specialix_board  *bp = port_Board(port);
 
@@ -1703,7 +1703,7 @@
 
 static int sx_write_room(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	int	ret;
 
 	func_enter();
@@ -1724,7 +1724,7 @@
 
 static int sx_chars_in_buffer(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 
 	func_enter();
 
@@ -1738,7 +1738,7 @@
 
 static int sx_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	unsigned char status;
 	unsigned int result;
@@ -1780,7 +1780,7 @@
 static int sx_tiocmset(struct tty_struct *tty, struct file *file,
 		       unsigned int set, unsigned int clear)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	unsigned long flags;
 	struct specialix_board *bp;
 
@@ -1820,7 +1820,7 @@
 
 static int sx_send_break(struct tty_struct *tty, int length)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp = port_Board(port);
 	unsigned long flags;
 
@@ -1931,7 +1931,7 @@
 static int sx_ioctl(struct tty_struct *tty, struct file *filp,
 				unsigned int cmd, unsigned long arg)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	void __user *argp = (void __user *)arg;
 
 	func_enter();
@@ -1959,7 +1959,7 @@
 
 static void sx_throttle(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	unsigned long flags;
 
@@ -2004,7 +2004,7 @@
 
 static void sx_unthrottle(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	unsigned long flags;
 
@@ -2045,7 +2045,7 @@
 
 static void sx_stop(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	unsigned long flags;
 
@@ -2072,7 +2072,7 @@
 
 static void sx_start(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	unsigned long flags;
 
@@ -2100,7 +2100,7 @@
 
 static void sx_hangup(struct tty_struct *tty)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	struct specialix_board *bp;
 	unsigned long flags;
 
@@ -2135,7 +2135,7 @@
 static void sx_set_termios(struct tty_struct *tty,
 					struct ktermios *old_termios)
 {
-	struct specialix_port *port = (struct specialix_port *)tty->driver_data;
+	struct specialix_port *port = tty->driver_data;
 	unsigned long flags;
 	struct specialix_board  *bp;
 
diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index 963b03f..e1e0dd8 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -130,6 +130,8 @@
 static DEFINE_MUTEX(stl_brdslock);
 static struct stlbrd		*stl_brds[STL_MAXBRDS];
 
+static const struct tty_port_operations stl_port_ops;
+
 /*
  *	Per board state flags. Used with the state field of the board struct.
  *	Not really much here!
@@ -407,7 +409,6 @@
 static int	stl_brdinit(struct stlbrd *brdp);
 static int	stl_getportstats(struct tty_struct *tty, struct stlport *portp, comstats_t __user *cp);
 static int	stl_clrportstats(struct stlport *portp, comstats_t __user *cp);
-static int	stl_waitcarrier(struct tty_struct *tty, struct stlport *portp, struct file *filp);
 
 /*
  *	CD1400 uart specific handling functions.
@@ -703,8 +704,9 @@
 {
 	struct stlport	*portp;
 	struct stlbrd	*brdp;
+	struct tty_port *port;
 	unsigned int	minordev, brdnr, panelnr;
-	int		portnr, rc;
+	int		portnr;
 
 	pr_debug("stl_open(tty=%p,filp=%p): device=%s\n", tty, filp, tty->name);
 
@@ -715,6 +717,7 @@
 	brdp = stl_brds[brdnr];
 	if (brdp == NULL)
 		return -ENODEV;
+
 	minordev = MINOR2PORT(minordev);
 	for (portnr = -1, panelnr = 0; panelnr < STL_MAXPANELS; panelnr++) {
 		if (brdp->panels[panelnr] == NULL)
@@ -731,16 +734,17 @@
 	portp = brdp->panels[panelnr]->ports[portnr];
 	if (portp == NULL)
 		return -ENODEV;
+	port = &portp->port;
 
 /*
  *	On the first open of the device setup the port hardware, and
  *	initialize the per port data structure.
  */
-	tty_port_tty_set(&portp->port, tty);
+	tty_port_tty_set(port, tty);
 	tty->driver_data = portp;
-	portp->port.count++;
+	port->count++;
 
-	if ((portp->port.flags & ASYNC_INITIALIZED) == 0) {
+	if ((port->flags & ASYNC_INITIALIZED) == 0) {
 		if (!portp->tx.buf) {
 			portp->tx.buf = kmalloc(STL_TXBUFSIZE, GFP_KERNEL);
 			if (!portp->tx.buf)
@@ -754,91 +758,24 @@
 		stl_enablerxtx(portp, 1, 1);
 		stl_startrxtx(portp, 1, 0);
 		clear_bit(TTY_IO_ERROR, &tty->flags);
-		portp->port.flags |= ASYNC_INITIALIZED;
+		port->flags |= ASYNC_INITIALIZED;
 	}
-
-/*
- *	Check if this port is in the middle of closing. If so then wait
- *	until it is closed then return error status, based on flag settings.
- *	The sleep here does not need interrupt protection since the wakeup
- *	for it is done with the same context.
- */
-	if (portp->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->port.close_wait);
-		if (portp->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		return -ERESTARTSYS;
-	}
-
-/*
- *	Based on type of open being done check if it can overlap with any
- *	previous opens still in effect. If we are a normal serial device
- *	then also we might have to wait for carrier.
- */
-	if (!(filp->f_flags & O_NONBLOCK))
-		if ((rc = stl_waitcarrier(tty, portp, filp)) != 0)
-			return rc;
-
-	portp->port.flags |= ASYNC_NORMAL_ACTIVE;
-
-	return 0;
+	return tty_port_block_til_ready(port, tty, filp);
 }
 
 /*****************************************************************************/
 
-/*
- *	Possibly need to wait for carrier (DCD signal) to come high. Say
- *	maybe because if we are clocal then we don't need to wait...
- */
-
-static int stl_waitcarrier(struct tty_struct *tty, struct stlport *portp,
-							struct file *filp)
+static int stl_carrier_raised(struct tty_port *port)
 {
-	unsigned long	flags;
-	int		rc, doclocal;
+	struct stlport *portp = container_of(port, struct stlport, port);
+	return (portp->sigs & TIOCM_CD) ? 1 : 0;
+}
 
-	pr_debug("stl_waitcarrier(portp=%p,filp=%p)\n", portp, filp);
-
-	rc = 0;
-	doclocal = 0;
-
-	spin_lock_irqsave(&stallion_lock, flags);
-
-	if (tty->termios->c_cflag & CLOCAL)
-		doclocal++;
-
-	portp->openwaitcnt++;
-	if (! tty_hung_up_p(filp))
-		portp->port.count--;
-
-	for (;;) {
-		/* Takes brd_lock internally */
-		stl_setsignals(portp, 1, 1);
-		if (tty_hung_up_p(filp) ||
-		    ((portp->port.flags & ASYNC_INITIALIZED) == 0)) {
-			if (portp->port.flags & ASYNC_HUP_NOTIFY)
-				rc = -EBUSY;
-			else
-				rc = -ERESTARTSYS;
-			break;
-		}
-		if (((portp->port.flags & ASYNC_CLOSING) == 0) &&
-		    (doclocal || (portp->sigs & TIOCM_CD)))
-			break;
-		if (signal_pending(current)) {
-			rc = -ERESTARTSYS;
-			break;
-		}
-		/* FIXME */
-		interruptible_sleep_on(&portp->port.open_wait);
-	}
-
-	if (! tty_hung_up_p(filp))
-		portp->port.count++;
-	portp->openwaitcnt--;
-	spin_unlock_irqrestore(&stallion_lock, flags);
-
-	return rc;
+static void stl_raise_dtr_rts(struct tty_port *port)
+{
+	struct stlport *portp = container_of(port, struct stlport, port);
+	/* Takes brd_lock internally */
+	stl_setsignals(portp, 1, 1);
 }
 
 /*****************************************************************************/
@@ -890,47 +827,29 @@
 static void stl_close(struct tty_struct *tty, struct file *filp)
 {
 	struct stlport	*portp;
+	struct tty_port *port;
 	unsigned long	flags;
 
 	pr_debug("stl_close(tty=%p,filp=%p)\n", tty, filp);
 
 	portp = tty->driver_data;
-	if (portp == NULL)
-		return;
+	BUG_ON(portp == NULL);
 
-	spin_lock_irqsave(&stallion_lock, flags);
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&stallion_lock, flags);
-		return;
-	}
-	if ((tty->count == 1) && (portp->port.count != 1))
-		portp->port.count = 1;
-	if (portp->port.count-- > 1) {
-		spin_unlock_irqrestore(&stallion_lock, flags);
-		return;
-	}
+	port = &portp->port;
 
-	portp->port.count = 0;
-	portp->port.flags |= ASYNC_CLOSING;
-
+	if (tty_port_close_start(port, tty, filp) == 0)
+		return;
 /*
  *	May want to wait for any data to drain before closing. The BUSY
  *	flag keeps track of whether we are still sending or not - it is
  *	very accurate for the cd1400, not quite so for the sc26198.
  *	(The sc26198 has no "end-of-data" interrupt only empty FIFO)
  */
-	tty->closing = 1;
-
-	spin_unlock_irqrestore(&stallion_lock, flags);
-
-	if (portp->closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, portp->closing_wait);
 	stl_waituntilsent(tty, (HZ / 2));
 
-
-	spin_lock_irqsave(&stallion_lock, flags);
+	spin_lock_irqsave(&port->lock, flags);
 	portp->port.flags &= ~ASYNC_INITIALIZED;
-	spin_unlock_irqrestore(&stallion_lock, flags);
+	spin_unlock_irqrestore(&port->lock, flags);
 
 	stl_disableintrs(portp);
 	if (tty->termios->c_cflag & HUPCL)
@@ -944,20 +863,9 @@
 		portp->tx.head = NULL;
 		portp->tx.tail = NULL;
 	}
-	set_bit(TTY_IO_ERROR, &tty->flags);
-	tty_ldisc_flush(tty);
 
-	tty->closing = 0;
-	tty_port_tty_set(&portp->port, NULL);
-
-	if (portp->openwaitcnt) {
-		if (portp->close_delay)
-			msleep_interruptible(jiffies_to_msecs(portp->close_delay));
-		wake_up_interruptible(&portp->port.open_wait);
-	}
-
-	portp->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&portp->port.close_wait);
+	tty_port_close_end(port, tty);
+	tty_port_tty_set(port, NULL);
 }
 
 /*****************************************************************************/
@@ -1405,14 +1313,20 @@
 static void stl_hangup(struct tty_struct *tty)
 {
 	struct stlport	*portp;
+	struct tty_port *port;
+	unsigned long flags;
 
 	pr_debug("stl_hangup(tty=%p)\n", tty);
 
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
+	port = &portp->port;
 
-	portp->port.flags &= ~ASYNC_INITIALIZED;
+	spin_lock_irqsave(&port->lock, flags);
+	port->flags &= ~ASYNC_INITIALIZED;
+	spin_unlock_irqrestore(&port->lock, flags);
+
 	stl_disableintrs(portp);
 	if (tty->termios->c_cflag & HUPCL)
 		stl_setsignals(portp, 0, 0);
@@ -1426,10 +1340,7 @@
 		portp->tx.head = NULL;
 		portp->tx.tail = NULL;
 	}
-	tty_port_tty_set(&portp->port, NULL);
-	portp->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	portp->port.count = 0;
-	wake_up_interruptible(&portp->port.open_wait);
+	tty_port_hangup(port);
 }
 
 /*****************************************************************************/
@@ -1776,6 +1687,7 @@
 			break;
 		}
 		tty_port_init(&portp->port);
+		portp->port.ops = &stl_port_ops;
 		portp->magic = STL_PORTMAGIC;
 		portp->portnr = i;
 		portp->brdnr = panelp->brdnr;
@@ -2659,6 +2571,11 @@
 	.tiocmset = stl_tiocmset,
 };
 
+static const struct tty_port_operations stl_port_ops = {
+	.carrier_raised = stl_carrier_raised,
+	.raise_dtr_rts = stl_raise_dtr_rts,
+};
+
 /*****************************************************************************/
 /*                       CD1400 HARDWARE FUNCTIONS                           */
 /*****************************************************************************/
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index ba4e862..b60be7b 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -279,7 +279,7 @@
 static void sx_enable_tx_interrupts(void *ptr);
 static void sx_disable_rx_interrupts(void *ptr);
 static void sx_enable_rx_interrupts(void *ptr);
-static int sx_get_CD(void *ptr);
+static int sx_carrier_raised(struct tty_port *port);
 static void sx_shutdown_port(void *ptr);
 static int sx_set_real_termios(void *ptr);
 static void sx_close(void *ptr);
@@ -360,7 +360,6 @@
 	sx_enable_tx_interrupts,
 	sx_disable_rx_interrupts,
 	sx_enable_rx_interrupts,
-	sx_get_CD,
 	sx_shutdown_port,
 	sx_set_real_termios,
 	sx_chars_in_buffer,
@@ -791,7 +790,7 @@
 	sx_dprintk(SX_DEBUG_MODEMSIGNALS, "getsignals: %d/%d  (%d/%d) "
 			"%02x/%02x\n",
 			(o_stat & OP_DTR) != 0, (o_stat & OP_RTS) != 0,
-			port->c_dcd, sx_get_CD(port),
+			port->c_dcd, tty_port_carrier_raised(&port->gs.port),
 			sx_read_channel_byte(port, hi_ip),
 			sx_read_channel_byte(port, hi_state));
 
@@ -1190,7 +1189,7 @@
 
 	hi_state = sx_read_channel_byte(port, hi_state);
 	sx_dprintk(SX_DEBUG_MODEMSIGNALS, "Checking modem signals (%d/%d)\n",
-			port->c_dcd, sx_get_CD(port));
+			port->c_dcd, tty_port_carrier_raised(&port->gs.port));
 
 	if (hi_state & ST_BREAK) {
 		hi_state &= ~ST_BREAK;
@@ -1202,11 +1201,11 @@
 		hi_state &= ~ST_DCD;
 		sx_dprintk(SX_DEBUG_MODEMSIGNALS, "got a DCD change.\n");
 		sx_write_channel_byte(port, hi_state, hi_state);
-		c_dcd = sx_get_CD(port);
+		c_dcd = tty_port_carrier_raised(&port->gs.port);
 		sx_dprintk(SX_DEBUG_MODEMSIGNALS, "DCD is now %d\n", c_dcd);
 		if (c_dcd != port->c_dcd) {
 			port->c_dcd = c_dcd;
-			if (sx_get_CD(port)) {
+			if (tty_port_carrier_raised(&port->gs.port)) {
 				/* DCD went UP */
 				if ((sx_read_channel_byte(port, hi_hstat) !=
 						HS_IDLE_CLOSED) &&
@@ -1415,13 +1414,10 @@
 }
 
 /* Jeez. Isn't this simple? */
-static int sx_get_CD(void *ptr)
+static int sx_carrier_raised(struct tty_port *port)
 {
-	struct sx_port *port = ptr;
-	func_enter2();
-
-	func_exit();
-	return ((sx_read_channel_byte(port, hi_ip) & IP_DCD) != 0);
+	struct sx_port *sp = container_of(port, struct sx_port, gs.port);
+	return ((sx_read_channel_byte(sp, hi_ip) & IP_DCD) != 0);
 }
 
 /* Jeez. Isn't this simple? */
@@ -1536,7 +1532,7 @@
 	}
 	/* tty->low_latency = 1; */
 
-	port->c_dcd = sx_get_CD(port);
+	port->c_dcd = sx_carrier_raised(&port->gs.port);
 	sx_dprintk(SX_DEBUG_OPEN, "at open: cd=%d\n", port->c_dcd);
 
 	func_exit();
@@ -1945,7 +1941,7 @@
 
 static void sx_throttle(struct tty_struct *tty)
 {
-	struct sx_port *port = (struct sx_port *)tty->driver_data;
+	struct sx_port *port = tty->driver_data;
 
 	func_enter2();
 	/* If the port is using any type of input flow
@@ -1959,7 +1955,7 @@
 
 static void sx_unthrottle(struct tty_struct *tty)
 {
-	struct sx_port *port = (struct sx_port *)tty->driver_data;
+	struct sx_port *port = tty->driver_data;
 
 	func_enter2();
 	/* Always unthrottle even if flow control is not enabled on
@@ -2354,6 +2350,10 @@
 	.tiocmset = sx_tiocmset,
 };
 
+static const struct tty_port_operations sx_port_ops = {
+	.carrier_raised = sx_carrier_raised,
+};
+
 static int sx_init_drivers(void)
 {
 	int error;
@@ -2410,6 +2410,7 @@
 		for (j = 0; j < boards[i].nports; j++) {
 			sx_dprintk(SX_DEBUG_INIT, "initing port %d\n", j);
 			tty_port_init(&port->gs.port);
+			port->gs.port.ops = &sx_port_ops;
 			port->gs.magic = SX_MAGIC;
 			port->gs.close_delay = HZ / 2;
 			port->gs.closing_wait = 30 * HZ;
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index 500f517..b8063d4 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -977,7 +977,7 @@
  */
 static void mgsl_stop(struct tty_struct *tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 	
 	if (mgsl_paranoia_check(info, tty->name, "mgsl_stop"))
@@ -1000,7 +1000,7 @@
  */
 static void mgsl_start(struct tty_struct *tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 	
 	if (mgsl_paranoia_check(info, tty->name, "mgsl_start"))
@@ -2057,7 +2057,7 @@
  */
 static void mgsl_flush_chars(struct tty_struct *tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 				
 	if ( debug_level >= DEBUG_LEVEL_INFO )
@@ -2109,7 +2109,7 @@
 		    const unsigned char *buf, int count)
 {
 	int	c, ret = 0;
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 	
 	if ( debug_level >= DEBUG_LEVEL_INFO )
@@ -2232,7 +2232,7 @@
  */
 static int mgsl_write_room(struct tty_struct *tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	int	ret;
 				
 	if (mgsl_paranoia_check(info, tty->name, "mgsl_write_room"))
@@ -2267,7 +2267,7 @@
  */
 static int mgsl_chars_in_buffer(struct tty_struct *tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 			 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgsl_chars_in_buffer(%s)\n",
@@ -2301,7 +2301,7 @@
  */
 static void mgsl_flush_buffer(struct tty_struct *tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -2329,7 +2329,7 @@
  */
 static void mgsl_send_xchar(struct tty_struct *tty, char ch)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -2358,7 +2358,7 @@
  */
 static void mgsl_throttle(struct tty_struct * tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -2388,7 +2388,7 @@
  */
 static void mgsl_unthrottle(struct tty_struct * tty)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -2841,7 +2841,7 @@
  */
 static int tiocmget(struct tty_struct *tty, struct file *file)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned int result;
  	unsigned long flags;
 
@@ -2867,7 +2867,7 @@
 static int tiocmset(struct tty_struct *tty, struct file *file,
 		    unsigned int set, unsigned int clear)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
  	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -2898,7 +2898,7 @@
  */
 static int mgsl_break(struct tty_struct *tty, int break_state)
 {
-	struct mgsl_struct * info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct * info = tty->driver_data;
 	unsigned long flags;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -2932,7 +2932,7 @@
 static int mgsl_ioctl(struct tty_struct *tty, struct file * file,
 		    unsigned int cmd, unsigned long arg)
 {
-	struct mgsl_struct * info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct * info = tty->driver_data;
 	int ret;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -3042,7 +3042,7 @@
  */
 static void mgsl_set_termios(struct tty_struct *tty, struct ktermios *old_termios)
 {
-	struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct *info = tty->driver_data;
 	unsigned long flags;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -3096,7 +3096,7 @@
  */
 static void mgsl_close(struct tty_struct *tty, struct file * filp)
 {
-	struct mgsl_struct * info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct * info = tty->driver_data;
 
 	if (mgsl_paranoia_check(info, tty->name, "mgsl_close"))
 		return;
@@ -3104,70 +3104,18 @@
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgsl_close(%s) entry, count=%d\n",
 			 __FILE__,__LINE__, info->device_name, info->port.count);
-			 
-	if (!info->port.count)
-		return;
 
-	if (tty_hung_up_p(filp))
+	if (tty_port_close_start(&info->port, tty, filp) == 0)			 
 		goto cleanup;
 			
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * tty->count is 1 and the tty structure will be freed.
-		 * info->port.count should be one in this case.
-		 * if it's not, correct it so that the port is shutdown.
-		 */
-		printk("mgsl_close: bad refcount; tty->count is 1, "
-		       "info->port.count is %d\n", info->port.count);
-		info->port.count = 1;
-	}
-	
-	info->port.count--;
-	
-	/* if at least one open remaining, leave hardware active */
-	if (info->port.count)
-		goto cleanup;
-	
-	info->port.flags |= ASYNC_CLOSING;
-	
-	/* set tty->closing to notify line discipline to 
-	 * only process XON/XOFF characters. Only the N_TTY
-	 * discipline appears to use this (ppp does not).
-	 */
-	tty->closing = 1;
-	
-	/* wait for transmit data to clear all layers */
-	
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		if (debug_level >= DEBUG_LEVEL_INFO)
-			printk("%s(%d):mgsl_close(%s) calling tty_wait_until_sent\n",
-				 __FILE__,__LINE__, info->device_name );
-		tty_wait_until_sent(tty, info->port.closing_wait);
-	}
-		
  	if (info->port.flags & ASYNC_INITIALIZED)
  		mgsl_wait_until_sent(tty, info->timeout);
-
 	mgsl_flush_buffer(tty);
-
 	tty_ldisc_flush(tty);
-		
 	shutdown(info);
-	
-	tty->closing = 0;
+
+	tty_port_close_end(&info->port, tty);	
 	info->port.tty = NULL;
-	
-	if (info->port.blocked_open) {
-		if (info->port.close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
-		}
-		wake_up_interruptible(&info->port.open_wait);
-	}
-	
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-			 
-	wake_up_interruptible(&info->port.close_wait);
-	
 cleanup:			
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgsl_close(%s) exit, count=%d\n", __FILE__,__LINE__,
@@ -3188,7 +3136,7 @@
  */
 static void mgsl_wait_until_sent(struct tty_struct *tty, int timeout)
 {
-	struct mgsl_struct * info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct * info = tty->driver_data;
 	unsigned long orig_jiffies, char_time;
 
 	if (!info )
@@ -3261,7 +3209,7 @@
  */
 static void mgsl_hangup(struct tty_struct *tty)
 {
-	struct mgsl_struct * info = (struct mgsl_struct *)tty->driver_data;
+	struct mgsl_struct * info = tty->driver_data;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgsl_hangup(%s)\n",
@@ -3281,6 +3229,35 @@
 	
 }	/* end of mgsl_hangup() */
 
+/*
+ * carrier_raised()
+ *
+ *	Return true if carrier is raised
+ */
+
+static int carrier_raised(struct tty_port *port)
+{
+	unsigned long flags;
+	struct mgsl_struct *info = container_of(port, struct mgsl_struct, port);
+	
+	spin_lock_irqsave(&info->irq_spinlock, flags);
+ 	usc_get_serial_signals(info);
+	spin_unlock_irqrestore(&info->irq_spinlock, flags);
+	return (info->serial_signals & SerialSignal_DCD) ? 1 : 0;
+}
+
+static void raise_dtr_rts(struct tty_port *port)
+{
+	struct mgsl_struct *info = container_of(port, struct mgsl_struct, port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&info->irq_spinlock,flags);
+	info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
+ 	usc_set_serial_signals(info);
+	spin_unlock_irqrestore(&info->irq_spinlock,flags);
+}
+
+
 /* block_til_ready()
  * 
  * 	Block the current process until the specified port
@@ -3302,6 +3279,8 @@
 	bool		do_clocal = false;
 	bool		extra_count = false;
 	unsigned long	flags;
+	int		dcd;
+	struct tty_port *port = &info->port;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):block_til_ready on %s\n",
@@ -3309,7 +3288,7 @@
 
 	if (filp->f_flags & O_NONBLOCK || tty->flags & (1 << TTY_IO_ERROR)){
 		/* nonblock mode is set or port is not enabled */
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -3318,50 +3297,42 @@
 
 	/* Wait for carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * mgsl_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	 
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):block_til_ready before block on %s count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 
 	spin_lock_irqsave(&info->irq_spinlock, flags);
 	if (!tty_hung_up_p(filp)) {
 		extra_count = true;
-		info->port.count--;
+		port->count--;
 	}
 	spin_unlock_irqrestore(&info->irq_spinlock, flags);
-	info->port.blocked_open++;
+	port->blocked_open++;
 	
 	while (1) {
-		if (tty->termios->c_cflag & CBAUD) {
-			spin_lock_irqsave(&info->irq_spinlock,flags);
-			info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
-		 	usc_set_serial_signals(info);
-			spin_unlock_irqrestore(&info->irq_spinlock,flags);
-		}
+		if (tty->termios->c_cflag & CBAUD)
+			tty_port_raise_dtr_rts(port);
 		
 		set_current_state(TASK_INTERRUPTIBLE);
 		
-		if (tty_hung_up_p(filp) || !(info->port.flags & ASYNC_INITIALIZED)){
-			retval = (info->port.flags & ASYNC_HUP_NOTIFY) ?
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)){
+			retval = (port->flags & ASYNC_HUP_NOTIFY) ?
 					-EAGAIN : -ERESTARTSYS;
 			break;
 		}
 		
-		spin_lock_irqsave(&info->irq_spinlock,flags);
-	 	usc_get_serial_signals(info);
-		spin_unlock_irqrestore(&info->irq_spinlock,flags);
+		dcd = tty_port_carrier_raised(&info->port);
 		
- 		if (!(info->port.flags & ASYNC_CLOSING) &&
- 		    (do_clocal || (info->serial_signals & SerialSignal_DCD)) ) {
+ 		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || dcd))
  			break;
-		}
 			
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -3370,24 +3341,25 @@
 		
 		if (debug_level >= DEBUG_LEVEL_INFO)
 			printk("%s(%d):block_til_ready blocking on %s count=%d\n",
-				 __FILE__,__LINE__, tty->driver->name, info->port.count );
+				 __FILE__,__LINE__, tty->driver->name, port->count );
 				 
 		schedule();
 	}
 	
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	
+	/* FIXME: Racy on hangup during close wait */
 	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):block_til_ready after blocking on %s count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 			 
 	if (!retval)
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		
 	return retval;
 	
@@ -4304,6 +4276,12 @@
 
 }	/* end of mgsl_add_device() */
 
+static const struct tty_port_operations mgsl_port_ops = {
+	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
+};
+
+
 /* mgsl_allocate_device()
  * 
  * 	Allocate and initialize a device instance structure
@@ -4322,6 +4300,7 @@
 		printk("Error can't allocate device instance data\n");
 	} else {
 		tty_port_init(&info->port);
+		info->port.ops = &mgsl_port_ops;
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, mgsl_bh_handler);
 		info->max_frame_size = 4096;
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 08911ed..53544e2 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -720,44 +720,9 @@
 		return;
 	DBGINFO(("%s close entry, count=%d\n", info->device_name, info->port.count));
 
-	if (!info->port.count)
-		return;
-
-	if (tty_hung_up_p(filp))
+	if (tty_port_close_start(&info->port, tty, filp) == 0)
 		goto cleanup;
 
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * tty->count is 1 and the tty structure will be freed.
-		 * info->port.count should be one in this case.
-		 * if it's not, correct it so that the port is shutdown.
-		 */
-		DBGERR(("%s close: bad refcount; tty->count=1, "
-		       "info->port.count=%d\n", info->device_name, info->port.count));
-		info->port.count = 1;
-	}
-
-	info->port.count--;
-
-	/* if at least one open remaining, leave hardware active */
-	if (info->port.count)
-		goto cleanup;
-
-	info->port.flags |= ASYNC_CLOSING;
-
-	/* set tty->closing to notify line discipline to
-	 * only process XON/XOFF characters. Only the N_TTY
-	 * discipline appears to use this (ppp does not).
-	 */
-	tty->closing = 1;
-
-	/* wait for transmit data to clear all layers */
-
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		DBGINFO(("%s call tty_wait_until_sent\n", info->device_name));
-		tty_wait_until_sent(tty, info->port.closing_wait);
-	}
-
  	if (info->port.flags & ASYNC_INITIALIZED)
  		wait_until_sent(tty, info->timeout);
 	flush_buffer(tty);
@@ -765,20 +730,8 @@
 
 	shutdown(info);
 
-	tty->closing = 0;
+	tty_port_close_end(&info->port, tty);
 	info->port.tty = NULL;
-
-	if (info->port.blocked_open) {
-		if (info->port.close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
-		}
-		wake_up_interruptible(&info->port.open_wait);
-	}
-
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-
-	wake_up_interruptible(&info->port.close_wait);
-
 cleanup:
 	DBGINFO(("%s close exit, count=%d\n", tty->driver->name, info->port.count));
 }
@@ -3132,6 +3085,29 @@
 	return 0;
 }
 
+static int carrier_raised(struct tty_port *port)
+{
+	unsigned long flags;
+	struct slgt_info *info = container_of(port, struct slgt_info, port);
+
+	spin_lock_irqsave(&info->lock,flags);
+ 	get_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+	return (info->signals & SerialSignal_DCD) ? 1 : 0;
+}
+
+static void raise_dtr_rts(struct tty_port *port)
+{
+	unsigned long flags;
+	struct slgt_info *info = container_of(port, struct slgt_info, port);
+
+	spin_lock_irqsave(&info->lock,flags);
+	info->signals |= SerialSignal_RTS + SerialSignal_DTR;
+ 	set_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+}
+
+
 /*
  *  block current process until the device is ready to open
  */
@@ -3143,12 +3119,14 @@
 	bool		do_clocal = false;
 	bool		extra_count = false;
 	unsigned long	flags;
+	int		cd;
+	struct tty_port *port = &info->port;
 
 	DBGINFO(("%s block_til_ready\n", tty->driver->name));
 
 	if (filp->f_flags & O_NONBLOCK || tty->flags & (1 << TTY_IO_ERROR)){
 		/* nonblock mode is set or port is not enabled */
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -3157,46 +3135,38 @@
 
 	/* Wait for carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	spin_lock_irqsave(&info->lock, flags);
 	if (!tty_hung_up_p(filp)) {
 		extra_count = true;
-		info->port.count--;
+		port->count--;
 	}
 	spin_unlock_irqrestore(&info->lock, flags);
-	info->port.blocked_open++;
+	port->blocked_open++;
 
 	while (1) {
-		if ((tty->termios->c_cflag & CBAUD)) {
-			spin_lock_irqsave(&info->lock,flags);
-			info->signals |= SerialSignal_RTS + SerialSignal_DTR;
-		 	set_signals(info);
-			spin_unlock_irqrestore(&info->lock,flags);
-		}
+		if ((tty->termios->c_cflag & CBAUD))
+			tty_port_raise_dtr_rts(port);
 
 		set_current_state(TASK_INTERRUPTIBLE);
 
-		if (tty_hung_up_p(filp) || !(info->port.flags & ASYNC_INITIALIZED)){
-			retval = (info->port.flags & ASYNC_HUP_NOTIFY) ?
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)){
+			retval = (port->flags & ASYNC_HUP_NOTIFY) ?
 					-EAGAIN : -ERESTARTSYS;
 			break;
 		}
 
-		spin_lock_irqsave(&info->lock,flags);
-	 	get_signals(info);
-		spin_unlock_irqrestore(&info->lock,flags);
+		cd = tty_port_carrier_raised(port);
 
- 		if (!(info->port.flags & ASYNC_CLOSING) &&
- 		    (do_clocal || (info->signals & SerialSignal_DCD)) ) {
+ 		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || cd ))
  			break;
-		}
 
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -3208,14 +3178,14 @@
 	}
 
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 
 	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 
 	if (!retval)
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 
 	DBGINFO(("%s block_til_ready ready, rc=%d\n", tty->driver->name, retval));
 	return retval;
@@ -3444,6 +3414,11 @@
 #endif
 }
 
+static const struct tty_port_operations slgt_port_ops = {
+	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
+};
+
 /*
  *  allocate device instance structure, return NULL on failure
  */
@@ -3458,6 +3433,7 @@
 			driver_name, adapter_num, port_num));
 	} else {
 		tty_port_init(&info->port);
+		info->port.ops = &slgt_port_ops;
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, bh_handler);
 		info->max_frame_size = 4096;
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index 6bdb44f..7b0c5b2 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -558,6 +558,7 @@
 
 static int  startup(SLMP_INFO *info);
 static int  block_til_ready(struct tty_struct *tty, struct file * filp,SLMP_INFO *info);
+static int carrier_raised(struct tty_port *port);
 static void shutdown(SLMP_INFO *info);
 static void program_hw(SLMP_INFO *info);
 static void change_params(SLMP_INFO *info);
@@ -800,7 +801,7 @@
  */
 static void close(struct tty_struct *tty, struct file *filp)
 {
-	SLMP_INFO * info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO * info = tty->driver_data;
 
 	if (sanity_check(info, tty->name, "close"))
 		return;
@@ -809,70 +810,18 @@
 		printk("%s(%d):%s close() entry, count=%d\n",
 			 __FILE__,__LINE__, info->device_name, info->port.count);
 
-	if (!info->port.count)
-		return;
-
-	if (tty_hung_up_p(filp))
+	if (tty_port_close_start(&info->port, tty, filp) == 0)
 		goto cleanup;
-
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * tty->count is 1 and the tty structure will be freed.
-		 * info->port.count should be one in this case.
-		 * if it's not, correct it so that the port is shutdown.
-		 */
-		printk("%s(%d):%s close: bad refcount; tty->count is 1, "
-		       "info->port.count is %d\n",
-			 __FILE__,__LINE__, info->device_name, info->port.count);
-		info->port.count = 1;
-	}
-
-	info->port.count--;
-
-	/* if at least one open remaining, leave hardware active */
-	if (info->port.count)
-		goto cleanup;
-
-	info->port.flags |= ASYNC_CLOSING;
-
-	/* set tty->closing to notify line discipline to
-	 * only process XON/XOFF characters. Only the N_TTY
-	 * discipline appears to use this (ppp does not).
-	 */
-	tty->closing = 1;
-
-	/* wait for transmit data to clear all layers */
-
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		if (debug_level >= DEBUG_LEVEL_INFO)
-			printk("%s(%d):%s close() calling tty_wait_until_sent\n",
-				 __FILE__,__LINE__, info->device_name );
-		tty_wait_until_sent(tty, info->port.closing_wait);
-	}
-
+		
  	if (info->port.flags & ASYNC_INITIALIZED)
  		wait_until_sent(tty, info->timeout);
 
 	flush_buffer(tty);
-
 	tty_ldisc_flush(tty);
-
 	shutdown(info);
 
-	tty->closing = 0;
+	tty_port_close_end(&info->port, tty);
 	info->port.tty = NULL;
-
-	if (info->port.blocked_open) {
-		if (info->port.close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
-		}
-		wake_up_interruptible(&info->port.open_wait);
-	}
-
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-
-	wake_up_interruptible(&info->port.close_wait);
-
 cleanup:
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s close() exit, count=%d\n", __FILE__,__LINE__,
@@ -884,7 +833,7 @@
  */
 static void hangup(struct tty_struct *tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s hangup()\n",
@@ -907,7 +856,7 @@
  */
 static void set_termios(struct tty_struct *tty, struct ktermios *old_termios)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -960,7 +909,7 @@
 		 const unsigned char *buf, int count)
 {
 	int	c, ret = 0;
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1038,7 +987,7 @@
  */
 static int put_char(struct tty_struct *tty, unsigned char ch)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 	int ret = 0;
 
@@ -1075,7 +1024,7 @@
  */
 static void send_xchar(struct tty_struct *tty, char ch)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1099,7 +1048,7 @@
  */
 static void wait_until_sent(struct tty_struct *tty, int timeout)
 {
-	SLMP_INFO * info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO * info = tty->driver_data;
 	unsigned long orig_jiffies, char_time;
 
 	if (!info )
@@ -1166,7 +1115,7 @@
  */
 static int write_room(struct tty_struct *tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	int ret;
 
 	if (sanity_check(info, tty->name, "write_room"))
@@ -1193,7 +1142,7 @@
  */
 static void flush_chars(struct tty_struct *tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if ( debug_level >= DEBUG_LEVEL_INFO )
@@ -1232,7 +1181,7 @@
  */
 static void flush_buffer(struct tty_struct *tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1254,7 +1203,7 @@
  */
 static void tx_hold(struct tty_struct *tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (sanity_check(info, tty->name, "tx_hold"))
@@ -1274,7 +1223,7 @@
  */
 static void tx_release(struct tty_struct *tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (sanity_check(info, tty->name, "tx_release"))
@@ -1304,7 +1253,7 @@
 static int do_ioctl(struct tty_struct *tty, struct file *file,
 		 unsigned int cmd, unsigned long arg)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	int error;
 	struct mgsl_icount cnow;	/* kernel counter temps */
 	struct serial_icounter_struct __user *p_cuser;	/* user space */
@@ -1515,7 +1464,7 @@
  */
 static int chars_in_buffer(struct tty_struct *tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 
 	if (sanity_check(info, tty->name, "chars_in_buffer"))
 		return 0;
@@ -1531,7 +1480,7 @@
  */
 static void throttle(struct tty_struct * tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1556,7 +1505,7 @@
  */
 static void unthrottle(struct tty_struct * tty)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1587,7 +1536,7 @@
 static int set_break(struct tty_struct *tty, int break_state)
 {
 	unsigned char RegValue;
-	SLMP_INFO * info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO * info = tty->driver_data;
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -3269,7 +3218,7 @@
  */
 static int tiocmget(struct tty_struct *tty, struct file *file)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
 	unsigned int result;
  	unsigned long flags;
 
@@ -3295,7 +3244,7 @@
 static int tiocmset(struct tty_struct *tty, struct file *file,
 		    unsigned int set, unsigned int clear)
 {
-	SLMP_INFO *info = (SLMP_INFO *)tty->driver_data;
+	SLMP_INFO *info = tty->driver_data;
  	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -3318,7 +3267,28 @@
 	return 0;
 }
 
+static int carrier_raised(struct tty_port *port)
+{
+	SLMP_INFO *info = container_of(port, SLMP_INFO, port);
+	unsigned long flags;
 
+	spin_lock_irqsave(&info->lock,flags);
+ 	get_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+
+	return (info->serial_signals & SerialSignal_DCD) ? 1 : 0;
+}
+
+static void raise_dtr_rts(struct tty_port *port)
+{
+	SLMP_INFO *info = container_of(port, SLMP_INFO, port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&info->lock,flags);
+	info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
+ 	set_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+}
 
 /* Block the current process until the specified port is ready to open.
  */
@@ -3330,6 +3300,8 @@
 	bool		do_clocal = false;
 	bool		extra_count = false;
 	unsigned long	flags;
+	int		cd;
+	struct tty_port *port = &info->port;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s block_til_ready()\n",
@@ -3338,7 +3310,7 @@
 	if (filp->f_flags & O_NONBLOCK || tty->flags & (1 << TTY_IO_ERROR)){
 		/* nonblock mode is set or port is not enabled */
 		/* just verify that callout device is not active */
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -3347,50 +3319,42 @@
 
 	/* Wait for carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s block_til_ready() before block, count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 
 	spin_lock_irqsave(&info->lock, flags);
 	if (!tty_hung_up_p(filp)) {
 		extra_count = true;
-		info->port.count--;
+		port->count--;
 	}
 	spin_unlock_irqrestore(&info->lock, flags);
-	info->port.blocked_open++;
+	port->blocked_open++;
 
 	while (1) {
-		if ((tty->termios->c_cflag & CBAUD)) {
-			spin_lock_irqsave(&info->lock,flags);
-			info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
-		 	set_signals(info);
-			spin_unlock_irqrestore(&info->lock,flags);
-		}
+		if (tty->termios->c_cflag & CBAUD)
+			tty_port_raise_dtr_rts(port);
 
 		set_current_state(TASK_INTERRUPTIBLE);
 
-		if (tty_hung_up_p(filp) || !(info->port.flags & ASYNC_INITIALIZED)){
-			retval = (info->port.flags & ASYNC_HUP_NOTIFY) ?
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)){
+			retval = (port->flags & ASYNC_HUP_NOTIFY) ?
 					-EAGAIN : -ERESTARTSYS;
 			break;
 		}
 
-		spin_lock_irqsave(&info->lock,flags);
-	 	get_signals(info);
-		spin_unlock_irqrestore(&info->lock,flags);
+		cd = tty_port_carrier_raised(port);
 
- 		if (!(info->port.flags & ASYNC_CLOSING) &&
- 		    (do_clocal || (info->serial_signals & SerialSignal_DCD)) ) {
+ 		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || cd))
  			break;
-		}
 
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -3399,24 +3363,24 @@
 
 		if (debug_level >= DEBUG_LEVEL_INFO)
 			printk("%s(%d):%s block_til_ready() count=%d\n",
-				 __FILE__,__LINE__, tty->driver->name, info->port.count );
+				 __FILE__,__LINE__, tty->driver->name, port->count );
 
 		schedule();
 	}
 
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 
 	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s block_til_ready() after, count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 
 	if (!retval)
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 
 	return retval;
 }
@@ -3782,6 +3746,11 @@
 #endif
 }
 
+static const struct tty_port_operations port_ops = {
+	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
+};
+
 /* Allocate and initialize a device instance structure
  *
  * Return Value:	pointer to SLMP_INFO if success, otherwise NULL
@@ -3798,6 +3767,7 @@
 			__FILE__,__LINE__, adapter_num, port_num);
 	} else {
 		tty_port_init(&info->port);
+		info->port.ops = &port_ops;
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, bh_handler);
 		info->max_frame_size = 4096;
@@ -3940,6 +3910,7 @@
 	.tiocmset = tiocmset,
 };
 
+
 static void synclinkmp_cleanup(void)
 {
 	int rc;
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index db15f9b..d33e5ab 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -1111,9 +1111,7 @@
  *		Locks the line discipline as required
  *		Writes to the tty driver are serialized by the atomic_write_lock
  *	and are then processed in chunks to the device. The line discipline
- *	write method will not be involked in parallel for each device
- *		The line discipline write method is called under the big
- *	kernel lock for historical reasons. New code should not rely on this.
+ *	write method will not be invoked in parallel for each device.
  */
 
 static ssize_t tty_write(struct file *file, const char __user *buf,
@@ -1213,7 +1211,7 @@
  *	be held until the 'fast-open' is also done. Will change once we
  *	have refcounting in the driver and per driver locking
  */
-struct tty_struct *tty_driver_lookup_tty(struct tty_driver *driver,
+static struct tty_struct *tty_driver_lookup_tty(struct tty_driver *driver,
 		struct inode *inode, int idx)
 {
 	struct tty_struct *tty;
@@ -2050,7 +2048,6 @@
 /**
  *	tty_do_resize		-	resize event
  *	@tty: tty being resized
- *	@real_tty: real tty (not the same as tty if using a pty/tty pair)
  *	@rows: rows (character)
  *	@cols: cols (character)
  *
@@ -2058,41 +2055,34 @@
  *	peform a terminal resize correctly
  */
 
-int tty_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-					struct winsize *ws)
+int tty_do_resize(struct tty_struct *tty, struct winsize *ws)
 {
-	struct pid *pgrp, *rpgrp;
+	struct pid *pgrp;
 	unsigned long flags;
 
-	/* For a PTY we need to lock the tty side */
-	mutex_lock(&real_tty->termios_mutex);
-	if (!memcmp(ws, &real_tty->winsize, sizeof(*ws)))
+	/* Lock the tty */
+	mutex_lock(&tty->termios_mutex);
+	if (!memcmp(ws, &tty->winsize, sizeof(*ws)))
 		goto done;
 	/* Get the PID values and reference them so we can
 	   avoid holding the tty ctrl lock while sending signals */
 	spin_lock_irqsave(&tty->ctrl_lock, flags);
 	pgrp = get_pid(tty->pgrp);
-	rpgrp = get_pid(real_tty->pgrp);
 	spin_unlock_irqrestore(&tty->ctrl_lock, flags);
 
 	if (pgrp)
 		kill_pgrp(pgrp, SIGWINCH, 1);
-	if (rpgrp != pgrp && rpgrp)
-		kill_pgrp(rpgrp, SIGWINCH, 1);
-
 	put_pid(pgrp);
-	put_pid(rpgrp);
 
 	tty->winsize = *ws;
-	real_tty->winsize = *ws;
 done:
-	mutex_unlock(&real_tty->termios_mutex);
+	mutex_unlock(&tty->termios_mutex);
 	return 0;
 }
 
 /**
  *	tiocswinsz		-	implement window size set ioctl
- *	@tty; tty
+ *	@tty; tty side of tty
  *	@arg: user buffer for result
  *
  *	Copies the user idea of the window size to the kernel. Traditionally
@@ -2105,17 +2095,16 @@
  *	then calls into the default method.
  */
 
-static int tiocswinsz(struct tty_struct *tty, struct tty_struct *real_tty,
-	struct winsize __user *arg)
+static int tiocswinsz(struct tty_struct *tty, struct winsize __user *arg)
 {
 	struct winsize tmp_ws;
 	if (copy_from_user(&tmp_ws, arg, sizeof(*arg)))
 		return -EFAULT;
 
 	if (tty->ops->resize)
-		return tty->ops->resize(tty, real_tty, &tmp_ws);
+		return tty->ops->resize(tty, &tmp_ws);
 	else
-		return tty_do_resize(tty, real_tty, &tmp_ws);
+		return tty_do_resize(tty, &tmp_ws);
 }
 
 /**
@@ -2540,7 +2529,7 @@
 	case TIOCGWINSZ:
 		return tiocgwinsz(real_tty, p);
 	case TIOCSWINSZ:
-		return tiocswinsz(tty, real_tty, p);
+		return tiocswinsz(real_tty, p);
 	case TIOCCONS:
 		return real_tty != tty ? -EINVAL : tioccons(file);
 	case FIONBIO:
@@ -2785,6 +2774,8 @@
 	INIT_WORK(&tty->hangup_work, do_tty_hangup);
 	mutex_init(&tty->atomic_read_lock);
 	mutex_init(&tty->atomic_write_lock);
+	mutex_init(&tty->output_lock);
+	mutex_init(&tty->echo_lock);
 	spin_lock_init(&tty->read_lock);
 	spin_lock_init(&tty->ctrl_lock);
 	INIT_LIST_HEAD(&tty->tty_files);
diff --git a/drivers/char/tty_ldisc.c b/drivers/char/tty_ldisc.c
index f307f13..7a84b40 100644
--- a/drivers/char/tty_ldisc.c
+++ b/drivers/char/tty_ldisc.c
@@ -316,8 +316,7 @@
 {
 	/* wait_event is a macro */
 	wait_event(tty_ldisc_wait, tty_ldisc_try(tty));
-	if (tty->ldisc.refcount == 0)
-		printk(KERN_ERR "tty_ldisc_ref_wait\n");
+	WARN_ON(tty->ldisc.refcount == 0);
 	return &tty->ldisc;
 }
 
@@ -376,15 +375,17 @@
  *	@tty: terminal to activate ldisc on
  *
  *	Set the TTY_LDISC flag when the line discipline can be called
- *	again. Do necessary wakeups for existing sleepers.
+ *	again. Do necessary wakeups for existing sleepers. Clear the LDISC
+ *	changing flag to indicate any ldisc change is now over.
  *
- *	Note: nobody should set this bit except via this function. Clearing
- *	directly is allowed.
+ *	Note: nobody should set the TTY_LDISC bit except via this function.
+ *	Clearing directly is allowed.
  */
 
 void tty_ldisc_enable(struct tty_struct *tty)
 {
 	set_bit(TTY_LDISC, &tty->flags);
+	clear_bit(TTY_LDISC_CHANGING, &tty->flags);
 	wake_up(&tty_ldisc_wait);
 }
 
@@ -496,7 +497,14 @@
 	 *	reference to the line discipline. The TTY_LDISC bit
 	 *	prevents anyone taking a reference once it is clear.
 	 *	We need the lock to avoid racing reference takers.
+	 *
+	 *	We must clear the TTY_LDISC bit here to avoid a livelock
+	 *	with a userspace app continually trying to use the tty in
+	 *	parallel to the change and re-referencing the tty.
 	 */
+	clear_bit(TTY_LDISC, &tty->flags);
+	if (o_tty)
+		clear_bit(TTY_LDISC, &o_tty->flags);
 
 	spin_lock_irqsave(&tty_ldisc_lock, flags);
 	if (tty->ldisc.refcount || (o_tty && o_tty->ldisc.refcount)) {
@@ -528,7 +536,7 @@
 	 *	If the TTY_LDISC bit is set, then we are racing against
 	 *	another ldisc change
 	 */
-	if (!test_bit(TTY_LDISC, &tty->flags)) {
+	if (test_bit(TTY_LDISC_CHANGING, &tty->flags)) {
 		struct tty_ldisc *ld;
 		spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 		tty_ldisc_put(new_ldisc.ops);
@@ -536,10 +544,14 @@
 		tty_ldisc_deref(ld);
 		goto restart;
 	}
-
-	clear_bit(TTY_LDISC, &tty->flags);
+	/*
+	 *	This flag is used to avoid two parallel ldisc changes. Once
+	 *	open and close are fine grained locked this may work better
+	 *	as a mutex shared with the open/close/hup paths
+	 */
+	set_bit(TTY_LDISC_CHANGING, &tty->flags);
 	if (o_tty)
-		clear_bit(TTY_LDISC, &o_tty->flags);
+		set_bit(TTY_LDISC_CHANGING, &o_tty->flags);
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 	
 	/*
diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index c8f8024..9b8004c 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -7,6 +7,7 @@
 #include <linux/tty.h>
 #include <linux/tty_driver.h>
 #include <linux/tty_flip.h>
+#include <linux/serial.h>
 #include <linux/timer.h>
 #include <linux/string.h>
 #include <linux/slab.h>
@@ -94,3 +95,227 @@
 	spin_unlock_irqrestore(&port->lock, flags);
 }
 EXPORT_SYMBOL(tty_port_tty_set);
+
+/**
+ *	tty_port_hangup		-	hangup helper
+ *	@port: tty port
+ *
+ *	Perform port level tty hangup flag and count changes. Drop the tty
+ *	reference.
+ */
+
+void tty_port_hangup(struct tty_port *port)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&port->lock, flags);
+	port->count = 0;
+	port->flags &= ~ASYNC_NORMAL_ACTIVE;
+	if (port->tty)
+		tty_kref_put(port->tty);
+	port->tty = NULL;
+	spin_unlock_irqrestore(&port->lock, flags);
+	wake_up_interruptible(&port->open_wait);
+}
+EXPORT_SYMBOL(tty_port_hangup);
+
+/**
+ *	tty_port_carrier_raised	-	carrier raised check
+ *	@port: tty port
+ *
+ *	Wrapper for the carrier detect logic. For the moment this is used
+ *	to hide some internal details. This will eventually become entirely
+ *	internal to the tty port.
+ */
+
+int tty_port_carrier_raised(struct tty_port *port)
+{
+	if (port->ops->carrier_raised == NULL)
+		return 1;
+	return port->ops->carrier_raised(port);
+}
+EXPORT_SYMBOL(tty_port_carrier_raised);
+
+/**
+ *	tty_port_raise_dtr_rts	-	Riase DTR/RTS
+ *	@port: tty port
+ *
+ *	Wrapper for the DTR/RTS raise logic. For the moment this is used
+ *	to hide some internal details. This will eventually become entirely
+ *	internal to the tty port.
+ */
+
+void tty_port_raise_dtr_rts(struct tty_port *port)
+{
+	if (port->ops->raise_dtr_rts)
+		port->ops->raise_dtr_rts(port);
+}
+EXPORT_SYMBOL(tty_port_raise_dtr_rts);
+
+/**
+ *	tty_port_block_til_ready	-	Waiting logic for tty open
+ *	@port: the tty port being opened
+ *	@tty: the tty device being bound
+ *	@filp: the file pointer of the opener
+ *
+ *	Implement the core POSIX/SuS tty behaviour when opening a tty device.
+ *	Handles:
+ *		- hangup (both before and during)
+ *		- non blocking open
+ *		- rts/dtr/dcd
+ *		- signals
+ *		- port flags and counts
+ *
+ *	The passed tty_port must implement the carrier_raised method if it can
+ *	do carrier detect and the raise_dtr_rts method if it supports software
+ *	management of these lines. Note that the dtr/rts raise is done each
+ *	iteration as a hangup may have previously dropped them while we wait.
+ */
+ 
+int tty_port_block_til_ready(struct tty_port *port,
+				struct tty_struct *tty, struct file *filp)
+{
+	int do_clocal = 0, retval;
+	unsigned long flags;
+	DECLARE_WAITQUEUE(wait, current);
+	int cd;
+
+	/* block if port is in the process of being closed */
+	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&port->close_wait);
+		if (port->flags & ASYNC_HUP_NOTIFY)
+			return -EAGAIN;
+		else
+			return -ERESTARTSYS;
+	}
+
+	/* if non-blocking mode is set we can pass directly to open unless
+	   the port has just hung up or is in another error state */
+	if ((filp->f_flags & O_NONBLOCK) ||
+			(tty->flags & (1 << TTY_IO_ERROR))) {
+		port->flags |= ASYNC_NORMAL_ACTIVE;
+		return 0;
+	}
+
+	if (C_CLOCAL(tty))
+		do_clocal = 1;
+
+	/* Block waiting until we can proceed. We may need to wait for the
+	   carrier, but we must also wait for any close that is in progress
+	   before the next open may complete */
+
+	retval = 0;
+	add_wait_queue(&port->open_wait, &wait);
+
+	/* The port lock protects the port counts */
+	spin_lock_irqsave(&port->lock, flags);
+	if (!tty_hung_up_p(filp))
+		port->count--;
+	port->blocked_open++;
+	spin_unlock_irqrestore(&port->lock, flags);
+
+	while (1) {
+		/* Indicate we are open */
+		if (tty->termios->c_cflag & CBAUD)
+			tty_port_raise_dtr_rts(port);
+
+		set_current_state(TASK_INTERRUPTIBLE);
+		/* Check for a hangup or uninitialised port. Return accordingly */
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
+				retval = -EAGAIN;
+			else
+				retval = -ERESTARTSYS;
+			break;
+		}
+		/* Probe the carrier. For devices with no carrier detect this
+		   will always return true */
+		cd = tty_port_carrier_raised(port);
+		if (!(port->flags & ASYNC_CLOSING) &&
+				(do_clocal || cd))
+			break;
+		if (signal_pending(current)) {
+			retval = -ERESTARTSYS;
+			break;
+		}
+		schedule();
+	}
+	set_current_state(TASK_RUNNING);
+	remove_wait_queue(&port->open_wait, &wait);
+
+	/* Update counts. A parallel hangup will have set count to zero and
+	   we must not mess that up further */
+	spin_lock_irqsave(&port->lock, flags);
+	if (!tty_hung_up_p(filp))
+		port->count++;
+	port->blocked_open--;
+	if (retval == 0)
+		port->flags |= ASYNC_NORMAL_ACTIVE;
+	spin_unlock_irqrestore(&port->lock, flags);
+	return 0;
+	
+}
+EXPORT_SYMBOL(tty_port_block_til_ready);
+
+int tty_port_close_start(struct tty_port *port, struct tty_struct *tty, struct file *filp)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&port->lock, flags);
+	if (tty_hung_up_p(filp)) {
+		spin_unlock_irqrestore(&port->lock, flags);
+		return 0;
+	}
+
+	if( tty->count == 1 && port->count != 1) {
+		printk(KERN_WARNING
+		    "tty_port_close_start: tty->count = 1 port count = %d.\n",
+								port->count);
+		port->count = 1;
+	}
+	if (--port->count < 0) {
+		printk(KERN_WARNING "tty_port_close_start: count = %d\n",
+								port->count);
+		port->count = 0;
+	}
+
+	if (port->count) {
+		spin_unlock_irqrestore(&port->lock, flags);
+		return 0;
+	}
+	port->flags |= ASYNC_CLOSING;
+	tty->closing = 1;
+	spin_unlock_irqrestore(&port->lock, flags);
+	/* Don't block on a stalled port, just pull the chain */
+	if (tty->flow_stopped)
+		tty_driver_flush_buffer(tty);
+	if (port->flags & ASYNC_INITIALIZED &&
+			port->closing_wait != ASYNC_CLOSING_WAIT_NONE)
+		tty_wait_until_sent(tty, port->closing_wait);
+	return 1;
+}
+EXPORT_SYMBOL(tty_port_close_start);
+
+void tty_port_close_end(struct tty_port *port, struct tty_struct *tty)
+{
+	unsigned long flags;
+
+	tty_ldisc_flush(tty);
+
+	spin_lock_irqsave(&port->lock, flags);
+	tty->closing = 0;
+
+	if (port->blocked_open) {
+		spin_unlock_irqrestore(&port->lock, flags);
+		if (port->close_delay) {
+			msleep_interruptible(
+				jiffies_to_msecs(port->close_delay));
+		}
+		spin_lock_irqsave(&port->lock, flags);
+		wake_up_interruptible(&port->open_wait);
+	}
+	port->flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
+	wake_up_interruptible(&port->close_wait);
+	spin_unlock_irqrestore(&port->lock, flags);
+}
+EXPORT_SYMBOL(tty_port_close_end);
diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c
index 1718b3c..0e8234b 100644
--- a/drivers/char/vme_scc.c
+++ b/drivers/char/vme_scc.c
@@ -69,7 +69,7 @@
 static void scc_enable_tx_interrupts(void * ptr);
 static void scc_disable_rx_interrupts(void * ptr);
 static void scc_enable_rx_interrupts(void * ptr);
-static int  scc_get_CD(void * ptr);
+static int  scc_carrier_raised(struct tty_port *port);
 static void scc_shutdown_port(void * ptr);
 static int scc_set_real_termios(void  *ptr);
 static void scc_hungup(void  *ptr);
@@ -100,7 +100,6 @@
         scc_enable_tx_interrupts,
         scc_disable_rx_interrupts,
         scc_enable_rx_interrupts,
-        scc_get_CD,
         scc_shutdown_port,
         scc_set_real_termios,
         scc_chars_in_buffer,
@@ -129,6 +128,10 @@
 	.break_ctl = scc_break_ctl,
 };
 
+static const struct tty_port_operations scc_port_ops = {
+	.carrier_raised = scc_carrier_raised,
+};
+
 /*----------------------------------------------------------------------------
  * vme_scc_init() and support functions
  *---------------------------------------------------------------------------*/
@@ -176,6 +179,8 @@
 
 	for (i = 0; i < 2; i++) {
 		port = scc_ports + i;
+		tty_port_init(&port->gs.port);
+		port->gs.port.ops = &scc_port_ops;
 		port->gs.magic = SCC_MAGIC;
 		port->gs.close_delay = HZ/2;
 		port->gs.closing_wait = 30 * HZ;
@@ -624,10 +629,10 @@
 }
 
 
-static int scc_get_CD(void *ptr)
+static int scc_carrier_raised(struct tty_port *port)
 {
-	struct scc_port *port = ptr;
-	unsigned channel = port->channel;
+	struct scc_port *sc = container_of(port, struct scc_port, gs.port);
+	unsigned channel = sc->channel;
 
 	return !!(scc_last_status_reg[channel] & SR_DCD);
 }
@@ -638,7 +643,7 @@
 	struct scc_port *port = ptr;
 
 	port->gs.port.flags &= ~ GS_ACTIVE;
-	if (port->gs.port.tty && port->gs.port.tty->termios->c_cflag & HUPCL) {
+	if (port->gs.port.tty && (port->gs.port.tty->termios->c_cflag & HUPCL)) {
 		scc_setsignals (port, 0, 0);
 	}
 }
@@ -779,7 +784,7 @@
 
 static void scc_send_xchar(struct tty_struct *tty, char ch)
 {
-	struct scc_port *port = (struct scc_port *)tty->driver_data;
+	struct scc_port *port = tty->driver_data;
 
 	port->x_char = ch;
 	if (ch)
@@ -896,7 +901,7 @@
 		return retval;
 	}
 
-	port->c_dcd = scc_get_CD (port);
+	port->c_dcd = tty_port_carrier_raised(&port->gs.port);
 
 	scc_enable_rx_interrupts(port);
 
@@ -906,7 +911,7 @@
 
 static void scc_throttle (struct tty_struct * tty)
 {
-	struct scc_port *port = (struct scc_port *)tty->driver_data;
+	struct scc_port *port = tty->driver_data;
 	unsigned long	flags;
 	SCC_ACCESS_INIT(port);
 
@@ -922,7 +927,7 @@
 
 static void scc_unthrottle (struct tty_struct * tty)
 {
-	struct scc_port *port = (struct scc_port *)tty->driver_data;
+	struct scc_port *port = tty->driver_data;
 	unsigned long	flags;
 	SCC_ACCESS_INIT(port);
 
@@ -945,7 +950,7 @@
 
 static int scc_break_ctl(struct tty_struct *tty, int break_state)
 {
-	struct scc_port *port = (struct scc_port *)tty->driver_data;
+	struct scc_port *port = tty->driver_data;
 	unsigned long	flags;
 	SCC_ACCESS_INIT(port);
 
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 008176e..8001421 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -819,8 +819,8 @@
  *	ctrl_lock of the tty IFF a tty is passed.
  */
 
-static int vc_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-		struct vc_data *vc, unsigned int cols, unsigned int lines)
+static int vc_do_resize(struct tty_struct *tty, struct vc_data *vc,
+				unsigned int cols, unsigned int lines)
 {
 	unsigned long old_origin, new_origin, new_scr_end, rlth, rrem, err = 0;
 	unsigned int old_cols, old_rows, old_row_size, old_screen_size;
@@ -932,7 +932,7 @@
 		ws.ws_row = vc->vc_rows;
 		ws.ws_col = vc->vc_cols;
 		ws.ws_ypixel = vc->vc_scan_lines;
-		tty_do_resize(tty, real_tty, &ws);
+		tty_do_resize(tty, &ws);
 	}
 
 	if (CON_IS_VISIBLE(vc))
@@ -954,13 +954,12 @@
 
 int vc_resize(struct vc_data *vc, unsigned int cols, unsigned int rows)
 {
-	return vc_do_resize(vc->vc_tty, vc->vc_tty, vc, cols, rows);
+	return vc_do_resize(vc->vc_tty, vc, cols, rows);
 }
 
 /**
  *	vt_resize		-	resize a VT
  *	@tty: tty to resize
- *	@real_tty: tty if a pty/tty pair
  *	@ws: winsize attributes
  *
  *	Resize a virtual terminal. This is called by the tty layer as we
@@ -971,14 +970,13 @@
  *	termios_mutex and the tty ctrl_lock in that order.
  */
 
-int vt_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-	struct winsize *ws)
+int vt_resize(struct tty_struct *tty, struct winsize *ws)
 {
 	struct vc_data *vc = tty->driver_data;
 	int ret;
 
 	acquire_console_sem();
-	ret = vc_do_resize(tty, real_tty, vc, ws->ws_col, ws->ws_row);
+	ret = vc_do_resize(tty, vc, ws->ws_col, ws->ws_row);
 	release_console_sem();
 	return ret;
 }
@@ -2679,7 +2677,7 @@
 {
 	if (tty->stopped)
 		return 0;
-	return 4096;		/* No limit, really; we're not buffering */
+	return 32768;		/* No limit, really; we're not buffering */
 }
 
 static int con_chars_in_buffer(struct tty_struct *tty)
diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c
index 8944ce5..a2dee0e 100644
--- a/drivers/char/vt_ioctl.c
+++ b/drivers/char/vt_ioctl.c
@@ -366,7 +366,7 @@
 int vt_ioctl(struct tty_struct *tty, struct file * file,
 	     unsigned int cmd, unsigned long arg)
 {
-	struct vc_data *vc = (struct vc_data *)tty->driver_data;
+	struct vc_data *vc = tty->driver_data;
 	struct console_font_op op;	/* used in multiple places here */
 	struct kbd_struct * kbd;
 	unsigned int console;
diff --git a/drivers/clocksource/tcb_clksrc.c b/drivers/clocksource/tcb_clksrc.c
index f450588..254f106 100644
--- a/drivers/clocksource/tcb_clksrc.c
+++ b/drivers/clocksource/tcb_clksrc.c
@@ -154,7 +154,6 @@
 		.shift		= 32,
 		/* Should be lower than at91rm9200's system timer */
 		.rating		= 125,
-		.cpumask	= CPU_MASK_CPU0,
 		.set_next_event	= tc_next_event,
 		.set_mode	= tc_mode,
 	},
@@ -195,6 +194,7 @@
 	clkevt.clkevt.max_delta_ns
 		= clockevent_delta2ns(0xffff, &clkevt.clkevt);
 	clkevt.clkevt.min_delta_ns = clockevent_delta2ns(1, &clkevt.clkevt) + 1;
+	clkevt.clkevt.cpumask = cpumask_of(0);
 
 	setup_irq(irq, &tc_irqaction);
 
diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index c9f21e3..4ee85fc 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -137,6 +137,7 @@
 
 config BLK_DEV_IDECD
 	tristate "Include IDE/ATAPI CDROM support"
+	select IDE_ATAPI
 	---help---
 	  If you have a CD-ROM drive using the ATAPI protocol, say Y. ATAPI is
 	  a newer protocol used by IDE CD-ROM and TAPE drives, similar to the
@@ -185,23 +186,6 @@
 	  To compile this driver as a module, choose M here: the
 	  module will be called ide-tape.
 
-config BLK_DEV_IDESCSI
-	tristate "SCSI emulation support (DEPRECATED)"
-	depends on SCSI
-	select IDE_ATAPI
-	---help---
-	  WARNING: ide-scsi is no longer needed for cd writing applications!
-	  The 2.6 kernel supports direct writing to ide-cd, which eliminates
-	  the need for ide-scsi + the entire scsi stack just for writing a
-	  cd. The new method is more efficient in every way.
-
-	  This will provide SCSI host adapter emulation for IDE ATAPI devices,
-	  and will allow you to use a SCSI device driver instead of a native
-	  ATAPI driver.
-
-	  If both this SCSI emulation and native ATAPI support are compiled
-	  into the kernel, the native support will be used.
-
 config BLK_DEV_IDEACPI
 	bool "IDE ACPI support"
 	depends on ACPI
diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index 177e3f8..4107289 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -5,7 +5,7 @@
 EXTRA_CFLAGS				+= -Idrivers/ide
 
 ide-core-y += ide.o ide-ioctls.o ide-io.o ide-iops.o ide-lib.o ide-probe.o \
-	      ide-taskfile.o ide-pm.o ide-park.o ide-pio-blacklist.o
+	      ide-taskfile.o ide-pm.o ide-park.o ide-pio-blacklist.o ide-sysfs.o
 
 # core IDE code
 ide-core-$(CONFIG_IDE_TIMINGS)		+= ide-timings.o
diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 4e58b9e..e8688c0 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -3,6 +3,7 @@
  */
 
 #include <linux/kernel.h>
+#include <linux/cdrom.h>
 #include <linux/delay.h>
 #include <linux/ide.h>
 #include <scsi/scsi.h>
@@ -14,6 +15,13 @@
 #define debug_log(fmt, args...) do {} while (0)
 #endif
 
+#define ATAPI_MIN_CDB_BYTES	12
+
+static inline int dev_is_idecd(ide_drive_t *drive)
+{
+	return drive->media == ide_cdrom || drive->media == ide_optical;
+}
+
 /*
  * Check whether we can support a device,
  * based on the ATAPI IDENTIFY command results.
@@ -233,18 +241,49 @@
 }
 EXPORT_SYMBOL_GPL(ide_retry_pc);
 
-int ide_scsi_expiry(ide_drive_t *drive)
+int ide_cd_expiry(ide_drive_t *drive)
 {
-	struct ide_atapi_pc *pc = drive->pc;
+	struct request *rq = HWGROUP(drive)->rq;
+	unsigned long wait = 0;
 
-	debug_log("%s called for %lu at %lu\n", __func__,
-		  pc->scsi_cmd->serial_number, jiffies);
+	debug_log("%s: rq->cmd[0]: 0x%x\n", __func__, rq->cmd[0]);
 
-	pc->flags |= PC_FLAG_TIMEDOUT;
-
-	return 0; /* we do not want the IDE subsystem to retry */
+	/*
+	 * Some commands are *slow* and normally take a long time to complete.
+	 * Usually we can use the ATAPI "disconnect" to bypass this, but not all
+	 * commands/drives support that. Let ide_timer_expiry keep polling us
+	 * for these.
+	 */
+	switch (rq->cmd[0]) {
+	case GPCMD_BLANK:
+	case GPCMD_FORMAT_UNIT:
+	case GPCMD_RESERVE_RZONE_TRACK:
+	case GPCMD_CLOSE_TRACK:
+	case GPCMD_FLUSH_CACHE:
+		wait = ATAPI_WAIT_PC;
+		break;
+	default:
+		if (!(rq->cmd_flags & REQ_QUIET))
+			printk(KERN_INFO "cmd 0x%x timed out\n",
+					 rq->cmd[0]);
+		wait = 0;
+		break;
+	}
+	return wait;
 }
-EXPORT_SYMBOL_GPL(ide_scsi_expiry);
+EXPORT_SYMBOL_GPL(ide_cd_expiry);
+
+int ide_cd_get_xferlen(struct request *rq)
+{
+	if (blk_fs_request(rq))
+		return 32768;
+	else if (blk_sense_request(rq) || blk_pc_request(rq) ||
+			 rq->cmd_type == REQ_TYPE_ATA_PC)
+		return rq->data_len;
+	else
+		return 0;
+}
+EXPORT_SYMBOL_GPL(ide_cd_get_xferlen);
 
 /*
  * This is the usual interrupt handler which will be called during a packet
@@ -258,21 +297,14 @@
 	struct request *rq = hwif->hwgroup->rq;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	xfer_func_t *xferfunc;
-	ide_expiry_t *expiry;
 	unsigned int timeout, temp;
 	u16 bcount;
-	u8 stat, ireason, scsi = !!(drive->dev_flags & IDE_DFLAG_SCSI), dsc = 0;
+	u8 stat, ireason, dsc = 0;
 
 	debug_log("Enter %s - interrupt handler\n", __func__);
 
-	if (scsi) {
-		timeout = ide_scsi_get_timeout(pc);
-		expiry = ide_scsi_expiry;
-	} else {
-		timeout = (drive->media == ide_floppy) ? WAIT_FLOPPY_CMD
-						       : WAIT_TAPE_CMD;
-		expiry = NULL;
-	}
+	timeout = (drive->media == ide_floppy) ? WAIT_FLOPPY_CMD
+					       : WAIT_TAPE_CMD;
 
 	if (pc->flags & PC_FLAG_TIMEDOUT) {
 		drive->pc_callback(drive, 0);
@@ -284,8 +316,8 @@
 
 	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
 		if (hwif->dma_ops->dma_end(drive) ||
-		    (drive->media == ide_tape && !scsi && (stat & ATA_ERR))) {
-			if (drive->media == ide_floppy && !scsi)
+		    (drive->media == ide_tape && (stat & ATA_ERR))) {
+			if (drive->media == ide_floppy)
 				printk(KERN_ERR "%s: DMA %s error\n",
 					drive->name, rq_data_dir(pc->rq)
 						     ? "write" : "read");
@@ -307,7 +339,7 @@
 
 		local_irq_enable_in_hardirq();
 
-		if (drive->media == ide_tape && !scsi &&
+		if (drive->media == ide_tape &&
 		    (stat & ATA_ERR) && rq->cmd[0] == REQUEST_SENSE)
 			stat &= ~ATA_ERR;
 
@@ -315,11 +347,8 @@
 			/* Error detected */
 			debug_log("%s: I/O error\n", drive->name);
 
-			if (drive->media != ide_tape || scsi) {
+			if (drive->media != ide_tape)
 				pc->rq->errors++;
-				if (scsi)
-					goto cmd_finished;
-			}
 
 			if (rq->cmd[0] == REQUEST_SENSE) {
 				printk(KERN_ERR "%s: I/O error in request sense"
@@ -335,7 +364,6 @@
 			/* queued, but not started */
 			return ide_stopped;
 		}
-cmd_finished:
 		pc->error = 0;
 
 		if ((pc->flags & PC_FLAG_WAIT_FOR_DSC) && (stat & ATA_DSC) == 0)
@@ -382,25 +410,8 @@
 						"us more data than expected - "
 						"discarding data\n",
 						drive->name);
-				if (scsi)
-					temp = pc->buf_size - pc->xferred;
-				else
-					temp = 0;
-				if (temp) {
-					if (pc->sg)
-						drive->pc_io_buffers(drive, pc,
-								     temp, 0);
-					else
-						tp_ops->input_data(drive, NULL,
-							pc->cur_pos, temp);
-					printk(KERN_ERR "%s: transferred %d of "
-							"%d bytes\n",
-							drive->name,
-							temp, bcount);
-				}
-				pc->xferred += temp;
-				pc->cur_pos += temp;
-				ide_pad_transfer(drive, 0, bcount - temp);
+
+				ide_pad_transfer(drive, 0, bcount);
 				goto next_irq;
 			}
 			debug_log("The device wants to send us more data than "
@@ -410,14 +421,13 @@
 	} else
 		xferfunc = tp_ops->output_data;
 
-	if ((drive->media == ide_floppy && !scsi && !pc->buf) ||
-	    (drive->media == ide_tape && !scsi && pc->bh) ||
-	    (scsi && pc->sg)) {
+	if ((drive->media == ide_floppy && !pc->buf) ||
+	    (drive->media == ide_tape && pc->bh)) {
 		int done = drive->pc_io_buffers(drive, pc, bcount,
 				  !!(pc->flags & PC_FLAG_WRITING));
 
 		/* FIXME: don't do partial completions */
-		if (drive->media == ide_floppy && !scsi)
+		if (drive->media == ide_floppy)
 			ide_end_request(drive, 1, done >> 9);
 	} else
 		xferfunc(drive, NULL, pc->cur_pos, bcount);
@@ -430,7 +440,7 @@
 		  rq->cmd[0], bcount);
 next_irq:
 	/* And set the interrupt handler again */
-	ide_set_handler(drive, ide_pc_intr, timeout, expiry);
+	ide_set_handler(drive, ide_pc_intr, timeout, NULL);
 	return ide_started;
 }
 
@@ -479,11 +489,12 @@
 
 static ide_startstop_t ide_transfer_pc(ide_drive_t *drive)
 {
-	struct ide_atapi_pc *pc = drive->pc;
+	struct ide_atapi_pc *uninitialized_var(pc);
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq = hwif->hwgroup->rq;
 	ide_expiry_t *expiry;
 	unsigned int timeout;
+	int cmd_len;
 	ide_startstop_t startstop;
 	u8 ireason;
 
@@ -493,101 +504,124 @@
 		return startstop;
 	}
 
-	ireason = ide_read_ireason(drive);
-	if (drive->media == ide_tape &&
-	    (drive->dev_flags & IDE_DFLAG_SCSI) == 0)
-		ireason = ide_wait_ireason(drive, ireason);
-
-	if ((ireason & ATAPI_COD) == 0 || (ireason & ATAPI_IO)) {
-		printk(KERN_ERR "%s: (IO,CoD) != (0,1) while issuing "
-				"a packet command\n", drive->name);
-		return ide_do_reset(drive);
+	if (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT) {
+		if (drive->dma)
+			drive->waiting_for_dma = 1;
 	}
 
-	/*
-	 * If necessary schedule the packet transfer to occur 'timeout'
-	 * miliseconds later in ide_delayed_transfer_pc() after the device
-	 * says it's ready for a packet.
-	 */
-	if (drive->atapi_flags & IDE_AFLAG_ZIP_DRIVE) {
-		timeout = drive->pc_delay;
-		expiry = &ide_delayed_transfer_pc;
+	if (dev_is_idecd(drive)) {
+		/* ATAPI commands get padded out to 12 bytes minimum */
+		cmd_len = COMMAND_SIZE(rq->cmd[0]);
+		if (cmd_len < ATAPI_MIN_CDB_BYTES)
+			cmd_len = ATAPI_MIN_CDB_BYTES;
+
+		timeout = rq->timeout;
+		expiry  = ide_cd_expiry;
 	} else {
-		if (drive->dev_flags & IDE_DFLAG_SCSI) {
-			timeout = ide_scsi_get_timeout(pc);
-			expiry = ide_scsi_expiry;
+		pc = drive->pc;
+
+		cmd_len = ATAPI_MIN_CDB_BYTES;
+
+		/*
+		 * If necessary schedule the packet transfer to occur 'timeout'
+		 * miliseconds later in ide_delayed_transfer_pc() after the
+		 * device says it's ready for a packet.
+		 */
+		if (drive->atapi_flags & IDE_AFLAG_ZIP_DRIVE) {
+			timeout = drive->pc_delay;
+			expiry = &ide_delayed_transfer_pc;
 		} else {
 			timeout = (drive->media == ide_floppy) ? WAIT_FLOPPY_CMD
 							       : WAIT_TAPE_CMD;
 			expiry = NULL;
 		}
+
+		ireason = ide_read_ireason(drive);
+		if (drive->media == ide_tape)
+			ireason = ide_wait_ireason(drive, ireason);
+
+		if ((ireason & ATAPI_COD) == 0 || (ireason & ATAPI_IO)) {
+			printk(KERN_ERR "%s: (IO,CoD) != (0,1) while issuing "
+					"a packet command\n", drive->name);
+
+			return ide_do_reset(drive);
+		}
 	}
 
 	/* Set the interrupt routine */
 	ide_set_handler(drive, ide_pc_intr, timeout, expiry);
 
 	/* Begin DMA, if necessary */
-	if (pc->flags & PC_FLAG_DMA_OK) {
-		pc->flags |= PC_FLAG_DMA_IN_PROGRESS;
-		hwif->dma_ops->dma_start(drive);
+	if (dev_is_idecd(drive)) {
+		if (drive->dma)
+			hwif->dma_ops->dma_start(drive);
+	} else {
+		if (pc->flags & PC_FLAG_DMA_OK) {
+			pc->flags |= PC_FLAG_DMA_IN_PROGRESS;
+			hwif->dma_ops->dma_start(drive);
+		}
 	}
 
 	/* Send the actual packet */
 	if ((drive->atapi_flags & IDE_AFLAG_ZIP_DRIVE) == 0)
-		hwif->tp_ops->output_data(drive, NULL, rq->cmd, 12);
+		hwif->tp_ops->output_data(drive, NULL, rq->cmd, cmd_len);
 
 	return ide_started;
 }
 
-ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout,
-			     ide_expiry_t *expiry)
+ide_startstop_t ide_issue_pc(ide_drive_t *drive)
 {
-	struct ide_atapi_pc *pc = drive->pc;
+	struct ide_atapi_pc *pc;
 	ide_hwif_t *hwif = drive->hwif;
+	ide_expiry_t *expiry = NULL;
+	unsigned int timeout;
 	u32 tf_flags;
 	u16 bcount;
-	u8 scsi = !!(drive->dev_flags & IDE_DFLAG_SCSI);
 
-	/* We haven't transferred any data yet */
-	pc->xferred = 0;
-	pc->cur_pos = pc->buf;
-
-	/* Request to transfer the entire buffer at once */
-	if (drive->media == ide_tape && scsi == 0)
-		bcount = pc->req_xfer;
-	else
-		bcount = min(pc->req_xfer, 63 * 1024);
-
-	if (pc->flags & PC_FLAG_DMA_ERROR) {
-		pc->flags &= ~PC_FLAG_DMA_ERROR;
-		ide_dma_off(drive);
-	}
-
-	if ((pc->flags & PC_FLAG_DMA_OK) &&
-	    (drive->dev_flags & IDE_DFLAG_USING_DMA)) {
-		if (scsi)
-			hwif->sg_mapped = 1;
-		drive->dma = !hwif->dma_ops->dma_setup(drive);
-		if (scsi)
-			hwif->sg_mapped = 0;
-	}
-
-	if (!drive->dma)
-		pc->flags &= ~PC_FLAG_DMA_OK;
-
-	if (scsi)
-		tf_flags = 0;
-	else if (drive->media == ide_cdrom || drive->media == ide_optical)
+	if (dev_is_idecd(drive)) {
 		tf_flags = IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL;
-	else
+		bcount = ide_cd_get_xferlen(hwif->hwgroup->rq);
+		expiry = ide_cd_expiry;
+		timeout = ATAPI_WAIT_PC;
+
+		if (drive->dma)
+			drive->dma = !hwif->dma_ops->dma_setup(drive);
+	} else {
+		pc = drive->pc;
+
+		/* We haven't transferred any data yet */
+		pc->xferred = 0;
+		pc->cur_pos = pc->buf;
+
 		tf_flags = IDE_TFLAG_OUT_DEVICE;
+		bcount = ((drive->media == ide_tape) ?
+				pc->req_xfer :
+				min(pc->req_xfer, 63 * 1024));
+
+		if (pc->flags & PC_FLAG_DMA_ERROR) {
+			pc->flags &= ~PC_FLAG_DMA_ERROR;
+			ide_dma_off(drive);
+		}
+
+		if ((pc->flags & PC_FLAG_DMA_OK) &&
+		     (drive->dev_flags & IDE_DFLAG_USING_DMA))
+			drive->dma = !hwif->dma_ops->dma_setup(drive);
+
+		if (!drive->dma)
+			pc->flags &= ~PC_FLAG_DMA_OK;
+
+		timeout = (drive->media == ide_floppy) ? WAIT_FLOPPY_CMD
+						       : WAIT_TAPE_CMD;
+	}
 
 	ide_pktcmd_tf_load(drive, tf_flags, bcount, drive->dma);
 
 	/* Issue the packet command */
 	if (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT) {
+		if (drive->dma)
+			drive->waiting_for_dma = 0;
 		ide_execute_command(drive, ATA_CMD_PACKET, ide_transfer_pc,
-				    timeout, NULL);
+				    timeout, expiry);
 		return ide_started;
 	} else {
 		ide_execute_pkt_cmd(drive);
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 5daa4dd..1a7410f 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -53,14 +53,6 @@
 
 #include "ide-cd.h"
 
-#define IDECD_DEBUG_LOG		1
-
-#if IDECD_DEBUG_LOG
-#define ide_debug_log(lvl, fmt, args...) __ide_debug_log(lvl, fmt, args)
-#else
-#define ide_debug_log(lvl, fmt, args...) do {} while (0)
-#endif
-
 static DEFINE_MUTEX(idecd_ref_mutex);
 
 static void ide_cd_release(struct kref *);
@@ -519,37 +511,8 @@
 	return 1;
 }
 
-static int cdrom_timer_expiry(ide_drive_t *drive)
-{
-	struct request *rq = HWGROUP(drive)->rq;
-	unsigned long wait = 0;
-
-	ide_debug_log(IDE_DBG_RQ, "Call %s: rq->cmd[0]: 0x%x\n", __func__,
-		      rq->cmd[0]);
-
-	/*
-	 * Some commands are *slow* and normally take a long time to complete.
-	 * Usually we can use the ATAPI "disconnect" to bypass this, but not all
-	 * commands/drives support that. Let ide_timer_expiry keep polling us
-	 * for these.
-	 */
-	switch (rq->cmd[0]) {
-	case GPCMD_BLANK:
-	case GPCMD_FORMAT_UNIT:
-	case GPCMD_RESERVE_RZONE_TRACK:
-	case GPCMD_CLOSE_TRACK:
-	case GPCMD_FLUSH_CACHE:
-		wait = ATAPI_WAIT_PC;
-		break;
-	default:
-		if (!(rq->cmd_flags & REQ_QUIET))
-			printk(KERN_INFO PFX "cmd 0x%x timed out\n",
-					 rq->cmd[0]);
-		wait = 0;
-		break;
-	}
-	return wait;
-}
+static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *);
+static ide_startstop_t cdrom_newpc_intr(ide_drive_t *);
 
 /*
  * Set up the device registers for transferring a packet command on DEV,
@@ -559,11 +522,13 @@
  * called when the interrupt from the drive arrives.  Otherwise, HANDLER
  * will be called immediately after the drive is prepared for the transfer.
  */
-static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive,
-						  int xferlen,
-						  ide_handler_t *handler)
+static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
+	struct request *rq = hwif->hwgroup->rq;
+	int xferlen;
+
+	xferlen = ide_cd_get_xferlen(rq);
 
 	ide_debug_log(IDE_DBG_PC, "Call %s, xferlen: %d\n", __func__, xferlen);
 
@@ -581,13 +546,14 @@
 			drive->waiting_for_dma = 0;
 
 		/* packet command */
-		ide_execute_command(drive, ATA_CMD_PACKET, handler,
-				    ATAPI_WAIT_PC, cdrom_timer_expiry);
+		ide_execute_command(drive, ATA_CMD_PACKET,
+				    cdrom_transfer_packet_command,
+				    ATAPI_WAIT_PC, ide_cd_expiry);
 		return ide_started;
 	} else {
 		ide_execute_pkt_cmd(drive);
 
-		return (*handler) (drive);
+		return cdrom_transfer_packet_command(drive);
 	}
 }
 
@@ -598,11 +564,10 @@
  * there's data ready.
  */
 #define ATAPI_MIN_CDB_BYTES 12
-static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive,
-					  struct request *rq,
-					  ide_handler_t *handler)
+static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
+	struct request *rq = hwif->hwgroup->rq;
 	int cmd_len;
 	ide_startstop_t startstop;
 
@@ -629,7 +594,7 @@
 	}
 
 	/* arm the interrupt handler */
-	ide_set_handler(drive, handler, rq->timeout, cdrom_timer_expiry);
+	ide_set_handler(drive, cdrom_newpc_intr, rq->timeout, ide_cd_expiry);
 
 	/* ATAPI commands get padded out to 12 bytes minimum */
 	cmd_len = COMMAND_SIZE(rq->cmd[0]);
@@ -717,8 +682,6 @@
 	return 1;
 }
 
-static ide_startstop_t cdrom_newpc_intr(ide_drive_t *);
-
 static ide_startstop_t ide_cd_prepare_rw_request(ide_drive_t *drive,
 						 struct request *rq)
 {
@@ -761,20 +724,6 @@
 }
 
 /*
- * Routine to send a read/write packet command to the drive. This is usually
- * called directly from cdrom_start_{read,write}(). However, for drq_interrupt
- * devices, it is called from an interrupt when the drive is ready to accept
- * the command.
- */
-static ide_startstop_t cdrom_start_rw_cont(ide_drive_t *drive)
-{
-	struct request *rq = drive->hwif->hwgroup->rq;
-
-	/* send the command to the drive and return */
-	return cdrom_transfer_packet_command(drive, rq, cdrom_newpc_intr);
-}
-
-/*
  * Fix up a possibly partially-processed request so that we can start it over
  * entirely, or even put it back on the request queue.
  */
@@ -1096,7 +1045,7 @@
 	} else {
 		timeout = ATAPI_WAIT_PC;
 		if (!blk_fs_request(rq))
-			expiry = cdrom_timer_expiry;
+			expiry = ide_cd_expiry;
 	}
 
 	ide_set_handler(drive, cdrom_newpc_intr, timeout, expiry);
@@ -1163,13 +1112,6 @@
 	return ide_started;
 }
 
-static ide_startstop_t cdrom_do_newpc_cont(ide_drive_t *drive)
-{
-	struct request *rq = HWGROUP(drive)->rq;
-
-	return cdrom_transfer_packet_command(drive, rq, cdrom_newpc_intr);
-}
-
 static void cdrom_do_block_pc(ide_drive_t *drive, struct request *rq)
 {
 
@@ -1214,18 +1156,12 @@
 static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 					sector_t block)
 {
-	ide_handler_t *fn;
-	int xferlen;
-
 	ide_debug_log(IDE_DBG_RQ, "Call %s, rq->cmd[0]: 0x%x, "
 		      "rq->cmd_type: 0x%x, block: %llu\n",
 		      __func__, rq->cmd[0], rq->cmd_type,
 		      (unsigned long long)block);
 
 	if (blk_fs_request(rq)) {
-		xferlen = 32768;
-		fn = cdrom_start_rw_cont;
-
 		if (cdrom_start_rw(drive, rq) == ide_stopped)
 			return ide_stopped;
 
@@ -1233,9 +1169,6 @@
 			return ide_stopped;
 	} else if (blk_sense_request(rq) || blk_pc_request(rq) ||
 		   rq->cmd_type == REQ_TYPE_ATA_PC) {
-		xferlen = rq->data_len;
-		fn = cdrom_do_newpc_cont;
-
 		if (!rq->timeout)
 			rq->timeout = ATAPI_WAIT_PC;
 
@@ -1250,7 +1183,7 @@
 		return ide_stopped;
 	}
 
-	return cdrom_start_packet_command(drive, xferlen, fn);
+	return cdrom_start_packet_command(drive);
 }
 
 /*
diff --git a/drivers/ide/ide-cd.h b/drivers/ide/ide-cd.h
index d5ce336..bf676b2 100644
--- a/drivers/ide/ide-cd.h
+++ b/drivers/ide/ide-cd.h
@@ -8,10 +8,14 @@
 #include <linux/cdrom.h>
 #include <asm/byteorder.h>
 
-/*
- * typical timeout for packet command
- */
-#define ATAPI_WAIT_PC		(60 * HZ)
+#define IDECD_DEBUG_LOG		0
+
+#if IDECD_DEBUG_LOG
+#define ide_debug_log(lvl, fmt, args...) __ide_debug_log(lvl, fmt, args)
+#else
+#define ide_debug_log(lvl, fmt, args...) do {} while (0)
+#endif
+
 #define ATAPI_WAIT_WRITE_BUSY	(10 * HZ)
 
 /************************************************************************/
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index aeb1ad7..0a48e2d 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -197,7 +197,7 @@
 
 	pc->retries++;
 
-	return ide_issue_pc(drive, WAIT_FLOPPY_CMD, NULL);
+	return ide_issue_pc(drive);
 }
 
 void ide_floppy_create_read_capacity_cmd(struct ide_atapi_pc *pc)
@@ -342,38 +342,38 @@
  * Look at the flexible disk page parameters. We ignore the CHS capacity
  * parameters and use the LBA parameters instead.
  */
-static int ide_floppy_get_flexible_disk_page(ide_drive_t *drive)
+static int ide_floppy_get_flexible_disk_page(ide_drive_t *drive,
+					     struct ide_atapi_pc *pc)
 {
 	struct ide_disk_obj *floppy = drive->driver_data;
 	struct gendisk *disk = floppy->disk;
-	struct ide_atapi_pc pc;
 	u8 *page;
 	int capacity, lba_capacity;
 	u16 transfer_rate, sector_size, cyls, rpm;
 	u8 heads, sectors;
 
-	ide_floppy_create_mode_sense_cmd(&pc, IDEFLOPPY_FLEXIBLE_DISK_PAGE);
+	ide_floppy_create_mode_sense_cmd(pc, IDEFLOPPY_FLEXIBLE_DISK_PAGE);
 
-	if (ide_queue_pc_tail(drive, disk, &pc)) {
+	if (ide_queue_pc_tail(drive, disk, pc)) {
 		printk(KERN_ERR PFX "Can't get flexible disk page params\n");
 		return 1;
 	}
 
-	if (pc.buf[3] & 0x80)
+	if (pc->buf[3] & 0x80)
 		drive->dev_flags |= IDE_DFLAG_WP;
 	else
 		drive->dev_flags &= ~IDE_DFLAG_WP;
 
 	set_disk_ro(disk, !!(drive->dev_flags & IDE_DFLAG_WP));
 
-	page = &pc.buf[8];
+	page = &pc->buf[8];
 
-	transfer_rate = be16_to_cpup((__be16 *)&pc.buf[8 + 2]);
-	sector_size   = be16_to_cpup((__be16 *)&pc.buf[8 + 6]);
-	cyls          = be16_to_cpup((__be16 *)&pc.buf[8 + 8]);
-	rpm           = be16_to_cpup((__be16 *)&pc.buf[8 + 28]);
-	heads         = pc.buf[8 + 4];
-	sectors       = pc.buf[8 + 5];
+	transfer_rate = be16_to_cpup((__be16 *)&pc->buf[8 + 2]);
+	sector_size   = be16_to_cpup((__be16 *)&pc->buf[8 + 6]);
+	cyls          = be16_to_cpup((__be16 *)&pc->buf[8 + 8]);
+	rpm           = be16_to_cpup((__be16 *)&pc->buf[8 + 28]);
+	heads         = pc->buf[8 + 4];
+	sectors       = pc->buf[8 + 5];
 
 	capacity = cyls * heads * sectors * sector_size;
 
@@ -499,7 +499,7 @@
 
 	/* Clik! disk does not support get_flexible_disk_page */
 	if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE))
-		(void) ide_floppy_get_flexible_disk_page(drive);
+		(void) ide_floppy_get_flexible_disk_page(drive, &pc);
 
 	return rc;
 }
diff --git a/drivers/ide/ide-floppy_ioctl.c b/drivers/ide/ide-floppy_ioctl.c
index 2bc51ff..8f8be85 100644
--- a/drivers/ide/ide-floppy_ioctl.c
+++ b/drivers/ide/ide-floppy_ioctl.c
@@ -31,10 +31,11 @@
  * On exit we set nformats to the number of records we've actually initialized.
  */
 
-static int ide_floppy_get_format_capacities(ide_drive_t *drive, int __user *arg)
+static int ide_floppy_get_format_capacities(ide_drive_t *drive,
+					    struct ide_atapi_pc *pc,
+					    int __user *arg)
 {
 	struct ide_disk_obj *floppy = drive->driver_data;
-	struct ide_atapi_pc pc;
 	u8 header_len, desc_cnt;
 	int i, blocks, length, u_array_size, u_index;
 	int __user *argp;
@@ -45,13 +46,13 @@
 	if (u_array_size <= 0)
 		return -EINVAL;
 
-	ide_floppy_create_read_capacity_cmd(&pc);
-	if (ide_queue_pc_tail(drive, floppy->disk, &pc)) {
+	ide_floppy_create_read_capacity_cmd(pc);
+	if (ide_queue_pc_tail(drive, floppy->disk, pc)) {
 		printk(KERN_ERR "ide-floppy: Can't get floppy parameters\n");
 		return -EIO;
 	}
 
-	header_len = pc.buf[3];
+	header_len = pc->buf[3];
 	desc_cnt = header_len / 8; /* capacity descriptor of 8 bytes */
 
 	u_index = 0;
@@ -68,8 +69,8 @@
 		if (u_index >= u_array_size)
 			break;	/* User-supplied buffer too small */
 
-		blocks = be32_to_cpup((__be32 *)&pc.buf[desc_start]);
-		length = be16_to_cpup((__be16 *)&pc.buf[desc_start + 6]);
+		blocks = be32_to_cpup((__be32 *)&pc->buf[desc_start]);
+		length = be16_to_cpup((__be16 *)&pc->buf[desc_start + 6]);
 
 		if (put_user(blocks, argp))
 			return -EFAULT;
@@ -111,29 +112,28 @@
 	pc->flags |= PC_FLAG_WRITING;
 }
 
-static int ide_floppy_get_sfrp_bit(ide_drive_t *drive)
+static int ide_floppy_get_sfrp_bit(ide_drive_t *drive, struct ide_atapi_pc *pc)
 {
 	struct ide_disk_obj *floppy = drive->driver_data;
-	struct ide_atapi_pc pc;
 
 	drive->atapi_flags &= ~IDE_AFLAG_SRFP;
 
-	ide_floppy_create_mode_sense_cmd(&pc, IDEFLOPPY_CAPABILITIES_PAGE);
-	pc.flags |= PC_FLAG_SUPPRESS_ERROR;
+	ide_floppy_create_mode_sense_cmd(pc, IDEFLOPPY_CAPABILITIES_PAGE);
+	pc->flags |= PC_FLAG_SUPPRESS_ERROR;
 
-	if (ide_queue_pc_tail(drive, floppy->disk, &pc))
+	if (ide_queue_pc_tail(drive, floppy->disk, pc))
 		return 1;
 
-	if (pc.buf[8 + 2] & 0x40)
+	if (pc->buf[8 + 2] & 0x40)
 		drive->atapi_flags |= IDE_AFLAG_SRFP;
 
 	return 0;
 }
 
-static int ide_floppy_format_unit(ide_drive_t *drive, int __user *arg)
+static int ide_floppy_format_unit(ide_drive_t *drive, struct ide_atapi_pc *pc,
+				  int __user *arg)
 {
 	struct ide_disk_obj *floppy = drive->driver_data;
-	struct ide_atapi_pc pc;
 	int blocks, length, flags, err = 0;
 
 	if (floppy->openers > 1) {
@@ -166,10 +166,10 @@
 		goto out;
 	}
 
-	(void)ide_floppy_get_sfrp_bit(drive);
-	ide_floppy_create_format_unit_cmd(&pc, blocks, length, flags);
+	ide_floppy_get_sfrp_bit(drive, pc);
+	ide_floppy_create_format_unit_cmd(pc, blocks, length, flags);
 
-	if (ide_queue_pc_tail(drive, floppy->disk, &pc))
+	if (ide_queue_pc_tail(drive, floppy->disk, pc))
 		err = -EIO;
 
 out:
@@ -188,15 +188,16 @@
  * the dsc bit, and return either 0 or 65536.
  */
 
-static int ide_floppy_get_format_progress(ide_drive_t *drive, int __user *arg)
+static int ide_floppy_get_format_progress(ide_drive_t *drive,
+					  struct ide_atapi_pc *pc,
+					  int __user *arg)
 {
 	struct ide_disk_obj *floppy = drive->driver_data;
-	struct ide_atapi_pc pc;
 	int progress_indication = 0x10000;
 
 	if (drive->atapi_flags & IDE_AFLAG_SRFP) {
-		ide_create_request_sense_cmd(drive, &pc);
-		if (ide_queue_pc_tail(drive, floppy->disk, &pc))
+		ide_create_request_sense_cmd(drive, pc);
+		if (ide_queue_pc_tail(drive, floppy->disk, pc))
 			return -EIO;
 
 		if (floppy->sense_key == 2 &&
@@ -241,20 +242,21 @@
 	return 0;
 }
 
-static int ide_floppy_format_ioctl(ide_drive_t *drive, fmode_t mode,
-				   unsigned int cmd, void __user *argp)
+static int ide_floppy_format_ioctl(ide_drive_t *drive, struct ide_atapi_pc *pc,
+				   fmode_t mode, unsigned int cmd,
+				   void __user *argp)
 {
 	switch (cmd) {
 	case IDEFLOPPY_IOCTL_FORMAT_SUPPORTED:
 		return 0;
 	case IDEFLOPPY_IOCTL_FORMAT_GET_CAPACITY:
-		return ide_floppy_get_format_capacities(drive, argp);
+		return ide_floppy_get_format_capacities(drive, pc, argp);
 	case IDEFLOPPY_IOCTL_FORMAT_START:
 		if (!(mode & FMODE_WRITE))
 			return -EPERM;
-		return ide_floppy_format_unit(drive, (int __user *)argp);
+		return ide_floppy_format_unit(drive, pc, (int __user *)argp);
 	case IDEFLOPPY_IOCTL_FORMAT_GET_PROGRESS:
-		return ide_floppy_get_format_progress(drive, argp);
+		return ide_floppy_get_format_progress(drive, pc, argp);
 	default:
 		return -ENOTTY;
 	}
@@ -270,7 +272,7 @@
 	if (cmd == CDROMEJECT || cmd == CDROM_LOCKDOOR)
 		return ide_floppy_lockdoor(drive, &pc, arg, cmd);
 
-	err = ide_floppy_format_ioctl(drive, mode, cmd, argp);
+	err = ide_floppy_format_ioctl(drive, &pc, mode, cmd, argp);
 	if (err != -ENOTTY)
 		return err;
 
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index ecacc00..1c36a8e 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -426,9 +426,6 @@
 	ide_hwif_t *hwif = drive->hwif;
 	struct scatterlist *sg = hwif->sg_table;
 
-	if (hwif->sg_mapped)	/* needed by ide-scsi */
-		return;
-
 	if (rq->cmd_type != REQ_TYPE_ATA_TASKFILE) {
 		hwif->sg_nents = blk_rq_map_sg(drive->queue, rq, sg);
 	} else {
@@ -667,85 +664,10 @@
 	drive->sleep = timeout + jiffies;
 	drive->dev_flags |= IDE_DFLAG_SLEEPING;
 }
-
 EXPORT_SYMBOL(ide_stall_queue);
 
-#define WAKEUP(drive)	((drive)->service_start + 2 * (drive)->service_time)
-
-/**
- *	choose_drive		-	select a drive to service
- *	@hwgroup: hardware group to select on
- *
- *	choose_drive() selects the next drive which will be serviced.
- *	This is necessary because the IDE layer can't issue commands
- *	to both drives on the same cable, unlike SCSI.
- */
- 
-static inline ide_drive_t *choose_drive (ide_hwgroup_t *hwgroup)
-{
-	ide_drive_t *drive, *best;
-
-repeat:	
-	best = NULL;
-	drive = hwgroup->drive;
-
-	/*
-	 * drive is doing pre-flush, ordered write, post-flush sequence. even
-	 * though that is 3 requests, it must be seen as a single transaction.
-	 * we must not preempt this drive until that is complete
-	 */
-	if (blk_queue_flushing(drive->queue)) {
-		/*
-		 * small race where queue could get replugged during
-		 * the 3-request flush cycle, just yank the plug since
-		 * we want it to finish asap
-		 */
-		blk_remove_plug(drive->queue);
-		return drive;
-	}
-
-	do {
-		u8 dev_s = !!(drive->dev_flags & IDE_DFLAG_SLEEPING);
-		u8 best_s = (best && !!(best->dev_flags & IDE_DFLAG_SLEEPING));
-
-		if ((dev_s == 0 || time_after_eq(jiffies, drive->sleep)) &&
-		    !elv_queue_empty(drive->queue)) {
-			if (best == NULL ||
-			    (dev_s && (best_s == 0 || time_before(drive->sleep, best->sleep))) ||
-			    (best_s == 0 && time_before(WAKEUP(drive), WAKEUP(best)))) {
-				if (!blk_queue_plugged(drive->queue))
-					best = drive;
-			}
-		}
-	} while ((drive = drive->next) != hwgroup->drive);
-
-	if (best && (best->dev_flags & IDE_DFLAG_NICE1) &&
-	    (best->dev_flags & IDE_DFLAG_SLEEPING) == 0 &&
-	    best != hwgroup->drive && best->service_time > WAIT_MIN_SLEEP) {
-		long t = (signed long)(WAKEUP(best) - jiffies);
-		if (t >= WAIT_MIN_SLEEP) {
-		/*
-		 * We *may* have some time to spare, but first let's see if
-		 * someone can potentially benefit from our nice mood today..
-		 */
-			drive = best->next;
-			do {
-				if ((drive->dev_flags & IDE_DFLAG_SLEEPING) == 0
-				 && time_before(jiffies - best->service_time, WAKEUP(drive))
-				 && time_before(WAKEUP(drive), jiffies + t))
-				{
-					ide_stall_queue(best, min_t(long, t, 10 * WAIT_MIN_SLEEP));
-					goto repeat;
-				}
-			} while ((drive = drive->next) != best);
-		}
-	}
-	return best;
-}
-
 /*
  * Issue a new request to a drive from hwgroup
- * Caller must have already done spin_lock_irqsave(&hwgroup->lock, ..);
  *
  * A hwgroup is a serialized group of IDE interfaces.  Usually there is
  * exactly one hwif (interface) per hwgroup, but buggy controllers (eg. CMD640)
@@ -757,8 +679,7 @@
  * possibly along with many other devices.  This is especially common in
  * PCI-based systems with off-board IDE controller cards.
  *
- * The IDE driver uses a per-hwgroup spinlock to protect
- * access to the request queues, and to protect the hwgroup->busy flag.
+ * The IDE driver uses a per-hwgroup lock to protect the hwgroup->busy flag.
  *
  * The first thread into the driver for a particular hwgroup sets the
  * hwgroup->busy flag to indicate that this hwgroup is now active,
@@ -778,69 +699,41 @@
  * the driver.  This makes the driver much more friendlier to shared IRQs
  * than previous designs, while remaining 100% (?) SMP safe and capable.
  */
-static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
+void do_ide_request(struct request_queue *q)
 {
-	ide_drive_t	*drive;
-	ide_hwif_t	*hwif;
+	ide_drive_t	*drive = q->queuedata;
+	ide_hwif_t	*hwif = drive->hwif;
+	ide_hwgroup_t	*hwgroup = hwif->hwgroup;
 	struct request	*rq;
 	ide_startstop_t	startstop;
-	int             loops = 0;
 
-	/* caller must own hwgroup->lock */
-	BUG_ON(!irqs_disabled());
-
-	while (!hwgroup->busy) {
-		hwgroup->busy = 1;
-		/* for atari only */
-		ide_get_lock(ide_intr, hwgroup);
-		drive = choose_drive(hwgroup);
-		if (drive == NULL) {
-			int sleeping = 0;
-			unsigned long sleep = 0; /* shut up, gcc */
-			hwgroup->rq = NULL;
-			drive = hwgroup->drive;
-			do {
-				if ((drive->dev_flags & IDE_DFLAG_SLEEPING) &&
-				    (sleeping == 0 ||
-				     time_before(drive->sleep, sleep))) {
-					sleeping = 1;
-					sleep = drive->sleep;
-				}
-			} while ((drive = drive->next) != hwgroup->drive);
-			if (sleeping) {
+	/*
+	 * drive is doing pre-flush, ordered write, post-flush sequence. even
+	 * though that is 3 requests, it must be seen as a single transaction.
+	 * we must not preempt this drive until that is complete
+	 */
+	if (blk_queue_flushing(q))
 		/*
-		 * Take a short snooze, and then wake up this hwgroup again.
-		 * This gives other hwgroups on the same a chance to
-		 * play fairly with us, just in case there are big differences
-		 * in relative throughputs.. don't want to hog the cpu too much.
+		 * small race where queue could get replugged during
+		 * the 3-request flush cycle, just yank the plug since
+		 * we want it to finish asap
 		 */
-				if (time_before(sleep, jiffies + WAIT_MIN_SLEEP))
-					sleep = jiffies + WAIT_MIN_SLEEP;
-#if 1
-				if (timer_pending(&hwgroup->timer))
-					printk(KERN_CRIT "ide_set_handler: timer already active\n");
-#endif
-				/* so that ide_timer_expiry knows what to do */
-				hwgroup->sleeping = 1;
-				hwgroup->req_gen_timer = hwgroup->req_gen;
-				mod_timer(&hwgroup->timer, sleep);
-				/* we purposely leave hwgroup->busy==1
-				 * while sleeping */
-			} else {
-				/* Ugly, but how can we sleep for the lock
-				 * otherwise? perhaps from tq_disk?
-				 */
+		blk_remove_plug(q);
 
-				/* for atari only */
-				ide_release_lock();
-				hwgroup->busy = 0;
+	spin_unlock_irq(q->queue_lock);
+	spin_lock_irq(&hwgroup->lock);
+
+	if (!ide_lock_hwgroup(hwgroup)) {
+repeat:
+		hwgroup->rq = NULL;
+
+		if (drive->dev_flags & IDE_DFLAG_SLEEPING) {
+			if (time_before(drive->sleep, jiffies)) {
+				ide_unlock_hwgroup(hwgroup);
+				goto plug_device;
 			}
-
-			/* no more work for this hwgroup (for now) */
-			return;
 		}
-	again:
-		hwif = HWIF(drive);
+
 		if (hwif != hwgroup->hwif) {
 			/*
 			 * set nIEN for previous hwif, drives in the
@@ -852,16 +745,20 @@
 		hwgroup->hwif = hwif;
 		hwgroup->drive = drive;
 		drive->dev_flags &= ~(IDE_DFLAG_SLEEPING | IDE_DFLAG_PARKED);
-		drive->service_start = jiffies;
 
+		spin_unlock_irq(&hwgroup->lock);
+		spin_lock_irq(q->queue_lock);
 		/*
 		 * we know that the queue isn't empty, but this can happen
 		 * if the q->prep_rq_fn() decides to kill a request
 		 */
 		rq = elv_next_request(drive->queue);
+		spin_unlock_irq(q->queue_lock);
+		spin_lock_irq(&hwgroup->lock);
+
 		if (!rq) {
-			hwgroup->busy = 0;
-			break;
+			ide_unlock_hwgroup(hwgroup);
+			goto out;
 		}
 
 		/*
@@ -876,53 +773,36 @@
 		 * though. I hope that doesn't happen too much, hopefully not
 		 * unless the subdriver triggers such a thing in its own PM
 		 * state machine.
-		 *
-		 * We count how many times we loop here to make sure we service
-		 * all drives in the hwgroup without looping for ever
 		 */
 		if ((drive->dev_flags & IDE_DFLAG_BLOCKED) &&
 		    blk_pm_request(rq) == 0 &&
 		    (rq->cmd_flags & REQ_PREEMPT) == 0) {
-			drive = drive->next ? drive->next : hwgroup->drive;
-			if (loops++ < 4 && !blk_queue_plugged(drive->queue))
-				goto again;
-			/* We clear busy, there should be no pending ATA command at this point. */
-			hwgroup->busy = 0;
-			break;
+			/* there should be no pending command at this point */
+			ide_unlock_hwgroup(hwgroup);
+			goto plug_device;
 		}
 
 		hwgroup->rq = rq;
 
-		/*
-		 * Some systems have trouble with IDE IRQs arriving while
-		 * the driver is still setting things up.  So, here we disable
-		 * the IRQ used by this interface while the request is being started.
-		 * This may look bad at first, but pretty much the same thing
-		 * happens anyway when any interrupt comes in, IDE or otherwise
-		 *  -- the kernel masks the IRQ while it is being handled.
-		 */
-		if (masked_irq != IDE_NO_IRQ && hwif->irq != masked_irq)
-			disable_irq_nosync(hwif->irq);
-		spin_unlock(&hwgroup->lock);
-		local_irq_enable_in_hardirq();
-			/* allow other IRQs while we start this request */
+		spin_unlock_irq(&hwgroup->lock);
 		startstop = start_request(drive, rq);
 		spin_lock_irq(&hwgroup->lock);
-		if (masked_irq != IDE_NO_IRQ && hwif->irq != masked_irq)
-			enable_irq(hwif->irq);
+
 		if (startstop == ide_stopped)
-			hwgroup->busy = 0;
-	}
-}
+			goto repeat;
+	} else
+		goto plug_device;
+out:
+	spin_unlock_irq(&hwgroup->lock);
+	spin_lock_irq(q->queue_lock);
+	return;
 
-/*
- * Passes the stuff to ide_do_request
- */
-void do_ide_request(struct request_queue *q)
-{
-	ide_drive_t *drive = q->queuedata;
+plug_device:
+	spin_unlock_irq(&hwgroup->lock);
+	spin_lock_irq(q->queue_lock);
 
-	ide_do_request(HWGROUP(drive), IDE_NO_IRQ);
+	if (!elv_queue_empty(q))
+		blk_plug_device(q);
 }
 
 /*
@@ -983,6 +863,17 @@
 	return ret;
 }
 
+static void ide_plug_device(ide_drive_t *drive)
+{
+	struct request_queue *q = drive->queue;
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	if (!elv_queue_empty(q))
+		blk_plug_device(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
 /**
  *	ide_timer_expiry	-	handle lack of an IDE interrupt
  *	@data: timer callback magic (hwgroup)
@@ -1000,10 +891,12 @@
 void ide_timer_expiry (unsigned long data)
 {
 	ide_hwgroup_t	*hwgroup = (ide_hwgroup_t *) data;
+	ide_drive_t	*uninitialized_var(drive);
 	ide_handler_t	*handler;
 	ide_expiry_t	*expiry;
 	unsigned long	flags;
 	unsigned long	wait = -1;
+	int		plug_device = 0;
 
 	spin_lock_irqsave(&hwgroup->lock, flags);
 
@@ -1015,22 +908,15 @@
 		 * or we were "sleeping" to give other devices a chance.
 		 * Either way, we don't really want to complain about anything.
 		 */
-		if (hwgroup->sleeping) {
-			hwgroup->sleeping = 0;
-			hwgroup->busy = 0;
-		}
 	} else {
-		ide_drive_t *drive = hwgroup->drive;
+		drive = hwgroup->drive;
 		if (!drive) {
 			printk(KERN_ERR "ide_timer_expiry: hwgroup->drive was NULL\n");
 			hwgroup->handler = NULL;
 		} else {
 			ide_hwif_t *hwif;
 			ide_startstop_t startstop = ide_stopped;
-			if (!hwgroup->busy) {
-				hwgroup->busy = 1;	/* paranoia */
-				printk(KERN_ERR "%s: ide_timer_expiry: hwgroup->busy was 0 ??\n", drive->name);
-			}
+
 			if ((expiry = hwgroup->expiry) != NULL) {
 				/* continue */
 				if ((wait = expiry(drive)) > 0) {
@@ -1071,15 +957,18 @@
 					ide_error(drive, "irq timeout",
 						  hwif->tp_ops->read_status(hwif));
 			}
-			drive->service_time = jiffies - drive->service_start;
 			spin_lock_irq(&hwgroup->lock);
 			enable_irq(hwif->irq);
-			if (startstop == ide_stopped)
-				hwgroup->busy = 0;
+			if (startstop == ide_stopped) {
+				ide_unlock_hwgroup(hwgroup);
+				plug_device = 1;
+			}
 		}
 	}
-	ide_do_request(hwgroup, IDE_NO_IRQ);
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
+
+	if (plug_device)
+		ide_plug_device(drive);
 }
 
 /**
@@ -1173,10 +1062,11 @@
 	unsigned long flags;
 	ide_hwgroup_t *hwgroup = (ide_hwgroup_t *)dev_id;
 	ide_hwif_t *hwif = hwgroup->hwif;
-	ide_drive_t *drive;
+	ide_drive_t *uninitialized_var(drive);
 	ide_handler_t *handler;
 	ide_startstop_t startstop;
 	irqreturn_t irq_ret = IRQ_NONE;
+	int plug_device = 0;
 
 	spin_lock_irqsave(&hwgroup->lock, flags);
 
@@ -1241,10 +1131,6 @@
 		 */
 		goto out;
 
-	if (!hwgroup->busy) {
-		hwgroup->busy = 1;	/* paranoia */
-		printk(KERN_ERR "%s: ide_intr: hwgroup->busy was 0 ??\n", drive->name);
-	}
 	hwgroup->handler = NULL;
 	hwgroup->req_gen++;
 	del_timer(&hwgroup->timer);
@@ -1267,20 +1153,22 @@
 	 * same irq as is currently being serviced here, and Linux
 	 * won't allow another of the same (on any CPU) until we return.
 	 */
-	drive->service_time = jiffies - drive->service_start;
 	if (startstop == ide_stopped) {
 		if (hwgroup->handler == NULL) {	/* paranoia */
-			hwgroup->busy = 0;
-			ide_do_request(hwgroup, hwif->irq);
-		} else {
-			printk(KERN_ERR "%s: ide_intr: huh? expected NULL handler "
-				"on exit\n", drive->name);
-		}
+			ide_unlock_hwgroup(hwgroup);
+			plug_device = 1;
+		} else
+			printk(KERN_ERR "%s: %s: huh? expected NULL handler "
+					"on exit\n", __func__, drive->name);
 	}
 out_handled:
 	irq_ret = IRQ_HANDLED;
 out:
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
+
+	if (plug_device)
+		ide_plug_device(drive);
+
 	return irq_ret;
 }
 
diff --git a/drivers/ide/ide-ioctls.c b/drivers/ide/ide-ioctls.c
index 28232c6..1be263e 100644
--- a/drivers/ide/ide-ioctls.c
+++ b/drivers/ide/ide-ioctls.c
@@ -95,8 +95,7 @@
 		return -EPERM;
 
 	if (((arg >> IDE_NICE_DSC_OVERLAP) & 1) &&
-	    (drive->media != ide_tape ||
-	     (drive->dev_flags & IDE_DFLAG_SCSI)))
+	    (drive->media != ide_tape))
 		return -EPERM;
 
 	if ((arg >> IDE_NICE_DSC_OVERLAP) & 1)
diff --git a/drivers/ide/ide-park.c b/drivers/ide/ide-park.c
index 63d01c5..678454a 100644
--- a/drivers/ide/ide-park.c
+++ b/drivers/ide/ide-park.c
@@ -16,16 +16,19 @@
 	spin_lock_irq(&hwgroup->lock);
 	if (drive->dev_flags & IDE_DFLAG_PARKED) {
 		int reset_timer = time_before(timeout, drive->sleep);
+		int start_queue = 0;
 
 		drive->sleep = timeout;
 		wake_up_all(&ide_park_wq);
-		if (reset_timer && hwgroup->sleeping &&
-		    del_timer(&hwgroup->timer)) {
-			hwgroup->sleeping = 0;
-			hwgroup->busy = 0;
-			blk_start_queueing(q);
-		}
+		if (reset_timer && del_timer(&hwgroup->timer))
+			start_queue = 1;
 		spin_unlock_irq(&hwgroup->lock);
+
+		if (start_queue) {
+			spin_lock_irq(q->queue_lock);
+			blk_start_queueing(q);
+			spin_unlock_irq(q->queue_lock);
+		}
 		return;
 	}
 	spin_unlock_irq(&hwgroup->lock);
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index a64ec25..c5adb7b 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -101,6 +101,82 @@
 	}
 }
 
+static void ide_classify_ata_dev(ide_drive_t *drive)
+{
+	u16 *id = drive->id;
+	char *m = (char *)&id[ATA_ID_PROD];
+	int is_cfa = ata_id_is_cfa(id);
+
+	/* CF devices are *not* removable in Linux definition of the term */
+	if (is_cfa == 0 && (id[ATA_ID_CONFIG] & (1 << 7)))
+		drive->dev_flags |= IDE_DFLAG_REMOVABLE;
+
+	drive->media = ide_disk;
+
+	if (!ata_id_has_unload(drive->id))
+		drive->dev_flags |= IDE_DFLAG_NO_UNLOAD;
+
+	printk(KERN_INFO "%s: %s, %s DISK drive\n", drive->name, m,
+		is_cfa ? "CFA" : "ATA");
+}
+
+static void ide_classify_atapi_dev(ide_drive_t *drive)
+{
+	u16 *id = drive->id;
+	char *m = (char *)&id[ATA_ID_PROD];
+	u8 type = (id[ATA_ID_CONFIG] >> 8) & 0x1f;
+
+	printk(KERN_INFO "%s: %s, ATAPI ", drive->name, m);
+	switch (type) {
+	case ide_floppy:
+		if (!strstr(m, "CD-ROM")) {
+			if (!strstr(m, "oppy") &&
+			    !strstr(m, "poyp") &&
+			    !strstr(m, "ZIP"))
+				printk(KERN_CONT "cdrom or floppy?, assuming ");
+			if (drive->media != ide_cdrom) {
+				printk(KERN_CONT "FLOPPY");
+				drive->dev_flags |= IDE_DFLAG_REMOVABLE;
+				break;
+			}
+		}
+		/* Early cdrom models used zero */
+		type = ide_cdrom;
+	case ide_cdrom:
+		drive->dev_flags |= IDE_DFLAG_REMOVABLE;
+#ifdef CONFIG_PPC
+		/* kludge for Apple PowerBook internal zip */
+		if (!strstr(m, "CD-ROM") && strstr(m, "ZIP")) {
+			printk(KERN_CONT "FLOPPY");
+			type = ide_floppy;
+			break;
+		}
+#endif
+		printk(KERN_CONT "CD/DVD-ROM");
+		break;
+	case ide_tape:
+		printk(KERN_CONT "TAPE");
+		break;
+	case ide_optical:
+		printk(KERN_CONT "OPTICAL");
+		drive->dev_flags |= IDE_DFLAG_REMOVABLE;
+		break;
+	default:
+		printk(KERN_CONT "UNKNOWN (type %d)", type);
+		break;
+	}
+
+	printk(KERN_CONT " drive\n");
+	drive->media = type;
+	/* an ATAPI device ignores DRDY */
+	drive->ready_stat = 0;
+	if (ata_id_cdb_intr(id))
+		drive->atapi_flags |= IDE_AFLAG_DRQ_INTERRUPT;
+	drive->dev_flags |= IDE_DFLAG_DOORLOCKING;
+	/* we don't do head unloading on ATAPI devices */
+	drive->dev_flags |= IDE_DFLAG_NO_UNLOAD;
+}
+
 /**
  *	do_identify	-	identify a drive
  *	@drive: drive to identify 
@@ -117,7 +193,7 @@
 	u16 *id = drive->id;
 	char *m = (char *)&id[ATA_ID_PROD];
 	unsigned long flags;
-	int bswap = 1, is_cfa;
+	int bswap = 1;
 
 	/* local CPU only; some systems need this */
 	local_irq_save(flags);
@@ -154,91 +230,23 @@
 	if (strstr(m, "E X A B Y T E N E S T"))
 		goto err_misc;
 
-	printk(KERN_INFO "%s: %s, ", drive->name, m);
-
 	drive->dev_flags |= IDE_DFLAG_PRESENT;
 	drive->dev_flags &= ~IDE_DFLAG_DEAD;
 
 	/*
 	 * Check for an ATAPI device
 	 */
-	if (cmd == ATA_CMD_ID_ATAPI) {
-		u8 type = (id[ATA_ID_CONFIG] >> 8) & 0x1f;
-
-		printk(KERN_CONT "ATAPI ");
-		switch (type) {
-			case ide_floppy:
-				if (!strstr(m, "CD-ROM")) {
-					if (!strstr(m, "oppy") &&
-					    !strstr(m, "poyp") &&
-					    !strstr(m, "ZIP"))
-						printk(KERN_CONT "cdrom or floppy?, assuming ");
-					if (drive->media != ide_cdrom) {
-						printk(KERN_CONT "FLOPPY");
-						drive->dev_flags |= IDE_DFLAG_REMOVABLE;
-						break;
-					}
-				}
-				/* Early cdrom models used zero */
-				type = ide_cdrom;
-			case ide_cdrom:
-				drive->dev_flags |= IDE_DFLAG_REMOVABLE;
-#ifdef CONFIG_PPC
-				/* kludge for Apple PowerBook internal zip */
-				if (!strstr(m, "CD-ROM") && strstr(m, "ZIP")) {
-					printk(KERN_CONT "FLOPPY");
-					type = ide_floppy;
-					break;
-				}
-#endif
-				printk(KERN_CONT "CD/DVD-ROM");
-				break;
-			case ide_tape:
-				printk(KERN_CONT "TAPE");
-				break;
-			case ide_optical:
-				printk(KERN_CONT "OPTICAL");
-				drive->dev_flags |= IDE_DFLAG_REMOVABLE;
-				break;
-			default:
-				printk(KERN_CONT "UNKNOWN (type %d)", type);
-				break;
-		}
-		printk(KERN_CONT " drive\n");
-		drive->media = type;
-		/* an ATAPI device ignores DRDY */
-		drive->ready_stat = 0;
-		if (ata_id_cdb_intr(id))
-			drive->atapi_flags |= IDE_AFLAG_DRQ_INTERRUPT;
-		drive->dev_flags |= IDE_DFLAG_DOORLOCKING;
-		/* we don't do head unloading on ATAPI devices */
-		drive->dev_flags |= IDE_DFLAG_NO_UNLOAD;
-		return;
-	}
-
+	if (cmd == ATA_CMD_ID_ATAPI)
+		ide_classify_atapi_dev(drive);
+	else
 	/*
 	 * Not an ATAPI device: looks like a "regular" hard disk
 	 */
-
-	is_cfa = ata_id_is_cfa(id);
-
-	/* CF devices are *not* removable in Linux definition of the term */
-	if (is_cfa == 0 && (id[ATA_ID_CONFIG] & (1 << 7)))
-		drive->dev_flags |= IDE_DFLAG_REMOVABLE;
-
-	drive->media = ide_disk;
-
-	if (!ata_id_has_unload(drive->id))
-		drive->dev_flags |= IDE_DFLAG_NO_UNLOAD;
-
-	printk(KERN_CONT "%s DISK drive\n", is_cfa ? "CFA" : "ATA");
-
+		ide_classify_ata_dev(drive);
 	return;
-
 err_misc:
 	kfree(id);
 	drive->dev_flags &= ~IDE_DFLAG_PRESENT;
-	return;
 }
 
 /**
@@ -641,14 +649,9 @@
 	/* register with global device tree */
 	dev_set_name(&hwif->gendev, hwif->name);
 	hwif->gendev.driver_data = hwif;
-	if (hwif->gendev.parent == NULL) {
-		if (hwif->dev)
-			hwif->gendev.parent = hwif->dev;
-		else
-			/* Would like to do = &device_legacy */
-			hwif->gendev.parent = NULL;
-	}
+	hwif->gendev.parent = hwif->dev;
 	hwif->gendev.release = hwif_release_dev;
+
 	ret = device_register(&hwif->gendev);
 	if (ret < 0) {
 		printk(KERN_WARNING "IDE: %s: device_register error: %d\n",
@@ -878,8 +881,7 @@
 	 *	do not.
 	 */
 
-	q = blk_init_queue_node(do_ide_request, &hwif->hwgroup->lock,
-				hwif_to_node(hwif));
+	q = blk_init_queue_node(do_ide_request, NULL, hwif_to_node(hwif));
 	if (!q)
 		return 1;
 
@@ -1139,8 +1141,6 @@
 
 	if (drive->media == ide_disk)
 		request_module("ide-disk");
-	if (drive->dev_flags & IDE_DFLAG_SCSI)
-		request_module("ide-scsi");
 	if (drive->media == ide_cdrom || drive->media == ide_optical)
 		request_module("ide-cd");
 	if (drive->media == ide_tape)
@@ -1417,58 +1417,6 @@
 	}
 }
 
-static ssize_t store_delete_devices(struct device *portdev,
-				    struct device_attribute *attr,
-				    const char *buf, size_t n)
-{
-	ide_hwif_t *hwif = dev_get_drvdata(portdev);
-
-	if (strncmp(buf, "1", n))
-		return -EINVAL;
-
-	ide_port_unregister_devices(hwif);
-
-	return n;
-};
-
-static DEVICE_ATTR(delete_devices, S_IWUSR, NULL, store_delete_devices);
-
-static ssize_t store_scan(struct device *portdev,
-			  struct device_attribute *attr,
-			  const char *buf, size_t n)
-{
-	ide_hwif_t *hwif = dev_get_drvdata(portdev);
-
-	if (strncmp(buf, "1", n))
-		return -EINVAL;
-
-	ide_port_unregister_devices(hwif);
-	ide_port_scan(hwif);
-
-	return n;
-};
-
-static DEVICE_ATTR(scan, S_IWUSR, NULL, store_scan);
-
-static struct device_attribute *ide_port_attrs[] = {
-	&dev_attr_delete_devices,
-	&dev_attr_scan,
-	NULL
-};
-
-static int ide_sysfs_register_port(ide_hwif_t *hwif)
-{
-	int i, uninitialized_var(rc);
-
-	for (i = 0; ide_port_attrs[i]; i++) {
-		rc = device_create_file(hwif->portdev, ide_port_attrs[i]);
-		if (rc)
-			break;
-	}
-
-	return rc;
-}
-
 static unsigned int ide_indexes;
 
 /**
@@ -1655,9 +1603,6 @@
 		if (hwif == NULL)
 			continue;
 
-		if (hwif->chipset == ide_unknown)
-			hwif->chipset = ide_generic;
-
 		if (hwif->present)
 			hwif_register_devices(hwif);
 	}
diff --git a/drivers/ide/ide-sysfs.c b/drivers/ide/ide-sysfs.c
new file mode 100644
index 0000000..883ffac
--- /dev/null
+++ b/drivers/ide/ide-sysfs.c
@@ -0,0 +1,125 @@
+#include <linux/kernel.h>
+#include <linux/ide.h>
+
+char *ide_media_string(ide_drive_t *drive)
+{
+	switch (drive->media) {
+	case ide_disk:
+		return "disk";
+	case ide_cdrom:
+		return "cdrom";
+	case ide_tape:
+		return "tape";
+	case ide_floppy:
+		return "floppy";
+	case ide_optical:
+		return "optical";
+	default:
+		return "UNKNOWN";
+	}
+}
+
+static ssize_t media_show(struct device *dev, struct device_attribute *attr,
+			  char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", ide_media_string(drive));
+}
+
+static ssize_t drivename_show(struct device *dev, struct device_attribute *attr,
+			      char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", drive->name);
+}
+
+static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "ide:m-%s\n", ide_media_string(drive));
+}
+
+static ssize_t model_show(struct device *dev, struct device_attribute *attr,
+			  char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_PROD]);
+}
+
+static ssize_t firmware_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_FW_REV]);
+}
+
+static ssize_t serial_show(struct device *dev, struct device_attribute *attr,
+			   char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_SERNO]);
+}
+
+struct device_attribute ide_dev_attrs[] = {
+	__ATTR_RO(media),
+	__ATTR_RO(drivename),
+	__ATTR_RO(modalias),
+	__ATTR_RO(model),
+	__ATTR_RO(firmware),
+	__ATTR(serial, 0400, serial_show, NULL),
+	__ATTR(unload_heads, 0644, ide_park_show, ide_park_store),
+	__ATTR_NULL
+};
+
+static ssize_t store_delete_devices(struct device *portdev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t n)
+{
+	ide_hwif_t *hwif = dev_get_drvdata(portdev);
+
+	if (strncmp(buf, "1", n))
+		return -EINVAL;
+
+	ide_port_unregister_devices(hwif);
+
+	return n;
+};
+
+static DEVICE_ATTR(delete_devices, S_IWUSR, NULL, store_delete_devices);
+
+static ssize_t store_scan(struct device *portdev,
+			  struct device_attribute *attr,
+			  const char *buf, size_t n)
+{
+	ide_hwif_t *hwif = dev_get_drvdata(portdev);
+
+	if (strncmp(buf, "1", n))
+		return -EINVAL;
+
+	ide_port_unregister_devices(hwif);
+	ide_port_scan(hwif);
+
+	return n;
+};
+
+static DEVICE_ATTR(scan, S_IWUSR, NULL, store_scan);
+
+static struct device_attribute *ide_port_attrs[] = {
+	&dev_attr_delete_devices,
+	&dev_attr_scan,
+	NULL
+};
+
+int ide_sysfs_register_port(ide_hwif_t *hwif)
+{
+	int i, uninitialized_var(rc);
+
+	for (i = 0; ide_port_attrs[i]; i++) {
+		rc = device_create_file(hwif->portdev, ide_port_attrs[i]);
+		if (rc)
+			break;
+	}
+
+	return rc;
+}
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index a2d470e..5d2aa22 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -694,7 +694,7 @@
 
 	pc->retries++;
 
-	return ide_issue_pc(drive, WAIT_TAPE_CMD, NULL);
+	return ide_issue_pc(drive);
 }
 
 /* A mode sense command is used to "sense" tape parameters. */
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index f0f09f7..46a2d4c 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -440,81 +440,13 @@
 	return 1;
 }
 
-static char *media_string(ide_drive_t *drive)
-{
-	switch (drive->media) {
-	case ide_disk:
-		return "disk";
-	case ide_cdrom:
-		return "cdrom";
-	case ide_tape:
-		return "tape";
-	case ide_floppy:
-		return "floppy";
-	case ide_optical:
-		return "optical";
-	default:
-		return "UNKNOWN";
-	}
-}
-
-static ssize_t media_show(struct device *dev, struct device_attribute *attr, char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", media_string(drive));
-}
-
-static ssize_t drivename_show(struct device *dev, struct device_attribute *attr, char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", drive->name);
-}
-
-static ssize_t modalias_show(struct device *dev, struct device_attribute *attr, char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "ide:m-%s\n", media_string(drive));
-}
-
-static ssize_t model_show(struct device *dev, struct device_attribute *attr,
-			  char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_PROD]);
-}
-
-static ssize_t firmware_show(struct device *dev, struct device_attribute *attr,
-			     char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_FW_REV]);
-}
-
-static ssize_t serial_show(struct device *dev, struct device_attribute *attr,
-			   char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_SERNO]);
-}
-
-static struct device_attribute ide_dev_attrs[] = {
-	__ATTR_RO(media),
-	__ATTR_RO(drivename),
-	__ATTR_RO(modalias),
-	__ATTR_RO(model),
-	__ATTR_RO(firmware),
-	__ATTR(serial, 0400, serial_show, NULL),
-	__ATTR(unload_heads, 0644, ide_park_show, ide_park_store),
-	__ATTR_NULL
-};
-
 static int ide_uevent(struct device *dev, struct kobj_uevent_env *env)
 {
 	ide_drive_t *drive = to_ide_device(dev);
 
-	add_uevent_var(env, "MEDIA=%s", media_string(drive));
+	add_uevent_var(env, "MEDIA=%s", ide_media_string(drive));
 	add_uevent_var(env, "DRIVENAME=%s", drive->name);
-	add_uevent_var(env, "MODALIAS=ide:m-%s", media_string(drive));
+	add_uevent_var(env, "MODALIAS=ide:m-%s", ide_media_string(drive));
 	return 0;
 }
 
diff --git a/drivers/ide/tx4938ide.c b/drivers/ide/tx4938ide.c
index 13b63e7..b4ef218 100644
--- a/drivers/ide/tx4938ide.c
+++ b/drivers/ide/tx4938ide.c
@@ -216,16 +216,17 @@
 #endif	/* __BIG_ENDIAN */
 
 static const struct ide_port_ops tx4938ide_port_ops = {
-	.set_pio_mode = tx4938ide_set_pio_mode,
+	.set_pio_mode		= tx4938ide_set_pio_mode,
 };
 
 static const struct ide_port_info tx4938ide_port_info __initdata = {
-	.port_ops = &tx4938ide_port_ops,
+	.port_ops		= &tx4938ide_port_ops,
 #ifdef __BIG_ENDIAN
-	.tp_ops = &tx4938ide_tp_ops,
+	.tp_ops			= &tx4938ide_tp_ops,
 #endif
-	.host_flags = IDE_HFLAG_MMIO | IDE_HFLAG_NO_DMA,
-	.pio_mask = ATA_PIO5,
+	.host_flags		= IDE_HFLAG_MMIO | IDE_HFLAG_NO_DMA,
+	.pio_mask		= ATA_PIO5,
+	.chipset		= ide_generic,
 };
 
 static int __init tx4938ide_probe(struct platform_device *pdev)
diff --git a/drivers/ide/tx4939ide.c b/drivers/ide/tx4939ide.c
index 97cd9e0..4a8c5a2 100644
--- a/drivers/ide/tx4939ide.c
+++ b/drivers/ide/tx4939ide.c
@@ -623,33 +623,34 @@
 #endif	/* __LITTLE_ENDIAN */
 
 static const struct ide_port_ops tx4939ide_port_ops = {
-	.set_pio_mode = tx4939ide_set_pio_mode,
-	.set_dma_mode = tx4939ide_set_dma_mode,
-	.clear_irq = tx4939ide_clear_irq,
-	.cable_detect = tx4939ide_cable_detect,
+	.set_pio_mode		= tx4939ide_set_pio_mode,
+	.set_dma_mode		= tx4939ide_set_dma_mode,
+	.clear_irq		= tx4939ide_clear_irq,
+	.cable_detect		= tx4939ide_cable_detect,
 };
 
 static const struct ide_dma_ops tx4939ide_dma_ops = {
-	.dma_host_set = tx4939ide_dma_host_set,
-	.dma_setup = tx4939ide_dma_setup,
-	.dma_exec_cmd = ide_dma_exec_cmd,
-	.dma_start = ide_dma_start,
-	.dma_end = tx4939ide_dma_end,
-	.dma_test_irq = tx4939ide_dma_test_irq,
-	.dma_lost_irq = ide_dma_lost_irq,
-	.dma_timeout = ide_dma_timeout,
+	.dma_host_set		= tx4939ide_dma_host_set,
+	.dma_setup		= tx4939ide_dma_setup,
+	.dma_exec_cmd		= ide_dma_exec_cmd,
+	.dma_start		= ide_dma_start,
+	.dma_end		= tx4939ide_dma_end,
+	.dma_test_irq		= tx4939ide_dma_test_irq,
+	.dma_lost_irq		= ide_dma_lost_irq,
+	.dma_timeout		= ide_dma_timeout,
 };
 
 static const struct ide_port_info tx4939ide_port_info __initdata = {
-	.init_hwif = tx4939ide_init_hwif,
-	.init_dma = tx4939ide_init_dma,
-	.port_ops = &tx4939ide_port_ops,
-	.dma_ops = &tx4939ide_dma_ops,
-	.tp_ops = &tx4939ide_tp_ops,
-	.host_flags = IDE_HFLAG_MMIO,
-	.pio_mask = ATA_PIO4,
-	.mwdma_mask = ATA_MWDMA2,
-	.udma_mask = ATA_UDMA5,
+	.init_hwif		= tx4939ide_init_hwif,
+	.init_dma		= tx4939ide_init_dma,
+	.port_ops		= &tx4939ide_port_ops,
+	.dma_ops		= &tx4939ide_dma_ops,
+	.tp_ops			= &tx4939ide_tp_ops,
+	.host_flags		= IDE_HFLAG_MMIO,
+	.pio_mask		= ATA_PIO4,
+	.mwdma_mask		= ATA_MWDMA2,
+	.udma_mask		= ATA_UDMA5,
+	.chipset		= ide_generic,
 };
 
 static int __init tx4939ide_probe(struct platform_device *pdev)
diff --git a/drivers/lguest/interrupts_and_traps.c b/drivers/lguest/interrupts_and_traps.c
index a103906..415fab0 100644
--- a/drivers/lguest/interrupts_and_traps.c
+++ b/drivers/lguest/interrupts_and_traps.c
@@ -222,11 +222,16 @@
 int init_interrupts(void)
 {
 	/* If they want some strange system call vector, reserve it now */
-	if (syscall_vector != SYSCALL_VECTOR
-	    && test_and_set_bit(syscall_vector, used_vectors)) {
-		printk("lg: couldn't reserve syscall %u\n", syscall_vector);
-		return -EBUSY;
+	if (syscall_vector != SYSCALL_VECTOR) {
+		if (test_bit(syscall_vector, used_vectors) ||
+		    vector_used_by_percpu_irq(syscall_vector)) {
+			printk(KERN_ERR "lg: couldn't reserve syscall %u\n",
+				 syscall_vector);
+			return -EBUSY;
+		}
+		set_bit(syscall_vector, used_vectors);
 	}
+
 	return 0;
 }
 
diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index c7630a2..ba0bd3d 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -815,19 +815,20 @@
 	if (err)
 		goto out_free;
 
+	err = -ENOMEM;
 	ubi->peb_buf1 = vmalloc(ubi->peb_size);
 	if (!ubi->peb_buf1)
 		goto out_free;
 
 	ubi->peb_buf2 = vmalloc(ubi->peb_size);
 	if (!ubi->peb_buf2)
-		 goto out_free;
+		goto out_free;
 
 #ifdef CONFIG_MTD_UBI_DEBUG
 	mutex_init(&ubi->dbg_buf_mutex);
 	ubi->dbg_peb_buf = vmalloc(ubi->peb_size);
 	if (!ubi->dbg_peb_buf)
-		 goto out_free;
+		goto out_free;
 #endif
 
 	err = attach_by_scanning(ubi);
diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index b30a0b8..98cf31e 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -721,7 +721,8 @@
 		 * It seems we need to remove volume with name @re->new_name,
 		 * if it exists.
 		 */
-		desc = ubi_open_volume_nm(ubi->ubi_num, re->new_name, UBI_EXCLUSIVE);
+		desc = ubi_open_volume_nm(ubi->ubi_num, re->new_name,
+					  UBI_EXCLUSIVE);
 		if (IS_ERR(desc)) {
 			err = PTR_ERR(desc);
 			if (err == -ENODEV)
diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h
index 78e914d..13777e5 100644
--- a/drivers/mtd/ubi/debug.h
+++ b/drivers/mtd/ubi/debug.h
@@ -27,11 +27,11 @@
 #define dbg_err(fmt, ...) ubi_err(fmt, ##__VA_ARGS__)
 
 #define ubi_assert(expr)  do {                                               \
-        if (unlikely(!(expr))) {                                             \
-                printk(KERN_CRIT "UBI assert failed in %s at %u (pid %d)\n", \
-                       __func__, __LINE__, current->pid);                    \
-                ubi_dbg_dump_stack();                                        \
-        }                                                                    \
+	if (unlikely(!(expr))) {                                             \
+		printk(KERN_CRIT "UBI assert failed in %s at %u (pid %d)\n", \
+		       __func__, __LINE__, current->pid);                    \
+		ubi_dbg_dump_stack();                                        \
+	}                                                                    \
 } while (0)
 
 #define dbg_msg(fmt, ...)                                    \
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index d8966ba..048a606 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -504,12 +504,9 @@
 	if (!vid_hdr)
 		return -ENOMEM;
 
-	mutex_lock(&ubi->buf_mutex);
-
 retry:
 	new_pnum = ubi_wl_get_peb(ubi, UBI_UNKNOWN);
 	if (new_pnum < 0) {
-		mutex_unlock(&ubi->buf_mutex);
 		ubi_free_vid_hdr(ubi, vid_hdr);
 		return new_pnum;
 	}
@@ -529,20 +526,23 @@
 		goto write_error;
 
 	data_size = offset + len;
+	mutex_lock(&ubi->buf_mutex);
 	memset(ubi->peb_buf1 + offset, 0xFF, len);
 
 	/* Read everything before the area where the write failure happened */
 	if (offset > 0) {
 		err = ubi_io_read_data(ubi, ubi->peb_buf1, pnum, 0, offset);
 		if (err && err != UBI_IO_BITFLIPS)
-			goto out_put;
+			goto out_unlock;
 	}
 
 	memcpy(ubi->peb_buf1 + offset, buf, len);
 
 	err = ubi_io_write_data(ubi, ubi->peb_buf1, new_pnum, 0, data_size);
-	if (err)
+	if (err) {
+		mutex_unlock(&ubi->buf_mutex);
 		goto write_error;
+	}
 
 	mutex_unlock(&ubi->buf_mutex);
 	ubi_free_vid_hdr(ubi, vid_hdr);
@@ -553,8 +553,9 @@
 	ubi_msg("data was successfully recovered");
 	return 0;
 
-out_put:
+out_unlock:
 	mutex_unlock(&ubi->buf_mutex);
+out_put:
 	ubi_wl_put_peb(ubi, new_pnum, 1);
 	ubi_free_vid_hdr(ubi, vid_hdr);
 	return err;
@@ -567,7 +568,6 @@
 	ubi_warn("failed to write to PEB %d", new_pnum);
 	ubi_wl_put_peb(ubi, new_pnum, 1);
 	if (++tries > UBI_IO_RETRIES) {
-		mutex_unlock(&ubi->buf_mutex);
 		ubi_free_vid_hdr(ubi, vid_hdr);
 		return err;
 	}
@@ -949,10 +949,14 @@
  * This function copies logical eraseblock from physical eraseblock @from to
  * physical eraseblock @to. The @vid_hdr buffer may be changed by this
  * function. Returns:
- *   o %0  in case of success;
- *   o %1 if the operation was canceled and should be tried later (e.g.,
- *     because a bit-flip was detected at the target PEB);
- *   o %2 if the volume is being deleted and this LEB should not be moved.
+ *   o %0 in case of success;
+ *   o %1 if the operation was canceled because the volume is being deleted
+ *        or because the PEB was put meanwhile;
+ *   o %2 if the operation was canceled because there was a write error to the
+ *        target PEB;
+ *   o %-EAGAIN if the operation was canceled because a bit-flip was detected
+ *     in the target PEB;
+ *   o a negative error code in case of failure.
  */
 int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
 		     struct ubi_vid_hdr *vid_hdr)
@@ -978,7 +982,7 @@
 	/*
 	 * Note, we may race with volume deletion, which means that the volume
 	 * this logical eraseblock belongs to might be being deleted. Since the
-	 * volume deletion unmaps all the volume's logical eraseblocks, it will
+	 * volume deletion un-maps all the volume's logical eraseblocks, it will
 	 * be locked in 'ubi_wl_put_peb()' and wait for the WL worker to finish.
 	 */
 	vol = ubi->volumes[idx];
@@ -986,7 +990,7 @@
 		/* No need to do further work, cancel */
 		dbg_eba("volume %d is being removed, cancel", vol_id);
 		spin_unlock(&ubi->volumes_lock);
-		return 2;
+		return 1;
 	}
 	spin_unlock(&ubi->volumes_lock);
 
@@ -1023,7 +1027,7 @@
 
 	/*
 	 * OK, now the LEB is locked and we can safely start moving it. Since
-	 * this function utilizes thie @ubi->peb1_buf buffer which is shared
+	 * this function utilizes the @ubi->peb1_buf buffer which is shared
 	 * with some other functions, so lock the buffer by taking the
 	 * @ubi->buf_mutex.
 	 */
@@ -1068,8 +1072,11 @@
 	vid_hdr->sqnum = cpu_to_be64(next_sqnum(ubi));
 
 	err = ubi_io_write_vid_hdr(ubi, to, vid_hdr);
-	if (err)
+	if (err) {
+		if (err == -EIO)
+			err = 2;
 		goto out_unlock_buf;
+	}
 
 	cond_resched();
 
@@ -1079,14 +1086,17 @@
 		if (err != UBI_IO_BITFLIPS)
 			ubi_warn("cannot read VID header back from PEB %d", to);
 		else
-			err = 1;
+			err = -EAGAIN;
 		goto out_unlock_buf;
 	}
 
 	if (data_size > 0) {
 		err = ubi_io_write_data(ubi, ubi->peb_buf1, to, 0, aldata_size);
-		if (err)
+		if (err) {
+			if (err == -EIO)
+				err = 2;
 			goto out_unlock_buf;
+		}
 
 		cond_resched();
 
@@ -1101,15 +1111,16 @@
 				ubi_warn("cannot read data back from PEB %d",
 					 to);
 			else
-				err = 1;
+				err = -EAGAIN;
 			goto out_unlock_buf;
 		}
 
 		cond_resched();
 
 		if (memcmp(ubi->peb_buf1, ubi->peb_buf2, aldata_size)) {
-			ubi_warn("read data back from PEB %d - it is different",
-				 to);
+			ubi_warn("read data back from PEB %d and it is "
+				 "different", to);
+			err = -EINVAL;
 			goto out_unlock_buf;
 		}
 	}
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 2fb64be..a74118c 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -637,8 +637,6 @@
 
 	dbg_io("read EC header from PEB %d", pnum);
 	ubi_assert(pnum >= 0 && pnum < ubi->peb_count);
-	if (UBI_IO_DEBUG)
-		verbose = 1;
 
 	err = ubi_io_read(ubi, ec_hdr, pnum, 0, UBI_EC_HDR_SIZE);
 	if (err) {
@@ -685,6 +683,9 @@
 			if (verbose)
 				ubi_warn("no EC header found at PEB %d, "
 					 "only 0xFF bytes", pnum);
+			else if (UBI_IO_DEBUG)
+				dbg_msg("no EC header found at PEB %d, "
+					"only 0xFF bytes", pnum);
 			return UBI_IO_PEB_EMPTY;
 		}
 
@@ -696,7 +697,9 @@
 			ubi_warn("bad magic number at PEB %d: %08x instead of "
 				 "%08x", pnum, magic, UBI_EC_HDR_MAGIC);
 			ubi_dbg_dump_ec_hdr(ec_hdr);
-		}
+		} else if (UBI_IO_DEBUG)
+			dbg_msg("bad magic number at PEB %d: %08x instead of "
+				"%08x", pnum, magic, UBI_EC_HDR_MAGIC);
 		return UBI_IO_BAD_EC_HDR;
 	}
 
@@ -708,7 +711,9 @@
 			ubi_warn("bad EC header CRC at PEB %d, calculated "
 				 "%#08x, read %#08x", pnum, crc, hdr_crc);
 			ubi_dbg_dump_ec_hdr(ec_hdr);
-		}
+		} else if (UBI_IO_DEBUG)
+			dbg_msg("bad EC header CRC at PEB %d, calculated "
+				"%#08x, read %#08x", pnum, crc, hdr_crc);
 		return UBI_IO_BAD_EC_HDR;
 	}
 
@@ -912,8 +917,6 @@
 
 	dbg_io("read VID header from PEB %d", pnum);
 	ubi_assert(pnum >= 0 &&  pnum < ubi->peb_count);
-	if (UBI_IO_DEBUG)
-		verbose = 1;
 
 	p = (char *)vid_hdr - ubi->vid_hdr_shift;
 	err = ubi_io_read(ubi, p, pnum, ubi->vid_hdr_aloffset,
@@ -960,6 +963,9 @@
 			if (verbose)
 				ubi_warn("no VID header found at PEB %d, "
 					 "only 0xFF bytes", pnum);
+			else if (UBI_IO_DEBUG)
+				dbg_msg("no VID header found at PEB %d, "
+					"only 0xFF bytes", pnum);
 			return UBI_IO_PEB_FREE;
 		}
 
@@ -971,7 +977,9 @@
 			ubi_warn("bad magic number at PEB %d: %08x instead of "
 				 "%08x", pnum, magic, UBI_VID_HDR_MAGIC);
 			ubi_dbg_dump_vid_hdr(vid_hdr);
-		}
+		} else if (UBI_IO_DEBUG)
+			dbg_msg("bad magic number at PEB %d: %08x instead of "
+				"%08x", pnum, magic, UBI_VID_HDR_MAGIC);
 		return UBI_IO_BAD_VID_HDR;
 	}
 
@@ -983,7 +991,9 @@
 			ubi_warn("bad CRC at PEB %d, calculated %#08x, "
 				 "read %#08x", pnum, crc, hdr_crc);
 			ubi_dbg_dump_vid_hdr(vid_hdr);
-		}
+		} else if (UBI_IO_DEBUG)
+			dbg_msg("bad CRC at PEB %d, calculated %#08x, "
+				"read %#08x", pnum, crc, hdr_crc);
 		return UBI_IO_BAD_VID_HDR;
 	}
 
@@ -1024,7 +1034,7 @@
 
 	err = paranoid_check_peb_ec_hdr(ubi, pnum);
 	if (err)
-		return err > 0 ? -EINVAL: err;
+		return err > 0 ? -EINVAL : err;
 
 	vid_hdr->magic = cpu_to_be32(UBI_VID_HDR_MAGIC);
 	vid_hdr->version = UBI_VERSION;
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 1c3fa18..4a8ec48 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -74,6 +74,13 @@
 #define UBI_IO_RETRIES 3
 
 /*
+ * Length of the protection queue. The length is effectively equivalent to the
+ * number of (global) erase cycles PEBs are protected from the wear-leveling
+ * worker.
+ */
+#define UBI_PROT_QUEUE_LEN 10
+
+/*
  * Error codes returned by the I/O sub-system.
  *
  * UBI_IO_PEB_EMPTY: the physical eraseblock is empty, i.e. it contains only
@@ -95,7 +102,8 @@
 
 /**
  * struct ubi_wl_entry - wear-leveling entry.
- * @rb: link in the corresponding RB-tree
+ * @u.rb: link in the corresponding (free/used) RB-tree
+ * @u.list: link in the protection queue
  * @ec: erase counter
  * @pnum: physical eraseblock number
  *
@@ -104,7 +112,10 @@
  * RB-trees. See WL sub-system for details.
  */
 struct ubi_wl_entry {
-	struct rb_node rb;
+	union {
+		struct rb_node rb;
+		struct list_head list;
+	} u;
 	int ec;
 	int pnum;
 };
@@ -288,7 +299,7 @@
  * @beb_rsvd_level: normal level of PEBs reserved for bad PEB handling
  *
  * @autoresize_vol_id: ID of the volume which has to be auto-resized at the end
- *                     of UBI ititializetion
+ *                     of UBI initialization
  * @vtbl_slots: how many slots are available in the volume table
  * @vtbl_size: size of the volume table in bytes
  * @vtbl: in-RAM volume table copy
@@ -306,18 +317,17 @@
  * @used: RB-tree of used physical eraseblocks
  * @free: RB-tree of free physical eraseblocks
  * @scrub: RB-tree of physical eraseblocks which need scrubbing
- * @prot: protection trees
- * @prot.pnum: protection tree indexed by physical eraseblock numbers
- * @prot.aec: protection tree indexed by absolute erase counter value
- * @wl_lock: protects the @used, @free, @prot, @lookuptbl, @abs_ec, @move_from,
- *           @move_to, @move_to_put @erase_pending, @wl_scheduled, and @works
- *           fields
+ * @pq: protection queue (contain physical eraseblocks which are temporarily
+ *      protected from the wear-leveling worker)
+ * @pq_head: protection queue head
+ * @wl_lock: protects the @used, @free, @pq, @pq_head, @lookuptbl, @move_from,
+ * 	     @move_to, @move_to_put @erase_pending, @wl_scheduled and @works
+ * 	     fields
  * @move_mutex: serializes eraseblock moves
- * @work_sem: sycnhronizes the WL worker with use tasks
+ * @work_sem: synchronizes the WL worker with use tasks
  * @wl_scheduled: non-zero if the wear-leveling was scheduled
  * @lookuptbl: a table to quickly find a &struct ubi_wl_entry object for any
  *             physical eraseblock
- * @abs_ec: absolute erase counter
  * @move_from: physical eraseblock from where the data is being moved
  * @move_to: physical eraseblock where the data is being moved to
  * @move_to_put: if the "to" PEB was put
@@ -351,11 +361,11 @@
  *
  * @peb_buf1: a buffer of PEB size used for different purposes
  * @peb_buf2: another buffer of PEB size used for different purposes
- * @buf_mutex: proptects @peb_buf1 and @peb_buf2
+ * @buf_mutex: protects @peb_buf1 and @peb_buf2
  * @ckvol_mutex: serializes static volume checking when opening
- * @mult_mutex: serializes operations on multiple volumes, like re-nameing
+ * @mult_mutex: serializes operations on multiple volumes, like re-naming
  * @dbg_peb_buf: buffer of PEB size used for debugging
- * @dbg_buf_mutex: proptects @dbg_peb_buf
+ * @dbg_buf_mutex: protects @dbg_peb_buf
  */
 struct ubi_device {
 	struct cdev cdev;
@@ -392,16 +402,13 @@
 	struct rb_root used;
 	struct rb_root free;
 	struct rb_root scrub;
-	struct {
-		struct rb_root pnum;
-		struct rb_root aec;
-	} prot;
+	struct list_head pq[UBI_PROT_QUEUE_LEN];
+	int pq_head;
 	spinlock_t wl_lock;
 	struct mutex move_mutex;
 	struct rw_semaphore work_sem;
 	int wl_scheduled;
 	struct ubi_wl_entry **lookuptbl;
-	unsigned long long abs_ec;
 	struct ubi_wl_entry *move_from;
 	struct ubi_wl_entry *move_to;
 	int move_to_put;
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index dcb6dac..14901cb 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -22,7 +22,7 @@
  * UBI wear-leveling sub-system.
  *
  * This sub-system is responsible for wear-leveling. It works in terms of
- * physical* eraseblocks and erase counters and knows nothing about logical
+ * physical eraseblocks and erase counters and knows nothing about logical
  * eraseblocks, volumes, etc. From this sub-system's perspective all physical
  * eraseblocks are of two types - used and free. Used physical eraseblocks are
  * those that were "get" by the 'ubi_wl_get_peb()' function, and free physical
@@ -55,8 +55,39 @@
  *
  * As it was said, for the UBI sub-system all physical eraseblocks are either
  * "free" or "used". Free eraseblock are kept in the @wl->free RB-tree, while
- * used eraseblocks are kept in a set of different RB-trees: @wl->used,
- * @wl->prot.pnum, @wl->prot.aec, and @wl->scrub.
+ * used eraseblocks are kept in @wl->used or @wl->scrub RB-trees, or
+ * (temporarily) in the @wl->pq queue.
+ *
+ * When the WL sub-system returns a physical eraseblock, the physical
+ * eraseblock is protected from being moved for some "time". For this reason,
+ * the physical eraseblock is not directly moved from the @wl->free tree to the
+ * @wl->used tree. There is a protection queue in between where this
+ * physical eraseblock is temporarily stored (@wl->pq).
+ *
+ * All this protection stuff is needed because:
+ *  o we don't want to move physical eraseblocks just after we have given them
+ *    to the user; instead, we first want to let users fill them up with data;
+ *
+ *  o there is a chance that the user will put the physical eraseblock very
+ *    soon, so it makes sense not to move it for some time, but wait; this is
+ *    especially important in case of "short term" physical eraseblocks.
+ *
+ * Physical eraseblocks stay protected only for limited time. But the "time" is
+ * measured in erase cycles in this case. This is implemented with help of the
+ * protection queue. Eraseblocks are put to the tail of this queue when they
+ * are returned by the 'ubi_wl_get_peb()', and eraseblocks are removed from the
+ * head of the queue on each erase operation (for any eraseblock). So the
+ * length of the queue defines how may (global) erase cycles PEBs are protected.
+ *
+ * To put it differently, each physical eraseblock has 2 main states: free and
+ * used. The former state corresponds to the @wl->free tree. The latter state
+ * is split up on several sub-states:
+ * o the WL movement is allowed (@wl->used tree);
+ * o the WL movement is temporarily prohibited (@wl->pq queue);
+ * o scrubbing is needed (@wl->scrub tree).
+ *
+ * Depending on the sub-state, wear-leveling entries of the used physical
+ * eraseblocks may be kept in one of those structures.
  *
  * Note, in this implementation, we keep a small in-RAM object for each physical
  * eraseblock. This is surely not a scalable solution. But it appears to be good
@@ -70,9 +101,6 @@
  * target PEB, we pick a PEB with the highest EC if our PEB is "old" and we
  * pick target PEB with an average EC if our PEB is not very "old". This is a
  * room for future re-works of the WL sub-system.
- *
- * Note: the stuff with protection trees looks too complex and is difficult to
- * understand. Should be fixed.
  */
 
 #include <linux/slab.h>
@@ -85,14 +113,6 @@
 #define WL_RESERVED_PEBS 1
 
 /*
- * How many erase cycles are short term, unknown, and long term physical
- * eraseblocks protected.
- */
-#define ST_PROTECTION 16
-#define U_PROTECTION  10
-#define LT_PROTECTION 4
-
-/*
  * Maximum difference between two erase counters. If this threshold is
  * exceeded, the WL sub-system starts moving data from used physical
  * eraseblocks with low erase counter to free physical eraseblocks with high
@@ -120,64 +140,9 @@
 #define WL_MAX_FAILURES 32
 
 /**
- * struct ubi_wl_prot_entry - PEB protection entry.
- * @rb_pnum: link in the @wl->prot.pnum RB-tree
- * @rb_aec: link in the @wl->prot.aec RB-tree
- * @abs_ec: the absolute erase counter value when the protection ends
- * @e: the wear-leveling entry of the physical eraseblock under protection
- *
- * When the WL sub-system returns a physical eraseblock, the physical
- * eraseblock is protected from being moved for some "time". For this reason,
- * the physical eraseblock is not directly moved from the @wl->free tree to the
- * @wl->used tree. There is one more tree in between where this physical
- * eraseblock is temporarily stored (@wl->prot).
- *
- * All this protection stuff is needed because:
- *  o we don't want to move physical eraseblocks just after we have given them
- *    to the user; instead, we first want to let users fill them up with data;
- *
- *  o there is a chance that the user will put the physical eraseblock very
- *    soon, so it makes sense not to move it for some time, but wait; this is
- *    especially important in case of "short term" physical eraseblocks.
- *
- * Physical eraseblocks stay protected only for limited time. But the "time" is
- * measured in erase cycles in this case. This is implemented with help of the
- * absolute erase counter (@wl->abs_ec). When it reaches certain value, the
- * physical eraseblocks are moved from the protection trees (@wl->prot.*) to
- * the @wl->used tree.
- *
- * Protected physical eraseblocks are searched by physical eraseblock number
- * (when they are put) and by the absolute erase counter (to check if it is
- * time to move them to the @wl->used tree). So there are actually 2 RB-trees
- * storing the protected physical eraseblocks: @wl->prot.pnum and
- * @wl->prot.aec. They are referred to as the "protection" trees. The
- * first one is indexed by the physical eraseblock number. The second one is
- * indexed by the absolute erase counter. Both trees store
- * &struct ubi_wl_prot_entry objects.
- *
- * Each physical eraseblock has 2 main states: free and used. The former state
- * corresponds to the @wl->free tree. The latter state is split up on several
- * sub-states:
- * o the WL movement is allowed (@wl->used tree);
- * o the WL movement is temporarily prohibited (@wl->prot.pnum and
- * @wl->prot.aec trees);
- * o scrubbing is needed (@wl->scrub tree).
- *
- * Depending on the sub-state, wear-leveling entries of the used physical
- * eraseblocks may be kept in one of those trees.
- */
-struct ubi_wl_prot_entry {
-	struct rb_node rb_pnum;
-	struct rb_node rb_aec;
-	unsigned long long abs_ec;
-	struct ubi_wl_entry *e;
-};
-
-/**
  * struct ubi_work - UBI work description data structure.
  * @list: a link in the list of pending works
  * @func: worker function
- * @priv: private data of the worker function
  * @e: physical eraseblock to erase
  * @torture: if the physical eraseblock has to be tortured
  *
@@ -198,9 +163,11 @@
 static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec);
 static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e,
 				     struct rb_root *root);
+static int paranoid_check_in_pq(struct ubi_device *ubi, struct ubi_wl_entry *e);
 #else
 #define paranoid_check_ec(ubi, pnum, ec) 0
 #define paranoid_check_in_wl_tree(e, root)
+#define paranoid_check_in_pq(ubi, e) 0
 #endif
 
 /**
@@ -220,7 +187,7 @@
 		struct ubi_wl_entry *e1;
 
 		parent = *p;
-		e1 = rb_entry(parent, struct ubi_wl_entry, rb);
+		e1 = rb_entry(parent, struct ubi_wl_entry, u.rb);
 
 		if (e->ec < e1->ec)
 			p = &(*p)->rb_left;
@@ -235,8 +202,8 @@
 		}
 	}
 
-	rb_link_node(&e->rb, parent, p);
-	rb_insert_color(&e->rb, root);
+	rb_link_node(&e->u.rb, parent, p);
+	rb_insert_color(&e->u.rb, root);
 }
 
 /**
@@ -331,7 +298,7 @@
 	while (p) {
 		struct ubi_wl_entry *e1;
 
-		e1 = rb_entry(p, struct ubi_wl_entry, rb);
+		e1 = rb_entry(p, struct ubi_wl_entry, u.rb);
 
 		if (e->pnum == e1->pnum) {
 			ubi_assert(e == e1);
@@ -355,50 +322,24 @@
 }
 
 /**
- * prot_tree_add - add physical eraseblock to protection trees.
+ * prot_queue_add - add physical eraseblock to the protection queue.
  * @ubi: UBI device description object
  * @e: the physical eraseblock to add
- * @pe: protection entry object to use
- * @abs_ec: absolute erase counter value when this physical eraseblock has
- * to be removed from the protection trees.
  *
- * @wl->lock has to be locked.
+ * This function adds @e to the tail of the protection queue @ubi->pq, where
+ * @e will stay for %UBI_PROT_QUEUE_LEN erase operations and will be
+ * temporarily protected from the wear-leveling worker. Note, @wl->lock has to
+ * be locked.
  */
-static void prot_tree_add(struct ubi_device *ubi, struct ubi_wl_entry *e,
-			  struct ubi_wl_prot_entry *pe, int abs_ec)
+static void prot_queue_add(struct ubi_device *ubi, struct ubi_wl_entry *e)
 {
-	struct rb_node **p, *parent = NULL;
-	struct ubi_wl_prot_entry *pe1;
+	int pq_tail = ubi->pq_head - 1;
 
-	pe->e = e;
-	pe->abs_ec = ubi->abs_ec + abs_ec;
-
-	p = &ubi->prot.pnum.rb_node;
-	while (*p) {
-		parent = *p;
-		pe1 = rb_entry(parent, struct ubi_wl_prot_entry, rb_pnum);
-
-		if (e->pnum < pe1->e->pnum)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-	rb_link_node(&pe->rb_pnum, parent, p);
-	rb_insert_color(&pe->rb_pnum, &ubi->prot.pnum);
-
-	p = &ubi->prot.aec.rb_node;
-	parent = NULL;
-	while (*p) {
-		parent = *p;
-		pe1 = rb_entry(parent, struct ubi_wl_prot_entry, rb_aec);
-
-		if (pe->abs_ec < pe1->abs_ec)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-	rb_link_node(&pe->rb_aec, parent, p);
-	rb_insert_color(&pe->rb_aec, &ubi->prot.aec);
+	if (pq_tail < 0)
+		pq_tail = UBI_PROT_QUEUE_LEN - 1;
+	ubi_assert(pq_tail >= 0 && pq_tail < UBI_PROT_QUEUE_LEN);
+	list_add_tail(&e->u.list, &ubi->pq[pq_tail]);
+	dbg_wl("added PEB %d EC %d to the protection queue", e->pnum, e->ec);
 }
 
 /**
@@ -414,14 +355,14 @@
 	struct rb_node *p;
 	struct ubi_wl_entry *e;
 
-	e = rb_entry(rb_first(root), struct ubi_wl_entry, rb);
+	e = rb_entry(rb_first(root), struct ubi_wl_entry, u.rb);
 	max += e->ec;
 
 	p = root->rb_node;
 	while (p) {
 		struct ubi_wl_entry *e1;
 
-		e1 = rb_entry(p, struct ubi_wl_entry, rb);
+		e1 = rb_entry(p, struct ubi_wl_entry, u.rb);
 		if (e1->ec >= max)
 			p = p->rb_left;
 		else {
@@ -443,17 +384,12 @@
  */
 int ubi_wl_get_peb(struct ubi_device *ubi, int dtype)
 {
-	int err, protect, medium_ec;
+	int err, medium_ec;
 	struct ubi_wl_entry *e, *first, *last;
-	struct ubi_wl_prot_entry *pe;
 
 	ubi_assert(dtype == UBI_LONGTERM || dtype == UBI_SHORTTERM ||
 		   dtype == UBI_UNKNOWN);
 
-	pe = kmalloc(sizeof(struct ubi_wl_prot_entry), GFP_NOFS);
-	if (!pe)
-		return -ENOMEM;
-
 retry:
 	spin_lock(&ubi->wl_lock);
 	if (!ubi->free.rb_node) {
@@ -461,16 +397,13 @@
 			ubi_assert(list_empty(&ubi->works));
 			ubi_err("no free eraseblocks");
 			spin_unlock(&ubi->wl_lock);
-			kfree(pe);
 			return -ENOSPC;
 		}
 		spin_unlock(&ubi->wl_lock);
 
 		err = produce_free_peb(ubi);
-		if (err < 0) {
-			kfree(pe);
+		if (err < 0)
 			return err;
-		}
 		goto retry;
 	}
 
@@ -483,7 +416,6 @@
 		 * %WL_FREE_MAX_DIFF.
 		 */
 		e = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
-		protect = LT_PROTECTION;
 		break;
 	case UBI_UNKNOWN:
 		/*
@@ -492,81 +424,63 @@
 		 * eraseblock with erase counter greater or equivalent than the
 		 * lowest erase counter plus %WL_FREE_MAX_DIFF.
 		 */
-		first = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry, rb);
-		last = rb_entry(rb_last(&ubi->free), struct ubi_wl_entry, rb);
+		first = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry,
+					u.rb);
+		last = rb_entry(rb_last(&ubi->free), struct ubi_wl_entry, u.rb);
 
 		if (last->ec - first->ec < WL_FREE_MAX_DIFF)
 			e = rb_entry(ubi->free.rb_node,
-					struct ubi_wl_entry, rb);
+					struct ubi_wl_entry, u.rb);
 		else {
 			medium_ec = (first->ec + WL_FREE_MAX_DIFF)/2;
 			e = find_wl_entry(&ubi->free, medium_ec);
 		}
-		protect = U_PROTECTION;
 		break;
 	case UBI_SHORTTERM:
 		/*
 		 * For short term data we pick a physical eraseblock with the
 		 * lowest erase counter as we expect it will be erased soon.
 		 */
-		e = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry, rb);
-		protect = ST_PROTECTION;
+		e = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry, u.rb);
 		break;
 	default:
-		protect = 0;
-		e = NULL;
 		BUG();
 	}
 
+	paranoid_check_in_wl_tree(e, &ubi->free);
+
 	/*
-	 * Move the physical eraseblock to the protection trees where it will
+	 * Move the physical eraseblock to the protection queue where it will
 	 * be protected from being moved for some time.
 	 */
-	paranoid_check_in_wl_tree(e, &ubi->free);
-	rb_erase(&e->rb, &ubi->free);
-	prot_tree_add(ubi, e, pe, protect);
-
-	dbg_wl("PEB %d EC %d, protection %d", e->pnum, e->ec, protect);
+	rb_erase(&e->u.rb, &ubi->free);
+	dbg_wl("PEB %d EC %d", e->pnum, e->ec);
+	prot_queue_add(ubi, e);
 	spin_unlock(&ubi->wl_lock);
-
 	return e->pnum;
 }
 
 /**
- * prot_tree_del - remove a physical eraseblock from the protection trees
+ * prot_queue_del - remove a physical eraseblock from the protection queue.
  * @ubi: UBI device description object
  * @pnum: the physical eraseblock to remove
  *
- * This function returns PEB @pnum from the protection trees and returns zero
- * in case of success and %-ENODEV if the PEB was not found in the protection
- * trees.
+ * This function deletes PEB @pnum from the protection queue and returns zero
+ * in case of success and %-ENODEV if the PEB was not found.
  */
-static int prot_tree_del(struct ubi_device *ubi, int pnum)
+static int prot_queue_del(struct ubi_device *ubi, int pnum)
 {
-	struct rb_node *p;
-	struct ubi_wl_prot_entry *pe = NULL;
+	struct ubi_wl_entry *e;
 
-	p = ubi->prot.pnum.rb_node;
-	while (p) {
+	e = ubi->lookuptbl[pnum];
+	if (!e)
+		return -ENODEV;
 
-		pe = rb_entry(p, struct ubi_wl_prot_entry, rb_pnum);
+	if (paranoid_check_in_pq(ubi, e))
+		return -ENODEV;
 
-		if (pnum == pe->e->pnum)
-			goto found;
-
-		if (pnum < pe->e->pnum)
-			p = p->rb_left;
-		else
-			p = p->rb_right;
-	}
-
-	return -ENODEV;
-
-found:
-	ubi_assert(pe->e->pnum == pnum);
-	rb_erase(&pe->rb_aec, &ubi->prot.aec);
-	rb_erase(&pe->rb_pnum, &ubi->prot.pnum);
-	kfree(pe);
+	list_del(&e->u.list);
+	dbg_wl("deleted PEB %d from the protection queue", e->pnum);
 	return 0;
 }
 
@@ -632,47 +546,47 @@
 }
 
 /**
- * check_protection_over - check if it is time to stop protecting some PEBs.
+ * serve_prot_queue - check if it is time to stop protecting PEBs.
  * @ubi: UBI device description object
  *
- * This function is called after each erase operation, when the absolute erase
- * counter is incremented, to check if some physical eraseblock  have not to be
- * protected any longer. These physical eraseblocks are moved from the
- * protection trees to the used tree.
+ * This function is called after each erase operation and removes PEBs from the
+ * tail of the protection queue. These PEBs have been protected for long enough
+ * and should be moved to the used tree.
  */
-static void check_protection_over(struct ubi_device *ubi)
+static void serve_prot_queue(struct ubi_device *ubi)
 {
-	struct ubi_wl_prot_entry *pe;
+	struct ubi_wl_entry *e, *tmp;
+	int count;
 
 	/*
 	 * There may be several protected physical eraseblock to remove,
 	 * process them all.
 	 */
-	while (1) {
-		spin_lock(&ubi->wl_lock);
-		if (!ubi->prot.aec.rb_node) {
+repeat:
+	count = 0;
+	spin_lock(&ubi->wl_lock);
+	list_for_each_entry_safe(e, tmp, &ubi->pq[ubi->pq_head], u.list) {
+		dbg_wl("PEB %d EC %d protection over, move to used tree",
+			e->pnum, e->ec);
+
+		list_del(&e->u.list);
+		wl_tree_add(e, &ubi->used);
+		if (count++ > 32) {
+			/*
+			 * Let's be nice and avoid holding the spinlock for
+			 * too long.
+			 */
 			spin_unlock(&ubi->wl_lock);
-			break;
+			cond_resched();
+			goto repeat;
 		}
-
-		pe = rb_entry(rb_first(&ubi->prot.aec),
-			      struct ubi_wl_prot_entry, rb_aec);
-
-		if (pe->abs_ec > ubi->abs_ec) {
-			spin_unlock(&ubi->wl_lock);
-			break;
-		}
-
-		dbg_wl("PEB %d protection over, abs_ec %llu, PEB abs_ec %llu",
-		       pe->e->pnum, ubi->abs_ec, pe->abs_ec);
-		rb_erase(&pe->rb_aec, &ubi->prot.aec);
-		rb_erase(&pe->rb_pnum, &ubi->prot.pnum);
-		wl_tree_add(pe->e, &ubi->used);
-		spin_unlock(&ubi->wl_lock);
-
-		kfree(pe);
-		cond_resched();
 	}
+
+	ubi->pq_head += 1;
+	if (ubi->pq_head == UBI_PROT_QUEUE_LEN)
+		ubi->pq_head = 0;
+	ubi_assert(ubi->pq_head >= 0 && ubi->pq_head < UBI_PROT_QUEUE_LEN);
+	spin_unlock(&ubi->wl_lock);
 }
 
 /**
@@ -680,8 +594,8 @@
  * @ubi: UBI device description object
  * @wrk: the work to schedule
  *
- * This function enqueues a work defined by @wrk to the tail of the pending
- * works list.
+ * This function adds a work defined by @wrk to the tail of the pending works
+ * list.
  */
 static void schedule_ubi_work(struct ubi_device *ubi, struct ubi_work *wrk)
 {
@@ -739,13 +653,11 @@
 static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 				int cancel)
 {
-	int err, put = 0, scrubbing = 0, protect = 0;
-	struct ubi_wl_prot_entry *uninitialized_var(pe);
+	int err, scrubbing = 0, torture = 0;
 	struct ubi_wl_entry *e1, *e2;
 	struct ubi_vid_hdr *vid_hdr;
 
 	kfree(wrk);
-
 	if (cancel)
 		return 0;
 
@@ -781,7 +693,7 @@
 		 * highly worn-out free physical eraseblock. If the erase
 		 * counters differ much enough, start wear-leveling.
 		 */
-		e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, rb);
+		e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, u.rb);
 		e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
 
 		if (!(e2->ec - e1->ec >= UBI_WL_THRESHOLD)) {
@@ -790,21 +702,21 @@
 			goto out_cancel;
 		}
 		paranoid_check_in_wl_tree(e1, &ubi->used);
-		rb_erase(&e1->rb, &ubi->used);
+		rb_erase(&e1->u.rb, &ubi->used);
 		dbg_wl("move PEB %d EC %d to PEB %d EC %d",
 		       e1->pnum, e1->ec, e2->pnum, e2->ec);
 	} else {
 		/* Perform scrubbing */
 		scrubbing = 1;
-		e1 = rb_entry(rb_first(&ubi->scrub), struct ubi_wl_entry, rb);
+		e1 = rb_entry(rb_first(&ubi->scrub), struct ubi_wl_entry, u.rb);
 		e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
 		paranoid_check_in_wl_tree(e1, &ubi->scrub);
-		rb_erase(&e1->rb, &ubi->scrub);
+		rb_erase(&e1->u.rb, &ubi->scrub);
 		dbg_wl("scrub PEB %d to PEB %d", e1->pnum, e2->pnum);
 	}
 
 	paranoid_check_in_wl_tree(e2, &ubi->free);
-	rb_erase(&e2->rb, &ubi->free);
+	rb_erase(&e2->u.rb, &ubi->free);
 	ubi->move_from = e1;
 	ubi->move_to = e2;
 	spin_unlock(&ubi->wl_lock);
@@ -844,46 +756,67 @@
 
 	err = ubi_eba_copy_leb(ubi, e1->pnum, e2->pnum, vid_hdr);
 	if (err) {
-
+		if (err == -EAGAIN)
+			goto out_not_moved;
 		if (err < 0)
 			goto out_error;
-		if (err == 1)
+		if (err == 2) {
+			/* Target PEB write error, torture it */
+			torture = 1;
 			goto out_not_moved;
-
-		/*
-		 * For some reason the LEB was not moved - it might be because
-		 * the volume is being deleted. We should prevent this PEB from
-		 * being selected for wear-levelling movement for some "time",
-		 * so put it to the protection tree.
-		 */
-
-		dbg_wl("cancelled moving PEB %d", e1->pnum);
-		pe = kmalloc(sizeof(struct ubi_wl_prot_entry), GFP_NOFS);
-		if (!pe) {
-			err = -ENOMEM;
-			goto out_error;
 		}
 
-		protect = 1;
+		/*
+		 * The LEB has not been moved because the volume is being
+		 * deleted or the PEB has been put meanwhile. We should prevent
+		 * this PEB from being selected for wear-leveling movement
+		 * again, so put it to the protection queue.
+		 */
+
+		dbg_wl("canceled moving PEB %d", e1->pnum);
+		ubi_assert(err == 1);
+
+		ubi_free_vid_hdr(ubi, vid_hdr);
+		vid_hdr = NULL;
+
+		spin_lock(&ubi->wl_lock);
+		prot_queue_add(ubi, e1);
+		ubi_assert(!ubi->move_to_put);
+		ubi->move_from = ubi->move_to = NULL;
+		ubi->wl_scheduled = 0;
+		spin_unlock(&ubi->wl_lock);
+
+		e1 = NULL;
+		err = schedule_erase(ubi, e2, 0);
+		if (err)
+			goto out_error;
+		mutex_unlock(&ubi->move_mutex);
+		return 0;
 	}
 
+	/* The PEB has been successfully moved */
 	ubi_free_vid_hdr(ubi, vid_hdr);
-	if (scrubbing && !protect)
+	vid_hdr = NULL;
+	if (scrubbing)
 		ubi_msg("scrubbed PEB %d, data moved to PEB %d",
 			e1->pnum, e2->pnum);
 
 	spin_lock(&ubi->wl_lock);
-	if (protect)
-		prot_tree_add(ubi, e1, pe, protect);
-	if (!ubi->move_to_put)
+	if (!ubi->move_to_put) {
 		wl_tree_add(e2, &ubi->used);
-	else
-		put = 1;
+		e2 = NULL;
+	}
 	ubi->move_from = ubi->move_to = NULL;
 	ubi->move_to_put = ubi->wl_scheduled = 0;
 	spin_unlock(&ubi->wl_lock);
 
-	if (put) {
+	err = schedule_erase(ubi, e1, 0);
+	if (err) {
+		e1 = NULL;
+		goto out_error;
+	}
+
+	if (e2) {
 		/*
 		 * Well, the target PEB was put meanwhile, schedule it for
 		 * erasure.
@@ -894,13 +827,6 @@
 			goto out_error;
 	}
 
-	if (!protect) {
-		err = schedule_erase(ubi, e1, 0);
-		if (err)
-			goto out_error;
-	}
-
-
 	dbg_wl("done");
 	mutex_unlock(&ubi->move_mutex);
 	return 0;
@@ -908,20 +834,24 @@
 	/*
 	 * For some reasons the LEB was not moved, might be an error, might be
 	 * something else. @e1 was not changed, so return it back. @e2 might
-	 * be changed, schedule it for erasure.
+	 * have been changed, schedule it for erasure.
 	 */
 out_not_moved:
+	dbg_wl("canceled moving PEB %d", e1->pnum);
 	ubi_free_vid_hdr(ubi, vid_hdr);
+	vid_hdr = NULL;
 	spin_lock(&ubi->wl_lock);
 	if (scrubbing)
 		wl_tree_add(e1, &ubi->scrub);
 	else
 		wl_tree_add(e1, &ubi->used);
+	ubi_assert(!ubi->move_to_put);
 	ubi->move_from = ubi->move_to = NULL;
-	ubi->move_to_put = ubi->wl_scheduled = 0;
+	ubi->wl_scheduled = 0;
 	spin_unlock(&ubi->wl_lock);
 
-	err = schedule_erase(ubi, e2, 0);
+	e1 = NULL;
+	err = schedule_erase(ubi, e2, torture);
 	if (err)
 		goto out_error;
 
@@ -938,8 +868,10 @@
 	ubi->move_to_put = ubi->wl_scheduled = 0;
 	spin_unlock(&ubi->wl_lock);
 
-	kmem_cache_free(ubi_wl_entry_slab, e1);
-	kmem_cache_free(ubi_wl_entry_slab, e2);
+	if (e1)
+		kmem_cache_free(ubi_wl_entry_slab, e1);
+	if (e2)
+		kmem_cache_free(ubi_wl_entry_slab, e2);
 	ubi_ro_mode(ubi);
 
 	mutex_unlock(&ubi->move_mutex);
@@ -988,7 +920,7 @@
 		 * erase counter of free physical eraseblocks is greater then
 		 * %UBI_WL_THRESHOLD.
 		 */
-		e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, rb);
+		e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, u.rb);
 		e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
 
 		if (!(e2->ec - e1->ec >= UBI_WL_THRESHOLD))
@@ -1050,7 +982,6 @@
 		kfree(wl_wrk);
 
 		spin_lock(&ubi->wl_lock);
-		ubi->abs_ec += 1;
 		wl_tree_add(e, &ubi->free);
 		spin_unlock(&ubi->wl_lock);
 
@@ -1058,7 +989,7 @@
 		 * One more erase operation has happened, take care about
 		 * protected physical eraseblocks.
 		 */
-		check_protection_over(ubi);
+		serve_prot_queue(ubi);
 
 		/* And take care about wear-leveling */
 		err = ensure_wear_leveling(ubi);
@@ -1190,12 +1121,12 @@
 	} else {
 		if (in_wl_tree(e, &ubi->used)) {
 			paranoid_check_in_wl_tree(e, &ubi->used);
-			rb_erase(&e->rb, &ubi->used);
+			rb_erase(&e->u.rb, &ubi->used);
 		} else if (in_wl_tree(e, &ubi->scrub)) {
 			paranoid_check_in_wl_tree(e, &ubi->scrub);
-			rb_erase(&e->rb, &ubi->scrub);
+			rb_erase(&e->u.rb, &ubi->scrub);
 		} else {
-			err = prot_tree_del(ubi, e->pnum);
+			err = prot_queue_del(ubi, e->pnum);
 			if (err) {
 				ubi_err("PEB %d not found", pnum);
 				ubi_ro_mode(ubi);
@@ -1255,11 +1186,11 @@
 
 	if (in_wl_tree(e, &ubi->used)) {
 		paranoid_check_in_wl_tree(e, &ubi->used);
-		rb_erase(&e->rb, &ubi->used);
+		rb_erase(&e->u.rb, &ubi->used);
 	} else {
 		int err;
 
-		err = prot_tree_del(ubi, e->pnum);
+		err = prot_queue_del(ubi, e->pnum);
 		if (err) {
 			ubi_err("PEB %d not found", pnum);
 			ubi_ro_mode(ubi);
@@ -1290,7 +1221,7 @@
 	int err;
 
 	/*
-	 * Erase while the pending works queue is not empty, but not more then
+	 * Erase while the pending works queue is not empty, but not more than
 	 * the number of currently pending works.
 	 */
 	dbg_wl("flush (%d pending works)", ubi->works_count);
@@ -1308,7 +1239,7 @@
 	up_write(&ubi->work_sem);
 
 	/*
-	 * And in case last was the WL worker and it cancelled the LEB
+	 * And in case last was the WL worker and it canceled the LEB
 	 * movement, flush again.
 	 */
 	while (ubi->works_count) {
@@ -1337,11 +1268,11 @@
 		else if (rb->rb_right)
 			rb = rb->rb_right;
 		else {
-			e = rb_entry(rb, struct ubi_wl_entry, rb);
+			e = rb_entry(rb, struct ubi_wl_entry, u.rb);
 
 			rb = rb_parent(rb);
 			if (rb) {
-				if (rb->rb_left == &e->rb)
+				if (rb->rb_left == &e->u.rb)
 					rb->rb_left = NULL;
 				else
 					rb->rb_right = NULL;
@@ -1436,15 +1367,13 @@
  */
 int ubi_wl_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si)
 {
-	int err;
+	int err, i;
 	struct rb_node *rb1, *rb2;
 	struct ubi_scan_volume *sv;
 	struct ubi_scan_leb *seb, *tmp;
 	struct ubi_wl_entry *e;
 
-
 	ubi->used = ubi->free = ubi->scrub = RB_ROOT;
-	ubi->prot.pnum = ubi->prot.aec = RB_ROOT;
 	spin_lock_init(&ubi->wl_lock);
 	mutex_init(&ubi->move_mutex);
 	init_rwsem(&ubi->work_sem);
@@ -1458,6 +1387,10 @@
 	if (!ubi->lookuptbl)
 		return err;
 
+	for (i = 0; i < UBI_PROT_QUEUE_LEN; i++)
+		INIT_LIST_HEAD(&ubi->pq[i]);
+	ubi->pq_head = 0;
+
 	list_for_each_entry_safe(seb, tmp, &si->erase, u.list) {
 		cond_resched();
 
@@ -1552,33 +1485,18 @@
 }
 
 /**
- * protection_trees_destroy - destroy the protection RB-trees.
+ * protection_queue_destroy - destroy the protection queue.
  * @ubi: UBI device description object
  */
-static void protection_trees_destroy(struct ubi_device *ubi)
+static void protection_queue_destroy(struct ubi_device *ubi)
 {
-	struct rb_node *rb;
-	struct ubi_wl_prot_entry *pe;
+	int i;
+	struct ubi_wl_entry *e, *tmp;
 
-	rb = ubi->prot.aec.rb_node;
-	while (rb) {
-		if (rb->rb_left)
-			rb = rb->rb_left;
-		else if (rb->rb_right)
-			rb = rb->rb_right;
-		else {
-			pe = rb_entry(rb, struct ubi_wl_prot_entry, rb_aec);
-
-			rb = rb_parent(rb);
-			if (rb) {
-				if (rb->rb_left == &pe->rb_aec)
-					rb->rb_left = NULL;
-				else
-					rb->rb_right = NULL;
-			}
-
-			kmem_cache_free(ubi_wl_entry_slab, pe->e);
-			kfree(pe);
+	for (i = 0; i < UBI_PROT_QUEUE_LEN; ++i) {
+		list_for_each_entry_safe(e, tmp, &ubi->pq[i], u.list) {
+			list_del(&e->u.list);
+			kmem_cache_free(ubi_wl_entry_slab, e);
 		}
 	}
 }
@@ -1591,7 +1509,7 @@
 {
 	dbg_wl("close the WL sub-system");
 	cancel_pending(ubi);
-	protection_trees_destroy(ubi);
+	protection_queue_destroy(ubi);
 	tree_destroy(&ubi->used);
 	tree_destroy(&ubi->free);
 	tree_destroy(&ubi->scrub);
@@ -1661,4 +1579,27 @@
 	return 1;
 }
 
+/**
+ * paranoid_check_in_pq - check if wear-leveling entry is in the protection
+ *                        queue.
+ * @ubi: UBI device description object
+ * @e: the wear-leveling entry to check
+ *
+ * This function returns zero if @e is in @ubi->pq and %1 if it is not.
+ */
+static int paranoid_check_in_pq(struct ubi_device *ubi, struct ubi_wl_entry *e)
+{
+	struct ubi_wl_entry *p;
+	int i;
+
+	for (i = 0; i < UBI_PROT_QUEUE_LEN; ++i)
+		list_for_each_entry(p, &ubi->pq[i], u.list)
+			if (p == e)
+				return 0;
+
+	ubi_err("paranoid check failed for PEB %d, EC %d, Protect queue",
+		e->pnum, e->ec);
+	ubi_dbg_dump_stack();
+	return 1;
+}
 #endif /* CONFIG_MTD_UBI_DEBUG_PARANOID */
diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c
index 9f7896a..c4918b8 100644
--- a/drivers/net/usb/hso.c
+++ b/drivers/net/usb/hso.c
@@ -3,6 +3,8 @@
  * Driver for Option High Speed Mobile Devices.
  *
  *  Copyright (C) 2008 Option International
+ *                     Filip Aben <f.aben@option.com>
+ *                     Denis Joseph Barrow <d.barow@option.com>
  *  Copyright (C) 2007 Andrew Bird (Sphere Systems Ltd)
  *  			<ajb@spheresystems.co.uk>
  *  Copyright (C) 2008 Greg Kroah-Hartman <gregkh@suse.de>
@@ -39,8 +41,11 @@
  *		port is opened, as this have a huge impact on the network port
  *		throughput.
  *
- * Interface 2:	Standard modem interface - circuit switched interface, should
- *		not be used.
+ * Interface 2:	Standard modem interface - circuit switched interface, this
+ *		can be used to make a standard ppp connection however it
+ *              should not be used in conjunction with the IP network interface
+ *              enabled for USB performance reasons i.e. if using this set
+ *              ideally disable_net=1.
  *
  *****************************************************************************/
 
@@ -63,6 +68,8 @@
 #include <linux/usb/cdc.h>
 #include <net/arp.h>
 #include <asm/byteorder.h>
+#include <linux/serial_core.h>
+#include <linux/serial.h>
 
 
 #define DRIVER_VERSION			"1.2"
@@ -182,6 +189,41 @@
 	RX_PENDING
 };
 
+#define BM_REQUEST_TYPE (0xa1)
+#define B_NOTIFICATION  (0x20)
+#define W_VALUE         (0x0)
+#define W_INDEX         (0x2)
+#define W_LENGTH        (0x2)
+
+#define B_OVERRUN       (0x1<<6)
+#define B_PARITY        (0x1<<5)
+#define B_FRAMING       (0x1<<4)
+#define B_RING_SIGNAL   (0x1<<3)
+#define B_BREAK         (0x1<<2)
+#define B_TX_CARRIER    (0x1<<1)
+#define B_RX_CARRIER    (0x1<<0)
+
+struct hso_serial_state_notification {
+	u8 bmRequestType;
+	u8 bNotification;
+	u16 wValue;
+	u16 wIndex;
+	u16 wLength;
+	u16 UART_state_bitmap;
+} __attribute__((packed));
+
+struct hso_tiocmget {
+	struct mutex mutex;
+	wait_queue_head_t waitq;
+	int    intr_completed;
+	struct usb_endpoint_descriptor *endp;
+	struct urb *urb;
+	struct hso_serial_state_notification serial_state_notification;
+	u16    prev_UART_state_bitmap;
+	struct uart_icount icount;
+};
+
+
 struct hso_serial {
 	struct hso_device *parent;
 	int magic;
@@ -219,6 +261,7 @@
 	spinlock_t serial_lock;
 
 	int (*write_data) (struct hso_serial *serial);
+	struct hso_tiocmget  *tiocmget;
 	/* Hacks required to get flow control
 	 * working on the serial receive buffers
 	 * so as not to drop characters on the floor.
@@ -305,7 +348,7 @@
 static void async_put_intf(struct work_struct *data);
 static int hso_put_activity(struct hso_device *hso_dev);
 static int hso_get_activity(struct hso_device *hso_dev);
-
+static void tiocmget_intr_callback(struct urb *urb);
 /*****************************************************************************/
 /* Helping functions                                                         */
 /*****************************************************************************/
@@ -362,8 +405,6 @@
 static struct hso_device *serial_table[HSO_SERIAL_TTY_MINORS];
 static struct hso_device *network_table[HSO_MAX_NET_DEVICES];
 static spinlock_t serial_table_lock;
-static struct ktermios *hso_serial_termios[HSO_SERIAL_TTY_MINORS];
-static struct ktermios *hso_serial_termios_locked[HSO_SERIAL_TTY_MINORS];
 
 static const s32 default_port_spec[] = {
 	HSO_INTF_MUX | HSO_PORT_NETWORK,
@@ -1009,23 +1050,11 @@
 
 /* Serial driver functions */
 
-static void _hso_serial_set_termios(struct tty_struct *tty,
-				    struct ktermios *old)
+static void hso_init_termios(struct ktermios *termios)
 {
-	struct hso_serial *serial = get_serial_by_tty(tty);
-	struct ktermios *termios;
-
-	if ((!tty) || (!tty->termios) || (!serial)) {
-		printk(KERN_ERR "%s: no tty structures", __func__);
-		return;
-	}
-
-	D4("port %d", serial->minor);
-
 	/*
 	 * The default requirements for this device are:
 	 */
-	termios = tty->termios;
 	termios->c_iflag &=
 		~(IGNBRK	/* disable ignore break */
 		| BRKINT	/* disable break causes interrupt */
@@ -1057,15 +1086,38 @@
 	termios->c_cflag |= CS8;	/* character size 8 bits */
 
 	/* baud rate 115200 */
-	tty_encode_baud_rate(serial->tty, 115200, 115200);
+	tty_termios_encode_baud_rate(termios, 115200, 115200);
+}
+
+static void _hso_serial_set_termios(struct tty_struct *tty,
+				    struct ktermios *old)
+{
+	struct hso_serial *serial = get_serial_by_tty(tty);
+	struct ktermios *termios;
+
+	if (!serial) {
+		printk(KERN_ERR "%s: no tty structures", __func__);
+		return;
+	}
+
+	D4("port %d", serial->minor);
 
 	/*
-	 * Force low_latency on; otherwise the pushes are scheduled;
-	 * this is bad as it opens up the possibility of dropping bytes
-	 * on the floor.  We don't want to drop bytes on the floor. :)
+	 *	Fix up unsupported bits
 	 */
-	serial->tty->low_latency = 1;
-	return;
+	termios = tty->termios;
+	termios->c_iflag &= ~IXON; /* disable enable XON/XOFF flow control */
+
+	termios->c_cflag &=
+		~(CSIZE		/* no size */
+		| PARENB	/* disable parity bit */
+		| CBAUD		/* clear current baud rate */
+		| CBAUDEX);	/* clear current buad rate */
+
+	termios->c_cflag |= CS8;	/* character size 8 bits */
+
+	/* baud rate 115200 */
+	tty_encode_baud_rate(tty, 115200, 115200);
 }
 
 static void hso_resubmit_rx_bulk_urb(struct hso_serial *serial, struct urb *urb)
@@ -1228,6 +1280,7 @@
 
 	/* sanity check */
 	if (serial == NULL || serial->magic != HSO_SERIAL_MAGIC) {
+		WARN_ON(1);
 		tty->driver_data = NULL;
 		D1("Failed to open port");
 		return -ENODEV;
@@ -1242,8 +1295,10 @@
 	kref_get(&serial->parent->ref);
 
 	/* setup */
+	spin_lock_irq(&serial->serial_lock);
 	tty->driver_data = serial;
-	serial->tty = tty;
+	serial->tty = tty_kref_get(tty);
+	spin_unlock_irq(&serial->serial_lock);
 
 	/* check for port already opened, if not set the termios */
 	serial->open_count++;
@@ -1285,6 +1340,10 @@
 
 	D1("Closing serial port");
 
+	/* Open failed, no close cleanup required */
+	if (serial == NULL)
+		return;
+
 	mutex_lock(&serial->parent->mutex);
 	usb_gone = serial->parent->usb_gone;
 
@@ -1297,10 +1356,13 @@
 	kref_put(&serial->parent->ref, hso_serial_ref_free);
 	if (serial->open_count <= 0) {
 		serial->open_count = 0;
-		if (serial->tty) {
+		spin_lock_irq(&serial->serial_lock);
+		if (serial->tty == tty) {
 			serial->tty->driver_data = NULL;
 			serial->tty = NULL;
+			tty_kref_put(tty);
 		}
+		spin_unlock_irq(&serial->serial_lock);
 		if (!usb_gone)
 			hso_stop_serial_device(serial->parent);
 		tasklet_kill(&serial->unthrottle_tasklet);
@@ -1400,25 +1462,217 @@
 
 	return chars;
 }
+int tiocmget_submit_urb(struct hso_serial *serial,
+			struct hso_tiocmget  *tiocmget,
+			struct usb_device *usb)
+{
+	int result;
+
+	if (serial->parent->usb_gone)
+		return -ENODEV;
+	usb_fill_int_urb(tiocmget->urb, usb,
+			 usb_rcvintpipe(usb,
+					tiocmget->endp->
+					bEndpointAddress & 0x7F),
+			 &tiocmget->serial_state_notification,
+			 sizeof(struct hso_serial_state_notification),
+			 tiocmget_intr_callback, serial,
+			 tiocmget->endp->bInterval);
+	result = usb_submit_urb(tiocmget->urb, GFP_ATOMIC);
+	if (result) {
+		dev_warn(&usb->dev, "%s usb_submit_urb failed %d\n", __func__,
+			 result);
+	}
+	return result;
+
+}
+
+static void tiocmget_intr_callback(struct urb *urb)
+{
+	struct hso_serial *serial = urb->context;
+	struct hso_tiocmget *tiocmget;
+	int status = urb->status;
+	u16 UART_state_bitmap, prev_UART_state_bitmap;
+	struct uart_icount *icount;
+	struct hso_serial_state_notification *serial_state_notification;
+	struct usb_device *usb;
+
+	/* Sanity checks */
+	if (!serial)
+		return;
+	if (status) {
+		log_usb_status(status, __func__);
+		return;
+	}
+	tiocmget = serial->tiocmget;
+	if (!tiocmget)
+		return;
+	usb = serial->parent->usb;
+	serial_state_notification = &tiocmget->serial_state_notification;
+	if (serial_state_notification->bmRequestType != BM_REQUEST_TYPE ||
+	    serial_state_notification->bNotification != B_NOTIFICATION ||
+	    le16_to_cpu(serial_state_notification->wValue) != W_VALUE ||
+	    le16_to_cpu(serial_state_notification->wIndex) != W_INDEX ||
+	    le16_to_cpu(serial_state_notification->wLength) != W_LENGTH) {
+		dev_warn(&usb->dev,
+			 "hso received invalid serial state notification\n");
+		DUMP(serial_state_notification,
+		     sizeof(hso_serial_state_notifation))
+	} else {
+
+		UART_state_bitmap = le16_to_cpu(serial_state_notification->
+						UART_state_bitmap);
+		prev_UART_state_bitmap = tiocmget->prev_UART_state_bitmap;
+		icount = &tiocmget->icount;
+		spin_lock(&serial->serial_lock);
+		if ((UART_state_bitmap & B_OVERRUN) !=
+		   (prev_UART_state_bitmap & B_OVERRUN))
+			icount->parity++;
+		if ((UART_state_bitmap & B_PARITY) !=
+		   (prev_UART_state_bitmap & B_PARITY))
+			icount->parity++;
+		if ((UART_state_bitmap & B_FRAMING) !=
+		   (prev_UART_state_bitmap & B_FRAMING))
+			icount->frame++;
+		if ((UART_state_bitmap & B_RING_SIGNAL) &&
+		   !(prev_UART_state_bitmap & B_RING_SIGNAL))
+			icount->rng++;
+		if ((UART_state_bitmap & B_BREAK) !=
+		   (prev_UART_state_bitmap & B_BREAK))
+			icount->brk++;
+		if ((UART_state_bitmap & B_TX_CARRIER) !=
+		   (prev_UART_state_bitmap & B_TX_CARRIER))
+			icount->dsr++;
+		if ((UART_state_bitmap & B_RX_CARRIER) !=
+		   (prev_UART_state_bitmap & B_RX_CARRIER))
+			icount->dcd++;
+		tiocmget->prev_UART_state_bitmap = UART_state_bitmap;
+		spin_unlock(&serial->serial_lock);
+		tiocmget->intr_completed = 1;
+		wake_up_interruptible(&tiocmget->waitq);
+	}
+	memset(serial_state_notification, 0,
+	       sizeof(struct hso_serial_state_notification));
+	tiocmget_submit_urb(serial,
+			    tiocmget,
+			    serial->parent->usb);
+}
+
+/*
+ * next few functions largely stolen from drivers/serial/serial_core.c
+ */
+/* Wait for any of the 4 modem inputs (DCD,RI,DSR,CTS) to change
+ * - mask passed in arg for lines of interest
+ *   (use |'ed TIOCM_RNG/DSR/CD/CTS for masking)
+ * Caller should use TIOCGICOUNT to see which one it was
+ */
+static int
+hso_wait_modem_status(struct hso_serial *serial, unsigned long arg)
+{
+	DECLARE_WAITQUEUE(wait, current);
+	struct uart_icount cprev, cnow;
+	struct hso_tiocmget  *tiocmget;
+	int ret;
+
+	tiocmget = serial->tiocmget;
+	if (!tiocmget)
+		return -ENOENT;
+	/*
+	 * note the counters on entry
+	 */
+	spin_lock_irq(&serial->serial_lock);
+	memcpy(&cprev, &tiocmget->icount, sizeof(struct uart_icount));
+	spin_unlock_irq(&serial->serial_lock);
+	add_wait_queue(&tiocmget->waitq, &wait);
+	for (;;) {
+		spin_lock_irq(&serial->serial_lock);
+		memcpy(&cnow, &tiocmget->icount, sizeof(struct uart_icount));
+		spin_unlock_irq(&serial->serial_lock);
+		set_current_state(TASK_INTERRUPTIBLE);
+		if (((arg & TIOCM_RNG) && (cnow.rng != cprev.rng)) ||
+		    ((arg & TIOCM_DSR) && (cnow.dsr != cprev.dsr)) ||
+		    ((arg & TIOCM_CD)  && (cnow.dcd != cprev.dcd))) {
+			ret = 0;
+			break;
+		}
+		schedule();
+		/* see if a signal did it */
+		if (signal_pending(current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+		cprev = cnow;
+	}
+	current->state = TASK_RUNNING;
+	remove_wait_queue(&tiocmget->waitq, &wait);
+
+	return ret;
+}
+
+/*
+ * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
+ * Return: write counters to the user passed counter struct
+ * NB: both 1->0 and 0->1 transitions are counted except for
+ *     RI where only 0->1 is counted.
+ */
+static int hso_get_count(struct hso_serial *serial,
+			  struct serial_icounter_struct __user *icnt)
+{
+	struct serial_icounter_struct icount;
+	struct uart_icount cnow;
+	struct hso_tiocmget  *tiocmget = serial->tiocmget;
+
+	if (!tiocmget)
+		 return -ENOENT;
+	spin_lock_irq(&serial->serial_lock);
+	memcpy(&cnow, &tiocmget->icount, sizeof(struct uart_icount));
+	spin_unlock_irq(&serial->serial_lock);
+
+	icount.cts         = cnow.cts;
+	icount.dsr         = cnow.dsr;
+	icount.rng         = cnow.rng;
+	icount.dcd         = cnow.dcd;
+	icount.rx          = cnow.rx;
+	icount.tx          = cnow.tx;
+	icount.frame       = cnow.frame;
+	icount.overrun     = cnow.overrun;
+	icount.parity      = cnow.parity;
+	icount.brk         = cnow.brk;
+	icount.buf_overrun = cnow.buf_overrun;
+
+	return copy_to_user(icnt, &icount, sizeof(icount)) ? -EFAULT : 0;
+}
+
 
 static int hso_serial_tiocmget(struct tty_struct *tty, struct file *file)
 {
-	unsigned int value;
+	int retval;
 	struct hso_serial *serial = get_serial_by_tty(tty);
-	unsigned long flags;
+	struct hso_tiocmget  *tiocmget;
+	u16 UART_state_bitmap;
 
 	/* sanity check */
 	if (!serial) {
 		D1("no tty structures");
 		return -EINVAL;
 	}
-
-	spin_lock_irqsave(&serial->serial_lock, flags);
-	value = ((serial->rts_state) ? TIOCM_RTS : 0) |
+	spin_lock_irq(&serial->serial_lock);
+	retval = ((serial->rts_state) ? TIOCM_RTS : 0) |
 	    ((serial->dtr_state) ? TIOCM_DTR : 0);
-	spin_unlock_irqrestore(&serial->serial_lock, flags);
+	tiocmget = serial->tiocmget;
+	if (tiocmget) {
 
-	return value;
+		UART_state_bitmap = le16_to_cpu(
+			tiocmget->prev_UART_state_bitmap);
+		if (UART_state_bitmap & B_RING_SIGNAL)
+			retval |=  TIOCM_RNG;
+		if (UART_state_bitmap & B_RX_CARRIER)
+			retval |=  TIOCM_CD;
+		if (UART_state_bitmap & B_TX_CARRIER)
+			retval |=  TIOCM_DSR;
+	}
+	spin_unlock_irq(&serial->serial_lock);
+	return retval;
 }
 
 static int hso_serial_tiocmset(struct tty_struct *tty, struct file *file,
@@ -1460,6 +1714,32 @@
 			       USB_CTRL_SET_TIMEOUT);
 }
 
+static int hso_serial_ioctl(struct tty_struct *tty, struct file *file,
+			    unsigned int cmd, unsigned long arg)
+{
+	struct hso_serial *serial =  get_serial_by_tty(tty);
+	void __user *uarg = (void __user *)arg;
+	int ret = 0;
+	D4("IOCTL cmd: %d, arg: %ld", cmd, arg);
+
+	if (!serial)
+		return -ENODEV;
+	switch (cmd) {
+	case TIOCMIWAIT:
+		ret = hso_wait_modem_status(serial, arg);
+		break;
+
+	case TIOCGICOUNT:
+		ret = hso_get_count(serial, uarg);
+		break;
+	default:
+		ret = -ENOIOCTLCMD;
+		break;
+	}
+	return ret;
+}
+
+
 /* starts a transmit */
 static void hso_kick_transmit(struct hso_serial *serial)
 {
@@ -1653,6 +1933,7 @@
 {
 	struct hso_serial *serial = urb->context;
 	int status = urb->status;
+	struct tty_struct *tty;
 
 	/* sanity check */
 	if (!serial) {
@@ -1662,14 +1943,18 @@
 
 	spin_lock(&serial->serial_lock);
 	serial->tx_urb_used = 0;
+	tty = tty_kref_get(serial->tty);
 	spin_unlock(&serial->serial_lock);
 	if (status) {
 		log_usb_status(status, __func__);
+		tty_kref_put(tty);
 		return;
 	}
 	hso_put_activity(serial->parent);
-	if (serial->tty)
-		tty_wakeup(serial->tty);
+	if (tty) {
+		tty_wakeup(tty);
+		tty_kref_put(tty);
+	}
 	hso_kick_transmit(serial);
 
 	D1(" ");
@@ -1706,6 +1991,7 @@
 	struct hso_serial *serial = urb->context;
 	struct usb_ctrlrequest *req;
 	int status = urb->status;
+	struct tty_struct *tty;
 
 	/* sanity check */
 	if (!serial)
@@ -1713,9 +1999,11 @@
 
 	spin_lock(&serial->serial_lock);
 	serial->tx_urb_used = 0;
+	tty = tty_kref_get(serial->tty);
 	spin_unlock(&serial->serial_lock);
 	if (status) {
 		log_usb_status(status, __func__);
+		tty_kref_put(tty);
 		return;
 	}
 
@@ -1734,25 +2022,31 @@
 		spin_unlock(&serial->serial_lock);
 	} else {
 		hso_put_activity(serial->parent);
-		if (serial->tty)
-			tty_wakeup(serial->tty);
+		if (tty)
+			tty_wakeup(tty);
 		/* response to a write command */
 		hso_kick_transmit(serial);
 	}
+	tty_kref_put(tty);
 }
 
 /* handle RX data for serial port */
 static int put_rxbuf_data(struct urb *urb, struct hso_serial *serial)
 {
-	struct tty_struct *tty = serial->tty;
+	struct tty_struct *tty;
 	int write_length_remaining = 0;
 	int curr_write_len;
+
 	/* Sanity check */
 	if (urb == NULL || serial == NULL) {
 		D1("serial = NULL");
 		return -2;
 	}
 
+	spin_lock(&serial->serial_lock);
+	tty = tty_kref_get(serial->tty);
+	spin_unlock(&serial->serial_lock);
+
 	/* Push data to tty */
 	if (tty) {
 		write_length_remaining = urb->actual_length -
@@ -1774,6 +2068,7 @@
 		serial->curr_rx_urb_offset = 0;
 		serial->rx_urb_filled[hso_urb_to_index(serial, urb)] = 0;
 	}
+	tty_kref_put(tty);
 	return write_length_remaining;
 }
 
@@ -1922,7 +2217,10 @@
 		serial->shared_int->use_count++;
 		mutex_unlock(&serial->shared_int->shared_int_lock);
 	}
-
+	if (serial->tiocmget)
+		tiocmget_submit_urb(serial,
+				    serial->tiocmget,
+				    serial->parent->usb);
 	return result;
 }
 
@@ -1930,6 +2228,7 @@
 {
 	int i;
 	struct hso_serial *serial = dev2ser(hso_dev);
+	struct hso_tiocmget  *tiocmget;
 
 	if (!serial)
 		return -ENODEV;
@@ -1958,6 +2257,11 @@
 		}
 		mutex_unlock(&serial->shared_int->shared_int_lock);
 	}
+	tiocmget = serial->tiocmget;
+	if (tiocmget) {
+		wake_up_interruptible(&tiocmget->waitq);
+		usb_kill_urb(tiocmget->urb);
+	}
 
 	return 0;
 }
@@ -2304,6 +2608,20 @@
 	return NULL;
 }
 
+static void hso_free_tiomget(struct hso_serial *serial)
+{
+	struct hso_tiocmget *tiocmget = serial->tiocmget;
+	if (tiocmget) {
+		kfree(tiocmget);
+		if (tiocmget->urb) {
+			usb_free_urb(tiocmget->urb);
+			tiocmget->urb = NULL;
+		}
+		serial->tiocmget = NULL;
+
+	}
+}
+
 /* Frees an AT channel ( goes for both mux and non-mux ) */
 static void hso_free_serial_device(struct hso_device *hso_dev)
 {
@@ -2322,6 +2640,7 @@
 		else
 			mutex_unlock(&serial->shared_int->shared_int_lock);
 	}
+	hso_free_tiomget(serial);
 	kfree(serial);
 	hso_free_device(hso_dev);
 }
@@ -2333,6 +2652,7 @@
 	struct hso_device *hso_dev;
 	struct hso_serial *serial;
 	int num_urbs;
+	struct hso_tiocmget *tiocmget;
 
 	hso_dev = hso_create_device(interface, port);
 	if (!hso_dev)
@@ -2345,8 +2665,27 @@
 	serial->parent = hso_dev;
 	hso_dev->port_data.dev_serial = serial;
 
-	if (port & HSO_PORT_MODEM)
+	if ((port & HSO_PORT_MASK) == HSO_PORT_MODEM) {
 		num_urbs = 2;
+		serial->tiocmget = kzalloc(sizeof(struct hso_tiocmget),
+					   GFP_KERNEL);
+		/* it isn't going to break our heart if serial->tiocmget
+		 *  allocation fails don't bother checking this.
+		 */
+		if (serial->tiocmget) {
+			tiocmget = serial->tiocmget;
+			tiocmget->urb = usb_alloc_urb(0, GFP_KERNEL);
+			if (tiocmget->urb) {
+				mutex_init(&tiocmget->mutex);
+				init_waitqueue_head(&tiocmget->waitq);
+				tiocmget->endp = hso_get_ep(
+					interface,
+					USB_ENDPOINT_XFER_INT,
+					USB_DIR_IN);
+			} else
+				hso_free_tiomget(serial);
+		}
+	}
 	else
 		num_urbs = 1;
 
@@ -2382,6 +2721,7 @@
 exit2:
 	hso_serial_common_free(serial);
 exit:
+	hso_free_tiomget(serial);
 	kfree(serial);
 	hso_free_device(hso_dev);
 	return NULL;
@@ -2786,15 +3126,20 @@
 static void hso_free_interface(struct usb_interface *interface)
 {
 	struct hso_serial *hso_dev;
+	struct tty_struct *tty;
 	int i;
 
 	for (i = 0; i < HSO_SERIAL_TTY_MINORS; i++) {
 		if (serial_table[i]
 		    && (serial_table[i]->interface == interface)) {
 			hso_dev = dev2ser(serial_table[i]);
-			if (hso_dev->tty)
-				tty_hangup(hso_dev->tty);
+			spin_lock_irq(&hso_dev->serial_lock);
+			tty = tty_kref_get(hso_dev->tty);
+			spin_unlock_irq(&hso_dev->serial_lock);
+			if (tty)
+				tty_hangup(tty);
 			mutex_lock(&hso_dev->parent->mutex);
+			tty_kref_put(tty);
 			hso_dev->parent->usb_gone = 1;
 			mutex_unlock(&hso_dev->parent->mutex);
 			kref_put(&serial_table[i]->ref, hso_serial_ref_free);
@@ -2887,6 +3232,7 @@
 	.close = hso_serial_close,
 	.write = hso_serial_write,
 	.write_room = hso_serial_write_room,
+	.ioctl = hso_serial_ioctl,
 	.set_termios = hso_serial_set_termios,
 	.chars_in_buffer = hso_serial_chars_in_buffer,
 	.tiocmget = hso_serial_tiocmget,
@@ -2939,9 +3285,7 @@
 	tty_drv->subtype = SERIAL_TYPE_NORMAL;
 	tty_drv->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV;
 	tty_drv->init_termios = tty_std_termios;
-	tty_drv->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL;
-	tty_drv->termios = hso_serial_termios;
-	tty_drv->termios_locked = hso_serial_termios_locked;
+	hso_init_termios(&tty_drv->init_termios);
 	tty_set_operations(tty_drv, &hso_serial_ops);
 
 	/* register the tty driver */
diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c
index 737bd94..65e8294 100644
--- a/drivers/oprofile/buffer_sync.c
+++ b/drivers/oprofile/buffer_sync.c
@@ -200,7 +200,7 @@
 {
 	unsigned long cookie;
 
-	if (path->dentry->d_cookie)
+	if (path->dentry->d_flags & DCACHE_COOKIE)
 		return (unsigned long)path->dentry;
 	get_dcookie(path, &cookie);
 	return cookie;
diff --git a/drivers/parisc/iosapic.c b/drivers/parisc/iosapic.c
index 7beffca..9dedbbd 100644
--- a/drivers/parisc/iosapic.c
+++ b/drivers/parisc/iosapic.c
@@ -704,16 +704,17 @@
 }
 
 #ifdef CONFIG_SMP
-static void iosapic_set_affinity_irq(unsigned int irq, cpumask_t dest)
+static void iosapic_set_affinity_irq(unsigned int irq,
+				     const struct cpumask *dest)
 {
 	struct vector_info *vi = iosapic_get_vector(irq);
 	u32 d0, d1, dummy_d0;
 	unsigned long flags;
 
-	if (cpu_check_affinity(irq, &dest))
+	if (cpu_check_affinity(irq, dest))
 		return;
 
-	vi->txn_addr = txn_affinity_addr(irq, first_cpu(dest));
+	vi->txn_addr = txn_affinity_addr(irq, cpumask_first(dest));
 
 	spin_lock_irqsave(&iosapic_lock, flags);
 	/* d1 contains the destination CPU, so only want to set that
diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 8514c3a..c2e1bcb 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -45,7 +45,7 @@
 
 #include "cpqphp.h"
 #include "cpqphp_nvram.h"
-#include "../../../arch/x86/pci/pci.h"	/* horrible hack showing how processor dependent we are... */
+#include <asm/pci_x86.h>
 
 
 /* Global variables */
diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 0902193..df146be 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -37,7 +37,7 @@
 #include "../pci.h"
 #include "cpqphp.h"
 #include "cpqphp_nvram.h"
-#include "../../../arch/x86/pci/pci.h"	/* horrible hack showing how processor dependent we are... */
+#include <asm/pci_x86.h>
 
 
 u8 cpqhp_nic_irq;
diff --git a/drivers/pci/hotplug/ibmphp_core.c b/drivers/pci/hotplug/ibmphp_core.c
index 633e743..dd18f85 100644
--- a/drivers/pci/hotplug/ibmphp_core.c
+++ b/drivers/pci/hotplug/ibmphp_core.c
@@ -35,7 +35,7 @@
 #include <linux/delay.h>
 #include <linux/wait.h>
 #include "../pci.h"
-#include "../../../arch/x86/pci/pci.h"	/* for struct irq_routing_table */
+#include <asm/pci_x86.h>		/* for struct irq_routing_table */
 #include "ibmphp.h"
 
 #define attn_on(sl)  ibmphp_hpc_writeslot (sl, HPC_SLOT_ATTNON)
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 5d72866..c884858 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -74,7 +74,7 @@
 	int len;
 
 	mask = pcibus_to_cpumask(to_pci_dev(dev)->bus);
-	len = cpumask_scnprintf(buf, PAGE_SIZE-2, mask);
+	len = cpumask_scnprintf(buf, PAGE_SIZE-2, &mask);
 	buf[len++] = '\n';
 	buf[len] = '\0';
 	return len;
@@ -88,7 +88,7 @@
 	int len;
 
 	mask = pcibus_to_cpumask(to_pci_dev(dev)->bus);
-	len = cpulist_scnprintf(buf, PAGE_SIZE-2, mask);
+	len = cpulist_scnprintf(buf, PAGE_SIZE-2, &mask);
 	buf[len++] = '\n';
 	buf[len] = '\0';
 	return len;
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 003a9b3..5b3f593 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -55,8 +55,8 @@
 
 	cpumask = pcibus_to_cpumask(to_pci_bus(dev));
 	ret = type?
-		cpulist_scnprintf(buf, PAGE_SIZE-2, cpumask):
-		cpumask_scnprintf(buf, PAGE_SIZE-2, cpumask);
+		cpulist_scnprintf(buf, PAGE_SIZE-2, &cpumask) :
+		cpumask_scnprintf(buf, PAGE_SIZE-2, &cpumask);
 	buf[ret++] = '\n';
 	buf[ret] = '\0';
 	return ret;
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 8a8df75..06b7182 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -632,8 +632,8 @@
 	struct pt_regs *old_regs;
 
 	old_regs = set_irq_regs(regs);
-	irq_enter();
 	s390_idle_check();
+	irq_enter();
 	if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator)
 		/* Serve timer interrupts first. */
 		clock_comparator_work();
diff --git a/drivers/s390/s390mach.c b/drivers/s390/s390mach.c
index 834e9ee..92b0417 100644
--- a/drivers/s390/s390mach.c
+++ b/drivers/s390/s390mach.c
@@ -18,6 +18,7 @@
 #include <asm/etr.h>
 #include <asm/lowcore.h>
 #include <asm/cio.h>
+#include <asm/cpu.h>
 #include "s390mach.h"
 
 static struct semaphore m_sem;
@@ -369,6 +370,8 @@
 
 	lockdep_off();
 
+	s390_idle_check();
+
 	mci = (struct mci *) &S390_lowcore.mcck_interruption_code;
 	mcck = &__get_cpu_var(cpu_mcck);
 	umode = user_mode(regs);
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index 152d4aa..b732297 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -21,7 +21,7 @@
 	  You also need to say Y here if you have a device which speaks
 	  the SCSI protocol.  Examples of this include the parallel port
 	  version of the IOMEGA ZIP drive, USB storage devices, Fibre
-	  Channel, FireWire storage and the IDE-SCSI emulation driver.
+	  Channel, and FireWire storage.
 
 	  To compile this driver as a module, choose M here and read
 	  <file:Documentation/scsi/scsi.txt>.
@@ -101,9 +101,9 @@
 	---help---
 	  The OnStream SC-x0 SCSI tape drives cannot be driven by the
 	  standard st driver, but instead need this special osst driver and
-	  use the  /dev/osstX char device nodes (major 206).  Via usb-storage
-	  and ide-scsi, you may be able to drive the USB-x0 and DI-x0 drives
-	  as well.  Note that there is also a second generation of OnStream
+	  use the  /dev/osstX char device nodes (major 206).  Via usb-storage,
+	  you may be able to drive the USB-x0 and DI-x0 drives as well.
+	  Note that there is also a second generation of OnStream
 	  tape drives (ADR-x0) that supports the standard SCSI-2 commands for
 	  tapes (QIC-157) and can be driven by the standard driver st.
 	  For more information, you may have a look at the SCSI-HOWTO
diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile
index 1410697..7461eb0 100644
--- a/drivers/scsi/Makefile
+++ b/drivers/scsi/Makefile
@@ -105,7 +105,6 @@
 obj-$(CONFIG_SCSI_INITIO)	+= initio.o
 obj-$(CONFIG_SCSI_INIA100)	+= a100u2w.o
 obj-$(CONFIG_SCSI_QLOGICPTI)	+= qlogicpti.o
-obj-$(CONFIG_BLK_DEV_IDESCSI)	+= ide-scsi.o
 obj-$(CONFIG_SCSI_MESH)		+= mesh.o
 obj-$(CONFIG_SCSI_MAC53C94)	+= mac53c94.o
 obj-$(CONFIG_BLK_DEV_3W_XXXX_RAID) += 3w-xxxx.o
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
deleted file mode 100644
index c24140a..0000000
--- a/drivers/scsi/ide-scsi.c
+++ /dev/null
@@ -1,840 +0,0 @@
-/*
- * Copyright (C) 1996-1999  Gadi Oxman <gadio@netvision.net.il>
- * Copyright (C) 2004-2005  Bartlomiej Zolnierkiewicz
- */
-
-/*
- * Emulation of a SCSI host adapter for IDE ATAPI devices.
- *
- * With this driver, one can use the Linux SCSI drivers instead of the
- * native IDE ATAPI drivers.
- *
- * Ver 0.1   Dec  3 96   Initial version.
- * Ver 0.2   Jan 26 97   Fixed bug in cleanup_module() and added emulation
- *                        of MODE_SENSE_6/MODE_SELECT_6 for cdroms. Thanks
- *                        to Janos Farkas for pointing this out.
- *                       Avoid using bitfields in structures for m68k.
- *                       Added Scatter/Gather and DMA support.
- * Ver 0.4   Dec  7 97   Add support for ATAPI PD/CD drives.
- *                       Use variable timeout for each command.
- * Ver 0.5   Jan  2 98   Fix previous PD/CD support.
- *                       Allow disabling of SCSI-6 to SCSI-10 transformation.
- * Ver 0.6   Jan 27 98   Allow disabling of SCSI command translation layer
- *                        for access through /dev/sg.
- *                       Fix MODE_SENSE_6/MODE_SELECT_6/INQUIRY translation.
- * Ver 0.7   Dec 04 98   Ignore commands where lun != 0 to avoid multiple
- *                        detection of devices with CONFIG_SCSI_MULTI_LUN
- * Ver 0.8   Feb 05 99   Optical media need translation too. Reverse 0.7.
- * Ver 0.9   Jul 04 99   Fix a bug in SG_SET_TRANSFORM.
- * Ver 0.91  Jun 10 02   Fix "off by one" error in transforms
- * Ver 0.92  Dec 31 02   Implement new SCSI mid level API
- */
-
-#define IDESCSI_VERSION "0.92"
-
-#include <linux/module.h>
-#include <linux/types.h>
-#include <linux/string.h>
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/ioport.h>
-#include <linux/blkdev.h>
-#include <linux/errno.h>
-#include <linux/slab.h>
-#include <linux/ide.h>
-#include <linux/scatterlist.h>
-#include <linux/delay.h>
-#include <linux/mutex.h>
-#include <linux/bitops.h>
-
-#include <asm/io.h>
-#include <asm/uaccess.h>
-
-#include <scsi/scsi.h>
-#include <scsi/scsi_cmnd.h>
-#include <scsi/scsi_device.h>
-#include <scsi/scsi_host.h>
-#include <scsi/scsi_tcq.h>
-#include <scsi/sg.h>
-
-#define IDESCSI_DEBUG_LOG		0
-
-#if IDESCSI_DEBUG_LOG
-#define debug_log(fmt, args...) \
-	printk(KERN_INFO "ide-scsi: " fmt, ## args)
-#else
-#define debug_log(fmt, args...) do {} while (0)
-#endif
-
-/*
- *	SCSI command transformation layer
- */
-#define IDESCSI_SG_TRANSFORM		1	/* /dev/sg transformation */
-
-/*
- *	Log flags
- */
-#define IDESCSI_LOG_CMD			0	/* Log SCSI commands */
-
-typedef struct ide_scsi_obj {
-	ide_drive_t		*drive;
-	ide_driver_t		*driver;
-	struct gendisk		*disk;
-	struct Scsi_Host	*host;
-
-	unsigned long transform;		/* SCSI cmd translation layer */
-	unsigned long log;			/* log flags */
-} idescsi_scsi_t;
-
-static DEFINE_MUTEX(idescsi_ref_mutex);
-/* Set by module param to skip cd */
-static int idescsi_nocd;
-
-#define ide_scsi_g(disk) \
-	container_of((disk)->private_data, struct ide_scsi_obj, driver)
-
-static struct ide_scsi_obj *ide_scsi_get(struct gendisk *disk)
-{
-	struct ide_scsi_obj *scsi = NULL;
-
-	mutex_lock(&idescsi_ref_mutex);
-	scsi = ide_scsi_g(disk);
-	if (scsi) {
-		if (ide_device_get(scsi->drive))
-			scsi = NULL;
-		else
-			scsi_host_get(scsi->host);
-	}
-	mutex_unlock(&idescsi_ref_mutex);
-	return scsi;
-}
-
-static void ide_scsi_put(struct ide_scsi_obj *scsi)
-{
-	ide_drive_t *drive = scsi->drive;
-
-	mutex_lock(&idescsi_ref_mutex);
-	scsi_host_put(scsi->host);
-	ide_device_put(drive);
-	mutex_unlock(&idescsi_ref_mutex);
-}
-
-static inline idescsi_scsi_t *scsihost_to_idescsi(struct Scsi_Host *host)
-{
-	return (idescsi_scsi_t*) (&host[1]);
-}
-
-static inline idescsi_scsi_t *drive_to_idescsi(ide_drive_t *ide_drive)
-{
-	return scsihost_to_idescsi(ide_drive->driver_data);
-}
-
-static void ide_scsi_hex_dump(u8 *data, int len)
-{
-	print_hex_dump(KERN_CONT, "", DUMP_PREFIX_NONE, 16, 1, data, len, 0);
-}
-
-static int idescsi_end_request(ide_drive_t *, int, int);
-
-static void ide_scsi_callback(ide_drive_t *drive, int dsc)
-{
-	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-	struct ide_atapi_pc *pc = drive->pc;
-
-	if (pc->flags & PC_FLAG_TIMEDOUT)
-		debug_log("%s: got timed out packet %lu at %lu\n", __func__,
-			  pc->scsi_cmd->serial_number, jiffies);
-		/* end this request now - scsi should retry it*/
-	else if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
-		printk(KERN_INFO "Packet command completed, %d bytes"
-				 " transferred\n", pc->xferred);
-
-	idescsi_end_request(drive, 1, 0);
-}
-
-static int idescsi_check_condition(ide_drive_t *drive,
-		struct request *failed_cmd)
-{
-	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-	struct ide_atapi_pc   *pc;
-	struct request *rq;
-	u8             *buf;
-
-	/* stuff a sense request in front of our current request */
-	pc = kzalloc(sizeof(struct ide_atapi_pc), GFP_ATOMIC);
-	rq = blk_get_request(drive->queue, READ, GFP_ATOMIC);
-	buf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_ATOMIC);
-	if (!pc || !rq || !buf) {
-		kfree(buf);
-		if (rq)
-			blk_put_request(rq);
-		kfree(pc);
-		return -ENOMEM;
-	}
-	rq->special = (char *) pc;
-	pc->rq = rq;
-	pc->buf = buf;
-	pc->c[0] = REQUEST_SENSE;
-	pc->c[4] = pc->req_xfer = pc->buf_size = SCSI_SENSE_BUFFERSIZE;
-	rq->cmd_type = REQ_TYPE_SENSE;
-	rq->cmd_flags |= REQ_PREEMPT;
-	pc->timeout = jiffies + WAIT_READY;
-	/* NOTE! Save the failed packet command in "rq->buffer" */
-	rq->buffer = (void *) failed_cmd->special;
-	pc->scsi_cmd = ((struct ide_atapi_pc *) failed_cmd->special)->scsi_cmd;
-	if (test_bit(IDESCSI_LOG_CMD, &scsi->log)) {
-		printk ("ide-scsi: %s: queue cmd = ", drive->name);
-		ide_scsi_hex_dump(pc->c, 6);
-	}
-	rq->rq_disk = scsi->disk;
-	rq->ref_count++;
-	memcpy(rq->cmd, pc->c, 12);
-	ide_do_drive_cmd(drive, rq);
-	return 0;
-}
-
-static ide_startstop_t
-idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
-{
-	ide_hwif_t *hwif = drive->hwif;
-
-	if (hwif->tp_ops->read_status(hwif) & (ATA_BUSY | ATA_DRQ))
-		/* force an abort */
-		hwif->tp_ops->exec_command(hwif, ATA_CMD_IDLEIMMEDIATE);
-
-	rq->errors++;
-
-	idescsi_end_request(drive, 0, 0);
-
-	return ide_stopped;
-}
-
-static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs)
-{
-	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-	struct request *rq = HWGROUP(drive)->rq;
-	struct ide_atapi_pc *pc = (struct ide_atapi_pc *) rq->special;
-	int log = test_bit(IDESCSI_LOG_CMD, &scsi->log);
-	struct Scsi_Host *host;
-	int errors = rq->errors;
-	unsigned long flags;
-
-	if (!blk_special_request(rq) && !blk_sense_request(rq)) {
-		ide_end_request(drive, uptodate, nrsecs);
-		return 0;
-	}
-	ide_end_drive_cmd (drive, 0, 0);
-	if (blk_sense_request(rq)) {
-		struct ide_atapi_pc *opc = (struct ide_atapi_pc *) rq->buffer;
-		if (log) {
-			printk ("ide-scsi: %s: wrap up check %lu, rst = ", drive->name, opc->scsi_cmd->serial_number);
-			ide_scsi_hex_dump(pc->buf, 16);
-		}
-		memcpy((void *) opc->scsi_cmd->sense_buffer, pc->buf,
-			SCSI_SENSE_BUFFERSIZE);
-		kfree(pc->buf);
-		kfree(pc);
-		blk_put_request(rq);
-		pc = opc;
-		rq = pc->rq;
-		pc->scsi_cmd->result = (CHECK_CONDITION << 1) |
-				(((pc->flags & PC_FLAG_TIMEDOUT) ?
-				  DID_TIME_OUT :
-				  DID_OK) << 16);
-	} else if (pc->flags & PC_FLAG_TIMEDOUT) {
-		if (log)
-			printk (KERN_WARNING "ide-scsi: %s: timed out for %lu\n",
-					drive->name, pc->scsi_cmd->serial_number);
-		pc->scsi_cmd->result = DID_TIME_OUT << 16;
-	} else if (errors >= ERROR_MAX) {
-		pc->scsi_cmd->result = DID_ERROR << 16;
-		if (log)
-			printk ("ide-scsi: %s: I/O error for %lu\n", drive->name, pc->scsi_cmd->serial_number);
-	} else if (errors) {
-		if (log)
-			printk ("ide-scsi: %s: check condition for %lu\n", drive->name, pc->scsi_cmd->serial_number);
-		if (!idescsi_check_condition(drive, rq))
-			/* we started a request sense, so we'll be back, exit for now */
-			return 0;
-		pc->scsi_cmd->result = (CHECK_CONDITION << 1) | (DID_OK << 16);
-	} else {
-		pc->scsi_cmd->result = DID_OK << 16;
-	}
-	host = pc->scsi_cmd->device->host;
-	spin_lock_irqsave(host->host_lock, flags);
-	pc->done(pc->scsi_cmd);
-	spin_unlock_irqrestore(host->host_lock, flags);
-	kfree(pc);
-	blk_put_request(rq);
-	drive->pc = NULL;
-	return 0;
-}
-
-static inline int idescsi_set_direction(struct ide_atapi_pc *pc)
-{
-	switch (pc->c[0]) {
-		case READ_6: case READ_10: case READ_12:
-			pc->flags &= ~PC_FLAG_WRITING;
-			return 0;
-		case WRITE_6: case WRITE_10: case WRITE_12:
-			pc->flags |= PC_FLAG_WRITING;
-			return 0;
-		default:
-			return 1;
-	}
-}
-
-static int idescsi_map_sg(ide_drive_t *drive, struct ide_atapi_pc *pc)
-{
-	ide_hwif_t *hwif = drive->hwif;
-	struct scatterlist *sg, *scsi_sg;
-	int segments;
-
-	if (!pc->req_xfer || pc->req_xfer % 1024)
-		return 1;
-
-	if (idescsi_set_direction(pc))
-		return 1;
-
-	sg = hwif->sg_table;
-	scsi_sg = scsi_sglist(pc->scsi_cmd);
-	segments = scsi_sg_count(pc->scsi_cmd);
-
-	if (segments > hwif->sg_max_nents)
-		return 1;
-
-	hwif->sg_nents = segments;
-	memcpy(sg, scsi_sg, sizeof(*sg) * segments);
-
-	return 0;
-}
-
-static ide_startstop_t idescsi_issue_pc(ide_drive_t *drive,
-		struct ide_atapi_pc *pc)
-{
-	/* Set the current packet command */
-	drive->pc = pc;
-
-	return ide_issue_pc(drive, ide_scsi_get_timeout(pc), ide_scsi_expiry);
-}
-
-/*
- *	idescsi_do_request is our request handling function.
- */
-static ide_startstop_t idescsi_do_request (ide_drive_t *drive, struct request *rq, sector_t block)
-{
-	debug_log("dev: %s, cmd: %x, errors: %d\n", rq->rq_disk->disk_name,
-		  rq->cmd[0], rq->errors);
-	debug_log("sector: %ld, nr_sectors: %ld, current_nr_sectors: %d\n",
-		  rq->sector, rq->nr_sectors, rq->current_nr_sectors);
-
-	if (blk_sense_request(rq) || blk_special_request(rq)) {
-		struct ide_atapi_pc *pc = (struct ide_atapi_pc *)rq->special;
-
-		if ((drive->dev_flags & IDE_DFLAG_USING_DMA) &&
-		    idescsi_map_sg(drive, pc) == 0)
-			pc->flags |= PC_FLAG_DMA_OK;
-
-		return idescsi_issue_pc(drive, pc);
-	}
-	blk_dump_rq_flags(rq, "ide-scsi: unsup command");
-	idescsi_end_request (drive, 0, 0);
-	return ide_stopped;
-}
-
-#ifdef CONFIG_IDE_PROC_FS
-static ide_proc_entry_t idescsi_proc[] = {
-	{ "capacity", S_IFREG|S_IRUGO, proc_ide_read_capacity, NULL },
-	{ NULL, 0, NULL, NULL }
-};
-
-#define ide_scsi_devset_get(name, field) \
-static int get_##name(ide_drive_t *drive) \
-{ \
-	idescsi_scsi_t *scsi = drive_to_idescsi(drive); \
-	return scsi->field; \
-}
-
-#define ide_scsi_devset_set(name, field) \
-static int set_##name(ide_drive_t *drive, int arg) \
-{ \
-	idescsi_scsi_t *scsi = drive_to_idescsi(drive); \
-	scsi->field = arg; \
-	return 0; \
-}
-
-#define ide_scsi_devset_rw_field(_name, _field) \
-ide_scsi_devset_get(_name, _field); \
-ide_scsi_devset_set(_name, _field); \
-IDE_DEVSET(_name, DS_SYNC, get_##_name, set_##_name);
-
-ide_devset_rw_field(bios_cyl, bios_cyl);
-ide_devset_rw_field(bios_head, bios_head);
-ide_devset_rw_field(bios_sect, bios_sect);
-
-ide_scsi_devset_rw_field(transform, transform);
-ide_scsi_devset_rw_field(log, log);
-
-static const struct ide_proc_devset idescsi_settings[] = {
-	IDE_PROC_DEVSET(bios_cyl,  0, 1023),
-	IDE_PROC_DEVSET(bios_head, 0,  255),
-	IDE_PROC_DEVSET(bios_sect, 0,	63),
-	IDE_PROC_DEVSET(log,	   0,	 1),
-	IDE_PROC_DEVSET(transform, 0,	 3),
-	{ 0 },
-};
-
-static ide_proc_entry_t *ide_scsi_proc_entries(ide_drive_t *drive)
-{
-	return idescsi_proc;
-}
-
-static const struct ide_proc_devset *ide_scsi_proc_devsets(ide_drive_t *drive)
-{
-	return idescsi_settings;
-}
-#endif
-
-/*
- *	Driver initialization.
- */
-static void idescsi_setup (ide_drive_t *drive, idescsi_scsi_t *scsi)
-{
-	clear_bit(IDESCSI_SG_TRANSFORM, &scsi->transform);
-#if IDESCSI_DEBUG_LOG
-	set_bit(IDESCSI_LOG_CMD, &scsi->log);
-#endif /* IDESCSI_DEBUG_LOG */
-
-	drive->pc_callback	 = ide_scsi_callback;
-	drive->pc_update_buffers = NULL;
-	drive->pc_io_buffers	 = ide_io_buffers;
-
-	ide_proc_register_driver(drive, scsi->driver);
-}
-
-static void ide_scsi_remove(ide_drive_t *drive)
-{
-	struct Scsi_Host *scsihost = drive->driver_data;
-	struct ide_scsi_obj *scsi = scsihost_to_idescsi(scsihost);
-	struct gendisk *g = scsi->disk;
-
-	scsi_remove_host(scsihost);
-	ide_proc_unregister_driver(drive, scsi->driver);
-
-	ide_unregister_region(g);
-
-	drive->driver_data = NULL;
-	g->private_data = NULL;
-	put_disk(g);
-
-	ide_scsi_put(scsi);
-
-	drive->dev_flags &= ~IDE_DFLAG_SCSI;
-}
-
-static int ide_scsi_probe(ide_drive_t *);
-
-static ide_driver_t idescsi_driver = {
-	.gen_driver = {
-		.owner		= THIS_MODULE,
-		.name		= "ide-scsi",
-		.bus		= &ide_bus_type,
-	},
-	.probe			= ide_scsi_probe,
-	.remove			= ide_scsi_remove,
-	.version		= IDESCSI_VERSION,
-	.do_request		= idescsi_do_request,
-	.end_request		= idescsi_end_request,
-	.error                  = idescsi_atapi_error,
-#ifdef CONFIG_IDE_PROC_FS
-	.proc_entries		= ide_scsi_proc_entries,
-	.proc_devsets		= ide_scsi_proc_devsets,
-#endif
-};
-
-static int idescsi_ide_open(struct block_device *bdev, fmode_t mode)
-{
-	struct ide_scsi_obj *scsi = ide_scsi_get(bdev->bd_disk);
-
-	if (!scsi)
-		return -ENXIO;
-
-	return 0;
-}
-
-static int idescsi_ide_release(struct gendisk *disk, fmode_t mode)
-{
-	ide_scsi_put(ide_scsi_g(disk));
-	return 0;
-}
-
-static int idescsi_ide_ioctl(struct block_device *bdev, fmode_t mode,
-			unsigned int cmd, unsigned long arg)
-{
-	struct ide_scsi_obj *scsi = ide_scsi_g(bdev->bd_disk);
-	return generic_ide_ioctl(scsi->drive, bdev, cmd, arg);
-}
-
-static struct block_device_operations idescsi_ops = {
-	.owner		= THIS_MODULE,
-	.open		= idescsi_ide_open,
-	.release	= idescsi_ide_release,
-	.locked_ioctl	= idescsi_ide_ioctl,
-};
-
-static int idescsi_slave_configure(struct scsi_device * sdp)
-{
-	/* Configure detected device */
-	sdp->use_10_for_rw = 1;
-	sdp->use_10_for_ms = 1;
-	scsi_adjust_queue_depth(sdp, MSG_SIMPLE_TAG, sdp->host->cmd_per_lun);
-	return 0;
-}
-
-static const char *idescsi_info (struct Scsi_Host *host)
-{
-	return "SCSI host adapter emulation for IDE ATAPI devices";
-}
-
-static int idescsi_ioctl (struct scsi_device *dev, int cmd, void __user *arg)
-{
-	idescsi_scsi_t *scsi = scsihost_to_idescsi(dev->host);
-
-	if (cmd == SG_SET_TRANSFORM) {
-		if (arg)
-			set_bit(IDESCSI_SG_TRANSFORM, &scsi->transform);
-		else
-			clear_bit(IDESCSI_SG_TRANSFORM, &scsi->transform);
-		return 0;
-	} else if (cmd == SG_GET_TRANSFORM)
-		return put_user(test_bit(IDESCSI_SG_TRANSFORM, &scsi->transform), (int __user *) arg);
-	return -EINVAL;
-}
-
-static int idescsi_queue (struct scsi_cmnd *cmd,
-		void (*done)(struct scsi_cmnd *))
-{
-	struct Scsi_Host *host = cmd->device->host;
-	idescsi_scsi_t *scsi = scsihost_to_idescsi(host);
-	ide_drive_t *drive = scsi->drive;
-	struct request *rq = NULL;
-	struct ide_atapi_pc *pc = NULL;
-	int write = cmd->sc_data_direction == DMA_TO_DEVICE;
-
-	if (!drive) {
-		scmd_printk (KERN_ERR, cmd, "drive not present\n");
-		goto abort;
-	}
-	scsi = drive_to_idescsi(drive);
-	pc = kmalloc(sizeof(struct ide_atapi_pc), GFP_ATOMIC);
-	rq = blk_get_request(drive->queue, write, GFP_ATOMIC);
-	if (rq == NULL || pc == NULL) {
-		printk (KERN_ERR "ide-scsi: %s: out of memory\n", drive->name);
-		goto abort;
-	}
-
-	memset (pc->c, 0, 12);
-	pc->flags = 0;
-	if (cmd->sc_data_direction == DMA_TO_DEVICE)
-		pc->flags |= PC_FLAG_WRITING;
-	pc->rq = rq;
-	memcpy (pc->c, cmd->cmnd, cmd->cmd_len);
-	pc->buf = NULL;
-	pc->sg = scsi_sglist(cmd);
-	pc->sg_cnt = scsi_sg_count(cmd);
-	pc->b_count = 0;
-	pc->req_xfer = pc->buf_size = scsi_bufflen(cmd);
-	pc->scsi_cmd = cmd;
-	pc->done = done;
-	pc->timeout = jiffies + cmd->request->timeout;
-
-	if (test_bit(IDESCSI_LOG_CMD, &scsi->log)) {
-		printk ("ide-scsi: %s: que %lu, cmd = ", drive->name, cmd->serial_number);
-		ide_scsi_hex_dump(cmd->cmnd, cmd->cmd_len);
-		if (memcmp(pc->c, cmd->cmnd, cmd->cmd_len)) {
-			printk ("ide-scsi: %s: que %lu, tsl = ", drive->name, cmd->serial_number);
-			ide_scsi_hex_dump(pc->c, 12);
-		}
-	}
-
-	rq->special = (char *) pc;
-	rq->cmd_type = REQ_TYPE_SPECIAL;
-	spin_unlock_irq(host->host_lock);
-	rq->ref_count++;
-	memcpy(rq->cmd, pc->c, 12);
-	blk_execute_rq_nowait(drive->queue, scsi->disk, rq, 0, NULL);
-	spin_lock_irq(host->host_lock);
-	return 0;
-abort:
-	kfree (pc);
-	if (rq)
-		blk_put_request(rq);
-	cmd->result = DID_ERROR << 16;
-	done(cmd);
-	return 0;
-}
-
-static int idescsi_eh_abort (struct scsi_cmnd *cmd)
-{
-	idescsi_scsi_t *scsi  = scsihost_to_idescsi(cmd->device->host);
-	ide_drive_t    *drive = scsi->drive;
-	ide_hwif_t     *hwif;
-	ide_hwgroup_t  *hwgroup;
-	int		busy;
-	int             ret   = FAILED;
-
-	struct ide_atapi_pc *pc;
-
-	/* In idescsi_eh_abort we try to gently pry our command from the ide subsystem */
-
-	if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
-		printk (KERN_WARNING "ide-scsi: abort called for %lu\n", cmd->serial_number);
-
-	if (!drive) {
-		printk (KERN_WARNING "ide-scsi: Drive not set in idescsi_eh_abort\n");
-		WARN_ON(1);
-		goto no_drive;
-	}
-
-	hwif = drive->hwif;
-	hwgroup = hwif->hwgroup;
-
-	/* First give it some more time, how much is "right" is hard to say :-(
-	   FIXME - uses mdelay which causes latency? */
-	busy = ide_wait_not_busy(hwif, 100);
-	if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
-		printk (KERN_WARNING "ide-scsi: drive did%s become ready\n", busy?" not":"");
-
-	spin_lock_irq(&hwgroup->lock);
-
-	/* If there is no pc running we're done (our interrupt took care of it) */
-	pc = drive->pc;
-	if (pc == NULL) {
-		ret = SUCCESS;
-		goto ide_unlock;
-	}
-
-	/* It's somewhere in flight. Does ide subsystem agree? */
-	if (pc->scsi_cmd->serial_number == cmd->serial_number && !busy &&
-	    elv_queue_empty(drive->queue) && HWGROUP(drive)->rq != pc->rq) {
-		/*
-		 * FIXME - not sure this condition can ever occur
-		 */
-		printk (KERN_ERR "ide-scsi: cmd aborted!\n");
-
-		if (blk_sense_request(pc->rq))
-			kfree(pc->buf);
-		/* we need to call blk_put_request twice. */
-		blk_put_request(pc->rq);
-		blk_put_request(pc->rq);
-		kfree(pc);
-		drive->pc = NULL;
-
-		ret = SUCCESS;
-	}
-
-ide_unlock:
-	spin_unlock_irq(&hwgroup->lock);
-no_drive:
-	if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
-		printk (KERN_WARNING "ide-scsi: abort returns %s\n", ret == SUCCESS?"success":"failed");
-
-	return ret;
-}
-
-static int idescsi_eh_reset (struct scsi_cmnd *cmd)
-{
-	struct request *req;
-	idescsi_scsi_t *scsi  = scsihost_to_idescsi(cmd->device->host);
-	ide_drive_t    *drive = scsi->drive;
-	ide_hwgroup_t  *hwgroup;
-	int             ready = 0;
-	int             ret   = SUCCESS;
-
-	struct ide_atapi_pc *pc;
-
-	/* In idescsi_eh_reset we forcefully remove the command from the ide subsystem and reset the device. */
-
-	if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
-		printk (KERN_WARNING "ide-scsi: reset called for %lu\n", cmd->serial_number);
-
-	if (!drive) {
-		printk (KERN_WARNING "ide-scsi: Drive not set in idescsi_eh_reset\n");
-		WARN_ON(1);
-		return FAILED;
-	}
-
-	hwgroup = drive->hwif->hwgroup;
-
-	spin_lock_irq(cmd->device->host->host_lock);
-	spin_lock(&hwgroup->lock);
-
-	pc = drive->pc;
-	if (pc)
-		req = pc->rq;
-
-	if (pc == NULL || req != hwgroup->rq || hwgroup->handler == NULL) {
-		printk (KERN_WARNING "ide-scsi: No active request in idescsi_eh_reset\n");
-		spin_unlock(&hwgroup->lock);
-		spin_unlock_irq(cmd->device->host->host_lock);
-		return FAILED;
-	}
-
-	/* kill current request */
-	if (__blk_end_request(req, -EIO, 0))
-		BUG();
-	if (blk_sense_request(req))
-		kfree(pc->buf);
-	kfree(pc);
-	drive->pc = NULL;
-	blk_put_request(req);
-
-	/* now nuke the drive queue */
-	while ((req = elv_next_request(drive->queue))) {
-		if (__blk_end_request(req, -EIO, 0))
-			BUG();
-	}
-
-	hwgroup->rq = NULL;
-	hwgroup->handler = NULL;
-	hwgroup->busy = 1; /* will set this to zero when ide reset finished */
-	spin_unlock(&hwgroup->lock);
-
-	ide_do_reset(drive);
-
-	/* ide_do_reset starts a polling handler which restarts itself every 50ms until the reset finishes */
-
-	do {
-		spin_unlock_irq(cmd->device->host->host_lock);
-		msleep(50);
-		spin_lock_irq(cmd->device->host->host_lock);
-	} while ( HWGROUP(drive)->handler );
-
-	ready = drive_is_ready(drive);
-	HWGROUP(drive)->busy--;
-	if (!ready) {
-		printk (KERN_ERR "ide-scsi: reset failed!\n");
-		ret = FAILED;
-	}
-
-	spin_unlock_irq(cmd->device->host->host_lock);
-	return ret;
-}
-
-static int idescsi_bios(struct scsi_device *sdev, struct block_device *bdev,
-		sector_t capacity, int *parm)
-{
-	idescsi_scsi_t *idescsi = scsihost_to_idescsi(sdev->host);
-	ide_drive_t *drive = idescsi->drive;
-
-	if (drive->bios_cyl && drive->bios_head && drive->bios_sect) {
-		parm[0] = drive->bios_head;
-		parm[1] = drive->bios_sect;
-		parm[2] = drive->bios_cyl;
-	}
-	return 0;
-}
-
-static struct scsi_host_template idescsi_template = {
-	.module			= THIS_MODULE,
-	.name			= "idescsi",
-	.info			= idescsi_info,
-	.slave_configure        = idescsi_slave_configure,
-	.ioctl			= idescsi_ioctl,
-	.queuecommand		= idescsi_queue,
-	.eh_abort_handler	= idescsi_eh_abort,
-	.eh_host_reset_handler  = idescsi_eh_reset,
-	.bios_param		= idescsi_bios,
-	.can_queue		= 40,
-	.this_id		= -1,
-	.sg_tablesize		= 256,
-	.cmd_per_lun		= 5,
-	.max_sectors		= 128,
-	.use_clustering		= DISABLE_CLUSTERING,
-	.emulated		= 1,
-	.proc_name		= "ide-scsi",
-};
-
-static int ide_scsi_probe(ide_drive_t *drive)
-{
-	idescsi_scsi_t *idescsi;
-	struct Scsi_Host *host;
-	struct gendisk *g;
-	static int warned;
-	int err = -ENOMEM;
-	u16 last_lun;
-
-	if (!warned && drive->media == ide_cdrom) {
-		printk(KERN_WARNING "ide-scsi is deprecated for cd burning! Use ide-cd and give dev=/dev/hdX as device\n");
-		warned = 1;
-	}
-
-	if (idescsi_nocd && drive->media == ide_cdrom)
-		return -ENODEV;
-
-	if (!strstr("ide-scsi", drive->driver_req) ||
-	    drive->media == ide_disk ||
-	    !(host = scsi_host_alloc(&idescsi_template,sizeof(idescsi_scsi_t))))
-		return -ENODEV;
-
-	drive->dev_flags |= IDE_DFLAG_SCSI;
-
-	g = alloc_disk(1 << PARTN_BITS);
-	if (!g)
-		goto out_host_put;
-
-	ide_init_disk(g, drive);
-
-	host->max_id = 1;
-
-	last_lun = drive->id[ATA_ID_LAST_LUN];
-	if (last_lun)
-		debug_log("%s: last_lun=%u\n", drive->name, last_lun);
-
-	if ((last_lun & 7) != 7)
-		host->max_lun = (last_lun & 7) + 1;
-	else
-		host->max_lun = 1;
-
-	drive->driver_data = host;
-	idescsi = scsihost_to_idescsi(host);
-	idescsi->drive = drive;
-	idescsi->driver = &idescsi_driver;
-	idescsi->host = host;
-	idescsi->disk = g;
-	g->private_data = &idescsi->driver;
-	err = 0;
-	idescsi_setup(drive, idescsi);
-	g->fops = &idescsi_ops;
-	ide_register_region(g);
-	err = scsi_add_host(host, &drive->gendev);
-	if (!err) {
-		scsi_scan_host(host);
-		return 0;
-	}
-	/* fall through on error */
-	ide_unregister_region(g);
-	ide_proc_unregister_driver(drive, &idescsi_driver);
-
-	put_disk(g);
-out_host_put:
-	drive->dev_flags &= ~IDE_DFLAG_SCSI;
-	scsi_host_put(host);
-	return err;
-}
-
-static int __init init_idescsi_module(void)
-{
-	return driver_register(&idescsi_driver.gen_driver);
-}
-
-static void __exit exit_idescsi_module(void)
-{
-	driver_unregister(&idescsi_driver.gen_driver);
-}
-
-module_param(idescsi_nocd, int, 0600);
-MODULE_PARM_DESC(idescsi_nocd, "Disable handling of CD-ROMs so they may be driven by ide-cd");
-module_init(init_idescsi_module);
-module_exit(exit_idescsi_module);
-MODULE_LICENSE("GPL");
diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 303272a..daa0056 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -279,6 +279,13 @@
 		.fcr		= UART_FCR_ENABLE_FIFO | UART_FCR_R_TRIG_10,
 		.flags		= UART_CAP_FIFO,
 	},
+	[PORT_OCTEON] = {
+		.name		= "OCTEON",
+		.fifo_size	= 64,
+		.tx_loadsz	= 64,
+		.fcr		= UART_FCR_ENABLE_FIFO | UART_FCR_R_TRIG_10,
+		.flags		= UART_CAP_FIFO,
+	},
 };
 
 #if defined (CONFIG_SERIAL_8250_AU1X00)
@@ -303,16 +310,16 @@
 };
 
 /* sane hardware needs no mapping */
-static inline int map_8250_in_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_in_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_AU)
+	if (p->iotype != UPIO_AU)
 		return offset;
 	return au_io_in_map[offset];
 }
 
-static inline int map_8250_out_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_out_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_AU)
+	if (p->iotype != UPIO_AU)
 		return offset;
 	return au_io_out_map[offset];
 }
@@ -341,16 +348,16 @@
 		[UART_SCR]	= 0x2c
 	};
 
-static inline int map_8250_in_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_in_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_RM9000)
+	if (p->iotype != UPIO_RM9000)
 		return offset;
 	return regmap_in[offset];
 }
 
-static inline int map_8250_out_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_out_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_RM9000)
+	if (p->iotype != UPIO_RM9000)
 		return offset;
 	return regmap_out[offset];
 }
@@ -363,108 +370,170 @@
 
 #endif
 
-static unsigned int serial_in(struct uart_8250_port *up, int offset)
+static unsigned int hub6_serial_in(struct uart_port *p, int offset)
 {
-	unsigned int tmp;
-	offset = map_8250_in_reg(up, offset) << up->port.regshift;
-
-	switch (up->port.iotype) {
-	case UPIO_HUB6:
-		outb(up->port.hub6 - 1 + offset, up->port.iobase);
-		return inb(up->port.iobase + 1);
-
-	case UPIO_MEM:
-	case UPIO_DWAPB:
-		return readb(up->port.membase + offset);
-
-	case UPIO_RM9000:
-	case UPIO_MEM32:
-		return readl(up->port.membase + offset);
-
-#ifdef CONFIG_SERIAL_8250_AU1X00
-	case UPIO_AU:
-		return __raw_readl(up->port.membase + offset);
-#endif
-
-	case UPIO_TSI:
-		if (offset == UART_IIR) {
-			tmp = readl(up->port.membase + (UART_IIR & ~3));
-			return (tmp >> 16) & 0xff; /* UART_IIR % 4 == 2 */
-		} else
-			return readb(up->port.membase + offset);
-
-	default:
-		return inb(up->port.iobase + offset);
-	}
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	outb(p->hub6 - 1 + offset, p->iobase);
+	return inb(p->iobase + 1);
 }
 
-static void
-serial_out(struct uart_8250_port *up, int offset, int value)
+static void hub6_serial_out(struct uart_port *p, int offset, int value)
 {
-	/* Save the offset before it's remapped */
-	int save_offset = offset;
-	offset = map_8250_out_reg(up, offset) << up->port.regshift;
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	outb(p->hub6 - 1 + offset, p->iobase);
+	outb(value, p->iobase + 1);
+}
 
-	switch (up->port.iotype) {
+static unsigned int mem_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return readb(p->membase + offset);
+}
+
+static void mem_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	writeb(value, p->membase + offset);
+}
+
+static void mem32_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	writel(value, p->membase + offset);
+}
+
+static unsigned int mem32_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return readl(p->membase + offset);
+}
+
+#ifdef CONFIG_SERIAL_8250_AU1X00
+static unsigned int au_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return __raw_readl(p->membase + offset);
+}
+
+static void au_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	__raw_writel(value, p->membase + offset);
+}
+#endif
+
+static unsigned int tsi_serial_in(struct uart_port *p, int offset)
+{
+	unsigned int tmp;
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	if (offset == UART_IIR) {
+		tmp = readl(p->membase + (UART_IIR & ~3));
+		return (tmp >> 16) & 0xff; /* UART_IIR % 4 == 2 */
+	} else
+		return readb(p->membase + offset);
+}
+
+static void tsi_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	if (!((offset == UART_IER) && (value & UART_IER_UUE)))
+		writeb(value, p->membase + offset);
+}
+
+static void dwapb_serial_out(struct uart_port *p, int offset, int value)
+{
+	int save_offset = offset;
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	/* Save the LCR value so it can be re-written when a
+	 * Busy Detect interrupt occurs. */
+	if (save_offset == UART_LCR) {
+		struct uart_8250_port *up = (struct uart_8250_port *)p;
+		up->lcr = value;
+	}
+	writeb(value, p->membase + offset);
+	/* Read the IER to ensure any interrupt is cleared before
+	 * returning from ISR. */
+	if (save_offset == UART_TX || save_offset == UART_IER)
+		value = p->serial_in(p, UART_IER);
+}
+
+static unsigned int io_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return inb(p->iobase + offset);
+}
+
+static void io_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	outb(value, p->iobase + offset);
+}
+
+static void set_io_from_upio(struct uart_port *p)
+{
+	switch (p->iotype) {
 	case UPIO_HUB6:
-		outb(up->port.hub6 - 1 + offset, up->port.iobase);
-		outb(value, up->port.iobase + 1);
+		p->serial_in = hub6_serial_in;
+		p->serial_out = hub6_serial_out;
 		break;
 
 	case UPIO_MEM:
-		writeb(value, up->port.membase + offset);
+		p->serial_in = mem_serial_in;
+		p->serial_out = mem_serial_out;
 		break;
 
 	case UPIO_RM9000:
 	case UPIO_MEM32:
-		writel(value, up->port.membase + offset);
+		p->serial_in = mem32_serial_in;
+		p->serial_out = mem32_serial_out;
 		break;
 
 #ifdef CONFIG_SERIAL_8250_AU1X00
 	case UPIO_AU:
-		__raw_writel(value, up->port.membase + offset);
+		p->serial_in = au_serial_in;
+		p->serial_out = au_serial_out;
 		break;
 #endif
 	case UPIO_TSI:
-		if (!((offset == UART_IER) && (value & UART_IER_UUE)))
-			writeb(value, up->port.membase + offset);
+		p->serial_in = tsi_serial_in;
+		p->serial_out = tsi_serial_out;
 		break;
 
 	case UPIO_DWAPB:
-		/* Save the LCR value so it can be re-written when a
-		 * Busy Detect interrupt occurs. */
-		if (save_offset == UART_LCR)
-			up->lcr = value;
-		writeb(value, up->port.membase + offset);
-		/* Read the IER to ensure any interrupt is cleared before
-		 * returning from ISR. */
-		if (save_offset == UART_TX || save_offset == UART_IER)
-			value = serial_in(up, UART_IER);
+		p->serial_in = mem_serial_in;
+		p->serial_out = dwapb_serial_out;
 		break;
 
 	default:
-		outb(value, up->port.iobase + offset);
+		p->serial_in = io_serial_in;
+		p->serial_out = io_serial_out;
+		break;
 	}
 }
 
 static void
 serial_out_sync(struct uart_8250_port *up, int offset, int value)
 {
-	switch (up->port.iotype) {
+	struct uart_port *p = &up->port;
+	switch (p->iotype) {
 	case UPIO_MEM:
 	case UPIO_MEM32:
 #ifdef CONFIG_SERIAL_8250_AU1X00
 	case UPIO_AU:
 #endif
 	case UPIO_DWAPB:
-		serial_out(up, offset, value);
-		serial_in(up, UART_LCR);	/* safe, no side-effects */
+		p->serial_out(p, offset, value);
+		p->serial_in(p, UART_LCR);	/* safe, no side-effects */
 		break;
 	default:
-		serial_out(up, offset, value);
+		p->serial_out(p, offset, value);
 	}
 }
 
+#define serial_in(up, offset)		\
+	(up->port.serial_in(&(up)->port, (offset)))
+#define serial_out(up, offset, value)	\
+	(up->port.serial_out(&(up)->port, (offset), (value)))
 /*
  * We used to support using pause I/O for certain machines.  We
  * haven't supported this for a while, but just in case it's badly
@@ -2576,6 +2645,7 @@
 		up->port.membase  = old_serial_port[i].iomem_base;
 		up->port.iotype   = old_serial_port[i].io_type;
 		up->port.regshift = old_serial_port[i].iomem_reg_shift;
+		set_io_from_upio(&up->port);
 		if (share_irqs)
 			up->port.flags |= UPF_SHARE_IRQ;
 	}
@@ -2752,12 +2822,30 @@
  */
 int __init early_serial_setup(struct uart_port *port)
 {
+	struct uart_port *p;
+
 	if (port->line >= ARRAY_SIZE(serial8250_ports))
 		return -ENODEV;
 
 	serial8250_isa_init_ports();
-	serial8250_ports[port->line].port	= *port;
-	serial8250_ports[port->line].port.ops	= &serial8250_pops;
+	p = &serial8250_ports[port->line].port;
+	p->iobase       = port->iobase;
+	p->membase      = port->membase;
+	p->irq          = port->irq;
+	p->uartclk      = port->uartclk;
+	p->fifosize     = port->fifosize;
+	p->regshift     = port->regshift;
+	p->iotype       = port->iotype;
+	p->flags        = port->flags;
+	p->mapbase      = port->mapbase;
+	p->private_data = port->private_data;
+
+	set_io_from_upio(p);
+	if (port->serial_in)
+		p->serial_in = port->serial_in;
+	if (port->serial_out)
+		p->serial_out = port->serial_out;
+
 	return 0;
 }
 
@@ -2822,6 +2910,9 @@
 		port.mapbase		= p->mapbase;
 		port.hub6		= p->hub6;
 		port.private_data	= p->private_data;
+		port.type		= p->type;
+		port.serial_in		= p->serial_in;
+		port.serial_out		= p->serial_out;
 		port.dev		= &dev->dev;
 		if (share_irqs)
 			port.flags |= UPF_SHARE_IRQ;
@@ -2976,6 +3067,20 @@
 		if (port->dev)
 			uart->port.dev = port->dev;
 
+		if (port->flags & UPF_FIXED_TYPE) {
+			uart->port.type = port->type;
+			uart->port.fifosize = uart_config[port->type].fifo_size;
+			uart->capabilities = uart_config[port->type].flags;
+			uart->tx_loadsz = uart_config[port->type].tx_loadsz;
+		}
+
+		set_io_from_upio(&uart->port);
+		/* Possibly override default I/O functions.  */
+		if (port->serial_in)
+			uart->port.serial_in = port->serial_in;
+		if (port->serial_out)
+			uart->port.serial_out = port->serial_out;
+
 		ret = uart_add_one_port(&serial8250_reg, &uart->port);
 		if (ret == 0)
 			ret = uart->port.line;
diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index 5450a0e..c088146 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -42,7 +42,8 @@
 	u32	subvendor;
 	u32	subdevice;
 	int	(*init)(struct pci_dev *dev);
-	int	(*setup)(struct serial_private *, struct pciserial_board *,
+	int	(*setup)(struct serial_private *,
+			 const struct pciserial_board *,
 			 struct uart_port *, int);
 	void	(*exit)(struct pci_dev *dev);
 };
@@ -107,7 +108,7 @@
  * ADDI-DATA GmbH communication cards <info@addi-data.com>
  */
 static int addidata_apci7800_setup(struct serial_private *priv,
-				struct pciserial_board *board,
+				const struct pciserial_board *board,
 				struct uart_port *port, int idx)
 {
 	unsigned int bar = 0, offset = board->first_offset;
@@ -134,7 +135,7 @@
  * Not that ugly ;) -- HW
  */
 static int
-afavlab_setup(struct serial_private *priv, struct pciserial_board *board,
+afavlab_setup(struct serial_private *priv, const struct pciserial_board *board,
 	      struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset;
@@ -188,8 +189,9 @@
  * some serial ports are supposed to be hidden on certain models.
  */
 static int
-pci_hp_diva_setup(struct serial_private *priv, struct pciserial_board *board,
-	      struct uart_port *port, int idx)
+pci_hp_diva_setup(struct serial_private *priv,
+		const struct pciserial_board *board,
+		struct uart_port *port, int idx)
 {
 	unsigned int offset = board->first_offset;
 	unsigned int bar = FL_GET_BASE(board->flags);
@@ -306,7 +308,7 @@
 
 /* SBS Technologies Inc. PMC-OCTPRO and P-OCTAL cards */
 static int
-sbs_setup(struct serial_private *priv, struct pciserial_board *board,
+sbs_setup(struct serial_private *priv, const struct pciserial_board *board,
 		struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset;
@@ -463,7 +465,7 @@
 }
 
 static int pci_siig_setup(struct serial_private *priv,
-			  struct pciserial_board *board,
+			  const struct pciserial_board *board,
 			  struct uart_port *port, int idx)
 {
 	unsigned int bar = FL_GET_BASE(board->flags) + idx, offset = 0;
@@ -534,7 +536,8 @@
  * Ugh, this is ugly as all hell --- TYT
  */
 static int
-pci_timedia_setup(struct serial_private *priv, struct pciserial_board *board,
+pci_timedia_setup(struct serial_private *priv,
+		  const struct pciserial_board *board,
 		  struct uart_port *port, int idx)
 {
 	unsigned int bar = 0, offset = board->first_offset;
@@ -568,7 +571,7 @@
  */
 static int
 titan_400l_800l_setup(struct serial_private *priv,
-		      struct pciserial_board *board,
+		      const struct pciserial_board *board,
 		      struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset;
@@ -737,8 +740,41 @@
 	release_region(ioport, ITE_887x_IOSIZE);
 }
 
+/*
+ * Oxford Semiconductor Inc.
+ * Check that device is part of the Tornado range of devices, then determine
+ * the number of ports available on the device.
+ */
+static int pci_oxsemi_tornado_init(struct pci_dev *dev)
+{
+	u8 __iomem *p;
+	unsigned long deviceID;
+	unsigned int  number_uarts = 0;
+
+	/* OxSemi Tornado devices are all 0xCxxx */
+	if (dev->vendor == PCI_VENDOR_ID_OXSEMI &&
+	    (dev->device & 0xF000) != 0xC000)
+		return 0;
+
+	p = pci_iomap(dev, 0, 5);
+	if (p == NULL)
+		return -ENOMEM;
+
+	deviceID = ioread32(p);
+	/* Tornado device */
+	if (deviceID == 0x07000200) {
+		number_uarts = ioread8(p + 4);
+		printk(KERN_DEBUG
+			"%d ports detected on Oxford PCI Express device\n",
+								number_uarts);
+	}
+	pci_iounmap(dev, p);
+	return number_uarts;
+}
+
 static int
-pci_default_setup(struct serial_private *priv, struct pciserial_board *board,
+pci_default_setup(struct serial_private *priv,
+		  const struct pciserial_board *board,
 		  struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset, maxnr;
@@ -1018,6 +1054,25 @@
 		.setup		= pci_default_setup,
 	},
 	/*
+	 * For Oxford Semiconductor and Mainpine
+	 */
+	{
+		.vendor		= PCI_VENDOR_ID_OXSEMI,
+		.device		= PCI_ANY_ID,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.init		= pci_oxsemi_tornado_init,
+		.setup		= pci_default_setup,
+	},
+	{
+		.vendor		= PCI_VENDOR_ID_MAINPINE,
+		.device		= PCI_ANY_ID,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.init		= pci_oxsemi_tornado_init,
+		.setup		= pci_default_setup,
+	},
+	/*
 	 * Default "match everything" terminator entry
 	 */
 	{
@@ -1048,7 +1103,7 @@
 }
 
 static inline int get_pci_irq(struct pci_dev *dev,
-				struct pciserial_board *board)
+				const struct pciserial_board *board)
 {
 	if (board->flags & FL_NOIRQ)
 		return 0;
@@ -1843,8 +1898,8 @@
 }
 
 static inline int
-serial_pci_matches(struct pciserial_board *board,
-		   struct pciserial_board *guessed)
+serial_pci_matches(const struct pciserial_board *board,
+		   const struct pciserial_board *guessed)
 {
 	return
 	    board->num_ports == guessed->num_ports &&
@@ -1854,54 +1909,14 @@
 	    board->first_offset == guessed->first_offset;
 }
 
-/*
- * Oxford Semiconductor Inc.
- * Check that device is part of the Tornado range of devices, then determine
- * the number of ports available on the device.
- */
-static int pci_oxsemi_tornado_init(struct pci_dev *dev, struct pciserial_board *board)
-{
-	u8 __iomem *p;
-	unsigned long deviceID;
-	unsigned int  number_uarts;
-
-	/* OxSemi Tornado devices are all 0xCxxx */
-	if (dev->vendor == PCI_VENDOR_ID_OXSEMI &&
-	    (dev->device & 0xF000) != 0xC000)
-		return 0;
-
-	p = pci_iomap(dev, 0, 5);
-	if (p == NULL)
-		return -ENOMEM;
-
-	deviceID = ioread32(p);
-	/* Tornado device */
-	if (deviceID == 0x07000200) {
-		number_uarts = ioread8(p + 4);
-		board->num_ports = number_uarts;
-		printk(KERN_DEBUG
-			"%d ports detected on Oxford PCI Express device\n",
-								number_uarts);
-	}
-	pci_iounmap(dev, p);
-	return 0;
-}
-
 struct serial_private *
-pciserial_init_ports(struct pci_dev *dev, struct pciserial_board *board)
+pciserial_init_ports(struct pci_dev *dev, const struct pciserial_board *board)
 {
 	struct uart_port serial_port;
 	struct serial_private *priv;
 	struct pci_serial_quirk *quirk;
 	int rc, nr_ports, i;
 
-	/*
-	 * Find number of ports on board
-	 */
-	if (dev->vendor == PCI_VENDOR_ID_OXSEMI ||
-	    dev->vendor == PCI_VENDOR_ID_MAINPINE)
-		pci_oxsemi_tornado_init(dev, board);
-
 	nr_ports = board->num_ports;
 
 	/*
@@ -2028,7 +2043,8 @@
 pciserial_init_one(struct pci_dev *dev, const struct pci_device_id *ent)
 {
 	struct serial_private *priv;
-	struct pciserial_board *board, tmp;
+	const struct pciserial_board *board;
+	struct pciserial_board tmp;
 	int rc;
 
 	if (ent->driver_data >= ARRAY_SIZE(pci_boards)) {
@@ -2055,7 +2071,7 @@
 		 * We matched one of our class entries.  Try to
 		 * determine the parameters of this board.
 		 */
-		rc = serial_pci_guess_board(dev, board);
+		rc = serial_pci_guess_board(dev, &tmp);
 		if (rc)
 			goto disable;
 	} else {
@@ -2271,6 +2287,9 @@
 	{	PCI_VENDOR_ID_SEALEVEL, PCI_DEVICE_ID_SEALEVEL_COMM8,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b2_8_115200 },
+	{	PCI_VENDOR_ID_SEALEVEL, PCI_DEVICE_ID_SEALEVEL_7803,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
+		pbn_b2_8_460800 },
 	{	PCI_VENDOR_ID_SEALEVEL, PCI_DEVICE_ID_SEALEVEL_UCOMM8,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b2_8_115200 },
@@ -2372,6 +2391,9 @@
 		 * For now just used the hex ID 0x950a.
 		 */
 	{	PCI_VENDOR_ID_OXSEMI, 0x950a,
+		PCI_SUBVENDOR_ID_SIIG, PCI_SUBDEVICE_ID_SIIG_DUAL_SERIAL, 0, 0,
+		pbn_b0_2_115200 },
+	{	PCI_VENDOR_ID_OXSEMI, 0x950a,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b0_2_1130000 },
 	{	PCI_VENDOR_ID_OXSEMI, PCI_DEVICE_ID_OXSEMI_16PCI954,
diff --git a/drivers/serial/bfin_5xx.c b/drivers/serial/bfin_5xx.c
index 569f0e2..318d69d 100644
--- a/drivers/serial/bfin_5xx.c
+++ b/drivers/serial/bfin_5xx.c
@@ -22,7 +22,8 @@
 #include <linux/tty_flip.h>
 #include <linux/serial_core.h>
 
-#ifdef CONFIG_KGDB_UART
+#if defined(CONFIG_KGDB_SERIAL_CONSOLE) || \
+	defined(CONFIG_KGDB_SERIAL_CONSOLE_MODULE)
 #include <linux/kgdb.h>
 #include <asm/irq_regs.h>
 #endif
@@ -45,6 +46,16 @@
 static struct bfin_serial_port bfin_serial_ports[BFIN_UART_NR_PORTS];
 static int nr_active_ports = ARRAY_SIZE(bfin_serial_resource);
 
+#if defined(CONFIG_KGDB_SERIAL_CONSOLE) || \
+	defined(CONFIG_KGDB_SERIAL_CONSOLE_MODULE)
+
+# ifndef CONFIG_SERIAL_BFIN_PIO
+#  error KGDB only support UART in PIO mode.
+# endif
+
+static int kgdboc_port_line;
+static int kgdboc_break_enabled;
+#endif
 /*
  * Setup for console. Argument comes from the menuconfig
  */
@@ -62,13 +73,17 @@
 
 static void bfin_serial_mctrl_check(struct bfin_serial_port *uart);
 
+static void bfin_serial_reset_irda(struct uart_port *port);
+
 /*
  * interrupts are disabled on entry
  */
 static void bfin_serial_stop_tx(struct uart_port *port)
 {
 	struct bfin_serial_port *uart = (struct bfin_serial_port *)port;
+#ifdef CONFIG_SERIAL_BFIN_DMA
 	struct circ_buf *xmit = &uart->port.info->xmit;
+#endif
 
 	while (!(UART_GET_LSR(uart) & TEMT))
 		cpu_relax();
@@ -94,6 +109,14 @@
 static void bfin_serial_start_tx(struct uart_port *port)
 {
 	struct bfin_serial_port *uart = (struct bfin_serial_port *)port;
+	struct tty_struct *tty = uart->port.info->port.tty;
+
+	/*
+	 * To avoid losting RX interrupt, we reset IR function
+	 * before sending data.
+	 */
+	if (tty->termios->c_line == N_IRDA)
+		bfin_serial_reset_irda(port);
 
 #ifdef CONFIG_SERIAL_BFIN_DMA
 	if (uart->tx_done)
@@ -110,9 +133,7 @@
 static void bfin_serial_stop_rx(struct uart_port *port)
 {
 	struct bfin_serial_port *uart = (struct bfin_serial_port *)port;
-#ifdef CONFIG_KGDB_UART
-	if (uart->port.line != CONFIG_KGDB_UART_PORT)
-#endif
+
 	UART_CLEAR_IER(uart, ERBFI);
 }
 
@@ -123,49 +144,6 @@
 {
 }
 
-#ifdef CONFIG_KGDB_UART
-static int kgdb_entry_state;
-
-void kgdb_put_debug_char(int chr)
-{
-	struct bfin_serial_port *uart;
-
-	if (CONFIG_KGDB_UART_PORT < 0
-		|| CONFIG_KGDB_UART_PORT >= BFIN_UART_NR_PORTS)
-		uart = &bfin_serial_ports[0];
-	else
-		uart = &bfin_serial_ports[CONFIG_KGDB_UART_PORT];
-
-	while (!(UART_GET_LSR(uart) & THRE)) {
-		SSYNC();
-	}
-
-	UART_CLEAR_DLAB(uart);
-	UART_PUT_CHAR(uart, (unsigned char)chr);
-	SSYNC();
-}
-
-int kgdb_get_debug_char(void)
-{
-	struct bfin_serial_port *uart;
-	unsigned char chr;
-
-	if (CONFIG_KGDB_UART_PORT < 0
-		|| CONFIG_KGDB_UART_PORT >= BFIN_UART_NR_PORTS)
-		uart = &bfin_serial_ports[0];
-	else
-		uart = &bfin_serial_ports[CONFIG_KGDB_UART_PORT];
-
-	while(!(UART_GET_LSR(uart) & DR)) {
-		SSYNC();
-	}
-	UART_CLEAR_DLAB(uart);
-	chr = UART_GET_CHAR(uart);
-	SSYNC();
-
-	return chr;
-}
-#endif
 
 #if ANOMALY_05000363 && defined(CONFIG_SERIAL_BFIN_PIO)
 # define UART_GET_ANOMALY_THRESHOLD(uart)    ((uart)->anomaly_threshold)
@@ -178,7 +156,7 @@
 #ifdef CONFIG_SERIAL_BFIN_PIO
 static void bfin_serial_rx_chars(struct bfin_serial_port *uart)
 {
-	struct tty_struct *tty = uart->port.info->port.tty;
+	struct tty_struct *tty = NULL;
 	unsigned int status, ch, flg;
 	static struct timeval anomaly_start = { .tv_sec = 0 };
 
@@ -188,27 +166,18 @@
  	ch = UART_GET_CHAR(uart);
  	uart->port.icount.rx++;
 
-#ifdef CONFIG_KGDB_UART
-	if (uart->port.line == CONFIG_KGDB_UART_PORT) {
-		struct pt_regs *regs = get_irq_regs();
-		if (uart->port.cons->index == CONFIG_KGDB_UART_PORT && ch == 0x1) { /* Ctrl + A */
-			kgdb_breakkey_pressed(regs);
+#if defined(CONFIG_KGDB_SERIAL_CONSOLE) || \
+	defined(CONFIG_KGDB_SERIAL_CONSOLE_MODULE)
+	if (kgdb_connected && kgdboc_port_line == uart->port.line)
+		if (ch == 0x3) {/* Ctrl + C */
+			kgdb_breakpoint();
 			return;
-		} else if (kgdb_entry_state == 0 && ch == '$') {/* connection from KGDB */
-			kgdb_entry_state = 1;
-		} else if (kgdb_entry_state == 1 && ch == 'q') {
-			kgdb_entry_state = 0;
-			kgdb_breakkey_pressed(regs);
-			return;
-		} else if (ch == 0x3) {/* Ctrl + C */
-			kgdb_entry_state = 0;
-			kgdb_breakkey_pressed(regs);
-			return;
-		} else {
-			kgdb_entry_state = 0;
 		}
-	}
+
+	if (!uart->port.info || !uart->port.info->tty)
+		return;
 #endif
+	tty = uart->port.info->tty;
 
 	if (ANOMALY_05000363) {
 		/* The BF533 (and BF561) family of processors have a nice anomaly
@@ -250,6 +219,7 @@
 			return;
 
  known_good_char:
+			status &= ~BI;
 			anomaly_start.tv_sec = 0;
 		}
 	}
@@ -445,7 +415,9 @@
 
 void bfin_serial_rx_dma_timeout(struct bfin_serial_port *uart)
 {
-	int x_pos, pos;
+	int x_pos, pos, flags;
+
+	spin_lock_irqsave(&uart->port.lock, flags);
 
 	uart->rx_dma_nrows = get_dma_curr_ycount(uart->rx_dma_channel);
 	x_pos = get_dma_curr_xcount(uart->rx_dma_channel);
@@ -463,6 +435,8 @@
 		uart->rx_dma_buf.tail = uart->rx_dma_buf.head;
 	}
 
+	spin_unlock_irqrestore(&uart->port.lock, flags);
+
 	mod_timer(&(uart->rx_dma_timer), jiffies + DMA_RX_FLUSH_JIFFIES);
 }
 
@@ -497,10 +471,9 @@
 	spin_lock(&uart->port.lock);
 	irqstat = get_dma_curr_irqstat(uart->rx_dma_channel);
 	clear_dma_irqstat(uart->rx_dma_channel);
+	bfin_serial_dma_rx_chars(uart);
 	spin_unlock(&uart->port.lock);
 
-	mod_timer(&(uart->rx_dma_timer), jiffies);
-
 	return IRQ_HANDLED;
 }
 #endif
@@ -630,16 +603,16 @@
 	uart->rx_dma_timer.expires = jiffies + DMA_RX_FLUSH_JIFFIES;
 	add_timer(&(uart->rx_dma_timer));
 #else
+#if defined(CONFIG_KGDB_SERIAL_CONSOLE) || \
+	defined(CONFIG_KGDB_SERIAL_CONSOLE_MODULE)
+	if (kgdboc_port_line == uart->port.line && kgdboc_break_enabled)
+		kgdboc_break_enabled = 0;
+	else {
+# endif
 	if (request_irq(uart->port.irq, bfin_serial_rx_int, IRQF_DISABLED,
 	     "BFIN_UART_RX", uart)) {
-# ifdef	CONFIG_KGDB_UART
-		if (uart->port.line != CONFIG_KGDB_UART_PORT) {
-# endif
 		printk(KERN_NOTICE "Unable to attach BlackFin UART RX interrupt\n");
 		return -EBUSY;
-# ifdef	CONFIG_KGDB_UART
-		}
-# endif
 	}
 
 	if (request_irq
@@ -685,6 +658,10 @@
 		}
 	}
 # endif
+#if defined(CONFIG_KGDB_SERIAL_CONSOLE) || \
+	defined(CONFIG_KGDB_SERIAL_CONSOLE_MODULE)
+	}
+# endif
 #endif
 	UART_SET_IER(uart, ERBFI);
 	return 0;
@@ -716,9 +693,6 @@
 		break;
 	};
 #endif
-#ifdef	CONFIG_KGDB_UART
-	if (uart->port.line != CONFIG_KGDB_UART_PORT)
-#endif
 	free_irq(uart->port.irq, uart);
 	free_irq(uart->port.irq+1, uart);
 #endif
@@ -887,6 +861,65 @@
 	}
 }
 
+#ifdef CONFIG_CONSOLE_POLL
+static void bfin_serial_poll_put_char(struct uart_port *port, unsigned char chr)
+{
+	struct bfin_serial_port *uart = (struct bfin_serial_port *)port;
+
+	while (!(UART_GET_LSR(uart) & THRE))
+		cpu_relax();
+
+	UART_CLEAR_DLAB(uart);
+	UART_PUT_CHAR(uart, (unsigned char)chr);
+}
+
+static int bfin_serial_poll_get_char(struct uart_port *port)
+{
+	struct bfin_serial_port *uart = (struct bfin_serial_port *)port;
+	unsigned char chr;
+
+	while (!(UART_GET_LSR(uart) & DR))
+		cpu_relax();
+
+	UART_CLEAR_DLAB(uart);
+	chr = UART_GET_CHAR(uart);
+
+	return chr;
+}
+#endif
+
+#if defined(CONFIG_KGDB_SERIAL_CONSOLE) || \
+	defined(CONFIG_KGDB_SERIAL_CONSOLE_MODULE)
+static void bfin_kgdboc_port_shutdown(struct uart_port *port)
+{
+	if (kgdboc_break_enabled) {
+		kgdboc_break_enabled = 0;
+		bfin_serial_shutdown(port);
+	}
+}
+
+static int bfin_kgdboc_port_startup(struct uart_port *port)
+{
+	kgdboc_port_line = port->line;
+	kgdboc_break_enabled = !bfin_serial_startup(port);
+	return 0;
+}
+#endif
+
+static void bfin_serial_reset_irda(struct uart_port *port)
+{
+	int line = port->line;
+	unsigned short val;
+
+	val = UART_GET_GCTL(&bfin_serial_ports[line]);
+	val &= ~(IREN | RPOLC);
+	UART_PUT_GCTL(&bfin_serial_ports[line], val);
+	SSYNC();
+	val |= (IREN | RPOLC);
+	UART_PUT_GCTL(&bfin_serial_ports[line], val);
+	SSYNC();
+}
+
 static struct uart_ops bfin_serial_pops = {
 	.tx_empty	= bfin_serial_tx_empty,
 	.set_mctrl	= bfin_serial_set_mctrl,
@@ -905,6 +938,15 @@
 	.request_port	= bfin_serial_request_port,
 	.config_port	= bfin_serial_config_port,
 	.verify_port	= bfin_serial_verify_port,
+#if defined(CONFIG_KGDB_SERIAL_CONSOLE) || \
+	defined(CONFIG_KGDB_SERIAL_CONSOLE_MODULE)
+	.kgdboc_port_startup	= bfin_kgdboc_port_startup,
+	.kgdboc_port_shutdown	= bfin_kgdboc_port_shutdown,
+#endif
+#ifdef CONFIG_CONSOLE_POLL
+	.poll_put_char	= bfin_serial_poll_put_char,
+	.poll_get_char	= bfin_serial_poll_get_char,
+#endif
 };
 
 static void __init bfin_serial_init_ports(void)
@@ -950,7 +992,7 @@
 
 }
 
-#ifdef CONFIG_SERIAL_BFIN_CONSOLE
+#if defined(CONFIG_SERIAL_BFIN_CONSOLE) || defined(CONFIG_EARLY_PRINTK)
 /*
  * If the port was already initialised (eg, by a boot loader),
  * try to determine the current setup.
@@ -994,24 +1036,20 @@
 	}
 	pr_debug("%s:baud = %d, parity = %c, bits= %d\n", __func__, *baud, *parity, *bits);
 }
-#endif
 
-#if defined(CONFIG_SERIAL_BFIN_CONSOLE) || defined(CONFIG_EARLY_PRINTK)
 static struct uart_driver bfin_serial_reg;
 
 static int __init
 bfin_serial_console_setup(struct console *co, char *options)
 {
 	struct bfin_serial_port *uart;
-# ifdef CONFIG_SERIAL_BFIN_CONSOLE
 	int baud = 57600;
 	int bits = 8;
 	int parity = 'n';
-#  ifdef CONFIG_SERIAL_BFIN_CTSRTS
+# ifdef CONFIG_SERIAL_BFIN_CTSRTS
 	int flow = 'r';
-#  else
+# else
 	int flow = 'n';
-#  endif
 # endif
 
 	/*
@@ -1023,16 +1061,12 @@
 		co->index = 0;
 	uart = &bfin_serial_ports[co->index];
 
-# ifdef CONFIG_SERIAL_BFIN_CONSOLE
 	if (options)
 		uart_parse_options(options, &baud, &parity, &bits, &flow);
 	else
 		bfin_serial_console_get_options(uart, &baud, &parity, &bits);
 
 	return uart_set_options(&uart->port, co, baud, parity, bits, flow);
-# else
-	return 0;
-# endif
 }
 #endif /* defined (CONFIG_SERIAL_BFIN_CONSOLE) ||
 				 defined (CONFIG_EARLY_PRINTK) */
@@ -1076,10 +1110,7 @@
 {
 	bfin_serial_init_ports();
 	register_console(&bfin_serial_console);
-#ifdef CONFIG_KGDB_UART
-	kgdb_entry_state = 0;
-	init_kgdb_uart();
-#endif
+
 	return 0;
 }
 console_initcall(bfin_serial_rs_console_init);
@@ -1144,7 +1175,7 @@
 	return &bfin_early_serial_console;
 }
 
-#endif /* CONFIG_SERIAL_BFIN_CONSOLE */
+#endif /* CONFIG_EARLY_PRINTK */
 
 static struct uart_driver bfin_serial_reg = {
 	.owner			= THIS_MODULE,
@@ -1235,10 +1266,6 @@
 static int __init bfin_serial_init(void)
 {
 	int ret;
-#ifdef CONFIG_KGDB_UART
-	struct bfin_serial_port *uart = &bfin_serial_ports[CONFIG_KGDB_UART_PORT];
-	struct ktermios t;
-#endif
 
 	pr_info("Serial: Blackfin serial driver\n");
 
@@ -1252,21 +1279,6 @@
 			uart_unregister_driver(&bfin_serial_reg);
 		}
 	}
-#ifdef CONFIG_KGDB_UART
-	if (uart->port.cons->index != CONFIG_KGDB_UART_PORT) {
-		request_irq(uart->port.irq, bfin_serial_rx_int,
-			IRQF_DISABLED, "BFIN_UART_RX", uart);
-		pr_info("Request irq for kgdb uart port\n");
-		UART_SET_IER(uart, ERBFI);
-		SSYNC();
-		t.c_cflag = CS8|B57600;
-		t.c_iflag = 0;
-		t.c_oflag = 0;
-		t.c_lflag = ICANON;
-		t.c_line = CONFIG_KGDB_UART_PORT;
-		bfin_serial_set_termios(&uart->port, &t, &t);
-	}
-#endif
 	return ret;
 }
 
@@ -1276,6 +1288,7 @@
 	uart_unregister_driver(&bfin_serial_reg);
 }
 
+
 module_init(bfin_serial_init);
 module_exit(bfin_serial_exit);
 
diff --git a/drivers/serial/bfin_sport_uart.c b/drivers/serial/bfin_sport_uart.c
index dd8564d..529c0ff 100644
--- a/drivers/serial/bfin_sport_uart.c
+++ b/drivers/serial/bfin_sport_uart.c
@@ -99,7 +99,7 @@
 
 static inline void tx_one_byte(struct sport_uart_port *up, unsigned int value)
 {
-	pr_debug("%s value:%x\n", __FUNCTION__, value);
+	pr_debug("%s value:%x\n", __func__, value);
 	/* Place a Start and Stop bit */
 	__asm__ volatile (
 		"R2 = b#01111111100;\n\t"
@@ -110,7 +110,7 @@
 		:"=r"(value)
 		:"0"(value)
 		:"R2", "R3");
-	pr_debug("%s value:%x\n", __FUNCTION__, value);
+	pr_debug("%s value:%x\n", __func__, value);
 
 	SPORT_PUT_TX(up, value);
 }
@@ -120,7 +120,7 @@
 	unsigned int value, extract;
 
 	value = SPORT_GET_RX32(up);
-	pr_debug("%s value:%x\n", __FUNCTION__, value);
+	pr_debug("%s value:%x\n", __func__, value);
 
 	/* Extract 8 bits data */
 	__asm__ volatile (
@@ -151,12 +151,12 @@
 	/* Set TCR1 and TCR2 */
 	SPORT_PUT_TCR1(up, (LTFS | ITFS | TFSR | TLSBIT | ITCLK));
 	SPORT_PUT_TCR2(up, 10);
-	pr_debug("%s TCR1:%x, TCR2:%x\n", __FUNCTION__, SPORT_GET_TCR1(up), SPORT_GET_TCR2(up));
+	pr_debug("%s TCR1:%x, TCR2:%x\n", __func__, SPORT_GET_TCR1(up), SPORT_GET_TCR2(up));
 
 	/* Set RCR1 and RCR2 */
 	SPORT_PUT_RCR1(up, (RCKFE | LARFS | LRFS | RFSR | IRCLK));
 	SPORT_PUT_RCR2(up, 28);
-	pr_debug("%s RCR1:%x, RCR2:%x\n", __FUNCTION__, SPORT_GET_RCR1(up), SPORT_GET_RCR2(up));
+	pr_debug("%s RCR1:%x, RCR2:%x\n", __func__, SPORT_GET_RCR1(up), SPORT_GET_RCR2(up));
 
 	tclkdiv = sclk/(2 * baud_rate) - 1;
 	tfsdiv = 12;
@@ -166,7 +166,7 @@
 	SPORT_PUT_RCLKDIV(up, rclkdiv);
 	SSYNC();
 	pr_debug("%s sclk:%d, baud_rate:%d, tclkdiv:%d, tfsdiv:%d, rclkdiv:%d\n",
-			__FUNCTION__, sclk, baud_rate, tclkdiv, tfsdiv, rclkdiv);
+			__func__, sclk, baud_rate, tclkdiv, tfsdiv, rclkdiv);
 
 	return 0;
 }
@@ -231,7 +231,7 @@
 	char buffer[20];
 	int retval;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	memset(buffer, 20, '\0');
 	snprintf(buffer, 20, "%s rx", up->name);
 	retval = request_irq(up->rx_irq, sport_uart_rx_irq, IRQF_SAMPLE_RANDOM, buffer, up);
@@ -320,7 +320,7 @@
 	unsigned int stat;
 
 	stat = SPORT_GET_STAT(up);
-	pr_debug("%s stat:%04x\n", __FUNCTION__, stat);
+	pr_debug("%s stat:%04x\n", __func__, stat);
 	if (stat & TXHRE) {
 		return TIOCSER_TEMT;
 	} else
@@ -329,13 +329,13 @@
 
 static unsigned int sport_get_mctrl(struct uart_port *port)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	return (TIOCM_CTS | TIOCM_CD | TIOCM_DSR);
 }
 
 static void sport_set_mctrl(struct uart_port *port, unsigned int mctrl)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 }
 
 static void sport_stop_tx(struct uart_port *port)
@@ -343,7 +343,7 @@
 	struct sport_uart_port *up = (struct sport_uart_port *)port;
 	unsigned int stat;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 
 	stat = SPORT_GET_STAT(up);
 	while(!(stat & TXHRE)) {
@@ -366,21 +366,21 @@
 {
 	struct sport_uart_port *up = (struct sport_uart_port *)port;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	/* Write data into SPORT FIFO before enable SPROT to transmit */
 	sport_uart_tx_chars(up);
 
 	/* Enable transmit, then an interrupt will generated */
 	SPORT_PUT_TCR1(up, (SPORT_GET_TCR1(up) | TSPEN));
 	SSYNC();
-	pr_debug("%s exit\n", __FUNCTION__);
+	pr_debug("%s exit\n", __func__);
 }
 
 static void sport_stop_rx(struct uart_port *port)
 {
 	struct sport_uart_port *up = (struct sport_uart_port *)port;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	/* Disable sport to stop rx */
 	SPORT_PUT_RCR1(up, (SPORT_GET_RCR1(up) & ~RSPEN));
 	SSYNC();
@@ -388,19 +388,19 @@
 
 static void sport_enable_ms(struct uart_port *port)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 }
 
 static void sport_break_ctl(struct uart_port *port, int break_state)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 }
 
 static void sport_shutdown(struct uart_port *port)
 {
 	struct sport_uart_port *up = (struct sport_uart_port *)port;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 
 	/* Disable sport */
 	SPORT_PUT_TCR1(up, (SPORT_GET_TCR1(up) & ~TSPEN));
@@ -421,7 +421,7 @@
 static void sport_set_termios(struct uart_port *port,
 		struct termios *termios, struct termios *old)
 {
-	pr_debug("%s enter, c_cflag:%08x\n", __FUNCTION__, termios->c_cflag);
+	pr_debug("%s enter, c_cflag:%08x\n", __func__, termios->c_cflag);
 	uart_update_timeout(port, CS8 ,port->uartclk);
 }
 
@@ -429,18 +429,18 @@
 {
 	struct sport_uart_port *up = (struct sport_uart_port *)port;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	return up->name;
 }
 
 static void sport_release_port(struct uart_port *port)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 }
 
 static int sport_request_port(struct uart_port *port)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	return 0;
 }
 
@@ -448,13 +448,13 @@
 {
 	struct sport_uart_port *up = (struct sport_uart_port *)port;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	up->port.type = PORT_BFIN_SPORT;
 }
 
 static int sport_verify_port(struct uart_port *port, struct serial_struct *ser)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	return 0;
 }
 
@@ -527,7 +527,7 @@
 {
 	struct sport_uart_port *sport = platform_get_drvdata(dev);
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	if (sport)
 		uart_suspend_port(&sport_uart_reg, &sport->port);
 
@@ -538,7 +538,7 @@
 {
 	struct sport_uart_port *sport = platform_get_drvdata(dev);
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	if (sport)
 		uart_resume_port(&sport_uart_reg, &sport->port);
 
@@ -547,7 +547,7 @@
 
 static int sport_uart_probe(struct platform_device *dev)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	sport_uart_ports[dev->id].port.dev = &dev->dev;
 	uart_add_one_port(&sport_uart_reg, &sport_uart_ports[dev->id].port);
 	platform_set_drvdata(dev, &sport_uart_ports[dev->id]);
@@ -559,7 +559,7 @@
 {
 	struct sport_uart_port *sport = platform_get_drvdata(dev);
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	platform_set_drvdata(dev, NULL);
 
 	if (sport)
@@ -582,7 +582,7 @@
 {
 	int ret;
 
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	ret = uart_register_driver(&sport_uart_reg);
 	if (ret != 0) {
 		printk(KERN_ERR "Failed to register %s:%d\n",
@@ -597,13 +597,13 @@
 	}
 
 
-	pr_debug("%s exit\n", __FUNCTION__);
+	pr_debug("%s exit\n", __func__);
 	return ret;
 }
 
 static void __exit sport_uart_exit(void)
 {
-	pr_debug("%s enter\n", __FUNCTION__);
+	pr_debug("%s enter\n", __func__);
 	platform_driver_unregister(&sport_uart_driver);
 	uart_unregister_driver(&sport_uart_reg);
 }
diff --git a/drivers/serial/jsm/jsm_tty.c b/drivers/serial/jsm/jsm_tty.c
index a697914..3547558 100644
--- a/drivers/serial/jsm/jsm_tty.c
+++ b/drivers/serial/jsm/jsm_tty.c
@@ -272,7 +272,7 @@
 	jsm_printk(CLOSE, INFO, &channel->ch_bd->pci_dev, "start\n");
 
 	bd = channel->ch_bd;
-	ts = channel->uart_port.info->port.tty->termios;
+	ts = port->info->port.tty->termios;
 
 	channel->ch_flags &= ~(CH_STOPI);
 
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 874786a..dc68b7e 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -50,7 +50,7 @@
 
 #define HIGH_BITS_OFFSET	((sizeof(long)-sizeof(int))*8)
 
-#define uart_users(state)	((state)->count + ((state)->info ? (state)->info->port.blocked_open : 0))
+#define uart_users(state)	((state)->count + (state)->info.port.blocked_open)
 
 #ifdef CONFIG_SERIAL_CORE_CONSOLE
 #define uart_console(port)	((port)->cons && (port)->cons->index == (port)->line)
@@ -94,7 +94,7 @@
 	struct uart_state *state = tty->driver_data;
 	struct uart_port *port = state->port;
 
-	if (!uart_circ_empty(&state->info->xmit) && state->info->xmit.buf &&
+	if (!uart_circ_empty(&state->info.xmit) && state->info.xmit.buf &&
 	    !tty->stopped && !tty->hw_stopped)
 		port->ops->start_tx(port);
 }
@@ -113,7 +113,7 @@
 static void uart_tasklet_action(unsigned long data)
 {
 	struct uart_state *state = (struct uart_state *)data;
-	tty_wakeup(state->info->port.tty);
+	tty_wakeup(state->info.port.tty);
 }
 
 static inline void
@@ -139,7 +139,7 @@
  */
 static int uart_startup(struct uart_state *state, int init_hw)
 {
-	struct uart_info *info = state->info;
+	struct uart_info *info = &state->info;
 	struct uart_port *port = state->port;
 	unsigned long page;
 	int retval = 0;
@@ -212,14 +212,15 @@
  */
 static void uart_shutdown(struct uart_state *state)
 {
-	struct uart_info *info = state->info;
+	struct uart_info *info = &state->info;
 	struct uart_port *port = state->port;
+	struct tty_struct *tty = info->port.tty;
 
 	/*
 	 * Set the TTY IO error marker
 	 */
-	if (info->port.tty)
-		set_bit(TTY_IO_ERROR, &info->port.tty->flags);
+	if (tty)
+		set_bit(TTY_IO_ERROR, &tty->flags);
 
 	if (info->flags & UIF_INITIALIZED) {
 		info->flags &= ~UIF_INITIALIZED;
@@ -227,7 +228,7 @@
 		/*
 		 * Turn off DTR and RTS early.
 		 */
-		if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL))
+		if (!tty || (tty->termios->c_cflag & HUPCL))
 			uart_clear_mctrl(port, TIOCM_DTR | TIOCM_RTS);
 
 		/*
@@ -427,7 +428,7 @@
 static void
 uart_change_speed(struct uart_state *state, struct ktermios *old_termios)
 {
-	struct tty_struct *tty = state->info->port.tty;
+	struct tty_struct *tty = state->info.port.tty;
 	struct uart_port *port = state->port;
 	struct ktermios *termios;
 
@@ -444,14 +445,14 @@
 	 * Set flags based on termios cflag
 	 */
 	if (termios->c_cflag & CRTSCTS)
-		state->info->flags |= UIF_CTS_FLOW;
+		state->info.flags |= UIF_CTS_FLOW;
 	else
-		state->info->flags &= ~UIF_CTS_FLOW;
+		state->info.flags &= ~UIF_CTS_FLOW;
 
 	if (termios->c_cflag & CLOCAL)
-		state->info->flags &= ~UIF_CHECK_CD;
+		state->info.flags &= ~UIF_CHECK_CD;
 	else
-		state->info->flags |= UIF_CHECK_CD;
+		state->info.flags |= UIF_CHECK_CD;
 
 	port->ops->set_termios(port, termios, old_termios);
 }
@@ -479,7 +480,7 @@
 {
 	struct uart_state *state = tty->driver_data;
 
-	return __uart_put_char(state->port, &state->info->xmit, ch);
+	return __uart_put_char(state->port, &state->info.xmit, ch);
 }
 
 static void uart_flush_chars(struct tty_struct *tty)
@@ -500,13 +501,13 @@
 	 * This means you called this function _after_ the port was
 	 * closed.  No cookie for you.
 	 */
-	if (!state || !state->info) {
+	if (!state) {
 		WARN_ON(1);
 		return -EL3HLT;
 	}
 
 	port = state->port;
-	circ = &state->info->xmit;
+	circ = &state->info.xmit;
 
 	if (!circ->buf)
 		return 0;
@@ -537,7 +538,7 @@
 	int ret;
 
 	spin_lock_irqsave(&state->port->lock, flags);
-	ret = uart_circ_chars_free(&state->info->xmit);
+	ret = uart_circ_chars_free(&state->info.xmit);
 	spin_unlock_irqrestore(&state->port->lock, flags);
 	return ret;
 }
@@ -549,7 +550,7 @@
 	int ret;
 
 	spin_lock_irqsave(&state->port->lock, flags);
-	ret = uart_circ_chars_pending(&state->info->xmit);
+	ret = uart_circ_chars_pending(&state->info.xmit);
 	spin_unlock_irqrestore(&state->port->lock, flags);
 	return ret;
 }
@@ -564,7 +565,7 @@
 	 * This means you called this function _after_ the port was
 	 * closed.  No cookie for you.
 	 */
-	if (!state || !state->info) {
+	if (!state) {
 		WARN_ON(1);
 		return;
 	}
@@ -573,7 +574,7 @@
 	pr_debug("uart_flush_buffer(%d) called\n", tty->index);
 
 	spin_lock_irqsave(&port->lock, flags);
-	uart_circ_clear(&state->info->xmit);
+	uart_circ_clear(&state->info.xmit);
 	if (port->ops->flush_buffer)
 		port->ops->flush_buffer(port);
 	spin_unlock_irqrestore(&port->lock, flags);
@@ -837,15 +838,15 @@
 	state->closing_wait    = closing_wait;
 	if (new_serial.xmit_fifo_size)
 		port->fifosize = new_serial.xmit_fifo_size;
-	if (state->info->port.tty)
-		state->info->port.tty->low_latency =
+	if (state->info.port.tty)
+		state->info.port.tty->low_latency =
 			(port->flags & UPF_LOW_LATENCY) ? 1 : 0;
 
  check_and_exit:
 	retval = 0;
 	if (port->type == PORT_UNKNOWN)
 		goto exit;
-	if (state->info->flags & UIF_INITIALIZED) {
+	if (state->info.flags & UIF_INITIALIZED) {
 		if (((old_flags ^ port->flags) & UPF_SPD_MASK) ||
 		    old_custom_divisor != port->custom_divisor) {
 			/*
@@ -858,7 +859,7 @@
 				printk(KERN_NOTICE
 				       "%s sets custom speed on %s. This "
 				       "is deprecated.\n", current->comm,
-				       tty_name(state->info->port.tty, buf));
+				       tty_name(state->info.port.tty, buf));
 			}
 			uart_change_speed(state, NULL);
 		}
@@ -889,8 +890,8 @@
 	 * interrupt happens).
 	 */
 	if (port->x_char ||
-	    ((uart_circ_chars_pending(&state->info->xmit) > 0) &&
-	     !state->info->port.tty->stopped && !state->info->port.tty->hw_stopped))
+	    ((uart_circ_chars_pending(&state->info.xmit) > 0) &&
+	     !state->info.port.tty->stopped && !state->info.port.tty->hw_stopped))
 		result &= ~TIOCSER_TEMT;
 
 	return put_user(result, value);
@@ -1017,7 +1018,7 @@
 	port->ops->enable_ms(port);
 	spin_unlock_irq(&port->lock);
 
-	add_wait_queue(&state->info->delta_msr_wait, &wait);
+	add_wait_queue(&state->info.delta_msr_wait, &wait);
 	for (;;) {
 		spin_lock_irq(&port->lock);
 		memcpy(&cnow, &port->icount, sizeof(struct uart_icount));
@@ -1045,7 +1046,7 @@
 	}
 
 	current->state = TASK_RUNNING;
-	remove_wait_queue(&state->info->delta_msr_wait, &wait);
+	remove_wait_queue(&state->info.delta_msr_wait, &wait);
 
 	return ret;
 }
@@ -1241,7 +1242,7 @@
 	 */
 	if (!(old_termios->c_cflag & CLOCAL) &&
 	    (tty->termios->c_cflag & CLOCAL))
-		wake_up_interruptible(&state->info->port.open_wait);
+		wake_up_interruptible(&info->port.open_wait);
 #endif
 }
 
@@ -1303,7 +1304,7 @@
 	 * At this point, we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts.
 	 */
-	if (state->info->flags & UIF_INITIALIZED) {
+	if (state->info.flags & UIF_INITIALIZED) {
 		unsigned long flags;
 		spin_lock_irqsave(&port->lock, flags);
 		port->ops->stop_rx(port);
@@ -1322,9 +1323,9 @@
 	tty_ldisc_flush(tty);
 
 	tty->closing = 0;
-	state->info->port.tty = NULL;
+	state->info.port.tty = NULL;
 
-	if (state->info->port.blocked_open) {
+	if (state->info.port.blocked_open) {
 		if (state->close_delay)
 			msleep_interruptible(state->close_delay);
 	} else if (!uart_console(port)) {
@@ -1334,8 +1335,8 @@
 	/*
 	 * Wake up anyone trying to open this port.
 	 */
-	state->info->flags &= ~UIF_NORMAL_ACTIVE;
-	wake_up_interruptible(&state->info->port.open_wait);
+	state->info.flags &= ~UIF_NORMAL_ACTIVE;
+	wake_up_interruptible(&state->info.port.open_wait);
 
  done:
 	mutex_unlock(&state->mutex);
@@ -1409,19 +1410,20 @@
 static void uart_hangup(struct tty_struct *tty)
 {
 	struct uart_state *state = tty->driver_data;
+	struct uart_info *info = &state->info;
 
 	BUG_ON(!kernel_locked());
 	pr_debug("uart_hangup(%d)\n", state->port->line);
 
 	mutex_lock(&state->mutex);
-	if (state->info && state->info->flags & UIF_NORMAL_ACTIVE) {
+	if (info->flags & UIF_NORMAL_ACTIVE) {
 		uart_flush_buffer(tty);
 		uart_shutdown(state);
 		state->count = 0;
-		state->info->flags &= ~UIF_NORMAL_ACTIVE;
-		state->info->port.tty = NULL;
-		wake_up_interruptible(&state->info->port.open_wait);
-		wake_up_interruptible(&state->info->delta_msr_wait);
+		info->flags &= ~UIF_NORMAL_ACTIVE;
+		info->port.tty = NULL;
+		wake_up_interruptible(&info->port.open_wait);
+		wake_up_interruptible(&info->delta_msr_wait);
 	}
 	mutex_unlock(&state->mutex);
 }
@@ -1434,7 +1436,7 @@
  */
 static void uart_update_termios(struct uart_state *state)
 {
-	struct tty_struct *tty = state->info->port.tty;
+	struct tty_struct *tty = state->info.port.tty;
 	struct uart_port *port = state->port;
 
 	if (uart_console(port) && port->cons->cflag) {
@@ -1469,7 +1471,7 @@
 uart_block_til_ready(struct file *filp, struct uart_state *state)
 {
 	DECLARE_WAITQUEUE(wait, current);
-	struct uart_info *info = state->info;
+	struct uart_info *info = &state->info;
 	struct uart_port *port = state->port;
 	unsigned int mctrl;
 
@@ -1563,28 +1565,6 @@
 		ret = -ENXIO;
 		goto err_unlock;
 	}
-
-	/* BKL: RACE HERE - LEAK */
-	/* We should move this into the uart_state structure and kill off
-	   this whole complexity */
-	if (!state->info) {
-		state->info = kzalloc(sizeof(struct uart_info), GFP_KERNEL);
-		if (state->info) {
-			init_waitqueue_head(&state->info->port.open_wait);
-			init_waitqueue_head(&state->info->delta_msr_wait);
-
-			/*
-			 * Link the info into the other structures.
-			 */
-			state->port->info = state->info;
-
-			tasklet_init(&state->info->tlet, uart_tasklet_action,
-				     (unsigned long)state);
-		} else {
-			ret = -ENOMEM;
-			goto err_unlock;
-		}
-	}
 	return state;
 
  err_unlock:
@@ -1641,9 +1621,10 @@
 	 * Any failures from here onwards should not touch the count.
 	 */
 	tty->driver_data = state;
+	state->port->info = &state->info;
 	tty->low_latency = (state->port->flags & UPF_LOW_LATENCY) ? 1 : 0;
 	tty->alt_speed = 0;
-	state->info->port.tty = tty;
+	state->info.port.tty = tty;
 
 	/*
 	 * If the port is in the middle of closing, bail out now.
@@ -1676,8 +1657,8 @@
 	/*
 	 * If this is the first open to succeed, adjust things to suit.
 	 */
-	if (retval == 0 && !(state->info->flags & UIF_NORMAL_ACTIVE)) {
-		state->info->flags |= UIF_NORMAL_ACTIVE;
+	if (retval == 0 && !(state->info.flags & UIF_NORMAL_ACTIVE)) {
+		state->info.flags |= UIF_NORMAL_ACTIVE;
 
 		uart_update_termios(state);
 	}
@@ -2028,11 +2009,11 @@
 	}
 	port->suspended = 1;
 
-	if (state->info && state->info->flags & UIF_INITIALIZED) {
+	if (state->info.flags & UIF_INITIALIZED) {
 		const struct uart_ops *ops = port->ops;
 		int tries;
 
-		state->info->flags = (state->info->flags & ~UIF_INITIALIZED)
+		state->info.flags = (state->info.flags & ~UIF_INITIALIZED)
 				     | UIF_SUSPENDED;
 
 		spin_lock_irq(&port->lock);
@@ -2107,15 +2088,15 @@
 		/*
 		 * If that's unset, use the tty termios setting.
 		 */
-		if (state->info && state->info->port.tty && termios.c_cflag == 0)
-			termios = *state->info->port.tty->termios;
+		if (state->info.port.tty && termios.c_cflag == 0)
+			termios = *state->info.port.tty->termios;
 
 		uart_change_pm(state, 0);
 		port->ops->set_termios(port, &termios, NULL);
 		console_start(port->cons);
 	}
 
-	if (state->info && state->info->flags & UIF_SUSPENDED) {
+	if (state->info.flags & UIF_SUSPENDED) {
 		const struct uart_ops *ops = port->ops;
 		int ret;
 
@@ -2130,7 +2111,7 @@
 			ops->set_mctrl(port, port->mctrl);
 			ops->start_tx(port);
 			spin_unlock_irq(&port->lock);
-			state->info->flags |= UIF_INITIALIZED;
+			state->info.flags |= UIF_INITIALIZED;
 		} else {
 			/*
 			 * Failed to resume - maybe hardware went away?
@@ -2140,7 +2121,7 @@
 			uart_shutdown(state);
 		}
 
-		state->info->flags &= ~UIF_SUSPENDED;
+		state->info.flags &= ~UIF_SUSPENDED;
 	}
 
 	mutex_unlock(&state->mutex);
@@ -2198,11 +2179,14 @@
 	 * Now do the auto configuration stuff.  Note that config_port
 	 * is expected to claim the resources and map the port for us.
 	 */
-	flags = UART_CONFIG_TYPE;
+	flags = 0;
 	if (port->flags & UPF_AUTO_IRQ)
 		flags |= UART_CONFIG_IRQ;
 	if (port->flags & UPF_BOOT_AUTOCONF) {
-		port->type = PORT_UNKNOWN;
+		if (!(port->flags & UPF_FIXED_TYPE)) {
+			port->type = PORT_UNKNOWN;
+			flags |= UART_CONFIG_TYPE;
+		}
 		port->ops->config_port(port, flags);
 	}
 
@@ -2383,8 +2367,12 @@
 
 		state->close_delay     = 500;	/* .5 seconds */
 		state->closing_wait    = 30000;	/* 30 seconds */
-
 		mutex_init(&state->mutex);
+
+		tty_port_init(&state->info.port);
+		init_waitqueue_head(&state->info.delta_msr_wait);
+		tasklet_init(&state->info.tlet, uart_tasklet_action,
+			     (unsigned long)state);
 	}
 
 	retval = tty_register_driver(normal);
@@ -2455,7 +2443,7 @@
 	state->pm_state = -1;
 
 	port->cons = drv->cons;
-	port->info = state->info;
+	port->info = &state->info;
 
 	/*
 	 * If this port is a console, then the spinlock is already
@@ -2527,18 +2515,11 @@
 	 */
 	tty_unregister_device(drv->tty_driver, port->line);
 
-	info = state->info;
+	info = &state->info;
 	if (info && info->port.tty)
 		tty_vhangup(info->port.tty);
 
 	/*
-	 * All users of this port should now be disconnected from
-	 * this driver, and the port shut down.  We should be the
-	 * only thread fiddling with this port from now on.
-	 */
-	state->info = NULL;
-
-	/*
 	 * Free the port IO and memory resources, if any.
 	 */
 	if (port->type != PORT_UNKNOWN)
@@ -2552,10 +2533,8 @@
 	/*
 	 * Kill the tasklet, and free resources.
 	 */
-	if (info) {
+	if (info)
 		tasklet_kill(&info->tlet);
-		kfree(info);
-	}
 
 	state->port = NULL;
 	mutex_unlock(&port_mutex);
diff --git a/drivers/usb/host/hwa-hc.c b/drivers/usb/host/hwa-hc.c
index 64be4d8..8582236 100644
--- a/drivers/usb/host/hwa-hc.c
+++ b/drivers/usb/host/hwa-hc.c
@@ -54,7 +54,6 @@
  *                      DWA).
  */
 #include <linux/kernel.h>
-#include <linux/version.h>
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/workqueue.h>
@@ -63,16 +62,12 @@
 #include "../wusbcore/wa-hc.h"
 #include "../wusbcore/wusbhc.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
 struct hwahc {
 	struct wusbhc wusbhc;	/* has to be 1st */
 	struct wahc wa;
-	u8 buffer[16];		/* for misc usb transactions */
 };
 
-/**
+/*
  * FIXME should be wusbhc
  *
  * NOTE: we need to cache the Cluster ID because later...there is no
@@ -126,7 +121,6 @@
 	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
 	struct device *dev = &hwahc->wa.usb_iface->dev;
 
-	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
 	mutex_lock(&wusbhc->mutex);
 	wa_nep_disarm(&hwahc->wa);
 	result = __wa_set_feature(&hwahc->wa, WA_RESET);
@@ -134,7 +128,6 @@
 		dev_err(dev, "error commanding HC to reset: %d\n", result);
 		goto error_unlock;
 	}
-	d_printf(3, dev, "reset: waiting for device to change state\n");
 	result = __wa_wait_status(&hwahc->wa, WA_STATUS_RESETTING, 0);
 	if (result < 0) {
 		dev_err(dev, "error waiting for HC to reset: %d\n", result);
@@ -142,7 +135,6 @@
 	}
 error_unlock:
 	mutex_unlock(&wusbhc->mutex);
-	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
 	return result;
 }
 
@@ -155,15 +147,9 @@
 	int result;
 	struct wusbhc *wusbhc = usb_hcd_to_wusbhc(usb_hcd);
 	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
-	struct device *dev = &hwahc->wa.usb_iface->dev;
 
-	/* Set up a Host Info WUSB Information Element */
-	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
 	result = -ENOSPC;
 	mutex_lock(&wusbhc->mutex);
-	/* Start the numbering from the top so that the bottom
-	 * range of the unauth addr space is used for devices,
-	 * the top for HCs; use 0xfe - RC# */
 	addr = wusb_cluster_id_get();
 	if (addr == 0)
 		goto error_cluster_id_get;
@@ -171,22 +157,14 @@
 	if (result < 0)
 		goto error_set_cluster_id;
 
-	result = wa_nep_arm(&hwahc->wa, GFP_KERNEL);
-	if (result < 0) {
-		dev_err(dev, "cannot listen to notifications: %d\n", result);
-		goto error_stop;
-	}
 	usb_hcd->uses_new_polling = 1;
 	usb_hcd->poll_rh = 1;
 	usb_hcd->state = HC_STATE_RUNNING;
 	result = 0;
 out:
 	mutex_unlock(&wusbhc->mutex);
-	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
 	return result;
 
-error_stop:
-	__wa_stop(&hwahc->wa);
 error_set_cluster_id:
 	wusb_cluster_id_put(wusbhc->cluster_id);
 error_cluster_id_get:
@@ -194,39 +172,6 @@
 
 }
 
-/*
- * FIXME: break this function up
- */
-static int __hwahc_op_wusbhc_start(struct wusbhc *wusbhc)
-{
-	int result;
-	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
-	struct device *dev = &hwahc->wa.usb_iface->dev;
-
-	/* Set up a Host Info WUSB Information Element */
-	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
-	result = -ENOSPC;
-
-	result = __wa_set_feature(&hwahc->wa, WA_ENABLE);
-	if (result < 0) {
-		dev_err(dev, "error commanding HC to start: %d\n", result);
-		goto error_stop;
-	}
-	result = __wa_wait_status(&hwahc->wa, WA_ENABLE, WA_ENABLE);
-	if (result < 0) {
-		dev_err(dev, "error waiting for HC to start: %d\n", result);
-		goto error_stop;
-	}
-	result = 0;
-out:
-	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
-	return result;
-
-error_stop:
-	result = __wa_clear_feature(&hwahc->wa, WA_ENABLE);
-	goto out;
-}
-
 static int hwahc_op_suspend(struct usb_hcd *usb_hcd, pm_message_t msg)
 {
 	struct wusbhc *wusbhc = usb_hcd_to_wusbhc(usb_hcd);
@@ -246,18 +191,6 @@
 	return -ENOSYS;
 }
 
-static void __hwahc_op_wusbhc_stop(struct wusbhc *wusbhc)
-{
-	int result;
-	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
-	struct device *dev = &hwahc->wa.usb_iface->dev;
-
-	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
-	/* Nothing for now */
-	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
-	return;
-}
-
 /*
  * No need to abort pipes, as when this is called, all the children
  * has been disconnected and that has done it [through
@@ -266,21 +199,11 @@
  */
 static void hwahc_op_stop(struct usb_hcd *usb_hcd)
 {
-	int result;
 	struct wusbhc *wusbhc = usb_hcd_to_wusbhc(usb_hcd);
-	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
-	struct wahc *wa = &hwahc->wa;
-	struct device *dev = &wa->usb_iface->dev;
 
-	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
 	mutex_lock(&wusbhc->mutex);
-	wusbhc_stop(wusbhc);
-	wa_nep_disarm(&hwahc->wa);
-	result = __wa_stop(&hwahc->wa);
 	wusb_cluster_id_put(wusbhc->cluster_id);
 	mutex_unlock(&wusbhc->mutex);
-	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
-	return;
 }
 
 static int hwahc_op_get_frame_number(struct usb_hcd *usb_hcd)
@@ -325,6 +248,54 @@
 	rpipe_ep_disable(&hwahc->wa, ep);
 }
 
+static int __hwahc_op_wusbhc_start(struct wusbhc *wusbhc)
+{
+	int result;
+	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
+	struct device *dev = &hwahc->wa.usb_iface->dev;
+
+	result = __wa_set_feature(&hwahc->wa, WA_ENABLE);
+	if (result < 0) {
+		dev_err(dev, "error commanding HC to start: %d\n", result);
+		goto error_stop;
+	}
+	result = __wa_wait_status(&hwahc->wa, WA_ENABLE, WA_ENABLE);
+	if (result < 0) {
+		dev_err(dev, "error waiting for HC to start: %d\n", result);
+		goto error_stop;
+	}
+	result = wa_nep_arm(&hwahc->wa, GFP_KERNEL);
+	if (result < 0) {
+		dev_err(dev, "cannot listen to notifications: %d\n", result);
+		goto error_stop;
+	}
+	return result;
+
+error_stop:
+	__wa_clear_feature(&hwahc->wa, WA_ENABLE);
+	return result;
+}
+
+static void __hwahc_op_wusbhc_stop(struct wusbhc *wusbhc, int delay)
+{
+	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
+	struct wahc *wa = &hwahc->wa;
+	u8 iface_no = wa->usb_iface->cur_altsetting->desc.bInterfaceNumber;
+	int ret;
+
+	ret = usb_control_msg(wa->usb_dev, usb_sndctrlpipe(wa->usb_dev, 0),
+			      WUSB_REQ_CHAN_STOP,
+			      USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_INTERFACE,
+			      delay * 1000,
+			      iface_no,
+			      NULL, 0, 1000 /* FIXME: arbitrary */);
+	if (ret == 0)
+		msleep(delay);
+
+	wa_nep_disarm(&hwahc->wa);
+	__wa_stop(&hwahc->wa);
+}
+
 /*
  * Set the UWB MAS allocation for the WUSB cluster
  *
@@ -581,11 +552,11 @@
 	itr_size = le16_to_cpu(usb_dev->actconfig->desc.wTotalLength);
 	while (itr_size >= sizeof(*hdr)) {
 		hdr = (struct usb_descriptor_header *) itr;
-		d_printf(3, dev, "Extra device descriptor: "
-			 "type %02x/%u bytes @ %zu (%zu left)\n",
-			 hdr->bDescriptorType, hdr->bLength,
-			 (itr - usb_dev->rawdescriptors[actconfig_idx]),
-			 itr_size);
+		dev_dbg(dev, "Extra device descriptor: "
+			"type %02x/%u bytes @ %zu (%zu left)\n",
+			hdr->bDescriptorType, hdr->bLength,
+			(itr - usb_dev->rawdescriptors[actconfig_idx]),
+			itr_size);
 		if (hdr->bDescriptorType == USB_DT_WIRE_ADAPTER)
 			goto found;
 		itr += hdr->bLength;
@@ -794,7 +765,6 @@
 {
 	struct wusbhc *wusbhc = &hwahc->wusbhc;
 
-	d_fnstart(1, NULL, "(hwahc %p)\n", hwahc);
 	mutex_lock(&wusbhc->mutex);
 	__wa_destroy(&hwahc->wa);
 	wusbhc_destroy(&hwahc->wusbhc);
@@ -804,7 +774,6 @@
 	usb_put_intf(hwahc->wa.usb_iface);
 	usb_put_dev(hwahc->wa.usb_dev);
 	mutex_unlock(&wusbhc->mutex);
-	d_fnend(1, NULL, "(hwahc %p) = void\n", hwahc);
 }
 
 static void hwahc_init(struct hwahc *hwahc)
@@ -821,7 +790,6 @@
 	struct hwahc *hwahc;
 	struct device *dev = &usb_iface->dev;
 
-	d_fnstart(4, dev, "(%p, %p)\n", usb_iface, id);
 	result = -ENOMEM;
 	usb_hcd = usb_create_hcd(&hwahc_hc_driver, &usb_iface->dev, "wusb-hwa");
 	if (usb_hcd == NULL) {
@@ -848,7 +816,6 @@
 		dev_err(dev, "Cannot setup phase B of WUSBHC: %d\n", result);
 		goto error_wusbhc_b_create;
 	}
-	d_fnend(4, dev, "(%p, %p) = 0\n", usb_iface, id);
 	return 0;
 
 error_wusbhc_b_create:
@@ -858,7 +825,6 @@
 error_hwahc_create:
 	usb_put_hcd(usb_hcd);
 error_alloc:
-	d_fnend(4, dev, "(%p, %p) = %d\n", usb_iface, id, result);
 	return result;
 }
 
@@ -872,16 +838,12 @@
 	wusbhc = usb_hcd_to_wusbhc(usb_hcd);
 	hwahc = container_of(wusbhc, struct hwahc, wusbhc);
 
-	d_fnstart(1, NULL, "(hwahc %p [usb_iface %p])\n", hwahc, usb_iface);
 	wusbhc_b_destroy(&hwahc->wusbhc);
 	usb_remove_hcd(usb_hcd);
 	hwahc_destroy(hwahc);
 	usb_put_hcd(usb_hcd);
-	d_fnend(1, NULL, "(hwahc %p [usb_iface %p]) = void\n", hwahc,
-		usb_iface);
 }
 
-/** USB device ID's that we handle */
 static struct usb_device_id hwahc_id_table[] = {
 	/* FIXME: use class labels for this */
 	{ USB_INTERFACE_INFO(0xe0, 0x02, 0x01), },
@@ -898,18 +860,7 @@
 
 static int __init hwahc_driver_init(void)
 {
-	int result;
-	result = usb_register(&hwahc_driver);
-	if (result < 0) {
-		printk(KERN_ERR "WA-CDS: Cannot register USB driver: %d\n",
-		       result);
-		goto error_usb_register;
-	}
-	return 0;
-
-error_usb_register:
-	return result;
-
+	return usb_register(&hwahc_driver);
 }
 module_init(hwahc_driver_init);
 
diff --git a/drivers/usb/host/whci/Kbuild b/drivers/usb/host/whci/Kbuild
index 26a3871..11e5040 100644
--- a/drivers/usb/host/whci/Kbuild
+++ b/drivers/usb/host/whci/Kbuild
@@ -2,6 +2,7 @@
 
 whci-hcd-y := \
 	asl.o	\
+	debug.o \
 	hcd.o 	\
 	hw.o	\
 	init.o	\
diff --git a/drivers/usb/host/whci/asl.c b/drivers/usb/host/whci/asl.c
index 4d7078e5..577c0d2 100644
--- a/drivers/usb/host/whci/asl.c
+++ b/drivers/usb/host/whci/asl.c
@@ -19,32 +19,11 @@
 #include <linux/dma-mapping.h>
 #include <linux/uwb/umc.h>
 #include <linux/usb.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 #include "../../wusbcore/wusbhc.h"
 
 #include "whcd.h"
 
-#if D_LOCAL >= 4
-static void dump_asl(struct whc *whc, const char *tag)
-{
-	struct device *dev = &whc->umc->dev;
-	struct whc_qset *qset;
-
-	d_printf(4, dev, "ASL %s\n", tag);
-
-	list_for_each_entry(qset, &whc->async_list, list_node) {
-		dump_qset(qset, dev);
-	}
-}
-#else
-static inline void dump_asl(struct whc *whc, const char *tag)
-{
-}
-#endif
-
-
 static void qset_get_next_prev(struct whc *whc, struct whc_qset *qset,
 			       struct whc_qset **next, struct whc_qset **prev)
 {
@@ -179,11 +158,26 @@
 		      1000, "stop ASL");
 }
 
+/**
+ * asl_update - request an ASL update and wait for the hardware to be synced
+ * @whc: the WHCI HC
+ * @wusbcmd: WUSBCMD value to start the update.
+ *
+ * If the WUSB HC is inactive (i.e., the ASL is stopped) then the
+ * update must be skipped as the hardware may not respond to update
+ * requests.
+ */
 void asl_update(struct whc *whc, uint32_t wusbcmd)
 {
-	whc_write_wusbcmd(whc, wusbcmd, wusbcmd);
-	wait_event(whc->async_list_wq,
-		   (le_readl(whc->base + WUSBCMD) & WUSBCMD_ASYNC_UPDATED) == 0);
+	struct wusbhc *wusbhc = &whc->wusbhc;
+
+	mutex_lock(&wusbhc->mutex);
+	if (wusbhc->active) {
+		whc_write_wusbcmd(whc, wusbcmd, wusbcmd);
+		wait_event(whc->async_list_wq,
+			   (le_readl(whc->base + WUSBCMD) & WUSBCMD_ASYNC_UPDATED) == 0);
+	}
+	mutex_unlock(&wusbhc->mutex);
 }
 
 /**
@@ -202,8 +196,6 @@
 
 	spin_lock_irq(&whc->lock);
 
-	dump_asl(whc, "before processing");
-
 	/*
 	 * Transerve the software list backwards so new qsets can be
 	 * safely inserted into the ASL without making it non-circular.
@@ -217,8 +209,6 @@
 		update |= process_qset(whc, qset);
 	}
 
-	dump_asl(whc, "after processing");
-
 	spin_unlock_irq(&whc->lock);
 
 	if (update) {
diff --git a/drivers/usb/host/whci/debug.c b/drivers/usb/host/whci/debug.c
new file mode 100644
index 0000000..cf2d459
--- /dev/null
+++ b/drivers/usb/host/whci/debug.c
@@ -0,0 +1,189 @@
+/*
+ * Wireless Host Controller (WHC) debug.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#include <linux/kernel.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#include "../../wusbcore/wusbhc.h"
+
+#include "whcd.h"
+
+struct whc_dbg {
+	struct dentry *di_f;
+	struct dentry *asl_f;
+	struct dentry *pzl_f;
+};
+
+void qset_print(struct seq_file *s, struct whc_qset *qset)
+{
+	struct whc_std *std;
+	struct urb *urb = NULL;
+	int i;
+
+	seq_printf(s, "qset %08x\n", (u32)qset->qset_dma);
+	seq_printf(s, "  -> %08x\n", (u32)qset->qh.link);
+	seq_printf(s, "  info: %08x %08x %08x\n",
+		qset->qh.info1, qset->qh.info2,  qset->qh.info3);
+	seq_printf(s, "  sts: %04x errs: %d\n", qset->qh.status, qset->qh.err_count);
+	seq_printf(s, "  TD: sts: %08x opts: %08x\n",
+		qset->qh.overlay.qtd.status, qset->qh.overlay.qtd.options);
+
+	for (i = 0; i < WHCI_QSET_TD_MAX; i++) {
+		seq_printf(s, "  %c%c TD[%d]: sts: %08x opts: %08x ptr: %08x\n",
+			i == qset->td_start ? 'S' : ' ',
+			i == qset->td_end ? 'E' : ' ',
+			i, qset->qtd[i].status, qset->qtd[i].options,
+			(u32)qset->qtd[i].page_list_ptr);
+	}
+	seq_printf(s, "  ntds: %d\n", qset->ntds);
+	list_for_each_entry(std, &qset->stds, list_node) {
+		if (urb != std->urb) {
+			urb = std->urb;
+			seq_printf(s, "  urb %p transferred: %d bytes\n", urb,
+				urb->actual_length);
+		}
+		if (std->qtd)
+			seq_printf(s, "    sTD[%td]: %zu bytes @ %08x\n",
+				std->qtd - &qset->qtd[0],
+				std->len, std->num_pointers ?
+				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
+		else
+			seq_printf(s, "    sTD[-]: %zd bytes @ %08x\n",
+				std->len, std->num_pointers ?
+				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
+	}
+}
+
+static int di_print(struct seq_file *s, void *p)
+{
+	struct whc *whc = s->private;
+	char buf[72];
+	int d;
+
+	for (d = 0; d < whc->n_devices; d++) {
+		struct di_buf_entry *di = &whc->di_buf[d];
+
+		bitmap_scnprintf(buf, sizeof(buf),
+				 (unsigned long *)di->availability_info, UWB_NUM_MAS);
+
+		seq_printf(s, "DI[%d]\n", d);
+		seq_printf(s, "  availability: %s\n", buf);
+		seq_printf(s, "  %c%c key idx: %d dev addr: %d\n",
+			   (di->addr_sec_info & WHC_DI_SECURE) ? 'S' : ' ',
+			   (di->addr_sec_info & WHC_DI_DISABLE) ? 'D' : ' ',
+			   (di->addr_sec_info & WHC_DI_KEY_IDX_MASK) >> 8,
+			   (di->addr_sec_info & WHC_DI_DEV_ADDR_MASK));
+	}
+	return 0;
+}
+
+static int asl_print(struct seq_file *s, void *p)
+{
+	struct whc *whc = s->private;
+	struct whc_qset *qset;
+
+	list_for_each_entry(qset, &whc->async_list, list_node) {
+		qset_print(s, qset);
+	}
+
+	return 0;
+}
+
+static int pzl_print(struct seq_file *s, void *p)
+{
+	struct whc *whc = s->private;
+	struct whc_qset *qset;
+	int period;
+
+	for (period = 0; period < 5; period++) {
+		seq_printf(s, "Period %d\n", period);
+		list_for_each_entry(qset, &whc->periodic_list[period], list_node) {
+			qset_print(s, qset);
+		}
+	}
+	return 0;
+}
+
+static int di_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, di_print, inode->i_private);
+}
+
+static int asl_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, asl_print, inode->i_private);
+}
+
+static int pzl_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, pzl_print, inode->i_private);
+}
+
+static struct file_operations di_fops = {
+	.open    = di_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+	.owner   = THIS_MODULE,
+};
+
+static struct file_operations asl_fops = {
+	.open    = asl_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+	.owner   = THIS_MODULE,
+};
+
+static struct file_operations pzl_fops = {
+	.open    = pzl_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+	.owner   = THIS_MODULE,
+};
+
+void whc_dbg_init(struct whc *whc)
+{
+	if (whc->wusbhc.pal.debugfs_dir == NULL)
+		return;
+
+	whc->dbg = kzalloc(sizeof(struct whc_dbg), GFP_KERNEL);
+	if (whc->dbg == NULL)
+		return;
+
+	whc->dbg->di_f = debugfs_create_file("di", 0444,
+					      whc->wusbhc.pal.debugfs_dir, whc,
+					      &di_fops);
+	whc->dbg->asl_f = debugfs_create_file("asl", 0444,
+					      whc->wusbhc.pal.debugfs_dir, whc,
+					      &asl_fops);
+	whc->dbg->pzl_f = debugfs_create_file("pzl", 0444,
+					      whc->wusbhc.pal.debugfs_dir, whc,
+					      &pzl_fops);
+}
+
+void whc_dbg_clean_up(struct whc *whc)
+{
+	if (whc->dbg) {
+		debugfs_remove(whc->dbg->pzl_f);
+		debugfs_remove(whc->dbg->asl_f);
+		debugfs_remove(whc->dbg->di_f);
+		kfree(whc->dbg);
+	}
+}
diff --git a/drivers/usb/host/whci/hcd.c b/drivers/usb/host/whci/hcd.c
index ef3ad4d..1569afd 100644
--- a/drivers/usb/host/whci/hcd.c
+++ b/drivers/usb/host/whci/hcd.c
@@ -15,7 +15,6 @@
  * You should have received a copy of the GNU General Public License
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/uwb/umc.h>
@@ -92,8 +91,6 @@
 
 	mutex_lock(&wusbhc->mutex);
 
-	wusbhc_stop(wusbhc);
-
 	/* stop HC */
 	le_writel(0, whc->base + WUSBINTR);
 	whc_write_wusbcmd(whc, WUSBCMD_RUN, 0);
@@ -276,6 +273,8 @@
 		goto error_wusbhc_b_create;
 	}
 
+	whc_dbg_init(whc);
+
 	return 0;
 
 error_wusbhc_b_create:
@@ -299,6 +298,7 @@
 	struct whc *whc = wusbhc_to_whc(wusbhc);
 
 	if (usb_hcd) {
+		whc_dbg_clean_up(whc);
 		wusbhc_b_destroy(wusbhc);
 		usb_remove_hcd(usb_hcd);
 		wusbhc_destroy(wusbhc);
diff --git a/drivers/usb/host/whci/hw.c b/drivers/usb/host/whci/hw.c
index ac86e59..d498e72 100644
--- a/drivers/usb/host/whci/hw.c
+++ b/drivers/usb/host/whci/hw.c
@@ -50,6 +50,7 @@
 	unsigned long flags;
 	dma_addr_t dma_addr;
 	int t;
+	int ret = 0;
 
 	mutex_lock(&whc->mutex);
 
@@ -61,7 +62,8 @@
 		dev_err(&whc->umc->dev, "generic command timeout (%04x/%04x)\n",
 			le_readl(whc->base + WUSBGENCMDSTS),
 			le_readl(whc->base + WUSBGENCMDPARAMS));
-		return -ETIMEDOUT;
+		ret = -ETIMEDOUT;
+		goto out;
 	}
 
 	if (addr) {
@@ -80,8 +82,8 @@
 		  whc->base + WUSBGENCMDSTS);
 
 	spin_unlock_irqrestore(&whc->lock, flags);
-
+out:
 	mutex_unlock(&whc->mutex);
 
-	return 0;
+	return ret;
 }
diff --git a/drivers/usb/host/whci/int.c b/drivers/usb/host/whci/int.c
index fce0117..6aae700 100644
--- a/drivers/usb/host/whci/int.c
+++ b/drivers/usb/host/whci/int.c
@@ -15,7 +15,6 @@
  * You should have received a copy of the GNU General Public License
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/uwb/umc.h>
diff --git a/drivers/usb/host/whci/pzl.c b/drivers/usb/host/whci/pzl.c
index 8d62df0..2ae5abf 100644
--- a/drivers/usb/host/whci/pzl.c
+++ b/drivers/usb/host/whci/pzl.c
@@ -19,35 +19,11 @@
 #include <linux/dma-mapping.h>
 #include <linux/uwb/umc.h>
 #include <linux/usb.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 #include "../../wusbcore/wusbhc.h"
 
 #include "whcd.h"
 
-#if D_LOCAL >= 4
-static void dump_pzl(struct whc *whc, const char *tag)
-{
-	struct device *dev = &whc->umc->dev;
-	struct whc_qset *qset;
-	int period = 0;
-
-	d_printf(4, dev, "PZL %s\n", tag);
-
-	for (period = 0; period < 5; period++) {
-		d_printf(4, dev, "Period %d\n", period);
-		list_for_each_entry(qset, &whc->periodic_list[period], list_node) {
-			dump_qset(qset, dev);
-		}
-	}
-}
-#else
-static inline void dump_pzl(struct whc *whc, const char *tag)
-{
-}
-#endif
-
 static void update_pzl_pointers(struct whc *whc, int period, u64 addr)
 {
 	switch (period) {
@@ -195,11 +171,26 @@
 		      1000, "stop PZL");
 }
 
+/**
+ * pzl_update - request a PZL update and wait for the hardware to be synced
+ * @whc: the WHCI HC
+ * @wusbcmd: WUSBCMD value to start the update.
+ *
+ * If the WUSB HC is inactive (i.e., the PZL is stopped) then the
+ * update must be skipped as the hardware may not respond to update
+ * requests.
+ */
 void pzl_update(struct whc *whc, uint32_t wusbcmd)
 {
-	whc_write_wusbcmd(whc, wusbcmd, wusbcmd);
-	wait_event(whc->periodic_list_wq,
-		   (le_readl(whc->base + WUSBCMD) & WUSBCMD_PERIODIC_UPDATED) == 0);
+	struct wusbhc *wusbhc = &whc->wusbhc;
+
+	mutex_lock(&wusbhc->mutex);
+	if (wusbhc->active) {
+		whc_write_wusbcmd(whc, wusbcmd, wusbcmd);
+		wait_event(whc->periodic_list_wq,
+			   (le_readl(whc->base + WUSBCMD) & WUSBCMD_PERIODIC_UPDATED) == 0);
+	}
+	mutex_unlock(&wusbhc->mutex);
 }
 
 static void update_pzl_hw_view(struct whc *whc)
@@ -235,8 +226,6 @@
 
 	spin_lock_irq(&whc->lock);
 
-	dump_pzl(whc, "before processing");
-
 	for (period = 4; period >= 0; period--) {
 		list_for_each_entry_safe(qset, t, &whc->periodic_list[period], list_node) {
 			if (!qset->in_hw_list)
@@ -248,8 +237,6 @@
 	if (update & (WHC_UPDATE_ADDED | WHC_UPDATE_REMOVED))
 		update_pzl_hw_view(whc);
 
-	dump_pzl(whc, "after processing");
-
 	spin_unlock_irq(&whc->lock);
 
 	if (update) {
diff --git a/drivers/usb/host/whci/qset.c b/drivers/usb/host/whci/qset.c
index 0420037..7be7431 100644
--- a/drivers/usb/host/whci/qset.c
+++ b/drivers/usb/host/whci/qset.c
@@ -24,46 +24,6 @@
 
 #include "whcd.h"
 
-void dump_qset(struct whc_qset *qset, struct device *dev)
-{
-	struct whc_std *std;
-	struct urb *urb = NULL;
-	int i;
-
-	dev_dbg(dev, "qset %08x\n", (u32)qset->qset_dma);
-	dev_dbg(dev, "  -> %08x\n", (u32)qset->qh.link);
-	dev_dbg(dev, "  info: %08x %08x %08x\n",
-		qset->qh.info1, qset->qh.info2,  qset->qh.info3);
-	dev_dbg(dev, "  sts: %04x errs: %d\n", qset->qh.status, qset->qh.err_count);
-	dev_dbg(dev, "  TD: sts: %08x opts: %08x\n",
-		qset->qh.overlay.qtd.status, qset->qh.overlay.qtd.options);
-
-	for (i = 0; i < WHCI_QSET_TD_MAX; i++) {
-		dev_dbg(dev, "  %c%c TD[%d]: sts: %08x opts: %08x ptr: %08x\n",
-			i == qset->td_start ? 'S' : ' ',
-			i == qset->td_end ? 'E' : ' ',
-			i, qset->qtd[i].status, qset->qtd[i].options,
-			(u32)qset->qtd[i].page_list_ptr);
-	}
-	dev_dbg(dev, "  ntds: %d\n", qset->ntds);
-	list_for_each_entry(std, &qset->stds, list_node) {
-		if (urb != std->urb) {
-			urb = std->urb;
-			dev_dbg(dev, "  urb %p transferred: %d bytes\n", urb,
-				urb->actual_length);
-		}
-		if (std->qtd)
-			dev_dbg(dev, "    sTD[%td]: %zu bytes @ %08x\n",
-				std->qtd - &qset->qtd[0],
-				std->len, std->num_pointers ?
-				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
-		else
-			dev_dbg(dev, "    sTD[-]: %zd bytes @ %08x\n",
-				std->len, std->num_pointers ?
-				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
-	}
-}
-
 struct whc_qset *qset_alloc(struct whc *whc, gfp_t mem_flags)
 {
 	struct whc_qset *qset;
diff --git a/drivers/usb/host/whci/whcd.h b/drivers/usb/host/whci/whcd.h
index 1d2a53b..0f3540f 100644
--- a/drivers/usb/host/whci/whcd.h
+++ b/drivers/usb/host/whci/whcd.h
@@ -21,6 +21,7 @@
 #define __WHCD_H
 
 #include <linux/uwb/whci.h>
+#include <linux/uwb/umc.h>
 #include <linux/workqueue.h>
 
 #include "whci-hc.h"
@@ -28,6 +29,7 @@
 /* Generic command timeout. */
 #define WHC_GENCMD_TIMEOUT_MS 100
 
+struct whc_dbg;
 
 struct whc {
 	struct wusbhc wusbhc;
@@ -69,6 +71,8 @@
 	struct list_head periodic_removed_list;
 	wait_queue_head_t periodic_list_wq;
 	struct work_struct periodic_work;
+
+	struct whc_dbg *dbg;
 };
 
 #define wusbhc_to_whc(w) (container_of((w), struct whc, wusbhc))
@@ -136,7 +140,7 @@
 
 /* wusb.c */
 int whc_wusbhc_start(struct wusbhc *wusbhc);
-void whc_wusbhc_stop(struct wusbhc *wusbhc);
+void whc_wusbhc_stop(struct wusbhc *wusbhc, int delay);
 int whc_mmcie_add(struct wusbhc *wusbhc, u8 interval, u8 repeat_cnt,
 		  u8 handle, struct wuie_hdr *wuie);
 int whc_mmcie_rm(struct wusbhc *wusbhc, u8 handle);
@@ -190,8 +194,11 @@
 				 struct whc_qtd *qtd);
 enum whc_update qset_add_qtds(struct whc *whc, struct whc_qset *qset);
 void qset_remove_complete(struct whc *whc, struct whc_qset *qset);
-void dump_qset(struct whc_qset *qset, struct device *dev);
 void pzl_update(struct whc *whc, uint32_t wusbcmd);
 void asl_update(struct whc *whc, uint32_t wusbcmd);
 
+/* debug.c */
+void whc_dbg_init(struct whc *whc);
+void whc_dbg_clean_up(struct whc *whc);
+
 #endif /* #ifndef __WHCD_H */
diff --git a/drivers/usb/host/whci/whci-hc.h b/drivers/usb/host/whci/whci-hc.h
index bff1eb7..51df7e3 100644
--- a/drivers/usb/host/whci/whci-hc.h
+++ b/drivers/usb/host/whci/whci-hc.h
@@ -410,6 +410,8 @@
 #  define WUSBDNTSCTRL_SLOTS(s)    ((s) << 0)
 
 #define WUSBTIME             0x68
+#  define WUSBTIME_CHANNEL_TIME_MASK 0x00ffffff
+
 #define WUSBBPST             0x6c
 #define WUSBDIBUPDATED       0x70
 
diff --git a/drivers/usb/host/whci/wusb.c b/drivers/usb/host/whci/wusb.c
index 66e4ddc..f24efde 100644
--- a/drivers/usb/host/whci/wusb.c
+++ b/drivers/usb/host/whci/wusb.c
@@ -15,47 +15,19 @@
  * You should have received a copy of the GNU General Public License
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/uwb/umc.h>
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
 
 #include "../../wusbcore/wusbhc.h"
 
 #include "whcd.h"
 
-#if D_LOCAL >= 1
-static void dump_di(struct whc *whc, int idx)
-{
-	struct di_buf_entry *di = &whc->di_buf[idx];
-	struct device *dev = &whc->umc->dev;
-	char buf[128];
-
-	bitmap_scnprintf(buf, sizeof(buf), (unsigned long *)di->availability_info, UWB_NUM_MAS);
-
-	d_printf(1, dev, "DI[%d]\n", idx);
-	d_printf(1, dev, "  availability: %s\n", buf);
-	d_printf(1, dev, "  %c%c key idx: %d dev addr: %d\n",
-		 (di->addr_sec_info & WHC_DI_SECURE) ? 'S' : ' ',
-		 (di->addr_sec_info & WHC_DI_DISABLE) ? 'D' : ' ',
-		 (di->addr_sec_info & WHC_DI_KEY_IDX_MASK) >> 8,
-		 (di->addr_sec_info & WHC_DI_DEV_ADDR_MASK));
-}
-#else
-static inline void dump_di(struct whc *whc, int idx)
-{
-}
-#endif
-
 static int whc_update_di(struct whc *whc, int idx)
 {
 	int offset = idx / 32;
 	u32 bit = 1 << (idx % 32);
 
-	dump_di(whc, idx);
-
 	le_writel(bit, whc->base + WUSBDIBUPDATED + offset);
 
 	return whci_wait_for(&whc->umc->dev,
@@ -64,8 +36,9 @@
 }
 
 /*
- * WHCI starts and stops MMCs based on there being a valid GTK so
- * these need only start/stop the asynchronous and periodic schedules.
+ * WHCI starts MMCs based on there being a valid GTK so these need
+ * only start/stop the asynchronous and periodic schedules and send a
+ * channel stop command.
  */
 
 int whc_wusbhc_start(struct wusbhc *wusbhc)
@@ -78,12 +51,20 @@
 	return 0;
 }
 
-void whc_wusbhc_stop(struct wusbhc *wusbhc)
+void whc_wusbhc_stop(struct wusbhc *wusbhc, int delay)
 {
 	struct whc *whc = wusbhc_to_whc(wusbhc);
+	u32 stop_time, now_time;
+	int ret;
 
 	pzl_stop(whc);
 	asl_stop(whc);
+
+	now_time = le_readl(whc->base + WUSBTIME) & WUSBTIME_CHANNEL_TIME_MASK;
+	stop_time = (now_time + ((delay * 8) << 7)) & 0x00ffffff;
+	ret = whc_do_gencmd(whc, WUSBGENCMDSTS_CHAN_STOP, stop_time, NULL, 0);
+	if (ret == 0)
+		msleep(delay);
 }
 
 int whc_mmcie_add(struct wusbhc *wusbhc, u8 interval, u8 repeat_cnt,
diff --git a/drivers/usb/serial/console.c b/drivers/usb/serial/console.c
index 5b95009..19e2404 100644
--- a/drivers/usb/serial/console.c
+++ b/drivers/usb/serial/console.c
@@ -241,12 +241,25 @@
 	}
 }
 
+static struct tty_driver *usb_console_device(struct console *co, int *index)
+{
+	struct tty_driver **p = (struct tty_driver **)co->data;
+
+	if (!*p)
+		return NULL;
+
+	*index = co->index;
+	return *p;
+}
+
 static struct console usbcons = {
 	.name =		"ttyUSB",
 	.write =	usb_console_write,
+	.device =	usb_console_device,
 	.setup =	usb_console_setup,
 	.flags =	CON_PRINTBUFFER,
 	.index =	-1,
+	.data = 	&usb_serial_tty_driver,
 };
 
 void usb_serial_console_disconnect(struct usb_serial *serial)
diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index fb6f293..ef6cfa5 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -1054,6 +1054,8 @@
 
 	if (copy_from_user(&new_serial, newinfo, sizeof(new_serial)))
 		return -EFAULT;
+
+	lock_kernel();
 	old_priv = *priv;
 
 	/* Do error checking and permission checking */
@@ -1069,8 +1071,10 @@
 	}
 
 	if ((new_serial.baud_base != priv->baud_base) &&
-	    (new_serial.baud_base < 9600))
+	    (new_serial.baud_base < 9600)) {
+	    	unlock_kernel();
 		return -EINVAL;
+	}
 
 	/* Make the changes - these are privileged changes! */
 
@@ -1098,8 +1102,11 @@
 	     (priv->flags & ASYNC_SPD_MASK)) ||
 	    (((priv->flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST) &&
 	     (old_priv.custom_divisor != priv->custom_divisor))) {
+		unlock_kernel();
 		change_speed(tty, port);
 	}
+	else
+		unlock_kernel();
 	return 0;
 
 } /* set_serial_info */
diff --git a/drivers/usb/serial/kl5kusb105.c b/drivers/usb/serial/kl5kusb105.c
index dc36a05..fcd9082 100644
--- a/drivers/usb/serial/kl5kusb105.c
+++ b/drivers/usb/serial/kl5kusb105.c
@@ -878,6 +878,7 @@
 
 	dbg("%sstate=%d", __func__, break_state);
 
+	/* LOCKING */
 	if (break_state)
 		lcr |= MCT_U232_SET_BREAK;
 
diff --git a/drivers/usb/serial/mct_u232.c b/drivers/usb/serial/mct_u232.c
index 07710cf..82930a7 100644
--- a/drivers/usb/serial/mct_u232.c
+++ b/drivers/usb/serial/mct_u232.c
@@ -721,10 +721,10 @@
 
 	spin_lock_irqsave(&priv->lock, flags);
 	lcr = priv->last_lcr;
-	spin_unlock_irqrestore(&priv->lock, flags);
 
 	if (break_state)
 		lcr |= MCT_U232_SET_BREAK;
+	spin_unlock_irqrestore(&priv->lock, flags);
 
 	mct_u232_set_line_ctrl(serial, lcr);
 } /* mct_u232_break_ctl */
diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c
index fda4a64..96a8c77 100644
--- a/drivers/usb/serial/mos7840.c
+++ b/drivers/usb/serial/mos7840.c
@@ -1343,6 +1343,7 @@
 	else
 		data = mos7840_port->shadowLCR & ~LCR_SET_BREAK;
 
+	/* FIXME: no locking on shadowLCR anywhere in driver */
 	mos7840_port->shadowLCR = data;
 	dbg("mcs7840_break mos7840_port->shadowLCR is %x\n",
 	    mos7840_port->shadowLCR);
@@ -2214,10 +2215,12 @@
 		break;
 	}
 
+	lock_kernel();
 	mos7840_port->shadowMCR = mcr;
 
 	Data = mos7840_port->shadowMCR;
 	status = mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+	unlock_kernel();
 	if (status < 0) {
 		dbg("setting MODEM_CONTROL_REGISTER Failed\n");
 		return -1;
diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c
index 0f2b672..d9bf9a5 100644
--- a/drivers/usb/serial/sierra.c
+++ b/drivers/usb/serial/sierra.c
@@ -442,7 +442,7 @@
 		    " endpoint %02x.", __func__, status, endpoint);
 	} else {
 		if (urb->actual_length) {
-		tty = tty_port_tty_get(&port->port);
+			tty = tty_port_tty_get(&port->port);
 			tty_buffer_request_room(tty, urb->actual_length);
 			tty_insert_flip_string(tty, data, urb->actual_length);
 			tty_flip_buffer_push(tty);
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 794b5ff..080ade2 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -269,15 +269,19 @@
 		return;
 	}
 
-	--port->port.count;
-	if (port->port.count == 0)
+	if (port->port.count == 1)
 		/* only call the device specific close if this
-		 * port is being closed by the last owner */
+		 * port is being closed by the last owner. Ensure we do
+		 * this before we drop the port count. The call is protected
+		 * by the port mutex
+		 */
 		port->serial->type->close(tty, port, filp);
 
-	if (port->port.count == (port->console? 1 : 0)) {
+	if (port->port.count == (port->console ? 2 : 1)) {
 		struct tty_struct *tty = tty_port_tty_get(&port->port);
 		if (tty) {
+			/* We must do this before we drop the port count to
+			   zero. */
 			if (tty->driver_data)
 				tty->driver_data = NULL;
 			tty_port_tty_set(&port->port, NULL);
@@ -285,13 +289,14 @@
 		}
 	}
 
-	if (port->port.count == 0) {
+	if (port->port.count == 1) {
 		mutex_lock(&port->serial->disc_mutex);
 		if (!port->serial->disconnected)
 			usb_autopm_put_interface(port->serial->interface);
 		mutex_unlock(&port->serial->disc_mutex);
 		module_put(port->serial->type->driver.owner);
 	}
+	--port->port.count;
 
 	mutex_unlock(&port->mutex);
 	usb_serial_put(port->serial);
@@ -334,6 +339,10 @@
 	dbg("%s = port %d", __func__, port->number);
 
 	WARN_ON(!port->port.count);
+	/* if the device was unplugged then any remaining characters
+	   fell out of the connector ;) */
+	if (port->serial->disconnected)
+		return 0;
 	/* pass on to the driver specific version of this function */
 	return port->serial->type->chars_in_buffer(tty);
 }
@@ -373,9 +382,7 @@
 	/* pass on to the driver specific version of this function
 	   if it is available */
 	if (port->serial->type->ioctl) {
-		lock_kernel();
 		retval = port->serial->type->ioctl(tty, file, cmd, arg);
-		unlock_kernel();
 	} else
 		retval = -ENOIOCTLCMD;
 	return retval;
@@ -404,11 +411,8 @@
 	WARN_ON(!port->port.count);
 	/* pass on to the driver specific version of this function
 	   if it is available */
-	if (port->serial->type->break_ctl) {
-		lock_kernel();
+	if (port->serial->type->break_ctl)
 		port->serial->type->break_ctl(tty, break_state);
-		unlock_kernel();
-	}
 	return 0;
 }
 
diff --git a/drivers/usb/wusbcore/cbaf.c b/drivers/usb/wusbcore/cbaf.c
index ab4788d..1335cbe 100644
--- a/drivers/usb/wusbcore/cbaf.c
+++ b/drivers/usb/wusbcore/cbaf.c
@@ -88,7 +88,6 @@
  */
 #include <linux/module.h>
 #include <linux/ctype.h>
-#include <linux/version.h>
 #include <linux/usb.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
diff --git a/drivers/usb/wusbcore/crypto.c b/drivers/usb/wusbcore/crypto.c
index c36c438..9ec7fd5 100644
--- a/drivers/usb/wusbcore/crypto.c
+++ b/drivers/usb/wusbcore/crypto.c
@@ -51,9 +51,17 @@
 #include <linux/uwb.h>
 #include <linux/usb/wusb.h>
 #include <linux/scatterlist.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
+static int debug_crypto_verify = 0;
+
+module_param(debug_crypto_verify, int, 0);
+MODULE_PARM_DESC(debug_crypto_verify, "verify the key generation algorithms");
+
+static void wusb_key_dump(const void *buf, size_t len)
+{
+	print_hex_dump(KERN_ERR, "  ", DUMP_PREFIX_OFFSET, 16, 1,
+		       buf, len, 0);
+}
 
 /*
  * Block of data, as understood by AES-CCM
@@ -203,9 +211,6 @@
 	const u8 bzero[16] = { 0 };
 	size_t zero_padding;
 
-	d_fnstart(3, NULL, "(tfm_cbc %p, tfm_aes %p, mic %p, "
-		  "n %p, a %p, b %p, blen %zu)\n",
-		  tfm_cbc, tfm_aes, mic, n, a, b, blen);
 	/*
 	 * These checks should be compile time optimized out
 	 * ensure @a fills b1's mac_header and following fields
@@ -247,16 +252,6 @@
 	b1.la = cpu_to_be16(blen + 14);
 	memcpy(&b1.mac_header, a, sizeof(*a));
 
-	d_printf(4, NULL, "I: B0 (%zu bytes)\n", sizeof(b0));
-	d_dump(4, NULL, &b0, sizeof(b0));
-	d_printf(4, NULL, "I: B1 (%zu bytes)\n", sizeof(b1));
-	d_dump(4, NULL, &b1, sizeof(b1));
-	d_printf(4, NULL, "I: B (%zu bytes)\n", blen);
-	d_dump(4, NULL, b, blen);
-	d_printf(4, NULL, "I: B 0-padding (%zu bytes)\n", zero_padding);
-	d_printf(4, NULL, "D: IV before crypto (%zu)\n", ivsize);
-	d_dump(4, NULL, iv, ivsize);
-
 	sg_init_table(sg, ARRAY_SIZE(sg));
 	sg_set_buf(&sg[0], &b0, sizeof(b0));
 	sg_set_buf(&sg[1], &b1, sizeof(b1));
@@ -273,8 +268,6 @@
 		       result);
 		goto error_cbc_crypt;
 	}
-	d_printf(4, NULL, "D: MIC tag\n");
-	d_dump(4, NULL, iv, ivsize);
 
 	/* Now we crypt the MIC Tag (*iv) with Ax -- values per WUSB1.0[6.5]
 	 * The procedure is to AES crypt the A0 block and XOR the MIC
@@ -289,17 +282,10 @@
 	ax.counter = 0;
 	crypto_cipher_encrypt_one(tfm_aes, (void *)&ax, (void *)&ax);
 	bytewise_xor(mic, &ax, iv, 8);
-	d_printf(4, NULL, "D: CTR[MIC]\n");
-	d_dump(4, NULL, &ax, 8);
-	d_printf(4, NULL, "D: CCM-MIC tag\n");
-	d_dump(4, NULL, mic, 8);
 	result = 8;
 error_cbc_crypt:
 	kfree(dst_buf);
 error_dst_buf:
-	d_fnend(3, NULL, "(tfm_cbc %p, tfm_aes %p, mic %p, "
-		"n %p, a %p, b %p, blen %zu)\n",
-		tfm_cbc, tfm_aes, mic, n, a, b, blen);
 	return result;
 }
 
@@ -321,10 +307,6 @@
 	u64 sfn = 0;
 	__le64 sfn_le;
 
-	d_fnstart(3, NULL, "(out %p, out_size %zu, key %p, _n %p, "
-		  "a %p, b %p, blen %zu, len %zu)\n", out, out_size,
-		  key, _n, a, b, blen, len);
-
 	tfm_cbc = crypto_alloc_blkcipher("cbc(aes)", 0, CRYPTO_ALG_ASYNC);
 	if (IS_ERR(tfm_cbc)) {
 		result = PTR_ERR(tfm_cbc);
@@ -366,9 +348,6 @@
 error_setkey_cbc:
 	crypto_free_blkcipher(tfm_cbc);
 error_alloc_cbc:
-	d_fnend(3, NULL, "(out %p, out_size %zu, key %p, _n %p, "
-		"a %p, b %p, blen %zu, len %zu) = %d\n", out, out_size,
-		key, _n, a, b, blen, len, (int)bytes);
 	return result;
 }
 
@@ -422,14 +401,14 @@
 		       "mismatch between MIC result and WUSB1.0[A2]\n");
 		hs_size = sizeof(stv_hsmic_hs) - sizeof(stv_hsmic_hs.MIC);
 		printk(KERN_ERR "E: Handshake2 in: (%zu bytes)\n", hs_size);
-		dump_bytes(NULL, &stv_hsmic_hs, hs_size);
+		wusb_key_dump(&stv_hsmic_hs, hs_size);
 		printk(KERN_ERR "E: CCM Nonce in: (%zu bytes)\n",
 		       sizeof(stv_hsmic_n));
-		dump_bytes(NULL, &stv_hsmic_n, sizeof(stv_hsmic_n));
+		wusb_key_dump(&stv_hsmic_n, sizeof(stv_hsmic_n));
 		printk(KERN_ERR "E: MIC out:\n");
-		dump_bytes(NULL, mic, sizeof(mic));
+		wusb_key_dump(mic, sizeof(mic));
 		printk(KERN_ERR "E: MIC out (from WUSB1.0[A.2]):\n");
-		dump_bytes(NULL, stv_hsmic_hs.MIC, sizeof(stv_hsmic_hs.MIC));
+		wusb_key_dump(stv_hsmic_hs.MIC, sizeof(stv_hsmic_hs.MIC));
 		result = -EINVAL;
 	} else
 		result = 0;
@@ -497,19 +476,16 @@
 		printk(KERN_ERR "E: WUSB key derivation test: "
 		       "mismatch between key derivation result "
 		       "and WUSB1.0[A1] Errata 2006/12\n");
-		printk(KERN_ERR "E: keydvt in: key (%zu bytes)\n",
-		       sizeof(stv_key_a1));
-		dump_bytes(NULL, stv_key_a1, sizeof(stv_key_a1));
-		printk(KERN_ERR "E: keydvt in: nonce (%zu bytes)\n",
-		       sizeof(stv_keydvt_n_a1));
-		dump_bytes(NULL, &stv_keydvt_n_a1, sizeof(stv_keydvt_n_a1));
-		printk(KERN_ERR "E: keydvt in: hnonce & dnonce (%zu bytes)\n",
-		       sizeof(stv_keydvt_in_a1));
-		dump_bytes(NULL, &stv_keydvt_in_a1, sizeof(stv_keydvt_in_a1));
+		printk(KERN_ERR "E: keydvt in: key\n");
+		wusb_key_dump(stv_key_a1, sizeof(stv_key_a1));
+		printk(KERN_ERR "E: keydvt in: nonce\n");
+		wusb_key_dump( &stv_keydvt_n_a1, sizeof(stv_keydvt_n_a1));
+		printk(KERN_ERR "E: keydvt in: hnonce & dnonce\n");
+		wusb_key_dump(&stv_keydvt_in_a1, sizeof(stv_keydvt_in_a1));
 		printk(KERN_ERR "E: keydvt out: KCK\n");
-		dump_bytes(NULL, &keydvt_out.kck, sizeof(keydvt_out.kck));
+		wusb_key_dump(&keydvt_out.kck, sizeof(keydvt_out.kck));
 		printk(KERN_ERR "E: keydvt out: PTK\n");
-		dump_bytes(NULL, &keydvt_out.ptk, sizeof(keydvt_out.ptk));
+		wusb_key_dump(&keydvt_out.ptk, sizeof(keydvt_out.ptk));
 		result = -EINVAL;
 	} else
 		result = 0;
@@ -526,10 +502,13 @@
 {
 	int result;
 
-	result = wusb_key_derive_verify();
-	if (result < 0)
-		return result;
-	return wusb_oob_mic_verify();
+	if (debug_crypto_verify) {
+		result = wusb_key_derive_verify();
+		if (result < 0)
+			return result;
+		return wusb_oob_mic_verify();
+	}
+	return 0;
 }
 
 void wusb_crypto_exit(void)
diff --git a/drivers/usb/wusbcore/dev-sysfs.c b/drivers/usb/wusbcore/dev-sysfs.c
index 7897a19..1018345 100644
--- a/drivers/usb/wusbcore/dev-sysfs.c
+++ b/drivers/usb/wusbcore/dev-sysfs.c
@@ -28,10 +28,6 @@
 #include <linux/workqueue.h>
 #include "wusbhc.h"
 
-#undef D_LOCAL
-#define D_LOCAL 4
-#include <linux/uwb/debug.h>
-
 static ssize_t wusb_disconnect_store(struct device *dev,
 				     struct device_attribute *attr,
 				     const char *buf, size_t size)
diff --git a/drivers/usb/wusbcore/devconnect.c b/drivers/usb/wusbcore/devconnect.c
index f45d777..e2e7e4b 100644
--- a/drivers/usb/wusbcore/devconnect.c
+++ b/drivers/usb/wusbcore/devconnect.c
@@ -57,9 +57,6 @@
  *                              Called by notif.c:wusb_handle_dn_connect()
  *                              when a DN_Connect is received.
  *
- *   wusbhc_devconnect_auth()   Called by rh.c:wusbhc_rh_port_reset() when
- *                              doing the device connect sequence.
- *
  *     wusb_devconnect_acked()  Ack done, release resources.
  *
  *   wusb_handle_dn_alive()     Called by notif.c:wusb_handle_dn()
@@ -69,9 +66,6 @@
  *                              process a disconenct request from a
  *                              device.
  *
- *   wusb_dev_reset()           Called by rh.c:wusbhc_rh_port_reset() when
- *                              resetting a device.
- *
  *   __wusb_dev_disable()       Called by rh.c:wusbhc_rh_clear_port_feat() when
  *                              disabling a port.
  *
@@ -97,10 +91,6 @@
 #include <linux/workqueue.h>
 #include "wusbhc.h"
 
-#undef D_LOCAL
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
-
 static void wusbhc_devconnect_acked_work(struct work_struct *work);
 
 static void wusb_dev_free(struct wusb_dev *wusb_dev)
@@ -240,6 +230,7 @@
 	list_add_tail(&wusb_dev->cack_node, &wusbhc->cack_list);
 	wusbhc->cack_count++;
 	wusbhc_fill_cack_ie(wusbhc);
+
 	return wusb_dev;
 }
 
@@ -250,12 +241,9 @@
  */
 static void wusbhc_cack_rm(struct wusbhc *wusbhc, struct wusb_dev *wusb_dev)
 {
-	struct device *dev = wusbhc->dev;
-	d_fnstart(3, dev, "(wusbhc %p wusb_dev %p)\n", wusbhc, wusb_dev);
 	list_del_init(&wusb_dev->cack_node);
 	wusbhc->cack_count--;
 	wusbhc_fill_cack_ie(wusbhc);
-	d_fnend(3, dev, "(wusbhc %p wusb_dev %p) = void\n", wusbhc, wusb_dev);
 }
 
 /*
@@ -263,14 +251,11 @@
 static
 void wusbhc_devconnect_acked(struct wusbhc *wusbhc, struct wusb_dev *wusb_dev)
 {
-	struct device *dev = wusbhc->dev;
-	d_fnstart(3, dev, "(wusbhc %p wusb_dev %p)\n", wusbhc, wusb_dev);
 	wusbhc_cack_rm(wusbhc, wusb_dev);
 	if (wusbhc->cack_count)
 		wusbhc_mmcie_set(wusbhc, 0, 0, &wusbhc->cack_ie.hdr);
 	else
 		wusbhc_mmcie_rm(wusbhc, &wusbhc->cack_ie.hdr);
-	d_fnend(3, dev, "(wusbhc %p wusb_dev %p) = void\n", wusbhc, wusb_dev);
 }
 
 static void wusbhc_devconnect_acked_work(struct work_struct *work)
@@ -320,7 +305,6 @@
 	struct wusb_port *port;
 	unsigned idx, devnum;
 
-	d_fnstart(3, dev, "(%p, %p, %s)\n", wusbhc, dnc, pr_cdid);
 	mutex_lock(&wusbhc->mutex);
 
 	/* Check we are not handling it already */
@@ -366,16 +350,13 @@
 	port->wusb_dev = wusb_dev;
 	port->status |= USB_PORT_STAT_CONNECTION;
 	port->change |= USB_PORT_STAT_C_CONNECTION;
-	port->reset_count = 0;
 	/* Now the port status changed to connected; khubd will
 	 * pick the change up and try to reset the port to bring it to
 	 * the enabled state--so this process returns up to the stack
-	 * and it calls back into wusbhc_rh_port_reset() who will call
-	 * devconnect_auth().
+	 * and it calls back into wusbhc_rh_port_reset().
 	 */
 error_unlock:
 	mutex_unlock(&wusbhc->mutex);
-	d_fnend(3, dev, "(%p, %p, %s) = void\n", wusbhc, dnc, pr_cdid);
 	return;
 
 }
@@ -398,10 +379,8 @@
 static void __wusbhc_dev_disconnect(struct wusbhc *wusbhc,
 				    struct wusb_port *port)
 {
-	struct device *dev = wusbhc->dev;
 	struct wusb_dev *wusb_dev = port->wusb_dev;
 
-	d_fnstart(3, dev, "(wusbhc %p, port %p)\n", wusbhc, port);
 	port->status &= ~(USB_PORT_STAT_CONNECTION | USB_PORT_STAT_ENABLE
 			  | USB_PORT_STAT_SUSPEND | USB_PORT_STAT_RESET
 			  | USB_PORT_STAT_LOW_SPEED | USB_PORT_STAT_HIGH_SPEED);
@@ -413,15 +392,11 @@
 		wusb_dev_put(wusb_dev);
 	}
 	port->wusb_dev = NULL;
-	/* don't reset the reset_count to zero or wusbhc_rh_port_reset will get
-	 * confused! We only reset to zero when we connect a new device.
-	 */
 
 	/* After a device disconnects, change the GTK (see [WUSB]
 	 * section 6.2.11.2). */
 	wusbhc_gtk_rekey(wusbhc);
 
-	d_fnend(3, dev, "(wusbhc %p, port %p) = void\n", wusbhc, port);
 	/* The Wireless USB part has forgotten about the device already; now
 	 * khubd's timer will pick up the disconnection and remove the USB
 	 * device from the system
@@ -429,39 +404,6 @@
 }
 
 /*
- * Authenticate a device into the WUSB Cluster
- *
- * Called from the Root Hub code (rh.c:wusbhc_rh_port_reset()) when
- * asking for a reset on a port that is not enabled (ie: first connect
- * on the port).
- *
- * Performs the 4way handshake to allow the device to comunicate w/ the
- * WUSB Cluster securely; once done, issue a request to the device for
- * it to change to address 0.
- *
- * This mimics the reset step of Wired USB that once resetting a
- * device, leaves the port in enabled state and the dev with the
- * default address (0).
- *
- * WUSB1.0[7.1.2]
- *
- * @port_idx: port where the change happened--This is the index into
- *            the wusbhc port array, not the USB port number.
- */
-int wusbhc_devconnect_auth(struct wusbhc *wusbhc, u8 port_idx)
-{
-	struct device *dev = wusbhc->dev;
-	struct wusb_port *port = wusb_port_by_idx(wusbhc, port_idx);
-
-	d_fnstart(3, dev, "(%p, %u)\n", wusbhc, port_idx);
-	port->status &= ~USB_PORT_STAT_RESET;
-	port->status |= USB_PORT_STAT_ENABLE;
-	port->change |= USB_PORT_STAT_C_RESET | USB_PORT_STAT_C_ENABLE;
-	d_fnend(3, dev, "(%p, %u) = 0\n", wusbhc, port_idx);
-	return 0;
-}
-
-/*
  * Refresh the list of keep alives to emit in the MMC
  *
  * Some devices don't respond to keep alives unless they've been
@@ -528,21 +470,15 @@
  */
 static void wusbhc_keep_alive_run(struct work_struct *ws)
 {
-	struct delayed_work *dw =
-		container_of(ws, struct delayed_work, work);
-	struct wusbhc *wusbhc =
-		container_of(dw, struct wusbhc, keep_alive_timer);
+	struct delayed_work *dw = container_of(ws, struct delayed_work, work);
+	struct wusbhc *wusbhc =	container_of(dw, struct wusbhc, keep_alive_timer);
 
-	d_fnstart(5, wusbhc->dev, "(wusbhc %p)\n", wusbhc);
-	if (wusbhc->active) {
-		mutex_lock(&wusbhc->mutex);
-		__wusbhc_keep_alive(wusbhc);
-		mutex_unlock(&wusbhc->mutex);
-		queue_delayed_work(wusbd, &wusbhc->keep_alive_timer,
-				   (wusbhc->trust_timeout * CONFIG_HZ)/1000/2);
-	}
-	d_fnend(5, wusbhc->dev, "(wusbhc %p) = void\n", wusbhc);
-	return;
+	mutex_lock(&wusbhc->mutex);
+	__wusbhc_keep_alive(wusbhc);
+	mutex_unlock(&wusbhc->mutex);
+
+	queue_delayed_work(wusbd, &wusbhc->keep_alive_timer,
+			   msecs_to_jiffies(wusbhc->trust_timeout / 2));
 }
 
 /*
@@ -585,10 +521,6 @@
  */
 static void wusbhc_handle_dn_alive(struct wusbhc *wusbhc, struct wusb_dev *wusb_dev)
 {
-	struct device *dev = wusbhc->dev;
-
-	d_printf(2, dev, "DN ALIVE: device 0x%02x pong\n", wusb_dev->addr);
-
 	mutex_lock(&wusbhc->mutex);
 	wusb_dev->entry_ts = jiffies;
 	__wusbhc_keep_alive(wusbhc);
@@ -621,11 +553,10 @@
 		"no-beacon"
 	};
 
-	d_fnstart(3, dev, "(%p, %p, %zu)\n", wusbhc, dn_hdr, size);
 	if (size < sizeof(*dnc)) {
 		dev_err(dev, "DN CONNECT: short notification (%zu < %zu)\n",
 			size, sizeof(*dnc));
-		goto out;
+		return;
 	}
 
 	dnc = container_of(dn_hdr, struct wusb_dn_connect, hdr);
@@ -637,10 +568,6 @@
 		 wusb_dn_connect_new_connection(dnc) ? "connect" : "reconnect");
 	/* ACK the connect */
 	wusbhc_devconnect_ack(wusbhc, dnc, pr_cdid);
-out:
-	d_fnend(3, dev, "(%p, %p, %zu) = void\n",
-		wusbhc, dn_hdr, size);
-	return;
 }
 
 /*
@@ -662,60 +589,6 @@
 }
 
 /*
- * Reset a WUSB device on a HWA
- *
- * @wusbhc
- * @port_idx   Index of the port where the device is
- *
- * In Wireless USB, a reset is more or less equivalent to a full
- * disconnect; so we just do a full disconnect and send the device a
- * Device Reset IE (WUSB1.0[7.5.11]) giving it a few millisecs (6 MMCs).
- *
- * @wusbhc should be refcounted and unlocked
- */
-int wusbhc_dev_reset(struct wusbhc *wusbhc, u8 port_idx)
-{
-	int result;
-	struct device *dev = wusbhc->dev;
-	struct wusb_dev *wusb_dev;
-	struct wuie_reset *ie;
-
-	d_fnstart(3, dev, "(%p, %u)\n", wusbhc, port_idx);
-	mutex_lock(&wusbhc->mutex);
-	result = 0;
-	wusb_dev = wusb_port_by_idx(wusbhc, port_idx)->wusb_dev;
-	if (wusb_dev == NULL) {
-		/* reset no device? ignore */
-		dev_dbg(dev, "RESET: no device at port %u, ignoring\n",
-			port_idx);
-		goto error_unlock;
-	}
-	result = -ENOMEM;
-	ie = kzalloc(sizeof(*ie), GFP_KERNEL);
-	if (ie == NULL)
-		goto error_unlock;
-	ie->hdr.bLength = sizeof(ie->hdr) + sizeof(ie->CDID);
-	ie->hdr.bIEIdentifier = WUIE_ID_RESET_DEVICE;
-	ie->CDID = wusb_dev->cdid;
-	result = wusbhc_mmcie_set(wusbhc, 0xff, 6, &ie->hdr);
-	if (result < 0) {
-		dev_err(dev, "RESET: cant's set MMC: %d\n", result);
-		goto error_kfree;
-	}
-	__wusbhc_dev_disconnect(wusbhc, wusb_port_by_idx(wusbhc, port_idx));
-
-	/* 120ms, hopefully 6 MMCs (FIXME) */
-	msleep(120);
-	wusbhc_mmcie_rm(wusbhc, &ie->hdr);
-error_kfree:
-	kfree(ie);
-error_unlock:
-	mutex_unlock(&wusbhc->mutex);
-	d_fnend(3, dev, "(%p, %u) = %d\n", wusbhc, port_idx, result);
-	return result;
-}
-
-/*
  * Handle a Device Notification coming a host
  *
  * The Device Notification comes from a host (HWA, DWA or WHCI)
@@ -735,19 +608,17 @@
 	struct device *dev = wusbhc->dev;
 	struct wusb_dev *wusb_dev;
 
-	d_fnstart(3, dev, "(%p, %p)\n", wusbhc, dn_hdr);
-
 	if (size < sizeof(struct wusb_dn_hdr)) {
 		dev_err(dev, "DN data shorter than DN header (%d < %d)\n",
 			(int)size, (int)sizeof(struct wusb_dn_hdr));
-		goto out;
+		return;
 	}
 
 	wusb_dev = wusbhc_find_dev_by_addr(wusbhc, srcaddr);
 	if (wusb_dev == NULL && dn_hdr->bType != WUSB_DN_CONNECT) {
 		dev_dbg(dev, "ignoring DN %d from unconnected device %02x\n",
 			dn_hdr->bType, srcaddr);
-		goto out;
+		return;
 	}
 
 	switch (dn_hdr->bType) {
@@ -772,9 +643,6 @@
 		dev_warn(dev, "unknown DN %u (%d octets) from %u\n",
 			 dn_hdr->bType, (int)size, srcaddr);
 	}
-out:
-	d_fnend(3, dev, "(%p, %p) = void\n", wusbhc, dn_hdr);
-	return;
 }
 EXPORT_SYMBOL_GPL(wusbhc_handle_dn);
 
@@ -804,59 +672,30 @@
 	struct wusb_dev *wusb_dev;
 	struct wuie_disconnect *ie;
 
-	d_fnstart(3, dev, "(%p, %u)\n", wusbhc, port_idx);
-	result = 0;
 	wusb_dev = wusb_port_by_idx(wusbhc, port_idx)->wusb_dev;
 	if (wusb_dev == NULL) {
 		/* reset no device? ignore */
 		dev_dbg(dev, "DISCONNECT: no device at port %u, ignoring\n",
 			port_idx);
-		goto error;
+		return;
 	}
 	__wusbhc_dev_disconnect(wusbhc, wusb_port_by_idx(wusbhc, port_idx));
 
-	result = -ENOMEM;
 	ie = kzalloc(sizeof(*ie), GFP_KERNEL);
 	if (ie == NULL)
-		goto error;
+		return;
 	ie->hdr.bLength = sizeof(*ie);
 	ie->hdr.bIEIdentifier = WUIE_ID_DEVICE_DISCONNECT;
 	ie->bDeviceAddress = wusb_dev->addr;
 	result = wusbhc_mmcie_set(wusbhc, 0, 0, &ie->hdr);
-	if (result < 0) {
+	if (result < 0)
 		dev_err(dev, "DISCONNECT: can't set MMC: %d\n", result);
-		goto error_kfree;
+	else {
+		/* At least 6 MMCs, assuming at least 1 MMC per zone. */
+		msleep(7*4);
+		wusbhc_mmcie_rm(wusbhc, &ie->hdr);
 	}
-
-	/* 120ms, hopefully 6 MMCs */
-	msleep(100);
-	wusbhc_mmcie_rm(wusbhc, &ie->hdr);
-error_kfree:
 	kfree(ie);
-error:
-	d_fnend(3, dev, "(%p, %u) = %d\n", wusbhc, port_idx, result);
-	return;
-}
-
-static void wusb_cap_descr_printf(const unsigned level, struct device *dev,
-				  const struct usb_wireless_cap_descriptor *wcd)
-{
-	d_printf(level, dev,
-		 "WUSB Capability Descriptor\n"
-		 "  bDevCapabilityType          0x%02x\n"
-		 "  bmAttributes                0x%02x\n"
-		 "  wPhyRates                   0x%04x\n"
-		 "  bmTFITXPowerInfo            0x%02x\n"
-		 "  bmFFITXPowerInfo            0x%02x\n"
-		 "  bmBandGroup                 0x%04x\n"
-		 "  bReserved                   0x%02x\n",
-		 wcd->bDevCapabilityType,
-		 wcd->bmAttributes,
-		 le16_to_cpu(wcd->wPHYRates),
-		 wcd->bmTFITXPowerInfo,
-		 wcd->bmFFITXPowerInfo,
-		 wcd->bmBandGroup,
-		 wcd->bReserved);
 }
 
 /*
@@ -899,8 +738,6 @@
 		}
 		cap_size = cap_hdr->bLength;
 		cap_type = cap_hdr->bDevCapabilityType;
-		d_printf(4, dev, "BOS Capability: 0x%02x (%zu bytes)\n",
-			 cap_type, cap_size);
 		if (cap_size == 0)
 			break;
 		if (cap_size > top - itr) {
@@ -912,7 +749,6 @@
 			result = -EBADF;
 			goto error_bad_cap;
 		}
-		d_dump(3, dev, itr, cap_size);
 		switch (cap_type) {
 		case USB_CAP_TYPE_WIRELESS_USB:
 			if (cap_size != sizeof(*wusb_dev->wusb_cap_descr))
@@ -920,10 +756,8 @@
 					"descriptor is %zu bytes vs %zu "
 					"needed\n", cap_size,
 					sizeof(*wusb_dev->wusb_cap_descr));
-			else {
+			else
 				wusb_dev->wusb_cap_descr = itr;
-				wusb_cap_descr_printf(3, dev, itr);
-			}
 			break;
 		default:
 			dev_err(dev, "BUG? Unknown BOS capability 0x%02x "
@@ -988,9 +822,7 @@
 			"%zu bytes): %zd\n", desc_size, result);
 		goto error_get_descriptor;
 	}
-	d_printf(2, dev, "Got BOS descriptor %zd bytes, %u capabilities\n",
-		 result, bos->bNumDeviceCaps);
-	d_dump(2, dev, bos, result);
+
 	result = wusb_dev_bos_grok(usb_dev, wusb_dev, bos, result);
 	if (result < 0)
 		goto error_bad_bos;
@@ -1056,8 +888,6 @@
 	if (usb_dev->wusb == 0 || usb_dev->devnum == 1)
 		return;		/* skip non wusb and wusb RHs */
 
-	d_fnstart(3, dev, "(usb_dev %p)\n", usb_dev);
-
 	wusbhc = wusbhc_get_by_usb_dev(usb_dev);
 	if (wusbhc == NULL)
 		goto error_nodev;
@@ -1087,7 +917,6 @@
 	wusb_dev_put(wusb_dev);
 	wusbhc_put(wusbhc);
 error_nodev:
-	d_fnend(3, dev, "(usb_dev %p) = void\n", usb_dev);
 	return;
 
 	wusb_dev_sysfs_rm(wusb_dev);
@@ -1174,11 +1003,10 @@
 
 void wusb_dev_destroy(struct kref *_wusb_dev)
 {
-	struct wusb_dev *wusb_dev
-		= container_of(_wusb_dev, struct wusb_dev, refcnt);
+	struct wusb_dev *wusb_dev = container_of(_wusb_dev, struct wusb_dev, refcnt);
+
 	list_del_init(&wusb_dev->cack_node);
 	wusb_dev_free(wusb_dev);
-	d_fnend(1, NULL, "%s (wusb_dev %p) = void\n", __func__, wusb_dev);
 }
 EXPORT_SYMBOL_GPL(wusb_dev_destroy);
 
@@ -1190,8 +1018,6 @@
  */
 int wusbhc_devconnect_create(struct wusbhc *wusbhc)
 {
-	d_fnstart(3, wusbhc->dev, "(wusbhc %p)\n", wusbhc);
-
 	wusbhc->keep_alive_ie.hdr.bIEIdentifier = WUIE_ID_KEEP_ALIVE;
 	wusbhc->keep_alive_ie.hdr.bLength = sizeof(wusbhc->keep_alive_ie.hdr);
 	INIT_DELAYED_WORK(&wusbhc->keep_alive_timer, wusbhc_keep_alive_run);
@@ -1200,7 +1026,6 @@
 	wusbhc->cack_ie.hdr.bLength = sizeof(wusbhc->cack_ie.hdr);
 	INIT_LIST_HEAD(&wusbhc->cack_list);
 
-	d_fnend(3, wusbhc->dev, "(wusbhc %p) = void\n", wusbhc);
 	return 0;
 }
 
@@ -1209,8 +1034,7 @@
  */
 void wusbhc_devconnect_destroy(struct wusbhc *wusbhc)
 {
-	d_fnstart(3, wusbhc->dev, "(wusbhc %p)\n", wusbhc);
-	d_fnend(3, wusbhc->dev, "(wusbhc %p) = void\n", wusbhc);
+	/* no op */
 }
 
 /*
@@ -1222,8 +1046,7 @@
  * FIXME: This also enables the keep alives but this is not necessary
  * until there are connected and authenticated devices.
  */
-int wusbhc_devconnect_start(struct wusbhc *wusbhc,
-			    const struct wusb_ckhdid *chid)
+int wusbhc_devconnect_start(struct wusbhc *wusbhc)
 {
 	struct device *dev = wusbhc->dev;
 	struct wuie_host_info *hi;
@@ -1236,7 +1059,7 @@
 	hi->hdr.bLength       = sizeof(*hi);
 	hi->hdr.bIEIdentifier = WUIE_ID_HOST_INFO;
 	hi->attributes        = cpu_to_le16((wusbhc->rsv->stream << 3) | WUIE_HI_CAP_ALL);
-	hi->CHID              = *chid;
+	hi->CHID              = wusbhc->chid;
 	result = wusbhc_mmcie_set(wusbhc, 0, 0, &hi->hdr);
 	if (result < 0) {
 		dev_err(dev, "Cannot add Host Info MMCIE: %d\n", result);
diff --git a/drivers/usb/wusbcore/mmc.c b/drivers/usb/wusbcore/mmc.c
index cfa77a0..3b52161 100644
--- a/drivers/usb/wusbcore/mmc.c
+++ b/drivers/usb/wusbcore/mmc.c
@@ -159,15 +159,35 @@
 }
 EXPORT_SYMBOL_GPL(wusbhc_mmcie_rm);
 
+static int wusbhc_mmc_start(struct wusbhc *wusbhc)
+{
+	int ret;
+
+	mutex_lock(&wusbhc->mutex);
+	ret = wusbhc->start(wusbhc);
+	if (ret >= 0)
+		wusbhc->active = 1;
+	mutex_unlock(&wusbhc->mutex);
+
+	return ret;
+}
+
+static void wusbhc_mmc_stop(struct wusbhc *wusbhc)
+{
+	mutex_lock(&wusbhc->mutex);
+	wusbhc->active = 0;
+	wusbhc->stop(wusbhc, WUSB_CHANNEL_STOP_DELAY_MS);
+	mutex_unlock(&wusbhc->mutex);
+}
+
 /*
  * wusbhc_start - start transmitting MMCs and accepting connections
  * @wusbhc: the HC to start
- * @chid: the CHID to use for this host
  *
  * Establishes a cluster reservation, enables device connections, and
  * starts MMCs with appropriate DNTS parameters.
  */
-int wusbhc_start(struct wusbhc *wusbhc, const struct wusb_ckhdid *chid)
+int wusbhc_start(struct wusbhc *wusbhc)
 {
 	int result;
 	struct device *dev = wusbhc->dev;
@@ -181,7 +201,7 @@
 		goto error_rsv_establish;
 	}
 
-	result = wusbhc_devconnect_start(wusbhc, chid);
+	result = wusbhc_devconnect_start(wusbhc);
 	if (result < 0) {
 		dev_err(dev, "error enabling device connections: %d\n", result);
 		goto error_devconnect_start;
@@ -199,12 +219,12 @@
 		dev_err(dev, "Cannot set DNTS parameters: %d\n", result);
 		goto error_set_num_dnts;
 	}
-	result = wusbhc->start(wusbhc);
+	result = wusbhc_mmc_start(wusbhc);
 	if (result < 0) {
 		dev_err(dev, "error starting wusbch: %d\n", result);
 		goto error_wusbhc_start;
 	}
-	wusbhc->active = 1;
+
 	return 0;
 
 error_wusbhc_start:
@@ -219,76 +239,17 @@
 }
 
 /*
- * Disconnect all from the WUSB Channel
- *
- * Send a Host Disconnect IE in the MMC, wait, don't send it any more
- */
-static int __wusbhc_host_disconnect_ie(struct wusbhc *wusbhc)
-{
-	int result = -ENOMEM;
-	struct wuie_host_disconnect *host_disconnect_ie;
-	might_sleep();
-	host_disconnect_ie = kmalloc(sizeof(*host_disconnect_ie), GFP_KERNEL);
-	if (host_disconnect_ie == NULL)
-		goto error_alloc;
-	host_disconnect_ie->hdr.bLength       = sizeof(*host_disconnect_ie);
-	host_disconnect_ie->hdr.bIEIdentifier = WUIE_ID_HOST_DISCONNECT;
-	result = wusbhc_mmcie_set(wusbhc, 0, 0, &host_disconnect_ie->hdr);
-	if (result < 0)
-		goto error_mmcie_set;
-
-	/* WUSB1.0[8.5.3.1 & 7.5.2] */
-	msleep(100);
-	wusbhc_mmcie_rm(wusbhc, &host_disconnect_ie->hdr);
-error_mmcie_set:
-	kfree(host_disconnect_ie);
-error_alloc:
-	return result;
-}
-
-/*
  * wusbhc_stop - stop transmitting MMCs
  * @wusbhc: the HC to stop
  *
- * Send a Host Disconnect IE, wait, remove all the MMCs (stop sending MMCs).
- *
- * If we can't allocate a Host Stop IE, screw it, we don't notify the
- * devices we are disconnecting...
+ * Stops the WUSB channel and removes the cluster reservation.
  */
 void wusbhc_stop(struct wusbhc *wusbhc)
 {
-	if (wusbhc->active) {
-		wusbhc->active = 0;
-		wusbhc->stop(wusbhc);
-		wusbhc_sec_stop(wusbhc);
-		__wusbhc_host_disconnect_ie(wusbhc);
-		wusbhc_devconnect_stop(wusbhc);
-		wusbhc_rsv_terminate(wusbhc);
-	}
-}
-EXPORT_SYMBOL_GPL(wusbhc_stop);
-
-/*
- * Change the CHID in a WUSB Channel
- *
- * If it is just a new CHID, send a Host Disconnect IE and then change
- * the CHID IE.
- */
-static int __wusbhc_chid_change(struct wusbhc *wusbhc,
-				const struct wusb_ckhdid *chid)
-{
-	int result = -ENOSYS;
-	struct device *dev = wusbhc->dev;
-	dev_err(dev, "%s() not implemented yet\n", __func__);
-	return result;
-
-	BUG_ON(wusbhc->wuie_host_info == NULL);
-	__wusbhc_host_disconnect_ie(wusbhc);
-	wusbhc->wuie_host_info->CHID = *chid;
-	result = wusbhc_mmcie_set(wusbhc, 0, 0, &wusbhc->wuie_host_info->hdr);
-	if (result < 0)
-		dev_err(dev, "Can't update Host Info WUSB IE: %d\n", result);
-	return result;
+	wusbhc_mmc_stop(wusbhc);
+	wusbhc_sec_stop(wusbhc);
+	wusbhc_devconnect_stop(wusbhc);
+	wusbhc_rsv_terminate(wusbhc);
 }
 
 /*
@@ -306,16 +267,19 @@
 		chid = NULL;
 
 	mutex_lock(&wusbhc->mutex);
-	if (wusbhc->active) {
-		if (chid)
-			result = __wusbhc_chid_change(wusbhc, chid);
-		else
-			wusbhc_stop(wusbhc);
-	} else {
-		if (chid)
-			wusbhc_start(wusbhc, chid);
+	if (chid) {
+		if (wusbhc->active) {
+			mutex_unlock(&wusbhc->mutex);
+			return -EBUSY;
+		}
+		wusbhc->chid = *chid;
 	}
 	mutex_unlock(&wusbhc->mutex);
+
+	if (chid)
+		result = uwb_radio_start(&wusbhc->pal);
+	else
+		uwb_radio_stop(&wusbhc->pal);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wusbhc_chid_set);
diff --git a/drivers/usb/wusbcore/pal.c b/drivers/usb/wusbcore/pal.c
index 7cc51e9..d0b172c 100644
--- a/drivers/usb/wusbcore/pal.c
+++ b/drivers/usb/wusbcore/pal.c
@@ -18,6 +18,16 @@
  */
 #include "wusbhc.h"
 
+static void wusbhc_channel_changed(struct uwb_pal *pal, int channel)
+{
+	struct wusbhc *wusbhc = container_of(pal, struct wusbhc, pal);
+
+	if (channel < 0)
+		wusbhc_stop(wusbhc);
+	else
+		wusbhc_start(wusbhc);
+}
+
 /**
  * wusbhc_pal_register - register the WUSB HC as a UWB PAL
  * @wusbhc: the WUSB HC
@@ -28,8 +38,10 @@
 
 	wusbhc->pal.name   = "wusbhc";
 	wusbhc->pal.device = wusbhc->usb_hcd.self.controller;
+	wusbhc->pal.rc     = wusbhc->uwb_rc;
+	wusbhc->pal.channel_changed = wusbhc_channel_changed;
 
-	return uwb_pal_register(wusbhc->uwb_rc, &wusbhc->pal);
+	return uwb_pal_register(&wusbhc->pal);
 }
 
 /**
@@ -38,5 +50,5 @@
  */
 void wusbhc_pal_unregister(struct wusbhc *wusbhc)
 {
-	uwb_pal_unregister(wusbhc->uwb_rc, &wusbhc->pal);
+	uwb_pal_unregister(&wusbhc->pal);
 }
diff --git a/drivers/usb/wusbcore/reservation.c b/drivers/usb/wusbcore/reservation.c
index fc63e77..4ed9736 100644
--- a/drivers/usb/wusbcore/reservation.c
+++ b/drivers/usb/wusbcore/reservation.c
@@ -48,18 +48,19 @@
 {
 	struct wusbhc *wusbhc = rsv->pal_priv;
 	struct device *dev = wusbhc->dev;
+	struct uwb_mas_bm mas;
 	char buf[72];
 
 	switch (rsv->state) {
 	case UWB_RSV_STATE_O_ESTABLISHED:
-		bitmap_scnprintf(buf, sizeof(buf), rsv->mas.bm, UWB_NUM_MAS);
+		uwb_rsv_get_usable_mas(rsv, &mas);
+		bitmap_scnprintf(buf, sizeof(buf), mas.bm, UWB_NUM_MAS);
 		dev_dbg(dev, "established reservation: %s\n", buf);
-		wusbhc_bwa_set(wusbhc, rsv->stream, &rsv->mas);
+		wusbhc_bwa_set(wusbhc, rsv->stream, &mas);
 		break;
 	case UWB_RSV_STATE_NONE:
 		dev_dbg(dev, "removed reservation\n");
 		wusbhc_bwa_set(wusbhc, 0, NULL);
-		wusbhc->rsv = NULL;
 		break;
 	default:
 		dev_dbg(dev, "unexpected reservation state: %d\n", rsv->state);
@@ -86,13 +87,12 @@
 	bcid.data[0] = wusbhc->cluster_id;
 	bcid.data[1] = 0;
 
-	rsv->owner = &rc->uwb_dev;
 	rsv->target.type = UWB_RSV_TARGET_DEVADDR;
 	rsv->target.devaddr = bcid;
 	rsv->type = UWB_DRP_TYPE_PRIVATE;
-	rsv->max_mas = 256;
-	rsv->min_mas = 16;  /* one MAS per zone? */
-	rsv->sparsity = 16; /* at least one MAS in each zone? */
+	rsv->max_mas = 256; /* try to get as much as possible */
+	rsv->min_mas = 15;  /* one MAS per zone */
+	rsv->max_interval = 1; /* max latency is one zone */
 	rsv->is_multicast = true;
 
 	ret = uwb_rsv_establish(rsv);
@@ -105,11 +105,14 @@
 
 
 /**
- * wusbhc_rsv_terminate - terminate any cluster reservation
+ * wusbhc_rsv_terminate - terminate the cluster reservation
  * @wusbhc: the WUSB host whose reservation is to be terminated
  */
 void wusbhc_rsv_terminate(struct wusbhc *wusbhc)
 {
-	if (wusbhc->rsv)
+	if (wusbhc->rsv) {
 		uwb_rsv_terminate(wusbhc->rsv);
+		uwb_rsv_destroy(wusbhc->rsv);
+		wusbhc->rsv = NULL;
+	}
 }
diff --git a/drivers/usb/wusbcore/rh.c b/drivers/usb/wusbcore/rh.c
index 267a643..95c6fa3 100644
--- a/drivers/usb/wusbcore/rh.c
+++ b/drivers/usb/wusbcore/rh.c
@@ -71,19 +71,20 @@
  */
 #include "wusbhc.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
 /*
  * Reset a fake port
  *
- * This can be called to reset a port from any other state or to reset
- * it when connecting. In Wireless USB they are different; when doing
- * a new connect that involves going over the authentication. When
- * just reseting, its a different story.
+ * Using a Reset Device IE is too heavyweight as it causes the device
+ * to enter the UnConnected state and leave the cluster, this can mean
+ * that when the device reconnects it is connected to a different fake
+ * port.
  *
- * The Linux USB stack resets a port twice before it considers it
- * enabled, so we have to detect and ignore that.
+ * Instead, reset authenticated devices with a SetAddress(0), followed
+ * by a SetAddresss(AuthAddr).
+ *
+ * For unauthenticated devices just pretend to reset but do nothing.
+ * If the device initialization continues to fail it will eventually
+ * time out after TrustTimeout and enter the UnConnected state.
  *
  * @wusbhc is assumed referenced and @wusbhc->mutex unlocked.
  *
@@ -97,20 +98,20 @@
 {
 	int result = 0;
 	struct wusb_port *port = wusb_port_by_idx(wusbhc, port_idx);
+	struct wusb_dev *wusb_dev = port->wusb_dev;
 
-	d_fnstart(3, wusbhc->dev, "(wusbhc %p port_idx %u)\n",
-		  wusbhc, port_idx);
-	if (port->reset_count == 0) {
-		wusbhc_devconnect_auth(wusbhc, port_idx);
-		port->reset_count++;
-	} else if (port->reset_count == 1)
-		/* see header */
-		d_printf(2, wusbhc->dev, "Ignoring second reset on port_idx "
-			"%u\n", port_idx);
+	port->status |= USB_PORT_STAT_RESET;
+	port->change |= USB_PORT_STAT_C_RESET;
+
+	if (wusb_dev->addr & WUSB_DEV_ADDR_UNAUTH)
+		result = 0;
 	else
-		result = wusbhc_dev_reset(wusbhc, port_idx);
-	d_fnend(3, wusbhc->dev, "(wusbhc %p port_idx %u) = %d\n",
-		wusbhc, port_idx, result);
+		result = wusb_dev_update_address(wusbhc, wusb_dev);
+
+	port->status &= ~USB_PORT_STAT_RESET;
+	port->status |= USB_PORT_STAT_ENABLE;
+	port->change |= USB_PORT_STAT_C_RESET | USB_PORT_STAT_C_ENABLE;	
+
 	return result;
 }
 
@@ -138,7 +139,6 @@
 	size_t cnt, size;
 	unsigned long *buf = (unsigned long *) _buf;
 
-	d_fnstart(1, wusbhc->dev, "(wusbhc %p)\n", wusbhc);
 	/* WE DON'T LOCK, see comment */
 	size = wusbhc->ports_max + 1 /* hub bit */;
 	size = (size + 8 - 1) / 8;	/* round to bytes */
@@ -147,8 +147,6 @@
 			set_bit(cnt + 1, buf);
 		else
 			clear_bit(cnt + 1, buf);
-	d_fnend(1, wusbhc->dev, "(wusbhc %p) %u, buffer:\n", wusbhc, (int)size);
-	d_dump(1, wusbhc->dev, _buf, size);
 	return size;
 }
 EXPORT_SYMBOL_GPL(wusbhc_rh_status_data);
@@ -197,9 +195,7 @@
 static int wusbhc_rh_clear_hub_feat(struct wusbhc *wusbhc, u16 feature)
 {
 	int result;
-	struct device *dev = wusbhc->dev;
 
-	d_fnstart(4, dev, "(%p, feature 0x%04u)\n", wusbhc, feature);
 	switch (feature) {
 	case C_HUB_LOCAL_POWER:
 		/* FIXME: maybe plug bit 0 to the power input status,
@@ -211,7 +207,6 @@
 	default:
 		result = -EPIPE;
 	}
-	d_fnend(4, dev, "(%p, feature 0x%04u), %d\n", wusbhc, feature, result);
 	return result;
 }
 
@@ -238,14 +233,10 @@
 static int wusbhc_rh_set_port_feat(struct wusbhc *wusbhc, u16 feature,
 				   u8 selector, u8 port_idx)
 {
-	int result = -EINVAL;
 	struct device *dev = wusbhc->dev;
 
-	d_fnstart(4, dev, "(feat 0x%04u, selector 0x%u, port_idx %d)\n",
-		  feature, selector, port_idx);
-
 	if (port_idx > wusbhc->ports_max)
-		goto error;
+		return -EINVAL;
 
 	switch (feature) {
 		/* According to USB2.0[11.24.2.13]p2, these features
@@ -255,35 +246,27 @@
 	case USB_PORT_FEAT_C_SUSPEND:
 	case USB_PORT_FEAT_C_CONNECTION:
 	case USB_PORT_FEAT_C_RESET:
-		result = 0;
-		break;
-
+		return 0;
 	case USB_PORT_FEAT_POWER:
 		/* No such thing, but we fake it works */
 		mutex_lock(&wusbhc->mutex);
 		wusb_port_by_idx(wusbhc, port_idx)->status |= USB_PORT_STAT_POWER;
 		mutex_unlock(&wusbhc->mutex);
-		result = 0;
-		break;
+		return 0;
 	case USB_PORT_FEAT_RESET:
-		result = wusbhc_rh_port_reset(wusbhc, port_idx);
-		break;
+		return wusbhc_rh_port_reset(wusbhc, port_idx);
 	case USB_PORT_FEAT_ENABLE:
 	case USB_PORT_FEAT_SUSPEND:
 		dev_err(dev, "(port_idx %d) set feat %d/%d UNIMPLEMENTED\n",
 			port_idx, feature, selector);
-		result = -ENOSYS;
-		break;
+		return -ENOSYS;
 	default:
 		dev_err(dev, "(port_idx %d) set feat %d/%d UNKNOWN\n",
 			port_idx, feature, selector);
-		result = -EPIPE;
-		break;
+		return -EPIPE;
 	}
-error:
-	d_fnend(4, dev, "(feat 0x%04u, selector 0x%u, port_idx %d) = %d\n",
-		feature, selector, port_idx, result);
-	return result;
+
+	return 0;
 }
 
 /*
@@ -294,17 +277,13 @@
 static int wusbhc_rh_clear_port_feat(struct wusbhc *wusbhc, u16 feature,
 				     u8 selector, u8 port_idx)
 {
-	int result = -EINVAL;
+	int result = 0;
 	struct device *dev = wusbhc->dev;
 
-	d_fnstart(4, dev, "(wusbhc %p feat 0x%04x selector %d port_idx %d)\n",
-		  wusbhc, feature, selector, port_idx);
-
 	if (port_idx > wusbhc->ports_max)
-		goto error;
+		return -EINVAL;
 
 	mutex_lock(&wusbhc->mutex);
-	result = 0;
 	switch (feature) {
 	case USB_PORT_FEAT_POWER:	/* fake port always on */
 		/* According to USB2.0[11.24.2.7.1.4], no need to implement? */
@@ -324,10 +303,8 @@
 		break;
 	case USB_PORT_FEAT_SUSPEND:
 	case USB_PORT_FEAT_C_SUSPEND:
-	case 0xffff:		/* ??? FIXME */
 		dev_err(dev, "(port_idx %d) Clear feat %d/%d UNIMPLEMENTED\n",
 			port_idx, feature, selector);
-		/* dump_stack(); */
 		result = -ENOSYS;
 		break;
 	default:
@@ -337,9 +314,7 @@
 		break;
 	}
 	mutex_unlock(&wusbhc->mutex);
-error:
-	d_fnend(4, dev, "(wusbhc %p feat 0x%04x selector %d port_idx %d) = "
-		"%d\n", wusbhc, feature, selector, port_idx, result);
+
 	return result;
 }
 
@@ -351,22 +326,17 @@
 static int wusbhc_rh_get_port_status(struct wusbhc *wusbhc, u16 port_idx,
 				     u32 *_buf, u16 wLength)
 {
-	int result = -EINVAL;
 	u16 *buf = (u16 *) _buf;
 
-	d_fnstart(1, wusbhc->dev, "(wusbhc %p port_idx %u wLength %u)\n",
-		  wusbhc, port_idx, wLength);
 	if (port_idx > wusbhc->ports_max)
-		goto error;
+		return -EINVAL;
+
 	mutex_lock(&wusbhc->mutex);
 	buf[0] = cpu_to_le16(wusb_port_by_idx(wusbhc, port_idx)->status);
 	buf[1] = cpu_to_le16(wusb_port_by_idx(wusbhc, port_idx)->change);
-	result = 0;
 	mutex_unlock(&wusbhc->mutex);
-error:
-	d_fnend(1, wusbhc->dev, "(wusbhc %p) = %d, buffer:\n", wusbhc, result);
-	d_dump(1, wusbhc->dev, _buf, wLength);
-	return result;
+
+	return 0;
 }
 
 /*
diff --git a/drivers/usb/wusbcore/security.c b/drivers/usb/wusbcore/security.c
index a101cad..f4aa28e 100644
--- a/drivers/usb/wusbcore/security.c
+++ b/drivers/usb/wusbcore/security.c
@@ -27,19 +27,6 @@
 #include <linux/random.h>
 #include "wusbhc.h"
 
-/*
- * DEBUG & SECURITY WARNING!!!!
- *
- * If you enable this past 1, the debug code will weaken the
- * cryptographic safety of the system (on purpose, for debugging).
- *
- * Weaken means:
- *   we print secret keys and intermediate values all the way,
- */
-#undef D_LOCAL
-#define D_LOCAL 2
-#include <linux/uwb/debug.h>
-
 static void wusbhc_set_gtk_callback(struct urb *urb);
 static void wusbhc_gtk_rekey_done_work(struct work_struct *work);
 
@@ -219,7 +206,6 @@
 	const void *itr, *top;
 	char buf[64];
 
-	d_fnstart(3, dev, "(usb_dev %p, wusb_dev %p)\n", usb_dev, wusb_dev);
 	result = usb_get_descriptor(usb_dev, USB_DT_SECURITY,
 				    0, &secd, sizeof(secd));
 	if (result < sizeof(secd)) {
@@ -228,8 +214,6 @@
 		goto error_secd;
 	}
 	secd_size = le16_to_cpu(secd.wTotalLength);
-	d_printf(5, dev, "got %d bytes of sec descriptor, total is %d\n",
-		 result, secd_size);
 	secd_buf = kmalloc(secd_size, GFP_KERNEL);
 	if (secd_buf == NULL) {
 		dev_err(dev, "Can't allocate space for security descriptors\n");
@@ -242,7 +226,6 @@
 			"not enough data: %d\n", result);
 		goto error_secd_all;
 	}
-	d_printf(5, dev, "got %d bytes of sec descriptors\n", result);
 	bytes = 0;
 	itr = secd_buf + sizeof(secd);
 	top = secd_buf + result;
@@ -279,14 +262,12 @@
 		goto error_no_ccm1;
 	}
 	wusb_dev->ccm1_etd = *ccm1_etd;
-	dev_info(dev, "supported encryption: %s; using %s (0x%02x/%02x)\n",
-		 buf, wusb_et_name(ccm1_etd->bEncryptionType),
-		 ccm1_etd->bEncryptionValue, ccm1_etd->bAuthKeyIndex);
+	dev_dbg(dev, "supported encryption: %s; using %s (0x%02x/%02x)\n",
+		buf, wusb_et_name(ccm1_etd->bEncryptionType),
+		ccm1_etd->bEncryptionValue, ccm1_etd->bAuthKeyIndex);
 	result = 0;
 	kfree(secd_buf);
 out:
-	d_fnend(3, dev, "(usb_dev %p, wusb_dev %p) = %d\n",
-		usb_dev, wusb_dev, result);
 	return result;
 
 
@@ -303,32 +284,6 @@
 	/* Nothing so far */
 }
 
-static void hs_printk(unsigned level, struct device *dev,
-		      struct usb_handshake *hs)
-{
-	d_printf(level, dev,
-		 "  bMessageNumber: %u\n"
-		 "  bStatus:        %u\n"
-		 "  tTKID:          %02x %02x %02x\n"
-		 "  CDID:           %02x %02x %02x %02x %02x %02x %02x %02x\n"
-		 "                  %02x %02x %02x %02x %02x %02x %02x %02x\n"
-		 "  nonce:          %02x %02x %02x %02x %02x %02x %02x %02x\n"
-		 "                  %02x %02x %02x %02x %02x %02x %02x %02x\n"
-		 "  MIC:            %02x %02x %02x %02x %02x %02x %02x %02x\n",
-		 hs->bMessageNumber, hs->bStatus,
-		 hs->tTKID[2], hs->tTKID[1], hs->tTKID[0],
-		 hs->CDID[0], hs->CDID[1], hs->CDID[2], hs->CDID[3],
-		 hs->CDID[4], hs->CDID[5], hs->CDID[6], hs->CDID[7],
-		 hs->CDID[8], hs->CDID[9], hs->CDID[10], hs->CDID[11],
-		 hs->CDID[12], hs->CDID[13], hs->CDID[14], hs->CDID[15],
-		 hs->nonce[0], hs->nonce[1], hs->nonce[2], hs->nonce[3],
-		 hs->nonce[4], hs->nonce[5], hs->nonce[6], hs->nonce[7],
-		 hs->nonce[8], hs->nonce[9], hs->nonce[10], hs->nonce[11],
-		 hs->nonce[12], hs->nonce[13], hs->nonce[14], hs->nonce[15],
-		 hs->MIC[0], hs->MIC[1], hs->MIC[2], hs->MIC[3],
-		 hs->MIC[4], hs->MIC[5], hs->MIC[6], hs->MIC[7]);
-}
-
 /**
  * Update the address of an unauthenticated WUSB device
  *
@@ -338,8 +293,7 @@
  * Before the device's address (as known by it) was usb_dev->devnum |
  * 0x80 (unauthenticated address). With this we update it to usb_dev->devnum.
  */
-static int wusb_dev_update_address(struct wusbhc *wusbhc,
-				   struct wusb_dev *wusb_dev)
+int wusb_dev_update_address(struct wusbhc *wusbhc, struct wusb_dev *wusb_dev)
 {
 	int result = -ENOMEM;
 	struct usb_device *usb_dev = wusb_dev->usb_dev;
@@ -422,9 +376,6 @@
 	get_random_bytes(&hs[0].nonce, sizeof(hs[0].nonce));
 	memset(hs[0].MIC, 0, sizeof(hs[0].MIC));	/* Per WUSB1.0[T7-22] */
 
-	d_printf(1, dev, "I: sending hs1:\n");
-	hs_printk(2, dev, &hs[0]);
-
 	result = usb_control_msg(
 		usb_dev, usb_sndctrlpipe(usb_dev, 0),
 		USB_REQ_SET_HANDSHAKE,
@@ -445,8 +396,6 @@
 		dev_err(dev, "Handshake2: request failed: %d\n", result);
 		goto error_hs2;
 	}
-	d_printf(1, dev, "got HS2:\n");
-	hs_printk(2, dev, &hs[1]);
 
 	result = -EINVAL;
 	if (hs[1].bMessageNumber != 2) {
@@ -487,10 +436,6 @@
 			result);
 		goto error_hs2;
 	}
-	d_printf(2, dev, "KCK:\n");
-	d_dump(2, dev, keydvt_out.kck, sizeof(keydvt_out.kck));
-	d_printf(2, dev, "PTK:\n");
-	d_dump(2, dev, keydvt_out.ptk, sizeof(keydvt_out.ptk));
 
 	/* Compute MIC and verify it */
 	result = wusb_oob_mic(mic, keydvt_out.kck, &ccm_n, &hs[1]);
@@ -500,8 +445,6 @@
 		goto error_hs2;
 	}
 
-	d_printf(2, dev, "MIC:\n");
-	d_dump(2, dev, mic, sizeof(mic));
 	if (memcmp(hs[1].MIC, mic, sizeof(hs[1].MIC))) {
 		dev_err(dev, "Handshake2 failed: MIC mismatch\n");
 		goto error_hs2;
@@ -521,9 +464,6 @@
 		goto error_hs2;
 	}
 
-	d_printf(1, dev, "I: sending hs3:\n");
-	hs_printk(2, dev, &hs[2]);
-
 	result = usb_control_msg(
 		usb_dev, usb_sndctrlpipe(usb_dev, 0),
 		USB_REQ_SET_HANDSHAKE,
@@ -534,14 +474,11 @@
 		goto error_hs3;
 	}
 
-	d_printf(1, dev, "I: turning on encryption on host for device\n");
-	d_dump(2, dev, keydvt_out.ptk, sizeof(keydvt_out.ptk));
 	result = wusbhc->set_ptk(wusbhc, wusb_dev->port_idx, tkid,
 				 keydvt_out.ptk, sizeof(keydvt_out.ptk));
 	if (result < 0)
 		goto error_wusbhc_set_ptk;
 
-	d_printf(1, dev, "I: setting a GTK\n");
 	result = wusb_dev_set_gtk(wusbhc, wusb_dev);
 	if (result < 0) {
 		dev_err(dev, "Set GTK for device: request failed: %d\n",
@@ -551,13 +488,12 @@
 
 	/* Update the device's address from unauth to auth */
 	if (usb_dev->authenticated == 0) {
-		d_printf(1, dev, "I: updating addres to auth from non-auth\n");
 		result = wusb_dev_update_address(wusbhc, wusb_dev);
 		if (result < 0)
 			goto error_dev_update_address;
 	}
 	result = 0;
-	d_printf(1, dev, "I: 4way handshke done, device authenticated\n");
+	dev_info(dev, "device authenticated\n");
 
 error_dev_update_address:
 error_wusbhc_set_gtk:
@@ -570,10 +506,8 @@
 	memset(&keydvt_in, 0, sizeof(keydvt_in));
 	memset(&ccm_n, 0, sizeof(ccm_n));
 	memset(mic, 0, sizeof(mic));
-	if (result < 0) {
-		/* error path */
+	if (result < 0)
 		wusb_dev_set_encryption(usb_dev, 0);
-	}
 error_dev_set_encryption:
 	kfree(hs);
 error_kzalloc:
diff --git a/drivers/usb/wusbcore/wa-nep.c b/drivers/usb/wusbcore/wa-nep.c
index 3f54299..17d2626 100644
--- a/drivers/usb/wusbcore/wa-nep.c
+++ b/drivers/usb/wusbcore/wa-nep.c
@@ -51,7 +51,7 @@
  */
 #include <linux/workqueue.h>
 #include <linux/ctype.h>
-#include <linux/uwb/debug.h>
+
 #include "wa-hc.h"
 #include "wusbhc.h"
 
@@ -139,13 +139,10 @@
 			/* FIXME: unimplemented WA NOTIFs */
 			/* fallthru */
 		default:
-			if (printk_ratelimit()) {
-				dev_err(dev, "HWA: unknown notification 0x%x, "
-					"%zu bytes; discarding\n",
-					notif_hdr->bNotifyType,
-					(size_t)notif_hdr->bLength);
-				dump_bytes(dev, notif_hdr, 16);
-			}
+			dev_err(dev, "HWA: unknown notification 0x%x, "
+				"%zu bytes; discarding\n",
+				notif_hdr->bNotifyType,
+				(size_t)notif_hdr->bLength);
 			break;
 		}
 	}
@@ -160,12 +157,9 @@
 	 * discard the data, as this should not happen.
 	 */
 exhausted_buffer:
-	if (!printk_ratelimit())
-		goto out;
 	dev_warn(dev, "HWA: device sent short notification, "
 		 "%d bytes missing; discarding %d bytes.\n",
 		 missing, (int)size);
-	dump_bytes(dev, itr, size);
 	goto out;
 }
 
diff --git a/drivers/usb/wusbcore/wa-rpipe.c b/drivers/usb/wusbcore/wa-rpipe.c
index f18e4aa..7369655 100644
--- a/drivers/usb/wusbcore/wa-rpipe.c
+++ b/drivers/usb/wusbcore/wa-rpipe.c
@@ -60,13 +60,10 @@
 #include <linux/init.h>
 #include <asm/atomic.h>
 #include <linux/bitmap.h>
+
 #include "wusbhc.h"
 #include "wa-hc.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
-
 static int __rpipe_get_descr(struct wahc *wa,
 			     struct usb_rpipe_descriptor *descr, u16 index)
 {
@@ -76,7 +73,6 @@
 	/* Get the RPIPE descriptor -- we cannot use the usb_get_descriptor()
 	 * function because the arguments are different.
 	 */
-	d_printf(1, dev, "rpipe %u: get descr\n", index);
 	result = usb_control_msg(
 		wa->usb_dev, usb_rcvctrlpipe(wa->usb_dev, 0),
 		USB_REQ_GET_DESCRIPTOR,
@@ -115,7 +111,6 @@
 	/* we cannot use the usb_get_descriptor() function because the
 	 * arguments are different.
 	 */
-	d_printf(1, dev, "rpipe %u: set descr\n", index);
 	result = usb_control_msg(
 		wa->usb_dev, usb_sndctrlpipe(wa->usb_dev, 0),
 		USB_REQ_SET_DESCRIPTOR,
@@ -174,13 +169,12 @@
 {
 	struct wa_rpipe *rpipe = container_of(_rpipe, struct wa_rpipe, refcnt);
 	u8 index = le16_to_cpu(rpipe->descr.wRPipeIndex);
-	d_fnstart(1, NULL, "(rpipe %p %u)\n", rpipe, index);
+
 	if (rpipe->ep)
 		rpipe->ep->hcpriv = NULL;
 	rpipe_put_idx(rpipe->wa, index);
 	wa_put(rpipe->wa);
 	kfree(rpipe);
-	d_fnend(1, NULL, "(rpipe %p %u)\n", rpipe, index);
 }
 EXPORT_SYMBOL_GPL(rpipe_destroy);
 
@@ -202,7 +196,6 @@
 	struct wa_rpipe *rpipe;
 	struct device *dev = &wa->usb_iface->dev;
 
-	d_fnstart(3, dev, "(wa %p crs 0x%02x)\n", wa, crs);
 	rpipe = kzalloc(sizeof(*rpipe), gfp);
 	if (rpipe == NULL)
 		return -ENOMEM;
@@ -223,14 +216,12 @@
 	}
 	*prpipe = NULL;
 	kfree(rpipe);
-	d_fnend(3, dev, "(wa %p crs 0x%02x) = -ENXIO\n", wa, crs);
 	return -ENXIO;
 
 found:
 	set_bit(rpipe_idx, wa->rpipe_bm);
 	rpipe->wa = wa_get(wa);
 	*prpipe = rpipe;
-	d_fnstart(3, dev, "(wa %p crs 0x%02x) = 0\n", wa, crs);
 	return 0;
 }
 
@@ -239,7 +230,6 @@
 	int result;
 	struct device *dev = &wa->usb_iface->dev;
 
-	d_printf(1, dev, "rpipe %u: reset\n", index);
 	result = usb_control_msg(
 		wa->usb_dev, usb_sndctrlpipe(wa->usb_dev, 0),
 		USB_REQ_RPIPE_RESET,
@@ -276,7 +266,6 @@
 	struct usb_descriptor_header *hdr;
 	struct usb_wireless_ep_comp_descriptor *epcd;
 
-	d_fnstart(3, dev, "(ep %p)\n", ep);
 	if (ep->desc.bEndpointAddress == 0) {
 		epcd = &epc0;
 		goto out;
@@ -310,7 +299,6 @@
 		itr_size -= hdr->bDescriptorType;
 	}
 out:
-	d_fnend(3, dev, "(ep %p) = %p\n", ep, epcd);
 	return epcd;
 }
 
@@ -329,8 +317,6 @@
 	struct usb_wireless_ep_comp_descriptor *epcd;
 	u8 unauth;
 
-	d_fnstart(3, dev, "(rpipe %p wa %p ep %p, urb %p)\n",
-		    rpipe, wa, ep, urb);
 	epcd = rpipe_epc_find(dev, ep);
 	if (epcd == NULL) {
 		dev_err(dev, "ep 0x%02x: can't find companion descriptor\n",
@@ -350,10 +336,12 @@
 	/* FIXME: use maximum speed as supported or recommended by device */
 	rpipe->descr.bSpeed = usb_pipeendpoint(urb->pipe) == 0 ?
 		UWB_PHY_RATE_53 : UWB_PHY_RATE_200;
-	d_printf(2, dev, "addr %u (0x%02x) rpipe #%u ep# %u speed %d\n",
-		 urb->dev->devnum, urb->dev->devnum | unauth,
-		 le16_to_cpu(rpipe->descr.wRPipeIndex),
-		 usb_pipeendpoint(urb->pipe), rpipe->descr.bSpeed);
+
+	dev_dbg(dev, "addr %u (0x%02x) rpipe #%u ep# %u speed %d\n",
+		urb->dev->devnum, urb->dev->devnum | unauth,
+		le16_to_cpu(rpipe->descr.wRPipeIndex),
+		usb_pipeendpoint(urb->pipe), rpipe->descr.bSpeed);
+
 	/* see security.c:wusb_update_address() */
 	if (unlikely(urb->dev->devnum == 0x80))
 		rpipe->descr.bDeviceAddress = 0;
@@ -384,8 +372,6 @@
 	}
 	result = 0;
 error:
-	d_fnend(3, dev, "(rpipe %p wa %p ep %p urb %p) = %d\n",
-		  rpipe, wa, ep, urb, result);
 	return result;
 }
 
@@ -405,8 +391,6 @@
 	u8 unauth = (usb_dev->wusb && !usb_dev->authenticated) ? 0x80 : 0;
 	u8 portnum = wusb_port_no_to_idx(urb->dev->portnum);
 
-	d_fnstart(3, dev, "(rpipe %p wa %p ep %p, urb %p)\n",
-		    rpipe, wa, ep, urb);
 #define AIM_CHECK(rdf, val, text)					\
 	do {								\
 		if (rpipe->descr.rdf != (val)) {			\
@@ -451,8 +435,6 @@
 	struct wa_rpipe *rpipe;
 	u8 eptype;
 
-	d_fnstart(3, dev, "(wa %p ep %p urb %p gfp 0x%08x)\n", wa, ep, urb,
-		  gfp);
 	mutex_lock(&wa->rpipe_mutex);
 	rpipe = ep->hcpriv;
 	if (rpipe != NULL) {
@@ -462,9 +444,9 @@
 				goto error;
 		}
 		__rpipe_get(rpipe);
-		d_printf(2, dev, "ep 0x%02x: reusing rpipe %u\n",
-			 ep->desc.bEndpointAddress,
-			 le16_to_cpu(rpipe->descr.wRPipeIndex));
+		dev_dbg(dev, "ep 0x%02x: reusing rpipe %u\n",
+			ep->desc.bEndpointAddress,
+			le16_to_cpu(rpipe->descr.wRPipeIndex));
 	} else {
 		/* hmm, assign idle rpipe, aim it */
 		result = -ENOBUFS;
@@ -480,14 +462,12 @@
 		ep->hcpriv = rpipe;
 		rpipe->ep = ep;
 		__rpipe_get(rpipe);	/* for caching into ep->hcpriv */
-		d_printf(2, dev, "ep 0x%02x: using rpipe %u\n",
-			 ep->desc.bEndpointAddress,
-			 le16_to_cpu(rpipe->descr.wRPipeIndex));
+		dev_dbg(dev, "ep 0x%02x: using rpipe %u\n",
+			ep->desc.bEndpointAddress,
+			le16_to_cpu(rpipe->descr.wRPipeIndex));
 	}
-	d_dump(4, dev, &rpipe->descr, sizeof(rpipe->descr));
 error:
 	mutex_unlock(&wa->rpipe_mutex);
-	d_fnend(3, dev, "(wa %p ep %p urb %p gfp 0x%08x)\n", wa, ep, urb, gfp);
 	return result;
 }
 
@@ -507,7 +487,7 @@
 void wa_rpipes_destroy(struct wahc *wa)
 {
 	struct device *dev = &wa->usb_iface->dev;
-	d_fnstart(3, dev, "(wa %p)\n", wa);
+
 	if (!bitmap_empty(wa->rpipe_bm, wa->rpipes)) {
 		char buf[256];
 		WARN_ON(1);
@@ -515,7 +495,6 @@
 		dev_err(dev, "BUG: pipes not released on exit: %s\n", buf);
 	}
 	kfree(wa->rpipe_bm);
-	d_fnend(3, dev, "(wa %p)\n", wa);
 }
 
 /*
@@ -530,33 +509,20 @@
  */
 void rpipe_ep_disable(struct wahc *wa, struct usb_host_endpoint *ep)
 {
-	struct device *dev = &wa->usb_iface->dev;
 	struct wa_rpipe *rpipe;
-	d_fnstart(2, dev, "(wa %p ep %p)\n", wa, ep);
+
 	mutex_lock(&wa->rpipe_mutex);
 	rpipe = ep->hcpriv;
 	if (rpipe != NULL) {
-		unsigned rc = atomic_read(&rpipe->refcnt.refcount);
-		int result;
 		u16 index = le16_to_cpu(rpipe->descr.wRPipeIndex);
 
-		if (rc != 1)
-			d_printf(1, dev, "(wa %p ep %p) rpipe %p refcnt %u\n",
-				 wa, ep, rpipe, rc);
-
-		d_printf(1, dev, "rpipe %u: abort\n", index);
-		result = usb_control_msg(
+		usb_control_msg(
 			wa->usb_dev, usb_rcvctrlpipe(wa->usb_dev, 0),
 			USB_REQ_RPIPE_ABORT,
 			USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_RPIPE,
 			0, index, NULL, 0, 1000 /* FIXME: arbitrary */);
-		if (result < 0 && result != -ENODEV /* dev is gone */)
-			d_printf(1, dev, "(wa %p rpipe %u): abort failed: %d\n",
-				 wa, index, result);
 		rpipe_put(rpipe);
 	}
 	mutex_unlock(&wa->rpipe_mutex);
-	d_fnend(2, dev, "(wa %p ep %p)\n", wa, ep);
-	return;
 }
 EXPORT_SYMBOL_GPL(rpipe_ep_disable);
diff --git a/drivers/usb/wusbcore/wa-xfer.c b/drivers/usb/wusbcore/wa-xfer.c
index c038635..238a96a 100644
--- a/drivers/usb/wusbcore/wa-xfer.c
+++ b/drivers/usb/wusbcore/wa-xfer.c
@@ -82,13 +82,10 @@
 #include <linux/init.h>
 #include <linux/spinlock.h>
 #include <linux/hash.h>
+
 #include "wa-hc.h"
 #include "wusbhc.h"
 
-#undef D_LOCAL
-#define D_LOCAL 0 /* 0 disabled, > 0 different levels... */
-#include <linux/uwb/debug.h>
-
 enum {
 	WA_SEGS_MAX = 255,
 };
@@ -180,7 +177,6 @@
 		}
 	}
 	kfree(xfer);
-	d_printf(2, NULL, "xfer %p destroyed\n", xfer);
 }
 
 static void wa_xfer_get(struct wa_xfer *xfer)
@@ -190,10 +186,7 @@
 
 static void wa_xfer_put(struct wa_xfer *xfer)
 {
-	d_fnstart(3, NULL, "(xfer %p) -- ref count bef put %d\n",
-		    xfer, atomic_read(&xfer->refcnt.refcount));
 	kref_put(&xfer->refcnt, wa_xfer_destroy);
-	d_fnend(3, NULL, "(xfer %p) = void\n", xfer);
 }
 
 /*
@@ -209,7 +202,7 @@
 static void wa_xfer_giveback(struct wa_xfer *xfer)
 {
 	unsigned long flags;
-	d_fnstart(3, NULL, "(xfer %p)\n", xfer);
+
 	spin_lock_irqsave(&xfer->wa->xfer_list_lock, flags);
 	list_del_init(&xfer->list_node);
 	spin_unlock_irqrestore(&xfer->wa->xfer_list_lock, flags);
@@ -217,7 +210,6 @@
 	wusbhc_giveback_urb(xfer->wa->wusb, xfer->urb, xfer->result);
 	wa_put(xfer->wa);
 	wa_xfer_put(xfer);
-	d_fnend(3, NULL, "(xfer %p) = void\n", xfer);
 }
 
 /*
@@ -227,13 +219,10 @@
  */
 static void wa_xfer_completion(struct wa_xfer *xfer)
 {
-	d_fnstart(3, NULL, "(xfer %p)\n", xfer);
 	if (xfer->wusb_dev)
 		wusb_dev_put(xfer->wusb_dev);
 	rpipe_put(xfer->ep->hcpriv);
 	wa_xfer_giveback(xfer);
-	d_fnend(3, NULL, "(xfer %p) = void\n", xfer);
-	return;
 }
 
 /*
@@ -243,12 +232,12 @@
  */
 static unsigned __wa_xfer_is_done(struct wa_xfer *xfer)
 {
+	struct device *dev = &xfer->wa->usb_iface->dev;
 	unsigned result, cnt;
 	struct wa_seg *seg;
 	struct urb *urb = xfer->urb;
 	unsigned found_short = 0;
 
-	d_fnstart(3, NULL, "(xfer %p)\n", xfer);
 	result = xfer->segs_done == xfer->segs_submitted;
 	if (result == 0)
 		goto out;
@@ -258,10 +247,8 @@
 		switch (seg->status) {
 		case WA_SEG_DONE:
 			if (found_short && seg->result > 0) {
-				if (printk_ratelimit())
-					printk(KERN_ERR "xfer %p#%u: bad short "
-					       "segments (%zu)\n", xfer, cnt,
-					       seg->result);
+				dev_dbg(dev, "xfer %p#%u: bad short segments (%zu)\n",
+					xfer, cnt, seg->result);
 				urb->status = -EINVAL;
 				goto out;
 			}
@@ -269,36 +256,30 @@
 			if (seg->result < xfer->seg_size
 			    && cnt != xfer->segs-1)
 				found_short = 1;
-			d_printf(2, NULL, "xfer %p#%u: DONE short %d "
-				 "result %zu urb->actual_length %d\n",
-				 xfer, seg->index, found_short, seg->result,
-				 urb->actual_length);
+			dev_dbg(dev, "xfer %p#%u: DONE short %d "
+				"result %zu urb->actual_length %d\n",
+				xfer, seg->index, found_short, seg->result,
+				urb->actual_length);
 			break;
 		case WA_SEG_ERROR:
 			xfer->result = seg->result;
-			d_printf(2, NULL, "xfer %p#%u: ERROR result %zu\n",
-				 xfer, seg->index, seg->result);
+			dev_dbg(dev, "xfer %p#%u: ERROR result %zu\n",
+				xfer, seg->index, seg->result);
 			goto out;
 		case WA_SEG_ABORTED:
-			WARN_ON(urb->status != -ECONNRESET
-				&& urb->status != -ENOENT);
-			d_printf(2, NULL, "xfer %p#%u ABORTED: result %d\n",
-				 xfer, seg->index, urb->status);
+			dev_dbg(dev, "xfer %p#%u ABORTED: result %d\n",
+				xfer, seg->index, urb->status);
 			xfer->result = urb->status;
 			goto out;
 		default:
-			/* if (printk_ratelimit()) */
-				printk(KERN_ERR "xfer %p#%u: "
-				       "is_done bad state %d\n",
-				       xfer, cnt, seg->status);
+			dev_warn(dev, "xfer %p#%u: is_done bad state %d\n",
+				 xfer, cnt, seg->status);
 			xfer->result = -EINVAL;
-			WARN_ON(1);
 			goto out;
 		}
 	}
 	xfer->result = 0;
 out:
-	d_fnend(3, NULL, "(xfer %p) = void\n", xfer);
 	return result;
 }
 
@@ -424,8 +405,6 @@
 	struct urb *urb = xfer->urb;
 	struct wa_rpipe *rpipe = xfer->ep->hcpriv;
 
-	d_fnstart(3, dev, "(xfer %p [rpipe %p] urb %p)\n",
-		  xfer, rpipe, urb);
 	switch (rpipe->descr.bmAttribute & 0x3) {
 	case USB_ENDPOINT_XFER_CONTROL:
 		*pxfer_type = WA_XFER_TYPE_CTL;
@@ -472,12 +451,10 @@
 	if (xfer->segs == 0 && *pxfer_type == WA_XFER_TYPE_CTL)
 		xfer->segs = 1;
 error:
-	d_fnend(3, dev, "(xfer %p [rpipe %p] urb %p) = %d\n",
-		xfer, rpipe, urb, (int)result);
 	return result;
 }
 
-/** Fill in the common request header and xfer-type specific data. */
+/* Fill in the common request header and xfer-type specific data. */
 static void __wa_xfer_setup_hdr0(struct wa_xfer *xfer,
 				 struct wa_xfer_hdr *xfer_hdr0,
 				 enum wa_xfer_type xfer_type,
@@ -534,14 +511,13 @@
 	unsigned rpipe_ready = 0;
 	u8 done = 0;
 
-	d_fnstart(3, NULL, "(urb %p [%d])\n", urb, urb->status);
 	switch (urb->status) {
 	case 0:
 		spin_lock_irqsave(&xfer->lock, flags);
 		wa = xfer->wa;
 		dev = &wa->usb_iface->dev;
-		d_printf(2, dev, "xfer %p#%u: data out done (%d bytes)\n",
-			   xfer, seg->index, urb->actual_length);
+		dev_dbg(dev, "xfer %p#%u: data out done (%d bytes)\n",
+			xfer, seg->index, urb->actual_length);
 		if (seg->status < WA_SEG_PENDING)
 			seg->status = WA_SEG_PENDING;
 		seg->result = urb->actual_length;
@@ -555,9 +531,8 @@
 		wa = xfer->wa;
 		dev = &wa->usb_iface->dev;
 		rpipe = xfer->ep->hcpriv;
-		if (printk_ratelimit())
-			dev_err(dev, "xfer %p#%u: data out error %d\n",
-				xfer, seg->index, urb->status);
+		dev_dbg(dev, "xfer %p#%u: data out error %d\n",
+			xfer, seg->index, urb->status);
 		if (edc_inc(&wa->nep_edc, EDC_MAX_ERRORS,
 			    EDC_ERROR_TIMEFRAME)){
 			dev_err(dev, "DTO: URB max acceptable errors "
@@ -578,7 +553,6 @@
 		if (rpipe_ready)
 			wa_xfer_delayed_run(rpipe);
 	}
-	d_fnend(3, NULL, "(urb %p [%d]) = void\n", urb, urb->status);
 }
 
 /*
@@ -610,14 +584,12 @@
 	unsigned rpipe_ready;
 	u8 done = 0;
 
-	d_fnstart(3, NULL, "(urb %p [%d])\n", urb, urb->status);
 	switch (urb->status) {
 	case 0:
 		spin_lock_irqsave(&xfer->lock, flags);
 		wa = xfer->wa;
 		dev = &wa->usb_iface->dev;
-		d_printf(2, dev, "xfer %p#%u: request done\n",
-			   xfer, seg->index);
+		dev_dbg(dev, "xfer %p#%u: request done\n", xfer, seg->index);
 		if (xfer->is_inbound && seg->status < WA_SEG_PENDING)
 			seg->status = WA_SEG_PENDING;
 		spin_unlock_irqrestore(&xfer->lock, flags);
@@ -652,7 +624,6 @@
 		if (rpipe_ready)
 			wa_xfer_delayed_run(rpipe);
 	}
-	d_fnend(3, NULL, "(urb %p [%d]) = void\n", urb, urb->status);
 }
 
 /*
@@ -750,9 +721,6 @@
 	size_t xfer_hdr_size, cnt, transfer_size;
 	struct wa_xfer_hdr *xfer_hdr0, *xfer_hdr;
 
-	d_fnstart(3, dev, "(xfer %p [rpipe %p] urb %p)\n",
-		  xfer, xfer->ep->hcpriv, urb);
-
 	result = __wa_xfer_setup_sizes(xfer, &xfer_type);
 	if (result < 0)
 		goto error_setup_sizes;
@@ -788,8 +756,6 @@
 	result = 0;
 error_setup_segs:
 error_setup_sizes:
-	d_fnend(3, dev, "(xfer %p [rpipe %p] urb %p) = %d\n",
-		xfer, xfer->ep->hcpriv, urb, result);
 	return result;
 }
 
@@ -843,9 +809,6 @@
 	struct wa_xfer *xfer;
 	unsigned long flags;
 
-	d_fnstart(1, dev, "(rpipe #%d) %d segments available\n",
-		  le16_to_cpu(rpipe->descr.wRPipeIndex),
-		  atomic_read(&rpipe->segs_available));
 	spin_lock_irqsave(&rpipe->seg_lock, flags);
 	while (atomic_read(&rpipe->segs_available) > 0
 	      && !list_empty(&rpipe->seg_list)) {
@@ -854,10 +817,8 @@
 		list_del(&seg->list_node);
 		xfer = seg->xfer;
 		result = __wa_seg_submit(rpipe, xfer, seg);
-		d_printf(1, dev, "xfer %p#%u submitted from delayed "
-			 "[%d segments available] %d\n",
-			 xfer, seg->index,
-			 atomic_read(&rpipe->segs_available), result);
+		dev_dbg(dev, "xfer %p#%u submitted from delayed [%d segments available] %d\n",
+			xfer, seg->index, atomic_read(&rpipe->segs_available), result);
 		if (unlikely(result < 0)) {
 			spin_unlock_irqrestore(&rpipe->seg_lock, flags);
 			spin_lock_irqsave(&xfer->lock, flags);
@@ -868,10 +829,6 @@
 		}
 	}
 	spin_unlock_irqrestore(&rpipe->seg_lock, flags);
-	d_fnend(1, dev, "(rpipe #%d) = void, %d segments available\n",
-		le16_to_cpu(rpipe->descr.wRPipeIndex),
-		atomic_read(&rpipe->segs_available));
-
 }
 
 /*
@@ -894,9 +851,6 @@
 	u8 available;
 	u8 empty;
 
-	d_fnstart(3, dev, "(xfer %p [rpipe %p])\n",
-		  xfer, xfer->ep->hcpriv);
-
 	spin_lock_irqsave(&wa->xfer_list_lock, flags);
 	list_add_tail(&xfer->list_node, &wa->xfer_list);
 	spin_unlock_irqrestore(&wa->xfer_list_lock, flags);
@@ -908,30 +862,24 @@
 		available = atomic_read(&rpipe->segs_available);
 		empty = list_empty(&rpipe->seg_list);
 		seg = xfer->seg[cnt];
-		d_printf(2, dev, "xfer %p#%u: available %u empty %u (%s)\n",
-			 xfer, cnt, available, empty,
-			 available == 0 || !empty ? "delayed" : "submitted");
+		dev_dbg(dev, "xfer %p#%u: available %u empty %u (%s)\n",
+			xfer, cnt, available, empty,
+			available == 0 || !empty ? "delayed" : "submitted");
 		if (available == 0 || !empty) {
-			d_printf(1, dev, "xfer %p#%u: delayed\n", xfer, cnt);
+			dev_dbg(dev, "xfer %p#%u: delayed\n", xfer, cnt);
 			seg->status = WA_SEG_DELAYED;
 			list_add_tail(&seg->list_node, &rpipe->seg_list);
 		} else {
 			result = __wa_seg_submit(rpipe, xfer, seg);
-			if (result < 0)
+			if (result < 0) {
+				__wa_xfer_abort(xfer);
 				goto error_seg_submit;
+			}
 		}
 		xfer->segs_submitted++;
 	}
-	spin_unlock_irqrestore(&rpipe->seg_lock, flags);
-	d_fnend(3, dev, "(xfer %p [rpipe %p]) = void\n", xfer,
-		xfer->ep->hcpriv);
-	return result;
-
 error_seg_submit:
-	__wa_xfer_abort(xfer);
 	spin_unlock_irqrestore(&rpipe->seg_lock, flags);
-	d_fnend(3, dev, "(xfer %p [rpipe %p]) = void\n", xfer,
-		xfer->ep->hcpriv);
 	return result;
 }
 
@@ -964,11 +912,9 @@
 	struct urb *urb = xfer->urb;
 	struct wahc *wa = xfer->wa;
 	struct wusbhc *wusbhc = wa->wusb;
-	struct device *dev = &wa->usb_iface->dev;
 	struct wusb_dev *wusb_dev;
 	unsigned done;
 
-	d_fnstart(3, dev, "(wa %p urb %p)\n", wa, urb);
 	result = rpipe_get_by_ep(wa, xfer->ep, urb, xfer->gfp);
 	if (result < 0)
 		goto error_rpipe_get;
@@ -997,7 +943,6 @@
 	if (result < 0)
 		goto error_xfer_submit;
 	spin_unlock_irqrestore(&xfer->lock, flags);
-	d_fnend(3, dev, "(wa %p urb %p) = void\n", wa, urb);
 	return;
 
 	/* this is basically wa_xfer_completion() broken up wa_xfer_giveback()
@@ -1015,7 +960,6 @@
 error_rpipe_get:
 	xfer->result = result;
 	wa_xfer_giveback(xfer);
-	d_fnend(3, dev, "(wa %p urb %p) = (void) %d\n", wa, urb, result);
 	return;
 
 error_xfer_submit:
@@ -1024,8 +968,6 @@
 	spin_unlock_irqrestore(&xfer->lock, flags);
 	if (done)
 		wa_xfer_completion(xfer);
-	d_fnend(3, dev, "(wa %p urb %p) = (void) %d\n", wa, urb, result);
-	return;
 }
 
 /*
@@ -1041,11 +983,9 @@
 void wa_urb_enqueue_run(struct work_struct *ws)
 {
 	struct wahc *wa = container_of(ws, struct wahc, xfer_work);
-	struct device *dev = &wa->usb_iface->dev;
 	struct wa_xfer *xfer, *next;
 	struct urb *urb;
 
-	d_fnstart(3, dev, "(wa %p)\n", wa);
 	spin_lock_irq(&wa->xfer_list_lock);
 	list_for_each_entry_safe(xfer, next, &wa->xfer_delayed_list,
 				 list_node) {
@@ -1059,7 +999,6 @@
 		spin_lock_irq(&wa->xfer_list_lock);
 	}
 	spin_unlock_irq(&wa->xfer_list_lock);
-	d_fnend(3, dev, "(wa %p) = void\n", wa);
 }
 EXPORT_SYMBOL_GPL(wa_urb_enqueue_run);
 
@@ -1084,9 +1023,6 @@
 	unsigned long my_flags;
 	unsigned cant_sleep = irqs_disabled() | in_atomic();
 
-	d_fnstart(3, dev, "(wa %p ep %p urb %p [%d] gfp 0x%x)\n",
-		  wa, ep, urb, urb->transfer_buffer_length, gfp);
-
 	if (urb->transfer_buffer == NULL
 	    && !(urb->transfer_flags & URB_NO_TRANSFER_DMA_MAP)
 	    && urb->transfer_buffer_length != 0) {
@@ -1108,11 +1044,13 @@
 	xfer->gfp = gfp;
 	xfer->ep = ep;
 	urb->hcpriv = xfer;
-	d_printf(2, dev, "xfer %p urb %p pipe 0x%02x [%d bytes] %s %s %s\n",
-		 xfer, urb, urb->pipe, urb->transfer_buffer_length,
-		 urb->transfer_flags & URB_NO_TRANSFER_DMA_MAP ? "dma" : "nodma",
-		 urb->pipe & USB_DIR_IN ? "inbound" : "outbound",
-		 cant_sleep ? "deferred" : "inline");
+
+	dev_dbg(dev, "xfer %p urb %p pipe 0x%02x [%d bytes] %s %s %s\n",
+		xfer, urb, urb->pipe, urb->transfer_buffer_length,
+		urb->transfer_flags & URB_NO_TRANSFER_DMA_MAP ? "dma" : "nodma",
+		urb->pipe & USB_DIR_IN ? "inbound" : "outbound",
+		cant_sleep ? "deferred" : "inline");
+
 	if (cant_sleep) {
 		usb_get_urb(urb);
 		spin_lock_irqsave(&wa->xfer_list_lock, my_flags);
@@ -1122,15 +1060,11 @@
 	} else {
 		wa_urb_enqueue_b(xfer);
 	}
-	d_fnend(3, dev, "(wa %p ep %p urb %p [%d] gfp 0x%x) = 0\n",
-		wa, ep, urb, urb->transfer_buffer_length, gfp);
 	return 0;
 
 error_dequeued:
 	kfree(xfer);
 error_kmalloc:
-	d_fnend(3, dev, "(wa %p ep %p urb %p [%d] gfp 0x%x) = %d\n",
-		wa, ep, urb, urb->transfer_buffer_length, gfp, result);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wa_urb_enqueue);
@@ -1155,7 +1089,6 @@
  */
 int wa_urb_dequeue(struct wahc *wa, struct urb *urb)
 {
-	struct device *dev = &wa->usb_iface->dev;
 	unsigned long flags, flags2;
 	struct wa_xfer *xfer;
 	struct wa_seg *seg;
@@ -1163,9 +1096,6 @@
 	unsigned cnt;
 	unsigned rpipe_ready = 0;
 
-	d_fnstart(3, dev, "(wa %p, urb %p)\n", wa, urb);
-
-	d_printf(1, dev, "xfer %p urb %p: aborting\n", urb->hcpriv, urb);
 	xfer = urb->hcpriv;
 	if (xfer == NULL) {
 		/* NOthing setup yet enqueue will see urb->status !=
@@ -1234,13 +1164,11 @@
 	wa_xfer_completion(xfer);
 	if (rpipe_ready)
 		wa_xfer_delayed_run(rpipe);
-	d_fnend(3, dev, "(wa %p, urb %p) = 0\n", wa, urb);
 	return 0;
 
 out_unlock:
 	spin_unlock_irqrestore(&xfer->lock, flags);
 out:
-	d_fnend(3, dev, "(wa %p, urb %p) = 0\n", wa, urb);
 	return 0;
 
 dequeue_delayed:
@@ -1250,7 +1178,6 @@
 	spin_unlock_irqrestore(&xfer->lock, flags);
 	wa_xfer_giveback(xfer);
 	usb_put_urb(urb);		/* we got a ref in enqueue() */
-	d_fnend(3, dev, "(wa %p, urb %p) = 0\n", wa, urb);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(wa_urb_dequeue);
@@ -1326,7 +1253,6 @@
 	u8 usb_status;
 	unsigned rpipe_ready = 0;
 
-	d_fnstart(3, dev, "(wa %p xfer %p)\n", wa, xfer);
 	spin_lock_irqsave(&xfer->lock, flags);
 	seg_idx = xfer_result->bTransferSegment & 0x7f;
 	if (unlikely(seg_idx >= xfer->segs))
@@ -1334,8 +1260,8 @@
 	seg = xfer->seg[seg_idx];
 	rpipe = xfer->ep->hcpriv;
 	usb_status = xfer_result->bTransferStatus;
-	d_printf(2, dev, "xfer %p#%u: bTransferStatus 0x%02x (seg %u)\n",
-		 xfer, seg_idx, usb_status, seg->status);
+	dev_dbg(dev, "xfer %p#%u: bTransferStatus 0x%02x (seg %u)\n",
+		xfer, seg_idx, usb_status, seg->status);
 	if (seg->status == WA_SEG_ABORTED
 	    || seg->status == WA_SEG_ERROR)	/* already handled */
 		goto segment_aborted;
@@ -1391,10 +1317,8 @@
 		wa_xfer_completion(xfer);
 	if (rpipe_ready)
 		wa_xfer_delayed_run(rpipe);
-	d_fnend(3, dev, "(wa %p xfer %p) = void\n", wa, xfer);
 	return;
 
-
 error_submit_buf_in:
 	if (edc_inc(&wa->dti_edc, EDC_MAX_ERRORS, EDC_ERROR_TIMEFRAME)) {
 		dev_err(dev, "DTI: URB max acceptable errors "
@@ -1416,11 +1340,8 @@
 		wa_xfer_completion(xfer);
 	if (rpipe_ready)
 		wa_xfer_delayed_run(rpipe);
-	d_fnend(3, dev, "(wa %p xfer %p) = void [segment/DTI-submit error]\n",
-		wa, xfer);
 	return;
 
-
 error_bad_seg:
 	spin_unlock_irqrestore(&xfer->lock, flags);
 	wa_urb_dequeue(wa, xfer->urb);
@@ -1431,17 +1352,11 @@
 			"exceeded, resetting device\n");
 		wa_reset_all(wa);
 	}
-	d_fnend(3, dev, "(wa %p xfer %p) = void [bad seg]\n", wa, xfer);
 	return;
 
-
 segment_aborted:
 	/* nothing to do, as the aborter did the completion */
 	spin_unlock_irqrestore(&xfer->lock, flags);
-	d_fnend(3, dev, "(wa %p xfer %p) = void [segment aborted]\n",
-		wa, xfer);
-	return;
-
 }
 
 /*
@@ -1465,15 +1380,14 @@
 	unsigned long flags;
 	u8 done = 0;
 
-	d_fnstart(3, NULL, "(urb %p [%d])\n", urb, urb->status);
 	switch (urb->status) {
 	case 0:
 		spin_lock_irqsave(&xfer->lock, flags);
 		wa = xfer->wa;
 		dev = &wa->usb_iface->dev;
 		rpipe = xfer->ep->hcpriv;
-		d_printf(2, dev, "xfer %p#%u: data in done (%zu bytes)\n",
-			   xfer, seg->index, (size_t)urb->actual_length);
+		dev_dbg(dev, "xfer %p#%u: data in done (%zu bytes)\n",
+			xfer, seg->index, (size_t)urb->actual_length);
 		seg->status = WA_SEG_DONE;
 		seg->result = urb->actual_length;
 		xfer->segs_done++;
@@ -1514,7 +1428,6 @@
 		if (rpipe_ready)
 			wa_xfer_delayed_run(rpipe);
 	}
-	d_fnend(3, NULL, "(urb %p [%d]) = void\n", urb, urb->status);
 }
 
 /*
@@ -1553,14 +1466,12 @@
 	struct wa_xfer *xfer;
 	u8 usb_status;
 
-	d_fnstart(3, dev, "(%p)\n", wa);
 	BUG_ON(wa->dti_urb != urb);
 	switch (wa->dti_urb->status) {
 	case 0:
 		/* We have a xfer result buffer; check it */
-		d_printf(2, dev, "DTI: xfer result %d bytes at %p\n",
-			   urb->actual_length, urb->transfer_buffer);
-		d_dump(3, dev, urb->transfer_buffer, urb->actual_length);
+		dev_dbg(dev, "DTI: xfer result %d bytes at %p\n",
+			urb->actual_length, urb->transfer_buffer);
 		if (wa->dti_urb->actual_length != sizeof(*xfer_result)) {
 			dev_err(dev, "DTI Error: xfer result--bad size "
 				"xfer result (%d bytes vs %zu needed)\n",
@@ -1622,7 +1533,6 @@
 		wa_reset_all(wa);
 	}
 out:
-	d_fnend(3, dev, "(%p) = void\n", wa);
 	return;
 }
 
@@ -1653,7 +1563,6 @@
 	struct wa_notif_xfer *notif_xfer;
 	const struct usb_endpoint_descriptor *dti_epd = wa->dti_epd;
 
-	d_fnstart(4, dev, "(%p, %p)\n", wa, notif_hdr);
 	notif_xfer = container_of(notif_hdr, struct wa_notif_xfer, hdr);
 	BUG_ON(notif_hdr->bNotifyType != WA_NOTIF_TRANSFER);
 
@@ -1693,7 +1602,6 @@
 		goto error_dti_urb_submit;
 	}
 out:
-	d_fnend(4, dev, "(%p, %p) = void\n", wa, notif_hdr);
 	return;
 
 error_dti_urb_submit:
@@ -1704,6 +1612,4 @@
 error_dti_urb_alloc:
 error:
 	wa_reset_all(wa);
-	d_fnend(4, dev, "(%p, %p) = void\n", wa, notif_hdr);
-	return;
 }
diff --git a/drivers/usb/wusbcore/wusbhc.h b/drivers/usb/wusbcore/wusbhc.h
index d0c1324..797c245 100644
--- a/drivers/usb/wusbcore/wusbhc.h
+++ b/drivers/usb/wusbcore/wusbhc.h
@@ -64,6 +64,13 @@
 #include <linux/uwb.h>
 #include <linux/usb/wusb.h>
 
+/*
+ * Time from a WUSB channel stop request to the last transmitted MMC.
+ *
+ * This needs to be > 4.096 ms in case no MMCs can be transmitted in
+ * zone 0.
+ */
+#define WUSB_CHANNEL_STOP_DELAY_MS 8
 
 /**
  * Wireless USB device
@@ -147,7 +154,6 @@
 	u16 status;
 	u16 change;
 	struct wusb_dev *wusb_dev;	/* connected device's info */
-	unsigned reset_count;
 	u32 ptk_tkid;
 };
 
@@ -198,21 +204,18 @@
  * @mmcies_max	   Max number of Information Elements this HC can send
  *                 in its MMC. Read-only.
  *
+ * @start          Start the WUSB channel.
+ *
+ * @stop           Stop the WUSB channel after the specified number of
+ *                 milliseconds.  Channel Stop IEs should be transmitted
+ *                 as required by [WUSB] 4.16.2.1.
+ *
  * @mmcie_add	   HC specific operation (WHCI or HWA) for adding an
  *                 MMCIE.
  *
  * @mmcie_rm	   HC specific operation (WHCI or HWA) for removing an
  *                 MMCIE.
  *
- * @enc_types	   Array which describes the encryptions methods
- *                 supported by the host as described in WUSB1.0 --
- *                 one entry per supported method. As of WUSB1.0 there
- *                 is only four methods, we make space for eight just in
- *                 case they decide to add some more (and pray they do
- *                 it in sequential order). if 'enc_types[enc_method]
- *                 != 0', then it is supported by the host. enc_method
- *                 is USB_ENC_TYPE*.
- *
  * @set_ptk:       Set the PTK and enable encryption for a device. Or, if
  *                 the supplied key is NULL, disable encryption for that
  *                 device.
@@ -249,7 +252,8 @@
 	struct uwb_pal pal;
 
 	unsigned trust_timeout;			/* in jiffies */
-	struct wuie_host_info *wuie_host_info;	/* Includes CHID */
+	struct wusb_ckhdid chid;
+	struct wuie_host_info *wuie_host_info;
 
 	struct mutex mutex;			/* locks everything else */
 	u16 cluster_id;				/* Wireless USB Cluster ID */
@@ -269,7 +273,7 @@
 	u8 mmcies_max;
 	/* FIXME: make wusbhc_ops? */
 	int (*start)(struct wusbhc *wusbhc);
-	void (*stop)(struct wusbhc *wusbhc);
+	void (*stop)(struct wusbhc *wusbhc, int delay);
 	int (*mmcie_add)(struct wusbhc *wusbhc, u8 interval, u8 repeat_cnt,
 			 u8 handle, struct wuie_hdr *wuie);
 	int (*mmcie_rm)(struct wusbhc *wusbhc, u8 handle);
@@ -373,20 +377,17 @@
 	usb_put_hcd(&wusbhc->usb_hcd);
 }
 
-int wusbhc_start(struct wusbhc *wusbhc, const struct wusb_ckhdid *chid);
+int wusbhc_start(struct wusbhc *wusbhc);
 void wusbhc_stop(struct wusbhc *wusbhc);
 extern int wusbhc_chid_set(struct wusbhc *, const struct wusb_ckhdid *);
 
 /* Device connect handling */
 extern int wusbhc_devconnect_create(struct wusbhc *);
 extern void wusbhc_devconnect_destroy(struct wusbhc *);
-extern int wusbhc_devconnect_start(struct wusbhc *wusbhc,
-				   const struct wusb_ckhdid *chid);
+extern int wusbhc_devconnect_start(struct wusbhc *wusbhc);
 extern void wusbhc_devconnect_stop(struct wusbhc *wusbhc);
-extern int wusbhc_devconnect_auth(struct wusbhc *, u8);
 extern void wusbhc_handle_dn(struct wusbhc *, u8 srcaddr,
 			     struct wusb_dn_hdr *dn_hdr, size_t size);
-extern int wusbhc_dev_reset(struct wusbhc *wusbhc, u8 port);
 extern void __wusbhc_dev_disable(struct wusbhc *wusbhc, u8 port);
 extern int wusb_usb_ncb(struct notifier_block *nb, unsigned long val,
 			void *priv);
@@ -432,6 +433,7 @@
 extern int wusb_dev_4way_handshake(struct wusbhc *, struct wusb_dev *,
 				   struct wusb_ckhdid *ck);
 void wusbhc_gtk_rekey(struct wusbhc *wusbhc);
+int wusb_dev_update_address(struct wusbhc *wusbhc, struct wusb_dev *wusb_dev);
 
 
 /* WUSB Cluster ID handling */
diff --git a/drivers/uwb/Makefile b/drivers/uwb/Makefile
index 257e690..2f98d08 100644
--- a/drivers/uwb/Makefile
+++ b/drivers/uwb/Makefile
@@ -6,6 +6,7 @@
 
 uwb-objs :=		\
 	address.o	\
+	allocator.o	\
 	beacon.o	\
 	driver.o	\
 	drp.o		\
@@ -13,10 +14,12 @@
 	drp-ie.o	\
 	est.o		\
 	ie.o		\
+	ie-rcv.o	\
 	lc-dev.o	\
 	lc-rc.o		\
 	neh.o		\
 	pal.o		\
+	radio.o		\
 	reset.o		\
 	rsv.o		\
 	scan.o		\
diff --git a/drivers/uwb/address.c b/drivers/uwb/address.c
index 1664ae5..ad21b1d7 100644
--- a/drivers/uwb/address.c
+++ b/drivers/uwb/address.c
@@ -28,7 +28,7 @@
 #include <linux/device.h>
 #include <linux/random.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
+
 #include "uwb-internal.h"
 
 
diff --git a/drivers/uwb/allocator.c b/drivers/uwb/allocator.c
new file mode 100644
index 0000000..c8185e6
--- /dev/null
+++ b/drivers/uwb/allocator.c
@@ -0,0 +1,386 @@
+/*
+ * UWB reservation management.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#include <linux/version.h>
+#include <linux/kernel.h>
+#include <linux/uwb.h>
+
+#include "uwb-internal.h"
+
+static void uwb_rsv_fill_column_alloc(struct uwb_rsv_alloc_info *ai)
+{
+	int col, mas, safe_mas, unsafe_mas;
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_col_info *ci = ai->ci;
+	unsigned char c;
+
+	for (col = ci->csi.start_col; col < UWB_NUM_ZONES; col += ci->csi.interval) {
+    
+		safe_mas   = ci->csi.safe_mas_per_col;
+		unsafe_mas = ci->csi.unsafe_mas_per_col;
+    
+		for (mas = 0; mas < UWB_MAS_PER_ZONE; mas++ ) {
+			if (bm[col * UWB_MAS_PER_ZONE + mas] == 0) {
+	
+				if (safe_mas > 0) {
+					safe_mas--;
+					c = UWB_RSV_MAS_SAFE;
+				} else if (unsafe_mas > 0) {
+					unsafe_mas--;
+					c = UWB_RSV_MAS_UNSAFE;
+				} else {
+					break;
+				}
+				bm[col * UWB_MAS_PER_ZONE + mas] = c;
+			}
+		}
+	}
+}
+
+static void uwb_rsv_fill_row_alloc(struct uwb_rsv_alloc_info *ai)
+{
+	int mas, col, rows;
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_row_info *ri = &ai->ri;
+	unsigned char c;
+
+	rows = 1;
+	c = UWB_RSV_MAS_SAFE;
+	for (mas = UWB_MAS_PER_ZONE - 1; mas >= 0; mas--) {
+		if (ri->avail[mas] == 1) {
+      
+			if (rows > ri->used_rows) {
+				break;
+			} else if (rows > 7) {
+				c = UWB_RSV_MAS_UNSAFE;
+			}
+
+			for (col = 0; col < UWB_NUM_ZONES; col++) {
+				if (bm[col * UWB_NUM_ZONES + mas] != UWB_RSV_MAS_NOT_AVAIL) {
+					bm[col * UWB_NUM_ZONES + mas] = c;
+					if(c == UWB_RSV_MAS_SAFE)
+						ai->safe_allocated_mases++;
+					else
+						ai->unsafe_allocated_mases++;
+				}
+			}
+			rows++;
+		}
+	}
+	ai->total_allocated_mases = ai->safe_allocated_mases + ai->unsafe_allocated_mases;
+}
+
+/*
+ * Find the best column set for a given availability, interval, num safe mas and
+ * num unsafe mas.
+ *
+ * The different sets are tried in order as shown below, depending on the interval.
+ *
+ * interval = 16
+ *	deep = 0
+ *		set 1 ->  {  8 }
+ *	deep = 1
+ *		set 1 ->  {  4 }
+ *		set 2 ->  { 12 }
+ *	deep = 2
+ *		set 1 ->  {  2 }
+ *		set 2 ->  {  6 }
+ *		set 3 ->  { 10 }
+ *		set 4 ->  { 14 }
+ *	deep = 3
+ *		set 1 ->  {  1 }
+ *		set 2 ->  {  3 }
+ *		set 3 ->  {  5 }
+ *		set 4 ->  {  7 }
+ *		set 5 ->  {  9 }
+ *		set 6 ->  { 11 }
+ *		set 7 ->  { 13 }
+ *		set 8 ->  { 15 }
+ *
+ * interval = 8
+ *	deep = 0
+ *		set 1 ->  {  4  12 }
+ *	deep = 1
+ *		set 1 ->  {  2  10 }
+ *		set 2 ->  {  6  14 }
+ *	deep = 2
+ *		set 1 ->  {  1   9 }
+ *		set 2 ->  {  3  11 }
+ *		set 3 ->  {  5  13 }
+ *		set 4 ->  {  7  15 }
+ *
+ * interval = 4
+ *	deep = 0
+ *		set 1 ->  {  2   6  10  14 }
+ *	deep = 1
+ *		set 1 ->  {  1   5   9  13 }
+ *		set 2 ->  {  3   7  11  15 }
+ *
+ * interval = 2
+ *	deep = 0
+ *		set 1 ->  {  1   3   5   7   9  11  13  15 }
+ */
+static int uwb_rsv_find_best_column_set(struct uwb_rsv_alloc_info *ai, int interval, 
+					int num_safe_mas, int num_unsafe_mas)
+{
+	struct uwb_rsv_col_info *ci = ai->ci;
+	struct uwb_rsv_col_set_info *csi = &ci->csi;
+	struct uwb_rsv_col_set_info tmp_csi;
+	int deep, set, col, start_col_deep, col_start_set;
+	int start_col, max_mas_in_set, lowest_max_mas_in_deep;
+	int n_mas;
+	int found = UWB_RSV_ALLOC_NOT_FOUND; 
+
+	tmp_csi.start_col = 0;
+	start_col_deep = interval;
+	n_mas = num_unsafe_mas + num_safe_mas;
+
+	for (deep = 0; ((interval >> deep) & 0x1) == 0; deep++) {
+		start_col_deep /= 2;
+		col_start_set = 0;
+		lowest_max_mas_in_deep = UWB_MAS_PER_ZONE;
+
+		for (set = 1; set <= (1 << deep); set++) {
+			max_mas_in_set = 0;
+			start_col = start_col_deep + col_start_set;
+			for (col = start_col; col < UWB_NUM_ZONES; col += interval) {
+                
+				if (ci[col].max_avail_safe >= num_safe_mas &&
+				    ci[col].max_avail_unsafe >= n_mas) {
+					if (ci[col].highest_mas[n_mas] > max_mas_in_set)
+						max_mas_in_set = ci[col].highest_mas[n_mas];
+				} else {
+					max_mas_in_set = 0;
+					break;
+				}
+			}
+			if ((lowest_max_mas_in_deep > max_mas_in_set) && max_mas_in_set) {
+				lowest_max_mas_in_deep = max_mas_in_set;
+
+				tmp_csi.start_col = start_col;
+			}
+			col_start_set += (interval >> deep);
+		}
+
+		if (lowest_max_mas_in_deep < 8) {
+			csi->start_col = tmp_csi.start_col;
+			found = UWB_RSV_ALLOC_FOUND;
+			break;
+		} else if ((lowest_max_mas_in_deep > 8) && 
+			   (lowest_max_mas_in_deep != UWB_MAS_PER_ZONE) &&
+			   (found == UWB_RSV_ALLOC_NOT_FOUND)) {
+			csi->start_col = tmp_csi.start_col;
+			found = UWB_RSV_ALLOC_FOUND;
+		}
+	}
+
+	if (found == UWB_RSV_ALLOC_FOUND) {
+		csi->interval = interval;
+		csi->safe_mas_per_col = num_safe_mas;
+		csi->unsafe_mas_per_col = num_unsafe_mas;
+
+		ai->safe_allocated_mases = (UWB_NUM_ZONES / interval) * num_safe_mas;
+		ai->unsafe_allocated_mases = (UWB_NUM_ZONES / interval) * num_unsafe_mas;
+		ai->total_allocated_mases = ai->safe_allocated_mases + ai->unsafe_allocated_mases;
+		ai->interval = interval;		
+	}
+	return found;
+}
+
+static void get_row_descriptors(struct uwb_rsv_alloc_info *ai)
+{
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_row_info *ri = &ai->ri;
+	int col, mas;
+  
+	ri->free_rows = 16;
+	for (mas = 0; mas < UWB_MAS_PER_ZONE; mas ++) {
+		ri->avail[mas] = 1;
+		for (col = 1; col < UWB_NUM_ZONES; col++) {
+			if (bm[col * UWB_NUM_ZONES + mas] == UWB_RSV_MAS_NOT_AVAIL) {
+				ri->free_rows--;
+				ri->avail[mas]=0;
+				break;
+			}
+		}
+	}
+}
+
+static void uwb_rsv_fill_column_info(unsigned char *bm, int column, struct uwb_rsv_col_info *rci)
+{
+	int mas;
+	int block_count = 0, start_block = 0; 
+	int previous_avail = 0;
+	int available = 0;
+	int safe_mas_in_row[UWB_MAS_PER_ZONE] = {
+		8, 7, 6, 5, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 2, 1,
+	};
+
+	rci->max_avail_safe = 0;
+
+	for (mas = 0; mas < UWB_MAS_PER_ZONE; mas ++) {
+		if (!bm[column * UWB_NUM_ZONES + mas]) {
+			available++;
+			rci->max_avail_unsafe = available;
+
+			rci->highest_mas[available] = mas;
+
+			if (previous_avail) {
+				block_count++;
+				if ((block_count > safe_mas_in_row[start_block]) &&
+				    (!rci->max_avail_safe))
+					rci->max_avail_safe = available - 1;
+			} else {
+				previous_avail = 1;
+				start_block = mas;
+				block_count = 1;
+			}
+		} else {
+			previous_avail = 0;
+		}
+	}
+	if (!rci->max_avail_safe)
+		rci->max_avail_safe = rci->max_avail_unsafe;
+}
+
+static void get_column_descriptors(struct uwb_rsv_alloc_info *ai)
+{
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_col_info *ci = ai->ci;
+	int col;
+
+	for (col = 1; col < UWB_NUM_ZONES; col++) {
+		uwb_rsv_fill_column_info(bm, col, &ci[col]);
+	}
+}
+
+static int uwb_rsv_find_best_row_alloc(struct uwb_rsv_alloc_info *ai)
+{
+	int n_rows;
+	int max_rows = ai->max_mas / UWB_USABLE_MAS_PER_ROW;
+	int min_rows = ai->min_mas / UWB_USABLE_MAS_PER_ROW;
+	if (ai->min_mas % UWB_USABLE_MAS_PER_ROW)
+		min_rows++;
+	for (n_rows = max_rows; n_rows >= min_rows; n_rows--) {
+		if (n_rows <= ai->ri.free_rows) {
+			ai->ri.used_rows = n_rows;
+			ai->interval = 1; /* row reservation */
+			uwb_rsv_fill_row_alloc(ai);
+			return UWB_RSV_ALLOC_FOUND;
+		}
+	}  
+	return UWB_RSV_ALLOC_NOT_FOUND;
+}
+
+static int uwb_rsv_find_best_col_alloc(struct uwb_rsv_alloc_info *ai, int interval)
+{
+	int n_safe, n_unsafe, n_mas;  
+	int n_column = UWB_NUM_ZONES / interval;
+	int max_per_zone = ai->max_mas / n_column;
+	int min_per_zone = ai->min_mas / n_column;
+
+	if (ai->min_mas % n_column)
+		min_per_zone++;
+
+	if (min_per_zone > UWB_MAS_PER_ZONE) {
+		return UWB_RSV_ALLOC_NOT_FOUND;
+	}
+    
+	if (max_per_zone > UWB_MAS_PER_ZONE) {
+		max_per_zone = UWB_MAS_PER_ZONE;
+	}
+    
+	for (n_mas = max_per_zone; n_mas >= min_per_zone; n_mas--) {
+		if (uwb_rsv_find_best_column_set(ai, interval, 0, n_mas) == UWB_RSV_ALLOC_NOT_FOUND)
+			continue;
+		for (n_safe = n_mas; n_safe >= 0; n_safe--) {
+			n_unsafe = n_mas - n_safe;
+			if (uwb_rsv_find_best_column_set(ai, interval, n_safe, n_unsafe) == UWB_RSV_ALLOC_FOUND) {
+				uwb_rsv_fill_column_alloc(ai);
+				return UWB_RSV_ALLOC_FOUND;
+			}
+		}
+	}
+	return UWB_RSV_ALLOC_NOT_FOUND;
+}
+
+int uwb_rsv_find_best_allocation(struct uwb_rsv *rsv, struct uwb_mas_bm *available, 
+				 struct uwb_mas_bm *result)
+{
+	struct uwb_rsv_alloc_info *ai;
+	int interval;
+	int bit_index;
+
+	ai = kzalloc(sizeof(struct uwb_rsv_alloc_info), GFP_KERNEL);
+	
+	ai->min_mas = rsv->min_mas;
+	ai->max_mas = rsv->max_mas;
+	ai->max_interval = rsv->max_interval;
+
+
+	/* fill the not available vector from the available bm */
+	for (bit_index = 0; bit_index < UWB_NUM_MAS; bit_index++) {
+		if (!test_bit(bit_index, available->bm))
+			ai->bm[bit_index] = UWB_RSV_MAS_NOT_AVAIL;
+	}
+
+	if (ai->max_interval == 1) {
+		get_row_descriptors(ai);
+		if (uwb_rsv_find_best_row_alloc(ai) == UWB_RSV_ALLOC_FOUND)
+			goto alloc_found;
+		else
+			goto alloc_not_found;
+	}
+
+	get_column_descriptors(ai);
+        
+	for (interval = 16; interval >= 2; interval>>=1) {
+		if (interval > ai->max_interval)
+			continue;
+		if (uwb_rsv_find_best_col_alloc(ai, interval) == UWB_RSV_ALLOC_FOUND)
+			goto alloc_found;
+	}
+
+	/* try row reservation if no column is found */
+	get_row_descriptors(ai);
+	if (uwb_rsv_find_best_row_alloc(ai) == UWB_RSV_ALLOC_FOUND)
+		goto alloc_found;
+	else
+		goto alloc_not_found;
+
+  alloc_found:
+	bitmap_zero(result->bm, UWB_NUM_MAS);
+	bitmap_zero(result->unsafe_bm, UWB_NUM_MAS);
+	/* fill the safe and unsafe bitmaps */
+	for (bit_index = 0; bit_index < UWB_NUM_MAS; bit_index++) {
+		if (ai->bm[bit_index] == UWB_RSV_MAS_SAFE)
+			set_bit(bit_index, result->bm);
+		else if (ai->bm[bit_index] == UWB_RSV_MAS_UNSAFE)
+			set_bit(bit_index, result->unsafe_bm);
+	}
+	bitmap_or(result->bm, result->bm, result->unsafe_bm, UWB_NUM_MAS);
+
+	result->safe   = ai->safe_allocated_mases;
+	result->unsafe = ai->unsafe_allocated_mases;
+	
+	kfree(ai);		
+	return UWB_RSV_ALLOC_FOUND;
+  
+  alloc_not_found:
+	kfree(ai);
+	return UWB_RSV_ALLOC_NOT_FOUND;
+}
diff --git a/drivers/uwb/beacon.c b/drivers/uwb/beacon.c
index 46b18ee..36bc315 100644
--- a/drivers/uwb/beacon.c
+++ b/drivers/uwb/beacon.c
@@ -22,19 +22,16 @@
  *
  * FIXME: docs
  */
-
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/device.h>
 #include <linux/err.h>
 #include <linux/kdev_t.h>
+
 #include "uwb-internal.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
-/** Start Beaconing command structure */
+/* Start Beaconing command structure */
 struct uwb_rc_cmd_start_beacon {
 	struct uwb_rccb rccb;
 	__le16 wBPSTOffset;
@@ -119,7 +116,6 @@
 	int result;
 	struct device *dev = &rc->uwb_dev.dev;
 
-	mutex_lock(&rc->uwb_dev.mutex);
 	if (channel < 0)
 		channel = -1;
 	if (channel == -1)
@@ -128,7 +124,7 @@
 		/* channel >= 0...dah */
 		result = uwb_rc_start_beacon(rc, bpst_offset, channel);
 		if (result < 0)
-			goto out_up;
+			return result;
 		if (le16_to_cpu(rc->ies->wIELength) > 0) {
 			result = uwb_rc_set_ie(rc, rc->ies);
 			if (result < 0) {
@@ -137,19 +133,12 @@
 				result = uwb_rc_stop_beacon(rc);
 				channel = -1;
 				bpst_offset = 0;
-			} else
-				result = 0;
+			}
 		}
 	}
 
-	if (result < 0)
-		goto out_up;
-	rc->beaconing = channel;
-
-	uwb_notify(rc, NULL, uwb_bg_joined(rc) ? UWB_NOTIF_BG_JOIN : UWB_NOTIF_BG_LEAVE);
-
-out_up:
-	mutex_unlock(&rc->uwb_dev.mutex);
+	if (result >= 0)
+		rc->beaconing = channel;
 	return result;
 }
 
@@ -168,12 +157,6 @@
  * FIXME: use something faster for search than a list
  */
 
-struct uwb_beca uwb_beca = {
-	.list = LIST_HEAD_INIT(uwb_beca.list),
-	.mutex = __MUTEX_INITIALIZER(uwb_beca.mutex)
-};
-
-
 void uwb_bce_kfree(struct kref *_bce)
 {
 	struct uwb_beca_e *bce = container_of(_bce, struct uwb_beca_e, refcnt);
@@ -185,13 +168,11 @@
 
 /* Find a beacon by dev addr in the cache */
 static
-struct uwb_beca_e *__uwb_beca_find_bydev(const struct uwb_dev_addr *dev_addr)
+struct uwb_beca_e *__uwb_beca_find_bydev(struct uwb_rc *rc,
+					 const struct uwb_dev_addr *dev_addr)
 {
 	struct uwb_beca_e *bce, *next;
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
-		d_printf(6, NULL, "looking for addr %02x:%02x in %02x:%02x\n",
-			 dev_addr->data[0], dev_addr->data[1],
-			 bce->dev_addr.data[0], bce->dev_addr.data[1]);
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		if (!memcmp(&bce->dev_addr, dev_addr, sizeof(bce->dev_addr)))
 			goto out;
 	}
@@ -202,10 +183,11 @@
 
 /* Find a beacon by dev addr in the cache */
 static
-struct uwb_beca_e *__uwb_beca_find_bymac(const struct uwb_mac_addr *mac_addr)
+struct uwb_beca_e *__uwb_beca_find_bymac(struct uwb_rc *rc, 
+					 const struct uwb_mac_addr *mac_addr)
 {
 	struct uwb_beca_e *bce, *next;
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		if (!memcmp(bce->mac_addr, mac_addr->data,
 			    sizeof(struct uwb_mac_addr)))
 			goto out;
@@ -229,11 +211,11 @@
 	struct uwb_dev *found = NULL;
 	struct uwb_beca_e *bce;
 
-	mutex_lock(&uwb_beca.mutex);
-	bce = __uwb_beca_find_bydev(devaddr);
+	mutex_lock(&rc->uwb_beca.mutex);
+	bce = __uwb_beca_find_bydev(rc, devaddr);
 	if (bce)
 		found = uwb_dev_try_get(rc, bce->uwb_dev);
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 
 	return found;
 }
@@ -249,11 +231,11 @@
 	struct uwb_dev *found = NULL;
 	struct uwb_beca_e *bce;
 
-	mutex_lock(&uwb_beca.mutex);
-	bce = __uwb_beca_find_bymac(macaddr);
+	mutex_lock(&rc->uwb_beca.mutex);
+	bce = __uwb_beca_find_bymac(rc, macaddr);
 	if (bce)
 		found = uwb_dev_try_get(rc, bce->uwb_dev);
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 
 	return found;
 }
@@ -274,7 +256,9 @@
  * @bf:         Beacon frame (part of b, really)
  * @ts_jiffies: Timestamp (in jiffies) when the beacon was received
  */
-struct uwb_beca_e *__uwb_beca_add(struct uwb_rc_evt_beacon *be,
+static
+struct uwb_beca_e *__uwb_beca_add(struct uwb_rc *rc,
+				  struct uwb_rc_evt_beacon *be,
 				  struct uwb_beacon_frame *bf,
 				  unsigned long ts_jiffies)
 {
@@ -286,7 +270,7 @@
 	uwb_beca_e_init(bce);
 	bce->ts_jiffies = ts_jiffies;
 	bce->uwb_dev = NULL;
-	list_add(&bce->node, &uwb_beca.list);
+	list_add(&bce->node, &rc->uwb_beca.list);
 	return bce;
 }
 
@@ -295,33 +279,32 @@
  *
  * Remove associated devicest too.
  */
-void uwb_beca_purge(void)
+void uwb_beca_purge(struct uwb_rc *rc)
 {
 	struct uwb_beca_e *bce, *next;
 	unsigned long expires;
 
-	mutex_lock(&uwb_beca.mutex);
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
+	mutex_lock(&rc->uwb_beca.mutex);
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		expires = bce->ts_jiffies + msecs_to_jiffies(beacon_timeout_ms);
 		if (time_after(jiffies, expires)) {
 			uwbd_dev_offair(bce);
-			list_del(&bce->node);
-			uwb_bce_put(bce);
 		}
 	}
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 }
 
 /* Clean up the whole beacon cache. Called on shutdown */
-void uwb_beca_release(void)
+void uwb_beca_release(struct uwb_rc *rc)
 {
 	struct uwb_beca_e *bce, *next;
-	mutex_lock(&uwb_beca.mutex);
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
+
+	mutex_lock(&rc->uwb_beca.mutex);
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		list_del(&bce->node);
 		uwb_bce_put(bce);
 	}
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 }
 
 static void uwb_beacon_print(struct uwb_rc *rc, struct uwb_rc_evt_beacon *be,
@@ -349,22 +332,22 @@
 	ssize_t result = 0;
 	struct uwb_rc_evt_beacon *be;
 	struct uwb_beacon_frame *bf;
-	struct uwb_buf_ctx ctx = {
-		.buf = buf,
-		.bytes = 0,
-		.size = size
-	};
+	int ies_len;
+	struct uwb_ie_hdr *ies;
 
 	mutex_lock(&bce->mutex);
+
 	be = bce->be;
-	if (be == NULL)
-		goto out;
-	bf = (void *) be->BeaconInfo;
-	uwb_ie_for_each(uwb_dev, uwb_ie_dump_hex, &ctx,
-			bf->IEData, be->wBeaconInfoLength - sizeof(*bf));
-	result = ctx.bytes;
-out:
+	if (be) {
+		bf = (struct uwb_beacon_frame *)bce->be->BeaconInfo;
+		ies_len = be->wBeaconInfoLength - sizeof(struct uwb_beacon_frame);
+		ies = (struct uwb_ie_hdr *)bf->IEData;
+
+		result = uwb_ie_dump_hex(ies, ies_len, buf, size);
+	}
+
 	mutex_unlock(&bce->mutex);
+
 	return result;
 }
 
@@ -437,18 +420,18 @@
 	if (uwb_mac_addr_bcast(&bf->Device_Identifier))
 		return 0;
 
-	mutex_lock(&uwb_beca.mutex);
-	bce = __uwb_beca_find_bymac(&bf->Device_Identifier);
+	mutex_lock(&rc->uwb_beca.mutex);
+	bce = __uwb_beca_find_bymac(rc, &bf->Device_Identifier);
 	if (bce == NULL) {
 		/* Not in there, a new device is pinging */
 		uwb_beacon_print(evt->rc, be, bf);
-		bce = __uwb_beca_add(be, bf, evt->ts_jiffies);
+		bce = __uwb_beca_add(rc, be, bf, evt->ts_jiffies);
 		if (bce == NULL) {
-			mutex_unlock(&uwb_beca.mutex);
+			mutex_unlock(&rc->uwb_beca.mutex);
 			return -ENOMEM;
 		}
 	}
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 
 	mutex_lock(&bce->mutex);
 	/* purge old beacon data */
@@ -588,19 +571,6 @@
 	return result;
 }
 
-/**
- * uwb_bg_joined - is the RC in a beacon group?
- * @rc: the radio controller
- *
- * Returns true if the radio controller is in a beacon group (even if
- * it's the sole member).
- */
-int uwb_bg_joined(struct uwb_rc *rc)
-{
-	return rc->beaconing != -1;
-}
-EXPORT_SYMBOL_GPL(uwb_bg_joined);
-
 /*
  * Print beaconing state.
  */
@@ -619,9 +589,6 @@
 
 /*
  * Start beaconing on the specified channel, or stop beaconing.
- *
- * The BPST offset of when to start searching for a beacon group to
- * join may be specified.
  */
 static ssize_t uwb_rc_beacon_store(struct device *dev,
 				   struct device_attribute *attr,
@@ -630,12 +597,11 @@
 	struct uwb_dev *uwb_dev = to_uwb_dev(dev);
 	struct uwb_rc *rc = uwb_dev->rc;
 	int channel;
-	unsigned bpst_offset = 0;
 	ssize_t result = -EINVAL;
 
-	result = sscanf(buf, "%d %u\n", &channel, &bpst_offset);
+	result = sscanf(buf, "%d", &channel);
 	if (result >= 1)
-		result = uwb_rc_beacon(rc, channel, bpst_offset);
+		result = uwb_radio_force_channel(rc, channel);
 
 	return result < 0 ? result : size;
 }
diff --git a/drivers/uwb/driver.c b/drivers/uwb/driver.c
index 521cdeb..da77e41 100644
--- a/drivers/uwb/driver.c
+++ b/drivers/uwb/driver.c
@@ -53,7 +53,7 @@
 #include <linux/err.h>
 #include <linux/kdev_t.h>
 #include <linux/random.h>
-#include <linux/uwb/debug.h>
+
 #include "uwb-internal.h"
 
 
@@ -118,7 +118,6 @@
 	result = class_register(&uwb_rc_class);
 	if (result < 0)
 		goto error_uwb_rc_class_register;
-	uwbd_start();
 	uwb_dbg_init();
 	return 0;
 
@@ -132,7 +131,6 @@
 static void __exit uwb_subsys_exit(void)
 {
 	uwb_dbg_exit();
-	uwbd_stop();
 	class_unregister(&uwb_rc_class);
 	uwb_est_destroy();
 	return;
diff --git a/drivers/uwb/drp-avail.c b/drivers/uwb/drp-avail.c
index 3febd855..40a540a 100644
--- a/drivers/uwb/drp-avail.c
+++ b/drivers/uwb/drp-avail.c
@@ -58,7 +58,7 @@
  *
  * avail = global & local & pending
  */
-static void uwb_drp_available(struct uwb_rc *rc, struct uwb_mas_bm *avail)
+void uwb_drp_available(struct uwb_rc *rc, struct uwb_mas_bm *avail)
 {
 	bitmap_and(avail->bm, rc->drp_avail.global, rc->drp_avail.local, UWB_NUM_MAS);
 	bitmap_and(avail->bm, avail->bm, rc->drp_avail.pending, UWB_NUM_MAS);
@@ -105,6 +105,7 @@
 	bitmap_or(rc->drp_avail.local, rc->drp_avail.local, mas->bm, UWB_NUM_MAS);
 	bitmap_or(rc->drp_avail.pending, rc->drp_avail.pending, mas->bm, UWB_NUM_MAS);
 	rc->drp_avail.ie_valid = false;
+	uwb_rsv_handle_drp_avail_change(rc);
 }
 
 /**
@@ -280,6 +281,7 @@
 	mutex_lock(&rc->rsvs_mutex);
 	bitmap_copy(rc->drp_avail.global, bmp, UWB_NUM_MAS);
 	rc->drp_avail.ie_valid = false;
+	uwb_rsv_handle_drp_avail_change(rc);
 	mutex_unlock(&rc->rsvs_mutex);
 
 	uwb_rsv_sched_update(rc);
diff --git a/drivers/uwb/drp-ie.c b/drivers/uwb/drp-ie.c
index 882724c..2840d7b 100644
--- a/drivers/uwb/drp-ie.c
+++ b/drivers/uwb/drp-ie.c
@@ -16,13 +16,102 @@
  * You should have received a copy of the GNU General Public License
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/random.h>
 #include <linux/uwb.h>
 
 #include "uwb-internal.h"
 
+
+/*
+ * Return the reason code for a reservations's DRP IE.
+ */
+int uwb_rsv_reason_code(struct uwb_rsv *rsv)
+{
+	static const int reason_codes[] = {
+		[UWB_RSV_STATE_O_INITIATED]          = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_PENDING]            = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_MODIFIED]           = UWB_DRP_REASON_MODIFIED,
+		[UWB_RSV_STATE_O_ESTABLISHED]        = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_TO_BE_MOVED]        = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_MOVE_COMBINING]     = UWB_DRP_REASON_MODIFIED,
+		[UWB_RSV_STATE_O_MOVE_REDUCING]      = UWB_DRP_REASON_MODIFIED,
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_ACCEPTED]           = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_CONFLICT]           = UWB_DRP_REASON_CONFLICT,
+		[UWB_RSV_STATE_T_PENDING]            = UWB_DRP_REASON_PENDING,
+		[UWB_RSV_STATE_T_DENIED]             = UWB_DRP_REASON_DENIED,
+		[UWB_RSV_STATE_T_RESIZED]            = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = UWB_DRP_REASON_CONFLICT,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = UWB_DRP_REASON_PENDING,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = UWB_DRP_REASON_DENIED,
+	};
+
+	return reason_codes[rsv->state];
+}
+
+/*
+ * Return the reason code for a reservations's companion DRP IE .
+ */
+int uwb_rsv_companion_reason_code(struct uwb_rsv *rsv)
+{
+	static const int companion_reason_codes[] = {
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = UWB_DRP_REASON_CONFLICT,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = UWB_DRP_REASON_PENDING,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = UWB_DRP_REASON_DENIED,
+	};
+
+	return companion_reason_codes[rsv->state];
+}
+
+/*
+ * Return the status bit for a reservations's DRP IE.
+ */
+int uwb_rsv_status(struct uwb_rsv *rsv)
+{
+	static const int statuses[] = {
+		[UWB_RSV_STATE_O_INITIATED]          = 0,
+		[UWB_RSV_STATE_O_PENDING]            = 0,
+		[UWB_RSV_STATE_O_MODIFIED]           = 1,
+		[UWB_RSV_STATE_O_ESTABLISHED]        = 1,
+		[UWB_RSV_STATE_O_TO_BE_MOVED]        = 0,
+		[UWB_RSV_STATE_O_MOVE_COMBINING]     = 1,
+		[UWB_RSV_STATE_O_MOVE_REDUCING]      = 1,
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = 1,
+		[UWB_RSV_STATE_T_ACCEPTED]           = 1,
+		[UWB_RSV_STATE_T_CONFLICT]           = 0,
+		[UWB_RSV_STATE_T_PENDING]            = 0,
+		[UWB_RSV_STATE_T_DENIED]             = 0,
+		[UWB_RSV_STATE_T_RESIZED]            = 1,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = 1,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = 1,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = 1,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = 1,
+
+	};
+
+	return statuses[rsv->state];
+}
+
+/*
+ * Return the status bit for a reservations's companion DRP IE .
+ */
+int uwb_rsv_companion_status(struct uwb_rsv *rsv)
+{
+	static const int companion_statuses[] = {
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = 0,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = 1,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = 0,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = 0,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = 0,
+	};
+
+	return companion_statuses[rsv->state];
+}
+
 /*
  * Allocate a DRP IE.
  *
@@ -34,16 +123,12 @@
 static struct uwb_ie_drp *uwb_drp_ie_alloc(void)
 {
 	struct uwb_ie_drp *drp_ie;
-	unsigned tiebreaker;
 
 	drp_ie = kzalloc(sizeof(struct uwb_ie_drp) +
 			UWB_NUM_ZONES * sizeof(struct uwb_drp_alloc),
 			GFP_KERNEL);
 	if (drp_ie) {
 		drp_ie->hdr.element_id = UWB_IE_DRP;
-
-		get_random_bytes(&tiebreaker, sizeof(unsigned));
-		uwb_ie_drp_set_tiebreaker(drp_ie, tiebreaker & 1);
 	}
 	return drp_ie;
 }
@@ -104,43 +189,17 @@
  */
 int uwb_drp_ie_update(struct uwb_rsv *rsv)
 {
-	struct device *dev = &rsv->rc->uwb_dev.dev;
 	struct uwb_ie_drp *drp_ie;
-	int reason_code, status;
+	struct uwb_rsv_move *mv;
+	int unsafe;
 
-	switch (rsv->state) {
-	case UWB_RSV_STATE_NONE:
+	if (rsv->state == UWB_RSV_STATE_NONE) {
 		kfree(rsv->drp_ie);
 		rsv->drp_ie = NULL;
 		return 0;
-	case UWB_RSV_STATE_O_INITIATED:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 0;
-		break;
-	case UWB_RSV_STATE_O_PENDING:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 0;
-		break;
-	case UWB_RSV_STATE_O_MODIFIED:
-		reason_code = UWB_DRP_REASON_MODIFIED;
-		status = 1;
-		break;
-	case UWB_RSV_STATE_O_ESTABLISHED:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 1;
-		break;
-	case UWB_RSV_STATE_T_ACCEPTED:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 1;
-		break;
-	case UWB_RSV_STATE_T_DENIED:
-		reason_code = UWB_DRP_REASON_DENIED;
-		status = 0;
-		break;
-	default:
-		dev_dbg(dev, "rsv with unhandled state (%d)\n", rsv->state);
-		return -EINVAL;
 	}
+	
+	unsafe = rsv->mas.unsafe ? 1 : 0;
 
 	if (rsv->drp_ie == NULL) {
 		rsv->drp_ie = uwb_drp_ie_alloc();
@@ -149,9 +208,11 @@
 	}
 	drp_ie = rsv->drp_ie;
 
+	uwb_ie_drp_set_unsafe(drp_ie,       unsafe);
+	uwb_ie_drp_set_tiebreaker(drp_ie,   rsv->tiebreaker);
 	uwb_ie_drp_set_owner(drp_ie,        uwb_rsv_is_owner(rsv));
-	uwb_ie_drp_set_status(drp_ie,       status);
-	uwb_ie_drp_set_reason_code(drp_ie,  reason_code);
+	uwb_ie_drp_set_status(drp_ie,       uwb_rsv_status(rsv));
+	uwb_ie_drp_set_reason_code(drp_ie,  uwb_rsv_reason_code(rsv));
 	uwb_ie_drp_set_stream_index(drp_ie, rsv->stream);
 	uwb_ie_drp_set_type(drp_ie,         rsv->type);
 
@@ -169,6 +230,27 @@
 
 	uwb_drp_ie_from_bm(drp_ie, &rsv->mas);
 
+	if (uwb_rsv_has_two_drp_ies(rsv)) {
+		mv = &rsv->mv; 
+		if (mv->companion_drp_ie == NULL) {
+			mv->companion_drp_ie = uwb_drp_ie_alloc();
+			if (mv->companion_drp_ie == NULL)
+				return -ENOMEM;
+		}
+		drp_ie = mv->companion_drp_ie;
+		
+		/* keep all the same configuration of the main drp_ie */
+		memcpy(drp_ie, rsv->drp_ie, sizeof(struct uwb_ie_drp));
+		
+
+		/* FIXME: handle properly the unsafe bit */
+		uwb_ie_drp_set_unsafe(drp_ie,       1);
+		uwb_ie_drp_set_status(drp_ie,       uwb_rsv_companion_status(rsv));
+		uwb_ie_drp_set_reason_code(drp_ie,  uwb_rsv_companion_reason_code(rsv));
+	
+		uwb_drp_ie_from_bm(drp_ie, &mv->companion_mas);
+	}
+
 	rsv->ie_valid = true;
 	return 0;
 }
@@ -219,6 +301,8 @@
 	u8 zone;
 	u16 zone_mask;
 
+	bitmap_zero(bm->bm, UWB_NUM_MAS);
+
 	for (cnt = 0; cnt < numallocs; cnt++) {
 		alloc = &drp_ie->allocs[cnt];
 		zone_bm = le16_to_cpu(alloc->zone_bm);
@@ -230,3 +314,4 @@
 		}
 	}
 }
+
diff --git a/drivers/uwb/drp.c b/drivers/uwb/drp.c
index c0b1e5e..2b4f940 100644
--- a/drivers/uwb/drp.c
+++ b/drivers/uwb/drp.c
@@ -23,6 +23,59 @@
 #include <linux/delay.h>
 #include "uwb-internal.h"
 
+
+/* DRP Conflict Actions ([ECMA-368 2nd Edition] 17.4.6) */
+enum uwb_drp_conflict_action {
+	/* Reservation is mantained, no action needed */
+	UWB_DRP_CONFLICT_MANTAIN = 0,
+	
+	/* the device shall not transmit frames in conflicting MASs in
+	 * the following superframe. If the device is the reservation
+	 * target, it shall also set the Reason Code in its DRP IE to
+	 * Conflict in its beacon in the following superframe.
+	 */
+	UWB_DRP_CONFLICT_ACT1,
+	
+	/* the device shall not set the Reservation Status bit to ONE
+	 * and shall not transmit frames in conflicting MASs. If the
+	 * device is the reservation target, it shall also set the
+	 * Reason Code in its DRP IE to Conflict.
+	 */	
+	UWB_DRP_CONFLICT_ACT2,
+
+	/* the device shall not transmit frames in conflicting MASs in
+	 * the following superframe. It shall remove the conflicting
+	 * MASs from the reservation or set the Reservation Status to
+	 * ZERO in its beacon in the following superframe. If the
+	 * device is the reservation target, it shall also set the
+	 * Reason Code in its DRP IE to Conflict.
+	 */
+	UWB_DRP_CONFLICT_ACT3,
+};
+
+
+static void uwb_rc_set_drp_cmd_done(struct uwb_rc *rc, void *arg,
+				    struct uwb_rceb *reply, ssize_t reply_size)
+{
+	struct uwb_rc_evt_set_drp_ie *r = (struct uwb_rc_evt_set_drp_ie *)reply;
+
+	if (r != NULL) {
+		if (r->bResultCode != UWB_RC_RES_SUCCESS)
+			dev_err(&rc->uwb_dev.dev, "SET-DRP-IE failed: %s (%d)\n",
+				uwb_rc_strerror(r->bResultCode), r->bResultCode);
+	} else
+		dev_err(&rc->uwb_dev.dev, "SET-DRP-IE: timeout\n");
+
+	spin_lock(&rc->rsvs_lock);
+	if (rc->set_drp_ie_pending > 1) {
+		rc->set_drp_ie_pending = 0;
+		uwb_rsv_queue_update(rc);	
+	} else {
+		rc->set_drp_ie_pending = 0;	
+	}
+	spin_unlock(&rc->rsvs_lock);
+}
+
 /**
  * Construct and send the SET DRP IE
  *
@@ -37,28 +90,32 @@
  *
  * A DRP Availability IE is appended.
  *
- * rc->uwb_dev.mutex is held
+ * rc->rsvs_mutex is held
  *
  * FIXME We currently ignore the returned value indicating the remaining space
  * in beacon. This could be used to deny reservation requests earlier if
  * determined that they would cause the beacon space to be exceeded.
  */
-static
-int uwb_rc_gen_send_drp_ie(struct uwb_rc *rc)
+int uwb_rc_send_all_drp_ie(struct uwb_rc *rc)
 {
 	int result;
-	struct device *dev = &rc->uwb_dev.dev;
 	struct uwb_rc_cmd_set_drp_ie *cmd;
-	struct uwb_rc_evt_set_drp_ie reply;
 	struct uwb_rsv *rsv;
+	struct uwb_rsv_move *mv;
 	int num_bytes = 0;
 	u8 *IEDataptr;
 
 	result = -ENOMEM;
 	/* First traverse all reservations to determine memory needed. */
 	list_for_each_entry(rsv, &rc->reservations, rc_node) {
-		if (rsv->drp_ie != NULL)
+		if (rsv->drp_ie != NULL) {
 			num_bytes += rsv->drp_ie->hdr.length + 2;
+			if (uwb_rsv_has_two_drp_ies(rsv) &&
+				(rsv->mv.companion_drp_ie != NULL)) {
+				mv = &rsv->mv;
+				num_bytes += mv->companion_drp_ie->hdr.length + 2;	
+			}
+		}
 	}
 	num_bytes += sizeof(rc->drp_avail.ie);
 	cmd = kzalloc(sizeof(*cmd) + num_bytes, GFP_KERNEL);
@@ -69,128 +126,322 @@
 	cmd->wIELength = num_bytes;
 	IEDataptr = (u8 *)&cmd->IEData[0];
 
+	/* FIXME: DRV avail IE is not always needed */
+	/* put DRP avail IE first */
+	memcpy(IEDataptr, &rc->drp_avail.ie, sizeof(rc->drp_avail.ie));
+	IEDataptr += sizeof(struct uwb_ie_drp_avail);
+
 	/* Next traverse all reservations to place IEs in allocated memory. */
 	list_for_each_entry(rsv, &rc->reservations, rc_node) {
 		if (rsv->drp_ie != NULL) {
 			memcpy(IEDataptr, rsv->drp_ie,
 			       rsv->drp_ie->hdr.length + 2);
 			IEDataptr += rsv->drp_ie->hdr.length + 2;
+			
+			if (uwb_rsv_has_two_drp_ies(rsv) &&
+				(rsv->mv.companion_drp_ie != NULL)) {
+				mv = &rsv->mv;
+				memcpy(IEDataptr, mv->companion_drp_ie,
+				       mv->companion_drp_ie->hdr.length + 2);
+				IEDataptr += mv->companion_drp_ie->hdr.length + 2;	
+			}
 		}
 	}
-	memcpy(IEDataptr, &rc->drp_avail.ie, sizeof(rc->drp_avail.ie));
 
-	reply.rceb.bEventType = UWB_RC_CET_GENERAL;
-	reply.rceb.wEvent = UWB_RC_CMD_SET_DRP_IE;
-	result = uwb_rc_cmd(rc, "SET-DRP-IE", &cmd->rccb,
-			sizeof(*cmd) + num_bytes, &reply.rceb,
-			sizeof(reply));
-	if (result < 0)
-		goto error_cmd;
-	result = le16_to_cpu(reply.wRemainingSpace);
-	if (reply.bResultCode != UWB_RC_RES_SUCCESS) {
-		dev_err(&rc->uwb_dev.dev, "SET-DRP-IE: command execution "
-				"failed: %s (%d). RemainingSpace in beacon "
-				"= %d\n", uwb_rc_strerror(reply.bResultCode),
-				reply.bResultCode, result);
-		result = -EIO;
-	} else {
-		dev_dbg(dev, "SET-DRP-IE sent. RemainingSpace in beacon "
-			     "= %d.\n", result);
-		result = 0;
-	}
-error_cmd:
+	result = uwb_rc_cmd_async(rc, "SET-DRP-IE", &cmd->rccb, sizeof(*cmd) + num_bytes,
+				  UWB_RC_CET_GENERAL, UWB_RC_CMD_SET_DRP_IE,
+				  uwb_rc_set_drp_cmd_done, NULL);
+	
+	rc->set_drp_ie_pending = 1;
+
 	kfree(cmd);
 error:
 	return result;
-
 }
-/**
- * Send all DRP IEs associated with this host
+
+/*
+ * Evaluate the action to perform using conflict resolution rules
  *
- * @returns:    >= 0 number of bytes still available in the beacon
- *              < 0 errno code on error.
- *
- * As per the protocol we obtain the host controller device lock to access
- * bandwidth structures.
+ * Return a uwb_drp_conflict_action.
  */
-int uwb_rc_send_all_drp_ie(struct uwb_rc *rc)
+static int evaluate_conflict_action(struct uwb_ie_drp *ext_drp_ie, int ext_beacon_slot,
+				    struct uwb_rsv *rsv, int our_status)
 {
-	int result;
+	int our_tie_breaker = rsv->tiebreaker;
+	int our_type        = rsv->type;
+	int our_beacon_slot = rsv->rc->uwb_dev.beacon_slot;
 
-	mutex_lock(&rc->uwb_dev.mutex);
-	result = uwb_rc_gen_send_drp_ie(rc);
-	mutex_unlock(&rc->uwb_dev.mutex);
-	return result;
-}
-
-void uwb_drp_handle_timeout(struct uwb_rsv *rsv)
-{
-	struct device *dev = &rsv->rc->uwb_dev.dev;
-
-	dev_dbg(dev, "reservation timeout in state %s (%d)\n",
-		uwb_rsv_state_str(rsv->state), rsv->state);
-
-	switch (rsv->state) {
-	case UWB_RSV_STATE_O_INITIATED:
-		if (rsv->is_multicast) {
-			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
-			return;
-		}
-		break;
-	case UWB_RSV_STATE_O_ESTABLISHED:
-		if (rsv->is_multicast)
-			return;
-		break;
-	default:
-		break;
+	int ext_tie_breaker = uwb_ie_drp_tiebreaker(ext_drp_ie);
+	int ext_status      = uwb_ie_drp_status(ext_drp_ie);
+	int ext_type        = uwb_ie_drp_type(ext_drp_ie);
+	
+	
+	/* [ECMA-368 2nd Edition] 17.4.6 */
+	if (ext_type == UWB_DRP_TYPE_PCA && our_type == UWB_DRP_TYPE_PCA) {
+		return UWB_DRP_CONFLICT_MANTAIN;
 	}
-	uwb_rsv_remove(rsv);
+
+	/* [ECMA-368 2nd Edition] 17.4.6-1 */
+	if (our_type == UWB_DRP_TYPE_ALIEN_BP) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+	
+	/* [ECMA-368 2nd Edition] 17.4.6-2 */
+	if (ext_type == UWB_DRP_TYPE_ALIEN_BP) {
+		/* here we know our_type != UWB_DRP_TYPE_ALIEN_BP */
+		return UWB_DRP_CONFLICT_ACT1;
+	}
+
+	/* [ECMA-368 2nd Edition] 17.4.6-3 */
+	if (our_status == 0 && ext_status == 1) {
+		return UWB_DRP_CONFLICT_ACT2;
+	}
+
+	/* [ECMA-368 2nd Edition] 17.4.6-4 */
+	if (our_status == 1 && ext_status == 0) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+
+	/* [ECMA-368 2nd Edition] 17.4.6-5a */
+	if (our_tie_breaker == ext_tie_breaker &&
+	    our_beacon_slot <  ext_beacon_slot) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+
+	/* [ECMA-368 2nd Edition] 17.4.6-5b */
+	if (our_tie_breaker != ext_tie_breaker &&
+	    our_beacon_slot >  ext_beacon_slot) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+	
+	if (our_status == 0) {
+		if (our_tie_breaker == ext_tie_breaker) {
+			/* [ECMA-368 2nd Edition] 17.4.6-6a */
+			if (our_beacon_slot > ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT2;
+			}
+		} else  {
+			/* [ECMA-368 2nd Edition] 17.4.6-6b */
+			if (our_beacon_slot < ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT2;
+			}
+		}
+	} else {
+		if (our_tie_breaker == ext_tie_breaker) {
+			/* [ECMA-368 2nd Edition] 17.4.6-7a */
+			if (our_beacon_slot > ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT3;
+			}
+		} else {
+			/* [ECMA-368 2nd Edition] 17.4.6-7b */
+			if (our_beacon_slot < ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT3;
+			}
+		}
+	}
+	return UWB_DRP_CONFLICT_MANTAIN;
 }
 
+static void handle_conflict_normal(struct uwb_ie_drp *drp_ie, 
+				   int ext_beacon_slot, 
+				   struct uwb_rsv *rsv, 
+				   struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rc *rc = rsv->rc;
+	struct uwb_rsv_move *mv = &rsv->mv;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	int action;
+
+	action = evaluate_conflict_action(drp_ie, ext_beacon_slot, rsv, uwb_rsv_status(rsv));
+
+	if (uwb_rsv_is_owner(rsv)) {
+		switch(action) {
+		case UWB_DRP_CONFLICT_ACT2:
+			/* try move */
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_TO_BE_MOVED);
+			if (bow->can_reserve_extra_mases == false)
+				uwb_rsv_backoff_win_increment(rc);
+			
+			break;
+		case UWB_DRP_CONFLICT_ACT3:
+			uwb_rsv_backoff_win_increment(rc);
+			/* drop some mases with reason modified */
+			/* put in the companion the mases to be dropped */
+			bitmap_and(mv->companion_mas.bm, rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS);
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MODIFIED);
+		default:
+			break;
+		}
+	} else {
+		switch(action) {
+		case UWB_DRP_CONFLICT_ACT2:
+		case UWB_DRP_CONFLICT_ACT3:
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_CONFLICT);	
+		default:
+			break;
+		}
+
+	}
+	
+}
+
+static void handle_conflict_expanding(struct uwb_ie_drp *drp_ie, int ext_beacon_slot,
+				      struct uwb_rsv *rsv, bool companion_only,
+				      struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rc *rc = rsv->rc;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct uwb_rsv_move *mv = &rsv->mv;
+	int action;
+	
+	if (companion_only) {
+		/* status of companion is 0 at this point */
+		action = evaluate_conflict_action(drp_ie, ext_beacon_slot, rsv, 0);
+		if (uwb_rsv_is_owner(rsv)) {
+			switch(action) {
+			case UWB_DRP_CONFLICT_ACT2:
+			case UWB_DRP_CONFLICT_ACT3:
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				rsv->needs_release_companion_mas = false;
+				if (bow->can_reserve_extra_mases == false)
+					uwb_rsv_backoff_win_increment(rc);
+				uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+			}
+		} else { /* rsv is target */			
+			switch(action) {
+			case UWB_DRP_CONFLICT_ACT2:
+			case UWB_DRP_CONFLICT_ACT3:
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_EXPANDING_CONFLICT);
+                                /* send_drp_avail_ie = true; */
+			}
+		}
+	} else { /* also base part of the reservation is conflicting */		
+		if (uwb_rsv_is_owner(rsv)) {
+			uwb_rsv_backoff_win_increment(rc);
+			/* remove companion part */
+			uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+
+			/* drop some mases with reason modified */
+
+			/* put in the companion the mases to be dropped */
+			bitmap_andnot(mv->companion_mas.bm, rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS);
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MODIFIED);
+		} else { /* it is a target rsv */
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_CONFLICT);
+                        /* send_drp_avail_ie = true; */
+		}
+	}
+}
+
+static void uwb_drp_handle_conflict_rsv(struct uwb_rc *rc, struct uwb_rsv *rsv,
+					struct uwb_rc_evt_drp *drp_evt, 
+					struct uwb_ie_drp *drp_ie,
+					struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rsv_move *mv;
+
+	/* check if the conflicting reservation has two drp_ies */
+	if (uwb_rsv_has_two_drp_ies(rsv)) {
+		mv = &rsv->mv;
+		if (bitmap_intersects(rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS)) {
+			handle_conflict_expanding(drp_ie, drp_evt->beacon_slot_number,
+						  rsv, false, conflicting_mas);
+		} else {
+			if (bitmap_intersects(mv->companion_mas.bm, conflicting_mas->bm, UWB_NUM_MAS)) {
+				handle_conflict_expanding(drp_ie, drp_evt->beacon_slot_number,
+							  rsv, true, conflicting_mas);	
+			}
+		}
+	} else if (bitmap_intersects(rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS)) {
+		handle_conflict_normal(drp_ie, drp_evt->beacon_slot_number, rsv, conflicting_mas);
+	}
+}
+
+static void uwb_drp_handle_all_conflict_rsv(struct uwb_rc *rc,
+					    struct uwb_rc_evt_drp *drp_evt, 
+					    struct uwb_ie_drp *drp_ie,
+					    struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rsv *rsv;
+	
+	list_for_each_entry(rsv, &rc->reservations, rc_node) {
+		uwb_drp_handle_conflict_rsv(rc, rsv, drp_evt, drp_ie, conflicting_mas);	
+	}
+}
+	
 /*
  * Based on the DRP IE, transition a target reservation to a new
  * state.
  */
 static void uwb_drp_process_target(struct uwb_rc *rc, struct uwb_rsv *rsv,
-				   struct uwb_ie_drp *drp_ie)
+				   struct uwb_ie_drp *drp_ie, struct uwb_rc_evt_drp *drp_evt)
 {
 	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_rsv_move *mv = &rsv->mv;
 	int status;
 	enum uwb_drp_reason reason_code;
-
+	struct uwb_mas_bm mas;
+	
 	status = uwb_ie_drp_status(drp_ie);
 	reason_code = uwb_ie_drp_reason_code(drp_ie);
+	uwb_drp_ie_to_bm(&mas, drp_ie);
 
-	if (status) {
-		switch (reason_code) {
-		case UWB_DRP_REASON_ACCEPTED:
+	switch (reason_code) {
+	case UWB_DRP_REASON_ACCEPTED:
+
+		if (rsv->state == UWB_RSV_STATE_T_CONFLICT) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_CONFLICT);
+			break;
+		}
+
+		if (rsv->state == UWB_RSV_STATE_T_EXPANDING_ACCEPTED) {
+			/* drp_ie is companion */
+			if (!bitmap_equal(rsv->mas.bm, mas.bm, UWB_NUM_MAS))
+				/* stroke companion */
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_EXPANDING_ACCEPTED);	
+		} else {
+			if (!bitmap_equal(rsv->mas.bm, mas.bm, UWB_NUM_MAS)) {
+				if (uwb_drp_avail_reserve_pending(rc, &mas) == -EBUSY) {
+					/* FIXME: there is a conflict, find
+					 * the conflicting reservations and
+					 * take a sensible action. Consider
+					 * that in drp_ie there is the
+					 * "neighbour" */
+					uwb_drp_handle_all_conflict_rsv(rc, drp_evt, drp_ie, &mas);
+				} else {
+					/* accept the extra reservation */
+					bitmap_copy(mv->companion_mas.bm, mas.bm, UWB_NUM_MAS);
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_EXPANDING_ACCEPTED);
+				}
+			} else {
+				if (status) {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_ACCEPTED);
+				}
+			}
+			
+		}
+		break;
+
+	case UWB_DRP_REASON_MODIFIED:
+		/* check to see if we have already modified the reservation */
+		if (bitmap_equal(rsv->mas.bm, mas.bm, UWB_NUM_MAS)) {
 			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_ACCEPTED);
 			break;
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
-			break;
-		default:
-			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
-				 reason_code, status);
 		}
-	} else {
-		switch (reason_code) {
-		case UWB_DRP_REASON_ACCEPTED:
-			/* New reservations are handled in uwb_rsv_find(). */
-			break;
-		case UWB_DRP_REASON_DENIED:
-			uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
-			break;
-		case UWB_DRP_REASON_CONFLICT:
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
-			break;
-		default:
-			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
-				 reason_code, status);
+
+		/* find if the owner wants to expand or reduce */
+		if (bitmap_subset(mas.bm, rsv->mas.bm, UWB_NUM_MAS)) {
+			/* owner is reducing */
+			bitmap_andnot(mv->companion_mas.bm, rsv->mas.bm, mas.bm, UWB_NUM_MAS);
+			uwb_drp_avail_release(rsv->rc, &mv->companion_mas);
 		}
+
+		bitmap_copy(rsv->mas.bm, mas.bm, UWB_NUM_MAS);
+		uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_RESIZED);
+		break;
+	default:
+		dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
+			 reason_code, status);
 	}
 }
 
@@ -199,23 +450,60 @@
  * state.
  */
 static void uwb_drp_process_owner(struct uwb_rc *rc, struct uwb_rsv *rsv,
-				  struct uwb_ie_drp *drp_ie)
+				  struct uwb_dev *src, struct uwb_ie_drp *drp_ie,
+				  struct uwb_rc_evt_drp *drp_evt)
 {
 	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_rsv_move *mv = &rsv->mv;
 	int status;
 	enum uwb_drp_reason reason_code;
+	struct uwb_mas_bm mas;
 
 	status = uwb_ie_drp_status(drp_ie);
 	reason_code = uwb_ie_drp_reason_code(drp_ie);
+	uwb_drp_ie_to_bm(&mas, drp_ie);
 
 	if (status) {
 		switch (reason_code) {
 		case UWB_DRP_REASON_ACCEPTED:
-			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
-			break;
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
+			switch (rsv->state) {
+			case UWB_RSV_STATE_O_PENDING:
+			case UWB_RSV_STATE_O_INITIATED:
+			case UWB_RSV_STATE_O_ESTABLISHED:
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				break;
+			case UWB_RSV_STATE_O_MODIFIED:
+				if (bitmap_equal(mas.bm, rsv->mas.bm, UWB_NUM_MAS)) {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				} else {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MODIFIED);	
+				}
+				break;
+				
+			case UWB_RSV_STATE_O_MOVE_REDUCING: /* shouldn' t be a problem */
+				if (bitmap_equal(mas.bm, rsv->mas.bm, UWB_NUM_MAS)) {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				} else {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);	
+				}
+				break;
+			case UWB_RSV_STATE_O_MOVE_EXPANDING:
+				if (bitmap_equal(mas.bm, mv->companion_mas.bm, UWB_NUM_MAS)) {
+					/* Companion reservation accepted */
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+				} else {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_EXPANDING);
+				}
+				break;
+			case UWB_RSV_STATE_O_MOVE_COMBINING:
+				if (bitmap_equal(mas.bm, rsv->mas.bm, UWB_NUM_MAS))
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);
+				else
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+				break;
+			default:
+				break;	
+			}
 			break;
 		default:
 			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
@@ -230,9 +518,10 @@
 			uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
 			break;
 		case UWB_DRP_REASON_CONFLICT:
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
+			/* resolve the conflict */
+			bitmap_complement(mas.bm, src->last_availability_bm,
+					  UWB_NUM_MAS);
+			uwb_drp_handle_conflict_rsv(rc, rsv, drp_evt, drp_ie, &mas);
 			break;
 		default:
 			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
@@ -241,12 +530,110 @@
 	}
 }
 
+static void uwb_cnflt_alien_stroke_timer(struct uwb_cnflt_alien *cnflt)
+{
+	unsigned timeout_us = UWB_MAX_LOST_BEACONS * UWB_SUPERFRAME_LENGTH_US;
+	mod_timer(&cnflt->timer, jiffies + usecs_to_jiffies(timeout_us));
+}
+
+static void uwb_cnflt_update_work(struct work_struct *work)
+{
+	struct uwb_cnflt_alien *cnflt = container_of(work,
+						     struct uwb_cnflt_alien,
+						     cnflt_update_work);
+	struct uwb_cnflt_alien *c;
+	struct uwb_rc *rc = cnflt->rc;
+	
+	unsigned long delay_us = UWB_MAS_LENGTH_US * UWB_MAS_PER_ZONE;
+	
+	mutex_lock(&rc->rsvs_mutex);
+
+	list_del(&cnflt->rc_node);
+
+	/* update rc global conflicting alien bitmap */
+	bitmap_zero(rc->cnflt_alien_bitmap.bm, UWB_NUM_MAS);
+
+	list_for_each_entry(c, &rc->cnflt_alien_list, rc_node) {
+		bitmap_or(rc->cnflt_alien_bitmap.bm, rc->cnflt_alien_bitmap.bm, c->mas.bm, UWB_NUM_MAS);			
+	}
+	
+	queue_delayed_work(rc->rsv_workq, &rc->rsv_alien_bp_work, usecs_to_jiffies(delay_us));
+
+	kfree(cnflt);
+	mutex_unlock(&rc->rsvs_mutex);
+}
+
+static void uwb_cnflt_timer(unsigned long arg)
+{
+	struct uwb_cnflt_alien *cnflt = (struct uwb_cnflt_alien *)arg;
+
+	queue_work(cnflt->rc->rsv_workq, &cnflt->cnflt_update_work);
+}
+
 /*
- * Process a received DRP IE, it's either for a reservation owned by
- * the RC or targeted at it (or it's for a WUSB cluster reservation).
+ * We have received an DRP_IE of type Alien BP and we need to make
+ * sure we do not transmit in conflicting MASs.
  */
-static void uwb_drp_process(struct uwb_rc *rc, struct uwb_dev *src,
-		     struct uwb_ie_drp *drp_ie)
+static void uwb_drp_handle_alien_drp(struct uwb_rc *rc, struct uwb_ie_drp *drp_ie)
+{
+	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_mas_bm mas;
+	struct uwb_cnflt_alien *cnflt;
+	char buf[72];
+	unsigned long delay_us = UWB_MAS_LENGTH_US * UWB_MAS_PER_ZONE;
+	
+	uwb_drp_ie_to_bm(&mas, drp_ie);
+	bitmap_scnprintf(buf, sizeof(buf), mas.bm, UWB_NUM_MAS);
+	
+	list_for_each_entry(cnflt, &rc->cnflt_alien_list, rc_node) {
+		if (bitmap_equal(cnflt->mas.bm, mas.bm, UWB_NUM_MAS)) {
+			/* Existing alien BP reservation conflicting
+			 * bitmap, just reset the timer */
+			uwb_cnflt_alien_stroke_timer(cnflt);
+			return;
+		}
+	}
+
+	/* New alien BP reservation conflicting bitmap */
+
+	/* alloc and initialize new uwb_cnflt_alien */
+	cnflt = kzalloc(sizeof(struct uwb_cnflt_alien), GFP_KERNEL);
+	if (!cnflt)
+		dev_err(dev, "failed to alloc uwb_cnflt_alien struct\n");
+	INIT_LIST_HEAD(&cnflt->rc_node);
+	init_timer(&cnflt->timer);
+	cnflt->timer.function = uwb_cnflt_timer;
+	cnflt->timer.data     = (unsigned long)cnflt;
+
+	cnflt->rc = rc;
+	INIT_WORK(&cnflt->cnflt_update_work, uwb_cnflt_update_work);
+	
+	bitmap_copy(cnflt->mas.bm, mas.bm, UWB_NUM_MAS);
+
+	list_add_tail(&cnflt->rc_node, &rc->cnflt_alien_list);
+
+	/* update rc global conflicting alien bitmap */
+	bitmap_or(rc->cnflt_alien_bitmap.bm, rc->cnflt_alien_bitmap.bm, mas.bm, UWB_NUM_MAS);
+
+	queue_delayed_work(rc->rsv_workq, &rc->rsv_alien_bp_work, usecs_to_jiffies(delay_us));
+	
+	/* start the timer */
+	uwb_cnflt_alien_stroke_timer(cnflt);
+}
+
+static void uwb_drp_process_not_involved(struct uwb_rc *rc,
+					 struct uwb_rc_evt_drp *drp_evt, 
+					 struct uwb_ie_drp *drp_ie)
+{
+	struct uwb_mas_bm mas;
+	
+	uwb_drp_ie_to_bm(&mas, drp_ie);
+	uwb_drp_handle_all_conflict_rsv(rc, drp_evt, drp_ie, &mas);
+}
+
+static void uwb_drp_process_involved(struct uwb_rc *rc, struct uwb_dev *src,
+				     struct uwb_rc_evt_drp *drp_evt,
+				     struct uwb_ie_drp *drp_ie)
 {
 	struct uwb_rsv *rsv;
 
@@ -259,7 +646,7 @@
 		 */
 		return;
 	}
-
+	
 	/*
 	 * Do nothing with DRP IEs for reservations that have been
 	 * terminated.
@@ -268,14 +655,44 @@
 		uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
 		return;
 	}
-
+			
 	if (uwb_ie_drp_owner(drp_ie))
-		uwb_drp_process_target(rc, rsv, drp_ie);
+		uwb_drp_process_target(rc, rsv, drp_ie, drp_evt);
 	else
-		uwb_drp_process_owner(rc, rsv, drp_ie);
+		uwb_drp_process_owner(rc, rsv, src, drp_ie, drp_evt);
+	
 }
 
 
+static bool uwb_drp_involves_us(struct uwb_rc *rc, struct uwb_ie_drp *drp_ie)
+{
+	return uwb_dev_addr_cmp(&rc->uwb_dev.dev_addr, &drp_ie->dev_addr) == 0;
+}
+
+/*
+ * Process a received DRP IE.
+ */
+static void uwb_drp_process(struct uwb_rc *rc, struct uwb_rc_evt_drp *drp_evt,
+			    struct uwb_dev *src, struct uwb_ie_drp *drp_ie)
+{
+	if (uwb_ie_drp_type(drp_ie) == UWB_DRP_TYPE_ALIEN_BP)
+		uwb_drp_handle_alien_drp(rc, drp_ie);
+	else if (uwb_drp_involves_us(rc, drp_ie))
+		uwb_drp_process_involved(rc, src, drp_evt, drp_ie);
+	else
+		uwb_drp_process_not_involved(rc, drp_evt, drp_ie);
+}
+
+/*
+ * Process a received DRP Availability IE
+ */
+static void uwb_drp_availability_process(struct uwb_rc *rc, struct uwb_dev *src,
+					 struct uwb_ie_drp_avail *drp_availability_ie)
+{
+	bitmap_copy(src->last_availability_bm,
+		    drp_availability_ie->bmp, UWB_NUM_MAS);
+}
+
 /*
  * Process all the DRP IEs (both DRP IEs and the DRP Availability IE)
  * from a device.
@@ -296,10 +713,10 @@
 
 		switch (ie_hdr->element_id) {
 		case UWB_IE_DRP_AVAILABILITY:
-			/* FIXME: does something need to be done with this? */
+			uwb_drp_availability_process(rc, src_dev, (struct uwb_ie_drp_avail *)ie_hdr);
 			break;
 		case UWB_IE_DRP:
-			uwb_drp_process(rc, src_dev, (struct uwb_ie_drp *)ie_hdr);
+			uwb_drp_process(rc, drp_evt, src_dev, (struct uwb_ie_drp *)ie_hdr);
 			break;
 		default:
 			dev_warn(dev, "unexpected IE in DRP notification\n");
@@ -312,55 +729,6 @@
 			 (int)ielen);
 }
 
-
-/*
- * Go through all the DRP IEs and find the ones that conflict with our
- * reservations.
- *
- * FIXME: must resolve the conflict according the the rules in
- * [ECMA-368].
- */
-static
-void uwb_drp_process_conflict_all(struct uwb_rc *rc, struct uwb_rc_evt_drp *drp_evt,
-				  size_t ielen, struct uwb_dev *src_dev)
-{
-	struct device *dev = &rc->uwb_dev.dev;
-	struct uwb_ie_hdr *ie_hdr;
-	struct uwb_ie_drp *drp_ie;
-	void *ptr;
-
-	ptr = drp_evt->ie_data;
-	for (;;) {
-		ie_hdr = uwb_ie_next(&ptr, &ielen);
-		if (!ie_hdr)
-			break;
-
-		drp_ie = container_of(ie_hdr, struct uwb_ie_drp, hdr);
-
-		/* FIXME: check if this DRP IE conflicts. */
-	}
-
-	if (ielen > 0)
-		dev_warn(dev, "%d octets remaining in DRP notification\n",
-			 (int)ielen);
-}
-
-
-/*
- * Terminate all reservations owned by, or targeted at, 'uwb_dev'.
- */
-static void uwb_drp_terminate_all(struct uwb_rc *rc, struct uwb_dev *uwb_dev)
-{
-	struct uwb_rsv *rsv;
-
-	list_for_each_entry(rsv, &rc->reservations, rc_node) {
-		if (rsv->owner == uwb_dev
-		    || (rsv->target.type == UWB_RSV_TARGET_DEV && rsv->target.dev == uwb_dev))
-			uwb_rsv_remove(rsv);
-	}
-}
-
-
 /**
  * uwbd_evt_handle_rc_drp - handle a DRP_IE event
  * @evt: the DRP_IE event from the radio controller
@@ -401,7 +769,6 @@
 	size_t ielength, bytes_left;
 	struct uwb_dev_addr src_addr;
 	struct uwb_dev *src_dev;
-	int reason;
 
 	/* Is there enough data to decode the event (and any IEs in
 	   its payload)? */
@@ -437,22 +804,8 @@
 
 	mutex_lock(&rc->rsvs_mutex);
 
-	reason = uwb_rc_evt_drp_reason(drp_evt);
-
-	switch (reason) {
-	case UWB_DRP_NOTIF_DRP_IE_RCVD:
-		uwb_drp_process_all(rc, drp_evt, ielength, src_dev);
-		break;
-	case UWB_DRP_NOTIF_CONFLICT:
-		uwb_drp_process_conflict_all(rc, drp_evt, ielength, src_dev);
-		break;
-	case UWB_DRP_NOTIF_TERMINATE:
-		uwb_drp_terminate_all(rc, src_dev);
-		break;
-	default:
-		dev_warn(dev, "ignored DRP event with reason code: %d\n", reason);
-		break;
-	}
+	/* We do not distinguish from the reason */
+	uwb_drp_process_all(rc, drp_evt, ielength, src_dev);
 
 	mutex_unlock(&rc->rsvs_mutex);
 
diff --git a/drivers/uwb/est.c b/drivers/uwb/est.c
index 5fe566b..328fcc2 100644
--- a/drivers/uwb/est.c
+++ b/drivers/uwb/est.c
@@ -40,10 +40,8 @@
  *   uwb_est_get_size()
  */
 #include <linux/spinlock.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-#include "uwb-internal.h"
 
+#include "uwb-internal.h"
 
 struct uwb_est {
 	u16 type_event_high;
@@ -52,7 +50,6 @@
 	const struct uwb_est_entry *entry;
 };
 
-
 static struct uwb_est *uwb_est;
 static u8 uwb_est_size;
 static u8 uwb_est_used;
@@ -440,21 +437,12 @@
 	u8 *ptr = (u8 *) rceb;
 
 	read_lock_irqsave(&uwb_est_lock, flags);
-	d_printf(2, dev, "Size query for event 0x%02x/%04x/%02x,"
-		 " buffer size %ld\n",
-		 (unsigned) rceb->bEventType,
-		 (unsigned) le16_to_cpu(rceb->wEvent),
-		 (unsigned) rceb->bEventContext,
-		 (long) rceb_size);
 	size = -ENOSPC;
 	if (rceb_size < sizeof(*rceb))
 		goto out;
 	event = le16_to_cpu(rceb->wEvent);
 	type_event_high = rceb->bEventType << 8 | (event & 0xff00) >> 8;
 	for (itr = 0; itr < uwb_est_used; itr++) {
-		d_printf(3, dev, "Checking EST 0x%04x/%04x/%04x\n",
-			uwb_est[itr].type_event_high, uwb_est[itr].vendor,
-			uwb_est[itr].product);
 		if (uwb_est[itr].type_event_high != type_event_high)
 			continue;
 		size = uwb_est_get_size(rc, &uwb_est[itr],
diff --git a/drivers/uwb/hwa-rc.c b/drivers/uwb/hwa-rc.c
index 3d26fa0..559f878 100644
--- a/drivers/uwb/hwa-rc.c
+++ b/drivers/uwb/hwa-rc.c
@@ -51,16 +51,14 @@
  *
  *
  */
-#include <linux/version.h>
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/usb.h>
 #include <linux/usb/wusb.h>
 #include <linux/usb/wusb-wa.h>
 #include <linux/uwb.h>
+
 #include "uwb-internal.h"
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
 
 /* The device uses commands and events from the WHCI specification, although
  * reporting itself as WUSB compliant. */
@@ -631,17 +629,13 @@
 
 	switch (result = urb->status) {
 	case 0:
-		d_printf(3, dev, "NEEP: receive stat %d, %zu bytes\n",
-			 urb->status, (size_t)urb->actual_length);
 		uwb_rc_neh_grok(hwarc->uwb_rc, urb->transfer_buffer,
 				urb->actual_length);
 		break;
 	case -ECONNRESET:	/* Not an error, but a controlled situation; */
 	case -ENOENT:		/* (we killed the URB)...so, no broadcast */
-		d_printf(2, dev, "NEEP: URB reset/noent %d\n", urb->status);
 		goto out;
 	case -ESHUTDOWN:	/* going away! */
-		d_printf(2, dev, "NEEP: URB down %d\n", urb->status);
 		goto out;
 	default:		/* On general errors, retry unless it gets ugly */
 		if (edc_inc(&hwarc->neep_edc, EDC_MAX_ERRORS,
@@ -650,7 +644,6 @@
 		dev_err(dev, "NEEP: URB error %d\n", urb->status);
 	}
 	result = usb_submit_urb(urb, GFP_ATOMIC);
-	d_printf(3, dev, "NEEP: submit %d\n", result);
 	if (result < 0) {
 		dev_err(dev, "NEEP: Can't resubmit URB (%d) resetting device\n",
 			result);
@@ -759,11 +752,11 @@
 	itr_size = le16_to_cpu(usb_dev->actconfig->desc.wTotalLength);
 	while (itr_size >= sizeof(*hdr)) {
 		hdr = (struct usb_descriptor_header *) itr;
-		d_printf(3, dev, "Extra device descriptor: "
-			 "type %02x/%u bytes @ %zu (%zu left)\n",
-			 hdr->bDescriptorType, hdr->bLength,
-			 (itr - usb_dev->rawdescriptors[actconfig_idx]),
-			 itr_size);
+		dev_dbg(dev, "Extra device descriptor: "
+			"type %02x/%u bytes @ %zu (%zu left)\n",
+			hdr->bDescriptorType, hdr->bLength,
+			(itr - usb_dev->rawdescriptors[actconfig_idx]),
+			itr_size);
 		if (hdr->bDescriptorType == USB_DT_CS_RADIO_CONTROL)
 			goto found;
 		itr += hdr->bLength;
@@ -795,8 +788,7 @@
 		goto error;
 	}
 	rc->version = version;
-	d_printf(3, dev, "Device supports WUSB protocol version 0x%04x \n",
-		 rc->version);
+	dev_dbg(dev, "Device supports WUSB protocol version 0x%04x \n",	rc->version);
 	result = 0;
 error:
 	return result;
@@ -877,11 +869,28 @@
 	uwb_rc_rm(uwb_rc);
 	usb_put_intf(hwarc->usb_iface);
 	usb_put_dev(hwarc->usb_dev);
-	d_printf(1, &hwarc->usb_iface->dev, "freed hwarc %p\n", hwarc);
 	kfree(hwarc);
 	uwb_rc_put(uwb_rc);	/* when creating the device, refcount = 1 */
 }
 
+static int hwarc_pre_reset(struct usb_interface *iface)
+{
+	struct hwarc *hwarc = usb_get_intfdata(iface);
+	struct uwb_rc *uwb_rc = hwarc->uwb_rc;
+
+	uwb_rc_pre_reset(uwb_rc);
+	return 0;
+}
+
+static int hwarc_post_reset(struct usb_interface *iface)
+{
+	struct hwarc *hwarc = usb_get_intfdata(iface);
+	struct uwb_rc *uwb_rc = hwarc->uwb_rc;
+
+	uwb_rc_post_reset(uwb_rc);
+	return 0;
+}
+
 /** USB device ID's that we handle */
 static struct usb_device_id hwarc_id_table[] = {
 	/* D-Link DUB-1210 */
@@ -898,20 +907,16 @@
 
 static struct usb_driver hwarc_driver = {
 	.name =		"hwa-rc",
+	.id_table =	hwarc_id_table,
 	.probe =	hwarc_probe,
 	.disconnect =	hwarc_disconnect,
-	.id_table =	hwarc_id_table,
+	.pre_reset =    hwarc_pre_reset,
+	.post_reset =   hwarc_post_reset,
 };
 
 static int __init hwarc_driver_init(void)
 {
-	int result;
-	result = usb_register(&hwarc_driver);
-	if (result < 0)
-		printk(KERN_ERR "HWA-RC: Cannot register USB driver: %d\n",
-		       result);
-	return result;
-
+	return usb_register(&hwarc_driver);
 }
 module_init(hwarc_driver_init);
 
diff --git a/drivers/uwb/i1480/dfu/dfu.c b/drivers/uwb/i1480/dfu/dfu.c
index 9097b3b..da7b1d0 100644
--- a/drivers/uwb/i1480/dfu/dfu.c
+++ b/drivers/uwb/i1480/dfu/dfu.c
@@ -34,10 +34,7 @@
 #include <linux/uwb.h>
 #include <linux/random.h>
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
-/**
+/*
  * i1480_rceb_check - Check RCEB for expected field values
  * @i1480: pointer to device for which RCEB is being checked
  * @rceb: RCEB being checked
@@ -83,7 +80,7 @@
 EXPORT_SYMBOL_GPL(i1480_rceb_check);
 
 
-/**
+/*
  * Execute a Radio Control Command
  *
  * Command data has to be in i1480->cmd_buf.
@@ -101,7 +98,6 @@
 	u8 expected_type = reply->bEventType;
 	u8 context;
 
-	d_fnstart(3, i1480->dev, "(%p, %s, %zu)\n", i1480, cmd_name, cmd_size);
 	init_completion(&i1480->evt_complete);
 	i1480->evt_result = -EINPROGRESS;
 	do {
@@ -150,8 +146,6 @@
 	result = i1480_rceb_check(i1480, i1480->evt_buf, cmd_name, context,
 				  expected_type, expected_event);
 error:
-	d_fnend(3, i1480->dev, "(%p, %s, %zu) = %zd\n",
-		i1480, cmd_name, cmd_size, result);
 	return result;
 }
 EXPORT_SYMBOL_GPL(i1480_cmd);
diff --git a/drivers/uwb/i1480/dfu/mac.c b/drivers/uwb/i1480/dfu/mac.c
index 2e4d8f0..694d0da 100644
--- a/drivers/uwb/i1480/dfu/mac.c
+++ b/drivers/uwb/i1480/dfu/mac.c
@@ -31,9 +31,6 @@
 #include <linux/uwb.h>
 #include "i1480-dfu.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
 /*
  * Descriptor for a continuous segment of MAC fw data
  */
@@ -184,10 +181,6 @@
 		}
 		if (memcmp(i1480->cmd_buf, bin + src_itr, result)) {
 			u8 *buf = i1480->cmd_buf;
-			d_printf(2, i1480->dev,
-				 "original data @ %p + %u, %zu bytes\n",
-				 bin, src_itr, result);
-			d_dump(4, i1480->dev, bin + src_itr, result);
 			for (cnt = 0; cnt < result; cnt++)
 				if (bin[src_itr + cnt] != buf[cnt]) {
 					dev_err(i1480->dev, "byte failed at "
@@ -224,7 +217,6 @@
 	struct fw_hdr *hdr_itr;
 	int verif_retry_count;
 
-	d_fnstart(3, dev, "(%p, %p)\n", i1480, hdr);
 	/* Now, header by header, push them to the hw */
 	for (hdr_itr = hdr; hdr_itr != NULL; hdr_itr = hdr_itr->next) {
 		verif_retry_count = 0;
@@ -264,7 +256,6 @@
 			break;
 		}
 	}
-	d_fnend(3, dev, "(%zd)\n", result);
 	return result;
 }
 
@@ -337,11 +328,9 @@
 	const struct firmware *fw;
 	struct fw_hdr *fw_hdrs;
 
-	d_fnstart(3, i1480->dev, "(%p, %s, %s)\n", i1480, fw_name, fw_tag);
 	result = request_firmware(&fw, fw_name, i1480->dev);
 	if (result < 0)	/* Up to caller to complain on -ENOENT */
 		goto out;
-	d_printf(3, i1480->dev, "%s fw '%s': uploading\n", fw_tag, fw_name);
 	result = fw_hdrs_load(i1480, &fw_hdrs, fw->data, fw->size);
 	if (result < 0) {
 		dev_err(i1480->dev, "%s fw '%s': failed to parse firmware "
@@ -363,8 +352,6 @@
 out_release:
 	release_firmware(fw);
 out:
-	d_fnend(3, i1480->dev, "(%p, %s, %s) = %d\n", i1480, fw_name, fw_tag,
-		result);
 	return result;
 }
 
@@ -433,7 +420,6 @@
 	int result;
 	u32 *val = (u32 *) i1480->cmd_buf;
 
-	d_fnstart(3, i1480->dev, "(i1480 %p)\n", i1480);
 	for (cnt = 0; cnt < 10; cnt++) {
 		msleep(100);
 		result = i1480->read(i1480, 0x80080000, 4);
@@ -447,7 +433,6 @@
 	dev_err(i1480->dev, "Timed out waiting for fw to start\n");
 	result = -ETIMEDOUT;
 out:
-	d_fnend(3, i1480->dev, "(i1480 %p) = %d\n", i1480, result);
 	return result;
 
 }
@@ -467,7 +452,6 @@
 	int result = 0, deprecated_name = 0;
 	struct i1480_rceb *rcebe = (void *) i1480->evt_buf;
 
-	d_fnstart(3, i1480->dev, "(%p)\n", i1480);
 	result = __mac_fw_upload(i1480, i1480->mac_fw_name, "MAC");
 	if (result == -ENOENT) {
 		result = __mac_fw_upload(i1480, i1480->mac_fw_name_deprecate,
@@ -501,7 +485,6 @@
 		dev_err(i1480->dev, "MAC fw '%s': initialization event returns "
 			"wrong size (%zu bytes vs %zu needed)\n",
 			i1480->mac_fw_name, i1480->evt_result, sizeof(*rcebe));
-		dump_bytes(i1480->dev, rcebe, min(i1480->evt_result, (ssize_t)32));
 		goto error_size;
 	}
 	result = -EIO;
@@ -522,6 +505,5 @@
 error_init_timeout:
 error_size:
 error_setup:
-	d_fnend(3, i1480->dev, "(i1480 %p) = %d\n", i1480, result);
 	return result;
 }
diff --git a/drivers/uwb/i1480/dfu/usb.c b/drivers/uwb/i1480/dfu/usb.c
index 98eeeff..686795e 100644
--- a/drivers/uwb/i1480/dfu/usb.c
+++ b/drivers/uwb/i1480/dfu/usb.c
@@ -35,7 +35,6 @@
  * the functions are i1480_usb_NAME().
  */
 #include <linux/module.h>
-#include <linux/version.h>
 #include <linux/usb.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
@@ -44,10 +43,6 @@
 #include <linux/usb/wusb-wa.h>
 #include "i1480-dfu.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
-
 struct i1480_usb {
 	struct i1480 i1480;
 	struct usb_device *usb_dev;
@@ -118,8 +113,6 @@
 	struct i1480_usb *i1480_usb = container_of(i1480, struct i1480_usb, i1480);
 	size_t buffer_size, itr = 0;
 
-	d_fnstart(3, i1480->dev, "(%p, 0x%08x, %p, %zu)\n",
-		  i1480, memory_address, buffer, size);
 	BUG_ON(size & 0x3); /* Needs to be a multiple of 4 */
 	while (size > 0) {
 		buffer_size = size < i1480->buf_size ? size : i1480->buf_size;
@@ -132,16 +125,10 @@
 			i1480->cmd_buf, buffer_size, 100 /* FIXME: arbitrary */);
 		if (result < 0)
 			break;
-		d_printf(3, i1480->dev,
-			 "wrote @ 0x%08x %u bytes (of %zu bytes requested)\n",
-			 memory_address, result, buffer_size);
-		d_dump(4, i1480->dev, i1480->cmd_buf, result);
 		itr += result;
 		memory_address += result;
 		size -= result;
 	}
-	d_fnend(3, i1480->dev, "(%p, 0x%08x, %p, %zu) = %d\n",
-		i1480, memory_address, buffer, size, result);
 	return result;
 }
 
@@ -166,8 +153,6 @@
 	size_t itr, read_size = i1480->buf_size;
 	struct i1480_usb *i1480_usb = container_of(i1480, struct i1480_usb, i1480);
 
-	d_fnstart(3, i1480->dev, "(%p, 0x%08x, %zu)\n",
-		  i1480, addr, size);
 	BUG_ON(size > i1480->buf_size);
 	BUG_ON(size & 0x3); /* Needs to be a multiple of 4 */
 	BUG_ON(read_size > 512);
@@ -201,10 +186,6 @@
 	}
 	result = bytes;
 out:
-	d_fnend(3, i1480->dev, "(%p, 0x%08x, %zu) = %zd\n",
-		i1480, addr, size, result);
-	if (result > 0)
-		d_dump(4, i1480->dev, i1480->cmd_buf, result);
 	return result;
 }
 
@@ -260,7 +241,6 @@
 	struct i1480_usb *i1480_usb = container_of(i1480, struct i1480_usb, i1480);
 	struct usb_endpoint_descriptor *epd;
 
-	d_fnstart(3, dev, "(%p)\n", i1480);
 	init_completion(&i1480->evt_complete);
 	i1480->evt_result = -EINPROGRESS;
 	epd = &i1480_usb->usb_iface->cur_altsetting->endpoint[0].desc;
@@ -282,14 +262,12 @@
 		goto error_wait;
 	}
 	usb_kill_urb(i1480_usb->neep_urb);
-	d_fnend(3, dev, "(%p) = 0\n", i1480);
 	return 0;
 
 error_wait:
 	usb_kill_urb(i1480_usb->neep_urb);
 error_submit:
 	i1480->evt_result = result;
-	d_fnend(3, dev, "(%p) = %d\n", i1480, result);
 	return result;
 }
 
@@ -320,7 +298,6 @@
 	struct uwb_rccb *cmd = i1480->cmd_buf;
 	u8 iface_no;
 
-	d_fnstart(3, dev, "(%p, %s, %zu)\n", i1480, cmd_name, cmd_size);
 	/* Post a read on the notification & event endpoint */
 	iface_no = i1480_usb->usb_iface->cur_altsetting->desc.bInterfaceNumber;
 	epd = &i1480_usb->usb_iface->cur_altsetting->endpoint[0].desc;
@@ -348,15 +325,11 @@
 			cmd_name, result);
 		goto error_submit_ep0;
 	}
-	d_fnend(3, dev, "(%p, %s, %zu) = %d\n",
-		i1480, cmd_name, cmd_size, result);
 	return result;
 
 error_submit_ep0:
 	usb_kill_urb(i1480_usb->neep_urb);
 error_submit_ep1:
-	d_fnend(3, dev, "(%p, %s, %zu) = %d\n",
-		i1480, cmd_name, cmd_size, result);
 	return result;
 }
 
diff --git a/drivers/uwb/i1480/i1480u-wlp/lc.c b/drivers/uwb/i1480/i1480u-wlp/lc.c
index 737d60c..049c05d 100644
--- a/drivers/uwb/i1480/i1480u-wlp/lc.c
+++ b/drivers/uwb/i1480/i1480u-wlp/lc.c
@@ -55,10 +55,9 @@
  *                          is being removed.
  *         i1480u_rm()
  */
-#include <linux/version.h>
 #include <linux/if_arp.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
+
 #include "i1480u-wlp.h"
 
 
@@ -207,7 +206,7 @@
 	wlp->fill_device_info = i1480u_fill_device_info;
 	wlp->stop_queue = i1480u_stop_queue;
 	wlp->start_queue = i1480u_start_queue;
-	result = wlp_setup(wlp, rc);
+	result = wlp_setup(wlp, rc, net_dev);
 	if (result < 0) {
 		dev_err(&iface->dev, "Cannot setup WLP\n");
 		goto error_wlp_setup;
diff --git a/drivers/uwb/i1480/i1480u-wlp/netdev.c b/drivers/uwb/i1480/i1480u-wlp/netdev.c
index 8802ac4..e3873ff 100644
--- a/drivers/uwb/i1480/i1480u-wlp/netdev.c
+++ b/drivers/uwb/i1480/i1480u-wlp/netdev.c
@@ -41,7 +41,7 @@
 
 #include <linux/if_arp.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
+
 #include "i1480u-wlp.h"
 
 struct i1480u_cmd_set_ip_mas {
@@ -207,6 +207,11 @@
 	result = i1480u_rx_setup(i1480u);		/* Alloc RX stuff */
 	if (result < 0)
 		goto error_rx_setup;
+
+	result = uwb_radio_start(&wlp->pal);
+	if (result < 0)
+		goto error_radio_start;
+
 	netif_wake_queue(net_dev);
 #ifdef i1480u_FLOW_CONTROL
 	result = usb_submit_urb(i1480u->notif_urb, GFP_KERNEL);;
@@ -215,25 +220,20 @@
 		goto error_notif_urb_submit;
 	}
 #endif
-	i1480u->uwb_notifs_handler.cb = i1480u_uwb_notifs_cb;
-	i1480u->uwb_notifs_handler.data = i1480u;
-	if (uwb_bg_joined(rc))
-		netif_carrier_on(net_dev);
-	else
-		netif_carrier_off(net_dev);
-	uwb_notifs_register(rc, &i1480u->uwb_notifs_handler);
 	/* Interface is up with an address, now we can create WSS */
 	result = wlp_wss_setup(net_dev, &wlp->wss);
 	if (result < 0) {
 		dev_err(dev, "Can't create WSS: %d. \n", result);
-		goto error_notif_deregister;
+		goto error_wss_setup;
 	}
 	return 0;
-error_notif_deregister:
-	uwb_notifs_deregister(rc, &i1480u->uwb_notifs_handler);
+error_wss_setup:
 #ifdef i1480u_FLOW_CONTROL
+	usb_kill_urb(i1480u->notif_urb);
 error_notif_urb_submit:
 #endif
+	uwb_radio_stop(&wlp->pal);
+error_radio_start:
 	netif_stop_queue(net_dev);
 	i1480u_rx_release(i1480u);
 error_rx_setup:
@@ -248,16 +248,15 @@
 {
 	struct i1480u *i1480u = netdev_priv(net_dev);
 	struct wlp *wlp = &i1480u->wlp;
-	struct uwb_rc *rc = wlp->rc;
 
 	BUG_ON(wlp->rc == NULL);
 	wlp_wss_remove(&wlp->wss);
-	uwb_notifs_deregister(rc, &i1480u->uwb_notifs_handler);
 	netif_carrier_off(net_dev);
 #ifdef i1480u_FLOW_CONTROL
 	usb_kill_urb(i1480u->notif_urb);
 #endif
 	netif_stop_queue(net_dev);
+	uwb_radio_stop(&wlp->pal);
 	i1480u_rx_release(i1480u);
 	i1480u_tx_release(i1480u);
 	return 0;
@@ -303,34 +302,6 @@
 	return 0;
 }
 
-
-/**
- * Callback function to handle events from UWB
- * When we see other devices we know the carrier is ok,
- * if we are the only device in the beacon group we set the carrier
- * state to off.
- * */
-void i1480u_uwb_notifs_cb(void *data, struct uwb_dev *uwb_dev,
-			  enum uwb_notifs event)
-{
-	struct i1480u *i1480u = data;
-	struct net_device *net_dev = i1480u->net_dev;
-	struct device *dev = &i1480u->usb_iface->dev;
-	switch (event) {
-	case UWB_NOTIF_BG_JOIN:
-		netif_carrier_on(net_dev);
-		dev_info(dev, "Link is up\n");
-		break;
-	case UWB_NOTIF_BG_LEAVE:
-		netif_carrier_off(net_dev);
-		dev_info(dev, "Link is down\n");
-		break;
-	default:
-		dev_err(dev, "don't know how to handle event %d from uwb\n",
-				event);
-	}
-}
-
 /**
  * Stop the network queue
  *
diff --git a/drivers/uwb/i1480/i1480u-wlp/rx.c b/drivers/uwb/i1480/i1480u-wlp/rx.c
index 9fc0353..34f4cf9 100644
--- a/drivers/uwb/i1480/i1480u-wlp/rx.c
+++ b/drivers/uwb/i1480/i1480u-wlp/rx.c
@@ -68,11 +68,7 @@
 #include <linux/etherdevice.h>
 #include "i1480u-wlp.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
-
-
-/**
+/*
  * Setup the RX context
  *
  * Each URB is provided with a transfer_buffer that is the data field
@@ -129,7 +125,7 @@
 }
 
 
-/** Release resources associated to the rx context */
+/* Release resources associated to the rx context */
 void i1480u_rx_release(struct i1480u *i1480u)
 {
 	int cnt;
@@ -155,7 +151,7 @@
 	}
 }
 
-/** Fix an out-of-sequence packet */
+/* Fix an out-of-sequence packet */
 #define i1480u_fix(i1480u, msg...)			\
 do {							\
 	if (printk_ratelimit())				\
@@ -166,7 +162,7 @@
 } while (0)
 
 
-/** Drop an out-of-sequence packet */
+/* Drop an out-of-sequence packet */
 #define i1480u_drop(i1480u, msg...)			\
 do {							\
 	if (printk_ratelimit())				\
@@ -177,7 +173,7 @@
 
 
 
-/** Finalizes setting up the SKB and delivers it
+/* Finalizes setting up the SKB and delivers it
  *
  * We first pass the incoming frame to WLP substack for verification. It
  * may also be a WLP association frame in which case WLP will take over the
@@ -192,18 +188,11 @@
 	struct net_device *net_dev = i1480u->net_dev;
 	struct device *dev = &i1480u->usb_iface->dev;
 
-	d_printf(6, dev, "RX delivered pre skb(%p), %u bytes\n",
-		 i1480u->rx_skb, i1480u->rx_skb->len);
-	d_dump(7, dev, i1480u->rx_skb->data, i1480u->rx_skb->len);
 	should_parse = wlp_receive_frame(dev, &i1480u->wlp, i1480u->rx_skb,
 					 &i1480u->rx_srcaddr);
 	if (!should_parse)
 		goto out;
 	i1480u->rx_skb->protocol = eth_type_trans(i1480u->rx_skb, net_dev);
-	d_printf(5, dev, "RX delivered skb(%p), %u bytes\n",
-		 i1480u->rx_skb, i1480u->rx_skb->len);
-	d_dump(7, dev, i1480u->rx_skb->data,
-	       i1480u->rx_skb->len > 72 ? 72 : i1480u->rx_skb->len);
 	i1480u->stats.rx_packets++;
 	i1480u->stats.rx_bytes += i1480u->rx_untd_pkt_size;
 	net_dev->last_rx = jiffies;
@@ -216,7 +205,7 @@
 }
 
 
-/**
+/*
  * Process a buffer of data received from the USB RX endpoint
  *
  * First fragment arrives with next or last fragment. All other fragments
@@ -404,7 +393,7 @@
 }
 
 
-/**
+/*
  * Called when an RX URB has finished receiving or has found some kind
  * of error condition.
  *
diff --git a/drivers/uwb/i1480/i1480u-wlp/sysfs.c b/drivers/uwb/i1480/i1480u-wlp/sysfs.c
index a1d8ca6..4ffaf54 100644
--- a/drivers/uwb/i1480/i1480u-wlp/sysfs.c
+++ b/drivers/uwb/i1480/i1480u-wlp/sysfs.c
@@ -25,8 +25,8 @@
 
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
 #include <linux/device.h>
+
 #include "i1480u-wlp.h"
 
 
@@ -226,7 +226,6 @@
  * (CLASS_DEVICE_ATTR or DEVICE_ATTR) and i1480u_ATTR_NAME produces a
  * class_device_attr_NAME or device_attr_NAME (for group registration).
  */
-#include <linux/version.h>
 
 #define i1480u_SHOW(name, fn, param)				\
 static ssize_t i1480u_show_##name(struct device *dev,		\
diff --git a/drivers/uwb/i1480/i1480u-wlp/tx.c b/drivers/uwb/i1480/i1480u-wlp/tx.c
index 3426bfb..39032cc 100644
--- a/drivers/uwb/i1480/i1480u-wlp/tx.c
+++ b/drivers/uwb/i1480/i1480u-wlp/tx.c
@@ -55,8 +55,6 @@
  */
 
 #include "i1480u-wlp.h"
-#define D_LOCAL 5
-#include <linux/uwb/debug.h>
 
 enum {
 	/* This is only for Next and Last TX packets */
@@ -64,7 +62,7 @@
 		- sizeof(struct untd_hdr_rst),
 };
 
-/** Free resources allocated to a i1480u tx context. */
+/* Free resources allocated to a i1480u tx context. */
 static
 void i1480u_tx_free(struct i1480u_tx *wtx)
 {
@@ -99,7 +97,7 @@
 }
 
 
-/**
+/*
  * Callback for a completed tx USB URB.
  *
  * TODO:
@@ -149,8 +147,6 @@
 	    <= i1480u->tx_inflight.threshold
 	    && netif_queue_stopped(net_dev)
 	    && i1480u->tx_inflight.threshold != 0) {
-		if (d_test(2) && printk_ratelimit())
-			d_printf(2, dev, "Restart queue. \n");
 		netif_start_queue(net_dev);
 		atomic_inc(&i1480u->tx_inflight.restart_count);
 	}
@@ -158,7 +154,7 @@
 }
 
 
-/**
+/*
  * Given a buffer that doesn't fit in a single fragment, create an
  * scatter/gather structure for delivery to the USB pipe.
  *
@@ -253,15 +249,11 @@
 	/* Now do each remaining fragment */
 	result = -EINVAL;
 	while (pl_size_left > 0) {
-		d_printf(5, NULL, "ITR HDR: pl_size_left %zu buf_itr %zu\n",
-			 pl_size_left, buf_itr - wtx->buf);
 		if (buf_itr + sizeof(*untd_hdr_rst) - wtx->buf
 		    > wtx->buf_size) {
 			printk(KERN_ERR "BUG: no space for header\n");
 			goto error_bug;
 		}
-		d_printf(5, NULL, "ITR HDR 2: pl_size_left %zu buf_itr %zu\n",
-			 pl_size_left, buf_itr - wtx->buf);
 		untd_hdr_rst = buf_itr;
 		buf_itr += sizeof(*untd_hdr_rst);
 		if (pl_size_left > i1480u_MAX_PL_SIZE) {
@@ -271,9 +263,6 @@
 			frg_pl_size = pl_size_left;
 			untd_hdr_set_type(&untd_hdr_rst->hdr, i1480u_PKT_FRAG_LST);
 		}
-		d_printf(5, NULL,
-			 "ITR PL: pl_size_left %zu buf_itr %zu frg_pl_size %zu\n",
-			 pl_size_left, buf_itr - wtx->buf, frg_pl_size);
 		untd_hdr_set_rx_tx(&untd_hdr_rst->hdr, 0);
 		untd_hdr_rst->hdr.len = cpu_to_le16(frg_pl_size);
 		untd_hdr_rst->padding = 0;
@@ -286,9 +275,6 @@
 		buf_itr += frg_pl_size;
 		pl_itr += frg_pl_size;
 		pl_size_left -= frg_pl_size;
-		d_printf(5, NULL,
-			 "ITR PL 2: pl_size_left %zu buf_itr %zu frg_pl_size %zu\n",
-			 pl_size_left, buf_itr - wtx->buf, frg_pl_size);
 	}
 	dev_kfree_skb_irq(skb);
 	return 0;
@@ -308,7 +294,7 @@
 }
 
 
-/**
+/*
  * Given a buffer that fits in a single fragment, fill out a @wtx
  * struct for transmitting it down the USB pipe.
  *
@@ -346,7 +332,7 @@
 }
 
 
-/**
+/*
  * Given a skb to transmit, massage it to become palatable for the TX pipe
  *
  * This will break the buffer in chunks smaller than
@@ -425,7 +411,7 @@
 	return NULL;
 }
 
-/**
+/*
  * Actual fragmentation and transmission of frame
  *
  * @wlp:  WLP substack data structure
@@ -447,20 +433,12 @@
 	struct i1480u_tx *wtx;
 	struct wlp_tx_hdr *wlp_tx_hdr;
 	static unsigned char dev_bcast[2] = { 0xff, 0xff };
-#if 0
-	int lockup = 50;
-#endif
 
-	d_fnstart(6, dev, "(skb %p (%u), net_dev %p)\n", skb, skb->len,
-		  net_dev);
 	BUG_ON(i1480u->wlp.rc == NULL);
 	if ((net_dev->flags & IFF_UP) == 0)
 		goto out;
 	result = -EBUSY;
 	if (atomic_read(&i1480u->tx_inflight.count) >= i1480u->tx_inflight.max) {
-		if (d_test(2) && printk_ratelimit())
-			d_printf(2, dev, "Max frames in flight "
-				 "stopping queue.\n");
 		netif_stop_queue(net_dev);
 		goto error_max_inflight;
 	}
@@ -489,21 +467,6 @@
 		wlp_tx_hdr_set_delivery_id_type(wlp_tx_hdr, i1480u->options.pca_base_priority);
 	}
 
-#if 0
-	dev_info(dev, "TX delivering skb -> USB, %zu bytes\n", skb->len);
-	dump_bytes(dev, skb->data, skb->len > 72 ? 72 : skb->len);
-#endif
-#if 0
-	/* simulates a device lockup after every lockup# packets */
-	if (lockup && ((i1480u->stats.tx_packets + 1) % lockup) == 0) {
-		/* Simulate a dropped transmit interrupt */
-		net_dev->trans_start = jiffies;
-		netif_stop_queue(net_dev);
-		dev_err(dev, "Simulate lockup at %ld\n", jiffies);
-		return result;
-	}
-#endif
-
 	result = usb_submit_urb(wtx->urb, GFP_ATOMIC);		/* Go baby */
 	if (result < 0) {
 		dev_err(dev, "TX: cannot submit URB: %d\n", result);
@@ -513,8 +476,6 @@
 	}
 	atomic_inc(&i1480u->tx_inflight.count);
 	net_dev->trans_start = jiffies;
-	d_fnend(6, dev, "(skb %p (%u), net_dev %p) = %d\n", skb, skb->len,
-		net_dev, result);
 	return result;
 
 error_tx_urb_submit:
@@ -522,13 +483,11 @@
 error_wtx_alloc:
 error_max_inflight:
 out:
-	d_fnend(6, dev, "(skb %p (%u), net_dev %p) = %d\n", skb, skb->len,
-		net_dev, result);
 	return result;
 }
 
 
-/**
+/*
  * Transmit an skb  Called when an skbuf has to be transmitted
  *
  * The skb is first passed to WLP substack to ensure this is a valid
@@ -551,9 +510,6 @@
 	struct device *dev = &i1480u->usb_iface->dev;
 	struct uwb_dev_addr dst;
 
-	d_fnstart(6, dev, "(skb %p (%u), net_dev %p)\n", skb, skb->len,
-		  net_dev);
-	BUG_ON(i1480u->wlp.rc == NULL);
 	if ((net_dev->flags & IFF_UP) == 0)
 		goto error;
 	result = wlp_prepare_tx_frame(dev, &i1480u->wlp, skb, &dst);
@@ -562,31 +518,25 @@
 			"Dropping packet.\n", result);
 		goto error;
 	} else if (result == 1) {
-		d_printf(6, dev, "WLP will transmit frame. \n");
 		/* trans_start time will be set when WLP actually transmits
 		 * the frame */
 		goto out;
 	}
-	d_printf(6, dev, "Transmitting frame. \n");
 	result = i1480u_xmit_frame(&i1480u->wlp, skb, &dst);
 	if (result < 0) {
 		dev_err(dev, "Frame TX failed (%d).\n", result);
 		goto error;
 	}
-	d_fnend(6, dev, "(skb %p (%u), net_dev %p) = %d\n", skb, skb->len,
-		net_dev, result);
 	return NETDEV_TX_OK;
 error:
 	dev_kfree_skb_any(skb);
 	i1480u->stats.tx_dropped++;
 out:
-	d_fnend(6, dev, "(skb %p (%u), net_dev %p) = %d\n", skb, skb->len,
-		net_dev, result);
 	return NETDEV_TX_OK;
 }
 
 
-/**
+/*
  * Called when a pkt transmission doesn't complete in a reasonable period
  * Device reset may sleep - do it outside of interrupt context (delayed)
  */
diff --git a/drivers/uwb/ie-rcv.c b/drivers/uwb/ie-rcv.c
new file mode 100644
index 0000000..917e6d7
--- /dev/null
+++ b/drivers/uwb/ie-rcv.c
@@ -0,0 +1,55 @@
+/*
+ * Ultra Wide Band
+ * IE Received notification handling.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/bitmap.h>
+#include "uwb-internal.h"
+
+/*
+ * Process an incoming IE Received notification.
+ */
+int uwbd_evt_handle_rc_ie_rcv(struct uwb_event *evt)
+{
+	int result = -EINVAL;
+	struct device *dev = &evt->rc->uwb_dev.dev;
+	struct uwb_rc_evt_ie_rcv *iercv;
+	size_t iesize;
+
+	/* Is there enough data to decode it? */
+	if (evt->notif.size < sizeof(*iercv)) {
+		dev_err(dev, "IE Received notification: Not enough data to "
+			"decode (%zu vs %zu bytes needed)\n",
+			evt->notif.size, sizeof(*iercv));
+		goto error;
+	}
+	iercv = container_of(evt->notif.rceb, struct uwb_rc_evt_ie_rcv, rceb);
+	iesize = le16_to_cpu(iercv->wIELength);
+
+	dev_dbg(dev, "IE received, element ID=%d\n", iercv->IEData[0]);
+
+	if (iercv->IEData[0] == UWB_RELINQUISH_REQUEST_IE) {
+		dev_warn(dev, "unhandled Relinquish Request IE\n");
+	}
+
+	return 0;
+error:
+	return result;
+}
diff --git a/drivers/uwb/ie.c b/drivers/uwb/ie.c
index cf6f3d1..ab97668 100644
--- a/drivers/uwb/ie.c
+++ b/drivers/uwb/ie.c
@@ -25,8 +25,6 @@
  */
 
 #include "uwb-internal.h"
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 /**
  * uwb_ie_next - get the next IE in a buffer
@@ -61,6 +59,42 @@
 EXPORT_SYMBOL_GPL(uwb_ie_next);
 
 /**
+ * uwb_ie_dump_hex - print IEs to a character buffer
+ * @ies: the IEs to print.
+ * @len: length of all the IEs.
+ * @buf: the destination buffer.
+ * @size: size of @buf.
+ *
+ * Returns the number of characters written.
+ */
+int uwb_ie_dump_hex(const struct uwb_ie_hdr *ies, size_t len,
+		    char *buf, size_t size)
+{
+	void *ptr;
+	const struct uwb_ie_hdr *ie;
+	int r = 0;
+	u8 *d;
+
+	ptr = (void *)ies;
+	for (;;) {
+		ie = uwb_ie_next(&ptr, &len);
+		if (!ie)
+			break;
+
+		r += scnprintf(buf + r, size - r, "%02x %02x",
+			       (unsigned)ie->element_id,
+			       (unsigned)ie->length);
+		d = (uint8_t *)ie + sizeof(struct uwb_ie_hdr);
+		while (d != ptr && r < size)
+			r += scnprintf(buf + r, size - r, " %02x", (unsigned)*d++);
+		if (r < size)
+			buf[r++] = '\n';
+	};
+
+	return r;
+}
+
+/**
  * Get the IEs that a radio controller is sending in its beacon
  *
  * @uwb_rc:  UWB Radio Controller
@@ -70,6 +104,7 @@
  * anything. Once done with the iedata buffer, call
  * uwb_rc_ie_release(iedata). Don't call kfree on it.
  */
+static
 ssize_t uwb_rc_get_ie(struct uwb_rc *uwb_rc, struct uwb_rc_evt_get_ie **pget_ie)
 {
 	ssize_t result;
@@ -78,148 +113,35 @@
 	struct uwb_rceb *reply = NULL;
 	struct uwb_rc_evt_get_ie *get_ie;
 
-	d_fnstart(3, dev, "(%p, %p)\n", uwb_rc, pget_ie);
-	result = -ENOMEM;
 	cmd = kzalloc(sizeof(*cmd), GFP_KERNEL);
 	if (cmd == NULL)
-		goto error_kzalloc;
+		return -ENOMEM;
+
 	cmd->bCommandType = UWB_RC_CET_GENERAL;
 	cmd->wCommand = cpu_to_le16(UWB_RC_CMD_GET_IE);
 	result = uwb_rc_vcmd(uwb_rc, "GET_IE", cmd, sizeof(*cmd),
 			     UWB_RC_CET_GENERAL, UWB_RC_CMD_GET_IE,
 			     &reply);
+	kfree(cmd);
 	if (result < 0)
-		goto error_cmd;
+		return result;
+
 	get_ie = container_of(reply, struct uwb_rc_evt_get_ie, rceb);
 	if (result < sizeof(*get_ie)) {
 		dev_err(dev, "not enough data returned for decoding GET IE "
 			"(%zu bytes received vs %zu needed)\n",
 			result, sizeof(*get_ie));
-		result = -EINVAL;
+		return -EINVAL;
 	} else if (result < sizeof(*get_ie) + le16_to_cpu(get_ie->wIELength)) {
 		dev_err(dev, "not enough data returned for decoding GET IE "
 			"payload (%zu bytes received vs %zu needed)\n", result,
 			sizeof(*get_ie) + le16_to_cpu(get_ie->wIELength));
-		result = -EINVAL;
-	} else
-		*pget_ie = get_ie;
-error_cmd:
-	kfree(cmd);
-error_kzalloc:
-	d_fnend(3, dev, "(%p, %p) = %d\n", uwb_rc, pget_ie, (int)result);
-	return result;
-}
-EXPORT_SYMBOL_GPL(uwb_rc_get_ie);
-
-
-/*
- * Given a pointer to an IE, print it in ASCII/hex followed by a new line
- *
- * @ie_hdr: pointer to the IE header. Length is in there, and it is
- *          guaranteed that the ie_hdr->length bytes following it are
- *          safely accesible.
- *
- * @_data: context data passed from uwb_ie_for_each(), an struct output_ctx
- */
-int uwb_ie_dump_hex(struct uwb_dev *uwb_dev, const struct uwb_ie_hdr *ie_hdr,
-		    size_t offset, void *_ctx)
-{
-	struct uwb_buf_ctx *ctx = _ctx;
-	const u8 *pl = (void *)(ie_hdr + 1);
-	u8 pl_itr;
-
-	ctx->bytes += scnprintf(ctx->buf + ctx->bytes, ctx->size - ctx->bytes,
-				"%02x %02x ", (unsigned) ie_hdr->element_id,
-				(unsigned) ie_hdr->length);
-	pl_itr = 0;
-	while (pl_itr < ie_hdr->length && ctx->bytes < ctx->size)
-		ctx->bytes += scnprintf(ctx->buf + ctx->bytes,
-					ctx->size - ctx->bytes,
-					"%02x ", (unsigned) pl[pl_itr++]);
-	if (ctx->bytes < ctx->size)
-		ctx->buf[ctx->bytes++] = '\n';
-	return 0;
-}
-EXPORT_SYMBOL_GPL(uwb_ie_dump_hex);
-
-
-/**
- * Verify that a pointer in a buffer points to valid IE
- *
- * @start: pointer to start of buffer in which IE appears
- * @itr:   pointer to IE inside buffer that will be verified
- * @top:   pointer to end of buffer
- *
- * @returns: 0 if IE is valid, <0 otherwise
- *
- * Verification involves checking that the buffer can contain a
- * header and the amount of data reported in the IE header can be found in
- * the buffer.
- */
-static
-int uwb_rc_ie_verify(struct uwb_dev *uwb_dev, const void *start,
-		     const void *itr, const void *top)
-{
-	struct device *dev = &uwb_dev->dev;
-	const struct uwb_ie_hdr *ie_hdr;
-
-	if (top - itr < sizeof(*ie_hdr)) {
-		dev_err(dev, "Bad IE: no data to decode header "
-			"(%zu bytes left vs %zu needed) at offset %zu\n",
-			top - itr, sizeof(*ie_hdr), itr - start);
 		return -EINVAL;
 	}
-	ie_hdr = itr;
-	itr += sizeof(*ie_hdr);
-	if (top - itr < ie_hdr->length) {
-		dev_err(dev, "Bad IE: not enough data for payload "
-			"(%zu bytes left vs %zu needed) at offset %zu\n",
-			top - itr, (size_t)ie_hdr->length,
-			(void *)ie_hdr - start);
-		return -EINVAL;
-	}
-	return 0;
-}
 
-
-/**
- * Walk a buffer filled with consecutive IE's a buffer
- *
- * @uwb_dev: UWB device this IEs belong to (for err messages mainly)
- *
- * @fn: function to call with each IE; if it returns 0, we keep
- *      traversing the buffer. If it returns !0, we'll stop and return
- *      that value.
- *
- * @data: pointer passed to @fn
- *
- * @buf: buffer where the consecutive IEs are located
- *
- * @size: size of @buf
- *
- * Each IE is checked for basic correctness (there is space left for
- * the header and the payload). If that test is failed, we stop
- * processing. For every good IE, @fn is called.
- */
-ssize_t uwb_ie_for_each(struct uwb_dev *uwb_dev, uwb_ie_f fn, void *data,
-			const void *buf, size_t size)
-{
-	ssize_t result = 0;
-	const struct uwb_ie_hdr *ie_hdr;
-	const void *itr = buf, *top = itr + size;
-
-	while (itr < top) {
-		if (uwb_rc_ie_verify(uwb_dev, buf, itr, top) != 0)
-			break;
-		ie_hdr = itr;
-		itr += sizeof(*ie_hdr) + ie_hdr->length;
-		result = fn(uwb_dev, ie_hdr, itr - buf, data);
-		if (result != 0)
-			break;
-	}
+	*pget_ie = get_ie;
 	return result;
 }
-EXPORT_SYMBOL_GPL(uwb_ie_for_each);
 
 
 /**
@@ -256,70 +178,6 @@
 	return result;
 }
 
-/**
- * Determine by IE id if IE is host settable
- * WUSB 1.0 [8.6.2.8 Table 8.85]
- *
- * EXCEPTION:
- * All but UWB_IE_WLP appears in Table 8.85 from WUSB 1.0. Setting this IE
- * is required for the WLP substack to perform association with its WSS so
- * we hope that the WUSB spec will be changed to reflect this.
- */
-static
-int uwb_rc_ie_is_host_settable(enum uwb_ie element_id)
-{
-	if (element_id == UWB_PCA_AVAILABILITY ||
-	    element_id == UWB_BP_SWITCH_IE ||
-	    element_id == UWB_MAC_CAPABILITIES_IE ||
-	    element_id == UWB_PHY_CAPABILITIES_IE ||
-	    element_id == UWB_APP_SPEC_PROBE_IE ||
-	    element_id == UWB_IDENTIFICATION_IE ||
-	    element_id == UWB_MASTER_KEY_ID_IE ||
-	    element_id == UWB_IE_WLP ||
-	    element_id == UWB_APP_SPEC_IE)
-		return 1;
-	return 0;
-}
-
-
-/**
- * Extract Host Settable IEs from IE
- *
- * @ie_data: pointer to buffer containing all IEs
- * @size:    size of buffer
- *
- * @returns: length of buffer that only includes host settable IEs
- *
- * Given a buffer of IEs we move all Host Settable IEs to front of buffer
- * by overwriting the IEs that are not Host Settable.
- * Buffer length is adjusted accordingly.
- */
-static
-ssize_t uwb_rc_parse_host_settable_ie(struct uwb_dev *uwb_dev,
-				      void *ie_data, size_t size)
-{
-	size_t new_len = size;
-	struct uwb_ie_hdr *ie_hdr;
-	size_t ie_length;
-	void *itr = ie_data, *top = itr + size;
-
-	while (itr < top) {
-		if (uwb_rc_ie_verify(uwb_dev, ie_data, itr, top) != 0)
-			break;
-		ie_hdr = itr;
-		ie_length = sizeof(*ie_hdr) + ie_hdr->length;
-		if (uwb_rc_ie_is_host_settable(ie_hdr->element_id)) {
-			itr += ie_length;
-		} else {
-			memmove(itr, itr + ie_length, top - (itr + ie_length));
-			new_len -= ie_length;
-			top -= ie_length;
-		}
-	}
-	return new_len;
-}
-
-
 /* Cleanup the whole IE management subsystem */
 void uwb_rc_ie_init(struct uwb_rc *uwb_rc)
 {
@@ -328,49 +186,34 @@
 
 
 /**
- * Set up cache for host settable IEs currently being transmitted
+ * uwb_rc_ie_setup - setup a radio controller's IE manager
+ * @uwb_rc: the radio controller.
  *
- * First we just call GET-IE to get the current IEs being transmitted
- * (or we workaround and pretend we did) and (because the format is
- * the same) reuse that as the IE cache (with the command prefix, as
- * explained in 'struct uwb_rc').
+ * The current set of IEs are obtained from the hardware with a GET-IE
+ * command (since the radio controller is not yet beaconing this will
+ * be just the hardware's MAC and PHY Capability IEs).
  *
- * @returns: size of cache created
+ * Returns 0 on success; -ve on an error.
  */
-ssize_t uwb_rc_ie_setup(struct uwb_rc *uwb_rc)
+int uwb_rc_ie_setup(struct uwb_rc *uwb_rc)
 {
-	struct device *dev = &uwb_rc->uwb_dev.dev;
-	ssize_t result;
-	size_t capacity;
-	struct uwb_rc_evt_get_ie *ie_info;
+	struct uwb_rc_evt_get_ie *ie_info = NULL;
+	int capacity;
 
-	d_fnstart(3, dev, "(%p)\n", uwb_rc);
+	capacity = uwb_rc_get_ie(uwb_rc, &ie_info);
+	if (capacity < 0)
+		return capacity;
+
 	mutex_lock(&uwb_rc->ies_mutex);
-	result = uwb_rc_get_ie(uwb_rc, &ie_info);
-	if (result < 0)
-		goto error_get_ie;
-	capacity = result;
-	d_printf(5, dev, "Got IEs %zu bytes (%zu long at %p)\n", result,
-		 (size_t)le16_to_cpu(ie_info->wIELength), ie_info);
 
-	/* Remove IEs that host should not set. */
-	result = uwb_rc_parse_host_settable_ie(&uwb_rc->uwb_dev,
-			ie_info->IEData, le16_to_cpu(ie_info->wIELength));
-	if (result < 0)
-		goto error_parse;
-	d_printf(5, dev, "purged non-settable IEs to %zu bytes\n", result);
-	uwb_rc->ies = (void *) ie_info;
+	uwb_rc->ies = (struct uwb_rc_cmd_set_ie *)ie_info;
 	uwb_rc->ies->rccb.bCommandType = UWB_RC_CET_GENERAL;
 	uwb_rc->ies->rccb.wCommand = cpu_to_le16(UWB_RC_CMD_SET_IE);
 	uwb_rc->ies_capacity = capacity;
-	d_printf(5, dev, "IE cache at %p %zu bytes, %zu capacity\n",
-		 ie_info, result, capacity);
-	result = 0;
-error_parse:
-error_get_ie:
+
 	mutex_unlock(&uwb_rc->ies_mutex);
-	d_fnend(3, dev, "(%p) = %zu\n", uwb_rc, result);
-	return result;
+
+	return 0;
 }
 
 
@@ -383,26 +226,47 @@
 }
 
 
-static
-int __acc_size(struct uwb_dev *uwb_dev, const struct uwb_ie_hdr *ie_hdr,
-	       size_t offset, void *_ctx)
+static int uwb_rc_ie_add_one(struct uwb_rc *rc, const struct uwb_ie_hdr *new_ie)
 {
-	size_t *acc_size = _ctx;
-	*acc_size += sizeof(*ie_hdr) + ie_hdr->length;
-	d_printf(6, &uwb_dev->dev, "new acc size %zu\n", *acc_size);
+	struct uwb_rc_cmd_set_ie *new_ies;
+	void *ptr, *prev_ie;
+	struct uwb_ie_hdr *ie;
+	size_t length, new_ie_len, new_capacity, size, prev_size;
+
+	length = le16_to_cpu(rc->ies->wIELength);
+	new_ie_len = sizeof(struct uwb_ie_hdr) + new_ie->length;
+	new_capacity = sizeof(struct uwb_rc_cmd_set_ie) + length + new_ie_len;
+
+	if (new_capacity > rc->ies_capacity) {
+		new_ies = krealloc(rc->ies, new_capacity, GFP_KERNEL);
+		if (!new_ies)
+			return -ENOMEM;
+		rc->ies = new_ies;
+	}
+
+	ptr = rc->ies->IEData;
+	size = length;
+	for (;;) {
+		prev_ie = ptr;
+		prev_size = size;
+		ie = uwb_ie_next(&ptr, &size);
+		if (!ie || ie->element_id > new_ie->element_id)
+			break;
+	}
+
+	memmove(prev_ie + new_ie_len, prev_ie, prev_size);
+	memcpy(prev_ie, new_ie, new_ie_len);
+	rc->ies->wIELength = cpu_to_le16(length + new_ie_len);
+
 	return 0;
 }
 
-
 /**
- * Add a new IE to IEs currently being transmitted by device
- *
+ * uwb_rc_ie_add - add new IEs to the radio controller's beacon
+ * @uwb_rc: the radio controller.
  * @ies: the buffer containing the new IE or IEs to be added to
- *       the device's beacon. The buffer will be verified for
- *       consistence (meaning the headers should be right) and
- *       consistent with the buffer size.
- * @size: size of @ies (in bytes, total buffer size)
- * @returns: 0 if ok, <0 errno code on error
+ *       the device's beacon.
+ * @size: length of all the IEs.
  *
  * According to WHCI 0.95 [4.13.6] the driver will only receive the RCEB
  * after the device sent the first beacon that includes the IEs specified
@@ -411,66 +275,40 @@
  * we start beaconing.
  *
  * Setting an IE on the device will overwrite all current IEs in device. So
- * we take the current IEs being transmitted by the device, append the
+ * we take the current IEs being transmitted by the device, insert the
  * new one, and call SET IE with all the IEs needed.
  *
- * The local IE cache will only be updated with the new IE if SET IE
- * completed successfully.
+ * Returns 0 on success; or -ENOMEM.
  */
 int uwb_rc_ie_add(struct uwb_rc *uwb_rc,
 		  const struct uwb_ie_hdr *ies, size_t size)
 {
 	int result = 0;
-	struct device *dev = &uwb_rc->uwb_dev.dev;
-	struct uwb_rc_cmd_set_ie *new_ies;
-	size_t ies_size, total_size, acc_size = 0;
+	void *ptr;
+	const struct uwb_ie_hdr *ie;
 
-	if (uwb_rc->ies == NULL)
-		return -ESHUTDOWN;
-	uwb_ie_for_each(&uwb_rc->uwb_dev, __acc_size, &acc_size, ies, size);
-	if (acc_size != size) {
-		dev_err(dev, "BUG: bad IEs, misconstructed headers "
-			"[%zu bytes reported vs %zu calculated]\n",
-			size, acc_size);
-		WARN_ON(1);
-		return -EINVAL;
-	}
 	mutex_lock(&uwb_rc->ies_mutex);
-	ies_size = le16_to_cpu(uwb_rc->ies->wIELength);
-	total_size = sizeof(*uwb_rc->ies) + ies_size;
-	if (total_size + size > uwb_rc->ies_capacity) {
-		d_printf(4, dev, "Reallocating IE cache from %p capacity %zu "
-			 "to capacity %zu\n", uwb_rc->ies, uwb_rc->ies_capacity,
-			 total_size + size);
-		new_ies = kzalloc(total_size + size, GFP_KERNEL);
-		if (new_ies == NULL) {
-			dev_err(dev, "No memory for adding new IE\n");
-			result = -ENOMEM;
-			goto error_alloc;
-		}
-		memcpy(new_ies, uwb_rc->ies, total_size);
-		uwb_rc->ies_capacity = total_size + size;
-		kfree(uwb_rc->ies);
-		uwb_rc->ies = new_ies;
-		d_printf(4, dev, "New IE cache at %p capacity %zu\n",
-			 uwb_rc->ies, uwb_rc->ies_capacity);
+
+	ptr = (void *)ies;
+	for (;;) {
+		ie = uwb_ie_next(&ptr, &size);
+		if (!ie)
+			break;
+
+		result = uwb_rc_ie_add_one(uwb_rc, ie);
+		if (result < 0)
+			break;
 	}
-	memcpy((void *)uwb_rc->ies + total_size, ies, size);
-	uwb_rc->ies->wIELength = cpu_to_le16(ies_size + size);
-	if (uwb_rc->beaconing != -1) {
-		result = uwb_rc_set_ie(uwb_rc, uwb_rc->ies);
-		if (result < 0) {
-			dev_err(dev, "Cannot set new IE on device: %d\n",
-				result);
-			uwb_rc->ies->wIELength = cpu_to_le16(ies_size);
+	if (result >= 0) {
+		if (size == 0) {
+			if (uwb_rc->beaconing != -1)
+				result = uwb_rc_set_ie(uwb_rc, uwb_rc->ies);
 		} else
-			result = 0;
+			result = -EINVAL;
 	}
-	d_printf(4, dev, "IEs now occupy %hu bytes of %zu capacity at %p\n",
-		 le16_to_cpu(uwb_rc->ies->wIELength), uwb_rc->ies_capacity,
-		 uwb_rc->ies);
-error_alloc:
+
 	mutex_unlock(&uwb_rc->ies_mutex);
+
 	return result;
 }
 EXPORT_SYMBOL_GPL(uwb_rc_ie_add);
@@ -489,53 +327,52 @@
  * beacon. We don't reallocate, we just mark the size smaller.
  */
 static
-int uwb_rc_ie_cache_rm(struct uwb_rc *uwb_rc, enum uwb_ie to_remove)
+void uwb_rc_ie_cache_rm(struct uwb_rc *uwb_rc, enum uwb_ie to_remove)
 {
-	struct uwb_ie_hdr *ie_hdr;
-	size_t new_len = le16_to_cpu(uwb_rc->ies->wIELength);
-	void *itr = uwb_rc->ies->IEData;
-	void *top = itr + new_len;
+	struct uwb_ie_hdr *ie;
+	size_t len = le16_to_cpu(uwb_rc->ies->wIELength);
+	void *ptr;
+	size_t size;
 
-	while (itr < top) {
-		ie_hdr = itr;
-		if (ie_hdr->element_id != to_remove) {
-			itr += sizeof(*ie_hdr) + ie_hdr->length;
-		} else {
-			int ie_length;
-			ie_length = sizeof(*ie_hdr) + ie_hdr->length;
-			if (top - itr != ie_length)
-				memmove(itr, itr + ie_length, top - itr + ie_length);
-			top -= ie_length;
-			new_len -= ie_length;
+	ptr = uwb_rc->ies->IEData;
+	size = len;
+	for (;;) {
+		ie = uwb_ie_next(&ptr, &size);
+		if (!ie)
+			break;
+		if (ie->element_id == to_remove) {
+			len -= sizeof(struct uwb_ie_hdr) + ie->length;
+			memmove(ie, ptr, size);
+			ptr = ie;
 		}
 	}
-	uwb_rc->ies->wIELength = cpu_to_le16(new_len);
-	return 0;
+	uwb_rc->ies->wIELength = cpu_to_le16(len);
 }
 
 
 /**
- * Remove an IE currently being transmitted by device
+ * uwb_rc_ie_rm - remove an IE from the radio controller's beacon
+ * @uwb_rc: the radio controller.
+ * @element_id: the element ID of the IE to remove.
  *
- * @element_id: id of IE to be removed from device's beacon
+ * Only IEs previously added with uwb_rc_ie_add() may be removed.
+ *
+ * Returns 0 on success; or -ve the SET-IE command to the radio
+ * controller failed.
  */
 int uwb_rc_ie_rm(struct uwb_rc *uwb_rc, enum uwb_ie element_id)
 {
-	struct device *dev = &uwb_rc->uwb_dev.dev;
-	int result;
+	int result = 0;
 
-	if (uwb_rc->ies == NULL)
-		return -ESHUTDOWN;
 	mutex_lock(&uwb_rc->ies_mutex);
-	result = uwb_rc_ie_cache_rm(uwb_rc, element_id);
-	if (result < 0)
-		dev_err(dev, "Cannot remove IE from cache.\n");
-	if (uwb_rc->beaconing != -1) {
+
+	uwb_rc_ie_cache_rm(uwb_rc, element_id);
+
+	if (uwb_rc->beaconing != -1)
 		result = uwb_rc_set_ie(uwb_rc, uwb_rc->ies);
-		if (result < 0)
-			dev_err(dev, "Cannot set new IE on device.\n");
-	}
+
 	mutex_unlock(&uwb_rc->ies_mutex);
+
 	return result;
 }
 EXPORT_SYMBOL_GPL(uwb_rc_ie_rm);
diff --git a/drivers/uwb/lc-dev.c b/drivers/uwb/lc-dev.c
index 15f856c..e9fe1bb 100644
--- a/drivers/uwb/lc-dev.c
+++ b/drivers/uwb/lc-dev.c
@@ -22,7 +22,6 @@
  *
  * FIXME: docs
  */
-
 #include <linux/kernel.h>
 #include <linux/device.h>
 #include <linux/err.h>
@@ -30,10 +29,6 @@
 #include <linux/random.h>
 #include "uwb-internal.h"
 
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
-
-
 /* We initialize addresses to 0xff (invalid, as it is bcast) */
 static inline void uwb_dev_addr_init(struct uwb_dev_addr *addr)
 {
@@ -104,12 +99,9 @@
 {
 	struct uwb_dev *uwb_dev = to_uwb_dev(dev);
 
-	d_fnstart(4, NULL, "(dev %p uwb_dev %p)\n", dev, uwb_dev);
 	uwb_bce_put(uwb_dev->bce);
-	d_printf(0, &uwb_dev->dev, "uwb_dev %p freed\n", uwb_dev);
 	memset(uwb_dev, 0x69, sizeof(*uwb_dev));
 	kfree(uwb_dev);
-	d_fnend(4, NULL, "(dev %p uwb_dev %p) = void\n", dev, uwb_dev);
 }
 
 /*
@@ -275,12 +267,8 @@
  */
 static int __uwb_dev_sys_add(struct uwb_dev *uwb_dev, struct device *parent_dev)
 {
-	int result;
 	struct device *dev;
 
-	d_fnstart(4, NULL, "(uwb_dev %p parent_dev %p)\n", uwb_dev, parent_dev);
-	BUG_ON(parent_dev == NULL);
-
 	dev = &uwb_dev->dev;
 	/* Device sysfs files are only useful for neighbor devices not
 	   local radio controllers. */
@@ -289,18 +277,14 @@
 	dev->parent = parent_dev;
 	dev_set_drvdata(dev, uwb_dev);
 
-	result = device_add(dev);
-	d_fnend(4, NULL, "(uwb_dev %p parent_dev %p) = %d\n", uwb_dev, parent_dev, result);
-	return result;
+	return device_add(dev);
 }
 
 
 static void __uwb_dev_sys_rm(struct uwb_dev *uwb_dev)
 {
-	d_fnstart(4, NULL, "(uwb_dev %p)\n", uwb_dev);
 	dev_set_drvdata(&uwb_dev->dev, NULL);
 	device_del(&uwb_dev->dev);
-	d_fnend(4, NULL, "(uwb_dev %p) = void\n", uwb_dev);
 }
 
 
@@ -384,7 +368,6 @@
 	struct device *dev = &uwb_dev->dev;
 	char macbuf[UWB_ADDR_STRSIZE], devbuf[UWB_ADDR_STRSIZE];
 
-	d_fnstart(3, NULL, "(dev %p [uwb_dev %p], uwb_rc %p)\n", dev, uwb_dev, rc);
 	uwb_mac_addr_print(macbuf, sizeof(macbuf), &uwb_dev->mac_addr);
 	uwb_dev_addr_print(devbuf, sizeof(devbuf), &uwb_dev->dev_addr);
 	dev_info(dev, "uwb device (mac %s dev %s) disconnected from %s %s\n",
@@ -392,8 +375,10 @@
 		 rc ? rc->uwb_dev.dev.parent->bus->name : "n/a",
 		 rc ? dev_name(rc->uwb_dev.dev.parent) : "");
 	uwb_dev_rm(uwb_dev);
+	list_del(&uwb_dev->bce->node);
+	uwb_bce_put(uwb_dev->bce);
 	uwb_dev_put(uwb_dev);	/* for the creation in _onair() */
-	d_fnend(3, NULL, "(dev %p [uwb_dev %p], uwb_rc %p) = 0\n", dev, uwb_dev, rc);
+
 	return 0;
 }
 
diff --git a/drivers/uwb/lc-rc.c b/drivers/uwb/lc-rc.c
index ee5772f..9cf21e6 100644
--- a/drivers/uwb/lc-rc.c
+++ b/drivers/uwb/lc-rc.c
@@ -36,8 +36,6 @@
 #include <linux/etherdevice.h>
 #include <linux/usb.h>
 
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
 #include "uwb-internal.h"
 
 static int uwb_rc_index_match(struct device *dev, void *data)
@@ -81,9 +79,7 @@
 	struct uwb_dev *uwb_dev = container_of(dev, struct uwb_dev, dev);
 	struct uwb_rc *rc = container_of(uwb_dev, struct uwb_rc, uwb_dev);
 
-	uwb_rc_neh_destroy(rc);
 	uwb_rc_ie_release(rc);
-	d_printf(1, dev, "freed uwb_rc %p\n", rc);
 	kfree(rc);
 }
 
@@ -100,6 +96,8 @@
 	rc->scan_type = UWB_SCAN_DISABLED;
 	INIT_LIST_HEAD(&rc->notifs_chain.list);
 	mutex_init(&rc->notifs_chain.mutex);
+	INIT_LIST_HEAD(&rc->uwb_beca.list);
+	mutex_init(&rc->uwb_beca.mutex);
 	uwb_drp_avail_init(rc);
 	uwb_rc_ie_init(rc);
 	uwb_rsv_init(rc);
@@ -191,9 +189,9 @@
 	int result;
 	struct device *dev = &rc->uwb_dev.dev;
 
-	result = uwb_rc_reset(rc);
+	result = uwb_radio_setup(rc);
 	if (result < 0) {
-		dev_err(dev, "cannot reset UWB radio: %d\n", result);
+		dev_err(dev, "cannot setup UWB radio: %d\n", result);
 		goto error;
 	}
 	result = uwb_rc_mac_addr_setup(rc);
@@ -250,6 +248,12 @@
 
 	rc->priv = priv;
 
+	init_waitqueue_head(&rc->uwbd.wq);
+	INIT_LIST_HEAD(&rc->uwbd.event_list);
+	spin_lock_init(&rc->uwbd.event_list_lock);
+
+	uwbd_start(rc);
+
 	result = rc->start(rc);
 	if (result < 0)
 		goto error_rc_start;
@@ -284,7 +288,7 @@
 error_dev_add:
 error_rc_setup:
 	rc->stop(rc);
-	uwbd_flush(rc);
+	uwbd_stop(rc);
 error_rc_start:
 	return result;
 }
@@ -306,25 +310,24 @@
 	rc->ready = 0;
 
 	uwb_dbg_del_rc(rc);
-	uwb_rsv_cleanup(rc);
-	uwb_rc_ie_rm(rc, UWB_IDENTIFICATION_IE);
-	if (rc->beaconing >= 0)
-		uwb_rc_beacon(rc, -1, 0);
-	if (rc->scan_type != UWB_SCAN_DISABLED)
-		uwb_rc_scan(rc, rc->scanning, UWB_SCAN_DISABLED, 0);
-	uwb_rc_reset(rc);
+	uwb_rsv_remove_all(rc);
+	uwb_radio_shutdown(rc);
 
 	rc->stop(rc);
-	uwbd_flush(rc);
+
+	uwbd_stop(rc);
+	uwb_rc_neh_destroy(rc);
 
 	uwb_dev_lock(&rc->uwb_dev);
 	rc->priv = NULL;
 	rc->cmd = NULL;
 	uwb_dev_unlock(&rc->uwb_dev);
-	mutex_lock(&uwb_beca.mutex);
+	mutex_lock(&rc->uwb_beca.mutex);
 	uwb_dev_for_each(rc, uwb_dev_offair_helper, NULL);
 	__uwb_rc_sys_rm(rc);
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
+	uwb_rsv_cleanup(rc);
+ 	uwb_beca_release(rc);
 	uwb_dev_rm(&rc->uwb_dev);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_rm);
@@ -468,28 +471,3 @@
 	__uwb_rc_put(rc);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_put);
-
-/*
- *
- *
- */
-ssize_t uwb_rc_print_IEs(struct uwb_rc *uwb_rc, char *buf, size_t size)
-{
-	ssize_t result;
-	struct uwb_rc_evt_get_ie *ie_info;
-	struct uwb_buf_ctx ctx;
-
-	result = uwb_rc_get_ie(uwb_rc, &ie_info);
-	if (result < 0)
-		goto error_get_ie;
-	ctx.buf = buf;
-	ctx.size = size;
-	ctx.bytes = 0;
-	uwb_ie_for_each(&uwb_rc->uwb_dev, uwb_ie_dump_hex, &ctx,
-			ie_info->IEData, result - sizeof(*ie_info));
-	result = ctx.bytes;
-	kfree(ie_info);
-error_get_ie:
-	return result;
-}
-
diff --git a/drivers/uwb/neh.c b/drivers/uwb/neh.c
index 9b4eb64..0af8916 100644
--- a/drivers/uwb/neh.c
+++ b/drivers/uwb/neh.c
@@ -86,8 +86,6 @@
 #include <linux/err.h>
 
 #include "uwb-internal.h"
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 /*
  * UWB Radio Controller Notification/Event Handle
@@ -254,7 +252,6 @@
 
 static void __uwb_rc_neh_rm(struct uwb_rc *rc, struct uwb_rc_neh *neh)
 {
-	del_timer(&neh->timer);
 	__uwb_rc_ctx_put(rc, neh);
 	list_del(&neh->list_node);
 }
@@ -275,6 +272,7 @@
 	__uwb_rc_neh_rm(rc, neh);
 	spin_unlock_irqrestore(&rc->neh_lock, flags);
 
+	del_timer_sync(&neh->timer);
 	uwb_rc_neh_put(neh);
 }
 
@@ -349,7 +347,7 @@
 }
 
 
-/**
+/*
  * Process notifications coming from the radio control interface
  *
  * @rc:    UWB Radio Control Interface descriptor
@@ -401,23 +399,6 @@
 	uwb_evt->notif.size = size;
 	uwb_evt->notif.rceb = rceb;
 
-	switch (le16_to_cpu(rceb->wEvent)) {
-		/* Trap some vendor specific events
-		 *
-		 * FIXME: move this to handling in ptc-est, where we
-		 * register a NULL event handler for these two guys
-		 * using the Intel IDs.
-		 */
-	case 0x0103:
-		dev_info(dev, "FIXME: DEVICE ADD\n");
-		return;
-	case 0x0104:
-		dev_info(dev, "FIXME: DEVICE RM\n");
-		return;
-	default:
-		break;
-	}
-
 	uwbd_event_queue(uwb_evt);
 }
 
@@ -438,9 +419,10 @@
 				rceb->bEventContext, size);
 	} else {
 		neh = uwb_rc_neh_lookup(rc, rceb);
-		if (neh)
+		if (neh) {
+			del_timer_sync(&neh->timer);
 			uwb_rc_neh_cb(neh, rceb, size);
-		else
+		} else
 			dev_warn(dev, "event 0x%02x/%04x/%02x (%zu bytes): nobody cared\n",
 				 rceb->bEventType, le16_to_cpu(rceb->wEvent),
 				 rceb->bEventContext, size);
@@ -495,8 +477,6 @@
 	size_t size, real_size, event_size;
 	int needtofree;
 
-	d_fnstart(3, dev, "(rc %p buf %p %zu buf_size)\n", rc, buf, buf_size);
-	d_printf(2, dev, "groking event block: %zu bytes\n", buf_size);
 	itr = buf;
 	size = buf_size;
 	while (size > 0) {
@@ -544,10 +524,7 @@
 
 		itr += real_size;
 		size -= real_size;
-		d_printf(2, dev, "consumed %zd bytes, %zu left\n",
-			 event_size, size);
 	}
-	d_fnend(3, dev, "(rc %p buf %p %zu buf_size) = void\n", rc, buf, buf_size);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_neh_grok);
 
@@ -562,16 +539,22 @@
  */
 void uwb_rc_neh_error(struct uwb_rc *rc, int error)
 {
-	struct uwb_rc_neh *neh, *next;
+	struct uwb_rc_neh *neh;
 	unsigned long flags;
 
-	BUG_ON(error >= 0);
-	spin_lock_irqsave(&rc->neh_lock, flags);
-	list_for_each_entry_safe(neh, next, &rc->neh_list, list_node) {
+	for (;;) {
+		spin_lock_irqsave(&rc->neh_lock, flags);
+		if (list_empty(&rc->neh_list)) {
+			spin_unlock_irqrestore(&rc->neh_lock, flags);
+			break;
+		}
+		neh = list_first_entry(&rc->neh_list, struct uwb_rc_neh, list_node);
 		__uwb_rc_neh_rm(rc, neh);
+		spin_unlock_irqrestore(&rc->neh_lock, flags);
+
+		del_timer_sync(&neh->timer);
 		uwb_rc_neh_cb(neh, NULL, error);
 	}
-	spin_unlock_irqrestore(&rc->neh_lock, flags);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_neh_error);
 
@@ -583,10 +566,14 @@
 	unsigned long flags;
 
 	spin_lock_irqsave(&rc->neh_lock, flags);
-	__uwb_rc_neh_rm(rc, neh);
+	if (neh->context)
+		__uwb_rc_neh_rm(rc, neh);
+	else
+		neh = NULL;
 	spin_unlock_irqrestore(&rc->neh_lock, flags);
 
-	uwb_rc_neh_cb(neh, NULL, -ETIMEDOUT);
+	if (neh)
+		uwb_rc_neh_cb(neh, NULL, -ETIMEDOUT);
 }
 
 /** Initializes the @rc's neh subsystem
@@ -605,12 +592,19 @@
 void uwb_rc_neh_destroy(struct uwb_rc *rc)
 {
 	unsigned long flags;
-	struct uwb_rc_neh *neh, *next;
+	struct uwb_rc_neh *neh;
 
-	spin_lock_irqsave(&rc->neh_lock, flags);
-	list_for_each_entry_safe(neh, next, &rc->neh_list, list_node) {
+	for (;;) {
+		spin_lock_irqsave(&rc->neh_lock, flags);
+		if (list_empty(&rc->neh_list)) {
+			spin_unlock_irqrestore(&rc->neh_lock, flags);
+			break;
+		}
+		neh = list_first_entry(&rc->neh_list, struct uwb_rc_neh, list_node);
 		__uwb_rc_neh_rm(rc, neh);
+		spin_unlock_irqrestore(&rc->neh_lock, flags);
+
+		del_timer_sync(&neh->timer);
 		uwb_rc_neh_put(neh);
 	}
-	spin_unlock_irqrestore(&rc->neh_lock, flags);
 }
diff --git a/drivers/uwb/pal.c b/drivers/uwb/pal.c
index 1afb38e..99a19c1 100644
--- a/drivers/uwb/pal.c
+++ b/drivers/uwb/pal.c
@@ -16,6 +16,7 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 #include <linux/kernel.h>
+#include <linux/debugfs.h>
 #include <linux/uwb.h>
 
 #include "uwb-internal.h"
@@ -32,13 +33,13 @@
 
 /**
  * uwb_pal_register - register a UWB PAL
- * @rc: the radio controller the PAL will be using
  * @pal: the PAL
  *
  * The PAL must be initialized with uwb_pal_init().
  */
-int uwb_pal_register(struct uwb_rc *rc, struct uwb_pal *pal)
+int uwb_pal_register(struct uwb_pal *pal)
 {
+	struct uwb_rc *rc = pal->rc;
 	int ret;
 
 	if (pal->device) {
@@ -54,9 +55,11 @@
 		}
 	}
 
-	spin_lock(&rc->pal_lock);
+	pal->debugfs_dir = uwb_dbg_create_pal_dir(pal);
+
+	mutex_lock(&rc->uwb_dev.mutex);
 	list_add(&pal->node, &rc->pals);
-	spin_unlock(&rc->pal_lock);
+	mutex_unlock(&rc->uwb_dev.mutex);
 
 	return 0;
 }
@@ -64,14 +67,19 @@
 
 /**
  * uwb_pal_register - unregister a UWB PAL
- * @rc: the radio controller the PAL was using
  * @pal: the PAL
  */
-void uwb_pal_unregister(struct uwb_rc *rc, struct uwb_pal *pal)
+void uwb_pal_unregister(struct uwb_pal *pal)
 {
-	spin_lock(&rc->pal_lock);
+	struct uwb_rc *rc = pal->rc;
+
+	uwb_radio_stop(pal);
+
+	mutex_lock(&rc->uwb_dev.mutex);
 	list_del(&pal->node);
-	spin_unlock(&rc->pal_lock);
+	mutex_unlock(&rc->uwb_dev.mutex);
+
+	debugfs_remove(pal->debugfs_dir);
 
 	if (pal->device) {
 		sysfs_remove_link(&rc->uwb_dev.dev.kobj, pal->name);
@@ -86,6 +94,5 @@
  */
 void uwb_rc_pal_init(struct uwb_rc *rc)
 {
-	spin_lock_init(&rc->pal_lock);
 	INIT_LIST_HEAD(&rc->pals);
 }
diff --git a/drivers/uwb/radio.c b/drivers/uwb/radio.c
new file mode 100644
index 0000000..f0d5549
--- /dev/null
+++ b/drivers/uwb/radio.c
@@ -0,0 +1,202 @@
+/*
+ * UWB radio (channel) management.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#include <linux/kernel.h>
+#include <linux/uwb.h>
+
+#include "uwb-internal.h"
+
+
+static int uwb_radio_select_channel(struct uwb_rc *rc)
+{
+	/*
+	 * Default to channel 9 (BG1, TFC1) unless the user has
+	 * selected a specific channel or there are no active PALs.
+	 */
+	if (rc->active_pals == 0)
+		return -1;
+	if (rc->beaconing_forced)
+		return rc->beaconing_forced;
+	return 9;
+}
+
+
+/*
+ * Notify all active PALs that the channel has changed.
+ */
+static void uwb_radio_channel_changed(struct uwb_rc *rc, int channel)
+{
+	struct uwb_pal *pal;
+
+	list_for_each_entry(pal, &rc->pals, node) {
+		if (pal->channel && channel != pal->channel) {
+			pal->channel = channel;
+			if (pal->channel_changed)
+				pal->channel_changed(pal, pal->channel);
+		}
+	}
+}
+
+/*
+ * Change to a new channel and notify any active PALs of the new
+ * channel.
+ *
+ * When stopping the radio, PALs need to be notified first so they can
+ * terminate any active reservations.
+ */
+static int uwb_radio_change_channel(struct uwb_rc *rc, int channel)
+{
+	int ret = 0;
+
+	if (channel == -1)
+		uwb_radio_channel_changed(rc, channel);
+
+	if (channel != rc->beaconing) {
+		if (rc->beaconing != -1 && channel != -1) {
+			/*
+			 * FIXME: should signal the channel change
+			 * with a Channel Change IE.
+			 */
+			ret = uwb_radio_change_channel(rc, -1);
+			if (ret < 0)
+				return ret;
+		}
+		ret = uwb_rc_beacon(rc, channel, 0);
+	}
+
+	if (channel != -1)
+		uwb_radio_channel_changed(rc, rc->beaconing);
+
+	return ret;
+}
+
+/**
+ * uwb_radio_start - request that the radio be started
+ * @pal: the PAL making the request.
+ *
+ * If the radio is not already active, aa suitable channel is selected
+ * and beacons are started.
+ */
+int uwb_radio_start(struct uwb_pal *pal)
+{
+	struct uwb_rc *rc = pal->rc;
+	int ret = 0;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	if (!pal->channel) {
+		pal->channel = -1;
+		rc->active_pals++;
+		ret = uwb_radio_change_channel(rc, uwb_radio_select_channel(rc));
+	}
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(uwb_radio_start);
+
+/**
+ * uwb_radio_stop - request tha the radio be stopped.
+ * @pal: the PAL making the request.
+ *
+ * Stops the radio if no other PAL is making use of it.
+ */
+void uwb_radio_stop(struct uwb_pal *pal)
+{
+	struct uwb_rc *rc = pal->rc;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	if (pal->channel) {
+		rc->active_pals--;
+		uwb_radio_change_channel(rc, uwb_radio_select_channel(rc));
+		pal->channel = 0;
+	}
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+}
+EXPORT_SYMBOL_GPL(uwb_radio_stop);
+
+/*
+ * uwb_radio_force_channel - force a specific channel to be used
+ * @rc: the radio controller.
+ * @channel: the channel to use; -1 to force the radio to stop; 0 to
+ *   use the default channel selection algorithm.
+ */
+int uwb_radio_force_channel(struct uwb_rc *rc, int channel)
+{
+	int ret = 0;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	rc->beaconing_forced = channel;
+	ret = uwb_radio_change_channel(rc, uwb_radio_select_channel(rc));
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+	return ret;
+}
+
+/*
+ * uwb_radio_setup - setup the radio manager
+ * @rc: the radio controller.
+ *
+ * The radio controller is reset to ensure it's in a known state
+ * before it's used.
+ */
+int uwb_radio_setup(struct uwb_rc *rc)
+{
+	return uwb_rc_reset(rc);
+}
+
+/*
+ * uwb_radio_reset_state - reset any radio manager state
+ * @rc: the radio controller.
+ *
+ * All internal radio manager state is reset to values corresponding
+ * to a reset radio controller.
+ */
+void uwb_radio_reset_state(struct uwb_rc *rc)
+{
+	struct uwb_pal *pal;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	list_for_each_entry(pal, &rc->pals, node) {
+		if (pal->channel) {
+			pal->channel = -1;
+			if (pal->channel_changed)
+				pal->channel_changed(pal, -1);
+		}
+	}
+
+	rc->beaconing = -1;
+	rc->scanning = -1;
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+}
+
+/*
+ * uwb_radio_shutdown - shutdown the radio manager
+ * @rc: the radio controller.
+ *
+ * The radio controller is reset.
+ */
+void uwb_radio_shutdown(struct uwb_rc *rc)
+{
+	uwb_radio_reset_state(rc);
+	uwb_rc_reset(rc);
+}
diff --git a/drivers/uwb/reset.c b/drivers/uwb/reset.c
index 8de856f..70f8050 100644
--- a/drivers/uwb/reset.c
+++ b/drivers/uwb/reset.c
@@ -32,8 +32,6 @@
 #include <linux/err.h>
 
 #include "uwb-internal.h"
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 /**
  * Command result codes (WUSB1.0[T8-69])
@@ -323,17 +321,16 @@
 	struct uwb_rc *rc = evt->rc;
 	int ret;
 
-	/* Need to prevent the RC hardware module going away while in
-	   the rc->reset() call. */
-	if (!try_module_get(rc->owner))
-		return 0;
-
 	dev_info(&rc->uwb_dev.dev, "resetting radio controller\n");
 	ret = rc->reset(rc);
-	if (ret)
+	if (ret) {
 		dev_err(&rc->uwb_dev.dev, "failed to reset hardware: %d\n", ret);
-
-	module_put(rc->owner);
+		goto error;
+	}
+	return 0;
+error:
+	/* Nothing can be done except try the reset again. */
+	uwb_rc_reset_all(rc);
 	return ret;
 }
 
@@ -360,3 +357,33 @@
 	uwbd_event_queue(evt);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_reset_all);
+
+void uwb_rc_pre_reset(struct uwb_rc *rc)
+{
+	rc->stop(rc);
+	uwbd_flush(rc);
+
+	uwb_radio_reset_state(rc);
+	uwb_rsv_remove_all(rc);
+}
+EXPORT_SYMBOL_GPL(uwb_rc_pre_reset);
+
+void uwb_rc_post_reset(struct uwb_rc *rc)
+{
+	int ret;
+
+	ret = rc->start(rc);
+	if (ret)
+		goto error;
+	ret = uwb_rc_mac_addr_set(rc, &rc->uwb_dev.mac_addr);
+	if (ret)
+		goto error;
+	ret = uwb_rc_dev_addr_set(rc, &rc->uwb_dev.dev_addr);
+	if (ret)
+		goto error;
+	return;
+error:
+	/* Nothing can be done except try the reset again. */
+	uwb_rc_reset_all(rc);
+}
+EXPORT_SYMBOL_GPL(uwb_rc_post_reset);
diff --git a/drivers/uwb/rsv.c b/drivers/uwb/rsv.c
index bae16204..ec6eecb 100644
--- a/drivers/uwb/rsv.c
+++ b/drivers/uwb/rsv.c
@@ -15,23 +15,33 @@
  * You should have received a copy of the GNU General Public License
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/uwb.h>
+#include <linux/random.h>
 
 #include "uwb-internal.h"
 
 static void uwb_rsv_timer(unsigned long arg);
 
 static const char *rsv_states[] = {
-	[UWB_RSV_STATE_NONE]          = "none",
-	[UWB_RSV_STATE_O_INITIATED]   = "initiated",
-	[UWB_RSV_STATE_O_PENDING]     = "pending",
-	[UWB_RSV_STATE_O_MODIFIED]    = "modified",
-	[UWB_RSV_STATE_O_ESTABLISHED] = "established",
-	[UWB_RSV_STATE_T_ACCEPTED]    = "accepted",
-	[UWB_RSV_STATE_T_DENIED]      = "denied",
-	[UWB_RSV_STATE_T_PENDING]     = "pending",
+	[UWB_RSV_STATE_NONE]                 = "none            ",
+	[UWB_RSV_STATE_O_INITIATED]          = "o initiated     ",
+	[UWB_RSV_STATE_O_PENDING]            = "o pending       ",
+	[UWB_RSV_STATE_O_MODIFIED]           = "o modified      ",
+	[UWB_RSV_STATE_O_ESTABLISHED]        = "o established   ",
+	[UWB_RSV_STATE_O_TO_BE_MOVED]        = "o to be moved   ",
+	[UWB_RSV_STATE_O_MOVE_EXPANDING]     = "o move expanding",
+	[UWB_RSV_STATE_O_MOVE_COMBINING]     = "o move combining",
+	[UWB_RSV_STATE_O_MOVE_REDUCING]      = "o move reducing ",
+	[UWB_RSV_STATE_T_ACCEPTED]           = "t accepted      ",
+	[UWB_RSV_STATE_T_CONFLICT]           = "t conflict      ",
+	[UWB_RSV_STATE_T_PENDING]            = "t pending       ",
+	[UWB_RSV_STATE_T_DENIED]             = "t denied        ",
+	[UWB_RSV_STATE_T_RESIZED]            = "t resized       ",
+	[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = "t expanding acc ",
+	[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = "t expanding conf",
+	[UWB_RSV_STATE_T_EXPANDING_PENDING]  = "t expanding pend",
+	[UWB_RSV_STATE_T_EXPANDING_DENIED]   = "t expanding den ",
 };
 
 static const char *rsv_types[] = {
@@ -42,6 +52,31 @@
 	[UWB_DRP_TYPE_PCA]      = "pca",
 };
 
+bool uwb_rsv_has_two_drp_ies(struct uwb_rsv *rsv)
+{
+	static const bool has_two_drp_ies[] = {
+		[UWB_RSV_STATE_O_INITIATED]               = false,
+		[UWB_RSV_STATE_O_PENDING]                 = false,
+		[UWB_RSV_STATE_O_MODIFIED]                = false,
+		[UWB_RSV_STATE_O_ESTABLISHED]             = false,
+		[UWB_RSV_STATE_O_TO_BE_MOVED]             = false,
+		[UWB_RSV_STATE_O_MOVE_COMBINING]          = false,
+		[UWB_RSV_STATE_O_MOVE_REDUCING]           = false,
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]          = true,
+		[UWB_RSV_STATE_T_ACCEPTED]                = false,
+		[UWB_RSV_STATE_T_CONFLICT]                = false,
+		[UWB_RSV_STATE_T_PENDING]                 = false,
+		[UWB_RSV_STATE_T_DENIED]                  = false,
+		[UWB_RSV_STATE_T_RESIZED]                 = false,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED]      = true,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT]      = true,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]       = true,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]        = true,
+	};
+
+	return has_two_drp_ies[rsv->state];
+}
+
 /**
  * uwb_rsv_state_str - return a string for a reservation state
  * @state: the reservation state.
@@ -66,7 +101,7 @@
 }
 EXPORT_SYMBOL_GPL(uwb_rsv_type_str);
 
-static void uwb_rsv_dump(struct uwb_rsv *rsv)
+void uwb_rsv_dump(char *text, struct uwb_rsv *rsv)
 {
 	struct device *dev = &rsv->rc->uwb_dev.dev;
 	struct uwb_dev_addr devaddr;
@@ -82,6 +117,23 @@
 	dev_dbg(dev, "rsv %s -> %s: %s\n", owner, target, uwb_rsv_state_str(rsv->state));
 }
 
+static void uwb_rsv_release(struct kref *kref)
+{
+	struct uwb_rsv *rsv = container_of(kref, struct uwb_rsv, kref);
+
+	kfree(rsv);
+}
+
+void uwb_rsv_get(struct uwb_rsv *rsv)
+{
+	kref_get(&rsv->kref);
+}
+
+void uwb_rsv_put(struct uwb_rsv *rsv)
+{
+	kref_put(&rsv->kref, uwb_rsv_release);
+}
+
 /*
  * Get a free stream index for a reservation.
  *
@@ -92,6 +144,7 @@
 static int uwb_rsv_get_stream(struct uwb_rsv *rsv)
 {
 	struct uwb_rc *rc = rsv->rc;
+	struct device *dev = &rc->uwb_dev.dev;
 	unsigned long *streams_bm;
 	int stream;
 
@@ -113,12 +166,15 @@
 	rsv->stream = stream;
 	set_bit(stream, streams_bm);
 
+	dev_dbg(dev, "get stream %d\n", rsv->stream);
+
 	return 0;
 }
 
 static void uwb_rsv_put_stream(struct uwb_rsv *rsv)
 {
 	struct uwb_rc *rc = rsv->rc;
+	struct device *dev = &rc->uwb_dev.dev;
 	unsigned long *streams_bm;
 
 	switch (rsv->target.type) {
@@ -133,86 +189,52 @@
 	}
 
 	clear_bit(rsv->stream, streams_bm);
+
+	dev_dbg(dev, "put stream %d\n", rsv->stream);
 }
 
-/*
- * Generate a MAS allocation with a single row component.
- */
-static void uwb_rsv_gen_alloc_row(struct uwb_mas_bm *mas,
-				  int first_mas, int mas_per_zone,
-				  int zs, int ze)
+void uwb_rsv_backoff_win_timer(unsigned long arg)
 {
-	struct uwb_mas_bm col;
-	int z;
+	struct uwb_drp_backoff_win *bow = (struct uwb_drp_backoff_win *)arg;
+	struct uwb_rc *rc = container_of(bow, struct uwb_rc, bow);
+	struct device *dev = &rc->uwb_dev.dev;
 
-	bitmap_zero(mas->bm, UWB_NUM_MAS);
-	bitmap_zero(col.bm, UWB_NUM_MAS);
-	bitmap_fill(col.bm, mas_per_zone);
-	bitmap_shift_left(col.bm, col.bm, first_mas + zs * UWB_MAS_PER_ZONE, UWB_NUM_MAS);
-
-	for (z = zs; z <= ze; z++) {
-		bitmap_or(mas->bm, mas->bm, col.bm, UWB_NUM_MAS);
-		bitmap_shift_left(col.bm, col.bm, UWB_MAS_PER_ZONE, UWB_NUM_MAS);
+	bow->can_reserve_extra_mases = true;
+	if (bow->total_expired <= 4) {
+		bow->total_expired++;
+	} else {
+		/* after 4 backoff window has expired we can exit from
+		 * the backoff procedure */
+		bow->total_expired = 0;
+		bow->window = UWB_DRP_BACKOFF_WIN_MIN >> 1;
 	}
+	dev_dbg(dev, "backoff_win_timer total_expired=%d, n=%d\n: ", bow->total_expired, bow->n);
+
+	/* try to relocate all the "to be moved" relocations */
+	uwb_rsv_handle_drp_avail_change(rc);
 }
 
-/*
- * Allocate some MAS for this reservation based on current local
- * availability, the reservation parameters (max_mas, min_mas,
- * sparsity), and the WiMedia rules for MAS allocations.
- *
- * Returns -EBUSY is insufficient free MAS are available.
- *
- * FIXME: to simplify this, only safe reservations with a single row
- * component in zones 1 to 15 are tried (zone 0 is skipped to avoid
- * problems with the MAS reserved for the BP).
- *
- * [ECMA-368] section B.2.
- */
-static int uwb_rsv_alloc_mas(struct uwb_rsv *rsv)
+void uwb_rsv_backoff_win_increment(struct uwb_rc *rc)
 {
-	static const int safe_mas_in_row[UWB_NUM_ZONES] = {
-		8, 7, 6, 5, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 2, 1,
-	};
-	int n, r;
-	struct uwb_mas_bm mas;
-	bool found = false;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct device *dev = &rc->uwb_dev.dev;
+	unsigned timeout_us;
 
-	/*
-	 * Search all valid safe allocations until either: too few MAS
-	 * are available; or the smallest allocation with sufficient
-	 * MAS is found.
-	 *
-	 * The top of the zones are preferred, so space for larger
-	 * allocations is available in the bottom of the zone (e.g., a
-	 * 15 MAS allocation should start in row 14 leaving space for
-	 * a 120 MAS allocation at row 0).
-	 */
-	for (n = safe_mas_in_row[0]; n >= 1; n--) {
-		int num_mas;
+	dev_dbg(dev, "backoff_win_increment: window=%d\n", bow->window);
 
-		num_mas = n * (UWB_NUM_ZONES - 1);
-		if (num_mas < rsv->min_mas)
-			break;
-		if (found && num_mas < rsv->max_mas)
-			break;
+	bow->can_reserve_extra_mases = false;
 
-		for (r = UWB_MAS_PER_ZONE-1;  r >= 0; r--) {
-			if (safe_mas_in_row[r] < n)
-				continue;
-			uwb_rsv_gen_alloc_row(&mas, r, n, 1, UWB_NUM_ZONES);
-			if (uwb_drp_avail_reserve_pending(rsv->rc, &mas) == 0) {
-				found = true;
-				break;
-			}
-		}
-	}
+	if((bow->window << 1) == UWB_DRP_BACKOFF_WIN_MAX)
+		return;
 
-	if (!found)
-		return -EBUSY;
+	bow->window <<= 1;
+	bow->n = random32() & (bow->window - 1);
+	dev_dbg(dev, "new_window=%d, n=%d\n: ", bow->window, bow->n);
 
-	bitmap_copy(rsv->mas.bm, mas.bm, UWB_NUM_MAS);
-	return 0;
+	/* reset the timer associated variables */
+	timeout_us = bow->n * UWB_SUPERFRAME_LENGTH_US;
+	bow->total_expired = 0;
+	mod_timer(&bow->timer, jiffies + usecs_to_jiffies(timeout_us));		
 }
 
 static void uwb_rsv_stroke_timer(struct uwb_rsv *rsv)
@@ -225,13 +247,16 @@
 	 * received.
 	 */
 	if (rsv->is_multicast) {
-		if (rsv->state == UWB_RSV_STATE_O_INITIATED)
+		if (rsv->state == UWB_RSV_STATE_O_INITIATED
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_EXPANDING
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_COMBINING
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_REDUCING)
 			sframes = 1;
 		if (rsv->state == UWB_RSV_STATE_O_ESTABLISHED)
 			sframes = 0;
+		
 	}
 
-	rsv->expired = false;
 	if (sframes > 0) {
 		/*
 		 * Add an additional 2 superframes to account for the
@@ -253,7 +278,7 @@
 	rsv->state = new_state;
 	rsv->ie_valid = false;
 
-	uwb_rsv_dump(rsv);
+	uwb_rsv_dump("SU", rsv);
 
 	uwb_rsv_stroke_timer(rsv);
 	uwb_rsv_sched_update(rsv->rc);
@@ -267,10 +292,17 @@
 
 void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state)
 {
+	struct uwb_rsv_move *mv = &rsv->mv;
+
 	if (rsv->state == new_state) {
 		switch (rsv->state) {
 		case UWB_RSV_STATE_O_ESTABLISHED:
+		case UWB_RSV_STATE_O_MOVE_EXPANDING:
+		case UWB_RSV_STATE_O_MOVE_COMBINING:
+		case UWB_RSV_STATE_O_MOVE_REDUCING:
 		case UWB_RSV_STATE_T_ACCEPTED:
+		case UWB_RSV_STATE_T_EXPANDING_ACCEPTED:
+		case UWB_RSV_STATE_T_RESIZED:
 		case UWB_RSV_STATE_NONE:
 			uwb_rsv_stroke_timer(rsv);
 			break;
@@ -282,10 +314,10 @@
 		return;
 	}
 
+	uwb_rsv_dump("SC", rsv);
+
 	switch (new_state) {
 	case UWB_RSV_STATE_NONE:
-		uwb_drp_avail_release(rsv->rc, &rsv->mas);
-		uwb_rsv_put_stream(rsv);
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_NONE);
 		uwb_rsv_callback(rsv);
 		break;
@@ -295,12 +327,45 @@
 	case UWB_RSV_STATE_O_PENDING:
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_PENDING);
 		break;
+	case UWB_RSV_STATE_O_MODIFIED:
+		/* in the companion there are the MASes to drop */
+		bitmap_andnot(rsv->mas.bm, rsv->mas.bm, mv->companion_mas.bm, UWB_NUM_MAS);
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MODIFIED);
+		break;
 	case UWB_RSV_STATE_O_ESTABLISHED:
+		if (rsv->state == UWB_RSV_STATE_O_MODIFIED
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_REDUCING) {
+			uwb_drp_avail_release(rsv->rc, &mv->companion_mas);
+			rsv->needs_release_companion_mas = false;
+		}
 		uwb_drp_avail_reserve(rsv->rc, &rsv->mas);
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_ESTABLISHED);
 		uwb_rsv_callback(rsv);
 		break;
+	case UWB_RSV_STATE_O_MOVE_EXPANDING:
+		rsv->needs_release_companion_mas = true;
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MOVE_EXPANDING);
+		break;
+	case UWB_RSV_STATE_O_MOVE_COMBINING:
+		rsv->needs_release_companion_mas = false;
+		uwb_drp_avail_reserve(rsv->rc, &mv->companion_mas);
+		bitmap_or(rsv->mas.bm, rsv->mas.bm, mv->companion_mas.bm, UWB_NUM_MAS);
+		rsv->mas.safe   += mv->companion_mas.safe;
+		rsv->mas.unsafe += mv->companion_mas.unsafe;
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+		break;
+	case UWB_RSV_STATE_O_MOVE_REDUCING:
+		bitmap_andnot(mv->companion_mas.bm, rsv->mas.bm, mv->final_mas.bm, UWB_NUM_MAS);
+		rsv->needs_release_companion_mas = true;
+		rsv->mas.safe   = mv->final_mas.safe;
+		rsv->mas.unsafe = mv->final_mas.unsafe;
+		bitmap_copy(rsv->mas.bm, mv->final_mas.bm, UWB_NUM_MAS);
+		bitmap_copy(rsv->mas.unsafe_bm, mv->final_mas.unsafe_bm, UWB_NUM_MAS);
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);
+		break;
 	case UWB_RSV_STATE_T_ACCEPTED:
+	case UWB_RSV_STATE_T_RESIZED:
+		rsv->needs_release_companion_mas = false;
 		uwb_drp_avail_reserve(rsv->rc, &rsv->mas);
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_ACCEPTED);
 		uwb_rsv_callback(rsv);
@@ -308,12 +373,82 @@
 	case UWB_RSV_STATE_T_DENIED:
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_DENIED);
 		break;
+	case UWB_RSV_STATE_T_CONFLICT:
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_CONFLICT);
+		break;
+	case UWB_RSV_STATE_T_PENDING:
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_PENDING);
+		break;
+	case UWB_RSV_STATE_T_EXPANDING_ACCEPTED:
+		rsv->needs_release_companion_mas = true;
+		uwb_drp_avail_reserve(rsv->rc, &mv->companion_mas);
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_EXPANDING_ACCEPTED);
+		break;
 	default:
 		dev_err(&rsv->rc->uwb_dev.dev, "unhandled state: %s (%d)\n",
 			uwb_rsv_state_str(new_state), new_state);
 	}
 }
 
+static void uwb_rsv_handle_timeout_work(struct work_struct *work)
+{
+	struct uwb_rsv *rsv = container_of(work, struct uwb_rsv,
+					   handle_timeout_work);
+	struct uwb_rc *rc = rsv->rc;
+
+	mutex_lock(&rc->rsvs_mutex);
+
+	uwb_rsv_dump("TO", rsv);
+
+	switch (rsv->state) {
+	case UWB_RSV_STATE_O_INITIATED:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_MOVE_EXPANDING:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_MOVE_COMBINING:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_MOVE_REDUCING:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_ESTABLISHED:
+		if (rsv->is_multicast)
+			goto unlock;
+		break;
+	case UWB_RSV_STATE_T_EXPANDING_ACCEPTED:
+		/*
+		 * The time out could be for the main or of the
+		 * companion DRP, assume it's for the companion and
+		 * drop that first.  A further time out is required to
+		 * drop the main.
+		 */
+		uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_ACCEPTED);
+		uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+		goto unlock;
+	default:
+		break;
+	}
+
+	uwb_rsv_remove(rsv);
+
+unlock:
+	mutex_unlock(&rc->rsvs_mutex);
+}
+
 static struct uwb_rsv *uwb_rsv_alloc(struct uwb_rc *rc)
 {
 	struct uwb_rsv *rsv;
@@ -324,23 +459,17 @@
 
 	INIT_LIST_HEAD(&rsv->rc_node);
 	INIT_LIST_HEAD(&rsv->pal_node);
+	kref_init(&rsv->kref);
 	init_timer(&rsv->timer);
 	rsv->timer.function = uwb_rsv_timer;
 	rsv->timer.data     = (unsigned long)rsv;
 
 	rsv->rc = rc;
+	INIT_WORK(&rsv->handle_timeout_work, uwb_rsv_handle_timeout_work);
 
 	return rsv;
 }
 
-static void uwb_rsv_free(struct uwb_rsv *rsv)
-{
-	uwb_dev_put(rsv->owner);
-	if (rsv->target.type == UWB_RSV_TARGET_DEV)
-		uwb_dev_put(rsv->target.dev);
-	kfree(rsv);
-}
-
 /**
  * uwb_rsv_create - allocate and initialize a UWB reservation structure
  * @rc: the radio controller
@@ -371,26 +500,36 @@
 
 void uwb_rsv_remove(struct uwb_rsv *rsv)
 {
+	uwb_rsv_dump("RM", rsv);
+
 	if (rsv->state != UWB_RSV_STATE_NONE)
 		uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
+
+	if (rsv->needs_release_companion_mas)
+		uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+	uwb_drp_avail_release(rsv->rc, &rsv->mas);
+
+	if (uwb_rsv_is_owner(rsv))
+		uwb_rsv_put_stream(rsv);
+	
 	del_timer_sync(&rsv->timer);
-	list_del(&rsv->rc_node);
-	uwb_rsv_free(rsv);
+	uwb_dev_put(rsv->owner);
+	if (rsv->target.type == UWB_RSV_TARGET_DEV)
+		uwb_dev_put(rsv->target.dev);
+
+	list_del_init(&rsv->rc_node);
+	uwb_rsv_put(rsv);
 }
 
 /**
  * uwb_rsv_destroy - free a UWB reservation structure
  * @rsv: the reservation to free
  *
- * The reservation will be terminated if it is pending or established.
+ * The reservation must already be terminated.
  */
 void uwb_rsv_destroy(struct uwb_rsv *rsv)
 {
-	struct uwb_rc *rc = rsv->rc;
-
-	mutex_lock(&rc->rsvs_mutex);
-	uwb_rsv_remove(rsv);
-	mutex_unlock(&rc->rsvs_mutex);
+	uwb_rsv_put(rsv);
 }
 EXPORT_SYMBOL_GPL(uwb_rsv_destroy);
 
@@ -399,7 +538,7 @@
  * @rsv: the reservation
  *
  * The PAL should fill in @rsv's owner, target, type, max_mas,
- * min_mas, sparsity and is_multicast fields.  If the target is a
+ * min_mas, max_interval and is_multicast fields.  If the target is a
  * uwb_dev it must be referenced.
  *
  * The reservation's callback will be called when the reservation is
@@ -408,20 +547,32 @@
 int uwb_rsv_establish(struct uwb_rsv *rsv)
 {
 	struct uwb_rc *rc = rsv->rc;
+	struct uwb_mas_bm available;
 	int ret;
 
 	mutex_lock(&rc->rsvs_mutex);
-
 	ret = uwb_rsv_get_stream(rsv);
 	if (ret)
 		goto out;
 
-	ret = uwb_rsv_alloc_mas(rsv);
-	if (ret) {
+	rsv->tiebreaker = random32() & 1;
+	/* get available mas bitmap */
+	uwb_drp_available(rc, &available);
+
+	ret = uwb_rsv_find_best_allocation(rsv, &available, &rsv->mas);
+	if (ret == UWB_RSV_ALLOC_NOT_FOUND) {
+		ret = -EBUSY;
 		uwb_rsv_put_stream(rsv);
 		goto out;
 	}
 
+	ret = uwb_drp_avail_reserve_pending(rc, &rsv->mas);
+	if (ret != 0) {
+		uwb_rsv_put_stream(rsv);
+		goto out;
+	}
+
+	uwb_rsv_get(rsv);
 	list_add_tail(&rsv->rc_node, &rc->reservations);
 	rsv->owner = &rc->uwb_dev;
 	uwb_dev_get(rsv->owner);
@@ -437,16 +588,71 @@
  * @rsv: the reservation to modify
  * @max_mas: new maximum MAS to reserve
  * @min_mas: new minimum MAS to reserve
- * @sparsity: new sparsity to use
+ * @max_interval: new max_interval to use
  *
  * FIXME: implement this once there are PALs that use it.
  */
-int uwb_rsv_modify(struct uwb_rsv *rsv, int max_mas, int min_mas, int sparsity)
+int uwb_rsv_modify(struct uwb_rsv *rsv, int max_mas, int min_mas, int max_interval)
 {
 	return -ENOSYS;
 }
 EXPORT_SYMBOL_GPL(uwb_rsv_modify);
 
+/*
+ * move an already established reservation (rc->rsvs_mutex must to be
+ * taken when tis function is called)
+ */
+int uwb_rsv_try_move(struct uwb_rsv *rsv, struct uwb_mas_bm *available)
+{
+	struct uwb_rc *rc = rsv->rc;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_rsv_move *mv;
+	int ret = 0;
+ 
+	if (bow->can_reserve_extra_mases == false)
+		return -EBUSY;
+
+	mv = &rsv->mv;
+
+	if (uwb_rsv_find_best_allocation(rsv, available, &mv->final_mas) == UWB_RSV_ALLOC_FOUND) {
+
+		if (!bitmap_equal(rsv->mas.bm, mv->final_mas.bm, UWB_NUM_MAS)) {
+			/* We want to move the reservation */
+			bitmap_andnot(mv->companion_mas.bm, mv->final_mas.bm, rsv->mas.bm, UWB_NUM_MAS);
+			uwb_drp_avail_reserve_pending(rc, &mv->companion_mas);
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_EXPANDING);
+		}
+	} else {
+		dev_dbg(dev, "new allocation not found\n");
+	}
+	
+	return ret;
+}
+
+/* It will try to move every reservation in state O_ESTABLISHED giving
+ * to the MAS allocator algorithm an availability that is the real one
+ * plus the allocation already established from the reservation. */
+void uwb_rsv_handle_drp_avail_change(struct uwb_rc *rc)
+{
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct uwb_rsv *rsv;
+	struct uwb_mas_bm mas;
+	
+	if (bow->can_reserve_extra_mases == false)
+		return;
+
+	list_for_each_entry(rsv, &rc->reservations, rc_node) {
+		if (rsv->state == UWB_RSV_STATE_O_ESTABLISHED ||
+		    rsv->state == UWB_RSV_STATE_O_TO_BE_MOVED) {
+			uwb_drp_available(rc, &mas);
+			bitmap_or(mas.bm, mas.bm, rsv->mas.bm, UWB_NUM_MAS);
+			uwb_rsv_try_move(rsv, &mas);
+		}
+	}
+	
+}
+
 /**
  * uwb_rsv_terminate - terminate an established reservation
  * @rsv: the reservation to terminate
@@ -463,7 +669,8 @@
 
 	mutex_lock(&rc->rsvs_mutex);
 
-	uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
+	if (rsv->state != UWB_RSV_STATE_NONE)
+		uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
 
 	mutex_unlock(&rc->rsvs_mutex);
 }
@@ -477,9 +684,14 @@
  *
  * Reservation requests from peers are denied unless a PAL accepts it
  * by calling this function.
+ *
+ * The PAL call uwb_rsv_destroy() for all accepted reservations before
+ * calling uwb_pal_unregister().
  */
 void uwb_rsv_accept(struct uwb_rsv *rsv, uwb_rsv_cb_f cb, void *pal_priv)
 {
+	uwb_rsv_get(rsv);
+
 	rsv->callback = cb;
 	rsv->pal_priv = pal_priv;
 	rsv->state    = UWB_RSV_STATE_T_ACCEPTED;
@@ -530,9 +742,9 @@
 	uwb_dev_get(rsv->owner);
 	rsv->target.type = UWB_RSV_TARGET_DEV;
 	rsv->target.dev  = &rc->uwb_dev;
+	uwb_dev_get(&rc->uwb_dev);
 	rsv->type        = uwb_ie_drp_type(drp_ie);
 	rsv->stream      = uwb_ie_drp_stream_index(drp_ie);
-	set_bit(rsv->stream, rsv->owner->streams);
 	uwb_drp_ie_to_bm(&rsv->mas, drp_ie);
 
 	/*
@@ -540,24 +752,46 @@
 	 * deny the request.
 	 */
 	rsv->state = UWB_RSV_STATE_T_DENIED;
-	spin_lock(&rc->pal_lock);
+	mutex_lock(&rc->uwb_dev.mutex);
 	list_for_each_entry(pal, &rc->pals, node) {
 		if (pal->new_rsv)
-			pal->new_rsv(rsv);
+			pal->new_rsv(pal, rsv);
 		if (rsv->state == UWB_RSV_STATE_T_ACCEPTED)
 			break;
 	}
-	spin_unlock(&rc->pal_lock);
+	mutex_unlock(&rc->uwb_dev.mutex);
 
 	list_add_tail(&rsv->rc_node, &rc->reservations);
 	state = rsv->state;
 	rsv->state = UWB_RSV_STATE_NONE;
-	uwb_rsv_set_state(rsv, state);
+
+	/* FIXME: do something sensible here */
+	if (state == UWB_RSV_STATE_T_ACCEPTED
+	    && uwb_drp_avail_reserve_pending(rc, &rsv->mas) == -EBUSY) {
+		/* FIXME: do something sensible here */
+	} else {
+		uwb_rsv_set_state(rsv, state);
+	}
 
 	return rsv;
 }
 
 /**
+ * uwb_rsv_get_usable_mas - get the bitmap of the usable MAS of a reservations
+ * @rsv: the reservation.
+ * @mas: returns the available MAS.
+ *
+ * The usable MAS of a reservation may be less than the negotiated MAS
+ * if alien BPs are present.
+ */
+void uwb_rsv_get_usable_mas(struct uwb_rsv *rsv, struct uwb_mas_bm *mas)
+{
+	bitmap_zero(mas->bm, UWB_NUM_MAS);
+	bitmap_andnot(mas->bm, rsv->mas.bm, rsv->rc->cnflt_alien_bitmap.bm, UWB_NUM_MAS);
+}
+EXPORT_SYMBOL_GPL(uwb_rsv_get_usable_mas);
+
+/**
  * uwb_rsv_find - find a reservation for a received DRP IE.
  * @rc: the radio controller
  * @src: source of the DRP IE
@@ -596,8 +830,6 @@
 	bool ie_updated = false;
 
 	list_for_each_entry_safe(rsv, t, &rc->reservations, rc_node) {
-		if (rsv->expired)
-			uwb_drp_handle_timeout(rsv);
 		if (!rsv->ie_valid) {
 			uwb_drp_ie_update(rsv);
 			ie_updated = true;
@@ -607,9 +839,47 @@
 	return ie_updated;
 }
 
+void uwb_rsv_queue_update(struct uwb_rc *rc)
+{
+	unsigned long delay_us = UWB_MAS_LENGTH_US * UWB_MAS_PER_ZONE;
+
+	queue_delayed_work(rc->rsv_workq, &rc->rsv_update_work, usecs_to_jiffies(delay_us));
+}
+
+/**
+ * uwb_rsv_sched_update - schedule an update of the DRP IEs
+ * @rc: the radio controller.
+ *
+ * To improve performance and ensure correctness with [ECMA-368] the
+ * number of SET-DRP-IE commands that are done are limited.
+ *
+ * DRP IEs update come from two sources: DRP events from the hardware
+ * which all occur at the beginning of the superframe ('syncronous'
+ * events) and reservation establishment/termination requests from
+ * PALs or timers ('asynchronous' events).
+ *
+ * A delayed work ensures that all the synchronous events result in
+ * one SET-DRP-IE command.
+ *
+ * Additional logic (the set_drp_ie_pending and rsv_updated_postponed
+ * flags) will prevent an asynchrous event starting a SET-DRP-IE
+ * command if one is currently awaiting a response.
+ *
+ * FIXME: this does leave a window where an asynchrous event can delay
+ * the SET-DRP-IE for a synchronous event by one superframe.
+ */
 void uwb_rsv_sched_update(struct uwb_rc *rc)
 {
-	queue_work(rc->rsv_workq, &rc->rsv_update_work);
+	spin_lock(&rc->rsvs_lock);
+	if (!delayed_work_pending(&rc->rsv_update_work)) {
+		if (rc->set_drp_ie_pending > 0) {
+			rc->set_drp_ie_pending++;
+			goto unlock;
+		}
+		uwb_rsv_queue_update(rc);
+	}
+unlock:
+	spin_unlock(&rc->rsvs_lock);
 }
 
 /*
@@ -618,7 +888,8 @@
  */
 static void uwb_rsv_update_work(struct work_struct *work)
 {
-	struct uwb_rc *rc = container_of(work, struct uwb_rc, rsv_update_work);
+	struct uwb_rc *rc = container_of(work, struct uwb_rc,
+					 rsv_update_work.work);
 	bool ie_updated;
 
 	mutex_lock(&rc->rsvs_mutex);
@@ -630,25 +901,71 @@
 		ie_updated = true;
 	}
 
-	if (ie_updated)
+	if (ie_updated && (rc->set_drp_ie_pending == 0))
 		uwb_rc_send_all_drp_ie(rc);
 
 	mutex_unlock(&rc->rsvs_mutex);
 }
 
+static void uwb_rsv_alien_bp_work(struct work_struct *work)
+{
+	struct uwb_rc *rc = container_of(work, struct uwb_rc,
+					 rsv_alien_bp_work.work);
+	struct uwb_rsv *rsv;
+
+	mutex_lock(&rc->rsvs_mutex);
+	
+	list_for_each_entry(rsv, &rc->reservations, rc_node) {
+		if (rsv->type != UWB_DRP_TYPE_ALIEN_BP) {
+			rsv->callback(rsv);
+		}
+	}
+
+	mutex_unlock(&rc->rsvs_mutex);
+}
+
 static void uwb_rsv_timer(unsigned long arg)
 {
 	struct uwb_rsv *rsv = (struct uwb_rsv *)arg;
 
-	rsv->expired = true;
-	uwb_rsv_sched_update(rsv->rc);
+	queue_work(rsv->rc->rsv_workq, &rsv->handle_timeout_work);
+}
+
+/**
+ * uwb_rsv_remove_all - remove all reservations
+ * @rc: the radio controller
+ *
+ * A DRP IE update is not done.
+ */
+void uwb_rsv_remove_all(struct uwb_rc *rc)
+{
+	struct uwb_rsv *rsv, *t;
+
+	mutex_lock(&rc->rsvs_mutex);
+	list_for_each_entry_safe(rsv, t, &rc->reservations, rc_node) {
+		uwb_rsv_remove(rsv);
+	}
+	/* Cancel any postponed update. */
+	rc->set_drp_ie_pending = 0;
+	mutex_unlock(&rc->rsvs_mutex);
+
+	cancel_delayed_work_sync(&rc->rsv_update_work);
 }
 
 void uwb_rsv_init(struct uwb_rc *rc)
 {
 	INIT_LIST_HEAD(&rc->reservations);
+	INIT_LIST_HEAD(&rc->cnflt_alien_list);
 	mutex_init(&rc->rsvs_mutex);
-	INIT_WORK(&rc->rsv_update_work, uwb_rsv_update_work);
+	spin_lock_init(&rc->rsvs_lock);
+	INIT_DELAYED_WORK(&rc->rsv_update_work, uwb_rsv_update_work);
+	INIT_DELAYED_WORK(&rc->rsv_alien_bp_work, uwb_rsv_alien_bp_work);
+	rc->bow.can_reserve_extra_mases = true;
+	rc->bow.total_expired = 0;
+	rc->bow.window = UWB_DRP_BACKOFF_WIN_MIN >> 1;
+	init_timer(&rc->bow.timer);
+	rc->bow.timer.function = uwb_rsv_backoff_win_timer;
+	rc->bow.timer.data     = (unsigned long)&rc->bow;
 
 	bitmap_complement(rc->uwb_dev.streams, rc->uwb_dev.streams, UWB_NUM_STREAMS);
 }
@@ -667,14 +984,6 @@
 
 void uwb_rsv_cleanup(struct uwb_rc *rc)
 {
-	struct uwb_rsv *rsv, *t;
-
-	mutex_lock(&rc->rsvs_mutex);
-	list_for_each_entry_safe(rsv, t, &rc->reservations, rc_node) {
-		uwb_rsv_remove(rsv);
-	}
-	mutex_unlock(&rc->rsvs_mutex);
-
-	cancel_work_sync(&rc->rsv_update_work);
+	uwb_rsv_remove_all(rc);
 	destroy_workqueue(rc->rsv_workq);
 }
diff --git a/drivers/uwb/umc-bus.c b/drivers/uwb/umc-bus.c
index 2d8d62d..5ad3616 100644
--- a/drivers/uwb/umc-bus.c
+++ b/drivers/uwb/umc-bus.c
@@ -11,23 +11,48 @@
 #include <linux/uwb/umc.h>
 #include <linux/pci.h>
 
-static int umc_bus_unbind_helper(struct device *dev, void *data)
+static int umc_bus_pre_reset_helper(struct device *dev, void *data)
 {
-	struct device *parent = data;
+	int ret = 0;
 
-	if (dev->parent == parent && dev->driver)
-		device_release_driver(dev);
-	return 0;
+	if (dev->driver) {
+		struct umc_dev *umc = to_umc_dev(dev);
+		struct umc_driver *umc_drv = to_umc_driver(dev->driver);
+
+		if (umc_drv->pre_reset)
+			ret = umc_drv->pre_reset(umc);
+		else
+			device_release_driver(dev);
+	}
+	return ret;
+}
+
+static int umc_bus_post_reset_helper(struct device *dev, void *data)
+{
+	int ret = 0;
+
+	if (dev->driver) {
+		struct umc_dev *umc = to_umc_dev(dev);
+		struct umc_driver *umc_drv = to_umc_driver(dev->driver);
+
+		if (umc_drv->post_reset)
+			ret = umc_drv->post_reset(umc);
+	} else
+		ret = device_attach(dev);
+
+	return ret;
 }
 
 /**
  * umc_controller_reset - reset the whole UMC controller
  * @umc: the UMC device for the radio controller.
  *
- * Drivers will be unbound from all UMC devices belonging to the
- * controller and then the radio controller will be rebound.  The
- * radio controller is expected to do a full hardware reset when it is
- * probed.
+ * Drivers or all capabilities of the controller will have their
+ * pre_reset methods called or be unbound from their device.  Then all
+ * post_reset methods will be called or the drivers will be rebound.
+ *
+ * Radio controllers must provide pre_reset and post_reset methods and
+ * reset the hardware in their start method.
  *
  * If this is called while a probe() or remove() is in progress it
  * will return -EAGAIN and not perform the reset.
@@ -35,14 +60,13 @@
 int umc_controller_reset(struct umc_dev *umc)
 {
 	struct device *parent = umc->dev.parent;
-	int ret;
+	int ret = 0;
 
-	if (down_trylock(&parent->sem))
+	if(down_trylock(&parent->sem))
 		return -EAGAIN;
-	bus_for_each_dev(&umc_bus_type, NULL, parent, umc_bus_unbind_helper);
-	ret = device_attach(&umc->dev);
-	if (ret == 1)
-		ret = 0;
+	ret = device_for_each_child(parent, parent, umc_bus_pre_reset_helper);
+	if (ret >= 0)
+		device_for_each_child(parent, parent, umc_bus_post_reset_helper);
 	up(&parent->sem);
 
 	return ret;
@@ -75,10 +99,10 @@
 	if (!dev->driver)
 		ret = device_attach(dev);
 
-	return ret < 0 ? ret : 0;
+	return ret;
 }
 
-static void umc_bus_rescan(void)
+static void umc_bus_rescan(struct device *parent)
 {
 	int err;
 
@@ -86,7 +110,7 @@
 	 * We can't use bus_rescan_devices() here as it deadlocks when
 	 * it tries to retake the dev->parent semaphore.
 	 */
-	err = bus_for_each_dev(&umc_bus_type, NULL, NULL, umc_bus_rescan_helper);
+	err = device_for_each_child(parent, NULL, umc_bus_rescan_helper);
 	if (err < 0)
 		printk(KERN_WARNING "%s: rescan of bus failed: %d\n",
 		       KBUILD_MODNAME, err);
@@ -120,7 +144,7 @@
 	if (err)
 		put_device(dev);
 	else
-		umc_bus_rescan();
+		umc_bus_rescan(dev->parent);
 
 	return err;
 }
diff --git a/drivers/uwb/umc-dev.c b/drivers/uwb/umc-dev.c
index aa44e1c..1fc7d82 100644
--- a/drivers/uwb/umc-dev.c
+++ b/drivers/uwb/umc-dev.c
@@ -7,8 +7,6 @@
  */
 #include <linux/kernel.h>
 #include <linux/uwb/umc.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 static void umc_device_release(struct device *dev)
 {
@@ -31,8 +29,7 @@
 
 	umc = kzalloc(sizeof(struct umc_dev), GFP_KERNEL);
 	if (umc) {
-		snprintf(umc->dev.bus_id, sizeof(umc->dev.bus_id), "%s-%d",
-			 parent->bus_id, n);
+		dev_set_name(&umc->dev, "%s-%d", dev_name(parent), n);
 		umc->dev.parent  = parent;
 		umc->dev.bus     = &umc_bus_type;
 		umc->dev.release = umc_device_release;
@@ -54,8 +51,6 @@
 {
 	int err;
 
-	d_fnstart(3, &umc->dev, "(umc_dev %p)\n", umc);
-
 	err = request_resource(umc->resource.parent, &umc->resource);
 	if (err < 0) {
 		dev_err(&umc->dev, "can't allocate resource range "
@@ -69,13 +64,11 @@
 	err = device_register(&umc->dev);
 	if (err < 0)
 		goto error_device_register;
-	d_fnend(3, &umc->dev, "(umc_dev %p) = 0\n", umc);
 	return 0;
 
 error_device_register:
 	release_resource(&umc->resource);
 error_request_resource:
-	d_fnend(3, &umc->dev, "(umc_dev %p) = %d\n", umc, err);
 	return err;
 }
 EXPORT_SYMBOL_GPL(umc_device_register);
@@ -95,10 +88,8 @@
 	if (!umc)
 		return;
 	dev = get_device(&umc->dev);
-	d_fnstart(3, dev, "(umc_dev %p)\n", umc);
 	device_unregister(&umc->dev);
 	release_resource(&umc->resource);
-	d_fnend(3, dev, "(umc_dev %p) = void\n", umc);
 	put_device(dev);
 }
 EXPORT_SYMBOL_GPL(umc_device_unregister);
diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index 6d232c3..4a42993 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -4,6 +4,7 @@
  *
  * Copyright (C) 2005-2006 Intel Corporation
  * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version
@@ -33,31 +34,9 @@
 #include <linux/seq_file.h>
 
 #include <linux/uwb/debug-cmd.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 #include "uwb-internal.h"
 
-void dump_bytes(struct device *dev, const void *_buf, size_t rsize)
-{
-	const char *buf = _buf;
-	char line[32];
-	size_t offset = 0;
-	int cnt, cnt2;
-	for (cnt = 0; cnt < rsize; cnt += 8) {
-		size_t rtop = rsize - cnt < 8 ? rsize - cnt : 8;
-		for (offset = cnt2 = 0; cnt2 < rtop; cnt2++) {
-			offset += scnprintf(line + offset, sizeof(line) - offset,
-					    "%02x ", buf[cnt + cnt2] & 0xff);
-		}
-		if (dev)
-			dev_info(dev, "%s\n", line);
-		else
-			printk(KERN_INFO "%s\n", line);
-	}
-}
-EXPORT_SYMBOL_GPL(dump_bytes);
-
 /*
  * Debug interface
  *
@@ -84,26 +63,23 @@
 	struct dentry *reservations_f;
 	struct dentry *accept_f;
 	struct dentry *drp_avail_f;
+	spinlock_t list_lock;
 };
 
 static struct dentry *root_dir;
 
 static void uwb_dbg_rsv_cb(struct uwb_rsv *rsv)
 {
-	struct uwb_rc *rc = rsv->rc;
-	struct device *dev = &rc->uwb_dev.dev;
-	struct uwb_dev_addr devaddr;
-	char owner[UWB_ADDR_STRSIZE], target[UWB_ADDR_STRSIZE];
+	struct uwb_dbg *dbg = rsv->pal_priv;
 
-	uwb_dev_addr_print(owner, sizeof(owner), &rsv->owner->dev_addr);
-	if (rsv->target.type == UWB_RSV_TARGET_DEV)
-		devaddr = rsv->target.dev->dev_addr;
-	else
-		devaddr = rsv->target.devaddr;
-	uwb_dev_addr_print(target, sizeof(target), &devaddr);
+	uwb_rsv_dump("debug", rsv);
 
-	dev_dbg(dev, "debug: rsv %s -> %s: %s\n",
-		owner, target, uwb_rsv_state_str(rsv->state));
+	if (rsv->state == UWB_RSV_STATE_NONE) {
+		spin_lock(&dbg->list_lock);
+		list_del(&rsv->pal_node);
+		spin_unlock(&dbg->list_lock);
+		uwb_rsv_destroy(rsv);
+	}
 }
 
 static int cmd_rsv_establish(struct uwb_rc *rc,
@@ -119,26 +95,27 @@
 	if (target == NULL)
 		return -ENODEV;
 
-	rsv = uwb_rsv_create(rc, uwb_dbg_rsv_cb, NULL);
+	rsv = uwb_rsv_create(rc, uwb_dbg_rsv_cb, rc->dbg);
 	if (rsv == NULL) {
 		uwb_dev_put(target);
 		return -ENOMEM;
 	}
 
-	rsv->owner       = &rc->uwb_dev;
-	rsv->target.type = UWB_RSV_TARGET_DEV;
-	rsv->target.dev  = target;
-	rsv->type        = cmd->type;
-	rsv->max_mas     = cmd->max_mas;
-	rsv->min_mas     = cmd->min_mas;
-	rsv->sparsity    = cmd->sparsity;
+	rsv->target.type  = UWB_RSV_TARGET_DEV;
+	rsv->target.dev   = target;
+	rsv->type         = cmd->type;
+	rsv->max_mas      = cmd->max_mas;
+	rsv->min_mas      = cmd->min_mas;
+	rsv->max_interval = cmd->max_interval;
 
 	ret = uwb_rsv_establish(rsv);
 	if (ret)
 		uwb_rsv_destroy(rsv);
-	else
+	else {
+		spin_lock(&(rc->dbg)->list_lock);
 		list_add_tail(&rsv->pal_node, &rc->dbg->rsvs);
-
+		spin_unlock(&(rc->dbg)->list_lock);
+	}
 	return ret;
 }
 
@@ -148,21 +125,40 @@
 	struct uwb_rsv *rsv, *found = NULL;
 	int i = 0;
 
+	spin_lock(&(rc->dbg)->list_lock);
+
 	list_for_each_entry(rsv, &rc->dbg->rsvs, pal_node) {
 		if (i == cmd->index) {
 			found = rsv;
+			uwb_rsv_get(found);
 			break;
 		}
+		i++;
 	}
+
+	spin_unlock(&(rc->dbg)->list_lock);
+
 	if (!found)
 		return -EINVAL;
 
-	list_del(&found->pal_node);
 	uwb_rsv_terminate(found);
+	uwb_rsv_put(found);
 
 	return 0;
 }
 
+static int cmd_ie_add(struct uwb_rc *rc, struct uwb_dbg_cmd_ie *ie_to_add)
+{
+	return uwb_rc_ie_add(rc,
+			     (const struct uwb_ie_hdr *) ie_to_add->data,
+			     ie_to_add->len);
+}
+
+static int cmd_ie_rm(struct uwb_rc *rc, struct uwb_dbg_cmd_ie *ie_to_rm)
+{
+	return uwb_rc_ie_rm(rc, ie_to_rm->data[0]);
+}
+
 static int command_open(struct inode *inode, struct file *file)
 {
 	file->private_data = inode->i_private;
@@ -175,8 +171,8 @@
 {
 	struct uwb_rc *rc = file->private_data;
 	struct uwb_dbg_cmd cmd;
-	int ret;
-
+	int ret = 0;
+	
 	if (len != sizeof(struct uwb_dbg_cmd))
 		return -EINVAL;
 
@@ -190,6 +186,18 @@
 	case UWB_DBG_CMD_RSV_TERMINATE:
 		ret = cmd_rsv_terminate(rc, &cmd.rsv_terminate);
 		break;
+	case UWB_DBG_CMD_IE_ADD:
+		ret = cmd_ie_add(rc, &cmd.ie_add);
+		break;
+	case UWB_DBG_CMD_IE_RM:
+		ret = cmd_ie_rm(rc, &cmd.ie_rm);
+		break;
+	case UWB_DBG_CMD_RADIO_START:
+		ret = uwb_radio_start(&rc->dbg->pal);
+		break;
+	case UWB_DBG_CMD_RADIO_STOP:
+		uwb_radio_stop(&rc->dbg->pal);
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -283,12 +291,26 @@
 	.owner   = THIS_MODULE,
 };
 
-static void uwb_dbg_new_rsv(struct uwb_rsv *rsv)
+static void uwb_dbg_channel_changed(struct uwb_pal *pal, int channel)
 {
-	struct uwb_rc *rc = rsv->rc;
+	struct device *dev = &pal->rc->uwb_dev.dev;
 
-	if (rc->dbg->accept)
-		uwb_rsv_accept(rsv, uwb_dbg_rsv_cb, NULL);
+	if (channel > 0)
+		dev_info(dev, "debug: channel %d started\n", channel);
+	else
+		dev_info(dev, "debug: channel stopped\n");
+}
+
+static void uwb_dbg_new_rsv(struct uwb_pal *pal, struct uwb_rsv *rsv)
+{
+	struct uwb_dbg *dbg = container_of(pal, struct uwb_dbg, pal);
+
+	if (dbg->accept) {
+		spin_lock(&dbg->list_lock);
+		list_add_tail(&rsv->pal_node, &dbg->rsvs);
+		spin_unlock(&dbg->list_lock);
+		uwb_rsv_accept(rsv, uwb_dbg_rsv_cb, dbg);
+	}
 }
 
 /**
@@ -302,10 +324,14 @@
 		return;
 
 	INIT_LIST_HEAD(&rc->dbg->rsvs);
+	spin_lock_init(&(rc->dbg)->list_lock);
 
 	uwb_pal_init(&rc->dbg->pal);
+	rc->dbg->pal.rc = rc;
+	rc->dbg->pal.channel_changed = uwb_dbg_channel_changed;
 	rc->dbg->pal.new_rsv = uwb_dbg_new_rsv;
-	uwb_pal_register(rc, &rc->dbg->pal);
+	uwb_pal_register(&rc->dbg->pal);
+
 	if (root_dir) {
 		rc->dbg->root_d = debugfs_create_dir(dev_name(&rc->uwb_dev.dev),
 						     root_dir);
@@ -325,7 +351,7 @@
 }
 
 /**
- * uwb_dbg_add_rc - remove a radio controller's debug interface
+ * uwb_dbg_del_rc - remove a radio controller's debug interface
  * @rc: the radio controller
  */
 void uwb_dbg_del_rc(struct uwb_rc *rc)
@@ -336,10 +362,10 @@
 		return;
 
 	list_for_each_entry_safe(rsv, t, &rc->dbg->rsvs, pal_node) {
-		uwb_rsv_destroy(rsv);
+		uwb_rsv_terminate(rsv);
 	}
 
-	uwb_pal_unregister(rc, &rc->dbg->pal);
+	uwb_pal_unregister(&rc->dbg->pal);
 
 	if (root_dir) {
 		debugfs_remove(rc->dbg->drp_avail_f);
@@ -365,3 +391,16 @@
 {
 	debugfs_remove(root_dir);
 }
+
+/**
+ * uwb_dbg_create_pal_dir - create a debugfs directory for a PAL
+ * @pal: The PAL.
+ */
+struct dentry *uwb_dbg_create_pal_dir(struct uwb_pal *pal)
+{
+	struct uwb_rc *rc = pal->rc;
+
+	if (root_dir && rc->dbg && rc->dbg->root_d && pal->name)
+		return debugfs_create_dir(pal->name, rc->dbg->root_d);
+	return NULL;
+}
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index 2ad307d..d5bcfc1 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -66,14 +66,14 @@
 		       unsigned channel, enum uwb_scan_type type,
 		       unsigned bpst_offset);
 extern int uwb_rc_send_all_drp_ie(struct uwb_rc *rc);
-extern ssize_t uwb_rc_print_IEs(struct uwb_rc *rc, char *, size_t);
-extern void uwb_rc_ie_init(struct uwb_rc *);
-extern void uwb_rc_ie_init(struct uwb_rc *);
-extern ssize_t uwb_rc_ie_setup(struct uwb_rc *);
-extern void uwb_rc_ie_release(struct uwb_rc *);
-extern int uwb_rc_ie_add(struct uwb_rc *,
-			 const struct uwb_ie_hdr *, size_t);
-extern int uwb_rc_ie_rm(struct uwb_rc *, enum uwb_ie);
+
+void uwb_rc_ie_init(struct uwb_rc *);
+int uwb_rc_ie_setup(struct uwb_rc *);
+void uwb_rc_ie_release(struct uwb_rc *);
+int uwb_ie_dump_hex(const struct uwb_ie_hdr *ies, size_t len,
+		    char *buf, size_t size);
+int uwb_rc_set_ie(struct uwb_rc *, struct uwb_rc_cmd_set_ie *);
+
 
 extern const char *uwb_rc_strerror(unsigned code);
 
@@ -92,6 +92,12 @@
 
 struct uwb_rc_neh;
 
+extern int uwb_rc_cmd_async(struct uwb_rc *rc, const char *cmd_name,
+			    struct uwb_rccb *cmd, size_t cmd_size,
+			    u8 expected_type, u16 expected_event,
+			    uwb_rc_cmd_cb_f cb, void *arg);
+
+
 void uwb_rc_neh_create(struct uwb_rc *rc);
 void uwb_rc_neh_destroy(struct uwb_rc *rc);
 
@@ -106,7 +112,69 @@
 extern int uwb_est_create(void);
 extern void uwb_est_destroy(void);
 
+/*
+ * UWB conflicting alien reservations
+ */
+struct uwb_cnflt_alien {
+	struct uwb_rc *rc;
+	struct list_head rc_node;
+	struct uwb_mas_bm mas;
+	struct timer_list timer;
+	struct work_struct cnflt_update_work;
+};
 
+enum uwb_uwb_rsv_alloc_result {
+	UWB_RSV_ALLOC_FOUND = 0,
+	UWB_RSV_ALLOC_NOT_FOUND,
+};
+
+enum uwb_rsv_mas_status {
+	UWB_RSV_MAS_NOT_AVAIL = 1,
+	UWB_RSV_MAS_SAFE,
+	UWB_RSV_MAS_UNSAFE,
+};
+
+struct uwb_rsv_col_set_info {
+	unsigned char start_col;
+	unsigned char interval;
+	unsigned char safe_mas_per_col;
+	unsigned char unsafe_mas_per_col;
+};
+
+struct uwb_rsv_col_info {
+	unsigned char max_avail_safe;
+	unsigned char max_avail_unsafe;
+	unsigned char highest_mas[UWB_MAS_PER_ZONE];
+	struct uwb_rsv_col_set_info csi;
+};
+
+struct uwb_rsv_row_info {
+	unsigned char avail[UWB_MAS_PER_ZONE];
+	unsigned char free_rows;
+	unsigned char used_rows;
+};
+
+/*
+ * UWB find allocation
+ */
+struct uwb_rsv_alloc_info {
+	unsigned char bm[UWB_MAS_PER_ZONE * UWB_NUM_ZONES];
+	struct uwb_rsv_col_info ci[UWB_NUM_ZONES];
+	struct uwb_rsv_row_info ri;
+	struct uwb_mas_bm *not_available;
+	struct uwb_mas_bm *result;
+	int min_mas;
+	int max_mas;
+	int max_interval;
+	int total_allocated_mases;
+	int safe_allocated_mases;
+	int unsafe_allocated_mases;
+	int interval;
+};
+
+int uwb_rsv_find_best_allocation(struct uwb_rsv *rsv, struct uwb_mas_bm *available, 
+				 struct uwb_mas_bm *result);
+void uwb_rsv_handle_drp_avail_change(struct uwb_rc *rc);
 /*
  * UWB Events & management daemon
  */
@@ -160,13 +228,14 @@
 	};
 };
 
-extern void uwbd_start(void);
-extern void uwbd_stop(void);
+extern void uwbd_start(struct uwb_rc *rc);
+extern void uwbd_stop(struct uwb_rc *rc);
 extern struct uwb_event *uwb_event_alloc(size_t, gfp_t gfp_mask);
 extern void uwbd_event_queue(struct uwb_event *);
 void uwbd_flush(struct uwb_rc *rc);
 
 /* UWB event handlers */
+extern int uwbd_evt_handle_rc_ie_rcv(struct uwb_event *);
 extern int uwbd_evt_handle_rc_beacon(struct uwb_event *);
 extern int uwbd_evt_handle_rc_beacon_size(struct uwb_event *);
 extern int uwbd_evt_handle_rc_bpoie_change(struct uwb_event *);
@@ -193,15 +262,6 @@
 
 extern unsigned long beacon_timeout_ms;
 
-/** Beacon cache list */
-struct uwb_beca {
-	struct list_head list;
-	size_t entries;
-	struct mutex mutex;
-};
-
-extern struct uwb_beca uwb_beca;
-
 /**
  * Beacon cache entry
  *
@@ -228,9 +288,6 @@
 struct uwb_beacon_frame;
 extern ssize_t uwb_bce_print_IEs(struct uwb_dev *, struct uwb_beca_e *,
 				 char *, size_t);
-extern struct uwb_beca_e *__uwb_beca_add(struct uwb_rc_evt_beacon *,
-					 struct uwb_beacon_frame *,
-					 unsigned long);
 
 extern void uwb_bce_kfree(struct kref *_bce);
 static inline void uwb_bce_get(struct uwb_beca_e *bce)
@@ -241,14 +298,19 @@
 {
 	kref_put(&bce->refcnt, uwb_bce_kfree);
 }
-extern void uwb_beca_purge(void);
-extern void uwb_beca_release(void);
+extern void uwb_beca_purge(struct uwb_rc *rc);
+extern void uwb_beca_release(struct uwb_rc *rc);
 
 struct uwb_dev *uwb_dev_get_by_devaddr(struct uwb_rc *rc,
 				       const struct uwb_dev_addr *devaddr);
 struct uwb_dev *uwb_dev_get_by_macaddr(struct uwb_rc *rc,
 				       const struct uwb_mac_addr *macaddr);
 
+int uwb_radio_setup(struct uwb_rc *rc);
+void uwb_radio_reset_state(struct uwb_rc *rc);
+void uwb_radio_shutdown(struct uwb_rc *rc);
+int uwb_radio_force_channel(struct uwb_rc *rc, int channel);
+
 /* -- UWB Sysfs representation */
 extern struct class uwb_rc_class;
 extern struct device_attribute dev_attr_mac_address;
@@ -259,18 +321,29 @@
 void uwb_rsv_init(struct uwb_rc *rc);
 int uwb_rsv_setup(struct uwb_rc *rc);
 void uwb_rsv_cleanup(struct uwb_rc *rc);
+void uwb_rsv_remove_all(struct uwb_rc *rc);
+void uwb_rsv_get(struct uwb_rsv *rsv);
+void uwb_rsv_put(struct uwb_rsv *rsv);
+bool uwb_rsv_has_two_drp_ies(struct uwb_rsv *rsv);
+void uwb_rsv_dump(char *text, struct uwb_rsv *rsv);
+int uwb_rsv_try_move(struct uwb_rsv *rsv, struct uwb_mas_bm *available);
+void uwb_rsv_backoff_win_timer(unsigned long arg);
+void uwb_rsv_backoff_win_increment(struct uwb_rc *rc);
+int uwb_rsv_status(struct uwb_rsv *rsv);
+int uwb_rsv_companion_status(struct uwb_rsv *rsv);
 
 void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state);
 void uwb_rsv_remove(struct uwb_rsv *rsv);
 struct uwb_rsv *uwb_rsv_find(struct uwb_rc *rc, struct uwb_dev *src,
 			     struct uwb_ie_drp *drp_ie);
 void uwb_rsv_sched_update(struct uwb_rc *rc);
+void uwb_rsv_queue_update(struct uwb_rc *rc);
 
-void uwb_drp_handle_timeout(struct uwb_rsv *rsv);
 int uwb_drp_ie_update(struct uwb_rsv *rsv);
 void uwb_drp_ie_to_bm(struct uwb_mas_bm *bm, const struct uwb_ie_drp *drp_ie);
 
 void uwb_drp_avail_init(struct uwb_rc *rc);
+void uwb_drp_available(struct uwb_rc *rc, struct uwb_mas_bm *avail);
 int  uwb_drp_avail_reserve_pending(struct uwb_rc *rc, struct uwb_mas_bm *mas);
 void uwb_drp_avail_reserve(struct uwb_rc *rc, struct uwb_mas_bm *mas);
 void uwb_drp_avail_release(struct uwb_rc *rc, struct uwb_mas_bm *mas);
@@ -289,8 +362,7 @@
 void uwb_dbg_exit(void);
 void uwb_dbg_add_rc(struct uwb_rc *rc);
 void uwb_dbg_del_rc(struct uwb_rc *rc);
-
-/* Workarounds for version specific stuff */
+struct dentry *uwb_dbg_create_pal_dir(struct uwb_pal *pal);
 
 static inline void uwb_dev_lock(struct uwb_dev *uwb_dev)
 {
diff --git a/drivers/uwb/uwbd.c b/drivers/uwb/uwbd.c
index 7890841..57bd6bf 100644
--- a/drivers/uwb/uwbd.c
+++ b/drivers/uwb/uwbd.c
@@ -68,17 +68,13 @@
  *
  * Handler functions are called normally uwbd_evt_handle_*().
  */
-
 #include <linux/kthread.h>
 #include <linux/module.h>
 #include <linux/freezer.h>
+
 #include "uwb-internal.h"
 
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
-
-
-/**
+/*
  * UWBD Event handler function signature
  *
  * Return !0 if the event needs not to be freed (ie the handler
@@ -101,9 +97,12 @@
 	const char *name;
 };
 
-/** Table of handlers for and properties of the UWBD Radio Control Events */
-static
-struct uwbd_event uwbd_events[] = {
+/* Table of handlers for and properties of the UWBD Radio Control Events */
+static struct uwbd_event uwbd_urc_events[] = {
+	[UWB_RC_EVT_IE_RCV] = {
+		.handler = uwbd_evt_handle_rc_ie_rcv,
+		.name = "IE_RECEIVED"
+	},
 	[UWB_RC_EVT_BEACON] = {
 		.handler = uwbd_evt_handle_rc_beacon,
 		.name = "BEACON_RECEIVED"
@@ -142,23 +141,15 @@
 	size_t size;
 };
 
-#define UWBD_EVT_TYPE_HANDLER(n,a) {		\
-	.name = (n),				\
-	.uwbd_events = (a),			\
-	.size = sizeof(a)/sizeof((a)[0])	\
-}
-
-
-/** Table of handlers for each UWBD Event type. */
-static
-struct uwbd_evt_type_handler uwbd_evt_type_handlers[] = {
-	[UWB_RC_CET_GENERAL] = UWBD_EVT_TYPE_HANDLER("RC", uwbd_events)
+/* Table of handlers for each UWBD Event type. */
+static struct uwbd_evt_type_handler uwbd_urc_evt_type_handlers[] = {
+	[UWB_RC_CET_GENERAL] = {
+		.name        = "URC",
+		.uwbd_events = uwbd_urc_events,
+		.size        = ARRAY_SIZE(uwbd_urc_events),
+	},
 };
 
-static const
-size_t uwbd_evt_type_handlers_len =
-	sizeof(uwbd_evt_type_handlers) / sizeof(uwbd_evt_type_handlers[0]);
-
 static const struct uwbd_event uwbd_message_handlers[] = {
 	[UWB_EVT_MSG_RESET] = {
 		.handler = uwbd_msg_handle_reset,
@@ -166,9 +157,7 @@
 	},
 };
 
-static DEFINE_MUTEX(uwbd_event_mutex);
-
-/**
+/*
  * Handle an URC event passed to the UWB Daemon
  *
  * @evt: the event to handle
@@ -188,6 +177,7 @@
 static
 int uwbd_event_handle_urc(struct uwb_event *evt)
 {
+	int result = -EINVAL;
 	struct uwbd_evt_type_handler *type_table;
 	uwbd_evt_handler_f handler;
 	u8 type, context;
@@ -197,26 +187,24 @@
 	event = le16_to_cpu(evt->notif.rceb->wEvent);
 	context = evt->notif.rceb->bEventContext;
 
-	if (type > uwbd_evt_type_handlers_len) {
-		printk(KERN_ERR "UWBD: event type %u: unknown (too high)\n", type);
-		return -EINVAL;
-	}
-	type_table = &uwbd_evt_type_handlers[type];
-	if (type_table->uwbd_events == NULL) {
-		printk(KERN_ERR "UWBD: event type %u: unknown\n", type);
-		return -EINVAL;
-	}
-	if (event > type_table->size) {
-		printk(KERN_ERR "UWBD: event %s[%u]: unknown (too high)\n",
-		       type_table->name, event);
-		return -EINVAL;
-	}
+	if (type > ARRAY_SIZE(uwbd_urc_evt_type_handlers))
+		goto out;
+	type_table = &uwbd_urc_evt_type_handlers[type];
+	if (type_table->uwbd_events == NULL)
+		goto out;
+	if (event > type_table->size)
+		goto out;
 	handler = type_table->uwbd_events[event].handler;
-	if (handler == NULL) {
-		printk(KERN_ERR "UWBD: event %s[%u]: unknown\n", type_table->name, event);
-		return -EINVAL;
-	}
-	return (*handler)(evt);
+	if (handler == NULL)
+		goto out;
+
+	result = (*handler)(evt);
+out:
+	if (result < 0)
+		dev_err(&evt->rc->uwb_dev.dev,
+			"UWBD: event 0x%02x/%04x/%02x, handling failed: %d\n",
+			type, event, context, result);
+	return result;
 }
 
 static void uwbd_event_handle_message(struct uwb_event *evt)
@@ -231,19 +219,10 @@
 		return;
 	}
 
-	/* If this is a reset event we need to drop the
-	 * uwbd_event_mutex or it deadlocks when the reset handler
-	 * attempts to flush the uwbd events. */
-	if (evt->message == UWB_EVT_MSG_RESET)
-		mutex_unlock(&uwbd_event_mutex);
-
 	result = uwbd_message_handlers[evt->message].handler(evt);
 	if (result < 0)
 		dev_err(&rc->uwb_dev.dev, "UWBD: '%s' message failed: %d\n",
 			uwbd_message_handlers[evt->message].name, result);
-
-	if (evt->message == UWB_EVT_MSG_RESET)
-		mutex_lock(&uwbd_event_mutex);
 }
 
 static void uwbd_event_handle(struct uwb_event *evt)
@@ -271,20 +250,6 @@
 
 	__uwb_rc_put(rc);	/* for the __uwb_rc_get() in uwb_rc_notif_cb() */
 }
-/* The UWB Daemon */
-
-
-/** Daemon's PID: used to decide if we can queue or not */
-static int uwbd_pid;
-/** Daemon's task struct for managing the kthread */
-static struct task_struct *uwbd_task;
-/** Daemon's waitqueue for waiting for new events */
-static DECLARE_WAIT_QUEUE_HEAD(uwbd_wq);
-/** Daemon's list of events; we queue/dequeue here */
-static struct list_head uwbd_event_list = LIST_HEAD_INIT(uwbd_event_list);
-/** Daemon's list lock to protect concurent access */
-static DEFINE_SPINLOCK(uwbd_event_list_lock);
-
 
 /**
  * UWB Daemon
@@ -298,65 +263,58 @@
  * FIXME: should change so we don't have a 1HZ timer all the time, but
  *        only if there are devices.
  */
-static int uwbd(void *unused)
+static int uwbd(void *param)
 {
+	struct uwb_rc *rc = param;
 	unsigned long flags;
-	struct list_head list = LIST_HEAD_INIT(list);
-	struct uwb_event *evt, *nxt;
+	struct uwb_event *evt;
 	int should_stop = 0;
+
 	while (1) {
 		wait_event_interruptible_timeout(
-			uwbd_wq,
-			!list_empty(&uwbd_event_list)
+			rc->uwbd.wq,
+			!list_empty(&rc->uwbd.event_list)
 			  || (should_stop = kthread_should_stop()),
 			HZ);
 		if (should_stop)
 			break;
 		try_to_freeze();
 
-		mutex_lock(&uwbd_event_mutex);
-		spin_lock_irqsave(&uwbd_event_list_lock, flags);
-		list_splice_init(&uwbd_event_list, &list);
-		spin_unlock_irqrestore(&uwbd_event_list_lock, flags);
-		list_for_each_entry_safe(evt, nxt, &list, list_node) {
+		spin_lock_irqsave(&rc->uwbd.event_list_lock, flags);
+		if (!list_empty(&rc->uwbd.event_list)) {
+			evt = list_first_entry(&rc->uwbd.event_list, struct uwb_event, list_node);
 			list_del(&evt->list_node);
+		} else
+			evt = NULL;
+		spin_unlock_irqrestore(&rc->uwbd.event_list_lock, flags);
+
+		if (evt) {
 			uwbd_event_handle(evt);
 			kfree(evt);
 		}
-		mutex_unlock(&uwbd_event_mutex);
 
-		uwb_beca_purge();	/* Purge devices that left */
+		uwb_beca_purge(rc);	/* Purge devices that left */
 	}
 	return 0;
 }
 
 
 /** Start the UWB daemon */
-void uwbd_start(void)
+void uwbd_start(struct uwb_rc *rc)
 {
-	uwbd_task = kthread_run(uwbd, NULL, "uwbd");
-	if (uwbd_task == NULL)
+	rc->uwbd.task = kthread_run(uwbd, rc, "uwbd");
+	if (rc->uwbd.task == NULL)
 		printk(KERN_ERR "UWB: Cannot start management daemon; "
 		       "UWB won't work\n");
 	else
-		uwbd_pid = uwbd_task->pid;
+		rc->uwbd.pid = rc->uwbd.task->pid;
 }
 
 /* Stop the UWB daemon and free any unprocessed events */
-void uwbd_stop(void)
+void uwbd_stop(struct uwb_rc *rc)
 {
-	unsigned long flags;
-	struct uwb_event *evt, *nxt;
-	kthread_stop(uwbd_task);
-	spin_lock_irqsave(&uwbd_event_list_lock, flags);
-	uwbd_pid = 0;
-	list_for_each_entry_safe(evt, nxt, &uwbd_event_list, list_node) {
-		if (evt->type == UWB_EVT_TYPE_NOTIF)
-			kfree(evt->notif.rceb);
-		kfree(evt);
-	}
-	spin_unlock_irqrestore(&uwbd_event_list_lock, flags);
-	uwb_beca_release();
+	kthread_stop(rc->uwbd.task);
+	uwbd_flush(rc);
 }
 
 /*
@@ -373,18 +331,20 @@
  */
 void uwbd_event_queue(struct uwb_event *evt)
 {
+	struct uwb_rc *rc = evt->rc;
 	unsigned long flags;
-	spin_lock_irqsave(&uwbd_event_list_lock, flags);
-	if (uwbd_pid != 0) {
-		list_add(&evt->list_node, &uwbd_event_list);
-		wake_up_all(&uwbd_wq);
+
+	spin_lock_irqsave(&rc->uwbd.event_list_lock, flags);
+	if (rc->uwbd.pid != 0) {
+		list_add(&evt->list_node, &rc->uwbd.event_list);
+		wake_up_all(&rc->uwbd.wq);
 	} else {
 		__uwb_rc_put(evt->rc);
 		if (evt->type == UWB_EVT_TYPE_NOTIF)
 			kfree(evt->notif.rceb);
 		kfree(evt);
 	}
-	spin_unlock_irqrestore(&uwbd_event_list_lock, flags);
+	spin_unlock_irqrestore(&rc->uwbd.event_list_lock, flags);
 	return;
 }
 
@@ -392,10 +352,8 @@
 {
 	struct uwb_event *evt, *nxt;
 
-	mutex_lock(&uwbd_event_mutex);
-
-	spin_lock_irq(&uwbd_event_list_lock);
-	list_for_each_entry_safe(evt, nxt, &uwbd_event_list, list_node) {
+	spin_lock_irq(&rc->uwbd.event_list_lock);
+	list_for_each_entry_safe(evt, nxt, &rc->uwbd.event_list, list_node) {
 		if (evt->rc == rc) {
 			__uwb_rc_put(rc);
 			list_del(&evt->list_node);
@@ -404,7 +362,5 @@
 			kfree(evt);
 		}
 	}
-	spin_unlock_irq(&uwbd_event_list_lock);
-
-	mutex_unlock(&uwbd_event_mutex);
+	spin_unlock_irq(&rc->uwbd.event_list_lock);
 }
diff --git a/drivers/uwb/whc-rc.c b/drivers/uwb/whc-rc.c
index 1711dea..19a1dd1 100644
--- a/drivers/uwb/whc-rc.c
+++ b/drivers/uwb/whc-rc.c
@@ -39,7 +39,6 @@
  * them to the hw and transfer the replies/notifications back to the
  * UWB stack through the UWB daemon (UWBD).
  */
-#include <linux/version.h>
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/pci.h>
@@ -49,10 +48,8 @@
 #include <linux/uwb.h>
 #include <linux/uwb/whci.h>
 #include <linux/uwb/umc.h>
-#include "uwb-internal.h"
 
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
+#include "uwb-internal.h"
 
 /**
  * Descriptor for an instance of the UWB Radio Control Driver that
@@ -98,13 +95,8 @@
 	struct device *dev = &whcrc->umc_dev->dev;
 	u32 urccmd;
 
-	d_fnstart(3, dev, "(%p, %p, %zu)\n", uwb_rc, cmd, cmd_size);
-	might_sleep();
-
-	if (cmd_size >= 4096) {
-		result = -E2BIG;
-		goto error;
-	}
+	if (cmd_size >= 4096)
+		return -EINVAL;
 
 	/*
 	 * If the URC is halted, then the hardware has reset itself.
@@ -115,16 +107,14 @@
 	if (le_readl(whcrc->rc_base + URCSTS) & URCSTS_HALTED) {
 		dev_err(dev, "requesting reset of halted radio controller\n");
 		uwb_rc_reset_all(uwb_rc);
-		result = -EIO;
-		goto error;
+		return -EIO;
 	}
 
 	result = wait_event_timeout(whcrc->cmd_wq,
 		!(le_readl(whcrc->rc_base + URCCMD) & URCCMD_ACTIVE), HZ/2);
 	if (result == 0) {
 		dev_err(dev, "device is not ready to execute commands\n");
-		result = -ETIMEDOUT;
-		goto error;
+		return -ETIMEDOUT;
 	}
 
 	memmove(whcrc->cmd_buf, cmd, cmd_size);
@@ -137,10 +127,7 @@
 		  whcrc->rc_base + URCCMD);
 	spin_unlock(&whcrc->irq_lock);
 
-error:
-	d_fnend(3, dev, "(%p, %p, %zu) = %d\n",
-		uwb_rc, cmd, cmd_size, result);
-	return result;
+	return 0;
 }
 
 static int whcrc_reset(struct uwb_rc *rc)
@@ -167,34 +154,25 @@
 static
 void whcrc_enable_events(struct whcrc *whcrc)
 {
-	struct device *dev = &whcrc->umc_dev->dev;
 	u32 urccmd;
 
-	d_fnstart(4, dev, "(whcrc %p)\n", whcrc);
-
 	le_writeq(whcrc->evt_dma_buf, whcrc->rc_base + URCEVTADDR);
 
 	spin_lock(&whcrc->irq_lock);
 	urccmd = le_readl(whcrc->rc_base + URCCMD) & ~URCCMD_ACTIVE;
 	le_writel(urccmd | URCCMD_EARV, whcrc->rc_base + URCCMD);
 	spin_unlock(&whcrc->irq_lock);
-
-	d_fnend(4, dev, "(whcrc %p) = void\n", whcrc);
 }
 
 static void whcrc_event_work(struct work_struct *work)
 {
 	struct whcrc *whcrc = container_of(work, struct whcrc, event_work);
-	struct device *dev = &whcrc->umc_dev->dev;
 	size_t size;
 	u64 urcevtaddr;
 
 	urcevtaddr = le_readq(whcrc->rc_base + URCEVTADDR);
 	size = urcevtaddr & URCEVTADDR_OFFSET_MASK;
 
-	d_printf(3, dev, "received %zu octet event\n", size);
-	d_dump(4, dev, whcrc->evt_buf, size > 32 ? 32 : size);
-
 	uwb_rc_neh_grok(whcrc->uwb_rc, whcrc->evt_buf, size);
 	whcrc_enable_events(whcrc);
 }
@@ -217,22 +195,15 @@
 		return IRQ_NONE;
 	le_writel(urcsts & URCSTS_INT_MASK, whcrc->rc_base + URCSTS);
 
-	d_printf(4, dev, "acked 0x%08x, urcsts 0x%08x\n",
-		 le_readl(whcrc->rc_base + URCSTS), urcsts);
-
 	if (urcsts & URCSTS_HSE) {
 		dev_err(dev, "host system error -- hardware halted\n");
 		/* FIXME: do something sensible here */
 		goto out;
 	}
-	if (urcsts & URCSTS_ER) {
-		d_printf(3, dev, "ER: event ready\n");
+	if (urcsts & URCSTS_ER)
 		schedule_work(&whcrc->event_work);
-	}
-	if (urcsts & URCSTS_RCI) {
-		d_printf(3, dev, "RCI: ready to execute another command\n");
+	if (urcsts & URCSTS_RCI)
 		wake_up_all(&whcrc->cmd_wq);
-	}
 out:
 	return IRQ_HANDLED;
 }
@@ -251,8 +222,7 @@
 	whcrc->area = umc_dev->resource.start;
 	whcrc->rc_len = umc_dev->resource.end - umc_dev->resource.start + 1;
 	result = -EBUSY;
-	if (request_mem_region(whcrc->area, whcrc->rc_len, KBUILD_MODNAME)
-	    == NULL) {
+	if (request_mem_region(whcrc->area, whcrc->rc_len, KBUILD_MODNAME) == NULL) {
 		dev_err(dev, "can't request URC region (%zu bytes @ 0x%lx): %d\n",
 			whcrc->rc_len, whcrc->area, result);
 		goto error_request_region;
@@ -287,8 +257,6 @@
 		dev_err(dev, "Can't allocate evt transfer buffer\n");
 		goto error_evt_buffer;
 	}
-	d_printf(3, dev, "UWB RC Interface: %zu bytes at 0x%p, irq %u\n",
-		 whcrc->rc_len, whcrc->rc_base, umc_dev->irq);
 	return 0;
 
 error_evt_buffer:
@@ -333,47 +301,23 @@
 static int whcrc_start_rc(struct uwb_rc *rc)
 {
 	struct whcrc *whcrc = rc->priv;
-	int result = 0;
 	struct device *dev = &whcrc->umc_dev->dev;
-	unsigned long start, duration;
 
 	/* Reset the thing */
 	le_writel(URCCMD_RESET, whcrc->rc_base + URCCMD);
-	if (d_test(3))
-		start = jiffies;
 	if (whci_wait_for(dev, whcrc->rc_base + URCCMD, URCCMD_RESET, 0,
-			  5000, "device to reset at init") < 0) {
-		result = -EBUSY;
-		goto error;
-	} else if (d_test(3)) {
-		duration = jiffies - start;
-		if (duration > msecs_to_jiffies(40))
-			dev_err(dev, "Device took %ums to "
-				     "reset. MAX expected: 40ms\n",
-				     jiffies_to_msecs(duration));
-	}
+			  5000, "hardware reset") < 0)
+		return -EBUSY;
 
 	/* Set the event buffer, start the controller (enable IRQs later) */
 	le_writel(0, whcrc->rc_base + URCINTR);
 	le_writel(URCCMD_RS, whcrc->rc_base + URCCMD);
-	result = -ETIMEDOUT;
-	if (d_test(3))
-		start = jiffies;
 	if (whci_wait_for(dev, whcrc->rc_base + URCSTS, URCSTS_HALTED, 0,
-			  5000, "device to start") < 0)
-		goto error;
-	if (d_test(3)) {
-		duration = jiffies - start;
-		if (duration > msecs_to_jiffies(40))
-			dev_err(dev, "Device took %ums to start. "
-				     "MAX expected: 40ms\n",
-				     jiffies_to_msecs(duration));
-	}
+			  5000, "radio controller start") < 0)
+		return -ETIMEDOUT;
 	whcrc_enable_events(whcrc);
-	result = 0;
 	le_writel(URCINTR_EN_ALL, whcrc->rc_base + URCINTR);
-error:
-	return result;
+	return 0;
 }
 
 
@@ -395,7 +339,7 @@
 
 	le_writel(0, whcrc->rc_base + URCCMD);
 	whci_wait_for(&umc_dev->dev, whcrc->rc_base + URCSTS,
-		      URCSTS_HALTED, 0, 40, "URCSTS.HALTED");
+		      URCSTS_HALTED, URCSTS_HALTED, 100, "radio controller stop");
 }
 
 static void whcrc_init(struct whcrc *whcrc)
@@ -421,7 +365,6 @@
 	struct whcrc *whcrc;
 	struct device *dev = &umc_dev->dev;
 
-	d_fnstart(3, dev, "(umc_dev %p)\n", umc_dev);
 	result = -ENOMEM;
 	uwb_rc = uwb_rc_alloc();
 	if (uwb_rc == NULL) {
@@ -453,7 +396,6 @@
 	if (result < 0)
 		goto error_rc_add;
 	umc_set_drvdata(umc_dev, whcrc);
-	d_fnend(3, dev, "(umc_dev %p) = 0\n", umc_dev);
 	return 0;
 
 error_rc_add:
@@ -463,7 +405,6 @@
 error_alloc:
 	uwb_rc_put(uwb_rc);
 error_rc_alloc:
-	d_fnend(3, dev, "(umc_dev %p) = %d\n", umc_dev, result);
 	return result;
 }
 
@@ -486,7 +427,24 @@
 	whcrc_release_rc_umc(whcrc);
 	kfree(whcrc);
 	uwb_rc_put(uwb_rc);
-	d_printf(1, &umc_dev->dev, "freed whcrc %p\n", whcrc);
+}
+
+static int whcrc_pre_reset(struct umc_dev *umc)
+{
+	struct whcrc *whcrc = umc_get_drvdata(umc);
+	struct uwb_rc *uwb_rc = whcrc->uwb_rc;
+
+	uwb_rc_pre_reset(uwb_rc);
+	return 0;
+}
+
+static int whcrc_post_reset(struct umc_dev *umc)
+{
+	struct whcrc *whcrc = umc_get_drvdata(umc);
+	struct uwb_rc *uwb_rc = whcrc->uwb_rc;
+
+	uwb_rc_post_reset(uwb_rc);
+	return 0;
 }
 
 /* PCI device ID's that we handle [so it gets loaded] */
@@ -497,10 +455,12 @@
 MODULE_DEVICE_TABLE(pci, whcrc_id_table);
 
 static struct umc_driver whcrc_driver = {
-	.name   = "whc-rc",
-	.cap_id = UMC_CAP_ID_WHCI_RC,
-	.probe  = whcrc_probe,
-	.remove = whcrc_remove,
+	.name       = "whc-rc",
+	.cap_id     = UMC_CAP_ID_WHCI_RC,
+	.probe      = whcrc_probe,
+	.remove     = whcrc_remove,
+	.pre_reset  = whcrc_pre_reset,
+	.post_reset = whcrc_post_reset,
 };
 
 static int __init whcrc_driver_init(void)
diff --git a/drivers/uwb/whci.c b/drivers/uwb/whci.c
index 3df2388..1f8964e 100644
--- a/drivers/uwb/whci.c
+++ b/drivers/uwb/whci.c
@@ -67,11 +67,11 @@
 		val = le_readl(reg);
 		if ((val & mask) == result)
 			break;
-		msleep(10);
 		if (t >= max_ms) {
-			dev_err(dev, "timed out waiting for %s ", tag);
+			dev_err(dev, "%s timed out\n", tag);
 			return -ETIMEDOUT;
 		}
+		msleep(10);
 		t += 10;
 	}
 	return 0;
@@ -111,7 +111,7 @@
 		+ UWBCAPDATA_TO_OFFSET(capdata);
 	umc->resource.end    = umc->resource.start
 		+ (n == 0 ? 0x20 : UWBCAPDATA_TO_SIZE(capdata)) - 1;
-	umc->resource.name   = umc->dev.bus_id;
+	umc->resource.name   = dev_name(&umc->dev);
 	umc->resource.flags  = card->pci->resource[bar].flags;
 	umc->resource.parent = &card->pci->resource[bar];
 	umc->irq             = card->pci->irq;
diff --git a/drivers/uwb/wlp/eda.c b/drivers/uwb/wlp/eda.c
index 10985fa..69e0200 100644
--- a/drivers/uwb/wlp/eda.c
+++ b/drivers/uwb/wlp/eda.c
@@ -51,9 +51,7 @@
  * the tag and address of the transmitting neighbor.
  */
 
-#define D_LOCAL 5
 #include <linux/netdevice.h>
-#include <linux/uwb/debug.h>
 #include <linux/etherdevice.h>
 #include <linux/wlp.h>
 #include "wlp-internal.h"
@@ -304,7 +302,6 @@
 {
 	int result = 0;
 	struct wlp *wlp = container_of(eda, struct wlp, eda);
-	struct device *dev = &wlp->rc->uwb_dev.dev;
 	struct wlp_eda_node *itr;
 	unsigned long flags;
 	int found = 0;
@@ -313,26 +310,14 @@
 	list_for_each_entry(itr, &eda->cache, list_node) {
 		if (!memcmp(itr->virt_addr, virt_addr,
 			   sizeof(itr->virt_addr))) {
-			d_printf(6, dev, "EDA: looking for %pM hit %02x:%02x "
-			       "wss %p tag 0x%02x state %u\n",
-			       virt_addr,
-			       itr->dev_addr.data[1],
-			       itr->dev_addr.data[0], itr->wss,
-			       itr->tag, itr->state);
 			result = (*function)(wlp, itr, priv);
 			*dev_addr = itr->dev_addr;
 			found = 1;
 			break;
-		} else
-			d_printf(6, dev, "EDA: looking for %pM against %pM miss\n",
-			         virt_addr, itr->virt_addr);
+		}
 	}
-	if (!found) {
-		if (printk_ratelimit())
-			dev_err(dev, "EDA: Eth addr %pM not found.\n",
-				virt_addr);
+	if (!found)
 		result = -ENODEV;
-	}
 	spin_unlock_irqrestore(&eda->lock, flags);
 	return result;
 }
diff --git a/drivers/uwb/wlp/messages.c b/drivers/uwb/wlp/messages.c
index a64cb82..aa42fce 100644
--- a/drivers/uwb/wlp/messages.c
+++ b/drivers/uwb/wlp/messages.c
@@ -24,8 +24,7 @@
  */
 
 #include <linux/wlp.h>
-#define D_LOCAL 6
-#include <linux/uwb/debug.h>
+
 #include "wlp-internal.h"
 
 static
@@ -105,24 +104,18 @@
 #define wlp_set(type, type_code, name)					\
 static size_t wlp_set_##name(struct wlp_attr_##name *attr, type value)	\
 {									\
-	d_fnstart(6, NULL, "(attribute %p)\n", attr);			\
 	wlp_set_attr_hdr(&attr->hdr, type_code,				\
 			 sizeof(*attr) - sizeof(struct wlp_attr_hdr));	\
 	attr->name = value;						\
-	d_dump(6, NULL, attr, sizeof(*attr));				\
-	d_fnend(6, NULL, "(attribute %p)\n", attr);			\
 	return sizeof(*attr);						\
 }
 
 #define wlp_pset(type, type_code, name)					\
 static size_t wlp_set_##name(struct wlp_attr_##name *attr, type value)	\
 {									\
-	d_fnstart(6, NULL, "(attribute %p)\n", attr);			\
 	wlp_set_attr_hdr(&attr->hdr, type_code,				\
 			 sizeof(*attr) - sizeof(struct wlp_attr_hdr));	\
 	attr->name = *value;						\
-	d_dump(6, NULL, attr, sizeof(*attr));				\
-	d_fnend(6, NULL, "(attribute %p)\n", attr);			\
 	return sizeof(*attr);						\
 }
 
@@ -139,11 +132,8 @@
 static size_t wlp_set_##name(struct wlp_attr_##name *attr, type value,	\
 				size_t len)				\
 {									\
-	d_fnstart(6, NULL, "(attribute %p)\n", attr);			\
 	wlp_set_attr_hdr(&attr->hdr, type_code, len);			\
 	memcpy(attr->name, value, len);					\
-	d_dump(6, NULL, attr, sizeof(*attr) + len);			\
-	d_fnend(6, NULL, "(attribute %p)\n", attr);			\
 	return sizeof(*attr) + len;					\
 }
 
@@ -182,7 +172,7 @@
 	size_t datalen;
 	void *ptr = attr->wss_info;
 	size_t used = sizeof(*attr);
-	d_fnstart(6, NULL, "(attribute %p)\n", attr);
+
 	datalen = sizeof(struct wlp_wss_info) + strlen(wss->name);
 	wlp_set_attr_hdr(&attr->hdr, WLP_ATTR_WSS_INFO, datalen);
 	used = wlp_set_wssid(ptr, &wss->wssid);
@@ -190,9 +180,6 @@
 	used += wlp_set_accept_enrl(ptr + used, wss->accept_enroll);
 	used += wlp_set_wss_sec_status(ptr + used, wss->secure_status);
 	used += wlp_set_wss_bcast(ptr + used, &wss->bcast);
-	d_dump(6, NULL, attr, sizeof(*attr) + datalen);
-	d_fnend(6, NULL, "(attribute %p, used %d)\n",
-		attr, (int)(sizeof(*attr) + used));
 	return sizeof(*attr) + used;
 }
 
@@ -414,7 +401,6 @@
 	size_t used = 0;
 	ssize_t result = -EINVAL;
 
-	d_printf(6, dev, "WLP: WSS info: Retrieving WSS name\n");
 	result = wlp_get_wss_name(wlp, ptr, info->name, buflen);
 	if (result < 0) {
 		dev_err(dev, "WLP: unable to obtain WSS name from "
@@ -422,7 +408,7 @@
 		goto error_parse;
 	}
 	used += result;
-	d_printf(6, dev, "WLP: WSS info: Retrieving accept enroll\n");
+
 	result = wlp_get_accept_enrl(wlp, ptr + used, &info->accept_enroll,
 				     buflen - used);
 	if (result < 0) {
@@ -437,7 +423,7 @@
 		goto error_parse;
 	}
 	used += result;
-	d_printf(6, dev, "WLP: WSS info: Retrieving secure status\n");
+
 	result = wlp_get_wss_sec_status(wlp, ptr + used, &info->sec_status,
 					buflen - used);
 	if (result < 0) {
@@ -452,7 +438,7 @@
 		goto error_parse;
 	}
 	used += result;
-	d_printf(6, dev, "WLP: WSS info: Retrieving broadcast\n");
+
 	result = wlp_get_wss_bcast(wlp, ptr + used, &info->bcast,
 				   buflen - used);
 	if (result < 0) {
@@ -530,7 +516,7 @@
 	len = result;
 	used = sizeof(*attr);
 	ptr = attr;
-	d_printf(6, dev, "WLP: WSS info: Retrieving WSSID\n");
+
 	result = wlp_get_wssid(wlp, ptr + used, wssid, buflen - used);
 	if (result < 0) {
 		dev_err(dev, "WLP: unable to obtain WSSID from WSS info.\n");
@@ -553,8 +539,6 @@
 		goto out;
 	}
 	result = used;
-	d_printf(6, dev, "WLP: Successfully parsed WLP information "
-		 "attribute. used %zu bytes\n", used);
 out:
 	return result;
 }
@@ -598,8 +582,6 @@
 	struct wlp_wssid_e *wssid_e;
 	char buf[WLP_WSS_UUID_STRSIZE];
 
-	d_fnstart(6, dev, "wlp %p, attr %p, neighbor %p, wss %p, buflen %d \n",
-		  wlp, attr, neighbor, wss, (int)buflen);
 	if (buflen < 0)
 		goto out;
 
@@ -638,8 +620,7 @@
 			wss->accept_enroll = wss_info.accept_enroll;
 			wss->state = WLP_WSS_STATE_PART_ENROLLED;
 			wlp_wss_uuid_print(buf, sizeof(buf), &wssid);
-			d_printf(2, dev, "WLP: Found WSS %s. Enrolling.\n",
-				 buf);
+			dev_dbg(dev, "WLP: Found WSS %s. Enrolling.\n", buf);
 		} else {
 			wssid_e = wlp_create_wssid_e(wlp, neighbor);
 			if (wssid_e == NULL) {
@@ -660,9 +641,6 @@
 	if (result < 0 && !enroll) /* this was a discovery */
 		wlp_remove_neighbor_tmp_info(neighbor);
 out:
-	d_fnend(6, dev, "wlp %p, attr %p, neighbor %p, wss %p, buflen %d, "
-		"result %d \n", wlp, attr, neighbor, wss, (int)buflen,
-		(int)result);
 	return result;
 
 }
@@ -718,7 +696,6 @@
 	struct sk_buff *_skb;
 	void *d1_itr;
 
-	d_fnstart(6, dev, "wlp %p\n", wlp);
 	if (wlp->dev_info == NULL) {
 		result = __wlp_setup_device_info(wlp);
 		if (result < 0) {
@@ -728,24 +705,6 @@
 		}
 	}
 	info = wlp->dev_info;
-	d_printf(6, dev, "Local properties:\n"
-		 "Device name (%d bytes): %s\n"
-		 "Model name (%d bytes): %s\n"
-		 "Manufacturer (%d bytes): %s\n"
-		 "Model number (%d bytes): %s\n"
-		 "Serial number (%d bytes): %s\n"
-		 "Primary device type: \n"
-		 " Category: %d \n"
-		 " OUI: %02x:%02x:%02x \n"
-		 " OUI Subdivision: %u \n",
-		 (int)strlen(info->name), info->name,
-		 (int)strlen(info->model_name), info->model_name,
-		 (int)strlen(info->manufacturer), info->manufacturer,
-		 (int)strlen(info->model_nr),  info->model_nr,
-		 (int)strlen(info->serial), info->serial,
-		 info->prim_dev_type.category,
-		 info->prim_dev_type.OUI[0], info->prim_dev_type.OUI[1],
-		 info->prim_dev_type.OUI[2], info->prim_dev_type.OUIsubdiv);
 	_skb = dev_alloc_skb(sizeof(*_d1)
 		      + sizeof(struct wlp_attr_uuid_e)
 		      + sizeof(struct wlp_attr_wss_sel_mthd)
@@ -768,7 +727,6 @@
 		goto error;
 	}
 	_d1 = (void *) _skb->data;
-	d_printf(6, dev, "D1 starts at %p \n", _d1);
 	_d1->hdr.mux_hdr = cpu_to_le16(WLP_PROTOCOL_ID);
 	_d1->hdr.type = WLP_FRAME_ASSOCIATION;
 	_d1->type = WLP_ASSOC_D1;
@@ -791,25 +749,8 @@
 	used += wlp_set_prim_dev_type(d1_itr + used, &info->prim_dev_type);
 	used += wlp_set_wlp_assc_err(d1_itr + used, WLP_ASSOC_ERROR_NONE);
 	skb_put(_skb, sizeof(*_d1) + used);
-	d_printf(6, dev, "D1 message:\n");
-	d_dump(6, dev, _d1, sizeof(*_d1)
-		     + sizeof(struct wlp_attr_uuid_e)
-		     + sizeof(struct wlp_attr_wss_sel_mthd)
-		     + sizeof(struct wlp_attr_dev_name)
-		     + strlen(info->name)
-		     + sizeof(struct wlp_attr_manufacturer)
-		     + strlen(info->manufacturer)
-		     + sizeof(struct wlp_attr_model_name)
-		     + strlen(info->model_name)
-		     + sizeof(struct wlp_attr_model_nr)
-		     + strlen(info->model_nr)
-		     + sizeof(struct wlp_attr_serial)
-		     + strlen(info->serial)
-		     + sizeof(struct wlp_attr_prim_dev_type)
-		     + sizeof(struct wlp_attr_wlp_assc_err));
 	*skb = _skb;
 error:
-	d_fnend(6, dev, "wlp %p, result = %d\n", wlp, result);
 	return result;
 }
 
@@ -837,7 +778,6 @@
 	void *d2_itr;
 	size_t mem_needed;
 
-	d_fnstart(6, dev, "wlp %p\n", wlp);
 	if (wlp->dev_info == NULL) {
 		result = __wlp_setup_device_info(wlp);
 		if (result < 0) {
@@ -847,24 +787,6 @@
 		}
 	}
 	info = wlp->dev_info;
-	d_printf(6, dev, "Local properties:\n"
-		 "Device name (%d bytes): %s\n"
-		 "Model name (%d bytes): %s\n"
-		 "Manufacturer (%d bytes): %s\n"
-		 "Model number (%d bytes): %s\n"
-		 "Serial number (%d bytes): %s\n"
-		 "Primary device type: \n"
-		 " Category: %d \n"
-		 " OUI: %02x:%02x:%02x \n"
-		 " OUI Subdivision: %u \n",
-		 (int)strlen(info->name), info->name,
-		 (int)strlen(info->model_name), info->model_name,
-		 (int)strlen(info->manufacturer), info->manufacturer,
-		 (int)strlen(info->model_nr),  info->model_nr,
-		 (int)strlen(info->serial), info->serial,
-		 info->prim_dev_type.category,
-		 info->prim_dev_type.OUI[0], info->prim_dev_type.OUI[1],
-		 info->prim_dev_type.OUI[2], info->prim_dev_type.OUIsubdiv);
 	mem_needed = sizeof(*_d2)
 		      + sizeof(struct wlp_attr_uuid_e)
 		      + sizeof(struct wlp_attr_uuid_r)
@@ -892,7 +814,6 @@
 		goto error;
 	}
 	_d2 = (void *) _skb->data;
-	d_printf(6, dev, "D2 starts at %p \n", _d2);
 	_d2->hdr.mux_hdr = cpu_to_le16(WLP_PROTOCOL_ID);
 	_d2->hdr.type = WLP_FRAME_ASSOCIATION;
 	_d2->type = WLP_ASSOC_D2;
@@ -917,11 +838,8 @@
 	used += wlp_set_prim_dev_type(d2_itr + used, &info->prim_dev_type);
 	used += wlp_set_wlp_assc_err(d2_itr + used, WLP_ASSOC_ERROR_NONE);
 	skb_put(_skb, sizeof(*_d2) + used);
-	d_printf(6, dev, "D2 message:\n");
-	d_dump(6, dev, _d2, mem_needed);
 	*skb = _skb;
 error:
-	d_fnend(6, dev, "wlp %p, result = %d\n", wlp, result);
 	return result;
 }
 
@@ -947,7 +865,6 @@
 	struct sk_buff *_skb;
 	struct wlp_nonce tmp;
 
-	d_fnstart(6, dev, "wlp %p\n", wlp);
 	_skb = dev_alloc_skb(sizeof(*f0));
 	if (_skb == NULL) {
 		dev_err(dev, "WLP: Unable to allocate memory for F0 "
@@ -955,7 +872,6 @@
 		goto error_alloc;
 	}
 	f0 = (void *) _skb->data;
-	d_printf(6, dev, "F0 starts at %p \n", f0);
 	f0->f0_hdr.hdr.mux_hdr = cpu_to_le16(WLP_PROTOCOL_ID);
 	f0->f0_hdr.hdr.type = WLP_FRAME_ASSOCIATION;
 	f0->f0_hdr.type = WLP_ASSOC_F0;
@@ -969,7 +885,6 @@
 	*skb = _skb;
 	result = 0;
 error_alloc:
-	d_fnend(6, dev, "wlp %p, result %d \n", wlp, result);
 	return result;
 }
 
@@ -1242,12 +1157,9 @@
 	enum wlp_wss_sel_mthd sel_mthd = 0;
 	struct wlp_device_info dev_info;
 	enum wlp_assc_error assc_err;
-	char uuid[WLP_WSS_UUID_STRSIZE];
 	struct sk_buff *resp = NULL;
 
 	/* Parse D1 frame */
-	d_fnstart(6, dev, "WLP: handle D1 frame. wlp = %p, skb = %p\n",
-		  wlp, skb);
 	mutex_lock(&wss->mutex);
 	mutex_lock(&wlp->mutex); /* to access wlp->uuid */
 	memset(&dev_info, 0, sizeof(dev_info));
@@ -1258,30 +1170,6 @@
 		kfree_skb(skb);
 		goto out;
 	}
-	wlp_wss_uuid_print(uuid, sizeof(uuid), &uuid_e);
-	d_printf(6, dev, "From D1 frame:\n"
-		 "UUID-E: %s\n"
-		 "Selection method: %d\n"
-		 "Device name (%d bytes): %s\n"
-		 "Model name (%d bytes): %s\n"
-		 "Manufacturer (%d bytes): %s\n"
-		 "Model number (%d bytes): %s\n"
-		 "Serial number (%d bytes): %s\n"
-		 "Primary device type: \n"
-		 " Category: %d \n"
-		 " OUI: %02x:%02x:%02x \n"
-		 " OUI Subdivision: %u \n",
-		 uuid, sel_mthd,
-		 (int)strlen(dev_info.name), dev_info.name,
-		 (int)strlen(dev_info.model_name), dev_info.model_name,
-		 (int)strlen(dev_info.manufacturer), dev_info.manufacturer,
-		 (int)strlen(dev_info.model_nr),  dev_info.model_nr,
-		 (int)strlen(dev_info.serial), dev_info.serial,
-		 dev_info.prim_dev_type.category,
-		 dev_info.prim_dev_type.OUI[0],
-		 dev_info.prim_dev_type.OUI[1],
-		 dev_info.prim_dev_type.OUI[2],
-		 dev_info.prim_dev_type.OUIsubdiv);
 
 	kfree_skb(skb);
 	if (!wlp_uuid_is_set(&wlp->uuid)) {
@@ -1316,7 +1204,6 @@
 	kfree(frame_ctx);
 	mutex_unlock(&wlp->mutex);
 	mutex_unlock(&wss->mutex);
-	d_fnend(6, dev, "WLP: handle D1 frame. wlp = %p\n", wlp);
 }
 
 /**
@@ -1546,10 +1433,8 @@
 	void *ptr = skb->data;
 	size_t len = skb->len;
 	size_t used;
-	char buf[WLP_WSS_UUID_STRSIZE];
 	struct wlp_frame_assoc *assoc = ptr;
 
-	d_fnstart(6, dev, "wlp %p, skb %p \n", wlp, skb);
 	used = sizeof(*assoc);
 	result = wlp_get_wssid(wlp, ptr + used, wssid, len - used);
 	if (result < 0) {
@@ -1572,14 +1457,7 @@
 			wlp_assoc_frame_str(assoc->type));
 		goto error_parse;
 	}
-	wlp_wss_uuid_print(buf, sizeof(buf), wssid);
-	d_printf(6, dev, "WLP: parsed: WSSID %s, tag 0x%02x, virt "
-		 "%02x:%02x:%02x:%02x:%02x:%02x \n", buf, *tag,
-		 virt_addr->data[0], virt_addr->data[1], virt_addr->data[2],
-		 virt_addr->data[3], virt_addr->data[4], virt_addr->data[5]);
-
 error_parse:
-	d_fnend(6, dev, "wlp %p, skb %p, result = %d \n", wlp, skb, result);
 	return result;
 }
 
@@ -1600,7 +1478,6 @@
 	} *c;
 	struct sk_buff *_skb;
 
-	d_fnstart(6, dev, "wlp %p, wss %p \n", wlp, wss);
 	_skb = dev_alloc_skb(sizeof(*c));
 	if (_skb == NULL) {
 		dev_err(dev, "WLP: Unable to allocate memory for C1/C2 "
@@ -1608,7 +1485,6 @@
 		goto error_alloc;
 	}
 	c = (void *) _skb->data;
-	d_printf(6, dev, "C1/C2 starts at %p \n", c);
 	c->c_hdr.hdr.mux_hdr = cpu_to_le16(WLP_PROTOCOL_ID);
 	c->c_hdr.hdr.type = WLP_FRAME_ASSOCIATION;
 	c->c_hdr.type = type;
@@ -1616,12 +1492,9 @@
 	wlp_set_msg_type(&c->c_hdr.msg_type, type);
 	wlp_set_wssid(&c->wssid, &wss->wssid);
 	skb_put(_skb, sizeof(*c));
-	d_printf(6, dev, "C1/C2 message:\n");
-	d_dump(6, dev, c, sizeof(*c));
 	*skb = _skb;
 	result = 0;
 error_alloc:
-	d_fnend(6, dev, "wlp %p, wss %p, result %d \n", wlp, wss, result);
 	return result;
 }
 
@@ -1660,7 +1533,6 @@
 	} *c;
 	struct sk_buff *_skb;
 
-	d_fnstart(6, dev, "wlp %p, wss %p \n", wlp, wss);
 	_skb = dev_alloc_skb(sizeof(*c));
 	if (_skb == NULL) {
 		dev_err(dev, "WLP: Unable to allocate memory for C3/C4 "
@@ -1668,7 +1540,6 @@
 		goto error_alloc;
 	}
 	c = (void *) _skb->data;
-	d_printf(6, dev, "C3/C4 starts at %p \n", c);
 	c->c_hdr.hdr.mux_hdr = cpu_to_le16(WLP_PROTOCOL_ID);
 	c->c_hdr.hdr.type = WLP_FRAME_ASSOCIATION;
 	c->c_hdr.type = type;
@@ -1678,12 +1549,9 @@
 	wlp_set_wss_tag(&c->wss_tag, wss->tag);
 	wlp_set_wss_virt(&c->wss_virt, &wss->virtual_addr);
 	skb_put(_skb, sizeof(*c));
-	d_printf(6, dev, "C3/C4 message:\n");
-	d_dump(6, dev, c, sizeof(*c));
 	*skb = _skb;
 	result = 0;
 error_alloc:
-	d_fnend(6, dev, "wlp %p, wss %p, result %d \n", wlp, wss, result);
 	return result;
 }
 
@@ -1709,10 +1577,7 @@
 	struct device *dev = &wlp->rc->uwb_dev.dev;			\
 	int result;							\
 	struct sk_buff *skb = NULL;					\
-	d_fnstart(6, dev, "wlp %p, wss %p, neighbor: %02x:%02x\n",	\
-		  wlp, wss, dev_addr->data[1], dev_addr->data[0]);	\
-	d_printf(6, dev, "WLP: Constructing %s frame. \n",		\
-		 wlp_assoc_frame_str(id));				\
+									\
 	/* Build the frame */						\
 	result = wlp_build_assoc_##type(wlp, wss, &skb);		\
 	if (result < 0) {						\
@@ -1721,9 +1586,6 @@
 		goto error_build_assoc;					\
 	}								\
 	/* Send the frame */						\
-	d_printf(6, dev, "Transmitting %s frame to %02x:%02x \n",	\
-		 wlp_assoc_frame_str(id),				\
-		 dev_addr->data[1], dev_addr->data[0]);			\
 	BUG_ON(wlp->xmit_frame == NULL);				\
 	result = wlp->xmit_frame(wlp, skb, dev_addr);			\
 	if (result < 0) {						\
@@ -1740,8 +1602,6 @@
 	/* We could try again ... */					\
 	dev_kfree_skb_any(skb);/*we need to free if tx fails*/		\
 error_build_assoc:							\
-	d_fnend(6, dev, "wlp %p, wss %p, neighbor: %02x:%02x\n",	\
-		wlp, wss, dev_addr->data[1], dev_addr->data[0]);	\
 	return result;							\
 }
 
@@ -1794,12 +1654,9 @@
 	struct uwb_dev_addr *src = &frame_ctx->src;
 	int result;
 	struct wlp_uuid wssid;
-	char buf[WLP_WSS_UUID_STRSIZE];
 	struct sk_buff *resp = NULL;
 
 	/* Parse C1 frame */
-	d_fnstart(6, dev, "WLP: handle C1 frame. wlp = %p, c1 = %p\n",
-		  wlp, c1);
 	mutex_lock(&wss->mutex);
 	result = wlp_get_wssid(wlp, (void *)c1 + sizeof(*c1), &wssid,
 			       len - sizeof(*c1));
@@ -1807,12 +1664,8 @@
 		dev_err(dev, "WLP: unable to obtain WSSID from C1 frame.\n");
 		goto out;
 	}
-	wlp_wss_uuid_print(buf, sizeof(buf), &wssid);
-	d_printf(6, dev, "Received C1 frame with WSSID %s \n", buf);
 	if (!memcmp(&wssid, &wss->wssid, sizeof(wssid))
 	    && wss->state == WLP_WSS_STATE_ACTIVE) {
-		d_printf(6, dev, "WSSID from C1 frame is known locally "
-			 "and is active\n");
 		/* Construct C2 frame */
 		result = wlp_build_assoc_c2(wlp, wss, &resp);
 		if (result < 0) {
@@ -1820,8 +1673,6 @@
 			goto out;
 		}
 	} else {
-		d_printf(6, dev, "WSSID from C1 frame is not known locally "
-			 "or is not active\n");
 		/* Construct F0 frame */
 		result = wlp_build_assoc_f0(wlp, &resp, WLP_ASSOC_ERROR_INV);
 		if (result < 0) {
@@ -1830,8 +1681,6 @@
 		}
 	}
 	/* Send C2 frame */
-	d_printf(6, dev, "Transmitting response (C2/F0) frame to %02x:%02x \n",
-		 src->data[1], src->data[0]);
 	BUG_ON(wlp->xmit_frame == NULL);
 	result = wlp->xmit_frame(wlp, resp, src);
 	if (result < 0) {
@@ -1846,7 +1695,6 @@
 	kfree_skb(frame_ctx->skb);
 	kfree(frame_ctx);
 	mutex_unlock(&wss->mutex);
-	d_fnend(6, dev, "WLP: handle C1 frame. wlp = %p\n", wlp);
 }
 
 /**
@@ -1868,27 +1716,20 @@
 	struct sk_buff *skb = frame_ctx->skb;
 	struct uwb_dev_addr *src = &frame_ctx->src;
 	int result;
-	char buf[WLP_WSS_UUID_STRSIZE];
 	struct sk_buff *resp = NULL;
 	struct wlp_uuid wssid;
 	u8 tag;
 	struct uwb_mac_addr virt_addr;
 
 	/* Parse C3 frame */
-	d_fnstart(6, dev, "WLP: handle C3 frame. wlp = %p, skb = %p\n",
-		  wlp, skb);
 	mutex_lock(&wss->mutex);
 	result = wlp_parse_c3c4_frame(wlp, skb, &wssid, &tag, &virt_addr);
 	if (result < 0) {
 		dev_err(dev, "WLP: unable to obtain values from C3 frame.\n");
 		goto out;
 	}
-	wlp_wss_uuid_print(buf, sizeof(buf), &wssid);
-	d_printf(6, dev, "Received C3 frame with WSSID %s \n", buf);
 	if (!memcmp(&wssid, &wss->wssid, sizeof(wssid))
 	    && wss->state >= WLP_WSS_STATE_ACTIVE) {
-		d_printf(6, dev, "WSSID from C3 frame is known locally "
-			 "and is active\n");
 		result = wlp_eda_update_node(&wlp->eda, src, wss,
 					     (void *) virt_addr.data, tag,
 					     WLP_WSS_CONNECTED);
@@ -1913,8 +1754,6 @@
 			}
 		}
 	} else {
-		d_printf(6, dev, "WSSID from C3 frame is not known locally "
-			 "or is not active\n");
 		/* Construct F0 frame */
 		result = wlp_build_assoc_f0(wlp, &resp, WLP_ASSOC_ERROR_INV);
 		if (result < 0) {
@@ -1923,8 +1762,6 @@
 		}
 	}
 	/* Send C4 frame */
-	d_printf(6, dev, "Transmitting response (C4/F0) frame to %02x:%02x \n",
-		 src->data[1], src->data[0]);
 	BUG_ON(wlp->xmit_frame == NULL);
 	result = wlp->xmit_frame(wlp, resp, src);
 	if (result < 0) {
@@ -1939,8 +1776,6 @@
 	kfree_skb(frame_ctx->skb);
 	kfree(frame_ctx);
 	mutex_unlock(&wss->mutex);
-	d_fnend(6, dev, "WLP: handle C3 frame. wlp = %p, skb = %p\n",
-		wlp, skb);
 }
 
 
diff --git a/drivers/uwb/wlp/sysfs.c b/drivers/uwb/wlp/sysfs.c
index 1bb9b1f..0370399 100644
--- a/drivers/uwb/wlp/sysfs.c
+++ b/drivers/uwb/wlp/sysfs.c
@@ -23,8 +23,8 @@
  * FIXME: Docs
  *
  */
-
 #include <linux/wlp.h>
+
 #include "wlp-internal.h"
 
 static
diff --git a/drivers/uwb/wlp/txrx.c b/drivers/uwb/wlp/txrx.c
index c701bd1..cd20357 100644
--- a/drivers/uwb/wlp/txrx.c
+++ b/drivers/uwb/wlp/txrx.c
@@ -26,12 +26,10 @@
 
 #include <linux/etherdevice.h>
 #include <linux/wlp.h>
-#define D_LOCAL 5
-#include <linux/uwb/debug.h>
+
 #include "wlp-internal.h"
 
-
-/**
+/*
  * Direct incoming association msg to correct parsing routine
  *
  * We only expect D1, E1, C1, C3 messages as new. All other incoming
@@ -48,35 +46,31 @@
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	struct wlp_frame_assoc *assoc = (void *) skb->data;
 	struct wlp_assoc_frame_ctx *frame_ctx;
-	d_fnstart(5, dev, "wlp %p, skb %p\n", wlp, skb);
+
 	frame_ctx = kmalloc(sizeof(*frame_ctx), GFP_ATOMIC);
 	if (frame_ctx == NULL) {
 		dev_err(dev, "WLP: Unable to allocate memory for association "
 			"frame handling.\n");
 		kfree_skb(skb);
-		goto out;
+		return;
 	}
 	frame_ctx->wlp = wlp;
 	frame_ctx->skb = skb;
 	frame_ctx->src = *src;
 	switch (assoc->type) {
 	case WLP_ASSOC_D1:
-		d_printf(5, dev, "Received a D1 frame.\n");
 		INIT_WORK(&frame_ctx->ws, wlp_handle_d1_frame);
 		schedule_work(&frame_ctx->ws);
 		break;
 	case WLP_ASSOC_E1:
-		d_printf(5, dev, "Received a E1 frame. FIXME?\n");
 		kfree_skb(skb); /* Temporary until we handle it */
 		kfree(frame_ctx); /* Temporary until we handle it */
 		break;
 	case WLP_ASSOC_C1:
-		d_printf(5, dev, "Received a C1 frame.\n");
 		INIT_WORK(&frame_ctx->ws, wlp_handle_c1_frame);
 		schedule_work(&frame_ctx->ws);
 		break;
 	case WLP_ASSOC_C3:
-		d_printf(5, dev, "Received a C3 frame.\n");
 		INIT_WORK(&frame_ctx->ws, wlp_handle_c3_frame);
 		schedule_work(&frame_ctx->ws);
 		break;
@@ -87,11 +81,9 @@
 		kfree(frame_ctx);
 		break;
 	}
-out:
-	d_fnend(5, dev, "wlp %p\n", wlp);
 }
 
-/**
+/*
  * Process incoming association frame
  *
  * Although it could be possible to deal with some incoming association
@@ -112,7 +104,6 @@
 	struct wlp_frame_assoc *assoc = (void *) skb->data;
 	struct wlp_session *session = wlp->session;
 	u8 version;
-	d_fnstart(5, dev, "wlp %p, skb %p\n", wlp, skb);
 
 	if (wlp_get_version(wlp, &assoc->version, &version,
 			    sizeof(assoc->version)) < 0)
@@ -150,14 +141,12 @@
 	} else {
 		wlp_direct_assoc_frame(wlp, skb, src);
 	}
-	d_fnend(5, dev, "wlp %p\n", wlp);
 	return;
 error:
 	kfree_skb(skb);
-	d_fnend(5, dev, "wlp %p\n", wlp);
 }
 
-/**
+/*
  * Verify incoming frame is from connected neighbor, prep to pass to WLP client
  *
  * Verification proceeds according to WLP 0.99 [7.3.1]. The source address
@@ -176,7 +165,6 @@
 	struct wlp_eda_node eda_entry;
 	struct wlp_frame_std_abbrv_hdr *hdr = (void *) skb->data;
 
-	d_fnstart(6, dev, "wlp %p, skb %p \n", wlp, skb);
 	/*verify*/
 	result = wlp_copy_eda_node(&wlp->eda, src, &eda_entry);
 	if (result < 0) {
@@ -207,11 +195,10 @@
 	/*prep*/
 	skb_pull(skb, sizeof(*hdr));
 out:
-	d_fnend(6, dev, "wlp %p, skb %p, result = %d \n", wlp, skb, result);
 	return result;
 }
 
-/**
+/*
  * Receive a WLP frame from device
  *
  * @returns: 1 if calling function should free the skb
@@ -226,14 +213,12 @@
 	struct wlp_frame_hdr *hdr;
 	int result = 0;
 
-	d_fnstart(6, dev, "skb (%p), len (%u)\n", skb, len);
 	if (len < sizeof(*hdr)) {
 		dev_err(dev, "Not enough data to parse WLP header.\n");
 		result = -EINVAL;
 		goto out;
 	}
 	hdr = ptr;
-	d_dump(6, dev, hdr, sizeof(*hdr));
 	if (le16_to_cpu(hdr->mux_hdr) != WLP_PROTOCOL_ID) {
 		dev_err(dev, "Not a WLP frame type.\n");
 		result = -EINVAL;
@@ -270,7 +255,6 @@
 				"WLP header.\n");
 			goto out;
 		}
-		d_printf(5, dev, "Association frame received.\n");
 		wlp_receive_assoc_frame(wlp, skb, src);
 		break;
 	default:
@@ -283,13 +267,12 @@
 		kfree_skb(skb);
 		result = 0;
 	}
-	d_fnend(6, dev, "skb (%p)\n", skb);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wlp_receive_frame);
 
 
-/**
+/*
  * Verify frame from network stack, prepare for further transmission
  *
  * @skb:   the socket buffer that needs to be prepared for transmission (it
@@ -343,9 +326,7 @@
 	int result = -EINVAL;
 	struct ethhdr *eth_hdr = (void *) skb->data;
 
-	d_fnstart(6, dev, "wlp (%p), skb (%p) \n", wlp, skb);
 	if (is_broadcast_ether_addr(eth_hdr->h_dest)) {
-		d_printf(6, dev, "WLP: handling broadcast frame. \n");
 		result = wlp_eda_for_each(&wlp->eda, wlp_wss_send_copy, skb);
 		if (result < 0) {
 			if (printk_ratelimit())
@@ -357,7 +338,6 @@
 		result = 1;
 		/* Frame will be transmitted by WLP. */
 	} else {
-		d_printf(6, dev, "WLP: handling unicast frame. \n");
 		result = wlp_eda_for_virtual(&wlp->eda, eth_hdr->h_dest, dst,
 					     wlp_wss_prep_hdr, skb);
 		if (unlikely(result < 0)) {
@@ -368,7 +348,6 @@
 		}
 	}
 out:
-	d_fnend(6, dev, "wlp (%p), skb (%p). result = %d \n", wlp, skb, result);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wlp_prepare_tx_frame);
diff --git a/drivers/uwb/wlp/wlp-internal.h b/drivers/uwb/wlp/wlp-internal.h
index 1c94fab..3e8d5de 100644
--- a/drivers/uwb/wlp/wlp-internal.h
+++ b/drivers/uwb/wlp/wlp-internal.h
@@ -42,10 +42,6 @@
 extern struct kobj_type wss_ktype;
 extern struct attribute_group wss_attr_group;
 
-extern int uwb_rc_ie_add(struct uwb_rc *, const struct uwb_ie_hdr *, size_t);
-extern int uwb_rc_ie_rm(struct uwb_rc *, enum uwb_ie);
-
-
 /* This should be changed to a dynamic array where entries are sorted
  * by eth_addr and search is done in a binary form
  *
diff --git a/drivers/uwb/wlp/wlp-lc.c b/drivers/uwb/wlp/wlp-lc.c
index 0799402e..13db739 100644
--- a/drivers/uwb/wlp/wlp-lc.c
+++ b/drivers/uwb/wlp/wlp-lc.c
@@ -21,12 +21,9 @@
  *
  * FIXME: docs
  */
-
 #include <linux/wlp.h>
-#define D_LOCAL 6
-#include <linux/uwb/debug.h>
-#include "wlp-internal.h"
 
+#include "wlp-internal.h"
 
 static
 void wlp_neighbor_init(struct wlp_neighbor_e *neighbor)
@@ -61,11 +58,6 @@
 static
 void __wlp_fill_device_info(struct wlp *wlp)
 {
-	struct device *dev = &wlp->rc->uwb_dev.dev;
-
-	BUG_ON(wlp->fill_device_info == NULL);
-	d_printf(6, dev, "Retrieving device information "
-			 "from device driver.\n");
 	wlp->fill_device_info(wlp, wlp->dev_info);
 }
 
@@ -127,7 +119,7 @@
 	}
 }
 
-/**
+/*
  * Populate WLP neighborhood cache with neighbor information
  *
  * A new neighbor is found. If it is discoverable then we add it to the
@@ -141,10 +133,7 @@
 	int discoverable;
 	struct wlp_neighbor_e *neighbor;
 
-	d_fnstart(6, &dev->dev, "uwb %p \n", dev);
-	d_printf(6, &dev->dev, "Found neighbor device %02x:%02x \n",
-		 dev->dev_addr.data[1], dev->dev_addr.data[0]);
-	/**
+	/*
 	 * FIXME:
 	 * Use contents of WLP IE found in beacon cache to determine if
 	 * neighbor is discoverable.
@@ -167,7 +156,6 @@
 		list_add(&neighbor->node, &wlp->neighbors);
 	}
 error_no_mem:
-	d_fnend(6, &dev->dev, "uwb %p, result = %d \n", dev, result);
 	return result;
 }
 
@@ -255,8 +243,6 @@
 		dev_err(dev, "Unable to send D1 frame to neighbor "
 			"%02x:%02x (%d)\n", dev_addr->data[1],
 			dev_addr->data[0], result);
-		d_printf(6, dev, "Add placeholders into buffer next to "
-			 "neighbor information we have (dev address).\n");
 		goto out;
 	}
 	/* Create session, wait for response */
@@ -284,8 +270,6 @@
 	/* Parse message in session->data: it will be either D2 or F0 */
 	skb = session.data;
 	resp = (void *) skb->data;
-	d_printf(6, dev, "Received response to D1 frame. \n");
-	d_dump(6, dev, skb->data, skb->len > 72 ? 72 : skb->len);
 
 	if (resp->type == WLP_ASSOC_F0) {
 		result = wlp_parse_f0(wlp, skb);
@@ -337,10 +321,9 @@
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	char buf[WLP_WSS_UUID_STRSIZE];
 	struct uwb_dev_addr *dev_addr = &neighbor->uwb_dev->dev_addr;
+
 	wlp_wss_uuid_print(buf, sizeof(buf), wssid);
-	d_fnstart(6, dev, "wlp %p, neighbor %p, wss %p, wssid %p (%s)\n",
-		  wlp, neighbor, wss, wssid, buf);
-	d_printf(6, dev, "Complete me.\n");
+
 	result =  wlp_d1d2_exchange(wlp, neighbor, wss, wssid);
 	if (result < 0) {
 		dev_err(dev, "WLP: D1/D2 message exchange for enrollment "
@@ -360,13 +343,10 @@
 		goto error;
 	} else {
 		wss->state = WLP_WSS_STATE_ENROLLED;
-		d_printf(2, dev, "WLP: Success Enrollment into unsecure WSS "
-			 "%s using neighbor %02x:%02x. \n", buf,
-			 dev_addr->data[1], dev_addr->data[0]);
+		dev_dbg(dev, "WLP: Success Enrollment into unsecure WSS "
+			"%s using neighbor %02x:%02x. \n",
+			buf, dev_addr->data[1], dev_addr->data[0]);
 	}
-
-	d_fnend(6, dev, "wlp %p, neighbor %p, wss %p, wssid %p (%s)\n",
-		  wlp, neighbor, wss, wssid, buf);
 out:
 	return result;
 error:
@@ -449,7 +429,6 @@
 	int result = 0;
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 
-	d_fnstart(6, dev, "wlp %p \n", wlp);
 	mutex_lock(&wlp->nbmutex);
 	/* Clear current neighborhood cache. */
 	__wlp_neighbors_release(wlp);
@@ -469,7 +448,6 @@
 	}
 error_dev_for_each:
 	mutex_unlock(&wlp->nbmutex);
-	d_fnend(6, dev, "wlp %p \n", wlp);
 	return result;
 }
 
@@ -492,9 +470,6 @@
 	int result;
 	switch (event) {
 	case UWB_NOTIF_ONAIR:
-		d_printf(6, dev, "UWB device %02x:%02x is onair\n",
-				uwb_dev->dev_addr.data[1],
-				uwb_dev->dev_addr.data[0]);
 		result = wlp_eda_create_node(&wlp->eda,
 					     uwb_dev->mac_addr.data,
 					     &uwb_dev->dev_addr);
@@ -505,18 +480,11 @@
 				uwb_dev->dev_addr.data[0]);
 		break;
 	case UWB_NOTIF_OFFAIR:
-		d_printf(6, dev, "UWB device %02x:%02x is offair\n",
-				uwb_dev->dev_addr.data[1],
-				uwb_dev->dev_addr.data[0]);
 		wlp_eda_rm_node(&wlp->eda, &uwb_dev->dev_addr);
 		mutex_lock(&wlp->nbmutex);
-		list_for_each_entry_safe(neighbor, next, &wlp->neighbors,
-					 node) {
-			if (neighbor->uwb_dev == uwb_dev) {
-				d_printf(6, dev, "Removing device from "
-					 "neighborhood.\n");
+		list_for_each_entry_safe(neighbor, next, &wlp->neighbors, node) {
+			if (neighbor->uwb_dev == uwb_dev)
 				__wlp_neighbor_release(neighbor);
-			}
 		}
 		mutex_unlock(&wlp->nbmutex);
 		break;
@@ -526,38 +494,47 @@
 	}
 }
 
-int wlp_setup(struct wlp *wlp, struct uwb_rc *rc)
+static void wlp_channel_changed(struct uwb_pal *pal, int channel)
 {
-	struct device *dev = &rc->uwb_dev.dev;
+	struct wlp *wlp = container_of(pal, struct wlp, pal);
+
+	if (channel < 0)
+		netif_carrier_off(wlp->ndev);
+	else
+		netif_carrier_on(wlp->ndev);
+}
+
+int wlp_setup(struct wlp *wlp, struct uwb_rc *rc, struct net_device *ndev)
+{
 	int result;
 
-	d_fnstart(6, dev, "wlp %p\n", wlp);
 	BUG_ON(wlp->fill_device_info == NULL);
 	BUG_ON(wlp->xmit_frame == NULL);
 	BUG_ON(wlp->stop_queue == NULL);
 	BUG_ON(wlp->start_queue == NULL);
+
 	wlp->rc = rc;
+	wlp->ndev = ndev;
 	wlp_eda_init(&wlp->eda);/* Set up address cache */
 	wlp->uwb_notifs_handler.cb = wlp_uwb_notifs_cb;
 	wlp->uwb_notifs_handler.data = wlp;
 	uwb_notifs_register(rc, &wlp->uwb_notifs_handler);
 
 	uwb_pal_init(&wlp->pal);
-	result = uwb_pal_register(rc, &wlp->pal);
+	wlp->pal.rc = rc;
+	wlp->pal.channel_changed = wlp_channel_changed;
+	result = uwb_pal_register(&wlp->pal);
 	if (result < 0)
 		uwb_notifs_deregister(wlp->rc, &wlp->uwb_notifs_handler);
 
-	d_fnend(6, dev, "wlp %p, result = %d\n", wlp, result);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wlp_setup);
 
 void wlp_remove(struct wlp *wlp)
 {
-	struct device *dev = &wlp->rc->uwb_dev.dev;
-	d_fnstart(6, dev, "wlp %p\n", wlp);
 	wlp_neighbors_release(wlp);
-	uwb_pal_unregister(wlp->rc, &wlp->pal);
+	uwb_pal_unregister(&wlp->pal);
 	uwb_notifs_deregister(wlp->rc, &wlp->uwb_notifs_handler);
 	wlp_eda_release(&wlp->eda);
 	mutex_lock(&wlp->mutex);
@@ -565,9 +542,6 @@
 		kfree(wlp->dev_info);
 	mutex_unlock(&wlp->mutex);
 	wlp->rc = NULL;
-	/* We have to use NULL here because this function can be called
-	 * when the device disappeared. */
-	d_fnend(6, NULL, "wlp %p\n", wlp);
 }
 EXPORT_SYMBOL_GPL(wlp_remove);
 
diff --git a/drivers/uwb/wlp/wss-lc.c b/drivers/uwb/wlp/wss-lc.c
index 96b18c9..5913c7a 100644
--- a/drivers/uwb/wlp/wss-lc.c
+++ b/drivers/uwb/wlp/wss-lc.c
@@ -43,14 +43,11 @@
  * 	wlp_wss_release()
  * 		wlp_wss_reset()
  */
-
 #include <linux/etherdevice.h> /* for is_valid_ether_addr */
 #include <linux/skbuff.h>
 #include <linux/wlp.h>
-#define D_LOCAL 5
-#include <linux/uwb/debug.h>
-#include "wlp-internal.h"
 
+#include "wlp-internal.h"
 
 size_t wlp_wss_key_print(char *buf, size_t bufsize, u8 *key)
 {
@@ -116,9 +113,6 @@
  */
 void wlp_wss_reset(struct wlp_wss *wss)
 {
-	struct wlp *wlp = container_of(wss, struct wlp, wss);
-	struct device *dev = &wlp->rc->uwb_dev.dev;
-	d_fnstart(5, dev, "wss (%p) \n", wss);
 	memset(&wss->wssid, 0, sizeof(wss->wssid));
 	wss->hash = 0;
 	memset(&wss->name[0], 0, sizeof(wss->name));
@@ -127,7 +121,6 @@
 	memset(&wss->master_key[0], 0, sizeof(wss->master_key));
 	wss->tag = 0;
 	wss->state = WLP_WSS_STATE_NONE;
-	d_fnend(5, dev, "wss (%p) \n", wss);
 }
 
 /**
@@ -145,7 +138,6 @@
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	int result;
 
-	d_fnstart(5, dev, "wss (%p), wssid: %s\n", wss, wssid_str);
 	result = kobject_set_name(&wss->kobj, "wss-%s", wssid_str);
 	if (result < 0)
 		return result;
@@ -162,7 +154,6 @@
 			result);
 		goto error_sysfs_create_group;
 	}
-	d_fnend(5, dev, "Completed. result = %d \n", result);
 	return 0;
 error_sysfs_create_group:
 
@@ -214,22 +205,14 @@
 	struct wlp *wlp = container_of(wss, struct wlp, wss);
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	struct wlp_neighbor_e *neighbor;
-	char buf[WLP_WSS_UUID_STRSIZE];
 	int result = -ENXIO;
 	struct uwb_dev_addr *dev_addr;
 
-	wlp_wss_uuid_print(buf, sizeof(buf), wssid);
-	d_fnstart(5, dev, "wss %p, wssid %s, registrar %02x:%02x \n",
-		  wss, buf, dest->data[1], dest->data[0]);
 	mutex_lock(&wlp->nbmutex);
 	list_for_each_entry(neighbor, &wlp->neighbors, node) {
 		dev_addr = &neighbor->uwb_dev->dev_addr;
 		if (!memcmp(dest, dev_addr, sizeof(*dest))) {
-			d_printf(5, dev, "Neighbor %02x:%02x is valid, "
-				 "enrolling. \n",
-				 dev_addr->data[1], dev_addr->data[0]);
-			result = wlp_enroll_neighbor(wlp, neighbor, wss,
-						     wssid);
+			result = wlp_enroll_neighbor(wlp, neighbor, wss, wssid);
 			break;
 		}
 	}
@@ -237,8 +220,6 @@
 		dev_err(dev, "WLP: Cannot find neighbor %02x:%02x. \n",
 			dest->data[1], dest->data[0]);
 	mutex_unlock(&wlp->nbmutex);
-	d_fnend(5, dev, "wss %p, wssid %s, registrar %02x:%02x, result %d \n",
-		  wss, buf, dest->data[1], dest->data[0], result);
 	return result;
 }
 
@@ -260,16 +241,11 @@
 	char buf[WLP_WSS_UUID_STRSIZE];
 	int result = -ENXIO;
 
-	wlp_wss_uuid_print(buf, sizeof(buf), wssid);
-	d_fnstart(5, dev, "wss %p, wssid %s \n", wss, buf);
+
 	mutex_lock(&wlp->nbmutex);
 	list_for_each_entry(neighbor, &wlp->neighbors, node) {
 		list_for_each_entry(wssid_e, &neighbor->wssid, node) {
 			if (!memcmp(wssid, &wssid_e->wssid, sizeof(*wssid))) {
-				d_printf(5, dev, "Found WSSID %s in neighbor "
-					 "%02x:%02x cache. \n", buf,
-					 neighbor->uwb_dev->dev_addr.data[1],
-					 neighbor->uwb_dev->dev_addr.data[0]);
 				result = wlp_enroll_neighbor(wlp, neighbor,
 							     wss, wssid);
 				if (result == 0) /* enrollment success */
@@ -279,10 +255,11 @@
 		}
 	}
 out:
-	if (result == -ENXIO)
+	if (result == -ENXIO) {
+		wlp_wss_uuid_print(buf, sizeof(buf), wssid);
 		dev_err(dev, "WLP: Cannot find WSSID %s in cache. \n", buf);
+	}
 	mutex_unlock(&wlp->nbmutex);
-	d_fnend(5, dev, "wss %p, wssid %s, result %d \n", wss, buf, result);
 	return result;
 }
 
@@ -307,27 +284,22 @@
 	struct uwb_dev_addr bcast = {.data = {0xff, 0xff} };
 
 	wlp_wss_uuid_print(buf, sizeof(buf), wssid);
+
 	if (wss->state != WLP_WSS_STATE_NONE) {
 		dev_err(dev, "WLP: Already enrolled in WSS %s.\n", buf);
 		result = -EEXIST;
 		goto error;
 	}
-	if (!memcmp(&bcast, devaddr, sizeof(bcast))) {
-		d_printf(5, dev, "Request to enroll in discovered WSS "
-			 "with WSSID %s \n", buf);
+	if (!memcmp(&bcast, devaddr, sizeof(bcast)))
 		result = wlp_wss_enroll_discovered(wss, wssid);
-	} else {
-		d_printf(5, dev, "Request to enroll in WSSID %s with "
-			 "registrar %02x:%02x\n", buf, devaddr->data[1],
-			 devaddr->data[0]);
+	else
 		result = wlp_wss_enroll_target(wss, wssid, devaddr);
-	}
 	if (result < 0) {
 		dev_err(dev, "WLP: Unable to enroll into WSS %s, result %d \n",
 			buf, result);
 		goto error;
 	}
-	d_printf(2, dev, "Successfully enrolled into WSS %s \n", buf);
+	dev_dbg(dev, "Successfully enrolled into WSS %s \n", buf);
 	result = wlp_wss_sysfs_add(wss, buf);
 	if (result < 0) {
 		dev_err(dev, "WLP: Unable to set up sysfs for WSS kobject.\n");
@@ -363,7 +335,6 @@
 		u8 hash; /* only include one hash */
 	} ie_data;
 
-	d_fnstart(5, dev, "Activating WSS %p. \n", wss);
 	BUG_ON(wss->state != WLP_WSS_STATE_ENROLLED);
 	wss->hash = wlp_wss_comp_wssid_hash(&wss->wssid);
 	wss->tag = wss->hash;
@@ -382,7 +353,6 @@
 	wss->state = WLP_WSS_STATE_ACTIVE;
 	result = 0;
 error_wlp_ie:
-	d_fnend(5, dev, "Activating WSS %p, result = %d \n", wss, result);
 	return result;
 }
 
@@ -405,7 +375,6 @@
 	int result = 0;
 	char buf[WLP_WSS_UUID_STRSIZE];
 
-	d_fnstart(5, dev, "Enrollment and activation requested. \n");
 	mutex_lock(&wss->mutex);
 	result = wlp_wss_enroll(wss, wssid, devaddr);
 	if (result < 0) {
@@ -424,7 +393,6 @@
 error_activate:
 error_enroll:
 	mutex_unlock(&wss->mutex);
-	d_fnend(5, dev, "Completed. result = %d \n", result);
 	return result;
 }
 
@@ -447,11 +415,9 @@
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	int result = 0;
 	char buf[WLP_WSS_UUID_STRSIZE];
-	d_fnstart(5, dev, "Request to create new WSS.\n");
+
 	result = wlp_wss_uuid_print(buf, sizeof(buf), wssid);
-	d_printf(5, dev, "Request to create WSS: WSSID=%s, name=%s, "
-		 "sec_status=%u, accepting enrollment=%u \n",
-		 buf, name, sec_status, accept);
+
 	if (!mutex_trylock(&wss->mutex)) {
 		dev_err(dev, "WLP: WLP association session in progress.\n");
 		return -EBUSY;
@@ -498,7 +464,6 @@
 	result = 0;
 out:
 	mutex_unlock(&wss->mutex);
-	d_fnend(5, dev, "Completed. result = %d \n", result);
 	return result;
 }
 
@@ -520,16 +485,12 @@
 {
 	int result = 0;
 	struct device *dev = &wlp->rc->uwb_dev.dev;
-	char buf[WLP_WSS_UUID_STRSIZE];
 	DECLARE_COMPLETION_ONSTACK(completion);
 	struct wlp_session session;
 	struct sk_buff  *skb;
 	struct wlp_frame_assoc *resp;
 	struct wlp_uuid wssid;
 
-	wlp_wss_uuid_print(buf, sizeof(buf), &wss->wssid);
-	d_fnstart(5, dev, "wlp %p, wss %p (wssid %s), neighbor %02x:%02x \n",
-		  wlp, wss, buf, dev_addr->data[1], dev_addr->data[0]);
 	mutex_lock(&wlp->mutex);
 	/* Send C1 association frame */
 	result = wlp_send_assoc_frame(wlp, wss, dev_addr, WLP_ASSOC_C1);
@@ -565,8 +526,6 @@
 	/* Parse message in session->data: it will be either C2 or F0 */
 	skb = session.data;
 	resp = (void *) skb->data;
-	d_printf(5, dev, "Received response to C1 frame. \n");
-	d_dump(5, dev, skb->data, skb->len > 72 ? 72 : skb->len);
 	if (resp->type == WLP_ASSOC_F0) {
 		result = wlp_parse_f0(wlp, skb);
 		if (result < 0)
@@ -584,11 +543,9 @@
 		result = 0;
 		goto error_resp_parse;
 	}
-	if (!memcmp(&wssid, &wss->wssid, sizeof(wssid))) {
-		d_printf(5, dev, "WSSID in C2 frame matches local "
-			 "active WSS.\n");
+	if (!memcmp(&wssid, &wss->wssid, sizeof(wssid)))
 		result = 1;
-	} else {
+	else {
 		dev_err(dev, "WLP: Received a C2 frame without matching "
 			"WSSID.\n");
 		result = 0;
@@ -598,8 +555,6 @@
 out:
 	wlp->session = NULL;
 	mutex_unlock(&wlp->mutex);
-	d_fnend(5, dev, "wlp %p, wss %p (wssid %s), neighbor %02x:%02x \n",
-		  wlp, wss, buf, dev_addr->data[1], dev_addr->data[0]);
 	return result;
 }
 
@@ -620,16 +575,8 @@
 {
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	int result = 0;
-	char buf[WLP_WSS_UUID_STRSIZE];
-	wlp_wss_uuid_print(buf, sizeof(buf), wssid);
-	d_fnstart(5, dev, "wlp %p, wss %p, wssid %s, tag %u, virtual "
-		  "%02x:%02x:%02x:%02x:%02x:%02x \n", wlp, wss, buf, *tag,
-		  virt_addr->data[0], virt_addr->data[1], virt_addr->data[2],
-		  virt_addr->data[3], virt_addr->data[4], virt_addr->data[5]);
 
 	if (!memcmp(wssid, &wss->wssid, sizeof(*wssid))) {
-		d_printf(5, dev, "WSSID from neighbor frame matches local "
-			 "active WSS.\n");
 		/* Update EDA cache */
 		result = wlp_eda_update_node(&wlp->eda, dev_addr, wss,
 					     (void *) virt_addr->data, *tag,
@@ -638,18 +585,9 @@
 			dev_err(dev, "WLP: Unable to update EDA cache "
 				"with new connected neighbor information.\n");
 	} else {
-		dev_err(dev, "WLP: Neighbor does not have matching "
-			"WSSID.\n");
+		dev_err(dev, "WLP: Neighbor does not have matching WSSID.\n");
 		result = -EINVAL;
 	}
-
-	d_fnend(5, dev, "wlp %p, wss %p, wssid %s, tag %u, virtual "
-		  "%02x:%02x:%02x:%02x:%02x:%02x, result = %d \n",
-		  wlp, wss, buf, *tag,
-		  virt_addr->data[0], virt_addr->data[1], virt_addr->data[2],
-		  virt_addr->data[3], virt_addr->data[4], virt_addr->data[5],
-		  result);
-
 	return result;
 }
 
@@ -665,7 +603,6 @@
 {
 	int result;
 	struct device *dev = &wlp->rc->uwb_dev.dev;
-	char buf[WLP_WSS_UUID_STRSIZE];
 	struct wlp_uuid wssid;
 	u8 tag;
 	struct uwb_mac_addr virt_addr;
@@ -674,9 +611,6 @@
 	struct wlp_frame_assoc *resp;
 	struct sk_buff *skb;
 
-	wlp_wss_uuid_print(buf, sizeof(buf), &wss->wssid);
-	d_fnstart(5, dev, "wlp %p, wss %p (wssid %s), neighbor %02x:%02x \n",
-		  wlp, wss, buf, dev_addr->data[1], dev_addr->data[0]);
 	mutex_lock(&wlp->mutex);
 	/* Send C3 association frame */
 	result = wlp_send_assoc_frame(wlp, wss, dev_addr, WLP_ASSOC_C3);
@@ -711,8 +645,6 @@
 	/* Parse message in session->data: it will be either C4 or F0 */
 	skb = session.data;
 	resp = (void *) skb->data;
-	d_printf(5, dev, "Received response to C3 frame. \n");
-	d_dump(5, dev, skb->data, skb->len > 72 ? 72 : skb->len);
 	if (resp->type == WLP_ASSOC_F0) {
 		result = wlp_parse_f0(wlp, skb);
 		if (result < 0)
@@ -744,8 +676,6 @@
 					  WLP_WSS_CONNECT_FAILED);
 	wlp->session = NULL;
 	mutex_unlock(&wlp->mutex);
-	d_fnend(5, dev, "wlp %p, wss %p (wssid %s), neighbor %02x:%02x \n",
-		  wlp, wss, buf, dev_addr->data[1], dev_addr->data[0]);
 	return result;
 }
 
@@ -780,12 +710,8 @@
 	struct wlp_wss *wss = &wlp->wss;
 	int result;
 	struct device *dev = &wlp->rc->uwb_dev.dev;
-	char buf[WLP_WSS_UUID_STRSIZE];
 
 	mutex_lock(&wss->mutex);
-	wlp_wss_uuid_print(buf, sizeof(buf), &wss->wssid);
-	d_fnstart(5, dev, "wlp %p, wss %p (wssid %s), neighbor %02x:%02x \n",
-		  wlp, wss, buf, dev_addr->data[1], dev_addr->data[0]);
 	if (wss->state < WLP_WSS_STATE_ACTIVE) {
 		if (printk_ratelimit())
 			dev_err(dev, "WLP: Attempting to connect with "
@@ -836,7 +762,6 @@
 	BUG_ON(wlp->start_queue == NULL);
 	wlp->start_queue(wlp);
 	mutex_unlock(&wss->mutex);
-	d_fnend(5, dev, "wlp %p, wss %p (wssid %s)\n", wlp, wss, buf);
 }
 
 /**
@@ -855,7 +780,6 @@
 	struct sk_buff *skb = _skb;
 	struct wlp_frame_std_abbrv_hdr *std_hdr;
 
-	d_fnstart(6, dev, "wlp %p \n", wlp);
 	if (eda_entry->state == WLP_WSS_CONNECTED) {
 		/* Add WLP header */
 		BUG_ON(skb_headroom(skb) < sizeof(*std_hdr));
@@ -873,7 +797,6 @@
 				dev_addr->data[0]);
 		result = -EINVAL;
 	}
-	d_fnend(6, dev, "wlp %p \n", wlp);
 	return result;
 }
 
@@ -893,16 +816,9 @@
 {
 	int result = 0;
 	struct device *dev = &wlp->rc->uwb_dev.dev;
-	struct uwb_dev_addr *dev_addr = &eda_entry->dev_addr;
-	unsigned char *eth_addr = eda_entry->eth_addr;
 	struct sk_buff *skb = _skb;
 	struct wlp_assoc_conn_ctx *conn_ctx;
 
-	d_fnstart(5, dev, "wlp %p\n", wlp);
-	d_printf(5, dev, "To neighbor %02x:%02x with eth "
-		  "%02x:%02x:%02x:%02x:%02x:%02x\n", dev_addr->data[1],
-		  dev_addr->data[0], eth_addr[0], eth_addr[1], eth_addr[2],
-		  eth_addr[3], eth_addr[4], eth_addr[5]);
 	if (eda_entry->state == WLP_WSS_UNCONNECTED) {
 		/* We don't want any more packets while we set up connection */
 		BUG_ON(wlp->stop_queue == NULL);
@@ -929,12 +845,9 @@
 			 "previously. Not retrying. \n");
 		result = -ENONET;
 		goto out;
-	} else { /* eda_entry->state == WLP_WSS_CONNECTED */
-		d_printf(5, dev, "Neighbor is connected, preparing frame.\n");
+	} else /* eda_entry->state == WLP_WSS_CONNECTED */
 		result = wlp_wss_prep_hdr(wlp, eda_entry, skb);
-	}
 out:
-	d_fnend(5, dev, "wlp %p, result = %d \n", wlp, result);
 	return result;
 }
 
@@ -957,8 +870,6 @@
 	struct sk_buff *copy;
 	struct uwb_dev_addr *dev_addr = &eda_entry->dev_addr;
 
-	d_fnstart(5, dev, "to neighbor %02x:%02x, skb (%p) \n",
-		  dev_addr->data[1], dev_addr->data[0], skb);
 	copy = skb_copy(skb, GFP_ATOMIC);
 	if (copy == NULL) {
 		if (printk_ratelimit())
@@ -988,8 +899,6 @@
 		dev_kfree_skb_irq(copy);/*we need to free if tx fails */
 	}
 out:
-	d_fnend(5, dev, "to neighbor %02x:%02x \n", dev_addr->data[1],
-		  dev_addr->data[0]);
 	return result;
 }
 
@@ -1005,7 +914,7 @@
 	struct wlp *wlp = container_of(wss, struct wlp, wss);
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	int result = 0;
-	d_fnstart(5, dev, "wss (%p) \n", wss);
+
 	mutex_lock(&wss->mutex);
 	wss->kobj.parent = &net_dev->dev.kobj;
 	if (!is_valid_ether_addr(net_dev->dev_addr)) {
@@ -1018,7 +927,6 @@
 	       sizeof(wss->virtual_addr.data));
 out:
 	mutex_unlock(&wss->mutex);
-	d_fnend(5, dev, "wss (%p) \n", wss);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wlp_wss_setup);
@@ -1035,8 +943,7 @@
 void wlp_wss_remove(struct wlp_wss *wss)
 {
 	struct wlp *wlp = container_of(wss, struct wlp, wss);
-	struct device *dev = &wlp->rc->uwb_dev.dev;
-	d_fnstart(5, dev, "wss (%p) \n", wss);
+
 	mutex_lock(&wss->mutex);
 	if (wss->state == WLP_WSS_STATE_ACTIVE)
 		uwb_rc_ie_rm(wlp->rc, UWB_IE_WLP);
@@ -1050,6 +957,5 @@
 	wlp_eda_release(&wlp->eda);
 	wlp_eda_init(&wlp->eda);
 	mutex_unlock(&wss->mutex);
-	d_fnend(5, dev, "wss (%p) \n", wss);
 }
 EXPORT_SYMBOL_GPL(wlp_wss_remove);
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 4fd3fa5..ec68c74 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -55,6 +55,13 @@
 	  To compile this driver as a module, choose M here: the
 	  module will be called softdog.
 
+config WM8350_WATCHDOG
+	tristate "WM8350 watchdog"
+	depends on MFD_WM8350
+	help
+	  Support for the watchdog in the WM8350 AudioPlus PMIC.  When
+	  the watchdog triggers the system will be reset.
+
 # ALPHA Architecture
 
 # ARM Architecture
@@ -551,6 +558,18 @@
 	  To compile this driver as a module, choose M here: the
 	  module will be called cpu5wdt.
 
+config SMSC_SCH311X_WDT
+	tristate "SMSC SCH311X Watchdog Timer"
+	depends on X86
+	---help---
+	  This is the driver for the hardware watchdog timer on the
+	  SMSC SCH3112, SCH3114 and SCH3116 Super IO chipset
+	  (LPC IO with 8042 KBC, Reset Generation, HWM and multiple
+	  serial ports).
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called sch311x_wdt.
+
 config SMSC37B787_WDT
 	tristate "Winbond SMsC37B787 Watchdog Timer"
 	depends on X86
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index e352bbb..c19b866 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -83,6 +83,7 @@
 obj-$(CONFIG_SBC8360_WDT) += sbc8360.o
 obj-$(CONFIG_SBC7240_WDT) += sbc7240_wdt.o
 obj-$(CONFIG_CPU5_WDT) += cpu5wdt.o
+obj-$(CONFIG_SMSC_SCH311X_WDT) += sch311x_wdt.o
 obj-$(CONFIG_SMSC37B787_WDT) += smsc37b787_wdt.o
 obj-$(CONFIG_W83627HF_WDT) += w83627hf_wdt.o
 obj-$(CONFIG_W83697HF_WDT) += w83697hf_wdt.o
@@ -133,4 +134,5 @@
 # XTENSA Architecture
 
 # Architecture Independant
+obj-$(CONFIG_WM8350_WATCHDOG) += wm8350_wdt.o
 obj-$(CONFIG_SOFT_WATCHDOG) += softdog.o
diff --git a/drivers/watchdog/ib700wdt.c b/drivers/watchdog/ib700wdt.c
index 317ef2b..4bef3dd 100644
--- a/drivers/watchdog/ib700wdt.c
+++ b/drivers/watchdog/ib700wdt.c
@@ -91,32 +91,16 @@
  *
  */
 
-static int wd_times[] = {
-	30,	/* 0x0 */
-	28,	/* 0x1 */
-	26,	/* 0x2 */
-	24,	/* 0x3 */
-	22,	/* 0x4 */
-	20,	/* 0x5 */
-	18,	/* 0x6 */
-	16,	/* 0x7 */
-	14,	/* 0x8 */
-	12,	/* 0x9 */
-	10,	/* 0xA */
-	8,	/* 0xB */
-	6,	/* 0xC */
-	4,	/* 0xD */
-	2,	/* 0xE */
-	0,	/* 0xF */
-};
-
 #define WDT_STOP 0x441
 #define WDT_START 0x443
 
 /* Default timeout */
-#define WD_TIMO 0		/* 30 seconds +/- 20%, from table */
-
-static int wd_margin = WD_TIMO;
+#define WATCHDOG_TIMEOUT 30		/* 30 seconds +/- 20% */
+static int timeout = WATCHDOG_TIMEOUT;	/* in seconds */
+module_param(timeout, int, 0);
+MODULE_PARM_DESC(timeout,
+	"Watchdog timeout in seconds. 0<= timeout <=30, default="
+		__MODULE_STRING(WATCHDOG_TIMEOUT) ".");
 
 static int nowayout = WATCHDOG_NOWAYOUT;
 module_param(nowayout, int, 0);
@@ -131,6 +115,8 @@
 
 static void ibwdt_ping(void)
 {
+	int wd_margin = 15 - ((timeout + 1) / 2);
+
 	spin_lock(&ibwdt_lock);
 
 	/* Write a watchdog value */
@@ -148,15 +134,10 @@
 
 static int ibwdt_set_heartbeat(int t)
 {
-	int i;
-
-	if ((t < 0) || (t > 30))
+	if (t < 0 || t > 30)
 		return -EINVAL;
 
-	for (i = 0x0F; i > -1; i--)
-		if (wd_times[i] >= t)
-			break;
-	wd_margin = i;
+	timeout = t;
 	return 0;
 }
 
@@ -240,7 +221,7 @@
 		/* Fall */
 
 	case WDIOC_GETTIMEOUT:
-		return put_user(wd_times[wd_margin], p);
+		return put_user(timeout, p);
 
 	default:
 		return -ENOTTY;
@@ -317,6 +298,14 @@
 		goto out_nostartreg;
 	}
 
+	/* Check that the heartbeat value is within it's range ;
+	 * if not reset to the default */
+	if (ibwdt_set_heartbeat(timeout)) {
+		ibwdt_set_heartbeat(WATCHDOG_TIMEOUT);
+		printk(KERN_INFO PFX
+			"timeout value must be 0<=x<=30, using %d\n", timeout);
+	}
+
 	res = misc_register(&ibwdt_miscdev);
 	if (res) {
 		printk(KERN_ERR PFX "failed to register misc device\n");
diff --git a/drivers/watchdog/sch311x_wdt.c b/drivers/watchdog/sch311x_wdt.c
new file mode 100644
index 0000000..569eb29
--- /dev/null
+++ b/drivers/watchdog/sch311x_wdt.c
@@ -0,0 +1,578 @@
+/*
+ *	sch311x_wdt.c - Driver for the SCH311x Super-I/O chips
+ *			integrated watchdog.
+ *
+ *	(c) Copyright 2008 Wim Van Sebroeck <wim@iguana.be>.
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ *	Neither Wim Van Sebroeck nor Iguana vzw. admit liability nor
+ *	provide warranty for any of this software. This material is
+ *	provided "AS-IS" and at no charge.
+ */
+
+/*
+ *	Includes, defines, variables, module parameters, ...
+ */
+
+/* Includes */
+#include <linux/module.h>		/* For module specific items */
+#include <linux/moduleparam.h>		/* For new moduleparam's */
+#include <linux/types.h>		/* For standard types (like size_t) */
+#include <linux/errno.h>		/* For the -ENODEV/... values */
+#include <linux/kernel.h>		/* For printk/... */
+#include <linux/miscdevice.h>		/* For MODULE_ALIAS_MISCDEV
+							(WATCHDOG_MINOR) */
+#include <linux/watchdog.h>		/* For the watchdog specific items */
+#include <linux/init.h>			/* For __init/__exit/... */
+#include <linux/fs.h>			/* For file operations */
+#include <linux/platform_device.h>	/* For platform_driver framework */
+#include <linux/ioport.h>		/* For io-port access */
+#include <linux/spinlock.h>		/* For spin_lock/spin_unlock/... */
+#include <linux/uaccess.h>		/* For copy_to_user/put_user/... */
+#include <linux/io.h>			/* For inb/outb/... */
+
+/* Module and version information */
+#define DRV_NAME	"sch311x_wdt"
+#define PFX		DRV_NAME ": "
+
+/* Runtime registers */
+#define RESGEN			0x1d
+#define GP60			0x47
+#define WDT_TIME_OUT		0x65
+#define WDT_VAL			0x66
+#define WDT_CFG			0x67
+#define WDT_CTRL		0x68
+
+/* internal variables */
+static unsigned long sch311x_wdt_is_open;
+static char sch311x_wdt_expect_close;
+static struct platform_device *sch311x_wdt_pdev;
+
+static int sch311x_ioports[] = { 0x2e, 0x4e, 0x162e, 0x164e, 0x00 };
+
+static struct {	/* The devices private data */
+	/* the Runtime Register base address */
+	unsigned short runtime_reg;
+	/* The card's boot status */
+	int boot_status;
+	/* the lock for io operations */
+	spinlock_t io_lock;
+} sch311x_wdt_data;
+
+/* Module load parameters */
+static unsigned short force_id;
+module_param(force_id, ushort, 0);
+MODULE_PARM_DESC(force_id, "Override the detected device ID");
+
+static unsigned short therm_trip;
+module_param(therm_trip, ushort, 0);
+MODULE_PARM_DESC(therm_trip, "Should a ThermTrip trigger the reset generator");
+
+#define WATCHDOG_TIMEOUT 60		/* 60 sec default timeout */
+static int timeout = WATCHDOG_TIMEOUT;	/* in seconds */
+module_param(timeout, int, 0);
+MODULE_PARM_DESC(timeout,
+	"Watchdog timeout in seconds. 1<= timeout <=15300, default="
+		__MODULE_STRING(WATCHDOG_TIMEOUT) ".");
+
+static int nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout,
+	"Watchdog cannot be stopped once started (default="
+		__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+/*
+ *	Super-IO functions
+ */
+
+static inline void sch311x_sio_enter(int sio_config_port)
+{
+	outb(0x55, sio_config_port);
+}
+
+static inline void sch311x_sio_exit(int sio_config_port)
+{
+	outb(0xaa, sio_config_port);
+}
+
+static inline int sch311x_sio_inb(int sio_config_port, int reg)
+{
+	outb(reg, sio_config_port);
+	return inb(sio_config_port + 1);
+}
+
+static inline void sch311x_sio_outb(int sio_config_port, int reg, int val)
+{
+	outb(reg, sio_config_port);
+	outb(val, sio_config_port + 1);
+}
+
+/*
+ *	Watchdog Operations
+ */
+
+static void sch311x_wdt_set_timeout(int t)
+{
+	unsigned char timeout_unit = 0x80;
+
+	/* When new timeout is bigger then 255 seconds, we will use minutes */
+	if (t > 255) {
+		timeout_unit = 0;
+		t /= 60;
+	}
+
+	/* -- Watchdog Timeout --
+	 * Bit 0-6 (Reserved)
+	 * Bit 7   WDT Time-out Value Units Select
+	 *         (0 = Minutes, 1 = Seconds)
+	 */
+	outb(timeout_unit, sch311x_wdt_data.runtime_reg + WDT_TIME_OUT);
+
+	/* -- Watchdog Timer Time-out Value --
+	 * Bit 0-7 Binary coded units (0=Disabled, 1..255)
+	 */
+	outb(t, sch311x_wdt_data.runtime_reg + WDT_VAL);
+}
+
+static void sch311x_wdt_start(void)
+{
+	spin_lock(&sch311x_wdt_data.io_lock);
+
+	/* set watchdog's timeout */
+	sch311x_wdt_set_timeout(timeout);
+	/* enable the watchdog */
+	/* -- General Purpose I/O Bit 6.0 --
+	 * Bit 0,   In/Out: 0 = Output, 1 = Input
+	 * Bit 1,   Polarity: 0 = No Invert, 1 = Invert
+	 * Bit 2-3, Function select: 00 = GPI/O, 01 = LED1, 11 = WDT,
+	 *                           10 = Either Edge Triggered Intr.4
+	 * Bit 4-6  (Reserved)
+	 * Bit 7,   Output Type: 0 = Push Pull Bit, 1 = Open Drain
+	 */
+	outb(0x0e, sch311x_wdt_data.runtime_reg + GP60);
+
+	spin_unlock(&sch311x_wdt_data.io_lock);
+
+}
+
+static void sch311x_wdt_stop(void)
+{
+	spin_lock(&sch311x_wdt_data.io_lock);
+
+	/* stop the watchdog */
+	outb(0x01, sch311x_wdt_data.runtime_reg + GP60);
+	/* disable timeout by setting it to 0 */
+	sch311x_wdt_set_timeout(0);
+
+	spin_unlock(&sch311x_wdt_data.io_lock);
+}
+
+static void sch311x_wdt_keepalive(void)
+{
+	spin_lock(&sch311x_wdt_data.io_lock);
+	sch311x_wdt_set_timeout(timeout);
+	spin_unlock(&sch311x_wdt_data.io_lock);
+}
+
+static int sch311x_wdt_set_heartbeat(int t)
+{
+	if (t < 1 || t > (255*60))
+		return -EINVAL;
+
+	/* When new timeout is bigger then 255 seconds,
+	 * we will round up to minutes (with a max of 255) */
+	if (t > 255)
+		t = (((t - 1) / 60) + 1) * 60;
+
+	timeout = t;
+	return 0;
+}
+
+static void sch311x_wdt_get_status(int *status)
+{
+	unsigned char new_status;
+
+	*status = 0;
+
+	spin_lock(&sch311x_wdt_data.io_lock);
+
+	/* -- Watchdog timer control --
+	 * Bit 0   Status Bit: 0 = Timer counting, 1 = Timeout occured
+	 * Bit 1   Reserved
+	 * Bit 2   Force Timeout: 1 = Forces WD timeout event (self-cleaning)
+	 * Bit 3   P20 Force Timeout enabled:
+	 *          0 = P20 activity does not generate the WD timeout event
+	 *          1 = P20 Allows rising edge of P20, from the keyboard
+	 *              controller, to force the WD timeout event.
+	 * Bit 4-7 Reserved
+	 */
+	new_status = inb(sch311x_wdt_data.runtime_reg + WDT_CTRL);
+	if (new_status & 0x01)
+		*status |= WDIOF_CARDRESET;
+
+	spin_unlock(&sch311x_wdt_data.io_lock);
+}
+
+/*
+ *	/dev/watchdog handling
+ */
+
+static ssize_t sch311x_wdt_write(struct file *file, const char __user *buf,
+						size_t count, loff_t *ppos)
+{
+	if (count) {
+		if (!nowayout) {
+			size_t i;
+
+			sch311x_wdt_expect_close = 0;
+
+			for (i = 0; i != count; i++) {
+				char c;
+				if (get_user(c, buf + i))
+					return -EFAULT;
+				if (c == 'V')
+					sch311x_wdt_expect_close = 42;
+			}
+		}
+		sch311x_wdt_keepalive();
+	}
+	return count;
+}
+
+static long sch311x_wdt_ioctl(struct file *file, unsigned int cmd,
+							unsigned long arg)
+{
+	int status;
+	int new_timeout;
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	static struct watchdog_info ident = {
+		.options		= WDIOF_KEEPALIVEPING |
+					  WDIOF_SETTIMEOUT |
+					  WDIOF_MAGICCLOSE,
+		.firmware_version	= 1,
+		.identity		= DRV_NAME,
+	};
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		if (copy_to_user(argp, &ident, sizeof(ident)))
+			return -EFAULT;
+		break;
+
+	case WDIOC_GETSTATUS:
+	{
+		sch311x_wdt_get_status(&status);
+		return put_user(status, p);
+	}
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(sch311x_wdt_data.boot_status, p);
+
+	case WDIOC_SETOPTIONS:
+	{
+		int options, retval = -EINVAL;
+
+		if (get_user(options, p))
+			return -EFAULT;
+		if (options & WDIOS_DISABLECARD) {
+			sch311x_wdt_stop();
+			retval = 0;
+		}
+		if (options & WDIOS_ENABLECARD) {
+			sch311x_wdt_start();
+			retval = 0;
+		}
+		return retval;
+	}
+	case WDIOC_KEEPALIVE:
+		sch311x_wdt_keepalive();
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		if (get_user(new_timeout, p))
+			return -EFAULT;
+		if (sch311x_wdt_set_heartbeat(new_timeout))
+			return -EINVAL;
+		sch311x_wdt_keepalive();
+		/* Fall */
+	case WDIOC_GETTIMEOUT:
+		return put_user(timeout, p);
+	default:
+		return -ENOTTY;
+	}
+	return 0;
+}
+
+static int sch311x_wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(0, &sch311x_wdt_is_open))
+		return -EBUSY;
+	/*
+	 *	Activate
+	 */
+	sch311x_wdt_start();
+	return nonseekable_open(inode, file);
+}
+
+static int sch311x_wdt_close(struct inode *inode, struct file *file)
+{
+	if (sch311x_wdt_expect_close == 42) {
+		sch311x_wdt_stop();
+	} else {
+		printk(KERN_CRIT PFX
+				"Unexpected close, not stopping watchdog!\n");
+		sch311x_wdt_keepalive();
+	}
+	clear_bit(0, &sch311x_wdt_is_open);
+	sch311x_wdt_expect_close = 0;
+	return 0;
+}
+
+/*
+ *	Kernel Interfaces
+ */
+
+static const struct file_operations sch311x_wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.write		= sch311x_wdt_write,
+	.unlocked_ioctl	= sch311x_wdt_ioctl,
+	.open		= sch311x_wdt_open,
+	.release	= sch311x_wdt_close,
+};
+
+static struct miscdevice sch311x_wdt_miscdev = {
+	.minor	= WATCHDOG_MINOR,
+	.name	= "watchdog",
+	.fops	= &sch311x_wdt_fops,
+};
+
+/*
+ *	Init & exit routines
+ */
+
+static int __devinit sch311x_wdt_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	unsigned char val;
+	int err;
+
+	spin_lock_init(&sch311x_wdt_data.io_lock);
+
+	if (!request_region(sch311x_wdt_data.runtime_reg + RESGEN, 1,
+								DRV_NAME)) {
+		dev_err(dev, "Failed to request region 0x%04x-0x%04x.\n",
+			sch311x_wdt_data.runtime_reg + RESGEN,
+			sch311x_wdt_data.runtime_reg + RESGEN);
+		err = -EBUSY;
+		goto exit;
+	}
+
+	if (!request_region(sch311x_wdt_data.runtime_reg + GP60, 1, DRV_NAME)) {
+		dev_err(dev, "Failed to request region 0x%04x-0x%04x.\n",
+			sch311x_wdt_data.runtime_reg + GP60,
+			sch311x_wdt_data.runtime_reg + GP60);
+		err = -EBUSY;
+		goto exit_release_region;
+	}
+
+	if (!request_region(sch311x_wdt_data.runtime_reg + WDT_TIME_OUT, 4,
+								DRV_NAME)) {
+		dev_err(dev, "Failed to request region 0x%04x-0x%04x.\n",
+			sch311x_wdt_data.runtime_reg + WDT_TIME_OUT,
+			sch311x_wdt_data.runtime_reg + WDT_CTRL);
+		err = -EBUSY;
+		goto exit_release_region2;
+	}
+
+	/* Make sure that the watchdog is not running */
+	sch311x_wdt_stop();
+
+	/* Disable keyboard and mouse interaction and interrupt */
+	/* -- Watchdog timer configuration --
+	 * Bit 0   Reserved
+	 * Bit 1   Keyboard enable: 0* = No Reset, 1 = Reset WDT upon KBD Intr.
+	 * Bit 2   Mouse enable: 0* = No Reset, 1 = Reset WDT upon Mouse Intr
+	 * Bit 3   Reserved
+	 * Bit 4-7 WDT Interrupt Mapping: (0000* = Disabled,
+	 *            0001=IRQ1, 0010=(Invalid), 0011=IRQ3 to 1111=IRQ15)
+	 */
+	outb(0, sch311x_wdt_data.runtime_reg + WDT_CFG);
+
+	/* Check that the heartbeat value is within it's range ;
+	 * if not reset to the default */
+	if (sch311x_wdt_set_heartbeat(timeout)) {
+		sch311x_wdt_set_heartbeat(WATCHDOG_TIMEOUT);
+		dev_info(dev, "timeout value must be 1<=x<=15300, using %d\n",
+			timeout);
+	}
+
+	/* Get status at boot */
+	sch311x_wdt_get_status(&sch311x_wdt_data.boot_status);
+
+	/* enable watchdog */
+	/* -- Reset Generator --
+	 * Bit 0   Enable Watchdog Timer Generation: 0* = Enabled, 1 = Disabled
+	 * Bit 1   Thermtrip Source Select: O* = No Source, 1 = Source
+	 * Bit 2   WDT2_CTL: WDT input bit
+	 * Bit 3-7 Reserved
+	 */
+	outb(0, sch311x_wdt_data.runtime_reg + RESGEN);
+	val = therm_trip ? 0x06 : 0x04;
+	outb(val, sch311x_wdt_data.runtime_reg + RESGEN);
+
+	err = misc_register(&sch311x_wdt_miscdev);
+	if (err != 0) {
+		dev_err(dev, "cannot register miscdev on minor=%d (err=%d)\n",
+							WATCHDOG_MINOR, err);
+		goto exit_release_region3;
+	}
+
+	sch311x_wdt_miscdev.parent = dev;
+
+	dev_info(dev,
+		"SMSC SCH311x WDT initialized. timeout=%d sec (nowayout=%d)\n",
+		timeout, nowayout);
+
+	return 0;
+
+exit_release_region3:
+	release_region(sch311x_wdt_data.runtime_reg + WDT_TIME_OUT, 4);
+exit_release_region2:
+	release_region(sch311x_wdt_data.runtime_reg + GP60, 1);
+exit_release_region:
+	release_region(sch311x_wdt_data.runtime_reg + RESGEN, 1);
+	sch311x_wdt_data.runtime_reg = 0;
+exit:
+	return err;
+}
+
+static int __devexit sch311x_wdt_remove(struct platform_device *pdev)
+{
+	/* Stop the timer before we leave */
+	if (!nowayout)
+		sch311x_wdt_stop();
+
+	/* Deregister */
+	misc_deregister(&sch311x_wdt_miscdev);
+	release_region(sch311x_wdt_data.runtime_reg + WDT_TIME_OUT, 4);
+	release_region(sch311x_wdt_data.runtime_reg + GP60, 1);
+	release_region(sch311x_wdt_data.runtime_reg + RESGEN, 1);
+	sch311x_wdt_data.runtime_reg = 0;
+	return 0;
+}
+
+static void sch311x_wdt_shutdown(struct platform_device *dev)
+{
+	/* Turn the WDT off if we have a soft shutdown */
+	sch311x_wdt_stop();
+}
+
+#define sch311x_wdt_suspend NULL
+#define sch311x_wdt_resume  NULL
+
+static struct platform_driver sch311x_wdt_driver = {
+	.probe		= sch311x_wdt_probe,
+	.remove		= __devexit_p(sch311x_wdt_remove),
+	.shutdown	= sch311x_wdt_shutdown,
+	.suspend	= sch311x_wdt_suspend,
+	.resume		= sch311x_wdt_resume,
+	.driver		= {
+		.owner = THIS_MODULE,
+		.name = DRV_NAME,
+	},
+};
+
+static int __init sch311x_detect(int sio_config_port, unsigned short *addr)
+{
+	int err = 0, reg;
+	unsigned short base_addr;
+	unsigned char dev_id;
+
+	sch311x_sio_enter(sio_config_port);
+
+	/* Check device ID. We currently know about:
+	 * SCH3112 (0x7c), SCH3114 (0x7d), and SCH3116 (0x7f). */
+	reg = force_id ? force_id : sch311x_sio_inb(sio_config_port, 0x20);
+	if (!(reg == 0x7c || reg == 0x7d || reg == 0x7f)) {
+		err = -ENODEV;
+		goto exit;
+	}
+	dev_id = reg == 0x7c ? 2 : reg == 0x7d ? 4 : 6;
+
+	/* Select logical device A (runtime registers) */
+	sch311x_sio_outb(sio_config_port, 0x07, 0x0a);
+
+	/* Check if Logical Device Register is currently active */
+	if (sch311x_sio_inb(sio_config_port, 0x30) && 0x01 == 0)
+		printk(KERN_INFO PFX "Seems that LDN 0x0a is not active...\n");
+
+	/* Get the base address of the runtime registers */
+	base_addr = (sch311x_sio_inb(sio_config_port, 0x60) << 8) |
+			   sch311x_sio_inb(sio_config_port, 0x61);
+	if (!base_addr) {
+		printk(KERN_ERR PFX "Base address not set.\n");
+		err = -ENODEV;
+		goto exit;
+	}
+	*addr = base_addr;
+
+	printk(KERN_INFO PFX "Found an SMSC SCH311%d chip at 0x%04x\n",
+		dev_id, base_addr);
+
+exit:
+	sch311x_sio_exit(sio_config_port);
+	return err;
+}
+
+static int __init sch311x_wdt_init(void)
+{
+	int err, i, found = 0;
+	unsigned short addr = 0;
+
+	for (i = 0; !found && sch311x_ioports[i]; i++)
+		if (sch311x_detect(sch311x_ioports[i], &addr) == 0)
+			found++;
+
+	if (!found)
+		return -ENODEV;
+
+	sch311x_wdt_data.runtime_reg = addr;
+
+	err = platform_driver_register(&sch311x_wdt_driver);
+	if (err)
+		return err;
+
+	sch311x_wdt_pdev = platform_device_register_simple(DRV_NAME, addr,
+								NULL, 0);
+
+	if (IS_ERR(sch311x_wdt_pdev)) {
+		err = PTR_ERR(sch311x_wdt_pdev);
+		goto unreg_platform_driver;
+	}
+
+	return 0;
+
+unreg_platform_driver:
+	platform_driver_unregister(&sch311x_wdt_driver);
+	return err;
+}
+
+static void __exit sch311x_wdt_exit(void)
+{
+	platform_device_unregister(sch311x_wdt_pdev);
+	platform_driver_unregister(&sch311x_wdt_driver);
+}
+
+module_init(sch311x_wdt_init);
+module_exit(sch311x_wdt_exit);
+
+MODULE_AUTHOR("Wim Van Sebroeck <wim@iguana.be>");
+MODULE_DESCRIPTION("SMSC SCH311x WatchDog Timer Driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
+
diff --git a/drivers/watchdog/wm8350_wdt.c b/drivers/watchdog/wm8350_wdt.c
new file mode 100644
index 0000000..2bc0d4d
--- /dev/null
+++ b/drivers/watchdog/wm8350_wdt.c
@@ -0,0 +1,329 @@
+/*
+ * Watchdog driver for the wm8350
+ *
+ * Copyright (C) 2007, 2008 Wolfson Microelectronics <linux@wolfsonmicro.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/platform_device.h>
+#include <linux/watchdog.h>
+#include <linux/uaccess.h>
+#include <linux/mfd/wm8350/core.h>
+
+static int nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout,
+		 "Watchdog cannot be stopped once started (default="
+		 __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static unsigned long wm8350_wdt_users;
+static struct miscdevice wm8350_wdt_miscdev;
+static int wm8350_wdt_expect_close;
+static DEFINE_MUTEX(wdt_mutex);
+
+static struct {
+	int time;  /* Seconds */
+	u16 val;   /* To be set in WM8350_SYSTEM_CONTROL_2 */
+} wm8350_wdt_cfgs[] = {
+	{ 1, 0x02 },
+	{ 2, 0x04 },
+	{ 4, 0x05 },
+};
+
+static struct wm8350 *get_wm8350(void)
+{
+	return dev_get_drvdata(wm8350_wdt_miscdev.parent);
+}
+
+static int wm8350_wdt_set_timeout(struct wm8350 *wm8350, u16 value)
+{
+	int ret;
+	u16 reg;
+
+	mutex_lock(&wdt_mutex);
+	wm8350_reg_unlock(wm8350);
+
+	reg = wm8350_reg_read(wm8350, WM8350_SYSTEM_CONTROL_2);
+	reg &= ~WM8350_WDOG_TO_MASK;
+	reg |= value;
+	ret = wm8350_reg_write(wm8350, WM8350_SYSTEM_CONTROL_2, reg);
+
+	wm8350_reg_lock(wm8350);
+	mutex_unlock(&wdt_mutex);
+
+	return ret;
+}
+
+static int wm8350_wdt_start(struct wm8350 *wm8350)
+{
+	int ret;
+	u16 reg;
+
+	mutex_lock(&wdt_mutex);
+	wm8350_reg_unlock(wm8350);
+
+	reg = wm8350_reg_read(wm8350, WM8350_SYSTEM_CONTROL_2);
+	reg &= ~WM8350_WDOG_MODE_MASK;
+	reg |= 0x20;
+	ret = wm8350_reg_write(wm8350, WM8350_SYSTEM_CONTROL_2, reg);
+
+	wm8350_reg_lock(wm8350);
+	mutex_unlock(&wdt_mutex);
+
+	return ret;
+}
+
+static int wm8350_wdt_stop(struct wm8350 *wm8350)
+{
+	int ret;
+	u16 reg;
+
+	mutex_lock(&wdt_mutex);
+	wm8350_reg_unlock(wm8350);
+
+	reg = wm8350_reg_read(wm8350, WM8350_SYSTEM_CONTROL_2);
+	reg &= ~WM8350_WDOG_MODE_MASK;
+	ret = wm8350_reg_write(wm8350, WM8350_SYSTEM_CONTROL_2, reg);
+
+	wm8350_reg_lock(wm8350);
+	mutex_unlock(&wdt_mutex);
+
+	return ret;
+}
+
+static int wm8350_wdt_kick(struct wm8350 *wm8350)
+{
+	int ret;
+	u16 reg;
+
+	mutex_lock(&wdt_mutex);
+
+	reg = wm8350_reg_read(wm8350, WM8350_SYSTEM_CONTROL_2);
+	ret = wm8350_reg_write(wm8350, WM8350_SYSTEM_CONTROL_2, reg);
+
+	mutex_unlock(&wdt_mutex);
+
+	return ret;
+}
+
+static int wm8350_wdt_open(struct inode *inode, struct file *file)
+{
+	struct wm8350 *wm8350 = get_wm8350();
+	int ret;
+
+	if (!wm8350)
+		return -ENODEV;
+
+	if (test_and_set_bit(0, &wm8350_wdt_users))
+		return -EBUSY;
+
+	ret = wm8350_wdt_start(wm8350);
+	if (ret != 0)
+		return ret;
+
+	return nonseekable_open(inode, file);
+}
+
+static int wm8350_wdt_release(struct inode *inode, struct file *file)
+{
+	struct wm8350 *wm8350 = get_wm8350();
+
+	if (wm8350_wdt_expect_close)
+		wm8350_wdt_stop(wm8350);
+	else {
+		dev_warn(wm8350->dev, "Watchdog device closed uncleanly\n");
+		wm8350_wdt_kick(wm8350);
+	}
+
+	clear_bit(0, &wm8350_wdt_users);
+
+	return 0;
+}
+
+static ssize_t wm8350_wdt_write(struct file *file,
+				const char __user *data, size_t count,
+				loff_t *ppos)
+{
+	struct wm8350 *wm8350 = get_wm8350();
+	size_t i;
+
+	if (count) {
+		wm8350_wdt_kick(wm8350);
+
+		if (!nowayout) {
+			/* In case it was set long ago */
+			wm8350_wdt_expect_close = 0;
+
+			/* scan to see whether or not we got the magic
+			   character */
+			for (i = 0; i != count; i++) {
+				char c;
+				if (get_user(c, data + i))
+					return -EFAULT;
+				if (c == 'V')
+					wm8350_wdt_expect_close = 42;
+			}
+		}
+	}
+	return count;
+}
+
+static struct watchdog_info ident = {
+	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE,
+	.identity = "WM8350 Watchdog",
+};
+
+static long wm8350_wdt_ioctl(struct file *file, unsigned int cmd,
+			     unsigned long arg)
+{
+	struct wm8350 *wm8350 = get_wm8350();
+	int ret = -ENOTTY, time, i;
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	u16 reg;
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		ret = copy_to_user(argp, &ident, sizeof(ident)) ? -EFAULT : 0;
+		break;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		ret = put_user(0, p);
+		break;
+
+	case WDIOC_SETOPTIONS:
+	{
+		int options;
+
+		if (get_user(options, p))
+			return -EFAULT;
+
+		ret = -EINVAL;
+
+		/* Setting both simultaneously means at least one must fail */
+		if (options == WDIOS_DISABLECARD)
+			ret = wm8350_wdt_start(wm8350);
+
+		if (options == WDIOS_ENABLECARD)
+			ret = wm8350_wdt_stop(wm8350);
+		break;
+	}
+
+	case WDIOC_KEEPALIVE:
+		ret = wm8350_wdt_kick(wm8350);
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		ret = get_user(time, p);
+		if (ret)
+			break;
+
+		if (time == 0) {
+			if (nowayout)
+				ret = -EINVAL;
+			else
+				wm8350_wdt_stop(wm8350);
+			break;
+		}
+
+		for (i = 0; i < ARRAY_SIZE(wm8350_wdt_cfgs); i++)
+			if (wm8350_wdt_cfgs[i].time == time)
+				break;
+		if (i == ARRAY_SIZE(wm8350_wdt_cfgs))
+			ret = -EINVAL;
+		else
+			ret = wm8350_wdt_set_timeout(wm8350,
+						     wm8350_wdt_cfgs[i].val);
+		break;
+
+	case WDIOC_GETTIMEOUT:
+		reg = wm8350_reg_read(wm8350, WM8350_SYSTEM_CONTROL_2);
+		reg &= WM8350_WDOG_TO_MASK;
+		for (i = 0; i < ARRAY_SIZE(wm8350_wdt_cfgs); i++)
+			if (wm8350_wdt_cfgs[i].val == reg)
+				break;
+		if (i == ARRAY_SIZE(wm8350_wdt_cfgs)) {
+			dev_warn(wm8350->dev,
+				 "Unknown watchdog configuration: %x\n", reg);
+			ret = -EINVAL;
+		} else
+			ret = put_user(wm8350_wdt_cfgs[i].time, p);
+
+	}
+
+	return ret;
+}
+
+static const struct file_operations wm8350_wdt_fops = {
+	.owner = THIS_MODULE,
+	.llseek = no_llseek,
+	.write = wm8350_wdt_write,
+	.unlocked_ioctl = wm8350_wdt_ioctl,
+	.open = wm8350_wdt_open,
+	.release = wm8350_wdt_release,
+};
+
+static struct miscdevice wm8350_wdt_miscdev = {
+	.minor = WATCHDOG_MINOR,
+	.name = "watchdog",
+	.fops = &wm8350_wdt_fops,
+};
+
+static int wm8350_wdt_probe(struct platform_device *pdev)
+{
+	struct wm8350 *wm8350 = platform_get_drvdata(pdev);
+
+	if (!wm8350) {
+		dev_err(wm8350->dev, "No driver data supplied\n");
+		return -ENODEV;
+	}
+
+	/* Default to 4s timeout */
+	wm8350_wdt_set_timeout(wm8350, 0x05);
+
+	wm8350_wdt_miscdev.parent = &pdev->dev;
+
+	return misc_register(&wm8350_wdt_miscdev);
+}
+
+static int __exit wm8350_wdt_remove(struct platform_device *pdev)
+{
+	misc_deregister(&wm8350_wdt_miscdev);
+
+	return 0;
+}
+
+static struct platform_driver wm8350_wdt_driver = {
+	.probe = wm8350_wdt_probe,
+	.remove = wm8350_wdt_remove,
+	.driver = {
+		.name = "wm8350-wdt",
+	},
+};
+
+static int __init wm8350_wdt_init(void)
+{
+	return platform_driver_register(&wm8350_wdt_driver);
+}
+module_init(wm8350_wdt_init);
+
+static void __exit wm8350_wdt_exit(void)
+{
+	platform_driver_unregister(&wm8350_wdt_driver);
+}
+module_exit(wm8350_wdt_exit);
+
+MODULE_AUTHOR("Mark Brown");
+MODULE_DESCRIPTION("WM8350 Watchdog");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:wm8350-wdt");
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 46625cd..eb0dfde 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -142,9 +142,6 @@
 
 	/* By default all event channels notify CPU#0. */
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		desc->affinity = cpumask_of_cpu(0);
 	}
 #endif
@@ -588,7 +585,7 @@
 	spin_unlock(&irq_mapping_update_lock);
 
 	/* new event channels are always bound to cpu 0 */
-	irq_set_affinity(irq, cpumask_of_cpu(0));
+	irq_set_affinity(irq, cpumask_of(0));
 
 	/* Unmask the event channel. */
 	enable_irq(irq);
@@ -617,9 +614,9 @@
 }
 
 
-static void set_affinity_irq(unsigned irq, cpumask_t dest)
+static void set_affinity_irq(unsigned irq, const struct cpumask *dest)
 {
-	unsigned tcpu = first_cpu(dest);
+	unsigned tcpu = cpumask_first(dest);
 	rebind_irq_to_cpu(irq, tcpu);
 }
 
diff --git a/fs/Kconfig b/fs/Kconfig
index 522469a..ff0e819 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -270,44 +270,7 @@
 
 endif # BLOCK
 
-config DNOTIFY
-	bool "Dnotify support"
-	default y
-	help
-	  Dnotify is a directory-based per-fd file change notification system
-	  that uses signals to communicate events to user-space.  There exist
-	  superior alternatives, but some applications may still rely on
-	  dnotify.
-
-	  If unsure, say Y.
-
-config INOTIFY
-	bool "Inotify file change notification support"
-	default y
-	---help---
-	  Say Y here to enable inotify support.  Inotify is a file change
-	  notification system and a replacement for dnotify.  Inotify fixes
-	  numerous shortcomings in dnotify and introduces several new features
-	  including multiple file events, one-shot support, and unmount
-	  notification.
-
-	  For more information, see <file:Documentation/filesystems/inotify.txt>
-
-	  If unsure, say Y.
-
-config INOTIFY_USER
-	bool "Inotify support for userspace"
-	depends on INOTIFY
-	default y
-	---help---
-	  Say Y here to enable inotify support for userspace, including the
-	  associated system calls.  Inotify allows monitoring of both files and
-	  directories via a single open fd.  Events are read from the file
-	  descriptor, which is also select()- and poll()-able.
-
-	  For more information, see <file:Documentation/filesystems/inotify.txt>
-
-	  If unsure, say Y.
+source "fs/notify/Kconfig"
 
 config QUOTA
 	bool "Quota support"
diff --git a/fs/Makefile b/fs/Makefile
index d9f8afe..e6f423d 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -20,8 +20,7 @@
 endif
 
 obj-$(CONFIG_BLK_DEV_INTEGRITY) += bio-integrity.o
-obj-$(CONFIG_INOTIFY)		+= inotify.o
-obj-$(CONFIG_INOTIFY_USER)	+= inotify_user.o
+obj-y				+= notify/
 obj-$(CONFIG_EPOLL)		+= eventpoll.o
 obj-$(CONFIG_ANON_INODES)	+= anon_inodes.o
 obj-$(CONFIG_SIGNALFD)		+= signalfd.o
@@ -57,8 +56,6 @@
 obj-$(CONFIG_QFMT_V2)		+= quota_v2.o
 obj-$(CONFIG_QUOTACTL)		+= quota.o
 
-obj-$(CONFIG_DNOTIFY)		+= dnotify.o
-
 obj-$(CONFIG_PROC_FS)		+= proc/
 obj-y				+= partitions/
 obj-$(CONFIG_SYSFS)		+= sysfs/
diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c
index c16d9be..3bbdb9d 100644
--- a/fs/anon_inodes.c
+++ b/fs/anon_inodes.c
@@ -79,9 +79,12 @@
 	if (IS_ERR(anon_inode_inode))
 		return -ENODEV;
 
+	if (fops->owner && !try_module_get(fops->owner))
+		return -ENOENT;
+
 	error = get_unused_fd_flags(flags);
 	if (error < 0)
-		return error;
+		goto err_module;
 	fd = error;
 
 	/*
@@ -128,6 +131,8 @@
 	dput(dentry);
 err_put_unused_fd:
 	put_unused_fd(fd);
+err_module:
+	module_put(fops->owner);
 	return error;
 }
 EXPORT_SYMBOL_GPL(anon_inode_getfd);
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index 5f1538c..a05287a 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -132,11 +132,6 @@
 	return -EIO;
 }
 
-static int bad_file_dir_notify(struct file *file, unsigned long arg)
-{
-	return -EIO;
-}
-
 static int bad_file_flock(struct file *filp, int cmd, struct file_lock *fl)
 {
 	return -EIO;
@@ -179,7 +174,6 @@
 	.sendpage	= bad_file_sendpage,
 	.get_unmapped_area = bad_file_get_unmapped_area,
 	.check_flags	= bad_file_check_flags,
-	.dir_notify	= bad_file_dir_notify,
 	.flock		= bad_file_flock,
 	.splice_write	= bad_file_splice_write,
 	.splice_read	= bad_file_splice_read,
diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index b6dfee3..d06cb02 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -378,7 +378,8 @@
 		inode->i_size = 0;
 		inode->i_blocks = befs_sb->block_size / VFS_BLOCK_SIZE;
 		strncpy(befs_ino->i_data.symlink, raw_inode->data.symlink,
-			BEFS_SYMLINK_LEN);
+			BEFS_SYMLINK_LEN - 1);
+		befs_ino->i_data.symlink[BEFS_SYMLINK_LEN - 1] = '\0';
 	} else {
 		int num_blks;
 
@@ -477,6 +478,8 @@
 			kfree(link);
 			befs_error(sb, "Failed to read entire long symlink");
 			link = ERR_PTR(-EIO);
+		} else {
+			link[len - 1] = '\0';
 		}
 	} else {
 		link = befs_ino->i_data.symlink;
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index f1f3f41..b639dcf 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -95,92 +95,55 @@
 	int has_dumped = 0;
 	unsigned long dump_start, dump_size;
 	struct user dump;
-#if defined(__alpha__)
+#ifdef __alpha__
 #       define START_DATA(u)	(u.start_data)
-#elif defined(__arm__)
-#	define START_DATA(u)	((u.u_tsize << PAGE_SHIFT) + u.start_code)
-#elif defined(__sparc__)
-#       define START_DATA(u)    (u.u_tsize)
-#elif defined(__i386__) || defined(__mc68000__) || defined(__arch_um__)
-#       define START_DATA(u)	(u.u_tsize << PAGE_SHIFT)
-#endif
-#ifdef __sparc__
-#       define START_STACK(u)   ((regs->u_regs[UREG_FP]) & ~(PAGE_SIZE - 1))
 #else
-#       define START_STACK(u)   (u.start_stack)
+#	define START_DATA(u)	((u.u_tsize << PAGE_SHIFT) + u.start_code)
 #endif
+#       define START_STACK(u)   (u.start_stack)
 
 	fs = get_fs();
 	set_fs(KERNEL_DS);
 	has_dumped = 1;
 	current->flags |= PF_DUMPCORE;
        	strncpy(dump.u_comm, current->comm, sizeof(dump.u_comm));
-#ifndef __sparc__
 	dump.u_ar0 = offsetof(struct user, regs);
-#endif
 	dump.signal = signr;
 	aout_dump_thread(regs, &dump);
 
 /* If the size of the dump file exceeds the rlimit, then see what would happen
    if we wrote the stack, but not the data area.  */
-#ifdef __sparc__
-	if ((dump.u_dsize + dump.u_ssize) > limit)
-		dump.u_dsize = 0;
-#else
 	if ((dump.u_dsize + dump.u_ssize+1) * PAGE_SIZE > limit)
 		dump.u_dsize = 0;
-#endif
 
 /* Make sure we have enough room to write the stack and data areas. */
-#ifdef __sparc__
-	if (dump.u_ssize > limit)
-		dump.u_ssize = 0;
-#else
 	if ((dump.u_ssize + 1) * PAGE_SIZE > limit)
 		dump.u_ssize = 0;
-#endif
 
 /* make sure we actually have a data and stack area to dump */
 	set_fs(USER_DS);
-#ifdef __sparc__
-	if (!access_ok(VERIFY_READ, (void __user *)START_DATA(dump), dump.u_dsize))
-		dump.u_dsize = 0;
-	if (!access_ok(VERIFY_READ, (void __user *)START_STACK(dump), dump.u_ssize))
-		dump.u_ssize = 0;
-#else
 	if (!access_ok(VERIFY_READ, (void __user *)START_DATA(dump), dump.u_dsize << PAGE_SHIFT))
 		dump.u_dsize = 0;
 	if (!access_ok(VERIFY_READ, (void __user *)START_STACK(dump), dump.u_ssize << PAGE_SHIFT))
 		dump.u_ssize = 0;
-#endif
 
 	set_fs(KERNEL_DS);
 /* struct user */
 	DUMP_WRITE(&dump,sizeof(dump));
 /* Now dump all of the user data.  Include malloced stuff as well */
-#ifndef __sparc__
 	DUMP_SEEK(PAGE_SIZE);
-#endif
 /* now we start writing out the user space info */
 	set_fs(USER_DS);
 /* Dump the data area */
 	if (dump.u_dsize != 0) {
 		dump_start = START_DATA(dump);
-#ifdef __sparc__
-		dump_size = dump.u_dsize;
-#else
 		dump_size = dump.u_dsize << PAGE_SHIFT;
-#endif
 		DUMP_WRITE(dump_start,dump_size);
 	}
 /* Now prepare to dump the stack area */
 	if (dump.u_ssize != 0) {
 		dump_start = START_STACK(dump);
-#ifdef __sparc__
-		dump_size = dump.u_ssize;
-#else
 		dump_size = dump.u_ssize << PAGE_SHIFT;
-#endif
 		DUMP_WRITE(dump_start,dump_size);
 	}
 /* Finally dump the task struct.  Not be used by gdb, but could be useful */
@@ -205,29 +168,24 @@
 	int envc = bprm->envc;
 
 	sp = (void __user *)((-(unsigned long)sizeof(char *)) & (unsigned long) p);
-#ifdef __sparc__
-	/* This imposes the proper stack alignment for a new process. */
-	sp = (void __user *) (((unsigned long) sp) & ~7);
-	if ((envc+argc+3)&1) --sp;
-#endif
 #ifdef __alpha__
 /* whee.. test-programs are so much fun. */
 	put_user(0, --sp);
 	put_user(0, --sp);
 	if (bprm->loader) {
 		put_user(0, --sp);
-		put_user(0x3eb, --sp);
+		put_user(1003, --sp);
 		put_user(bprm->loader, --sp);
-		put_user(0x3ea, --sp);
+		put_user(1002, --sp);
 	}
 	put_user(bprm->exec, --sp);
-	put_user(0x3e9, --sp);
+	put_user(1001, --sp);
 #endif
 	sp -= envc+1;
 	envp = (char __user * __user *) sp;
 	sp -= argc+1;
 	argv = (char __user * __user *) sp;
-#if defined(__i386__) || defined(__mc68000__) || defined(__arm__) || defined(__arch_um__)
+#ifndef __alpha__
 	put_user((unsigned long) envp,--sp);
 	put_user((unsigned long) argv,--sp);
 #endif
@@ -300,13 +258,8 @@
 		return retval;
 
 	/* OK, This is the point of no return */
-#if defined(__alpha__)
+#ifdef __alpha__
 	SET_AOUT_PERSONALITY(bprm, ex);
-#elif defined(__sparc__)
-	set_personality(PER_SUNOS);
-#if !defined(__sparc_v9__)
-	memcpy(&current->thread.core_exec, &ex, sizeof(struct exec));
-#endif
 #else
 	set_personality(PER_LINUX);
 #endif
@@ -322,24 +275,6 @@
 
 	install_exec_creds(bprm);
  	current->flags &= ~PF_FORKNOEXEC;
-#ifdef __sparc__
-	if (N_MAGIC(ex) == NMAGIC) {
-		loff_t pos = fd_offset;
-		/* Fuck me plenty... */
-		/* <AOL></AOL> */
-		down_write(&current->mm->mmap_sem);	
-		error = do_brk(N_TXTADDR(ex), ex.a_text);
-		up_write(&current->mm->mmap_sem);
-		bprm->file->f_op->read(bprm->file, (char *) N_TXTADDR(ex),
-			  ex.a_text, &pos);
-		down_write(&current->mm->mmap_sem);
-		error = do_brk(N_DATADDR(ex), ex.a_data);
-		up_write(&current->mm->mmap_sem);
-		bprm->file->f_op->read(bprm->file, (char *) N_DATADDR(ex),
-			  ex.a_data, &pos);
-		goto beyond_if;
-	}
-#endif
 
 	if (N_MAGIC(ex) == OMAGIC) {
 		unsigned long text_addr, map_size;
@@ -347,7 +282,7 @@
 
 		text_addr = N_TXTADDR(ex);
 
-#if defined(__alpha__) || defined(__sparc__)
+#ifdef __alpha__
 		pos = fd_offset;
 		map_size = ex.a_text+ex.a_data + PAGE_SIZE - 1;
 #else
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 99e0ae1..349a26c 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -326,12 +326,13 @@
 	.kill_sb	= kill_anon_super,
 };
 
-static struct vfsmount *bd_mnt __read_mostly;
-struct super_block *blockdev_superblock;
+struct super_block *blockdev_superblock __read_mostly;
 
 void __init bdev_cache_init(void)
 {
 	int err;
+	struct vfsmount *bd_mnt;
+
 	bdev_cachep = kmem_cache_create("bdev_cache", sizeof(struct bdev_inode),
 			0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
 				SLAB_MEM_SPREAD|SLAB_PANIC),
@@ -373,7 +374,7 @@
 	struct block_device *bdev;
 	struct inode *inode;
 
-	inode = iget5_locked(bd_mnt->mnt_sb, hash(dev),
+	inode = iget5_locked(blockdev_superblock, hash(dev),
 			bdev_test, bdev_set, &dev);
 
 	if (!inode)
@@ -463,7 +464,7 @@
 
 	spin_lock(&bdev_lock);
 	if (inode->i_bdev) {
-		if (inode->i_sb != blockdev_superblock)
+		if (!sb_is_blkdev_sb(inode->i_sb))
 			bdev = inode->i_bdev;
 		__bd_forget(inode);
 	}
diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile
index 6ba43fb..9948c003 100644
--- a/fs/cifs/Makefile
+++ b/fs/cifs/Makefile
@@ -5,7 +5,7 @@
 
 cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \
 	  link.o misc.o netmisc.o smbdes.o smbencrypt.o transport.o asn1.o \
-	  md4.o md5.o cifs_unicode.o nterr.o xattr.o cifsencrypt.o fcntl.o \
+	  md4.o md5.o cifs_unicode.o nterr.o xattr.o cifsencrypt.o \
 	  readdir.o ioctl.o sess.o export.o cifsacl.o
 
 cifs-$(CONFIG_CIFS_UPCALL) += cifs_spnego.o
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 0005a19..13ea532 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -747,7 +747,6 @@
 #endif /* CONFIG_CIFS_POSIX */
 
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -768,7 +767,6 @@
 #endif /* CONFIG_CIFS_POSIX */
 	.llseek = cifs_llseek,
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -789,7 +787,6 @@
 #endif /* CONFIG_CIFS_POSIX */
 
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -809,7 +806,6 @@
 #endif /* CONFIG_CIFS_POSIX */
 	.llseek = cifs_llseek,
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -818,9 +814,6 @@
 	.readdir = cifs_readdir,
 	.release = cifs_closedir,
 	.read    = generic_read_dir,
-#ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
-#endif /* CONFIG_CIFS_EXPERIMENTAL */
 	.unlocked_ioctl  = cifs_ioctl,
 	.llseek = generic_file_llseek,
 };
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 2ce04c7..7ac4818 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -76,7 +76,6 @@
 extern const struct file_operations cifs_dir_ops;
 extern int cifs_dir_open(struct inode *inode, struct file *file);
 extern int cifs_readdir(struct file *file, void *direntry, filldir_t filldir);
-extern int cifs_dir_notify(struct file *, unsigned long arg);
 
 /* Functions related to dir entries */
 extern struct dentry_operations cifs_dentry_ops;
diff --git a/fs/cifs/fcntl.c b/fs/cifs/fcntl.c
deleted file mode 100644
index 5a57581..0000000
--- a/fs/cifs/fcntl.c
+++ /dev/null
@@ -1,118 +0,0 @@
-/*
- *   fs/cifs/fcntl.c
- *
- *   vfs operations that deal with the file control API
- *
- *   Copyright (C) International Business Machines  Corp., 2003,2004
- *   Author(s): Steve French (sfrench@us.ibm.com)
- *
- *   This library is free software; you can redistribute it and/or modify
- *   it under the terms of the GNU Lesser General Public License as published
- *   by the Free Software Foundation; either version 2.1 of the License, or
- *   (at your option) any later version.
- *
- *   This library is distributed in the hope that it will be useful,
- *   but WITHOUT ANY WARRANTY; without even the implied warranty of
- *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
- *   the GNU Lesser General Public License for more details.
- *
- *   You should have received a copy of the GNU Lesser General Public License
- *   along with this library; if not, write to the Free Software
- *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-#include <linux/fs.h>
-#include <linux/stat.h>
-#include <linux/fcntl.h>
-#include "cifsglob.h"
-#include "cifsproto.h"
-#include "cifs_unicode.h"
-#include "cifs_debug.h"
-#include "cifsfs.h"
-
-static __u32 convert_to_cifs_notify_flags(unsigned long fcntl_notify_flags)
-{
-	__u32 cifs_ntfy_flags = 0;
-
-	/* No way on Linux VFS to ask to monitor xattr
-	changes (and no stream support either */
-	if (fcntl_notify_flags & DN_ACCESS)
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_ACCESS;
-	if (fcntl_notify_flags & DN_MODIFY) {
-		/* What does this mean on directories? */
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_WRITE |
-			FILE_NOTIFY_CHANGE_SIZE;
-	}
-	if (fcntl_notify_flags & DN_CREATE) {
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_CREATION |
-			FILE_NOTIFY_CHANGE_LAST_WRITE;
-	}
-	if (fcntl_notify_flags & DN_DELETE)
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_WRITE;
-	if (fcntl_notify_flags & DN_RENAME) {
-		/* BB review this - checking various server behaviors */
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_DIR_NAME |
-			FILE_NOTIFY_CHANGE_FILE_NAME;
-	}
-	if (fcntl_notify_flags & DN_ATTRIB) {
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_SECURITY |
-			FILE_NOTIFY_CHANGE_ATTRIBUTES;
-	}
-/*	if (fcntl_notify_flags & DN_MULTISHOT) {
-		cifs_ntfy_flags |= ;
-	} */ /* BB fixme - not sure how to handle this with CIFS yet */
-
-	return cifs_ntfy_flags;
-}
-
-int cifs_dir_notify(struct file *file, unsigned long arg)
-{
-	int xid;
-	int rc = -EINVAL;
-	int oplock = 0;
-	struct cifs_sb_info *cifs_sb;
-	struct cifsTconInfo *pTcon;
-	char *full_path = NULL;
-	__u32 filter = FILE_NOTIFY_CHANGE_NAME | FILE_NOTIFY_CHANGE_ATTRIBUTES;
-	__u16 netfid;
-
-	if (experimEnabled == 0)
-		return 0;
-
-	xid = GetXid();
-	cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
-	pTcon = cifs_sb->tcon;
-
-	full_path = build_path_from_dentry(file->f_path.dentry);
-
-	if (full_path == NULL) {
-		rc = -ENOMEM;
-	} else {
-		cFYI(1, ("dir notify on file %s Arg 0x%lx", full_path, arg));
-		rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN,
-			GENERIC_READ | SYNCHRONIZE, 0 /* create options */,
-			&netfid, &oplock, NULL, cifs_sb->local_nls,
-			cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
-		/* BB fixme - add this handle to a notify handle list */
-		if (rc) {
-			cFYI(1, ("Could not open directory for notify"));
-		} else {
-			filter = convert_to_cifs_notify_flags(arg);
-			if (filter != 0) {
-				rc = CIFSSMBNotify(xid, pTcon,
-					0 /* no subdirs */, netfid,
-					filter, file, arg & DN_MULTISHOT,
-					cifs_sb->local_nls);
-			} else {
-				rc = -EINVAL;
-			}
-			/* BB add code to close file eventually (at unmount
-			it would close automatically but may be a way
-			to do it easily when inode freed or when
-			notify info is cleared/changed */
-			cFYI(1, ("notify rc %d", rc));
-		}
-	}
-
-	FreeXid(xid);
-	return rc;
-}
diff --git a/fs/dcache.c b/fs/dcache.c
index a1d86c7..e88c23b 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -34,7 +34,6 @@
 #include <linux/bootmem.h>
 #include "internal.h"
 
-
 int sysctl_vfs_cache_pressure __read_mostly = 100;
 EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure);
 
@@ -948,9 +947,6 @@
 	dentry->d_op = NULL;
 	dentry->d_fsdata = NULL;
 	dentry->d_mounted = 0;
-#ifdef CONFIG_PROFILING
-	dentry->d_cookie = NULL;
-#endif
 	INIT_HLIST_NODE(&dentry->d_hash);
 	INIT_LIST_HEAD(&dentry->d_lru);
 	INIT_LIST_HEAD(&dentry->d_subdirs);
@@ -1336,7 +1332,7 @@
  *
  * Searches the children of the parent dentry for the name in question. If
  * the dentry is found its reference count is incremented and the dentry
- * is returned. The caller must use d_put to free the entry when it has
+ * is returned. The caller must use dput to free the entry when it has
  * finished using it. %NULL is returned on failure.
  *
  * __d_lookup is dcache_lock free. The hash list is protected using RCU.
@@ -1620,8 +1616,11 @@
 			 */
 			memcpy(dentry->d_iname, target->d_name.name,
 					target->d_name.len + 1);
+			dentry->d_name.len = target->d_name.len;
+			return;
 		}
 	}
+	do_switch(dentry->d_name.len, target->d_name.len);
 }
 
 /*
@@ -1681,7 +1680,6 @@
 
 	/* Switch the names.. */
 	switch_names(dentry, target);
-	do_switch(dentry->d_name.len, target->d_name.len);
 	do_switch(dentry->d_name.hash, target->d_name.hash);
 
 	/* ... and switch the parents */
@@ -1791,7 +1789,6 @@
 	struct dentry *dparent, *aparent;
 
 	switch_names(dentry, anon);
-	do_switch(dentry->d_name.len, anon->d_name.len);
 	do_switch(dentry->d_name.hash, anon->d_name.hash);
 
 	dparent = dentry->d_parent;
@@ -1911,7 +1908,8 @@
  * Convert a dentry into an ASCII path name. If the entry has been deleted
  * the string " (deleted)" is appended. Note that this is ambiguous.
  *
- * Returns the buffer or an error code if the path was too long.
+ * Returns a pointer into the buffer or an error code if the
+ * path was too long.
  *
  * "buflen" should be positive. Caller holds the dcache_lock.
  *
@@ -1987,7 +1985,10 @@
  * Convert a dentry into an ASCII path name. If the entry has been deleted
  * the string " (deleted)" is appended. Note that this is ambiguous.
  *
- * Returns the buffer or an error code if the path was too long.
+ * Returns a pointer into the buffer or an error code if the path was
+ * too long. Note: Callers should use the returned pointer, not the passed
+ * in buffer, to use the name! The implementation often starts at an offset
+ * into the buffer, and may leave 0 bytes at the start.
  *
  * "buflen" should be positive.
  */
@@ -2313,9 +2314,6 @@
 /* SLAB cache for __getname() consumers */
 struct kmem_cache *names_cachep __read_mostly;
 
-/* SLAB cache for file structures */
-struct kmem_cache *filp_cachep __read_mostly;
-
 EXPORT_SYMBOL(d_genocide);
 
 void __init vfs_caches_init_early(void)
@@ -2337,9 +2335,6 @@
 	names_cachep = kmem_cache_create("names_cache", PATH_MAX, 0,
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
 
-	filp_cachep = kmem_cache_create("filp", sizeof(struct file), 0,
-			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
-
 	dcache_init();
 	inode_init();
 	files_init(mempages);
diff --git a/fs/dcookies.c b/fs/dcookies.c
index 855d4b1..180e9fe 100644
--- a/fs/dcookies.c
+++ b/fs/dcookies.c
@@ -93,10 +93,15 @@
 {
 	struct dcookie_struct *dcs = kmem_cache_alloc(dcookie_cache,
 							GFP_KERNEL);
+	struct dentry *d;
 	if (!dcs)
 		return NULL;
 
-	path->dentry->d_cookie = dcs;
+	d = path->dentry;
+	spin_lock(&d->d_lock);
+	d->d_flags |= DCACHE_COOKIE;
+	spin_unlock(&d->d_lock);
+
 	dcs->path = *path;
 	path_get(path);
 	hash_dcookie(dcs);
@@ -119,14 +124,14 @@
 		goto out;
 	}
 
-	dcs = path->dentry->d_cookie;
-
-	if (!dcs)
+	if (path->dentry->d_flags & DCACHE_COOKIE) {
+		dcs = find_dcookie((unsigned long)path->dentry);
+	} else {
 		dcs = alloc_dcookie(path);
-
-	if (!dcs) {
-		err = -ENOMEM;
-		goto out;
+		if (!dcs) {
+			err = -ENOMEM;
+			goto out;
+		}
 	}
 
 	*cookie = dcookie_value(dcs);
@@ -251,7 +256,12 @@
 
 static void free_dcookie(struct dcookie_struct * dcs)
 {
-	dcs->path.dentry->d_cookie = NULL;
+	struct dentry *d = dcs->path.dentry;
+
+	spin_lock(&d->d_lock);
+	d->d_flags &= ~DCACHE_COOKIE;
+	spin_unlock(&d->d_lock);
+
 	path_put(&dcs->path);
 	kmem_cache_free(dcookie_cache, dcs);
 }
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 5d61b7c..fff96e1 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -27,25 +27,32 @@
 #define DEVPTS_SUPER_MAGIC 0x1cd1
 
 #define DEVPTS_DEFAULT_MODE 0600
+/*
+ * ptmx is a new node in /dev/pts and will be unused in legacy (single-
+ * instance) mode. To prevent surprises in user space, set permissions of
+ * ptmx to 0. Use 'chmod' or remount with '-o ptmxmode' to set meaningful
+ * permissions.
+ */
+#define DEVPTS_DEFAULT_PTMX_MODE 0000
 #define PTMX_MINOR	2
 
 extern int pty_limit;			/* Config limit on Unix98 ptys */
-static DEFINE_IDA(allocated_ptys);
 static DEFINE_MUTEX(allocated_ptys_lock);
 
 static struct vfsmount *devpts_mnt;
-static struct dentry *devpts_root;
 
-static struct {
+struct pts_mount_opts {
 	int setuid;
 	int setgid;
 	uid_t   uid;
 	gid_t   gid;
 	umode_t mode;
-} config = {.mode = DEVPTS_DEFAULT_MODE};
+	umode_t ptmxmode;
+	int newinstance;
+};
 
 enum {
-	Opt_uid, Opt_gid, Opt_mode,
+	Opt_uid, Opt_gid, Opt_mode, Opt_ptmxmode, Opt_newinstance,
 	Opt_err
 };
 
@@ -53,18 +60,50 @@
 	{Opt_uid, "uid=%u"},
 	{Opt_gid, "gid=%u"},
 	{Opt_mode, "mode=%o"},
+#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
+	{Opt_ptmxmode, "ptmxmode=%o"},
+	{Opt_newinstance, "newinstance"},
+#endif
 	{Opt_err, NULL}
 };
 
-static int devpts_remount(struct super_block *sb, int *flags, char *data)
+struct pts_fs_info {
+	struct ida allocated_ptys;
+	struct pts_mount_opts mount_opts;
+	struct dentry *ptmx_dentry;
+};
+
+static inline struct pts_fs_info *DEVPTS_SB(struct super_block *sb)
+{
+	return sb->s_fs_info;
+}
+
+static inline struct super_block *pts_sb_from_inode(struct inode *inode)
+{
+#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
+	if (inode->i_sb->s_magic == DEVPTS_SUPER_MAGIC)
+		return inode->i_sb;
+#endif
+	return devpts_mnt->mnt_sb;
+}
+
+#define PARSE_MOUNT	0
+#define PARSE_REMOUNT	1
+
+static int parse_mount_options(char *data, int op, struct pts_mount_opts *opts)
 {
 	char *p;
 
-	config.setuid  = 0;
-	config.setgid  = 0;
-	config.uid     = 0;
-	config.gid     = 0;
-	config.mode    = DEVPTS_DEFAULT_MODE;
+	opts->setuid  = 0;
+	opts->setgid  = 0;
+	opts->uid     = 0;
+	opts->gid     = 0;
+	opts->mode    = DEVPTS_DEFAULT_MODE;
+	opts->ptmxmode = DEVPTS_DEFAULT_PTMX_MODE;
+
+	/* newinstance makes sense only on initial mount */
+	if (op == PARSE_MOUNT)
+		opts->newinstance = 0;
 
 	while ((p = strsep(&data, ",")) != NULL) {
 		substring_t args[MAX_OPT_ARGS];
@@ -79,20 +118,32 @@
 		case Opt_uid:
 			if (match_int(&args[0], &option))
 				return -EINVAL;
-			config.uid = option;
-			config.setuid = 1;
+			opts->uid = option;
+			opts->setuid = 1;
 			break;
 		case Opt_gid:
 			if (match_int(&args[0], &option))
 				return -EINVAL;
-			config.gid = option;
-			config.setgid = 1;
+			opts->gid = option;
+			opts->setgid = 1;
 			break;
 		case Opt_mode:
 			if (match_octal(&args[0], &option))
 				return -EINVAL;
-			config.mode = option & S_IALLUGO;
+			opts->mode = option & S_IALLUGO;
 			break;
+#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
+		case Opt_ptmxmode:
+			if (match_octal(&args[0], &option))
+				return -EINVAL;
+			opts->ptmxmode = option & S_IALLUGO;
+			break;
+		case Opt_newinstance:
+			/* newinstance makes sense only on initial mount */
+			if (op == PARSE_MOUNT)
+				opts->newinstance = 1;
+			break;
+#endif
 		default:
 			printk(KERN_ERR "devpts: called with bogus options\n");
 			return -EINVAL;
@@ -102,13 +153,108 @@
 	return 0;
 }
 
+#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
+static int mknod_ptmx(struct super_block *sb)
+{
+	int mode;
+	int rc = -ENOMEM;
+	struct dentry *dentry;
+	struct inode *inode;
+	struct dentry *root = sb->s_root;
+	struct pts_fs_info *fsi = DEVPTS_SB(sb);
+	struct pts_mount_opts *opts = &fsi->mount_opts;
+
+	mutex_lock(&root->d_inode->i_mutex);
+
+	/* If we have already created ptmx node, return */
+	if (fsi->ptmx_dentry) {
+		rc = 0;
+		goto out;
+	}
+
+	dentry = d_alloc_name(root, "ptmx");
+	if (!dentry) {
+		printk(KERN_NOTICE "Unable to alloc dentry for ptmx node\n");
+		goto out;
+	}
+
+	/*
+	 * Create a new 'ptmx' node in this mount of devpts.
+	 */
+	inode = new_inode(sb);
+	if (!inode) {
+		printk(KERN_ERR "Unable to alloc inode for ptmx node\n");
+		dput(dentry);
+		goto out;
+	}
+
+	inode->i_ino = 2;
+	inode->i_uid = inode->i_gid = 0;
+	inode->i_blocks = 0;
+	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
+
+	mode = S_IFCHR|opts->ptmxmode;
+	init_special_inode(inode, mode, MKDEV(TTYAUX_MAJOR, 2));
+
+	d_add(dentry, inode);
+
+	fsi->ptmx_dentry = dentry;
+	rc = 0;
+
+	printk(KERN_DEBUG "Created ptmx node in devpts ino %lu\n",
+			inode->i_ino);
+out:
+	mutex_unlock(&root->d_inode->i_mutex);
+	return rc;
+}
+
+static void update_ptmx_mode(struct pts_fs_info *fsi)
+{
+	struct inode *inode;
+	if (fsi->ptmx_dentry) {
+		inode = fsi->ptmx_dentry->d_inode;
+		inode->i_mode = S_IFCHR|fsi->mount_opts.ptmxmode;
+	}
+}
+#else
+static inline void update_ptmx_mode(struct pts_fs_info *fsi)
+{
+       return;
+}
+#endif
+
+static int devpts_remount(struct super_block *sb, int *flags, char *data)
+{
+	int err;
+	struct pts_fs_info *fsi = DEVPTS_SB(sb);
+	struct pts_mount_opts *opts = &fsi->mount_opts;
+
+	err = parse_mount_options(data, PARSE_REMOUNT, opts);
+
+	/*
+	 * parse_mount_options() restores options to default values
+	 * before parsing and may have changed ptmxmode. So, update the
+	 * mode in the inode too. Bogus options don't fail the remount,
+	 * so do this even on error return.
+	 */
+	update_ptmx_mode(fsi);
+
+	return err;
+}
+
 static int devpts_show_options(struct seq_file *seq, struct vfsmount *vfs)
 {
-	if (config.setuid)
-		seq_printf(seq, ",uid=%u", config.uid);
-	if (config.setgid)
-		seq_printf(seq, ",gid=%u", config.gid);
-	seq_printf(seq, ",mode=%03o", config.mode);
+	struct pts_fs_info *fsi = DEVPTS_SB(vfs->mnt_sb);
+	struct pts_mount_opts *opts = &fsi->mount_opts;
+
+	if (opts->setuid)
+		seq_printf(seq, ",uid=%u", opts->uid);
+	if (opts->setgid)
+		seq_printf(seq, ",gid=%u", opts->gid);
+	seq_printf(seq, ",mode=%03o", opts->mode);
+#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
+	seq_printf(seq, ",ptmxmode=%03o", opts->ptmxmode);
+#endif
 
 	return 0;
 }
@@ -119,10 +265,25 @@
 	.show_options	= devpts_show_options,
 };
 
+static void *new_pts_fs_info(void)
+{
+	struct pts_fs_info *fsi;
+
+	fsi = kzalloc(sizeof(struct pts_fs_info), GFP_KERNEL);
+	if (!fsi)
+		return NULL;
+
+	ida_init(&fsi->allocated_ptys);
+	fsi->mount_opts.mode = DEVPTS_DEFAULT_MODE;
+	fsi->mount_opts.ptmxmode = DEVPTS_DEFAULT_PTMX_MODE;
+
+	return fsi;
+}
+
 static int
 devpts_fill_super(struct super_block *s, void *data, int silent)
 {
-	struct inode * inode;
+	struct inode *inode;
 
 	s->s_blocksize = 1024;
 	s->s_blocksize_bits = 10;
@@ -130,9 +291,13 @@
 	s->s_op = &devpts_sops;
 	s->s_time_gran = 1;
 
+	s->s_fs_info = new_pts_fs_info();
+	if (!s->s_fs_info)
+		goto fail;
+
 	inode = new_inode(s);
 	if (!inode)
-		goto fail;
+		goto free_fsi;
 	inode->i_ino = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	inode->i_blocks = 0;
@@ -142,27 +307,226 @@
 	inode->i_fop = &simple_dir_operations;
 	inode->i_nlink = 2;
 
-	devpts_root = s->s_root = d_alloc_root(inode);
+	s->s_root = d_alloc_root(inode);
 	if (s->s_root)
 		return 0;
-	
-	printk("devpts: get root dentry failed\n");
+
+	printk(KERN_ERR "devpts: get root dentry failed\n");
 	iput(inode);
+
+free_fsi:
+	kfree(s->s_fs_info);
 fail:
 	return -ENOMEM;
 }
 
+#ifdef CONFIG_DEVPTS_MULTIPLE_INSTANCES
+static int compare_init_pts_sb(struct super_block *s, void *p)
+{
+	if (devpts_mnt)
+		return devpts_mnt->mnt_sb == s;
+	return 0;
+}
+
+/*
+ * Safely parse the mount options in @data and update @opts.
+ *
+ * devpts ends up parsing options two times during mount, due to the
+ * two modes of operation it supports. The first parse occurs in
+ * devpts_get_sb() when determining the mode (single-instance or
+ * multi-instance mode). The second parse happens in devpts_remount()
+ * or new_pts_mount() depending on the mode.
+ *
+ * Parsing of options modifies the @data making subsequent parsing
+ * incorrect. So make a local copy of @data and parse it.
+ *
+ * Return: 0 On success, -errno on error
+ */
+static int safe_parse_mount_options(void *data, struct pts_mount_opts *opts)
+{
+	int rc;
+	void *datacp;
+
+	if (!data)
+		return 0;
+
+	/* Use kstrdup() ?  */
+	datacp = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!datacp)
+		return -ENOMEM;
+
+	memcpy(datacp, data, PAGE_SIZE);
+	rc = parse_mount_options((char *)datacp, PARSE_MOUNT, opts);
+	kfree(datacp);
+
+	return rc;
+}
+
+/*
+ * Mount a new (private) instance of devpts.  PTYs created in this
+ * instance are independent of the PTYs in other devpts instances.
+ */
+static int new_pts_mount(struct file_system_type *fs_type, int flags,
+		void *data, struct vfsmount *mnt)
+{
+	int err;
+	struct pts_fs_info *fsi;
+	struct pts_mount_opts *opts;
+
+	printk(KERN_NOTICE "devpts: newinstance mount\n");
+
+	err = get_sb_nodev(fs_type, flags, data, devpts_fill_super, mnt);
+	if (err)
+		return err;
+
+	fsi = DEVPTS_SB(mnt->mnt_sb);
+	opts = &fsi->mount_opts;
+
+	err = parse_mount_options(data, PARSE_MOUNT, opts);
+	if (err)
+		goto fail;
+
+	err = mknod_ptmx(mnt->mnt_sb);
+	if (err)
+		goto fail;
+
+	return 0;
+
+fail:
+	dput(mnt->mnt_sb->s_root);
+	deactivate_super(mnt->mnt_sb);
+	return err;
+}
+
+/*
+ * Check if 'newinstance' mount option was specified in @data.
+ *
+ * Return: -errno  	on error (eg: invalid mount options specified)
+ * 	 : 1 		if 'newinstance' mount option was specified
+ * 	 : 0 		if 'newinstance' mount option was NOT specified
+ */
+static int is_new_instance_mount(void *data)
+{
+	int rc;
+	struct pts_mount_opts opts;
+
+	if (!data)
+		return 0;
+
+	rc = safe_parse_mount_options(data, &opts);
+	if (!rc)
+		rc = opts.newinstance;
+
+	return rc;
+}
+
+/*
+ * get_init_pts_sb()
+ *
+ *     This interface is needed to support multiple namespace semantics in
+ *     devpts while preserving backward compatibility of the current 'single-
+ *     namespace' semantics. i.e all mounts of devpts without the 'newinstance'
+ *     mount option should bind to the initial kernel mount, like
+ *     get_sb_single().
+ *
+ *     Mounts with 'newinstance' option create a new private namespace.
+ *
+ *     But for single-mount semantics, devpts cannot use get_sb_single(),
+ *     because get_sb_single()/sget() find and use the super-block from
+ *     the most recent mount of devpts. But that recent mount may be a
+ *     'newinstance' mount and get_sb_single() would pick the newinstance
+ *     super-block instead of the initial super-block.
+ *
+ *     This interface is identical to get_sb_single() except that it
+ *     consistently selects the 'single-namespace' superblock even in the
+ *     presence of the private namespace (i.e 'newinstance') super-blocks.
+ */
+static int get_init_pts_sb(struct file_system_type *fs_type, int flags,
+		void *data, struct vfsmount *mnt)
+{
+	struct super_block *s;
+	int error;
+
+	s = sget(fs_type, compare_init_pts_sb, set_anon_super, NULL);
+	if (IS_ERR(s))
+		return PTR_ERR(s);
+
+	if (!s->s_root) {
+		s->s_flags = flags;
+		error = devpts_fill_super(s, data, flags & MS_SILENT ? 1 : 0);
+		if (error) {
+			up_write(&s->s_umount);
+			deactivate_super(s);
+			return error;
+		}
+		s->s_flags |= MS_ACTIVE;
+	}
+	do_remount_sb(s, flags, data, 0);
+	return simple_set_mnt(mnt, s);
+}
+
+/*
+ * Mount or remount the initial kernel mount of devpts. This type of
+ * mount maintains the legacy, single-instance semantics, while the
+ * kernel still allows multiple-instances.
+ */
+static int init_pts_mount(struct file_system_type *fs_type, int flags,
+		void *data, struct vfsmount *mnt)
+{
+	int err;
+
+	err = get_init_pts_sb(fs_type, flags, data, mnt);
+	if (err)
+		return err;
+
+	err = mknod_ptmx(mnt->mnt_sb);
+	if (err) {
+		dput(mnt->mnt_sb->s_root);
+		deactivate_super(mnt->mnt_sb);
+	}
+
+	return err;
+}
+
 static int devpts_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *data, struct vfsmount *mnt)
 {
+	int new;
+
+	new = is_new_instance_mount(data);
+	if (new < 0)
+		return new;
+
+	if (new)
+		return new_pts_mount(fs_type, flags, data, mnt);
+
+	return init_pts_mount(fs_type, flags, data, mnt);
+}
+#else
+/*
+ * This supports only the legacy single-instance semantics (no
+ * multiple-instance semantics)
+ */
+static int devpts_get_sb(struct file_system_type *fs_type, int flags,
+		const char *dev_name, void *data, struct vfsmount *mnt)
+{
 	return get_sb_single(fs_type, flags, data, devpts_fill_super, mnt);
 }
+#endif
+
+static void devpts_kill_sb(struct super_block *sb)
+{
+	struct pts_fs_info *fsi = DEVPTS_SB(sb);
+
+	kfree(fsi);
+	kill_litter_super(sb);
+}
 
 static struct file_system_type devpts_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "devpts",
 	.get_sb		= devpts_get_sb,
-	.kill_sb	= kill_anon_super,
+	.kill_sb	= devpts_kill_sb,
 };
 
 /*
@@ -172,16 +536,17 @@
 
 int devpts_new_index(struct inode *ptmx_inode)
 {
+	struct super_block *sb = pts_sb_from_inode(ptmx_inode);
+	struct pts_fs_info *fsi = DEVPTS_SB(sb);
 	int index;
 	int ida_ret;
 
 retry:
-	if (!ida_pre_get(&allocated_ptys, GFP_KERNEL)) {
+	if (!ida_pre_get(&fsi->allocated_ptys, GFP_KERNEL))
 		return -ENOMEM;
-	}
 
 	mutex_lock(&allocated_ptys_lock);
-	ida_ret = ida_get_new(&allocated_ptys, &index);
+	ida_ret = ida_get_new(&fsi->allocated_ptys, &index);
 	if (ida_ret < 0) {
 		mutex_unlock(&allocated_ptys_lock);
 		if (ida_ret == -EAGAIN)
@@ -190,7 +555,7 @@
 	}
 
 	if (index >= pty_limit) {
-		ida_remove(&allocated_ptys, index);
+		ida_remove(&fsi->allocated_ptys, index);
 		mutex_unlock(&allocated_ptys_lock);
 		return -EIO;
 	}
@@ -200,18 +565,26 @@
 
 void devpts_kill_index(struct inode *ptmx_inode, int idx)
 {
+	struct super_block *sb = pts_sb_from_inode(ptmx_inode);
+	struct pts_fs_info *fsi = DEVPTS_SB(sb);
+
 	mutex_lock(&allocated_ptys_lock);
-	ida_remove(&allocated_ptys, idx);
+	ida_remove(&fsi->allocated_ptys, idx);
 	mutex_unlock(&allocated_ptys_lock);
 }
 
 int devpts_pty_new(struct inode *ptmx_inode, struct tty_struct *tty)
 {
-	int number = tty->index; /* tty layer puts index from devpts_new_index() in here */
+	/* tty layer puts index from devpts_new_index() in here */
+	int number = tty->index;
 	struct tty_driver *driver = tty->driver;
 	dev_t device = MKDEV(driver->major, driver->minor_start+number);
 	struct dentry *dentry;
-	struct inode *inode = new_inode(devpts_mnt->mnt_sb);
+	struct super_block *sb = pts_sb_from_inode(ptmx_inode);
+	struct inode *inode = new_inode(sb);
+	struct dentry *root = sb->s_root;
+	struct pts_fs_info *fsi = DEVPTS_SB(sb);
+	struct pts_mount_opts *opts = &fsi->mount_opts;
 	char s[12];
 
 	/* We're supposed to be given the slave end of a pty */
@@ -221,25 +594,25 @@
 	if (!inode)
 		return -ENOMEM;
 
-	inode->i_ino = number+2;
-	inode->i_uid = config.setuid ? config.uid : current_fsuid();
-	inode->i_gid = config.setgid ? config.gid : current_fsgid();
+	inode->i_ino = number + 3;
+	inode->i_uid = opts->setuid ? opts->uid : current_fsuid();
+	inode->i_gid = opts->setgid ? opts->gid : current_fsgid();
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
-	init_special_inode(inode, S_IFCHR|config.mode, device);
+	init_special_inode(inode, S_IFCHR|opts->mode, device);
 	inode->i_private = tty;
 	tty->driver_data = inode;
 
 	sprintf(s, "%d", number);
 
-	mutex_lock(&devpts_root->d_inode->i_mutex);
+	mutex_lock(&root->d_inode->i_mutex);
 
-	dentry = d_alloc_name(devpts_root, s);
+	dentry = d_alloc_name(root, s);
 	if (!IS_ERR(dentry)) {
 		d_add(dentry, inode);
-		fsnotify_create(devpts_root->d_inode, dentry);
+		fsnotify_create(root->d_inode, dentry);
 	}
 
-	mutex_unlock(&devpts_root->d_inode->i_mutex);
+	mutex_unlock(&root->d_inode->i_mutex);
 
 	return 0;
 }
@@ -256,20 +629,27 @@
 void devpts_pty_kill(struct tty_struct *tty)
 {
 	struct inode *inode = tty->driver_data;
+	struct super_block *sb = pts_sb_from_inode(inode);
+	struct dentry *root = sb->s_root;
 	struct dentry *dentry;
 
 	BUG_ON(inode->i_rdev == MKDEV(TTYAUX_MAJOR, PTMX_MINOR));
 
-	mutex_lock(&devpts_root->d_inode->i_mutex);
+	mutex_lock(&root->d_inode->i_mutex);
 
 	dentry = d_find_alias(inode);
-	if (dentry && !IS_ERR(dentry)) {
+	if (IS_ERR(dentry))
+		goto out;
+
+	if (dentry) {
 		inode->i_nlink--;
 		d_delete(dentry);
-		dput(dentry);
+		dput(dentry);	/* d_alloc_name() in devpts_pty_new() */
 	}
 
-	mutex_unlock(&devpts_root->d_inode->i_mutex);
+	dput(dentry);		/* d_find_alias above */
+out:
+	mutex_unlock(&root->d_inode->i_mutex);
 }
 
 static int __init init_devpts_fs(void)
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 89209f0..5e78fc1 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -673,10 +673,11 @@
 	ecryptfs_printk(KERN_DEBUG, "Calling readlink w/ "
 			"dentry->d_name.name = [%s]\n", dentry->d_name.name);
 	rc = dentry->d_inode->i_op->readlink(dentry, (char __user *)buf, len);
-	buf[rc] = '\0';
 	set_fs(old_fs);
 	if (rc < 0)
 		goto out_free;
+	else
+		buf[rc] = '\0';
 	rc = 0;
 	nd_set_link(nd, buf);
 	goto out;
diff --git a/fs/exec.c b/fs/exec.c
index 02d2e12..3ef9cf9 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -57,11 +57,6 @@
 #include <asm/tlb.h>
 #include "internal.h"
 
-#ifdef __alpha__
-/* for /sbin/loader handling in search_binary_handler() */
-#include <linux/a.out.h>
-#endif
-
 int core_uses_pid;
 char core_pattern[CORENAME_MAX_SIZE] = "core";
 int suid_dumpable = 0;
@@ -127,7 +122,8 @@
 	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
 		goto exit;
 
-	error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN);
+	error = inode_permission(nd.path.dentry->d_inode,
+				 MAY_READ | MAY_EXEC | MAY_OPEN);
 	if (error)
 		goto exit;
 
@@ -680,7 +676,7 @@
 	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
 		goto out_path_put;
 
-	err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN);
+	err = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_OPEN);
 	if (err)
 		goto out_path_put;
 
@@ -1171,41 +1167,7 @@
 	unsigned int depth = bprm->recursion_depth;
 	int try,retval;
 	struct linux_binfmt *fmt;
-#ifdef __alpha__
-	/* handle /sbin/loader.. */
-	{
-	    struct exec * eh = (struct exec *) bprm->buf;
 
-	    if (!bprm->loader && eh->fh.f_magic == 0x183 &&
-		(eh->fh.f_flags & 0x3000) == 0x3000)
-	    {
-		struct file * file;
-		unsigned long loader;
-
-		allow_write_access(bprm->file);
-		fput(bprm->file);
-		bprm->file = NULL;
-
-		loader = bprm->vma->vm_end - sizeof(void *);
-
-		file = open_exec("/sbin/loader");
-		retval = PTR_ERR(file);
-		if (IS_ERR(file))
-			return retval;
-
-		/* Remember if the application is TASO.  */
-		bprm->taso = eh->ah.entry < 0x100000000UL;
-
-		bprm->file = file;
-		bprm->loader = loader;
-		retval = prepare_binprm(bprm);
-		if (retval<0)
-			return retval;
-		/* should call search_binary_handler recursively here,
-		   but it does not matter */
-	    }
-	}
-#endif
 	retval = security_bprm_check(bprm);
 	if (retval)
 		return retval;
diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index 8d0add6..c454d5d 100644
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -585,7 +585,10 @@
 	spin_lock(&sbi->s_next_gen_lock);
 	inode->i_generation = sbi->s_next_generation++;
 	spin_unlock(&sbi->s_next_gen_lock);
-	insert_inode_hash(inode);
+	if (insert_inode_locked(inode) < 0) {
+		err = -EINVAL;
+		goto fail_drop;
+	}
 
 	if (DQUOT_ALLOC_INODE(inode)) {
 		err = -EDQUOT;
@@ -612,6 +615,7 @@
 	DQUOT_DROP(inode);
 	inode->i_flags |= S_NOQUOTA;
 	inode->i_nlink = 0;
+	unlock_new_inode(inode);
 	iput(inode);
 	return ERR_PTR(err);
 
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 7658b33..02b39a5 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -32,6 +32,7 @@
 #include <linux/buffer_head.h>
 #include <linux/mpage.h>
 #include <linux/fiemap.h>
+#include <linux/namei.h>
 #include "ext2.h"
 #include "acl.h"
 #include "xip.h"
@@ -1286,9 +1287,11 @@
 		else
 			inode->i_mapping->a_ops = &ext2_aops;
 	} else if (S_ISLNK(inode->i_mode)) {
-		if (ext2_inode_is_fast_symlink(inode))
+		if (ext2_inode_is_fast_symlink(inode)) {
 			inode->i_op = &ext2_fast_symlink_inode_operations;
-		else {
+			nd_terminate_link(ei->i_data, inode->i_size,
+				sizeof(ei->i_data) - 1);
+		} else {
 			inode->i_op = &ext2_symlink_inode_operations;
 			if (test_opt(inode->i_sb, NOBH))
 				inode->i_mapping->a_ops = &ext2_nobh_aops;
diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index 2a74725..90ea179 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -41,9 +41,11 @@
 	int err = ext2_add_link(dentry, inode);
 	if (!err) {
 		d_instantiate(dentry, inode);
+		unlock_new_inode(inode);
 		return 0;
 	}
 	inode_dec_link_count(inode);
+	unlock_new_inode(inode);
 	iput(inode);
 	return err;
 }
@@ -170,6 +172,7 @@
 
 out_fail:
 	inode_dec_link_count(inode);
+	unlock_new_inode(inode);
 	iput (inode);
 	goto out;
 }
@@ -178,6 +181,7 @@
 	struct dentry *dentry)
 {
 	struct inode *inode = old_dentry->d_inode;
+	int err;
 
 	if (inode->i_nlink >= EXT2_LINK_MAX)
 		return -EMLINK;
@@ -186,7 +190,14 @@
 	inode_inc_link_count(inode);
 	atomic_inc(&inode->i_count);
 
-	return ext2_add_nondir(dentry, inode);
+	err = ext2_add_link(dentry, inode);
+	if (!err) {
+		d_instantiate(dentry, inode);
+		return 0;
+	}
+	inode_dec_link_count(inode);
+	iput(inode);
+	return err;
 }
 
 static int ext2_mkdir(struct inode * dir, struct dentry * dentry, int mode)
@@ -222,12 +233,14 @@
 		goto out_fail;
 
 	d_instantiate(dentry, inode);
+	unlock_new_inode(inode);
 out:
 	return err;
 
 out_fail:
 	inode_dec_link_count(inode);
 	inode_dec_link_count(inode);
+	unlock_new_inode(inode);
 	iput(inode);
 out_dir:
 	inode_dec_link_count(dir);
diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 490bd0e..5655fbc 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -579,7 +579,10 @@
 	ext3_set_inode_flags(inode);
 	if (IS_DIRSYNC(inode))
 		handle->h_sync = 1;
-	insert_inode_hash(inode);
+	if (insert_inode_locked(inode) < 0) {
+		err = -EINVAL;
+		goto fail_drop;
+	}
 	spin_lock(&sbi->s_next_gen_lock);
 	inode->i_generation = sbi->s_next_generation++;
 	spin_unlock(&sbi->s_next_gen_lock);
@@ -627,6 +630,7 @@
 	DQUOT_DROP(inode);
 	inode->i_flags |= S_NOQUOTA;
 	inode->i_nlink = 0;
+	unlock_new_inode(inode);
 	iput(inode);
 	brelse(bitmap_bh);
 	return ERR_PTR(err);
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index f8424ad..c4bdccf 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -37,6 +37,7 @@
 #include <linux/uio.h>
 #include <linux/bio.h>
 #include <linux/fiemap.h>
+#include <linux/namei.h>
 #include "xattr.h"
 #include "acl.h"
 
@@ -2817,9 +2818,11 @@
 		inode->i_op = &ext3_dir_inode_operations;
 		inode->i_fop = &ext3_dir_operations;
 	} else if (S_ISLNK(inode->i_mode)) {
-		if (ext3_inode_is_fast_symlink(inode))
+		if (ext3_inode_is_fast_symlink(inode)) {
 			inode->i_op = &ext3_fast_symlink_inode_operations;
-		else {
+			nd_terminate_link(ei->i_data, inode->i_size,
+				sizeof(ei->i_data) - 1);
+		} else {
 			inode->i_op = &ext3_symlink_inode_operations;
 			ext3_set_aops(inode);
 		}
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 3e5edc9..297ea8d 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -1652,9 +1652,11 @@
 	if (!err) {
 		ext3_mark_inode_dirty(handle, inode);
 		d_instantiate(dentry, inode);
+		unlock_new_inode(inode);
 		return 0;
 	}
 	drop_nlink(inode);
+	unlock_new_inode(inode);
 	iput(inode);
 	return err;
 }
@@ -1765,6 +1767,7 @@
 	dir_block = ext3_bread (handle, inode, 0, 1, &err);
 	if (!dir_block) {
 		drop_nlink(inode); /* is this nlink == 0? */
+		unlock_new_inode(inode);
 		ext3_mark_inode_dirty(handle, inode);
 		iput (inode);
 		goto out_stop;
@@ -1792,6 +1795,7 @@
 	err = ext3_add_entry (handle, dentry, inode);
 	if (err) {
 		inode->i_nlink = 0;
+		unlock_new_inode(inode);
 		ext3_mark_inode_dirty(handle, inode);
 		iput (inode);
 		goto out_stop;
@@ -1800,6 +1804,7 @@
 	ext3_update_dx_flag(dir);
 	ext3_mark_inode_dirty(handle, dir);
 	d_instantiate(dentry, inode);
+	unlock_new_inode(inode);
 out_stop:
 	ext3_journal_stop(handle);
 	if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries))
@@ -2174,6 +2179,7 @@
 				mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS);
 		if (err) {
 			drop_nlink(inode);
+			unlock_new_inode(inode);
 			ext3_mark_inode_dirty(handle, inode);
 			iput (inode);
 			goto out_stop;
@@ -2221,7 +2227,14 @@
 	inc_nlink(inode);
 	atomic_inc(&inode->i_count);
 
-	err = ext3_add_nondir(handle, dentry, inode);
+	err = ext3_add_entry(handle, dentry, inode);
+	if (!err) {
+		ext3_mark_inode_dirty(handle, inode);
+		d_instantiate(dentry, inode);
+	} else {
+		drop_nlink(inode);
+		iput(inode);
+	}
 	ext3_journal_stop(handle);
 	if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries))
 		goto retry;
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 08cac9f..6e60528 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -826,7 +826,10 @@
 	ext4_set_inode_flags(inode);
 	if (IS_DIRSYNC(inode))
 		handle->h_sync = 1;
-	insert_inode_hash(inode);
+	if (insert_inode_locked(inode) < 0) {
+		err = -EINVAL;
+		goto fail_drop;
+	}
 	spin_lock(&sbi->s_next_gen_lock);
 	inode->i_generation = sbi->s_next_generation++;
 	spin_unlock(&sbi->s_next_gen_lock);
@@ -881,6 +884,7 @@
 	DQUOT_DROP(inode);
 	inode->i_flags |= S_NOQUOTA;
 	inode->i_nlink = 0;
+	unlock_new_inode(inode);
 	iput(inode);
 	brelse(bitmap_bh);
 	return ERR_PTR(err);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index be21a5a..7c3325e 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -34,6 +34,7 @@
 #include <linux/writeback.h>
 #include <linux/pagevec.h>
 #include <linux/mpage.h>
+#include <linux/namei.h>
 #include <linux/uio.h>
 #include <linux/bio.h>
 #include "ext4_jbd2.h"
@@ -4164,9 +4165,11 @@
 		inode->i_op = &ext4_dir_inode_operations;
 		inode->i_fop = &ext4_dir_operations;
 	} else if (S_ISLNK(inode->i_mode)) {
-		if (ext4_inode_is_fast_symlink(inode))
+		if (ext4_inode_is_fast_symlink(inode)) {
 			inode->i_op = &ext4_fast_symlink_inode_operations;
-		else {
+			nd_terminate_link(ei->i_data, inode->i_size,
+				sizeof(ei->i_data) - 1);
+		} else {
 			inode->i_op = &ext4_symlink_inode_operations;
 			ext4_set_aops(inode);
 		}
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 63adcb7..da98a90 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -1693,9 +1693,11 @@
 	if (!err) {
 		ext4_mark_inode_dirty(handle, inode);
 		d_instantiate(dentry, inode);
+		unlock_new_inode(inode);
 		return 0;
 	}
 	drop_nlink(inode);
+	unlock_new_inode(inode);
 	iput(inode);
 	return err;
 }
@@ -1830,6 +1832,7 @@
 	if (err) {
 out_clear_inode:
 		clear_nlink(inode);
+		unlock_new_inode(inode);
 		ext4_mark_inode_dirty(handle, inode);
 		iput(inode);
 		goto out_stop;
@@ -1838,6 +1841,7 @@
 	ext4_update_dx_flag(dir);
 	ext4_mark_inode_dirty(handle, dir);
 	d_instantiate(dentry, inode);
+	unlock_new_inode(inode);
 out_stop:
 	ext4_journal_stop(handle);
 	if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries))
@@ -2212,6 +2216,7 @@
 				mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS);
 		if (err) {
 			clear_nlink(inode);
+			unlock_new_inode(inode);
 			ext4_mark_inode_dirty(handle, inode);
 			iput(inode);
 			goto out_stop;
@@ -2262,7 +2267,14 @@
 	ext4_inc_count(handle, inode);
 	atomic_inc(&inode->i_count);
 
-	err = ext4_add_nondir(handle, dentry, inode);
+	err = ext4_add_entry(handle, dentry, inode);
+	if (!err) {
+		ext4_mark_inode_dirty(handle, inode);
+		d_instantiate(dentry, inode);
+	} else {
+		drop_nlink(inode);
+		iput(inode);
+	}
 	ext4_journal_stop(handle);
 	if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries))
 		goto retry;
diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index 67e0583..3a7f603 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -841,7 +841,6 @@
 	.compat_ioctl	= fat_compat_dir_ioctl,
 #endif
 	.fsync		= file_fsync,
-	.llseek		= generic_file_llseek,
 };
 
 static int fat_get_short_entry(struct inode *dir, loff_t *pos,
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index d937aaf..6b74d09 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -749,6 +749,8 @@
 	brelse(bh);
 
 	parent = d_obtain_alias(inode);
+	if (!IS_ERR(parent))
+		parent->d_op = sb->s_root->d_op;
 out:
 	unlock_super(sb);
 
diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c
index bf326d4..8ae32e3 100644
--- a/fs/fat/namei_vfat.c
+++ b/fs/fat/namei_vfat.c
@@ -78,7 +78,7 @@
 	 * for creation.
 	 */
 	if (!(nd->flags & (LOOKUP_CONTINUE | LOOKUP_PARENT))) {
-		if (nd->flags & LOOKUP_CREATE)
+		if (nd->flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET))
 			return 0;
 	}
 
diff --git a/fs/file_table.c b/fs/file_table.c
index 0fbcacc..bbeeac6 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -32,6 +32,9 @@
 /* public. Not pretty! */
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(files_lock);
 
+/* SLAB cache for file structures */
+static struct kmem_cache *filp_cachep __read_mostly;
+
 static struct percpu_counter nr_files __cacheline_aligned_in_smp;
 
 static inline void file_free_rcu(struct rcu_head *head)
@@ -397,7 +400,12 @@
 void __init files_init(unsigned long mempages)
 { 
 	int n; 
-	/* One file with associated inode and dcache is very roughly 1K. 
+
+	filp_cachep = kmem_cache_create("filp", sizeof(struct file), 0,
+			SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL);
+
+	/*
+	 * One file with associated inode and dcache is very roughly 1K.
 	 * Per default don't use more than 10% of our memory for files. 
 	 */ 
 
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index 9f3f2ce..03a6ea5 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -325,8 +325,10 @@
 		if (!VXFS_ISIMMED(vip)) {
 			ip->i_op = &page_symlink_inode_operations;
 			ip->i_mapping->a_ops = &vxfs_aops;
-		} else
+		} else {
 			ip->i_op = &vxfs_immed_symlink_iops;
+			vip->vii_immed.vi_immed[ip->i_size] = '\0';
+		}
 	} else
 		init_special_inode(ip, ip->i_mode, old_decode_dev(vip->vii_rdev));
 
diff --git a/fs/inode.c b/fs/inode.c
index 098a244..7de1cda 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1032,6 +1032,65 @@
 
 EXPORT_SYMBOL(iget_locked);
 
+int insert_inode_locked(struct inode *inode)
+{
+	struct super_block *sb = inode->i_sb;
+	ino_t ino = inode->i_ino;
+	struct hlist_head *head = inode_hashtable + hash(sb, ino);
+	struct inode *old;
+
+	inode->i_state |= I_LOCK|I_NEW;
+	while (1) {
+		spin_lock(&inode_lock);
+		old = find_inode_fast(sb, head, ino);
+		if (likely(!old)) {
+			hlist_add_head(&inode->i_hash, head);
+			spin_unlock(&inode_lock);
+			return 0;
+		}
+		__iget(old);
+		spin_unlock(&inode_lock);
+		wait_on_inode(old);
+		if (unlikely(!hlist_unhashed(&old->i_hash))) {
+			iput(old);
+			return -EBUSY;
+		}
+		iput(old);
+	}
+}
+
+EXPORT_SYMBOL(insert_inode_locked);
+
+int insert_inode_locked4(struct inode *inode, unsigned long hashval,
+		int (*test)(struct inode *, void *), void *data)
+{
+	struct super_block *sb = inode->i_sb;
+	struct hlist_head *head = inode_hashtable + hash(sb, hashval);
+	struct inode *old;
+
+	inode->i_state |= I_LOCK|I_NEW;
+
+	while (1) {
+		spin_lock(&inode_lock);
+		old = find_inode(sb, head, test, data);
+		if (likely(!old)) {
+			hlist_add_head(&inode->i_hash, head);
+			spin_unlock(&inode_lock);
+			return 0;
+		}
+		__iget(old);
+		spin_unlock(&inode_lock);
+		wait_on_inode(old);
+		if (unlikely(!hlist_unhashed(&old->i_hash))) {
+			iput(old);
+			return -EBUSY;
+		}
+		iput(old);
+	}
+}
+
+EXPORT_SYMBOL(insert_inode_locked4);
+
 /**
  *	__insert_inode_hash - hash an inode
  *	@inode: unhashed inode
diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c
index 70022fd..d4d142c 100644
--- a/fs/jfs/jfs_inode.c
+++ b/fs/jfs/jfs_inode.c
@@ -79,7 +79,8 @@
 	inode = new_inode(sb);
 	if (!inode) {
 		jfs_warn("ialloc: new_inode returned NULL!");
-		return ERR_PTR(-ENOMEM);
+		rc = -ENOMEM;
+		goto fail;
 	}
 
 	jfs_inode = JFS_IP(inode);
@@ -89,8 +90,12 @@
 		jfs_warn("ialloc: diAlloc returned %d!", rc);
 		if (rc == -EIO)
 			make_bad_inode(inode);
-		iput(inode);
-		return ERR_PTR(rc);
+		goto fail_put;
+	}
+
+	if (insert_inode_locked(inode) < 0) {
+		rc = -EINVAL;
+		goto fail_unlock;
 	}
 
 	inode->i_uid = current_fsuid();
@@ -112,11 +117,8 @@
 	 * Allocate inode to quota.
 	 */
 	if (DQUOT_ALLOC_INODE(inode)) {
-		DQUOT_DROP(inode);
-		inode->i_flags |= S_NOQUOTA;
-		inode->i_nlink = 0;
-		iput(inode);
-		return ERR_PTR(-EDQUOT);
+		rc = -EDQUOT;
+		goto fail_drop;
 	}
 
 	inode->i_mode = mode;
@@ -158,4 +160,15 @@
 	jfs_info("ialloc returns inode = 0x%p\n", inode);
 
 	return inode;
+
+fail_drop:
+	DQUOT_DROP(inode);
+	inode->i_flags |= S_NOQUOTA;
+fail_unlock:
+	inode->i_nlink = 0;
+	unlock_new_inode(inode);
+fail_put:
+	iput(inode);
+fail:
+	return ERR_PTR(rc);
 }
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index cc3cedff..b4de56b 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -155,7 +155,6 @@
 	ip->i_fop = &jfs_file_operations;
 	ip->i_mapping->a_ops = &jfs_aops;
 
-	insert_inode_hash(ip);
 	mark_inode_dirty(ip);
 
 	dip->i_ctime = dip->i_mtime = CURRENT_TIME;
@@ -171,9 +170,12 @@
 	if (rc) {
 		free_ea_wmap(ip);
 		ip->i_nlink = 0;
+		unlock_new_inode(ip);
 		iput(ip);
-	} else
+	} else {
 		d_instantiate(dentry, ip);
+		unlock_new_inode(ip);
+	}
 
       out2:
 	free_UCSname(&dname);
@@ -289,7 +291,6 @@
 	ip->i_op = &jfs_dir_inode_operations;
 	ip->i_fop = &jfs_dir_operations;
 
-	insert_inode_hash(ip);
 	mark_inode_dirty(ip);
 
 	/* update parent directory inode */
@@ -306,9 +307,12 @@
 	if (rc) {
 		free_ea_wmap(ip);
 		ip->i_nlink = 0;
+		unlock_new_inode(ip);
 		iput(ip);
-	} else
+	} else {
 		d_instantiate(dentry, ip);
+		unlock_new_inode(ip);
+	}
 
       out2:
 	free_UCSname(&dname);
@@ -1019,7 +1023,6 @@
 		goto out3;
 	}
 
-	insert_inode_hash(ip);
 	mark_inode_dirty(ip);
 
 	dip->i_ctime = dip->i_mtime = CURRENT_TIME;
@@ -1039,9 +1042,12 @@
 	if (rc) {
 		free_ea_wmap(ip);
 		ip->i_nlink = 0;
+		unlock_new_inode(ip);
 		iput(ip);
-	} else
+	} else {
 		d_instantiate(dentry, ip);
+		unlock_new_inode(ip);
+	}
 
       out2:
 	free_UCSname(&dname);
@@ -1399,7 +1405,6 @@
 	jfs_ip->dev = new_encode_dev(rdev);
 	init_special_inode(ip, ip->i_mode, rdev);
 
-	insert_inode_hash(ip);
 	mark_inode_dirty(ip);
 
 	dir->i_ctime = dir->i_mtime = CURRENT_TIME;
@@ -1417,9 +1422,12 @@
 	if (rc) {
 		free_ea_wmap(ip);
 		ip->i_nlink = 0;
+		unlock_new_inode(ip);
 		iput(ip);
-	} else
+	} else {
 		d_instantiate(dentry, ip);
+		unlock_new_inode(ip);
+	}
 
       out1:
 	free_UCSname(&dname);
diff --git a/fs/namei.c b/fs/namei.c
index af3783f..dd5c9f0 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -226,6 +226,16 @@
 	return -EACCES;
 }
 
+/**
+ * inode_permission  -  check for access rights to a given inode
+ * @inode:	inode to check permission on
+ * @mask:	right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
+ *
+ * Used to check for read/write/execute permissions on an inode.
+ * We use "fsuid" for this, letting us set arbitrary permissions
+ * for filesystem access without changing the "normal" uids which
+ * are used for other things.
+ */
 int inode_permission(struct inode *inode, int mask)
 {
 	int retval;
@@ -247,7 +257,6 @@
 			return -EACCES;
 	}
 
-	/* Ordinary permission routines do not understand MAY_APPEND. */
 	if (inode->i_op && inode->i_op->permission)
 		retval = inode->i_op->permission(inode, mask);
 	else
@@ -265,21 +274,6 @@
 }
 
 /**
- * vfs_permission  -  check for access rights to a given path
- * @nd:		lookup result that describes the path
- * @mask:	right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
- *
- * Used to check for read/write/execute permissions on a path.
- * We use "fsuid" for this, letting us set arbitrary permissions
- * for filesystem access without changing the "normal" uids which
- * are used for other things.
- */
-int vfs_permission(struct nameidata *nd, int mask)
-{
-	return inode_permission(nd->path.dentry->d_inode, mask);
-}
-
-/**
  * file_permission  -  check for additional access rights to a given file
  * @file:	file to check access rights for
  * @mask:	right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
@@ -289,7 +283,7 @@
  *
  * Note:
  *	Do not use this function in new code.  All access checks should
- *	be done using vfs_permission().
+ *	be done using inode_permission().
  */
 int file_permission(struct file *file, int mask)
 {
@@ -527,18 +521,6 @@
 	return result;
 }
 
-/* SMP-safe */
-static __always_inline void
-walk_init_root(const char *name, struct nameidata *nd)
-{
-	struct fs_struct *fs = current->fs;
-
-	read_lock(&fs->lock);
-	nd->path = fs->root;
-	path_get(&fs->root);
-	read_unlock(&fs->lock);
-}
-
 /*
  * Wrapper to retry pathname resolution whenever the underlying
  * file system returns an ESTALE.
@@ -576,9 +558,16 @@
 		goto fail;
 
 	if (*link == '/') {
+		struct fs_struct *fs = current->fs;
+
 		path_put(&nd->path);
-		walk_init_root(link, nd);
+
+		read_lock(&fs->lock);
+		nd->path = fs->root;
+		path_get(&fs->root);
+		read_unlock(&fs->lock);
 	}
+
 	res = link_path_walk(link, nd);
 	if (nd->depth || res || nd->last_type!=LAST_NORM)
 		return res;
@@ -859,7 +848,8 @@
 		nd->flags |= LOOKUP_CONTINUE;
 		err = exec_permission_lite(inode);
 		if (err == -EAGAIN)
-			err = vfs_permission(nd, MAY_EXEC);
+			err = inode_permission(nd->path.dentry->d_inode,
+					       MAY_EXEC);
  		if (err)
 			break;
 
@@ -1493,9 +1483,9 @@
 	return error;
 }
 
-int may_open(struct nameidata *nd, int acc_mode, int flag)
+int may_open(struct path *path, int acc_mode, int flag)
 {
-	struct dentry *dentry = nd->path.dentry;
+	struct dentry *dentry = path->dentry;
 	struct inode *inode = dentry->d_inode;
 	int error;
 
@@ -1516,13 +1506,13 @@
 	if (S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) {
 	    	flag &= ~O_TRUNC;
 	} else if (S_ISBLK(inode->i_mode) || S_ISCHR(inode->i_mode)) {
-		if (nd->path.mnt->mnt_flags & MNT_NODEV)
+		if (path->mnt->mnt_flags & MNT_NODEV)
 			return -EACCES;
 
 		flag &= ~O_TRUNC;
 	}
 
-	error = vfs_permission(nd, acc_mode);
+	error = inode_permission(inode, acc_mode);
 	if (error)
 		return error;
 	/*
@@ -1556,6 +1546,9 @@
 		 * Refuse to truncate files with mandatory locks held on them.
 		 */
 		error = locks_verify_locked(inode);
+		if (!error)
+			error = security_path_truncate(path, 0,
+					       ATTR_MTIME|ATTR_CTIME|ATTR_OPEN);
 		if (!error) {
 			DQUOT_INIT(inode);
 
@@ -1586,14 +1579,18 @@
 
 	if (!IS_POSIXACL(dir->d_inode))
 		mode &= ~current->fs->umask;
+	error = security_path_mknod(&nd->path, path->dentry, mode, 0);
+	if (error)
+		goto out_unlock;
 	error = vfs_create(dir->d_inode, path->dentry, mode, nd);
+out_unlock:
 	mutex_unlock(&dir->d_inode->i_mutex);
 	dput(nd->path.dentry);
 	nd->path.dentry = path->dentry;
 	if (error)
 		return error;
 	/* Don't check for write permission, don't truncate */
-	return may_open(nd, 0, flag & ~O_TRUNC);
+	return may_open(&nd->path, 0, flag & ~O_TRUNC);
 }
 
 /*
@@ -1779,7 +1776,7 @@
 		if (error)
 			goto exit;
 	}
-	error = may_open(&nd, acc_mode, flag);
+	error = may_open(&nd.path, acc_mode, flag);
 	if (error) {
 		if (will_write)
 			mnt_drop_write(nd.path.mnt);
@@ -1999,6 +1996,9 @@
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_mknod(&nd.path, dentry, mode, dev);
+	if (error)
+		goto out_drop_write;
 	switch (mode & S_IFMT) {
 		case 0: case S_IFREG:
 			error = vfs_create(nd.path.dentry->d_inode,dentry,mode,&nd);
@@ -2011,6 +2011,7 @@
 			error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode,0);
 			break;
 	}
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(dentry);
@@ -2070,7 +2071,11 @@
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_mkdir(&nd.path, dentry, mode);
+	if (error)
+		goto out_drop_write;
 	error = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode);
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(dentry);
@@ -2180,7 +2185,11 @@
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto exit3;
+	error = security_path_rmdir(&nd.path, dentry);
+	if (error)
+		goto exit4;
 	error = vfs_rmdir(nd.path.dentry->d_inode, dentry);
+exit4:
 	mnt_drop_write(nd.path.mnt);
 exit3:
 	dput(dentry);
@@ -2265,7 +2274,11 @@
 		error = mnt_want_write(nd.path.mnt);
 		if (error)
 			goto exit2;
+		error = security_path_unlink(&nd.path, dentry);
+		if (error)
+			goto exit3;
 		error = vfs_unlink(nd.path.dentry->d_inode, dentry);
+exit3:
 		mnt_drop_write(nd.path.mnt);
 	exit2:
 		dput(dentry);
@@ -2346,7 +2359,11 @@
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_symlink(&nd.path, dentry, from);
+	if (error)
+		goto out_drop_write;
 	error = vfs_symlink(nd.path.dentry->d_inode, dentry, from);
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(dentry);
@@ -2443,7 +2460,11 @@
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_link(old_path.dentry, &nd.path, new_dentry);
+	if (error)
+		goto out_drop_write;
 	error = vfs_link(old_path.dentry, nd.path.dentry->d_inode, new_dentry);
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(new_dentry);
@@ -2679,8 +2700,13 @@
 	error = mnt_want_write(oldnd.path.mnt);
 	if (error)
 		goto exit5;
+	error = security_path_rename(&oldnd.path, old_dentry,
+				     &newnd.path, new_dentry);
+	if (error)
+		goto exit6;
 	error = vfs_rename(old_dir->d_inode, old_dentry,
 				   new_dir->d_inode, new_dentry);
+exit6:
 	mnt_drop_write(oldnd.path.mnt);
 exit5:
 	dput(new_dentry);
@@ -2750,13 +2776,16 @@
 /* get the link contents into pagecache */
 static char *page_getlink(struct dentry * dentry, struct page **ppage)
 {
-	struct page * page;
+	char *kaddr;
+	struct page *page;
 	struct address_space *mapping = dentry->d_inode->i_mapping;
 	page = read_mapping_page(mapping, 0, NULL);
 	if (IS_ERR(page))
 		return (char*)page;
 	*ppage = page;
-	return kmap(page);
+	kaddr = kmap(page);
+	nd_terminate_link(kaddr, dentry->d_inode->i_size, PAGE_SIZE - 1);
+	return kaddr;
 }
 
 int page_readlink(struct dentry *dentry, char __user *buffer, int buflen)
@@ -2849,7 +2878,6 @@
 EXPORT_SYMBOL(kern_path);
 EXPORT_SYMBOL(vfs_path_lookup);
 EXPORT_SYMBOL(inode_permission);
-EXPORT_SYMBOL(vfs_permission);
 EXPORT_SYMBOL(file_permission);
 EXPORT_SYMBOL(unlock_rename);
 EXPORT_SYMBOL(vfs_create);
@@ -2865,3 +2893,10 @@
 EXPORT_SYMBOL(vfs_unlink);
 EXPORT_SYMBOL(dentry_unhash);
 EXPORT_SYMBOL(generic_readlink);
+
+/* to be mentioned only in INIT_TASK */
+struct fs_struct init_fs = {
+	.count		= ATOMIC_INIT(1),
+	.lock		= __RW_LOCK_UNLOCKED(init_fs.lock),
+	.umask		= 0022,
+};
diff --git a/fs/namespace.c b/fs/namespace.c
index 1c09cab..a40685d 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1990,7 +1990,7 @@
 	if (!new_ns->root) {
 		up_write(&namespace_sem);
 		kfree(new_ns);
-		return ERR_PTR(-ENOMEM);;
+		return ERR_PTR(-ENOMEM);
 	}
 	spin_lock(&vfsmount_lock);
 	list_add_tail(&new_ns->list, &new_ns->root->mnt_list);
diff --git a/fs/nfsctl.c b/fs/nfsctl.c
index b1acbd6..b274519 100644
--- a/fs/nfsctl.c
+++ b/fs/nfsctl.c
@@ -38,9 +38,10 @@
 		return ERR_PTR(error);
 
 	if (flags == O_RDWR)
-		error = may_open(&nd,MAY_READ|MAY_WRITE,FMODE_READ|FMODE_WRITE);
+		error = may_open(&nd.path, MAY_READ|MAY_WRITE,
+					   FMODE_READ|FMODE_WRITE);
 	else
-		error = may_open(&nd, MAY_WRITE, FMODE_WRITE);
+		error = may_open(&nd.path, MAY_WRITE, FMODE_WRITE);
 
 	if (!error)
 		return dentry_open(nd.path.dentry, nd.path.mnt, flags,
diff --git a/fs/notify/Kconfig b/fs/notify/Kconfig
new file mode 100644
index 0000000..50914d7
--- /dev/null
+++ b/fs/notify/Kconfig
@@ -0,0 +1,2 @@
+source "fs/notify/dnotify/Kconfig"
+source "fs/notify/inotify/Kconfig"
diff --git a/fs/notify/Makefile b/fs/notify/Makefile
new file mode 100644
index 0000000..5a95b60
--- /dev/null
+++ b/fs/notify/Makefile
@@ -0,0 +1,2 @@
+obj-y			+= dnotify/
+obj-y			+= inotify/
diff --git a/fs/notify/dnotify/Kconfig b/fs/notify/dnotify/Kconfig
new file mode 100644
index 0000000..26adf5d
--- /dev/null
+++ b/fs/notify/dnotify/Kconfig
@@ -0,0 +1,10 @@
+config DNOTIFY
+	bool "Dnotify support"
+	default y
+	help
+	  Dnotify is a directory-based per-fd file change notification system
+	  that uses signals to communicate events to user-space.  There exist
+	  superior alternatives, but some applications may still rely on
+	  dnotify.
+
+	  If unsure, say Y.
diff --git a/fs/notify/dnotify/Makefile b/fs/notify/dnotify/Makefile
new file mode 100644
index 0000000..f145251
--- /dev/null
+++ b/fs/notify/dnotify/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_DNOTIFY)		+= dnotify.o
diff --git a/fs/dnotify.c b/fs/notify/dnotify/dnotify.c
similarity index 97%
rename from fs/dnotify.c
rename to fs/notify/dnotify/dnotify.c
index 676073b..b0aa2cd 100644
--- a/fs/dnotify.c
+++ b/fs/notify/dnotify/dnotify.c
@@ -115,9 +115,6 @@
 	dn->dn_next = inode->i_dnotify;
 	inode->i_dnotify = dn;
 	spin_unlock(&inode->i_lock);
-
-	if (filp->f_op && filp->f_op->dir_notify)
-		return filp->f_op->dir_notify(filp, arg);
 	return 0;
 
 out_free:
diff --git a/fs/notify/inotify/Kconfig b/fs/notify/inotify/Kconfig
new file mode 100644
index 0000000..4467928
--- /dev/null
+++ b/fs/notify/inotify/Kconfig
@@ -0,0 +1,27 @@
+config INOTIFY
+	bool "Inotify file change notification support"
+	default y
+	---help---
+	  Say Y here to enable inotify support.  Inotify is a file change
+	  notification system and a replacement for dnotify.  Inotify fixes
+	  numerous shortcomings in dnotify and introduces several new features
+	  including multiple file events, one-shot support, and unmount
+	  notification.
+
+	  For more information, see <file:Documentation/filesystems/inotify.txt>
+
+	  If unsure, say Y.
+
+config INOTIFY_USER
+	bool "Inotify support for userspace"
+	depends on INOTIFY
+	default y
+	---help---
+	  Say Y here to enable inotify support for userspace, including the
+	  associated system calls.  Inotify allows monitoring of both files and
+	  directories via a single open fd.  Events are read from the file
+	  descriptor, which is also select()- and poll()-able.
+
+	  For more information, see <file:Documentation/filesystems/inotify.txt>
+
+	  If unsure, say Y.
diff --git a/fs/notify/inotify/Makefile b/fs/notify/inotify/Makefile
new file mode 100644
index 0000000..e290f3b
--- /dev/null
+++ b/fs/notify/inotify/Makefile
@@ -0,0 +1,2 @@
+obj-$(CONFIG_INOTIFY)		+= inotify.o
+obj-$(CONFIG_INOTIFY_USER)	+= inotify_user.o
diff --git a/fs/inotify.c b/fs/notify/inotify/inotify.c
similarity index 100%
rename from fs/inotify.c
rename to fs/notify/inotify/inotify.c
diff --git a/fs/inotify_user.c b/fs/notify/inotify/inotify_user.c
similarity index 99%
rename from fs/inotify_user.c
rename to fs/notify/inotify/inotify_user.c
index e2425bb..400f806 100644
--- a/fs/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -76,10 +76,10 @@
 	struct mutex		ev_mutex;	/* protects event queue */
 	struct mutex		up_mutex;	/* synchronizes watch updates */
 	struct list_head 	events;		/* list of queued events */
-	atomic_t		count;		/* reference count */
 	struct user_struct	*user;		/* user who opened this dev */
 	struct inotify_handle	*ih;		/* inotify handle */
 	struct fasync_struct    *fa;            /* async notification */
+	atomic_t		count;		/* reference count */
 	unsigned int		queue_size;	/* size of the queue (bytes) */
 	unsigned int		event_count;	/* number of pending events */
 	unsigned int		max_events;	/* maximum number of events */
diff --git a/fs/open.c b/fs/open.c
index c0a426d..1cd7d40 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -272,6 +272,8 @@
 		goto put_write_and_out;
 
 	error = locks_verify_truncate(inode, NULL, length);
+	if (!error)
+		error = security_path_truncate(&path, length, 0);
 	if (!error) {
 		DQUOT_INIT(inode);
 		error = do_truncate(path.dentry, length, 0, NULL);
@@ -329,6 +331,9 @@
 
 	error = locks_verify_truncate(inode, file, length);
 	if (!error)
+		error = security_path_truncate(&file->f_path, length,
+					       ATTR_MTIME|ATTR_CTIME);
+	if (!error)
 		error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, file);
 out_putf:
 	fput(file);
diff --git a/fs/proc/stat.c b/fs/proc/stat.c
index 3bb1cf1..f75efa2 100644
--- a/fs/proc/stat.c
+++ b/fs/proc/stat.c
@@ -9,6 +9,7 @@
 #include <linux/seq_file.h>
 #include <linux/slab.h>
 #include <linux/time.h>
+#include <linux/irqnr.h>
 #include <asm/cputime.h>
 
 #ifndef arch_irq_stat_cpu
@@ -45,10 +46,6 @@
 		steal = cputime64_add(steal, kstat_cpu(i).cpustat.steal);
 		guest = cputime64_add(guest, kstat_cpu(i).cpustat.guest);
 		for_each_irq_nr(j) {
-#ifdef CONFIG_SPARSE_IRQ
-			if (!irq_to_desc(j))
-				continue;
-#endif
 			sum += kstat_irqs_cpu(j, i);
 		}
 		sum += arch_irq_stat_cpu(i);
@@ -95,12 +92,6 @@
 	/* sum again ? it could be updated? */
 	for_each_irq_nr(j) {
 		per_irq_sum = 0;
-#ifdef CONFIG_SPARSE_IRQ
-		if (!irq_to_desc(j)) {
-			seq_printf(p, " %u", per_irq_sum);
-			continue;
-		}
-#endif
 		for_each_possible_cpu(i)
 			per_irq_sum += kstat_irqs_cpu(j, i);
 
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 6c4c2c6..145c2d3 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1753,6 +1753,7 @@
 		       struct inode *inode)
 {
 	struct super_block *sb;
+	struct reiserfs_iget_args args;
 	INITIALIZE_PATH(path_to_key);
 	struct cpu_key key;
 	struct item_head ih;
@@ -1780,6 +1781,14 @@
 		err = -ENOMEM;
 		goto out_bad_inode;
 	}
+	args.objectid = inode->i_ino = le32_to_cpu(ih.ih_key.k_objectid);
+	memcpy(INODE_PKEY(inode), &(ih.ih_key), KEY_SIZE);
+	args.dirid = le32_to_cpu(ih.ih_key.k_dir_id);
+	if (insert_inode_locked4(inode, args.objectid,
+			     reiserfs_find_actor, &args) < 0) {
+		err = -EINVAL;
+		goto out_bad_inode;
+	}
 	if (old_format_only(sb))
 		/* not a perfect generation count, as object ids can be reused, but 
 		 ** this is as good as reiserfs can do right now.
@@ -1859,13 +1868,9 @@
 	} else {
 		inode2sd(&sd, inode, inode->i_size);
 	}
-	// these do not go to on-disk stat data
-	inode->i_ino = le32_to_cpu(ih.ih_key.k_objectid);
-
 	// store in in-core inode the key of stat data and version all
 	// object items will have (directory items will have old offset
 	// format, other new objects will consist of new items)
-	memcpy(INODE_PKEY(inode), &(ih.ih_key), KEY_SIZE);
 	if (old_format_only(sb) || S_ISDIR(mode) || S_ISLNK(mode))
 		set_inode_item_key_version(inode, KEY_FORMAT_3_5);
 	else
@@ -1929,7 +1934,6 @@
 		reiserfs_mark_inode_private(inode);
 	}
 
-	insert_inode_hash(inode);
 	reiserfs_update_sd(th, inode);
 	reiserfs_check_path(&path_to_key);
 
@@ -1956,6 +1960,7 @@
       out_inserted_sd:
 	inode->i_nlink = 0;
 	th->t_trans_id = 0;	/* so the caller can't use this handle later */
+	unlock_new_inode(inode); /* OK to do even if we hadn't locked it */
 
 	/* If we were inheriting an ACL, we need to release the lock so that
 	 * iput doesn't deadlock in reiserfs_delete_xattrs. The locking
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index 4f322e5..738967f 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -646,6 +646,7 @@
 		err = journal_end(&th, dir->i_sb, jbegin_count);
 		if (err)
 			retval = err;
+		unlock_new_inode(inode);
 		iput(inode);
 		goto out_failed;
 	}
@@ -653,6 +654,7 @@
 	reiserfs_update_inode_transaction(dir);
 
 	d_instantiate(dentry, inode);
+	unlock_new_inode(inode);
 	retval = journal_end(&th, dir->i_sb, jbegin_count);
 
       out_failed:
@@ -727,11 +729,13 @@
 		err = journal_end(&th, dir->i_sb, jbegin_count);
 		if (err)
 			retval = err;
+		unlock_new_inode(inode);
 		iput(inode);
 		goto out_failed;
 	}
 
 	d_instantiate(dentry, inode);
+	unlock_new_inode(inode);
 	retval = journal_end(&th, dir->i_sb, jbegin_count);
 
       out_failed:
@@ -812,6 +816,7 @@
 		err = journal_end(&th, dir->i_sb, jbegin_count);
 		if (err)
 			retval = err;
+		unlock_new_inode(inode);
 		iput(inode);
 		goto out_failed;
 	}
@@ -819,6 +824,7 @@
 	reiserfs_update_sd(&th, dir);
 
 	d_instantiate(dentry, inode);
+	unlock_new_inode(inode);
 	retval = journal_end(&th, dir->i_sb, jbegin_count);
       out_failed:
 	if (locked)
@@ -1096,11 +1102,13 @@
 		err = journal_end(&th, parent_dir->i_sb, jbegin_count);
 		if (err)
 			retval = err;
+		unlock_new_inode(inode);
 		iput(inode);
 		goto out_failed;
 	}
 
 	d_instantiate(dentry, inode);
+	unlock_new_inode(inode);
 	retval = journal_end(&th, parent_dir->i_sb, jbegin_count);
       out_failed:
 	reiserfs_write_unlock(parent_dir->i_sb);
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 16c2115..99d8b8c 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -389,8 +389,14 @@
 }
 EXPORT_SYMBOL(mangle_path);
 
-/*
- * return the absolute path of 'dentry' residing in mount 'mnt'.
+/**
+ * seq_path - seq_file interface to print a pathname
+ * @m: the seq_file handle
+ * @path: the struct path to print
+ * @esc: set of characters to escape in the output
+ *
+ * return the absolute path of 'path', as represented by the
+ * dentry / mnt pair in the path parameter.
  */
 int seq_path(struct seq_file *m, struct path *path, char *esc)
 {
diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index df0d435..3d81bf5 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -27,6 +27,7 @@
 #include <linux/init.h>
 #include <linux/buffer_head.h>
 #include <linux/vfs.h>
+#include <linux/namei.h>
 #include <asm/byteorder.h>
 #include "sysv.h"
 
@@ -163,8 +164,11 @@
 		if (inode->i_blocks) {
 			inode->i_op = &sysv_symlink_inode_operations;
 			inode->i_mapping->a_ops = &sysv_aops;
-		} else
+		} else {
 			inode->i_op = &sysv_fast_symlink_inode_operations;
+			nd_terminate_link(SYSV_I(inode)->i_data, inode->i_size,
+				sizeof(SYSV_I(inode)->i_data) - 1);
+		}
 	} else
 		init_special_inode(inode, inode->i_mode, rdev);
 }
diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c
index 4a18f08..0e5e54d 100644
--- a/fs/ubifs/budget.c
+++ b/fs/ubifs/budget.c
@@ -32,18 +32,15 @@
 
 #include "ubifs.h"
 #include <linux/writeback.h>
-#include <asm/div64.h>
+#include <linux/math64.h>
 
 /*
  * When pessimistic budget calculations say that there is no enough space,
  * UBIFS starts writing back dirty inodes and pages, doing garbage collection,
- * or committing. The below constants define maximum number of times UBIFS
+ * or committing. The below constant defines maximum number of times UBIFS
  * repeats the operations.
  */
-#define MAX_SHRINK_RETRIES 8
-#define MAX_GC_RETRIES     4
-#define MAX_CMT_RETRIES    2
-#define MAX_NOSPC_RETRIES  1
+#define MAX_MKSPC_RETRIES 3
 
 /*
  * The below constant defines amount of dirty pages which should be written
@@ -52,30 +49,6 @@
 #define NR_TO_WRITE 16
 
 /**
- * struct retries_info - information about re-tries while making free space.
- * @prev_liability: previous liability
- * @shrink_cnt: how many times the liability was shrinked
- * @shrink_retries: count of liability shrink re-tries (increased when
- *                  liability does not shrink)
- * @try_gc: GC should be tried first
- * @gc_retries: how many times GC was run
- * @cmt_retries: how many times commit has been done
- * @nospc_retries: how many times GC returned %-ENOSPC
- *
- * Since we consider budgeting to be the fast-path, and this structure has to
- * be allocated on stack and zeroed out, we make it smaller using bit-fields.
- */
-struct retries_info {
-	long long prev_liability;
-	unsigned int shrink_cnt;
-	unsigned int shrink_retries:5;
-	unsigned int try_gc:1;
-	unsigned int gc_retries:4;
-	unsigned int cmt_retries:3;
-	unsigned int nospc_retries:1;
-};
-
-/**
  * shrink_liability - write-back some dirty pages/inodes.
  * @c: UBIFS file-system description object
  * @nr_to_write: how many dirty pages to write-back
@@ -147,9 +120,25 @@
 }
 
 /**
+ * get_liability - calculate current liability.
+ * @c: UBIFS file-system description object
+ *
+ * This function calculates and returns current UBIFS liability, i.e. the
+ * amount of bytes UBIFS has "promised" to write to the media.
+ */
+static long long get_liability(struct ubifs_info *c)
+{
+	long long liab;
+
+	spin_lock(&c->space_lock);
+	liab = c->budg_idx_growth + c->budg_data_growth + c->budg_dd_growth;
+	spin_unlock(&c->space_lock);
+	return liab;
+}
+
+/**
  * make_free_space - make more free space on the file-system.
  * @c: UBIFS file-system description object
- * @ri: information about previous invocations of this function
  *
  * This function is called when an operation cannot be budgeted because there
  * is supposedly no free space. But in most cases there is some free space:
@@ -165,87 +154,42 @@
  * Returns %-ENOSPC if it couldn't do more free space, and other negative error
  * codes on failures.
  */
-static int make_free_space(struct ubifs_info *c, struct retries_info *ri)
+static int make_free_space(struct ubifs_info *c)
 {
-	int err;
+	int err, retries = 0;
+	long long liab1, liab2;
 
-	/*
-	 * If we have some dirty pages and inodes (liability), try to write
-	 * them back unless this was tried too many times without effect
-	 * already.
-	 */
-	if (ri->shrink_retries < MAX_SHRINK_RETRIES && !ri->try_gc) {
-		long long liability;
+	do {
+		liab1 = get_liability(c);
+		/*
+		 * We probably have some dirty pages or inodes (liability), try
+		 * to write them back.
+		 */
+		dbg_budg("liability %lld, run write-back", liab1);
+		shrink_liability(c, NR_TO_WRITE);
 
-		spin_lock(&c->space_lock);
-		liability = c->budg_idx_growth + c->budg_data_growth +
-			    c->budg_dd_growth;
-		spin_unlock(&c->space_lock);
+		liab2 = get_liability(c);
+		if (liab2 < liab1)
+			return -EAGAIN;
 
-		if (ri->prev_liability >= liability) {
-			/* Liability does not shrink, next time try GC then */
-			ri->shrink_retries += 1;
-			if (ri->gc_retries < MAX_GC_RETRIES)
-				ri->try_gc = 1;
-			dbg_budg("liability did not shrink: retries %d of %d",
-				 ri->shrink_retries, MAX_SHRINK_RETRIES);
-		}
+		dbg_budg("new liability %lld (not shrinked)", liab2);
 
-		dbg_budg("force write-back (count %d)", ri->shrink_cnt);
-		shrink_liability(c, NR_TO_WRITE + ri->shrink_cnt);
-
-		ri->prev_liability = liability;
-		ri->shrink_cnt += 1;
-		return -EAGAIN;
-	}
-
-	/*
-	 * Try to run garbage collector unless it was already tried too many
-	 * times.
-	 */
-	if (ri->gc_retries < MAX_GC_RETRIES) {
-		ri->gc_retries += 1;
-		dbg_budg("run GC, retries %d of %d",
-			 ri->gc_retries, MAX_GC_RETRIES);
-
-		ri->try_gc = 0;
+		/* Liability did not shrink again, try GC */
+		dbg_budg("Run GC");
 		err = run_gc(c);
 		if (!err)
 			return -EAGAIN;
 
-		if (err == -EAGAIN) {
-			dbg_budg("GC asked to commit");
-			err = ubifs_run_commit(c);
-			if (err)
-				return err;
-			return -EAGAIN;
-		}
-
-		if (err != -ENOSPC)
+		if (err != -EAGAIN && err != -ENOSPC)
+			/* Some real error happened */
 			return err;
 
-		/*
-		 * GC could not make any progress. If this is the first time,
-		 * then it makes sense to try to commit, because it might make
-		 * some dirty space.
-		 */
-		dbg_budg("GC returned -ENOSPC, retries %d",
-			 ri->nospc_retries);
-		if (ri->nospc_retries >= MAX_NOSPC_RETRIES)
-			return err;
-		ri->nospc_retries += 1;
-	}
-
-	/* Neither GC nor write-back helped, try to commit */
-	if (ri->cmt_retries < MAX_CMT_RETRIES) {
-		ri->cmt_retries += 1;
-		dbg_budg("run commit, retries %d of %d",
-			 ri->cmt_retries, MAX_CMT_RETRIES);
+		dbg_budg("Run commit (retries %d)", retries);
 		err = ubifs_run_commit(c);
 		if (err)
 			return err;
-		return -EAGAIN;
-	}
+	} while (retries++ < MAX_MKSPC_RETRIES);
+
 	return -ENOSPC;
 }
 
@@ -258,8 +202,8 @@
  */
 int ubifs_calc_min_idx_lebs(struct ubifs_info *c)
 {
-	int ret;
-	uint64_t idx_size;
+	int idx_lebs, eff_leb_size = c->leb_size - c->max_idx_node_sz;
+	long long idx_size;
 
 	idx_size = c->old_idx_sz + c->budg_idx_growth + c->budg_uncommitted_idx;
 
@@ -271,23 +215,16 @@
 	 * pair, nor similarly the two variables for the new index size, so we
 	 * have to do this costly 64-bit division on fast-path.
 	 */
-	if (do_div(idx_size, c->leb_size - c->max_idx_node_sz))
-		ret = idx_size + 1;
-	else
-		ret = idx_size;
+	idx_size += eff_leb_size - 1;
+	idx_lebs = div_u64(idx_size, eff_leb_size);
 	/*
 	 * The index head is not available for the in-the-gaps method, so add an
 	 * extra LEB to compensate.
 	 */
-	ret += 1;
-	/*
-	 * At present the index needs at least 2 LEBs: one for the index head
-	 * and one for in-the-gaps method (which currently does not cater for
-	 * the index head and so excludes it from consideration).
-	 */
-	if (ret < 2)
-		ret = 2;
-	return ret;
+	idx_lebs += 1;
+	if (idx_lebs < MIN_INDEX_LEBS)
+		idx_lebs = MIN_INDEX_LEBS;
+	return idx_lebs;
 }
 
 /**
@@ -530,8 +467,7 @@
 int ubifs_budget_space(struct ubifs_info *c, struct ubifs_budget_req *req)
 {
 	int uninitialized_var(cmt_retries), uninitialized_var(wb_retries);
-	int err, idx_growth, data_growth, dd_growth;
-	struct retries_info ri;
+	int err, idx_growth, data_growth, dd_growth, retried = 0;
 
 	ubifs_assert(req->new_page <= 1);
 	ubifs_assert(req->dirtied_page <= 1);
@@ -549,7 +485,6 @@
 	if (!data_growth && !dd_growth)
 		return 0;
 	idx_growth = calc_idx_growth(c, req);
-	memset(&ri, 0, sizeof(struct retries_info));
 
 again:
 	spin_lock(&c->space_lock);
@@ -587,12 +522,17 @@
 		return err;
 	}
 
-	err = make_free_space(c, &ri);
+	err = make_free_space(c);
+	cond_resched();
 	if (err == -EAGAIN) {
 		dbg_budg("try again");
-		cond_resched();
 		goto again;
 	} else if (err == -ENOSPC) {
+		if (!retried) {
+			retried = 1;
+			dbg_budg("-ENOSPC, but anyway try once again");
+			goto again;
+		}
 		dbg_budg("FS is full, -ENOSPC");
 		c->nospace = 1;
 		if (can_use_rp(c) || c->rp_size == 0)
@@ -712,9 +652,9 @@
  * user-space. User-space application tend to expect that if the file-system
  * (e.g., via the 'statfs()' call) reports that it has N bytes available, they
  * are able to write a file of size N. UBIFS attaches node headers to each data
- * node and it has to write indexind nodes as well. This introduces additional
- * overhead, and UBIFS it has to report sligtly less free space to meet the
- * above expectetion.
+ * node and it has to write indexing nodes as well. This introduces additional
+ * overhead, and UBIFS has to report slightly less free space to meet the above
+ * expectations.
  *
  * This function assumes free space is made up of uncompressed data nodes and
  * full index nodes (one per data node, tripled because we always allow enough
@@ -723,7 +663,7 @@
  * Note, the calculation is pessimistic, which means that most of the time
  * UBIFS reports less space than it actually has.
  */
-long long ubifs_reported_space(const struct ubifs_info *c, uint64_t free)
+long long ubifs_reported_space(const struct ubifs_info *c, long long free)
 {
 	int divisor, factor, f;
 
@@ -737,7 +677,7 @@
 	 * of data nodes, f - fanout. Because effective UBIFS fanout is twice
 	 * as less than maximum fanout, we assume that each data node
 	 * introduces 3 * @c->max_idx_node_sz / (@c->fanout/2 - 1) bytes.
-	 * Note, the multiplier 3 is because UBIFS reseves thrice as more space
+	 * Note, the multiplier 3 is because UBIFS reserves thrice as more space
 	 * for the index.
 	 */
 	f = c->fanout > 3 ? c->fanout >> 1 : 2;
@@ -745,8 +685,7 @@
 	divisor = UBIFS_MAX_DATA_NODE_SZ;
 	divisor += (c->max_idx_node_sz * 3) / (f - 1);
 	free *= factor;
-	do_div(free, divisor);
-	return free;
+	return div_u64(free, divisor);
 }
 
 /**
@@ -756,10 +695,10 @@
  * This function calculates amount of free space to report to user-space.
  *
  * Because UBIFS may introduce substantial overhead (the index, node headers,
- * alighment, wastage at the end of eraseblocks, etc), it cannot report real
+ * alignment, wastage at the end of eraseblocks, etc), it cannot report real
  * amount of free flash space it has (well, because not all dirty space is
- * reclamable, UBIFS does not actually know the real amount). If UBIFS did so,
- * it would bread user expectetion about what free space is. Users seem to
+ * reclaimable, UBIFS does not actually know the real amount). If UBIFS did so,
+ * it would bread user expectations about what free space is. Users seem to
  * accustomed to assume that if the file-system reports N bytes of free space,
  * they would be able to fit a file of N bytes to the FS. This almost works for
  * traditional file-systems, because they have way less overhead than UBIFS.
@@ -771,18 +710,9 @@
 	long long available, outstanding, free;
 
 	spin_lock(&c->space_lock);
-	min_idx_lebs = ubifs_calc_min_idx_lebs(c);
+	min_idx_lebs = c->min_idx_lebs;
+	ubifs_assert(min_idx_lebs == ubifs_calc_min_idx_lebs(c));
 	outstanding = c->budg_data_growth + c->budg_dd_growth;
-
-	/*
-	 * Force the amount available to the total size reported if the used
-	 * space is zero.
-	 */
-	if (c->lst.total_used <= UBIFS_INO_NODE_SZ && !outstanding) {
-		spin_unlock(&c->space_lock);
-		return (long long)c->block_cnt << UBIFS_BLOCK_SHIFT;
-	}
-
 	available = ubifs_calc_available(c, min_idx_lebs);
 
 	/*
diff --git a/fs/ubifs/commit.c b/fs/ubifs/commit.c
index b49884c..f3a7945 100644
--- a/fs/ubifs/commit.c
+++ b/fs/ubifs/commit.c
@@ -470,12 +470,12 @@
 {
 	struct ubifs_idx_node *idx;
 	int lnum, offs, len, err = 0;
+	struct ubifs_debug_info *d = c->dbg;
 
-	c->old_zroot = *zroot;
-
-	lnum = c->old_zroot.lnum;
-	offs = c->old_zroot.offs;
-	len = c->old_zroot.len;
+	d->old_zroot = *zroot;
+	lnum = d->old_zroot.lnum;
+	offs = d->old_zroot.offs;
+	len = d->old_zroot.len;
 
 	idx = kmalloc(c->max_idx_node_sz, GFP_NOFS);
 	if (!idx)
@@ -485,8 +485,8 @@
 	if (err)
 		goto out;
 
-	c->old_zroot_level = le16_to_cpu(idx->level);
-	c->old_zroot_sqnum = le64_to_cpu(idx->ch.sqnum);
+	d->old_zroot_level = le16_to_cpu(idx->level);
+	d->old_zroot_sqnum = le64_to_cpu(idx->ch.sqnum);
 out:
 	kfree(idx);
 	return err;
@@ -509,6 +509,7 @@
 {
 	int lnum, offs, len, err = 0, uninitialized_var(last_level), child_cnt;
 	int first = 1, iip;
+	struct ubifs_debug_info *d = c->dbg;
 	union ubifs_key lower_key, upper_key, l_key, u_key;
 	unsigned long long uninitialized_var(last_sqnum);
 	struct ubifs_idx_node *idx;
@@ -525,9 +526,9 @@
 	     UBIFS_IDX_NODE_SZ;
 
 	/* Start at the old zroot */
-	lnum = c->old_zroot.lnum;
-	offs = c->old_zroot.offs;
-	len = c->old_zroot.len;
+	lnum = d->old_zroot.lnum;
+	offs = d->old_zroot.offs;
+	len = d->old_zroot.len;
 	iip = 0;
 
 	/*
@@ -560,11 +561,11 @@
 		if (first) {
 			first = 0;
 			/* Check root level and sqnum */
-			if (le16_to_cpu(idx->level) != c->old_zroot_level) {
+			if (le16_to_cpu(idx->level) != d->old_zroot_level) {
 				err = 2;
 				goto out_dump;
 			}
-			if (le64_to_cpu(idx->ch.sqnum) != c->old_zroot_sqnum) {
+			if (le64_to_cpu(idx->ch.sqnum) != d->old_zroot_sqnum) {
 				err = 3;
 				goto out_dump;
 			}
diff --git a/fs/ubifs/compress.c b/fs/ubifs/compress.c
index a0ada59..11e4132 100644
--- a/fs/ubifs/compress.c
+++ b/fs/ubifs/compress.c
@@ -33,7 +33,7 @@
 /* Fake description object for the "none" compressor */
 static struct ubifs_compressor none_compr = {
 	.compr_type = UBIFS_COMPR_NONE,
-	.name = "no compression",
+	.name = "none",
 	.capi_name = "",
 };
 
@@ -43,13 +43,13 @@
 static struct ubifs_compressor lzo_compr = {
 	.compr_type = UBIFS_COMPR_LZO,
 	.comp_mutex = &lzo_mutex,
-	.name = "LZO",
+	.name = "lzo",
 	.capi_name = "lzo",
 };
 #else
 static struct ubifs_compressor lzo_compr = {
 	.compr_type = UBIFS_COMPR_LZO,
-	.name = "LZO",
+	.name = "lzo",
 };
 #endif
 
@@ -108,7 +108,7 @@
 	if (compr->comp_mutex)
 		mutex_lock(compr->comp_mutex);
 	err = crypto_comp_compress(compr->cc, in_buf, in_len, out_buf,
-				   out_len);
+				   (unsigned int *)out_len);
 	if (compr->comp_mutex)
 		mutex_unlock(compr->comp_mutex);
 	if (unlikely(err)) {
@@ -119,10 +119,10 @@
 	}
 
 	/*
-	 * Presently, we just require that compression results in less data,
-	 * rather than any defined minimum compression ratio or amount.
+	 * If the data compressed only slightly, it is better to leave it
+	 * uncompressed to improve read speed.
 	 */
-	if (ALIGN(*out_len, 8) >= ALIGN(in_len, 8))
+	if (in_len - *out_len < UBIFS_MIN_COMPRESS_DIFF)
 		goto no_compr;
 
 	return;
@@ -172,7 +172,7 @@
 	if (compr->decomp_mutex)
 		mutex_lock(compr->decomp_mutex);
 	err = crypto_comp_decompress(compr->cc, in_buf, in_len, out_buf,
-				     out_len);
+				     (unsigned int *)out_len);
 	if (compr->decomp_mutex)
 		mutex_unlock(compr->decomp_mutex);
 	if (err)
@@ -244,7 +244,7 @@
 /**
  * ubifs_compressors_exit - de-initialize UBIFS compressors.
  */
-void __exit ubifs_compressors_exit(void)
+void ubifs_compressors_exit(void)
 {
 	compr_exit(&lzo_compr);
 	compr_exit(&zlib_compr);
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index 510ffa0..792c5a1 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -32,6 +32,8 @@
 #include "ubifs.h"
 #include <linux/module.h>
 #include <linux/moduleparam.h>
+#include <linux/debugfs.h>
+#include <linux/math64.h>
 
 #ifdef CONFIG_UBIFS_FS_DEBUG
 
@@ -596,7 +598,9 @@
 	struct rb_node *rb;
 	struct ubifs_bud *bud;
 	struct ubifs_gced_idx_leb *idx_gc;
+	long long available, outstanding, free;
 
+	ubifs_assert(spin_is_locked(&c->space_lock));
 	spin_lock(&dbg_lock);
 	printk(KERN_DEBUG "(pid %d) Budgeting info: budg_data_growth %lld, "
 	       "budg_dd_growth %lld, budg_idx_growth %lld\n", current->pid,
@@ -629,6 +633,17 @@
 		printk(KERN_DEBUG "\tGC'ed idx LEB %d unmap %d\n",
 		       idx_gc->lnum, idx_gc->unmap);
 	printk(KERN_DEBUG "\tcommit state %d\n", c->cmt_state);
+
+	/* Print budgeting predictions */
+	available = ubifs_calc_available(c, c->min_idx_lebs);
+	outstanding = c->budg_data_growth + c->budg_dd_growth;
+	if (available > outstanding)
+		free = ubifs_reported_space(c, available - outstanding);
+	else
+		free = 0;
+	printk(KERN_DEBUG "Budgeting predictions:\n");
+	printk(KERN_DEBUG "\tavailable: %lld, outstanding %lld, free %lld\n",
+	       available, outstanding, free);
 	spin_unlock(&dbg_lock);
 }
 
@@ -645,7 +660,8 @@
 	struct ubifs_lprops lp;
 	struct ubifs_lp_stats lst;
 
-	printk(KERN_DEBUG "(pid %d) Dumping LEB properties\n", current->pid);
+	printk(KERN_DEBUG "(pid %d) start dumping LEB properties\n",
+	       current->pid);
 	ubifs_get_lp_stats(c, &lst);
 	dbg_dump_lstats(&lst);
 
@@ -656,6 +672,8 @@
 
 		dbg_dump_lprop(c, &lp);
 	}
+	printk(KERN_DEBUG "(pid %d) finish dumping LEB properties\n",
+	       current->pid);
 }
 
 void dbg_dump_lpt_info(struct ubifs_info *c)
@@ -663,6 +681,7 @@
 	int i;
 
 	spin_lock(&dbg_lock);
+	printk(KERN_DEBUG "(pid %d) dumping LPT information\n", current->pid);
 	printk(KERN_DEBUG "\tlpt_sz:        %lld\n", c->lpt_sz);
 	printk(KERN_DEBUG "\tpnode_sz:      %d\n", c->pnode_sz);
 	printk(KERN_DEBUG "\tnnode_sz:      %d\n", c->nnode_sz);
@@ -684,7 +703,8 @@
 	printk(KERN_DEBUG "\tLPT root is at %d:%d\n", c->lpt_lnum, c->lpt_offs);
 	printk(KERN_DEBUG "\tLPT head is at %d:%d\n",
 	       c->nhead_lnum, c->nhead_offs);
-	printk(KERN_DEBUG "\tLPT ltab is at %d:%d\n", c->ltab_lnum, c->ltab_offs);
+	printk(KERN_DEBUG "\tLPT ltab is at %d:%d\n",
+	       c->ltab_lnum, c->ltab_offs);
 	if (c->big_lpt)
 		printk(KERN_DEBUG "\tLPT lsave is at %d:%d\n",
 		       c->lsave_lnum, c->lsave_offs);
@@ -703,9 +723,9 @@
 	if (dbg_failure_mode)
 		return;
 
-	printk(KERN_DEBUG "(pid %d) Dumping LEB %d\n", current->pid, lnum);
-
-	sleb = ubifs_scan(c, lnum, 0, c->dbg_buf);
+	printk(KERN_DEBUG "(pid %d) start dumping LEB %d\n",
+	       current->pid, lnum);
+	sleb = ubifs_scan(c, lnum, 0, c->dbg->buf);
 	if (IS_ERR(sleb)) {
 		ubifs_err("scan error %d", (int)PTR_ERR(sleb));
 		return;
@@ -721,6 +741,8 @@
 		dbg_dump_node(c, snod->node);
 	}
 
+	printk(KERN_DEBUG "(pid %d) finish dumping LEB %d\n",
+	       current->pid, lnum);
 	ubifs_scan_destroy(sleb);
 	return;
 }
@@ -768,7 +790,7 @@
 {
 	int i;
 
-	printk(KERN_DEBUG "(pid %d) Dumping heap cat %d (%d elements)\n",
+	printk(KERN_DEBUG "(pid %d) start dumping heap cat %d (%d elements)\n",
 	       current->pid, cat, heap->cnt);
 	for (i = 0; i < heap->cnt; i++) {
 		struct ubifs_lprops *lprops = heap->arr[i];
@@ -777,6 +799,7 @@
 		       "flags %d\n", i, lprops->lnum, lprops->hpos,
 		       lprops->free, lprops->dirty, lprops->flags);
 	}
+	printk(KERN_DEBUG "(pid %d) finish dumping heap\n", current->pid);
 }
 
 void dbg_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
@@ -784,7 +807,7 @@
 {
 	int i;
 
-	printk(KERN_DEBUG "(pid %d) Dumping pnode:\n", current->pid);
+	printk(KERN_DEBUG "(pid %d) dumping pnode:\n", current->pid);
 	printk(KERN_DEBUG "\taddress %zx parent %zx cnext %zx\n",
 	       (size_t)pnode, (size_t)parent, (size_t)pnode->cnext);
 	printk(KERN_DEBUG "\tflags %lu iip %d level %d num %d\n",
@@ -803,7 +826,7 @@
 	int level;
 
 	printk(KERN_DEBUG "\n");
-	printk(KERN_DEBUG "(pid %d) Dumping the TNC tree\n", current->pid);
+	printk(KERN_DEBUG "(pid %d) start dumping TNC tree\n", current->pid);
 	znode = ubifs_tnc_levelorder_next(c->zroot.znode, NULL);
 	level = znode->level;
 	printk(KERN_DEBUG "== Level %d ==\n", level);
@@ -815,8 +838,7 @@
 		dbg_dump_znode(c, znode);
 		znode = ubifs_tnc_levelorder_next(c->zroot.znode, znode);
 	}
-
-	printk(KERN_DEBUG "\n");
+	printk(KERN_DEBUG "(pid %d) finish dumping TNC tree\n", current->pid);
 }
 
 static int dump_znode(struct ubifs_info *c, struct ubifs_znode *znode,
@@ -992,8 +1014,8 @@
 			zbr1->offs, DBGKEY(&key));
 		dbg_err("but it should have key %s according to tnc",
 			DBGKEY(&zbr1->key));
-			dbg_dump_node(c, dent1);
-			goto out_free;
+		dbg_dump_node(c, dent1);
+		goto out_free;
 	}
 
 	key_read(c, &dent2->key, &key);
@@ -1002,8 +1024,8 @@
 			zbr1->offs, DBGKEY(&key));
 		dbg_err("but it should have key %s according to tnc",
 			DBGKEY(&zbr2->key));
-			dbg_dump_node(c, dent2);
-			goto out_free;
+		dbg_dump_node(c, dent2);
+		goto out_free;
 	}
 
 	nlen1 = le16_to_cpu(dent1->nlen);
@@ -1020,9 +1042,9 @@
 		dbg_err("bad order of colliding key %s",
 			DBGKEY(&key));
 
-	dbg_msg("first node at %d:%d\n", zbr1->lnum, zbr1->offs);
+	ubifs_msg("first node at %d:%d\n", zbr1->lnum, zbr1->offs);
 	dbg_dump_node(c, dent1);
-	dbg_msg("second node at %d:%d\n", zbr2->lnum, zbr2->offs);
+	ubifs_msg("second node at %d:%d\n", zbr2->lnum, zbr2->offs);
 	dbg_dump_node(c, dent2);
 
 out_free:
@@ -2097,13 +2119,13 @@
 	return (next >> 16) & 32767;
 }
 
-void dbg_failure_mode_registration(struct ubifs_info *c)
+static void failure_mode_init(struct ubifs_info *c)
 {
 	struct failure_mode_info *fmi;
 
 	fmi = kmalloc(sizeof(struct failure_mode_info), GFP_NOFS);
 	if (!fmi) {
-		dbg_err("Failed to register failure mode - no memory");
+		ubifs_err("Failed to register failure mode - no memory");
 		return;
 	}
 	fmi->c = c;
@@ -2112,7 +2134,7 @@
 	spin_unlock(&fmi_lock);
 }
 
-void dbg_failure_mode_deregistration(struct ubifs_info *c)
+static void failure_mode_exit(struct ubifs_info *c)
 {
 	struct failure_mode_info *fmi, *tmp;
 
@@ -2146,42 +2168,44 @@
 	struct ubifs_info *c = dbg_find_info(desc);
 
 	if (c && dbg_failure_mode)
-		return c->failure_mode;
+		return c->dbg->failure_mode;
 	return 0;
 }
 
 static int do_fail(struct ubi_volume_desc *desc, int lnum, int write)
 {
 	struct ubifs_info *c = dbg_find_info(desc);
+	struct ubifs_debug_info *d;
 
 	if (!c || !dbg_failure_mode)
 		return 0;
-	if (c->failure_mode)
+	d = c->dbg;
+	if (d->failure_mode)
 		return 1;
-	if (!c->fail_cnt) {
+	if (!d->fail_cnt) {
 		/* First call - decide delay to failure */
 		if (chance(1, 2)) {
 			unsigned int delay = 1 << (simple_rand() >> 11);
 
 			if (chance(1, 2)) {
-				c->fail_delay = 1;
-				c->fail_timeout = jiffies +
+				d->fail_delay = 1;
+				d->fail_timeout = jiffies +
 						  msecs_to_jiffies(delay);
 				dbg_rcvry("failing after %ums", delay);
 			} else {
-				c->fail_delay = 2;
-				c->fail_cnt_max = delay;
+				d->fail_delay = 2;
+				d->fail_cnt_max = delay;
 				dbg_rcvry("failing after %u calls", delay);
 			}
 		}
-		c->fail_cnt += 1;
+		d->fail_cnt += 1;
 	}
 	/* Determine if failure delay has expired */
-	if (c->fail_delay == 1) {
-		if (time_before(jiffies, c->fail_timeout))
+	if (d->fail_delay == 1) {
+		if (time_before(jiffies, d->fail_timeout))
 			return 0;
-	} else if (c->fail_delay == 2)
-		if (c->fail_cnt++ < c->fail_cnt_max)
+	} else if (d->fail_delay == 2)
+		if (d->fail_cnt++ < d->fail_cnt_max)
 			return 0;
 	if (lnum == UBIFS_SB_LNUM) {
 		if (write) {
@@ -2239,7 +2263,7 @@
 		dbg_rcvry("failing in bud LEB %d commit not running", lnum);
 	}
 	ubifs_err("*** SETTING FAILURE MODE ON (LEB %d) ***", lnum);
-	c->failure_mode = 1;
+	d->failure_mode = 1;
 	dump_stack();
 	return 1;
 }
@@ -2344,4 +2368,181 @@
 	return 0;
 }
 
+/**
+ * ubifs_debugging_init - initialize UBIFS debugging.
+ * @c: UBIFS file-system description object
+ *
+ * This function initializes debugging-related data for the file system.
+ * Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+int ubifs_debugging_init(struct ubifs_info *c)
+{
+	c->dbg = kzalloc(sizeof(struct ubifs_debug_info), GFP_KERNEL);
+	if (!c->dbg)
+		return -ENOMEM;
+
+	c->dbg->buf = vmalloc(c->leb_size);
+	if (!c->dbg->buf)
+		goto out;
+
+	failure_mode_init(c);
+	return 0;
+
+out:
+	kfree(c->dbg);
+	return -ENOMEM;
+}
+
+/**
+ * ubifs_debugging_exit - free debugging data.
+ * @c: UBIFS file-system description object
+ */
+void ubifs_debugging_exit(struct ubifs_info *c)
+{
+	failure_mode_exit(c);
+	vfree(c->dbg->buf);
+	kfree(c->dbg);
+}
+
+/*
+ * Root directory for UBIFS stuff in debugfs. Contains sub-directories which
+ * contain the stuff specific to particular file-system mounts.
+ */
+static struct dentry *debugfs_rootdir;
+
+/**
+ * dbg_debugfs_init - initialize debugfs file-system.
+ *
+ * UBIFS uses debugfs file-system to expose various debugging knobs to
+ * user-space. This function creates "ubifs" directory in the debugfs
+ * file-system. Returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+int dbg_debugfs_init(void)
+{
+	debugfs_rootdir = debugfs_create_dir("ubifs", NULL);
+	if (IS_ERR(debugfs_rootdir)) {
+		int err = PTR_ERR(debugfs_rootdir);
+		ubifs_err("cannot create \"ubifs\" debugfs directory, "
+			  "error %d\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * dbg_debugfs_exit - remove the "ubifs" directory from debugfs file-system.
+ */
+void dbg_debugfs_exit(void)
+{
+	debugfs_remove(debugfs_rootdir);
+}
+
+static int open_debugfs_file(struct inode *inode, struct file *file)
+{
+	file->private_data = inode->i_private;
+	return 0;
+}
+
+static ssize_t write_debugfs_file(struct file *file, const char __user *buf,
+				  size_t count, loff_t *ppos)
+{
+	struct ubifs_info *c = file->private_data;
+	struct ubifs_debug_info *d = c->dbg;
+
+	if (file->f_path.dentry == d->dump_lprops)
+		dbg_dump_lprops(c);
+	else if (file->f_path.dentry == d->dump_budg) {
+		spin_lock(&c->space_lock);
+		dbg_dump_budg(c);
+		spin_unlock(&c->space_lock);
+	} else if (file->f_path.dentry == d->dump_tnc) {
+		mutex_lock(&c->tnc_mutex);
+		dbg_dump_tnc(c);
+		mutex_unlock(&c->tnc_mutex);
+	} else
+		return -EINVAL;
+
+	*ppos += count;
+	return count;
+}
+
+static const struct file_operations debugfs_fops = {
+	.open = open_debugfs_file,
+	.write = write_debugfs_file,
+	.owner = THIS_MODULE,
+};
+
+/**
+ * dbg_debugfs_init_fs - initialize debugfs for UBIFS instance.
+ * @c: UBIFS file-system description object
+ *
+ * This function creates all debugfs files for this instance of UBIFS. Returns
+ * zero in case of success and a negative error code in case of failure.
+ *
+ * Note, the only reason we have not merged this function with the
+ * 'ubifs_debugging_init()' function is because it is better to initialize
+ * debugfs interfaces at the very end of the mount process, and remove them at
+ * the very beginning of the mount process.
+ */
+int dbg_debugfs_init_fs(struct ubifs_info *c)
+{
+	int err;
+	const char *fname;
+	struct dentry *dent;
+	struct ubifs_debug_info *d = c->dbg;
+
+	sprintf(d->debugfs_dir_name, "ubi%d_%d", c->vi.ubi_num, c->vi.vol_id);
+	d->debugfs_dir = debugfs_create_dir(d->debugfs_dir_name,
+					      debugfs_rootdir);
+	if (IS_ERR(d->debugfs_dir)) {
+		err = PTR_ERR(d->debugfs_dir);
+		ubifs_err("cannot create \"%s\" debugfs directory, error %d\n",
+			  d->debugfs_dir_name, err);
+		goto out;
+	}
+
+	fname = "dump_lprops";
+	dent = debugfs_create_file(fname, S_IWUGO, d->debugfs_dir, c,
+				   &debugfs_fops);
+	if (IS_ERR(dent))
+		goto out_remove;
+	d->dump_lprops = dent;
+
+	fname = "dump_budg";
+	dent = debugfs_create_file(fname, S_IWUGO, d->debugfs_dir, c,
+				   &debugfs_fops);
+	if (IS_ERR(dent))
+		goto out_remove;
+	d->dump_budg = dent;
+
+	fname = "dump_tnc";
+	dent = debugfs_create_file(fname, S_IWUGO, d->debugfs_dir, c,
+				   &debugfs_fops);
+	if (IS_ERR(dent))
+		goto out_remove;
+	d->dump_tnc = dent;
+
+	return 0;
+
+out_remove:
+	err = PTR_ERR(dent);
+	ubifs_err("cannot create \"%s\" debugfs directory, error %d\n",
+		  fname, err);
+	debugfs_remove_recursive(d->debugfs_dir);
+out:
+	return err;
+}
+
+/**
+ * dbg_debugfs_exit_fs - remove all debugfs files.
+ * @c: UBIFS file-system description object
+ */
+void dbg_debugfs_exit_fs(struct ubifs_info *c)
+{
+	debugfs_remove_recursive(c->dbg->debugfs_dir);
+}
+
 #endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h
index 33d6b95..9820d69 100644
--- a/fs/ubifs/debug.h
+++ b/fs/ubifs/debug.h
@@ -25,7 +25,56 @@
 
 #ifdef CONFIG_UBIFS_FS_DEBUG
 
-#define UBIFS_DBG(op) op
+/**
+ * ubifs_debug_info - per-FS debugging information.
+ * @buf: a buffer of LEB size, used for various purposes
+ * @old_zroot: old index root - used by 'dbg_check_old_index()'
+ * @old_zroot_level: old index root level - used by 'dbg_check_old_index()'
+ * @old_zroot_sqnum: old index root sqnum - used by 'dbg_check_old_index()'
+ * @failure_mode: failure mode for recovery testing
+ * @fail_delay: 0=>don't delay, 1=>delay a time, 2=>delay a number of calls
+ * @fail_timeout: time in jiffies when delay of failure mode expires
+ * @fail_cnt: current number of calls to failure mode I/O functions
+ * @fail_cnt_max: number of calls by which to delay failure mode
+ * @chk_lpt_sz: used by LPT tree size checker
+ * @chk_lpt_sz2: used by LPT tree size checker
+ * @chk_lpt_wastage: used by LPT tree size checker
+ * @chk_lpt_lebs: used by LPT tree size checker
+ * @new_nhead_offs: used by LPT tree size checker
+ * @new_ihead_lnum: used by debugging to check ihead_lnum
+ * @new_ihead_offs: used by debugging to check ihead_offs
+ *
+ * debugfs_dir_name: name of debugfs directory containing this file-system's
+ *                   files
+ * debugfs_dir: direntry object of the file-system debugfs directory
+ * dump_lprops: "dump lprops" debugfs knob
+ * dump_budg: "dump budgeting information" debugfs knob
+ * dump_tnc: "dump TNC" debugfs knob
+ */
+struct ubifs_debug_info {
+	void *buf;
+	struct ubifs_zbranch old_zroot;
+	int old_zroot_level;
+	unsigned long long old_zroot_sqnum;
+	int failure_mode;
+	int fail_delay;
+	unsigned long fail_timeout;
+	unsigned int fail_cnt;
+	unsigned int fail_cnt_max;
+	long long chk_lpt_sz;
+	long long chk_lpt_sz2;
+	long long chk_lpt_wastage;
+	int chk_lpt_lebs;
+	int new_nhead_offs;
+	int new_ihead_lnum;
+	int new_ihead_offs;
+
+	char debugfs_dir_name[100];
+	struct dentry *debugfs_dir;
+	struct dentry *dump_lprops;
+	struct dentry *dump_budg;
+	struct dentry *dump_tnc;
+};
 
 #define ubifs_assert(expr) do {                                                \
 	if (unlikely(!(expr))) {                                               \
@@ -211,14 +260,18 @@
 extern unsigned int ubifs_chk_flags;
 extern unsigned int ubifs_tst_flags;
 
-/* Dump functions */
+int ubifs_debugging_init(struct ubifs_info *c);
+void ubifs_debugging_exit(struct ubifs_info *c);
 
+/* Dump functions */
 const char *dbg_ntype(int type);
 const char *dbg_cstate(int cmt_state);
 const char *dbg_get_key_dump(const struct ubifs_info *c,
 			     const union ubifs_key *key);
 void dbg_dump_inode(const struct ubifs_info *c, const struct inode *inode);
 void dbg_dump_node(const struct ubifs_info *c, const void *node);
+void dbg_dump_lpt_node(const struct ubifs_info *c, void *node, int lnum,
+		       int offs);
 void dbg_dump_budget_req(const struct ubifs_budget_req *req);
 void dbg_dump_lstats(const struct ubifs_lp_stats *lst);
 void dbg_dump_budg(struct ubifs_info *c);
@@ -233,9 +286,9 @@
 		    struct ubifs_nnode *parent, int iip);
 void dbg_dump_tnc(struct ubifs_info *c);
 void dbg_dump_index(struct ubifs_info *c);
+void dbg_dump_lpt_lebs(const struct ubifs_info *c);
 
 /* Checking helper functions */
-
 typedef int (*dbg_leaf_callback)(struct ubifs_info *c,
 				 struct ubifs_zbranch *zbr, void *priv);
 typedef int (*dbg_znode_callback)(struct ubifs_info *c,
@@ -274,9 +327,6 @@
 
 #define dbg_failure_mode (ubifs_tst_flags & UBIFS_TST_RCVRY)
 
-void dbg_failure_mode_registration(struct ubifs_info *c);
-void dbg_failure_mode_deregistration(struct ubifs_info *c);
-
 #ifndef UBIFS_DBG_PRESERVE_UBI
 
 #define ubi_leb_read   dbg_leb_read
@@ -318,9 +368,13 @@
 	return dbg_leb_change(desc, lnum, buf, len, UBI_UNKNOWN);
 }
 
-#else /* !CONFIG_UBIFS_FS_DEBUG */
+/* Debugfs-related stuff */
+int dbg_debugfs_init(void);
+void dbg_debugfs_exit(void);
+int dbg_debugfs_init_fs(struct ubifs_info *c);
+void dbg_debugfs_exit_fs(struct ubifs_info *c);
 
-#define UBIFS_DBG(op)
+#else /* !CONFIG_UBIFS_FS_DEBUG */
 
 /* Use "if (0)" to make compiler check arguments even if debugging is off */
 #define ubifs_assert(expr)  do {                                               \
@@ -360,23 +414,28 @@
 #define DBGKEY(key)  ((char *)(key))
 #define DBGKEY1(key) ((char *)(key))
 
-#define dbg_ntype(type)                       ""
-#define dbg_cstate(cmt_state)                 ""
-#define dbg_get_key_dump(c, key)              ({})
-#define dbg_dump_inode(c, inode)              ({})
-#define dbg_dump_node(c, node)                ({})
-#define dbg_dump_budget_req(req)              ({})
-#define dbg_dump_lstats(lst)                  ({})
-#define dbg_dump_budg(c)                      ({})
-#define dbg_dump_lprop(c, lp)                 ({})
-#define dbg_dump_lprops(c)                    ({})
-#define dbg_dump_lpt_info(c)                  ({})
-#define dbg_dump_leb(c, lnum)                 ({})
-#define dbg_dump_znode(c, znode)              ({})
-#define dbg_dump_heap(c, heap, cat)           ({})
-#define dbg_dump_pnode(c, pnode, parent, iip) ({})
-#define dbg_dump_tnc(c)                       ({})
-#define dbg_dump_index(c)                     ({})
+#define ubifs_debugging_init(c)                0
+#define ubifs_debugging_exit(c)                ({})
+
+#define dbg_ntype(type)                        ""
+#define dbg_cstate(cmt_state)                  ""
+#define dbg_get_key_dump(c, key)               ({})
+#define dbg_dump_inode(c, inode)               ({})
+#define dbg_dump_node(c, node)                 ({})
+#define dbg_dump_lpt_node(c, node, lnum, offs) ({})
+#define dbg_dump_budget_req(req)               ({})
+#define dbg_dump_lstats(lst)                   ({})
+#define dbg_dump_budg(c)                       ({})
+#define dbg_dump_lprop(c, lp)                  ({})
+#define dbg_dump_lprops(c)                     ({})
+#define dbg_dump_lpt_info(c)                   ({})
+#define dbg_dump_leb(c, lnum)                  ({})
+#define dbg_dump_znode(c, znode)               ({})
+#define dbg_dump_heap(c, heap, cat)            ({})
+#define dbg_dump_pnode(c, pnode, parent, iip)  ({})
+#define dbg_dump_tnc(c)                        ({})
+#define dbg_dump_index(c)                      ({})
+#define dbg_dump_lpt_lebs(c)                   ({})
 
 #define dbg_walk_index(c, leaf_cb, znode_cb, priv) 0
 #define dbg_old_index_check_init(c, zroot)         0
@@ -396,9 +455,11 @@
 #define dbg_force_in_the_gaps_enabled              0
 #define dbg_force_in_the_gaps()                    0
 #define dbg_failure_mode                           0
-#define dbg_failure_mode_registration(c)           ({})
-#define dbg_failure_mode_deregistration(c)         ({})
+
+#define dbg_debugfs_init()                         0
+#define dbg_debugfs_exit()
+#define dbg_debugfs_init_fs(c)                     0
+#define dbg_debugfs_exit_fs(c)                     0
 
 #endif /* !CONFIG_UBIFS_FS_DEBUG */
-
 #endif /* !__UBIFS_DEBUG_H__ */
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 2624411..fe82d24 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -72,8 +72,8 @@
 		return err;
 	}
 
-	ubifs_assert(le64_to_cpu(dn->ch.sqnum) > ubifs_inode(inode)->creat_sqnum);
-
+	ubifs_assert(le64_to_cpu(dn->ch.sqnum) >
+		     ubifs_inode(inode)->creat_sqnum);
 	len = le32_to_cpu(dn->size);
 	if (len <= 0 || len > UBIFS_BLOCK_SIZE)
 		goto dump;
@@ -254,7 +254,7 @@
 	}
 
 	if (!PageUptodate(page)) {
-		if (!(pos & PAGE_CACHE_MASK) && len == PAGE_CACHE_SIZE)
+		if (!(pos & ~PAGE_CACHE_MASK) && len == PAGE_CACHE_SIZE)
 			SetPageChecked(page);
 		else {
 			err = do_readpage(page);
@@ -444,7 +444,7 @@
 
 	if (!PageUptodate(page)) {
 		/* The page is not loaded from the flash */
-		if (!(pos & PAGE_CACHE_MASK) && len == PAGE_CACHE_SIZE)
+		if (!(pos & ~PAGE_CACHE_MASK) && len == PAGE_CACHE_SIZE)
 			/*
 			 * We change whole page so no need to load it. But we
 			 * have to set the @PG_checked flag to make the further
diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c
index 5e82cff..6db7a6b 100644
--- a/fs/ubifs/ioctl.c
+++ b/fs/ubifs/ioctl.c
@@ -154,6 +154,7 @@
 	case FS_IOC_GETFLAGS:
 		flags = ubifs2ioctl(ubifs_inode(inode)->flags);
 
+		dbg_gen("get flags: %#x, i_flags %#x", flags, inode->i_flags);
 		return put_user(flags, (int __user *) arg);
 
 	case FS_IOC_SETFLAGS: {
@@ -176,6 +177,7 @@
 		err = mnt_want_write(file->f_path.mnt);
 		if (err)
 			return err;
+		dbg_gen("set flags: %#x, i_flags %#x", flags, inode->i_flags);
 		err = setflags(inode, flags);
 		mnt_drop_write(file->f_path.mnt);
 		return err;
diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c
index f91b745..10ae25b 100644
--- a/fs/ubifs/journal.c
+++ b/fs/ubifs/journal.c
@@ -704,7 +704,7 @@
 	data->size = cpu_to_le32(len);
 	zero_data_node_unused(data);
 
-	if (!(ui->flags && UBIFS_COMPR_FL))
+	if (!(ui->flags & UBIFS_COMPR_FL))
 		/* Compression is disabled for this inode */
 		compr_type = UBIFS_COMPR_NONE;
 	else
@@ -1220,7 +1220,7 @@
 	data_key_init(c, &key, inum, blk);
 
 	bit = old_size & (UBIFS_BLOCK_SIZE - 1);
-	blk = (old_size >> UBIFS_BLOCK_SHIFT) - (bit ? 0: 1);
+	blk = (old_size >> UBIFS_BLOCK_SHIFT) - (bit ? 0 : 1);
 	data_key_init(c, &to_key, inum, blk);
 
 	err = ubifs_tnc_remove_range(c, &key, &to_key);
diff --git a/fs/ubifs/key.h b/fs/ubifs/key.h
index 3f1f16b..efb3430 100644
--- a/fs/ubifs/key.h
+++ b/fs/ubifs/key.h
@@ -38,6 +38,22 @@
 #define __UBIFS_KEY_H__
 
 /**
+ * key_mask_hash - mask a valid hash value.
+ * @val: value to be masked
+ *
+ * We use hash values as offset in directories, so values %0 and %1 are
+ * reserved for "." and "..". %2 is reserved for "end of readdir" marker. This
+ * function makes sure the reserved values are not used.
+ */
+static inline uint32_t key_mask_hash(uint32_t hash)
+{
+	hash &= UBIFS_S_KEY_HASH_MASK;
+	if (unlikely(hash <= 2))
+		hash += 3;
+	return hash;
+}
+
+/**
  * key_r5_hash - R5 hash function (borrowed from reiserfs).
  * @s: direntry name
  * @len: name length
@@ -54,16 +70,7 @@
 		str++;
 	}
 
-	a &= UBIFS_S_KEY_HASH_MASK;
-
-	/*
-	 * We use hash values as offset in directories, so values %0 and %1 are
-	 * reserved for "." and "..". %2 is reserved for "end of readdir"
-	 * marker.
-	 */
-	if (unlikely(a >= 0 && a <= 2))
-		a += 3;
-	return a;
+	return key_mask_hash(a);
 }
 
 /**
@@ -77,10 +84,7 @@
 
 	len = min_t(uint32_t, len, 4);
 	memcpy(&a, str, len);
-	a &= UBIFS_S_KEY_HASH_MASK;
-	if (unlikely(a >= 0 && a <= 2))
-		a += 3;
-	return a;
+	return key_mask_hash(a);
 }
 
 /**
diff --git a/fs/ubifs/lprops.c b/fs/ubifs/lprops.c
index f27176e..dfd2bce 100644
--- a/fs/ubifs/lprops.c
+++ b/fs/ubifs/lprops.c
@@ -520,13 +520,13 @@
  * @flags: new flags
  * @idx_gc_cnt: change to the count of idx_gc list
  *
- * This function changes LEB properties. This function does not change a LEB
- * property (@free, @dirty or @flag) if the value passed is %LPROPS_NC.
+ * This function changes LEB properties (@free, @dirty or @flag). However, the
+ * property which has the %LPROPS_NC value is not changed. Returns a pointer to
+ * the updated LEB properties on success and a negative error code on failure.
  *
- * This function returns a pointer to the updated LEB properties on success
- * and a negative error code on failure. N.B. the LEB properties may have had to
- * be copied (due to COW) and consequently the pointer returned may not be the
- * same as the pointer passed.
+ * Note, the LEB properties may have had to be copied (due to COW) and
+ * consequently the pointer returned may not be the same as the pointer
+ * passed.
  */
 const struct ubifs_lprops *ubifs_change_lp(struct ubifs_info *c,
 					   const struct ubifs_lprops *lp,
@@ -1088,7 +1088,7 @@
 		}
 	}
 
-	sleb = ubifs_scan(c, lnum, 0, c->dbg_buf);
+	sleb = ubifs_scan(c, lnum, 0, c->dbg->buf);
 	if (IS_ERR(sleb)) {
 		/*
 		 * After an unclean unmount, empty and freeable LEBs
diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c
index db8bd0e..b2792e8 100644
--- a/fs/ubifs/lpt.c
+++ b/fs/ubifs/lpt.c
@@ -36,15 +36,16 @@
  * can be written into a single eraseblock. In that case, garbage collection
  * consists of just writing the whole table, which therefore makes all other
  * eraseblocks reusable. In the case of the big model, dirty eraseblocks are
- * selected for garbage collection, which consists are marking the nodes in
+ * selected for garbage collection, which consists of marking the clean nodes in
  * that LEB as dirty, and then only the dirty nodes are written out. Also, in
  * the case of the big model, a table of LEB numbers is saved so that the entire
  * LPT does not to be scanned looking for empty eraseblocks when UBIFS is first
  * mounted.
  */
 
-#include <linux/crc16.h>
 #include "ubifs.h"
+#include <linux/crc16.h>
+#include <linux/math64.h>
 
 /**
  * do_calc_lpt_geom - calculate sizes for the LPT area.
@@ -135,15 +136,13 @@
 int ubifs_calc_lpt_geom(struct ubifs_info *c)
 {
 	int lebs_needed;
-	uint64_t sz;
+	long long sz;
 
 	do_calc_lpt_geom(c);
 
 	/* Verify that lpt_lebs is big enough */
 	sz = c->lpt_sz * 2; /* Must have at least 2 times the size */
-	sz += c->leb_size - 1;
-	do_div(sz, c->leb_size);
-	lebs_needed = sz;
+	lebs_needed = div_u64(sz + c->leb_size - 1, c->leb_size);
 	if (lebs_needed > c->lpt_lebs) {
 		ubifs_err("too few LPT LEBs");
 		return -EINVAL;
@@ -156,7 +155,6 @@
 	}
 
 	c->check_lpt_free = c->big_lpt;
-
 	return 0;
 }
 
@@ -176,7 +174,7 @@
 			      int *big_lpt)
 {
 	int i, lebs_needed;
-	uint64_t sz;
+	long long sz;
 
 	/* Start by assuming the minimum number of LPT LEBs */
 	c->lpt_lebs = UBIFS_MIN_LPT_LEBS;
@@ -203,9 +201,7 @@
 	/* Now check there are enough LPT LEBs */
 	for (i = 0; i < 64 ; i++) {
 		sz = c->lpt_sz * 4; /* Allow 4 times the size */
-		sz += c->leb_size - 1;
-		do_div(sz, c->leb_size);
-		lebs_needed = sz;
+		lebs_needed = div_u64(sz + c->leb_size - 1, c->leb_size);
 		if (lebs_needed > c->lpt_lebs) {
 			/* Not enough LPT LEBs so try again with more */
 			c->lpt_lebs = lebs_needed;
@@ -558,7 +554,7 @@
  * This function calculates and returns the nnode number based on the parent's
  * nnode number and the index in parent.
  */
-static int calc_nnode_num_from_parent(struct ubifs_info *c,
+static int calc_nnode_num_from_parent(const struct ubifs_info *c,
 				      struct ubifs_nnode *parent, int iip)
 {
 	int num, shft;
@@ -583,7 +579,7 @@
  * This function calculates and returns the pnode number based on the parent's
  * nnode number and the index in parent.
  */
-static int calc_pnode_num_from_parent(struct ubifs_info *c,
+static int calc_pnode_num_from_parent(const struct ubifs_info *c,
 				      struct ubifs_nnode *parent, int iip)
 {
 	int i, n = c->lpt_hght - 1, pnum = parent->num, num = 0;
@@ -966,7 +962,7 @@
  *
  * This function returns %0 on success and a negative error code on failure.
  */
-static int unpack_pnode(struct ubifs_info *c, void *buf,
+static int unpack_pnode(const struct ubifs_info *c, void *buf,
 			struct ubifs_pnode *pnode)
 {
 	uint8_t *addr = buf + UBIFS_LPT_CRC_BYTES;
@@ -996,15 +992,15 @@
 }
 
 /**
- * unpack_nnode - unpack a nnode.
+ * ubifs_unpack_nnode - unpack a nnode.
  * @c: UBIFS file-system description object
  * @buf: buffer containing packed nnode to unpack
  * @nnode: nnode structure to fill
  *
  * This function returns %0 on success and a negative error code on failure.
  */
-static int unpack_nnode(struct ubifs_info *c, void *buf,
-			struct ubifs_nnode *nnode)
+int ubifs_unpack_nnode(const struct ubifs_info *c, void *buf,
+		       struct ubifs_nnode *nnode)
 {
 	uint8_t *addr = buf + UBIFS_LPT_CRC_BYTES;
 	int i, pos = 0, err;
@@ -1036,7 +1032,7 @@
  *
  * This function returns %0 on success and a negative error code on failure.
  */
-static int unpack_ltab(struct ubifs_info *c, void *buf)
+static int unpack_ltab(const struct ubifs_info *c, void *buf)
 {
 	uint8_t *addr = buf + UBIFS_LPT_CRC_BYTES;
 	int i, pos = 0, err;
@@ -1068,7 +1064,7 @@
  *
  * This function returns %0 on success and a negative error code on failure.
  */
-static int unpack_lsave(struct ubifs_info *c, void *buf)
+static int unpack_lsave(const struct ubifs_info *c, void *buf)
 {
 	uint8_t *addr = buf + UBIFS_LPT_CRC_BYTES;
 	int i, pos = 0, err;
@@ -1096,7 +1092,7 @@
  *
  * This function returns %0 on success and a negative error code on failure.
  */
-static int validate_nnode(struct ubifs_info *c, struct ubifs_nnode *nnode,
+static int validate_nnode(const struct ubifs_info *c, struct ubifs_nnode *nnode,
 			  struct ubifs_nnode *parent, int iip)
 {
 	int i, lvl, max_offs;
@@ -1140,7 +1136,7 @@
  *
  * This function returns %0 on success and a negative error code on failure.
  */
-static int validate_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
+static int validate_pnode(const struct ubifs_info *c, struct ubifs_pnode *pnode,
 			  struct ubifs_nnode *parent, int iip)
 {
 	int i;
@@ -1174,7 +1170,8 @@
  * This function calculates the LEB numbers for the LEB properties it contains
  * based on the pnode number.
  */
-static void set_pnode_lnum(struct ubifs_info *c, struct ubifs_pnode *pnode)
+static void set_pnode_lnum(const struct ubifs_info *c,
+			   struct ubifs_pnode *pnode)
 {
 	int i, lnum;
 
@@ -1227,7 +1224,7 @@
 		err = ubi_read(c->ubi, lnum, buf, offs, c->nnode_sz);
 		if (err)
 			goto out;
-		err = unpack_nnode(c, buf, nnode);
+		err = ubifs_unpack_nnode(c, buf, nnode);
 		if (err)
 			goto out;
 	}
@@ -1816,7 +1813,7 @@
 			       c->nnode_sz);
 		if (err)
 			return ERR_PTR(err);
-		err = unpack_nnode(c, buf, nnode);
+		err = ubifs_unpack_nnode(c, buf, nnode);
 		if (err)
 			return ERR_PTR(err);
 	}
diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c
index a41434b..96ca957 100644
--- a/fs/ubifs/lpt_commit.c
+++ b/fs/ubifs/lpt_commit.c
@@ -320,6 +320,8 @@
 	dbg_err("LPT out of space at LEB %d:%d needing %d, done_ltab %d, "
 		"done_lsave %d", lnum, offs, len, done_ltab, done_lsave);
 	dbg_dump_lpt_info(c);
+	dbg_dump_lpt_lebs(c);
+	dump_stack();
 	return err;
 }
 
@@ -546,8 +548,10 @@
 no_space:
 	ubifs_err("LPT out of space mismatch");
 	dbg_err("LPT out of space mismatch at LEB %d:%d needing %d, done_ltab "
-	        "%d, done_lsave %d", lnum, offs, len, done_ltab, done_lsave);
+		"%d, done_lsave %d", lnum, offs, len, done_ltab, done_lsave);
 	dbg_dump_lpt_info(c);
+	dbg_dump_lpt_lebs(c);
+	dump_stack();
 	return err;
 }
 
@@ -749,7 +753,7 @@
  * LPT trivial garbage collection is where a LPT LEB contains only dirty and
  * free space and so may be reused as soon as the next commit is completed.
  * This function is called after the commit is completed (master node has been
- * written) and unmaps LPT LEBs that were marked for trivial GC.
+ * written) and un-maps LPT LEBs that were marked for trivial GC.
  */
 static int lpt_tgc_end(struct ubifs_info *c)
 {
@@ -1025,7 +1029,7 @@
  * @c: UBIFS file-system description object
  * @node_type: LPT node type
  */
-static int get_lpt_node_len(struct ubifs_info *c, int node_type)
+static int get_lpt_node_len(const struct ubifs_info *c, int node_type)
 {
 	switch (node_type) {
 	case UBIFS_LPT_NNODE:
@@ -1046,7 +1050,7 @@
  * @buf: buffer
  * @len: length of buffer
  */
-static int get_pad_len(struct ubifs_info *c, uint8_t *buf, int len)
+static int get_pad_len(const struct ubifs_info *c, uint8_t *buf, int len)
 {
 	int offs, pad_len;
 
@@ -1063,7 +1067,8 @@
  * @buf: buffer
  * @node_num: node number is returned here
  */
-static int get_lpt_node_type(struct ubifs_info *c, uint8_t *buf, int *node_num)
+static int get_lpt_node_type(const struct ubifs_info *c, uint8_t *buf,
+			     int *node_num)
 {
 	uint8_t *addr = buf + UBIFS_LPT_CRC_BYTES;
 	int pos = 0, node_type;
@@ -1081,7 +1086,7 @@
  *
  * This function returns %1 if the buffer contains a node or %0 if it does not.
  */
-static int is_a_node(struct ubifs_info *c, uint8_t *buf, int len)
+static int is_a_node(const struct ubifs_info *c, uint8_t *buf, int len)
 {
 	uint8_t *addr = buf + UBIFS_LPT_CRC_BYTES;
 	int pos = 0, node_type, node_len;
@@ -1105,7 +1110,6 @@
 	return 1;
 }
 
-
 /**
  * lpt_gc_lnum - garbage collect a LPT LEB.
  * @c: UBIFS file-system description object
@@ -1463,7 +1467,7 @@
 #ifdef CONFIG_UBIFS_FS_DEBUG
 
 /**
- * dbg_is_all_ff - determine if a buffer contains only 0xff bytes.
+ * dbg_is_all_ff - determine if a buffer contains only 0xFF bytes.
  * @buf: buffer
  * @len: buffer length
  */
@@ -1488,7 +1492,7 @@
 	struct ubifs_nnode *nnode;
 	int hght;
 
-	/* Entire tree is in memory so first_nnode / next_nnode are ok */
+	/* Entire tree is in memory so first_nnode / next_nnode are OK */
 	nnode = first_nnode(c, &hght);
 	for (; nnode; nnode = next_nnode(c, nnode, &hght)) {
 		struct ubifs_nbranch *branch;
@@ -1602,7 +1606,10 @@
 {
 	int err, len = c->leb_size, dirty = 0, node_type, node_num, node_len;
 	int ret;
-	void *buf = c->dbg_buf;
+	void *buf = c->dbg->buf;
+
+	if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
+		return 0;
 
 	dbg_lp("LEB %d", lnum);
 	err = ubi_read(c->ubi, lnum, buf, 0, c->leb_size);
@@ -1704,6 +1711,9 @@
 	long long free = 0;
 	int i;
 
+	if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
+		return 0;
+
 	for (i = 0; i < c->lpt_lebs; i++) {
 		if (c->ltab[i].tgc || c->ltab[i].cmt)
 			continue;
@@ -1716,6 +1726,8 @@
 		dbg_err("LPT space error: free %lld lpt_sz %lld",
 			free, c->lpt_sz);
 		dbg_dump_lpt_info(c);
+		dbg_dump_lpt_lebs(c);
+		dump_stack();
 		return -EINVAL;
 	}
 	return 0;
@@ -1731,15 +1743,19 @@
  */
 int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len)
 {
+	struct ubifs_debug_info *d = c->dbg;
 	long long chk_lpt_sz, lpt_sz;
 	int err = 0;
 
+	if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
+		return 0;
+
 	switch (action) {
 	case 0:
-		c->chk_lpt_sz = 0;
-		c->chk_lpt_sz2 = 0;
-		c->chk_lpt_lebs = 0;
-		c->chk_lpt_wastage = 0;
+		d->chk_lpt_sz = 0;
+		d->chk_lpt_sz2 = 0;
+		d->chk_lpt_lebs = 0;
+		d->chk_lpt_wastage = 0;
 		if (c->dirty_pn_cnt > c->pnode_cnt) {
 			dbg_err("dirty pnodes %d exceed max %d",
 				c->dirty_pn_cnt, c->pnode_cnt);
@@ -1752,35 +1768,35 @@
 		}
 		return err;
 	case 1:
-		c->chk_lpt_sz += len;
+		d->chk_lpt_sz += len;
 		return 0;
 	case 2:
-		c->chk_lpt_sz += len;
-		c->chk_lpt_wastage += len;
-		c->chk_lpt_lebs += 1;
+		d->chk_lpt_sz += len;
+		d->chk_lpt_wastage += len;
+		d->chk_lpt_lebs += 1;
 		return 0;
 	case 3:
 		chk_lpt_sz = c->leb_size;
-		chk_lpt_sz *= c->chk_lpt_lebs;
+		chk_lpt_sz *= d->chk_lpt_lebs;
 		chk_lpt_sz += len - c->nhead_offs;
-		if (c->chk_lpt_sz != chk_lpt_sz) {
+		if (d->chk_lpt_sz != chk_lpt_sz) {
 			dbg_err("LPT wrote %lld but space used was %lld",
-				c->chk_lpt_sz, chk_lpt_sz);
+				d->chk_lpt_sz, chk_lpt_sz);
 			err = -EINVAL;
 		}
-		if (c->chk_lpt_sz > c->lpt_sz) {
+		if (d->chk_lpt_sz > c->lpt_sz) {
 			dbg_err("LPT wrote %lld but lpt_sz is %lld",
-				c->chk_lpt_sz, c->lpt_sz);
+				d->chk_lpt_sz, c->lpt_sz);
 			err = -EINVAL;
 		}
-		if (c->chk_lpt_sz2 && c->chk_lpt_sz != c->chk_lpt_sz2) {
+		if (d->chk_lpt_sz2 && d->chk_lpt_sz != d->chk_lpt_sz2) {
 			dbg_err("LPT layout size %lld but wrote %lld",
-				c->chk_lpt_sz, c->chk_lpt_sz2);
+				d->chk_lpt_sz, d->chk_lpt_sz2);
 			err = -EINVAL;
 		}
-		if (c->chk_lpt_sz2 && c->new_nhead_offs != len) {
+		if (d->chk_lpt_sz2 && d->new_nhead_offs != len) {
 			dbg_err("LPT new nhead offs: expected %d was %d",
-				c->new_nhead_offs, len);
+				d->new_nhead_offs, len);
 			err = -EINVAL;
 		}
 		lpt_sz = (long long)c->pnode_cnt * c->pnode_sz;
@@ -1788,26 +1804,146 @@
 		lpt_sz += c->ltab_sz;
 		if (c->big_lpt)
 			lpt_sz += c->lsave_sz;
-		if (c->chk_lpt_sz - c->chk_lpt_wastage > lpt_sz) {
+		if (d->chk_lpt_sz - d->chk_lpt_wastage > lpt_sz) {
 			dbg_err("LPT chk_lpt_sz %lld + waste %lld exceeds %lld",
-				c->chk_lpt_sz, c->chk_lpt_wastage, lpt_sz);
+				d->chk_lpt_sz, d->chk_lpt_wastage, lpt_sz);
 			err = -EINVAL;
 		}
-		if (err)
+		if (err) {
 			dbg_dump_lpt_info(c);
-		c->chk_lpt_sz2 = c->chk_lpt_sz;
-		c->chk_lpt_sz = 0;
-		c->chk_lpt_wastage = 0;
-		c->chk_lpt_lebs = 0;
-		c->new_nhead_offs = len;
+			dbg_dump_lpt_lebs(c);
+			dump_stack();
+		}
+		d->chk_lpt_sz2 = d->chk_lpt_sz;
+		d->chk_lpt_sz = 0;
+		d->chk_lpt_wastage = 0;
+		d->chk_lpt_lebs = 0;
+		d->new_nhead_offs = len;
 		return err;
 	case 4:
-		c->chk_lpt_sz += len;
-		c->chk_lpt_wastage += len;
+		d->chk_lpt_sz += len;
+		d->chk_lpt_wastage += len;
 		return 0;
 	default:
 		return -EINVAL;
 	}
 }
 
+/**
+ * dbg_dump_lpt_leb - dump an LPT LEB.
+ * @c: UBIFS file-system description object
+ * @lnum: LEB number to dump
+ *
+ * This function dumps an LEB from LPT area. Nodes in this area are very
+ * different to nodes in the main area (e.g., they do not have common headers,
+ * they do not have 8-byte alignments, etc), so we have a separate function to
+ * dump LPT area LEBs. Note, LPT has to be locked by the caller.
+ */
+static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
+{
+	int err, len = c->leb_size, node_type, node_num, node_len, offs;
+	void *buf = c->dbg->buf;
+
+	printk(KERN_DEBUG "(pid %d) start dumping LEB %d\n",
+	       current->pid, lnum);
+	err = ubi_read(c->ubi, lnum, buf, 0, c->leb_size);
+	if (err) {
+		ubifs_err("cannot read LEB %d, error %d", lnum, err);
+		return;
+	}
+	while (1) {
+		offs = c->leb_size - len;
+		if (!is_a_node(c, buf, len)) {
+			int pad_len;
+
+			pad_len = get_pad_len(c, buf, len);
+			if (pad_len) {
+				printk(KERN_DEBUG "LEB %d:%d, pad %d bytes\n",
+				       lnum, offs, pad_len);
+				buf += pad_len;
+				len -= pad_len;
+				continue;
+			}
+			if (len)
+				printk(KERN_DEBUG "LEB %d:%d, free %d bytes\n",
+				       lnum, offs, len);
+			break;
+		}
+
+		node_type = get_lpt_node_type(c, buf, &node_num);
+		switch (node_type) {
+		case UBIFS_LPT_PNODE:
+		{
+			node_len = c->pnode_sz;
+			if (c->big_lpt)
+				printk(KERN_DEBUG "LEB %d:%d, pnode num %d\n",
+				       lnum, offs, node_num);
+			else
+				printk(KERN_DEBUG "LEB %d:%d, pnode\n",
+				       lnum, offs);
+			break;
+		}
+		case UBIFS_LPT_NNODE:
+		{
+			int i;
+			struct ubifs_nnode nnode;
+
+			node_len = c->nnode_sz;
+			if (c->big_lpt)
+				printk(KERN_DEBUG "LEB %d:%d, nnode num %d, ",
+				       lnum, offs, node_num);
+			else
+				printk(KERN_DEBUG "LEB %d:%d, nnode, ",
+				       lnum, offs);
+			err = ubifs_unpack_nnode(c, buf, &nnode);
+			for (i = 0; i < UBIFS_LPT_FANOUT; i++) {
+				printk("%d:%d", nnode.nbranch[i].lnum,
+				       nnode.nbranch[i].offs);
+				if (i != UBIFS_LPT_FANOUT - 1)
+					printk(", ");
+			}
+			printk("\n");
+			break;
+		}
+		case UBIFS_LPT_LTAB:
+			node_len = c->ltab_sz;
+			printk(KERN_DEBUG "LEB %d:%d, ltab\n",
+			       lnum, offs);
+			break;
+		case UBIFS_LPT_LSAVE:
+			node_len = c->lsave_sz;
+			printk(KERN_DEBUG "LEB %d:%d, lsave len\n", lnum, offs);
+			break;
+		default:
+			ubifs_err("LPT node type %d not recognized", node_type);
+			return;
+		}
+
+		buf += node_len;
+		len -= node_len;
+	}
+
+	printk(KERN_DEBUG "(pid %d) finish dumping LEB %d\n",
+	       current->pid, lnum);
+}
+
+/**
+ * dbg_dump_lpt_lebs - dump LPT lebs.
+ * @c: UBIFS file-system description object
+ *
+ * This function dumps all LPT LEBs. The caller has to make sure the LPT is
+ * locked.
+ */
+void dbg_dump_lpt_lebs(const struct ubifs_info *c)
+{
+	int i;
+
+	printk(KERN_DEBUG "(pid %d) start dumping all LPT LEBs\n",
+	       current->pid);
+	for (i = 0; i < c->lpt_lebs; i++)
+		dump_lpt_leb(c, i + c->lpt_first);
+	printk(KERN_DEBUG "(pid %d) finish dumping all LPT LEBs\n",
+	       current->pid);
+}
+
 #endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/orphan.c b/fs/ubifs/orphan.c
index 9bd5a43..9e6f403 100644
--- a/fs/ubifs/orphan.c
+++ b/fs/ubifs/orphan.c
@@ -899,7 +899,7 @@
 	for (lnum = c->orph_first; lnum <= c->orph_last; lnum++) {
 		struct ubifs_scan_leb *sleb;
 
-		sleb = ubifs_scan(c, lnum, 0, c->dbg_buf);
+		sleb = ubifs_scan(c, lnum, 0, c->dbg->buf);
 		if (IS_ERR(sleb)) {
 			err = PTR_ERR(sleb);
 			break;
diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c
index 21f7d04..ce42a7b 100644
--- a/fs/ubifs/replay.c
+++ b/fs/ubifs/replay.c
@@ -144,7 +144,7 @@
 		/*
 		 * If the replay order was perfect the dirty space would now be
 		 * zero. The order is not perfect because the the journal heads
-		 * race with eachother. This is not a problem but is does mean
+		 * race with each other. This is not a problem but is does mean
 		 * that the dirty space may temporarily exceed c->leb_size
 		 * during the replay.
 		 */
@@ -656,7 +656,7 @@
  * @dirty: amount of dirty space from padding and deletion nodes
  *
  * This function inserts a reference node to the replay tree and returns zero
- * in case of success ort a negative error code in case of failure.
+ * in case of success or a negative error code in case of failure.
  */
 static int insert_ref_node(struct ubifs_info *c, int lnum, int offs,
 			   unsigned long long sqnum, int free, int dirty)
@@ -883,7 +883,7 @@
 		 * This means that we reached end of log and now
 		 * look to the older log data, which was already
 		 * committed but the eraseblock was not erased (UBIFS
-		 * only unmaps it). So this basically means we have to
+		 * only un-maps it). So this basically means we have to
 		 * exit with "end of log" code.
 		 */
 		err = 1;
@@ -1062,6 +1062,15 @@
 	if (err)
 		goto out;
 
+	/*
+	 * UBIFS budgeting calculations use @c->budg_uncommitted_idx variable
+	 * to roughly estimate index growth. Things like @c->min_idx_lebs
+	 * depend on it. This means we have to initialize it to make sure
+	 * budgeting works properly.
+	 */
+	c->budg_uncommitted_idx = atomic_long_read(&c->dirty_zn_cnt);
+	c->budg_uncommitted_idx *= c->max_idx_node_sz;
+
 	ubifs_assert(c->bud_bytes <= c->max_bud_bytes || c->need_recovery);
 	dbg_mnt("finished, log head LEB %d:%d, max_sqnum %llu, "
 		"highest_inum %lu", c->lhead_lnum, c->lhead_offs, c->max_sqnum,
diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c
index 0f39235..e070c64 100644
--- a/fs/ubifs/sb.c
+++ b/fs/ubifs/sb.c
@@ -28,6 +28,7 @@
 
 #include "ubifs.h"
 #include <linux/random.h>
+#include <linux/math64.h>
 
 /*
  * Default journal size in logical eraseblocks as a percent of total
@@ -80,7 +81,7 @@
 	int err, tmp, jnl_lebs, log_lebs, max_buds, main_lebs, main_first;
 	int lpt_lebs, lpt_first, orph_lebs, big_lpt, ino_waste, sup_flags = 0;
 	int min_leb_cnt = UBIFS_MIN_LEB_CNT;
-	uint64_t tmp64, main_bytes;
+	long long tmp64, main_bytes;
 	__le64 tmp_le64;
 
 	/* Some functions called from here depend on the @c->key_len filed */
@@ -160,7 +161,7 @@
 	if (!sup)
 		return -ENOMEM;
 
-	tmp64 = (uint64_t)max_buds * c->leb_size;
+	tmp64 = (long long)max_buds * c->leb_size;
 	if (big_lpt)
 		sup_flags |= UBIFS_FLG_BIGLPT;
 
@@ -179,14 +180,16 @@
 	sup->fanout        = cpu_to_le32(DEFAULT_FANOUT);
 	sup->lsave_cnt     = cpu_to_le32(c->lsave_cnt);
 	sup->fmt_version   = cpu_to_le32(UBIFS_FORMAT_VERSION);
-	sup->default_compr = cpu_to_le16(UBIFS_COMPR_LZO);
 	sup->time_gran     = cpu_to_le32(DEFAULT_TIME_GRAN);
+	if (c->mount_opts.override_compr)
+		sup->default_compr = cpu_to_le16(c->mount_opts.compr_type);
+	else
+		sup->default_compr = cpu_to_le16(UBIFS_COMPR_LZO);
 
 	generate_random_uuid(sup->uuid);
 
-	main_bytes = (uint64_t)main_lebs * c->leb_size;
-	tmp64 = main_bytes * DEFAULT_RP_PERCENT;
-	do_div(tmp64, 100);
+	main_bytes = (long long)main_lebs * c->leb_size;
+	tmp64 = div_u64(main_bytes * DEFAULT_RP_PERCENT, 100);
 	if (tmp64 > DEFAULT_MAX_RP_SIZE)
 		tmp64 = DEFAULT_MAX_RP_SIZE;
 	sup->rp_size = cpu_to_le64(tmp64);
@@ -582,16 +585,15 @@
 	c->jhead_cnt     = le32_to_cpu(sup->jhead_cnt) + NONDATA_JHEADS_CNT;
 	c->fanout        = le32_to_cpu(sup->fanout);
 	c->lsave_cnt     = le32_to_cpu(sup->lsave_cnt);
-	c->default_compr = le16_to_cpu(sup->default_compr);
 	c->rp_size       = le64_to_cpu(sup->rp_size);
 	c->rp_uid        = le32_to_cpu(sup->rp_uid);
 	c->rp_gid        = le32_to_cpu(sup->rp_gid);
 	sup_flags        = le32_to_cpu(sup->flags);
+	if (!c->mount_opts.override_compr)
+		c->default_compr = le16_to_cpu(sup->default_compr);
 
 	c->vfs_sb->s_time_gran = le32_to_cpu(sup->time_gran);
-
 	memcpy(&c->uuid, &sup->uuid, 16);
-
 	c->big_lpt = !!(sup_flags & UBIFS_FLG_BIGLPT);
 
 	/* Automatically increase file system size to the maximum size */
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index d80b2ae..0d7564b 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -34,6 +34,8 @@
 #include <linux/parser.h>
 #include <linux/seq_file.h>
 #include <linux/mount.h>
+#include <linux/math64.h>
+#include <linux/writeback.h>
 #include "ubifs.h"
 
 /*
@@ -417,39 +419,54 @@
 	else if (c->mount_opts.chk_data_crc == 1)
 		seq_printf(s, ",no_chk_data_crc");
 
+	if (c->mount_opts.override_compr) {
+		seq_printf(s, ",compr=");
+		seq_printf(s, ubifs_compr_name(c->mount_opts.compr_type));
+	}
+
 	return 0;
 }
 
 static int ubifs_sync_fs(struct super_block *sb, int wait)
 {
+	int i, err;
 	struct ubifs_info *c = sb->s_fs_info;
-	int i, ret = 0, err;
-	long long bud_bytes;
+	struct writeback_control wbc = {
+		.sync_mode   = wait ? WB_SYNC_ALL : WB_SYNC_HOLD,
+		.range_start = 0,
+		.range_end   = LLONG_MAX,
+		.nr_to_write = LONG_MAX,
+	};
 
-	if (c->jheads) {
-		for (i = 0; i < c->jhead_cnt; i++) {
-			err = ubifs_wbuf_sync(&c->jheads[i].wbuf);
-			if (err && !ret)
-				ret = err;
-		}
+	if (sb->s_flags & MS_RDONLY)
+		return 0;
 
-		/* Commit the journal unless it has too little data */
-		spin_lock(&c->buds_lock);
-		bud_bytes = c->bud_bytes;
-		spin_unlock(&c->buds_lock);
-		if (bud_bytes > c->leb_size) {
-			err = ubifs_run_commit(c);
-			if (err)
-				return err;
-		}
+	/*
+	 * Synchronize write buffers, because 'ubifs_run_commit()' does not
+	 * do this if it waits for an already running commit.
+	 */
+	for (i = 0; i < c->jhead_cnt; i++) {
+		err = ubifs_wbuf_sync(&c->jheads[i].wbuf);
+		if (err)
+			return err;
 	}
 
 	/*
-	 * We ought to call sync for c->ubi but it does not have one. If it had
-	 * it would in turn call mtd->sync, however mtd operations are
-	 * synchronous anyway, so we don't lose any sleep here.
+	 * VFS calls '->sync_fs()' before synchronizing all dirty inodes and
+	 * pages, so synchronize them first, then commit the journal. Strictly
+	 * speaking, it is not necessary to commit the journal here,
+	 * synchronizing write-buffers would be enough. But committing makes
+	 * UBIFS free space predictions much more accurate, so we want to let
+	 * the user be able to get more accurate results of 'statfs()' after
+	 * they synchronize the file system.
 	 */
-	return ret;
+	generic_sync_sb_inodes(sb, &wbc);
+
+	err = ubifs_run_commit(c);
+	if (err)
+		return err;
+
+	return ubi_sync(c->vi.ubi_num);
 }
 
 /**
@@ -596,7 +613,7 @@
 }
 
 /*
- * init_constants_late - initialize UBIFS constants.
+ * init_constants_sb - initialize UBIFS constants.
  * @c: UBIFS file-system description object
  *
  * This is a helper function which initializes various UBIFS constants after
@@ -604,10 +621,10 @@
  * makes sure they are all right. Returns zero in case of success and a
  * negative error code in case of failure.
  */
-static int init_constants_late(struct ubifs_info *c)
+static int init_constants_sb(struct ubifs_info *c)
 {
 	int tmp, err;
-	uint64_t tmp64;
+	long long tmp64;
 
 	c->main_bytes = (long long)c->main_lebs * c->leb_size;
 	c->max_znode_sz = sizeof(struct ubifs_znode) +
@@ -634,9 +651,8 @@
 	 * Make sure that the log is large enough to fit reference nodes for
 	 * all buds plus one reserved LEB.
 	 */
-	tmp64 = c->max_bud_bytes;
-	tmp = do_div(tmp64, c->leb_size);
-	c->max_bud_cnt = tmp64 + !!tmp;
+	tmp64 = c->max_bud_bytes + c->leb_size - 1;
+	c->max_bud_cnt = div_u64(tmp64, c->leb_size);
 	tmp = (c->ref_node_alsz * c->max_bud_cnt + c->leb_size - 1);
 	tmp /= c->leb_size;
 	tmp += 1;
@@ -672,7 +688,7 @@
 	 * Consequently, if the journal is too small, UBIFS will treat it as
 	 * always full.
 	 */
-	tmp64 = (uint64_t)(c->jhead_cnt + 1) * c->leb_size + 1;
+	tmp64 = (long long)(c->jhead_cnt + 1) * c->leb_size + 1;
 	if (c->bg_bud_bytes < tmp64)
 		c->bg_bud_bytes = tmp64;
 	if (c->max_bud_bytes < tmp64 + c->leb_size)
@@ -682,6 +698,21 @@
 	if (err)
 		return err;
 
+	return 0;
+}
+
+/*
+ * init_constants_master - initialize UBIFS constants.
+ * @c: UBIFS file-system description object
+ *
+ * This is a helper function which initializes various UBIFS constants after
+ * the master node has been read. It also checks various UBIFS parameters and
+ * makes sure they are all right.
+ */
+static void init_constants_master(struct ubifs_info *c)
+{
+	long long tmp64;
+
 	c->min_idx_lebs = ubifs_calc_min_idx_lebs(c);
 
 	/*
@@ -690,14 +721,13 @@
 	 * necessary to report something for the 'statfs()' call.
 	 *
 	 * Subtract the LEB reserved for GC, the LEB which is reserved for
-	 * deletions, and assume only one journal head is available.
+	 * deletions, minimum LEBs for the index, and assume only one journal
+	 * head is available.
 	 */
-	tmp64 = c->main_lebs - 2 - c->jhead_cnt + 1;
-	tmp64 *= (uint64_t)c->leb_size - c->leb_overhead;
+	tmp64 = c->main_lebs - 1 - 1 - MIN_INDEX_LEBS - c->jhead_cnt + 1;
+	tmp64 *= (long long)c->leb_size - c->leb_overhead;
 	tmp64 = ubifs_reported_space(c, tmp64);
 	c->block_cnt = tmp64 >> UBIFS_BLOCK_SHIFT;
-
-	return 0;
 }
 
 /**
@@ -878,6 +908,7 @@
  * Opt_no_bulk_read: disable bulk-reads
  * Opt_chk_data_crc: check CRCs when reading data nodes
  * Opt_no_chk_data_crc: do not check CRCs when reading data nodes
+ * Opt_override_compr: override default compressor
  * Opt_err: just end of array marker
  */
 enum {
@@ -887,6 +918,7 @@
 	Opt_no_bulk_read,
 	Opt_chk_data_crc,
 	Opt_no_chk_data_crc,
+	Opt_override_compr,
 	Opt_err,
 };
 
@@ -897,6 +929,7 @@
 	{Opt_no_bulk_read, "no_bulk_read"},
 	{Opt_chk_data_crc, "chk_data_crc"},
 	{Opt_no_chk_data_crc, "no_chk_data_crc"},
+	{Opt_override_compr, "compr=%s"},
 	{Opt_err, NULL},
 };
 
@@ -950,6 +983,28 @@
 			c->mount_opts.chk_data_crc = 1;
 			c->no_chk_data_crc = 1;
 			break;
+		case Opt_override_compr:
+		{
+			char *name = match_strdup(&args[0]);
+
+			if (!name)
+				return -ENOMEM;
+			if (!strcmp(name, "none"))
+				c->mount_opts.compr_type = UBIFS_COMPR_NONE;
+			else if (!strcmp(name, "lzo"))
+				c->mount_opts.compr_type = UBIFS_COMPR_LZO;
+			else if (!strcmp(name, "zlib"))
+				c->mount_opts.compr_type = UBIFS_COMPR_ZLIB;
+			else {
+				ubifs_err("unknown compressor \"%s\"", name);
+				kfree(name);
+				return -EINVAL;
+			}
+			kfree(name);
+			c->mount_opts.override_compr = 1;
+			c->default_compr = c->mount_opts.compr_type;
+			break;
+		}
 		default:
 			ubifs_err("unrecognized mount option \"%s\" "
 				  "or missing value", p);
@@ -1019,6 +1074,30 @@
 }
 
 /**
+ * check_free_space - check if there is enough free space to mount.
+ * @c: UBIFS file-system description object
+ *
+ * This function makes sure UBIFS has enough free space to be mounted in
+ * read/write mode. UBIFS must always have some free space to allow deletions.
+ */
+static int check_free_space(struct ubifs_info *c)
+{
+	ubifs_assert(c->dark_wm > 0);
+	if (c->lst.total_free + c->lst.total_dirty < c->dark_wm) {
+		ubifs_err("insufficient free space to mount in read/write mode");
+		dbg_dump_budg(c);
+		dbg_dump_lprops(c);
+		/*
+		 * We return %-EINVAL instead of %-ENOSPC because it seems to
+		 * be the closest error code mentioned in the mount function
+		 * documentation.
+		 */
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/**
  * mount_ubifs - mount UBIFS file-system.
  * @c: UBIFS file-system description object
  *
@@ -1039,11 +1118,9 @@
 	if (err)
 		return err;
 
-#ifdef CONFIG_UBIFS_FS_DEBUG
-	c->dbg_buf = vmalloc(c->leb_size);
-	if (!c->dbg_buf)
-		return -ENOMEM;
-#endif
+	err = ubifs_debugging_init(c);
+	if (err)
+		return err;
 
 	err = check_volume_empty(c);
 	if (err)
@@ -1100,27 +1177,25 @@
 		goto out_free;
 
 	/*
-	 * Make sure the compressor which is set as the default on in the
-	 * superblock was actually compiled in.
+	 * Make sure the compressor which is set as default in the superblock
+	 * or overridden by mount options is actually compiled in.
 	 */
 	if (!ubifs_compr_present(c->default_compr)) {
-		ubifs_warn("'%s' compressor is set by superblock, but not "
-			   "compiled in", ubifs_compr_name(c->default_compr));
-		c->default_compr = UBIFS_COMPR_NONE;
+		ubifs_err("'compressor \"%s\" is not compiled in",
+			  ubifs_compr_name(c->default_compr));
+		goto out_free;
 	}
 
-	dbg_failure_mode_registration(c);
-
-	err = init_constants_late(c);
+	err = init_constants_sb(c);
 	if (err)
-		goto out_dereg;
+		goto out_free;
 
 	sz = ALIGN(c->max_idx_node_sz, c->min_io_size);
 	sz = ALIGN(sz + c->max_idx_node_sz, c->min_io_size);
 	c->cbuf = kmalloc(sz, GFP_NOFS);
 	if (!c->cbuf) {
 		err = -ENOMEM;
-		goto out_dereg;
+		goto out_free;
 	}
 
 	sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, c->vi.vol_id);
@@ -1145,6 +1220,8 @@
 	if (err)
 		goto out_master;
 
+	init_constants_master(c);
+
 	if ((c->mst_node->flags & cpu_to_le32(UBIFS_MST_DIRTY)) != 0) {
 		ubifs_msg("recovery needed");
 		c->need_recovery = 1;
@@ -1183,12 +1260,9 @@
 	if (!mounted_read_only) {
 		int lnum;
 
-		/* Check for enough free space */
-		if (ubifs_calc_available(c, c->min_idx_lebs) <= 0) {
-			ubifs_err("insufficient available space");
-			err = -EINVAL;
+		err = check_free_space(c);
+		if (err)
 			goto out_orphans;
-		}
 
 		/* Check for enough log space */
 		lnum = c->lhead_lnum + 1;
@@ -1232,6 +1306,10 @@
 		}
 	}
 
+	err = dbg_debugfs_init_fs(c);
+	if (err)
+		goto out_infos;
+
 	err = dbg_check_filesystem(c);
 	if (err)
 		goto out_infos;
@@ -1283,8 +1361,20 @@
 	dbg_msg("tree fanout:         %d", c->fanout);
 	dbg_msg("reserved GC LEB:     %d", c->gc_lnum);
 	dbg_msg("first main LEB:      %d", c->main_first);
+	dbg_msg("max. znode size      %d", c->max_znode_sz);
+	dbg_msg("max. index node size %d", c->max_idx_node_sz);
+	dbg_msg("node sizes:          data %zu, inode %zu, dentry %zu",
+		UBIFS_DATA_NODE_SZ, UBIFS_INO_NODE_SZ, UBIFS_DENT_NODE_SZ);
+	dbg_msg("node sizes:          trun %zu, sb %zu, master %zu",
+		UBIFS_TRUN_NODE_SZ, UBIFS_SB_NODE_SZ, UBIFS_MST_NODE_SZ);
+	dbg_msg("node sizes:          ref %zu, cmt. start %zu, orph %zu",
+		UBIFS_REF_NODE_SZ, UBIFS_CS_NODE_SZ, UBIFS_ORPH_NODE_SZ);
+	dbg_msg("max. node sizes:     data %zu, inode %zu dentry %zu",
+	        UBIFS_MAX_DATA_NODE_SZ, UBIFS_MAX_INO_NODE_SZ,
+		UBIFS_MAX_DENT_NODE_SZ);
 	dbg_msg("dead watermark:      %d", c->dead_wm);
 	dbg_msg("dark watermark:      %d", c->dark_wm);
+	dbg_msg("LEB overhead:        %d", c->leb_overhead);
 	x = (long long)c->main_lebs * c->dark_wm;
 	dbg_msg("max. dark space:     %lld (%lld KiB, %lld MiB)",
 		x, x >> 10, x >> 20);
@@ -1320,14 +1410,12 @@
 	free_wbufs(c);
 out_cbuf:
 	kfree(c->cbuf);
-out_dereg:
-	dbg_failure_mode_deregistration(c);
 out_free:
 	kfree(c->bu.buf);
 	vfree(c->ileb_buf);
 	vfree(c->sbuf);
 	kfree(c->bottom_up_buf);
-	UBIFS_DBG(vfree(c->dbg_buf));
+	ubifs_debugging_exit(c);
 	return err;
 }
 
@@ -1345,6 +1433,7 @@
 	dbg_gen("un-mounting UBI device %d, volume %d", c->vi.ubi_num,
 		c->vi.vol_id);
 
+	dbg_debugfs_exit_fs(c);
 	spin_lock(&ubifs_infos_lock);
 	list_del(&c->infos_list);
 	spin_unlock(&ubifs_infos_lock);
@@ -1364,8 +1453,7 @@
 	vfree(c->ileb_buf);
 	vfree(c->sbuf);
 	kfree(c->bottom_up_buf);
-	UBIFS_DBG(vfree(c->dbg_buf));
-	dbg_failure_mode_deregistration(c);
+	ubifs_debugging_exit(c);
 }
 
 /**
@@ -1387,12 +1475,9 @@
 	c->remounting_rw = 1;
 	c->always_chk_crc = 1;
 
-	/* Check for enough free space */
-	if (ubifs_calc_available(c, c->min_idx_lebs) <= 0) {
-		ubifs_err("insufficient available space");
-		err = -EINVAL;
+	err = check_free_space(c);
+	if (err)
 		goto out;
-	}
 
 	if (c->old_leb_cnt != c->leb_cnt) {
 		struct ubifs_sb_node *sup;
@@ -1515,20 +1600,24 @@
  * @c: UBIFS file-system description object
  *
  * This function is called during un-mounting and re-mounting, and it commits
- * the journal unless the "fast unmount" mode is enabled. It also avoids
- * committing the journal if it contains too few data.
+ * the journal unless the "fast unmount" mode is enabled.
  */
 static void commit_on_unmount(struct ubifs_info *c)
 {
-	if (!c->fast_unmount) {
-		long long bud_bytes;
+	struct super_block *sb = c->vfs_sb;
+	long long bud_bytes;
 
-		spin_lock(&c->buds_lock);
-		bud_bytes = c->bud_bytes;
-		spin_unlock(&c->buds_lock);
-		if (bud_bytes > c->leb_size)
-			ubifs_run_commit(c);
-	}
+	/*
+	 * This function is called before the background thread is stopped, so
+	 * we may race with ongoing commit, which means we have to take
+	 * @c->bud_lock to access @c->bud_bytes.
+	 */
+	spin_lock(&c->buds_lock);
+	bud_bytes = c->bud_bytes;
+	spin_unlock(&c->buds_lock);
+
+	if (!c->fast_unmount && !(sb->s_flags & MS_RDONLY) && bud_bytes)
+		ubifs_run_commit(c);
 }
 
 /**
@@ -1849,7 +1938,6 @@
 		goto out_iput;
 
 	mutex_unlock(&c->umount_mutex);
-
 	return 0;
 
 out_iput:
@@ -1955,7 +2043,7 @@
 	 * We do 'commit_on_unmount()' here instead of 'ubifs_put_super()'
 	 * in order to be outside BKL.
 	 */
-	if (sb->s_root && !(sb->s_flags & MS_RDONLY))
+	if (sb->s_root)
 		commit_on_unmount(c);
 	/* The un-mount routine is actually done in put_super() */
 	generic_shutdown_super(sb);
@@ -2021,6 +2109,14 @@
 	BUILD_BUG_ON(UBIFS_REF_NODE_SZ != 64);
 
 	/*
+	 * We use 2 bit wide bit-fields to store compression type, which should
+	 * be amended if more compressors are added. The bit-fields are:
+	 * @compr_type in 'struct ubifs_inode', @default_compr in
+	 * 'struct ubifs_info' and @compr_type in 'struct ubifs_mount_opts'.
+	 */
+	BUILD_BUG_ON(UBIFS_COMPR_TYPES_CNT > 4);
+
+	/*
 	 * We require that PAGE_CACHE_SIZE is greater-than-or-equal-to
 	 * UBIFS_BLOCK_SIZE. It is assumed that both are powers of 2.
 	 */
@@ -2049,11 +2145,17 @@
 
 	err = ubifs_compressors_init();
 	if (err)
+		goto out_shrinker;
+
+	err = dbg_debugfs_init();
+	if (err)
 		goto out_compr;
 
 	return 0;
 
 out_compr:
+	ubifs_compressors_exit();
+out_shrinker:
 	unregister_shrinker(&ubifs_shrinker_info);
 	kmem_cache_destroy(ubifs_inode_slab);
 out_reg:
@@ -2068,6 +2170,7 @@
 	ubifs_assert(list_empty(&ubifs_infos));
 	ubifs_assert(atomic_long_read(&ubifs_clean_zn_cnt) == 0);
 
+	dbg_debugfs_exit();
 	ubifs_compressors_exit();
 	unregister_shrinker(&ubifs_shrinker_info);
 	kmem_cache_destroy(ubifs_inode_slab);
diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c
index 6eef534..f7e36f5 100644
--- a/fs/ubifs/tnc.c
+++ b/fs/ubifs/tnc.c
@@ -2245,12 +2245,11 @@
 			if (found) {
 				/* Ensure the znode is dirtied */
 				if (znode->cnext || !ubifs_zn_dirty(znode)) {
-					    znode = dirty_cow_bottom_up(c,
-									znode);
-					    if (IS_ERR(znode)) {
-						    err = PTR_ERR(znode);
-						    goto out_unlock;
-					    }
+					znode = dirty_cow_bottom_up(c, znode);
+					if (IS_ERR(znode)) {
+						err = PTR_ERR(znode);
+						goto out_unlock;
+					}
 				}
 				zbr = &znode->zbranch[n];
 				lnc_free(zbr);
@@ -2317,11 +2316,11 @@
 
 		/* Ensure the znode is dirtied */
 		if (znode->cnext || !ubifs_zn_dirty(znode)) {
-			    znode = dirty_cow_bottom_up(c, znode);
-			    if (IS_ERR(znode)) {
-				    err = PTR_ERR(znode);
-				    goto out_unlock;
-			    }
+			znode = dirty_cow_bottom_up(c, znode);
+			if (IS_ERR(znode)) {
+				err = PTR_ERR(znode);
+				goto out_unlock;
+			}
 		}
 
 		if (found == 1) {
@@ -2627,11 +2626,11 @@
 
 		/* Ensure the znode is dirtied */
 		if (znode->cnext || !ubifs_zn_dirty(znode)) {
-			    znode = dirty_cow_bottom_up(c, znode);
-			    if (IS_ERR(znode)) {
-				    err = PTR_ERR(znode);
-				    goto out_unlock;
-			    }
+			znode = dirty_cow_bottom_up(c, znode);
+			if (IS_ERR(znode)) {
+				err = PTR_ERR(znode);
+				goto out_unlock;
+			}
 		}
 
 		/* Remove all keys in range except the first */
diff --git a/fs/ubifs/tnc_commit.c b/fs/ubifs/tnc_commit.c
index 8ac76b1..fde8d12 100644
--- a/fs/ubifs/tnc_commit.c
+++ b/fs/ubifs/tnc_commit.c
@@ -553,8 +553,8 @@
 	}
 
 #ifdef CONFIG_UBIFS_FS_DEBUG
-	c->new_ihead_lnum = lnum;
-	c->new_ihead_offs = buf_offs;
+	c->dbg->new_ihead_lnum = lnum;
+	c->dbg->new_ihead_offs = buf_offs;
 #endif
 
 	return 0;
@@ -802,8 +802,10 @@
 	 * budgeting subsystem to assume the index is already committed,
 	 * even though it is not.
 	 */
+	ubifs_assert(c->min_idx_lebs == ubifs_calc_min_idx_lebs(c));
 	c->old_idx_sz = c->calc_idx_sz;
 	c->budg_uncommitted_idx = 0;
+	c->min_idx_lebs = ubifs_calc_min_idx_lebs(c);
 	spin_unlock(&c->space_lock);
 	mutex_unlock(&c->tnc_mutex);
 
@@ -1002,7 +1004,8 @@
 	}
 
 #ifdef CONFIG_UBIFS_FS_DEBUG
-	if (lnum != c->new_ihead_lnum || buf_offs != c->new_ihead_offs) {
+	if (lnum != c->dbg->new_ihead_lnum ||
+	    buf_offs != c->dbg->new_ihead_offs) {
 		ubifs_err("inconsistent ihead");
 		return -EINVAL;
 	}
diff --git a/fs/ubifs/ubifs-media.h b/fs/ubifs/ubifs-media.h
index 0b37804..b25fc36 100644
--- a/fs/ubifs/ubifs-media.h
+++ b/fs/ubifs/ubifs-media.h
@@ -51,6 +51,13 @@
  */
 #define UBIFS_MIN_COMPR_LEN 128
 
+/*
+ * If compressed data length is less than %UBIFS_MIN_COMPRESS_DIFF bytes
+ * shorter than uncompressed data length, UBIFS preferes to leave this data
+ * node uncompress, because it'll be read faster.
+ */
+#define UBIFS_MIN_COMPRESS_DIFF 64
+
 /* Root inode number */
 #define UBIFS_ROOT_INO 1
 
diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h
index 46b1725..fc2a4cc 100644
--- a/fs/ubifs/ubifs.h
+++ b/fs/ubifs/ubifs.h
@@ -63,6 +63,14 @@
 #define SQNUM_WARN_WATERMARK 0xFFFFFFFF00000000ULL
 #define SQNUM_WATERMARK      0xFFFFFFFFFF000000ULL
 
+/*
+ * Minimum amount of LEBs reserved for the index. At present the index needs at
+ * least 2 LEBs: one for the index head and one for in-the-gaps method (which
+ * currently does not cater for the index head and so excludes it from
+ * consideration).
+ */
+#define MIN_INDEX_LEBS 2
+
 /* Minimum amount of data UBIFS writes to the flash */
 #define MIN_WRITE_SZ (UBIFS_DATA_NODE_SZ + 8)
 
@@ -386,12 +394,12 @@
 	unsigned int dirty:1;
 	unsigned int xattr:1;
 	unsigned int bulk_read:1;
+	unsigned int compr_type:2;
 	struct mutex ui_mutex;
 	spinlock_t ui_lock;
 	loff_t synced_i_size;
 	loff_t ui_size;
 	int flags;
-	int compr_type;
 	pgoff_t last_page_read;
 	pgoff_t read_in_a_row;
 	int data_len;
@@ -419,7 +427,7 @@
  *
  * LPROPS_UNCAT: not categorized
  * LPROPS_DIRTY: dirty > 0, not index
- * LPROPS_DIRTY_IDX: dirty + free > UBIFS_CH_SZ and index
+ * LPROPS_DIRTY_IDX: dirty + free > @c->min_idx_node_sze and index
  * LPROPS_FREE: free > 0, not empty, not index
  * LPROPS_HEAP_CNT: number of heaps used for storing categorized LEBs
  * LPROPS_EMPTY: LEB is empty, not taken
@@ -473,8 +481,8 @@
 struct ubifs_lpt_lprops {
 	int free;
 	int dirty;
-	unsigned tgc : 1;
-	unsigned cmt : 1;
+	unsigned tgc:1;
+	unsigned cmt:1;
 };
 
 /**
@@ -482,24 +490,26 @@
  * @empty_lebs: number of empty LEBs
  * @taken_empty_lebs: number of taken LEBs
  * @idx_lebs: number of indexing LEBs
- * @total_free: total free space in bytes
- * @total_dirty: total dirty space in bytes
- * @total_used: total used space in bytes (includes only data LEBs)
- * @total_dead: total dead space in bytes (includes only data LEBs)
- * @total_dark: total dark space in bytes (includes only data LEBs)
+ * @total_free: total free space in bytes (includes all LEBs)
+ * @total_dirty: total dirty space in bytes (includes all LEBs)
+ * @total_used: total used space in bytes (does not include index LEBs)
+ * @total_dead: total dead space in bytes (does not include index LEBs)
+ * @total_dark: total dark space in bytes (does not include index LEBs)
  *
- * N.B. total_dirty and total_used are different to other total_* fields,
- * because they account _all_ LEBs, not just data LEBs.
+ * The @taken_empty_lebs field counts the LEBs that are in the transient state
+ * of having been "taken" for use but not yet written to. @taken_empty_lebs is
+ * needed to account correctly for @gc_lnum, otherwise @empty_lebs could be
+ * used by itself (in which case 'unused_lebs' would be a better name). In the
+ * case of @gc_lnum, it is "taken" at mount time or whenever a LEB is retained
+ * by GC, but unlike other empty LEBs that are "taken", it may not be written
+ * straight away (i.e. before the next commit start or unmount), so either
+ * @gc_lnum must be specially accounted for, or the current approach followed
+ * i.e. count it under @taken_empty_lebs.
  *
- * 'taken_empty_lebs' counts the LEBs that are in the transient state of having
- * been 'taken' for use but not yet written to. 'taken_empty_lebs' is needed
- * to account correctly for gc_lnum, otherwise 'empty_lebs' could be used
- * by itself (in which case 'unused_lebs' would be a better name). In the case
- * of gc_lnum, it is 'taken' at mount time or whenever a LEB is retained by GC,
- * but unlike other empty LEBs that are 'taken', it may not be written straight
- * away (i.e. before the next commit start or unmount), so either gc_lnum must
- * be specially accounted for, or the current approach followed i.e. count it
- * under 'taken_empty_lebs'.
+ * @empty_lebs includes @taken_empty_lebs.
+ *
+ * @total_used, @total_dead and @total_dark fields do not account indexing
+ * LEBs.
  */
 struct ubifs_lp_stats {
 	int empty_lebs;
@@ -893,15 +903,25 @@
 /**
  * struct ubifs_mount_opts - UBIFS-specific mount options information.
  * @unmount_mode: selected unmount mode (%0 default, %1 normal, %2 fast)
- * @bulk_read: enable bulk-reads
- * @chk_data_crc: check CRCs when reading data nodes
+ * @bulk_read: enable/disable bulk-reads (%0 default, %1 disabe, %2 enable)
+ * @chk_data_crc: enable/disable CRC data checking when reading data nodes
+ *                (%0 default, %1 disabe, %2 enable)
+ * @override_compr: override default compressor (%0 - do not override and use
+ *                  superblock compressor, %1 - override and use compressor
+ *                  specified in @compr_type)
+ * @compr_type: compressor type to override the superblock compressor with
+ *              (%UBIFS_COMPR_NONE, etc)
  */
 struct ubifs_mount_opts {
 	unsigned int unmount_mode:2;
 	unsigned int bulk_read:2;
 	unsigned int chk_data_crc:2;
+	unsigned int override_compr:1;
+	unsigned int compr_type:2;
 };
 
+struct ubifs_debug_info;
+
 /**
  * struct ubifs_info - UBIFS file-system description data structure
  * (per-superblock).
@@ -946,6 +966,7 @@
  * @no_chk_data_crc: do not check CRCs when reading data nodes (except during
  *                   recovery)
  * @bulk_read: enable bulk-reads
+ * @default_compr: default compression algorithm (%UBIFS_COMPR_LZO, etc)
  *
  * @tnc_mutex: protects the Tree Node Cache (TNC), @zroot, @cnext, @enext, and
  *             @calc_idx_sz
@@ -963,8 +984,6 @@
  * @ileb_nxt: next pre-allocated index LEBs
  * @old_idx: tree of index nodes obsoleted since the last commit start
  * @bottom_up_buf: a buffer which is used by 'dirty_cow_bottom_up()' in tnc.c
- * @new_ihead_lnum: used by debugging to check ihead_lnum
- * @new_ihead_offs: used by debugging to check ihead_offs
  *
  * @mst_node: master node
  * @mst_offs: offset of valid master node
@@ -986,7 +1005,6 @@
  * @main_lebs: count of LEBs in the main area
  * @main_first: first LEB of the main area
  * @main_bytes: main area size in bytes
- * @default_compr: default compression algorithm (%UBIFS_COMPR_LZO, etc)
  *
  * @key_hash_type: type of the key hash
  * @key_hash: direntry key hash function
@@ -1149,15 +1167,7 @@
  * @always_chk_crc: always check CRCs (while mounting and remounting rw)
  * @mount_opts: UBIFS-specific mount options
  *
- * @dbg_buf: a buffer of LEB size used for debugging purposes
- * @old_zroot: old index root - used by 'dbg_check_old_index()'
- * @old_zroot_level: old index root level - used by 'dbg_check_old_index()'
- * @old_zroot_sqnum: old index root sqnum - used by 'dbg_check_old_index()'
- * @failure_mode: failure mode for recovery testing
- * @fail_delay: 0=>don't delay, 1=>delay a time, 2=>delay a number of calls
- * @fail_timeout: time in jiffies when delay of failure mode expires
- * @fail_cnt: current number of calls to failure mode I/O functions
- * @fail_cnt_max: number of calls by which to delay failure mode
+ * @dbg: debugging-related information
  */
 struct ubifs_info {
 	struct super_block *vfs_sb;
@@ -1196,6 +1206,7 @@
 	unsigned int big_lpt:1;
 	unsigned int no_chk_data_crc:1;
 	unsigned int bulk_read:1;
+	unsigned int default_compr:2;
 
 	struct mutex tnc_mutex;
 	struct ubifs_zbranch zroot;
@@ -1212,10 +1223,6 @@
 	int ileb_nxt;
 	struct rb_root old_idx;
 	int *bottom_up_buf;
-#ifdef CONFIG_UBIFS_FS_DEBUG
-	int new_ihead_lnum;
-	int new_ihead_offs;
-#endif
 
 	struct ubifs_mst_node *mst_node;
 	int mst_offs;
@@ -1237,7 +1244,6 @@
 	int main_lebs;
 	int main_first;
 	long long main_bytes;
-	int default_compr;
 
 	uint8_t key_hash_type;
 	uint32_t (*key_hash)(const char *str, int len);
@@ -1315,8 +1321,8 @@
 	void *sbuf;
 	struct list_head idx_gc;
 	int idx_gc_cnt;
-	volatile int gc_seq;
-	volatile int gced_lnum;
+	int gc_seq;
+	int gced_lnum;
 
 	struct list_head infos_list;
 	struct mutex umount_mutex;
@@ -1391,21 +1397,7 @@
 	struct ubifs_mount_opts mount_opts;
 
 #ifdef CONFIG_UBIFS_FS_DEBUG
-	void *dbg_buf;
-	struct ubifs_zbranch old_zroot;
-	int old_zroot_level;
-	unsigned long long old_zroot_sqnum;
-	int failure_mode;
-	int fail_delay;
-	unsigned long fail_timeout;
-	unsigned int fail_cnt;
-	unsigned int fail_cnt_max;
-	long long chk_lpt_sz;
-	long long chk_lpt_sz2;
-	long long chk_lpt_wastage;
-	int chk_lpt_lebs;
-	int new_nhead_lnum;
-	int new_nhead_offs;
+	struct ubifs_debug_info *dbg;
 #endif
 };
 
@@ -1505,7 +1497,7 @@
 long long ubifs_get_free_space(struct ubifs_info *c);
 int ubifs_calc_min_idx_lebs(struct ubifs_info *c);
 void ubifs_convert_page_budget(struct ubifs_info *c);
-long long ubifs_reported_space(const struct ubifs_info *c, uint64_t free);
+long long ubifs_reported_space(const struct ubifs_info *c, long long free);
 long long ubifs_calc_available(const struct ubifs_info *c, int min_idx_lebs);
 
 /* find.c */
@@ -1639,6 +1631,9 @@
 void ubifs_add_nnode_dirt(struct ubifs_info *c, struct ubifs_nnode *nnode);
 uint32_t ubifs_unpack_bits(uint8_t **addr, int *pos, int nrbits);
 struct ubifs_nnode *ubifs_first_nnode(struct ubifs_info *c, int *hght);
+/* Needed only in debugging code in lpt_commit.c */
+int ubifs_unpack_nnode(const struct ubifs_info *c, void *buf,
+		       struct ubifs_nnode *nnode);
 
 /* lpt_commit.c */
 int ubifs_lpt_start_commit(struct ubifs_info *c);
@@ -1714,7 +1709,7 @@
 
 /* compressor.c */
 int __init ubifs_compressors_init(void);
-void __exit ubifs_compressors_exit(void);
+void ubifs_compressors_exit(void);
 void ubifs_compress(const void *in_buf, int in_len, void *out_buf, int *out_len,
 		    int *compr_type);
 int ubifs_decompress(const void *buf, int len, void *out, int *out_len,
diff --git a/include/asm-generic/topology.h b/include/asm-generic/topology.h
index 54bbf6e..0e9e2bc 100644
--- a/include/asm-generic/topology.h
+++ b/include/asm-generic/topology.h
@@ -40,6 +40,9 @@
 #ifndef node_to_cpumask
 #define node_to_cpumask(node)	((void)node, cpu_online_map)
 #endif
+#ifndef cpumask_of_node
+#define cpumask_of_node(node)	((void)node, cpu_online_mask)
+#endif
 #ifndef node_to_first_cpu
 #define node_to_first_cpu(node)	((void)(node),0)
 #endif
@@ -54,9 +57,18 @@
 				)
 #endif
 
+#ifndef cpumask_of_pcibus
+#define cpumask_of_pcibus(bus)	(pcibus_to_node(bus) == -1 ?		\
+				 cpu_all_mask :				\
+				 cpumask_of_node(pcibus_to_node(bus)))
+#endif
+
 #endif	/* CONFIG_NUMA */
 
-/* returns pointer to cpumask for specified node */
+/*
+ * returns pointer to cpumask for specified node
+ * Deprecated: use "const struct cpumask *mask = cpumask_of_node(node)"
+ */
 #ifndef node_to_cpumask_ptr
 
 #define	node_to_cpumask_ptr(v, node) 					\
diff --git a/include/asm-m32r/smp.h b/include/asm-m32r/smp.h
index c5dd669..b96a6d2 100644
--- a/include/asm-m32r/smp.h
+++ b/include/asm-m32r/smp.h
@@ -63,8 +63,6 @@
 #define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 extern cpumask_t cpu_callout_map;
-extern cpumask_t cpu_possible_map;
-extern cpumask_t cpu_present_map;
 
 static __inline__ int hard_smp_processor_id(void)
 {
diff --git a/include/linux/8250_pci.h b/include/linux/8250_pci.h
index 3209dd4..b24ff08 100644
--- a/include/linux/8250_pci.h
+++ b/include/linux/8250_pci.h
@@ -31,7 +31,7 @@
 struct serial_private;
 
 struct serial_private *
-pciserial_init_ports(struct pci_dev *dev, struct pciserial_board *board);
+pciserial_init_ports(struct pci_dev *dev, const struct pciserial_board *board);
 void pciserial_remove_ports(struct serial_private *priv);
 void pciserial_suspend_ports(struct serial_private *priv);
 void pciserial_resume_ports(struct serial_private *priv);
diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index ed3a5d4..cea1536 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -82,13 +82,13 @@
 	int			shift;
 	int			rating;
 	int			irq;
-	cpumask_t		cpumask;
+	const struct cpumask	*cpumask;
 	int			(*set_next_event)(unsigned long evt,
 						  struct clock_event_device *);
 	void			(*set_mode)(enum clock_event_mode mode,
 					    struct clock_event_device *);
 	void			(*event_handler)(struct clock_event_device *);
-	void			(*broadcast)(cpumask_t mask);
+	void			(*broadcast)(const struct cpumask *mask);
 	struct list_head	list;
 	enum clock_event_mode	mode;
 	ktime_t			next_event;
diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index 5c8351b..af40f8e 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -61,3 +61,8 @@
 #define  noinline			__attribute__((noinline))
 #define __attribute_const__		__attribute__((__const__))
 #define __maybe_unused			__attribute__((unused))
+
+#define __gcc_header(x) #x
+#define _gcc_header(x) __gcc_header(linux/compiler-gcc##x.h)
+#define gcc_header(x) _gcc_header(x)
+#include gcc_header(__GNUC__)
diff --git a/include/linux/compiler-gcc3.h b/include/linux/compiler-gcc3.h
index e5eb795..8005eff 100644
--- a/include/linux/compiler-gcc3.h
+++ b/include/linux/compiler-gcc3.h
@@ -2,8 +2,9 @@
 #error "Please don't include <linux/compiler-gcc3.h> directly, include <linux/compiler.h> instead."
 #endif
 
-/* These definitions are for GCC v3.x.  */
-#include <linux/compiler-gcc.h>
+#if __GNUC_MINOR__ < 2
+# error Sorry, your compiler is too old - please upgrade it.
+#endif
 
 #if __GNUC_MINOR__ >= 3
 # define __used			__attribute__((__used__))
diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h
index 974f5b7..0999271 100644
--- a/include/linux/compiler-gcc4.h
+++ b/include/linux/compiler-gcc4.h
@@ -2,8 +2,10 @@
 #error "Please don't include <linux/compiler-gcc4.h> directly, include <linux/compiler.h> instead."
 #endif
 
-/* These definitions are for GCC v4.x.  */
-#include <linux/compiler-gcc.h>
+/* GCC 4.1.[01] miscompiles __weak */
+#if __GNUC_MINOR__ == 1 && __GNUC_PATCHLEVEL__ <= 1
+# error Your version of gcc miscompiles the __weak directive
+#endif
 
 #define __used			__attribute__((__used__))
 #define __must_check 		__attribute__((warn_unused_result))
@@ -16,7 +18,7 @@
  */
 #define uninitialized_var(x) x = x
 
-#if !(__GNUC__ == 4 && __GNUC_MINOR__ < 3)
+#if __GNUC_MINOR__ >= 3
 /* Mark functions as cold. gcc will assume any path leading to a call
    to them will be unlikely.  This means a lot of manual unlikely()s
    are unnecessary now for any paths leading to the usual suspects
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index ea7c6be..d95da10 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -36,12 +36,8 @@
 
 #ifdef __KERNEL__
 
-#if __GNUC__ >= 4
-# include <linux/compiler-gcc4.h>
-#elif __GNUC__ == 3 && __GNUC_MINOR__ >= 2
-# include <linux/compiler-gcc3.h>
-#else
-# error Sorry, your compiler is too old/not recognized.
+#ifdef __GNUC__
+#include <linux/compiler-gcc.h>
 #endif
 
 #define notrace __attribute__((no_instrument_function))
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 21e1dd4..d4bf526 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -339,36 +339,6 @@
 #endif
 #define	CPUMASK_PTR(v, m) 	cpumask_t *v = &(m->v)
 
-#define cpumask_scnprintf(buf, len, src) \
-			__cpumask_scnprintf((buf), (len), &(src), NR_CPUS)
-static inline int __cpumask_scnprintf(char *buf, int len,
-					const cpumask_t *srcp, int nbits)
-{
-	return bitmap_scnprintf(buf, len, srcp->bits, nbits);
-}
-
-#define cpumask_parse_user(ubuf, ulen, dst) \
-			__cpumask_parse_user((ubuf), (ulen), &(dst), NR_CPUS)
-static inline int __cpumask_parse_user(const char __user *buf, int len,
-					cpumask_t *dstp, int nbits)
-{
-	return bitmap_parse_user(buf, len, dstp->bits, nbits);
-}
-
-#define cpulist_scnprintf(buf, len, src) \
-			__cpulist_scnprintf((buf), (len), &(src), NR_CPUS)
-static inline int __cpulist_scnprintf(char *buf, int len,
-					const cpumask_t *srcp, int nbits)
-{
-	return bitmap_scnlistprintf(buf, len, srcp->bits, nbits);
-}
-
-#define cpulist_parse(buf, dst) __cpulist_parse((buf), &(dst), NR_CPUS)
-static inline int __cpulist_parse(const char *buf, cpumask_t *dstp, int nbits)
-{
-	return bitmap_parselist(buf, dstp->bits, nbits);
-}
-
 #define cpu_remap(oldbit, old, new) \
 		__cpu_remap((oldbit), &(old), &(new), NR_CPUS)
 static inline int __cpu_remap(int oldbit,
@@ -540,9 +510,6 @@
 	[BITS_TO_LONGS(NR_CPUS)-1] = CPU_MASK_LAST_WORD	\
 }
 
-/* This produces more efficient code. */
-#define nr_cpumask_bits	NR_CPUS
-
 #else /* NR_CPUS > BITS_PER_LONG */
 
 #define CPU_BITS_ALL						\
@@ -550,10 +517,16 @@
 	[0 ... BITS_TO_LONGS(NR_CPUS)-2] = ~0UL,		\
 	[BITS_TO_LONGS(NR_CPUS)-1] = CPU_MASK_LAST_WORD		\
 }
-
-#define nr_cpumask_bits	nr_cpu_ids
 #endif /* NR_CPUS > BITS_PER_LONG */
 
+#ifdef CONFIG_CPUMASK_OFFSTACK
+/* Assuming NR_CPUS is huge, a runtime limit is more efficient.  Also,
+ * not all bits may be allocated. */
+#define nr_cpumask_bits	nr_cpu_ids
+#else
+#define nr_cpumask_bits	NR_CPUS
+#endif
+
 /* verify cpu argument to cpumask_* operators */
 static inline unsigned int cpumask_check(unsigned int cpu)
 {
@@ -946,6 +919,63 @@
 #define cpumask_of(cpu) (get_cpu_mask(cpu))
 
 /**
+ * cpumask_scnprintf - print a cpumask into a string as comma-separated hex
+ * @buf: the buffer to sprintf into
+ * @len: the length of the buffer
+ * @srcp: the cpumask to print
+ *
+ * If len is zero, returns zero.  Otherwise returns the length of the
+ * (nul-terminated) @buf string.
+ */
+static inline int cpumask_scnprintf(char *buf, int len,
+				    const struct cpumask *srcp)
+{
+	return bitmap_scnprintf(buf, len, srcp->bits, nr_cpumask_bits);
+}
+
+/**
+ * cpumask_parse_user - extract a cpumask from a user string
+ * @buf: the buffer to extract from
+ * @len: the length of the buffer
+ * @dstp: the cpumask to set.
+ *
+ * Returns -errno, or 0 for success.
+ */
+static inline int cpumask_parse_user(const char __user *buf, int len,
+				     struct cpumask *dstp)
+{
+	return bitmap_parse_user(buf, len, dstp->bits, nr_cpumask_bits);
+}
+
+/**
+ * cpulist_scnprintf - print a cpumask into a string as comma-separated list
+ * @buf: the buffer to sprintf into
+ * @len: the length of the buffer
+ * @srcp: the cpumask to print
+ *
+ * If len is zero, returns zero.  Otherwise returns the length of the
+ * (nul-terminated) @buf string.
+ */
+static inline int cpulist_scnprintf(char *buf, int len,
+				    const struct cpumask *srcp)
+{
+	return bitmap_scnlistprintf(buf, len, srcp->bits, nr_cpumask_bits);
+}
+
+/**
+ * cpulist_parse_user - extract a cpumask from a user string of ranges
+ * @buf: the buffer to extract from
+ * @len: the length of the buffer
+ * @dstp: the cpumask to set.
+ *
+ * Returns -errno, or 0 for success.
+ */
+static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
+{
+	return bitmap_parselist(buf, dstp->bits, nr_cpumask_bits);
+}
+
+/**
  * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
  * @bitmap: the bitmap
  *
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index a37359d..c66d224 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -75,14 +75,22 @@
 	return end_name_hash(hash);
 }
 
-struct dcookie_struct;
-
-#define DNAME_INLINE_LEN_MIN 36
+/*
+ * Try to keep struct dentry aligned on 64 byte cachelines (this will
+ * give reasonable cacheline footprint with larger lines without the
+ * large memory footprint increase).
+ */
+#ifdef CONFIG_64BIT
+#define DNAME_INLINE_LEN_MIN 32 /* 192 bytes */
+#else
+#define DNAME_INLINE_LEN_MIN 40 /* 128 bytes */
+#endif
 
 struct dentry {
 	atomic_t d_count;
 	unsigned int d_flags;		/* protected by d_lock */
 	spinlock_t d_lock;		/* per dentry lock */
+	int d_mounted;
 	struct inode *d_inode;		/* Where the name belongs to - NULL is
 					 * negative */
 	/*
@@ -107,10 +115,7 @@
 	struct dentry_operations *d_op;
 	struct super_block *d_sb;	/* The root of the dentry tree */
 	void *d_fsdata;			/* fs-specific data */
-#ifdef CONFIG_PROFILING
-	struct dcookie_struct *d_cookie; /* cookie, if any */
-#endif
-	int d_mounted;
+
 	unsigned char d_iname[DNAME_INLINE_LEN_MIN];	/* small names */
 };
 
@@ -177,6 +182,8 @@
 
 #define DCACHE_INOTIFY_PARENT_WATCHED	0x0020 /* Parent inode is watched */
 
+#define DCACHE_COOKIE		0x0040	/* For use by dcookie subsystem */
+
 extern spinlock_t dcache_lock;
 extern seqlock_t rename_lock;
 
diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h
index 4aab6f1..09d6c5b 100644
--- a/include/linux/fdtable.h
+++ b/include/linux/fdtable.h
@@ -57,8 +57,6 @@
 
 #define files_fdtable(files) (rcu_dereference((files)->fdt))
 
-extern struct kmem_cache *filp_cachep;
-
 struct file_operations;
 struct vfsmount;
 struct dentry;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 001ded4..e2170ee 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -21,7 +21,6 @@
 
 /* Fixed constants first: */
 #undef NR_OPEN
-extern int sysctl_nr_open;
 #define INR_OPEN 1024		/* Initial setting for nfile rlimits */
 
 #define BLOCK_SIZE_BITS 10
@@ -38,21 +37,13 @@
 	int nr_free_files;	/* read only */
 	int max_files;		/* tunable */
 };
-extern struct files_stat_struct files_stat;
-extern int get_max_files(void);
 
 struct inodes_stat_t {
 	int nr_inodes;
 	int nr_unused;
 	int dummy[5];		/* padding for sysctl ABI compatibility */
 };
-extern struct inodes_stat_t inodes_stat;
 
-extern int leases_enable, lease_break_time;
-
-#ifdef CONFIG_DNOTIFY
-extern int dir_notify_enable;
-#endif
 
 #define NR_FILE  8192	/* this can well be larger on a larger system */
 
@@ -330,6 +321,15 @@
 extern void __init inode_init_early(void);
 extern void __init files_init(unsigned long);
 
+extern struct files_stat_struct files_stat;
+extern int get_max_files(void);
+extern int sysctl_nr_open;
+extern struct inodes_stat_t inodes_stat;
+extern int leases_enable, lease_break_time;
+#ifdef CONFIG_DNOTIFY
+extern int dir_notify_enable;
+#endif
+
 struct buffer_head;
 typedef int (get_block_t)(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh_result, int create);
@@ -1212,7 +1212,6 @@
 /*
  * VFS helper functions..
  */
-extern int vfs_permission(struct nameidata *, int);
 extern int vfs_create(struct inode *, struct dentry *, int, struct nameidata *);
 extern int vfs_mkdir(struct inode *, struct dentry *, int);
 extern int vfs_mknod(struct inode *, struct dentry *, int, dev_t);
@@ -1310,7 +1309,6 @@
 	ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
 	unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
 	int (*check_flags)(int);
-	int (*dir_notify)(struct file *filp, unsigned long arg);
 	int (*flock) (struct file *, int, struct file_lock *);
 	ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int);
 	ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int);
@@ -1869,7 +1867,7 @@
 
 extern struct file *do_filp_open(int dfd, const char *pathname,
 		int open_flag, int mode);
-extern int may_open(struct nameidata *, int, int);
+extern int may_open(struct path *, int, int);
 
 extern int kernel_read(struct file *, unsigned long, char *, unsigned long);
 extern struct file * open_exec(const char *);
@@ -1904,6 +1902,8 @@
 
 extern struct inode * iget5_locked(struct super_block *, unsigned long, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *);
 extern struct inode * iget_locked(struct super_block *, unsigned long);
+extern int insert_inode_locked4(struct inode *, unsigned long, int (*test)(struct inode *, void *), void *);
+extern int insert_inode_locked(struct inode *);
 extern void unlock_new_inode(struct inode *);
 
 extern void __iget(struct inode * inode);
diff --git a/include/linux/fs_struct.h b/include/linux/fs_struct.h
index 9e5a06e..a97c053 100644
--- a/include/linux/fs_struct.h
+++ b/include/linux/fs_struct.h
@@ -10,12 +10,6 @@
 	struct path root, pwd;
 };
 
-#define INIT_FS {				\
-	.count		= ATOMIC_INIT(1),	\
-	.lock		= RW_LOCK_UNLOCKED,	\
-	.umask		= 0022, \
-}
-
 extern struct kmem_cache *fs_cachep;
 
 extern void exit_fs(struct task_struct *);
diff --git a/include/linux/generic_serial.h b/include/linux/generic_serial.h
index 4cc9139..fadff28 100644
--- a/include/linux/generic_serial.h
+++ b/include/linux/generic_serial.h
@@ -21,7 +21,6 @@
   void                    (*enable_tx_interrupts) (void *);
   void                    (*disable_rx_interrupts) (void *);
   void                    (*enable_rx_interrupts) (void *);
-  int                     (*get_CD) (void *);
   void                    (*shutdown_port) (void*);
   int                     (*set_real_termios) (void*);
   int                     (*chars_in_buffer) (void*);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e99c56d..db5ef8a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -32,13 +32,6 @@
 # define SUPPORT_VLB_SYNC 1
 #endif
 
-/*
- * Used to indicate "no IRQ", should be a value that cannot be an IRQ
- * number.
- */
- 
-#define IDE_NO_IRQ		(-1)
-
 typedef unsigned char	byte;	/* used everywhere */
 
 /*
@@ -403,6 +396,7 @@
  * This is used for several packet commands (not for READ/WRITE commands).
  */
 #define IDE_PC_BUFFER_SIZE	256
+#define ATAPI_WAIT_PC		(60 * HZ)
 
 struct ide_atapi_pc {
 	/* actual packet bytes */
@@ -480,53 +474,53 @@
 
 	/* ide-cd */
 	/* Drive cannot eject the disc. */
-	IDE_AFLAG_NO_EJECT		= (1 << 3),
+	IDE_AFLAG_NO_EJECT		= (1 << 1),
 	/* Drive is a pre ATAPI 1.2 drive. */
-	IDE_AFLAG_PRE_ATAPI12		= (1 << 4),
+	IDE_AFLAG_PRE_ATAPI12		= (1 << 2),
 	/* TOC addresses are in BCD. */
-	IDE_AFLAG_TOCADDR_AS_BCD	= (1 << 5),
+	IDE_AFLAG_TOCADDR_AS_BCD	= (1 << 3),
 	/* TOC track numbers are in BCD. */
-	IDE_AFLAG_TOCTRACKS_AS_BCD	= (1 << 6),
+	IDE_AFLAG_TOCTRACKS_AS_BCD	= (1 << 4),
 	/*
 	 * Drive does not provide data in multiples of SECTOR_SIZE
 	 * when more than one interrupt is needed.
 	 */
-	IDE_AFLAG_LIMIT_NFRAMES		= (1 << 7),
+	IDE_AFLAG_LIMIT_NFRAMES		= (1 << 5),
 	/* Saved TOC information is current. */
-	IDE_AFLAG_TOC_VALID		= (1 << 9),
+	IDE_AFLAG_TOC_VALID		= (1 << 6),
 	/* We think that the drive door is locked. */
-	IDE_AFLAG_DOOR_LOCKED		= (1 << 10),
+	IDE_AFLAG_DOOR_LOCKED		= (1 << 7),
 	/* SET_CD_SPEED command is unsupported. */
-	IDE_AFLAG_NO_SPEED_SELECT	= (1 << 11),
-	IDE_AFLAG_VERTOS_300_SSD	= (1 << 12),
-	IDE_AFLAG_VERTOS_600_ESD	= (1 << 13),
-	IDE_AFLAG_SANYO_3CD		= (1 << 14),
-	IDE_AFLAG_FULL_CAPS_PAGE	= (1 << 15),
-	IDE_AFLAG_PLAY_AUDIO_OK		= (1 << 16),
-	IDE_AFLAG_LE_SPEED_FIELDS	= (1 << 17),
+	IDE_AFLAG_NO_SPEED_SELECT	= (1 << 8),
+	IDE_AFLAG_VERTOS_300_SSD	= (1 << 9),
+	IDE_AFLAG_VERTOS_600_ESD	= (1 << 10),
+	IDE_AFLAG_SANYO_3CD		= (1 << 11),
+	IDE_AFLAG_FULL_CAPS_PAGE	= (1 << 12),
+	IDE_AFLAG_PLAY_AUDIO_OK		= (1 << 13),
+	IDE_AFLAG_LE_SPEED_FIELDS	= (1 << 14),
 
 	/* ide-floppy */
 	/* Avoid commands not supported in Clik drive */
-	IDE_AFLAG_CLIK_DRIVE		= (1 << 19),
+	IDE_AFLAG_CLIK_DRIVE		= (1 << 15),
 	/* Requires BH algorithm for packets */
-	IDE_AFLAG_ZIP_DRIVE		= (1 << 20),
+	IDE_AFLAG_ZIP_DRIVE		= (1 << 16),
 	/* Supports format progress report */
-	IDE_AFLAG_SRFP			= (1 << 22),
+	IDE_AFLAG_SRFP			= (1 << 17),
 
 	/* ide-tape */
-	IDE_AFLAG_IGNORE_DSC		= (1 << 23),
+	IDE_AFLAG_IGNORE_DSC		= (1 << 18),
 	/* 0 When the tape position is unknown */
-	IDE_AFLAG_ADDRESS_VALID		= (1 <<	24),
+	IDE_AFLAG_ADDRESS_VALID		= (1 <<	19),
 	/* Device already opened */
-	IDE_AFLAG_BUSY			= (1 << 25),
+	IDE_AFLAG_BUSY			= (1 << 20),
 	/* Attempt to auto-detect the current user block size */
-	IDE_AFLAG_DETECT_BS		= (1 << 26),
+	IDE_AFLAG_DETECT_BS		= (1 << 21),
 	/* Currently on a filemark */
-	IDE_AFLAG_FILEMARK		= (1 << 27),
+	IDE_AFLAG_FILEMARK		= (1 << 22),
 	/* 0 = no tape is loaded, so we don't rewind after ejecting */
-	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 28),
+	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 23),
 
-	IDE_AFLAG_NO_AUTOCLOSE		= (1 << 29),
+	IDE_AFLAG_NO_AUTOCLOSE		= (1 << 24),
 };
 
 /* device flags */
@@ -565,28 +559,26 @@
 	IDE_DFLAG_NODMA			= (1 << 16),
 	/* powermanagment told us not to do anything, so sleep nicely */
 	IDE_DFLAG_BLOCKED		= (1 << 17),
-	/* ide-scsi emulation */
-	IDE_DFLAG_SCSI			= (1 << 18),
 	/* sleeping & sleep field valid */
-	IDE_DFLAG_SLEEPING		= (1 << 19),
-	IDE_DFLAG_POST_RESET		= (1 << 20),
-	IDE_DFLAG_UDMA33_WARNED		= (1 << 21),
-	IDE_DFLAG_LBA48			= (1 << 22),
+	IDE_DFLAG_SLEEPING		= (1 << 18),
+	IDE_DFLAG_POST_RESET		= (1 << 19),
+	IDE_DFLAG_UDMA33_WARNED		= (1 << 20),
+	IDE_DFLAG_LBA48			= (1 << 21),
 	/* status of write cache */
-	IDE_DFLAG_WCACHE		= (1 << 23),
+	IDE_DFLAG_WCACHE		= (1 << 22),
 	/* used for ignoring ATA_DF */
-	IDE_DFLAG_NOWERR		= (1 << 24),
+	IDE_DFLAG_NOWERR		= (1 << 23),
 	/* retrying in PIO */
-	IDE_DFLAG_DMA_PIO_RETRY		= (1 << 25),
-	IDE_DFLAG_LBA			= (1 << 26),
+	IDE_DFLAG_DMA_PIO_RETRY		= (1 << 24),
+	IDE_DFLAG_LBA			= (1 << 25),
 	/* don't unload heads */
-	IDE_DFLAG_NO_UNLOAD		= (1 << 27),
+	IDE_DFLAG_NO_UNLOAD		= (1 << 26),
 	/* heads unloaded, please don't reset port */
-	IDE_DFLAG_PARKED		= (1 << 28),
-	IDE_DFLAG_MEDIA_CHANGED		= (1 << 29),
+	IDE_DFLAG_PARKED		= (1 << 27),
+	IDE_DFLAG_MEDIA_CHANGED		= (1 << 28),
 	/* write protect */
-	IDE_DFLAG_WP			= (1 << 30),
-	IDE_DFLAG_FORMAT_IN_PROGRESS	= (1 << 31),
+	IDE_DFLAG_WP			= (1 << 29),
+	IDE_DFLAG_FORMAT_IN_PROGRESS	= (1 << 30),
 };
 
 struct ide_drive_s {
@@ -610,8 +602,6 @@
 	unsigned long dev_flags;
 
 	unsigned long sleep;		/* sleep until this time */
-	unsigned long service_start;	/* time we started last request */
-	unsigned long service_time;	/* service time of last request */
 	unsigned long timeout;		/* max time to wait for irq */
 
 	special_t	special;	/* special action flags */
@@ -879,8 +869,6 @@
 
 		/* BOOL: protects all fields below */
 	volatile int busy;
-		/* BOOL: wake us up on timer expiry */
-	unsigned int sleeping	: 1;
 		/* BOOL: polling active & poll_timeout field valid */
 	unsigned int polling	: 1;
 
@@ -1258,14 +1246,11 @@
 void ide_create_request_sense_cmd(ide_drive_t *, struct ide_atapi_pc *);
 void ide_retry_pc(ide_drive_t *, struct gendisk *);
 
-static inline unsigned long ide_scsi_get_timeout(struct ide_atapi_pc *pc)
-{
-	return max_t(unsigned long, WAIT_CMD, pc->timeout - jiffies);
-}
+int ide_cd_expiry(ide_drive_t *);
 
-int ide_scsi_expiry(ide_drive_t *);
+int ide_cd_get_xferlen(struct request *);
 
-ide_startstop_t ide_issue_pc(ide_drive_t *, unsigned int, ide_expiry_t *);
+ide_startstop_t ide_issue_pc(ide_drive_t *);
 
 ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
@@ -1287,6 +1272,26 @@
 
 extern void ide_timer_expiry(unsigned long);
 extern irqreturn_t ide_intr(int irq, void *dev_id);
+
+static inline int ide_lock_hwgroup(ide_hwgroup_t *hwgroup)
+{
+	if (hwgroup->busy)
+		return 1;
+
+	hwgroup->busy = 1;
+	/* for atari only */
+	ide_get_lock(ide_intr, hwgroup);
+
+	return 0;
+}
+
+static inline void ide_unlock_hwgroup(ide_hwgroup_t *hwgroup)
+{
+	/* for atari only */
+	ide_release_lock();
+	hwgroup->busy = 0;
+}
+
 extern void do_ide_request(struct request_queue *);
 
 void ide_init_disk(struct gendisk *, ide_drive_t *);
@@ -1533,6 +1538,7 @@
 void ide_undecoded_slave(ide_drive_t *);
 
 void ide_port_apply_params(ide_hwif_t *);
+int ide_sysfs_register_port(ide_hwif_t *);
 
 struct ide_host *ide_host_alloc(const struct ide_port_info *, hw_regs_t **);
 void ide_host_free(struct ide_host *);
@@ -1627,6 +1633,9 @@
 
 #define local_irq_set(flags)	do { local_save_flags((flags)); local_irq_enable_in_hardirq(); } while (0)
 
+char *ide_media_string(ide_drive_t *);
+
+extern struct device_attribute ide_dev_attrs[];
 extern struct bus_type ide_bus_type;
 extern struct class *ide_port_class;
 
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 959f552..2f3c2d4 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -12,6 +12,7 @@
 #include <net/net_namespace.h>
 
 extern struct files_struct init_files;
+extern struct fs_struct init_fs;
 
 #define INIT_KIOCTX(name, which_mm) \
 {							\
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index be3c484..990355f 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -111,13 +111,13 @@
 
 extern cpumask_t irq_default_affinity;
 
-extern int irq_set_affinity(unsigned int irq, cpumask_t cpumask);
+extern int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask);
 extern int irq_can_set_affinity(unsigned int irq);
 extern int irq_select_affinity(unsigned int irq);
 
 #else /* CONFIG_SMP */
 
-static inline int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
+static inline int irq_set_affinity(unsigned int irq, const struct cpumask *m)
 {
 	return -EINVAL;
 }
@@ -464,4 +464,10 @@
 
 int show_interrupts(struct seq_file *p, void *v);
 
+struct irq_desc;
+
+extern int early_irq_init(void);
+extern int arch_early_irq_init(void);
+extern int arch_init_chip_data(struct irq_desc *desc, int cpu);
+
 #endif
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 98564dc..f899b50 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -113,7 +113,8 @@
 	void		(*eoi)(unsigned int irq);
 
 	void		(*end)(unsigned int irq);
-	void		(*set_affinity)(unsigned int irq, cpumask_t dest);
+	void		(*set_affinity)(unsigned int irq,
+					const struct cpumask *dest);
 	int		(*retrigger)(unsigned int irq);
 	int		(*set_type)(unsigned int irq, unsigned int flow_type);
 	int		(*set_wake)(unsigned int irq, unsigned int on);
@@ -193,42 +194,23 @@
 	const char		*name;
 } ____cacheline_internodealigned_in_smp;
 
-extern void early_irq_init(void);
-extern void arch_early_irq_init(void);
-extern void arch_init_chip_data(struct irq_desc *desc, int cpu);
 extern void arch_init_copy_chip_data(struct irq_desc *old_desc,
 					struct irq_desc *desc, int cpu);
 extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc);
 
 #ifndef CONFIG_SPARSE_IRQ
 extern struct irq_desc irq_desc[NR_IRQS];
-
-static inline struct irq_desc *irq_to_desc(unsigned int irq)
-{
-	return (irq < NR_IRQS) ? irq_desc + irq : NULL;
-}
-static inline struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu)
-{
-	return irq_to_desc(irq);
-}
-
-#else
-
-extern struct irq_desc *irq_to_desc(unsigned int irq);
-extern struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu);
+#else /* CONFIG_SPARSE_IRQ */
 extern struct irq_desc *move_irq_desc(struct irq_desc *old_desc, int cpu);
 
-# define for_each_irq_desc(irq, desc)		\
-	for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs; irq++, desc = irq_to_desc(irq))
-# define for_each_irq_desc_reverse(irq, desc)                          \
-	for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0; irq--, desc = irq_to_desc(irq))
-
 #define kstat_irqs_this_cpu(DESC) \
 	((DESC)->kstat_irqs[smp_processor_id()])
 #define kstat_incr_irqs_this_cpu(irqno, DESC) \
 	((DESC)->kstat_irqs[smp_processor_id()]++)
 
-#endif
+#endif /* CONFIG_SPARSE_IRQ */
+
+extern struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu);
 
 static inline struct irq_desc *
 irq_remap_to_desc(unsigned int irq, struct irq_desc *desc)
diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h
index 95d2b74..5504a5c 100644
--- a/include/linux/irqnr.h
+++ b/include/linux/irqnr.h
@@ -15,20 +15,23 @@
 
 # define for_each_irq_desc_reverse(irq, desc)                          \
 	for (irq = nr_irqs - 1; irq >= 0; irq--)
-#else
+#else /* CONFIG_GENERIC_HARDIRQS */
 
 extern int nr_irqs;
+extern struct irq_desc *irq_to_desc(unsigned int irq);
 
-#ifndef CONFIG_SPARSE_IRQ
+# define for_each_irq_desc(irq, desc)					\
+	for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs;		\
+	     irq++, desc = irq_to_desc(irq))				\
+		if (desc)
 
-struct irq_desc;
-# define for_each_irq_desc(irq, desc)		\
-	for (irq = 0, desc = irq_desc; irq < nr_irqs; irq++, desc++)
-# define for_each_irq_desc_reverse(irq, desc)                          \
-	for (irq = nr_irqs - 1, desc = irq_desc + (nr_irqs - 1);        \
-	    irq >= 0; irq--, desc--)
-#endif
-#endif
+
+# define for_each_irq_desc_reverse(irq, desc)				\
+	for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0;	\
+	     irq--, desc = irq_to_desc(irq))				\
+		if (desc)
+
+#endif /* CONFIG_GENERIC_HARDIRQS */
 
 #define for_each_irq_nr(irq)                   \
        for (irq = 0; irq < nr_irqs; irq++)
diff --git a/include/linux/istallion.h b/include/linux/istallion.h
index 0d18407..7faca98 100644
--- a/include/linux/istallion.h
+++ b/include/linux/istallion.h
@@ -59,9 +59,7 @@
 	unsigned int		devnr;
 	int			baud_base;
 	int			custom_divisor;
-	int			close_delay;
 	int			closing_wait;
-	int			openwaitcnt;
 	int			rc;
 	int			argsize;
 	void			*argp;
diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index 4ee4b3d..570d204 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -79,10 +79,13 @@
 }
 
 extern unsigned long long task_delta_exec(struct task_struct *);
-extern void account_user_time(struct task_struct *, cputime_t);
-extern void account_user_time_scaled(struct task_struct *, cputime_t);
-extern void account_system_time(struct task_struct *, int, cputime_t);
-extern void account_system_time_scaled(struct task_struct *, cputime_t);
-extern void account_steal_time(struct task_struct *, cputime_t);
+extern void account_user_time(struct task_struct *, cputime_t, cputime_t);
+extern void account_system_time(struct task_struct *, int, cputime_t, cputime_t);
+extern void account_steal_time(cputime_t);
+extern void account_idle_time(cputime_t);
+
+extern void account_process_tick(struct task_struct *, int user);
+extern void account_steal_ticks(unsigned long ticks);
+extern void account_idle_ticks(unsigned long ticks);
 
 #endif /* _LINUX_KERNEL_STAT_H */
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index f18b86f..35525ac 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -83,6 +83,7 @@
 #define KVM_EXIT_S390_SIEIC       13
 #define KVM_EXIT_S390_RESET       14
 #define KVM_EXIT_DCR              15
+#define KVM_EXIT_NMI              16
 
 /* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */
 struct kvm_run {
@@ -387,6 +388,14 @@
 #define KVM_CAP_DEVICE_ASSIGNMENT 17
 #endif
 #define KVM_CAP_IOMMU 18
+#if defined(CONFIG_X86)
+#define KVM_CAP_DEVICE_MSI 20
+#endif
+/* Bug in KVM_SET_USER_MEMORY_REGION fixed: */
+#define KVM_CAP_DESTROY_MEMORY_REGION_WORKS 21
+#if defined(CONFIG_X86)
+#define KVM_CAP_USER_NMI 22
+#endif
 
 /*
  * ioctls for VM fds
@@ -458,6 +467,8 @@
 #define KVM_S390_INITIAL_RESET    _IO(KVMIO,  0x97)
 #define KVM_GET_MP_STATE          _IOR(KVMIO,  0x98, struct kvm_mp_state)
 #define KVM_SET_MP_STATE          _IOW(KVMIO,  0x99, struct kvm_mp_state)
+/* Available with KVM_CAP_NMI */
+#define KVM_NMI                   _IO(KVMIO,  0x9a)
 
 #define KVM_TRC_INJ_VIRQ         (KVM_TRC_HANDLER + 0x02)
 #define KVM_TRC_REDELIVER_EVT    (KVM_TRC_HANDLER + 0x03)
@@ -500,10 +511,17 @@
 	__u32 guest_irq;
 	__u32 flags;
 	union {
+		struct {
+			__u32 addr_lo;
+			__u32 addr_hi;
+			__u32 data;
+		} guest_msi;
 		__u32 reserved[12];
 	};
 };
 
 #define KVM_DEV_ASSIGN_ENABLE_IOMMU	(1 << 0)
 
+#define KVM_DEV_IRQ_ASSIGN_ENABLE_MSI	(1 << 0)
+
 #endif
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index bb92be2..eafabd5 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -16,6 +16,7 @@
 #include <linux/mm.h>
 #include <linux/preempt.h>
 #include <linux/marker.h>
+#include <linux/msi.h>
 #include <asm/signal.h>
 
 #include <linux/kvm.h>
@@ -306,8 +307,14 @@
 	int host_busnr;
 	int host_devfn;
 	int host_irq;
+	bool host_irq_disabled;
 	int guest_irq;
-	int irq_requested;
+	struct msi_msg guest_msi;
+#define KVM_ASSIGNED_DEV_GUEST_INTX	(1 << 0)
+#define KVM_ASSIGNED_DEV_GUEST_MSI	(1 << 1)
+#define KVM_ASSIGNED_DEV_HOST_INTX	(1 << 8)
+#define KVM_ASSIGNED_DEV_HOST_MSI	(1 << 9)
+	unsigned long irq_requested_type;
 	int irq_source_id;
 	struct pci_dev *dev;
 	struct kvm *kvm;
@@ -316,8 +323,7 @@
 void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi);
 void kvm_register_irq_ack_notifier(struct kvm *kvm,
 				   struct kvm_irq_ack_notifier *kian);
-void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
-				     struct kvm_irq_ack_notifier *kian);
+void kvm_unregister_irq_ack_notifier(struct kvm_irq_ack_notifier *kian);
 int kvm_request_irq_source_id(struct kvm *kvm);
 void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
 
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 99eb803..fc2e035 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -94,4 +94,9 @@
 	return nd->saved_names[nd->depth];
 }
 
+static inline void nd_terminate_link(void *name, size_t len, size_t maxlen)
+{
+	((char *) name)[min(len, maxlen)] = '\0';
+}
+
 #endif /* _LINUX_NAMEI_H */
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index b6e6944..218c73b 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1766,6 +1766,7 @@
 #define PCI_DEVICE_ID_SIIG_8S_20x_650	0x2081
 #define PCI_DEVICE_ID_SIIG_8S_20x_850	0x2082
 #define PCI_SUBDEVICE_ID_SIIG_QUARTET_SERIAL	0x2050
+#define PCI_SUBDEVICE_ID_SIIG_DUAL_SERIAL	0x2530
 
 #define PCI_VENDOR_ID_RADISYS		0x1331
 
@@ -1795,6 +1796,7 @@
 #define PCI_DEVICE_ID_SEALEVEL_UCOMM232	0x7202
 #define PCI_DEVICE_ID_SEALEVEL_COMM4	0x7401
 #define PCI_DEVICE_ID_SEALEVEL_COMM8	0x7801
+#define PCI_DEVICE_ID_SEALEVEL_7803	0x7803
 #define PCI_DEVICE_ID_SEALEVEL_UCOMM8	0x7804
 
 #define PCI_VENDOR_ID_HYPERCOPE		0x1365
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 8395e71..38a3f4b 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -250,7 +250,7 @@
 extern int runqueue_is_locked(void);
 extern void task_rq_unlock_wait(struct task_struct *p);
 
-extern cpumask_t nohz_cpu_mask;
+extern cpumask_var_t nohz_cpu_mask;
 #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ)
 extern int select_nohz_load_balancer(int cpu);
 #else
@@ -284,7 +284,6 @@
 
 extern void cpu_init (void);
 extern void trap_init(void);
-extern void account_process_tick(struct task_struct *task, int user);
 extern void update_process_times(int user);
 extern void scheduler_tick(void);
 
@@ -758,20 +757,51 @@
 #define SD_SERIALIZE		1024	/* Only a single load balancing instance */
 #define SD_WAKE_IDLE_FAR	2048	/* Gain latency sacrificing cache hit */
 
-#define BALANCE_FOR_MC_POWER	\
-	(sched_smt_power_savings ? SD_POWERSAVINGS_BALANCE : 0)
+enum powersavings_balance_level {
+	POWERSAVINGS_BALANCE_NONE = 0,  /* No power saving load balance */
+	POWERSAVINGS_BALANCE_BASIC,	/* Fill one thread/core/package
+					 * first for long running threads
+					 */
+	POWERSAVINGS_BALANCE_WAKEUP,	/* Also bias task wakeups to semi-idle
+					 * cpu package for power savings
+					 */
+	MAX_POWERSAVINGS_BALANCE_LEVELS
+};
 
-#define BALANCE_FOR_PKG_POWER	\
-	((sched_mc_power_savings || sched_smt_power_savings) ?	\
-	 SD_POWERSAVINGS_BALANCE : 0)
+extern int sched_mc_power_savings, sched_smt_power_savings;
 
-#define test_sd_parent(sd, flag)	((sd->parent &&		\
-					 (sd->parent->flags & flag)) ? 1 : 0)
+static inline int sd_balance_for_mc_power(void)
+{
+	if (sched_smt_power_savings)
+		return SD_POWERSAVINGS_BALANCE;
 
+	return 0;
+}
+
+static inline int sd_balance_for_package_power(void)
+{
+	if (sched_mc_power_savings | sched_smt_power_savings)
+		return SD_POWERSAVINGS_BALANCE;
+
+	return 0;
+}
+
+/*
+ * Optimise SD flags for power savings:
+ * SD_BALANCE_NEWIDLE helps agressive task consolidation and power savings.
+ * Keep default SD flags if sched_{smt,mc}_power_saving=0
+ */
+
+static inline int sd_power_saving_flags(void)
+{
+	if (sched_mc_power_savings | sched_smt_power_savings)
+		return SD_BALANCE_NEWIDLE;
+
+	return 0;
+}
 
 struct sched_group {
 	struct sched_group *next;	/* Must be a circular list */
-	cpumask_t cpumask;
 
 	/*
 	 * CPU power of this group, SCHED_LOAD_SCALE being max power for a
@@ -784,8 +814,15 @@
 	 * (see include/linux/reciprocal_div.h)
 	 */
 	u32 reciprocal_cpu_power;
+
+	unsigned long cpumask[];
 };
 
+static inline struct cpumask *sched_group_cpus(struct sched_group *sg)
+{
+	return to_cpumask(sg->cpumask);
+}
+
 enum sched_domain_level {
 	SD_LV_NONE = 0,
 	SD_LV_SIBLING,
@@ -809,7 +846,6 @@
 	struct sched_domain *parent;	/* top domain must be null terminated */
 	struct sched_domain *child;	/* bottom domain must be null terminated */
 	struct sched_group *groups;	/* the balancing groups of the domain */
-	cpumask_t span;			/* span of all CPUs in this domain */
 	unsigned long min_interval;	/* Minimum balance interval ms */
 	unsigned long max_interval;	/* Maximum balance interval ms */
 	unsigned int busy_factor;	/* less balancing by factor if busy */
@@ -864,18 +900,35 @@
 #ifdef CONFIG_SCHED_DEBUG
 	char *name;
 #endif
+
+	/* span of all CPUs in this domain */
+	unsigned long span[];
 };
 
-extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
+{
+	return to_cpumask(sd->span);
+}
+
+extern void partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 				    struct sched_domain_attr *dattr_new);
 extern int arch_reinit_sched_domains(void);
 
+/* Test a flag in parent sched domain */
+static inline int test_sd_parent(struct sched_domain *sd, int flag)
+{
+	if (sd->parent && (sd->parent->flags & flag))
+		return 1;
+
+	return 0;
+}
+
 #else /* CONFIG_SMP */
 
 struct sched_domain_attr;
 
 static inline void
-partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 			struct sched_domain_attr *dattr_new)
 {
 }
@@ -926,7 +979,7 @@
 	void (*task_wake_up) (struct rq *this_rq, struct task_struct *task);
 
 	void (*set_cpus_allowed)(struct task_struct *p,
-				 const cpumask_t *newmask);
+				 const struct cpumask *newmask);
 
 	void (*rq_online)(struct rq *rq);
 	void (*rq_offline)(struct rq *rq);
@@ -1579,12 +1632,12 @@
 
 #ifdef CONFIG_SMP
 extern int set_cpus_allowed_ptr(struct task_struct *p,
-				const cpumask_t *new_mask);
+				const struct cpumask *new_mask);
 #else
 static inline int set_cpus_allowed_ptr(struct task_struct *p,
-				       const cpumask_t *new_mask)
+				       const struct cpumask *new_mask)
 {
-	if (!cpu_isset(0, *new_mask))
+	if (!cpumask_test_cpu(0, new_mask))
 		return -EINVAL;
 	return 0;
 }
@@ -2195,10 +2248,8 @@
 }
 #endif
 
-extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask);
-extern long sched_getaffinity(pid_t pid, cpumask_t *mask);
-
-extern int sched_mc_power_savings, sched_smt_power_savings;
+extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
+extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
 
 extern void normalize_rt_tasks(void);
 
diff --git a/include/linux/security.h b/include/linux/security.h
index 3416cb8..b92b5e4 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -335,17 +335,37 @@
  *	@dir contains the inode structure of the parent directory of the new link.
  *	@new_dentry contains the dentry structure for the new link.
  *	Return 0 if permission is granted.
+ * @path_link:
+ *	Check permission before creating a new hard link to a file.
+ *	@old_dentry contains the dentry structure for an existing link
+ *	to the file.
+ *	@new_dir contains the path structure of the parent directory of
+ *	the new link.
+ *	@new_dentry contains the dentry structure for the new link.
+ *	Return 0 if permission is granted.
  * @inode_unlink:
  *	Check the permission to remove a hard link to a file.
  *	@dir contains the inode structure of parent directory of the file.
  *	@dentry contains the dentry structure for file to be unlinked.
  *	Return 0 if permission is granted.
+ * @path_unlink:
+ *	Check the permission to remove a hard link to a file.
+ *	@dir contains the path structure of parent directory of the file.
+ *	@dentry contains the dentry structure for file to be unlinked.
+ *	Return 0 if permission is granted.
  * @inode_symlink:
  *	Check the permission to create a symbolic link to a file.
  *	@dir contains the inode structure of parent directory of the symbolic link.
  *	@dentry contains the dentry structure of the symbolic link.
  *	@old_name contains the pathname of file.
  *	Return 0 if permission is granted.
+ * @path_symlink:
+ *	Check the permission to create a symbolic link to a file.
+ *	@dir contains the path structure of parent directory of
+ *	the symbolic link.
+ *	@dentry contains the dentry structure of the symbolic link.
+ *	@old_name contains the pathname of file.
+ *	Return 0 if permission is granted.
  * @inode_mkdir:
  *	Check permissions to create a new directory in the existing directory
  *	associated with inode strcture @dir.
@@ -353,11 +373,25 @@
  *	@dentry contains the dentry structure of new directory.
  *	@mode contains the mode of new directory.
  *	Return 0 if permission is granted.
+ * @path_mkdir:
+ *	Check permissions to create a new directory in the existing directory
+ *	associated with path strcture @path.
+ *	@dir containst the path structure of parent of the directory
+ *	to be created.
+ *	@dentry contains the dentry structure of new directory.
+ *	@mode contains the mode of new directory.
+ *	Return 0 if permission is granted.
  * @inode_rmdir:
  *	Check the permission to remove a directory.
  *	@dir contains the inode structure of parent of the directory to be removed.
  *	@dentry contains the dentry structure of directory to be removed.
  *	Return 0 if permission is granted.
+ * @path_rmdir:
+ *	Check the permission to remove a directory.
+ *	@dir contains the path structure of parent of the directory to be
+ *	removed.
+ *	@dentry contains the dentry structure of directory to be removed.
+ *	Return 0 if permission is granted.
  * @inode_mknod:
  *	Check permissions when creating a special file (or a socket or a fifo
  *	file created via the mknod system call).  Note that if mknod operation
@@ -368,6 +402,15 @@
  *	@mode contains the mode of the new file.
  *	@dev contains the device number.
  *	Return 0 if permission is granted.
+ * @path_mknod:
+ *	Check permissions when creating a file. Note that this hook is called
+ *	even if mknod operation is being done for a regular file.
+ *	@dir contains the path structure of parent of the new file.
+ *	@dentry contains the dentry structure of the new file.
+ *	@mode contains the mode of the new file.
+ *	@dev contains the undecoded device number. Use new_decode_dev() to get
+ *	the decoded device number.
+ *	Return 0 if permission is granted.
  * @inode_rename:
  *	Check for permission to rename a file or directory.
  *	@old_dir contains the inode structure for parent of the old link.
@@ -375,6 +418,13 @@
  *	@new_dir contains the inode structure for parent of the new link.
  *	@new_dentry contains the dentry structure of the new link.
  *	Return 0 if permission is granted.
+ * @path_rename:
+ *	Check for permission to rename a file or directory.
+ *	@old_dir contains the path structure for parent of the old link.
+ *	@old_dentry contains the dentry structure of the old link.
+ *	@new_dir contains the path structure for parent of the new link.
+ *	@new_dentry contains the dentry structure of the new link.
+ *	Return 0 if permission is granted.
  * @inode_readlink:
  *	Check the permission to read the symbolic link.
  *	@dentry contains the dentry structure for the file link.
@@ -403,6 +453,12 @@
  *	@dentry contains the dentry structure for the file.
  *	@attr is the iattr structure containing the new file attributes.
  *	Return 0 if permission is granted.
+ * @path_truncate:
+ *	Check permission before truncating a file.
+ *	@path contains the path structure for the file.
+ *	@length is the new length of the file.
+ *	@time_attrs is the flags passed to do_truncate().
+ *	Return 0 if permission is granted.
  * @inode_getattr:
  *	Check permission before obtaining file attributes.
  *	@mnt is the vfsmount where the dentry was looked up
@@ -1331,6 +1387,22 @@
 				   struct super_block *newsb);
 	int (*sb_parse_opts_str) (char *options, struct security_mnt_opts *opts);
 
+#ifdef CONFIG_SECURITY_PATH
+	int (*path_unlink) (struct path *dir, struct dentry *dentry);
+	int (*path_mkdir) (struct path *dir, struct dentry *dentry, int mode);
+	int (*path_rmdir) (struct path *dir, struct dentry *dentry);
+	int (*path_mknod) (struct path *dir, struct dentry *dentry, int mode,
+			   unsigned int dev);
+	int (*path_truncate) (struct path *path, loff_t length,
+			      unsigned int time_attrs);
+	int (*path_symlink) (struct path *dir, struct dentry *dentry,
+			     const char *old_name);
+	int (*path_link) (struct dentry *old_dentry, struct path *new_dir,
+			  struct dentry *new_dentry);
+	int (*path_rename) (struct path *old_dir, struct dentry *old_dentry,
+			    struct path *new_dir, struct dentry *new_dentry);
+#endif
+
 	int (*inode_alloc_security) (struct inode *inode);
 	void (*inode_free_security) (struct inode *inode);
 	int (*inode_init_security) (struct inode *inode, struct inode *dir,
@@ -2705,6 +2777,71 @@
 
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 
+#ifdef CONFIG_SECURITY_PATH
+int security_path_unlink(struct path *dir, struct dentry *dentry);
+int security_path_mkdir(struct path *dir, struct dentry *dentry, int mode);
+int security_path_rmdir(struct path *dir, struct dentry *dentry);
+int security_path_mknod(struct path *dir, struct dentry *dentry, int mode,
+			unsigned int dev);
+int security_path_truncate(struct path *path, loff_t length,
+			   unsigned int time_attrs);
+int security_path_symlink(struct path *dir, struct dentry *dentry,
+			  const char *old_name);
+int security_path_link(struct dentry *old_dentry, struct path *new_dir,
+		       struct dentry *new_dentry);
+int security_path_rename(struct path *old_dir, struct dentry *old_dentry,
+			 struct path *new_dir, struct dentry *new_dentry);
+#else	/* CONFIG_SECURITY_PATH */
+static inline int security_path_unlink(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static inline int security_path_mkdir(struct path *dir, struct dentry *dentry,
+				      int mode)
+{
+	return 0;
+}
+
+static inline int security_path_rmdir(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static inline int security_path_mknod(struct path *dir, struct dentry *dentry,
+				      int mode, unsigned int dev)
+{
+	return 0;
+}
+
+static inline int security_path_truncate(struct path *path, loff_t length,
+					 unsigned int time_attrs)
+{
+	return 0;
+}
+
+static inline int security_path_symlink(struct path *dir, struct dentry *dentry,
+					const char *old_name)
+{
+	return 0;
+}
+
+static inline int security_path_link(struct dentry *old_dentry,
+				     struct path *new_dir,
+				     struct dentry *new_dentry)
+{
+	return 0;
+}
+
+static inline int security_path_rename(struct path *old_dir,
+				       struct dentry *old_dentry,
+				       struct path *new_dir,
+				       struct dentry *new_dentry)
+{
+	return 0;
+}
+#endif	/* CONFIG_SECURITY_PATH */
+
 #ifdef CONFIG_KEYS
 #ifdef CONFIG_SECURITY
 
diff --git a/include/linux/serial.h b/include/linux/serial.h
index 1ea8d92..9136cc5 100644
--- a/include/linux/serial.h
+++ b/include/linux/serial.h
@@ -10,8 +10,9 @@
 #ifndef _LINUX_SERIAL_H
 #define _LINUX_SERIAL_H
 
-#ifdef __KERNEL__
 #include <linux/types.h>
+
+#ifdef __KERNEL__
 #include <asm/page.h>
 
 /*
diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h
index 3d37c94..d4d2a78 100644
--- a/include/linux/serial_8250.h
+++ b/include/linux/serial_8250.h
@@ -28,6 +28,9 @@
 	unsigned char	iotype;		/* UPIO_* */
 	unsigned char	hub6;
 	upf_t		flags;		/* UPF_* flags */
+	unsigned int	type;		/* If UPF_FIXED_TYPE */
+	unsigned int	(*serial_in)(struct uart_port *, int);
+	void		(*serial_out)(struct uart_port *, int, int);
 };
 
 /*
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index feb3b93..b419984 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -40,7 +40,8 @@
 #define PORT_NS16550A	14
 #define PORT_XSCALE	15
 #define PORT_RM9000	16	/* PMC-Sierra RM9xxx internal UART */
-#define PORT_MAX_8250	16	/* max port ID */
+#define PORT_OCTEON	17	/* Cavium OCTEON internal UART */
+#define PORT_MAX_8250	17	/* max port ID */
 
 /*
  * ARM specific type numbers.  These are not currently guaranteed
@@ -248,6 +249,8 @@
 	spinlock_t		lock;			/* port lock */
 	unsigned long		iobase;			/* in/out[bwl] */
 	unsigned char __iomem	*membase;		/* read/write[bwl] */
+	unsigned int		(*serial_in)(struct uart_port *, int);
+	void			(*serial_out)(struct uart_port *, int, int);
 	unsigned int		irq;			/* irq number */
 	unsigned int		uartclk;		/* base uart clock */
 	unsigned int		fifosize;		/* tx fifo size */
@@ -293,6 +296,8 @@
 #define UPF_MAGIC_MULTIPLIER	((__force upf_t) (1 << 16))
 #define UPF_CONS_FLOW		((__force upf_t) (1 << 23))
 #define UPF_SHARE_IRQ		((__force upf_t) (1 << 24))
+/* The exact UART type is known and should not be probed.  */
+#define UPF_FIXED_TYPE		((__force upf_t) (1 << 27))
 #define UPF_BOOT_AUTOCONF	((__force upf_t) (1 << 28))
 #define UPF_FIXED_PORT		((__force upf_t) (1 << 29))
 #define UPF_DEAD		((__force upf_t) (1 << 30))
@@ -316,35 +321,13 @@
 };
 
 /*
- * This is the state information which is persistent across opens.
- * The low level driver must not to touch any elements contained
- * within.
- */
-struct uart_state {
-	unsigned int		close_delay;		/* msec */
-	unsigned int		closing_wait;		/* msec */
-
-#define USF_CLOSING_WAIT_INF	(0)
-#define USF_CLOSING_WAIT_NONE	(~0U)
-
-	int			count;
-	int			pm_state;
-	struct uart_info	*info;
-	struct uart_port	*port;
-
-	struct mutex		mutex;
-};
-
-#define UART_XMIT_SIZE	PAGE_SIZE
-
-typedef unsigned int __bitwise__ uif_t;
-
-/*
  * This is the state information which is only valid when the port
- * is open; it may be freed by the core driver once the device has
+ * is open; it may be cleared the core driver once the device has
  * been closed.  Either the low level driver or the core can modify
  * stuff here.
  */
+typedef unsigned int __bitwise__ uif_t;
+
 struct uart_info {
 	struct tty_port		port;
 	struct circ_buf		xmit;
@@ -366,6 +349,29 @@
 	wait_queue_head_t	delta_msr_wait;
 };
 
+/*
+ * This is the state information which is persistent across opens.
+ * The low level driver must not to touch any elements contained
+ * within.
+ */
+struct uart_state {
+	unsigned int		close_delay;		/* msec */
+	unsigned int		closing_wait;		/* msec */
+
+#define USF_CLOSING_WAIT_INF	(0)
+#define USF_CLOSING_WAIT_NONE	(~0U)
+
+	int			count;
+	int			pm_state;
+	struct uart_info	info;
+	struct uart_port	*port;
+
+	struct mutex		mutex;
+};
+
+#define UART_XMIT_SIZE	PAGE_SIZE
+
+
 /* number of characters left in xmit buffer before we ask for more */
 #define WAKEUP_CHARS		256
 
@@ -439,8 +445,13 @@
 #define uart_circ_chars_free(circ)	\
 	(CIRC_SPACE((circ)->head, (circ)->tail, UART_XMIT_SIZE))
 
-#define uart_tx_stopped(portp)		\
-	((portp)->info->port.tty->stopped || (portp)->info->port.tty->hw_stopped)
+static inline int uart_tx_stopped(struct uart_port *port)
+{
+	struct tty_struct *tty = port->info->port.tty;
+	if(tty->stopped || tty->hw_stopped)
+		return 1;
+	return 0;
+}
 
 /*
  * The following are helper functions for the low level drivers.
@@ -451,7 +462,7 @@
 #ifdef SUPPORT_SYSRQ
 	if (port->sysrq) {
 		if (ch && time_before(jiffies, port->sysrq)) {
-			handle_sysrq(ch, port->info ? port->info->port.tty : NULL);
+			handle_sysrq(ch, port->info->port.tty);
 			port->sysrq = 0;
 			return 1;
 		}
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 0c5b5ac..e632d29 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -125,7 +125,8 @@
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_BALANCE	\
 				| SD_SHARE_PKG_RESOURCES\
-				| BALANCE_FOR_MC_POWER,	\
+				| sd_balance_for_mc_power()\
+				| sd_power_saving_flags(),\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
 }
@@ -150,7 +151,8 @@
 				| SD_BALANCE_FORK	\
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_BALANCE	\
-				| BALANCE_FOR_PKG_POWER,\
+				| sd_balance_for_package_power()\
+				| sd_power_saving_flags(),\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
 }
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 3f4954c..fc39db95 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -180,8 +180,17 @@
  * until a hangup so don't use the wrong path.
  */
 
+struct tty_port;
+
+struct tty_port_operations {
+	/* Return 1 if the carrier is raised */
+	int (*carrier_raised)(struct tty_port *port);
+	void (*raise_dtr_rts)(struct tty_port *port);
+};
+	
 struct tty_port {
 	struct tty_struct	*tty;		/* Back pointer */
+	const struct tty_port_operations *ops;	/* Port operations */
 	spinlock_t		lock;		/* Lock protecting tty field */
 	int			blocked_open;	/* Waiting to open */
 	int			count;		/* Usage count */
@@ -253,6 +262,7 @@
 	unsigned int column;
 	unsigned char lnext:1, erasing:1, raw:1, real_raw:1, icanon:1;
 	unsigned char closing:1;
+	unsigned char echo_overrun:1;
 	unsigned short minimum_to_wake;
 	unsigned long overrun_time;
 	int num_overrun;
@@ -262,11 +272,16 @@
 	int read_tail;
 	int read_cnt;
 	unsigned long read_flags[N_TTY_BUF_SIZE/(8*sizeof(unsigned long))];
+	unsigned char *echo_buf;
+	unsigned int echo_pos;
+	unsigned int echo_cnt;
 	int canon_data;
 	unsigned long canon_head;
 	unsigned int canon_column;
 	struct mutex atomic_read_lock;
 	struct mutex atomic_write_lock;
+	struct mutex output_lock;
+	struct mutex echo_lock;
 	unsigned char *write_buf;
 	int write_cnt;
 	spinlock_t read_lock;
@@ -295,6 +310,7 @@
 #define TTY_PUSH 		6	/* n_tty private */
 #define TTY_CLOSING 		7	/* ->close() in progress */
 #define TTY_LDISC 		9	/* Line discipline attached */
+#define TTY_LDISC_CHANGING 	10	/* Line discipline changing */
 #define TTY_HW_COOK_OUT 	14	/* Hardware can do output cooking */
 #define TTY_HW_COOK_IN 		15	/* Hardware can do input cooking */
 #define TTY_PTY_LOCK 		16	/* pty private */
@@ -354,8 +370,7 @@
 extern void tty_driver_flush_buffer(struct tty_struct *tty);
 extern void tty_throttle(struct tty_struct *tty);
 extern void tty_unthrottle(struct tty_struct *tty);
-extern int tty_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-						struct winsize *ws);
+extern int tty_do_resize(struct tty_struct *tty, struct winsize *ws);
 extern void tty_shutdown(struct tty_struct *tty);
 extern void tty_free_termios(struct tty_struct *tty);
 extern int is_current_pgrp_orphaned(void);
@@ -421,6 +436,14 @@
 extern void tty_port_free_xmit_buf(struct tty_port *port);
 extern struct tty_struct *tty_port_tty_get(struct tty_port *port);
 extern void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty);
+extern int tty_port_carrier_raised(struct tty_port *port);
+extern void tty_port_raise_dtr_rts(struct tty_port *port);
+extern void tty_port_hangup(struct tty_port *port);
+extern int tty_port_block_til_ready(struct tty_port *port,
+				struct tty_struct *tty, struct file *filp);
+extern int tty_port_close_start(struct tty_port *port,
+				struct tty_struct *tty, struct file *filp);
+extern void tty_port_close_end(struct tty_port *port, struct tty_struct *tty);
 
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index 78416b9..08e0883 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -196,8 +196,7 @@
  *	Optional: If not provided then the write method is called under
  *	the atomic write lock to keep it serialized with the ldisc.
  *
- * int (*resize)(struct tty_struct *tty, struct tty_struct *real_tty,
- *				unsigned int rows, unsigned int cols);
+ * int (*resize)(struct tty_struct *tty, struct winsize *ws)
  *
  *	Called when a termios request is issued which changes the
  *	requested terminal geometry.
@@ -258,8 +257,7 @@
 	int (*tiocmget)(struct tty_struct *tty, struct file *file);
 	int (*tiocmset)(struct tty_struct *tty, struct file *file,
 			unsigned int set, unsigned int clear);
-	int (*resize)(struct tty_struct *tty, struct tty_struct *real_tty,
-				struct winsize *ws);
+	int (*resize)(struct tty_struct *tty, struct winsize *ws);
 	int (*set_termiox)(struct tty_struct *tty, struct termiox *tnew);
 #ifdef CONFIG_CONSOLE_POLL
 	int (*poll_init)(struct tty_driver *driver, int line, char *options);
diff --git a/include/linux/usb/wusb-wa.h b/include/linux/usb/wusb-wa.h
index a102561..fb7c359 100644
--- a/include/linux/usb/wusb-wa.h
+++ b/include/linux/usb/wusb-wa.h
@@ -51,6 +51,7 @@
 	WUSB_REQ_GET_TIME       = 25,
 	WUSB_REQ_SET_STREAM_IDX = 26,
 	WUSB_REQ_SET_WUSB_MAS   = 27,
+	WUSB_REQ_CHAN_STOP      = 28,
 };
 
 
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index f9ccbd9..c021289 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -30,6 +30,7 @@
 #include <linux/device.h>
 #include <linux/mutex.h>
 #include <linux/timer.h>
+#include <linux/wait.h>
 #include <linux/workqueue.h>
 #include <linux/uwb/spec.h>
 
@@ -66,6 +67,7 @@
 	struct uwb_dev_addr dev_addr;
 	int beacon_slot;
 	DECLARE_BITMAP(streams, UWB_NUM_STREAMS);
+	DECLARE_BITMAP(last_availability_bm, UWB_NUM_MAS);
 };
 #define to_uwb_dev(d) container_of(d, struct uwb_dev, dev)
 
@@ -86,12 +88,31 @@
 	struct mutex mutex;
 };
 
+/* Beacon cache list */
+struct uwb_beca {
+	struct list_head list;
+	size_t entries;
+	struct mutex mutex;
+};
+
+/* Event handling thread. */
+struct uwbd {
+	int pid;
+	struct task_struct *task;
+	wait_queue_head_t wq;
+	struct list_head event_list;
+	spinlock_t event_list_lock;
+};
+
 /**
  * struct uwb_mas_bm - a bitmap of all MAS in a superframe
  * @bm: a bitmap of length #UWB_NUM_MAS
  */
 struct uwb_mas_bm {
 	DECLARE_BITMAP(bm, UWB_NUM_MAS);
+	DECLARE_BITMAP(unsafe_bm, UWB_NUM_MAS);
+	int safe;
+	int unsafe;
 };
 
 /**
@@ -117,14 +138,24 @@
  * FIXME: further target states TBD.
  */
 enum uwb_rsv_state {
-	UWB_RSV_STATE_NONE,
+	UWB_RSV_STATE_NONE = 0,
 	UWB_RSV_STATE_O_INITIATED,
 	UWB_RSV_STATE_O_PENDING,
 	UWB_RSV_STATE_O_MODIFIED,
 	UWB_RSV_STATE_O_ESTABLISHED,
+	UWB_RSV_STATE_O_TO_BE_MOVED,
+	UWB_RSV_STATE_O_MOVE_EXPANDING,
+	UWB_RSV_STATE_O_MOVE_COMBINING,
+	UWB_RSV_STATE_O_MOVE_REDUCING,
 	UWB_RSV_STATE_T_ACCEPTED,
 	UWB_RSV_STATE_T_DENIED,
+	UWB_RSV_STATE_T_CONFLICT,
 	UWB_RSV_STATE_T_PENDING,
+	UWB_RSV_STATE_T_EXPANDING_ACCEPTED,
+	UWB_RSV_STATE_T_EXPANDING_CONFLICT,
+	UWB_RSV_STATE_T_EXPANDING_PENDING,
+	UWB_RSV_STATE_T_EXPANDING_DENIED,
+	UWB_RSV_STATE_T_RESIZED,
 
 	UWB_RSV_STATE_LAST,
 };
@@ -149,6 +180,12 @@
 	};
 };
 
+struct uwb_rsv_move {
+	struct uwb_mas_bm final_mas;
+	struct uwb_ie_drp *companion_drp_ie;
+	struct uwb_mas_bm companion_mas;
+};
+
 /*
  * Number of streams reserved for reservations targeted at DevAddrs.
  */
@@ -186,6 +223,7 @@
  *
  * @status:         negotiation status
  * @stream:         stream index allocated for this reservation
+ * @tiebreaker:     conflict tiebreaker for this reservation
  * @mas:            reserved MAS
  * @drp_ie:         the DRP IE
  * @ie_valid:       true iff the DRP IE matches the reservation parameters
@@ -201,25 +239,29 @@
 	struct uwb_rc *rc;
 	struct list_head rc_node;
 	struct list_head pal_node;
+	struct kref kref;
 
 	struct uwb_dev *owner;
 	struct uwb_rsv_target target;
 	enum uwb_drp_type type;
 	int max_mas;
 	int min_mas;
-	int sparsity;
+	int max_interval;
 	bool is_multicast;
 
 	uwb_rsv_cb_f callback;
 	void *pal_priv;
 
 	enum uwb_rsv_state state;
+	bool needs_release_companion_mas;
 	u8 stream;
+	u8 tiebreaker;
 	struct uwb_mas_bm mas;
 	struct uwb_ie_drp *drp_ie;
+	struct uwb_rsv_move mv;
 	bool ie_valid;
 	struct timer_list timer;
-	bool expired;
+	struct work_struct handle_timeout_work;
 };
 
 static const
@@ -261,6 +303,13 @@
 	bool ie_valid;
 };
 
+struct uwb_drp_backoff_win {
+	u8 window;
+	u8 n;
+	int total_expired;
+	struct timer_list timer;
+	bool can_reserve_extra_mases;
+};
 
 const char *uwb_rsv_state_str(enum uwb_rsv_state state);
 const char *uwb_rsv_type_str(enum uwb_drp_type type);
@@ -276,6 +325,8 @@
 
 void uwb_rsv_accept(struct uwb_rsv *rsv, uwb_rsv_cb_f cb, void *pal_priv);
 
+void uwb_rsv_get_usable_mas(struct uwb_rsv *orig_rsv, struct uwb_mas_bm *mas);
+
 /**
  * Radio Control Interface instance
  *
@@ -337,23 +388,33 @@
 	u8 ctx_roll;
 
 	int beaconing;			/* Beaconing state [channel number] */
+	int beaconing_forced;
 	int scanning;
 	enum uwb_scan_type scan_type:3;
 	unsigned ready:1;
 	struct uwb_notifs_chain notifs_chain;
+	struct uwb_beca uwb_beca;
 
+	struct uwbd uwbd;
+
+	struct uwb_drp_backoff_win bow;
 	struct uwb_drp_avail drp_avail;
 	struct list_head reservations;
+	struct list_head cnflt_alien_list;
+	struct uwb_mas_bm cnflt_alien_bitmap;
 	struct mutex rsvs_mutex;
+	spinlock_t rsvs_lock;
 	struct workqueue_struct *rsv_workq;
-	struct work_struct rsv_update_work;
 
+	struct delayed_work rsv_update_work;
+	struct delayed_work rsv_alien_bp_work;
+	int set_drp_ie_pending;
 	struct mutex ies_mutex;
 	struct uwb_rc_cmd_set_ie *ies;
 	size_t ies_capacity;
 
-	spinlock_t pal_lock;
 	struct list_head pals;
+	int active_pals;
 
 	struct uwb_dbg *dbg;
 };
@@ -361,11 +422,19 @@
 
 /**
  * struct uwb_pal - a UWB PAL
- * @name:    descriptive name for this PAL (wushc, wlp, etc.).
+ * @name:    descriptive name for this PAL (wusbhc, wlp, etc.).
  * @device:  a device for the PAL.  Used to link the PAL and the radio
  *           controller in sysfs.
+ * @rc:      the radio controller the PAL uses.
+ * @channel_changed: called when the channel used by the radio changes.
+ *           A channel of -1 means the channel has been stopped.
  * @new_rsv: called when a peer requests a reservation (may be NULL if
  *           the PAL cannot accept reservation requests).
+ * @channel: channel being used by the PAL; 0 if the PAL isn't using
+ *           the radio; -1 if the PAL wishes to use the radio but
+ *           cannot.
+ * @debugfs_dir: a debugfs directory which the PAL can use for its own
+ *           debugfs files.
  *
  * A Protocol Adaptation Layer (PAL) is a user of the WiMedia UWB
  * radio platform (e.g., WUSB, WLP or Bluetooth UWB AMP).
@@ -384,12 +453,21 @@
 	struct list_head node;
 	const char *name;
 	struct device *device;
-	void (*new_rsv)(struct uwb_rsv *rsv);
+	struct uwb_rc *rc;
+
+	void (*channel_changed)(struct uwb_pal *pal, int channel);
+	void (*new_rsv)(struct uwb_pal *pal, struct uwb_rsv *rsv);
+
+	int channel;
+	struct dentry *debugfs_dir;
 };
 
 void uwb_pal_init(struct uwb_pal *pal);
-int uwb_pal_register(struct uwb_rc *rc, struct uwb_pal *pal);
-void uwb_pal_unregister(struct uwb_rc *rc, struct uwb_pal *pal);
+int uwb_pal_register(struct uwb_pal *pal);
+void uwb_pal_unregister(struct uwb_pal *pal);
+
+int uwb_radio_start(struct uwb_pal *pal);
+void uwb_radio_stop(struct uwb_pal *pal);
 
 /*
  * General public API
@@ -443,8 +521,6 @@
 		    struct uwb_rccb *cmd, size_t cmd_size,
 		    u8 expected_type, u16 expected_event,
 		    struct uwb_rceb **preply);
-ssize_t uwb_rc_get_ie(struct uwb_rc *, struct uwb_rc_evt_get_ie **);
-int uwb_bg_joined(struct uwb_rc *rc);
 
 size_t __uwb_addr_print(char *, size_t, const unsigned char *, int);
 
@@ -520,6 +596,8 @@
 void uwb_rc_neh_grok(struct uwb_rc *, void *, size_t);
 void uwb_rc_neh_error(struct uwb_rc *, int);
 void uwb_rc_reset_all(struct uwb_rc *rc);
+void uwb_rc_pre_reset(struct uwb_rc *rc);
+void uwb_rc_post_reset(struct uwb_rc *rc);
 
 /**
  * uwb_rsv_is_owner - is the owner of this reservation the RC?
@@ -531,7 +609,9 @@
 }
 
 /**
- * Events generated by UWB that can be passed to any listeners
+ * enum uwb_notifs - UWB events that can be passed to any listeners
+ * @UWB_NOTIF_ONAIR: a new neighbour has joined the beacon group.
+ * @UWB_NOTIF_OFFAIR: a neighbour has left the beacon group.
  *
  * Higher layers can register callback functions with the radio
  * controller using uwb_notifs_register(). The radio controller
@@ -539,8 +619,6 @@
  * nodes when an event occurs.
  */
 enum uwb_notifs {
-	UWB_NOTIF_BG_JOIN = 0,	/* radio controller joined a beacon group */
-	UWB_NOTIF_BG_LEAVE = 1,	/* radio controller left a beacon group */
 	UWB_NOTIF_ONAIR,
 	UWB_NOTIF_OFFAIR,
 };
@@ -652,22 +730,9 @@
 
 /* Information Element handling */
 
-/* For representing the state of writing to a buffer when iterating */
-struct uwb_buf_ctx {
-	char *buf;
-	size_t bytes, size;
-};
-
-typedef int (*uwb_ie_f)(struct uwb_dev *, const struct uwb_ie_hdr *,
-			size_t, void *);
 struct uwb_ie_hdr *uwb_ie_next(void **ptr, size_t *len);
-ssize_t uwb_ie_for_each(struct uwb_dev *uwb_dev, uwb_ie_f fn, void *data,
-			const void *buf, size_t size);
-int uwb_ie_dump_hex(struct uwb_dev *, const struct uwb_ie_hdr *,
-		    size_t, void *);
-int uwb_rc_set_ie(struct uwb_rc *, struct uwb_rc_cmd_set_ie *);
-struct uwb_ie_hdr *uwb_ie_next(void **ptr, size_t *len);
-
+int uwb_rc_ie_add(struct uwb_rc *uwb_rc, const struct uwb_ie_hdr *ies, size_t size);
+int uwb_rc_ie_rm(struct uwb_rc *uwb_rc, enum uwb_ie element_id);
 
 /*
  * Transmission statistics
diff --git a/include/linux/uwb/debug-cmd.h b/include/linux/uwb/debug-cmd.h
index 1141f41..8da004e 100644
--- a/include/linux/uwb/debug-cmd.h
+++ b/include/linux/uwb/debug-cmd.h
@@ -32,6 +32,10 @@
 enum uwb_dbg_cmd_type {
 	UWB_DBG_CMD_RSV_ESTABLISH = 1,
 	UWB_DBG_CMD_RSV_TERMINATE = 2,
+	UWB_DBG_CMD_IE_ADD = 3,
+	UWB_DBG_CMD_IE_RM = 4,
+	UWB_DBG_CMD_RADIO_START = 5,
+	UWB_DBG_CMD_RADIO_STOP = 6,
 };
 
 struct uwb_dbg_cmd_rsv_establish {
@@ -39,18 +43,25 @@
 	__u8  type;
 	__u16 max_mas;
 	__u16 min_mas;
-	__u8  sparsity;
+	__u8  max_interval;
 };
 
 struct uwb_dbg_cmd_rsv_terminate {
 	int index;
 };
 
+struct uwb_dbg_cmd_ie {
+	__u8 data[128];
+	int len;
+};
+
 struct uwb_dbg_cmd {
 	__u32 type;
 	union {
 		struct uwb_dbg_cmd_rsv_establish rsv_establish;
 		struct uwb_dbg_cmd_rsv_terminate rsv_terminate;
+		struct uwb_dbg_cmd_ie ie_add;
+		struct uwb_dbg_cmd_ie ie_rm;
 	};
 };
 
diff --git a/include/linux/uwb/debug.h b/include/linux/uwb/debug.h
deleted file mode 100644
index a86a73f..0000000
--- a/include/linux/uwb/debug.h
+++ /dev/null
@@ -1,82 +0,0 @@
-/*
- * Ultra Wide Band
- * Debug Support
- *
- * Copyright (C) 2005-2006 Intel Corporation
- * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version
- * 2 as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
- * 02110-1301, USA.
- *
- *
- * FIXME: doc
- * Invoke like:
- *
- * #define D_LOCAL 4
- * #include <linux/uwb/debug.h>
- *
- * At the end of your include files.
- */
-#include <linux/types.h>
-
-struct device;
-extern void dump_bytes(struct device *dev, const void *_buf, size_t rsize);
-
-/* Master debug switch; !0 enables, 0 disables */
-#define D_MASTER (!0)
-
-/* Local (per-file) debug switch; #define before #including */
-#ifndef D_LOCAL
-#define D_LOCAL 0
-#endif
-
-#undef __d_printf
-#undef d_fnstart
-#undef d_fnend
-#undef d_printf
-#undef d_dump
-
-#define __d_printf(l, _tag, _dev, f, a...)				\
-do {									\
-	struct device *__dev = (_dev);					\
-	if (D_MASTER && D_LOCAL >= (l)) {				\
-		char __head[64] = "";					\
-		if (_dev != NULL) {					\
-			if ((unsigned long)__dev < 4096)		\
-				printk(KERN_ERR "E: Corrupt dev %p\n",	\
-					__dev);				\
-			else						\
-				snprintf(__head, sizeof(__head),	\
-					 "%s %s: ",			\
-					 dev_driver_string(__dev),	\
-					 __dev->bus_id);		\
-		}							\
-		printk(KERN_ERR "%s%s" _tag ": " f, __head,		\
-			__func__, ## a);				\
-	}								\
-} while (0 && _dev)
-
-#define d_fnstart(l, _dev, f, a...)	\
-	__d_printf(l, " FNSTART", _dev, f, ## a)
-#define d_fnend(l, _dev, f, a...)	\
-	__d_printf(l, " FNEND", _dev, f, ## a)
-#define d_printf(l, _dev, f, a...)	\
-	__d_printf(l, "", _dev, f, ## a)
-#define d_dump(l, _dev, ptr, size)		\
-do {						\
-	struct device *__dev = _dev;		\
-	if (D_MASTER && D_LOCAL >= (l))		\
-		dump_bytes(__dev, ptr, size);	\
-} while (0 && _dev)
-#define d_test(l) (D_MASTER && D_LOCAL >= (l))
diff --git a/include/linux/uwb/spec.h b/include/linux/uwb/spec.h
index 198c15f..b52e44f 100644
--- a/include/linux/uwb/spec.h
+++ b/include/linux/uwb/spec.h
@@ -59,6 +59,11 @@
 #define UWB_MAS_PER_ZONE (UWB_NUM_MAS / UWB_NUM_ZONES)
 
 /*
+ * Number of MAS required before a row can be considered available.
+ */
+#define UWB_USABLE_MAS_PER_ROW (UWB_NUM_ZONES - 1)
+
+/*
  * Number of streams per DRP reservation between a pair of devices.
  *
  * [ECMA-368] section 16.8.6.
@@ -94,6 +99,26 @@
 enum { UWB_MAX_LOST_BEACONS = 3 };
 
 /*
+ * mDRPBackOffWinMin
+ *
+ * The minimum number of superframes to wait before trying to reserve
+ * extra MAS.
+ *
+ * [ECMA-368] section 17.16
+ */
+enum { UWB_DRP_BACKOFF_WIN_MIN = 2 };
+
+/*
+ * mDRPBackOffWinMax
+ *
+ * The maximum number of superframes to wait before trying to reserve
+ * extra MAS.
+ *
+ * [ECMA-368] section 17.16
+ */
+enum { UWB_DRP_BACKOFF_WIN_MAX = 16 };
+
+/*
  * Length of a superframe in microseconds.
  */
 #define UWB_SUPERFRAME_LENGTH_US (UWB_MAS_LENGTH_US * UWB_NUM_MAS)
@@ -200,6 +225,12 @@
 	UWB_DRP_REASON_MODIFIED,
 };
 
+/** Relinquish Request Reason Codes ([ECMA-368] table 113) */
+enum uwb_relinquish_req_reason {
+	UWB_RELINQUISH_REQ_REASON_NON_SPECIFIC = 0,
+	UWB_RELINQUISH_REQ_REASON_OVER_ALLOCATION,
+};
+
 /**
  *  DRP Notification Reason Codes (WHCI 0.95 [3.1.4.9])
  */
@@ -252,6 +283,7 @@
 	UWB_APP_SPEC_PROBE_IE = 15,
 	UWB_IDENTIFICATION_IE = 19,
 	UWB_MASTER_KEY_ID_IE = 20,
+	UWB_RELINQUISH_REQUEST_IE = 21,
 	UWB_IE_WLP = 250, /* WiMedia Logical Link Control Protocol WLP 0.99 */
 	UWB_APP_SPEC_IE = 255,
 };
@@ -365,6 +397,27 @@
 	DECLARE_BITMAP(bmp, UWB_NUM_MAS);
 } __attribute__((packed));
 
+/* Relinqish Request IE ([ECMA-368] section 16.8.19). */
+struct uwb_relinquish_request_ie {
+        struct uwb_ie_hdr       hdr;
+        __le16                  relinquish_req_control;
+        struct uwb_dev_addr     dev_addr;
+        struct uwb_drp_alloc    allocs[];
+} __attribute__((packed));
+
+static inline int uwb_ie_relinquish_req_reason_code(struct uwb_relinquish_request_ie *ie)
+{
+	return (le16_to_cpu(ie->relinquish_req_control) >> 0) & 0xf;
+}
+
+static inline void uwb_ie_relinquish_req_set_reason_code(struct uwb_relinquish_request_ie *ie,
+							 int reason_code)
+{
+	u16 ctrl = le16_to_cpu(ie->relinquish_req_control);
+	ctrl = (ctrl & ~(0xf << 0)) | (reason_code << 0);
+	ie->relinquish_req_control = cpu_to_le16(ctrl);
+}
+
 /**
  * The Vendor ID is set to an OUI that indicates the vendor of the device.
  * ECMA-368 [16.8.10]
diff --git a/include/linux/uwb/umc.h b/include/linux/uwb/umc.h
index 36a39e3..4b4fc0f 100644
--- a/include/linux/uwb/umc.h
+++ b/include/linux/uwb/umc.h
@@ -89,6 +89,8 @@
 	void (*remove)(struct umc_dev *);
 	int  (*suspend)(struct umc_dev *, pm_message_t state);
 	int  (*resume)(struct umc_dev *);
+	int  (*pre_reset)(struct umc_dev *);
+	int  (*post_reset)(struct umc_dev *);
 
 	struct device_driver driver;
 };
diff --git a/include/linux/wlp.h b/include/linux/wlp.h
index 033545e..ac95ce6 100644
--- a/include/linux/wlp.h
+++ b/include/linux/wlp.h
@@ -646,6 +646,7 @@
 struct wlp {
 	struct mutex mutex;
 	struct uwb_rc *rc;		/* UWB radio controller */
+	struct net_device *ndev;
 	struct uwb_pal pal;
 	struct wlp_eda eda;
 	struct wlp_uuid uuid;
@@ -675,7 +676,7 @@
 static struct wlp_wss_attribute wss_attr_##_name = __ATTR(_name, _mode,	\
 							  _show, _store)
 
-extern int wlp_setup(struct wlp *, struct uwb_rc *);
+extern int wlp_setup(struct wlp *, struct uwb_rc *, struct net_device *ndev);
 extern void wlp_remove(struct wlp *);
 extern ssize_t wlp_neighborhood_show(struct wlp *, char *);
 extern int wlp_wss_setup(struct net_device *, struct wlp_wss *);
diff --git a/init/Kconfig b/init/Kconfig
index 1362719..f628171 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -924,6 +924,15 @@
 
 endif # MODULES
 
+config INIT_ALL_POSSIBLE
+	bool
+	help
+	  Back when each arch used to define their own cpu_online_map and
+	  cpu_possible_map, some of them chose to initialize cpu_possible_map
+	  with all 1s, and others with all 0s.  When they were centralised,
+	  it was better to provide this option than to break all the archs
+	  and have several arch maintainers persuing me down dark alleys.
+
 config STOP_MACHINE
 	bool
 	default y
diff --git a/init/main.c b/init/main.c
index 2a7ce0f..ad8f9f5 100644
--- a/init/main.c
+++ b/init/main.c
@@ -75,15 +75,6 @@
 #include <asm/smp.h>
 #endif
 
-/*
- * This is one of the first .c files built. Error out early if we have compiler
- * trouble.
- */
-
-#if __GNUC__ == 4 && __GNUC_MINOR__ == 1 && __GNUC_PATCHLEVEL__ == 0
-#warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
-#endif
-
 static int kernel_init(void *);
 
 extern void init_IRQ(void);
@@ -540,15 +531,6 @@
 {
 }
 
-void __init __weak arch_early_irq_init(void)
-{
-}
-
-void __init __weak early_irq_init(void)
-{
-	arch_early_irq_init();
-}
-
 asmlinkage void __init start_kernel(void)
 {
 	char * command_line;
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 8ea32e8..bae131a 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -24,19 +24,20 @@
 cpumask_t cpu_present_map __read_mostly;
 EXPORT_SYMBOL(cpu_present_map);
 
-#ifndef CONFIG_SMP
-
 /*
  * Represents all cpu's that are currently online.
  */
-cpumask_t cpu_online_map __read_mostly = CPU_MASK_ALL;
+cpumask_t cpu_online_map __read_mostly;
 EXPORT_SYMBOL(cpu_online_map);
 
+#ifdef CONFIG_INIT_ALL_POSSIBLE
 cpumask_t cpu_possible_map __read_mostly = CPU_MASK_ALL;
+#else
+cpumask_t cpu_possible_map __read_mostly;
+#endif
 EXPORT_SYMBOL(cpu_possible_map);
 
-#else /* CONFIG_SMP */
-
+#ifdef CONFIG_SMP
 /* Serializes the updates to cpu_online_map, cpu_present_map */
 static DEFINE_MUTEX(cpu_add_remove_lock);
 
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 96c0ba1..39c1a4c 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -896,7 +896,7 @@
 	if (!*buf) {
 		cpus_clear(trialcs.cpus_allowed);
 	} else {
-		retval = cpulist_parse(buf, trialcs.cpus_allowed);
+		retval = cpulist_parse(buf, &trialcs.cpus_allowed);
 		if (retval < 0)
 			return retval;
 
@@ -1482,7 +1482,7 @@
 	mask = cs->cpus_allowed;
 	mutex_unlock(&callback_mutex);
 
-	return cpulist_scnprintf(page, PAGE_SIZE, mask);
+	return cpulist_scnprintf(page, PAGE_SIZE, &mask);
 }
 
 static int cpuset_sprintf_memlist(char *page, struct cpuset *cs)
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index bda9cb9..eb2bfef 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -32,7 +32,6 @@
  */
 
 #include <linux/cpu.h>
-#include <linux/irq.h>
 #include <linux/module.h>
 #include <linux/percpu.h>
 #include <linux/hrtimer.h>
diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c
index 650ce41..cc0f732 100644
--- a/kernel/irq/autoprobe.c
+++ b/kernel/irq/autoprobe.c
@@ -40,9 +40,6 @@
 	 * flush such a longstanding irq before considering it as spurious.
 	 */
 	for_each_irq_desc_reverse(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		if (!desc->action && !(desc->status & IRQ_NOPROBE)) {
 			/*
@@ -71,9 +68,6 @@
 	 * happened in the previous stage, it may have masked itself)
 	 */
 	for_each_irq_desc_reverse(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		if (!desc->action && !(desc->status & IRQ_NOPROBE)) {
 			desc->status |= IRQ_AUTODETECT | IRQ_WAITING;
@@ -92,9 +86,6 @@
 	 * Now filter out any obviously spurious interrupts
 	 */
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		status = desc->status;
 
@@ -133,9 +124,6 @@
 	int i;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		status = desc->status;
 
@@ -178,9 +166,6 @@
 	unsigned int status;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		status = desc->status;
 
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 6eb3c79..f63c706 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -46,7 +46,7 @@
 	desc->irq_count = 0;
 	desc->irqs_unhandled = 0;
 #ifdef CONFIG_SMP
-	cpus_setall(desc->affinity);
+	cpumask_setall(&desc->affinity);
 #endif
 	spin_unlock_irqrestore(&desc->lock, flags);
 }
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 6492400..c20db0b 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -56,10 +56,6 @@
 int nr_irqs = NR_IRQS;
 EXPORT_SYMBOL_GPL(nr_irqs);
 
-void __init __attribute__((weak)) arch_early_irq_init(void)
-{
-}
-
 #ifdef CONFIG_SPARSE_IRQ
 static struct irq_desc irq_desc_init = {
 	.irq	    = -1,
@@ -90,13 +86,11 @@
 		desc->kstat_irqs = (unsigned int *)ptr;
 }
 
-void __attribute__((weak)) arch_init_chip_data(struct irq_desc *desc, int cpu)
-{
-}
-
 static void init_one_irq_desc(int irq, struct irq_desc *desc, int cpu)
 {
 	memcpy(desc, &irq_desc_init, sizeof(struct irq_desc));
+
+	spin_lock_init(&desc->lock);
 	desc->irq = irq;
 #ifdef CONFIG_SMP
 	desc->cpu = cpu;
@@ -134,7 +128,7 @@
 /* FIXME: use bootmem alloc ...*/
 static unsigned int kstat_irqs_legacy[NR_IRQS_LEGACY][NR_CPUS];
 
-void __init early_irq_init(void)
+int __init early_irq_init(void)
 {
 	struct irq_desc *desc;
 	int legacy_count;
@@ -146,6 +140,7 @@
 	for (i = 0; i < legacy_count; i++) {
 		desc[i].irq = i;
 		desc[i].kstat_irqs = kstat_irqs_legacy[i];
+		lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
 
 		irq_desc_ptrs[i] = desc + i;
 	}
@@ -153,7 +148,7 @@
 	for (i = legacy_count; i < NR_IRQS; i++)
 		irq_desc_ptrs[i] = NULL;
 
-	arch_early_irq_init();
+	return arch_early_irq_init();
 }
 
 struct irq_desc *irq_to_desc(unsigned int irq)
@@ -203,7 +198,7 @@
 	return desc;
 }
 
-#else
+#else /* !CONFIG_SPARSE_IRQ */
 
 struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
 	[0 ... NR_IRQS-1] = {
@@ -218,7 +213,31 @@
 	}
 };
 
-#endif
+int __init early_irq_init(void)
+{
+	struct irq_desc *desc;
+	int count;
+	int i;
+
+	desc = irq_desc;
+	count = ARRAY_SIZE(irq_desc);
+
+	for (i = 0; i < count; i++)
+		desc[i].irq = i;
+
+	return arch_early_irq_init();
+}
+
+struct irq_desc *irq_to_desc(unsigned int irq)
+{
+	return (irq < NR_IRQS) ? irq_desc + irq : NULL;
+}
+
+struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu)
+{
+	return irq_to_desc(irq);
+}
+#endif /* !CONFIG_SPARSE_IRQ */
 
 /*
  * What should we do if we get a hw irq event on an illegal vector?
@@ -428,9 +447,6 @@
 	int i;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		lockdep_set_class(&desc->lock, &irq_desc_lock_class);
 	}
 }
@@ -439,7 +455,7 @@
 unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-	return desc->kstat_irqs[cpu];
+	return desc ? desc->kstat_irqs[cpu] : 0;
 }
 #endif
 EXPORT_SYMBOL(kstat_irqs_cpu);
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 540f6c4..61c4a9b 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -79,7 +79,7 @@
  *	@cpumask:	cpumask
  *
  */
-int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
+int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	unsigned long flags;
@@ -91,14 +91,14 @@
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) {
-		desc->affinity = cpumask;
+		cpumask_copy(&desc->affinity, cpumask);
 		desc->chip->set_affinity(irq, cpumask);
 	} else {
 		desc->status |= IRQ_MOVE_PENDING;
-		desc->pending_mask = cpumask;
+		cpumask_copy(&desc->pending_mask, cpumask);
 	}
 #else
-	desc->affinity = cpumask;
+	cpumask_copy(&desc->affinity, cpumask);
 	desc->chip->set_affinity(irq, cpumask);
 #endif
 	desc->status |= IRQ_AFFINITY_SET;
@@ -112,26 +112,24 @@
  */
 int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc)
 {
-	cpumask_t mask;
-
 	if (!irq_can_set_affinity(irq))
 		return 0;
 
-	cpus_and(mask, cpu_online_map, irq_default_affinity);
-
 	/*
 	 * Preserve an userspace affinity setup, but make sure that
 	 * one of the targets is online.
 	 */
 	if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) {
-		if (cpus_intersects(desc->affinity, cpu_online_map))
-			mask = desc->affinity;
+		if (cpumask_any_and(&desc->affinity, cpu_online_mask)
+		    < nr_cpu_ids)
+			goto set_affinity;
 		else
 			desc->status &= ~IRQ_AFFINITY_SET;
 	}
 
-	desc->affinity = mask;
-	desc->chip->set_affinity(irq, mask);
+	cpumask_and(&desc->affinity, cpu_online_mask, &irq_default_affinity);
+set_affinity:
+	desc->chip->set_affinity(irq, &desc->affinity);
 
 	return 0;
 }
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 9db681d..bd72329 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -4,7 +4,6 @@
 void move_masked_irq(int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-	cpumask_t tmp;
 
 	if (likely(!(desc->status & IRQ_MOVE_PENDING)))
 		return;
@@ -19,7 +18,7 @@
 
 	desc->status &= ~IRQ_MOVE_PENDING;
 
-	if (unlikely(cpus_empty(desc->pending_mask)))
+	if (unlikely(cpumask_empty(&desc->pending_mask)))
 		return;
 
 	if (!desc->chip->set_affinity)
@@ -27,8 +26,6 @@
 
 	assert_spin_locked(&desc->lock);
 
-	cpus_and(tmp, desc->pending_mask, cpu_online_map);
-
 	/*
 	 * If there was a valid mask to work with, please
 	 * do the disable, re-program, enable sequence.
@@ -41,10 +38,13 @@
 	 * For correct operation this depends on the caller
 	 * masking the irqs.
 	 */
-	if (likely(!cpus_empty(tmp))) {
-		desc->chip->set_affinity(irq,tmp);
+	if (likely(cpumask_any_and(&desc->pending_mask, cpu_online_mask)
+		   < nr_cpu_ids)) {
+		cpumask_and(&desc->affinity,
+			    &desc->pending_mask, cpu_online_mask);
+		desc->chip->set_affinity(irq, &desc->affinity);
 	}
-	cpus_clear(desc->pending_mask);
+	cpumask_clear(&desc->pending_mask);
 }
 
 void move_native_irq(int irq)
diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c
index 089c3746..ecf765c 100644
--- a/kernel/irq/numa_migrate.c
+++ b/kernel/irq/numa_migrate.c
@@ -42,6 +42,7 @@
 		 struct irq_desc *desc, int cpu)
 {
 	memcpy(desc, old_desc, sizeof(struct irq_desc));
+	spin_lock_init(&desc->lock);
 	desc->cpu = cpu;
 	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
 	init_copy_kstat_irqs(old_desc, desc, cpu, nr_cpu_ids);
@@ -74,10 +75,8 @@
 
 	node = cpu_to_node(cpu);
 	desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node);
-	printk(KERN_DEBUG "  move irq_desc for %d to cpu %d node %d\n",
-		 irq, cpu, node);
 	if (!desc) {
-		printk(KERN_ERR "can not get new irq_desc for moving\n");
+		printk(KERN_ERR "irq %d: can not get new irq_desc for migration.\n", irq);
 		/* still use old one */
 		desc = old_desc;
 		goto out_unlock;
@@ -106,8 +105,6 @@
 		return desc;
 
 	old_cpu = desc->cpu;
-	printk(KERN_DEBUG
-		 "try to move irq_desc from cpu %d to %d\n", old_cpu, cpu);
 	if (old_cpu != cpu) {
 		node = cpu_to_node(cpu);
 		old_node = cpu_to_node(old_cpu);
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index f6b3440..d2c0e5e 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -40,33 +40,42 @@
 		const char __user *buffer, size_t count, loff_t *pos)
 {
 	unsigned int irq = (int)(long)PDE(file->f_path.dentry->d_inode)->data;
-	cpumask_t new_value;
+	cpumask_var_t new_value;
 	int err;
 
 	if (!irq_to_desc(irq)->chip->set_affinity || no_irq_affinity ||
 	    irq_balancing_disabled(irq))
 		return -EIO;
 
+	if (!alloc_cpumask_var(&new_value, GFP_KERNEL))
+		return -ENOMEM;
+
 	err = cpumask_parse_user(buffer, count, new_value);
 	if (err)
-		return err;
+		goto free_cpumask;
 
-	if (!is_affinity_mask_valid(new_value))
-		return -EINVAL;
+	if (!is_affinity_mask_valid(*new_value)) {
+		err = -EINVAL;
+		goto free_cpumask;
+	}
 
 	/*
 	 * Do not allow disabling IRQs completely - it's a too easy
 	 * way to make the system unusable accidentally :-) At least
 	 * one online CPU still has to be targeted.
 	 */
-	if (!cpus_intersects(new_value, cpu_online_map))
+	if (!cpumask_intersects(new_value, cpu_online_mask)) {
 		/* Special case for empty set - allow the architecture
 		   code to set default SMP affinity. */
-		return irq_select_affinity_usr(irq) ? -EINVAL : count;
+		err = irq_select_affinity_usr(irq) ? -EINVAL : count;
+	} else {
+		irq_set_affinity(irq, new_value);
+		err = count;
+	}
 
-	irq_set_affinity(irq, new_value);
-
-	return count;
+free_cpumask:
+	free_cpumask_var(new_value);
+	return err;
 }
 
 static int irq_affinity_proc_open(struct inode *inode, struct file *file)
@@ -95,7 +104,7 @@
 	cpumask_t new_value;
 	int err;
 
-	err = cpumask_parse_user(buffer, count, new_value);
+	err = cpumask_parse_user(buffer, count, &new_value);
 	if (err)
 		return err;
 
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index 3738107..dd364c1 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -91,9 +91,6 @@
 	int i, ok = 0;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		if (!i)
 			 continue;
 
@@ -115,8 +112,6 @@
 	for_each_irq_desc(i, desc) {
 		unsigned int status;
 
-		if (!desc)
-			continue;
 		if (!i)
 			 continue;
 
diff --git a/kernel/profile.c b/kernel/profile.c
index 60adefb..4cb7d68 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -442,7 +442,7 @@
 static int prof_cpu_mask_read_proc(char *page, char **start, off_t off,
 			int count, int *eof, void *data)
 {
-	int len = cpumask_scnprintf(page, count, *(cpumask_t *)data);
+	int len = cpumask_scnprintf(page, count, (cpumask_t *)data);
 	if (count - len < 2)
 		return -EINVAL;
 	len += sprintf(page + len, "\n");
@@ -456,7 +456,7 @@
 	unsigned long full_count = count, err;
 	cpumask_t new_value;
 
-	err = cpumask_parse_user(buffer, count, new_value);
+	err = cpumask_parse_user(buffer, count, &new_value);
 	if (err)
 		return err;
 
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index e503a00..c03ca3e 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -393,7 +393,7 @@
 		 * unnecessarily.
 		 */
 		smp_mb();
-		cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask);
+		cpumask_andnot(&rcp->cpumask, cpu_online_mask, nohz_cpu_mask);
 
 		rcp->signaled = 0;
 	}
diff --git a/kernel/sched.c b/kernel/sched.c
index fff1c4a..930bf2e 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -498,18 +498,26 @@
  */
 struct root_domain {
 	atomic_t refcount;
-	cpumask_t span;
-	cpumask_t online;
+	cpumask_var_t span;
+	cpumask_var_t online;
 
 	/*
 	 * The "RT overload" flag: it gets set if a CPU has more than
 	 * one runnable RT task.
 	 */
-	cpumask_t rto_mask;
+	cpumask_var_t rto_mask;
 	atomic_t rto_count;
 #ifdef CONFIG_SMP
 	struct cpupri cpupri;
 #endif
+#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
+	/*
+	 * Preferred wake up cpu nominated by sched_mc balance that will be
+	 * used when most cpus are idle in the system indicating overall very
+	 * low system utilisation. Triggered at POWERSAVINGS_BALANCE_WAKEUP(2)
+	 */
+	unsigned int sched_mc_preferred_wakeup_cpu;
+#endif
 };
 
 /*
@@ -1514,7 +1522,7 @@
 	struct sched_domain *sd = data;
 	int i;
 
-	for_each_cpu_mask(i, sd->span) {
+	for_each_cpu(i, sched_domain_span(sd)) {
 		/*
 		 * If there are currently no tasks on the cpu pretend there
 		 * is one of average load so that when a new task gets to
@@ -1535,7 +1543,7 @@
 	if (!sd->parent || !(sd->parent->flags & SD_LOAD_BALANCE))
 		shares = tg->shares;
 
-	for_each_cpu_mask(i, sd->span)
+	for_each_cpu(i, sched_domain_span(sd))
 		update_group_shares_cpu(tg, i, shares, rq_weight);
 
 	return 0;
@@ -2101,15 +2109,17 @@
 		int i;
 
 		/* Skip over this group if it has no CPUs allowed */
-		if (!cpus_intersects(group->cpumask, p->cpus_allowed))
+		if (!cpumask_intersects(sched_group_cpus(group),
+					&p->cpus_allowed))
 			continue;
 
-		local_group = cpu_isset(this_cpu, group->cpumask);
+		local_group = cpumask_test_cpu(this_cpu,
+					       sched_group_cpus(group));
 
 		/* Tally up the load of all CPUs in the group */
 		avg_load = 0;
 
-		for_each_cpu_mask_nr(i, group->cpumask) {
+		for_each_cpu(i, sched_group_cpus(group)) {
 			/* Bias balancing toward cpus of our domain */
 			if (local_group)
 				load = source_load(i, load_idx);
@@ -2141,17 +2151,14 @@
  * find_idlest_cpu - find the idlest cpu among the cpus in group.
  */
 static int
-find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu,
-		cpumask_t *tmp)
+find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 {
 	unsigned long load, min_load = ULONG_MAX;
 	int idlest = -1;
 	int i;
 
 	/* Traverse only the allowed CPUs */
-	cpus_and(*tmp, group->cpumask, p->cpus_allowed);
-
-	for_each_cpu_mask_nr(i, *tmp) {
+	for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) {
 		load = weighted_cpuload(i);
 
 		if (load < min_load || (load == min_load && i == this_cpu)) {
@@ -2193,7 +2200,6 @@
 		update_shares(sd);
 
 	while (sd) {
-		cpumask_t span, tmpmask;
 		struct sched_group *group;
 		int new_cpu, weight;
 
@@ -2202,14 +2208,13 @@
 			continue;
 		}
 
-		span = sd->span;
 		group = find_idlest_group(sd, t, cpu);
 		if (!group) {
 			sd = sd->child;
 			continue;
 		}
 
-		new_cpu = find_idlest_cpu(group, t, cpu, &tmpmask);
+		new_cpu = find_idlest_cpu(group, t, cpu);
 		if (new_cpu == -1 || new_cpu == cpu) {
 			/* Now try balancing at a lower domain level of cpu */
 			sd = sd->child;
@@ -2218,10 +2223,10 @@
 
 		/* Now try balancing at a lower domain level of new_cpu */
 		cpu = new_cpu;
+		weight = cpumask_weight(sched_domain_span(sd));
 		sd = NULL;
-		weight = cpus_weight(span);
 		for_each_domain(cpu, tmp) {
-			if (weight <= cpus_weight(tmp->span))
+			if (weight <= cpumask_weight(sched_domain_span(tmp)))
 				break;
 			if (tmp->flags & flag)
 				sd = tmp;
@@ -2266,7 +2271,7 @@
 		cpu = task_cpu(p);
 
 		for_each_domain(this_cpu, sd) {
-			if (cpu_isset(cpu, sd->span)) {
+			if (cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 				update_shares(sd);
 				break;
 			}
@@ -2315,7 +2320,7 @@
 	else {
 		struct sched_domain *sd;
 		for_each_domain(this_cpu, sd) {
-			if (cpu_isset(cpu, sd->span)) {
+			if (cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 				schedstat_inc(sd, ttwu_wake_remote);
 				break;
 			}
@@ -2846,7 +2851,7 @@
 	struct rq *rq;
 
 	rq = task_rq_lock(p, &flags);
-	if (!cpu_isset(dest_cpu, p->cpus_allowed)
+	if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed)
 	    || unlikely(!cpu_active(dest_cpu)))
 		goto out;
 
@@ -2911,7 +2916,7 @@
 	 * 2) cannot be migrated to this CPU due to cpus_allowed, or
 	 * 3) are cache-hot on their current CPU.
 	 */
-	if (!cpu_isset(this_cpu, p->cpus_allowed)) {
+	if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) {
 		schedstat_inc(p, se.nr_failed_migrations_affine);
 		return 0;
 	}
@@ -3086,7 +3091,7 @@
 static struct sched_group *
 find_busiest_group(struct sched_domain *sd, int this_cpu,
 		   unsigned long *imbalance, enum cpu_idle_type idle,
-		   int *sd_idle, const cpumask_t *cpus, int *balance)
+		   int *sd_idle, const struct cpumask *cpus, int *balance)
 {
 	struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups;
 	unsigned long max_load, avg_load, total_load, this_load, total_pwr;
@@ -3122,10 +3127,11 @@
 		unsigned long sum_avg_load_per_task;
 		unsigned long avg_load_per_task;
 
-		local_group = cpu_isset(this_cpu, group->cpumask);
+		local_group = cpumask_test_cpu(this_cpu,
+					       sched_group_cpus(group));
 
 		if (local_group)
-			balance_cpu = first_cpu(group->cpumask);
+			balance_cpu = cpumask_first(sched_group_cpus(group));
 
 		/* Tally up the load of all CPUs in the group */
 		sum_weighted_load = sum_nr_running = avg_load = 0;
@@ -3134,13 +3140,8 @@
 		max_cpu_load = 0;
 		min_cpu_load = ~0UL;
 
-		for_each_cpu_mask_nr(i, group->cpumask) {
-			struct rq *rq;
-
-			if (!cpu_isset(i, *cpus))
-				continue;
-
-			rq = cpu_rq(i);
+		for_each_cpu_and(i, sched_group_cpus(group), cpus) {
+			struct rq *rq = cpu_rq(i);
 
 			if (*sd_idle && rq->nr_running)
 				*sd_idle = 0;
@@ -3251,8 +3252,8 @@
 		 */
 		if ((sum_nr_running < min_nr_running) ||
 		    (sum_nr_running == min_nr_running &&
-		     first_cpu(group->cpumask) <
-		     first_cpu(group_min->cpumask))) {
+		     cpumask_first(sched_group_cpus(group)) >
+		     cpumask_first(sched_group_cpus(group_min)))) {
 			group_min = group;
 			min_nr_running = sum_nr_running;
 			min_load_per_task = sum_weighted_load /
@@ -3267,8 +3268,8 @@
 		if (sum_nr_running <= group_capacity - 1) {
 			if (sum_nr_running > leader_nr_running ||
 			    (sum_nr_running == leader_nr_running &&
-			     first_cpu(group->cpumask) >
-			      first_cpu(group_leader->cpumask))) {
+			     cpumask_first(sched_group_cpus(group)) <
+			     cpumask_first(sched_group_cpus(group_leader)))) {
 				group_leader = group;
 				leader_nr_running = sum_nr_running;
 			}
@@ -3394,6 +3395,10 @@
 
 	if (this == group_leader && group_leader != group_min) {
 		*imbalance = min_load_per_task;
+		if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP) {
+			cpu_rq(this_cpu)->rd->sched_mc_preferred_wakeup_cpu =
+				cpumask_first(sched_group_cpus(group_leader));
+		}
 		return group_min;
 	}
 #endif
@@ -3407,16 +3412,16 @@
  */
 static struct rq *
 find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
-		   unsigned long imbalance, const cpumask_t *cpus)
+		   unsigned long imbalance, const struct cpumask *cpus)
 {
 	struct rq *busiest = NULL, *rq;
 	unsigned long max_load = 0;
 	int i;
 
-	for_each_cpu_mask_nr(i, group->cpumask) {
+	for_each_cpu(i, sched_group_cpus(group)) {
 		unsigned long wl;
 
-		if (!cpu_isset(i, *cpus))
+		if (!cpumask_test_cpu(i, cpus))
 			continue;
 
 		rq = cpu_rq(i);
@@ -3446,7 +3451,7 @@
  */
 static int load_balance(int this_cpu, struct rq *this_rq,
 			struct sched_domain *sd, enum cpu_idle_type idle,
-			int *balance, cpumask_t *cpus)
+			int *balance, struct cpumask *cpus)
 {
 	int ld_moved, all_pinned = 0, active_balance = 0, sd_idle = 0;
 	struct sched_group *group;
@@ -3454,7 +3459,7 @@
 	struct rq *busiest;
 	unsigned long flags;
 
-	cpus_setall(*cpus);
+	cpumask_setall(cpus);
 
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
@@ -3514,8 +3519,8 @@
 
 		/* All tasks on this runqueue were pinned by CPU affinity */
 		if (unlikely(all_pinned)) {
-			cpu_clear(cpu_of(busiest), *cpus);
-			if (!cpus_empty(*cpus))
+			cpumask_clear_cpu(cpu_of(busiest), cpus);
+			if (!cpumask_empty(cpus))
 				goto redo;
 			goto out_balanced;
 		}
@@ -3532,7 +3537,8 @@
 			/* don't kick the migration_thread, if the curr
 			 * task on busiest cpu can't be moved to this_cpu
 			 */
-			if (!cpu_isset(this_cpu, busiest->curr->cpus_allowed)) {
+			if (!cpumask_test_cpu(this_cpu,
+					      &busiest->curr->cpus_allowed)) {
 				spin_unlock_irqrestore(&busiest->lock, flags);
 				all_pinned = 1;
 				goto out_one_pinned;
@@ -3607,7 +3613,7 @@
  */
 static int
 load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd,
-			cpumask_t *cpus)
+			struct cpumask *cpus)
 {
 	struct sched_group *group;
 	struct rq *busiest = NULL;
@@ -3616,7 +3622,7 @@
 	int sd_idle = 0;
 	int all_pinned = 0;
 
-	cpus_setall(*cpus);
+	cpumask_setall(cpus);
 
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
@@ -3660,17 +3666,71 @@
 		double_unlock_balance(this_rq, busiest);
 
 		if (unlikely(all_pinned)) {
-			cpu_clear(cpu_of(busiest), *cpus);
-			if (!cpus_empty(*cpus))
+			cpumask_clear_cpu(cpu_of(busiest), cpus);
+			if (!cpumask_empty(cpus))
 				goto redo;
 		}
 	}
 
 	if (!ld_moved) {
+		int active_balance = 0;
+
 		schedstat_inc(sd, lb_failed[CPU_NEWLY_IDLE]);
 		if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
 		    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
 			return -1;
+
+		if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
+			return -1;
+
+		if (sd->nr_balance_failed++ < 2)
+			return -1;
+
+		/*
+		 * The only task running in a non-idle cpu can be moved to this
+		 * cpu in an attempt to completely freeup the other CPU
+		 * package. The same method used to move task in load_balance()
+		 * have been extended for load_balance_newidle() to speedup
+		 * consolidation at sched_mc=POWERSAVINGS_BALANCE_WAKEUP (2)
+		 *
+		 * The package power saving logic comes from
+		 * find_busiest_group().  If there are no imbalance, then
+		 * f_b_g() will return NULL.  However when sched_mc={1,2} then
+		 * f_b_g() will select a group from which a running task may be
+		 * pulled to this cpu in order to make the other package idle.
+		 * If there is no opportunity to make a package idle and if
+		 * there are no imbalance, then f_b_g() will return NULL and no
+		 * action will be taken in load_balance_newidle().
+		 *
+		 * Under normal task pull operation due to imbalance, there
+		 * will be more than one task in the source run queue and
+		 * move_tasks() will succeed.  ld_moved will be true and this
+		 * active balance code will not be triggered.
+		 */
+
+		/* Lock busiest in correct order while this_rq is held */
+		double_lock_balance(this_rq, busiest);
+
+		/*
+		 * don't kick the migration_thread, if the curr
+		 * task on busiest cpu can't be moved to this_cpu
+		 */
+		if (!cpu_isset(this_cpu, busiest->curr->cpus_allowed)) {
+			double_unlock_balance(this_rq, busiest);
+			all_pinned = 1;
+			return ld_moved;
+		}
+
+		if (!busiest->active_balance) {
+			busiest->active_balance = 1;
+			busiest->push_cpu = this_cpu;
+			active_balance = 1;
+		}
+
+		double_unlock_balance(this_rq, busiest);
+		if (active_balance)
+			wake_up_process(busiest->migration_thread);
+
 	} else
 		sd->nr_balance_failed = 0;
 
@@ -3696,7 +3756,10 @@
 	struct sched_domain *sd;
 	int pulled_task = 0;
 	unsigned long next_balance = jiffies + HZ;
-	cpumask_t tmpmask;
+	cpumask_var_t tmpmask;
+
+	if (!alloc_cpumask_var(&tmpmask, GFP_ATOMIC))
+		return;
 
 	for_each_domain(this_cpu, sd) {
 		unsigned long interval;
@@ -3707,7 +3770,7 @@
 		if (sd->flags & SD_BALANCE_NEWIDLE)
 			/* If we've pulled tasks over stop searching: */
 			pulled_task = load_balance_newidle(this_cpu, this_rq,
-							   sd, &tmpmask);
+							   sd, tmpmask);
 
 		interval = msecs_to_jiffies(sd->balance_interval);
 		if (time_after(next_balance, sd->last_balance + interval))
@@ -3722,6 +3785,7 @@
 		 */
 		this_rq->next_balance = next_balance;
 	}
+	free_cpumask_var(tmpmask);
 }
 
 /*
@@ -3759,7 +3823,7 @@
 	/* Search for an sd spanning us and the target CPU. */
 	for_each_domain(target_cpu, sd) {
 		if ((sd->flags & SD_LOAD_BALANCE) &&
-		    cpu_isset(busiest_cpu, sd->span))
+		    cpumask_test_cpu(busiest_cpu, sched_domain_span(sd)))
 				break;
 	}
 
@@ -3778,10 +3842,9 @@
 #ifdef CONFIG_NO_HZ
 static struct {
 	atomic_t load_balancer;
-	cpumask_t cpu_mask;
+	cpumask_var_t cpu_mask;
 } nohz ____cacheline_aligned = {
 	.load_balancer = ATOMIC_INIT(-1),
-	.cpu_mask = CPU_MASK_NONE,
 };
 
 /*
@@ -3809,7 +3872,7 @@
 	int cpu = smp_processor_id();
 
 	if (stop_tick) {
-		cpu_set(cpu, nohz.cpu_mask);
+		cpumask_set_cpu(cpu, nohz.cpu_mask);
 		cpu_rq(cpu)->in_nohz_recently = 1;
 
 		/*
@@ -3823,7 +3886,7 @@
 		}
 
 		/* time for ilb owner also to sleep */
-		if (cpus_weight(nohz.cpu_mask) == num_online_cpus()) {
+		if (cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
 			if (atomic_read(&nohz.load_balancer) == cpu)
 				atomic_set(&nohz.load_balancer, -1);
 			return 0;
@@ -3836,10 +3899,10 @@
 		} else if (atomic_read(&nohz.load_balancer) == cpu)
 			return 1;
 	} else {
-		if (!cpu_isset(cpu, nohz.cpu_mask))
+		if (!cpumask_test_cpu(cpu, nohz.cpu_mask))
 			return 0;
 
-		cpu_clear(cpu, nohz.cpu_mask);
+		cpumask_clear_cpu(cpu, nohz.cpu_mask);
 
 		if (atomic_read(&nohz.load_balancer) == cpu)
 			if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
@@ -3867,7 +3930,11 @@
 	unsigned long next_balance = jiffies + 60*HZ;
 	int update_next_balance = 0;
 	int need_serialize;
-	cpumask_t tmp;
+	cpumask_var_t tmp;
+
+	/* Fails alloc?  Rebalancing probably not a priority right now. */
+	if (!alloc_cpumask_var(&tmp, GFP_ATOMIC))
+		return;
 
 	for_each_domain(cpu, sd) {
 		if (!(sd->flags & SD_LOAD_BALANCE))
@@ -3892,7 +3959,7 @@
 		}
 
 		if (time_after_eq(jiffies, sd->last_balance + interval)) {
-			if (load_balance(cpu, rq, sd, idle, &balance, &tmp)) {
+			if (load_balance(cpu, rq, sd, idle, &balance, tmp)) {
 				/*
 				 * We've pulled tasks over so either we're no
 				 * longer idle, or one of our SMT siblings is
@@ -3926,6 +3993,8 @@
 	 */
 	if (likely(update_next_balance))
 		rq->next_balance = next_balance;
+
+	free_cpumask_var(tmp);
 }
 
 /*
@@ -3950,12 +4019,13 @@
 	 */
 	if (this_rq->idle_at_tick &&
 	    atomic_read(&nohz.load_balancer) == this_cpu) {
-		cpumask_t cpus = nohz.cpu_mask;
 		struct rq *rq;
 		int balance_cpu;
 
-		cpu_clear(this_cpu, cpus);
-		for_each_cpu_mask_nr(balance_cpu, cpus) {
+		for_each_cpu(balance_cpu, nohz.cpu_mask) {
+			if (balance_cpu == this_cpu)
+				continue;
+
 			/*
 			 * If this cpu gets work to do, stop the load balancing
 			 * work being done for other cpus. Next load
@@ -3993,7 +4063,7 @@
 		rq->in_nohz_recently = 0;
 
 		if (atomic_read(&nohz.load_balancer) == cpu) {
-			cpu_clear(cpu, nohz.cpu_mask);
+			cpumask_clear_cpu(cpu, nohz.cpu_mask);
 			atomic_set(&nohz.load_balancer, -1);
 		}
 
@@ -4006,7 +4076,7 @@
 			 * TBD: Traverse the sched domains and nominate
 			 * the nearest cpu in the nohz.cpu_mask.
 			 */
-			int ilb = first_cpu(nohz.cpu_mask);
+			int ilb = cpumask_first(nohz.cpu_mask);
 
 			if (ilb < nr_cpu_ids)
 				resched_cpu(ilb);
@@ -4018,7 +4088,7 @@
 	 * cpus with ticks stopped, is it time for that to stop?
 	 */
 	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) == cpu &&
-	    cpus_weight(nohz.cpu_mask) == num_online_cpus()) {
+	    cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
 		resched_cpu(cpu);
 		return;
 	}
@@ -4028,7 +4098,7 @@
 	 * someone else, then no need raise the SCHED_SOFTIRQ
 	 */
 	if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) != cpu &&
-	    cpu_isset(cpu, nohz.cpu_mask))
+	    cpumask_test_cpu(cpu, nohz.cpu_mask))
 		return;
 #endif
 	if (time_after_eq(jiffies, rq->next_balance))
@@ -4080,13 +4150,17 @@
  * Account user cpu time to a process.
  * @p: the process that the cpu time gets accounted to
  * @cputime: the cpu time spent in user space since the last update
+ * @cputime_scaled: cputime scaled by cpu frequency
  */
-void account_user_time(struct task_struct *p, cputime_t cputime)
+void account_user_time(struct task_struct *p, cputime_t cputime,
+		       cputime_t cputime_scaled)
 {
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
 	cputime64_t tmp;
 
+	/* Add user time to process. */
 	p->utime = cputime_add(p->utime, cputime);
+	p->utimescaled = cputime_add(p->utimescaled, cputime_scaled);
 	account_group_user_time(p, cputime);
 
 	/* Add user time to cpustat. */
@@ -4103,51 +4177,48 @@
  * Account guest cpu time to a process.
  * @p: the process that the cpu time gets accounted to
  * @cputime: the cpu time spent in virtual machine since the last update
+ * @cputime_scaled: cputime scaled by cpu frequency
  */
-static void account_guest_time(struct task_struct *p, cputime_t cputime)
+static void account_guest_time(struct task_struct *p, cputime_t cputime,
+			       cputime_t cputime_scaled)
 {
 	cputime64_t tmp;
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
 
 	tmp = cputime_to_cputime64(cputime);
 
+	/* Add guest time to process. */
 	p->utime = cputime_add(p->utime, cputime);
+	p->utimescaled = cputime_add(p->utimescaled, cputime_scaled);
 	account_group_user_time(p, cputime);
 	p->gtime = cputime_add(p->gtime, cputime);
 
+	/* Add guest time to cpustat. */
 	cpustat->user = cputime64_add(cpustat->user, tmp);
 	cpustat->guest = cputime64_add(cpustat->guest, tmp);
 }
 
 /*
- * Account scaled user cpu time to a process.
- * @p: the process that the cpu time gets accounted to
- * @cputime: the cpu time spent in user space since the last update
- */
-void account_user_time_scaled(struct task_struct *p, cputime_t cputime)
-{
-	p->utimescaled = cputime_add(p->utimescaled, cputime);
-}
-
-/*
  * Account system cpu time to a process.
  * @p: the process that the cpu time gets accounted to
  * @hardirq_offset: the offset to subtract from hardirq_count()
  * @cputime: the cpu time spent in kernel space since the last update
+ * @cputime_scaled: cputime scaled by cpu frequency
  */
 void account_system_time(struct task_struct *p, int hardirq_offset,
-			 cputime_t cputime)
+			 cputime_t cputime, cputime_t cputime_scaled)
 {
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
-	struct rq *rq = this_rq();
 	cputime64_t tmp;
 
 	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
-		account_guest_time(p, cputime);
+		account_guest_time(p, cputime, cputime_scaled);
 		return;
 	}
 
+	/* Add system time to process. */
 	p->stime = cputime_add(p->stime, cputime);
+	p->stimescaled = cputime_add(p->stimescaled, cputime_scaled);
 	account_group_system_time(p, cputime);
 
 	/* Add system time to cpustat. */
@@ -4156,48 +4227,84 @@
 		cpustat->irq = cputime64_add(cpustat->irq, tmp);
 	else if (softirq_count())
 		cpustat->softirq = cputime64_add(cpustat->softirq, tmp);
-	else if (p != rq->idle)
-		cpustat->system = cputime64_add(cpustat->system, tmp);
-	else if (atomic_read(&rq->nr_iowait) > 0)
-		cpustat->iowait = cputime64_add(cpustat->iowait, tmp);
 	else
-		cpustat->idle = cputime64_add(cpustat->idle, tmp);
+		cpustat->system = cputime64_add(cpustat->system, tmp);
+
 	/* Account for system time used */
 	acct_update_integrals(p);
 }
 
 /*
- * Account scaled system cpu time to a process.
- * @p: the process that the cpu time gets accounted to
- * @hardirq_offset: the offset to subtract from hardirq_count()
- * @cputime: the cpu time spent in kernel space since the last update
+ * Account for involuntary wait time.
+ * @steal: the cpu time spent in involuntary wait
  */
-void account_system_time_scaled(struct task_struct *p, cputime_t cputime)
+void account_steal_time(cputime_t cputime)
 {
-	p->stimescaled = cputime_add(p->stimescaled, cputime);
+	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
+	cputime64_t cputime64 = cputime_to_cputime64(cputime);
+
+	cpustat->steal = cputime64_add(cpustat->steal, cputime64);
 }
 
 /*
- * Account for involuntary wait time.
- * @p: the process from which the cpu time has been stolen
- * @steal: the cpu time spent in involuntary wait
+ * Account for idle time.
+ * @cputime: the cpu time spent in idle wait
  */
-void account_steal_time(struct task_struct *p, cputime_t steal)
+void account_idle_time(cputime_t cputime)
 {
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
-	cputime64_t tmp = cputime_to_cputime64(steal);
+	cputime64_t cputime64 = cputime_to_cputime64(cputime);
 	struct rq *rq = this_rq();
 
-	if (p == rq->idle) {
-		p->stime = cputime_add(p->stime, steal);
-		if (atomic_read(&rq->nr_iowait) > 0)
-			cpustat->iowait = cputime64_add(cpustat->iowait, tmp);
-		else
-			cpustat->idle = cputime64_add(cpustat->idle, tmp);
-	} else
-		cpustat->steal = cputime64_add(cpustat->steal, tmp);
+	if (atomic_read(&rq->nr_iowait) > 0)
+		cpustat->iowait = cputime64_add(cpustat->iowait, cputime64);
+	else
+		cpustat->idle = cputime64_add(cpustat->idle, cputime64);
 }
 
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+
+/*
+ * Account a single tick of cpu time.
+ * @p: the process that the cpu time gets accounted to
+ * @user_tick: indicates if the tick is a user or a system tick
+ */
+void account_process_tick(struct task_struct *p, int user_tick)
+{
+	cputime_t one_jiffy = jiffies_to_cputime(1);
+	cputime_t one_jiffy_scaled = cputime_to_scaled(one_jiffy);
+	struct rq *rq = this_rq();
+
+	if (user_tick)
+		account_user_time(p, one_jiffy, one_jiffy_scaled);
+	else if (p != rq->idle)
+		account_system_time(p, HARDIRQ_OFFSET, one_jiffy,
+				    one_jiffy_scaled);
+	else
+		account_idle_time(one_jiffy);
+}
+
+/*
+ * Account multiple ticks of steal time.
+ * @p: the process from which the cpu time has been stolen
+ * @ticks: number of stolen ticks
+ */
+void account_steal_ticks(unsigned long ticks)
+{
+	account_steal_time(jiffies_to_cputime(ticks));
+}
+
+/*
+ * Account multiple ticks of idle time.
+ * @ticks: number of stolen ticks
+ */
+void account_idle_ticks(unsigned long ticks)
+{
+	account_idle_time(jiffies_to_cputime(ticks));
+}
+
+#endif
+
 /*
  * Use precise platform statistics if available:
  */
@@ -5401,10 +5508,9 @@
 	return retval;
 }
 
-long sched_setaffinity(pid_t pid, const cpumask_t *in_mask)
+long sched_setaffinity(pid_t pid, const struct cpumask *in_mask)
 {
-	cpumask_t cpus_allowed;
-	cpumask_t new_mask = *in_mask;
+	cpumask_var_t cpus_allowed, new_mask;
 	struct task_struct *p;
 	int retval;
 
@@ -5426,6 +5532,14 @@
 	get_task_struct(p);
 	read_unlock(&tasklist_lock);
 
+	if (!alloc_cpumask_var(&cpus_allowed, GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto out_put_task;
+	}
+	if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto out_free_cpus_allowed;
+	}
 	retval = -EPERM;
 	if (!check_same_owner(p) && !capable(CAP_SYS_NICE))
 		goto out_unlock;
@@ -5434,37 +5548,41 @@
 	if (retval)
 		goto out_unlock;
 
-	cpuset_cpus_allowed(p, &cpus_allowed);
-	cpus_and(new_mask, new_mask, cpus_allowed);
+	cpuset_cpus_allowed(p, cpus_allowed);
+	cpumask_and(new_mask, in_mask, cpus_allowed);
  again:
-	retval = set_cpus_allowed_ptr(p, &new_mask);
+	retval = set_cpus_allowed_ptr(p, new_mask);
 
 	if (!retval) {
-		cpuset_cpus_allowed(p, &cpus_allowed);
-		if (!cpus_subset(new_mask, cpus_allowed)) {
+		cpuset_cpus_allowed(p, cpus_allowed);
+		if (!cpumask_subset(new_mask, cpus_allowed)) {
 			/*
 			 * We must have raced with a concurrent cpuset
 			 * update. Just reset the cpus_allowed to the
 			 * cpuset's cpus_allowed
 			 */
-			new_mask = cpus_allowed;
+			cpumask_copy(new_mask, cpus_allowed);
 			goto again;
 		}
 	}
 out_unlock:
+	free_cpumask_var(new_mask);
+out_free_cpus_allowed:
+	free_cpumask_var(cpus_allowed);
+out_put_task:
 	put_task_struct(p);
 	put_online_cpus();
 	return retval;
 }
 
 static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len,
-			     cpumask_t *new_mask)
+			     struct cpumask *new_mask)
 {
-	if (len < sizeof(cpumask_t)) {
-		memset(new_mask, 0, sizeof(cpumask_t));
-	} else if (len > sizeof(cpumask_t)) {
-		len = sizeof(cpumask_t);
-	}
+	if (len < cpumask_size())
+		cpumask_clear(new_mask);
+	else if (len > cpumask_size())
+		len = cpumask_size();
+
 	return copy_from_user(new_mask, user_mask_ptr, len) ? -EFAULT : 0;
 }
 
@@ -5477,17 +5595,20 @@
 asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len,
 				      unsigned long __user *user_mask_ptr)
 {
-	cpumask_t new_mask;
+	cpumask_var_t new_mask;
 	int retval;
 
-	retval = get_user_cpu_mask(user_mask_ptr, len, &new_mask);
-	if (retval)
-		return retval;
+	if (!alloc_cpumask_var(&new_mask, GFP_KERNEL))
+		return -ENOMEM;
 
-	return sched_setaffinity(pid, &new_mask);
+	retval = get_user_cpu_mask(user_mask_ptr, len, new_mask);
+	if (retval == 0)
+		retval = sched_setaffinity(pid, new_mask);
+	free_cpumask_var(new_mask);
+	return retval;
 }
 
-long sched_getaffinity(pid_t pid, cpumask_t *mask)
+long sched_getaffinity(pid_t pid, struct cpumask *mask)
 {
 	struct task_struct *p;
 	int retval;
@@ -5504,7 +5625,7 @@
 	if (retval)
 		goto out_unlock;
 
-	cpus_and(*mask, p->cpus_allowed, cpu_online_map);
+	cpumask_and(mask, &p->cpus_allowed, cpu_online_mask);
 
 out_unlock:
 	read_unlock(&tasklist_lock);
@@ -5523,19 +5644,24 @@
 				      unsigned long __user *user_mask_ptr)
 {
 	int ret;
-	cpumask_t mask;
+	cpumask_var_t mask;
 
-	if (len < sizeof(cpumask_t))
+	if (len < cpumask_size())
 		return -EINVAL;
 
-	ret = sched_getaffinity(pid, &mask);
-	if (ret < 0)
-		return ret;
+	if (!alloc_cpumask_var(&mask, GFP_KERNEL))
+		return -ENOMEM;
 
-	if (copy_to_user(user_mask_ptr, &mask, sizeof(cpumask_t)))
-		return -EFAULT;
+	ret = sched_getaffinity(pid, mask);
+	if (ret == 0) {
+		if (copy_to_user(user_mask_ptr, mask, cpumask_size()))
+			ret = -EFAULT;
+		else
+			ret = cpumask_size();
+	}
+	free_cpumask_var(mask);
 
-	return sizeof(cpumask_t);
+	return ret;
 }
 
 /**
@@ -5877,7 +6003,7 @@
 	idle->se.exec_start = sched_clock();
 
 	idle->prio = idle->normal_prio = MAX_PRIO;
-	idle->cpus_allowed = cpumask_of_cpu(cpu);
+	cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu));
 	__set_task_cpu(idle, cpu);
 
 	rq->curr = rq->idle = idle;
@@ -5904,9 +6030,9 @@
  * indicates which cpus entered this state. This is used
  * in the rcu update to wait only for active cpus. For system
  * which do not switch off the HZ timer nohz_cpu_mask should
- * always be CPU_MASK_NONE.
+ * always be CPU_BITS_NONE.
  */
-cpumask_t nohz_cpu_mask = CPU_MASK_NONE;
+cpumask_var_t nohz_cpu_mask;
 
 /*
  * Increase the granularity value when there are more CPUs,
@@ -5961,7 +6087,7 @@
  * task must not exit() & deallocate itself prematurely. The
  * call is not atomic; no spinlocks may be held.
  */
-int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
+int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask)
 {
 	struct migration_req req;
 	unsigned long flags;
@@ -5969,13 +6095,13 @@
 	int ret = 0;
 
 	rq = task_rq_lock(p, &flags);
-	if (!cpus_intersects(*new_mask, cpu_online_map)) {
+	if (!cpumask_intersects(new_mask, cpu_online_mask)) {
 		ret = -EINVAL;
 		goto out;
 	}
 
 	if (unlikely((p->flags & PF_THREAD_BOUND) && p != current &&
-		     !cpus_equal(p->cpus_allowed, *new_mask))) {
+		     !cpumask_equal(&p->cpus_allowed, new_mask))) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -5983,15 +6109,15 @@
 	if (p->sched_class->set_cpus_allowed)
 		p->sched_class->set_cpus_allowed(p, new_mask);
 	else {
-		p->cpus_allowed = *new_mask;
-		p->rt.nr_cpus_allowed = cpus_weight(*new_mask);
+		cpumask_copy(&p->cpus_allowed, new_mask);
+		p->rt.nr_cpus_allowed = cpumask_weight(new_mask);
 	}
 
 	/* Can the task run on the task's current CPU? If so, we're done */
-	if (cpu_isset(task_cpu(p), *new_mask))
+	if (cpumask_test_cpu(task_cpu(p), new_mask))
 		goto out;
 
-	if (migrate_task(p, any_online_cpu(*new_mask), &req)) {
+	if (migrate_task(p, cpumask_any_and(cpu_online_mask, new_mask), &req)) {
 		/* Need help from migration thread: drop lock and wait. */
 		task_rq_unlock(rq, &flags);
 		wake_up_process(rq->migration_thread);
@@ -6033,7 +6159,7 @@
 	if (task_cpu(p) != src_cpu)
 		goto done;
 	/* Affinity changed (again). */
-	if (!cpu_isset(dest_cpu, p->cpus_allowed))
+	if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed))
 		goto fail;
 
 	on_rq = p->se.on_rq;
@@ -6130,50 +6256,43 @@
  */
 static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p)
 {
-	unsigned long flags;
-	cpumask_t mask;
-	struct rq *rq;
 	int dest_cpu;
+	/* FIXME: Use cpumask_of_node here. */
+	cpumask_t _nodemask = node_to_cpumask(cpu_to_node(dead_cpu));
+	const struct cpumask *nodemask = &_nodemask;
 
-	do {
-		/* On same node? */
-		mask = node_to_cpumask(cpu_to_node(dead_cpu));
-		cpus_and(mask, mask, p->cpus_allowed);
-		dest_cpu = any_online_cpu(mask);
+again:
+	/* Look for allowed, online CPU in same node. */
+	for_each_cpu_and(dest_cpu, nodemask, cpu_online_mask)
+		if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed))
+			goto move;
 
-		/* On any allowed CPU? */
-		if (dest_cpu >= nr_cpu_ids)
-			dest_cpu = any_online_cpu(p->cpus_allowed);
+	/* Any allowed, online CPU? */
+	dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_online_mask);
+	if (dest_cpu < nr_cpu_ids)
+		goto move;
 
-		/* No more Mr. Nice Guy. */
-		if (dest_cpu >= nr_cpu_ids) {
-			cpumask_t cpus_allowed;
+	/* No more Mr. Nice Guy. */
+	if (dest_cpu >= nr_cpu_ids) {
+		cpuset_cpus_allowed_locked(p, &p->cpus_allowed);
+		dest_cpu = cpumask_any_and(cpu_online_mask, &p->cpus_allowed);
 
-			cpuset_cpus_allowed_locked(p, &cpus_allowed);
-			/*
-			 * Try to stay on the same cpuset, where the
-			 * current cpuset may be a subset of all cpus.
-			 * The cpuset_cpus_allowed_locked() variant of
-			 * cpuset_cpus_allowed() will not block. It must be
-			 * called within calls to cpuset_lock/cpuset_unlock.
-			 */
-			rq = task_rq_lock(p, &flags);
-			p->cpus_allowed = cpus_allowed;
-			dest_cpu = any_online_cpu(p->cpus_allowed);
-			task_rq_unlock(rq, &flags);
-
-			/*
-			 * Don't tell them about moving exiting tasks or
-			 * kernel threads (both mm NULL), since they never
-			 * leave kernel.
-			 */
-			if (p->mm && printk_ratelimit()) {
-				printk(KERN_INFO "process %d (%s) no "
-				       "longer affine to cpu%d\n",
-					task_pid_nr(p), p->comm, dead_cpu);
-			}
+		/*
+		 * Don't tell them about moving exiting tasks or
+		 * kernel threads (both mm NULL), since they never
+		 * leave kernel.
+		 */
+		if (p->mm && printk_ratelimit()) {
+			printk(KERN_INFO "process %d (%s) no "
+			       "longer affine to cpu%d\n",
+			       task_pid_nr(p), p->comm, dead_cpu);
 		}
-	} while (!__migrate_task_irq(p, dead_cpu, dest_cpu));
+	}
+
+move:
+	/* It can have affinity changed while we were choosing. */
+	if (unlikely(!__migrate_task_irq(p, dead_cpu, dest_cpu)))
+		goto again;
 }
 
 /*
@@ -6185,7 +6304,7 @@
  */
 static void migrate_nr_uninterruptible(struct rq *rq_src)
 {
-	struct rq *rq_dest = cpu_rq(any_online_cpu(*CPU_MASK_ALL_PTR));
+	struct rq *rq_dest = cpu_rq(cpumask_any(cpu_online_mask));
 	unsigned long flags;
 
 	local_irq_save(flags);
@@ -6475,7 +6594,7 @@
 	if (!rq->online) {
 		const struct sched_class *class;
 
-		cpu_set(rq->cpu, rq->rd->online);
+		cpumask_set_cpu(rq->cpu, rq->rd->online);
 		rq->online = 1;
 
 		for_each_class(class) {
@@ -6495,7 +6614,7 @@
 				class->rq_offline(rq);
 		}
 
-		cpu_clear(rq->cpu, rq->rd->online);
+		cpumask_clear_cpu(rq->cpu, rq->rd->online);
 		rq->online = 0;
 	}
 }
@@ -6536,7 +6655,7 @@
 		rq = cpu_rq(cpu);
 		spin_lock_irqsave(&rq->lock, flags);
 		if (rq->rd) {
-			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
 
 			set_rq_online(rq);
 		}
@@ -6550,7 +6669,7 @@
 			break;
 		/* Unbind it from offline cpu so it can run. Fall thru. */
 		kthread_bind(cpu_rq(cpu)->migration_thread,
-			     any_online_cpu(cpu_online_map));
+			     cpumask_any(cpu_online_mask));
 		kthread_stop(cpu_rq(cpu)->migration_thread);
 		cpu_rq(cpu)->migration_thread = NULL;
 		break;
@@ -6600,7 +6719,7 @@
 		rq = cpu_rq(cpu);
 		spin_lock_irqsave(&rq->lock, flags);
 		if (rq->rd) {
-			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
 			set_rq_offline(rq);
 		}
 		spin_unlock_irqrestore(&rq->lock, flags);
@@ -6639,13 +6758,13 @@
 #ifdef CONFIG_SCHED_DEBUG
 
 static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
-				  cpumask_t *groupmask)
+				  struct cpumask *groupmask)
 {
 	struct sched_group *group = sd->groups;
 	char str[256];
 
-	cpulist_scnprintf(str, sizeof(str), sd->span);
-	cpus_clear(*groupmask);
+	cpulist_scnprintf(str, sizeof(str), sched_domain_span(sd));
+	cpumask_clear(groupmask);
 
 	printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
 
@@ -6659,11 +6778,11 @@
 
 	printk(KERN_CONT "span %s level %s\n", str, sd->name);
 
-	if (!cpu_isset(cpu, sd->span)) {
+	if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 		printk(KERN_ERR "ERROR: domain->span does not contain "
 				"CPU%d\n", cpu);
 	}
-	if (!cpu_isset(cpu, group->cpumask)) {
+	if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) {
 		printk(KERN_ERR "ERROR: domain->groups does not contain"
 				" CPU%d\n", cpu);
 	}
@@ -6683,31 +6802,32 @@
 			break;
 		}
 
-		if (!cpus_weight(group->cpumask)) {
+		if (!cpumask_weight(sched_group_cpus(group))) {
 			printk(KERN_CONT "\n");
 			printk(KERN_ERR "ERROR: empty group\n");
 			break;
 		}
 
-		if (cpus_intersects(*groupmask, group->cpumask)) {
+		if (cpumask_intersects(groupmask, sched_group_cpus(group))) {
 			printk(KERN_CONT "\n");
 			printk(KERN_ERR "ERROR: repeated CPUs\n");
 			break;
 		}
 
-		cpus_or(*groupmask, *groupmask, group->cpumask);
+		cpumask_or(groupmask, groupmask, sched_group_cpus(group));
 
-		cpulist_scnprintf(str, sizeof(str), group->cpumask);
+		cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group));
 		printk(KERN_CONT " %s", str);
 
 		group = group->next;
 	} while (group != sd->groups);
 	printk(KERN_CONT "\n");
 
-	if (!cpus_equal(sd->span, *groupmask))
+	if (!cpumask_equal(sched_domain_span(sd), groupmask))
 		printk(KERN_ERR "ERROR: groups don't span domain->span\n");
 
-	if (sd->parent && !cpus_subset(*groupmask, sd->parent->span))
+	if (sd->parent &&
+	    !cpumask_subset(groupmask, sched_domain_span(sd->parent)))
 		printk(KERN_ERR "ERROR: parent span is not a superset "
 			"of domain->span\n");
 	return 0;
@@ -6715,7 +6835,7 @@
 
 static void sched_domain_debug(struct sched_domain *sd, int cpu)
 {
-	cpumask_t *groupmask;
+	cpumask_var_t groupmask;
 	int level = 0;
 
 	if (!sd) {
@@ -6725,8 +6845,7 @@
 
 	printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu);
 
-	groupmask = kmalloc(sizeof(cpumask_t), GFP_KERNEL);
-	if (!groupmask) {
+	if (!alloc_cpumask_var(&groupmask, GFP_KERNEL)) {
 		printk(KERN_DEBUG "Cannot load-balance (out of memory)\n");
 		return;
 	}
@@ -6739,7 +6858,7 @@
 		if (!sd)
 			break;
 	}
-	kfree(groupmask);
+	free_cpumask_var(groupmask);
 }
 #else /* !CONFIG_SCHED_DEBUG */
 # define sched_domain_debug(sd, cpu) do { } while (0)
@@ -6747,7 +6866,7 @@
 
 static int sd_degenerate(struct sched_domain *sd)
 {
-	if (cpus_weight(sd->span) == 1)
+	if (cpumask_weight(sched_domain_span(sd)) == 1)
 		return 1;
 
 	/* Following flags need at least 2 groups */
@@ -6778,7 +6897,7 @@
 	if (sd_degenerate(parent))
 		return 1;
 
-	if (!cpus_equal(sd->span, parent->span))
+	if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent)))
 		return 0;
 
 	/* Does parent contain flags not in child? */
@@ -6802,6 +6921,16 @@
 	return 1;
 }
 
+static void free_rootdomain(struct root_domain *rd)
+{
+	cpupri_cleanup(&rd->cpupri);
+
+	free_cpumask_var(rd->rto_mask);
+	free_cpumask_var(rd->online);
+	free_cpumask_var(rd->span);
+	kfree(rd);
+}
+
 static void rq_attach_root(struct rq *rq, struct root_domain *rd)
 {
 	unsigned long flags;
@@ -6811,38 +6940,63 @@
 	if (rq->rd) {
 		struct root_domain *old_rd = rq->rd;
 
-		if (cpu_isset(rq->cpu, old_rd->online))
+		if (cpumask_test_cpu(rq->cpu, old_rd->online))
 			set_rq_offline(rq);
 
-		cpu_clear(rq->cpu, old_rd->span);
+		cpumask_clear_cpu(rq->cpu, old_rd->span);
 
 		if (atomic_dec_and_test(&old_rd->refcount))
-			kfree(old_rd);
+			free_rootdomain(old_rd);
 	}
 
 	atomic_inc(&rd->refcount);
 	rq->rd = rd;
 
-	cpu_set(rq->cpu, rd->span);
-	if (cpu_isset(rq->cpu, cpu_online_map))
+	cpumask_set_cpu(rq->cpu, rd->span);
+	if (cpumask_test_cpu(rq->cpu, cpu_online_mask))
 		set_rq_online(rq);
 
 	spin_unlock_irqrestore(&rq->lock, flags);
 }
 
-static void init_rootdomain(struct root_domain *rd)
+static int init_rootdomain(struct root_domain *rd, bool bootmem)
 {
 	memset(rd, 0, sizeof(*rd));
 
-	cpus_clear(rd->span);
-	cpus_clear(rd->online);
+	if (bootmem) {
+		alloc_bootmem_cpumask_var(&def_root_domain.span);
+		alloc_bootmem_cpumask_var(&def_root_domain.online);
+		alloc_bootmem_cpumask_var(&def_root_domain.rto_mask);
+		cpupri_init(&rd->cpupri, true);
+		return 0;
+	}
 
-	cpupri_init(&rd->cpupri);
+	if (!alloc_cpumask_var(&rd->span, GFP_KERNEL))
+		goto free_rd;
+	if (!alloc_cpumask_var(&rd->online, GFP_KERNEL))
+		goto free_span;
+	if (!alloc_cpumask_var(&rd->rto_mask, GFP_KERNEL))
+		goto free_online;
+
+	if (cpupri_init(&rd->cpupri, false) != 0)
+		goto free_rto_mask;
+	return 0;
+
+free_rto_mask:
+	free_cpumask_var(rd->rto_mask);
+free_online:
+	free_cpumask_var(rd->online);
+free_span:
+	free_cpumask_var(rd->span);
+free_rd:
+	kfree(rd);
+	return -ENOMEM;
 }
 
 static void init_defrootdomain(void)
 {
-	init_rootdomain(&def_root_domain);
+	init_rootdomain(&def_root_domain, true);
+
 	atomic_set(&def_root_domain.refcount, 1);
 }
 
@@ -6854,7 +7008,10 @@
 	if (!rd)
 		return NULL;
 
-	init_rootdomain(rd);
+	if (init_rootdomain(rd, false) != 0) {
+		kfree(rd);
+		return NULL;
+	}
 
 	return rd;
 }
@@ -6896,19 +7053,12 @@
 }
 
 /* cpus with isolated domains */
-static cpumask_t cpu_isolated_map = CPU_MASK_NONE;
+static cpumask_var_t cpu_isolated_map;
 
 /* Setup the mask of cpus configured for isolated domains */
 static int __init isolated_cpu_setup(char *str)
 {
-	static int __initdata ints[NR_CPUS];
-	int i;
-
-	str = get_options(str, ARRAY_SIZE(ints), ints);
-	cpus_clear(cpu_isolated_map);
-	for (i = 1; i <= ints[0]; i++)
-		if (ints[i] < NR_CPUS)
-			cpu_set(ints[i], cpu_isolated_map);
+	cpulist_parse(str, cpu_isolated_map);
 	return 1;
 }
 
@@ -6917,42 +7067,43 @@
 /*
  * init_sched_build_groups takes the cpumask we wish to span, and a pointer
  * to a function which identifies what group(along with sched group) a CPU
- * belongs to. The return value of group_fn must be a >= 0 and < NR_CPUS
- * (due to the fact that we keep track of groups covered with a cpumask_t).
+ * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids
+ * (due to the fact that we keep track of groups covered with a struct cpumask).
  *
  * init_sched_build_groups will build a circular linked list of the groups
  * covered by the given span, and will set each group's ->cpumask correctly,
  * and ->cpu_power to 0.
  */
 static void
-init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map,
-			int (*group_fn)(int cpu, const cpumask_t *cpu_map,
+init_sched_build_groups(const struct cpumask *span,
+			const struct cpumask *cpu_map,
+			int (*group_fn)(int cpu, const struct cpumask *cpu_map,
 					struct sched_group **sg,
-					cpumask_t *tmpmask),
-			cpumask_t *covered, cpumask_t *tmpmask)
+					struct cpumask *tmpmask),
+			struct cpumask *covered, struct cpumask *tmpmask)
 {
 	struct sched_group *first = NULL, *last = NULL;
 	int i;
 
-	cpus_clear(*covered);
+	cpumask_clear(covered);
 
-	for_each_cpu_mask_nr(i, *span) {
+	for_each_cpu(i, span) {
 		struct sched_group *sg;
 		int group = group_fn(i, cpu_map, &sg, tmpmask);
 		int j;
 
-		if (cpu_isset(i, *covered))
+		if (cpumask_test_cpu(i, covered))
 			continue;
 
-		cpus_clear(sg->cpumask);
+		cpumask_clear(sched_group_cpus(sg));
 		sg->__cpu_power = 0;
 
-		for_each_cpu_mask_nr(j, *span) {
+		for_each_cpu(j, span) {
 			if (group_fn(j, cpu_map, NULL, tmpmask) != group)
 				continue;
 
-			cpu_set(j, *covered);
-			cpu_set(j, sg->cpumask);
+			cpumask_set_cpu(j, covered);
+			cpumask_set_cpu(j, sched_group_cpus(sg));
 		}
 		if (!first)
 			first = sg;
@@ -7016,9 +7167,10 @@
  * should be one that prevents unnecessary balancing, but also spreads tasks
  * out optimally.
  */
-static void sched_domain_node_span(int node, cpumask_t *span)
+static void sched_domain_node_span(int node, struct cpumask *span)
 {
 	nodemask_t used_nodes;
+	/* FIXME: use cpumask_of_node() */
 	node_to_cpumask_ptr(nodemask, node);
 	int i;
 
@@ -7040,18 +7192,33 @@
 int sched_smt_power_savings = 0, sched_mc_power_savings = 0;
 
 /*
+ * The cpus mask in sched_group and sched_domain hangs off the end.
+ * FIXME: use cpumask_var_t or dynamic percpu alloc to avoid wasting space
+ * for nr_cpu_ids < CONFIG_NR_CPUS.
+ */
+struct static_sched_group {
+	struct sched_group sg;
+	DECLARE_BITMAP(cpus, CONFIG_NR_CPUS);
+};
+
+struct static_sched_domain {
+	struct sched_domain sd;
+	DECLARE_BITMAP(span, CONFIG_NR_CPUS);
+};
+
+/*
  * SMT sched-domains:
  */
 #ifdef CONFIG_SCHED_SMT
-static DEFINE_PER_CPU(struct sched_domain, cpu_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_cpus);
+static DEFINE_PER_CPU(struct static_sched_domain, cpu_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_cpus);
 
 static int
-cpu_to_cpu_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		 cpumask_t *unused)
+cpu_to_cpu_group(int cpu, const struct cpumask *cpu_map,
+		 struct sched_group **sg, struct cpumask *unused)
 {
 	if (sg)
-		*sg = &per_cpu(sched_group_cpus, cpu);
+		*sg = &per_cpu(sched_group_cpus, cpu).sg;
 	return cpu;
 }
 #endif /* CONFIG_SCHED_SMT */
@@ -7060,56 +7227,55 @@
  * multi-core sched-domains:
  */
 #ifdef CONFIG_SCHED_MC
-static DEFINE_PER_CPU(struct sched_domain, core_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_core);
+static DEFINE_PER_CPU(struct static_sched_domain, core_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_core);
 #endif /* CONFIG_SCHED_MC */
 
 #if defined(CONFIG_SCHED_MC) && defined(CONFIG_SCHED_SMT)
 static int
-cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		  cpumask_t *mask)
+cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *mask)
 {
 	int group;
 
-	*mask = per_cpu(cpu_sibling_map, cpu);
-	cpus_and(*mask, *mask, *cpu_map);
-	group = first_cpu(*mask);
+	cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map);
+	group = cpumask_first(mask);
 	if (sg)
-		*sg = &per_cpu(sched_group_core, group);
+		*sg = &per_cpu(sched_group_core, group).sg;
 	return group;
 }
 #elif defined(CONFIG_SCHED_MC)
 static int
-cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		  cpumask_t *unused)
+cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *unused)
 {
 	if (sg)
-		*sg = &per_cpu(sched_group_core, cpu);
+		*sg = &per_cpu(sched_group_core, cpu).sg;
 	return cpu;
 }
 #endif
 
-static DEFINE_PER_CPU(struct sched_domain, phys_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_phys);
+static DEFINE_PER_CPU(struct static_sched_domain, phys_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys);
 
 static int
-cpu_to_phys_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		  cpumask_t *mask)
+cpu_to_phys_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *mask)
 {
 	int group;
 #ifdef CONFIG_SCHED_MC
+	/* FIXME: Use cpu_coregroup_mask. */
 	*mask = cpu_coregroup_map(cpu);
 	cpus_and(*mask, *mask, *cpu_map);
-	group = first_cpu(*mask);
+	group = cpumask_first(mask);
 #elif defined(CONFIG_SCHED_SMT)
-	*mask = per_cpu(cpu_sibling_map, cpu);
-	cpus_and(*mask, *mask, *cpu_map);
-	group = first_cpu(*mask);
+	cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map);
+	group = cpumask_first(mask);
 #else
 	group = cpu;
 #endif
 	if (sg)
-		*sg = &per_cpu(sched_group_phys, group);
+		*sg = &per_cpu(sched_group_phys, group).sg;
 	return group;
 }
 
@@ -7123,19 +7289,21 @@
 static struct sched_group ***sched_group_nodes_bycpu;
 
 static DEFINE_PER_CPU(struct sched_domain, allnodes_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_allnodes);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
 
-static int cpu_to_allnodes_group(int cpu, const cpumask_t *cpu_map,
-				 struct sched_group **sg, cpumask_t *nodemask)
+static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
+				 struct sched_group **sg,
+				 struct cpumask *nodemask)
 {
 	int group;
+	/* FIXME: use cpumask_of_node */
+	node_to_cpumask_ptr(pnodemask, cpu_to_node(cpu));
 
-	*nodemask = node_to_cpumask(cpu_to_node(cpu));
-	cpus_and(*nodemask, *nodemask, *cpu_map);
-	group = first_cpu(*nodemask);
+	cpumask_and(nodemask, pnodemask, cpu_map);
+	group = cpumask_first(nodemask);
 
 	if (sg)
-		*sg = &per_cpu(sched_group_allnodes, group);
+		*sg = &per_cpu(sched_group_allnodes, group).sg;
 	return group;
 }
 
@@ -7147,11 +7315,11 @@
 	if (!sg)
 		return;
 	do {
-		for_each_cpu_mask_nr(j, sg->cpumask) {
+		for_each_cpu(j, sched_group_cpus(sg)) {
 			struct sched_domain *sd;
 
-			sd = &per_cpu(phys_domains, j);
-			if (j != first_cpu(sd->groups->cpumask)) {
+			sd = &per_cpu(phys_domains, j).sd;
+			if (j != cpumask_first(sched_group_cpus(sd->groups))) {
 				/*
 				 * Only add "power" once for each
 				 * physical package.
@@ -7168,11 +7336,12 @@
 
 #ifdef CONFIG_NUMA
 /* Free memory allocated for various sched_group structures */
-static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask)
+static void free_sched_groups(const struct cpumask *cpu_map,
+			      struct cpumask *nodemask)
 {
 	int cpu, i;
 
-	for_each_cpu_mask_nr(cpu, *cpu_map) {
+	for_each_cpu(cpu, cpu_map) {
 		struct sched_group **sched_group_nodes
 			= sched_group_nodes_bycpu[cpu];
 
@@ -7181,10 +7350,11 @@
 
 		for (i = 0; i < nr_node_ids; i++) {
 			struct sched_group *oldsg, *sg = sched_group_nodes[i];
+			/* FIXME: Use cpumask_of_node */
+			node_to_cpumask_ptr(pnodemask, i);
 
-			*nodemask = node_to_cpumask(i);
-			cpus_and(*nodemask, *nodemask, *cpu_map);
-			if (cpus_empty(*nodemask))
+			cpus_and(*nodemask, *pnodemask, *cpu_map);
+			if (cpumask_empty(nodemask))
 				continue;
 
 			if (sg == NULL)
@@ -7202,7 +7372,8 @@
 	}
 }
 #else /* !CONFIG_NUMA */
-static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask)
+static void free_sched_groups(const struct cpumask *cpu_map,
+			      struct cpumask *nodemask)
 {
 }
 #endif /* CONFIG_NUMA */
@@ -7228,7 +7399,7 @@
 
 	WARN_ON(!sd || !sd->groups);
 
-	if (cpu != first_cpu(sd->groups->cpumask))
+	if (cpu != cpumask_first(sched_group_cpus(sd->groups)))
 		return;
 
 	child = sd->child;
@@ -7293,48 +7464,6 @@
  SD_INIT_FUNC(MC)
 #endif
 
-/*
- * To minimize stack usage kmalloc room for cpumasks and share the
- * space as the usage in build_sched_domains() dictates.  Used only
- * if the amount of space is significant.
- */
-struct allmasks {
-	cpumask_t tmpmask;			/* make this one first */
-	union {
-		cpumask_t nodemask;
-		cpumask_t this_sibling_map;
-		cpumask_t this_core_map;
-	};
-	cpumask_t send_covered;
-
-#ifdef CONFIG_NUMA
-	cpumask_t domainspan;
-	cpumask_t covered;
-	cpumask_t notcovered;
-#endif
-};
-
-#if	NR_CPUS > 128
-#define SCHED_CPUMASK_DECLARE(v)	struct allmasks *v
-static inline void sched_cpumask_alloc(struct allmasks **masks)
-{
-	*masks = kmalloc(sizeof(**masks), GFP_KERNEL);
-}
-static inline void sched_cpumask_free(struct allmasks *masks)
-{
-	kfree(masks);
-}
-#else
-#define SCHED_CPUMASK_DECLARE(v)	struct allmasks _v, *v = &_v
-static inline void sched_cpumask_alloc(struct allmasks **masks)
-{ }
-static inline void sched_cpumask_free(struct allmasks *masks)
-{ }
-#endif
-
-#define	SCHED_CPUMASK_VAR(v, a) 	cpumask_t *v = (cpumask_t *) \
-			((unsigned long)(a) + offsetof(struct allmasks, v))
-
 static int default_relax_domain_level = -1;
 
 static int __init setup_relax_domain_level(char *str)
@@ -7374,17 +7503,38 @@
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
  */
-static int __build_sched_domains(const cpumask_t *cpu_map,
+static int __build_sched_domains(const struct cpumask *cpu_map,
 				 struct sched_domain_attr *attr)
 {
-	int i;
+	int i, err = -ENOMEM;
 	struct root_domain *rd;
-	SCHED_CPUMASK_DECLARE(allmasks);
-	cpumask_t *tmpmask;
+	cpumask_var_t nodemask, this_sibling_map, this_core_map, send_covered,
+		tmpmask;
 #ifdef CONFIG_NUMA
+	cpumask_var_t domainspan, covered, notcovered;
 	struct sched_group **sched_group_nodes = NULL;
 	int sd_allnodes = 0;
 
+	if (!alloc_cpumask_var(&domainspan, GFP_KERNEL))
+		goto out;
+	if (!alloc_cpumask_var(&covered, GFP_KERNEL))
+		goto free_domainspan;
+	if (!alloc_cpumask_var(&notcovered, GFP_KERNEL))
+		goto free_covered;
+#endif
+
+	if (!alloc_cpumask_var(&nodemask, GFP_KERNEL))
+		goto free_notcovered;
+	if (!alloc_cpumask_var(&this_sibling_map, GFP_KERNEL))
+		goto free_nodemask;
+	if (!alloc_cpumask_var(&this_core_map, GFP_KERNEL))
+		goto free_this_sibling_map;
+	if (!alloc_cpumask_var(&send_covered, GFP_KERNEL))
+		goto free_this_core_map;
+	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
+		goto free_send_covered;
+
+#ifdef CONFIG_NUMA
 	/*
 	 * Allocate the per-node list of sched groups
 	 */
@@ -7392,54 +7542,37 @@
 				    GFP_KERNEL);
 	if (!sched_group_nodes) {
 		printk(KERN_WARNING "Can not alloc sched group node list\n");
-		return -ENOMEM;
+		goto free_tmpmask;
 	}
 #endif
 
 	rd = alloc_rootdomain();
 	if (!rd) {
 		printk(KERN_WARNING "Cannot alloc root domain\n");
-#ifdef CONFIG_NUMA
-		kfree(sched_group_nodes);
-#endif
-		return -ENOMEM;
+		goto free_sched_groups;
 	}
 
-	/* get space for all scratch cpumask variables */
-	sched_cpumask_alloc(&allmasks);
-	if (!allmasks) {
-		printk(KERN_WARNING "Cannot alloc cpumask array\n");
-		kfree(rd);
 #ifdef CONFIG_NUMA
-		kfree(sched_group_nodes);
-#endif
-		return -ENOMEM;
-	}
-
-	tmpmask = (cpumask_t *)allmasks;
-
-
-#ifdef CONFIG_NUMA
-	sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes;
+	sched_group_nodes_bycpu[cpumask_first(cpu_map)] = sched_group_nodes;
 #endif
 
 	/*
 	 * Set up domains for cpus specified by the cpu_map.
 	 */
-	for_each_cpu_mask_nr(i, *cpu_map) {
+	for_each_cpu(i, cpu_map) {
 		struct sched_domain *sd = NULL, *p;
-		SCHED_CPUMASK_VAR(nodemask, allmasks);
 
+		/* FIXME: use cpumask_of_node */
 		*nodemask = node_to_cpumask(cpu_to_node(i));
 		cpus_and(*nodemask, *nodemask, *cpu_map);
 
 #ifdef CONFIG_NUMA
-		if (cpus_weight(*cpu_map) >
-				SD_NODES_PER_DOMAIN*cpus_weight(*nodemask)) {
+		if (cpumask_weight(cpu_map) >
+				SD_NODES_PER_DOMAIN*cpumask_weight(nodemask)) {
 			sd = &per_cpu(allnodes_domains, i);
 			SD_INIT(sd, ALLNODES);
 			set_domain_attribute(sd, attr);
-			sd->span = *cpu_map;
+			cpumask_copy(sched_domain_span(sd), cpu_map);
 			cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask);
 			p = sd;
 			sd_allnodes = 1;
@@ -7449,18 +7582,19 @@
 		sd = &per_cpu(node_domains, i);
 		SD_INIT(sd, NODE);
 		set_domain_attribute(sd, attr);
-		sched_domain_node_span(cpu_to_node(i), &sd->span);
+		sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
 		sd->parent = p;
 		if (p)
 			p->child = sd;
-		cpus_and(sd->span, sd->span, *cpu_map);
+		cpumask_and(sched_domain_span(sd),
+			    sched_domain_span(sd), cpu_map);
 #endif
 
 		p = sd;
-		sd = &per_cpu(phys_domains, i);
+		sd = &per_cpu(phys_domains, i).sd;
 		SD_INIT(sd, CPU);
 		set_domain_attribute(sd, attr);
-		sd->span = *nodemask;
+		cpumask_copy(sched_domain_span(sd), nodemask);
 		sd->parent = p;
 		if (p)
 			p->child = sd;
@@ -7468,11 +7602,12 @@
 
 #ifdef CONFIG_SCHED_MC
 		p = sd;
-		sd = &per_cpu(core_domains, i);
+		sd = &per_cpu(core_domains, i).sd;
 		SD_INIT(sd, MC);
 		set_domain_attribute(sd, attr);
-		sd->span = cpu_coregroup_map(i);
-		cpus_and(sd->span, sd->span, *cpu_map);
+		*sched_domain_span(sd) = cpu_coregroup_map(i);
+		cpumask_and(sched_domain_span(sd),
+			    sched_domain_span(sd), cpu_map);
 		sd->parent = p;
 		p->child = sd;
 		cpu_to_core_group(i, cpu_map, &sd->groups, tmpmask);
@@ -7480,11 +7615,11 @@
 
 #ifdef CONFIG_SCHED_SMT
 		p = sd;
-		sd = &per_cpu(cpu_domains, i);
+		sd = &per_cpu(cpu_domains, i).sd;
 		SD_INIT(sd, SIBLING);
 		set_domain_attribute(sd, attr);
-		sd->span = per_cpu(cpu_sibling_map, i);
-		cpus_and(sd->span, sd->span, *cpu_map);
+		cpumask_and(sched_domain_span(sd),
+			    &per_cpu(cpu_sibling_map, i), cpu_map);
 		sd->parent = p;
 		p->child = sd;
 		cpu_to_cpu_group(i, cpu_map, &sd->groups, tmpmask);
@@ -7493,13 +7628,10 @@
 
 #ifdef CONFIG_SCHED_SMT
 	/* Set up CPU (sibling) groups */
-	for_each_cpu_mask_nr(i, *cpu_map) {
-		SCHED_CPUMASK_VAR(this_sibling_map, allmasks);
-		SCHED_CPUMASK_VAR(send_covered, allmasks);
-
-		*this_sibling_map = per_cpu(cpu_sibling_map, i);
-		cpus_and(*this_sibling_map, *this_sibling_map, *cpu_map);
-		if (i != first_cpu(*this_sibling_map))
+	for_each_cpu(i, cpu_map) {
+		cpumask_and(this_sibling_map,
+			    &per_cpu(cpu_sibling_map, i), cpu_map);
+		if (i != cpumask_first(this_sibling_map))
 			continue;
 
 		init_sched_build_groups(this_sibling_map, cpu_map,
@@ -7510,13 +7642,11 @@
 
 #ifdef CONFIG_SCHED_MC
 	/* Set up multi-core groups */
-	for_each_cpu_mask_nr(i, *cpu_map) {
-		SCHED_CPUMASK_VAR(this_core_map, allmasks);
-		SCHED_CPUMASK_VAR(send_covered, allmasks);
-
+	for_each_cpu(i, cpu_map) {
+		/* FIXME: Use cpu_coregroup_mask */
 		*this_core_map = cpu_coregroup_map(i);
 		cpus_and(*this_core_map, *this_core_map, *cpu_map);
-		if (i != first_cpu(*this_core_map))
+		if (i != cpumask_first(this_core_map))
 			continue;
 
 		init_sched_build_groups(this_core_map, cpu_map,
@@ -7527,12 +7657,10 @@
 
 	/* Set up physical groups */
 	for (i = 0; i < nr_node_ids; i++) {
-		SCHED_CPUMASK_VAR(nodemask, allmasks);
-		SCHED_CPUMASK_VAR(send_covered, allmasks);
-
+		/* FIXME: Use cpumask_of_node */
 		*nodemask = node_to_cpumask(i);
 		cpus_and(*nodemask, *nodemask, *cpu_map);
-		if (cpus_empty(*nodemask))
+		if (cpumask_empty(nodemask))
 			continue;
 
 		init_sched_build_groups(nodemask, cpu_map,
@@ -7543,8 +7671,6 @@
 #ifdef CONFIG_NUMA
 	/* Set up node groups */
 	if (sd_allnodes) {
-		SCHED_CPUMASK_VAR(send_covered, allmasks);
-
 		init_sched_build_groups(cpu_map, cpu_map,
 					&cpu_to_allnodes_group,
 					send_covered, tmpmask);
@@ -7553,58 +7679,58 @@
 	for (i = 0; i < nr_node_ids; i++) {
 		/* Set up node groups */
 		struct sched_group *sg, *prev;
-		SCHED_CPUMASK_VAR(nodemask, allmasks);
-		SCHED_CPUMASK_VAR(domainspan, allmasks);
-		SCHED_CPUMASK_VAR(covered, allmasks);
 		int j;
 
+		/* FIXME: Use cpumask_of_node */
 		*nodemask = node_to_cpumask(i);
-		cpus_clear(*covered);
+		cpumask_clear(covered);
 
 		cpus_and(*nodemask, *nodemask, *cpu_map);
-		if (cpus_empty(*nodemask)) {
+		if (cpumask_empty(nodemask)) {
 			sched_group_nodes[i] = NULL;
 			continue;
 		}
 
 		sched_domain_node_span(i, domainspan);
-		cpus_and(*domainspan, *domainspan, *cpu_map);
+		cpumask_and(domainspan, domainspan, cpu_map);
 
-		sg = kmalloc_node(sizeof(struct sched_group), GFP_KERNEL, i);
+		sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(),
+				  GFP_KERNEL, i);
 		if (!sg) {
 			printk(KERN_WARNING "Can not alloc domain group for "
 				"node %d\n", i);
 			goto error;
 		}
 		sched_group_nodes[i] = sg;
-		for_each_cpu_mask_nr(j, *nodemask) {
+		for_each_cpu(j, nodemask) {
 			struct sched_domain *sd;
 
 			sd = &per_cpu(node_domains, j);
 			sd->groups = sg;
 		}
 		sg->__cpu_power = 0;
-		sg->cpumask = *nodemask;
+		cpumask_copy(sched_group_cpus(sg), nodemask);
 		sg->next = sg;
-		cpus_or(*covered, *covered, *nodemask);
+		cpumask_or(covered, covered, nodemask);
 		prev = sg;
 
 		for (j = 0; j < nr_node_ids; j++) {
-			SCHED_CPUMASK_VAR(notcovered, allmasks);
 			int n = (i + j) % nr_node_ids;
+			/* FIXME: Use cpumask_of_node */
 			node_to_cpumask_ptr(pnodemask, n);
 
-			cpus_complement(*notcovered, *covered);
-			cpus_and(*tmpmask, *notcovered, *cpu_map);
-			cpus_and(*tmpmask, *tmpmask, *domainspan);
-			if (cpus_empty(*tmpmask))
+			cpumask_complement(notcovered, covered);
+			cpumask_and(tmpmask, notcovered, cpu_map);
+			cpumask_and(tmpmask, tmpmask, domainspan);
+			if (cpumask_empty(tmpmask))
 				break;
 
-			cpus_and(*tmpmask, *tmpmask, *pnodemask);
-			if (cpus_empty(*tmpmask))
+			cpumask_and(tmpmask, tmpmask, pnodemask);
+			if (cpumask_empty(tmpmask))
 				continue;
 
-			sg = kmalloc_node(sizeof(struct sched_group),
+			sg = kmalloc_node(sizeof(struct sched_group) +
+					  cpumask_size(),
 					  GFP_KERNEL, i);
 			if (!sg) {
 				printk(KERN_WARNING
@@ -7612,9 +7738,9 @@
 				goto error;
 			}
 			sg->__cpu_power = 0;
-			sg->cpumask = *tmpmask;
+			cpumask_copy(sched_group_cpus(sg), tmpmask);
 			sg->next = prev->next;
-			cpus_or(*covered, *covered, *tmpmask);
+			cpumask_or(covered, covered, tmpmask);
 			prev->next = sg;
 			prev = sg;
 		}
@@ -7623,22 +7749,22 @@
 
 	/* Calculate CPU power for physical packages and nodes */
 #ifdef CONFIG_SCHED_SMT
-	for_each_cpu_mask_nr(i, *cpu_map) {
-		struct sched_domain *sd = &per_cpu(cpu_domains, i);
+	for_each_cpu(i, cpu_map) {
+		struct sched_domain *sd = &per_cpu(cpu_domains, i).sd;
 
 		init_sched_groups_power(i, sd);
 	}
 #endif
 #ifdef CONFIG_SCHED_MC
-	for_each_cpu_mask_nr(i, *cpu_map) {
-		struct sched_domain *sd = &per_cpu(core_domains, i);
+	for_each_cpu(i, cpu_map) {
+		struct sched_domain *sd = &per_cpu(core_domains, i).sd;
 
 		init_sched_groups_power(i, sd);
 	}
 #endif
 
-	for_each_cpu_mask_nr(i, *cpu_map) {
-		struct sched_domain *sd = &per_cpu(phys_domains, i);
+	for_each_cpu(i, cpu_map) {
+		struct sched_domain *sd = &per_cpu(phys_domains, i).sd;
 
 		init_sched_groups_power(i, sd);
 	}
@@ -7650,53 +7776,78 @@
 	if (sd_allnodes) {
 		struct sched_group *sg;
 
-		cpu_to_allnodes_group(first_cpu(*cpu_map), cpu_map, &sg,
+		cpu_to_allnodes_group(cpumask_first(cpu_map), cpu_map, &sg,
 								tmpmask);
 		init_numa_sched_groups_power(sg);
 	}
 #endif
 
 	/* Attach the domains */
-	for_each_cpu_mask_nr(i, *cpu_map) {
+	for_each_cpu(i, cpu_map) {
 		struct sched_domain *sd;
 #ifdef CONFIG_SCHED_SMT
-		sd = &per_cpu(cpu_domains, i);
+		sd = &per_cpu(cpu_domains, i).sd;
 #elif defined(CONFIG_SCHED_MC)
-		sd = &per_cpu(core_domains, i);
+		sd = &per_cpu(core_domains, i).sd;
 #else
-		sd = &per_cpu(phys_domains, i);
+		sd = &per_cpu(phys_domains, i).sd;
 #endif
 		cpu_attach_domain(sd, rd, i);
 	}
 
-	sched_cpumask_free(allmasks);
-	return 0;
+	err = 0;
+
+free_tmpmask:
+	free_cpumask_var(tmpmask);
+free_send_covered:
+	free_cpumask_var(send_covered);
+free_this_core_map:
+	free_cpumask_var(this_core_map);
+free_this_sibling_map:
+	free_cpumask_var(this_sibling_map);
+free_nodemask:
+	free_cpumask_var(nodemask);
+free_notcovered:
+#ifdef CONFIG_NUMA
+	free_cpumask_var(notcovered);
+free_covered:
+	free_cpumask_var(covered);
+free_domainspan:
+	free_cpumask_var(domainspan);
+out:
+#endif
+	return err;
+
+free_sched_groups:
+#ifdef CONFIG_NUMA
+	kfree(sched_group_nodes);
+#endif
+	goto free_tmpmask;
 
 #ifdef CONFIG_NUMA
 error:
 	free_sched_groups(cpu_map, tmpmask);
-	sched_cpumask_free(allmasks);
-	kfree(rd);
-	return -ENOMEM;
+	free_rootdomain(rd);
+	goto free_tmpmask;
 #endif
 }
 
-static int build_sched_domains(const cpumask_t *cpu_map)
+static int build_sched_domains(const struct cpumask *cpu_map)
 {
 	return __build_sched_domains(cpu_map, NULL);
 }
 
-static cpumask_t *doms_cur;	/* current sched domains */
+static struct cpumask *doms_cur;	/* current sched domains */
 static int ndoms_cur;		/* number of sched domains in 'doms_cur' */
 static struct sched_domain_attr *dattr_cur;
 				/* attribues of custom domains in 'doms_cur' */
 
 /*
  * Special case: If a kmalloc of a doms_cur partition (array of
- * cpumask_t) fails, then fallback to a single sched domain,
- * as determined by the single cpumask_t fallback_doms.
+ * cpumask) fails, then fallback to a single sched domain,
+ * as determined by the single cpumask fallback_doms.
  */
-static cpumask_t fallback_doms;
+static cpumask_var_t fallback_doms;
 
 /*
  * arch_update_cpu_topology lets virtualized architectures update the
@@ -7713,16 +7864,16 @@
  * For now this just excludes isolated cpus, but could be used to
  * exclude other special cases in the future.
  */
-static int arch_init_sched_domains(const cpumask_t *cpu_map)
+static int arch_init_sched_domains(const struct cpumask *cpu_map)
 {
 	int err;
 
 	arch_update_cpu_topology();
 	ndoms_cur = 1;
-	doms_cur = kmalloc(sizeof(cpumask_t), GFP_KERNEL);
+	doms_cur = kmalloc(cpumask_size(), GFP_KERNEL);
 	if (!doms_cur)
-		doms_cur = &fallback_doms;
-	cpus_andnot(*doms_cur, *cpu_map, cpu_isolated_map);
+		doms_cur = fallback_doms;
+	cpumask_andnot(doms_cur, cpu_map, cpu_isolated_map);
 	dattr_cur = NULL;
 	err = build_sched_domains(doms_cur);
 	register_sched_domain_sysctl();
@@ -7730,8 +7881,8 @@
 	return err;
 }
 
-static void arch_destroy_sched_domains(const cpumask_t *cpu_map,
-				       cpumask_t *tmpmask)
+static void arch_destroy_sched_domains(const struct cpumask *cpu_map,
+				       struct cpumask *tmpmask)
 {
 	free_sched_groups(cpu_map, tmpmask);
 }
@@ -7740,15 +7891,16 @@
  * Detach sched domains from a group of cpus specified in cpu_map
  * These cpus will now be attached to the NULL domain
  */
-static void detach_destroy_domains(const cpumask_t *cpu_map)
+static void detach_destroy_domains(const struct cpumask *cpu_map)
 {
-	cpumask_t tmpmask;
+	/* Save because hotplug lock held. */
+	static DECLARE_BITMAP(tmpmask, CONFIG_NR_CPUS);
 	int i;
 
-	for_each_cpu_mask_nr(i, *cpu_map)
+	for_each_cpu(i, cpu_map)
 		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
-	arch_destroy_sched_domains(cpu_map, &tmpmask);
+	arch_destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
 }
 
 /* handle null as "default" */
@@ -7773,7 +7925,7 @@
  * doms_new[] to the current sched domain partitioning, doms_cur[].
  * It destroys each deleted domain and builds each new domain.
  *
- * 'doms_new' is an array of cpumask_t's of length 'ndoms_new'.
+ * 'doms_new' is an array of cpumask's of length 'ndoms_new'.
  * The masks don't intersect (don't overlap.) We should setup one
  * sched domain for each mask. CPUs not in any of the cpumasks will
  * not be load balanced. If the same cpumask appears both in the
@@ -7787,13 +7939,14 @@
  * the single partition 'fallback_doms', it also forces the domains
  * to be rebuilt.
  *
- * If doms_new == NULL it will be replaced with cpu_online_map.
+ * If doms_new == NULL it will be replaced with cpu_online_mask.
  * ndoms_new == 0 is a special case for destroying existing domains,
  * and it will not create the default domain.
  *
  * Call with hotplug lock held
  */
-void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+/* FIXME: Change to struct cpumask *doms_new[] */
+void partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 			     struct sched_domain_attr *dattr_new)
 {
 	int i, j, n;
@@ -7812,7 +7965,7 @@
 	/* Destroy deleted domains */
 	for (i = 0; i < ndoms_cur; i++) {
 		for (j = 0; j < n && !new_topology; j++) {
-			if (cpus_equal(doms_cur[i], doms_new[j])
+			if (cpumask_equal(&doms_cur[i], &doms_new[j])
 			    && dattrs_equal(dattr_cur, i, dattr_new, j))
 				goto match1;
 		}
@@ -7824,15 +7977,15 @@
 
 	if (doms_new == NULL) {
 		ndoms_cur = 0;
-		doms_new = &fallback_doms;
-		cpus_andnot(doms_new[0], cpu_online_map, cpu_isolated_map);
+		doms_new = fallback_doms;
+		cpumask_andnot(&doms_new[0], cpu_online_mask, cpu_isolated_map);
 		WARN_ON_ONCE(dattr_new);
 	}
 
 	/* Build new domains */
 	for (i = 0; i < ndoms_new; i++) {
 		for (j = 0; j < ndoms_cur && !new_topology; j++) {
-			if (cpus_equal(doms_new[i], doms_cur[j])
+			if (cpumask_equal(&doms_new[i], &doms_cur[j])
 			    && dattrs_equal(dattr_new, i, dattr_cur, j))
 				goto match2;
 		}
@@ -7844,7 +7997,7 @@
 	}
 
 	/* Remember the new sched domains */
-	if (doms_cur != &fallback_doms)
+	if (doms_cur != fallback_doms)
 		kfree(doms_cur);
 	kfree(dattr_cur);	/* kfree(NULL) is safe */
 	doms_cur = doms_new;
@@ -7873,14 +8026,25 @@
 static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
 {
 	int ret;
+	unsigned int level = 0;
 
-	if (buf[0] != '0' && buf[0] != '1')
+	if (sscanf(buf, "%u", &level) != 1)
+		return -EINVAL;
+
+	/*
+	 * level is always be positive so don't check for
+	 * level < POWERSAVINGS_BALANCE_NONE which is 0
+	 * What happens on 0 or 1 byte write,
+	 * need to check for count as well?
+	 */
+
+	if (level >= MAX_POWERSAVINGS_BALANCE_LEVELS)
 		return -EINVAL;
 
 	if (smt)
-		sched_smt_power_savings = (buf[0] == '1');
+		sched_smt_power_savings = level;
 	else
-		sched_mc_power_savings = (buf[0] == '1');
+		sched_mc_power_savings = level;
 
 	ret = arch_reinit_sched_domains();
 
@@ -7984,7 +8148,9 @@
 
 void __init sched_init_smp(void)
 {
-	cpumask_t non_isolated_cpus;
+	cpumask_var_t non_isolated_cpus;
+
+	alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL);
 
 #if defined(CONFIG_NUMA)
 	sched_group_nodes_bycpu = kzalloc(nr_cpu_ids * sizeof(void **),
@@ -7993,10 +8159,10 @@
 #endif
 	get_online_cpus();
 	mutex_lock(&sched_domains_mutex);
-	arch_init_sched_domains(&cpu_online_map);
-	cpus_andnot(non_isolated_cpus, cpu_possible_map, cpu_isolated_map);
-	if (cpus_empty(non_isolated_cpus))
-		cpu_set(smp_processor_id(), non_isolated_cpus);
+	arch_init_sched_domains(cpu_online_mask);
+	cpumask_andnot(non_isolated_cpus, cpu_possible_mask, cpu_isolated_map);
+	if (cpumask_empty(non_isolated_cpus))
+		cpumask_set_cpu(smp_processor_id(), non_isolated_cpus);
 	mutex_unlock(&sched_domains_mutex);
 	put_online_cpus();
 
@@ -8011,9 +8177,13 @@
 	init_hrtick();
 
 	/* Move init over to a non-isolated CPU */
-	if (set_cpus_allowed_ptr(current, &non_isolated_cpus) < 0)
+	if (set_cpus_allowed_ptr(current, non_isolated_cpus) < 0)
 		BUG();
 	sched_init_granularity();
+	free_cpumask_var(non_isolated_cpus);
+
+	alloc_cpumask_var(&fallback_doms, GFP_KERNEL);
+	init_sched_rt_class();
 }
 #else
 void __init sched_init_smp(void)
@@ -8328,6 +8498,15 @@
 	 */
 	current->sched_class = &fair_sched_class;
 
+	/* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */
+	alloc_bootmem_cpumask_var(&nohz_cpu_mask);
+#ifdef CONFIG_SMP
+#ifdef CONFIG_NO_HZ
+	alloc_bootmem_cpumask_var(&nohz.cpu_mask);
+#endif
+	alloc_bootmem_cpumask_var(&cpu_isolated_map);
+#endif /* SMP */
+
 	scheduler_running = 1;
 }
 
diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c
index 52154fe..018b7be 100644
--- a/kernel/sched_cpupri.c
+++ b/kernel/sched_cpupri.c
@@ -67,24 +67,21 @@
  * Returns: (int)bool - CPUs were found
  */
 int cpupri_find(struct cpupri *cp, struct task_struct *p,
-		cpumask_t *lowest_mask)
+		struct cpumask *lowest_mask)
 {
 	int                  idx      = 0;
 	int                  task_pri = convert_prio(p->prio);
 
 	for_each_cpupri_active(cp->pri_active, idx) {
 		struct cpupri_vec *vec  = &cp->pri_to_cpu[idx];
-		cpumask_t mask;
 
 		if (idx >= task_pri)
 			break;
 
-		cpus_and(mask, p->cpus_allowed, vec->mask);
-
-		if (cpus_empty(mask))
+		if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids)
 			continue;
 
-		*lowest_mask = mask;
+		cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask);
 		return 1;
 	}
 
@@ -126,7 +123,7 @@
 		vec->count--;
 		if (!vec->count)
 			clear_bit(oldpri, cp->pri_active);
-		cpu_clear(cpu, vec->mask);
+		cpumask_clear_cpu(cpu, vec->mask);
 
 		spin_unlock_irqrestore(&vec->lock, flags);
 	}
@@ -136,7 +133,7 @@
 
 		spin_lock_irqsave(&vec->lock, flags);
 
-		cpu_set(cpu, vec->mask);
+		cpumask_set_cpu(cpu, vec->mask);
 		vec->count++;
 		if (vec->count == 1)
 			set_bit(newpri, cp->pri_active);
@@ -150,10 +147,11 @@
 /**
  * cpupri_init - initialize the cpupri structure
  * @cp: The cpupri context
+ * @bootmem: true if allocations need to use bootmem
  *
- * Returns: (void)
+ * Returns: -ENOMEM if memory fails.
  */
-void cpupri_init(struct cpupri *cp)
+int cpupri_init(struct cpupri *cp, bool bootmem)
 {
 	int i;
 
@@ -164,11 +162,30 @@
 
 		spin_lock_init(&vec->lock);
 		vec->count = 0;
-		cpus_clear(vec->mask);
+		if (bootmem)
+			alloc_bootmem_cpumask_var(&vec->mask);
+		else if (!alloc_cpumask_var(&vec->mask, GFP_KERNEL))
+			goto cleanup;
 	}
 
 	for_each_possible_cpu(i)
 		cp->cpu_to_pri[i] = CPUPRI_INVALID;
+	return 0;
+
+cleanup:
+	for (i--; i >= 0; i--)
+		free_cpumask_var(cp->pri_to_cpu[i].mask);
+	return -ENOMEM;
 }
 
+/**
+ * cpupri_cleanup - clean up the cpupri structure
+ * @cp: The cpupri context
+ */
+void cpupri_cleanup(struct cpupri *cp)
+{
+	int i;
 
+	for (i = 0; i < CPUPRI_NR_PRIORITIES; i++)
+		free_cpumask_var(cp->pri_to_cpu[i].mask);
+}
diff --git a/kernel/sched_cpupri.h b/kernel/sched_cpupri.h
index f25811b0..642a94e 100644
--- a/kernel/sched_cpupri.h
+++ b/kernel/sched_cpupri.h
@@ -14,7 +14,7 @@
 struct cpupri_vec {
 	spinlock_t lock;
 	int        count;
-	cpumask_t  mask;
+	cpumask_var_t mask;
 };
 
 struct cpupri {
@@ -27,7 +27,8 @@
 int  cpupri_find(struct cpupri *cp,
 		 struct task_struct *p, cpumask_t *lowest_mask);
 void cpupri_set(struct cpupri *cp, int cpu, int pri);
-void cpupri_init(struct cpupri *cp);
+int cpupri_init(struct cpupri *cp, bool bootmem);
+void cpupri_cleanup(struct cpupri *cp);
 #else
 #define cpupri_set(cp, cpu, pri) do { } while (0)
 #define cpupri_init() do { } while (0)
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 5ad4440..56c0efe 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1019,16 +1019,33 @@
  * search starts with cpus closest then further out as needed,
  * so we always favor a closer, idle cpu.
  * Domains may include CPUs that are not usable for migration,
- * hence we need to mask them out (cpu_active_map)
+ * hence we need to mask them out (cpu_active_mask)
  *
  * Returns the CPU we should wake onto.
  */
 #if defined(ARCH_HAS_SCHED_WAKE_IDLE)
 static int wake_idle(int cpu, struct task_struct *p)
 {
-	cpumask_t tmp;
 	struct sched_domain *sd;
 	int i;
+	unsigned int chosen_wakeup_cpu;
+	int this_cpu;
+
+	/*
+	 * At POWERSAVINGS_BALANCE_WAKEUP level, if both this_cpu and prev_cpu
+	 * are idle and this is not a kernel thread and this task's affinity
+	 * allows it to be moved to preferred cpu, then just move!
+	 */
+
+	this_cpu = smp_processor_id();
+	chosen_wakeup_cpu =
+		cpu_rq(this_cpu)->rd->sched_mc_preferred_wakeup_cpu;
+
+	if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP &&
+		idle_cpu(cpu) && idle_cpu(this_cpu) &&
+		p->mm && !(p->flags & PF_KTHREAD) &&
+		cpu_isset(chosen_wakeup_cpu, p->cpus_allowed))
+		return chosen_wakeup_cpu;
 
 	/*
 	 * If it is idle, then it is the best cpu to run this task.
@@ -1046,10 +1063,9 @@
 		if ((sd->flags & SD_WAKE_IDLE)
 		    || ((sd->flags & SD_WAKE_IDLE_FAR)
 			&& !task_hot(p, task_rq(p)->clock, sd))) {
-			cpus_and(tmp, sd->span, p->cpus_allowed);
-			cpus_and(tmp, tmp, cpu_active_map);
-			for_each_cpu_mask_nr(i, tmp) {
-				if (idle_cpu(i)) {
+			for_each_cpu_and(i, sched_domain_span(sd),
+					 &p->cpus_allowed) {
+				if (cpu_active(i) && idle_cpu(i)) {
 					if (i != task_cpu(p)) {
 						schedstat_inc(p,
 						       se.nr_wakeups_idle);
@@ -1242,13 +1258,13 @@
 	 * this_cpu and prev_cpu are present in:
 	 */
 	for_each_domain(this_cpu, sd) {
-		if (cpu_isset(prev_cpu, sd->span)) {
+		if (cpumask_test_cpu(prev_cpu, sched_domain_span(sd))) {
 			this_sd = sd;
 			break;
 		}
 	}
 
-	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
+	if (unlikely(!cpumask_test_cpu(this_cpu, &p->cpus_allowed)))
 		goto out;
 
 	/*
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 51d2af3..833b6d4 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -15,7 +15,7 @@
 	if (!rq->online)
 		return;
 
-	cpu_set(rq->cpu, rq->rd->rto_mask);
+	cpumask_set_cpu(rq->cpu, rq->rd->rto_mask);
 	/*
 	 * Make sure the mask is visible before we set
 	 * the overload count. That is checked to determine
@@ -34,7 +34,7 @@
 
 	/* the order here really doesn't matter */
 	atomic_dec(&rq->rd->rto_count);
-	cpu_clear(rq->cpu, rq->rd->rto_mask);
+	cpumask_clear_cpu(rq->cpu, rq->rd->rto_mask);
 }
 
 static void update_rt_migration(struct rq *rq)
@@ -139,14 +139,14 @@
 }
 
 #ifdef CONFIG_SMP
-static inline cpumask_t sched_rt_period_mask(void)
+static inline const struct cpumask *sched_rt_period_mask(void)
 {
 	return cpu_rq(smp_processor_id())->rd->span;
 }
 #else
-static inline cpumask_t sched_rt_period_mask(void)
+static inline const struct cpumask *sched_rt_period_mask(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 #endif
 
@@ -212,9 +212,9 @@
 	return rt_rq->rt_throttled;
 }
 
-static inline cpumask_t sched_rt_period_mask(void)
+static inline const struct cpumask *sched_rt_period_mask(void)
 {
-	return cpu_online_map;
+	return cpu_online_mask;
 }
 
 static inline
@@ -241,11 +241,11 @@
 	int i, weight, more = 0;
 	u64 rt_period;
 
-	weight = cpus_weight(rd->span);
+	weight = cpumask_weight(rd->span);
 
 	spin_lock(&rt_b->rt_runtime_lock);
 	rt_period = ktime_to_ns(rt_b->rt_period);
-	for_each_cpu_mask_nr(i, rd->span) {
+	for_each_cpu(i, rd->span) {
 		struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i);
 		s64 diff;
 
@@ -324,7 +324,7 @@
 		/*
 		 * Greedy reclaim, take back as much as we can.
 		 */
-		for_each_cpu_mask(i, rd->span) {
+		for_each_cpu(i, rd->span) {
 			struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i);
 			s64 diff;
 
@@ -429,13 +429,13 @@
 static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 {
 	int i, idle = 1;
-	cpumask_t span;
+	const struct cpumask *span;
 
 	if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF)
 		return 1;
 
 	span = sched_rt_period_mask();
-	for_each_cpu_mask(i, span) {
+	for_each_cpu(i, span) {
 		int enqueue = 0;
 		struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i);
 		struct rq *rq = rq_of_rt_rq(rt_rq);
@@ -805,17 +805,20 @@
 
 static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p)
 {
-	cpumask_t mask;
+	cpumask_var_t mask;
 
 	if (rq->curr->rt.nr_cpus_allowed == 1)
 		return;
 
-	if (p->rt.nr_cpus_allowed != 1
-	    && cpupri_find(&rq->rd->cpupri, p, &mask))
+	if (!alloc_cpumask_var(&mask, GFP_ATOMIC))
 		return;
 
-	if (!cpupri_find(&rq->rd->cpupri, rq->curr, &mask))
-		return;
+	if (p->rt.nr_cpus_allowed != 1
+	    && cpupri_find(&rq->rd->cpupri, p, mask))
+		goto free;
+
+	if (!cpupri_find(&rq->rd->cpupri, rq->curr, mask))
+		goto free;
 
 	/*
 	 * There appears to be other cpus that can accept
@@ -824,6 +827,8 @@
 	 */
 	requeue_task_rt(rq, p, 1);
 	resched_task(rq->curr);
+free:
+	free_cpumask_var(mask);
 }
 
 #endif /* CONFIG_SMP */
@@ -914,7 +919,7 @@
 static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
 	if (!task_running(rq, p) &&
-	    (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)) &&
+	    (cpu < 0 || cpumask_test_cpu(cpu, &p->cpus_allowed)) &&
 	    (p->rt.nr_cpus_allowed > 1))
 		return 1;
 	return 0;
@@ -953,7 +958,7 @@
 	return next;
 }
 
-static DEFINE_PER_CPU(cpumask_t, local_cpu_mask);
+static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask);
 
 static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask)
 {
@@ -973,7 +978,7 @@
 static int find_lowest_rq(struct task_struct *task)
 {
 	struct sched_domain *sd;
-	cpumask_t *lowest_mask = &__get_cpu_var(local_cpu_mask);
+	struct cpumask *lowest_mask = __get_cpu_var(local_cpu_mask);
 	int this_cpu = smp_processor_id();
 	int cpu      = task_cpu(task);
 
@@ -988,7 +993,7 @@
 	 * I guess we might want to change cpupri_find() to ignore those
 	 * in the first place.
 	 */
-	cpus_and(*lowest_mask, *lowest_mask, cpu_active_map);
+	cpumask_and(lowest_mask, lowest_mask, cpu_active_mask);
 
 	/*
 	 * At this point we have built a mask of cpus representing the
@@ -998,7 +1003,7 @@
 	 * We prioritize the last cpu that the task executed on since
 	 * it is most likely cache-hot in that location.
 	 */
-	if (cpu_isset(cpu, *lowest_mask))
+	if (cpumask_test_cpu(cpu, lowest_mask))
 		return cpu;
 
 	/*
@@ -1013,7 +1018,8 @@
 			cpumask_t domain_mask;
 			int       best_cpu;
 
-			cpus_and(domain_mask, sd->span, *lowest_mask);
+			cpumask_and(&domain_mask, sched_domain_span(sd),
+				    lowest_mask);
 
 			best_cpu = pick_optimal_cpu(this_cpu,
 						    &domain_mask);
@@ -1054,8 +1060,8 @@
 			 * Also make sure that it wasn't scheduled on its rq.
 			 */
 			if (unlikely(task_rq(task) != rq ||
-				     !cpu_isset(lowest_rq->cpu,
-						task->cpus_allowed) ||
+				     !cpumask_test_cpu(lowest_rq->cpu,
+						       &task->cpus_allowed) ||
 				     task_running(rq, task) ||
 				     !task->se.on_rq)) {
 
@@ -1176,7 +1182,7 @@
 
 	next = pick_next_task_rt(this_rq);
 
-	for_each_cpu_mask_nr(cpu, this_rq->rd->rto_mask) {
+	for_each_cpu(cpu, this_rq->rd->rto_mask) {
 		if (this_cpu == cpu)
 			continue;
 
@@ -1305,9 +1311,9 @@
 }
 
 static void set_cpus_allowed_rt(struct task_struct *p,
-				const cpumask_t *new_mask)
+				const struct cpumask *new_mask)
 {
-	int weight = cpus_weight(*new_mask);
+	int weight = cpumask_weight(new_mask);
 
 	BUG_ON(!rt_task(p));
 
@@ -1328,7 +1334,7 @@
 		update_rt_migration(rq);
 	}
 
-	p->cpus_allowed    = *new_mask;
+	cpumask_copy(&p->cpus_allowed, new_mask);
 	p->rt.nr_cpus_allowed = weight;
 }
 
@@ -1371,6 +1377,14 @@
 	if (!rq->rt.rt_nr_running)
 		pull_rt_task(rq);
 }
+
+static inline void init_sched_rt_class(void)
+{
+	unsigned int i;
+
+	for_each_possible_cpu(i)
+		alloc_cpumask_var(&per_cpu(local_cpu_mask, i), GFP_KERNEL);
+}
 #endif /* CONFIG_SMP */
 
 /*
@@ -1541,3 +1555,4 @@
 	rcu_read_unlock();
 }
 #endif /* CONFIG_SCHED_DEBUG */
+
diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h
index 3b01098..f2773b5 100644
--- a/kernel/sched_stats.h
+++ b/kernel/sched_stats.h
@@ -42,7 +42,8 @@
 		for_each_domain(cpu, sd) {
 			enum cpu_idle_type itype;
 
-			cpumask_scnprintf(mask_str, mask_len, sd->span);
+			cpumask_scnprintf(mask_str, mask_len,
+					  sched_domain_span(sd));
 			seq_printf(seq, "domain%d %s", dcount++, mask_str);
 			for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES;
 					itype++) {
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 466e75c..670c1ec 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -784,3 +784,23 @@
 }
 EXPORT_SYMBOL(on_each_cpu);
 #endif
+
+/*
+ * [ These __weak aliases are kept in a separate compilation unit, so that
+ *   GCC does not inline them incorrectly. ]
+ */
+
+int __init __weak early_irq_init(void)
+{
+	return 0;
+}
+
+int __init __weak arch_early_irq_init(void)
+{
+	return 0;
+}
+
+int __weak arch_init_chip_data(struct irq_desc *desc, int cpu)
+{
+	return 0;
+}
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index bd6be76..6d7dc4e 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -352,7 +352,7 @@
 	if (!data)
 		return -ENOMEM;
 	nla_strlcpy(data, na, len);
-	ret = cpulist_parse(data, *mask);
+	ret = cpulist_parse(data, mask);
 	kfree(data);
 	return ret;
 }
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index f8d9680..ea2f48a 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -166,6 +166,8 @@
 void clockevents_register_device(struct clock_event_device *dev)
 {
 	BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
+	BUG_ON(!dev->cpumask);
+
 	/*
 	 * A nsec2cyc multiplicator of 0 is invalid and we'd crash
 	 * on it, so fix it up and emit a warning:
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index f98a1b7..9590af2 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -150,7 +150,7 @@
 		 */
 		cpu = first_cpu(mask);
 		td = &per_cpu(tick_cpu_device, cpu);
-		td->evtdev->broadcast(mask);
+		td->evtdev->broadcast(&mask);
 	}
 }
 
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index df12434..f8372be 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -136,7 +136,7 @@
  */
 static void tick_setup_device(struct tick_device *td,
 			      struct clock_event_device *newdev, int cpu,
-			      const cpumask_t *cpumask)
+			      const struct cpumask *cpumask)
 {
 	ktime_t next_event;
 	void (*handler)(struct clock_event_device *) = NULL;
@@ -171,8 +171,8 @@
 	 * When the device is not per cpu, pin the interrupt to the
 	 * current cpu:
 	 */
-	if (!cpus_equal(newdev->cpumask, *cpumask))
-		irq_set_affinity(newdev->irq, *cpumask);
+	if (!cpumask_equal(newdev->cpumask, cpumask))
+		irq_set_affinity(newdev->irq, cpumask);
 
 	/*
 	 * When global broadcasting is active, check if the current
@@ -202,14 +202,14 @@
 	spin_lock_irqsave(&tick_device_lock, flags);
 
 	cpu = smp_processor_id();
-	if (!cpu_isset(cpu, newdev->cpumask))
+	if (!cpumask_test_cpu(cpu, newdev->cpumask))
 		goto out_bc;
 
 	td = &per_cpu(tick_cpu_device, cpu);
 	curdev = td->evtdev;
 
 	/* cpu local device ? */
-	if (!cpus_equal(newdev->cpumask, cpumask_of_cpu(cpu))) {
+	if (!cpumask_equal(newdev->cpumask, cpumask_of(cpu))) {
 
 		/*
 		 * If the cpu affinity of the device interrupt can not
@@ -222,7 +222,7 @@
 		 * If we have a cpu local device already, do not replace it
 		 * by a non cpu local device
 		 */
-		if (curdev && cpus_equal(curdev->cpumask, cpumask_of_cpu(cpu)))
+		if (curdev && cpumask_equal(curdev->cpumask, cpumask_of(cpu)))
 			goto out_bc;
 	}
 
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 8f3fc25..1b6c05b 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -144,7 +144,7 @@
 	if (!ts->tick_stopped)
 		return;
 
-	cpu_clear(cpu, nohz_cpu_mask);
+	cpumask_clear_cpu(cpu, nohz_cpu_mask);
 	now = ktime_get();
 	ts->idle_waketime = now;
 
@@ -301,7 +301,7 @@
 			tick_do_timer_cpu = TICK_DO_TIMER_NONE;
 
 		if (delta_jiffies > 1)
-			cpu_set(cpu, nohz_cpu_mask);
+			cpumask_set_cpu(cpu, nohz_cpu_mask);
 
 		/* Skip reprogram of event if its not changed */
 		if (ts->tick_stopped && ktime_equal(expires, dev->next_event))
@@ -319,7 +319,7 @@
 				/*
 				 * sched tick not stopped!
 				 */
-				cpu_clear(cpu, nohz_cpu_mask);
+				cpumask_clear_cpu(cpu, nohz_cpu_mask);
 				goto out;
 			}
 
@@ -361,7 +361,7 @@
 		 * softirq.
 		 */
 		tick_do_update_jiffies64(ktime_get());
-		cpu_clear(cpu, nohz_cpu_mask);
+		cpumask_clear_cpu(cpu, nohz_cpu_mask);
 	}
 	raise_softirq_irqoff(TIMER_SOFTIRQ);
 out:
@@ -419,7 +419,9 @@
 {
 	int cpu = smp_processor_id();
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
 	unsigned long ticks;
+#endif
 	ktime_t now;
 
 	local_irq_disable();
@@ -439,8 +441,9 @@
 	select_nohz_load_balancer(0);
 	now = ktime_get();
 	tick_do_update_jiffies64(now);
-	cpu_clear(cpu, nohz_cpu_mask);
+	cpumask_clear_cpu(cpu, nohz_cpu_mask);
 
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
 	/*
 	 * We stopped the tick in idle. Update process times would miss the
 	 * time we slept as update_process_times does only a 1 tick
@@ -450,12 +453,9 @@
 	/*
 	 * We might be one off. Do not randomly account a huge number of ticks!
 	 */
-	if (ticks && ticks < LONG_MAX) {
-		add_preempt_count(HARDIRQ_OFFSET);
-		account_system_time(current, HARDIRQ_OFFSET,
-				    jiffies_to_cputime(ticks));
-		sub_preempt_count(HARDIRQ_OFFSET);
-	}
+	if (ticks && ticks < LONG_MAX)
+		account_idle_ticks(ticks);
+#endif
 
 	touch_softlockup_watchdog();
 	/*
diff --git a/kernel/timer.c b/kernel/timer.c
index 566257d..dee3f64 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1018,21 +1018,6 @@
 }
 #endif
 
-#ifndef CONFIG_VIRT_CPU_ACCOUNTING
-void account_process_tick(struct task_struct *p, int user_tick)
-{
-	cputime_t one_jiffy = jiffies_to_cputime(1);
-
-	if (user_tick) {
-		account_user_time(p, one_jiffy);
-		account_user_time_scaled(p, cputime_to_scaled(one_jiffy));
-	} else {
-		account_system_time(p, HARDIRQ_OFFSET, one_jiffy);
-		account_system_time_scaled(p, cputime_to_scaled(one_jiffy));
-	}
-}
-#endif
-
 /*
  * Called from the timer interrupt handler to charge one tick to the current
  * process.  user_tick is 1 if the tick is user time, 0 for system.
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 4185d52..0e91f43 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -2674,7 +2674,7 @@
 
 	mutex_lock(&tracing_cpumask_update_lock);
 
-	len = cpumask_scnprintf(mask_str, count, tracing_cpumask);
+	len = cpumask_scnprintf(mask_str, count, &tracing_cpumask);
 	if (count - len < 2) {
 		count = -EINVAL;
 		goto out_err;
@@ -2695,7 +2695,7 @@
 	int err, cpu;
 
 	mutex_lock(&tracing_cpumask_update_lock);
-	err = cpumask_parse_user(ubuf, count, tracing_cpumask_new);
+	err = cpumask_parse_user(ubuf, count, &tracing_cpumask_new);
 	if (err)
 		goto err_unlock;
 
diff --git a/lib/Kconfig b/lib/Kconfig
index fd4118e..2ba43c4 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -159,4 +159,11 @@
 config HAVE_LMB
 	boolean
 
+config CPUMASK_OFFSTACK
+	bool "Force CPU masks off stack" if DEBUG_PER_CPU_MAPS
+	help
+	  Use dynamic allocation for cpumask_var_t, instead of putting
+	  them on the stack.  This is a bit more expensive, but avoids
+	  stack overflow.
+
 endmenu
diff --git a/lib/vsprintf.c b/lib/vsprintf.c
index 3b77702..98d63227 100644
--- a/lib/vsprintf.c
+++ b/lib/vsprintf.c
@@ -661,6 +661,9 @@
  */
 static char *pointer(const char *fmt, char *buf, char *end, void *ptr, int field_width, int precision, int flags)
 {
+	if (!ptr)
+		return string(buf, end, "(null)", field_width, precision, flags);
+
 	switch (*fmt) {
 	case 'F':
 		ptr = dereference_function_descriptor(ptr);
diff --git a/mm/slub.c b/mm/slub.c
index 6cb7ad1..0d861c3 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -3642,7 +3642,7 @@
 				len < PAGE_SIZE - 60) {
 			len += sprintf(buf + len, " cpus=");
 			len += cpulist_scnprintf(buf + len, PAGE_SIZE - len - 50,
-					l->cpus);
+					&l->cpus);
 		}
 
 		if (num_online_nodes() > 1 && !nodes_empty(l->nodes) &&
diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c
index e4e2cae..086d5ef 100644
--- a/net/irda/ircomm/ircomm_tty.c
+++ b/net/irda/ircomm/ircomm_tty.c
@@ -371,9 +371,8 @@
 	IRDA_DEBUG(2, "%s()\n", __func__ );
 
 	line = tty->index;
-	if ((line < 0) || (line >= IRCOMM_TTY_PORTS)) {
+	if (line >= IRCOMM_TTY_PORTS)
 		return -ENODEV;
-	}
 
 	/* Check if instance already exists */
 	self = hashbin_lock_find(ircomm_tty, line, NULL);
@@ -405,6 +404,8 @@
 		 * Force TTY into raw mode by default which is usually what
 		 * we want for IrCOMM and IrLPT. This way applications will
 		 * not have to twiddle with printcap etc.
+		 *
+		 * Note this is completely usafe and doesn't work properly
 		 */
 		tty->termios->c_iflag = 0;
 		tty->termios->c_oflag = 0;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index c6250d0..d1b8982 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -836,7 +836,11 @@
 		err = mnt_want_write(nd.path.mnt);
 		if (err)
 			goto out_mknod_dput;
+		err = security_path_mknod(&nd.path, dentry, mode, 0);
+		if (err)
+			goto out_mknod_drop_write;
 		err = vfs_mknod(nd.path.dentry->d_inode, dentry, mode, 0);
+out_mknod_drop_write:
 		mnt_drop_write(nd.path.mnt);
 		if (err)
 			goto out_mknod_dput;
diff --git a/scripts/headers_check.pl b/scripts/headers_check.pl
index 488a3b1..db30fac 100644
--- a/scripts/headers_check.pl
+++ b/scripts/headers_check.pl
@@ -14,7 +14,9 @@
 #    Only include files located in asm* and linux* are checked.
 #    The rest are assumed to be system include files.
 #
-# 2) TODO: check for leaked CONFIG_ symbols
+# 2) It is checked that prototypes does not use "extern"
+#
+# 3) Check for leaked CONFIG_ symbols
 
 use strict;
 
@@ -32,7 +34,11 @@
 	$lineno = 0;
 	while ($line = <FH>) {
 		$lineno++;
-		check_include();
+		&check_include();
+		&check_asm_types();
+		&check_sizetypes();
+		&check_prototypes();
+		&check_config();
 	}
 	close FH;
 }
@@ -54,3 +60,63 @@
 		}
 	}
 }
+
+sub check_prototypes
+{
+	if ($line =~ m/^\s*extern\b/) {
+		printf STDERR "$filename:$lineno: extern's make no sense in userspace\n";
+	}
+}
+
+sub check_config
+{
+	if ($line =~ m/[^a-zA-Z0-9_]+CONFIG_([a-zA-Z0-9]+)[^a-zA-Z0-9]/) {
+		printf STDERR "$filename:$lineno: leaks CONFIG_$1 to userspace where it is not valid\n";
+	}
+}
+
+my $linux_asm_types;
+sub check_asm_types()
+{
+	if ($filename =~ /types.h|int-l64.h|int-ll64.h/o) {
+		return;
+	}
+	if ($lineno == 1) {
+		$linux_asm_types = 0;
+	} elsif ($linux_asm_types >= 1) {
+		return;
+	}
+	if ($line =~ m/^\s*#\s*include\s+<asm\/types.h>/) {
+		$linux_asm_types = 1;
+		printf STDERR "$filename:$lineno: " .
+		"include of <linux/types.h> is preferred over <asm/types.h>\n"
+		# Warn until headers are all fixed
+		#$ret = 1;
+	}
+}
+
+my $linux_types;
+sub check_sizetypes
+{
+	if ($filename =~ /types.h|int-l64.h|int-ll64.h/o) {
+		return;
+	}
+	if ($lineno == 1) {
+		$linux_types = 0;
+	} elsif ($linux_types >= 1) {
+		return;
+	}
+	if ($line =~ m/^\s*#\s*include\s+<linux\/types.h>/) {
+		$linux_types = 1;
+		return;
+	}
+	if ($line =~ m/__[us](8|16|32|64)\b/) {
+		printf STDERR "$filename:$lineno: " .
+		              "found __[us]{8,16,32,64} type " .
+		              "without #include <linux/types.h>\n";
+		$linux_types = 2;
+		# Warn until headers are all fixed
+		#$ret = 1;
+	}
+}
+
diff --git a/scripts/headers_install.pl b/scripts/headers_install.pl
index 7d2b414..c6ae405 100644
--- a/scripts/headers_install.pl
+++ b/scripts/headers_install.pl
@@ -36,6 +36,9 @@
 		$line =~ s/\s__attribute_const__\s/ /g;
 		$line =~ s/\s__attribute_const__$//g;
 		$line =~ s/^#include <linux\/compiler.h>//;
+		$line =~ s/(^|\s)(inline)\b/$1__$2__/g;
+		$line =~ s/(^|\s)(asm)\b(\s|[(]|$)/$1__$2__$3/g;
+		$line =~ s/(^|\s|[(])(volatile)\b(\s|[(]|$)/$1__$2__$3/g;
 		printf OUTFILE "%s", $line;
 	}
 	close OUTFILE;
diff --git a/scripts/kconfig/expr.h b/scripts/kconfig/expr.h
index 9d4cba1..6408fef 100644
--- a/scripts/kconfig/expr.h
+++ b/scripts/kconfig/expr.h
@@ -65,9 +65,13 @@
 	S_UNKNOWN, S_BOOLEAN, S_TRISTATE, S_INT, S_HEX, S_STRING, S_OTHER
 };
 
+/* enum values are used as index to symbol.def[] */
 enum {
 	S_DEF_USER,		/* main user value */
-	S_DEF_AUTO,
+	S_DEF_AUTO,		/* values read from auto.conf */
+	S_DEF_DEF3,		/* Reserved for UI usage */
+	S_DEF_DEF4,		/* Reserved for UI usage */
+	S_DEF_COUNT
 };
 
 struct symbol {
@@ -75,7 +79,7 @@
 	char *name;
 	enum symbol_type type;
 	struct symbol_value curr;
-	struct symbol_value def[4];
+	struct symbol_value def[S_DEF_COUNT];
 	tristate visible;
 	int flags;
 	struct property *prop;
@@ -84,42 +88,64 @@
 
 #define for_all_symbols(i, sym) for (i = 0; i < 257; i++) for (sym = symbol_hash[i]; sym; sym = sym->next) if (sym->type != S_OTHER)
 
-#define SYMBOL_CONST		0x0001
-#define SYMBOL_CHECK		0x0008
-#define SYMBOL_CHOICE		0x0010
-#define SYMBOL_CHOICEVAL	0x0020
-#define SYMBOL_VALID		0x0080
-#define SYMBOL_OPTIONAL		0x0100
-#define SYMBOL_WRITE		0x0200
-#define SYMBOL_CHANGED		0x0400
-#define SYMBOL_AUTO		0x1000
-#define SYMBOL_CHECKED		0x2000
-#define SYMBOL_WARNED		0x8000
-#define SYMBOL_DEF		0x10000
-#define SYMBOL_DEF_USER		0x10000
-#define SYMBOL_DEF_AUTO		0x20000
-#define SYMBOL_DEF3		0x40000
-#define SYMBOL_DEF4		0x80000
+#define SYMBOL_CONST      0x0001  /* symbol is const */
+#define SYMBOL_CHECK      0x0008  /* used during dependency checking */
+#define SYMBOL_CHOICE     0x0010  /* start of a choice block (null name) */
+#define SYMBOL_CHOICEVAL  0x0020  /* used as a value in a choice block */
+#define SYMBOL_VALID      0x0080  /* set when symbol.curr is calculated */
+#define SYMBOL_OPTIONAL   0x0100  /* choice is optional - values can be 'n' */
+#define SYMBOL_WRITE      0x0200  /* ? */
+#define SYMBOL_CHANGED    0x0400  /* ? */
+#define SYMBOL_AUTO       0x1000  /* value from environment variable */
+#define SYMBOL_CHECKED    0x2000  /* used during dependency checking */
+#define SYMBOL_WARNED     0x8000  /* warning has been issued */
+
+/* Set when symbol.def[] is used */
+#define SYMBOL_DEF        0x10000  /* First bit of SYMBOL_DEF */
+#define SYMBOL_DEF_USER   0x10000  /* symbol.def[S_DEF_USER] is valid */
+#define SYMBOL_DEF_AUTO   0x20000  /* symbol.def[S_DEF_AUTO] is valid */
+#define SYMBOL_DEF3       0x40000  /* symbol.def[S_DEF_3] is valid */
+#define SYMBOL_DEF4       0x80000  /* symbol.def[S_DEF_4] is valid */
 
 #define SYMBOL_MAXLENGTH	256
 #define SYMBOL_HASHSIZE		257
 #define SYMBOL_HASHMASK		0xff
 
+/* A property represent the config options that can be associated
+ * with a config "symbol".
+ * Sample:
+ * config FOO
+ *         default y
+ *         prompt "foo prompt"
+ *         select BAR
+ * config BAZ
+ *         int "BAZ Value"
+ *         range 1..255
+ */
 enum prop_type {
-	P_UNKNOWN, P_PROMPT, P_COMMENT, P_MENU, P_DEFAULT, P_CHOICE,
-	P_SELECT, P_RANGE, P_ENV
+	P_UNKNOWN,
+	P_PROMPT,   /* prompt "foo prompt" or "BAZ Value" */
+	P_COMMENT,  /* text associated with a comment */
+	P_MENU,     /* prompt associated with a menuconfig option */
+	P_DEFAULT,  /* default y */
+	P_CHOICE,   /* choice value */
+	P_SELECT,   /* select BAR */
+	P_RANGE,    /* range 7..100 (for a symbol) */
+	P_ENV,      /* value from environment variable */
 };
 
 struct property {
-	struct property *next;
-	struct symbol *sym;
-	enum prop_type type;
-	const char *text;
+	struct property *next;     /* next property - null if last */
+	struct symbol *sym;        /* the symbol for which the property is associated */
+	enum prop_type type;       /* type of property */
+	const char *text;          /* the prompt value - P_PROMPT, P_MENU, P_COMMENT */
 	struct expr_value visible;
-	struct expr *expr;
-	struct menu *menu;
-	struct file *file;
-	int lineno;
+	struct expr *expr;         /* the optional conditional part of the property */
+	struct menu *menu;         /* the menu the property are associated with
+	                            * valid for: P_SELECT, P_RANGE, P_CHOICE,
+	                            * P_PROMPT, P_DEFAULT, P_MENU, P_COMMENT */
+	struct file *file;         /* what file was this property defined */
+	int lineno;                /* what lineno was this property defined */
 };
 
 #define for_all_properties(sym, st, tok) \
diff --git a/scripts/kconfig/lex.zconf.c_shipped b/scripts/kconfig/lex.zconf.c_shipped
index 7342ce0..dc3e818 100644
--- a/scripts/kconfig/lex.zconf.c_shipped
+++ b/scripts/kconfig/lex.zconf.c_shipped
@@ -2370,11 +2370,14 @@
 	current_buf = buf;
 
 	if (file->flags & FILE_BUSY) {
-		printf("recursive scan (%s)?\n", name);
+		printf("%s:%d: do not source '%s' from itself\n",
+		       zconf_curname(), zconf_lineno(), name);
 		exit(1);
 	}
 	if (file->flags & FILE_SCANNED) {
-		printf("file %s already scanned?\n", name);
+		printf("%s:%d: file '%s' is already sourced from '%s'\n",
+		       zconf_curname(), zconf_lineno(), name,
+		       file->parent->name);
 		exit(1);
 	}
 	file->flags |= FILE_BUSY;
diff --git a/scripts/kconfig/zconf.l b/scripts/kconfig/zconf.l
index 5164ef7..21ff69c 100644
--- a/scripts/kconfig/zconf.l
+++ b/scripts/kconfig/zconf.l
@@ -314,11 +314,14 @@
 	current_buf = buf;
 
 	if (file->flags & FILE_BUSY) {
-		printf("recursive scan (%s)?\n", name);
+		printf("%s:%d: do not source '%s' from itself\n",
+		       zconf_curname(), zconf_lineno(), name);
 		exit(1);
 	}
 	if (file->flags & FILE_SCANNED) {
-		printf("file %s already scanned?\n", name);
+		printf("%s:%d: file '%s' is already sourced from '%s'\n",
+		       zconf_curname(), zconf_lineno(), name,
+		       file->parent->name);
 		exit(1);
 	}
 	file->flags |= FILE_BUSY;
diff --git a/scripts/tags.sh b/scripts/tags.sh
index 4e75472..9e3451d 100755
--- a/scripts/tags.sh
+++ b/scripts/tags.sh
@@ -84,7 +84,6 @@
 
 exuberant()
 {
-	all_sources > all
 	all_sources | xargs $1 -a                               \
 	-I __initdata,__exitdata,__acquires,__releases          \
 	-I __read_mostly,____cacheline_aligned                  \
diff --git a/security/Kconfig b/security/Kconfig
index d9f47ce..9438535 100644
--- a/security/Kconfig
+++ b/security/Kconfig
@@ -81,6 +81,15 @@
 	  IPSec.
 	  If you are unsure how to answer this question, answer N.
 
+config SECURITY_PATH
+	bool "Security hooks for pathname based access control"
+	depends on SECURITY
+	help
+	  This enables the security hooks for pathname based access control.
+	  If enabled, a security module can use these hooks to
+	  implement pathname based access controls.
+	  If you are unsure how to answer this question, answer N.
+
 config SECURITY_FILE_CAPABILITIES
 	bool "File POSIX Capabilities"
 	default n
diff --git a/security/capability.c b/security/capability.c
index 2dce66f..c545bd1 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -263,6 +263,53 @@
 	*secid = 0;
 }
 
+#ifdef CONFIG_SECURITY_PATH
+static int cap_path_mknod(struct path *dir, struct dentry *dentry, int mode,
+			  unsigned int dev)
+{
+	return 0;
+}
+
+static int cap_path_mkdir(struct path *dir, struct dentry *dentry, int mode)
+{
+	return 0;
+}
+
+static int cap_path_rmdir(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static int cap_path_unlink(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static int cap_path_symlink(struct path *dir, struct dentry *dentry,
+			    const char *old_name)
+{
+	return 0;
+}
+
+static int cap_path_link(struct dentry *old_dentry, struct path *new_dir,
+			 struct dentry *new_dentry)
+{
+	return 0;
+}
+
+static int cap_path_rename(struct path *old_path, struct dentry *old_dentry,
+			   struct path *new_path, struct dentry *new_dentry)
+{
+	return 0;
+}
+
+static int cap_path_truncate(struct path *path, loff_t length,
+			     unsigned int time_attrs)
+{
+	return 0;
+}
+#endif
+
 static int cap_file_permission(struct file *file, int mask)
 {
 	return 0;
@@ -883,6 +930,16 @@
 	set_to_cap_if_null(ops, inode_setsecurity);
 	set_to_cap_if_null(ops, inode_listsecurity);
 	set_to_cap_if_null(ops, inode_getsecid);
+#ifdef CONFIG_SECURITY_PATH
+	set_to_cap_if_null(ops, path_mknod);
+	set_to_cap_if_null(ops, path_mkdir);
+	set_to_cap_if_null(ops, path_rmdir);
+	set_to_cap_if_null(ops, path_unlink);
+	set_to_cap_if_null(ops, path_symlink);
+	set_to_cap_if_null(ops, path_link);
+	set_to_cap_if_null(ops, path_rename);
+	set_to_cap_if_null(ops, path_truncate);
+#endif
 	set_to_cap_if_null(ops, file_permission);
 	set_to_cap_if_null(ops, file_alloc_security);
 	set_to_cap_if_null(ops, file_free_security);
diff --git a/security/security.c b/security/security.c
index d85dbb3..678d4d0 100644
--- a/security/security.c
+++ b/security/security.c
@@ -355,6 +355,72 @@
 }
 EXPORT_SYMBOL(security_inode_init_security);
 
+#ifdef CONFIG_SECURITY_PATH
+int security_path_mknod(struct path *path, struct dentry *dentry, int mode,
+			unsigned int dev)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_mknod(path, dentry, mode, dev);
+}
+EXPORT_SYMBOL(security_path_mknod);
+
+int security_path_mkdir(struct path *path, struct dentry *dentry, int mode)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_mkdir(path, dentry, mode);
+}
+
+int security_path_rmdir(struct path *path, struct dentry *dentry)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_rmdir(path, dentry);
+}
+
+int security_path_unlink(struct path *path, struct dentry *dentry)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_unlink(path, dentry);
+}
+
+int security_path_symlink(struct path *path, struct dentry *dentry,
+			  const char *old_name)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_symlink(path, dentry, old_name);
+}
+
+int security_path_link(struct dentry *old_dentry, struct path *new_dir,
+		       struct dentry *new_dentry)
+{
+	if (unlikely(IS_PRIVATE(old_dentry->d_inode)))
+		return 0;
+	return security_ops->path_link(old_dentry, new_dir, new_dentry);
+}
+
+int security_path_rename(struct path *old_dir, struct dentry *old_dentry,
+			 struct path *new_dir, struct dentry *new_dentry)
+{
+	if (unlikely(IS_PRIVATE(old_dentry->d_inode) ||
+		     (new_dentry->d_inode && IS_PRIVATE(new_dentry->d_inode))))
+		return 0;
+	return security_ops->path_rename(old_dir, old_dentry, new_dir,
+					 new_dentry);
+}
+
+int security_path_truncate(struct path *path, loff_t length,
+			   unsigned int time_attrs)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_truncate(path, length, time_attrs);
+}
+#endif
+
 int security_inode_create(struct inode *dir, struct dentry *dentry, int mode)
 {
 	if (unlikely(IS_PRIVATE(dir)))
diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c
index 53772bb..23b81cf 100644
--- a/virt/kvm/ioapic.c
+++ b/virt/kvm/ioapic.c
@@ -150,10 +150,11 @@
 static void ioapic_inj_nmi(struct kvm_vcpu *vcpu)
 {
 	kvm_inject_nmi(vcpu);
+	kvm_vcpu_kick(vcpu);
 }
 
-static u32 ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
-				       u8 dest_mode)
+u32 kvm_ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
+				    u8 dest_mode)
 {
 	u32 mask = 0;
 	int i;
@@ -207,7 +208,8 @@
 		     "vector=%x trig_mode=%x\n",
 		     dest, dest_mode, delivery_mode, vector, trig_mode);
 
-	deliver_bitmask = ioapic_get_delivery_bitmask(ioapic, dest, dest_mode);
+	deliver_bitmask = kvm_ioapic_get_delivery_bitmask(ioapic, dest,
+							  dest_mode);
 	if (!deliver_bitmask) {
 		ioapic_debug("no target on destination\n");
 		return 0;
diff --git a/virt/kvm/ioapic.h b/virt/kvm/ioapic.h
index cd7ae76..49c9581 100644
--- a/virt/kvm/ioapic.h
+++ b/virt/kvm/ioapic.h
@@ -85,5 +85,7 @@
 int kvm_ioapic_init(struct kvm *kvm);
 void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level);
 void kvm_ioapic_reset(struct kvm_ioapic *ioapic);
+u32 kvm_ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
+				u8 dest_mode);
 
 #endif
diff --git a/virt/kvm/irq_comm.c b/virt/kvm/irq_comm.c
index 55ad76e..aa5d1e5 100644
--- a/virt/kvm/irq_comm.c
+++ b/virt/kvm/irq_comm.c
@@ -61,10 +61,9 @@
 	hlist_add_head(&kian->link, &kvm->arch.irq_ack_notifier_list);
 }
 
-void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
-				     struct kvm_irq_ack_notifier *kian)
+void kvm_unregister_irq_ack_notifier(struct kvm_irq_ack_notifier *kian)
 {
-	hlist_del(&kian->link);
+	hlist_del_init(&kian->link);
 }
 
 /* The caller must hold kvm->lock mutex */
@@ -73,11 +72,15 @@
 	unsigned long *bitmap = &kvm->arch.irq_sources_bitmap;
 	int irq_source_id = find_first_zero_bit(bitmap,
 				sizeof(kvm->arch.irq_sources_bitmap));
+
 	if (irq_source_id >= sizeof(kvm->arch.irq_sources_bitmap)) {
 		printk(KERN_WARNING "kvm: exhaust allocatable IRQ sources!\n");
-		irq_source_id = -EFAULT;
-	} else
-		set_bit(irq_source_id, bitmap);
+		return -EFAULT;
+	}
+
+	ASSERT(irq_source_id != KVM_USERSPACE_IRQ_SOURCE_ID);
+	set_bit(irq_source_id, bitmap);
+
 	return irq_source_id;
 }
 
@@ -85,7 +88,9 @@
 {
 	int i;
 
-	if (irq_source_id <= 0 ||
+	ASSERT(irq_source_id != KVM_USERSPACE_IRQ_SOURCE_ID);
+
+	if (irq_source_id < 0 ||
 	    irq_source_id >= sizeof(kvm->arch.irq_sources_bitmap)) {
 		printk(KERN_ERR "kvm: IRQ source ID out of range!\n");
 		return;
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index a87f45e..fc6127c 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -47,6 +47,10 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
+#ifdef CONFIG_X86
+#include <asm/msidef.h>
+#endif
+
 #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
 #include "coalesced_mmio.h"
 #endif
@@ -60,10 +64,13 @@
 MODULE_AUTHOR("Qumranet");
 MODULE_LICENSE("GPL");
 
+static int msi2intx = 1;
+module_param(msi2intx, bool, 0);
+
 DEFINE_SPINLOCK(kvm_lock);
 LIST_HEAD(vm_list);
 
-static cpumask_t cpus_hardware_enabled;
+static cpumask_var_t cpus_hardware_enabled;
 
 struct kmem_cache *kvm_vcpu_cache;
 EXPORT_SYMBOL_GPL(kvm_vcpu_cache);
@@ -75,9 +82,60 @@
 static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl,
 			   unsigned long arg);
 
-bool kvm_rebooting;
+static bool kvm_rebooting;
 
 #ifdef KVM_CAP_DEVICE_ASSIGNMENT
+
+#ifdef CONFIG_X86
+static void assigned_device_msi_dispatch(struct kvm_assigned_dev_kernel *dev)
+{
+	int vcpu_id;
+	struct kvm_vcpu *vcpu;
+	struct kvm_ioapic *ioapic = ioapic_irqchip(dev->kvm);
+	int dest_id = (dev->guest_msi.address_lo & MSI_ADDR_DEST_ID_MASK)
+			>> MSI_ADDR_DEST_ID_SHIFT;
+	int vector = (dev->guest_msi.data & MSI_DATA_VECTOR_MASK)
+			>> MSI_DATA_VECTOR_SHIFT;
+	int dest_mode = test_bit(MSI_ADDR_DEST_MODE_SHIFT,
+				(unsigned long *)&dev->guest_msi.address_lo);
+	int trig_mode = test_bit(MSI_DATA_TRIGGER_SHIFT,
+				(unsigned long *)&dev->guest_msi.data);
+	int delivery_mode = test_bit(MSI_DATA_DELIVERY_MODE_SHIFT,
+				(unsigned long *)&dev->guest_msi.data);
+	u32 deliver_bitmask;
+
+	BUG_ON(!ioapic);
+
+	deliver_bitmask = kvm_ioapic_get_delivery_bitmask(ioapic,
+				dest_id, dest_mode);
+	/* IOAPIC delivery mode value is the same as MSI here */
+	switch (delivery_mode) {
+	case IOAPIC_LOWEST_PRIORITY:
+		vcpu = kvm_get_lowest_prio_vcpu(ioapic->kvm, vector,
+				deliver_bitmask);
+		if (vcpu != NULL)
+			kvm_apic_set_irq(vcpu, vector, trig_mode);
+		else
+			printk(KERN_INFO "kvm: null lowest priority vcpu!\n");
+		break;
+	case IOAPIC_FIXED:
+		for (vcpu_id = 0; deliver_bitmask != 0; vcpu_id++) {
+			if (!(deliver_bitmask & (1 << vcpu_id)))
+				continue;
+			deliver_bitmask &= ~(1 << vcpu_id);
+			vcpu = ioapic->kvm->vcpus[vcpu_id];
+			if (vcpu)
+				kvm_apic_set_irq(vcpu, vector, trig_mode);
+		}
+		break;
+	default:
+		printk(KERN_INFO "kvm: unsupported MSI delivery mode\n");
+	}
+}
+#else
+static void assigned_device_msi_dispatch(struct kvm_assigned_dev_kernel *dev) {}
+#endif
+
 static struct kvm_assigned_dev_kernel *kvm_find_assigned_dev(struct list_head *head,
 						      int assigned_dev_id)
 {
@@ -104,9 +162,16 @@
 	 * finer-grained lock, update this
 	 */
 	mutex_lock(&assigned_dev->kvm->lock);
-	kvm_set_irq(assigned_dev->kvm,
-		    assigned_dev->irq_source_id,
-		    assigned_dev->guest_irq, 1);
+	if (assigned_dev->irq_requested_type & KVM_ASSIGNED_DEV_GUEST_INTX)
+		kvm_set_irq(assigned_dev->kvm,
+			    assigned_dev->irq_source_id,
+			    assigned_dev->guest_irq, 1);
+	else if (assigned_dev->irq_requested_type &
+				KVM_ASSIGNED_DEV_GUEST_MSI) {
+		assigned_device_msi_dispatch(assigned_dev);
+		enable_irq(assigned_dev->host_irq);
+		assigned_dev->host_irq_disabled = false;
+	}
 	mutex_unlock(&assigned_dev->kvm->lock);
 	kvm_put_kvm(assigned_dev->kvm);
 }
@@ -117,8 +182,12 @@
 		(struct kvm_assigned_dev_kernel *) dev_id;
 
 	kvm_get_kvm(assigned_dev->kvm);
+
 	schedule_work(&assigned_dev->interrupt_work);
+
 	disable_irq_nosync(irq);
+	assigned_dev->host_irq_disabled = true;
+
 	return IRQ_HANDLED;
 }
 
@@ -132,19 +201,32 @@
 
 	dev = container_of(kian, struct kvm_assigned_dev_kernel,
 			   ack_notifier);
+
 	kvm_set_irq(dev->kvm, dev->irq_source_id, dev->guest_irq, 0);
-	enable_irq(dev->host_irq);
+
+	/* The guest irq may be shared so this ack may be
+	 * from another device.
+	 */
+	if (dev->host_irq_disabled) {
+		enable_irq(dev->host_irq);
+		dev->host_irq_disabled = false;
+	}
 }
 
-static void kvm_free_assigned_device(struct kvm *kvm,
-				     struct kvm_assigned_dev_kernel
-				     *assigned_dev)
+static void kvm_free_assigned_irq(struct kvm *kvm,
+				  struct kvm_assigned_dev_kernel *assigned_dev)
 {
-	if (irqchip_in_kernel(kvm) && assigned_dev->irq_requested)
-		free_irq(assigned_dev->host_irq, (void *)assigned_dev);
+	if (!irqchip_in_kernel(kvm))
+		return;
 
-	kvm_unregister_irq_ack_notifier(kvm, &assigned_dev->ack_notifier);
-	kvm_free_irq_source_id(kvm, assigned_dev->irq_source_id);
+	kvm_unregister_irq_ack_notifier(&assigned_dev->ack_notifier);
+
+	if (assigned_dev->irq_source_id != -1)
+		kvm_free_irq_source_id(kvm, assigned_dev->irq_source_id);
+	assigned_dev->irq_source_id = -1;
+
+	if (!assigned_dev->irq_requested_type)
+		return;
 
 	if (cancel_work_sync(&assigned_dev->interrupt_work))
 		/* We had pending work. That means we will have to take
@@ -152,6 +234,23 @@
 		 */
 		kvm_put_kvm(kvm);
 
+	free_irq(assigned_dev->host_irq, (void *)assigned_dev);
+
+	if (assigned_dev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI)
+		pci_disable_msi(assigned_dev->dev);
+
+	assigned_dev->irq_requested_type = 0;
+}
+
+
+static void kvm_free_assigned_device(struct kvm *kvm,
+				     struct kvm_assigned_dev_kernel
+				     *assigned_dev)
+{
+	kvm_free_assigned_irq(kvm, assigned_dev);
+
+	pci_reset_function(assigned_dev->dev);
+
 	pci_release_regions(assigned_dev->dev);
 	pci_disable_device(assigned_dev->dev);
 	pci_dev_put(assigned_dev->dev);
@@ -174,6 +273,95 @@
 	}
 }
 
+static int assigned_device_update_intx(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *adev,
+			struct kvm_assigned_irq *airq)
+{
+	adev->guest_irq = airq->guest_irq;
+	adev->ack_notifier.gsi = airq->guest_irq;
+
+	if (adev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_INTX)
+		return 0;
+
+	if (irqchip_in_kernel(kvm)) {
+		if (!msi2intx &&
+		    adev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI) {
+			free_irq(adev->host_irq, (void *)kvm);
+			pci_disable_msi(adev->dev);
+		}
+
+		if (!capable(CAP_SYS_RAWIO))
+			return -EPERM;
+
+		if (airq->host_irq)
+			adev->host_irq = airq->host_irq;
+		else
+			adev->host_irq = adev->dev->irq;
+
+		/* Even though this is PCI, we don't want to use shared
+		 * interrupts. Sharing host devices with guest-assigned devices
+		 * on the same interrupt line is not a happy situation: there
+		 * are going to be long delays in accepting, acking, etc.
+		 */
+		if (request_irq(adev->host_irq, kvm_assigned_dev_intr,
+				0, "kvm_assigned_intx_device", (void *)adev))
+			return -EIO;
+	}
+
+	adev->irq_requested_type = KVM_ASSIGNED_DEV_GUEST_INTX |
+				   KVM_ASSIGNED_DEV_HOST_INTX;
+	return 0;
+}
+
+#ifdef CONFIG_X86
+static int assigned_device_update_msi(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *adev,
+			struct kvm_assigned_irq *airq)
+{
+	int r;
+
+	if (airq->flags & KVM_DEV_IRQ_ASSIGN_ENABLE_MSI) {
+		/* x86 don't care upper address of guest msi message addr */
+		adev->irq_requested_type |= KVM_ASSIGNED_DEV_GUEST_MSI;
+		adev->irq_requested_type &= ~KVM_ASSIGNED_DEV_GUEST_INTX;
+		adev->guest_msi.address_lo = airq->guest_msi.addr_lo;
+		adev->guest_msi.data = airq->guest_msi.data;
+		adev->ack_notifier.gsi = -1;
+	} else if (msi2intx) {
+		adev->irq_requested_type |= KVM_ASSIGNED_DEV_GUEST_INTX;
+		adev->irq_requested_type &= ~KVM_ASSIGNED_DEV_GUEST_MSI;
+		adev->guest_irq = airq->guest_irq;
+		adev->ack_notifier.gsi = airq->guest_irq;
+	}
+
+	if (adev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI)
+		return 0;
+
+	if (irqchip_in_kernel(kvm)) {
+		if (!msi2intx) {
+			if (adev->irq_requested_type &
+					KVM_ASSIGNED_DEV_HOST_INTX)
+				free_irq(adev->host_irq, (void *)adev);
+
+			r = pci_enable_msi(adev->dev);
+			if (r)
+				return r;
+		}
+
+		adev->host_irq = adev->dev->irq;
+		if (request_irq(adev->host_irq, kvm_assigned_dev_intr, 0,
+				"kvm_assigned_msi_device", (void *)adev))
+			return -EIO;
+	}
+
+	if (!msi2intx)
+		adev->irq_requested_type = KVM_ASSIGNED_DEV_GUEST_MSI;
+
+	adev->irq_requested_type |= KVM_ASSIGNED_DEV_HOST_MSI;
+	return 0;
+}
+#endif
+
 static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 				   struct kvm_assigned_irq
 				   *assigned_irq)
@@ -190,49 +378,68 @@
 		return -EINVAL;
 	}
 
-	if (match->irq_requested) {
-		match->guest_irq = assigned_irq->guest_irq;
-		match->ack_notifier.gsi = assigned_irq->guest_irq;
-		mutex_unlock(&kvm->lock);
-		return 0;
+	if (!match->irq_requested_type) {
+		INIT_WORK(&match->interrupt_work,
+				kvm_assigned_dev_interrupt_work_handler);
+		if (irqchip_in_kernel(kvm)) {
+			/* Register ack nofitier */
+			match->ack_notifier.gsi = -1;
+			match->ack_notifier.irq_acked =
+					kvm_assigned_dev_ack_irq;
+			kvm_register_irq_ack_notifier(kvm,
+					&match->ack_notifier);
+
+			/* Request IRQ source ID */
+			r = kvm_request_irq_source_id(kvm);
+			if (r < 0)
+				goto out_release;
+			else
+				match->irq_source_id = r;
+
+#ifdef CONFIG_X86
+			/* Determine host device irq type, we can know the
+			 * result from dev->msi_enabled */
+			if (msi2intx)
+				pci_enable_msi(match->dev);
+#endif
+		}
 	}
 
-	INIT_WORK(&match->interrupt_work,
-		  kvm_assigned_dev_interrupt_work_handler);
-
-	if (irqchip_in_kernel(kvm)) {
-		if (!capable(CAP_SYS_RAWIO)) {
-			r = -EPERM;
+	if ((!msi2intx &&
+	     (assigned_irq->flags & KVM_DEV_IRQ_ASSIGN_ENABLE_MSI)) ||
+	    (msi2intx && match->dev->msi_enabled)) {
+#ifdef CONFIG_X86
+		r = assigned_device_update_msi(kvm, match, assigned_irq);
+		if (r) {
+			printk(KERN_WARNING "kvm: failed to enable "
+					"MSI device!\n");
 			goto out_release;
 		}
-
-		if (assigned_irq->host_irq)
-			match->host_irq = assigned_irq->host_irq;
-		else
-			match->host_irq = match->dev->irq;
-		match->guest_irq = assigned_irq->guest_irq;
-		match->ack_notifier.gsi = assigned_irq->guest_irq;
-		match->ack_notifier.irq_acked = kvm_assigned_dev_ack_irq;
-		kvm_register_irq_ack_notifier(kvm, &match->ack_notifier);
-		r = kvm_request_irq_source_id(kvm);
-		if (r < 0)
+#else
+		r = -ENOTTY;
+#endif
+	} else if (assigned_irq->host_irq == 0 && match->dev->irq == 0) {
+		/* Host device IRQ 0 means don't support INTx */
+		if (!msi2intx) {
+			printk(KERN_WARNING
+			       "kvm: wait device to enable MSI!\n");
+			r = 0;
+		} else {
+			printk(KERN_WARNING
+			       "kvm: failed to enable MSI device!\n");
+			r = -ENOTTY;
 			goto out_release;
-		else
-			match->irq_source_id = r;
-
-		/* Even though this is PCI, we don't want to use shared
-		 * interrupts. Sharing host devices with guest-assigned devices
-		 * on the same interrupt line is not a happy situation: there
-		 * are going to be long delays in accepting, acking, etc.
-		 */
-		if (request_irq(match->host_irq, kvm_assigned_dev_intr, 0,
-				"kvm_assigned_device", (void *)match)) {
-			r = -EIO;
+		}
+	} else {
+		/* Non-sharing INTx mode */
+		r = assigned_device_update_intx(kvm, match, assigned_irq);
+		if (r) {
+			printk(KERN_WARNING "kvm: failed to enable "
+					"INTx device!\n");
 			goto out_release;
 		}
 	}
 
-	match->irq_requested = true;
 	mutex_unlock(&kvm->lock);
 	return r;
 out_release:
@@ -283,11 +490,14 @@
 		       __func__);
 		goto out_disable;
 	}
+
+	pci_reset_function(dev);
+
 	match->assigned_dev_id = assigned_dev->assigned_dev_id;
 	match->host_busnr = assigned_dev->busnr;
 	match->host_devfn = assigned_dev->devfn;
 	match->dev = dev;
-
+	match->irq_source_id = -1;
 	match->kvm = kvm;
 
 	list_add(&match->list, &kvm->arch.assigned_dev_head);
@@ -355,58 +565,49 @@
 {
 }
 
-void kvm_flush_remote_tlbs(struct kvm *kvm)
+static bool make_all_cpus_request(struct kvm *kvm, unsigned int req)
 {
 	int i, cpu, me;
-	cpumask_t cpus;
+	cpumask_var_t cpus;
+	bool called = true;
 	struct kvm_vcpu *vcpu;
 
+	if (alloc_cpumask_var(&cpus, GFP_ATOMIC))
+		cpumask_clear(cpus);
+
 	me = get_cpu();
-	cpus_clear(cpus);
 	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
 		vcpu = kvm->vcpus[i];
 		if (!vcpu)
 			continue;
-		if (test_and_set_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
+		if (test_and_set_bit(req, &vcpu->requests))
 			continue;
 		cpu = vcpu->cpu;
-		if (cpu != -1 && cpu != me)
-			cpu_set(cpu, cpus);
+		if (cpus != NULL && cpu != -1 && cpu != me)
+			cpumask_set_cpu(cpu, cpus);
 	}
-	if (cpus_empty(cpus))
-		goto out;
-	++kvm->stat.remote_tlb_flush;
-	smp_call_function_mask(cpus, ack_flush, NULL, 1);
-out:
+	if (unlikely(cpus == NULL))
+		smp_call_function_many(cpu_online_mask, ack_flush, NULL, 1);
+	else if (!cpumask_empty(cpus))
+		smp_call_function_many(cpus, ack_flush, NULL, 1);
+	else
+		called = false;
 	put_cpu();
+	free_cpumask_var(cpus);
+	return called;
+}
+
+void kvm_flush_remote_tlbs(struct kvm *kvm)
+{
+	if (make_all_cpus_request(kvm, KVM_REQ_TLB_FLUSH))
+		++kvm->stat.remote_tlb_flush;
 }
 
 void kvm_reload_remote_mmus(struct kvm *kvm)
 {
-	int i, cpu, me;
-	cpumask_t cpus;
-	struct kvm_vcpu *vcpu;
-
-	me = get_cpu();
-	cpus_clear(cpus);
-	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
-		vcpu = kvm->vcpus[i];
-		if (!vcpu)
-			continue;
-		if (test_and_set_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests))
-			continue;
-		cpu = vcpu->cpu;
-		if (cpu != -1 && cpu != me)
-			cpu_set(cpu, cpus);
-	}
-	if (cpus_empty(cpus))
-		goto out;
-	smp_call_function_mask(cpus, ack_flush, NULL, 1);
-out:
-	put_cpu();
+	make_all_cpus_request(kvm, KVM_REQ_MMU_RELOAD);
 }
 
-
 int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
 {
 	struct page *page;
@@ -710,6 +911,8 @@
 		goto out;
 	if (mem->guest_phys_addr & (PAGE_SIZE - 1))
 		goto out;
+	if (user_alloc && (mem->userspace_addr & (PAGE_SIZE - 1)))
+		goto out;
 	if (mem->slot >= KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS)
 		goto out;
 	if (mem->guest_phys_addr + mem->memory_size < mem->guest_phys_addr)
@@ -821,7 +1024,10 @@
 		goto out_free;
 	}
 
-	kvm_free_physmem_slot(&old, &new);
+	kvm_free_physmem_slot(&old, npages ? &new : NULL);
+	/* Slot deletion case: we have to update the current slot */
+	if (!npages)
+		*memslot = old;
 #ifdef CONFIG_DMAR
 	/* map the pages in iommu page table */
 	r = kvm_iommu_map_pages(kvm, base_gfn, npages);
@@ -918,7 +1124,7 @@
 }
 EXPORT_SYMBOL_GPL(kvm_is_error_hva);
 
-static struct kvm_memory_slot *__gfn_to_memslot(struct kvm *kvm, gfn_t gfn)
+struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn)
 {
 	int i;
 
@@ -931,11 +1137,12 @@
 	}
 	return NULL;
 }
+EXPORT_SYMBOL_GPL(gfn_to_memslot_unaliased);
 
 struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn)
 {
 	gfn = unalias_gfn(kvm, gfn);
-	return __gfn_to_memslot(kvm, gfn);
+	return gfn_to_memslot_unaliased(kvm, gfn);
 }
 
 int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn)
@@ -959,7 +1166,7 @@
 	struct kvm_memory_slot *slot;
 
 	gfn = unalias_gfn(kvm, gfn);
-	slot = __gfn_to_memslot(kvm, gfn);
+	slot = gfn_to_memslot_unaliased(kvm, gfn);
 	if (!slot)
 		return bad_hva();
 	return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE);
@@ -1210,7 +1417,7 @@
 	struct kvm_memory_slot *memslot;
 
 	gfn = unalias_gfn(kvm, gfn);
-	memslot = __gfn_to_memslot(kvm, gfn);
+	memslot = gfn_to_memslot_unaliased(kvm, gfn);
 	if (memslot && memslot->dirty_bitmap) {
 		unsigned long rel_gfn = gfn - memslot->base_gfn;
 
@@ -1295,7 +1502,7 @@
 	return 0;
 }
 
-static const struct file_operations kvm_vcpu_fops = {
+static struct file_operations kvm_vcpu_fops = {
 	.release        = kvm_vcpu_release,
 	.unlocked_ioctl = kvm_vcpu_ioctl,
 	.compat_ioctl   = kvm_vcpu_ioctl,
@@ -1689,7 +1896,7 @@
 	return 0;
 }
 
-static const struct file_operations kvm_vm_fops = {
+static struct file_operations kvm_vm_fops = {
 	.release        = kvm_vm_release,
 	.unlocked_ioctl = kvm_vm_ioctl,
 	.compat_ioctl   = kvm_vm_ioctl,
@@ -1711,6 +1918,18 @@
 	return fd;
 }
 
+static long kvm_dev_ioctl_check_extension_generic(long arg)
+{
+	switch (arg) {
+	case KVM_CAP_USER_MEMORY:
+	case KVM_CAP_DESTROY_MEMORY_REGION_WORKS:
+		return 1;
+	default:
+		break;
+	}
+	return kvm_dev_ioctl_check_extension(arg);
+}
+
 static long kvm_dev_ioctl(struct file *filp,
 			  unsigned int ioctl, unsigned long arg)
 {
@@ -1730,7 +1949,7 @@
 		r = kvm_dev_ioctl_create_vm();
 		break;
 	case KVM_CHECK_EXTENSION:
-		r = kvm_dev_ioctl_check_extension(arg);
+		r = kvm_dev_ioctl_check_extension_generic(arg);
 		break;
 	case KVM_GET_VCPU_MMAP_SIZE:
 		r = -EINVAL;
@@ -1771,9 +1990,9 @@
 {
 	int cpu = raw_smp_processor_id();
 
-	if (cpu_isset(cpu, cpus_hardware_enabled))
+	if (cpumask_test_cpu(cpu, cpus_hardware_enabled))
 		return;
-	cpu_set(cpu, cpus_hardware_enabled);
+	cpumask_set_cpu(cpu, cpus_hardware_enabled);
 	kvm_arch_hardware_enable(NULL);
 }
 
@@ -1781,9 +2000,9 @@
 {
 	int cpu = raw_smp_processor_id();
 
-	if (!cpu_isset(cpu, cpus_hardware_enabled))
+	if (!cpumask_test_cpu(cpu, cpus_hardware_enabled))
 		return;
-	cpu_clear(cpu, cpus_hardware_enabled);
+	cpumask_clear_cpu(cpu, cpus_hardware_enabled);
 	kvm_arch_hardware_disable(NULL);
 }
 
@@ -2017,9 +2236,14 @@
 
 	bad_pfn = page_to_pfn(bad_page);
 
+	if (!alloc_cpumask_var(&cpus_hardware_enabled, GFP_KERNEL)) {
+		r = -ENOMEM;
+		goto out_free_0;
+	}
+
 	r = kvm_arch_hardware_setup();
 	if (r < 0)
-		goto out_free_0;
+		goto out_free_0a;
 
 	for_each_online_cpu(cpu) {
 		smp_call_function_single(cpu,
@@ -2053,6 +2277,8 @@
 	}
 
 	kvm_chardev_ops.owner = module;
+	kvm_vm_fops.owner = module;
+	kvm_vcpu_fops.owner = module;
 
 	r = misc_register(&kvm_dev);
 	if (r) {
@@ -2062,6 +2288,9 @@
 
 	kvm_preempt_ops.sched_in = kvm_sched_in;
 	kvm_preempt_ops.sched_out = kvm_sched_out;
+#ifndef CONFIG_X86
+	msi2intx = 0;
+#endif
 
 	return 0;
 
@@ -2078,6 +2307,8 @@
 	on_each_cpu(hardware_disable, NULL, 1);
 out_free_1:
 	kvm_arch_hardware_unsetup();
+out_free_0a:
+	free_cpumask_var(cpus_hardware_enabled);
 out_free_0:
 	__free_page(bad_page);
 out:
@@ -2101,6 +2332,7 @@
 	kvm_arch_hardware_unsetup();
 	kvm_arch_exit();
 	kvm_exit_debug();
+	free_cpumask_var(cpus_hardware_enabled);
 	__free_page(bad_page);
 }
 EXPORT_SYMBOL_GPL(kvm_exit);
diff --git a/virt/kvm/kvm_trace.c b/virt/kvm/kvm_trace.c
index 41dcc84..f598744 100644
--- a/virt/kvm/kvm_trace.c
+++ b/virt/kvm/kvm_trace.c
@@ -252,6 +252,7 @@
 			struct kvm_trace_probe *p = &kvm_trace_probes[i];
 			marker_probe_unregister(p->name, p->probe_func, p);
 		}
+		marker_synchronize_unregister();
 
 		relay_close(kt->rchan);
 		debugfs_remove(kt->lost_file);