diff --git a/Documentation/scsi/cxgb3i.txt b/Documentation/scsi/cxgb3i.txt
index 8141fa0..7ac8032 100644
--- a/Documentation/scsi/cxgb3i.txt
+++ b/Documentation/scsi/cxgb3i.txt
@@ -4,7 +4,7 @@
 ============
 
 The Chelsio T3 ASIC based Adapters (S310, S320, S302, S304, Mezz cards, etc.
-series of products) supports iSCSI acceleration and iSCSI Direct Data Placement
+series of products) support iSCSI acceleration and iSCSI Direct Data Placement
 (DDP) where the hardware handles the expensive byte touching operations, such
 as CRC computation and verification, and direct DMA to the final host memory
 destination:
@@ -31,9 +31,9 @@
 	  the TCP segments onto the wire. It handles TCP retransmission if
 	  needed.
 
-	  On receving, S3 h/w recovers the iSCSI PDU by reassembling TCP
+	  On receiving, S3 h/w recovers the iSCSI PDU by reassembling TCP
 	  segments, separating the header and data, calculating and verifying
-	  the digests, then forwards the header to the host. The payload data,
+	  the digests, then forwarding the header to the host. The payload data,
 	  if possible, will be directly placed into the pre-posted host DDP
 	  buffer. Otherwise, the payload data will be sent to the host too.
 
@@ -68,9 +68,8 @@
 	sure the ip address is unique in the network.
 
 3. edit /etc/iscsi/iscsid.conf
-   The default setting for MaxRecvDataSegmentLength (131072) is too big,
-   replace "node.conn[0].iscsi.MaxRecvDataSegmentLength" to be a value no
-   bigger than 15360 (for example 8192):
+   The default setting for MaxRecvDataSegmentLength (131072) is too big;
+   replace with a value no bigger than 15360 (for example 8192):
 
 	node.conn[0].iscsi.MaxRecvDataSegmentLength = 8192
 
diff --git a/Documentation/x86/boot.txt b/Documentation/x86/boot.txt
index 1229969..e020366 100644
--- a/Documentation/x86/boot.txt
+++ b/Documentation/x86/boot.txt
@@ -543,7 +543,10 @@
 
   The payload may be compressed. The format of both the compressed and
   uncompressed data should be determined using the standard magic
-  numbers. Currently only gzip compressed ELF is used.
+  numbers.  The currently supported compression formats are gzip
+  (magic numbers 1F 8B or 1F 9E), bzip2 (magic number 42 5A) and LZMA
+  (magic number 5D 00).  The uncompressed payload is currently always ELF
+  (magic number 7F 45 4C 46).
   
 Field name:	payload_length
 Type:		read
diff --git a/MAINTAINERS b/MAINTAINERS
index 59fd2d1..1c2ca1d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2464,7 +2464,7 @@
 
 ISDN SUBSYSTEM
 P:	Karsten Keil
-M:	kkeil@suse.de
+M:	isdn@linux-pingi.de
 L:	isdn4linux@listserv.isdn4linux.de (subscribers-only)
 W:	http://www.isdn4linux.de
 T:	git kernel.org:/pub/scm/linux/kernel/kkeil/isdn-2.6.git
diff --git a/arch/mips/include/asm/seccomp.h b/arch/mips/include/asm/seccomp.h
index 36ed440..a6772e9 100644
--- a/arch/mips/include/asm/seccomp.h
+++ b/arch/mips/include/asm/seccomp.h
@@ -1,6 +1,5 @@
 #ifndef __ASM_SECCOMP_H
 
-#include <linux/thread_info.h>
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/powerpc/include/asm/compat.h b/arch/powerpc/include/asm/compat.h
index d811a8c..4774c2f 100644
--- a/arch/powerpc/include/asm/compat.h
+++ b/arch/powerpc/include/asm/compat.h
@@ -210,5 +210,10 @@
 	compat_ulong_t __unused6;
 };
 
+static inline int is_compat_task(void)
+{
+	return test_thread_flag(TIF_32BIT);
+}
+
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_COMPAT_H */
diff --git a/arch/powerpc/include/asm/seccomp.h b/arch/powerpc/include/asm/seccomp.h
index 853765e..00c1d91 100644
--- a/arch/powerpc/include/asm/seccomp.h
+++ b/arch/powerpc/include/asm/seccomp.h
@@ -1,10 +1,6 @@
 #ifndef _ASM_POWERPC_SECCOMP_H
 #define _ASM_POWERPC_SECCOMP_H
 
-#ifdef __KERNEL__
-#include <linux/thread_info.h>
-#endif
-
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/sparc/include/asm/compat.h b/arch/sparc/include/asm/compat.h
index f260b58..0e70625 100644
--- a/arch/sparc/include/asm/compat.h
+++ b/arch/sparc/include/asm/compat.h
@@ -240,4 +240,9 @@
 	unsigned int	__unused2;
 };
 
+static inline int is_compat_task(void)
+{
+	return test_thread_flag(TIF_32BIT);
+}
+
 #endif /* _ASM_SPARC64_COMPAT_H */
diff --git a/arch/sparc/include/asm/seccomp.h b/arch/sparc/include/asm/seccomp.h
index 7fcd996..adca1bc 100644
--- a/arch/sparc/include/asm/seccomp.h
+++ b/arch/sparc/include/asm/seccomp.h
@@ -1,11 +1,5 @@
 #ifndef _ASM_SECCOMP_H
 
-#include <linux/thread_info.h> /* already defines TIF_32BIT */
-
-#ifndef TIF_32BIT
-#error "unexpected TIF_32BIT on sparc64"
-#endif
-
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index d908432..469f345 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -40,6 +40,9 @@
 	select HAVE_GENERIC_DMA_COHERENT if X86_32
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS
 	select USER_STACKTRACE_SUPPORT
+	select HAVE_KERNEL_GZIP
+	select HAVE_KERNEL_BZIP2
+	select HAVE_KERNEL_LZMA
 
 config ARCH_DEFCONFIG
 	string
diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index 1771c80..3ca4c19 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -4,7 +4,7 @@
 # create a compressed vmlinux image from the original vmlinux
 #
 
-targets := vmlinux vmlinux.bin vmlinux.bin.gz head_$(BITS).o misc.o piggy.o
+targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma head_$(BITS).o misc.o piggy.o
 
 KBUILD_CFLAGS := -m$(BITS) -D__KERNEL__ $(LINUX_INCLUDE) -O2
 KBUILD_CFLAGS += -fno-strict-aliasing -fPIC
@@ -47,18 +47,35 @@
 ifdef CONFIG_RELOCATABLE
 $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin.all FORCE
 	$(call if_changed,gzip)
+$(obj)/vmlinux.bin.bz2: $(obj)/vmlinux.bin.all FORCE
+	$(call if_changed,bzip2)
+$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin.all FORCE
+	$(call if_changed,lzma)
 else
 $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE
 	$(call if_changed,gzip)
+$(obj)/vmlinux.bin.bz2: $(obj)/vmlinux.bin FORCE
+	$(call if_changed,bzip2)
+$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin FORCE
+	$(call if_changed,lzma)
 endif
 LDFLAGS_piggy.o := -r --format binary --oformat elf32-i386 -T
 
 else
+
 $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE
 	$(call if_changed,gzip)
+$(obj)/vmlinux.bin.bz2: $(obj)/vmlinux.bin FORCE
+	$(call if_changed,bzip2)
+$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin FORCE
+	$(call if_changed,lzma)
 
 LDFLAGS_piggy.o := -r --format binary --oformat elf64-x86-64 -T
 endif
 
-$(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE
+suffix_$(CONFIG_KERNEL_GZIP)  = gz
+suffix_$(CONFIG_KERNEL_BZIP2) = bz2
+suffix_$(CONFIG_KERNEL_LZMA)  = lzma
+
+$(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.$(suffix_y) FORCE
 	$(call if_changed,ld)
diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
index da06221..e45be73 100644
--- a/arch/x86/boot/compressed/misc.c
+++ b/arch/x86/boot/compressed/misc.c
@@ -116,71 +116,13 @@
 /*
  * gzip declarations
  */
-
-#define OF(args)	args
 #define STATIC		static
 
 #undef memset
 #undef memcpy
 #define memzero(s, n)	memset((s), 0, (n))
 
-typedef unsigned char	uch;
-typedef unsigned short	ush;
-typedef unsigned long	ulg;
 
-/*
- * Window size must be at least 32k, and a power of two.
- * We don't actually have a window just a huge output buffer,
- * so we report a 2G window size, as that should always be
- * larger than our output buffer:
- */
-#define WSIZE		0x80000000
-
-/* Input buffer: */
-static unsigned char	*inbuf;
-
-/* Sliding window buffer (and final output buffer): */
-static unsigned char	*window;
-
-/* Valid bytes in inbuf: */
-static unsigned		insize;
-
-/* Index of next byte to be processed in inbuf: */
-static unsigned		inptr;
-
-/* Bytes in output buffer: */
-static unsigned		outcnt;
-
-/* gzip flag byte */
-#define ASCII_FLAG	0x01 /* bit 0 set: file probably ASCII text */
-#define CONTINUATION	0x02 /* bit 1 set: continuation of multi-part gz file */
-#define EXTRA_FIELD	0x04 /* bit 2 set: extra field present */
-#define ORIG_NAM	0x08 /* bit 3 set: original file name present */
-#define COMMENT		0x10 /* bit 4 set: file comment present */
-#define ENCRYPTED	0x20 /* bit 5 set: file is encrypted */
-#define RESERVED	0xC0 /* bit 6, 7:  reserved */
-
-#define get_byte()	(inptr < insize ? inbuf[inptr++] : fill_inbuf())
-
-/* Diagnostic functions */
-#ifdef DEBUG
-#  define Assert(cond, msg) do { if (!(cond)) error(msg); } while (0)
-#  define Trace(x)	do { fprintf x; } while (0)
-#  define Tracev(x)	do { if (verbose) fprintf x ; } while (0)
-#  define Tracevv(x)	do { if (verbose > 1) fprintf x ; } while (0)
-#  define Tracec(c, x)	do { if (verbose && (c)) fprintf x ; } while (0)
-#  define Tracecv(c, x)	do { if (verbose > 1 && (c)) fprintf x ; } while (0)
-#else
-#  define Assert(cond, msg)
-#  define Trace(x)
-#  define Tracev(x)
-#  define Tracevv(x)
-#  define Tracec(c, x)
-#  define Tracecv(c, x)
-#endif
-
-static int  fill_inbuf(void);
-static void flush_window(void);
 static void error(char *m);
 
 /*
@@ -189,13 +131,8 @@
 static struct boot_params *real_mode;		/* Pointer to real-mode data */
 static int quiet;
 
-extern unsigned char input_data[];
-extern int input_len;
-
-static long bytes_out;
-
 static void *memset(void *s, int c, unsigned n);
-static void *memcpy(void *dest, const void *src, unsigned n);
+void *memcpy(void *dest, const void *src, unsigned n);
 
 static void __putstr(int, const char *);
 #define putstr(__x)  __putstr(0, __x)
@@ -213,7 +150,17 @@
 static int vidport;
 static int lines, cols;
 
-#include "../../../../lib/inflate.c"
+#ifdef CONFIG_KERNEL_GZIP
+#include "../../../../lib/decompress_inflate.c"
+#endif
+
+#ifdef CONFIG_KERNEL_BZIP2
+#include "../../../../lib/decompress_bunzip2.c"
+#endif
+
+#ifdef CONFIG_KERNEL_LZMA
+#include "../../../../lib/decompress_unlzma.c"
+#endif
 
 static void scroll(void)
 {
@@ -282,7 +229,7 @@
 	return s;
 }
 
-static void *memcpy(void *dest, const void *src, unsigned n)
+void *memcpy(void *dest, const void *src, unsigned n)
 {
 	int i;
 	const char *s = src;
@@ -293,38 +240,6 @@
 	return dest;
 }
 
-/* ===========================================================================
- * Fill the input buffer. This is called only when the buffer is empty
- * and at least one byte is really needed.
- */
-static int fill_inbuf(void)
-{
-	error("ran out of input data");
-	return 0;
-}
-
-/* ===========================================================================
- * Write the output window window[0..outcnt-1] and update crc and bytes_out.
- * (Used for the decompressed data only.)
- */
-static void flush_window(void)
-{
-	/* With my window equal to my output buffer
-	 * I only need to compute the crc here.
-	 */
-	unsigned long c = crc;         /* temporary variable */
-	unsigned n;
-	unsigned char *in, ch;
-
-	in = window;
-	for (n = 0; n < outcnt; n++) {
-		ch = *in++;
-		c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8);
-	}
-	crc = c;
-	bytes_out += (unsigned long)outcnt;
-	outcnt = 0;
-}
 
 static void error(char *x)
 {
@@ -407,12 +322,8 @@
 	lines = real_mode->screen_info.orig_video_lines;
 	cols = real_mode->screen_info.orig_video_cols;
 
-	window = output;		/* Output buffer (Normally at 1M) */
 	free_mem_ptr     = heap;	/* Heap */
 	free_mem_end_ptr = heap + BOOT_HEAP_SIZE;
-	inbuf  = input_data;		/* Input buffer */
-	insize = input_len;
-	inptr  = 0;
 
 #ifdef CONFIG_X86_64
 	if ((unsigned long)output & (__KERNEL_ALIGN - 1))
@@ -430,10 +341,9 @@
 #endif
 #endif
 
-	makecrc();
 	if (!quiet)
 		putstr("\nDecompressing Linux... ");
-	gunzip();
+	decompress(input_data, input_len, NULL, NULL, output, NULL, error);
 	parse_elf(output);
 	if (!quiet)
 		putstr("done.\nBooting the kernel.\n");
diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index a6208dc..4ef949c 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -75,7 +75,14 @@
 #define setup_secondary_clock setup_secondary_APIC_clock
 #endif
 
+#ifdef CONFIG_X86_VSMP
 extern int is_vsmp_box(void);
+#else
+static inline int is_vsmp_box(void)
+{
+	return 0;
+}
+#endif
 extern void xapic_wait_icr_idle(void);
 extern u32 safe_xapic_wait_icr_idle(void);
 extern void xapic_icr_write(u32, u32);
@@ -306,7 +313,7 @@
 	void (*send_IPI_self)(int vector);
 
 	/* wakeup_secondary_cpu */
-	int (*wakeup_cpu)(int apicid, unsigned long start_eip);
+	int (*wakeup_secondary_cpu)(int apicid, unsigned long start_eip);
 
 	int trampoline_phys_low;
 	int trampoline_phys_high;
@@ -324,8 +331,21 @@
 	u32 (*safe_wait_icr_idle)(void);
 };
 
+/*
+ * Pointer to the local APIC driver in use on this system (there's
+ * always just one such driver in use - the kernel decides via an
+ * early probing process which one it picks - and then sticks to it):
+ */
 extern struct apic *apic;
 
+/*
+ * APIC functionality to boot other CPUs - only used on SMP:
+ */
+#ifdef CONFIG_SMP
+extern atomic_t init_deasserted;
+extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip);
+#endif
+
 static inline u32 apic_read(u32 reg)
 {
 	return apic->read(reg);
@@ -384,9 +404,7 @@
 #define DEFAULT_TRAMPOLINE_PHYS_LOW		0x467
 #define DEFAULT_TRAMPOLINE_PHYS_HIGH		0x469
 
-#ifdef CONFIG_X86_32
-extern void es7000_update_apic_to_cluster(void);
-#else
+#ifdef CONFIG_X86_64
 extern struct apic apic_flat;
 extern struct apic apic_physflat;
 extern struct apic apic_x2apic_cluster;
diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
index dd61616..6526cf0 100644
--- a/arch/x86/include/asm/boot.h
+++ b/arch/x86/include/asm/boot.h
@@ -10,17 +10,31 @@
 #define EXTENDED_VGA	0xfffe		/* 80x50 mode */
 #define ASK_VGA		0xfffd		/* ask for it at bootup */
 
+#ifdef __KERNEL__
+
 /* Physical address where kernel should be loaded. */
 #define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \
 				+ (CONFIG_PHYSICAL_ALIGN - 1)) \
 				& ~(CONFIG_PHYSICAL_ALIGN - 1))
 
+#ifdef CONFIG_KERNEL_BZIP2
+#define BOOT_HEAP_SIZE             0x400000
+#else /* !CONFIG_KERNEL_BZIP2 */
+
 #ifdef CONFIG_X86_64
 #define BOOT_HEAP_SIZE	0x7000
-#define BOOT_STACK_SIZE	0x4000
 #else
 #define BOOT_HEAP_SIZE	0x4000
+#endif
+
+#endif /* !CONFIG_KERNEL_BZIP2 */
+
+#ifdef CONFIG_X86_64
+#define BOOT_STACK_SIZE	0x4000
+#else
 #define BOOT_STACK_SIZE	0x1000
 #endif
 
+#endif /* __KERNEL__ */
+
 #endif /* _ASM_X86_BOOT_H */
diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h
index 23696d4..dca8f03 100644
--- a/arch/x86/include/asm/fixmap.h
+++ b/arch/x86/include/asm/fixmap.h
@@ -1,11 +1,155 @@
+/*
+ * fixmap.h: compile-time virtual memory allocation
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1998 Ingo Molnar
+ *
+ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
+ * x86_32 and x86_64 integration by Gustavo F. Padovan, February 2009
+ */
+
 #ifndef _ASM_X86_FIXMAP_H
 #define _ASM_X86_FIXMAP_H
 
+#ifndef __ASSEMBLY__
+#include <linux/kernel.h>
+#include <asm/acpi.h>
+#include <asm/apicdef.h>
+#include <asm/page.h>
 #ifdef CONFIG_X86_32
-# include "fixmap_32.h"
+#include <linux/threads.h>
+#include <asm/kmap_types.h>
 #else
-# include "fixmap_64.h"
+#include <asm/vsyscall.h>
+#ifdef CONFIG_EFI
+#include <asm/efi.h>
 #endif
+#endif
+
+/*
+ * We can't declare FIXADDR_TOP as variable for x86_64 because vsyscall
+ * uses fixmaps that relies on FIXADDR_TOP for proper address calculation.
+ * Because of this, FIXADDR_TOP x86 integration was left as later work.
+ */
+#ifdef CONFIG_X86_32
+/* used by vmalloc.c, vsyscall.lds.S.
+ *
+ * Leave one empty page between vmalloc'ed areas and
+ * the start of the fixmap.
+ */
+extern unsigned long __FIXADDR_TOP;
+#define FIXADDR_TOP	((unsigned long)__FIXADDR_TOP)
+
+#define FIXADDR_USER_START     __fix_to_virt(FIX_VDSO)
+#define FIXADDR_USER_END       __fix_to_virt(FIX_VDSO - 1)
+#else
+#define FIXADDR_TOP	(VSYSCALL_END-PAGE_SIZE)
+
+/* Only covers 32bit vsyscalls currently. Need another set for 64bit. */
+#define FIXADDR_USER_START	((unsigned long)VSYSCALL32_VSYSCALL)
+#define FIXADDR_USER_END	(FIXADDR_USER_START + PAGE_SIZE)
+#endif
+
+
+/*
+ * Here we define all the compile-time 'special' virtual
+ * addresses. The point is to have a constant address at
+ * compile time, but to set the physical address only
+ * in the boot process.
+ * for x86_32: We allocate these special addresses
+ * from the end of virtual memory (0xfffff000) backwards.
+ * Also this lets us do fail-safe vmalloc(), we
+ * can guarantee that these special addresses and
+ * vmalloc()-ed addresses never overlap.
+ *
+ * These 'compile-time allocated' memory buffers are
+ * fixed-size 4k pages (or larger if used with an increment
+ * higher than 1). Use set_fixmap(idx,phys) to associate
+ * physical memory with fixmap indices.
+ *
+ * TLB entries of such buffers will not be flushed across
+ * task switches.
+ */
+enum fixed_addresses {
+#ifdef CONFIG_X86_32
+	FIX_HOLE,
+	FIX_VDSO,
+#else
+	VSYSCALL_LAST_PAGE,
+	VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE
+			    + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
+	VSYSCALL_HPET,
+#endif
+	FIX_DBGP_BASE,
+	FIX_EARLYCON_MEM_BASE,
+#ifdef CONFIG_X86_LOCAL_APIC
+	FIX_APIC_BASE,	/* local (CPU) APIC) -- required for SMP or not */
+#endif
+#ifdef CONFIG_X86_IO_APIC
+	FIX_IO_APIC_BASE_0,
+	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS - 1,
+#endif
+#ifdef CONFIG_X86_64
+#ifdef CONFIG_EFI
+	FIX_EFI_IO_MAP_LAST_PAGE,
+	FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE
+				  + MAX_EFI_IO_PAGES - 1,
+#endif
+#endif
+#ifdef CONFIG_X86_VISWS_APIC
+	FIX_CO_CPU,	/* Cobalt timer */
+	FIX_CO_APIC,	/* Cobalt APIC Redirection Table */
+	FIX_LI_PCIA,	/* Lithium PCI Bridge A */
+	FIX_LI_PCIB,	/* Lithium PCI Bridge B */
+#endif
+#ifdef CONFIG_X86_F00F_BUG
+	FIX_F00F_IDT,	/* Virtual mapping for IDT */
+#endif
+#ifdef CONFIG_X86_CYCLONE_TIMER
+	FIX_CYCLONE_TIMER, /*cyclone timer register*/
+#endif
+#ifdef CONFIG_X86_32
+	FIX_KMAP_BEGIN,	/* reserved pte's for temporary kernel mappings */
+	FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
+#ifdef CONFIG_PCI_MMCONFIG
+	FIX_PCIE_MCFG,
+#endif
+#endif
+#ifdef CONFIG_PARAVIRT
+	FIX_PARAVIRT_BOOTMAP,
+#endif
+	__end_of_permanent_fixed_addresses,
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+	FIX_OHCI1394_BASE,
+#endif
+	/*
+	 * 256 temporary boot-time mappings, used by early_ioremap(),
+	 * before ioremap() is functional.
+	 *
+	 * We round it up to the next 256 pages boundary so that we
+	 * can have a single pgd entry and a single pte table:
+	 */
+#define NR_FIX_BTMAPS		64
+#define FIX_BTMAPS_SLOTS	4
+	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
+			(__end_of_permanent_fixed_addresses & 255),
+	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
+#ifdef CONFIG_X86_32
+	FIX_WP_TEST,
+#endif
+	__end_of_fixed_addresses
+};
+
+
+extern void reserve_top_address(unsigned long reserve);
+
+#define FIXADDR_SIZE	(__end_of_permanent_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_BOOT_SIZE	(__end_of_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_START		(FIXADDR_TOP - FIXADDR_SIZE)
+#define FIXADDR_BOOT_START	(FIXADDR_TOP - FIXADDR_BOOT_SIZE)
 
 extern int fixmaps_set;
 
@@ -69,4 +213,5 @@
 	BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
 	return __virt_to_fix(vaddr);
 }
+#endif /* !__ASSEMBLY__ */
 #endif /* _ASM_X86_FIXMAP_H */
diff --git a/arch/x86/include/asm/fixmap_32.h b/arch/x86/include/asm/fixmap_32.h
deleted file mode 100644
index 047d9ba..0000000
--- a/arch/x86/include/asm/fixmap_32.h
+++ /dev/null
@@ -1,115 +0,0 @@
-/*
- * fixmap.h: compile-time virtual memory allocation
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1998 Ingo Molnar
- *
- * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
- */
-
-#ifndef _ASM_X86_FIXMAP_32_H
-#define _ASM_X86_FIXMAP_32_H
-
-
-/* used by vmalloc.c, vsyscall.lds.S.
- *
- * Leave one empty page between vmalloc'ed areas and
- * the start of the fixmap.
- */
-extern unsigned long __FIXADDR_TOP;
-#define FIXADDR_USER_START     __fix_to_virt(FIX_VDSO)
-#define FIXADDR_USER_END       __fix_to_virt(FIX_VDSO - 1)
-
-#ifndef __ASSEMBLY__
-#include <linux/kernel.h>
-#include <asm/acpi.h>
-#include <asm/apicdef.h>
-#include <asm/page.h>
-#include <linux/threads.h>
-#include <asm/kmap_types.h>
-
-/*
- * Here we define all the compile-time 'special' virtual
- * addresses. The point is to have a constant address at
- * compile time, but to set the physical address only
- * in the boot process. We allocate these special addresses
- * from the end of virtual memory (0xfffff000) backwards.
- * Also this lets us do fail-safe vmalloc(), we
- * can guarantee that these special addresses and
- * vmalloc()-ed addresses never overlap.
- *
- * these 'compile-time allocated' memory buffers are
- * fixed-size 4k pages. (or larger if used with an increment
- * highger than 1) use fixmap_set(idx,phys) to associate
- * physical memory with fixmap indices.
- *
- * TLB entries of such buffers will not be flushed across
- * task switches.
- */
-enum fixed_addresses {
-	FIX_HOLE,
-	FIX_VDSO,
-	FIX_DBGP_BASE,
-	FIX_EARLYCON_MEM_BASE,
-#ifdef CONFIG_X86_LOCAL_APIC
-	FIX_APIC_BASE,	/* local (CPU) APIC) -- required for SMP or not */
-#endif
-#ifdef CONFIG_X86_IO_APIC
-	FIX_IO_APIC_BASE_0,
-	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
-#endif
-#ifdef CONFIG_X86_VISWS_APIC
-	FIX_CO_CPU,	/* Cobalt timer */
-	FIX_CO_APIC,	/* Cobalt APIC Redirection Table */
-	FIX_LI_PCIA,	/* Lithium PCI Bridge A */
-	FIX_LI_PCIB,	/* Lithium PCI Bridge B */
-#endif
-#ifdef CONFIG_X86_F00F_BUG
-	FIX_F00F_IDT,	/* Virtual mapping for IDT */
-#endif
-#ifdef CONFIG_X86_CYCLONE_TIMER
-	FIX_CYCLONE_TIMER, /*cyclone timer register*/
-#endif
-	FIX_KMAP_BEGIN,	/* reserved pte's for temporary kernel mappings */
-	FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
-#ifdef CONFIG_PCI_MMCONFIG
-	FIX_PCIE_MCFG,
-#endif
-#ifdef CONFIG_PARAVIRT
-	FIX_PARAVIRT_BOOTMAP,
-#endif
-	__end_of_permanent_fixed_addresses,
-	/*
-	 * 256 temporary boot-time mappings, used by early_ioremap(),
-	 * before ioremap() is functional.
-	 *
-	 * We round it up to the next 256 pages boundary so that we
-	 * can have a single pgd entry and a single pte table:
-	 */
-#define NR_FIX_BTMAPS		64
-#define FIX_BTMAPS_SLOTS	4
-	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
-			(__end_of_permanent_fixed_addresses & 255),
-	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
-	FIX_WP_TEST,
-#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
-	FIX_OHCI1394_BASE,
-#endif
-	__end_of_fixed_addresses
-};
-
-extern void reserve_top_address(unsigned long reserve);
-
-
-#define FIXADDR_TOP	((unsigned long)__FIXADDR_TOP)
-
-#define __FIXADDR_SIZE	(__end_of_permanent_fixed_addresses << PAGE_SHIFT)
-#define __FIXADDR_BOOT_SIZE	(__end_of_fixed_addresses << PAGE_SHIFT)
-#define FIXADDR_START		(FIXADDR_TOP - __FIXADDR_SIZE)
-#define FIXADDR_BOOT_START	(FIXADDR_TOP - __FIXADDR_BOOT_SIZE)
-
-#endif /* !__ASSEMBLY__ */
-#endif /* _ASM_X86_FIXMAP_32_H */
diff --git a/arch/x86/include/asm/fixmap_64.h b/arch/x86/include/asm/fixmap_64.h
deleted file mode 100644
index 298d9ba..0000000
--- a/arch/x86/include/asm/fixmap_64.h
+++ /dev/null
@@ -1,79 +0,0 @@
-/*
- * fixmap.h: compile-time virtual memory allocation
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1998 Ingo Molnar
- */
-
-#ifndef _ASM_X86_FIXMAP_64_H
-#define _ASM_X86_FIXMAP_64_H
-
-#include <linux/kernel.h>
-#include <asm/acpi.h>
-#include <asm/apicdef.h>
-#include <asm/page.h>
-#include <asm/vsyscall.h>
-#include <asm/efi.h>
-
-/*
- * Here we define all the compile-time 'special' virtual
- * addresses. The point is to have a constant address at
- * compile time, but to set the physical address only
- * in the boot process.
- *
- * These 'compile-time allocated' memory buffers are
- * fixed-size 4k pages (or larger if used with an increment
- * higher than 1). Use set_fixmap(idx,phys) to associate
- * physical memory with fixmap indices.
- *
- * TLB entries of such buffers will not be flushed across
- * task switches.
- */
-
-enum fixed_addresses {
-	VSYSCALL_LAST_PAGE,
-	VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE
-			    + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
-	VSYSCALL_HPET,
-	FIX_DBGP_BASE,
-	FIX_EARLYCON_MEM_BASE,
-	FIX_APIC_BASE,	/* local (CPU) APIC) -- required for SMP or not */
-	FIX_IO_APIC_BASE_0,
-	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS - 1,
-	FIX_EFI_IO_MAP_LAST_PAGE,
-	FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE
-				  + MAX_EFI_IO_PAGES - 1,
-#ifdef CONFIG_PARAVIRT
-	FIX_PARAVIRT_BOOTMAP,
-#endif
-	__end_of_permanent_fixed_addresses,
-#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
-	FIX_OHCI1394_BASE,
-#endif
-	/*
-	 * 256 temporary boot-time mappings, used by early_ioremap(),
-	 * before ioremap() is functional.
-	 *
-	 * We round it up to the next 256 pages boundary so that we
-	 * can have a single pgd entry and a single pte table:
-	 */
-#define NR_FIX_BTMAPS		64
-#define FIX_BTMAPS_SLOTS	4
-	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
-			(__end_of_permanent_fixed_addresses & 255),
-	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
-	__end_of_fixed_addresses
-};
-
-#define FIXADDR_TOP	(VSYSCALL_END-PAGE_SIZE)
-#define FIXADDR_SIZE	(__end_of_fixed_addresses << PAGE_SHIFT)
-#define FIXADDR_START	(FIXADDR_TOP - FIXADDR_SIZE)
-
-/* Only covers 32bit vsyscalls currently. Need another set for 64bit. */
-#define FIXADDR_USER_START	((unsigned long)VSYSCALL32_VSYSCALL)
-#define FIXADDR_USER_END	(FIXADDR_USER_START + PAGE_SIZE)
-
-#endif /* _ASM_X86_FIXMAP_64_H */
diff --git a/arch/x86/include/asm/numa_32.h b/arch/x86/include/asm/numa_32.h
index e9f5db7..a372290 100644
--- a/arch/x86/include/asm/numa_32.h
+++ b/arch/x86/include/asm/numa_32.h
@@ -4,8 +4,12 @@
 extern int pxm_to_nid(int pxm);
 extern void numa_remove_cpu(int cpu);
 
-#ifdef CONFIG_NUMA
+#ifdef CONFIG_HIGHMEM
 extern void set_highmem_pages_init(void);
+#else
+static inline void set_highmem_pages_init(void)
+{
+}
 #endif
 
 #endif /* _ASM_X86_NUMA_32_H */
diff --git a/arch/x86/include/asm/seccomp_32.h b/arch/x86/include/asm/seccomp_32.h
index a6ad87b..b811d6f 100644
--- a/arch/x86/include/asm/seccomp_32.h
+++ b/arch/x86/include/asm/seccomp_32.h
@@ -1,12 +1,6 @@
 #ifndef _ASM_X86_SECCOMP_32_H
 #define _ASM_X86_SECCOMP_32_H
 
-#include <linux/thread_info.h>
-
-#ifdef TIF_32BIT
-#error "unexpected TIF_32BIT on i386"
-#endif
-
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/x86/include/asm/seccomp_64.h b/arch/x86/include/asm/seccomp_64.h
index 4171bb7..84ec1bd 100644
--- a/arch/x86/include/asm/seccomp_64.h
+++ b/arch/x86/include/asm/seccomp_64.h
@@ -1,14 +1,6 @@
 #ifndef _ASM_X86_SECCOMP_64_H
 #define _ASM_X86_SECCOMP_64_H
 
-#include <linux/thread_info.h>
-
-#ifdef TIF_32BIT
-#error "unexpected TIF_32BIT on x86_64"
-#else
-#define TIF_32BIT TIF_IA32
-#endif
-
 #include <linux/unistd.h>
 #include <asm/ia32_unistd.h>
 
diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h
index 66801cb..05c6f6b 100644
--- a/arch/x86/include/asm/setup.h
+++ b/arch/x86/include/asm/setup.h
@@ -31,7 +31,6 @@
 	void (*smp_read_mpc_oem)(struct mpc_oemtable *oemtable,
 				unsigned short oemsize);
 	int (*setup_ioapic_ids)(void);
-	int (*update_apic)(void);
 };
 
 extern void x86_quirk_pre_intr_init(void);
@@ -65,7 +64,11 @@
 #include <asm/bootparam.h>
 
 /* Interrupt control for vSMPowered x86_64 systems */
+#ifdef CONFIG_X86_VSMP
 void vsmp_init(void);
+#else
+static inline void vsmp_init(void) { }
+#endif
 
 void setup_bios_corruption_check(void);
 
@@ -77,8 +80,6 @@
 static inline int is_visws_box(void) { return 0; }
 #endif
 
-extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip);
-extern int wakeup_secondary_cpu_via_init(int apicid, unsigned long start_eip);
 extern struct x86_quirks *x86_quirks;
 extern unsigned long saved_video_mode;
 
diff --git a/arch/x86/include/asm/uaccess_32.h b/arch/x86/include/asm/uaccess_32.h
index a0ba613..5e06259 100644
--- a/arch/x86/include/asm/uaccess_32.h
+++ b/arch/x86/include/asm/uaccess_32.h
@@ -157,7 +157,7 @@
 }
 
 static __always_inline unsigned long __copy_from_user_nocache(void *to,
-		const void __user *from, unsigned long n, unsigned long total)
+				const void __user *from, unsigned long n)
 {
 	might_fault();
 	if (__builtin_constant_p(n)) {
@@ -180,7 +180,7 @@
 
 static __always_inline unsigned long
 __copy_from_user_inatomic_nocache(void *to, const void __user *from,
-				  unsigned long n, unsigned long total)
+				  unsigned long n)
 {
        return __copy_from_user_ll_nocache_nozero(to, from, n);
 }
diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h
index dcaa0404..8cc6873 100644
--- a/arch/x86/include/asm/uaccess_64.h
+++ b/arch/x86/include/asm/uaccess_64.h
@@ -188,29 +188,18 @@
 extern long __copy_user_nocache(void *dst, const void __user *src,
 				unsigned size, int zerorest);
 
-static inline int __copy_from_user_nocache(void *dst, const void __user *src,
-				   unsigned size, unsigned long total)
+static inline int
+__copy_from_user_nocache(void *dst, const void __user *src, unsigned size)
 {
 	might_sleep();
-	/*
-	 * In practice this limit means that large file write()s
-	 * which get chunked to 4K copies get handled via
-	 * non-temporal stores here. Smaller writes get handled
-	 * via regular __copy_from_user():
-	 */
-	if (likely(total >= PAGE_SIZE))
-		return __copy_user_nocache(dst, src, size, 1);
-	else
-		return __copy_from_user(dst, src, size);
+	return __copy_user_nocache(dst, src, size, 1);
 }
 
-static inline int __copy_from_user_inatomic_nocache(void *dst,
-	    const void __user *src, unsigned size, unsigned total)
+static inline int
+__copy_from_user_inatomic_nocache(void *dst, const void __user *src,
+				  unsigned size)
 {
-	if (likely(total >= PAGE_SIZE))
-		return __copy_user_nocache(dst, src, size, 0);
-	else
-		return __copy_from_user_inatomic(dst, src, size);
+	return __copy_user_nocache(dst, src, size, 0);
 }
 
 unsigned long
diff --git a/arch/x86/include/asm/uv/uv.h b/arch/x86/include/asm/uv/uv.h
index 8242bf9..c0a01b5 100644
--- a/arch/x86/include/asm/uv/uv.h
+++ b/arch/x86/include/asm/uv/uv.h
@@ -12,7 +12,6 @@
 extern int is_uv_system(void);
 extern void uv_cpu_init(void);
 extern void uv_system_init(void);
-extern int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip);
 extern const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask,
 						 struct mm_struct *mm,
 						 unsigned long va,
@@ -24,8 +23,6 @@
 static inline int is_uv_system(void)	{ return 0; }
 static inline void uv_cpu_init(void)	{ }
 static inline void uv_system_init(void)	{ }
-static inline int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
-{ return 1; }
 static inline const struct cpumask *
 uv_flush_tlb_others(const struct cpumask *cpumask, struct mm_struct *mm,
 		    unsigned long va, unsigned int cpu)
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index de5657c..95f216b 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -70,7 +70,7 @@
 obj-$(CONFIG_KEXEC)		+= machine_kexec_$(BITS).o
 obj-$(CONFIG_KEXEC)		+= relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_$(BITS).o
-obj-y				+= vsmp_64.o
+obj-$(CONFIG_X86_VSMP)		+= vsmp_64.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_MODULES)		+= module_$(BITS).o
 obj-$(CONFIG_EFI) 		+= efi.o efi_$(BITS).o efi_stub_$(BITS).o
diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c
index 3b00299..f933822 100644
--- a/arch/x86/kernel/apic/apic_flat_64.c
+++ b/arch/x86/kernel/apic/apic_flat_64.c
@@ -222,7 +222,6 @@
 	.send_IPI_all			= flat_send_IPI_all,
 	.send_IPI_self			= apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
@@ -373,7 +372,6 @@
 	.send_IPI_all			= physflat_send_IPI_all,
 	.send_IPI_self			= apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c
index 0b10933..d806eca 100644
--- a/arch/x86/kernel/apic/bigsmp_32.c
+++ b/arch/x86/kernel/apic/bigsmp_32.c
@@ -16,17 +16,17 @@
 #include <asm/apic.h>
 #include <asm/ipi.h>
 
-static inline unsigned bigsmp_get_apic_id(unsigned long x)
+static unsigned bigsmp_get_apic_id(unsigned long x)
 {
 	return (x >> 24) & 0xFF;
 }
 
-static inline int bigsmp_apic_id_registered(void)
+static int bigsmp_apic_id_registered(void)
 {
 	return 1;
 }
 
-static inline const cpumask_t *bigsmp_target_cpus(void)
+static const cpumask_t *bigsmp_target_cpus(void)
 {
 #ifdef CONFIG_SMP
 	return &cpu_online_map;
@@ -35,13 +35,12 @@
 #endif
 }
 
-static inline unsigned long
-bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid)
+static unsigned long bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid)
 {
 	return 0;
 }
 
-static inline unsigned long bigsmp_check_apicid_present(int bit)
+static unsigned long bigsmp_check_apicid_present(int bit)
 {
 	return 1;
 }
@@ -64,7 +63,7 @@
  * an APIC.  See e.g. "AP-388 82489DX User's Manual" (Intel
  * document number 292116).  So here it goes...
  */
-static inline void bigsmp_init_apic_ldr(void)
+static void bigsmp_init_apic_ldr(void)
 {
 	unsigned long val;
 	int cpu = smp_processor_id();
@@ -74,19 +73,19 @@
 	apic_write(APIC_LDR, val);
 }
 
-static inline void bigsmp_setup_apic_routing(void)
+static void bigsmp_setup_apic_routing(void)
 {
 	printk(KERN_INFO
 		"Enabling APIC mode:  Physflat.  Using %d I/O APICs\n",
 		nr_ioapics);
 }
 
-static inline int bigsmp_apicid_to_node(int logical_apicid)
+static int bigsmp_apicid_to_node(int logical_apicid)
 {
 	return apicid_2_node[hard_smp_processor_id()];
 }
 
-static inline int bigsmp_cpu_present_to_apicid(int mps_cpu)
+static int bigsmp_cpu_present_to_apicid(int mps_cpu)
 {
 	if (mps_cpu < nr_cpu_ids)
 		return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu);
@@ -94,7 +93,7 @@
 	return BAD_APICID;
 }
 
-static inline physid_mask_t bigsmp_apicid_to_cpu_present(int phys_apicid)
+static physid_mask_t bigsmp_apicid_to_cpu_present(int phys_apicid)
 {
 	return physid_mask_of_physid(phys_apicid);
 }
@@ -107,29 +106,24 @@
 	return cpu_physical_id(cpu);
 }
 
-static inline physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map)
+static physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map)
 {
 	/* For clustered we don't have a good way to do this yet - hack */
 	return physids_promote(0xFFL);
 }
 
-static inline void bigsmp_setup_portio_remap(void)
-{
-}
-
-static inline int bigsmp_check_phys_apicid_present(int boot_cpu_physical_apicid)
+static int bigsmp_check_phys_apicid_present(int boot_cpu_physical_apicid)
 {
 	return 1;
 }
 
 /* As we are using single CPU as destination, pick only one CPU here */
-static inline unsigned int bigsmp_cpu_mask_to_apicid(const cpumask_t *cpumask)
+static unsigned int bigsmp_cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
 	return bigsmp_cpu_to_logical_apicid(first_cpu(*cpumask));
 }
 
-static inline unsigned int
-bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+static unsigned int bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
 			      const struct cpumask *andmask)
 {
 	int cpu;
@@ -148,7 +142,7 @@
 	return BAD_APICID;
 }
 
-static inline int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
+static int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
 {
 	return cpuid_apic >> index_msb;
 }
@@ -158,12 +152,12 @@
 	default_send_IPI_mask_sequence_phys(mask, vector);
 }
 
-static inline void bigsmp_send_IPI_allbutself(int vector)
+static void bigsmp_send_IPI_allbutself(int vector)
 {
 	default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector);
 }
 
-static inline void bigsmp_send_IPI_all(int vector)
+static void bigsmp_send_IPI_all(int vector)
 {
 	bigsmp_send_IPI_mask(cpu_online_mask, vector);
 }
@@ -256,7 +250,6 @@
 	.send_IPI_all			= bigsmp_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c
index 320f2d2..19588f2 100644
--- a/arch/x86/kernel/apic/es7000_32.c
+++ b/arch/x86/kernel/apic/es7000_32.c
@@ -163,22 +163,17 @@
 	return 0;
 }
 
-static int __init es7000_update_apic(void)
+static int es7000_apic_is_cluster(void)
 {
-	apic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
-
 	/* MPENTIUMIII */
 	if (boot_cpu_data.x86 == 6 &&
-	    (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11)) {
-		es7000_update_apic_to_cluster();
-		apic->wait_for_init_deassert = NULL;
-		apic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
-	}
+	    (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11))
+		return 1;
 
 	return 0;
 }
 
-static void __init setup_unisys(void)
+static void setup_unisys(void)
 {
 	/*
 	 * Determine the generation of the ES7000 currently running.
@@ -192,14 +187,12 @@
 	else
 		es7000_plat = ES7000_CLASSIC;
 	ioapic_renumber_irq = es7000_rename_gsi;
-
-	x86_quirks->update_apic = es7000_update_apic;
 }
 
 /*
  * Parse the OEM Table:
  */
-static int __init parse_unisys_oem(char *oemptr)
+static int parse_unisys_oem(char *oemptr)
 {
 	int			i;
 	int 			success = 0;
@@ -261,7 +254,7 @@
 }
 
 #ifdef CONFIG_ACPI
-static int __init find_unisys_acpi_oem_table(unsigned long *oem_addr)
+static int find_unisys_acpi_oem_table(unsigned long *oem_addr)
 {
 	struct acpi_table_header *header = NULL;
 	struct es7000_oem_table *table;
@@ -292,7 +285,7 @@
 	return 0;
 }
 
-static void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr)
+static void unmap_unisys_acpi_oem_table(unsigned long oem_addr)
 {
 	if (!oem_addr)
 		return;
@@ -310,8 +303,10 @@
 	return 0;
 }
 
+static int es7000_acpi_ret;
+
 /* Hook from generic ACPI tables.c */
-static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 {
 	unsigned long oem_addr = 0;
 	int check_dsdt;
@@ -332,10 +327,26 @@
 		 */
 		unmap_unisys_acpi_oem_table(oem_addr);
 	}
-	return ret;
+
+	es7000_acpi_ret = ret;
+
+	return ret && !es7000_apic_is_cluster();
 }
+
+static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id)
+{
+	int ret = es7000_acpi_ret;
+
+	return ret && es7000_apic_is_cluster();
+}
+
 #else /* !CONFIG_ACPI: */
-static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	return 0;
+}
+
+static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id)
 {
 	return 0;
 }
@@ -349,8 +360,7 @@
 		rep_nop();
 }
 
-static int __init
-es7000_mip_write(struct mip_reg *mip_reg)
+static int es7000_mip_write(struct mip_reg *mip_reg)
 {
 	int status = 0;
 	int spin;
@@ -383,7 +393,7 @@
 	return status;
 }
 
-static void __init es7000_enable_apic_mode(void)
+static void es7000_enable_apic_mode(void)
 {
 	struct mip_reg es7000_mip_reg;
 	int mip_status;
@@ -416,11 +426,8 @@
 
 static void es7000_wait_for_init_deassert(atomic_t *deassert)
 {
-#ifndef CONFIG_ES7000_CLUSTERED_APIC
 	while (!atomic_read(deassert))
 		cpu_relax();
-#endif
-	return;
 }
 
 static unsigned int es7000_get_apic_id(unsigned long x)
@@ -565,72 +572,24 @@
 	return 1;
 }
 
-static unsigned int
-es7000_cpu_mask_to_apicid_cluster(const struct cpumask *cpumask)
-{
-	int cpus_found = 0;
-	int num_bits_set;
-	int apicid;
-	int cpu;
-
-	num_bits_set = cpumask_weight(cpumask);
-	/* Return id to all */
-	if (num_bits_set == nr_cpu_ids)
-		return 0xFF;
-	/*
-	 * 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 = es7000_cpu_to_logical_apicid(cpu);
-
-	while (cpus_found < num_bits_set) {
-		if (cpumask_test_cpu(cpu, cpumask)) {
-			int new_apicid = es7000_cpu_to_logical_apicid(cpu);
-
-			if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-				WARN(1, "Not a valid mask!");
-
-				return 0xFF;
-			}
-			apicid = new_apicid;
-			cpus_found++;
-		}
-		cpu++;
-	}
-	return apicid;
-}
-
 static unsigned int es7000_cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
-	int cpus_found = 0;
-	int num_bits_set;
-	int apicid;
-	int cpu;
+	unsigned int round = 0;
+	int cpu, uninitialized_var(apicid);
 
-	num_bits_set = cpus_weight(*cpumask);
-	/* Return id to all */
-	if (num_bits_set == nr_cpu_ids)
-		return es7000_cpu_to_logical_apicid(0);
 	/*
-	 * 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():
+	 * The cpus in the mask must all be on the apic cluster.
 	 */
-	cpu = first_cpu(*cpumask);
-	apicid = es7000_cpu_to_logical_apicid(cpu);
-	while (cpus_found < num_bits_set) {
-		if (cpu_isset(cpu, *cpumask)) {
-			int new_apicid = es7000_cpu_to_logical_apicid(cpu);
+	for_each_cpu(cpu, cpumask) {
+		int new_apicid = es7000_cpu_to_logical_apicid(cpu);
 
-			if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-				printk("%s: Not a valid mask!\n", __func__);
+		if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
+			WARN(1, "Not a valid mask!");
 
-				return es7000_cpu_to_logical_apicid(0);
-			}
-			apicid = new_apicid;
-			cpus_found++;
+			return BAD_APICID;
 		}
-		cpu++;
+		apicid = new_apicid;
+		round++;
 	}
 	return apicid;
 }
@@ -659,37 +618,103 @@
 	return cpuid_apic >> index_msb;
 }
 
-void __init es7000_update_apic_to_cluster(void)
-{
-	apic->target_cpus = target_cpus_cluster;
-	apic->irq_delivery_mode = dest_LowestPrio;
-	/* logical delivery broadcast to all procs: */
-	apic->irq_dest_mode = 1;
-
-	apic->init_apic_ldr = es7000_init_apic_ldr_cluster;
-
-	apic->cpu_mask_to_apicid = es7000_cpu_mask_to_apicid_cluster;
-}
-
 static int probe_es7000(void)
 {
 	/* probed later in mptable/ACPI hooks */
 	return 0;
 }
 
-static __init int
-es7000_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid)
+static int es7000_mps_ret;
+static int es7000_mps_oem_check(struct mpc_table *mpc, char *oem,
+		char *productid)
 {
+	int ret = 0;
+
 	if (mpc->oemptr) {
 		struct mpc_oemtable *oem_table =
 			(struct mpc_oemtable *)mpc->oemptr;
 
 		if (!strncmp(oem, "UNISYS", 6))
-			return parse_unisys_oem((char *)oem_table);
+			ret = parse_unisys_oem((char *)oem_table);
 	}
-	return 0;
+
+	es7000_mps_ret = ret;
+
+	return ret && !es7000_apic_is_cluster();
 }
 
+static int es7000_mps_oem_check_cluster(struct mpc_table *mpc, char *oem,
+		char *productid)
+{
+	int ret = es7000_mps_ret;
+
+	return ret && es7000_apic_is_cluster();
+}
+
+struct apic apic_es7000_cluster = {
+
+	.name				= "es7000",
+	.probe				= probe_es7000,
+	.acpi_madt_oem_check		= es7000_acpi_madt_oem_check_cluster,
+	.apic_id_registered		= es7000_apic_id_registered,
+
+	.irq_delivery_mode		= dest_LowestPrio,
+	/* logical delivery broadcast to all procs: */
+	.irq_dest_mode			= 1,
+
+	.target_cpus			= target_cpus_cluster,
+	.disable_esr			= 1,
+	.dest_logical			= 0,
+	.check_apicid_used		= es7000_check_apicid_used,
+	.check_apicid_present		= es7000_check_apicid_present,
+
+	.vector_allocation_domain	= es7000_vector_allocation_domain,
+	.init_apic_ldr			= es7000_init_apic_ldr_cluster,
+
+	.ioapic_phys_id_map		= es7000_ioapic_phys_id_map,
+	.setup_apic_routing		= es7000_setup_apic_routing,
+	.multi_timer_check		= NULL,
+	.apicid_to_node			= es7000_apicid_to_node,
+	.cpu_to_logical_apicid		= es7000_cpu_to_logical_apicid,
+	.cpu_present_to_apicid		= es7000_cpu_present_to_apicid,
+	.apicid_to_cpu_present		= es7000_apicid_to_cpu_present,
+	.setup_portio_remap		= NULL,
+	.check_phys_apicid_present	= es7000_check_phys_apicid_present,
+	.enable_apic_mode		= es7000_enable_apic_mode,
+	.phys_pkg_id			= es7000_phys_pkg_id,
+	.mps_oem_check			= es7000_mps_oem_check_cluster,
+
+	.get_apic_id			= es7000_get_apic_id,
+	.set_apic_id			= NULL,
+	.apic_id_mask			= 0xFF << 24,
+
+	.cpu_mask_to_apicid		= es7000_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid_and		= es7000_cpu_mask_to_apicid_and,
+
+	.send_IPI_mask			= es7000_send_IPI_mask,
+	.send_IPI_mask_allbutself	= NULL,
+	.send_IPI_allbutself		= es7000_send_IPI_allbutself,
+	.send_IPI_all			= es7000_send_IPI_all,
+	.send_IPI_self			= default_send_IPI_self,
+
+	.wakeup_secondary_cpu		= wakeup_secondary_cpu_via_mip,
+
+	.trampoline_phys_low		= 0x467,
+	.trampoline_phys_high		= 0x469,
+
+	.wait_for_init_deassert		= NULL,
+
+	/* Nothing to do for most platforms, since cleared by the INIT cycle: */
+	.smp_callin_clear_local_apic	= NULL,
+	.inquire_remote_apic		= default_inquire_remote_apic,
+
+	.read				= native_apic_mem_read,
+	.write				= native_apic_mem_write,
+	.icr_read			= native_apic_icr_read,
+	.icr_write			= native_apic_icr_write,
+	.wait_icr_idle			= native_apic_wait_icr_idle,
+	.safe_wait_icr_idle		= native_safe_apic_wait_icr_idle,
+};
 
 struct apic apic_es7000 = {
 
@@ -737,8 +762,6 @@
 	.send_IPI_all			= es7000_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
-
 	.trampoline_phys_low		= 0x467,
 	.trampoline_phys_high		= 0x469,
 
diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c
index d9d6d61..ba2fc64 100644
--- a/arch/x86/kernel/apic/numaq_32.c
+++ b/arch/x86/kernel/apic/numaq_32.c
@@ -69,7 +69,7 @@
 /* x86_quirks member */
 static int				mpc_record;
 
-static __cpuinitdata struct mpc_trans	*translation_table[MAX_MPC_ENTRY];
+static struct mpc_trans			*translation_table[MAX_MPC_ENTRY];
 
 int					mp_bus_id_to_node[MAX_MP_BUSSES];
 int					mp_bus_id_to_local[MAX_MP_BUSSES];
@@ -256,13 +256,6 @@
 	return 1;
 }
 
-static int __init numaq_update_apic(void)
-{
-	apic->wakeup_cpu = wakeup_secondary_cpu_via_nmi;
-
-	return 0;
-}
-
 static struct x86_quirks numaq_x86_quirks __initdata = {
 	.arch_pre_time_init		= numaq_pre_time_init,
 	.arch_time_init			= NULL,
@@ -278,7 +271,6 @@
 	.mpc_oem_pci_bus		= mpc_oem_pci_bus,
 	.smp_read_mpc_oem		= smp_read_mpc_oem,
 	.setup_ioapic_ids		= numaq_setup_ioapic_ids,
-	.update_apic			= numaq_update_apic,
 };
 
 static __init void early_check_numaq(void)
@@ -546,7 +538,7 @@
 	.send_IPI_all			= numaq_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
+	.wakeup_secondary_cpu		= wakeup_secondary_cpu_via_nmi,
 	.trampoline_phys_low		= NUMAQ_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= NUMAQ_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c
index 3a730fa..141c99a 100644
--- a/arch/x86/kernel/apic/probe_32.c
+++ b/arch/x86/kernel/apic/probe_32.c
@@ -138,7 +138,6 @@
 	.send_IPI_all			= default_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
@@ -159,6 +158,7 @@
 extern struct apic apic_summit;
 extern struct apic apic_bigsmp;
 extern struct apic apic_es7000;
+extern struct apic apic_es7000_cluster;
 extern struct apic apic_default;
 
 struct apic *apic = &apic_default;
@@ -176,6 +176,7 @@
 #endif
 #ifdef CONFIG_X86_ES7000
 	&apic_es7000,
+	&apic_es7000_cluster,
 #endif
 	&apic_default,	/* must be last */
 	NULL,
@@ -197,9 +198,6 @@
 		}
 	}
 
-	if (x86_quirks->update_apic)
-		x86_quirks->update_apic();
-
 	/* Parsed again by __setup for debug/verbose */
 	return 0;
 }
@@ -218,8 +216,6 @@
 	if (!cmdline_apic && apic == &apic_default) {
 		if (apic_bigsmp.probe()) {
 			apic = &apic_bigsmp;
-			if (x86_quirks->update_apic)
-				x86_quirks->update_apic();
 			printk(KERN_INFO "Overriding APIC driver with %s\n",
 			       apic->name);
 		}
@@ -240,9 +236,6 @@
 		/* Not visible without early console */
 		if (!apic_probe[i])
 			panic("Didn't find an APIC driver");
-
-		if (x86_quirks->update_apic)
-			x86_quirks->update_apic();
 	}
 	printk(KERN_INFO "Using APIC driver %s\n", apic->name);
 }
@@ -262,8 +255,6 @@
 
 		if (!cmdline_apic) {
 			apic = apic_probe[i];
-			if (x86_quirks->update_apic)
-				x86_quirks->update_apic();
 			printk(KERN_INFO "Switched to APIC driver `%s'.\n",
 			       apic->name);
 		}
@@ -284,8 +275,6 @@
 
 		if (!cmdline_apic) {
 			apic = apic_probe[i];
-			if (x86_quirks->update_apic)
-				x86_quirks->update_apic();
 			printk(KERN_INFO "Switched to APIC driver `%s'.\n",
 			       apic->name);
 		}
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index e7c1636..8d7748e 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -68,9 +68,6 @@
 			apic = &apic_physflat;
 		printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
 	}
-
-	if (x86_quirks->update_apic)
-		x86_quirks->update_apic();
 }
 
 /* Same for both flat and physical. */
diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c
index 32838b5..aac52fa 100644
--- a/arch/x86/kernel/apic/summit_32.c
+++ b/arch/x86/kernel/apic/summit_32.c
@@ -77,9 +77,9 @@
 extern int use_cyclone;
 
 #ifdef CONFIG_X86_SUMMIT_NUMA
-extern void setup_summit(void);
+static void setup_summit(void);
 #else
-#define setup_summit()	{}
+static inline void setup_summit(void) {}
 #endif
 
 static int summit_mps_oem_check(struct mpc_table *mpc, char *oem,
@@ -291,33 +291,21 @@
 
 static unsigned int summit_cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
-	int cpus_found = 0;
-	int num_bits_set;
-	int apicid;
-	int cpu;
+	unsigned int round = 0;
+	int cpu, apicid = 0;
 
-	num_bits_set = cpus_weight(*cpumask);
-	if (num_bits_set >= nr_cpu_ids)
-		return BAD_APICID;
 	/*
 	 * The cpus in the mask must all be on the apic cluster.
 	 */
-	cpu = first_cpu(*cpumask);
-	apicid = summit_cpu_to_logical_apicid(cpu);
+	for_each_cpu(cpu, cpumask) {
+		int new_apicid = summit_cpu_to_logical_apicid(cpu);
 
-	while (cpus_found < num_bits_set) {
-		if (cpu_isset(cpu, *cpumask)) {
-			int new_apicid = summit_cpu_to_logical_apicid(cpu);
-
-			if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-				printk("%s: Not a valid mask!\n", __func__);
-
-				return BAD_APICID;
-			}
-			apicid = apicid | new_apicid;
-			cpus_found++;
+		if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
+			printk("%s: Not a valid mask!\n", __func__);
+			return BAD_APICID;
 		}
-		cpu++;
+		apicid |= new_apicid;
+		round++;
 	}
 	return apicid;
 }
@@ -372,15 +360,15 @@
 }
 
 #ifdef CONFIG_X86_SUMMIT_NUMA
-static struct rio_table_hdr *rio_table_hdr __initdata;
-static struct scal_detail   *scal_devs[MAX_NUMNODES] __initdata;
-static struct rio_detail    *rio_devs[MAX_NUMNODES*4] __initdata;
+static struct rio_table_hdr *rio_table_hdr;
+static struct scal_detail   *scal_devs[MAX_NUMNODES];
+static struct rio_detail    *rio_devs[MAX_NUMNODES*4];
 
 #ifndef CONFIG_X86_NUMAQ
-static int mp_bus_id_to_node[MAX_MP_BUSSES] __initdata;
+static int mp_bus_id_to_node[MAX_MP_BUSSES];
 #endif
 
-static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
+static int setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
 {
 	int twister = 0, node = 0;
 	int i, bus, num_buses;
@@ -442,7 +430,7 @@
 	return bus;
 }
 
-static int __init build_detail_arrays(void)
+static int build_detail_arrays(void)
 {
 	unsigned long ptr;
 	int i, scal_detail_size, rio_detail_size;
@@ -476,7 +464,7 @@
 	return 1;
 }
 
-void __init setup_summit(void)
+void setup_summit(void)
 {
 	unsigned long		ptr;
 	unsigned short		offset;
@@ -574,7 +562,6 @@
 	.send_IPI_all			= summit_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c
index 354b9c4..8fb87b6 100644
--- a/arch/x86/kernel/apic/x2apic_cluster.c
+++ b/arch/x86/kernel/apic/x2apic_cluster.c
@@ -224,7 +224,6 @@
 	.send_IPI_all			= x2apic_send_IPI_all,
 	.send_IPI_self			= x2apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c
index 5bcb174..23625b9 100644
--- a/arch/x86/kernel/apic/x2apic_phys.c
+++ b/arch/x86/kernel/apic/x2apic_phys.c
@@ -213,7 +213,6 @@
 	.send_IPI_all			= x2apic_send_IPI_all,
 	.send_IPI_self			= x2apic_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index 20b4ad0..1bd6da1 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -7,28 +7,28 @@
  *
  * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved.
  */
-
-#include <linux/kernel.h>
-#include <linux/threads.h>
-#include <linux/cpu.h>
 #include <linux/cpumask.h>
+#include <linux/hardirq.h>
+#include <linux/proc_fs.h>
+#include <linux/threads.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/string.h>
 #include <linux/ctype.h>
-#include <linux/init.h>
 #include <linux/sched.h>
-#include <linux/module.h>
-#include <linux/hardirq.h>
 #include <linux/timer.h>
-#include <linux/proc_fs.h>
-#include <asm/current.h>
-#include <asm/smp.h>
-#include <asm/apic.h>
-#include <asm/ipi.h>
-#include <asm/pgtable.h>
-#include <asm/uv/uv.h>
+#include <linux/cpu.h>
+#include <linux/init.h>
+
 #include <asm/uv/uv_mmrs.h>
 #include <asm/uv/uv_hub.h>
+#include <asm/current.h>
+#include <asm/pgtable.h>
 #include <asm/uv/bios.h>
+#include <asm/uv/uv.h>
+#include <asm/apic.h>
+#include <asm/ipi.h>
+#include <asm/smp.h>
 
 DEFINE_PER_CPU(int, x2apic_extra_bits);
 
@@ -91,24 +91,28 @@
 	cpumask_set_cpu(cpu, retmask);
 }
 
-int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
+static int uv_wakeup_secondary(int phys_apicid, unsigned long start_rip)
 {
+#ifdef CONFIG_SMP
 	unsigned long val;
 	int pnode;
 
 	pnode = uv_apicid_to_pnode(phys_apicid);
 	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
 	    (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
-	    (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+	    ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
 	    APIC_DM_INIT;
 	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
 	mdelay(10);
 
 	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
 	    (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
-	    (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+	    ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
 	    APIC_DM_STARTUP;
 	uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
+
+	atomic_set(&init_deasserted, 1);
+#endif
 	return 0;
 }
 
@@ -285,7 +289,7 @@
 	.send_IPI_all			= uv_send_IPI_all,
 	.send_IPI_self			= uv_send_IPI_self,
 
-	.wakeup_cpu			= NULL,
+	.wakeup_secondary_cpu		= uv_wakeup_secondary,
 	.trampoline_phys_low		= DEFAULT_TRAMPOLINE_PHYS_LOW,
 	.trampoline_phys_high		= DEFAULT_TRAMPOLINE_PHYS_HIGH,
 	.wait_for_init_deassert		= NULL,
@@ -365,7 +369,7 @@
 	paddr = base << shift;
 	bytes = (1UL << shift) * (max_pnode + 1);
 	printk(KERN_INFO "UV: Map %s_HI 0x%lx - 0x%lx\n", id, paddr,
-	       					paddr + bytes);
+						paddr + bytes);
 	if (map_type == map_uc)
 		init_extra_mapping_uc(paddr, bytes);
 	else
@@ -528,7 +532,7 @@
 
 /*
  * Called on each cpu to initialize the per_cpu UV data area.
- * 	ZZZ hotplug not supported yet
+ * FIXME: hotplug not supported yet
  */
 void __cpuinit uv_cpu_init(void)
 {
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index 01b1244..d67e0e4 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -7,11 +7,10 @@
 /*
  *	Get CPU information for use by the procfs.
  */
-#ifdef CONFIG_X86_32
 static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c,
 			      unsigned int cpu)
 {
-#ifdef CONFIG_X86_HT
+#ifdef CONFIG_SMP
 	if (c->x86_max_cores * smp_num_siblings > 1) {
 		seq_printf(m, "physical id\t: %d\n", c->phys_proc_id);
 		seq_printf(m, "siblings\t: %d\n",
@@ -24,6 +23,7 @@
 #endif
 }
 
+#ifdef CONFIG_X86_32
 static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c)
 {
 	/*
@@ -50,22 +50,6 @@
 		   c->wp_works_ok ? "yes" : "no");
 }
 #else
-static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c,
-			      unsigned int cpu)
-{
-#ifdef CONFIG_SMP
-	if (c->x86_max_cores * smp_num_siblings > 1) {
-		seq_printf(m, "physical id\t: %d\n", c->phys_proc_id);
-		seq_printf(m, "siblings\t: %d\n",
-			   cpus_weight(per_cpu(cpu_core_map, cpu)));
-		seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id);
-		seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
-		seq_printf(m, "apicid\t\t: %d\n", c->apicid);
-		seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid);
-	}
-#endif
-}
-
 static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c)
 {
 	seq_printf(m,
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index fb2159a..3d9672e 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -1383,7 +1383,7 @@
 #ifdef CONFIG_X86_32
 # define IS_IA32	1
 #elif defined CONFIG_IA32_EMULATION
-# define IS_IA32	test_thread_flag(TIF_IA32)
+# define IS_IA32	is_compat_task()
 #else
 # define IS_IA32	0
 #endif
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 5b85759..4c54bc0 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -600,19 +600,7 @@
 early_param("elfcorehdr", setup_elfcorehdr);
 #endif
 
-static int __init default_update_apic(void)
-{
-#ifdef CONFIG_SMP
-	if (!apic->wakeup_cpu)
-		apic->wakeup_cpu = wakeup_secondary_cpu_via_init;
-#endif
-
-	return 0;
-}
-
-static struct x86_quirks default_x86_quirks __initdata = {
-	.update_apic         = default_update_apic,
-};
+static struct x86_quirks default_x86_quirks __initdata;
 
 struct x86_quirks *x86_quirks __initdata = &default_x86_quirks;
 
@@ -875,9 +863,7 @@
 
 	reserve_initrd();
 
-#ifdef CONFIG_X86_64
 	vsmp_init();
-#endif
 
 	io_delay_init();
 
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index 7cdcd16..d2cc642 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -187,6 +187,71 @@
 /*
  * Set up a signal frame.
  */
+
+/*
+ * Determine which stack to use..
+ */
+static unsigned long align_sigframe(unsigned long sp)
+{
+#ifdef CONFIG_X86_32
+	/*
+	 * Align the stack pointer according to the i386 ABI,
+	 * i.e. so that on function entry ((sp + 4) & 15) == 0.
+	 */
+	sp = ((sp + 4) & -16ul) - 4;
+#else /* !CONFIG_X86_32 */
+	sp = round_down(sp, 16) - 8;
+#endif
+	return sp;
+}
+
+static inline void __user *
+get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
+	     void __user **fpstate)
+{
+	/* Default to using normal stack */
+	unsigned long sp = regs->sp;
+
+#ifdef CONFIG_X86_64
+	/* redzone */
+	sp -= 128;
+#endif /* CONFIG_X86_64 */
+
+	/*
+	 * If we are on the alternate signal stack and would overflow it, don't.
+	 * Return an always-bogus address instead so we will die with SIGSEGV.
+	 */
+	if (on_sig_stack(sp) && !likely(on_sig_stack(sp - frame_size)))
+		return (void __user *) -1L;
+
+	/* This is the X/Open sanctioned signal stack switching.  */
+	if (ka->sa.sa_flags & SA_ONSTACK) {
+		if (sas_ss_flags(sp) == 0)
+			sp = current->sas_ss_sp + current->sas_ss_size;
+	} else {
+#ifdef CONFIG_X86_32
+		/* This is the legacy signal stack switching. */
+		if ((regs->ss & 0xffff) != __USER_DS &&
+			!(ka->sa.sa_flags & SA_RESTORER) &&
+				ka->sa.sa_restorer)
+			sp = (unsigned long) ka->sa.sa_restorer;
+#endif /* CONFIG_X86_32 */
+	}
+
+	if (used_math()) {
+		sp -= sig_xstate_size;
+#ifdef CONFIG_X86_64
+		sp = round_down(sp, 64);
+#endif /* CONFIG_X86_64 */
+		*fpstate = (void __user *)sp;
+
+		if (save_i387_xstate(*fpstate) < 0)
+			return (void __user *)-1L;
+	}
+
+	return (void __user *)align_sigframe(sp - frame_size);
+}
+
 #ifdef CONFIG_X86_32
 static const struct {
 	u16 poplmovl;
@@ -210,54 +275,6 @@
 	0
 };
 
-/*
- * Determine which stack to use..
- */
-static inline void __user *
-get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
-	     void **fpstate)
-{
-	unsigned long sp;
-
-	/* Default to using normal stack */
-	sp = regs->sp;
-
-	/*
-	 * If we are on the alternate signal stack and would overflow it, don't.
-	 * Return an always-bogus address instead so we will die with SIGSEGV.
-	 */
-	if (on_sig_stack(sp) && !likely(on_sig_stack(sp - frame_size)))
-		return (void __user *) -1L;
-
-	/* This is the X/Open sanctioned signal stack switching.  */
-	if (ka->sa.sa_flags & SA_ONSTACK) {
-		if (sas_ss_flags(sp) == 0)
-			sp = current->sas_ss_sp + current->sas_ss_size;
-	} else {
-		/* This is the legacy signal stack switching. */
-		if ((regs->ss & 0xffff) != __USER_DS &&
-			!(ka->sa.sa_flags & SA_RESTORER) &&
-				ka->sa.sa_restorer)
-			sp = (unsigned long) ka->sa.sa_restorer;
-	}
-
-	if (used_math()) {
-		sp = sp - sig_xstate_size;
-		*fpstate = (struct _fpstate *) sp;
-		if (save_i387_xstate(*fpstate) < 0)
-			return (void __user *)-1L;
-	}
-
-	sp -= frame_size;
-	/*
-	 * Align the stack pointer according to the i386 ABI,
-	 * i.e. so that on function entry ((sp + 4) & 15) == 0.
-	 */
-	sp = ((sp + 4) & -16ul) - 4;
-
-	return (void __user *) sp;
-}
-
 static int
 __setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 	      struct pt_regs *regs)
@@ -388,24 +405,6 @@
 	return 0;
 }
 #else /* !CONFIG_X86_32 */
-/*
- * Determine which stack to use..
- */
-static void __user *
-get_stack(struct k_sigaction *ka, unsigned long sp, unsigned long size)
-{
-	/* Default to using normal stack - redzone*/
-	sp -= 128;
-
-	/* This is the X/Open sanctioned signal stack switching.  */
-	if (ka->sa.sa_flags & SA_ONSTACK) {
-		if (sas_ss_flags(sp) == 0)
-			sp = current->sas_ss_sp + current->sas_ss_size;
-	}
-
-	return (void __user *)round_down(sp - size, 64);
-}
-
 static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 			    sigset_t *set, struct pt_regs *regs)
 {
@@ -414,15 +413,7 @@
 	int err = 0;
 	struct task_struct *me = current;
 
-	if (used_math()) {
-		fp = get_stack(ka, regs->sp, sig_xstate_size);
-		frame = (void __user *)round_down(
-			(unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8;
-
-		if (save_i387_xstate(fp) < 0)
-			return -EFAULT;
-	} else
-		frame = get_stack(ka, regs->sp, sizeof(struct rt_sigframe)) - 8;
+	frame = get_sigframe(ka, regs, sizeof(struct rt_sigframe), &fp);
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
 		return -EFAULT;
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 9ce6663..249334f 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -112,7 +112,7 @@
 DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
 EXPORT_PER_CPU_SYMBOL(cpu_info);
 
-static atomic_t init_deasserted;
+atomic_t init_deasserted;
 
 
 /* Set if we find a B stepping CPU */
@@ -614,12 +614,6 @@
 	unsigned long send_status, accept_status = 0;
 	int maxlvt, num_starts, j;
 
-	if (get_uv_system_type() == UV_NON_UNIQUE_APIC) {
-		send_status = uv_wakeup_secondary(phys_apicid, start_eip);
-		atomic_set(&init_deasserted, 1);
-		return send_status;
-	}
-
 	maxlvt = lapic_get_maxlvt();
 
 	/*
@@ -748,7 +742,8 @@
 /*
  * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
  * (ie clustered apic addressing mode), this is a LOGICAL apic ID.
- * Returns zero if CPU booted OK, else error code from ->wakeup_cpu.
+ * Returns zero if CPU booted OK, else error code from
+ * ->wakeup_secondary_cpu.
  */
 static int __cpuinit do_boot_cpu(int apicid, int cpu)
 {
@@ -835,9 +830,13 @@
 	}
 
 	/*
-	 * Starting actual IPI sequence...
+	 * Kick the secondary CPU. Use the method in the APIC driver
+	 * if it's defined - or use an INIT boot APIC message otherwise:
 	 */
-	boot_error = apic->wakeup_cpu(apicid, start_ip);
+	if (apic->wakeup_secondary_cpu)
+		boot_error = apic->wakeup_secondary_cpu(apicid, start_ip);
+	else
+		boot_error = wakeup_secondary_cpu_via_init(apicid, start_ip);
 
 	if (!boot_error) {
 		/*
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index c609205..74de562 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -22,7 +22,7 @@
 #include <asm/paravirt.h>
 #include <asm/setup.h>
 
-#if defined CONFIG_PCI && defined CONFIG_PARAVIRT
+#ifdef CONFIG_PARAVIRT
 /*
  * Interrupt control on vSMPowered systems:
  * ~AC is a shadow of IF.  If IF is 'on' AC should be 'off'
@@ -114,7 +114,6 @@
 }
 #endif
 
-#ifdef CONFIG_PCI
 static int is_vsmp = -1;
 
 static void __init detect_vsmp_box(void)
@@ -139,15 +138,6 @@
 		return 0;
 	}
 }
-#else
-static void __init detect_vsmp_box(void)
-{
-}
-int is_vsmp_box(void)
-{
-	return 0;
-}
-#endif
 
 void __init vsmp_init(void)
 {
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index 2b938a3..0853774 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -1,4 +1,4 @@
-obj-y	:=  init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
+obj-y	:=  init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
 	    pat.o pgtable.o gup.o
 
 obj-$(CONFIG_SMP)		+= tlb.o
diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index bcc079c..00f127c 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -1,5 +1,6 @@
 #include <linux/highmem.h>
 #include <linux/module.h>
+#include <linux/swap.h> /* for totalram_pages */
 
 void *kmap(struct page *page)
 {
@@ -156,3 +157,36 @@
 EXPORT_SYMBOL(kunmap);
 EXPORT_SYMBOL(kmap_atomic);
 EXPORT_SYMBOL(kunmap_atomic);
+
+#ifdef CONFIG_NUMA
+void __init set_highmem_pages_init(void)
+{
+	struct zone *zone;
+	int nid;
+
+	for_each_zone(zone) {
+		unsigned long zone_start_pfn, zone_end_pfn;
+
+		if (!is_highmem(zone))
+			continue;
+
+		zone_start_pfn = zone->zone_start_pfn;
+		zone_end_pfn = zone_start_pfn + zone->spanned_pages;
+
+		nid = zone_to_nid(zone);
+		printk(KERN_INFO "Initializing %s for node %d (%08lx:%08lx)\n",
+				zone->name, nid, zone_start_pfn, zone_end_pfn);
+
+		add_highpages_with_active_regions(nid, zone_start_pfn,
+				 zone_end_pfn);
+	}
+	totalram_pages += totalhigh_pages;
+}
+#else
+void __init set_highmem_pages_init(void)
+{
+	add_highpages_with_active_regions(0, highstart_pfn, highend_pfn);
+
+	totalram_pages += totalhigh_pages;
+}
+#endif /* CONFIG_NUMA */
diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
new file mode 100644
index 0000000..ce6a722
--- /dev/null
+++ b/arch/x86/mm/init.c
@@ -0,0 +1,49 @@
+#include <linux/swap.h>
+#include <asm/cacheflush.h>
+#include <asm/page.h>
+#include <asm/sections.h>
+#include <asm/system.h>
+
+void free_init_pages(char *what, unsigned long begin, unsigned long end)
+{
+	unsigned long addr = begin;
+
+	if (addr >= end)
+		return;
+
+	/*
+	 * If debugging page accesses then do not free this memory but
+	 * mark them not present - any buggy init-section access will
+	 * create a kernel page fault:
+	 */
+#ifdef CONFIG_DEBUG_PAGEALLOC
+	printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
+		begin, PAGE_ALIGN(end));
+	set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
+#else
+	/*
+	 * We just marked the kernel text read only above, now that
+	 * we are going to free part of that, we need to make that
+	 * writeable first.
+	 */
+	set_memory_rw(begin, (end - begin) >> PAGE_SHIFT);
+
+	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
+
+	for (; addr < end; addr += PAGE_SIZE) {
+		ClearPageReserved(virt_to_page(addr));
+		init_page_count(virt_to_page(addr));
+		memset((void *)(addr & ~(PAGE_SIZE-1)),
+			POISON_FREE_INITMEM, PAGE_SIZE);
+		free_page(addr);
+		totalram_pages++;
+	}
+#endif
+}
+
+void free_initmem(void)
+{
+	free_init_pages("unused kernel memory",
+			(unsigned long)(&__init_begin),
+			(unsigned long)(&__init_end));
+}
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 06708ee..0b087dc 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -50,8 +50,6 @@
 #include <asm/setup.h>
 #include <asm/cacheflush.h>
 
-unsigned int __VMALLOC_RESERVE = 128 << 20;
-
 unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
 
@@ -469,22 +467,10 @@
 	work_with_active_regions(nid, add_highpages_work_fn, &data);
 }
 
-#ifndef CONFIG_NUMA
-static void __init set_highmem_pages_init(void)
-{
-	add_highpages_with_active_regions(0, highstart_pfn, highend_pfn);
-
-	totalram_pages += totalhigh_pages;
-}
-#endif /* !CONFIG_NUMA */
-
 #else
 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)
@@ -847,10 +833,10 @@
 	unsigned long puds, pmds, ptes, tables, start;
 
 	puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
-	tables = PAGE_ALIGN(puds * sizeof(pud_t));
+	tables = roundup(puds * sizeof(pud_t), PAGE_SIZE);
 
 	pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
-	tables += PAGE_ALIGN(pmds * sizeof(pmd_t));
+	tables += roundup(pmds * sizeof(pmd_t), PAGE_SIZE);
 
 	if (use_pse) {
 		unsigned long extra;
@@ -861,10 +847,10 @@
 	} else
 		ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
-	tables += PAGE_ALIGN(ptes * sizeof(pte_t));
+	tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE);
 
 	/* for fixmap */
-	tables += PAGE_ALIGN(__end_of_fixed_addresses * sizeof(pte_t));
+	tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE);
 
 	/*
 	 * RED-PEN putting page tables only on node 0 could
@@ -1214,45 +1200,6 @@
 }
 #endif
 
-void free_init_pages(char *what, unsigned long begin, unsigned long end)
-{
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	/*
-	 * If debugging page accesses then do not free this memory but
-	 * mark them not present - any buggy init-section access will
-	 * create a kernel page fault:
-	 */
-	printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
-		begin, PAGE_ALIGN(end));
-	set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
-#else
-	unsigned long addr;
-
-	/*
-	 * We just marked the kernel text read only above, now that
-	 * we are going to free part of that, we need to make that
-	 * writeable first.
-	 */
-	set_memory_rw(begin, (end - begin) >> PAGE_SHIFT);
-
-	for (addr = begin; addr < end; addr += PAGE_SIZE) {
-		ClearPageReserved(virt_to_page(addr));
-		init_page_count(virt_to_page(addr));
-		memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
-		free_page(addr);
-		totalram_pages++;
-	}
-	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
-#endif
-}
-
-void free_initmem(void)
-{
-	free_init_pages("unused kernel memory",
-			(unsigned long)(&__init_begin),
-			(unsigned long)(&__init_end));
-}
-
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
 {
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index e6d36b4..724e537 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -714,6 +714,8 @@
 	pos = start_pfn << PAGE_SHIFT;
 	end_pfn = ((pos + (PMD_SIZE - 1)) >> PMD_SHIFT)
 			<< (PMD_SHIFT - PAGE_SHIFT);
+	if (end_pfn > (end >> PAGE_SHIFT))
+		end_pfn = end >> PAGE_SHIFT;
 	if (start_pfn < end_pfn) {
 		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);
 		pos = end_pfn << PAGE_SHIFT;
@@ -945,43 +947,6 @@
 		initsize >> 10);
 }
 
-void free_init_pages(char *what, unsigned long begin, unsigned long end)
-{
-	unsigned long addr = begin;
-
-	if (addr >= end)
-		return;
-
-	/*
-	 * If debugging page accesses then do not free this memory but
-	 * mark them not present - any buggy init-section access will
-	 * create a kernel page fault:
-	 */
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
-		begin, PAGE_ALIGN(end));
-	set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
-#else
-	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
-
-	for (; addr < end; addr += PAGE_SIZE) {
-		ClearPageReserved(virt_to_page(addr));
-		init_page_count(virt_to_page(addr));
-		memset((void *)(addr & ~(PAGE_SIZE-1)),
-			POISON_FREE_INITMEM, PAGE_SIZE);
-		free_page(addr);
-		totalram_pages++;
-	}
-#endif
-}
-
-void free_initmem(void)
-{
-	free_init_pages("unused kernel memory",
-			(unsigned long)(&__init_begin),
-			(unsigned long)(&__init_end));
-}
-
 #ifdef CONFIG_DEBUG_RODATA
 const int rodata_test_data = 0xC3;
 EXPORT_SYMBOL_GPL(rodata_test_data);
diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c
index 3957cd6..451fe95 100644
--- a/arch/x86/mm/numa_32.c
+++ b/arch/x86/mm/numa_32.c
@@ -423,32 +423,6 @@
 	setup_bootmem_allocator();
 }
 
-void __init set_highmem_pages_init(void)
-{
-#ifdef CONFIG_HIGHMEM
-	struct zone *zone;
-	int nid;
-
-	for_each_zone(zone) {
-		unsigned long zone_start_pfn, zone_end_pfn;
-
-		if (!is_highmem(zone))
-			continue;
-
-		zone_start_pfn = zone->zone_start_pfn;
-		zone_end_pfn = zone_start_pfn + zone->spanned_pages;
-
-		nid = zone_to_nid(zone);
-		printk(KERN_INFO "Initializing %s for node %d (%08lx:%08lx)\n",
-				zone->name, nid, zone_start_pfn, zone_end_pfn);
-
-		add_highpages_with_active_regions(nid, zone_start_pfn,
-				 zone_end_pfn);
-	}
-	totalram_pages += totalhigh_pages;
-#endif
-}
-
 #ifdef CONFIG_MEMORY_HOTPLUG
 static int paddr_to_nid(u64 addr)
 {
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index 86f2ffc..5b7c7c84 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -313,6 +313,24 @@
 	return young;
 }
 
+/**
+ * reserve_top_address - reserves a hole in the top of kernel address space
+ * @reserve - size of hole to reserve
+ *
+ * Can be used to relocate the fixmap area and poke a hole in the top
+ * of kernel address space to make room for a hypervisor.
+ */
+void __init reserve_top_address(unsigned long reserve)
+{
+#ifdef CONFIG_X86_32
+	BUG_ON(fixmaps_set > 0);
+	printk(KERN_INFO "Reserving virtual address space above 0x%08x\n",
+	       (int)-reserve);
+	__FIXADDR_TOP = -reserve - PAGE_SIZE;
+	__VMALLOC_RESERVE += reserve;
+#endif
+}
+
 int fixmaps_set;
 
 void __native_set_fixmap(enum fixed_addresses idx, pte_t pte)
diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index 0951db9..f2e477c 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -20,6 +20,8 @@
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
 
+unsigned int __VMALLOC_RESERVE = 128 << 20;
+
 /*
  * Associate a virtual page frame with a given physical page frame 
  * and protection flags for that frame.
@@ -97,22 +99,6 @@
 unsigned long __FIXADDR_TOP = 0xfffff000;
 EXPORT_SYMBOL(__FIXADDR_TOP);
 
-/**
- * reserve_top_address - reserves a hole in the top of kernel address space
- * @reserve - size of hole to reserve
- *
- * Can be used to relocate the fixmap area and poke a hole in the top
- * of kernel address space to make room for a hypervisor.
- */
-void __init reserve_top_address(unsigned long reserve)
-{
-	BUG_ON(fixmaps_set > 0);
-	printk(KERN_INFO "Reserving virtual address space above 0x%08x\n",
-	       (int)-reserve);
-	__FIXADDR_TOP = -reserve - PAGE_SIZE;
-	__VMALLOC_RESERVE += reserve;
-}
-
 /*
  * vmalloc=size forces the vmalloc area to be exactly 'size'
  * bytes. This can be used to increase (or decrease) the
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index e9f80c7..10131fb 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -78,8 +78,18 @@
 	if (cpu_has_arch_perfmon) {
 		union cpuid10_eax eax;
 		eax.full = cpuid_eax(0xa);
-		if (counter_width < eax.split.bit_width)
-			counter_width = eax.split.bit_width;
+
+		/*
+		 * For Core2 (family 6, model 15), don't reset the
+		 * counter width:
+		 */
+		if (!(eax.split.version_id == 0 &&
+			current_cpu_data.x86 == 6 &&
+				current_cpu_data.x86_model == 15)) {
+
+			if (counter_width < eax.split.bit_width)
+				counter_width = eax.split.bit_width;
+		}
 	}
 
 	/* clear all counters */
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index 72c667f..12715d3 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -420,7 +420,7 @@
 				dev->sigdata.lock = NULL;
 			master->lock.hw_lock = NULL;   /* SHM removed */
 			master->lock.file_priv = NULL;
-			wake_up_interruptible(&master->lock.lock_queue);
+			wake_up_interruptible_all(&master->lock.lock_queue);
 		}
 		break;
 	case _DRM_AGP:
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 6c020fe..f52663e 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -484,6 +484,7 @@
 	mutex_lock(&dev->struct_mutex);
 
 	if (file_priv->is_master) {
+		struct drm_master *master = file_priv->master;
 		struct drm_file *temp;
 		list_for_each_entry(temp, &dev->filelist, lhead) {
 			if ((temp->master == file_priv->master) &&
@@ -491,6 +492,19 @@
 				temp->authenticated = 0;
 		}
 
+		/**
+		 * Since the master is disappearing, so is the
+		 * possibility to lock.
+		 */
+
+		if (master->lock.hw_lock) {
+			if (dev->sigdata.lock == master->lock.hw_lock)
+				dev->sigdata.lock = NULL;
+			master->lock.hw_lock = NULL;
+			master->lock.file_priv = NULL;
+			wake_up_interruptible_all(&master->lock.lock_queue);
+		}
+
 		if (file_priv->minor->master == file_priv->master) {
 			/* drop the reference held my the minor */
 			drm_master_put(&file_priv->minor->master);
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index 46e7b28..e2f70a5 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -80,6 +80,7 @@
 		__set_current_state(TASK_INTERRUPTIBLE);
 		if (!master->lock.hw_lock) {
 			/* Device has been unregistered */
+			send_sig(SIGTERM, current, 0);
 			ret = -EINTR;
 			break;
 		}
@@ -93,7 +94,7 @@
 		/* Contention */
 		schedule();
 		if (signal_pending(current)) {
-			ret = -ERESTARTSYS;
+			ret = -EINTR;
 			break;
 		}
 	}
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index 46bb923..096e2a3 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -146,14 +146,6 @@
 
 	drm_ht_remove(&master->magiclist);
 
-	if (master->lock.hw_lock) {
-		if (dev->sigdata.lock == master->lock.hw_lock)
-			dev->sigdata.lock = NULL;
-		master->lock.hw_lock = NULL;
-		master->lock.file_priv = NULL;
-		wake_up_interruptible(&master->lock.lock_queue);
-	}
-
 	drm_free(master, sizeof(*master), DRM_MEM_DRIVER);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index c29feb9..85685bf 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -211,7 +211,7 @@
 
 	vaddr_atomic = io_mapping_map_atomic_wc(mapping, page_base);
 	unwritten = __copy_from_user_inatomic_nocache(vaddr_atomic + page_offset,
-						      user_data, length, length);
+						      user_data, length);
 	io_mapping_unmap_atomic(vaddr_atomic);
 	if (unwritten)
 		return -EFAULT;
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 548ff2c..87b6b60 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -383,12 +383,13 @@
 	drm_i915_irq_emit_t *emit = data;
 	int result;
 
-	RING_LOCK_TEST_WITH_RETURN(dev, file_priv);
-
 	if (!dev_priv) {
 		DRM_ERROR("called with no initialization\n");
 		return -EINVAL;
 	}
+
+	RING_LOCK_TEST_WITH_RETURN(dev, file_priv);
+
 	mutex_lock(&dev->struct_mutex);
 	result = i915_emit_irq(dev);
 	mutex_unlock(&dev->struct_mutex);
diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c
index c3c8b9b..45470f1 100644
--- a/drivers/input/keyboard/atkbd.c
+++ b/drivers/input/keyboard/atkbd.c
@@ -839,7 +839,7 @@
  */
 static void atkbd_dell_laptop_keymap_fixup(struct atkbd *atkbd)
 {
-	const unsigned int forced_release_keys[] = {
+	static const unsigned int forced_release_keys[] = {
 		0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x8b, 0x8f, 0x93,
 	};
 	int i;
@@ -856,7 +856,7 @@
  */
 static void atkbd_hp_keymap_fixup(struct atkbd *atkbd)
 {
-	const unsigned int forced_release_keys[] = {
+	static const unsigned int forced_release_keys[] = {
 		0x94,
 	};
 	int i;
diff --git a/drivers/input/keyboard/bf54x-keys.c b/drivers/input/keyboard/bf54x-keys.c
index 1928401..ee855c5 100644
--- a/drivers/input/keyboard/bf54x-keys.c
+++ b/drivers/input/keyboard/bf54x-keys.c
@@ -209,8 +209,8 @@
 		goto out;
 	}
 
-	if (!pdata->debounce_time || !pdata->debounce_time > MAX_MULT ||
-	    !pdata->coldrive_time || !pdata->coldrive_time > MAX_MULT) {
+	if (!pdata->debounce_time || pdata->debounce_time > MAX_MULT ||
+	    !pdata->coldrive_time || pdata->coldrive_time > MAX_MULT) {
 		printk(KERN_ERR DRV_NAME
 			": Invalid Debounce/Columdrive Time from pdata\n");
 		bfin_write_KPAD_MSEL(0xFF0);	/* Default MSEL	*/
diff --git a/drivers/input/keyboard/corgikbd.c b/drivers/input/keyboard/corgikbd.c
index c8ed065..abb04c8 100644
--- a/drivers/input/keyboard/corgikbd.c
+++ b/drivers/input/keyboard/corgikbd.c
@@ -288,7 +288,7 @@
 #define corgikbd_resume		NULL
 #endif
 
-static int __init corgikbd_probe(struct platform_device *pdev)
+static int __devinit corgikbd_probe(struct platform_device *pdev)
 {
 	struct corgikbd *corgikbd;
 	struct input_dev *input_dev;
@@ -368,7 +368,7 @@
 	return err;
 }
 
-static int corgikbd_remove(struct platform_device *pdev)
+static int __devexit corgikbd_remove(struct platform_device *pdev)
 {
 	int i;
 	struct corgikbd *corgikbd = platform_get_drvdata(pdev);
@@ -388,7 +388,7 @@
 
 static struct platform_driver corgikbd_driver = {
 	.probe		= corgikbd_probe,
-	.remove		= corgikbd_remove,
+	.remove		= __devexit_p(corgikbd_remove),
 	.suspend	= corgikbd_suspend,
 	.resume		= corgikbd_resume,
 	.driver		= {
@@ -397,7 +397,7 @@
 	},
 };
 
-static int __devinit corgikbd_init(void)
+static int __init corgikbd_init(void)
 {
 	return platform_driver_register(&corgikbd_driver);
 }
diff --git a/drivers/input/keyboard/omap-keypad.c b/drivers/input/keyboard/omap-keypad.c
index 3f3d119..058fa8b 100644
--- a/drivers/input/keyboard/omap-keypad.c
+++ b/drivers/input/keyboard/omap-keypad.c
@@ -279,7 +279,7 @@
 #define omap_kp_resume	NULL
 #endif
 
-static int __init omap_kp_probe(struct platform_device *pdev)
+static int __devinit omap_kp_probe(struct platform_device *pdev)
 {
 	struct omap_kp *omap_kp;
 	struct input_dev *input_dev;
@@ -422,7 +422,7 @@
 	return -EINVAL;
 }
 
-static int omap_kp_remove(struct platform_device *pdev)
+static int __devexit omap_kp_remove(struct platform_device *pdev)
 {
 	struct omap_kp *omap_kp = platform_get_drvdata(pdev);
 
@@ -454,7 +454,7 @@
 
 static struct platform_driver omap_kp_driver = {
 	.probe		= omap_kp_probe,
-	.remove		= omap_kp_remove,
+	.remove		= __devexit_p(omap_kp_remove),
 	.suspend	= omap_kp_suspend,
 	.resume		= omap_kp_resume,
 	.driver		= {
@@ -463,7 +463,7 @@
 	},
 };
 
-static int __devinit omap_kp_init(void)
+static int __init omap_kp_init(void)
 {
 	printk(KERN_INFO "OMAP Keypad Driver\n");
 	return platform_driver_register(&omap_kp_driver);
diff --git a/drivers/input/keyboard/spitzkbd.c b/drivers/input/keyboard/spitzkbd.c
index c48b76a..9d1781a 100644
--- a/drivers/input/keyboard/spitzkbd.c
+++ b/drivers/input/keyboard/spitzkbd.c
@@ -343,7 +343,7 @@
 #define spitzkbd_resume		NULL
 #endif
 
-static int __init spitzkbd_probe(struct platform_device *dev)
+static int __devinit spitzkbd_probe(struct platform_device *dev)
 {
 	struct spitzkbd *spitzkbd;
 	struct input_dev *input_dev;
@@ -444,7 +444,7 @@
 	return err;
 }
 
-static int spitzkbd_remove(struct platform_device *dev)
+static int __devexit spitzkbd_remove(struct platform_device *dev)
 {
 	int i;
 	struct spitzkbd *spitzkbd = platform_get_drvdata(dev);
@@ -470,7 +470,7 @@
 
 static struct platform_driver spitzkbd_driver = {
 	.probe		= spitzkbd_probe,
-	.remove		= spitzkbd_remove,
+	.remove		= __devexit_p(spitzkbd_remove),
 	.suspend	= spitzkbd_suspend,
 	.resume		= spitzkbd_resume,
 	.driver		= {
@@ -479,7 +479,7 @@
 	},
 };
 
-static int __devinit spitzkbd_init(void)
+static int __init spitzkbd_init(void)
 {
 	return platform_driver_register(&spitzkbd_driver);
 }
diff --git a/drivers/input/mouse/Kconfig b/drivers/input/mouse/Kconfig
index 9bef935..4f38e6f 100644
--- a/drivers/input/mouse/Kconfig
+++ b/drivers/input/mouse/Kconfig
@@ -70,7 +70,7 @@
 config MOUSE_PS2_LIFEBOOK
 	bool "Fujitsu Lifebook PS/2 mouse protocol extension" if EMBEDDED
 	default y
-	depends on MOUSE_PS2
+	depends on MOUSE_PS2 && X86
 	help
 	  Say Y here if you have a Fujitsu B-series Lifebook PS/2
 	  TouchScreen connected to your system.
diff --git a/drivers/input/mouse/elantech.c b/drivers/input/mouse/elantech.c
index b9a25d5..6ab0eb1a 100644
--- a/drivers/input/mouse/elantech.c
+++ b/drivers/input/mouse/elantech.c
@@ -542,7 +542,7 @@
 	    ps2_command(ps2dev,  NULL, PSMOUSE_CMD_SETSCALE11) ||
 	    ps2_command(ps2dev,  NULL, PSMOUSE_CMD_SETSCALE11) ||
 	    ps2_command(ps2dev, param, PSMOUSE_CMD_GETINFO)) {
-		pr_err("elantech.c: sending Elantech magic knock failed.\n");
+		pr_debug("elantech.c: sending Elantech magic knock failed.\n");
 		return -1;
 	}
 
@@ -551,8 +551,27 @@
 	 * set of magic numbers
 	 */
 	if (param[0] != 0x3c || param[1] != 0x03 || param[2] != 0xc8) {
-		pr_info("elantech.c: unexpected magic knock result 0x%02x, 0x%02x, 0x%02x.\n",
-			param[0], param[1], param[2]);
+		pr_debug("elantech.c: "
+			 "unexpected magic knock result 0x%02x, 0x%02x, 0x%02x.\n",
+			 param[0], param[1], param[2]);
+		return -1;
+	}
+
+	/*
+	 * Query touchpad's firmware version and see if it reports known
+	 * value to avoid mis-detection. Logitech mice are known to respond
+	 * to Elantech magic knock and there might be more.
+	 */
+	if (synaptics_send_cmd(psmouse, ETP_FW_VERSION_QUERY, param)) {
+		pr_debug("elantech.c: failed to query firmware version.\n");
+		return -1;
+	}
+
+	pr_debug("elantech.c: Elantech version query result 0x%02x, 0x%02x, 0x%02x.\n",
+		 param[0], param[1], param[2]);
+
+	if (param[0] == 0 || param[1] != 0) {
+		pr_debug("elantech.c: Probably not a real Elantech touchpad. Aborting.\n");
 		return -1;
 	}
 
@@ -600,8 +619,7 @@
 	int i, error;
 	unsigned char param[3];
 
-	etd = kzalloc(sizeof(struct elantech_data), GFP_KERNEL);
-	psmouse->private = etd;
+	psmouse->private = etd = kzalloc(sizeof(struct elantech_data), GFP_KERNEL);
 	if (!etd)
 		return -1;
 
@@ -610,14 +628,12 @@
 		etd->parity[i] = etd->parity[i & (i - 1)] ^ 1;
 
 	/*
-	 * Find out what version hardware this is
+	 * Do the version query again so we can store the result
 	 */
 	if (synaptics_send_cmd(psmouse, ETP_FW_VERSION_QUERY, param)) {
 		pr_err("elantech.c: failed to query firmware version.\n");
 		goto init_fail;
 	}
-	pr_info("elantech.c: Elantech version query result 0x%02x, 0x%02x, 0x%02x.\n",
-		param[0], param[1], param[2]);
 	etd->fw_version_maj = param[0];
 	etd->fw_version_min = param[2];
 
diff --git a/drivers/input/mouse/pxa930_trkball.c b/drivers/input/mouse/pxa930_trkball.c
index d297acc..1e827ad 100644
--- a/drivers/input/mouse/pxa930_trkball.c
+++ b/drivers/input/mouse/pxa930_trkball.c
@@ -83,7 +83,7 @@
 
 	__raw_writel(v, trkball->mmio_base + TBCR);
 
-	while (i--) {
+	while (--i) {
 		if (__raw_readl(trkball->mmio_base + TBCR) == v)
 			break;
 		msleep(1);
diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c
index 865fc69..f3e4f7b 100644
--- a/drivers/input/mouse/synaptics.c
+++ b/drivers/input/mouse/synaptics.c
@@ -182,11 +182,6 @@
 
 static int synaptics_query_hardware(struct psmouse *psmouse)
 {
-	int retries = 0;
-
-	while ((retries++ < 3) && psmouse_reset(psmouse))
-		/* empty */;
-
 	if (synaptics_identify(psmouse))
 		return -1;
 	if (synaptics_model_id(psmouse))
@@ -582,6 +577,8 @@
 	struct synaptics_data *priv = psmouse->private;
 	struct synaptics_data old_priv = *priv;
 
+	psmouse_reset(psmouse);
+
 	if (synaptics_detect(psmouse, 0))
 		return -1;
 
@@ -640,6 +637,8 @@
 	if (!priv)
 		return -1;
 
+	psmouse_reset(psmouse);
+
 	if (synaptics_query_hardware(psmouse)) {
 		printk(KERN_ERR "Unable to query Synaptics hardware.\n");
 		goto init_fail;
diff --git a/drivers/input/serio/ambakmi.c b/drivers/input/serio/ambakmi.c
index b10ffae..e29cdc1 100644
--- a/drivers/input/serio/ambakmi.c
+++ b/drivers/input/serio/ambakmi.c
@@ -57,7 +57,7 @@
 	struct amba_kmi_port *kmi = io->port_data;
 	unsigned int timeleft = 10000; /* timeout in 100ms */
 
-	while ((readb(KMISTAT) & KMISTAT_TXEMPTY) == 0 && timeleft--)
+	while ((readb(KMISTAT) & KMISTAT_TXEMPTY) == 0 && --timeleft)
 		udelay(10);
 
 	if (timeleft)
@@ -129,8 +129,8 @@
 	io->write	= amba_kmi_write;
 	io->open	= amba_kmi_open;
 	io->close	= amba_kmi_close;
-	strlcpy(io->name, dev->dev.bus_id, sizeof(io->name));
-	strlcpy(io->phys, dev->dev.bus_id, sizeof(io->phys));
+	strlcpy(io->name, dev_name(&dev->dev), sizeof(io->name));
+	strlcpy(io->phys, dev_name(&dev->dev), sizeof(io->phys));
 	io->port_data	= kmi;
 	io->dev.parent	= &dev->dev;
 
diff --git a/drivers/input/serio/gscps2.c b/drivers/input/serio/gscps2.c
index adc3bd6..bd0f92d 100644
--- a/drivers/input/serio/gscps2.c
+++ b/drivers/input/serio/gscps2.c
@@ -359,7 +359,7 @@
 
 	snprintf(serio->name, sizeof(serio->name), "GSC PS/2 %s",
 		 (ps2port->id == GSC_ID_KEYBOARD) ? "keyboard" : "mouse");
-	strlcpy(serio->phys, dev->dev.bus_id, sizeof(serio->phys));
+	strlcpy(serio->phys, dev_name(&dev->dev), sizeof(serio->phys));
 	serio->id.type		= SERIO_8042;
 	serio->write		= gscps2_write;
 	serio->open		= gscps2_open;
diff --git a/drivers/input/serio/sa1111ps2.c b/drivers/input/serio/sa1111ps2.c
index 2ad8878..57953c0 100644
--- a/drivers/input/serio/sa1111ps2.c
+++ b/drivers/input/serio/sa1111ps2.c
@@ -246,8 +246,8 @@
 	serio->write		= ps2_write;
 	serio->open		= ps2_open;
 	serio->close		= ps2_close;
-	strlcpy(serio->name, dev->dev.bus_id, sizeof(serio->name));
-	strlcpy(serio->phys, dev->dev.bus_id, sizeof(serio->phys));
+	strlcpy(serio->name, dev_name(&dev->dev), sizeof(serio->name));
+	strlcpy(serio->phys, dev_name(&dev->dev), sizeof(serio->phys));
 	serio->port_data	= ps2if;
 	serio->dev.parent	= &dev->dev;
 	ps2if->io		= serio;
diff --git a/drivers/input/touchscreen/atmel_tsadcc.c b/drivers/input/touchscreen/atmel_tsadcc.c
index a89a6a8..055969e 100644
--- a/drivers/input/touchscreen/atmel_tsadcc.c
+++ b/drivers/input/touchscreen/atmel_tsadcc.c
@@ -236,7 +236,7 @@
 	ts_dev->bufferedmeasure = 0;
 
 	snprintf(ts_dev->phys, sizeof(ts_dev->phys),
-		 "%s/input0", pdev->dev.bus_id);
+		 "%s/input0", dev_name(&pdev->dev));
 
 	input_dev->name = "atmel touch screen controller";
 	input_dev->phys = ts_dev->phys;
diff --git a/drivers/input/touchscreen/corgi_ts.c b/drivers/input/touchscreen/corgi_ts.c
index 65202c9..3fb51b5 100644
--- a/drivers/input/touchscreen/corgi_ts.c
+++ b/drivers/input/touchscreen/corgi_ts.c
@@ -268,7 +268,7 @@
 #define corgits_resume		NULL
 #endif
 
-static int __init corgits_probe(struct platform_device *pdev)
+static int __devinit corgits_probe(struct platform_device *pdev)
 {
 	struct corgi_ts *corgi_ts;
 	struct input_dev *input_dev;
@@ -343,7 +343,7 @@
 	return err;
 }
 
-static int corgits_remove(struct platform_device *pdev)
+static int __devexit corgits_remove(struct platform_device *pdev)
 {
 	struct corgi_ts *corgi_ts = platform_get_drvdata(pdev);
 
@@ -352,12 +352,13 @@
 	corgi_ts->machinfo->put_hsync();
 	input_unregister_device(corgi_ts->input);
 	kfree(corgi_ts);
+
 	return 0;
 }
 
 static struct platform_driver corgits_driver = {
 	.probe		= corgits_probe,
-	.remove		= corgits_remove,
+	.remove		= __devexit_p(corgits_remove),
 	.suspend	= corgits_suspend,
 	.resume		= corgits_resume,
 	.driver		= {
@@ -366,7 +367,7 @@
 	},
 };
 
-static int __devinit corgits_init(void)
+static int __init corgits_init(void)
 {
 	return platform_driver_register(&corgits_driver);
 }
diff --git a/drivers/input/touchscreen/tsc2007.c b/drivers/input/touchscreen/tsc2007.c
index b75dc29..4ab0702 100644
--- a/drivers/input/touchscreen/tsc2007.c
+++ b/drivers/input/touchscreen/tsc2007.c
@@ -289,7 +289,8 @@
 
 	pdata->init_platform_hw();
 
-	snprintf(ts->phys, sizeof(ts->phys), "%s/input0", client->dev.bus_id);
+	snprintf(ts->phys, sizeof(ts->phys),
+		 "%s/input0", dev_name(&client->dev));
 
 	input_dev->name = "TSC2007 Touchscreen";
 	input_dev->phys = ts->phys;
diff --git a/drivers/input/touchscreen/usbtouchscreen.c b/drivers/input/touchscreen/usbtouchscreen.c
index 5080b26..fb7cb9b 100644
--- a/drivers/input/touchscreen/usbtouchscreen.c
+++ b/drivers/input/touchscreen/usbtouchscreen.c
@@ -60,6 +60,10 @@
 module_param(swap_xy, bool, 0644);
 MODULE_PARM_DESC(swap_xy, "If set X and Y axes are swapped.");
 
+static int hwcalib_xy;
+module_param(hwcalib_xy, bool, 0644);
+MODULE_PARM_DESC(hwcalib_xy, "If set hw-calibrated X/Y are used if available");
+
 /* device specifc data/functions */
 struct usbtouch_usb;
 struct usbtouch_device_info {
@@ -118,6 +122,7 @@
 
 #define USB_DEVICE_HID_CLASS(vend, prod) \
 	.match_flags = USB_DEVICE_ID_MATCH_INT_CLASS \
+		| USB_DEVICE_ID_MATCH_INT_PROTOCOL \
 		| USB_DEVICE_ID_MATCH_DEVICE, \
 	.idVendor = (vend), \
 	.idProduct = (prod), \
@@ -260,8 +265,13 @@
 
 static int mtouch_read_data(struct usbtouch_usb *dev, unsigned char *pkt)
 {
-	dev->x = (pkt[8] << 8) | pkt[7];
-	dev->y = (pkt[10] << 8) | pkt[9];
+	if (hwcalib_xy) {
+		dev->x = (pkt[4] << 8) | pkt[3];
+		dev->y = 0xffff - ((pkt[6] << 8) | pkt[5]);
+	} else {
+		dev->x = (pkt[8] << 8) | pkt[7];
+		dev->y = (pkt[10] << 8) | pkt[9];
+	}
 	dev->touch = (pkt[2] & 0x40) ? 1 : 0;
 
 	return 1;
@@ -294,6 +304,12 @@
 			return ret;
 	}
 
+	/* Default min/max xy are the raw values, override if using hw-calib */
+	if (hwcalib_xy) {
+		input_set_abs_params(usbtouch->input, ABS_X, 0, 0xffff, 0, 0);
+		input_set_abs_params(usbtouch->input, ABS_Y, 0, 0xffff, 0, 0);
+	}
+
 	return 0;
 }
 #endif
diff --git a/drivers/media/video/uvc/uvc_status.c b/drivers/media/video/uvc/uvc_status.c
index c1e4ae2..c705f24 100644
--- a/drivers/media/video/uvc/uvc_status.c
+++ b/drivers/media/video/uvc/uvc_status.c
@@ -46,8 +46,8 @@
 	usb_to_input_id(udev, &input->id);
 	input->dev.parent = &dev->intf->dev;
 
-	set_bit(EV_KEY, input->evbit);
-	set_bit(BTN_0, input->keybit);
+	__set_bit(EV_KEY, input->evbit);
+	__set_bit(KEY_CAMERA, input->keybit);
 
 	if ((ret = input_register_device(input)) < 0)
 		goto error;
@@ -70,8 +70,10 @@
 static void uvc_input_report_key(struct uvc_device *dev, unsigned int code,
 	int value)
 {
-	if (dev->input)
+	if (dev->input) {
 		input_report_key(dev->input, code, value);
+		input_sync(dev->input);
+	}
 }
 
 #else
@@ -96,7 +98,7 @@
 			return;
 		uvc_trace(UVC_TRACE_STATUS, "Button (intf %u) %s len %d\n",
 			data[1], data[3] ? "pressed" : "released", len);
-		uvc_input_report_key(dev, BTN_0, data[3]);
+		uvc_input_report_key(dev, KEY_CAMERA, data[3]);
 	} else {
 		uvc_trace(UVC_TRACE_STATUS, "Stream %u error event %02x %02x "
 			"len %d.\n", data[1], data[2], data[3], len);
diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c
index 96ac883..ea3aafb 100644
--- a/drivers/message/fusion/mptbase.c
+++ b/drivers/message/fusion/mptbase.c
@@ -91,9 +91,9 @@
 		controllers (default=0)");
 
 static int mpt_msi_enable_sas;
-module_param(mpt_msi_enable_sas, int, 1);
+module_param(mpt_msi_enable_sas, int, 0);
 MODULE_PARM_DESC(mpt_msi_enable_sas, " Enable MSI Support for SAS \
-		controllers (default=1)");
+		controllers (default=0)");
 
 
 static int mpt_channel_mapping;
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index 8cff5f5..406da9a 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -107,6 +107,7 @@
 
 static const struct sdhci_pci_fixes sdhci_cafe = {
 	.quirks		= SDHCI_QUIRK_NO_SIMULT_VDD_AND_POWER |
+			  SDHCI_QUIRK_NO_BUSY_IRQ |
 			  SDHCI_QUIRK_BROKEN_TIMEOUT_VAL,
 };
 
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index f52f305..accb592 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1291,8 +1291,11 @@
 		if (host->cmd->data)
 			DBG("Cannot wait for busy signal when also "
 				"doing a data transfer");
-		else
+		else if (!(host->quirks & SDHCI_QUIRK_NO_BUSY_IRQ))
 			return;
+
+		/* The controller does not support the end-of-busy IRQ,
+		 * fall through and take the SDHCI_INT_RESPONSE */
 	}
 
 	if (intmask & SDHCI_INT_RESPONSE)
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index ebb8365..43c37c6 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -208,6 +208,8 @@
 #define SDHCI_QUIRK_BROKEN_TIMEOUT_VAL			(1<<12)
 /* Controller has an issue with buffer bits for small transfers */
 #define SDHCI_QUIRK_BROKEN_SMALL_PIO			(1<<13)
+/* Controller does not provide transfer-complete interrupt when not busy */
+#define SDHCI_QUIRK_NO_BUSY_IRQ				(1<<14)
 
 	int			irq;		/* Device IRQ */
 	void __iomem *		ioaddr;		/* Mapped address */
diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index c38512e..dc5f051 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -1264,8 +1264,14 @@
 static void b44_chip_reset(struct b44 *bp, int reset_kind)
 {
 	struct ssb_device *sdev = bp->sdev;
+	bool was_enabled;
 
-	if (ssb_device_is_enabled(bp->sdev)) {
+	was_enabled = ssb_device_is_enabled(bp->sdev);
+
+	ssb_device_enable(bp->sdev, 0);
+	ssb_pcicore_dev_irqvecs_enable(&sdev->bus->pcicore, sdev);
+
+	if (was_enabled) {
 		bw32(bp, B44_RCV_LAZY, 0);
 		bw32(bp, B44_ENET_CTRL, ENET_CTRL_DISABLE);
 		b44_wait_bit(bp, B44_ENET_CTRL, ENET_CTRL_DISABLE, 200, 1);
@@ -1277,10 +1283,8 @@
 		}
 		bw32(bp, B44_DMARX_CTRL, 0);
 		bp->rx_prod = bp->rx_cons = 0;
-	} else
-		ssb_pcicore_dev_irqvecs_enable(&sdev->bus->pcicore, sdev);
+	}
 
-	ssb_device_enable(bp->sdev, 0);
 	b44_clear_stats(bp);
 
 	/*
@@ -2236,6 +2240,7 @@
 	struct net_device *dev = ssb_get_drvdata(sdev);
 
 	unregister_netdev(dev);
+	ssb_device_disable(sdev, 0);
 	ssb_bus_may_powerdown(sdev->bus);
 	free_netdev(dev);
 	ssb_pcihost_set_power_state(sdev, PCI_D3hot);
diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index 9b12a13..9831b3f 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -1284,7 +1284,7 @@
 	spin_lock_irqsave(&priv->txlock, flags);
 
 	/* check if there is space to queue this packet */
-	if (nr_frags > priv->num_txbdfree) {
+	if ((nr_frags+1) > priv->num_txbdfree) {
 		/* no space, stop the queue */
 		netif_stop_queue(dev);
 		dev->stats.tx_fifo_errors++;
diff --git a/drivers/net/hp-plus.c b/drivers/net/hp-plus.c
index 5e070f4..0486cbe 100644
--- a/drivers/net/hp-plus.c
+++ b/drivers/net/hp-plus.c
@@ -467,7 +467,7 @@
 			if (this_dev != 0) break; /* only autoprobe 1st one */
 			printk(KERN_NOTICE "hp-plus.c: Presently autoprobing (not recommended) for a single card.\n");
 		}
-		dev = alloc_ei_netdev();
+		dev = alloc_eip_netdev();
 		if (!dev)
 			break;
 		dev->irq = irq[this_dev];
diff --git a/drivers/net/netxen/netxen_nic_main.c b/drivers/net/netxen/netxen_nic_main.c
index 9f33e44..1308778 100644
--- a/drivers/net/netxen/netxen_nic_main.c
+++ b/drivers/net/netxen/netxen_nic_main.c
@@ -588,7 +588,12 @@
 		adapter->pci_mem_read = netxen_nic_pci_mem_read_2M;
 		adapter->pci_mem_write = netxen_nic_pci_mem_write_2M;
 
-		mem_ptr0 = ioremap(mem_base, mem_len);
+		mem_ptr0 = pci_ioremap_bar(pdev, 0);
+		if (mem_ptr0 == NULL) {
+			dev_err(&pdev->dev, "failed to map PCI bar 0\n");
+			return -EIO;
+		}
+
 		pci_len0 = mem_len;
 		first_page_group_start = 0;
 		first_page_group_end   = 0;
@@ -795,9 +800,12 @@
 	 * See if the firmware gave us a virtual-physical port mapping.
 	 */
 	adapter->physical_port = adapter->portnum;
-	i = adapter->pci_read_normalize(adapter, CRB_V2P(adapter->portnum));
-	if (i != 0x55555555)
-		adapter->physical_port = i;
+	if (adapter->fw_major < 4) {
+		i = adapter->pci_read_normalize(adapter,
+				CRB_V2P(adapter->portnum));
+		if (i != 0x55555555)
+			adapter->physical_port = i;
+	}
 
 	adapter->flags &= ~(NETXEN_NIC_MSI_ENABLED | NETXEN_NIC_MSIX_ENABLED);
 
diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 0771eb6..b347340 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -81,9 +81,9 @@
 #define RTL8169_TX_TIMEOUT	(6*HZ)
 #define RTL8169_PHY_TIMEOUT	(10*HZ)
 
-#define RTL_EEPROM_SIG		cpu_to_le32(0x8129)
-#define RTL_EEPROM_SIG_MASK	cpu_to_le32(0xffff)
+#define RTL_EEPROM_SIG		0x8129
 #define RTL_EEPROM_SIG_ADDR	0x0000
+#define RTL_EEPROM_MAC_ADDR	0x0007
 
 /* write/read MMIO register */
 #define RTL_W8(reg, val8)	writeb ((val8), ioaddr + (reg))
@@ -293,6 +293,11 @@
 	/* Cfg9346Bits */
 	Cfg9346_Lock	= 0x00,
 	Cfg9346_Unlock	= 0xc0,
+	Cfg9346_Program	= 0x80,		/* Programming mode */
+	Cfg9346_EECS	= 0x08,		/* Chip select */
+	Cfg9346_EESK	= 0x04,		/* Serial data clock */
+	Cfg9346_EEDI	= 0x02,		/* Data input */
+	Cfg9346_EEDO	= 0x01,		/* Data output */
 
 	/* rx_mode_bits */
 	AcceptErr	= 0x20,
@@ -305,6 +310,7 @@
 	/* RxConfigBits */
 	RxCfgFIFOShift	= 13,
 	RxCfgDMAShift	=  8,
+	RxCfg9356SEL	=  6, 		/* EEPROM type: 0 = 9346, 1 = 9356 */
 
 	/* TxConfigBits */
 	TxInterFrameGapShift = 24,
@@ -1963,6 +1969,108 @@
 
 };
 
+/* Delay between EEPROM clock transitions. Force out buffered PCI writes. */
+#define RTL_EEPROM_DELAY()	RTL_R8(Cfg9346)
+#define RTL_EEPROM_READ_CMD	6
+
+/* read 16bit word stored in EEPROM. EEPROM is addressed by words. */
+static u16 rtl_eeprom_read(void __iomem *ioaddr, int addr)
+{
+	u16 result = 0;
+	int cmd, cmd_len, i;
+
+	/* check for EEPROM address size (in bits) */
+	if (RTL_R32(RxConfig) & (1 << RxCfg9356SEL)) {
+		/* EEPROM is 93C56 */
+		cmd_len = 3 + 8; /* 3 bits for command id and 8 for address */
+		cmd = (RTL_EEPROM_READ_CMD << 8) | (addr & 0xff);
+	} else {
+		/* EEPROM is 93C46 */
+		cmd_len = 3 + 6; /* 3 bits for command id and 6 for address */
+		cmd = (RTL_EEPROM_READ_CMD << 6) | (addr & 0x3f);
+	}
+
+	/* enter programming mode */
+	RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS);
+	RTL_EEPROM_DELAY();
+
+	/* write command and requested address */
+	while (cmd_len--) {
+		u8 x = Cfg9346_Program | Cfg9346_EECS;
+
+		x |= (cmd & (1 << cmd_len)) ? Cfg9346_EEDI : 0;
+
+		/* write a bit */
+		RTL_W8(Cfg9346, x);
+		RTL_EEPROM_DELAY();
+
+		/* raise clock */
+		RTL_W8(Cfg9346, x | Cfg9346_EESK);
+		RTL_EEPROM_DELAY();
+	}
+
+	/* lower clock */
+	RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS);
+	RTL_EEPROM_DELAY();
+
+	/* read back 16bit value */
+	for (i = 16; i > 0; i--) {
+		/* raise clock */
+		RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS | Cfg9346_EESK);
+		RTL_EEPROM_DELAY();
+
+		result <<= 1;
+		result |= (RTL_R8(Cfg9346) & Cfg9346_EEDO) ? 1 : 0;
+
+		/* lower clock */
+		RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS);
+		RTL_EEPROM_DELAY();
+	}
+
+	RTL_W8(Cfg9346, Cfg9346_Program);
+	/* leave programming mode */
+	RTL_W8(Cfg9346, Cfg9346_Lock);
+
+	return result;
+}
+
+static void rtl_init_mac_address(struct rtl8169_private *tp,
+				 void __iomem *ioaddr)
+{
+	struct pci_dev *pdev = tp->pci_dev;
+	u16 x;
+	u8 mac[8];
+
+	/* read EEPROM signature */
+	x = rtl_eeprom_read(ioaddr, RTL_EEPROM_SIG_ADDR);
+
+	if (x != RTL_EEPROM_SIG) {
+		dev_info(&pdev->dev, "Missing EEPROM signature: %04x\n", x);
+		return;
+	}
+
+	/* read MAC address */
+	x = rtl_eeprom_read(ioaddr, RTL_EEPROM_MAC_ADDR);
+	mac[0] = x & 0xff;
+	mac[1] = x >> 8;
+	x = rtl_eeprom_read(ioaddr, RTL_EEPROM_MAC_ADDR + 1);
+	mac[2] = x & 0xff;
+	mac[3] = x >> 8;
+	x = rtl_eeprom_read(ioaddr, RTL_EEPROM_MAC_ADDR + 2);
+	mac[4] = x & 0xff;
+	mac[5] = x >> 8;
+
+	if (netif_msg_probe(tp)) {
+		DECLARE_MAC_BUF(buf);
+
+		dev_info(&pdev->dev, "MAC address found in EEPROM: %s\n",
+			 print_mac(buf, mac));
+	}
+
+	if (is_valid_ether_addr(mac))
+		rtl_rar_set(tp, mac);
+}
+
 static int __devinit
 rtl8169_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
@@ -2141,6 +2249,8 @@
 
 	tp->mmio_addr = ioaddr;
 
+	rtl_init_mac_address(tp, ioaddr);
+
 	/* Get MAC address */
 	for (i = 0; i < MAC_ADDR_LEN; i++)
 		dev->dev_addr[i] = RTL_R8(MAC0 + i);
diff --git a/drivers/net/usb/asix.c b/drivers/net/usb/asix.c
index e009481..396f821 100644
--- a/drivers/net/usb/asix.c
+++ b/drivers/net/usb/asix.c
@@ -1451,6 +1451,14 @@
 	// Cables-to-Go USB Ethernet Adapter
 	USB_DEVICE(0x0b95, 0x772a),
 	.driver_info = (unsigned long) &ax88772_info,
+}, {
+	// ABOCOM for pci
+	USB_DEVICE(0x14ea, 0xab11),
+	.driver_info = (unsigned long) &ax88178_info,
+}, {
+	// ASIX 88772a
+	USB_DEVICE(0x0db0, 0xa877),
+	.driver_info = (unsigned long) &ax88772_info,
 },
 	{ },		// END
 };
diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c
index 0e061df..55e8ecc 100644
--- a/drivers/net/usb/cdc_ether.c
+++ b/drivers/net/usb/cdc_ether.c
@@ -559,6 +559,11 @@
 	USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_ETHERNET,
 			USB_CDC_PROTO_NONE),
 	.driver_info = (unsigned long) &cdc_info,
+}, {
+	/* Ericsson F3507g */
+	USB_DEVICE_AND_INTERFACE_INFO(0x0bdb, 0x1900, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+	.driver_info = (unsigned long) &cdc_info,
 },
 	{ },		// END
 };
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index aa31490..c32284f 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -723,8 +723,8 @@
 	if (dev->mii.mdio_read)
 		return mii_link_ok(&dev->mii);
 
-	/* Otherwise, say we're up (to avoid breaking scripts) */
-	return 1;
+	/* Otherwise, dtrt for drivers calling netif_carrier_{on,off} */
+	return ethtool_op_get_link(net);
 }
 EXPORT_SYMBOL_GPL(usbnet_get_link);
 
diff --git a/drivers/net/usb/zaurus.c b/drivers/net/usb/zaurus.c
index e24f7b3..04882c8 100644
--- a/drivers/net/usb/zaurus.c
+++ b/drivers/net/usb/zaurus.c
@@ -341,6 +341,11 @@
 	USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_MDLM,
 			USB_CDC_PROTO_NONE),
 	.driver_info = (unsigned long) &bogus_mdlm_info,
+}, {
+	/* Motorola MOTOMAGX phones */
+	USB_DEVICE_AND_INTERFACE_INFO(0x22b8, 0x6425, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+	.driver_info = (unsigned long) &bogus_mdlm_info,
 },
 
 /* Olympus has some models with a Zaurus-compatible option.
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 108bbbea..124fe75 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -239,6 +239,16 @@
 	return 0;
 }
 
+static int veth_close(struct net_device *dev)
+{
+	struct veth_priv *priv = netdev_priv(dev);
+
+	netif_carrier_off(dev);
+	netif_carrier_off(priv->peer);
+
+	return 0;
+}
+
 static int veth_dev_init(struct net_device *dev)
 {
 	struct veth_net_stats *stats;
@@ -265,6 +275,7 @@
 static const struct net_device_ops veth_netdev_ops = {
 	.ndo_init            = veth_dev_init,
 	.ndo_open            = veth_open,
+	.ndo_stop            = veth_close,
 	.ndo_start_xmit      = veth_xmit,
 	.ndo_get_stats       = veth_get_stats,
 	.ndo_set_mac_address = eth_mac_addr,
@@ -280,44 +291,6 @@
 	dev->destructor = veth_dev_free;
 }
 
-static void veth_change_state(struct net_device *dev)
-{
-	struct net_device *peer;
-	struct veth_priv *priv;
-
-	priv = netdev_priv(dev);
-	peer = priv->peer;
-
-	if (netif_carrier_ok(peer)) {
-		if (!netif_carrier_ok(dev))
-			netif_carrier_on(dev);
-	} else {
-		if (netif_carrier_ok(dev))
-			netif_carrier_off(dev);
-	}
-}
-
-static int veth_device_event(struct notifier_block *unused,
-			     unsigned long event, void *ptr)
-{
-	struct net_device *dev = ptr;
-
-	if (dev->netdev_ops->ndo_open != veth_open)
-		goto out;
-
-	switch (event) {
-	case NETDEV_CHANGE:
-		veth_change_state(dev);
-		break;
-	}
-out:
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block veth_notifier_block __read_mostly = {
-	.notifier_call	= veth_device_event,
-};
-
 /*
  * netlink interface
  */
@@ -468,14 +441,12 @@
 
 static __init int veth_init(void)
 {
-	register_netdevice_notifier(&veth_notifier_block);
 	return rtnl_link_register(&veth_link_ops);
 }
 
 static __exit void veth_exit(void)
 {
 	rtnl_link_unregister(&veth_link_ops);
-	unregister_netdevice_notifier(&veth_notifier_block);
 }
 
 module_init(veth_init);
diff --git a/drivers/net/wireless/ath9k/main.c b/drivers/net/wireless/ath9k/main.c
index 727f067..0e80990 100644
--- a/drivers/net/wireless/ath9k/main.c
+++ b/drivers/net/wireless/ath9k/main.c
@@ -1538,6 +1538,7 @@
 bad:
 	if (ah)
 		ath9k_hw_detach(ah);
+	ath9k_exit_debug(sc);
 
 	return error;
 }
@@ -1545,7 +1546,7 @@
 static int ath_attach(u16 devid, struct ath_softc *sc)
 {
 	struct ieee80211_hw *hw = sc->hw;
-	int error = 0;
+	int error = 0, i;
 
 	DPRINTF(sc, ATH_DBG_CONFIG, "Attach ATH hw\n");
 
@@ -1589,11 +1590,11 @@
 	/* initialize tx/rx engine */
 	error = ath_tx_init(sc, ATH_TXBUF);
 	if (error != 0)
-		goto detach;
+		goto error_attach;
 
 	error = ath_rx_init(sc, ATH_RXBUF);
 	if (error != 0)
-		goto detach;
+		goto error_attach;
 
 #if defined(CONFIG_RFKILL) || defined(CONFIG_RFKILL_MODULE)
 	/* Initialze h/w Rfkill */
@@ -1601,8 +1602,9 @@
 		INIT_DELAYED_WORK(&sc->rf_kill.rfkill_poll, ath_rfkill_poll);
 
 	/* Initialize s/w rfkill */
-	if (ath_init_sw_rfkill(sc))
-		goto detach;
+	error = ath_init_sw_rfkill(sc);
+	if (error)
+		goto error_attach;
 #endif
 
 	error = ieee80211_register_hw(hw);
@@ -1611,8 +1613,16 @@
 	ath_init_leds(sc);
 
 	return 0;
-detach:
-	ath_detach(sc);
+
+error_attach:
+	/* cleanup tx queues */
+	for (i = 0; i < ATH9K_NUM_TX_QUEUES; i++)
+		if (ATH_TXQ_SETUP(sc, i))
+			ath_tx_cleanupq(sc, &sc->tx.txq[i]);
+
+	ath9k_hw_detach(sc->sc_ah);
+	ath9k_exit_debug(sc);
+
 	return error;
 }
 
diff --git a/drivers/net/wireless/libertas/ethtool.c b/drivers/net/wireless/libertas/ethtool.c
index 61d2f50..b118a35 100644
--- a/drivers/net/wireless/libertas/ethtool.c
+++ b/drivers/net/wireless/libertas/ethtool.c
@@ -23,7 +23,7 @@
 static void lbs_ethtool_get_drvinfo(struct net_device *dev,
 					 struct ethtool_drvinfo *info)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	snprintf(info->fw_version, 32, "%u.%u.%u.p%u",
 		priv->fwrelease >> 24 & 0xff,
@@ -47,7 +47,7 @@
 static int lbs_ethtool_get_eeprom(struct net_device *dev,
                                   struct ethtool_eeprom *eeprom, u8 * bytes)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct cmd_ds_802_11_eeprom_access cmd;
 	int ret;
 
@@ -76,7 +76,7 @@
 static void lbs_ethtool_get_stats(struct net_device *dev,
 				  struct ethtool_stats *stats, uint64_t *data)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct cmd_ds_mesh_access mesh_access;
 	int ret;
 
@@ -113,7 +113,7 @@
 
 static int lbs_ethtool_get_sset_count(struct net_device *dev, int sset)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	if (sset == ETH_SS_STATS && dev == priv->mesh_dev)
 		return MESH_STATS_NUM;
@@ -143,7 +143,7 @@
 static void lbs_ethtool_get_wol(struct net_device *dev,
 				struct ethtool_wolinfo *wol)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	if (priv->wol_criteria == 0xffffffff) {
 		/* Interface driver didn't configure wake */
@@ -166,7 +166,7 @@
 static int lbs_ethtool_set_wol(struct net_device *dev,
 			       struct ethtool_wolinfo *wol)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	uint32_t criteria = 0;
 
 	if (priv->wol_criteria == 0xffffffff && wol->wolopts)
diff --git a/drivers/net/wireless/libertas/if_usb.c b/drivers/net/wireless/libertas/if_usb.c
index 2fc637a..ea3dc038 100644
--- a/drivers/net/wireless/libertas/if_usb.c
+++ b/drivers/net/wireless/libertas/if_usb.c
@@ -59,7 +59,7 @@
 static ssize_t if_usb_firmware_set(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct if_usb_card *cardp = priv->card;
 	char fwname[FIRMWARE_NAME_MAX];
 	int ret;
@@ -86,7 +86,7 @@
 static ssize_t if_usb_boot2_set(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct if_usb_card *cardp = priv->card;
 	char fwname[FIRMWARE_NAME_MAX];
 	int ret;
diff --git a/drivers/net/wireless/libertas/main.c b/drivers/net/wireless/libertas/main.c
index 4e0007d..f76623e 100644
--- a/drivers/net/wireless/libertas/main.c
+++ b/drivers/net/wireless/libertas/main.c
@@ -222,7 +222,7 @@
 static ssize_t lbs_anycast_get(struct device *dev,
 		struct device_attribute *attr, char * buf)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_access mesh_access;
 	int ret;
 
@@ -241,7 +241,7 @@
 static ssize_t lbs_anycast_set(struct device *dev,
 		struct device_attribute *attr, const char * buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_access mesh_access;
 	uint32_t datum;
 	int ret;
@@ -263,7 +263,7 @@
 static ssize_t lbs_prb_rsp_limit_get(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_access mesh_access;
 	int ret;
 	u32 retry_limit;
@@ -286,7 +286,7 @@
 static ssize_t lbs_prb_rsp_limit_set(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_access mesh_access;
 	int ret;
 	unsigned long retry_limit;
@@ -321,7 +321,7 @@
 static ssize_t lbs_rtap_get(struct device *dev,
 		struct device_attribute *attr, char * buf)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	return snprintf(buf, 5, "0x%X\n", priv->monitormode);
 }
 
@@ -332,7 +332,7 @@
 		struct device_attribute *attr, const char * buf, size_t count)
 {
 	int monitor_mode;
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 
 	sscanf(buf, "%x", &monitor_mode);
 	if (monitor_mode) {
@@ -383,7 +383,7 @@
 static ssize_t lbs_mesh_get(struct device *dev,
 		struct device_attribute *attr, char * buf)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	return snprintf(buf, 5, "0x%X\n", !!priv->mesh_dev);
 }
 
@@ -393,7 +393,7 @@
 static ssize_t lbs_mesh_set(struct device *dev,
 		struct device_attribute *attr, const char * buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	int enable;
 	int ret, action = CMD_ACT_MESH_CONFIG_STOP;
 
@@ -452,7 +452,7 @@
  */
 static int lbs_dev_open(struct net_device *dev)
 {
-	struct lbs_private *priv = netdev_priv(dev) ;
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 
 	lbs_deb_enter(LBS_DEB_NET);
@@ -521,7 +521,7 @@
  */
 static int lbs_eth_stop(struct net_device *dev)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_NET);
 
@@ -538,7 +538,7 @@
 
 static void lbs_tx_timeout(struct net_device *dev)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_TX);
 
@@ -590,7 +590,7 @@
  */
 static struct net_device_stats *lbs_get_stats(struct net_device *dev)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_NET);
 	return &priv->stats;
@@ -599,7 +599,7 @@
 static int lbs_set_mac_address(struct net_device *dev, void *addr)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct sockaddr *phwaddr = addr;
 	struct cmd_ds_802_11_mac_address cmd;
 
@@ -732,7 +732,7 @@
 
 static void lbs_set_multicast_list(struct net_device *dev)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	schedule_work(&priv->mcast_work);
 }
@@ -748,7 +748,7 @@
 static int lbs_thread(void *data)
 {
 	struct net_device *dev = data;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	wait_queue_t wait;
 
 	lbs_deb_enter(LBS_DEB_THREAD);
@@ -1184,6 +1184,7 @@
 		goto done;
 	}
 	priv = netdev_priv(dev);
+	dev->ml_priv = priv;
 
 	if (lbs_init_adapter(priv)) {
 		lbs_pr_err("failed to initialize adapter structure.\n");
diff --git a/drivers/net/wireless/libertas/persistcfg.c b/drivers/net/wireless/libertas/persistcfg.c
index d42b7a5..18fe29f 100644
--- a/drivers/net/wireless/libertas/persistcfg.c
+++ b/drivers/net/wireless/libertas/persistcfg.c
@@ -18,7 +18,7 @@
 static int mesh_get_default_parameters(struct device *dev,
 				       struct mrvl_mesh_defaults *defs)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_config cmd;
 	int ret;
 
@@ -57,7 +57,7 @@
 static ssize_t bootflag_set(struct device *dev, struct device_attribute *attr,
 			    const char *buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_config cmd;
 	uint32_t datum;
 	int ret;
@@ -100,7 +100,7 @@
 static ssize_t boottime_set(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_config cmd;
 	uint32_t datum;
 	int ret;
@@ -152,7 +152,7 @@
 static ssize_t channel_set(struct device *dev, struct device_attribute *attr,
 			   const char *buf, size_t count)
 {
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	struct cmd_ds_mesh_config cmd;
 	uint32_t datum;
 	int ret;
@@ -210,7 +210,7 @@
 	struct cmd_ds_mesh_config cmd;
 	struct mrvl_mesh_defaults defs;
 	struct mrvl_meshie *ie;
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	int len;
 	int ret;
 
@@ -269,7 +269,7 @@
 	struct cmd_ds_mesh_config cmd;
 	struct mrvl_mesh_defaults defs;
 	struct mrvl_meshie *ie;
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	uint32_t datum;
 	int ret;
 
@@ -323,7 +323,7 @@
 	struct cmd_ds_mesh_config cmd;
 	struct mrvl_mesh_defaults defs;
 	struct mrvl_meshie *ie;
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	uint32_t datum;
 	int ret;
 
@@ -377,7 +377,7 @@
 	struct cmd_ds_mesh_config cmd;
 	struct mrvl_mesh_defaults defs;
 	struct mrvl_meshie *ie;
-	struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+	struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 	uint32_t datum;
 	int ret;
 
diff --git a/drivers/net/wireless/libertas/scan.c b/drivers/net/wireless/libertas/scan.c
index 57f6c12..9014950 100644
--- a/drivers/net/wireless/libertas/scan.c
+++ b/drivers/net/wireless/libertas/scan.c
@@ -945,7 +945,7 @@
 		 union iwreq_data *wrqu, char *extra)
 {
 	DECLARE_SSID_BUF(ssid);
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
@@ -1008,7 +1008,7 @@
 		 struct iw_point *dwrq, char *extra)
 {
 #define SCAN_ITEM_SIZE 128
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int err = 0;
 	char *ev = extra;
 	char *stop = ev + dwrq->length;
diff --git a/drivers/net/wireless/libertas/tx.c b/drivers/net/wireless/libertas/tx.c
index dac4626..68bec31 100644
--- a/drivers/net/wireless/libertas/tx.c
+++ b/drivers/net/wireless/libertas/tx.c
@@ -60,7 +60,7 @@
 int lbs_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	unsigned long flags;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct txpd *txpd;
 	char *p802x_hdr;
 	uint16_t pkt_len;
diff --git a/drivers/net/wireless/libertas/wext.c b/drivers/net/wireless/libertas/wext.c
index c6102e0..f16d136 100644
--- a/drivers/net/wireless/libertas/wext.c
+++ b/drivers/net/wireless/libertas/wext.c
@@ -163,7 +163,7 @@
 static int lbs_get_freq(struct net_device *dev, struct iw_request_info *info,
 			 struct iw_freq *fwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct chan_freq_power *cfp;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
@@ -189,7 +189,7 @@
 static int lbs_get_wap(struct net_device *dev, struct iw_request_info *info,
 			struct sockaddr *awrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -207,7 +207,7 @@
 static int lbs_set_nick(struct net_device *dev, struct iw_request_info *info,
 			 struct iw_point *dwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -231,7 +231,7 @@
 static int lbs_get_nick(struct net_device *dev, struct iw_request_info *info,
 			 struct iw_point *dwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -248,7 +248,7 @@
 static int mesh_get_nick(struct net_device *dev, struct iw_request_info *info,
 			 struct iw_point *dwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -273,7 +273,7 @@
 			struct iw_param *vwrq, char *extra)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	u32 val = vwrq->value;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
@@ -293,7 +293,7 @@
 static int lbs_get_rts(struct net_device *dev, struct iw_request_info *info,
 			struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 	u16 val = 0;
 
@@ -315,7 +315,7 @@
 static int lbs_set_frag(struct net_device *dev, struct iw_request_info *info,
 			 struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 	u32 val = vwrq->value;
 
@@ -336,7 +336,7 @@
 static int lbs_get_frag(struct net_device *dev, struct iw_request_info *info,
 			 struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 	u16 val = 0;
 
@@ -359,7 +359,7 @@
 static int lbs_get_mode(struct net_device *dev,
 			 struct iw_request_info *info, u32 * uwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -385,7 +385,7 @@
 			  struct iw_request_info *info,
 			  struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	s16 curlevel = 0;
 	int ret = 0;
 
@@ -418,7 +418,7 @@
 static int lbs_set_retry(struct net_device *dev, struct iw_request_info *info,
 			  struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 	u16 slimit = 0, llimit = 0;
 
@@ -466,7 +466,7 @@
 static int lbs_get_retry(struct net_device *dev, struct iw_request_info *info,
 			  struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 	u16 val = 0;
 
@@ -542,7 +542,7 @@
 			  struct iw_point *dwrq, char *extra)
 {
 	int i, j;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct iw_range *range = (struct iw_range *)extra;
 	struct chan_freq_power *cfp;
 	u8 rates[MAX_RATES + 1];
@@ -708,7 +708,7 @@
 static int lbs_set_power(struct net_device *dev, struct iw_request_info *info,
 			  struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -758,7 +758,7 @@
 static int lbs_get_power(struct net_device *dev, struct iw_request_info *info,
 			  struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -781,7 +781,7 @@
 		EXCELLENT = 95,
 		PERFECT = 100
 	};
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	u32 rssi_qual;
 	u32 tx_qual;
 	u32 quality = 0;
@@ -886,7 +886,7 @@
 		  struct iw_freq *fwrq, char *extra)
 {
 	int ret = -EINVAL;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct chan_freq_power *cfp;
 	struct assoc_request * assoc_req;
 
@@ -943,7 +943,7 @@
 			     struct iw_request_info *info,
 			     struct iw_freq *fwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct chan_freq_power *cfp;
 	int ret = -EINVAL;
 
@@ -994,7 +994,7 @@
 static int lbs_set_rate(struct net_device *dev, struct iw_request_info *info,
 		  struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	u8 new_rate = 0;
 	int ret = -EINVAL;
 	u8 rates[MAX_RATES + 1];
@@ -1054,7 +1054,7 @@
 static int lbs_get_rate(struct net_device *dev, struct iw_request_info *info,
 		  struct iw_param *vwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1079,7 +1079,7 @@
 		  struct iw_request_info *info, u32 * uwrq, char *extra)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct assoc_request * assoc_req;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
@@ -1124,7 +1124,7 @@
 			   struct iw_request_info *info,
 			   struct iw_point *dwrq, u8 * extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int index = (dwrq->flags & IW_ENCODE_INDEX) - 1;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
@@ -1319,7 +1319,7 @@
 		    struct iw_point *dwrq, char *extra)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct assoc_request * assoc_req;
 	u16 is_default = 0, index = 0, set_tx_key = 0;
 
@@ -1395,7 +1395,7 @@
 			      char *extra)
 {
 	int ret = -EINVAL;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct iw_encode_ext *ext = (struct iw_encode_ext *)extra;
 	int index, max_key_len;
 
@@ -1501,7 +1501,7 @@
 			      char *extra)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct iw_encode_ext *ext = (struct iw_encode_ext *)extra;
 	int alg = ext->alg;
 	struct assoc_request * assoc_req;
@@ -1639,7 +1639,7 @@
 			  struct iw_point *dwrq,
 			  char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 	struct assoc_request * assoc_req;
 
@@ -1685,7 +1685,7 @@
 			  char *extra)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1713,7 +1713,7 @@
 			 struct iw_param *dwrq,
 			 char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct assoc_request * assoc_req;
 	int ret = 0;
 	int updated = 0;
@@ -1816,7 +1816,7 @@
 			 char *extra)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1857,7 +1857,7 @@
 		   struct iw_param *vwrq, char *extra)
 {
 	int ret = 0;
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	s16 dbm = (s16) vwrq->value;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
@@ -1936,7 +1936,7 @@
 static int lbs_get_essid(struct net_device *dev, struct iw_request_info *info,
 		   struct iw_point *dwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1971,7 +1971,7 @@
 static int lbs_set_essid(struct net_device *dev, struct iw_request_info *info,
 		   struct iw_point *dwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 	u8 ssid[IW_ESSID_MAX_SIZE];
 	u8 ssid_len = 0;
@@ -2040,7 +2040,7 @@
 			      struct iw_request_info *info,
 			      struct iw_point *dwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -2058,7 +2058,7 @@
 			      struct iw_request_info *info,
 			      struct iw_point *dwrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	int ret = 0;
 
 	lbs_deb_enter(LBS_DEB_WEXT);
@@ -2102,7 +2102,7 @@
 static int lbs_set_wap(struct net_device *dev, struct iw_request_info *info,
 		 struct sockaddr *awrq, char *extra)
 {
-	struct lbs_private *priv = netdev_priv(dev);
+	struct lbs_private *priv = dev->ml_priv;
 	struct assoc_request * assoc_req;
 	int ret = 0;
 
diff --git a/drivers/net/wireless/orinoco/orinoco.c b/drivers/net/wireless/orinoco/orinoco.c
index 45a04fa..067d1a9 100644
--- a/drivers/net/wireless/orinoco/orinoco.c
+++ b/drivers/net/wireless/orinoco/orinoco.c
@@ -3157,8 +3157,20 @@
 
 	return NOTIFY_DONE;
 }
+
+static void orinoco_register_pm_notifier(struct orinoco_private *priv)
+{
+	priv->pm_notifier.notifier_call = orinoco_pm_notifier;
+	register_pm_notifier(&priv->pm_notifier);
+}
+
+static void orinoco_unregister_pm_notifier(struct orinoco_private *priv)
+{
+	unregister_pm_notifier(&priv->pm_notifier);
+}
 #else /* !PM_SLEEP || HERMES_CACHE_FW_ON_INIT */
-#define orinoco_pm_notifier NULL
+#define orinoco_register_pm_notifier(priv) do { } while(0)
+#define orinoco_unregister_pm_notifier(priv) do { } while(0)
 #endif
 
 /********************************************************************/
@@ -3648,8 +3660,7 @@
 	priv->cached_fw = NULL;
 
 	/* Register PM notifiers */
-	priv->pm_notifier.notifier_call = orinoco_pm_notifier;
-	register_pm_notifier(&priv->pm_notifier);
+	orinoco_register_pm_notifier(priv);
 
 	return dev;
 }
@@ -3673,7 +3684,7 @@
 		kfree(rx_data);
 	}
 
-	unregister_pm_notifier(&priv->pm_notifier);
+	orinoco_unregister_pm_notifier(priv);
 	orinoco_uncache_fw(priv);
 
 	priv->wpa_ie_len = 0;
diff --git a/drivers/net/wireless/rtl818x/rtl8187_dev.c b/drivers/net/wireless/rtl818x/rtl8187_dev.c
index 22bc07e..f4747a1 100644
--- a/drivers/net/wireless/rtl818x/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl818x/rtl8187_dev.c
@@ -48,6 +48,10 @@
 	{USB_DEVICE(0x0bda, 0x8189), .driver_info = DEVICE_RTL8187B},
 	{USB_DEVICE(0x0bda, 0x8197), .driver_info = DEVICE_RTL8187B},
 	{USB_DEVICE(0x0bda, 0x8198), .driver_info = DEVICE_RTL8187B},
+	/* Surecom */
+	{USB_DEVICE(0x0769, 0x11F2), .driver_info = DEVICE_RTL8187},
+	/* Logitech */
+	{USB_DEVICE(0x0789, 0x010C), .driver_info = DEVICE_RTL8187},
 	/* Netgear */
 	{USB_DEVICE(0x0846, 0x6100), .driver_info = DEVICE_RTL8187},
 	{USB_DEVICE(0x0846, 0x6a00), .driver_info = DEVICE_RTL8187},
@@ -57,8 +61,16 @@
 	/* Sitecom */
 	{USB_DEVICE(0x0df6, 0x000d), .driver_info = DEVICE_RTL8187},
 	{USB_DEVICE(0x0df6, 0x0028), .driver_info = DEVICE_RTL8187B},
+	/* Sphairon Access Systems GmbH */
+	{USB_DEVICE(0x114B, 0x0150), .driver_info = DEVICE_RTL8187},
+	/* Dick Smith Electronics */
+	{USB_DEVICE(0x1371, 0x9401), .driver_info = DEVICE_RTL8187},
 	/* Abocom */
 	{USB_DEVICE(0x13d1, 0xabe6), .driver_info = DEVICE_RTL8187},
+	/* Qcom */
+	{USB_DEVICE(0x18E8, 0x6232), .driver_info = DEVICE_RTL8187},
+	/* AirLive */
+	{USB_DEVICE(0x1b75, 0x8187), .driver_info = DEVICE_RTL8187},
 	{}
 };
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i.h b/drivers/scsi/cxgb3i/cxgb3i.h
index fde6e4c..a7cf550 100644
--- a/drivers/scsi/cxgb3i/cxgb3i.h
+++ b/drivers/scsi/cxgb3i/cxgb3i.h
@@ -20,6 +20,7 @@
 #include <linux/list.h>
 #include <linux/netdevice.h>
 #include <linux/scatterlist.h>
+#include <linux/skbuff.h>
 #include <scsi/libiscsi_tcp.h>
 
 /* from cxgb3 LLD */
@@ -113,6 +114,26 @@
 	struct cxgb3i_conn *cconn;
 };
 
+/**
+ * struct cxgb3i_task_data - private iscsi task data
+ *
+ * @nr_frags:	# of coalesced page frags (from scsi sgl)
+ * @frags:	coalesced page frags (from scsi sgl)
+ * @skb:	tx pdu skb
+ * @offset:	data offset for the next pdu
+ * @count:	max. possible pdu payload
+ * @sgoffset:	offset to the first sg entry for a given offset
+ */
+#define MAX_PDU_FRAGS	((ULP2_MAX_PDU_PAYLOAD + 512 - 1) / 512)
+struct cxgb3i_task_data {
+	unsigned short nr_frags;
+	skb_frag_t frags[MAX_PDU_FRAGS];
+	struct sk_buff *skb;
+	unsigned int offset;
+	unsigned int count;
+	unsigned int sgoffset;
+};
+
 int cxgb3i_iscsi_init(void);
 void cxgb3i_iscsi_cleanup(void);
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i_ddp.c b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
index 08f3a09..a83d36e 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_ddp.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
@@ -639,10 +639,11 @@
 	write_unlock(&cxgb3i_ddp_rwlock);
 
 	ddp_log_info("nppods %u (0x%x ~ 0x%x), bits %u, mask 0x%x,0x%x "
-			"pkt %u,%u.\n",
+			"pkt %u/%u, %u/%u.\n",
 			ppmax, ddp->llimit, ddp->ulimit, ddp->idx_bits,
 			ddp->idx_mask, ddp->rsvd_tag_mask,
-			ddp->max_txsz, ddp->max_rxsz);
+			ddp->max_txsz, uinfo.max_txsz,
+			ddp->max_rxsz, uinfo.max_rxsz);
 	return 0;
 
 free_ddp_map:
@@ -654,8 +655,8 @@
  * cxgb3i_adapter_ddp_init - initialize the adapter's ddp resource
  * @tdev: t3cdev adapter
  * @tformat: tag format
- * @txsz: max tx pkt size, filled in by this func.
- * @rxsz: max rx pkt size, filled in by this func.
+ * @txsz: max tx pdu payload size, filled in by this func.
+ * @rxsz: max rx pdu payload size, filled in by this func.
  * initialize the ddp pagepod manager for a given adapter if needed and
  * setup the tag format for a given iscsi entity
  */
@@ -685,10 +686,12 @@
 		      tformat->sw_bits, tformat->rsvd_bits,
 		      tformat->rsvd_shift, tformat->rsvd_mask);
 
-	*txsz = ddp->max_txsz;
-	*rxsz = ddp->max_rxsz;
-	ddp_log_info("ddp max pkt size: %u, %u.\n",
-		     ddp->max_txsz, ddp->max_rxsz);
+	*txsz = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
+			ddp->max_txsz - ISCSI_PDU_NONPAYLOAD_LEN);
+	*rxsz = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
+			ddp->max_rxsz - ISCSI_PDU_NONPAYLOAD_LEN);
+	ddp_log_info("max payload size: %u/%u, %u/%u.\n",
+		     *txsz, ddp->max_txsz, *rxsz, ddp->max_rxsz);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(cxgb3i_adapter_ddp_init);
diff --git a/drivers/scsi/cxgb3i/cxgb3i_ddp.h b/drivers/scsi/cxgb3i/cxgb3i_ddp.h
index 5c7c4d9..3faae78 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_ddp.h
+++ b/drivers/scsi/cxgb3i/cxgb3i_ddp.h
@@ -13,6 +13,8 @@
 #ifndef __CXGB3I_ULP2_DDP_H__
 #define __CXGB3I_ULP2_DDP_H__
 
+#include <linux/vmalloc.h>
+
 /**
  * struct cxgb3i_tag_format - cxgb3i ulp tag format for an iscsi entity
  *
@@ -85,8 +87,9 @@
 	struct sk_buff **gl_skb;
 };
 
+#define ISCSI_PDU_NONPAYLOAD_LEN	312 /* bhs(48) + ahs(256) + digest(8) */
 #define ULP2_MAX_PKT_SIZE	16224
-#define ULP2_MAX_PDU_PAYLOAD	(ULP2_MAX_PKT_SIZE - ISCSI_PDU_NONPAYLOAD_MAX)
+#define ULP2_MAX_PDU_PAYLOAD	(ULP2_MAX_PKT_SIZE - ISCSI_PDU_NONPAYLOAD_LEN)
 #define PPOD_PAGES_MAX		4
 #define PPOD_PAGES_SHIFT	2	/* 4 pages per pod */
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i_init.c b/drivers/scsi/cxgb3i/cxgb3i_init.c
index 091ecb4..1ce9f24 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_init.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_init.c
@@ -12,8 +12,8 @@
 #include "cxgb3i.h"
 
 #define DRV_MODULE_NAME         "cxgb3i"
-#define DRV_MODULE_VERSION	"1.0.0"
-#define DRV_MODULE_RELDATE	"Jun. 1, 2008"
+#define DRV_MODULE_VERSION	"1.0.1"
+#define DRV_MODULE_RELDATE	"Jan. 2009"
 
 static char version[] =
 	"Chelsio S3xx iSCSI Driver " DRV_MODULE_NAME
diff --git a/drivers/scsi/cxgb3i/cxgb3i_iscsi.c b/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
index d83464b..fa2a44f 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
@@ -364,7 +364,8 @@
 
 	cls_session = iscsi_session_setup(&cxgb3i_iscsi_transport, shost,
 					  cmds_max,
-					  sizeof(struct iscsi_tcp_task),
+					  sizeof(struct iscsi_tcp_task) +
+					  sizeof(struct cxgb3i_task_data),
 					  initial_cmdsn, ISCSI_MAX_TARGET);
 	if (!cls_session)
 		return NULL;
@@ -402,17 +403,15 @@
 {
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
 	struct cxgb3i_conn *cconn = tcp_conn->dd_data;
-	unsigned int max = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
-				 cconn->hba->snic->tx_max_size -
-				 ISCSI_PDU_NONPAYLOAD_MAX);
+	unsigned int max = max(512 * MAX_SKB_FRAGS, SKB_TX_HEADROOM);
 
+	max = min(cconn->hba->snic->tx_max_size, max);
 	if (conn->max_xmit_dlength)
-		conn->max_xmit_dlength = min_t(unsigned int,
-						conn->max_xmit_dlength, max);
+		conn->max_xmit_dlength = min(conn->max_xmit_dlength, max);
 	else
 		conn->max_xmit_dlength = max;
 	align_pdu_size(conn->max_xmit_dlength);
-	cxgb3i_log_info("conn 0x%p, max xmit %u.\n",
+	cxgb3i_api_debug("conn 0x%p, max xmit %u.\n",
 			 conn, conn->max_xmit_dlength);
 	return 0;
 }
@@ -427,9 +426,7 @@
 {
 	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
 	struct cxgb3i_conn *cconn = tcp_conn->dd_data;
-	unsigned int max = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
-				 cconn->hba->snic->rx_max_size -
-				 ISCSI_PDU_NONPAYLOAD_MAX);
+	unsigned int max = cconn->hba->snic->rx_max_size;
 
 	align_pdu_size(max);
 	if (conn->max_recv_dlength) {
@@ -439,8 +436,7 @@
 					 conn->max_recv_dlength, max);
 			return -EINVAL;
 		}
-		conn->max_recv_dlength = min_t(unsigned int,
-						conn->max_recv_dlength, max);
+		conn->max_recv_dlength = min(conn->max_recv_dlength, max);
 		align_pdu_size(conn->max_recv_dlength);
 	} else
 		conn->max_recv_dlength = max;
@@ -844,7 +840,7 @@
 	.proc_name		= "cxgb3i",
 	.queuecommand		= iscsi_queuecommand,
 	.change_queue_depth	= iscsi_change_queue_depth,
-	.can_queue		= 128 * (ISCSI_DEF_XMIT_CMDS_MAX - 1),
+	.can_queue		= CXGB3I_SCSI_QDEPTH_DFLT - 1,
 	.sg_tablesize		= SG_ALL,
 	.max_sectors		= 0xFFFF,
 	.cmd_per_lun		= ISCSI_DEF_CMD_PER_LUN,
diff --git a/drivers/scsi/cxgb3i/cxgb3i_offload.c b/drivers/scsi/cxgb3i/cxgb3i_offload.c
index a865f1f..de3b3b6 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_offload.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_offload.c
@@ -23,19 +23,19 @@
 #include "cxgb3i_ddp.h"
 
 #ifdef __DEBUG_C3CN_CONN__
-#define c3cn_conn_debug         cxgb3i_log_info
+#define c3cn_conn_debug		cxgb3i_log_debug
 #else
 #define c3cn_conn_debug(fmt...)
 #endif
 
 #ifdef __DEBUG_C3CN_TX__
-#define c3cn_tx_debug         cxgb3i_log_debug
+#define c3cn_tx_debug		cxgb3i_log_debug
 #else
 #define c3cn_tx_debug(fmt...)
 #endif
 
 #ifdef __DEBUG_C3CN_RX__
-#define c3cn_rx_debug         cxgb3i_log_debug
+#define c3cn_rx_debug		cxgb3i_log_debug
 #else
 #define c3cn_rx_debug(fmt...)
 #endif
@@ -47,9 +47,9 @@
 module_param(cxgb3_rcv_win, int, 0644);
 MODULE_PARM_DESC(cxgb3_rcv_win, "TCP receive window in bytes (default=256KB)");
 
-static int cxgb3_snd_win = 64 * 1024;
+static int cxgb3_snd_win = 128 * 1024;
 module_param(cxgb3_snd_win, int, 0644);
-MODULE_PARM_DESC(cxgb3_snd_win, "TCP send window in bytes (default=64KB)");
+MODULE_PARM_DESC(cxgb3_snd_win, "TCP send window in bytes (default=128KB)");
 
 static int cxgb3_rx_credit_thres = 10 * 1024;
 module_param(cxgb3_rx_credit_thres, int, 0644);
@@ -301,8 +301,8 @@
 static void skb_entail(struct s3_conn *c3cn, struct sk_buff *skb,
 		       int flags)
 {
-	CXGB3_SKB_CB(skb)->seq = c3cn->write_seq;
-	CXGB3_SKB_CB(skb)->flags = flags;
+	skb_tcp_seq(skb) = c3cn->write_seq;
+	skb_flags(skb) = flags;
 	__skb_queue_tail(&c3cn->write_queue, skb);
 }
 
@@ -457,12 +457,9 @@
  * The number of WRs needed for an skb depends on the number of fragments
  * in the skb and whether it has any payload in its main body.  This maps the
  * length of the gather list represented by an skb into the # of necessary WRs.
- *
- * The max. length of an skb is controlled by the max pdu size which is ~16K.
- * Also, assume the min. fragment length is the sector size (512), then add
- * extra fragment counts for iscsi bhs and payload padding.
+ * The extra two fragments are for iscsi bhs and payload padding.
  */
-#define SKB_WR_LIST_SIZE	(16384/512 + 3)
+#define SKB_WR_LIST_SIZE	(MAX_SKB_FRAGS + 2)
 static unsigned int skb_wrs[SKB_WR_LIST_SIZE] __read_mostly;
 
 static void s3_init_wr_tab(unsigned int wr_len)
@@ -485,7 +482,7 @@
 
 static inline void reset_wr_list(struct s3_conn *c3cn)
 {
-	c3cn->wr_pending_head = NULL;
+	c3cn->wr_pending_head = c3cn->wr_pending_tail = NULL;
 }
 
 /*
@@ -496,7 +493,7 @@
 static inline void enqueue_wr(struct s3_conn *c3cn,
 			      struct sk_buff *skb)
 {
-	skb_wr_data(skb) = NULL;
+	skb_tx_wr_next(skb) = NULL;
 
 	/*
 	 * We want to take an extra reference since both us and the driver
@@ -509,10 +506,22 @@
 	if (!c3cn->wr_pending_head)
 		c3cn->wr_pending_head = skb;
 	else
-		skb_wr_data(skb) = skb;
+		skb_tx_wr_next(c3cn->wr_pending_tail) = skb;
 	c3cn->wr_pending_tail = skb;
 }
 
+static int count_pending_wrs(struct s3_conn *c3cn)
+{
+	int n = 0;
+	const struct sk_buff *skb = c3cn->wr_pending_head;
+
+	while (skb) {
+		n += skb->csum;
+		skb = skb_tx_wr_next(skb);
+	}
+	return n;
+}
+
 static inline struct sk_buff *peek_wr(const struct s3_conn *c3cn)
 {
 	return c3cn->wr_pending_head;
@@ -529,8 +538,8 @@
 
 	if (likely(skb)) {
 		/* Don't bother clearing the tail */
-		c3cn->wr_pending_head = skb_wr_data(skb);
-		skb_wr_data(skb) = NULL;
+		c3cn->wr_pending_head = skb_tx_wr_next(skb);
+		skb_tx_wr_next(skb) = NULL;
 	}
 	return skb;
 }
@@ -543,13 +552,14 @@
 }
 
 static inline void make_tx_data_wr(struct s3_conn *c3cn, struct sk_buff *skb,
-				   int len)
+				   int len, int req_completion)
 {
 	struct tx_data_wr *req;
 
 	skb_reset_transport_header(skb);
 	req = (struct tx_data_wr *)__skb_push(skb, sizeof(*req));
-	req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA));
+	req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA) |
+			(req_completion ? F_WR_COMPL : 0));
 	req->wr_lo = htonl(V_WR_TID(c3cn->tid));
 	req->sndseq = htonl(c3cn->snd_nxt);
 	/* len includes the length of any HW ULP additions */
@@ -592,7 +602,7 @@
 
 	if (unlikely(c3cn->state == C3CN_STATE_CONNECTING ||
 		     c3cn->state == C3CN_STATE_CLOSE_WAIT_1 ||
-		     c3cn->state == C3CN_STATE_ABORTING)) {
+		     c3cn->state >= C3CN_STATE_ABORTING)) {
 		c3cn_tx_debug("c3cn 0x%p, in closing state %u.\n",
 			      c3cn, c3cn->state);
 		return 0;
@@ -615,7 +625,7 @@
 		if (c3cn->wr_avail < wrs_needed) {
 			c3cn_tx_debug("c3cn 0x%p, skb len %u/%u, frag %u, "
 				      "wr %d < %u.\n",
-				      c3cn, skb->len, skb->datalen, frags,
+				      c3cn, skb->len, skb->data_len, frags,
 				      wrs_needed, c3cn->wr_avail);
 			break;
 		}
@@ -627,20 +637,24 @@
 		c3cn->wr_unacked += wrs_needed;
 		enqueue_wr(c3cn, skb);
 
-		if (likely(CXGB3_SKB_CB(skb)->flags & C3CB_FLAG_NEED_HDR)) {
-			len += ulp_extra_len(skb);
-			make_tx_data_wr(c3cn, skb, len);
-			c3cn->snd_nxt += len;
-			if ((req_completion
-			     && c3cn->wr_unacked == wrs_needed)
-			    || (CXGB3_SKB_CB(skb)->flags & C3CB_FLAG_COMPL)
-			    || c3cn->wr_unacked >= c3cn->wr_max / 2) {
-				struct work_request_hdr *wr = cplhdr(skb);
+		c3cn_tx_debug("c3cn 0x%p, enqueue, skb len %u/%u, frag %u, "
+				"wr %d, left %u, unack %u.\n",
+				c3cn, skb->len, skb->data_len, frags,
+				wrs_needed, c3cn->wr_avail, c3cn->wr_unacked);
 
-				wr->wr_hi |= htonl(F_WR_COMPL);
+
+		if (likely(skb_flags(skb) & C3CB_FLAG_NEED_HDR)) {
+			if ((req_completion &&
+				c3cn->wr_unacked == wrs_needed) ||
+			    (skb_flags(skb) & C3CB_FLAG_COMPL) ||
+			    c3cn->wr_unacked >= c3cn->wr_max / 2) {
+				req_completion = 1;
 				c3cn->wr_unacked = 0;
 			}
-			CXGB3_SKB_CB(skb)->flags &= ~C3CB_FLAG_NEED_HDR;
+			len += ulp_extra_len(skb);
+			make_tx_data_wr(c3cn, skb, len, req_completion);
+			c3cn->snd_nxt += len;
+			skb_flags(skb) &= ~C3CB_FLAG_NEED_HDR;
 		}
 
 		total_size += skb->truesize;
@@ -735,8 +749,11 @@
 	if (unlikely(c3cn_flag(c3cn, C3CN_ACTIVE_CLOSE_NEEDED)))
 		/* upper layer has requested closing */
 		send_abort_req(c3cn);
-	else if (c3cn_push_tx_frames(c3cn, 1))
+	else {
+		if (skb_queue_len(&c3cn->write_queue))
+			c3cn_push_tx_frames(c3cn, 1);
 		cxgb3i_conn_tx_open(c3cn);
+	}
 }
 
 static int do_act_establish(struct t3cdev *cdev, struct sk_buff *skb,
@@ -1082,8 +1099,8 @@
 		return;
 	}
 
-	CXGB3_SKB_CB(skb)->seq = ntohl(hdr_cpl->seq);
-	CXGB3_SKB_CB(skb)->flags = 0;
+	skb_tcp_seq(skb) = ntohl(hdr_cpl->seq);
+	skb_flags(skb) = 0;
 
 	skb_reset_transport_header(skb);
 	__skb_pull(skb, sizeof(struct cpl_iscsi_hdr));
@@ -1103,12 +1120,12 @@
 		goto abort_conn;
 
 	skb_ulp_mode(skb) = ULP2_FLAG_DATA_READY;
-	skb_ulp_pdulen(skb) = ntohs(ddp_cpl.len);
-	skb_ulp_ddigest(skb) = ntohl(ddp_cpl.ulp_crc);
+	skb_rx_pdulen(skb) = ntohs(ddp_cpl.len);
+	skb_rx_ddigest(skb) = ntohl(ddp_cpl.ulp_crc);
 	status = ntohl(ddp_cpl.ddp_status);
 
 	c3cn_rx_debug("rx skb 0x%p, len %u, pdulen %u, ddp status 0x%x.\n",
-		      skb, skb->len, skb_ulp_pdulen(skb), status);
+		      skb, skb->len, skb_rx_pdulen(skb), status);
 
 	if (status & (1 << RX_DDP_STATUS_HCRC_SHIFT))
 		skb_ulp_mode(skb) |= ULP2_FLAG_HCRC_ERROR;
@@ -1126,7 +1143,7 @@
 	} else if (status & (1 << RX_DDP_STATUS_DDP_SHIFT))
 		skb_ulp_mode(skb) |= ULP2_FLAG_DATA_DDPED;
 
-	c3cn->rcv_nxt = ntohl(ddp_cpl.seq) + skb_ulp_pdulen(skb);
+	c3cn->rcv_nxt = ntohl(ddp_cpl.seq) + skb_rx_pdulen(skb);
 	__pskb_trim(skb, len);
 	__skb_queue_tail(&c3cn->receive_queue, skb);
 	cxgb3i_conn_pdu_ready(c3cn);
@@ -1151,12 +1168,27 @@
  * Process an acknowledgment of WR completion.  Advance snd_una and send the
  * next batch of work requests from the write queue.
  */
+static void check_wr_invariants(struct s3_conn *c3cn)
+{
+	int pending = count_pending_wrs(c3cn);
+
+	if (unlikely(c3cn->wr_avail + pending != c3cn->wr_max))
+		cxgb3i_log_error("TID %u: credit imbalance: avail %u, "
+				"pending %u, total should be %u\n",
+				c3cn->tid, c3cn->wr_avail, pending,
+				c3cn->wr_max);
+}
+
 static void process_wr_ack(struct s3_conn *c3cn, struct sk_buff *skb)
 {
 	struct cpl_wr_ack *hdr = cplhdr(skb);
 	unsigned int credits = ntohs(hdr->credits);
 	u32 snd_una = ntohl(hdr->snd_una);
 
+	c3cn_tx_debug("%u WR credits, avail %u, unack %u, TID %u, state %u.\n",
+			credits, c3cn->wr_avail, c3cn->wr_unacked,
+			c3cn->tid, c3cn->state);
+
 	c3cn->wr_avail += credits;
 	if (c3cn->wr_unacked > c3cn->wr_max - c3cn->wr_avail)
 		c3cn->wr_unacked = c3cn->wr_max - c3cn->wr_avail;
@@ -1171,6 +1203,17 @@
 			break;
 		}
 		if (unlikely(credits < p->csum)) {
+			struct tx_data_wr *w = cplhdr(p);
+			cxgb3i_log_error("TID %u got %u WR credits need %u, "
+					 "len %u, main body %u, frags %u, "
+					 "seq # %u, ACK una %u, ACK nxt %u, "
+					 "WR_AVAIL %u, WRs pending %u\n",
+					 c3cn->tid, credits, p->csum, p->len,
+					 p->len - p->data_len,
+					 skb_shinfo(p)->nr_frags,
+					 ntohl(w->sndseq), snd_una,
+					 ntohl(hdr->snd_nxt), c3cn->wr_avail,
+					 count_pending_wrs(c3cn) - credits);
 			p->csum -= credits;
 			break;
 		} else {
@@ -1180,15 +1223,24 @@
 		}
 	}
 
-	if (unlikely(before(snd_una, c3cn->snd_una)))
+	check_wr_invariants(c3cn);
+
+	if (unlikely(before(snd_una, c3cn->snd_una))) {
+		cxgb3i_log_error("TID %u, unexpected sequence # %u in WR_ACK "
+				 "snd_una %u\n",
+				 c3cn->tid, snd_una, c3cn->snd_una);
 		goto out_free;
+	}
 
 	if (c3cn->snd_una != snd_una) {
 		c3cn->snd_una = snd_una;
 		dst_confirm(c3cn->dst_cache);
 	}
 
-	if (skb_queue_len(&c3cn->write_queue) && c3cn_push_tx_frames(c3cn, 0))
+	if (skb_queue_len(&c3cn->write_queue)) {
+		if (c3cn_push_tx_frames(c3cn, 0))
+			cxgb3i_conn_tx_open(c3cn);
+	} else
 		cxgb3i_conn_tx_open(c3cn);
 out_free:
 	__kfree_skb(skb);
@@ -1452,7 +1504,7 @@
 			      struct dst_entry *dst)
 {
 	BUG_ON(c3cn->cdev != cdev);
-	c3cn->wr_max = c3cn->wr_avail = T3C_DATA(cdev)->max_wrs;
+	c3cn->wr_max = c3cn->wr_avail = T3C_DATA(cdev)->max_wrs - 1;
 	c3cn->wr_unacked = 0;
 	c3cn->mss_idx = select_mss(c3cn, dst_mtu(dst));
 
@@ -1671,9 +1723,17 @@
 		goto out_err;
 	}
 
-	err = -EPIPE;
 	if (c3cn->err) {
 		c3cn_tx_debug("c3cn 0x%p, err %d.\n", c3cn, c3cn->err);
+		err = -EPIPE;
+		goto out_err;
+	}
+
+	if (c3cn->write_seq - c3cn->snd_una >= cxgb3_snd_win) {
+		c3cn_tx_debug("c3cn 0x%p, snd %u - %u > %u.\n",
+				c3cn, c3cn->write_seq, c3cn->snd_una,
+				cxgb3_snd_win);
+		err = -EAGAIN;
 		goto out_err;
 	}
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i_offload.h b/drivers/scsi/cxgb3i/cxgb3i_offload.h
index d231569..6344b9e 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_offload.h
+++ b/drivers/scsi/cxgb3i/cxgb3i_offload.h
@@ -178,25 +178,33 @@
  * @flag:	see C3CB_FLAG_* below
  * @ulp_mode:	ULP mode/submode of sk_buff
  * @seq:	tcp sequence number
- * @ddigest:	pdu data digest
- * @pdulen:	recovered pdu length
- * @wr_data:	scratch area for tx wr
  */
+struct cxgb3_skb_rx_cb {
+	__u32 ddigest;			/* data digest */
+	__u32 pdulen;			/* recovered pdu length */
+};
+
+struct cxgb3_skb_tx_cb {
+	struct sk_buff *wr_next;	/* next wr */
+};
+
 struct cxgb3_skb_cb {
 	__u8 flags;
 	__u8 ulp_mode;
 	__u32 seq;
-	__u32 ddigest;
-	__u32 pdulen;
-	struct sk_buff *wr_data;
+	union {
+		struct cxgb3_skb_rx_cb rx;
+		struct cxgb3_skb_tx_cb tx;
+	};
 };
 
 #define CXGB3_SKB_CB(skb)	((struct cxgb3_skb_cb *)&((skb)->cb[0]))
-
+#define skb_flags(skb)		(CXGB3_SKB_CB(skb)->flags)
 #define skb_ulp_mode(skb)	(CXGB3_SKB_CB(skb)->ulp_mode)
-#define skb_ulp_ddigest(skb)	(CXGB3_SKB_CB(skb)->ddigest)
-#define skb_ulp_pdulen(skb)	(CXGB3_SKB_CB(skb)->pdulen)
-#define skb_wr_data(skb)	(CXGB3_SKB_CB(skb)->wr_data)
+#define skb_tcp_seq(skb)	(CXGB3_SKB_CB(skb)->seq)
+#define skb_rx_ddigest(skb)	(CXGB3_SKB_CB(skb)->rx.ddigest)
+#define skb_rx_pdulen(skb)	(CXGB3_SKB_CB(skb)->rx.pdulen)
+#define skb_tx_wr_next(skb)	(CXGB3_SKB_CB(skb)->tx.wr_next)
 
 enum c3cb_flags {
 	C3CB_FLAG_NEED_HDR = 1 << 0,	/* packet needs a TX_DATA_WR header */
@@ -217,6 +225,7 @@
 /* for TX: a skb must have a headroom of at least TX_HEADER_LEN bytes */
 #define TX_HEADER_LEN \
 		(sizeof(struct tx_data_wr) + sizeof(struct sge_opaque_hdr))
+#define SKB_TX_HEADROOM		SKB_MAX_HEAD(TX_HEADER_LEN)
 
 /*
  * get and set private ip for iscsi traffic
diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.c b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
index ce7ce8c..17115c2 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
@@ -32,6 +32,10 @@
 #define cxgb3i_tx_debug(fmt...)
 #endif
 
+/* always allocate rooms for AHS */
+#define SKB_TX_PDU_HEADER_LEN	\
+	(sizeof(struct iscsi_hdr) + ISCSI_MAX_AHS_SIZE)
+static unsigned int skb_extra_headroom;
 static struct page *pad_page;
 
 /*
@@ -146,12 +150,13 @@
 
 void cxgb3i_conn_cleanup_task(struct iscsi_task *task)
 {
-	struct iscsi_tcp_task *tcp_task = task->dd_data;
+	struct cxgb3i_task_data *tdata = task->dd_data +
+					sizeof(struct iscsi_tcp_task);
 
 	/* never reached the xmit task callout */
-	if (tcp_task->dd_data)
-		kfree_skb(tcp_task->dd_data);
-	tcp_task->dd_data = NULL;
+	if (tdata->skb)
+		__kfree_skb(tdata->skb);
+	memset(tdata, 0, sizeof(struct cxgb3i_task_data));
 
 	/* MNC - Do we need a check in case this is called but
 	 * cxgb3i_conn_alloc_pdu has never been called on the task */
@@ -159,28 +164,102 @@
 	iscsi_tcp_cleanup_task(task);
 }
 
-/*
- * We do not support ahs yet
- */
+static int sgl_seek_offset(struct scatterlist *sgl, unsigned int sgcnt,
+				unsigned int offset, unsigned int *off,
+				struct scatterlist **sgp)
+{
+	int i;
+	struct scatterlist *sg;
+
+	for_each_sg(sgl, sg, sgcnt, i) {
+		if (offset < sg->length) {
+			*off = offset;
+			*sgp = sg;
+			return 0;
+		}
+		offset -= sg->length;
+	}
+	return -EFAULT;
+}
+
+static int sgl_read_to_frags(struct scatterlist *sg, unsigned int sgoffset,
+				unsigned int dlen, skb_frag_t *frags,
+				int frag_max)
+{
+	unsigned int datalen = dlen;
+	unsigned int sglen = sg->length - sgoffset;
+	struct page *page = sg_page(sg);
+	int i;
+
+	i = 0;
+	do {
+		unsigned int copy;
+
+		if (!sglen) {
+			sg = sg_next(sg);
+			if (!sg) {
+				cxgb3i_log_error("%s, sg NULL, len %u/%u.\n",
+						 __func__, datalen, dlen);
+				return -EINVAL;
+			}
+			sgoffset = 0;
+			sglen = sg->length;
+			page = sg_page(sg);
+
+		}
+		copy = min(datalen, sglen);
+		if (i && page == frags[i - 1].page &&
+		    sgoffset + sg->offset ==
+			frags[i - 1].page_offset + frags[i - 1].size) {
+			frags[i - 1].size += copy;
+		} else {
+			if (i >= frag_max) {
+				cxgb3i_log_error("%s, too many pages %u, "
+						 "dlen %u.\n", __func__,
+						 frag_max, dlen);
+				return -EINVAL;
+			}
+
+			frags[i].page = page;
+			frags[i].page_offset = sg->offset + sgoffset;
+			frags[i].size = copy;
+			i++;
+		}
+		datalen -= copy;
+		sgoffset += copy;
+		sglen -= copy;
+	} while (datalen);
+
+	return i;
+}
+
 int cxgb3i_conn_alloc_pdu(struct iscsi_task *task, u8 opcode)
 {
+	struct iscsi_conn *conn = task->conn;
 	struct iscsi_tcp_task *tcp_task = task->dd_data;
-	struct sk_buff *skb;
+	struct cxgb3i_task_data *tdata = task->dd_data + sizeof(*tcp_task);
+	struct scsi_cmnd *sc = task->sc;
+	int headroom = SKB_TX_PDU_HEADER_LEN;
 
+	tcp_task->dd_data = tdata;
 	task->hdr = NULL;
-	/* always allocate rooms for AHS */
-	skb = alloc_skb(sizeof(struct iscsi_hdr) + ISCSI_MAX_AHS_SIZE +
-			TX_HEADER_LEN,  GFP_ATOMIC);
-	if (!skb)
+
+	/* write command, need to send data pdus */
+	if (skb_extra_headroom && (opcode == ISCSI_OP_SCSI_DATA_OUT ||
+	    (opcode == ISCSI_OP_SCSI_CMD &&
+	    (scsi_bidi_cmnd(sc) || sc->sc_data_direction == DMA_TO_DEVICE))))
+		headroom += min(skb_extra_headroom, conn->max_xmit_dlength);
+
+	tdata->skb = alloc_skb(TX_HEADER_LEN + headroom, GFP_ATOMIC);
+	if (!tdata->skb)
 		return -ENOMEM;
+	skb_reserve(tdata->skb, TX_HEADER_LEN);
 
 	cxgb3i_tx_debug("task 0x%p, opcode 0x%x, skb 0x%p.\n",
-			task, opcode, skb);
+			task, opcode, tdata->skb);
 
-	tcp_task->dd_data = skb;
-	skb_reserve(skb, TX_HEADER_LEN);
-	task->hdr = (struct iscsi_hdr *)skb->data;
-	task->hdr_max = sizeof(struct iscsi_hdr);
+	task->hdr = (struct iscsi_hdr *)tdata->skb->data;
+	task->hdr_max = SKB_TX_PDU_HEADER_LEN;
 
 	/* data_out uses scsi_cmd's itt */
 	if (opcode != ISCSI_OP_SCSI_DATA_OUT)
@@ -192,13 +271,13 @@
 int cxgb3i_conn_init_pdu(struct iscsi_task *task, unsigned int offset,
 			      unsigned int count)
 {
-	struct iscsi_tcp_task *tcp_task = task->dd_data;
-	struct sk_buff *skb = tcp_task->dd_data;
 	struct iscsi_conn *conn = task->conn;
-	struct page *pg;
+	struct iscsi_tcp_task *tcp_task = task->dd_data;
+	struct cxgb3i_task_data *tdata = tcp_task->dd_data;
+	struct sk_buff *skb = tdata->skb;
 	unsigned int datalen = count;
 	int i, padlen = iscsi_padding(count);
-	skb_frag_t *frag;
+	struct page *pg;
 
 	cxgb3i_tx_debug("task 0x%p,0x%p, offset %u, count %u, skb 0x%p.\n",
 			task, task->sc, offset, count, skb);
@@ -209,90 +288,94 @@
 		return 0;
 
 	if (task->sc) {
-		struct scatterlist *sg;
-		struct scsi_data_buffer *sdb;
-		unsigned int sgoffset = offset;
-		struct page *sgpg;
-		unsigned int sglen;
+		struct scsi_data_buffer *sdb = scsi_out(task->sc);
+		struct scatterlist *sg = NULL;
+		int err;
 
-		sdb = scsi_out(task->sc);
-		sg = sdb->table.sgl;
-
-		for_each_sg(sdb->table.sgl, sg, sdb->table.nents, i) {
-			cxgb3i_tx_debug("sg %d, page 0x%p, len %u offset %u\n",
-					i, sg_page(sg), sg->length, sg->offset);
-
-			if (sgoffset < sg->length)
-				break;
-			sgoffset -= sg->length;
+		tdata->offset = offset;
+		tdata->count = count;
+		err = sgl_seek_offset(sdb->table.sgl, sdb->table.nents,
+					tdata->offset, &tdata->sgoffset, &sg);
+		if (err < 0) {
+			cxgb3i_log_warn("tpdu, sgl %u, bad offset %u/%u.\n",
+					sdb->table.nents, tdata->offset,
+					sdb->length);
+			return err;
 		}
-		sgpg = sg_page(sg);
-		sglen = sg->length - sgoffset;
+		err = sgl_read_to_frags(sg, tdata->sgoffset, tdata->count,
+					tdata->frags, MAX_PDU_FRAGS);
+		if (err < 0) {
+			cxgb3i_log_warn("tpdu, sgl %u, bad offset %u + %u.\n",
+					sdb->table.nents, tdata->offset,
+					tdata->count);
+			return err;
+		}
+		tdata->nr_frags = err;
 
-		do {
-			int j = skb_shinfo(skb)->nr_frags;
-			unsigned int copy;
+		if (tdata->nr_frags > MAX_SKB_FRAGS ||
+		    (padlen && tdata->nr_frags == MAX_SKB_FRAGS)) {
+			char *dst = skb->data + task->hdr_len;
+			skb_frag_t *frag = tdata->frags;
 
-			if (!sglen) {
-				sg = sg_next(sg);
-				sgpg = sg_page(sg);
-				sgoffset = 0;
-				sglen = sg->length;
-				++i;
+			/* data fits in the skb's headroom */
+			for (i = 0; i < tdata->nr_frags; i++, frag++) {
+				char *src = kmap_atomic(frag->page,
+							KM_SOFTIRQ0);
+
+				memcpy(dst, src+frag->page_offset, frag->size);
+				dst += frag->size;
+				kunmap_atomic(src, KM_SOFTIRQ0);
 			}
-			copy = min(sglen, datalen);
-			if (j && skb_can_coalesce(skb, j, sgpg,
-						  sg->offset + sgoffset)) {
-				skb_shinfo(skb)->frags[j - 1].size += copy;
-			} else {
-				get_page(sgpg);
-				skb_fill_page_desc(skb, j, sgpg,
-						   sg->offset + sgoffset, copy);
+			if (padlen) {
+				memset(dst, 0, padlen);
+				padlen = 0;
 			}
-			sgoffset += copy;
-			sglen -= copy;
-			datalen -= copy;
-		} while (datalen);
+			skb_put(skb, count + padlen);
+		} else {
+			/* data fit into frag_list */
+			for (i = 0; i < tdata->nr_frags; i++)
+				get_page(tdata->frags[i].page);
+
+			memcpy(skb_shinfo(skb)->frags, tdata->frags,
+				sizeof(skb_frag_t) * tdata->nr_frags);
+			skb_shinfo(skb)->nr_frags = tdata->nr_frags;
+			skb->len += count;
+			skb->data_len += count;
+			skb->truesize += count;
+		}
+
 	} else {
 		pg = virt_to_page(task->data);
 
-		while (datalen) {
-			i = skb_shinfo(skb)->nr_frags;
-			frag = &skb_shinfo(skb)->frags[i];
-
-			get_page(pg);
-			frag->page = pg;
-			frag->page_offset = 0;
-			frag->size = min((unsigned int)PAGE_SIZE, datalen);
-
-			skb_shinfo(skb)->nr_frags++;
-			datalen -= frag->size;
-			pg++;
-		}
+		get_page(pg);
+		skb_fill_page_desc(skb, 0, pg, offset_in_page(task->data),
+					count);
+		skb->len += count;
+		skb->data_len += count;
+		skb->truesize += count;
 	}
 
 	if (padlen) {
 		i = skb_shinfo(skb)->nr_frags;
-		frag = &skb_shinfo(skb)->frags[i];
-		frag->page = pad_page;
-		frag->page_offset = 0;
-		frag->size = padlen;
-		skb_shinfo(skb)->nr_frags++;
+		get_page(pad_page);
+		skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags, pad_page, 0,
+				 padlen);
+
+		skb->data_len += padlen;
+		skb->truesize += padlen;
+		skb->len += padlen;
 	}
 
-	datalen = count + padlen;
-	skb->data_len += datalen;
-	skb->truesize += datalen;
-	skb->len += datalen;
 	return 0;
 }
 
 int cxgb3i_conn_xmit_pdu(struct iscsi_task *task)
 {
-	struct iscsi_tcp_task *tcp_task = task->dd_data;
-	struct sk_buff *skb = tcp_task->dd_data;
 	struct iscsi_tcp_conn *tcp_conn = task->conn->dd_data;
 	struct cxgb3i_conn *cconn = tcp_conn->dd_data;
+	struct iscsi_tcp_task *tcp_task = task->dd_data;
+	struct cxgb3i_task_data *tdata = tcp_task->dd_data;
+	struct sk_buff *skb = tdata->skb;
 	unsigned int datalen;
 	int err;
 
@@ -300,13 +383,14 @@
 		return 0;
 
 	datalen = skb->data_len;
-	tcp_task->dd_data = NULL;
+	tdata->skb = NULL;
 	err = cxgb3i_c3cn_send_pdus(cconn->cep->c3cn, skb);
-	cxgb3i_tx_debug("task 0x%p, skb 0x%p, len %u/%u, rv %d.\n",
-			task, skb, skb->len, skb->data_len, err);
 	if (err > 0) {
 		int pdulen = err;
 
+	cxgb3i_tx_debug("task 0x%p, skb 0x%p, len %u/%u, rv %d.\n",
+			task, skb, skb->len, skb->data_len, err);
+
 		if (task->conn->hdrdgst_en)
 			pdulen += ISCSI_DIGEST_SIZE;
 		if (datalen && task->conn->datadgst_en)
@@ -325,12 +409,14 @@
 		return err;
 	}
 	/* reset skb to send when we are called again */
-	tcp_task->dd_data = skb;
+	tdata->skb = skb;
 	return -EAGAIN;
 }
 
 int cxgb3i_pdu_init(void)
 {
+	if (SKB_TX_HEADROOM > (512 * MAX_SKB_FRAGS))
+		skb_extra_headroom = SKB_TX_HEADROOM;
 	pad_page = alloc_page(GFP_KERNEL);
 	if (!pad_page)
 		return -ENOMEM;
@@ -366,7 +452,9 @@
 	skb = skb_peek(&c3cn->receive_queue);
 	while (!err && skb) {
 		__skb_unlink(skb, &c3cn->receive_queue);
-		read += skb_ulp_pdulen(skb);
+		read += skb_rx_pdulen(skb);
+		cxgb3i_rx_debug("conn 0x%p, cn 0x%p, rx skb 0x%p, pdulen %u.\n",
+				conn, c3cn, skb, skb_rx_pdulen(skb));
 		err = cxgb3i_conn_read_pdu_skb(conn, skb);
 		__kfree_skb(skb);
 		skb = skb_peek(&c3cn->receive_queue);
@@ -377,6 +465,11 @@
 		cxgb3i_c3cn_rx_credits(c3cn, read);
 	}
 	conn->rxdata_octets += read;
+
+	if (err) {
+		cxgb3i_log_info("conn 0x%p rx failed err %d.\n", conn, err);
+		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+	}
 }
 
 void cxgb3i_conn_tx_open(struct s3_conn *c3cn)
diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.h b/drivers/scsi/cxgb3i/cxgb3i_pdu.h
index a3f685c..0770b23 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_pdu.h
+++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.h
@@ -53,7 +53,7 @@
 #define ULP2_FLAG_DCRC_ERROR		0x20
 #define ULP2_FLAG_PAD_ERROR		0x40
 
-void cxgb3i_conn_closing(struct s3_conn *);
+void cxgb3i_conn_closing(struct s3_conn *c3cn);
 void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn);
 void cxgb3i_conn_tx_open(struct s3_conn *c3cn);
 #endif
diff --git a/drivers/scsi/hptiop.c b/drivers/scsi/hptiop.c
index a48e499..34be88d 100644
--- a/drivers/scsi/hptiop.c
+++ b/drivers/scsi/hptiop.c
@@ -1251,6 +1251,7 @@
 	{ PCI_VDEVICE(TTI, 0x3530), (kernel_ulong_t)&hptiop_itl_ops },
 	{ PCI_VDEVICE(TTI, 0x3560), (kernel_ulong_t)&hptiop_itl_ops },
 	{ PCI_VDEVICE(TTI, 0x4322), (kernel_ulong_t)&hptiop_itl_ops },
+	{ PCI_VDEVICE(TTI, 0x4321), (kernel_ulong_t)&hptiop_itl_ops },
 	{ PCI_VDEVICE(TTI, 0x4210), (kernel_ulong_t)&hptiop_itl_ops },
 	{ PCI_VDEVICE(TTI, 0x4211), (kernel_ulong_t)&hptiop_itl_ops },
 	{ PCI_VDEVICE(TTI, 0x4310), (kernel_ulong_t)&hptiop_itl_ops },
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 940dc32..b82ffd9 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1040,12 +1040,11 @@
 				action = ACTION_FAIL;
 			break;
 		case ABORTED_COMMAND:
+			action = ACTION_FAIL;
 			if (sshdr.asc == 0x10) { /* DIF */
 				description = "Target Data Integrity Failure";
-				action = ACTION_FAIL;
 				error = -EILSEQ;
-			} else
-				action = ACTION_RETRY;
+			}
 			break;
 		case NOT_READY:
 			/* If the device is in the process of becoming
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index d57566b..55310db 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -107,6 +107,7 @@
 static void sd_print_sense_hdr(struct scsi_disk *, struct scsi_sense_hdr *);
 static void sd_print_result(struct scsi_disk *, int);
 
+static DEFINE_SPINLOCK(sd_index_lock);
 static DEFINE_IDA(sd_index_ida);
 
 /* This semaphore is used to mediate the 0->1 reference get in the
@@ -1914,7 +1915,9 @@
 		if (!ida_pre_get(&sd_index_ida, GFP_KERNEL))
 			goto out_put;
 
+		spin_lock(&sd_index_lock);
 		error = ida_get_new(&sd_index_ida, &index);
+		spin_unlock(&sd_index_lock);
 	} while (error == -EAGAIN);
 
 	if (error)
@@ -1936,7 +1939,9 @@
 	return 0;
 
  out_free_index:
+	spin_lock(&sd_index_lock);
 	ida_remove(&sd_index_ida, index);
+	spin_unlock(&sd_index_lock);
  out_put:
 	put_disk(gd);
  out_free:
@@ -1986,7 +1991,9 @@
 	struct scsi_disk *sdkp = to_scsi_disk(dev);
 	struct gendisk *disk = sdkp->disk;
 	
+	spin_lock(&sd_index_lock);
 	ida_remove(&sd_index_ida, sdkp->index);
+	spin_unlock(&sd_index_lock);
 
 	disk->private_data = NULL;
 	put_disk(disk);
diff --git a/fs/Makefile b/fs/Makefile
index 38bc735..dc20db3 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -69,10 +69,12 @@
 # Do not add any filesystems before this line
 obj-$(CONFIG_REISERFS_FS)	+= reiserfs/
 obj-$(CONFIG_EXT3_FS)		+= ext3/ # Before ext2 so root fs can be ext3
-obj-$(CONFIG_EXT4_FS)		+= ext4/ # Before ext2 so root fs can be ext4
+obj-$(CONFIG_EXT2_FS)		+= ext2/
+# We place ext4 after ext2 so plain ext2 root fs's are mounted using ext2
+# unless explicitly requested by rootfstype
+obj-$(CONFIG_EXT4_FS)		+= ext4/
 obj-$(CONFIG_JBD)		+= jbd/
 obj-$(CONFIG_JBD2)		+= jbd2/
-obj-$(CONFIG_EXT2_FS)		+= ext2/
 obj-$(CONFIG_CRAMFS)		+= cramfs/
 obj-$(CONFIG_SQUASHFS)		+= squashfs/
 obj-y				+= ramfs/
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 9a50b80..de9459b 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -609,7 +609,9 @@
  */
 int ext4_should_retry_alloc(struct super_block *sb, int *retries)
 {
-	if (!ext4_has_free_blocks(EXT4_SB(sb), 1) || (*retries)++ > 3)
+	if (!ext4_has_free_blocks(EXT4_SB(sb), 1) ||
+	    (*retries)++ > 3 ||
+	    !EXT4_SB(sb)->s_journal)
 		return 0;
 
 	jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 51cdd13..c7fed5b 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2544,7 +2544,7 @@
 
 		ext4_journal_stop(handle);
 
-		if (mpd.retval == -ENOSPC) {
+		if ((mpd.retval == -ENOSPC) && sbi->s_journal) {
 			/* commit the transaction which would
 			 * free blocks released in the transaction
 			 * and try again
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index a5732c5..39d1993 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -3091,7 +3091,6 @@
 
 		/* Journal blocked and flushed, clear needs_recovery flag. */
 		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
-		ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
 		error = ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
 		if (error)
 			goto out;
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index b97cdc5..106c3ba 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -52,6 +52,7 @@
 header-y += cgroupstats.h
 header-y += cramfs_fs.h
 header-y += cycx_cfm.h
+header-y += dcbnl.h
 header-y += dlmconstants.h
 header-y += dlm_device.h
 header-y += dlm_netlink.h
diff --git a/include/linux/dcbnl.h b/include/linux/dcbnl.h
index b0ef274..7d2e100 100644
--- a/include/linux/dcbnl.h
+++ b/include/linux/dcbnl.h
@@ -20,10 +20,12 @@
 #ifndef __LINUX_DCBNL_H__
 #define __LINUX_DCBNL_H__
 
+#include <linux/types.h>
+
 #define DCB_PROTO_VERSION 1
 
 struct dcbmsg {
-	unsigned char      dcb_family;
+	__u8               dcb_family;
 	__u8               cmd;
 	__u16              dcb_pad;
 };
diff --git a/include/linux/decompress/bunzip2.h b/include/linux/decompress/bunzip2.h
new file mode 100644
index 0000000..1152721
--- /dev/null
+++ b/include/linux/decompress/bunzip2.h
@@ -0,0 +1,10 @@
+#ifndef DECOMPRESS_BUNZIP2_H
+#define DECOMPRESS_BUNZIP2_H
+
+int bunzip2(unsigned char *inbuf, int len,
+	    int(*fill)(void*, unsigned int),
+	    int(*flush)(void*, unsigned int),
+	    unsigned char *output,
+	    int *pos,
+	    void(*error)(char *x));
+#endif
diff --git a/include/linux/decompress/generic.h b/include/linux/decompress/generic.h
new file mode 100644
index 0000000..6dfb856
--- /dev/null
+++ b/include/linux/decompress/generic.h
@@ -0,0 +1,33 @@
+#ifndef DECOMPRESS_GENERIC_H
+#define DECOMPRESS_GENERIC_H
+
+/* Minimal chunksize to be read.
+ *Bzip2 prefers at least 4096
+ *Lzma prefers 0x10000 */
+#define COMPR_IOBUF_SIZE	4096
+
+typedef int (*decompress_fn) (unsigned char *inbuf, int len,
+			      int(*fill)(void*, unsigned int),
+			      int(*writebb)(void*, unsigned int),
+			      unsigned char *output,
+			      int *posp,
+			      void(*error)(char *x));
+
+/* inbuf   - input buffer
+ *len     - len of pre-read data in inbuf
+ *fill    - function to fill inbuf if empty
+ *writebb - function to write out outbug
+ *posp    - if non-null, input position (number of bytes read) will be
+ *	  returned here
+ *
+ *If len != 0, the inbuf is initialized (with as much data), and fill
+ *should not be called
+ *If len = 0, the inbuf is allocated, but empty. Its size is IOBUF_SIZE
+ *fill should be called (repeatedly...) to read data, at most IOBUF_SIZE
+ */
+
+/* Utility routine to detect the decompression method */
+decompress_fn decompress_method(const unsigned char *inbuf, int len,
+				const char **name);
+
+#endif
diff --git a/include/linux/decompress/inflate.h b/include/linux/decompress/inflate.h
new file mode 100644
index 0000000..f9b06cc
--- /dev/null
+++ b/include/linux/decompress/inflate.h
@@ -0,0 +1,13 @@
+#ifndef INFLATE_H
+#define INFLATE_H
+
+/* Other housekeeping constants */
+#define INBUFSIZ 4096
+
+int gunzip(unsigned char *inbuf, int len,
+	   int(*fill)(void*, unsigned int),
+	   int(*flush)(void*, unsigned int),
+	   unsigned char *output,
+	   int *pos,
+	   void(*error_fn)(char *x));
+#endif
diff --git a/include/linux/decompress/mm.h b/include/linux/decompress/mm.h
new file mode 100644
index 0000000..12ff8c3
--- /dev/null
+++ b/include/linux/decompress/mm.h
@@ -0,0 +1,87 @@
+/*
+ * linux/compr_mm.h
+ *
+ * Memory management for pre-boot and ramdisk uncompressors
+ *
+ * Authors: Alain Knaff <alain@knaff.lu>
+ *
+ */
+
+#ifndef DECOMPR_MM_H
+#define DECOMPR_MM_H
+
+#ifdef STATIC
+
+/* Code active when included from pre-boot environment: */
+
+/* A trivial malloc implementation, adapted from
+ *  malloc by Hannu Savolainen 1993 and Matthias Urlichs 1994
+ */
+static unsigned long malloc_ptr;
+static int malloc_count;
+
+static void *malloc(int size)
+{
+	void *p;
+
+	if (size < 0)
+		error("Malloc error");
+	if (!malloc_ptr)
+		malloc_ptr = free_mem_ptr;
+
+	malloc_ptr = (malloc_ptr + 3) & ~3;     /* Align */
+
+	p = (void *)malloc_ptr;
+	malloc_ptr += size;
+
+	if (free_mem_end_ptr && malloc_ptr >= free_mem_end_ptr)
+		error("Out of memory");
+
+	malloc_count++;
+	return p;
+}
+
+static void free(void *where)
+{
+	malloc_count--;
+	if (!malloc_count)
+		malloc_ptr = free_mem_ptr;
+}
+
+#define large_malloc(a) malloc(a)
+#define large_free(a) free(a)
+
+#define set_error_fn(x)
+
+#define INIT
+
+#else /* STATIC */
+
+/* Code active when compiled standalone for use when loading ramdisk: */
+
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/vmalloc.h>
+
+/* Use defines rather than static inline in order to avoid spurious
+ * warnings when not needed (indeed large_malloc / large_free are not
+ * needed by inflate */
+
+#define malloc(a) kmalloc(a, GFP_KERNEL)
+#define free(a) kfree(a)
+
+#define large_malloc(a) vmalloc(a)
+#define large_free(a) vfree(a)
+
+static void(*error)(char *m);
+#define set_error_fn(x) error = x;
+
+#define INIT __init
+#define STATIC
+
+#include <linux/init.h>
+
+#endif /* STATIC */
+
+#endif /* DECOMPR_MM_H */
diff --git a/include/linux/decompress/unlzma.h b/include/linux/decompress/unlzma.h
new file mode 100644
index 0000000..7796538
--- /dev/null
+++ b/include/linux/decompress/unlzma.h
@@ -0,0 +1,12 @@
+#ifndef DECOMPRESS_UNLZMA_H
+#define DECOMPRESS_UNLZMA_H
+
+int unlzma(unsigned char *, int,
+	   int(*fill)(void*, unsigned int),
+	   int(*flush)(void*, unsigned int),
+	   unsigned char *output,
+	   int *posp,
+	   void(*error)(char *x)
+	);
+
+#endif
diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h
index cbc2f0c..0adb0f9 100644
--- a/include/linux/io-mapping.h
+++ b/include/linux/io-mapping.h
@@ -91,8 +91,11 @@
 static inline void *
 io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset)
 {
+	resource_size_t phys_addr;
+
 	BUG_ON(offset >= mapping->size);
-	resource_size_t phys_addr = mapping->base + offset;
+	phys_addr = mapping->base + offset;
+
 	return ioremap_wc(phys_addr, PAGE_SIZE);
 }
 
diff --git a/include/linux/netfilter/xt_NFLOG.h b/include/linux/netfilter/xt_NFLOG.h
index cdcd0ed..4b36aeb 100644
--- a/include/linux/netfilter/xt_NFLOG.h
+++ b/include/linux/netfilter/xt_NFLOG.h
@@ -2,7 +2,7 @@
 #define _XT_NFLOG_TARGET
 
 #define XT_NFLOG_DEFAULT_GROUP		0x1
-#define XT_NFLOG_DEFAULT_THRESHOLD	1
+#define XT_NFLOG_DEFAULT_THRESHOLD	0
 
 #define XT_NFLOG_MASK			0x0
 
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 6f3c603..6b58367 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -41,13 +41,13 @@
 #ifndef ARCH_HAS_NOCACHE_UACCESS
 
 static inline unsigned long __copy_from_user_inatomic_nocache(void *to,
-		const void __user *from, unsigned long n, unsigned long total)
+				const void __user *from, unsigned long n)
 {
 	return __copy_from_user_inatomic(to, from, n);
 }
 
 static inline unsigned long __copy_from_user_nocache(void *to,
-		const void __user *from, unsigned long n, unsigned long total)
+				const void __user *from, unsigned long n)
 {
 	return __copy_from_user(to, from, n);
 }
diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h
index e78afe7..c25068e 100644
--- a/include/net/netfilter/nf_conntrack_core.h
+++ b/include/net/netfilter/nf_conntrack_core.h
@@ -59,7 +59,7 @@
 	struct nf_conn *ct = (struct nf_conn *)skb->nfct;
 	int ret = NF_ACCEPT;
 
-	if (ct) {
+	if (ct && ct != &nf_conntrack_untracked) {
 		if (!nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct))
 			ret = __nf_conntrack_confirm(skb);
 		nf_ct_deliver_cached_events(ct);
diff --git a/init/Kconfig b/init/Kconfig
index f068071..95a6613 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -101,6 +101,66 @@
 
 	  which is done within the script "scripts/setlocalversion".)
 
+config HAVE_KERNEL_GZIP
+	bool
+
+config HAVE_KERNEL_BZIP2
+	bool
+
+config HAVE_KERNEL_LZMA
+	bool
+
+choice
+	prompt "Kernel compression mode"
+	default KERNEL_GZIP
+	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA
+	help
+	  The linux kernel is a kind of self-extracting executable.
+	  Several compression algorithms are available, which differ
+	  in efficiency, compression and decompression speed.
+	  Compression speed is only relevant when building a kernel.
+	  Decompression speed is relevant at each boot.
+
+	  If you have any problems with bzip2 or lzma compressed
+	  kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
+	  version of this functionality (bzip2 only), for 2.4, was
+	  supplied by Christian Ludwig)
+
+	  High compression options are mostly useful for users, who
+	  are low on disk space (embedded systems), but for whom ram
+	  size matters less.
+
+	  If in doubt, select 'gzip'
+
+config KERNEL_GZIP
+	bool "Gzip"
+	depends on HAVE_KERNEL_GZIP
+	help
+	  The old and tried gzip compression. Its compression ratio is
+	  the poorest among the 3 choices; however its speed (both
+	  compression and decompression) is the fastest.
+
+config KERNEL_BZIP2
+	bool "Bzip2"
+	depends on HAVE_KERNEL_BZIP2
+	help
+	  Its compression ratio and speed is intermediate.
+	  Decompression speed is slowest among the three.  The kernel
+	  size is about 10% smaller with bzip2, in comparison to gzip.
+	  Bzip2 uses a large amount of memory. For modern kernels you
+	  will need at least 8MB RAM or more for booting.
+
+config KERNEL_LZMA
+	bool "LZMA"
+	depends on HAVE_KERNEL_LZMA
+	help
+	  The most recent compression algorithm.
+	  Its ratio is best, decompression speed is between the other
+	  two. Compression is slowest.	The kernel size is about 33%
+	  smaller with LZMA in comparison to gzip.
+
+endchoice
+
 config SWAP
 	bool "Support for paging of anonymous memory (swap)"
 	depends on MMU && BLOCK
diff --git a/init/do_mounts_rd.c b/init/do_mounts_rd.c
index 0f0f0cf..027a402 100644
--- a/init/do_mounts_rd.c
+++ b/init/do_mounts_rd.c
@@ -11,6 +11,9 @@
 #include "do_mounts.h"
 #include "../fs/squashfs/squashfs_fs.h"
 
+#include <linux/decompress/generic.h>
+
+
 int __initdata rd_prompt = 1;/* 1 = prompt for RAM disk, 0 = don't prompt */
 
 static int __init prompt_ramdisk(char *str)
@@ -29,7 +32,7 @@
 }
 __setup("ramdisk_start=", ramdisk_start_setup);
 
-static int __init crd_load(int in_fd, int out_fd);
+static int __init crd_load(int in_fd, int out_fd, decompress_fn deco);
 
 /*
  * This routine tries to find a RAM disk image to load, and returns the
@@ -38,15 +41,15 @@
  * numbers could not be found.
  *
  * We currently check for the following magic numbers:
- * 	minix
- * 	ext2
+ *	minix
+ *	ext2
  *	romfs
  *	cramfs
  *	squashfs
- * 	gzip
+ *	gzip
  */
-static int __init 
-identify_ramdisk_image(int fd, int start_block)
+static int __init
+identify_ramdisk_image(int fd, int start_block, decompress_fn *decompressor)
 {
 	const int size = 512;
 	struct minix_super_block *minixsb;
@@ -56,6 +59,7 @@
 	struct squashfs_super_block *squashfsb;
 	int nblocks = -1;
 	unsigned char *buf;
+	const char *compress_name;
 
 	buf = kmalloc(size, GFP_KERNEL);
 	if (!buf)
@@ -69,18 +73,19 @@
 	memset(buf, 0xe5, size);
 
 	/*
-	 * Read block 0 to test for gzipped kernel
+	 * Read block 0 to test for compressed kernel
 	 */
 	sys_lseek(fd, start_block * BLOCK_SIZE, 0);
 	sys_read(fd, buf, size);
 
-	/*
-	 * If it matches the gzip magic numbers, return 0
-	 */
-	if (buf[0] == 037 && ((buf[1] == 0213) || (buf[1] == 0236))) {
-		printk(KERN_NOTICE
-		       "RAMDISK: Compressed image found at block %d\n",
-		       start_block);
+	*decompressor = decompress_method(buf, size, &compress_name);
+	if (compress_name) {
+		printk(KERN_NOTICE "RAMDISK: %s image found at block %d\n",
+		       compress_name, start_block);
+		if (!*decompressor)
+			printk(KERN_EMERG
+			       "RAMDISK: %s decompressor not configured!\n",
+			       compress_name);
 		nblocks = 0;
 		goto done;
 	}
@@ -142,7 +147,7 @@
 	printk(KERN_NOTICE
 	       "RAMDISK: Couldn't find valid RAM disk image starting at %d.\n",
 	       start_block);
-	
+
 done:
 	sys_lseek(fd, start_block * BLOCK_SIZE, 0);
 	kfree(buf);
@@ -157,6 +162,7 @@
 	int nblocks, i, disk;
 	char *buf = NULL;
 	unsigned short rotate = 0;
+	decompress_fn decompressor = NULL;
 #if !defined(CONFIG_S390) && !defined(CONFIG_PPC_ISERIES)
 	char rotator[4] = { '|' , '/' , '-' , '\\' };
 #endif
@@ -169,12 +175,12 @@
 	if (in_fd < 0)
 		goto noclose_input;
 
-	nblocks = identify_ramdisk_image(in_fd, rd_image_start);
+	nblocks = identify_ramdisk_image(in_fd, rd_image_start, &decompressor);
 	if (nblocks < 0)
 		goto done;
 
 	if (nblocks == 0) {
-		if (crd_load(in_fd, out_fd) == 0)
+		if (crd_load(in_fd, out_fd, decompressor) == 0)
 			goto successful_load;
 		goto done;
 	}
@@ -200,7 +206,7 @@
 		       nblocks, rd_blocks);
 		goto done;
 	}
-		
+
 	/*
 	 * OK, time to copy in the data
 	 */
@@ -273,138 +279,48 @@
 	return rd_load_image("/dev/root");
 }
 
-/*
- * gzip declarations
- */
-
-#define OF(args)  args
-
-#ifndef memzero
-#define memzero(s, n)     memset ((s), 0, (n))
-#endif
-
-typedef unsigned char  uch;
-typedef unsigned short ush;
-typedef unsigned long  ulg;
-
-#define INBUFSIZ 4096
-#define WSIZE 0x8000    /* window size--must be a power of two, and */
-			/*  at least 32K for zip's deflate method */
-
-static uch *inbuf;
-static uch *window;
-
-static unsigned insize;  /* valid bytes in inbuf */
-static unsigned inptr;   /* index of next byte to be processed in inbuf */
-static unsigned outcnt;  /* bytes in output buffer */
 static int exit_code;
-static int unzip_error;
-static long bytes_out;
+static int decompress_error;
 static int crd_infd, crd_outfd;
 
-#define get_byte()  (inptr < insize ? inbuf[inptr++] : fill_inbuf())
-		
-/* Diagnostic functions (stubbed out) */
-#define Assert(cond,msg)
-#define Trace(x)
-#define Tracev(x)
-#define Tracevv(x)
-#define Tracec(c,x)
-#define Tracecv(c,x)
-
-#define STATIC static
-#define INIT __init
-
-static int  __init fill_inbuf(void);
-static void __init flush_window(void);
-static void __init error(char *m);
-
-#define NO_INFLATE_MALLOC
-
-#include "../lib/inflate.c"
-
-/* ===========================================================================
- * Fill the input buffer. This is called only when the buffer is empty
- * and at least one byte is really needed.
- * Returning -1 does not guarantee that gunzip() will ever return.
- */
-static int __init fill_inbuf(void)
+static int __init compr_fill(void *buf, unsigned int len)
 {
-	if (exit_code) return -1;
-	
-	insize = sys_read(crd_infd, inbuf, INBUFSIZ);
-	if (insize == 0) {
-		error("RAMDISK: ran out of compressed data");
-		return -1;
-	}
-
-	inptr = 1;
-
-	return inbuf[0];
+	int r = sys_read(crd_infd, buf, len);
+	if (r < 0)
+		printk(KERN_ERR "RAMDISK: error while reading compressed data");
+	else if (r == 0)
+		printk(KERN_ERR "RAMDISK: EOF while reading compressed data");
+	return r;
 }
 
-/* ===========================================================================
- * Write the output window window[0..outcnt-1] and update crc and bytes_out.
- * (Used for the decompressed data only.)
- */
-static void __init flush_window(void)
+static int __init compr_flush(void *window, unsigned int outcnt)
 {
-    ulg c = crc;         /* temporary variable */
-    unsigned n, written;
-    uch *in, ch;
-    
-    written = sys_write(crd_outfd, window, outcnt);
-    if (written != outcnt && unzip_error == 0) {
-	printk(KERN_ERR "RAMDISK: incomplete write (%d != %d) %ld\n",
-	       written, outcnt, bytes_out);
-	unzip_error = 1;
-    }
-    in = window;
-    for (n = 0; n < outcnt; n++) {
-	    ch = *in++;
-	    c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8);
-    }
-    crc = c;
-    bytes_out += (ulg)outcnt;
-    outcnt = 0;
+	int written = sys_write(crd_outfd, window, outcnt);
+	if (written != outcnt) {
+		if (decompress_error == 0)
+			printk(KERN_ERR
+			       "RAMDISK: incomplete write (%d != %d)\n",
+			       written, outcnt);
+		decompress_error = 1;
+		return -1;
+	}
+	return outcnt;
 }
 
 static void __init error(char *x)
 {
 	printk(KERN_ERR "%s\n", x);
 	exit_code = 1;
-	unzip_error = 1;
+	decompress_error = 1;
 }
 
-static int __init crd_load(int in_fd, int out_fd)
+static int __init crd_load(int in_fd, int out_fd, decompress_fn deco)
 {
 	int result;
-
-	insize = 0;		/* valid bytes in inbuf */
-	inptr = 0;		/* index of next byte to be processed in inbuf */
-	outcnt = 0;		/* bytes in output buffer */
-	exit_code = 0;
-	bytes_out = 0;
-	crc = (ulg)0xffffffffL; /* shift register contents */
-
 	crd_infd = in_fd;
 	crd_outfd = out_fd;
-	inbuf = kmalloc(INBUFSIZ, GFP_KERNEL);
-	if (!inbuf) {
-		printk(KERN_ERR "RAMDISK: Couldn't allocate gzip buffer\n");
-		return -1;
-	}
-	window = kmalloc(WSIZE, GFP_KERNEL);
-	if (!window) {
-		printk(KERN_ERR "RAMDISK: Couldn't allocate gzip window\n");
-		kfree(inbuf);
-		return -1;
-	}
-	makecrc();
-	result = gunzip();
-	if (unzip_error)
+	result = deco(NULL, 0, compr_fill, compr_flush, NULL, NULL, error);
+	if (decompress_error)
 		result = 1;
-	kfree(inbuf);
-	kfree(window);
 	return result;
 }
diff --git a/init/initramfs.c b/init/initramfs.c
index d9c941c..7dcde7e 100644
--- a/init/initramfs.c
+++ b/init/initramfs.c
@@ -390,11 +390,13 @@
 	return len - count;
 }
 
-static void __init flush_buffer(char *buf, unsigned len)
+static int __init flush_buffer(void *bufv, unsigned len)
 {
+	char *buf = (char *) bufv;
 	int written;
+	int origLen = len;
 	if (message)
-		return;
+		return -1;
 	while ((written = write_buffer(buf, len)) < len && !message) {
 		char c = buf[written];
 		if (c == '0') {
@@ -408,84 +410,28 @@
 		} else
 			error("junk in compressed archive");
 	}
+	return origLen;
 }
 
-/*
- * gzip declarations
- */
+static unsigned my_inptr;   /* index of next byte to be processed in inbuf */
 
-#define OF(args)  args
-
-#ifndef memzero
-#define memzero(s, n)     memset ((s), 0, (n))
-#endif
-
-typedef unsigned char  uch;
-typedef unsigned short ush;
-typedef unsigned long  ulg;
-
-#define WSIZE 0x8000    /* window size--must be a power of two, and */
-			/*  at least 32K for zip's deflate method */
-
-static uch *inbuf;
-static uch *window;
-
-static unsigned insize;  /* valid bytes in inbuf */
-static unsigned inptr;   /* index of next byte to be processed in inbuf */
-static unsigned outcnt;  /* bytes in output buffer */
-static long bytes_out;
-
-#define get_byte()  (inptr < insize ? inbuf[inptr++] : -1)
-		
-/* Diagnostic functions (stubbed out) */
-#define Assert(cond,msg)
-#define Trace(x)
-#define Tracev(x)
-#define Tracevv(x)
-#define Tracec(c,x)
-#define Tracecv(c,x)
-
-#define STATIC static
-#define INIT __init
-
-static void __init flush_window(void);
-static void __init error(char *m);
-
-#define NO_INFLATE_MALLOC
-
-#include "../lib/inflate.c"
-
-/* ===========================================================================
- * Write the output window window[0..outcnt-1] and update crc and bytes_out.
- * (Used for the decompressed data only.)
- */
-static void __init flush_window(void)
-{
-	ulg c = crc;         /* temporary variable */
-	unsigned n;
-	uch *in, ch;
-
-	flush_buffer(window, outcnt);
-	in = window;
-	for (n = 0; n < outcnt; n++) {
-		ch = *in++;
-		c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8);
-	}
-	crc = c;
-	bytes_out += (ulg)outcnt;
-	outcnt = 0;
-}
+#include <linux/decompress/generic.h>
 
 static char * __init unpack_to_rootfs(char *buf, unsigned len, int check_only)
 {
 	int written;
+	decompress_fn decompress;
+	const char *compress_name;
+	static __initdata char msg_buf[64];
+
 	dry_run = check_only;
 	header_buf = kmalloc(110, GFP_KERNEL);
 	symlink_buf = kmalloc(PATH_MAX + N_ALIGN(PATH_MAX) + 1, GFP_KERNEL);
 	name_buf = kmalloc(N_ALIGN(PATH_MAX), GFP_KERNEL);
-	window = kmalloc(WSIZE, GFP_KERNEL);
-	if (!window || !header_buf || !symlink_buf || !name_buf)
+
+	if (!header_buf || !symlink_buf || !name_buf)
 		panic("can't allocate buffers");
+
 	state = Start;
 	this_header = 0;
 	message = NULL;
@@ -505,22 +451,25 @@
 			continue;
 		}
 		this_header = 0;
-		insize = len;
-		inbuf = buf;
-		inptr = 0;
-		outcnt = 0;		/* bytes in output buffer */
-		bytes_out = 0;
-		crc = (ulg)0xffffffffL; /* shift register contents */
-		makecrc();
-		gunzip();
+		decompress = decompress_method(buf, len, &compress_name);
+		if (decompress)
+			decompress(buf, len, NULL, flush_buffer, NULL,
+				   &my_inptr, error);
+		else if (compress_name) {
+			if (!message) {
+				snprintf(msg_buf, sizeof msg_buf,
+					 "compression method %s not configured",
+					 compress_name);
+				message = msg_buf;
+			}
+		}
 		if (state != Reset)
-			error("junk in gzipped archive");
-		this_header = saved_offset + inptr;
-		buf += inptr;
-		len -= inptr;
+			error("junk in compressed archive");
+		this_header = saved_offset + my_inptr;
+		buf += my_inptr;
+		len -= my_inptr;
 	}
 	dir_utime();
-	kfree(window);
 	kfree(name_buf);
 	kfree(symlink_buf);
 	kfree(header_buf);
@@ -579,7 +528,7 @@
 	char *err = unpack_to_rootfs(__initramfs_start,
 			 __initramfs_end - __initramfs_start, 0);
 	if (err)
-		panic(err);
+		panic(err);	/* Failed to decompress INTERNAL initramfs */
 	if (initrd_start) {
 #ifdef CONFIG_BLK_DEV_RAM
 		int fd;
@@ -605,9 +554,12 @@
 		printk(KERN_INFO "Unpacking initramfs...");
 		err = unpack_to_rootfs((char *)initrd_start,
 			initrd_end - initrd_start, 0);
-		if (err)
-			panic(err);
-		printk(" done\n");
+		if (err) {
+			printk(" failed!\n");
+			printk(KERN_EMERG "%s\n", err);
+		} else {
+			printk(" done\n");
+		}
 		free_initrd();
 #endif
 	}
diff --git a/kernel/seccomp.c b/kernel/seccomp.c
index ad64fcb..57d4b13 100644
--- a/kernel/seccomp.c
+++ b/kernel/seccomp.c
@@ -8,6 +8,7 @@
 
 #include <linux/seccomp.h>
 #include <linux/sched.h>
+#include <linux/compat.h>
 
 /* #define SECCOMP_DEBUG 1 */
 #define NR_SECCOMP_MODES 1
@@ -22,7 +23,7 @@
 	0, /* null terminated */
 };
 
-#ifdef TIF_32BIT
+#ifdef CONFIG_COMPAT
 static int mode1_syscalls_32[] = {
 	__NR_seccomp_read_32, __NR_seccomp_write_32, __NR_seccomp_exit_32, __NR_seccomp_sigreturn_32,
 	0, /* null terminated */
@@ -37,8 +38,8 @@
 	switch (mode) {
 	case 1:
 		syscall = mode1_syscalls;
-#ifdef TIF_32BIT
-		if (test_thread_flag(TIF_32BIT))
+#ifdef CONFIG_COMPAT
+		if (is_compat_task())
 			syscall = mode1_syscalls_32;
 #endif
 		do {
diff --git a/lib/Kconfig b/lib/Kconfig
index 03c2c24..daa4818 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -98,6 +98,20 @@
 	tristate
 
 #
+# These all provide a common interface (hence the apparent duplication with
+# ZLIB_INFLATE; DECOMPRESS_GZIP is just a wrapper.)
+#
+config DECOMPRESS_GZIP
+	select ZLIB_INFLATE
+	tristate
+
+config DECOMPRESS_BZIP2
+	tristate
+
+config DECOMPRESS_LZMA
+	tristate
+
+#
 # Generic allocator support is selected if needed
 #
 config GENERIC_ALLOCATOR
diff --git a/lib/Makefile b/lib/Makefile
index 32b0e64..790de7c 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -11,7 +11,8 @@
 	 rbtree.o radix-tree.o dump_stack.o \
 	 idr.o int_sqrt.o extable.o prio_tree.o \
 	 sha1.o irq_regs.o reciprocal_div.o argv_split.o \
-	 proportions.o prio_heap.o ratelimit.o show_mem.o is_single_threaded.o
+	 proportions.o prio_heap.o ratelimit.o show_mem.o \
+	 is_single_threaded.o decompress.o
 
 lib-$(CONFIG_MMU) += ioremap.o
 lib-$(CONFIG_SMP) += cpumask.o
@@ -65,6 +66,10 @@
 obj-$(CONFIG_LZO_COMPRESS) += lzo/
 obj-$(CONFIG_LZO_DECOMPRESS) += lzo/
 
+lib-$(CONFIG_DECOMPRESS_GZIP) += decompress_inflate.o
+lib-$(CONFIG_DECOMPRESS_BZIP2) += decompress_bunzip2.o
+lib-$(CONFIG_DECOMPRESS_LZMA) += decompress_unlzma.o
+
 obj-$(CONFIG_TEXTSEARCH) += textsearch.o
 obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
 obj-$(CONFIG_TEXTSEARCH_BM) += ts_bm.o
diff --git a/lib/decompress.c b/lib/decompress.c
new file mode 100644
index 0000000..d2842f5
--- /dev/null
+++ b/lib/decompress.c
@@ -0,0 +1,54 @@
+/*
+ * decompress.c
+ *
+ * Detect the decompression method based on magic number
+ */
+
+#include <linux/decompress/generic.h>
+
+#include <linux/decompress/bunzip2.h>
+#include <linux/decompress/unlzma.h>
+#include <linux/decompress/inflate.h>
+
+#include <linux/types.h>
+#include <linux/string.h>
+
+#ifndef CONFIG_DECOMPRESS_GZIP
+# define gunzip NULL
+#endif
+#ifndef CONFIG_DECOMPRESS_BZIP2
+# define bunzip2 NULL
+#endif
+#ifndef CONFIG_DECOMPRESS_LZMA
+# define unlzma NULL
+#endif
+
+static const struct compress_format {
+	unsigned char magic[2];
+	const char *name;
+	decompress_fn decompressor;
+} compressed_formats[] = {
+	{ {037, 0213}, "gzip", gunzip },
+	{ {037, 0236}, "gzip", gunzip },
+	{ {0x42, 0x5a}, "bzip2", bunzip2 },
+	{ {0x5d, 0x00}, "lzma", unlzma },
+	{ {0, 0}, NULL, NULL }
+};
+
+decompress_fn decompress_method(const unsigned char *inbuf, int len,
+				const char **name)
+{
+	const struct compress_format *cf;
+
+	if (len < 2)
+		return NULL;	/* Need at least this much... */
+
+	for (cf = compressed_formats; cf->name; cf++) {
+		if (!memcmp(inbuf, cf->magic, 2))
+			break;
+
+	}
+	if (name)
+		*name = cf->name;
+	return cf->decompressor;
+}
diff --git a/lib/decompress_bunzip2.c b/lib/decompress_bunzip2.c
new file mode 100644
index 0000000..5d3ddb5
--- /dev/null
+++ b/lib/decompress_bunzip2.c
@@ -0,0 +1,735 @@
+/* vi: set sw = 4 ts = 4: */
+/*	Small bzip2 deflate implementation, by Rob Landley (rob@landley.net).
+
+	Based on bzip2 decompression code by Julian R Seward (jseward@acm.org),
+	which also acknowledges contributions by Mike Burrows, David Wheeler,
+	Peter Fenwick, Alistair Moffat, Radford Neal, Ian H. Witten,
+	Robert Sedgewick, and Jon L. Bentley.
+
+	This code is licensed under the LGPLv2:
+		LGPL (http://www.gnu.org/copyleft/lgpl.html
+*/
+
+/*
+	Size and speed optimizations by Manuel Novoa III  (mjn3@codepoet.org).
+
+	More efficient reading of Huffman codes, a streamlined read_bunzip()
+	function, and various other tweaks.  In (limited) tests, approximately
+	20% faster than bzcat on x86 and about 10% faster on arm.
+
+	Note that about 2/3 of the time is spent in read_unzip() reversing
+	the Burrows-Wheeler transformation.  Much of that time is delay
+	resulting from cache misses.
+
+	I would ask that anyone benefiting from this work, especially those
+	using it in commercial products, consider making a donation to my local
+	non-profit hospice organization in the name of the woman I loved, who
+	passed away Feb. 12, 2003.
+
+		In memory of Toni W. Hagan
+
+		Hospice of Acadiana, Inc.
+		2600 Johnston St., Suite 200
+		Lafayette, LA 70503-3240
+
+		Phone (337) 232-1234 or 1-800-738-2226
+		Fax   (337) 232-1297
+
+		http://www.hospiceacadiana.com/
+
+	Manuel
+ */
+
+/*
+	Made it fit for running in Linux Kernel by Alain Knaff (alain@knaff.lu)
+*/
+
+
+#ifndef STATIC
+#include <linux/decompress/bunzip2.h>
+#endif /* !STATIC */
+
+#include <linux/decompress/mm.h>
+
+#ifndef INT_MAX
+#define INT_MAX 0x7fffffff
+#endif
+
+/* Constants for Huffman coding */
+#define MAX_GROUPS		6
+#define GROUP_SIZE   		50	/* 64 would have been more efficient */
+#define MAX_HUFCODE_BITS 	20	/* Longest Huffman code allowed */
+#define MAX_SYMBOLS 		258	/* 256 literals + RUNA + RUNB */
+#define SYMBOL_RUNA		0
+#define SYMBOL_RUNB		1
+
+/* Status return values */
+#define RETVAL_OK			0
+#define RETVAL_LAST_BLOCK		(-1)
+#define RETVAL_NOT_BZIP_DATA		(-2)
+#define RETVAL_UNEXPECTED_INPUT_EOF	(-3)
+#define RETVAL_UNEXPECTED_OUTPUT_EOF	(-4)
+#define RETVAL_DATA_ERROR		(-5)
+#define RETVAL_OUT_OF_MEMORY		(-6)
+#define RETVAL_OBSOLETE_INPUT		(-7)
+
+/* Other housekeeping constants */
+#define BZIP2_IOBUF_SIZE		4096
+
+/* This is what we know about each Huffman coding group */
+struct group_data {
+	/* We have an extra slot at the end of limit[] for a sentinal value. */
+	int limit[MAX_HUFCODE_BITS+1];
+	int base[MAX_HUFCODE_BITS];
+	int permute[MAX_SYMBOLS];
+	int minLen, maxLen;
+};
+
+/* Structure holding all the housekeeping data, including IO buffers and
+   memory that persists between calls to bunzip */
+struct bunzip_data {
+	/* State for interrupting output loop */
+	int writeCopies, writePos, writeRunCountdown, writeCount, writeCurrent;
+	/* I/O tracking data (file handles, buffers, positions, etc.) */
+	int (*fill)(void*, unsigned int);
+	int inbufCount, inbufPos /*, outbufPos*/;
+	unsigned char *inbuf /*,*outbuf*/;
+	unsigned int inbufBitCount, inbufBits;
+	/* The CRC values stored in the block header and calculated from the
+	data */
+	unsigned int crc32Table[256], headerCRC, totalCRC, writeCRC;
+	/* Intermediate buffer and its size (in bytes) */
+	unsigned int *dbuf, dbufSize;
+	/* These things are a bit too big to go on the stack */
+	unsigned char selectors[32768];		/* nSelectors = 15 bits */
+	struct group_data groups[MAX_GROUPS];	/* Huffman coding tables */
+	int io_error;			/* non-zero if we have IO error */
+};
+
+
+/* Return the next nnn bits of input.  All reads from the compressed input
+   are done through this function.  All reads are big endian */
+static unsigned int INIT get_bits(struct bunzip_data *bd, char bits_wanted)
+{
+	unsigned int bits = 0;
+
+	/* If we need to get more data from the byte buffer, do so.
+	   (Loop getting one byte at a time to enforce endianness and avoid
+	   unaligned access.) */
+	while (bd->inbufBitCount < bits_wanted) {
+		/* If we need to read more data from file into byte buffer, do
+		   so */
+		if (bd->inbufPos == bd->inbufCount) {
+			if (bd->io_error)
+				return 0;
+			bd->inbufCount = bd->fill(bd->inbuf, BZIP2_IOBUF_SIZE);
+			if (bd->inbufCount <= 0) {
+				bd->io_error = RETVAL_UNEXPECTED_INPUT_EOF;
+				return 0;
+			}
+			bd->inbufPos = 0;
+		}
+		/* Avoid 32-bit overflow (dump bit buffer to top of output) */
+		if (bd->inbufBitCount >= 24) {
+			bits = bd->inbufBits&((1 << bd->inbufBitCount)-1);
+			bits_wanted -= bd->inbufBitCount;
+			bits <<= bits_wanted;
+			bd->inbufBitCount = 0;
+		}
+		/* Grab next 8 bits of input from buffer. */
+		bd->inbufBits = (bd->inbufBits << 8)|bd->inbuf[bd->inbufPos++];
+		bd->inbufBitCount += 8;
+	}
+	/* Calculate result */
+	bd->inbufBitCount -= bits_wanted;
+	bits |= (bd->inbufBits >> bd->inbufBitCount)&((1 << bits_wanted)-1);
+
+	return bits;
+}
+
+/* Unpacks the next block and sets up for the inverse burrows-wheeler step. */
+
+static int INIT get_next_block(struct bunzip_data *bd)
+{
+	struct group_data *hufGroup = NULL;
+	int *base = NULL;
+	int *limit = NULL;
+	int dbufCount, nextSym, dbufSize, groupCount, selector,
+		i, j, k, t, runPos, symCount, symTotal, nSelectors,
+		byteCount[256];
+	unsigned char uc, symToByte[256], mtfSymbol[256], *selectors;
+	unsigned int *dbuf, origPtr;
+
+	dbuf = bd->dbuf;
+	dbufSize = bd->dbufSize;
+	selectors = bd->selectors;
+
+	/* Read in header signature and CRC, then validate signature.
+	   (last block signature means CRC is for whole file, return now) */
+	i = get_bits(bd, 24);
+	j = get_bits(bd, 24);
+	bd->headerCRC = get_bits(bd, 32);
+	if ((i == 0x177245) && (j == 0x385090))
+		return RETVAL_LAST_BLOCK;
+	if ((i != 0x314159) || (j != 0x265359))
+		return RETVAL_NOT_BZIP_DATA;
+	/* We can add support for blockRandomised if anybody complains.
+	   There was some code for this in busybox 1.0.0-pre3, but nobody ever
+	   noticed that it didn't actually work. */
+	if (get_bits(bd, 1))
+		return RETVAL_OBSOLETE_INPUT;
+	origPtr = get_bits(bd, 24);
+	if (origPtr > dbufSize)
+		return RETVAL_DATA_ERROR;
+	/* mapping table: if some byte values are never used (encoding things
+	   like ascii text), the compression code removes the gaps to have fewer
+	   symbols to deal with, and writes a sparse bitfield indicating which
+	   values were present.  We make a translation table to convert the
+	   symbols back to the corresponding bytes. */
+	t = get_bits(bd, 16);
+	symTotal = 0;
+	for (i = 0; i < 16; i++) {
+		if (t&(1 << (15-i))) {
+			k = get_bits(bd, 16);
+			for (j = 0; j < 16; j++)
+				if (k&(1 << (15-j)))
+					symToByte[symTotal++] = (16*i)+j;
+		}
+	}
+	/* How many different Huffman coding groups does this block use? */
+	groupCount = get_bits(bd, 3);
+	if (groupCount < 2 || groupCount > MAX_GROUPS)
+		return RETVAL_DATA_ERROR;
+	/* nSelectors: Every GROUP_SIZE many symbols we select a new
+	   Huffman coding group.  Read in the group selector list,
+	   which is stored as MTF encoded bit runs.  (MTF = Move To
+	   Front, as each value is used it's moved to the start of the
+	   list.) */
+	nSelectors = get_bits(bd, 15);
+	if (!nSelectors)
+		return RETVAL_DATA_ERROR;
+	for (i = 0; i < groupCount; i++)
+		mtfSymbol[i] = i;
+	for (i = 0; i < nSelectors; i++) {
+		/* Get next value */
+		for (j = 0; get_bits(bd, 1); j++)
+			if (j >= groupCount)
+				return RETVAL_DATA_ERROR;
+		/* Decode MTF to get the next selector */
+		uc = mtfSymbol[j];
+		for (; j; j--)
+			mtfSymbol[j] = mtfSymbol[j-1];
+		mtfSymbol[0] = selectors[i] = uc;
+	}
+	/* Read the Huffman coding tables for each group, which code
+	   for symTotal literal symbols, plus two run symbols (RUNA,
+	   RUNB) */
+	symCount = symTotal+2;
+	for (j = 0; j < groupCount; j++) {
+		unsigned char length[MAX_SYMBOLS], temp[MAX_HUFCODE_BITS+1];
+		int	minLen,	maxLen, pp;
+		/* Read Huffman code lengths for each symbol.  They're
+		   stored in a way similar to mtf; record a starting
+		   value for the first symbol, and an offset from the
+		   previous value for everys symbol after that.
+		   (Subtracting 1 before the loop and then adding it
+		   back at the end is an optimization that makes the
+		   test inside the loop simpler: symbol length 0
+		   becomes negative, so an unsigned inequality catches
+		   it.) */
+		t = get_bits(bd, 5)-1;
+		for (i = 0; i < symCount; i++) {
+			for (;;) {
+				if (((unsigned)t) > (MAX_HUFCODE_BITS-1))
+					return RETVAL_DATA_ERROR;
+
+				/* If first bit is 0, stop.  Else
+				   second bit indicates whether to
+				   increment or decrement the value.
+				   Optimization: grab 2 bits and unget
+				   the second if the first was 0. */
+
+				k = get_bits(bd, 2);
+				if (k < 2) {
+					bd->inbufBitCount++;
+					break;
+				}
+				/* Add one if second bit 1, else
+				 * subtract 1.  Avoids if/else */
+				t += (((k+1)&2)-1);
+			}
+			/* Correct for the initial -1, to get the
+			 * final symbol length */
+			length[i] = t+1;
+		}
+		/* Find largest and smallest lengths in this group */
+		minLen = maxLen = length[0];
+
+		for (i = 1; i < symCount; i++) {
+			if (length[i] > maxLen)
+				maxLen = length[i];
+			else if (length[i] < minLen)
+				minLen = length[i];
+		}
+
+		/* Calculate permute[], base[], and limit[] tables from
+		 * length[].
+		 *
+		 * permute[] is the lookup table for converting
+		 * Huffman coded symbols into decoded symbols.  base[]
+		 * is the amount to subtract from the value of a
+		 * Huffman symbol of a given length when using
+		 * permute[].
+		 *
+		 * limit[] indicates the largest numerical value a
+		 * symbol with a given number of bits can have.  This
+		 * is how the Huffman codes can vary in length: each
+		 * code with a value > limit[length] needs another
+		 * bit.
+		 */
+		hufGroup = bd->groups+j;
+		hufGroup->minLen = minLen;
+		hufGroup->maxLen = maxLen;
+		/* Note that minLen can't be smaller than 1, so we
+		   adjust the base and limit array pointers so we're
+		   not always wasting the first entry.  We do this
+		   again when using them (during symbol decoding).*/
+		base = hufGroup->base-1;
+		limit = hufGroup->limit-1;
+		/* Calculate permute[].  Concurently, initialize
+		 * temp[] and limit[]. */
+		pp = 0;
+		for (i = minLen; i <= maxLen; i++) {
+			temp[i] = limit[i] = 0;
+			for (t = 0; t < symCount; t++)
+				if (length[t] == i)
+					hufGroup->permute[pp++] = t;
+		}
+		/* Count symbols coded for at each bit length */
+		for (i = 0; i < symCount; i++)
+			temp[length[i]]++;
+		/* Calculate limit[] (the largest symbol-coding value
+		 *at each bit length, which is (previous limit <<
+		 *1)+symbols at this level), and base[] (number of
+		 *symbols to ignore at each bit length, which is limit
+		 *minus the cumulative count of symbols coded for
+		 *already). */
+		pp = t = 0;
+		for (i = minLen; i < maxLen; i++) {
+			pp += temp[i];
+			/* We read the largest possible symbol size
+			   and then unget bits after determining how
+			   many we need, and those extra bits could be
+			   set to anything.  (They're noise from
+			   future symbols.)  At each level we're
+			   really only interested in the first few
+			   bits, so here we set all the trailing
+			   to-be-ignored bits to 1 so they don't
+			   affect the value > limit[length]
+			   comparison. */
+			limit[i] = (pp << (maxLen - i)) - 1;
+			pp <<= 1;
+			base[i+1] = pp-(t += temp[i]);
+		}
+		limit[maxLen+1] = INT_MAX; /* Sentinal value for
+					    * reading next sym. */
+		limit[maxLen] = pp+temp[maxLen]-1;
+		base[minLen] = 0;
+	}
+	/* We've finished reading and digesting the block header.  Now
+	   read this block's Huffman coded symbols from the file and
+	   undo the Huffman coding and run length encoding, saving the
+	   result into dbuf[dbufCount++] = uc */
+
+	/* Initialize symbol occurrence counters and symbol Move To
+	 * Front table */
+	for (i = 0; i < 256; i++) {
+		byteCount[i] = 0;
+		mtfSymbol[i] = (unsigned char)i;
+	}
+	/* Loop through compressed symbols. */
+	runPos = dbufCount = symCount = selector = 0;
+	for (;;) {
+		/* Determine which Huffman coding group to use. */
+		if (!(symCount--)) {
+			symCount = GROUP_SIZE-1;
+			if (selector >= nSelectors)
+				return RETVAL_DATA_ERROR;
+			hufGroup = bd->groups+selectors[selector++];
+			base = hufGroup->base-1;
+			limit = hufGroup->limit-1;
+		}
+		/* Read next Huffman-coded symbol. */
+		/* Note: It is far cheaper to read maxLen bits and
+		   back up than it is to read minLen bits and then an
+		   additional bit at a time, testing as we go.
+		   Because there is a trailing last block (with file
+		   CRC), there is no danger of the overread causing an
+		   unexpected EOF for a valid compressed file.  As a
+		   further optimization, we do the read inline
+		   (falling back to a call to get_bits if the buffer
+		   runs dry).  The following (up to got_huff_bits:) is
+		   equivalent to j = get_bits(bd, hufGroup->maxLen);
+		 */
+		while (bd->inbufBitCount < hufGroup->maxLen) {
+			if (bd->inbufPos == bd->inbufCount) {
+				j = get_bits(bd, hufGroup->maxLen);
+				goto got_huff_bits;
+			}
+			bd->inbufBits =
+				(bd->inbufBits << 8)|bd->inbuf[bd->inbufPos++];
+			bd->inbufBitCount += 8;
+		};
+		bd->inbufBitCount -= hufGroup->maxLen;
+		j = (bd->inbufBits >> bd->inbufBitCount)&
+			((1 << hufGroup->maxLen)-1);
+got_huff_bits:
+		/* Figure how how many bits are in next symbol and
+		 * unget extras */
+		i = hufGroup->minLen;
+		while (j > limit[i])
+			++i;
+		bd->inbufBitCount += (hufGroup->maxLen - i);
+		/* Huffman decode value to get nextSym (with bounds checking) */
+		if ((i > hufGroup->maxLen)
+			|| (((unsigned)(j = (j>>(hufGroup->maxLen-i))-base[i]))
+				>= MAX_SYMBOLS))
+			return RETVAL_DATA_ERROR;
+		nextSym = hufGroup->permute[j];
+		/* We have now decoded the symbol, which indicates
+		   either a new literal byte, or a repeated run of the
+		   most recent literal byte.  First, check if nextSym
+		   indicates a repeated run, and if so loop collecting
+		   how many times to repeat the last literal. */
+		if (((unsigned)nextSym) <= SYMBOL_RUNB) { /* RUNA or RUNB */
+			/* If this is the start of a new run, zero out
+			 * counter */
+			if (!runPos) {
+				runPos = 1;
+				t = 0;
+			}
+			/* Neat trick that saves 1 symbol: instead of
+			   or-ing 0 or 1 at each bit position, add 1
+			   or 2 instead.  For example, 1011 is 1 << 0
+			   + 1 << 1 + 2 << 2.  1010 is 2 << 0 + 2 << 1
+			   + 1 << 2.  You can make any bit pattern
+			   that way using 1 less symbol than the basic
+			   or 0/1 method (except all bits 0, which
+			   would use no symbols, but a run of length 0
+			   doesn't mean anything in this context).
+			   Thus space is saved. */
+			t += (runPos << nextSym);
+			/* +runPos if RUNA; +2*runPos if RUNB */
+
+			runPos <<= 1;
+			continue;
+		}
+		/* When we hit the first non-run symbol after a run,
+		   we now know how many times to repeat the last
+		   literal, so append that many copies to our buffer
+		   of decoded symbols (dbuf) now.  (The last literal
+		   used is the one at the head of the mtfSymbol
+		   array.) */
+		if (runPos) {
+			runPos = 0;
+			if (dbufCount+t >= dbufSize)
+				return RETVAL_DATA_ERROR;
+
+			uc = symToByte[mtfSymbol[0]];
+			byteCount[uc] += t;
+			while (t--)
+				dbuf[dbufCount++] = uc;
+		}
+		/* Is this the terminating symbol? */
+		if (nextSym > symTotal)
+			break;
+		/* At this point, nextSym indicates a new literal
+		   character.  Subtract one to get the position in the
+		   MTF array at which this literal is currently to be
+		   found.  (Note that the result can't be -1 or 0,
+		   because 0 and 1 are RUNA and RUNB.  But another
+		   instance of the first symbol in the mtf array,
+		   position 0, would have been handled as part of a
+		   run above.  Therefore 1 unused mtf position minus 2
+		   non-literal nextSym values equals -1.) */
+		if (dbufCount >= dbufSize)
+			return RETVAL_DATA_ERROR;
+		i = nextSym - 1;
+		uc = mtfSymbol[i];
+		/* Adjust the MTF array.  Since we typically expect to
+		 *move only a small number of symbols, and are bound
+		 *by 256 in any case, using memmove here would
+		 *typically be bigger and slower due to function call
+		 *overhead and other assorted setup costs. */
+		do {
+			mtfSymbol[i] = mtfSymbol[i-1];
+		} while (--i);
+		mtfSymbol[0] = uc;
+		uc = symToByte[uc];
+		/* We have our literal byte.  Save it into dbuf. */
+		byteCount[uc]++;
+		dbuf[dbufCount++] = (unsigned int)uc;
+	}
+	/* At this point, we've read all the Huffman-coded symbols
+	   (and repeated runs) for this block from the input stream,
+	   and decoded them into the intermediate buffer.  There are
+	   dbufCount many decoded bytes in dbuf[].  Now undo the
+	   Burrows-Wheeler transform on dbuf.  See
+	   http://dogma.net/markn/articles/bwt/bwt.htm
+	 */
+	/* Turn byteCount into cumulative occurrence counts of 0 to n-1. */
+	j = 0;
+	for (i = 0; i < 256; i++) {
+		k = j+byteCount[i];
+		byteCount[i] = j;
+		j = k;
+	}
+	/* Figure out what order dbuf would be in if we sorted it. */
+	for (i = 0; i < dbufCount; i++) {
+		uc = (unsigned char)(dbuf[i] & 0xff);
+		dbuf[byteCount[uc]] |= (i << 8);
+		byteCount[uc]++;
+	}
+	/* Decode first byte by hand to initialize "previous" byte.
+	   Note that it doesn't get output, and if the first three
+	   characters are identical it doesn't qualify as a run (hence
+	   writeRunCountdown = 5). */
+	if (dbufCount) {
+		if (origPtr >= dbufCount)
+			return RETVAL_DATA_ERROR;
+		bd->writePos = dbuf[origPtr];
+		bd->writeCurrent = (unsigned char)(bd->writePos&0xff);
+		bd->writePos >>= 8;
+		bd->writeRunCountdown = 5;
+	}
+	bd->writeCount = dbufCount;
+
+	return RETVAL_OK;
+}
+
+/* Undo burrows-wheeler transform on intermediate buffer to produce output.
+   If start_bunzip was initialized with out_fd =-1, then up to len bytes of
+   data are written to outbuf.  Return value is number of bytes written or
+   error (all errors are negative numbers).  If out_fd!=-1, outbuf and len
+   are ignored, data is written to out_fd and return is RETVAL_OK or error.
+*/
+
+static int INIT read_bunzip(struct bunzip_data *bd, char *outbuf, int len)
+{
+	const unsigned int *dbuf;
+	int pos, xcurrent, previous, gotcount;
+
+	/* If last read was short due to end of file, return last block now */
+	if (bd->writeCount < 0)
+		return bd->writeCount;
+
+	gotcount = 0;
+	dbuf = bd->dbuf;
+	pos = bd->writePos;
+	xcurrent = bd->writeCurrent;
+
+	/* We will always have pending decoded data to write into the output
+	   buffer unless this is the very first call (in which case we haven't
+	   Huffman-decoded a block into the intermediate buffer yet). */
+
+	if (bd->writeCopies) {
+		/* Inside the loop, writeCopies means extra copies (beyond 1) */
+		--bd->writeCopies;
+		/* Loop outputting bytes */
+		for (;;) {
+			/* If the output buffer is full, snapshot
+			 * state and return */
+			if (gotcount >= len) {
+				bd->writePos = pos;
+				bd->writeCurrent = xcurrent;
+				bd->writeCopies++;
+				return len;
+			}
+			/* Write next byte into output buffer, updating CRC */
+			outbuf[gotcount++] = xcurrent;
+			bd->writeCRC = (((bd->writeCRC) << 8)
+				^bd->crc32Table[((bd->writeCRC) >> 24)
+				^xcurrent]);
+			/* Loop now if we're outputting multiple
+			 * copies of this byte */
+			if (bd->writeCopies) {
+				--bd->writeCopies;
+				continue;
+			}
+decode_next_byte:
+			if (!bd->writeCount--)
+				break;
+			/* Follow sequence vector to undo
+			 * Burrows-Wheeler transform */
+			previous = xcurrent;
+			pos = dbuf[pos];
+			xcurrent = pos&0xff;
+			pos >>= 8;
+			/* After 3 consecutive copies of the same
+			   byte, the 4th is a repeat count.  We count
+			   down from 4 instead *of counting up because
+			   testing for non-zero is faster */
+			if (--bd->writeRunCountdown) {
+				if (xcurrent != previous)
+					bd->writeRunCountdown = 4;
+			} else {
+				/* We have a repeated run, this byte
+				 * indicates the count */
+				bd->writeCopies = xcurrent;
+				xcurrent = previous;
+				bd->writeRunCountdown = 5;
+				/* Sometimes there are just 3 bytes
+				 * (run length 0) */
+				if (!bd->writeCopies)
+					goto decode_next_byte;
+				/* Subtract the 1 copy we'd output
+				 * anyway to get extras */
+				--bd->writeCopies;
+			}
+		}
+		/* Decompression of this block completed successfully */
+		bd->writeCRC = ~bd->writeCRC;
+		bd->totalCRC = ((bd->totalCRC << 1) |
+				(bd->totalCRC >> 31)) ^ bd->writeCRC;
+		/* If this block had a CRC error, force file level CRC error. */
+		if (bd->writeCRC != bd->headerCRC) {
+			bd->totalCRC = bd->headerCRC+1;
+			return RETVAL_LAST_BLOCK;
+		}
+	}
+
+	/* Refill the intermediate buffer by Huffman-decoding next
+	 * block of input */
+	/* (previous is just a convenient unused temp variable here) */
+	previous = get_next_block(bd);
+	if (previous) {
+		bd->writeCount = previous;
+		return (previous != RETVAL_LAST_BLOCK) ? previous : gotcount;
+	}
+	bd->writeCRC = 0xffffffffUL;
+	pos = bd->writePos;
+	xcurrent = bd->writeCurrent;
+	goto decode_next_byte;
+}
+
+static int INIT nofill(void *buf, unsigned int len)
+{
+	return -1;
+}
+
+/* Allocate the structure, read file header.  If in_fd ==-1, inbuf must contain
+   a complete bunzip file (len bytes long).  If in_fd!=-1, inbuf and len are
+   ignored, and data is read from file handle into temporary buffer. */
+static int INIT start_bunzip(struct bunzip_data **bdp, void *inbuf, int len,
+			     int (*fill)(void*, unsigned int))
+{
+	struct bunzip_data *bd;
+	unsigned int i, j, c;
+	const unsigned int BZh0 =
+		(((unsigned int)'B') << 24)+(((unsigned int)'Z') << 16)
+		+(((unsigned int)'h') << 8)+(unsigned int)'0';
+
+	/* Figure out how much data to allocate */
+	i = sizeof(struct bunzip_data);
+
+	/* Allocate bunzip_data.  Most fields initialize to zero. */
+	bd = *bdp = malloc(i);
+	memset(bd, 0, sizeof(struct bunzip_data));
+	/* Setup input buffer */
+	bd->inbuf = inbuf;
+	bd->inbufCount = len;
+	if (fill != NULL)
+		bd->fill = fill;
+	else
+		bd->fill = nofill;
+
+	/* Init the CRC32 table (big endian) */
+	for (i = 0; i < 256; i++) {
+		c = i << 24;
+		for (j = 8; j; j--)
+			c = c&0x80000000 ? (c << 1)^0x04c11db7 : (c << 1);
+		bd->crc32Table[i] = c;
+	}
+
+	/* Ensure that file starts with "BZh['1'-'9']." */
+	i = get_bits(bd, 32);
+	if (((unsigned int)(i-BZh0-1)) >= 9)
+		return RETVAL_NOT_BZIP_DATA;
+
+	/* Fourth byte (ascii '1'-'9'), indicates block size in units of 100k of
+	   uncompressed data.  Allocate intermediate buffer for block. */
+	bd->dbufSize = 100000*(i-BZh0);
+
+	bd->dbuf = large_malloc(bd->dbufSize * sizeof(int));
+	return RETVAL_OK;
+}
+
+/* Example usage: decompress src_fd to dst_fd.  (Stops at end of bzip2 data,
+   not end of file.) */
+STATIC int INIT bunzip2(unsigned char *buf, int len,
+			int(*fill)(void*, unsigned int),
+			int(*flush)(void*, unsigned int),
+			unsigned char *outbuf,
+			int *pos,
+			void(*error_fn)(char *x))
+{
+	struct bunzip_data *bd;
+	int i = -1;
+	unsigned char *inbuf;
+
+	set_error_fn(error_fn);
+	if (flush)
+		outbuf = malloc(BZIP2_IOBUF_SIZE);
+	else
+		len -= 4; /* Uncompressed size hack active in pre-boot
+			     environment */
+	if (!outbuf) {
+		error("Could not allocate output bufer");
+		return -1;
+	}
+	if (buf)
+		inbuf = buf;
+	else
+		inbuf = malloc(BZIP2_IOBUF_SIZE);
+	if (!inbuf) {
+		error("Could not allocate input bufer");
+		goto exit_0;
+	}
+	i = start_bunzip(&bd, inbuf, len, fill);
+	if (!i) {
+		for (;;) {
+			i = read_bunzip(bd, outbuf, BZIP2_IOBUF_SIZE);
+			if (i <= 0)
+				break;
+			if (!flush)
+				outbuf += i;
+			else
+				if (i != flush(outbuf, i)) {
+					i = RETVAL_UNEXPECTED_OUTPUT_EOF;
+					break;
+				}
+		}
+	}
+	/* Check CRC and release memory */
+	if (i == RETVAL_LAST_BLOCK) {
+		if (bd->headerCRC != bd->totalCRC)
+			error("Data integrity error when decompressing.");
+		else
+			i = RETVAL_OK;
+	} else if (i == RETVAL_UNEXPECTED_OUTPUT_EOF) {
+		error("Compressed file ends unexpectedly");
+	}
+	if (bd->dbuf)
+		large_free(bd->dbuf);
+	if (pos)
+		*pos = bd->inbufPos;
+	free(bd);
+	if (!buf)
+		free(inbuf);
+exit_0:
+	if (flush)
+		free(outbuf);
+	return i;
+}
+
+#define decompress bunzip2
diff --git a/lib/decompress_inflate.c b/lib/decompress_inflate.c
new file mode 100644
index 0000000..839a329
--- /dev/null
+++ b/lib/decompress_inflate.c
@@ -0,0 +1,167 @@
+#ifdef STATIC
+/* Pre-boot environment: included */
+
+/* prevent inclusion of _LINUX_KERNEL_H in pre-boot environment: lots
+ * errors about console_printk etc... on ARM */
+#define _LINUX_KERNEL_H
+
+#include "zlib_inflate/inftrees.c"
+#include "zlib_inflate/inffast.c"
+#include "zlib_inflate/inflate.c"
+
+#else /* STATIC */
+/* initramfs et al: linked */
+
+#include <linux/zutil.h>
+
+#include "zlib_inflate/inftrees.h"
+#include "zlib_inflate/inffast.h"
+#include "zlib_inflate/inflate.h"
+
+#include "zlib_inflate/infutil.h"
+
+#endif /* STATIC */
+
+#include <linux/decompress/mm.h>
+
+#define INBUF_LEN (16*1024)
+
+/* Included from initramfs et al code */
+STATIC int INIT gunzip(unsigned char *buf, int len,
+		       int(*fill)(void*, unsigned int),
+		       int(*flush)(void*, unsigned int),
+		       unsigned char *out_buf,
+		       int *pos,
+		       void(*error_fn)(char *x)) {
+	u8 *zbuf;
+	struct z_stream_s *strm;
+	int rc;
+	size_t out_len;
+
+	set_error_fn(error_fn);
+	rc = -1;
+	if (flush) {
+		out_len = 0x8000; /* 32 K */
+		out_buf = malloc(out_len);
+	} else {
+		out_len = 0x7fffffff; /* no limit */
+	}
+	if (!out_buf) {
+		error("Out of memory while allocating output buffer");
+		goto gunzip_nomem1;
+	}
+
+	if (buf)
+		zbuf = buf;
+	else {
+		zbuf = malloc(INBUF_LEN);
+		len = 0;
+	}
+	if (!zbuf) {
+		error("Out of memory while allocating input buffer");
+		goto gunzip_nomem2;
+	}
+
+	strm = malloc(sizeof(*strm));
+	if (strm == NULL) {
+		error("Out of memory while allocating z_stream");
+		goto gunzip_nomem3;
+	}
+
+	strm->workspace = malloc(flush ? zlib_inflate_workspacesize() :
+				 sizeof(struct inflate_state));
+	if (strm->workspace == NULL) {
+		error("Out of memory while allocating workspace");
+		goto gunzip_nomem4;
+	}
+
+	if (len == 0)
+		len = fill(zbuf, INBUF_LEN);
+
+	/* verify the gzip header */
+	if (len < 10 ||
+	   zbuf[0] != 0x1f || zbuf[1] != 0x8b || zbuf[2] != 0x08) {
+		if (pos)
+			*pos = 0;
+		error("Not a gzip file");
+		goto gunzip_5;
+	}
+
+	/* skip over gzip header (1f,8b,08... 10 bytes total +
+	 * possible asciz filename)
+	 */
+	strm->next_in = zbuf + 10;
+	/* skip over asciz filename */
+	if (zbuf[3] & 0x8) {
+		while (strm->next_in[0])
+			strm->next_in++;
+		strm->next_in++;
+	}
+	strm->avail_in = len - (strm->next_in - zbuf);
+
+	strm->next_out = out_buf;
+	strm->avail_out = out_len;
+
+	rc = zlib_inflateInit2(strm, -MAX_WBITS);
+
+	if (!flush) {
+		WS(strm)->inflate_state.wsize = 0;
+		WS(strm)->inflate_state.window = NULL;
+	}
+
+	while (rc == Z_OK) {
+		if (strm->avail_in == 0) {
+			/* TODO: handle case where both pos and fill are set */
+			len = fill(zbuf, INBUF_LEN);
+			if (len < 0) {
+				rc = -1;
+				error("read error");
+				break;
+			}
+			strm->next_in = zbuf;
+			strm->avail_in = len;
+		}
+		rc = zlib_inflate(strm, 0);
+
+		/* Write any data generated */
+		if (flush && strm->next_out > out_buf) {
+			int l = strm->next_out - out_buf;
+			if (l != flush(out_buf, l)) {
+				rc = -1;
+				error("write error");
+				break;
+			}
+			strm->next_out = out_buf;
+			strm->avail_out = out_len;
+		}
+
+		/* after Z_FINISH, only Z_STREAM_END is "we unpacked it all" */
+		if (rc == Z_STREAM_END) {
+			rc = 0;
+			break;
+		} else if (rc != Z_OK) {
+			error("uncompression error");
+			rc = -1;
+		}
+	}
+
+	zlib_inflateEnd(strm);
+	if (pos)
+		/* add + 8 to skip over trailer */
+		*pos = strm->next_in - zbuf+8;
+
+gunzip_5:
+	free(strm->workspace);
+gunzip_nomem4:
+	free(strm);
+gunzip_nomem3:
+	if (!buf)
+		free(zbuf);
+gunzip_nomem2:
+	if (flush)
+		free(out_buf);
+gunzip_nomem1:
+	return rc; /* returns Z_OK (0) if successful */
+}
+
+#define decompress gunzip
diff --git a/lib/decompress_unlzma.c b/lib/decompress_unlzma.c
new file mode 100644
index 0000000..546f2f4
--- /dev/null
+++ b/lib/decompress_unlzma.c
@@ -0,0 +1,647 @@
+/* Lzma decompressor for Linux kernel. Shamelessly snarfed
+ *from busybox 1.1.1
+ *
+ *Linux kernel adaptation
+ *Copyright (C) 2006  Alain < alain@knaff.lu >
+ *
+ *Based on small lzma deflate implementation/Small range coder
+ *implementation for lzma.
+ *Copyright (C) 2006  Aurelien Jacobs < aurel@gnuage.org >
+ *
+ *Based on LzmaDecode.c from the LZMA SDK 4.22 (http://www.7-zip.org/)
+ *Copyright (C) 1999-2005  Igor Pavlov
+ *
+ *Copyrights of the parts, see headers below.
+ *
+ *
+ *This program 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 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
+ *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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#ifndef STATIC
+#include <linux/decompress/unlzma.h>
+#endif /* STATIC */
+
+#include <linux/decompress/mm.h>
+
+#define	MIN(a, b) (((a) < (b)) ? (a) : (b))
+
+static long long INIT read_int(unsigned char *ptr, int size)
+{
+	int i;
+	long long ret = 0;
+
+	for (i = 0; i < size; i++)
+		ret = (ret << 8) | ptr[size-i-1];
+	return ret;
+}
+
+#define ENDIAN_CONVERT(x) \
+  x = (typeof(x))read_int((unsigned char *)&x, sizeof(x))
+
+
+/* Small range coder implementation for lzma.
+ *Copyright (C) 2006  Aurelien Jacobs < aurel@gnuage.org >
+ *
+ *Based on LzmaDecode.c from the LZMA SDK 4.22 (http://www.7-zip.org/)
+ *Copyright (c) 1999-2005  Igor Pavlov
+ */
+
+#include <linux/compiler.h>
+
+#define LZMA_IOBUF_SIZE	0x10000
+
+struct rc {
+	int (*fill)(void*, unsigned int);
+	uint8_t *ptr;
+	uint8_t *buffer;
+	uint8_t *buffer_end;
+	int buffer_size;
+	uint32_t code;
+	uint32_t range;
+	uint32_t bound;
+};
+
+
+#define RC_TOP_BITS 24
+#define RC_MOVE_BITS 5
+#define RC_MODEL_TOTAL_BITS 11
+
+
+/* Called twice: once at startup and once in rc_normalize() */
+static void INIT rc_read(struct rc *rc)
+{
+	rc->buffer_size = rc->fill((char *)rc->buffer, LZMA_IOBUF_SIZE);
+	if (rc->buffer_size <= 0)
+		error("unexpected EOF");
+	rc->ptr = rc->buffer;
+	rc->buffer_end = rc->buffer + rc->buffer_size;
+}
+
+/* Called once */
+static inline void INIT rc_init(struct rc *rc,
+				       int (*fill)(void*, unsigned int),
+				       char *buffer, int buffer_size)
+{
+	rc->fill = fill;
+	rc->buffer = (uint8_t *)buffer;
+	rc->buffer_size = buffer_size;
+	rc->buffer_end = rc->buffer + rc->buffer_size;
+	rc->ptr = rc->buffer;
+
+	rc->code = 0;
+	rc->range = 0xFFFFFFFF;
+}
+
+static inline void INIT rc_init_code(struct rc *rc)
+{
+	int i;
+
+	for (i = 0; i < 5; i++) {
+		if (rc->ptr >= rc->buffer_end)
+			rc_read(rc);
+		rc->code = (rc->code << 8) | *rc->ptr++;
+	}
+}
+
+
+/* Called once. TODO: bb_maybe_free() */
+static inline void INIT rc_free(struct rc *rc)
+{
+	free(rc->buffer);
+}
+
+/* Called twice, but one callsite is in inline'd rc_is_bit_0_helper() */
+static void INIT rc_do_normalize(struct rc *rc)
+{
+	if (rc->ptr >= rc->buffer_end)
+		rc_read(rc);
+	rc->range <<= 8;
+	rc->code = (rc->code << 8) | *rc->ptr++;
+}
+static inline void INIT rc_normalize(struct rc *rc)
+{
+	if (rc->range < (1 << RC_TOP_BITS))
+		rc_do_normalize(rc);
+}
+
+/* Called 9 times */
+/* Why rc_is_bit_0_helper exists?
+ *Because we want to always expose (rc->code < rc->bound) to optimizer
+ */
+static inline uint32_t INIT rc_is_bit_0_helper(struct rc *rc, uint16_t *p)
+{
+	rc_normalize(rc);
+	rc->bound = *p * (rc->range >> RC_MODEL_TOTAL_BITS);
+	return rc->bound;
+}
+static inline int INIT rc_is_bit_0(struct rc *rc, uint16_t *p)
+{
+	uint32_t t = rc_is_bit_0_helper(rc, p);
+	return rc->code < t;
+}
+
+/* Called ~10 times, but very small, thus inlined */
+static inline void INIT rc_update_bit_0(struct rc *rc, uint16_t *p)
+{
+	rc->range = rc->bound;
+	*p += ((1 << RC_MODEL_TOTAL_BITS) - *p) >> RC_MOVE_BITS;
+}
+static inline void rc_update_bit_1(struct rc *rc, uint16_t *p)
+{
+	rc->range -= rc->bound;
+	rc->code -= rc->bound;
+	*p -= *p >> RC_MOVE_BITS;
+}
+
+/* Called 4 times in unlzma loop */
+static int INIT rc_get_bit(struct rc *rc, uint16_t *p, int *symbol)
+{
+	if (rc_is_bit_0(rc, p)) {
+		rc_update_bit_0(rc, p);
+		*symbol *= 2;
+		return 0;
+	} else {
+		rc_update_bit_1(rc, p);
+		*symbol = *symbol * 2 + 1;
+		return 1;
+	}
+}
+
+/* Called once */
+static inline int INIT rc_direct_bit(struct rc *rc)
+{
+	rc_normalize(rc);
+	rc->range >>= 1;
+	if (rc->code >= rc->range) {
+		rc->code -= rc->range;
+		return 1;
+	}
+	return 0;
+}
+
+/* Called twice */
+static inline void INIT
+rc_bit_tree_decode(struct rc *rc, uint16_t *p, int num_levels, int *symbol)
+{
+	int i = num_levels;
+
+	*symbol = 1;
+	while (i--)
+		rc_get_bit(rc, p + *symbol, symbol);
+	*symbol -= 1 << num_levels;
+}
+
+
+/*
+ * Small lzma deflate implementation.
+ * Copyright (C) 2006  Aurelien Jacobs < aurel@gnuage.org >
+ *
+ * Based on LzmaDecode.c from the LZMA SDK 4.22 (http://www.7-zip.org/)
+ * Copyright (C) 1999-2005  Igor Pavlov
+ */
+
+
+struct lzma_header {
+	uint8_t pos;
+	uint32_t dict_size;
+	uint64_t dst_size;
+} __attribute__ ((packed)) ;
+
+
+#define LZMA_BASE_SIZE 1846
+#define LZMA_LIT_SIZE 768
+
+#define LZMA_NUM_POS_BITS_MAX 4
+
+#define LZMA_LEN_NUM_LOW_BITS 3
+#define LZMA_LEN_NUM_MID_BITS 3
+#define LZMA_LEN_NUM_HIGH_BITS 8
+
+#define LZMA_LEN_CHOICE 0
+#define LZMA_LEN_CHOICE_2 (LZMA_LEN_CHOICE + 1)
+#define LZMA_LEN_LOW (LZMA_LEN_CHOICE_2 + 1)
+#define LZMA_LEN_MID (LZMA_LEN_LOW \
+		      + (1 << (LZMA_NUM_POS_BITS_MAX + LZMA_LEN_NUM_LOW_BITS)))
+#define LZMA_LEN_HIGH (LZMA_LEN_MID \
+		       +(1 << (LZMA_NUM_POS_BITS_MAX + LZMA_LEN_NUM_MID_BITS)))
+#define LZMA_NUM_LEN_PROBS (LZMA_LEN_HIGH + (1 << LZMA_LEN_NUM_HIGH_BITS))
+
+#define LZMA_NUM_STATES 12
+#define LZMA_NUM_LIT_STATES 7
+
+#define LZMA_START_POS_MODEL_INDEX 4
+#define LZMA_END_POS_MODEL_INDEX 14
+#define LZMA_NUM_FULL_DISTANCES (1 << (LZMA_END_POS_MODEL_INDEX >> 1))
+
+#define LZMA_NUM_POS_SLOT_BITS 6
+#define LZMA_NUM_LEN_TO_POS_STATES 4
+
+#define LZMA_NUM_ALIGN_BITS 4
+
+#define LZMA_MATCH_MIN_LEN 2
+
+#define LZMA_IS_MATCH 0
+#define LZMA_IS_REP (LZMA_IS_MATCH + (LZMA_NUM_STATES << LZMA_NUM_POS_BITS_MAX))
+#define LZMA_IS_REP_G0 (LZMA_IS_REP + LZMA_NUM_STATES)
+#define LZMA_IS_REP_G1 (LZMA_IS_REP_G0 + LZMA_NUM_STATES)
+#define LZMA_IS_REP_G2 (LZMA_IS_REP_G1 + LZMA_NUM_STATES)
+#define LZMA_IS_REP_0_LONG (LZMA_IS_REP_G2 + LZMA_NUM_STATES)
+#define LZMA_POS_SLOT (LZMA_IS_REP_0_LONG \
+		       + (LZMA_NUM_STATES << LZMA_NUM_POS_BITS_MAX))
+#define LZMA_SPEC_POS (LZMA_POS_SLOT \
+		       +(LZMA_NUM_LEN_TO_POS_STATES << LZMA_NUM_POS_SLOT_BITS))
+#define LZMA_ALIGN (LZMA_SPEC_POS \
+		    + LZMA_NUM_FULL_DISTANCES - LZMA_END_POS_MODEL_INDEX)
+#define LZMA_LEN_CODER (LZMA_ALIGN + (1 << LZMA_NUM_ALIGN_BITS))
+#define LZMA_REP_LEN_CODER (LZMA_LEN_CODER + LZMA_NUM_LEN_PROBS)
+#define LZMA_LITERAL (LZMA_REP_LEN_CODER + LZMA_NUM_LEN_PROBS)
+
+
+struct writer {
+	uint8_t *buffer;
+	uint8_t previous_byte;
+	size_t buffer_pos;
+	int bufsize;
+	size_t global_pos;
+	int(*flush)(void*, unsigned int);
+	struct lzma_header *header;
+};
+
+struct cstate {
+	int state;
+	uint32_t rep0, rep1, rep2, rep3;
+};
+
+static inline size_t INIT get_pos(struct writer *wr)
+{
+	return
+		wr->global_pos + wr->buffer_pos;
+}
+
+static inline uint8_t INIT peek_old_byte(struct writer *wr,
+						uint32_t offs)
+{
+	if (!wr->flush) {
+		int32_t pos;
+		while (offs > wr->header->dict_size)
+			offs -= wr->header->dict_size;
+		pos = wr->buffer_pos - offs;
+		return wr->buffer[pos];
+	} else {
+		uint32_t pos = wr->buffer_pos - offs;
+		while (pos >= wr->header->dict_size)
+			pos += wr->header->dict_size;
+		return wr->buffer[pos];
+	}
+
+}
+
+static inline void INIT write_byte(struct writer *wr, uint8_t byte)
+{
+	wr->buffer[wr->buffer_pos++] = wr->previous_byte = byte;
+	if (wr->flush && wr->buffer_pos == wr->header->dict_size) {
+		wr->buffer_pos = 0;
+		wr->global_pos += wr->header->dict_size;
+		wr->flush((char *)wr->buffer, wr->header->dict_size);
+	}
+}
+
+
+static inline void INIT copy_byte(struct writer *wr, uint32_t offs)
+{
+	write_byte(wr, peek_old_byte(wr, offs));
+}
+
+static inline void INIT copy_bytes(struct writer *wr,
+					 uint32_t rep0, int len)
+{
+	do {
+		copy_byte(wr, rep0);
+		len--;
+	} while (len != 0 && wr->buffer_pos < wr->header->dst_size);
+}
+
+static inline void INIT process_bit0(struct writer *wr, struct rc *rc,
+				     struct cstate *cst, uint16_t *p,
+				     int pos_state, uint16_t *prob,
+				     int lc, uint32_t literal_pos_mask) {
+	int mi = 1;
+	rc_update_bit_0(rc, prob);
+	prob = (p + LZMA_LITERAL +
+		(LZMA_LIT_SIZE
+		 * (((get_pos(wr) & literal_pos_mask) << lc)
+		    + (wr->previous_byte >> (8 - lc))))
+		);
+
+	if (cst->state >= LZMA_NUM_LIT_STATES) {
+		int match_byte = peek_old_byte(wr, cst->rep0);
+		do {
+			int bit;
+			uint16_t *prob_lit;
+
+			match_byte <<= 1;
+			bit = match_byte & 0x100;
+			prob_lit = prob + 0x100 + bit + mi;
+			if (rc_get_bit(rc, prob_lit, &mi)) {
+				if (!bit)
+					break;
+			} else {
+				if (bit)
+					break;
+			}
+		} while (mi < 0x100);
+	}
+	while (mi < 0x100) {
+		uint16_t *prob_lit = prob + mi;
+		rc_get_bit(rc, prob_lit, &mi);
+	}
+	write_byte(wr, mi);
+	if (cst->state < 4)
+		cst->state = 0;
+	else if (cst->state < 10)
+		cst->state -= 3;
+	else
+		cst->state -= 6;
+}
+
+static inline void INIT process_bit1(struct writer *wr, struct rc *rc,
+					    struct cstate *cst, uint16_t *p,
+					    int pos_state, uint16_t *prob) {
+  int offset;
+	uint16_t *prob_len;
+	int num_bits;
+	int len;
+
+	rc_update_bit_1(rc, prob);
+	prob = p + LZMA_IS_REP + cst->state;
+	if (rc_is_bit_0(rc, prob)) {
+		rc_update_bit_0(rc, prob);
+		cst->rep3 = cst->rep2;
+		cst->rep2 = cst->rep1;
+		cst->rep1 = cst->rep0;
+		cst->state = cst->state < LZMA_NUM_LIT_STATES ? 0 : 3;
+		prob = p + LZMA_LEN_CODER;
+	} else {
+		rc_update_bit_1(rc, prob);
+		prob = p + LZMA_IS_REP_G0 + cst->state;
+		if (rc_is_bit_0(rc, prob)) {
+			rc_update_bit_0(rc, prob);
+			prob = (p + LZMA_IS_REP_0_LONG
+				+ (cst->state <<
+				   LZMA_NUM_POS_BITS_MAX) +
+				pos_state);
+			if (rc_is_bit_0(rc, prob)) {
+				rc_update_bit_0(rc, prob);
+
+				cst->state = cst->state < LZMA_NUM_LIT_STATES ?
+					9 : 11;
+				copy_byte(wr, cst->rep0);
+				return;
+			} else {
+				rc_update_bit_1(rc, prob);
+			}
+		} else {
+			uint32_t distance;
+
+			rc_update_bit_1(rc, prob);
+			prob = p + LZMA_IS_REP_G1 + cst->state;
+			if (rc_is_bit_0(rc, prob)) {
+				rc_update_bit_0(rc, prob);
+				distance = cst->rep1;
+			} else {
+				rc_update_bit_1(rc, prob);
+				prob = p + LZMA_IS_REP_G2 + cst->state;
+				if (rc_is_bit_0(rc, prob)) {
+					rc_update_bit_0(rc, prob);
+					distance = cst->rep2;
+				} else {
+					rc_update_bit_1(rc, prob);
+					distance = cst->rep3;
+					cst->rep3 = cst->rep2;
+				}
+				cst->rep2 = cst->rep1;
+			}
+			cst->rep1 = cst->rep0;
+			cst->rep0 = distance;
+		}
+		cst->state = cst->state < LZMA_NUM_LIT_STATES ? 8 : 11;
+		prob = p + LZMA_REP_LEN_CODER;
+	}
+
+	prob_len = prob + LZMA_LEN_CHOICE;
+	if (rc_is_bit_0(rc, prob_len)) {
+		rc_update_bit_0(rc, prob_len);
+		prob_len = (prob + LZMA_LEN_LOW
+			    + (pos_state <<
+			       LZMA_LEN_NUM_LOW_BITS));
+		offset = 0;
+		num_bits = LZMA_LEN_NUM_LOW_BITS;
+	} else {
+		rc_update_bit_1(rc, prob_len);
+		prob_len = prob + LZMA_LEN_CHOICE_2;
+		if (rc_is_bit_0(rc, prob_len)) {
+			rc_update_bit_0(rc, prob_len);
+			prob_len = (prob + LZMA_LEN_MID
+				    + (pos_state <<
+				       LZMA_LEN_NUM_MID_BITS));
+			offset = 1 << LZMA_LEN_NUM_LOW_BITS;
+			num_bits = LZMA_LEN_NUM_MID_BITS;
+		} else {
+			rc_update_bit_1(rc, prob_len);
+			prob_len = prob + LZMA_LEN_HIGH;
+			offset = ((1 << LZMA_LEN_NUM_LOW_BITS)
+				  + (1 << LZMA_LEN_NUM_MID_BITS));
+			num_bits = LZMA_LEN_NUM_HIGH_BITS;
+		}
+	}
+
+	rc_bit_tree_decode(rc, prob_len, num_bits, &len);
+	len += offset;
+
+	if (cst->state < 4) {
+		int pos_slot;
+
+		cst->state += LZMA_NUM_LIT_STATES;
+		prob =
+			p + LZMA_POS_SLOT +
+			((len <
+			  LZMA_NUM_LEN_TO_POS_STATES ? len :
+			  LZMA_NUM_LEN_TO_POS_STATES - 1)
+			 << LZMA_NUM_POS_SLOT_BITS);
+		rc_bit_tree_decode(rc, prob,
+				   LZMA_NUM_POS_SLOT_BITS,
+				   &pos_slot);
+		if (pos_slot >= LZMA_START_POS_MODEL_INDEX) {
+			int i, mi;
+			num_bits = (pos_slot >> 1) - 1;
+			cst->rep0 = 2 | (pos_slot & 1);
+			if (pos_slot < LZMA_END_POS_MODEL_INDEX) {
+				cst->rep0 <<= num_bits;
+				prob = p + LZMA_SPEC_POS +
+					cst->rep0 - pos_slot - 1;
+			} else {
+				num_bits -= LZMA_NUM_ALIGN_BITS;
+				while (num_bits--)
+					cst->rep0 = (cst->rep0 << 1) |
+						rc_direct_bit(rc);
+				prob = p + LZMA_ALIGN;
+				cst->rep0 <<= LZMA_NUM_ALIGN_BITS;
+				num_bits = LZMA_NUM_ALIGN_BITS;
+			}
+			i = 1;
+			mi = 1;
+			while (num_bits--) {
+				if (rc_get_bit(rc, prob + mi, &mi))
+					cst->rep0 |= i;
+				i <<= 1;
+			}
+		} else
+			cst->rep0 = pos_slot;
+		if (++(cst->rep0) == 0)
+			return;
+	}
+
+	len += LZMA_MATCH_MIN_LEN;
+
+	copy_bytes(wr, cst->rep0, len);
+}
+
+
+
+STATIC inline int INIT unlzma(unsigned char *buf, int in_len,
+			      int(*fill)(void*, unsigned int),
+			      int(*flush)(void*, unsigned int),
+			      unsigned char *output,
+			      int *posp,
+			      void(*error_fn)(char *x)
+	)
+{
+	struct lzma_header header;
+	int lc, pb, lp;
+	uint32_t pos_state_mask;
+	uint32_t literal_pos_mask;
+	uint16_t *p;
+	int num_probs;
+	struct rc rc;
+	int i, mi;
+	struct writer wr;
+	struct cstate cst;
+	unsigned char *inbuf;
+	int ret = -1;
+
+	set_error_fn(error_fn);
+	if (!flush)
+		in_len -= 4; /* Uncompressed size hack active in pre-boot
+				environment */
+	if (buf)
+		inbuf = buf;
+	else
+		inbuf = malloc(LZMA_IOBUF_SIZE);
+	if (!inbuf) {
+		error("Could not allocate input bufer");
+		goto exit_0;
+	}
+
+	cst.state = 0;
+	cst.rep0 = cst.rep1 = cst.rep2 = cst.rep3 = 1;
+
+	wr.header = &header;
+	wr.flush = flush;
+	wr.global_pos = 0;
+	wr.previous_byte = 0;
+	wr.buffer_pos = 0;
+
+	rc_init(&rc, fill, inbuf, in_len);
+
+	for (i = 0; i < sizeof(header); i++) {
+		if (rc.ptr >= rc.buffer_end)
+			rc_read(&rc);
+		((unsigned char *)&header)[i] = *rc.ptr++;
+	}
+
+	if (header.pos >= (9 * 5 * 5))
+		error("bad header");
+
+	mi = 0;
+	lc = header.pos;
+	while (lc >= 9) {
+		mi++;
+		lc -= 9;
+	}
+	pb = 0;
+	lp = mi;
+	while (lp >= 5) {
+		pb++;
+		lp -= 5;
+	}
+	pos_state_mask = (1 << pb) - 1;
+	literal_pos_mask = (1 << lp) - 1;
+
+	ENDIAN_CONVERT(header.dict_size);
+	ENDIAN_CONVERT(header.dst_size);
+
+	if (header.dict_size == 0)
+		header.dict_size = 1;
+
+	if (output)
+		wr.buffer = output;
+	else {
+		wr.bufsize = MIN(header.dst_size, header.dict_size);
+		wr.buffer = large_malloc(wr.bufsize);
+	}
+	if (wr.buffer == NULL)
+		goto exit_1;
+
+	num_probs = LZMA_BASE_SIZE + (LZMA_LIT_SIZE << (lc + lp));
+	p = (uint16_t *) large_malloc(num_probs * sizeof(*p));
+	if (p == 0)
+		goto exit_2;
+	num_probs = LZMA_LITERAL + (LZMA_LIT_SIZE << (lc + lp));
+	for (i = 0; i < num_probs; i++)
+		p[i] = (1 << RC_MODEL_TOTAL_BITS) >> 1;
+
+	rc_init_code(&rc);
+
+	while (get_pos(&wr) < header.dst_size) {
+		int pos_state =	get_pos(&wr) & pos_state_mask;
+		uint16_t *prob = p + LZMA_IS_MATCH +
+			(cst.state << LZMA_NUM_POS_BITS_MAX) + pos_state;
+		if (rc_is_bit_0(&rc, prob))
+			process_bit0(&wr, &rc, &cst, p, pos_state, prob,
+				     lc, literal_pos_mask);
+		else {
+			process_bit1(&wr, &rc, &cst, p, pos_state, prob);
+			if (cst.rep0 == 0)
+				break;
+		}
+	}
+
+	if (posp)
+		*posp = rc.ptr-rc.buffer;
+	if (wr.flush)
+		wr.flush(wr.buffer, wr.buffer_pos);
+	ret = 0;
+	large_free(p);
+exit_2:
+	if (!output)
+		large_free(wr.buffer);
+exit_1:
+	if (!buf)
+		free(inbuf);
+exit_0:
+	return ret;
+}
+
+#define decompress unlzma
diff --git a/lib/zlib_inflate/inflate.h b/lib/zlib_inflate/inflate.h
index df8a6c9..3d17b3d 100644
--- a/lib/zlib_inflate/inflate.h
+++ b/lib/zlib_inflate/inflate.h
@@ -1,3 +1,6 @@
+#ifndef INFLATE_H
+#define INFLATE_H
+
 /* inflate.h -- internal inflate state definition
  * Copyright (C) 1995-2004 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
@@ -105,3 +108,4 @@
     unsigned short work[288];   /* work area for code table building */
     code codes[ENOUGH];         /* space for code tables */
 };
+#endif
diff --git a/lib/zlib_inflate/inftrees.h b/lib/zlib_inflate/inftrees.h
index 5f5219b..b70b473 100644
--- a/lib/zlib_inflate/inftrees.h
+++ b/lib/zlib_inflate/inftrees.h
@@ -1,3 +1,6 @@
+#ifndef INFTREES_H
+#define INFTREES_H
+
 /* inftrees.h -- header to use inftrees.c
  * Copyright (C) 1995-2005 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
@@ -53,3 +56,4 @@
 extern int zlib_inflate_table (codetype type, unsigned short *lens,
                              unsigned codes, code **table,
                              unsigned *bits, unsigned short *work);
+#endif
diff --git a/mm/filemap.c b/mm/filemap.c
index 60fd567..126d397 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1816,14 +1816,14 @@
 static size_t __iovec_copy_from_user_inatomic(char *vaddr,
 			const struct iovec *iov, size_t base, size_t bytes)
 {
-	size_t copied = 0, left = 0, total = bytes;
+	size_t copied = 0, left = 0;
 
 	while (bytes) {
 		char __user *buf = iov->iov_base + base;
 		int copy = min(bytes, iov->iov_len - base);
 
 		base = 0;
-		left = __copy_from_user_inatomic_nocache(vaddr, buf, copy, total);
+		left = __copy_from_user_inatomic(vaddr, buf, copy);
 		copied += copy;
 		bytes -= copy;
 		vaddr += copy;
@@ -1851,9 +1851,7 @@
 	if (likely(i->nr_segs == 1)) {
 		int left;
 		char __user *buf = i->iov->iov_base + i->iov_offset;
-
-		left = __copy_from_user_inatomic_nocache(kaddr + offset,
-							buf, bytes, bytes);
+		left = __copy_from_user_inatomic(kaddr + offset, buf, bytes);
 		copied = bytes - left;
 	} else {
 		copied = __iovec_copy_from_user_inatomic(kaddr + offset,
@@ -1881,8 +1879,7 @@
 	if (likely(i->nr_segs == 1)) {
 		int left;
 		char __user *buf = i->iov->iov_base + i->iov_offset;
-
-		left = __copy_from_user_nocache(kaddr + offset, buf, bytes, bytes);
+		left = __copy_from_user(kaddr + offset, buf, bytes);
 		copied = bytes - left;
 	} else {
 		copied = __iovec_copy_from_user_inatomic(kaddr + offset,
diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c
index bf54f8a..0c04615 100644
--- a/mm/filemap_xip.c
+++ b/mm/filemap_xip.c
@@ -354,7 +354,7 @@
 			break;
 
 		copied = bytes -
-			__copy_from_user_nocache(xip_mem + offset, buf, bytes, bytes);
+			__copy_from_user_nocache(xip_mem + offset, buf, bytes);
 
 		if (likely(copied > 0)) {
 			status = copied;
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index e9db889..2886d2f 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -1,12 +1,16 @@
 #include <linux/skbuff.h>
 #include <linux/netdevice.h>
 #include <linux/if_vlan.h>
+#include <linux/netpoll.h>
 #include "vlan.h"
 
 /* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
 int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		      u16 vlan_tci, int polling)
 {
+	if (netpoll_rx(skb))
+		return NET_RX_DROP;
+
 	if (skb_bond_should_drop(skb))
 		goto drop;
 
@@ -100,6 +104,9 @@
 {
 	int err = NET_RX_SUCCESS;
 
+	if (netpoll_receive_skb(skb))
+		return NET_RX_DROP;
+
 	switch (vlan_gro_common(napi, grp, vlan_tci, skb)) {
 	case -1:
 		return netif_receive_skb(skb);
@@ -126,6 +133,9 @@
 	if (!skb)
 		goto out;
 
+	if (netpoll_receive_skb(skb))
+		goto out;
+
 	err = NET_RX_SUCCESS;
 
 	switch (vlan_gro_common(napi, grp, vlan_tci, skb)) {
diff --git a/net/core/dev.c b/net/core/dev.c
index a17e006..72b0d26f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2488,6 +2488,9 @@
 
 int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 {
+	if (netpoll_receive_skb(skb))
+		return NET_RX_DROP;
+
 	switch (__napi_gro_receive(napi, skb)) {
 	case -1:
 		return netif_receive_skb(skb);
@@ -2558,6 +2561,9 @@
 	if (!skb)
 		goto out;
 
+	if (netpoll_receive_skb(skb))
+		goto out;
+
 	err = NET_RX_SUCCESS;
 
 	switch (__napi_gro_receive(napi, skb)) {
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a6961d7..c28976a7 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1374,7 +1374,8 @@
 
 static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
 			   struct tcp_sacktag_state *state,
-			   unsigned int pcount, int shifted, int mss)
+			   unsigned int pcount, int shifted, int mss,
+			   int dup_sack)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *prev = tcp_write_queue_prev(sk, skb);
@@ -1410,7 +1411,7 @@
 	}
 
 	/* We discard results */
-	tcp_sacktag_one(skb, sk, state, 0, pcount);
+	tcp_sacktag_one(skb, sk, state, dup_sack, pcount);
 
 	/* Difference in this won't matter, both ACKed by the same cumul. ACK */
 	TCP_SKB_CB(prev)->sacked |= (TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS);
@@ -1561,7 +1562,7 @@
 
 	if (!skb_shift(prev, skb, len))
 		goto fallback;
-	if (!tcp_shifted_skb(sk, skb, state, pcount, len, mss))
+	if (!tcp_shifted_skb(sk, skb, state, pcount, len, mss, dup_sack))
 		goto out;
 
 	/* Hole filled allows collapsing with the next as well, this is very
@@ -1580,7 +1581,7 @@
 	len = skb->len;
 	if (skb_shift(prev, skb, len)) {
 		pcount += tcp_skb_pcount(skb);
-		tcp_shifted_skb(sk, skb, state, tcp_skb_pcount(skb), len, mss);
+		tcp_shifted_skb(sk, skb, state, tcp_skb_pcount(skb), len, mss, 0);
 	}
 
 out:
diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c
index 2747ec7..4660b08 100644
--- a/net/ipv4/tcp_scalable.c
+++ b/net/ipv4/tcp_scalable.c
@@ -1,6 +1,6 @@
 /* Tom Kelly's Scalable TCP
  *
- * See htt://www-lce.eng.cam.ac.uk/~ctk21/scalable/
+ * See http://www.deneholme.net/tom/scalable/
  *
  * John Heffner <jheffner@sc.edu>
  */
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 8fe267f..1bcc343 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -258,11 +258,11 @@
 
 	if (twp != NULL) {
 		*twp = tw;
-		NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITRECYCLED);
+		NET_INC_STATS_BH(net, LINUX_MIB_TIMEWAITRECYCLED);
 	} else if (tw != NULL) {
 		/* Silly. Should hash-dance instead... */
 		inet_twsk_deschedule(tw, death_row);
-		NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITRECYCLED);
+		NET_INC_STATS_BH(net, LINUX_MIB_TIMEWAITRECYCLED);
 
 		inet_twsk_put(tw);
 	}
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index c323643..72dbb6d 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -201,8 +201,9 @@
 
 	if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING &&
 	    nf_ip6_checksum(skb, hooknum, dataoff, IPPROTO_ICMPV6)) {
-		nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL,
-			      "nf_ct_icmpv6: ICMPv6 checksum failed\n");
+		if (LOG_INVALID(net, IPPROTO_ICMPV6))
+			nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL,
+				      "nf_ct_icmpv6: ICMPv6 checksum failed ");
 		return -NF_ACCEPT;
 	}
 
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index fa49dc7..c712e9f 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -39,7 +39,7 @@
 #endif
 
 #define NFULNL_NLBUFSIZ_DEFAULT	NLMSG_GOODSIZE
-#define NFULNL_TIMEOUT_DEFAULT 	HZ	/* every second */
+#define NFULNL_TIMEOUT_DEFAULT 	100	/* every second */
 #define NFULNL_QTHRESH_DEFAULT 	100	/* 100 packets */
 #define NFULNL_COPY_RANGE_MAX	0xFFFF	/* max packet size is limited by 16-bit struct nfattr nfa_len field */
 
@@ -590,8 +590,10 @@
 
 	qthreshold = inst->qthreshold;
 	/* per-rule qthreshold overrides per-instance */
-	if (qthreshold > li->u.ulog.qthreshold)
-		qthreshold = li->u.ulog.qthreshold;
+	if (li->u.ulog.qthreshold)
+		if (qthreshold > li->u.ulog.qthreshold)
+			qthreshold = li->u.ulog.qthreshold;
+
 
 	switch (inst->copy_mode) {
 	case NFULNL_COPY_META:
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index bfbf521..5baccfa 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -827,59 +827,143 @@
 	.release = seq_release_net,
 };
 
+/*
+ * Traverse state for ip{,6}_{tables,matches} for helping crossing
+ * the multi-AF mutexes.
+ */
+struct nf_mttg_trav {
+	struct list_head *head, *curr;
+	uint8_t class, nfproto;
+};
+
+enum {
+	MTTG_TRAV_INIT,
+	MTTG_TRAV_NFP_UNSPEC,
+	MTTG_TRAV_NFP_SPEC,
+	MTTG_TRAV_DONE,
+};
+
+static void *xt_mttg_seq_next(struct seq_file *seq, void *v, loff_t *ppos,
+    bool is_target)
+{
+	static const uint8_t next_class[] = {
+		[MTTG_TRAV_NFP_UNSPEC] = MTTG_TRAV_NFP_SPEC,
+		[MTTG_TRAV_NFP_SPEC]   = MTTG_TRAV_DONE,
+	};
+	struct nf_mttg_trav *trav = seq->private;
+
+	switch (trav->class) {
+	case MTTG_TRAV_INIT:
+		trav->class = MTTG_TRAV_NFP_UNSPEC;
+		mutex_lock(&xt[NFPROTO_UNSPEC].mutex);
+		trav->head = trav->curr = is_target ?
+			&xt[NFPROTO_UNSPEC].target : &xt[NFPROTO_UNSPEC].match;
+ 		break;
+	case MTTG_TRAV_NFP_UNSPEC:
+		trav->curr = trav->curr->next;
+		if (trav->curr != trav->head)
+			break;
+		mutex_unlock(&xt[NFPROTO_UNSPEC].mutex);
+		mutex_lock(&xt[trav->nfproto].mutex);
+		trav->head = trav->curr = is_target ?
+			&xt[trav->nfproto].target : &xt[trav->nfproto].match;
+		trav->class = next_class[trav->class];
+		break;
+	case MTTG_TRAV_NFP_SPEC:
+		trav->curr = trav->curr->next;
+		if (trav->curr != trav->head)
+			break;
+		/* fallthru, _stop will unlock */
+	default:
+		return NULL;
+	}
+
+	if (ppos != NULL)
+		++*ppos;
+	return trav;
+}
+
+static void *xt_mttg_seq_start(struct seq_file *seq, loff_t *pos,
+    bool is_target)
+{
+	struct nf_mttg_trav *trav = seq->private;
+	unsigned int j;
+
+	trav->class = MTTG_TRAV_INIT;
+	for (j = 0; j < *pos; ++j)
+		if (xt_mttg_seq_next(seq, NULL, NULL, is_target) == NULL)
+			return NULL;
+	return trav;
+}
+
+static void xt_mttg_seq_stop(struct seq_file *seq, void *v)
+{
+	struct nf_mttg_trav *trav = seq->private;
+
+	switch (trav->class) {
+	case MTTG_TRAV_NFP_UNSPEC:
+		mutex_unlock(&xt[NFPROTO_UNSPEC].mutex);
+		break;
+	case MTTG_TRAV_NFP_SPEC:
+		mutex_unlock(&xt[trav->nfproto].mutex);
+		break;
+	}
+}
+
 static void *xt_match_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-	u_int16_t af = (unsigned long)pde->data;
-
-	mutex_lock(&xt[af].mutex);
-	return seq_list_start(&xt[af].match, *pos);
+	return xt_mttg_seq_start(seq, pos, false);
 }
 
-static void *xt_match_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+static void *xt_match_seq_next(struct seq_file *seq, void *v, loff_t *ppos)
 {
-	struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-	u_int16_t af = (unsigned long)pde->data;
-
-	return seq_list_next(v, &xt[af].match, pos);
-}
-
-static void xt_match_seq_stop(struct seq_file *seq, void *v)
-{
-	struct proc_dir_entry *pde = seq->private;
-	u_int16_t af = (unsigned long)pde->data;
-
-	mutex_unlock(&xt[af].mutex);
+	return xt_mttg_seq_next(seq, v, ppos, false);
 }
 
 static int xt_match_seq_show(struct seq_file *seq, void *v)
 {
-	struct xt_match *match = list_entry(v, struct xt_match, list);
+	const struct nf_mttg_trav *trav = seq->private;
+	const struct xt_match *match;
 
-	if (strlen(match->name))
-		return seq_printf(seq, "%s\n", match->name);
-	else
-		return 0;
+	switch (trav->class) {
+	case MTTG_TRAV_NFP_UNSPEC:
+	case MTTG_TRAV_NFP_SPEC:
+		if (trav->curr == trav->head)
+			return 0;
+		match = list_entry(trav->curr, struct xt_match, list);
+		return (*match->name == '\0') ? 0 :
+		       seq_printf(seq, "%s\n", match->name);
+	}
+	return 0;
 }
 
 static const struct seq_operations xt_match_seq_ops = {
 	.start	= xt_match_seq_start,
 	.next	= xt_match_seq_next,
-	.stop	= xt_match_seq_stop,
+	.stop	= xt_mttg_seq_stop,
 	.show	= xt_match_seq_show,
 };
 
 static int xt_match_open(struct inode *inode, struct file *file)
 {
+	struct seq_file *seq;
+	struct nf_mttg_trav *trav;
 	int ret;
 
-	ret = seq_open(file, &xt_match_seq_ops);
-	if (!ret) {
-		struct seq_file *seq = file->private_data;
+	trav = kmalloc(sizeof(*trav), GFP_KERNEL);
+	if (trav == NULL)
+		return -ENOMEM;
 
-		seq->private = PDE(inode);
+	ret = seq_open(file, &xt_match_seq_ops);
+	if (ret < 0) {
+		kfree(trav);
+		return ret;
 	}
-	return ret;
+
+	seq = file->private_data;
+	seq->private = trav;
+	trav->nfproto = (unsigned long)PDE(inode)->data;
+	return 0;
 }
 
 static const struct file_operations xt_match_ops = {
@@ -887,62 +971,63 @@
 	.open	 = xt_match_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = seq_release,
+	.release = seq_release_private,
 };
 
 static void *xt_target_seq_start(struct seq_file *seq, loff_t *pos)
 {
-	struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-	u_int16_t af = (unsigned long)pde->data;
-
-	mutex_lock(&xt[af].mutex);
-	return seq_list_start(&xt[af].target, *pos);
+	return xt_mttg_seq_start(seq, pos, true);
 }
 
-static void *xt_target_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+static void *xt_target_seq_next(struct seq_file *seq, void *v, loff_t *ppos)
 {
-	struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-	u_int16_t af = (unsigned long)pde->data;
-
-	return seq_list_next(v, &xt[af].target, pos);
-}
-
-static void xt_target_seq_stop(struct seq_file *seq, void *v)
-{
-	struct proc_dir_entry *pde = seq->private;
-	u_int16_t af = (unsigned long)pde->data;
-
-	mutex_unlock(&xt[af].mutex);
+	return xt_mttg_seq_next(seq, v, ppos, true);
 }
 
 static int xt_target_seq_show(struct seq_file *seq, void *v)
 {
-	struct xt_target *target = list_entry(v, struct xt_target, list);
+	const struct nf_mttg_trav *trav = seq->private;
+	const struct xt_target *target;
 
-	if (strlen(target->name))
-		return seq_printf(seq, "%s\n", target->name);
-	else
-		return 0;
+	switch (trav->class) {
+	case MTTG_TRAV_NFP_UNSPEC:
+	case MTTG_TRAV_NFP_SPEC:
+		if (trav->curr == trav->head)
+			return 0;
+		target = list_entry(trav->curr, struct xt_target, list);
+		return (*target->name == '\0') ? 0 :
+		       seq_printf(seq, "%s\n", target->name);
+	}
+	return 0;
 }
 
 static const struct seq_operations xt_target_seq_ops = {
 	.start	= xt_target_seq_start,
 	.next	= xt_target_seq_next,
-	.stop	= xt_target_seq_stop,
+	.stop	= xt_mttg_seq_stop,
 	.show	= xt_target_seq_show,
 };
 
 static int xt_target_open(struct inode *inode, struct file *file)
 {
+	struct seq_file *seq;
+	struct nf_mttg_trav *trav;
 	int ret;
 
-	ret = seq_open(file, &xt_target_seq_ops);
-	if (!ret) {
-		struct seq_file *seq = file->private_data;
+	trav = kmalloc(sizeof(*trav), GFP_KERNEL);
+	if (trav == NULL)
+		return -ENOMEM;
 
-		seq->private = PDE(inode);
+	ret = seq_open(file, &xt_target_seq_ops);
+	if (ret < 0) {
+		kfree(trav);
+		return ret;
 	}
-	return ret;
+
+	seq = file->private_data;
+	seq->private = trav;
+	trav->nfproto = (unsigned long)PDE(inode)->data;
+	return 0;
 }
 
 static const struct file_operations xt_target_ops = {
@@ -950,7 +1035,7 @@
 	.open	 = xt_target_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = seq_release,
+	.release = seq_release_private,
 };
 
 #define FORMAT_TABLES	"_tables_names"
diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c
index fe80b61..791e030 100644
--- a/net/netfilter/xt_recent.c
+++ b/net/netfilter/xt_recent.c
@@ -542,7 +542,7 @@
 	struct recent_entry *e;
 	char buf[sizeof("+b335:1d35:1e55:dead:c0de:1715:5afe:c0de")];
 	const char *c = buf;
-	union nf_inet_addr addr;
+	union nf_inet_addr addr = {};
 	u_int16_t family;
 	bool add, succ;
 
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index f6b4fa9..e36e94a 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -66,11 +66,15 @@
 {
 	struct drr_sched *q = qdisc_priv(sch);
 	struct drr_class *cl = (struct drr_class *)*arg;
+	struct nlattr *opt = tca[TCA_OPTIONS];
 	struct nlattr *tb[TCA_DRR_MAX + 1];
 	u32 quantum;
 	int err;
 
-	err = nla_parse_nested(tb, TCA_DRR_MAX, tca[TCA_OPTIONS], drr_policy);
+	if (!opt)
+		return -EINVAL;
+
+	err = nla_parse_nested(tb, TCA_DRR_MAX, opt, drr_policy);
 	if (err < 0)
 		return err;
 
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index e0636577..3b949a3 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -186,3 +186,17 @@
 cmd_gzip = gzip -f -9 < $< > $@
 
 
+# Bzip2
+# ---------------------------------------------------------------------------
+
+# Bzip2 does not include size in file... so we have to fake that
+size_append=$(CONFIG_SHELL) $(srctree)/scripts/bin_size
+
+quiet_cmd_bzip2 = BZIP2    $@
+cmd_bzip2 = (bzip2 -9 < $< && $(size_append) $<) > $@ || (rm -f $@ ; false)
+
+# Lzma
+# ---------------------------------------------------------------------------
+
+quiet_cmd_lzma = LZMA    $@
+cmd_lzma = (lzma -9 -c $< && $(size_append) $<) >$@ || (rm -f $@ ; false)
diff --git a/scripts/bin_size b/scripts/bin_size
new file mode 100644
index 0000000..43e1b36
--- /dev/null
+++ b/scripts/bin_size
@@ -0,0 +1,10 @@
+#!/bin/sh
+
+if [ $# = 0 ] ; then
+   echo Usage: $0 file
+fi
+
+size_dec=`stat -c "%s" $1`
+size_hex_echo_string=`printf "%08x" $size_dec |
+     sed 's/\(..\)\(..\)\(..\)\(..\)/\\\\x\4\\\\x\3\\\\x\2\\\\x\1/g'`
+/bin/echo -ne $size_hex_echo_string
diff --git a/scripts/gen_initramfs_list.sh b/scripts/gen_initramfs_list.sh
index 5f3415f..3eea8f1 100644
--- a/scripts/gen_initramfs_list.sh
+++ b/scripts/gen_initramfs_list.sh
@@ -5,7 +5,7 @@
 # Released under the terms of the GNU GPL
 #
 # Generate a cpio packed initramfs. It uses gen_init_cpio to generate
-# the cpio archive, and gzip to pack it.
+# the cpio archive, and then compresses it.
 # The script may also be used to generate the inputfile used for gen_init_cpio
 # This script assumes that gen_init_cpio is located in usr/ directory
 
@@ -16,8 +16,8 @@
 cat << EOF
 Usage:
 $0 [-o <file>] [-u <uid>] [-g <gid>] {-d | <cpio_source>} ...
-	-o <file>      Create gzipped initramfs file named <file> using
-		       gen_init_cpio and gzip
+	-o <file>      Create compressed initramfs file named <file> using
+		       gen_init_cpio and compressor depending on the extension
 	-u <uid>       User ID to map to user ID 0 (root).
 		       <uid> is only meaningful if <cpio_source> is a
 		       directory.  "squash" forces all files to uid 0.
@@ -225,6 +225,7 @@
 output="/dev/stdout"
 output_file=""
 is_cpio_compressed=
+compr="gzip -9 -f"
 
 arg="$1"
 case "$arg" in
@@ -233,11 +234,15 @@
 		echo "deps_initramfs := \\"
 		shift
 		;;
-	"-o")	# generate gzipped cpio image named $1
+	"-o")	# generate compressed cpio image named $1
 		shift
 		output_file="$1"
 		cpio_list="$(mktemp ${TMPDIR:-/tmp}/cpiolist.XXXXXX)"
 		output=${cpio_list}
+		echo "$output_file" | grep -q "\.gz$" && compr="gzip -9 -f"
+		echo "$output_file" | grep -q "\.bz2$" && compr="bzip2 -9 -f"
+		echo "$output_file" | grep -q "\.lzma$" && compr="lzma -9 -f"
+		echo "$output_file" | grep -q "\.cpio$" && compr="cat"
 		shift
 		;;
 esac
@@ -274,7 +279,7 @@
 	esac
 done
 
-# If output_file is set we will generate cpio archive and gzip it
+# If output_file is set we will generate cpio archive and compress it
 # we are carefull to delete tmp files
 if [ ! -z ${output_file} ]; then
 	if [ -z ${cpio_file} ]; then
@@ -287,7 +292,8 @@
 	if [ "${is_cpio_compressed}" = "compressed" ]; then
 		cat ${cpio_tfile} > ${output_file}
 	else
-		cat ${cpio_tfile} | gzip -f -9 - > ${output_file}
+		(cat ${cpio_tfile} | ${compr}  - > ${output_file}) \
+		|| (rm -f ${output_file} ; false)
 	fi
 	[ -z ${cpio_file} ] && rm ${cpio_tfile}
 fi
diff --git a/security/selinux/netlabel.c b/security/selinux/netlabel.c
index 3f4b266..350794a 100644
--- a/security/selinux/netlabel.c
+++ b/security/selinux/netlabel.c
@@ -386,11 +386,12 @@
 	if (!S_ISSOCK(inode->i_mode) ||
 	    ((mask & (MAY_WRITE | MAY_APPEND)) == 0))
 		return 0;
-
 	sock = SOCKET_I(inode);
 	sk = sock->sk;
+	if (sk == NULL)
+		return 0;
 	sksec = sk->sk_security;
-	if (sksec->nlbl_state != NLBL_REQUIRE)
+	if (sksec == NULL || sksec->nlbl_state != NLBL_REQUIRE)
 		return 0;
 
 	local_bh_disable();
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index c8d9178..5e909e0 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -2095,6 +2095,8 @@
 	SND_PCI_QUIRK(0x1028, 0x20ac, "Dell Studio Desktop", 0x01),
 	/* including bogus ALC268 in slot#2 that conflicts with ALC888 */
 	SND_PCI_QUIRK(0x17c0, 0x4085, "Medion MD96630", 0x01),
+	/* conflict of ALC268 in slot#3 (digital I/O); a temporary fix */
+	SND_PCI_QUIRK(0x1179, 0xff00, "Toshiba laptop", 0x03),
 	{}
 };
 
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index a680be0..6c26afc 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -10557,6 +10557,7 @@
 	SND_PCI_QUIRK(0x103c, 0x1309, "HP xw4*00", ALC262_HP_BPC),
 	SND_PCI_QUIRK(0x103c, 0x130a, "HP xw6*00", ALC262_HP_BPC),
 	SND_PCI_QUIRK(0x103c, 0x130b, "HP xw8*00", ALC262_HP_BPC),
+	SND_PCI_QUIRK(0x103c, 0x170b, "HP xw*", ALC262_HP_BPC),
 	SND_PCI_QUIRK(0x103c, 0x2800, "HP D7000", ALC262_HP_BPC_D7000_WL),
 	SND_PCI_QUIRK(0x103c, 0x2801, "HP D7000", ALC262_HP_BPC_D7000_WF),
 	SND_PCI_QUIRK(0x103c, 0x2802, "HP D7000", ALC262_HP_BPC_D7000_WL),
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 8027edf..3bc4276 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -4989,7 +4989,7 @@
 	case STAC_DELL_M4_3:
 		spec->num_dmics = 1;
 		spec->num_smuxes = 0;
-		spec->num_dmuxes = 0;
+		spec->num_dmuxes = 1;
 		break;
 	default:
 		spec->num_dmics = STAC92HD71BXX_NUM_DMICS;
diff --git a/usr/Kconfig b/usr/Kconfig
index 86cecb5..43a3a0f 100644
--- a/usr/Kconfig
+++ b/usr/Kconfig
@@ -44,3 +44,92 @@
 	  owned by group root in the initial ramdisk image.
 
 	  If you are not sure, leave it set to "0".
+
+config RD_GZIP
+	bool "Initial ramdisk compressed using gzip"
+	default y
+	depends on BLK_DEV_INITRD=y
+	select DECOMPRESS_GZIP
+	help
+	  Support loading of a gzip encoded initial ramdisk or cpio buffer.
+	  If unsure, say Y.
+
+config RD_BZIP2
+	bool "Initial ramdisk compressed using bzip2"
+	default n
+	depends on BLK_DEV_INITRD=y
+	select DECOMPRESS_BZIP2
+	help
+	  Support loading of a bzip2 encoded initial ramdisk or cpio buffer
+	  If unsure, say N.
+
+config RD_LZMA
+	bool "Initial ramdisk compressed using lzma"
+	default n
+	depends on BLK_DEV_INITRD=y
+	select DECOMPRESS_LZMA
+	help
+	  Support loading of a lzma encoded initial ramdisk or cpio buffer
+	  If unsure, say N.
+
+choice
+	prompt "Built-in initramfs compression mode"
+	help
+	  This setting is only meaningful if the INITRAMFS_SOURCE is
+	  set. It decides by which algorithm the INITRAMFS_SOURCE will
+	  be compressed.
+	  Several compression algorithms are available, which differ
+	  in efficiency, compression and decompression speed.
+	  Compression speed is only relevant when building a kernel.
+	  Decompression speed is relevant at each boot.
+
+	  If you have any problems with bzip2 or lzma compressed
+	  initramfs, mail me (Alain Knaff) <alain@knaff.lu>.
+
+	  High compression options are mostly useful for users who
+	  are low on disk space (embedded systems), but for whom ram
+	  size matters less.
+
+	  If in doubt, select 'gzip'
+
+config INITRAMFS_COMPRESSION_NONE
+	bool "None"
+	help
+	  Do not compress the built-in initramfs at all. This may
+	  sound wasteful in space, but, you should be aware that the
+	  built-in initramfs will be compressed at a later stage
+	  anyways along with the rest of the kernel, on those
+	  architectures that support this.
+	  However, not compressing the initramfs may lead to slightly
+	  higher memory consumption during a short time at boot, while
+	  both the cpio image and the unpacked filesystem image will
+	  be present in memory simultaneously
+
+config INITRAMFS_COMPRESSION_GZIP
+	bool "Gzip"
+	depends on RD_GZIP
+	help
+	  The old and tried gzip compression. Its compression ratio is
+	  the poorest among the 3 choices; however its speed (both
+	  compression and decompression) is the fastest.
+
+config INITRAMFS_COMPRESSION_BZIP2
+	bool "Bzip2"
+	depends on RD_BZIP2
+	help
+	  Its compression ratio and speed is intermediate.
+	  Decompression speed is slowest among the three.  The initramfs
+	  size is about 10% smaller with bzip2, in comparison to gzip.
+	  Bzip2 uses a large amount of memory. For modern kernels you
+	  will need at least 8MB RAM or more for booting.
+
+config INITRAMFS_COMPRESSION_LZMA
+	bool "LZMA"
+	depends on RD_LZMA
+	help
+	  The most recent compression algorithm.
+	  Its ratio is best, decompression speed is between the other
+	  two. Compression is slowest.	The initramfs size is about 33%
+	  smaller with LZMA in comparison to gzip.
+
+endchoice
diff --git a/usr/Makefile b/usr/Makefile
index 201f27f..b84894b 100644
--- a/usr/Makefile
+++ b/usr/Makefile
@@ -6,13 +6,25 @@
 PHONY += klibcdirs
 
 
-# Generate builtin.o based on initramfs_data.o
-obj-$(CONFIG_BLK_DEV_INITRD) := initramfs_data.o
+# No compression
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_NONE)   =
 
-# initramfs_data.o contains the initramfs_data.cpio.gz image.
+# Gzip, but no bzip2
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_GZIP)   = .gz
+
+# Bzip2
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_BZIP2)  = .bz2
+
+# Lzma
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_LZMA)   = .lzma
+
+# Generate builtin.o based on initramfs_data.o
+obj-$(CONFIG_BLK_DEV_INITRD) := initramfs_data$(suffix_y).o
+
+# initramfs_data.o contains the compressed initramfs_data.cpio image.
 # The image is included using .incbin, a dependency which is not
 # tracked automatically.
-$(obj)/initramfs_data.o: $(obj)/initramfs_data.cpio.gz FORCE
+$(obj)/initramfs_data$(suffix_y).o: $(obj)/initramfs_data.cpio$(suffix_y) FORCE
 
 #####
 # Generate the initramfs cpio archive
@@ -25,28 +37,28 @@
         $(if $(CONFIG_INITRAMFS_ROOT_UID), -u $(CONFIG_INITRAMFS_ROOT_UID)) \
         $(if $(CONFIG_INITRAMFS_ROOT_GID), -g $(CONFIG_INITRAMFS_ROOT_GID))
 
-# .initramfs_data.cpio.gz.d is used to identify all files included
+# .initramfs_data.cpio.d is used to identify all files included
 # in initramfs and to detect if any files are added/removed.
 # Removed files are identified by directory timestamp being updated
 # The dependency list is generated by gen_initramfs.sh -l
-ifneq ($(wildcard $(obj)/.initramfs_data.cpio.gz.d),)
-	include $(obj)/.initramfs_data.cpio.gz.d
+ifneq ($(wildcard $(obj)/.initramfs_data.cpio.d),)
+	include $(obj)/.initramfs_data.cpio.d
 endif
 
 quiet_cmd_initfs = GEN     $@
       cmd_initfs = $(initramfs) -o $@ $(ramfs-args) $(ramfs-input)
 
-targets := initramfs_data.cpio.gz
+targets := initramfs_data.cpio.gz initramfs_data.cpio.bz2 initramfs_data.cpio.lzma initramfs_data.cpio
 # do not try to update files included in initramfs
 $(deps_initramfs): ;
 
 $(deps_initramfs): klibcdirs
-# We rebuild initramfs_data.cpio.gz if:
-# 1) Any included file is newer then initramfs_data.cpio.gz
+# We rebuild initramfs_data.cpio if:
+# 1) Any included file is newer then initramfs_data.cpio
 # 2) There are changes in which files are included (added or deleted)
-# 3) If gen_init_cpio are newer than initramfs_data.cpio.gz
+# 3) If gen_init_cpio are newer than initramfs_data.cpio
 # 4) arguments to gen_initramfs.sh changes
-$(obj)/initramfs_data.cpio.gz: $(obj)/gen_init_cpio $(deps_initramfs) klibcdirs
-	$(Q)$(initramfs) -l $(ramfs-input) > $(obj)/.initramfs_data.cpio.gz.d
+$(obj)/initramfs_data.cpio$(suffix_y): $(obj)/gen_init_cpio $(deps_initramfs) klibcdirs
+	$(Q)$(initramfs) -l $(ramfs-input) > $(obj)/.initramfs_data.cpio.d
 	$(call if_changed,initfs)
 
diff --git a/usr/initramfs_data.S b/usr/initramfs_data.S
index c2e1ad4..7c6973d 100644
--- a/usr/initramfs_data.S
+++ b/usr/initramfs_data.S
@@ -26,5 +26,5 @@
 */
 
 .section .init.ramfs,"a"
-.incbin "usr/initramfs_data.cpio.gz"
+.incbin "usr/initramfs_data.cpio"
 
diff --git a/usr/initramfs_data.bz2.S b/usr/initramfs_data.bz2.S
new file mode 100644
index 0000000..bc54d09
--- /dev/null
+++ b/usr/initramfs_data.bz2.S
@@ -0,0 +1,29 @@
+/*
+  initramfs_data includes the compressed binary that is the
+  filesystem used for early user space.
+  Note: Older versions of "as" (prior to binutils 2.11.90.0.23
+  released on 2001-07-14) dit not support .incbin.
+  If you are forced to use older binutils than that then the
+  following trick can be applied to create the resulting binary:
+
+
+  ld -m elf_i386  --format binary --oformat elf32-i386 -r \
+  -T initramfs_data.scr initramfs_data.cpio.gz -o initramfs_data.o
+   ld -m elf_i386  -r -o built-in.o initramfs_data.o
+
+  initramfs_data.scr looks like this:
+SECTIONS
+{
+       .init.ramfs : { *(.data) }
+}
+
+  The above example is for i386 - the parameters vary from architectures.
+  Eventually look up LDFLAGS_BLOB in an older version of the
+  arch/$(ARCH)/Makefile to see the flags used before .incbin was introduced.
+
+  Using .incbin has the advantage over ld that the correct flags are set
+  in the ELF header, as required by certain architectures.
+*/
+
+.section .init.ramfs,"a"
+.incbin "usr/initramfs_data.cpio.bz2"
diff --git a/usr/initramfs_data.gz.S b/usr/initramfs_data.gz.S
new file mode 100644
index 0000000..890c8dd
--- /dev/null
+++ b/usr/initramfs_data.gz.S
@@ -0,0 +1,29 @@
+/*
+  initramfs_data includes the compressed binary that is the
+  filesystem used for early user space.
+  Note: Older versions of "as" (prior to binutils 2.11.90.0.23
+  released on 2001-07-14) dit not support .incbin.
+  If you are forced to use older binutils than that then the
+  following trick can be applied to create the resulting binary:
+
+
+  ld -m elf_i386  --format binary --oformat elf32-i386 -r \
+  -T initramfs_data.scr initramfs_data.cpio.gz -o initramfs_data.o
+   ld -m elf_i386  -r -o built-in.o initramfs_data.o
+
+  initramfs_data.scr looks like this:
+SECTIONS
+{
+       .init.ramfs : { *(.data) }
+}
+
+  The above example is for i386 - the parameters vary from architectures.
+  Eventually look up LDFLAGS_BLOB in an older version of the
+  arch/$(ARCH)/Makefile to see the flags used before .incbin was introduced.
+
+  Using .incbin has the advantage over ld that the correct flags are set
+  in the ELF header, as required by certain architectures.
+*/
+
+.section .init.ramfs,"a"
+.incbin "usr/initramfs_data.cpio.gz"
diff --git a/usr/initramfs_data.lzma.S b/usr/initramfs_data.lzma.S
new file mode 100644
index 0000000..e11469e
--- /dev/null
+++ b/usr/initramfs_data.lzma.S
@@ -0,0 +1,29 @@
+/*
+  initramfs_data includes the compressed binary that is the
+  filesystem used for early user space.
+  Note: Older versions of "as" (prior to binutils 2.11.90.0.23
+  released on 2001-07-14) dit not support .incbin.
+  If you are forced to use older binutils than that then the
+  following trick can be applied to create the resulting binary:
+
+
+  ld -m elf_i386  --format binary --oformat elf32-i386 -r \
+  -T initramfs_data.scr initramfs_data.cpio.gz -o initramfs_data.o
+   ld -m elf_i386  -r -o built-in.o initramfs_data.o
+
+  initramfs_data.scr looks like this:
+SECTIONS
+{
+       .init.ramfs : { *(.data) }
+}
+
+  The above example is for i386 - the parameters vary from architectures.
+  Eventually look up LDFLAGS_BLOB in an older version of the
+  arch/$(ARCH)/Makefile to see the flags used before .incbin was introduced.
+
+  Using .incbin has the advantage over ld that the correct flags are set
+  in the ELF header, as required by certain architectures.
+*/
+
+.section .init.ramfs,"a"
+.incbin "usr/initramfs_data.cpio.lzma"