Merge branches 'sh/ftrace' and 'sh/stable-updates'
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index e2bdd7b..120bd31 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -10,6 +10,7 @@
 	select EMBEDDED
 	select HAVE_CLK
 	select HAVE_IDE
+	select HAVE_LMB
 	select HAVE_OPROFILE
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_IOREMAP_PROT if MMU
@@ -31,6 +32,9 @@
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FTRACE_MCOUNT_RECORD
 	select HAVE_DYNAMIC_FTRACE
+	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
+	select HAVE_FTRACE_SYSCALLS
+	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_ARCH_KGDB
 	select ARCH_HIBERNATION_POSSIBLE if MMU
 
diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index 39224b5..52a132c 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -123,4 +123,15 @@
 	bool "Debug: set SR.WATCH to enable hardware watchpoints and trace"
 	depends on SUPERH64
 
+config STACK_DEBUG
+	bool "Enable diagnostic checks of the kernel stack"
+	depends on FUNCTION_TRACER
+	select DEBUG_STACKOVERFLOW
+	default n
+	help
+	  This option allows checks to be performed on the kernel stack
+	  at runtime. Saying Y here will add overhead to every function
+	  call and will therefore incur a major performance hit. Most
+	  users should say N.
+
 endmenu
diff --git a/arch/sh/include/asm/ftrace.h b/arch/sh/include/asm/ftrace.h
index 8fea7d8..7e0bcc4 100644
--- a/arch/sh/include/asm/ftrace.h
+++ b/arch/sh/include/asm/ftrace.h
@@ -11,10 +11,13 @@
 #define MCOUNT_ADDR		((long)(mcount))
 
 #ifdef CONFIG_DYNAMIC_FTRACE
-#define CALLER_ADDR		((long)(ftrace_caller))
+#define CALL_ADDR		((long)(ftrace_call))
 #define STUB_ADDR		((long)(ftrace_stub))
+#define GRAPH_ADDR		((long)(ftrace_graph_call))
+#define CALLER_ADDR		((long)(ftrace_caller))
 
-#define MCOUNT_INSN_OFFSET	((STUB_ADDR - CALLER_ADDR) >> 1)
+#define MCOUNT_INSN_OFFSET	((STUB_ADDR - CALL_ADDR) - 4)
+#define GRAPH_INSN_OFFSET	((CALLER_ADDR - GRAPH_ADDR) - 4)
 
 struct dyn_arch_ftrace {
 	/* No extra data needed on sh */
diff --git a/arch/sh/include/asm/hwblk.h b/arch/sh/include/asm/hwblk.h
new file mode 100644
index 0000000..51a46f4
--- /dev/null
+++ b/arch/sh/include/asm/hwblk.h
@@ -0,0 +1,61 @@
+#ifndef __ASM_SH_HWBLK_H
+#define __ASM_SH_HWBLK_H
+
+#include <asm/clock.h>
+#include <asm/io.h>
+
+#define HWBLK_AREA_FLAG_PARENT (1 << 0) /* valid parent */
+
+#define HWBLK_AREA(_flags, _parent)		\
+{						\
+	.flags = _flags,			\
+	.parent = _parent,			\
+}
+
+struct hwblk_area {
+	unsigned long cnt;
+	unsigned char parent;
+	unsigned char flags;
+};
+
+#define HWBLK(_mstp, _bit, _area)		\
+{						\
+	.mstp = (void __iomem *)_mstp,		\
+	.bit = _bit,				\
+	.area = _area,				\
+}
+
+struct hwblk {
+	void __iomem *mstp;
+	unsigned char bit;
+	unsigned char area;
+	unsigned long cnt;
+};
+
+struct hwblk_info {
+	struct hwblk_area *areas;
+	int nr_areas;
+	struct hwblk *hwblks;
+	int nr_hwblks;
+};
+
+/* Should be defined by processor-specific code */
+int arch_hwblk_init(void);
+int arch_hwblk_sleep_mode(void);
+
+int hwblk_register(struct hwblk_info *info);
+int hwblk_init(void);
+
+/* allow clocks to enable and disable hardware blocks */
+#define SH_HWBLK_CLK(_name, _id, _parent, _hwblk, _flags)	\
+{							\
+	.name		= _name,			\
+	.id		= _id,				\
+	.parent		= _parent,			\
+	.arch_flags	= _hwblk,			\
+	.flags		= _flags,			\
+}
+
+int sh_hwblk_clk_register(struct clk *clks, int nr);
+
+#endif /* __ASM_SH_HWBLK_H */
diff --git a/arch/sh/include/asm/lmb.h b/arch/sh/include/asm/lmb.h
new file mode 100644
index 0000000..9b437f6
--- /dev/null
+++ b/arch/sh/include/asm/lmb.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_SH_LMB_H
+#define __ASM_SH_LMB_H
+
+#define LMB_REAL_LIMIT	0
+
+#endif /* __ASM_SH_LMB_H */
diff --git a/arch/sh/include/asm/suspend.h b/arch/sh/include/asm/suspend.h
index b1b9953..5c8ea28 100644
--- a/arch/sh/include/asm/suspend.h
+++ b/arch/sh/include/asm/suspend.h
@@ -10,6 +10,15 @@
 	struct pt_regs user_regs;
 	unsigned long bank1_regs[8];
 };
+
+void sh_mobile_call_standby(unsigned long mode);
+
+#ifdef CONFIG_CPU_IDLE
+void sh_mobile_setup_cpuidle(void);
+#else
+static inline void sh_mobile_setup_cpuidle(void) {}
+#endif
+
 #endif
 
 /* flags passed to assembly suspend code */
diff --git a/arch/sh/include/asm/syscall_32.h b/arch/sh/include/asm/syscall_32.h
index 6f83f2c..7d80df4 100644
--- a/arch/sh/include/asm/syscall_32.h
+++ b/arch/sh/include/asm/syscall_32.h
@@ -65,6 +65,7 @@
 	case 3: args[2] = regs->regs[6];
 	case 2: args[1] = regs->regs[5];
 	case 1:	args[0] = regs->regs[4];
+	case 0:
 		break;
 	default:
 		BUG();
diff --git a/arch/sh/include/asm/thread_info.h b/arch/sh/include/asm/thread_info.h
index d570ac2..5123bca 100644
--- a/arch/sh/include/asm/thread_info.h
+++ b/arch/sh/include/asm/thread_info.h
@@ -97,7 +97,7 @@
 
 extern struct thread_info *alloc_thread_info(struct task_struct *tsk);
 extern void free_thread_info(struct thread_info *ti);
- 
+
 #endif /* THREAD_SHIFT < PAGE_SHIFT */
 
 #endif /* __ASSEMBLY__ */
@@ -116,6 +116,7 @@
 #define TIF_SYSCALL_AUDIT	5	/* syscall auditing active */
 #define TIF_SECCOMP		6	/* secure computing */
 #define TIF_NOTIFY_RESUME	7	/* callback before returning to user */
+#define TIF_SYSCALL_FTRACE	8	/* for ftrace syscall instrumentation */
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_MEMDIE		18
@@ -129,25 +130,27 @@
 #define _TIF_SYSCALL_AUDIT	(1 << TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1 << TIF_SECCOMP)
 #define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
+#define _TIF_SYSCALL_FTRACE	(1 << TIF_SYSCALL_FTRACE)
 #define _TIF_USEDFPU		(1 << TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1 << TIF_POLLING_NRFLAG)
 #define _TIF_FREEZE		(1 << TIF_FREEZE)
 
 /*
- * _TIF_ALLWORK_MASK and _TIF_WORK_MASK need to fit within a byte, or we
+ * _TIF_ALLWORK_MASK and _TIF_WORK_MASK need to fit within 2 bytes, or we
  * blow the tst immediate size constraints and need to fix up
  * arch/sh/kernel/entry-common.S.
  */
 
 /* work to do in syscall trace */
 #define _TIF_WORK_SYSCALL_MASK	(_TIF_SYSCALL_TRACE | _TIF_SINGLESTEP | \
-				 _TIF_SYSCALL_AUDIT | _TIF_SECCOMP)
+				 _TIF_SYSCALL_AUDIT | _TIF_SECCOMP    | \
+				 _TIF_SYSCALL_FTRACE)
 
 /* work to do on any return to u-space */
 #define _TIF_ALLWORK_MASK	(_TIF_SYSCALL_TRACE | _TIF_SIGPENDING      | \
 				 _TIF_NEED_RESCHED  | _TIF_SYSCALL_AUDIT   | \
 				 _TIF_SINGLESTEP    | _TIF_RESTORE_SIGMASK | \
-				 _TIF_NOTIFY_RESUME)
+				 _TIF_NOTIFY_RESUME | _TIF_SYSCALL_FTRACE)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK		(_TIF_ALLWORK_MASK & ~(_TIF_SYSCALL_TRACE | \
diff --git a/arch/sh/include/cpu-sh4/cpu/sh7722.h b/arch/sh/include/cpu-sh4/cpu/sh7722.h
index 738ea43..4856040 100644
--- a/arch/sh/include/cpu-sh4/cpu/sh7722.h
+++ b/arch/sh/include/cpu-sh4/cpu/sh7722.h
@@ -221,4 +221,18 @@
 	GPIO_FN_KEYOUT3, GPIO_FN_KEYOUT4_IN6, GPIO_FN_KEYOUT5_IN5,
 };
 
+enum {
+	HWBLK_UNKNOWN = 0,
+	HWBLK_TLB, HWBLK_IC, HWBLK_OC, HWBLK_URAM, HWBLK_XYMEM,
+	HWBLK_INTC, HWBLK_DMAC, HWBLK_SHYWAY, HWBLK_HUDI,
+	HWBLK_UBC, HWBLK_TMU, HWBLK_CMT, HWBLK_RWDT, HWBLK_FLCTL,
+	HWBLK_SCIF0, HWBLK_SCIF1, HWBLK_SCIF2, HWBLK_SIO,
+	HWBLK_SIOF0, HWBLK_SIOF1, HWBLK_IIC, HWBLK_RTC,
+	HWBLK_TPU, HWBLK_IRDA, HWBLK_SDHI, HWBLK_SIM, HWBLK_KEYSC,
+	HWBLK_TSIF, HWBLK_USBF, HWBLK_2DG, HWBLK_SIU, HWBLK_VOU,
+	HWBLK_JPU, HWBLK_BEU, HWBLK_CEU, HWBLK_VEU, HWBLK_VPU,
+	HWBLK_LCDC,
+	HWBLK_NR,
+};
+
 #endif /* __ASM_SH7722_H__ */
diff --git a/arch/sh/kernel/Makefile_32 b/arch/sh/kernel/Makefile_32
index 9411e3e..94ed99b 100644
--- a/arch/sh/kernel/Makefile_32
+++ b/arch/sh/kernel/Makefile_32
@@ -29,6 +29,8 @@
 obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_GENERIC_GPIO)	+= gpio.o
 obj-$(CONFIG_DYNAMIC_FTRACE)	+= ftrace.o
+obj-$(CONFIG_FTRACE_SYSCALLS)	+= ftrace.o
+obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o
 obj-$(CONFIG_DUMP_CODE)		+= disassemble.o
 obj-$(CONFIG_HIBERNATION)	+= swsusp.o
 
diff --git a/arch/sh/kernel/asm-offsets.c b/arch/sh/kernel/asm-offsets.c
index 99aceb28..d218e80 100644
--- a/arch/sh/kernel/asm-offsets.c
+++ b/arch/sh/kernel/asm-offsets.c
@@ -26,6 +26,7 @@
 	DEFINE(TI_CPU,		offsetof(struct thread_info, cpu));
 	DEFINE(TI_PRE_COUNT,	offsetof(struct thread_info, preempt_count));
 	DEFINE(TI_RESTART_BLOCK,offsetof(struct thread_info, restart_block));
+	DEFINE(TI_SIZE,		sizeof(struct thread_info));
 
 #ifdef CONFIG_HIBERNATION
 	DEFINE(PBE_ADDRESS, offsetof(struct pbe, address));
diff --git a/arch/sh/kernel/cpu/Makefile b/arch/sh/kernel/cpu/Makefile
index eecad7c..3d6b931 100644
--- a/arch/sh/kernel/cpu/Makefile
+++ b/arch/sh/kernel/cpu/Makefile
@@ -19,4 +19,4 @@
 obj-$(CONFIG_SH_ADC)		+= adc.o
 obj-$(CONFIG_SH_CLK_CPG)	+= clock-cpg.o
 
-obj-y	+= irq/ init.o clock.o
+obj-y	+= irq/ init.o clock.o hwblk.o
diff --git a/arch/sh/kernel/cpu/hwblk.c b/arch/sh/kernel/cpu/hwblk.c
new file mode 100644
index 0000000..7c3a73d
--- /dev/null
+++ b/arch/sh/kernel/cpu/hwblk.c
@@ -0,0 +1,130 @@
+#include <linux/clk.h>
+#include <linux/compiler.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/spinlock.h>
+#include <asm/suspend.h>
+#include <asm/hwblk.h>
+#include <asm/clock.h>
+
+static DEFINE_SPINLOCK(hwblk_lock);
+
+static void hwblk_area_inc(struct hwblk_info *info, int area)
+{
+	struct hwblk_area *hap = info->areas + area;
+
+	hap->cnt++;
+	if (hap->cnt == 1)
+		if (hap->flags & HWBLK_AREA_FLAG_PARENT)
+			hwblk_area_inc(info, hap->parent);
+}
+
+static void hwblk_area_dec(struct hwblk_info *info, int area)
+{
+	struct hwblk_area *hap = info->areas + area;
+
+	if (hap->cnt == 1)
+		if (hap->flags & HWBLK_AREA_FLAG_PARENT)
+			hwblk_area_dec(info, hap->parent);
+	hap->cnt--;
+}
+
+static void hwblk_enable(struct hwblk_info *info, int hwblk)
+{
+	struct hwblk *hp = info->hwblks + hwblk;
+	unsigned long tmp;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hwblk_lock, flags);
+
+	hp->cnt++;
+	if (hp->cnt == 1) {
+		hwblk_area_inc(info, hp->area);
+
+		tmp = __raw_readl(hp->mstp);
+		tmp &= ~(1 << hp->bit);
+		__raw_writel(tmp, hp->mstp);
+	}
+
+	spin_unlock_irqrestore(&hwblk_lock, flags);
+}
+
+static void hwblk_disable(struct hwblk_info *info, int hwblk)
+{
+	struct hwblk *hp = info->hwblks + hwblk;
+	unsigned long tmp;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hwblk_lock, flags);
+
+	if (hp->cnt == 1) {
+		hwblk_area_dec(info, hp->area);
+
+		tmp = __raw_readl(hp->mstp);
+		tmp |= 1 << hp->bit;
+		__raw_writel(tmp, hp->mstp);
+	}
+	hp->cnt--;
+
+	spin_unlock_irqrestore(&hwblk_lock, flags);
+}
+
+static struct hwblk_info *hwblk_info;
+
+int __init hwblk_register(struct hwblk_info *info)
+{
+	hwblk_info = info;
+	return 0;
+}
+
+int __init __weak arch_hwblk_init(void)
+{
+	return 0;
+}
+
+int __weak arch_hwblk_sleep_mode(void)
+{
+	return SUSP_SH_SLEEP;
+}
+
+int __init hwblk_init(void)
+{
+	return arch_hwblk_init();
+}
+
+/* allow clocks to enable and disable hardware blocks */
+static int sh_hwblk_clk_enable(struct clk *clk)
+{
+	if (!hwblk_info)
+		return -ENOENT;
+
+	hwblk_enable(hwblk_info, clk->arch_flags);
+	return 0;
+}
+
+static void sh_hwblk_clk_disable(struct clk *clk)
+{
+	if (hwblk_info)
+		hwblk_disable(hwblk_info, clk->arch_flags);
+}
+
+static struct clk_ops sh_hwblk_clk_ops = {
+	.enable		= sh_hwblk_clk_enable,
+	.disable	= sh_hwblk_clk_disable,
+	.recalc		= followparent_recalc,
+};
+
+int __init sh_hwblk_clk_register(struct clk *clks, int nr)
+{
+	struct clk *clkp;
+	int ret = 0;
+	int k;
+
+	for (k = 0; !ret && (k < nr); k++) {
+		clkp = clks + k;
+		clkp->ops = &sh_hwblk_clk_ops;
+		ret |= clk_register(clkp);
+	}
+
+	return ret;
+}
diff --git a/arch/sh/kernel/cpu/sh4a/Makefile b/arch/sh/kernel/cpu/sh4a/Makefile
index ebdd391..3cafda6 100644
--- a/arch/sh/kernel/cpu/sh4a/Makefile
+++ b/arch/sh/kernel/cpu/sh4a/Makefile
@@ -25,7 +25,7 @@
 clock-$(CONFIG_CPU_SUBTYPE_SH7785)	:= clock-sh7785.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7786)	:= clock-sh7786.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7343)	:= clock-sh7343.o
-clock-$(CONFIG_CPU_SUBTYPE_SH7722)	:= clock-sh7722.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7722)	:= clock-sh7722.o hwblk-sh7722.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7723)	:= clock-sh7723.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7724)	:= clock-sh7724.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7366)	:= clock-sh7366.o
diff --git a/arch/sh/kernel/cpu/sh4a/clock-sh7722.c b/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
index 40f8593..1fa9e1d 100644
--- a/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
+++ b/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
@@ -22,6 +22,8 @@
 #include <linux/kernel.h>
 #include <linux/io.h>
 #include <asm/clock.h>
+#include <asm/hwblk.h>
+#include <cpu/sh7722.h>
 
 /* SH7722 registers */
 #define FRQCR		0xa4150000
@@ -140,35 +142,37 @@
 	SH_CLK_DIV6("video_clk", &pll_clk, VCLKCR, 0),
 };
 
-#define MSTP(_str, _parent, _reg, _bit, _flags) \
-  SH_CLK_MSTP32(_str, -1, _parent, _reg, _bit, _flags)
+#define R_CLK &r_clk
+#define P_CLK &div4_clks[DIV4_P]
+#define B_CLK &div4_clks[DIV4_B]
+#define U_CLK &div4_clks[DIV4_U]
 
 static struct clk mstp_clks[] = {
-	MSTP("uram0", &div4_clks[DIV4_U], MSTPCR0, 28, CLK_ENABLE_ON_INIT),
-	MSTP("xymem0", &div4_clks[DIV4_B], MSTPCR0, 26, CLK_ENABLE_ON_INIT),
-	MSTP("tmu0", &div4_clks[DIV4_P], MSTPCR0, 15, 0),
-	MSTP("cmt0", &r_clk, MSTPCR0, 14, 0),
-	MSTP("rwdt0", &r_clk, MSTPCR0, 13, 0),
-	MSTP("flctl0", &div4_clks[DIV4_P], MSTPCR0, 10, 0),
-	MSTP("scif0", &div4_clks[DIV4_P], MSTPCR0, 7, 0),
-	MSTP("scif1", &div4_clks[DIV4_P], MSTPCR0, 6, 0),
-	MSTP("scif2", &div4_clks[DIV4_P], MSTPCR0, 5, 0),
+	SH_HWBLK_CLK("uram0", -1, U_CLK, HWBLK_URAM, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("xymem0", -1, B_CLK, HWBLK_XYMEM, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("tmu0", -1, P_CLK, HWBLK_TMU, 0),
+	SH_HWBLK_CLK("cmt0", -1, R_CLK, HWBLK_CMT, 0),
+	SH_HWBLK_CLK("rwdt0", -1, R_CLK, HWBLK_RWDT, 0),
+	SH_HWBLK_CLK("flctl0", -1, P_CLK, HWBLK_FLCTL, 0),
+	SH_HWBLK_CLK("scif0", -1, P_CLK, HWBLK_SCIF0, 0),
+	SH_HWBLK_CLK("scif1", -1, P_CLK, HWBLK_SCIF1, 0),
+	SH_HWBLK_CLK("scif2", -1, P_CLK, HWBLK_SCIF2, 0),
 
-	MSTP("i2c0", &div4_clks[DIV4_P], MSTPCR1, 9, 0),
-	MSTP("rtc0", &r_clk, MSTPCR1, 8, 0),
+	SH_HWBLK_CLK("i2c0", -1, P_CLK, HWBLK_IIC, 0),
+	SH_HWBLK_CLK("rtc0", -1, R_CLK, HWBLK_RTC, 0),
 
-	MSTP("sdhi0", &div4_clks[DIV4_P], MSTPCR2, 18, 0),
-	MSTP("keysc0", &r_clk, MSTPCR2, 14, 0),
-	MSTP("usbf0", &div4_clks[DIV4_P], MSTPCR2, 11, 0),
-	MSTP("2dg0", &div4_clks[DIV4_B], MSTPCR2, 9, 0),
-	MSTP("siu0", &div4_clks[DIV4_B], MSTPCR2, 8, 0),
-	MSTP("vou0", &div4_clks[DIV4_B], MSTPCR2, 5, 0),
-	MSTP("jpu0", &div4_clks[DIV4_B], MSTPCR2, 6, CLK_ENABLE_ON_INIT),
-	MSTP("beu0", &div4_clks[DIV4_B], MSTPCR2, 4, 0),
-	MSTP("ceu0", &div4_clks[DIV4_B], MSTPCR2, 3, 0),
-	MSTP("veu0", &div4_clks[DIV4_B], MSTPCR2, 2, CLK_ENABLE_ON_INIT),
-	MSTP("vpu0", &div4_clks[DIV4_B], MSTPCR2, 1, CLK_ENABLE_ON_INIT),
-	MSTP("lcdc0", &div4_clks[DIV4_B], MSTPCR2, 0, 0),
+	SH_HWBLK_CLK("sdhi0", -1, P_CLK, HWBLK_SDHI, 0),
+	SH_HWBLK_CLK("keysc0", -1, R_CLK, HWBLK_KEYSC, 0),
+	SH_HWBLK_CLK("usbf0", -1, P_CLK, HWBLK_USBF, 0),
+	SH_HWBLK_CLK("2dg0", -1, B_CLK, HWBLK_2DG, 0),
+	SH_HWBLK_CLK("siu0", -1, B_CLK, HWBLK_SIU, 0),
+	SH_HWBLK_CLK("vou0", -1, B_CLK, HWBLK_VOU, 0),
+	SH_HWBLK_CLK("jpu0", -1, B_CLK, HWBLK_JPU, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("beu0", -1, B_CLK, HWBLK_BEU, 0),
+	SH_HWBLK_CLK("ceu0", -1, B_CLK, HWBLK_CEU, 0),
+	SH_HWBLK_CLK("veu0", -1, B_CLK, HWBLK_VEU, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("vpu0", -1, B_CLK, HWBLK_VPU, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("lcdc0", -1, P_CLK, HWBLK_LCDC, 0),
 };
 
 int __init arch_clk_init(void)
@@ -191,7 +195,7 @@
 		ret = sh_clk_div6_register(div6_clks, ARRAY_SIZE(div6_clks));
 
 	if (!ret)
-		ret = sh_clk_mstp32_register(mstp_clks, ARRAY_SIZE(mstp_clks));
+		ret = sh_hwblk_clk_register(mstp_clks, ARRAY_SIZE(mstp_clks));
 
 	return ret;
 }
diff --git a/arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c b/arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c
new file mode 100644
index 0000000..00a1c02
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c
@@ -0,0 +1,106 @@
+/*
+ * arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c
+ *
+ * SH7722 hardware block support
+ *
+ * Copyright (C) 2009 Magnus Damm
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <asm/suspend.h>
+#include <asm/hwblk.h>
+#include <cpu/sh7722.h>
+
+/* SH7722 registers */
+#define MSTPCR0		0xa4150030
+#define MSTPCR1		0xa4150034
+#define MSTPCR2		0xa4150038
+
+/* SH7722 Power Domains */
+enum { CORE_AREA, SUB_AREA, CORE_AREA_BM };
+static struct hwblk_area sh7722_hwblk_area[] = {
+	[CORE_AREA] = HWBLK_AREA(0, 0),
+	[CORE_AREA_BM] = HWBLK_AREA(HWBLK_AREA_FLAG_PARENT, CORE_AREA),
+	[SUB_AREA] = HWBLK_AREA(0, 0),
+};
+
+/* Table mapping HWBLK to Module Stop Bit and Power Domain */
+static struct hwblk sh7722_hwblk[HWBLK_NR] = {
+	[HWBLK_TLB] = HWBLK(MSTPCR0, 31, CORE_AREA),
+	[HWBLK_IC] = HWBLK(MSTPCR0, 30, CORE_AREA),
+	[HWBLK_OC] = HWBLK(MSTPCR0, 29, CORE_AREA),
+	[HWBLK_URAM] = HWBLK(MSTPCR0, 28, CORE_AREA),
+	[HWBLK_XYMEM] = HWBLK(MSTPCR0, 26, CORE_AREA),
+	[HWBLK_INTC] = HWBLK(MSTPCR0, 22, CORE_AREA),
+	[HWBLK_DMAC] = HWBLK(MSTPCR0, 21, CORE_AREA_BM),
+	[HWBLK_SHYWAY] = HWBLK(MSTPCR0, 20, CORE_AREA),
+	[HWBLK_HUDI] = HWBLK(MSTPCR0, 19, CORE_AREA),
+	[HWBLK_UBC] = HWBLK(MSTPCR0, 17, CORE_AREA),
+	[HWBLK_TMU] = HWBLK(MSTPCR0, 15, CORE_AREA),
+	[HWBLK_CMT] = HWBLK(MSTPCR0, 14, SUB_AREA),
+	[HWBLK_RWDT] = HWBLK(MSTPCR0, 13, SUB_AREA),
+	[HWBLK_FLCTL] = HWBLK(MSTPCR0, 10, CORE_AREA),
+	[HWBLK_SCIF0] = HWBLK(MSTPCR0, 7, CORE_AREA),
+	[HWBLK_SCIF1] = HWBLK(MSTPCR0, 6, CORE_AREA),
+	[HWBLK_SCIF2] = HWBLK(MSTPCR0, 5, CORE_AREA),
+	[HWBLK_SIO] = HWBLK(MSTPCR0, 3, CORE_AREA),
+	[HWBLK_SIOF0] = HWBLK(MSTPCR0, 2, CORE_AREA),
+	[HWBLK_SIOF1] = HWBLK(MSTPCR0, 1, CORE_AREA),
+
+	[HWBLK_IIC] = HWBLK(MSTPCR1, 9, CORE_AREA),
+	[HWBLK_RTC] = HWBLK(MSTPCR1, 8, SUB_AREA),
+
+	[HWBLK_TPU] = HWBLK(MSTPCR2, 25, CORE_AREA),
+	[HWBLK_IRDA] = HWBLK(MSTPCR2, 24, CORE_AREA),
+	[HWBLK_SDHI] = HWBLK(MSTPCR2, 18, CORE_AREA),
+	[HWBLK_SIM] = HWBLK(MSTPCR2, 16, CORE_AREA),
+	[HWBLK_KEYSC] = HWBLK(MSTPCR2, 14, SUB_AREA),
+	[HWBLK_TSIF] = HWBLK(MSTPCR2, 13, SUB_AREA),
+	[HWBLK_USBF] = HWBLK(MSTPCR2, 11, CORE_AREA),
+	[HWBLK_2DG] = HWBLK(MSTPCR2, 9, CORE_AREA_BM),
+	[HWBLK_SIU] = HWBLK(MSTPCR2, 8, CORE_AREA),
+	[HWBLK_JPU] = HWBLK(MSTPCR2, 6, CORE_AREA_BM),
+	[HWBLK_VOU] = HWBLK(MSTPCR2, 5, CORE_AREA_BM),
+	[HWBLK_BEU] = HWBLK(MSTPCR2, 4, CORE_AREA_BM),
+	[HWBLK_CEU] = HWBLK(MSTPCR2, 3, CORE_AREA_BM),
+	[HWBLK_VEU] = HWBLK(MSTPCR2, 2, CORE_AREA_BM),
+	[HWBLK_VPU] = HWBLK(MSTPCR2, 1, CORE_AREA_BM),
+	[HWBLK_LCDC] = HWBLK(MSTPCR2, 0, CORE_AREA_BM),
+};
+
+static struct hwblk_info sh7722_hwblk_info = {
+	.areas = sh7722_hwblk_area,
+	.nr_areas = ARRAY_SIZE(sh7722_hwblk_area),
+	.hwblks = sh7722_hwblk,
+	.nr_hwblks = ARRAY_SIZE(sh7722_hwblk),
+};
+
+int arch_hwblk_sleep_mode(void)
+{
+	if (!sh7722_hwblk_area[CORE_AREA].cnt)
+		return SUSP_SH_STANDBY | SUSP_SH_SF;
+
+	if (!sh7722_hwblk_area[CORE_AREA_BM].cnt)
+		return SUSP_SH_SLEEP | SUSP_SH_SF;
+
+	return SUSP_SH_SLEEP;
+}
+
+int __init arch_hwblk_init(void)
+{
+	return hwblk_register(&sh7722_hwblk_info);
+}
diff --git a/arch/sh/kernel/cpu/shmobile/Makefile b/arch/sh/kernel/cpu/shmobile/Makefile
index 08bfa7c..e8a5111 100644
--- a/arch/sh/kernel/cpu/shmobile/Makefile
+++ b/arch/sh/kernel/cpu/shmobile/Makefile
@@ -4,3 +4,4 @@
 
 # Power Management & Sleep mode
 obj-$(CONFIG_PM)	+= pm.o sleep.o
+obj-$(CONFIG_CPU_IDLE)	+= cpuidle.o
diff --git a/arch/sh/kernel/cpu/shmobile/cpuidle.c b/arch/sh/kernel/cpu/shmobile/cpuidle.c
new file mode 100644
index 0000000..4afdd97
--- /dev/null
+++ b/arch/sh/kernel/cpu/shmobile/cpuidle.c
@@ -0,0 +1,102 @@
+/*
+ * arch/sh/kernel/cpu/shmobile/cpuidle.c
+ *
+ * Cpuidle support code for SuperH Mobile
+ *
+ *  Copyright (C) 2009 Magnus Damm
+ *
+ * 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.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <linux/suspend.h>
+#include <linux/cpuidle.h>
+#include <asm/suspend.h>
+#include <asm/uaccess.h>
+#include <asm/hwblk.h>
+
+static unsigned long cpuidle_mode[] = {
+	SUSP_SH_SLEEP, /* regular sleep mode */
+	SUSP_SH_SLEEP | SUSP_SH_SF, /* sleep mode + self refresh */
+};
+
+static int cpuidle_sleep_enter(struct cpuidle_device *dev,
+			       struct cpuidle_state *state)
+{
+	unsigned long allowed_mode = arch_hwblk_sleep_mode();
+	ktime_t before, after;
+	int requested_state = state - &dev->states[0];
+	int allowed_state;
+	int k;
+
+	/* convert allowed mode to allowed state */
+	for (k = ARRAY_SIZE(cpuidle_mode) - 1; k > 0; k--)
+		if (cpuidle_mode[k] == allowed_mode)
+			break;
+
+	allowed_state = k;
+
+	/* take the following into account for sleep mode selection:
+	 * - allowed_state: best mode allowed by hardware (clock deps)
+	 * - requested_state: best mode allowed by software (latencies)
+	 */
+	k = min_t(int, allowed_state, requested_state);
+
+	dev->last_state = &dev->states[k];
+	before = ktime_get();
+	sh_mobile_call_standby(cpuidle_mode[k]);
+	after = ktime_get();
+	return ktime_to_ns(ktime_sub(after, before)) >> 10;
+}
+
+static struct cpuidle_device cpuidle_dev;
+static struct cpuidle_driver cpuidle_driver = {
+	.name =		"sh_idle",
+	.owner =	THIS_MODULE,
+};
+
+void sh_mobile_setup_cpuidle(void)
+{
+	struct cpuidle_device *dev = &cpuidle_dev;
+	struct cpuidle_state *state;
+	int i;
+
+	cpuidle_register_driver(&cpuidle_driver);
+
+	for (i = 0; i < CPUIDLE_STATE_MAX; i++) {
+		dev->states[i].name[0] = '\0';
+		dev->states[i].desc[0] = '\0';
+	}
+
+	i = CPUIDLE_DRIVER_STATE_START;
+
+	state = &dev->states[i++];
+	snprintf(state->name, CPUIDLE_NAME_LEN, "C0");
+	strncpy(state->desc, "SuperH Sleep Mode", CPUIDLE_DESC_LEN);
+	state->exit_latency = 1;
+	state->target_residency = 1 * 2;
+	state->power_usage = 3;
+	state->flags = 0;
+	state->flags |= CPUIDLE_FLAG_SHALLOW;
+	state->flags |= CPUIDLE_FLAG_TIME_VALID;
+	state->enter = cpuidle_sleep_enter;
+
+	dev->safe_state = state;
+
+	state = &dev->states[i++];
+	snprintf(state->name, CPUIDLE_NAME_LEN, "C1");
+	strncpy(state->desc, "SuperH Sleep Mode [SF]", CPUIDLE_DESC_LEN);
+	state->exit_latency = 100;
+	state->target_residency = 1 * 2;
+	state->power_usage = 1;
+	state->flags = 0;
+	state->flags |= CPUIDLE_FLAG_TIME_VALID;
+	state->enter = cpuidle_sleep_enter;
+
+	dev->state_count = i;
+
+	cpuidle_register_device(dev);
+}
diff --git a/arch/sh/kernel/cpu/shmobile/pm.c b/arch/sh/kernel/cpu/shmobile/pm.c
index 8c067ad..de078d2 100644
--- a/arch/sh/kernel/cpu/shmobile/pm.c
+++ b/arch/sh/kernel/cpu/shmobile/pm.c
@@ -1,5 +1,5 @@
 /*
- * arch/sh/kernel/cpu/sh4a/pm-sh_mobile.c
+ * arch/sh/kernel/cpu/shmobile/pm.c
  *
  * Power management support code for SuperH Mobile
  *
@@ -32,20 +32,17 @@
  *
  * R-standby mode is unsupported, but will be added in the future
  * U-standby mode is low priority since it needs bootloader hacks
- *
- * All modes should be tied in with cpuidle. But before that can
- * happen we need to keep track of enabled hardware blocks so we
- * can avoid entering sleep modes that stop clocks to hardware
- * blocks that are in use even though the cpu core is idle.
  */
 
+#define ILRAM_BASE 0xe5200000
+
 extern const unsigned char sh_mobile_standby[];
 extern const unsigned int sh_mobile_standby_size;
 
-static void sh_mobile_call_standby(unsigned long mode)
+void sh_mobile_call_standby(unsigned long mode)
 {
 	extern void *vbr_base;
-	void *onchip_mem = (void *)0xe5200000; /* ILRAM */
+	void *onchip_mem = (void *)ILRAM_BASE;
 	void (*standby_onchip_mem)(unsigned long) = onchip_mem;
 
 	/* Note: Wake up from sleep may generate exceptions!
@@ -55,11 +52,6 @@
 	if (mode & SUSP_SH_SF)
 		asm volatile("ldc %0, vbr" : : "r" (onchip_mem) : "memory");
 
-	/* Copy the assembly snippet to the otherwise ununsed ILRAM */
-	memcpy(onchip_mem, sh_mobile_standby, sh_mobile_standby_size);
-	wmb();
-	ctrl_barrier();
-
 	/* Let assembly snippet in on-chip memory handle the rest */
 	standby_onchip_mem(mode);
 
@@ -85,7 +77,15 @@
 
 static int __init sh_pm_init(void)
 {
+	void *onchip_mem = (void *)ILRAM_BASE;
+
+	/* Copy the assembly snippet to the otherwise ununsed ILRAM */
+	memcpy(onchip_mem, sh_mobile_standby, sh_mobile_standby_size);
+	wmb();
+	ctrl_barrier();
+
 	suspend_set_ops(&sh_pm_ops);
+	sh_mobile_setup_cpuidle();
 	return 0;
 }
 
diff --git a/arch/sh/kernel/entry-common.S b/arch/sh/kernel/entry-common.S
index d62359c..d621756 100644
--- a/arch/sh/kernel/entry-common.S
+++ b/arch/sh/kernel/entry-common.S
@@ -131,7 +131,7 @@
 	 nop
 #endif
 	mov.l	@(TI_FLAGS,r8), r0		! current_thread_info->flags
-	tst	#_TIF_WORK_MASK, r0
+	tst	#(_TIF_WORK_MASK & 0xff), r0
 	bt/s	__restore_all
 	 tst	#_TIF_NEED_RESCHED, r0
 
@@ -163,7 +163,7 @@
 #endif
 	!
 	mov.l	@(TI_FLAGS,r8), r0		! current_thread_info->flags
-	tst	#_TIF_WORK_MASK, r0
+	tst	#(_TIF_WORK_MASK & 0xff), r0
 	bt	__restore_all
 	bra	work_pending
 	 tst	#_TIF_NEED_RESCHED, r0
@@ -181,7 +181,7 @@
 syscall_exit_work:
 	! r0: current_thread_info->flags
 	! r8: current_thread_info
-	tst	#_TIF_WORK_SYSCALL_MASK, r0
+	tst	#(_TIF_WORK_SYSCALL_MASK & 0xff), r0
 	bt/s	work_pending
 	 tst	#_TIF_NEED_RESCHED, r0
 #ifdef CONFIG_TRACE_IRQFLAGS
@@ -331,8 +331,12 @@
 	!
 	get_current_thread_info r8, r10
 	mov.l	@(TI_FLAGS,r8), r8
-	mov	#_TIF_WORK_SYSCALL_MASK, r10
+	mov	#(_TIF_WORK_SYSCALL_MASK & 0xff), r10
+	mov	#(_TIF_WORK_SYSCALL_MASK >> 8), r9
 	tst	r10, r8
+	shll8	r9
+	bf	syscall_trace_entry
+	tst	r9, r8
 	bf	syscall_trace_entry
 	!
 	mov.l	2f, r8			! Number of syscalls
@@ -359,7 +363,11 @@
 	!
 	get_current_thread_info r8, r0
 	mov.l	@(TI_FLAGS,r8), r0		! current_thread_info->flags
-	tst	#_TIF_ALLWORK_MASK, r0
+	tst	#(_TIF_ALLWORK_MASK & 0xff), r0
+	mov	#(_TIF_ALLWORK_MASK >> 8), r1
+	bf	syscall_exit_work
+	shlr8	r0
+	tst	r0, r1
 	bf	syscall_exit_work
 	bra	__restore_all
 	 nop
diff --git a/arch/sh/kernel/ftrace.c b/arch/sh/kernel/ftrace.c
index 066f37d..6647dfc 100644
--- a/arch/sh/kernel/ftrace.c
+++ b/arch/sh/kernel/ftrace.c
@@ -16,9 +16,13 @@
 #include <linux/string.h>
 #include <linux/init.h>
 #include <linux/io.h>
+#include <linux/kernel.h>
 #include <asm/ftrace.h>
 #include <asm/cacheflush.h>
+#include <asm/unistd.h>
+#include <trace/syscall.h>
 
+#ifdef CONFIG_DYNAMIC_FTRACE
 static unsigned char ftrace_replaced_code[MCOUNT_INSN_SIZE];
 
 static unsigned char ftrace_nop[4];
@@ -131,3 +135,189 @@
 
 	return 0;
 }
+#endif /* CONFIG_DYNAMIC_FTRACE */
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+#ifdef CONFIG_DYNAMIC_FTRACE
+extern void ftrace_graph_call(void);
+
+static int ftrace_mod(unsigned long ip, unsigned long old_addr,
+		      unsigned long new_addr)
+{
+	unsigned char code[MCOUNT_INSN_SIZE];
+
+	if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE))
+		return -EFAULT;
+
+	if (old_addr != __raw_readl((unsigned long *)code))
+		return -EINVAL;
+
+	__raw_writel(new_addr, ip);
+	return 0;
+}
+
+int ftrace_enable_ftrace_graph_caller(void)
+{
+	unsigned long ip, old_addr, new_addr;
+
+	ip = (unsigned long)(&ftrace_graph_call) + GRAPH_INSN_OFFSET;
+	old_addr = (unsigned long)(&skip_trace);
+	new_addr = (unsigned long)(&ftrace_graph_caller);
+
+	return ftrace_mod(ip, old_addr, new_addr);
+}
+
+int ftrace_disable_ftrace_graph_caller(void)
+{
+	unsigned long ip, old_addr, new_addr;
+
+	ip = (unsigned long)(&ftrace_graph_call) + GRAPH_INSN_OFFSET;
+	old_addr = (unsigned long)(&ftrace_graph_caller);
+	new_addr = (unsigned long)(&skip_trace);
+
+	return ftrace_mod(ip, old_addr, new_addr);
+}
+#endif /* CONFIG_DYNAMIC_FTRACE */
+
+/*
+ * Hook the return address and push it in the stack of return addrs
+ * in the current thread info.
+ *
+ * This is the main routine for the function graph tracer. The function
+ * graph tracer essentially works like this:
+ *
+ * parent is the stack address containing self_addr's return address.
+ * We pull the real return address out of parent and store it in
+ * current's ret_stack. Then, we replace the return address on the stack
+ * with the address of return_to_handler. self_addr is the function that
+ * called mcount.
+ *
+ * When self_addr returns, it will jump to return_to_handler which calls
+ * ftrace_return_to_handler. ftrace_return_to_handler will pull the real
+ * return address off of current's ret_stack and jump to it.
+ */
+void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
+{
+	unsigned long old;
+	int faulted, err;
+	struct ftrace_graph_ent trace;
+	unsigned long return_hooker = (unsigned long)&return_to_handler;
+
+	if (unlikely(atomic_read(&current->tracing_graph_pause)))
+		return;
+
+	/*
+	 * Protect against fault, even if it shouldn't
+	 * happen. This tool is too much intrusive to
+	 * ignore such a protection.
+	 */
+	__asm__ __volatile__(
+		"1:						\n\t"
+		"mov.l		@%2, %0				\n\t"
+		"2:						\n\t"
+		"mov.l		%3, @%2				\n\t"
+		"mov		#0, %1				\n\t"
+		"3:						\n\t"
+		".section .fixup, \"ax\"			\n\t"
+		"4:						\n\t"
+		"mov.l		5f, %0				\n\t"
+		"jmp		@%0				\n\t"
+		" mov		#1, %1				\n\t"
+		".balign 4					\n\t"
+		"5:	.long 3b				\n\t"
+		".previous					\n\t"
+		".section __ex_table,\"a\"			\n\t"
+		".long 1b, 4b					\n\t"
+		".long 2b, 4b					\n\t"
+		".previous					\n\t"
+		: "=&r" (old), "=r" (faulted)
+		: "r" (parent), "r" (return_hooker)
+	);
+
+	if (unlikely(faulted)) {
+		ftrace_graph_stop();
+		WARN_ON(1);
+		return;
+	}
+
+	err = ftrace_push_return_trace(old, self_addr, &trace.depth, 0);
+	if (err == -EBUSY) {
+		__raw_writel(old, parent);
+		return;
+	}
+
+	trace.func = self_addr;
+
+	/* Only trace if the calling function expects to */
+	if (!ftrace_graph_entry(&trace)) {
+		current->curr_ret_stack--;
+		__raw_writel(old, parent);
+	}
+}
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+
+#ifdef CONFIG_FTRACE_SYSCALLS
+
+extern unsigned long __start_syscalls_metadata[];
+extern unsigned long __stop_syscalls_metadata[];
+extern unsigned long *sys_call_table;
+
+static struct syscall_metadata **syscalls_metadata;
+
+static struct syscall_metadata *find_syscall_meta(unsigned long *syscall)
+{
+	struct syscall_metadata *start;
+	struct syscall_metadata *stop;
+	char str[KSYM_SYMBOL_LEN];
+
+
+	start = (struct syscall_metadata *)__start_syscalls_metadata;
+	stop = (struct syscall_metadata *)__stop_syscalls_metadata;
+	kallsyms_lookup((unsigned long) syscall, NULL, NULL, NULL, str);
+
+	for ( ; start < stop; start++) {
+		if (start->name && !strcmp(start->name, str))
+			return start;
+	}
+
+	return NULL;
+}
+
+#define FTRACE_SYSCALL_MAX	(NR_syscalls - 1)
+
+struct syscall_metadata *syscall_nr_to_meta(int nr)
+{
+	if (!syscalls_metadata || nr >= FTRACE_SYSCALL_MAX || nr < 0)
+		return NULL;
+
+	return syscalls_metadata[nr];
+}
+
+void arch_init_ftrace_syscalls(void)
+{
+	int i;
+	struct syscall_metadata *meta;
+	unsigned long **psys_syscall_table = &sys_call_table;
+	static atomic_t refs;
+
+	if (atomic_inc_return(&refs) != 1)
+		goto end;
+
+	syscalls_metadata = kzalloc(sizeof(*syscalls_metadata) *
+					FTRACE_SYSCALL_MAX, GFP_KERNEL);
+	if (!syscalls_metadata) {
+		WARN_ON(1);
+		return;
+	}
+
+	for (i = 0; i < FTRACE_SYSCALL_MAX; i++) {
+		meta = find_syscall_meta(psys_syscall_table[i]);
+		syscalls_metadata[i] = meta;
+	}
+	return;
+
+	/* Paranoid: avoid overflow */
+end:
+	atomic_dec(&refs);
+}
+#endif /* CONFIG_FTRACE_SYSCALLS */
diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c
index 92d7740..9fee977 100644
--- a/arch/sh/kernel/process_32.c
+++ b/arch/sh/kernel/process_32.c
@@ -23,6 +23,7 @@
 #include <linux/tick.h>
 #include <linux/reboot.h>
 #include <linux/fs.h>
+#include <linux/ftrace.h>
 #include <linux/preempt.h>
 #include <asm/uaccess.h>
 #include <asm/mmu_context.h>
@@ -264,8 +265,8 @@
  *	switch_to(x,y) should switch tasks from x to y.
  *
  */
-struct task_struct *__switch_to(struct task_struct *prev,
-				struct task_struct *next)
+__notrace_funcgraph struct task_struct *
+__switch_to(struct task_struct *prev, struct task_struct *next)
 {
 #if defined(CONFIG_SH_FPU)
 	unlazy_fpu(prev, task_pt_regs(prev));
diff --git a/arch/sh/kernel/ptrace_32.c b/arch/sh/kernel/ptrace_32.c
index 3392e83..c198ece 100644
--- a/arch/sh/kernel/ptrace_32.c
+++ b/arch/sh/kernel/ptrace_32.c
@@ -34,6 +34,8 @@
 #include <asm/syscalls.h>
 #include <asm/fpu.h>
 
+#include <trace/syscall.h>
+
 /*
  * This routine will get a word off of the process kernel stack.
  */
@@ -459,6 +461,9 @@
 		 */
 		ret = -1L;
 
+	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+		ftrace_syscall_enter(regs);
+
 	if (unlikely(current->audit_context))
 		audit_syscall_entry(audit_arch(), regs->regs[3],
 				    regs->regs[4], regs->regs[5],
@@ -475,6 +480,9 @@
 		audit_syscall_exit(AUDITSC_RESULT(regs->regs[0]),
 				   regs->regs[0]);
 
+	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+		ftrace_syscall_exit(regs);
+
 	step = test_thread_flag(TIF_SINGLESTEP);
 	if (step || test_thread_flag(TIF_SYSCALL_TRACE))
 		tracehook_report_syscall_exit(regs, step);
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index dd38338..ceb409b 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -30,6 +30,7 @@
 #include <linux/clk.h>
 #include <linux/delay.h>
 #include <linux/platform_device.h>
+#include <linux/lmb.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
 #include <asm/page.h>
@@ -233,39 +234,45 @@
 void __init setup_bootmem_allocator(unsigned long free_pfn)
 {
 	unsigned long bootmap_size;
+	unsigned long bootmap_pages, bootmem_paddr;
+	u64 total_pages = (lmb_end_of_DRAM() - __MEMORY_START) >> PAGE_SHIFT;
+	int i;
+
+	bootmap_pages = bootmem_bootmap_pages(total_pages);
+
+	bootmem_paddr = lmb_alloc(bootmap_pages << PAGE_SHIFT, PAGE_SIZE);
 
 	/*
 	 * Find a proper area for the bootmem bitmap. After this
 	 * bootstrap step all allocations (until the page allocator
 	 * is intact) must be done via bootmem_alloc().
 	 */
-	bootmap_size = init_bootmem_node(NODE_DATA(0), free_pfn,
+	bootmap_size = init_bootmem_node(NODE_DATA(0),
+					 bootmem_paddr >> PAGE_SHIFT,
 					 min_low_pfn, max_low_pfn);
 
-	__add_active_range(0, min_low_pfn, max_low_pfn);
+	/* Add active regions with valid PFNs. */
+	for (i = 0; i < lmb.memory.cnt; i++) {
+		unsigned long start_pfn, end_pfn;
+		start_pfn = lmb.memory.region[i].base >> PAGE_SHIFT;
+		end_pfn = start_pfn + lmb_size_pages(&lmb.memory, i);
+		__add_active_range(0, start_pfn, end_pfn);
+	}
+
+	/*
+	 * Add all physical memory to the bootmem map and mark each
+	 * area as present.
+	 */
 	register_bootmem_low_pages();
 
-	node_set_online(0);
-
-	/*
-	 * Reserve the kernel text and
-	 * Reserve the bootmem bitmap. We do this in two steps (first step
-	 * was init_bootmem()), because this catches the (definitely buggy)
-	 * case of us accidentally initializing the bootmem allocator with
-	 * an invalid RAM area.
-	 */
-	reserve_bootmem(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET,
-			(PFN_PHYS(free_pfn) + bootmap_size + PAGE_SIZE - 1) -
-			(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET),
-			BOOTMEM_DEFAULT);
-
-	/*
-	 * Reserve physical pages below CONFIG_ZERO_PAGE_OFFSET.
-	 */
-	if (CONFIG_ZERO_PAGE_OFFSET != 0)
-		reserve_bootmem(__MEMORY_START, CONFIG_ZERO_PAGE_OFFSET,
+	/* Reserve the sections we're already using. */
+	for (i = 0; i < lmb.reserved.cnt; i++)
+		reserve_bootmem(lmb.reserved.region[i].base,
+				lmb_size_bytes(&lmb.reserved, i),
 				BOOTMEM_DEFAULT);
 
+	node_set_online(0);
+
 	sparse_memory_present_with_active_regions(0);
 
 #ifdef CONFIG_BLK_DEV_INITRD
@@ -296,12 +303,37 @@
 static void __init setup_memory(void)
 {
 	unsigned long start_pfn;
+	u64 base = min_low_pfn << PAGE_SHIFT;
+	u64 size = (max_low_pfn << PAGE_SHIFT) - base;
 
 	/*
 	 * Partially used pages are not usable - thus
 	 * we are rounding upwards:
 	 */
 	start_pfn = PFN_UP(__pa(_end));
+
+	lmb_add(base, size);
+
+	/*
+	 * Reserve the kernel text and
+	 * Reserve the bootmem bitmap. We do this in two steps (first step
+	 * was init_bootmem()), because this catches the (definitely buggy)
+	 * case of us accidentally initializing the bootmem allocator with
+	 * an invalid RAM area.
+	 */
+	lmb_reserve(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET,
+		    (PFN_PHYS(start_pfn) + PAGE_SIZE - 1) -
+		    (__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET));
+
+	/*
+	 * Reserve physical pages below CONFIG_ZERO_PAGE_OFFSET.
+	 */
+	if (CONFIG_ZERO_PAGE_OFFSET != 0)
+		lmb_reserve(__MEMORY_START, CONFIG_ZERO_PAGE_OFFSET);
+
+	lmb_analyze();
+	lmb_dump_all();
+
 	setup_bootmem_allocator(start_pfn);
 }
 #else
@@ -402,6 +434,7 @@
 	nodes_clear(node_online_map);
 
 	/* Setup bootmem with available RAM */
+	lmb_init();
 	setup_memory();
 	sparse_init();
 
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index 9b352a1..d2424b0 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -21,6 +21,7 @@
 #include <linux/smp.h>
 #include <linux/rtc.h>
 #include <asm/clock.h>
+#include <asm/hwblk.h>
 #include <asm/rtc.h>
 
 /* Dummy RTC ops */
@@ -96,6 +97,7 @@
 	if (board_time_init)
 		board_time_init();
 
+	hwblk_init();
 	clk_init();
 
 	rtc_sh_get_time(&xtime);
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index f53c76a..674ed8f 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -50,12 +50,7 @@
 		_etext = .;		/* End of text section */
 	} = 0x0009
 
-	. = ALIGN(16);		/* Exception table */
-	__ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) {
-		__start___ex_table = .;
-		*(__ex_table)
-		__stop___ex_table = .;
-	}
+	EXCEPTION_TABLE(16)
 
 	NOTES
 	RO_DATA(PAGE_SIZE)
@@ -71,69 +66,14 @@
 		__uncached_end = .;
 	}
 
-	. = ALIGN(THREAD_SIZE);
-	.data : AT(ADDR(.data) - LOAD_OFFSET) {		/* Data */
-		*(.data.init_task)
-
-		. = ALIGN(L1_CACHE_BYTES);
-		*(.data.cacheline_aligned)
-
-		. = ALIGN(L1_CACHE_BYTES);
-		*(.data.read_mostly)
-
-		. = ALIGN(PAGE_SIZE);
-		*(.data.page_aligned)
-
-		__nosave_begin = .;
-		*(.data.nosave)
-		. = ALIGN(PAGE_SIZE);
-		__nosave_end = .;
-
-		DATA_DATA
-		CONSTRUCTORS
-	}
+	RW_DATA_SECTION(L1_CACHE_BYTES, PAGE_SIZE, THREAD_SIZE)
 
 	_edata = .;			/* End of data section */
 
 	. = ALIGN(PAGE_SIZE);		/* Init code and data */
-	.init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
-		__init_begin = .;
-		_sinittext = .;
-		INIT_TEXT
-		_einittext = .;
-	}
-
-	.init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { INIT_DATA }
-
-	. = ALIGN(16);
-	.init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) {
-		__setup_start = .;
-		*(.init.setup)
-		__setup_end = .;
-	}
-
-	.initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) {
-		__initcall_start = .;
-		INITCALLS
-		__initcall_end = .;
-	}
-
-	.con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) {
-		__con_initcall_start = .;
-		*(.con_initcall.init)
-		__con_initcall_end = .;
-	}
-
-	SECURITY_INIT
-
-#ifdef CONFIG_BLK_DEV_INITRD
-	. = ALIGN(PAGE_SIZE);
-	.init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) {
-		__initramfs_start = .;
-		*(.init.ramfs)
-		__initramfs_end = .;
-	}
-#endif
+	__init_begin = .;
+	INIT_TEXT_SECTION(PAGE_SIZE)
+	INIT_DATA_SECTION(16)
 
 	. = ALIGN(4);
 	.machvec.init : AT(ADDR(.machvec.init) - LOAD_OFFSET) {
@@ -152,16 +92,11 @@
 	.exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { EXIT_DATA }
 
 	. = ALIGN(PAGE_SIZE);
-	.bss : AT(ADDR(.bss) - LOAD_OFFSET) {
-		__init_end = .;
-		__bss_start = .;		/* BSS */
-		*(.bss.page_aligned)
-		*(.bss)
-		*(COMMON)
-		. = ALIGN(4);
-		_ebss = .;			/* uClinux MTD sucks */
-		_end = . ;
-	}
+	__init_end = .;
+	BSS(PAGE_SIZE)
+	. = ALIGN(4);
+	_ebss = .;			/* uClinux MTD sucks */
+	_end = . ;
 
 	/*
 	 * When something in the kernel is NOT compiled as a module, the
@@ -170,7 +105,7 @@
 	 * it's a module.
 	 */
 	/DISCARD/ : {
-		*(.exitcall.exit)
+		EXIT_CALL
 	}
 
 	STABS_DEBUG
diff --git a/arch/sh/kernel/vmlinux_64.lds.S b/arch/sh/kernel/vmlinux_64.lds.S
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/arch/sh/kernel/vmlinux_64.lds.S
diff --git a/arch/sh/lib/Makefile b/arch/sh/lib/Makefile
index aaea580..19328d9 100644
--- a/arch/sh/lib/Makefile
+++ b/arch/sh/lib/Makefile
@@ -25,6 +25,7 @@
 
 lib-$(CONFIG_MMU)		+= copy_page.o clear_page.o
 lib-$(CONFIG_FUNCTION_TRACER)	+= mcount.o
+lib-$(CONFIG_FUNCTION_GRAPH_TRACER) += mcount.o
 lib-y				+= $(memcpy-y) $(udivsi3-y)
 
 EXTRA_CFLAGS += -Werror
diff --git a/arch/sh/lib/mcount.S b/arch/sh/lib/mcount.S
index 110fbfe..bd3ec64 100644
--- a/arch/sh/lib/mcount.S
+++ b/arch/sh/lib/mcount.S
@@ -2,13 +2,15 @@
  * arch/sh/lib/mcount.S
  *
  *  Copyright (C) 2008  Paul Mundt
- *  Copyright (C) 2008  Matt Fleming
+ *  Copyright (C) 2008, 2009  Matt Fleming
  *
  * 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.
  */
 #include <asm/ftrace.h>
+#include <asm/thread_info.h>
+#include <asm/asm-offsets.h>
 
 #define MCOUNT_ENTER()		\
 	mov.l	r4, @-r15;	\
@@ -28,6 +30,55 @@
 	rts;			\
 	 mov.l	@r15+, r4
 
+#ifdef CONFIG_STACK_DEBUG
+/*
+ * Perform diagnostic checks on the state of the kernel stack.
+ *
+ * Check for stack overflow. If there is less than 1KB free
+ * then it has overflowed.
+ *
+ * Make sure the stack pointer contains a valid address. Valid
+ * addresses for kernel stacks are anywhere after the bss
+ * (after _ebss) and anywhere in init_thread_union (init_stack).
+ */
+#define STACK_CHECK()					\
+	mov	#(THREAD_SIZE >> 10), r0;		\
+	shll8	r0;					\
+	shll2	r0;					\
+							\
+	/* r1 = sp & (THREAD_SIZE - 1) */		\
+	mov	#-1, r1;				\
+	add	r0, r1;					\
+	and	r15, r1;				\
+							\
+	mov	#TI_SIZE, r3;				\
+	mov	#(STACK_WARN >> 8), r2;			\
+	shll8	r2;					\
+	add	r3, r2;					\
+							\
+	/* Is the stack overflowing? */			\
+	cmp/hi	r2, r1;					\
+	bf	stack_panic;				\
+							\
+	/* If sp > _ebss then we're OK. */		\
+	mov.l	.L_ebss, r1;				\
+	cmp/hi	r1, r15;				\
+	bt	1f;					\
+							\
+	/* If sp < init_stack, we're not OK. */		\
+	mov.l	.L_init_thread_union, r1;		\
+	cmp/hs	r1, r15;				\
+	bf	stack_panic;				\
+							\
+	/* If sp > init_stack && sp < _ebss, not OK. */	\
+	add	r0, r1;					\
+	cmp/hs	r1, r15;				\
+	bt	stack_panic;				\
+1:
+#else
+#define STACK_CHECK()
+#endif /* CONFIG_STACK_DEBUG */
+
 	.align 2
 	.globl	_mcount
 	.type	_mcount,@function
@@ -35,6 +86,14 @@
 	.type	mcount,@function
 _mcount:
 mcount:
+#ifndef CONFIG_DYNAMIC_FTRACE
+	mov.l	.Lfunction_trace_stop, r0
+	mov.l	@r0, r0
+	tst	r0, r0
+	bf	ftrace_stub
+#endif
+	STACK_CHECK()
+
 	MCOUNT_ENTER()
 
 #ifdef CONFIG_DYNAMIC_FTRACE
@@ -52,16 +111,71 @@
 	jsr	@r6
 	 nop
 
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	mov.l   .Lftrace_graph_return, r6
+	mov.l   .Lftrace_stub, r7
+	cmp/eq  r6, r7
+	bt      1f
+
+	mov.l   .Lftrace_graph_caller, r0
+	jmp     @r0
+	 nop
+
+1:
+	mov.l	.Lftrace_graph_entry, r6
+	mov.l	.Lftrace_graph_entry_stub, r7
+	cmp/eq	r6, r7
+	bt	skip_trace
+
+	mov.l   .Lftrace_graph_caller, r0
+	jmp	@r0
+	 nop
+
+	.align 2
+.Lftrace_graph_return:
+	.long   ftrace_graph_return
+.Lftrace_graph_entry:
+	.long   ftrace_graph_entry
+.Lftrace_graph_entry_stub:
+	.long   ftrace_graph_entry_stub
+.Lftrace_graph_caller:
+	.long   ftrace_graph_caller
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+
+	.globl skip_trace
 skip_trace:
 	MCOUNT_LEAVE()
 
 	.align 2
 .Lftrace_trace_function:
-	.long	ftrace_trace_function
+	.long   ftrace_trace_function
 
 #ifdef CONFIG_DYNAMIC_FTRACE
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+/*
+ * NOTE: Do not move either ftrace_graph_call or ftrace_caller
+ * as this will affect the calculation of GRAPH_INSN_OFFSET.
+ */
+	.globl ftrace_graph_call
+ftrace_graph_call:
+	mov.l	.Lskip_trace, r0
+	jmp	@r0
+	 nop
+
+	.align 2
+.Lskip_trace:
+	.long	skip_trace
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+
 	.globl ftrace_caller
 ftrace_caller:
+	mov.l	.Lfunction_trace_stop, r0
+	mov.l	@r0, r0
+	tst	r0, r0
+	bf	ftrace_stub
+
+	STACK_CHECK()
+
 	MCOUNT_ENTER()
 
 	.globl ftrace_call
@@ -70,7 +184,12 @@
 	jsr	@r6
 	 nop
 
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	bra	ftrace_graph_call
+	 nop
+#else
 	MCOUNT_LEAVE()
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 /*
@@ -88,3 +207,99 @@
 ftrace_stub:
 	rts
 	 nop
+
+#ifdef CONFIG_STACK_DEBUG
+	.globl	stack_panic
+stack_panic:
+	mov.l	.Ldump_stack, r0
+	jsr	@r0
+	 nop
+
+	mov.l	.Lpanic, r0
+	jsr	@r0
+	 mov.l	.Lpanic_s, r4
+
+	rts
+	 nop
+
+	.align 2
+.Lfunction_trace_stop:
+	.long	function_trace_stop
+.L_ebss:
+	.long	_ebss
+.L_init_thread_union:
+	.long	init_thread_union
+.Lpanic:
+	.long	panic
+.Lpanic_s:
+	.long	.Lpanic_str
+.Ldump_stack:
+	.long	dump_stack
+
+	.section	.rodata
+	.align 2
+.Lpanic_str:
+	.string "Stack error"
+#endif /* CONFIG_STACK_DEBUG */
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	.globl	ftrace_graph_caller
+ftrace_graph_caller:
+	mov.l	2f, r0
+	mov.l	@r0, r0
+	tst	r0, r0
+	bt	1f
+
+	mov.l	3f, r1
+	jmp	@r1
+	 nop
+1:
+	/*
+	 * MCOUNT_ENTER() pushed 5 registers onto the stack, so
+	 * the stack address containing our return address is
+	 * r15 + 20.
+	 */
+	mov	#20, r0
+	add	r15, r0
+	mov	r0, r4
+
+	mov.l	.Lprepare_ftrace_return, r0
+	jsr	@r0
+	 nop
+
+	MCOUNT_LEAVE()
+
+	.align 2
+2:	.long	function_trace_stop
+3:	.long	skip_trace
+.Lprepare_ftrace_return:
+	.long	prepare_ftrace_return
+
+	.globl	return_to_handler
+return_to_handler:
+	/*
+	 * Save the return values.
+	 */
+	mov.l	r0, @-r15
+	mov.l	r1, @-r15
+
+	mov	#0, r4
+
+	mov.l	.Lftrace_return_to_handler, r0
+	jsr	@r0
+	 nop
+
+	/*
+	 * The return value from ftrace_return_handler has the real
+	 * address that we should return to.
+	 */
+	lds	r0, pr
+	mov.l	@r15+, r1
+	rts
+	 mov.l	@r15+, r0
+
+
+	.align 2
+.Lftrace_return_to_handler:
+	.long	ftrace_return_to_handler
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
diff --git a/arch/sh/mm/fault_32.c b/arch/sh/mm/fault_32.c
index 7192594..08d0117 100644
--- a/arch/sh/mm/fault_32.c
+++ b/arch/sh/mm/fault_32.c
@@ -2,7 +2,7 @@
  * Page fault handler for SH with an MMU.
  *
  *  Copyright (C) 1999  Niibe Yutaka
- *  Copyright (C) 2003 - 2008  Paul Mundt
+ *  Copyright (C) 2003 - 2009  Paul Mundt
  *
  *  Based on linux/arch/i386/mm/fault.c:
  *   Copyright (C) 1995  Linus Torvalds
@@ -25,18 +25,84 @@
 {
 	int ret = 0;
 
-#ifdef CONFIG_KPROBES
-	if (!user_mode(regs)) {
+	if (kprobes_built_in() && !user_mode(regs)) {
 		preempt_disable();
 		if (kprobe_running() && kprobe_fault_handler(regs, trap))
 			ret = 1;
 		preempt_enable();
 	}
-#endif
 
 	return ret;
 }
 
+static inline pmd_t *vmalloc_sync_one(pgd_t *pgd, unsigned long address)
+{
+	unsigned index = pgd_index(address);
+	pgd_t *pgd_k;
+	pud_t *pud, *pud_k;
+	pmd_t *pmd, *pmd_k;
+
+	pgd += index;
+	pgd_k = init_mm.pgd + index;
+
+	if (!pgd_present(*pgd_k))
+		return NULL;
+
+	pud = pud_offset(pgd, address);
+	pud_k = pud_offset(pgd_k, address);
+	if (!pud_present(*pud_k))
+		return NULL;
+
+	pmd = pmd_offset(pud, address);
+	pmd_k = pmd_offset(pud_k, address);
+	if (!pmd_present(*pmd_k))
+		return NULL;
+
+	if (!pmd_present(*pmd))
+		set_pmd(pmd, *pmd_k);
+	else
+		BUG_ON(pmd_page(*pmd) != pmd_page(*pmd_k));
+
+	return pmd_k;
+}
+
+/*
+ * Handle a fault on the vmalloc or module mapping area
+ */
+static noinline int vmalloc_fault(unsigned long address)
+{
+	pgd_t *pgd_k;
+	pmd_t *pmd_k;
+	pte_t *pte_k;
+
+	/* Make sure we are in vmalloc area: */
+	if (!(address >= VMALLOC_START && address < VMALLOC_END))
+		return -1;
+
+	/*
+	 * Synchronize this task's top level page-table
+	 * with the 'reference' page table.
+	 *
+	 * Do _not_ use "current" here. We might be inside
+	 * an interrupt in the middle of a task switch..
+	 */
+	pgd_k = get_TTB();
+	pmd_k = vmalloc_sync_one(__va((unsigned long)pgd_k), address);
+	if (!pmd_k)
+		return -1;
+
+	pte_k = pte_offset_kernel(pmd_k, address);
+	if (!pte_present(*pte_k))
+		return -1;
+
+	return 0;
+}
+
+static int fault_in_kernel_space(unsigned long address)
+{
+	return address >= TASK_SIZE;
+}
+
 /*
  * This routine handles page faults.  It determines the address,
  * and the problem, and then passes it off to one of the appropriate
@@ -46,6 +112,7 @@
 					unsigned long writeaccess,
 					unsigned long address)
 {
+	unsigned long vec;
 	struct task_struct *tsk;
 	struct mm_struct *mm;
 	struct vm_area_struct * vma;
@@ -53,59 +120,30 @@
 	int fault;
 	siginfo_t info;
 
-	/*
-	 * We don't bother with any notifier callbacks here, as they are
-	 * all handled through the __do_page_fault() fast-path.
-	 */
-
 	tsk = current;
+	mm = tsk->mm;
 	si_code = SEGV_MAPERR;
+	vec = lookup_exception_vector();
 
-	if (unlikely(address >= TASK_SIZE)) {
-		/*
-		 * Synchronize this task's top level page-table
-		 * with the 'reference' page table.
-		 *
-		 * Do _not_ use "tsk" here. We might be inside
-		 * an interrupt in the middle of a task switch..
-		 */
-		int offset = pgd_index(address);
-		pgd_t *pgd, *pgd_k;
-		pud_t *pud, *pud_k;
-		pmd_t *pmd, *pmd_k;
-
-		pgd = get_TTB() + offset;
-		pgd_k = swapper_pg_dir + offset;
-
-		if (!pgd_present(*pgd)) {
-			if (!pgd_present(*pgd_k))
-				goto bad_area_nosemaphore;
-			set_pgd(pgd, *pgd_k);
+	/*
+	 * We fault-in kernel-space virtual memory on-demand. The
+	 * 'reference' page table is init_mm.pgd.
+	 *
+	 * NOTE! We MUST NOT take any locks for this case. We may
+	 * be in an interrupt or a critical region, and should
+	 * only copy the information from the master page table,
+	 * nothing more.
+	 */
+	if (unlikely(fault_in_kernel_space(address))) {
+		if (vmalloc_fault(address) >= 0)
 			return;
-		}
-
-		pud = pud_offset(pgd, address);
-		pud_k = pud_offset(pgd_k, address);
-
-		if (!pud_present(*pud)) {
-			if (!pud_present(*pud_k))
-				goto bad_area_nosemaphore;
-			set_pud(pud, *pud_k);
+		if (notify_page_fault(regs, vec))
 			return;
-		}
 
-		pmd = pmd_offset(pud, address);
-		pmd_k = pmd_offset(pud_k, address);
-		if (pmd_present(*pmd) || !pmd_present(*pmd_k))
-			goto bad_area_nosemaphore;
-		set_pmd(pmd, *pmd_k);
-
-		return;
+		goto bad_area_nosemaphore;
 	}
 
-	mm = tsk->mm;
-
-	if (unlikely(notify_page_fault(regs, lookup_exception_vector())))
+	if (unlikely(notify_page_fault(regs, vec)))
 		return;
 
 	/* Only enable interrupts if they were on before the fault */
@@ -115,8 +153,8 @@
 	perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, address);
 
 	/*
-	 * If we're in an interrupt or have no user
-	 * context, we must not take the fault..
+	 * If we're in an interrupt, have no user context or are running
+	 * in an atomic region then we must not take the fault:
 	 */
 	if (in_atomic() || !mm)
 		goto no_context;
@@ -132,10 +170,11 @@
 		goto bad_area;
 	if (expand_stack(vma, address))
 		goto bad_area;
-/*
- * Ok, we have a good vm_area for this memory access, so
- * we can handle it..
- */
+
+	/*
+	 * Ok, we have a good vm_area for this memory access, so
+	 * we can handle it..
+	 */
 good_area:
 	si_code = SEGV_ACCERR;
 	if (writeaccess) {
@@ -173,10 +212,10 @@
 	up_read(&mm->mmap_sem);
 	return;
 
-/*
- * Something tried to access memory that isn't in our memory map..
- * Fix it, but check if it's kernel or user first..
- */
+	/*
+	 * Something tried to access memory that isn't in our memory map..
+	 * Fix it, but check if it's kernel or user first..
+	 */
 bad_area:
 	up_read(&mm->mmap_sem);
 
diff --git a/arch/sh/mm/numa.c b/arch/sh/mm/numa.c
index 095d93b..9b784fd 100644
--- a/arch/sh/mm/numa.c
+++ b/arch/sh/mm/numa.c
@@ -9,6 +9,7 @@
  */
 #include <linux/module.h>
 #include <linux/bootmem.h>
+#include <linux/lmb.h>
 #include <linux/mm.h>
 #include <linux/numa.h>
 #include <linux/pfn.h>
@@ -26,6 +27,15 @@
 void __init setup_memory(void)
 {
 	unsigned long free_pfn = PFN_UP(__pa(_end));
+	u64 base = min_low_pfn << PAGE_SHIFT;
+	u64 size = (max_low_pfn << PAGE_SHIFT) - min_low_pfn;
+
+	lmb_add(base, size);
+
+	/* Reserve the LMB regions used by the kernel, initrd, etc.. */
+	lmb_reserve(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET,
+		    (PFN_PHYS(free_pfn) + PAGE_SIZE - 1) -
+		    (__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET));
 
 	/*
 	 * Node 0 sets up its pgdat at the first available pfn,
@@ -45,24 +55,23 @@
 
 void __init setup_bootmem_node(int nid, unsigned long start, unsigned long end)
 {
-	unsigned long bootmap_pages, bootmap_start, bootmap_size;
-	unsigned long start_pfn, free_pfn, end_pfn;
+	unsigned long bootmap_pages;
+	unsigned long start_pfn, end_pfn;
+	unsigned long bootmem_paddr;
 
 	/* Don't allow bogus node assignment */
 	BUG_ON(nid > MAX_NUMNODES || nid == 0);
 
-	/*
-	 * The free pfn starts at the beginning of the range, and is
-	 * advanced as necessary for pgdat and node map allocations.
-	 */
-	free_pfn = start_pfn = start >> PAGE_SHIFT;
+	start_pfn = start >> PAGE_SHIFT;
 	end_pfn = end >> PAGE_SHIFT;
 
+	lmb_add(start, end - start);
+
 	__add_active_range(nid, start_pfn, end_pfn);
 
 	/* Node-local pgdat */
-	NODE_DATA(nid) = pfn_to_kaddr(free_pfn);
-	free_pfn += PFN_UP(sizeof(struct pglist_data));
+	NODE_DATA(nid) = __va(lmb_alloc_base(sizeof(struct pglist_data),
+					     SMP_CACHE_BYTES, end_pfn));
 	memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
 
 	NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
@@ -71,16 +80,17 @@
 
 	/* Node-local bootmap */
 	bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn);
-	bootmap_start = (unsigned long)pfn_to_kaddr(free_pfn);
-	bootmap_size = init_bootmem_node(NODE_DATA(nid), free_pfn, start_pfn,
-				    end_pfn);
+	bootmem_paddr = lmb_alloc_base(bootmap_pages << PAGE_SHIFT,
+				       PAGE_SIZE, end_pfn);
+	init_bootmem_node(NODE_DATA(nid), bootmem_paddr >> PAGE_SHIFT,
+			  start_pfn, end_pfn);
 
 	free_bootmem_with_active_regions(nid, end_pfn);
 
 	/* Reserve the pgdat and bootmap space with the bootmem allocator */
 	reserve_bootmem_node(NODE_DATA(nid), start_pfn << PAGE_SHIFT,
 			     sizeof(struct pglist_data), BOOTMEM_DEFAULT);
-	reserve_bootmem_node(NODE_DATA(nid), free_pfn << PAGE_SHIFT,
+	reserve_bootmem_node(NODE_DATA(nid), bootmem_paddr,
 			     bootmap_pages << PAGE_SHIFT, BOOTMEM_DEFAULT);
 
 	/* It's up */
diff --git a/drivers/usb/gadget/m66592-udc.c b/drivers/usb/gadget/m66592-udc.c
index 43dcf9e..0dddd2f 100644
--- a/drivers/usb/gadget/m66592-udc.c
+++ b/drivers/usb/gadget/m66592-udc.c
@@ -37,7 +37,7 @@
 MODULE_AUTHOR("Yoshihiro Shimoda");
 MODULE_ALIAS("platform:m66592_udc");
 
-#define DRIVER_VERSION	"18 Oct 2007"
+#define DRIVER_VERSION	"26 Jun 2009"
 
 /* module parameters */
 #if defined(CONFIG_SUPERH_BUILT_IN_M66592)
@@ -276,24 +276,27 @@
 		buf_bsize = 0;
 		break;
 	case M66592_BULK:
-		bufnum = m66592->bi_bufnum +
-			 (info->pipe - M66592_BASE_PIPENUM_BULK) * 16;
-		m66592->bi_bufnum += 16;
+		/* isochronous pipes may be used as bulk pipes */
+		if (info->pipe > M66592_BASE_PIPENUM_BULK)
+			bufnum = info->pipe - M66592_BASE_PIPENUM_BULK;
+		else
+			bufnum = info->pipe - M66592_BASE_PIPENUM_ISOC;
+
+		bufnum = M66592_BASE_BUFNUM + (bufnum * 16);
 		buf_bsize = 7;
 		pipecfg |= M66592_DBLB;
 		if (!info->dir_in)
 			pipecfg |= M66592_SHTNAK;
 		break;
 	case M66592_ISO:
-		bufnum = m66592->bi_bufnum +
+		bufnum = M66592_BASE_BUFNUM +
 			 (info->pipe - M66592_BASE_PIPENUM_ISOC) * 16;
-		m66592->bi_bufnum += 16;
 		buf_bsize = 7;
 		break;
 	}
-	if (m66592->bi_bufnum > M66592_MAX_BUFNUM) {
-		pr_err("m66592 pipe memory is insufficient(%d)\n",
-				m66592->bi_bufnum);
+
+	if (buf_bsize && ((bufnum + 16) >= M66592_MAX_BUFNUM)) {
+		pr_err("m66592 pipe memory is insufficient\n");
 		return -ENOMEM;
 	}
 
@@ -313,17 +316,6 @@
 	if (info->pipe == 0)
 		return;
 
-	switch (info->type) {
-	case M66592_BULK:
-		if (is_bulk_pipe(info->pipe))
-			m66592->bi_bufnum -= 16;
-		break;
-	case M66592_ISO:
-		if (is_isoc_pipe(info->pipe))
-			m66592->bi_bufnum -= 16;
-		break;
-	}
-
 	if (is_bulk_pipe(info->pipe)) {
 		m66592->bulk--;
 	} else if (is_interrupt_pipe(info->pipe))
@@ -1603,8 +1595,6 @@
 	m66592->timer.data = (unsigned long)m66592;
 	m66592->reg = reg;
 
-	m66592->bi_bufnum = M66592_BASE_BUFNUM;
-
 	ret = request_irq(irq, m66592_irq, IRQF_DISABLED | IRQF_SHARED,
 			udc_name, m66592);
 	if (ret < 0) {
diff --git a/drivers/usb/gadget/m66592-udc.h b/drivers/usb/gadget/m66592-udc.h
index 286ce07..9a9c2bf 100644
--- a/drivers/usb/gadget/m66592-udc.h
+++ b/drivers/usb/gadget/m66592-udc.h
@@ -506,7 +506,6 @@
 	int interrupt;
 	int isochronous;
 	int num_dma;
-	int bi_bufnum;	/* bulk and isochronous's bufnum */
 };
 
 #define gadget_to_m66592(_gadget) container_of(_gadget, struct m66592, gadget)
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 8afcf08..ca330b1 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -1866,7 +1866,7 @@
 
 config FB_SH_MOBILE_LCDC
 	tristate "SuperH Mobile LCDC framebuffer support"
-	depends on FB && SUPERH
+	depends on FB && SUPERH && HAVE_CLK
 	select FB_SYS_FILLRECT
 	select FB_SYS_COPYAREA
 	select FB_SYS_IMAGEBLIT
diff --git a/drivers/video/sh_mobile_lcdcfb.c b/drivers/video/sh_mobile_lcdcfb.c
index 8f24564..98fb82f 100644
--- a/drivers/video/sh_mobile_lcdcfb.c
+++ b/drivers/video/sh_mobile_lcdcfb.c
@@ -42,11 +42,9 @@
 struct sh_mobile_lcdc_priv {
 	void __iomem *base;
 	int irq;
-#ifdef CONFIG_HAVE_CLK
 	atomic_t clk_usecnt;
 	struct clk *dot_clk;
 	struct clk *clk;
-#endif
 	unsigned long lddckr;
 	struct sh_mobile_lcdc_chan ch[2];
 	int started;
@@ -185,7 +183,6 @@
 	lcdc_sys_read_data,
 };
 
-#ifdef CONFIG_HAVE_CLK
 static void sh_mobile_lcdc_clk_on(struct sh_mobile_lcdc_priv *priv)
 {
 	if (atomic_inc_and_test(&priv->clk_usecnt)) {
@@ -203,10 +200,6 @@
 		clk_disable(priv->clk);
 	}
 }
-#else
-static void sh_mobile_lcdc_clk_on(struct sh_mobile_lcdc_priv *priv) {}
-static void sh_mobile_lcdc_clk_off(struct sh_mobile_lcdc_priv *priv) {}
-#endif
 
 static int sh_mobile_lcdc_sginit(struct fb_info *info,
 				  struct list_head *pagelist)
@@ -515,7 +508,6 @@
 		board_cfg = &ch->cfg.board_cfg;
 		if (board_cfg->display_off)
 			board_cfg->display_off(board_cfg->board_data);
-
 	}
 
 	/* stop the lcdc */
@@ -574,9 +566,7 @@
 				       int clock_source,
 				       struct sh_mobile_lcdc_priv *priv)
 {
-#ifdef CONFIG_HAVE_CLK
 	char clk_name[8];
-#endif
 	char *str;
 	int icksel;
 
@@ -590,7 +580,6 @@
 
 	priv->lddckr = icksel << 16;
 
-#ifdef CONFIG_HAVE_CLK
 	atomic_set(&priv->clk_usecnt, -1);
 	snprintf(clk_name, sizeof(clk_name), "lcdc%d", pdev->id);
 	priv->clk = clk_get(&pdev->dev, clk_name);
@@ -598,7 +587,7 @@
 		dev_err(&pdev->dev, "cannot get clock \"%s\"\n", clk_name);
 		return PTR_ERR(priv->clk);
 	}
-	
+
 	if (str) {
 		priv->dot_clk = clk_get(&pdev->dev, str);
 		if (IS_ERR(priv->dot_clk)) {
@@ -607,7 +596,6 @@
 			return PTR_ERR(priv->dot_clk);
 		}
 	}
-#endif
 
 	return 0;
 }
@@ -942,11 +930,9 @@
 		framebuffer_release(info);
 	}
 
-#ifdef CONFIG_HAVE_CLK
 	if (priv->dot_clk)
 		clk_put(priv->dot_clk);
 	clk_put(priv->clk);
-#endif
 
 	if (priv->base)
 		iounmap(priv->base);