12 files changed, 287 insertions, 323 deletions

diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile
index ad455ff5aeb..eb9092ca800 100644
--- a/arch/arm/kernel/Makefile
+++ b/arch/arm/kernel/Makefile
@@ -4,6 +4,10 @@
 
 AFLAGS_head.o := -DTEXT_OFFSET=$(TEXT_OFFSET)
 
+ifdef CONFIG_DYNAMIC_FTRACE
+CFLAGS_REMOVE_ftrace.o = -pg
+endif
+
 # Object file lists.
 
 obj-y		:= compat.o entry-armv.o entry-common.o irq.o \
@@ -18,6 +22,7 @@ obj-$(CONFIG_ARTHUR)		+= arthur.o
 obj-$(CONFIG_ISA_DMA)		+= dma-isa.o
 obj-$(CONFIG_PCI)		+= bios32.o isa.o
 obj-$(CONFIG_SMP)		+= smp.o
+obj-$(CONFIG_DYNAMIC_FTRACE)	+= ftrace.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o kprobes-decode.o
 obj-$(CONFIG_ATAGS_PROC)	+= atags.o
diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c
index 688b7b1ee41..cc7b246e965 100644
--- a/arch/arm/kernel/armksyms.c
+++ b/arch/arm/kernel/armksyms.c
@@ -18,6 +18,7 @@
 #include <asm/io.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
+#include <asm/ftrace.h>
 
 /*
  * libgcc functions - functions that are used internally by the
@@ -181,3 +182,7 @@ EXPORT_SYMBOL(_find_next_bit_be);
 #endif
 
 EXPORT_SYMBOL(copy_page);
+
+#ifdef CONFIG_FTRACE
+EXPORT_SYMBOL(mcount);
+#endif
diff --git a/arch/arm/kernel/atags.c b/arch/arm/kernel/atags.c
index 64c420805e6..42a1a1415fa 100644
--- a/arch/arm/kernel/atags.c
+++ b/arch/arm/kernel/atags.c
@@ -1,5 +1,4 @@
 #include <linux/slab.h>
-#include <linux/kexec.h>
 #include <linux/proc_fs.h>
 #include <asm/setup.h>
 #include <asm/types.h>
@@ -7,9 +6,8 @@
 
 struct buffer {
 	size_t size;
-	char *data;
+	char data[];
 };
-static struct buffer tags_buffer;
 
 static int
 read_buffer(char* page, char** start, off_t off, int count,
@@ -29,58 +27,57 @@ read_buffer(char* page, char** start, off_t off, int count,
 	return count;
 }
 
-
-static int
-create_proc_entries(void)
-{
-	struct proc_dir_entry* tags_entry;
-
-	tags_entry = create_proc_read_entry("atags", 0400, NULL, read_buffer, &tags_buffer);
-	if (!tags_entry)
-		return -ENOMEM;
-
-	return 0;
-}
-
-
-static char __initdata atags_copy_buf[KEXEC_BOOT_PARAMS_SIZE];
-static char __initdata *atags_copy;
+#define BOOT_PARAMS_SIZE 1536
+static char __initdata atags_copy[BOOT_PARAMS_SIZE];
 
 void __init save_atags(const struct tag *tags)
 {
-	atags_copy = atags_copy_buf;
-	memcpy(atags_copy, tags, KEXEC_BOOT_PARAMS_SIZE);
+	memcpy(atags_copy, tags, sizeof(atags_copy));
 }
 
-
 static int __init init_atags_procfs(void)
 {
-	struct tag *tag;
-	int error;
+	/*
+	 * This cannot go into save_atags() because kmalloc and proc don't work
+	 * yet when it is called.
+	 */
+	struct proc_dir_entry *tags_entry;
+	struct tag *tag = (struct tag *)atags_copy;
+	struct buffer *b;
+	size_t size;
 
-	if (!atags_copy) {
-		printk(KERN_WARNING "Exporting ATAGs: No saved tags found\n");
-		return -EIO;
+	if (tag->hdr.tag != ATAG_CORE) {
+		printk(KERN_INFO "No ATAGs?");
+		return -EINVAL;
 	}
 
-	for (tag = (struct tag *) atags_copy; tag->hdr.size; tag = tag_next(tag))
+	for (; tag->hdr.size; tag = tag_next(tag))
 		;
 
-	tags_buffer.size = ((char *) tag - atags_copy) + sizeof(tag->hdr);
-	tags_buffer.data = kmalloc(tags_buffer.size, GFP_KERNEL);
-	if (tags_buffer.data == NULL)
-		return -ENOMEM;
-	memcpy(tags_buffer.data, atags_copy, tags_buffer.size);
-
-	error = create_proc_entries();
-	if (error) {
-		printk(KERN_ERR "Exporting ATAGs: not enough memory\n");
-		kfree(tags_buffer.data);
-		tags_buffer.size = 0;
-		tags_buffer.data = NULL;
-	}
+	/* include the terminating ATAG_NONE */
+	size = (char *)tag - atags_copy + sizeof(struct tag_header);
 
-	return error;
-}
+	WARN_ON(tag->hdr.tag != ATAG_NONE);
+
+	b = kmalloc(sizeof(*b) + size, GFP_KERNEL);
+	if (!b)
+		goto nomem;
 
+	b->size = size;
+	memcpy(b->data, atags_copy, size);
+
+	tags_entry = create_proc_read_entry("atags", 0400,
+			NULL, read_buffer, b);
+
+	if (!tags_entry)
+		goto nomem;
+
+	return 0;
+
+nomem:
+	kfree(b);
+	printk(KERN_ERR "Exporting ATAGs: not enough memory\n");
+
+	return -ENOMEM;
+}
 arch_initcall(init_atags_procfs);
diff --git a/arch/arm/kernel/ecard.c b/arch/arm/kernel/ecard.c
index a53c0aba5c1..8bfd299bfe7 100644
--- a/arch/arm/kernel/ecard.c
+++ b/arch/arm/kernel/ecard.c
@@ -680,7 +680,7 @@ static int __init ecard_probeirqhw(void)
 #define IO_EC_MEMC8_BASE 0
 #endif
 
-unsigned int __ecard_address(ecard_t *ec, card_type_t type, card_speed_t speed)
+static unsigned int __ecard_address(ecard_t *ec, card_type_t type, card_speed_t speed)
 {
 	unsigned long address = 0;
 	int slot = ec->slot_no;
@@ -1002,7 +1002,7 @@ ecard_probe(int slot, card_type_t type)
 	}
 
 	rc = -ENODEV;
-	if ((ec->podaddr = ecard_address(ec, type, ECARD_SYNC)) == 0)
+	if ((ec->podaddr = __ecard_address(ec, type, ECARD_SYNC)) == 0)
 		goto nodev;
 
 	cid.r_zero = 1;
@@ -1141,10 +1141,10 @@ static int ecard_drv_probe(struct device *dev)
 
 	id = ecard_match_device(drv->id_table, ec);
 
-	ecard_claim(ec);
+	ec->claimed = 1;
 	ret = drv->probe(ec, id);
 	if (ret)
-		ecard_release(ec);
+		ec->claimed = 0;
 	return ret;
 }
 
@@ -1154,7 +1154,7 @@ static int ecard_drv_remove(struct device *dev)
 	struct ecard_driver *drv = ECARD_DRV(dev->driver);
 
 	drv->remove(ec);
-	ecard_release(ec);
+	ec->claimed = 0;
 
 	/*
 	 * Restore the default operations.  We ensure that the
@@ -1182,7 +1182,7 @@ static void ecard_drv_shutdown(struct device *dev)
 	if (dev->driver) {
 		if (drv->shutdown)
 			drv->shutdown(ec);
-		ecard_release(ec);
+		ec->claimed = 0;
 	}
 
 	/*
@@ -1239,7 +1239,6 @@ static int ecard_bus_init(void)
 postcore_initcall(ecard_bus_init);
 
 EXPORT_SYMBOL(ecard_readchunk);
-EXPORT_SYMBOL(__ecard_address);
 EXPORT_SYMBOL(ecard_register_driver);
 EXPORT_SYMBOL(ecard_remove_driver);
 EXPORT_SYMBOL(ecard_bus_type);
diff --git a/arch/arm/kernel/ecard.h b/arch/arm/kernel/ecard.h
index d7c2dacf935..4642d436be2 100644
--- a/arch/arm/kernel/ecard.h
+++ b/arch/arm/kernel/ecard.h
@@ -54,3 +54,16 @@ struct ex_chunk_dir {
 #define c_len(x)	((x)->r_len[0]|((x)->r_len[1]<<8)|((x)->r_len[2]<<16))
 #define c_start(x)	((x)->r_start)
 };
+
+typedef enum ecard_type {		/* Cards address space		*/
+	ECARD_IOC,
+	ECARD_MEMC,
+	ECARD_EASI
+} card_type_t;
+
+typedef enum {				/* Speed for ECARD_IOC space	*/
+	ECARD_SLOW	 = 0,
+	ECARD_MEDIUM	 = 1,
+	ECARD_FAST	 = 2,
+	ECARD_SYNC	 = 3
+} card_speed_t;
diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S
index 597ed00a08d..84694e88b42 100644
--- a/arch/arm/kernel/entry-common.S
+++ b/arch/arm/kernel/entry-common.S
@@ -9,6 +9,7 @@
  */
 
 #include <asm/unistd.h>
+#include <asm/ftrace.h>
 #include <asm/arch/entry-macro.S>
 
 #include "entry-header.S"
@@ -99,6 +100,56 @@ ENTRY(ret_from_fork)
 #undef CALL
 #define CALL(x) .long x
 
+#ifdef CONFIG_FTRACE
+#ifdef CONFIG_DYNAMIC_FTRACE
+ENTRY(mcount)
+	stmdb sp!, {r0-r3, lr}
+	mov r0, lr
+	sub r0, r0, #MCOUNT_INSN_SIZE
+
+	.globl mcount_call
+mcount_call:
+	bl ftrace_stub
+	ldmia sp!, {r0-r3, pc}
+
+ENTRY(ftrace_caller)
+	stmdb sp!, {r0-r3, lr}
+	ldr r1, [fp, #-4]
+	mov r0, lr
+	sub r0, r0, #MCOUNT_INSN_SIZE
+
+	.globl ftrace_call
+ftrace_call:
+	bl ftrace_stub
+	ldmia sp!, {r0-r3, pc}
+
+#else
+
+ENTRY(mcount)
+	stmdb sp!, {r0-r3, lr}
+	ldr r0, =ftrace_trace_function
+	ldr r2, [r0]
+	adr r0, ftrace_stub
+	cmp r0, r2
+	bne trace
+	ldmia sp!, {r0-r3, pc}
+
+trace:
+	ldr r1, [fp, #-4]
+	mov r0, lr
+	sub r0, r0, #MCOUNT_INSN_SIZE
+	mov lr, pc
+	mov pc, r2
+	ldmia sp!, {r0-r3, pc}
+
+#endif /* CONFIG_DYNAMIC_FTRACE */
+
+	.globl ftrace_stub
+ftrace_stub:
+	mov pc, lr
+
+#endif /* CONFIG_FTRACE */
+
 /*=============================================================================
  * SWI handler
  *-----------------------------------------------------------------------------
diff --git a/arch/arm/kernel/ftrace.c b/arch/arm/kernel/ftrace.c
new file mode 100644
index 00000000000..76d50e6091b
--- /dev/null
+++ b/arch/arm/kernel/ftrace.c
@@ -0,0 +1,116 @@
+/*
+ * Dynamic function tracing support.
+ *
+ * Copyright (C) 2008 Abhishek Sagar <sagar.abhishek@gmail.com>
+ *
+ * For licencing details, see COPYING.
+ *
+ * Defines low-level handling of mcount calls when the kernel
+ * is compiled with the -pg flag. When using dynamic ftrace, the
+ * mcount call-sites get patched lazily with NOP till they are
+ * enabled. All code mutation routines here take effect atomically.
+ */
+
+#include <linux/ftrace.h>
+
+#include <asm/cacheflush.h>
+#include <asm/ftrace.h>
+
+#define PC_OFFSET      8
+#define BL_OPCODE      0xeb000000
+#define BL_OFFSET_MASK 0x00ffffff
+
+static unsigned long bl_insn;
+static const unsigned long NOP = 0xe1a00000; /* mov r0, r0 */
+
+unsigned char *ftrace_nop_replace(void)
+{
+	return (char *)&NOP;
+}
+
+/* construct a branch (BL) instruction to addr */
+unsigned char *ftrace_call_replace(unsigned long pc, unsigned long addr)
+{
+	long offset;
+
+	offset = (long)addr - (long)(pc + PC_OFFSET);
+	if (unlikely(offset < -33554432 || offset > 33554428)) {
+		/* Can't generate branches that far (from ARM ARM). Ftrace
+		 * doesn't generate branches outside of kernel text.
+		 */
+		WARN_ON_ONCE(1);
+		return NULL;
+	}
+	offset = (offset >> 2) & BL_OFFSET_MASK;
+	bl_insn = BL_OPCODE | offset;
+	return (unsigned char *)&bl_insn;
+}
+
+int ftrace_modify_code(unsigned long pc, unsigned char *old_code,
+		       unsigned char *new_code)
+{
+	unsigned long err = 0, replaced = 0, old, new;
+
+	old = *(unsigned long *)old_code;
+	new = *(unsigned long *)new_code;
+
+	__asm__ __volatile__ (
+		"1:  ldr    %1, [%2]  \n"
+		"    cmp    %1, %4    \n"
+		"2:  streq  %3, [%2]  \n"
+		"    cmpne  %1, %3    \n"
+		"    movne  %0, #2    \n"
+		"3:\n"
+
+		".section .fixup, \"ax\"\n"
+		"4:  mov  %0, #1  \n"
+		"    b    3b      \n"
+		".previous\n"
+
+		".section __ex_table, \"a\"\n"
+		"    .long 1b, 4b \n"
+		"    .long 2b, 4b \n"
+		".previous\n"
+
+		: "=r"(err), "=r"(replaced)
+		: "r"(pc), "r"(new), "r"(old), "0"(err), "1"(replaced)
+		: "memory");
+
+	if (!err && (replaced == old))
+		flush_icache_range(pc, pc + MCOUNT_INSN_SIZE);
+
+	return err;
+}
+
+int ftrace_update_ftrace_func(ftrace_func_t func)
+{
+	int ret;
+	unsigned long pc, old;
+	unsigned char *new;
+
+	pc = (unsigned long)&ftrace_call;
+	memcpy(&old, &ftrace_call, MCOUNT_INSN_SIZE);
+	new = ftrace_call_replace(pc, (unsigned long)func);
+	ret = ftrace_modify_code(pc, (unsigned char *)&old, new);
+	return ret;
+}
+
+int ftrace_mcount_set(unsigned long *data)
+{
+	unsigned long pc, old;
+	unsigned long *addr = data;
+	unsigned char *new;
+
+	pc = (unsigned long)&mcount_call;
+	memcpy(&old, &mcount_call, MCOUNT_INSN_SIZE);
+	new = ftrace_call_replace(pc, *addr);
+	*addr = ftrace_modify_code(pc, (unsigned char *)&old, new);
+	return 0;
+}
+
+/* run from kstop_machine */
+int __init ftrace_dyn_arch_init(void *data)
+{
+	ftrace_mcount_set(data);
+	return 0;
+}
diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c
index 5593dd20721..5ee39e10c8d 100644
--- a/arch/arm/kernel/kprobes.c
+++ b/arch/arm/kernel/kprobes.c
@@ -274,7 +274,7 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
  * for kretprobe handlers which should normally be interested in r0 only
  * anyway.
  */
-static void __attribute__((naked)) __kprobes kretprobe_trampoline(void)
+void __naked __kprobes kretprobe_trampoline(void)
 {
 	__asm__ __volatile__ (
 		"stmdb	sp!, {r0 - r11}		\n\t"
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 46bf2ede612..199b3680118 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -133,10 +133,8 @@ static void default_idle(void)
 		cpu_relax();
 	else {
 		local_irq_disable();
-		if (!need_resched()) {
-			timer_dyn_reprogram();
+		if (!need_resched())
 			arch_idle();
-		}
 		local_irq_enable();
 	}
 }
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index eefae1de334..5a7c09564d1 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -68,20 +68,10 @@ enum ipi_msg_type {
 	IPI_TIMER,
 	IPI_RESCHEDULE,
 	IPI_CALL_FUNC,
+	IPI_CALL_FUNC_SINGLE,
 	IPI_CPU_STOP,
 };
 
-struct smp_call_struct {
-	void (*func)(void *info);
-	void *info;
-	int wait;
-	cpumask_t pending;
-	cpumask_t unfinished;
-};
-
-static struct smp_call_struct * volatile smp_call_function_data;
-static DEFINE_SPINLOCK(smp_call_function_lock);
-
 int __cpuinit __cpu_up(unsigned int cpu)
 {
 	struct cpuinfo_arm *ci = &per_cpu(cpu_data, cpu);
@@ -366,114 +356,15 @@ static void send_ipi_message(cpumask_t callmap, enum ipi_msg_type msg)
 	local_irq_restore(flags);
 }
 
-/*
- * You must not call this function with disabled interrupts, from a
- * hardware interrupt handler, nor from a bottom half handler.
- */
-static int smp_call_function_on_cpu(void (*func)(void *info), void *info,
-				    int retry, int wait, cpumask_t callmap)
-{
-	struct smp_call_struct data;
-	unsigned long timeout;
-	int ret = 0;
-
-	data.func = func;
-	data.info = info;
-	data.wait = wait;
-
-	cpu_clear(smp_processor_id(), callmap);
-	if (cpus_empty(callmap))
-		goto out;
-
-	data.pending = callmap;
-	if (wait)
-		data.unfinished = callmap;
-
-	/*
-	 * try to get the mutex on smp_call_function_data
-	 */
-	spin_lock(&smp_call_function_lock);
-	smp_call_function_data = &data;
-
-	send_ipi_message(callmap, IPI_CALL_FUNC);
-
-	timeout = jiffies + HZ;
-	while (!cpus_empty(data.pending) && time_before(jiffies, timeout))
-		barrier();
-
-	/*
-	 * did we time out?
-	 */
-	if (!cpus_empty(data.pending)) {
-		/*
-		 * this may be causing our panic - report it
-		 */
-		printk(KERN_CRIT
-		       "CPU%u: smp_call_function timeout for %p(%p)\n"
-		       "      callmap %lx pending %lx, %swait\n",
-		       smp_processor_id(), func, info, *cpus_addr(callmap),
-		       *cpus_addr(data.pending), wait ? "" : "no ");
-
-		/*
-		 * TRACE
-		 */
-		timeout = jiffies + (5 * HZ);
-		while (!cpus_empty(data.pending) && time_before(jiffies, timeout))
-			barrier();
-
-		if (cpus_empty(data.pending))
-			printk(KERN_CRIT "     RESOLVED\n");
-		else
-			printk(KERN_CRIT "     STILL STUCK\n");
-	}
-
-	/*
-	 * whatever happened, we're done with the data, so release it
-	 */
-	smp_call_function_data = NULL;
-	spin_unlock(&smp_call_function_lock);
-
-	if (!cpus_empty(data.pending)) {
-		ret = -ETIMEDOUT;
-		goto out;
-	}
-
-	if (wait)
-		while (!cpus_empty(data.unfinished))
-			barrier();
- out:
-
-	return 0;
-}
-
-int smp_call_function(void (*func)(void *info), void *info, int retry,
-                      int wait)
+void arch_send_call_function_ipi(cpumask_t mask)
 {
-	return smp_call_function_on_cpu(func, info, retry, wait,
-					cpu_online_map);
+	send_ipi_message(mask, IPI_CALL_FUNC);
 }
-EXPORT_SYMBOL_GPL(smp_call_function);
 
-int smp_call_function_single(int cpu, void (*func)(void *info), void *info,
-			     int retry, int wait)
+void arch_send_call_function_single_ipi(int cpu)
 {
-	/* prevent preemption and reschedule on another processor */
-	int current_cpu = get_cpu();
-	int ret = 0;
-
-	if (cpu == current_cpu) {
-		local_irq_disable();
-		func(info);
-		local_irq_enable();
-	} else
-		ret = smp_call_function_on_cpu(func, info, retry, wait,
-					       cpumask_of_cpu(cpu));
-
-	put_cpu();
-
-	return ret;
+	send_ipi_message(cpumask_of_cpu(cpu), IPI_CALL_FUNC_SINGLE);
 }
-EXPORT_SYMBOL_GPL(smp_call_function_single);
 
 void show_ipi_list(struct seq_file *p)
 {
@@ -521,27 +412,6 @@ asmlinkage void __exception do_local_timer(struct pt_regs *regs)
 }
 #endif
 
-/*
- * ipi_call_function - handle IPI from smp_call_function()
- *
- * Note that we copy data out of the cross-call structure and then
- * let the caller know that we're here and have done with their data
- */
-static void ipi_call_function(unsigned int cpu)
-{
-	struct smp_call_struct *data = smp_call_function_data;
-	void (*func)(void *info) = data->func;
-	void *info = data->info;
-	int wait = data->wait;
-
-	cpu_clear(cpu, data->pending);
-
-	func(info);
-
-	if (wait)
-		cpu_clear(cpu, data->unfinished);
-}
-
 static DEFINE_SPINLOCK(stop_lock);
 
 /*
@@ -611,7 +481,11 @@ asmlinkage void __exception do_IPI(struct pt_regs *regs)
 				break;
 
 			case IPI_CALL_FUNC:
-				ipi_call_function(cpu);
+				generic_smp_call_function_interrupt();
+				break;
+
+			case IPI_CALL_FUNC_SINGLE:
+				generic_smp_call_function_single_interrupt();
 				break;
 
 			case IPI_CPU_STOP:
@@ -662,14 +536,13 @@ int setup_profiling_timer(unsigned int multiplier)
 }
 
 static int
-on_each_cpu_mask(void (*func)(void *), void *info, int retry, int wait,
-		 cpumask_t mask)
+on_each_cpu_mask(void (*func)(void *), void *info, int wait, cpumask_t mask)
 {
 	int ret = 0;
 
 	preempt_disable();
 
-	ret = smp_call_function_on_cpu(func, info, retry, wait, mask);
+	ret = smp_call_function_mask(mask, func, info, wait);
 	if (cpu_isset(smp_processor_id(), mask))
 		func(info);
 
@@ -731,14 +604,14 @@ static inline void ipi_flush_tlb_kernel_range(void *arg)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(ipi_flush_tlb_all, NULL, 1, 1);
+	on_each_cpu(ipi_flush_tlb_all, NULL, 1);
 }
 
 void flush_tlb_mm(struct mm_struct *mm)
 {
 	cpumask_t mask = mm->cpu_vm_mask;
 
-	on_each_cpu_mask(ipi_flush_tlb_mm, mm, 1, 1, mask);
+	on_each_cpu_mask(ipi_flush_tlb_mm, mm, 1, mask);
 }
 
 void flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr)
@@ -749,7 +622,7 @@ void flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr)
 	ta.ta_vma = vma;
 	ta.ta_start = uaddr;
 
-	on_each_cpu_mask(ipi_flush_tlb_page, &ta, 1, 1, mask);
+	on_each_cpu_mask(ipi_flush_tlb_page, &ta, 1, mask);
 }
 
 void flush_tlb_kernel_page(unsigned long kaddr)
@@ -758,7 +631,7 @@ void flush_tlb_kernel_page(unsigned long kaddr)
 
 	ta.ta_start = kaddr;
 
-	on_each_cpu(ipi_flush_tlb_kernel_page, &ta, 1, 1);
+	on_each_cpu(ipi_flush_tlb_kernel_page, &ta, 1);
 }
 
 void flush_tlb_range(struct vm_area_struct *vma,
@@ -771,7 +644,7 @@ void flush_tlb_range(struct vm_area_struct *vma,
 	ta.ta_start = start;
 	ta.ta_end = end;
 
-	on_each_cpu_mask(ipi_flush_tlb_range, &ta, 1, 1, mask);
+	on_each_cpu_mask(ipi_flush_tlb_range, &ta, 1, mask);
 }
 
 void flush_tlb_kernel_range(unsigned long start, unsigned long end)
@@ -781,5 +654,5 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 	ta.ta_start = start;
 	ta.ta_end = end;
 
-	on_each_cpu(ipi_flush_tlb_kernel_range, &ta, 1, 1);
+	on_each_cpu(ipi_flush_tlb_kernel_range, &ta, 1);
 }
diff --git a/arch/arm/kernel/stacktrace.c b/arch/arm/kernel/stacktrace.c
index ae31deb2d06..fc650f64df4 100644
--- a/arch/arm/kernel/stacktrace.c
+++ b/arch/arm/kernel/stacktrace.c
@@ -36,6 +36,7 @@ EXPORT_SYMBOL(walk_stackframe);
 #ifdef CONFIG_STACKTRACE
 struct stack_trace_data {
 	struct stack_trace *trace;
+	unsigned int no_sched_functions;
 	unsigned int skip;
 };
 
@@ -43,27 +44,53 @@ static int save_trace(struct stackframe *frame, void *d)
 {
 	struct stack_trace_data *data = d;
 	struct stack_trace *trace = data->trace;
+	unsigned long addr = frame->lr;
 
+	if (data->no_sched_functions && in_sched_functions(addr))
+		return 0;
 	if (data->skip) {
 		data->skip--;
 		return 0;
 	}
 
-	trace->entries[trace->nr_entries++] = frame->lr;
+	trace->entries[trace->nr_entries++] = addr;
 
 	return trace->nr_entries >= trace->max_entries;
 }
 
-void save_stack_trace(struct stack_trace *trace)
+void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 {
 	struct stack_trace_data data;
 	unsigned long fp, base;
 
 	data.trace = trace;
 	data.skip = trace->skip;
-	base = (unsigned long)task_stack_page(current);
-	asm("mov %0, fp" : "=r" (fp));
+	base = (unsigned long)task_stack_page(tsk);
+
+	if (tsk != current) {
+#ifdef CONFIG_SMP
+		/*
+		 * What guarantees do we have here that 'tsk'
+		 * is not running on another CPU?
+		 */
+		BUG();
+#else
+		data.no_sched_functions = 1;
+		fp = thread_saved_fp(tsk);
+#endif
+	} else {
+		data.no_sched_functions = 0;
+		asm("mov %0, fp" : "=r" (fp));
+	}
 
 	walk_stackframe(fp, base, base + THREAD_SIZE, save_trace, &data);
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
+
+void save_stack_trace(struct stack_trace *trace)
+{
+	save_stack_trace_tsk(current, trace);
 }
+EXPORT_SYMBOL_GPL(save_stack_trace);
 #endif
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index b5867eca1d0..cc5145b28e7 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -365,108 +365,6 @@ static struct sysdev_class timer_sysclass = {
 	.resume		= timer_resume,
 };
 
-#ifdef CONFIG_NO_IDLE_HZ
-static int timer_dyn_tick_enable(void)
-{
-	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
-	unsigned long flags;
-	int ret = -ENODEV;
-
-	if (dyn_tick) {
-		spin_lock_irqsave(&dyn_tick->lock, flags);
-		ret = 0;
-		if (!(dyn_tick->state & DYN_TICK_ENABLED)) {
-			ret = dyn_tick->enable();
-
-			if (ret == 0)
-				dyn_tick->state |= DYN_TICK_ENABLED;
-		}
-		spin_unlock_irqrestore(&dyn_tick->lock, flags);
-	}
-
-	return ret;
-}
-
-static int timer_dyn_tick_disable(void)
-{
-	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
-	unsigned long flags;
-	int ret = -ENODEV;
-
-	if (dyn_tick) {
-		spin_lock_irqsave(&dyn_tick->lock, flags);
-		ret = 0;
-		if (dyn_tick->state & DYN_TICK_ENABLED) {
-			ret = dyn_tick->disable();
-
-			if (ret == 0)
-				dyn_tick->state &= ~DYN_TICK_ENABLED;
-		}
-		spin_unlock_irqrestore(&dyn_tick->lock, flags);
-	}
-
-	return ret;
-}
-
-/*
- * Reprogram the system timer for at least the calculated time interval.
- * This function should be called from the idle thread with IRQs disabled,
- * immediately before sleeping.
- */
-void timer_dyn_reprogram(void)
-{
-	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
-	unsigned long next, seq, flags;
-
-	if (!dyn_tick)
-		return;
-
-	spin_lock_irqsave(&dyn_tick->lock, flags);
-	if (dyn_tick->state & DYN_TICK_ENABLED) {
-		next = next_timer_interrupt();
-		do {
-			seq = read_seqbegin(&xtime_lock);
-			dyn_tick->reprogram(next - jiffies);
-		} while (read_seqretry(&xtime_lock, seq));
-	}
-	spin_unlock_irqrestore(&dyn_tick->lock, flags);
-}
-
-static ssize_t timer_show_dyn_tick(struct sys_device *dev, char *buf)
-{
-	return sprintf(buf, "%i\n",
-		       (system_timer->dyn_tick->state & DYN_TICK_ENABLED) >> 1);
-}
-
-static ssize_t timer_set_dyn_tick(struct sys_device *dev, const char *buf,
-				  size_t count)
-{
-	unsigned int enable = simple_strtoul(buf, NULL, 2);
-
-	if (enable)
-		timer_dyn_tick_enable();
-	else
-		timer_dyn_tick_disable();
-
-	return count;
-}
-static SYSDEV_ATTR(dyn_tick, 0644, timer_show_dyn_tick, timer_set_dyn_tick);
-
-/*
- * dyntick=enable|disable
- */
-static char dyntick_str[4] __initdata = "";
-
-static int __init dyntick_setup(char *str)
-{
-	if (str)
-		strlcpy(dyntick_str, str, sizeof(dyntick_str));
-	return 1;
-}
-
-__setup("dyntick=", dyntick_setup);
-#endif
-
 static int __init timer_init_sysfs(void)
 {
 	int ret = sysdev_class_register(&timer_sysclass);
@@ -475,19 +373,6 @@ static int __init timer_init_sysfs(void)
 		ret = sysdev_register(&system_timer->dev);
 	}
 
-#ifdef CONFIG_NO_IDLE_HZ
-	if (ret == 0 && system_timer->dyn_tick) {
-		ret = sysdev_create_file(&system_timer->dev, &attr_dyn_tick);
-
-		/*
-		 * Turn on dynamic tick after calibrate delay
-		 * for correct bogomips
-		 */
-		if (ret == 0 && dyntick_str[0] == 'e')
-			ret = timer_dyn_tick_enable();
-	}
-#endif
-
 	return ret;
 }
 
@@ -500,10 +385,5 @@ void __init time_init(void)
 		system_timer->offset = dummy_gettimeoffset;
 #endif
 	system_timer->init();
-
-#ifdef CONFIG_NO_IDLE_HZ
-	if (system_timer->dyn_tick)
-		spin_lock_init(&system_timer->dyn_tick->lock);
-#endif
 }