Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull perf changes from Ingo Molnar:
 "There are lots of improvements, the biggest changes are:

  Main kernel side changes:

   - Improve uprobes performance by adding 'pre-filtering' support, by
     Oleg Nesterov.

   - Make some POWER7 events available in sysfs, equivalent to what was
     done on x86, from Sukadev Bhattiprolu.

   - tracing updates by Steve Rostedt - mostly misc fixes and smaller
     improvements.

   - Use perf/event tracing to report PCI Express advanced errors, by
     Tony Luck.

   - Enable northbridge performance counters on AMD family 15h, by Jacob
     Shin.

   - This tracing commit:

        tracing: Remove the extra 4 bytes of padding in events

     changes the ABI.  All involved parties (PowerTop in particular)
     seem to agree that it's safe to do now with the introduction of
     libtraceevent, but the devil is in the details ...

  Main tooling side changes:

   - Add 'event group view', from Namyung Kim:

     To use it, 'perf record' should group events when recording.  And
     then perf report parses the saved group relation from file header
     and prints them together if --group option is provided.  You can
     use the 'perf evlist' command to see event group information:

        $ perf record -e '{ref-cycles,cycles}' noploop 1
        [ perf record: Woken up 2 times to write data ]
        [ perf record: Captured and wrote 0.385 MB perf.data (~16807 samples) ]

        $ perf evlist --group
        {ref-cycles,cycles}

     With this example, default perf report will show you each event
     separately.

     You can use --group option to enable event group view:

        $ perf report --group
        ...
        # group: {ref-cycles,cycles}
        # ========
        # Samples: 7K of event 'anon group { ref-cycles, cycles }'
        # Event count (approx.): 6876107743
        #
        #         Overhead  Command      Shared Object                      Symbol
        # ................  .......  .................  ..........................
            99.84%  99.76%  noploop  noploop            [.] main
             0.07%   0.00%  noploop  ld-2.15.so         [.] strcmp
             0.03%   0.00%  noploop  [kernel.kallsyms]  [k] timerqueue_del
             0.03%   0.03%  noploop  [kernel.kallsyms]  [k] sched_clock_cpu
             0.02%   0.00%  noploop  [kernel.kallsyms]  [k] account_user_time
             0.01%   0.00%  noploop  [kernel.kallsyms]  [k] __alloc_pages_nodemask
             0.00%   0.00%  noploop  [kernel.kallsyms]  [k] native_write_msr_safe
             0.00%   0.11%  noploop  [kernel.kallsyms]  [k] _raw_spin_lock
             0.00%   0.06%  noploop  [kernel.kallsyms]  [k] find_get_page
             0.00%   0.02%  noploop  [kernel.kallsyms]  [k] rcu_check_callbacks
             0.00%   0.02%  noploop  [kernel.kallsyms]  [k] __current_kernel_time

     As you can see the Overhead column now contains both of ref-cycles
     and cycles and header line shows group information also - 'anon
     group { ref-cycles, cycles }'.  The output is sorted by period of
     group leader first.

   - Initial GTK+ annotate browser, from Namhyung Kim.

   - Add option for runtime switching perf data file in perf report,
     just press 's' and a menu with the valid files found in the current
     directory will be presented, from Feng Tang.

   - Add support to display whole group data for raw columns, from Jiri
     Olsa.

   - Add per processor socket count aggregation in perf stat, from
     Stephane Eranian.

   - Add interval printing in 'perf stat', from Stephane Eranian.

   - 'perf test' improvements

   - Add support for wildcards in tracepoint system name, from Jiri
     Olsa.

   - Add anonymous huge page recognition, from Joshua Zhu.

   - perf build-id cache now can show DSOs present in a perf.data file
     that are not in the cache, to integrate with build-id servers being
     put in place by organizations such as Fedora.

   - perf top now shares more of the evsel config/creation routines with
     'record', paving the way for further integration like 'top'
     snapshots, etc.

   - perf top now supports DWARF callchains.

   - Fix mmap limitations on 32-bit, fix from David Miller.

   - 'perf bench numa mem' NUMA performance measurement suite

   - ... and lots of fixes, performance improvements, cleanups and other
     improvements I failed to list - see the shortlog and git log for
     details."

* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (270 commits)
  perf/x86/amd: Enable northbridge performance counters on AMD family 15h
  perf/hwbp: Fix cleanup in case of kzalloc failure
  perf tools: Fix build with bison 2.3 and older.
  perf tools: Limit unwind support to x86 archs
  perf annotate: Make it to be able to skip unannotatable symbols
  perf gtk/annotate: Fail early if it can't annotate
  perf gtk/annotate: Show source lines with gray color
  perf gtk/annotate: Support multiple event annotation
  perf ui/gtk: Implement basic GTK2 annotation browser
  perf annotate: Fix warning message on a missing vmlinux
  perf buildid-cache: Add --update option
  uprobes/perf: Avoid uprobe_apply() whenever possible
  uprobes/perf: Teach trace_uprobe/perf code to use UPROBE_HANDLER_REMOVE
  uprobes/perf: Teach trace_uprobe/perf code to pre-filter
  uprobes/perf: Teach trace_uprobe/perf code to track the active perf_event's
  uprobes: Introduce uprobe_apply()
  perf: Introduce hw_perf_event->tp_target and ->tp_list
  uprobes/perf: Always increment trace_uprobe->nhit
  uprobes/tracing: Kill uprobe_trace_consumer, embed uprobe_consumer into trace_uprobe
  uprobes/tracing: Introduce is_trace_uprobe_enabled()
  ...

14 files changed, 725 insertions, 268 deletions

diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 5d89335a485f..36567564e221 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -39,6 +39,9 @@ config HAVE_DYNAMIC_FTRACE
 	help
 	  See Documentation/trace/ftrace-design.txt
 
+config HAVE_DYNAMIC_FTRACE_WITH_REGS
+	bool
+
 config HAVE_FTRACE_MCOUNT_RECORD
 	bool
 	help
@@ -250,6 +253,16 @@ config FTRACE_SYSCALLS
 	help
 	  Basic tracer to catch the syscall entry and exit events.
 
+config TRACER_SNAPSHOT
+	bool "Create a snapshot trace buffer"
+	select TRACER_MAX_TRACE
+	help
+	  Allow tracing users to take snapshot of the current buffer using the
+	  ftrace interface, e.g.:
+
+	      echo 1 > /sys/kernel/debug/tracing/snapshot
+	      cat snapshot
+
 config TRACE_BRANCH_PROFILING
 	bool
 	select GENERIC_TRACER
@@ -434,6 +447,11 @@ config DYNAMIC_FTRACE
 	  were made. If so, it runs stop_machine (stops all CPUS)
 	  and modifies the code to jump over the call to ftrace.
 
+config DYNAMIC_FTRACE_WITH_REGS
+	def_bool y
+	depends on DYNAMIC_FTRACE
+	depends on HAVE_DYNAMIC_FTRACE_WITH_REGS
+
 config FUNCTION_PROFILER
 	bool "Kernel function profiler"
 	depends on FUNCTION_TRACER
diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c
index c0bd0308741c..71259e2b6b61 100644
--- a/kernel/trace/blktrace.c
+++ b/kernel/trace/blktrace.c
@@ -147,7 +147,7 @@ void __trace_note_message(struct blk_trace *bt, const char *fmt, ...)
 		return;
 
 	local_irq_save(flags);
-	buf = per_cpu_ptr(bt->msg_data, smp_processor_id());
+	buf = this_cpu_ptr(bt->msg_data);
 	va_start(args, fmt);
 	n = vscnprintf(buf, BLK_TN_MAX_MSG, fmt, args);
 	va_end(args);
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 41473b4ad7a4..ce8c3d68292f 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -111,6 +111,26 @@ static void ftrace_ops_no_ops(unsigned long ip, unsigned long parent_ip);
 #define ftrace_ops_list_func ((ftrace_func_t)ftrace_ops_no_ops)
 #endif
 
+/*
+ * Traverse the ftrace_global_list, invoking all entries.  The reason that we
+ * can use rcu_dereference_raw() is that elements removed from this list
+ * are simply leaked, so there is no need to interact with a grace-period
+ * mechanism.  The rcu_dereference_raw() calls are needed to handle
+ * concurrent insertions into the ftrace_global_list.
+ *
+ * Silly Alpha and silly pointer-speculation compiler optimizations!
+ */
+#define do_for_each_ftrace_op(op, list)			\
+	op = rcu_dereference_raw(list);			\
+	do
+
+/*
+ * Optimized for just a single item in the list (as that is the normal case).
+ */
+#define while_for_each_ftrace_op(op)				\
+	while (likely(op = rcu_dereference_raw((op)->next)) &&	\
+	       unlikely((op) != &ftrace_list_end))
+
 /**
  * ftrace_nr_registered_ops - return number of ops registered
  *
@@ -132,29 +152,21 @@ int ftrace_nr_registered_ops(void)
 	return cnt;
 }
 
-/*
- * Traverse the ftrace_global_list, invoking all entries.  The reason that we
- * can use rcu_dereference_raw() is that elements removed from this list
- * are simply leaked, so there is no need to interact with a grace-period
- * mechanism.  The rcu_dereference_raw() calls are needed to handle
- * concurrent insertions into the ftrace_global_list.
- *
- * Silly Alpha and silly pointer-speculation compiler optimizations!
- */
 static void
 ftrace_global_list_func(unsigned long ip, unsigned long parent_ip,
 			struct ftrace_ops *op, struct pt_regs *regs)
 {
-	if (unlikely(trace_recursion_test(TRACE_GLOBAL_BIT)))
+	int bit;
+
+	bit = trace_test_and_set_recursion(TRACE_GLOBAL_START, TRACE_GLOBAL_MAX);
+	if (bit < 0)
 		return;
 
-	trace_recursion_set(TRACE_GLOBAL_BIT);
-	op = rcu_dereference_raw(ftrace_global_list); /*see above*/
-	while (op != &ftrace_list_end) {
+	do_for_each_ftrace_op(op, ftrace_global_list) {
 		op->func(ip, parent_ip, op, regs);
-		op = rcu_dereference_raw(op->next); /*see above*/
-	};
-	trace_recursion_clear(TRACE_GLOBAL_BIT);
+	} while_for_each_ftrace_op(op);
+
+	trace_clear_recursion(bit);
 }
 
 static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip,
@@ -221,10 +233,24 @@ static void update_global_ops(void)
 	 * registered callers.
 	 */
 	if (ftrace_global_list == &ftrace_list_end ||
-	    ftrace_global_list->next == &ftrace_list_end)
+	    ftrace_global_list->next == &ftrace_list_end) {
 		func = ftrace_global_list->func;
-	else
+		/*
+		 * As we are calling the function directly.
+		 * If it does not have recursion protection,
+		 * the function_trace_op needs to be updated
+		 * accordingly.
+		 */
+		if (ftrace_global_list->flags & FTRACE_OPS_FL_RECURSION_SAFE)
+			global_ops.flags |= FTRACE_OPS_FL_RECURSION_SAFE;
+		else
+			global_ops.flags &= ~FTRACE_OPS_FL_RECURSION_SAFE;
+	} else {
 		func = ftrace_global_list_func;
+		/* The list has its own recursion protection. */
+		global_ops.flags |= FTRACE_OPS_FL_RECURSION_SAFE;
+	}
+
 
 	/* If we filter on pids, update to use the pid function */
 	if (!list_empty(&ftrace_pids)) {
@@ -337,7 +363,7 @@ static int __register_ftrace_function(struct ftrace_ops *ops)
 	if ((ops->flags & FL_GLOBAL_CONTROL_MASK) == FL_GLOBAL_CONTROL_MASK)
 		return -EINVAL;
 
-#ifndef ARCH_SUPPORTS_FTRACE_SAVE_REGS
+#ifndef CONFIG_DYNAMIC_FTRACE_WITH_REGS
 	/*
 	 * If the ftrace_ops specifies SAVE_REGS, then it only can be used
 	 * if the arch supports it, or SAVE_REGS_IF_SUPPORTED is also set.
@@ -4090,14 +4116,11 @@ ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip,
 	 */
 	preempt_disable_notrace();
 	trace_recursion_set(TRACE_CONTROL_BIT);
-	op = rcu_dereference_raw(ftrace_control_list);
-	while (op != &ftrace_list_end) {
+	do_for_each_ftrace_op(op, ftrace_control_list) {
 		if (!ftrace_function_local_disabled(op) &&
 		    ftrace_ops_test(op, ip))
 			op->func(ip, parent_ip, op, regs);
-
-		op = rcu_dereference_raw(op->next);
-	};
+	} while_for_each_ftrace_op(op);
 	trace_recursion_clear(TRACE_CONTROL_BIT);
 	preempt_enable_notrace();
 }
@@ -4112,27 +4135,26 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *ignored, struct pt_regs *regs)
 {
 	struct ftrace_ops *op;
+	int bit;
 
 	if (function_trace_stop)
 		return;
 
-	if (unlikely(trace_recursion_test(TRACE_INTERNAL_BIT)))
+	bit = trace_test_and_set_recursion(TRACE_LIST_START, TRACE_LIST_MAX);
+	if (bit < 0)
 		return;
 
-	trace_recursion_set(TRACE_INTERNAL_BIT);
 	/*
 	 * Some of the ops may be dynamically allocated,
 	 * they must be freed after a synchronize_sched().
 	 */
 	preempt_disable_notrace();
-	op = rcu_dereference_raw(ftrace_ops_list);
-	while (op != &ftrace_list_end) {
+	do_for_each_ftrace_op(op, ftrace_ops_list) {
 		if (ftrace_ops_test(op, ip))
 			op->func(ip, parent_ip, op, regs);
-		op = rcu_dereference_raw(op->next);
-	};
+	} while_for_each_ftrace_op(op);
 	preempt_enable_notrace();
-	trace_recursion_clear(TRACE_INTERNAL_BIT);
+	trace_clear_recursion(bit);
 }
 
 /*
@@ -4143,8 +4165,8 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip,
  * Archs are to support both the regs and ftrace_ops at the same time.
  * If they support ftrace_ops, it is assumed they support regs.
  * If call backs want to use regs, they must either check for regs
- * being NULL, or ARCH_SUPPORTS_FTRACE_SAVE_REGS.
- * Note, ARCH_SUPPORT_SAVE_REGS expects a full regs to be saved.
+ * being NULL, or CONFIG_DYNAMIC_FTRACE_WITH_REGS.
+ * Note, CONFIG_DYNAMIC_FTRACE_WITH_REGS expects a full regs to be saved.
  * An architecture can pass partial regs with ftrace_ops and still
  * set the ARCH_SUPPORT_FTARCE_OPS.
  */
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index ce8514feedcd..7244acde77b0 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -3,8 +3,10 @@
  *
  * Copyright (C) 2008 Steven Rostedt <srostedt@redhat.com>
  */
+#include <linux/ftrace_event.h>
 #include <linux/ring_buffer.h>
 #include <linux/trace_clock.h>
+#include <linux/trace_seq.h>
 #include <linux/spinlock.h>
 #include <linux/debugfs.h>
 #include <linux/uaccess.h>
@@ -21,7 +23,6 @@
 #include <linux/fs.h>
 
 #include <asm/local.h>
-#include "trace.h"
 
 static void update_pages_handler(struct work_struct *work);
 
@@ -2432,41 +2433,76 @@ rb_reserve_next_event(struct ring_buffer *buffer,
 
 #ifdef CONFIG_TRACING
 
-#define TRACE_RECURSIVE_DEPTH 16
+/*
+ * The lock and unlock are done within a preempt disable section.
+ * The current_context per_cpu variable can only be modified
+ * by the current task between lock and unlock. But it can
+ * be modified more than once via an interrupt. To pass this
+ * information from the lock to the unlock without having to
+ * access the 'in_interrupt()' functions again (which do show
+ * a bit of overhead in something as critical as function tracing,
+ * we use a bitmask trick.
+ *
+ *  bit 0 =  NMI context
+ *  bit 1 =  IRQ context
+ *  bit 2 =  SoftIRQ context
+ *  bit 3 =  normal context.
+ *
+ * This works because this is the order of contexts that can
+ * preempt other contexts. A SoftIRQ never preempts an IRQ
+ * context.
+ *
+ * When the context is determined, the corresponding bit is
+ * checked and set (if it was set, then a recursion of that context
+ * happened).
+ *
+ * On unlock, we need to clear this bit. To do so, just subtract
+ * 1 from the current_context and AND it to itself.
+ *
+ * (binary)
+ *  101 - 1 = 100
+ *  101 & 100 = 100 (clearing bit zero)
+ *
+ *  1010 - 1 = 1001
+ *  1010 & 1001 = 1000 (clearing bit 1)
+ *
+ * The least significant bit can be cleared this way, and it
+ * just so happens that it is the same bit corresponding to
+ * the current context.
+ */
+static DEFINE_PER_CPU(unsigned int, current_context);
 
-/* Keep this code out of the fast path cache */
-static noinline void trace_recursive_fail(void)
+static __always_inline int trace_recursive_lock(void)
 {
-	/* Disable all tracing before we do anything else */
-	tracing_off_permanent();
-
-	printk_once(KERN_WARNING "Tracing recursion: depth[%ld]:"
-		    "HC[%lu]:SC[%lu]:NMI[%lu]\n",
-		    trace_recursion_buffer(),
-		    hardirq_count() >> HARDIRQ_SHIFT,
-		    softirq_count() >> SOFTIRQ_SHIFT,
-		    in_nmi());
-
-	WARN_ON_ONCE(1);
-}
+	unsigned int val = this_cpu_read(current_context);
+	int bit;
 
-static inline int trace_recursive_lock(void)
-{
-	trace_recursion_inc();
+	if (in_interrupt()) {
+		if (in_nmi())
+			bit = 0;
+		else if (in_irq())
+			bit = 1;
+		else
+			bit = 2;
+	} else
+		bit = 3;
 
-	if (likely(trace_recursion_buffer() < TRACE_RECURSIVE_DEPTH))
-		return 0;
+	if (unlikely(val & (1 << bit)))
+		return 1;
 
-	trace_recursive_fail();
+	val |= (1 << bit);
+	this_cpu_write(current_context, val);
 
-	return -1;
+	return 0;
 }
 
-static inline void trace_recursive_unlock(void)
+static __always_inline void trace_recursive_unlock(void)
 {
-	WARN_ON_ONCE(!trace_recursion_buffer());
+	unsigned int val = this_cpu_read(current_context);
 
-	trace_recursion_dec();
+	val--;
+	val &= this_cpu_read(current_context);
+	this_cpu_write(current_context, val);
 }
 
 #else
@@ -3067,6 +3103,24 @@ ring_buffer_dropped_events_cpu(struct ring_buffer *buffer, int cpu)
 EXPORT_SYMBOL_GPL(ring_buffer_dropped_events_cpu);
 
 /**
+ * ring_buffer_read_events_cpu - get the number of events successfully read
+ * @buffer: The ring buffer
+ * @cpu: The per CPU buffer to get the number of events read
+ */
+unsigned long
+ring_buffer_read_events_cpu(struct ring_buffer *buffer, int cpu)
+{
+	struct ring_buffer_per_cpu *cpu_buffer;
+
+	if (!cpumask_test_cpu(cpu, buffer->cpumask))
+		return 0;
+
+	cpu_buffer = buffer->buffers[cpu];
+	return cpu_buffer->read;
+}
+EXPORT_SYMBOL_GPL(ring_buffer_read_events_cpu);
+
+/**
  * ring_buffer_entries - get the number of entries in a buffer
  * @buffer: The ring buffer
  *
@@ -3425,7 +3479,7 @@ static void rb_advance_iter(struct ring_buffer_iter *iter)
 	/* check for end of page padding */
 	if ((iter->head >= rb_page_size(iter->head_page)) &&
 	    (iter->head_page != cpu_buffer->commit_page))
-		rb_advance_iter(iter);
+		rb_inc_iter(iter);
 }
 
 static int rb_lost_events(struct ring_buffer_per_cpu *cpu_buffer)
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 3c13e46d7d24..5d520b7bb4c5 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -249,7 +249,7 @@ static unsigned long		trace_buf_size = TRACE_BUF_SIZE_DEFAULT;
 static struct tracer		*trace_types __read_mostly;
 
 /* current_trace points to the tracer that is currently active */
-static struct tracer		*current_trace __read_mostly;
+static struct tracer		*current_trace __read_mostly = &nop_trace;
 
 /*
  * trace_types_lock is used to protect the trace_types list.
@@ -709,10 +709,13 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
 		return;
 
 	WARN_ON_ONCE(!irqs_disabled());
-	if (!current_trace->use_max_tr) {
-		WARN_ON_ONCE(1);
+
+	if (!current_trace->allocated_snapshot) {
+		/* Only the nop tracer should hit this when disabling */
+		WARN_ON_ONCE(current_trace != &nop_trace);
 		return;
 	}
+
 	arch_spin_lock(&ftrace_max_lock);
 
 	tr->buffer = max_tr.buffer;
@@ -739,10 +742,8 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
 		return;
 
 	WARN_ON_ONCE(!irqs_disabled());
-	if (!current_trace->use_max_tr) {
-		WARN_ON_ONCE(1);
+	if (WARN_ON_ONCE(!current_trace->allocated_snapshot))
 		return;
-	}
 
 	arch_spin_lock(&ftrace_max_lock);
 
@@ -862,10 +863,13 @@ int register_tracer(struct tracer *type)
 
 		current_trace = type;
 
-		/* If we expanded the buffers, make sure the max is expanded too */
-		if (ring_buffer_expanded && type->use_max_tr)
-			ring_buffer_resize(max_tr.buffer, trace_buf_size,
-						RING_BUFFER_ALL_CPUS);
+		if (type->use_max_tr) {
+			/* If we expanded the buffers, make sure the max is expanded too */
+			if (ring_buffer_expanded)
+				ring_buffer_resize(max_tr.buffer, trace_buf_size,
+						   RING_BUFFER_ALL_CPUS);
+			type->allocated_snapshot = true;
+		}
 
 		/* the test is responsible for initializing and enabling */
 		pr_info("Testing tracer %s: ", type->name);
@@ -881,10 +885,14 @@ int register_tracer(struct tracer *type)
 		/* Only reset on passing, to avoid touching corrupted buffers */
 		tracing_reset_online_cpus(tr);
 
-		/* Shrink the max buffer again */
-		if (ring_buffer_expanded && type->use_max_tr)
-			ring_buffer_resize(max_tr.buffer, 1,
-						RING_BUFFER_ALL_CPUS);
+		if (type->use_max_tr) {
+			type->allocated_snapshot = false;
+
+			/* Shrink the max buffer again */
+			if (ring_buffer_expanded)
+				ring_buffer_resize(max_tr.buffer, 1,
+						   RING_BUFFER_ALL_CPUS);
+		}
 
 		printk(KERN_CONT "PASSED\n");
 	}
@@ -922,6 +930,9 @@ void tracing_reset(struct trace_array *tr, int cpu)
 {
 	struct ring_buffer *buffer = tr->buffer;
 
+	if (!buffer)
+		return;
+
 	ring_buffer_record_disable(buffer);
 
 	/* Make sure all commits have finished */
@@ -936,6 +947,9 @@ void tracing_reset_online_cpus(struct trace_array *tr)
 	struct ring_buffer *buffer = tr->buffer;
 	int cpu;
 
+	if (!buffer)
+		return;
+
 	ring_buffer_record_disable(buffer);
 
 	/* Make sure all commits have finished */
@@ -1167,7 +1181,6 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
 
 	entry->preempt_count		= pc & 0xff;
 	entry->pid			= (tsk) ? tsk->pid : 0;
-	entry->padding			= 0;
 	entry->flags =
 #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
 		(irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) |
@@ -1335,7 +1348,7 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer,
 	 */
 	preempt_disable_notrace();
 
-	use_stack = ++__get_cpu_var(ftrace_stack_reserve);
+	use_stack = __this_cpu_inc_return(ftrace_stack_reserve);
 	/*
 	 * We don't need any atomic variables, just a barrier.
 	 * If an interrupt comes in, we don't care, because it would
@@ -1389,7 +1402,7 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer,
  out:
 	/* Again, don't let gcc optimize things here */
 	barrier();
-	__get_cpu_var(ftrace_stack_reserve)--;
+	__this_cpu_dec(ftrace_stack_reserve);
 	preempt_enable_notrace();
 
 }
@@ -1517,7 +1530,6 @@ static struct trace_buffer_struct *trace_percpu_nmi_buffer;
 static char *get_trace_buf(void)
 {
 	struct trace_buffer_struct *percpu_buffer;
-	struct trace_buffer_struct *buffer;
 
 	/*
 	 * If we have allocated per cpu buffers, then we do not
@@ -1535,9 +1547,7 @@ static char *get_trace_buf(void)
 	if (!percpu_buffer)
 		return NULL;
 
-	buffer = per_cpu_ptr(percpu_buffer, smp_processor_id());
-
-	return buffer->buffer;
+	return this_cpu_ptr(&percpu_buffer->buffer[0]);
 }
 
 static int alloc_percpu_trace_buffer(void)
@@ -1942,21 +1952,27 @@ void tracing_iter_reset(struct trace_iterator *iter, int cpu)
 static void *s_start(struct seq_file *m, loff_t *pos)
 {
 	struct trace_iterator *iter = m->private;
-	static struct tracer *old_tracer;
 	int cpu_file = iter->cpu_file;
 	void *p = NULL;
 	loff_t l = 0;
 	int cpu;
 
-	/* copy the tracer to avoid using a global lock all around */
+	/*
+	 * copy the tracer to avoid using a global lock all around.
+	 * iter->trace is a copy of current_trace, the pointer to the
+	 * name may be used instead of a strcmp(), as iter->trace->name
+	 * will point to the same string as current_trace->name.
+	 */
 	mutex_lock(&trace_types_lock);
-	if (unlikely(old_tracer != current_trace && current_trace)) {
-		old_tracer = current_trace;
+	if (unlikely(current_trace && iter->trace->name != current_trace->name))
 		*iter->trace = *current_trace;
-	}
 	mutex_unlock(&trace_types_lock);
 
-	atomic_inc(&trace_record_cmdline_disabled);
+	if (iter->snapshot && iter->trace->use_max_tr)
+		return ERR_PTR(-EBUSY);
+
+	if (!iter->snapshot)
+		atomic_inc(&trace_record_cmdline_disabled);
 
 	if (*pos != iter->pos) {
 		iter->ent = NULL;
@@ -1995,7 +2011,11 @@ static void s_stop(struct seq_file *m, void *p)
 {
 	struct trace_iterator *iter = m->private;
 
-	atomic_dec(&trace_record_cmdline_disabled);
+	if (iter->snapshot && iter->trace->use_max_tr)
+		return;
+
+	if (!iter->snapshot)
+		atomic_dec(&trace_record_cmdline_disabled);
 	trace_access_unlock(iter->cpu_file);
 	trace_event_read_unlock();
 }
@@ -2080,8 +2100,7 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter)
 	unsigned long total;
 	const char *name = "preemption";
 
-	if (type)
-		name = type->name;
+	name = type->name;
 
 	get_total_entries(tr, &total, &entries);
 
@@ -2430,7 +2449,7 @@ static const struct seq_operations tracer_seq_ops = {
 };
 
 static struct trace_iterator *
-__tracing_open(struct inode *inode, struct file *file)
+__tracing_open(struct inode *inode, struct file *file, bool snapshot)
 {
 	long cpu_file = (long) inode->i_private;
 	struct trace_iterator *iter;
@@ -2457,16 +2476,16 @@ __tracing_open(struct inode *inode, struct file *file)
 	if (!iter->trace)
 		goto fail;
 
-	if (current_trace)
-		*iter->trace = *current_trace;
+	*iter->trace = *current_trace;
 
 	if (!zalloc_cpumask_var(&iter->started, GFP_KERNEL))
 		goto fail;
 
-	if (current_trace && current_trace->print_max)
+	if (current_trace->print_max || snapshot)
 		iter->tr = &max_tr;
 	else
 		iter->tr = &global_trace;
+	iter->snapshot = snapshot;
 	iter->pos = -1;
 	mutex_init(&iter->mutex);
 	iter->cpu_file = cpu_file;
@@ -2483,8 +2502,9 @@ __tracing_open(struct inode *inode, struct file *file)
 	if (trace_clocks[trace_clock_id].in_ns)
 		iter->iter_flags |= TRACE_FILE_TIME_IN_NS;
 
-	/* stop the trace while dumping */
-	tracing_stop();
+	/* stop the trace while dumping if we are not opening "snapshot" */
+	if (!iter->snapshot)
+		tracing_stop();
 
 	if (iter->cpu_file == TRACE_PIPE_ALL_CPU) {
 		for_each_tracing_cpu(cpu) {
@@ -2547,8 +2567,9 @@ static int tracing_release(struct inode *inode, struct file *file)
 	if (iter->trace && iter->trace->close)
 		iter->trace->close(iter);
 
-	/* reenable tracing if it was previously enabled */
-	tracing_start();
+	if (!iter->snapshot)
+		/* reenable tracing if it was previously enabled */
+		tracing_start();
 	mutex_unlock(&trace_types_lock);
 
 	mutex_destroy(&iter->mutex);
@@ -2576,7 +2597,7 @@ static int tracing_open(struct inode *inode, struct file *file)
 	}
 
 	if (file->f_mode & FMODE_READ) {
-		iter = __tracing_open(inode, file);
+		iter = __tracing_open(inode, file, false);
 		if (IS_ERR(iter))
 			ret = PTR_ERR(iter);
 		else if (trace_flags & TRACE_ITER_LATENCY_FMT)
@@ -3014,10 +3035,7 @@ tracing_set_trace_read(struct file *filp, char __user *ubuf,
 	int r;
 
 	mutex_lock(&trace_types_lock);
-	if (current_trace)
-		r = sprintf(buf, "%s\n", current_trace->name);
-	else
-		r = sprintf(buf, "\n");
+	r = sprintf(buf, "%s\n", current_trace->name);
 	mutex_unlock(&trace_types_lock);
 
 	return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
@@ -3183,6 +3201,7 @@ static int tracing_set_tracer(const char *buf)
 	static struct trace_option_dentry *topts;
 	struct trace_array *tr = &global_trace;
 	struct tracer *t;
+	bool had_max_tr;
 	int ret = 0;
 
 	mutex_lock(&trace_types_lock);
@@ -3207,9 +3226,21 @@ static int tracing_set_tracer(const char *buf)
 		goto out;
 
 	trace_branch_disable();
-	if (current_trace && current_trace->reset)
+	if (current_trace->reset)
 		current_trace->reset(tr);
-	if (current_trace && current_trace->use_max_tr) {
+
+	had_max_tr = current_trace->allocated_snapshot;
+	current_trace = &nop_trace;
+
+	if (had_max_tr && !t->use_max_tr) {
+		/*
+		 * We need to make sure that the update_max_tr sees that
+		 * current_trace changed to nop_trace to keep it from
+		 * swapping the buffers after we resize it.
+		 * The update_max_tr is called from interrupts disabled
+		 * so a synchronized_sched() is sufficient.
+		 */
+		synchronize_sched();
 		/*
 		 * We don't free the ring buffer. instead, resize it because
 		 * The max_tr ring buffer has some state (e.g. ring->clock) and
@@ -3217,18 +3248,19 @@ static int tracing_set_tracer(const char *buf)
 		 */
 		ring_buffer_resize(max_tr.buffer, 1, RING_BUFFER_ALL_CPUS);
 		set_buffer_entries(&max_tr, 1);
+		tracing_reset_online_cpus(&max_tr);
+		current_trace->allocated_snapshot = false;
 	}
 	destroy_trace_option_files(topts);
 
-	current_trace = &nop_trace;
-
 	topts = create_trace_option_files(t);
-	if (t->use_max_tr) {
+	if (t->use_max_tr && !had_max_tr) {
 		/* we need to make per cpu buffer sizes equivalent */
 		ret = resize_buffer_duplicate_size(&max_tr, &global_trace,
 						   RING_BUFFER_ALL_CPUS);
 		if (ret < 0)
 			goto out;
+		t->allocated_snapshot = true;
 	}
 
 	if (t->init) {
@@ -3336,8 +3368,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp)
 		ret = -ENOMEM;
 		goto fail;
 	}
-	if (current_trace)
-		*iter->trace = *current_trace;
+	*iter->trace = *current_trace;
 
 	if (!alloc_cpumask_var(&iter->started, GFP_KERNEL)) {
 		ret = -ENOMEM;
@@ -3477,7 +3508,6 @@ tracing_read_pipe(struct file *filp, char __user *ubuf,
 		  size_t cnt, loff_t *ppos)
 {
 	struct trace_iterator *iter = filp->private_data;
-	static struct tracer *old_tracer;
 	ssize_t sret;
 
 	/* return any leftover data */
@@ -3489,10 +3519,8 @@ tracing_read_pipe(struct file *filp, char __user *ubuf,
 
 	/* copy the tracer to avoid using a global lock all around */
 	mutex_lock(&trace_types_lock);
-	if (unlikely(old_tracer != current_trace && current_trace)) {
-		old_tracer = current_trace;
+	if (unlikely(iter->trace->name != current_trace->name))
 		*iter->trace = *current_trace;
-	}
 	mutex_unlock(&trace_types_lock);
 
 	/*
@@ -3648,7 +3676,6 @@ static ssize_t tracing_splice_read_pipe(struct file *filp,
 		.ops		= &tracing_pipe_buf_ops,
 		.spd_release	= tracing_spd_release_pipe,
 	};
-	static struct tracer *old_tracer;
 	ssize_t ret;
 	size_t rem;
 	unsigned int i;
@@ -3658,10 +3685,8 @@ static ssize_t tracing_splice_read_pipe(struct file *filp,
 
 	/* copy the tracer to avoid using a global lock all around */
 	mutex_lock(&trace_types_lock);
-	if (unlikely(old_tracer != current_trace && current_trace)) {
-		old_tracer = current_trace;
+	if (unlikely(iter->trace->name != current_trace->name))
 		*iter->trace = *current_trace;
-	}
 	mutex_unlock(&trace_types_lock);
 
 	mutex_lock(&iter->mutex);
@@ -4037,8 +4062,7 @@ static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf,
 	 * Reset the buffer so that it doesn't have incomparable timestamps.
 	 */
 	tracing_reset_online_cpus(&global_trace);
-	if (max_tr.buffer)
-		tracing_reset_online_cpus(&max_tr);
+	tracing_reset_online_cpus(&max_tr);
 
 	mutex_unlock(&trace_types_lock);
 
@@ -4054,6 +4078,87 @@ static int tracing_clock_open(struct inode *inode, struct file *file)
 	return single_open(file, tracing_clock_show, NULL);
 }
 
+#ifdef CONFIG_TRACER_SNAPSHOT
+static int tracing_snapshot_open(struct inode *inode, struct file *file)
+{
+	struct trace_iterator *iter;
+	int ret = 0;
+
+	if (file->f_mode & FMODE_READ) {
+		iter = __tracing_open(inode, file, true);
+		if (IS_ERR(iter))
+			ret = PTR_ERR(iter);
+	}
+	return ret;
+}
+
+static ssize_t
+tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt,
+		       loff_t *ppos)
+{
+	unsigned long val;
+	int ret;
+
+	ret = tracing_update_buffers();
+	if (ret < 0)
+		return ret;
+
+	ret = kstrtoul_from_user(ubuf, cnt, 10, &val);
+	if (ret)
+		return ret;
+
+	mutex_lock(&trace_types_lock);
+
+	if (current_trace->use_max_tr) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	switch (val) {
+	case 0:
+		if (current_trace->allocated_snapshot) {
+			/* free spare buffer */
+			ring_buffer_resize(max_tr.buffer, 1,
+					   RING_BUFFER_ALL_CPUS);
+			set_buffer_entries(&max_tr, 1);
+			tracing_reset_online_cpus(&max_tr);
+			current_trace->allocated_snapshot = false;
+		}
+		break;
+	case 1:
+		if (!current_trace->allocated_snapshot) {
+			/* allocate spare buffer */
+			ret = resize_buffer_duplicate_size(&max_tr,
+					&global_trace, RING_BUFFER_ALL_CPUS);
+			if (ret < 0)
+				break;
+			current_trace->allocated_snapshot = true;
+		}
+
+		local_irq_disable();
+		/* Now, we're going to swap */
+		update_max_tr(&global_trace, current, smp_processor_id());
+		local_irq_enable();
+		break;
+	default:
+		if (current_trace->allocated_snapshot)
+			tracing_reset_online_cpus(&max_tr);
+		else
+			ret = -EINVAL;