linux-toradex.git/kernel/trace/trace_event_profile.c, branch v2.6.33.1

tracing: Change event->profile_count to be int type

2009-12-13T17:37:28+00:00

Like total_profile_count, struct ftrace_event_call::profile_count
is protected by event_mutex, so it doesn't need to be atomic_t.

Signed-off-by: Li Zefan 
Acked-by: Steven Rostedt 
Cc: Jason Baron 
Cc: Masami Hiramatsu 
Cc: Peter Zijlstra 
LKML-Reference: <4B1DC549.5010705@cn.fujitsu.com>
Signed-off-by: Frederic Weisbecker

tracing: Use the perf recursion protection from trace event

2009-11-22T08:03:42+00:00

When we commit a trace to perf, we first check if we are
recursing in the same buffer so that we don't mess-up the buffer
with a recursing trace. But later on, we do the same check from
perf to avoid commit recursion. The recursion check is desired
early before we touch the buffer but we want to do this check
only once.

Then export the recursion protection from perf and use it from
the trace events before submitting a trace.

v2: Put appropriate Reported-by tag

Reported-by: Peter Zijlstra 
Signed-off-by: Frederic Weisbecker 
Cc: Arnaldo Carvalho de Melo 
Cc: Paul Mackerras 
Cc: Steven Rostedt 
Cc: Masami Hiramatsu 
Cc: Jason Baron 
LKML-Reference: <1258864015-10579-1-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar

tracing, perf_events: Protect the buffer from recursion in perf

2009-11-08T09:31:42+00:00

While tracing using events with perf, if one enables the
lockdep:lock_acquire event, it will infect every other perf
trace events.

Basically, you can enable whatever set of trace events through
perf but if this event is part of the set, the only result we
can get is a long list of lock_acquire events of rcu read lock,
and only that.

This is because of a recursion inside perf.

1) When a trace event is triggered, it will fill a per cpu
   buffer and submit it to perf.

2) Perf will commit this event but will also protect some data
   using rcu_read_lock

3) A recursion appears: rcu_read_lock triggers a lock_acquire
   event that will fill the per cpu event and then submit the
   buffer to perf.

4) Perf detects a recursion and ignores it

5) Perf continues its work on the previous event, but its buffer
   has been overwritten by the lock_acquire event, it has then
   been turned into a lock_acquire event of rcu read lock

Such scenario also happens with lock_release with
rcu_read_unlock().

We could turn the rcu_read_lock() into __rcu_read_lock() to drop
the lock debugging from perf fast path, but that would make us
lose the rcu debugging and that doesn't prevent from other
possible kind of recursion from perf in the future.

This patch adds a recursion protection based on a counter on the
perf trace per cpu buffers to solve the problem.

-v2: Fixed lost whitespace, added reviewed-by tag

Signed-off-by: Frederic Weisbecker 
Reviewed-by: Masami Hiramatsu 
Cc: Peter Zijlstra 
Cc: Arnaldo Carvalho de Melo 
Cc: Mike Galbraith 
Cc: Paul Mackerras 
Cc: Steven Rostedt 
Cc: Li Zefan 
Cc: Jason Baron 
LKML-Reference: <1257477185-7838-1-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar

Merge commit 'v2.6.32-rc5' into perf/probes

2009-10-17T07:58:25+00:00

Conflicts:
	kernel/trace/trace_event_profile.c

Merge reason: update to -rc5 and resolve conflict.

Signed-off-by: Ingo Molnar

tracing: Use free_percpu instead of kfree

2009-10-05T08:57:56+00:00

In the event->profile_enable() failure path, we release the per cpu
buffers using kfree which is wrong because they are per cpu pointers.
Although free_percpu only wraps kfree for now, that may change in the
future so lets use the correct way.

Reported-by: Paul Mackerras 
Signed-off-by: Frederic Weisbecker 
Cc: Steven Rostedt 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Cc: Li Zefan

tracing: Check total refcount before releasing bufs in profile_enable failure

2009-10-05T08:57:41+00:00

When we call the profile_enable() callback of an event, we release the
shared perf event tracing buffers unconditionnaly in the failure path.
This is wrong because there may be other users of these. Then check the
total refcount before doing this.

Reported-by: Paul Mackerras 
Signed-off-by: Frederic Weisbecker 
Cc: Steven Rostedt 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Cc: Li Zefan

Merge commit 'linus/master' into tracing/kprobes

2009-09-23T21:08:43+00:00

Conflicts:
	kernel/trace/Makefile
	kernel/trace/trace.h
	kernel/trace/trace_event_types.h
	kernel/trace/trace_export.c

Merge reason:
	Sync with latest significant tracing core changes.

tracing: Export trace_profile_buf symbols

2009-09-20T17:15:57+00:00

ERROR: "trace_profile_buf_nmi" [fs/jbd2/jbd2.ko] undefined!
ERROR: "trace_profile_buf" [fs/jbd2/jbd2.ko] undefined!
ERROR: "trace_profile_buf_nmi" [fs/ext4/ext4.ko] undefined!
ERROR: "trace_profile_buf" [fs/ext4/ext4.ko] undefined!
ERROR: "trace_profile_buf_nmi" [arch/x86/kvm/kvm.ko] undefined!
ERROR: "trace_profile_buf" [arch/x86/kvm/kvm.ko] undefined!

Signed-off-by: Peter Zijlstra 
Acked-by: Steven Rostedt 
Cc: Frederic Weisbecker 
LKML-Reference: <1253442878.7542.3.camel@laptop>
[ fixed whitespace noise and checkpatch complaint ]
Signed-off-by: Ingo Molnar

tracing: Allocate the ftrace event profile buffer dynamically

2009-09-18T05:25:44+00:00

Currently the trace event profile buffer is allocated in the stack. But
this may be too much for the stack, as the events can have large
statically defined field size and can also grow with dynamic arrays.

Allocate two per cpu buffer for all profiled events. The first cpu
buffer is used to host every non-nmi context traces. It is protected
by disabling the interrupts while writing and committing the trace.

The second buffer is reserved for nmi. So that there is no race between
them and the first buffer.

The whole write/commit section is rcu protected because we release
these buffers while deactivating the last profiling trace event.

v2: Move the buffers from trace_event to be global, as pointed by
    Steven Rostedt.

v3: Fix the syscall events to handle the profiling buffer races
    by disabling interrupts, now that the buffers are globals.

Suggested-by: Steven Rostedt 
Signed-off-by: Frederic Weisbecker 
Cc: Steven Rostedt 
Cc: Peter Zijlstra 
Cc: Li Zefan 
Cc: Jason Baron 
Cc: Masami Hiramatsu 
Cc: Ingo Molnar

tracing: Factorize the events profile accounting

2009-09-18T04:14:32+00:00

Factorize the events enabling accounting in a common tracing core
helper. This reduces the size of the profile_enable() and
profile_disable() callbacks for each trace events.

Signed-off-by: Frederic Weisbecker 
Acked-by: Steven Rostedt 
Acked-by: Li Zefan 
Cc: Peter Zijlstra 
Cc: Jason Baron 
Cc: Masami Hiramatsu 
Cc: Ingo Molnar