Hello, Steven
Thanks for your comments for v1 patch.
I send v2 patch which includes your suggestion.
B.R
Zhaolei
We need check sched_stopped in probe_sched_wakeup() to stop tracing when an
user stops it.
But current code can also do the right thing(stop tracing after
echo 0 > tracing_enabled) just because tracing_sched_wakeup_trace() can return
on buffer->record_disabled.
Signed-off-by: Zhao Lei <[email protected]>
---
kernel/trace/trace_sched_switch.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index de35f20..9117cea 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -62,6 +62,9 @@ probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success)
pc = preempt_count();
tracing_record_cmdline(current);
+ if (sched_stopped)
+ return;
+
local_irq_save(flags);
cpu = raw_smp_processor_id();
data = ctx_trace->data[cpu];
--
1.5.5.3
Unify sched_switch and sched_wakeup's action to following logic:
Do record_cmdline when start_cmdline_record() called.
Trace event when trace is inited and started.
Signed-off-by: Zhao Lei <[email protected]>
---
kernel/trace/trace_sched_switch.c | 10 +++++-----
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index 9117cea..9d8cccd 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -29,13 +29,13 @@ probe_sched_switch(struct rq *__rq, struct task_struct *prev,
int cpu;
int pc;
- if (!sched_ref || sched_stopped)
+ if (unlikely(!sched_ref))
return;
tracing_record_cmdline(prev);
tracing_record_cmdline(next);
- if (!tracer_enabled)
+ if (!tracer_enabled || sched_stopped)
return;
pc = preempt_count();
@@ -56,15 +56,15 @@ probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success)
unsigned long flags;
int cpu, pc;
- if (!likely(tracer_enabled))
+ if (unlikely(!sched_ref))
return;
- pc = preempt_count();
tracing_record_cmdline(current);
- if (sched_stopped)
+ if (!tracer_enabled || sched_stopped)
return;
+ pc = preempt_count();
local_irq_save(flags);
cpu = raw_smp_processor_id();
data = ctx_trace->data[cpu];
--
1.5.5.3
On Tue, 31 Mar 2009, Zhaolei wrote:
> We need check sched_stopped in probe_sched_wakeup() to stop tracing when an
> user stops it.
>
> But current code can also do the right thing(stop tracing after
> echo 0 > tracing_enabled) just because tracing_sched_wakeup_trace() can return
> on buffer->record_disabled.
>
> Signed-off-by: Zhao Lei <[email protected]>
Thanks, applied.
-- Steve
On Tue, 31 Mar 2009, Zhaolei wrote:
> Unify sched_switch and sched_wakeup's action to following logic:
> Do record_cmdline when start_cmdline_record() called.
> Trace event when trace is inited and started.
>
> Signed-off-by: Zhao Lei <[email protected]>
Thanks, applied.
-- Steve
Commit-ID: 8bcae09b93e7f96f700b6bb372c2b3f2b36636dc
Gitweb: http://git.kernel.org/tip/8bcae09b93e7f96f700b6bb372c2b3f2b36636dc
Author: Zhaolei <[email protected]>
AuthorDate: Tue, 31 Mar 2009 15:24:51 +0800
Committer: Ingo Molnar <[email protected]>
CommitDate: Tue, 7 Apr 2009 14:01:11 +0200
ftrace: Add check of sched_stopped for probe_sched_wakeup
The wakeup tracing in sched_switch does not stop when a user
disables tracing. This is because the probe_sched_wakeup() is missing
the check to prevent the wakeup from being traced.
Signed-off-by: Zhao Lei <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Steven Rostedt <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
---
kernel/trace/trace_sched_switch.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index de35f20..9117cea 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -62,6 +62,9 @@ probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success)
pc = preempt_count();
tracing_record_cmdline(current);
+ if (sched_stopped)
+ return;
+
local_irq_save(flags);
cpu = raw_smp_processor_id();
data = ctx_trace->data[cpu];
Commit-ID: dcef788eb9659b61a2110284fcce3ca6e63480d2
Gitweb: http://git.kernel.org/tip/dcef788eb9659b61a2110284fcce3ca6e63480d2
Author: Zhaolei <[email protected]>
AuthorDate: Tue, 31 Mar 2009 15:26:14 +0800
Committer: Ingo Molnar <[email protected]>
CommitDate: Tue, 7 Apr 2009 14:43:09 +0200
ftrace: clean up enable logic for sched_switch
Unify sched_switch and sched_wakeup's action to following logic:
Do record_cmdline when start_cmdline_record() is called.
Start tracing events when the tracer is started.
Signed-off-by: Zhao Lei <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Steven Rostedt <[email protected]>
---
kernel/trace/trace_sched_switch.c | 10 +++++-----
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index 9117cea..9d8cccd 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -29,13 +29,13 @@ probe_sched_switch(struct rq *__rq, struct task_struct *prev,
int cpu;
int pc;
- if (!sched_ref || sched_stopped)
+ if (unlikely(!sched_ref))
return;
tracing_record_cmdline(prev);
tracing_record_cmdline(next);
- if (!tracer_enabled)
+ if (!tracer_enabled || sched_stopped)
return;
pc = preempt_count();
@@ -56,15 +56,15 @@ probe_sched_wakeup(struct rq *__rq, struct task_struct *wakee, int success)
unsigned long flags;
int cpu, pc;
- if (!likely(tracer_enabled))
+ if (unlikely(!sched_ref))
return;
- pc = preempt_count();
tracing_record_cmdline(current);
- if (sched_stopped)
+ if (!tracer_enabled || sched_stopped)
return;
+ pc = preempt_count();
local_irq_save(flags);
cpu = raw_smp_processor_id();
data = ctx_trace->data[cpu];