lkml.org 
[lkml]   [2008]   [Nov]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[PATCH] tracing/function-return-tracer: clean up task start/exit callbacks

* Frederic Weisbecker <fweisbec@gmail.com> wrote:

> Impact: use deeper tracing depth safely
>
> Some tests showed that function return tracing needed a more deeper depth
> of function calls. But it could be unsafe to store these return addresses
> to the stack.
> So these arrays will now be allocated dynamically into task_struct of current
> only when the tracer is activated.
>
> Typical scheme when tracer is activated:
> _ allocate a return stack for each task in global list.
> _ fork: allocate the return stack for the newly created task
> _ exit: free return stack of current
> _ idle init: same as fork
>
> I chose a default depth of 50. I don't have anymore overruns (didn't tested for
> a long time).
>
> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> ---
> arch/x86/include/asm/ftrace.h | 1 -
> arch/x86/include/asm/thread_info.h | 29 -----------
> arch/x86/kernel/ftrace.c | 29 ++++++-----
> include/linux/ftrace.h | 5 ++
> include/linux/sched.h | 23 ++++-----
> kernel/exit.c | 5 ++-
> kernel/fork.c | 4 ++
> kernel/sched.c | 3 +
> kernel/trace/ftrace.c | 96 +++++++++++++++++++++++++++++++++++-
> 9 files changed, 137 insertions(+), 58 deletions(-)

nice! Applied to tip/tracing/function-return-tracer and i started
testing it as well.

I've done the small cleanup below on top of your patch - we try to
avoid #ifdefs in core kernel code as much as possible, by hiding
.config details into include files.

Ingo

-------------->
From 82f60f0bc854aada696f27d863c03bef91f1509d Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 23 Nov 2008 09:18:56 +0100
Subject: [PATCH] tracing/function-return-tracer: clean up task start/exit callbacks

Impact: cleanup

Eliminate #ifdefs in core code by using empty inline functions.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
include/linux/ftrace.h | 3 +++
kernel/exit.c | 2 --
kernel/fork.c | 2 --
kernel/sched.c | 2 --
4 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 2ba259b..938ca19 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -335,6 +335,9 @@ extern void unregister_ftrace_return(void);

extern void ftrace_retfunc_init_task(struct task_struct *t);
extern void ftrace_retfunc_exit_task(struct task_struct *t);
+#else
+static inline void ftrace_retfunc_init_task(struct task_struct *t) { }
+static inline void ftrace_retfunc_exit_task(struct task_struct *t) { }
#endif

#endif /* _LINUX_FTRACE_H */
diff --git a/kernel/exit.c b/kernel/exit.c
index b9d4463..ef04d03 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -1128,9 +1128,7 @@ NORET_TYPE void do_exit(long code)
preempt_disable();
/* causes final put_task_struct in finish_task_switch(). */
tsk->state = TASK_DEAD;
-#ifdef CONFIG_FUNCTION_RET_TRACER
ftrace_retfunc_exit_task(tsk);
-#endif
schedule();
BUG();
/* Avoid "noreturn function does return". */
diff --git a/kernel/fork.c b/kernel/fork.c
index d1eb30e..fbf4a4c 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1270,9 +1270,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
total_forks++;
spin_unlock(&current->sighand->siglock);
write_unlock_irq(&tasklist_lock);
-#ifdef CONFIG_FUNCTION_RET_TRACER
ftrace_retfunc_init_task(p);
-#endif
proc_fork_connector(p);
cgroup_post_fork(p);
return p;
diff --git a/kernel/sched.c b/kernel/sched.c
index fb17205..388d9db 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5901,9 +5901,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
* The idle tasks have their own, simple scheduling class:
*/
idle->sched_class = &idle_sched_class;
-#ifdef CONFIG_FUNCTION_RET_TRACER
ftrace_retfunc_init_task(idle);
-#endif
}

/*

\
 
 \ /
  Last update: 2008-11-23 09:25    [W:0.084 / U:0.916 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site