[GIT PULL] tracing: A couple of minor fixes

[GIT PULL] tracing: A couple of minor fixes

[Steven Rostedt]

Linus,

Two minor fixes:

 - A race in perf trace initialization (missing mutexes)

 - Minor fix to represent gfp_t in synthetic events as properly signed


Please pull the latest trace-v5.4-rc3 tree, which can be found at:


  git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git
trace-v5.4-rc3

Tag SHA1: 46467c95be6062b0c66aae0efbab5746bee09f1e
Head SHA1: 668993e5696ddc5db01940d66818c7c038895893


Prateek Sood (1):
      tracing: Fix race in perf_trace_buf initialization

Zhengjun Xing (1):
      tracing: Fix "gfp_t" format for synthetic events

----
 kernel/trace/trace_event_perf.c  | 4 ++++
 kernel/trace/trace_events_hist.c | 2 ++
 2 files changed, 6 insertions(+)
---------------------------
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 0892e38ed6fb..a9dfa04ffa44 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -272,9 +272,11 @@ int perf_kprobe_init(struct perf_event *p_event, bool is_retprobe)
 		goto out;
 	}
 
+	mutex_lock(&event_mutex);
 	ret = perf_trace_event_init(tp_event, p_event);
 	if (ret)
 		destroy_local_trace_kprobe(tp_event);
+	mutex_unlock(&event_mutex);
 out:
 	kfree(func);
 	return ret;
@@ -282,8 +284,10 @@ int perf_kprobe_init(struct perf_event *p_event, bool is_retprobe)
 
 void perf_kprobe_destroy(struct perf_event *p_event)
 {
+	mutex_lock(&event_mutex);
 	perf_trace_event_close(p_event);
 	perf_trace_event_unreg(p_event);
+	mutex_unlock(&event_mutex);
 
 	destroy_local_trace_kprobe(p_event->tp_event);
 }
diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index 57648c5aa679..7482a1466ebf 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -679,6 +679,8 @@ static bool synth_field_signed(char *type)
 {
 	if (str_has_prefix(type, "u"))
 		return false;
+	if (strcmp(type, "gfp_t") == 0)
+		return false;
 
 	return true;
 }

Re: [GIT PULL] tracing: A couple of minor fixes

[Steven Rostedt]

On Mon, 21 Oct 2019 12:45:08 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:

> Linus,
> 

Hi Linus,

Please hold off on this. Someone just pointed out to me that one of the
change logs has a bunch of crap attached to it that shouldn't be
committed.

I need to rebase (just the change logs, so no retesting is needed).

I'll go and delete this tag and create a new (different one) when
ready.

-- Steve


> Two minor fixes:
> 
>  - A race in perf trace initialization (missing mutexes)
> 
>  - Minor fix to represent gfp_t in synthetic events as properly signed
> 
> 
> Please pull the latest trace-v5.4-rc3 tree, which can be found at:
> 
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git
> trace-v5.4-rc3
> 
> Tag SHA1: 46467c95be6062b0c66aae0efbab5746bee09f1e
> Head SHA1: 668993e5696ddc5db01940d66818c7c038895893
> 
> 
> Prateek Sood (1):
>       tracing: Fix race in perf_trace_buf initialization
> 
> Zhengjun Xing (1):
>       tracing: Fix "gfp_t" format for synthetic events
> 
> ----
>  kernel/trace/trace_event_perf.c  | 4 ++++
>  kernel/trace/trace_events_hist.c | 2 ++
>  2 files changed, 6 insertions(+)
> ---------------------------
> diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
> index 0892e38ed6fb..a9dfa04ffa44 100644
> --- a/kernel/trace/trace_event_perf.c
> +++ b/kernel/trace/trace_event_perf.c
> @@ -272,9 +272,11 @@ int perf_kprobe_init(struct perf_event *p_event, bool is_retprobe)
>  		goto out;
>  	}
>  
> +	mutex_lock(&event_mutex);
>  	ret = perf_trace_event_init(tp_event, p_event);
>  	if (ret)
>  		destroy_local_trace_kprobe(tp_event);
> +	mutex_unlock(&event_mutex);
>  out:
>  	kfree(func);
>  	return ret;
> @@ -282,8 +284,10 @@ int perf_kprobe_init(struct perf_event *p_event, bool is_retprobe)
>  
>  void perf_kprobe_destroy(struct perf_event *p_event)
>  {
> +	mutex_lock(&event_mutex);
>  	perf_trace_event_close(p_event);
>  	perf_trace_event_unreg(p_event);
> +	mutex_unlock(&event_mutex);
>  
>  	destroy_local_trace_kprobe(p_event->tp_event);
>  }
> diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
> index 57648c5aa679..7482a1466ebf 100644
> --- a/kernel/trace/trace_events_hist.c
> +++ b/kernel/trace/trace_events_hist.c
> @@ -679,6 +679,8 @@ static bool synth_field_signed(char *type)
>  {
>  	if (str_has_prefix(type, "u"))
>  		return false;
> +	if (strcmp(type, "gfp_t") == 0)
> +		return false;
>  
>  	return true;
>  }

[GIT PULL v2] tracing: A couple of minor fixes

[Steven Rostedt]

Linus,

Two minor fixes:

 - A race in perf trace initialization (missing mutexes)

 - Minor fix to represent gfp_t in synthetic events as properly signed


Please pull the latest trace-v5.4-rc3-2 tree, which can be found at:


  git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git
trace-v5.4-rc3-2

Tag SHA1: e8c07438dfe315733667f93b15f866c45f8ce192
Head SHA1: 5dacd386363b9a286002740cfa0a16f3358ae990


Prateek Sood (1):
      tracing: Fix race in perf_trace_buf initialization

Zhengjun Xing (1):
      tracing: Fix "gfp_t" format for synthetic events

----
 kernel/trace/trace_event_perf.c  | 4 ++++
 kernel/trace/trace_events_hist.c | 2 ++
 2 files changed, 6 insertions(+)
---------------------------
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 0892e38ed6fb..a9dfa04ffa44 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -272,9 +272,11 @@ int perf_kprobe_init(struct perf_event *p_event, bool is_retprobe)
 		goto out;
 	}
 
+	mutex_lock(&event_mutex);
 	ret = perf_trace_event_init(tp_event, p_event);
 	if (ret)
 		destroy_local_trace_kprobe(tp_event);
+	mutex_unlock(&event_mutex);
 out:
 	kfree(func);
 	return ret;
@@ -282,8 +284,10 @@ int perf_kprobe_init(struct perf_event *p_event, bool is_retprobe)
 
 void perf_kprobe_destroy(struct perf_event *p_event)
 {
+	mutex_lock(&event_mutex);
 	perf_trace_event_close(p_event);
 	perf_trace_event_unreg(p_event);
+	mutex_unlock(&event_mutex);
 
 	destroy_local_trace_kprobe(p_event->tp_event);
 }
diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index 57648c5aa679..7482a1466ebf 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -679,6 +679,8 @@ static bool synth_field_signed(char *type)
 {
 	if (str_has_prefix(type, "u"))
 		return false;
+	if (strcmp(type, "gfp_t") == 0)
+		return false;
 
 	return true;
 }

Re: [GIT PULL v2] tracing: A couple of minor fixes

[Linus Torvalds]

On Mon, Oct 21, 2019 at 3:19 PM Steven Rostedt <rostedt@goodmis.org> wrote:
>
> Two minor fixes:
>
>  - A race in perf trace initialization (missing mutexes)

Are you sure you fixed the crud?

That commit still has a very odd commit message with an oops and
ramdump at the top, before the explanation.

             Linus

Re: [GIT PULL v2] tracing: A couple of minor fixes

[Steven Rostedt]

On Mon, 21 Oct 2019 16:30:14 -0400
Linus Torvalds <torvalds@linux-foundation.org> wrote:

> On Mon, Oct 21, 2019 at 3:19 PM Steven Rostedt <rostedt@goodmis.org> wrote:
> >
> > Two minor fixes:
> >
> >  - A race in perf trace initialization (missing mutexes)  
> 
> Are you sure you fixed the crud?
> 
> That commit still has a very odd commit message with an oops and
> ramdump at the top, before the explanation.
> 

I can clean it up some more. I just deleted from the original:

"Could you please confirm:
1) the race mentioned below exists or not.
2) if following patch fixes it.

>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8>8  "

That was in it.

These are minor fixes, so there's no rush. I'll look at it tonight and
see if I can make the change log a little nicer.

-- Steve