Re: [PATCH 0/2] tracing: Fixes to event_triggers patches found by Fengguang's test bot

[Tom Zanussi <tom.zanussi@linux.intel.com>]

Hi Steve,

On Sat, 2013-12-21 at 22:07 -0500, Steven Rostedt wrote:
> Tom,
> 
> This is the changes I made to fix the reports that Fengguang's kbuild test bot
> found. I folded in your change that fixes the bug with -ENODEV used in kernel.h.
> 

These look fine to me, and I didn't see any problems after running
through my normal event trigger testing with them applied.

So you can add my Acked-by and/or Tested-by..

Thanks for writing these improved versions!

Tom

> Paul,
> 
> I've Cc'd you because the second patch is fixing up RCU notation. The filter
> is protect by rcu_sched, and the first hunk of the patch is performed
> within a rcu_read_lock_sched(), and the second hunk is done on the write
> side. The tmp variable holds the old value, the value is updated, and
> then the new value gets updated. I emailed you so that you can verify
> that this all works.
> 
> Thanks!
> 
> -- Steve
> 
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace.git
> ftrace/core
> 
> Head SHA1: 05741e383228a79eeec83644181255d2171f9f7c
> 
> 
> Steven Rostedt (Red Hat) (2):
>       tracing: Add generic tracing_lseek() function
>       tracing: Fix rcu handling of event_trigger_data filter field
> 
> ----
>  include/linux/ftrace.h              |  2 --
>  kernel/trace/ftrace.c               | 25 ++++++-------------------
>  kernel/trace/trace.c                | 14 +++++++++-----
>  kernel/trace/trace.h                |  4 +++-
>  kernel/trace/trace_events_trigger.c | 10 ++++++----
>  kernel/trace/trace_stack.c          |  2 +-
>  6 files changed, 25 insertions(+), 32 deletions(-)