[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 5 May 2016 08:40:41 -0700
From: Alexei Starovoitov <alexei.starovoitov@...il.com>
To: Steven Rostedt <rostedt@...dmis.org>
Cc: linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...nel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Alexei Starovoitov <ast@...nel.org>
Subject: Re: [for-next][PATCH 2/2] tracing: Use temp buffer when filtering
events
On Thu, May 05, 2016 at 11:35:43AM -0400, Steven Rostedt wrote:
> On Thu, 5 May 2016 11:32:51 -0400
> Steven Rostedt <rostedt@...dmis.org> wrote:
>
> > Think about it, what's the difference if the interrupt came in just
> > before the trace or just after? It still came in the same location with
> > respect to the normal flow of the code. The only difference is, where
> > we recorded it.
>
> Also, if we used a temp buffer for each one, the same thing would
> happen. The interrupt would be committed first before returning back to
> the interrupted event. Perf does the same thing, but all the time.
yeah. good point. there is no actual 'order' here.
The whole thing looks good to me.
Acked-by: Alexei Starovoitov <ast@...nel.org>
Powered by blists - more mailing lists