[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20210806181044.6tqzlanvxdihaguy@linutronix.de>
Date: Fri, 6 Aug 2021 20:10:44 +0200
From: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To: Steven Rostedt <rostedt@...dmis.org>
Cc: linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...hat.com>,
Thomas Gleixner <tglx@...utronix.de>,
Peter Zijlstra <peterz@...radead.org>
Subject: Re: [PATCH] trace: Add migrate-disabled counter to tracing output.
On 2021-08-06 13:51:24 [-0400], Steven Rostedt wrote:
> > diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h
> > index ad413b382a3ca..7c4280b4c6be7 100644
> > --- a/include/linux/trace_events.h
> > +++ b/include/linux/trace_events.h
> > @@ -69,6 +69,7 @@ struct trace_entry {
> > unsigned char flags;
> > unsigned char preempt_count;
> > int pid;
> > + unsigned char migrate_disable;
>
> The only issue I have with this patch is this part. It extends the
> header of all events beyond 8 bytes, and actually adds another 4 or 8
> bytes despite being just a char in size. That's because this is a
> header of a structure, which depending on the first field of an event,
> padding may be added to have 4 or 8 byte alignment.
>
> I'll be fine with merging this counter with either flags or
> preempt_count (neither needs all 8 bits). I can figure out how to
> update libtraceevent to handle this case.
I was kind of expecting that. Let me try 4 bits for the preempt counter
and 4 bits for the migrate counter.
> -- Steve
>
Sebastian
Powered by blists - more mailing lists