[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <5e6c71ca24103a54c4909ae57e58544aec4a56be.camel@redhat.com>
Date: Fri, 16 Jan 2026 16:01:30 +0100
From: Gabriele Monaco <gmonaco@...hat.com>
To: K Prateek Nayak <kprateek.nayak@....com>
Cc: Tomas Glozar <tglozar@...hat.com>, Clark Williams <williams@...hat.com>,
John Kacur <jkacur@...hat.com>, linux-kernel@...r.kernel.org, Steven
Rostedt <rostedt@...dmis.org>, Nam Cao <namcao@...utronix.de>, Juri Lelli
<jlelli@...hat.com>, Masami Hiramatsu <mhiramat@...nel.org>, Ingo Molnar
<mingo@...hat.com>, Peter Zijlstra <peterz@...radead.org>,
linux-trace-kernel@...r.kernel.org
Subject: Re: [PATCH v4 10/15] sched: Add task enqueue/dequeue trace points
On Fri, 2026-01-16 at 20:20 +0530, K Prateek Nayak wrote:
> Hello Gabriele, Nam,
>
> On 1/16/2026 6:09 PM, Gabriele Monaco wrote:
> > @@ -2087,6 +2087,8 @@ unsigned long get_wchan(struct task_struct *p)
> >
> > void enqueue_task(struct rq *rq, struct task_struct *p, int flags)
> > {
>
> For delayed task, I think you'll need:
>
> if (trace_sched_enqueue_tp_enabled() && !(flags & ENQUEUE_DELAYED))
>
> > + trace_sched_enqueue_tp(p, rq->cpu);
> > +
> > if (!(flags & ENQUEUE_NOCLOCK))
> > update_rq_clock(rq);
>
> Since delayed tasks haven't hit __block_task(), they are essentially
> still enqueued. Peter should be able to confirm. Other than that,
> the placements of the tracepoints look good now. Feel free to include:
>
> Reviewed-by: K Prateek Nayak <kprateek.nayak@....com>
Mmh, I was sure I missed something after the comments on Nam's patch, thanks for
the heads up and review!
Going to try your suggestion.
Thanks,
Gabriele
Powered by blists - more mailing lists