[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220309105403.66d2b6ef@gandalf.local.home>
Date: Wed, 9 Mar 2022 10:54:03 -0500
From: Steven Rostedt <rostedt@...dmis.org>
To: Daniel Bristot de Oliveira <bristot@...nel.org>
Cc: Nicolas Saenz Julienne <nsaenzju@...hat.com>,
linux-kernel <linux-kernel@...r.kernel.org>,
linux-trace-devel@...r.kernel.org, mtosatti <mtosatti@...hat.com>
Subject: Re: Warning when exiting osnoise tracer
On Wed, 9 Mar 2022 08:25:56 +0100
Daniel Bristot de Oliveira <bristot@...nel.org> wrote:
> On 3/8/22 18:30, Nicolas Saenz Julienne wrote:
> > While running next-20220307, on a system with isolated CPUs (don't know if
> > relevant but less tested code-path), I see this after killing (ctrl-C) the
> > following trace command:
> >
> > trace-cmd record -e all -M 10 -p osnoise --poll
> >
> > Note that this is *without* my recent osnoise patch.
>
> yeah, your patch does not hit this part of the code.
>
> > [ 129.925474] ------------[ cut here ]------------
I'm surprised that my tests did not hit this.
I've been running a lot of trace-cmd tests lately (to try and release 3.0!)
and I have osnoise tracer enabled in these tests.
Hmm, it's just a warning that doesn't appear to cause any damage. I just
tried it out, and sure enough it triggered. The problem is that my
trace-cmd tests do not check dmesg :-/ Maybe I need to change that, as I
only looked at the result of the test, and because the tests spits out a
lot of noise to dmesg, I can easily miss warnings there. I'll add a test to
look for Call Traces at the end of the test.
Thanks for the report.
-- Steve
Powered by blists - more mailing lists