[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMsRxfL7cSsWsmzo67BXN3JxtmeH+wVv-dkY0xGLb-ouZNu4zg@mail.gmail.com>
Date: Thu, 2 Jul 2015 21:48:12 +0200
From: Stephane Eranian <eranian@...glemail.com>
To: Vince Weaver <vincent.weaver@...ne.edu>
Cc: Peter Zijlstra <peterz@...radead.org>,
LKML <linux-kernel@...r.kernel.org>,
Ingo Molnar <mingo@...hat.com>,
Arnaldo Carvalho de Melo <acme@...nel.org>
Subject: Re: perf: fuzzer, lots of warnings
On Thu, Jul 2, 2015 at 7:59 PM, Vince Weaver <vincent.weaver@...ne.edu> wrote:
>
> On Thu, 2 Jul 2015, Peter Zijlstra wrote:
>
> > On Thu, Jul 02, 2015 at 11:42:43AM -0400, Vince Weaver wrote:
> > > (#4)
> > >
> > > [ 1638.702578] WARNING: CPU: 2 PID: 25543 at arch/x86/kernel/cpu/perf_event_intel_lbr.c:349 intel_pmu_lbr_disable+0xbc/0xd0()
> > >
> > > WARN_ON_ONCE(cpuc->lbr_users < 0);
> > >
> > > [ 1638.857921] [<ffffffff8102e56c>] intel_pmu_lbr_disable+0xbc/0xd0
> > > [ 1638.864874] [<ffffffff81031812>] intel_pmu_disable_event+0xc2/0x130
> > > [ 1638.872125] [<ffffffff81030a49>] ? __intel_pmu_disable_all+0x49/0x70
> > > [ 1638.879455] [<ffffffff81029175>] x86_pmu_stop+0x75/0xd0
> > > [ 1638.885560] [<ffffffff81159f1a>] perf_ioctl+0x40a/0x470
> > > [ 1638.891650] [<ffffffff811f199a>] do_vfs_ioctl+0x2ba/0x4a0
> > > [ 1638.897945] [<ffffffff811dd9ac>] ? vfs_read+0x11c/0x130
> > > [ 1638.904069] [<ffffffff811f27bd>] ? poll_select_set_timeout+0x4d/0x70
> > > [ 1638.911417] [<ffffffff811f1bf9>] SyS_ioctl+0x79/0x90
> > > [ 1638.917295] [<ffffffff816a64b2>] entry_SYSCALL_64_fastpath+0x16/0x7a
> >
> > This might maybe help:
> >
> > http://lkml.kernel.org/r/20150624144750.GJ18673@twins.programming.kicks-ass.net
>
> ah, OK, I missed that one somehow. It would also explain the
>
> [ 3971.156004] Uhhuh. NMI received for unknown reason 21 on CPU 2.
> [ 3971.162812] Do you have a strange power saving mode enabled?
> [ 3971.169363] Dazed and confused, but trying to continue
>
> messages I am getting.
>
>
> I'm also still getting a lot of
> perfevents: irq loop stuck!
Last I saw this was on WSM. I also suspect this may not happen if the
event is only measured at the user level.
> messages, I thought the workaround for that had gone in for 4.2 but I
> guess not.
>
> Vince
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists