Messages in this thread | | | Date | Tue, 16 Jul 2013 11:02:17 -0400 | From | Dave Jones <> | Subject | Re: splice vs execve lockdep trace. |
| |
On Tue, Jul 16, 2013 at 09:59:35AM -0400, Vince Weaver wrote: > On Mon, 15 Jul 2013, Linus Torvalds wrote: > > > On Mon, Jul 15, 2013 at 7:38 PM, Dave Jones <davej@redhat.com> wrote: > > > > > Interestingly, the 'soft lockups' I was > > > seeing all the time on that box seem to have gone into hiding. > > > > Honestly, I'm somewhat inclined to blame the whole perf situation, and > > saying that we hopefully got that fixed. In between the silly do_div() > > buglets and all the indications that the time was spent in nmi > > handlers, I'd be willing to just ignore them as false positives > > brought on by the whole switch to the perf irq.. > > Did the perf soft lockups go away with 3.11-rc1? > > 734df5ab549ca44f40de0f07af1c8803856dfb18 finally got committed, > and it fixes a major long-standing perf-related NMI soft lockup bug I > found when fuzzing.
That could be it. I never got around to trying that commit standalone when you first pointed it out.
thanks,
Dave
| |