linux-kernel - Re: [PATCH 4/5] perf report: Fix dynamic entry display in hierarchy

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20160226231208.GD19398@krava.redhat.com>
Date:	Sat, 27 Feb 2016 00:12:08 +0100
From:	Jiri Olsa <jolsa@...hat.com>
To:	Arnaldo Carvalho de Melo <acme@...nel.org>
Cc:	Steven Rostedt <rostedt@...dmis.org>,
	Namhyung Kim <namhyung@...nel.org>,
	Ingo Molnar <mingo@...nel.org>,
	Peter Zijlstra <peterz@...radead.org>,
	Jiri Olsa <jolsa@...nel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	David Ahern <dsahern@...il.com>,
	Andi Kleen <andi@...stfloor.org>,
	Stephane Eranian <eranian@...gle.com>,
	Wang Nan <wangnan0@...wei.com>
Subject: Re: [PATCH 4/5] perf report: Fix dynamic entry display in hierarchy

On Fri, Feb 26, 2016 at 07:37:05PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Fri, Feb 26, 2016 at 05:26:02PM -0500, Steven Rostedt escreveu:
> > On Fri, 26 Feb 2016 19:08:19 -0300
> > Arnaldo Carvalho de Melo <acme@...nel.org> wrote:
> > 
> > 
> > > It is exploding at this depth, not liking how gfp_flags is described, problem
> > > seems to be in this function, this __print_flags() long thing.
> > > 
> > > Steven, does this look familiar?
> > > 
> > >
> > 
> > Hmm, trace-cmd has the same issue. I just having noticed.
> > 
> > I'll see if I can fix it.
> 
> Ok, this works well with an RHEL kernel being tested with tons of perf
> backports by Jiri, its just with 4.5-rc kernels that this is exploding.

thanks, I isolated this down to the:
  156174999dd1 perf/intel/x86: Enlarge the PEBS buffer

it's related to the FREERUNNING PEBS patchset that allows
store more than 1 event in PEBS buffer

seems like Core CPU doesn't like the default setup and
delays in msr access, which makes the watchdog scream

jirka