linux-kernel - Re: [RFC PATCH 6/7] perf, x86: large PEBS interrupt threshold

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: <20140528171031.GV30445@twins.programming.kicks-ass.net>
Date:	Wed, 28 May 2014 19:10:31 +0200
From:	Peter Zijlstra <peterz@...radead.org>
To:	Stephane Eranian <eranian@...gle.com>
Cc:	Andi Kleen <ak@...ux.intel.com>,
	"Yan, Zheng" <zheng.z.yan@...el.com>,
	LKML <linux-kernel@...r.kernel.org>,
	"mingo@...e.hu" <mingo@...e.hu>
Subject: Re: [RFC PATCH 6/7] perf, x86: large PEBS interrupt threshold

On Wed, May 28, 2014 at 07:05:51PM +0200, Stephane Eranian wrote:
> On Wed, May 28, 2014 at 6:51 PM, Andi Kleen <ak@...ux.intel.com> wrote:
> >> The only part I don't quite follow here is this:
> >>                       if (__test_and_set_bit(bit, (unsigned long *)&status))
> >>                                 continue;
> >>
> >> Which seems to indicate the code is making sure each counter is
> >> processed only once. But it can only be processed once, if you have
> >> only one record. And if you have multiple, you want to be able to
> >> handle the same counter multiple times, at least once perf PEBS
> >> record. So I am a bit confused about this test.
> >
> > Each PEBS record is only for a single counter overflow. So it
> > always should only be a single perf event.
> >
> So, you're telling me this is a sanity check. That p->status can
> only have one bit set. Somehow that's not how I recall it working.
> 
> The point is that a single PEBS record is enough for multiple
> events when the overflows occur simultaneously because they
> all get the same machine state which is correct.  A single entry
> also saves space in the buffer.

Confusion reigns.. now if only someone would write all this down in a
coherent manner and give it an official Intel stamp of approval ;-)

Content of type "application/pgp-signature" skipped