linux-kernel - Re: [PATCH] perf_events: fix and improve x86 event scheduling

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: <1320949910.13800.35.camel@twins>
Date:	Thu, 10 Nov 2011 19:31:50 +0100
From:	Peter Zijlstra <peterz@...radead.org>
To:	Stephane Eranian <eranian@...gle.com>
Cc:	linux-kernel@...r.kernel.org, robert.richter@....com,
	mingo@...e.hu, ming.m.lin@...el.com, ak@...ux.intel.com
Subject: Re: [PATCH] perf_events: fix and improve x86 event scheduling

On Thu, 2011-11-10 at 16:09 +0100, Stephane Eranian wrote:
> What are the configs for which you have failures on Intel?
> 
Continuing to use Core2 as an example platform:

  cycles,cycles,instructions,instructions

They all have weight 3, yet the masks are not the same, it'll assign
them like: pmc0, pmc1, fp-instructions, fail

Counter rotation will make sure the next attempt looks like:

  instructions,cycles,cycles,instructions

Which will work correctly since it'll do: pmc0, pmc1, fp-cycles,
fp-instructions.

Its usually not a really big deal, but I ran into it few months ago and
noticed it because the scaled values weren't what I was expecting them
to be, took me a while to figure out wth happened.


--
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/