linux-kernel - Re: [PATCH 04/12] perf, x86: Support the TSX intx/intx

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: <20130126210000.GK30577@one.firstfloor.org>
Date:	Sat, 26 Jan 2013 22:00:00 +0100
From:	Andi Kleen <andi@...stfloor.org>
To:	Ingo Molnar <mingo@...nel.org>
Cc:	Andi Kleen <andi@...stfloor.org>, linux-kernel@...r.kernel.org,
	a.p.zijlstra@...llo.nl, akpm@...ux-foundation.org, acme@...hat.com,
	eranian@...gle.com, jolsa@...hat.com, namhyung@...nel.org,
	Andi Kleen <ak@...ux.intel.com>
Subject: Re: [PATCH 04/12] perf, x86: Support the TSX intx/intx_cp qualifiers v2

On Sat, Jan 26, 2013 at 12:54:02PM +0100, Ingo Molnar wrote:
> 
> * Andi Kleen <andi@...stfloor.org> wrote:
> 
> > From: Andi Kleen <ak@...ux.intel.com>
> > 
> > Implement the TSX transaction and checkpointed transaction 
> > qualifiers for Haswell. This allows e.g. to profile the number 
> > of cycles in transactions.
> 
> The changelog should explain why this is a basic feature that is 
> a must-have to have basic perf support. (i.e. default 'perf 
> record', 'perf report', 'perf stat' and 'perf top' works and is 
> meaningful.)

The patch is not needed for primitive cycle sampling only.

These qualifiers are a fundamental feature for any TSX tuning. If you don't
care about TSX it's not fundamental. TSX tuning relies heavily on the PMU,
so for any TSX work it's important. The 0/0 description has more
references on the topic.

It's not easy to use however with just this patch, the "perf stat -T" patch
in the advanced series makes it easy to use.

I will assume the description is sufficient and will not repost with
a new changelog, as "basicness" of the patch is not a useful property
to document in the long term git history.

-Andi
-- 
ak@...ux.intel.com -- Speaking for myself only.
--
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/