linux-kernel - Re: [RFC][PATCH] PERF_COUNT_SW_RUNNABLE_TASKS: measure and act upon parallellism

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]

Date:	Mon, 8 Feb 2010 12:52:15 +0100
From:	Stijn Devriendt <highguy@...il.com>
To:	Peter Zijlstra <peterz@...radead.org>
Cc:	mingo@...e.hu, linux-kernel@...r.kernel.org,
	torvalds@...ux-foundation.org, efault@....de, andrea@...e.de,
	akpm@...ux-foundation.org, Tejun Heo <tj@...nel.org>,
	Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [RFC][PATCH] PERF_COUNT_SW_RUNNABLE_TASKS: measure and act upon 
	parallellism

On Mon, Feb 8, 2010 at 11:00 AM, Peter Zijlstra <peterz@...radead.org> wrote:
> On Sun, 2010-02-07 at 12:30 +0100, highguy@...il.com wrote:
>
>> Here's an initial RFC patch for the parallallism
>> events for perf_events.
>
> OK, so you managed to rub me totally the wrong way with posting this
> yesterday:
>  - you send me each patch twice
>  - you used the horrible git sendmail default of --chain-reply-to
>   (some day I'll write a script that will detect and auto-bounce
>    series sent to me like that)
>  - you failed to provide a changelog for any of the patches
>  - some subjects were long enough to be a changelog
>
> Please don't do that again ;-)

Sorry for that. I'm still getting used to the thing called 'git'. Apart
from obviously sharing its name, there's not so
much we have in common so far... VCS experiences currently
don't go far beyond svn and the horrid CVS.
I just used the reply-list from the initial series about this subject,
perhaps you were listed twice in that one as well.

Isn't no longer making --chain-reply-to the default an option?

>
> Anyway, it did get me thinking, how about something like the below?
> <SNIP>

That pretty much looks like what I need at first sight. I'll definately
give it a very close look and test it out.
The perf_event approach would be nice, but the implementations
just seem to collide.

Thanks for taking the time anyway,
Stijn
--
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/