linux-kernel - Re: [PATCH 13/40] autonuma: CPU follow memory algorithm

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] [day] [month] [year] [list]

Date:	Sun, 01 Jul 2012 11:10:11 +0800
From:	Nai Xia <nai.xia@...il.com>
To:	Benjamin Herrenschmidt <benh@...nel.crashing.org>
CC:	dlaor@...hat.com, Andrea Arcangeli <aarcange@...hat.com>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Ingo Molnar <mingo@...nel.org>,
	Hillf Danton <dhillf@...il.com>, linux-kernel@...r.kernel.org,
	linux-mm@...ck.org, Dan Smith <danms@...ibm.com>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...e.hu>, Paul Turner <pjt@...gle.com>,
	Suresh Siddha <suresh.b.siddha@...el.com>,
	Mike Galbraith <efault@....de>,
	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
	Lai Jiangshan <laijs@...fujitsu.com>,
	Bharata B Rao <bharata.rao@...il.com>,
	Lee Schermerhorn <Lee.Schermerhorn@...com>,
	Rik van Riel <riel@...hat.com>,
	Johannes Weiner <hannes@...xchg.org>,
	Srivatsa Vaddagiri <vatsa@...ux.vnet.ibm.com>,
	Christoph Lameter <cl@...ux.com>,
	Alex Shi <alex.shi@...el.com>,
	Mauricio Faria de Oliveira <mauricfo@...ux.vnet.ibm.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
	Don Morris <don.morris@...com>
Subject: Re: [PATCH 13/40] autonuma: CPU follow memory algorithm

On 2012年07月01日 07:55, Benjamin Herrenschmidt wrote:
> On Sat, 2012-06-30 at 14:58 +0800, Nai Xia wrote:
>> If you insist on ignoring any constructive suggestions from others,
>
> But there is nothing constructive about your criticism.
>
> You are basically saying that the whole thing cannot work unless it's
> based on 20 years of research. Duh !

1. You quote me wrong: I said "group all pages to one node" is correct,
and highly possible to play the major role in your benchmarks.
Sampling is completely broken from my point of view. PZ's patch also
has similar idea of "group all pages to one node" which I think
is also correct.

2. My suggestion to Andrea: Do some more comparative benchmarks to
see what's really happening inside, instead of only macro benchmarks.
You need to have 20 hours of carefully designed survey research
for a new algorithm, instead of reading my mail and spending 20min
to give a conclusion.

If you cannot see the constructiveness of my suggestion. That's
your problem, not mine.

I understand the hard feelings of seeing the possible brokenness of a
thing you've already spend a lot of time. But that's the way people
seeking for truth.

You see, you guys has spent quite sometime to defend your points,
if this time were used to follow my advise doing some further
analysis maybe you've already got some valuable information.

Dor was right, we all made our points. And we are all busy.
Let's stop it. Thanks.

>
> Ben.
>
>
--
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/