linux-kernel - Re: [PATCH] aio: optimize free kioctx slot search

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]

Message-ID: <20131227123250.GA22587@parisc-linux.org>
Date:	Fri, 27 Dec 2013 05:32:50 -0700
From:	Matthew Wilcox <matthew@....cx>
To:	Gu Zheng <guz.fnst@...fujitsu.com>
Cc:	Benjamin <bcrl@...ck.org>, linux-aio@...ck.org,
	fsdevel <linux-fsdevel@...r.kernel.org>,
	linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] aio: optimize free kioctx slot search

On Fri, Dec 27, 2013 at 06:32:24PM +0800, Gu Zheng wrote:
> Hi Matthew,
> On 12/26/2013 10:13 PM, Matthew Wilcox wrote:
> 
> > On Thu, Dec 26, 2013 at 06:22:10PM +0800, Gu Zheng wrote:
> >> to the new free slot if it's smaller than 'next_free'. So that we can
> >> ensure that the slots before 'next_free' are all used, and the search can start
> >> from 'next_free' to reduce the search and improve the performance.
> > 
> > Have you benchmarked it?  What kinds of improvements are you seeing?
> 
> Not yet. But the improvement is obvious, we can always reduce the search region and
> improve the hit rate when trying to find a free slot with this change.

It's not so obvious to me.  You've added extra stores, which may dirty
an additional cacheline.  That may end up costing more than it saves.
I suspect the difference is lost in the noise, but I've been wrong about
these things before.

-- 
Matthew Wilcox				Intel Open Source Technology Centre
"Bill, look, we understand that you're interested in selling us this
operating system, but compare it to ours.  We can't possibly take such
a retrograde step."
--
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/