linux-kernel - Re: ext4 write performance regression in 3.6-rc1 on RAID0/5

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20120822160025.272188d1@notabene.brown>
Date:	Wed, 22 Aug 2012 16:00:25 +1000
From:	NeilBrown <neilb@...e.de>
To:	Yuanhan Liu <yuanhan.liu@...ux.intel.com>
Cc:	Fengguang Wu <fengguang.wu@...el.com>,
	Li Shaohua <shli@...ionio.com>, Theodore Ts'o <tytso@....edu>,
	Marti Raudsepp <marti@...fo.org>,
	Kernel hackers <linux-kernel@...r.kernel.org>,
	ext4 hackers <linux-ext4@...r.kernel.org>, maze@...gle.com,
	"Shi, Alex" <alex.shi@...el.com>, linux-fsdevel@...r.kernel.org,
	linux RAID <linux-raid@...r.kernel.org>
Subject: Re: ext4 write performance regression in 3.6-rc1 on RAID0/5

On Wed, 22 Aug 2012 11:57:02 +0800 Yuanhan Liu <yuanhan.liu@...ux.intel.com>
wrote:

>  
> -#define NR_STRIPES		256
> +#define NR_STRIPES		1024

Changing one magic number into another magic number might help your case, but
it not really a general solution.

Possibly making sure that max_nr_stripes is at least some multiple of the
chunk size might make sense, but I wouldn't want to see a very large multiple.

I thing the problems with RAID5 are deeper than that.  Hopefully I'll figure
out exactly what the best fix is soon - I'm trying to look into it.

I don't think the size of the cache is a big part of the solution.  I think
correct scheduling of IO is the real answer.

Thanks,
NeilBrown

Download attachment "signature.asc" of type "application/pgp-signature" (829 bytes)