linux-ext4 - Re: [PATCH -v2 0/3] jbd2 scalability patches

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: <1280973537.2678.37.camel@localhost.localdomain>
Date:	Wed, 04 Aug 2010 18:58:57 -0700
From:	john stultz <johnstul@...ibm.com>
To:	"Theodore Ts'o" <tytso@....edu>
Cc:	Ext4 Developers List <linux-ext4@...r.kernel.org>,
	ocfs2-devel@....oracle.com, Keith Maanthey <kmannth@...ibm.com>,
	Eric Whitney <eric.whitney@...com>
Subject: Re: [PATCH -v2 0/3] jbd2 scalability patches

On Wed, 2010-08-04 at 12:39 -0400, Theodore Ts'o wrote:
> This version fixes three bugs in the 2nd patch of this series that
> caused kernel BUG when the system was under race.  We weren't accounting
> with t_oustanding_credits correctly, and there were race conditions
> caused by the fact the I had overlooked the fact that
> __jbd2_log_wait_for_space() and jbd2_get_transaction() requires
> j_state_lock to be write locked.

So without the vfs patches, I don't see much change with this patchset
(similar to the last).

novfs + j_state lock
Throughput 763.105 MB/sec 8 procs
Throughput 1056.81 MB/sec 4 procs
Throughput 681.761 MB/sec 2 procs
Throughput 409.25 MB/sec 1 procs

vs

no vfs + j_state lock + jdb2 scalability queue
Throughput 767.778 MB/sec 8 procs
Throughput 1069.58 MB/sec 4 procs
Throughput 679.786 MB/sec 2 procs
Throughput 401.419 MB/sec 1 procs


But with the vfs patchset, there's a nice increase @8cpus.
 vfs + j_state lock
Throughput 1061.44 MB/sec 8 procs
Throughput 1126.55 MB/sec 4 procs
Throughput 706.306 MB/sec 2 procs
Throughput 402.102 MB/sec 1 procs

vs

vfs + j_state lock + jdb2 scalability queue
Throughput 1214.21 MB/sec 8 procs
Throughput 1175.49 MB/sec 4 procs
Throughput 716.294 MB/sec 2 procs
Throughput 402.988 MB/sec 1 procs


I'll post perf log data tomorrow.

thanks
-john



--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html