From: "Amir G." Subject: Re: ext4: removal of alloc_sem locks from block allocation paths Date: Thu, 24 Mar 2011 07:19:46 +0200 Message-ID: References: <1300656770-14994-1-git-send-email-amir73il@users.sourceforge.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Ext4 Developers List , Yongqiang Yang To: tytso@mit.edu Return-path: Received: from mail-qw0-f46.google.com ([209.85.216.46]:44382 "EHLO mail-qw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751489Ab1CXFTr convert rfc822-to-8bit (ORCPT ); Thu, 24 Mar 2011 01:19:47 -0400 Received: by qwk3 with SMTP id 3so6423420qwk.19 for ; Wed, 23 Mar 2011 22:19:46 -0700 (PDT) In-Reply-To: Sender: linux-ext4-owner@vger.kernel.org List-ID: On Wed, Mar 23, 2011 at 11:04 PM, Amir G. wrote: > Hi Ted, > > I have been running some more tests, including 1K, with no problems o= bserved Take that back. My test setup was wrong. xfstest 131 does hit a BUG in ext4_mb_init_cache() with 1K blocks. I will post a fix when I have it. > and Yongqiang has also reviewed the patches. > > Is there any chance to merge these patches in current merge window? > > Thanks, > Amir. > > On Sun, Mar 20, 2011 at 11:32 PM, =A0= wrote: >> The purpose of this patch set is the removal of grp->alloc_sem locks >> from block allocation paths. >> >> This resulting code is cleaner and should perform better in concurre= nt >> allocating tasks workloads. >> >> I ran several xfstests runs with these patches (4K block only). >> I tried several online resizes and verifyed that both in-core and on= -disk >> group counters are correct. >> >> ext4: move ext4_add_groupblocks() to mballoc.c >> ext4: implement ext4_add_groupblocks() by freeing blocks >> ext4: synchronize ext4_mb_init_group() with buddy page lock >> ext4: teach ext4_mb_init_cache() to skip uptodate buddy caches >> ext4: remove alloc_semp >> > -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html