From: Andreas Dilger Subject: Re: [PATCH] improve jbd fsync batching Date: Mon, 03 Nov 2008 22:24:28 -0700 Message-ID: <20081104052428.GP3184@webber.adilger.int> References: <20081028201614.GA21600@unused.rdu.redhat.com> <20081103122729.60582692.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7BIT Cc: Josef Bacik , linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org, rwheeler@redhat.com To: Andrew Morton Return-path: Received: from sca-es-mail-1.Sun.COM ([192.18.43.132]:59580 "EHLO sca-es-mail-1.sun.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751102AbYKDFYb (ORCPT ); Tue, 4 Nov 2008 00:24:31 -0500 In-reply-to: <20081103122729.60582692.akpm@linux-foundation.org> Content-disposition: inline Sender: linux-ext4-owner@vger.kernel.org List-ID: On Nov 03, 2008 12:27 -0800, Andrew Morton wrote: > On Tue, 28 Oct 2008 16:16:15 -0400 > Josef Bacik wrote: > > + spin_lock(&journal->j_state_lock); > > + commit_time = journal->j_average_commit_time; > > + spin_unlock(&journal->j_state_lock); > > OK, the lock is needed on 32-bit machines, I guess. Should we pessimize the 64-bit performance in that case, for 32-bit increasingly rare 32-bit platforms? It might be useful to have a spin_{,un}lock_64bit_word() helper that evaluates to a no-op on plaforms that don't need a hammer to do an atomic 64-bit update. Cheers, Andreas -- Andreas Dilger Sr. Staff Engineer, Lustre Group Sun Microsystems of Canada, Inc.