From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id pB17LsFb249094 for ; Thu, 1 Dec 2011 01:21:54 -0600 Date: Thu, 1 Dec 2011 02:21:49 -0500 From: Christoph Hellwig Subject: Re: [PATCH 4/4] xfs: fix the logspace waiting algorithm Message-ID: <20111201072149.GA1319@infradead.org> References: <20111128081732.350228200@bombadil.infradead.org> <20111128081925.981681380@bombadil.infradead.org> <20111130235641.GX29840@sgi.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20111130235641.GX29840@sgi.com> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: Ben Myers Cc: Christoph Hellwig , xfs@oss.sgi.com On Wed, Nov 30, 2011 at 05:56:41PM -0600, Ben Myers wrote: > Christoph, > > I feel that the ordering of the accesses to l_grant_head and the writeq > list may be important in the lockless path, and notice that they used to > be in opposite order. It could use a comment explaining why (if) it is > ok. Do you mean moving the xlog_space_left after the first list_empty_careful check in xlog_grant_log_space? It doesn't matter given that we re-check the available space after each wakeup. _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs