From mboxrd@z Thu Jan 1 00:00:00 1970 From: Al Viro Subject: Re: [PATCH 4/4] fs: remove inode_lock from iput_final and prune_icache Date: Wed, 27 Oct 2010 06:25:20 +0100 Message-ID: <20101027052520.GF19804@ZenIV.linux.org.uk> References: <1288153384-8878-1-git-send-email-david@fromorbit.com> <1288153384-8878-5-git-send-email-david@fromorbit.com> <20101027044038.GE19804@ZenIV.linux.org.uk> <1288154866.2652.84.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Dave Chinner , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org To: Eric Dumazet Return-path: Received: from zeniv.linux.org.uk ([195.92.253.2]:41518 "EHLO ZenIV.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751928Ab0J0FZW (ORCPT ); Wed, 27 Oct 2010 01:25:22 -0400 Content-Disposition: inline In-Reply-To: <1288154866.2652.84.camel@edumazet-laptop> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed, Oct 27, 2010 at 06:47:46AM +0200, Eric Dumazet wrote: > Le mercredi 27 octobre 2010 ?? 05:40 +0100, Al Viro a ??crit : > > On Wed, Oct 27, 2010 at 03:23:04PM +1100, Dave Chinner wrote: > > > From: Dave Chinner > > > > > > Now that inode state changes are protected by the inode->i_lock and > > > the inode LRU manipulations by the inode_lru_lock, we can remove the > > > inode_lock from prune_icache and the initial part of iput_final(). > > > > > > instead of using the inode_lock to protect the inode during > > > iput_final, use the inode->i_lock instead. This protects the inode > > > against new references being taken while we change the inode state > > > to I_FREEING, as well as preventing prune_icache from grabbing the > > > inode while we are manipulating it. Hence we no longer need the > > > i???ode_lock in iput_final prior to setting I_FREEING on the inode. > > ^^^^^^^^^^^^ > > > > ... the hell? There's more such damage elsewhere in the thread; what's > > going on? > > -- > > Maybe its on your side, no problem here on my copy. "i\xe1\xb9\x89ode_lock", i.e. 'n' turned into U+1E49, aka "latin small letter n with line below". I doubt that it's MTA braindamage. In the first patch there's - * invalidate_inodes - attempt to free all inodes on a + * nvalidate_inodes - attempt to free all inodes on a and I _really_ doubt that anything in mail system is capable of something that elaborate.