From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [patch 09/14] fs: use RCU / seqlock logic for reverse and multi-step operaitons Date: Mon, 30 Mar 2009 14:16:49 +0200 Message-ID: <874oxbnr2m.fsf@basil.nowhere.org> References: <20090329155539.275927173@nick.local0.net> <20090329155749.609663166@nick.local0.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org To: npiggin@suse.de Return-path: Received: from one.firstfloor.org ([213.235.205.2]:44854 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751253AbZC3MQx (ORCPT ); Mon, 30 Mar 2009 08:16:53 -0400 In-Reply-To: <20090329155749.609663166@nick.local0.net> (npiggin@suse.de's message of "Mon, 30 Mar 2009 02:55:48 +1100") Sender: linux-fsdevel-owner@vger.kernel.org List-ID: npiggin@suse.de writes: > The remaining usages for dcache_lock is to allow atomic, multi-step read-side > operations over the directory tree by excluding modifications to the tree. > Also, to walk in the leaf->root direction in the tree where we don't have > a natural d_lock ordering. This is the hardest bit. General thoughts: is there a way to add a self testing infrastructure to this. e.g. by having more sequence counts per object (only enabled in the debug case, so it doesn't matter when cache line bounces) and lots of checks? I suppose that would lower the work needed of actually fixing this to work significantly. -Andi -- ak@linux.intel.com -- Speaking for myself only.