From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [RFC 2/3] BKL: remove from ext2 Date: Wed, 07 Oct 2009 07:44:56 +0200 Message-ID: <87y6nnzrjb.fsf@basil.nowhere.org> References: <1254324604-20243-1-git-send-email-jblunck@suse.de> <1254324604-20243-3-git-send-email-jblunck@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-fsdevel@vger.kernel.org To: Jan Blunck Return-path: Received: from one.firstfloor.org ([213.235.205.2]:57919 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750755AbZJGFpm (ORCPT ); Wed, 7 Oct 2009 01:45:42 -0400 In-Reply-To: <1254324604-20243-3-git-send-email-jblunck@suse.de> (Jan Blunck's message of "Wed, 30 Sep 2009 17:30:03 +0200") Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Jan Blunck writes: > Remove the big kernel lock from ext2 and replace it with a per superblock > mutex. The problem is that this can easily deadlock if the code sleeps and relies on someone else making process on the same sb (BKL avoids this problem by getting transparently dropped on schedule) So it would need more auditing or like it was suggested porting from extX, X>2. -Andi -- ak@linux.intel.com -- Speaking for myself only.