From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759722Ab0JHVGL (ORCPT ); Fri, 8 Oct 2010 17:06:11 -0400 Received: from cantor2.suse.de ([195.135.220.15]:60831 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759664Ab0JHVGK (ORCPT ); Fri, 8 Oct 2010 17:06:10 -0400 From: Andreas Gruenbacher Organization: SUSE Labs, Novell Inc. To: Tvrtko Ursulin Subject: Re: [BUG][PATCH][2.6.36-rc3] fanotify: Do not ignore result of permission decisions Date: Fri, 8 Oct 2010 23:05:35 +0200 User-Agent: KMail/1.13.5 (Linux/2.6.34-12-desktop; KDE/4.4.4; x86_64; ; ) Cc: Eric Paris , "linux-kernel@vger.kernel.org" , "linux-fsdevel@vger.kernel.org" References: <201009071608.02290.tvrtko.ursulin@sophos.com> <201010081500.40494.agruen@suse.de> <201010081711.24373.tvrtko.ursulin@sophos.com> In-Reply-To: <201010081711.24373.tvrtko.ursulin@sophos.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201010082305.36195.agruen@suse.de> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Friday 08 October 2010 18:11:23 Tvrtko Ursulin wrote: > On Friday 08 Oct 2010 14:00:40 Andreas Gruenbacher wrote: > > Tvrtko, > > > > On Wednesday 08 September 2010 10:24:04 Tvrtko Ursulin wrote: > > > Improved version of the fix which does not include the check > > > when permission events are not enabled in configuration and > > > stops processing if no interesting events remain. > > > > > > Current code ignores access replies to permission decisions so > > > fix it in a way which will allow all listeners to still receive > > > non permission events. > > > > I agree with the patch (see comments below), but this explanation is close > > to incomprehensible and not good as a commit message. > > How about this: > > Current code incorrectly ignores responses to permission decisions. When a > single deny response has been received record it and do not send more > permission events. However still send non-permission events to other > clients. Much better, thanks. > Also if ENOMEM, why not try sending to other listeners? My point, exactly. Thanks, Andreas