From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Moyer Subject: Re: [PATCH] autofs4 - bad lookup fix Date: Thu, 9 Jun 2005 14:26:06 -0400 Message-ID: <17064.35262.790619.560855@segfault.boston.redhat.com> References: Reply-To: jmoyer@redhat.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: Andrew Morton , "Steinar H. Gunderson" , linux-fsdevel , Michael Blandford , Kernel Mailing List Return-path: Received: from mx1.redhat.com ([66.187.233.31]:12698 "EHLO mx1.redhat.com") by vger.kernel.org with ESMTP id S262440AbVFIS0d (ORCPT ); Thu, 9 Jun 2005 14:26:33 -0400 To: raven@themaw.net In-Reply-To: Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org ==> Regarding [PATCH] autofs4 - bad lookup fix; raven@themaw.net adds: raven> For browsable autofs maps, a mount request that arrives at the same raven> time an expire is happening can fail to perform the needed mount. raven> This happens becuase the directory exists and so the revalidate raven> succeeds when we need it to fail so that lookup is called on the raven> same dentry to do the mount. Instead lookup is called on the next raven> path component which should be whithin the mount, but the parent raven> isn't mounted. raven> The solution is to allow the revalidate to continue and perform the raven> mount as no directory creation (at mount time) is needed for raven> browsable mount entries. I have a reproducer case for this, and the patch provided fixes it in my environment. -Jeff