From: Peter Staubach Subject: Re: Should truncated READDIR replies return -EIO? Date: Fri, 08 Feb 2008 10:42:05 -0500 Message-ID: <47AC784D.1020003@redhat.com> References: <1202483082-5334-1-git-send-email-jlayton@redhat.com> <1202483596.8914.13.camel@heimdal.trondhjem.org> <1202483883.10337.2.camel@heimdal.trondhjem.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Cc: Jeff Layton , linux-nfs@vger.kernel.org To: Trond Myklebust Return-path: Received: from mx1.redhat.com ([66.187.233.31]:49765 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756783AbYBHPmP (ORCPT ); Fri, 8 Feb 2008 10:42:15 -0500 In-Reply-To: <1202483883.10337.2.camel-rJ7iovZKK19ZJLDQqaL3InhyD016LWXt@public.gmane.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: Trond Myklebust wrote: > On Fri, 2008-02-08 at 10:13 -0500, Trond Myklebust wrote: > > >> So whereas I agree that it might be correct to flag a READDIR reply that >> contains no entries due to XDR encoding bugs, I'm not sure that we >> should be flagging errors in the case where the XDR is correct. >> > > Another useful error change in the readdir code is that we may consider > flagging an ENAMETOOLONG error instead of EIO in the cases where the > filename is too large for us to parse. This brings up another question that I've had -- how to correctly differentiate between XDR decoding errors and issues having to do with local implementation limitations. I think that it would be good to be able to make a determination, to properly know how to do recovery from the situation. Thanx... ps