Return-Path: Received: from fieldses.org ([173.255.197.46]:60876 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756359AbbKEWDq (ORCPT ); Thu, 5 Nov 2015 17:03:46 -0500 Date: Thu, 5 Nov 2015 17:03:46 -0500 From: "J. Bruce Fields" To: Andrew W Elble Cc: linux-nfs@vger.kernel.org, jlayton@poochiereds.net Subject: Re: [PATCH RFCv2] nfsd: fix race with open / open upgrade stateids Message-ID: <20151105220346.GK9210@fieldses.org> References: <1446657225-72903-1-git-send-email-aweits@rit.edu> <20151105212108.GH9210@fieldses.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: Sender: linux-nfs-owner@vger.kernel.org List-ID: On Thu, Nov 05, 2015 at 04:54:45PM -0500, Andrew W Elble wrote: > > > Is there some lock imbalance?: > > Hmmmm, I'll have to poke at that a bit. > > > When you send a v2 patch, would you mind also describing what's changed? > > If you stick the description here (between the --- and the diff), it'll > > be discarded when git applies the patch (which is what we want). > > Ah. In this case, v1 missed unlocking the rwsem if nfs4_upgrade_open() > returned status. OK. Just to make sure, I checked again, and I was indeed testing with your v2 (on top of some patches of my own which I'm fairly certain weren't the cause, but I'll double check that as well). --b.