Return-Path: linux-nfs-owner@vger.kernel.org Received: from aa.linuxbox.com ([69.128.83.226]:4141 "EHLO aa.linuxbox.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932326AbaHZPYj (ORCPT ); Tue, 26 Aug 2014 11:24:39 -0400 Date: Tue, 26 Aug 2014 11:24:22 -0400 (EDT) From: "Matt W. Benjamin" To: Boaz Harrosh Cc: Christoph Hellwig , Linux NFS Mailing List , "Adam C. Emerson" , Trond Myklebust Message-ID: <1435166875.78.1409066662291.JavaMail.root@thunderbeast.private.linuxbox.com> In-Reply-To: <53FCA183.8000605@plexistor.com> Subject: Re: [PATCH] pnfs: Kick a pnfs_layoutcommit_inode on recall MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Sender: linux-nfs-owner@vger.kernel.org List-ID: IIUC, the problem is the forechannel slot count, since the call you want to make synchronously is on the forechannel? Matt ----- "Boaz Harrosh" wrote: > On 08/26/2014 05:55 PM, Trond Myklebust wrote: > > On Tue, Aug 26, 2014 at 10:37 AM, Boaz Harrosh > wrote: > > > > The above can deadlock if there are no session slots available to > send > > the layoutcommit, in which case the recall won't complete, and the > > layoutget won't get a reply (which would free up the slot). > > > > What? the back-channel and the fore-channel do not use the same > slots. these are two different slots, No? > > Matt, Adam you need to chip in here. > > If it is as you say, then yes it must be as Christoph wrote it. > > And the Ganesha server must be fixed because it has a slot system per > channel. > > Thanks > Boaz -- Matt Benjamin The Linux Box 206 South Fifth Ave. Suite 150 Ann Arbor, MI 48104 http://linuxbox.com tel. 734-761-4689 fax. 734-769-8938 cel. 734-216-5309