Return-Path: linux-nfs-owner@vger.kernel.org Received: from mail-we0-f182.google.com ([74.125.82.182]:48513 "EHLO mail-we0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752114AbaHZSeJ (ORCPT ); Tue, 26 Aug 2014 14:34:09 -0400 Received: by mail-we0-f182.google.com with SMTP id k48so15106865wev.41 for ; Tue, 26 Aug 2014 11:34:08 -0700 (PDT) Message-ID: <53FCD31D.9090705@gmail.com> Date: Tue, 26 Aug 2014 21:34:05 +0300 From: Boaz Harrosh MIME-Version: 1.0 To: Boaz Harrosh , Trond Myklebust CC: "Matt W. Benjamin" , Christoph Hellwig , Linux NFS Mailing List , "Adam C. Emerson" Subject: Re: [PATCH] pnfs: Kick a pnfs_layoutcommit_inode on recall References: <53FCA183.8000605@plexistor.com> <1435166875.78.1409066662291.JavaMail.root@thunderbeast.private.linuxbox.com> <53FCBC5A.4060304@plexistor.com> <53FCBE9C.40906@plexistor.com> <53FCCFBD.3080305@plexistor.com> In-Reply-To: <53FCCFBD.3080305@plexistor.com> Content-Type: text/plain; charset=UTF-8 Sender: linux-nfs-owner@vger.kernel.org List-ID: On 08/26/2014 09:19 PM, Boaz Harrosh wrote: <> > Beep! but this is exactly what I'm trying to say. The STD specifically > forbids that. The server is not allowed to wait here, it must return > imitatively, with an error that frees the slot and then later issue the > RECALL. > > This is what I said exactly three times in my mail, and what I have > depicted in my flow: > Server async operation (mandated by the STD) > Client back-channel can be sync with for channel (Not mentioned by the STD) > BTW Both Ganesha and kpnfsd behave the same. A recall issued while in layout_get the layout-get reply will be sent first and then the recall will be sent on the wire. If the recall was sent before the receive of the layout_get then an error/success is returned imitatively without ever waiting for recall_reply to return. Thanks Boaz