From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Christie Subject: Re: [PATCH 2/2] ceph osd: add support for new op cmpext Date: Thu, 10 Mar 2016 11:06:22 -0600 Message-ID: <56E1A98E.6030405@redhat.com> References: <[PATCH 0/2] ceph osd: initial VMware VAAI support> <1457591672-17430-1-git-send-email-mchristi@redhat.com> <1457591672-17430-3-git-send-email-mchristi@redhat.com> <20160310130341.444c76af@g21.suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Return-path: Received: from mx1.redhat.com ([209.132.183.28]:44943 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752734AbcCJRGX (ORCPT ); Thu, 10 Mar 2016 12:06:23 -0500 In-Reply-To: <20160310130341.444c76af@g21.suse.de> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: David Disseldorp Cc: ceph-devel@vger.kernel.org On 03/10/2016 06:03 AM, David Disseldorp wrote: > Hi Mike, > > On Thu, 10 Mar 2016 00:34:32 -0600, Mike Christie wrote: > >> This adds support for a new op cmpext. The request will read >> extent.length bytes and compare them to extent.length bytes at >> extent.offset on disk. If there is a miscompare the osd will return >> -EILSEQ, and the mismatched buffer that was read. >> >> rbd will use this in a multi op request to implement the >> SCSI COMPARE_AND_WRITE request which is used by VMware for >> its atomic test and set request. >> >> v2: >> - Merge David's tracing fixes. >> - Instead of returning the mismatch offset and buffer on matching >> failure just return the buffer. The client can figure out the offset >> if it needs it. > > What's your reason for dropping the mismatch offset? I was not sure if anyone else was going to use it. I can add it back. It does not matter to me.