From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Martin K. Petersen" Subject: Re: Write same support Date: Thu, 16 Feb 2012 15:46:39 -0500 Message-ID: References: <1327969892-5090-1-git-send-email-martin.petersen@oracle.com> <20120216200202.GA27311@redhat.com> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from rcsinet15.oracle.com ([148.87.113.117]:27293 "EHLO rcsinet15.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752511Ab2BPUqr (ORCPT ); Thu, 16 Feb 2012 15:46:47 -0500 In-Reply-To: <20120216200202.GA27311@redhat.com> (Mike Snitzer's message of "Thu, 16 Feb 2012 15:02:03 -0500") Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Mike Snitzer Cc: "Martin K. Petersen" , linux-scsi@vger.kernel.org, James.Bottomley@hansenpartnership.com, jaxboe@fusionio.com >>>>> "Mike" == Mike Snitzer writes: Mike> Given the feedback you've gotten so far I'd imagine you're well on Mike> your way with a new patchset (building on your discard merge Mike> cleanup, etc). Yeah, tweaked a few things based on your comments. Mike> I think it'd be more appropriate to put patch 4 and 5 (scsi Mike> support) at the beginning of the patchset (block interfaces aren't Mike> functional without underlying scsi support). The SCSI patches depend on the block layer ditto (topology parameters need to be present, etc.). Also, we always do it this way so we can get the block layer changes in during the merge window and then add the SCSI bits to the post-merge tree. Mike> FYI, I'll bounce a message detailing the iSCSI scatter-gather NULL Mike> pointer I _always_ hit with dm-io issuing async WRITE_SAME. Interesting, please do. I've only tested on SAS and FC... -- Martin K. Petersen Oracle Linux Engineering