From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Martin K. Petersen" Subject: Re: scsi: convert discard to REQ_TYPE_FS instead of REQ_TYPE_BLOCK_PC Date: Fri, 09 Jul 2010 12:22:51 -0400 Message-ID: References: <20100706160106C.fujita.tomonori@lab.ntt.co.jp> <20100706213136.GA21246@redhat.com> <4C33BEDF.7050602@interlog.com> <20100707004748.GA3068@redhat.com> <4C33F619.4010302@interlog.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from rcsinet10.oracle.com ([148.87.113.121]:26253 "EHLO rcsinet10.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752260Ab0GIQXl (ORCPT ); Fri, 9 Jul 2010 12:23:41 -0400 In-Reply-To: <4C33F619.4010302@interlog.com> (Douglas Gilbert's message of "Tue, 06 Jul 2010 23:35:53 -0400") Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: dgilbert@interlog.com Cc: "Martin K. Petersen" , Mike Snitzer , FUJITA Tomonori , linux-scsi@vger.kernel.org, James.Bottomley@suse.de, hch@lst.de, axboe@kernel.dk >>>>> "Doug" == Douglas Gilbert writes: >> Maximum Unmap LBA Count> 0 (in combination with the descriptor count) >> are what indicate that the device server supports UNMAP. Doug> That has been superseded by the TPU and TPWS bits in the Thin Doug> provisioning VPD page (B2h) in sbc3r22. TPU and TPWS indicate Doug> support for the UNMAP and WRITE SAME (16) with UNMAP bit ** Doug> commands respectively. *sigh* this is getting more and more convoluted. I really wish T10 had been able to agree on a single approach instead of 2. Doug> I think sbc3r22 is just flaky in that area and will be cleaned up Doug> soon. As the words stand now, in the Block limits VPD page Doug> "maximum unmap lba count" only applies to the UNMAP command while Doug> "optimal unmap granularity" applies to both the UNMAP command and Doug> the WRITE SAME(16) command. Inconsistent. And "maximum unmap lba Doug> count"==0 implying no UNMAP command is pointless given the TPU Doug> bit. I'll ask around and see what the plans are in T10 when I get back from vacation next week. And I'll try to adjust our heuristics accordingly. -- Martin K. Petersen Oracle Linux Engineering