From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:35191 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752377AbcKQCsj (ORCPT ); Wed, 16 Nov 2016 21:48:39 -0500 To: Keith Busch Cc: Chaitanya Kulkarni , axboe@fb.com, martin.petersen@oracle.com, linux-nvme@lists.infradead.org, linux-block@vger.kernel.org, Chaitanya Kulkarni Subject: Re: [PATCH 2/5] block: add support for REQ_OP_WRITE_ZEROES From: "Martin K. Petersen" References: <1479279039-25818-1-git-send-email-chaitanya.kulkarni@hgst.com> <1479279039-25818-3-git-send-email-chaitanya.kulkarni@hgst.com> <20161116175906.GA10050@localhost.localdomain> Date: Wed, 16 Nov 2016 21:48:02 -0500 In-Reply-To: <20161116175906.GA10050@localhost.localdomain> (Keith Busch's message of "Wed, 16 Nov 2016 12:59:07 -0500") Message-ID: MIME-Version: 1.0 Content-Type: text/plain Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org >>>>> "Keith" == Keith Busch writes: Keith> Your maximum bi_size exceeds the 2-bytes an NVMe Write Zeroes Keith> command provides for the block count. Instead of having a simple Keith> queue flag for write zeroes support, have it take a max sectors Keith> value instead. Yes, please. Make it a queue limit and make sure it is properly stacked. The rest looks sensible to me. I'll wire up the SCSI pieces. -- Martin K. Petersen Oracle Linux Engineering From mboxrd@z Thu Jan 1 00:00:00 1970 From: martin.petersen@oracle.com (Martin K. Petersen) Date: Wed, 16 Nov 2016 21:48:02 -0500 Subject: [PATCH 2/5] block: add support for REQ_OP_WRITE_ZEROES In-Reply-To: <20161116175906.GA10050@localhost.localdomain> (Keith Busch's message of "Wed, 16 Nov 2016 12:59:07 -0500") References: <1479279039-25818-1-git-send-email-chaitanya.kulkarni@hgst.com> <1479279039-25818-3-git-send-email-chaitanya.kulkarni@hgst.com> <20161116175906.GA10050@localhost.localdomain> Message-ID: >>>>> "Keith" == Keith Busch writes: Keith> Your maximum bi_size exceeds the 2-bytes an NVMe Write Zeroes Keith> command provides for the block count. Instead of having a simple Keith> queue flag for write zeroes support, have it take a max sectors Keith> value instead. Yes, please. Make it a queue limit and make sure it is properly stacked. The rest looks sensible to me. I'll wire up the SCSI pieces. -- Martin K. Petersen Oracle Linux Engineering