From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.2 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 73160C432C3 for ; Fri, 22 Nov 2019 18:14:55 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 541462071C for ; Fri, 22 Nov 2019 18:14:55 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726739AbfKVSOy (ORCPT ); Fri, 22 Nov 2019 13:14:54 -0500 Received: from mail-pl1-f196.google.com ([209.85.214.196]:42278 "EHLO mail-pl1-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726046AbfKVSOy (ORCPT ); Fri, 22 Nov 2019 13:14:54 -0500 Received: by mail-pl1-f196.google.com with SMTP id j12so3408249plt.9; Fri, 22 Nov 2019 10:14:53 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:subject:to:cc:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=55VA5utMsaCpnMBvNC7m6fZAbp1E5a9KUtFys5/MAWY=; b=bgilaGMtoySukqJ1WbOj29EzCHsSB0KdPGJ2CTINBW7sguo6XXMe8HNEs/DvDBO+5D Om8Y0A6Ts3sK6QALTWa1hW4e2zgB+Z+GKv91RXe+GC1uKtlAFYnLngpYVRDYFoI7luFT gE1eOYlB84y1mOeFHBVTzhgqOC+VczBdi9rhmmJzD38UYWDw+LpY+hBFuT6t85vrRsNZ xi2v4avkSh+Gd7WAmMJhaQSTwSsRVWKFPXmHGEEaaVye43i13m5aAw+Go6Jbb8nY7MG2 YD33xk2TXeGA0PXm2TX8iKJ6bz9KFtfb7xd1fhWgHnnT3Vzc0rR7UNLJIOrWf0ryCBKa RXeQ== X-Gm-Message-State: APjAAAV1WEIF2VEe3QR2iuBDMxhdtr7qTD86xXMvx165tfSYokUptACi u4DSd8dDxb1qTh6JiD1mM2H3G6DjqeY= X-Google-Smtp-Source: APXvYqziXlxI76ejrLEHAfcNhLzhn96hXvZK5LIgs8RNfsQqmDBNfUVfKxQg4FHBQxVNXkuIUqKOCA== X-Received: by 2002:a17:902:900b:: with SMTP id a11mr15185440plp.116.1574446493342; Fri, 22 Nov 2019 10:14:53 -0800 (PST) Received: from desktop-bart.svl.corp.google.com ([2620:15c:2cd:202:4308:52a3:24b6:2c60]) by smtp.gmail.com with ESMTPSA id f24sm3634578pjp.12.2019.11.22.10.14.51 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Fri, 22 Nov 2019 10:14:52 -0800 (PST) Subject: Re: [PATCH 4/4] scsi: core: don't limit per-LUN queue depth for SSD To: Ming Lei , Hannes Reinecke Cc: Jens Axboe , linux-block@vger.kernel.org, "James E . J . Bottomley" , "Martin K . Petersen" , linux-scsi@vger.kernel.org, Sathya Prakash , Chaitra P B , Suganath Prabu Subramani , Kashyap Desai , Sumit Saxena , Shivasharan S , "Ewan D . Milne" , Christoph Hellwig , Bart Van Assche References: <20191118103117.978-1-ming.lei@redhat.com> <20191118103117.978-5-ming.lei@redhat.com> <1081145f-3e17-9bc1-2332-50a4b5621ef7@suse.de> <20191121005323.GB24548@ming.t460p> <336f35fc-2e22-c615-9405-50297b9737ea@suse.de> <20191122080959.GC903@ming.t460p> From: Bart Van Assche Message-ID: <5f84476f-95b4-79b6-f72d-4e2de447065c@acm.org> Date: Fri, 22 Nov 2019 10:14:51 -0800 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.9.0 MIME-Version: 1.0 In-Reply-To: <20191122080959.GC903@ming.t460p> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org On 11/22/19 12:09 AM, Ming Lei wrote: > On Thu, Nov 21, 2019 at 04:45:48PM +0100, Hannes Reinecke wrote: >> On 11/21/19 1:53 AM, Ming Lei wrote: >>> On Wed, Nov 20, 2019 at 11:05:24AM +0100, Hannes Reinecke wrote: >>>> I would far prefer if we could delegate any queueing decision to the >>>> elevators, and completely drop the device_busy flag for all devices. >>> >>> If you drop it, you may create big sequential IO performance drop >>> on HDD., that is why this patch only bypasses sdev->queue_depth on >>> SSD. NVMe bypasses it because no one uses HDD. via NVMe. >>> >> I still wonder how much performance drop we actually see; what seems to >> happen is that device_busy just arbitrary pushes back to the block >> layer, giving it more time to do merging. >> I do think we can do better then that... > > For example, running the following script[1] on 4-core VM: > > ------------------------------------------ > | QD:255 | QD: 32 | > ------------------------------------------ > fio read throughput | 825MB/s | 1432MB/s| > ------------------------------------------ > > [ ... ] Hi Ming, Thanks for having shared these numbers. I think this is very useful information. Do these results show the performance drop that happens if /sys/block/.../device/queue_depth exceeds .can_queue? What I am wondering about is how important these results are in the context of this discussion. Are there any modern SCSI devices for which a SCSI LLD sets scsi_host->can_queue and scsi_host->cmd_per_lun such that the device responds with BUSY? What surprised me is that only three SCSI LLDs call scsi_track_queue_full() (mptsas, bfa, esp_scsi). Does that mean that BUSY responses from a SCSI device or HBA are rare? Thanks, Bart.