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=-5.3 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,NICE_REPLY_A,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 953FCC43470 for ; Tue, 11 May 2021 01:35:08 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 78A94616EA for ; Tue, 11 May 2021 01:35:08 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230343AbhEKBgM (ORCPT ); Mon, 10 May 2021 21:36:12 -0400 Received: from mail-1.ca.inter.net ([208.85.220.69]:47891 "EHLO mail-1.ca.inter.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229628AbhEKBgJ (ORCPT ); Mon, 10 May 2021 21:36:09 -0400 Received: from localhost (offload-3.ca.inter.net [208.85.220.70]) by mail-1.ca.inter.net (Postfix) with ESMTP id 5992B2EA173; Mon, 10 May 2021 21:35:03 -0400 (EDT) Received: from mail-1.ca.inter.net ([208.85.220.69]) by localhost (offload-3.ca.inter.net [208.85.220.70]) (amavisd-new, port 10024) with ESMTP id 8FcpycSND7un; Mon, 10 May 2021 21:14:00 -0400 (EDT) Received: from [192.168.48.23] (host-45-58-219-4.dyn.295.ca [45.58.219.4]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) (Authenticated sender: dgilbert@interlog.com) by mail-1.ca.inter.net (Postfix) with ESMTPSA id 464DE2EA078; Mon, 10 May 2021 21:35:02 -0400 (EDT) Reply-To: dgilbert@interlog.com Subject: Re: [PATCH] blk-mq: Use request queue-wide tags for tagset-wide sbitmap To: Ming Lei , John Garry Cc: axboe@kernel.dk, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org, kashyap.desai@broadcom.com, chenxiang66@hisilicon.com, yama@redhat.com References: <1620037333-2495-1-git-send-email-john.garry@huawei.com> From: Douglas Gilbert Message-ID: <0faa2ad6-4ae9-02b2-2f34-cf58f1e23c5b@interlog.com> Date: Mon, 10 May 2021 21:35:01 -0400 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.10.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-CA Content-Transfer-Encoding: 7bit Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 2021-05-10 8:52 p.m., Ming Lei wrote: > On Mon, May 03, 2021 at 06:22:13PM +0800, John Garry wrote: >> The tags used for an IO scheduler are currently per hctx. >> >> As such, when q->nr_hw_queues grows, so does the request queue total IO >> scheduler tag depth. >> >> This may cause problems for SCSI MQ HBAs whose total driver depth is >> fixed. >> >> Ming and Yanhui report higher CPU usage and lower throughput in scenarios >> where the fixed total driver tag depth is appreciably lower than the total >> scheduler tag depth: >> https://lore.kernel.org/linux-block/440dfcfc-1a2c-bd98-1161-cec4d78c6dfc@huawei.com/T/#mc0d6d4f95275a2743d1c8c3e4dc9ff6c9aa3a76b >> > > No difference any more wrt. fio running on scsi_debug with this patch in > Yanhui's test machine: > > modprobe scsi_debug host_max_queue=128 submit_queues=32 virtual_gb=256 delay=1 > vs. > modprobe scsi_debug max_queue=128 submit_queues=1 virtual_gb=256 delay=1 > > Without this patch, the latter's result is 30% higher than the former's. > > note: scsi_debug's queue depth needs to be updated to 128 for avoiding io hang, > which is another scsi issue. "scsi_debug: Fix cmd_per_lun, set to max_queue" made it into lk 5.13.0-rc1 as commit fc09acb7de31badb2ea9e85d21e071be1a5736e4 . Is this the issue you are referring to, or is there a separate issue in the wider scsi stack? BTW Martin's 5.14/scsi-queue is up and running with lk 5.13.0-rc1 . Doug Gilbert