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 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4CAFCC43334 for ; Tue, 26 Jul 2022 02:48:34 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231214AbiGZCs3 (ORCPT ); Mon, 25 Jul 2022 22:48:29 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43218 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230471AbiGZCs2 (ORCPT ); Mon, 25 Jul 2022 22:48:28 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 5CE01222B0 for ; Mon, 25 Jul 2022 19:48:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1658803705; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=TrVML+RmKiQO+KVpqG2+JGeEF1NTrW9VY/z2r6IxLSg=; b=G48XoUhTpODceI9/k9OLO0mzzQMoGe9/QeWZ2vUVlTBKUjc/N69IqDPtVEhvCuslFAg3oK O2V0qDcv4bXGoO9twq18gX2IEAG34tEcNILq6cdO1gmmm35H6DWHnt2XI52r8rmvrqT3Uy PMDxmWiZ82q46I0rgp6gbATkaNPDl2s= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-520-SJwa8gNbODmGhHnmbGbOTw-1; Mon, 25 Jul 2022 22:48:22 -0400 X-MC-Unique: SJwa8gNbODmGhHnmbGbOTw-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.rdu2.redhat.com [10.11.54.2]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id A7535801585; Tue, 26 Jul 2022 02:48:21 +0000 (UTC) Received: from T590 (ovpn-8-27.pek2.redhat.com [10.72.8.27]) by smtp.corp.redhat.com (Postfix) with ESMTPS id D57F440357BA; Tue, 26 Jul 2022 02:48:15 +0000 (UTC) Date: Tue, 26 Jul 2022 10:48:10 +0800 From: Ming Lei To: Yu Kuai Cc: axboe@kernel.dk, osandov@fb.com, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, yukuai3@huawei.com, yi.zhang@huawei.com, ming.lei@redhat.com Subject: Re: [PATCH] blk-mq: fix io hung due to missing commit_rqs while scheduler is none Message-ID: References: <20220726023852.3413784-1-yukuai1@huaweicloud.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20220726023852.3413784-1-yukuai1@huaweicloud.com> X-Scanned-By: MIMEDefang 2.84 on 10.11.54.2 Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org On Tue, Jul 26, 2022 at 10:38:52AM +0800, Yu Kuai wrote: > From: Yu Kuai > > Currently, in virtio_scsi, if 'bd->last' is not set to true while > dispatching request, such io will stay in driver's queue, and driver > will wait for block layer to dispatch more rqs. However, if block > layer failed to dispatch more rq, it should trigger commit_rqs to > inform driver. > > There is a problem in blk_mq_try_issue_list_directly() that commit_rqs > won't be called: > > // assume that queue_depth is set to 1, list contains two rq > blk_mq_try_issue_list_directly > blk_mq_request_issue_directly > // dispatch first rq > // last is false > __blk_mq_try_issue_directly > blk_mq_get_dispatch_budget > // succeed to get first budget > __blk_mq_issue_directly > scsi_queue_rq > cmd->flags |= SCMD_LAST > virtscsi_queuecommand > kick = (sc->flags & SCMD_LAST) != 0 > // kick is false, first rq won't issue to disk > queued++ > > blk_mq_request_issue_directly > // dispatch second rq > __blk_mq_try_issue_directly > blk_mq_get_dispatch_budget > // failed to get second budget > ret == BLK_STS_RESOURCE > blk_mq_request_bypass_insert > // errors is still 0 > > if (!list_empty(list) || errors && ...) > // won't pass, commit_rqs won't be called > > In this situation, first rq relied on second rq to dispatch, while > second rq relied on first rq to complete, thus they will both hung. > > Fix the problem by also treat 'BLK_STS_RESOURCE' and > 'BLK_STS_DEV_RESOURCE' as 'errors' in blk_mq_try_issue_list_directly(), > so that 'commit_rqs' will be called when dispatch of last rq failed. > > Fixes: d666ba98f849 ("blk-mq: add mq_ops->commit_rqs()") > Signed-off-by: Yu Kuai > --- > block/blk-mq.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/block/blk-mq.c b/block/blk-mq.c > index 70177ee74295..752b0fe4c128 100644 > --- a/block/blk-mq.c > +++ b/block/blk-mq.c > @@ -2680,6 +2680,7 @@ void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx, > list_del_init(&rq->queuelist); > ret = blk_mq_request_issue_directly(rq, list_empty(list)); > if (ret != BLK_STS_OK) { > + errors++; OK, it is because that list becomes empty, but .queue_rq(last_rq_in_list) returns BLK_STS_*RESOURCE, but scsi can't call ->queuecommand() for this real last rq. Then blk_mq_try_issue_list_directly() doesn't call ->commit_rqs() too. Here errors means that request not queued successfully, so this patch looks fine. Also I think blk_mq_dispatch_rq_list() needs similar handling too, right? thanks, Ming