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 bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id AB6E5C0218D for ; Wed, 29 Jan 2025 09:55:17 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:List-Subscribe:List-Help :List-Post:List-Archive:List-Unsubscribe:List-Id:Content-Transfer-Encoding: Content-Type:In-Reply-To:From:References:Cc:To:Subject:MIME-Version:Date: Message-ID:Reply-To:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=usmjl7ABJRKotUzXG1g98T82o0226T/3yDvMFxw9pTY=; b=QJS4+xoVStbD0PS76/5zNCBQ7q 5hM+7INcEx2aunSfX+otTJzMsw9Qtf2tDiwEMjCnEp8g3KNpB9ysE3LJmFb170EL/jq9ZhLYKaimx mF6mL/2bFyTQp6UJg87BY1vzBTgrQ/uPtJYkd7t0H+MLU7d9h+RZ15+SlHFLPQx0v5ac/yslWJWfR SNZWe23v8GMZ6M0wpktoVr7WRmKbugKvsOOBpQ0ANb9giK8Di0HgetgOZ4/itgALZnc+6/MqOFAU0 zr52FtN2UAbFj99gbtFms9jLWmSudzZ/t1bXfpOZBnRu9iiNcI6vzFSrSqcDzjGBq7lCST8RESkUN 7Z0Vn9Wg==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.98 #2 (Red Hat Linux)) id 1td4mv-00000006h9z-1hCW; Wed, 29 Jan 2025 09:55:13 +0000 Received: from mx0b-001b2d01.pphosted.com ([148.163.158.5]) by bombadil.infradead.org with esmtps (Exim 4.98 #2 (Red Hat Linux)) id 1td4mp-00000006h9c-3APF for linux-nvme@lists.infradead.org; Wed, 29 Jan 2025 09:55:09 +0000 Received: from pps.filterd (m0353725.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 50T21IAc029637; Wed, 29 Jan 2025 09:54:59 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=cc :content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pp1; bh=usmjl7 ABJRKotUzXG1g98T82o0226T/3yDvMFxw9pTY=; b=S69PLcWWqcpiDgXDcVKjNE O+QG7CPlpqlE2AnHa0AcE2cCiuc3+fLNOvVhttBKsdKa7r4Q0Kelv6FhZBF0blMa E3/IIr0I5XZHpPMbJrwq8fva1TvH9XgwV5+LqKQcM7cYpCq7jHSWuvYWniBM7hxC /fQ/cTIClK0kqMVUhmCrBbH0H7S5JICvXTgJDzpz+lZkBudzXHaE7kOXjZ2mWgG6 huvuicH2a99EPnnSFLT7DbxiRiOzhmc7u0RBaKprQvXRUsrKMP/Bkgt1rqnkDmrX sgAuqPMNQnO7zX7Yq9546JpPm3FyTk5+NEqspNNnwO5Z+LM/7NuLRLA/ENFbbRZg == Received: from ppma11.dal12v.mail.ibm.com (db.9e.1632.ip4.static.sl-reverse.com [50.22.158.219]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 44fb609hg7-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 29 Jan 2025 09:54:58 +0000 (GMT) Received: from pps.filterd (ppma11.dal12v.mail.ibm.com [127.0.0.1]) by ppma11.dal12v.mail.ibm.com (8.18.1.2/8.18.1.2) with ESMTP id 50T6P3Zc018922; Wed, 29 Jan 2025 09:54:58 GMT Received: from smtprelay02.wdc07v.mail.ibm.com ([172.16.1.69]) by ppma11.dal12v.mail.ibm.com (PPS) with ESMTPS id 44dd01fksn-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 29 Jan 2025 09:54:58 +0000 Received: from smtpav06.dal12v.mail.ibm.com (smtpav06.dal12v.mail.ibm.com [10.241.53.105]) by smtprelay02.wdc07v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 50T9svG731195802 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 29 Jan 2025 09:54:57 GMT Received: from smtpav06.dal12v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 3494258055; Wed, 29 Jan 2025 09:54:57 +0000 (GMT) Received: from smtpav06.dal12v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 2061A58043; Wed, 29 Jan 2025 09:54:54 +0000 (GMT) Received: from [9.109.198.253] (unknown [9.109.198.253]) by smtpav06.dal12v.mail.ibm.com (Postfix) with ESMTP; Wed, 29 Jan 2025 09:54:53 +0000 (GMT) Message-ID: <669a9e1c-bde2-4323-b997-cdbd82a26eab@linux.ibm.com> Date: Wed, 29 Jan 2025 15:24:52 +0530 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH 3/3] blk-mq: fix wait condition for tagset wait completed check To: Daniel Wagner , Keith Busch , Jens Axboe , Christoph Hellwig , Sagi Grimberg , Ming Lei Cc: James Smart , Hannes Reinecke , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, linux-block@vger.kernel.org References: <20250128-nvme-misc-fixes-v1-0-40c586581171@kernel.org> <20250128-nvme-misc-fixes-v1-3-40c586581171@kernel.org> Content-Language: en-US From: Nilay Shroff In-Reply-To: <20250128-nvme-misc-fixes-v1-3-40c586581171@kernel.org> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-TM-AS-GCONF: 00 X-Proofpoint-GUID: d2EY-MP-VCV9R6ZjNtuHoq6fqyixjhZ_ X-Proofpoint-ORIG-GUID: d2EY-MP-VCV9R6ZjNtuHoq6fqyixjhZ_ X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1057,Hydra:6.0.680,FMLib:17.12.68.34 definitions=2025-01-28_04,2025-01-27_01,2024-11-22_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 impostorscore=0 malwarescore=0 clxscore=1011 lowpriorityscore=0 mlxlogscore=999 bulkscore=0 spamscore=0 suspectscore=0 phishscore=0 priorityscore=1501 mlxscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2411120000 definitions=main-2501290077 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20250129_015507_937001_D0E3AF13 X-CRM114-Status: GOOD ( 28.41 ) X-BeenThere: linux-nvme@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "Linux-nvme" Errors-To: linux-nvme-bounces+linux-nvme=archiver.kernel.org@lists.infradead.org On 1/28/25 10:04 PM, Daniel Wagner wrote: > blk_mq_tagset_count_completed_reqs returns the number of completed > requests. The only user of this function is > blk_mq_tagset_wait_completed_request which wants to know how many > request are not yet completed. Thus return the number of in flight > requests and terminate the wait loop when there is no inflight request. > > Fixes: f9934a80f91d ("blk-mq: introduce blk_mq_tagset_wait_completed_request()") > Cc: Ming Lei > Signed-off-by: Daniel Wagner > --- > block/blk-mq-tag.c | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) > > diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c > index b9f417d980b46d54b74dec8adcb5b04e6a78635c..3ce46afb65e3c3de9f11ca440bf0f335f21d0998 100644 > --- a/block/blk-mq-tag.c > +++ b/block/blk-mq-tag.c > @@ -450,11 +450,11 @@ void blk_mq_tagset_busy_iter(struct blk_mq_tag_set *tagset, > } > EXPORT_SYMBOL(blk_mq_tagset_busy_iter); > > -static bool blk_mq_tagset_count_completed_rqs(struct request *rq, void *data) > +static bool blk_mq_tagset_count_inflight_rqs(struct request *rq, void *data) > { > unsigned *count = data; > > - if (blk_mq_request_completed(rq)) > + if (blk_mq_rq_state(rq) == MQ_RQ_IN_FLIGHT) > (*count)++; > return true; > } > @@ -472,7 +472,7 @@ void blk_mq_tagset_wait_completed_request(struct blk_mq_tag_set *tagset) > unsigned count = 0; > > blk_mq_tagset_busy_iter(tagset, > - blk_mq_tagset_count_completed_rqs, &count); > + blk_mq_tagset_count_inflight_rqs, &count); > if (!count) > break; > msleep(5); > I see that blk_mq_tagset_wait_completed_request() is called from nvme_cancel_tagset() and nvme_cancel_admin_tagset(). And it seems to me that the intention here's to wait until each completed requests are freed (or change its state to MQ_RQ_IDLE). Looking at code, the nvme_cancel_xxx() first invokes blk_mq_tagset_busy_iter() which iterates through tagset and cancels each in-flight request and marks the request state to MQ_RQ_COMPLETE. Next in blk_mq_tagset_wait_completed_request(), we wait for each completed request state changed to anything but MQ_RQ_COMPLETE. The next state of the request would be naturally MQ_RQ_IDLE once that request is freed. So in blk_mq_tagset_ wait_completed_request(), essentially we wait for request state change from MQ_RQ_COMPLETE to MQ_RQ_IDLE. So now if the proposal is that blk_mq_tagset_wait_completed_request() has to wait only if there's any in-flight request then it seems this function would never need to wait and looks redundant because req->state would never be MQ_RQ_IN_FLIGHT as those would have been already changed to MQ_RQ_COMPLETE when nvme_cancel_xxx() invokes blk_mq_tagset_ busy_iter(ctrl->tagset, nvme_cancel_request, ctrl). Having said that, I am not sure what was the real intention here, in nvme_cancel_xxx(), do we really need to wait only until in-flight requests are completed or synchronize with request's completion callback (i.e. wait until all completed requests are freed)? Thanks, --Nilay