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=-8.3 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI, SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_SANE_1 autolearn=unavailable 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 6503FC432C3 for ; Thu, 21 Nov 2019 00:53:50 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 3CB942089D for ; Thu, 21 Nov 2019 00:53:50 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="VoYmzusf" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726757AbfKUAxt (ORCPT ); Wed, 20 Nov 2019 19:53:49 -0500 Received: from us-smtp-delivery-1.mimecast.com ([205.139.110.120]:60432 "EHLO us-smtp-1.mimecast.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726658AbfKUAxt (ORCPT ); Wed, 20 Nov 2019 19:53:49 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1574297626; 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: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=HG0/OMA4IePs0RwQcFpc3v10JQC01dLQlpMZrLx9Fik=; b=VoYmzusf8qRucvaC9NJcwHK4wy56Kd7P8W9aI8dO+xwiJOqzkAag6PjW0toOFKOqthzR/t 516beU2qzbc6OTmzjaPis2K2N9mqTvYe1Mj6kLcv7C54g4b8XPCs/s4Ap1IPhM26szTt0a BcVsNnGocS1jqTssHtBjSrNijTEX/jY= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-224-5Q7zU7eZOsaArDUQb6CWuA-1; Wed, 20 Nov 2019 19:53:38 -0500 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id E455910CE792; Thu, 21 Nov 2019 00:53:36 +0000 (UTC) Received: from ming.t460p (ovpn-8-21.pek2.redhat.com [10.72.8.21]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 670046E501; Thu, 21 Nov 2019 00:53:27 +0000 (UTC) Date: Thu, 21 Nov 2019 08:53:23 +0800 From: Ming Lei To: 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 Subject: Re: [PATCH 4/4] scsi: core: don't limit per-LUN queue depth for SSD Message-ID: <20191121005323.GB24548@ming.t460p> References: <20191118103117.978-1-ming.lei@redhat.com> <20191118103117.978-5-ming.lei@redhat.com> <1081145f-3e17-9bc1-2332-50a4b5621ef7@suse.de> MIME-Version: 1.0 In-Reply-To: <1081145f-3e17-9bc1-2332-50a4b5621ef7@suse.de> User-Agent: Mutt/1.12.1 (2019-06-15) X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-MC-Unique: 5Q7zU7eZOsaArDUQb6CWuA-1 X-Mimecast-Spam-Score: 0 Content-Type: text/plain; charset=WINDOWS-1252 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Sender: linux-scsi-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-scsi@vger.kernel.org On Wed, Nov 20, 2019 at 11:05:24AM +0100, Hannes Reinecke wrote: > On 11/18/19 11:31 AM, Ming Lei wrote: > > SCSI core uses the atomic variable of sdev->device_busy to track > > in-flight IO requests dispatched to this scsi device. IO request may be > > submitted from any CPU, so the cost for maintaining the shared atomic > > counter can be very big on big NUMA machine with lots of CPU cores. > >=20 > > sdev->queue_depth is usually used for two purposes: 1) improve IO merge= ; > > 2) fair IO request scattered among all LUNs. > >=20 > > blk-mq already provides fair request allocation among all active shared > > request queues(LUNs), see hctx_may_queue(). > >=20 > > NVMe doesn't have such per-request-queue(namespace) queue depth, so it > > is reasonable to ignore the limit for SCSI SSD too. Also IO merge won't > > play big role for reaching top SSD performance. > >=20 > > With this patch, big cost for tracking in-flight per-LUN requests via > > atomic variable can be saved. > >=20 > > Given QUEUE_FLAG_NONROT is read in IO path, we have to freeze queue > > before changing this flag. > >=20 > > Cc: Sathya Prakash > > Cc: Chaitra P B > > Cc: Suganath Prabu Subramani > > Cc: Kashyap Desai > > Cc: Sumit Saxena > > Cc: Shivasharan S > > Cc: Ewan D. Milne > > Cc: Christoph Hellwig , > > Cc: Hannes Reinecke > > Cc: Bart Van Assche > > Signed-off-by: Ming Lei > > --- > > block/blk-sysfs.c | 14 +++++++++++++- > > drivers/scsi/scsi_lib.c | 24 ++++++++++++++++++------ > > drivers/scsi/sd.c | 4 ++++ > > 3 files changed, 35 insertions(+), 7 deletions(-) > >=20 > > diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c > > index fca9b158f4a0..9cc0dd5f88a1 100644 > > --- a/block/blk-sysfs.c > > +++ b/block/blk-sysfs.c > > @@ -281,6 +281,18 @@ QUEUE_SYSFS_BIT_FNS(random, ADD_RANDOM, 0); > > QUEUE_SYSFS_BIT_FNS(iostats, IO_STAT, 0); > > #undef QUEUE_SYSFS_BIT_FNS > > =20 > > +static ssize_t queue_store_nonrot_freeze(struct request_queue *q, > > +=09=09const char *page, size_t count) > > +{ > > +=09size_t ret; > > + > > +=09blk_mq_freeze_queue(q); > > +=09ret =3D queue_store_nonrot(q, page, count); > > +=09blk_mq_unfreeze_queue(q); > > + > > +=09return ret; > > +} > > + > > static ssize_t queue_zoned_show(struct request_queue *q, char *page) > > { > > =09switch (blk_queue_zoned_model(q)) { > > @@ -642,7 +654,7 @@ static struct queue_sysfs_entry queue_write_zeroes_= max_entry =3D { > > static struct queue_sysfs_entry queue_nonrot_entry =3D { > > =09.attr =3D {.name =3D "rotational", .mode =3D 0644 }, > > =09.show =3D queue_show_nonrot, > > -=09.store =3D queue_store_nonrot, > > +=09.store =3D queue_store_nonrot_freeze, > > }; > > =20 > > static struct queue_sysfs_entry queue_zoned_entry =3D { > > diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c > > index 62a86a82c38d..72655a049efd 100644 > > --- a/drivers/scsi/scsi_lib.c > > +++ b/drivers/scsi/scsi_lib.c > > @@ -354,7 +354,8 @@ void scsi_device_unbusy(struct scsi_device *sdev, s= truct scsi_cmnd *cmd) > > =09if (starget->can_queue > 0) > > =09=09atomic_dec(&starget->target_busy); > > =20 > > -=09atomic_dec(&sdev->device_busy); > > +=09if (!blk_queue_nonrot(sdev->request_queue)) > > +=09=09atomic_dec(&sdev->device_busy); > > } > > =20 > > static void scsi_kick_queue(struct request_queue *q) > > @@ -410,7 +411,8 @@ static void scsi_single_lun_run(struct scsi_device = *current_sdev) > > =20 > > static inline bool scsi_device_is_busy(struct scsi_device *sdev) > > { > > -=09if (atomic_read(&sdev->device_busy) >=3D sdev->queue_depth) > > +=09if (!blk_queue_nonrot(sdev->request_queue) && > > +=09=09=09atomic_read(&sdev->device_busy) >=3D sdev->queue_depth) > > =09=09return true; > > =09if (atomic_read(&sdev->device_blocked) > 0) > > =09=09return true; > > @@ -1283,8 +1285,12 @@ static inline int scsi_dev_queue_ready(struct re= quest_queue *q, > > =09=09=09=09 struct scsi_device *sdev) > > { > > =09unsigned int busy; > > +=09bool bypass =3D blk_queue_nonrot(sdev->request_queue); > > =20 > > -=09busy =3D atomic_inc_return(&sdev->device_busy) - 1; > > +=09if (!bypass) > > +=09=09busy =3D atomic_inc_return(&sdev->device_busy) - 1; > > +=09else > > +=09=09busy =3D 0; > > =09if (atomic_read(&sdev->device_blocked)) { > > =09=09if (busy) > > =09=09=09goto out_dec; > > @@ -1298,12 +1304,16 @@ static inline int scsi_dev_queue_ready(struct r= equest_queue *q, > > =09=09=09=09 "unblocking device at zero depth\n")); > > =09} > > =20 > > +=09if (bypass) > > +=09=09return 1; > > + > > =09if (busy >=3D sdev->queue_depth) > > =09=09goto out_dec; > > =20 > > =09return 1; > > out_dec: > > -=09atomic_dec(&sdev->device_busy); > > +=09if (!bypass) > > +=09=09atomic_dec(&sdev->device_busy); > > =09return 0; > > } > > =20 > > @@ -1624,7 +1634,8 @@ static void scsi_mq_put_budget(struct blk_mq_hw_c= tx *hctx) > > =09struct request_queue *q =3D hctx->queue; > > =09struct scsi_device *sdev =3D q->queuedata; > > =20 > > -=09atomic_dec(&sdev->device_busy); > > +=09if (!blk_queue_nonrot(sdev->request_queue)) > > +=09=09atomic_dec(&sdev->device_busy); > > } > > =20 > > static bool scsi_mq_get_budget(struct blk_mq_hw_ctx *hctx) > > @@ -1731,7 +1742,8 @@ static blk_status_t scsi_queue_rq(struct blk_mq_h= w_ctx *hctx, > > =09case BLK_STS_OK: > > =09=09break; > > =09case BLK_STS_RESOURCE: > > -=09=09if (atomic_read(&sdev->device_busy) || > > +=09=09if ((!blk_queue_nonrot(sdev->request_queue) && > > +=09=09 atomic_read(&sdev->device_busy)) || > > =09=09 scsi_device_blocked(sdev)) > > =09=09=09ret =3D BLK_STS_DEV_RESOURCE; > > =09=09break; > > diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c > > index 0744c34468e1..c3d47117700d 100644 > > --- a/drivers/scsi/sd.c > > +++ b/drivers/scsi/sd.c > > @@ -2927,7 +2927,9 @@ static void sd_read_block_characteristics(struct = scsi_disk *sdkp) > > =09rot =3D get_unaligned_be16(&buffer[4]); > > =20 > > =09if (rot =3D=3D 1) { > > +=09=09blk_mq_freeze_queue(q); > > =09=09blk_queue_flag_set(QUEUE_FLAG_NONROT, q); > > +=09=09blk_mq_unfreeze_queue(q); > > =09=09blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, q); > > =09} > > =20 > > @@ -3123,7 +3125,9 @@ static int sd_revalidate_disk(struct gendisk *dis= k) > > =09=09 * cause this to be updated correctly and any device which > > =09=09 * doesn't support it should be treated as rotational. > > =09=09 */ > > +=09=09blk_mq_freeze_queue(q); > > =09=09blk_queue_flag_clear(QUEUE_FLAG_NONROT, q); > > +=09=09blk_mq_unfreeze_queue(q); > > =09=09blk_queue_flag_set(QUEUE_FLAG_ADD_RANDOM, q); > > =20 > > =09=09if (scsi_device_supports_vpd(sdp)) { > >=20 > Hmm. >=20 > I must admit I patently don't like this explicit dependency on > blk_nonrot(). Having a conditional counter is just an open invitation to > getting things wrong... That won't be an issue given this patch freezes queue when the NONROT queue flag is changed. >=20 > 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. > But I do see that this might not be easy to do, nor necessarily > something which others agree on. > So alternatively, can't we make this a per-host flag, instead of having > it per device characteristic? Are you sure each LUN shares the same NONROT flag? I think it can be true for each LUN to have different NONROT flag. >=20 > I'm just thinking of a storage array with a massive cache; it's anyone's > guess if 'nonrot' is set there, but at the same time these things can be > fast, so they might be profiting from dropping the device_busy counter > there, too. If you are worrying about slow queue freeze, that won't be an issue, given we don't switch to percpu mode until the flag is updated during sd probe, and blk_mq_freeze_queue_wait() is pretty quick in ATOMIC mode. Thanks, Ming