From patchwork Thu Nov 9 06:57:05 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Damien Le Moal X-Patchwork-Id: 10050271 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 99F4B601EB for ; Thu, 9 Nov 2017 06:57:27 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 88C7A2AB38 for ; Thu, 9 Nov 2017 06:57:27 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 7DD742AB3B; Thu, 9 Nov 2017 06:57:27 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.8 required=2.0 tests=BAYES_00,DKIM_SIGNED, RCVD_IN_DNSWL_HI,T_DKIM_INVALID autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 16CB92AB38 for ; Thu, 9 Nov 2017 06:57:27 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752431AbdKIG5V (ORCPT ); Thu, 9 Nov 2017 01:57:21 -0500 Received: from esa4.hgst.iphmx.com ([216.71.154.42]:32355 "EHLO esa4.hgst.iphmx.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752498AbdKIG5R (ORCPT ); Thu, 9 Nov 2017 01:57:17 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=wdc.com; i=@wdc.com; q=dns/txt; s=dkim.wdc.com; t=1510210638; x=1541746638; h=from:to:cc:subject:date:message-id:in-reply-to: references; bh=+P1tsMT3vHgdS74AH4khbk2rzqxmssl5iCD+GHjh8Pk=; b=T2hCBDVPKysBJn43DiTj3C5z10duNZXzcw7hamf7A+U/JDSqXNPgNi4h FAvNCIhllos6SOGcMsURHGEOQ+LN4e8OEvbiDY/hsR9Lur/7K81uRltTU aHvUO1Q4PDTWwUjInKTi3PRSrL7sqdSIYizDgr5wCfvXoo8XngTUEB+j7 IkYVXuGLyB4zx6uq9l9lYUi2oyc+Un6JyJ+WXjSjwqt0thoqHSTuurapn g/+yG2Uobzel24GY6d2CU0UaZqd7eVscyfYkNsblsOc1GcabYeX29FZMJ BmEkq+QXWy64Ojed5pI0+7oFzDgRc23Mkj157BYXLmiFka3epf7pGDWJd w==; X-IronPort-AV: E=Sophos;i="5.44,368,1505750400"; d="scan'208";a="61715400" Received: from sjappemgw11.hgst.com (HELO sjappemgw12.hgst.com) ([199.255.44.62]) by ob1.hgst.iphmx.com with ESMTP; 09 Nov 2017 14:57:18 +0800 Received: from washi.fujisawa.hgst.com ([10.149.53.254]) by sjappemgw12.hgst.com with ESMTP; 08 Nov 2017 22:57:16 -0800 From: Damien Le Moal To: linux-scsi@vger.kernel.org, "Martin K . Petersen" , linux-block@vger.kernel.org, Jens Axboe Cc: Christoph Hellwig , Bart Van Assche Subject: [PATCH V8 5/7] deadline-iosched: Introduce zone locking support Date: Thu, 9 Nov 2017 15:57:05 +0900 Message-Id: <20171109065707.17210-6-damien.lemoal@wdc.com> X-Mailer: git-send-email 2.13.6 In-Reply-To: <20171109065707.17210-1-damien.lemoal@wdc.com> References: <20171109065707.17210-1-damien.lemoal@wdc.com> Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Introduce zone write locking to avoid write request reordering with zoned block devices. This is achieved using a finer selection of the next request to dispatch: 1) Any non-write request is always allowed to proceed. 2) Any write to a conventional zone is always allowed to proceed. 3) For a write to a sequential zone, the zone lock is first checked. a) If the zone is not locked, the write is allowed to proceed after its target zone is locked. b) If the zone is locked, the write request is skipped and the next request in the dispatch queue tested (back to step 1). For a write request that has locked its target zone, the zone is unlocked either when the request completes and the method deadline_request_completed() is called, or when the request is requeued using the method deadline_add_request(). Requests targeting a locked zone are always left in the scheduler queue to preserve the initial write order. If no write request can be dispatched, allow reads to be dispatched even if the write batch is not done. If the device used is not a zoned block device, or if zoned block device support is disabled, this patch does not modify deadline behavior. Signed-off-by: Damien Le Moal Reviewed-by: Christoph Hellwig Reviewed-by: Martin K. Petersen --- block/deadline-iosched.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 68 insertions(+), 3 deletions(-) diff --git a/block/deadline-iosched.c b/block/deadline-iosched.c index 81e3f0897457..9de9f156e203 100644 --- a/block/deadline-iosched.c +++ b/block/deadline-iosched.c @@ -98,6 +98,12 @@ deadline_add_request(struct request_queue *q, struct request *rq) struct deadline_data *dd = q->elevator->elevator_data; const int data_dir = rq_data_dir(rq); + /* + * This may be a requeue of a write request that has locked its + * target zone. If it is the case, this releases the zone lock. + */ + blk_req_zone_write_unlock(rq); + deadline_add_rq_rb(dd, rq); /* @@ -188,6 +194,12 @@ deadline_move_to_dispatch(struct deadline_data *dd, struct request *rq) { struct request_queue *q = rq->q; + /* + * For a zoned block device, write requests must write lock their + * target zone. + */ + blk_req_zone_write_lock(rq); + deadline_remove_request(q, rq); elv_dispatch_add_tail(q, rq); } @@ -235,13 +247,28 @@ static inline int deadline_check_fifo(struct deadline_data *dd, int ddir) static struct request * deadline_fifo_request(struct deadline_data *dd, int data_dir) { + struct request *rq; + if (WARN_ON_ONCE(data_dir != READ && data_dir != WRITE)) return NULL; if (list_empty(&dd->fifo_list[data_dir])) return NULL; - return rq_entry_fifo(dd->fifo_list[data_dir].next); + rq = rq_entry_fifo(dd->fifo_list[data_dir].next); + if (data_dir == READ || !blk_queue_is_zoned(rq->q)) + return rq; + + /* + * Look for a write request that can be dispatched, that is one with + * an unlocked target zone. + */ + list_for_each_entry(rq, &dd->fifo_list[WRITE], queuelist) { + if (blk_req_can_dispatch_to_zone(rq)) + return rq; + } + + return NULL; } /* @@ -251,10 +278,29 @@ deadline_fifo_request(struct deadline_data *dd, int data_dir) static struct request * deadline_next_request(struct deadline_data *dd, int data_dir) { + struct request *rq; + if (WARN_ON_ONCE(data_dir != READ && data_dir != WRITE)) return NULL; - return dd->next_rq[data_dir]; + rq = dd->next_rq[data_dir]; + if (!rq) + return NULL; + + if (data_dir == READ || !blk_queue_is_zoned(rq->q)) + return rq; + + /* + * Look for a write request that can be dispatched, that is one with + * an unlocked target zone. + */ + while (rq) { + if (blk_req_can_dispatch_to_zone(rq)) + return rq; + rq = deadline_latter_request(rq); + } + + return NULL; } /* @@ -288,7 +334,8 @@ static int deadline_dispatch_requests(struct request_queue *q, int force) if (reads) { BUG_ON(RB_EMPTY_ROOT(&dd->sort_list[READ])); - if (writes && (dd->starved++ >= dd->writes_starved)) + if (deadline_fifo_request(dd, WRITE) && + (dd->starved++ >= dd->writes_starved)) goto dispatch_writes; data_dir = READ; @@ -333,6 +380,13 @@ static int deadline_dispatch_requests(struct request_queue *q, int force) rq = next_rq; } + /* + * For a zoned block device, if we only have writes queued and none of + * them can be dispatched, rq will be NULL. + */ + if (!rq) + return 0; + dd->batching = 0; dispatch_request: @@ -345,6 +399,16 @@ static int deadline_dispatch_requests(struct request_queue *q, int force) return 1; } +/* + * For zoned block devices, write unlock the target zone of completed + * write requests. + */ +static void +deadline_completed_request(struct request_queue *q, struct request *rq) +{ + blk_req_zone_write_unlock(rq); +} + static void deadline_exit_queue(struct elevator_queue *e) { struct deadline_data *dd = e->elevator_data; @@ -466,6 +530,7 @@ static struct elevator_type iosched_deadline = { .elevator_merged_fn = deadline_merged_request, .elevator_merge_req_fn = deadline_merged_requests, .elevator_dispatch_fn = deadline_dispatch_requests, + .elevator_completed_req_fn = deadline_completed_request, .elevator_add_req_fn = deadline_add_request, .elevator_former_req_fn = elv_rb_former_request, .elevator_latter_req_fn = elv_rb_latter_request,