From nobody Fri Oct 3 21:07:24 2025 Received: from dggsgout11.his.huawei.com (dggsgout11.his.huawei.com [45.249.212.51]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id E8AA2272E5A for ; Tue, 26 Aug 2025 07:50:36 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=45.249.212.51 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1756194640; cv=none; b=M//WBkcPU09ofCDonJDLOSfDWuf9gQFpU+MiE4DbMm2gw3TUarZI/CYsX8G2ItEKf7o7ej/434U/xQMNjgSm/xgBnrzyWwizbzwxvO9jwq3VdVRzQqiSiTne6415dvED0OixnYcIm1Y3T5gV5XxBiFX6rOQwI9iFMEbV6K0U44w= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1756194640; c=relaxed/simple; bh=kceaIyp+7/ZAzh1vrgBWSvlBf1/V10tz7Bt1vXDC+Lw=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=eSt3BEn/7Aq4ToPv1feLe0uCrOHRmtc9Cbvlo2VXAOW9xlecSFrHnfoBcKDEeCB8Dv/4SM8lJhQ+CfBdTOY3X6ouVaKtrpJq38PQ9U+Kn8YEam38/Agu99hyQEPtwc8irHzGnJ1IkPujh1nvdQ9OsOURVd+Q5MPplXKg6jVjluY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=huaweicloud.com; spf=pass smtp.mailfrom=huaweicloud.com; arc=none smtp.client-ip=45.249.212.51 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=huaweicloud.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=huaweicloud.com Received: from mail.maildlp.com (unknown [172.19.163.235]) by dggsgout11.his.huawei.com (SkyGuard) with ESMTPS id 4cB0Fm42tbzYQvWs for ; Tue, 26 Aug 2025 15:50:36 +0800 (CST) Received: from mail02.huawei.com (unknown [10.116.40.128]) by mail.maildlp.com (Postfix) with ESMTP id 1DDDF1A084A for ; Tue, 26 Aug 2025 15:50:35 +0800 (CST) Received: from huaweicloud.com (unknown [10.175.104.67]) by APP4 (Coremail) with SMTP id gCh0CgDnMY5IZ61oTGJbAQ--.24858S5; Tue, 26 Aug 2025 15:50:34 +0800 (CST) From: Zheng Qixing To: agk@redhat.com, snitzer@kernel.org, mpatocka@redhat.com, axboe@kernel.dk, ming.lei@redhat.com Cc: dm-devel@lists.linux.dev, linux-kernel@vger.kernel.org, yukuai3@huawei.com, yi.zhang@huawei.com, yangerkun@huawei.com, houtao1@huawei.com, zhengqixing@huawei.com Subject: [PATCH-next 1/2] dm: fix queue start/stop imbalance under suspend/load/resume races Date: Tue, 26 Aug 2025 15:42:03 +0800 Message-Id: <20250826074204.390111-2-zhengqixing@huaweicloud.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20250826074204.390111-1-zhengqixing@huaweicloud.com> References: <20250826074204.390111-1-zhengqixing@huaweicloud.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-CM-TRANSID: gCh0CgDnMY5IZ61oTGJbAQ--.24858S5 X-Coremail-Antispam: 1UD129KBjvJXoWxAFWxGr4DXrWfXw47tr4ktFb_yoW5XF4xpw 4UXFWakw18Cr429w409rWxuFy3ta13t3y7CFyxC3sI9a90934rZF1UJFykXrWxAFs5AFW3 JF1UZrWDua4Utr7anT9S1TB71UUUUU7qnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUPYb4IE77IF4wAFF20E14v26rWj6s0DM7CY07I20VC2zVCF04k2 6cxKx2IYs7xG6rWj6s0DM7CIcVAFz4kK6r1j6r18M28IrcIa0xkI8VA2jI8067AKxVWUGw A2048vs2IY020Ec7CjxVAFwI0_Gr0_Xr1l8cAvFVAK0II2c7xJM28CjxkF64kEwVA0rcxS w2x7M28EF7xvwVC0I7IYx2IY67AKxVW7JVWDJwA2z4x0Y4vE2Ix0cI8IcVCY1x0267AKxV WxJVW8Jr1l84ACjcxK6I8E87Iv67AKxVW0oVCq3wA2z4x0Y4vEx4A2jsIEc7CjxVAFwI0_ GcCE3s1le2I262IYc4CY6c8Ij28IcVAaY2xG8wAqx4xG64xvF2IEw4CE5I8CrVC2j2WlYx 0E2Ix0cI8IcVAFwI0_Jr0_Jr4lYx0Ex4A2jsIE14v26r1j6r4UMcvjeVCFs4IE7xkEbVWU JVW8JwACjcxG0xvY0x0EwIxGrwACI402YVCY1x02628vn2kIc2xKxwCY1x0262kKe7AKxV WUtVW8ZwCF04k20xvY0x0EwIxGrwCFx2IqxVCFs4IE7xkEbVWUJVW8JwC20s026c02F40E 14v26r1j6r18MI8I3I0E7480Y4vE14v26r106r1rMI8E67AF67kF1VAFwI0_Jw0_GFylIx kGc2Ij64vIr41lIxAIcVC0I7IYx2IY67AKxVWUJVWUCwCI42IY6xIIjxv20xvEc7CjxVAF wI0_Gr0_Cr1lIxAIcVCF04k26cxKx2IYs7xG6r1j6r1xMIIF0xvEx4A2jsIE14v26r1j6r 4UMIIF0xvEx4A2jsIEc7CjxVAFwI0_Gr0_Gr1UYxBIdaVFxhVjvjDU0xZFpf9x07jnpnQU UUUU= X-CM-SenderInfo: x2kh0wptl0x03j6k3tpzhluzxrxghudrp/ Content-Type: text/plain; charset="utf-8" From: Zheng Qixing When suspend and load run concurrently, before q->mq_ops is set in blk_mq_init_allocated_queue(), __dm_suspend() skip dm_stop_queue(). As a result, the queue's quiesce depth is not incremented. Later, once table load has finished and __dm_resume() runs, which triggers q->quiesce_depth =3D=3D0 warning in blk_mq_unquiesce_queue(): Call Trace: dm_start_queue+0x16/0x20 [dm_mod] __dm_resume+0xac/0xb0 [dm_mod] dm_resume+0x12d/0x150 [dm_mod] do_resume+0x2c2/0x420 [dm_mod] dev_suspend+0x30/0x130 [dm_mod] ctl_ioctl+0x402/0x570 [dm_mod] dm_ctl_ioctl+0x23/0x30 [dm_mod] Fix this by explicitly tracking whether the request queue was stopped in __dm_suspend() via a new DMF_QUEUE_STOPPED flag. Only call dm_start_queue() in __dm_resume() if the queue was actually stopped. Fixes: e70feb8b3e68 ("blk-mq: support concurrent queue quiesce/unquiesce") Signed-off-by: Zheng Qixing --- drivers/md/dm-core.h | 1 + drivers/md/dm.c | 8 +++++--- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/md/dm-core.h b/drivers/md/dm-core.h index c889332e533b..0070e4462ee2 100644 --- a/drivers/md/dm-core.h +++ b/drivers/md/dm-core.h @@ -162,6 +162,7 @@ struct mapped_device { #define DMF_SUSPENDED_INTERNALLY 7 #define DMF_POST_SUSPENDING 8 #define DMF_EMULATE_ZONE_APPEND 9 +#define DMF_QUEUE_STOPPED 10 =20 static inline sector_t dm_get_size(struct mapped_device *md) { diff --git a/drivers/md/dm.c b/drivers/md/dm.c index a44e8c2dccee..7222f20c1a83 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -2960,8 +2960,10 @@ static int __dm_suspend(struct mapped_device *md, st= ruct dm_table *map, * Stop md->queue before flushing md->wq in case request-based * dm defers requests to md->wq from md->queue. */ - if (dm_request_based(md)) + if (dm_request_based(md)) { dm_stop_queue(md->queue); + set_bit(DMF_QUEUE_STOPPED, &md->flags); + } =20 flush_workqueue(md->wq); =20 @@ -2983,7 +2985,7 @@ static int __dm_suspend(struct mapped_device *md, str= uct dm_table *map, if (r < 0) { dm_queue_flush(md); =20 - if (dm_request_based(md)) + if (test_and_clear_bit(DMF_QUEUE_STOPPED, &md->flags)) dm_start_queue(md->queue); =20 unlock_fs(md); @@ -3067,7 +3069,7 @@ static int __dm_resume(struct mapped_device *md, stru= ct dm_table *map) * so that mapping of targets can work correctly. * Request-based dm is queueing the deferred I/Os in its request_queue. */ - if (dm_request_based(md)) + if (test_and_clear_bit(DMF_QUEUE_STOPPED, &md->flags)) dm_start_queue(md->queue); =20 unlock_fs(md); --=20 2.39.2