From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mta0.migadu.com (out-185.mta0.migadu.com [91.218.175.185]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7D1BD32BF4B for ; Sat, 19 Sep 2026 10:47:13 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=91.218.175.185 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789814836; cv=none; b=kuVOzNW0TJeWXCWu5pOpimy8R+wkr3FAJNG0w1vDm22l9S6Oqw6WaARQ5VrtIL6ZctxyR7cdpfBpk29aJanuOK31ncPsLsxzUpVlmce4jSQrimacQDW4tQ2cZilyMg+FZULr0rgLp9C28YbtbpyWSYv0eXgwA+8g5t9YAAgiM30= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789814836; c=relaxed/simple; bh=WP2pU0lcLiLTGhnsY0C/w1WlODK/RB5avIeRttjFroY=; h=From:To:Cc:Subject:Date:Message-ID:MIME-Version; b=ZIh+M18sailBZIlQ2KTH3otYCAhI/dLEyv5DHdx6FINEOEbvQqvM9PWKY6q3VZpbZwTPRQrh6keH4Fny+cjzokt60fbvnUW4lx81z3equyOUHHp8d/oPkHhZk5zl1yVgm9A0heZe/dyh+H1GBqm/YaZM0BLw9+3ohsFLwcVk9t8= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b=C9TEDVgm; arc=none smtp.client-ip=91.218.175.185 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b="C9TEDVgm" X-Envelope-To: linux-kernel@vger.kernel.org DKIM-Signature: a=rsa-sha256; bh=WP2pU0lcLiLTGhnsY0C/w1WlODK/RB5avIeRttjFroY=; c=simple/simple; d=linux.dev; h=from:to:subject:date:message-id:mime-version:content-type; s=key1; t=1789814831; v=1; x=1790419631; b=C9TEDVgmB4ZV7KJdGjHVNm0LTA1tqETKZk18vNSKNZqV8kKS8hOtFPtPEtqRxzeu3NhBb7aL tX8sDgDYwiFGq3yGX0kDIFC9dWrpOqZsTAJ8/XWlW54wtbCfAmgO5guPmjBMc7WOqohJGu9q9xm awTFDS+d/RUqnYNRj7ioe59s= X-Envelope-To: linux-kernel@vger.kernel.org Received: by smtp.migadu.com with ESMTPS id 72562e3c9ca8669f; Sat, 19 Sep 2026 10:47:11 +0000 X-Mizu-Trace-ID: 72562e3c9ca8669f X-Migadu-Flow: FLOW_OUT From: Tao Cui To: Bart Van Assche , axboe@kernel.dk, hch@lst.de, Tetsuo Handa Cc: cui.tao@linux.dev, cuitao@kylinos.cn, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH v4] loop: defer the queue limits clear to a workqueue Date: Sat, 19 Sep 2026 18:47:00 +0800 Message-ID: <20260919104700.1578247-1-cui.tao@linux.dev> X-Mailer: git-send-email 2.43.0 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit loop_clear_limits() calls queue_limits_commit_update() directly from the loop workqueue that processes the request. That does a non-atomic struct assignment to q->limits without freezing the queue, which races with lockless readers of q->limits on other CPUs - bio splitting reads max_hw_sectors, the discard path reads max_hw_discard_sectors - and can let them observe torn values. The trigger is a discard or write-zeroes request on a loop device whose backing file does not support the corresponding fallocate operation. The code already has an XXX comment saying this should move to a workqueue. Do that: schedule a work item on the system workqueue, where it is safe to freeze the queue around the limits update. The pending modes and the rebind generation live under a new mutex, lo->clear_limits_lock. The work item takes the mutex with the queue frozen, so a rebind cannot slip in between the generation check and the limits commit. Rebinding the device drops the accumulated modes and invalidates an already scheduled work item, so a stale clear cannot hit the new backing file. The work item is cancelled before the device is freed. Suggested-by: Bart Van Assche Signed-off-by: Tao Cui --- Changes since v3: - Replace the three atomic variables (clear_limits_mode, rebind_gen, clear_limits_gen) with plain variables protected by a new clear_limits_lock mutex, as suggested by Bart. The mutex is taken after the queue freeze in the work item, which keeps the same freeze -> mutex ordering as loop_change_fd(), the only rebinding path that freezes the queue. Changes since v2: - Skip the clear when the device was rebound since the work item was scheduled: the modes used to be captured before the freeze, so a LOOP_CHANGE_FD completing in between could apply the old modes to the new backing file. The rebind generation is now checked with the queue frozen, which excludes loop_change_fd() because it assigns the new backing file under the same freeze, so a rebind cannot slip in between the check and the commit. Cancelling from loop_assign_backing_file() would instead deadlock on the freeze held by loop_change_fd(). - Also bump the rebind generation from __loop_clr_fd(): unbinding does not go through loop_assign_backing_file(), so a work item scheduled before the last close could otherwise commit a stale clear to the queue limits of the unbound device. Changes since v1: - Reset clear_limits_mode when assigning a new backing file, so stale modes do not clear limits of the new file. - Cancel the work item from loop_remove() before del_gendisk(): the queue can already be in RCU-delayed freeing when lo_free_disk() cancels it. Tested on x86-64 (qemu, vfat-backed loop device): a 30s discard and reconfigure loop exercises the clear path 260 times, no torn sysfs reads, no difference against the unpatched kernel. Link: https://lore.kernel.org/r/20260828072004.273519-1-cui.tao@linux.dev/ --- drivers/block/loop.c | 45 +++++++++++++++++++++++++++++--------------- 1 file changed, 30 insertions(+), 15 deletions(-) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 9f6e518e60d28..d6460f359c0a8 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -68,9 +68,10 @@ struct loop_device { struct list_head idle_worker_list; struct rb_root worker_tree; struct work_struct clear_limits_work; - atomic_t clear_limits_mode; - atomic_t rebind_gen; - atomic_t clear_limits_gen; + struct mutex clear_limits_lock; + unsigned int clear_limits_mode; + unsigned int rebind_gen; + unsigned int clear_limits_gen; struct timer_list timer; bool sysfs_inited; @@ -241,11 +242,13 @@ static void loop_clear_limits_workfn(struct work_struct *work) * under the freeze; the other rebinding paths, loop_configure() * and __loop_clr_fd(), do not freeze the queue, but they bump * rebind_gen and reset clear_limits_mode themselves, so the - * generation check and the atomic_xchg() below cover them. + * generation check and the reset below cover them. */ memflags = blk_mq_freeze_queue(lo->lo_queue); - if (atomic_read(&lo->clear_limits_gen) == atomic_read(&lo->rebind_gen)) { - mode = atomic_xchg(&lo->clear_limits_mode, 0); + mutex_lock(&lo->clear_limits_lock); + if (lo->clear_limits_gen == lo->rebind_gen) { + mode = lo->clear_limits_mode; + lo->clear_limits_mode = 0; if (mode & FALLOC_FL_ZERO_RANGE) lim.max_write_zeroes_sectors = 0; @@ -255,14 +258,17 @@ static void loop_clear_limits_workfn(struct work_struct *work) lim.discard_granularity = 0; } } + mutex_unlock(&lo->clear_limits_lock); queue_limits_commit_update(lo->lo_queue, &lim); blk_mq_unfreeze_queue(lo->lo_queue, memflags); } static void loop_clear_limits(struct loop_device *lo, int mode) { - atomic_set(&lo->clear_limits_gen, atomic_read(&lo->rebind_gen)); - atomic_or(mode, &lo->clear_limits_mode); + mutex_lock(&lo->clear_limits_lock); + lo->clear_limits_gen = lo->rebind_gen; + lo->clear_limits_mode |= mode; + mutex_unlock(&lo->clear_limits_lock); schedule_work(&lo->clear_limits_work); } @@ -538,8 +544,10 @@ static int loop_validate_file(struct file *file, struct block_device *bdev) static void loop_assign_backing_file(struct loop_device *lo, struct file *file) { lo->lo_backing_file = file; - atomic_inc(&lo->rebind_gen); - atomic_set(&lo->clear_limits_mode, 0); + mutex_lock(&lo->clear_limits_lock); + lo->rebind_gen++; + lo->clear_limits_mode = 0; + mutex_unlock(&lo->clear_limits_lock); lo->old_gfp_mask = mapping_gfp_mask(file->f_mapping); mapping_set_gfp_mask(file->f_mapping, lo->old_gfp_mask & ~(__GFP_IO | __GFP_FS)); @@ -1168,13 +1176,16 @@ static void __loop_clr_fd(struct loop_device *lo) spin_lock_irq(&lo->lo_lock); filp = lo->lo_backing_file; lo->lo_backing_file = NULL; + spin_unlock_irq(&lo->lo_lock); + /* * Invalidate any pending clear that was scheduled against the old * backing file, like loop_assign_backing_file() does on rebind. */ - atomic_inc(&lo->rebind_gen); - atomic_set(&lo->clear_limits_mode, 0); - spin_unlock_irq(&lo->lo_lock); + mutex_lock(&lo->clear_limits_lock); + lo->rebind_gen++; + lo->clear_limits_mode = 0; + mutex_unlock(&lo->clear_limits_lock); lo->lo_device = NULL; lo->lo_offset = 0; @@ -1813,6 +1824,7 @@ static void lo_free_disk(struct gendisk *disk) timer_shutdown_sync(&lo->timer); cancel_work_sync(&lo->clear_limits_work); mutex_destroy(&lo->lo_mutex); + mutex_destroy(&lo->clear_limits_lock); kfree(lo); } @@ -2132,6 +2144,7 @@ static int loop_add(int i) spin_lock_init(&lo->lo_work_lock); INIT_WORK(&lo->rootcg_work, loop_rootcg_workfn); INIT_WORK(&lo->clear_limits_work, loop_clear_limits_workfn); + mutex_init(&lo->clear_limits_lock); INIT_LIST_HEAD(&lo->rootcg_cmd_list); disk->major = LOOP_MAJOR; disk->first_minor = i << part_shift; @@ -2170,8 +2183,10 @@ static int loop_add(int i) static void loop_remove(struct loop_device *lo) { - /* Cancel early: the queue may be in RCU-delayed freeing - * by the time lo_free_disk() runs. */ + /* + * Cancel early: the queue may already be in RCU-delayed freeing + * by the time lo_free_disk() cancels the work item. + */ cancel_work_sync(&lo->clear_limits_work); /* Make this loop device unreachable from pathname. */ -- 2.43.0