kernel/linux.git/drivers/md/dm.c, branch v5.15.208

dm: fix NULL pointer dereference in __dm_suspend()

2025-10-19T14:21:51+00:00

[ Upstream commit 8d33a030c566e1f105cd5bf27f37940b6367f3be ] There is a race condition between dm device suspend and table load that can lead to null pointer dereference. The issue occurs when suspend is invoked before table load completes: BUG: kernel NULL pointer dereference, address: 0000000000000054 Oops: 0000 [#1] PREEMPT SMP PTI CPU: 6 PID: 6798 Comm: dmsetup Not tainted 6.6.0-g7e52f5f0ca9b #62 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.1-2.fc37 04/01/2014 RIP: 0010:blk_mq_wait_quiesce_done+0x0/0x50 Call Trace: blk_mq_quiesce_queue+0x2c/0x50 dm_stop_queue+0xd/0x20 __dm_suspend+0x130/0x330 dm_suspend+0x11a/0x180 dev_suspend+0x27e/0x560 ctl_ioctl+0x4cf/0x850 dm_ctl_ioctl+0xd/0x20 vfs_ioctl+0x1d/0x50 __se_sys_ioctl+0x9b/0xc0 __x64_sys_ioctl+0x19/0x30 x64_sys_call+0x2c4a/0x4620 do_syscall_64+0x9e/0x1b0 The issue can be triggered as below: T1 T2 dm_suspend table_load __dm_suspend dm_setup_md_queue dm_mq_init_request_queue blk_mq_init_allocated_queue => q->mq_ops = set->ops; (1) dm_stop_queue / dm_wait_for_completion => q->tag_set NULL pointer! (2) => q->tag_set = set; (3) Fix this by checking if a valid table (map) exists before performing request-based suspend and waiting for target I/O. When map is NULL, skip these table-dependent suspend steps. Even when map is NULL, no I/O can reach any target because there is no table loaded; I/O submitted in this state will fail early in the DM layer. Skipping the table-dependent suspend logic in this case is safe and avoids NULL pointer dereferences. Fixes: c4576aed8d85 ("dm: fix request-based dm's use of dm_wait_for_completion") Cc: stable@vger.kernel.org Signed-off-by: Zheng Qixing Signed-off-by: Mikulas Patocka [ omitted DMF_QUEUE_STOPPED flag setting and braces absent in 5.15 ] Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

Revert "dm: requeue IO if mapping table not yet available"

2024-10-17T13:11:06+00:00

[ Upstream commit c8691cd0fc11197515ed148de0780d927bfca38b ] This reverts commit fa247089de9936a46e290d4724cb5f0b845600f5. The following sequence of commands causes a livelock - there will be workqueue process looping and consuming 100% CPU: dmsetup create --notable test truncate -s 1MiB testdata losetup /dev/loop0 testdata dmsetup load test --table '0 2048 linear /dev/loop0 0' dd if=/dev/zero of=/dev/dm-0 bs=16k count=1 conv=fdatasync The livelock is caused by the commit fa247089de99. The commit claims that it fixes a race condition, however, it is unknown what the actual race condition is and what program is involved in the race condition. When the inactive table is loaded, the nodes /dev/dm-0 and /sys/block/dm-0 are created. /dev/dm-0 has zero size at this point. When the device is suspended and resumed, the nodes /dev/mapper/test and /dev/disk/* are created. If some program opens a block device before it is created by dmsetup or lvm, the program is buggy, so dm could just report an error as it used to do before. Reported-by: Zdenek Kabelac Signed-off-by: Mikulas Patocka Fixes: fa247089de99 ("dm: requeue IO if mapping table not yet available") Signed-off-by: Sasha Levin

dm suspend: return -ERESTARTSYS instead of -EINTR

2024-09-04T11:23:31+00:00

[ Upstream commit 1e1fd567d32fcf7544c6e09e0e5bc6c650da6e23 ] This commit changes device mapper, so that it returns -ERESTARTSYS instead of -EINTR when it is interrupted by a signal (so that the ioctl can be restarted). The manpage signal(7) says that the ioctl function should be restarted if the signal was handled with SA_RESTART. Signed-off-by: Mikulas Patocka Cc: stable@vger.kernel.org Signed-off-by: Sasha Levin

dm: call the resume method on internal suspend

2024-03-26T22:21:24+00:00

[ Upstream commit 65e8fbde64520001abf1c8d0e573561b4746ef38 ] There is this reported crash when experimenting with the lvm2 testsuite. The list corruption is caused by the fact that the postsuspend and resume methods were not paired correctly; there were two consecutive calls to the origin_postsuspend function. The second call attempts to remove the "hash_list" entry from a list, while it was already removed by the first call. Fix __dm_internal_resume so that it calls the preresume and resume methods of the table's targets. If a preresume method of some target fails, we are in a tricky situation. We can't return an error because dm_internal_resume isn't supposed to return errors. We can't return success, because then the "resume" and "postsuspend" methods would not be paired correctly. So, we set the DMF_SUSPENDED flag and we fake normal suspend - it may confuse userspace tools, but it won't cause a kernel crash. ------------[ cut here ]------------ kernel BUG at lib/list_debug.c:56! invalid opcode: 0000 [#1] PREEMPT SMP CPU: 1 PID: 8343 Comm: dmsetup Not tainted 6.8.0-rc6 #4 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-2 04/01/2014 RIP: 0010:__list_del_entry_valid_or_report+0x77/0xc0 RSP: 0018:ffff8881b831bcc0 EFLAGS: 00010282 RAX: 000000000000004e RBX: ffff888143b6eb80 RCX: 0000000000000000 RDX: 0000000000000001 RSI: ffffffff819053d0 RDI: 00000000ffffffff RBP: ffff8881b83a3400 R08: 00000000fffeffff R09: 0000000000000058 R10: 0000000000000000 R11: ffffffff81a24080 R12: 0000000000000001 R13: ffff88814538e000 R14: ffff888143bc6dc0 R15: ffffffffa02e4bb0 FS: 00000000f7c0f780(0000) GS:ffff8893f0a40000(0000) knlGS:0000000000000000 CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 CR2: 0000000057fb5000 CR3: 0000000143474000 CR4: 00000000000006b0 Call Trace: ? die+0x2d/0x80 ? do_trap+0xeb/0xf0 ? __list_del_entry_valid_or_report+0x77/0xc0 ? do_error_trap+0x60/0x80 ? __list_del_entry_valid_or_report+0x77/0xc0 ? exc_invalid_op+0x49/0x60 ? __list_del_entry_valid_or_report+0x77/0xc0 ? asm_exc_invalid_op+0x16/0x20 ? table_deps+0x1b0/0x1b0 [dm_mod] ? __list_del_entry_valid_or_report+0x77/0xc0 origin_postsuspend+0x1a/0x50 [dm_snapshot] dm_table_postsuspend_targets+0x34/0x50 [dm_mod] dm_suspend+0xd8/0xf0 [dm_mod] dev_suspend+0x1f2/0x2f0 [dm_mod] ? table_deps+0x1b0/0x1b0 [dm_mod] ctl_ioctl+0x300/0x5f0 [dm_mod] dm_compat_ctl_ioctl+0x7/0x10 [dm_mod] __x64_compat_sys_ioctl+0x104/0x170 do_syscall_64+0x184/0x1b0 entry_SYSCALL_64_after_hwframe+0x46/0x4e RIP: 0033:0xf7e6aead ---[ end trace 0000000000000000 ]--- Fixes: ffcc39364160 ("dm: enhance internal suspend and resume interface") Signed-off-by: Mikulas Patocka Signed-off-by: Mike Snitzer Signed-off-by: Sasha Levin

dm: don't lock fs when the map is NULL during suspend or resume

2023-06-21T13:59:18+00:00

[ Upstream commit 2760904d895279f87196f0fa9ec570c79fe6a2e4 ] As described in commit 38d11da522aa ("dm: don't lock fs when the map is NULL in process of resume"), a deadlock may be triggered between do_resume() and do_mount(). This commit preserves the fix from commit 38d11da522aa but moves it to where it also serves to fix a similar deadlock between do_suspend() and do_mount(). It does so, if the active map is NULL, by clearing DM_SUSPEND_LOCKFS_FLAG in dm_suspend() which is called by both do_suspend() and do_resume(). Fixes: 38d11da522aa ("dm: don't lock fs when the map is NULL in process of resume") Signed-off-by: Li Lingfeng Signed-off-by: Mike Snitzer Signed-off-by: Sasha Levin

dm stats: check for and propagate alloc_percpu failure

2023-03-30T10:48:00+00:00

commit d3aa3e060c4a80827eb801fc448debc9daa7c46b upstream. Check alloc_precpu()'s return value and return an error from dm_stats_init() if it fails. Update alloc_dev() to fail if dm_stats_init() does. Otherwise, a NULL pointer dereference will occur in dm_stats_cleanup() even if dm-stats isn't being actively used. Fixes: fd2ed4d25270 ("dm: add statistics support") Cc: stable@vger.kernel.org Signed-off-by: Jiasheng Jiang Signed-off-by: Mike Snitzer Signed-off-by: Greg Kroah-Hartman

dm: add cond_resched() to dm_wq_work()

2023-03-10T08:40:06+00:00

commit 0ca44fcef241768fd25ee763b3d203b9852f269b upstream. Otherwise the while() loop in dm_wq_work() can result in a "dead loop" on systems that have preemption disabled. This is particularly problematic on single cpu systems. Cc: stable@vger.kernel.org Signed-off-by: Pingfan Liu Acked-by: Ming Lei Signed-off-by: Mike Snitzer Signed-off-by: Greg Kroah-Hartman

dm: send just one event on resize, not two

2023-03-10T08:40:05+00:00

commit 7533afa1d27ba1234146d31d2402c195cf195962 upstream. Device mapper sends an uevent when the device is suspended, using the function set_capacity_and_notify. However, this causes a race condition with udev. Udev skips scanning dm devices that are suspended. If we send an uevent while we are suspended, udev will be racing with device mapper resume code. If the device mapper resume code wins the race, udev will process the uevent after the device is resumed and it will properly scan the device. However, if udev wins the race, it will receive the uevent, find out that the dm device is suspended and skip scanning the device. This causes bugs such as systemd unmounting the device - see https://bugzilla.redhat.com/show_bug.cgi?id=2158628 This commit fixes this race. We replace the function set_capacity_and_notify with set_capacity, so that the uevent is not sent at this point. In do_resume, we detect if the capacity has changed and we pass a boolean variable need_resize_uevent to dm_kobject_uevent. dm_kobject_uevent adds "RESIZE=1" to the uevent if need_resize_uevent is set. Signed-off-by: Mikulas Patocka Tested-by: Peter Rajnoha Cc: stable@vger.kernel.org Signed-off-by: Mike Snitzer Signed-off-by: Greg Kroah-Hartman

dm: remove flush_scheduled_work() during local_exit()

2023-03-10T08:39:30+00:00

[ Upstream commit 0b22ff5360f5c4e11050b89206370fdf7dc0a226 ] Commit acfe0ad74d2e1 ("dm: allocate a special workqueue for deferred device removal") switched from using system workqueue to a single workqueue local to DM. But it didn't eliminate the call to flush_scheduled_work() that was introduced purely for the benefit of deferred device removal with commit 2c140a246dc ("dm: allow remove to be deferred"). Since DM core uses its own workqueue (and queue_work) there is no need to call flush_scheduled_work() from local_exit(). local_exit()'s destroy_workqueue(deferred_remove_workqueue) handles flushing work started with queue_work(). Fixes: acfe0ad74d2e1 ("dm: allocate a special workqueue for deferred device removal") Signed-off-by: Mike Snitzer Signed-off-by: Sasha Levin

dm: remove unnecessary assignment statement in alloc_dev()

2022-10-29T08:12:56+00:00

[ Upstream commit 99f4f5bcb975527508eb7a5e3e34bdb91d576746 ] Fixes: 74fe6ba923949 ("dm: convert to blk_alloc_disk/blk_cleanup_disk") Signed-off-by: Genjian Zhang Signed-off-by: Mike Snitzer Signed-off-by: Sasha Levin