INFO: task hung in rdma_dev_exit_net

Status: upstream: reported on 2026/10/15 02:19
Reported-by: syzbot+78f4296869b73f68534f@syzkaller.appspotmail.com
First crash: 3d08h, last: 1d04h
▶ ▼ Similar bugs (7)
Kernel Title Rank 🛈 Repro Cause bisect Fix bisect Count Last Reported Patched Status
upstream INFO: task hung in rdma_dev_exit_net (2) rdma 1 6 2076d 2210d 0/29 auto-closed as invalid on 2021/05/10 22:49
upstream INFO: task hung in rdma_dev_exit_net (3) rdma 1 5 1824d 1859d 0/29 auto-closed as invalid on 2022/01/17 19:05
upstream INFO: task hung in rdma_dev_exit_net rdma 1 1 2369d 2369d 0/29 auto-closed as invalid on 2020/07/21 12:21
upstream INFO: task hung in rdma_dev_exit_net (5) rdma 1 4 1000d 1015d 0/29 auto-obsoleted due to no activity on 2024/04/21 00:36
linux-5.15 INFO: task hung in rdma_dev_exit_net 1 1 389d 389d 0/3 auto-obsoleted due to no activity on 2026/01/02 06:39
upstream INFO: task hung in rdma_dev_exit_net (4) rdma 1 7 1198d 1361d 0/29 auto-obsoleted due to no activity on 2023/10/05 19:47
upstream INFO: task hung in rdma_dev_exit_net (6) prio:high rdma 1 277 1h25m 667d 0/29 upstream: reported on 2024/12/20 08:21

Sample crash report:
INFO: task kworker/u4:4:14912 blocked for more than 143 seconds.
      Not tainted syzkaller #0
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:kworker/u4:4    state:D stack:22664 pid:14912 ppid:2      flags:0x00004000
Workqueue: netns cleanup_net
Call Trace:
 <TASK>
 context_switch kernel/sched/core.c:5382 [inline]
 __schedule+0x15ae/0x4660 kernel/sched/core.c:6701
 schedule+0xbd/0x170 kernel/sched/core.c:6775
 schedule_preempt_disabled+0x13/0x20 kernel/sched/core.c:6834
 rwsem_down_write_slowpath+0xc3f/0x1020 kernel/locking/rwsem.c:1178
 __down_write_common kernel/locking/rwsem.c:1306 [inline]
 __down_write kernel/locking/rwsem.c:1315 [inline]
 down_write+0x1b9/0x200 kernel/locking/rwsem.c:1574
 rdma_dev_exit_net+0x90/0x340 drivers/infiniband/core/device.c:1133
 ops_exit_list net/core/net_namespace.c:173 [inline]
 cleanup_net+0x6f7/0xb00 net/core/net_namespace.c:656
 process_one_work kernel/workqueue.c:2657 [inline]
 process_scheduled_works+0xa60/0x1600 kernel/workqueue.c:2734
 worker_thread+0xa5e/0xfe0 kernel/workqueue.c:2815
 kthread+0x2fa/0x390 kernel/kthread.c:388
 ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:152
 ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:293
 </TASK>

Showing all locks held in the system:
2 locks held by kworker/0:1/9:
3 locks held by kworker/u4:0/11:
3 locks held by kworker/u4:1/12:
1 lock held by khungtaskd/29:
 #0: ffffffff8df320e0 (rcu_read_lock){....}-{1:2}, at: rcu_lock_acquire include/linux/rcupdate.h:334 [inline]
 #0: ffffffff8df320e0 (rcu_read_lock){....}-{1:2}, at: rcu_read_lock include/linux/rcupdate.h:786 [inline]
 #0: ffffffff8df320e0 (rcu_read_lock){....}-{1:2}, at: debug_show_all_locks+0x55/0x290 kernel/locking/lockdep.c:6669
3 locks held by kworker/u4:5/1083:
2 locks held by kworker/u4:7/2969:
5 locks held by kworker/u4:11/3479:
1 lock held by klogd/5131:
2 locks held by kworker/u4:8/9788:
 #0: ffff888019368538 ((wq_completion)iou_exit){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:2632 [inline]
 #0: ffff888019368538 ((wq_completion)iou_exit){+.+.}-{0:0}, at: process_scheduled_works+0x975/0x1600 kernel/workqueue.c:2734
 #1: ffffc9000b46fd00 ((work_completion)(&ctx->exit_work)){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:2632 [inline]
 #1: ffffc9000b46fd00 ((work_completion)(&ctx->exit_work)){+.+.}-{0:0}, at: process_scheduled_works+0x975/0x1600 kernel/workqueue.c:2734
2 locks held by kworker/0:2/13836:
 #0: ffff888018872d38 ((wq_completion)rcu_gp){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:2632 [inline]
 #0: ffff888018872d38 ((wq_completion)rcu_gp){+.+.}-{0:0}, at: process_scheduled_works+0x975/0x1600 kernel/workqueue.c:2734
 #1: ffffc9000db9fd00 ((work_completion)(&rew->rew_work)){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:2632 [inline]
 #1: ffffc9000db9fd00 ((work_completion)(&rew->rew_work)){+.+.}-{0:0}, at: process_scheduled_works+0x975/0x1600 kernel/workqueue.c:2734
4 locks held by kworker/u4:4/14912:
 #0: ffff88801ae5d138 ((wq_completion)netns){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:2632 [inline]
 #0: ffff88801ae5d138 ((wq_completion)netns){+.+.}-{0:0}, at: process_scheduled_works+0x975/0x1600 kernel/workqueue.c:2734
 #1: ffffc900035a7d00 (net_cleanup_work){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:2632 [inline]
 #1: ffffc900035a7d00 (net_cleanup_work){+.+.}-{0:0}, at: process_scheduled_works+0x975/0x1600 kernel/workqueue.c:2734
 #2: ffffffff8f1c5810 (pernet_ops_rwsem){++++}-{3:3}, at: cleanup_net+0x140/0xb00 net/core/net_namespace.c:610
 #3: ffffffff8ef7d190 (rdma_nets_rwsem){++++}-{3:3}, at: rdma_dev_exit_net+0x90/0x340 drivers/infiniband/core/device.c:1133
6 locks held by kworker/u4:9/15432:
4 locks held by kworker/u4:10/17986:
2 locks held by getty/23404:
 #0: ffff88802ea860a0 (&tty->ldisc_sem){++++}-{0:0}, at: tty_ldisc_ref_wait+0x25/0x70 drivers/tty/tty_ldisc.c:243
 #1: ffffc9000bdaf2f0 (&ldata->atomic_read_lock){+.+.}-{3:3}, at: n_tty_read+0x428/0x1370 drivers/tty/n_tty.c:2217
2 locks held by kworker/1:5/24593:
5 locks held by syz.6.5742/24920:
8 locks held by syz.7.5754/24962:
3 locks held by kworker/u4:13/24971:
4 locks held by kworker/u4:14/24972:
 #0: ffff888018871938 ((wq_completion)events_unbound){+.+.}-{0:0}

Crashes (2):
Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Assets (help?) Manager Title
2026/10/17 06:42 linux-6.6.y cecd39af8a1a 432c0077 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan INFO: task hung in rdma_dev_exit_net
2026/10/15 02:18 linux-6.6.y a91a6aa9253c 8fb0ae89 .config console log report info [disk image] [vmlinux] [kernel image] ci2-linux-6-6-kasan INFO: task hung in rdma_dev_exit_net
* Struck through repros no longer work on HEAD.