Hi,
We run CRIU tests for cgroup/for-next regularly, and today they
detected this warning:
[ 380.301034] ======================================================
[ 380.301096] WARNING: possible circular locking dependency detected
[ 380.301159] 4.15.0-rc1+ #1 Not tainted
[ 380.301223] ------------------------------------------------------
[ 380.301284] criu/14277 is trying to acquire lock:
[ 380.301346] (cpuset_mutex){+.+.}, at: [<ffffffffb3160191>]
cpuset_write_resmask+0x61/0x1100
[ 380.301423]
[ 380.301423] but task is already holding lock:
[ 380.301489] (cpu_hotplug_lock.rw_sem){++++}, at:
[<ffffffffb3160183>] cpuset_write_resmask+0x53/0x1100
[ 380.301559]
[ 380.301559] which lock already depends on the new lock.
[ 380.301559]
[ 380.301632]
[ 380.301632] the existing dependency chain (in reverse order) is:
[ 380.301699]
[ 380.301699] -> #1 (cpu_hotplug_lock.rw_sem){++++}:
[ 380.301770] cpus_read_lock+0x3e/0x80
[ 380.301835] static_key_slow_inc+0xe/0xa0
[ 380.301897] cpuset_css_online+0x62/0x330
[ 380.301961] online_css+0x26/0x80
[ 380.302022] cgroup_apply_control_enable+0x266/0x3d0
[ 380.302084] cgroup_mkdir+0x37d/0x4f0
[ 380.302148] kernfs_iop_mkdir+0x53/0x80
[ 380.302213] vfs_mkdir+0x10e/0x1a0
[ 380.302275] SyS_mkdirat+0xb3/0xe0
[ 380.302338] entry_SYSCALL_64_fastpath+0x23/0x9a
[ 380.302399]
[ 380.302399] -> #0 (cpuset_mutex){+.+.}:
[ 380.302468] lock_acquire+0xec/0x1e0
[ 380.302531] __mutex_lock+0x89/0x920
[ 380.302593] cpuset_write_resmask+0x61/0x1100
[ 380.302655] cgroup_file_write+0x7b/0x200
[ 380.302716] kernfs_fop_write+0x112/0x1a0
[ 380.302779] __vfs_write+0x23/0x150
[ 380.302840] vfs_write+0xc8/0x1c0
[ 380.302901] SyS_write+0x45/0xa0
[ 380.302963] entry_SYSCALL_64_fastpath+0x23/0x9a
[ 380.303023]
[ 380.303023] other info that might help us debug this:
[ 380.303023]
[ 380.303095] Possible unsafe locking scenario:
[ 380.303095]
[ 380.303162] CPU0 CPU1
[ 380.303225] ---- ----
[ 380.303285] lock(cpu_hotplug_lock.rw_sem);
[ 380.303347] lock(cpuset_mutex);
[ 380.303408] lock(cpu_hotplug_lock.rw_sem);
[ 380.303470] lock(cpuset_mutex);
[ 380.303531]
[ 380.303531] *** DEADLOCK ***
[ 380.303531]
[ 380.303609] 3 locks held by criu/14277:
[ 380.303677] #0: (sb_writers#12){.+.+}, at: [<ffffffffb3284a70>]
vfs_write+0x1a0/0x1c0
[ 380.303746] #1: (&of->mutex){+.+.}, at: [<ffffffffb3317dc2>]
kernfs_fop_write+0xe2/0x1a0
[ 380.303816] #2: (cpu_hotplug_lock.rw_sem){++++}, at:
[<ffffffffb3160183>] cpuset_write_resmask+0x53/0x1100
[ 380.303886]
[ 380.303886] stack backtrace:
[ 380.303953] CPU: 0 PID: 14277 Comm: criu Not tainted 4.15.0-rc1+ #1
[ 380.304015] Hardware name: Google Google Compute Engine/Google
Compute Engine, BIOS Google 01/01/2011
[ 380.304081] Call Trace:
[ 380.304146] dump_stack+0x7c/0xbe
[ 380.304212] print_circular_bug+0x20d/0x320
[ 380.304276] ? copy_trace+0xb0/0xb0
[ 380.304337] check_prev_add+0x41d/0x770
[ 380.304400] ? __lock_acquire+0x1091/0x1130
[ 380.304461] __lock_acquire+0x1091/0x1130
[ 380.304523] lock_acquire+0xec/0x1e0
[ 380.304586] ? cpuset_write_resmask+0x61/0x1100
[ 380.304649] __mutex_lock+0x89/0x920
[ 380.304711] ? cpuset_write_resmask+0x61/0x1100
[ 380.304774] ? cpuset_write_resmask+0x61/0x1100
[ 380.304837] ? cpuset_write_resmask+0x27f/0x1100
[ 380.304900] ? cpuset_write_resmask+0x27f/0x1100
[ 380.304962] ? cpuset_write_resmask+0x61/0x1100
[ 380.305024] cpuset_write_resmask+0x61/0x1100
[ 380.305086] ? kernfs_fop_write+0xea/0x1a0
[ 380.305148] cgroup_file_write+0x7b/0x200
[ 380.305215] kernfs_fop_write+0x112/0x1a0
[ 380.305277] __vfs_write+0x23/0x150
[ 380.305341] ? rcu_read_lock_sched_held+0x3f/0x70
[ 380.305404] ? rcu_sync_lockdep_assert+0x2a/0x50
[ 380.305466] ? __sb_start_write+0xd2/0x1b0
[ 380.305527] ? vfs_write+0x1a0/0x1c0
[ 380.305588] vfs_write+0xc8/0x1c0
[ 380.305650] SyS_write+0x45/0xa0
[ 380.305712] entry_SYSCALL_64_fastpath+0x23/0x9a
[ 380.305774] RIP: 0033:0x145971dc4330
[ 380.305835] RSP: 002b:00007ffeaffb30b8 EFLAGS: 00000246 ORIG_RAX:
0000000000000001
[ 380.305903] RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000145971dc4330
[ 380.305964] RDX: 0000000000000003 RSI: 000000000137aac0 RDI: 0000000000000003
[ 380.306026] RBP: 0000000000000000 R08: 737570632e746573 R09: 6f72702070756f72
[ 380.306087] R10: 0000000000000000 R11: 0000000000000246 R12: 000000000137b440
[ 380.306149] R13: 0000000001373448 R14: 0000000001373450 R15: 000000000137a970
[avagin@laptop linux]$ git describe cgroup/for-next
v4.15-rc1-3-gddf7005f3221
All logs are here: https://travis-ci.org/avagin/linux/jobs/308154268
Thanks,
Andrei
--
To unsubscribe from this list: send the line "unsubscribe cgroups" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
