Re: printk() + memory offline deadlock (WAS Re: page_alloc.shuffle=1 + CONFIG_PROVE_LOCKING=y = arm64 hang)
From: Sergey Senozhatsky
Date: Wed Sep 18 2019 - 11:51:06 EST
On (09/18/19 10:39), Qian Cai wrote:
> > Perhaps for a quick fix (and a comment that says this needs to be fixed
> > properly). I think the changes to printk() that was discussed at
> > Plumbers may also solve this properly.
>
> I assume that the new printk() stuff will also fix this deadlock between
> printk() and memory offline.
Mother chicken...
Do you actually see a deadlock? I'd rather expect a lockdep splat, but
anyway...
> [ 317.337595] WARNING: possible circular locking dependency detected
> [ 317.337596] 5.3.0-next-20190917+ #9 Not tainted
> [ 317.337597] ------------------------------------------------------
> [ 317.337597] test.sh/8738 is trying to acquire lock:
> [ 317.337598] ffffffffb33a4978 ((console_sem).lock){-.-.}, at:> down_trylock+0x16/0x50
>
> [ 317.337602] but task is already holding lock:
> [ 317.337602] ffff88883fff4318 (&(&zone->lock)->rlock){-.-.}, at:> start_isolate_page_range+0x1f7/0x570
>
> [ 317.337606] which lock already depends on the new lock.
>
> [ 317.337608] the existing dependency chain (in reverse order) is:
>
> [ 317.337609] -> #3 (&(&zone->lock)->rlock){-.-.}:
> [ 317.337612] __lock_acquire+0x5b3/0xb40
> [ 317.337613] lock_acquire+0x126/0x280
> [ 317.337613] _raw_spin_lock+0x2f/0x40
> [ 317.337614] rmqueue_bulk.constprop.21+0xb6/0x1160
> [ 317.337615] get_page_from_freelist+0x898/0x22c0
> [ 317.337616] __alloc_pages_nodemask+0x2f3/0x1cd0
> [ 317.337617] alloc_page_interleave+0x18/0x130
> [ 317.337618] alloc_pages_current+0xf6/0x110
> [ 317.337619] allocate_slab+0x4c6/0x19c0
> [ 317.337620] new_slab+0x46/0x70
> [ 317.337621] ___slab_alloc+0x58b/0x960
> [ 317.337621] __slab_alloc+0x43/0x70
> [ 317.337622] kmem_cache_alloc+0x354/0x460
> [ 317.337623] fill_pool+0x272/0x4b0
> [ 317.337624] __debug_object_init+0x86/0x790
> [ 317.337624] debug_object_init+0x16/0x20
> [ 317.337625] hrtimer_init+0x27/0x1e0
> [ 317.337626] init_dl_task_timer+0x20/0x40
> [ 317.337627] __sched_fork+0x10b/0x1f0
> [ 317.337627] init_idle+0xac/0x520
> [ 317.337628] idle_thread_get+0x7c/0xc0
> [ 317.337629] bringup_cpu+0x1a/0x1e0
> [ 317.337630] cpuhp_invoke_callback+0x197/0x1120
> [ 317.337630] _cpu_up+0x171/0x280
> [ 317.337631] do_cpu_up+0xb1/0x120
> [ 317.337632] cpu_up+0x13/0x20
> [ 317.337632] smp_init+0xa4/0x12d
> [ 317.337633] kernel_init_freeable+0x37e/0x76e
> [ 317.337634] kernel_init+0x11/0x12f
> [ 317.337635] ret_from_fork+0x3a/0x50
So you have debug objects enabled. Right? This thing does not behave
when it comes to printing. debug_objects are slightly problematic.
This thing does
rq->lock --> zone->lock
It takes rq->lock and then calls into __sched_fork()->hrtimer_init()->debug_objects()->MM
This doesn't look very right - a dive into MM under rq->lock.
Peter, Thomas am I wrong?
> [ 317.337635] -> #2 (&rq->lock){-.-.}:
> [ 317.337638] __lock_acquire+0x5b3/0xb40
> [ 317.337639] lock_acquire+0x126/0x280
> [ 317.337639] _raw_spin_lock+0x2f/0x40
> [ 317.337640] task_fork_fair+0x43/0x200
> [ 317.337641] sched_fork+0x29b/0x420
> [ 317.337642] copy_process+0xf3c/0x2fd0
> [ 317.337642] _do_fork+0xef/0x950
> [ 317.337643] kernel_thread+0xa8/0xe0
> [ 317.337644] rest_init+0x28/0x311
> [ 317.337645] arch_call_rest_init+0xe/0x1b
> [ 317.337645] start_kernel+0x6eb/0x724
> [ 317.337646] x86_64_start_reservations+0x24/0x26
> [ 317.337647] x86_64_start_kernel+0xf4/0xfb
> [ 317.337648] secondary_startup_64+0xb6/0xc0
pi_lock --> rq->lock
> [ 317.337649] -> #1 (&p->pi_lock){-.-.}:
> [ 317.337651] __lock_acquire+0x5b3/0xb40
> [ 317.337652] lock_acquire+0x126/0x280
> [ 317.337653] _raw_spin_lock_irqsave+0x3a/0x50
> [ 317.337653] try_to_wake_up+0xb4/0x1030
> [ 317.337654] wake_up_process+0x15/0x20
> [ 317.337655] __up+0xaa/0xc0
> [ 317.337655] up+0x55/0x60
> [ 317.337656] __up_console_sem+0x37/0x60
> [ 317.337657] console_unlock+0x3a0/0x750
> [ 317.337658] vprintk_emit+0x10d/0x340
> [ 317.337658] vprintk_default+0x1f/0x30
> [ 317.337659] vprintk_func+0x44/0xd4
> [ 317.337660] printk+0x9f/0xc5
> [ 317.337660] crng_reseed+0x3cc/0x440
> [ 317.337661] credit_entropy_bits+0x3e8/0x4f0
> [ 317.337662] random_ioctl+0x1eb/0x250
> [ 317.337663] do_vfs_ioctl+0x13e/0xa70
> [ 317.337663] ksys_ioctl+0x41/0x80
> [ 317.337664] __x64_sys_ioctl+0x43/0x4c
> [ 317.337665] do_syscall_64+0xcc/0x76c
> [ 317.337666] entry_SYSCALL_64_after_hwframe+0x49/0xbe
console_sem->lock --> pi_lock
This also covers console_sem->lock --> rq->lock, and maintains
pi_lock --> rq->lock
So we have
console_sem->lock --> pi_lock --> rq->lock
> [ 317.337667] -> #0 ((console_sem).lock){-.-.}:
> [ 317.337669] check_prev_add+0x107/0xea0
> [ 317.337670] validate_chain+0x8fc/0x1200
> [ 317.337671] __lock_acquire+0x5b3/0xb40
> [ 317.337671] lock_acquire+0x126/0x280
> [ 317.337672] _raw_spin_lock_irqsave+0x3a/0x50
> [ 317.337673] down_trylock+0x16/0x50
> [ 317.337674] __down_trylock_console_sem+0x2b/0xa0
> [ 317.337675] console_trylock+0x16/0x60
> [ 317.337676] vprintk_emit+0x100/0x340
> [ 317.337677] vprintk_default+0x1f/0x30
> [ 317.337678] vprintk_func+0x44/0xd4
> [ 317.337678] printk+0x9f/0xc5
> [ 317.337679] __dump_page.cold.2+0x73/0x210
> [ 317.337680] dump_page+0x12/0x50
> [ 317.337680] has_unmovable_pages+0x3e9/0x4b0
> [ 317.337681] start_isolate_page_range+0x3b4/0x570
> [ 317.337682] __offline_pages+0x1ad/0xa10
> [ 317.337683] offline_pages+0x11/0x20
> [ 317.337683] memory_subsys_offline+0x7e/0xc0
> [ 317.337684] device_offline+0xd5/0x110
> [ 317.337685] state_store+0xc6/0xe0
> [ 317.337686] dev_attr_store+0x3f/0x60
> [ 317.337686] sysfs_kf_write+0x89/0xb0
> [ 317.337687] kernfs_fop_write+0x188/0x240
> [ 317.337688] __vfs_write+0x50/0xa0
> [ 317.337688] vfs_write+0x105/0x290
> [ 317.337689] ksys_write+0xc6/0x160
> [ 317.337690] __x64_sys_write+0x43/0x50
> [ 317.337691] do_syscall_64+0xcc/0x76c
> [ 317.337691] entry_SYSCALL_64_after_hwframe+0x49/0xbe
zone->lock --> console_sem->lock
So then we have
zone->lock --> console_sem->lock --> pi_lock --> rq->lock
vs. the reverse chain
rq->lock --> console_sem->lock
If I get this right.
> [ 317.337693] other info that might help us debug this:
>
> [ 317.337694] Chain exists of:
> [ 317.337694] (console_sem).lock --> &rq->lock --> &(&zone->lock)->rlock
>
> [ 317.337699] Possible unsafe locking scenario:
>
> [ 317.337700] CPU0 CPU1
> [ 317.337701] ---- ----
> [ 317.337701] lock(&(&zone->lock)->rlock);
> [ 317.337703] lock(&rq->lock);
> [ 317.337705] lock(&(&zone->lock)->rlock);
> [ 317.337706] lock((console_sem).lock);
>
> [ 317.337708] *** DEADLOCK ***
>
> [ 317.337710] 8 locks held by test.sh/8738:
> [ 317.337710] #0: ffff8883940b5408 (sb_writers#4){.+.+}, at: vfs_write+0x25f/0x290
> [ 317.337713] #1: ffff889fce310280 (&of->mutex){+.+.}, at: kernfs_fop_write+0x128/0x240
> [ 317.337716] #2: ffff889feb6d4830 (kn->count#115){.+.+}, at: kernfs_fop_write+0x138/0x240
> [ 317.337720] #3: ffffffffb3762d40 (device_hotplug_lock){+.+.}, at: lock_device_hotplug_sysfs+0x16/0x50
> [ 317.337723] #4: ffff88981f0dc990 (&dev->mutex){....}, at: device_offline+0x70/0x110
> [ 317.337726] #5: ffffffffb3315250 (cpu_hotplug_lock.rw_sem){++++}, at: __offline_pages+0xbf/0xa10
> [ 317.337729] #6: ffffffffb35408b0 (mem_hotplug_lock.rw_sem){++++}, at: percpu_down_write+0x87/0x2f0
> [ 317.337732] #7: ffff88883fff4318 (&(&zone->lock)->rlock){-.-.}, at: start_isolate_page_range+0x1f7/0x570
> [ 317.337736] stack backtrace:
> [ 317.337737] CPU: 58 PID: 8738 Comm: test.sh Not tainted 5.3.0-next-20190917+ #9
> [ 317.337738] Hardware name: HPE ProLiant DL560 Gen10/ProLiant DL560 Gen10, BIOS U34 05/21/2019
> [ 317.337739] Call Trace:
> [ 317.337739] dump_stack+0x86/0xca
> [ 317.337740] print_circular_bug.cold.31+0x243/0x26e
> [ 317.337741] check_noncircular+0x29e/0x2e0
> [ 317.337742] ? debug_lockdep_rcu_enabled+0x4b/0x60
> [ 317.337742] ? print_circular_bug+0x120/0x120
> [ 317.337743] ? is_ftrace_trampoline+0x9/0x20
> [ 317.337744] ? kernel_text_address+0x59/0xc0
> [ 317.337744] ? __kernel_text_address+0x12/0x40
> [ 317.337745] check_prev_add+0x107/0xea0
> [ 317.337746] validate_chain+0x8fc/0x1200
> [ 317.337746] ? check_prev_add+0xea0/0xea0
> [ 317.337747] ? format_decode+0xd6/0x600
> [ 317.337748] ? file_dentry_name+0xe0/0xe0
> [ 317.337749] __lock_acquire+0x5b3/0xb40
> [ 317.337749] lock_acquire+0x126/0x280
> [ 317.337750] ? down_trylock+0x16/0x50
> [ 317.337751] ? vprintk_emit+0x100/0x340
> [ 317.337752] _raw_spin_lock_irqsave+0x3a/0x50
> [ 317.337753] ? down_trylock+0x16/0x50
> [ 317.337753] down_trylock+0x16/0x50
> [ 317.337754] ? vprintk_emit+0x100/0x340
> [ 317.337755] __down_trylock_console_sem+0x2b/0xa0
> [ 317.337756] console_trylock+0x16/0x60
> [ 317.337756] vprintk_emit+0x100/0x340
> [ 317.337757] vprintk_default+0x1f/0x30
> [ 317.337758] vprintk_func+0x44/0xd4
> [ 317.337758] printk+0x9f/0xc5
> [ 317.337759] ? kmsg_dump_rewind_nolock+0x64/0x64
> [ 317.337760] ? __dump_page+0x1d7/0x430
> [ 317.337760] __dump_page.cold.2+0x73/0x210
> [ 317.337761] dump_page+0x12/0x50
> [ 317.337762] has_unmovable_pages+0x3e9/0x4b0
> [ 317.337763] start_isolate_page_range+0x3b4/0x570
> [ 317.337763] ? unset_migratetype_isolate+0x280/0x280
> [ 317.337764] ? rcu_read_lock_bh_held+0xc0/0xc0
> [ 317.337765] __offline_pages+0x1ad/0xa10
> [ 317.337765] ? lock_acquire+0x126/0x280
> [ 317.337766] ? __add_memory+0xc0/0xc0
> [ 317.337767] ? __kasan_check_write+0x14/0x20
> [ 317.337767] ? __mutex_lock+0x344/0xcd0
> [ 317.337768] ? _raw_spin_unlock_irqrestore+0x49/0x50
> [ 317.337769] ? device_offline+0x70/0x110
> [ 317.337770] ? klist_next+0x1c1/0x1e0
> [ 317.337770] ? __mutex_add_waiter+0xc0/0xc0
> [ 317.337771] ? klist_next+0x10b/0x1e0
> [ 317.337772] ? klist_iter_exit+0x16/0x40
> [ 317.337772] ? device_for_each_child+0xd0/0x110
> [ 317.337773] offline_pages+0x11/0x20
> [ 317.337774] memory_subsys_offline+0x7e/0xc0
> [ 317.337774] device_offline+0xd5/0x110
> [ 317.337775] ? auto_online_blocks_show+0x70/0x70
> [ 317.337776] state_store+0xc6/0xe0
> [ 317.337776] dev_attr_store+0x3f/0x60
> [ 317.337777] ? device_match_name+0x40/0x40
> [ 317.337778] sysfs_kf_write+0x89/0xb0
> [ 317.337778] ? sysfs_file_ops+0xa0/0xa0
> [ 317.337779] kernfs_fop_write+0x188/0x240
> [ 317.337780] __vfs_write+0x50/0xa0
> [ 317.337780] vfs_write+0x105/0x290
> [ 317.337781] ksys_write+0xc6/0x160
> [ 317.337782] ? __x64_sys_read+0x50/0x50
> [ 317.337782] ? do_syscall_64+0x79/0x76c
> [ 317.337783] ? do_syscall_64+0x79/0x76c
> [ 317.337784] __x64_sys_write+0x43/0x50
> [ 317.337784] do_syscall_64+0xcc/0x76c
> [ 317.337785] ? trace_hardirqs_on_thunk+0x1a/0x20
> [ 317.337786] ? syscall_return_slowpath+0x210/0x210
> [ 317.337787] ? entry_SYSCALL_64_after_hwframe+0x3e/0xbe
> [ 317.337787] ? trace_hardirqs_off_caller+0x3a/0x150
> [ 317.337788] ? trace_hardirqs_off_thunk+0x1a/0x20
> [ 317.337789] entry_SYSCALL_64_after_hwframe+0x49/0xbe
Lovely.
-ss