INFO: rcu detected stall in ext4_file_write_iter

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit: 1ef64dae Merge 4.9.170 into android-4.9
git tree: android-4.9
console output: https://syzkaller.appspot.com/x/log.txt?x=17069a3d200000
kernel config: https://syzkaller.appspot.com/x/.config?x=bbc07cb9576201de
dashboard link: https://syzkaller.appspot.com/bug?extid=829a98851a62c9e37e2c
compiler: gcc (GCC) 9.0.0 20181231 (experimental)
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=1581a7a0a00000

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+829a98...@syzkaller.appspotmail.com

hrtimer: interrupt took 62263 ns
sched: RT throttling activated for rt_rq 00000000ddd2b4a1 (cpu 0)
potential CPU hogs:
syz-executor.3 (2987)
INFO: rcu_preempt detected stalls on CPUs/tasks:
(detected by 0, t=10502 jiffies, g=570, c=569, q=90)
All QSes seen, last rcu_preempt kthread activity 10499
(4294964452-4294953953), jiffies_till_next_fqs=1, root ->qsmask 0x0
syz-executor.2 R running task 26504 3212 2112 0x00000000
ffff8801db607cc8 ffffffff813fcbf1 ffffffff813fc9f0 0000000000002903
00000000ffffcbe1 00000000fffff4e4 dffffc0000000000 ffff8801db621c80
ffffffff830d5980 ffff8801db607d98 ffffffff81408e6c 0000000000000000
Call Trace:
<IRQ> [ 272.854926] [<0000000089bc4351>]
sched_show_task.cold+0x27f/0x327 kernel/sched/core.c:5312
[<000000000ca14c79>] print_other_cpu_stall kernel/rcu/tree.c:1403 [inline]
[<000000000ca14c79>] check_cpu_stall kernel/rcu/tree.c:1520 [inline]
[<000000000ca14c79>] __rcu_pending kernel/rcu/tree.c:3495 [inline]
[<000000000ca14c79>] rcu_pending kernel/rcu/tree.c:3559 [inline]
[<000000000ca14c79>] rcu_check_callbacks.cold+0xcd6/0xd2f
kernel/rcu/tree.c:2888
[<00000000aef91a0d>] update_process_times+0x31/0x70
kernel/time/timer.c:1629
[<0000000060854cba>] tick_sched_handle.isra.0+0x4a/0xf0
kernel/time/tick-sched.c:151
[<00000000e690576e>] tick_sched_timer+0x7a/0x130
kernel/time/tick-sched.c:1190
[<0000000088d9e626>] __run_hrtimer kernel/time/hrtimer.c:1255 [inline]
[<0000000088d9e626>] __hrtimer_run_queues+0x313/0xe00
kernel/time/hrtimer.c:1319
[<00000000743fa032>] hrtimer_interrupt+0x1b6/0x450
kernel/time/hrtimer.c:1353
[<00000000f16888de>] local_apic_timer_interrupt+0x76/0xa0
arch/x86/kernel/apic/apic.c:937
[<0000000087370bb6>] smp_apic_timer_interrupt+0x79/0xb0
arch/x86/kernel/apic/apic.c:961
[<0000000042a7829a>] apic_timer_interrupt+0x9d/0xb0
arch/x86/entry/entry_64.S:648
<EOI> [ 272.952169] [<0000000012335029>] ? arch_local_irq_restore
arch/x86/include/asm/paravirt.h:768 [inline]
<EOI> [ 272.952169] [<0000000012335029>] ? lock_acquire+0x176/0x3d0
kernel/locking/lockdep.c:3762
[<00000000ddee93f9>] __raw_spin_lock include/linux/spinlock_api_smp.h:144
[inline]
[<00000000ddee93f9>] _raw_spin_lock+0x38/0x50 kernel/locking/spinlock.c:151
[<00000000b492c4d4>] spin_lock include/linux/spinlock.h:302 [inline]
[<00000000b492c4d4>] ext4_do_update_inode fs/ext4/inode.c:4905 [inline]
[<00000000b492c4d4>] ext4_mark_iloc_dirty+0x1d1/0x2850 fs/ext4/inode.c:5472
[<00000000261e215c>] ext4_mark_inode_dirty+0x3db/0xb30 fs/ext4/inode.c:5588
[<0000000087c10fbd>] ext4_dirty_inode+0x73/0xa0 fs/ext4/inode.c:5619
[<000000001485eb38>] __mark_inode_dirty+0x10e/0x10a0 fs/fs-writeback.c:2097
[<00000000f1fbc14a>] mark_inode_dirty include/linux/fs.h:2013 [inline]
[<00000000f1fbc14a>] generic_write_end+0x195/0x260 fs/buffer.c:2180
[<000000004fe3a8a6>] ext4_da_write_end+0x461/0xc20 fs/ext4/inode.c:3086
[<00000000960403a2>] generic_perform_write+0x2fa/0x500 mm/filemap.c:2950
[<000000009db89091>] __generic_file_write_iter+0x340/0x530
mm/filemap.c:3064
[<00000000e05d7fa8>] ext4_file_write_iter+0x66e/0xdd0 fs/ext4/file.c:165
[<000000002aa19711>] vfs_iter_write+0x2e2/0x580 fs/read_write.c:390
[<00000000dcdca8dd>] iter_file_splice_write+0x5a9/0xb10 fs/splice.c:768
[<00000000a362389d>] do_splice_from fs/splice.c:870 [inline]
[<00000000a362389d>] direct_splice_actor+0x126/0x1a0 fs/splice.c:1037
[<000000005eed3f39>] splice_direct_to_actor+0x2c8/0x820 fs/splice.c:992
[<000000009878c073>] do_splice_direct+0x1a5/0x260 fs/splice.c:1080
[<00000000834eef08>] do_sendfile+0x503/0xc00 fs/read_write.c:1401
[<00000000e8b6f267>] SYSC_sendfile64 fs/read_write.c:1462 [inline]
[<00000000e8b6f267>] SyS_sendfile64+0x145/0x160 fs/read_write.c:1448
[<000000000e528bc4>] do_syscall_64+0x1ad/0x570 arch/x86/entry/common.c:285
[<0000000084d38d86>] entry_SYSCALL_64_after_swapgs+0x5d/0xdb
rcu_preempt kthread starved for 10555 jiffies! g570 c569 f0x2
RCU_GP_WAIT_FQS(3) ->state=0x0
rcu_preempt R running task 28856 7 2 0x80000000
ffff8801da6817c0 ffff8801d1a2c780 ffff8801db721000 ffff8801cc93df00
ffff8801db721018 ffff8801da697a70 ffffffff8280185e ffff8801da6817c0
ffff8801da6979b8 00ffffff812445d1 ffff8801db7218f0 1ffff1003b4d2f3d
Call Trace:
[<00000000781472c1>] schedule+0x92/0x1c0 kernel/sched/core.c:3546
[<000000001fb81289>] schedule_timeout+0x513/0xe50 kernel/time/timer.c:1794
[<0000000093c2983c>] rcu_gp_kthread+0xba5/0x20e0 kernel/rcu/tree.c:2235
[<000000004a8d5e78>] kthread+0x278/0x310 kernel/kthread.c:211
[<00000000aa4b7d82>] ret_from_fork+0x5c/0x70 arch/x86/entry/entry_64.S:373


---
This bug is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzk...@googlegroups.com.

syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches