[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] rcu_sched self-detected stall on CPU on kernel 4.4.5 in PV DomU


  • To: xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxxx>
  • From: Steven Haigh <netwiz@xxxxxxxxx>
  • Date: Thu, 17 Mar 2016 08:39:22 +1100
  • Delivery-date: Wed, 16 Mar 2016 21:39:59 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xen.org>

Hi all,

I've noticed the following problem that ends up with a non-repsonsive PV
DomU using kernel 4.4.5 under heavy disk IO:

INFO: rcu_sched self-detected stall on CPU
        0-...: (6759098 ticks this GP) idle=cb3/140000000000001/0
softirq=3244615/3244615 fqs=4
         (t=6762321 jiffies g=2275626 c=2275625 q=54)
rcu_sched kthread starved for 6762309 jiffies! g2275626 c2275625 f0x0 s3
->state=0x0
Task dump for CPU 0:
updatedb        R  running task        0  6027   6021 0x00000088
 ffffffff818d0c00 ffff88007fc03c58 ffffffff810a625f 0000000000000000
 ffffffff818d0c00 ffff88007fc03c70 ffffffff810a8699 0000000000000001
 ffff88007fc03ca0 ffffffff810d0e5a ffff88007fc170c0 ffffffff818d0c00
Call Trace:
 <IRQ>  [<ffffffff810a625f>] sched_show_task+0xaf/0x110
 [<ffffffff810a8699>] dump_cpu_task+0x39/0x40
 [<ffffffff810d0e5a>] rcu_dump_cpu_stacks+0x8a/0xc0
 [<ffffffff810d4884>] rcu_check_callbacks+0x424/0x7a0
 [<ffffffff810a91e1>] ? account_system_time+0x81/0x110
 [<ffffffff810a9481>] ? account_process_tick+0x61/0x160
 [<ffffffff810e8050>] ? tick_sched_do_timer+0x30/0x30
 [<ffffffff810d9749>] update_process_times+0x39/0x60
 [<ffffffff810e7aa6>] tick_sched_handle.isra.15+0x36/0x50
 [<ffffffff810e808d>] tick_sched_timer+0x3d/0x70
 [<ffffffff810da342>] __hrtimer_run_queues+0xf2/0x250
 [<ffffffff810da698>] hrtimer_interrupt+0xa8/0x190
 [<ffffffff8100c61e>] xen_timer_interrupt+0x2e/0x140
 [<ffffffff810c8555>] handle_irq_event_percpu+0x55/0x1e0
 [<ffffffff810cbbca>] handle_percpu_irq+0x3a/0x50
 [<ffffffff810c7d22>] generic_handle_irq+0x22/0x30
 [<ffffffff813e51af>] __evtchn_fifo_handle_events+0x15f/0x180
 [<ffffffff813e51e0>] evtchn_fifo_handle_events+0x10/0x20
 [<ffffffff813e22e3>] __xen_evtchn_do_upcall+0x43/0x80
 [<ffffffff813e3f50>] xen_evtchn_do_upcall+0x30/0x50
 [<ffffffff8165def2>] xen_hvm_callback_vector+0x82/0x90
 <EOI>  [<ffffffff810b9cd0>] ? queued_spin_lock_slowpath+0x10/0x170
 [<ffffffff8165bcc0>] _raw_spin_lock+0x20/0x30
 [<ffffffff811df221>] find_inode_fast+0x61/0xa0
 [<ffffffff811df7ce>] iget_locked+0x6e/0x170
 [<ffffffff81249203>] ext4_iget+0x33/0xae0
 [<ffffffff81658f02>] ? out_of_line_wait_on_bit+0x72/0x80
 [<ffffffff81249ce0>] ext4_iget_normal+0x30/0x40
 [<ffffffff81253415>] ext4_lookup+0xd5/0x140
 [<ffffffff811ceffd>] lookup_real+0x1d/0x50
 [<ffffffff811cf3b3>] __lookup_hash+0x33/0x40
 [<ffffffff811d1af7>] walk_component+0x177/0x280
 [<ffffffff811d2c50>] path_lookupat+0x60/0x110
 [<ffffffff811d47dc>] filename_lookup+0x9c/0x150
 [<ffffffff811a982d>] ? kfree+0x10d/0x290
 [<ffffffff8124122c>] ? call_filldir+0x9c/0x130
 [<ffffffff811d441f>] ? getname_flags+0x4f/0x1f0
 [<ffffffff811d4946>] user_path_at_empty+0x36/0x40
 [<ffffffff811ca6f3>] vfs_fstatat+0x53/0xa0
 [<ffffffff811c7919>] ? __fput+0x169/0x1d0
 [<ffffffff811cab42>] SYSC_newlstat+0x22/0x40
 [<ffffffff8110cad0>] ? __audit_syscall_exit+0x1f0/0x270
 [<ffffffff81078115>] ? syscall_slow_exit_work+0x3f/0xc0
 [<ffffffff8110c88f>] ? __audit_syscall_entry+0xaf/0x100
 [<ffffffff811cad4e>] SyS_newlstat+0xe/0x10
 [<ffffffff8165c06e>] entry_SYSCALL_64_fastpath+0x12/0x71

This ends up with the system not responding at 100% CPU usage.

Has anyone else seen this using kernel 4.4.5 in a DomU?

-- 
Steven Haigh

Email: netwiz@xxxxxxxxx
Web: https://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.