Re: [Qemu-devel] guest: NO_HZ FULL will not work with unstable sched clock

[Stefan Priebe <s.priebe@profihost.ag>]

same with kvm clock:

[    0.195073] Switched to clocksource kvm-clock
[    0.195509] ------------[ cut here ]------------
[    0.195509] WARNING: CPU: 1 PID: 1 at kernel/time/tick-sched.c:192 
can_stop_full_tick+0x1c1/0x200()
[    0.195515] NO_HZ FULL will not work with unstable sched clock
[    0.195518] Modules linked in:
[    0.195531] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 4.1.5+11-ph #1
[    0.195535] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), 
BIOS rel-1.8.1-0-g4adadbd-20150316_085822-nilsson.home.kraxel.org 04/01/2014
[    0.195544]  ffffffff819fc200 ffff88023fc43e88 ffffffff8162dd36 
000000000000121a
[    0.195551]  ffff88023fc43ed8 ffff88023fc43ec8 ffffffff8107dcb7 
ffff88023fc43ea8
[    0.195558]  0000000000000000 0000000000000001 ffffffff81e667c0 
ffffffff81cf1540
[    0.195564] Call Trace:
[    0.195568]  <IRQ>  [<ffffffff8162dd36>] dump_stack+0x45/0x57
[    0.195591]  [<ffffffff8107dcb7>] warn_slowpath_common+0x97/0xe0
[    0.195596]  [<ffffffff8107ddb6>] warn_slowpath_fmt+0x46/0x50
[    0.195600]  [<ffffffff810f08c1>] can_stop_full_tick+0x1c1/0x200
[    0.195604]  [<ffffffff810f0c87>] tick_nohz_irq_exit+0x97/0xd0
[    0.195609]  [<ffffffff81081c3d>] irq_exit+0xcd/0x130
[    0.195620]  [<ffffffff8103aeaa>] smp_apic_timer_interrupt+0x4a/0x60
[    0.195627]  [<ffffffff816351fe>] apic_timer_interrupt+0x6e/0x80
[    0.195631]  <EOI>  [<ffffffff81132e87>] ? 
trace_event_enum_update+0x1c7/0x3b0
[    0.195644]  [<ffffffff81132ecc>] ? trace_event_enum_update+0x20c/0x3b0
[    0.195649]  [<ffffffff81d77d01>] ? stop_trace_on_warning+0x3f/0x3f
[    0.195654]  [<ffffffff81d77e1b>] tracer_init_tracefs+0x11a/0x23b
[    0.195661]  [<ffffffff81000328>] do_one_initcall+0xd8/0x210
[    0.195669]  [<ffffffff8109b700>] ? parse_args+0x1b0/0x450
[    0.195676]  [<ffffffff810c0d23>] ? __wake_up+0x53/0x70
[    0.195684]  [<ffffffff81d5a265>] kernel_init_freeable+0x178/0x20f
[    0.195689]  [<ffffffff81d5997d>] ? do_early_param+0x8c/0x8c
[    0.195700]  [<ffffffff811c76c2>] ? create_pipe_files+0x1f2/0x200
[    0.195706]  [<ffffffff8161f6a0>] ? rest_init+0x80/0x80
[    0.195711]  [<ffffffff8161f6ae>] kernel_init+0xe/0xf0
[    0.195715]  [<ffffffff816347a2>] ret_from_fork+0x42/0x70
[    0.195719]  [<ffffffff8161f6a0>] ? rest_init+0x80/0x80
[    0.195729] ---[ end trace cf665146248feec1 ]---

Stefan

Am 15.08.2015 um 20:44 schrieb Stefan Priebe:
> Hi,
>
> while switching to a FULL tickless kernel i detected that all our VMs
> produce the following stack trace while running under qemu 2.3.0.
>
> [    0.195160] HPET: 3 timers in total, 0 timers will be used for
> per-cpu timer
> [    0.195181] hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
> [    0.195188] hpet0: 3 comparators, 64-bit 100.000000 MHz counter
> [    0.200119] Switched to clocksource hpet
> [    0.201164] ------------[ cut here ]------------
> [    0.201193] WARNING: CPU: 1 PID: 1 at kernel/time/tick-sched.c:192
> can_stop_full_tick+0x1c1/0x200()
> [    0.201199] NO_HZ FULL will not work with unstable sched clock
> [    0.201203] Modules linked in:
> [    0.201214] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 3.18.20+55-ph #1
> [    0.201219] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),
> BIOS rel-1.8.1-0-g4adadbd-20150316_085822-nilsson.home.kraxel.org
> 04/01/2014
> [    0.201226]  0000000000000009 ffff88023fc43e88 ffffffff815f5889
> 00000000000008df
> [    0.201230]  ffff88023fc43ed8 ffff88023fc43ec8 ffffffff8107433c
> ffff88023fc43ea8
> [    0.201233]  0000000000000000 0000000000000001 ffff880236c41300
> 00000000cbbde78b
> [    0.201237] Call Trace:
> [    0.201240]  <IRQ>  [<ffffffff815f5889>] dump_stack+0x46/0x58
> [    0.201260]  [<ffffffff8107433c>] warn_slowpath_common+0x8c/0xc0
> [    0.201264]  [<ffffffff81074426>] warn_slowpath_fmt+0x46/0x50
> [    0.201270]  [<ffffffff810ddc31>] can_stop_full_tick+0x1c1/0x200
> [    0.201274]  [<ffffffff810ddfc7>] tick_nohz_irq_exit+0x97/0xd0
> [    0.201277]  [<ffffffff81077f0d>] irq_exit+0xcd/0x130
> [    0.201286]  [<ffffffff8103419a>] smp_apic_timer_interrupt+0x4a/0x60
> [    0.201293]  [<ffffffff815fc4ea>] apic_timer_interrupt+0x6a/0x70
> [    0.201296]  <EOI>  [<ffffffff811bc1d5>] ? __d_lookup+0x55/0x160
> [    0.201309]  [<ffffffff811bc315>] d_lookup+0x35/0x60
> [    0.201312]  [<ffffffff811adc1b>] lookup_dcache+0x3b/0xd0
> [    0.201315]  [<ffffffff811adcd3>] __lookup_hash+0x23/0x50
> [    0.201319]  [<ffffffff811aeaae>] lookup_one_len+0xce/0x120
> [    0.201328]  [<ffffffff812d062f>] __create_file+0x8f/0x260
> [    0.201332]  [<ffffffff812d089b>] debugfs_create_dir+0x1b/0x20
> [    0.201338]  [<ffffffff8111d9a8>] event_create_dir+0x68/0x520
> [    0.201342]  [<ffffffff81d5bdb0>] event_trace_init+0x188/0x210
> [    0.201348]  [<ffffffff81113e1b>] ? trace_create_file+0x1b/0x50
> [    0.201352]  [<ffffffff81d5bc28>] ? event_trace_enable+0x1a2/0x1a2
> [    0.201359]  [<ffffffff81000328>] do_one_initcall+0xd8/0x200
> [    0.201368]  [<ffffffff81090200>] ? parse_args+0x1f0/0x470
> [    0.201373]  [<ffffffff810b1c33>] ? __wake_up+0x53/0x70
> [    0.201378]  [<ffffffff81d3e1ce>] kernel_init_freeable+0x168/0x1ff
> [    0.201381]  [<ffffffff81d3d917>] ? do_early_param+0x8c/0x8c
> [    0.201389]  [<ffffffff812f0680>] ? cryptomgr_probe+0x60/0xe0
> [    0.201398]  [<ffffffff815e6f30>] ? rest_init+0x80/0x80
> [    0.201401]  [<ffffffff815e6f3e>] kernel_init+0xe/0xf0
> [    0.201405]  [<ffffffff815fb708>] ret_from_fork+0x58/0x90
> [    0.201408]  [<ffffffff815e6f30>] ? rest_init+0x80/0x80
> [    0.201415] ---[ end trace a15ba3bc1d247b4c ]---
>
> Stefan