Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1735000 > unrolled thread

[regression 4.14rc] 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it)

Started byYanko Kaneti <yaneti@declera.com>
First post2017-09-19 17:20 +0200
Last post2017-09-20 02:30 +0200
Articles 5 — 3 participants

Back to article view | Back to linux.kernel

Contents

  [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Yanko Kaneti <yaneti@declera.com> - 2017-09-19 17:20 +0200
    Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Marc Zyngier <marc.zyngier@arm.com> - 2017-09-19 17:40 +0200
      Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Yanko Kaneti <yaneti@declera.com> - 2017-09-19 17:50 +0200
        Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Marc Zyngier <marc.zyngier@arm.com> - 2017-09-19 18:00 +0200
          Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Chuck Ebbert <cebbert.lkml@gmail.com> - 2017-09-20 02:30 +0200

#1735000 — [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it)

FromYanko Kaneti <yaneti@declera.com>
Date2017-09-19 17:20 +0200
Subject[regression 4.14rc] 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it)
Message-ID<urstz-77o-15@gated-at.bofh.it>
Hello, 

Fedora rawhide config here. 
AMD FX-8370E

Bisected a problem to:
74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it) 

It seems to be causing stalls, short lived or long lived lockups very shortly after boot. 
Everything becomes jerky.

The only visible in the log indication is something like :
....
[   59.802129] clocksource: timekeeping watchdog on CPU3: Marking clocksource 'tsc' as unstable because the skew is too large:
[   59.802134] clocksource:                       'hpet' wd_now: 3326e7aa wd_last: 329956f8 mask: ffffffff
[   59.802137] clocksource:                       'tsc' cs_now: 423662bc6f cs_last: 41dfc91650 mask: ffffffffffffffff
[   59.802140] tsc: Marking TSC unstable due to clocksource watchdog
[   59.802158] TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.
[   59.802161] sched_clock: Marking unstable (59802142067, 15510)<-(59920871789, -118714277)
[   60.015604] clocksource: Switched to clocksource hpet
[   89.015994] INFO: NMI handler (perf_event_nmi_handler) took too long to run: 209.660 msecs
[   89.016003] perf: interrupt took too long (1638003 > 2500), lowering kernel.perf_event_max_sample_rate to 1000
....

Just reverting that commit on top of linus mainline cures all the symptoms


Regards
- Yanko

[toc] | [next] | [standalone]

#1735017

FromMarc Zyngier <marc.zyngier@arm.com>
Date2017-09-19 17:40 +0200
Message-ID<ursMW-7dF-15@gated-at.bofh.it>
In reply to#1735000
On 19/09/17 16:12, Yanko Kaneti wrote:
> Hello, 
> 
> Fedora rawhide config here. 
> AMD FX-8370E
> 
> Bisected a problem to:
> 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it) 
> 
> It seems to be causing stalls, short lived or long lived lockups very shortly after boot. 
> Everything becomes jerky.
> 
> The only visible in the log indication is something like :
> ....
> [   59.802129] clocksource: timekeeping watchdog on CPU3: Marking clocksource 'tsc' as unstable because the skew is too large:
> [   59.802134] clocksource:                       'hpet' wd_now: 3326e7aa wd_last: 329956f8 mask: ffffffff
> [   59.802137] clocksource:                       'tsc' cs_now: 423662bc6f cs_last: 41dfc91650 mask: ffffffffffffffff
> [   59.802140] tsc: Marking TSC unstable due to clocksource watchdog
> [   59.802158] TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.
> [   59.802161] sched_clock: Marking unstable (59802142067, 15510)<-(59920871789, -118714277)
> [   60.015604] clocksource: Switched to clocksource hpet
> [   89.015994] INFO: NMI handler (perf_event_nmi_handler) took too long to run: 209.660 msecs
> [   89.016003] perf: interrupt took too long (1638003 > 2500), lowering kernel.perf_event_max_sample_rate to 1000
> ....
> 
> Just reverting that commit on top of linus mainline cures all the symptoms
Interesting. Do you still get HPET interrupts?

Thanks,

	M.
-- 
Jazz is not dead. It just smells funny...

[toc] | [prev] | [next] | [standalone]

#1735029

FromYanko Kaneti <yaneti@declera.com>
Date2017-09-19 17:50 +0200
Message-ID<ursWC-7h8-25@gated-at.bofh.it>
In reply to#1735017
On Tue, 2017-09-19 at 16:33 +0100, Marc Zyngier wrote:
> On 19/09/17 16:12, Yanko Kaneti wrote:
> > Hello, 
> > 
> > Fedora rawhide config here. 
> > AMD FX-8370E
> > 
> > Bisected a problem to:
> > 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it) 
> > 
> > It seems to be causing stalls, short lived or long lived lockups very shortly after boot. 
> > Everything becomes jerky.
> > 
> > The only visible in the log indication is something like :
> > ....
> > [   59.802129] clocksource: timekeeping watchdog on CPU3: Marking clocksource 'tsc' as unstable because the skew is too large:
> > [   59.802134] clocksource:                       'hpet' wd_now: 3326e7aa wd_last: 329956f8 mask: ffffffff
> > [   59.802137] clocksource:                       'tsc' cs_now: 423662bc6f cs_last: 41dfc91650 mask: ffffffffffffffff
> > [   59.802140] tsc: Marking TSC unstable due to clocksource watchdog
> > [   59.802158] TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.
> > [   59.802161] sched_clock: Marking unstable (59802142067, 15510)<-(59920871789, -118714277)
> > [   60.015604] clocksource: Switched to clocksource hpet
> > [   89.015994] INFO: NMI handler (perf_event_nmi_handler) took too long to run: 209.660 msecs
> > [   89.016003] perf: interrupt took too long (1638003 > 2500), lowering kernel.perf_event_max_sample_rate to 1000
> > ....
> > 
> > Just reverting that commit on top of linus mainline cures all the symptoms
> 
> Interesting. Do you still get HPET interrupts?

Sorry, I might need some basic help here (i.e where do I count them...)

After the watchdog switches the clocksource to hpet the system is still
 somewhat alive, so I'll guess some clock is still ticking....

-Yanko

[toc] | [prev] | [next] | [standalone]

#1735039

FromMarc Zyngier <marc.zyngier@arm.com>
Date2017-09-19 18:00 +0200
Message-ID<urt6i-7kk-25@gated-at.bofh.it>
In reply to#1735029
On 19/09/17 16:40, Yanko Kaneti wrote:
> On Tue, 2017-09-19 at 16:33 +0100, Marc Zyngier wrote:
>> On 19/09/17 16:12, Yanko Kaneti wrote:
>>> Hello, 
>>>
>>> Fedora rawhide config here. 
>>> AMD FX-8370E
>>>
>>> Bisected a problem to:
>>> 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it) 
>>>
>>> It seems to be causing stalls, short lived or long lived lockups very shortly after boot. 
>>> Everything becomes jerky.
>>>
>>> The only visible in the log indication is something like :
>>> ....
>>> [   59.802129] clocksource: timekeeping watchdog on CPU3: Marking clocksource 'tsc' as unstable because the skew is too large:
>>> [   59.802134] clocksource:                       'hpet' wd_now: 3326e7aa wd_last: 329956f8 mask: ffffffff
>>> [   59.802137] clocksource:                       'tsc' cs_now: 423662bc6f cs_last: 41dfc91650 mask: ffffffffffffffff
>>> [   59.802140] tsc: Marking TSC unstable due to clocksource watchdog
>>> [   59.802158] TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.
>>> [   59.802161] sched_clock: Marking unstable (59802142067, 15510)<-(59920871789, -118714277)
>>> [   60.015604] clocksource: Switched to clocksource hpet
>>> [   89.015994] INFO: NMI handler (perf_event_nmi_handler) took too long to run: 209.660 msecs
>>> [   89.016003] perf: interrupt took too long (1638003 > 2500), lowering kernel.perf_event_max_sample_rate to 1000
>>> ....
>>>
>>> Just reverting that commit on top of linus mainline cures all the symptoms
>>
>> Interesting. Do you still get HPET interrupts?
> 
> Sorry, I might need some basic help here (i.e where do I count them...)

/proc/interrupts should display them.

> After the watchdog switches the clocksource to hpet the system is still
>  somewhat alive, so I'll guess some clock is still ticking....
Probably, but I suspect they're not hitting the right CPU, hence the
lockups.

Unfortunately, my x86-foo is pretty minimal, and I'm about to drop off
the net for a few days.

Thomas, any insight?

Thanks,

	M.
-- 
Jazz is not dead. It just smells funny...

[toc] | [prev] | [next] | [standalone]

#1735369

FromChuck Ebbert <cebbert.lkml@gmail.com>
Date2017-09-20 02:30 +0200
Message-ID<urB3R-3Vn-13@gated-at.bofh.it>
In reply to#1735039
On Tue, 19 Sep 2017 16:51:06 +0100
Marc Zyngier <marc.zyngier@arm.com> wrote:

> On 19/09/17 16:40, Yanko Kaneti wrote:
> > On Tue, 2017-09-19 at 16:33 +0100, Marc Zyngier wrote:  
> >> On 19/09/17 16:12, Yanko Kaneti wrote:  
> >>> Hello, 
> >>>
> >>> Fedora rawhide config here. 
> >>> AMD FX-8370E
> >>>
> >>> Bisected a problem to:
> >>> 74def747bcd0 (genirq: Restrict effective affinity to interrupts
> >>> actually using it) 
> >>>
> >>> It seems to be causing stalls, short lived or long lived lockups
> >>> very shortly after boot. Everything becomes jerky.
> >>>
> >>> The only visible in the log indication is something like :
> >>> ....
> >>> [   59.802129] clocksource: timekeeping watchdog on CPU3: Marking
> >>> clocksource 'tsc' as unstable because the skew is too large:
> >>> [   59.802134] clocksource:                       'hpet' wd_now:
> >>> 3326e7aa wd_last: 329956f8 mask: ffffffff [   59.802137]
> >>> clocksource:                       'tsc' cs_now: 423662bc6f
> >>> cs_last: 41dfc91650 mask: ffffffffffffffff [   59.802140] tsc:
> >>> Marking TSC unstable due to clocksource watchdog [   59.802158]
> >>> TSC found unstable after boot, most likely due to broken BIOS.
> >>> Use 'tsc=unstable'. [   59.802161] sched_clock: Marking unstable
> >>> (59802142067, 15510)<-(59920871789, -118714277) [   60.015604]
> >>> clocksource: Switched to clocksource hpet [   89.015994] INFO:
> >>> NMI handler (perf_event_nmi_handler) took too long to run:
> >>> 209.660 msecs [   89.016003] perf: interrupt took too long
> >>> (1638003 > 2500), lowering kernel.perf_event_max_sample_rate to
> >>> 1000 ....
> >>>
> >>> Just reverting that commit on top of linus mainline cures all the
> >>> symptoms  
> >>
> >> Interesting. Do you still get HPET interrupts?  
> > 
> > Sorry, I might need some basic help here (i.e where do I count
> > them...)  
> 
> /proc/interrupts should display them.
> 
> > After the watchdog switches the clocksource to hpet the system is
> > still somewhat alive, so I'll guess some clock is still
> > ticking....  
> Probably, but I suspect they're not hitting the right CPU, hence the
> lockups.
> 
> Unfortunately, my x86-foo is pretty minimal, and I'm about to drop off
> the net for a few days.
> 
> Thomas, any insight?

Looking at flat_cpu_mask_to_apicid(), I don't see how 74def747bcd0
can be correct:

	struct cpumask *effmsk =
	irq_data_get_effective_affinity_mask(irqdata); unsigned long
	cpu_mask = cpumask_bits(mask)[0] & APIC_ALL_CPUS;

	if (!cpu_mask)
		return -EINVAL;
	*apicid = (unsigned int)cpu_mask;
	cpumask_bits(effmsk)[0] = cpu_mask;

Before that patch, this function wrote to the effective mask
unconditionally. After, it only writes to effective_mask if it is
already non-zero.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web