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


Groups > linux.kernel > #1742787 > unrolled thread

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

Started byThorsten Leemhuis <regressions@leemhuis.info>
First post2017-10-01 14:50 +0200
Last post2017-10-01 15:20 +0200
Articles 3 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Thorsten Leemhuis <regressions@leemhuis.info> - 2017-10-01 14:50 +0200
    Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Yanko Kaneti <yaneti@declera.com> - 2017-10-01 15:10 +0200
      Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective  affinity to interrupts actually using it) Thorsten Leemhuis <regressions@leemhuis.info> - 2017-10-01 15:20 +0200

#1742787 — Re: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it)

FromThorsten Leemhuis <regressions@leemhuis.info>
Date2017-10-01 14:50 +0200
SubjectRe: [regression 4.14rc] 74def747bcd0 (genirq: Restrict effective affinity to interrupts actually using it)
Message-ID<uvLR0-2em-7@gated-at.bofh.it>
Hi, the regression tracker here. What's the status of this issue? Was
the problem fixed? It seems nothing happened for more than 10 days -- or
did the discussion move somewhere else? Ciao, Thorsten

On 20.09.2017 02:30, Chuck Ebbert wrote:
> 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.
> 
> 
> http://news.gmane.org/find-root.php?message_id=20170919203044.560cb9f1%40gmail.com 
> http://mid.gmane.org/20170919203044.560cb9f1%40gmail.com
> 

[toc] | [next] | [standalone]


#1742791

FromYanko Kaneti <yaneti@declera.com>
Date2017-10-01 15:10 +0200
Message-ID<uvMam-2AV-15@gated-at.bofh.it>
In reply to#1742787
On Sun, 2017-10-01 at 14:46 +0200, Thorsten Leemhuis wrote:
> Hi, the regression tracker here. What's the status of this issue? Was
> the problem fixed? It seems nothing happened for more than 10 days -- or
> did the discussion move somewhere else? Ciao, Thorsten

The commit was reverted last week before rc2

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=0551968add53777fddd18f4ffb4e3bbc1f646d79

Thanks for tracking it
-Yanko

> 
> On 20.09.2017 02:30, Chuck Ebbert wrote:
> > 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.
> > 
> > 
> > http://news.gmane.org/find-root.php?message_id=20170919203044.560cb9f1%40gmail.com 
> > http://mid.gmane.org/20170919203044.560cb9f1%40gmail.com
> > 

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


#1742795

FromThorsten Leemhuis <regressions@leemhuis.info>
Date2017-10-01 15:20 +0200
Message-ID<uvMk1-2E8-7@gated-at.bofh.it>
In reply to#1742791
On 01.10.2017 15:06, Yanko Kaneti wrote:
> On Sun, 2017-10-01 at 14:46 +0200, Thorsten Leemhuis wrote:
>> Hi, the regression tracker here. What's the status of this issue? Was
>> the problem fixed? It seems nothing happened for more than 10 days -- or
>> did the discussion move somewhere else? Ciao, Thorsten
> The commit was reverted last week before rc2
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=0551968add53777fddd18f4ffb4e3bbc1f646d79

I could have sworn I checked that :-/ Thx for the hint and sorry for the
noise! Ciao, Thorsten

>> On 20.09.2017 02:30, Chuck Ebbert wrote:
>>> 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.
>>>
>>>
>>> http://news.gmane.org/find-root.php?message_id=20170919203044.560cb9f1%40gmail.com 
>>> http://mid.gmane.org/20170919203044.560cb9f1%40gmail.com
>>>
> 

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web