Re: smp_call_function_single lockups

From: Chris J Arges
Date: Wed Apr 01 2015 - 17:59:48 EST




On 04/01/2015 11:14 AM, Linus Torvalds wrote:
> On Wed, Apr 1, 2015 at 9:10 AM, Chris J Arges
> <chris.j.arges@xxxxxxxxxxxxx> wrote:
>>
>> Even with irqbalance removed from the L0/L1 machines the hang still occurs.
>>
>> This results in no 'apic: vector* or 'ack_APIC*' messages being displayed.
>
> Ok. So the ack_APIC debug patch found *something*, but it seems to be
> unrelated to the hang.
>
> Dang. Oh well. Back to square one.
>
> Linus
>

With my L0 testing I've normally used a 3.13 series kernel since it
tends to reproduce the hang very quickly with the testcase. Note we have
reproduced an identical hang with newer kernels (3.19+patches) using the
openstack tempest on openstack reproducer, but the timing can vary
between hours and days.

Installing a v4.0-rc6+patch kernel on L0 makes the problem very slow to
reproduce, so I am running these tests now which may take day(s).

It is worthwhile to do a 'bisect' to see where on average it takes
longer to reproduce? Perhaps it will point to a relevant change, or it
may be completely useless.

--chris

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/