Re: NetBSD dom0 PVH: hardware interrupts stalls

[Jan Beulich <jbeulich@xxxxxxxx>]

On 24.11.2020 15:27, Manuel Bouyer wrote:
> On Tue, Nov 24, 2020 at 03:09:55PM +0100, Jan Beulich wrote:
>>>> [...]
>>>>> What we're missing is LAPIC information, since the masked status logged
>>>>> is unclear: (-MM) isn't fully matching up with "mask=0". But of course
>>>>> the former is just a software representation, while the latter is what
>>>>> the RTE holds. IOW for the interrupt to not get delivered, there needs
>>>>> to be this or a higher ISR bit set (considering we don't use the TPR),
>>>>> or (I think we can pretty much exclude this) we'd need to be running
>>>>> with IRQs off for extended periods of time.
>>>>
>>>> Let's dump the physical lapic(s) IRR and ISR together with the
>>>> IO-APIC state. Can you please apply the following patch and use the
>>>> 'i' key again? (please keep the previous patch applied)
>>>
>>> Done, you'll find the log at
>>> http://www-soc.lip6.fr/~bouyer/xen-log6.txt
>>
>> Hmm, I can't spot respective output. Are you sure you did this with
>> a hypervisor with Roger's latest patch in place?
> 
> Ops, sorry I copied xen.gz to the wrong place.
> new log at
> http://www-soc.lip6.fr/~bouyer/xen-log7.txt
> 
> this one ends up in a panic,

Argh - too much output triggered the watchdog. I guess we need to
cut down on the vIO-APIC dumping, permaps limiting it to just the
one RTE we care about. But let me (and Roger) see if there's
anything to be derived from the LAPIC state...

Jan