On 22.11.2023 10:29, Krzysztof Kozlowski wrote:
> On 22/11/2023 10:06, AngeloGioacchino Del Regno wrote:
>>>>> Hey Krzysztof,
>>>>>
>>>>> This is interesting. It might be about the cores that are missing from the partial
>>>>> core_mask raising interrupts, but an external abort on non-linefetch is strange to
>>>>> see here.
>>>> I've seen such external aborts in the past, and the fault type has
>>>> often been misleading. It's unlikely to have anything to do with a
>>> Yeah, often accessing device with power or clocks gated.
>>>
>> Except my commit does *not* gate SoC power, nor SoC clocks 🙂
> It could be that something (like clocks or power supplies) was missing
> on this board/SoC, which was not critical till your patch came.
>
>> What the "Really power off ..." commit does is to ask the GPU to internally power
>> off the shaders, tilers and L2, that's why I say that it is strange to see that
>> kind of abort.
>>
>> The GPU_INT_CLEAR GPU_INT_STAT, GPU_FAULT_STATUS and GPU_FAULT_ADDRESS_{HI/LO}
>> registers should still be accessible even with shaders, tilers and cache OFF.
>>
>> Anyway, yes, synchronizing IRQs before calling the poweroff sequence would also
>> work, but that'd add up quite a bit of latency on the runtime_suspend() call, so
>> in this case I'd be more for avoiding to execute any register r/w in the handler
>> by either checking if the GPU is supposed to be OFF, or clearing interrupts, which
>> may not work if those are generated after the execution of the poweroff function.
>> Or we could simply disable the irq after power_off, but that'd be hacky (as well).
>>
>>
>> Let's see if asking to poweroff *everything* works:
> Worked.
Yes, I also got into this issue some time ago, but I didn't report it
because I also had some power supply related problems on my test farm
and everything was a bit unstable. I wasn't 100% sure that the $subject
patch is responsible for the observed issues. Now, after fixing power
supply, I confirm that the issue was revealed by the $subject patch and
above mentioned change fixes the problem. Feel free to add:
Tested-by: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx>
Best regards
--
Marek Szyprowski, PhD
Samsung R&D Institute Poland
