Am 02.05.23 um 11:12 schrieb Timur
Kristóf:
Hi Christian,
Christian König <christian.koenig@xxxxxxx> ezt írta (időpont: 2023. máj. 2., Ke 9:59):
Am 02.05.23 um 03:26 schrieb André Almeida:
> Em 01/05/2023 16:24, Alex Deucher escreveu:
>> On Mon, May 1, 2023 at 2:58 PM André Almeida <andrealmeid@xxxxxxxxxx>
>> wrote:
>>>
>>> I know that devcoredump is also used for this kind of information,
>>> but I believe
>>> that using an IOCTL is better for interfacing Mesa + Linux rather
>>> than parsing
>>> a file that its contents are subjected to be changed.
>>
>> Can you elaborate a bit on that? Isn't the whole point of devcoredump
>> to store this sort of information?
>>
>
> I think that devcoredump is something that you could use to submit to
> a bug report as it is, and then people can read/parse as they want,
> not as an interface to be read by Mesa... I'm not sure that it's
> something that I would call an API. But I might be wrong, if you know
> something that uses that as an API please share.
>
> Anyway, relying on that for Mesa would mean that we would need to
> ensure stability for the file content and format, making it less
> flexible to modify in the future and probe to bugs, while the IOCTL is
> well defined and extensible. Maybe the dump from Mesa + devcoredump
> could be complementary information to a bug report.
Neither using an IOCTL nor devcoredump is a good approach for this since
the values read from the hw register are completely unreliable. They
could not be available because of GFXOFF or they could be overwritten or
not even updated by the CP in the first place because of a hang etc....
If you want to track progress inside an IB what you do instead is to
insert intermediate fence write commands into the IB. E.g. something
like write value X to location Y when this executes.
This way you can not only track how far the IB processed, but also in
which stages of processing we where when the hang occurred. E.g. End of
Pipe, End of Shaders, specific shader stages etc...
Currently our biggest challenge in the userspace driver is debugging "random" GPU hangs. We have many dozens of bug reports from users which are like: "play the game for X hours and it will eventually hang the GPU". With the currently available tools, it is impossible for us to tackle these issues. André's proposal would be a step in improving this situation.
We already do something like what you suggest, but there are multiple problems with that approach:
1. we can only submit 1 command buffer at a time because we won't know which IB hanged2. we can't use chaining because we don't know where in the IB it hanged3. it needs userspace to insert (a lot of) extra commands such as extra synchronization and memory writes4. It doesn't work when GPU recovery is enabled because the information is already gone when we detect the hang
You can still submit multiple IBs and even chain them. All you need to do is to insert into each IB commands which write to an extra memory location with the IB executed and the position inside the IB.
The write data command allows to write as many dw as you want (up to multiple kb). The only potential problem is when you submit the same IB multiple times.
And yes that is of course quite some extra overhead, but I think that should be manageable.
Consequences:
A. It has a huge perf impact, so we can't enable it alwaysB. Thanks to the extra synchronization, some issues can't be reproduced when this kind of debugging is enabledC. We have to ask users to disable GPU recovery to collect logs for us
In my opinion, the correct solution to those problems would be if the kernel could give userspace the necessary information about a GPU hang before a GPU reset.
The fundamental problem here is that the kernel doesn't have that information either. We know which IB timed out and can potentially do a devcoredump when that happens, but that's it.
The devcoredump can contain register values and memory locations, but since the ASIC is hung reading back stuff like the CP or power management state is dangerous. Alex already noted as well that we potentially need to disable GFXOFF to be able to read back register values.
To avoid the massive peformance cost, it would be best if we could know which IB hung and what were the commands being executed when it hung (perhaps pointers to the VA of the commands), along with which shaders were in flight (perhaps pointers to the VA of the shader binaries).
If such an interface could be created, that would mean we could easily query this information and create useful logs of GPU hangs without much userspace overhead and without requiring the user to disable GPU resets etc.
Disabling GPU reset is actually not something you can reliably do either. If the core memory management runs OOM it can chose to wait for the GPU reset to finish and when that timeout is infinite the system just hard hangs.
That's also the reason why you can't wait for userspace to read back state or something like that.
If it's not possible to do this, we'd appreciate some suggestions on how to properly solve this without the massive performance cost and without requiring the user to disable GPU recovery.
The most doable option I can see is to say instead of resetting the GPU we tell the OS the GPU was hot unplugged and disable system memory access for the ASIC to prevent random memory corruption.
This way you can investigate the GPU state with tools like umr and only after that is done we do a hot add event and start over from scratch.
The game will obviously crash and needs to be restarted, but that is still better than a full system crash.
Side note, it is also extremely difficult to even determine whether the problem is in userspace or the kernel. While kernel developers usually dismiss all GPU hangs as userspace problems, we've seen many issues where the problem was in the kernel (eg. bugs where wrong voltages were set, etc.) - any idea for tackling those kind of issues is also welcome.
No idea either. If I had a better idea how to triage things my live would be much simpler as well :)
Regards,
Christian.
Thanks & best regards,Timur
