Command "clinfo" causes BUG: kernel NULL pointer dereference, address: 0000000000000008 on driver amdgpu
Mikhail Gavrilov
mikhail.v.gavrilov at gmail.com
Tue Jul 19 11:26:09 UTC 2022
On Tue, Jul 19, 2022 at 1:40 PM Mike Lothian <mike at fireburn.co.uk> wrote:
>
> I was told that this patch replaces the patch you mentioned
> https://patchwork.freedesktop.org/series/106078/ and it the one
> that'll hopefully land in Linus's tree
>
Great, I confirm that both patches solve the issue.
As I understand the second patch [1] is more right and it should be
land merged 5.19 soon, right?
And since we are talking about clinfo, there is a question.
No one has encountered the problem that on configurations with two
GPUs, it hangs in a cycle since it completely occupies one processor
core. In my case, one GPU is in the RENOIR processor, and the other is
a discrete AMD Radeon 6800M. In the BIOS there is no ability to turn
off the integrated GPU in the processor, so there is no way to check
this configuration with each GPU separately. In the kernel log there
is no error so it is most likely a user space issue , but I am not
sure about it.
clinfo backtrace is here [2]
[1] https://patchwork.freedesktop.org/series/106078/
[2] https://pastebin.com/wv5iGibi
--
Best Regards,
Mike Gavrilov.
More information about the amd-gfx
mailing list