LIMA kernel bug - help!

LIMA kernel bug - help!

[Linus Walleij]

Hi Qiang,

I am using Lima on the ST-Ericsson U8500 SoC. It is one
of the very earliest versions of MALI 400 MP. It mostly works
on the mobile phones I have using PostmarkeOS and
the Phosh UI (Wayland with MESA), but now I have an issue with
one single specimen.

I have this bug in my dmesg and after this the graphics
start to act weird:

[   48.662336] lima a0300000.gpu: mmu page fault at 0x2020400 from bus
id 0 of type read on ppmmu0
[   48.662464] lima a0300000.gpu: fail to save task state from phoc
pid 1581: error task list is full
[   48.662481] lima a0300000.gpu: pp task error 0 int_state=0 status=1
[   48.662494] lima a0300000.gpu: mmu resume

In interrupts:

 75:          5          0     GIC-0 147 Level     gp
 76:          0          0     GIC-0 146 Level     gpmmu
 77:          3          0     GIC-0 145 Level     pp0
 78:          1          0     GIC-0 144 Level     ppmmu0

My UI definitely want to render some more stuff so it gets
stuck here.

Any hints on how to troubleshoot this?

Yours,
Linus Walleij

Re: [Lima] LIMA kernel bug - help!

[Vasily Khoruzhick]

On Thu, Jun 3, 2021 at 11:51 PM Linus Walleij <linus.walleij@linaro.org> wrote:
>
> Hi Qiang,

Hey Linus,

> I am using Lima on the ST-Ericsson U8500 SoC. It is one
> of the very earliest versions of MALI 400 MP. It mostly works
> on the mobile phones I have using PostmarkeOS and
> the Phosh UI (Wayland with MESA), but now I have an issue with
> one single specimen.
>
> I have this bug in my dmesg and after this the graphics
> start to act weird:
>
> [   48.662336] lima a0300000.gpu: mmu page fault at 0x2020400 from bus
> id 0 of type read on ppmmu0
> [   48.662464] lima a0300000.gpu: fail to save task state from phoc
> pid 1581: error task list is full
> [   48.662481] lima a0300000.gpu: pp task error 0 int_state=0 status=1
> [   48.662494] lima a0300000.gpu: mmu resume

It's not a kernel bug, GPU MMU just catched invalid memory access by
fragment shader. It's likely a mesa bug.

What mesa version do you use?

> In interrupts:
>
>  75:          5          0     GIC-0 147 Level     gp
>  76:          0          0     GIC-0 146 Level     gpmmu
>  77:          3          0     GIC-0 145 Level     pp0
>  78:          1          0     GIC-0 144 Level     ppmmu0
>
> My UI definitely want to render some more stuff so it gets
> stuck here.

It's stuck because context is marked as tainted and all the jobs that
are submitted from this context afterwards are essentially no-op.

> Any hints on how to troubleshoot this?

Please make sure that you still can reproduce the bug with mesa from
git main (ex-master), if it's still reproducible please open an issue
at https://gitlab.freedesktop.org/mesa/mesa/issues
Please also try to capture an apitrace of the failing app.

See "Bug Reporting" on [1].

[1] https://docs.mesa3d.org/drivers/lima.html

> Yours,
> Linus Walleij
> _______________________________________________
> lima mailing list
> lima@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/lima