Hello Xi,
On 7/4/22 12:29, Xi Ruoyao wrote:
> On Mon, 2022-07-04 at 17:36 +0800, Xi Ruoyao wrote:
>
>>> Yes, please do. Either with CONFIG_SYSFB_SIMPLEFB disabled and CONFIG_FB_EFI
>>> enabled (so that "efi-framebuffer" is registered and efifb probed) or with
>>> CONFIG_SYSFB_SIMPLEFB but CONFIG_FB_SIMPLE enabled (so "simple-framebuffer
>>> is used too but with simplefb instead of simpledrm).
>>>
>>> I'm not able to reproduce, it would be useful to have another data point.
>>
>> Also happening for me with CONFIG_SYSFB_SIMPLEFB, on a Intel Core i7-
>> 1065G7 (with iGPU).
>>
>> Reverting this commit on top of 5.19-rc5 "fixes" the issue.
>
> With CONFIG_SYSFB_SIMPLEFB and CONFIG_FB_SIMPLE enabled, there is no
> issue.
>
> I guess it's something going wrong on a "drm -> drm" pass over. For now
> I'll continue to use simpledrm with this commit reverted.
>
Yes, we need to also cherry-pick b84efa28a48 ("drm/aperture: Run fbdev
removal before internal helpers") now that the sysfb_disable() patches
are in v5.19-rc5.
--
Best regards,
Javier Martinez Canillas
Linux Engineering
Red Hat
