Re: [PATCH 5.18 0309/1095] drm/panel: Fix build error when CONFIG_DRM_PANEL_SAMSUNG_ATNA33XC20=y && CONFIG_DRM_DISPLAY_HELPER=m

[Doug Anderson]

Hi,

On Mon, Aug 15, 2022 at 12:09 PM Greg Kroah-Hartman
<gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
>
> From: Gao Chao <gaochao49@xxxxxxxxxx>
>
> [ Upstream commit a67664860f7833015a683ea295f7c79ac2901332 ]
>
> If CONFIG_DRM_PANEL_SAMSUNG_ATNA33XC20=y && CONFIG_DRM_DISPLAY_HELPER=m,
> bulding fails:
>
> drivers/gpu/drm/panel/panel-samsung-atna33xc20.o: In function `atana33xc20_probe':
> panel-samsung-atna33xc20.c:(.text+0x744): undefined reference to
>  `drm_panel_dp_aux_backlight'
> make: *** [vmlinux] Error 1
>
> Let CONFIG_DRM_PANEL_SAMSUNG_ATNA33XC20 select DRM_DISPLAY_DP_HELPER and
> CONFIG_DRM_DISPLAY_HELPER to fix this error.
>
> Fixes: 32ce3b320343 ("drm/panel: atna33xc20: Introduce the Samsung ATNA33XC20 panel")
> Reported-by: Hulk Robot <hulkci@xxxxxxxxxx>
> Signed-off-by: Gao Chao <gaochao49@xxxxxxxxxx>
> Reviewed-by: Douglas Anderson <dianders@xxxxxxxxxxxx>
> Signed-off-by: Douglas Anderson <dianders@xxxxxxxxxxxx>
> Link: https://patchwork.freedesktop.org/patch/msgid/20220524024551.539-1-gaochao49@xxxxxxxxxx
> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
> ---
>  drivers/gpu/drm/panel/Kconfig | 2 ++
>  1 file changed, 2 insertions(+)

While it doesn't hurt to land this patch as-is on 5.18 and older
kernels, it's not quite right. The symbols that this patch 'select'
don't actually exist on 5.18. ;-) Doing a `git grep` of
`DRM_DISPLAY_DP_HELPER` shows no hits except the one introduced in
this patch...

If you want the equivalent fix for v5.18 and older, I believe you'd want:

select DRM_DP_HELPER
select DRM_KMS_HELPER

See commit 3755d35ee1d2 ("drm/panel: Select DRM_DP_HELPER for
DRM_PANEL_EDP") and commit 3c3384050d68 ("drm: Don't make
DRM_PANEL_BRIDGE dependent on DRM_KMS_HELPERS") which added those for
the (very similar) panel-edp.

The first of those is what got changed in v5.19 in commit 1e0f66420b13
("drm/display: Introduce a DRM display-helper module")

So I guess the tl;dr:

* If you leave this patch in 5.18 (and 5.15), nothing bad will happen
but the broken "randconfig" won't be fixed.

* If you revert this patch in 5.18 (and 5.15) also nothing bad will
happen but also the broken "randconfig" won't be fixed.

* If someone cares about the randconfig on 5.15 / 5.18, we need a
backport that adapts what's selected to the old symbol names.