[PATCH 0/4] STi device-tree display subsystem rework

[Raphaël Gallais-Pou]

Le 17/07/2025 à 21:15, Raphael Gallais-Pou a écrit :
> This serie aims to rework the display-subsystem node, which was
> previously included directly within the SoC node.  This was wrong
> because it is an abstraction and describes how IPs behave together, not
> what the hardware is.  Instead, extract display-subsystem outside of the
> SoC node, and let IPs describe their connections.  Doing so helps the
> readability, and eases the understanding of the hardware.
> 
> Several nodes have been renamed to stick to the generic names defined in
> the device-tree specification.
> 
> This series depends on another sent a few days ago.  It is not critical
> though, since not having it only triggers warnings when building
> deprecated device-trees.  Please see link below.
> 
> Link: https://lore.kernel.org/lkml/20250714-sti-rework-v2-0-f4274920858b@gmail.com
> 
> Signed-off-by: Raphael Gallais-Pou <rgallaispou at gmail.com>
> ---
> Raphael Gallais-Pou (4):
>        drm/sti: check dma_set_coherent_mask return value
>        drm/sti: make use of drm_of_component_probe
>        ARM: dts: sti: extract display subsystem out of soc
>        ARM: dts: sti: remove useless cells fields
> 
Hi,

@Patrice
Would you agree to make the device-tree patches go through drm-misc 
instead of your tree ? So that the breaking change lands in -next in the 
same time as the driver changes.

I will send another series to convert display subsystem bindings to DT 
schema as soon as the series is merged.

@Alain
Do you prefer to merge it yourself or you rather let me do it ?

Best regards,
Raphaël>   arch/arm/boot/dts/st/stih410.dtsi | 316 
++++++++++++++++++++++----------------
>   drivers/gpu/drm/sti/sti_drv.c     |  18 +--
>   2 files changed, 192 insertions(+), 142 deletions(-)
> ---
> base-commit: b9a572f471993d3e8bf874fcb57f331d66650440
> change-id: 20250401-sti-rework-b009551a362c
> 
> Best regards,