Hi Marek, On 6/1/23 01:10, Marek Vasut wrote: > All boards using the DSI node duplicate the same pattern common pattern > in board DTs, that pattern is ports with endpoint labels and the same > in-SoC regulator connection. Move that common pattern into stm32mp157.dtsi > instead. I think that is a good idea, while it did not crossed my mind implementing it this way on the first time. I'll let my peers Philippe and Yannick review and ack it this time. I tested it on DK2, it seems fine. Thanks :) > The two boards which do define panel@0 directly in the DSI bridge node now > have #address-cells/#size-cells in their board DT instead of it being in > stm32mp157.dtsi and activated incorrectly for all boards, even the ones > which use e.g. another DSI-to-something bridge. > > Signed-off-by: Marek Vasut <marex@xxxxxxx> Reviewed-by: Raphael Gallais-Pou <raphael.gallais-pou@xxxxxxxxxxx> > --- > Cc: Alexandre Torgue <alexandre.torgue@xxxxxxxxxxx> > Cc: Conor Dooley <conor+dt@xxxxxxxxxx> > Cc: Krzysztof Kozlowski <krzysztof.kozlowski+dt@xxxxxxxxxx> > Cc: Maxime Coquelin <mcoquelin.stm32@xxxxxxxxx> > Cc: Raphael Gallais-Pou <raphael.gallais-pou@xxxxxxxxxxx> > Cc: Rob Herring <robh+dt@xxxxxxxxxx> > Cc: devicetree@xxxxxxxxxxxxxxx > Cc: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx > Cc: linux-stm32@xxxxxxxxxxxxxxxxxxxxxxxxxxxx > --- > arch/arm/boot/dts/stm32mp157.dtsi | 18 ++++++++++++ > ...tm32mp157a-icore-stm32mp1-ctouch2-of10.dts | 24 ++++------------ > .../stm32mp157a-icore-stm32mp1-edimm2.2.dts | 28 ++++++------------- > arch/arm/boot/dts/stm32mp157c-dk2.dts | 28 ++++++------------- > arch/arm/boot/dts/stm32mp157c-ev1.dts | 28 ++++++------------- > 5 files changed, 48 insertions(+), 78 deletions(-)