On 09/12/2022 18:53, Doug Anderson wrote: > Hi, > > On Fri, Dec 9, 2022 at 2:25 AM Krzysztof Kozlowski > <krzysztof.kozlowski@xxxxxxxxxx> wrote: >> >> On 02/12/2022 17:53, Doug Anderson wrote: >>> Hi, >>> >>> On Fri, Dec 2, 2022 at 7:57 AM Krzysztof Kozlowski >>> <krzysztof.kozlowski@xxxxxxxxxx> wrote: >>>> >>>> DT schema expects TLMM pin configuration nodes to be named with >>>> '-state' suffix and their optional children with '-pins' suffix. >>>> >>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx> >>>> >>>> --- >>>> >>>> Cc: Doug Anderson <dianders@xxxxxxxxxxxx> >>>> >>>> Tested on Qualcomm RB3. Please kndly test a bit more on other devices. >>>> This should not have an functional impact. >>>> >>>> Changes since v2: >>>> 1. Bring back UART6 4-pin bias/drive strength to DTSI. >>> >>> Just to be clear, it doesn't actually belong in the DTSI, but it was >>> there before your patch and it's fine if your patch series doesn't fix >>> the whole world. I'm OK with this one staying in the DTSI for now just >>> to keep things simpler. >>> >>> One change missing in v3 that I would have expected based on our >>> discussion in the previous version would be to "Add UART3 4-pin mux >>> settings for use in db845c." I think you said you would do this, but I >>> don't see it done. >> >> Hm, I don't recall that. Changing db845c to usage of RTS/CTS is >> independent problem, not related to fixes or aligning with DT schema. > > It was in the message: > > https://lore.kerne.org/r/68bcdf25-e8e3-f817-f213-efb0bce3f43a@xxxxxxxxxx > > I said: > >> FWIW, I would have expected that the SoC dtsi file would get a "4-pin" >> definition (similar to what you did with qup_uart6_4pin) and then we'd >> use that here. > > You said: > >> Sure. > Yes, indeed and it see now I still keep there cts-rts-pins. Somehow I thought now we talk about UART6 or UART9... However, the UART3 is disabled, so it will not have any effect, except for some downstream users. Best regards, Krzysztof
