Hi Jan, On Tue, Jan 23, 2018 at 8:56 PM, Jan Kundrát <jan.kundrat@xxxxxxxxx> wrote: > Commit b28b9149b37f added support for using an additional GPIO pin for > Chip Select. According to that commit and to my understanding of a > semi-public datahseet, it seems that the SPI hardware really insists on > driving *some* HW CS signal whenever a SPI transaction is in progress. > > The old code effectively forced the HW CS pin to be CS0. That means that > there could not be anything connected to the real CS0 signal when one > uses a GPIO CS. That assumption does not hold on e.g. Solidrun Clearfog > where some SOM models have a built-in SPI flash on SPI1, CS0. > > Signed-off-by: Jan Kundrát <jan.kundrat@xxxxxxxxx> Thanks for your patch! > --- a/Documentation/devicetree/bindings/spi/spi-orion.txt > +++ b/Documentation/devicetree/bindings/spi/spi-orion.txt > @@ -28,6 +28,11 @@ Optional properties: > - clock-names : names of used clocks, mandatory if the second clock is > used, the name must be "core", and "axi" (the latter > is only for Armada 7K/8K). > +- orion-spi,cs-gpio-hw-cs : Because the SoC always wants to drive some HW Chip > + Select pin, it is necessary to specify which one > + should be used when Linux drives an additional > + GPIO as a software-controlled CS. Defaults to HW > + CS 0. Can't you just deduce an unused HW CS from the cs-gpios property? Renesas MSIOF also requires driving a native chip select, cfr. commit b8761434bdec32fa ("spi: sh-msiof: Implement cs-gpios configuration"). Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds -- To unsubscribe from this list: send the line "unsubscribe linux-spi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html