On Thu, May 11, 2017 at 05:18:00PM +0200, Geert Uytterhoeven wrote: > Hi Simon, > > On Thu, May 11, 2017 at 4:53 PM, Simon Horman <horms@xxxxxxxxxxxx> wrote: > > On Thu, May 11, 2017 at 01:58:53PM +0200, Geert Uytterhoeven wrote: > >> Split off support for Salvator-X boards with the ES1.x revision of the > >> R-Car H3 SoC into a separate file. The main r8a7795-salvator-x.dts file > >> now corresponds to Salvator-X with R-Car H3 ES2.0 or later. > >> > >> Signed-off-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx> > > >> --- a/arch/arm64/boot/dts/renesas/Makefile > >> +++ b/arch/arm64/boot/dts/renesas/Makefile > >> @@ -1,4 +1,5 @@ > >> dtb-$(CONFIG_ARCH_R8A7795) += r8a7795-salvator-x.dtb r8a7795-h3ulcb.dtb > >> +dtb-$(CONFIG_ARCH_R8A7795) += r8a7795-es1-salvator-x.dtb > > > > Probably this was already discussed but I wonder if there is any value > > in a CONFIG_ARCH_R8A7795_ES1 Kconfig symbol to allow more fine grained > > control over what is compiled into the kernel. > > The plan was always to have a single binary supporting both. > So I think it makes sense to have a single DT build (as in both DTBs are > built), too. > > Note that if we add and use CONFIG_ARCH_R8A7795_ES1 here, the ES1 DTB will > no longer be always created, making it even harder for people who want to > support multiple kernel versions. > > E.g. to pick the correct DTB for Salvator-X with H3 ES1.0, I have the > following in my install script: > > DTB=arch/arm64/boot/dts/renesas/r8a7795-es1-salvator-x.dtb > if [ ! -e $DTB ]; then > DTB=arch/arm64/boot/dts/renesas/r8a7795-salvator-x.dtb > fi > > If we change our mind, it means we can/should sprinkle some > "#ifdef CONFIG_ARCH_R8A7795_ES1" checks in the clk/pfc/sysc driver code? Maybe it would. Thanks for your explanation above. I think it reasonable to leave things as-is and not introduce CONFIG_ARCH_R8A7795_ES1 for now.
