On Mon, May 14, 2018 at 1:20 AM, Jagan Teki <jagan@xxxxxxxxxxxxxxxxxxxx> wrote: > On Tue, May 1, 2018 at 9:53 PM, Chen-Yu Tsai <wens@xxxxxxxx> wrote: >> On Mon, Apr 30, 2018 at 7:40 PM, Jagan Teki <jagan@xxxxxxxxxxxxxxxxxxxx> wrote: >>> Allwinner 64-bit SoC like H5/A64 has DE2 CCU so enable them >>> as default. >>> >>> Signed-off-by: Jagan Teki <jagan@xxxxxxxxxxxxxxxxxxxx> >>> --- >>> drivers/clk/sunxi-ng/Kconfig | 2 ++ >>> 1 file changed, 2 insertions(+) >>> >>> diff --git a/drivers/clk/sunxi-ng/Kconfig b/drivers/clk/sunxi-ng/Kconfig >>> index 79dfd296c3d1..1fffd3bf6ff3 100644 >>> --- a/drivers/clk/sunxi-ng/Kconfig >>> +++ b/drivers/clk/sunxi-ng/Kconfig >>> @@ -58,6 +58,8 @@ config SUN8I_V3S_CCU >>> >>> config SUN8I_DE2_CCU >>> bool "Support for the Allwinner SoCs DE2 CCU" >>> + default ARM64 && ARCH_SUNXI >>> + depends on (DRM_SUN4I && (ARM64 && ARCH_SUNXI)) || COMPILE_TEST >> >> There is no reason to depend on DRM_SUN4I. There is no compile dependency. > > Since this CCU is for DE2 I've make DRM_SUN4I for that matter, any problem? Users may very well be just using simplefb, but with all the clock drivers enabled. That is a valid, if very limited, use case. So again, you should not limit this driver based on DRM_SUN4I. There is no compile dependency. > >> >> Also, this is needed on SUN8I as well, pretty much anything with DE 2.0. >> So you shouldn't limit it to ARM64. That pretty much breaks things for >> people with A83T's or H3's. In fact you should enable it by default for >> these as well. > > True, I've skipped SUN8I since this series for A64. And yet you are breaking existing users, which is a big no-no. ChenYu _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel
