Hi Maxime,
On Wed, 27 Mar 2024, Maxime Ripard wrote:
Jani recently pointed out that the Kconfig symbols are a bit difficult
to work with at the moment when they depend on each other, and that
using depends on would be a better idea, but no one really did the work
so far.
So here it goes :)
It's been tested by comparing the riscv defconfig, arm
multi_v7_defconfig, arm64 defconfig, drm-misc-arm, drm-misc-arm64 and
drm-misc-x86 before and after this series and making sure they are
identical.
That is not true: comparing drm-misc/for-linux-next to v6.9-rc2,
arm/multi_v7_defconfig, arm64/defconfig, and riscv/defconfig lost
several of:
- CONFIG_DRM_DW_HDMI,
- CONFIG_DRM_DW_HDMI_AHB_AUDIO,
- CONFIG_DRM_DW_HDMI_CEC,
- CONFIG_DRM_DW_HDMI_I2S_AUDIO,
- CONFIG_DRM_IMX_HDMI.
- CONFIG_DRM_MESON_DW_HDMI,
- CONFIG_DRM_RCAR_DW_HDMI,
- CONFIG_DRM_SUN8I_DW_HDMI,
- CONFIG_ROCKCHIP_DW_HDMI,
- CONFIG_SND_MESON_G12A_TOHDMITX,
Let me know what you think,
IMHO this series looks like a big usuability issue for anyone
configuring the kernel, and you try to work around this by sprinkling
"default y" around.
The user should not need to know which helpers are needed for the driver
he is interested in. When a symbol selects another symbol, it should
just make sure the dependencies of the target symbol are met.
Thanks for reverting ;-)
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
