DRM_PANEL_LD9040 and DRM_PANEL_S6E8AA0 both explicitly depended on DRM_PANEL && DRM. This is nonsense for two reasons:
(a) DRM_PANEL already depends on DRM, so DRM_PANEL can not be enabled without DRM first being enabled. Hence the && DRM is useless.
(b) These two configs are already beneath a menu which depends on DRM_PANEL being enabled; these options will not be visible without DRM_PANEL already being enabled.
Let's just remove this unnecessary verbosity.
Signed-off-by: Russell King rmk+kernel@arm.linux.org.uk --- drivers/gpu/drm/panel/Kconfig | 2 -- 1 file changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig index 4ec874da5668..44aea730b3aa 100644 --- a/drivers/gpu/drm/panel/Kconfig +++ b/drivers/gpu/drm/panel/Kconfig @@ -18,14 +18,12 @@ config DRM_PANEL_SIMPLE
config DRM_PANEL_LD9040 tristate "LD9040 RGB/SPI panel" - depends on DRM && DRM_PANEL depends on OF select SPI select VIDEOMODE_HELPERS
config DRM_PANEL_S6E8AA0 tristate "S6E8AA0 DSI video mode panel" - depends on DRM && DRM_PANEL depends on OF select DRM_MIPI_DSI select VIDEOMODE_HELPERS
On Thursday 12 June 2014 16:50:15 Russell King wrote:
DRM_PANEL_LD9040 and DRM_PANEL_S6E8AA0 both explicitly depended on DRM_PANEL && DRM. This is nonsense for two reasons:
(a) DRM_PANEL already depends on DRM, so DRM_PANEL can not be enabled without DRM first being enabled. Hence the && DRM is useless.
(b) These two configs are already beneath a menu which depends on DRM_PANEL being enabled; these options will not be visible without DRM_PANEL already being enabled.
Let's just remove this unnecessary verbosity.
Signed-off-by: Russell King rmk+kernel@arm.linux.org.uk
Unfortunately, this is wrong, because the current expression exploits one of the finer details of Kconfig syntax:
drivers/gpu/drm/panel/Kconfig | 2 -- 1 file changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig index 4ec874da5668..44aea730b3aa 100644 --- a/drivers/gpu/drm/panel/Kconfig +++ b/drivers/gpu/drm/panel/Kconfig @@ -18,14 +18,12 @@ config DRM_PANEL_SIMPLE
config DRM_PANEL_LD9040 tristate "LD9040 RGB/SPI panel"
depends on DRM && DRM_PANEL depends on OF select SPI select VIDEOMODE_HELPERS
config DRM_PANEL_S6E8AA0 tristate "S6E8AA0 DSI video mode panel"
depends on DRM && DRM_PANEL depends on OF select DRM_MIPI_DSI select VIDEOMODE_HELPERS
DRM_PANEL is bool, while DRM is tristate. If DRM=m and DRM_PANEL=y, these two options can only be 'n' or 'm' but not 'y' with the current code, but they can be 'y' if you drop the explicit DRM dependency, which will cause a link failure.
The DRM_PANEL dependency can be removed, it is indeed unnecessary.
Arnd
On Thu, Jun 12, 2014 at 06:00:07PM +0200, Arnd Bergmann wrote:
On Thursday 12 June 2014 16:50:15 Russell King wrote:
DRM_PANEL_LD9040 and DRM_PANEL_S6E8AA0 both explicitly depended on DRM_PANEL && DRM. This is nonsense for two reasons:
(a) DRM_PANEL already depends on DRM, so DRM_PANEL can not be enabled without DRM first being enabled. Hence the && DRM is useless.
(b) These two configs are already beneath a menu which depends on DRM_PANEL being enabled; these options will not be visible without DRM_PANEL already being enabled.
Let's just remove this unnecessary verbosity.
Signed-off-by: Russell King rmk+kernel@arm.linux.org.uk
Unfortunately, this is wrong, because the current expression exploits one of the finer details of Kconfig syntax:
drivers/gpu/drm/panel/Kconfig | 2 -- 1 file changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig index 4ec874da5668..44aea730b3aa 100644 --- a/drivers/gpu/drm/panel/Kconfig +++ b/drivers/gpu/drm/panel/Kconfig @@ -18,14 +18,12 @@ config DRM_PANEL_SIMPLE
config DRM_PANEL_LD9040 tristate "LD9040 RGB/SPI panel"
depends on DRM && DRM_PANEL depends on OF select SPI select VIDEOMODE_HELPERS
config DRM_PANEL_S6E8AA0 tristate "S6E8AA0 DSI video mode panel"
depends on DRM && DRM_PANEL depends on OF select DRM_MIPI_DSI select VIDEOMODE_HELPERS
DRM_PANEL is bool, while DRM is tristate. If DRM=m and DRM_PANEL=y, these two options can only be 'n' or 'm' but not 'y' with the current code, but they can be 'y' if you drop the explicit DRM dependency, which will cause a link failure.
The DRM_PANEL dependency can be removed, it is indeed unnecessary.
In which case, that's a bug for DRM_PANEL_SIMPLE which uses the same mechanism. Since everything is going to be in the same boat, the menu should depend on DRM && DRM_PANEL.
On Thursday 12 June 2014 17:04:34 Russell King - ARM Linux wrote:
In which case, that's a bug for DRM_PANEL_SIMPLE which uses the same mechanism. Since everything is going to be in the same boat, the menu should depend on DRM && DRM_PANEL.
Right, I actually have a patch for DRM_PANEL_SIMPLE in my randconfig series that I haven't sent out. Making the menu depend on DRM is probably correct, but I haven't tried.
Arnd
dri-devel@lists.freedesktop.org