On Fri, Jun 28, 2013 at 7:59 PM, Tomi Valkeinen tomi.valkeinen@iki.fi wrote:
On 15/06/13 15:20, Rob Clark wrote:
On Thu, Jun 13, 2013 at 6:31 AM, Paul Bolle pebolle@tiscali.nl wrote:
On Wed, 2013-03-13 at 20:48 +0100, Paul Bolle wrote:
Signed-off-by: Paul Bolle pebolle@tiscali.nl
Untested. Perhaps the first test that people with access to the relevant hardware might do, is to test _before applying this patch_ with FB_OMAP2 set. Perhaps this negative dependency isn't needed at all. Or is it obvious?
drivers/gpu/drm/omapdrm/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
This patch was sent exactly three months ago, shortly after v3.9-rc2 was released. This obvious typo is still present in v3.10-rc5.
I didn't received any feedback on this patch. Did anyone had a look at it? Is it perhaps queued somewhere?
oh, sorry, at the time I was in the middle of moving, and wasn't watching dri-devel so much.
Yeah, this should probably either be merged, or we should just drop the negative dependency. (I guess in theory you could build both omapfb and omapdrm as modules and just load one.)
Hmm, fixing that creates a problem:
drivers/video/Kconfig:42:error: recursive dependency detected! drivers/video/Kconfig:42: symbol FB is selected by DRM_KMS_HELPER drivers/gpu/drm/Kconfig:29: symbol DRM_KMS_HELPER is selected by DRM_OMAP drivers/gpu/drm/omapdrm/Kconfig:2: symbol DRM_OMAP depends on FB_OMAP2 drivers/video/omap2/omapfb/Kconfig:1: symbol FB_OMAP2 depends on FB
And it makes selecting omapfb not possible...
omapfb already has a !DRM_OMAP dependency. I think it's enough to have that one there, and remove the !FB_OMAP dependency from omapdrm.
I just nuked it upstream in drm-next.
commit 2644ee9614be67abe155f1073bb9e1b737bbca53 Author: Dave Airlie airlied@redhat.com Date: Fri Jun 28 12:08:10 2013 +1000
drm/omap: drop the !FB_OMAP2 dep
This ends up causing circularity and really let people shoot themselves in the foot.
Acked-by: Rob Clark robdclark@gmail.com Signed-off-by: Dave Airlie airlied@redhat.com
Dave.
Tomi