Hi On Wed, Aug 7, 2013 at 2:39 PM, Laurent Pinchart <laurent.pinchart at ideasonboard.com> wrote: > Hi Daniel, > > On Wednesday 07 August 2013 14:19:34 Daniel Vetter wrote: >> On Wed, Aug 7, 2013 at 2:17 PM, Laurent Pinchart wrote: >> > The drm_agp_clear() function is only defined on platforms with AGP >> > support. Move the drm_core_has_AGP() check from drm_agp_clear() to the >> > caller to let the compiler optimize the drm_agp_clear() call away. >> > >> > Signed-off-by: Laurent Pinchart >> > <laurent.pinchart+renesas at ideasonboard.com> >> >> Can't we use the usual approach of an empty static inline helper for >> the !CONFIG_AGP case here? > > Sure, that's possible as well. I find my solution slightly more explicit as it > shows that drm_agp_clear() will only be called for platforms that have AGP, > but I would be fine with a static inline as well.
But the compile fails with -O0.. we don't depend on optimizations to avoid compiler errors. Besides, "static inline" is standard kernel coding-style so I think we should stick to it. I've attached a patch, feel free to pick it up. Otherwise, I will give it some test-compiles once I get home and send it to dri-devel. Thanks David -------------- next part -------------- A non-text attachment was scrubbed... Name: drm_agp_fix.patch Type: application/octet-stream Size: 8616 bytes Desc: not available URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20130807/1f11e9d2/attachment-0001.obj>
