On Don, 2010-10-28 at 19:59 -0400, Matt Turner wrote: > Can we use gcc's __sync_val_compare_and_swap to implement DRM_CAS? > (If so, do we actually need the __sync_val version, or can we use > __sync_bool?) > > I just threw the patch together in two minutes, so I've got no idea > if it's right, just looking for feedback. The purpose of this is to > remove a lot of inline assembly that hasn't been touched since the > file was added in 2004, including some awesome already-assembled > SPARC instructions. Given that DRM_CAS is part of a kernel<->userspace ABI and is only used for the DRI1 'hardware lock', I suspect this is more risky than it's worth. At the least though, it would need to be verified for each architecture which has a specific DRM_CAS implementation now that this change doesn't result in DRM_CAS always failing and the HW lock falling back to calling the slow path ioctl every time. -- Earthling Michel DÃnzer | http://www.vmware.com Libre software enthusiast | Debian, X and DRI developer _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
