Here's my rebase of Ken's shader_time updates for Haswell, attempting to incorporate review feedback plus my discomfort with the discussion of addressing changes (it was superfluous -- the messages we're sending are the untyped byte offsets same both ways, and the RAW vs RGBA32F in the surface state doesn't matter). I hope this accurately reflects review, so I'd appreciate an ack from ken/paul if they agree. While doing so, I also made the simple change I'd been thinking about to reduce the overhead of shader_time.
Available at the shader_time branch of my tree. _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
