On 13 March 2017 at 00:29, Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl> wrote: > Less IFETCH latency on misses. Shader code is write once read many, > so GTT doesn't make much sense anyway. > > If it turns out to fragment the CPU visible VRAM too much, we can upload with SDMA. Reviewed-by:Dave Airlie <airlied at redhat.com>