[Intel-gfx] Residual RPS tweaks

[Chris Wilson]

When Daniel applied the RPS patches, only then did I realise that I
hadn't send the fresh series that incorporated the bit of feedback and
more importantly a couple of bug fixes!

Hopefully better late than never...

And now with the RPS in, we can get on to lots of fun patches to speed
up execbuffer/requests by a factor of two (in throughput/latency sensitive
tests such as gem_exec_lut_handle -- which directly translates to
improvements in CPU bound synthetics like OglBatch7).
-Chris