[Intel-gfx] [PATCH 8/8] drm/i915: Extract the actual workload submission mechanism from execbuffer

[Chris Wilson]

On Thu, Jun 26, 2014 at 02:24:19PM +0100, oscar.mateo at intel.com wrote:
> +	i915_gem_execbuffer_move_to_active(vmas, ring);
> +	i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);

This is where I start freaking out over the mix of global vs logical
state and the implications of reordering.

The key question for me is how clean busy-ioctl is when it has to pick
up the pieces from a partial failure to submit.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre