[Intel-gfx] Prep work for preemption and GEM bugfixes

Chris Wilson chris at chris-wilson.co.uk
Tue Feb 14 09:53:59 UTC 2017

Previous message: [Intel-gfx] [PATCH] drm/i915: Avoid overflow in computing pot_hole loop termination
Next message: [Intel-gfx] [PATCH v2 01/14] drm/i915: Keep a global seqno per-engine
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Tidy up the global seqno in preparation for requests being cancelled,
and their seqno returned - even if they are currently being waited upon
by third parties. In the process one annoying GEM_DEBUG sporadic failure
is fixed and we are able to take another pass through
i915_wait_request() and make its loops more uniform.
-Chris

Previous message: [Intel-gfx] [PATCH] drm/i915: Avoid overflow in computing pot_hole loop termination
Next message: [Intel-gfx] [PATCH v2 01/14] drm/i915: Keep a global seqno per-engine
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the Intel-gfx mailing list