On 12/01/2016 00:20, Chris Wilson wrote:
On Mon, Jan 11, 2016 at 06:42:31PM +0000, John.C.Harrison@xxxxxxxxx wrote:
From: John Harrison <John.C.Harrison@xxxxxxxxx>
A later patch in this series re-organises the batch buffer submission
code. Part of that is to reduce the scope of a pm_get/put pair.
Specifically, they previously wrapped the entire submission path from
the very start to the very end, now they only wrap the actual hardware
submission part in the back half.
However, as you haven't fixed the ordering issue that requires rpm_get
before struct_mutex, this is broken.
Why does 'intel_runtime_pm_get' require the struct mutex to be held? It
has certainly not complained at me about trying to do stuff without it.
When we have rpm fixed, you don't
need this patch as we can take the wakeref around the GSM access itself.
We still want the prolonged rpm wakeref for the GPU activity, ofc.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
