[Intel-gfx] [PATCH v2 1/2] drm/i915: add render state initialization

Mika Kuoppala mika.kuoppala at linux.intel.com
Tue May 6 16:34:30 CEST 2014


Chris Wilson <chris at chris-wilson.co.uk> writes:

> On Tue, May 06, 2014 at 04:26:05PM +0300, Mika Kuoppala wrote:
>> HW guys say that it is not a cool idea to let device
>> go into rc6 without proper 3d pipeline state.
>
> * shrug
>
> What's improper 3d state and what prevents userspace from triggering
> badness later?

I would guess improver is 'whats is there after powerup'. But yes,
we dont even know (yet?) what is the proper minimal state :P

What comes to userspace triggering badness later is that the ring
will hang and hangcheck will cleanup the mess.

> The only problem I see in the patch is that you don't move the so->obj
> to the GPU before execution - the code is only coherent thanks to LLC
> atm.

Fixed in v3.

Thanks,
-Mika

> -Chris
>
> -- 
> Chris Wilson, Intel Open Source Technology Centre



More information about the Intel-gfx mailing list