On Fri, Jan 22, 2016 at 04:33:18PM +, Tvrtko Ursulin wrote:
> I was confused by the current code which does the reset after unpinning:
>
> ...
> i915_gem_reset(dev);
>
> simulated = dev_priv->gpu_error.stop_rings != 0;
>
> ret = intel_gpu_reset(dev);
> ...
>
>
> What is r
On 21/01/16 11:02, Chris Wilson wrote:
On Thu, Jan 21, 2016 at 10:10:42AM +, Tvrtko Ursulin wrote:
From: Tvrtko Ursulin
In GuC mode LRC pinning lifetime depends exclusively on the
request liftime. Since that is terminated by the seqno update
that opens up a race condition between GPU fini
On Thu, Jan 21, 2016 at 11:02:50AM +, Chris Wilson wrote:
> intel_gpu_reset();
> for_each_engine() unpin(engine->last_context);
> unpin(dev_priv->default_context);
>
> After a few more passes, we should be able to call the reset and cleanup
> functions higher up (as part of the stopping the GP
On Thu, Jan 21, 2016 at 10:10:42AM +, Tvrtko Ursulin wrote:
> From: Tvrtko Ursulin
>
> In GuC mode LRC pinning lifetime depends exclusively on the
> request liftime. Since that is terminated by the seqno update
> that opens up a race condition between GPU finishing writing
> out the context i
From: Tvrtko Ursulin
In GuC mode LRC pinning lifetime depends exclusively on the
request liftime. Since that is terminated by the seqno update
that opens up a race condition between GPU finishing writing
out the context image and the driver unpinning the LRC.
To extend the LRC lifetime we will e
