On Mon, Apr 25, 2016 at 04:21:58PM +0100, Dave Gordon wrote:
> On 24/04/16 19:10, Chris Wilson wrote:
> >Rather than reuse the current location of the context in the global GTT
> >for its hardware identifier, use the context's unique ID assigned to it
> >for its whole lifetime.
> >
> >Signed-off-by: Chris Wilson <ch...@chris-wilson.co.uk>
> >Reviewed-by: Tvrtko Ursulin <tvrtko.ursu...@intel.com>
> 
> Hmmm I wonder whether this will work with the GuC?
> IIRC the GuC firmware requires the LRCA to be used as the CTX ID :(
> Or at least it used to, maybe that has changed now.

I know you raised it last time, as far as I can tell the two are
distinct in i915_guc_submission and you never pass the high 32 bits of
the LRC context descriptor (the context id) to the GuC. i.e.

guc_init_ctx_desc:
        ctx_desc = intel_lr_context_descriptor(ctx, engine);
        lrc->context_desc = (u32)ctx_desc;

        lrc->ring_lrca = i915_gem_object_gtt_offset(ctx_obj) + PAGE;
        lrc->context_id = client->ctx_index | engine->guc_id;

The guc hasn't fared any worse then it did before applying the change...
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Reply via email to