On Tue, Apr 12, 2016 at 09:52:28AM +0100, Tvrtko Ursulin wrote:
> From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
>
> We can use the new pin/lazy unpin API for simplicity
> and more performance in the execlist submission paths.
>
> v2:
> * Fix error handling and convert more users.
> * Compact some names for readability.
>
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
> Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Hmm, the stress tests that would exercise this are already currently
fail (thinking of gem_ctx_create etc). But this should not excerbate it
much!
> - lrc_state_page = i915_gem_object_get_dirty_page(ctx_obj, LRC_STATE_PN);
> - if (WARN_ON(!lrc_state_page)) {
> - ret = -ENODEV;
> + obj_addr = i915_gem_object_pin_map(ctx_obj);
Oops, there's a bug in pin_map in that we don't mark the object as
dirty. Can you send a quick obj->dirty = true patch?
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
