Mir

Code review comment for lp:~vanvugt/mir/unocclude

Revision history for this message
Alberto Aguirre (albaguirre) wrote :

" loosing the logic puts us in a worse position if we're ever told to send 'occluded' messages to the client."

We already have that need. It should go hand-in-hand with the non-blocking swap behavior IMHO.

Clients need to know when their surfaces are no longer visible so they can stop their rendering loop - this is either from having a surface that is no longer visible in any display or from true occlusion.
In the Unity8 case this is currently being handled by the app lifecycle since we don't currently produce those events.

Anpok has a branch in papi that would benefit from the occlusion logic:
https://code.launchpad.net/~andreas-pokorny/platform-api/opaque_by_default/+merge/205259
We should still push through to get that branch merged.

review: Disapprove

« Back to merge proposal