summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorKristian Høgsberg <krh@bitplanet.net>2010-08-02 12:45:38 -0400
committerKristian Høgsberg <krh@bitplanet.net>2010-08-02 12:45:38 -0400
commitc37c57aec7128be829fa80640d7236ca0a4b56fc (patch)
tree7a90922ae4eeb8b3b34a175124e8cdf7177ad35a
parent723b2852d22851d0cf1dfe737f0fc375cc692f12 (diff)
TODO: Add a few lines about removing commit request
-rw-r--r--TODO16
1 files changed, 15 insertions, 1 deletions
diff --git a/TODO b/TODO
index 71a3ece..2ddcabf 100644
--- a/TODO
+++ b/TODO
@@ -95,7 +95,21 @@ Core wayland protocol
- xml based description instead?
- - actually make batch/commit batch up commands
+ - Figure out if we need the batch/commit scheme and what to do
+ instead. Since dropping the "copy" request, we have a race between
+ copy from back to front and reporting damage. "copy" did this
+ atomically, but copy is a rendering operation (wayland doesn't do
+ rendering) and requires synchronization between server and client
+ before client can reuse backbuffer.
+
+ The race condition happens when a client copies new content into
+ its window and then, before the client reports the damage, the
+ compositor then does a partial repaint (triggered by another
+ client) that only pulls in part of the repainted area. It's only a
+ one-frame glitch, as the client will submit the damage and the
+ compositor will repaint the damaged area next frame. And ideally
+ clients should do all rendering as early in the frame as possible
+ to avoid this race.
- auth; We need to generate a random socket name and advertise that
on dbus along with a connection cookie. Something like a method