~raof/mir/threadsafe-death-tests

Viewing all changes in revision 2673.

  • Committer: Tarmac
  • Author(s): Daniel van Vugt
  • Date: 2015-06-17 05:20:42 UTC
  • mfrom: (2094.7.67 ddouble)
  • Revision ID: tarmac-20150617052042-9d2ihmqx6i9hu9hd
Introducing dynamic buffer queue scaling to minimize lag.
(LP: #1240909)

Actually the size of the queue doesn't really change with this branch.
What it does do is reliably detect when a client is staying ahead of
the compositor and throttle it so that no more than one future frame
is ever pre-rendered. Hence you get the same single-frame latency as
double buffering. But the moment the client fails to keep up, it will
get to use all three buffers again. Fixes: https://bugs.launchpad.net/bugs/1240909, https://bugs.launchpad.net/bugs/1274408.

Approved by Cemil Azizoglu, Alexandros Frantzis, PS Jenkins bot.

expand all expand all

Show diffs side-by-side

added added

removed removed

Lines of Context: