Mir

[regression] [multimonitor] Clients (sometimes) run at double frame rate if overlapping two monitors.

Bug #1420678 reported by Daniel van Vugt
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Mir
Fix Released
Medium
Daniel van Vugt
mir (Ubuntu)
Fix Released
Medium
Unassigned

Bug Description

[regression] [multimonitor] Clients sometimes run at double frame rate if overlapping two monitors. Whether or not they do seems somewhat unpredictable. Depends how many times they cross monitor boundaries.

Related branches

tags: added: multimonitor
summary: - [regression] [multimonitor] Clients run at double frame rate if
- overlapping two monitors.
+ [regression] [multimonitor] Clients (sometimes) run at double frame rate
+ if overlapping two monitors.
description: updated
Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Even weirder - the clients stay at double framerate any time they come within 50-100 pixels of the shared edge (!?). Not even touching the screen edge.

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

OK, the weirdness is less weird. It's just the edge of the shadow touching the screen edge I think.

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Found the source of the regression:

------------------------------------------------------------
revno: 2183 [merge]
author: Daniel van Vugt <email address hidden>
committer: Tarmac
branch nick: development-branch
timestamp: Fri 2014-12-19 21:52:32 +0000
message:
  Split post_update() into separate gl_swap_buffers() and flip().

  On some platforms this provides a significant performance boost, as
  GL resources and client buffers can be released after the gl_swap_buffers
  without having to wait for the flip.

  This solves most of LP: #1264934, and unblocks work on LP: #1350716 and
  possibly unblocks LP: #1350725 too. Fixes: https://bugs.launchpad.net/bugs/1264934, https://bugs.launchpad.net/bugs/1350716.

  Approved by Alberto Aguirre, Cemil Azizoglu, Alan Griffiths, PS Jenkins bot, Robert Carr.
------------------------------------------------------------

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Oh, I see. The timerless multi-monitor frame sync algorithm (which is awesome) can't coexist with the speedup of r2183 (which is also awesome).

We need a new approach to multi-monitor frame sync if we're going to continue living with high-performance compositing (r2183). Ideally without losing either.

Revision history for this message
PS Jenkins bot (ps-jenkins) wrote :

Fix committed into lp:mir at revision None, scheduled for release in mir, milestone 0.12.0

Changed in mir:
status: In Progress → Fix Committed
Kevin DuBois (kdub)
Changed in mir:
milestone: 0.12.0 → 0.13.0
Changed in mir (Ubuntu):
importance: Undecided → Medium
status: New → Triaged
Changed in mir:
status: Fix Committed → Fix Released
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package mir - 0.13.1+15.10.20150520-0ubuntu1

---------------
mir (0.13.1+15.10.20150520-0ubuntu1) wily; urgency=medium

  [ Cemil Azizoglu ]
  * New upstream release 0.13.1 (https://launchpad.net/mir/+milestone/0.13.1)
    - ABI summary: No ABI break. Servers and clients do not need rebuilding.
      . Mirclient ABI unchanged at 8
      . Mircommon ABI unchanged at 4
      . Mirplatform ABI unchanged at 7
      . Mirserver ABI unchanged at 31
    - Bug fixes:
      . Can't load app purchase UI without a U1 account (LP: #1450377)
      . Crash because uncaught exception in mir::events::add_touch (LP: #1437357)

 -- CI Train Bot <email address hidden> Wed, 20 May 2015 21:20:15 +0000

Changed in mir (Ubuntu):
status: Triaged → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.