- Jun 19, 2012
-
-
Didier Roche authored
-
- Jun 17, 2012
- Jun 14, 2012
-
-
MC Return authored
-
- Jun 11, 2012
-
-
Daniel van Vugt authored
availablePlugins is unused except by dbus, and flawed by designed. It makes no sense to have a function that claims to return the list of available plugins, when that list is not complete. You could easily load other plugins from LD_LIBRARY_PATH that availablePlugins doesn't know about. And you could add or remove plugin binaries at runtime which would also invalidate what availablePlugins knows. ListPlugins was only used by availablePlugins. If you want a list of known plugins, call CompPlugin::getPlugins() instead.
-
- Jun 10, 2012
-
-
Daniel van Vugt authored
-
Daniel van Vugt authored
its only caller (availablePlugins) is now also removed. Re-implement the small part of the dbus plugin that used availablePlugins() to use getPlugins() instead.
-
Daniel van Vugt authored
(LP: #1009338) The fallback is still only used if sync-to-vblank is not available (disabled in the driver) AND if xrandr doesn't provide a refresh rate.
-
Daniel van Vugt authored
the underlying window. (LP: #1009320)
-
- Jun 08, 2012
-
-
Daniel van Vugt authored
the underlying window. (LP: #1009320)
-
Daniel van Vugt authored
availablePlugins is unused except by dbus, and flawed by designed. It makes no sense to have a function that claims to return the list of available plugins, when that list is not complete. You could easily load other plugins from LD_LIBRARY_PATH that availablePlugins doesn't know about. And you could add or remove plugin binaries at runtime which would also invalidate what availablePlugins knows. If you want a list of known plugins, call CompPlugin::getPlugins() instead.
-
- Jun 06, 2012
-
-
Daniel van Vugt authored
Though doing so does have the added benefit of generating a symbol you can see in a debugger.
-
Daniel van Vugt authored
(LP: #1009338) The fallback is still only used if sync-to-vblank is not available (disabled in the driver) AND if xrandr doesn't provide a refresh rate.
-
refactoring was done.
-
(LP: #1006335)
-
-
states (which won't exist if not decorated) (LP: #1007754)
-
- Jun 05, 2012
-
-
Daniel van Vugt authored
damage artifacts.
-
Sam Spilsbury authored
-
Daniel van Vugt authored
-
Daniel van Vugt authored
used). This dramatically reduces CPU usage, reduces wakeups, and increases frame rates. It also solves at least one observed performance bug (LP: #1007299) and probably several more.
-
Daniel van Vugt authored
* Use else instead of unconditional reset (makes the first call faster, but all others slower, negligibly) * Don't mask locals by reusing the same name * Remember to reset arg[7] for XkbStateNotify * Check for NULL strings. Don't dereference them, just treat as empty.
-
Daniel van Vugt authored
-
- Jun 02, 2012
- Jun 01, 2012
-
-
Daniel van Vugt authored
artefact that seems to be specific to xchat + Delta Rectangles.
-
smspillaz authored
-
smspillaz authored
-
Daniel van Vugt authored
-
Daniel van Vugt authored
perform at least as fast as Bounding Box, and both much faster than Raw Rectangles, Delta Rectangles should be fastest in more cases I think.
-
Daniel van Vugt authored
-
Daniel van Vugt authored
-
- May 31, 2012
-
-
Daniel van Vugt authored
-
Daniel van Vugt authored
-
Daniel van Vugt authored
Previously with glxgears (or whatever) rendering 3700 FPS, compiz would get 7400 XDamageNotify's per second. Now it's only around 120 per second (the refresh rate x2).
-
smspillaz authored
buffer them ourselves
-
smspillaz authored
-
Daniel van Vugt authored
Calling CompOption::setName should not implicitly construct a new string object every time, when the name is not changing. This is also an alternative fix to the previous commit, but both together don't hurt either.
-
smspillaz authored
-
- May 30, 2012
-
-
Daniel van Vugt authored
which was wasting lots of CPU (LP: #1005569)
-