[Bug 689741] [NEW] libgtk2.0-0 performance problem due to unused cruft from 10 years ago

2010-12-13 Thread eXtace
Public bug reported: Binary package hint: libgtk2.0-0 After discussions with Owen Taylor it was found that GTK+ has a longstanding performance problem in cases where a program has a large number of GTK+ widgets on a window. https://bugzilla.gnome.org/show_bug.cgi?id=637155 https://bugzilla.gnom

[Bug 689741] Re: libgtk2.0-0 performance problem due to unused cruft from 10 years ago

2010-12-13 Thread eXtace
** Patch added: "diff" https://bugs.launchpad.net/bugs/689741/+attachment/1764898/+files/diff -- You received this bug notification because you are a member of Ubuntu Desktop Bugs, which is subscribed to gtk+2.0 in ubuntu. https://bugs.launchpad.net/bugs/689741 Title: libgtk2.0-0 performan

[Bug 689741] Re: libgtk2.0-0 performance problem due to unused cruft from 10 years ago

2010-12-13 Thread eXtace
Removing a call to a cpu eating yet UNNEEDED function isn't trivial? I don't understand your reasoning. Removing 10 year old UNUSED cruft that results in a significant speedup to Applications with large(r) numbers of widgets, seems like a win, esp since it appears to be unused and unneeded.

[Bug 689741] Re: libgtk2.0-0 performance problem due to unused cruft from 10 years ago

2010-12-14 Thread eXtace
The reason why I posted to ubuntu/launchpad, is that I don't expect the upstream to really care about this enough to release a new version in a timely manner, as its not something that directly causes crashes or a security leak, it just causes a big performance drop for some applications. I would

Re: [Bug 689741] Re: libgtk2.0-0 performance problem due to unused cruft from 10 years ago

2010-12-15 Thread eXtace
I'm sorry, I don't have a natty system to test against, however you can do some testing with owen's python testcase app, which I've attached. Just change the SIZE = XX line and test, i.e. setting it to 12, creates a 12x12 grid, changing that to 48, creates a 48x48 grid and so on. You sho