Bug 37262 - Extreme IO using libre/openoffice
Summary: Extreme IO using libre/openoffice
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/R100 (show other bugs)
Version: git
Hardware: x86-64 (AMD64) Linux (All)
: high critical
Assignee: Default DRI bug account
QA Contact:
Depends on:
Reported: 2011-05-16 08:53 UTC by dE
Modified: 2019-09-18 18:40 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Note You need to log in before you can comment on or make changes to this bug.
Description dE 2011-05-16 08:53:02 UTC
Since the last few days and in many places, I've been wandering around posting this - 

"I've been using KDE (QT) build of Openoffice on Gentoo and suddenly came across
this strange problem which also persist with GTK openoffice and even

It appears Openoffice takes too much disk I/O when drawing toolbars, it takes 2
minutes to cold start OOo, and in the mean time the whole system is unusable (I
can hardly move the mouse) + the kernel hangs when I doing some other I/O
intensive tasks while OOo is loading, and sometimes it hangs even if I'm
nothing doing anything.

Since Base doesn't have many toolbars by default, it opens OK, but other things
are just horrible. 

Most important fact is that all this happens when composting is enabled with

Also I've seen, this problem persists only if Kwin render method is OpenGL, if
it's XRender, the problem's solved."

As bug reports or in forms.

The problem appears to come from the latest mesa (May 16th 2011), downgrading to 7.10.2 solves the problem.
Comment 1 dE 2011-07-17 20:45:32 UTC
Ok, a new version has been made off this -- 7.11_rc1-r1, which has this problem. Looks like this problem's gonna persist till it's stabilization like all other problems.
Comment 2 Alex Deucher 2011-07-17 21:53:26 UTC
Can you bisect?
Comment 3 dE 2011-07-18 08:27:22 UTC
I hope you mean the commit. As said, the problem persists with mesa greater than 7.10.x.

This problem appears be of low level, I tried a systemtap script to see what was causing the I/O but there was hardly anything suspicious.
Comment 4 Alex Deucher 2011-07-18 08:31:58 UTC
Bisect the mesa git tree between when 7.10 was branched and the earliest master commit you've tested.
Comment 5 dE 2013-01-14 18:22:44 UTC
Looks like no one uses R690 (x1270) anymore. And mine is dead.
Comment 6 GitLab Migration User 2019-09-18 18:40:25 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/280.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.