suspend/gpu_lockup_after_suspend: Failed to close window
Bug #1237199 reported by
Yung Shen
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| checkbox (Ubuntu) |
Fix Released
|
High
|
Daniel Manrique | ||
Bug Description
Unable to finish gpu_lockup test, which prevents following tests.
Manually closed glxgears and browser windows will cause testing fail in submission, but next tests will be continue without problems.
From submission log:
WARNING: Got an exception Command 'wmctrl -l | grep glxgears' returned non-zero exit status 1
WARNING: Window 0 not found, not rotating it.
Traceback (most recent call last):
File "/usr/share/
sys.
File "/usr/share/
GlxRotate[
IndexError: list index out of range
Related branches
lp:~roadmr/checkbox/1237199-more-gpu-test
- Zygmunt Krynicki (community): Approve
-
Diff: 61 lines (+20/-11)2 files modifiedcheckbox-old/debian/changelog (+4/-0)
checkbox-old/scripts/gpu_test (+16/-11)
Revision history for this message
| Daniel Manrique (roadmr) wrote | #1 |
| Changed in checkbox (Ubuntu): | |
| assignee: | nobody → Daniel Manrique (roadmr) |
| importance: | Undecided → High |
| status: | New → In Progress |
| Changed in checkbox (Ubuntu): | |
| status: | In Progress → Fix Committed |
| status: | Fix Committed → Fix Released |
To post a comment you must log in.
Rev 2398 introduced a fix for this kind of problem, but what you're seeing here is a case where the fix doesn't work (actually it was wrong from the beginning:
if hasattr(
Say we try to start two windows, 0 and 1. For some reason, we fail to obtain the id for window 0. So:
i=0
GlxWindows[0].id does not exist, we do nothing in the above check.
i=1
GlxWindows[1].id exists, so:
- We append a RotateGlxThread to GlxRotate (which was empty as window 0 wasn't added), pointing to GlxWindows[1] which does exist.
- We try to start rotation on GlxRotate[1] <- But notice, this element does NOT exist because window 0 was never added, so window 1 is actually in slot 0.
I'll get this fixed right away, thanks for reporting this!