On Fri, 2009-10-02 at 18:38 +0200, Kjartan Maraas wrote:
- Suspend / resume stopped working for me recently. Works from runlevel
3, but not from within Xorg https://bugzilla.redhat.com/show_bug.cgi?id=526757
I wouldn't really consider that serious enough. suspend is always pretty flaky and we've never held releases for suspend problems until now.
Suspend + resume has worked flawlessly on this laptop for a long time with the occacional glitch in rawhide of course. The machine is a HP Compaq nc6400 and was quite popular as a "business" laptop over here a couple of years ago at least. I've got another one just like it so I can test whether this is bad hardware at least :-)
Sure, understood, but unfortunately I'm not sure we can practically consider even regressions in suspend support to be blockers :/ just my opinion though, you could certainly stick it on the list to be reviewed at a meeting.
- Evolution has a tendency to hang after a suspend / resume cycle and
take the whole machine down with it when it does so https://bugzilla.redhat.com/show_bug.cgi?id=492686
see above, but hanging the whole system is icky. If this isn't limited to your hardware it's worth considering.
See the thread I pointed to in the last comment. This looked like it's a general kernel problem and is quite similar to one of the top reported oops by kerneloops:
http://kerneloops.org/searchweek.php?search=warn_slowpath_null
Right. I'd probably consider that a final release blocker, then.
- Machine freezes from time to time. Possibly related to intel graphics
xorg driver again. Filed upstream bug here: https://bugs.freedesktop.org/show_bug.cgi?id=23969
ditto with #2, depends how hardware-dependent it is.
Still trying to get more information out of this bug. Any pointers to things I can do to get more output from xorg, the intel driver or the kernel here?
You could perhaps also attach Xorg.0.log after a hang? it may have something useful at the end. you'd have to boot to runlevel 3 to take a copy of it, of course.