Messages in this thread | | | Date | Wed, 14 May 2014 17:57:59 +0200 | From | Pavel Machek <> | Subject | Re: 3.15-rc: regression in suspend |
| |
On Wed 2014-05-14 14:39:19, Jiri Kosina wrote: > On Tue, 13 May 2014, Pavel Machek wrote: > > > > > There seems to be regression in 3.15 w.r.t. suspend to ram and suspend > > > > to disk. > > > > > > > > Resume sometiems works, but often, system is half-usable, with "what > > > > do you want to do suspend/hibernate" dialog staying on screen. > > > > > > > > I did echo disk > /sys/power/state, and echo never returned to the > > > > shell prompt. (X was not really working, either.) > > > > > > > > It is still there on 3.15.0-rc3+. I'll update to rc5 now. > > > > > > 3.15.0-rc5+ seems to behave a bit better. Lets see if it reappears. > > > > It did. > > So it might in the end of the day the same thing as I have reported in > > https://bugs.freedesktop.org/show_bug.cgi?id=76554 > > which is not 100% reproducible either (although with newer kernels it's > *much* more reproducible for me than with older). > > > It worked few times on -rc5, but then I rebooted and problem > > re-appeared. > > If you could capture dmesg of the resume where X fails to resume properly > to see whether you have i915 ring initialization errors there as well?
I did dmesg and I believe I would have noticed "*ERROR*" there.. but I'll of course capture the buffer when it happens.
Note that X do work somehow after resume (I can't switch virtual desktops and dialog is stuck on screen, but it is not complete failure). I can do ctrl-alt-f1 and get to useful prompt.
Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
| |