KDE no longer works properly after resuming from suspend-to-RAM. The first suspend/resume cycle still works ok some of the time; otherwise, there's a little pixel junk stuck to a GUI element or the desktop. The next time around, the GUI is no longer drawn completely and/or is laggy/unresponsive. The screens keep going blank intermittently (as in "no signal"). The mouse gets "stuck" a lot.
If I manage to run "kwin --replace", everything's magically okay again. Would that suggest it's not, in fact, time to buy yet another AGP graphics card (of great antiquity)?
I see a lot of these in Xorg log:
Backtrace:
[ 1129.747] 0: /usr/bin/Xorg (xorg_backtrace+0x37) [0x982f87]
[ 1129.747] 1: /usr/bin/Xorg (mieqEnqueue+0x223) [0x9612d3]
[ 1129.747] 2: /usr/bin/Xorg (0x7fb000+0x4cac5) [0x847ac5]
[ 1129.748] 3: /usr/bin/Xorg (xf86PostMotionEventM+0xf9) [0x888b69]
[ 1129.748] 4: /usr/lib/xorg/modules/input/evdev_drv.so (0x42d000+0x3c3d) [0x430c3d]
[ 1129.748] 5: /usr/lib/xorg/modules/input/evdev_drv.so (0x42d000+0x524f) [0x43224f]
[ 1129.748] 6: /usr/bin/Xorg (0x7fb000+0x78081) [0x873081]
[ 1129.748] 7: /usr/bin/Xorg (0x7fb000+0x9fab8) [0x89aab8]
[ 1129.748] 8: (vdso) (__kernel_sigreturn+0x0) [0x110400]
[ 1129.748] 9: /usr/lib/i386-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x1c42b000+0x61bf9) [0x1c48cbf9]
[ 1129.748] [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
[ 1129.748] [mi] mieq is *NOT* the cause. It is a victim.
[ 1129.748] [mi] Increasing EQ size to 512 to prevent dropped events.
[ 1129.749] [mi] EQ processing has resumed after 13 dropped events.
[ 1129.749] [mi] This may be caused my a misbehaving driver monopolizing the server's resources.
Is this happening to anyone else? Googling just brings up Flash issues, but I don't seem to have them. The ttys are unaffected, too. Stuff like VLC or glxgears runs fine. Desktop effects run fine. It's just that after resuming from suspend I have to restart kwin. Very annoying!
I'm using the proprietary Nvidia drivers. Everything's up-to-date. I can't test it with Nouveau because that won't let me suspend successfully in the first place.
If I manage to run "kwin --replace", everything's magically okay again. Would that suggest it's not, in fact, time to buy yet another AGP graphics card (of great antiquity)?
I see a lot of these in Xorg log:
Backtrace:
[ 1129.747] 0: /usr/bin/Xorg (xorg_backtrace+0x37) [0x982f87]
[ 1129.747] 1: /usr/bin/Xorg (mieqEnqueue+0x223) [0x9612d3]
[ 1129.747] 2: /usr/bin/Xorg (0x7fb000+0x4cac5) [0x847ac5]
[ 1129.748] 3: /usr/bin/Xorg (xf86PostMotionEventM+0xf9) [0x888b69]
[ 1129.748] 4: /usr/lib/xorg/modules/input/evdev_drv.so (0x42d000+0x3c3d) [0x430c3d]
[ 1129.748] 5: /usr/lib/xorg/modules/input/evdev_drv.so (0x42d000+0x524f) [0x43224f]
[ 1129.748] 6: /usr/bin/Xorg (0x7fb000+0x78081) [0x873081]
[ 1129.748] 7: /usr/bin/Xorg (0x7fb000+0x9fab8) [0x89aab8]
[ 1129.748] 8: (vdso) (__kernel_sigreturn+0x0) [0x110400]
[ 1129.748] 9: /usr/lib/i386-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x1c42b000+0x61bf9) [0x1c48cbf9]
[ 1129.748] [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
[ 1129.748] [mi] mieq is *NOT* the cause. It is a victim.
[ 1129.748] [mi] Increasing EQ size to 512 to prevent dropped events.
[ 1129.749] [mi] EQ processing has resumed after 13 dropped events.
[ 1129.749] [mi] This may be caused my a misbehaving driver monopolizing the server's resources.
Is this happening to anyone else? Googling just brings up Flash issues, but I don't seem to have them. The ttys are unaffected, too. Stuff like VLC or glxgears runs fine. Desktop effects run fine. It's just that after resuming from suspend I have to restart kwin. Very annoying!
I'm using the proprietary Nvidia drivers. Everything's up-to-date. I can't test it with Nouveau because that won't let me suspend successfully in the first place.
Comment