Version: 4.6 (using KDE 4.6.2) OS: Linux using kubuntu 11.04 x64 When browsing the internet and doing nothing in particular, I encountered a ui lockup. I could only move the mouse cursor. Nothing else worked, including ctlr-alt-f1. This was in kernel.log: May 5 10:29:31 compy kernel: [63008.200848] Btrfs loaded May 5 17:37:42 compy kernel: [88699.779915] hub 4-0:1.0: port 3 disabled by hub (EMI?), re-enabling... May 5 17:37:42 compy kernel: [88699.779920] usb 4-3: USB disconnect, address 5 May 5 17:37:43 compy kernel: [88700.430039] usb 4-3: new low speed USB device using ohci_hcd and address 6 May 5 17:37:43 compy kernel: [88700.680159] input: CHESEN USB Keyboard as /devices/pci0000:00/0000:00:12.0/usb4/4-3/4-3:1.0/input/input11 May 5 17:37:43 compy kernel: [88700.680286] generic-usb 0003:0A81:0101.0009: input,hidraw0: USB HID v1.10 Keyboard [CHESEN USB Keyboard] on usb-0000:00:12.0-3/input0 May 5 17:37:43 compy kernel: [88700.689952] input: CHESEN USB Keyboard as /devices/pci0000:00/0000:00:12.0/usb4/4-3/4-3:1.1/input/input12 May 5 17:37:43 compy kernel: [88700.690077] generic-usb 0003:0A81:0101.000A: input,hidraw1: USB HID v1.10 Device [CHESEN USB Keyboard] on usb-0000:00:12.0-3/input1 May 5 18:48:53 compy kernel: [92970.256456] NVRM: Xid (0000:01:00): 13, 0001 00000000 00008597 000015e0 00000000 00000100 May 5 18:51:16 compy kernel: imklog 4.6.4, log source = /proc/kmsg started. May 5 18:51:16 compy kernel: [ 0.000000] Initializing cgroup subsys cpuset Reproducible: Sometimes
If this happens again, could you add more information, such as: - if it only happens when using the browser (which one?) - if desktop effects are enabled - if yes, which video drivers you use If you cannot switch to console, it is usually the X server which hangs because of video driver bugs.
I have the same problem on my Ubuntu 11.04 x64 box. And I mean the VERY same, i.e. it is precisely the NVRM: Xid (0000:01:00): 13, 0001 00000000 00008597 000015e0 00000000 00000100 message I get in my dmesg just before this happens. I can reproduce this with near certainty by starting up emacs in GTK mode (in a KDE 4.6.3 session) and "half-maximizing" the window (by dragging it to the middle of the left screen edge). Just about any kind of resizing the emacs window, sometimes even moving it around, will do though. Even if the system does not lock up immediately, I experience the random black "snow" other people also reported as a by-product of an nvidia driver going mad (i.e. black pixel artifacts throughout the screen). Once this snow has started appearing, even when I close emacs again, the "snow" problem persists; random black pixels appear everywhere when the screen content changes (e.g., by moving a window around). If I manage to suspend the box (the nvidia driver sometimes causes lockups when trying to suspend, but this is another, independent issue) and then resume again, the snow is gone and everything works fine again (until I start emacs-gtk again). All this happens both with desktop effects enabled and with effects disabled (with the Shift-Alt-F12 toggle). Symptoms (of the lockup, after the "black snow" phase) are in both cases: The mouse cursor moves, but anything else locks up hard. I can still ssh into the box (for some time, at least), but there is no way to restart the X server. Magic Sysrq-B works for rebooting, unless I have tried too many other things before (like Ctrl-Alt-F1 or hitting the suspend hotkey when it's already locked up -- please note that the above remedy of suspending and resuming only works while there is still the black snow that comes before the hard lockup). Once I attempt a suspend when the lockup is already there, only holding down the power button will work, nothing else. My machine is a Sony Vaio VPC F12 S1E laptop with an onboard 01:00.0 VGA compatible controller: nVidia Corporation GT216 [GeForce GT 330M] (rev a2) and I am using nvidia-current-275.09-0ubuntu1~edgers~natty, but the exact same phenomenon occurs also with nvidia-current-260.19.06-0ubuntu1, nvidia-current-270.41.06-0ubuntu1, and nvidia-current-270.41.19-0ubuntu1~xup~natty. My emacs version is emacs23-23.2+17ubuntu2, but it also happens with no difference in emacs-snapshot-1:20090909-1. I have not yet encountered any other program provoking the same behavior (nor does emacs in a konsole window -- but that is a pain ... Yes, vim addicts, I know what you are thinking now ... don't). With the nouveau drivers all this does not happen, but they exhibit different problems like random lockups not related to any specific application and thus are not an option for me. Thanks a lot for your time and patience to read through my crap -- I was just so happy that there was another user with the same NVRM: Xid ... line, so I finally hope it's not just me ... If there is anything I can do to help debug this problem, please let me know.
One more detail: This issue is not related to anything in my .emacs file, it also happens when I invoke `emacs -q` (i.e. in vanilla mode). And another: It happens no matter which GTK style I used (tried oxygen-molecule and QtCurve).
Good news: With yesterday's update to nvidia-current-275.09.04-0ubuntu1~edgers~natty, the problem went away. Thanks to whomever it may concern! :-)
It does seem to have been an nvidia driver problem, as I haven't had snow or lockup problems since an update a while back. Thanks lpetersen
Have the same problem, KDE locks up and only the mouse cursor moves. The only way out is a restart via the big red button. However, this only happens on our two new Intel i7 and i5 Sandy Bridge boxes. The AMD boxes never see this problem. The i7 is an HP Pvilion dv6 portable with a Radeon 6k series video card and the i5 box is a desktop with a GeForce card so I do not think that it is x related, rather something to do with Intel and Sandy Bridge. Sorry about the lack of a log ....