|
|
#1 | |
|
Registered User
Join Date: Nov 2010
Posts: 2
|
After having invested in a Geforce 480 card, I found that none of the latest Nvidia drivers work on my machine with it. A few days ago the new 260.19.29 driver came out, and trying that gives the same errors as all the previous 260.* series drivers:
In /var/log/messages: Dec 17 12:30:26 sedna kernel: NVRM: os_schedule: Attempted to yield the CPU while in atomic or interrupt context Dec 17 12:30:49 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000002 Dec 17 12:30:57 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000003 Dec 17 12:31:05 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000004 Dec 17 12:31:13 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000005 Dec 17 12:31:21 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000006 Dec 17 12:31:29 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000007 Dec 17 12:31:37 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000008 Dec 17 12:31:45 sedna kernel: NVRM: Xid (0006:00): 16, Head 00000000 Count 00000009 In /var/log/Xorg.0.log: ... [ 123.187] (II) NVIDIA(0): Initialized GPU GART. [ 123.192] (II) NVIDIA(0): Setting mode "nvidia-auto-select" [ 131.228] (II) Loading extension NV-GLX [ 135.229] (EE) NVIDIA(0): WAIT: (E, 0, 0x857d, 0) After these, the machine freezes with a blank screen. The last driver that works stably for me was 195.36.31 (and 195.36.24). The drivers since then were all unstable, either locking the machine up completely, or producing messages like above. I would really appreciate some help on how to proceed, since the latest Fedora distribution doesn't support the old 195.* drivers even. The nvidia-bug-report output is attached. Thanks, Andras |
|
|
|
|
|
|
#2 | |
|
Registered User
Join Date: May 2004
Location: Norway
Posts: 117
|
I'm getting somewhat similar errors (i.e. the "os_schedule.. " part), and the only drivers that work stably (no errors) are from the 195-series. Ever since, there's been problems. X tends to hang for long periods of time here when the errors occur, though it rarely crashes. Hardware is not the same, I'm on a Dell laptop with an NVS 3100M. I can most easily trigger the problem with console switching. Here's the thread in which I reported the problem:
http://www.nvnews.net/vbulletin/showthread.php?t=157041 I keep that thread updated with testing from the latest releases, to determine if the problem is solved or not. Never gotten any official response though. Hardware fault is not likely, since everything is very stable with the older drivers, and Windows 7 works fine. |
|
|
|
|
![]() |
| Thread Tools | |
|
|