Bug 35344 - nouveau: oops in nv50_vm_map_pgt
Summary: nouveau: oops in nv50_vm_map_pgt
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/nouveau (show other bugs)
Version: 7.4 (2008.09)
Hardware: x86-64 (AMD64) Linux (All)
: medium blocker
Assignee: Nouveau Project
QA Contact: Xorg Project Team
Depends on:
Reported: 2011-03-15 17:02 UTC by somethingsome2000
Modified: 2012-06-26 09:54 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:

dmesg logfile from boot (72.53 KB, text/plain)
2011-03-15 17:04 UTC, somethingsome2000
no flags Details
xorg log file (16.76 KB, text/plain)
2011-03-15 17:04 UTC, somethingsome2000
no flags Details
Dmesg log (62.38 KB, text/plain)
2011-03-16 13:12 UTC, Michał Lipski
no flags Details

Description somethingsome2000 2011-03-15 17:02:15 UTC
This did not happen with kernel 2.6.38-rc8 but started with the 2.6.38 release
Comment 1 somethingsome2000 2011-03-15 17:04:06 UTC
Created attachment 44489 [details]
dmesg logfile from boot
Comment 2 somethingsome2000 2011-03-15 17:04:41 UTC
Created attachment 44490 [details]
xorg log file
Comment 3 Michał Lipski 2011-03-16 13:11:55 UTC
I saw it only ones. Just after kernel upgrade from 2.6.38-rc7 to 2.6.38-rc8. I couldn't reproduce it.

kernel: 2.6.38
xorg-server: 1.9.4

Right now I use 2.6.38.
Comment 4 Michał Lipski 2011-03-16 13:12:42 UTC
Created attachment 44520 [details]
Dmesg log
Comment 5 somethingsome2000 2011-03-25 23:56:05 UTC
Very odd. I noticed that my kernel was compiled without debugging, so I decided to recompile with debugging enabled to get useful output for you guys. The strange thing is that with debugging enabled, my kernel can now boot properly
Comment 6 Marcin Slusarz 2011-12-06 09:58:07 UTC
Is it still crashing on newer kernel? I think we fixed it in / 2.6.39-rc5.

(Logs are from 2.6.38 / 2.6.38-rc8 and in both oops triggered on address which had high word set to 0xffffffff. Considering that between 2.6.38-rc7 and we had a memory corruption bug which could overwrite one word with 0xffffffff, it's quite possible it's the same bug...)

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.