| Summary: | >nvidia-drivers-100.14.23 causing xorg-server segfaults w/ for Nvidia 7900gs&7600gt | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Georg Witwer <georg.witwer> |
| Component: | Current packages | Assignee: | Tony Vroon (RETIRED) <chainsaw> |
| Status: | RESOLVED TEST-REQUEST | ||
| Severity: | critical | CC: | gentoo, tommy |
| Priority: | High | ||
| Version: | 2006.1 | ||
| Hardware: | AMD64 | ||
| OS: | Linux | ||
| URL: | http://forums.gentoo.org/viewtopic-p-4945187.html#4945187 | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
|
Description
Georg Witwer
2008-03-09 21:48:55 UTC
We can't fix nvidia drivers at all; submit this to nVidia together w/ nvidia-bug-report.sh output and a backtrace. http://www.gentoo.org/proj/en/qa/backtraces.xml (In reply to comment #1) > We can't fix nvidia drivers at all; submit this to nVidia together w/ > nvidia-bug-report.sh output and a backtrace. > > http://www.gentoo.org/proj/en/qa/backtraces.xml > OKIDO, thx and sorry to have you bothered ;) We can however pmask or keyword the 169.* series again until nvidia fix the problem. I'm also experiencing this with the 169* series along with other issues after rebooting when it hard freezes the machine. Here is the sequence of events that follow for me: * X11 hard freezes * Upon reboot 'startx' puts monitor to sleep * Upon subsequent reboots monitor stays asleep until power is cycled to monitor Very odd bug(s?), but always reproducible :( Same monitor sleeping bug touched on here -> http://www.nvnews.net/vbulletin/showthread.php?t=106886 Rolling back to nvidia-drivers-100.14.19 fixes the problem for me. I have segfaults problem with 169.* on 32-bit gentoo as well. GDM fails to start with a segfault and gtk+ fails to compile (gtk-query-immodules-2.0 bails out with a segfult).
>=169.0 should be hard masked IMHO...
Chainsaw is taking over this package. (In reply to comment #3) > We can however pmask or keyword the 169.* series again until nvidia fix the > problem. That assumes that others have the problem, which is not the case. Please let me know whether this issue exists on 173.14.05 We already do some level of detection to suggest specific masks to the user, I have no problem detecting the 7600GT & 7900GS adapters through their PCI IDs and suggesting a masking that does not offer the 169 drivers to you. Note that this will deprive you of kernels newer then 2.6.25 Requesting testing on 173.14.05 which hasn't happened yet. Please try 173.14.09 and reopen the bug if that does not resolve the issue for you. |