Bug 36742 – nvidia-kernel 1.0.5328 causes oops

• Status: RESOLVED FIXED
• Product: Gentoo Linux

[Description Axel Reimann 2003-12-29 09:54:14 UTC]

The nvidia module 1.0.5328 compiles fine 
but causes kernel message printed below when loaded.
A side effect is that the module stays listed (lsmod)
and cannot be 'unloaded' (device or resource busy).

I had no trouble whatsoever going back to 1.0.4496-r3.

Reproducible: Didn't try
Steps to Reproduce:
1. emerge nvidia-kernel and nvidia-glx 1.0.5328
2. stop X, unload old nvidia kernel module
3. load new kernel module or restart machine alternatively

Actual Results:  
Dec 29 17:18:42 juliette kernel: 0: nvidia: loading NVIDIA Linux x86 nvidia.o 
Kernel Module  1.0-5328  Wed Dec 17 13:54:51 PST 2003
Dec 29 17:18:50 juliette kernel:  printing eip:
Dec 29 17:18:50 juliette kernel: c03163e0
Dec 29 17:18:50 juliette kernel: Oops: 0000
Dec 29 17:18:50 juliette kernel: CPU:    0
Dec 29 17:18:50 juliette kernel: EIP:    0010:[<c03163e0>]    Tainted: P
Dec 29 17:18:50 juliette kernel: EFLAGS: 00013046
Dec 29 17:18:50 juliette kernel: eax: 00000087   ebx: 00003246   ecx: 00000048   
edx: 00000000
Dec 29 17:18:50 juliette kernel: esi: 00000000   edi: dffe0000   ebp: c79bb738   
esp: c79bb708
Dec 29 17:18:50 juliette kernel: ds: 0018   es: 0018   ss: 0018
Dec 29 17:18:50 juliette kernel: Process X (pid: 22389, stackpage=c79bb000)
Dec 29 17:18:50 juliette kernel: Stack: dcffc800 c79bb764 c01c642c dfef9c00 
00003099 e117b358 00000000 00000048
Dec 29 17:18:50 juliette kernel:        c79bb734 e118a45e dd5cb000 00000000 
c79bb748 e119500d 00000000 00000048
Dec 29 17:18:50 juliette kernel:        c79bb774 e118d626 dfef9c00 00000000 
00000048 00000400 d93f8000 dcffc800
Dec 29 17:18:50 juliette kernel: Call Trace:    [<c01c642c>] [<e117b358>] 
[<e118a45e>] [<e119500d>] [<e118d626>]
Dec 29 17:18:50 juliette kernel:   [<e1190ef7>] [<e11928b0>] [<e13564e0>] 
[<e119935e>] [<e1179f7f>] [<e13564e0>]
Dec 29 17:18:50 juliette kernel:   [<e13152de>] [<e1356d40>] [<e13564e0>] 
[<e1356d40>] [<e13564e0>] [<e1196d9c>]
Dec 29 17:18:50 juliette kernel:   [<e13564e0>] [<e1356dac>] [<e1356dd0>] 
[<e1356de8>] [<e13564e0>] [<e1199f49>]
Dec 29 17:18:50 juliette kernel:   [<e13564e0>] [<e13564e0>] [<e118a45e>] 
[<e1194f58>] [<e1304891>] [<e119691b>]
Dec 29 17:18:50 juliette kernel:   [<e1304891>] [<e1304891>] [<e119696c>] 
[<e132e800>] [<e1304891>] [<e11cac3b>]
Dec 29 17:18:50 juliette kernel:   [<e11cac3b>] [<e11ad4db>] [<e11baead>] 
[<e1260b5b>] [<e1260b5b>] [<e118ab5a>]
Dec 29 17:18:50 juliette kernel:   [<e117b8d5>] [<e117c09b>] [<e117bf70>] 
[<e118ace5>] [<e11964c2>] [<e13564e0>]
Dec 29 17:18:50 juliette kernel:   [<e1227d67>] [<e1260e7f>] [<e1260b5b>] 
[<e1260b5b>] [<e1260b5b>] [<e1260b5b>]
Dec 29 17:18:50 juliette kernel:   [<e126125d>] [<e12a98d0>] [<e124f398>] 
[<e127d847>] [<e1242f73>] [<e11964c2>]
Dec 29 17:18:50 juliette kernel:   [<e13564e0>] [<e1244bc1>] [<e1273705>] 
[<e119aac9>] [<e1244a85>] [<c020ba89>]
Dec 29 17:18:50 juliette kernel:   [<c039c14f>] [<c020cb61>] [<c039c74c>] 
[<c021b797>] [<c0219765>] [<e1199331>]
Dec 29 17:18:50 juliette kernel:   [<e13564e0>] [<e1178c01>] [<e13564e0>] 
[<c01ca6f4>] [<c020a78c>] [<c01f79b6>]
Dec 29 17:18:50 juliette kernel:   [<e1178943>] [<c0205c0d>] [<c01be553>]
Dec 29 17:18:50 juliette kernel:
Dec 29 17:18:50 juliette kernel: Code: 8b 46 10 8b 50 30 89 34 24 89 4c 24 04 8b 
44 24 20 89 44 24
Dec 29 17:18:50 juliette kernel:  <1>Unable to handle kernel NULL pointer 
dereference at virtual address 00000010

Expected Results:  
load fine :)

Portage 2.0.49-r18 (default-x86-1.4, gcc-3.3.2, glibc-2.3.3_pre20031222-r0, 2.4.
23-grsec-1.9.13)
=================================================================
System uname: 2.4.23-grsec-1.9.13 i686 AMD Athlon(tm) XP 1800+
Gentoo Base System version 1.4.3.12
ACCEPT_KEYWORDS="x86 ~x86"
AUTOCLEAN="yes"
CFLAGS="-march=i686 -O3 -pipe"
CHOST="i686-pc-linux-gnu"
COMPILER="gcc3"
CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.
1/share/config /usr/kde/3/share/config /usr/share/config 
/usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ 
/usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ 
/usr/share/texmf/xdvi/ /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d"
CXXFLAGS="-march=i686 -O3 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs ccache fixpackages sandbox"
GENTOO_MIRRORS="http://gentoo.oregonstate.edu http://distro.ibiblio.
org/pub/Linux/distributions/gentoo"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY=""
SYNC="rsync://rsync.namerica.gentoo.org/gentoo-portage"
USE="X aalib aavm acl acpi acpi4linux alsa apache2 apm arts artswrappersuid avi 
berkdb cdr cjk crypt cups dga doc dvd encode esd foomaticdb gdbm gif gnome 
gphoto2 gpm gps gtk gtk2 guile hbci imap imlib ipv6 java javascript jpeg kde 
lcms ldap libg++ libwww mad maildir mbox md5sum mikmod motif mozilla mpeg mule 
mysql nas ncurses nls oci8 odbc oggvorbis opengl oss pam pda pdflib perl pic 
plotutils png ppds python qt quicktime readline ruby sasl scanner sdl slang 
spell sse ssl svga tcltk tcpd tetex tiff truetype trusted type1 usb v4l videos 
wmf x86 xml xml2 xmms xosd xv xvid zlib"

[Comment 1 Olivier Castan 2003-12-29 13:09:11 UTC]

According to Minion (http://www.minion.de/nvidia.html) their patch may be applied even with 2.4 kernels.
On nvidia forums I found at least 1 post telling it solved the problem (http://www.nvnews.net/vbulletin/showthread.php?s=7f11acc7f0009a5ea5b4d6b68b66a8e3&threadid=22598&perpage=20&pagenumber=2)
So we have to find what are the criterias to apply the patch on 2.4 kernels...

[Comment 2 Seemant Kulleen (RETIRED) 2003-12-29 13:30:53 UTC]

the way I see it, minion patches should be applied regardless of kernel version (that's what I understand from the nv forums)

[Comment 3 Olivier Castan 2003-12-29 13:41:38 UTC]

As far as I understand the problem seems related to (recent ?) VIA chipsets...

[Comment 4 Seemant Kulleen (RETIRED) 2003-12-29 14:33:57 UTC]

*** Bug 36758 has been marked as a duplicate of this bug. ***

[Comment 5 Donnie Berkholz (RETIRED) 2003-12-29 22:24:06 UTC]

At least for me, minion.de patch caused compiles to be impossible on 2.4 (as mentioned on other bug).

[Comment 6 Axel Reimann 2003-12-30 00:17:29 UTC]

I'm running Gentoo on an EPOX board with VIA KT333 (KT333 + VT8235 + HPT374 (unused)) chipset, if that helps. It's not exactly new, I guess.

[Comment 7 Martin Schlemmer (RETIRED) 2003-12-31 16:35:11 UTC]

*** Bug 36689 has been marked as a duplicate of this bug. ***

[Comment 8 Martin Schlemmer (RETIRED) 2003-12-31 16:40:07 UTC]

Oops, wrong bug.

[Comment 9 David Oftedal 2004-01-03 06:04:10 UTC]

On my machine it causes ZSNES to segfault when setting the mode to 1024x768 GL. It must not be released to the general public.

[Comment 10 Andreas Boeyer 2004-01-04 11:27:23 UTC]

I have also a problem with nvidia-kernel 1.0.5328
After I emerge the nvidia-kernel 1.0.5328 and nvidia-glx 1.0.5328
my KDE didn't start and what was more worse that even my regular terminal
1 to 6 had no display. So I took the gentoo cd and booted from cd.
Then I moved back to nvidia-kernel 1.0.4496-r4 and nvidia-glx 1.0.4496-r1.
Now every thing working fine.

[Comment 11 Donnie Berkholz (RETIRED) 2004-01-05 12:50:47 UTC]

Please test 1.0.5328-r1 of nvidia-kernel and nvidia-glx, I just committed them. Should fix a TLS issue, an issue on VIA chipsets, and whatever the updated minion.de patch fixed for 2.6 kernels.

[Comment 12 Olivier Castan 2004-01-06 13:41:08 UTC]

New ebuilds solved the problem on my box :
- linux-2.4.22-gentoo-r2
- via KT333 chipset
- Nvidia GeForce2 MX/MX 400

[Comment 13 Olivier Castan 2004-01-08 13:18:55 UTC]

My computer freezed when running The GLForestFire screensaver... that may be a nvidia-glx/kernel problem. Unfortunately I didn't found any oops...

[Comment 14 Axel Reimann 2004-01-09 12:23:38 UTC]

Same for me: new ebuilds solved the problem on my box as well.
The Ooops is gone. :)

[Comment 15 Olivier Castan 2004-01-09 12:42:07 UTC]

In fact any OpenGL screensaver freezes my box. nvidia-glx-1.0.5328* and nvidia-kernel-1.0.5328* won a place in my /etc/portage/package.mask, I give up.

[Comment 16 Andrew Bevitt 2004-04-13 07:29:59 UTC]

What about 5336 does it work for you?

[Comment 17 Andrew Bevitt 2004-05-10 20:40:08 UTC]

Please test nvidia-kernel-5336-r3 (preferably with 2.6.6 kernel, but not essential) and report back please.

[Comment 18 Axel Reimann 2004-05-10 23:02:56 UTC]

Works nicely on 2.4.25 kernel, and
Open-GL is doing fine as well.

Thank you!

[Comment 19 Andrew Bevitt 2004-05-21 07:26:16 UTC]

Closing.

Fixed in newer versions.