Nvidia OpenGL/GLX libraries with 'new-style' ELF
Short version:
At www.minion.de the description of the 2.6 updates for the nvidia
kernel module sources state:
Users of Debian GNU/Linux (unstable) with libc6 2.3.2.ds1-10 or
later installed will need to install the set of NVIDIA OpenGL/GLX
libraries with support for the "new-style" ELF TLS mechanism in
/usr/lib/tls and /usr/X11R6/lib/modules/extensions respectively.
I am running the nvidia kernel module and nvidia XFree86 driver on a
Debian testing system with a 2.6.0-test11 kernel and libc6
2.3.2.ds1-10. How do I find or create these "NVIDIA OpenGL/GLX
libraries with support for the "new-style" ELF TLS mechanism"?
(Suggestions of alternative approaches are welcome.)
Long version:
I am running Debian testing on a computer using the Nvidia Nforce2
chipset. I installed Knoppix 3.3 then did an apt-get dist-upgrade to
Debian testing.
Still running the 2.4 series kernel I installed the official Nvidia
driver release with kernel modules nvnet, nvidia, and nvaudio.
I then prepared and installed a 2.6.0-test9-mm5 kernel, since upgraded
to 2.6.0-test11 with the forcedeth patch (Open Source network driver
for Nforce chipsets) applied.
Because I only have one PCI slot (it's a small form-factor computer)
in which I have a LeadTeK TV tuner card (bttv driver), I must use the
onboard networking, audio, and video. At present the forcedeth
networking is running fine and ALSA sound works with the snd_intel8x0
module. (Well, I wish it supported SPDIF/OUT lack of SPDIF/OUT on that
sound module for this chipset is a known issue.)
I installed the 2.6 kernel series patches for the nvidia kernel module
from www.minion.de and compiled and installed that module. I can run
either the nv and the nvidia XFree86 drivers but both give problems.
The xv driver spontaneously shuts down at random times. (It seems
that the shutdowns are related to mouse activity but I do not get log
messages in either /var/log/messages or /var/log/XFree86.0.log so it
is hard to diagnose.)
The nvidia XFree86 driver is stable but every 5 seconds messages like
Nov 27 09:16:06 bates1_home kernel: Badness in pci_find_subsys at drivers/pci/search.c:132
Nov 27 09:16:06 bates1_home kernel: Call Trace:
Nov 27 09:16:06 bates1_home kernel: [pci_find_subsys+232/240] pci_find_subsys+0xe8/0xf0
Nov 27 09:16:06 bates1_home kernel: [pci_find_device+47/64] pci_find_device+0x2f/0x40
Nov 27 09:16:06 bates1_home kernel: [pci_find_slot+40/80] pci_find_slot+0x28/0x50
Nov 27 09:16:06 bates1_home kernel: [_end+948821101/1069627092] os_pci_init_handle+0x39/0x70 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947595651/1069627092] __nvsym00057+0x1f/0x24 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+948464703/1069627092] __nvsym04236+0x1f/0x24 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+948301697/1069627092] __nvsym03984+0x215/0x1d88 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656654/1069627092] __nvsym01580+0x3a/0x70 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656654/1069627092] __nvsym01580+0x3a/0x70 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947783096/1069627092] __nvsym01575+0x28/0x34 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656781/1069627092] __nvsym01540+0x49/0x90 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947783295/1069627092] __nvsym01480+0x2b/0x34 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947655032/1069627092] __nvsym01583+0x58/0x68 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656654/1069627092] __nvsym01580+0x3a/0x70 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656654/1069627092] __nvsym01580+0x3a/0x70 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947783096/1069627092] __nvsym01575+0x28/0x34 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656781/1069627092] __nvsym01540+0x49/0x90 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656654/1069627092] __nvsym01580+0x3a/0x70 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947656654/1069627092] __nvsym01580+0x3a/0x70 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947783096/1069627092] __nvsym01575+0x28/0x34 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [__wake_up_common+49/96] __wake_up_common+0x31/0x60
Nov 27 09:16:06 bates1_home kernel: [_end+948314304/1069627092] __nvsym03993+0x15cc/0x15d8 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [__wake_up_common+49/96] __wake_up_common+0x31/0x60
Nov 27 09:16:06 bates1_home kernel: [_end+948300986/1069627092] __nvsym03986+0x21a/0x2cc [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+948324225/1069627092] __nvsym04031+0xc5/0x418 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+948323956/1069627092] __nvsym04015+0x68/0xb0 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+948280897/1069627092] __nvsym00610+0x85/0x954 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [sock_aio_read+184/208] sock_aio_read+0xb8/0xd0
Nov 27 09:16:06 bates1_home kernel: [do_sync_read+139/192] do_sync_read+0x8b/0xc0
Nov 27 09:16:06 bates1_home kernel: [_end+948476046/1069627092] __nvsym00688+0x16a/0x338 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947605645/1069627092] __nvsym00827+0xd/0x1c [nvidia]
Nov 27 09:16:06 bates1_home kernel: [_end+947611432/1069627092] rm_isr_bh+0xc/0x10 [nvidia]
Nov 27 09:16:06 bates1_home kernel: [tasklet_action+70/112] tasklet_action+0x46/0x70
Nov 27 09:16:06 bates1_home kernel: [do_softirq+144/160] do_softirq+0x90/0xa0
Nov 27 09:16:06 bates1_home kernel: [do_IRQ+253/304] do_IRQ+0xfd/0x130
Nov 27 09:16:06 bates1_home kernel: [common_interrupt+24/32] common_interrupt+0x18/0x20
Nov 27 09:16:06 bates1_home kernel:
appear in /var/log/messages, which is disconcerting. In case this is
related to the advice given at www.minion.de on upgrading the
OpenGL/GLX libraries, I wanted to start there but I am open to any
other suggestions.
--
Douglas Bates bates@stat.wisc.edu
Statistics Department 608/262-2598
University of Wisconsin - Madison http://www.stat.wisc.edu/~bates/
Reply to: