Go Back   nV News Forums > Linux Support Forums > NVIDIA Linux

Newegg Daily Deals

Reply
 
Thread Tools
Old 06-22-05, 09:16 AM   #1
zander
NVIDIA Corporation
 
zander's Avatar
 
Join Date: Aug 2002
Posts: 3,740
Default 1.0-7667 for Linux x86 released

Release Highlights:

* Added support for GeForce 7800 GTX.

* Fixed problem with certain flatpanels running at 1600x1200.

* Fixed an unresolved symbol issue in libXvMCNVIDIA.so.

* Worked around a bug in Java2D that was exposed in 1.0-7664.

* Improved support for Quadro FX 4000 SDI.
The 1.0-7667 NVIDIA Accelerated Linux Graphics Driver Set for Linux/x86 is available for download via HTTP and FTP.
zander is offline   Reply With Quote
Old 06-22-05, 10:07 AM   #2
metamatics
Free Software Zealot
 
Join Date: Apr 2004
Posts: 39
Default Re: 1.0-7667 for Linux x86 released

No bugfixes?
Damnit!!!!
metamatics is offline   Reply With Quote
Old 06-22-05, 10:28 AM   #3
zyphr
Registered User
 
Join Date: Mar 2004
Posts: 21
Default Re: 1.0-7667 for Linux x86 released

very nice move from nvidia with linux support for the GeForce 7800 GTX on release day.
I think I'll get one very soon cuz of this release.
zyphr is offline   Reply With Quote
Old 06-22-05, 01:09 PM   #4
Evil Penguin
Registered User
 
Join Date: May 2004
Posts: 49
Default Re: 1.0-7667 for Linux x86 released

Quote:
Originally Posted by metamatics
No bugfixes?
Damnit!!!!
"Fixed...
Fixed...
Worked around..."

No bugfixes, yeah right.. :P
Evil Penguin is offline   Reply With Quote
Old 06-22-05, 01:19 PM   #5
blueworm
[DEADAGAIN]blueworm
 
blueworm's Avatar
 
Join Date: Mar 2003
Posts: 416
Default Re: 1.0-7667 for Linux x86 released

Quote:
Originally Posted by zyphr
very nice move from nvidia with linux support for the GeForce 7800 GTX on release day.
I think I'll get one very soon cuz of this release.
Its a first! YAY
And Maybe I wont get one!
__________________
Registered linux user #271604 Get Counted! http://counter.li.org/
lianLi PC60, OCZ powerStream 520W PSU, Asus P5N-E SLI, Core 2 duo E6550, 2Gb GEIL, XFX 8800GT, TB SantaCruz
blueworm is offline   Reply With Quote
Old 06-22-05, 01:42 PM   #6
metamatics
Free Software Zealot
 
Join Date: Apr 2004
Posts: 39
Default Re: 1.0-7667 for Linux x86 released

Quote:
Originally Posted by Evil Penguin
"Fixed...
Fixed...
Worked around..."

No bugfixes, yeah right.. :P
Are you kidding?
I want fixes for crashes, not for "problem with certain flatpanels running at 1600x1200.",
or XVMC that is not so important (ok, it is, but less then the stability of the X server).
Like a fix for RenderAccel that is currently almost impossible to enable.

BTW I am not sure of what means "Worked around a bug in Java2D that was exposed in 1.0-7664":
Was it a bug of 1.0-7664 that was triggered by Java2D?
Or is it just something to make work a buggy program?
metamatics is offline   Reply With Quote
Old 06-22-05, 04:35 PM   #7
zap
Registered User
 
Join Date: Feb 2003
Location: St. Petersburg, Russia
Posts: 20
Default Re: 1.0-7667 for Linux x86 released

Okay, today I've tried to upgrade to the new driver and had to return back to version 6629 that I was using before. It locks up so often that it makes impossible any work. In fact, it doesn't "crash" so much but the screen is completely locked even that the mouse cursor is moving.

After rebooting (with hard reset, argh!) I can find in kernel log lines like:

00:23:32 zap kernel: NVRM: Xid: 13, 0000 02005600 00001796 00000c28 003e0007 00000080
00:44:57 zap kernel: NVRM: Xid: 13, 0000 02005600 00001796 00000c2c 00010001 00000080

My configuration: VIA KT333A, Athlon XP 1700+, nVidia GeForce 440MX SE, RenderAccel enabled (maybe this was the cause? was so angry after second lockup that didn't tried to check).

The 6629 driver has an annoying bug as well, and I'm not even sure this version fixes it: I have xscreensaver running, and since it detects all available screens (and there's no way to disable it ) and I have a dual screen configuration (LCD + TV-out) it runs two GL screensavers in parallel. When this happens it looks like something bad happens with OpenGL (like reciprocal interlocks of two simultaneous GL programs). Music, if it's playing starts to hiccup annoyingly, all processes stop for a second-two with regular intervals of 2-3 seconds. Can somebody confirm this has been or haven't been fixed in the 7xxx series of drivers?
zap is offline   Reply With Quote
Old 06-22-05, 08:45 PM   #8
gamucf
Registered User
 
Join Date: Jun 2005
Posts: 3
Default Re: 1.0-7667 for Linux x86 released

I agree. They need to cease all new development and get these crucial bugs fixed before going on. I have an entire lab full of useless 6600GT cards because the driver locks up the machine literally every 10 minutes. Thousands of dollars going down the drain. I've told everybody that we're not getting any 7800's until the 6600's work.

Glenn
gamucf is offline   Reply With Quote

Old 06-23-05, 07:19 AM   #9
arokh
Registered User
 
arokh's Avatar
 
Join Date: Jun 2004
Posts: 287
Default Re: 1.0-7667 for Linux x86 released

Quote:
Originally Posted by gamucf
I agree. They need to cease all new development and get these crucial bugs fixed before going on. I have an entire lab full of useless 6600GT cards because the driver locks up the machine literally every 10 minutes. Thousands of dollars going down the drain. I've told everybody that we're not getting any 7800's until the 6600's work.

Glenn
How come I'm running 6600GT's without problems on several different machines and distributions at work then? Are you sure these instabilities you are talking about is because of the nvidia driver?
__________________
Asus A8N32-SLI : AMD64 3700+ @ 2.85GHz : 2GB PC4000 OCZ Platinum EB : 2x 7800GT SLI
arokh is offline   Reply With Quote
Old 06-23-05, 10:30 AM   #10
kennyz
Registered User
 
Join Date: Dec 2003
Posts: 10
Default Re: 1.0-7667 for Linux x86 released

I cannot get the kernel module (either 7667 or 7174) to load on my system, with kernel 2.6.12.1 and a GeForce3 Ti500.

Both 7667 and 7174 compiled and loaded on kernel 2.6.11.11 with the GeForce3 Ti500, but after upgrading to kernel 2.6.12.1, neither module will load (although the compilation works fine - with a few warnings).

Also interesting to note is that 7667 DOES compile and load on my main system, which is running 2.6.12.1 with a GeForce 6600GT.

Therefore, it seems that some combination of kernel 2.6.12.1 with the older GeForce3 Ti500 is causing the module (either 7174 or 7667) to fail loading.

Note that I assured that rivafb is not included in the kernel.

The log message is:

Quote:
Jun 23 11:25:09 jupiter kernel: [ 395.915932] NVRM: The NVIDIA probe routine was not called for 1 device(s).
Jun 23 11:25:09 jupiter kernel: [ 395.916045] NVRM: This can occur when a driver such as rivafb or rivatv was
Jun 23 11:25:09 jupiter kernel: [ 395.916048] NVRM: loaded and obtained ownership of the NVIDIA device(s).
Jun 23 11:25:09 jupiter kernel: [ 395.916147] NVRM: Try unloading the rivafb (and/or the rivatv) kernel module
Jun 23 11:25:09 jupiter kernel: [ 395.916150] NVRM: (or reconfigure your kernel without rivafb support), then
Jun 23 11:25:09 jupiter kernel: [ 395.916152] NVRM: try loading the NVIDIA kernel module again.
Jun 23 11:25:09 jupiter kernel: [ 395.916329] NVRM: No NVIDIA graphics adapter probed!
kennyz is offline   Reply With Quote
Old 06-23-05, 11:00 AM   #11
lyceel
Registered User
 
Join Date: Feb 2005
Posts: 10
Default Re: 1.0-7667 for Linux x86 released

Quote:
Originally Posted by arokh
How come I'm running 6600GT's without problems on several different machines and distributions at work then? Are you sure these instabilities you are talking about is because of the nvidia driver?
We're pretty sure it's at least a combination of the card and the driver (I work in the same lab). When we switch back to the Quadro NVS cards that the machines came with, the crashes stop. We've had minor issues before with 5600's, but not nearly as bad as they are with the 6600GT PCIe cards.

The machines in question are Dell Precision 670's with dual Xeon 3.2GHz processors and 1GB Ram. The cards we bought for them are PNY 6600GT PCI-e cards. We've had a problem with every single workstation we put the card in. Going from 6629 to 7174 to 7664 to 7667 has not helped at all. Whenever the machine is doing something OpenGL-intensive (even as simple as glxgears), it tends to lock up in under 5 minutes, nearly 100% reproducible.

Unlike a lot of people, we don't typically get any Xid errors or any pertinent log messages at all. It just hangs, black screen, no keyboard LED's, nothing. Only a cold start will bring it back. I've attached a bug report (from a reboot after a crash) in hopes that it will be useful to someone.
Attached Files
File Type: zip nvidia-bug-report.zip (18.6 KB, 322 views)
lyceel is offline   Reply With Quote
Old 06-23-05, 11:34 AM   #12
ghostwalker
Registered User
 
Join Date: Sep 2003
Location: Tacoma, WA
Posts: 76
Default Re: 1.0-7667 for Linux x86 released

If you have renderaccel enabled? disable it and try again. This sounds like the problems I was having.
__________________
P4 2.8Ghz @ 2.8Ghz SL6WT
Zalman CNPS7000-Cu
865PE Neo-2 LS BIOs 2.4
512mb Mushkin Level II 256x2
160GB Maxtor HD
Geforce 4 440
Antec 3700 1 exhaust and 1 intake fans
Linux 2.6.x ArchLinux
ghostwalker is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 04:22 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 1998 - 2014, nV News.