nV News Forums

 
 

nV News Forums (http://www.nvnews.net/vbulletin/index.php)
-   NVIDIA Linux (http://www.nvnews.net/vbulletin/forumdisplay.php?f=14)
-   -   Strange RedHat 7.3 issues with A7N266-C (http://www.nvnews.net/vbulletin/showthread.php?t=2188)

nforceuser 09-25-02 01:31 AM

Strange RedHat 7.3 issues with A7N266-C
 
This is a really strange problem. I have two systems using this board that are very similar.

System 1:

Asus A7N266-C rev 1.03 board - bios 1.001E (came with this board)
Athlon Xp 1900+ OC to 1680 MHz
512 MB PC 2100
GeForce 4 Ti 4200 video card

This was the first system I built. After installing Win2k, I used the freespace to install Redhat 7.3 (using GRUB and adding Win2k to the .conf file.) I then loaded the nForce chipset and nVidia graphics drivers, and changed the X11Config-4 file. This system works fine under Linux.

I ended up building a second system with:

Asus A7N266-C rev 1.03 board - bios 1.03 (came with this version)
Athlon XP 1900+ not OC
384 MB PC 2100
GeForce 2 MX 200 video card

I expected this system to work fine, but after installing RH 7.3 grub would fail to boot linux, reporting an error 28. I did some research and found nothing, but I did discover a workaround by using LILO and dumping the first 512 bytes of the linux partion and linking to them in the Win2k boot.ini file. Now Linux will boot on this system, but the fun doesn't end there. (Note that system one works fine with grub, the only real difference is the bios revision as far as I can tell. Bios 1.03 seems to have broken grub.) The next problem was that the system would COMPLETELY hang when APM was loaded as linux started. (System 1 loads it fine.) I stopped APM from loading and got Linux to start. I then installed the 3 nVidia drivers (one for chipset, 2 for graphics) and modified the config file. The system will always lock hard (only reset switch works) after typing startx. I just get a gray screen with a black X and then the system hangs. Note that system 1 works fine! I also get the same lockup just using the "nv" driver and before the nVidia drivers were ever installed. I'm thinking of flashing back system 2 to 1.001E bios, but I'm afraid that might break my Win2k install.

Just a quick rundown of what's above: System 1 works great, system 2, which differs realistically only in the bios revision has/had the following issues with RH 7.3: Grub reported error 28 and would not start linux. I worked around that one. The system would hang when starting the APM daemon. Typing startx will also hang the system. (For reference, system 2 is totally stable under Win2k SP2.)

As far as I can tell, bios 1.03 is the only real diffence between the systems that could be causing the problem! Have any of you seen this before???

Klaus-P 09-25-02 10:12 AM

Before you flash BIOS etc,...(i would like to call it the VERY LAST OPTION,
and not without risk) simply try to connect the harddrive of System 1
to System 2, check
the BIOS HD recognition before the first booting.

nforceuser 09-25-02 01:44 PM

It appears other people are seeing that Grub error, judging by the A7N266-VM post today. That's the exact same "out of memory" type error that I had.

I think Asus changed something with APM or APIC that is causing problems. I know 1.03 added support for LBA48 harddrives (i.e. >137 GB drives). I'm almost certain flashing system 2 to 1.001E will fix the problem, but will probably require a Win2k and Linux reinstall to work. I was hoping a newer kernal or driver might fix this...

It's strange that both systems are totally stable under Win2k, but only system 1 is trouble-free under Linux.

The only real difference is Video card and bios revision.

urak 09-26-02 04:00 AM

hi
 
I got the same problem
i got a a7n266-e motherboard with an integrated geforce2 mx 200 video adapter.

i am using debian 3.0 r0, athlon amn 1800+, 512 mb ddr ram .. and some other stuff ...

when i try to run x, it all hangs. Gray/black screen, and only power down helps..

i have installed the latest glx and kernel drivers for nvidia (from debian.org, .dbs packages) and recompiled my kernel (2.5.0). I am using xfree86 4.2.0, but it still doesnt work... in my log i get this:


(II) NVIDIA: NVIDIA driver for: RIVA TNT, RIVA TNT2, Vanta,
RIVA TNT2 Ultra, RIVA TNT2 Model 64, Aladdin TNT2, GeForce 256,
GeForce DDR, Quadro, GeForce2 MX/MX 400, GeForce2 MX 100/200,
Quadro2 MXR/EX/Go, GeForce2 Integrated GPU, GeForce2 GTS,
GeForce2 Ti, GeForce2 Ultra, Quadro2 Pro, GeForce4 MX 460,
GeForce4 MX 440, GeForce4 MX 420, Quadro4 500/550 XGL,
Quadro4 200/400 NVS, GeForce3, GeForce3 Ti 200, GeForce3 Ti 500,
Quadro DCC, GeForce4 Ti 4600, GeForce4 Ti 4400, GeForce4 Ti 4200,
Quadro4 900 XGL, Quadro4 750 XGL, Quadro4 700 XGL, GeForce2 Go,
GeForce4 440 Go, GeForce4 420 Go, GeForce4 420 Go 32M,
GeForce4 440 Go 64M, Quadro4 500 GoGL
(II) Primary Device is: PCI 02:00:0
(--) Assigning device section with no busID to primary device
(--) Chipset GeForce2 Integrated GPU found
(II) Setting vga for screen 0.
(II) Loading sub module "vgahw"
(II) LoadModule: "vgahw"
(II) Loading /usr/X11R6/lib/modules/libvgahw.a
(II) Module vgahw: vendor="The XFree86 Project"
compiled for 4.2.0, module version = 0.1.0
ABI class: XFree86 Video Driver, version 0.5
(**) NVIDIA(0): Depth 24, (--) framebuffer bpp 32
(==) NVIDIA(0): RGB weight 888
(==) NVIDIA(0): Default visual is TrueColor
(==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
(--) NVIDIA(0): Linear framebuffer at 0xF0000000
(--) NVIDIA(0): MMIO registers at 0xEC000000
(EE) NVIDIA(0): Failed to initialize the NVdriver kernel module!
(EE) NVIDIA(0): *** Aborting ***
(II) UnloadModule: "nvidia"
(II) UnloadModule: "vgahw"
(II) Unloading /usr/X11R6/lib/modules/libvgahw.a
(EE) Screen(s) found, but none have a usable configuration.

Fatal server error:
no screens found

When reporting a problem related to a server crash, please send
the full server output, not just the last messages.
This can be found in the log file "/var/log/XFree86.0.log".
Please report problems to xfree86@xfree86.org.


It seems like x frinds my card, and is supporting it. But somehow he doesnt like my NVdriver kernel module.
I dont know why not when the kernel installation was successfully....

any ideas?

.urak.

Klaus-P 09-26-02 05:31 AM

" ... EE) Screen(s) found, but none have a usable configuration..."

Errors of that kind occur when f.i. the NVdriver has a wrong ownership
all nvidia stuff must have root ownership. Set it to permission mode 644.
Then run a depmod and reboot your computer.
The system BIOS has nothing to do with that.
Perhaps this will help.

nforceuser 09-26-02 12:08 PM

Unfortunately that's not the issue with my system. Both can actually start an x server manually. (My log file actually shows no errors and is indentical to the working system 1. If I remember right the last line refers to the mouse on both.) The problem is when startx tries to start kde. I just find it very strange that the system would hang when APM was started on bootup for system 2, but works fine for system 1. On nearly identical systems the bios seems to be the likely culprit for at least that problem. I'm assuming KDE may also make use of APM features, and it would follow that this may cause problems on system 2.

I'm also considering installing ACPI for Linux support.

The Grub error #28 ("Selected item does not fit into memory.") is also strange, as it only affected system 2 as well. I just did some Google searching and found that it seems to affect Asus A7N series systems (2 results at least).

Edit: I thought I'd mention startx caused system 2 to hang before the nVidia drivers were ever installed. (I've installed RH 7.3 5 times now, so I've tried a lot of different combinations.)

It really bugs me that two identical (except for video card and bios revision) systems act so differently.

nforceuser 09-27-02 02:30 PM

Ok, here's an update. I switched all the components around so the two systems were TOTALLY IDENTICAL except one board had bios 1.001E, the other has 1.03. The 1.001E system works, 1.03 freezes.

I'm now nearly 100% certain this is a bios issue!

I see there are 2 A7N266-VM users with the same problem.

bwkaz 09-27-02 02:52 PM

When the 1.03 system "freezes", does Ctrl-Alt-Backspace get you back to a console?

The "gray screen, black X-shaped thing in the middle" sounds like what X looks like when it doesn't load a window manager. If you just run X, you'll probably get the same thing. If Ctrl-Alt-Bksp works, then you should be able to get a working system by creating a .xinitrc file in your home directory. If you use KDE, then the file should consist of exec startkde by itself. If you use Gnome, then it should be exec gnome-session. If all else fails, you can try this:

Code:

twm &
xterm -geometry 100x66-0+0 &
xterm -geometry 20x20+0+0 -name login

Exit the xterm named "login" to get out of X.

nforceuser 09-27-02 03:16 PM

It's a total system freeze, cntl-alt-backspace doesn't work. Even telnet sessions into the box to try to diagnose the problem stop responding.

I should also mention the 1.03 bios system reported a Grub error 28 ("Selected item does not fit into memory") where the 1.001E system did not, grub worked fine on that system. (I used dd to extract the first 512 bytes of the Linux /boot drive and used the NTloader to load linux by linking to the extracted file which I placed on the Windows drive and adding a line to the boot.ini file under Win2k.) I've seen this issue mentioned on this forum elsewhere.


All times are GMT -5. The time now is 11:33 AM.

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