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

Newegg Daily Deals

Reply
 
Thread Tools
Old 04-11-10, 04:51 PM   #1
infallible
Registered User
 
Join Date: Apr 2010
Posts: 3
Default graphics card fail

I've been having issues with a 9500GT card, and they didn't go away with an upgrade to a GT250. Running Fedora 12 x86_64; issue persists with every attempted version of kernel and nvidia driver.

I am able to use generic drivers, ie vesa or nouveau, but cannot use the nvidia driver. If the driver is properly installed, I must append 'vga=0x0317' to the boot parameters, essentially specifying the card's use as a framebuffer. If I do not, boot hangs during the udev process. With this workaround, I have console login only.

From here, i was able to look at logs, try different kernel parameters, etc. I can't remember where I read the information suggesting it (late night!) but it seemed that the driver module ran, but didn't attatch itself to the hardware. I find these reports especially concerning:

Expansion ROM unassigned and disabled?
Code:
04:00.0 VGA compatible controller: nVidia Corporation G92+ [GeForce GTS 250] (rev a2) (prog-if 00 [VGA controller])
	Subsystem: Giga-byte Technology Device 34cb
	Flags: bus master, fast devsel, latency 0, IRQ 5
	Memory at c1000000 (32-bit, non-prefetchable) [size=16M]
	Memory at d0000000 (64-bit, prefetchable) [size=256M]
	Memory at c2000000 (64-bit, non-prefetchable) [size=32M]
	I/O ports at 4000 [size=128]
	Expansion ROM at <unassigned> [disabled]
	Capabilities: [60] Power Management version 3
	Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
	Capabilities: [78] Express Endpoint, MSI 00
	Capabilities: [100] Virtual Channel
	Capabilities: [128] Power Budgeting <?>
	Capabilities: [600] Vendor Specific Information: ID=0001 Rev=1 Len=024 <?>
	Kernel modules: nouveau, nvidiafb
Not sure what's gong on here...
Code:
Apr 11 15:23:45 rutherford kernel: vgaarb: device added: PCI:0000:04:00.0,decodes=io+mem,owns=io+mem,locks=none
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1000-0x107f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1080-0x10ff has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1400-0x147f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1480-0x14ff has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1800-0x187f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1880-0x18ff has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x2440-0x247f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x2400-0x243f has been reserved
Apr 11 15:23:45 rutherford kernel: pci 0000:04:00.0: BAR 6: can't allocate mem resource [0xe0000000-0xdfffffff]
So, that's where I'm aiming my shots in the dark. I'd appreciate any insight.
Attached Files
File Type: gz nvidia-bug-report.log.gz (28.7 KB, 54 views)
infallible is offline   Reply With Quote
Old 04-12-10, 12:38 AM   #2
Dizzle7677
Registered User
 
Dizzle7677's Avatar
 
Join Date: May 2008
Location: Relativity
Posts: 194
Default Re: graphics card fail

Ask Fedora.
http://fedoraproject.org/wiki/Bugzilla
Dizzle7677 is offline   Reply With Quote
Old 04-12-10, 02:17 AM   #3
gradinaruvasile
Registered User
 
gradinaruvasile's Avatar
 
Join Date: May 2008
Posts: 199
Default Re: graphics card fail

Quote:
Originally Posted by infallible View Post
I've been having issues with a 9500GT card, and they didn't go away with an upgrade to a GT250. Running Fedora 12 x86_64; issue persists with every attempted version of kernel and nvidia driver.

I am able to use generic drivers, ie vesa or nouveau, but cannot use the nvidia driver. If the driver is properly installed, I must append 'vga=0x0317' to the boot parameters, essentially specifying the card's use as a framebuffer. If I do not, boot hangs during the udev process. With this workaround, I have console login only.

From here, i was able to look at logs, try different kernel parameters, etc. I can't remember where I read the information suggesting it (late night!) but it seemed that the driver module ran, but didn't attatch itself to the hardware. I find these reports especially concerning:

Expansion ROM unassigned and disabled?
Code:
04:00.0 VGA compatible controller: nVidia Corporation G92+ [GeForce GTS 250] (rev a2) (prog-if 00 [VGA controller])
	Subsystem: Giga-byte Technology Device 34cb
	Flags: bus master, fast devsel, latency 0, IRQ 5
	Memory at c1000000 (32-bit, non-prefetchable) [size=16M]
	Memory at d0000000 (64-bit, prefetchable) [size=256M]
	Memory at c2000000 (64-bit, non-prefetchable) [size=32M]
	I/O ports at 4000 [size=128]
	Expansion ROM at <unassigned> [disabled]
	Capabilities: [60] Power Management version 3
	Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
	Capabilities: [78] Express Endpoint, MSI 00
	Capabilities: [100] Virtual Channel
	Capabilities: [128] Power Budgeting <?>
	Capabilities: [600] Vendor Specific Information: ID=0001 Rev=1 Len=024 <?>
	Kernel modules: nouveau, nvidiafb
Not sure what's gong on here...
Code:
Apr 11 15:23:45 rutherford kernel: vgaarb: device added: PCI:0000:04:00.0,decodes=io+mem,owns=io+mem,locks=none
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1000-0x107f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1080-0x10ff has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1400-0x147f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1480-0x14ff has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1800-0x187f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x1880-0x18ff has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x2440-0x247f has been reserved
Apr 11 15:23:45 rutherford kernel: system 00:04: ioport range 0x2400-0x243f has been reserved
Apr 11 15:23:45 rutherford kernel: pci 0000:04:00.0: BAR 6: can't allocate mem resource [0xe0000000-0xdfffffff]
So, that's where I'm aiming my shots in the dark. I'd appreciate any insight.
Try this:

1. Blacklist the nouveau driver:

Edit the /etc/modprobe.d/blacklist.conf file (or its equivalent on fedora if it is) with root rights. The idea is to prevent this kernel module from loading, i found it wont let other modules access the card.
Append to the end:

Code:
blacklist nouveau
Save.

2. Disable kms:

Append as kernel parameter "nomodeset"

Reboot. Log in a tty (text mode), make sure gdm is stopped.

Install nvidia driver, make sure you set up your xorg.conf (or let the installer do it)
Reboot or start gdm.

This worked for me on Debian with the 2.6.33 kernel.
gradinaruvasile is offline   Reply With Quote
Old 04-12-10, 11:11 AM   #4
infallible
Registered User
 
Join Date: Apr 2010
Posts: 3
Default Re: graphics card fail

Quote:
Originally Posted by gradinaruvasile View Post
Try this:

1. Blacklist the nouveau driver:

Edit the /etc/modprobe.d/blacklist.conf file (or its equivalent on fedora if it is) with root rights. The idea is to prevent this kernel module from loading, i found it wont let other modules access the card.
Append to the end:

Code:
blacklist nouveau
Save.

2. Disable kms:

Append as kernel parameter "nomodeset"

Reboot. Log in a tty (text mode), make sure gdm is stopped.

Install nvidia driver, make sure you set up your xorg.conf (or let the installer do it)
Reboot or start gdm.

This worked for me on Debian with the 2.6.33 kernel.
I've blacklisted the nouveau driver, recompiled initramfs with dracut to ensure it does not load, and blacklisted the radeon driver for the onboard chipset.

I've tried a number of kernel parameters and combinations thereof at boot, among them nomodeset, iommu=off, noapic, nolapic, pci=routeirq, pci=alloptionsthatseemedrelevant. In fact, I have a copy of kernel_parameters.txt for each kernel version I've attempted, and tried them in a semi-methodical way.

Under the assumption that my board ran out of addressing space, I disabled, in the BIOS, everything but for USB and networking, powercyled, and booted to a USB live image with various kernel parameters. In each case, the drivers fail to load and the conditions presented in the below {code} blocks persist.

Under the assumption that this is a fedora-specific condition, I am currently downloading live images of other distributions to test in this same method.

Thanks for your attention,

Pete
infallible 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 08:15 PM.


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