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

Newegg Daily Deals

Reply
 
Thread Tools
Old 07-21-04, 02:22 AM   #181
tmarble
Registered User
 
Join Date: Jul 2004
Posts: 4
Default Re: 1.0-6106 for X86-32 thread (debian on hp7260us)

Quote:
Originally Posted by zander
I have just submitted this information to linux-bugs@nvidia.com

I just purchased a HP zd7260us laptop and I am really hoping
to get X running on this machine.

I have followed the basic advice of the README as well as
read information in the nvNews forum. As I am running debian
(unstable) the following two threads have been most helpful:
1) 1.0-6106 for X86-32 thread
2) new 6101 drivers crashing 2 different debian sid systems - same error

I suspect that the problem may simply involve not having
the correct modelines. If it turns out to be more complex
I am more than happy to provide additional debugging details
and/or try different software combinations.

Here's what I have done:

1. Upgraded the BIOS from the HP website:
http://h10025.www1.hp.com/ewfrf/wc/s...tem=hb-22160-1
The laptop shipped with version F.30, this
update is for version F.33

2. Written down some version information from Windows
BIOS: 4.36.20.31.B1
TV Encoder Type: NVIDIA integrated
IRQ: 16
Windows Drivers Version: 6.14.10.5405
Screen refresh rate: 60 Hz

3. Installed debian unstable 2.6.7 with a custom built
kernel and (with the automatically built) matching
nVidia drivers 6106. There has been some discussion of
things changing from 2.6.6 to 2.6.7 ???
see attachment: config-2.6.7-tm3.txt

4. Booted the kernel in uniprocessor mode (not SMP)
and with the noapic option.
see attachment: dmesg.txt

5. Tested agp using the testgart program listed at
http://dri.sourceforge.net/res/
see attachment: testgart.txt

6. Insured that the usb 2.0 driver (ehci_ucd) was not loaded
as some have suggested that has caused a timing problem.

7. Insured that the agpgart driver was not loaded
to avoid any problems here (NOTE: configuration is
set to no agp with Option "NvAgp" "0")

8. Tried X with the vesa driver and that works at
dimensions: 1280x1024 pixels (325x260 millimeters)
resolution: 100x100 dots per inch
depths (7): 24, 1, 4, 8, 15, 16, 32

9. Tried the nv driver, but it doesn't work as this
GPU is too new to be supported.

10. Configured X very conservatively using a minimum number
of loaded X modules and no GLX.
(as I didn't have a previous installation of the debian
drivers on this machine I did not have any problem with
nvidia-glx installation)
see attachment: XF86Config-4.txt

11. Started X with the verbose option: startx -- -logverbose 5
see attachment: XFree86.0.log.txt
NOTE: shouldn't this be 96 like in windows (or 100)?
(==) NV(0): DPI set to (75, 75)

12. The display stays blank, I do not get the Gnome startup sound
and then the machine is hung. Ctrl-Alt-Backspace does not
exit the X session. Ctrl-F2 does not take me to another
text window. Ctrl-Alt-Delete does not reboot. The only
solution is power cycling.

13. I have included other system information:
dmesg.txt
nvidia.txt (information from /proc/driver/nvidia after insertion)

Thanks in advance,

--Tom
Attached Files
File Type: txt XF86Config-4.txt (10.0 KB, 272 views)
File Type: txt XFree86.0.log.txt (29.5 KB, 244 views)
File Type: txt config-2.6.7-tm3.txt (44.2 KB, 257 views)
File Type: txt dmesg.txt (13.0 KB, 220 views)
File Type: txt nvidia.txt (927 Bytes, 255 views)
tmarble is offline   Reply With Quote
Old 07-21-04, 09:46 AM   #182
wmontez
Registered User
 
Join Date: Jul 2004
Posts: 2
Default Re: 1.0-6106 for X86-32 thread

Have installed the 6106 nvidia drivers on Mandrake 10.0 kernel 2.6-13 with GeForce4 440MX AGP 8x and it works fine except for two problems: First is with video playback (avi, dvd) which will completely freeze the system after about one or two minutes forcing a reset. Second is with the GL screensavers, specifically Euphoria. All goes well in normal mode but when I try to change to other configuration options the screen goes almost blank. In random mode it will just lock the whole system also. Other GL screensavers run without problems.
A check using nv_check tells me NVdriver is not correctly loaded. Seems the system is using nvidia_drv.o but not loading its counterpart nvidia.ko neither nvidia.o though the first was normally installed.
What I´ve done so far: Reinstalled the driver, tried to load nvidia.ko and nvidia.o via insmod and script without success (see attached XFree86.0 log), added "alias char-major 195 nvidia* NVdriver" to modules.conf.
I´ve upgraded ther kernel from 2.6.7 including the sources, via rpm and already checked that the current versions in use are the upgraded ones. Any help is welcome!!!

Regards

wmontez
Attached Files
File Type: txt nvidia-installer.txt (1.9 KB, 207 views)
File Type: txt XF86Config-4.txt (2.8 KB, 229 views)
File Type: txt XFree86.0.txt (26.3 KB, 188 views)
wmontez is offline   Reply With Quote
Old 07-22-04, 07:29 PM   #183
Antonio P.
5700 Ultra
 
Join Date: Jan 2004
Location: PARIS
Posts: 12
Default Re: 1.0-6106 for X86-32 thread

Hi all,
I'm trying to setup the TVout and I'm having some trouble.
First, I've to say that I tried a long time with the previous version of the driver. I never had any output on my TV with this driver. I've a 5700 Ultra.

Now, I'm trying with the 6106. I've better results but it's still not satisfying.
First, my DFP seems to be the *second* monitor, while my tv appears to be the first monitor.
(It's bizarre, because on ms-windows, the TV is the second one)
So the specs gived for the monitor section are taken for the TV while the "secondmonitor" options are taken for the DFP.
My problem is that my TV is not always connected to my pc, then if I unplug it, the DFP will be the first monitor. Then, how can I give the specs of the monitors 1 and 2 ?

To solve this, I'm trying to configure twinview with the new device options where you can specify the monitor (as described in the doc) :
Option "HorizSync" "TV-0: 30-50, DFP-0: 30-80"
Option "VertRefresh" "TV-0: 60, DFP-0: 50-75"

Well, these options seem just not to work.

I had the following lines in the log :
(**) NVIDIA(0): Option "HorizSync" "TV-0: 30-50, DFP-0: 30-80"
(**) NVIDIA(0): Option "VertRefresh" "TV-0: 60, DFP-0: 50-75"

this is ok, but after :

(WW) NVIDIA(0): Unable to parse range in HorizSync string
(WW) NVIDIA(0): "TV-0:30-50,DFP-0:30-80"; ignoring
(WW) NVIDIA(0): Unable to parse range in VertRefresh string
(WW) NVIDIA(0): "TV-0:60,DFP-0:50-75"; ignoring

If someone can light my candle...
Antonio P. is offline   Reply With Quote
Old 07-23-04, 05:45 PM   #184
MadEgg
Registered User
 
Join Date: Apr 2003
Posts: 88
Default Re: 1.0-6106 for X86-32 thread

Can't you fix that using the "ConnectedMonitor" option, and then using "SecondMonitorHorizSync" and "SecondMonitorVertRefresh" to set these settings up right?

I thought if you made it:

Option "ConnectedMonitor" "DFP-0, TV-0"

The DFP will always be the first monitor. I never had this problem so I can't confirm it though...
__________________
Proud owner of GF 6800 Ultra AGP and GF 6600 AGP
MadEgg is offline   Reply With Quote
Old 07-23-04, 07:17 PM   #185
Antonio P.
5700 Ultra
 
Join Date: Jan 2004
Location: PARIS
Posts: 12
Default Re: 1.0-6106 for X86-32 thread

That's what I already did.
I made a lot of changes after the trial with the not recognized syntax.
I joined the last XF86Config and the log. The point is, with this config file which seems correct to me, I have a display on the TV but not on the DFP. When I switch to console mode, the two displays are unreadable.
Attached Files
File Type: txt XF86Config-4_TV3.txt (3.8 KB, 241 views)
File Type: txt XF2.log2.txt (31.4 KB, 259 views)
Antonio P. is offline   Reply With Quote
Old 07-24-04, 12:08 PM   #186
graffic
nobody nogroup
 
Join Date: Jul 2004
Posts: 3
Exclamation Re: 1.0-6106 for X86-32 thread

Quote:
Originally Posted by erek
alright, i want to request a change:
I have a BFG GeForce 6800 GT OC the PCI (device) ID is 0x0045 and is not recognized by the 6106 drivers:
Code:
OpenGL vendor string: NVIDIA Corporation
OpenGL renderer string: unknown board/AGP/SSE/3DNOW!
OpenGL version string: 1.5.1 NVIDIA 61.06
also i'd like to know specifically not based on speculation why my core temperatures differ by ~20C in Linux from Win32, my gpu idles at ~53C in win32, but a constant 70+ C in Linux, "ripperda" speculates that because the drivers /statically/ start engines such as "overlay" up front instead of when the engine is needed /dynamically/ (win32) that the heat could be from that, so could you PLEASE take a more win32 approach to this, even though 70+ C is WELL within specifications of operation, i still find it quite WARM for this type of weather in my area which is only going to get a lot warmer...
My Point of View Geforce 6800 GT idles at 66ºC in windows xp, but in linux at 78ºC. When running 3D ap's: 73ºC in win and 86ºC in linux. My card is really hot but it's a hot summer here
graffic is offline   Reply With Quote
Old 07-28-04, 02:10 PM   #187
NuMeR07
Registered User
 
Join Date: Nov 2003
Posts: 3
Default Re: 1.0-6106 for X86-32 thread

About the problem reported often : glxinfos reporting "direct rendering :No", one trick that worked for me :

- the install of the drivers was ok, I spent hours looking for solutions....

--> till I tried to check ldd /usr/X11R6/bin/glxinfo
===> the libGL it was looking for was in /usr/lib/opengl/xfree/lib (or something),
but NVIDIA GL ara in /usr/lib. After a Sherlock Holmes job following symlinks, I understand that it linked toward /usr/X11R6/lib

=> solution, I went in /etc/ld.so.conf to put /usr/lib BEFORE /usr/X11R6/lib

--> ran ldconfig

-> glxinfo no more complains !

Hope it helps
NuMeR07 is offline   Reply With Quote
Old 08-04-04, 03:05 PM   #188
sobers_2002
Registered User
 
Join Date: Aug 2004
Posts: 4
Default Re: 1.0-6106 for X86-32 thread

hi all

this is something i couldn't understand with the new drivers.......everything works fine but the glxgears shows a slowdown in performance..........i have nvidia nforce2 motherboard and onboard graphics geforce2.........earlier fps was ~1100 but now its just half ~560 with glxgears........ i just don't get it. i have agpart enabled as before......also framebuffer hasn;t worked eversince with the new drivers.....using kerel 2.6.8-r2-love3 right now in gentoo.

thnx in adv for help
Saurabh
sobers_2002 is offline   Reply With Quote

Old 08-05-04, 01:41 AM   #189
large_satchell
Registered User
 
Join Date: Aug 2004
Posts: 10
Thumbs down Re: 1.0-6106 for X86-32 thread

Why will my GeForce FX 5900 XT not work in AGP 8X mode ??
I have a SiS 648 AGP bridge that is fully capable of doing 8x, yet the /proc/whatever says that I am in 4X mode....

I'm using kernel 2.6.7. I tried using both agpgart and NVGART and I always get 4x and not 8x..... WHAT GIVES? (BTW, With agpgart I get 5700 fps [with glxgears], and with NVGART I get 5100 fps...)

also, when using agpgart, I get this message in the syslog:

Aug 5 00:12:10 titan kernel: Linux agpgart interface v0.100 (c) Dave Jones
Aug 5 00:12:10 titan kernel: agpgart: Detected SiS 648 chipset
Aug 5 00:12:10 titan kernel: agpgart: Maximum main memory to use for agp memory: 439M
Aug 5 00:12:10 titan kernel: agpgart: AGP aperture is 64M @ 0xe8000000
Aug 5 00:14:55 titan kernel: agpgart: Found an AGP 3.0 compliant device at 0000:00:00.0.
Aug 5 00:14:55 titan kernel: agpgart: Putting AGP V3 device at 0000:00:00.0 into 4x mode
Aug 5 00:14:55 titan kernel: agpgart: SiS delay workaround: giving bridge time to recover.
Aug 5 00:14:56 titan kernel: agpgart: Putting AGP V3 device at 0000:01:00.0 into 4x mode

heres some more info:

[root@titan agp]# more *
::::::::::::::
card
::::::::::::::
Fast Writes: Supported
SBA: Supported
AGP Rates: 8x 4x
Registers: 0x1f000e1b:0x1f004301
::::::::::::::
host-bridge
::::::::::::::
Host Bridge: Silicon Integrated Systems [SiS] SiS 645xx
Fast Writes: Not Supported
SBA: Supported
AGP Rates: 4x
Registers: 0x1f004e09:0x00000f01
::::::::::::::
status
::::::::::::::
Status: Enabled
Driver: NVIDIA
AGP Rate: 4x
Fast Writes: Disabled
SBA: Enabled
[root@titan agp]#

Help me fix this please.
large_satchell is offline   Reply With Quote
Old 08-05-04, 02:49 AM   #190
PrakashP
gentoo ~x86_64 user
 
PrakashP's Avatar
 
Join Date: Jul 2004
Location: Germania
Posts: 213
Default Re: 1.0-6106 for X86-32 thread

Your chipset doesn't support 8x, according to its status.
PrakashP is offline   Reply With Quote
Old 08-09-04, 08:53 AM   #191
shirsch
Registered User
 
Join Date: Jul 2004
Posts: 2
Default Re: 1.0-6106 for X86-32 thread

I am seeing this as well, with totally different hardware and on a SuSE 9.1 system. Did you ever get a solution?
shirsch is offline   Reply With Quote
Old 08-24-04, 12:00 PM   #192
TheSauceman
Registered User
 
Join Date: Aug 2004
Posts: 1
Default Re: 1.0-6106 for X86-32 thread

I've been trying to get the Nvidia drivers working on an HP zd7260us laptop on Fedora Core 2 without success. All attempts to use the nvidia driver in place of nv, results in the black screen of death. No recovery, just power reset. Using nv "works", but without OpenGL, which is central to my job.

I've tried earlier drivers, earlier kernels, to no avail.

Has anyone successfully installed the new drivers on a zd7260us or similar in any flavor of linux? If so, which distro?

Thanks in advance!
TheSauceman 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 07:30 PM.


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