Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230
Nvidia driver - Debian way, custom (or not) kernel.
- Jackiebrown
- Posts: 1246
- Joined: 2007-01-02 04:46
- Location: San Antonio, TX
Reboot.estus wrote:i cant use terminal and ctrl+alt+f2 dont work
At the boot prompt add this to the end:
Code: Select all
single
Code: Select all
update-rc.d -f gdm remove
Code: Select all
nano /etc/X11/xorg.conf
Code: Select all
init 5
Code: Select all
startx
Code: Select all
dpkg -l|grep nvidia
Code: Select all
apt-get install nvidia-xconfig
nvidia-xconfig
Debian Sid Laptops:
AMD Athlon(tm) 64 X2 Dual-Core Processor TK-55 / 1.5G
Intel(R) Pentium(R) Dual CPU T2390 @ 1.86GHz / 3G
AMD Athlon(tm) 64 X2 Dual-Core Processor TK-55 / 1.5G
Intel(R) Pentium(R) Dual CPU T2390 @ 1.86GHz / 3G
I just installed Lenny from the Lenny beta 2 install CD. I followed the tutorial from the Debian wiki:
http://wiki.debian.org/NvidiaGraphicsDrivers
I also had to install nvidia-xconfig, mesa-utils, and nvidia-settings. I then ran "nvidia-xconfig".
I also apparently ran into this bug mentioned in the Debian wiki article:
This has been a very informative thread. I learned a lot.
http://wiki.debian.org/NvidiaGraphicsDrivers
Code: Select all
# apt-get install module-assistant gcc nvidia-kernel-common
# m-a update
# m-a prepare
# m-a auto-install nvidia
I also apparently ran into this bug mentioned in the Debian wiki article:
Well, I did not read that far in the wiki untill I had already installed the driver. Fortunately, it all went well. The driver runs well.BUG: There is a bug when you 'apt-get install nvidia-kernel-common' you are asked to insall 40+ packages. (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=303521)
Code: Select all
tom@debian:/data/sauerbraten$ cd /data/
tom@debian:/data$ glxinfo | grep -i direct
direct rendering: Yes
tom@debian:/data$
Doing this HowTo in fresh Lenny install...
I'm following the HowTo in hopes of enabling 3d acceleration on my fresh Lenny net install from the current weekly build image.
The first hitch I ran into was that, by default, only the main repositories are included in sources.list. mzilikazi, you might want to mention that contrib and/or non-free need to be added before you can install the nvidia kernel source and other pakages, for the benefit of others working on a fresh install?
The next hitch is that, with the newfangled stripped down version of xorg.conf, there is no 'nv' or 'vesa' driver line to change to 'nvidia.' So I'm off to find a representative xorg.conf "Device" entry...
In the Debian Wiki, nVidia Graphics Drivers page,the section on how to configure X to use the nvidia driver the manual way says:
Adding a little more detail to this part of the HowTo would be helpful, I think.
Now to save, then restart gnome... It says there's a display manager running on display 0 and do I want to go ahead anyway... yes... some flickering and transient blackness, a grey screen with the "X" in the middle, and another dialog warning that this session was started on display 1...
Finally, I've managed to log into a new session which seems to be functional.
Lets try:
Okay, Debian's package database tells me that glxinfo is in package mesa-utils... Load Synaptic and install it.
Is this another thing to add to the HowTo?
Try again:
Wahoo! YES!
Allright, now to try nvidia-settings... Opens a very pretty looking GUI configuration window showing all sorts of stuff. It seems to have identified my GeForce 8600 GT; the thermal monitor is even working. This is good.
Okay, now it's off to see if I can break it on something that uses 3d... Hmmm... Synaptic... Neverball...
Lordy, it works!!! I've never seen how the game's supposed to behave before!
Okay then, the upgrade to Lenny is paying off!
Thanks to mzilikazi for maintaining this HowTo, and anarchyinc666 for his index of HowTo's.
Now to bring over some more of my configuration stuff and data!! Oh, and maybe install a few more 3d games!
Mike D.
The first hitch I ran into was that, by default, only the main repositories are included in sources.list. mzilikazi, you might want to mention that contrib and/or non-free need to be added before you can install the nvidia kernel source and other pakages, for the benefit of others working on a fresh install?
The next hitch is that, with the newfangled stripped down version of xorg.conf, there is no 'nv' or 'vesa' driver line to change to 'nvidia.' So I'm off to find a representative xorg.conf "Device" entry...
In the Debian Wiki, nVidia Graphics Drivers page,the section on how to configure X to use the nvidia driver the manual way says:
Well, I don't have a "Module" section, but I added the Driver line so my Device section now looks like:Edit /etc/X11/xorg.conf:
* If it contains a "Module" section, be sure that it contains a line:
... and remove or comment out (prepend with a #) any lines that refer to the "dri" or "GLCore" modules.Code: Select all
Load "glx"
* In the "Device" section for your video card, change the driver (normally nv or vesa) to nvidia. If the driver is not defined, add the line
Code: Select all
Driver "nvidia"
Code: Select all
Section "Device"
Identifier "Configured Video Device"
Driver "nvidia"
EndSection
Now to save, then restart gnome... It says there's a display manager running on display 0 and do I want to go ahead anyway... yes... some flickering and transient blackness, a grey screen with the "X" in the middle, and another dialog warning that this session was started on display 1...
Finally, I've managed to log into a new session which seems to be functional.
Lets try:
Code: Select all
# glxinfo|grep rendering
bash: glxinfo: command not found
Is this another thing to add to the HowTo?
Try again:
Code: Select all
# glxinfo|grep rendering
direct rendering: Yes
Allright, now to try nvidia-settings... Opens a very pretty looking GUI configuration window showing all sorts of stuff. It seems to have identified my GeForce 8600 GT; the thermal monitor is even working. This is good.
Okay, now it's off to see if I can break it on something that uses 3d... Hmmm... Synaptic... Neverball...
Lordy, it works!!! I've never seen how the game's supposed to behave before!
Okay then, the upgrade to Lenny is paying off!
Thanks to mzilikazi for maintaining this HowTo, and anarchyinc666 for his index of HowTo's.
Now to bring over some more of my configuration stuff and data!! Oh, and maybe install a few more 3d games!
Mike D.
-
- Posts: 47
- Joined: 2007-09-01 22:02
Hello,
I just installed a kernel 2.6.27.8 #1 SMP Mon Dec 29 15:13:50 CET 2008 i686 GNU/Linux
method debian & method lavene, it's working so great !
So the nvidia isnt working via the debian way.
I get those error messages :
:/var/cache/modass$ for each in nvid* ; do echo "$each" >> /tmp/nvidialog ; echo "------------" >> /tmp/nvidialog ; cat "$each" >> /tmp/nvidialog ; done ; gedit /tmp/nvidialog
http://pastebin.com/f23dfe3ab
EDIT UPDATE, some success in some way :
Hi !
Succcess.
I just installed thislinux-image-2.6.26-bpo.1-686 from the backport:
and m-a a-i gspca failed but :
http://mxhaard.free.fr/spca50x/Download ... 224.tar.gz
succeeded !
It seems that the method to compile kernel we did (lavene method) is far different to the kernel in the backport (made by linux professionals)
So, what is it ?
The deb is here :
http://www.backports.org/debian/pool/main/l/linux-2.6/
those bpo kernels
How to compile kernel well, like very debian way/professionals ?
at least for the webcam running lol
I just installed a kernel 2.6.27.8 #1 SMP Mon Dec 29 15:13:50 CET 2008 i686 GNU/Linux
method debian & method lavene, it's working so great !
So the nvidia isnt working via the debian way.
Code: Select all
aptitude install module assistant build-essential
m-a update
m-a prepare
m-a auto-install nvidia -f
modprobe nvidia
:/var/cache/modass$ for each in nvid* ; do echo "$each" >> /tmp/nvidialog ; echo "------------" >> /tmp/nvidialog ; cat "$each" >> /tmp/nvidialog ; done ; gedit /tmp/nvidialog
http://pastebin.com/f23dfe3ab
Code: Select all
nvidia-kernel-legacy-source.avail_version
------------
1.0.7184-3nvidia-kernel-source.avail_version
------------
1.0.8776-4nvidia-kernel-source.buildlog.2.6.27.8.1230590616
------------
/usr/bin/make -f debian/rules clean
make[1]: Entering directory `/usr/src/modules/nvidia-kernel'
# select which makefile to use.
rm -f /usr/src/modules/nvidia-kernel/nv/Makefile || true
if [ 6 = 6 ]; then \
cd /usr/src/modules/nvidia-kernel/nv ; \
ln -s Makefile.kbuild Makefile ; \
cd .. ; \
if [ 0 = 1 ] ; then \
dpatch apply 04_minion ; \
fi ; \
if [ 0 = 1 ]; then \
dpatch apply 01_sysfs ; \
dpatch status 01_sysfs >patch-stamp ; \
dpatch apply 02_pcialias ; \
dpatch status 02_pcialias >>patch-stamp ; \
fi ; \
fi....
bla bla
...
inter type
make[4]: *** [/usr/src/modules/nvidia-kernel/nv/nv.o] Error 1
make[3]: *** [_module_/usr/src/modules/nvidia-kernel/nv] Error 2
make[3]: Leaving directory `/usr/src/linux-2.6.27.8'
NVIDIA: left KBUILD.
nvidia.ko failed to build!
make[2]: *** [mdl] Error 1
make[2]: Leaving directory `/usr/src/modules/nvidia-kernel/nv'
make[1]: *** [build-stamp] Error 2
make[1]: Leaving directory `/usr/src/modules/nvidia-kernel'
make: *** [kdist_image] Error 2
nvidia-kernel-source.cur_version
------------
1.0.8776-4
EDIT UPDATE, some success in some way :
Hi !
Succcess.
I just installed thislinux-image-2.6.26-bpo.1-686 from the backport:
Code: Select all
i libsilc-1.1-2 1.1.7-1~bpo40+1 SILC library (silc-toolkit)
ii linux-headers-2.6.26-bpo.1-686 2.6.26-12~bpo40+1 Header files for Linux 2.6.26-bpo.1-686
ii linux-headers-2.6.26-bpo.1-common 2.6.26-12~bpo40+1 Common header files for Linux 2.6.26-bpo.1
ii linux-image-2.6.26-bpo.1-686 2.6.26-12~bpo40+1 Linux 2.6.26 image on PPro/Celeron/PII/PIII/
ii
http://mxhaard.free.fr/spca50x/Download ... 224.tar.gz
succeeded !
It seems that the method to compile kernel we did (lavene method) is far different to the kernel in the backport (made by linux professionals)
So, what is it ?
The deb is here :
http://www.backports.org/debian/pool/main/l/linux-2.6/
those bpo kernels
How to compile kernel well, like very debian way/professionals ?
at least for the webcam running lol
Re: Nvidia driver - Debian way, custom (or not) kernel.
I believe that that is a folder you are trying to write to. Your source list is in /etc/apt/sources.listFibonacci wrote:Well first problem.. I am logged in as a root, but I can not add the source.list. And there seems to be no list atm. I just installed Debian, quite new to Linux, first time on Debian.. Maybe I am doing something wrong. Please try to explain me..mzilikazi wrote:
Standard method
Add contrib and non-free to /etc/apt/sources.list.d/sources.listCode: Select all
deb ftp://ftp.us.debian.org/debian/ sid main contrib non-free deb ftp://ftp.us.debian.org/debian/ testing main contrib non-free deb ftp://ftp.us.debian.org/debian/ stable main contrib non-free
Here is screenshot
So I logged in as a root, wrote in editor the above quite from the creator of this thread so that I can actually get smth but it won't allow me to add this list of sources.
Edit: Link
Cheers
You are very fast Bro.Tiag
As I managed to delete that post, but you were already into replay to it. I have solved that issue but thanks a lot anyways. I was mixing folders, you were right
But here is the thing, I managed to install Nvidia drivers, properly or so I think, because I got Nvidia splash screen. Is that enough to prove installation went well?
What I would like to point out is the following. Maybe for begginers like me, there should be a line saying that when we change xorg.conf the PC Bus, we should check ourselves before for our specific graphic card. Because what happened is I used the one supplied in this topic. And it didn't work, the X wouldn't start.. went on internet through phone and found PCI Bus ID for my graphic card, edited xorg.conf and it seems that it worked out.
Eitherway, the last thing.. I suppose this line is supposed to check if installation went well..
But all I get as an answer is..
or logged as root
Hence, my question is, beside Nvidia splash screen while booting, how else to check 3D works, as I havn't noticed any difference from what my desktop looked like before?
As I managed to delete that post, but you were already into replay to it. I have solved that issue but thanks a lot anyways. I was mixing folders, you were right
But here is the thing, I managed to install Nvidia drivers, properly or so I think, because I got Nvidia splash screen. Is that enough to prove installation went well?
What I would like to point out is the following. Maybe for begginers like me, there should be a line saying that when we change xorg.conf the PC Bus, we should check ourselves before for our specific graphic card. Because what happened is I used the one supplied in this topic. And it didn't work, the X wouldn't start.. went on internet through phone and found PCI Bus ID for my graphic card, edited xorg.conf and it seems that it worked out.
Eitherway, the last thing.. I suppose this line is supposed to check if installation went well..
Code: Select all
glxinfo|grep rendering
Code: Select all
bash: glxinfo: command not found
Code: Select all
-su: glxinfo: command not found
Hence, my question is, beside Nvidia splash screen while booting, how else to check 3D works, as I havn't noticed any difference from what my desktop looked like before?
- Telemachus
- Posts: 4574
- Joined: 2006-12-25 15:53
- Been thanked: 2 times
You have to install a package called mesa-utils in order to have glxinfo.Fibonacci wrote:Either way, the last thing.. I suppose this line is supposed to check if installation went well..But all I get as an answer is..Code: Select all
glxinfo|grep rendering
or logged as rootCode: Select all
bash: glxinfo: command not found
Code: Select all
-su: glxinfo: command not found
"We have not been faced with the need to satisfy someone else's requirements, and for this freedom we are grateful."
Dennis Ritchie and Ken Thompson, The UNIX Time-Sharing System
Dennis Ritchie and Ken Thompson, The UNIX Time-Sharing System
Not to be a pain in the arse, and maybe I missed it in the OP, but installing mesa-utils is not included there, yet it is suggest to do the the following.Telemachus wrote:You have to install a package called mesa-utils in order to have glxinfo.Fibonacci wrote:Either way, the last thing.. I suppose this line is supposed to check if installation went well..But all I get as an answer is..Code: Select all
glxinfo|grep rendering
or logged as rootCode: Select all
bash: glxinfo: command not found
Code: Select all
-su: glxinfo: command not found
Cheersmzilikazi wrote:
After you've gotten the driver to build and Xwindows working you can do a quick test for Direct Rendering:Also of interest is nvidia-settingsCode: Select all
glxinfo|grep rendering
-
- Posts: 349
- Joined: 2008-03-31 15:55
- Location: Still trying to figure out for what this field is...
A simple "apt-get install nvidia-glx" worked for me on a fresh testing install with non-free and contrib added to the sources.list. After install nvidia-glx all I had to do was add the appropriate driver line to xorg.conf and restart x. Maybe I'm missing something but it seems like you guys are making the process harder than it really is.
And now try that on a custom kernel with the latest nvidia driver. Yes, you are missing plenty.monacelli wrote:A simple "apt-get install nvidia-glx" worked for me on a fresh testing install with non-free and contrib added to the sources.list. After install nvidia-glx all I had to do was add the appropriate driver line to xorg.conf and restart x. Maybe I'm missing something but it seems like you guys are making the process harder than it really is.
Wisdom from my inbox: "do not mock at your pottenocy"
Are you saying that you didn't use module-assistant? What does this say:monacelli wrote:A simple "apt-get install nvidia-glx" worked for me on a fresh testing install with non-free and contrib added to the sources.list. After install nvidia-glx all I had to do was add the appropriate driver line to xorg.conf and restart x. Maybe I'm missing something but it seems like you guys are making the process harder than it really is.
Code: Select all
lsmod|grep nv
Debian Sid Laptops:
AMD Athlon(tm) 64 X2 Dual-Core Processor TK-55 / 1.5G
Intel(R) Pentium(R) Dual CPU T2390 @ 1.86GHz / 3G
AMD Athlon(tm) 64 X2 Dual-Core Processor TK-55 / 1.5G
Intel(R) Pentium(R) Dual CPU T2390 @ 1.86GHz / 3G
Indeed, I didn't use module-assistant.mzilikazi wrote:Are you saying that you didn't use module-assistant? What does this say:Code: Select all
lsmod|grep nv
lsmod|grep nv returns:
Code: Select all
nvidia 7086148 26
i2c_core 19828 2 nvidia,i2c_i801
agpgart 28776 2 nvidia,intel_agp
OK then how about this:monacelli wrote:Indeed, I didn't use module-assistant.mzilikazi wrote:Are you saying that you didn't use module-assistant? What does this say:Code: Select all
lsmod|grep nv
lsmod|grep nv returns:Code: Select all
nvidia 7086148 26 i2c_core 19828 2 nvidia,i2c_i801 agpgart 28776 2 nvidia,intel_agp
Code: Select all
modinfo nvidia
Debian Sid Laptops:
AMD Athlon(tm) 64 X2 Dual-Core Processor TK-55 / 1.5G
Intel(R) Pentium(R) Dual CPU T2390 @ 1.86GHz / 3G
AMD Athlon(tm) 64 X2 Dual-Core Processor TK-55 / 1.5G
Intel(R) Pentium(R) Dual CPU T2390 @ 1.86GHz / 3G
I can confirm that you can (also) install it by just using aptitude, did it recently on my laptop.
Code: Select all
# modinfo nvidia
filename: /lib/modules/2.6.26-1-686-bigmem/nvidia/nvidia.ko
license: NVIDIA
alias: char-major-195-*
alias: pci:v000010DEd*sv*sd*bc03sc02i00*
alias: pci:v000010DEd*sv*sd*bc03sc00i00*
depends: agpgart,i2c-core
vermagic: 2.6.26-1-686-bigmem SMP mod_unload modversions 686
parm: NVreg_EnableVia4x:int
parm: NVreg_EnableALiAGP:int
parm: NVreg_ReqAGPRate:int
parm: NVreg_EnableAGPSBA:int
parm: NVreg_EnableAGPFW:int
parm: NVreg_Mobile:int
parm: NVreg_ResmanDebugLevel:int
parm: NVreg_RmLogonRC:int
parm: NVreg_ModifyDeviceFiles:int
parm: NVreg_DeviceFileUID:int
parm: NVreg_DeviceFileGID:int
parm: NVreg_DeviceFileMode:int
parm: NVreg_RemapLimit:int
parm: NVreg_UpdateMemoryTypes:int
parm: NVreg_UseVBios:int
parm: NVreg_RMEdgeIntrCheck:int
parm: NVreg_UsePageAttributeTable:int
parm: NVreg_RegistryDwords:charp
parm: NVreg_NvAGP:int