Debian :: Building Kernel - Initrd Fail And Nvidia Can't Find Source
Feb 25, 2010
So I'm building a custom kernel cuz I want the fbcondecor patch in my kernel. I use the same .config that 2.6.32 debian kernel package comes with. This kernel runs perfectly. I pass --initrd to make-kpkg when building the package but no initrd is built when I install it so I have to make it using "mkinitramfs -o /boot/initrd.img-2.6.33 2.6.33" to get it to boot. Now when I try to build the nvidia drivers it complains it can't find the source. I did build kernel_headers and installed them also the source is in /usr/src/linux. I also tried to specify the path by passing --kernel-source-path= to the nvidia script but no change. What is going on? I've done this fifty times before and never had any problems. Has there been some changes to how debian kernel packages are built? EDIT: Just thought I'd add some info about the steps I took.
Code:
tar xjvf linux-2.6.33.tar.bz2
ln -s linux-2.6.33 linux
cd linux
patch -p1 < ../fbcondecor-0.9.6-2.6.33-rc7.patch
cp /boot/config-2.6.32-trunk-686 ./.config
make menuconfig Loaded .config then I removed support for maxtorfb, tile blitting and some sirrusfb thing, nothing thats relevant to my system. Changed cpu from Pentium 4 to Core 2 and added framebuffer decor
support from my patch, exited and saved. Then:
[Code]....
EDIT2: I have now tried to build 2.6.32.8 in the same way with the same strange results, anyone have any thoughts as to what I'm doing wrong just throw it at me, I'm getting desperate and running out of ideas. I've checked all the kernel source symlinks and everything looks good.
I have a problem with PAE kernel sources and builing nvidia driver. uname -a returns Linux myX 2.6.31.12-174.2.3.fc12.i686.PAE #1 SMP Mon Jan 18 20:06:44 UTC 2010 i686 athlon i386 GNU/Linux
1. I run ./NVIDIA-Linux-x86-173.14.22-pkg1.run and I get an error message that kernel sources cannot be found.
2. My folders: /lib/modules had these two subfolders 2.6.31.5-127.fc12.i686.PAE 2.6.31.12-74.2.3.fc12.i686.PAE so what I did, I run: yum install kernel and got inside /lib/modules this extra 3rd subfolder (no PAE) 2.6.31.12-174.2.3.fc12.i686 On the other hand /usr/src/kernels has one subdirectory (no PAE): 2.6.31.12-174.2.3.fc12.i686
3. Now, when I build nvidia driver, it still gives me the error, that no sources found, so I do ./NVIDIA-Linux-x86-173.14.22-pkg1.run --kernel-source-path /usr/src/kernels/2.6.31.12-174.2.3.fc12.i686 and get an error that this seems to be the incorrect version. And I guess this is true as I have no PAE sources.
Code: Linux quad 2.6.33 #1 SMP Sun Mar 7 18:22:02 CET 2010 x86_64 GNU/Linux I am using Trisquel GNU/Linux 3.0. I am asking my questions here, since this involves non-free drivers. I succesfully installed the nvidia driver on the default kernel. But the default kernel has removed all support for DVB USB sticks, so I had to compile my own kernel.
I got the newest version from kernel.org. Saved the archieve to /usr/src/. unzipped the file in the directory (so my kernel source is now in /usr/src/linux-2.6.33/.) Made a symlink with ln -s linux-2.6.33 linux. I compiled the kernel succesfully. Did a "make install" and "make modules_install" and ran "update-grub". Restarted system. Cd'ed to my source directory and ran "make headers_install" succesfully. Looking at my timestamps, it looks like the kernel headers has been installed to /usr/src/linux-2.6.33/usr/include/linux/. I downloaded the latest x86_64 drivers from nvidias website. Went to console 1 and closed up X. If I start the installer without any parameters (sh NVIDIA*.run) I get the following error:
Code: ERROR: Unable to determine the version of the kernel sources located in '/lib/modules/2.6.33/source'. Please make sure you have installed the kernel source files for your kernel and that they are properly configured; on Red Hat Linux systems, for example, be sure you have the 'kernel-source' or 'kernel-devel' RPM installed. If you know the correct kernel source files are installed, you may specify the kernel source path with the '--kernel-source-path' command line option. lib/modules/2.6.33/source is a symlink which point to /usr/src/linux-2.6.33
I get the same error if using --kernel-source-path=/usr/src/linux/, /usr/src/linux-2.6.33/ and similar options which link to this directory through symlinks. If I use --kernel-source-path=/usr/src/linux-2.6.33/usr/include, I get the following error:
So after getting around the Fakeraid bug, and the grub bug, and all the other bugs killing my system fixed and then i moved on to install my video driver. There was nothing in Administration>Hardware Drivers, so i downloaded the latest driver for my two 8600m gt cards. Did the whole ctrl+alt+f2 and then stopping xserver and then running the driver install only to run into yet another damn bug(see log below)
I have tried fixing it by doing what other threads have said to do e.g:[URL] still nothing. Below is the Nvidia log.
I installed the latest kernel liquorix (2.6.35) but when i want to install the Nvidia driver downloaded on the Nvidia website (256.53), i have an error message because Nvidia doesn't found the kernel source tree.
I install linux-image-2.6.35-6.dmz.2-liquorix-686_2.6.35-16_i386.deb, linux-headers-2.6.35-6.dmz.2-liquorix-686_2.6.35-16_i386.deb and build-essential. I don't understand why the installation doesn't works.
Squeeze-beta was my first foray into Debian, and I love it. I changed my setup to a rolling setup with testing (Wheezy), and have done that for several months. Lately, I got a new kernel, but it reboots to a terminal rather than GUI (I'm a simple laptop user). I think it's because of the NVIDIA drivers, and here is what I've tried (meanwhile, I'm using the previous kernel):
# apt-get install module-assistant nvidia-kernel-common # m-a auto-install nvidia-kernel${VERSION}-source A blue screen appears that says:
module-assistant error message Bad luck, the kernel headers for the target kernel version could not be found and you did not specify other valid kernel headers to use.
You can try:
module-assistant prepare or apt-get install linux-headers-2.6.38-2-amd64
I have done both, rebooted, and I still get the blue screen. I also see this message:
nvidia-kernel-source was not built successfully, see:
/var/cache/modass/nvidia-kernel-source*buildlog*
...and I have copy/pasted the file below (which omits lines 101-200 because this message is too long then):
I'm attempting to install ndiswrapper-dkms package. The installation fails due to the following error:Module build for the currently running kernel was skipped since the kernel source for this kernel does not seem to be installed. dpkg reports that the source is installed
[code]....
I've been searching google for the past couple of days but haven't found anything specific. Any ideas to get me going in the right direction?
uname -a reports
Linux debtop 2.6.32-trunk-686 #1 SMP Sun Jan 10 06:32:16 UTC 2010 i686 GNU/Linux
I'd like to have a shot at building FBReader 12.1 from source for use within F12 (kernel 2.6.31.9-174.fc12.i686.PAE) but am having problems locating the following .rpm files:
libz and libbz2 -- libraries for zip and bzip2 (de)compression libfribidi -- for bidirectional text support lincurl, version >= 7.17 -- for network libraries integration
I have a brand new installation of 13.37 but I can't seem to get the nvidia-kernel (64 bit only) Slackbuild from slackbuilds.org to build. I have the nvidia driver installed and working on 13.1 so I am not desperate yet: although I thought that it could just be something with the older Nvidia driver so I waited for the updated release but it fails in the same way.I have uninstalled xf86-video-nouveau and installed xf86-video-nouveau-blacklist-noarch-1.txz from /extra.I have also installed libvdpau from SBo. The nvidia-driver Slackbuild builds fine.I am using the stock 2.6.37.6 huge kernel but have also tried the 2.6.38.4 kernel from /extra and the 2.6.38.7 kernel from Slackware-current.Here is the error message I am seeing:
I've only a small /boot sector and rpm -qa | grep kernel kernel-2.6.31.12-174.2.22.fc12.i686 kernel-PAE-devel-2.6.32.9-70.fc12.i686 kernel-PAE-2.6.32.9-70.fc12.i686 kernel-2.6.32.9-67.fc12.i686 abrt-addon-kerneloops-1.0.8-2.fc12.i686
I'm using the PAE kernels, need the devs for nvidia kernel building,can I remove all the non PAE kernels without damage please?
Problem: computer and graphics card had been running for months without problem. For some reason, last night, things locked up...wouldn't think it's related, but happened when I noticed I was locked out of creating new folders on my usb attached My Book storage drive. So, from terminal I tried to sudo change permissions to read/write. Everything locked up. Now, every time I reboot it states needing to use low graphics mode "Screen, graphics card, and input device settings could not be detected correctly."
When I check the error log I see: "(EE) Failed to initialize the NVIDIA kernel module.."(EE) NVIDIA(0): ***Aborting* Screens found but none have a usable configuration. Fatal Server Error - no screens found"
If I try to then start in safe low graphics mode or restart X, I get the Ubuntu splash screen with the 5 dots that turn red to show progress...but nothing ever happens. I can switch to VT 1, login at command line, and type "startx" and then the graphical desktop loads just fine. Video playback, etc. is perfect and fine - not in low graphics mode at all. System seems totally normal.
However, I have to go through these steps each time I reboot. Also, I noticed that when I try to boot from Ubuntu Live CD I can't. I get as far as picking English as a language, then screen goes blank and goes to power saving mode (no signal being sent). This greatly concerns me as even though I have a work around for now - getting to command line then startx'ing - if I ever needed to reinstall from the CD, it seems I can't.
Some info on my pc:Running 10.04 on Compaq Presario SR1950NX AMD 64 GeForce 6150 LE/PCI/SSE2/3DNow! <- believe built onto motherboard on what I can do to fix? Is this a sign that something is wrong with my video card?
I would like to create some packages for programs that use the Git software to manage the source, I want to do this directly using the source code in the Git repo as the source code form the package. Does anyone have any hints or directions on how to do this? I been looking round the web for information about this, but all I seem to find is for about storing and build Debian packages from git, this is not what I want to do as these packages are for personal use and I have no need for this.
Apparently, the wireless adapter Realtek 8171 uses the rtl8192se driver, which is currently available only as source from Realtek's website. Is building a driver equal to building a package? Should I follow the same cautions, say these ones?
When I compile a custom kernel with this command: make-kpkg --initrd kernel_image kernel_headers and then install the .deb, there's no initrd in /boot and I have to create it manually. I've thought that the --initrd option should take care about this, but somehow it doesn't.
It behaves like this for about two years at least (since I've compiled my first kernel). Of course, it's no big deal to create it manually, I was just wondering whether do I do anything wrong or whether should I fill a bug report..
I have made a full install of Slack 13.37.0. When I try to run the installer of the NVIDIA GF 8400GS card (NVIDIA-Linux-x86-180.29-pkg1.run) downloaded from [URL], I receive an error message. It says that the kernel source cannot be indetified/found. /lib/modules/2.6.37.6-smp/source and .../build links to /usr/src/linux-2.6.37.6-smp, which contains the full kernel source (can be compiled), including the header .h files in include/linux. The same NVIDIA installer can be run successfully on my previous Slack 11.
In order to configure a dual display in a fresh 11.4 installation, I am attempting to install the nVidia driver according to the instructions in SDB:NVIDIA the hard way - openSUSE. I have not previously compiled or configured a kernel so I largely followed the instructions from OpenSUSE 11.2 - How to compile a Kernel for Newbies.The currently installed kernel is 2.6.37.1-1.2-desktop. The graphics card is an ASUS EN9600GSO (512MB).
In Yast, I installed gcc, make, kernel-devel (v. 2.6.37.1-1.2) and kernel-desktop-devel (v. 2.6.37.1-1.2). Per the "hard way" instructions, I did not install kernel-source.
I'm trying to build the kernel module for dvbhdhomerun. The problem is the linux kernel headers provided by Debian do not have the dvb header files. Trying to set up my own kernel source tree to use is not working out so well.
I'm following the instructions in the Debian section here: [URL] ....
My first try compiled, but the modules ender up under /lib/modules/3.2.51/extra instead of /lib/modules/3.2.0-4-amd64/extra. Why?
I tried to start over, but noticed this while running apt-get source linux-image-3.2.0-4-amd64:
Code: Select allReading package lists... Done Building dependency tree    Reading state information... Done Picking 'linux' as source package instead of 'linux-image-3.2.0-4-amd64' NOTICE: 'linux' packaging is maintained in the 'Svn' version control system at: svn://svn.debian.org/svn/kernel/dists/trunk/linux/
[Code] ....
The failure to verify the signature has me concerned, but beyond that, what do I do about the different version number when compiling and installing the kernel modules?
I was trying to install VPN client for my Ubuntu 9.10 64-bit. During the installation process the terminal reads:
"Directory containing linux kernel source code [/lib/modules/2.6.31-21-generic/build]"
On that I pressed enter for the default option (in bold). After a few more steps I reached the following error:
Making module sh: Can't open ./driver_build.sh Failed to make module "cisco_ipsec.ko".
[/lib/modules/2.6.31-21-generic/build] is the location where the installer expects the kernel source to be (I am guessing). So unless I correct the terminal (by providing the location of the kernel source), I think I will keep on getting the same error message.
So to get the kernel source I visited: [URL]From there I copy pasted the command:
I upgraded my kernel in my Debian Lenny to version 2.6.32-bpo.4-amd64 to fix a problem with a software and ALSA-related issues. The thing is that now, when I boot, eth1 is not enabled and I have to "load" (I guess) the driver each time. To do this, I go to the folder where all the files for the drivers are and I run:
modprobe lib80211 and then insmod wl.ko
That I got from the readme file. That gets the wireless working perfectly again, but when I reboot the wireless is dead again and I have to run the commands above again. I imagine that what I should do is to build the driver again, for my new kernel. However, according to the readme file, before I need to install linux-headers-generic and build-essential. With build-essential, it says it's already the newest version, and with the linux-headers, I don't know which one should I install for my kernel! I thought it would be the linux-headers-2.6.32-bpo.4-xen-amd64, but when I tried that, this is what happened:
debian:/home/hernan/Programas/hybrid_wl# apt-get install linux-headers-2.6.32-bpo.4-xen-amd64 Reading package lists... Done Building dependency tree
[code]....
So the warning says something about the MODULE_LICENSE being missing... how to build the driver again to make it load each time I boot?
When trying to install the nvidia graphics driver I get the following error-message:
ERROR: The kernel header file '/usr/src/linux/include/linux/version.h' does not exist. The most likely reason for this is that the kernel source files in '/usr/src/linux' have not been configured. How do I configure the kernel source files?
When i finish install debian7.8 wheezy in my Acer computer(graphics :GT750M+ Inter HD Graphics Family),
it appear:
"GNOME3 Failed to Load" " Unfortunately GNOME 3failed to start properly anmode" "This most likely means your system(graphics hardwcapable of ....."
i check system's Driver status:Gallium 0.4 on llvmpipe(LLVM 0x209) Experience Fallback so i try to install nvidia drivers in debian,but i have failed many times. everytimes i finish install nvidia drivers i 'startx " failed
How do I force recompilation of the kernel .deb packages. After a small change I make to the sources without having to clean the sources and recompile the whole kernel again?
Code: Select all$ fakeroot make -f debian/rules.gen setup_i386_none_686 $ fakeroot make -f debian/rules.gen binary-arch_i386_none_686 binary-indep
Calling the second command again does not recompile the modified code, just recreates the .deb packages.
If I use:
Code: Select all$ make -f debian/rules clean before the build command, then it will recompile everything which takes ages.
How can I force recompilation of the files/objects I changed (and dependencies)?
I use this guide: [URL] .... ('Building only a single kernel variant' section)
Got this while installing the kernel development tools on my Squeeze system being used as a router/firewall. Setting up update-inetd (4.36) ... Setting up cvs (1:1.12.13-12) ... Ignoring install-info called from maintainer script The package cvs should be rebuilt with new debhelper to get trigger support Ignoring install-info called from maintainer script The package cvs should be rebuilt with new debhelper to get trigger support Setting up autopoint (0.17-11) ...
Is that an issue that I need to address prior to building the latest kernel (2.6.33.4)?
I'm I seeing this wrong or is the initrd file in kernel-2.6.27.25-170.2.72.fc10.x86_64 a zero byte file and that's why I can't boot with it( get this "kernel panic not syncing VFS unable to mount root fs on known -block (0,0)" message)
The RHEL 4 update 7 installation CD installs the kernel with version 2.6.9-78.EL and therefore must contain corresponding vmlinuz and initrd.img. However I would like to have an all modules initrd that is used during installation but for a different version (2.6.9-78.0.13.EL). Is there a way in which I could obtain/make such an initrd?
I am running Lenny and using VirtualBox OSE 1.6.6. In order to try squeeze I installed it in a VM. I tried to install Guest Additions also, but i get this error:
VirtualBox 1.6.6 Guest Additions installation Building the VirtualBox Guest Additions kernel module... Building the shared folder support kernel module... Unable to build the kernel module.
I checked what /var/log/vboxadd-install.log had to say:
/usr/src/linux-headers-2.6.32-5-common/include/net/neighbour.h:115: error: variably modified a at file scope /tmp/vbox.0/utils.c: In function 'sf_path_from_dentry:
[code].....
I tried with 3 versions of the kernel 2.6.32-5-686,2.6.32-5-484,2.6.32-3-686 but I cannot get it to work using any of them. Also I tried to install version 2.6.30-2-686 but I couldn't since synaptic said that it Depends: linux-kbuild-2.6.30 but it is not installable.