My OpenWRT kernel version is 4.14.22. I downloaded the wireGuard module source code and modified its Makefile to compile using the OpenWRT module. I also compiled successfully, but only after installing the module. Execute command “ip link add wg0 type wireguard” report segment error, what could possibly go wrong
Related
I am new in kernel programming.
For implementing my project work, I have downloaded the latest stable kernel (v4.3) from kernel.org.
Just for checking I have copied a few files from the kernel directories into my project directory. Made changes to it and inserted a few more code to it.
Then I compiled on SLES11 Linux kernel using
make -C /lib/modules/$(uname -r)/build M=$PWD modules
I have used the below makefile
obj-m := my_module.o
my_module-objs := module_main.0 other_module1.o other_module2.o other_module3.o
It compiled successfully.
But when I tried to insert into the kernel using
insmod my_sample.ko
It showed the following
disagrees about version of symbol symbol_name
You need to build your kernel module against the same version kernel you are going to run. Thus if you have kernel 4.3 sources that you have downloaded you need to compile that version of the kernel and boot with that running before trying to load your kernel.
You have two solutions then:
Download the kernel sources for the kernel you are currently running (you can install those with zypper install kernel-source on SLES or an equivalent command on other distributions.)
Compile and install the 4.3 kernel in to your operating system. If you need help with this then ask a separate question (and it probably belongs on superuser not here). Note that if kernel and glibc are tightly coupled, and it is possible that you can't run a new kernel if you have a very old C library.
The problem here is that your Kernel module is using the exported symbols of other kernel modules which in this case appears to be the linux InfiniBand RDMA stack's exported methods or symbols.
To solve the symbol version problems, copy the Module.symvers file from the
/usr/src/ofa-kernel
directory and paste it to your current working directory. Then you make your modules again. Now the insmod should work perfectly fine.
NOTE: The Module.symvers file contains information of all the kernel
module exported symbol. So by copying it to your working directory,
you are helping kbuild to know more about the used exported symbols.
And if you don't find Module.symvers or it is empty, then create one using create_Module.symvers.sh
make -C /lib/modules/$(uname -r)/build M=$PWD modules,
"$(uname -r)" shows that you are compiling against the kernel version you are running now so you should be able to insmod the module in the current kernel if you haven't changed the headers.
From your text,
"Just for checking I have copied a few files from the kernel directories into my project directory. Made changes to it and inserted a few more code to it."
If you have made modifications to the kernel source then you may need to recompile the new kernel and boot with the new updated kernel. Then you should be able to compile your kernel module with the modified headers.
Looks like you built agAinst right kernel.something to do with how your kernel is compiled. (See Config_conversions). Try --force
I am new to contiki os, I tried to compile and run hello-world.c file located in examples directory in contiki os 3.0. But I got the error,
hello-world.c:40:21: fatal error: contiki.h: No such file or directory".
I tried this in root user.
Any kind of help to run very beginner program in contiki is appreciated.
Use the make command to build Contiki. Doing that invokes architecture-specific version of gcc with compiler flags pointing to the the proper header file paths:
By default, make is going to build for the "native" platform (that is, x86) and create an application image that is executable on the host PC.
To build for a specific hardware platform, set the TARGET variable. For example, to build for Tmote Sky/TelosB nodes:
make TARGET=sky hello-world
To build and run for the native platform:
make
Output:
TARGET not defined, using target 'native'
mkdir obj_native
...
CC hello-world.c
LD hello-world.native
rm hello-world.co
To run it:
./hello-world.native
Output:
Contiki-3.x-1457-g552408b started with IPV6, RPL
Rime started with address 1.2.3.4.5.6.7.8
MAC nullmac RDC nullrdc NETWORK sicslowpan
Tentative link-local IPv6 address fe80:0000:0000:0000:0302:0304:0506:0708
Hello, world
Thanks for all, who helped to answer my question. Finally, I am able to run first hello-world program in Contiki OS. I have use the following code syntax.
$cd contiki
$cd examples/hello-world
$make TARGET=native
To run the compiled code use the following code.
$ ./hello-world.native
Out put:
Contiki 3.0 started with IPV6, RPL
Rime started with address 1.2.3.4.5.6.7.8
MAC nullmac RDC nullrdc NETWORK sicslowpan
Tentative link-local IPv6 address fe80:0000:0000:0000:0302:0304:0506:0708
Hello, world
I have an embedded Linux application that can find and install some drivers from my C based application, but there are other drivers that it can't find at all.
From application:
system("insmod i2c_core.ko");
Debug error message:
insmod: i2c_core.ko: no module by that name found
However I have no problem installing any of the same drivers from the command line using the insmod command.
It's not a driver sequence dependency problem because the error message would be different than the above error message.
Linux version is 2.6.10 (old yes, but from reference design)
All the drivers are under this path '/lib/modules' and are built as part of the kernel build.
insmod wants the correct path (absolute or relative, insmod i2c_core.ko is same as insmod ./i2c_core.ko). If the module is properly installed for the kernel, try using modprobe:
system("modprobe i2c_core");
Otherwise, you need to figure out how to find the correct path to the module file.
I would ultimately like to modify and compile the existing Linux USB storage driver and test it. For the first step, I wanted to compile the module as is.
I downloaded the latest Linux kernel (version 3.12) and extracted it to ~/linux-3.12.
I found the driver I wanted to compile: drivers/usb/storage, but when I ran make, I got the following error:
make: *** No targets. Stop.
I found many guides online, but none of them worked for the USB storage driver. All I want is to compile this one module and get the .ko so I can test it out.
NOTE: I'm running Ubuntu 13.04 64-bit, and uname -r outputs 3.8.0-30-generic - I'm not sure if that's the problem, but I managed to compile the whole Kernel before. I don't want to do that now because it takes an eon.
If you wanted to build the drivers/usb/storage module you would do this:
make M=drivers/usb/storage
from the root directory of the kernel tree. Before doing so, you will need to make sure that your configuration is the same as the config of the running kernel.
You can't simply take the source code for one kernel and use it to build modules for another one. The module needs to be built from the same source and with the same configuration as the kernel itself.
Basically, you need to find the source code for the Ubuntu kernel you're running. In Ubuntu, as in Debian, that can be done with 'apt-get source '. The package name is probably something like 'linux-image-3.8-2-amd64'.
Once you have the source code you need to find the configuration of your running kernel. Fortunately Ubuntu keeps that in /boot/config-3.8-....
Copy that config to your kernel source tree as .config and run 'make oldconfig'. Now you should be able to build the module (assuming it's not already built into your kernel!).
I have an OLinuXino board. I downloaded the ArchLinux img file (ArchLinuxARM-2013.02-olinuxino-rootfs.img) and wrote it to the SD card using dd and booted the board using the card. I connected the board to the internet using ethernet and installed gcc and make on it using pacman. I was able to build userspace program for the board o n the board.
The ArchLinux SD card image already had the kernel headers directory in the rootfs (/lib/modules/linux-3.7.2-2-ARCH/build). And so I was able to build loadable kernel modules for the board on the board itself too.
I have an Ubuntu 12.04.1 development PC. I have installed Sourcery CodeBench Lite for ARM GNU/Linux (arm-2012.09-64-arm-none-linux-gnueabi.bin) on it. I am able to cross compile userspace programs for OLinuXino on this development PC and transfer it to the board over SFTP and run it on the board (using console over ttyAMA0 Serial port).
Now I want to cross compile kernel modules for the OLinuXino board. I have done this earlier for another custom build imx233 board - in that case I had configured the kernel build system (LTIB) to leave the kernel sources and rootfs intact after building the image. That way I was able to specify the kernel headers build directory for cross compiling the kernel module and it worked.
This time for OLinuXino I don't have the build sources so I copied the rootfs (using cp -dR) to my Ubuntu PC and tried cross compiling a hello world kernel module by specifying the kernel headers directory as /lib/modules/linux-3.7.2-2-ARCH/build and it threw the following error:
anurag#anurag-VirtualBox:~/HelloKS$ make
make ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- -C /mnt/ArchOL/lib/modules/3.7.2-2- ARCH/build M=/home/anurag/HelloKS modules
make[1]: Entering directory `/mnt/ArchOL/usr/src/linux-3.7.2-2-ARCH'
CC [M] /home/anurag/HelloKS/khello.o
/bin/sh: scripts/basic/fixdep: cannot execute binary file
make[2]: *** [/home/anurag/HelloKS/khello.o] Error 126
make[1]: *** [_module_/home/anurag/HelloKS] Error 2
make[1]: Leaving directory `/mnt/ArchOL/usr/src/linux-3.7.2-2-ARCH'
make: *** [all] Error 2
The fixdep binary in the scripts folder was precompiled for ARM, so I deleted the binary and recompiled it for x86 and placed it there. When I tried cross compiling the kernel module again, a similar error was thrown complaining about another executable in the scripts folder (this time modpost).
My question is how can I replace these arm binaries in the kernel-header / build folder with x86 version? is there a script in the build folder to do this? Can I replace the scripts folder in the copied ArchLinux with the scripts folder from my ubuntu's kernel module build folder? Or do I modify kernel modules's makefile to instruct the build script to rebuild binaries in the scripts folder or use a different scripts folder for this binary (I would specify) the path to ubuntu's scripts folder in its kernel headers folder?
Or am I going about this the wrong way and there a better way to setup cross compilation for the board and setup I have?
PS. FYI: Cross compiler uses libc 2.16 and the ArchLinux img for OLinuXino has libc 2.17 on it
I've successfully compiled a module without kernel sources, only with kernel headers. I used qemu and ARM version library files from gcc-libs and glibc.
Just copy over the host's scripts directory to override the ARM one, except the mod/modpost program. Edit mod/Makefile.modpost to run the modpost program with qemu-arm. If you don't want to actually change those files, you can of cause use aufs / overlayfs / bind mount to achieve the same.
Set QEMU_LD_PREFIX to where your ARM library files are, make sure qemu-arm .../modpost actually work. Then run make as you've tried to do.
If you can read Chinese, you can read my experience here.
If you are compiling for x86 should specify the kernel headers directory as /lib/modules/linux-3.7.2-2-ARCH/build. For cross-compiling either we will be downloading the linux source or use the linux source provided by the SOC manufacturer. Kernel header of the downloaded linux source has to be specified for compilation. Suppose have your downloaded linux source under /opt directory then sample "Makefile" would look like this
obj-m += name-of-driver.o
make -C /opt/linux(specify full version) M=${PWD} modules
Have to install the cross compiler and export i.e. export PATH=$PATH:<absolute-path-of-cross-compiler-binaries>. While compiling using make utility provide make ARCH=arm(Target for which you are compiling) CROSS_COMPILE=arm-none-linux-gnueabi-. Once all these procedure are followed you will be successfully compiling your kernel module for target.