I am compiling a linux kernel (4.9.15) in my host machine and installing it in an embedded device. All works fine but I have a question about the proper way to install all the include files in the system (/usr/include/linux with updated version.h)
This is how I proceed:
I compile the sources in my host machine.
My compilation script generates boot.tar.gz, modules-4.9.15.tar.gz, linux-4.9.15.tar.gz and linux-headers-4.9.15.tar.gz and copy them to the embedded system
In the embedded system I extract boot.tar.gz (containing System.map-4.9.15, config-4.9.15, initrd.img-4.9.15 and vmlinuz-4.9.15) to /boot folder, the modules-4.9.15.tar.gz to /lib/modules and the source and headers files to /usr/src/
I update the /lib/modules/4.9.15 "source" and "build" links to point to /usr/src/4.9.15 folder
In /usr/src/linux-4.9.15 folder I do make install_headers
I update grub with update-grub2 and reboot
My doubt is how to update the system's /usr/include/linux/ folder. I thought that executing make headers_install would do it but it just installs the include files in the sources folder. Should I copy the generated /usr/src/linux-4.9.15/usr/include/linux folder to the system's /usr/include/linux manually? Is that the proper way to do it?
Any suggestion to do the process in a better way?
Thanks!
Related
What's the purpose of the .bin directory within node_modules?
In another question, the answerer stated:
"it's where your binaries (executables) from your node modules are located."
So additionally can someone explain to me the following: What are binaries/executables?
Any help would be greatly appreciated!
Binary or executable files are files which have already been compiled for your specific computer architecture and, once installed, these files can be ran directly on your computer. Common instruction set architectures are: X86 and ARM, which most computer processors are based upon. Contrary to binary files, source files are the actual source code itself and these files need to be compiled prior to installing.
As for the .bin directory, within ./node_modules/.bin, this directory stores all the executables of your node_modules for which your project is dependent upon to run. This allows your project to just 'run' the libraries which are necessary, for your project, without you having to worry about compiling these files yourself. By compiling I mean transforming the source code down to executable code (machine code) which can understood by your computer's underlying processor.
Hopefully that helps!
According to npm docmentation
When in global mode, executables are linked into {prefix}/bin on Unix, or directly into {prefix} on Windows.
When in local mode, executables are linked into ./node_modules/.bin so that they can be made available to scripts run through npm. (For example, so that a test runner will be in the path when you run npm test.)
Binaries are executable files ( the compiled version of your file for a specific computer architecture) and once installed they can be run directly from your machine
I have a Yocto build for i.mx6 and I want to modify its Kernel. I figured that if I copy Kernel source outside the Yocto project and make my modifications without dealing with patches, I can speed things up significantly. But the thing is, the Kernel source I have to use is already patched and I want to fetch and continue working from there. I will work on the already-patched source files and re-arranging them is a painful process.
For starting point, my patches work fine, and I can get a working image using bitbake fsl-image-multimedia-full command. The Kernel source I want to use is created after this process.
I have tried copying the source under ..../tmp/work-shared/imx6qsabresd/kernel-source. Although make zImage and make modules finished without any trouble, manual building was not successful with an error in a dtsi file (Unable to parse...). Of course, I have checked the file and there was no syntax error.
Also, I checked the kernel source files I copied and it seems that the patches are successfully implemented.
Am I doing something wrong with the patches? With my manual build routine, I can build unpatched kernel source with no errors. I am sure that there are experienced Yocto users here that have their own workarounds to make this process shorter. So, any help is appreciated. Thanks in advance.
You can also edit files in tmp/work-shared/<machine>/kernel-source then compile modified kernel with bitbake -C compile virtual/kernel
My favorite method of doing kernel development in a Yocto project is to create an SDK and build the kernel outside of the Yocto system. This allows more rapid builds because make will only build new changes, whereas a kernel build within Yocto always starts from scratch.
Here are some of my notes on compiling the Linux kernel outside of the Yocto system. The exact paths for this will depend on your exact configuration and software versions. In your case, IMAGE_NAME=fsl-image-multimedia-full
Run bitbake -c populate_sdk ${IMAGE_NAME}. You will get a
self-extracting and self-installing shell script.
Run the shell script (for me it was
tmp/deploy/sdk/${NAME}-glibc-i686-${IMAGE_NAME}-cortexa9hf-neon-toolchain-1.0.0.sh),
and agree to the default SDK location (for me it was
usr/local/oecore-i686).
Source the scripts generated by the install script. I use the
following helper script to load the SDK so I don't have to keep
track of the paths involved. You need to source this in each time
you want to use the SDK.
enable_sdk.sh:
#!/bin/bash
if [[ "$0" = "$BASH_SOURCE" ]]
then
echo "Error: you must source this script."
exit 1
fi
source /usr/local/oecore-i686/environment-setup-corei7-32-${NAME}-linux
source /usr/local/oecore-i686/environment-setup-cortexa9hf-neon-${NAME}-linux-gnueabi
Copy the defconfig file from your Yocto directory to your kernel
directory (checked out somewhere outside of the Yocto tree) as
.config.
Run make oldconfig in your kernel directory so that the Linux
kernel build system picks up the existing .config.
Note: you may have to answer questions about config options that
are not set in the .config file.
Note: running make menuconfig will fail when the SDK is enabled,
because the SDK does not have the ncurses libraries set up
correctly. For this command, run it in a new terminal that has not
enabled the SDK so that it uses the local ncurses-dev packages you
have installed.
Run make -jN to build the kernel.
To run the new kernel, copy the zImage and ${NAME}.dtb files to
your NFS/TFTP share or boot device. I use another script to speed
up the process.
update_kernel.sh:
#!/bin/bash
set -x
sudo cp /path-to-linux-source/arch/arm/boot/dts/${NAME}.dtb /srv/nfs/${DEVICE}/boot/
sudo cp /path-to-linux-source/arch/arm/boot/zImage /srv/nfs/${DEVICE}/boot/
set +x
You can also point Yocto to your local Linux repo in your .bb
file. This is useful for making sure your kernel changes still
build correctly within Yocto.
SRC_URI = "git:///path-to-linux-source/.git/;branch=${KBRANCH};protocol=file"
UPDATE: Over a year later, I realize that I completely missed the question about broken patches. Without more information, I can't be sure what went wrong copying the kernel source from Yocto to an external build. I'd suggest opening a Bitbake devshell for the kernel and doing a diff with the external directory after manually applying patches to see what went wrong, or just copy the source from inside the devshell to your external directory.
https://www.yoctoproject.org/docs/1.4.2/dev-manual/dev-manual.html#platdev-appdev-devshell
When debugging certain commands or even when just editing packages, devshell can be a useful tool. When you invoke devshell, source files are extracted into your working directory and patches are applied.
Since it can't parse it, there seems to be a problem with patch. How do you patch the device tree? Are you patching it in the .bb file?
If so, check your patch for possible syntax errors, it's very easy to overlook the syntax errors in device tree. You can remove the patch and do it manually from bitbake -c devshell <kernel-name>
If not, please try to do it there and check again. Please share results if any of these helps you.
I am trying to build a Linux uImage and device tree for the zynq but my computer cant find the aio_abi.h file. I have moved the Linux folder which contains the file into the shared library folder that my system is looking but the file is still not being found. I think this is because there is no .so file for the folder containing the aio_abi file. I installed libaio-dev using the Ubuntu software installer tool because get-apt isn't working on my machine because of a proxy.
Is there a way to directly create a .so file for the folder I need to include or is there something wrong with the libaio-dev install? I have an older libaio1.so file but the older version doesn't contain the aio_abi file required.
Thanks
I am running ubuntu 13.10 on linux kernel version 3.11.0-12.I have to add a system call in this but i am facing a problem. The very first step says that I have to change my current working directory to kernel directory.
I used the command " cd linux-3.11.0-12 " but it is showing that no such file or directory exists. Please tell me where am i going wrong and how do I correct this mistake.
Wait, you want to add a system call to the Linux kernel, but you don't know how to reach the source code? Are you sure you are able to modify, configure, build, install and boot the Linux kernel?
Assuming yes, you need to get the source code of Linux first (for example, by cloning https://github.com/torvalds/linux or just downloading the version you are interested in), extract it somewhere and then cd to the path where you extracted it. Then you can begin modifying to your heart's content.
Perhaps this blog post could help you.
To get the source of the installed kernel on ubuntu, you can use the command [for ubuntu 13.04+ ]
apt-get source linux-image-`uname -r`
and should be typically be placed under /usr/src
Reference:
[1] https://help.ubuntu.com/community/Kernel/Compile
[2] https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel
I am going to setup build environment to make my own linux embedded system for AT91SAM9X25 Board. I am using buildroot to do this. The make command build all targets, the first it build toolchain then packages and then rootfs and images of rootfs (tar, cpio ...).
To rebuild rootfs I usually use make clean and then make. The make clean command removes all and including toolchain.
So the first my question is: Is there some way to remake rootfs without building toolchain? It takes a lot of time.
Also I am building linux kernel within buildroot. I have turned on BR2_LINUX_KERNEL [=y] in buildroot. The linux configured to use Initial RAM filesystem, so to build kernel it required image of rootfs (which should be created by buildroot). When I run make under root of buildroot the building fails with error Cannot open 'buildroot-2013.05/output/images/rootfs.cpio'. Because (if I understand correctly) the building sequence is toolchain - pakages - rootfs - linux kernel - images of rootfs. When it tries to build linux kernel the rootfs.cpio image is not created.
So the second question is: How to build linux within buildroot if I want to use Initial RAM filesystem?
Possibly are there more efficient alternatives than buildroot?
Thanks in advance.
The make command build all targets
You do not want to do that (until Buildroot is configured).
You first need to configure Buildroot by specifying the target board.
Per the manual you can start from scratch, or create a Buildroot config file for your AT91SAM9X25 board derived from a similar board such as configs/at91sam9g20dfc_defconfig
Besides the Buildroot config file, you will also need a Linux kernel config file (unless you want to try configuring the kernel from scratch).
The kernel config file for Atmel's eval board with a AT91SAM9x5 is at91sam9x5ek_defconfig
You should also read section 3.4.2. Creating your own board support
So the first my question is: Is there some way to remake rootfs without building toolchain? It takes a lot of time.
The answer depends on how you define "remake rootfs".
If you delete the directory output/images/, then the files of the rootfs are rewritten.
If you delete directories in output/build/, then those packages or subsystems are recompiled from source.
If you configure Buildroot to use your own or an external tool chain, then make clean would not remove them. If you configure Buildroot to install the toolchain it builds outside of its directory, then it may leave it alone during a make clean.
Of course the Buildroot make is smart enough to know what has changed since the last build and what has to be recompiled.
It should be the rare case that you need to delete directories in output/build/ to force recompilation.
So the second question is: How to build linux within buildroot if I want to use Initial RAM filesystem?
You need to properly configure both Buildroot and the Linux kernel.
make menuconfig
Filesystem images --->
make linux-menuconfig
General setup --->
make
More concise information on using Buildroot for AT91SAM9x5 is this Linx4SAM page
Possibly are there more efficient alternatives than buildroot?
There are other tools such as Open Embedded, but describing them as "more efficient" is subjective.
ADDENDUM
how to rebuild rootfs in buildroot
To force the rootfs to be rebuilt (in this case an initramfs) delete three hidden files in the output/build/linux-x.xx.xx directory
.stamp_images_installed
.stamp_initramfs_rebuilt
.stamp_target_installed
Rebuild rootfs in buildroot
View dependencies:
make show-targets
Example output:
rootfs-cpio rootfs-tar rootfs-ubi rootfs-ubifs
Rebuild the target. Example:
# Tell buildroot to rebuild `rootfs-ubi`
make rootfs-ubi rebuild
If you only want to regenerate the rootfs partition, do
rm -r output/target && make
This helps, e.g., if you removed files in a rootfs overlay.