Hot answers tagged

10

The ${D} variable allows the software being built to be installed in a directory other than its real target. For example, you might configure the software so that libdir is /usr/lib, but that's for the target device; when you run the installation on your build system, you don't want the newly-built files to actually be installed in /usr/lib, you want the ...


8

When U-Boot executes the boot command, it provides a memory address for the kernel and a memory address for the device tree blob. Therefore, prior to this command, it must load these files into memory. Based on the messages you provided we see that two files failed to be loaded from the eMMC/SD card: /boot/boot.ub /boot/imx6dl-ts4900-13.dtb Its possible ...


8

As per this section of Bitbake manual ?= is: You can use the "?=" operator to achieve a "softer" assignment for a variable. This type of assignment allows you to define a variable if it is undefined when the statement is parsed, but to leave the value alone if the variable has a value. Here is an example: A ?= "aval" If A is set at the time this ...


8

Sometimes Samba fails to install or update it's dependencies correctly. Inspect /var/log/samba/log.smbd, looking for anything regarding talloc2 errors. Try running the following command to ensure all the dependencies are available, this is on Ubuntu, modify for your Linux flavour: apt-get install --reinstall libsmbclient libsmbclient-dev libtevent0 ...


7

I'm way late on this, but I implemented this script and I'll address this for anyone who finds this using an internet search engine. This computer on module can be put on almost any off the shelf TS or custom baseboard, and we wanted it to automatically work without users having to adjust the device tree used. We have an 8-input shift register on any given ...


7

/etc/inittab is a feature of the init system in use, and not part of the kernel. After the kernel has completed its part of the system initialisation it will create the first user-space process. Typically this is /sbin/init but it can change. Now what /sbin/init does depends on the type of init system you have. The traditional "sysvinit" will read /etc/...


7

From what I can understand, you want to create a rule when you build the OS and have it reside in the rootfs. If that's correct, then you need to add a recipe for it. Store your rule in a directory such as the following "distro/meta-myproject/recipes-core/rfs-myproject/rfs-myproject/myrule.rule". Then, up a level in the rfs-myproject folder you need a rfs-...


5

Last time I had to do this I overrode the default gnome-terminal, which I don't have, with xterm. Assuming you have installed xterm then edit your local.conf file, eg with echo 'OE_TERMINAL = "xterm"' >>conf/local.conf


5

The choice between sudoers and sudoers.d has nothing to do with security, but everything with maintainability. By uncommenting the sudo group line in /etc/sudoers, you can add all users that need to have sudo access to the sudo group. This may or may not be easier to do than adding a new file in sudoers.d, depending on your setup. However, changing the ...


4

Yocto can create images utilizing either systemd or SysV init. In either case the program used as init will have pid 1. As you are seeing a systemd process on your system you will have a configuration utilizing systemd. Systemd has no run levels perse.


4

You need to enable it a boot time: systemctl enable boondocks-agent


4

You say, In YOCTO, the equivalent to calling systemctl enable is: SYSTEMD_SERVICE_${PN} = " \ boondocks-agent.service " I think this is the equivalent to installing the service.  To have the service default to enabled you would want to add: SYSTEMD_AUTO_ENABLE_${PN} = "enable"


3

${D} is the usual top of the path used by a Yocto recipe where the recipe installs files that belong to packages produced by this recipe. These directories under ${D} are scanned by the build system for files to include into packages (like RPM files) and to be included in the so-called sysroot tree where files (e.g. libraries and headers) go that are used by ...


3

I had the same problem (also using yocto fido). I couldn't use the proposed answer because i didn't manage to build xscreensaver (has way to many dependencies) I finally solved it by using xssstart and modifying slock which is a screen locker. I called it clicklock https://github.com/zpfvo/clicklock I also made a fork of xssstart with autotools build ...


3

The problem may come from the common miss-use of the word Linux. Linux is just the kernel, however it is often used to mean the whole system (kernel, shell, tools, libraries, compiler, and user applications) see here https://www.gnu.org/gnu/linux-and-gnu.en.html it may clear up some confusion. Debian, Ubuntu, Mint, CentOs, Suse, Vector, … are all ...


3

There is a package called haveged that serves just this purpose.


2

This is a rather old question. However I won't be the last one struggling to solve this. In short: I've found a solution. I'm running chromium in kiosk mode with fluxbox as WM. Instead of relying on Xorg to put the screen to sleep I'm using xscreensaver with blanking only. This enables me to listen for state changes of the screen with a small perl script. ...


2

I had the exact same problem and found that subdir and name didn't work, but destsuffix=<somename> worked. As far as I know this is not documented but can be understood from the bitbake fetch code. Example like airfishey: SRC_URI = "git://<git_repo_uri>;protocol=ssh;rev=<revision>;destsuffix=first_repo \ git://<git_repo_uri&...


2

Have you tried adding subdir= to your git lines? Something like this? SRC_URI = "git://<git_repo_uri>;protocol=ssh;rev=<revision>;subdir=first_repo \ git://<git_repo_uri>;protocol=ssh;rev=<revision>;subdir=second_repo \ file://<some file to grab>" It looks like there was a bug filed against this ...


2

Assuming you want a script that will send a password to the remote when logging in via ssh, here's some very minimal python code that should not need any extra libraries and so on. Obviously, it is just an example of what is possible using just os forkpty execlp read write. #!/usr/bin/python # simplest builtin python pseudo-tty for ssh password. meuh # ...


2

According to https://cateee.net/lkddb/web-lkddb/MWIFIEX.html, you have to have CONFIG_MWIFIEX in order to support the Marvell Technology Group Ltd. 88W8897 [AVASTAR] 802.11ac Wireless. Most probably you have to reconfigure your kernel to contain this. EDIT: usually the default kernels provide the options to check their configuration like this (on the ...


2

(deleted) appended to a file name output by the kernel is the standard behaviour for files which have been deleted (see d_path in fs/dcache.c — it’s well-documented, you don’t need to read C). In this case, it means the device node /dev/mmcblk0p4 has been deleted for some reason. (This doesn’t mean that the partition itself has been removed, it’s just the ...


2

You could try using UART1 or UART2 on the Mira board. According to the manual they use TTL level signals. The GPS module outputs 3 V, but tolerates 5 V on input. The +3 V should be enough to be interpreted as a "1" on a TTL input. The RS-232 signals are not suitable without a buffer, because RS-232 specifies +3..+12 V for the space state and -3..-15 V (that'...


2

You ask, "Is there any alternate method to install phpmyadmin other than using commands?" No.


2

x86_64 and i686 instruction sets are not the same (for more details). Use this command to compile source on your host : gcc -m32 -mtune=i686 source.c -o source if getting error like this fatal error: *.h : No such file or directory this means that you don't have proper a library for cross compilation sudo apt install g++-multilib And run gcc ...


2

Enabling TPM support in the kernel is only part of the setup; you also need to provide access to either a real TPM or a TPM emulator. If you have a real TPM which isn’t used by anything else, you can forward it to QEMU as follows: runqemu qemuparams="-tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0" If you want to use a TPM emulator (there might be ...


1

First, I'd observe that both of the outputs in your question are from SysV init (INIT: version 2.88), so at some point you have become confused. Anyway, WRT the timing of various services, you can get a pretty snazzy SVG with systemd-analyze plot > sysd.svg (the redirection here is necessary, otherwise it prints SVG data to standard output, which is not ...


1

You probably need to mount the debugfs filesystem: mount -t debugfs none /sys/kernel/debug If you already have a startup script of your own in /etc/init.d/ it should be possible to add it there, or you might add it to /etc/fstab as debugfs /sys/kernel/debug debugfs defaults


1

Check out the large but comprehensive Yocto Reference Manual and the simple example there. It shows a typical install script will move files to a subdirectory of ${D}, eg install -d ${D}${bindir} You must use a similar tactic, eg install -d ${D}/test and so on. This variable is explained here. Read section 5.3.13. Installing for what is expected to ...


1

You should look up what git is: it's a version control system. For you, the basic usage is to download a copy of a repository with git clone …. This downloads not only a particular version of the source code, but all the versions of the source code, including proposed variants and change logs. All this extra information is in a hidden directory called .git. ...


Only top voted, non community-wiki answers of a minimum length are eligible