You are not logged in.
Head_on_a_Stick wrote:Change ...
... and changeLike this?
For the firmware file ---> UPD72020X_FW=[path-to-file] K2026.mem
For the executable ---> UPD72020X_CMD=[path-to-file]./upd72020x-load
Q: where in the system should these two (firmware and executable) files be saved?
I would place the script under /usr/local/bin/ and place the firmware under /usr/local/lib/firmware/ then change the script like this:
# set paths to loader and firmware, if not provided by environment
readonly UPD72020X_CMD="${UPD72020X_CMD:-/usr/local/bin/upd72020x-load}"
readonly UPD72020X_FW="${UPD72020X_FW:-/usr/local/lib/firmware/K2026.mem}"^ That replaces lines 7-13 (inclusive) in the original script. It does the same thing without having to run two if...fi loops.
Head_on_a_Stick wrote:[0] Not "SystemD"!
Quite obvious ...
Just who do you take me for?
Sorry, that was a comment about the README in the linked GitHub repository. No offence intended.
Check your RAM health.
The GitHub repository supplies a systemd[0] unit file to load the firmware at boot but that won't work for Devuan.
You could call the upd72020x-check-and-init script using /etc/rc.local instead. Change this line so that it points to the correct location of the firmware file.
EDIT: and change this line so that it points to the correct location of the upd72020x-load executable.
[0] Not "SystemD"! ![]()
Perhaps also check iotop.
Have you tried changing this line:
convert $1 -strip -rotate "$rot" -set filename:copy '%t-rotated.%e' '%[filename:copy]'
To this:
convert "$@" -strip -rotate "$rot" -set filename:copy '%t-rotated.%e' '%[filename:copy]'If the file name is separated by spaces then it will be interpreted as multiple arguments for the script and "$@" will pass them all through whereas $1 (or "$1", as it should be) will only pass the first part of the file name, which won't be recognised.
We're trying to rebuild the package on a Devuan beowulf base and so avoid creating a FrankenDevuan by adding the Ubuntu repositories.
Having said that I've given up on this. Sorry OP.
Ownership of an NFT does not confer control of copyright. Most NFTs can be freely copied by anyone because doing so will usually increase the value of the NFT.
Add amd-iommu=off as well (as per my link).
Just install the debhelper package from the beowulf-backports repository, instructions here (s/debian/devuan/g):
https://backports.debian.org/Instructions/
But I really don't think this will work.
Why do you think you need the newer obs-studio version anyway?
https://www.gentoofan.org/blog/index.ph … llers.html?
tl;dr: add iommu=soft as a kernel command line parameter.
echo 'ondemand' > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
That change will be reverted after a reboot.
To make it permanent add this line to the end of /etc/sysfs.conf (or in it's own file under /etc/sysfs.d/):
devices/system/cpu/cpu*/cpufreq/scaling_governor = ondemandNot sure what you've done to encounter that error, sorry.
I'm just waiting for a new mono version to build in my repository, which might make the meson package build, which will get the backported obs-studio package built. I hope. I'll post back when it finishes.
But I really don't think this is going to work because the installation dependencies don't seem to be satisfiable. Sorry to waste your time with this, I should have checked more carefully yesterday.
Try using "$@" instead of "$1" in the convert line of the shell script.
Off topic but I would also suggest using whiptail (in a terminal) rather than yad because the output could be parsed without resorting to awk:
Yeah, it needs the backported debhelper version.
But I'm trying to build a version in my openSUSE Build Service repository and it needs dwz as well, which needs a newer meson version. Oh dear. This might not work, I can't seem to build a newer meson, not sure why.
I'll try again tomorrow when I'm actually booted into a Debian system.
Doing the backports steps I got this:
sudo mk-build-deps --install --remove mk-build-deps: Unable to install obs-studio-build-deps at /usr/bin/mk-build-deps line 416. mk-build-deps: Unable to install all build-dep packages
Is that the full error message?
I think you need the debhelper package from the beowulf-backports repository.
Edit debian/control and change the linux-generic dependency to linux-image-amd64.
EDIT: or just backport the testing/unstable package, as I suggested earlier. That will probably be simpler.
Only one way to find out ![]()
I haven't tested my method but we can troubleshoot it here from any errors if needed. Takes a bit longer but it makes the thread a bit more useful for others to see the process.
The version from OBS has been compiled in an Ubuntu system so it is not compatible with Devuan. Building a version in your Devuan system from the same source will make it compatible.
I can't follow that guide because they add a ppa...
D'oh! Sorry...
Add the (source) repository manually:
sudo tee /etc/apt/sources.list.d/obs.list <<<"deb-src deb-src http://ppa.launchpad.net/obsproject/obs-studio/ubuntu bionic main"Then download and add the key:
gpg --keyserver keyserver.ubuntu.com --recv bc7345f522079769f5bbe987efc71127f425e228
gpg -a --export EFC71127F425E228 | sudo apt-key add -And finally update the database and build the package:
sudo apt update
apt source --build obs-studioEDIT: or backport the version in testing/unstable by following https://wiki.debian.org/SimpleBackportCreation
EDIT2: no, it is not "safe" to add the binary package repository. Use the source, Luke.
what do I do?
Don't add PPAs to Devuan, it is not binary-compatible with Ubuntu so they can break your system.
Follow this guide if you really need something from a PPA: https://wiki.debian.org/CreatePackageFromPPA
Should I enable the Ubuntu repo?
No. You'll have to download the individual .debs from the Ubuntu repositories and install them in order. Use the URL for the package you already have and change the package name (and version if necessary).
Or perhaps follow https://wiki.debian.org/CreatePackageFromPPA to build some local Devuan packages from the Ubuntu source.
I'd appreciate some guidance as to which to choose between the ones available for Ubuntu 18.04 LTS.
I think the latest version would be best (fwts_18.03.00-0ubuntu5_amd64.deb).
But you might be able to fix things with acpica-tools instead. See https://wiki.archlinux.org/index.php/DSDT for an overview of the technique.
If you can fix it then remember to submit a bug report with your findings so that the upstream kernel developers can make things work for everybody else.
is there something that can be done so this does not happen again?
Not without knowing what caused it in the first place.
You shouldn't see interface names like that for internal interfaces.
That interface name is derived from the MAC address and is used when the firmware doesn't provide onboard or hotplug index numbers or the physical location of the device connector is unknown. This is most common for USB devices but it can also happen with internal cards.
I don't know the difference between apt and apt-get, looking around I found that "apt is better way in devuan/debian shell".
Do I'm wrong?
The apt command offers a more "user-friendly" interface for APT and is currently recommended over apt-get for interactive use but not for scripting (because the command interface isn't stable yet).
after reboot and log in xfce, to connect to network, dhclient was needed again
Edit /etc/network/interfaces to configure the device:
auto enx34298f882acf
iface enx34298f882acf inet dhcpSee also https://wiki.debian.org/NetworkConfigur … _interface & interfaces(5).