Installing Ubuntu 22.04 or Linux Mint on an HP Chromebook G5 EE works fine, except for the audio hanging and producing a loud beep until pulseaudio is restarted. The solution :
On 12/18/22 7:02 AM, Mark Hills wrote:
HP Chromebook (11a, G5 “Setzer”) uses snd_soc_sst_cht_bsw_rt5645 module
and presents a device “rt5650”.
can you share the result of ‘aplay -l’, wondering which platform driver you are using.
Audio works briefly, until after about 30 seconds it’s interrupted by a
square wave or short buffer cycling.
If this is with the SOF driver, there’s a known issue and a known workaround, please add this to /etc/modprobe.d/alsa-base.conf
I’ve found multiple guides on how to enable NVIDIA GPU access from lxc containers, however I had to combine the information from multiple sources to get a fully working setup. Here are the steps that worked for me.
Install dkms on your Proxmox host to ensure the nvidia driver can be auto-updated with new kernel versions. # apt install dkms
Download the latest nvidia binary driver supported by nvidia-patch. Head over to https://github.com/keylase/nvidia-patch and get the latest supported driver version listed there.
Download the nvidia-patch repo git clone https://github.com/keylase/nvidia-patch.git
Install the driver from step 1 on the host. For example, ./NVIDIA-Linux-x86_64-455.45.01.run
Run the nvidia-patch/patch.sh script on the host.
Install the same driver in each container that needs access to the Nvidia GPU, but without the kernel module. ./NVIDIA-Linux-x86_64-455.45.01.run --no-kernel-module
Run the nvidia-patch/patch.sh script on the lxc container.
On the host, create a script to initialize the nvidia-uvm devices. Normally these are created on the fly when a program such as ffmpeg calls upon the GPU, but since we need to pass the device nodes through to the containers, they must exist before the containers are started.
I saved the following script as /usr/local/bin/nvidia-uvm-init. Make sure to chmod +x !
#!/bin/bash
## Script to initialize nvidia device nodes.
## https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#runfile-verifications
/sbin/modprobe nvidia
if [ "$?" -eq 0 ]; then
# Count the number of NVIDIA controllers found.
NVDEVS=`lspci | grep -i NVIDIA`
N3D=`echo "$NVDEVS" | grep "3D controller" | wc -l`
NVGA=`echo "$NVDEVS" | grep "VGA compatible controller" | wc -l`
N=`expr $N3D + $NVGA - 1`
for i in `seq 0 $N`; do
mknod -m 666 /dev/nvidia$i c 195 $i
done
mknod -m 666 /dev/nvidiactl c 195 255
else
exit 1
fi
/sbin/modprobe nvidia-uvm
if [ "$?" -eq 0 ]; then
# Find out the major device number used by the nvidia-uvm driver
D=`grep nvidia-uvm /proc/devices | awk '{print $1}'`
mknod -m 666 /dev/nvidia-uvm c $D 0
mknod -m 666 /dev/nvidia-uvm-tools c $D 0
else
exit 1
fi
Next, we create the systemd service files to start this script, and the nvidia-persistenced.
# NVIDIA Persistence Daemon Init Script
#
# Copyright (c) 2013 NVIDIA Corporation
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
#
# This is a sample systemd service file, designed to show how the NVIDIA
# Persistence Daemon can be started.
#
[Unit]
Description=NVIDIA Persistence Daemon
Wants=syslog.target
[Service]
Type=forking
ExecStart=/usr/bin/nvidia-persistenced --user nvidia-persistenced
ExecStopPost=/bin/rm -rf /var/run/nvidia-persistenced
[Install]
WantedBy=multi-user.target
Next, symlink the two service definition files into /etc/systemd/system cd /etc/systemd/system # ln -s /usr/local/lib/systemd/system/nvidia-uvm-init.service # ln -s /usr/local/lib/systemd/system/nvidia-persistenced.service and load the services # systemctl daemon-reload # systemctl start nvidia-uvm-init.service # systemctl start nvidia-persistenced.service
Now you should see all the nvidia device nodes have been created # ls -l /dev/nvidia* crw-rw-rw- 1 root root 195, 0 Dec 6 18:07 /dev/nvidia0 crw-rw-rw- 1 root root 195, 1 Dec 6 18:10 /dev/nvidia1 crw-rw-rw- 1 root root 195, 255 Dec 6 18:07 /dev/nvidiactl crw-rw-rw- 1 root root 195, 254 Dec 6 18:12 /dev/nvidia-modeset crw-rw-rw- 1 root root 511, 0 Dec 6 19:00 /dev/nvidia-uvm crw-rw-rw- 1 root root 511, 0 Dec 6 19:00 /dev/nvidia-uvm-tools
/dev/nvidia-caps: total 0 cr-------- 1 root root 236, 1 Dec 6 18:07 nvidia-cap1 cr--r--r-- 1 root root 236, 2 Dec 6 18:07 nvidia-cap2
Check the dri devices as well # ls -l /dev/dri* total 0 drwxr-xr-x 2 root root 100 Dec 6 17:00 by-path crw-rw---- 1 root video 226, 0 Dec 6 17:00 card0 crw-rw---- 1 root video 226, 1 Dec 6 17:00 card1 crw-rw---- 1 root render 226, 128 Dec 6 17:00 renderD128
Take note of the first number of each device after the group name. In the listings above I have 195, 511, 236 and 226.
Now we need to edit the lxc container configuration file to pass through the devices. Shut down your container, then edit the config file – example /etc/pve/lxc/117.conf. The relevant lines are below the swap: 8192 line
Now, start your container back up. You should be able to use NVENC features. You can test by using ffmpeg: $ ffmpeg -i dQw4w9WgXcQ.mp4 -c:v h264_nvenc -c:a copy /tmp/rickroll.mp4
You should now have working GPU transcode in your lxc container!
If you get the following error, recheck and make sure you have set the correct numeric values for lxc.cgroup.devices.allow and restart your container.
[h264_nvenc @ 0x559f2a536b40] Cannot init CUDA
Error initializing output stream 0:0 -- Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width
or height
Conversion failed!
Another way to tell the values are incorrect is having blank (———) permission lines for the nvidia device nodes. You will get this inside any containers that are started before the nvidia devices are initialized by the nvidia-uvm-init script on the host.
$ ls -l /dev/nvidia*
---------- 1 root root 0 Dec 6 18:04 /dev/nvidia0
crw-rw-rw- 1 root root 195, 255 Dec 6 19:02 /dev/nvidiactl
---------- 1 root root 0 Dec 6 18:04 /dev/nvidia-modeset
crw-rw-rw- 1 root root 511, 0 Dec 6 19:02 /dev/nvidia-uvm
crw-rw-rw- 1 root root 511, 1 Dec 6 19:02 /dev/nvidia-uvm-tools
Sometimes, after the host has been up for a long time, the /dev/nvidia-uvm or other device nodes may disappear. In this case, simply run the nvidia-uvm-init script, perhaps schedule it to run as a cron job.
Install pulseaudio-dlna from Ubuntu default repos, then run pulseaudio-dlna. The first time, it failed, but I think that’s because the receiver was switched off. Console output did show it had been detected though. After switching on the receiver then running pulseaudio-dlna again, it redirected my currently playing audio to the receiver after a few seconds! So far this is the first time I’ve managed to get DLNA live streaming working with this receiver in 5 years of owning it.
Note: this post is more intended for personal reference; as such the quality of writing is not up to my standard, but I’ve decided to publish it as-is anyway since I don’t have time to review and figure it could be useful to someone.
I was looking for a good portable Linux laptop to that I could carry around without being too worried about battery life or the device getting scratched up in my bag. I tried a few different devices, including the HP Folio 13, ThinkPad X131e, GPD Pocket, Dell Latitude E7240. Finally I settled on the Acer C720 Chromebook. The main points in favour of this device are, for me:
Compact, thin and light (11.6″)
Relatively low cost (4GB RAM model can be had for under $150 CAD)
Proper modern low-voltage Intel Core processor (Celeron 2957U – Haswell) and not an Atom core or high-end i7
Dual-band Wi-Fi and Bluetooth
and the main downsides:
Basic TN LCD panel
16GB storage
Runs Chrome OS
Chrome OS keyboard layout
The good news is the downsides are fairly easily remedied.
TN LCD Panel
The Acer C720 uses a standard 11.6″ LED backlit eDP panel. I was able to find an IPS panel to replace it on laptopscreen.com (model N116BCA-EA1). The panel replacement is fairly straightforward; care needs to be taken not to break the LCD bezel or any of the clips.
16GB SSD
The SSD on the C720 is a standard M.2 2242 SATA SSD. The M.2 slot is easily accessible after removing the Chromebook’s bottom cover.
Runs Chrome OS
The Chrome OS/Linux dual-boot process is done by using the chrx script (https://github.com/reynhout/chrx). Remember to remove the write-protect screw while the bottom cover is off.
Chrome OS keyboard layout
This was arguably the hardest part. I chose to run Xubuntu, so the steps I went through apply to that environment. Chrx by default provides good defaults, mapping the back, forward, refresh, brightness and volume keys. The big issue for me was the placement of the power button (above the backspace key)
Prevent systemd-logind from handling the power button
edit /etc/systemd/logind.conf and add the following line:
HandlePowerKey=ignore
Prevent xfce power manager from handling the power button
xfconf-query -c xfce4-power-manager -p /xfce4-power-manager/logind-handle-power-key -s true
Allow xfce power manager to handle lid switch
xfconf-query -c xfce4-power-manager -p /xfce4-power-manager/logind-handle-lid-switch -s false
Remap power key to Delete
xmodmap -e “keycode 124 = Delete”
To find the keycode, stop lightdm, startx to start a basic X session, then use xev to get the keycode for the power key.
Other tweaks
Consistent custom wallpaper/background for boot splash, login, and lock screen
gnome-screensaver uses the Gnome DE’s wallaper setting. Configure this via CLI:
gsettings set org.gnome.desktop.background picture-uri file:///usr/local/share/wallpaper/animevillage.jpg
lightdm-gtk-greeter can be configured using GUI tool lightdm-gtk-greeter-settings or by editing /etc/lightdm/lightdm-gtk-greeter.conf
plymouth/boot splash:
go to /usr/share/plymouth/themes and create a copy of the xubuntu-logo directory. I called mine xubuntu-logo-mod
sudo cp -R xubuntu-logo xubuntu-logo-mod
within the new xubuntu-logo-mod directory, rename and edit config files accordingly
325 cd xubuntu-logo-mod/
326 ls
327 sudo mv xubuntu-logo.plymouth xubuntu-logo-mod.plymouth
328 sudo mv xubuntu-logo.script xubuntu-logo-mod.script
329 sudo nano xubuntu-logo-mod.script
replace wallpaper.png in xubuntu-logo-mod directory with desired image. I used ImageMagick’s convert tool to convert my jpg format image to png.
add new theme we have created as an “alternative” with higher priority
Find out priority and config for current theme
sudo update-alternatives –query default.plymouth
Name: default.plymouth
Link: /usr/share/plymouth/themes/default.plymouth
Status: auto
Best: /usr/share/plymouth/themes/xubuntu-logo-mod/xubuntu-logo-mod.plymouth
Value: /usr/share/plymouth/themes/xubuntu-logo-mod/xubuntu-logo-mod.plymouth
Create new alternative with higher priority
sudo update-alternatives –install /usr/share/plymouth/themes/default.plymouth default.plymouth /usr/share/plymouth/themes/xubuntu-logo-mod/xubuntu-logo-mod.plymouth 200
Where “/usr/local/bin/monitorlayout.sh” is the path to the script saved from arandr.
For sddm, instead edit /usr/share/sddm/scripts/Xsetup.sh and add a line to have it run the same “/usr/local/bin/monitorlayout.sh” script.
#!/bin/sh
# Xsetup - run as root before the login dialog appears
# set screen layout using Xrandr
/usr/local/bin/monitorlayout.sh
Since ending my colo server plan, my Subsonic server hasn’t been running. I used to just have all my media synced to the Proxmox server I had in colo, then bind-mounted that folder to the OpenVZ container that was running Subsonic.
I’ve now moved my Proxmox/Subsonic server to a little Zotac ZBOX on my LAN that only has one local disk so I figured I wouldn’t sync all my music over locally when it could just directly access it from my new FreeNAS box. Turns out this is a little bit more complex than I thought – here are the steps I did to get it working:
Get all music onto the FreeNAS box. For this I’m still using BTsync. One client is on my Mac, sharing my iTunes library, the other is in a FreeNAS jail.
Install Subsonic in an LXC container on the Proxmox server.
Create an NFS export on the FreeNAS server and mount it as storage on Proxmox. I just used the “Add storage” option in the Proxmox Web UI rather than adding it to fstab, because I was lazy.
Configure the LXC bind mount. Essentially, in /etc/pve/104.conf :
mp1: /mnt/pve/freenas,mp=/mnt/freenas
Set up Subsonic to scan the /mnt/freenas/music folder.
Realize that Subsonic has no access to write to the NFS share, since the GIDs on the FreeNAS box and the Subsonic container don’t match. I use GID 500 = users as my main group on FreeNAS, so I created a group freenasusers with GID 500 in the Subsonic container and added the subsonic user to this group.
Probably missed some steps, but that’s the gist of it. I really need to get my blogging and documentation back on track.
I don’t really play DVDs on my PCs much anymore, but my brother have me the full boxed set of Freaks and Geeks (great show, cancelled too soon). Running on a fresh install of Ubuntu 14.04, I popped in the DVD and was prompted by “Video Player” aka Totem to install some additional codecs. I obliged, but after the installation, the DVD still wouldn’t play. It turns out that because of legal issues, the libdvdcss2 library is no longer included in the Ubuntu repositories. However, the libdvdread4 package does provide a script to easily install it.
Once you have libdvdread4 installed, run following script:
This system also hasn’t been re-installed since November 2012. The initial install was done using the Ubuntu 12.10 “Quantal Quetzal” CD on November 20, 2012. I’ve since upgraded to every interim release – 13.04, 13.10 and now 14.04, and used a host of different desktop environments and window managers (currently using i3). The PC actually hasn’t been rebooted since installing 14.04; I only restarted X.
matt@work:/var/log/installer$ ls -l
total 1532
-rw------- 1 root root 1303 Nov 20 2012 casper.log
-rw------- 1 root root 3856 Nov 20 2012 debug
-rw-r--r-- 1 root root 420116 Nov 20 2012 initial-status.gz
-rw-r--r-- 1 root root 59 Nov 20 2012 media-info
-rw------- 1 root root 747016 Nov 20 2012 partman
-rw------- 1 syslog adm 377562 Nov 20 2012 syslog
-rw------- 1 root root 17 Nov 20 2012 version
matt@work:/var/log/installer$ sudo cat media-info
Ubuntu 12.10 "Quantal Quetzal" - Release amd64 (20121017.5)
matt@work:/var/log/installer$
This script will find all .rar files in a given directory and extract them into the specified directory. Requires the unrar binary installed in PATH.
usage: unrar-recursive.sh [directory to scan] [extraction destination dir]
Running the script without arguments will search for all .rar files in the current directory and subdirectories and extract them all to the current working directory.
Running the script with only 1 argument will search for all .rar files in the specified directory and extract them into the same directory.
#!/bin/bashif["$#"-eq0]; thendest=$1; elsedest=$2; fifind$1-type f -iname"*.rar"-exec unrar e {}$dest \;
#!/bin/bash
if [ "$#" -eq 0 ]; then dest=$1; else dest=$2; fi
find $1 -type f -iname "*.rar" -exec unrar e {} $dest \;
I have a few old reliable AR5005G (5212, PCI 168c:0013) cards, both MiniPCI and Cardbus from the heyday of Wireless-G back in the early ’00s. Back then, Atheros 500x series cards were THE cards to have if you wanted to have some fun with aircrack, or if you just wanted your WiFi to actually WORK, especially under Linux. The madwifi (aka ath_pci) drivers were probably the most stable wireless drivers at the time. Even on Windows, you could use 3rd party drivers to put the cards in monitor mode and capture packets.
Times have changed and madwifi has been superseded by ath5k (and indirectly ath9k).
For some reason I decided to install Ubuntu 12.04 on an old Fujitsu Lifebook (Pentium III 600MHz, upgraded 512MB RAM, ATI Mobility Radeon M4) without built-in wireless, using a Netgear WPN511 Cardbus adapter. I expected everything to work as it did in the old days, but for some reason the WiFi wouldn’t stay connected.
It seems the hardware encryption capabilites on the card don’t quite support WPA2/CCMP-AES, even though the ath5k driver says it does. So the solution is to disable hardware encryption support:
echo "options ath5k nohwcrypt=1" | sudo tee /etc/modprobe.d/ath5k.conf
Then reboot, or reload the ath5k module (modprobe -rv ath5k, modprobe -v ath5k).
Now I can enjoy my surprisingly not-extremely-slow 10-year old laptop wirelessly.
Opt-out complete; your visits to this website will not be recorded by the Web Analytics tool. Note that if you clear your cookies, delete the opt-out cookie, or if you change computers or Web browsers, you will need to perform the opt-out procedure again.
You may choose to prevent this website from aggregating and analyzing the actions you take here. Doing so will protect your privacy, but will also prevent the owner from learning from your actions and creating a better experience for you and other users.
The tracking opt-out feature requires cookies to be enabled.
Custom theme by me. Based on Panorama by Themocracy
Comments (0) 
Linux, Uncategorized
| 
fix, linux, ubuntu 
Themocracy