Skip to content

Latest commit

 

History

History
109 lines (87 loc) · 5.5 KB

README.md

File metadata and controls

109 lines (87 loc) · 5.5 KB
Raw

Configuring and running CamillaDSP

Running as a systemd service under Fedora 31 and newer

This is a way to run CamillaDSP as a systemd service to provide system wide filtering. This uses the Alsa backend for both capture and playback. These steps work on Fedora 31 and newer. Other distributions are probably similar, but there are likely some differences. If you try on another distribution and have to do something differently, please let me know and I will add that here.

Step 1: Install CamillaDSP to /usr/local/bin

  • Install the CamillaDSP binary:
    • Option 1: download a pre-build binary from releases and copy it to /usr/local/bin/.
    • Option 2: clone the CamillaDSP repo and install with sudo cargo install --path . --root /usr/local/

Step 2: Load loopback driver on boot

  • Copy aloop.conf to /etc/modules-load.d/aloop.conf
  • Reboot or load it manually with sudo modprobe snd-aloop
  • This creates a loopback device, so sound sent to "hw:Loopback,0" can be captured from "hw:Loopback,1".

Step 3: Add an Alsa "plug" for sample format conversion etc

This creates a "plug" device named "camilladsp" that sends its output to the Loopback device.

  • Copy asound.conf to /etc/asound.conf

Step 4: Create the CamillaDSP configuration

  • Use Alsa for both playback and capture, see alsaconfig.yml for an example.
  • Use "hw:Loopback,1" for capture device.
  • Set the capture sample format and rate to the same as for the plug device
  • Set the playback device to your dac, and the best sample format it supports.

Step 4: Set up systemd service for CamillaDSP

  • Edit as needed and copy camilladsp.service to /etc/systemd/system/camilladsp.service.
  • Reload the config: sudo systemctl daemon-reload
  • Start the service: sudo systemctl start camilladsp
  • Check the service status: sudo systemctl status camilladsp
  • If all ok, enable the service so it runs automatically: sudo systemctl enable camilladsp
  • Optionally for the GUI, repeat this procedure for the camillagui.service file.

Step 5: (Optional,for usb dac) Add udev rule to start service when dac is connected

  • Figure out the vendor id of the dac using udevadm. Run udevadm monitor --subsystem-match=usb --property --udev and plug in the dac. There should be a few pages of output (see udevadm_output.txt for an example).
  • Look for the vendor id of the dac. In the example it's "20b1".
  • Edit 90-camilladsp.rules and replace the id with the one for your dac.
  • Copy the file to /etc/udev/rules.d/90-camilladsp.rules
  • Reload the udev config: sudo udevadm control --reload-rules
  • Unplug the dac, wait a few seconds and plug it back in. Then check that the service started: sudo systemctl status camilladsp

Step 6: (optional) Send desktop audio to CamillaDSP

Follow either the PulseAudio or the Pipewire steps.

To check what your system uses, run pactl info in a terminal and look at the line starting with "Server Name":

> pactl info
...
Server Name: PulseAudio (on PipeWire 0.3.30)
...

In the above output, Pipewire is in use. If Pulseaudio is used by the system, the server name will normally be "pulseaudio" only.

PulseAudio

  • Open /etc/pulse/default.pa in a text editor (see default.pa for an example).
  • Add these two lines (lines 43-44 in example): 
    load-module module-alsa-sink device="camilladsp" sink_name="CamillaDSP"
update-sink-proplist CamillaDSP device.description=CamillaDSP
  • Comment out the autodetection module (lines 47-54 in example). This stops PulseAudio from finding and trying to use the Loopback and dac directly.
  • Set camilladsp as the default sink (line 152): set-default-sink camilladsp
  • Save the file.

Pipewire

  • Open ~/.config/pipewire/pipewire.conf in a text editor. If the file doesn't exist, create it by copying the template from: /usr/share/pipewire/pipewire.conf

  • Under context.properties, set the desired sample rate:

        default.clock.rate        = 44100

  • Add this block under context.objects:

        {   factory = adapter
        args = {
            factory.name            = api.alsa.pcm.sink
            node.name               = "alsa-sink"
            node.description        = "Alsa Loopback"
            media.class             = "Audio/Sink"
            api.alsa.path           = "hw:Loopback,0,0"
            #api.alsa.period-size   = 1024
            #api.alsa.headroom      = 0
            #api.alsa.disable-mmap  = false
            #api.alsa.disable-batch = false
            audio.format           = "S32LE"
            audio.rate             = 44100
            audio.channels         = 2
            #audio.position         = "FL,FR"
        }
    }

  • Set api.alsa.path to the name of the loopback to output to.

  • Set audio.rate to match the rate set above.

  • If unsure, compare with the pipewire.conf in this repository.

  • Save the file.

  • After reboot, the audio settings should show a new output device called "Alsa Loopback".

Step 7: Reboot to verify that everything starts ok

Other files included

  • asound_multidevice.conf: An example Alsa configuration for combining two stereo dacs into a 4-channel one. This only works well if the two dacs share the same clock. Using two synchronous USB dacs connected to the same USB bus should work, as they will both lock their sample rate to the USB bus clock.