• Dandroid@dandroid.app
    link
    fedilink
    arrow-up
    12
    ·
    1 year ago

    I am running into a bug with pipewire where every time my laptop wakes from sleep, my audio devices (both input and output) get duplicated. So if I close my laptop lid and open it 5 times, I will have 5 sets of speakers and 5 microphones. And at that point, the volume controls stop working and everything kinda goes to shit. If I restart the pipewire service, everything starts working fine again.

    The workaround I am using is I made a systemd service that restarts the pipewire service each time the device wakes from sleep. And that works fine. But I was wondering if anyone else ran into this and if there is a better solution available.

    • Laser@feddit.de
      link
      fedilink
      arrow-up
      7
      ·
      1 year ago

      Which version are you currently on? I thought this was in 0.3.77 and fixed in 0.3.78

      • Dandroid@dandroid.app
        link
        fedilink
        arrow-up
        4
        ·
        edit-2
        1 year ago

        Oh. I just checked. I’m on 0.3.79, and yes it’s fixed. I just had to disable my service to check. Thanks for the heads up!

  • AutoTL;DR@lemmings.worldB
    link
    fedilink
    English
    arrow-up
    7
    ·
    1 year ago

    This is the best summary I could come up with:


    Red Hat multimedia expert Wim Taymans has released PipeWire 0.3.80 as the newest version of this open-source solution for efficiently handling video and audio streams on the Linux desktop.

    With PipeWire 0.3.80 its Vulkan SPA plug-in has added DMA-BUF support for efficient handling of buffers.

    The work is part of this merge request that had been open for six months on introducing Vulkan modifiers support for this PipeWire plug-in to enable DMA-BUF usage.

    A set of 22 patches made up this change authored by the wlroots developers.

    PipeWire 0.3.80 also adds a new “Tag” parameter to allow for arbitrary metadata to be transported out-of-band in the graph.

    PipeWire’s echo cancelling code was also ported over to the WebRTC audio processing code, a regression in locating monitor sources by ID were fixed in the PulseAudio server code, and updates to avoid potential crashes.


    The original article contains 167 words, the summary contains 142 words. Saved 15%. I’m a bot and I’m open source!