Ultralite AVB

Talk about your MIDI interfaces, microphones, keyboards...

Moderators: MattKingUSA, khz

shellwalker
Established Member
Posts: 68
Joined: Sun Jan 03, 2021 1:54 pm
Has thanked: 21 times
Been thanked: 11 times

Re: Ultralite AVB

Post by shellwalker »

AudioNarwhal wrote: Wed Nov 09, 2022 5:42 pm

I am still looking into all of this as I have time, but I wanted to give an update.
...
... I think this will result in the best long term solution for Linux users being able to use class compliant devices that rely on this synchronization method.
...

Hi Dylan,
Thanks so much for driving this further and for keeping us updated. In particular I want to honor your mindset of contributing to the overall improvement of Linux for even other CC devices!!!

Let's hope you get that breakthrough on the firmware side of things. I'd be willing to test your firmware if you like. I didn't test your patch because I've been so busy with my paid work (and other stuff) that I simply had no time for patching kernels ☹
But flashing a new MOTU Firmware does not require so much effort, so that I can squeeze in the extra time.

Have you had the chance to think about a test utility that can be used for setting up a loopback and for monitoring against drop-outs/hopping?

MOTU 828mk3, MOTU 828ES, Cakewalk by Bandlab, Kubuntu 22.04.1 LTS, https://github.com/shellwalker-coder/motu_patch_testing

buonamorte.records
Established Member
Posts: 98
Joined: Tue Apr 07, 2020 4:31 am
Has thanked: 11 times
Been thanked: 11 times

Re: Ultralite AVB

Post by buonamorte.records »

Thank you @AudioNarwhal to join this thread, I'm on a tour so I've not so much time to continue testing the patch, but it seems to work...I'm sure that with your help we'll have a fully working devices! Piermarco

baptiste
Established Member
Posts: 26
Joined: Sat Oct 06, 2018 7:18 am
Has thanked: 10 times
Been thanked: 8 times

Re: Ultralite AVB

Post by baptiste »

@AudioNarwhal
I have a non-ESS Ultralite AVB and i compiled a patched vanilla 6.0.3 kernel. I had to fiddle a bit with the patch as part of it was already applied, it seems.

With the latest firmware, (1.4), i got a 16-channel hopping very quickly. I had to unplug and replug the usb to get it to work again correctly.
I flashed 1.3.2+520, which is said to be the latest explicit version (iirc) and things have been good so far (>48h), which is surprising to me because it has never been so stable since i own the interface.

Am i right if i say that we are supposed to test the patch with the latest firmware version ?

User avatar
AudioNarwhal
Established Member
Posts: 39
Joined: Thu Apr 29, 2021 12:43 pm
Has thanked: 2 times
Been thanked: 35 times

Re: Ultralite AVB

Post by AudioNarwhal »

@baptiste
Can you post the changes you had to make to the patch? The patch I shared with @buonamorte.records was based off of the zen-kernel 6.0.3-zen2 which might be slightly different than the 6.0.3 vanilla code, but I would be surprised if any of the patch was already applied.

With the latest firmware, you should see the printk "MOTU Patch" emitted in the kernel log (dmesg) whenever an audio stream starts. Can you also confirm the presence of this log?

While the channel hop could still occur with the patch, I'm a bit surprised you ran into it so quickly, so I want to double check that the patch is truly active.

baptiste wrote: Tue Nov 15, 2022 2:47 pm

Am i right if i say that we are supposed to test the patch with the latest firmware version ?

Yes, the patch should work the latest firmware version.

baptiste
Established Member
Posts: 26
Joined: Sat Oct 06, 2018 7:18 am
Has thanked: 10 times
Been thanked: 8 times

Re: Ultralite AVB

Post by baptiste »

AudioNarwhal wrote: Tue Nov 15, 2022 4:14 pm

@baptiste
Can you post the changes you had to make to the patch? The patch I shared with @buonamorte.records was based off of the zen-kernel 6.0.3-zen2 which might be slightly different than the 6.0.3 vanilla code, but I would be surprised if any of the patch was already applied.

So it might be a good idea to build the zen kernel instead. I remember the second part of the patch wouldn't be applied completely because the line number didn't match. I looked for the code and found it, and IIRC, it was already applied. It's been two weeks since i did it, so i can't remember exactly. Fact is, it was applied finally anyway, i triple checked.

AudioNarwhal wrote: Tue Nov 15, 2022 4:14 pm

With the latest firmware, you should see the printk "MOTU Patch" emitted in the kernel log (dmesg) whenever an audio stream starts. Can you also confirm the presence of this log?

While the channel hop could still occur with the patch, I'm a bit surprised you ran into it so quickly, so I want to double check that the patch is truly active.

I've just reflashed the 1.40 firmware :

Code: Select all

[63425.497116] usb 1-1: new high-speed USB device number 20 using xhci_hcd
[63425.652595] usb 1-1: config 1 interface 6 altsetting 1 has a duplicate endpoint with address 0x9, skipping
[63425.652606] usb 1-1: config 1 interface 7 altsetting 1 has a duplicate endpoint with address 0x87, skipping
[63425.661730] usb 1-1: New USB device found, idVendor=07fd, idProduct=0005, bcdDevice= 2.00
[63425.661740] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[63425.661744] usb 1-1: Product: Ultralite AVB
[63425.661747] usb 1-1: Manufacturer: MOTU
[63425.661750] usb 1-1: SerialNumber: super_secret_serial_number
[63430.549086] usb 1-1: clock source 1 is not valid, cannot use
[63437.973515] usb 1-1: clock source 1 is not valid, cannot use
[63438.793885] usb 1-1: clock source 1 is not valid, cannot use
[63441.263550] MOTU Patch
[63442.106602] usb 1-1: clock source 1 is not valid, cannot use
[63442.929115] usb 1-1: clock source 1 is not valid, cannot use
[63442.943604] usb 1-1: clock source 1 is not valid, cannot use
[63442.957740] usb 1-1: clock source 1 is not valid, cannot use
[63442.974019] usb 1-1: clock source 1 is not valid, cannot use
[63442.988066] usb 1-1: clock source 1 is not valid, cannot use
[63443.002566] usb 1-1: clock source 1 is not valid, cannot use
[63443.017630] usb 1-1: clock source 1 is not valid, cannot use
[63443.034077] usb 1-1: clock source 1 is not valid, cannot use
[63443.048640] usb 1-1: clock source 1 is not valid, cannot use
[63443.062864] usb 1-1: clock source 1 is not valid, cannot use
[63443.077307] usb 1-1: clock source 1 is not valid, cannot use
[63443.094460] usb 1-1: clock source 1 is not valid, cannot use
[63443.109085] usb 1-1: clock source 1 is not valid, cannot use
[63443.123752] usb 1-1: clock source 1 is not valid, cannot use
[63443.138084] usb 1-1: clock source 1 is not valid, cannot use
[63443.154822] usb 1-1: clock source 1 is not valid, cannot use
[63443.169012] usb 1-1: clock source 1 is not valid, cannot use
[63443.183248] usb 1-1: clock source 1 is not valid, cannot use
[63443.203103] usb 1-1: clock source 1 is not valid, cannot use
[63443.220455] usb 1-1: clock source 1 is not valid, cannot use
[63443.234643] usb 1-1: clock source 1 is not valid, cannot use
[63443.249058] usb 1-1: clock source 1 is not valid, cannot use
[63443.263545] usb 1-1: clock source 1 is not valid, cannot use
[63443.280981] usb 1-1: clock source 1 is not valid, cannot use
[63443.295265] usb 1-1: clock source 1 is not valid, cannot use
[63443.309858] usb 1-1: clock source 1 is not valid, cannot use
[63443.324009] usb 1-1: clock source 1 is not valid, cannot use
[63443.340159] usb 1-1: clock source 1 is not valid, cannot use
[63443.356925] usb 1-1: clock source 1 is not valid, cannot use
AudioNarwhal wrote: Tue Nov 15, 2022 4:14 pm
baptiste wrote: Tue Nov 15, 2022 2:47 pm

Am i right if i say that we are supposed to test the patch with the latest firmware version ?

Yes, the patch should work the latest firmware version.

Ok, thanks again !

User avatar
AudioNarwhal
Established Member
Posts: 39
Joined: Thu Apr 29, 2021 12:43 pm
Has thanked: 2 times
Been thanked: 35 times

Re: Ultralite AVB

Post by AudioNarwhal »

@baptiste Thank you for confirming the log!

baptiste wrote: Tue Nov 15, 2022 5:08 pm

So it might be a good idea to build the zen kernel instead. I remember the second part of the patch wouldn't be applied completely because the line number didn't match. I looked for the code and found it, and IIRC, it was already applied. It's been two weeks since i did it, so i can't remember exactly. Fact is, it was applied finally anyway, i triple checked.

I'd like to get a patch that can be applied as-is to the vanilla kernel so that we can tweak it and find out what exactly is going on. I'm not sure the best way to get a copy of the exact source that built you kernel. Do you think you could zip and share the unpatched contents of the sound/usb directory? Or is there a link you could share?

Also, it sounds like some of you are recompiling the whole kernel to apply the patch, which may be easier in some regards, but a whole kernel build can take a bit of time. The snd-usb-audio.ko can be built out of tree and reloaded without even having to reboot the computer. It can be a bit fiddly, but if you already have the source for your running kernel, it isn't too bad and might let us iterate on changes a bit faster. Is this something you are already doing or have tried in the past? Maybe we can make a fork of the snd-usb-audio that we can work on until things seem to be stable.

baptiste
Established Member
Posts: 26
Joined: Sat Oct 06, 2018 7:18 am
Has thanked: 10 times
Been thanked: 8 times

Re: Ultralite AVB

Post by baptiste »

AudioNarwhal wrote: Tue Nov 15, 2022 8:38 pm

@baptiste Thank you for confirming the log!

I'd like to get a patch that can be applied as-is to the vanilla kernel so that we can tweak it and find out what exactly is going on. I'm not sure the best way to get a copy of the exact source that built you kernel. Do you think you could zip and share the unpatched contents of the sound/usb directory? Or is there a link you could share?

the vanilla kernel :
http://cdn.kernel.org/pub/linux/kernel/ ... 0.3.tar.gz

How i build the kernel :

Code: Select all

make mrproper &&
make olddefconfig &&
scripts/config --set-str SYSTEM_TRUSTED_KEYS "" &&
scripts/config --set-str SYSTEM_REVOCATION_KEYS "" &&
scripts/config --undefine DEBUG_INFO &&
scripts/config --undefine DEBUG_INFO_COMPRESSED &&
scripts/config --undefine DEBUG_INFO_REDUCED &&
scripts/config --undefine DEBUG_INFO_SPLIT &&
scripts/config --undefine GDB_SCRIPTS &&
scripts/config --set-val  DEBUG_INFO_DWARF5     n &&
scripts/config --set-val  DEBUG_INFO_NONE       y &&
version_suffix="nodbg-motu" &&

make -j "$(nproc)" bindeb-pkg LOCALVERSION=-"$version_suffix"
#easier debug make bindeb-pkg LOCALVERSION=-"$version_suffix"

sudo make modules_install INSTALL_MOD_STRIP=1
sudo make install

I checked the unpatched source and i actually messed up something when patching : as you could expect, it was not already patched :oops:
unpatched sound/usb : https://nextcloud.nosoco.duckdns.org/s/ySAtMy56yPzwGXY

AudioNarwhal wrote: Tue Nov 15, 2022 8:38 pm

Also, it sounds like some of you are recompiling the whole kernel to apply the patch, which may be easier in some regards, but a whole kernel build can take a bit of time. The snd-usb-audio.ko can be built out of tree and reloaded without even having to reboot the computer. It can be a bit fiddly, but if you already have the source for your running kernel, it isn't too bad and might let us iterate on changes a bit faster. Is this something you are already doing or have tried in the past? Maybe we can make a fork of the snd-usb-audio that we can work on until things seem to be stable.

I think i can do that !

User avatar
AudioNarwhal
Established Member
Posts: 39
Joined: Thu Apr 29, 2021 12:43 pm
Has thanked: 2 times
Been thanked: 35 times

Re: Ultralite AVB

Post by AudioNarwhal »

baptiste wrote: Wed Nov 16, 2022 8:47 am

the vanilla kernel :
http://cdn.kernel.org/pub/linux/kernel/ ... 0.3.tar.gz

baptiste wrote: Wed Nov 16, 2022 8:47 am

I checked the unpatched source and i actually messed up something when patching : as you could expect, it was not already patched
unpatched sound/usb : https://nextcloud.nosoco.duckdns.org/s/ySAtMy56yPzwGXY

Thank you for these!

Here is a patch based off of this source. It looked to me like all the line numbers lined up with the zen kernel, so this should essentially be the same thing except I modified the log to include the string 6.0.3.

https://drive.google.com/file/d/1-00_qF ... sp=sharing

Please let me know if the channel hopping persists.

tseaver
Established Member
Posts: 398
Joined: Mon Mar 13, 2017 6:07 am
Has thanked: 11 times
Been thanked: 98 times

Re: Ultralite AVB

Post by tseaver »

@AudioNarwhal:

I really appreciate your efforts to improve Linux support for the Ultralite AVB device: as a professional FLOSS developer, I know first-hand how important and valuable such interactions are with the user community.

FWIW, although I've never owned any MOTU gear, your work here definitely puts the company on the short-list for my future buying decisions.

Tres.

Ubuntu, Mixbus32C; acoustic blues / country / jazz
shellwalker
Established Member
Posts: 68
Joined: Sun Jan 03, 2021 1:54 pm
Has thanked: 21 times
Been thanked: 11 times

Re: Ultralite AVB

Post by shellwalker »

baptiste wrote: Wed Nov 16, 2022 8:47 am

I think i can do that !

Hi Baptiste,
Would you mind sharing simple step by step instructions for patching and compiling the module out if tree?
Less experienced members could then easily pick that up and give it a try.

Cheers!

MOTU 828mk3, MOTU 828ES, Cakewalk by Bandlab, Kubuntu 22.04.1 LTS, https://github.com/shellwalker-coder/motu_patch_testing

baptiste
Established Member
Posts: 26
Joined: Sat Oct 06, 2018 7:18 am
Has thanked: 10 times
Been thanked: 8 times

Re: Ultralite AVB

Post by baptiste »

After much googling and sweating and swearing :

Code: Select all

[ 6563.884433] MOTU Patch - 6.0.3

I finally built snd-usb-audio for vanilla 6.0.3.
I can provide a download link for the module (x86_64).
I might try to built it for a less recent kernel if the patch can be applied (i run Mint 20.3 with 5.15.0-53)

As for a procedure, it's kinda wonky because i've built kernel 6.0.3 first (less than 90min on my 2015 i7), so i had everything already installed to do this kind of build.
Everything might not be necessary and some parts could be entirely wrong, so feel free to correct :roll: :

  • if you're working with a kernel downloaded from kernel.org, uncompress it in a folder (~/kernel in my case)

  • place the patch into that folder and in a terminal, cd to the source folder (linux-6.0.3 in my case) and do :

    Code: Select all

    patch -p1 < ../snd-usb-audio-printk-6.0.3.patch

    results should be:

    Code: Select all

    patching file sound/usb/card.h
    patching file sound/usb/endpoint.c
  • Now let's prepare and clean the source

    Code: Select all

    make mrproper
  • Code: Select all

    cp /usr/src/linux-headers-$(uname -r)/Module.symvers .
  • Let's build this thing !

    Code: Select all

    make -C /lib/modules/`uname -r`/build M=$PWD
  • Unplug any usb sound interface so we can replace the module

  • Make a backup of the current module :

    Code: Select all

    cd /usr/lib/modules/$(uname -r)/kernel/sound/usb

    Code: Select all

    sudo mv snd-usb-audio.ko snd-usb-audio.ko.bkp
  • Then copy the module we built to where it should be :

    Code: Select all

    sudo cp ~/kernel/linux-6.0.3/sound/usb/snd-usb-audio.ko /usr/lib/modules/$(uname -r)/kernel/sound/usb
  • Now plug your Ultralite and do

    Code: Select all

    dmesg

    you should see

    Code: Select all

    MOTU Patch - 6.0.3
buonamorte.records
Established Member
Posts: 98
Joined: Tue Apr 07, 2020 4:31 am
Has thanked: 11 times
Been thanked: 11 times

Re: Ultralite AVB

Post by buonamorte.records »

@AudioNarwhal
I've found some problems in different contests:

1 - Two days ago I've recorded 2 bands using the ultralite, the avb switch and the stage b-16. In the beginning, only an heavy sound of crackling. Than i turn off the wifi connection and everything worked. I'm not sure about the exact problem, i'll try to redo the same to understand if it was that or not.

2 - Today, i tried to start a project for a sound design on mixbus 32c, 5 tracks with synths with an arturia midi keyboard, 88.200/256 unusable, 88200/512 too, 88200/1024 after a few minutes start crackling and channel-hopping. I tried with and without wifi connection and with/without a bluetooth mouse. It seems like something happen with the midi note. :roll:

I hope to solve this problem, i've a week to send the track for an exposition...

Tonight i'll make other experiments, trying to understand the exact behavior of the problem. Can i make some test that can help you in understanding this problems?

User avatar
AudioNarwhal
Established Member
Posts: 39
Joined: Thu Apr 29, 2021 12:43 pm
Has thanked: 2 times
Been thanked: 35 times

Re: Ultralite AVB

Post by AudioNarwhal »

buonamorte.records wrote: Fri Nov 18, 2022 6:09 pm

1 - Two days ago I've recorded 2 bands using the ultralite, the avb switch and the stage b-16. In the beginning, only an heavy sound of crackling. Than i turn off the wifi connection and everything worked. I'm not sure about the exact problem, i'll try to redo the same to understand if it was that or not.

Thank you for the feedback, I have a few questions that might help me understand the setup better.

  1. Was the crackling on input (capture) or output (playback) or both?

  2. If you were passing audio over AVB, how was the clocking configured, and which device was connected to the PC?

  3. What sampling rate were you running at?

buonamorte.records wrote: Fri Nov 18, 2022 6:09 pm

2 - Today, i tried to start a project for a sound design on mixbus 32c, 5 tracks with synths with an arturia midi keyboard, 88.200/256 unusable, 88200/512 too, 88200/1024 after a few minutes start crackling and channel-hopping. I tried with and without wifi connection and with/without a bluetooth mouse. It seems like something happen with the midi note. :roll:

It would be helpful to see some of kernel logs that are emitted (if any are) when the crackling and channel-hopping start to occur. Unfortunately, all the useful logs are hidden in the dynamic debug feature of the kernel. I'll attempt to explain how to enable them, although any corrections are welcome.

  • First things first, lets make sure the kernel was actually built with dynamic debug support. To check we need to see if CONFIG_DYNAMIC_DEBUG was enabled when the kernel was built. You should be able to check that by running the following command:

    Code: Select all

    cat /boot/config-$(uname -r) | grep DYNAMIC_DEBUG

    My kernel was built with this support, so the output looks like this:

    Code: Select all

    CONFIG_DYNAMIC_DEBUG=y
    CONFIG_DYNAMIC_DEBUG_CORE=y
    

    If you see "CONFIG_DYNAMIC_DEBUG=y" then you should good to go. If it isn't set, you'll need to enable this feature and rebuild the kernel.

  • Next, lets make sure debugfs is mounted. Ubuntu seems to mount this by default, but this isn't the case for every distribution. To see if it is already mounted, you can run this command:

    Code: Select all

    mount | grep debugfs

    Since I am on Ubuntu, debugfs was already mounted at /sys/kernel/debug. This is confirmed by the following output:

    Code: Select all

    debugfs on /sys/kernel/debug type debugfs (rw,nosuid,nodev,noexec,relatime)
  • If debugfs is not mounted, you'll need to mount it. That can be done with the following command:

    Code: Select all

    sudo mount -t debugfs none /sys/kernel/debug

    To check that its mounted, you can run the command from the previous step.

  • Once debugfs is mounted, we can do some sanity checks to make sure the filesystem looks correct. If you run:

    Code: Select all

    sudo ls /sys/kernel/debug

    You should see that this directory contains a folder called dynamic_debug. Here is the directory listing output on my computer:

    Code: Select all

    acpi		 devfreq	   dynamic_debug       hid		 kvm		   pm_genpd    slab		 ttm
    bdi		 device_component  energy_model        ieee80211	 mce		   pwm	       sleep_time	 usb
    block		 devices_deferred  error_injection     intel_powerclamp  mei0		   ras	       split_huge_pages  virtio-ports
    cec		 dma_buf	   extfrag	       interconnect	 mtd		   regmap      suspend_stats	 wakeup_sources
    cleancache	 dmaengine	   fault_around_bytes  iosf_sb		 opp		   regulator   swiotlb		 x86
    clear_warn_once  dma_pools	   frontswap	       kfence		 pinctrl	   remoteproc  sync		 zswap
    clk		 dri		   gpio		       kprobes		 pkg_temp_thermal  sched       tracing
    
  • Now we can enable the dynamic debug logging for the snd-usb-audio module. I do this with the following command:

    Code: Select all

    sudo sh -c 'echo "module snd_usb_audio +p" > /sys/kernel/debug/dynamic_debug/control'

    Basically, you need to write the string "module snd_usb_audio +p" into the file "/sys/kernel/debug/dynamic_debug/control" as root. The "sudo sh -c", is a little trick I use to do that without having to switch user to root, but there may be other ways to do this. In my experience, if snd-usb-audio is ever removed and re-inserted, this needs to be written again to re-enable the dynamic logs for the module.

  • Next is to enable the logs for the xhci_hcd module. I suspect these will be the more interesting logs. I do this with the following command:

    Code: Select all

    sudo sh -c 'echo "module xhci_hcd +p" > /sys/kernel/debug/dynamic_debug/control'
  • For me, this immediately makes the kernel log very noisy, as there are logs emitted for every short bulk packet received. Those logs look like this:

    Code: Select all

    [26923.879607] xhci_hcd 0000:00:14.0: ep 0x83 - asked for 64 bytes, 48 bytes untransferred
    [26923.880252] xhci_hcd 0000:00:14.0: ep 0x83 - asked for 64 bytes, 48 bytes untransferred
    [26923.880756] xhci_hcd 0000:00:14.0: ep 0x83 - asked for 64 bytes, 28 bytes untransferred
    [26923.881232] xhci_hcd 0000:00:14.0: ep 0x83 - asked for 64 bytes, 44 bytes untransferred
    [26923.881612] xhci_hcd 0000:00:14.0: ep 0x83 - asked for 64 bytes, 44 bytes untransferred
    [26923.888433] xhci_hcd 0000:00:14.0: ep 0x82 - asked for 16384 bytes, 16056 bytes untransferred
    

    Short bulk packets are a normal part of bulk operation, so they are probably unrelated to any issues we are seeing. We can silence these by removing the specific line number generating these logs from the dynamic debug configuration.

  • To figure out the line number generating these logs, the following command can be used:

    Code: Select all

    sudo cat /sys/kernel/debug/dynamic_debug/control | grep "asked for"

    The output of this should reveal the offending line numbers:

    Code: Select all

    /home/dylan/linux-5.15.0/drivers/usb/host/xhci-ring.c:2538 [xhci_hcd]process_bulk_intr_td =p "ep %#x - asked for %d bytes, %d bytes untransferred\012"
    /home/dylan/linux-5.15.0/drivers/usb/host/xhci-ring.c:2531 [xhci_hcd]process_bulk_intr_td =p "ep %#x - asked for %d bytes, %d bytes untransferred\012"
    

    Here, you can see that line numbers 2538 and 2531 in the file xhci-ring.c contains that string. To silence these messages, remove them from the dynamic debug configuration. Please note the specific line numbers for your kernel source as they may be different from mine. The following two commands should remove these logs:

    Code: Select all

    sudo sh -c 'echo "file xhci-ring.c line 2538 -p" > /sys/kernel/debug/dynamic_debug/control'
    sudo sh -c 'echo "file xhci-ring.c line 2531 -p" > /sys/kernel/debug/dynamic_debug/control'
    
  • Finally, you can live-view the kernel logs while trying to reproduce the problem. To watch the kernel logs I use:

    Code: Select all

    sudo dmesg -w

    Once the issue starts to occur, I hope there will be some corresponding interesting log. Let me know what you see!

baptiste
Established Member
Posts: 26
Joined: Sat Oct 06, 2018 7:18 am
Has thanked: 10 times
Been thanked: 8 times

Re: Ultralite AVB

Post by baptiste »

@AudioNarwhal thanks for the debug info !

My kernel is built with debug enabled and the ultralite decided to play tricks this morning, so here we go :

Code: Select all

[83553.286633] xhci_hcd 0000:00:14.0: Babble error for slot 6 ep 14 on endpoint
[83559.521379] xhci_hcd 0000:00:14.0: Babble error for slot 6 ep 14 on endpoint

I was quite lucky to watch this happen as i was watching the logs and a video on youtube with firefox/pulseaudio, so i can describe the events.

  • at first line, sound was quickly decimated, and i have sound to outputs 1 to 8 and 17 to 24

  • then sound becomes normal again and i have sound to outputs to 17 to 24 only

  • then i can have a few more identical log lines, i may hear a loud click or nothing

  • then it goes on, decimated and sound to 9 to 16 and 17 to 24, then normal and sound to 9 to 16 only, etc.

buonamorte.records
Established Member
Posts: 98
Joined: Tue Apr 07, 2020 4:31 am
Has thanked: 11 times
Been thanked: 11 times

Re: Ultralite AVB

Post by buonamorte.records »

Hi @AudioNarwhal , when I mount debugfs and I check if it is mounted, it gives me

Code: Select all

none on /sys/kernel/debug type debugfs (rw,relatime)

Code: Select all

Was the crackling on input (capture) or output (playback) or both?

Mostly on playback, the recordings are good.

Code: Select all

If you were passing audio over AVB, how was the clocking configured, and which device was connected to the PC?

I was using ultralite avb ess as clock master connected to PC

Code: Select all

What sampling rate were you running at?

88.200

Post Reply