fxthomas commented 3 years ago

I had this issue while trying to debug https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/530 on an HP Detachable x2 ultrabook, where we're trying to see if the SOF driver is somehow more stable than SST.

This issue happens with raw ALSA (neither Pipewire/Pulseaudio running) as well as with both PW/PA.

The following sequence reproduces the issue 100% of the time:

Enable SOF with options snd-intel-dspcfg dsp_driver=3 on a 5.11 kernel
Check that the SOF driver is actually used with dmesg | grep sof
Play a relatively long sound file for a few minutes (aplay -D <device name> file.wav for example)
After sometimes (usually 2-3 minutes, though not always), speakers/headphones stop playing anything and emit a loud persistent "beeeep" that doesn't stop

As far as I can tell the beep only goes away when starting (if running raw ALSA) or restarting (if already running) the Pipewire/Pulseaudio server, which restores proper playback until the next beep.

The default SST driver is stable for basic audio output in S16LE, despite issues in other areas.

Environment 1) Branch name and commit hash of the 2 repositories: sof (firmware/topology) and linux (kernel driver).

Kernel: v5.11-arch2 in testing
SOF: 1.6.1-1 (3dffc2ae on the Arch package source, sof-bin@b77c851) 2) Name of the topology file
Topology: ? (I don't know how to get that) 3) Name of the platform(s) on which the bug is observed.
Platform: Atom x5-Z8530 (Cherrytrail) / rt5640

Logs:

There are no logs in dmesg at all
Regarding sof-logger, is there a way for me to compile it without bringing up the whole toolchain?

I'm not familiar at all with SOF, so if this is the wrong place for a report do let me know!

plbossart commented 3 years ago

Copying everyone who has a BYT/CHT device or worked on DMAs before: I have no explanation for the change in behavior with the trace

@lgirdwood @keyonjie @kv2019i @bardliao @jwrdegoede @ujfalusi @lyakh

plbossart commented 3 years ago

This issue is not related to a specific codec as mentioned in the title, this also happens with a WM5102 https://github.com/thesofproject/linux/issues/2485#issuecomment-858246382

Clearly an SOF issue.

fxthomas commented 3 years ago

In a way that's good news, this means it'll be somewhat easier to reproduce!

These are just idle hypotheses, and I'm probably wrong, but could it be that either:

some part of the code/firmware is causing the issue by sometimes writing a bit too far outside of their allowed range, which becomes less likely when sof_debug reserves some additional memory regions, or
the issue is a race that's highly dependent on timing and the activation of firmware traces makes it far less likely?

Can I test for memory corruption, e.g. by adding padding/guard data like we do in userland? Any other idea?

lgirdwood commented 3 years ago

@fxthomas I suspect its a DMA issue, we need to do this DMA channel mapping (we have first come first served today on CHT).

        /* Init DMAC1 for Mem2Mem Transfers */
        /* 
         *    Channel 0 - Unused
         *    Channel 1 - Unused
         *    Channels 2-7 - Mem to Mem 
         */
       /* Init DMAC0 for Mem2Phe and Phe2Mem Transfers 
        /* 
         *    Channel 0 - SSP2 to Mem - Src_Id  = 0
         *    Channel 1 - Mem to SSP2 - Dest_Id = 5
         *    Channels 2-7 - Unused
         */

carlos-ferreira commented 3 years ago

Just wanted to confirm that i experienced the same issue on my HP Detachable x2 p000np on Arch Linux Kernel 5.12.11-arch1-1. I can also confirm that adding options snd_sof sof_debug=1 to /etc/modprobe.d/sof-debug.conf solves the issue for me.

marc-hb commented 3 years ago

Linking to another, "stuck DMA" issue, sorry if not related #4333

a5a5aa555oo commented 3 years ago

Having the same issue and adding the following module parameters doesn't solve it.

$ cat /etc/modprobe.d/snd* blacklist snd_hdmi_lpe_audio options snd-intel-dspcfg dsp_driver=3 options snd_sof sof_debug=1 $ cat /sys/module/snd_sof/parameters/sof_debug 1 $ cat /sys/module/snd_intel_dspcfg/parameters/dsp_driver 3

My tablet notebook is Genuine E10T3W with a ATOM Z3735F CPU and ALC5642 audio chip. kernel: 5.13.6-arch1-1 sof-firmware: 1.8-1

lamyseba commented 3 years ago

Hi, just to say I'm experiencing the same bug on a surface3, with a Kernel : Linux surface3 5.13.7-arch1-1-surface with an rt5645 card. sof-firmware: 1.8-1

plbossart commented 3 years ago

Hi, just to say I'm experiencing the same bug on a surface3, with a Kernel : Linux surface3 5.13.7-arch1-1-surface with an rt5645 card. sof-firmware: 1.8-1

@lamyseba how did you end-up using SOF? is the CONFIG_SND_INTEL_BYT_PREFER_SOF option set, or did you use options snd-intel-dspcfg dsp_driver=3

What happens if you use a) legacy driver options snd-intel-dspcfg dsp_driver=1

b) SOF with trace enabled options snd-intel-dspcfg dsp_driver=3 options snd_sof sof_debug=1

lamyseba commented 3 years ago

@plbossart thank your for your answer. I'm just an average user, so my linux start with the default options set by manjaro at installation time. In fact I didn't even know how to change this options.

But when looking for some answer to the bug I experienced, I found this page, that said:

On newer kernels (after 5.6 or so) you may need the sof-firmware package to make the correct audio device appear. Try installing it if the rt5645 device is missing.

The SOF driver may cause an issue where the audio output is replaced by a single tone, see this bug report. It can possibly be resolved by changing back to the older HDA driver using the kernel option 'snd_intel_dspcfg.dsp_driver=2' (see the Fedora wiki).

Reading this, I realized that it was probably a kernel update that made the bug appear. This seem to be confirmed, as switching back to the LTS kernel 4.19 made the audio bug disappear. This workaround using a LTS kernel is ok for me, but if it helps solve the bug on newer kernel, I tried your suggestions on the current kernel (5.13.10-arch1-1-surface). Here are the results. As I told I'm not used to change kernel options, not sure I did this right, so I put the output of cat /proc/cmdline for you to check if I did it the right way.

Config solving the bug with LTS kernel

$ cat /proc/cmdline

BOOT_IMAGE=/boot/vmlinuz-linux-surface-lts root=UUID=9991d280-9ba2-481b-b8c5-b4444195ef8c rw quiet udev.log_priority=3

The command sudo sysctl -a listed no options set for snd-intel-dspcfg

With legacy drivers and kernel 5.13.10-arch1-1-surface

$ cat /proc/cmdline

BOOT_IMAGE=/boot/vmlinuz-linux-surface root=UUID=9991d280-9ba2-481b-b8c5-b4444195ef8c rw quiet udev.log_priority=3 snd-intel-dspcfg dsp_driver=1

The bug is still there. And sudo sysctl -a still gives me no output for snd-intel-dspcfg

With SOF trace enabled and kernel 5.13.10-arch1-1-surface

$ cat /proc/cmdline

BOOT_IMAGE=/boot/vmlinuz-linux-surface root=UUID=9991d280-9ba2-481b-b8c5-b4444195ef8c rw quiet udev.log_priority=3 snd-intel-dspcfg dsp_driver=3 snd_sof sof_debug=1

The bug is still there. And sudo sysctl -a still gives me no output for snd-intel-dspcfg

nrndda commented 3 years ago

Also experiencing this bug. After beeping starts video stream freezes. Couple of seconds after pausing video playback beeping stops and I can resume normal playback. Of course beeping starts again after couple of minutes of playback.

DMI: HP HP Pro Tablet 608 G1/8042, BIOS N70 Ver. 01.21 8/20/2019 Kernel: 5.14.0-rc6-g7c60610d4767-dirty with builtin snd_sof without snd_sst.

kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware info: version 1:8:0-9e7a8 kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware: ABI 3:18:1 Kernel ABI 3:18:0 kernel: sof-audio-acpi-intel-byt 808622A8:00: unknown sof_ext_man header type 3 size 0x30 kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware info: version 1:8:0-9e7a8 kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware: ABI 3:18:1 Kernel ABI 3:18:0 kernel: sof-audio-acpi-intel-byt 808622A8:00: Topology: ABI 3:18:1 Kernel ABI 3:18:0 kernel: sof-audio-acpi-intel-byt 808622A8:00: ASoC: Parent card not yet available, widget card binding deferred kernel: input: sof-bytcht cx2072x Headset as /devices/pci0000:00/808622A8:00/bytcht_cx2072x/sound/card0/input6

Setting snd_sof.sof_debug=1 in cmdline prevents beeping so far.

Semperverus commented 3 years ago

Writing to share that I am also experiencing this with sof-firmware 1.8-1 on 5.13.10-arch1-1 with sound card rt5672 [sof-bytcht rt5672].

My device is a Dell Venue 10 Pro 5055 A08.

Shen-long87 commented 3 years ago

Hello. I have a surface rt 3 model 1645 tablet. With ubuntu mate installed on 21.04. Recently updated the kernel on your converted 5.13.13 is the same bug with pip at some time. After checking what the sof-bytcht rt5645.

hdwiLpeaudio intel hdmi / dp lpe audio is not visible.

Too. There is a problem with the hardware setting to "1output / 2inputs -play hifi quality miusic" and the output tab "Headphones + stereo speaker built-in".

I installed alsamixer v1.2.4 to check the drivers and which driver is detected for the hardware.

What I noticed is that the "pip temporary interrupt error" music play signal starts at kernel 5.11.31 and continues into kernel 5.14.

This is the first time I am writing here on the github forum.

marc-hb commented 3 years ago

The command sudo sysctl -a listed no options set for snd-intel-dspcfg

Run-time sysctl is different and unrelated to load-time module parameters. It's possible to make both point at the same setting and some drivers do that but others don't.

You need grep -H ^ /sys/module/snd_intel_dspcfg/parameters/*

grep -H ^ /sys/module/*sof*/parameters/* is also useful, see https://github.com/thesofproject/sof-test/commit/cbb42e7bd19e

plbossart commented 3 years ago

Sorry @Shen-long87. I am not able to understand your problem report, it could be a formatting issue or a translation issue. Please open a new issue unless you are experiencing the 'persistent beeping sound' listed in the subject of this thread.

Shen-long87 commented 3 years ago

I can see that something translated wrong. Yes, I have a problem with the above topic, annoying sudden sound. Caused by selecting hardware settings: Too. There is a problem with the hardware setting to "1output / 2inputs -play hifi quality miusic" and the output tab "Headphones + stereo speaker built-in".

I have Ubuntu mate 21.04 installed with kernel 5.13.13 surface. As I mentioned: ms surface rt3 1645

plbossart commented 3 years ago

I can see that something translated wrong. Yes, I have a problem with the above topic, annoying sudden sound. Caused by selecting hardware settings: Too. There is a problem with the hardware setting to "1output / 2inputs -play hifi quality miusic" and the output tab "Headphones + stereo speaker built-in".

I have Ubuntu mate 21.04 installed with kernel 5.13.13 surface. As I mentioned: ms surface rt3 1645

Does the workaround mentioned earlier work for you: add this to /etc/modprobe.d/alsa-base.conf

options snd_sof sof_debug=1

Shen-long87 commented 3 years ago

I think there is something else. I'm sending the file. Today I only had time to write back. Thank you very much for help. Therefore, WiFi works very well with this kernelm 5.13.13. Without breaking the connection. Now after 10 x Linux Ubuntu mate started working.

Shen-long87 commented 3 years ago

# autoloader aliases
install sound-slot-0 /sbin/modprobe snd-card-0
install sound-slot-1 /sbin/modprobe snd-card-1
install sound-slot-2 /sbin/modprobe snd-card-2
install sound-slot-3 /sbin/modprobe snd-card-3
install sound-slot-4 /sbin/modprobe snd-card-4
install sound-slot-5 /sbin/modprobe snd-card-5
install sound-slot-6 /sbin/modprobe snd-card-6
install sound-slot-7 /sbin/modprobe snd-card-7

# Cause optional modules to be loaded above generic modules
install snd /sbin/modprobe --ignore-install snd $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist snd-ioctl32 ; /sbin/modprobe --quiet --use-blacklist snd-seq ; }
#
# Workaround at bug #499695 (reverted in Ubuntu see LP #319505)
install snd-pcm /sbin/modprobe --ignore-install snd-pcm $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist snd-pcm-oss ; : ; }
install snd-mixer /sbin/modprobe --ignore-install snd-mixer $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist snd-mixer-oss ; : ; }
install snd-seq /sbin/modprobe --ignore-install snd-seq $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist snd-seq-midi ; /sbin/modprobe --quiet --use-blacklist snd-seq-oss ; : ; }
#
install snd-rawmidi /sbin/modprobe --ignore-install snd-rawmidi $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist snd-seq-midi ; : ; }
# Cause optional modules to be loaded above sound card driver modules
install snd-emu10k1 /sbin/modprobe --ignore-install snd-emu10k1 $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist snd-emu10k1-synth ; }
install snd-via82xx /sbin/modprobe --ignore-install snd-via82xx $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist snd-seq ; }

# Load saa7134-alsa instead of saa7134 (which gets dragged in by it anyway)
install saa7134 /sbin/modprobe --ignore-install saa7134 $CMDLINE_OPTS && { /sbin/modprobe --quiet --use-blacklist saa7134-alsa ; : ; }
# Prevent abnormal drivers from grabbing index 0
options bt87x index=-2
options cx88_alsa index=-2
options saa7134-alsa index=-2
options snd-atiixp-modem index=-2
options snd-intel8x0m index=-2
options snd-via82xx-modem index=-2
options snd-usb-audio index=-2
options snd-usb-caiaq index=-2
options snd-usb-ua101 index=-2
options snd-usb-us122l index=-2
options snd-usb-usx2y index=-2
# Ubuntu #62691, enable MPU for snd-cmipci
options snd-cmipci mpu_port=0x330 fm_port=0x388
# Keep snd-pcsp from being loaded as first soundcard
options snd-pcsp index=-2
# Keep snd-usb-audio from beeing loaded as first soundcard
options snd-usb-audio index=-2

lgirdwood commented 3 years ago

Please see https://github.com/thesofproject/sof/pull/4775. This aligns the DMA mappings to the closed source FW, but I have no BYT/CHT HW so cant validate. Also attached GCC built binaries for BYT and CHT sof-cht.zip sof-byt.zip .

Shen-long87 commented 3 years ago

I'm just wondering what to do with this. Forgive me. I read link 4775 what exactly is happening. I will be monitoring the progress.

lgirdwood commented 3 years ago

I'm just wondering what to do with this. Forgive me. I read link 4775 what exactly is happening. I will be monitoring the progress.

Thanks, let me know if you can reproduce with the binaries I've attached above. These will need unzipped and copied to the SOF firmware directory (please rename the old FW file as you may need to restore it). To test you will need to reload the SOF driver or reboot.

nrndda commented 3 years ago

Replaced /lib/firmware/intel/sof/sof-cht.ri, removed sof_debug=1 and stil reproducing. Same beep but a bit loader and earlier. Kernel: 5.14.6. sof-cht

kernel: sof-audio-acpi-intel-byt 808622A8:00: unknown sof_ext_man header type 3 size 0x30 kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware info: version 1:9:0-24391 kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware: ABI 3:19:0 Kernel ABI 3:18:0 kernel: sof-audio-acpi-intel-byt 808622A8:00: warn: FW ABI is more recent than kernel

Shen-long87 commented 3 years ago

Hello everyone. Isn't it better to use the audio settings from kernel 5.8? this error did not occur.

lgirdwood commented 3 years ago

Replaced /lib/firmware/intel/sof/sof-cht.ri, removed sof_debug=1 and stil reproducing. Same beep but a bit loader and earlier. Kernel: 5.14.6. sof-cht

kernel: sof-audio-acpi-intel-byt 808622A8:00: unknown sof_ext_man header type 3 size 0x30 kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware info: version 1:9:0-24391 kernel: sof-audio-acpi-intel-byt 808622A8:00: Firmware: ABI 3:19:0 Kernel ABI 3:18:0 kernel: sof-audio-acpi-intel-byt 808622A8:00: warn: FW ABI is more recent than kernel

Thanks, it looks like we may need the debug atm. Btw, v1.9 will have this update but it will be built with XCC as GCC builds are slow and this can also introduce issues (so we should retest with debug disabled with v1.9 next week).

Pandemonious commented 3 years ago

hello. I am running arch kernel 5.14.8-arch1-1 on a Dell Chromebook 11 3180.

I am experiencing the same issue with the loud ringing noise after a few minutes of audio playback. Causes the video stream (if there is one) to freeze up as well. Sometimes it happens without anything else happening but I will say that it usually happens when I am moving around different programs and working the CPU.

I am using the sof-firmware and alsa-ucm-conf packages. Without them, the audio device is not detected in the OS by KDE or by alsa-utils, and no audio works at all. with my other normal laptop, I didn't need the sof firmware at all, just pulseaudio and pulseaudio-alsa.

I previously had Debian 11 non-free installed on here and the audio worked just fine, not even the cracking issue.

I am quite invested in the idea that this is a kernel issue

not a linux expert and kinda new to the whole scene so idk what else you guys need but I am happy to help in any way I can

plbossart commented 3 years ago

hello. I am running arch kernel 5.14.8-arch1-1 on a Dell Chromebook 11 3180.

I am experiencing the same issue with the loud ringing noise after a few minutes of audio playback. Causes the video stream (if there is one) to freeze up as well. Sometimes it happens without anything else happening but I will say that it usually happens when I am moving around different programs and working the CPU.

@Pandemonious same recommendation as above: add this to /etc/modprobe.d/alsa-base.conf

options snd_sof sof_debug=1

Shen-long87 commented 3 years ago

Hello everyone! Today I had time to investigate the above subject of the problem. I am an average Linux user from the Ubuntu / debian family. Sorry for using translator, I don't use English every day.

I have provided information from the file /etc/modprobe.d/alsa-base.conf

in the file, these are the settings for searching for outgoing audio calls.

autoloader aliases: install sound-slot-0 /sbin/modprobe snd-card-0 install sound-slot-1 /sbin/modprobe snd-card-1 install sound-slot-2 /sbin/modprobe snd-card-2 install sound-slot-3 /sbin/modprobe snd-card-3 install sound-slot-4 /sbin/modprobe snd-card-4 install sound-slot-5 /sbin/modprobe snd-card-5 install sound-slot-6 /sbin/modprobe snd-card-6 install sound-slot-7 /sbin/modprobe snd-card-7

i set it to = -3 this is the output from my audio: options bt87x index=-3 options cx88_alsa index=-3 options saa7134-alsa index=-3 options snd-atiixp-modem index=-3 options snd-intel8x0m index=-3 options snd-via82xx-modem index=-3 options snd-usb-audio index=-3 options snd-usb-caiaq index=-3 options snd-usb-ua101 index=-3 options snd-usb-us122l index=-3 options snd-usb-usx2y index=-3

Ubuntu #62691, enable MPU for snd-cmipci

options snd-cmipci mpu_port=0x330 fm_port=0x388

Keep snd-pcsp from being loaded as first soundcard

options snd-pcsp index=-3

Keep snd-usb-audio from beeing loaded as first soundcard

options snd-usb-audio index=-3 options snd-sof sof_debug=-3

I entered the folder as administrato: /lib/firmware/intel/sof/ swapped two files from: lgirdwood sof-cht.zip sof-byt.zip i advise you to make a copy of these files there is a difference in kb.

I opened the sound and there was no sound error.

Shen-long87 commented 3 years ago

dzień dobry. Używam jądra arch 5.14.8-arch1-1 na Dell Chromebook 11 3180. Ten sam problem występuje z głośnym dzwonkiem po kilku minutach odtwarzania dźwięku. Powoduje również zamrożenie strumienia wideo (jeśli taki istnieje). Czasami dzieje się to bez niczego innego, ale powiem, że zwykle dzieje się tak, gdy poruszam się po różnych programach i pracuję na procesorze.

@Pandemonious to samo zalecenie co powyżej: dodaj to do /etc/modprobe.d/alsa-base.conf

opcje snd_sof sof_debug=1

I would like to thank you for your help. I have a little problem with changing the card hard in the settings

I changed the settings in the added command: options snd_sof sof_debug = 1 on snd-sof options sof_debug = -3

plbossart commented 3 years ago

options snd-sof sof_debug=-3

@Shen-long87 That's not a valid value, this value is a bitmap. This looks like a copy-paste mistake

Shen-long87 commented 3 years ago

yet check it out if it crashes an error with the sound. I'm using ubuntu mate 21.04, with krenel as I wrote earlier.

lamyseba commented 2 years ago

Hi, any news on this bug? Is it solved in newer kernels ?

Pandemonious commented 2 years ago

Hi, any news on this bug? Is it solved in newer kernels ?

for me it stopped happening with a kernel update, so yeah ... Wish I knew more.

jwrdegoede commented 2 years ago

for me it stopped happening with a kernel update, so yeah ... Wish I knew more.

Maybe the new kernel is no longer using the SOF driver? What is the output of "dmesg | grep sof" ?

nrndda commented 2 years ago

The bug still persist. Kernel: 5.16.0-rc8-gb2b436ec0205-dirty sof-audio-acpi-intel-byt 808622A8:00: Firmware info: version 2:0:0-b678a sof-audio-acpi-intel-byt 808622A8:00: Firmware: ABI 3:20:0 Kernel ABI 3:18:0

lamyseba commented 2 years ago

Hi, just asking about setting options. In his message, @plbossart suggested to set option snd_sof sof_debug=1. Reading this and not knowing about what "options" is, I supposed it was kernel options, and that is what my further message was showing. But reading further message from @plbossart , I think I was wrong : @plbossart was talking about modprobe/alsa options that should be set in the /etc/modprobe.d/alsa-base.conf file. This file does not exist in my filesystem (manjaro install), should I create this file, or does this mean this options should be set somewhere else? The message from @Shen-long87 show that he has this file and that it is full of config text.

plbossart commented 2 years ago

@lamyseba the name of the file does not matter, you can use /etc/modprobe.d/mysoftest.conf if you prefer. The .conf extension is however required.

rdft4e commented 2 years ago

Hi, it was difficult to find you. The term "screech" made me the clue. I confirm it in Ubuntu 20.04.3 LTS, Kernel release: 5.13.0-28-generic,x86_64 snd_soc_sst_bytcr_rt5651 (card 1) SOF - sof-bytcht rt5651, BESSTARTech-Z83_F-Defaultstring-T3MRD The persistent beep appeared after a large updating/upgrading of this LTS.

The modification of alsa-base.conf with options snd_sof sof_debug=1 solved the problem. More than one day without beeping.

A question: options snd_sof sof_debug=1 is equivalent to snd-intel-dspcfg dsp_driver=2 ?

Another question: What I can expect after a kernel or related update to alsa ?

Thank you very much for solving this annoying thing.

plbossart commented 2 years ago

A question: options snd_sof sof_debug=1 is equivalent to snd-intel-dspcfg dsp_driver=2 ?

no. the former is related to a debug option of the SOF driver, while the latter will select the SST driver - i.e. NOT SOF. if you want options snd_sof sof_debug=1 to have any effect you need options snd-intel-dspcfg dsp_driver=3

Another question: What I can expect after a kernel or related update to alsa ?

It's been more than a year and we haven't root caused the issue. Don't hold your breath.

rdft4e commented 2 years ago

A question: options snd_sof sof_debug=1 is equivalent to snd-intel-dspcfg dsp_driver=2 ?

no. the former is related to a debug option of the SOF driver, while the latter will select the SST driver - i.e. NOT SOF. if you want options snd_sof sof_debug=1 to have any effect you need options snd-intel-dspcfg dsp_driver=3

I do not have snd-intel-dspcfg dsp_driver=3 explicit in the alsa-base.conf but I am "pretty sure" that options snd_sof sof_debug=1 solve the arbitrary continous beeping. If it come back I will try the options snd-intel-dspcfg dsp_driver=3 but now I am not touching nothing for the momento and crossfingering.

Another question: What I can expect after a kernel or related update to alsa ?

It's been more than a year and we haven't root caused the issue. Don't hold your breath.

plbossart commented 2 years ago

check your kernel config @rdft4e, you probably have the CONFIG_SND_INTEL_BYT_PREFER_SOF option set. that would override the default which is to use snd-intel-dspcfg dsp_driver=2 on those devices.

d03j commented 2 years ago

Hi,

I'm running Ubuntu MATE 21.10 x86_64 kernel 5.13.0-28-generic Intel Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Series Imaging Unit alsa k5.13.0-28-generic PipeWire 0.3.32 sof-bytcht nau8824

And am also getting a loud beep/screech on the speakers and headphones after a minute or two playing audio.

d03j commented 2 years ago

FWIW, installing v1.9 from https://github.com/thesofproject/sof-bin/archive/refs/heads/main.zip solved my issue. v2.0 doesn't seem to fix it.

sturn42 commented 2 years ago

Is it possible that the SOF firmware playing audio in stereo could be part of the problem? I am curious as I noticed the SST firmware plays back in mono when I changed to it.

plbossart commented 2 years ago

@sturn42 That would be a bug of the SST firmware/driver or a configuration issue. On some platforms there's also a mono speaker only, so there is a downmix in the codec IIRC. The I2S link is always configured as stereo btw. The word clock aka frame sync toggles and defines the position of left and right channels.

wwangsa commented 2 years ago

Hi, Hopefully my experiments helps at the higher level. I experienced the same issue with fxthomas and Pandemonious. I encountered continuous beep after several minutes of any video/music playback and the sound stops after the playback is paused. Finally, the problem went away with Manjaro Qonos 21.2.6 with Kernel 5.15.32-1 XFCE on Acer Chromebook (Edgar CB3-341 -- Intel Braswell). Before I landed with Manjaro XFCE, I installed several Ubuntu distributions and none of them working except the Gallium OS and somehow I believe this is related to the Intel Braswell (hardware accelaration) driver rather than kernel issue.

Here were my attempts:

Ubuntu Budgie 21.10 and tried with different kernel 5.15, 5.16 and 5.17. Installed ubuntu-restricted-extras. YouTube video froze after several minutes followed by continous beep. MOV file is lagging. Has to use MPV player to play it smoothly. I installed the SOF driver but it only lower down the beep volume.
Ubuntu Mate 21.10 and tried with different kernel 5.15, 5.16 and 5.17. Installed ubuntu-restricted-extras. YouTube video froze after several minutes followed by continous beep. MOV file is lagging.
Gallium OS Xubuntu 18.04. Playback were smooth and similar experience like using chromebook. There is no issue with the sound and playback but the OS is not maintained. Lots of broken packages and I coudn't upgrade to the latest ubuntu.
Manjaro Qonos Budgie live stick would not boot. It only shows black screen.
Lubuntu 22.04 with kernel 5.15. MOV file plays fine but YouTube video froze after several minutes followed by continous beep.

plbossart commented 2 years ago

@wwangsa I would bet the differences in results is largely due to the kernel configurations, some kernels enable the DSP firmware trace and some don't, but as can be read above the only current/known work-around is to enable the trace...

Danct12 commented 2 years ago

I haven't seen the issue coming back lately on Arch Linux, kernel 5.18.2 and SOF 2.1.1-1. I was able to play games and watch YouTube videos without running into this issue.

nrndda commented 2 years ago

Still present without snd_sof.sof_debug=1 Kernel 5.18.3 sof-audio-acpi-intel-byt 808622A8:00: Firmware info: version 2:1:1-3964a sof-audio-acpi-intel-byt 808622A8:00: Firmware: ABI 3:21:0 Kernel ABI 3:19:1

mengdonglin commented 2 years ago

Move BYT and CHT platform to maintenance mode and SOF v2.2 will be the last upstream release with support for this platform. There will be no more FW bug fixing effort on this platform.

thesofproject / sof

[BUG][BYT][CHT] Linux 5.11 Cherrytrail/rt5640 emits persistent beeping sound after a short while #3868

Config solving the bug with LTS kernel

With legacy drivers and kernel 5.13.10-arch1-1-surface

With SOF trace enabled and kernel 5.13.10-arch1-1-surface

Ubuntu #62691, enable MPU for snd-cmipci

Keep snd-pcsp from being loaded as first soundcard

Keep snd-usb-audio from beeing loaded as first soundcard