Thanks for report Dave.  Could you attach output of generate_logs and tarball of /usr/local/autotest/results dir as well for additional debug.

Looking at the automated test lab I'm not seeing any recent failures w/ that signature.

Was the failure repeatable?  How about after reboot?  Did audio work when tested manually?

Hi,

I've just tested against developer builds of kevin (dated 2018-01-17) and caroline (2018-01-24), using a CrOS checkout from 2018-01-17 - the failure only occurred on caroline. It happens every time on caroline.

Using the volume buttons on caroline to change volume does not work properly - they just set volume to mute (lower vol button) or to unmuted-but-lowest-setting (increase vol button). The slider appears on screen but doesn't move - they just toggle the strike-through on the speaker logo.

Volume buttons work fine on kevin.

I've attached a tarball of the results directory from caroline - although based on the issues with interactively setting volume, it looks like this is not a test issue.

Deleted: plt_results.tar.bz2 11.8 MB

plt_results.tar.bz2 11.8 MB Download

Agreed (non-test issue).

I see stderr has,

01/24 02:54:27.425 INFO |      audio_helper:0093| Setting capture gain to 10
01/24 02:54:27.487 ERROR|             utils:0282| [stderr] Invalid card number.

And in messages,

2018-01-23T23:51:06.337060-08:00 ERR kernel: [    8.350651] snd_soc_skl 0000:00:1f.3: ASoC: failed to probe component -2
2018-01-23T23:51:06.337061-08:00 ERR kernel: [    8.350697] skl_nau88l25_ssm4567_i2s skl_nau88l25_ssm4567_i2s: ASoC: failed to instantiate card -2
2018-01-23T23:51:06.337062-08:00 WARNING kernel: [    8.350736] skl_nau88l25_ssm4567_i2s: probe of skl_nau88l25_ssm4567_i2s failed with error -2

Your version,
CHROMEOS_RELEASE_DESCRIPTION=10336.0.2018_01_24_0738 (Test Build - pdswauto) developer-build caroline

dCHROMEOS_RELEASE_DESCRIPTION=10332.0.0 (Official Build) dev-channel caroline test

So failure may be specific to your unit.  Did power cycling help?

No, it's been power-cycled and reflashed with the latest dev build since the first time I observed this error, so it seems persistent across power-cycling. 

Unfortunately we only have one caroline here so can't look at behaviour on other devices (if you mean it may be specific to an individual device). Were you thinking that a hardware fault may have stopped sound from working?

certainly possible its a HW issue specfic to your device as I can't locate any reports of similar failure to-date. I'll look to scrape available logfiles for similar devices to see if the signature turns up there.

Have you tried recovering the device?

I've tried a recovery image (downloaded today). The volume controls worked fine and the device made sound as expected. I then re-flashed with a test image (dated 2018-01-17); volume controls did not work (as above) and the device did not make any sound.

So it looks like there is a sound issue for caroline on the dev channel?

I got a caroline & installed 10334.0.0 test image and audio worked fine for me.

Could you,

1. cd /var/log && rm -rf /var/log/* && reboot
2. boot, login as guest, test audio (youtube.com) 
3. from crosh, shell, sudo su
4. ./generate_logs & attach here.

Wondering if there's something firmware related thats caused some incompat on your device.

Please see attached. I played a youtube video, then used the on-screen interface to try and adjust volume, then captured the logs.

Deleted: debug-logs_20180129-013011.tgz 829 KB

debug-logs_20180129-013011.tgz 829 KB Download

These look worrisome as before in #c5

2018-01-29T01:26:57.411037-08:00 ERR kernel: [    5.425314] skl_nau88l25_ssm4567_i2s skl_nau88l25_ssm4567_i2s: ASoC: CODEC DAI dmic-hifi not registered
2018-01-29T01:26:57.411051-08:00 INFO kernel: [    5.425508] platform skl_nau88l25_ssm4567_i2s: Driver skl_nau88l25_ssm4567_i2s requests probe deferral
2018-01-29T01:26:57.412593-08:00 ERR kernel: [    5.427164] skl_nau88l25_ssm4567_i2s skl_nau88l25_ssm4567_i2s: ASoC: CODEC DAI intel-hdmi-hifi1 not registered
2018-01-29T01:26:57.412599-08:00 INFO kernel: [    5.427181] platform skl_nau88l25_ssm4567_i2s: Driver skl_nau88l25_ssm4567_i2s requests probe deferral
2018-01-29T01:26:57.412958-08:00 WARNING kernel: [    5.427426] HDMI HDA Codec ehdaudio0D2: Max dais supported: 3
2018-01-29T01:26:57.412969-08:00 WARNING kernel: [    5.427547] snd_soc_skl 0000:00:1f.3: Direct firmware load for 9d70-CORE-COREBOOT-0-tplg.bin failed with error -2
2018-01-29T01:26:57.412970-08:00 INFO kernel: [    5.427559] snd_soc_skl 0000:00:1f.3: tplg fw 9d70-CORE-COREBOOT-0-tplg.bin load failed with -2, falling back to dfw_sst.bin
2018-01-29T01:26:57.412971-08:00 WARNING kernel: [    5.427577] snd_soc_skl 0000:00:1f.3: Direct firmware load for dfw_sst.bin failed with error -2
2018-01-29T01:26:57.412973-08:00 ERR kernel: [    5.427587] snd_soc_skl 0000:00:1f.3: Fallback tplg fw dfw_sst.bin load failed with -2
2018-01-29T01:26:57.412974-08:00 ERR kernel: [    5.427595] snd_soc_skl 0000:00:1f.3: ASoC: failed to probe component -2
2018-01-29T01:26:57.412975-08:00 ERR kernel: [    5.427611] skl_nau88l25_ssm4567_i2s skl_nau88l25_ssm4567_i2s: ASoC: failed to instantiate card -2
2018-01-29T01:26:57.412976-08:00 WARNING kernel: [    5.427643] skl_nau88l25_ssm4567_i2s: probe of skl_nau88l25_ssm4567_i2s failed with error -2

Are you able to try the 10334.0.0 test image by downloading from CPFE?

+Dylan as well.

benzh, firmware issue on caroline?

We probably should get this resolved since it's affecting multiple devices

So far we've only been able to identify one failure on Dave's system.  Please attach logs for the other failures as well.

This issue has an owner, a component and a priority, but is still listed as untriaged or unconfirmed. By definition, this bug is triaged. Changing status to "assigned". Please reach out to me if you disagree with how I've done this.