-
Notifications
You must be signed in to change notification settings - Fork 131
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
LNL SDW "signal too weak" alsabat capture failure #5076
Comments
Still happening with B0 and April BIOS: |
In this run, alsabat playback failed first: https://sof-ci.01.org/softestpr/PR1204/build488/devicetest/index.html?model=LNLM_SDW_AIOC&testcase=check-alsabat-headset-playback-997
Then all capture failed too but not as "usual". The playback failure seems to have left a bad state behind. |
@singalsu @plbossart could this be DC ? |
I'd like to park this issue until we've solved the more important Controller Timeout and initialization failed. |
"Signal too weak" just spotted after the SDW clock fix was merged: https://sof-ci.01.org/softestpr/PR1205/build514/devicetest/index.html |
If you look at instances seen today in PR testing like this: There's a seemingly a fairly clean reference sine tone recorded, it's just very weak signal just barely above noise floor (so the test fails as it should). Another alsabat did pass just on the same machine and all alsamixer settings were the same. Cannot immediately explain how this could happen. If we compare to other open alsabat cases like thesofproject/sof#9164 (on HDA), this looks distinctly different. The occurence rate seems higher, and the failure looks (and sounds) very different. |
This starts to look like a codec/setup issue with rt711. I logged on to one CI machine that triggers the issue and run the tests manually. I can obverse:
|
Seems there is connection to headset detection. In failing cases, the rt711_sdca_headset_detect() function is called twice. The results are the same, but in failing case this is just done twice upon codec runtime resume.
|
The kernel patch seems to help:
This confirms the hypothesis, so moving the bug to kernel. |
I noticed that Latest daily run today: July 12th reproduction: https://sof-ci.01.org/sofpr/PR9298/build6450/devicetest/index.html?model=LNLM_SDW_AIOC&testcase=check-alsabat-headset-playback-599 |
TWL is 2 generations older than LNL, this should be a different issue. |
Today's failure. Same or different bug?
Same in Aug 6th daily 44550?model=LNLM_SDW_AIOC&testcase=check-alsabat-headset-capture-997 EDIT: still failing in August 12th daily run 44671?model=LNLM_SDW_AIOC&testcase=check-alsabat-headset-capture-599 |
Tested alsabat on another LNL system with a headset-codec <-> dmic acoustic loop, and I could not reproduce these failures. I was using sof-dev as of today and SOF2.10 FW release. The DUT configuration is different, so this is not fully comparable to the original setup, but at least one positive test setup with USB-audio excluded from test setup. FYI @ssavati @jsarha |
Originally posted by @marc-hb in #4681 (comment)
"signal too weak" failure in daily test run 41404?model=LNLM_SDW_AIOC&testcase=check-alsabat-headset-capture-599, see logs below.
EDIT: also in https://sof-ci.01.org/sofpr/PR9159/build4831/devicetest/index.html?model=LNLM_SDW_AIOC&testcase=check-alsabat-headset-capture-821
cc:
The text was updated successfully, but these errors were encountered: