You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[BUG][GLK] HDMI failed to get afg sub nodes, no soundcard #1001
Daily build of 0530 has no soundcard due to HDMI failures.
linux-80a2cb89
sof-92fc7067
I tried daily build of 0530 onto my GLK Ubuntu RT7219 platform but I see hdmi codec failure, no soundcard. (full log attached)
[ 5.018689] HDMI HDA Codec ehdaudio0D2: HDMI: failed to get afg sub nodes
[ 5.018693] HDMI HDA Codec ehdaudio0D2: Failed in parse and map nid with err: -22
[ 5.018710] HDMI HDA Codec: probe of ehdaudio0D2 failed with error -22
In terms of audio playback in GLK(sof-dev), 072cf06 Merge pull request #970 from plbossart/merge/broonie-next-20190521 <== bad d807d30 ASoC: SOF: Intel: ipc: don't check for HIPCCTL register value <== good
After git bisect, the result is, e57ccca is the first bad commit e57ccca (refs/bisect/bad) Merge tag 'sound-5.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
This error started from e57ccca
[ 5.018689] HDMI HDA Codec ehdaudio0D2: HDMI: failed to get afg sub nodes
[ 5.018693] HDMI HDA Codec ehdaudio0D2: Failed in parse and map nid with err: -22
[ 5.018710] HDMI HDA Codec: probe of ehdaudio0D2 failed with error -22
One more update, I tried today's daily build(20190601, China time), surprisingly it works. For more build info,
linux build : 0a747dc
sof build : 3e3f1d0
I am still seeing this errors in today's latest commits:
[ 5.227101] HDMI HDA Codec ehdaudio0D2: HDMI: failed to get afg sub nodes
[ 5.227105] HDMI HDA Codec ehdaudio0D2: Failed in parse and map nid with err: -22
[ 5.227121] HDMI HDA Codec: probe of ehdaudio0D2 failed with error -22
I've seen coupe of CML boot instances failing with same error signature.
HDMI HDA Codec ehdaudio0D2: HDMI: failed to get afg sub nodes
HDMI HDA Codec ehdaudio0D2: Failed in parse and map nid with err: -22
HDMI HDA Codec: probe of ehdaudio0D2 failed with error -22
Test tried : run reboot script and check card lists everytime after boot. Happens 1 in 10 machines and 1 in 1000 iterations..
Activity
fredoh9 commentedon May 31, 2019
@ranj063 reported she found same issue in CML chromebook. And Chrome team found suspicious commit.
I tried to revert the commit id, but looks same to me.
333ee22 ASoC: SOF: Intel: hda: fix the hda init chip
I can try git bisect and update the candidate commit here.
fredoh9 commentedon Jun 1, 2019
In terms of audio playback in GLK(sof-dev),
072cf06 Merge pull request #970 from plbossart/merge/broonie-next-20190521 <== bad
d807d30 ASoC: SOF: Intel: ipc: don't check for HIPCCTL register value <== good
After git bisect, the result is,
e57ccca is the first bad commit
e57ccca (refs/bisect/bad) Merge tag 'sound-5.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
This error started from e57ccca
[ 5.018689] HDMI HDA Codec ehdaudio0D2: HDMI: failed to get afg sub nodes
[ 5.018693] HDMI HDA Codec ehdaudio0D2: Failed in parse and map nid with err: -22
[ 5.018710] HDMI HDA Codec: probe of ehdaudio0D2 failed with error -22
One more update, I tried today's daily build(20190601, China time), surprisingly it works. For more build info,
linux build : 0a747dc
sof build : 3e3f1d0
keqiaozhang commentedon Jun 3, 2019
@fredoh9
The latest code base works on my GLK-Ubuntu.
Kernel: sof-dev/0a747dc
sof/tplg: master/8901e0
keqiaozhang commentedon Jun 4, 2019
This issue cannot be reproduced on GLK. let's close this bug.
Kernel: sof-dev/0a747dc
sof/tplg: master/14b724
keyonjie commentedon Jun 11, 2019
I am still seeing this errors in today's latest commits:
@fredoh9 @keqiaozhang how about you?
sathyap-chrome commentedon Sep 4, 2019
I've seen coupe of CML boot instances failing with same error signature.
HDMI HDA Codec ehdaudio0D2: HDMI: failed to get afg sub nodes
HDMI HDA Codec ehdaudio0D2: Failed in parse and map nid with err: -22
HDMI HDA Codec: probe of ehdaudio0D2 failed with error -22
Test tried : run reboot script and check card lists everytime after boot. Happens 1 in 10 machines and 1 in 1000 iterations..
cc @ranj063