TGL: Chrome-Ref: SDW: DMIC & HDMI Audio Fails with Generic Machine Driver

[naveen-manohar]

Issue Description:
Post switching to Generic Machine driver, Observe issue with DMIC Capture & HDMI Audio Playback

Details:
Kernel: 9b5b6c2 (HEAD -> ww12.5, origin/topic/sof-dev, origin/HEAD)
Topology: thesofproject/sof@master...naveen-manohar:tgl_rt5682-sdw
DSP-FW#1: TGL_SoF_V0.3 Release

Error Log:
dmesg_dmic_hdmi_panic.log

DMIC Capture

[   97.416021]  DMIC: ASoC: open FE DMIC
[   97.416027] sof-audio-pci 0000:00:1f.3: pcm: open stream 2 dir 1
[   97.416030] sof-audio-pci 0000:00:1f.3: period min 192 max 16384 bytes
[   97.416031] sof-audio-pci 0000:00:1f.3: period count 2 max 16
[   97.416033] sof-audio-pci 0000:00:1f.3: buffer max 65536 bytes
[   97.416261]  dmic01: ASoC: hw_params BE dmic01
[   97.416267] dmic-codec dmic-codec: Update DAI routes for dmic-hifi capture
[   97.416271] dmic-codec dmic-codec: Connecting DAI route DMIC AIF -> Capture
[   97.416274] sof-audio-pci 0000:00:1f.3: Update DAI routes for DMIC01 Pin capture
[   97.416277]  DMIC: ASoC: hw_params FE DMIC rate 48000 chan 4 fmt 10
[   97.416279] sof-audio-pci 0000:00:1f.3: Update DAI routes for DMIC 2 capture
[   97.416282] sof-audio-pci 0000:00:1f.3: Connecting DAI route PCM3C -> Passthrough Capture 3
[   97.416284] sof-audio-pci 0000:00:1f.3: pcm: hw params stream 2 dir 1
[   97.416288] sof-audio-pci 0000:00:1f.3: generating page table for 0000000090acddd6 size 0xffc0 pages 16
[   97.416330] sof-audio-pci 0000:00:1f.3: stream_tag 1
[   97.416339] sof-audio-pci 0000:00:1f.3: ipc tx: 0x60010000
[   97.416408] sof-audio-pci 0000:00:1f.3: error: ipc error for 0x60010000 size 20
[   97.416435] sof-audio-pci 0000:00:1f.3: error: hw params ipc failed for stream 1
[   97.416440] sof-audio-pci 0000:00:1f.3: ASoC: 0000:00:1f.3 hw params failed: -22
[   97.416443]  DMIC: ASoC: hw_params FE failed -22

HDMI Playback

[  215.649570]  HDMI2: ASoC: open FE HDMI2
[  215.649575] sof-audio-pci 0000:00:1f.3: pcm: open stream 5 dir 0
[  215.649576] sof-audio-pci 0000:00:1f.3: period min 192 max 16384 bytes
[  215.649577] sof-audio-pci 0000:00:1f.3: period count 2 max 16
[  215.649578] sof-audio-pci 0000:00:1f.3: buffer max 65536 bytes
[  215.651333]  iDisp2: ASoC: hw_params BE iDisp2
[  215.651336] snd_hda_codec_hdmi ehdaudio0D2: Update DAI routes for intel-hdmi-hifi2 playback
[  215.651348] sof-audio-pci 0000:00:1f.3: ipc tx: 0x80010000
[  215.651469] sof-audio-pci 0000:00:1f.3: ipc tx succeeded: 0x80010000
[  215.651495] sof-audio-pci 0000:00:1f.3: Update DAI routes for iDisp2 Pin playback
[  215.651496]  HDMI2: ASoC: hw_params FE HDMI2 rate 48000 chan 2 fmt 2
[  215.651497] sof-audio-pci 0000:00:1f.3: Update DAI routes for HDMI2 5 playback
[  215.651501] sof-audio-pci 0000:00:1f.3: Connecting DAI route Passthrough Playback 6 -> PCM6P
[  215.651502] sof-audio-pci 0000:00:1f.3: pcm: hw params stream 5 dir 0
[  215.651504] sof-audio-pci 0000:00:1f.3: generating page table for 0000000071e9840a size 0xffc0 pages 16
[  215.651566] sof-audio-pci 0000:00:1f.3: stream_tag 1
[  215.651570] sof-audio-pci 0000:00:1f.3: ipc tx: 0x60010000
[  215.651672] sof-audio-pci 0000:00:1f.3: error : DSP panic!
[  215.651675] sof-audio-pci 0000:00:1f.3: panic: dsp_oops_offset 788480 offset 788480
[  215.651679] sof-audio-pci 0000:00:1f.3: status: fw entered - code 00000005
[  215.651889] sof-audio-pci 0000:00:1f.3: error: runtime exception
[  215.651890] sof-audio-pci 0000:00:1f.3: error: trace point 00004000
[  215.651892] sof-audio-pci 0000:00:1f.3: error: panic at :0
[  215.651893] sof-audio-pci 0000:00:1f.3: error: DSP Firmware Oops
[  215.651895] sof-audio-pci 0000:00:1f.3: error: Exception Cause: LoadStorePIFDataErrorCause, Synchronous PIF data error during LoadStore access

@keqiaozhang : Please share your observation.

NOTE:
#1. With the mentioned recipe, RT5682 Headset Playback & Capture works fine
#2. With the same Topology and DSP FW on Old machine driver(PR:1885), DMIC Capture and HDMI Audio worked fine.

[plbossart]

@naveen-manohar I would have expected you to test prior to merging?

At any rate, did you make sure the topology is now aligned with the machine driver? the dai link IDs will be different so there's a risk of mismatch here.

Regardless we should trap this so you should also provide the sof-logger output to harden kernel+firmware.

[naveen-manohar]

@plbossart As I was posting rt5682 helper file(remotely), was concentrating headset functionality and this works perfect(verified through loop-back)
While submitting the corresponding topology file, noticed these issues and raised a GH right-away.
Apologies for the miss..

I ensured the Topology matches the new mach driver wrt link_id, Also any wrong ID-configuration will hit null-ptr deref issue.
console-ramoops-0.txt

@RanderWang was using the DUT, shall align and get the sof-logger tmrw.

[RanderWang]

sdw5682.diff.txt
pcm id should match with pipeline id.

[naveen-manohar]

Many Thanks @RanderWang
Verified all the end-points(remotely through loop-back) and seems all works fine.

@keqiaozhang : Have updated PR:2527

[keqiaozhang]

Both DMIC and HDMI can work with new TPLG
thesofproject/sof#2527

[naveen-manohar]

Thanks All