34 lines
3 KiB
Markdown
34 lines
3 KiB
Markdown
# B12A — audio audit (read-only, desk only), 2026-10-04
|
||
|
||
B11 LED topic closed by the operator: the 15-LED array under the display frame (BOT side, logs/b11/b11c/B11C2-annotated-bot.jpg) is the display backlight.
|
||
|
||
## Stock
|
||
- DT dtb_01: internal codec. Analog part = PM8916 "Tombak" `msm8x16_wcd_codec@f000/@f100` (qcom,wcd-spmi): speaker class-D IRQs (spk_cnp/clip/ocp),
|
||
headset detect/MBHC (ins_rem_det, but_press/rel_det, mbhc_int), earpiece/HPH OCP. Supplies (rails inferred from voltages): VDDA_CP/VDD_PX 1.8–2.2 V (≈S4), VDDA_H/VDD_PA 1.8 V (≈L5), micbias 3.075 V (≈L13).
|
||
Card `sound` qcom,msm8x16-audio-codec, model **msm8x16-snd-card-mtp** (generic MTP), ext-pa "primary", US/Euro switch GPIO113, cdc-pdm GPIO63–68.
|
||
- Android: the card probed (card0 + "Button Jack"/"Headset Jack" inputs), audiod/mediaserver ran. /system/etc mixer_paths.xml = stock Qualcomm 8916 MTP
|
||
(speaker = SPK DAC, handset = EAR PA, headphones = HPHL/R, handset-mic ADC1, headset-mic ADC2); mixer_paths_mtp.xml differs only in speaker-mic → ADC3.
|
||
audio_policy.conf: attached EARPIECE|SPEAKER, BUILTIN_MIC|BACK_MIC. **All of it is the unmodified reference config → no evidence the board has any transducer.**
|
||
- No Aurora-specific audio code or config found.
|
||
|
||
## Photos (logs/b11/b11c/photos, 3 shots)
|
||
- No speaker, earpiece, microphone, 3.5 mm jack, or spring contacts labelled SPK/MIC visible. The edge clips at the bottom of TOP = antenna feeds (PRX, next to the RF test connector).
|
||
Top-left empty pad field + "DIV" connector = unpopulated diversity RX chain, not audio. BOT test pads: DP, DM, VBUS, VBAT, … (USB/battery).
|
||
- Under the shield cans: not visible. The PM8916 codec pins (SPK_P/N, EAR, HPH, MIC_IN1–3) may be routed to nothing.
|
||
|
||
## Current Linux
|
||
- 7.2.7-aurora-b10b5: `CONFIG_SOUND` not set. QCOM_APR=y, RPMSG SMD=y.
|
||
- Upstream path without the DSP: msm8916.dtsi `&lpass` (apq8016-lpass-cpu) + `&lpass_codec` (msm8916-wcd-digital) + `&pm8916_codec` (analog) + `&sound` (qcom,apq8016-sbc-sndcard).
|
||
msm8916-pm8916.dtsi already sets codec supplies L5/L5/L13. Many upstream 8916 phones use exactly this.
|
||
|
||
## Risks for any later enablement (needs GO)
|
||
- micbias on L13 = the USB PHY 3.075 V rail (NCM). The codec must not change the L13 voltage; verify regulator constraints before any boot.
|
||
- GPIO63–68 (cdc-pdm) and 113 stay codec-only. The rest of the B11 DO-NOT-TOUCH list stays as it was.
|
||
- Without a transducer a test proves only that the driver probes, the DAPM paths and the PMIC registers. No sound can be heard.
|
||
|
||
## Verdict
|
||
Codec silicon = present (PM8916 + MSM8916 digital codec, always). Audio HW on the board = **not found** (no speaker, mic, or jack visible). Next step depends on the operator:
|
||
are there speaker/mic openings in the case, or SPK/MIC pads under the display or near the cans?
|
||
|
||
## Closure (2026-10-04)
|
||
Operator: the case has no speaker and no microphone. **B12 audio CLOSED — not applicable**: codec silicon present, no transducers. No enablement planned; CONFIG_SOUND stays off. If ever needed: upstream apq8016-sbc path, mind micbias ≈ L13 shared with USB PHY.
|