Skip to content

barbet: audio and Bluetooth firmware - #8

Open
Gibbz wants to merge 7 commits into
denialwm:mainfrom
bit-shift-io:devices/barbet-audio-bt
Open

Gibbz wants to merge 7 commits into
denialwm:mainfrom
bit-shift-io:devices/barbet-audio-bt

Conversation

@Gibbz

@Gibbz Gibbz commented Nov 19, 2026 •

Copy link
Copy Markdown

The audio and Bluetooth part of the split, stacked on the firmware PR (#7),
which is stacked on the device slice (#4).

What is here

  • build-boot.sh adds the RT5514P DSP blobs (rt5514p_dsp_fw{1..4}.bin,
    requested by bare name) and the WCN3990 patch/NVM (qca/crbtfw32.tlv,
    qca/crnv32.bin, qca/crnv32u.bin) to the tail.
  • The device README records the verified speaker route, the ADSP's place in
    it, the microphone's remaining DC offset, and the Bluetooth probe's
    dependence on the WLAN powering the combo chip.

What is not here, by agreement

No barbet-audio-start or barbet-bt-start. The ADSP-offline/WLAN-firmware
order and the post-WLAN Bluetooth rebind are the start-order issue (#10); the first
belongs in the service that starts the modem, and the second in the Bluetooth
probe. The README keeps the findings until that home is agreed.

Stacked on #7; rebase onto main when the stack lands. Not built here: no
vault and no phone in this environment.

A phone with no init_boot partition carries the early userspace in its boot
image, after the kernel. --ramdisk writes just the legacy-lz4 ramdisk for
that image. The boot tail then starts after the ramdisk as well as the
kernel: the early userspace and phone-reboot add the ramdisk size from the
header to the offset. A boot image with no ramdisk comes out the same.

PHONEWAVE_KERNEL names the linux-perpetua source the rescue's font comes
from, in place of the workstation path; it defaults to that path.
Add the device facts and the boot layout: SM7250, the SOFEF01 panel, ST FTS
touch and DRV2624 haptics, the WCN3990 behind the modem, and the RT5514P and
TAS2562 audio parts. The phone boots from slot B with a version 2 image that
carries the kernel, the ramdisk, the DTB and the tail; slot A is stock
Android and is never written.
Take a release number like the other devices, build it unless it exists, and
pack the modules and boot/tail/ with image/boot-tail. The ramdisk is the
shared early userspace written by image/init-boot --ramdisk, since this phone
keeps it in boot, and the image is packed by linux-perpetua's android-v2
from the verified working image in the vault. The kernel carries the Barbet
tree it must boot, so --dtb replaces the template's Google tree, and --tail
puts the modules and firmware after the DTB.
A version 2 boot image lays out the kernel, the ramdisk and the DTB in that
order, and the tail follows. The early userspace and phone-reboot computed
the offset from the kernel and ramdisk sizes alone, so on a v2 image they
looked one DTB too early. Read the DTB size from the header as well; a
version 4 template has zero there and keeps the old offset.
Record what the hardware run showed, with the battery as a voltage-derived
estimate and the microphone as partial: the DSP loads and streams, but the
samples carry a DC offset.
The kernel reads the modem, ADSP and WCN3990 firmware from one directory, but
the phone keeps the modem files in the modem partition and the ADSP and WLAN
files in vendor_a inside super. stock-firmware.conf binds one partition folder
over the firmware directory, which would hide the other source, so the vault
holds the extracted set with its hashes and build-boot.sh packs it into the
tail. Nothing is installed on the phone.

This is the firmware half of the Wi-Fi work. The modem still needs to be
started in the agreed order; that is the start-order issue.
Add the RT5514P DSP blobs and the WCN3990 patch/NVM to the tail, and record
what the audio and Bluetooth paths need. The speaker route is the verified
one; the microphone stays partial because of the DC offset. Neither the ADSP
start nor the Bluetooth rebind ships as a device script: the order they need
is the start-order issue, and it belongs in the modem service or the kernel
probe.
@Gibbz
Gibbz force-pushed the devices/barbet-audio-bt branch from f7a56ac to 74557aa Compare November 19, 2026 05:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants