From dc5ca41b7ee9a4fbf89c5c2834851b9448fd2897 Mon Sep 17 00:00:00 2001 From: Colin Lundholm <265166644+gitColinX@users.noreply.github.com> Date: Fri, 13 Nov 2026 13:31:23 -0700 Subject: [PATCH 1/2] extract-vendor-firmware: husky follow-ups from #722 Derek's review of #722 asked for three small things: - The refusal named the pattern ("is not a G5SN_V[0-9]*_HK3 Bluetooth patch"). BT_NAME gives it a readable name where BT_CHIP is a pattern: "BCM.hcd is not a Pixel 8 Pro (G5SN_V_HK3) Bluetooth patch". Other devices keep their chip name. - No test fed husky a wrong cs40l26.wmfw. A Halo WMFW that isn't the CS40L26's is now refused for husky; with husky taken out of the haptics check, that test fails. - The device README said "also" before the Wi-Fi paragraph it referred to. The sentence follows it now, with a line that pairings made on the placeholder Bluetooth address won't survive the address step. Co-Authored-By: Claude Opus 5.5 --- packages/omamobile-device-husky/README.md | 6 +++++- tests/extract-vendor-firmware | 5 ++++- tools/extract-vendor-firmware | 5 ++++- 3 files changed, 13 insertions(+), 3 deletions(-) diff --git a/packages/omamobile-device-husky/README.md b/packages/omamobile-device-husky/README.md index 0f8c81840..bcdb74d2b 100644 --- a/packages/omamobile-device-husky/README.md +++ b/packages/omamobile-device-husky/README.md @@ -28,7 +28,7 @@ Not in this package: the boot-success marker, the charger, cameras, and the slot ## Firmware the phone needs -Nothing here is in the package. `tools/extract-vendor-firmware -d husky -s SKU` also stages the Bluetooth patch (`brcm/BCM.google,zuma-husky.hcd`, stock's `BCM.hcd`) and the five CS40L26 haptics files (`cs40l26.wmfw`, `cs40l26.bin` and the three tuning files) from the same image. Wi-Fi needs four files from Google's firmware for the BCM4398 D0, which `tools/extract-vendor-firmware -d husky -s SKU` takes from the owner's own factory image or OTA `vendor.img`. They're installed under `/usr/lib/firmware/updates/brcm`, with the board-suffixed names brcmfmac loads first: +Nothing here is in the package. Wi-Fi needs four files from Google's firmware for the BCM4398 D0, which `tools/extract-vendor-firmware -d husky -s SKU` takes from the owner's own factory image or OTA `vendor.img`. They're installed under `/usr/lib/firmware/updates/brcm`, with the board-suffixed names brcmfmac loads first: - `brcmfmac4398d0-pcie.google,zuma-husky.bin`: the firmware; - `.txt`: the NVRAM for the phone's region, which `-s` picks from the hardware SKU (`grep hardware.sku /proc/bootconfig`); @@ -36,3 +36,7 @@ Nothing here is in the package. `tools/extract-vendor-firmware -d husky -s SKU` - `.txcap_blob`: the TX capabilities. The kernel side is [kernel #156](https://github.com/derekross/linux-omarchy-mobile/pull/156) (a draft). Nothing in this package starts or configures Wi-Fi: NetworkManager finds `wlp1s0` on its own. + +The same run also stages the Bluetooth patch (`brcm/BCM.google,zuma-husky.hcd`, stock's `BCM.hcd`) and the five CS40L26 haptics files (`cs40l26.wmfw`, `cs40l26.bin` and the three tuning files) from the same image. + +Until this package has its own Bluetooth address step, the controller comes up on its placeholder address, the same on every 8 Pro. A pairing made now won't survive the change to the phone's own address: pair the device again after it. diff --git a/tests/extract-vendor-firmware b/tests/extract-vendor-firmware index 7532ce5da..4703432ef 100755 --- a/tests/extract-vendor-firmware +++ b/tests/extract-vendor-firmware @@ -251,7 +251,10 @@ refuses "panther: a missing MFC firmware" "/vendor/firmware/mfc_fw.bin is missin # husky's BCM.hcd names the module and board, not the chip: another phone's # patch (here the 7 Pro's) is refused rm -rf "$t/husky-otherbt"; cp -r "$t/husky" "$t/husky-otherbt"; echo "BCM4389C1 fictional" > "$t/husky-otherbt/firmware/brcm/BCM.hcd" -refuses "husky: another chip's Bluetooth patch" "BCM.hcd is not a" "$t/husky-otherbt" "$t/ext4.img" -d husky +refuses "husky: another chip's Bluetooth patch" "BCM.hcd is not a Pixel 8 Pro (G5SN_V_HK3) Bluetooth patch" "$t/husky-otherbt" "$t/ext4.img" -d husky +# husky runs the haptics checks too: a Halo WMFW that isn't the CS40L26's is refused +rm -rf "$t/husky-otherhap"; cp -r "$t/husky" "$t/husky-otherhap"; halo "calib fictional" > "$t/husky-otherhap/firmware/cs40l26.wmfw" +refuses "husky: another part's haptics firmware" "cs40l26.wmfw is not a CS40L26 firmware" "$t/husky-otherhap" "$t/ext4.img" -d husky rm -rf "$t/v2-missing"; cp -r "$t/v2" "$t/v2-missing"; rm "$t/v2-missing/firmware/aoc.bin" refuses "a missing ext4 file with an older copy in OUT" "/vendor/firmware/aoc.bin is missing" "$t/v2-missing" "$t/ext4.img" refuses "a missing EROFS file with an older copy in OUT" "/vendor/firmware/aoc.bin is missing" "$t/v2-missing" "$t/erofs.img" diff --git a/tools/extract-vendor-firmware b/tools/extract-vendor-firmware index 2efcaa3a8..d256625a5 100755 --- a/tools/extract-vendor-firmware +++ b/tools/extract-vendor-firmware @@ -130,6 +130,8 @@ FW_BASE="brcmfmac4389c1-pcie" # and in BCM.hcd ("BCM ") WIFI_CHIP="4389c1" BT_CHIP="BCM4389C1" +# What the Bluetooth refusal calls the patch, where BT_CHIP is a pattern +BT_NAME="" DEST_REL="usr/lib/firmware/updates" # source path in the vendor partition -> path under $DEST_REL @@ -258,6 +260,7 @@ case $DEVICE_OPT in FW_BASE="brcmfmac4398d0-pcie" WIFI_CHIP="4398d0" BT_CHIP="G5SN_V[0-9]*_HK3" + BT_NAME="Pixel 8 Pro (G5SN_V_HK3)" # Google's SKU table (dhd_custom_google.c); husky's vendor has NVRAMs for # MMW, ROW and NA only, so the others take the module's default, as on stock case $SKU in @@ -466,7 +469,7 @@ brcm="$dest/brcm" grep -aq "$WIFI_CHIP-roml" "$brcm/$FW_BASE.$BOARD.bin" || die "fw_bcmdhd.bin is not a $WIFI_CHIP PCIe firmware" [[ -z $BT_CHIP ]] || grep -aq "$BT_CHIP" "$brcm/BCM.$BOARD.hcd" || - die "BCM.hcd is not a $BT_CHIP Bluetooth patch" + die "BCM.hcd is not a ${BT_NAME:-$BT_CHIP} Bluetooth patch" grep -aq '^NVRAMRev=' "$brcm/$FW_BASE.$BOARD.txt" || die "bcmdhd.cal does not look like a Broadcom text NVRAM" wmfw_halo() { # WMFW, Halo core (4): what cs_dsp accepts for this part From e0afef7b8995cc9062c271caab40c78919af1dce Mon Sep 17 00:00:00 2001 From: Colin Lundholm <265166644+gitColinX@users.noreply.github.com> Date: Sat, 14 Nov 2026 22:01:11 -0700 Subject: [PATCH 2/2] omamobile-device-husky README: the address step is in, Wi-Fi merged #752 gave the package its Bluetooth address step, so the paragraph that said a pairing wouldn't survive "until this package has its own address step" now says that pairings made before 0.1.13 have to be made again once. The Wi-Fi line named kernel #156 as a draft; it merged through kernel #187 on 2026-11-09. Co-Authored-By: Claude Opus 5.5 --- packages/omamobile-device-husky/README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/packages/omamobile-device-husky/README.md b/packages/omamobile-device-husky/README.md index bcdb74d2b..bfbf7fc14 100644 --- a/packages/omamobile-device-husky/README.md +++ b/packages/omamobile-device-husky/README.md @@ -35,8 +35,8 @@ Nothing here is in the package. Wi-Fi needs four files from Google's firmware fo - `.clm_blob`: the regulatory data; - `.txcap_blob`: the TX capabilities. -The kernel side is [kernel #156](https://github.com/derekross/linux-omarchy-mobile/pull/156) (a draft). Nothing in this package starts or configures Wi-Fi: NetworkManager finds `wlp1s0` on its own. +The kernel side merged on 2026-11-09 through [kernel #187](https://github.com/derekross/linux-omarchy-mobile/pull/187), which carried [#156](https://github.com/derekross/linux-omarchy-mobile/pull/156)'s commits. Nothing in this package starts or configures Wi-Fi: NetworkManager finds `wlp1s0` on its own. The same run also stages the Bluetooth patch (`brcm/BCM.google,zuma-husky.hcd`, stock's `BCM.hcd`) and the five CS40L26 haptics files (`cs40l26.wmfw`, `cs40l26.bin` and the three tuning files) from the same image. -Until this package has its own Bluetooth address step, the controller comes up on its placeholder address, the same on every 8 Pro. A pairing made now won't survive the change to the phone's own address: pair the device again after it. +Since 0.1.13 the controller takes the phone's own address at boot (`omamobile-bt-addr`, above). Pairings made before 0.1.13, on the placeholder address, have to be made again once.