Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions AGENTS.md
Original file line number Diff line number Diff line change
Expand Up @@ -115,6 +115,7 @@ Each entry gives the gotcha, then the reason or the fix. Details are in the link
- **ABL ignores the `vendor_boot` cmdline:** put kernel arguments in `/chosen/bootargs`. ABL appends `console=null`, so a panic is invisible: use `panic=10`. [BRINGUP-M1](device/cheetah/BRINGUP-M1.md)
- **ABL leaves both cluster watchdogs running:** the phone resets at about 60 s of uptime unless `s3c2410_wdt` takes them over at probe.
- **ABL picks the base DTB by `soc_id`:** `bin/omamobile images` ships both silicon revisions. It also needs an empty `/ect` node and `/chosen/bootargs`, and applies a DPM overlay that needs label stubs (build with `-@`).
- **Pixel 7 Pro: once slot B has booted with AVB footers, it needs them.** A stock install's `vbmeta_b` chains `boot` and `init_boot`. After one boot of #104's footer-carrying images, ABL refused footer-less ones (images built by any tool from before #104, or a saved older image set): it stopped in fastboot or used up B's tries and fell back to stock, whose recovery offered "Try again" or "Factory data reset". Build every image with the current `bin/omamobile`, keep rollback sets built the same way, and on that recovery screen choose neither: hold Volume Down + Power about 20 s for fastboot. A stock boot rewrites `efs`, `persist` and `modem_userdata`, so take a fresh read-only backup after one. [NOTES](device/cheetah/NOTES.md#2026-10-30-the-charger-reads-offline-at-full-and-slot-b-now-needs-its-avb-footers)
- **Pixel 9a: a fallback to slot A costs the Linux root.** `fastboot --set-active=b` (which `bin/omamobile flash` runs) clears B's successful flag and gives B three ABL tries; every boot uses one, and the initramfs USB safety net (no USB host in about 65 s: panic) uses them fast. At zero ABL boots stock on slot A, and stock overwrote `userdata` on 10-28, and most likely on 10-26, before anyone could choose anything in recovery. So: `flash` only a known-good image set, and on the 9a ask the maintainer before any `flash`, because it unmarks B. Marking B successful is a `devinfo` write, and on the 9a it is on hold until the maintainer decides how it's done: don't mark B, and don't run `omamobile-bootctl` or anything that makes `devinfo` writable. With B marked, try new images with `push`, which doesn't call `--set-active` and leaves the mark alone; ABL then no longer falls back, so a broken pushed image means fastboot by the buttons. With B unmarked, every boot spends one of the three tries: keep reboots few and don't test new images. [tegu NOTES](device/tegu/NOTES.md#2026-10-26-and-2026-10-28-slot-a-fallbacks-that-cost-userdata); the operation approved on 2026-10-22, now on hold: [NOTES](device/tegu/NOTES.md#approved-b-success-operation-after-layout-review-2026-10-22).
- **`reboot -f` and busybox `reboot` hang:** use `reboot-to <reason>` (`LINUX_REBOOT_CMD_RESTART2`). ABL often ignores the `bootloader` reason anyway.
- **Reserved memory on gs201 is `#address-cells = <2>`, `#size-cells = <1>`,** so each `reg`/`alloc-ranges` entry is 12 bytes. A 16-byte value is silently ignored, and the pool lands anywhere. The display pool's `alignment` must be two cells too, or `of_reserved_mem` rejects it. `/proc/device-tree` shows the raw property, so it *looks* right. [WIFI-BATTERY-PLAN](device/cheetah/WIFI-BATTERY-PLAN.md) 9.x, [NOTES](device/cheetah/NOTES.md)
Expand Down
63 changes: 63 additions & 0 deletions device/cheetah/NOTES.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,69 @@

Findings from real hardware, newest first. `BRINGUP-M1.md` is the research map written before first boot.

## 2026-10-30: the charger reads offline at Full, and slot B now needs its AVB footers

Kernels `7.3.0-rc6-omamobile-push1-00566-g66497d0197dc` (10-29),
`…-00568-gcfd00cdd22e6` (the gauge fix plus the fast-charge timer off),
`…-00571-gd40c36abfc0d` (`omamobile-int` with all three charging branches)
and the curated `…-gaae3b058ef98`; device-cheetah 0.1.34; the laptop's USB
port.

**Charging** [V]:

- **What looked broken on 10-29:** a phone at 0 % didn't gain a percent on
the laptop and then a wall charger, and USB dropped and came back. It was
neither a boot loop nor USB: the drops were upower's 0 % suspends, and SSH
failed because the laptop's Tailscale exit node routed 172.16.42.1 away
from the cable (`ip route get` showed `dev tailscale0 table 52`; fixed with
`tailscale set --exit-node-allow-lan-access=true`). A reboot by hand
charged at once. Why that boot's charger stayed offline after a replug in
the evening is still open: the journal has no entries from 10:24 to 22:00
(the boot clock read 1999 until the time was set), and the gauge model had
failed to load at that boot (-110). The flat-battery case is issue #169.
- **Charging works asleep:** in a fresh boot the battery took 226 mAh in 9.5
min of `rtcwake -m freeze` and 260 mAh in 10 min of `systemctl suspend`
(about 1.4-1.56 A), the same as awake.
- **Input current:** CNFG_09 reads 0x13 (500 mA, as `default_ilim_ua` sets
it), but NO_AUTOIBUS is clear, so the charger picks its input current
itself; Google's driver sets it. The gauge's 1.42 A is real.
- **At Full** (03:25 on 00568): CHG_DETAILS_01 went CC 0xb1, CV 0xb2, top-off
0xb3, DONE 0xb4, and CHG_INT_OK went 0xdd to 0xcd (CHG_OK clears) with
CNFG_00 still 0x05. The driver's `online` required CHG_OK as well as
CHGIN_OK, so a full phone on the cable read offline and upower counted it
as on battery. "power: supply: max77759: stay online while a full battery
is on the charger" reports the input from CHGIN_OK alone; on 00571 the
charger reads online at Full.
- **Fast-charge timer:** CNFG_01 0x99 to 0x98 (FCHGTIME off), as Google's
driver sets it; read back on 00568.
- **Replug after Full** (08:19 on 00568): after five hours at DONE, out for
a minute and back in; `omamobile-usb-rescue` closed the data switch and the
charger charged again (+373 mA at 98 %). A long unplug with the battery well
down, and a phone run flat, are still to be repeated.

**AVB footers, the hard way** [V]:

- `vbmeta_b` on this phone is byte-identical to `vbmeta_a` (stock CP2A's,
chaining `boot` and `init_boot` at rollback index locations 2 and 4), not
the one in the 10-11 backup: slot B got CP2A's factory bootloader and
`vbmeta` during the M1 bring-up (BRINGUP-M1 step 10). bl1, bl2, bl31, gsa,
ldfw, pbl, tzsw and modem match slot A's.
- #104's images (footers with CP2A's index) booted on 00571 at 09:15 with
every smoke check ok. After that boot, footer-less images no longer
booted: curated images built by a checkout from before #104 made ABL stop
in fastboot at once, and a saved 00571 set from before #104 used up B's
tries (no USB gadget, the initramfs safety net) and fell back to stock,
whose recovery offered "Try again" and "Factory data reset". Neither was
chosen; Volume Down + Power gave fastboot, and the 00571 set with footers
booted, then the curated kernel rebuilt with current tools.
- Stock's recovery ran twice and rewrote `efs`, `efs_backup`,
`modem_userdata` and `persist` (they matched the 10-12 backup that
morning); `modem_a/b` are unchanged. A fresh read-only backup was taken
and verified the same hour.
- So: build every image with the current `bin/omamobile`, keep rollback sets
built the same way, and never choose either option on stock's recovery
screen.

## 2026-10-29: footers with CP2A's rollback index boot on a phone with CP3A on slot A

- **A footer index below slot A's boots** [V]: the descriptor now carries
Expand Down
7 changes: 3 additions & 4 deletions kernel/configs/omamobile.config
Original file line number Diff line number Diff line change
Expand Up @@ -275,11 +275,10 @@ CONFIG_MEDIA_CONTROLLER=y
CONFIG_VIDEO_V4L2_SUBDEV_API=y
CONFIG_VIDEO_CAMERA_SENSOR=y
CONFIG_VIDEO_CCS=m
# Camera power bring-up (CAMERA-PLAN.md CAM1-CAM3): the read-only CSIS probe,
# the S2MPG13 GPIOs, the SLG51002 camera PMIC and the module power test
# driver. Blacklisted from autoload (modprobe.d/omamobile-camera.conf); their
# Camera power bring-up (CAMERA-PLAN.md CAM1-CAM3): the S2MPG13 GPIOs, the
# SLG51002 camera PMIC and the module power test driver (the read-only CSIS
# probe was dropped when the kernel was curated for publishing, 2026-10-30). Blacklisted from autoload (modprobe.d/omamobile-camera.conf); their
# DT nodes stay disabled until the camera-pwr-test overlay.
CONFIG_GS201_CAM_PROBE=m
CONFIG_GS201_CAM_POWER=m
CONFIG_GPIO_S2MPG13=m
CONFIG_REGULATOR_SLG51002=m
Expand Down