Skip to content

arm64: dts: exynos: google: gs201-panther: reboot=warm, so reboot-to reaches fastboot - #198

Merged
derekross merged 1 commit into
derekross:omamobilefrom
angelrenefm:pr-panther-reboot-warm
Nov 7, 2026
Merged

derekross merged 1 commit into
derekross:omamobilefrom
angelrenefm:pr-panther-reboot-warm

Conversation

@angelrenefm

Copy link
Copy Markdown
Contributor

reboot-to bootloader on the Pixel 7 boots slot B again instead of stopping in fastboot: a cold PSCI reset loses the reboot mode the kernel writes for the bootloader. husky already carries reboot=warm in its bootargs for the same reason; this adds it to panther's.

Measured on a second Pixel 7 (stock CP3A.261005.002.A1 bootloader), kernel 7.3.0-rc6-omamobile-push1-g44048a9aa795 (a local branch: omamobile at fe9627c plus #181's display series and other panther commits not yet proposed):

  • echo warm > /sys/kernel/reboot/mode, then reboot-to bootloader: fastboot 3 times out of 3, about 33 s each, with fastboot --set-active=b between tries.
  • Control, cold (the default): 0 of 1, it booted Linux again.
  • Then booted with a DTB carrying this line: /proc/cmdline has reboot=warm and /sys/kernel/reboot/mode reads warm.
  • From that boot, with nothing written by hand (mode warm from the bootargs alone), reboot-to bootloader: fastboot in about 33 s.

Not checked: this commit on omamobile without the other local commits; more than one try from the bootargs line alone.

Checks: tools/kernel-pr-check --build OUT --merge origin/omamobile passes: sign-off, no Co-Authored-By, prefix, checkpatch --strict (one LONG_LINE warning on the bootargs string, which was already over 100 columns), policy, and the 9 device trees with W=1 and dtbs_check with no new warnings.

tools/phone-smoke (the same phone, 2026-11-04T18:47Z)

phone-smoke: panther

Kernel 7.3.0-rc6-omamobile-push1-g44048a9aa795, tools at 3435ed4, 2026-11-04T18:47Z.

Check Result Detail
boot (required) ok kernel 7.3.0-rc6-omamobile-push1-g44048a9aa795, slot b, up 865s
storage guard (required) ok 55 partitions; only userdata and slot B's boot images writable
root (required) ok ext4 on userdata, 97G free
kernel health (required) ok no oops or BUG; 0 WARN, 8 error lines in dmesg (8 known, 0 new)
systemd units ok none failed
USB gadget (required) ok bound to 11210000.usb
display ok card2-DSI-1 connected, 1080x2400
GPU ok render node on panthor
session (required) ok Hyprland and the Omarchy shell running
lock ok PAM file present, the shell can lock (passwordPam true, locked true)
touch ok FocalTech FT3658
buttons ok gpio-keys
Wi-Fi ok wlp1s0 on brcmfmac, link up
Bluetooth ok hci0, powered true
battery ok 91% Charging, 33 °C
sound cards ok googleaocsndcar
sensors ok Accelerometer AmbientLight Proximity
cameras skip no libcamera tools
thermal ok hottest zone 49 °C

Packages: hyprland 0.57.0-3.1, libcamera 0.7.2-4.22, mesa 1:26.2.4-1.4, omacam 0.3.3-1, omamobile-apps 0.2.0-1, omamobile-audio 0.5.13-1, omamobile-device-panther 0.1.5-1, omamobile-firstboot 0.1.3-1, omamobile-gnss 0.2.7-1, omamobile-keyring 20261026-1, omamobile-mirror 0.1.1-1, omamobile-omarchy 0.1.28-1, omamobile-owl-widgets 0.1.1-1, omamobile-sensors 0.2.1-1, omamobile-shell 0.2.79-1, omarchy 4.0.6-1, quickshell 0.3.2.r7.g2b4501a-1

I used AI help (Claude) to work on this.

@derekross

Copy link
Copy Markdown
Owner

Thanks for this, and for the cold control; three out of three against zero out of one is a clear enough answer for the bootloader path. Two things before I merge it:

  1. Every reboot becomes warm, not just reboot-to bootloader. Could you check two more paths from a boot with reboot=warm in the bootargs: an ordinary reboot (reboot-to omamobile or systemctl reboot) comes back to Linux, and, with you watching the phone, what a panic does (echo c > /proc/sysrq-trigger, which with panic=10 resets after 10 s): does it come back to slot B, or stop somewhere? The initramfs's safety net (no USB host, then a panic) relies on that reset spending a bootloader try. On the Pixel 8 Pro it does with reboot=warm; nobody has recorded it on a gs201 phone yet. It costs a boot, so only if you're comfortable with that.
  2. The commit message lags the description. It says reboot-to wasn't repeated from the reboot=warm boot, while your description reports a fastboot from exactly that boot. The commit is what lands, so please update it, along with the results of point 1.

Small one: please add a sentence that a warm reset leaves the PMIC rails powered. On the 7 Pro that hid the accelerometer after a warm reboot until the AoC start power-cycled the sensor rails, and panther's sensor work will want to know.

…reaches fastboot

A cold PSCI reset loses the reboot mode the kernel writes for the
bootloader, so "reboot-to bootloader" boots slot B again. With
/sys/kernel/reboot/mode set to warm at runtime, the Pixel 7 reached
fastboot 3 times out of 3 (about 33 s each); a cold control booted
Linux again (0 of 1). husky carries the same option in its bootargs.

Booted with this line (/proc/cmdline has reboot=warm, and
/sys/kernel/reboot/mode reads warm), with nothing written by hand:
- reboot-to bootloader: fastboot, three times;
- an ordinary reboot: back to Linux on slot B;
- a panic (echo c > /proc/sysrq-trigger, with panic=10): reset, and
  back to Linux on slot B, the reset spending one of slot B's boot
  tries (devinfo retry count 2 to 1), as the initramfs's no-USB-host
  panic relies on.

Every reboot now becomes a warm reset, which leaves the PMIC rails
powered. On the Pixel 7 Pro that hid the accelerometer after a warm
reboot until the AoC start power-cycled the sensor rails; on the
Pixel 7 the sensor proxy still found the accelerometer, light and
proximity sensors after these warm resets.

Assisted-by: LLM
Signed-off-by: Ángel René <17342133+angelrenefm@users.noreply.github.com>
@angelrenefm
angelrenefm force-pushed the pr-panther-reboot-warm branch from d661ac5 to 05e2d83 Compare November 6, 2026 00:39
@angelrenefm

Copy link
Copy Markdown
Contributor Author

Thanks, Derek. Both paths are now checked from a boot with reboot=warm in the bootargs: /proc/cmdline has it, and /sys/kernel/reboot/mode reads warm with nothing written by hand. Same kernel as before, 7.3.0-rc6-omamobile-push1-g44048a9aa795.

  • Ordinary reboot: back to Linux on slot B.
  • Panic: echo c > /proc/sysrq-trigger with panic=10, and Ángel watched the phone. It reset by itself and came back to Linux on slot B. The reset spent one of slot B's boot tries: devinfo's retry count for B went from 2 to 1. So the initramfs's no-USB-host panic still spends a bootloader try on gs201 with a warm reset, as on the 8 Pro.
  • reboot-to bootloader: fastboot twice more today from such boots, both times with nothing written by hand.

I've updated the commit message with these results, and added a paragraph that a warm reset leaves the PMIC rails powered and what the 7 Pro's accelerometer did. On the Pixel 7, the sensor proxy still found the accelerometer, light and proximity sensors after three warm resets in a row. Turning the phone right after the last one rotated the screen (transform 3, then back to 0), so the accelerometer was working, not just present.

I used AI help (Claude) to work on this.

@derekross

Copy link
Copy Markdown
Owner

Thank you, @angelrenefm. That's everything I asked for: an ordinary reboot and a panic both come back to slot B with reboot=warm, and the panic still spends a bootloader try, which is what the initramfs's safety net depends on. Good to have that recorded for a gs201 phone at last, and the sensor check after three warm resets in a row is a nice touch.

On my side, tools/kernel-pr-check --build --merge passes in the project container, the device tree validates with no messages before and after, and this commit merges cleanly with your Wi-Fi and fingerprint PRs in any order. I'll merge it.

@derekross
derekross merged commit 5dec1b5 into derekross:omamobile Nov 7, 2026
1 check passed
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