Skip to content

omamobile-device-husky 0.1.10: Suspend means suspend-to-idle on the Pixel 8 Pro - #656

Merged
derekross merged 1 commit into
derekross:mainfrom
gitColinX:husky-s2idle
Nov 11, 2026
Merged

derekross merged 1 commit into
derekross:mainfrom
gitColinX:husky-s2idle

Conversation

@gitColinX

@gitColinX gitColinX commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

The 8 Pro's package had no sleep drop-in, so a tap on Suspend (or systemctl suspend) used the kernel's default, deep, which Linux can't wake from on zuma yet. The 8a's package has had the drop-in since #456; this gives husky the same one.

  • vendor/systemd/sleep.conf.d/10-omamobile.conf: MemorySleepMode=s2idle, no hibernation, hybrid sleep or suspend-then-hibernate. The settings are the Pixel 7 Pro's; only the comment names zuma and the 8 Pro.
  • install stages it to /usr/lib/systemd/sleep.conf.d.
  • test/all: the settings match cheetah's, install stages the file, and it holds MemorySleepMode=s2idle and AllowHibernation=no.
  • README: the file's row. ROADMAP, the Suspend row: deep is the default on my phone too, and the drop-in doesn't cover a test that writes /sys/power/state itself.

Checked on my phone (G1MNW, husky-test-20261109-t1, device package 0.1.8, read only): /sys/power/mem_sleep reads s2idle [deep], there is no mem_sleep_default on the command line, and no sleep.conf.d exists.

Tested: packages/omamobile-device-husky/test/all passes; it fails with the install line removed, and with MemorySleepMode=deep. tools/pr-check passes.

Not tested: suspend itself. It has never run on the 8 Pro (ROADMAP; UFS-PLAN U5 and U6), and I haven't run it. This only keeps Suspend from choosing deep; it doesn't make s2idle known-good. The kernel side, mem_sleep_default=s2idle in zuma-husky.dts as kernel #238 did for the 8a, is a separate PR I can send if you want it.

…ixel 8 Pro

The 8 Pro had no sleep drop-in, so systemd's Suspend used the kernel's
default, deep (/sys/power/mem_sleep reads "s2idle [deep]" on a second
phone too, on husky-test-20261109-t1), which Linux can't wake from on
zuma yet. The drop-in's settings are the Pixel 7 Pro's, as on the 8a
(derekross#456); the test checks that, that install stages the file, and the two
settings that matter. No hibernation: there is no swap.

Suspend itself is still untested on the 8 Pro (ROADMAP, UFS-PLAN U5 and
U6). This only keeps a tap on Suspend from choosing deep.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@derekross

Copy link
Copy Markdown
Owner

@gitColinX thank you, this closes a real gap: six other device packages ship this drop-in (the 8a's among them) and the 8 Pro's didn't, so a tap on Suspend went to deep, which doesn't wake on zuma. I checked it against the tree: the settings are the Pixel 7 Pro's line for line, the published kernel has mem_sleep_default=s2idle for the 7 Pro, the 7, the 8a and the 9a but not in zuma-husky.dts, which agrees with your read of /sys/power/mem_sleep, and 0.1.10 is free on main. I also like that the PR says what it doesn't do: suspend itself stays untested, and a test that writes /sys/power/state still has to select s2idle itself. Merging.

Yes please to the kernel PR that adds mem_sleep_default=s2idle to zuma-husky.dts, as #238 did for the 8a: that also covers rtcwake -m mem, which the drop-in can't.

One note for whoever runs the first suspend test on an 8 Pro, since this makes the Suspend button reachable: the package starts Wi-Fi and has no sleep hook for it (on the Pixel 7 Pro the hook unloads Wi-Fi around a sleep unless the D3 handshake is proven), UFS wants spm_lvl 5 checked first, and NFC must not be polling. That belongs in a short runbook in device/husky before the test, attended and on the cable.

@Somnius, for the record here too: your systemctl suspend runs from #640 and #642 are the first on this phone, so once they are in the husky notes this ROADMAP row can move off "never run".

@derekross
derekross merged commit 09e80bd into derekross:main Nov 11, 2026
2 checks passed
robertzas pushed a commit to robertzas/omarchy-mobile that referenced this pull request Nov 14, 2026
…ixel 8 Pro (derekross#656)

The 8 Pro's package had no sleep drop-in, so systemd's Suspend used the kernel's default, `deep`, which Linux can't wake from on zuma yet. It now ships the Pixel 7 Pro's `sleep.conf.d/10-omamobile.conf` settings (`MemorySleepMode=s2idle`, no hibernation), as the 8a's package does, with a test that the settings match cheetah's and that install stages the file. Suspend itself is not qualified on the 8 Pro by this change.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
gitColinX added a commit to gitColinX/omarchy-mobile that referenced this pull request Nov 15, 2026
Derek asked on derekross#656 for a short runbook in device/husky before an
attended suspend test on an 8 Pro: Wi-Fi across sleep, UFS spm_lvl 5,
NFC not polling. SUSPEND-PLAN.md has husky's read-only state for each
(no sleep hook, so every stage unloads Wi-Fi and the PCIe host;
spm_lvl 5 at boot; no NFC driver bound), the rest of what cheetah and
raven needed (s2idle only, no cpuidle on husky), recovery, and four
stages: pm_test dry runs, one RTC-woken s2idle, systemctl suspend woken
by Power, ten cycles. Nothing in it has run on this unit yet. The
ROADMAP Suspend row and UFS-PLAN U6 point to it.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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