Repository navigation
device/tegu SLEEP-PLAN, DEVICES: the overnight series, 126 of 127 sleeps woke - #347
Conversation
|
Thank you, 126 of 127 is a useful number, and the counts now agree between the plan and the matrix. One note for next time: AGENTS asks for attended sleep tests on the 9a, and this series ran unattended overnight; please check with me before another unattended run. I'll update the AGENTS gotcha's counts myself. Merging; whichever of this and #372 lands second resolves the DEVICES row. |
…eps woke Two hundred 120 s sleeps were planned overnight from 2026-11-01 to 11-02 with the console kept through the freeze: 126 woke clean and the 127th never did, and a Power + Volume Up reset kept nothing in ramoops. The status lines count six no-wakes now.
44f67c7 to
a6a32dc
Compare
|
I tried to merge this, but the DEVICES row now conflicts with main (other PRs landed in the same table today). Could you merge main and keep both changes in that row? Then it goes straight in. |
|
Done: rebased on main. The 9a's row keeps main's cells (the modem cell from #356 and the rest) with this PR's Suspend cell. |
|
Thanks for the rebase. The 9a row keeps main's modem cell and takes your Suspend cell, and the counts match the plan. Merging; one of the telephony PRs will have to rebase this row again when it lands after this. |
…eps woke (derekross#347) Two hundred 120 s sleeps were planned overnight from 2026-11-01 to 11-02 with the console kept through the freeze: 126 woke clean and the 127th never did, and a Power + Volume Up reset kept nothing in ramoops. The status lines count six no-wakes now.
The palmtop ran an overnight series of 120 s sleeps from 2026-11-01 to 11-02:
systemctl suspendwith an RTC alarm, auto-sleep off for the run, the console kept through the freeze, cable in, the modem never loaded, NFC polling stopped by the sleep hook. 126 woke clean (suspend_stats126 and 0, wake IRQ 56 each time, no UFS or ACPM lines, Wi-Fi back each time). The 127th never woke and needed Power + Volume Up in the morning; it entered that sleep exactly like the others, and nothing was kept in ramoops, so a button reset doesn't keep that region either.SLEEP-PLAN gets the record after "A day of use with auto-sleep on" and a status line that counts six no-wakes since 10-28. DEVICES' Suspend cell says six instead of five, with the two rates (one in 21 on 11-01, one in 127 overnight). Suspend stays at testing.