You are not logged in.
Hi there,
I recently had issues with my Samsung Galaxy Book4 related to very slow charging, and thanks to seth I was able to fix that by lying to the ACPI and set the OS to Windows in the bootloader (grub): https://bbs.archlinux.org/viewtopic.php … 5#p2183365.
However, since I did that, my laptop won't suspend correctly (or at least won't recover from a suspend, it won't react to any button press and I need to press the power button for 10-15s to turn it off then reboot)
I've tried removing the extra ACPI config, and suspend indeed works correctly.
However, lying to the ACPI is the one way I found for my laptop to not charge at an absurdly low rate, so I cannot just remove that...
I've tried finding differences in logs that could indicate what went wrong, but I've little experience with Arch so I don't know what to look for exactly...
For now, I've checked journalctl -b | grep acpi and here are the results:
Logs when using Windows 2022 ACPI
Logs when using default config (no acpi set in GRUB)
diff -u windows normal (please note that I removed the dates and times for diff purposes)
I'll continue investigating on my own, but I would appreciate it if someone with more knowledge could lead me to a solution ![]()
Thanks in advance,
Gurvan
UPDATE: I tried reading from dmesg logs, but there is no sensible difference (at least when piped in grep apci)
Last edited by Gurgur (2024-07-12 19:09:45)
Offline
Wow! Turns out that after more thorough searching, especially using "grep suspend" instead of "grep acpi", I found out that Windows uses deep suspend mode, and that default uses "s2idle". After changing the mem_sleep_default kernel parameter to "s2idle", suspend works along with Windows ACPI!!
Proud to have found that by myself, and ashamed to have opened a thread for that (didn't expect myself to find it),
Cheers,
Gurvan
Offline