https://bugzilla.kernel.org/show_bug.cgi?id=221866
--- Comment #9 from David Del Sol ([email protected]) --- Hi Armin, You're right: Linux has default-claimed _OSI("Windows 2022") since v6.12 (dd067afe3f8cb), and I've confirmed it's a silent no-op here - requesting it on 7.1.6-cachyos never even produces the "Added _OSI(Windows 2022)" dmesg line. acpidump attached (gzipped, 1.6 MB raw), plus the decompiled DSDT and the _OSI check summary. The only _OSI checks on this firmware are "Windows 2009"/"Windows 2015" at the DSDT root (writing TSOS, default 0x75), and every use of TSOS is a >= 0x70 comparison. There is no "Windows 2022" reference anywhere in the DSDT or any of the 35 SSDTs. The workaround turned out to be a coincidence. After a 30-second power-button EC reset the bug reproduces on an unmodified kernel: cold boot with the AC adapter plugged in -> ~0.2-2.5 W; cold boot without the adapter, plugging in after the OS is running -> sustained 31.7 W; a warm reboot with the adapter still plugged stays fast (30.5-30.9 W) until the EC reset. Conclusion: the EC firmware latches slow charging when AC is present at power-on. Not an OSI issue and not a kernel bug - I'm withdrawing the kernel patch and leaving this report as the documented firmware quirk (retitled to match). -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug. _______________________________________________ acpi-bugzilla mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla
