https://bugzilla.kernel.org/show_bug.cgi?id=218265
--- Comment #13 from Mark Haun (hau...@keteu.org) --- Update: I have not seen a recurrence of the immediate wake issue on 6.7-rc6 as long as all of the /proc/acpi/wakeup sources are set to disabled. The issue does happen most of the time with the boot-up defaults, so I guess it may be worth documenting/logging the results. Maybe someone can comment if that is appropriate here, or in a new bug report, or not considered a bug? At Thorstens suggestion I tried running the amd_s2idle.py script. Somehow it is able to tell me which interrupt/event caused the wake, information not available from dmesg, so that was helpful. In my case, this output ○ Wakeups triggered from IRQs: [7] ○ Woke up from IRQ 7 (IR-IO-APIC 7-fasteoi pinctrl_amd) ○ gpe07 increased from 13618 to 13632 ❌ Userspace suspended for 0:00:03.870541 (< minimum expected 0:00:09) led me to examine /sys/firmware/acpi/interrupts/gpe07: $ cat /sys/firmware/acpi/interrupts/gpe07 12843 EN enabled unmasked I spent some time testing with gpe07 masked/unmasked and disabled/enabled but the results were inconclusive. Sometimes the system remains in suspend and sometimes it doesn't. In *both* cases, the script usually reports that gpe07 was the cause of the wakeup, regardless of the mask or enable status of /sys/firmware/acpi/interrupts/gpe07. It seems like that is just a placebo AFAICT. Hope this helps. -- 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 acpi-bugzilla@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla
