On 2023-03-30 08:46, Andrei Borzenkov wrote:
On Wed, Mar 29, 2023 at 10:46 PM Carlos E. R. <robin.listas@telefonica.net> wrote:
Mar 29 20:47:06 Laicolasse kernel: Wakeup pending. Abort CPU freeze.
Ah, didn't see that one. Would be nice if the kernel reported which. With updated kernel from "Kernel:stable:Backports": Mar 30 10:13:22 Laicolasse kernel: ACPI: PM: Preparing to enter system sleep state S4 Mar 30 10:13:22 Laicolasse kernel: ACPI: EC: event blocked Mar 30 10:13:22 Laicolasse kernel: ACPI: EC: EC stopped Mar 30 10:13:22 Laicolasse kernel: ACPI: PM: Saving platform NVS memory Mar 30 10:13:22 Laicolasse kernel: Disabling non-boot CPUs ... Mar 30 10:13:22 Laicolasse kernel: Wakeup pending. Abort CPU freeze Mar 30 10:13:22 Laicolasse kernel: Non-boot CPUs are not disabled Mar 30 10:13:22 Laicolasse kernel: ACPI: EC: EC started Mar 30 10:13:22 Laicolasse kernel: ACPI: PM: Waking up from system sleep state S4 Mar 30 10:13:22 Laicolasse kernel: ACPI: EC: interrupt unblocked Mar 30 10:13:22 Laicolasse kernel: ACPI: EC: event unblocked Mar 30 10:13:22 Laicolasse kernel: pci 0000:00:00.2: can't derive routing for PCI INT A Mar 30 10:13:22 Laicolasse kernel: pci 0000:00:00.2: PCI INT A: no GSI
Some device tried to wake up at the wrong moment. You could try low hanging fruits like disabling wakeup sources in /etc/acpi/wakep
Ok. Oh :-( Laicolasse:~ # ls /etc/acpi/* ls: cannot access '/etc/acpi/*': No such file or directory Laicolasse:~ #
(https://wiki.archlinux.org/title/Power_management/Wakeup_triggers).
Ok, will read.
But there are a lot of reports of similar symptoms on AMD systems, just google for this message.
Ok.
Debugging suspend issues is more art than science. This usually needs second system to capture (debug) console output to start with.
Yes, I have done that, with a true RS232. Ethernet can do it, with some more limits. however, in this case I do have the log file, so maybe I don't need doing that? :-? Thanks for the pointers. Thinking aloud: Laicolasse:~ # cat /proc/acpi/wakeup Device S-state Status Sysfs node GPP4 S0 *enabled pci:0000:00:02.3 GPP7 S0 *disabled GP17 S0 *enabled pci:0000:00:08.1 LID S4 *enabled platform:PNP0C0D:00 SLPB S3 *enabled platform:PNP0C0E:00 Laicolasse:~ # Ok, LID I know what it is. What are the others? Laicolasse:~ # cat /sys/module/acpi/parameters/ec_no_wakeup N Laicolasse:~ # Laicolasse:~ # cat /sys/bus/usb/devices/usb?/power/wakeup disabled disabled disabled disabled disabled Laicolasse:~ # Hum. Must look in bios and disable wake on LAN. Laicolasse:~ # ls /sys/class/wakeup | wc -l 60 Laicolasse:~ # Laicolasse:~ # ls /sys/class/wakeup/*/device/physical_node/power/wakeup | wc -l 20 Laicolasse:~ # Laicolasse:~ # ls /sys/class/wakeup/*/device/power/wakeup | wc -l 15 Laicolasse:~ # Laicolasse:~ # ls /sys/class/wakeup/*/device/path | wc -l 48 Laicolasse:~ # Google: "Wakeup pending. Abort CPU freeze" "Lenovo" "L15" nothing found. "Wakeup pending. Abort CPU freeze" "Lenovo" There are some. [...] Nothing seems to be it. "Wakeup pending. Abort CPU freeze" Reading... -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)