On 2023-03-30 03:15, David C. Rankin wrote:

On 3/29/23 14:46, Carlos E. R. wrote:

...

ModemManager[1799]: <info>  [sleep-monitor] system is resuming

Looks like this is the first indication that a wakeup has occured. The next line is Network Manager. Why we have ModemManager running has always been a curiosity, and I've seen reasoning, but I don't recall the details.

I don't know if ModemManager is causing the wake up, or detecting the wake up.

Were it me, I'd kill it and try putting the laptop to sleep again and see if it stays that way. I doubt you have a modem anyway.

Let's see... Nope, didn't work.

(what I do recall is it is used by the likes of gphoto2, etc.. for some serial communications of USB. That is a vague recollection though)

Huh :-? Ah, maybe those usb things that had two behaviours, and something in the computer triggered a change of mode, then usb detected it as something different. Well, now it is NetworkManager who says waking up.

Mar 30 07:49:34 Laicolasse gdm-Xorg-:0[10939]: (II) AMDGPU(0): Modeline "1920x1080"x0.0 160.63 1920 1936 1952 2180 1080 1083 1097 1228 -hsync -vsync (73.7 kHz > Mar 30 07:49:34 Laicolasse systemd-sleep[18725]: INFO: Done. Mar 30 07:49:34 Laicolasse systemd[1]: systemd-hibernate.service: Deactivated successfully. Mar 30 07:49:34 Laicolasse systemd[1]: Finished Hibernate. Mar 30 07:49:34 Laicolasse systemd[1]: Reached target System Hibernation. Mar 30 07:49:34 Laicolasse systemd[1]: Stopped target Sleep. Mar 30 07:49:34 Laicolasse systemd[1]: Stopped target System Hibernation. Mar 30 07:49:34 Laicolasse systemd-logind[1725]: Operation 'sleep' finished. Mar 30 07:49:34 Laicolasse NetworkManager[1831]: <info> [1680155374.8546] manager: sleep: wake requested (sleeping: yes enabled: yes) Mar 30 07:49:34 Laicolasse NetworkManager[1831]: <info> [1680155374.8547] device (eth0): state change: unmanaged -> unavailable (reason 'managed', sys-iface-stat> Mar 30 07:49:34 Laicolasse kernel: Generic FE-GE Realtek PHY r8169-0-300:00: attached PHY driver (mii_bus:phy_addr=r8169-0-300:00, irq=MAC) Mar 30 07:49:34 Laicolasse kernel: Bluetooth: hci0: Device setup in 121088 usecs Mar 30 07:49:34 Laicolasse kernel: Bluetooth: hci0: HCI Enhanced Setup Synchronous Connection command is advertised, but not supported.

(this is with kernel 6.2.8-lp154.3 from Kernel:stable:Backports) (I notice that the ssh session, via wlan, now and then stops responding) Before, I suspected wake on line, so I tried disconnected the cable. At the moment, wlan was not working. Now I called hibernate from a ssh session using wlan. On other machines, the ssh would just stop responding and eventually die. Here... :-? But NetworkManager talks about etho, not wlan. -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)

On 2023-03-30 08:46, Andrei Borzenkov wrote:

On Wed, Mar 29, 2023 at 10:46 PM Carlos E. R. wrote:

...

https://paste.opensuse.org/pastes/46f774d01960

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)

On Fri, Mar 31, 2023 at 10:57 AM Carlos E. R. wrote: ...

...

(same new laptop, Lenovo L14 Gen3, new Leap 15.4 install) ...

There is known issue, reported here:

Bug 1207948 - Lenovo T14s Gen3 AMD resume from hibernation ends in kernel panic

How is a bug for T14s relevant for L14? You have already been told in another bug report that you have a different wireless chip.