System reboots ~10 seconds after I try to suspend/hibernate since 6.16

alba4k · 2025-11-12 17:35:32

Hey,

I am unable to properly suspend (or hibernate) my system.

After around 10 seconds of being suspended (10 seconds during which my system is able to wake up and starts running just fine) it randomly restarts. This happens for all kernels after 6.15.9 (in the core repo).

Here is an uploaded journal: http://0x0.st/KsAS.txt

go to

ago 18 22:16:36 dell-xps systemd-logind[1094]: The system will suspend now!

Some discussion has previously happened here.

I'm running an XPS 9320 with an i7-1260P. It has an ipu6 camera and I have intel-ivsc-firmware installed.

All I found after attempting a bisection using kernels found here is that version all the way up to 6.16.rc7.140 are able to maintain suspension properly (later ones fail).

6.17.7 is the latest official release as of today, and it's still broken.

Last edited by alba4k (2025-11-12 17:42:40)

seth · 2025-11-12 19:25:37

rc7 is the latest rc in https://github.com/gregkh/linux/commits/v6.16 (rc8 never got tagged)
There ~75 commits inbetween, so likely 6 bisection steps

ago 18 22:16:36 dell-xps systemd-logind[1094]: The system will suspend now!
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.0947] manager: sleep: sleep requested (sleeping: no  enabled: yes)
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.0949] device (p2p-dev-wlp0s20f3): state change: disconnected -> unmanaged (reason 'unmanaged-sleeping', managed-type: 'full')
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.0954] device (18:26:54:3C:0F:AD): state change: disconnected -> unmanaged (reason 'unmanaged-sleeping', managed-type: 'full')
ago 18 22:16:36 dell-xps polkitd[1122]: Unregistered Authentication Agent for unix-process:5065:7804 (system bus name :1.534, object path /org/freedesktop/PolicyKit1/AuthenticationAgent, locale it_CH.UTF-8) (disconnected from bus)
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.0958] manager: NetworkManager state is now ASLEEP
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.0960] device (wlp0s20f3): state change: activated -> deactivating (reason 'sleeping', managed-type: 'full')
ago 18 22:16:36 dell-xps systemd[1]: Starting Network Manager Script Dispatcher Service...
ago 18 22:16:36 dell-xps systemd[1]: Started Network Manager Script Dispatcher Service.
ago 18 22:16:36 dell-xps kernel: wlp0s20f3: deauthenticating from 00:31:92:28:11:86 by local choice (Reason: 3=DEAUTH_LEAVING)
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: nl80211: send_event_marker failed: Source based routing not supported
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: wlp0s20f3: CTRL-EVENT-DISCONNECTED bssid=00:31:92:28:11:86 reason=3 locally_generated=1
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: wlp0s20f3: CTRL-EVENT-DSCP-POLICY clear_all
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2324] device (wlp0s20f3): supplicant interface state: completed -> disconnected
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2326] device (wlp0s20f3): state change: deactivating -> disconnected (reason 'sleeping', managed-type: 'full')
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2330] dhcp4 (wlp0s20f3): canceled DHCP transaction
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2330] dhcp4 (wlp0s20f3): activation: beginning transaction (timeout in 45 seconds)
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2330] dhcp4 (wlp0s20f3): state changed no lease
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2332] dhcp6 (wlp0s20f3): canceled DHCP transaction
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2332] dhcp6 (wlp0s20f3): state changed no lease
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.2556] device (wlp0s20f3): set-hw-addr: set MAC address to 4A:17:29:16:A4:2B (scanning)
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.3058] device (wlp0s20f3): state change: disconnected -> unmanaged (reason 'unmanaged-sleeping', managed-type: 'full')
ago 18 22:16:36 dell-xps NetworkManager[1090]: <info>  [1755548196.4375] device (wlp0s20f3): set-hw-addr: reset MAC address to 38:7A:0E:DB:63:BB (unmanage)
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: p2p-dev-wlp0s20: CTRL-EVENT-DSCP-POLICY clear_all
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: p2p-dev-wlp0s20: CTRL-EVENT-DSCP-POLICY clear_all
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: nl80211: deinit ifname=p2p-dev-wlp0s20 disabled_11b_rates=0
ago 18 22:16:36 dell-xps systemd[1]: Reached target Sleep.
ago 18 22:16:36 dell-xps systemd[1]: Starting System Suspend...
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: wlp0s20f3: CTRL-EVENT-DSCP-POLICY clear_all
ago 18 22:16:36 dell-xps systemd[1]: user@1000.service: Unit now frozen-by-parent.
ago 18 22:16:36 dell-xps systemd[1]: session-1.scope: Unit now frozen-by-parent.
ago 18 22:16:36 dell-xps systemd[1]: user-1000.slice: Unit now frozen-by-parent.
ago 18 22:16:36 dell-xps systemd[1]: user.slice: Unit now frozen.
ago 18 22:16:36 dell-xps systemd-sleep[5109]: Successfully froze unit 'user.slice'.
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: wlp0s20f3: CTRL-EVENT-DSCP-POLICY clear_all
ago 18 22:16:36 dell-xps wpa_supplicant[1123]: nl80211: deinit ifname=wlp0s20f3 disabled_11b_rates=0
ago 18 22:16:36 dell-xps systemd-sleep[5109]: Performing sleep operation 'suspend'...
ago 18 22:16:36 dell-xps kernel: PM: suspend entry (s2idle)
-- Boot 4691f4c477e749e689c4d242473ec090 --

Can you enter S3/deep sleep?
https://wiki.archlinux.org/title/Power_ … end_method

ago 18 22:17:02 archlinux kernel:  nvme0n1: p1 p2
ago 18 22:17:02 archlinux systemd[1]: Found device PM9A1 NVMe Samsung 512GB 2.
ago 18 22:17:05 dell-xps systemd[1]: Condition check resulted in PM9A1 NVMe Samsung 512GB 1 being skipped.
ago 18 22:17:06 dell-xps kernel: FAT-fs (nvme0n1p1): Volume was not properly unmounted. Some data may be corrupt. Please run fsck.
ago 18 22:17:18 dell-xps kernel: nvme nvme0: using unchecked data buffer
ago 18 22:17:18 dell-xps kernel: block nvme0n1: No UUID available providing old NGUID

https://wiki.archlinux.org/title/Solid_ … leshooting

nvme_core.default_ps_max_latency_us=0 pcie_port_pm=off iommu=soft

You could also add "pcie_aspm=off" but

ago 17 00:17:20 archlinux kernel: ACPI FADT declares the system doesn't support PCIe ASPM, so disable it

ASPM isn't supported anyway?

alba4k · Yesterday 20:54:58

Regarding the kernel bisection, is there somewhere where I can find compiled kernels?

I don't this I can enter S3, as all I see In /sys/power/mem_sleep is "[s2idle]" (and I don't see anything in my fw settings to change this, just an option to block S3, but that's disabled)

I'll try adding those 4 kernel args but I don't expect much to change (edit: yeah no changes)

Last edited by alba4k (Yesterday 21:12:03)

seth · Yesterday 21:00:58

gromit or loqs might volunteer to assist you with the bisection as build-monkeys, but there's no repo that holds build of every single kernel patch precompiled
(the beauty of bisection is the logarithmic scale, ie the ability to run through tens of thousands of patches with some dozen actual builds)

alba4k · Yesterday 21:15:00

So what, should I just try contacting one of them (referencing this post) and kindly ask? That feels a bit off haha

Last edited by alba4k (Yesterday 21:15:08)

seth · Yesterday 21:28:29

"haha" - if you see no way to bisect the kernel yourself.
https://wiki.archlinux.org/title/Bisect … s_with_Git
https://wiki.gentoo.org/wiki/Kernel_git-bisect

Have you already checked the nvme behavior?

alba4k · Yesterday 21:39:08

I guess I'll take a look...

Regarding the NVME, those messages appeareven in LTS (http://0x0.st/Kp_m.txt) with

 nvme_core.default_ps_max_latency_us=0 pcie_port_pm=off iommu=soft pcie_aspm=off

I can still find these messages with 6.17.7 with that cmdline http://0x0.st/Kp_N.txt

I would say that not much is changed by those. I got rid of the fsck "reminder" by running fsck, but "sudo fsck.fat -r /dev/nvme0n1p1" still sometimes complains. I'm not sure this is relevant to the main issue however..?

No noteworthy difference when removing that cmdline (http://0x0.st/Kp_e.txt)

Last edited by alba4k (Yesterday 21:53:51)

seth · Yesterday 21:47:05

You're not looking for any specific messages, I just pointed out that the nvme is there and maybe not the most reliable one to reason the approach.
"Some data may be corrupt." is from unclean shutdowns leading to the dirty bit on the /boot vfat.

alba4k · Yesterday 21:49:15

That is probably my fault and not the ssd's.

Regarding the bisect, this is probably a useless check but I can just use the AUR linux-git right?

Last edited by alba4k (Yesterday 21:56:29)

seth · Yesterday 21:55:41

It's a direct consequence of the involuntary reboots and not the issue here.
So adding "nvme_core.default_ps_max_latency_us=0 pcie_port_pm=off iommu=soft pcie_aspm=off" didn't do anything about those?

You'll have to bisect this then - a cold reboot out of s2idle sounds like something triggers some hardware defect and we'll not get any data to analyze for that
My last idea would be https://wiki.archlinux.org/title/Intel_ … up_from_S3

alba4k · Yesterday 22:01:05

Actually, being able to use a PKGBUILD is 100% good enough for me even if it's not a compiled package, thanks for suggesting that. I don't know why I didn't think of it myself.

See my comment on the previous thread - I'll try bisecting based on linux-git from 89be9a8 (tag 6.16-rc7, should be good, will check) to 038d61f (tag 6.16, should be bad, will checked)

Let's hope this doesn't take an eternity, which it likely will

the nvme stuff is probably unrelated as I've seen it happen even before this whole issue, and honestly I don't care as long as it doesn't lead to further complications. I know it should be looked into. But it's not really the priority at the moment

Last edited by alba4k (Today 22:47:04)

alba4k · Today 16:18:24

please ignore my previous edits, I have no clue what was happening. rc7 is good now after testing it again, and so is rc5. I am 90% sure I saw a reboot on rc7 last time and 100% rc5 was misbehaving, but maybe those were unrelated things related to the many reboots on different kernels or smth. I will resume the bisection considering rc7 as good

Last edited by alba4k (Today 22:46:49)

Arch Linux

#1 2025-11-12 17:35:32

System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#2 2025-11-12 19:25:37

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#3 Yesterday 20:54:58

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#4 Yesterday 21:00:58

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#5 Yesterday 21:15:00

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#6 Yesterday 21:28:29

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#7 Yesterday 21:39:08

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#8 Yesterday 21:47:05

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#9 Yesterday 21:49:15

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#10 Yesterday 21:55:41

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#11 Yesterday 22:01:05

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

#12 Today 16:18:24

Re: System reboots ~10 seconds after I try to suspend/hibernate since 6.16

Board footer