i3 System hard freezes on suspend/resume

Jodi42 · 2026-01-23 18:07:38

I am running Arch Linux with i3wm on an HP laptop with an NVIDIA 3070Ti.
I am facing a hard system freeze after suspend (lid close or systemctl suspend).
The system does not wake, keyboard/mouse dead, and I must hard power cycle.

This is not just a black screen or Xorg issue — the whole system hangs.
I have tried disabling xsslock through config file but it still does not work.
https://termbin.com/my6fm

seth · 2026-01-23 19:52:45

I must hard power cycle

This renders any journal borderline useless, but

Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP                
Jan 23 22:45:59 arch kernel: pcieport 0000:00:01.1: AER: Correctable error message received from 0000:01:00.0
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:   device [10de:24a0] error status/mask=00000040/0000a000
Jan 23 22:45:59 arch kernel: nvidia 0000:01:00.0:    [ 6] BadTLP

looks suspicious enough.

You could try https://aur.archlinux.org/packages?O=0&K=580xx and in doubt disable the GSP, make sure the GPU has proper power supply, try to disable "pcie_aspm=off" (though no idea whether is's enabled itfp) and pay attention to any bus errors and there context.
Maybe re-seat the GPU

Jodi42 · 2026-01-24 13:36:25

What exactly is the problem here?I turned off the lidswitch in logind but and then tried again. It still showed the unresponsive black screen. And also can you please tell step by step what to do.

seth · 2026-01-24 14:46:52

Start by posting a complete and coherent journal.

sudo journalctl -b | curl -F 'file=@-' 0x0.st

Then address the bus problems

You could try https://aur.archlinux.org/packages?O=0&K=580xx and in doubt disable the GSP, make sure the GPU has proper power supply, try to disable "pcie_aspm=off"

https://wiki.archlinux.org/title/NVIDIA
https://wiki.archlinux.org/title/NVIDIA … P_firmware
https://wiki.archlinux.org/title/Kernel_parameters

Jodi42 · 2026-01-24 18:43:51

https://termbin.com/ljb8

This is the output for

sudo journalctl -b

seth · 2026-01-24 21:10:57

Jan 25 05:36:37 arch kernel: ACPI FADT declares the system doesn't support PCIe ASPM, so disable it

so "pcie_aspm=off" is gonna be pointless.
It's a hybrid graphics system, there're no outputs on the AMD GPU but noise on its bus as well

Jan 25 05:36:38 arch kernel: amdgpu 0000:06:00.0: [drm] Cannot find any crtc or sizes
Jan 25 05:36:38 arch kernel: pcie_mp2_amd 0000:06:00.7: enabling device (0000 -> 0002)
Jan 25 05:36:38 arch kernel: pcie_mp2_amd 0000:06:00.7: Failed to discover, sensors not enabled is 1
Jan 25 05:36:38 arch kernel: pcie_mp2_amd 0000:06:00.7: amd_sfh_hid_client_init failed err -95

This here might be a massive problem

Jan 25 05:37:07 arch tailscaled[732]: open-conn-track: timeout opening (TCP 100.88.31.10:42402 => 95.216.195.133:80); no associated peer node
Jan 25 05:37:09 arch tailscaled[732]: open-conn-track: timeout opening (TCP 100.88.31.10:42384 => 95.216.195.133:80); no associated peer node
Jan 25 00:07:10 arch systemd-timesyncd[466]: Contacted time server 15.207.248.194:123 (2.arch.pool.ntp.org).
Jan 25 00:07:10 arch systemd-timesyncd[466]: Initial clock synchronization to Sun 2026-01-25 00:07:10.048823 IST.
Jan 25 00:07:10 arch systemd-journald[449]: Time jumped backwards, rotating.

mostly since it strongly suggests that

Jan 25 05:36:38 arch kernel:  nvme0n1: p1 p2 p3 p4 p5 p6

there's a parallel windows installation.

3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.

Also see https://wiki.archlinux.org/title/Dual_b … e_standard

Jodi42 · 2026-01-25 14:52:19

I have windows fast startup disabled. Plus i start i3 with debug log on sddm so is there a way to access that debug log. Problem is display don't come after

 systemctl suspend

or after closing the lid. Rest if system sits idle and display closes then it opens up again with no problem.

seth · 2026-01-25 15:00:11

I do understand what your problem is, but

Plus i start i3 with debug log on sddm so is there a way to access that debug log.

Nothing meaningfull will be logged when you reboot w/ the power button, see whether you can reboot using https://wiki.archlinux.org/title/Keyboa … el_(SysRq) + REISUB

Jan 25 05:36:37 arch kernel: DMI: HP OMEN by HP 16.1 inch Gaming Laptop PC/8A44, BIOS F.25 08/15/2025

Since it's a notebook there's probably also no way to address the bus errors on a physical layer.

seth multiple times wrote:

You could try https://aur.archlinux.org/packages?O=0&K=580xx and in doubt disable the GSP

Then fix the system time maintainance (the drift heavily suggests that you at least ran windows between linux boots?)

Edit: can you disable the GPU or APU in the firmware (BIOS/UEFI)?

Last edited by seth (2026-01-25 15:01:32)

Jodi42 · 2026-01-25 16:28:19

I can't turn off gpu through bios. https://aur.archlinux.org/packages?O=0&K=580xx Do you want me install these packages?? Plus I don't have SysRq key on my keyboard so I can't go with that method you posted. i never encountered this problem before with any other arch distro or with arch KDE. Plus after the opening the lid I can't even ssh into a tty through termius on android. So i guess problem is more in deep than I can resolve

Last edited by Jodi42 (2026-01-25 16:29:19)

seth · 2026-01-25 16:34:13

What about the APU then?

sysrq is typically tthe print key (evtest would report it as such)

Can you still suspend and wake the system
1. from the multi-user.target
2. an install iso

Did you recently ensure that windows fast start is indeed *still* disabled (because it frequently will automatically be re-enabled w/ windows updates)

And yes, the idea would be to test the proprietary driver and notably then also to disable GSP

Jodi42 · 2026-01-25 16:50:12

Then lets reinstall the driver first and see if the issue is still there. I don't think i can even do 1. but yeah 2. install iso, that i can do. i run ML projects on arch dual boot. So will installing your said drivers are going to be a problem or not??

❯ lspci -k -d ::03xx
01:00.0 VGA compatible controller: NVIDIA Corporation GA104 [Geforce RTX 3070 Ti Laptop GPU] (rev a1)
	DeviceName: NVIDIA Graphics Device
	Subsystem: Hewlett-Packard Company Device 8a44
	Kernel driver in use: nvidia
	Kernel modules: nouveau, nvidia_drm, nvidia
06:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Rembrandt [Radeon 680M] (rev c8)
	DeviceName:  Onboard IGD
	Subsystem: Hewlett-Packard Company Device 8a44
	Kernel driver in use: amdgpu
	Kernel modules: amdgpu

This is the config generated by

sudo nvidia-xconfig

https://termbin.com/qju0

Last edited by Jodi42 (2026-01-25 16:57:50)

seth · 2026-01-25 17:00:19

I don't think i can even do 1.

2. link below, there you can run "systemctl suspend"

So will installing your said drivers are going to be a problem or not??

Not if you properly replace the driver and unless you depend on latest additions to the feature branch (ie. if your model training explicitly requires 590xx features - what is not likely)

Jodi42 · 2026-01-25 17:07:13

What do you mean by

2. link below, there you can run "systemctl suspend"

How to install them now. Do i remove the already installed ones before or does paru do that automatically.

https://termbin.com/gpz7 This is the /var/log/Xorg.0.log file

I have no way to access the system after running

 systemctl suspend

. Everything seems frozen and not running.

Last edited by Jodi42 (2026-01-25 17:55:17)

Jodi42 · 2026-01-25 18:13:13

Seems the problem is in systemd.special

 sudo systemctl mask sleep.target suspend.target hibernate.target hybrid-sleep.target

I used this and now after closing the lid i can resume work. although fans keep on working but earlier without this and even in /etc/systemd/logind.conf having following parameters

#HandleLidSwitch=ignore
#HandleLidSwitchDocked=ignore
#SuspendKeyIgnoreInhibited=no

I would still get unresponsive system after opening closing lid and opening lid.

seth · 2026-01-25 20:12:13

I used this and now after closing the lid

the system will simply not fall asleep at all.
Undo that to prevent the system from overheating when throwing up the fans in your backpack.

[    11.974] (==) Using config file: "/etc/X11/xorg.conf"

Remove that file, it's nvidia-settings generated cruft.

What do you mean by

There're 4 links in my signature, the 2nd one directs to http://wiki.archlinux.org/title/Systemd … _boot_into which explains how to move or boot to the multi-user.target

Do i remove the already installed ones before or does paru do that automatically.

I've no idea what paru does, but they'll cause a conflict and you'd typically be asked what to do to resolve it (one option being to remove the present packages)
Generally read https://wiki.archlinux.org/title/Pacman and https://wiki.archlinux.org/title/Arch_User_Repository because you quite frankly ought to use this kind of stuff for basic maintenance of your system.

Jodi42 · 2026-01-25 20:56:47

I think it is some error on hardware level if till now I am not able to resolve as i have tried many commands from chatgpt. Where do you think bug is that needs to be resolved I have provided xorg logs journalctl and etc.

seth · 2026-01-25 20:59:28

Since you're clearly not able to read my responses: https://deepl.com/

Jodi42 · 2026-01-25 21:02:00

I am used to chatting kind of talking. if you have some other platform to talk then fine with me. I don't have much expertise with forums.

seth · 2026-01-25 21:04:11

deepl.com is a translation service and it's increasingly becoming obvious that you don't actually speak English.

Jodi42 · 2026-01-25 21:04:19

How do i report this issue to Arch Developers??

seth · 2026-01-25 21:13:42

You don't.
This is going absolutely nowhere since you're either unable or unwilling to respond to simple questions.

Jodi42 · 2026-01-26 11:04:50

seth wrote:

You don't.
This is going absolutely nowhere since you're either unable or unwilling to respond to simple questions.

https://www.reddit.com/r/pop_os/comment … endresume/ This is also issue reported by many users with nvidia. It is an issue with drivers and system itself and hence i desire to report it to authorities. Why is there no way to do that?? Are you also a developer @Seth

seth · 2026-01-29 21:36:42

This is going nowhere because you're not leading any kind of meaningful discussion.
You ignore questions, you don't run suggested diagnostic steps and so, at post #23, we don't know the actual nature, let alone the vicinity of the cause of your issue.

There're *countless* thread on this forum along about systems not properly waking up from suspend for all sorts of reasons.
What exactly would you want to report and where?
This might be entirely because of a local misconfiguration or because god hates you personally. Nobody, including you, has the slightest idea about that.

I don't know whether that's a language barrier or you're just trolling or anything in between, but unless and until you change your approach there's no point in you reporting anything anywhere and if you're actually facing a problem and are not just trying to waste peoples time your best hope will be that other people run into this and fix it and you just get to benefit from that.

Jodi42 · 2026-02-10 15:21:58

I installed xfce and same issue there as well. this time even masking suspend target and sleep target is not working. As soon as lid is closed and lid is opened again, black screen is there and nothing is reponsive. From last time I have upgraded BIOS and am I running in discrete mode in arch. https://termbin.com/jx97 journal

 journalctl -b | nc termbin.com 9999

seth · 2026-02-10 15:36:19

xfce isn't the multi-user.target, the journal covers 8s and no attempt to sleep anything (so you probably ran that after rebooting) and you're also still on

Feb 10 20:39:37 arch kernel: nvidia-modeset: Loading NVIDIA UNIX Open Kernel Mode Setting Driver for x86_64  590.48.01  Release Build  (root@)

We're still at https://bbs.archlinux.org/viewtopic.php … 4#p2284634

Arch Linux

#1 2026-01-23 18:07:38

i3 System hard freezes on suspend/resume

#2 2026-01-23 19:52:45

Re: i3 System hard freezes on suspend/resume

#3 2026-01-24 13:36:25

Re: i3 System hard freezes on suspend/resume

#4 2026-01-24 14:46:52

Re: i3 System hard freezes on suspend/resume

#5 2026-01-24 18:43:51

Re: i3 System hard freezes on suspend/resume

#6 2026-01-24 21:10:57

Re: i3 System hard freezes on suspend/resume

#7 2026-01-25 14:52:19

Re: i3 System hard freezes on suspend/resume

#8 2026-01-25 15:00:11

Re: i3 System hard freezes on suspend/resume

#9 2026-01-25 16:28:19

Re: i3 System hard freezes on suspend/resume

#10 2026-01-25 16:34:13

Re: i3 System hard freezes on suspend/resume

#11 2026-01-25 16:50:12

Re: i3 System hard freezes on suspend/resume

#12 2026-01-25 17:00:19

Re: i3 System hard freezes on suspend/resume

#13 2026-01-25 17:07:13

Re: i3 System hard freezes on suspend/resume

#14 2026-01-25 18:13:13

Re: i3 System hard freezes on suspend/resume

#15 2026-01-25 20:12:13

Re: i3 System hard freezes on suspend/resume

#16 2026-01-25 20:56:47

Re: i3 System hard freezes on suspend/resume

#17 2026-01-25 20:59:28

Re: i3 System hard freezes on suspend/resume

#18 2026-01-25 21:02:00

Re: i3 System hard freezes on suspend/resume

#19 2026-01-25 21:04:11

Re: i3 System hard freezes on suspend/resume

#20 2026-01-25 21:04:19

Re: i3 System hard freezes on suspend/resume

#21 2026-01-25 21:13:42

Re: i3 System hard freezes on suspend/resume

#22 2026-01-26 11:04:50

Re: i3 System hard freezes on suspend/resume

#23 2026-01-29 21:36:42

Re: i3 System hard freezes on suspend/resume

#24 2026-02-10 15:21:58

Re: i3 System hard freezes on suspend/resume

#25 2026-02-10 15:36:19

Re: i3 System hard freezes on suspend/resume

Board footer