[RESOLVED]Suspend don't work after upgrade

barney · 2025-03-29 01:49:50

After upgrade I have problems with suspend. The system is trying to go in suspend mode but the power is not turned off, fans are spinning and the keyboard is blocked.
Here is my journal:
http://0x0.st/8eEH.txt
Here is dmesg:
http://0x0.st/8eE8.txt
What I have tried:
Disable/Enable ASPM in BIOS.
Also, when ASPM is enabled in BIOS I have tried 'ASPM controlled by BIOS' and 'ASPM controlled by OS'
Always the same result: suspend don't work.

Last edited by barney (Yesterday 18:52:49)

mackin_cheese · 2025-03-29 02:35:29

mind telling us your motherboard / cpu / gpu?

barney · 2025-03-29 03:10:32

Output from dmidecode:

# dmidecode 3.6
Getting SMBIOS data from sysfs.
SMBIOS 3.6.0 present.

Handle 0x0002, DMI type 2, 15 bytes
Base Board Information
	Manufacturer: Gigabyte Technology Co., Ltd.
	Product Name: B760M DS3H DDR4
	Version: x.x
	Serial Number: Default string
	Asset Tag: Default string
	Features:
		Board is a hosting board
		Board is replaceable
	Location In Chassis: Default string
	Chassis Handle: 0x0003
	Type: Motherboard
	Contained Object Handles: 0

Output from hwinfo:

14: PCI 100.0: 0300 VGA compatible controller (VGA)
  [Created at pci.386]
  Unique ID: VCu0.flkTNZehh12
  Parent ID: vSkL.LsEKNtpMAPB
  SysFS ID: /devices/pci0000:00/0000:00:01.0/0000:01:00.0
  SysFS BusID: 0000:01:00.0
  Hardware Class: graphics card
  Model: "ATI Lexa PRO [Radeon 540/540X/550/550X / RX 540X/550/550X]"
  Vendor: pci 0x1002 "ATI Technologies Inc"
  Device: pci 0x699f "Lexa PRO [Radeon 540/540X/550/550X / RX 540X/550/550X]"
  SubVendor: pci 0x1002 "ATI Technologies Inc"
  SubDevice: pci 0x0b04
  Revision: 0xc7
  Driver: "amdgpu"
  Driver Modules: "amdgpu"
  Memory Range: 0x4100000000-0x41ffffffff (ro,non-prefetchable)
  Memory Range: 0x4080000000-0x40801fffff (ro,non-prefetchable)
  I/O Ports: 0x5000-0x5fff (rw)
  Memory Range: 0x40e00000-0x40e3ffff (rw,non-prefetchable)
  Memory Range: 0x40e40000-0x40e5ffff (ro,non-prefetchable,disabled)
  IRQ: 143 (50030 events)
  Module Alias: "pci:v00001002d0000699Fsv00001002sd00000B04bc03sc00i00"
  Driver Info #0:
    Driver Status: amdgpu is active
    Driver Activation Cmd: "modprobe amdgpu"
  Config Status: cfg=new, avail=yes, need=no, active=unknown
  Attached to: #12 (PCI bridge)

Primary display adapter: #14

Also output from hwinfo:

13th Gen Intel(R) Core(TM) i5-13400F

seth · 2025-03-29 08:20:35

Try the LTS kernel, s2idle instead of S3/deep, https://wiki.archlinux.org/title/Power_ … end_method and "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off", https://wiki.archlinux.org/title/Solid_ … ST_support

barney · 2025-03-30 14:10:38

seth wrote:

Try the LTS kernel, s2idle instead of S3/deep, https://wiki.archlinux.org/title/Power_ … end_method and "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off", https://wiki.archlinux.org/title/Solid_ … ST_support

I have tried all possible combinations but no luck, suspend don't work.
I think that this is not kernel issue. This is driver issue with the amdgpu.
I have similar problems with Nvidia, now I have the same problems with AMD.

seth · 2025-03-30 20:10:34

amdgpu and nvidia are part of the kernel and it's not very likely that both show the same problem.

Are you suggesting s2idle produces the same results as S3?
Please post a journal of that where you're also using the LTS kernel and disabled APST.

Then there's https://bbs.archlinux.org/viewtopic.php?id=296954 …

GMP · 2025-03-31 09:32:43

You've tested a lot already, which is great—but to move forward, the key thing is to reduce variables and capture logs under a clearly defined setup.

As Seth suggested, try this config:
- LTS kernel
- Suspend mode: s2idle
- Kernel parameters:

  nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off

- APST disabled

Then suspend, and post the journal log from that session.

Also, from this related thread:
https://bbs.archlinux.org/viewtopic.php?id=296954

→ Disabling IOAPIC entries 24–119 in the BIOS resolved suspend issues for users with similar hardware.
This points to platform-level interrupt problems interfering with resume behavior.

If your BIOS exposes IOAPIC settings (sometimes under advanced/chipset), try disabling entries 24–119 if possible—or toggle any "IOAPIC support" setting entirely.

Make sure to test and log under the LTS+s2idle+no-APST config regardless, or it's impossible to isolate root causes.

barney · Yesterday 18:52:23

GMP wrote:

You've tested a lot already, which is great—but to move forward, the key thing is to reduce variables and capture logs under a clearly defined setup.
As Seth suggested, try this config:
- LTS kernel
- Suspend mode: s2idle
- Kernel parameters:
  nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off
  
- APST disabled
Then suspend, and post the journal log from that session.
Also, from this related thread:
https://bbs.archlinux.org/viewtopic.php?id=296954
→ Disabling IOAPIC entries 24–119 in the BIOS resolved suspend issues for users with similar hardware.
This points to platform-level interrupt problems interfering with resume behavior.
If your BIOS exposes IOAPIC settings (sometimes under advanced/chipset), try disabling entries 24–119 if possible—or toggle any "IOAPIC support" setting entirely.
Make sure to test and log under the LTS+s2idle+no-APST config regardless, or it's impossible to isolate root causes.

I have tested what Seth suggested, it is not working.
After that I have tested what Seth suggested + IOAPIC entries 24-119 disabled. It worked.
In the end I have removed kernel parameters and tested with the latest kernel + ASPM disabled + IOAPIC entries 24-119 disabled. It worked.
Probably the IOAPIC setting is the most important. I will mark this thread as resolved.
I don't know what you mean with APST?
Thanks @GMP.

barney · Today 06:41:06

Sorry about that I don't know what APST means. I have read the Arch Wiki about Solid State Drives/Nvme.
As I said I have tested the newest kernel without any kernel parameters, except for root, and it is working.
I have also tested the newest kernel + ASPM enabled + ASPM controled by OS + IOACPI entries 24-119 disabled and again it is working.
The only valid setting for my problem is IOACPI entries 24-119.
Now I know what they mean when they say "If you are using Arch, you will not have life outside Arch:)".
EDIT: If there is a need for my logs I will post them.

Last edited by barney (Today 06:43:37)

seth · Today 07:45:15

I don't know what APST means. I have read the Arch Wiki about Solid State Drives/Nvme

Including the paragrap that addresses APST issues?
https://wiki.archlinux.org/title/Solid_ … ST_support
And the one above that explains what APST is?
https://wiki.archlinux.org/title/Solid_ … tes_(APST)

However

The only valid setting for my problem is IOACPI entries 24-119.

The only other thread on this board addressing teh IOAPIC entries seems https://bbs.archlinux.org/viewtopic.php?id=299583
The one linked by GMP is actually about the user session freeze.

But google knows those entries quite a lot.

Arch Linux

#1 2025-03-29 01:49:50

[RESOLVED]Suspend don't work after upgrade

#2 2025-03-29 02:35:29

Re: [RESOLVED]Suspend don't work after upgrade

#3 2025-03-29 03:10:32

Re: [RESOLVED]Suspend don't work after upgrade

#4 2025-03-29 08:20:35

Re: [RESOLVED]Suspend don't work after upgrade

#5 2025-03-30 14:10:38

Re: [RESOLVED]Suspend don't work after upgrade

#6 2025-03-30 20:10:34

Re: [RESOLVED]Suspend don't work after upgrade

#7 2025-03-31 09:32:43

Re: [RESOLVED]Suspend don't work after upgrade

#8 Yesterday 18:52:23

Re: [RESOLVED]Suspend don't work after upgrade

#9 Today 06:41:06

Re: [RESOLVED]Suspend don't work after upgrade

#10 Today 07:45:15

Re: [RESOLVED]Suspend don't work after upgrade

Board footer