Bug 2426980
| Summary: | Unable to resume from sleep on TWO different computers, both running Fedora 43. | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Dominic Clifton <me> |
| Component: | kernel | Assignee: | Justin M. Forbes <jforbes> |
| Status: | NEW --- | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | high | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 43 | CC: | acaringi, adscvr, airlied, alexandre.andre, alfigueiredo, hans, hpa, jforbes, kernel-maint, linville, logistics.and.things, masami256, mchehab, ptalbert, steved, suraj.ghimire7 |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | --- | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | Type: | --- | |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Attachments: | |||
|
Description
Dominic Clifton
2026-01-03 22:40:30 UTC
Created attachment 2121003 [details]
full journalctl log from machine 1 (ASRock x570)
Created attachment 2121004 [details]
full journalctl log from machine 2 (Minisforum MS-S01)
Additionally, both systems report this: machine 1: ``` Jan 03 22:01:15 kwin_wayland[2726]: Pageflip timed out! This is a bug in the nvidia-drm kernel driver Jan 03 22:01:15 kwin_wayland[2726]: Please report this at https://forums.developer.nvidia.com/c/gpu-graphics/linux ``` machine 2: ``` Jan 03 22:15:23 kwin_wayland[2506]: Pageflip timed out! This is a bug in the amdgpu kernel driver Jan 03 22:15:23 kwin_wayland[2506]: Please report this at https://gitlab.freedesktop.org/drm/amd/-/issues ``` I have also tried various workarounds on both machines, such as enabling/disable global c-states in the bios, using kernel args like `mem_sleep_default=deep` and `pcie_aspm=off` independantly, but to no avail. Today, for fun, I updated Machine 2 (The Minisforum MS-S1, AMD Stryx Halo, AMD AI MAX 395+ CPU, using simply `sudo dnf update` This upgraded the kernel from: ``` Linux fedora 6.17.12-300.fc43.x86_64 ``` to ``` Linux fedora 6.18.4-200.fc43.x86_64 #1 SMP PREEMPT_DYNAMIC Thu Jan 8 17:35:49 UTC 2026 x86_64 GNU/Linux ``` ``` $ cat /proc/version Linux version 6.18.4-200.fc43.x86_64 (mockbuild@bc1e0bf216f84bd4841c551523010be9) (gcc (GCC) 15.2.1 20251211 (Red Hat 15.2.1-5), GNU ld version 2.45.1-1.fc43) #1 SMP PREEMPT_DYNAMIC Thu Jan 8 17:35:49 UTC 2026 ``` Many other packages were also updated. On this machine, s2idle is used. ``` $ cat /sys/power/mem_sleep [s2idle] ``` The system will now go to sleep and will also resume from sleep multiple times in a row. I will try and update Machine 1 to the 6.18.4-200 kernel, which is more complicated since it uses Bazzite, and report back. I will also attach a log from Machine 1 which includes the 2 sleep resume cycles. Created attachment 2121908 [details]
machine-2-20260112-two-sleeps.txt
full journalctl log from machine 2 after upgrading to kernel 6.18.4-200.
At the time the kernel update and successful sleep/resume cycles, Minisforum MS-S1 BIOS, version 1.05, had 'Global C-States: DISABLED'. I reset that back to: Global C-States: Auto' and performed the test again, however the machine would not go to sleep, it just woke up immediately, attempting to sleep again resulted in the display locking and unresponsive keyboard (no caps lock activity) and no mouse, the system was available over the network so I grabbed another log. While I was typing this the system became responsive again and I was able to login. The system still does not sleep though. Created attachment 2121909 [details]
machine-2-20260112-02-global-cstates-auto.txt
log from machine 2, with Global C-States set to AUTO in the bios. Machine does not sleep at all.
Related problem: When I close the screen of my laptop onto the keyboard, it goes into sleep mode, but I can't always restart it: the screen remains black and does not display the password prompt. However, if I lock the screen (Super+L) or wait a long time for the computer to go into sleep mode, I can always get the password prompt. Hardware: MacBook Air OS: Fedora 43 / GNOME Related problem: When I lock the screen (Super+L) or wait a long time for the computer to go into sleep mode, one of my screen remains black. When I lock again and unlock back normal. Hardware: AMD Ryzen 5 PRO 2600 x 12 / AMD Radeon RX 550 Series OS: Fedora 43 / GNOME UPDATE: After the last update, the problem is not occurring any more. --- afigueiredo@mac01065269:~$ uname -a Linux mac01065269 6.18.16-200.fc43.x86_64 #1 SMP PREEMPT_DYNAMIC Wed Mar 4 19:13:32 UTC 2026 x86_64 GNU/Linux afigueiredo@mac01065269:~$ more /etc/os-release NAME="Fedora Linux" VERSION="43 (Workstation Edition)" RELEASE_TYPE=stable ID=fedora VERSION_ID=43 VERSION_CODENAME="" PRETTY_NAME="Fedora Linux 43 (Workstation Edition)" ANSI_COLOR="0;38;2;60;110;180" LOGO=fedora-logo-icon CPE_NAME="cpe:/o:fedoraproject:fedora:43" DEFAULT_HOSTNAME="fedora" HOME_URL="https://fedoraproject.org/" DOCUMENTATION_URL="https://docs.fedoraproject.org/en-US/fedora/f43/" SUPPORT_URL="https://ask.fedoraproject.org/" BUG_REPORT_URL="https://bugzilla.redhat.com/" REDHAT_BUGZILLA_PRODUCT="Fedora" REDHAT_BUGZILLA_PRODUCT_VERSION=43 REDHAT_SUPPORT_PRODUCT="Fedora" REDHAT_SUPPORT_PRODUCT_VERSION=43 SUPPORT_END=2026-12-02 VARIANT="Workstation Edition" VARIANT_ID=workstation afigueiredo@mac01065269:~$ more /etc/redhat-release Fedora release 43 (Forty Three) --- As a follow up: The MS-S1 running Linux 6.18.4-200 still seems to suspend resume multiple times OK now. Regarding the ASRock x570 AMD 5950X + NVidia 4090 FE. I upgraded Bazzite to 43.20260313 on the ASRock today, kernel is the same Linux 6.17.7-ba28.fc43.x86_64 as before, system still does not wake from sleep. If someone in the know can point to a specific change that occurred between Linux fedora 6.17.12-300.fc43.x86_64 and Linux fedora 6.18.4-200.fc43.x86_64 (which fixed the MS-S1 box) that would really help with my understanding of the situation. |