[RESOLVED] Suspend does not work normally on ASUS FA507NV

Ask Fedora
1

Hi there! I’ve recently brought an ASUS FA507NV which I will definitely be carrying around. The issue is, suspending the laptop is slow (11 second re-wake with s2idle), and I have occasionally ran into kernel panics when trying to repair it. I’ve tried AI, but I fear it’s going nowhere, so most changed have been reverted and made sure not to mess up the system any further.

I have used a community tool by Mario Limonciello called amd-debug-tools to check where the issue lies. Here is the summary:
Summary
╒═════════════════════╤════════════╤══════════════════╤═════════════════╤═════════════════╤═════════════════╤════════════╤════
══════════════╕
│ Start Time │ Duration │ Hardware Sleep │ Battery Start │ Battery Delta │ Average Power │ Wake Pin │ Wak
e Interrupt │
╞═════════════════════╪════════════╪══════════════════╪═════════════════╪═════════════════╪═════════════════╪════════════╪════
══════════════╡
│ 2026-09-16 19:16:30 │ 0:00:25 │ 0.00% │ 92.43% │ 0.00% │ 0.00W │ │ ACP
I SCI │
╘═════════════════════╧════════════╧══════════════════╧═════════════════╧═════════════════╧═════════════════╧════════════╧════
══════════════╛
ACPI BIOS errors found
Did not reach hardware sleep state
:zzz: Hardware sleep cycle count: 1
:zzz: Notify devices ['BAT1', 'UBTC', 'NPCF', 'PEGP'] found during suspend

Here are the system specifications:

  • OS: Fedora KDE 44 x86_64
  • Host: ASUS TUF Gamibg A15 FA507NV_FA507NV (1.0)
  • Kernel: Linux 7.2.5-200.fc44.x86_64
  • CPU: AMD Ryzen 7 7735HS (16) @ 4.83 GHz
  • GPU 1: NVIDIA GeForce RTX 4060 Max-Q Mobile [Discrete]
  • GPU 2: AMD Radeon 680M [Integrated]

Would love to get this fixed as soon as possible, as I’ll need to start carrying this around in 3 weeks time, and I would definitely wish for suspend to work normally, or result to any alternative necessary. Thanks for your time reading!

2

Hi there, I wanted to follow up on this post as I have managed to resolve it myself, with further use of AI but mostly general research.

The error here actually came from the iGPU, the last place I would assume to be broken. After running more diagnostic tests on my system, namely, pm-graph, the component that took the full 11 seconds to suspend was infact the iGPU. Investigating dmesg, I found out a constant stream of ERROR messages, both during boot and after suspend wake. Here’s the exact log:

[ 1043.759119] amdgpu 0000:36:00.0: [drm] Wait for DMUB auto-load failed: 3
[ 1043.759124] amdgpu 0000:36:00.0: [drm] DMUB hardware initialized: version=0x0400004A
[ 1043.933963] amdgpu 0000:36:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error
...
[ 1055.499806] PM: resume devices took 12.280 seconds

Online research led to me finding out that this error was a problem of a recent linux firmware and amdgpu firmware patch, namely, 20260910, which affected all AMD Radeon 680M iGPUs across all distros and kernel versions. Hence, the solution to this was reverting the kernel version, and ensuring that dnf does not upgrade them again until a new firmware patch fixes this. Exact code used here:

sudo dnf downgrade amd-gpu-firmware linux-firmware
sudo dracut --regenerate-all --force
sudo dnf versionlock add amd-gpu-firmware linux-firmware
sudo reboot

Downgrading the firmware reversed any issues and left the machine running smoothly. If you are encountering similar issues, the commands above should be able to help, but do take in consideration the fact that this was generated by AI.

Thanks all for reading.