I use oculink to connect my RTX 3080 under fedora 43, but it can only run at PCie Gen1 x4 both at idle and under load.
Then i switched to Windows 11 and everything works great with full PCIe Gen4 x4 bandwidth, good performance.
I found something in kernel log
pci 0000:00:01.1: broken device, retraining non-functional downstream link at 2.5GT/s
pci 0000:01:00.0: 8.000 Gb/s available PCIe bandwidth, limited by 2.5 GT/s PCIe x4 link at 0000:00:01.1 (capable of 252.048 Gb/s with 16.0 GT/s PCIe x16 link)
pci 0000:00:01.1: ASPM: current common clock configuration is inconsistent, reconfiguring
One solution i found is manually changing PCie port speed using setpci with commands below
sudo setpci -s 0000:00:01.1 CAP_EXP+30.L=4:4
sudo setpci -s 0000:00:01.1 CAP_EXP+10.L=20:20
It didn’t work though. AI also advised me to try this script, PCie_set_speed, but maked no difference.
I also tried these kernel parameter, but none of them could help.
sudo grubby --update-kernel=ALL --args="pcie_aspm=off pcie_port_pm=off"
Then i switched to Linux Mint, and find same kernel log and same limited PCie port speed
For now, i find a kernel patch that might prevent this problem, but i’m sure.
Do you have any ideas?
Detailed system setup:
- Operating System: Fedora Linux 43
- KDE Plasma Version: 6.6.2
- KDE Frameworks Version: 6.24.0
- Qt Version: 6.10.2
- Kernel Version: 6.19.7-200.fc43.x86_64 (64-bit)
- Graphics Platform: Wayland
- Processors: 16 × AMD Ryzen 7 H 255 w/ Radeon 780M Graphics
- Memory: 32 GiB of RAM (27.2 GiB usable)
- Graphics Processor 1: AMD Radeon 780M Graphics
- Graphics Processor 2: NVIDIA GeForce RTX 3080
- Manufacturer: LENOVO
- Product Name: 21VS
- System Version: ThinkBook 16 G8+ AHP(with built-in tgx port)