Which Linux distro are you using?
NixOS
Which release version?
Unstable, last update November 27th, 2024
See my flake.lock for exact version.
Which kernel are you using?
Linux 6.11.10
Which BIOS version are you using?
3.03
Which Framework Laptop 16 model are you using?
CPU: AMD Ryzen 7 7840HS
GPU: AMD Radeon 780M [Integrated]
SSDs:
2280 slot: Samsung SSD 970 EVO Plus 500GB
2230 slot: Sabrent SB-2130-512
The OS is installed to the Samsung.
NixOS configuration:
https://github.com/slippyice/nixconfigs
Issue:
The FW will randomly decide to freeze/crash. Sometimes it happens right after waking up from sleep, and other times it just happens when I’m watching youtube. When it crashes the programs begin to become unresponsive, including the desktop (KDE Plasma 6) until it reaches a black screen. Sometimes I can switch into TTY mode, and sometimes I can’t. Even in TTY mode if I tried to log in, it’d freeze and go to a black screen with the cursor blinking in the top left corner.
It was really hard to get any useful debugging info as most of the times this is what was displayed in TTY
I finally was lucky enough to get error messages with meaning:
nvme0 refers to the Samsung SSD.
Most of the time nothing of interest pops up in dmesg while everything is working, but I was lucky enough to get this set of messages one time while gaming:
[ 163.796449] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
[ 344.088433] cros-ec-dev cros-ec-dev.1.auto: Some logs may have been dropped…
[ 1493.562450] i2c_designware AMDI0010:00: i2c_dw_handle_tx_abort: lost arbitration
[ 1539.918593] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
[ 1664.558723] nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10
[ 1664.558735] nvme nvme0: Does your device have a faulty power saving mode enabled?
[ 1664.558739] nvme nvme0: Try “nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off” and report a bug
[ 1664.592253] nvme 0000:03:00.0: enabling device (0000 → 0002)
[ 1664.594492] nvme nvme0: D3 entry latency set to 8 seconds
[ 1664.611686] nvme nvme0: 16/0/0 default/read/poll queues
[ 2377.816783] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
[ 2449.323744] systemd-ssh-generator[32166]: Disabling SSH generator logic, since sshd is not installed.
[ 2637.285267] systemd-ssh-generator[33470]: Disabling SSH generator logic, since sshd is not installed.
[ 3042.016815] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
[ 3087.181154] systemd-ssh-generator[51843]: Disabling SSH generator logic, since sshd is not installed.
[ 3402.659062] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
[ 3490.681897] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
[ 3494.439296] EXT4-fs (nvme1n1p2): mounted filesystem 132d7d7d-1812-4ebf-b837-ffe86441cf90 r/w with ordered data mode. Quota mode: none.
[ 6501.776987] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
[ 6584.592966] usb 1-4.1: reset full-speed USB device number 9 using xhci_hcd
So I am assuming something is up with power management, SSDs, and the FW16.
Ever since an update in like September I have been having issue with power management. With powertop enabled I was having issues the keyboard repeating keys frequently, and USB-A mouse disconnecting, and KDE Plasma was no longer able to communicate with powertop (improper power mode). So I disabled it by doing:
powerManagement.powertop.enable = false;
services.power-profiles-daemon.enable = false;
By doing that I believe it defaults to TLP.
I can revert to a flake.lock from August and the issues seem to go away. Which is not ideal and prevents me from updating. The issues seem to also be kernel independent.
