Unsolved
1 Rookie
•
1 Message
0
15
January 31st, 2026 13:19
m18 R1 AMD - Critical Boot Freezes requiring BIOS Reflash & ACPI Errors (USB4/GPP2)
Model: Alienware m18 R1 AMD
CPU: AMD Ryzen 9 7945HX
GPU: AMD Radeon RX 7900M
BIOS Version: 1.21.0
I am experiencing a recurring critical instability issue with my Alienware m18 R1 AMD running Linux. The system frequently freezes during the boot process (specifically during driver initialization/module loading).
The freeze is a "hard lockup" – the system becomes completely unresponsive (no CapsLock response, frozen screen). What makes this issue critical is that a simple power cycle (holding the power button for 30s) often does not resolve it. The system remains in a non-bootable state until I perform a BIOS Recovery/Reflash, which suggests the hardware, EC, or NVRAM enters an invalid/corrupted state that persists across power cycles.
Observations & Logs: While investigating the boot logs (tested on Kernel 6.6 LTS and 6.12), I have noticed consistent ACPI errors appearing right before the system loads chipset/USB drivers.
Specific Error observed:
ACPI Error: AE_NOT_FOUND, While resolving a named reference package element - \_SB_.PCI0.GPP2.UP00.DP00.U4UP.U4P3.UH10
This error points to the USB4 Upstream Port (UP00) on the GPP2 bridge. It appears the BIOS ACPI tables might be referencing a device path that is missing or incorrectly defined.
Troubleshooting Steps Taken (All Failed): I have performed extensive testing to isolate the issue. None of the following steps resolved the freezes permanently:
Kernel Parameters: Tested
nomodeset(graphics disabled),amdgpu.runpm=0,pcie_aspm=off,processor.max_cstate=1, andiommu=soft. The system still hangs even with graphics drivers disabled.BIOS Settings: Disabled "VT for Direct I/O", "USB Emulation", and "USB PowerShare" to rule out IOMMU conflicts and power states.
Kernel Version: Downgraded to Stable LTS Kernel (6.6) to rule out regressions in newer kernels.
ACPI OSI: Attempted
acpi_osi="Windows 2020"spoofing.Hardware Reset: Performed hard RTC resets (30s power button hold). This fails to recover the system; only a BIOS Reflash works.
Current Investigation: I suspect a correlation between the ACPI AE_NOT_FOUND error and the hardware lockups, as the freezes often occur when the OS attempts to initialize the USB4/UCSI subsystem. I am currently testing blacklisting ucsi_acpi, but the intermittent nature of the issue makes it hard to confirm.
Request for Engineering: Could the Alienware Engineering team please review the ACPI tables for the m18 R1 AMD regarding the USB4 controller implementation?
Is the
AE_NOT_FOUNDerror forGPP2.UP00a known issue in the firmware?Are there known conflicts with OS-directed power management (OSPM) for the USB4 controller that could cause such deep hardware lockups?
Any insights would be appreciated, as the requirement to reflash the BIOS to recover the system is very concerning and degrades the user experience.


