Hello Intel Community,
I'm experiencing a kernel panic on my Intel Core i7-13700 (13th Gen, Raptor Lake) when modifying the CR0 register in a C program with inline assembly. This issue occurs only on Linux kernel versions 6.8 and above (tested on 6.8 and 6.10). Interestingly, I tested the same code on systems with Intel Core i7-14700 (14th Gen) and i7-8700K (8th Gen), and no kernel panic occurs on those CPUs. I suspect this may be related to the reported 13th Gen instability issues.
System Details:
- CPU: Intel Core i7-13700
- OS: Ubuntu [24.04], Alma 10, Linux kernel 6.8+
- BIOS: Vendor: Phoenix Technologies LTD, Version: 6.00, Release Date: 11/12/2020
- Microcode: 0x12F (confirmed via cat /proc/cpuinfo | grep microcode)
Issue Description
My program toggles the CR0 Cache Disable (CD) bit using inline assembly. Below is the simplified code:
This code works fine on kernel 6.7 or lower, as well as on i7-14700 and i7-8700K systems running kernel 6.8+. However, on the i7-13700 with kernel 6.8+, it triggers a kernel panic immediately after the mov %0, %%cr0 instruction. The dmesg log shows a general protection fault or triple fault. I have attached the panic log capture below (via Pastebin or as an attachment).
Questions:
- Is this related to the known Raptor Lake voltage instability issue? Why does it only affect i7-13700 and not i7-14700 or i7-8700K?
- Are there specific kernel patches in 6.8+ that could affect CR0 handling?
- Any debugging tips (e.g., specific dmesg flags or tools like Bochs)?
- Should I contact Intel for an RMA, or is this likely a software issue?
Any insights or similar experiences would be greatly appreciated!
Thank you!
Link Copied
