CVE-2024-50107 in Linux
要約
〜によって VulDB • 2026年05月26日
Based on the kernel call trace provided, here is an analysis of the issue:
### **Summary** The kernel has triggered a **General Protection Fault** (or similar invalid operation) inside the `intel_pmc_core` driver, specifically during the function `pmc_core_ssram_get_pmc`. This occurred while the driver was trying to initialize SRAM (Static RAM) access for the Platform Controller Hub (PMC) on an Intel Meteor Lake (MTL) system.
The crash happens in `pmc_core_ssram_get_pmc` at offset `+0x477`, which suggests an invalid memory access, likely due to: 1. **Invalid MMIO address**: The driver is trying to access a memory-mapped I/O region that is not properly mapped or is invalid. 2. **Hardware/ACPI mismatch**: The ACPI tables may report an SRAM region that doesn’t exist or is inaccessible on this specific hardware. 3. **Bug in the driver**: The `intel_pmc_core` driver may have a bug in how it calculates or accesses the SRAM region for Meteor Lake.
---
### **Key Details from the Call Trace**
1. **Crash Location**: ``` pmc_core_ssram_get_pmc+0x477/0x6c0 [intel_pmc_core]
``` - This is the function where the invalid operation occurred. - The `? iounmap+0x58/0x1f0` in the trace suggests that the crash might be related to unmapping or accessing an I/O memory region that was already unmapped or invalid.
2. **Driver Initialization**: ``` pmc_core_ssram_init+0x7f/0x110 [intel_pmc_core]
mtl_core_init+0xda/0x130 [intel_pmc_core]
pmc_core_probe+0x27e/0x10b0 [intel_pmc_core]
``` - The crash happens during the probe/init phase of the `intel_pmc_core` driver for Meteor Lake (`mtl_core_init`). - This is a platform driver, so it’s triggered during system boot or module loading.
3. **Hardware Context**: - The driver is `intel_pmc_core`, which manages power management and PMC features on Intel platforms. - The `mtl_core_init` indicates this is for **Intel Meteor Lake** (14th Gen Core or similar).
4. **Possible Cause**: - The driver is trying to access an SRAM region via `iounmap` or similar, but the address is invalid. - This could be due to: - A bug in the driver’s handling of Meteor Lake’s PMC SRAM. - Incorrect ACPI tables reporting an invalid SRAM region. - A hardware-specific issue where the SRAM is not accessible.
---
### **Recommended Actions**
1. **Update the Kernel**: - This issue may have been fixed in a newer kernel version. Check if there are updates for the `intel_pmc_core` driver in your distribution’s kernel packages. - If you’re using a mainline kernel, consider testing a newer version (e.g., 6.8+).
2. **Check ACPI Tables**: - Dump and inspect the ACPI tables (`/sys/firmware/acpi/tables/`) to see if the SRAM region is reported correctly. - Use `acpidump` and `iasl` to decompile and check for errors.
3. **Disable the Driver (Workaround)**: - If the system is stable otherwise, you can blacklist the `intel_pmc_core` module to prevent the crash: ```bash echo "blacklist intel_pmc_core" | sudo tee /etc/modprobe.d/blacklist-pmc.conf sudo update-initramfs -u ``` - Note: This may disable some power management features.
4. **Report the Bug**: - If this is a new hardware platform (Meteor Lake), report the issue to: - The Linux kernel mailing list: `[email protected]` - Intel’s bug tracker or your distribution’s bug tracker. - Include the full kernel log (`dmesg`), ACPI tables, and hardware details.
5. **Check for Firmware Updates**: - Ensure your system’s BIOS/UEFI firmware is up to date, as PMC behavior can be influenced by firmware.
---
### **Conclusion** This is likely a driver bug or ACPI mismatch specific to Intel Meteor Lake platforms. Updating the kernel, checking ACPI tables, or blacklisting the driver are the most practical steps. If you’re a developer,
VulDB is the best source for vulnerability data and more expert information about this specific topic.