CVE-2026-23469 in Linux
Riassunto
di VulDB • 30/06/2026
Based on the kernel log provided, here is a detailed analysis of the crash and recommended troubleshooting steps.
### **Summary** The system experienced an **Asynchronous SError Interrupt (ASEI)** leading to a Kernel Panic. This occurred in the context of the GPU interrupt handler (`irq/234-gpu`) on a **Texas Instruments AM68 SK** board running Linux kernel version `6.17.7-B2C-00005`.
The crash was triggered by an asynchronous hardware error (likely from the SoC, memory subsystem, or GPU IP) while processing interrupts for the GPU driver. The presence of `[M]=MACHINE_CHECK` in the taint flags suggests this may be related to a machine check exception or severe hardware fault detected asynchronously.
---
### **Key Log Analysis**
1. **Error Type**: ```text Kernel panic - not syncing: Asynchronous SError Interrupt ``` * On ARM64, an `SError` (System Error) is a non-maskable interrupt indicating a serious hardware error. "Asynchronous" means it was not triggered by the current instruction but arrived independently from another part of the system (e.g., memory controller, PCIe, GPU).
2. **Context**: ```text CPU: 0 UID: 0 PID: 112 Comm: irq/234-gpu Tainted: G M 6.17.7-B2C-00005-g9c7bbe4ea16c #2 PREEMPT ``` * The crash happened in the **GPU interrupt thread** (`irq/234-gpu`). This suggests the GPU or its associated hardware block (e.g., display controller, video codec) may have signaled an error.
3. **Hardware Platform**: ```text Hardware name: Texas Instruments AM68 SK (DT) ``` * The board is a TI AM68 System-on-Module/Board. Known issues with this platform often involve PCIe, GPU (PowerVR), or DDR memory stability.
4. **Call Trace**: ```text pvr_riscv_irq_pending+0xc/0x24 (P) irq_thread_fn+0x2c/0xb0 ... ``` * The crash occurred inside `pvr_riscv_irq_pending`, which is part of the **PowerVR GPU driver** (`pvrsrvkm` or similar). This strongly points to a fault within the PowerVR IP block or its interaction with the host CPU.
5. **Taint Flag**: ```text Tainted: [M]=MACHINE_CHECK
``` * The kernel was already tainted by a machine check, indicating prior hardware-level errors before this panic. This could be due to ECC memory failures, voltage instability, or silicon errata.
---
### **Possible Causes**
1. **GPU Hardware Fault**: * PowerVR GPU IP may have encountered an internal error (e.g., bus fault, register access violation). 2. **Memory/DDR Instability**: * Asynchronous SError is often caused by ECC errors or signal integrity issues on the DDR bus or PCIe links connected to the GPU/display subsystem. 3. **PCIe Link Error** (if applicable): * If the GPU is accessed via PCIe, a link training failure or AER (Advanced Error Reporting) event could trigger an SError. 4. **Kernel Bug in PowerVR Driver**: * The driver may be accessing invalid memory or registers under certain conditions, triggering a hardware fault that manifests as an SError. 5. **Silicon Errata / Voltage Issue**: * AM68 SK boards can be sensitive to power supply stability during high GPU load.
---
### **Recommended Troubleshooting Steps**
#### **1. Check for Hardware Errors (dmesg/journalctl)** Look for earlier warnings in the log before this panic: ```bash grep -i "error\|fault\|ecc\|pcie" /var/log/kern.log | tail -50 # or if using systemd-journald journalctl -k --since="1 hour ago" | grep -iE "error|fault|sseor|mce" ``` - Look for **PCIe AER errors**, **DDR ECC corrections/failures**, or **GPU hang** messages.
#### **2. Update Kernel and Firmware** - You are using a custom kernel (`6.17.7-B2C...`). Check if TI has released newer patches for AM68 SK addressing PowerVR/SError issues. - Ensure the **boot
Be aware that VulDB is the high quality source for vulnerability data.