همهٔ گزارش‌ها
متوسط

kernel: drm/amdgpu/mes: Fix hung_queue_db_array loop limit for multi-XCC

متن اصلی

نسخه‌های متأثر

نسخه‌های متأثر
نامشخص
نسخهٔ اصلاح‌شده
اعلام نشده

بهره‌برداری

CVE-2026-93139

احتمال بهره‌برداری در ۳۰ روز آینده: 0.20%

FIRST EPSS · ۱۴۰۵/۰۸/۲۸

شرح آسیب‌پذیری

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/mes: Fix hung_queue_db_array loop limit for multi-XCC The loop iterated only AMDGPU_MAX_MES_PIPES times, leaving entries uninitialized for multi-XCC GPUs. This causes null pointer dereferences when accessing arrays indexed by XCC ID >= 2. Extend the loop to cover all XCCs (AMDGPU_MAX_MES_PIPES * num_xcc), matching other per-XCC arrays.

A flaw was found in the AMDGPU driver within the Linux kernel. An issue with the hung_queue_db_array loop limit for multi-XCC (Cross-Chiplet Interconnect) GPUs can leave array entries uninitialized. This can lead to a null pointer dereference when accessing arrays indexed by XCC ID greater than or equal to 2, potentially causing a system crash or denial of service.