CVE-2024-26943

What it actually does

Null pointer dereference in nouveau_dmem_evict_chunk when kcalloc fails under memory pressure, leading to GPU page eviction failure and potential kernel instability.

Why it matters to the business

Systems using the nouveau driver on NVIDIA GPUs may experience crashes or degraded performance during high memory usage, affecting server uptime and requiring kernel updates across fleets.

How an attack unfolds

An attacker triggers sustained memory pressure on a system with the nouveau driver loaded, forcing the eviction path in nouveau_dmem_evict_chunk to execute with a failed allocation, resulting in a null dereference crash.

How to fix it

Update the Linux kernel to a version containing commit 9acfd8b083a0ffbd387566800d89f55058a68af2 or later stable releases that incorporate the kvcalloc change with __GFP_NOFAIL handling.

How to confirm the fix worked

Verify kernel source or binary contains the kvcalloc replacement and __GFP_NOFAIL flag in nouveau_dmem_evict_chunk; test under memory pressure to confirm no null dereference occurs during eviction.

Typical attack path

1. Achieve local code execution or trigger memory exhaustion conditions on target system
2. Load or exercise nouveau driver paths that map GPU memory chunks
3. Force nouveau_dmem_evict_chunk to run while physical memory is exhausted
4. Trigger dereference of null pointers from failed kcalloc for src_pfns, dst_pfns or dma_addrs

How to detect it

Review kernel logs for nouveau-related null pointer dereference oops messages or GPU eviction failures under memory pressure; correlate with system memory exhaustion events in dmesg or syslog.

What your security team should do

• Prioritize kernel upgrades on all systems using the nouveau driver within the next maintenance window
• Monitor for memory pressure events on GPU-enabled hosts and enable kernel panic logging for early detection
• Review downstream kernel distributions for backported fixes and apply them before the CISA-style deadlines if applicable