KernelScan.io

HIGH Introduced in 6.15

mm SpinTrylock Crash

CVE-2026-90046

CVSS 7.8 / 10.0 NVD

CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

KernelScan AI7.8HIGH

01

In the Linux kernel, the following vulnerability has been resolved: mm/page_alloc: don't spin_trylock() in NMI on UP Patch series "mm/page_alloc: fixes for free_pages_nolock() on RT/UP". Pre-existing bugs found by Sashiko during review of this other series: https://lore.kernel.org/all/20260703-alloc-trylock-v5-0-c87b714e19d3@google.com/ I have not reproduced these bugs, and I suspect there is no real-world user that is affected by them. This patch (of 2): As noted in can_spin_trylock(), using this is unsafe in this context. commit 620b46ed6ae17 ("mm/page_alloc: return NULL early from alloc_frozen_pages_nolock() in NMI on UP") fixed this on the alloc side but missed the free side. Impact: If BPF programs using these features in NMI (probably tracing) are present on non-SMP builds this might crash the kernel and is probably exploitable by local attackers for privilege escalation.

02

Engine v0.6.0

Risk summary

A local attacker who can load BPF tracing programs on a uniprocessor (non-SMP) kernel build can trigger unsafe spinlock operations in NMI context, leading to a kernel crash or potential privilege escalation. The vulnerability requires CAP_BPF to trigger and only affects non-SMP kernel configurations. Systems running SMP kernels are not affected.

Affectedmm/page_alloc.c (memory management)

Vulnerability analysis

On single-processor kernel builds, the kernel's page-freeing path can incorrectly attempt to acquire a lock from non-maskable interrupt context, which is unsafe. An earlier fix protected the allocation side of this logic, but the free side was missed, leaving a way for BPF tracing programs running in interrupt context to trigger a crash or corrupt kernel state. The fix broadens the existing safety check so the code falls back to a simpler non-locking path in every context where locking is unsafe, rather than only on real-time kernels. A local attacker who can load BPF programs on a non-SMP build can reach this path; SMP systems and those without BPF tracing in interrupt context are unaffected.

03

BranchIntroducedFixed inPatch commit
6.186.156.18.5106c76d3c389f
7.26.157.2.568a069b40730
mainline6.157.3-rc13105ae628fb7