CVE-2026-42627: Integer Overflow
In Arm ArmNN through 2026-03-27, an integer overflow in TensorShape::GetNumElements() in armnn/Tensor.cpp allows a crafted TFLite model file to bypass buffer size validation and trigger a heap-based buffer over-read during model optimization. The overflow occurs when multiplying tensor dimensions using 32-bit unsigned arithmetic without overflow detection, causing GetNumBytes() to return an understated allocation size. During Optimize()->InferOutputShapes(), the BatchToSpaceNdLayer reads beyond the allocated buffer.
Affected Software
Event History
Frequently Asked Questions
What is the severity of CVE-2026-42627?
The severity of CVE-2026-42627 is medium with a CVSS score of 6.2.
How do I fix CVE-2026-42627?
To fix CVE-2026-42627, ensure that you update to the latest version of Arm ArmNN where this vulnerability has been addressed.
What type of vulnerability is CVE-2026-42627?
CVE-2026-42627 is classified as an integer overflow vulnerability.
What impact does CVE-2026-42627 have on Arm ArmNN?
CVE-2026-42627 can lead to a heap-based buffer over-read during model optimization, potentially allowing an attacker to exploit the overflow.
When was CVE-2026-42627 published?
CVE-2026-42627 was published on May 22, 2026.