ONNX: Denial of service via unchecked expansion of packed sub-byte tensors in ONNX Python APIs

影響: pip:onnx <= 1.21.0 → 1.22.0

ONNX's Python tensor-materialization helpers could allocate memory according to attacker-controlled tensor dimensions without verifying that a packed sub-byte payload contained the declared number of elements. A very small model could therefore cause a disproportionately large allocation when an application called onnx.numpy_helper.to_array() or constructed onnx.reference.ReferenceEvaluator with the model. The issue affected UINT2, INT2, UINT4, INT4, and FLOAT4E2M1 tensors stored in either raw_…

在 GitHub Advisory Database 阅读

2026/10/29 06:16，GitHub Advisory Database 发布了“ONNX: Denial of service via unchecked expansion of packed sub-byte tensors in ONNX Python APIs”（类型：漏洞公告）。严重程度为“中危”（medium）。相关技术：Python。CVE：CVE-2026-115922。