An integer overflow vulnerability in the loading of ExecuTorch models can cause smaller-than-expected memory regions to be allocated, potentially resulting in code execution or other undesirable effects. This issue affects ExecuTorch prior to commit 8f062d3f661e20bb19b24b767b9a9a46e8359f2b.
"https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2025-54952.json"
[
{
"deprecated": false,
"signature_type": "Line",
"signature_version": "v1",
"digest": {
"line_hashes": [
"32230930426288135781751580657657381743",
"27816795093322237428348240608559980770",
"166200815449907023341764011611882552768",
"31150020504940064754111791546867125044",
"177987325295771229544198077922036655720",
"331750139255521733435766596191056857196",
"121091089390543637236130583055923662272"
],
"threshold": 0.9
},
"source": "https://github.com/pytorch/executorch/commit/8f062d3f661e20bb19b24b767b9a9a46e8359f2b",
"id": "CVE-2025-54952-029ca469",
"target": {
"file": "runtime/core/memory_allocator.h"
}
}
]