Tensorflow is an Open Source Machine Learning Framework. The TFG dialect of TensorFlow (MLIR) makes several assumptions about the incoming GraphDef before converting it to the MLIR-based dialect. If an attacker changes the SavedModel format on disk to invalidate these assumptions and the GraphDef is then converted to MLIR-based IR then they can cause a crash in the Python interpreter. Under certain scenarios, heap OOB read/writes are possible. These issues have been discovered via fuzzing and it is possible that more weaknesses exist. We will patch them as they are discovered.
{
"cna_assigner": "GitHub_M",
"cwe_ids": [
"CWE-125"
],
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2022/23xxx/CVE-2022-23594.json"
}{
"cpe": "cpe:2.3:a:google:tensorflow:2.7.0:*:*:*:*:*:*:*",
"extracted_events": [
{
"introduced": "2.7.0"
},
{
"fixed": "2.8.0"
},
{
"last_affected": "2.7.0"
}
],
"source": [
"AFFECTED_FIELD",
"CPE_STRING"
]
}