CVE-2020-15193
In Tensorflow before versions 2.2.1 and 2.3.1, the implementation of `dlpack.to_dlpack` can be made to use uninitialized memory resulting in further memory corruption. This is because the pybind11 glue code assumes that the argument is a tensor. However, there is nothing stopping users from passing in a Python object instead of a tensor. The uninitialized memory address is due to a `reinterpret_cast` Since the `PyObject` is a Python object, not a TensorFlow Tensor, the cast to `EagerTensor` fails. The issue is patched in commit 22e07fb204386768e5bcbea563641ea11f96ceb8 and is released in TensorFlow versions 2.2.1, or 2.3.1.
CVSS
- Versión: 3.1
- Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L
- Puntuación base: 7.1
Probabilidad de explotación (EPSS)
- Probabilidad de explotación en los próximos 30 días: 0.83%
- Percentil entre todas las CVEs puntuadas: 56
- Fecha de la puntuación: 7/10/2026
EPSS (Exploit Prediction Scoring System, de FIRST) estima la probabilidad de que una vulnerabilidad sea explotada en 30 días. Complementa a CVSS (impacto) y a CISA KEV (explotación confirmada).
Tecnologías afectadas (2)
CWE
- CWE-908
- CWE-908
Referencias
- http://lists.opensuse.org/opensuse-security-announce/2020-10/msg00065.html
- https://github.com/tensorflow/tensorflow/commit/22e07fb204386768e5bcbea563641ea11f96ceb8
- https://github.com/tensorflow/tensorflow/releases/tag/v2.3.1
- https://github.com/tensorflow/tensorflow/security/advisories/GHSA-rjjg-hgv6-h69v
- http://lists.opensuse.org/opensuse-security-announce/2020-10/msg00065.html
- https://github.com/tensorflow/tensorflow/commit/22e07fb204386768e5bcbea563641ea11f96ceb8
- https://github.com/tensorflow/tensorflow/releases/tag/v2.3.1
- https://github.com/tensorflow/tensorflow/security/advisories/GHSA-rjjg-hgv6-h69v
JSON original (NVD)
Mostrar
{
"id": "CVE-2020-15193",
"cveTags": [],
"metrics": {
"cvssMetricV2": [
{
"type": "Primary",
"source": "nvd@nist.gov",
"cvssData": {
"version": "2.0",
"baseScore": 5.5,
"accessVector": "NETWORK",
"vectorString": "AV:N/AC:L/Au:S/C:N/I:P/A:P",
"authentication": "SINGLE",
"integrityImpact": "PARTIAL",
"accessComplexity": "LOW",
"availabilityImpact": "PARTIAL",
"confidentialityImpact": "NONE"
},
"acInsufInfo": false,
"impactScore": 4.9,
"baseSeverity": "MEDIUM",
"obtainAllPrivilege": false,
"exploitabilityScore": 8,
"obtainUserPrivilege": false,
"obtainOtherPrivilege": false,
"userInteractionRequired": false
}
],
"cvssMetricV31": [
{
"type": "Secondary",
"source": "security-advisories@github.com",
"cvssData": {
"scope": "UNCHANGED",
"version": "3.1",
"baseScore": 7.1,
"attackVector": "NETWORK",
"baseSeverity": "HIGH",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L",
"integrityImpact": "HIGH",
"userInteraction": "NONE",
"attackComplexity": "LOW",
"availabilityImpact": "LOW",
"privilegesRequired": "LOW",
"confidentialityImpact": "NONE"
},
"impactScore": 4.2,
"exploitabilityScore": 2.8
},
{
"type": "Primary",
"source": "nvd@nist.gov",
"cvssData": {
"scope": "UNCHANGED",
"version": "3.1",
"baseScore": 7.1,
"attackVector": "NETWORK",
"baseSeverity": "HIGH",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L",
"integrityImpact": "HIGH",
"userInteraction": "NONE",
"attackComplexity": "LOW",
"availabilityImpact": "LOW",
"privilegesRequired": "LOW",
"confidentialityImpact": "NONE"
},
"impactScore": 4.2,
"exploitabilityScore": 2.8
}
]
},
"affected": [
{
"source": "security-advisories@github.com",
"affectedData": [
{
"vendor": "tensorflow",
"product": "tensorflow",
"versions": [
{
"status": "affected",
"version": "= 2.2.0"
},
{
"status": "affected",
"version": "= 2.3.0"
}
]
}
]
}
],
"published": "2020-09-25T19:15:14.573",
"references": [
{
"url": "http://lists.opensuse.org/opensuse-security-announce/2020-10/msg00065.html",
"tags": [
"Mailing List",
"Third Party Advisory"
],
"source": "security-advisories@github.com"
},
{
"url": "https://github.com/tensorflow/tensorflow/commit/22e07fb204386768e5bcbea563641ea11f96ceb8",
"tags": [
"Patch",
"Third Party Advisory"
],
"source": "security-advisories@github.com"
},
{
"url": "https://github.com/tensorflow/tensorflow/releases/tag/v2.3.1",
"tags": [
"Third Party Advisory"
],
"source": "security-advisories@github.com"
},
{
"url": "https://github.com/tensorflow/tensorflow/security/advisories/GHSA-rjjg-hgv6-h69v",
"tags": [
"Exploit",
"Third Party Advisory"
],
"source": "security-advisories@github.com"
},
{
"url": "http://lists.opensuse.org/opensuse-security-announce/2020-10/msg00065.html",
"tags": [
"Mailing List",
"Third Party Advisory"
],
"source": "af854a3a-2127-422b-91ae-364da2661108"
},
{
"url": "https://github.com/tensorflow/tensorflow/commit/22e07fb204386768e5bcbea563641ea11f96ceb8",
"tags": [
"Patch",
"Third Party Advisory"
],
"source": "af854a3a-2127-422b-91ae-364da2661108"
},
{
"url": "https://github.com/tensorflow/tensorflow/releases/tag/v2.3.1",
"tags": [
"Third Party Advisory"
],
"source": "af854a3a-2127-422b-91ae-364da2661108"
},
{
"url": "https://github.com/tensorflow/tensorflow/security/advisories/GHSA-rjjg-hgv6-h69v",
"tags": [
"Exploit",
"Third Party Advisory"
],
"source": "af854a3a-2127-422b-91ae-364da2661108"
}
],
"vulnStatus": "Modified",
"weaknesses": [
{
"type": "Secondary",
"source": "security-advisories@github.com",
"description": [
{
"lang": "en",
"value": "CWE-908"
}
]
},
{
"type": "Primary",
"source": "nvd@nist.gov",
"description": [
{
"lang": "en",
"value": "CWE-908"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "In Tensorflow before versions 2.2.1 and 2.3.1, the implementation of `dlpack.to_dlpack` can be made to use uninitialized memory resulting in further memory corruption. This is because the pybind11 glue code assumes that the argument is a tensor. However, there is nothing stopping users from passing in a Python object instead of a tensor. The uninitialized memory address is due to a `reinterpret_cast` Since the `PyObject` is a Python object, not a TensorFlow Tensor, the cast to `EagerTensor` fails. The issue is patched in commit 22e07fb204386768e5bcbea563641ea11f96ceb8 and is released in TensorFlow versions 2.2.1, or 2.3.1."
},
{
"lang": "es",
"value": "En Tensorflow versiones anteriores a 2.2.1 y 2.3.1, la implementación de \"dlpack.to_dlpack\" puede ser realizada para usar la memoria no inicializada, lo que resulta en una mayor corrupción de la memoria. Esto es debido a que el código adhesivo pybind11 asume que el argumento es un tensor. Sin embargo, no existe nada que impida que los usuarios pasen un objeto Python en lugar de un tensor. La dirección de memoria no inicializada es debido a un \"reinterpret_cast\" Dado que el \"PyObject\" es un objeto de Python, no un Tensor de TensorFlow, la conversión a \"EagerTensor\" presenta un fallo. El problema es parcheado en el commit 22e07fb204386768e5bcbea563641ea11f96ceb8 y es publicado en TensorFlow versiones 2.2.1 o 2.3.1"
}
],
"lastModified": "2026-06-17T02:56:14.193",
"configurations": [
{
"nodes": [
{
"negate": false,
"cpeMatch": [
{
"criteria": "cpe:2.3:a:google:tensorflow:2.2.0:*:*:*:-:*:*:*",
"vulnerable": true,
"matchCriteriaId": "FB9BCD7D-1626-429F-B479-7D2F1E46B9C4"
},
{
"criteria": "cpe:2.3:a:google:tensorflow:2.3.0:*:*:*:-:*:*:*",
"vulnerable": true,
"matchCriteriaId": "D0A7B69E-9388-48F0-B744-49453EBAF5D5"
}
],
"operator": "OR"
}
]
},
{
"nodes": [
{
"negate": false,
"cpeMatch": [
{
"criteria": "cpe:2.3:o:opensuse:leap:15.2:*:*:*:*:*:*:*",
"vulnerable": true,
"matchCriteriaId": "B009C22E-30A4-4288-BCF6-C3E81DEAF45A"
}
],
"operator": "OR"
}
]
}
],
"sourceIdentifier": "security-advisories@github.com"
}