{"id":"CVE-2022-29208","aliases":["BIT-tensorflow-2022-29208","EUVD-2022-2039","GHSA-2r2f-g8mw-9gvr","PYSEC-2026-3088","PYSEC-2026-3266","PYSEC-2026-943"],"status":"published","title":"Segfault and Out-of-bounds Write write due to incomplete validation in TensorFlow","description":"TensorFlow is an open source platform for machine learning. Prior to versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, the implementation of `tf.raw_ops.EditDistance` has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service. In multiple places throughout the code, one may compute an index for a write operation. However, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`. Versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4 contain a patch for this issue.","severity":{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-10","score":0.00387,"percentile":0.33162,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-787"],"published":"2022-05-20T22:30:13Z","modified":"2026-10-28T10:28:44Z","fixes":[{"ecosystem":"PyPI","package":"tensorflow","fixed_in":["2.6.4","2.7.2","2.8.1"],"sources":["ghsa","osv"]},{"ecosystem":"PyPI","package":"tensorflow-cpu","fixed_in":["2.6.4","2.7.2","2.8.1"],"sources":["ghsa","osv"]},{"ecosystem":"PyPI","package":"tensorflow-gpu","fixed_in":["2.6.4","2.7.2","2.8.1"],"sources":["ghsa","osv"]},{"ecosystem":"Bitnami","package":"tensorflow","fixed_in":["2.6.4","2.7.2","2.8.1"],"sources":["osv"]}],"affected":[{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.7.0"},{"fixed":"2.7.2"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow-cpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow-cpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.7.0"},{"fixed":"2.7.2"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow-cpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow-gpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow-gpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.7.0"},{"fixed":"2.7.2"}]}]},{"source":"ghsa","ecosystem":"PyPI","package":"tensorflow-gpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}]},{"source":"osv","ecosystem":"Bitnami","package":"tensorflow","ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"2.6.4"},{"introduced":"2.7.0"},{"fixed":"2.7.2"},{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"}]}],"versions":["0.12.0","0.12.1","1.0.0","1.0.1","1.1.0","1.10.0","1.10.1","1.11.0","1.12.0","1.12.2","1.12.3","1.13.1","1.13.2","1.14.0","1.15.0","1.15.2","1.15.3","1.15.4","1.15.5","1.2.0","1.2.1","1.3.0","1.4.0","1.4.1","1.5.0","1.5.1","1.6.0","1.7.0","1.7.1","1.8.0","1.9.0","2.0.0","2.0.1","2.0.2","2.0.3","2.0.4","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.2.0","2.2.1","2.2.2","2.2.3","2.3.0","2.3.1","2.3.2","2.3.3","2.3.4","2.4.0","2.4.1","2.4.2","2.4.3","2.4.4","2.5.0","2.5.1","2.5.2","2.5.3","2.6.0","2.6.0rc0","2.6.0rc1","2.6.0rc2","2.6.1","2.6.2","2.6.3"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.7.0"},{"fixed":"2.7.2"}]}],"versions":["2.7.0","2.7.1"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}],"versions":["2.8.0"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-cpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"}]}],"versions":["1.15.0","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.2.0","2.2.1","2.2.2","2.2.3","2.3.0","2.3.1","2.3.2","2.3.3","2.3.4","2.4.0","2.4.1","2.4.2","2.4.3","2.4.4","2.5.0","2.5.1","2.5.2","2.5.3","2.6.0","2.6.1","2.6.2","2.6.3"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-cpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.7.0"},{"fixed":"2.7.2"}]}],"versions":["2.7.0","2.7.1"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-cpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}],"versions":["2.8.0"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-gpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"}]}],"versions":["0.12.0","0.12.1","1.0.0","1.0.1","1.1.0","1.10.0","1.10.1","1.11.0","1.12.0","1.12.2","1.12.3","1.13.1","1.13.2","1.14.0","1.15.0","1.15.2","1.15.3","1.15.4","1.15.5","1.2.0","1.2.1","1.3.0","1.4.0","1.4.1","1.5.0","1.5.1","1.6.0","1.7.0","1.7.1","1.8.0","1.9.0","2.0.0","2.0.1","2.0.2","2.0.3","2.0.4","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.2.0","2.2.1","2.2.2","2.2.3","2.3.0","2.3.1","2.3.2","2.3.3","2.3.4","2.4.0","2.4.1","2.4.2","2.4.3","2.4.4","2.5.0","2.5.1","2.5.2","2.5.3","2.6.0","2.6.1","2.6.2","2.6.3"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-gpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.7.0"},{"fixed":"2.7.2"}]}],"versions":["2.7.0","2.7.1"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-gpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}],"versions":["2.8.0"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"},{"introduced":"2.7.0"},{"fixed":"2.7.2"},{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}],"versions":["0.12.0","0.12.1","1.0.0","1.0.1","1.1.0","1.10.0","1.10.1","1.11.0","1.12.0","1.12.2","1.12.3","1.13.1","1.13.2","1.14.0","1.15.0","1.15.2","1.15.3","1.15.4","1.15.5","1.2.0","1.2.1","1.3.0","1.4.0","1.4.1","1.5.0","1.5.1","1.6.0","1.7.0","1.7.1","1.8.0","1.9.0","2.0.0","2.0.1","2.0.2","2.0.3","2.0.4","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.2.0","2.2.1","2.2.2","2.2.3","2.3.0","2.3.1","2.3.2","2.3.3","2.3.4","2.4.0","2.4.1","2.4.2","2.4.3","2.4.4","2.5.0","2.5.1","2.5.2","2.5.3","2.6.0","2.6.0rc0","2.6.0rc1","2.6.0rc2","2.6.1","2.6.2","2.6.3","2.7.0","2.7.1","2.8.0"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-cpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"},{"introduced":"2.7.0"},{"fixed":"2.7.2"},{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}],"versions":["1.15.0","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.2.0","2.2.1","2.2.2","2.2.3","2.3.0","2.3.1","2.3.2","2.3.3","2.3.4","2.4.0","2.4.1","2.4.2","2.4.3","2.4.4","2.5.0","2.5.1","2.5.2","2.5.3","2.6.0","2.6.1","2.6.2","2.6.3","2.7.0","2.7.1","2.8.0"]},{"source":"osv","ecosystem":"PyPI","package":"tensorflow-gpu","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.6.4"},{"introduced":"2.7.0"},{"fixed":"2.7.2"},{"introduced":"2.8.0"},{"fixed":"2.8.1"}]}],"versions":["0.12.0","0.12.1","1.0.0","1.0.1","1.1.0","1.10.0","1.10.1","1.11.0","1.12.0","1.12.2","1.12.3","1.13.1","1.13.2","1.14.0","1.15.0","1.15.2","1.15.3","1.15.4","1.15.5","1.2.0","1.2.1","1.3.0","1.4.0","1.4.1","1.5.0","1.5.1","1.6.0","1.7.0","1.7.1","1.8.0","1.9.0","2.0.0","2.0.1","2.0.2","2.0.3","2.0.4","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.2.0","2.2.1","2.2.2","2.2.3","2.3.0","2.3.1","2.3.2","2.3.3","2.3.4","2.4.0","2.4.1","2.4.2","2.4.3","2.4.4","2.5.0","2.5.1","2.5.2","2.5.3","2.6.0","2.6.1","2.6.2","2.6.3","2.7.0","2.7.1","2.8.0"]}],"products":[{"vendor":"google","product":"tensorflow"},{"vendor":"tensorflow","product":"tensorflow"}],"references":[{"url":"https://github.com/tensorflow/tensorflow/releases/tag/v2.6.4","tags":["x_refsource_MISC","x_transferred","Release Notes","Third Party Advisory","advisory","web"]},{"url":"https://github.com/tensorflow/tensorflow/releases/tag/v2.7.2","tags":["x_refsource_MISC","x_transferred","Release Notes","Third Party Advisory","advisory","web"]},{"url":"https://github.com/tensorflow/tensorflow/releases/tag/v2.8.1","tags":["x_refsource_MISC","x_transferred","Release Notes","Third Party Advisory","advisory","web"]},{"url":"https://github.com/tensorflow/tensorflow/releases/tag/v2.9.0","tags":["x_refsource_MISC","x_transferred","Release Notes","Third Party Advisory","advisory","web"]},{"url":"https://github.com/tensorflow/tensorflow/security/advisories/GHSA-2r2f-g8mw-9gvr","tags":["x_refsource_CONFIRM","x_transferred","Exploit","Patch","Third Party Advisory","advisory","web"]},{"url":"https://github.com/tensorflow/tensorflow/commit/30721cf564cb029d34535446d6a5a6357bebc8e7","tags":["x_refsource_MISC","x_transferred","Patch","Third Party Advisory","advisory","fix","web"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2022-29208","tags":["web","advisory"]},{"url":"https://github.com/advisories/GHSA-2r2f-g8mw-9gvr","tags":["advisory"]},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2022/29xxx/CVE-2022-29208.json","tags":["advisory"]},{"url":"https://github.com/tensorflow/tensorflow","tags":["package"]},{"url":"https://pypi.org/project/tensorflow","tags":["package"]},{"url":"https://pypi.org/project/tensorflow-cpu","tags":["package"]},{"url":"https://pypi.org/project/tensorflow-gpu","tags":["package"]}],"sources":{"cve":[{"id":"CVE-2022-29208","title":"Segfault and Out-of-bounds Write write due to incomplete validation in TensorFlow","description":"TensorFlow is an open source platform for machine learning. Prior to versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, the implementation of `tf.raw_ops.EditDistance` has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service. In multiple places throughout the code, one may compute an index for a write operation. However, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`. Versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4 contain a patch for this issue.","status":"published","published":"2022-05-20T22:30:13Z","modified":"2025-04-22T17:58:20.112Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high","assessor":"cna"}],"cwes":["CWE-787"],"updated_at":"2026-11-02T14:34:36.578327Z"}],"euvd":[{"id":"EUVD-2022-2039","description":"TensorFlow is an open source platform for machine learning. Prior to versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, the implementation of `tf.raw_ops.EditDistance` has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service. In multiple places throughout the code, one may compute an index for a write operation. However, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`. Versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4 contain a patch for this issue.","status":"published","published":"2022-05-20T22:30:13Z","modified":"2025-04-22T17:58:20Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high"}],"updated_at":"2026-11-02T20:52:15.540112Z"}],"ghsa":[{"id":"GHSA-2r2f-g8mw-9gvr","title":"Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow","description":"### Impact\nThe implementation of [`tf.raw_ops.EditDistance`]() has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service:\n\n```python\nimport tensorflow as tf\n\nhypothesis_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64) \nhypothesis_values = tf.constant(0, shape=[3], dtype=tf.int64)\nhypothesis_shape = tf.constant(0, shape=[3], dtype=tf.int64)\n\ntruth_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64)\ntruth_values = tf.constant(2, shape=[3], dtype=tf.int64)\ntruth_shape = tf.constant(2, shape=[3], dtype=tf.int64) \n\ntf.raw_ops.EditDistance(\n  hypothesis_indices=hypothesis_indices,\n  hypothesis_values=hypothesis_values,\n  hypothesis_shape=hypothesis_shape,\n  truth_indices=truth_indices,\n  truth_values=truth_values,\n  truth_shape=truth_shape)\n```\n\nIn multiple places throughout the code, we are computing an index for a write operation:\n\n```cc\nif (g_truth == g_hypothesis) {\n  auto loc = std::inner_product(g_truth.begin(), g_truth.end(),\n                                output_strides.begin(), int64_t{0});\n  OP_REQUIRES(\n      ctx, loc \u003c output_elements,\n      errors::Internal(\"Got an inner product \", loc,\n                       \" which would require in writing to outside of \"\n                       \"the buffer for the output tensor (max elements \",\n                       output_elements, \")\"));\n  output_t(loc) =\n      gtl::LevenshteinDistance\u003cT\u003e(truth_seq, hypothesis_seq, cmp);\n  // ...\n}\n```\n\nHowever, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`.\n\n### Patches\nWe have patched the issue in GitHub commit [30721cf564cb029d34535446d6a5a6357bebc8e7](https://github.com/tensorflow/tensorflow/commit/30721cf564cb029d34535446d6a5a6357bebc8e7).\n\nThe fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4, as these are also affected and still in supported range.\n\n### For more information\nPlease consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.\n\n### Attribution\nThis vulnerability has been reported by Neophytos Christou from Secure Systems Lab at Brown University.","status":"published","published":"2022-05-24T22:14:22Z","modified":"2023-01-30T05:05:50Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high"}],"cwes":["CWE-787"],"updated_at":"2026-11-02T14:36:28.424013Z"}],"nvd":[{"id":"CVE-2022-29208","description":"TensorFlow is an open source platform for machine learning. Prior to versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, the implementation of `tf.raw_ops.EditDistance` has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service. In multiple places throughout the code, one may compute an index for a write operation. However, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`. Versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4 contain a patch for this issue.","status":"published","published":"2022-05-20T23:15:45.15Z","modified":"2026-06-17T04:39:49.59Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high","assessor":"secondary"},{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high","assessor":"primary"},{"version":"2.0","score":3.6,"vector":"AV:L/AC:L/Au:N/C:N/I:P/A:P","level":"low","assessor":"primary"}],"cwes":["CWE-787"],"updated_at":"2026-11-02T14:07:21.261605Z"}],"osv":[{"id":"CVE-2022-29208","title":"Segfault and Out-of-bounds Write write due to incomplete validation in TensorFlow","description":"TensorFlow is an open source platform for machine learning. Prior to versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, the implementation of `tf.raw_ops.EditDistance` has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service. In multiple places throughout the code, one may compute an index for a write operation. However, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`. Versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4 contain a patch for this issue.","status":"published","published":"2022-05-20T22:30:13Z","modified":"2026-10-28T10:28:44.310104331Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high"}],"cwes":["CWE-787"],"updated_at":"2026-11-02T15:44:25.164673Z"},{"id":"GHSA-2r2f-g8mw-9gvr","title":"Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow","description":"### Impact\nThe implementation of [`tf.raw_ops.EditDistance`]() has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service:\n\n```python\nimport tensorflow as tf\n\nhypothesis_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64) \nhypothesis_values = tf.constant(0, shape=[3], dtype=tf.int64)\nhypothesis_shape = tf.constant(0, shape=[3], dtype=tf.int64)\n\ntruth_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64)\ntruth_values = tf.constant(2, shape=[3], dtype=tf.int64)\ntruth_shape = tf.constant(2, shape=[3], dtype=tf.int64) \n\ntf.raw_ops.EditDistance(\n  hypothesis_indices=hypothesis_indices,\n  hypothesis_values=hypothesis_values,\n  hypothesis_shape=hypothesis_shape,\n  truth_indices=truth_indices,\n  truth_values=truth_values,\n  truth_shape=truth_shape)\n```\n\nIn multiple places throughout the code, we are computing an index for a write operation:\n\n```cc\nif (g_truth == g_hypothesis) {\n  auto loc = std::inner_product(g_truth.begin(), g_truth.end(),\n                                output_strides.begin(), int64_t{0});\n  OP_REQUIRES(\n      ctx, loc \u003c output_elements,\n      errors::Internal(\"Got an inner product \", loc,\n                       \" which would require in writing to outside of \"\n                       \"the buffer for the output tensor (max elements \",\n                       output_elements, \")\"));\n  output_t(loc) =\n      gtl::LevenshteinDistance\u003cT\u003e(truth_seq, hypothesis_seq, cmp);\n  // ...\n}\n```\n\nHowever, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`.\n\n### Patches\nWe have patched the issue in GitHub commit [30721cf564cb029d34535446d6a5a6357bebc8e7](https://github.com/tensorflow/tensorflow/commit/30721cf564cb029d34535446d6a5a6357bebc8e7).\n\nThe fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4, as these are also affected and still in supported range.\n\n### For more information\nPlease consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.\n\n### Attribution\nThis vulnerability has been reported by Neophytos Christou from Secure Systems Lab at Brown University.","status":"published","published":"2022-05-24T22:14:22Z","modified":"2026-07-13T16:42:40.712669136Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high"}],"cwes":["CWE-787"],"updated_at":"2026-11-02T15:46:34.700741Z"},{"id":"PYSEC-2026-3088","title":"Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow","description":"### Impact\nThe implementation of [`tf.raw_ops.EditDistance`]() has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service:\n\n```python\nimport tensorflow as tf\n\nhypothesis_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64) \nhypothesis_values = tf.constant(0, shape=[3], dtype=tf.int64)\nhypothesis_shape = tf.constant(0, shape=[3], dtype=tf.int64)\n\ntruth_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64)\ntruth_values = tf.constant(2, shape=[3], dtype=tf.int64)\ntruth_shape = tf.constant(2, shape=[3], dtype=tf.int64) \n\ntf.raw_ops.EditDistance(\n  hypothesis_indices=hypothesis_indices,\n  hypothesis_values=hypothesis_values,\n  hypothesis_shape=hypothesis_shape,\n  truth_indices=truth_indices,\n  truth_values=truth_values,\n  truth_shape=truth_shape)\n```\n\nIn multiple places throughout the code, we are computing an index for a write operation:\n\n```cc\nif (g_truth == g_hypothesis) {\n  auto loc = std::inner_product(g_truth.begin(), g_truth.end(),\n                                output_strides.begin(), int64_t{0});\n  OP_REQUIRES(\n      ctx, loc \u003c output_elements,\n      errors::Internal(\"Got an inner product \", loc,\n                       \" which would require in writing to outside of \"\n                       \"the buffer for the output tensor (max elements \",\n                       output_elements, \")\"));\n  output_t(loc) =\n      gtl::LevenshteinDistance\u003cT\u003e(truth_seq, hypothesis_seq, cmp);\n  // ...\n}\n```\n\nHowever, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`.\n\n### Patches\nWe have patched the issue in GitHub commit [30721cf564cb029d34535446d6a5a6357bebc8e7](https://github.com/tensorflow/tensorflow/commit/30721cf564cb029d34535446d6a5a6357bebc8e7).\n\nThe fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4, as these are also affected and still in supported range.\n\n### For more information\nPlease consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.\n\n### Attribution\nThis vulnerability has been reported by Neophytos Christou from Secure Systems Lab at Brown University.","status":"published","published":"2026-07-09T16:49:50.315364Z","modified":"2026-07-13T16:42:40.712669136Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high"}],"updated_at":"2026-11-02T15:52:56.429374Z"},{"id":"PYSEC-2026-3266","title":"Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow","description":"### Impact\nThe implementation of [`tf.raw_ops.EditDistance`]() has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service:\n\n```python\nimport tensorflow as tf\n\nhypothesis_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64) \nhypothesis_values = tf.constant(0, shape=[3], dtype=tf.int64)\nhypothesis_shape = tf.constant(0, shape=[3], dtype=tf.int64)\n\ntruth_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64)\ntruth_values = tf.constant(2, shape=[3], dtype=tf.int64)\ntruth_shape = tf.constant(2, shape=[3], dtype=tf.int64) \n\ntf.raw_ops.EditDistance(\n  hypothesis_indices=hypothesis_indices,\n  hypothesis_values=hypothesis_values,\n  hypothesis_shape=hypothesis_shape,\n  truth_indices=truth_indices,\n  truth_values=truth_values,\n  truth_shape=truth_shape)\n```\n\nIn multiple places throughout the code, we are computing an index for a write operation:\n\n```cc\nif (g_truth == g_hypothesis) {\n  auto loc = std::inner_product(g_truth.begin(), g_truth.end(),\n                                output_strides.begin(), int64_t{0});\n  OP_REQUIRES(\n      ctx, loc \u003c output_elements,\n      errors::Internal(\"Got an inner product \", loc,\n                       \" which would require in writing to outside of \"\n                       \"the buffer for the output tensor (max elements \",\n                       output_elements, \")\"));\n  output_t(loc) =\n      gtl::LevenshteinDistance\u003cT\u003e(truth_seq, hypothesis_seq, cmp);\n  // ...\n}\n```\n\nHowever, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`.\n\n### Patches\nWe have patched the issue in GitHub commit [30721cf564cb029d34535446d6a5a6357bebc8e7](https://github.com/tensorflow/tensorflow/commit/30721cf564cb029d34535446d6a5a6357bebc8e7).\n\nThe fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4, as these are also affected and still in supported range.\n\n### For more information\nPlease consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.\n\n### Attribution\nThis vulnerability has been reported by Neophytos Christou from Secure Systems Lab at Brown University.","status":"published","published":"2026-07-09T16:49:50.405213Z","modified":"2026-07-13T16:42:40.712669136Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high"}],"updated_at":"2026-11-02T15:52:56.847083Z"},{"id":"PYSEC-2026-943","title":"Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow","description":"### Impact\nThe implementation of [`tf.raw_ops.EditDistance`]() has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service:\n\n```python\nimport tensorflow as tf\n\nhypothesis_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64) \nhypothesis_values = tf.constant(0, shape=[3], dtype=tf.int64)\nhypothesis_shape = tf.constant(0, shape=[3], dtype=tf.int64)\n\ntruth_indices = tf.constant(-1250999896764, shape=[3, 3], dtype=tf.int64)\ntruth_values = tf.constant(2, shape=[3], dtype=tf.int64)\ntruth_shape = tf.constant(2, shape=[3], dtype=tf.int64) \n\ntf.raw_ops.EditDistance(\n  hypothesis_indices=hypothesis_indices,\n  hypothesis_values=hypothesis_values,\n  hypothesis_shape=hypothesis_shape,\n  truth_indices=truth_indices,\n  truth_values=truth_values,\n  truth_shape=truth_shape)\n```\n\nIn multiple places throughout the code, we are computing an index for a write operation:\n\n```cc\nif (g_truth == g_hypothesis) {\n  auto loc = std::inner_product(g_truth.begin(), g_truth.end(),\n                                output_strides.begin(), int64_t{0});\n  OP_REQUIRES(\n      ctx, loc \u003c output_elements,\n      errors::Internal(\"Got an inner product \", loc,\n                       \" which would require in writing to outside of \"\n                       \"the buffer for the output tensor (max elements \",\n                       output_elements, \")\"));\n  output_t(loc) =\n      gtl::LevenshteinDistance\u003cT\u003e(truth_seq, hypothesis_seq, cmp);\n  // ...\n}\n```\n\nHowever, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`.\n\n### Patches\nWe have patched the issue in GitHub commit [30721cf564cb029d34535446d6a5a6357bebc8e7](https://github.com/tensorflow/tensorflow/commit/30721cf564cb029d34535446d6a5a6357bebc8e7).\n\nThe fix will be included in TensorFlow 2.9.0. We will also cherrypick this commit on TensorFlow 2.8.1, TensorFlow 2.7.2, and TensorFlow 2.6.4, as these are also affected and still in supported range.\n\n### For more information\nPlease consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.\n\n### Attribution\nThis vulnerability has been reported by Neophytos Christou from Secure Systems Lab at Brown University.","status":"published","published":"2026-07-06T08:03:30.019686Z","modified":"2026-07-13T16:42:40.712669136Z","cvss":[{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H","level":"high"}],"updated_at":"2026-11-02T15:52:58.117212Z"},{"id":"BIT-tensorflow-2022-29208","title":"Segfault and Out-of-bounds Write write due to incomplete validation in TensorFlow","description":"TensorFlow is an open source platform for machine learning. Prior to versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, the implementation of `tf.raw_ops.EditDistance` has incomplete validation. Users can pass negative values to cause a segmentation fault based denial of service. In multiple places throughout the code, one may compute an index for a write operation. However, the existing validation only checks against the upper bound of the array. Hence, it is possible to write before the array by massaging the input to generate negative values for `loc`. Versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4 contain a patch for this issue.","status":"published","published":"2024-03-06T11:14:20.113Z","modified":"2026-09-08T08:48:20.261839048Z","updated_at":"2026-11-02T15:28:35.51001Z"}]}}