{"id":"GHSA-4fg4-p75j-w5xj","summary":"Heap out of bounds in `QuantizedBatchNormWithGlobalNormalization`","details":"### Impact\nAn attacker can cause a segfault and denial of service via accessing data outside of bounds in `tf.raw_ops.QuantizedBatchNormWithGlobalNormalization`:\n\n```python\nimport tensorflow as tf\n\nt = tf.constant([1], shape=[1, 1, 1, 1], dtype=tf.quint8)\nt_min = tf.constant([], shape=[0], dtype=tf.float32)\nt_max = tf.constant([], shape=[0], dtype=tf.float32)\nm = tf.constant([1], shape=[1], dtype=tf.quint8)\nm_min = tf.constant([], shape=[0], dtype=tf.float32)\nm_max = tf.constant([], shape=[0], dtype=tf.float32)\nv = tf.constant([1], shape=[1], dtype=tf.quint8)\nv_min = tf.constant([], shape=[0], dtype=tf.float32)\nv_max = tf.constant([], shape=[0], dtype=tf.float32)\nbeta = tf.constant([1], shape=[1], dtype=tf.quint8)\nbeta_min = tf.constant([], shape=[0], dtype=tf.float32)\nbeta_max = tf.constant([], shape=[0], dtype=tf.float32)\ngamma = tf.constant([1], shape=[1], dtype=tf.quint8)\ngamma_min = tf.constant([], shape=[0], dtype=tf.float32)\ngamma_max = tf.constant([], shape=[0], dtype=tf.float32) \n\ntf.raw_ops.QuantizedBatchNormWithGlobalNormalization(\n  t=t, t_min=t_min, t_max=t_max, m=m, m_min=m_min, m_max=m_max,\n  v=v, v_min=v_min, v_max=v_max, beta=beta, beta_min=beta_min,\n  beta_max=beta_max, gamma=gamma, gamma_min=gamma_min,\n  gamma_max=gamma_max, out_type=tf.qint32,\n  variance_epsilon=0.1, scale_after_normalization=True)\n```                         \n                            \nThis is because the [implementation](https://github.com/tensorflow/tensorflow/blob/55a97caa9e99c7f37a0bbbeb414dc55553d3ae7f/tensorflow/core/kernels/quantized_batch_norm_op.cc#L176-L189) assumes the inputs are not empty: \n  \n```cc\nconst float input_min = context-\u003einput(1).flat\u003cfloat\u003e()(0);\nconst float input_max = context-\u003einput(2).flat\u003cfloat\u003e()(0);\n...\nconst float mean_min = context-\u003einput(4).flat\u003cfloat\u003e()(0);\nconst float mean_max = context-\u003einput(5).flat\u003cfloat\u003e()(0);\n...\nconst float var_min = context-\u003einput(7).flat\u003cfloat\u003e()(0);\nconst float var_max = context-\u003einput(8).flat\u003cfloat\u003e()(0);\n...\nconst float beta_min = context-\u003einput(10).flat\u003cfloat\u003e()(0);\nconst float beta_max = context-\u003einput(11).flat\u003cfloat\u003e()(0);\n...\nconst float gamma_min = context-\u003einput(13).flat\u003cfloat\u003e()(0);\nconst float gamma_max = context-\u003einput(14).flat\u003cfloat\u003e()(0);\n```\n\nIf any of these inputs is empty, `.flat\u003cT\u003e()` is an empty buffer, so accessing the element at index 0 is accessing data outside of bounds.\n\n### Patches\nWe have patched the issue in GitHub commit [d6ed5bcfe1dcab9e85a4d39931bd18d99018e75b](https://github.com/tensorflow/tensorflow/commit/d6ed5bcfe1dcab9e85a4d39931bd18d99018e75b).\n\nThe fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, 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 Yakun Zhang and Ying Wang of Baidu X-Team.","aliases":["BIT-tensorflow-2021-29547","CVE-2021-29547","PYSEC-2021-184","PYSEC-2021-475","PYSEC-2021-673"],"modified":"2026-03-13T22:15:58.733325Z","published":"2021-05-21T14:23:31Z","related":["CVE-2021-29547"],"database_specific":{"cwe_ids":["CWE-125"],"nvd_published_at":"2021-05-14T20:15:00Z","github_reviewed":true,"severity":"LOW","github_reviewed_at":"2021-05-18T21:42:27Z"},"references":[{"type":"WEB","url":"https://github.com/tensorflow/tensorflow/security/advisories/GHSA-4fg4-p75j-w5xj"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2021-29547"},{"type":"WEB","url":"https://github.com/tensorflow/tensorflow/commit/d6ed5bcfe1dcab9e85a4d39931bd18d99018e75b"},{"type":"WEB","url":"https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-cpu/PYSEC-2021-475.yaml"},{"type":"WEB","url":"https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-gpu/PYSEC-2021-673.yaml"},{"type":"WEB","url":"https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow/PYSEC-2021-184.yaml"},{"type":"PACKAGE","url":"https://github.com/tensorflow/tensorflow"}],"affected":[{"package":{"name":"tensorflow","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.1.4"}]}],"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"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.2.0"},{"fixed":"2.2.3"}]}],"versions":["2.2.0","2.2.1","2.2.2"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.3.0"},{"fixed":"2.3.3"}]}],"versions":["2.3.0","2.3.1","2.3.2"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.4.0"},{"fixed":"2.4.2"}]}],"versions":["2.4.0","2.4.1"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-cpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-cpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.1.4"}]}],"versions":["1.15.0","2.1.0","2.1.1","2.1.2","2.1.3"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-cpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-cpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.2.0"},{"fixed":"2.2.3"}]}],"versions":["2.2.0","2.2.1","2.2.2"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-cpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-cpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.3.0"},{"fixed":"2.3.3"}]}],"versions":["2.3.0","2.3.1","2.3.2"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-cpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-cpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.4.0"},{"fixed":"2.4.2"}]}],"versions":["2.4.0","2.4.1"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-gpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-gpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.1.4"}]}],"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"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-gpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-gpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.2.0"},{"fixed":"2.2.3"}]}],"versions":["2.2.0","2.2.1","2.2.2"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-gpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-gpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.3.0"},{"fixed":"2.3.3"}]}],"versions":["2.3.0","2.3.1","2.3.2"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}},{"package":{"name":"tensorflow-gpu","ecosystem":"PyPI","purl":"pkg:pypi/tensorflow-gpu"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.4.0"},{"fixed":"2.4.2"}]}],"versions":["2.4.0","2.4.1"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2021/05/GHSA-4fg4-p75j-w5xj/GHSA-4fg4-p75j-w5xj.json"}}],"schema_version":"1.7.5","severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:L"},{"type":"CVSS_V4","score":"CVSS:4.0/AV:L/AC:L/AT:P/PR:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N"}]}