Read the docs
TI-2021-15547 Aliases CVE-2021-37672GHSA-5hj3-vjjf-f5m7

Heap OOB in `SdcaOptimizerV2` in TensorFlow

Priority LOW 24CWE-125Published · Updated

Key facts

Attack vector
Local
Privileges required
Low
User interaction
None
Attack complexity
Low
Impact type
Information disclosure
Confidentiality · integrity · availability
High · None · None
Fix status
Fixed

Attack profile from CVSS 3.1 · CNA. Fix status from CNA and GitHub. Impact type from CWE-125 and the CVSS impact.

Summary

TensorFlow is an end-to-end open source platform for machine learning. In affected versions an attacker can read from outside of bounds of heap allocated data by sending specially crafted illegal arguments to `tf.raw_ops.SdcaOptimizerV2`. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/core/kernels/sdca_internal.cc#L320-L353) does not check that the length of `example_labels` is the same as the number of examples. We have patched the issue in GitHub commit a4e138660270e7599793fa438cd7b2fc2ce215a6. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

Why it matters now

EPSS 0.00172 6th percentileScored 2026-11-07
Fix Fixed tensorflow (tensorflow): 2.5.1, 2.4.3, 2.3.4tensorflow (PyPI): 2.3.4, 2.4.3, 2.5.1+2 more products

SSVC decision points

Not available: CISA has not published SSVC decision points for this record.

Affected products

VendorProductAffected versionsFixed inStatusSource
Googletensorflow2.5.0 to before 2.5.1; 2.4.0 to before 2.4.3; before 2.3.42.5.1, 2.4.3, 2.3.4Patch availableCNA
—tensorflow (PyPI)before 2.3.42.3.4Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow (PyPI)2.4.0 to before 2.4.32.4.3Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow (PyPI)2.5.0 to before 2.5.12.5.1Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow-cpu (PyPI)before 2.3.42.3.4Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow-cpu (PyPI)2.4.0 to before 2.4.32.4.3Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow-cpu (PyPI)2.5.0 to before 2.5.12.5.1Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow-gpu (PyPI)before 2.3.42.3.4Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow-gpu (PyPI)2.4.0 to before 2.4.32.4.3Patch availableGHSA-5hj3-vjjf-f5m7
—tensorflow-gpu (PyPI)2.5.0 to before 2.5.12.5.1Patch availableGHSA-5hj3-vjjf-f5m7
  • tensorflow VendorGoogle Affected 2.5.0 to before 2.5.1; 2.4.0 to before 2.4.3; before 2.3.4 Fixed in 2.5.1, 2.4.3, 2.3.4 Patch available CNA
  • tensorflow (PyPI) Vendor— Affected before 2.3.4 Fixed in 2.3.4 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow (PyPI) Vendor— Affected 2.4.0 to before 2.4.3 Fixed in 2.4.3 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow (PyPI) Vendor— Affected 2.5.0 to before 2.5.1 Fixed in 2.5.1 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow-cpu (PyPI) Vendor— Affected before 2.3.4 Fixed in 2.3.4 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow-cpu (PyPI) Vendor— Affected 2.4.0 to before 2.4.3 Fixed in 2.4.3 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow-cpu (PyPI) Vendor— Affected 2.5.0 to before 2.5.1 Fixed in 2.5.1 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow-gpu (PyPI) Vendor— Affected before 2.3.4 Fixed in 2.3.4 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow-gpu (PyPI) Vendor— Affected 2.4.0 to before 2.4.3 Fixed in 2.4.3 Patch available GHSA-5hj3-vjjf-f5m7
  • tensorflow-gpu (PyPI) Vendor— Affected 2.5.0 to before 2.5.1 Fixed in 2.5.1 Patch available GHSA-5hj3-vjjf-f5m7

CPE configurations NVD · 5

CPEVersionsRoleVendor
cpe:2.3:a:google:tensorflow:*:*:*:*:*:*:*:*2.3.0 to before 2.3.4; 2.4.0 to before 2.4.3VulnerableGoogle
cpe:2.3:a:google:tensorflow:2.5.0:*:*:*:*:*:*:*2.5.0VulnerableGoogle
cpe:2.3:a:google:tensorflow:2.6.0:rc0:*:*:*:*:*:*2.6.0 rc0VulnerableGoogle
cpe:2.3:a:google:tensorflow:2.6.0:rc1:*:*:*:*:*:*2.6.0 rc1VulnerableGoogle
cpe:2.3:a:google:tensorflow:2.6.0:rc2:*:*:*:*:*:*2.6.0 rc2VulnerableGoogle
  • cpe:2.3:a:google:tensorflow:*:*:*:*:*:*:*:* Versions 2.3.0 to before 2.3.4; 2.4.0 to before 2.4.3 Vulnerable ·Google
  • cpe:2.3:a:google:tensorflow:2.5.0:*:*:*:*:*:*:* Versions 2.5.0 Vulnerable ·Google
  • cpe:2.3:a:google:tensorflow:2.6.0:rc0:*:*:*:*:*:* Versions 2.6.0 rc0 Vulnerable ·Google
  • cpe:2.3:a:google:tensorflow:2.6.0:rc1:*:*:*:*:*:* Versions 2.6.0 rc1 Vulnerable ·Google
  • cpe:2.3:a:google:tensorflow:2.6.0:rc2:*:*:*:*:*:* Versions 2.6.0 rc2 Vulnerable ·Google

Remediation

  • Fix · CNA tensorflow (tensorflow) Fixed in 2.5.1, 2.4.3, 2.3.4
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow (PyPI) Fixed in 2.3.4 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow (PyPI) Fixed in 2.4.3 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow (PyPI) Fixed in 2.5.1 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow-cpu (PyPI) Fixed in 2.3.4 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow-cpu (PyPI) Fixed in 2.4.3 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow-cpu (PyPI) Fixed in 2.5.1 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow-gpu (PyPI) Fixed in 2.3.4 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow-gpu (PyPI) Fixed in 2.4.3 Advisory →
  • Fix · GHSA-5hj3-vjjf-f5m7 tensorflow-gpu (PyPI) Fixed in 2.5.1 Advisory →

Timeline

  1. 2026-10-02 Threatist record created from CVE-2021-37672
  2. 2026-10-03 Alias added: GHSA-5hj3-vjjf-f5m7

References

Patch 1

Advisory 2

Other 4

  • TI-2022-20241 FractionalMaxPool and FractionalAVGPool heap out-of-bounds acess in Tensorflow CVSS 3.1CRITICAL 9.8
  • TI-2022-07640 Integer overflow in Tensorflow CVSS 3.1CRITICAL 9.8
  • TI-2022-07627 Out of bounds read and write in Tensorflow CVSS 3.1HIGH 8.8
  • TI-2022-07619 Out of bounds write in Tensorflow CVSS 3.1HIGH 8.8
  • TI-2022-07613 Read and Write outside of bounds in TFLite CVSS 3.1HIGH 8.8
  • TI-2022-07612 Integer overflow in TFLite CVSS 3.1HIGH 8.8
See all 430 →

Sources

This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences