Advisory

PYSEC-2026-3121

Also known as: BIT-tensorflow-2022-21728, CVE-2022-21728, GHSA-6gmv-pjp9-p8w8, PYSEC-2022-107, PYSEC-2022-52

Summary

Out of bounds read in Tensorflow

Details

Impact
The implementation of shape inference for `ReverseSequence` does not fully validate the value of batch_dim and can result in a heap OOB read:

import tensorflow as tf

@tf.function
def test():
  y = tf.raw_ops.ReverseSequence(
    input = ['aaa','bbb'],
    seq_lengths = [1,1,1],
    seq_dim = -10,
    batch_dim = -10 )
  return y
    
test()

There is a check to make sure the value of batch_dim does not go over the rank of the input, but there is no check for negative values:

  const int32_t input_rank = c->Rank(input);
  if (batch_dim >= input_rank) {
    return errors::InvalidArgument( 
        "batch_dim must be < input rank: ", batch_dim, " vs. ", input_rank);
  }
  // ...
  
  DimensionHandle batch_dim_dim = c->Dim(input, batch_dim);

Negative dimensions are allowed in some cases to mimic Python's negative indexing (i.e., indexing from the end of the array), however if the value is too negative then the implementation of `Dim` would access elements before the start of an array:

  DimensionHandle Dim(ShapeHandle s, int64_t idx) {
    if (!s.Handle() || s->rank_ == kUnknownRank) {
      return UnknownDim();
    }
    return DimKnownRank(s, idx);
  } 
·
  static DimensionHandle DimKnownRank(ShapeHandle s, int64_t idx) {
    CHECK_NE(s->rank_, kUnknownRank);
    if (idx < 0) {
      return s->dims_[s->dims_.size() + idx];
    }
    return s->dims_[idx];
  }

Patches
We have patched the issue in GitHub commit 37c01fb5e25c3d80213060460196406c43d31995.

The fix will be included in TensorFlow 2.8.0. We will also cherrypick this commit on TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3, as these are also affected and still in supported range.

For more information
Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution
This vulnerability has been reported by Yu Tian of Qihoo 360 AIVul Team.

Severity

High Basis: CVSS v3 score 8.1, vector CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:H

Dates

Published
2026-07-09 16:49 UTC
Modified
2026-07-13 16:43 UTC

Affected packages and fixes

tensorflow

PyPI · All systems

Affected

  • introduced 0, fixed 2.5.3, introduced 2.6.0, fixed 2.6.3, introduced 2.7.0, fixed 2.7.1 (ecosystem)

Fixed versions

2.5.3, 2.6.3, 2.7.1

Fix

  • Fixed version 2.5.3
    pip install --upgrade tensorflow==2.5.3
  • Fixed version 2.6.3
    pip install --upgrade tensorflow==2.6.3
  • Fixed version 2.7.1
    pip install --upgrade tensorflow==2.7.1

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: PyPA Advisory Database · CC BY 4.0

Sources & licences