Unitialized access in `EinsumHelper::ParseEquation`
High7.8CVE-2021-41201 · Published Nov 10, 2021 · updated Jul 8, 2026
### Impact During execution, [`EinsumHelper::ParseEquation()`](https://github.com/tensorflow/tensorflow/blob/e0b6e58c328059829c3eb968136f17aa72b6c876/tensorflow/core/kernels/linalg/einsum_op_impl.h#L126-L181) is supposed to set the flags in `input_has_ellipsis` vector and `*output_has_ellipsis` boolean to indicate whether there is ellipsis in the corresponding inputs and output. However, the code only changes these flags to `true` and never assigns `false`. ```cc for (int i = 0; i < num_inputs; ++i) { input_label_counts->at(i).resize(num_labels); for (const int label : input_labels->at(i)) { if (label != kEllipsisLabel) input_label_counts->at(i)[label] += 1; else input_has_ellipsis->at(i) = true; } } output_label_counts->resize(num_labels); for (const int label : *output_labels) { if (label != kEllipsisLabel) output_label_counts->at(label) += 1; else *output_has_ellipsis = true; } ``` This results in unitialized variable access if callers assume that `EinsumHelper::ParseEquation()` always sets these flags. ### Patches We have patched the issue in GitHub commit [f09caa532b6e1ac8d2aa61b7832c78c5b79300c6](https://github.com/tensorflow/tensorfl...
Affected versions
| Package | Affected | Fixed in |
|---|---|---|
| tensorflow PyPI | >= 2.6.0, < 2.6.1 | 2.6.1 |
| >= 2.5.0, < 2.5.2 | 2.5.2 | |
| < 2.4.4 | 2.4.4 |
Details and references
### Impact During execution, [`EinsumHelper::ParseEquation()`](https://github.com/tensorflow/tensorflow/blob/e0b6e58c328059829c3eb968136f17aa72b6c876/tensorflow/core/kernels/linalg/einsum_op_impl.h#L126-L181) is supposed to set the flags in `input_has_ellipsis` vector and `*output_has_ellipsis` boolean to indicate whether there is ellipsis in the corresponding inputs and output. However, the code only changes these flags to `true` and never assigns `false`. ```cc for (int i = 0; i < num_inputs; ++i) { input_label_counts->at(i).resize(num_labels); for (const int label : input_labels->at(i)) { if (label != kEllipsisLabel) input_label_counts->at(i)[label] += 1; else input_has_ellipsis->at(i) = true; } } output_label_counts->resize(num_labels); for (const int label : *output_labels) { if (label != kEllipsisLabel) output_label_counts->at(label) += 1; else *output_has_ellipsis = true; } ``` This results in unitialized variable access if callers assume that `EinsumHelper::ParseEquation()` always sets these flags. ### Patches We have patched the issue in GitHub commit [f09caa532b6e1ac8d2aa61b7832c78c5b79300c6](https://github.com/tensorflow/tensorflow/commit/f09caa532b6e1ac8d2aa61b7832c78c5b79300c6). The fix will be included in TensorFlow 2.7.0. We will also cherrypick this commit on TensorFlow 2.6.1, TensorFlow 2.5.2, and TensorFlow 2.4.4, as these are also affected and still in supported range. ### For more information Please 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.
- CVSS 3.1
- CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
- Severity from
- GitHub (reviewed advisory)
- Weakness
- CWE-824
- Also known as
- BIT-tensorflow-2021-41201, CVE-2021-41201, PYSEC-2021-394, PYSEC-2021-611, PYSEC-2021-809
- github.com/tensorflow/tensorflow/security/advisories/GHSA-j86v-p27c-73fm
- nvd.nist.gov/vuln/detail/CVE-2021-41201
- github.com/tensorflow/tensorflow/commit/f09caa532b6e1ac8d2aa61b7832c78c5b79300c6
- github.com/pypa/advisory-database/tree/main/vulns/tensorflow-cpu/PYSEC-2021-611.yaml
- github.com/pypa/advisory-database/tree/main/vulns/tensorflow-gpu/PYSEC-2021-809.yaml
- github.com/pypa/advisory-database/tree/main/vulns/tensorflow/PYSEC-2021-394.yaml
- github.com/tensorflow/tensorflow
More TensorFlow advisories
All TensorFlow| Date | Advisory | Severity | Fixed in |
|---|---|---|---|
| Nov 102021 | Incomplete validation in boosted trees code | Critical9.3 | 2.4.4+2 more |
| Nov 102021 | Crash in `tf.math.segment_*` operations | Medium5.5 | 2.4.4+2 more |
| Nov 102021 | Crash in `max_pool3d` when size argument is 0 or negative | Medium5.5 | 2.4.4+2 more |
| Nov 102021 | Crashes due to overflow and `CHECK`-fail in ops with large tensor shapes | Medium5.5 | 2.4.4+2 more |
| Nov 102021 | Overflow/crash in `tf.tile` when tiling tensor is large | Medium5.5 | 2.4.4+2 more |
| Nov 102021 | Overflow/crash in `tf.image.resize` when size is large | Medium5.5 | 2.4.4+2 more |