https://cve.radiocsirt.org Contains only the most 10 recent entries. http://www.rssboard.org/rss-specification python-feedgen en Tue, 06 Oct 2026 18:54:46 +0000 https://cve.radiocsirt.org/vuln/bit-tensorflow-2020-26271 <p><strong>Affected:</strong> Bitnami: tensorflow</p> <p>In affected versions of TensorFlow under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The MakeEdge function creates an edge between one output tensor of the src node (given by output_index) and the input slot of the dst node (given by input_index). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding DataType values and comparing these for equality. However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays. In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.</p> <p><strong>Affected:</strong> Bitnami: tensorflow</p> <p>In affected versions of TensorFlow under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The MakeEdge function creates an edge between one output tensor of the src node (given by output_index) and the input slot of the dst node (given by input_index). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding DataType values and comparing these for equality. However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays. In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.</p> https://cve.radiocsirt.org/vuln/bit-tensorflow-2020-26271 https://cve.radiocsirt.org/vuln/cnvd-2021-00092 cnvd-2021-00092 https://cve.radiocsirt.org/vuln/cnvd-2021-00092 https://cve.radiocsirt.org/vuln/euvd-2026-46050 EUVD-2026-46050 https://cve.radiocsirt.org/vuln/euvd-2026-46050 https://cve.radiocsirt.org/vuln/fkie_cve-2020-26271 <p>In affected versions of TensorFlow under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The MakeEdge function creates an edge between one output tensor of the src node (given by output_index) and the input slot of the dst node (given by input_index). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding DataType values and comparing these for equality. However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays. In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.</p> <p>In affected versions of TensorFlow under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The MakeEdge function creates an edge between one output tensor of the src node (given by output_index) and the input slot of the dst node (given by input_index). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding DataType values and comparing these for equality. However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays. In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.</p> https://cve.radiocsirt.org/vuln/fkie_cve-2020-26271 https://cve.radiocsirt.org/vuln/ghsa-q263-fvxm-m5mw <p><strong>Affected:</strong> PyPI: tensorflow, PyPI: tensorflow-cpu, PyPI: tensorflow-gpu</p> <p>### Impact Under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The [`MakeEdge` function](https://github.com/tensorflow/tensorflow/blob/3616708cb866365301d8e67b43b32b46d94b08a0/tensorflow/core/common_runtime/graph_constructor.cc#L1426-L1438) creates an edge between one output tensor of the `src` node (given by `output_index`) and the input slot of the `dst` node (given by `input_index`). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding `DataType` values and comparing these for equality:</p> <p>```cc DataType src_out = src-&gt;output_type(output_index); DataType dst_in = dst-&gt;input_type(input_index); //... ```</p> <p>However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays.</p> <p>In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library.</p> <p>### Patches We have patched the issue in GitHub commit [0cc38aaa4064fd9e79101994ce9872c6d91f816b](https://github.com/tensorflow/tensorflow/commit/0cc38aaa4064fd9e79101994ce9872c6d91f816b) and will release TensorFlow 2.4.0 containing the patch. TensorFlow nightly packages after this commit will also have the issue resolved.</p> <p>Since this issue also impacts…</p> <p><strong>Affected:</strong> PyPI: tensorflow, PyPI: tensorflow-cpu, PyPI: tensorflow-gpu</p> <p>### Impact Under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The [`MakeEdge` function](https://github.com/tensorflow/tensorflow/blob/3616708cb866365301d8e67b43b32b46d94b08a0/tensorflow/core/common_runtime/graph_constructor.cc#L1426-L1438) creates an edge between one output tensor of the `src` node (given by `output_index`) and the input slot of the `dst` node (given by `input_index`). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding `DataType` values and comparing these for equality:</p> <p>```cc DataType src_out = src-&gt;output_type(output_index); DataType dst_in = dst-&gt;input_type(input_index); //... ```</p> <p>However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays.</p> <p>In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library.</p> <p>### Patches We have patched the issue in GitHub commit [0cc38aaa4064fd9e79101994ce9872c6d91f816b](https://github.com/tensorflow/tensorflow/commit/0cc38aaa4064fd9e79101994ce9872c6d91f816b) and will release TensorFlow 2.4.0 containing the patch. TensorFlow nightly packages after this commit will also have the issue resolved.</p> <p>Since this issue also impacts…</p> https://cve.radiocsirt.org/vuln/ghsa-q263-fvxm-m5mw https://cve.radiocsirt.org/vuln/gsd-2020-26271 gsd-2020-26271 https://cve.radiocsirt.org/vuln/gsd-2020-26271 https://cve.radiocsirt.org/vuln/opensuse-su-2022:10014-1 <p>Security update for tensorflow2</p> <p>Security update for tensorflow2</p> https://cve.radiocsirt.org/vuln/opensuse-su-2022:10014-1 https://cve.radiocsirt.org/vuln/pysec-2020-257 <p><strong>Affected:</strong> PyPI: tensorflow</p> <p>In affected versions of TensorFlow under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The MakeEdge function creates an edge between one output tensor of the src node (given by output_index) and the input slot of the dst node (given by input_index). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding DataType values and comparing these for equality. However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays. In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.</p> <p><strong>Affected:</strong> PyPI: tensorflow</p> <p>In affected versions of TensorFlow under certain cases, loading a saved model can result in accessing uninitialized memory while building the computation graph. The MakeEdge function creates an edge between one output tensor of the src node (given by output_index) and the input slot of the dst node (given by input_index). This is only possible if the types of the tensors on both sides coincide, so the function begins by obtaining the corresponding DataType values and comparing these for equality. However, there is no check that the indices point to inside of the arrays they index into. Thus, this can result in accessing data out of bounds of the corresponding heap allocated arrays. In most scenarios, this can manifest as unitialized data access, but if the index points far away from the boundaries of the arrays this can be used to leak addresses from the library. This is fixed in versions 1.15.5, 2.0.4, 2.1.3, 2.2.2, 2.3.2, and 2.4.0.</p> https://cve.radiocsirt.org/vuln/pysec-2020-257