In affected versions the shape inference code for `tf.raw_ops.Dequantize` has a vulnerability that could trigger a denial of service via a segfault if an attacker provides invalid arguments. TensorFlow is an end-to-end open source platform for machine learning.

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. The fix will be included in TensorFlow 2.6.0. We have patched the issue in GitHub commit d94ffe08a65400f898241c0374e9edc6fa8ed257. The check following the `if` statement will pass and the `for` loop would read one element before the start of `input_dims.data` (when `i = 0`). If `axis` is a large negative value (e.g., `-100000`), then after the first `if` it would still be negative. In affected versions TFLite's () contains a vulnerability which allows reading one element outside of bounds of heap allocated data. TensorFlow 2.6.0 is the only affected version. We have patched the issue in GitHub commit dfa22b348b70bb89d6d6ec0ff53973bacb4f4695.

In this case, the inner loop does not increase `i` and the `continue` statement causes execution to skip over the preincrement at the end of the outer loop. An attacker can craft a model such that `ellipsis_end_idx` is smaller than `i` (e.g., always negative). This arises from newly introduced support for (). In affected versions the strided slice implementation in TFLite has a logic bug which can allow an attacker to trigger an infinite loop.

Goodbye 3 / 5.2 Denial of Service Tool