VYPR
← All advisories
Low severityNVD Advisory· Published May 14, 2021· Updated Oct 31, 2024

CHECK-fail in `QuantizeAndDequantizeV4Grad`

CVE-2021-29544

Description

TensorFlow is an end-to-end open source platform for machine learning. An attacker can trigger a denial of service via a CHECK-fail in tf.raw_ops.QuantizeAndDequantizeV4Grad. This is because the implementation does not validate the rank of the input_* tensors. In turn, this results in the tensors being passes as they are to QuantizeAndDequantizePerChannelGradientImpl. However, the vec<T> method, requires the rank to 1 and triggers a CHECK failure otherwise. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2 as this is the only other affected version.

Affected packages

Versions sourced from the GitHub Security Advisory.

PackageAffected versionsPatched versions
tensorflowPyPI
>= 2.4.0, < 2.4.22.4.2
tensorflow-cpuPyPI
>= 2.4.0, < 2.4.22.4.2
tensorflow-gpuPyPI
>= 2.4.0, < 2.4.22.4.2

Affected products

1

Patches

1
20431e9044cf

Fix `tf.raw_ops.QuantizeAndDequantizeV4Grad` CHECK failure.

https://github.com/tensorflow/tensorflowAmit PatankarApr 26, 2021via ghsa
commit
1 file changed · +10 −0
  • tensorflow/core/kernels/quantize_and_dequantize_op.cc+10 −0 modified
    @@ -160,7 +160,17 @@ class QuantizeAndDequantizeV4GradientOp : public OpKernel {
         errors::InvalidArgument("gradient and input must be the same size"));
     const int depth = (axis_ == -1) ? 1 : input.dim_size(axis_);
     const Tensor& input_min_tensor = ctx->input(2);
+    OP_REQUIRES(ctx,
+                input_min_tensor.dims() == 0 || input_min_tensor.dims() == 1,
+                errors::InvalidArgument(
+                    "Input min tensor must have dimension 1. Recieved ",
+                    input_min_tensor.dims(), "."));
     const Tensor& input_max_tensor = ctx->input(3);
+    OP_REQUIRES(ctx,
+                input_max_tensor.dims() == 0 || input_max_tensor.dims() == 1,
+                errors::InvalidArgument(
+                    "Input max tensor must have dimension 1. Recieved ",
+                    input_max_tensor.dims(), "."));
     if (axis_ != -1) {
       OP_REQUIRES(
           ctx, input_min_tensor.dim_size(0) == depth,

Vulnerability mechanics

Generated by null/stub on May 9, 2026. Inputs: CWE entries + fix-commit diffs from this CVE's patches. Citations validated against bundle.

References

9

News mentions

0

No linked articles in our index yet.