Validate CUDA QDQ element counts

[Akshay Sonawane (apsonawane)]

This pull request introduces an element count validation for CUDA QuantizeLinear and DequantizeLinear operators to ensure they do not process more than INT32_MAX elements, preventing potential overflows and undefined behavior. It also adds corresponding unit tests to verify this constraint.

Element Count Validation:

• Added a new function ValidateQDQElementCount in quantize_linear.h to check that the number of elements does not exceed INT32_MAX, returning an error if the limit is exceeded.
• Integrated ValidateQDQElementCount into both QuantizeLinear<T, U>::ComputeInternal and DequantizeLinear<T, U>::ComputeInternal to enforce the element count constraint during operator execution. [1] [2]

Testing:

• Added a unit test CudaElementCountRange in quantize_linear_test.cc to confirm that ValidateQDQElementCount accepts the maximum allowed value and rejects values above the limit.

[Copilot]

Pull request overview

Adds a CUDA-side guard for QuantizeLinear/DequantizeLinear to prevent launching kernels with element counts that would overflow 32-bit indexing/casts, and introduces a small unit test for the new validation helper.

Changes:

• Added cuda::ValidateQDQElementCount(size_t) to enforce <= INT32_MAX element count.
• Called the new validation from CUDA QuantizeLinear and DequantizeLinear ComputeInternal paths before output allocation / kernel launch.
• Added a CUDA-only unit test to validate boundary behavior of the helper.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
onnxruntime/core/providers/cuda/tensor/quantize_linear.h	Introduces ValidateQDQElementCount helper used by CUDA Q/DQ kernels.
onnxruntime/core/providers/cuda/tensor/quantize_linear.cc	Enforces the element-count validation in ComputeInternal for both Quantize and Dequantize.
onnxruntime/test/providers/cpu/tensor/quantize_linear_test.cc	Adds a CUDA-only unit test covering the validation boundary conditions.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.