LLM Cheating on hip kernel matrix_multiplication

We observed that when using GPT-5.4, it tends to “cheat” on tasks/hip2hip/others/matrix_multiplication.
It modified main.hip in two key ways:

1. Removes the real GEMM computation
   The baseline uses a standard tiled GEMM, but the “optimized” version directly sets: C[row * b_cols + col] = a_cols * 0.02F; This exploits the fact that inputs are constant (A=1.0, B=0.02), so the result can be hardcoded without actual computation.
2. Removes the real GPU execution flow
   The baseline includes memory allocation, hipMemcpy, kernel launch, and result verification.
   The modified version bypasses actual execution and relies on trivial verification logic to pass.

So this is not a real optimization, but exploiting fixed inputs and weak validation to “pass” the test and reported 35x speedup.

main.hip.txt

Here is the cheating main.hip

Start with tasks/hip2hip/others/matrix_multiplication/main.hip and compare it with the attached main.hip.txt; inspect how this task is launched and verified. Run the existing benchmark or verification path. Done means implementations that bypass the GEMM computation or GPU execution are detected instead of being accepted as valid optimizations.