# Perform Matrix Multiplication
ptx_macro_generator.MMA(ptx_macro_generator, A_local, B_local, C_local)

  for ki in T.serial(0, (block_K // micro_size_k)):

      # Load A into fragment
      mma_emitter.ldmatrix_a(
          A_local,
          A_shared,
          ki,
          thread_bindings=thread_bindings,
      )

      # Load B into fragment
      mma_emitter.ldmatrix_b(
          B_local,
          B_shared,
          ki,
          thread_bindings=thread_bindings,
      )

      # Perform Matrix Multiplication
      mma_emitter.mma(A_local, B_local, C_local)

# Perform STMatrix
mma_emitter.stmatrix(
  C_local,
  C_shared,
  thread_bindings=thread_bindings,
)