Replacing the _mm_mul_epu32 Function

This function is used to multiply the unsigned numbers at the 32th bit of the 64-bit element in vectors a and b and store the result as a 64-bit unsigned number.

For details about _mm_mul_epu32, see Intrinsics Guide.

• Code on x86:

  __m128i _mm_mul_epu32 (__m128i a, __m128i b);

• Alternative for Kunpeng processors:

  #include <arm_neon.h>
  uint64x2_t Arm_mm_mul_epu32(int32X4_t a, int32x4_t b) 
  { 
  uint64x2_t result = { uint64_t(a[0] * b[0]), uint64_t(a[2] * b[2])}; 
  return result; 
  }