void test_vqaddQu32 (void)
{
  uint32x4_t out_uint32x4_t;
  uint32x4_t arg0_uint32x4_t;
  uint32x4_t arg1_uint32x4_t;

  out_uint32x4_t = vqaddq_u32 (arg0_uint32x4_t, arg1_uint32x4_t);
}
예제 #2
0
파일: mw_neon.c 프로젝트: Daniel-Hwang/eeg
/* u32x4 saturated add */
void mw_neon_mm_qadd_u32x4(unsigned int * A, int Row, int Col, unsigned int * B, unsigned int * C)
{
	uint32x4_t neon_a, neon_b, neon_c;
	int size = Row * Col;
	int i = 0;
	int k = 0;

	for (i = 4; i <= size ; i+=4)
	{
		k = i - 4;
		neon_a = vld1q_u32(A + k);
		neon_b = vld1q_u32(B + k);
		neon_c = vqaddq_u32(neon_a, neon_b);
		vst1q_u32(C + k, neon_c);
	}

	k = i - 4;
    for (i = 0; i < size % 4; i++)
	{
		C[k + i] = A[k + i] + B[k + i];
	}
}
예제 #3
0
inline  uint32x4_t vqaddq(const uint32x4_t  & v0, const uint32x4_t  & v1) { return vqaddq_u32(v0, v1); }