Defines
#define	_mm256_mpsadbw_epu8(X, Y, M) __builtin_ia32_mpsadbw256((X), (Y), (M))
#define	_mm256_alignr_epi8(a, b, n)
#define	_mm256_blend_epi16(V1, V2, M)
#define	_mm256_shuffle_epi32(a, imm)
#define	_mm256_shufflehi_epi16(a, imm)
#define	_mm256_shufflelo_epi16(a, imm)
#define	_mm256_slli_si256(a, count)
#define	_mm256_srli_si256(a, count)
#define	_mm_blend_epi32(V1, V2, M)
#define	_mm256_blend_epi32(V1, V2, M)
#define	_mm256_permute4x64_pd(V, M)
#define	_mm256_permute4x64_epi64(V, M)
#define	_mm256_permute2x128_si256(V1, V2, M)
#define	_mm256_extracti128_si256(A, O)
#define	_mm256_inserti128_si256(V1, V2, O)
#define	_mm_mask_i32gather_pd(a, m, i, mask, s)
#define	_mm256_mask_i32gather_pd(a, m, i, mask, s)
#define	_mm_mask_i64gather_pd(a, m, i, mask, s)
#define	_mm256_mask_i64gather_pd(a, m, i, mask, s)
#define	_mm_mask_i32gather_ps(a, m, i, mask, s)
#define	_mm256_mask_i32gather_ps(a, m, i, mask, s)
#define	_mm_mask_i64gather_ps(a, m, i, mask, s)
#define	_mm256_mask_i64gather_ps(a, m, i, mask, s)
#define	_mm_mask_i32gather_epi32(a, m, i, mask, s)
#define	_mm256_mask_i32gather_epi32(a, m, i, mask, s)
#define	_mm_mask_i64gather_epi32(a, m, i, mask, s)
#define	_mm256_mask_i64gather_epi32(a, m, i, mask, s)
#define	_mm_mask_i32gather_epi64(a, m, i, mask, s)
#define	_mm256_mask_i32gather_epi64(a, m, i, mask, s)
#define	_mm_mask_i64gather_epi64(a, m, i, mask, s)
#define	_mm256_mask_i64gather_epi64(a, m, i, mask, s)
#define	_mm_i32gather_pd(m, i, s)
#define	_mm256_i32gather_pd(m, i, s)
#define	_mm_i64gather_pd(m, i, s)
#define	_mm256_i64gather_pd(m, i, s)
#define	_mm_i32gather_ps(m, i, s)
#define	_mm256_i32gather_ps(m, i, s)
#define	_mm_i64gather_ps(m, i, s)
#define	_mm256_i64gather_ps(m, i, s)
#define	_mm_i32gather_epi32(m, i, s)
#define	_mm256_i32gather_epi32(m, i, s)
#define	_mm_i64gather_epi32(m, i, s)
#define	_mm256_i64gather_epi32(m, i, s)
#define	_mm_i32gather_epi64(m, i, s)
#define	_mm256_i32gather_epi64(m, i, s)
#define	_mm_i64gather_epi64(m, i, s)
#define	_mm256_i64gather_epi64(m, i, s)
Functions
static __inline__ __m256i	__attribute__ ((__always_inline__, __nodebug__)) _mm256_abs_epi8(__m256i __a)

Define Documentation

#define _mm256_alignr_epi8	(	a,
		b,
		n
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  __m256i __b = (b); \
  (__m256i)__builtin_ia32_palignr256((__v32qi)__a, (__v32qi)__b, (n)); })

#define _mm256_blend_epi16	(	V1,
		V2,
		M
	)

Value:

__extension__ ({ \
  __m256i __V1 = (V1); \
  __m256i __V2 = (V2); \
  (__m256d)__builtin_shufflevector((__v16hi)__V1, (__v16hi)__V2, \
                                   (((M) & 0x01) ? 16 : 0), \
                                   (((M) & 0x02) ? 17 : 1), \
                                   (((M) & 0x04) ? 18 : 2), \
                                   (((M) & 0x08) ? 19 : 3), \
                                   (((M) & 0x10) ? 20 : 4), \
                                   (((M) & 0x20) ? 21 : 5), \
                                   (((M) & 0x40) ? 22 : 6), \
                                   (((M) & 0x80) ? 23 : 7), \
                                   (((M) & 0x01) ? 24 : 8), \
                                   (((M) & 0x02) ? 25 : 9), \
                                   (((M) & 0x04) ? 26 : 10), \
                                   (((M) & 0x08) ? 27 : 11), \
                                   (((M) & 0x10) ? 28 : 12), \
                                   (((M) & 0x20) ? 29 : 13), \
                                   (((M) & 0x40) ? 30 : 14), \
                                   (((M) & 0x80) ? 31 : 15)); })

#define _mm256_blend_epi32	(	V1,
		V2,
		M
	)

Value:

__extension__ ({ \
  __m256i __V1 = (V1); \
  __m256i __V2 = (V2); \
  (__m256i)__builtin_shufflevector((__v8si)__V1, (__v8si)__V2, \
                                   (((M) & 0x01) ?  8 : 0), \
                                   (((M) & 0x02) ?  9 : 1), \
                                   (((M) & 0x04) ? 10 : 2), \
                                   (((M) & 0x08) ? 11 : 3), \
                                   (((M) & 0x10) ? 12 : 4), \
                                   (((M) & 0x20) ? 13 : 5), \
                                   (((M) & 0x40) ? 14 : 6), \
                                   (((M) & 0x80) ? 15 : 7)); })

#define _mm256_extracti128_si256	(	A,
		O
	)

Value:

__extension__ ({ \
  __m256i __A = (A); \
  (__m128i)__builtin_ia32_extract128i256(__A, (O)); })

#define _mm256_i32gather_epi32	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  int const *__m = (m); \
  __m256i __i = (i); \
  (__m256i)__builtin_ia32_gatherd_d256((__v8si)_mm256_setzero_si256(), \
            (const __v8si *)__m, (__v8si)__i, \
            (__v8si)_mm256_set1_epi32(-1), (s)); })

#define _mm256_i32gather_epi64	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  long long const *__m = (m); \
  __m128i __i = (i); \
  (__m256i)__builtin_ia32_gatherd_q256((__v4di)_mm256_setzero_si256(), \
             (const __v4di *)__m, (__v4si)__i, \
             (__v4di)_mm256_set1_epi64x(-1), (s)); })

#define _mm256_i32gather_pd	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  double const *__m = (m); \
  __m128i __i = (i); \
  (__m256d)__builtin_ia32_gatherd_pd256((__v4df)_mm256_setzero_pd(), \
             (const __v4df *)__m, (__v4si)__i, \
             (__v4df)_mm256_set1_pd((double)(long long int)-1), (s)); })

#define _mm256_i32gather_ps	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  float const *__m = (m); \
  __m256i __i = (i); \
  (__m256)__builtin_ia32_gatherd_ps256((__v8sf)_mm256_setzero_ps(), \
             (const __v8sf *)__m, (__v8si)__i, \
             (__v8sf)_mm256_set1_ps((float)(int)-1), (s)); })

#define _mm256_i64gather_epi32	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  int const *__m = (m); \
  __m256i __i = (i); \
  (__m128i)__builtin_ia32_gatherq_d256((__v4si)_mm_setzero_si128(), \
            (const __v4si *)__m, (__v4di)__i, \
            (__v4si)_mm_set1_epi32(-1), (s)); })

#define _mm256_i64gather_epi64	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  long long const *__m = (m); \
  __m256i __i = (i); \
  (__m256i)__builtin_ia32_gatherq_q256((__v4di)_mm256_setzero_si256(), \
             (const __v4di *)__m, (__v4di)__i, \
             (__v4di)_mm256_set1_epi64x(-1), (s)); })

#define _mm256_i64gather_pd	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  double const *__m = (m); \
  __m256i __i = (i); \
  (__m256d)__builtin_ia32_gatherq_pd256((__v4df)_mm256_setzero_pd(), \
             (const __v4df *)__m, (__v4di)__i, \
             (__v4df)_mm256_set1_pd((double)(long long int)-1), (s)); })

#define _mm256_i64gather_ps	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  float const *__m = (m); \
  __m256i __i = (i); \
  (__m128)__builtin_ia32_gatherq_ps256((__v4sf)_mm_setzero_ps(), \
             (const __v4sf *)__m, (__v4di)__i, \
             (__v4sf)_mm_set1_ps((float)(int)-1), (s)); })

#define _mm256_inserti128_si256	(	V1,
		V2,
		O
	)

Value:

__extension__ ({ \
  __m256i __V1 = (V1); \
  __m128i __V2 = (V2); \
  (__m256i)__builtin_ia32_insert128i256(__V1, __V2, (O)); })

#define _mm256_mask_i32gather_epi32	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  int const *__m = (m); \
  __m256i __i = (i); \
  __m256i __mask = (mask); \
  (__m256i)__builtin_ia32_gatherd_d256((__v8si)__a, (const __v8si *)__m, \
            (__v8si)__i, (__v8si)__mask, (s)); })

#define _mm256_mask_i32gather_epi64	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  long long const *__m = (m); \
  __m128i __i = (i); \
  __m256i __mask = (mask); \
  (__m256i)__builtin_ia32_gatherd_q256((__v4di)__a, (const __v4di *)__m, \
             (__v4si)__i, (__v4di)__mask, (s)); })

#define _mm256_mask_i32gather_pd	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m256d __a = (a); \
  double const *__m = (m); \
  __m128i __i = (i); \
  __m256d __mask = (mask); \
  (__m256d)__builtin_ia32_gatherd_pd256((__v4df)__a, (const __v4df *)__m, \
             (__v4si)__i, (__v4df)__mask, (s)); })

#define _mm256_mask_i32gather_ps	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m256 __a = (a); \
  float const *__m = (m); \
  __m256i __i = (i); \
  __m256 __mask = (mask); \
  (__m256)__builtin_ia32_gatherd_ps256((__v8sf)__a, (const __v8sf *)__m, \
            (__v8si)__i, (__v8sf)__mask, (s)); })

#define _mm256_mask_i64gather_epi32	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128i __a = (a); \
  int const *__m = (m); \
  __m256i __i = (i); \
  __m128i __mask = (mask); \
  (__m128i)__builtin_ia32_gatherq_d256((__v4si)__a, (const __v4si *)__m, \
            (__v4di)__i, (__v4si)__mask, (s)); })

#define _mm256_mask_i64gather_epi64	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  long long const *__m = (m); \
  __m256i __i = (i); \
  __m256i __mask = (mask); \
  (__m256i)__builtin_ia32_gatherq_q256((__v4di)__a, (const __v4di *)__m, \
             (__v4di)__i, (__v4di)__mask, (s)); })

#define _mm256_mask_i64gather_pd	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m256d __a = (a); \
  double const *__m = (m); \
  __m256i __i = (i); \
  __m256d __mask = (mask); \
  (__m256d)__builtin_ia32_gatherq_pd256((__v4df)__a, (const __v4df *)__m, \
             (__v4di)__i, (__v4df)__mask, (s)); })

#define _mm256_mask_i64gather_ps	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128 __a = (a); \
  float const *__m = (m); \
  __m256i __i = (i); \
  __m128 __mask = (mask); \
  (__m128)__builtin_ia32_gatherq_ps256((__v4sf)__a, (const __v4sf *)__m, \
            (__v4di)__i, (__v4sf)__mask, (s)); })

#define _mm256_mpsadbw_epu8	(	X,
		Y,
		M
	)	__builtin_ia32_mpsadbw256((X), (Y), (M))

Definition at line 32 of file avx2intrin.h.

#define _mm256_permute2x128_si256	(	V1,
		V2,
		M
	)

Value:

__extension__ ({ \
  __m256i __V1 = (V1); \
  __m256i __V2 = (V2); \
  (__m256i)__builtin_ia32_permti256(__V1, __V2, (M)); })

#define _mm256_permute4x64_epi64	(	V,
		M
	)

Value:

__extension__ ({ \
  __m256i __V = (V); \
  (__m256i)__builtin_shufflevector((__v4di)__V, (__v4di) _mm256_setzero_si256(), \
                                   (M) & 0x3, ((M) & 0xc) >> 2, \
                                   ((M) & 0x30) >> 4, ((M) & 0xc0) >> 6); })

#define _mm256_permute4x64_pd	(	V,
		M
	)

Value:

__extension__ ({ \
  __m256d __V = (V); \
  (__m256d)__builtin_shufflevector((__v4df)__V, (__v4df) _mm256_setzero_pd(), \
                                   (M) & 0x3, ((M) & 0xc) >> 2, \
                                   ((M) & 0x30) >> 4, ((M) & 0xc0) >> 6); })

#define _mm256_shuffle_epi32	(	a,
		imm
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  (__m256i)__builtin_shufflevector((__v8si)__a, (__v8si)_mm256_set1_epi32(0), \
                                   (imm) & 0x3, ((imm) & 0xc) >> 2, \
                                   ((imm) & 0x30) >> 4, ((imm) & 0xc0) >> 6, \
                                   4 + (((imm) & 0x03) >> 0), \
                                   4 + (((imm) & 0x0c) >> 2), \
                                   4 + (((imm) & 0x30) >> 4), \
                                   4 + (((imm) & 0xc0) >> 6)); })

#define _mm256_shufflehi_epi16	(	a,
		imm
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  (__m256i)__builtin_shufflevector((__v16hi)__a, (__v16hi)_mm256_set1_epi16(0), \
                                   0, 1, 2, 3, \
                                   4 + (((imm) & 0x03) >> 0), \
                                   4 + (((imm) & 0x0c) >> 2), \
                                   4 + (((imm) & 0x30) >> 4), \
                                   4 + (((imm) & 0xc0) >> 6), \
                                   8, 9, 10, 11, \
                                   12 + (((imm) & 0x03) >> 0), \
                                   12 + (((imm) & 0x0c) >> 2), \
                                   12 + (((imm) & 0x30) >> 4), \
                                   12 + (((imm) & 0xc0) >> 6)); })

#define _mm256_shufflelo_epi16	(	a,
		imm
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  (__m256i)__builtin_shufflevector((__v16hi)__a, (__v16hi)_mm256_set1_epi16(0), \
                                   (imm) & 0x3,((imm) & 0xc) >> 2, \
                                   ((imm) & 0x30) >> 4, ((imm) & 0xc0) >> 6, \
                                   4, 5, 6, 7, \
                                   8 + (((imm) & 0x03) >> 0), \
                                   8 + (((imm) & 0x0c) >> 2), \
                                   8 + (((imm) & 0x30) >> 4), \
                                   8 + (((imm) & 0xc0) >> 6), \
                                   12, 13, 14, 15); })

#define _mm256_slli_si256	(	a,
		count
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  (__m256i)__builtin_ia32_pslldqi256(__a, (count)*8); })

#define _mm256_srli_si256	(	a,
		count
	)

Value:

__extension__ ({ \
  __m256i __a = (a); \
  (__m256i)__builtin_ia32_psrldqi256(__a, (count)*8); })

#define _mm_blend_epi32	(	V1,
		V2,
		M
	)

Value:

__extension__ ({ \
  __m128i __V1 = (V1); \
  __m128i __V2 = (V2); \
  (__m128i)__builtin_shufflevector((__v4si)__V1, (__v4si)__V2, \
                                   (((M) & 0x01) ? 4 : 0), \
                                   (((M) & 0x02) ? 5 : 1), \
                                   (((M) & 0x04) ? 6 : 2), \
                                   (((M) & 0x08) ? 7 : 3)); })

#define _mm_i32gather_epi32	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  int const *__m = (m); \
  __m128i __i = (i); \
  (__m128i)__builtin_ia32_gatherd_d((__v4si)_mm_setzero_si128(), \
            (const __v4si *)__m, (__v4si)__i, \
            (__v4si)_mm_set1_epi32(-1), (s)); })

#define _mm_i32gather_epi64	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  long long const *__m = (m); \
  __m128i __i = (i); \
  (__m128i)__builtin_ia32_gatherd_q((__v2di)_mm_setzero_si128(), \
             (const __v2di *)__m, (__v4si)__i, \
             (__v2di)_mm_set1_epi64x(-1), (s)); })

#define _mm_i32gather_pd	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  double const *__m = (m); \
  __m128i __i = (i); \
  (__m128d)__builtin_ia32_gatherd_pd((__v2df)_mm_setzero_pd(), \
             (const __v2df *)__m, (__v4si)__i, \
             (__v2df)_mm_set1_pd((double)(long long int)-1), (s)); })

#define _mm_i32gather_ps	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  float const *__m = (m); \
  __m128i __i = (i); \
  (__m128)__builtin_ia32_gatherd_ps((__v4sf)_mm_setzero_ps(), \
             (const __v4sf *)__m, (__v4si)__i, \
             (__v4sf)_mm_set1_ps((float)(int)-1), (s)); })

#define _mm_i64gather_epi32	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  int const *__m = (m); \
  __m128i __i = (i); \
  (__m128i)__builtin_ia32_gatherq_d((__v4si)_mm_setzero_si128(), \
            (const __v4si *)__m, (__v2di)__i, \
            (__v4si)_mm_set1_epi32(-1), (s)); })

#define _mm_i64gather_epi64	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  long long const *__m = (m); \
  __m128i __i = (i); \
  (__m128i)__builtin_ia32_gatherq_q((__v2di)_mm_setzero_si128(), \
             (const __v2di *)__m, (__v2di)__i, \
             (__v2di)_mm_set1_epi64x(-1), (s)); })

#define _mm_i64gather_pd	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  double const *__m = (m); \
  __m128i __i = (i); \
  (__m128d)__builtin_ia32_gatherq_pd((__v2df)_mm_setzero_pd(), \
             (const __v2df *)__m, (__v2di)__i, \
             (__v2df)_mm_set1_pd((double)(long long int)-1), (s)); })

#define _mm_i64gather_ps	(	m,
		i,
		s
	)

Value:

__extension__ ({ \
  float const *__m = (m); \
  __m128i __i = (i); \
  (__m128)__builtin_ia32_gatherq_ps((__v4sf)_mm_setzero_ps(), \
             (const __v4sf *)__m, (__v2di)__i, \
             (__v4sf)_mm_set1_ps((float)(int)-1), (s)); })

#define _mm_mask_i32gather_epi32	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128i __a = (a); \
  int const *__m = (m); \
  __m128i __i = (i); \
  __m128i __mask = (mask); \
  (__m128i)__builtin_ia32_gatherd_d((__v4si)__a, (const __v4si *)__m, \
            (__v4si)__i, (__v4si)__mask, (s)); })

#define _mm_mask_i32gather_epi64	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128i __a = (a); \
  long long const *__m = (m); \
  __m128i __i = (i); \
  __m128i __mask = (mask); \
  (__m128i)__builtin_ia32_gatherd_q((__v2di)__a, (const __v2di *)__m, \
             (__v4si)__i, (__v2di)__mask, (s)); })

#define _mm_mask_i32gather_pd	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128d __a = (a); \
  double const *__m = (m); \
  __m128i __i = (i); \
  __m128d __mask = (mask); \
  (__m128d)__builtin_ia32_gatherd_pd((__v2df)__a, (const __v2df *)__m, \
             (__v4si)__i, (__v2df)__mask, (s)); })

#define _mm_mask_i32gather_ps	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128 __a = (a); \
  float const *__m = (m); \
  __m128i __i = (i); \
  __m128 __mask = (mask); \
  (__m128)__builtin_ia32_gatherd_ps((__v4sf)__a, (const __v4sf *)__m, \
            (__v4si)__i, (__v4sf)__mask, (s)); })

#define _mm_mask_i64gather_epi32	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128i __a = (a); \
  int const *__m = (m); \
  __m128i __i = (i); \
  __m128i __mask = (mask); \
  (__m128i)__builtin_ia32_gatherq_d((__v4si)__a, (const __v4si *)__m, \
            (__v2di)__i, (__v4si)__mask, (s)); })

#define _mm_mask_i64gather_epi64	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128i __a = (a); \
  long long const *__m = (m); \
  __m128i __i = (i); \
  __m128i __mask = (mask); \
  (__m128i)__builtin_ia32_gatherq_q((__v2di)__a, (const __v2di *)__m, \
             (__v2di)__i, (__v2di)__mask, (s)); })

#define _mm_mask_i64gather_pd	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128d __a = (a); \
  double const *__m = (m); \
  __m128i __i = (i); \
  __m128d __mask = (mask); \
  (__m128d)__builtin_ia32_gatherq_pd((__v2df)__a, (const __v2df *)__m, \
             (__v2di)__i, (__v2df)__mask, (s)); })

#define _mm_mask_i64gather_ps	(	a,
		m,
		i,
		mask,
		s
	)

Value:

__extension__ ({ \
  __m128 __a = (a); \
  float const *__m = (m); \
  __m128i __i = (i); \
  __m128 __mask = (mask); \
  (__m128)__builtin_ia32_gatherq_ps((__v4sf)__a, (const __v4sf *)__m, \
            (__v2di)__i, (__v4sf)__mask, (s)); })

Function Documentation

static __inline__ __m256i __attribute__ ( (__always_inline__, __nodebug__) ) [static]

Definition at line 34 of file avx2intrin.h.

Defines

Functions

Define Documentation

Function Documentation