Go to the documentation of this file.
13 #if !defined(__i386__) && !defined(__x86_64__)
14 #error "This header is only meant to be used on x86 and x64 architecture"
17 typedef long long __m64
__attribute__((__vector_size__(8), __aligned__(8)));
25 #define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("mmx"), __min_vector_width__(64)))
34 static __inline__
void __attribute__((__always_inline__, __nodebug__, __target__(
"mmx")))
37 __builtin_ia32_emms();
54 return (__m64)__builtin_ia32_vec_init_v2si(__i, 0);
71 return __builtin_ia32_vec_ext_v2si((__v2si)__m, 0);
103 return (
long long)__m;
133 return (__m64)__builtin_ia32_packsswb((__v4hi)__m1, (__v4hi)__m2);
163 return (__m64)__builtin_ia32_packssdw((__v2si)__m1, (__v2si)__m2);
193 return (__m64)__builtin_ia32_packuswb((__v4hi)__m1, (__v4hi)__m2);
220 return (__m64)__builtin_ia32_punpckhbw((__v8qi)__m1, (__v8qi)__m2);
243 return (__m64)__builtin_ia32_punpckhwd((__v4hi)__m1, (__v4hi)__m2);
264 return (__m64)__builtin_ia32_punpckhdq((__v2si)__m1, (__v2si)__m2);
291 return (__m64)__builtin_ia32_punpcklbw((__v8qi)__m1, (__v8qi)__m2);
314 return (__m64)__builtin_ia32_punpcklwd((__v4hi)__m1, (__v4hi)__m2);
335 return (__m64)__builtin_ia32_punpckldq((__v2si)__m1, (__v2si)__m2);
356 return (__m64)__builtin_ia32_paddb((__v8qi)__m1, (__v8qi)__m2);
377 return (__m64)__builtin_ia32_paddw((__v4hi)__m1, (__v4hi)__m2);
398 return (__m64)__builtin_ia32_paddd((__v2si)__m1, (__v2si)__m2);
420 return (__m64)__builtin_ia32_paddsb((__v8qi)__m1, (__v8qi)__m2);
443 return (__m64)__builtin_ia32_paddsw((__v4hi)__m1, (__v4hi)__m2);
465 return (__m64)__builtin_ia32_paddusb((__v8qi)__m1, (__v8qi)__m2);
487 return (__m64)__builtin_ia32_paddusw((__v4hi)__m1, (__v4hi)__m2);
508 return (__m64)__builtin_ia32_psubb((__v8qi)__m1, (__v8qi)__m2);
529 return (__m64)__builtin_ia32_psubw((__v4hi)__m1, (__v4hi)__m2);
550 return (__m64)__builtin_ia32_psubd((__v2si)__m1, (__v2si)__m2);
573 return (__m64)__builtin_ia32_psubsb((__v8qi)__m1, (__v8qi)__m2);
596 return (__m64)__builtin_ia32_psubsw((__v4hi)__m1, (__v4hi)__m2);
620 return (__m64)__builtin_ia32_psubusb((__v8qi)__m1, (__v8qi)__m2);
644 return (__m64)__builtin_ia32_psubusw((__v4hi)__m1, (__v4hi)__m2);
671 return (__m64)__builtin_ia32_pmaddwd((__v4hi)__m1, (__v4hi)__m2);
692 return (__m64)__builtin_ia32_pmulhw((__v4hi)__m1, (__v4hi)__m2);
713 return (__m64)__builtin_ia32_pmullw((__v4hi)__m1, (__v4hi)__m2);
736 return (__m64)__builtin_ia32_psllw((__v4hi)__m, __count);
758 return (__m64)__builtin_ia32_psllwi((__v4hi)__m, __count);
781 return (__m64)__builtin_ia32_pslld((__v2si)__m, __count);
803 return (__m64)__builtin_ia32_pslldi((__v2si)__m, __count);
823 return (__m64)__builtin_ia32_psllq((__v1di)__m, __count);
843 return (__m64)__builtin_ia32_psllqi((__v1di)__m, __count);
867 return (__m64)__builtin_ia32_psraw((__v4hi)__m, __count);
890 return (__m64)__builtin_ia32_psrawi((__v4hi)__m, __count);
914 return (__m64)__builtin_ia32_psrad((__v2si)__m, __count);
937 return (__m64)__builtin_ia32_psradi((__v2si)__m, __count);
960 return (__m64)__builtin_ia32_psrlw((__v4hi)__m, __count);
982 return (__m64)__builtin_ia32_psrlwi((__v4hi)__m, __count);
1005 return (__m64)__builtin_ia32_psrld((__v2si)__m, __count);
1027 return (__m64)__builtin_ia32_psrldi((__v2si)__m, __count);
1047 return (__m64)__builtin_ia32_psrlq((__v1di)__m, __count);
1068 return (__m64)__builtin_ia32_psrlqi((__v1di)__m, __count);
1086 return __builtin_ia32_pand((__v1di)__m1, (__v1di)__m2);
1107 return __builtin_ia32_pandn((__v1di)__m1, (__v1di)__m2);
1125 return __builtin_ia32_por((__v1di)__m1, (__v1di)__m2);
1143 return __builtin_ia32_pxor((__v1di)__m1, (__v1di)__m2);
1165 return (__m64)__builtin_ia32_pcmpeqb((__v8qi)__m1, (__v8qi)__m2);
1187 return (__m64)__builtin_ia32_pcmpeqw((__v4hi)__m1, (__v4hi)__m2);
1209 return (__m64)__builtin_ia32_pcmpeqd((__v2si)__m1, (__v2si)__m2);
1231 return (__m64)__builtin_ia32_pcmpgtb((__v8qi)__m1, (__v8qi)__m2);
1253 return (__m64)__builtin_ia32_pcmpgtw((__v4hi)__m1, (__v4hi)__m2);
1275 return (__m64)__builtin_ia32_pcmpgtd((__v2si)__m1, (__v2si)__m2);
1288 return __extension__ (__m64){ 0LL };
1309 return (__m64)__builtin_ia32_vec_init_v2si(__i0, __i1);
1332 return (__m64)__builtin_ia32_vec_init_v4hi(__s0, __s1, __s2, __s3);
1361 _mm_set_pi8(
char __b7,
char __b6,
char __b5,
char __b4,
char __b3,
char __b2,
1362 char __b1,
char __b0)
1364 return (__m64)__builtin_ia32_vec_init_v8qi(__b0, __b1, __b2, __b3,
1365 __b4, __b5, __b6, __b7);
1494 _mm_setr_pi8(
char __b0,
char __b1,
char __b2,
char __b3,
char __b4,
char __b5,
1495 char __b6,
char __b7)
1497 return _mm_set_pi8(__b7, __b6, __b5, __b4, __b3, __b2, __b1, __b0);
1500 #undef __DEFAULT_FN_ATTRS
1503 #define _m_empty _mm_empty
1504 #define _m_from_int _mm_cvtsi32_si64
1505 #define _m_from_int64 _mm_cvtsi64_m64
1506 #define _m_to_int _mm_cvtsi64_si32
1507 #define _m_to_int64 _mm_cvtm64_si64
1508 #define _m_packsswb _mm_packs_pi16
1509 #define _m_packssdw _mm_packs_pi32
1510 #define _m_packuswb _mm_packs_pu16
1511 #define _m_punpckhbw _mm_unpackhi_pi8
1512 #define _m_punpckhwd _mm_unpackhi_pi16
1513 #define _m_punpckhdq _mm_unpackhi_pi32
1514 #define _m_punpcklbw _mm_unpacklo_pi8
1515 #define _m_punpcklwd _mm_unpacklo_pi16
1516 #define _m_punpckldq _mm_unpacklo_pi32
1517 #define _m_paddb _mm_add_pi8
1518 #define _m_paddw _mm_add_pi16
1519 #define _m_paddd _mm_add_pi32
1520 #define _m_paddsb _mm_adds_pi8
1521 #define _m_paddsw _mm_adds_pi16
1522 #define _m_paddusb _mm_adds_pu8
1523 #define _m_paddusw _mm_adds_pu16
1524 #define _m_psubb _mm_sub_pi8
1525 #define _m_psubw _mm_sub_pi16
1526 #define _m_psubd _mm_sub_pi32
1527 #define _m_psubsb _mm_subs_pi8
1528 #define _m_psubsw _mm_subs_pi16
1529 #define _m_psubusb _mm_subs_pu8
1530 #define _m_psubusw _mm_subs_pu16
1531 #define _m_pmaddwd _mm_madd_pi16
1532 #define _m_pmulhw _mm_mulhi_pi16
1533 #define _m_pmullw _mm_mullo_pi16
1534 #define _m_psllw _mm_sll_pi16
1535 #define _m_psllwi _mm_slli_pi16
1536 #define _m_pslld _mm_sll_pi32
1537 #define _m_pslldi _mm_slli_pi32
1538 #define _m_psllq _mm_sll_si64
1539 #define _m_psllqi _mm_slli_si64
1540 #define _m_psraw _mm_sra_pi16
1541 #define _m_psrawi _mm_srai_pi16
1542 #define _m_psrad _mm_sra_pi32
1543 #define _m_psradi _mm_srai_pi32
1544 #define _m_psrlw _mm_srl_pi16
1545 #define _m_psrlwi _mm_srli_pi16
1546 #define _m_psrld _mm_srl_pi32
1547 #define _m_psrldi _mm_srli_pi32
1548 #define _m_psrlq _mm_srl_si64
1549 #define _m_psrlqi _mm_srli_si64
1550 #define _m_pand _mm_and_si64
1551 #define _m_pandn _mm_andnot_si64
1552 #define _m_por _mm_or_si64
1553 #define _m_pxor _mm_xor_si64
1554 #define _m_pcmpeqb _mm_cmpeq_pi8
1555 #define _m_pcmpeqw _mm_cmpeq_pi16
1556 #define _m_pcmpeqd _mm_cmpeq_pi32
1557 #define _m_pcmpgtb _mm_cmpgt_pi8
1558 #define _m_pcmpgtw _mm_cmpgt_pi16
1559 #define _m_pcmpgtd _mm_cmpgt_pi32
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srai_pi16(__m64 __m, int __count)
Right-shifts each 16-bit integer element of a 64-bit integer vector of [4 x i16] by the number of bit...
static __inline__ int __DEFAULT_FN_ATTRS _mm_cvtsi64_si32(__m64 __m)
Returns the lower 32 bits of a 64-bit integer vector as a 32-bit signed integer.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_xor_si64(__m64 __m1, __m64 __m2)
Performs a bitwise exclusive OR of two 64-bit integer vectors.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cmpgt_pi16(__m64 __m1, __m64 __m2)
Compares the 16-bit integer elements of two 64-bit integer vectors of [4 x i16] to determine if the e...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_subs_pi16(__m64 __m1, __m64 __m2)
Subtracts each 16-bit signed integer element of the second 64-bit integer vector of [4 x i16] from th...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_set1_pi8(char __b)
Constructs a 64-bit integer vector of [8 x i8], with each of the 8-bit integer vector elements set to...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_packs_pu16(__m64 __m1, __m64 __m2)
Converts 16-bit signed integers from both 64-bit integer vector parameters of [4 x i16] into 8-bit un...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srai_pi32(__m64 __m, int __count)
Right-shifts each 32-bit integer element of a 64-bit integer vector of [2 x i32] by the number of bit...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cmpeq_pi32(__m64 __m1, __m64 __m2)
Compares the 32-bit integer elements of two 64-bit integer vectors of [2 x i32] to determine if the e...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cmpeq_pi16(__m64 __m1, __m64 __m2)
Compares the 16-bit integer elements of two 64-bit integer vectors of [4 x i16] to determine if the e...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srli_pi16(__m64 __m, int __count)
Right-shifts each 16-bit integer element of a 64-bit integer vector of [4 x i16] by the number of bit...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srli_pi32(__m64 __m, int __count)
Right-shifts each 32-bit integer element of a 64-bit integer vector of [2 x i32] by the number of bit...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sub_pi16(__m64 __m1, __m64 __m2)
Subtracts each 16-bit integer element of the second 64-bit integer vector of [4 x i16] from the corre...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_unpackhi_pi32(__m64 __m1, __m64 __m2)
Unpacks the upper 32 bits from two 64-bit integer vectors of [2 x i32] and interleaves them into a 64...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_setzero_si64(void)
Constructs a 64-bit integer vector initialized to zero.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srli_si64(__m64 __m, int __count)
Right-shifts the first parameter, which is a 64-bit integer, by the number of bits specified by the s...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_setr_pi32(int __i0, int __i1)
Constructs a 64-bit integer vector, initialized in reverse order with the specified 32-bit integer va...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sll_si64(__m64 __m, __m64 __count)
Left-shifts the first 64-bit integer parameter by the number of bits specified by the second 64-bit i...
long long __m64 __attribute__((__vector_size__(8), __aligned__(8)))
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_slli_pi32(__m64 __m, int __count)
Left-shifts each 32-bit signed integer element of a 64-bit integer vector of [2 x i32] by the number ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_setr_pi8(char __b0, char __b1, char __b2, char __b3, char __b4, char __b5, char __b6, char __b7)
Constructs a 64-bit integer vector, initialized in reverse order with the specified 8-bit integer val...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_subs_pi8(__m64 __m1, __m64 __m2)
Subtracts each 8-bit signed integer element of the second 64-bit integer vector of [8 x i8] from the ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cmpeq_pi8(__m64 __m1, __m64 __m2)
Compares the 8-bit integer elements of two 64-bit integer vectors of [8 x i8] to determine if the ele...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cvtsi64_m64(long long __i)
Casts a 64-bit signed integer value into a 64-bit integer vector.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_slli_pi16(__m64 __m, int __count)
Left-shifts each 16-bit signed integer element of a 64-bit integer vector of [4 x i16] by the number ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_set_pi16(short __s3, short __s2, short __s1, short __s0)
Constructs a 64-bit integer vector initialized with the specified 16-bit integer values.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_add_pi32(__m64 __m1, __m64 __m2)
Adds each 32-bit integer element of the first 64-bit integer vector of [2 x i32] to the corresponding...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sub_pi32(__m64 __m1, __m64 __m2)
Subtracts each 32-bit integer element of the second 64-bit integer vector of [2 x i32] from the corre...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cvtsi32_si64(int __i)
Constructs a 64-bit integer vector, setting the lower 32 bits to the value of the 32-bit integer para...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_set1_pi16(short __w)
Constructs a 64-bit integer vector of [4 x i16], with each of the 16-bit integer vector elements set ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_subs_pu16(__m64 __m1, __m64 __m2)
Subtracts each 16-bit unsigned integer element of the second 64-bit integer vector of [4 x i16] from ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_packs_pi16(__m64 __m1, __m64 __m2)
Converts 16-bit signed integers from both 64-bit integer vector parameters of [4 x i16] into 8-bit si...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_packs_pi32(__m64 __m1, __m64 __m2)
Converts 32-bit signed integers from both 64-bit integer vector parameters of [2 x i32] into 16-bit s...
static __inline__ long long __DEFAULT_FN_ATTRS _mm_cvtm64_si64(__m64 __m)
Casts a 64-bit integer vector into a 64-bit signed integer value.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srl_si64(__m64 __m, __m64 __count)
Right-shifts the first 64-bit integer parameter by the number of bits specified by the second 64-bit ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_unpacklo_pi32(__m64 __m1, __m64 __m2)
Unpacks the lower 32 bits from two 64-bit integer vectors of [2 x i32] and interleaves them into a 64...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_and_si64(__m64 __m1, __m64 __m2)
Performs a bitwise AND of two 64-bit integer vectors.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_setr_pi16(short __w0, short __w1, short __w2, short __w3)
Constructs a 64-bit integer vector, initialized in reverse order with the specified 16-bit integer va...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_andnot_si64(__m64 __m1, __m64 __m2)
Performs a bitwise NOT of the first 64-bit integer vector, and then performs a bitwise AND of the int...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_add_pi16(__m64 __m1, __m64 __m2)
Adds each 16-bit integer element of the first 64-bit integer vector of [4 x i16] to the corresponding...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sll_pi32(__m64 __m, __m64 __count)
Left-shifts each 32-bit signed integer element of the first parameter, which is a 64-bit integer vect...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_adds_pu16(__m64 __m1, __m64 __m2)
Adds each 16-bit unsigned integer element of the first 64-bit integer vector of [4 x i16] to the corr...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cmpgt_pi32(__m64 __m1, __m64 __m2)
Compares the 32-bit integer elements of two 64-bit integer vectors of [2 x i32] to determine if the e...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sub_pi8(__m64 __m1, __m64 __m2)
Subtracts each 8-bit integer element of the second 64-bit integer vector of [8 x i8] from the corresp...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sll_pi16(__m64 __m, __m64 __count)
Left-shifts each 16-bit signed integer element of the first parameter, which is a 64-bit integer vect...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_unpackhi_pi8(__m64 __m1, __m64 __m2)
Unpacks the upper 32 bits from two 64-bit integer vectors of [8 x i8] and interleaves them into a 64-...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_madd_pi16(__m64 __m1, __m64 __m2)
Multiplies each 16-bit signed integer element of the first 64-bit integer vector of [4 x i16] by the ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sra_pi16(__m64 __m, __m64 __count)
Right-shifts each 16-bit integer element of the first parameter, which is a 64-bit integer vector of ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_sra_pi32(__m64 __m, __m64 __count)
Right-shifts each 32-bit integer element of the first parameter, which is a 64-bit integer vector of ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_mullo_pi16(__m64 __m1, __m64 __m2)
Multiplies each 16-bit signed integer element of the first 64-bit integer vector of [4 x i16] by the ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_or_si64(__m64 __m1, __m64 __m2)
Performs a bitwise OR of two 64-bit integer vectors.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_unpackhi_pi16(__m64 __m1, __m64 __m2)
Unpacks the upper 32 bits from two 64-bit integer vectors of [4 x i16] and interleaves them into a 64...
static __inline__ vector float vector float __b
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_mulhi_pi16(__m64 __m1, __m64 __m2)
Multiplies each 16-bit signed integer element of the first 64-bit integer vector of [4 x i16] by the ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_subs_pu8(__m64 __m1, __m64 __m2)
Subtracts each 8-bit unsigned integer element of the second 64-bit integer vector of [8 x i8] from th...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_set_pi32(int __i1, int __i0)
Constructs a 64-bit integer vector initialized with the specified 32-bit integer values.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srl_pi32(__m64 __m, __m64 __count)
Right-shifts each 32-bit integer element of the first parameter, which is a 64-bit integer vector of ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_srl_pi16(__m64 __m, __m64 __count)
Right-shifts each 16-bit integer element of the first parameter, which is a 64-bit integer vector of ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_add_pi8(__m64 __m1, __m64 __m2)
Adds each 8-bit integer element of the first 64-bit integer vector of [8 x i8] to the corresponding 8...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_slli_si64(__m64 __m, int __count)
Left-shifts the first parameter, which is a 64-bit integer, by the number of bits specified by the se...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_adds_pi16(__m64 __m1, __m64 __m2)
Adds each 16-bit signed integer element of the first 64-bit integer vector of [4 x i16] to the corres...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_set_pi8(char __b7, char __b6, char __b5, char __b4, char __b3, char __b2, char __b1, char __b0)
Constructs a 64-bit integer vector initialized with the specified 8-bit integer values.
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_unpacklo_pi8(__m64 __m1, __m64 __m2)
Unpacks the lower 32 bits from two 64-bit integer vectors of [8 x i8] and interleaves them into a 64-...
#define __DEFAULT_FN_ATTRS
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_cmpgt_pi8(__m64 __m1, __m64 __m2)
Compares the 8-bit integer elements of two 64-bit integer vectors of [8 x i8] to determine if the ele...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_set1_pi32(int __i)
Constructs a 64-bit integer vector of [2 x i32], with each of the 32-bit integer vector elements set ...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_adds_pu8(__m64 __m1, __m64 __m2)
Adds each 8-bit unsigned integer element of the first 64-bit integer vector of [8 x i8] to the corres...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_adds_pi8(__m64 __m1, __m64 __m2)
Adds each 8-bit signed integer element of the first 64-bit integer vector of [8 x i8] to the correspo...
static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_unpacklo_pi16(__m64 __m1, __m64 __m2)
Unpacks the lower 32 bits from two 64-bit integer vectors of [4 x i16] and interleaves them into a 64...