Avx512DQ 类
定义
重要
一些信息与预发行产品相关,相应产品在发行之前可能会进行重大修改。 对于此处提供的信息,Microsoft 不作任何明示或暗示的担保。
重要
此 API 不符合 CLS。
提供通过内部函数访问 X86 AVX512DQ硬件指令。
public ref class Avx512DQ abstract : System::Runtime::Intrinsics::X86::Avx512F
[System.CLSCompliant(false)]
public abstract class Avx512DQ : System.Runtime.Intrinsics.X86.Avx512F
[<System.CLSCompliant(false)>]
type Avx512DQ = class
inherit Avx512F
Public MustInherit Class Avx512DQ
Inherits Avx512F
- 继承
- 属性
属性
IsSupported |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
方法
And(Vector512<Double>, Vector512<Double>) |
__m512d _mm512_and_pd (__m512d a, __m512d b) VANDPD zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst |
And(Vector512<Single>, Vector512<Single>) |
__m512 _mm512_and_ps (__m512 a、 __m512 b) VANDPS zmm1 {k1}{z}, zmm2, zmm3/m512/m32bcst |
AndNot(Vector512<Double>, Vector512<Double>) |
__m512d _mm512_andnot_pd (__m512d a, __m512d b) VANDNPD zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst |
AndNot(Vector512<Single>, Vector512<Single>) |
__m512 _mm512_andnot_ps (__m512 a, __m512 b) VANDNPS zmm1 {k1}{z}, zmm2, zmm3/m512/m32bcst |
BroadcastPairScalarToVector512(Vector128<Int32>) |
__m512i _mm512_broadcast_i32x2 (__m128i) VBROADCASTI32x2 zmm1 {k1}{z}, xmm2/m64 |
BroadcastPairScalarToVector512(Vector128<Single>) |
__m512 _mm512_broadcast_f32x2 (__m128) VBROADCASTF32x2 zmm1 {k1}{z}, xmm2/m64 |
BroadcastPairScalarToVector512(Vector128<UInt32>) |
__m512i _mm512_broadcast_i32x2 (__m128i) VBROADCASTI32x2 zmm1 {k1}{z}, xmm2/m64 |
BroadcastVector128ToVector512(Double*) |
__m512d _mm512_broadcast_f64x2 (__m128d const * mem_addr) VBROADCASTF64x2 zmm1 {k1}{z}, m128 |
BroadcastVector128ToVector512(Int64*) |
__m512i _mm512_broadcast_i64x2 (__m128i const * mem_addr) VBROADCASTI64x2 zmm1 {k1}{z}, m128 |
BroadcastVector128ToVector512(UInt64*) |
__m512i _mm512_broadcast_i64x2 (__m128i const * mem_addr) VBROADCASTI64x2 zmm1 {k1}{z}, m128 |
BroadcastVector256ToVector512(Int32*) |
__m512i _mm512_broadcast_i32x8 (__m256i const * mem_addr) VBROADCASTI32x8 zmm1 {k1}{z}, m256 |
BroadcastVector256ToVector512(Single*) |
__m512 _mm512_broadcast_f32x8 (__m256 const * mem_addr) VBROADCASTF32x8 zmm1 {k1}{z}, m256 |
BroadcastVector256ToVector512(UInt32*) |
__m512i _mm512_broadcast_i32x8 (__m256i const * mem_addr) VBROADCASTI32x8 zmm1 {k1}{z}, m256 |
ConvertToVector256Single(Vector512<Int64>) |
__m512 _mm512_cvtepi64_ps (__m512i) VCVTQQ2PS ymm1 {k1}{z}, zmm2/m512/m64bcst |
ConvertToVector256Single(Vector512<Int64>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector256Single(Vector512<UInt64>) |
__m512 _mm512_cvtepu64_ps (__m512i) VCVTUQQ2PS ymm1 {k1}{z}, zmm2/m512/m64bcst |
ConvertToVector256Single(Vector512<UInt64>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector512Double(Vector512<Int64>) |
__m512d _mm512_cvtepi64_pd (__m512i) VCVTQQ2PD zmm1 {k1}{z}, zmm2/m512/m64bcst |
ConvertToVector512Double(Vector512<Int64>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector512Double(Vector512<UInt64>) |
__m512d _mm512_cvtepu64_pd (__m512i) VCVTUQQ2PD zmm1 {k1}{z}, zmm2/m512/m64bcst |
ConvertToVector512Double(Vector512<UInt64>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector512Int64(Vector256<Single>) |
__m512i _mm512_cvtps_epi64 (__m512) VCVTPS2QQ zmm1 {k1}{z}, ymm2/m256/m32bcst{er} |
ConvertToVector512Int64(Vector256<Single>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector512Int64(Vector512<Double>) |
__m512i _mm512_cvtpd_epi64 (__m512d) VCVTPD2QQ zmm1 {k1}{z}, zmm2/m512/m64bcst{er} |
ConvertToVector512Int64(Vector512<Double>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector512Int64WithTruncation(Vector256<Single>) |
__m512i _mm512_cvttps_epi64 (__m512) VCVTTPS2QQ zmm1 {k1}{z}, ymm2/m256/m32bcst{er} |
ConvertToVector512Int64WithTruncation(Vector512<Double>) |
__m512i _mm512_cvttpd_epi64 (__m512) VCVTTPD2QQ zmm1 {k1}{z}, zmm2/m512/m64bcst{sae} |
ConvertToVector512UInt64(Vector256<Single>) |
__m512i _mm512_cvtps_epu64 (__m512) VCVTPS2UQQ zmm1 {k1}{z}, ymm2/m256/m32bcst{er} |
ConvertToVector512UInt64(Vector256<Single>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector512UInt64(Vector512<Double>) |
__m512i _mm512_cvtpd_epu64 (__m512d) VCVTPD2UQQ zmm1 {k1}{z}, zmm2/m512/m64bcst{er} |
ConvertToVector512UInt64(Vector512<Double>, FloatRoundingMode) |
提供通过内部函数访问 X86 AVX512DQ硬件指令。 |
ConvertToVector512UInt64WithTruncation(Vector256<Single>) |
__m512i _mm512_cvttps_epu64 (__m512) VCVTTPS2UQQ zmm1 {k1}{z}, ymm2/m256/m32bcst{er} |
ConvertToVector512UInt64WithTruncation(Vector512<Double>) |
__m512i _mm512_cvttpd_epu64 (__m512d) VCVTTPD2UQQ zmm1 {k1}{z}, zmm2/m512/m64bcst{er} |
Equals(Object) |
确定指定对象是否等于当前对象。 (继承自 Object) |
ExtractVector128(Vector512<Double>, Byte) |
__m128d _mm512_extractf64x2_pd (__m512d a, const int imm8) VEXTRACTF64x2 xmm1/m128 {k1}{z}, zmm2, imm8 |
ExtractVector128(Vector512<Int64>, Byte) |
__m128i _mm512_extracti64x2_epi64 (__m512i a, const int imm8) VEXTRACTI64x2 xmm1/m128 {k1}{z}, zmm2, imm8 |
ExtractVector128(Vector512<UInt64>, Byte) |
__m128i _mm512_extracti64x2_epi64 (__m512i a, const int imm8) VEXTRACTI64x2 xmm1/m128 {k1}{z}, zmm2, imm8 |
ExtractVector256(Vector512<Int32>, Byte) |
__m256i _mm512_extracti32x8_epi32 (__m512i a, const int imm8) VEXTRACTI32x8 ymm1/m256 {k1}{z}, zmm2, imm8 |
ExtractVector256(Vector512<Single>, Byte) |
__m256 _mm512_extractf32x8_ps (__m512 a, const int imm8) VEXTRACTF32x8 ymm1/m256 {k1}{z}, zmm2, imm8 |
ExtractVector256(Vector512<UInt32>, Byte) |
__m256i _mm512_extracti32x8_epi32 (__m512i a, const int imm8) VEXTRACTI32x8 ymm1/m256 {k1}{z}, zmm2, imm8 |
GetHashCode() |
作为默认哈希函数。 (继承自 Object) |
GetType() |
获取当前实例的 Type。 (继承自 Object) |
InsertVector128(Vector512<Double>, Vector128<Double>, Byte) |
__m512d _mm512_insertf64x2_pd (__m512d a, __m128d b, int imm8) VINSERTF64x2 zmm1 {k1}{z}, zmm2, xmm3/m128, imm8 |
InsertVector128(Vector512<Int64>, Vector128<Int64>, Byte) |
__m512i _mm512_inserti64x2_si512 (__m512i a, __m128i b, const int imm8) VINSERTI64x2 zmm1 {k1}{z}, zmm2, xmm3/m128, imm8 |
InsertVector128(Vector512<UInt64>, Vector128<UInt64>, Byte) |
__m512i _mm512_inserti64x2_si512 (__m512i a, __m128i b, const int imm8) VINSERTI64x2 zmm1 {k1}{z}, zmm2, xmm3/m128, imm8 |
InsertVector256(Vector512<Int32>, Vector256<Int32>, Byte) |
__m512i _mm512_inserti32x8_si512 (__m512i a, __m256i b, const int imm8) VINSERTI32x8 zmm1 {k1}{z}, zmm2, xmm3/m256, imm8 |
InsertVector256(Vector512<Single>, Vector256<Single>, Byte) |
__m512 _mm512_insertf32x8_ps (__m512 a, __m256 b, int imm8) VINSERTF32x8 zmm1 {k1}{z}, zmm2, xmm3/m256, imm8 |
InsertVector256(Vector512<UInt32>, Vector256<UInt32>, Byte) |
__m512i _mm512_inserti32x8_si512 (__m512i a, __m256i b, const int imm8) VINSERTI32x8 zmm1 {k1}{z}, zmm2, xmm3/m256, imm8 |
MemberwiseClone() |
创建当前 Object 的浅表副本。 (继承自 Object) |
MultiplyLow(Vector512<Int64>, Vector512<Int64>) |
__m512i _mm512_mullo_epi64 (__m512i a, __m512i b) VPMULLQ zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst |
MultiplyLow(Vector512<UInt64>, Vector512<UInt64>) |
__m512i _mm512_mullo_epi64 (__m512i a, __m512i b) VPMULLQ zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst |
Or(Vector512<Double>, Vector512<Double>) |
__m512d _mm512_or_pd (__m512d a, __m512d b) VORPD zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst |
Or(Vector512<Single>, Vector512<Single>) |
__m512 _mm512_or_ps (__m512 a, __m512 b) VORPS zmm1 {k1}{z}, zmm2, zmm3/m512/m32bcst |
Range(Vector512<Double>, Vector512<Double>, Byte) |
__m512d _mm512_range_pd (__m512d a, __m512d b, int imm) ;VRANGEPD zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst{sae}, imm8 |
Range(Vector512<Single>, Vector512<Single>, Byte) |
__m512 _mm512_range_ps (__m512 a, __m512 b, int imm) ;VRANGEPS zmm1 {k1}{z}, zmm2, zmm3/m512/m32bcst{sae}, imm8 |
RangeScalar(Vector128<Double>, Vector128<Double>, Byte) |
__m128d _mm_range_sd (__m128d a, __m128d b, int imm) ;VRANGESD xmm1 {k1}{z}, xmm2, xmm3/m64{sae}, imm8 |
RangeScalar(Vector128<Single>, Vector128<Single>, Byte) |
__m128 _mm_range_ss (__m128 a, __m128 b, int imm) ;VRANGESS xmm1 {k1}{z}, xmm2, xmm3/m32{sae}, imm8 |
Reduce(Vector512<Double>, Byte) |
__m512d _mm512_reduce_pd (__m512d a, int imm) ;VREDUCEPD zmm1 {k1}{z}, zmm2/m512/m64bcst{sae}, imm8 |
Reduce(Vector512<Single>, Byte) |
__m512 _mm512_reduce_ps (__m512 a, int imm) ;VREDUCEPS zmm1 {k1}{z}, zmm2/m512/m32bcst{sae}, imm8 |
ReduceScalar(Vector128<Double>, Byte) |
__m128d _mm_reduce_sd (__m128d a, int imm) ;VREDUCESD xmm1 {k1}{z}, xmm2, xmm3/m64{sae}, imm8 |
ReduceScalar(Vector128<Double>, Vector128<Double>, Byte) |
__m128d _mm_reduce_sd (__m128d a, __m128d b, int imm) ;VREDUCESD xmm1 {k1}{z}, xmm2, xmm3/m64{sae}, imm8 上述本机签名不存在。 我们提供此额外的重载为的是与其他标量 API 保持一致。 |
ReduceScalar(Vector128<Single>, Byte) |
__m128 _mm_reduce_ss (__m128 a, int imm) ;VREDUCESS xmm1 {k1}{z}, xmm2, xmm3/m32{sae}, imm8 |
ReduceScalar(Vector128<Single>, Vector128<Single>, Byte) |
__m128 _mm_reduce_ss (__m128 a, __m128 b, int imm) ;VREDUCESS xmm1 {k1}{z}, xmm2, xmm3/m32{sae}, imm8 上述本机签名不存在。 我们提供此额外的重载为的是与其他标量 API 保持一致。 |
ToString() |
返回表示当前对象的字符串。 (继承自 Object) |
Xor(Vector512<Double>, Vector512<Double>) |
__m512d _mm512_xor_pd (__m512d a, __m512d b) VXORPD zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst |
Xor(Vector512<Single>, Vector512<Single>) |
__m512 _mm512_xor_ps (__m512 a, __m512 b) VXORPS zmm1 {k1}{z}, zmm2, zmm3/m512/m32bcst |