MOVSLDUP (Intel x86/64 assembly instruction)

모두의 코드
MOVSLDUP (Intel x86/64 assembly instruction)

작성일 : 2020-09-01 이 글은 460 번 읽혔습니다.

MOVSLDUP

Replicate Single FP Values

참고 사항

아래 표를 해석하는 방법은 x86-64 명령어 레퍼런스 읽는 법 글을 참조하시기 바랍니다.

Opcode/ Instruction	Op / En	64/32 bit Mode Support	CPUID Feature Flag	Description
`F3 0F 12 /r` MOVSLDUP xmm1 xmm2/m128	A	V/V	SSE3	Move even index single-precision floating-point values from xmm2/mem and duplicate each element into xmm1.
`VEX.128.F3.0F.WIG 12 /r` VMOVSLDUP xmm1 xmm2/m128	RM	V/V	AVX	Move even index single-precision floating-point values from xmm2/mem and duplicate each element into xmm1.
`VEX.256.F3.0F.WIG 12 /r` VMOVSLDUP ymm1 ymm2/m256	RM	V/V	AVX	Move even index single-precision floating-point values from ymm2/mem and duplicate each element into ymm1.
`EVEX.128.F3.0F.W0 12 /r` VMOVSLDUP xmm1 {k1}{z} xmm2/m128	FVM	V/V	AVX512VL AVX512F	Move even index single-precision floating-point values from xmm2/m128 and duplicate each element into xmm1 under writemask.
`EVEX.256.F3.0F.W0 12 /r` VMOVSLDUP ymm1 {k1}{z} ymm2/m256	FVM	V/V	AVX512VL AVX512F	Move even index single-precision floating-point values from ymm2/m256 and duplicate each element into ymm1 under writemask.
`EVEX.512.F3.0F.W0 12 /r` VMOVSLDUP zmm1 {k1}{z} zmm2/m512	FVM	V/V	AVX512F	Move even index single-precision floating-point values from zmm2/m512 and duplicate each element into zmm1 under writemask.

Instruction Operand Encoding

Op/En	Operand 1	Operand 2	Operand 3	Operand 4
RM	ModRM:reg (w)	ModRM:r/m (r)	NA	NA
FVM	ModRM:reg (w)	ModRM:r/m (r)	NA	NA

Description

Duplicates even-indexed single-precision floating-point values from the source operand (the second operand). See Figure 4-4. The source operand is an XMM, YMM or ZMM register or 128, 256 or 512-bit memory location and the destination operand is an XMM, YMM or ZMM register.

128-bit Legacy SSE version: Bits (MAXVL-1:128) of the corresponding destination register remain unchanged.

VEX.128 encoded version: Bits (MAXVL-1:128) of the destination register are zeroed.

VEX.256 encoded version: Bits (MAXVL-1:256) of the destination register are zeroed.

EVEX encoded version: The destination operand is updated at 32-bit granularity according to the writemask.

Note: VEX.vvvv and EVEX.vvvv are reserved and must be 1111b otherwise instructions will #UD.

Figure 4-4. MOVSLDUP Operation

Operation

VMOVSLDUP (EVEX encoded versions)

(KL, VL) = (4, 128), (8, 256), (16, 512)
TMP_SRC[31:0] <-  SRC[31:0]
TMP_SRC[63:32] <-  SRC[31:0]
TMP_SRC[95:64] <-  SRC[95:64]
TMP_SRC[127:96] <-  SRC[95:64]
IF VL >= 256
    TMP_SRC[159:128] <-  SRC[159:128]
    TMP_SRC[191:160] <-  SRC[159:128]
    TMP_SRC[223:192] <-  SRC[223:192]
    TMP_SRC[255:224] <-  SRC[223:192]
FI;
IF VL >= 512
    TMP_SRC[287:256] <-  SRC[287:256]
    TMP_SRC[319:288] <-  SRC[287:256]
    TMP_SRC[351:320] <-  SRC[351:320]
    TMP_SRC[383:352] <-  SRC[351:320]
    TMP_SRC[415:384] <-  SRC[415:384]
    TMP_SRC[447:416] <-  SRC[415:384]
    TMP_SRC[479:448] <-  SRC[479:448]
    TMP_SRC[511:480] <-  SRC[479:448]
FI;
FOR j <-  0 TO KL-1
    i <-  j * 32
    IF k1[j] OR *no writemask*
          THEN DEST[i+31:i] <-  TMP_SRC[i+31:i]
          ELSE 
                IF *merging-masking* ; merging-masking
                      THEN *DEST[i+31:i] remains unchanged*
                      ELSE  ; zeroing-masking
                            DEST[i+31:i] <-  0 
                FI
    FI;
ENDFOR
DEST[MAX_VL-1:VL] <-  0

VMOVSLDUP (VEX.256 encoded version)

DEST[31:0] <-  SRC[31:0]
DEST[63:32] <-  SRC[31:0]
DEST[95:64] <-  SRC[95:64]
DEST[127:96] <-  SRC[95:64]
DEST[159:128] <-  SRC[159:128]
DEST[191:160] <-  SRC[159:128]
DEST[223:192] <-  SRC[223:192]
DEST[255:224] <-  SRC[223:192]
DEST[MAX_VL-1:256] <-  0

VMOVSLDUP (VEX.128 encoded version)

DEST[31:0] <-  SRC[31:0]
DEST[63:32] <-  SRC[31:0]
DEST[95:64] <-  SRC[95:64]
DEST[127:96] <-  SRC[95:64]
DEST[MAX_VL-1:128] <-  0

MOVSLDUP (128-bit Legacy SSE version)

DEST[31:0] <- SRC[31:0]
DEST[63:32] <- SRC[31:0]
DEST[95:64] <- SRC[95:64]
DEST[127:96] <- SRC[95:64]
DEST[MAX_VL-1:128] (Unmodified)

Intel C/C++ Compiler Intrinsic Equivalent

VMOVSLDUP __m512 _mm512_moveldup_ps(__m512 a);
VMOVSLDUP __m512 _mm512_mask_moveldup_ps(__m512 s, __mmask16 k, __m512 a);
VMOVSLDUP __m512 _mm512_maskz_moveldup_ps(__mmask16 k, __m512 a);
VMOVSLDUP __m256 _mm256_mask_moveldup_ps(__m256 s, __mmask8 k, __m256 a);
VMOVSLDUP __m256 _mm256_maskz_moveldup_ps(__mmask8 k, __m256 a);
VMOVSLDUP __m128 _mm_mask_moveldup_ps(__m128 s, __mmask8 k, __m128 a);
VMOVSLDUP __m128 _mm_maskz_moveldup_ps(__mmask8 k, __m128 a);
VMOVSLDUP __m256 _mm256_moveldup_ps(__m256 a);
VMOVSLDUP __m128 _mm_moveldup_ps(__m128 a);

SIMD Floating-Point Exceptions

None

Other Exceptions

Non-EVEX-encoded instruction, see Exceptions Type 4;

EVEX-encoded instruction, see Exceptions Type E4NF.nb.

#UD If EVEX.vvvv != 1111B or VEX.vvvv != 1111B.

첫 댓글을 달아주세요!

강좌에 관련 없이 궁금한 내용은 여기를 사용해주세요

또는 직접 입력하세요 (댓글 수정시 비밀번호가 필요합니다)

댓글을 불러오는 중입니다..

모두의 코드 MOVSLDUP (Intel x86/64 assembly instruction)