MOVAPD

Move Aligned Packed Double Precision Floating-Point Values

stableVMJITAOTinstruction

Encodings

Operand encoding

Each mode is a value of the Op/En column above. It says which field of the encoded instruction carries each operand, in the order they are written, and whether the instruction reads it, writes it or both.

Measured cost

Description#

Moves 2, 4 or 8 double precision floating-point values from the source operand (second operand) to the destination operand (first operand). This instruction can be used to load an XMM, YMM or ZMM register from an 128-bit, 256-bit or 512-bit memory location, to store the contents of an XMM, YMM or ZMM register into a 128-bit, 256-bit or 512-bit memory location, or to move data between two XMM, two YMM or two ZMM registers.

When the source or destination operand is a memory operand, the operand must be aligned on a 16-byte (128-bit versions), 32-byte (256-bit version) or 64-byte (EVEX.512 encoded version) boundary or a general-protection

exception (#GP) will be generated. For EVEX encoded versions, the operand must be aligned to the size of the memory operand. To move double precision floating-point values to and from unaligned memory locations, use the VMOVUPD instruction.

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

EVEX.512 encoded version:

Moves 512 bits of packed double precision floating-point values from the source operand (second operand) to the destination operand (first operand). This instruction can be used to load a ZMM register from a 512-bit float64 memory location, to store the contents of a ZMM register into a 512-bit float64 memory location, or to move data between two ZMM registers. When the source or destination operand is a memory operand, the operand must be aligned on a 64-byte boundary or a general-protection exception (#GP) will be generated. To move single precision floating-point values to and from unaligned memory locations, use the VMOVUPD instruction.

VEX.256 and EVEX.256 encoded versions:

Moves 256 bits of packed double precision floating-point values from the source operand (second operand) to the destination operand (first operand). This instruction can be used to load a YMM register from a 256-bit memory location, to store the contents of a YMM register into a 256-bit memory location, or to move data between two YMM registers. When the source or destination operand is a memory operand, the operand must be aligned on a 32-byte boundary or a general-protection exception (#GP) will be generated. To move double precision floating-point values to and from unaligned memory locations, use the VMOVUPD instruction.

128-bit versions:

Moves 128 bits of packed double precision floating-point values from the source operand (second operand) to the destination operand (first operand). This instruction can be used to load an XMM register from a 128-bit memory location, to store the contents of an XMM register into a 128-bit memory location, or to move data between two XMM registers. When the source or destination operand is a memory operand, the operand must be aligned on a 16-byte boundary or a general-protection exception (#GP) will be generated. To move single precision floatingpoint values to and from unaligned memory locations, use the VMOVUPD instruction.

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

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

Operation#

VMOVAPD (EVEX Encoded Versions, Register-Copy Form)

(KL, VL) = (2, 128), (4, 256), (8, 512)

FOR j := 0 TO KL-1

i := j * 64

IF k1[j] OR *no writemask*

     THEN DEST[i+63:i] := SRC[i+63:i]

     ELSE

             IF *merging-masking*        ; merging-masking

                 THEN *DEST[i+63:i] remains unchanged*

                 ELSE DEST[i+63:i] := 0  ; zeroing-masking

             FI

FI;

ENDFOR

DEST[MAXVL-1:VL] := 0


VMOVAPD (EVEX Encoded Versions, Store-Form)      ; merging-masking
(KL, VL) = (2, 128), (4, 256), (8, 512)
FOR j := 0 TO KL-1

    i := j * 64
    IF k1[j] OR *no writemask*

          THEN DEST[i+63:i] := SRC[i+63:i]
          ELSE
          ELSE *DEST[i+63:i] remains unchanged*

    FI;
ENDFOR;

VMOVAPD (EVEX Encoded Versions, Load-Form)

(KL, VL) = (2, 128), (4, 256), (8, 512)

FOR j := 0 TO KL-1

i := j * 64

IF k1[j] OR *no writemask*

     THEN DEST[i+63:i] := SRC[i+63:i]

     ELSE

             IF *merging-masking*           ; merging-masking

                 THEN *DEST[i+63:i] remains unchanged*

                 ELSE DEST[i+63:i] := 0     ; zeroing-masking

             FI

FI;

ENDFOR

DEST[MAXVL-1:VL] := 0

VMOVAPD (VEX.256 Encoded Version, Load - and Register Copy)
DEST[255:0] := SRC[255:0]
DEST[MAXVL-1:256] := 0

VMOVAPD (VEX.256 Encoded Version, Store-Form)
DEST[255:0] := SRC[255:0]

VMOVAPD (VEX.128 Encoded Version, Load - and Register Copy)
DEST[127:0] := SRC[127:0]
DEST[MAXVL-1:128] := 0

MOVAPD (128-bit Load- and Register-Copy- Form Legacy SSE Version)
DEST[127:0] := SRC[127:0]
DEST[MAXVL-1:128] (Unmodified)

(V)MOVAPD (128-bit Store-Form Version)
DEST[127:0] := SRC[127:0]

Intel C/C++ compiler intrinsics#

VMOVAPD __m512d _mm512_load_pd( void * m);
VMOVAPD __m512d _mm512_mask_load_pd(__m512d s, __mmask8 k, void * m);
VMOVAPD __m512d _mm512_maskz_load_pd( __mmask8 k, void * m);
VMOVAPD void _mm512_store_pd( void * d, __m512d a);
VMOVAPD void _mm512_mask_store_pd( void * d, __mmask8 k, __m512d a);
VMOVAPD __m256d _mm256_mask_load_pd(__m256d s, __mmask8 k, void * m);
VMOVAPD __m256d _mm256_maskz_load_pd( __mmask8 k, void * m);
VMOVAPD void _mm256_mask_store_pd( void * d, __mmask8 k, __m256d a);
VMOVAPD __m128d _mm_mask_load_pd(__m128d s, __mmask8 k, void * m);
VMOVAPD __m128d _mm_maskz_load_pd( __mmask8 k, void * m);
VMOVAPD void _mm_mask_store_pd( void * d, __mmask8 k, __m128d a);
MOVAPD __m256d _mm256_load_pd (double * p);
MOVAPD void _mm256_store_pd(double * p, __m256d a);
MOVAPD __m128d _mm_load_pd (double * p);
MOVAPD void _mm_store_pd(double * p, __m128d a);

SIMD Floating-Point Exceptions#

None.

Other Exceptions#

Non-EVEX-encoded instruction, see Exceptions Type1.SSE2 in Table 2-18, "Type 1 Class Exception Conditions."

EVEX-encoded instruction, see Table 2-46, "Type E1 Class Exception Conditions."

Additionally:

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

Sources

• This instruction in the Intel manual
• Latency and throughput per microarchitecture