PEXTRW—Extract Word

NOTES:

1. See note in Section 2.4, “AVX and SSE Instruction Exception Specification” in the Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 2A and Section 22.25.3, “Exception Conditions of Legacy SIMD Instructions Operating on MMX Registers” in the Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 3A.

2. In 64-bit mode, VEX.W1 is ignored for VPEXTRW (similar to legacy REX.W=1 prefix in PEXTRW).

Instruction Operand Encoding

Description

Copies the word in the source operand (second operand) specified by the count operand (third operand) to the destination operand (first operand). The source operand can be an MMX technology register or an XMM register. The destination operand can be the low word of a general-purpose register or a 16-bit memory address. The count operand is an 8-bit immediate. When specifying a word location in an MMX technology register, the 2 least-signifi-cant bits of the count operand specify the location; for an XMM register, the 3 least-significant bits specify the loca-tion. The content of the destination register above bit 16 is cleared (set to all 0s).

In 64-bit mode, using a REX prefix in the form of REX.R permits this instruction to access additional registers (XMM8-XMM15, R8-15). If the destination operand is a general-purpose register, the default operand size is 64-bits in 64-bit mode.

Note: In VEX.128 encoded versions, VEX.vvvv is reserved and must be 1111b, VEX.L must be 0, otherwise the instruction will #UD. In EVEX.128 encoded versions, EVEX.vvvv is reserved and must be 1111b, EVEX.L must be 0, otherwise the instruction will #UD. If the destination operand is a register, the default operand size in 64-bit mode for VPEXTRW is 64 bits, the bits above the least significant byte/word/dword data are filled with zeros.

Operation

IF (DEST = Mem16)
    THEN
    SEL (cid:197) COUNT[2:0];
    TEMP (cid:197) (Src >> SEL*16) AND FFFFH;
    Mem16 (cid:197) TEMP[15:0];
    ELSE IF (64-Bit Mode and destination is a general-purpose register)
    THEN
    FOR (PEXTRW instruction with 64-bit source operand)
        {
        SEL ← COUNT[1:0];
        TEMP ← (SRC >> (SEL ∗ 16)) AND FFFFH;
        r64[15:0] ← TEMP[15:0];
        r64[63:16] ← ZERO_FILL; };
    FOR (PEXTRW instruction with 128-bit source operand)
        {
        SEL ← COUNT[2:0];
        TEMP ← (SRC >> (SEL ∗ 16)) AND FFFFH;
        r64[15:0] ← TEMP[15:0];
        r64[63:16] ← ZERO_FILL; }
    ELSE
    FOR (PEXTRW instruction with 64-bit source operand)
        {
        SEL ← COUNT[1:0];
        TEMP ← (SRC >> (SEL ∗ 16)) AND FFFFH;
        r32[15:0] ← TEMP[15:0];
        r32[31:16] ← ZERO_FILL; };
    FOR (PEXTRW instruction with 128-bit source operand)
        {
        SEL ← COUNT[2:0];
        TEMP ← (SRC >> (SEL ∗ 16)) AND FFFFH;
        r32[15:0] ← TEMP[15:0];
        r32[31:16] ← ZERO_FILL; };
FI;
FI;

SRC_Offset (cid:197) Imm8[2:0]
Mem16 (cid:197) (Src >> Src_Offset*16)

IF (64-Bit Mode )
    THEN
    SRC_Offset (cid:197) Imm8[2:0]
    DEST[15:0] (cid:197) ((Src >> Src_Offset*16) AND 0FFFFh)
    DEST[63:16] (cid:197)(cid:3)ZERO_FILL;
ELSE
SRC_Offset (cid:197) Imm8[2:0]
DEST[15:0] (cid:197) ((Src >> Src_Offset*16) AND 0FFFFh)
DEST[31:16] (cid:197)(cid:3)ZERO_FILL;
FI

Intel C/C++ Compiler Intrinsic Equivalent

PEXTRW:
int _mm_extract_pi16 (__m64 a, int n)
PEXTRW:
int _mm_extract_epi16 ( __m128i a, int imm)

Flags Affected

None.

Numeric Exceptions

None.

Other Exceptions

Non-EVEX-encoded instruction, see Exceptions Type 5;