SUBSD — Subtract Scalar Double-Precision Floating-Point Value

Opcode/Instruction Op / En 64/32 bit Mode Support CPUID Feature Flag Description
F2 0F 5C /r SUBSD xmm1, xmm2/m64 A V/V SSE2 Subtract the low double-precision floating-point value in xmm2/m64 from xmm1 and store the result in xmm1.
VEX.LIG.F2.0F.WIG 5C /r VSUBSD xmm1,xmm2, xmm3/m64 B V/V AVX Subtract the low double-precision floating-point value in xmm3/m64 from xmm2 and store the result in xmm1.
EVEX.LIG.F2.0F.W1 5C /r VSUBSD xmm1 {k1}{z}, xmm2, xmm3/m64{er} C V/V AVX512F Subtract the low double-precision floating-point value in xmm3/m64 from xmm2 and store the result in xmm1 under writemask k1.

Instruction Operand Encoding

Op/En Tuple Type Operand 1 Operand 2 Operand 3 Operand 4
A NA ModRM:reg (r, w) ModRM:r/m (r) NA NA
B NA ModRM:reg (w) VEX.vvvv (r) ModRM:r/m (r) NA
C Tuple1 Scalar ModRM:reg (w) EVEX.vvvv (r) ModRM:r/m (r) NA

Description

Subtract the low double-precision floating-point value in the second source operand from the first source operand and stores the double-precision floating-point result in the low quadword of the destination operand.

The second source operand can be an XMM register or a 64-bit memory location. The first source and destination operands are XMM registers.

128-bit Legacy SSE version: The destination and first source operand are the same. Bits (MAXVL-1:64) of the corresponding destination register remain unchanged.

VEX.128 and EVEX encoded versions: Bits (127:64) of the XMM register destination are copied from corresponding bits in the first source operand. Bits (MAXVL-1:128) of the destination register are zeroed.

EVEX encoded version: The low quadword element of the destination operand is updated according to the writemask.

Software should ensure VSUBSD is encoded with VEX.L=0. Encoding VSUBSD with VEX.L=1 may encounter unpredictable behavior across different processor generations.

Operation

VSUBSD (EVEX encoded version)

IF (SRC2 *is register*) AND (EVEX.b = 1)
    THEN
        SET_RM(EVEX.RC);
    ELSE
        SET_RM(MXCSR.RM);
FI;
IF k1[0] or *no writemask*
    THEN DEST[63:0]←SRC1[63:0] - SRC2[63:0]
    ELSE
        IF *merging-masking* ; merging-masking
            THEN *DEST[63:0] remains unchanged*
            ELSE ; zeroing-masking
                THEN DEST[63:0]←0
        FI;
FI;
DEST[127:64] ← SRC1[127:64]
DEST[MAXVL-1:128] ← 0

VSUBSD (VEX.128 encoded version)

DEST[63:0]←SRC1[63:0] - SRC2[63:0]
DEST[127:64] ←SRC1[127:64]
DEST[MAXVL-1:128] ←0

SUBSD (128-bit Legacy SSE version)

DEST[63:0]←DEST[63:0] - SRC[63:0]
DEST[MAXVL-1:64] (Unmodified)

Intel C/C++ Compiler Intrinsic Equivalent

VSUBSD __m128d _mm_mask_sub_sd (__m128d s, __mmask8 k, __m128d a, __m128d b);
VSUBSD __m128d _mm_maskz_sub_sd (__mmask8 k, __m128d a, __m128d b);
VSUBSD __m128d _mm_sub_round_sd (__m128d a, __m128d b, int);
VSUBSD __m128d _mm_mask_sub_round_sd (__m128d s, __mmask8 k, __m128d a, __m128d b, int);
VSUBSD __m128d _mm_maskz_sub_round_sd (__mmask8 k, __m128d a, __m128d b, int);
SUBSD __m128d _mm_sub_sd (__m128d a, __m128d b);

SIMD Floating-Point Exceptions

Overflow, Underflow, Invalid, Precision, Denormal

Other Exceptions

VEX-encoded instructions, see Exceptions Type 3.

EVEX-encoded instructions, see Exceptions Type E3.