Back to opcode table

VCVTPH2PS—Convert 16-bit FP values to Single-Precision FP values

Opcode/Instruction Op /En 64/32 bit Mode Support CPUID Feature Flag Description

VEX.128.66.0F38.W0 1313 /r

VCVTPH2PS xmm1, xmm2/m64

RM V/V F16C Convert four packed half precision (16-bit) floating-point values in xmm2/m64 to packed single-precision floating-point value in xmm1.

VEX.256.66.0F38.W0 1313 /r

VCVTPH2PS ymm1, xmm2/m128

RM V/V F16C Convert eight packed half precision (16-bit) floating-point values in xmm2/m128 to packed single-precision floating-point value in ymm1.

EVEX.128.66.0F38.W0 1313 /r

VCVTPH2PS xmm1 {k1}{z}, xmm2/m64

HVM V/V

AVX512VL

AVX512F

Convert four packed half precision (16-bit) floating-point values in xmm2/m64 to packed single-precision floating-point values in xmm1.

EVEX.256.66.0F38.W0 1313 /r

VCVTPH2PS ymm1 {k1}{z}, xmm2/m128

HVM V/V

AVX512VL

AVX512F

Convert eight packed half precision (16-bit) floating-point values in xmm2/m128 to packed single-precision floating-point values in ymm1.

EVEX.512.66.0F38.W0 1313 /r

VCVTPH2PS zmm1 {k1}{z}, ymm2/m256 {sae}

HVM V/V AVX512F Convert sixteen packed half precision (16-bit) floating-point values in ymm2/m256 to packed single-precision floating-point values in zmm1.

Instruction Operand Encoding

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

Description

Converts packed half precision (16-bits) floating-point values in the low-order bits of the source operand (the second operand) to packed single-precision floating-point values and writes the converted values into the destina-tion operand (the first operand).

If case of a denormal operand, the correct normal result is returned. MXCSR.DAZ is ignored and is treated as if it 0. No denormal exception is reported on MXCSR.

VEX.128 version: The source operand is a XMM register or 64-bit memory location. The destination operand is a XMM register. The upper bits (MAX_VL-1:128) of the corresponding destination register are zeroed.

VEX.256 version: The source operand is a XMM register or 128-bit memory location. The destination operand is a YMM register. Bits (MAX_VL-1:256) of the corresponding destination register are zeroed.

EVEX encoded versions: The source operand is a YMM/XMM/XMM (low 64-bits) register or a 256/128/64-bit memory location. The destination operand is a ZMM/YMM/XMM register conditionally updated with writemask k1.

The diagram below illustrates how data is converted from four packed half precision (in 64 bits) to four single preci-sion (in 128 bits) FP values.

Note: VEX.vvvv and EVEX.vvvv are reserved (must be 1111b).

Operation

vCvt_h2s(SRC1[15:0]) { RETURN Cvt_Half_Precision_To_Single_Precision(SRC1[15:0]); }
VCVTPH2PS (EVEX encoded versions)
(KL, VL) = (4, 128), (8, 256), (16, 512)
FOR j (cid:197) 0 TO KL-1
    i (cid:197) j * 32
    k (cid:197) j * 16
    IF k1[j] OR *no writemask*
        THEN DEST[i+31:i] (cid:197)
        vCvt_h2s(SRC[k+15:k])
        ELSE
        IF *merging-masking*
            ; merging-masking
            THEN *DEST[i+31:i] remains unchanged*
            ELSE
            ; zeroing-masking
            DEST[i+31:i] (cid:197) 0
        FI
    FI;
ENDFOR
DEST[MAX_VL-1:VL] (cid:197) 0
VCVTPH2PS (VEX.256 encoded version)
DEST[31:0] (cid:197)vCvt_h2s(SRC1[15:0]);
DEST[63:32] (cid:197)vCvt_h2s(SRC1[31:16]);
DEST[95:64] (cid:197)vCvt_h2s(SRC1[47:32]);
DEST[127:96] (cid:197)vCvt_h2s(SRC1[63:48]);
DEST[159:128] (cid:197)vCvt_h2s(SRC1[79:64]);
DEST[191:160] (cid:197)vCvt_h2s(SRC1[95:80]);
DEST[223:192] (cid:197)vCvt_h2s(SRC1[111:96]); DEST[255:224] (cid:197)vCvt_h2s(SRC1[127:112]);
DEST[MAX_VL-1:256] (cid:197) 0
VCVTPH2PS (VEX.128 encoded version)
DEST[31:0] (cid:197)vCvt_h2s(SRC1[15:0]);
DEST[63:32] (cid:197)vCvt_h2s(SRC1[31:16]);
DEST[95:64] (cid:197)vCvt_h2s(SRC1[47:32]); DEST[127:96] (cid:197)vCvt_h2s(SRC1[63:48]);
DEST[MAX_VL-1:128] (cid:197) 0

Flags Affected

None

Intel C/C++ Compiler Intrinsic Equivalent

VCVTPH2PS __m512 _mm512_cvtph_ps( __m256i a);
VCVTPH2PS __m512 _mm512_mask_cvtph_ps(__m512 s, __mmask16 k, __m256i a);
VCVTPH2PS __m512 _mm512_maskz_cvtph_ps(__mmask16 k, __m256i a);
VCVTPH2PS __m512 _mm512_cvt_roundph_ps( __m256i a, int sae);
VCVTPH2PS __m512 _mm512_mask_cvt_roundph_ps(__m512 s, __mmask16 k, __m256i a, int sae);
VCVTPH2PS __m512 _mm512_maskz_cvt_roundph_ps( __mmask16 k, __m256i a, int sae);
VCVTPH2PS __m256 _mm256_mask_cvtph_ps(__m256 s, __mmask8 k, __m128i a);
VCVTPH2PS __m256 _mm256_maskz_cvtph_ps(__mmask8 k, __m128i a);
VCVTPH2PS __m128 _mm_mask_cvtph_ps(__m128 s, __mmask8 k, __m128i a);
VCVTPH2PS __m128 _mm_maskz_cvtph_ps(__mmask8 k, __m128i a);
VCVTPH2PS __m128 _mm_cvtph_ps ( __m128i m1);
VCVTPH2PS __m256 _mm256_cvtph_ps ( __m128i m1)

SIMD Floating-Point Exceptions

Invalid

Other Exceptions

VEX-encoded instructions, see Exceptions Type 11 (do not report #AC);
EVEX-encoded instructions, see Exceptions Type E11.
If VEX.W=1.
If VEX.vvvv != 1111B or EVEX.vvvv != 1111B.