/src/qtbase/src/gui/painting/qdrawingprimitive_sse2_p.h

Source
// Copyright (C) 2016 The Qt Company Ltd.
// SPDX-License-Identifier: LicenseRef-Qt-Commercial OR LGPL-3.0-only OR GPL-2.0-only OR GPL-3.0-only
// Qt-Security score:significant reason:default

#ifndef QDRAWINGPRIMITIVE_SSE2_P_H
#define QDRAWINGPRIMITIVE_SSE2_P_H

#include <QtGui/private/qtguiglobal_p.h>
#include <private/qsimd_p.h>
#include "qdrawhelper_x86_p.h"
#include "qrgba64_p.h"

#ifdef __SSE2__

//
//  W A R N I N G
//  -------------
//
// This file is not part of the Qt API.  It exists purely as an
// implementation detail.  This header file may change from version to
// version without notice, or even be removed.
//
// We mean it.
//

QT_BEGIN_NAMESPACE

/*
 * Multiply the components of pixelVector by alphaChannel
 * Each 32bits components of alphaChannel must be in the form 0x00AA00AA
 * colorMask must have 0x00ff00ff on each 32 bits component
 * half must have the value 128 (0x80) for each 32 bits component
 */
#define BYTE_MUL_SSE2(result, pixelVector, alphaChannel, colorMask, half) \
{ \
    /* 1. separate the colors in 2 vectors so each color is on 16 bits \
       (in order to be multiplied by the alpha \
       each 32 bit of dstVectorAG are in the form 0x00AA00GG \
       each 32 bit of dstVectorRB are in the form 0x00RR00BB */\
    __m128i pixelVectorAG = _mm_srli_epi16(pixelVector, 8); \
    __m128i pixelVectorRB = _mm_and_si128(pixelVector, colorMask); \
 \
    /* 2. multiply the vectors by the alpha channel */\
    pixelVectorAG = _mm_mullo_epi16(pixelVectorAG, alphaChannel); \
    pixelVectorRB = _mm_mullo_epi16(pixelVectorRB, alphaChannel); \
 \
    /* 3. divide by 255, that's the tricky part. \
       we do it like for BYTE_MUL(), with bit shift: X/255 ~= (X + X/256 + rounding)/256 */ \
    /** so first (X + X/256 + rounding) */\
    pixelVectorRB = _mm_add_epi16(pixelVectorRB, _mm_srli_epi16(pixelVectorRB, 8)); \
    pixelVectorRB = _mm_add_epi16(pixelVectorRB, half); \
    pixelVectorAG = _mm_add_epi16(pixelVectorAG, _mm_srli_epi16(pixelVectorAG, 8)); \
    pixelVectorAG = _mm_add_epi16(pixelVectorAG, half); \
 \
    /** second divide by 256 */\
    pixelVectorRB = _mm_srli_epi16(pixelVectorRB, 8); \
    /** for AG, we could >> 8 to divide followed by << 8 to put the \
        bytes in the correct position. By masking instead, we execute \
        only one instruction */\
    pixelVectorAG = _mm_andnot_si128(colorMask, pixelVectorAG); \
 \
    /* 4. combine the 2 pairs of colors */ \
    result = _mm_or_si128(pixelVectorAG, pixelVectorRB); \
}

/*
 * Each 32bits components of alphaChannel must be in the form 0x00AA00AA
 * oneMinusAlphaChannel must be 255 - alpha for each 32 bits component
 * colorMask must have 0x00ff00ff on each 32 bits component
 * half must have the value 128 (0x80) for each 32 bits component
 */
#define INTERPOLATE_PIXEL_255_SSE2(result, srcVector, dstVector, alphaChannel, oneMinusAlphaChannel, colorMask, half) { \
    /* interpolate AG */\
    __m128i srcVectorAG = _mm_srli_epi16(srcVector, 8); \
    __m128i dstVectorAG = _mm_srli_epi16(dstVector, 8); \
    __m128i srcVectorAGalpha = _mm_mullo_epi16(srcVectorAG, alphaChannel); \
    __m128i dstVectorAGoneMinusAlphalpha = _mm_mullo_epi16(dstVectorAG, oneMinusAlphaChannel); \
    __m128i finalAG = _mm_add_epi16(srcVectorAGalpha, dstVectorAGoneMinusAlphalpha); \
    finalAG = _mm_add_epi16(finalAG, _mm_srli_epi16(finalAG, 8)); \
    finalAG = _mm_add_epi16(finalAG, half); \
    finalAG = _mm_andnot_si128(colorMask, finalAG); \
 \
    /* interpolate RB */\
    __m128i srcVectorRB = _mm_and_si128(srcVector, colorMask); \
    __m128i dstVectorRB = _mm_and_si128(dstVector, colorMask); \
    __m128i srcVectorRBalpha = _mm_mullo_epi16(srcVectorRB, alphaChannel); \
    __m128i dstVectorRBoneMinusAlphalpha = _mm_mullo_epi16(dstVectorRB, oneMinusAlphaChannel); \
    __m128i finalRB = _mm_add_epi16(srcVectorRBalpha, dstVectorRBoneMinusAlphalpha); \
    finalRB = _mm_add_epi16(finalRB, _mm_srli_epi16(finalRB, 8)); \
    finalRB = _mm_add_epi16(finalRB, half); \
    finalRB = _mm_srli_epi16(finalRB, 8); \
 \
    /* combine */\
    result = _mm_or_si128(finalAG, finalRB); \
}

// same as BLEND_SOURCE_OVER_ARGB32_SSE2, but for one vector srcVector
#define BLEND_SOURCE_OVER_ARGB32_SSE2_helper(dst, srcVector, nullVector, half, one, colorMask, alphaMask) { \
        const __m128i srcVectorAlpha = _mm_and_si128(srcVector, alphaMask); \
        if (_mm_movemask_epi8(_mm_cmpeq_epi32(srcVectorAlpha, alphaMask)) == 0xffff) { \
            /* all opaque */ \
            _mm_store_si128((__m128i *)&dst[x], srcVector); \
        } else if (_mm_movemask_epi8(_mm_cmpeq_epi32(srcVectorAlpha, nullVector)) != 0xffff) { \
            /* not fully transparent */ \
            /* extract the alpha channel on 2 x 16 bits */ \
            /* so we have room for the multiplication */ \
            /* each 32 bits will be in the form 0x00AA00AA */ \
            /* with A being the 1 - alpha */ \
            __m128i alphaChannel = _mm_srli_epi32(srcVector, 24); \
            alphaChannel = _mm_or_si128(alphaChannel, _mm_slli_epi32(alphaChannel, 16)); \
            alphaChannel = _mm_sub_epi16(one, alphaChannel); \
 \
            const __m128i dstVector = _mm_load_si128((__m128i *)&dst[x]); \
            __m128i destMultipliedByOneMinusAlpha; \
            BYTE_MUL_SSE2(destMultipliedByOneMinusAlpha, dstVector, alphaChannel, colorMask, half); \
 \
            /* result = s + d * (1-alpha) */\
            const __m128i result = _mm_add_epi8(srcVector, destMultipliedByOneMinusAlpha); \
            _mm_store_si128((__m128i *)&dst[x], result); \
        } \
    }


// Basically blend src over dst with the const alpha defined as constAlphaVector.
// nullVector, half, one, colorMask are constant across the whole image/texture, and should be defined as:
//const __m128i nullVector = _mm_set1_epi32(0);
//const __m128i half = _mm_set1_epi16(0x80);
//const __m128i one = _mm_set1_epi16(0xff);
//const __m128i colorMask = _mm_set1_epi32(0x00ff00ff);
//const __m128i alphaMask = _mm_set1_epi32(0xff000000);
//
// The computation being done is:
// result = s + d * (1-alpha)
// with shortcuts if fully opaque or fully transparent.
#define BLEND_SOURCE_OVER_ARGB32_SSE2(dst, src, length, nullVector, half, one, colorMask, alphaMask) { \
    int x = 0; \
\
    /* First, get dst aligned. */ \
    ALIGNMENT_PROLOGUE_16BYTES(dst, x, length) { \
        blend_pixel(dst[x], src[x]); \
    } \
\
    for (; x < length-3; x += 4) { \
        const __m128i srcVector = _mm_loadu_si128((const __m128i *)&src[x]); \
        BLEND_SOURCE_OVER_ARGB32_SSE2_helper(dst, srcVector, nullVector, half, one, colorMask, alphaMask) \
    } \
    SIMD_EPILOGUE(x, length, 3) { \
        blend_pixel(dst[x], src[x]); \
    } \
}

// Basically blend src over dst with the const alpha defined as constAlphaVector.
// nullVector, half, one, colorMask are constant across the whole image/texture, and should be defined as:
//const __m128i nullVector = _mm_set1_epi32(0);
//const __m128i half = _mm_set1_epi16(0x80);
//const __m128i one = _mm_set1_epi16(0xff);
//const __m128i colorMask = _mm_set1_epi32(0x00ff00ff);
//
// The computation being done is:
// dest = (s + d * sia) * ca + d * cia
//      = s * ca + d * (sia * ca + cia)
//      = s * ca + d * (1 - sa*ca)
#define BLEND_SOURCE_OVER_ARGB32_WITH_CONST_ALPHA_SSE2(dst, src, length, nullVector, half, one, colorMask, constAlphaVector) \
{ \
    int x = 0; \
\
    ALIGNMENT_PROLOGUE_16BYTES(dst, x, length) { \
        blend_pixel(dst[x], src[x], const_alpha); \
    } \
\
    for (; x < length-3; x += 4) { \
        __m128i srcVector = _mm_loadu_si128((const __m128i *)&src[x]); \
        if (_mm_movemask_epi8(_mm_cmpeq_epi32(srcVector, nullVector)) != 0xffff) { \
            BYTE_MUL_SSE2(srcVector, srcVector, constAlphaVector, colorMask, half); \
\
            __m128i alphaChannel = _mm_srli_epi32(srcVector, 24); \
            alphaChannel = _mm_or_si128(alphaChannel, _mm_slli_epi32(alphaChannel, 16)); \
            alphaChannel = _mm_sub_epi16(one, alphaChannel); \
 \
            const __m128i dstVector = _mm_load_si128((__m128i *)&dst[x]); \
            __m128i destMultipliedByOneMinusAlpha; \
            BYTE_MUL_SSE2(destMultipliedByOneMinusAlpha, dstVector, alphaChannel, colorMask, half); \
 \
            const __m128i result = _mm_add_epi8(srcVector, destMultipliedByOneMinusAlpha); \
            _mm_store_si128((__m128i *)&dst[x], result); \
        } \
    } \
    SIMD_EPILOGUE(x, length, 3) { \
        blend_pixel(dst[x], src[x], const_alpha); \
    } \
}

QT_END_NAMESPACE

#endif // __SSE2__

QT_BEGIN_NAMESPACE
#if QT_COMPILER_SUPPORTS_HERE(SSE4_1)
QT_FUNCTION_TARGET(SSE2)
static inline void Q_DECL_VECTORCALL reciprocal_mul_ss(__m128 &ia, const __m128 a, float mul)
{
    ia = _mm_rcp_ss(a); // Approximate 1/a
    // Improve precision of ia using Newton-Raphson
    ia = _mm_sub_ss(_mm_add_ss(ia, ia), _mm_mul_ss(ia, _mm_mul_ss(ia, a)));
    ia = _mm_mul_ss(ia, _mm_set_ss(mul));
    ia = _mm_shuffle_ps(ia, ia, _MM_SHUFFLE(0,0,0,0));
}

QT_FUNCTION_TARGET(SSE4_1)
static inline QRgb qUnpremultiply_sse4(QRgb p)
{
    const uint alpha = qAlpha(p);
    if (alpha == 255)
        return p;
    if (alpha == 0)
        return 0;
    const __m128 va = _mm_set1_ps(alpha);
    __m128 via;
    reciprocal_mul_ss(via, va, 255.0f); // Approximate 1/a
    __m128i vl = _mm_cvtepu8_epi32(_mm_cvtsi32_si128(p));
    vl = _mm_cvtps_epi32(_mm_mul_ps(_mm_cvtepi32_ps(vl), via));
    vl = _mm_packus_epi32(vl, vl);
    vl = _mm_insert_epi16(vl, alpha, 3);
    vl = _mm_packus_epi16(vl, vl);
    return _mm_cvtsi128_si32(vl);
}

template<enum QtPixelOrder PixelOrder>
QT_FUNCTION_TARGET(SSE4_1)
static inline uint qConvertArgb32ToA2rgb30_sse4(QRgb p)
{
    const uint alpha = qAlpha(p);
    if (alpha == 255)
        return qConvertRgb32ToRgb30<PixelOrder>(p);
    if (alpha == 0)
        return 0;
    constexpr float mult = 1023.0f / (255 >> 6);
    const uint newalpha = (alpha >> 6);
    const __m128 va = _mm_set1_ps(alpha);
    __m128 via;
    reciprocal_mul_ss(via, va, mult * newalpha);
    __m128i vl = _mm_cvtsi32_si128(p);
    vl = _mm_cvtepu8_epi32(vl);
    vl = _mm_cvtps_epi32(_mm_mul_ps(_mm_cvtepi32_ps(vl), via));
    vl = _mm_packus_epi32(vl, vl);
    uint rgb30 = (newalpha << 30);
    rgb30 |= ((uint)_mm_extract_epi16(vl, 1)) << 10;
    if (PixelOrder == PixelOrderRGB) {
        rgb30 |= ((uint)_mm_extract_epi16(vl, 2)) << 20;
        rgb30 |= ((uint)_mm_extract_epi16(vl, 0));
    } else {
        rgb30 |= ((uint)_mm_extract_epi16(vl, 0)) << 20;
        rgb30 |= ((uint)_mm_extract_epi16(vl, 2));
    }
    return rgb30;
}

template<enum QtPixelOrder PixelOrder>
QT_FUNCTION_TARGET(SSE4_1)
static inline uint qConvertRgba64ToRgb32_sse4(QRgba64 p)
{
    if (p.isTransparent())
        return 0;
    __m128i vl = _mm_loadl_epi64(reinterpret_cast<const __m128i *>(&p));
    if (!p.isOpaque()) {
        const __m128 va = _mm_set1_ps(p.alpha());
        __m128 via;
        reciprocal_mul_ss(via, va, 65535.0f);
        vl = _mm_unpacklo_epi16(vl, _mm_setzero_si128());
        vl = _mm_cvtps_epi32(_mm_mul_ps(_mm_cvtepi32_ps(vl) , via));
        vl = _mm_packus_epi32(vl, vl);
        vl = _mm_insert_epi16(vl, p.alpha(), 3);
    }
    if (PixelOrder == PixelOrderBGR)
        vl = _mm_shufflelo_epi16(vl, _MM_SHUFFLE(3, 0, 1, 2));
    return toArgb32(vl);
}
#endif
QT_END_NAMESPACE

#endif // QDRAWINGPRIMITIVE_SSE2_P_H

Coverage Report

Created: 2026-03-31 07:41

Line	Count	Source
1		// Copyright (C) 2016 The Qt Company Ltd.
2		// SPDX-License-Identifier: LicenseRef-Qt-Commercial OR LGPL-3.0-only OR GPL-2.0-only OR GPL-3.0-only
3		// Qt-Security score:significant reason:default
4
5		#ifndef QDRAWINGPRIMITIVE_SSE2_P_H
6		#define QDRAWINGPRIMITIVE_SSE2_P_H
7
8		#include <QtGui/private/qtguiglobal_p.h>
9		#include <private/qsimd_p.h>
10		#include "qdrawhelper_x86_p.h"
11		#include "qrgba64_p.h"
12
13		#ifdef __SSE2__
14
15		//
16		// W A R N I N G
17		// -------------
18		//
19		// This file is not part of the Qt API. It exists purely as an
20		// implementation detail. This header file may change from version to
21		// version without notice, or even be removed.
22		//
23		// We mean it.
24		//
25
26		QT_BEGIN_NAMESPACE
27
28		/*
29		* Multiply the components of pixelVector by alphaChannel
30		* Each 32bits components of alphaChannel must be in the form 0x00AA00AA
31		* colorMask must have 0x00ff00ff on each 32 bits component
32		* half must have the value 128 (0x80) for each 32 bits component
33		*/
34	0	#define BYTE_MUL_SSE2(result, pixelVector, alphaChannel, colorMask, half) \
35	0	{ \
36	0	/* 1. separate the colors in 2 vectors so each color is on 16 bits \
37	0	(in order to be multiplied by the alpha \
38	0	each 32 bit of dstVectorAG are in the form 0x00AA00GG \
39	0	each 32 bit of dstVectorRB are in the form 0x00RR00BB */\
40	0	__m128i pixelVectorAG = _mm_srli_epi16(pixelVector, 8); \
41	0	__m128i pixelVectorRB = _mm_and_si128(pixelVector, colorMask); \
42	0	\
43	0	/* 2. multiply the vectors by the alpha channel */\
44	0	pixelVectorAG = _mm_mullo_epi16(pixelVectorAG, alphaChannel); \
45	0	pixelVectorRB = _mm_mullo_epi16(pixelVectorRB, alphaChannel); \
46	0	\
47	0	/* 3. divide by 255, that's the tricky part. \
48	0	we do it like for BYTE_MUL(), with bit shift: X/255 ~= (X + X/256 + rounding)/256 */ \
49	0	/** so first (X + X/256 + rounding) */\
50	0	pixelVectorRB = _mm_add_epi16(pixelVectorRB, _mm_srli_epi16(pixelVectorRB, 8)); \
51	0	pixelVectorRB = _mm_add_epi16(pixelVectorRB, half); \
52	0	pixelVectorAG = _mm_add_epi16(pixelVectorAG, _mm_srli_epi16(pixelVectorAG, 8)); \
53	0	pixelVectorAG = _mm_add_epi16(pixelVectorAG, half); \
54	0	\
55	0	/** second divide by 256 */\
56	0	pixelVectorRB = _mm_srli_epi16(pixelVectorRB, 8); \
57	0	/** for AG, we could >> 8 to divide followed by << 8 to put the \
58	0	bytes in the correct position. By masking instead, we execute \
59	0	only one instruction */\
60	0	pixelVectorAG = _mm_andnot_si128(colorMask, pixelVectorAG); \
61	0	\
62	0	/* 4. combine the 2 pairs of colors */ \
63	0	result = _mm_or_si128(pixelVectorAG, pixelVectorRB); \
64	0	}
65
66		/*
67		* Each 32bits components of alphaChannel must be in the form 0x00AA00AA
68		* oneMinusAlphaChannel must be 255 - alpha for each 32 bits component
69		* colorMask must have 0x00ff00ff on each 32 bits component
70		* half must have the value 128 (0x80) for each 32 bits component
71		*/
72	0	#define INTERPOLATE_PIXEL_255_SSE2(result, srcVector, dstVector, alphaChannel, oneMinusAlphaChannel, colorMask, half) { \
73	0	/* interpolate AG */\
74	0	__m128i srcVectorAG = _mm_srli_epi16(srcVector, 8); \
75	0	__m128i dstVectorAG = _mm_srli_epi16(dstVector, 8); \
76	0	__m128i srcVectorAGalpha = _mm_mullo_epi16(srcVectorAG, alphaChannel); \
77	0	__m128i dstVectorAGoneMinusAlphalpha = _mm_mullo_epi16(dstVectorAG, oneMinusAlphaChannel); \
78	0	__m128i finalAG = _mm_add_epi16(srcVectorAGalpha, dstVectorAGoneMinusAlphalpha); \
79	0	finalAG = _mm_add_epi16(finalAG, _mm_srli_epi16(finalAG, 8)); \
80	0	finalAG = _mm_add_epi16(finalAG, half); \
81	0	finalAG = _mm_andnot_si128(colorMask, finalAG); \
82	0	\
83	0	/* interpolate RB */\
84	0	__m128i srcVectorRB = _mm_and_si128(srcVector, colorMask); \
85	0	__m128i dstVectorRB = _mm_and_si128(dstVector, colorMask); \
86	0	__m128i srcVectorRBalpha = _mm_mullo_epi16(srcVectorRB, alphaChannel); \
87	0	__m128i dstVectorRBoneMinusAlphalpha = _mm_mullo_epi16(dstVectorRB, oneMinusAlphaChannel); \
88	0	__m128i finalRB = _mm_add_epi16(srcVectorRBalpha, dstVectorRBoneMinusAlphalpha); \
89	0	finalRB = _mm_add_epi16(finalRB, _mm_srli_epi16(finalRB, 8)); \
90	0	finalRB = _mm_add_epi16(finalRB, half); \
91	0	finalRB = _mm_srli_epi16(finalRB, 8); \
92	0	\
93	0	/* combine */\
94	0	result = _mm_or_si128(finalAG, finalRB); \
95	0	}
96
97		// same as BLEND_SOURCE_OVER_ARGB32_SSE2, but for one vector srcVector
98	0	#define BLEND_SOURCE_OVER_ARGB32_SSE2_helper(dst, srcVector, nullVector, half, one, colorMask, alphaMask) { \
99	0	const __m128i srcVectorAlpha = _mm_and_si128(srcVector, alphaMask); \
100	0	if (_mm_movemask_epi8(_mm_cmpeq_epi32(srcVectorAlpha, alphaMask)) == 0xffff) { \
101	0	/* all opaque */ \
102	0	_mm_store_si128((__m128i *)&dst[x], srcVector); \
103	0	} else if (_mm_movemask_epi8(_mm_cmpeq_epi32(srcVectorAlpha, nullVector)) != 0xffff) { \
104	0	/* not fully transparent */ \
105	0	/* extract the alpha channel on 2 x 16 bits */ \
106	0	/* so we have room for the multiplication */ \
107	0	/* each 32 bits will be in the form 0x00AA00AA */ \
108	0	/* with A being the 1 - alpha */ \
109	0	__m128i alphaChannel = _mm_srli_epi32(srcVector, 24); \
110	0	alphaChannel = _mm_or_si128(alphaChannel, _mm_slli_epi32(alphaChannel, 16)); \
111	0	alphaChannel = _mm_sub_epi16(one, alphaChannel); \
112	0	\
113	0	const __m128i dstVector = _mm_load_si128((__m128i *)&dst[x]); \
114	0	__m128i destMultipliedByOneMinusAlpha; \
115	0	BYTE_MUL_SSE2(destMultipliedByOneMinusAlpha, dstVector, alphaChannel, colorMask, half); \
116	0	\
117	0	/* result = s + d * (1-alpha) */\
118	0	const __m128i result = _mm_add_epi8(srcVector, destMultipliedByOneMinusAlpha); \
119	0	_mm_store_si128((__m128i *)&dst[x], result); \
120	0	} \
121	0	}
122
123
124		// Basically blend src over dst with the const alpha defined as constAlphaVector.
125		// nullVector, half, one, colorMask are constant across the whole image/texture, and should be defined as:
126		//const __m128i nullVector = _mm_set1_epi32(0);
127		//const __m128i half = _mm_set1_epi16(0x80);
128		//const __m128i one = _mm_set1_epi16(0xff);
129		//const __m128i colorMask = _mm_set1_epi32(0x00ff00ff);
130		//const __m128i alphaMask = _mm_set1_epi32(0xff000000);
131		//
132		// The computation being done is:
133		// result = s + d * (1-alpha)
134		// with shortcuts if fully opaque or fully transparent.
135	0	#define BLEND_SOURCE_OVER_ARGB32_SSE2(dst, src, length, nullVector, half, one, colorMask, alphaMask) { \
136	0	int x = 0; \
137	0	\
138	0	/* First, get dst aligned. */ \
139	0	ALIGNMENT_PROLOGUE_16BYTES(dst, x, length) { \
140	0	blend_pixel(dst[x], src[x]); \
141	0	} \
142	0	\
143	0	for (; x < length-3; x += 4) { \
144	0	const __m128i srcVector = _mm_loadu_si128((const __m128i *)&src[x]); \
145	0	BLEND_SOURCE_OVER_ARGB32_SSE2_helper(dst, srcVector, nullVector, half, one, colorMask, alphaMask) \
146	0	} \
147	0	SIMD_EPILOGUE(x, length, 3) { \
148	0	blend_pixel(dst[x], src[x]); \
149	0	} \
150	0	}
151
152		// Basically blend src over dst with the const alpha defined as constAlphaVector.
153		// nullVector, half, one, colorMask are constant across the whole image/texture, and should be defined as:
154		//const __m128i nullVector = _mm_set1_epi32(0);
155		//const __m128i half = _mm_set1_epi16(0x80);
156		//const __m128i one = _mm_set1_epi16(0xff);
157		//const __m128i colorMask = _mm_set1_epi32(0x00ff00ff);
158		//
159		// The computation being done is:
160		// dest = (s + d * sia) * ca + d * cia
161		// = s * ca + d * (sia * ca + cia)
162		// = s * ca + d * (1 - sa*ca)
163	0	#define BLEND_SOURCE_OVER_ARGB32_WITH_CONST_ALPHA_SSE2(dst, src, length, nullVector, half, one, colorMask, constAlphaVector) \
164	0	{ \
165	0	int x = 0; \
166	0	\
167	0	ALIGNMENT_PROLOGUE_16BYTES(dst, x, length) { \
168	0	blend_pixel(dst[x], src[x], const_alpha); \
169	0	} \
170	0	\
171	0	for (; x < length-3; x += 4) { \
172	0	__m128i srcVector = _mm_loadu_si128((const __m128i *)&src[x]); \
173	0	if (_mm_movemask_epi8(_mm_cmpeq_epi32(srcVector, nullVector)) != 0xffff) { \
174	0	BYTE_MUL_SSE2(srcVector, srcVector, constAlphaVector, colorMask, half); \
175	0	\
176	0	__m128i alphaChannel = _mm_srli_epi32(srcVector, 24); \
177	0	alphaChannel = _mm_or_si128(alphaChannel, _mm_slli_epi32(alphaChannel, 16)); \
178	0	alphaChannel = _mm_sub_epi16(one, alphaChannel); \
179	0	\
180	0	const __m128i dstVector = _mm_load_si128((__m128i *)&dst[x]); \
181	0	__m128i destMultipliedByOneMinusAlpha; \
182	0	BYTE_MUL_SSE2(destMultipliedByOneMinusAlpha, dstVector, alphaChannel, colorMask, half); \
183	0	\
184	0	const __m128i result = _mm_add_epi8(srcVector, destMultipliedByOneMinusAlpha); \
185	0	_mm_store_si128((__m128i *)&dst[x], result); \
186	0	} \
187	0	} \
188	0	SIMD_EPILOGUE(x, length, 3) { \
189	0	blend_pixel(dst[x], src[x], const_alpha); \
190	0	} \
191	0	}
192
193		QT_END_NAMESPACE
194
195		#endif // __SSE2__
196
197		QT_BEGIN_NAMESPACE
198		#if QT_COMPILER_SUPPORTS_HERE(SSE4_1)
199		QT_FUNCTION_TARGET(SSE2)
200		static inline void Q_DECL_VECTORCALL reciprocal_mul_ss(__m128 &ia, const __m128 a, float mul)
201	0	{
202	0	ia = _mm_rcp_ss(a); // Approximate 1/a
203		// Improve precision of ia using Newton-Raphson
204	0	ia = _mm_sub_ss(_mm_add_ss(ia, ia), _mm_mul_ss(ia, _mm_mul_ss(ia, a)));
205	0	ia = _mm_mul_ss(ia, _mm_set_ss(mul));
206	0	ia = _mm_shuffle_ps(ia, ia, _MM_SHUFFLE(0,0,0,0));
207	0	} Unexecuted instantiation: qdrawhelper_ssse3.cpp:reciprocal_mul_ss(float __vector(4)&, float __vector(4), float) Unexecuted instantiation: qdrawhelper_sse4.cpp:reciprocal_mul_ss(float __vector(4)&, float __vector(4), float) Unexecuted instantiation: qdrawhelper.cpp:reciprocal_mul_ss(float __vector(4)&, float __vector(4), float) Unexecuted instantiation: qdrawhelper_sse2.cpp:reciprocal_mul_ss(float __vector(4)&, float __vector(4), float) Unexecuted instantiation: qdrawhelper_avx2.cpp:reciprocal_mul_ss(float __vector(4)&, float __vector(4), float)
208
209		QT_FUNCTION_TARGET(SSE4_1)
210		static inline QRgb qUnpremultiply_sse4(QRgb p)
211	0	{
212	0	const uint alpha = qAlpha(p);
213	0	if (alpha == 255)
214	0	return p;
215	0	if (alpha == 0)
216	0	return 0;
217	0	const __m128 va = _mm_set1_ps(alpha);
218	0	__m128 via;
219	0	reciprocal_mul_ss(via, va, 255.0f); // Approximate 1/a
220	0	__m128i vl = _mm_cvtepu8_epi32(_mm_cvtsi32_si128(p));
221	0	vl = _mm_cvtps_epi32(_mm_mul_ps(_mm_cvtepi32_ps(vl), via));
222	0	vl = _mm_packus_epi32(vl, vl);
223	0	vl = _mm_insert_epi16(vl, alpha, 3);
224	0	vl = _mm_packus_epi16(vl, vl);
225	0	return _mm_cvtsi128_si32(vl);
226	0	} Unexecuted instantiation: qdrawhelper_ssse3.cpp:qUnpremultiply_sse4(unsigned int) Unexecuted instantiation: qdrawhelper_sse4.cpp:qUnpremultiply_sse4(unsigned int) Unexecuted instantiation: qdrawhelper.cpp:qUnpremultiply_sse4(unsigned int) Unexecuted instantiation: qdrawhelper_sse2.cpp:qUnpremultiply_sse4(unsigned int) Unexecuted instantiation: qdrawhelper_avx2.cpp:qUnpremultiply_sse4(unsigned int)
227
228		template<enum QtPixelOrder PixelOrder>
229		QT_FUNCTION_TARGET(SSE4_1)
230		static inline uint qConvertArgb32ToA2rgb30_sse4(QRgb p)
231	0	{
232	0	const uint alpha = qAlpha(p);
233	0	if (alpha == 255)
234	0	return qConvertRgb32ToRgb30<PixelOrder>(p);
235	0	if (alpha == 0)
236	0	return 0;
237	0	constexpr float mult = 1023.0f / (255 >> 6);
238	0	const uint newalpha = (alpha >> 6);
239	0	const __m128 va = _mm_set1_ps(alpha);
240	0	__m128 via;
241	0	reciprocal_mul_ss(via, va, mult * newalpha);
242	0	__m128i vl = _mm_cvtsi32_si128(p);
243	0	vl = _mm_cvtepu8_epi32(vl);
244	0	vl = _mm_cvtps_epi32(_mm_mul_ps(_mm_cvtepi32_ps(vl), via));
245	0	vl = _mm_packus_epi32(vl, vl);
246	0	uint rgb30 = (newalpha << 30);
247	0	rgb30 \|= ((uint)_mm_extract_epi16(vl, 1)) << 10;
248	0	if (PixelOrder == PixelOrderRGB) {
249	0	rgb30 \|= ((uint)_mm_extract_epi16(vl, 2)) << 20;
250	0	rgb30 \|= ((uint)_mm_extract_epi16(vl, 0));
251	0	} else {
252	0	rgb30 \|= ((uint)_mm_extract_epi16(vl, 0)) << 20;
253	0	rgb30 \|= ((uint)_mm_extract_epi16(vl, 2));
254	0	}
255	0	return rgb30;
256	0	} Unexecuted instantiation: qdrawhelper_sse4.cpp:unsigned int qConvertArgb32ToA2rgb30_sse4<(QtPixelOrder)1>(unsigned int) Unexecuted instantiation: qdrawhelper_sse4.cpp:unsigned int qConvertArgb32ToA2rgb30_sse4<(QtPixelOrder)0>(unsigned int)
257
258		template<enum QtPixelOrder PixelOrder>
259		QT_FUNCTION_TARGET(SSE4_1)
260		static inline uint qConvertRgba64ToRgb32_sse4(QRgba64 p)
261	0	{
262	0	if (p.isTransparent())
263	0	return 0;
264	0	__m128i vl = _mm_loadl_epi64(reinterpret_cast<const __m128i *>(&p));
265	0	if (!p.isOpaque()) {
266	0	const __m128 va = _mm_set1_ps(p.alpha());
267	0	__m128 via;
268	0	reciprocal_mul_ss(via, va, 65535.0f);
269	0	vl = _mm_unpacklo_epi16(vl, _mm_setzero_si128());
270	0	vl = _mm_cvtps_epi32(_mm_mul_ps(_mm_cvtepi32_ps(vl) , via));
271	0	vl = _mm_packus_epi32(vl, vl);
272	0	vl = _mm_insert_epi16(vl, p.alpha(), 3);
273	0	}
274	0	if (PixelOrder == PixelOrderBGR)
275	0	vl = _mm_shufflelo_epi16(vl, _MM_SHUFFLE(3, 0, 1, 2));
276	0	return toArgb32(vl);
277	0	} Unexecuted instantiation: qdrawhelper_sse4.cpp:unsigned int qConvertRgba64ToRgb32_sse4<(QtPixelOrder)1>(QRgba64) Unexecuted instantiation: qdrawhelper_sse4.cpp:unsigned int qConvertRgba64ToRgb32_sse4<(QtPixelOrder)0>(QRgba64)
278		#endif
279		QT_END_NAMESPACE
280
281		#endif // QDRAWINGPRIMITIVE_SSE2_P_H