/src/Simd/src/Simd/SimdSse41Interleave.cpp

Source (jump to first uncovered line)
/*
* Simd Library (http://ermig1979.github.io/Simd).
*
* Copyright (c) 2011-2022 Yermalayeu Ihar.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include "Simd/SimdMemory.h"
#include "Simd/SimdStore.h"
#include "Simd/SimdInterleave.h"

namespace Simd
{
#ifdef SIMD_SSE41_ENABLE    
    namespace Sse41
    {
        template <bool align> SIMD_INLINE void InterleaveUv(const uint8_t* u, const uint8_t* v, uint8_t* uv)
        {
            __m128i _u = Load<align>((__m128i*)u);
            __m128i _v = Load<align>((__m128i*)v);
            Store<align>((__m128i*)uv + 0, _mm_unpacklo_epi8(_u, _v));
            Store<align>((__m128i*)uv + 1, _mm_unpackhi_epi8(_u, _v));
        }

        template <bool align> void InterleaveUv(const uint8_t* u, size_t uStride, const uint8_t* v, size_t vStride, size_t width, size_t height, uint8_t* uv, size_t uvStride)
        {
            assert(width >= A);
            if (align)
            {
                assert(Aligned(uv) && Aligned(uvStride) && Aligned(u) && Aligned(uStride) && Aligned(v) && Aligned(vStride));
            }

            size_t bodyWidth = AlignLo(width, A);
            size_t tail = width - bodyWidth;
            for (size_t row = 0; row < height; ++row)
            {
                for (size_t col = 0, offset = 0; col < bodyWidth; col += A, offset += DA)
                    InterleaveUv<align>(u + col, v + col, uv + offset);
                if (tail)
                {
                    size_t col = width - A;
                    size_t offset = 2 * col;
                    InterleaveUv<false>(u + col, v + col, uv + offset);
                }
                u += uStride;
                v += vStride;
                uv += uvStride;
            }
        }

        void InterleaveUv(const uint8_t* u, size_t uStride, const uint8_t* v, size_t vStride, size_t width, size_t height, uint8_t* uv, size_t uvStride)
        {
            if (Aligned(uv) && Aligned(uvStride) && Aligned(u) && Aligned(uStride) && Aligned(v) && Aligned(vStride))
                InterleaveUv<true>(u, uStride, v, vStride, width, height, uv, uvStride);
            else
                InterleaveUv<false>(u, uStride, v, vStride, width, height, uv, uvStride);
        }

        //-----------------------------------------------------------------------------------------

        template <bool align> SIMD_INLINE void InterleaveBgr(const uint8_t * b, const uint8_t * g, const uint8_t * r, size_t offset, uint8_t * bgr)
        {
            __m128i _b = Load<align>((__m128i*)(b + offset));
            __m128i _g = Load<align>((__m128i*)(g + offset));
            __m128i _r = Load<align>((__m128i*)(r + offset));
            Store<align>((__m128i*)bgr + 0, InterleaveBgr<0>(_b, _g, _r));
            Store<align>((__m128i*)bgr + 1, InterleaveBgr<1>(_b, _g, _r));
            Store<align>((__m128i*)bgr + 2, InterleaveBgr<2>(_b, _g, _r));
        }

        template <bool align> void InterleaveBgr(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, size_t width, size_t height, uint8_t * bgr, size_t bgrStride)
        {
            assert(width >= A);
            if (align)
            {
                assert(Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride));
                assert(Aligned(r) && Aligned(rStride) && Aligned(bgr) && Aligned(bgrStride));
            }

            size_t alignedWidth = AlignLo(width, A);
            size_t tail = width - alignedWidth;
            size_t A3 = A * 3;
            for (size_t row = 0; row < height; ++row)
            {
                for (size_t col = 0, offset = 0; col < alignedWidth; col += A, offset += A3)
                    InterleaveBgr<align>(b, g, r, col, bgr + offset);
                if (tail)
                    InterleaveBgr<false>(b, g, r, width - A, bgr + 3 * (width - A));
                b += bStride;
                g += gStride;
                r += rStride;
                bgr += bgrStride;
            }
        }

        void InterleaveBgr(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, size_t width, size_t height, uint8_t * bgr, size_t bgrStride)
        {
            if (Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride)
                && Aligned(r) && Aligned(rStride) && Aligned(bgr) && Aligned(bgrStride))
                InterleaveBgr<true>(b, bStride, g, gStride, r, rStride, width, height, bgr, bgrStride);
            else
                InterleaveBgr<false>(b, bStride, g, gStride, r, rStride, width, height, bgr, bgrStride);
        }

        //-----------------------------------------------------------------------------------------

        template <bool align> SIMD_INLINE void InterleaveBgra(const uint8_t * b, const uint8_t * g, const uint8_t * r, const uint8_t * a, size_t offset, uint8_t * bgra)
        {
            __m128i _b = Load<align>((__m128i*)(b + offset));
            __m128i _g = Load<align>((__m128i*)(g + offset));
            __m128i _r = Load<align>((__m128i*)(r + offset));
            __m128i _a = Load<align>((__m128i*)(a + offset));
            __m128i bg0 = _mm_unpacklo_epi8(_b, _g);
            __m128i bg1 = _mm_unpackhi_epi8(_b, _g);
            __m128i ra0 = _mm_unpacklo_epi8(_r, _a);
            __m128i ra1 = _mm_unpackhi_epi8(_r, _a);
            Store<align>((__m128i*)bgra + 0, _mm_unpacklo_epi16(bg0, ra0));
            Store<align>((__m128i*)bgra + 1, _mm_unpackhi_epi16(bg0, ra0));
            Store<align>((__m128i*)bgra + 2, _mm_unpacklo_epi16(bg1, ra1));
            Store<align>((__m128i*)bgra + 3, _mm_unpackhi_epi16(bg1, ra1));
        }

        template <bool align> void InterleaveBgra(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, const uint8_t * a, size_t aStride, size_t width, size_t height, uint8_t * bgra, size_t bgraStride)
        {
            assert(width >= A);
            if (align)
            {
                assert(Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride) && Aligned(r) && Aligned(rStride));
                assert(Aligned(a) && Aligned(aStride) && Aligned(bgra) && Aligned(bgraStride));
            }

            size_t alignedWidth = AlignLo(width, A);
            size_t tail = width - alignedWidth;
            for (size_t row = 0; row < height; ++row)
            {
                for (size_t col = 0, offset = 0; col < alignedWidth; col += A, offset += QA)
                    InterleaveBgra<align>(b, g, r, a, col, bgra + offset);
                if (tail)
                    InterleaveBgra<false>(b, g, r, a, width - A, bgra + 4 * (width - A));
                b += bStride;
                g += gStride;
                r += rStride;
                a += aStride;
                bgra += bgraStride;
            }
        }

        void InterleaveBgra(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, const uint8_t * a, size_t aStride, size_t width, size_t height, uint8_t * bgra, size_t bgraStride)
        {
            if (Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride) && Aligned(r) && Aligned(rStride) &&
                Aligned(a) && Aligned(aStride) && Aligned(bgra) && Aligned(bgraStride))
                InterleaveBgra<true>(b, bStride, g, gStride, r, rStride, a, aStride, width, height, bgra, bgraStride);
            else
                InterleaveBgra<false>(b, bStride, g, gStride, r, rStride, a, aStride, width, height, bgra, bgraStride);
        }
    }
#endif
}

Line	Count	Source (jump to first uncovered line)
1		/*
2		* Simd Library (http://ermig1979.github.io/Simd).
3		*
4		* Copyright (c) 2011-2022 Yermalayeu Ihar.
5		*
6		* Permission is hereby granted, free of charge, to any person obtaining a copy
7		* of this software and associated documentation files (the "Software"), to deal
8		* in the Software without restriction, including without limitation the rights
9		* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
10		* copies of the Software, and to permit persons to whom the Software is
11		* furnished to do so, subject to the following conditions:
12		*
13		* The above copyright notice and this permission notice shall be included in
14		* all copies or substantial portions of the Software.
15		*
16		* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17		* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18		* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19		* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20		* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21		* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
22		* SOFTWARE.
23		*/
24		#include "Simd/SimdMemory.h"
25		#include "Simd/SimdStore.h"
26		#include "Simd/SimdInterleave.h"
27
28		namespace Simd
29		{
30		#ifdef SIMD_SSE41_ENABLE
31		namespace Sse41
32		{
33		template <bool align> SIMD_INLINE void InterleaveUv(const uint8_t* u, const uint8_t* v, uint8_t* uv)
34	0	{
35	0	__m128i _u = Load<align>((__m128i*)u);
36	0	__m128i _v = Load<align>((__m128i*)v);
37	0	Store<align>((__m128i*)uv + 0, _mm_unpacklo_epi8(_u, _v));
38	0	Store<align>((__m128i*)uv + 1, _mm_unpackhi_epi8(_u, _v));
39	0	} Unexecuted instantiation: void Simd::Sse41::InterleaveUv<true>(unsigned char const, unsigned char const, unsigned char) Unexecuted instantiation: void Simd::Sse41::InterleaveUv<false>(unsigned char const, unsigned char const, unsigned char)
40
41		template <bool align> void InterleaveUv(const uint8_t* u, size_t uStride, const uint8_t* v, size_t vStride, size_t width, size_t height, uint8_t* uv, size_t uvStride)
42	0	{
43	0	assert(width >= A);
44	0	if (align)
45	0	{
46	0	assert(Aligned(uv) && Aligned(uvStride) && Aligned(u) && Aligned(uStride) && Aligned(v) && Aligned(vStride));
47	0	}
48
49	0	size_t bodyWidth = AlignLo(width, A);
50	0	size_t tail = width - bodyWidth;
51	0	for (size_t row = 0; row < height; ++row)
52	0	{
53	0	for (size_t col = 0, offset = 0; col < bodyWidth; col += A, offset += DA)
54	0	InterleaveUv<align>(u + col, v + col, uv + offset);
55	0	if (tail)
56	0	{
57	0	size_t col = width - A;
58	0	size_t offset = 2 * col;
59	0	InterleaveUv<false>(u + col, v + col, uv + offset);
60	0	}
61	0	u += uStride;
62	0	v += vStride;
63	0	uv += uvStride;
64	0	}
65	0	} Unexecuted instantiation: void Simd::Sse41::InterleaveUv<true>(unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned long, unsigned long, unsigned char, unsigned long) Unexecuted instantiation: void Simd::Sse41::InterleaveUv<false>(unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned long, unsigned long, unsigned char, unsigned long)
66
67		void InterleaveUv(const uint8_t* u, size_t uStride, const uint8_t* v, size_t vStride, size_t width, size_t height, uint8_t* uv, size_t uvStride)
68	0	{
69	0	if (Aligned(uv) && Aligned(uvStride) && Aligned(u) && Aligned(uStride) && Aligned(v) && Aligned(vStride))
70	0	InterleaveUv<true>(u, uStride, v, vStride, width, height, uv, uvStride);
71	0	else
72	0	InterleaveUv<false>(u, uStride, v, vStride, width, height, uv, uvStride);
73	0	}
74
75		//-----------------------------------------------------------------------------------------
76
77		template <bool align> SIMD_INLINE void InterleaveBgr(const uint8_t * b, const uint8_t * g, const uint8_t * r, size_t offset, uint8_t * bgr)
78	0	{
79	0	__m128i _b = Load<align>((__m128i*)(b + offset));
80	0	__m128i _g = Load<align>((__m128i*)(g + offset));
81	0	__m128i _r = Load<align>((__m128i*)(r + offset));
82	0	Store<align>((__m128i*)bgr + 0, InterleaveBgr<0>(_b, _g, _r));
83	0	Store<align>((__m128i*)bgr + 1, InterleaveBgr<1>(_b, _g, _r));
84	0	Store<align>((__m128i*)bgr + 2, InterleaveBgr<2>(_b, _g, _r));
85	0	} Unexecuted instantiation: void Simd::Sse41::InterleaveBgr<true>(unsigned char const, unsigned char const, unsigned char const, unsigned long, unsigned char) Unexecuted instantiation: void Simd::Sse41::InterleaveBgr<false>(unsigned char const, unsigned char const, unsigned char const, unsigned long, unsigned char)
86
87		template <bool align> void InterleaveBgr(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, size_t width, size_t height, uint8_t * bgr, size_t bgrStride)
88	0	{
89	0	assert(width >= A);
90	0	if (align)
91	0	{
92	0	assert(Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride));
93	0	assert(Aligned(r) && Aligned(rStride) && Aligned(bgr) && Aligned(bgrStride));
94	0	}
95
96	0	size_t alignedWidth = AlignLo(width, A);
97	0	size_t tail = width - alignedWidth;
98	0	size_t A3 = A * 3;
99	0	for (size_t row = 0; row < height; ++row)
100	0	{
101	0	for (size_t col = 0, offset = 0; col < alignedWidth; col += A, offset += A3)
102	0	InterleaveBgr<align>(b, g, r, col, bgr + offset);
103	0	if (tail)
104	0	InterleaveBgr<false>(b, g, r, width - A, bgr + 3 * (width - A));
105	0	b += bStride;
106	0	g += gStride;
107	0	r += rStride;
108	0	bgr += bgrStride;
109	0	}
110	0	} Unexecuted instantiation: void Simd::Sse41::InterleaveBgr<true>(unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned long, unsigned long, unsigned char, unsigned long) Unexecuted instantiation: void Simd::Sse41::InterleaveBgr<false>(unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned long, unsigned long, unsigned char, unsigned long)
111
112		void InterleaveBgr(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, size_t width, size_t height, uint8_t * bgr, size_t bgrStride)
113	0	{
114	0	if (Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride)
115	0	&& Aligned(r) && Aligned(rStride) && Aligned(bgr) && Aligned(bgrStride))
116	0	InterleaveBgr<true>(b, bStride, g, gStride, r, rStride, width, height, bgr, bgrStride);
117	0	else
118	0	InterleaveBgr<false>(b, bStride, g, gStride, r, rStride, width, height, bgr, bgrStride);
119	0	}
120
121		//-----------------------------------------------------------------------------------------
122
123		template <bool align> SIMD_INLINE void InterleaveBgra(const uint8_t * b, const uint8_t * g, const uint8_t * r, const uint8_t * a, size_t offset, uint8_t * bgra)
124	0	{
125	0	__m128i _b = Load<align>((__m128i*)(b + offset));
126	0	__m128i _g = Load<align>((__m128i*)(g + offset));
127	0	__m128i _r = Load<align>((__m128i*)(r + offset));
128	0	__m128i _a = Load<align>((__m128i*)(a + offset));
129	0	__m128i bg0 = _mm_unpacklo_epi8(_b, _g);
130	0	__m128i bg1 = _mm_unpackhi_epi8(_b, _g);
131	0	__m128i ra0 = _mm_unpacklo_epi8(_r, _a);
132	0	__m128i ra1 = _mm_unpackhi_epi8(_r, _a);
133	0	Store<align>((__m128i*)bgra + 0, _mm_unpacklo_epi16(bg0, ra0));
134	0	Store<align>((__m128i*)bgra + 1, _mm_unpackhi_epi16(bg0, ra0));
135	0	Store<align>((__m128i*)bgra + 2, _mm_unpacklo_epi16(bg1, ra1));
136	0	Store<align>((__m128i*)bgra + 3, _mm_unpackhi_epi16(bg1, ra1));
137	0	} Unexecuted instantiation: void Simd::Sse41::InterleaveBgra<true>(unsigned char const, unsigned char const, unsigned char const, unsigned char const, unsigned long, unsigned char) Unexecuted instantiation: void Simd::Sse41::InterleaveBgra<false>(unsigned char const, unsigned char const, unsigned char const, unsigned char const, unsigned long, unsigned char)
138
139		template <bool align> void InterleaveBgra(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, const uint8_t * a, size_t aStride, size_t width, size_t height, uint8_t * bgra, size_t bgraStride)
140	0	{
141	0	assert(width >= A);
142	0	if (align)
143	0	{
144	0	assert(Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride) && Aligned(r) && Aligned(rStride));
145	0	assert(Aligned(a) && Aligned(aStride) && Aligned(bgra) && Aligned(bgraStride));
146	0	}
147
148	0	size_t alignedWidth = AlignLo(width, A);
149	0	size_t tail = width - alignedWidth;
150	0	for (size_t row = 0; row < height; ++row)
151	0	{
152	0	for (size_t col = 0, offset = 0; col < alignedWidth; col += A, offset += QA)
153	0	InterleaveBgra<align>(b, g, r, a, col, bgra + offset);
154	0	if (tail)
155	0	InterleaveBgra<false>(b, g, r, a, width - A, bgra + 4 * (width - A));
156	0	b += bStride;
157	0	g += gStride;
158	0	r += rStride;
159	0	a += aStride;
160	0	bgra += bgraStride;
161	0	}
162	0	} Unexecuted instantiation: void Simd::Sse41::InterleaveBgra<true>(unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned long, unsigned long, unsigned char, unsigned long) Unexecuted instantiation: void Simd::Sse41::InterleaveBgra<false>(unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned char const, unsigned long, unsigned long, unsigned long, unsigned char, unsigned long)
163
164		void InterleaveBgra(const uint8_t * b, size_t bStride, const uint8_t * g, size_t gStride, const uint8_t * r, size_t rStride, const uint8_t * a, size_t aStride, size_t width, size_t height, uint8_t * bgra, size_t bgraStride)
165	0	{
166	0	if (Aligned(b) && Aligned(bStride) && Aligned(g) && Aligned(gStride) && Aligned(r) && Aligned(rStride) &&
167	0	Aligned(a) && Aligned(aStride) && Aligned(bgra) && Aligned(bgraStride))
168	0	InterleaveBgra<true>(b, bStride, g, gStride, r, rStride, a, aStride, width, height, bgra, bgraStride);
169	0	else
170	0	InterleaveBgra<false>(b, bStride, g, gStride, r, rStride, a, aStride, width, height, bgra, bgraStride);
171	0	}
172		}
173		#endif
174		}

Coverage Report

Created: 2025-07-23 07:53