Mercurial > libavcodec.hg
comparison sh4/qpel.c @ 11399:bb19a598f066 libavcodec
sh4: fix about 1000 warnings
author | mru |
---|---|
date | Mon, 08 Mar 2010 02:36:16 +0000 |
parents | 894404f1c2f0 |
children |
comparison
equal
deleted
inserted
replaced
11398:00b80bf84ad9 | 11399:bb19a598f066 |
---|---|
24 #define PIXOP2(OPNAME, OP) \ | 24 #define PIXOP2(OPNAME, OP) \ |
25 \ | 25 \ |
26 static inline void OPNAME ## _pixels4_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 26 static inline void OPNAME ## _pixels4_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
27 {\ | 27 {\ |
28 do {\ | 28 do {\ |
29 OP(LP(dst ),rnd_avg32(LP(src1 ),LP(src2 )) ); \ | 29 OP(LP(dst ),rnd_avg32(LPC(src1 ),LPC(src2 )) ); \ |
30 src1+=src_stride1; \ | 30 src1+=src_stride1; \ |
31 src2+=src_stride2; \ | 31 src2+=src_stride2; \ |
32 dst+=dst_stride; \ | 32 dst+=dst_stride; \ |
33 } while(--h); \ | 33 } while(--h); \ |
34 }\ | 34 }\ |
35 \ | 35 \ |
36 static inline void OPNAME ## _pixels4_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 36 static inline void OPNAME ## _pixels4_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
37 {\ | 37 {\ |
38 do {\ | 38 do {\ |
39 OP(LP(dst ),rnd_avg32(AV_RN32(src1 ),LP(src2 )) ); \ | 39 OP(LP(dst ),rnd_avg32(AV_RN32(src1 ),LPC(src2 )) ); \ |
40 src1+=src_stride1; \ | 40 src1+=src_stride1; \ |
41 src2+=src_stride2; \ | 41 src2+=src_stride2; \ |
42 dst+=dst_stride; \ | 42 dst+=dst_stride; \ |
43 } while(--h); \ | 43 } while(--h); \ |
44 }\ | 44 }\ |
45 \ | 45 \ |
46 static inline void OPNAME ## _no_rnd_pixels16_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 46 static inline void OPNAME ## _no_rnd_pixels16_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
47 {\ | 47 {\ |
48 do {\ | 48 do {\ |
49 OP(LP(dst ),no_rnd_avg32(AV_RN32(src1 ),LP(src2 )) ); \ | 49 OP(LP(dst ),no_rnd_avg32(AV_RN32(src1 ),LPC(src2 )) ); \ |
50 OP(LP(dst+4),no_rnd_avg32(AV_RN32(src1+4),LP(src2+4)) ); \ | 50 OP(LP(dst+4),no_rnd_avg32(AV_RN32(src1+4),LPC(src2+4)) ); \ |
51 OP(LP(dst+8),no_rnd_avg32(AV_RN32(src1+8),LP(src2+8)) ); \ | 51 OP(LP(dst+8),no_rnd_avg32(AV_RN32(src1+8),LPC(src2+8)) ); \ |
52 OP(LP(dst+12),no_rnd_avg32(AV_RN32(src1+12),LP(src2+12)) ); \ | 52 OP(LP(dst+12),no_rnd_avg32(AV_RN32(src1+12),LPC(src2+12)) ); \ |
53 src1+=src_stride1; \ | 53 src1+=src_stride1; \ |
54 src2+=src_stride2; \ | 54 src2+=src_stride2; \ |
55 dst+=dst_stride; \ | 55 dst+=dst_stride; \ |
56 } while(--h); \ | 56 } while(--h); \ |
57 }\ | 57 }\ |
58 \ | 58 \ |
59 static inline void OPNAME ## _pixels16_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 59 static inline void OPNAME ## _pixels16_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
60 {\ | 60 {\ |
61 do {\ | 61 do {\ |
62 OP(LP(dst ),rnd_avg32(AV_RN32(src1 ),LP(src2 )) ); \ | 62 OP(LP(dst ),rnd_avg32(AV_RN32(src1 ),LPC(src2 )) ); \ |
63 OP(LP(dst+4),rnd_avg32(AV_RN32(src1+4),LP(src2+4)) ); \ | 63 OP(LP(dst+4),rnd_avg32(AV_RN32(src1+4),LPC(src2+4)) ); \ |
64 OP(LP(dst+8),rnd_avg32(AV_RN32(src1+8),LP(src2+8)) ); \ | 64 OP(LP(dst+8),rnd_avg32(AV_RN32(src1+8),LPC(src2+8)) ); \ |
65 OP(LP(dst+12),rnd_avg32(AV_RN32(src1+12),LP(src2+12)) ); \ | 65 OP(LP(dst+12),rnd_avg32(AV_RN32(src1+12),LPC(src2+12)) ); \ |
66 src1+=src_stride1; \ | 66 src1+=src_stride1; \ |
67 src2+=src_stride2; \ | 67 src2+=src_stride2; \ |
68 dst+=dst_stride; \ | 68 dst+=dst_stride; \ |
69 } while(--h); \ | 69 } while(--h); \ |
70 }\ | 70 }\ |
71 \ | 71 \ |
72 static inline void OPNAME ## _no_rnd_pixels8_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 72 static inline void OPNAME ## _no_rnd_pixels8_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
73 {\ | 73 {\ |
74 do { /* onlye src2 aligned */\ | 74 do { /* onlye src2 aligned */\ |
75 OP(LP(dst ),no_rnd_avg32(AV_RN32(src1 ),LP(src2 )) ); \ | 75 OP(LP(dst ),no_rnd_avg32(AV_RN32(src1 ),LPC(src2 )) ); \ |
76 OP(LP(dst+4),no_rnd_avg32(AV_RN32(src1+4),LP(src2+4)) ); \ | 76 OP(LP(dst+4),no_rnd_avg32(AV_RN32(src1+4),LPC(src2+4)) ); \ |
77 src1+=src_stride1; \ | 77 src1+=src_stride1; \ |
78 src2+=src_stride2; \ | 78 src2+=src_stride2; \ |
79 dst+=dst_stride; \ | 79 dst+=dst_stride; \ |
80 } while(--h); \ | 80 } while(--h); \ |
81 }\ | 81 }\ |
82 \ | 82 \ |
83 static inline void OPNAME ## _pixels8_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 83 static inline void OPNAME ## _pixels8_l2_aligned2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
84 {\ | 84 {\ |
85 do {\ | 85 do {\ |
86 OP(LP(dst ),rnd_avg32(AV_RN32(src1 ),LP(src2 )) ); \ | 86 OP(LP(dst ),rnd_avg32(AV_RN32(src1 ),LPC(src2 )) ); \ |
87 OP(LP(dst+4),rnd_avg32(AV_RN32(src1+4),LP(src2+4)) ); \ | 87 OP(LP(dst+4),rnd_avg32(AV_RN32(src1+4),LPC(src2+4)) ); \ |
88 src1+=src_stride1; \ | 88 src1+=src_stride1; \ |
89 src2+=src_stride2; \ | 89 src2+=src_stride2; \ |
90 dst+=dst_stride; \ | 90 dst+=dst_stride; \ |
91 } while(--h); \ | 91 } while(--h); \ |
92 }\ | 92 }\ |
93 \ | 93 \ |
94 static inline void OPNAME ## _no_rnd_pixels8_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 94 static inline void OPNAME ## _no_rnd_pixels8_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
95 {\ | 95 {\ |
96 do {\ | 96 do {\ |
97 OP(LP(dst ),no_rnd_avg32(LP(src1 ),LP(src2 )) ); \ | 97 OP(LP(dst ),no_rnd_avg32(LPC(src1 ),LPC(src2 )) ); \ |
98 OP(LP(dst+4),no_rnd_avg32(LP(src1+4),LP(src2+4)) ); \ | 98 OP(LP(dst+4),no_rnd_avg32(LPC(src1+4),LPC(src2+4)) ); \ |
99 src1+=src_stride1; \ | 99 src1+=src_stride1; \ |
100 src2+=src_stride2; \ | 100 src2+=src_stride2; \ |
101 dst+=dst_stride; \ | 101 dst+=dst_stride; \ |
102 } while(--h); \ | 102 } while(--h); \ |
103 }\ | 103 }\ |
104 \ | 104 \ |
105 static inline void OPNAME ## _pixels8_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 105 static inline void OPNAME ## _pixels8_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
106 {\ | 106 {\ |
107 do {\ | 107 do {\ |
108 OP(LP(dst ),rnd_avg32(LP(src1 ),LP(src2 )) ); \ | 108 OP(LP(dst ),rnd_avg32(LPC(src1 ),LPC(src2 )) ); \ |
109 OP(LP(dst+4),rnd_avg32(LP(src1+4),LP(src2+4)) ); \ | 109 OP(LP(dst+4),rnd_avg32(LPC(src1+4),LPC(src2+4)) ); \ |
110 src1+=src_stride1; \ | 110 src1+=src_stride1; \ |
111 src2+=src_stride2; \ | 111 src2+=src_stride2; \ |
112 dst+=dst_stride; \ | 112 dst+=dst_stride; \ |
113 } while(--h); \ | 113 } while(--h); \ |
114 }\ | 114 }\ |
115 \ | 115 \ |
116 static inline void OPNAME ## _no_rnd_pixels16_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 116 static inline void OPNAME ## _no_rnd_pixels16_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
117 {\ | 117 {\ |
118 do {\ | 118 do {\ |
119 OP(LP(dst ),no_rnd_avg32(LP(src1 ),LP(src2 )) ); \ | 119 OP(LP(dst ),no_rnd_avg32(LPC(src1 ),LPC(src2 )) ); \ |
120 OP(LP(dst+4),no_rnd_avg32(LP(src1+4),LP(src2+4)) ); \ | 120 OP(LP(dst+4),no_rnd_avg32(LPC(src1+4),LPC(src2+4)) ); \ |
121 OP(LP(dst+8),no_rnd_avg32(LP(src1+8),LP(src2+8)) ); \ | 121 OP(LP(dst+8),no_rnd_avg32(LPC(src1+8),LPC(src2+8)) ); \ |
122 OP(LP(dst+12),no_rnd_avg32(LP(src1+12),LP(src2+12)) ); \ | 122 OP(LP(dst+12),no_rnd_avg32(LPC(src1+12),LPC(src2+12)) ); \ |
123 src1+=src_stride1; \ | 123 src1+=src_stride1; \ |
124 src2+=src_stride2; \ | 124 src2+=src_stride2; \ |
125 dst+=dst_stride; \ | 125 dst+=dst_stride; \ |
126 } while(--h); \ | 126 } while(--h); \ |
127 }\ | 127 }\ |
128 \ | 128 \ |
129 static inline void OPNAME ## _pixels16_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ | 129 static inline void OPNAME ## _pixels16_l2_aligned(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, int src_stride1, int src_stride2, int h) \ |
130 {\ | 130 {\ |
131 do {\ | 131 do {\ |
132 OP(LP(dst ),rnd_avg32(LP(src1 ),LP(src2 )) ); \ | 132 OP(LP(dst ),rnd_avg32(LPC(src1 ),LPC(src2 )) ); \ |
133 OP(LP(dst+4),rnd_avg32(LP(src1+4),LP(src2+4)) ); \ | 133 OP(LP(dst+4),rnd_avg32(LPC(src1+4),LPC(src2+4)) ); \ |
134 OP(LP(dst+8),rnd_avg32(LP(src1+8),LP(src2+8)) ); \ | 134 OP(LP(dst+8),rnd_avg32(LPC(src1+8),LPC(src2+8)) ); \ |
135 OP(LP(dst+12),rnd_avg32(LP(src1+12),LP(src2+12)) ); \ | 135 OP(LP(dst+12),rnd_avg32(LPC(src1+12),LPC(src2+12)) ); \ |
136 src1+=src_stride1; \ | 136 src1+=src_stride1; \ |
137 src2+=src_stride2; \ | 137 src2+=src_stride2; \ |
138 dst+=dst_stride; \ | 138 dst+=dst_stride; \ |
139 } while(--h); \ | 139 } while(--h); \ |
140 }\ | 140 }\ |
152 { OPNAME ## _pixels8_l2_aligned2(dst,src2,src1,dst_stride,src_stride2,src_stride1,h); } \ | 152 { OPNAME ## _pixels8_l2_aligned2(dst,src2,src1,dst_stride,src_stride2,src_stride1,h); } \ |
153 \ | 153 \ |
154 static inline void OPNAME ## _pixels8_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 154 static inline void OPNAME ## _pixels8_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
155 do { \ | 155 do { \ |
156 uint32_t a0,a1,a2,a3; \ | 156 uint32_t a0,a1,a2,a3; \ |
157 UNPACK(a0,a1,LP(src1),LP(src2)); \ | 157 UNPACK(a0,a1,LPC(src1),LPC(src2)); \ |
158 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 158 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
159 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ | 159 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ |
160 UNPACK(a0,a1,LP(src1+4),LP(src2+4)); \ | 160 UNPACK(a0,a1,LPC(src1+4),LPC(src2+4)); \ |
161 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 161 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
162 OP(LP(dst+4),rnd_PACK(a0,a1,a2,a3)); \ | 162 OP(LP(dst+4),rnd_PACK(a0,a1,a2,a3)); \ |
163 src1+=src_stride1;\ | 163 src1+=src_stride1;\ |
164 src2+=src_stride2;\ | 164 src2+=src_stride2;\ |
165 src3+=src_stride3;\ | 165 src3+=src_stride3;\ |
166 src4+=src_stride4;\ | 166 src4+=src_stride4;\ |
169 } \ | 169 } \ |
170 \ | 170 \ |
171 static inline void OPNAME ## _no_rnd_pixels8_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 171 static inline void OPNAME ## _no_rnd_pixels8_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
172 do { \ | 172 do { \ |
173 uint32_t a0,a1,a2,a3; \ | 173 uint32_t a0,a1,a2,a3; \ |
174 UNPACK(a0,a1,LP(src1),LP(src2)); \ | 174 UNPACK(a0,a1,LPC(src1),LPC(src2)); \ |
175 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 175 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
176 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ | 176 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ |
177 UNPACK(a0,a1,LP(src1+4),LP(src2+4)); \ | 177 UNPACK(a0,a1,LPC(src1+4),LPC(src2+4)); \ |
178 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 178 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
179 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ | 179 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ |
180 src1+=src_stride1;\ | 180 src1+=src_stride1;\ |
181 src2+=src_stride2;\ | 181 src2+=src_stride2;\ |
182 src3+=src_stride3;\ | 182 src3+=src_stride3;\ |
183 src4+=src_stride4;\ | 183 src4+=src_stride4;\ |
186 } \ | 186 } \ |
187 \ | 187 \ |
188 static inline void OPNAME ## _pixels8_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 188 static inline void OPNAME ## _pixels8_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
189 do { \ | 189 do { \ |
190 uint32_t a0,a1,a2,a3; /* src1 only not aligned */\ | 190 uint32_t a0,a1,a2,a3; /* src1 only not aligned */\ |
191 UNPACK(a0,a1,AV_RN32(src1),LP(src2)); \ | 191 UNPACK(a0,a1,AV_RN32(src1),LPC(src2)); \ |
192 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 192 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
193 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ | 193 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ |
194 UNPACK(a0,a1,AV_RN32(src1+4),LP(src2+4)); \ | 194 UNPACK(a0,a1,AV_RN32(src1+4),LPC(src2+4)); \ |
195 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 195 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
196 OP(LP(dst+4),rnd_PACK(a0,a1,a2,a3)); \ | 196 OP(LP(dst+4),rnd_PACK(a0,a1,a2,a3)); \ |
197 src1+=src_stride1;\ | 197 src1+=src_stride1;\ |
198 src2+=src_stride2;\ | 198 src2+=src_stride2;\ |
199 src3+=src_stride3;\ | 199 src3+=src_stride3;\ |
200 src4+=src_stride4;\ | 200 src4+=src_stride4;\ |
203 } \ | 203 } \ |
204 \ | 204 \ |
205 static inline void OPNAME ## _no_rnd_pixels8_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 205 static inline void OPNAME ## _no_rnd_pixels8_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
206 do { \ | 206 do { \ |
207 uint32_t a0,a1,a2,a3; \ | 207 uint32_t a0,a1,a2,a3; \ |
208 UNPACK(a0,a1,AV_RN32(src1),LP(src2)); \ | 208 UNPACK(a0,a1,AV_RN32(src1),LPC(src2)); \ |
209 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 209 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
210 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ | 210 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ |
211 UNPACK(a0,a1,AV_RN32(src1+4),LP(src2+4)); \ | 211 UNPACK(a0,a1,AV_RN32(src1+4),LPC(src2+4)); \ |
212 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 212 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
213 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ | 213 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ |
214 src1+=src_stride1;\ | 214 src1+=src_stride1;\ |
215 src2+=src_stride2;\ | 215 src2+=src_stride2;\ |
216 src3+=src_stride3;\ | 216 src3+=src_stride3;\ |
217 src4+=src_stride4;\ | 217 src4+=src_stride4;\ |
220 } \ | 220 } \ |
221 \ | 221 \ |
222 static inline void OPNAME ## _pixels16_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 222 static inline void OPNAME ## _pixels16_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
223 do { \ | 223 do { \ |
224 uint32_t a0,a1,a2,a3; \ | 224 uint32_t a0,a1,a2,a3; \ |
225 UNPACK(a0,a1,LP(src1),LP(src2)); \ | 225 UNPACK(a0,a1,LPC(src1),LPC(src2)); \ |
226 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 226 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
227 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ | 227 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ |
228 UNPACK(a0,a1,LP(src1+4),LP(src2+4)); \ | 228 UNPACK(a0,a1,LPC(src1+4),LPC(src2+4)); \ |
229 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 229 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
230 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ | 230 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ |
231 UNPACK(a0,a1,LP(src1+8),LP(src2+8)); \ | 231 UNPACK(a0,a1,LPC(src1+8),LPC(src2+8)); \ |
232 UNPACK(a2,a3,LP(src3+8),LP(src4+8)); \ | 232 UNPACK(a2,a3,LPC(src3+8),LPC(src4+8)); \ |
233 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ | 233 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ |
234 UNPACK(a0,a1,LP(src1+12),LP(src2+12)); \ | 234 UNPACK(a0,a1,LPC(src1+12),LPC(src2+12)); \ |
235 UNPACK(a2,a3,LP(src3+12),LP(src4+12)); \ | 235 UNPACK(a2,a3,LPC(src3+12),LPC(src4+12)); \ |
236 OP(LP(dst+12),rnd_PACK(a0,a1,a2,a3)); \ | 236 OP(LP(dst+12),rnd_PACK(a0,a1,a2,a3)); \ |
237 src1+=src_stride1;\ | 237 src1+=src_stride1;\ |
238 src2+=src_stride2;\ | 238 src2+=src_stride2;\ |
239 src3+=src_stride3;\ | 239 src3+=src_stride3;\ |
240 src4+=src_stride4;\ | 240 src4+=src_stride4;\ |
243 } \ | 243 } \ |
244 \ | 244 \ |
245 static inline void OPNAME ## _no_rnd_pixels16_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 245 static inline void OPNAME ## _no_rnd_pixels16_l4_aligned(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
246 do { \ | 246 do { \ |
247 uint32_t a0,a1,a2,a3; \ | 247 uint32_t a0,a1,a2,a3; \ |
248 UNPACK(a0,a1,LP(src1),LP(src2)); \ | 248 UNPACK(a0,a1,LPC(src1),LPC(src2)); \ |
249 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 249 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
250 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ | 250 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ |
251 UNPACK(a0,a1,LP(src1+4),LP(src2+4)); \ | 251 UNPACK(a0,a1,LPC(src1+4),LPC(src2+4)); \ |
252 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 252 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
253 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ | 253 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ |
254 UNPACK(a0,a1,LP(src1+8),LP(src2+8)); \ | 254 UNPACK(a0,a1,LPC(src1+8),LPC(src2+8)); \ |
255 UNPACK(a2,a3,LP(src3+8),LP(src4+8)); \ | 255 UNPACK(a2,a3,LPC(src3+8),LPC(src4+8)); \ |
256 OP(LP(dst+8),no_rnd_PACK(a0,a1,a2,a3)); \ | 256 OP(LP(dst+8),no_rnd_PACK(a0,a1,a2,a3)); \ |
257 UNPACK(a0,a1,LP(src1+12),LP(src2+12)); \ | 257 UNPACK(a0,a1,LPC(src1+12),LPC(src2+12)); \ |
258 UNPACK(a2,a3,LP(src3+12),LP(src4+12)); \ | 258 UNPACK(a2,a3,LPC(src3+12),LPC(src4+12)); \ |
259 OP(LP(dst+12),no_rnd_PACK(a0,a1,a2,a3)); \ | 259 OP(LP(dst+12),no_rnd_PACK(a0,a1,a2,a3)); \ |
260 src1+=src_stride1;\ | 260 src1+=src_stride1;\ |
261 src2+=src_stride2;\ | 261 src2+=src_stride2;\ |
262 src3+=src_stride3;\ | 262 src3+=src_stride3;\ |
263 src4+=src_stride4;\ | 263 src4+=src_stride4;\ |
266 } \ | 266 } \ |
267 \ | 267 \ |
268 static inline void OPNAME ## _pixels16_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 268 static inline void OPNAME ## _pixels16_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
269 do { /* src1 is unaligned */\ | 269 do { /* src1 is unaligned */\ |
270 uint32_t a0,a1,a2,a3; \ | 270 uint32_t a0,a1,a2,a3; \ |
271 UNPACK(a0,a1,AV_RN32(src1),LP(src2)); \ | 271 UNPACK(a0,a1,AV_RN32(src1),LPC(src2)); \ |
272 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 272 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
273 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ | 273 OP(LP(dst),rnd_PACK(a0,a1,a2,a3)); \ |
274 UNPACK(a0,a1,AV_RN32(src1+4),LP(src2+4)); \ | 274 UNPACK(a0,a1,AV_RN32(src1+4),LPC(src2+4)); \ |
275 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 275 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
276 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ | 276 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ |
277 UNPACK(a0,a1,AV_RN32(src1+8),LP(src2+8)); \ | 277 UNPACK(a0,a1,AV_RN32(src1+8),LPC(src2+8)); \ |
278 UNPACK(a2,a3,LP(src3+8),LP(src4+8)); \ | 278 UNPACK(a2,a3,LPC(src3+8),LPC(src4+8)); \ |
279 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ | 279 OP(LP(dst+8),rnd_PACK(a0,a1,a2,a3)); \ |
280 UNPACK(a0,a1,AV_RN32(src1+12),LP(src2+12)); \ | 280 UNPACK(a0,a1,AV_RN32(src1+12),LPC(src2+12)); \ |
281 UNPACK(a2,a3,LP(src3+12),LP(src4+12)); \ | 281 UNPACK(a2,a3,LPC(src3+12),LPC(src4+12)); \ |
282 OP(LP(dst+12),rnd_PACK(a0,a1,a2,a3)); \ | 282 OP(LP(dst+12),rnd_PACK(a0,a1,a2,a3)); \ |
283 src1+=src_stride1;\ | 283 src1+=src_stride1;\ |
284 src2+=src_stride2;\ | 284 src2+=src_stride2;\ |
285 src3+=src_stride3;\ | 285 src3+=src_stride3;\ |
286 src4+=src_stride4;\ | 286 src4+=src_stride4;\ |
289 } \ | 289 } \ |
290 \ | 290 \ |
291 static inline void OPNAME ## _no_rnd_pixels16_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ | 291 static inline void OPNAME ## _no_rnd_pixels16_l4_aligned0(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
292 do { \ | 292 do { \ |
293 uint32_t a0,a1,a2,a3; \ | 293 uint32_t a0,a1,a2,a3; \ |
294 UNPACK(a0,a1,AV_RN32(src1),LP(src2)); \ | 294 UNPACK(a0,a1,AV_RN32(src1),LPC(src2)); \ |
295 UNPACK(a2,a3,LP(src3),LP(src4)); \ | 295 UNPACK(a2,a3,LPC(src3),LPC(src4)); \ |
296 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ | 296 OP(LP(dst),no_rnd_PACK(a0,a1,a2,a3)); \ |
297 UNPACK(a0,a1,AV_RN32(src1+4),LP(src2+4)); \ | 297 UNPACK(a0,a1,AV_RN32(src1+4),LPC(src2+4)); \ |
298 UNPACK(a2,a3,LP(src3+4),LP(src4+4)); \ | 298 UNPACK(a2,a3,LPC(src3+4),LPC(src4+4)); \ |
299 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ | 299 OP(LP(dst+4),no_rnd_PACK(a0,a1,a2,a3)); \ |
300 UNPACK(a0,a1,AV_RN32(src1+8),LP(src2+8)); \ | 300 UNPACK(a0,a1,AV_RN32(src1+8),LPC(src2+8)); \ |
301 UNPACK(a2,a3,LP(src3+8),LP(src4+8)); \ | 301 UNPACK(a2,a3,LPC(src3+8),LPC(src4+8)); \ |
302 OP(LP(dst+8),no_rnd_PACK(a0,a1,a2,a3)); \ | 302 OP(LP(dst+8),no_rnd_PACK(a0,a1,a2,a3)); \ |
303 UNPACK(a0,a1,AV_RN32(src1+12),LP(src2+12)); \ | 303 UNPACK(a0,a1,AV_RN32(src1+12),LPC(src2+12)); \ |
304 UNPACK(a2,a3,LP(src3+12),LP(src4+12)); \ | 304 UNPACK(a2,a3,LPC(src3+12),LPC(src4+12)); \ |
305 OP(LP(dst+12),no_rnd_PACK(a0,a1,a2,a3)); \ | 305 OP(LP(dst+12),no_rnd_PACK(a0,a1,a2,a3)); \ |
306 src1+=src_stride1;\ | 306 src1+=src_stride1;\ |
307 src2+=src_stride2;\ | 307 src2+=src_stride2;\ |
308 src3+=src_stride3;\ | 308 src3+=src_stride3;\ |
309 src4+=src_stride4;\ | 309 src4+=src_stride4;\ |