notaz.gp2x.de / gpsp.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
use warm for cache control, avoid invalidating all icache
[gpsp.git] / gp2x / upscale_aspect.s
CommitLineData
788343bb 1/*
2 * 240x160 -> 320x213 upscaler for ARM with interpolation
3 *
4 * Written by GraÅžvydas "notaz" Ignotas
5 * Prototyped by Rokas
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions are met:
9 * * Redistributions of source code must retain the above copyright
10 * notice, this list of conditions and the following disclaimer.
11 * * Redistributions in binary form must reproduce the above copyright
12 * notice, this list of conditions and the following disclaimer in the
13 * documentation and/or other materials provided with the distribution.
14 * * Neither the name of the organization nor the
15 * names of its contributors may be used to endorse or promote products
16 * derived from this software without specific prior written permission.
17 *
18 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
19 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
20 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
21 * ARE DISCLAIMED. IN NO EVENT SHALL COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
22 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
23 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
24 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
25 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
26 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
27 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
28 *
29 * 0 1 2 : 3 4 5
30 * 6 7 8 : 9 10 11
31 * 12 13 14 : 15 16 17
32 * v
33 * 0 1 2 3 : 4 5 6 7
34 * 8 9 10 11 : 12 13 14 15
35 * 16 17 18 19 : 20 21 22 23
36 * 24 25 26 27 : 28 29 30 31
37 */
38
39.macro unpack_hi dst, src
40 mov \dst, \src, lsr #16
41 orr \dst, \dst, lsl #16
42 and \dst, \dst, lr
43.endm
44
45.macro unpack_lo dst, src
46 mov \dst, \src, lsl #16
47 orr \dst, \dst, lsr #16
48 and \dst, \dst, lr
49.endm
50
51@ do 3:5 summing: r2 = (s1*3 + s2*5 + 4) / 8
52@ s2 != r2
53.macro do_3_5 s1, s2
54 add r2,\s1,\s1, lsl #1 @ r2 = s1 * 3
55 add r2, r2,\s2, lsl #2
56 add r2, r2,\s2 @ r2 += s2 * 5
57 add r2, r2, r12,lsl #2 @ sum += round * 4
58 and r2, lr, r2, lsr #3 @ mask_to_unpacked(sum / 8)
59.endm
60
61@ do 14:7:7:4: r2 = (s1*14 + s2*7 + s3*7 + s4*4 + 16) / 32
62@ {s2,s3,s4} != r2
63.macro do_14_7_7_4 s1, s2, s3, s4
64 mov r2,\s1, lsl #4
65 sub r2, r2,\s1, lsl #1 @ r2 = s1 * 14
66 add r2, r2,\s2, lsl #3
67 sub r2, r2,\s2 @ r2 += s2 * 7
68 add r2, r2,\s3, lsl #3
69 sub r2, r2,\s3 @ r2 += s3 * 7
70 add r2, r2,\s4, lsl #2 @ r2 += s4 * 4
71 add r2, r2, r12,lsl #3 @ sum += round * 16
72 and r2, lr, r2, lsr #5 @ mask_to_unpacked(sum / 32)
73.endm
74
75.global upscale_aspect @ u16 *dst, u16 *src
76upscale_aspect:
77 stmfd sp!,{r4-r11,lr}
78 mov lr, #0x0000001f
79 orr lr, lr, #0x0000f800 @ for "unpacked" form of
80 orr lr, lr, #0x07e00000 @ 00000ggg'ggg00000'rrrrr000'000bbbbb
81 mov r12, #0x00000001
82 orr r12,r12,#0x00000800
83 orr r12,r12,#0x00200000 @ rounding constant
84
85 mov r8, #((240/6)-1) << 24 @ cols
86 orr r8, r8, #160/3 @ rows
87
88 add r0, r0, #320*2*13
89loop1:
90 ldr r10,[r1]
91 ldr r11,[r1, #320*2*1]
92
93 unpack_lo r4, r10
94 unpack_hi r5, r10
95 unpack_lo r6, r11
96 unpack_hi r7, r11
97
98 ldr r11,[r1, #4]
99
100 do_3_5 r4, r5
101 orr r2, r2, r2, lsr #16
102 mov r3, r10, lsl #16
103 mov r3, r3, lsr #16
104 orr r2, r3, r2, lsl #16
105 str r2, [r0] @ 0,1
106
107 ldr r10,[r1, #320*2*2]
108
109 do_3_5 r4, r6
110 orr r3, r2, r2, lsl #16
111 mov r3, r3, lsr #16 @ 8
112
113 do_14_7_7_4 r7, r5, r6, r4
114 orr r2, r2, r2, lsr #16
115 orr r2, r3, r2, lsl #16
116 str r2, [r0, #320*2*1] @ 8,9
117
118 unpack_lo r4, r10
119 unpack_hi r9, r10
120
121 do_3_5 r4, r6
122 orr r3, r2, r2, lsl #16
123 mov r3, r3, lsr #16
124
125 do_14_7_7_4 r7, r9, r6, r4
126 orr r2, r2, r2, lsr #16
127 orr r2, r3, r2, lsl #16
128 str r2, [r0, #320*2*2] @ 16,17
129
130 do_3_5 r4, r9
131 orr r2, r2, r2, lsr #16
132 mov r3, r10, lsl #16
133 mov r3, r3, lsr #16
134 orr r2, r3, r2, lsl #16
135 str r2, [r0, #320*2*3] @ 24,25
136
137 ldr r10,[r1, #320*2*1+4]
138
139 unpack_lo r6, r11
140 unpack_lo r4, r10
141
142 do_3_5 r6, r5
143 orr r2, r2, r2, lsl #16
144 mov r3, r11, lsl #16
145 orr r2, r3, r2, lsr #16
146 str r2, [r0, #4] @ 2,3
147
148 do_14_7_7_4 r7, r4, r5, r6
149 orr r2, r2, r2, lsl #16
150 mov r3, r2, lsr #16
151
152 ldr r5, [r1, #320*2*2+4]
153
154 do_3_5 r6, r4
155 orr r2, r2, r2, lsr #16
156 orr r2, r3, r2, lsl #16
157 str r2, [r0, #320*2*1+4] @ 10,11
158
159 unpack_lo r6, r5
160
161 do_14_7_7_4 r7, r4, r9, r6
162 orr r2, r2, r2, lsl #16
163 mov r3, r2, lsr #16
164
165 do_3_5 r6, r4
166 orr r2, r2, r2, lsr #16
167 orr r2, r3, r2, lsl #16
168 str r2, [r0, #320*2*2+4] @ 18,19
169
170 unpack_hi r4, r10
171
172 ldr r10,[r1, #8]
173
174 do_3_5 r6, r9
175 orr r2, r2, r2, lsl #16
176 mov r2, r2, lsr #16
177 orr r2, r2, r5, lsl #16
178 str r2, [r0, #320*2*3+4] @ 26,27
179
180 unpack_hi r6, r11
181 unpack_lo r7, r10
182
183 do_3_5 r6, r7
184 orr r2, r2, r2, lsr #16
185 mov r2, r2, lsl #16
186 orr r2, r2, r11,lsr #16
187 str r2, [r0, #8] @ 4,5
188
189 ldr r11,[r1, #320*2*1+8]
190
191 unpack_hi r9, r10
192
193 do_3_5 r9, r7
194 orr r2, r2, r2, lsr #16
195 mov r2, r2, lsl #16
196 orr r2, r2, r10,lsr #16
197 mov r2, r2, ror #16
198 str r2, [r0, #12] @ 6,7
199
200 unpack_lo r10,r11
201
202 do_3_5 r6, r4
203 orr r2, r2, r2, lsl #16
204 mov r3, r2, lsr #16
205
206 do_14_7_7_4 r10, r4, r7, r6
207 orr r2, r2, r2, lsr #16
208 orr r2, r3, r2, lsl #16
209 str r2, [r0, #320*2*1+8] @ 12,13
210
211 unpack_hi r6, r11
212
213 ldr r11,[r1, #320*2*2+8]
214
215 do_14_7_7_4 r10, r6, r7, r9
216 orr r2, r2, r2, lsl #16
217 mov r3, r2, lsr #16
218
219 do_3_5 r9, r6
220 orr r2, r2, r2, lsr #16
221 orr r2, r3, r2, lsl #16
222 str r2, [r0, #320*2*1+12] @ 14,15
223
224 unpack_hi r7, r5
225 unpack_lo r9, r11
226
227 do_3_5 r7, r4
228 orr r2, r2, r2, lsl #16
229 mov r3, r2, lsr #16
230
231 do_14_7_7_4 r10, r4, r9, r7
232 orr r2, r2, r2, lsr #16
233 orr r2, r3, r2, lsl #16
234 str r2, [r0, #320*2*2+8] @ 20,21
235
236 do_3_5 r7, r9
237 orr r2, r2, r2, lsr #16
238 mov r2, r2, lsl #16
239 orr r2, r2, r5, lsr #16
240 str r2, [r0, #320*2*3+8] @ 28,29
241
242 unpack_hi r5, r11
243
244 do_14_7_7_4 r10, r6, r9, r5
245 orr r2, r2, r2, lsl #16
246 mov r3, r2, lsr #16
247
248 do_3_5 r5, r6
249 orr r2, r2, r2, lsr #16
250 orr r2, r3, r2, lsl #16
251 str r2, [r0, #320*2*2+12] @ 22,23
252
253 do_3_5 r5, r9
254 orr r2, r2, r2, lsr #16
255 mov r3, r11, lsr #16
256 orr r2, r3, r2, lsl #16
257 mov r2, r2, ror #16
258 str r2, [r0, #320*2*3+12] @ 30,31
259
260 add r0, r0, #16
261 add r1, r1, #12
262
263 subs r8, r8, #1<<24
264 bpl loop1
265
266 add r0, r0, #320*3*2
267 add r1, r1, #(320*2+80)*2
268 sub r8, r8, #1
269 tst r8, #0xff
270 add r8, r8, #(240/6) << 24 @ cols
271 bne loop1
272
273 @@ last line
274 mov r8, #240/6
275
276loop2:
277 ldmia r1!,{r9,r10,r11}
278
279 unpack_lo r4, r9
280 unpack_hi r5, r9
281
282 do_3_5 r4, r5
283 orr r2, r2, r2, lsr #16
284 mov r3, r9, lsl #16
285 mov r3, r3, lsr #16
286 orr r2, r3, r2, lsl #16
287 str r2, [r0], #4
288
289 unpack_lo r6, r10
290 unpack_hi r7, r10
291
292 do_3_5 r6, r5
293 orr r2, r2, r2, lsl #16
294 mov r2, r2, lsr #16
295 orr r2, r2, r10,lsl #16
296 str r2, [r0], #4
297
298 unpack_lo r4, r11
299 unpack_hi r5, r11
300
301 do_3_5 r7, r4
302 orr r2, r2, r2, lsr #16
303 mov r3, r10, lsr #16
304 orr r2, r3, r2, lsl #16
305 str r2, [r0], #4
306
307 do_3_5 r5, r4
308 orr r2, r2, r2, lsr #16
309 mov r3, r11, lsr #16
310 orr r2, r3, r2, lsl #16
311 mov r2, r2, ror #16
312 str r2, [r0], #4
313
314 subs r8, r8, #1
315 bne loop2
316
317 ldmfd sp!,{r4-r11,pc}
318
319@ vim:filetype=armasm
320
gpSP fork tuned for ARM Linux