1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
|
@ imagescalearm.s is a hand tuned assembler version
@ of some of the imagescale functions targetted
@ for ARM based systems (any architecture version).
@
@ Copyright (C) 2010 Robin Watts for Artifex Software LLC.
@ <robin.watts@artifex.com> or <robin.watts@wss.co.uk>
.file "imagescalearm.s"
.global fz_srow4_arm
.global fz_scol4_arm
.type fz_srow4_arm, %function
.text
@ r0 = src
@ r1 = dst
@ r2 = w
@ r3 = denom
fz_srow4_arm:
STMFD r13!,{r3-r7,r9-r10,r14}
MOV r12,#1<<16 @ r12 = (1<<16)
MOV r14,#0 @ r14 = will contain invdenom
@ r14 = r12/r3
CMP r12,r3, LSL #16
SUBHS r12,r12,r3, LSL #16
ADDHS r14,r14,#1<<16
CMP r12,r3, LSL #15
SUBHS r12,r12,r3, LSL #15
ADDHS r14,r14,#1<<15
CMP r12,r3, LSL #14
SUBHS r12,r12,r3, LSL #14
ADDHS r14,r14,#1<<14
CMP r12,r3, LSL #13
SUBHS r12,r12,r3, LSL #13
ADDHS r14,r14,#1<<13
CMP r12,r3, LSL #12
SUBHS r12,r12,r3, LSL #12
ADDHS r14,r14,#1<<12
CMP r12,r3, LSL #11
SUBHS r12,r12,r3, LSL #11
ADDHS r14,r14,#1<<11
CMP r12,r3, LSL #10
SUBHS r12,r12,r3, LSL #10
ADDHS r14,r14,#1<<10
CMP r12,r3, LSL #9
SUBHS r12,r12,r3, LSL #9
ADDHS r14,r14,#1<<9
CMP r12,r3, LSL #8
SUBHS r12,r12,r3, LSL #8
ADDHS r14,r14,#1<<8
CMP r12,r3, LSL #7
SUBHS r12,r12,r3, LSL #7
ADDHS r14,r14,#1<<7
CMP r12,r3, LSL #6
SUBHS r12,r12,r3, LSL #6
ADDHS r14,r14,#1<<6
CMP r12,r3, LSL #5
SUBHS r12,r12,r3, LSL #5
ADDHS r14,r14,#1<<5
CMP r12,r3, LSL #4
SUBHS r12,r12,r3, LSL #4
ADDHS r14,r14,#1<<4
CMP r12,r3, LSL #3
SUBHS r12,r12,r3, LSL #3
ADDHS r14,r14,#1<<3
CMP r12,r3, LSL #2
SUBHS r12,r12,r3, LSL #2
ADDHS r14,r14,#1<<2
CMP r12,r3, LSL #1
SUBHS r12,r12,r3, LSL #1
ADDHS r14,r14,#1<<1
CMP r12,r3
SUBHS r12,r12,r3
ADDHS r14,r14,#1
@ r2 = x = w
@ r3 = left = denom
MOV r10,#1<<15 @ r10= 1<<15
B .L_enter_loop_r4 @ Enter the loop
.L_store_r4:
ADD r7, r7, r12 @ r7 = sum3 += r12
MLA r4, r14,r4, r10 @ r4 = sum0 * invdenom + (1<<15)
MLA r5, r14,r5, r10 @ r5 = sum1 * invdenom + (1<<15)
MLA r6, r14,r6, r10 @ r6 = sum2 * invdenom + (1<<15)
MLA r7, r14,r7, r10 @ r7 = sum3 * invdenom + (1<<15)
MOV r4, r4, LSR #16 @ r4 = r4 >> 16
MOV r5, r5, LSR #16 @ r5 = r5 >> 16
MOV r6, r6, LSR #16 @ r6 = r6 >> 16
MOV r7, r7, LSR #16 @ r7 = r7 >> 16
STRB r4, [r1], #1 @ *dst++ = r4
STRB r5, [r1], #1 @ *dst++ = r5
STRB r6, [r1], #1 @ *dst++ = r6
STRB r7, [r1], #1 @ *dst++ = r7
SUBS r2, r2, #1 @ x--
BEQ .L_end_r4
LDR r3, [r13] @ r3 = left = denom
.L_enter_loop_r4:
MOV r4, #0 @ r4 = sum0 = 0
MOV r5, #0 @ r5 = sum1 = 0
MOV r6, #0 @ r6 = sum2 = 0
MOV r7, #0 @ r7 = sum3 = 0
.L_x_loop_r4:
LDRB r9, [r0], #1 @ r9 = src++
LDRB r12,[r0], #1 @ r12= src++
SUBS r3, r3, #1 @ r3 = --left
ADD r4, r4, r9 @ r4 = sum0 += r9
LDRB r9, [r0], #1 @ r9 = src++
ADD r5, r5, r12 @ r5 = sum1 += r12
LDRB r12,[r0], #1 @ r12= src++
ADD r6, r6, r9 @ r9 = sum2 += r9
BEQ .L_store_r4
ADD r7, r7, r12 @ r7 = sum3 += r12
SUBS r2, r2, #1 @ x--
BNE .L_x_loop_r4
@ Trailers
LDR r0, [r13] @ r0 = denom
MOV r12,#1<<16 @ r12 = (1<<16)
MOV r14,#0 @ r14 = will contain invleft
SUB r3, r0, r3 @ r3 = denom-left
CMP r12,r3, LSL #16
SUBHS r12,r12,r3, LSL #16
ADDHS r14,r14,#1<<16
CMP r12,r3, LSL #15
SUBHS r12,r12,r3, LSL #15
ADDHS r14,r14,#1<<15
CMP r12,r3, LSL #14
SUBHS r12,r12,r3, LSL #14
ADDHS r14,r14,#1<<14
CMP r12,r3, LSL #13
SUBHS r12,r12,r3, LSL #13
ADDHS r14,r14,#1<<13
CMP r12,r3, LSL #12
SUBHS r12,r12,r3, LSL #12
ADDHS r14,r14,#1<<12
CMP r12,r3, LSL #11
SUBHS r12,r12,r3, LSL #11
ADDHS r14,r14,#1<<11
CMP r12,r3, LSL #10
SUBHS r12,r12,r3, LSL #10
ADDHS r14,r14,#1<<10
CMP r12,r3, LSL #9
SUBHS r12,r12,r3, LSL #9
ADDHS r14,r14,#1<<9
CMP r12,r3, LSL #8
SUBHS r12,r12,r3, LSL #8
ADDHS r14,r14,#1<<8
CMP r12,r3, LSL #7
SUBHS r12,r12,r3, LSL #7
ADDHS r14,r14,#1<<7
CMP r12,r3, LSL #6
SUBHS r12,r12,r3, LSL #6
ADDHS r14,r14,#1<<6
CMP r12,r3, LSL #5
SUBHS r12,r12,r3, LSL #5
ADDHS r14,r14,#1<<5
CMP r12,r3, LSL #4
SUBHS r12,r12,r3, LSL #4
ADDHS r14,r14,#1<<4
CMP r12,r3, LSL #3
SUBHS r12,r12,r3, LSL #3
ADDHS r14,r14,#1<<3
CMP r12,r3, LSL #2
SUBHS r12,r12,r3, LSL #2
ADDHS r14,r14,#1<<2
CMP r12,r3, LSL #1
SUBHS r12,r12,r3, LSL #1
ADDHS r14,r14,#1<<1
CMP r12,r3
SUBHS r12,r12,r3
ADDHS r14,r14,#1
MLA r4, r14,r4, r10 @ r4 = sum0 * invleft + (1<<15)
MLA r5, r14,r5, r10 @ r5 = sum1 * invleft + (1<<15)
MLA r6, r14,r6, r10 @ r6 = sum2 * invleft + (1<<15)
MLA r7, r14,r7, r10 @ r7 = sum3 * invleft + (1<<15)
MOV r4, r4, LSR #16 @ r4 = r4 >> 16
MOV r5, r5, LSR #16 @ r5 = r5 >> 16
MOV r6, r6, LSR #16 @ r6 = r6 >> 16
MOV r7, r7, LSR #16 @ r7 = r7 >> 16
STRB r4, [r1], #1 @ *dst++ = r4
STRB r5, [r1], #1 @ *dst++ = r5
STRB r6, [r1], #1 @ *dst++ = r6
STRB r7, [r1], #1 @ *dst++ = r7
.L_end_r4:
LDMFD r13!,{r3-r7,r9-r10,PC}
.fnend
.size fz_srow4_arm, .-fz_srow4_arm
.type fz_scol4_arm, %function
.text
@ r0 = src
@ r1 = dst
@ r2 = w
@ r3 = denom
fz_scol4_arm:
STMFD r13!,{r3-r7,r9-r11,r14}
MOV r12,#1<<16 @ r12 = (1<<16)
MOV r14,#0 @ r14 = will contain invdenom
@ r14 = r12/r3
CMP r12,r3, LSL #16
SUBHS r12,r12,r3, LSL #16
ADDHS r14,r14,#1<<16
CMP r12,r3, LSL #15
SUBHS r12,r12,r3, LSL #15
ADDHS r14,r14,#1<<15
CMP r12,r3, LSL #14
SUBHS r12,r12,r3, LSL #14
ADDHS r14,r14,#1<<14
CMP r12,r3, LSL #13
SUBHS r12,r12,r3, LSL #13
ADDHS r14,r14,#1<<13
CMP r12,r3, LSL #12
SUBHS r12,r12,r3, LSL #12
ADDHS r14,r14,#1<<12
CMP r12,r3, LSL #11
SUBHS r12,r12,r3, LSL #11
ADDHS r14,r14,#1<<11
CMP r12,r3, LSL #10
SUBHS r12,r12,r3, LSL #10
ADDHS r14,r14,#1<<10
CMP r12,r3, LSL #9
SUBHS r12,r12,r3, LSL #9
ADDHS r14,r14,#1<<9
CMP r12,r3, LSL #8
SUBHS r12,r12,r3, LSL #8
ADDHS r14,r14,#1<<8
CMP r12,r3, LSL #7
SUBHS r12,r12,r3, LSL #7
ADDHS r14,r14,#1<<7
CMP r12,r3, LSL #6
SUBHS r12,r12,r3, LSL #6
ADDHS r14,r14,#1<<6
CMP r12,r3, LSL #5
SUBHS r12,r12,r3, LSL #5
ADDHS r14,r14,#1<<5
CMP r12,r3, LSL #4
SUBHS r12,r12,r3, LSL #4
ADDHS r14,r14,#1<<4
CMP r12,r3, LSL #3
SUBHS r12,r12,r3, LSL #3
ADDHS r14,r14,#1<<3
CMP r12,r3, LSL #2
SUBHS r12,r12,r3, LSL #2
ADDHS r14,r14,#1<<2
CMP r12,r3, LSL #1
SUBHS r12,r12,r3, LSL #1
ADDHS r14,r14,#1<<1
CMP r12,r3
SUBHS r12,r12,r3
ADDHS r14,r14,#1
@ r2 = x = w
@ r3 = y = denom
MOV r11,r2, LSL #2 @ r11= w = w*n
RSB r11,r11,#0 @ r11=-w
MOV r10,#1<<15 @ r10= 1<<15
.L_x_loop_c4:
MOV r4, #0 @ r4 = sum0 = 0
MOV r5, #0 @ r5 = sum1 = 0
MOV r6, #0 @ r6 = sum2 = 0
MOV r7, #0 @ r7 = sum3 = 0
.L_y_loop_c4:
LDRB r9, [r0, #1] @ r9 = src[1]
LDRB r12,[r0, #2] @ r12= src[2]
SUBS r3, r3, #1 @ r3 = y--
ADD r5, r5, r9 @ r4 = sum1 += r9
LDRB r9, [r0, #3] @ r9 = src[3]
ADD r6, r6, r12 @ r5 = sum2 += r12
LDRB r12,[r0], -r11 @ r12= src[0] src += w
ADD r7, r7, r9 @ r9 = sum3 += r9
ADD r4, r4, r12 @ r7 = sum0 += r12
BGT .L_y_loop_c4
LDR r3, [r13] @ r3 = y = denom
MLA r4, r14,r4, r10 @ r4 = sum0 * invdenom + (1<<15)
MLA r5, r14,r5, r10 @ r5 = sum1 * invdenom + (1<<15)
MLA r6, r14,r6, r10 @ r6 = sum2 * invdenom + (1<<15)
MLA r7, r14,r7, r10 @ r7 = sum3 * invdenom + (1<<15)
MLA r0, r3, r11,r0 @ r0 = src += -denom*w
MOV r4, r4, LSR #16 @ r4 = r4 >> 16
MOV r5, r5, LSR #16 @ r5 = r5 >> 16
MOV r6, r6, LSR #16 @ r6 = r6 >> 16
MOV r7, r7, LSR #16 @ r7 = r7 >> 16
ADD r0, r0, #4 @ r0 = src += n
STRB r4, [r1], #1 @ *dst++ = r4
STRB r5, [r1], #1 @ *dst++ = r5
STRB r6, [r1], #1 @ *dst++ = r6
STRB r7, [r1], #1 @ *dst++ = r7
SUBS r2, r2, #1 @ x--
BNE .L_x_loop_c4
.L_end_c4:
LDMFD r13!,{r3-r7,r9-r11,PC}
.fnend
.size fz_scol4_arm, .-fz_scol4_arm
|
