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
|
/*
* idct BlackFin
*
* Copyright (C) 2007 Marc Hoffman <marc.hoffman@analog.com>
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/*
This blackfin DSP code implements an 8x8 inverse type II DCT.
Prototype : void ff_bfin_idct(int16_t *in)
Registers Used : A0, A1, R0-R7, I0-I3, B0, B2, B3, M0-M2, L0-L3, P0-P5, LC0.
Performance :
Code Size : 498 Bytes.
Cycle Count : 417 Cycles
-----------------------------------------------------------
FFMPEG conformance testing results
-----------------------------------------------------------
dct-test: modified with the following
dct_error("BFINidct", 1, ff_bfin_idct, idct, test);
produces the following output
root:/u/ffmpeg/bhead/libavcodec> ./dct-test -i
ffmpeg DCT/IDCT test
8 15 -2 21 24 17 0 10
2 -10 -5 -5 -3 7 -14 -3
2 -13 -10 -19 18 -6 6 -2
9 4 16 -3 9 12 10 15
15 -9 -2 10 1 16 0 -15
-15 5 7 3 13 0 13 20
-6 -15 24 9 -18 1 9 -22
-8 25 23 2 -7 0 30 13
IDCT BFINidct: err_inf=1 err2=0.01002344 syserr=0.00150000 maxout=266 blockSumErr=64
IDCT BFINidct: 88.3 kdct/s
*/
#include "libavutil/bfin/asm.h"
SECTION_L1_DATA_B
.align 4;
coefs:
.short 0x5a82; // C4
.short 0x5a82; // C4
.short 0x30FC; //cos(3pi/8) C6
.short 0x7642; //cos(pi/8) C2
.short 0x18F9; //cos(7pi/16)
.short 0x7D8A; //cos(pi/16)
.short 0x471D; //cos(5pi/16)
.short 0x6A6E; //cos(3pi/16)
.short 0x18F9; //cos(7pi/16)
.short 0x7D8A; //cos(pi/16)
SECTION_L1_DATA_A
vtmp: .space 256
#define TMP0 FP-8
#define TMP1 FP-12
#define TMP2 FP-16
.text
DEFUN(idct,mL1,
(int16_t *block)):
/********************** Function Prologue *********************************/
link 16;
[--SP] = (R7:4, P5:3); // Push the registers onto the stack.
B0 = R0; // Pointer to Input matrix
RELOC(R1, P3, coefs); // Pointer to Coefficients
RELOC(R2, P3, vtmp); // Pointer to Temporary matrix
B3 = R1;
B2 = R2;
L3 = 20; // L3 is used for making the coefficient array
// circular.
// MUST BE RESTORED TO ZERO at function exit.
M1 = 16 (X); // All these registers are initialized for
M3 = 8(X); // modifying address offsets.
I0 = B0; // I0 points to Input Element (0, 0).
I2 = B0; // I2 points to Input Element (0, 0).
I2 += M3 || R0.H = W[I0];
// Element 0 is read into R0.H
I1 = I2; // I1 points to input Element (0, 6).
I1 += 4 || R0.L = W[I2++];
// I2 points to input Element (0, 4).
// Element 4 is read into R0.L.
P2 = 8 (X);
P3 = 32 (X);
P4 = -32 (X);
P5 = 98 (X);
R7 = 0x8000(Z);
I3 = B3; // I3 points to Coefficients
P0 = B2; // P0 points to array Element (0, 0) of temp
P1 = B2;
R7 = [I3++] || [TMP2]=R7; // Coefficient C4 is read into R7.H and R7.L.
MNOP;
NOP;
/*
* A1 = Y0 * cos(pi/4)
* A0 = Y0 * cos(pi/4)
* A1 = A1 + Y4 * cos(pi/4)
* A0 = A0 - Y4 * cos(pi/4)
* load:
* R1=(Y2,Y6)
* R7=(C2,C6)
* res:
* R3=Y0, R2=Y4
*/
A1=R7.H*R0.H, A0=R7.H*R0.H (IS) || I0+= 4 || R1.L=W[I1++];
R3=(A1+=R7.H*R0.L), R2=(A0-=R7.H*R0.L) (IS) || R1.H=W[I0--] || R7=[I3++];
LSETUP (.0, .1) LC0 = P2; // perform 8 1d idcts
P2 = 112 (X);
P1 = P1 + P2; // P1 points to element (7, 0) of temp buffer.
P2 = -94(X);
.0:
/*
* A1 = Y2 * cos(3pi/8)
* A0 = Y2 * cos(pi/8)
* A1 = A1 - Y6 * cos(pi/8)
* A0 = A0 + Y6 * cos(3pi/8)
* R5 = (Y1,Y7)
* R7 = (C1,C7)
* res:
* R1=Y2, R0=Y6
*/
A1=R7.L*R1.H, A0=R7.H*R1.H (IS) || I0+=4 || R5.H=W[I0];
R1=(A1-=R7.H*R1.L), R0=(A0+=R7.L*R1.L) (IS) || R5.L=W[I1--] || R7=[I3++];
/*
* Y0 = Y0 + Y6.
* Y4 = Y4 + Y2.
* Y2 = Y4 - Y2.
* Y6 = Y0 - Y6.
* R3 is saved
* R6.l=Y3
* note: R3: Y0, R2: Y4, R1: Y2, R0: Y6
*/
R3=R3+R0, R0=R3-R0;
R2=R2+R1, R1=R2-R1 || [TMP0]=R3 || R6.L=W[I0--];
/*
* Compute the odd portion (1,3,5,7) even is done.
*
* Y1 = C7 * Y1 - C1 * Y7 + C3 * Y5 - C5 * Y3.
* Y7 = C1 * Y1 + C7 * Y7 + C5 * Y5 + C3 * Y3.
* Y5 = C5 * Y1 + C3 * Y7 + C7 * Y5 - C1 * Y3.
* Y3 = C3 * Y1 - C5 * Y7 - C1 * Y5 - C7 * Y3.
*/
// R5=(Y1,Y7) R6=(Y5,Y3) // R7=(C1,C7)
A1 =R7.L*R5.H, A0 =R7.H*R5.H (IS) || [TMP1]=R2 || R6.H=W[I2--];
A1-=R7.H*R5.L, A0+=R7.L*R5.L (IS) || I0-=4 || R7=[I3++];
A1+=R7.H*R6.H, A0+=R7.L*R6.H (IS) || I0+=M1; // R7=(C3,C5)
R3 =(A1-=R7.L*R6.L), R2 =(A0+=R7.H*R6.L) (IS);
A1 =R7.L*R5.H, A0 =R7.H*R5.H (IS) || R4=[TMP0];
A1+=R7.H*R5.L, A0-=R7.L*R5.L (IS) || I1+=M1 || R7=[I3++]; // R7=(C1,C7)
A1+=R7.L*R6.H, A0-=R7.H*R6.H (IS);
R7 =(A1-=R7.H*R6.L), R6 =(A0-=R7.L*R6.L) (IS) || I2+=M1;
// R3=Y1, R2=Y7, R7=Y5, R6=Y3
/* Transpose write column. */
R5.H=R4+R2 (RND12); // Y0=Y0+Y7
R5.L=R4-R2 (RND12) || R4 = [TMP1]; // Y7=Y7-Y0
R2.H=R1+R7 (RND12) || W[P0++P3]=R5.H; // Y2=Y2+Y5 st Y0
R2.L=R1-R7 (RND12) || W[P1++P4]=R5.L || R7=[I3++]; // Y5=Y2-Y5 st Y7
R5.H=R0-R3 (RND12) || W[P0++P3]=R2.H || R1.L=W[I1++]; // Y1=Y6-Y1 st Y2
R5.L=R0+R3 (RND12) || W[P1++P4]=R2.L || R0.H=W[I0++]; // Y6=Y6+Y1 st Y5
R3.H=R4-R6 (RND12) || W[P0++P3]=R5.H || R0.L=W[I2++]; // Y3=Y3-Y4 st Y1
R3.L=R4+R6 (RND12) || W[P1++P4]=R5.L || R1.H=W[I0++]; // Y4=Y3+Y4 st Y6
/* pipeline loop start, + drain Y3, Y4 */
A1=R7.H*R0.H, A0=R7.H*R0.H (IS) || W[P0++P2]= R3.H || R1.H = W[I0--];
.1: R3=(A1+=R7.H*R0.L), R2=(A0-=R7.H*R0.L) (IS) || W[P1++P5]= R3.L || R7 = [I3++];
I0 = B2; // I0 points to Input Element (0, 0)
I2 = B2; // I2 points to Input Element (0, 0)
I2 += M3 || R0.H = W[I0];
// Y0 is read in R0.H
I1 = I2; // I1 points to input Element (0, 6)
I1 += 4 || R0.L = W[I2++];
// I2 points to input Element (0, 4)
// Y4 is read in R0.L
P2 = 8 (X);
I3 = B3; // I3 points to Coefficients
P0 = B0; // P0 points to array Element (0, 0) for writing
// output
P1 = B0;
R7 = [I3++]; // R7.H = C4 and R7.L = C4
NOP;
/*
* A1 = Y0 * cos(pi/4)
* A0 = Y0 * cos(pi/4)
* A1 = A1 + Y4 * cos(pi/4)
* A0 = A0 - Y4 * cos(pi/4)
* load:
* R1=(Y2,Y6)
* R7=(C2,C6)
* res:
* R3=Y0, R2=Y4
*/
A1=R7.H*R0.H, A0=R7.H*R0.H (IS) || I0+=4 || R1.L=W[I1++];
R3=(A1+=R7.H*R0.L), R2=(A0-=R7.H*R0.L) (IS) || R1.H=W[I0--] || R7=[I3++];
LSETUP (.2, .3) LC0 = P2; // peform 8 1d idcts
P2 = 112 (X);
P1 = P1 + P2;
P2 = -94(X);
.2:
/*
* A1 = Y2 * cos(3pi/8)
* A0 = Y2 * cos(pi/8)
* A1 = A1 - Y6 * cos(pi/8)
* A0 = A0 + Y6 * cos(3pi/8)
* R5 = (Y1,Y7)
* R7 = (C1,C7)
* res:
* R1=Y2, R0=Y6
*/
A1=R7.L*R1.H, A0=R7.H*R1.H (IS) || I0+=4 || R5.H=W[I0];
R1=(A1-=R7.H*R1.L), R0=(A0+=R7.L*R1.L) (IS) || R5.L=W[I1--] || R7=[I3++];
/*
* Y0 = Y0 + Y6.
* Y4 = Y4 + Y2.
* Y2 = Y4 - Y2.
* Y6 = Y0 - Y6.
* R3 is saved
* R6.l=Y3
* note: R3: Y0, R2: Y4, R1: Y2, R0: Y6
*/
R3=R3+R0, R0=R3-R0;
R2=R2+R1, R1=R2-R1 || [TMP0]=R3 || R6.L=W[I0--];
/*
* Compute the odd portion (1,3,5,7) even is done.
*
* Y1 = C7 * Y1 - C1 * Y7 + C3 * Y5 - C5 * Y3.
* Y7 = C1 * Y1 + C7 * Y7 + C5 * Y5 + C3 * Y3.
* Y5 = C5 * Y1 + C3 * Y7 + C7 * Y5 - C1 * Y3.
* Y3 = C3 * Y1 - C5 * Y7 - C1 * Y5 - C7 * Y3.
*/
// R5=(Y1,Y7) R6=(Y5,Y3) // R7=(C1,C7)
A1 =R7.L*R5.H, A0 =R7.H*R5.H (IS) || [TMP1]=R2 || R6.H=W[I2--];
A1-=R7.H*R5.L, A0+=R7.L*R5.L (IS) || I0-=4 || R7=[I3++];
A1+=R7.H*R6.H, A0+=R7.L*R6.H (IS) || I0+=M1; // R7=(C3,C5)
R3 =(A1-=R7.L*R6.L), R2 =(A0+=R7.H*R6.L) (IS);
A1 =R7.L*R5.H, A0 =R7.H*R5.H (IS) || R4=[TMP0];
A1+=R7.H*R5.L, A0-=R7.L*R5.L (IS) || I1+=M1 || R7=[I3++]; // R7=(C1,C7)
A1+=R7.L*R6.H, A0-=R7.H*R6.H (IS);
R7 =(A1-=R7.H*R6.L), R6 =(A0-=R7.L*R6.L) (IS) || I2+=M1;
// R3=Y1, R2=Y7, R7=Y5, R6=Y3
/* Transpose write column. */
R5.H=R4+R2 (RND20); // Y0=Y0+Y7
R5.L=R4-R2 (RND20) || R4 = [TMP1]; // Y7=Y7-Y0
R2.H=R1+R7 (RND20) || W[P0++P3]=R5.H; // Y2=Y2+Y5 st Y0
R2.L=R1-R7 (RND20) || W[P1++P4]=R5.L || R7=[I3++]; // Y5=Y2-Y5 st Y7
R5.H=R0-R3 (RND20) || W[P0++P3]=R2.H || R1.L=W[I1++]; // Y1=Y6-Y1 st Y2
R5.L=R0+R3 (RND20) || W[P1++P4]=R2.L || R0.H=W[I0++]; // Y6=Y6+Y1 st Y5
R3.H=R4-R6 (RND20) || W[P0++P3]=R5.H || R0.L=W[I2++]; // Y3=Y3-Y4 st Y1
R3.L=R4+R6 (RND20) || W[P1++P4]=R5.L || R1.H=W[I0++]; // Y4=Y3+Y4 st Y6
/* pipeline loop start, + drain Y3, Y4 */
A1=R7.H*R0.H, A0=R7.H*R0.H (IS) || W[P0++P2]= R3.H || R1.H = W[I0--];
.3: R3=(A1+=R7.H*R0.L), R2=(A0-=R7.H*R0.L) (IS) || W[P1++P5]= R3.L || R7 = [I3++];
L3 = 0;
(R7:4,P5:3)=[SP++];
unlink;
RTS;
DEFUN_END(idct)
|
