/
synth_3dnow.asm
291 lines (277 loc) · 4.5 KB
/
synth_3dnow.asm
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
; 1 "synth_3dnow.S"
; 1 "<built-in>"
; 1 "<command line>"
; 1 "synth_3dnow.S"
; 34 "synth_3dnow.S"
; 1 "mangle.h" 1
; 13 "mangle.h"
; 1 "config.h" 1
; 14 "mangle.h" 2
; 1 "intsym.h" 1
; 15 "mangle.h" 2
; 35 "synth_3dnow.S" 2
; 53 "synth_3dnow.S"
%include "asm_nasm.inc"
_sym_mangle dct64_3dnow
_sym_mangle synth_1to1_3dnow_asm
EXTERN dct64_3dnow
%ifdef ACCURATE_ROUNDING
SECTION .data
ALIGN 8
max_s16:
dd 1191181824 ; 32767.0
dd 1191181824
min_s16:
dd -956301312 ; -32768.0
dd -956301312
ftoi_magic:
dd 1262485504 ; 2^23 + 2^22
dd 1262485504
%endif
SECTION .text
ALIGN 16
GLOBAL synth_1to1_3dnow_asm
synth_1to1_3dnow_asm:
sub esp,24
push ebp
push edi
xor ebp,ebp
push esi
push ebx
; 86 "synth_3dnow.S"
mov esi, [esp+52]
mov [esp+16],esi
mov ebx, [esp+48]
mov esi, [esp+60]
mov edx, [esi]
femms
test ebx,ebx
jne L26
dec edx
and edx,15
mov [esi],edx
mov ecx, [esp+56]
jmp L27
L26:
add dword [esp+16],2
mov ecx, [esp+56]
add ecx,2176
L27:
test dl,1
je L28
mov [esp+36],edx
mov ebx,ecx
mov esi, [esp+44]
mov edi,edx
push esi
sal edi,2
mov eax,ebx
mov [esp+24],edi
add eax,edi
push eax
mov eax,edx
inc eax
and eax,15
lea eax, [eax*4+1088]
add eax,ebx
push eax
call dct64_3dnow
add esp,12
jmp L29
L28:
lea esi, [edx+1]
mov edi, [esp+44]
mov [esp+36],esi
lea eax, [ecx+edx*4+1092]
push edi
lea ebx, [ecx+1088]
push eax
sal esi,2
lea eax, [ecx+edx*4]
push eax
call dct64_3dnow
add esp,12
mov [esp+20],esi
L29:
mov edx, [esp+64]
add edx,64
mov ecx,16
sub edx, [esp+20]
mov edi, [esp+16]
pcmpeqb mm7,mm7
pslld mm7,31
movq mm0,[edx]
movq mm1,[ebx]
ALIGN 32
L33:
movq mm3,[edx+8]
pfmul mm0,mm1
movq mm4,[ebx+8]
movq mm5,[edx+16]
pfmul mm3,mm4
movq mm6,[ebx+16]
pfadd mm0,mm3
movq mm1,[edx+24]
pfmul mm5,mm6
movq mm2,[ebx+24]
pfadd mm0,mm5
movq mm3,[edx+32]
pfmul mm1,mm2
movq mm4,[ebx+32]
pfadd mm0,mm1
movq mm5,[edx+40]
pfmul mm3,mm4
movq mm6,[ebx+40]
pfadd mm0,mm3
movq mm1,[edx+48]
pfmul mm5,mm6
movq mm2,[ebx+48]
pfadd mm5,mm0
movq mm3,[edx+56]
pfmul mm2,mm1
movq mm4,[ebx+56]
pfadd mm2,mm5
add ebx,64
sub edx,-128
movq mm0,[edx]
pfmul mm3,mm4
movq mm1,[ebx]
pfadd mm2,mm3
movq mm3,mm2
psrlq mm3,32
pfsub mm2,mm3
inc ebp
%ifdef ACCURATE_ROUNDING
pfmin mm2,[max_s16]
pfmax mm2,[min_s16]
pfadd mm2,[ftoi_magic]
%else
pf2id mm2,mm2
packssdw mm2,mm2
%endif
movd eax,mm2
mov [edi+0],ax
add edi,4
dec ecx
jnz L33
movd mm0,[ebx]
movd mm1,[edx]
punpckldq mm0,[ebx+8]
punpckldq mm1,[edx+8]
movd mm3,[ebx+16]
movd mm4,[edx+16]
pfmul mm0,mm1
punpckldq mm3,[ebx+24]
punpckldq mm4,[edx+24]
movd mm5,[ebx+32]
movd mm6,[edx+32]
pfmul mm3,mm4
punpckldq mm5,[ebx+40]
punpckldq mm6,[edx+40]
pfadd mm0,mm3
movd mm1,[ebx+48]
movd mm2,[edx+48]
pfmul mm5,mm6
punpckldq mm1,[ebx+56]
punpckldq mm2,[edx+56]
pfadd mm0,mm5
pfmul mm1,mm2
pfadd mm0,mm1
pfacc mm0,mm1
%ifdef ACCURATE_ROUNDING
pfmin mm0,[max_s16]
pfmax mm0,[min_s16]
pfadd mm0,[ftoi_magic]
%else
pf2id mm0,mm0
packssdw mm0,mm0
%endif
movd eax,mm0
mov [edi+0],ax
inc ebp
mov esi, [esp+36]
add ebx,-64
mov ebp,15
add edi,4
lea edx, [edx+esi*8-128]
mov ecx,15
movd mm0,[ebx]
movd mm1,[edx-4]
punpckldq mm0,[ebx+4]
punpckldq mm1,[edx-8]
ALIGN 32
L46:
movd mm3,[ebx+8]
movd mm4,[edx-12]
pfmul mm0,mm1
punpckldq mm3,[ebx+12]
punpckldq mm4,[edx-16]
movd mm5,[ebx+16]
movd mm6,[edx-20]
pfmul mm3,mm4
punpckldq mm5,[ebx+20]
punpckldq mm6,[edx-24]
pfadd mm0,mm3
movd mm1,[ebx+24]
movd mm2,[edx-28]
pfmul mm5,mm6
punpckldq mm1,[ebx+28]
punpckldq mm2,[edx-32]
pfadd mm0,mm5
movd mm3,[ebx+32]
movd mm4,[edx-36]
pfmul mm1,mm2
punpckldq mm3,[ebx+36]
punpckldq mm4,[edx-40]
pfadd mm0,mm1
movd mm5,[ebx+40]
movd mm6,[edx-44]
pfmul mm3,mm4
punpckldq mm5,[ebx+44]
punpckldq mm6,[edx-48]
pfadd mm0,mm3
movd mm1,[ebx+48]
movd mm2,[edx-52]
pfmul mm5,mm6
punpckldq mm1,[ebx+52]
punpckldq mm2,[edx-56]
pfadd mm5,mm0
movd mm3,[ebx+56]
movd mm4,[edx-60]
pfmul mm1,mm2
punpckldq mm3,[ebx+60]
punpckldq mm4,[edx]
pfadd mm5,mm1
add edx,-128
add ebx,-64
movd mm0,[ebx]
movd mm1,[edx-4]
pfmul mm3,mm4
punpckldq mm0,[ebx+4]
punpckldq mm1,[edx-8]
pfadd mm3,mm5
pfacc mm3,mm3
inc ebp
pxor mm3,mm7
%ifdef ACCURATE_ROUNDING
pfmin mm3,[max_s16]
pfmax mm3,[min_s16]
pfadd mm3,[ftoi_magic]
%else
pf2id mm3,mm3
packssdw mm3,mm3
%endif
movd eax,mm3
mov [edi],ax
add edi,4
dec ecx
jnz L46
femms
mov eax,ebp
pop ebx
pop esi
pop edi
pop ebp
add esp,24
ret
; 345 "synth_3dnow.S"