media/libvpx/vp8/common/x86/vp8_variance_impl_mmx.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
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353

;
;  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
;
;  Use of this source code is governed by a BSD-style license
;  that can be found in the LICENSE file in the root of the source
;  tree. An additional intellectual property rights grant can be found
;  in the file PATENTS.  All contributing project authors may
;  be found in the AUTHORS file in the root of the source tree.
;


%include "vpx_ports/x86_abi_support.asm"

%define mmx_filter_shift            7

;void vp8_filter_block2d_bil4x4_var_mmx
;(
;    unsigned char *ref_ptr,
;    int ref_pixels_per_line,
;    unsigned char *src_ptr,
;    int src_pixels_per_line,
;    unsigned short *HFilter,
;    unsigned short *VFilter,
;    int *sum,
;    unsigned int *sumsquared
;)
global sym(vp8_filter_block2d_bil4x4_var_mmx) PRIVATE
sym(vp8_filter_block2d_bil4x4_var_mmx):
    push        rbp
    mov         rbp, rsp
    SHADOW_ARGS_TO_STACK 8
    GET_GOT     rbx
    push rsi
    push rdi
    sub         rsp, 16
    ; end prolog


        pxor            mm6,            mm6                 ;
        pxor            mm7,            mm7                 ;

        mov             rax,            arg(4) ;HFilter             ;
        mov             rdx,            arg(5) ;VFilter             ;

        mov             rsi,            arg(0) ;ref_ptr              ;
        mov             rdi,            arg(2) ;src_ptr              ;

        mov             rcx,            4                   ;
        pxor            mm0,            mm0                 ;

        movd            mm1,            [rsi]               ;
        movd            mm3,            [rsi+1]             ;

        punpcklbw       mm1,            mm0                 ;
        pmullw          mm1,            [rax]               ;

        punpcklbw       mm3,            mm0                 ;
        pmullw          mm3,            [rax+8]             ;

        paddw           mm1,            mm3                 ;
        paddw           mm1,            [GLOBAL(mmx_bi_rd)] ;

        psraw           mm1,            mmx_filter_shift    ;
        movq            mm5,            mm1

%if ABI_IS_32BIT
        add             rsi, dword ptr  arg(1) ;ref_pixels_per_line    ;
%else
        movsxd          r8, dword ptr  arg(1) ;ref_pixels_per_line    ;
        add             rsi, r8
%endif

.filter_block2d_bil4x4_var_mmx_loop:

        movd            mm1,            [rsi]               ;
        movd            mm3,            [rsi+1]             ;

        punpcklbw       mm1,            mm0                 ;
        pmullw          mm1,            [rax]               ;

        punpcklbw       mm3,            mm0                 ;
        pmullw          mm3,            [rax+8]             ;

        paddw           mm1,            mm3                 ;
        paddw           mm1,            [GLOBAL(mmx_bi_rd)] ;

        psraw           mm1,            mmx_filter_shift    ;
        movq            mm3,            mm5                 ;

        movq            mm5,            mm1                 ;
        pmullw          mm3,            [rdx]               ;

        pmullw          mm1,            [rdx+8]             ;
        paddw           mm1,            mm3                 ;


        paddw           mm1,            [GLOBAL(mmx_bi_rd)] ;
        psraw           mm1,            mmx_filter_shift    ;

        movd            mm3,            [rdi]               ;
        punpcklbw       mm3,            mm0                 ;

        psubw           mm1,            mm3                 ;
        paddw           mm6,            mm1                 ;

        pmaddwd         mm1,            mm1                 ;
        paddd           mm7,            mm1                 ;

%if ABI_IS_32BIT
        add             rsi,            dword ptr arg(1) ;ref_pixels_per_line    ;
        add             rdi,            dword ptr arg(3) ;src_pixels_per_line    ;
%else
        movsxd          r8,             dword ptr arg(1) ;ref_pixels_per_line
        movsxd          r9,             dword ptr arg(3) ;src_pixels_per_line
        add             rsi,            r8
        add             rdi,            r9
%endif
        sub             rcx,            1                   ;
        jnz             .filter_block2d_bil4x4_var_mmx_loop       ;


        pxor            mm3,            mm3                 ;
        pxor            mm2,            mm2                 ;

        punpcklwd       mm2,            mm6                 ;
        punpckhwd       mm3,            mm6                 ;

        paddd           mm2,            mm3                 ;
        movq            mm6,            mm2                 ;

        psrlq           mm6,            32                  ;
        paddd           mm2,            mm6                 ;

        psrad           mm2,            16                  ;
        movq            mm4,            mm7                 ;

        psrlq           mm4,            32                  ;
        paddd           mm4,            mm7                 ;

        mov             rdi,            arg(6) ;sum
        mov             rsi,            arg(7) ;sumsquared

        movd            dword ptr [rdi],          mm2                 ;
        movd            dword ptr [rsi],          mm4                 ;


    ; begin epilog
    add rsp, 16
    pop rdi
    pop rsi
    RESTORE_GOT
    UNSHADOW_ARGS
    pop         rbp
    ret


;void vp8_filter_block2d_bil_var_mmx
;(
;    unsigned char *ref_ptr,
;    int ref_pixels_per_line,
;    unsigned char *src_ptr,
;    int src_pixels_per_line,
;    unsigned int Height,
;    unsigned short *HFilter,
;    unsigned short *VFilter,
;    int *sum,
;    unsigned int *sumsquared
;)
global sym(vp8_filter_block2d_bil_var_mmx) PRIVATE
sym(vp8_filter_block2d_bil_var_mmx):
    push        rbp
    mov         rbp, rsp
    SHADOW_ARGS_TO_STACK 9
    GET_GOT     rbx
    push rsi
    push rdi
    sub         rsp, 16
    ; end prolog

        pxor            mm6,            mm6                 ;
        pxor            mm7,            mm7                 ;
        mov             rax,            arg(5) ;HFilter             ;

        mov             rdx,            arg(6) ;VFilter             ;
        mov             rsi,            arg(0) ;ref_ptr              ;

        mov             rdi,            arg(2) ;src_ptr              ;
        movsxd          rcx,            dword ptr arg(4) ;Height              ;

        pxor            mm0,            mm0                 ;
        movq            mm1,            [rsi]               ;

        movq            mm3,            [rsi+1]             ;
        movq            mm2,            mm1                 ;

        movq            mm4,            mm3                 ;
        punpcklbw       mm1,            mm0                 ;

        punpckhbw       mm2,            mm0                 ;
        pmullw          mm1,            [rax]               ;

        pmullw          mm2,            [rax]               ;
        punpcklbw       mm3,            mm0                 ;

        punpckhbw       mm4,            mm0                 ;
        pmullw          mm3,            [rax+8]             ;

        pmullw          mm4,            [rax+8]             ;
        paddw           mm1,            mm3                 ;

        paddw           mm2,            mm4                 ;
        paddw           mm1,            [GLOBAL(mmx_bi_rd)] ;

        psraw           mm1,            mmx_filter_shift    ;
        paddw           mm2,            [GLOBAL(mmx_bi_rd)] ;

        psraw           mm2,            mmx_filter_shift    ;
        movq            mm5,            mm1

        packuswb        mm5,            mm2                 ;
%if ABI_IS_32BIT
        add             rsi,            dword ptr arg(1) ;ref_pixels_per_line
%else
        movsxd          r8,             dword ptr arg(1) ;ref_pixels_per_line
        add             rsi,            r8
%endif

.filter_block2d_bil_var_mmx_loop:

        movq            mm1,            [rsi]               ;
        movq            mm3,            [rsi+1]             ;

        movq            mm2,            mm1                 ;
        movq            mm4,            mm3                 ;

        punpcklbw       mm1,            mm0                 ;
        punpckhbw       mm2,            mm0                 ;

        pmullw          mm1,            [rax]               ;
        pmullw          mm2,            [rax]               ;

        punpcklbw       mm3,            mm0                 ;
        punpckhbw       mm4,            mm0                 ;

        pmullw          mm3,            [rax+8]             ;
        pmullw          mm4,            [rax+8]             ;

        paddw           mm1,            mm3                 ;
        paddw           mm2,            mm4                 ;

        paddw           mm1,            [GLOBAL(mmx_bi_rd)] ;
        psraw           mm1,            mmx_filter_shift    ;

        paddw           mm2,            [GLOBAL(mmx_bi_rd)] ;
        psraw           mm2,            mmx_filter_shift    ;

        movq            mm3,            mm5                 ;
        movq            mm4,            mm5                 ;

        punpcklbw       mm3,            mm0                 ;
        punpckhbw       mm4,            mm0                 ;

        movq            mm5,            mm1                 ;
        packuswb        mm5,            mm2                 ;

        pmullw          mm3,            [rdx]               ;
        pmullw          mm4,            [rdx]               ;

        pmullw          mm1,            [rdx+8]             ;
        pmullw          mm2,            [rdx+8]             ;

        paddw           mm1,            mm3                 ;
        paddw           mm2,            mm4                 ;

        paddw           mm1,            [GLOBAL(mmx_bi_rd)] ;
        paddw           mm2,            [GLOBAL(mmx_bi_rd)] ;

        psraw           mm1,            mmx_filter_shift    ;
        psraw           mm2,            mmx_filter_shift    ;

        movq            mm3,            [rdi]               ;
        movq            mm4,            mm3                 ;

        punpcklbw       mm3,            mm0                 ;
        punpckhbw       mm4,            mm0                 ;

        psubw           mm1,            mm3                 ;
        psubw           mm2,            mm4                 ;

        paddw           mm6,            mm1                 ;
        pmaddwd         mm1,            mm1                 ;

        paddw           mm6,            mm2                 ;
        pmaddwd         mm2,            mm2                 ;

        paddd           mm7,            mm1                 ;
        paddd           mm7,            mm2                 ;

%if ABI_IS_32BIT
        add             rsi,            dword ptr arg(1) ;ref_pixels_per_line    ;
        add             rdi,            dword ptr arg(3) ;src_pixels_per_line    ;
%else
        movsxd          r8,             dword ptr arg(1) ;ref_pixels_per_line    ;
        movsxd          r9,             dword ptr arg(3) ;src_pixels_per_line    ;
        add             rsi,            r8
        add             rdi,            r9
%endif
        sub             rcx,            1                   ;
        jnz             .filter_block2d_bil_var_mmx_loop       ;


        pxor            mm3,            mm3                 ;
        pxor            mm2,            mm2                 ;

        punpcklwd       mm2,            mm6                 ;
        punpckhwd       mm3,            mm6                 ;

        paddd           mm2,            mm3                 ;
        movq            mm6,            mm2                 ;

        psrlq           mm6,            32                  ;
        paddd           mm2,            mm6                 ;

        psrad           mm2,            16                  ;
        movq            mm4,            mm7                 ;

        psrlq           mm4,            32                  ;
        paddd           mm4,            mm7                 ;

        mov             rdi,            arg(7) ;sum
        mov             rsi,            arg(8) ;sumsquared

        movd            dword ptr [rdi],          mm2                 ;
        movd            dword ptr [rsi],          mm4                 ;

    ; begin epilog
    add rsp, 16
    pop rdi
    pop rsi
    RESTORE_GOT
    UNSHADOW_ARGS
    pop         rbp
    ret


SECTION_RODATA
;short mmx_bi_rd[4] = { 64, 64, 64, 64};
align 16
mmx_bi_rd:
    times 4 dw 64