Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(236)

Issues Search
    My Issues | Starred     Open | Closed | All

Side by Side Diff: source/libvpx/vp8/common/x86/variance_sse2.c

Issue 1154153005: Rename variance files (Closed) Base URL: https://chromium.googlesource.com/chromium/deps/libvpx.git@master
Patch Set: Created 5 years, 6 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View unified diff | Download patch
« no previous file with comments | « source/libvpx/vp8/common/x86/variance_mmx.c ('k') | source/libvpx/vp8/common/x86/vp8_variance_impl_mmx.asm » ('j') | no next file with comments »
Toggle Intra-line Diffs ('i') | Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
OLDNEW
(Empty)
1 /*
2 * Copyright (c) 2010 The WebM project authors. All Rights Reserved.
3 *
4 * Use of this source code is governed by a BSD-style license
5 * that can be found in the LICENSE file in the root of the source
6 * tree. An additional intellectual property rights grant can be found
7 * in the file PATENTS. All contributing project authors may
8 * be found in the AUTHORS file in the root of the source tree.
9 */
10
11 #include "./vp8_rtcd.h"
12 #include "vpx_config.h"
13 #include "vp8/common/variance.h"
14 #include "vpx_ports/mem.h"
15 #include "vp8/common/x86/filter_x86.h"
16
17 extern void filter_block1d_h6_mmx(const unsigned char *src_ptr, unsigned short * output_ptr, unsigned int src_pixels_per_line, unsigned int pixel_step, unsigned int output_height, unsigned int output_width, short *filter);
18 extern void filter_block1d_v6_mmx(const short *src_ptr, unsigned char *output_pt r, unsigned int pixels_per_line, unsigned int pixel_step, unsigned int output_he ight, unsigned int output_width, short *filter);
19 extern void filter_block1d8_h6_sse2(const unsigned char *src_ptr, unsigned short *output_ptr, unsigned int src_pixels_per_line, unsigned int pixel_step, unsigne d int output_height, unsigned int output_width, short *filter);
20 extern void filter_block1d8_v6_sse2(const short *src_ptr, unsigned char *output_ ptr, unsigned int pixels_per_line, unsigned int pixel_step, unsigned int output_ height, unsigned int output_width, short *filter);
21
22 extern void vp8_filter_block2d_bil4x4_var_mmx
23 (
24 const unsigned char *ref_ptr,
25 int ref_pixels_per_line,
26 const unsigned char *src_ptr,
27 int src_pixels_per_line,
28 const short *HFilter,
29 const short *VFilter,
30 int *sum,
31 unsigned int *sumsquared
32 );
33
34 void vp8_filter_block2d_bil_var_sse2
35 (
36 const unsigned char *ref_ptr,
37 int ref_pixels_per_line,
38 const unsigned char *src_ptr,
39 int src_pixels_per_line,
40 unsigned int Height,
41 int xoffset,
42 int yoffset,
43 int *sum,
44 unsigned int *sumsquared
45 );
46 void vp8_half_horiz_vert_variance8x_h_sse2
47 (
48 const unsigned char *ref_ptr,
49 int ref_pixels_per_line,
50 const unsigned char *src_ptr,
51 int src_pixels_per_line,
52 unsigned int Height,
53 int *sum,
54 unsigned int *sumsquared
55 );
56 void vp8_half_horiz_vert_variance16x_h_sse2
57 (
58 const unsigned char *ref_ptr,
59 int ref_pixels_per_line,
60 const unsigned char *src_ptr,
61 int src_pixels_per_line,
62 unsigned int Height,
63 int *sum,
64 unsigned int *sumsquared
65 );
66 void vp8_half_horiz_variance8x_h_sse2
67 (
68 const unsigned char *ref_ptr,
69 int ref_pixels_per_line,
70 const unsigned char *src_ptr,
71 int src_pixels_per_line,
72 unsigned int Height,
73 int *sum,
74 unsigned int *sumsquared
75 );
76 void vp8_half_horiz_variance16x_h_sse2
77 (
78 const unsigned char *ref_ptr,
79 int ref_pixels_per_line,
80 const unsigned char *src_ptr,
81 int src_pixels_per_line,
82 unsigned int Height,
83 int *sum,
84 unsigned int *sumsquared
85 );
86 void vp8_half_vert_variance8x_h_sse2
87 (
88 const unsigned char *ref_ptr,
89 int ref_pixels_per_line,
90 const unsigned char *src_ptr,
91 int src_pixels_per_line,
92 unsigned int Height,
93 int *sum,
94 unsigned int *sumsquared
95 );
96 void vp8_half_vert_variance16x_h_sse2
97 (
98 const unsigned char *ref_ptr,
99 int ref_pixels_per_line,
100 const unsigned char *src_ptr,
101 int src_pixels_per_line,
102 unsigned int Height,
103 int *sum,
104 unsigned int *sumsquared
105 );
106
107 unsigned int vp8_sub_pixel_variance4x4_wmt
108 (
109 const unsigned char *src_ptr,
110 int src_pixels_per_line,
111 int xoffset,
112 int yoffset,
113 const unsigned char *dst_ptr,
114 int dst_pixels_per_line,
115 unsigned int *sse
116 )
117 {
118 int xsum;
119 unsigned int xxsum;
120 vp8_filter_block2d_bil4x4_var_mmx(
121 src_ptr, src_pixels_per_line,
122 dst_ptr, dst_pixels_per_line,
123 vp8_bilinear_filters_x86_4[xoffset], vp8_bilinear_filters_x86_4[yoffset] ,
124 &xsum, &xxsum
125 );
126 *sse = xxsum;
127 return (xxsum - (((unsigned int)xsum * xsum) >> 4));
128 }
129
130
131 unsigned int vp8_sub_pixel_variance8x8_wmt
132 (
133 const unsigned char *src_ptr,
134 int src_pixels_per_line,
135 int xoffset,
136 int yoffset,
137 const unsigned char *dst_ptr,
138 int dst_pixels_per_line,
139 unsigned int *sse
140 )
141 {
142 int xsum;
143 unsigned int xxsum;
144
145 if (xoffset == 4 && yoffset == 0)
146 {
147 vp8_half_horiz_variance8x_h_sse2(
148 src_ptr, src_pixels_per_line,
149 dst_ptr, dst_pixels_per_line, 8,
150 &xsum, &xxsum);
151 }
152 else if (xoffset == 0 && yoffset == 4)
153 {
154 vp8_half_vert_variance8x_h_sse2(
155 src_ptr, src_pixels_per_line,
156 dst_ptr, dst_pixels_per_line, 8,
157 &xsum, &xxsum);
158 }
159 else if (xoffset == 4 && yoffset == 4)
160 {
161 vp8_half_horiz_vert_variance8x_h_sse2(
162 src_ptr, src_pixels_per_line,
163 dst_ptr, dst_pixels_per_line, 8,
164 &xsum, &xxsum);
165 }
166 else
167 {
168 vp8_filter_block2d_bil_var_sse2(
169 src_ptr, src_pixels_per_line,
170 dst_ptr, dst_pixels_per_line, 8,
171 xoffset, yoffset,
172 &xsum, &xxsum);
173 }
174
175 *sse = xxsum;
176 return (xxsum - (((unsigned int)xsum * xsum) >> 6));
177 }
178
179 unsigned int vp8_sub_pixel_variance16x16_wmt
180 (
181 const unsigned char *src_ptr,
182 int src_pixels_per_line,
183 int xoffset,
184 int yoffset,
185 const unsigned char *dst_ptr,
186 int dst_pixels_per_line,
187 unsigned int *sse
188 )
189 {
190 int xsum0, xsum1;
191 unsigned int xxsum0, xxsum1;
192
193
194 /* note we could avoid these if statements if the calling function
195 * just called the appropriate functions inside.
196 */
197 if (xoffset == 4 && yoffset == 0)
198 {
199 vp8_half_horiz_variance16x_h_sse2(
200 src_ptr, src_pixels_per_line,
201 dst_ptr, dst_pixels_per_line, 16,
202 &xsum0, &xxsum0);
203 }
204 else if (xoffset == 0 && yoffset == 4)
205 {
206 vp8_half_vert_variance16x_h_sse2(
207 src_ptr, src_pixels_per_line,
208 dst_ptr, dst_pixels_per_line, 16,
209 &xsum0, &xxsum0);
210 }
211 else if (xoffset == 4 && yoffset == 4)
212 {
213 vp8_half_horiz_vert_variance16x_h_sse2(
214 src_ptr, src_pixels_per_line,
215 dst_ptr, dst_pixels_per_line, 16,
216 &xsum0, &xxsum0);
217 }
218 else
219 {
220 vp8_filter_block2d_bil_var_sse2(
221 src_ptr, src_pixels_per_line,
222 dst_ptr, dst_pixels_per_line, 16,
223 xoffset, yoffset,
224 &xsum0, &xxsum0
225 );
226
227 vp8_filter_block2d_bil_var_sse2(
228 src_ptr + 8, src_pixels_per_line,
229 dst_ptr + 8, dst_pixels_per_line, 16,
230 xoffset, yoffset,
231 &xsum1, &xxsum1
232 );
233 xsum0 += xsum1;
234 xxsum0 += xxsum1;
235 }
236
237 *sse = xxsum0;
238 return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
239 }
240
241 unsigned int vp8_sub_pixel_mse16x16_wmt(
242 const unsigned char *src_ptr,
243 int src_pixels_per_line,
244 int xoffset,
245 int yoffset,
246 const unsigned char *dst_ptr,
247 int dst_pixels_per_line,
248 unsigned int *sse
249 )
250 {
251 vp8_sub_pixel_variance16x16_wmt(src_ptr, src_pixels_per_line, xoffset, yoffs et, dst_ptr, dst_pixels_per_line, sse);
252 return *sse;
253 }
254
255 unsigned int vp8_sub_pixel_variance16x8_wmt
256 (
257 const unsigned char *src_ptr,
258 int src_pixels_per_line,
259 int xoffset,
260 int yoffset,
261 const unsigned char *dst_ptr,
262 int dst_pixels_per_line,
263 unsigned int *sse
264
265 )
266 {
267 int xsum0, xsum1;
268 unsigned int xxsum0, xxsum1;
269
270 if (xoffset == 4 && yoffset == 0)
271 {
272 vp8_half_horiz_variance16x_h_sse2(
273 src_ptr, src_pixels_per_line,
274 dst_ptr, dst_pixels_per_line, 8,
275 &xsum0, &xxsum0);
276 }
277 else if (xoffset == 0 && yoffset == 4)
278 {
279 vp8_half_vert_variance16x_h_sse2(
280 src_ptr, src_pixels_per_line,
281 dst_ptr, dst_pixels_per_line, 8,
282 &xsum0, &xxsum0);
283 }
284 else if (xoffset == 4 && yoffset == 4)
285 {
286 vp8_half_horiz_vert_variance16x_h_sse2(
287 src_ptr, src_pixels_per_line,
288 dst_ptr, dst_pixels_per_line, 8,
289 &xsum0, &xxsum0);
290 }
291 else
292 {
293 vp8_filter_block2d_bil_var_sse2(
294 src_ptr, src_pixels_per_line,
295 dst_ptr, dst_pixels_per_line, 8,
296 xoffset, yoffset,
297 &xsum0, &xxsum0);
298
299 vp8_filter_block2d_bil_var_sse2(
300 src_ptr + 8, src_pixels_per_line,
301 dst_ptr + 8, dst_pixels_per_line, 8,
302 xoffset, yoffset,
303 &xsum1, &xxsum1);
304 xsum0 += xsum1;
305 xxsum0 += xxsum1;
306 }
307
308 *sse = xxsum0;
309 return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 7));
310 }
311
312 unsigned int vp8_sub_pixel_variance8x16_wmt
313 (
314 const unsigned char *src_ptr,
315 int src_pixels_per_line,
316 int xoffset,
317 int yoffset,
318 const unsigned char *dst_ptr,
319 int dst_pixels_per_line,
320 unsigned int *sse
321 )
322 {
323 int xsum;
324 unsigned int xxsum;
325
326 if (xoffset == 4 && yoffset == 0)
327 {
328 vp8_half_horiz_variance8x_h_sse2(
329 src_ptr, src_pixels_per_line,
330 dst_ptr, dst_pixels_per_line, 16,
331 &xsum, &xxsum);
332 }
333 else if (xoffset == 0 && yoffset == 4)
334 {
335 vp8_half_vert_variance8x_h_sse2(
336 src_ptr, src_pixels_per_line,
337 dst_ptr, dst_pixels_per_line, 16,
338 &xsum, &xxsum);
339 }
340 else if (xoffset == 4 && yoffset == 4)
341 {
342 vp8_half_horiz_vert_variance8x_h_sse2(
343 src_ptr, src_pixels_per_line,
344 dst_ptr, dst_pixels_per_line, 16,
345 &xsum, &xxsum);
346 }
347 else
348 {
349 vp8_filter_block2d_bil_var_sse2(
350 src_ptr, src_pixels_per_line,
351 dst_ptr, dst_pixels_per_line, 16,
352 xoffset, yoffset,
353 &xsum, &xxsum);
354 }
355
356 *sse = xxsum;
357 return (xxsum - (((unsigned int)xsum * xsum) >> 7));
358 }
359
360
361 unsigned int vp8_variance_halfpixvar16x16_h_wmt(
362 const unsigned char *src_ptr,
363 int src_pixels_per_line,
364 const unsigned char *dst_ptr,
365 int dst_pixels_per_line,
366 unsigned int *sse)
367 {
368 int xsum0;
369 unsigned int xxsum0;
370
371 vp8_half_horiz_variance16x_h_sse2(
372 src_ptr, src_pixels_per_line,
373 dst_ptr, dst_pixels_per_line, 16,
374 &xsum0, &xxsum0);
375
376 *sse = xxsum0;
377 return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
378 }
379
380
381 unsigned int vp8_variance_halfpixvar16x16_v_wmt(
382 const unsigned char *src_ptr,
383 int src_pixels_per_line,
384 const unsigned char *dst_ptr,
385 int dst_pixels_per_line,
386 unsigned int *sse)
387 {
388 int xsum0;
389 unsigned int xxsum0;
390 vp8_half_vert_variance16x_h_sse2(
391 src_ptr, src_pixels_per_line,
392 dst_ptr, dst_pixels_per_line, 16,
393 &xsum0, &xxsum0);
394
395 *sse = xxsum0;
396 return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
397 }
398
399
400 unsigned int vp8_variance_halfpixvar16x16_hv_wmt(
401 const unsigned char *src_ptr,
402 int src_pixels_per_line,
403 const unsigned char *dst_ptr,
404 int dst_pixels_per_line,
405 unsigned int *sse)
406 {
407 int xsum0;
408 unsigned int xxsum0;
409
410 vp8_half_horiz_vert_variance16x_h_sse2(
411 src_ptr, src_pixels_per_line,
412 dst_ptr, dst_pixels_per_line, 16,
413 &xsum0, &xxsum0);
414
415 *sse = xxsum0;
416 return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
417 }
OLDNEW
« no previous file with comments | « source/libvpx/vp8/common/x86/variance_mmx.c ('k') | source/libvpx/vp8/common/x86/vp8_variance_impl_mmx.asm » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698