| 2 |
# Ed.Gomez: This ChangeLog is generated from a personal tree maintained |
# Ed.Gomez: This ChangeLog is generated from a personal tree maintained |
| 3 |
# under the arch revision control tool. That's why dates may be skewed. I |
# under the arch revision control tool. That's why dates may be skewed. I |
| 4 |
# also removed all my email adresses from the output because they are not |
# also removed all my email adresses from the output because they are not |
| 5 |
# relevent |
# relevant. |
| 6 |
# |
# |
| 7 |
|
|
| 8 |
|
|
| 9 |
|
2003-11-13 23:09:34 GMT patch-95 |
| 10 |
|
|
| 11 |
|
Summary: |
| 12 |
|
8x8 16bit Block SSE optimization. |
| 13 |
|
Revision: |
| 14 |
|
xvidcore--devapi4--1.0--patch-95 |
| 15 |
|
|
| 16 |
|
MMXed the calculation of SSE for 8x8 16bit blocks. This helps quite |
| 17 |
|
a lot VHQ=4 mode. |
| 18 |
|
|
| 19 |
|
My tests show with trellis:chroma_me: |
| 20 |
|
- ~20% speed improvement for vhq=4. |
| 21 |
|
- at least 5% when using vhq=1. |
| 22 |
|
|
| 23 |
|
Of course this speedup vanishes if more CPU intensive features are used. |
| 24 |
|
CruNcher who used gmc/qpel, noticed "only" a ~5% speed improvement. |
| 25 |
|
|
| 26 |
|
NB: i'm of course talking about overall speed improvement. Such a small |
| 27 |
|
patch for such a big improvement :-) |
| 28 |
|
|
| 29 |
|
modified files: |
| 30 |
|
src/motion/estimation_rd_based.c src/motion/sad.c |
| 31 |
|
src/motion/sad.h src/motion/x86_asm/sad_mmx.asm src/xvid.c |
| 32 |
|
|
| 33 |
|
|
| 34 |
|
2003-11-13 22:34:33 GMT patch-94 |
| 35 |
|
|
| 36 |
|
Summary: |
| 37 |
|
Various small bug fixes. |
| 38 |
|
Revision: |
| 39 |
|
xvidcore--devapi4--1.0--patch-94 |
| 40 |
|
|
| 41 |
|
* encoder.c: GMC code fix in encoder.c. Now gmcval is initialized correctly |
| 42 |
|
when using GME. |
| 43 |
|
* xvid_decraw.c: Fix elementary stream output. |
| 44 |
|
* plugin_2pass2.c: Small parsing bug in stats reading in 2pass2. |
| 45 |
|
* decoder.c: Read resync markers in bframes. |
| 46 |
|
|
| 47 |
|
modified files: |
| 48 |
|
examples/xvid_decraw.c src/decoder.c src/encoder.c |
| 49 |
|
src/plugins/plugin_2pass2.c |
| 50 |
|
|
| 51 |
|
|
| 52 |
|
2003-11-11 16:24:05 GMT patch-93 |
| 53 |
|
|
| 54 |
|
Summary: |
| 55 |
|
VFW update for overflow control |
| 56 |
|
Revision: |
| 57 |
|
xvidcore--devapi4--1.0--patch-93 |
| 58 |
|
|
| 59 |
|
From Koepi. |
| 60 |
|
* Added widget and code for overflow control strength. |
| 61 |
|
* Removed widgets for payback options and kfthresholds. |
| 62 |
|
* Activated frame stats in DebugOutputView all the time. |
| 63 |
|
|
| 64 |
|
From me: |
| 65 |
|
* Activated static motion detection in cartoon mode. |
| 66 |
|
|
| 67 |
|
modified files: |
| 68 |
|
vfw/src/codec.c vfw/src/config.c vfw/src/resource.h |
| 69 |
|
vfw/src/resource.rc |
| 70 |
|
|
| 71 |
|
|
| 72 |
|
2003-11-09 20:47:47 GMT patch-92 |
| 73 |
|
|
| 74 |
|
Summary: |
| 75 |
|
New two pass code. |
| 76 |
|
Revision: |
| 77 |
|
xvidcore--devapi4--1.0--patch-92 |
| 78 |
|
|
| 79 |
|
New two pass code. I may say it's just a fixed version, though it looks |
| 80 |
|
more like a "take all the ideas and write it again" version. It performs |
| 81 |
|
better with all natural sequences i have and a bit worse with anime. |
| 82 |
|
|
| 83 |
|
Including it now, allow me improving the code during the beta releases. |
| 84 |
|
|
| 85 |
|
modified files: |
| 86 |
|
src/encoder.c src/plugins/plugin_2pass1.c |
| 87 |
|
src/plugins/plugin_2pass2.c src/xvid.h vfw/src/codec.c |
| 88 |
|
vfw/src/config.c vfw/src/config.h |
| 89 |
|
|
| 90 |
|
|
| 91 |
|
2003-11-09 17:07:16 GMT patch-91 |
| 92 |
|
|
| 93 |
|
Summary: |
| 94 |
|
Fixes for bframe compensation (used in psnr tests). |
| 95 |
|
Revision: |
| 96 |
|
xvidcore--devapi4--1.0--patch-91 |
| 97 |
|
|
| 98 |
|
* transfer_8to16_sub2_(c|mmx|xmm|3dne) write back the compensated |
| 99 |
|
result to current frame pointer. |
| 100 |
|
* transfer_8to16_sub2_mmx uses proper rounding (a+b+1)/2. The +1 |
| 101 |
|
operation was missing. |
| 102 |
|
* Blocks skipped in bframes must be compensated for psnr computing. |
| 103 |
|
|
| 104 |
|
modified files: |
| 105 |
|
src/encoder.c src/motion/estimation_bvop.c |
| 106 |
|
src/utils/mem_transfer.c |
| 107 |
|
src/utils/x86_asm/mem_transfer_3dne.asm |
| 108 |
|
src/utils/x86_asm/mem_transfer_mmx.asm |
| 109 |
|
|
| 110 |
|
|
| 111 |
|
2003-11-05 16:05:44 GMT patch-90 |
| 112 |
|
|
| 113 |
|
Summary: |
| 114 |
|
Speed improvement not wasting setedges and interpolate calls. |
| 115 |
|
Revision: |
| 116 |
|
xvidcore--devapi4--1.0--patch-90 |
| 117 |
|
|
| 118 |
|
Patch from syskin. |
| 119 |
|
|
| 120 |
|
* This patch avoids calling setedges and interpolate for uneeded cases: |
| 121 |
|
- setedges is only called once per frame. |
| 122 |
|
- interpolate is called only when the previous rounding |
| 123 |
|
was different from the one needed. |
| 124 |
|
* Interpolation has been optimized a bit for qpel case, we do the |
| 125 |
|
hv pass down to top to use the cache more efficiently (hope so). |
| 126 |
|
|
| 127 |
|
modified files: |
| 128 |
|
src/encoder.c src/encoder.h src/image/image.c |
| 129 |
|
|
| 130 |
|
|
| 131 |
|
2003-11-03 19:51:12 GMT patch-89 |
| 132 |
|
|
| 133 |
|
Summary: |
| 134 |
|
SSE2 dev16 fix + xvid_bench DCT block alignments. |
| 135 |
|
Revision: |
| 136 |
|
xvidcore--devapi4--1.0--patch-89 |
| 137 |
|
|
| 138 |
|
* Small error fixed by Skal in his dev16 code (missing pshufd). |
| 139 |
|
* Blocks used by DCT tests are now aligned with DECLARE_ALIGNED_MATRIX |
| 140 |
|
this avoids the well know segfaults when using SSE2 instructions that |
| 141 |
|
suppose data alignment. |
| 142 |
|
|
| 143 |
|
modified files: |
| 144 |
|
examples/xvid_bench.c src/motion/x86_asm/sad_sse2.asm |
| 145 |
|
|
| 146 |
|
|
| 147 |
|
2003-11-03 15:42:23 GMT patch-88 |
| 148 |
|
|
| 149 |
|
Summary: |
| 150 |
|
Align .rodata section for non coff objects |
| 151 |
|
Revision: |
| 152 |
|
xvidcore--devapi4--1.0--patch-88 |
| 153 |
|
|
| 154 |
|
Align .rodata section for non coff objects |
| 155 |
|
|
| 156 |
|
|
| 157 |
|
modified files: |
| 158 |
|
src/bitstream/x86_asm/cbp_mmx.asm |
| 159 |
|
src/bitstream/x86_asm/cbp_sse2.asm |
| 160 |
|
src/dct/x86_asm/fdct_mmx_ffmpeg.asm |
| 161 |
|
src/dct/x86_asm/fdct_mmx_skal.asm |
| 162 |
|
src/dct/x86_asm/fdct_sse2_skal.asm |
| 163 |
|
src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm |
| 164 |
|
src/dct/x86_asm/idct_sse2_dmitry.asm |
| 165 |
|
src/dct/x86_asm/simple_idct_mmx.asm |
| 166 |
|
src/image/x86_asm/colorspace_rgb_mmx.asm |
| 167 |
|
src/image/x86_asm/colorspace_yuyv_mmx.asm |
| 168 |
|
src/image/x86_asm/interpolate8x8_3dn.asm |
| 169 |
|
src/image/x86_asm/interpolate8x8_3dne.asm |
| 170 |
|
src/image/x86_asm/interpolate8x8_mmx.asm |
| 171 |
|
src/image/x86_asm/interpolate8x8_xmm.asm |
| 172 |
|
src/image/x86_asm/qpel_mmx.asm |
| 173 |
|
src/image/x86_asm/reduced_mmx.asm |
| 174 |
|
src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm |
| 175 |
|
src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm |
| 176 |
|
src/motion/x86_asm/sad_xmm.asm |
| 177 |
|
src/quant/x86_asm/quantize_h263_3dne.asm |
| 178 |
|
src/quant/x86_asm/quantize_h263_mmx.asm |
| 179 |
|
src/quant/x86_asm/quantize_mpeg_mmx.asm |
| 180 |
|
src/quant/x86_asm/quantize_mpeg_xmm.asm |
| 181 |
|
src/utils/x86_asm/cpuid.asm |
| 182 |
|
src/utils/x86_asm/interlacing_mmx.asm |
| 183 |
|
src/utils/x86_asm/mem_transfer_3dne.asm |
| 184 |
|
|
| 185 |
|
|
| 186 |
|
2003-11-02 23:01:43 GMT patch-87 |
| 187 |
|
|
| 188 |
|
Summary: |
| 189 |
|
SSE2 update |
| 190 |
|
Revision: |
| 191 |
|
xvidcore--devapi4--1.0--patch-87 |
| 192 |
|
|
| 193 |
|
* Added Dmitry SSE2 iDCT code back. |
| 194 |
|
* Plugged Dmitry iDCT as default for SSE2 |
| 195 |
|
* Fixed a bug in xvid_bench that was making it would test some CPU |
| 196 |
|
instruction set w/o host CPU support. xvidcore init was simply |
| 197 |
|
discarding irrelevant cpu flags. |
| 198 |
|
|
| 199 |
|
new files: |
| 200 |
|
src/dct/x86_asm/.arch-ids/idct_sse2_dmitry.asm.id |
| 201 |
|
src/dct/x86_asm/idct_sse2_dmitry.asm |
| 202 |
|
|
| 203 |
|
modified files: |
| 204 |
|
build/generic/sources.inc build/win32/libxvidcore.dsp |
| 205 |
|
examples/xvid_bench.c src/dct/fdct.h src/dct/idct.h src/xvid.c |
| 206 |
|
|
| 207 |
|
|
| 208 |
|
2003-10-31 14:53:26 GMT patch-86 |
| 209 |
|
|
| 210 |
|
Summary: |
| 211 |
|
Better handling of old windres versions + GNU make dependency. |
| 212 |
|
Revision: |
| 213 |
|
xvidcore--devapi4--1.0--patch-86 |
| 214 |
|
|
| 215 |
|
Old versions of GNU windres (<2.14) don't have the same short options. |
| 216 |
|
But long options remain the same so it's better to use long option names |
| 217 |
|
to have ful compatibility with older versions. |
| 218 |
|
|
| 219 |
|
The Makefile appears to be dependent on GNU make because shell expansion |
| 220 |
|
for retrieving the path of the Makefile is wrong when using `` even with |
| 221 |
|
a single expansion assignment :=. It keeps being expanded when used. |
| 222 |
|
|
| 223 |
|
modified files: |
| 224 |
|
doc/INSTALL vfw/bin/Makefile |
| 225 |
|
|
| 226 |
|
|
| 227 |
|
2003-10-29 11:31:28 GMT patch-85 |
| 228 |
|
|
| 229 |
|
Summary: |
| 230 |
|
Added sse2 f/iDCT code from skal |
| 231 |
|
Revision: |
| 232 |
|
xvidcore--devapi4--1.0--patch-85 |
| 233 |
|
|
| 234 |
|
* Added sse2 f/iDCT code from skal |
| 235 |
|
* Added hooking in xvid.c |
| 236 |
|
|
| 237 |
|
new files: |
| 238 |
|
src/dct/x86_asm/.arch-ids/fdct_sse2_skal.asm.id |
| 239 |
|
src/dct/x86_asm/fdct_sse2_skal.asm |
| 240 |
|
|
| 241 |
|
modified files: |
| 242 |
|
build/generic/sources.inc build/win32/libxvidcore.dsp |
| 243 |
|
src/xvid.c |
| 244 |
|
|
| 245 |
|
|
| 246 |
|
2003-10-29 00:19:10 GMT patch-84 |
| 247 |
|
|
| 248 |
|
Summary: |
| 249 |
|
Fix the static motion detection |
| 250 |
|
Revision: |
| 251 |
|
xvidcore--devapi4--1.0--patch-84 |
| 252 |
|
|
| 253 |
|
Fix the static motion detection |
| 254 |
|
|
| 255 |
|
|
| 256 |
|
modified files: |
| 257 |
|
src/motion/estimation_pvop.c |
| 258 |
|
|
| 259 |
|
|
| 260 |
|
2003-10-28 23:39:46 GMT patch-83 |
| 261 |
|
|
| 262 |
|
Summary: |
| 263 |
|
Added cartoon option handling. |
| 264 |
|
Revision: |
| 265 |
|
xvidcore--devapi4--1.0--patch-83 |
| 266 |
|
|
| 267 |
|
Added cartoon widgets + handling code. |
| 268 |
|
|
| 269 |
|
NB: static motion detection is disabled because of crashes on P4 |
| 270 |
|
cpus. |
| 271 |
|
|
| 272 |
|
modified files: |
| 273 |
|
vfw/src/codec.c vfw/src/config.c vfw/src/config.h |
| 274 |
|
vfw/src/resource.h vfw/src/resource.rc |
| 275 |
|
|
| 276 |
|
|
| 277 |
|
2003-10-28 17:44:09 GMT patch-82 |
| 278 |
|
|
| 279 |
|
Summary: |
| 280 |
|
ASM cleanups; |
| 281 |
|
Revision: |
| 282 |
|
xvidcore--devapi4--1.0--patch-82 |
| 283 |
|
|
| 284 |
|
* Applied same style to all asm files |
| 285 |
|
* Replaced current sad sse2 operators with skal's ones |
| 286 |
|
* Removed old and unused colorspace asm files |
| 287 |
|
|
| 288 |
|
removed files: |
| 289 |
|
src/image/x86_asm/.arch-ids/rgb_to_yv12_mmx.asm.id |
| 290 |
|
src/image/x86_asm/.arch-ids/yuv_to_yv12_mmx.asm.id |
| 291 |
|
src/image/x86_asm/.arch-ids/yuyv_to_yv12_mmx.asm.id |
| 292 |
|
src/image/x86_asm/.arch-ids/yv12_to_rgb24_mmx.asm.id |
| 293 |
|
src/image/x86_asm/.arch-ids/yv12_to_rgb32_mmx.asm.id |
| 294 |
|
src/image/x86_asm/.arch-ids/yv12_to_yuyv_mmx.asm.id |
| 295 |
|
src/image/x86_asm/rgb_to_yv12_mmx.asm |
| 296 |
|
src/image/x86_asm/yuv_to_yv12_mmx.asm |
| 297 |
|
src/image/x86_asm/yuyv_to_yv12_mmx.asm |
| 298 |
|
src/image/x86_asm/yv12_to_rgb24_mmx.asm |
| 299 |
|
src/image/x86_asm/yv12_to_rgb32_mmx.asm |
| 300 |
|
src/image/x86_asm/yv12_to_yuyv_mmx.asm |
| 301 |
|
|
| 302 |
|
modified files: |
| 303 |
|
build/generic/sources.inc src/bitstream/x86_asm/cbp_3dne.asm |
| 304 |
|
src/bitstream/x86_asm/cbp_mmx.asm |
| 305 |
|
src/bitstream/x86_asm/cbp_sse2.asm |
| 306 |
|
src/dct/x86_asm/fdct_mmx_ffmpeg.asm |
| 307 |
|
src/dct/x86_asm/fdct_mmx_skal.asm |
| 308 |
|
src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm |
| 309 |
|
src/dct/x86_asm/simple_idct_mmx.asm |
| 310 |
|
src/image/x86_asm/colorspace_mmx.inc |
| 311 |
|
src/image/x86_asm/colorspace_rgb_mmx.asm |
| 312 |
|
src/image/x86_asm/colorspace_yuv_mmx.asm |
| 313 |
|
src/image/x86_asm/colorspace_yuyv_mmx.asm |
| 314 |
|
src/image/x86_asm/interpolate8x8_3dn.asm |
| 315 |
|
src/image/x86_asm/interpolate8x8_3dne.asm |
| 316 |
|
src/image/x86_asm/interpolate8x8_mmx.asm |
| 317 |
|
src/image/x86_asm/interpolate8x8_xmm.asm |
| 318 |
|
src/image/x86_asm/qpel_mmx.asm |
| 319 |
|
src/image/x86_asm/reduced_mmx.asm |
| 320 |
|
src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm |
| 321 |
|
src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm |
| 322 |
|
src/motion/x86_asm/sad_xmm.asm |
| 323 |
|
src/quant/x86_asm/quantize_h263_3dne.asm |
| 324 |
|
src/quant/x86_asm/quantize_h263_mmx.asm |
| 325 |
|
src/quant/x86_asm/quantize_mpeg_mmx.asm |
| 326 |
|
src/quant/x86_asm/quantize_mpeg_xmm.asm |
| 327 |
|
src/utils/x86_asm/cpuid.asm |
| 328 |
|
src/utils/x86_asm/interlacing_mmx.asm |
| 329 |
|
src/utils/x86_asm/mem_transfer_3dne.asm |
| 330 |
|
src/utils/x86_asm/mem_transfer_mmx.asm |
| 331 |
|
|
| 332 |
|
|
| 333 |
|
2003-10-27 01:13:47 GMT patch-81 |
| 334 |
|
|
| 335 |
|
Summary: |
| 336 |
|
d_mv_bits speedup from sysKin |
| 337 |
|
Revision: |
| 338 |
|
xvidcore--devapi4--1.0--patch-81 |
| 339 |
|
|
| 340 |
|
d_mv_bits speedup from sysKin |
| 341 |
|
|
| 342 |
|
|
| 343 |
|
modified files: |
| 344 |
|
src/motion/motion_inlines.h |
| 345 |
|
|
| 346 |
|
|
| 347 |
|
2003-10-27 00:55:51 GMT patch-80 |
| 348 |
|
|
| 349 |
|
Summary: |
| 350 |
|
fDCT changes, new asm CodingStyle applied to dct dir |
| 351 |
|
Revision: |
| 352 |
|
xvidcore--devapi4--1.0--patch-80 |
| 353 |
|
|
| 354 |
|
* Ported the ffmpeg fDCT functions (mmx and xmm). |
| 355 |
|
* Modified the skal's versions a bit to allow rolling loops. |
| 356 |
|
* Activated Skal's fDCTs (unrolled versions) for mmx _and_ xmm |
| 357 |
|
(old code was ignoring xmm versions) |
| 358 |
|
* Removed the SSE2 versions (they'll be back later) |
| 359 |
|
* .data -> .rodata |
| 360 |
|
* Applied announced asm CodingStyle to the dct dir |
| 361 |
|
(I'll have to add a section with the said CodingStyle) |
| 362 |
|
|
| 363 |
|
modified files: |
| 364 |
|
build/generic/sources.inc build/win32/libxvidcore.dsp |
| 365 |
|
src/dct/fdct.h src/dct/idct.h |
| 366 |
|
src/dct/x86_asm/fdct_mmx_ffmpeg.asm |
| 367 |
|
src/dct/x86_asm/fdct_mmx_skal.asm |
| 368 |
|
src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm |
| 369 |
|
src/dct/x86_asm/simple_idct_mmx.asm src/xvid.c |
| 370 |
|
|
| 371 |
|
renamed files: |
| 372 |
|
src/dct/x86_asm/.arch-ids/fdct_mmx.asm.id |
| 373 |
|
==> src/dct/x86_asm/.arch-ids/fdct_mmx_ffmpeg.asm.id |
| 374 |
|
src/dct/x86_asm/.arch-ids/fdct_xmm.asm.id |
| 375 |
|
==> src/dct/x86_asm/.arch-ids/fdct_mmx_skal.asm.id |
| 376 |
|
src/dct/x86_asm/fdct_mmx.asm |
| 377 |
|
==> src/dct/x86_asm/fdct_mmx_ffmpeg.asm |
| 378 |
|
src/dct/x86_asm/fdct_xmm.asm |
| 379 |
|
==> src/dct/x86_asm/fdct_mmx_skal.asm |
| 380 |
|
|
| 381 |
|
|
| 382 |
|
2003-10-25 13:48:42 GMT patch-79 |
| 383 |
|
|
| 384 |
|
Summary: |
| 385 |
|
BQuant->PQuant fix. |
| 386 |
|
Revision: |
| 387 |
|
xvidcore--devapi4--1.0--patch-79 |
| 388 |
|
|
| 389 |
|
When using closed_gop, a BFrame before an IFrame is turned into a PFrame. |
| 390 |
|
Thus is original quant has to be computed back, a rounding was causing these |
| 391 |
|
frames to be orig_quant-1. As a consequence we had very big frames before |
| 392 |
|
the IFrame, loosing many bits for nearly no visual benefit. |
| 393 |
|
|
| 394 |
|
modified files: |
| 395 |
|
src/encoder.c |
| 396 |
|
|
| 397 |
|
|
| 398 |
|
2003-10-25 10:26:48 GMT patch-78 |
| 399 |
|
|
| 400 |
|
Summary: |
| 401 |
|
Added closed gop option to xvid_encraw |
| 402 |
|
Revision: |
| 403 |
|
xvidcore--devapi4--1.0--patch-78 |
| 404 |
|
|
| 405 |
|
Added closed gop option to xvid_encraw |
| 406 |
|
|
| 407 |
|
|
| 408 |
|
modified files: |
| 409 |
|
examples/xvid_encraw.c |
| 410 |
|
|
| 411 |
|
|
| 412 |
|
2003-10-24 17:39:53 GMT patch-77 |
| 413 |
|
|
| 414 |
|
Summary: |
| 415 |
|
RD fixes. |
| 416 |
|
Revision: |
| 417 |
|
xvidcore--devapi4--1.0--patch-77 |
| 418 |
|
|
| 419 |
|
Inter RD optimization relied on buggy functions to predict bitsize. |
| 420 |
|
|
| 421 |
|
modified files: |
| 422 |
|
src/motion/estimation_pvop.c src/motion/motion_inlines.h |
| 423 |
|
|
| 424 |
|
|
| 425 |
|
2003-10-22 15:48:01 GMT patch-76 |
| 426 |
|
|
| 427 |
|
Summary: |
| 428 |
|
Small INSTALL update due to previous build patches. |
| 429 |
|
Revision: |
| 430 |
|
xvidcore--devapi4--1.0--patch-76 |
| 431 |
|
|
| 432 |
|
Small INSTALL update due to previous build patches. |
| 433 |
|
|
| 434 |
|
modified files: |
| 435 |
|
doc/INSTALL |
| 436 |
|
|
| 437 |
|
|
| 438 |
|
2003-10-21 21:27:46 GMT patch-75 |
| 439 |
|
|
| 440 |
|
Summary: |
| 441 |
|
Removed unused next_block vars. |
| 442 |
|
Revision: |
| 443 |
|
xvidcore--devapi4--1.0--patch-75 |
| 444 |
|
|
| 445 |
|
Removed unused next_block vars. |
| 446 |
|
|
| 447 |
|
modified files: |
| 448 |
|
src/decoder.c |
| 449 |
|
|
| 450 |
|
|
| 451 |
|
2003-10-21 21:24:15 GMT patch-74 |
| 452 |
|
|
| 453 |
|
Summary: |
| 454 |
|
VFW build changes. |
| 455 |
|
Revision: |
| 456 |
|
xvidcore--devapi4--1.0--patch-74 |
| 457 |
|
|
| 458 |
|
The build system has been modified to look like the core lib one |
| 459 |
|
minus the configure system. |
| 460 |
|
|
| 461 |
|
modified files: |
| 462 |
|
vfw/bin/Makefile vfw/bin/sources.inc vfw/src/config.c |
| 463 |
|
vfw/vfw.dsp |
| 464 |
|
|
| 465 |
|
renamed files: |
| 466 |
|
vfw/bin/.arch-ids/Makefile.cygwin.id |
| 467 |
|
==> vfw/bin/.arch-ids/Makefile.id |
| 468 |
|
vfw/bin/.arch-ids/Makefile.inc.id |
| 469 |
|
==> vfw/bin/.arch-ids/sources.inc.id |
| 470 |
|
vfw/bin/Makefile.cygwin |
| 471 |
|
==> vfw/bin/Makefile |
| 472 |
|
vfw/bin/Makefile.inc |
| 473 |
|
==> vfw/bin/sources.inc |
| 474 |
|
vfw/src/.arch-ids/config.rc.id |
| 475 |
|
==> vfw/src/.arch-ids/resource.rc.id |
| 476 |
|
vfw/src/config.rc |
| 477 |
|
==> vfw/src/resource.rc |
| 478 |
|
|
| 479 |
|
|
| 480 |
|
2003-10-21 17:00:09 GMT patch-73 |
| 481 |
|
|
| 482 |
|
Summary: |
| 483 |
|
Decoder cleanups and speedup |
| 484 |
|
Revision: |
| 485 |
|
xvidcore--devapi4--1.0--patch-73 |
| 486 |
|
|
| 487 |
|
cleanups, speedups from sysKin |
| 488 |
|
|
| 489 |
|
modified files: |
| 490 |
|
src/decoder.c |
| 491 |
|
|
| 492 |
|
|
| 493 |
|
2003-10-21 16:22:15 GMT patch-72 |
| 494 |
|
|
| 495 |
|
Summary: |
| 496 |
|
Build fixes for newer nasm versions. |
| 497 |
|
Revision: |
| 498 |
|
xvidcore--devapi4--1.0--patch-72 |
| 499 |
|
|
| 500 |
|
nasm does not take care of adding trailing slashes to include paths. |
| 501 |
|
A patch to upstream authors has been refused because "the backslash() |
| 502 |
|
feature has been abandoned to get back to old nasm behavior" |
| 503 |
|
|
| 504 |
|
Their choice is kinda stupid as nasm is now open to user mistakes... :\ |
| 505 |
|
So we fix that on ou side. |
| 506 |
|
|
| 507 |
|
modified files: |
| 508 |
|
build/generic/configure.in build/win32/libxvidcore.dsp |
| 509 |
|
|
| 510 |
|
|
| 511 |
|
2003-10-17 15:13:12 GMT patch-71 |
| 512 |
|
|
| 513 |
|
Summary: |
| 514 |
|
Updated docs. |
| 515 |
|
Revision: |
| 516 |
|
xvidcore--devapi4--1.0--patch-71 |
| 517 |
|
|
| 518 |
|
The doc of devapi4 is mostly out dated, i t is much better not to |
| 519 |
|
keep it in the repository at the moment. We'll add new docs later. |
| 520 |
|
|
| 521 |
|
Added a INSTALL doc that explains the build/install process for |
| 522 |
|
supported platforms. It's a first try, things may be added later. |
| 523 |
|
|
| 524 |
|
new files: |
| 525 |
|
doc/.arch-ids/INSTALL.id doc/INSTALL |
| 526 |
|
|
| 527 |
|
removed files: |
| 528 |
|
doc/.arch-ids/API.dox.id doc/.arch-ids/Makefile.id |
| 529 |
|
doc/.arch-ids/foot.inc.in.id doc/.arch-ids/header.tex.in.id |
| 530 |
|
doc/.arch-ids/xvid-decoding.txt.id |
| 531 |
|
doc/.arch-ids/xvid-encoder.txt.id doc/API.dox doc/Makefile |
| 532 |
|
doc/foot.inc.in doc/header.tex.in doc/xvid-decoding.txt |
| 533 |
|
doc/xvid-encoder.txt |
| 534 |
|
|
| 535 |
|
modified files: |
| 536 |
|
CodingStyle README doc/README |
| 537 |
|
|
| 538 |
|
renamed files: |
| 539 |
|
.arch-ids/README.txt.id |
| 540 |
|
==> .arch-ids/README.id |
| 541 |
|
.arch-ids/changelog.txt.id |
| 542 |
|
==> .arch-ids/ChangeLog.id |
| 543 |
|
README.txt |
| 544 |
|
==> README |
| 545 |
|
changelog.txt |
| 546 |
|
==> ChangeLog |
| 547 |
|
|
| 548 |
|
|
| 549 |
|
2003-10-15 13:53:11 GMT patch-70 |
| 550 |
|
|
| 551 |
|
Summary: |
| 552 |
|
Better cross compilation handling. |
| 553 |
|
Revision: |
| 554 |
|
xvidcore--devapi4--1.0--patch-70 |
| 555 |
|
|
| 556 |
|
With this patch it is now possible to cross compile xvid |
| 557 |
|
quite easily for win32 platform on a build linux host. |
| 558 |
|
|
| 559 |
|
Recipe for debian system: |
| 560 |
|
$ apt-get install mingw32 |
| 561 |
|
(or create your own cross compiler/binutils suite and install mingw32 |
| 562 |
|
header files -- sorry i don't have a recipe for this, this is let as |
| 563 |
|
an exercice for the reader) |
| 564 |
|
$ cd ${xvidcore} |
| 565 |
|
$ cd build/generic |
| 566 |
|
$ ./bootstrap.sh |
| 567 |
|
$ ./configure --host=i586-mingw32msvc |
| 568 |
|
(all occurences of i586-mingw32msvc may be replaced with the right |
| 569 |
|
prefix you've choosen for your cross compiler and cross binutils) |
| 570 |
|
$ make |
| 571 |
|
$ cd ../../vfw/bin |
| 572 |
|
$ make -f Makefile.cygwin \ |
| 573 |
|
CC=i586-mingw32msvc-gcc WINDRES=i586-mingw32msvc-windres |
| 574 |
|
|
| 575 |
|
Enjoy your win32 xvid.dll build by free software, on a free OS, for a devil |
| 576 |
|
OS target. |
| 577 |
|
|
| 578 |
|
modified files: |
| 579 |
|
build/generic/configure.in vfw/bin/Makefile.cygwin |
| 580 |
|
|
| 581 |
|
|
| 582 |
|
2003-10-14 15:17:28 GMT patch-69 |
| 583 |
|
|
| 584 |
|
Summary: |
| 585 |
|
Fixed Qpel+Interpolation decoding. Cleaned up mb->mode usage. |
| 586 |
|
Revision: |
| 587 |
|
xvidcore--devapi4--1.0--patch-69 |
| 588 |
|
|
| 589 |
|
* Fixed interpolate mode + qpel decoding. |
| 590 |
|
* MB->mb_type completely replaced by MB->mode |
| 591 |
|
|
| 592 |
|
modified files: |
| 593 |
|
src/decoder.c |
| 594 |
|
|
| 595 |
|
|
| 596 |
|
2003-10-12 21:57:24 GMT patch-68 |
| 597 |
|
|
| 598 |
|
Summary: |
| 599 |
|
ac/dc prediction for intra RD search. |
| 600 |
|
Revision: |
| 601 |
|
xvidcore--devapi4--1.0--patch-68 |
| 602 |
|
|
| 603 |
|
From syskin, added real ac/dc prediction for INTRA's bitcount. |
| 604 |
|
|
| 605 |
|
modified files: |
| 606 |
|
src/motion/estimation_rd_based.c |
| 607 |
|
|
| 608 |
|
|
| 609 |
|
2003-10-09 18:15:50 GMT patch-67 |
| 610 |
|
|
| 611 |
|
Summary: |
| 612 |
|
Pigrated asm code to new quant API. |
| 613 |
|
Revision: |
| 614 |
|
xvidcore--devapi4--1.0--patch-67 |
| 615 |
|
|
| 616 |
|
Many changes that are mostly cosmetic in the asm files. |
| 617 |
|
* indent |
| 618 |
|
* added xor eax, eax in quant_(h263|mpeg)_intra_.* functions |
| 619 |
|
(just to make sure the returned value isn't random) |
| 620 |
|
* added xor eax, eax in dequant_(h263|mpeg)_.* functions |
| 621 |
|
(just to make sure the returned value isn't random) |
| 622 |
|
* synced cpuid.asm XVID_CPU_feature constants with the one |
| 623 |
|
defined in the C code (xvid.h) |
| 624 |
|
* enabled all cpu tests in xvid_bench.c |
| 625 |
|
|
| 626 |
|
modified files: |
| 627 |
|
examples/xvid_bench.c src/quant/quant_h263.c |
| 628 |
|
src/quant/quant_mpeg.c |
| 629 |
|
src/quant/x86_asm/quantize_h263_3dne.asm |
| 630 |
|
src/quant/x86_asm/quantize_h263_mmx.asm |
| 631 |
|
src/quant/x86_asm/quantize_mpeg_mmx.asm |
| 632 |
|
src/quant/x86_asm/quantize_mpeg_xmm.asm |
| 633 |
|
src/utils/x86_asm/cpuid.asm src/xvid.h |
| 634 |
|
|
| 635 |
|
|
| 636 |
|
2003-10-08 21:05:47 GMT patch-66 |
| 637 |
|
|
| 638 |
|
Summary: |
| 639 |
|
Updated xvid_bench for quant API changes |
| 640 |
|
Revision: |
| 641 |
|
xvidcore--devapi4--1.0--patch-66 |
| 642 |
|
|
| 643 |
|
Updated xvid_bench for quant API changes |
| 644 |
|
|
| 645 |
|
|
| 646 |
|
modified files: |
| 647 |
|
examples/xvid_bench.c |
| 648 |
|
|
| 649 |
|
|
| 650 |
|
2003-10-07 13:03:51 GMT patch-65 |
| 651 |
|
|
| 652 |
|
Summary: |
| 653 |
|
Quant functions API changes (first step) |
| 654 |
|
Revision: |
| 655 |
|
xvidcore--devapi4--1.0--patch-65 |
| 656 |
|
|
| 657 |
|
In the road to instance safe mpeg quantization, a small cleanup |
| 658 |
|
to the quant API was needed. It consists in changing the way we |
| 659 |
|
name the functions quant_{mpeg|h263}_{inter|intra}_{arch} and in |
| 660 |
|
a move to a more unified API (even intra functions return the sum |
| 661 |
|
of coefficients, it can be used as a complexity measure at a later |
| 662 |
|
time). |
| 663 |
|
|
| 664 |
|
This patch touch lot of files, but all changes are trivial. |
| 665 |
|
|
| 666 |
|
NB: we should check the IA64 asm validity, i changed things but |
| 667 |
|
i can't test them. |
| 668 |
|
|
| 669 |
|
new files: |
| 670 |
|
src/quant/.arch-ids/quant.h.id src/quant/quant.h |
| 671 |
|
|
| 672 |
|
removed files: |
| 673 |
|
src/quant/.arch-ids/quant_h263.h.id |
| 674 |
|
src/quant/.arch-ids/quant_mpeg4.h.id src/quant/quant_h263.h |
| 675 |
|
src/quant/quant_mpeg4.h |
| 676 |
|
|
| 677 |
|
modified files: |
| 678 |
|
build/generic/sources.inc build/win32/libxvidcore.dsp |
| 679 |
|
src/decoder.c src/encoder.c src/image/qpel.c |
| 680 |
|
src/motion/estimation_rd_based.c |
| 681 |
|
src/quant/ia64_asm/quant_h263_ia64.s src/quant/quant_h263.c |
| 682 |
|
src/quant/quant_matrix.c src/quant/quant_matrix.h |
| 683 |
|
src/quant/quant_mpeg.c |
| 684 |
|
src/quant/x86_asm/quantize_h263_3dne.asm |
| 685 |
|
src/quant/x86_asm/quantize_h263_mmx.asm |
| 686 |
|
src/quant/x86_asm/quantize_mpeg_mmx.asm |
| 687 |
|
src/quant/x86_asm/quantize_mpeg_xmm.asm |
| 688 |
|
src/utils/mbtransquant.c src/xvid.c src/xvid.h |
| 689 |
|
|
| 690 |
|
renamed files: |
| 691 |
|
src/quant/.arch-ids/quant_mpeg4.c.id |
| 692 |
|
==> src/quant/.arch-ids/quant_mpeg.c.id |
| 693 |
|
src/quant/quant_mpeg4.c |
| 694 |
|
==> src/quant/quant_mpeg.c |
| 695 |
|
src/quant/x86_asm/.arch-ids/quantize4_mmx.asm.id |
| 696 |
|
==> src/quant/x86_asm/.arch-ids/quantize_mpeg_mmx.asm.id |
| 697 |
|
src/quant/x86_asm/.arch-ids/quantize4_xmm.asm.id |
| 698 |
|
==> src/quant/x86_asm/.arch-ids/quantize_mpeg_xmm.asm.id |
| 699 |
|
src/quant/x86_asm/.arch-ids/quantize_3dne.asm.id |
| 700 |
|
==> src/quant/x86_asm/.arch-ids/quantize_h263_3dne.asm.id |
| 701 |
|
src/quant/x86_asm/.arch-ids/quantize_mmx.asm.id |
| 702 |
|
==> src/quant/x86_asm/.arch-ids/quantize_h263_mmx.asm.id |
| 703 |
|
src/quant/x86_asm/quantize4_mmx.asm |
| 704 |
|
==> src/quant/x86_asm/quantize_mpeg_mmx.asm |
| 705 |
|
src/quant/x86_asm/quantize4_xmm.asm |
| 706 |
|
==> src/quant/x86_asm/quantize_mpeg_xmm.asm |
| 707 |
|
src/quant/x86_asm/quantize_3dne.asm |
| 708 |
|
==> src/quant/x86_asm/quantize_h263_3dne.asm |
| 709 |
|
src/quant/x86_asm/quantize_mmx.asm |
| 710 |
|
==> src/quant/x86_asm/quantize_h263_mmx.asm |
| 711 |
|
|
| 712 |
|
|
| 713 |
|
2003-10-05 00:15:15 GMT patch-64 |
| 714 |
|
|
| 715 |
|
Summary: |
| 716 |
|
Updated ChangeLog |
| 717 |
|
Revision: |
| 718 |
|
xvidcore--devapi4--1.0--patch-64 |
| 719 |
|
|
| 720 |
|
Updated ChangeLog |
| 721 |
|
|
| 722 |
|
|
| 723 |
|
modified files: |
| 724 |
|
changelog.txt |
| 725 |
|
|
| 726 |
|
|
| 727 |
2003-10-04 16:04:30 GMT patch-63 |
2003-10-04 16:04:30 GMT patch-63 |
| 728 |
|
|
| 729 |
Summary: |
Summary: |
Parent Directory
| 
Revision Log
| 
Patch