| 4 |
# relevant. |
# relevant. |
| 5 |
|
|
| 6 |
|
|
| 7 |
|
######################################################################### |
| 8 |
|
# 1.0.3 (Bitstream Version 37) |
| 9 |
|
######################################################################### |
| 10 |
|
|
| 11 |
|
2004-12-19 11:25:10 GMT patch-63 |
| 12 |
|
|
| 13 |
|
Summary: |
| 14 |
|
Trellis overflow for quant<=2 |
| 15 |
|
Revision: |
| 16 |
|
xvidcore--stable--1.0--patch-63 |
| 17 |
|
|
| 18 |
|
From skal: |
| 19 |
|
* Don't call trellis optimization if quant <= 2 as the code |
| 20 |
|
overflows |
| 21 |
|
|
| 22 |
|
modified files: |
| 23 |
|
src/utils/mbtransquant.c |
| 24 |
|
|
| 25 |
|
|
| 26 |
|
2004-11-24 21:25:35 GMT patch-62 |
| 27 |
|
|
| 28 |
|
Summary: |
| 29 |
|
Fixed stride in vfw frontend. |
| 30 |
|
Revision: |
| 31 |
|
xvidcore--stable--1.0--patch-62 |
| 32 |
|
|
| 33 |
|
From pete: |
| 34 |
|
* Fixed the way stride is computed in the VFW frontend. |
| 35 |
|
(Same cure as for the DShow frontend) |
| 36 |
|
|
| 37 |
|
modified files: |
| 38 |
|
vfw/src/codec.c |
| 39 |
|
|
| 40 |
|
|
| 41 |
|
2004-11-24 21:09:45 GMT patch-61 |
| 42 |
|
|
| 43 |
|
Summary: |
| 44 |
|
Fixed stride in DShow decoder. |
| 45 |
|
Revision: |
| 46 |
|
xvidcore--stable--1.0--patch-61 |
| 47 |
|
|
| 48 |
|
From pete: |
| 49 |
|
* Fixed the way stride is computed in DShow filter |
| 50 |
|
|
| 51 |
|
modified files: |
| 52 |
|
dshow/src/CXvidDecoder.cpp |
| 53 |
|
|
| 54 |
|
|
| 55 |
|
2004-11-24 21:05:54 GMT patch-60 |
| 56 |
|
|
| 57 |
|
Summary: |
| 58 |
|
Fixed DiamondSearch |
| 59 |
|
Revision: |
| 60 |
|
xvidcore--stable--1.0--patch-60 |
| 61 |
|
|
| 62 |
|
From sysKin: |
| 63 |
|
* Fixed DiamondSearch, wrong directions were used in two cases. |
| 64 |
|
|
| 65 |
|
modified files: |
| 66 |
|
src/motion/estimation_common.c |
| 67 |
|
|
| 68 |
|
|
| 69 |
|
2004-10-12 20:59:17 GMT patch-59 |
| 70 |
|
|
| 71 |
|
Summary: |
| 72 |
|
Don't read too short streams. |
| 73 |
|
Revision: |
| 74 |
|
xvidcore--stable--1.0--patch-59 |
| 75 |
|
|
| 76 |
|
From sysKin: |
| 77 |
|
* Dont even try to read bitstreams shorter than 4 bytes |
| 78 |
|
(nb: 4 bytes == size of startcodes). |
| 79 |
|
|
| 80 |
|
modified files: |
| 81 |
|
src/bitstream/bitstream.c |
| 82 |
|
|
| 83 |
|
|
| 84 |
|
2004-10-12 20:33:59 GMT patch-58 |
| 85 |
|
|
| 86 |
|
Summary: |
| 87 |
|
64bit fixes |
| 88 |
|
Revision: |
| 89 |
|
xvidcore--stable--1.0--patch-58 |
| 90 |
|
|
| 91 |
|
From Andre Werthmann (wertmann at aei dot mpg dot de): |
| 92 |
|
- uint vs int cleanups for addresses. This fixes various problems |
| 93 |
|
for 64bit platforms. |
| 94 |
|
|
| 95 |
|
modified files: |
| 96 |
|
src/image/interpolate8x8.h src/image/qpel.h |
| 97 |
|
src/motion/estimation_bvop.c src/motion/motion_comp.c |
| 98 |
|
|
| 99 |
|
|
| 100 |
|
2004-10-12 19:22:53 GMT patch-57 |
| 101 |
|
|
| 102 |
|
Summary: |
| 103 |
|
ME fix. |
| 104 |
|
Revision: |
| 105 |
|
xvidcore--stable--1.0--patch-57 |
| 106 |
|
|
| 107 |
|
From gruel: |
| 108 |
|
* Diamond search sets iDirection to 0 preventing further searches. |
| 109 |
|
|
| 110 |
|
modified files: |
| 111 |
|
src/motion/estimation_common.c |
| 112 |
|
|
| 113 |
|
|
| 114 |
|
######################################################################### |
| 115 |
|
# 1.0.2 (Bitstream Version 36) |
| 116 |
|
######################################################################### |
| 117 |
|
|
| 118 |
|
2004-08-29 11:35:02 GMT patch-56 |
| 119 |
|
|
| 120 |
|
Summary: |
| 121 |
|
ChangeLog update |
| 122 |
|
Revision: |
| 123 |
|
xvidcore--stable--1.0--patch-56 |
| 124 |
|
|
| 125 |
|
ChangeLog update |
| 126 |
|
|
| 127 |
|
|
| 128 |
|
modified files: |
| 129 |
|
ChangeLog |
| 130 |
|
|
| 131 |
|
|
| 132 |
|
2004-08-29 11:24:26 GMT patch-55 |
| 133 |
|
|
| 134 |
|
Summary: |
| 135 |
|
Merged one important forgotten bugfix from head |
| 136 |
|
Revision: |
| 137 |
|
xvidcore--stable--1.0--patch-55 |
| 138 |
|
|
| 139 |
|
Merged one important forgotten bugfix from head |
| 140 |
|
|
| 141 |
|
Patches applied: |
| 142 |
|
|
| 143 |
|
* ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-70 |
| 144 |
|
Out of bounds MVs clipping |
| 145 |
|
|
| 146 |
|
* ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-71 |
| 147 |
|
Decoder optimization (fixing regression) |
| 148 |
|
|
| 149 |
|
|
| 150 |
|
modified files: |
| 151 |
|
src/decoder.c |
| 152 |
|
|
| 153 |
|
new patches: |
| 154 |
|
ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-70 |
| 155 |
|
ed.gomez@free.fr--2004-1/xvidcore--head--0.0--patch-71 |
| 156 |
|
|
| 157 |
|
|
| 158 |
|
2004-08-29 10:51:58 GMT patch-54 |
| 159 |
|
|
| 160 |
|
Summary: |
| 161 |
|
Marking 1.0.2 |
| 162 |
|
Revision: |
| 163 |
|
xvidcore--stable--1.0--patch-54 |
| 164 |
|
|
| 165 |
|
From ed.gomez: |
| 166 |
|
* Marking 1.0.2 |
| 167 |
|
|
| 168 |
|
modified files: |
| 169 |
|
ChangeLog build/generic/configure.in src/xvid.c src/xvid.h |
| 170 |
|
|
| 171 |
|
|
| 172 |
|
2004-08-22 13:08:44 GMT patch-53 |
| 173 |
|
|
| 174 |
|
Summary: |
| 175 |
|
Thread safety problem in idct C version |
| 176 |
|
Revision: |
| 177 |
|
xvidcore--stable--1.0--patch-53 |
| 178 |
|
|
| 179 |
|
From ed.gomez: |
| 180 |
|
* Fixed a thread safety problem in C version of the |
| 181 |
|
idct function. Added some comments on some static |
| 182 |
|
data not marked as RO. |
| 183 |
|
|
| 184 |
|
modified files: |
| 185 |
|
src/bitstream/mbcoding.c src/dct/idct.c |
| 186 |
|
|
| 187 |
|
|
| 188 |
|
2004-08-21 11:45:55 GMT patch-52 |
| 189 |
|
|
| 190 |
|
Summary: |
| 191 |
|
Stupid typo+error in fdct_xxx_skal macro generator. |
| 192 |
|
Revision: |
| 193 |
|
xvidcore--stable--1.0--patch-52 |
| 194 |
|
|
| 195 |
|
From Nicolas Boulay: |
| 196 |
|
* Found a typo mistake (ecx->eax) and an error in the same line |
| 197 |
|
But as we're lucky, the unrolled version was bugfree, and that |
| 198 |
|
is that one which is used. |
| 199 |
|
|
| 200 |
|
modified files: |
| 201 |
|
src/dct/x86_asm/fdct_mmx_skal.asm |
| 202 |
|
|
| 203 |
|
|
| 204 |
|
2004-07-26 20:21:24 GMT patch-51 |
| 205 |
|
|
| 206 |
|
Summary: |
| 207 |
|
ChangeLog Update |
| 208 |
|
Revision: |
| 209 |
|
xvidcore--stable--1.0--patch-51 |
| 210 |
|
|
| 211 |
|
ChangeLog Update |
| 212 |
|
|
| 213 |
|
|
| 214 |
|
modified files: |
| 215 |
|
ChangeLog |
| 216 |
|
|
| 217 |
|
|
| 218 |
|
2004-07-24 11:33:57 GMT patch-50 |
| 219 |
|
|
| 220 |
|
Summary: |
| 221 |
|
BVOP direct/interpolated ref block rounding fix. |
| 222 |
|
Revision: |
| 223 |
|
xvidcore--stable--1.0--patch-50 |
| 224 |
|
|
| 225 |
|
From ed.gomez: |
| 226 |
|
* BVOP direct/interpolated ref block rounding fix. |
| 227 |
|
It's been using rounding=1 for averaging stage since ever. |
| 228 |
|
The standard says it's rounding=0. |
| 229 |
|
|
| 230 |
|
See standard clause 7.6.9.4 for explicit code and Section 6.3.5 |
| 231 |
|
that says "rounding=0" in bframes as they don't set the |
| 232 |
|
vop_rounding_type in VOP header. Both sections match, xvid was |
| 233 |
|
wrong |
| 234 |
|
|
| 235 |
|
modified files: |
| 236 |
|
src/decoder.c |
| 237 |
|
|
| 238 |
|
|
| 239 |
|
2004-07-23 20:37:09 GMT patch-49 |
| 240 |
|
|
| 241 |
|
Summary: |
| 242 |
|
Removed data qualifer in .rodata |
| 243 |
|
Revision: |
| 244 |
|
xvidcore--stable--1.0--patch-49 |
| 245 |
|
|
| 246 |
|
From ed.gomez: |
| 247 |
|
* long standing warning by yasm... data isn't a keyword for |
| 248 |
|
.(ro)data sections. |
| 249 |
|
|
| 250 |
|
modified files: |
| 251 |
|
src/bitstream/x86_asm/cbp_mmx.asm |
| 252 |
|
src/bitstream/x86_asm/cbp_sse2.asm |
| 253 |
|
src/dct/x86_asm/fdct_mmx_ffmpeg.asm |
| 254 |
|
src/dct/x86_asm/fdct_mmx_skal.asm |
| 255 |
|
src/dct/x86_asm/fdct_sse2_skal.asm |
| 256 |
|
src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm |
| 257 |
|
src/dct/x86_asm/idct_sse2_dmitry.asm |
| 258 |
|
src/dct/x86_asm/simple_idct_mmx.asm |
| 259 |
|
src/image/x86_asm/colorspace_rgb_mmx.asm |
| 260 |
|
src/image/x86_asm/colorspace_yuyv_mmx.asm |
| 261 |
|
src/image/x86_asm/interpolate8x8_3dn.asm |
| 262 |
|
src/image/x86_asm/interpolate8x8_3dne.asm |
| 263 |
|
src/image/x86_asm/interpolate8x8_mmx.asm |
| 264 |
|
src/image/x86_asm/interpolate8x8_xmm.asm |
| 265 |
|
src/image/x86_asm/qpel_mmx.asm |
| 266 |
|
src/image/x86_asm/reduced_mmx.asm |
| 267 |
|
src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm |
| 268 |
|
src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm |
| 269 |
|
src/motion/x86_asm/sad_xmm.asm |
| 270 |
|
src/quant/x86_asm/quantize_h263_3dne.asm |
| 271 |
|
src/quant/x86_asm/quantize_h263_mmx.asm |
| 272 |
|
src/quant/x86_asm/quantize_mpeg_mmx.asm |
| 273 |
|
src/quant/x86_asm/quantize_mpeg_xmm.asm |
| 274 |
|
src/utils/x86_asm/cpuid.asm |
| 275 |
|
src/utils/x86_asm/interlacing_mmx.asm |
| 276 |
|
src/utils/x86_asm/mem_transfer_3dne.asm |
| 277 |
|
src/utils/x86_asm/mem_transfer_mmx.asm |
| 278 |
|
|
| 279 |
|
|
| 280 |
|
2004-07-19 18:45:14 GMT patch-48 |
| 281 |
|
|
| 282 |
|
Summary: |
| 283 |
|
Complete previous xvid_decraw patch |
| 284 |
|
Revision: |
| 285 |
|
xvidcore--stable--1.0--patch-48 |
| 286 |
|
|
| 287 |
|
Complete previous xvid_decraw patch |
| 288 |
|
|
| 289 |
|
|
| 290 |
|
modified files: |
| 291 |
|
examples/xvid_decraw.c |
| 292 |
|
|
| 293 |
|
|
| 294 |
|
2004-07-18 11:58:48 GMT patch-47 |
| 295 |
|
|
| 296 |
|
Summary: |
| 297 |
|
ISO C99'ism fix |
| 298 |
|
Revision: |
| 299 |
|
xvidcore--stable--1.0--patch-47 |
| 300 |
|
|
| 301 |
|
ISO C99'ism fix |
| 302 |
|
|
| 303 |
|
|
| 304 |
|
modified files: |
| 305 |
|
src/encoder.c |
| 306 |
|
|
| 307 |
|
|
| 308 |
|
2004-07-17 11:32:42 GMT patch-46 |
| 309 |
|
|
| 310 |
|
Summary: |
| 311 |
|
Make sure time incr is never larger than 16bit. |
| 312 |
|
Revision: |
| 313 |
|
xvidcore--stable--1.0--patch-46 |
| 314 |
|
|
| 315 |
|
From ed.gomez: |
| 316 |
|
* Keep both fbase and fincr under 16bit limit. |
| 317 |
|
|
| 318 |
|
modified files: |
| 319 |
|
src/encoder.c |
| 320 |
|
|
| 321 |
|
|
| 322 |
|
2004-07-17 10:00:42 GMT patch-45 |
| 323 |
|
|
| 324 |
|
Summary: |
| 325 |
|
Future version interoperability |
| 326 |
|
Revision: |
| 327 |
|
xvidcore--stable--1.0--patch-45 |
| 328 |
|
|
| 329 |
|
From ed.gomez: |
| 330 |
|
* Zeroing the structures is the best way not to pass |
| 331 |
|
wrong data when dealing with slightly new XviD (like |
| 332 |
|
head). If we don't zero here, then brightness is not |
| 333 |
|
initialized in HEAD leading to crash. |
| 334 |
|
|
| 335 |
|
modified files: |
| 336 |
|
examples/xvid_decraw.c |
| 337 |
|
|
| 338 |
|
|
| 339 |
|
2004-07-10 17:30:40 GMT patch-44 |
| 340 |
|
|
| 341 |
|
Summary: |
| 342 |
|
ChangeLog update |
| 343 |
|
Revision: |
| 344 |
|
xvidcore--stable--1.0--patch-44 |
| 345 |
|
|
| 346 |
|
ChangeLog update |
| 347 |
|
|
| 348 |
|
|
| 349 |
|
modified files: |
| 350 |
|
ChangeLog |
| 351 |
|
|
| 352 |
|
|
| 353 |
2004-07-10 17:27:06 GMT patch-43 |
2004-07-10 17:27:06 GMT patch-43 |
| 354 |
|
|
| 355 |
Summary: |
Summary: |
Parent Directory
| 
Revision Log
| 
Patch