API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Jul 18th, 2019
3,125
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 396.42 KB | None | 0 0
raw download clone embed print report
  1. commit 3759fcb7b48037a5169715ab89f80a0ab4801cdf [revision 2984]
  2. Author: Anton Mitrofanov <[email protected]>
  3. AuthorDate: Wed May 8 19:19:11 2019 +0300
  4. Commit: Anton Mitrofanov <[email protected]>
  5. CommitDate: Wed Jul 17 20:19:23 2019 +0300
  6.  
  7. Remove CRT objects use between DLL boundaries
  8.  
  9. Fix crash of MSVC builds compiled with --system-libx264 and /MT (default) CRT.
  10.  
  11. commit 76c5afc25b331cf98c63c6e313a90cd98c575858 [revision 2983]
  12. Author: Anton Mitrofanov <[email protected]>
  13. AuthorDate: Mon Apr 22 22:18:01 2019 +0300
  14. Commit: Anton Mitrofanov <[email protected]>
  15. CommitDate: Wed Jul 17 20:19:23 2019 +0300
  16.  
  17. Fix MSVS build with ./configure --enable-shared --system-libx264
  18.  
  19. commit a615f027ed172e2dd5380e736d487aa858a0c4ff [revision 2982]
  20. Author: Anton Mitrofanov <[email protected]>
  21. AuthorDate: Fri Mar 29 17:53:14 2019 +0300
  22. Commit: Anton Mitrofanov <[email protected]>
  23. CommitDate: Wed Jul 17 20:19:23 2019 +0300
  24.  
  25. Mark explicitly DSO public API symbols and hide all other by -fvisibility=hidden
  26.  
  27. Removes need for -Bsymbolic during linking.
  28.  
  29. commit b5bc5d69c580429ff716bafcd43655e855c31b02 [revision 2981]
  30. Author: Henrik Gramner <[email protected]>
  31. AuthorDate: Sat Mar 30 17:47:25 2019 +0100
  32. Commit: Anton Mitrofanov <[email protected]>
  33. CommitDate: Wed Jul 17 20:15:35 2019 +0300
  34.  
  35. x86: Perform stack realignment in C instead of assembly
  36.  
  37. Simplifies a lot of code and avoids having to export public asm functions.
  38.  
  39. Note that the force_align_arg_pointer function attribute is broken in clang
  40. versions prior to 6.0.1 which may result in crashes, so make sure to either
  41. use a newer clang version or a different compiler.
  42.  
  43. commit 34c06d1c17ad968fbdda153cb772f77ee31b3095 [revision 2980]
  44. Author: Anton Mitrofanov <[email protected]>
  45. AuthorDate: Fri Jul 12 15:23:29 2019 +0300
  46. Commit: Anton Mitrofanov <[email protected]>
  47. CommitDate: Wed Jul 17 20:15:34 2019 +0300
  48.  
  49. Strip git-hash from version in x264.pc
  50.  
  51. pkg-config doesn't like spaces in version string.
  52.  
  53. commit f9af2a0f71d0fca7c1cafa7657f03a302da0ca1c [revision 2979]
  54. Author: Anton Mitrofanov <[email protected]>
  55. AuthorDate: Mon Jul 8 15:46:56 2019 +0300
  56. Commit: Anton Mitrofanov <[email protected]>
  57. CommitDate: Wed Jul 17 20:15:34 2019 +0300
  58.  
  59. Revert r2959: Signal Progressive and Constrained profiles
  60.  
  61. Some hardware decoders reject to decode streams with non-zero
  62. constraint_set4_flag/constraint_set5_flag.
  63.  
  64. commit 6d4947083a712c7dc2efca569c8149ffc8667eda [revision 2978]
  65. Author: Anton Mitrofanov <[email protected]>
  66. AuthorDate: Fri Jun 14 19:57:36 2019 +0300
  67. Commit: Anton Mitrofanov <[email protected]>
  68. CommitDate: Wed Jul 17 20:15:34 2019 +0300
  69.  
  70. Fix x264_picture_alloc with X264_CSP_I400 colorspace
  71.  
  72. commit 6b1170cbbd4f5cf3170d9d79aa1182e863188b04 [revision 2977]
  73. Author: Anton Mitrofanov <[email protected]>
  74. AuthorDate: Wed May 8 17:52:15 2019 +0300
  75. Commit: Anton Mitrofanov <[email protected]>
  76. CommitDate: Wed Jul 17 20:15:34 2019 +0300
  77.  
  78. Shut up UBSan about uninitialized data read
  79.  
  80. Result was never used in that case.
  81.  
  82. commit f06062f51bc5928e6a364598357dbea2d7b83cd2 [revision 2976]
  83. Author: Anton Mitrofanov <[email protected]>
  84. AuthorDate: Mon Apr 22 21:41:43 2019 +0300
  85. Commit: Anton Mitrofanov <[email protected]>
  86. CommitDate: Wed Jul 17 20:15:34 2019 +0300
  87.  
  88. Fix integer overflow detected by UBSan in --weightp analysis
  89.  
  90. Bug report by Xuezhi Yan.
  91.  
  92. commit 3147fa431627f1a00e54c8701d5ac07f1857c981 [revision 2975]
  93. Author: Anton Mitrofanov <[email protected]>
  94. AuthorDate: Fri Apr 12 15:40:01 2019 +0300
  95. Commit: Anton Mitrofanov <[email protected]>
  96. CommitDate: Wed Jul 17 20:15:34 2019 +0300
  97.  
  98. checkasm: Fix heap-buffer-overflow read detected by ASan
  99.  
  100. commit 6381798d2d1339c0535732a764096b5345607981 [revision 2974]
  101. Author: Anton Mitrofanov <[email protected]>
  102. AuthorDate: Fri Apr 12 15:38:08 2019 +0300
  103. Commit: Anton Mitrofanov <[email protected]>
  104. CommitDate: Wed Jul 17 20:15:34 2019 +0300
  105.  
  106. Fix heap-buffer-overflow read detected by ASan with interlaced encoding
  107.  
  108. Bug report by Hongxu Chen.
  109.  
  110. commit bd8a88be426baa903427a10de9f9ddb5e7c32812 [revision 2973]
  111. Author: Konstantin Pavlov <[email protected]>
  112. AuthorDate: Tue Jul 16 22:38:32 2019 +0300
  113. Commit: Konstantin Pavlov <[email protected]>
  114. CommitDate: Wed Jul 17 00:06:53 2019 +0300
  115.  
  116. CI: Bump macos target to darwin18
  117.  
  118. commit 352c02634d6d004c1d79ff5ccbbd2414ad32b67c [revision 2972]
  119. Author: Konstantin Pavlov <[email protected]>
  120. AuthorDate: Tue Jul 16 22:24:46 2019 +0300
  121. Commit: Konstantin Pavlov <[email protected]>
  122. CommitDate: Wed Jul 17 00:06:24 2019 +0300
  123.  
  124. CI: Use a newer aarch64 image
  125.  
  126. It now includes pkg-config, so lavf can be detected.
  127.  
  128. commit 98ee9d2f215326feeb221a4434957fa586d55c18 [revision 2971]
  129. Author: Konstantin Pavlov <[email protected]>
  130. AuthorDate: Fri Apr 5 15:08:29 2019 +0300
  131. Commit: Konstantin Pavlov <[email protected]>
  132. CommitDate: Tue Jul 16 14:34:18 2019 +0300
  133.  
  134. Added gitlab CI
  135.  
  136. Supported targets:
  137. - debian amd64
  138. - debian aarch64
  139. - windows 32 bit
  140. - windows 64 bit
  141. - macos 64bit
  142.  
  143. The tests are ran on all supported targets (via wine on windows).
  144.  
  145. The release jobs are only available on master/stable branches in
  146. videolan/x264 repository, and must be ran manually when a developer
  147. wishes to upload the artifacts.
  148.  
  149. commit 5493be84cdccecee613236c31b1e3227681ce428 [revision 2970]
  150. Author: Henrik Gramner <[email protected]>
  151. AuthorDate: Thu Mar 14 14:31:22 2019 +0100
  152. Commit: Henrik Gramner <[email protected]>
  153. CommitDate: Thu Mar 14 14:31:22 2019 +0100
  154.  
  155. Fix warning in autocomplete.c when compiled with lavf
  156.  
  157. commit d4099dd4c722f52c4f3c14575d7d39eb8fadb97f [revision 2969]
  158. Author: Anton Mitrofanov <[email protected]>
  159. AuthorDate: Tue Jun 6 02:30:41 2017 +0300
  160. Commit: Anton Mitrofanov <[email protected]>
  161. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  162.  
  163. Remove compatibility workarounds
  164.  
  165. This will break decoding with older versions of FFmpeg/Libav.
  166.  
  167. commit 120ed3afe4bdef3f7f0ac2768e57da0d935e7536 [revision 2968]
  168. Author: Anton Mitrofanov <[email protected]>
  169. AuthorDate: Fri Nov 9 18:37:17 2018 +0300
  170. Commit: Anton Mitrofanov <[email protected]>
  171. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  172.  
  173. Remove h->rc dereferencing where possible
  174.  
  175. commit 3e5aed95cc470f37e2db3e6506a8deb89b527720 [revision 2967]
  176. Author: Henrik Gramner <[email protected]>
  177. AuthorDate: Sat Feb 16 21:02:01 2019 +0100
  178. Commit: Anton Mitrofanov <[email protected]>
  179. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  180.  
  181. x86inc: Add support for GFNI instructions
  182.  
  183. commit d3fa8b972557bad64c2e0247b0b5276c2d49961b [revision 2966]
  184. Author: Henrik Gramner <[email protected]>
  185. AuthorDate: Sat Feb 16 17:57:21 2019 +0100
  186. Commit: Anton Mitrofanov <[email protected]>
  187. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  188.  
  189. x86inc: Improve warnings for use of unsupported instructions
  190.  
  191. Warn when the following are used without the appropriate cpuflag:
  192. * YMM and ZMM registers
  193. * 'pextrw' with a memory operand
  194. * GPR instruction set extensions
  195.  
  196. commit 101bd27d89cc84c18845046c13a67ab39e443a25 [revision 2965]
  197. Author: Henrik Gramner <[email protected]>
  198. AuthorDate: Thu Jan 31 20:42:32 2019 +0100
  199. Commit: Anton Mitrofanov <[email protected]>
  200. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  201.  
  202. x86inc: Support N_PEXT bit on Mach-O
  203.  
  204. Allows for marking symbols as having limited global scope, similar to
  205. using 'hidden' symbol visibility on ELF.
  206.  
  207. commit 6f85b3c4961810427cc4e8f520e0b706a321114d [revision 2964]
  208. Author: Henrik Gramner <[email protected]>
  209. AuthorDate: Thu Jan 31 20:21:43 2019 +0100
  210. Commit: Anton Mitrofanov <[email protected]>
  211. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  212.  
  213. x86inc: Make 'non-adjacent' default in the TAIL_CALL macro
  214.  
  215. commit 82721eae6edddf4955634adc51bf6eb228cc1313 [revision 2963]
  216. Author: Henrik Gramner <[email protected]>
  217. AuthorDate: Thu Jan 31 20:17:56 2019 +0100
  218. Commit: Anton Mitrofanov <[email protected]>
  219. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  220.  
  221. x86inc: Add x86-32 PIC support macros
  222.  
  223. commit b7e9935c3f08055a67a0fdea498499c675d00054 [revision 2962]
  224. Author: Henrik Gramner <[email protected]>
  225. AuthorDate: Thu Jan 31 20:11:01 2019 +0100
  226. Commit: Anton Mitrofanov <[email protected]>
  227. CommitDate: Wed Mar 6 22:45:53 2019 +0300
  228.  
  229. x86inc: Turn 'movsxd' into 'movifnidn' on x86-32
  230.  
  231. commit ec1d32302d0f1f59d3882e0289126b8d897c9f57 [revision 2961]
  232. Author: Henrik Gramner <[email protected]>
  233. AuthorDate: Thu Jan 31 20:08:40 2019 +0100
  234. Commit: Anton Mitrofanov <[email protected]>
  235. CommitDate: Wed Mar 6 22:45:52 2019 +0300
  236.  
  237. Bump dates to 2019
  238.  
  239. commit 74c051f2c4945cf2a279e36051537a2a1897c120 [revision 2960]
  240. Author: Henrik Gramner <[email protected]>
  241. AuthorDate: Sun Jul 1 20:34:48 2018 +0200
  242. Commit: Anton Mitrofanov <[email protected]>
  243. CommitDate: Wed Mar 6 22:45:52 2019 +0300
  244.  
  245. cli: Bash autocomplete support
  246.  
  247. Allows for automatic command line completion for both options and values.
  248.  
  249. Options such as --input-csp and --input-fmt will dynamically retrieve
  250. supported values from libavformat when compiled with lavf support.
  251.  
  252. Execute 'source tools/bash-autocomplete.sh' in bash to enable.
  253.  
  254. commit 92d36908cbafd2a6edf7e61d69f341027b57f6f8 [revision 2959]
  255. Author: Yusuke Nakamura <[email protected]>
  256. AuthorDate: Mon Apr 9 11:01:28 2018 +0900
  257. Commit: Anton Mitrofanov <[email protected]>
  258. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  259.  
  260. Signal Progressive and Constrained profiles
  261.  
  262. Progressive High, Constrained High, and Progressive High 10.
  263.  
  264. Even in Main profile, constraint_set4_flag is now set to 1 if progressive,
  265. and constraint_set5_flag is set to 1 if no B-slices are present.
  266.  
  267. commit 57baac4ed7fe213a2c2bb07924c6c7cee8ac25f9 [revision 2958]
  268. Author: Alexandra Hájková <[email protected]>
  269. AuthorDate: Sat Sep 8 07:15:53 2018 +0000
  270. Commit: Anton Mitrofanov <[email protected]>
  271. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  272.  
  273. ppc: Use xxpermdi in sad_x3/x4 and use macros to avoid redundant code
  274.  
  275. commit de380f4aed75b0a9bf5bdfc298a9901646184375 [revision 2957]
  276. Author: Luca Barbato <[email protected]>
  277. AuthorDate: Thu Sep 6 12:25:14 2018 +0200
  278. Commit: Anton Mitrofanov <[email protected]>
  279. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  280.  
  281. ppc: Use the vec_xst_len for partial stores in mc
  282.  
  283. Around a ~1% speedup to the overall encoding for --slow.
  284.  
  285. commit 69dfb2896cf3180fd59233b124b5589f12fb6a94 [revision 2956]
  286. Author: Luca Barbato <[email protected]>
  287. AuthorDate: Thu Sep 6 12:25:13 2018 +0200
  288. Commit: Anton Mitrofanov <[email protected]>
  289. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  290.  
  291. ppc: Use vec_splats in mc
  292.  
  293. No overall speedup, just tidier code.
  294.  
  295. commit 40688108dd13fc0bf1847a6dfc1cf86a728654fb [revision 2955]
  296. Author: Luca Barbato <[email protected]>
  297. AuthorDate: Thu Aug 23 08:30:37 2018 +0000
  298. Commit: Anton Mitrofanov <[email protected]>
  299. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  300.  
  301. ppc: Use the vec_xst_len for partial stores
  302.  
  303. Seems to give about a 1-2% overall speedup on --slow.
  304.  
  305. commit 0d111333bbd65b1a76b5c646abf802f45dd41e96 [revision 2954]
  306. Author: Luca Barbato <[email protected]>
  307. AuthorDate: Sun Aug 19 17:27:55 2018 +0200
  308. Commit: Anton Mitrofanov <[email protected]>
  309. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  310.  
  311. ppc: Use xxpermdi in VEC_STORE8
  312.  
  313. Around a ~2% speedup to the overall encoding for --slow.
  314.  
  315. commit 18262ee37fedeb4d7b30d9a228f2f38ef0e13cc1 [revision 2953]
  316. Author: Luca Barbato <[email protected]>
  317. AuthorDate: Sun Aug 19 17:27:54 2018 +0200
  318. Commit: Anton Mitrofanov <[email protected]>
  319. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  320.  
  321. ppc: Use a single store to write the scores for sad_x4_8x8
  322.  
  323. Yet another use of xxpermdi, another 10% gain.
  324.  
  325. commit 28fb2661161c12ee20c29d9bb2a75509a5af5327 [revision 2952]
  326. Author: Luca Barbato <[email protected]>
  327. AuthorDate: Sun Aug 19 17:27:53 2018 +0200
  328. Commit: Anton Mitrofanov <[email protected]>
  329. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  330.  
  331. ppc: Use xxpermdi to halve the computation in sad_x4_8x8
  332.  
  333. About 20% faster.
  334.  
  335. commit 83acefef8990302caf962c77e5a8189bb620ca6f [revision 2951]
  336. Author: Luca Barbato <[email protected]>
  337. AuthorDate: Sun Aug 19 09:28:42 2018 +0200
  338. Commit: Anton Mitrofanov <[email protected]>
  339. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  340.  
  341. ppc: Rework satd_4* likewise
  342.  
  343. Now 4x4 is as slow as C and 4x8 is a 2% faster than before.
  344.  
  345. commit e0d846a63313e2a3d71faa703238b70385f6a5e4 [revision 2950]
  346. Author: Luca Barbato <[email protected]>
  347. AuthorDate: Sun Aug 19 09:28:41 2018 +0200
  348. Commit: Anton Mitrofanov <[email protected]>
  349. CommitDate: Wed Mar 6 22:45:51 2019 +0300
  350.  
  351. ppc: Factor out the sum of absolute
  352.  
  353. And use it on the other satd > 8.
  354.  
  355. 5-10% faster depending on the size.
  356.  
  357. commit 6e74eb5af2f28ab30d2c28a86f921b56e94f04f7 [revision 2949]
  358. Author: Luca Barbato <[email protected]>
  359. AuthorDate: Sun Aug 19 09:28:40 2018 +0200
  360. Commit: Anton Mitrofanov <[email protected]>
  361. CommitDate: Wed Mar 6 22:45:50 2019 +0300
  362.  
  363. ppc: Rework the adds in satd8x8
  364.  
  365. 10% faster.
  366.  
  367. commit 4dd83955b282e722fbeb3f4ee5cc05a45dc54c7f [revision 2948]
  368. Author: Luca Barbato <[email protected]>
  369. AuthorDate: Fri Aug 17 22:28:45 2018 +0200
  370. Commit: Anton Mitrofanov <[email protected]>
  371. CommitDate: Wed Mar 6 22:45:50 2019 +0300
  372.  
  373. ppc: Add quant_4x4x4
  374.  
  375. 4x faster than C.
  376.  
  377. commit 8f6ac77f325c70631359e5f173e76b41e3fb55d9 [revision 2947]
  378. Author: Luca Barbato <[email protected]>
  379. AuthorDate: Fri Aug 17 22:28:44 2018 +0200
  380. Commit: Anton Mitrofanov <[email protected]>
  381. CommitDate: Wed Mar 6 22:45:50 2019 +0300
  382.  
  383. ppc: Cleanup quant
  384.  
  385. commit 275ef5332dffec445a0c5a78dbc00c3e0766011d [revision 2946]
  386. Author: Henrik Gramner <[email protected]>
  387. AuthorDate: Sun Aug 12 17:00:13 2018 +0200
  388. Commit: Anton Mitrofanov <[email protected]>
  389. CommitDate: Wed Mar 6 22:45:50 2019 +0300
  390.  
  391. x86: Always use PIC in x86-64 asm
  392.  
  393. Most x86-64 operating systems nowadays doesn't even allow .text relocations
  394. in object files any more, and there is no measurable overall performance
  395. difference from using RIP-relative addressing in x264 asm.
  396.  
  397. Enforcing PIC reduces complexity and simplifies testing.
  398.  
  399. commit 72db437770fd1ce3961f624dd57a8e75ff65ae0b [revision 2945]
  400. Author: Henrik Gramner <[email protected]>
  401. AuthorDate: Sat Feb 23 20:15:33 2019 +0100
  402. Commit: Henrik Gramner <[email protected]>
  403. CommitDate: Sun Mar 3 23:52:42 2019 +0100
  404.  
  405. x86: Fix integer overflow in intra_sa8d_x3_8x8_sse2
  406.  
  407. commit 88943afa4ee6565370e0e7cdc475b3b2283ada4b [revision 2944]
  408. Author: Anton Mitrofanov <[email protected]>
  409. AuthorDate: Fri Nov 9 18:13:34 2018 +0300
  410. Commit: Henrik Gramner <[email protected]>
  411. CommitDate: Sun Mar 3 23:52:42 2019 +0100
  412.  
  413. Check that mbtree settings are consistent between passes
  414.  
  415. Also check that CQP mode is not used with 2-pass.
  416.  
  417. commit 6d8af5f0e390bbcd31a65dda04ef27d3f93821c1 [revision 2943]
  418. Author: Anton Mitrofanov <[email protected]>
  419. AuthorDate: Mon Feb 4 22:04:56 2019 +0300
  420. Commit: Henrik Gramner <[email protected]>
  421. CommitDate: Sun Mar 3 23:52:42 2019 +0100
  422.  
  423. Mark frame_size_estimated as volatile
  424.  
  425. Ensures that access is atomic and that other threads sees the actual
  426. value of the variable.
  427.  
  428. commit a6327f8a25b72f5edd3515aca82190046d18745b [revision 2942]
  429. Author: Anton Mitrofanov <[email protected]>
  430. AuthorDate: Mon Feb 4 21:46:12 2019 +0300
  431. Commit: Henrik Gramner <[email protected]>
  432. CommitDate: Sun Mar 3 23:52:42 2019 +0100
  433.  
  434. Fix data race detected by ThreadSanitizer
  435.  
  436. Bug report by Daniel Deptford.
  437.  
  438. commit 6172da4d77a574c831ed4710a10d945ea128528e [revision 2941]
  439. Author: Anton Mitrofanov <[email protected]>
  440. AuthorDate: Mon Dec 24 19:37:45 2018 +0300
  441. Commit: Henrik Gramner <[email protected]>
  442. CommitDate: Sun Mar 3 23:52:42 2019 +0100
  443.  
  444. Fix XAVC with sliced-threads
  445.  
  446. commit c7ec24cfbdf720dbf0806046cb5fb9302b941ec9 [revision 2940]
  447. Author: Anton Mitrofanov <[email protected]>
  448. AuthorDate: Fri Dec 21 18:54:56 2018 +0300
  449. Commit: Henrik Gramner <[email protected]>
  450. CommitDate: Sun Mar 3 23:52:42 2019 +0100
  451.  
  452. Fix XAVC slice pattern
  453.  
  454. commit 6aa4b5929d3ce92ab618e98c34ed6e0948b06bbf [revision 2939]
  455. Author: Henrik Gramner <[email protected]>
  456. AuthorDate: Sun Oct 21 14:28:59 2018 +0200
  457. Commit: Henrik Gramner <[email protected]>
  458. CommitDate: Sun Mar 3 23:52:40 2019 +0100
  459.  
  460. Eliminate the use of strtok()
  461.  
  462. Also fix the string parsing in param_apply_tune() to correctly compare
  463. the entire string, not just the first N characters.
  464.  
  465. commit d6af823959dc06f061e0a7b038dab83d9c1c9ea3 [revision 2938]
  466. Author: Anton Mitrofanov <[email protected]>
  467. AuthorDate: Thu Nov 8 22:01:54 2018 +0300
  468. Commit: Henrik Gramner <[email protected]>
  469. CommitDate: Sun Dec 23 20:13:43 2018 +0100
  470.  
  471. configure: Fix log2f misdetection on some systems
  472.  
  473. Bug report by Dirk Fieldhouse.
  474.  
  475. commit b763e338e0cec4dae13c4fc2fc49c63ac6f26df1 [revision 2937]
  476. Author: Anton Mitrofanov <[email protected]>
  477. AuthorDate: Thu Nov 8 21:53:17 2018 +0300
  478. Commit: Henrik Gramner <[email protected]>
  479. CommitDate: Sun Dec 23 20:13:43 2018 +0100
  480.  
  481. Fix ultrafast preset speed regression
  482.  
  483. --trellis 0 was missed for it during 8-bit and 10-bit unification.
  484. Bug report by Aleksey Vasenev.
  485.  
  486. commit b048e2658ad6aec55deceb0561db5796cdb64bd2 [revision 2936]
  487. Author: Anton Mitrofanov <[email protected]>
  488. AuthorDate: Wed Oct 10 19:41:08 2018 +0300
  489. Commit: Henrik Gramner <[email protected]>
  490. CommitDate: Sun Dec 23 20:13:43 2018 +0100
  491.  
  492. Fix --crop-rect top offset with --interlaced or --fake-interlaced
  493.  
  494. Bug report by Koby Shina.
  495.  
  496. commit 545de2ffec6ae9a80738de1b2c8cf820249a2530 [revision 2935]
  497. Author: Anton Mitrofanov <[email protected]>
  498. AuthorDate: Sun Sep 23 20:47:44 2018 +0300
  499. Commit: Henrik Gramner <[email protected]>
  500. CommitDate: Tue Sep 25 11:20:26 2018 +0200
  501.  
  502. Fix possible double transpose of custom CQM if --level is not set
  503.  
  504. Bug reported by Nicolas Gaullier
  505.  
  506. commit b63c73dc5c37e5405bf032c9113c1daced3e45a4 [revision 2934]
  507. Author: Henrik Gramner <[email protected]>
  508. AuthorDate: Tue Aug 7 22:42:22 2018 +0200
  509. Commit: Henrik Gramner <[email protected]>
  510. CommitDate: Thu Aug 23 01:34:51 2018 +0200
  511.  
  512. cli: Fix linking with --system-libx264 on x86
  513.  
  514. commit fb17a6b5b51d02020fb0cadea2b27c7803e734ba [revision 2933]
  515. Author: Anton Mitrofanov <[email protected]>
  516. AuthorDate: Tue Aug 21 15:11:21 2018 +0300
  517. Commit: Henrik Gramner <[email protected]>
  518. CommitDate: Thu Aug 23 01:34:51 2018 +0200
  519.  
  520. Fix CAVLC+RDO in 4:4:4
  521.  
  522. commit 303c484ec828ed0d8bfe743500e70314d026c3bd [revision 2932]
  523. Author: Alexandra Hájková <[email protected]>
  524. AuthorDate: Wed Jul 11 19:28:20 2018 +0000
  525. Commit: Henrik Gramner <[email protected]>
  526. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  527.  
  528. ppc: Optimize quant functions
  529.  
  530. 1) using xxpermdi + merge instead of 2 merges improves quant_8x8
  531. performance by 5%
  532.  
  533. 2) use vec_splats instead of vec_splat
  534.  
  535. checkasm timings when compiled with gcc:
  536. C: AltiVec:
  537. before: after:
  538. quant_2x2_dc: 57 163 46
  539. quant_4x4_dc: 141 162 57
  540.  
  541. dequant_4x4_cmp: 104 101 45
  542. dequant_4x4_flat: 104 106 46
  543. dequant_8x8_cmp: 412 208 147
  544. dequant_8x8_flat: 414 212 149
  545.  
  546. commit 44f1671369b54734db1775fe5155f17041344d8f [revision 2931]
  547. Author: Alexandra Hajkova <[email protected]>
  548. AuthorDate: Sun Jul 8 13:04:43 2018 -0500
  549. Commit: Henrik Gramner <[email protected]>
  550. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  551.  
  552. ppc: Add support for Power9-only vec_absd
  553.  
  554. Increases overall encoding speed on POWER9 by 8%.
  555.  
  556. commit f8afe3820c84798e9e50623cf7349bdb98765926 [revision 2930]
  557. Author: Alexandra Hájková <[email protected]>
  558. AuthorDate: Fri Jun 29 16:50:20 2018 +0000
  559. Commit: Henrik Gramner <[email protected]>
  560. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  561.  
  562. ppc: Optimize sub8x8_dct_dc
  563.  
  564. commit 411c957d82d357250f3a3099727b1a2c84caaee9 [revision 2929]
  565. Author: Alexandra Hájková <[email protected]>
  566. AuthorDate: Thu Jun 21 18:36:32 2018 +0000
  567. Commit: Henrik Gramner <[email protected]>
  568. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  569.  
  570. ppc: AltiVec add16x16_idct_dc
  571.  
  572. commit 53fe16e51349c43c483e81afb1f08a39f843a234 [revision 2928]
  573. Author: Alexandra Hájková <[email protected]>
  574. AuthorDate: Sat Jun 23 14:58:17 2018 +0000
  575. Commit: Henrik Gramner <[email protected]>
  576. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  577.  
  578. ppc: Optimize add8x8_idct_dc
  579.  
  580. commit 62dcebbce2c3f34998aeb2ea76b89f51306e78e9 [revision 2927]
  581. Author: Luca Barbato <[email protected]>
  582. AuthorDate: Thu Jul 12 10:41:22 2018 +0200
  583. Commit: Henrik Gramner <[email protected]>
  584. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  585.  
  586. ppc: Add compatibility macros for vec_xxpermdi
  587.  
  588. commit d1a53926fb90e9f4a4f1605f4b2a8a945a73e1d2 [revision 2926]
  589. Author: Henrik Gramner <[email protected]>
  590. AuthorDate: Mon Jun 25 00:09:51 2018 +0200
  591. Commit: Henrik Gramner <[email protected]>
  592. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  593.  
  594. Prefer a monotonic clock source if available
  595.  
  596. commit 1d18f0e025e994b93233b8e8afa0c691bccc8fda [revision 2925]
  597. Author: Kieran Kunhya <[email protected]>
  598. AuthorDate: Wed Aug 30 16:05:41 2017 +0100
  599. Commit: Henrik Gramner <[email protected]>
  600. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  601.  
  602. Add Sony XAVC, a flavour of AVC-Intra
  603.  
  604. commit bc136ec6a0f863c42686a3bc9fa4c7820f83d413 [revision 2924]
  605. Author: Anton Mitrofanov <[email protected]>
  606. AuthorDate: Mon Jul 2 20:20:03 2018 +0300
  607. Commit: Henrik Gramner <[email protected]>
  608. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  609.  
  610. Cosmetics: Fix indentation for multiline function prototypes
  611.  
  612. It was broken in "Drop the x264 prefix" patch.
  613.  
  614. commit 6dd1d3b5d9e16a5951ececb7351cd63f02b36435 [revision 2923]
  615. Author: Anton Mitrofanov <[email protected]>
  616. AuthorDate: Mon Apr 16 23:54:43 2018 +0300
  617. Commit: Henrik Gramner <[email protected]>
  618. CommitDate: Mon Aug 6 23:24:13 2018 +0200
  619.  
  620. Cosmetics: Use consistent "inline" attribute position
  621.  
  622. Place it immediately after "static".
  623.  
  624. commit 3d9ec58f27f1cd6732484246aaad59158b98af47 [revision 2922]
  625. Author: Henrik Gramner <[email protected]>
  626. AuthorDate: Thu Jan 25 22:17:57 2018 +0100
  627. Commit: Henrik Gramner <[email protected]>
  628. CommitDate: Mon Aug 6 23:24:12 2018 +0200
  629.  
  630. x86: AVX-512 plane_copy and plane_copy_swap
  631.  
  632. Avoid the scalar C wrapper by utilizing opmasks to prevent overreading the
  633. input buffer.
  634.  
  635. commit 698c5a32e63a3ed6b976ed196abe479efd78530b [revision 2921]
  636. Author: Emanuele Ruffaldi <[email protected]>
  637. AuthorDate: Sat Jan 6 02:34:39 2018 +0100
  638. Commit: Henrik Gramner <[email protected]>
  639. CommitDate: Mon Aug 6 23:24:12 2018 +0200
  640.  
  641. 4:0:0 (monochrome) encoding support
  642.  
  643. Virtually zero increase in compression efficiency compared to 4:2:0 with empty
  644. chroma planes. Performance is better though, especially with fast settings.
  645.  
  646. commit 814e61e88c809bb00d17c200a04e9c7d42a19bb5 [revision 2920]
  647. Author: Diego Biurrun <[email protected]>
  648. AuthorDate: Sun Feb 5 09:02:43 2017 +0100
  649. Commit: Henrik Gramner <[email protected]>
  650. CommitDate: Mon Aug 6 23:24:12 2018 +0200
  651.  
  652. Makefile improvements
  653.  
  654. * Coalesce some install recipe lines
  655.  
  656. * Remove empty addition of GPLed filters
  657.  
  658. * Install libdir in recipes that directly require it
  659.  
  660. * Coalesce etags/TAGS rules
  661.  
  662. * Simplify fprofiled rule
  663.  
  664. commit 28e4879842a86cc6bb63db0f5f386a3e9268fd46 [revision 2919]
  665. Author: Henrik Gramner <[email protected]>
  666. AuthorDate: Sun Apr 22 22:49:15 2018 +0200
  667. Commit: Henrik Gramner <[email protected]>
  668. CommitDate: Mon Aug 6 23:24:12 2018 +0200
  669.  
  670. x86inc: Improve SAVE/LOAD_MM_PERMUTATION macros
  671.  
  672. Use register numbers instead of copying the full register names. This makes it
  673. possible to change register widths in the middle of a function and keep the
  674. mmreg permutations intact which can be useful for code that only needs larger
  675. vectors for parts of the function in combination with macros etc.
  676.  
  677. Also change the LOAD_MM_PERMUTATION macro to use the same default name as the
  678. SAVE macro. This simplifies swapping from ymm to xmm registers or vice versa:
  679.  
  680. SAVE_MM_PERMUTATION
  681. INIT_XMM <cpuflags>
  682. LOAD_MM_PERMUTATION
  683.  
  684. commit 8badb910847e94abb66686009e424bdce355c9f4 [revision 2918]
  685. Author: Henrik Gramner <[email protected]>
  686. AuthorDate: Sat Mar 31 13:49:56 2018 +0200
  687. Commit: Henrik Gramner <[email protected]>
  688. CommitDate: Mon Aug 6 23:24:12 2018 +0200
  689.  
  690. x86inc: Optimize VEX instruction encoding
  691.  
  692. Most VEX-encoded instructions require an additional byte to encode when src2
  693. is a high register (e.g. x|ymm8..15). If the instruction is commutative we
  694. can swap src1 and src2 when doing so reduces the instruction length, e.g.
  695.  
  696. vpaddw xmm0, xmm0, xmm8 -> vpaddw xmm0, xmm8, xmm0
  697.  
  698. commit 0a84d986e7020f8344f00752e3600b9769cc1e85 [revision 2917]
  699. Author: Henrik Gramner <[email protected]>
  700. AuthorDate: Sat Mar 31 01:16:06 2018 +0200
  701. Commit: Henrik Gramner <[email protected]>
  702. CommitDate: Mon Aug 6 23:24:12 2018 +0200
  703.  
  704. x86inc: Fix VEX -> EVEX instruction conversion
  705.  
  706. There's an edge case that wasn't properly handled.
  707.  
  708. commit 9d33c8fefbb506377b943aba11cd99c74258c5de [revision 2916]
  709. Author: Anton Mitrofanov <[email protected]>
  710. AuthorDate: Tue Jul 31 22:54:33 2018 +0300
  711. Commit: Henrik Gramner <[email protected]>
  712. CommitDate: Mon Aug 6 23:24:12 2018 +0200
  713.  
  714. configure: Fix required version checks for lavf and swscale
  715.  
  716. commit 34843deb060248514ecd9edd88d72c2c2d6b906a [revision 2915]
  717. Author: Anton Mitrofanov <[email protected]>
  718. AuthorDate: Fri Jul 20 08:37:43 2018 +0300
  719. Commit: Henrik Gramner <[email protected]>
  720. CommitDate: Sat Jul 21 19:50:35 2018 +0200
  721.  
  722. Fix float division by zero in weightp analysis
  723.  
  724. commit 1c3174775c6c1789aaf10172e4cb619f91ecff4a [revision 2914]
  725. Author: Anton Mitrofanov <[email protected]>
  726. AuthorDate: Wed Jul 18 21:56:33 2018 +0300
  727. Commit: Henrik Gramner <[email protected]>
  728. CommitDate: Sat Jul 21 19:50:34 2018 +0200
  729.  
  730. Fix undefined behavior of left shift for CAVLC encoding
  731.  
  732. commit a0253ebee0f4d854cf89934b5f420275862d0b5b [revision 2913]
  733. Author: Anton Mitrofanov <[email protected]>
  734. AuthorDate: Mon Jul 2 20:59:16 2018 +0300
  735. Commit: Henrik Gramner <[email protected]>
  736. CommitDate: Sat Jul 21 19:50:19 2018 +0200
  737.  
  738. Fix integer overflow in slicetype_path_cost
  739.  
  740. The path cost for high resolutions can exceed COST_MAX.
  741.  
  742. commit 2af2742821f0b08a4295055b41875e660d5a7746 [revision 2912]
  743. Author: Henrik Gramner <[email protected]>
  744. AuthorDate: Fri Jun 29 13:14:01 2018 +0200
  745. Commit: Henrik Gramner <[email protected]>
  746. CommitDate: Fri Jun 29 13:19:57 2018 +0200
  747.  
  748. cli: Fix preset help listing
  749.  
  750. It was previously incorrect when --chroma-format or --bit-depth was
  751. specified in configure.
  752.  
  753. commit f5d929ab8faf2319dda10836f51803ba25f0ad07 [revision 2911]
  754. Author: Luca Barbato <[email protected]>
  755. AuthorDate: Sat Jun 23 13:14:28 2018 +0200
  756. Commit: Henrik Gramner <[email protected]>
  757. CommitDate: Mon Jun 25 00:14:30 2018 +0200
  758.  
  759. ppc: Fix zigzag_interleave
  760.  
  761. The permv array has 3 elements
  762.  
  763. commit 7737e6ad4acf1058aeb0f9802e2a3ca1e0a30d29 [revision 2910]
  764. Author: Henrik Gramner <[email protected]>
  765. AuthorDate: Sat Jun 2 20:35:10 2018 +0200
  766. Commit: Henrik Gramner <[email protected]>
  767. CommitDate: Sun Jun 3 01:41:38 2018 +0200
  768.  
  769. Fix clang stack alignment issues
  770.  
  771. Clang emits aligned AVX stores for things like zeroing stack-allocated
  772. variables when using -mavx even with -fno-tree-vectorize set which can
  773. result in crashes if this occurs before we've realigned the stack.
  774.  
  775. Previously we only ensured that the stack was realigned before calling
  776. assembly functions that accesses stack-allocated buffers but this is
  777. not sufficient. Fix the issue by changing the stack realignment to
  778. instead occur immediately in all CLI, API and thread entry points.
  779.  
  780. commit 26b99cce1f03f023dee98bef2ec3cd2eff319f8e [revision 2909]
  781. Author: Anton Mitrofanov <[email protected]>
  782. AuthorDate: Sun Apr 1 20:49:29 2018 +0300
  783. Commit: Henrik Gramner <[email protected]>
  784. CommitDate: Sun May 27 20:58:54 2018 +0200
  785.  
  786. Fix missing bs_flush in AUD writing
  787.  
  788. commit da6b29b553bb56e16e99527733849735c2ea264c [revision 2908]
  789. Author: Anton Mitrofanov <[email protected]>
  790. AuthorDate: Sun Apr 1 20:39:30 2018 +0300
  791. Commit: Henrik Gramner <[email protected]>
  792. CommitDate: Sun May 27 20:58:54 2018 +0200
  793.  
  794. Fix possible undefined behavior of right shift
  795.  
  796. 32-bit shifts are only defined for values in the range 0-31.
  797.  
  798. commit 7e457290cdd6da592ae63aa25facc47cd09d2128 [revision 2907]
  799. Author: Anton Mitrofanov <[email protected]>
  800. AuthorDate: Sun Apr 1 20:34:18 2018 +0300
  801. Commit: Henrik Gramner <[email protected]>
  802. CommitDate: Sun May 27 20:58:54 2018 +0200
  803.  
  804. Make bs_align_10 imply bs_flush
  805.  
  806. Now behaves the same as bs_align_0 and bs_align_1.
  807.  
  808. commit 6afb67c6d7b71fcc6fc14d167f1fcf55623846f4 [revision 2906]
  809. Author: Anton Mitrofanov <[email protected]>
  810. AuthorDate: Sun Apr 1 17:52:47 2018 +0300
  811. Commit: Henrik Gramner <[email protected]>
  812. CommitDate: Sun May 27 20:58:54 2018 +0200
  813.  
  814. Fix theoretically incorrect cost_mv_fpel free
  815.  
  816. commit 57dd6274e2da70bdb8220bc159976e3ac2aea017 [revision 2905]
  817. Author: Anton Mitrofanov <[email protected]>
  818. AuthorDate: Sun Apr 1 17:42:46 2018 +0300
  819. Commit: Henrik Gramner <[email protected]>
  820. CommitDate: Sun May 27 20:58:53 2018 +0200
  821.  
  822. configure: Fix ambiguous "$(("
  823.  
  824. commit 0e6425e03e28213e73ae770df5e08fffba72d290 [revision 2904]
  825. Author: Anton Mitrofanov <[email protected]>
  826. AuthorDate: Mon Feb 19 19:53:38 2018 +0300
  827. Commit: Henrik Gramner <[email protected]>
  828. CommitDate: Sun May 27 20:58:53 2018 +0200
  829.  
  830. Fix --qpmax default value in fullhelp
  831.  
  832. commit 5f7f950c80e330728ecb07bc133e17456870121a [revision 2903]
  833. Author: Henrik Gramner <[email protected]>
  834. AuthorDate: Sat Mar 31 01:31:57 2018 +0200
  835. Commit: Henrik Gramner <[email protected]>
  836. CommitDate: Sat Mar 31 21:53:42 2018 +0200
  837.  
  838. x86: Correctly use v-prefix for instructions with opmasks
  839.  
  840. This was always required, but accidentally happened to work correctly
  841. in a few cases.
  842.  
  843. commit 3d90057e15abf257320c89bb7146fb0c92687fa6 [revision 2902]
  844. Author: Martin Storsjö <[email protected]>
  845. AuthorDate: Sat Mar 31 00:10:14 2018 +0300
  846. Commit: Henrik Gramner <[email protected]>
  847. CommitDate: Sat Mar 31 21:53:42 2018 +0200
  848.  
  849. configure: Only use gas-preprocessor with armasm for compiler=CL
  850.  
  851. This picks the right assembler automatically for arm and aarch64
  852. llvm-mingw targets.
  853.  
  854. This doesn't get the right assembler for clang setups when clang
  855. acts like MSVC and uses MSVC headers though (where it perhaps
  856. should use armasm as before), but that's probably an even more
  857. obscure setup.
  858.  
  859. commit 7d0ff22e8c96de126be9d3de4952edd6d1b75a8c [revision 2901]
  860. Author: Anton Mitrofanov <[email protected]>
  861. AuthorDate: Wed Jan 17 22:03:06 2018 +0300
  862. Commit: Henrik Gramner <[email protected]>
  863. CommitDate: Thu Jan 18 21:03:22 2018 +0100
  864.  
  865. Remove ARRAY_SIZE macro which is identical to ARRAY_ELEMS
  866.  
  867. commit 4a158b00943c334ec9e0aabe6a919900c32e360e [revision 2900]
  868. Author: Henrik Gramner <[email protected]>
  869. AuthorDate: Sat Jan 6 17:47:42 2018 +0100
  870. Commit: Henrik Gramner <[email protected]>
  871. CommitDate: Wed Jan 17 18:31:04 2018 +0100
  872.  
  873. x86inc: Correctly set mmreg variables
  874.  
  875. commit 40b47eed1338cef1ac66c98b4e393dfcf5d998ae [revision 2899]
  876. Author: Diego Biurrun <[email protected]>
  877. AuthorDate: Sun Feb 5 09:02:49 2017 +0100
  878. Commit: Henrik Gramner <[email protected]>
  879. CommitDate: Wed Jan 17 18:31:04 2018 +0100
  880.  
  881. .gitignore: Ignore TAGS file
  882.  
  883. commit 6fce82284a0fb3edfa299b904b1559452a3b1094 [revision 2898]
  884. Author: Diego Biurrun <[email protected]>
  885. AuthorDate: Sun Feb 5 09:02:51 2017 +0100
  886. Commit: Henrik Gramner <[email protected]>
  887. CommitDate: Wed Jan 17 18:31:04 2018 +0100
  888.  
  889. Minor configure improvements
  890.  
  891. * Drop empty addition of GPLed filters
  892.  
  893. * Replace backticks with $()
  894.  
  895. commit ca5408b13cf0e58a7505051861f20a63a7a6aec1 [revision 2897]
  896. Author: Henrik Gramner <[email protected]>
  897. AuthorDate: Mon Jan 1 15:05:48 2018 +0100
  898. Commit: Henrik Gramner <[email protected]>
  899. CommitDate: Wed Jan 17 18:31:04 2018 +0100
  900.  
  901. Bump dates to 2018
  902.  
  903. commit b019515ef4ad77022b849283c62612157e8458a7 [revision 2896]
  904. Author: Henrik Gramner <[email protected]>
  905. AuthorDate: Tue Jan 16 17:43:24 2018 +0100
  906. Commit: Henrik Gramner <[email protected]>
  907. CommitDate: Wed Jan 17 18:31:03 2018 +0100
  908.  
  909. Merge zero buffers
  910.  
  911. Improves cache efficiency.
  912.  
  913. commit d75b93b0e82cefa93e5db2d6b0be475566101431 [revision 2895]
  914. Author: Anton Mitrofanov <[email protected]>
  915. AuthorDate: Wed Jan 17 18:19:44 2018 +0300
  916. Commit: Henrik Gramner <[email protected]>
  917. CommitDate: Wed Jan 17 18:31:01 2018 +0100
  918.  
  919. rdo: Use ALIGNED_ARRAY for stack arrays
  920.  
  921. commit 9384a7389b251b59a079ccc3d1af9edd42e3d5e6 [revision 2894]
  922. Author: Henrik Gramner <[email protected]>
  923. AuthorDate: Mon Jan 15 21:42:59 2018 +0100
  924. Commit: Henrik Gramner <[email protected]>
  925. CommitDate: Wed Jan 17 18:30:49 2018 +0100
  926.  
  927. Correctly align buffers for AVX and AVX-512
  928.  
  929. Fixes segfaults on Windows where the stack is only 16-byte aligned.
  930.  
  931. commit b00bcafe53a166b63a179a2f41470cd13b59f927 [revision 2893]
  932. Author: Anton Mitrofanov <[email protected]>
  933. AuthorDate: Sun Dec 24 22:59:09 2017 +0300
  934. Commit: Anton Mitrofanov <[email protected]>
  935. CommitDate: Sun Dec 24 23:47:27 2017 +0300
  936.  
  937. Cosmetics
  938.  
  939. commit 7c6b3ad50d9210d27be6953dfa6d24e5d183be18 [revision 2892]
  940. Author: Alexandra Hájková <[email protected]>
  941. AuthorDate: Sun May 21 17:40:45 2017 +0000
  942. Commit: Anton Mitrofanov <[email protected]>
  943. CommitDate: Sun Dec 24 23:47:27 2017 +0300
  944.  
  945. ppc: Add load_deinterleave_chroma_fenc_altivec
  946.  
  947. 5x speed up vs C code.
  948.  
  949. commit b461e015fd7efe3bb740ef0716bc41d76eff30c9 [revision 2891]
  950. Author: Martin Storsjö <[email protected]>
  951. AuthorDate: Thu Oct 26 13:09:46 2017 +0300
  952. Commit: Anton Mitrofanov <[email protected]>
  953. CommitDate: Sun Dec 24 23:47:27 2017 +0300
  954.  
  955. Update to the latest upstream version of gas-preprocessor
  956.  
  957. This version supports converting aarch64 assembly for MS armasm64.exe.
  958.  
  959. commit 61e8b5cc482b08d51e18b336081073736d963e7e [revision 2890]
  960. Author: Henrik Gramner <[email protected]>
  961. AuthorDate: Sun Oct 22 09:59:28 2017 +0200
  962. Commit: Anton Mitrofanov <[email protected]>
  963. CommitDate: Sun Dec 24 23:47:27 2017 +0300
  964.  
  965. input: Add a workaround for swscale overread bugs
  966.  
  967. swscale can read past the end of the input buffer, which may result in
  968. crashes if such a read crosses a page boundary into an invalid page.
  969.  
  970. Work around this by adding some padding space at the end of the buffer when
  971. using memory-mapped input frames. This may sometimes require copying the
  972. last frame into a new buffer on Windows since the Microsoft memory-mapping
  973. implementation has very limited capabilities compared to POSIX systems.
  974.  
  975. commit 1221f097473a049a52fbb47aff2733321bd4661a [revision 2889]
  976. Author: Henrik Gramner <[email protected]>
  977. AuthorDate: Sun Oct 22 10:50:46 2017 +0200
  978. Commit: Anton Mitrofanov <[email protected]>
  979. CommitDate: Sun Dec 24 23:47:27 2017 +0300
  980.  
  981. filters/resize: Upgrade to a newer libavutil API
  982.  
  983. Use the AVComponentDescriptor depth field instead of depth_minus1.
  984.  
  985. commit 99ca611d2e667553e026f487dc787b595dde84c5 [revision 2888]
  986. Author: Martin Storsjö <[email protected]>
  987. AuthorDate: Wed Oct 18 10:40:02 2017 +0300
  988. Commit: Anton Mitrofanov <[email protected]>
  989. CommitDate: Sun Dec 24 23:47:27 2017 +0300
  990.  
  991. aarch64: Use ldurb/sturb for loads/stores with negative offsets
  992.  
  993. The assembler (both gas and clang/llvm) automatically fixes this,
  994. armasm64 doesn't. We can fix it in gas-preprocessor, but we should
  995. also be using the right instruction form.
  996.  
  997. commit f745815e593b788d846182c8d42eed4f72f7c33c [revision 2887]
  998. Author: Martin Storsjö <[email protected]>
  999. AuthorDate: Mon Oct 16 22:50:27 2017 +0300
  1000. Commit: Anton Mitrofanov <[email protected]>
  1001. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1002.  
  1003. configure: Add support for building with MSVC/armasm for ARM64
  1004.  
  1005. commit 7b13b31be60ed65bee615bab28c422e2df027ee1 [revision 2886]
  1006. Author: Martin Storsjö <[email protected]>
  1007. AuthorDate: Mon Oct 16 22:50:26 2017 +0300
  1008. Commit: Anton Mitrofanov <[email protected]>
  1009. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1010.  
  1011. arm: Check for __ELF__ instead of !__APPLE__, for using .arch/.fpu
  1012.  
  1013. For windows, when building with armasm, we already filtered these out
  1014. with gas-preprocessor.
  1015.  
  1016. By filtering them out already in the source, we can also build directly
  1017. with clang for windows (which also require wrapping the assembler in
  1018. gas-preprocessor for converting instructions to thumb form, but
  1019. gas-preprocessor doesn't and shouldn't filter out them in the clang
  1020. configuration).
  1021.  
  1022. commit 12ca9a69e855c4d4b9000894f478bce665e4e02c [revision 2885]
  1023. Author: Martin Storsjö <[email protected]>
  1024. AuthorDate: Mon Oct 16 22:50:25 2017 +0300
  1025. Commit: Anton Mitrofanov <[email protected]>
  1026. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1027.  
  1028. aarch64: Don't .set a symbol named st2
  1029.  
  1030. This confuses gas-preprocessor, which tries to replace actual
  1031. st2 instructions by the integer 1 or 2.
  1032.  
  1033. commit 06c8f6bab0fc8fa9b2df9a1af5d10c87c515edb4 [revision 2884]
  1034. Author: Henrik Gramner <[email protected]>
  1035. AuthorDate: Sat Oct 14 14:11:26 2017 +0200
  1036. Commit: Anton Mitrofanov <[email protected]>
  1037. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1038.  
  1039. Shrink the i4x4_mode cost_table array
  1040.  
  1041. Only 17 elements are actually used. It was originally padded to 64 bytes to
  1042. avoid cache line splits in the x86 assembly, but those haven't really been
  1043. an issue on x86 CPU:s made in the past decade or so.
  1044.  
  1045. Benchmarking shows no performance impact from dropping the padding, so
  1046. might as well remove it and save some cache.
  1047.  
  1048. commit 344699fd386890ac1cf80a70a68a3ae16767ed62 [revision 2883]
  1049. Author: Henrik Gramner <[email protected]>
  1050. AuthorDate: Wed Oct 11 18:02:26 2017 +0200
  1051. Commit: Anton Mitrofanov <[email protected]>
  1052. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1053.  
  1054. x86: Remove some legacy CPU detection hacks
  1055.  
  1056. Some ancient Pentium-M and Core 1 CPU:s had slow SSE units, and using MMX
  1057. was preferable. Nowadays many assembly functions in x264 completely lack MMX
  1058. implementations and falling back to C code will likely make things worse.
  1059.  
  1060. Some misconfigured virtualized systems could sometimes also trigger this code
  1061. path and cause assertions.
  1062.  
  1063. commit 0fe75403d7b40c0209c3df992632956292065cdc [revision 2882]
  1064. Author: Henrik Gramner <[email protected]>
  1065. AuthorDate: Wed Oct 11 17:58:36 2017 +0200
  1066. Commit: Anton Mitrofanov <[email protected]>
  1067. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1068.  
  1069. lavf: Upgrade to the new core decoding API
  1070.  
  1071. commit dae7f18d2cc5c7eccfb73649cda458e3c8e2256e [revision 2881]
  1072. Author: Vittorio Giovara <[email protected]>
  1073. AuthorDate: Mon Oct 9 12:04:22 2017 -0400
  1074. Commit: Anton Mitrofanov <[email protected]>
  1075. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1076.  
  1077. lavf: Upgrade to some newer API:s
  1078.  
  1079. * Use the codec parameters API instead of the AVStream codec field.
  1080. * Use av_packet_unref() instead of av_free_packet().
  1081. * Use the AVFrame pts field instead of pkt_pts.
  1082.  
  1083. commit 12611ec99bb52f4f2c1b114138d867b3a2aa182b [revision 2880]
  1084. Author: Henrik Gramner <[email protected]>
  1085. AuthorDate: Sun Oct 8 21:41:16 2017 +0200
  1086. Commit: Anton Mitrofanov <[email protected]>
  1087. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1088.  
  1089. x86: AVX-512 load_deinterleave_chroma_fdec
  1090.  
  1091. commit d93851ec282eb069f91a6eddab3284f7766cd5bd [revision 2879]
  1092. Author: Henrik Gramner <[email protected]>
  1093. AuthorDate: Sun Oct 8 21:23:12 2017 +0200
  1094. Commit: Anton Mitrofanov <[email protected]>
  1095. CommitDate: Sun Dec 24 23:47:26 2017 +0300
  1096.  
  1097. x86: AVX-512 load_deinterleave_chroma_fenc
  1098.  
  1099. commit 5b62ab59be01579ab37033cc86527df922efb843 [revision 2878]
  1100. Author: Henrik Gramner <[email protected]>
  1101. AuthorDate: Sat Oct 7 12:06:51 2017 +0200
  1102. Commit: Anton Mitrofanov <[email protected]>
  1103. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1104.  
  1105. x86: AVX-512 mbtree_fix8_pack and mbtree_fix8_unpack
  1106.  
  1107. Takes advantage of opmasks to avoid having to use scalar code for the tail.
  1108.  
  1109. Also make some slight improvements to the checkasm test.
  1110.  
  1111. commit 08476ab1c0a9b741198677731373b173657fa079 [revision 2877]
  1112. Author: Henrik Gramner <[email protected]>
  1113. AuthorDate: Sat Oct 7 11:34:16 2017 +0200
  1114. Commit: Anton Mitrofanov <[email protected]>
  1115. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1116.  
  1117. x86: Faster mbtree_fix8_unpack
  1118.  
  1119. Use a different multiplier in order to eliminate some shifts.
  1120.  
  1121. About 25% faster than before.
  1122.  
  1123. commit e3fae10bf7db9571d5c69ad910f10df625bad73e [revision 2876]
  1124. Author: Anton Mitrofanov <[email protected]>
  1125. AuthorDate: Fri Sep 22 17:28:18 2017 +0300
  1126. Commit: Anton Mitrofanov <[email protected]>
  1127. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1128.  
  1129. Don't force fast-intra for subme < 3
  1130.  
  1131. It have caused significant quality hit without any meaningful (if any) speed up.
  1132.  
  1133. commit bdf27e783a8eb4a5bcae0cd0a950d6dc3d995bfe [revision 2875]
  1134. Author: Anton Mitrofanov <[email protected]>
  1135. AuthorDate: Fri Sep 22 17:18:55 2017 +0300
  1136. Commit: Anton Mitrofanov <[email protected]>
  1137. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1138.  
  1139. Make ref and i4x4_mode costs global instead of static
  1140.  
  1141. Fixes some thread safety doubts and makes code cleaner.
  1142. Downside: slightly higher memory usage when calling multiple encoders from the same application.
  1143.  
  1144. commit fefc3fa1fa98a7bac4eaf3c8e6e1c52b7e427ddd [revision 2874]
  1145. Author: Anton Mitrofanov <[email protected]>
  1146. AuthorDate: Fri Sep 22 17:05:06 2017 +0300
  1147. Commit: Anton Mitrofanov <[email protected]>
  1148. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1149.  
  1150. Fix thread safety of x264_threading_init() and use of X264_PTHREAD_MUTEX_INITIALIZER with win32thread
  1151.  
  1152. commit 694d031c1d120a8b578f60eeccf14fcf9ca4200e [revision 2873]
  1153. Author: Anton Mitrofanov <[email protected]>
  1154. AuthorDate: Fri Sep 22 16:59:13 2017 +0300
  1155. Commit: Anton Mitrofanov <[email protected]>
  1156. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1157.  
  1158. configure: Improvements
  1159.  
  1160. Log result of pkg-config checks to config.log.
  1161. Fix lavf support detection for pkg-config fallback case.
  1162. Fix detection of linking dependencies errors for lavf/lsmash/gpac.
  1163. Cosmetics.
  1164.  
  1165. commit 5d4031618e9feedcb527fd4e5a91bc06e30b70b4 [revision 2872]
  1166. Author: Anton Mitrofanov <[email protected]>
  1167. AuthorDate: Thu Aug 17 23:51:14 2017 +0300
  1168. Commit: Anton Mitrofanov <[email protected]>
  1169. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1170.  
  1171. flv: Fix one frame video total duration
  1172.  
  1173. commit 8b9c89d331f5a2d6335ff9b08abc8d5c94428731 [revision 2871]
  1174. Author: Anton Mitrofanov <[email protected]>
  1175. AuthorDate: Thu Aug 17 23:46:23 2017 +0300
  1176. Commit: Anton Mitrofanov <[email protected]>
  1177. CommitDate: Sun Dec 24 23:47:25 2017 +0300
  1178.  
  1179. flv: Split FrameType and CodecID values
  1180.  
  1181. commit 95cdb743463f723cea58c8ae01d7762f7ae9965c [revision 2870]
  1182. Author: Vittorio Giovara <[email protected]>
  1183. AuthorDate: Tue Aug 8 15:40:45 2017 +0200
  1184. Commit: Anton Mitrofanov <[email protected]>
  1185. CommitDate: Sun Dec 24 23:47:24 2017 +0300
  1186.  
  1187. Support writing the alternative transfer SEI message
  1188.  
  1189. commit c98d02bebd6dd04b61306ee27712aeff96f19f29 [revision 2869]
  1190. Author: Vittorio Giovara <[email protected]>
  1191. AuthorDate: Tue Aug 8 14:56:43 2017 +0200
  1192. Commit: Anton Mitrofanov <[email protected]>
  1193. CommitDate: Sun Dec 24 23:47:24 2017 +0300
  1194.  
  1195. Support 04/2017 color matrix and transfer values
  1196.  
  1197. commit 71ed44c7312438fac7c5c5301e45522e57127db4 [revision 2868]
  1198. Author: Vittorio Giovara <[email protected]>
  1199. AuthorDate: Fri Jan 6 15:23:38 2017 +0100
  1200. Commit: Anton Mitrofanov <[email protected]>
  1201. CommitDate: Sun Dec 24 23:47:24 2017 +0300
  1202.  
  1203. Unify 8-bit and 10-bit CLI and libraries
  1204.  
  1205. Add 'i_bitdepth' to x264_param_t with the corresponding '--output-depth' CLI
  1206. option to set the bit depth at runtime.
  1207.  
  1208. Drop the 'x264_bit_depth' global variable. Rather than hardcoding it to an
  1209. incorrect value, it's preferable to induce a linking failure. If applications
  1210. relies on this symbol this will make it more obvious where the problem is.
  1211.  
  1212. Add Makefile rules that compiles modules with different bit depths. Assembly
  1213. on x86 is prefixed with the 'private_prefix' define, while all other archs
  1214. modify their function prefix internally.
  1215.  
  1216. Templatize the main C library, x86/x86_64 assembly, ARM assembly, AARCH64
  1217. assembly, PowerPC assembly, and MIPS assembly.
  1218.  
  1219. The depth and cache CLI filters heavily depend on bit depth size, so they
  1220. need to be duplicated for each value. This means having to rename these
  1221. filters, and adjust the callers to use the right version.
  1222.  
  1223. Unfortunately the threaded input CLI module inherits a common.h dependency
  1224. (input/frame -> common/threadpool -> common/frame -> common/common) which
  1225. is extremely complicated to address in a sensible way. Instead duplicate
  1226. the module and select the appropriate one at run time.
  1227.  
  1228. Each bitdepth needs different checkasm compilation rules, so split the main
  1229. checkasm target into two executables.
  1230.  
  1231. commit 2451a7282463f68e532f2eee090a70ab139bb3e7 [revision 2867]
  1232. Author: Vittorio Giovara <[email protected]>
  1233. AuthorDate: Fri Jan 6 17:50:40 2017 +0100
  1234. Commit: Anton Mitrofanov <[email protected]>
  1235. CommitDate: Sun Dec 24 23:11:58 2017 +0300
  1236.  
  1237. Change default QP parameters initialization
  1238.  
  1239. qp is modified to require a valid value before use, while qp_max is set
  1240. to maximum allowable value (and clipped later on).
  1241.  
  1242. This is needed so that param functions do not depend on bit depth size.
  1243.  
  1244. commit 7839a9e1f03b49e3e0cbfcb3091093af7c6d54ee [revision 2866]
  1245. Author: Vittorio Giovara <[email protected]>
  1246. AuthorDate: Tue Jan 17 17:07:42 2017 +0100
  1247. Commit: Anton Mitrofanov <[email protected]>
  1248. CommitDate: Sun Dec 24 23:11:58 2017 +0300
  1249.  
  1250. aarch64: Set the function symbol prefix in a single location
  1251.  
  1252. commit 498cca0b74ab90c363b761083c7fdcf56fc60904 [revision 2865]
  1253. Author: Vittorio Giovara <[email protected]>
  1254. AuthorDate: Tue Jan 17 17:04:19 2017 +0100
  1255. Commit: Anton Mitrofanov <[email protected]>
  1256. CommitDate: Sun Dec 24 23:11:58 2017 +0300
  1257.  
  1258. arm: Set the function symbol prefix in a single location
  1259.  
  1260. commit 8f2437d33301faaf0e2fcaff16e2b01e9bbe27ae [revision 2864]
  1261. Author: Vittorio Giovara <[email protected]>
  1262. AuthorDate: Fri Jan 27 11:58:33 2017 +0100
  1263. Commit: Anton Mitrofanov <[email protected]>
  1264. CommitDate: Sun Dec 24 23:11:30 2017 +0300
  1265.  
  1266. Drop the x264 prefix from static functions and variables
  1267.  
  1268. commit 4e2ed4087ac1621f946b83366e1f53a1326d7424 [revision 2863]
  1269. Author: Anton Mitrofanov <[email protected]>
  1270. AuthorDate: Thu Aug 17 23:25:31 2017 +0300
  1271. Commit: Anton Mitrofanov <[email protected]>
  1272. CommitDate: Sun Dec 24 22:54:15 2017 +0300
  1273.  
  1274. configure: Check for strtok_r compiler support
  1275.  
  1276. commit d1eebb2927da15c41c7c180d398b0cdad3d1f396 [revision 2862]
  1277. Author: Henrik Gramner <[email protected]>
  1278. AuthorDate: Sun Aug 6 17:17:55 2017 +0200
  1279. Commit: Anton Mitrofanov <[email protected]>
  1280. CommitDate: Sun Dec 24 22:54:15 2017 +0300
  1281.  
  1282. cabac: Make the cabac_contexts array static
  1283.  
  1284. Also drop the x264 prefix from all static cabac arrays.
  1285.  
  1286. commit 3f9f6554a4cfa4189855756860a61ceb2f2a41a3 [revision 2861]
  1287. Author: Henrik Gramner <[email protected]>
  1288. AuthorDate: Thu Aug 17 18:04:13 2017 +0200
  1289. Commit: Anton Mitrofanov <[email protected]>
  1290. CommitDate: Sun Dec 24 22:54:15 2017 +0300
  1291.  
  1292. x86: AVX-512 pixel_satd_x3 and pixel_satd_x4
  1293.  
  1294. commit dd399ab862e2271e869bc8aefcb3166180ecdb10 [revision 2860]
  1295. Author: Henrik Gramner <[email protected]>
  1296. AuthorDate: Mon Aug 14 23:13:44 2017 +0200
  1297. Commit: Anton Mitrofanov <[email protected]>
  1298. CommitDate: Sun Dec 24 22:54:14 2017 +0300
  1299.  
  1300. x86: Shrink the x86-64 cabac coeff_last tables
  1301.  
  1302. Use dword instead of qword entries. Cuts the size of the tables in half
  1303. which allows each table fit inside a single cache line.
  1304.  
  1305. When PIC is disabled dwords are enough to store absolute addresses.
  1306.  
  1307. When PIC is enabled we can store dword offsets relative to the start of
  1308. the table and simply add the address of the table to the offset in order
  1309. to calculate the full address. This approach also have the advantage of
  1310. eliminating a whole bunch of run-time .data relocations.
  1311.  
  1312. commit d463a92e3b6f8ec04d54cc6c437892f9ffa98e29 [revision 2859]
  1313. Author: Henrik Gramner <[email protected]>
  1314. AuthorDate: Wed Aug 16 15:59:16 2017 +0200
  1315. Commit: Anton Mitrofanov <[email protected]>
  1316. CommitDate: Sun Dec 24 22:54:14 2017 +0300
  1317.  
  1318. x86inc: Support creating global symbols from local labels
  1319.  
  1320. On ELF platforms such symbols needs to be flagged as functions with the
  1321. correct visibility to please certain linkers in some scenarios.
  1322.  
  1323. commit 67b5c961046621a4554a9577e68cd9e31a212091 [revision 2858]
  1324. Author: Henrik Gramner <[email protected]>
  1325. AuthorDate: Tue Aug 15 16:11:32 2017 +0200
  1326. Commit: Anton Mitrofanov <[email protected]>
  1327. CommitDate: Sun Dec 24 22:54:14 2017 +0300
  1328.  
  1329. x86inc: Use .rdata instead of .rodata on Windows
  1330.  
  1331. The standard section for read-only data on Windows is .rdata. Nasm will
  1332. flag non-standard sections as executable by default which isn't ideal.
  1333.  
  1334. commit f15d366510cc60d9d9b2aeb576cade5b94509f37 [revision 2857]
  1335. Author: Henrik Gramner <[email protected]>
  1336. AuthorDate: Sat Aug 5 00:43:26 2017 +0200
  1337. Commit: Anton Mitrofanov <[email protected]>
  1338. CommitDate: Sun Dec 24 22:54:13 2017 +0300
  1339.  
  1340. x86inc: Set the correct cpuflag for AES-NI instructions
  1341.  
  1342. commit 1ae63361304e952ac625a7016f2cf4a64e39a314 [revision 2856]
  1343. Author: Henrik Gramner <[email protected]>
  1344. AuthorDate: Sat Aug 5 00:09:52 2017 +0200
  1345. Commit: Anton Mitrofanov <[email protected]>
  1346. CommitDate: Sun Dec 24 22:54:13 2017 +0300
  1347.  
  1348. x86inc: Enable AVX emulation for floating-point pseudo-instructions
  1349.  
  1350. There are 32 pseudo-instructions for each floating-point comparison
  1351. instruction, but only 8 of them are actually valid in legacy-encoded mode.
  1352. The remaining 24 requires the use of VEX-encoded (v-prefixed) instructions
  1353. and can therefore be disregarded for this purpose.
  1354.  
  1355. commit 1e27313c12154dd3922ef7ab9508a4320e83c2ac [revision 2855]
  1356. Author: Henrik Gramner <[email protected]>
  1357. AuthorDate: Fri Aug 4 23:09:00 2017 +0200
  1358. Commit: Anton Mitrofanov <[email protected]>
  1359. CommitDate: Sun Dec 24 22:54:13 2017 +0300
  1360.  
  1361. configure: Increase x86 stack alignment on clang
  1362.  
  1363. commit e9a5903edf8ca59ef20e6f4894c196f135af735e [revision 2854]
  1364. Author: Anton Mitrofanov <[email protected]>
  1365. AuthorDate: Sun Oct 22 20:18:39 2017 +0300
  1366. Commit: Anton Mitrofanov <[email protected]>
  1367. CommitDate: Sun Dec 24 22:54:12 2017 +0300
  1368.  
  1369. x86: Fix stack alignment for x264_cabac_encode_ue_bypass call
  1370.  
  1371. Fix MSVS fprofiled build for win64
  1372.  
  1373. commit 45e6eb6006d1d23b6f69a1cfb62a86dc67092a81 [revision 2853]
  1374. Author: Anton Mitrofanov <[email protected]>
  1375. AuthorDate: Sun Oct 22 16:18:29 2017 +0300
  1376. Commit: Anton Mitrofanov <[email protected]>
  1377. CommitDate: Sun Dec 24 22:54:12 2017 +0300
  1378.  
  1379. mips: Fix incorrect pointers to msa optimized functions
  1380.  
  1381. commit 09705c0b68232a05da8cc672c7c6092071eb4a21 [revision 2852]
  1382. Author: Henrik Gramner <[email protected]>
  1383. AuthorDate: Fri Aug 11 16:41:31 2017 +0200
  1384. Commit: Henrik Gramner <[email protected]>
  1385. CommitDate: Fri Aug 11 16:52:33 2017 +0200
  1386.  
  1387. Fix cpu capabilities listing on older x86 operating systems
  1388.  
  1389. Some cpuflags would previously be displayed incorrectly when running older
  1390. operating systems without AVX support on modern CPU:s.
  1391.  
  1392. commit ba24899b0bf23345921da022f7a51e0c57dbe73d [revision 2851]
  1393. Author: Henrik Gramner <[email protected]>
  1394. AuthorDate: Sat Jun 24 15:12:57 2017 +0200
  1395. Commit: Henrik Gramner <[email protected]>
  1396. CommitDate: Mon Jun 26 21:07:29 2017 +0200
  1397.  
  1398. x86: AVX-512 pixel_avg_weight_w8
  1399.  
  1400. commit d3214e6b102701911fc9d5fc92435e79e8b49100 [revision 2850]
  1401. Author: Henrik Gramner <[email protected]>
  1402. AuthorDate: Sat Jun 24 14:26:25 2017 +0200
  1403. Commit: Henrik Gramner <[email protected]>
  1404. CommitDate: Mon Jun 26 21:07:29 2017 +0200
  1405.  
  1406. x86: AVX-512 pixel_avg_weight_w16
  1407.  
  1408. commit 1d9dee2e9be717fcde416854f902db776312f141 [revision 2849]
  1409. Author: Henrik Gramner <[email protected]>
  1410. AuthorDate: Thu Jun 22 19:51:28 2017 +0200
  1411. Commit: Henrik Gramner <[email protected]>
  1412. CommitDate: Mon Jun 26 21:07:28 2017 +0200
  1413.  
  1414. x86: AVX-512 sub8x16_dct_dc
  1415.  
  1416. commit f672795407bf90045e399eb057e5b2426d79f961 [revision 2848]
  1417. Author: Henrik Gramner <[email protected]>
  1418. AuthorDate: Thu Jun 22 11:26:21 2017 +0200
  1419. Commit: Henrik Gramner <[email protected]>
  1420. CommitDate: Sat Jun 24 16:23:50 2017 +0200
  1421.  
  1422. x86: AVX-512 sub8x8_dct_dc
  1423.  
  1424. commit 0af1c6d0d0cc54ba4f888db39247774edcf19b44 [revision 2847]
  1425. Author: Henrik Gramner <[email protected]>
  1426. AuthorDate: Thu Jun 1 22:13:19 2017 +0200
  1427. Commit: Henrik Gramner <[email protected]>
  1428. CommitDate: Sat Jun 24 16:23:50 2017 +0200
  1429.  
  1430. x86: AVX-512 add8x8_idct
  1431.  
  1432. commit 9034085265e5ca56e801c3efbf5c538fcc17c82b [revision 2846]
  1433. Author: Henrik Gramner <[email protected]>
  1434. AuthorDate: Sat Jun 10 16:01:53 2017 +0200
  1435. Commit: Henrik Gramner <[email protected]>
  1436. CommitDate: Sat Jun 24 16:23:50 2017 +0200
  1437.  
  1438. x86: AVX-512 sub16x16_dct
  1439.  
  1440. commit 774c6c76d081305d9c891091e1d4694acb3f8a68 [revision 2845]
  1441. Author: Henrik Gramner <[email protected]>
  1442. AuthorDate: Wed Jun 7 16:55:48 2017 +0200
  1443. Commit: Henrik Gramner <[email protected]>
  1444. CommitDate: Sat Jun 24 16:23:50 2017 +0200
  1445.  
  1446. x86: AVX-512 sub8x8_dct
  1447.  
  1448. commit 2d653411c2135377fb8c956e897880ff997b50ec [revision 2844]
  1449. Author: Henrik Gramner <[email protected]>
  1450. AuthorDate: Thu Jun 8 21:14:08 2017 +0200
  1451. Commit: Henrik Gramner <[email protected]>
  1452. CommitDate: Sat Jun 24 16:23:50 2017 +0200
  1453.  
  1454. x86: AVX-512 sub4x4_dct
  1455.  
  1456. commit 07483f72d7e1a4f7079a429dd1370f4221006862 [revision 2843]
  1457. Author: Henrik Gramner <[email protected]>
  1458. AuthorDate: Sun May 28 16:12:33 2017 +0200
  1459. Commit: Henrik Gramner <[email protected]>
  1460. CommitDate: Sat Jun 24 16:23:50 2017 +0200
  1461.  
  1462. x86: AVX-512 mbtree_propagate_list
  1463.  
  1464. Uses gathers and scatters in combination with conflict detections to
  1465. vectorize the scalar part.
  1466.  
  1467. Also improve the checkasm test to try different mb_y values and check
  1468. for out-of-bounds writes.
  1469.  
  1470. commit 1a88481b85da964aadae1e05347342b03be17712 [revision 2842]
  1471. Author: James Darnley <[email protected]>
  1472. AuthorDate: Fri Jun 9 14:08:16 2017 +0200
  1473. Commit: Henrik Gramner <[email protected]>
  1474. CommitDate: Sat Jun 24 16:23:31 2017 +0200
  1475.  
  1476. x86inc: Add aesni cpuflag define
  1477.  
  1478. Upstreaming this from FFmpeg. Unused in x264.
  1479.  
  1480. commit 98e9543b4c39360326e6d5bf266c0c634cb9ee2e [revision 2841]
  1481. Author: Martin Storsjö <[email protected]>
  1482. AuthorDate: Mon May 29 12:13:03 2017 +0300
  1483. Commit: Anton Mitrofanov <[email protected]>
  1484. CommitDate: Wed Jun 14 23:24:38 2017 +0300
  1485.  
  1486. aarch64: Update the var2 functions to the new signature
  1487.  
  1488. The existing functions could easily be used by just calling them
  1489. twice - this would give the following cycle numbers from checkasm:
  1490.  
  1491. var2_8x8_c: 4110
  1492. var2_8x8_neon: 1505
  1493. var2_8x16_c: 8019
  1494. var2_8x16_neon: 2545
  1495.  
  1496. However, by merging both passes into the same function, we get the
  1497. following speedup:
  1498. var2_8x8_neon: 1205
  1499. var2_8x16_neon: 2327
  1500.  
  1501. commit 824802ad5a877244fb9eb48a892ed348736af5b0 [revision 2840]
  1502. Author: Martin Storsjö <[email protected]>
  1503. AuthorDate: Mon May 29 12:13:02 2017 +0300
  1504. Commit: Anton Mitrofanov <[email protected]>
  1505. CommitDate: Wed Jun 14 23:24:38 2017 +0300
  1506.  
  1507. arm: Update the var2 functions to the new signature
  1508.  
  1509. The existing functions could easily be used by just calling them
  1510. twice - this would give the following cycle numbers from checkasm:
  1511.  
  1512. Cortex A7 A8 A9 A53
  1513. var2_8x8_c: 7302 5342 5050 4400
  1514. var2_8x8_neon: 2645 1612 1932 1715
  1515. var2_8x16_c: 14300 10528 10020 8637
  1516. var2_8x16_neon: 5127 2695 3217 2651
  1517.  
  1518. However, by merging both passes into the same function, we get the
  1519. following speedup:
  1520. var2_8x8_neon: 2312 1190 1389 1300
  1521. var2_8x16_neon: 4862 2130 2293 2422
  1522.  
  1523. commit 6f8aa71ce797be01fd2ebe53c072a6696ea19828 [revision 2839]
  1524. Author: Henrik Gramner <[email protected]>
  1525. AuthorDate: Wed Feb 15 22:00:25 2017 +0100
  1526. Commit: Anton Mitrofanov <[email protected]>
  1527. CommitDate: Wed Jun 14 23:24:38 2017 +0300
  1528.  
  1529. Add support for levels 6, 6.1, and 6.2
  1530.  
  1531. These levels were added in the 2016-10 revision of the H.264 specification and
  1532. improves support for content with high resolutions and/or high frame rates.
  1533.  
  1534. Level 6.2 supports 8K resolution at 120 fps.
  1535.  
  1536. Also shrink the x264_levels array by using smaller data types.
  1537.  
  1538. commit 2baa28c880d11377115bbd5508e72053f6ba61f5 [revision 2838]
  1539. Author: Henrik Gramner <[email protected]>
  1540. AuthorDate: Thu Mar 23 17:51:09 2017 +0100
  1541. Commit: Anton Mitrofanov <[email protected]>
  1542. CommitDate: Wed Jun 14 23:22:58 2017 +0300
  1543.  
  1544. Use a larger integer type for the slice_table array
  1545.  
  1546. Makes it possible to use slicing with resolutions larger than 2^24 pixels.
  1547.  
  1548. commit c9d2c1c80b25c6ae15c41b200ec44ac2dabce725 [revision 2837]
  1549. Author: Henrik Gramner <[email protected]>
  1550. AuthorDate: Sun Feb 19 10:48:33 2017 +0100
  1551. Commit: Anton Mitrofanov <[email protected]>
  1552. CommitDate: Wed Jun 14 23:22:57 2017 +0300
  1553.  
  1554. analyse: Reduce the size the cost_mv arrays
  1555.  
  1556. Use a dynamic size depending on the MV range. Reduces memory consumption by
  1557. up to a few megabytes.
  1558.  
  1559. Drop a related old miscompilation check since it may otherwise cause an
  1560. out-of-bounds memory access.
  1561.  
  1562. Also remove an unused extern variable declaration.
  1563.  
  1564. commit d46a5a463f0de5ec479d256af72bba3de4ba2d1a [revision 2836]
  1565. Author: Anton Mitrofanov <[email protected]>
  1566. AuthorDate: Wed May 31 02:52:16 2017 +0300
  1567. Commit: Anton Mitrofanov <[email protected]>
  1568. CommitDate: Wed Jun 14 23:22:57 2017 +0300
  1569.  
  1570. Fix CABAC+8x8dct in 4:4:4
  1571.  
  1572. Use the correct ctxIdxInc calculation for coded_block_flag.
  1573.  
  1574. commit 79b36f27a57dd511eefead6d5422689220c767b5 [revision 2835]
  1575. Author: Anton Mitrofanov <[email protected]>
  1576. AuthorDate: Tue Jun 6 02:07:21 2017 +0300
  1577. Commit: Anton Mitrofanov <[email protected]>
  1578. CommitDate: Wed Jun 14 23:19:45 2017 +0300
  1579.  
  1580. Fix 8x8dct in lossless encoding
  1581.  
  1582. Change V and H intra prediction in lossless (TransformBypassModeFlag == 1)
  1583. macroblocks to correctly adhere to the specification. Affects lossless
  1584. encoding with 8x8dct or mix of lossless with normal macroblocks.
  1585.  
  1586. 8x8dct has already been disabled in lossless mode for some time due to
  1587. being out-of-spec but this will allow us to re-enable it again.
  1588.  
  1589. commit 68a550217c8d0fae6229c5b322b6810fe9652ef3 [revision 2834]
  1590. Author: Anton Mitrofanov <[email protected]>
  1591. AuthorDate: Thu Jun 8 18:35:21 2017 +0300
  1592. Commit: Anton Mitrofanov <[email protected]>
  1593. CommitDate: Wed Jun 14 23:19:45 2017 +0300
  1594.  
  1595. mbtree: Fix buffer overflow
  1596.  
  1597. Could occur on the 1st pass in combination with --fake-interlaced and
  1598. some input heights due to allocating a too small buffer.
  1599.  
  1600. commit df79067c0cf33da712d344b5f8869be7eaf326f3 [revision 2833]
  1601. Author: Henrik Gramner <[email protected]>
  1602. AuthorDate: Tue May 23 16:40:26 2017 +0200
  1603. Commit: Henrik Gramner <[email protected]>
  1604. CommitDate: Tue May 23 17:35:18 2017 +0200
  1605.  
  1606. x86: Avoid self-relative expressions on macho64
  1607.  
  1608. Functions that uses self-relative expressions in the form of [foo-$$]
  1609. appears to cause issues on 64-bit Mach-O systems when assembled with nasm.
  1610. Temporarily disable those functions on macho64 for the time being until
  1611. we've figured out the root cause.
  1612.  
  1613. commit f1ac7122645bbeb56e7a4401f71a7055cb2431c4 [revision 2832]
  1614. Author: Anton Mitrofanov <[email protected]>
  1615. AuthorDate: Mon May 22 23:59:32 2017 +0300
  1616. Commit: Henrik Gramner <[email protected]>
  1617. CommitDate: Tue May 23 16:57:45 2017 +0200
  1618.  
  1619. configure: Don't try to detect clang by $CC
  1620.  
  1621. Only check if option -Werror=unknown-warning-option is supported before adding it
  1622.  
  1623. commit b4d811df4fd7dbb9220fe2c8f2a2c2a6ba2bbc87 [revision 2831]
  1624. Author: Martin Storsjö <[email protected]>
  1625. AuthorDate: Mon May 22 13:10:46 2017 +0300
  1626. Commit: Henrik Gramner <[email protected]>
  1627. CommitDate: Tue May 23 16:57:44 2017 +0200
  1628.  
  1629. checkasm: Use the right variable in a loop condition
  1630.  
  1631. Prior to this, this loop hasn't run at all. The condition has been
  1632. the same since it was introduced in 5b0cb86f.
  1633.  
  1634. This issue was pointed out by a clang warning.
  1635.  
  1636. commit a3d24462ae284bf03958f0ed41e824dd7d48e15e [revision 2830]
  1637. Author: Anton Mitrofanov <[email protected]>
  1638. AuthorDate: Mon May 22 22:02:34 2017 +0300
  1639. Commit: Henrik Gramner <[email protected]>
  1640. CommitDate: Tue May 23 16:11:03 2017 +0200
  1641.  
  1642. x86: Fix linking with 8-bit depth shared libx264
  1643.  
  1644. commit d1fe6fd1c0930d88da90f23f6d5fdb6ceaf6b0a9 [revision 2829]
  1645. Author: Henrik Gramner <[email protected]>
  1646. AuthorDate: Mon May 15 00:18:36 2017 +0200
  1647. Commit: Henrik Gramner <[email protected]>
  1648. CommitDate: Sun May 21 23:14:25 2017 +0200
  1649.  
  1650. x86: Only enable AVX-512 in 8-bit mode
  1651.  
  1652. commit 6151882671b6f9e1ceec2cdb76dd1123c8dc766f [revision 2828]
  1653. Author: Henrik Gramner <[email protected]>
  1654. AuthorDate: Fri May 12 00:43:43 2017 +0200
  1655. Commit: Henrik Gramner <[email protected]>
  1656. CommitDate: Sun May 21 23:14:25 2017 +0200
  1657.  
  1658. x86: AVX-512 cabac_block_residual
  1659.  
  1660. commit 4579616543f2e701ee9510f5eb57e31a3ef99e10 [revision 2827]
  1661. Author: Henrik Gramner <[email protected]>
  1662. AuthorDate: Wed May 10 18:36:59 2017 +0200
  1663. Commit: Henrik Gramner <[email protected]>
  1664. CommitDate: Sun May 21 23:14:25 2017 +0200
  1665.  
  1666. x86: AVX-512 pixel_sad_x3 and pixel_sad_x4
  1667.  
  1668. Covers all variants: 4x4, 4x8, 8x4, 8x8, 8x16, 16x8, and 16x16.
  1669.  
  1670. commit 993eb2079e45619098241e14806fc70030968af6 [revision 2826]
  1671. Author: Henrik Gramner <[email protected]>
  1672. AuthorDate: Sun May 7 23:35:49 2017 +0200
  1673. Commit: Henrik Gramner <[email protected]>
  1674. CommitDate: Sun May 21 23:14:25 2017 +0200
  1675.  
  1676. x86: AVX-512 pixel_sad
  1677.  
  1678. Covers all variants: 4x4, 4x8, 4x16, 8x4, 8x8, 8x16, 16x8, and 16x16.
  1679.  
  1680. commit 2463174c119cef4f7e6a36a1151054fbb268b082 [revision 2825]
  1681. Author: Henrik Gramner <[email protected]>
  1682. AuthorDate: Thu May 4 21:53:28 2017 +0200
  1683. Commit: Henrik Gramner <[email protected]>
  1684. CommitDate: Sun May 21 23:14:25 2017 +0200
  1685.  
  1686. x86: AVX-512 decimate_score
  1687.  
  1688. Also drop the MMX versions and improve the SSE2, SSSE3 and AVX2 versions.
  1689.  
  1690. commit 49fb50a67cc41e4bed2dd66f7beed12797249cd9 [revision 2824]
  1691. Author: Henrik Gramner <[email protected]>
  1692. AuthorDate: Mon May 1 14:55:45 2017 +0200
  1693. Commit: Henrik Gramner <[email protected]>
  1694. CommitDate: Sun May 21 23:14:25 2017 +0200
  1695.  
  1696. x86: AVX-512 pixel_var2_8x8 and 8x16
  1697.  
  1698. commit 92c074e27f6bfccee42b41c183203b7b2763a94d [revision 2823]
  1699. Author: Henrik Gramner <[email protected]>
  1700. AuthorDate: Mon May 1 14:54:32 2017 +0200
  1701. Commit: Henrik Gramner <[email protected]>
  1702. CommitDate: Sun May 21 23:14:25 2017 +0200
  1703.  
  1704. Rework pixel_var2
  1705.  
  1706. The functions are only ever called with pointers to fenc and fdec and the
  1707. strides are always constant so there's no point in having them as parameters.
  1708.  
  1709. Cover both the U and V planes in a single function call. This is more
  1710. efficient with SIMD, especially with the wider vectors provided by AVX2 and
  1711. AVX-512, even when accounting for losing the possibility of early termination.
  1712.  
  1713. Drop the MMX and XOP implementations, update the rest of the x86 assembly
  1714. to match the new behavior. Also enable high bit-depth in the AVX2 version.
  1715.  
  1716. Comment out the ARM, AARCH64, and MIPS MSA assembly for now.
  1717.  
  1718. commit 4c48f9e751e969188d606eb15aeada7f652c9db9 [revision 2822]
  1719. Author: Henrik Gramner <[email protected]>
  1720. AuthorDate: Sat Apr 29 14:26:40 2017 +0200
  1721. Commit: Henrik Gramner <[email protected]>
  1722. CommitDate: Sun May 21 23:14:25 2017 +0200
  1723.  
  1724. x86: AVX-512 pixel_var_8x8, 8x16, and 16x16
  1725.  
  1726. Make the SSE2, AVX, and AVX2 versions a bit faster.
  1727.  
  1728. Drop the MMX and XOP versions.
  1729.  
  1730. commit 1cf7baa462ca52de7f07d6e4c795853900bb50bb [revision 2821]
  1731. Author: Henrik Gramner <[email protected]>
  1732. AuthorDate: Fri Apr 28 21:35:25 2017 +0200
  1733. Commit: Henrik Gramner <[email protected]>
  1734. CommitDate: Sun May 21 23:14:24 2017 +0200
  1735.  
  1736. x86: AVX-512 pixel_sa8d_8x8
  1737.  
  1738. commit 386050088a66aa66bcaebb9b6b4b0a2b6af76a73 [revision 2820]
  1739. Author: Henrik Gramner <[email protected]>
  1740. AuthorDate: Thu Apr 13 23:56:04 2017 +0200
  1741. Commit: Henrik Gramner <[email protected]>
  1742. CommitDate: Sun May 21 23:14:24 2017 +0200
  1743.  
  1744. x86: AVX-512 pixel_satd
  1745.  
  1746. Covers all variants: 4x4, 4x8, 4x16, 8x4, 8x8, 8x16, 16x8, and 16x16.
  1747.  
  1748. commit 2eceefe89fea91bbc7d5af2a1b4a9047d8da7805 [revision 2819]
  1749. Author: Henrik Gramner <[email protected]>
  1750. AuthorDate: Wed Apr 19 16:39:48 2017 +0200
  1751. Commit: Henrik Gramner <[email protected]>
  1752. CommitDate: Sun May 21 23:14:24 2017 +0200
  1753.  
  1754. x86: AVX-512 deblock_strength
  1755.  
  1756. Also drop the MMX version and make some slight improvements to the SSE2,
  1757. SSSE3, AVX, and AVX2 versions.
  1758.  
  1759. commit 3081ffa1c540d1df05123e0fab1937985573ac78 [revision 2818]
  1760. Author: Henrik Gramner <[email protected]>
  1761. AuthorDate: Wed Apr 12 16:21:09 2017 +0200
  1762. Commit: Henrik Gramner <[email protected]>
  1763. CommitDate: Sun May 21 23:14:24 2017 +0200
  1764.  
  1765. x86: AVX-512 plane_copy_deinterleave_v210
  1766.  
  1767. commit 95dc64c4efdf16404e58be9ff9da4e0acaa1a4b2 [revision 2817]
  1768. Author: Henrik Gramner <[email protected]>
  1769. AuthorDate: Sun Apr 9 20:34:28 2017 +0200
  1770. Commit: Henrik Gramner <[email protected]>
  1771. CommitDate: Sun May 21 23:14:24 2017 +0200
  1772.  
  1773. x86: AVX-512 memzero_aligned
  1774.  
  1775. Reorder some elements in the x264_t.mb.pic struct to reduce the amount
  1776. of padding required.
  1777.  
  1778. Also drop the MMX implementation in favor of SSE.
  1779.  
  1780. commit c0cd7650cb65164d183d8f77d0697b7569a52917 [revision 2816]
  1781. Author: Henrik Gramner <[email protected]>
  1782. AuthorDate: Fri Apr 7 21:34:40 2017 +0200
  1783. Commit: Henrik Gramner <[email protected]>
  1784. CommitDate: Sun May 21 23:14:24 2017 +0200
  1785.  
  1786. x86: AVX and AVX-512 memcpy_aligned
  1787.  
  1788. Reorder some elements in the x264_mb_analysis_list_t struct to reduce the
  1789. amount of padding required.
  1790.  
  1791. Also drop the MMX implementation in favor of SSE.
  1792.  
  1793. commit f29fbc6fd23e9bc2d800eb1246e8fa19a203b831 [revision 2815]
  1794. Author: Henrik Gramner <[email protected]>
  1795. AuthorDate: Thu Apr 6 16:06:34 2017 +0200
  1796. Commit: Henrik Gramner <[email protected]>
  1797. CommitDate: Sun May 21 22:42:17 2017 +0200
  1798.  
  1799. x86: AVX-512 dequant_8x8_flat16
  1800.  
  1801. commit 40aca29a164d5e5e6589d507bdcae6717d72f6bf [revision 2814]
  1802. Author: Henrik Gramner <[email protected]>
  1803. AuthorDate: Tue Apr 4 20:54:12 2017 +0200
  1804. Commit: Henrik Gramner <[email protected]>
  1805. CommitDate: Sun May 21 22:42:17 2017 +0200
  1806.  
  1807. x86: AVX-512 dequant_8x8
  1808.  
  1809. commit 74f7802bb7bd301299f8229a0552a7caf2b55434 [revision 2813]
  1810. Author: Henrik Gramner <[email protected]>
  1811. AuthorDate: Tue Apr 4 20:01:26 2017 +0200
  1812. Commit: Henrik Gramner <[email protected]>
  1813. CommitDate: Sun May 21 22:42:17 2017 +0200
  1814.  
  1815. x86: AVX-512 dequant_4x4
  1816.  
  1817. commit 3451ba3af49e58a720277615df3d8e4a4171986f [revision 2812]
  1818. Author: Henrik Gramner <[email protected]>
  1819. AuthorDate: Tue Mar 28 22:59:56 2017 +0200
  1820. Commit: Henrik Gramner <[email protected]>
  1821. CommitDate: Sun May 21 22:42:17 2017 +0200
  1822.  
  1823. x86: AVX-512 mbtree_propagate_cost
  1824.  
  1825. Also make the AVX and AVX2 implementations slightly faster.
  1826.  
  1827. commit 75f6f9b228c3498b8c9b0d97fc925c0a7e6e6f43 [revision 2811]
  1828. Author: Henrik Gramner <[email protected]>
  1829. AuthorDate: Mon Mar 27 18:19:53 2017 +0200
  1830. Commit: Henrik Gramner <[email protected]>
  1831. CommitDate: Sun May 21 22:42:17 2017 +0200
  1832.  
  1833. x86: AVX-512 coeff_last
  1834.  
  1835. commit c3a1d1d892a79bc460c7fc192b0bf7a32c2ce0b2 [revision 2810]
  1836. Author: Henrik Gramner <[email protected]>
  1837. AuthorDate: Sun Mar 26 18:29:37 2017 +0200
  1838. Commit: Henrik Gramner <[email protected]>
  1839. CommitDate: Sun May 21 22:42:17 2017 +0200
  1840.  
  1841. x86: AVX-512 zigzag_interleave_8x8_cavlc
  1842.  
  1843. commit edb22f57ba03718c1cb9781ba005aec20a1e50e0 [revision 2809]
  1844. Author: Henrik Gramner <[email protected]>
  1845. AuthorDate: Sun Mar 26 11:34:18 2017 +0200
  1846. Commit: Henrik Gramner <[email protected]>
  1847. CommitDate: Sun May 21 22:42:17 2017 +0200
  1848.  
  1849. x86: AVX-512 zigzag_scan_8x8_field
  1850.  
  1851. commit 77b9a818fc622d0cdaa96aeb37339fbd5b1ef857 [revision 2808]
  1852. Author: Henrik Gramner <[email protected]>
  1853. AuthorDate: Sat Mar 25 22:13:22 2017 +0100
  1854. Commit: Henrik Gramner <[email protected]>
  1855. CommitDate: Sun May 21 22:42:17 2017 +0200
  1856.  
  1857. x86: AVX-512 zigzag_scan_4x4_field
  1858.  
  1859. commit 724a577237f27cdb0c0fd18ef8ed32d39430796b [revision 2807]
  1860. Author: Henrik Gramner <[email protected]>
  1861. AuthorDate: Sat Mar 25 19:14:28 2017 +0100
  1862. Commit: Henrik Gramner <[email protected]>
  1863. CommitDate: Sun May 21 22:42:17 2017 +0200
  1864.  
  1865. x86: AVX-512 zigzag_scan_8x8_frame
  1866.  
  1867. The vperm* instructions ignores unused bits, so we can pack the permutation
  1868. indices together to save cache and just use a shift to get the right values.
  1869.  
  1870. commit 2b2f039512bde7c097280255c6376cf9a901e08e [revision 2806]
  1871. Author: Henrik Gramner <[email protected]>
  1872. AuthorDate: Sat Mar 25 19:14:22 2017 +0100
  1873. Commit: Henrik Gramner <[email protected]>
  1874. CommitDate: Sun May 21 22:42:17 2017 +0200
  1875.  
  1876. x86: AVX-512 zigzag_scan_4x4_frame
  1877.  
  1878. commit 1878c7f2af0a9c73e291488209109782c428cfcf [revision 2805]
  1879. Author: Henrik Gramner <[email protected]>
  1880. AuthorDate: Fri May 12 00:03:10 2017 +0200
  1881. Commit: Henrik Gramner <[email protected]>
  1882. CommitDate: Sun May 21 22:42:17 2017 +0200
  1883.  
  1884. checkasm: x86: More accurate ymm/zmm measurements
  1885.  
  1886. YMM and ZMM registers on x86 are turned off to save power when they haven't
  1887. been used for some period of time. When they are used there will be a
  1888. "warmup" period during which performance will be reduced and inconsistent
  1889. which is problematic when trying to benchmark individual functions.
  1890.  
  1891. Periodically issue "dummy" instructions that uses those registers to
  1892. prevent them from being powered down. The end result is more consitent
  1893. benchmark results.
  1894.  
  1895. commit 472ce3648aea3ddc16b7716eb114f4bcdb8fea8f [revision 2804]
  1896. Author: Henrik Gramner <[email protected]>
  1897. AuthorDate: Sat Mar 25 10:16:09 2017 +0100
  1898. Commit: Henrik Gramner <[email protected]>
  1899. CommitDate: Sun May 21 22:42:15 2017 +0200
  1900.  
  1901. x86: AVX-512 support
  1902.  
  1903. AVX-512 consists of a plethora of different extensions, but in order to keep
  1904. things a bit more manageable we group together the following extensions
  1905. under a single baseline cpu flag which should cover SKL-X and future CPUs:
  1906. * AVX-512 Foundation (F)
  1907. * AVX-512 Conflict Detection Instructions (CD)
  1908. * AVX-512 Byte and Word Instructions (BW)
  1909. * AVX-512 Doubleword and Quadword Instructions (DQ)
  1910. * AVX-512 Vector Length Extensions (VL)
  1911.  
  1912. On x86-64 AVX-512 provides 16 additional vector registers, prefer using
  1913. those over existing ones since it allows us to avoid using `vzeroupper`
  1914. unless more than 16 vector registers are required. They also happen to
  1915. be volatile on Windows which means that we don't need to save and restore
  1916. existing xmm register contents unless more than 22 vector registers are
  1917. required.
  1918.  
  1919. Also take the opportunity to drop X264_CPU_CMOV and X264_CPU_SLOW_CTZ while
  1920. we're breaking API by messing with the cpu flags since they weren't really
  1921. used for anything.
  1922.  
  1923. Big thanks to Intel for their support.
  1924.  
  1925. commit d2b5f4873e2147452a723b61b14f030b2ee760a5 [revision 2803]
  1926. Author: Henrik Gramner <[email protected]>
  1927. AuthorDate: Sat Mar 18 18:50:36 2017 +0100
  1928. Commit: Henrik Gramner <[email protected]>
  1929. CommitDate: Sun May 21 22:41:07 2017 +0200
  1930.  
  1931. x86: Change assembler from yasm to nasm
  1932.  
  1933. This is required to support AVX-512.
  1934.  
  1935. Drop `-Worphan-labels` from ASFLAGS since it's enabled by default in nasm.
  1936.  
  1937. Also change alignmode from `k8` to `p6` since it's more similar to `amdnop`
  1938. in yasm, e.g. use long nops without excessive prefixes.
  1939.  
  1940. commit 8c2974255b01728d4eda2434cc1997c4a3ca5eff [revision 2802]
  1941. Author: Henrik Gramner <[email protected]>
  1942. AuthorDate: Sat May 6 12:26:56 2017 +0200
  1943. Commit: Henrik Gramner <[email protected]>
  1944. CommitDate: Sun May 21 22:41:07 2017 +0200
  1945.  
  1946. x86: Add some additional cpuflag relations
  1947.  
  1948. Simplifies writing assembly code that depends on available instructions.
  1949.  
  1950. LZCNT implies SSE2
  1951. BMI1 implies AVX+LZCNT
  1952. AVX2 implies BMI2
  1953.  
  1954. Skip printing LZCNT under CPU capabilities when BMI1 or BMI2 is available,
  1955. and don't print FMA4 when FMA3 is available.
  1956.  
  1957. commit 93bc2cbc66f0bf4616965dcd7e0eba89201c8086 [revision 2801]
  1958. Author: Henrik Gramner <[email protected]>
  1959. AuthorDate: Fri Apr 14 16:16:49 2017 +0200
  1960. Commit: Henrik Gramner <[email protected]>
  1961. CommitDate: Sun May 21 22:41:07 2017 +0200
  1962.  
  1963. x86: Faster SSE2 pixel_sad_16x16 and 16x8
  1964.  
  1965. Also make the order of fenc/fdec arguments a bit more consistent.
  1966.  
  1967. commit 8ae2b62462176cd731a1cb8b5bdc9a38cba0fbe4 [revision 2800]
  1968. Author: Anton Mitrofanov <[email protected]>
  1969. AuthorDate: Mon May 15 00:40:52 2017 +0300
  1970. Commit: Henrik Gramner <[email protected]>
  1971. CommitDate: Sun May 21 22:41:07 2017 +0200
  1972.  
  1973. msvs/icl: Improve target host detection
  1974.  
  1975. commit 181a920ad5d0acdc3a08418c0e9c95be4785b814 [revision 2799]
  1976. Author: Alexandra Hájková <[email protected]>
  1977. AuthorDate: Sat May 13 17:14:52 2017 +0000
  1978. Commit: Henrik Gramner <[email protected]>
  1979. CommitDate: Sun May 21 22:41:07 2017 +0200
  1980.  
  1981. ppc: Optimize add8x8_idct_dc
  1982.  
  1983. Increases speedup compared to C from 2x to 6x.
  1984.  
  1985. commit d0b905b901c5ee5989777cf437a7f20c1fa0a794 [revision 2798]
  1986. Author: Henrik Gramner <[email protected]>
  1987. AuthorDate: Sun Feb 19 10:33:16 2017 +0100
  1988. Commit: Henrik Gramner <[email protected]>
  1989. CommitDate: Sun May 21 22:41:07 2017 +0200
  1990.  
  1991. analyse: Faster min/max MV clipping
  1992.  
  1993. Values only needs to be clipped in one direction.
  1994.  
  1995. commit 1bde30193eb91d1bc69b00a27e6874eb88ed4eab [revision 2797]
  1996. Author: Henrik Gramner <[email protected]>
  1997. AuthorDate: Thu Feb 16 20:04:10 2017 +0100
  1998. Commit: Henrik Gramner <[email protected]>
  1999. CommitDate: Sun May 21 22:41:07 2017 +0200
  2000.  
  2001. slicetype_mb_cost: Clip MVs based on MV range
  2002.  
  2003. Improves cost calculations, especially when a short MV range is used.
  2004.  
  2005. commit dcf406978b9dda5c2b8aab80af5c1c47c78efd92 [revision 2796]
  2006. Author: Henrik Gramner <[email protected]>
  2007. AuthorDate: Sun Jan 29 21:38:43 2017 +0100
  2008. Commit: Henrik Gramner <[email protected]>
  2009. CommitDate: Sun May 21 22:41:07 2017 +0200
  2010.  
  2011. Support YUYV and UYVY packed 4:2:2 raw input
  2012.  
  2013. Packed YUV is arguably more common than planar YUV when dealing with raw
  2014. 4:2:2 content.
  2015.  
  2016. We can utilize the existing plane_copy_deinterleave() functions with some
  2017. additional minor constraints (we cannot assume any particular alignment
  2018. or overread the input buffer).
  2019.  
  2020. Enables assembly optimizations on x86.
  2021.  
  2022. commit aaa9aa83a111ed6f1db253d5afa91c5fc844583f [revision 2795]
  2023. Author: Henrik Gramner <[email protected]>
  2024. AuthorDate: Thu Apr 20 21:58:23 2017 +0200
  2025. Commit: Henrik Gramner <[email protected]>
  2026. CommitDate: Sun May 21 22:41:07 2017 +0200
  2027.  
  2028. x86: Utilize 3-arg instructions in AVX deblock
  2029.  
  2030. Avoids some redundant register-register moves.
  2031.  
  2032. commit a52d41c4d135c79373a86c3a82dcc2ec3f88b025 [revision 2794]
  2033. Author: Martin Storsjö <[email protected]>
  2034. AuthorDate: Fri Mar 24 11:33:46 2017 +0200
  2035. Commit: Henrik Gramner <[email protected]>
  2036. CommitDate: Sun May 21 22:41:07 2017 +0200
  2037.  
  2038. configure: Support targeting ARM with MSVC tools
  2039.  
  2040. Set up the right gas-preprocessor as assembler frontend in these cases,
  2041. using armasm as actual assembler.
  2042.  
  2043. Don't try to add the -mcpu -mfpu options in this case.
  2044.  
  2045. Check whether the compiler actually supports inline assembly.
  2046.  
  2047. Check for the ARMv7 features in a different way for the MSVC compiler.
  2048.  
  2049. commit b22a5db3c481b10b4a6ec190978d97b377750a12 [revision 2793]
  2050. Author: Martin Storsjö <[email protected]>
  2051. AuthorDate: Fri Mar 24 11:33:45 2017 +0200
  2052. Commit: Henrik Gramner <[email protected]>
  2053. CommitDate: Sun May 21 22:41:07 2017 +0200
  2054.  
  2055. configure: Check for -lshell32 before forcibly adding it into LDFLAGSCLI
  2056.  
  2057. When targeting the Windows Phone API subset, there is no shell32.lib.
  2058.  
  2059. When targeting Windows Phone/RT, the CLI itself won't be built, but
  2060. LDFLAGSCLI are included in all later cases of cc_check within configure.
  2061. Therefore only add -lshell32 there if it actually is usable.
  2062.  
  2063. commit 0aed59e74808f1cd22ee47c055a8eb4f367b2f55 [revision 2792]
  2064. Author: Martin Storsjö <[email protected]>
  2065. AuthorDate: Thu May 4 22:00:51 2017 +0300
  2066. Commit: Henrik Gramner <[email protected]>
  2067. CommitDate: Sun May 21 22:41:07 2017 +0200
  2068.  
  2069. arm: Always unconditionally declare .arch armv7-a
  2070.  
  2071. We already unconditionally declare .fpu neon and try to build all the
  2072. neon codepaths (but only execute them conditionally based on a runtime
  2073. check).
  2074.  
  2075. This fixes builds targeting armv6, where the rbit instruction isn't
  2076. available. This instruction is only used within a neon function in
  2077. any case, so there's little point in emulating it.
  2078.  
  2079. commit 196d7676c8f40b7c1f8f2f4af64e09ebf4c9816b [revision 2791]
  2080. Author: Martin Storsjö <[email protected]>
  2081. AuthorDate: Fri Mar 24 11:33:44 2017 +0200
  2082. Commit: Henrik Gramner <[email protected]>
  2083. CommitDate: Sun May 21 22:41:07 2017 +0200
  2084.  
  2085. arm: Use .section .rodata for non-elf, non-mach platforms as well
  2086.  
  2087. If targeting windows with armasm, gas-preprocessor can rewrite the
  2088. .section .rodata into the right construct for that platform.
  2089.  
  2090. commit 9bffbabfecf0bda066362a1b76b62c5085257e18 [revision 2790]
  2091. Author: Martin Storsjö <[email protected]>
  2092. AuthorDate: Fri Mar 24 11:33:41 2017 +0200
  2093. Commit: Henrik Gramner <[email protected]>
  2094. CommitDate: Sun May 21 22:41:07 2017 +0200
  2095.  
  2096. gas-preprocessor: Support conversion of additional arm instructions into thumb
  2097.  
  2098. Convert muls into mul+cmp.
  2099.  
  2100. Convert "and r0, sp, #xx" into "mov r0, sp", "and r0, r0, #xx".
  2101.  
  2102. Convert ldr with a too large shift into add+ldr. This only works in the
  2103. special case when the base register is the same as the target for the ldr.
  2104.  
  2105. commit 2e9bd88f27ed8f5f058e7e220070b7a15965cb8e [revision 2789]
  2106. Author: Martin Storsjö <[email protected]>
  2107. AuthorDate: Fri Mar 24 11:33:40 2017 +0200
  2108. Commit: Henrik Gramner <[email protected]>
  2109. CommitDate: Sun May 21 22:41:07 2017 +0200
  2110.  
  2111. arm: Explicitly declare using the .text segment in the function macro
  2112.  
  2113. This fixes one issue in building with MS armasm via gas-preprocessor.
  2114. Without the .text segment specification, the object files assembled
  2115. fine, but linking failed. (armasm source files don't get the text/code
  2116. segment implied automatically if nothing is specified.)
  2117.  
  2118. commit 64843af913e76fd7fb590e9227f678add96e8a3c [revision 2788]
  2119. Author: Martin Storsjö <[email protected]>
  2120. AuthorDate: Fri Mar 24 11:33:39 2017 +0200
  2121. Commit: Henrik Gramner <[email protected]>
  2122. CommitDate: Sun May 21 22:41:07 2017 +0200
  2123.  
  2124. osdep: Use the EXPAND macro on other cases of ALIGNED_ARRAY_EMU
  2125.  
  2126. EXPAND is already used on the other cases where ALIGNED_ARRAY_EMU
  2127. is used on all platforms (originally needed for ICL, later also
  2128. required by MSVC); apply the same change (originally from 21ba91ae)
  2129. for the cases that only are used on ARM.
  2130.  
  2131. This fixes use of ALIGNED_ARRAY_16 with MSVC when targeting ARM.
  2132.  
  2133. commit 757091fe3abd0af0f45d11f52b652f0be2fb76f5 [revision 2787]
  2134. Author: Martin Storsjö <[email protected]>
  2135. AuthorDate: Fri Mar 24 11:33:38 2017 +0200
  2136. Commit: Henrik Gramner <[email protected]>
  2137. CommitDate: Sun May 21 22:41:07 2017 +0200
  2138.  
  2139. Update to the latest version of gas-preprocessor.pl
  2140.  
  2141. From http://git.libav.org/?p=gas-preprocessor.git
  2142.  
  2143. This update contains changes from myself only.
  2144.  
  2145. commit d13705191cdcbcd10d87524dbb0c26ba998d8dcc [revision 2786]
  2146. Author: Martin Storsjö <[email protected]>
  2147. AuthorDate: Fri Mar 24 11:33:37 2017 +0200
  2148. Commit: Henrik Gramner <[email protected]>
  2149. CommitDate: Sun May 21 22:41:06 2017 +0200
  2150.  
  2151. arm: Skip using gas-preprocessor for iOS on arm as well
  2152.  
  2153. The few constructs that differ can easily be handled within the
  2154. source itself - tested to be working since at least Xcode 6.
  2155.  
  2156. commit 3a3cfe32416efa4f966c0586411148236e4703c1 [revision 2785]
  2157. Author: Martin Storsjö <[email protected]>
  2158. AuthorDate: Fri Mar 24 11:33:36 2017 +0200
  2159. Commit: Henrik Gramner <[email protected]>
  2160. CommitDate: Sun May 21 22:41:06 2017 +0200
  2161.  
  2162. arm: Use const macros in arm assembly where applicable
  2163.  
  2164. This unifies the source code style, and allows building the code
  2165. with clang without gas-preprocessor.
  2166.  
  2167. commit 1e92821c5a52c80ca4d1a9b6d038bec84be48b0a [revision 2784]
  2168. Author: Martin Storsjö <[email protected]>
  2169. AuthorDate: Fri Mar 24 11:33:35 2017 +0200
  2170. Commit: Henrik Gramner <[email protected]>
  2171. CommitDate: Sun May 21 22:41:06 2017 +0200
  2172.  
  2173. arm: Use commas between all macro arguments in arm assembly
  2174.  
  2175. The clang built-in assembler requires proper commas between all macro
  2176. arguments. As long as gas-preprocessor is used when building with clang,
  2177. this isn't an issue.
  2178.  
  2179. commit a84e6a486b991bffb2cc9f86b6e236978d251d2c [revision 2783]
  2180. Author: Martin Storsjö <[email protected]>
  2181. AuthorDate: Fri Mar 24 11:33:34 2017 +0200
  2182. Commit: Henrik Gramner <[email protected]>
  2183. CommitDate: Sun May 21 22:41:06 2017 +0200
  2184.  
  2185. aarch64: Skip invoking gas-preprocessor for iOS
  2186.  
  2187. Clang can handle all the constructs used there these days, working
  2188. since Xcode 6 at least.
  2189.  
  2190. commit 535fd2ec9985b9874d6ed23904404d0d2f5d40d6 [revision 2782]
  2191. Author: Martin Storsjö <[email protected]>
  2192. AuthorDate: Fri Mar 24 11:33:33 2017 +0200
  2193. Commit: Henrik Gramner <[email protected]>
  2194. CommitDate: Sun May 21 22:41:06 2017 +0200
  2195.  
  2196. aarch64: Use the const macro in the aarch64 checkasm assembly source
  2197.  
  2198. This fixes building the source with clang for iOS without gas-preprocessor.
  2199.  
  2200. commit bec87ba69421572282e473cf8f2e11c77285ed88 [revision 2781]
  2201. Author: Henrik Gramner <[email protected]>
  2202. AuthorDate: Wed Apr 12 23:26:32 2017 +0200
  2203. Commit: Henrik Gramner <[email protected]>
  2204. CommitDate: Sun May 21 22:41:06 2017 +0200
  2205.  
  2206. Windows: Add support for MSVC compilation with WSL
  2207.  
  2208. In Windows 10 version 1703 (Creators Update) WSL supports calling native
  2209. Windows binaries from the Bash shell, but it requires using full file
  2210. names including extension, e.g. `cl.exe` instead of `cl`.
  2211.  
  2212. We also don't have access to `cygpath`, so use a simple regex for
  2213. converting the dependencies to Unix paths that `make` can understand.
  2214.  
  2215. commit 43e9a6157752c2a3c2cc6c6a7fa13da72033d1dd [revision 2780]
  2216. Author: Henrik Gramner <[email protected]>
  2217. AuthorDate: Sun Jan 29 22:58:24 2017 +0100
  2218. Commit: Henrik Gramner <[email protected]>
  2219. CommitDate: Sun May 21 22:41:06 2017 +0200
  2220.  
  2221. cli: Improve the --fullhelp raw demuxer input-csp listing
  2222.  
  2223. Use the same logic for indentation as the lavf demuxer.
  2224.  
  2225. commit 3538df12688fc4408f585c4e65ee92d5a4737b2c [revision 2779]
  2226. Author: Anton Mitrofanov <[email protected]>
  2227. AuthorDate: Sat May 20 21:17:59 2017 +0300
  2228. Commit: Henrik Gramner <[email protected]>
  2229. CommitDate: Sun May 21 17:11:06 2017 +0200
  2230.  
  2231. x86inc: Remove argument from WIN64_RESTORE_XMM
  2232.  
  2233. The use of rsp was pretty much hardcoded there and probably didn't work
  2234. otherwise with stack_size > 0.
  2235.  
  2236. commit e4b0974a4ea3a727f6cc8941e9accf7ef3ba0637 [revision 2778]
  2237. Author: Henrik Gramner <[email protected]>
  2238. AuthorDate: Sat Apr 22 20:30:35 2017 +0200
  2239. Commit: Henrik Gramner <[email protected]>
  2240. CommitDate: Fri May 19 16:12:15 2017 +0200
  2241.  
  2242. x86inc: Prefer r14/r15 over r12/r13 on x86-64
  2243.  
  2244. Due to a peculiarity in the ModR/M addressing encoding, the r12 and r13
  2245. registers sometimes requires an additional byte when used as a base register.
  2246.  
  2247. r14 and r15 doesn't have that issue, so prefer using them.
  2248.  
  2249. commit 46a489b5e21cae3b4fea5d41cc285dcaf79d19e3 [revision 2777]
  2250. Author: Henrik Gramner <[email protected]>
  2251. AuthorDate: Thu Apr 20 19:16:51 2017 +0200
  2252. Commit: Henrik Gramner <[email protected]>
  2253. CommitDate: Fri May 19 16:12:15 2017 +0200
  2254.  
  2255. x86inc: Make REP_RET identical to RET in SSSE3+ functions
  2256.  
  2257. There's no point in emitting a rep prefix before ret on modern CPUs.
  2258.  
  2259. commit 50a9dd78263191474c948d53e837348abd0bf316 [revision 2776]
  2260. Author: Henrik Gramner <[email protected]>
  2261. AuthorDate: Wed Mar 29 16:43:57 2017 +0200
  2262. Commit: Henrik Gramner <[email protected]>
  2263. CommitDate: Fri May 19 16:12:15 2017 +0200
  2264.  
  2265. x86inc: Fix call with memory operands
  2266.  
  2267. We overload the `call` instruction with a macro, but it would misbehave when
  2268. the macro argument wasn't a valid identifier. Fix it by explicitly checking
  2269. if the argument is an identifier.
  2270.  
  2271. commit d13b4c3a9574cd2fbd5407c7dfc58eeff72d2080 [revision 2775]
  2272. Author: Henrik Gramner <[email protected]>
  2273. AuthorDate: Sun Jan 29 16:41:33 2017 +0100
  2274. Commit: Henrik Gramner <[email protected]>
  2275. CommitDate: Fri May 19 16:12:15 2017 +0200
  2276.  
  2277. osdep: Rework alignment macros
  2278.  
  2279. Drop ALIGNED_N and ALIGNED_ARRAY_N in favor of using explicit alignment.
  2280.  
  2281. This will allow us to increase the native alignment without unnecessarily
  2282. increasing the alignment of everything that's currently 32-byte aligned.
  2283.  
  2284. commit 5840e200a0f1869a0596c5ed75c76f4d3221dd68 [revision 2774]
  2285. Author: Vittorio Giovara <[email protected]>
  2286. AuthorDate: Mon Jan 30 22:14:57 2017 +0100
  2287. Commit: Henrik Gramner <[email protected]>
  2288. CommitDate: Fri May 19 16:12:15 2017 +0200
  2289.  
  2290. Move cabac_block_residual function declarations
  2291.  
  2292. commit a2d2621cc5741414b1f1adfbc08f19f1cc763847 [revision 2773]
  2293. Author: Vittorio Giovara <[email protected]>
  2294. AuthorDate: Mon Jan 30 22:14:59 2017 +0100
  2295. Commit: Henrik Gramner <[email protected]>
  2296. CommitDate: Fri May 19 16:12:15 2017 +0200
  2297.  
  2298. Recursively delete conftest files
  2299.  
  2300. On OS X, one of the conftest files might be a directory named `conftest.dSYM`.
  2301.  
  2302. commit 988ce459433fd3f978d632e8fc0ef9c19c94a6a1 [revision 2772]
  2303. Author: Vittorio Giovara <[email protected]>
  2304. AuthorDate: Mon Jan 30 22:14:56 2017 +0100
  2305. Commit: Henrik Gramner <[email protected]>
  2306. CommitDate: Fri May 19 16:12:15 2017 +0200
  2307.  
  2308. Drop unused function declarations
  2309.  
  2310. commit fb3f97833cbe3305eb613633e604f424d6d2d096 [revision 2771]
  2311. Author: Vittorio Giovara <[email protected]>
  2312. AuthorDate: Fri Jan 27 18:06:39 2017 +0100
  2313. Commit: Henrik Gramner <[email protected]>
  2314. CommitDate: Fri May 19 16:12:14 2017 +0200
  2315.  
  2316. x86: Adjust cache64_ssse3 function suffixes
  2317.  
  2318. Makes those function names more consistent with other similar functions.
  2319.  
  2320. commit a77f3917cc6ba5e1d3c20ca649d4114217976d53 [revision 2770]
  2321. Author: Vittorio Giovara <[email protected]>
  2322. AuthorDate: Fri Jan 27 16:21:16 2017 +0100
  2323. Commit: Henrik Gramner <[email protected]>
  2324. CommitDate: Fri May 19 16:12:14 2017 +0200
  2325.  
  2326. mc: Mark a function only used within the file as static
  2327.  
  2328. commit 0ca36bfa3d2bf272da88b1df5abfc0406662989a [revision 2769]
  2329. Author: Vittorio Giovara <[email protected]>
  2330. AuthorDate: Fri Jan 27 16:21:15 2017 +0100
  2331. Commit: Henrik Gramner <[email protected]>
  2332. CommitDate: Fri May 19 16:12:14 2017 +0200
  2333.  
  2334. ppc: Drop two unused static functions
  2335.  
  2336. commit d32d7bf1c6923a42cbd5ac2fd540ecbb009ba681 [revision 2768]
  2337. Author: Henrik Gramner <[email protected]>
  2338. AuthorDate: Fri May 19 16:08:34 2017 +0200
  2339. Commit: Henrik Gramner <[email protected]>
  2340. CommitDate: Fri May 19 16:12:14 2017 +0200
  2341.  
  2342. cli: Verify that yuv/y4m input has at least one frame of data
  2343.  
  2344. Prevents a SIGBUS crash caused by attempting to access a memory-mapped
  2345. region beyond the end of the input file.
  2346.  
  2347. commit 959e869c20ea151917695930d9ad0a7a9a2f90c5 [revision 2767]
  2348. Author: Kaustubh Raste <[email protected]>
  2349. AuthorDate: Fri Apr 14 15:29:31 2017 +0530
  2350. Commit: Henrik Gramner <[email protected]>
  2351. CommitDate: Fri Apr 14 14:52:53 2017 +0200
  2352.  
  2353. mips: Fix out-of-tree build
  2354.  
  2355. Signed-off-by: Kaustubh Raste <[email protected]>
  2356.  
  2357. commit d6eb2c9630d40a2765d5092f87637f4e4d084ed1 [revision 2766]
  2358. Author: Henrik Gramner <[email protected]>
  2359. AuthorDate: Sat Mar 25 00:02:11 2017 +0100
  2360. Commit: Henrik Gramner <[email protected]>
  2361. CommitDate: Wed Mar 29 17:13:47 2017 +0200
  2362.  
  2363. checkasm: Fix load_deinterleave_chroma_fdec test
  2364.  
  2365. The function only writes to parts of the destination buffer but the test
  2366. verifies the content of the entire buffer. The problem is that some earlier
  2367. IDCT functions clobbers the same part of the buffer with garbage when
  2368. benchmarked which would incorrectly cause test failures.
  2369.  
  2370. Fix this by explicitly zeroing the buffers beforehand.
  2371.  
  2372. commit a472b60daae0cac17d91ddf62ad4f474ded63e5b [revision 2765]
  2373. Author: Henrik Gramner <[email protected]>
  2374. AuthorDate: Fri Mar 24 22:27:42 2017 +0100
  2375. Commit: Henrik Gramner <[email protected]>
  2376. CommitDate: Wed Mar 29 17:13:47 2017 +0200
  2377.  
  2378. checkasm: Fix compilation on hardened x86-64 ELF systems
  2379.  
  2380. Normal PC-relative relocations cannot be used for resolving the address of
  2381. external symbols on systems where ASLR results in the offset being larger
  2382. than 32 bits. We are required to to go through the PLT instead.
  2383.  
  2384. commit 469ad705b1064207b6b1068d1e25a0a591021007 [revision 2764]
  2385. Author: Martin Storsjö <[email protected]>
  2386. AuthorDate: Thu Mar 23 15:05:38 2017 +0200
  2387. Commit: Henrik Gramner <[email protected]>
  2388. CommitDate: Thu Mar 23 22:55:18 2017 +0100
  2389.  
  2390. aarch64: Fix building checkasm for iOS
  2391.  
  2392. On iOS, symbols are prefixed - this prefix gets added by the X()
  2393. macro.
  2394.  
  2395. commit 93340ca300e7ce66f49e41b7c2ef4a0492a7e57c [revision 2763]
  2396. Author: Martin Storsjö <[email protected]>
  2397. AuthorDate: Thu Mar 23 15:05:37 2017 +0200
  2398. Commit: Henrik Gramner <[email protected]>
  2399. CommitDate: Thu Mar 23 22:52:09 2017 +0100
  2400.  
  2401. configure: Always enable PIC in aarch64 assembly for apple platforms
  2402.  
  2403. This is similar to what we do for 32-bit ARM assembly as well.
  2404.  
  2405. Fixes linker errors such as `ld: Absolute addressing not allowed in
  2406. arm64 code but used in '_x264_cabac_encode_terminal_asm' referencing
  2407. '_x264_cabac_range_lps' for architecture arm64`.
  2408.  
  2409. commit 90a61ec76424778c050524f682a33f115024be96 [revision 2762]
  2410. Author: Alexandra Hájková <[email protected]>
  2411. AuthorDate: Mon Dec 5 10:28:53 2016 +0000
  2412. Commit: Henrik Gramner <[email protected]>
  2413. CommitDate: Mon Jan 23 22:46:04 2017 +0100
  2414.  
  2415. ppc: AltiVec plane_copy_deinterleave
  2416.  
  2417. commit bd6b66dbf9fcf67b7ebb23e4e9249083191fb984 [revision 2761]
  2418. Author: Alexandra Hájková <[email protected]>
  2419. AuthorDate: Mon Jan 2 12:56:48 2017 +0000
  2420. Commit: Henrik Gramner <[email protected]>
  2421. CommitDate: Mon Jan 23 22:46:03 2017 +0100
  2422.  
  2423. ppc: AltiVec plane_copy_deinterleave_v210
  2424.  
  2425. commit 00f1670087db1b025a8088289de8938bf88a0d8b [revision 2760]
  2426. Author: Alexandra Hájková <[email protected]>
  2427. AuthorDate: Wed Dec 7 19:48:02 2016 +0000
  2428. Commit: Henrik Gramner <[email protected]>
  2429. CommitDate: Mon Jan 23 22:46:03 2017 +0100
  2430.  
  2431. ppc: AltiVec plane_copy_deinterleave_rgb
  2432.  
  2433. Also add some missing vector types in ppccommon.h
  2434.  
  2435. commit 5e1ed367d725f895eeadf358861ab52521a420d3 [revision 2759]
  2436. Author: Vittorio Giovara <[email protected]>
  2437. AuthorDate: Thu Jan 19 17:43:57 2017 +0100
  2438. Commit: Henrik Gramner <[email protected]>
  2439. CommitDate: Mon Jan 23 22:46:03 2017 +0100
  2440.  
  2441. ppc: Adjust AltiVec function suffix
  2442.  
  2443. Architecture should always be the last element.
  2444.  
  2445. commit 28ebb95d92278069b80ee729eb1884fe0981c6ae [revision 2758]
  2446. Author: Vittorio Giovara <[email protected]>
  2447. AuthorDate: Mon Jan 9 22:28:20 2017 +0100
  2448. Commit: Henrik Gramner <[email protected]>
  2449. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2450.  
  2451. Move the x264_mdate() declaration to the appropriate header
  2452.  
  2453. commit 1d2420981aa004f051a0869c005776084f7d2a44 [revision 2757]
  2454. Author: Vittorio Giovara <[email protected]>
  2455. AuthorDate: Tue Jan 17 17:04:19 2017 +0100
  2456. Commit: Henrik Gramner <[email protected]>
  2457. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2458.  
  2459. arm/aarch64: Correctly prefix integral function symbols
  2460.  
  2461. commit 4c4c495d58dbdea46a23947e4f202fc3b82fb891 [revision 2756]
  2462. Author: Anton Mitrofanov <[email protected]>
  2463. AuthorDate: Fri Jan 13 14:57:51 2017 +0100
  2464. Commit: Henrik Gramner <[email protected]>
  2465. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2466.  
  2467. x86: Avoid using hardcoded function symbol prefixes
  2468.  
  2469. commit 2524fc3164d9f00b393d4254d2c5ea8f3b9d43b0 [revision 2755]
  2470. Author: Henrik Gramner <[email protected]>
  2471. AuthorDate: Wed Jan 18 21:57:14 2017 +0100
  2472. Commit: Henrik Gramner <[email protected]>
  2473. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2474.  
  2475. x86: AVX2 high bit-depth load_deinterleave_chroma
  2476.  
  2477. load_deinterleave_chroma_fenc: 50% faster than AVX
  2478. load_deinterleave_chroma_fdec: 25% faster than AVX
  2479.  
  2480. commit cce50082129d3c92bd41bc0afc5a8c8d93084c9c [revision 2754]
  2481. Author: Henrik Gramner <[email protected]>
  2482. AuthorDate: Wed Jan 18 21:46:55 2017 +0100
  2483. Commit: Henrik Gramner <[email protected]>
  2484. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2485.  
  2486. x86: AVX2 load_deinterleave_chroma_fenc
  2487.  
  2488. 20% faster than SSSE3.
  2489.  
  2490. commit c22c10ddb21e9f5af1da83d37122e6f7388e1342 [revision 2753]
  2491. Author: Henrik Gramner <[email protected]>
  2492. AuthorDate: Tue Jan 17 21:59:47 2017 +0100
  2493. Commit: Henrik Gramner <[email protected]>
  2494. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2495.  
  2496. x86: AVX2 plane_copy_deinterleave
  2497.  
  2498. 50% faster than SSSE3 in 8-bit.
  2499. 25% faster than AVX in high bit-depth.
  2500.  
  2501. Also drop the MMX versions of deinterleave functions in favor of SSE2.
  2502.  
  2503. commit f4890275ca6523dfe5b4ae60279ae8597d9dbd4b [revision 2752]
  2504. Author: Henrik Gramner <[email protected]>
  2505. AuthorDate: Thu Jan 12 22:16:53 2017 +0100
  2506. Commit: Henrik Gramner <[email protected]>
  2507. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2508.  
  2509. x86: AVX2 plane_copy_deinterleave_rgb
  2510.  
  2511. Around 15% faster than SSSE3.
  2512.  
  2513. commit da71b556730c8eb6c12a0d6950a221a4e4a99ca6 [revision 2751]
  2514. Author: Henrik Gramner <[email protected]>
  2515. AuthorDate: Thu Jan 12 21:36:28 2017 +0100
  2516. Commit: Henrik Gramner <[email protected]>
  2517. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2518.  
  2519. x86: Faster plane_copy_deinterleave_rgb_sse2
  2520.  
  2521. 50% faster than the previous SSE2 function.
  2522.  
  2523. commit 3c7bf52c5b0a849458a45b5628ed1cc4b898da5f [revision 2750]
  2524. Author: Henrik Gramner <[email protected]>
  2525. AuthorDate: Sun Jan 15 14:52:29 2017 +0100
  2526. Commit: Henrik Gramner <[email protected]>
  2527. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2528.  
  2529. x86util: Reduce code size of high bit-depth AVX LOAD_DIFF
  2530.  
  2531. AVX supports unaligned memory operands which makes the SATD code a bit denser.
  2532.  
  2533. commit c7a2e327bebd2b863c2620b6962fa18ab681e5dd [revision 2749]
  2534. Author: Henrik Gramner <[email protected]>
  2535. AuthorDate: Sun Jan 1 19:10:10 2017 +0100
  2536. Commit: Henrik Gramner <[email protected]>
  2537. CommitDate: Sat Jan 21 14:10:37 2017 +0100
  2538.  
  2539. Bump dates to 2017
  2540.  
  2541. commit 97eaef2ab82a46d13ea5e00270712d6475fbe42b [revision 2748]
  2542. Author: Alexandra Hájková <[email protected]>
  2543. AuthorDate: Sat Jan 21 12:34:49 2017 +0000
  2544. Commit: Henrik Gramner <[email protected]>
  2545. CommitDate: Sat Jan 21 14:10:36 2017 +0100
  2546.  
  2547. ppc: Fix the pre-VSX vec_vsx_st() fallback macro
  2548.  
  2549. It would previously only work correctly with 8-bit data types.
  2550.  
  2551. Fixes compilation with --disable-vsx.
  2552.  
  2553. commit 2ebe09a4f583d108c6ec1caf70b2a7a289a8820d [revision 2747]
  2554. Author: Alexandra Hájková <[email protected]>
  2555. AuthorDate: Wed Jan 18 09:13:39 2017 +0000
  2556. Commit: Henrik Gramner <[email protected]>
  2557. CommitDate: Wed Jan 18 16:38:16 2017 +0100
  2558.  
  2559. Fix plane_copy_deinterleave_v210 on big-endian
  2560.  
  2561. commit 79288d90471e246584d19054bdb5381982114126 [revision 2746]
  2562. Author: Alexandra Hájková <[email protected]>
  2563. AuthorDate: Wed Dec 21 13:13:43 2016 +0000
  2564. Commit: Henrik Gramner <[email protected]>
  2565. CommitDate: Wed Jan 18 16:35:36 2017 +0100
  2566.  
  2567. ppc: Avoid instantiating unused plane_copy functions
  2568.  
  2569. Those functions are currently only used in 8-bit mode and results in
  2570. warnings in other bit depths.
  2571.  
  2572. commit 2ebdb90bd32c3d1618b1c5b360bff750b82b1d0b [revision 2745]
  2573. Author: Martin Storsjö <[email protected]>
  2574. AuthorDate: Tue Dec 27 00:22:48 2016 +0200
  2575. Commit: Henrik Gramner <[email protected]>
  2576. CommitDate: Thu Dec 29 23:06:12 2016 +0100
  2577.  
  2578. arm: Load mb_y properly in mbtree_propagate_list_internal_neon
  2579.  
  2580. The previous version, attempting to load two stack parameters at once,
  2581. only would have worked if they were interpreted and loaded as 32 bit
  2582. elements, not when loading them as 16 bit elements.
  2583.  
  2584. commit b97ae0644f16bad2e2c9c9181264a946769a0aa0 [revision 2744]
  2585. Author: Anton Mitrofanov <[email protected]>
  2586. AuthorDate: Mon Oct 31 14:39:52 2016 +0300
  2587. Commit: Henrik Gramner <[email protected]>
  2588. CommitDate: Thu Dec 1 18:00:36 2016 +0100
  2589.  
  2590. analyse: Fix lambda table values
  2591.  
  2592. commit b2b39dae0bd891c8d150b4f4c3a2a24d8d6c1431 [revision 2743]
  2593. Author: Anton Mitrofanov <[email protected]>
  2594. AuthorDate: Sat Nov 26 15:30:58 2016 +0300
  2595. Commit: Henrik Gramner <[email protected]>
  2596. CommitDate: Thu Dec 1 18:00:07 2016 +0100
  2597.  
  2598. Cosmetics
  2599.  
  2600. Also make x264_weighted_reference_duplicate() static.
  2601.  
  2602. commit 9c82d2b65534e477c972b811a4dd5004d0dd262e [revision 2742]
  2603. Author: Alexandra Hájková <[email protected]>
  2604. AuthorDate: Mon Nov 28 14:04:10 2016 +0000
  2605. Commit: Henrik Gramner <[email protected]>
  2606. CommitDate: Thu Dec 1 17:59:59 2016 +0100
  2607.  
  2608. ppc: AltiVec store_interleave_chroma
  2609.  
  2610. commit ea1fee272b20e1bcff2a862ea9a29e151c9136a9 [revision 2741]
  2611. Author: Alexandra Hájková <[email protected]>
  2612. AuthorDate: Mon Nov 28 10:51:54 2016 +0000
  2613. Commit: Henrik Gramner <[email protected]>
  2614. CommitDate: Thu Dec 1 17:59:43 2016 +0100
  2615.  
  2616. ppc: AltiVec plane_copy_interleave
  2617.  
  2618. commit 42348a8e664b091203a05d3e15555b5085afcac1 [revision 2740]
  2619. Author: Alexandra Hájková <[email protected]>
  2620. AuthorDate: Sat Nov 26 20:03:34 2016 +0000
  2621. Commit: Henrik Gramner <[email protected]>
  2622. CommitDate: Thu Dec 1 17:59:22 2016 +0100
  2623.  
  2624. ppc: AltiVec plane_copy_swap
  2625.  
  2626. commit 2610019af8bfb8e71f813cd2188b9eccbc287c59 [revision 2739]
  2627. Author: Alexandra Hájková <[email protected]>
  2628. AuthorDate: Wed Nov 23 20:53:51 2016 +0100
  2629. Commit: Henrik Gramner <[email protected]>
  2630. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2631.  
  2632. ppc: AltiVec zigzag_interleave_8x8_cavlc
  2633.  
  2634. commit 25e4e06fe8151f627a953fbd2bd39302436bf689 [revision 2738]
  2635. Author: Alexandra Hájková <[email protected]>
  2636. AuthorDate: Wed Nov 23 20:53:50 2016 +0100
  2637. Commit: Henrik Gramner <[email protected]>
  2638. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2639.  
  2640. ppc: AltiVec zigzag_scan_8x8_frame
  2641.  
  2642. commit 99863c665a6d4ec58b7fcc4a8a791e9c8f35a86e [revision 2737]
  2643. Author: Alexandra Hájková <[email protected]>
  2644. AuthorDate: Mon Nov 14 15:06:06 2016 +0100
  2645. Commit: Henrik Gramner <[email protected]>
  2646. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2647.  
  2648. ppc: AltiVec sub8x8_dct_dc
  2649.  
  2650. commit 42cb0a6813714b5380e23871a155e3820846d991 [revision 2736]
  2651. Author: Alexandra Hájková <[email protected]>
  2652. AuthorDate: Mon Nov 14 15:06:05 2016 +0100
  2653. Commit: Henrik Gramner <[email protected]>
  2654. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2655.  
  2656. ppc: AltiVec add8x8_idct_dc
  2657.  
  2658. commit 983acc911543453449a65bd02bbdff4c8cfe8e6a [revision 2735]
  2659. Author: Martin Storsjö <[email protected]>
  2660. AuthorDate: Wed Nov 16 10:57:31 2016 +0200
  2661. Commit: Henrik Gramner <[email protected]>
  2662. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2663.  
  2664. checkasm: aarch64: Add filler args to make sure all parameters are passed on the stack
  2665.  
  2666. This, combined with clobbering the stack space prior to the call,
  2667. increases the chances of finding cases where 32 bit parameters
  2668. are erroneously treated as 64 bit.
  2669.  
  2670. commit 8ada354c9b5d72356c34c9ae3f787a6df4d61506 [revision 2734]
  2671. Author: Martin Storsjö <[email protected]>
  2672. AuthorDate: Wed Nov 16 10:57:30 2016 +0200
  2673. Commit: Henrik Gramner <[email protected]>
  2674. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2675.  
  2676. checkasm: aarch64: Clobber the stack before calling functions
  2677.  
  2678. commit 62d604ac6dddbf553c1ff2432d899b61cc50d95a [revision 2733]
  2679. Author: Alexandra Hájková <[email protected]>
  2680. AuthorDate: Tue Nov 1 23:16:17 2016 +0100
  2681. Commit: Henrik Gramner <[email protected]>
  2682. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2683.  
  2684. ppc: Use vec_vsx_ld instead of VEC_LOAD/STORE macros
  2685.  
  2686. Remove VEC_LOAD*, some of VEC_STORE* macros, some PREP* macros and
  2687. VEC_DIFF_H_OFFSET macro.
  2688.  
  2689. Make sure the functions do not use deprected primitives.
  2690.  
  2691. commit 16142d8ee2a974060ecbad0f495b5a5c6516a75e [revision 2732]
  2692. Author: Luca Barbato <[email protected]>
  2693. AuthorDate: Tue Nov 1 23:16:16 2016 +0100
  2694. Commit: Henrik Gramner <[email protected]>
  2695. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2696.  
  2697. ppc: Provide fallbacks for older architectures
  2698.  
  2699. commit 2b741f81e51f92d053d87a49f59ff1026553a0f6 [revision 2731]
  2700. Author: Luca Barbato <[email protected]>
  2701. AuthorDate: Tue Nov 1 23:16:14 2016 +0100
  2702. Commit: Henrik Gramner <[email protected]>
  2703. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2704.  
  2705. ppc: Add VSX support to configure
  2706.  
  2707. commit 1f7518182e3204cb14e87baffb0150a848167ddc [revision 2730]
  2708. Author: Luca Barbato <[email protected]>
  2709. AuthorDate: Tue Nov 1 23:16:13 2016 +0100
  2710. Commit: Henrik Gramner <[email protected]>
  2711. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2712.  
  2713. ppc: Manually unroll the horizontal prediction loop
  2714.  
  2715. Doubles the speedup from the function (from being slower to be over
  2716. twice as fast than C).
  2717.  
  2718. commit 0706ddb1df88d716cf73decba4d82b953011760c [revision 2729]
  2719. Author: Henrik Gramner <[email protected]>
  2720. AuthorDate: Sat Oct 8 17:20:18 2016 +0200
  2721. Commit: Henrik Gramner <[email protected]>
  2722. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2723.  
  2724. x86inc: Avoid using eax/rax for storing the stack pointer
  2725.  
  2726. When allocating stack space with an alignment requirement that is larger
  2727. than the current stack alignment we need to store a copy of the original
  2728. stack pointer in order to be able to restore it later.
  2729.  
  2730. If we chose to use another register for this purpose we should not pick
  2731. eax/rax since it can be overwritten as a return value.
  2732.  
  2733. commit 4d5c8b01a48f72f9c40651e92c39294326a0863f [revision 2728]
  2734. Author: Henrik Gramner <[email protected]>
  2735. AuthorDate: Thu Dec 1 16:05:16 2016 +0100
  2736. Commit: Henrik Gramner <[email protected]>
  2737. CommitDate: Thu Dec 1 16:10:37 2016 +0100
  2738.  
  2739. Show the correct settings for --preset slow in --fullhelp
  2740.  
  2741. The slow preset was recently adjusted but we forgot to update the
  2742. corresponding --fullhelp message to reflect the change.
  2743.  
  2744. commit c996ed202e2d17d1d8ae42c42d0707e51c29bb93 [revision 2727]
  2745. Author: Martin Storsjö <[email protected]>
  2746. AuthorDate: Mon Nov 14 23:54:51 2016 +0200
  2747. Commit: Henrik Gramner <[email protected]>
  2748. CommitDate: Mon Nov 21 19:52:24 2016 +0100
  2749.  
  2750. checkasm: arm/aarch64: Fix the amount of space reserved for stack parameters
  2751.  
  2752. Even if MAX_ARGS - 2 (for arm) or MAX_ARGS - 6 (for aarch64) parameters
  2753. are passed on the stack to checkasm_checked_call, we actually only
  2754. need to store MAX_ARGS - 4 (for arm) or MAX_ARGS - 8 (for aarch64)
  2755. parameters on the stack when calling the tested function.
  2756.  
  2757. commit cd15b354a887943d525e6fd8096ad4b75692d2b2 [revision 2726]
  2758. Author: Janne Grunau <[email protected]>
  2759. AuthorDate: Mon Nov 14 23:54:50 2016 +0200
  2760. Commit: Henrik Gramner <[email protected]>
  2761. CommitDate: Mon Nov 21 19:52:24 2016 +0100
  2762.  
  2763. checkasm: arm: preserve the stack alignment in x264_checkasm_checked_call
  2764.  
  2765. The stack used by x264_checkasm_checked_call_neon was a multiple of 4
  2766. when the checked function is called. AAPCS requires a double word (8 byte)
  2767. aligned stack public interfaces. Since both calls are public interfaces
  2768. the stack is misaligned when the checked is called.
  2769.  
  2770. This can cause issues if code called within this (which includes
  2771. the C implementations) relies on the stack alignment.
  2772.  
  2773. commit 834e1b11e174f2694a4c81b4922c0c5f8778796a [revision 2725]
  2774. Author: Martin Storsjö <[email protected]>
  2775. AuthorDate: Wed Nov 16 10:56:14 2016 +0200
  2776. Commit: Henrik Gramner <[email protected]>
  2777. CommitDate: Mon Nov 21 19:52:24 2016 +0100
  2778.  
  2779. arm: Don't use vcmp.f64 for testing for an all-zeros register
  2780.  
  2781. On iOS, vcmp.f64 can behave as if the register was zero, if the
  2782. register (interpreted as a f64), was a denormal number.
  2783.  
  2784. The vcmp.f64 (and other VFP instructions) will trap to the kernel
  2785. (which is supposed to implement the FP operation, which it apparently
  2786. doesn't do properly on iOS) if the value is a denormal. If this happens,
  2787. the whole comparison ends up way more costly.
  2788.  
  2789. commit a91e95fca2222ac0731e987a07f4b11c670f4556 [revision 2724]
  2790. Author: Janne Grunau <[email protected]>
  2791. AuthorDate: Wed Nov 16 10:49:14 2016 +0200
  2792. Commit: Henrik Gramner <[email protected]>
  2793. CommitDate: Mon Nov 21 19:52:24 2016 +0100
  2794.  
  2795. aarch64: Clear the upper half of int parameters in x264_plane_copy_core_neon
  2796.  
  2797. commit 1eab3b402e1d7729da295024fa7eec8b09e30c20 [revision 2723]
  2798. Author: Luca Barbato <[email protected]>
  2799. AuthorDate: Tue Nov 1 23:16:18 2016 +0100
  2800. Commit: Henrik Gramner <[email protected]>
  2801. CommitDate: Mon Nov 21 19:52:24 2016 +0100
  2802.  
  2803. ppc: Fix hadamard for little-endian
  2804.  
  2805. Extending to 16-bit works with flipped bytes.
  2806.  
  2807. commit 75918e1849e1286885bfcfb0c348de885a702fb3 [revision 2722]
  2808. Author: Anton Mitrofanov <[email protected]>
  2809. AuthorDate: Thu Sep 22 00:17:48 2016 +0300
  2810. Commit: Anton Mitrofanov <[email protected]>
  2811. CommitDate: Thu Sep 22 00:18:16 2016 +0300
  2812.  
  2813. Correctly signal max_dec_frame_buffering with --keyint 1
  2814.  
  2815. According to E.2.1 it is inferred to be equal to 0 only if profile_idc is equal
  2816. to 44, 86, 100, 110, 122, or 244 and constraint_set3_flag is equal to 1.
  2817.  
  2818. commit 72d53ab2ac7af24597a824e868f2ef363a22f5d4 [revision 2721]
  2819. Author: Henrik Gramner <[email protected]>
  2820. AuthorDate: Sat Sep 17 21:41:52 2016 +0200
  2821. Commit: Anton Mitrofanov <[email protected]>
  2822. CommitDate: Tue Sep 20 21:02:01 2016 +0300
  2823.  
  2824. x86: Faster pixel_ssim_4x4x2_core
  2825.  
  2826. commit 8c07263ad9218bdc3e0f5b84d578968513885df7 [revision 2720]
  2827. Author: Henrik Gramner <[email protected]>
  2828. AuthorDate: Sat Sep 17 21:14:35 2016 +0200
  2829. Commit: Anton Mitrofanov <[email protected]>
  2830. CommitDate: Tue Sep 20 21:02:01 2016 +0300
  2831.  
  2832. x86: Deduplicate a constant in hpel_filter_c
  2833.  
  2834. commit 9521b278adb92081f052c1b7bfc4b95651d88b07 [revision 2719]
  2835. Author: Henrik Gramner <[email protected]>
  2836. AuthorDate: Sat Sep 17 14:45:08 2016 +0200
  2837. Commit: Anton Mitrofanov <[email protected]>
  2838. CommitDate: Tue Sep 20 21:02:00 2016 +0300
  2839.  
  2840. x86: Faster pixel_ssd_nv12
  2841.  
  2842. Also drop the MMX2 version to simplify things.
  2843.  
  2844. commit 75d0f9cc8770bc4f36785062116757d24eb44604 [revision 2718]
  2845. Author: Henrik Gramner <[email protected]>
  2846. AuthorDate: Sun Sep 11 15:32:54 2016 +0200
  2847. Commit: Anton Mitrofanov <[email protected]>
  2848. CommitDate: Tue Sep 20 21:01:59 2016 +0300
  2849.  
  2850. x86: SSE zigzag_scan_4x4_field
  2851.  
  2852. Replaces the MMX2 version, one cycle faster.
  2853.  
  2854. Also change the checkasm test to use the correct alignment macro.
  2855.  
  2856. commit 0ce77f9eb71051c9a6121ec12c2abaac99ee628a [revision 2717]
  2857. Author: Henrik Gramner <[email protected]>
  2858. AuthorDate: Wed Sep 7 19:27:31 2016 +0200
  2859. Commit: Anton Mitrofanov <[email protected]>
  2860. CommitDate: Tue Sep 20 21:01:59 2016 +0300
  2861.  
  2862. x86: AVX2 mbtree_propagate_list
  2863.  
  2864. SIMD part is around 25% faster than AVX on Haswell, around 7%
  2865. faster when including the runtime of the scalar C wrapper.
  2866.  
  2867. commit 0c36239a4826f6e5a3cb873aca1814e389a46e29 [revision 2716]
  2868. Author: Henrik Gramner <[email protected]>
  2869. AuthorDate: Wed Sep 7 19:26:42 2016 +0200
  2870. Commit: Anton Mitrofanov <[email protected]>
  2871. CommitDate: Tue Sep 20 21:01:58 2016 +0300
  2872.  
  2873. x86: Move predict_16x16_dc_left calculations to asm
  2874.  
  2875. 1-2 cycles faster and avoids some code duplication to decrease code size.
  2876.  
  2877. Also drop the MMX2 implementation in favor of SSE2 to simplify things.
  2878.  
  2879. commit 0cc8afd31212de013b26b10f58c608c9adcff2fc [revision 2715]
  2880. Author: Anton Mitrofanov <[email protected]>
  2881. AuthorDate: Thu Aug 18 19:00:48 2016 +0300
  2882. Commit: Anton Mitrofanov <[email protected]>
  2883. CommitDate: Tue Sep 20 21:01:57 2016 +0300
  2884.  
  2885. avs: support for AviSynth+ high bit-depth pixel formats
  2886.  
  2887. commit dc0fe73636d34baeb3a64918b52db64d2a9e83bb [revision 2714]
  2888. Author: Janne Grunau <[email protected]>
  2889. AuthorDate: Fri Aug 26 20:26:56 2016 +0300
  2890. Commit: Anton Mitrofanov <[email protected]>
  2891. CommitDate: Tue Sep 20 21:01:57 2016 +0300
  2892.  
  2893. aarch64: implement x264_plane_copy_swap_neon
  2894.  
  2895. plane_copy_swap_c: 27054
  2896. plane_copy_swap_neon: 4152
  2897.  
  2898. commit eaf2fc20c8579714a48523b7ab8c05373708a25f [revision 2713]
  2899. Author: Anton Mitrofanov <[email protected]>
  2900. AuthorDate: Thu Aug 18 22:14:22 2016 +0300
  2901. Commit: Anton Mitrofanov <[email protected]>
  2902. CommitDate: Tue Sep 20 21:01:56 2016 +0300
  2903.  
  2904. Various cosmetics of semicolon use
  2905.  
  2906. commit aae177c55141460f442de0572c4a434bf2ae20bc [revision 2712]
  2907. Author: Henrik Gramner <[email protected]>
  2908. AuthorDate: Thu Jul 28 21:58:40 2016 +0200
  2909. Commit: Anton Mitrofanov <[email protected]>
  2910. CommitDate: Tue Sep 20 21:01:55 2016 +0300
  2911.  
  2912. cli: Prefetch yuv/y4m input frames on Windows 8 and newer
  2913.  
  2914. Use PrefetchVirtualMemory() (if available) on memory-mapped input frames.
  2915.  
  2916. Significantly improves performance when the source file is not already
  2917. present in the OS page cache by asking the OS to bring in those pages from
  2918. disk using large, concurrent I/O requests.
  2919.  
  2920. Most beneficial on fast encoding settings. Up to 40% faster overall with
  2921. --preset ultrafast, and up to 20% faster overall with --preset veryfast.
  2922.  
  2923. This API was introduced in Windows 8, so call it conditionally. On older
  2924. Windows systems the previous behavior remains unchanged.
  2925.  
  2926. commit 4e5adb87070c82b937c03e0cc030eae3578c251d [revision 2711]
  2927. Author: Henrik Gramner <[email protected]>
  2928. AuthorDate: Thu Jul 28 19:34:04 2016 +0200
  2929. Commit: Anton Mitrofanov <[email protected]>
  2930. CommitDate: Tue Sep 20 21:01:54 2016 +0300
  2931.  
  2932. Adjust --preset slow
  2933.  
  2934. * Swap --me umh for --trellis 2. They have a similar effect on performance
  2935. but the latter gives slightly better results in most cases.
  2936. * Change --b-adapt from 2 to 1. Negligible difference in quality since the
  2937. b-adapt 1 improvements, but it's significantly faster.
  2938.  
  2939. Also remove a redundant assignment from veryfast (--me hex is set by default).
  2940.  
  2941. commit 1e4fb55a283ba90fef346033027af851f2a04468 [revision 2710]
  2942. Author: Henrik Gramner <[email protected]>
  2943. AuthorDate: Thu Jul 28 19:33:57 2016 +0200
  2944. Commit: Anton Mitrofanov <[email protected]>
  2945. CommitDate: Tue Sep 20 21:01:54 2016 +0300
  2946.  
  2947. ratecontrol_new: Simplify an expression in HRD timescale calculation
  2948.  
  2949. Also gets rid of a false positive static analyser integer division warning.
  2950.  
  2951. commit 17378b2028146fa54a1b2b90da62554935d9dcc2 [revision 2709]
  2952. Author: Henrik Gramner <[email protected]>
  2953. AuthorDate: Thu Jul 28 19:33:44 2016 +0200
  2954. Commit: Anton Mitrofanov <[email protected]>
  2955. CommitDate: Tue Sep 20 21:01:53 2016 +0300
  2956.  
  2957. gcc: Enable __sync_fetch_and_add() on x86-64
  2958.  
  2959. It was previously only enabled on 32-bit x86 for no reason, so 64-bit
  2960. systems had to use a mutex instead of a simple `lock xadd` instruction.
  2961.  
  2962. Note that this code is only used in some very specific configurations
  2963. involving sliced threads.
  2964.  
  2965. commit 86b71982e131eaa70125f8d0e725fcade9c4c677 [revision 2708]
  2966. Author: Anton Mitrofanov <[email protected]>
  2967. AuthorDate: Tue Sep 20 18:48:22 2016 +0300
  2968. Commit: Anton Mitrofanov <[email protected]>
  2969. CommitDate: Tue Sep 20 21:01:52 2016 +0300
  2970.  
  2971. mips: Fix high bit-depth compilation
  2972.  
  2973. commit 1ea3c682ca12c7f13ea6f82b42bdc40afcfda87f [revision 2707]
  2974. Author: Henrik Gramner <[email protected]>
  2975. AuthorDate: Sat Sep 17 15:53:59 2016 +0200
  2976. Commit: Henrik Gramner <[email protected]>
  2977. CommitDate: Sat Sep 17 16:41:30 2016 +0200
  2978.  
  2979. checkasm: Fix compilation on Windows with --disable-thread
  2980.  
  2981. commit 5caef139cf7d6b41a95ee9568625d36d1ae1c107 [revision 2706]
  2982. Author: Janne Grunau <[email protected]>
  2983. AuthorDate: Fri Aug 26 20:26:55 2016 +0300
  2984. Commit: Henrik Gramner <[email protected]>
  2985. CommitDate: Sat Sep 17 15:10:14 2016 +0200
  2986.  
  2987. arm/aarch64: use plane_copy wrapper macros
  2988.  
  2989. Move the macros to common/mc.h to share them across all architectures.
  2990. Fixes possible buffer overreads if the width of the user supplied frames
  2991. is not a multiple of 16.
  2992.  
  2993. Reported-by: Kirill Batuzov <[email protected]>
  2994.  
  2995. commit 3f5ed56d4105f68c01b86f94f41bb9bbefa3433b [revision 2705]
  2996. Author: Henrik Gramner <[email protected]>
  2997. AuthorDate: Sun Apr 3 17:28:33 2016 +0200
  2998. Commit: Henrik Gramner <[email protected]>
  2999. CommitDate: Mon Jun 13 22:07:00 2016 +0200
  3000.  
  3001. configure: Support specifying a custom pkg-config
  3002.  
  3003. commit 7c9c687d8062f72b3ec300de8997bdae8277a741 [revision 2704]
  3004. Author: Anton Mitrofanov <[email protected]>
  3005. AuthorDate: Wed Jun 8 22:46:17 2016 +0300
  3006. Commit: Henrik Gramner <[email protected]>
  3007. CommitDate: Mon Jun 13 22:07:00 2016 +0200
  3008.  
  3009. Add support for new VUI parameters
  3010.  
  3011. Support the new color primaries, transfer characteristics, and matrix
  3012. coefficients defined in the 2016-02 edition of the H.264 specification.
  3013.  
  3014. commit 92515e8ff73491ef8a44c85e0bee265ba5791070 [revision 2703]
  3015. Author: Henrik Gramner <[email protected]>
  3016. AuthorDate: Sun Apr 24 14:10:22 2016 +0200
  3017. Commit: Henrik Gramner <[email protected]>
  3018. CommitDate: Mon Jun 13 22:07:00 2016 +0200
  3019.  
  3020. configure: Add link-time optimization support
  3021.  
  3022. Enabled by using the --enable-lto configuration option.
  3023.  
  3024. May give a slight performance improvement in some cases, but it can
  3025. also reduce performance in other cases (largely compiler-dependant)
  3026. so don't enable it by default. It also makes compilation (and linking
  3027. in particular) a fair bit slower.
  3028.  
  3029. Note that some older versions of GNU binutils will incorrectly warn
  3030. about "memset used with constant zero length parameter" when linking
  3031. using LTO. This is due to a bug in binutils and can safely be ignored.
  3032.  
  3033. commit b6267e0ff770545de88dfb5d3f176ea73f453730 [revision 2702]
  3034. Author: Henrik Gramner <[email protected]>
  3035. AuthorDate: Sun Apr 24 13:32:43 2016 +0200
  3036. Commit: Henrik Gramner <[email protected]>
  3037. CommitDate: Mon Jun 13 22:07:00 2016 +0200
  3038.  
  3039. configure: Fix clang detection with versioned binaries
  3040.  
  3041. Correctly detect clang binaries that has the version number appended
  3042. as a suffix to the file name, e.g. `clang38`.
  3043.  
  3044. commit 14a58532fea2c5f9e7b93c918476d842091c4268 [revision 2701]
  3045. Author: Janne Grunau <[email protected]>
  3046. AuthorDate: Sun Apr 24 14:38:56 2016 +0200
  3047. Commit: Henrik Gramner <[email protected]>
  3048. CommitDate: Mon Jun 13 22:07:00 2016 +0200
  3049.  
  3050. arm: Add asm for mbtree fixed point conversion
  3051.  
  3052. 7-8 times faster on a cortex-a53 vs. gcc-5.3.
  3053.  
  3054. mbtree_fix8_pack_c: 44114
  3055. mbtree_fix8_pack_neon: 5805
  3056. mbtree_fix8_unpack_c: 38924
  3057. mbtree_fix8_unpack_neon: 4870
  3058.  
  3059. commit b6f189eb4c5646483f7901293944695167e71ed9 [revision 2700]
  3060. Author: Janne Grunau <[email protected]>
  3061. AuthorDate: Sun Apr 24 14:38:55 2016 +0200
  3062. Commit: Henrik Gramner <[email protected]>
  3063. CommitDate: Mon Jun 13 22:06:51 2016 +0200
  3064.  
  3065. aarch64: Add asm for mbtree fixed point conversion
  3066.  
  3067. pack is ~7 times faster and unpack is ~9 times faster on a cortex-a53
  3068. compared to gcc-5.3.
  3069.  
  3070. mbtree_fix8_pack_c: 41534
  3071. mbtree_fix8_pack_neon: 5766
  3072. mbtree_fix8_unpack_c: 44102
  3073. mbtree_fix8_unpack_neon: 4868
  3074.  
  3075. commit a5e06b9a435852f0125de4ecb198ad47340483fa [revision 2699]
  3076. Author: Anton Mitrofanov <[email protected]>
  3077. AuthorDate: Sun May 22 22:33:58 2016 +0300
  3078. Commit: Henrik Gramner <[email protected]>
  3079. CommitDate: Mon Jun 13 22:06:51 2016 +0200
  3080.  
  3081. Fix p4x4 analyse for 4:4:4 encoding with chroma ME
  3082.  
  3083. commit 07221290db0a94bda1f6ece3fdf3c02675c8adce [revision 2698]
  3084. Author: Anton Mitrofanov <[email protected]>
  3085. AuthorDate: Sun May 22 22:18:34 2016 +0300
  3086. Commit: Henrik Gramner <[email protected]>
  3087. CommitDate: Mon Jun 13 22:06:51 2016 +0200
  3088.  
  3089. Fix 4:4:4 encoding with CQM
  3090.  
  3091. commit 23ebc1f763936b7fcfc81e21530e1b65dbc503b9 [revision 2697]
  3092. Author: Anton Mitrofanov <[email protected]>
  3093. AuthorDate: Sun May 22 19:36:05 2016 +0300
  3094. Commit: Henrik Gramner <[email protected]>
  3095. CommitDate: Mon Jun 13 22:06:51 2016 +0200
  3096.  
  3097. Fix p4x4 RDO with CAVLC
  3098.  
  3099. commit 740a8c556bd9b68e899d6991f3f987a443aa14aa [revision 2696]
  3100. Author: Anton Mitrofanov <[email protected]>
  3101. AuthorDate: Sat Apr 23 23:10:03 2016 +0300
  3102. Commit: Henrik Gramner <[email protected]>
  3103. CommitDate: Mon Jun 13 21:58:37 2016 +0200
  3104.  
  3105. Apply zone options a little bit earlier
  3106.  
  3107. This way things like SAR changes will have full effect from the start frame.
  3108.  
  3109. commit 928bd9d5def4f0ca5071ea176a11b816a01e6495 [revision 2695]
  3110. Author: Anton Mitrofanov <[email protected]>
  3111. AuthorDate: Sat Apr 23 22:45:44 2016 +0300
  3112. Commit: Henrik Gramner <[email protected]>
  3113. CommitDate: Mon Jun 13 21:53:45 2016 +0200
  3114.  
  3115. Fix corruption when using encoder_reconfig() with some parameters
  3116.  
  3117. Changing parameters that affects SPS, like --ref for example, wasn't
  3118. behaving correctly previously.
  3119.  
  3120. Probably a regression in r2373.
  3121.  
  3122. commit 3b70645597bea052d2398005bc723212aeea6875 [revision 2694]
  3123. Author: Anton Mitrofanov <[email protected]>
  3124. AuthorDate: Wed Apr 13 21:54:25 2016 +0300
  3125. Commit: Henrik Gramner <[email protected]>
  3126. CommitDate: Wed Apr 20 17:15:54 2016 +0200
  3127.  
  3128. Clean up header includes
  3129.  
  3130. commit 2102de2584e03fce4abac49eb37d5d7a0803380f [revision 2693]
  3131. Author: Henrik Gramner <[email protected]>
  3132. AuthorDate: Wed Apr 13 17:53:49 2016 +0200
  3133. Commit: Henrik Gramner <[email protected]>
  3134. CommitDate: Wed Apr 20 17:15:51 2016 +0200
  3135.  
  3136. Eliminate some compiler warnings on BSD
  3137.  
  3138. Include <strings.h> in addition to <string.h>. According to the POSIX
  3139. specification the prototypes for strcasecmp() and strncasecmp() are
  3140. declared in <strings.h>. On some systems they are also declared in
  3141. <string.h> for compatibility reasons but we shouldn't rely on that.
  3142.  
  3143. Define _POSIX_C_SOURCE only when it's required to do so. Some BSD
  3144. variants doesn't declare certain function prototypes otherwise.
  3145.  
  3146. commit 64f4e24909924fceeea6e154d71b7dfbf586c7ea [revision 2692]
  3147. Author: Henrik Gramner <[email protected]>
  3148. AuthorDate: Tue Apr 12 21:33:54 2016 +0200
  3149. Commit: Henrik Gramner <[email protected]>
  3150. CommitDate: Tue Apr 12 21:47:47 2016 +0200
  3151.  
  3152. osx: Add -D_DARWIN_C_SOURCE to CFLAGS
  3153.  
  3154. OSX doesn't like _POSIX_C_SOURCE being defined when _DARWIN_C_SOURCE isn't.
  3155.  
  3156. commit 00597d74c6223f3694e2c6614ef0574d7fca6b22 [revision 2691]
  3157. Author: Anton Mitrofanov <[email protected]>
  3158. AuthorDate: Tue Apr 12 20:33:42 2016 +0300
  3159. Commit: Henrik Gramner <[email protected]>
  3160. CommitDate: Tue Apr 12 19:58:07 2016 +0200
  3161.  
  3162. Remove an unused parameter from x264_slicetype_frame_cost()
  3163.  
  3164. The b_intra_penalty parameter is no longer used anywhere after the
  3165. improvements to the --b-adapt 1 algorithm.
  3166.  
  3167. commit aa26e880bc2cd04cc81c776051d5e21d03fc975a [revision 2690]
  3168. Author: Anton Mitrofanov <[email protected]>
  3169. AuthorDate: Sun Apr 10 20:17:32 2016 +0300
  3170. Commit: Henrik Gramner <[email protected]>
  3171. CommitDate: Tue Apr 12 19:25:17 2016 +0200
  3172.  
  3173. Improve the --b-adapt 1 algorithm
  3174.  
  3175. Roughly the same speed as before but with significantly better results,
  3176. comparable to --b-adapt 2.
  3177.  
  3178. commit 24f25b6afd21488a93bd86098f98dfaf229fc149 [revision 2689]
  3179. Author: Henrik Gramner <[email protected]>
  3180. AuthorDate: Sun Apr 3 15:49:26 2016 +0200
  3181. Commit: Henrik Gramner <[email protected]>
  3182. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3183.  
  3184. analyse: i_sub_partition write combining
  3185.  
  3186. commit 1507cfe80ecf5f8e240a35e9e9dc5a92bd25e792 [revision 2688]
  3187. Author: Henrik Gramner <[email protected]>
  3188. AuthorDate: Tue Mar 15 20:16:45 2016 +0100
  3189. Commit: Henrik Gramner <[email protected]>
  3190. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3191.  
  3192. x86: Use one less register in mbtree_propagate_cost_avx2
  3193.  
  3194. Avoids the need to save and restore xmm6 on 64-bit Windows.
  3195.  
  3196. commit c82c7374938f4342971adf8b2495c3a1bbe621c4 [revision 2687]
  3197. Author: Henrik Gramner <[email protected]>
  3198. AuthorDate: Fri Mar 4 17:53:08 2016 +0100
  3199. Commit: Henrik Gramner <[email protected]>
  3200. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3201.  
  3202. x86: Add asm for mbtree fixed point conversion
  3203.  
  3204. The QP offsets of each macroblock are stored as floats internally and
  3205. converted to big-endian Q8.8 fixed point numbers when written to the 2-pass
  3206. stats file, and converted back to floats when read from the stats file.
  3207.  
  3208. Add SSSE3 and AVX2 implementations for conversions in both directions.
  3209.  
  3210. About 8x faster than C on Haswell.
  3211.  
  3212. commit be677efc6313ade5eddf722fdf097cce56df1344 [revision 2686]
  3213. Author: Anton Mitrofanov <[email protected]>
  3214. AuthorDate: Thu Apr 7 13:09:03 2016 +0300
  3215. Commit: Henrik Gramner <[email protected]>
  3216. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3217.  
  3218. x86inc: Enable AVX emulation in additional cases
  3219.  
  3220. Allows emulation to work when dst is equal to src2 as long as the
  3221. instruction is commutative, e.g. `addps m0, m1, m0`.
  3222.  
  3223. commit b5661d322866df647e6084061a471eceac214c28 [revision 2685]
  3224. Author: Anton Mitrofanov <[email protected]>
  3225. AuthorDate: Thu Apr 7 12:48:29 2016 +0300
  3226. Commit: Henrik Gramner <[email protected]>
  3227. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3228.  
  3229. x86inc: Improve handling of %ifid with multi-token parameters
  3230.  
  3231. The yasm/nasm preprocessor only checks the first token, which means that
  3232. parameters such as `dword [rax]` are treated as identifiers, which is
  3233. generally not what we want.
  3234.  
  3235. commit 283663d4c13088f4811c78b75318bda59d696b2d [revision 2684]
  3236. Author: Anton Mitrofanov <[email protected]>
  3237. AuthorDate: Mon Mar 28 18:35:38 2016 +0300
  3238. Commit: Henrik Gramner <[email protected]>
  3239. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3240.  
  3241. x86inc: Fix AVX emulation of some instructions
  3242.  
  3243. commit 54fd697668d0a04246ad0b0e9955a6583b2bb8b6 [revision 2683]
  3244. Author: Henrik Gramner <[email protected]>
  3245. AuthorDate: Fri Mar 4 17:51:41 2016 +0100
  3246. Commit: Henrik Gramner <[email protected]>
  3247. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3248.  
  3249. x86inc: Fix AVX emulation of scalar float instructions
  3250.  
  3251. Those instructions are not commutative since they only change the first
  3252. element in the vector and leave the rest unmodified.
  3253.  
  3254. commit eeb9b66ddb0f27d8baaa8efa9597613e61140836 [revision 2682]
  3255. Author: Henrik Gramner <[email protected]>
  3256. AuthorDate: Sat Feb 27 20:34:39 2016 +0100
  3257. Commit: Henrik Gramner <[email protected]>
  3258. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3259.  
  3260. x86: dct2x4dc asm
  3261.  
  3262. Only used in 4:2:2. MMX2 version implemented for 8-bit, SSE2 and AVX
  3263. versions implemented for high bit-depth.
  3264.  
  3265. 2.5x faster on 32-bit and 1.6x faster on 64-bit compared to C on Ivy Bridge.
  3266.  
  3267. commit 23d1d8e89be2d99f5c6924a6055fc80d69429503 [revision 2681]
  3268. Author: Henrik Gramner <[email protected]>
  3269. AuthorDate: Sat Feb 20 20:31:22 2016 +0100
  3270. Commit: Henrik Gramner <[email protected]>
  3271. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3272.  
  3273. x86: SSE2/AVX idct_dequant_2x4_(dc|dconly)
  3274.  
  3275. Only used in 4:2:2. Both 8-bit and high bit-depth implemented.
  3276.  
  3277. Approximate performance improvement compared to C on Ivy Bridge:
  3278.  
  3279. x86-32 x86-64
  3280. idct_dequant_2x4_dc 2.1x 1.7x
  3281. idct_dequant_2x4_dconly 2.7x 2.0x
  3282.  
  3283. Helps more on 32-bit due to the C versions being register starved.
  3284.  
  3285. commit dbbf1dd2836a21b65178442c1fb7a00ea089d7ec [revision 2680]
  3286. Author: Henrik Gramner <[email protected]>
  3287. AuthorDate: Sat Feb 20 16:53:35 2016 +0100
  3288. Commit: Henrik Gramner <[email protected]>
  3289. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3290.  
  3291. checkasm: Fix idct_dequant_2x4_(dc|dconly) tests
  3292.  
  3293. They used the wrong qp values and the dconly test had the wrong name. This
  3294. was undetected before because there wasn't any assembly implementations.
  3295.  
  3296. commit 0db0ac3a05b80eee7994fab08cbce2d07e8b1586 [revision 2679]
  3297. Author: Henrik Gramner <[email protected]>
  3298. AuthorDate: Sun Feb 7 14:55:26 2016 +0100
  3299. Commit: Henrik Gramner <[email protected]>
  3300. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3301.  
  3302. checkasm: Disable Windows Error Reporting
  3303.  
  3304. When developing new assembly code it's expected that checkasm may crash,
  3305. and the error reporting dialog popup can be somewhat annoying.
  3306.  
  3307. commit deae1b1001d134f5babc4fad3208bd951a454951 [revision 2678]
  3308. Author: Henrik Gramner <[email protected]>
  3309. AuthorDate: Sat Feb 6 18:49:46 2016 +0100
  3310. Commit: Henrik Gramner <[email protected]>
  3311. CommitDate: Tue Apr 12 17:10:39 2016 +0200
  3312.  
  3313. windows: Flag debug builds in the resource file
  3314.  
  3315. commit 0082b717199bafb4abbb6638e7c30d50deaf2c1b [revision 2677]
  3316. Author: Henrik Gramner <[email protected]>
  3317. AuthorDate: Thu Feb 4 20:06:57 2016 +0100
  3318. Commit: Henrik Gramner <[email protected]>
  3319. CommitDate: Tue Apr 12 17:10:29 2016 +0200
  3320.  
  3321. cli: Refactor filter option parsing
  3322.  
  3323. The old code contained a whole bunch of memory leaks, unchecked mallocs,
  3324. sections of dead code, etc. and was generally overly complex.
  3325.  
  3326. Also consolidate some memory allocations into a single one.
  3327.  
  3328. commit dfe394cadc8a39752de5b3f4a0be222c1b9290f2 [revision 2676]
  3329. Author: Henrik Gramner <[email protected]>
  3330. AuthorDate: Sun Jan 31 21:50:52 2016 +0100
  3331. Commit: Henrik Gramner <[email protected]>
  3332. CommitDate: Mon Apr 11 16:59:51 2016 +0200
  3333.  
  3334. ffms: Various improvements
  3335.  
  3336. * Drop the MinGW Unicode workarounds. Those were required at the time
  3337. Windows Unicode support was added to x264 but the underlying problem
  3338. has since been fixed in FFMS.
  3339.  
  3340. * Use FFMS_IndexBelongsToFile() as an additional sanity check when reading
  3341. an index file to ensure that it belongs to the current source video.
  3342.  
  3343. * Upgrade to the new API to prevent deprecation warnings when compiling.
  3344.  
  3345. * Fix a resource leak that would occur if FFMS_GetFirstTrackOfType() or
  3346. FFMS_CreateVideoSource() failed.
  3347.  
  3348. * Minor string handling adjustments related to progress reporting.
  3349.  
  3350. This increases the FFMS version requirement from 2.16.2 to 2.21.0.
  3351.  
  3352. commit 215afdbd8ecc924f2028f79851458076683e97ad [revision 2675]
  3353. Author: Henrik Gramner <[email protected]>
  3354. AuthorDate: Mon Apr 11 16:59:46 2016 +0200
  3355. Commit: Henrik Gramner <[email protected]>
  3356. CommitDate: Mon Apr 11 16:59:46 2016 +0200
  3357.  
  3358. msvc: Add snprintf/vsnprintf replacements
  3359.  
  3360. MSVC pre-VS2015 has broken snprintf/vsnprintf implementations which are
  3361. incompatible with C99 and may lead to buffer overflows.
  3362.  
  3363. commit 5be32efc244d96aa56be462664b5c56d7318e86d [revision 2674]
  3364. Author: Henrik Gramner <[email protected]>
  3365. AuthorDate: Sun Jan 31 20:21:01 2016 +0100
  3366. Commit: Henrik Gramner <[email protected]>
  3367. CommitDate: Mon Apr 11 16:58:23 2016 +0200
  3368.  
  3369. configure: Define feature test macros for --std=gnu99
  3370.  
  3371. Makes the printf() family functions on MinGW use the correct C99 POSIX
  3372. versions instead of the broken pre-VS2015 Microsoft ones.
  3373.  
  3374. Also allows us to get rid of some _GNU_SOURCE and _ISOC99_SOURCE defines.
  3375.  
  3376. commit c01bf42117b811a0469f9f6c374f4a0daa98716d [revision 2673]
  3377. Author: Henrik Gramner <[email protected]>
  3378. AuthorDate: Thu Jan 28 18:37:37 2016 +0100
  3379. Commit: Henrik Gramner <[email protected]>
  3380. CommitDate: Mon Apr 11 16:58:23 2016 +0200
  3381.  
  3382. mingw: Enable high-entropy ASLR on 64-bit Windows
  3383.  
  3384. To fully utilize HEASLR the image base address must also be set above
  3385. 4 GiB. For consistency use the same address as MSVC uses by default.
  3386.  
  3387. This requires binutils 2.25 which isn't available on all common
  3388. distributions, so only enable it after checking that it's supported.
  3389.  
  3390. commit dd6b7b974e0057da726f71e10c24d057a339605b [revision 2672]
  3391. Author: Henrik Gramner <[email protected]>
  3392. AuthorDate: Sun Jan 24 01:48:18 2016 +0100
  3393. Commit: Henrik Gramner <[email protected]>
  3394. CommitDate: Mon Apr 11 16:58:22 2016 +0200
  3395.  
  3396. msvs: WinRT support
  3397.  
  3398. To compile x264 for WinRT the following additional steps has to be performed.
  3399.  
  3400. * Ensure that the necessary SDK is installed.
  3401.  
  3402. * Set the correct environment variables in the VS command prompt as shown at
  3403. https://trac.ffmpeg.org/wiki/CompilationGuide/WinRT
  3404.  
  3405. * Add one of the following to --extra-cflags depending on the target OS:
  3406. "-DWINAPI_FAMILY=WINAPI_FAMILY_PC_APP -D_WIN32_WINNT=0x0A00" (Windows 10)
  3407. "-DWINAPI_FAMILY=WINAPI_FAMILY_PC_APP -D_WIN32_WINNT=0x0603" (Windows 8.1)
  3408.  
  3409. commit 7650a1367003e24f4f1b831682c012b5ba3e6c69 [revision 2671]
  3410. Author: Henrik Gramner <[email protected]>
  3411. AuthorDate: Sun Jan 24 23:58:40 2016 +0100
  3412. Commit: Henrik Gramner <[email protected]>
  3413. CommitDate: Mon Apr 11 16:58:22 2016 +0200
  3414.  
  3415. configure: Disable CLI libraries when CLI is disabled
  3416.  
  3417. commit 1ce062abb47ac59621b402cb26a1f14c91bb52bc [revision 2670]
  3418. Author: Henrik Gramner <[email protected]>
  3419. AuthorDate: Fri Feb 5 18:46:13 2016 +0100
  3420. Commit: Henrik Gramner <[email protected]>
  3421. CommitDate: Mon Apr 11 16:58:22 2016 +0200
  3422.  
  3423. matroska: mk_close: Check fseek() return value
  3424.  
  3425. commit de7af9185e172122cd9b800845e1988a52ad7cc3 [revision 2669]
  3426. Author: Henrik Gramner <[email protected]>
  3427. AuthorDate: Fri Feb 5 18:46:02 2016 +0100
  3428. Commit: Henrik Gramner <[email protected]>
  3429. CommitDate: Mon Apr 11 16:58:22 2016 +0200
  3430.  
  3431. parse_qpfile: Check ftell() and fseek() return values
  3432.  
  3433. commit fd2c324731c2199e502ded9eff723d29c6eafe0b [revision 2668]
  3434. Author: Anton Mitrofanov <[email protected]>
  3435. AuthorDate: Sun Apr 10 20:13:59 2016 +0300
  3436. Commit: Henrik Gramner <[email protected]>
  3437. CommitDate: Mon Apr 11 16:58:22 2016 +0200
  3438.  
  3439. Use the correct default B-ref placement with B-pyramid
  3440.  
  3441. Cost analyse functions expects the placement of the B-ref in a sequence of
  3442. an even number of B-frames to be located towards the beginning while the
  3443. actual placement was towards the end.
  3444.  
  3445. Change the placement to be consistent with the analyse expectations, e.g.
  3446. PbbBbP -> PbBbbP.
  3447.  
  3448. commit e6a3f2989dd9eba3434c21fa94a6d9a5d1c7a9fe [revision 2667]
  3449. Author: Henrik Gramner <[email protected]>
  3450. AuthorDate: Fri Feb 5 18:45:47 2016 +0100
  3451. Commit: Henrik Gramner <[email protected]>
  3452. CommitDate: Fri Feb 5 18:48:13 2016 +0100
  3453.  
  3454. parse_zones: Fix memory leak
  3455.  
  3456. commit f86756985d42ac4a14866534c588061ede860b7b [revision 2666]
  3457. Author: Alexey Samsonov <[email protected]>
  3458. AuthorDate: Mon Jan 25 16:05:25 2016 -0800
  3459. Commit: Henrik Gramner <[email protected]>
  3460. CommitDate: Fri Feb 5 18:48:13 2016 +0100
  3461.  
  3462. Fix float-cast-overflow in x264_ratecontrol_end function
  3463.  
  3464. According to the C standard, it is undefined behavior to cast a negative
  3465. floating point number to an unsigned integer. Float-cast-overflow in
  3466. general is known to produce different results on different architectures.
  3467.  
  3468. Building x264 code with Clang and -fsanitize=float-cast-overflow
  3469. (http://clang.llvm.org/docs/UndefinedBehaviorSanitizer.html#availablle-checks)
  3470. and running it on some real-life examples occasionally produces errors
  3471. of the form:
  3472.  
  3473. encoder/ratecontrol.c:1892: runtime error: value -5011.14 is outside the
  3474. range of representable values of type 'unsigned short'
  3475.  
  3476. Fix these errors by explicitly coding the de-facto x86 behavior: casting
  3477. float to uint16_t through int16_t.
  3478.  
  3479. commit a01e33913655f983df7a4d64b0a4178abb1eb618 [revision 2665]
  3480. Author: Sebastian Dröge <[email protected]>
  3481. AuthorDate: Sun Dec 20 23:49:35 2015 +0300
  3482. Commit: Anton Mitrofanov <[email protected]>
  3483. CommitDate: Mon Jan 18 22:16:42 2016 +0300
  3484.  
  3485. Fix AVC-Intra padding for non-Annex B encoding
  3486.  
  3487. commit 1e4a24f305c006a95fec00131703d0e0ecae3a38 [revision 2664]
  3488. Author: Anton Mitrofanov <[email protected]>
  3489. AuthorDate: Mon Jan 11 21:39:22 2016 +0300
  3490. Commit: Anton Mitrofanov <[email protected]>
  3491. CommitDate: Sun Jan 17 23:05:52 2016 +0300
  3492.  
  3493. ppc: Only perform AltiVec detection if compiled with AltiVec enabled
  3494.  
  3495. commit b5953629117adc2b8d0d0eed6eb323c00587b428 [revision 2663]
  3496. Author: Anton Mitrofanov <[email protected]>
  3497. AuthorDate: Tue Oct 13 15:30:16 2015 +0300
  3498. Commit: Henrik Gramner <[email protected]>
  3499. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3500.  
  3501. 2-pass: Take into account possible frame reordering
  3502.  
  3503. commit 20821a26ec510979e49fcfd6becc6ad7e2d8b388 [revision 2662]
  3504. Author: Anton Mitrofanov <[email protected]>
  3505. AuthorDate: Tue Oct 13 12:54:05 2015 +0300
  3506. Commit: Henrik Gramner <[email protected]>
  3507. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3508.  
  3509. Revise the 2-pass algorithm
  3510.  
  3511. commit 065321c48d0d371c1735b3cc9d368b43e1b64aaa [revision 2661]
  3512. Author: Anton Mitrofanov <[email protected]>
  3513. AuthorDate: Tue Jan 5 02:41:43 2016 +0300
  3514. Commit: Henrik Gramner <[email protected]>
  3515. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3516.  
  3517. Revise the row VBV algorithm (part 2)
  3518.  
  3519. Should fix rare cases of VBV emergency mode activation caused by too much trust
  3520. to the row predictors.
  3521.  
  3522. commit d23d18655249944c1ca894b451e2c82c7a584c62 [revision 2660]
  3523. Author: Henrik Gramner <[email protected]>
  3524. AuthorDate: Fri Jan 1 12:44:31 2016 +0100
  3525. Commit: Henrik Gramner <[email protected]>
  3526. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3527.  
  3528. Bump dates to 2016
  3529.  
  3530. commit 3d972062c8a37d1a19586e2351e889b0a70beb40 [revision 2659]
  3531. Author: Henrik Gramner <[email protected]>
  3532. AuthorDate: Mon Oct 26 19:54:20 2015 +0100
  3533. Commit: Henrik Gramner <[email protected]>
  3534. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3535.  
  3536. cli: Use memory-mapped input frames for yuv and y4m
  3537.  
  3538. Improves performance by avoiding extraneous memory copying.
  3539. Most beneficial on fast settings.
  3540.  
  3541. On average around 5-10% faster overall on ultrafast but the
  3542. performance improvement can be even larger in some cases.
  3543.  
  3544. commit 38a5268dbec56adea750e05c4981f3bbb176e735 [revision 2658]
  3545. Author: Henrik Gramner <[email protected]>
  3546. AuthorDate: Thu Jan 7 01:59:24 2016 +0100
  3547. Commit: Henrik Gramner <[email protected]>
  3548. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3549.  
  3550. y4m: Support extended frame headers when seeking
  3551.  
  3552. Use the actual length of the frame header of the first frame instead of
  3553. assuming a header without extensions when calculating the frame size.
  3554.  
  3555. Also makes the frame counter more accurate with extended frame headers.
  3556.  
  3557. commit cc652c158c1fa65bfeafb6446b5be855850065d0 [revision 2657]
  3558. Author: Henrik Gramner <[email protected]>
  3559. AuthorDate: Tue Nov 3 17:55:08 2015 +0100
  3560. Commit: Henrik Gramner <[email protected]>
  3561. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3562.  
  3563. configure: Simplify cygwin/mingw/msys code
  3564.  
  3565. Avoids some code duplication.
  3566.  
  3567. Also drop the -mno-cygwin check since that option was removed back in 2008.
  3568.  
  3569. commit 8b2d2a6d51abf51ad38dd8705d280448fbe63aaf [revision 2656]
  3570. Author: Henrik Gramner <[email protected]>
  3571. AuthorDate: Mon Oct 26 18:52:46 2015 +0100
  3572. Commit: Henrik Gramner <[email protected]>
  3573. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3574.  
  3575. y4m: Avoid some redundant strlen() calls
  3576.  
  3577. commit 24f7705f15cf6d59028a76a894d866b9fad85f39 [revision 2655]
  3578. Author: Henrik Gramner <[email protected]>
  3579. AuthorDate: Sun Oct 25 17:15:10 2015 +0100
  3580. Commit: Henrik Gramner <[email protected]>
  3581. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3582.  
  3583. Simplify threadpool_wait
  3584.  
  3585. commit 30ba5dc22fd0ae359e144847f2636574f659627d [revision 2654]
  3586. Author: Henrik Gramner <[email protected]>
  3587. AuthorDate: Fri Oct 16 19:05:34 2015 +0200
  3588. Commit: Henrik Gramner <[email protected]>
  3589. CommitDate: Sun Jan 17 00:30:13 2016 +0100
  3590.  
  3591. windows: Use native threads by default
  3592.  
  3593. --disable-win32thread can be passed as an argument to configure to compile
  3594. with pthreads, which was the old default behavior.
  3595.  
  3596. commit 1637239a64f3ec9a491b91202bd37097f15a253d [revision 2653]
  3597. Author: Henrik Gramner <[email protected]>
  3598. AuthorDate: Sun Oct 11 22:32:11 2015 +0200
  3599. Commit: Henrik Gramner <[email protected]>
  3600. CommitDate: Sun Jan 17 00:30:12 2016 +0100
  3601.  
  3602. x86: Avoid some bypass delays and false dependencies
  3603.  
  3604. A bypass delay of 1-3 clock cycles may occur on some CPUs when transitioning
  3605. between int and float domains, so try to avoid that if possible.
  3606.  
  3607. commit 7688814a7ec994f8e5984d199b465ccc068b98af [revision 2652]
  3608. Author: Henrik Gramner <[email protected]>
  3609. AuthorDate: Sun Oct 11 22:32:03 2015 +0200
  3610. Commit: Henrik Gramner <[email protected]>
  3611. CommitDate: Sun Jan 17 00:30:12 2016 +0100
  3612.  
  3613. x86: Enable high bit-depth x264_coeff_last64_avx2_lzcnt
  3614.  
  3615. The function existed but was never enabled.
  3616.  
  3617. commit 366fa85885053c7b836a4272a4fbec1852103979 [revision 2651]
  3618. Author: Geza Lore <[email protected]>
  3619. AuthorDate: Mon Oct 12 13:13:42 2015 +0100
  3620. Commit: Henrik Gramner <[email protected]>
  3621. CommitDate: Sun Jan 17 00:30:12 2016 +0100
  3622.  
  3623. x86inc: Add debug symbols indicating sizes of compiled functions
  3624.  
  3625. Some debuggers/profilers use this metadata to determine which function a
  3626. given instruction is in; without it they get can confused by local labels
  3627. (if you haven't stripped those). On the other hand, some tools are still
  3628. confused even with this metadata. e.g. this fixes `gdb`, but not `perf`.
  3629.  
  3630. Currently only implemented for ELF.
  3631.  
  3632. commit 70c3ba42e610b4182edda4fdeb10b37a2a70eb8f [revision 2650]
  3633. Author: Henrik Gramner <[email protected]>
  3634. AuthorDate: Fri Oct 16 21:28:49 2015 +0200
  3635. Commit: Henrik Gramner <[email protected]>
  3636. CommitDate: Sun Jan 17 00:30:12 2016 +0100
  3637.  
  3638. x86inc: Avoid creating unnecessary local labels
  3639.  
  3640. The REP_RET workaround is only needed on old AMD cpus, and the labels clutter
  3641. up the symbol table and confuse debugging/profiling tools, so use EQU to
  3642. create SHN_ABS symbols instead of creating local labels. Furthermore, skip
  3643. the workaround completely in functions that definitely won't run on such cpus.
  3644.  
  3645. This patch doesn't modify any emitted instructions, and doesn't actually affect
  3646. x264 at all. It's only for other projects that use x86inc.asm without an
  3647. appropriate `strip` command in their buildsystem.
  3648.  
  3649. Note that EQU is just creating a local label when using nasm instead of yasm.
  3650. This is probably a bug, but at least it doesn't break anything.
  3651.  
  3652. commit 5c3d473a966e4b013759097fb98cd4a9cb5a34f5 [revision 2649]
  3653. Author: Henrik Gramner <[email protected]>
  3654. AuthorDate: Thu Oct 15 17:42:49 2015 +0200
  3655. Commit: Henrik Gramner <[email protected]>
  3656. CommitDate: Sun Jan 17 00:30:12 2016 +0100
  3657.  
  3658. x86inc: Simplify AUTO_REP_RET
  3659.  
  3660. cpuflags is never undefined any more, it's set to 0 instead.
  3661.  
  3662. Also fix an incorrect comment.
  3663.  
  3664. commit 28d68f090c0103704f5f6a86fcf362251774cd78 [revision 2648]
  3665. Author: Henrik Gramner <[email protected]>
  3666. AuthorDate: Mon Oct 12 21:55:11 2015 +0200
  3667. Commit: Henrik Gramner <[email protected]>
  3668. CommitDate: Sun Jan 17 00:30:12 2016 +0100
  3669.  
  3670. x86inc: Use more consistent indentation
  3671.  
  3672. commit 963b99efaaf1f0628b155e52b8a7c102cd1d37ff [revision 2647]
  3673. Author: Henrik Gramner <[email protected]>
  3674. AuthorDate: Mon Oct 12 20:15:18 2015 +0200
  3675. Commit: Henrik Gramner <[email protected]>
  3676. CommitDate: Sun Jan 17 00:30:12 2016 +0100
  3677.  
  3678. x86inc: Preserve arguments when allocating stack space
  3679.  
  3680. When allocating stack space with a larger alignment than the known stack
  3681. alignment a temporary register is used for storing the stack pointer.
  3682. Ensure that this isn't one of the registers used for passing arguments.
  3683.  
  3684. commit 6e5033417a53fa66d002665618a1350d7417725e [revision 2646]
  3685. Author: Henrik Gramner <[email protected]>
  3686. AuthorDate: Sun Jan 17 00:25:47 2016 +0100
  3687. Commit: Henrik Gramner <[email protected]>
  3688. CommitDate: Sun Jan 17 00:25:47 2016 +0100
  3689.  
  3690. x86inc: Improve FMA instruction handling
  3691.  
  3692. * Correctly handle FMA instructions with memory operands.
  3693. * Print a warning if FMA instructions are used without the correct cpuflag.
  3694. * Simplify the instantiation code.
  3695. * Clarify documentation.
  3696.  
  3697. Only the last operand in FMA3 instructions can be a memory operand. When
  3698. converting FMA4 instructions to FMA3 instructions we can utilize the fact
  3699. that multiply is a commutative operation and reorder operands if necessary
  3700. to ensure that a memory operand is used only as the last operand.
  3701.  
  3702. commit 93cba743c78959ad97812dbaf894903c608912d0 [revision 2645]
  3703. Author: Henrik Gramner <[email protected]>
  3704. AuthorDate: Sun Oct 11 22:31:53 2015 +0200
  3705. Commit: Henrik Gramner <[email protected]>
  3706. CommitDate: Sun Jan 3 17:15:47 2016 +0100
  3707.  
  3708. x86inc: Be more verbose in assertion failures
  3709.  
  3710. commit 8017b33454397d59b3285ec6d2ad35b6d0deb58a [revision 2644]
  3711. Author: Henrik Gramner <[email protected]>
  3712. AuthorDate: Wed Sep 30 23:17:00 2015 +0200
  3713. Commit: Henrik Gramner <[email protected]>
  3714. CommitDate: Sun Jan 3 17:15:47 2016 +0100
  3715.  
  3716. x86inc: Make cpuflag() and notcpuflag() return 0 or 1
  3717.  
  3718. Makes it possible to use them in arithmetic expressions.
  3719.  
  3720. commit 5c6570495f8f1c716b294aee1430d8766a4beb9c [revision 2643]
  3721. Author: Henrik Gramner <[email protected]>
  3722. AuthorDate: Fri Oct 30 16:55:49 2015 +0100
  3723. Commit: Henrik Gramner <[email protected]>
  3724. CommitDate: Sun Jan 3 17:15:47 2016 +0100
  3725.  
  3726. encoder_open: Fix memory leak
  3727.  
  3728. Furthermore, the x264_analyse_prepare_costs() and x264_analyse_init_costs()
  3729. functions were only used in x264_encoder_open(), so move that entire section
  3730. of code to analyse.c as well to simplify things.
  3731.  
  3732. commit 424534537a249dcf913e02560303f6afca423489 [revision 2642]
  3733. Author: Janne Grunau <[email protected]>
  3734. AuthorDate: Wed Nov 18 11:08:22 2015 +0100
  3735. Commit: Henrik Gramner <[email protected]>
  3736. CommitDate: Sun Dec 20 18:40:11 2015 +0100
  3737.  
  3738. arm: do not fill mc_weight*_neon tabs for HIGH_BIT_DEPTH
  3739.  
  3740. The asm is only for 8-bit and function prototypes reflect that. Avoids
  3741. numerous warnings with --bit-depth=9/10.
  3742.  
  3743. commit df51d8efa8ce9afcedda64acc69c1dba2648716d [revision 2641]
  3744. Author: Janne Grunau <[email protected]>
  3745. AuthorDate: Tue Oct 13 23:50:11 2015 +0200
  3746. Commit: Henrik Gramner <[email protected]>
  3747. CommitDate: Sun Dec 20 18:38:40 2015 +0100
  3748.  
  3749. arm: Eliminate text relocations in asm
  3750.  
  3751. Android 6 does not link shared libraries with text relocations.
  3752.  
  3753. Make the movrel macro position independent and add movrelx for indirect
  3754. loads of external symbols.
  3755.  
  3756. Move the function pointer table for the aligned memcpy variants to the
  3757. data.rel.ro section on Linux/Android.
  3758.  
  3759. commit a2fe237af1b68f2bd53d64ed3faed62429d3ee5a [revision 2640]
  3760. Author: Martin Storsjö <[email protected]>
  3761. AuthorDate: Thu Oct 15 11:50:33 2015 +0300
  3762. Commit: Henrik Gramner <[email protected]>
  3763. CommitDate: Sat Oct 17 14:07:44 2015 +0200
  3764.  
  3765. arm: Don't assume alignment in mbtree_propagate_list_internal where it isn't provided
  3766.  
  3767. commit 9f422c0cd9c0abcd6a7abb10b51f8be883c39b2b [revision 2639]
  3768. Author: Janne Grunau <[email protected]>
  3769. AuthorDate: Tue Oct 13 23:50:12 2015 +0200
  3770. Commit: Henrik Gramner <[email protected]>
  3771. CommitDate: Sat Oct 17 14:07:44 2015 +0200
  3772.  
  3773. arm: Fix checkasm register clobber check on iOS
  3774.  
  3775. r9 is a volatile register in the iOS ABI and will therefore not be
  3776. preserved by compiled functions like the luma motion compensation.
  3777.  
  3778. Add the symbol prefix to the puts() call and use blx since a switch
  3779. between arm and thumb mode might be required.
  3780.  
  3781. commit 75992107adcc8317ba2888e3957a7d56f16b5cd4 [revision 2638]
  3782. Author: Anton Mitrofanov <[email protected]>
  3783. AuthorDate: Thu Oct 1 01:02:16 2015 +0300
  3784. Commit: Henrik Gramner <[email protected]>
  3785. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3786.  
  3787. ppc: Add detection of AltiVec support for FreeBSD
  3788.  
  3789. Patch from FreeBSD ports.
  3790.  
  3791. commit 479d0c1fe73833ba65e0a10f6f5cf18df6def719 [revision 2637]
  3792. Author: Anton Mitrofanov <[email protected]>
  3793. AuthorDate: Mon Sep 28 21:07:55 2015 +0300
  3794. Commit: Henrik Gramner <[email protected]>
  3795. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3796.  
  3797. Don't assume 16-byte stack alignment by default on x86-32
  3798.  
  3799. Some compilers depending on target OS uses 4-byte stack alignment by default.
  3800. Explicitly check known good compilers and specific options for stack alignment.
  3801.  
  3802. commit fad44d59b3adeb29b9c92fde0b80116cde79020e [revision 2636]
  3803. Author: Anton Mitrofanov <[email protected]>
  3804. AuthorDate: Tue Sep 22 21:33:07 2015 +0300
  3805. Commit: Henrik Gramner <[email protected]>
  3806. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3807.  
  3808. Fix a few static analyzer performance hints
  3809.  
  3810. commit de24c8c189364013e62d58d1e8f2fef878eb62bf [revision 2635]
  3811. Author: Anton Mitrofanov <[email protected]>
  3812. AuthorDate: Tue Sep 22 20:19:23 2015 +0300
  3813. Commit: Henrik Gramner <[email protected]>
  3814. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3815.  
  3816. Revise the row VBV algorithm
  3817.  
  3818. commit 001d30598c75d9bbc3aa80f67f9bdac17692437d [revision 2634]
  3819. Author: Anton Mitrofanov <[email protected]>
  3820. AuthorDate: Tue Sep 22 19:26:25 2015 +0300
  3821. Commit: Henrik Gramner <[email protected]>
  3822. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3823.  
  3824. Fix high bit depth lookahead cost compensation algorithm
  3825.  
  3826. Now high bit depth VBV should act more like 8-bit depth one.
  3827.  
  3828. commit 91368390db9179226b5b4ed718a5788b754f9302 [revision 2633]
  3829. Author: Anton Mitrofanov <[email protected]>
  3830. AuthorDate: Tue Sep 22 19:05:52 2015 +0300
  3831. Commit: Henrik Gramner <[email protected]>
  3832. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3833.  
  3834. Correctly update the intra row predictor in B-frames
  3835.  
  3836. It was previously used but never updated from it's initialization value.
  3837.  
  3838. commit e0d722f85f8599e324be2bebef9430155b25c329 [revision 2632]
  3839. Author: Anton Mitrofanov <[email protected]>
  3840. AuthorDate: Tue Sep 22 18:58:24 2015 +0300
  3841. Commit: Henrik Gramner <[email protected]>
  3842. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3843.  
  3844. Change the predictors update algorithm
  3845.  
  3846. Keep predictor offsets more stable. This should fix VBV misprediction in frames
  3847. with a large difference in complexity between the top and bottom parts.
  3848.  
  3849. commit 6f04b146875c45e6f7845a7bb5fb7fdf8e7534f1 [revision 2631]
  3850. Author: Martin Storsjö <[email protected]>
  3851. AuthorDate: Thu Sep 3 09:30:44 2015 +0300
  3852. Commit: Henrik Gramner <[email protected]>
  3853. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3854.  
  3855. arm: Implement x264_mbtree_propagate_{cost, list}_neon
  3856.  
  3857. The cost function could be simplified to avoid having to clobber
  3858. q4/q5, but this requires reordering instructions which increase
  3859. the total runtime.
  3860.  
  3861. checkasm timing Cortex-A7 A8 A9
  3862. mbtree_propagate_cost_c 63702 155835 62829
  3863. mbtree_propagate_cost_neon 17199 10454 11106
  3864.  
  3865. mbtree_propagate_list_c 104203 108949 84532
  3866. mbtree_propagate_list_neon 82035 78348 60410
  3867.  
  3868. commit 3e25eab0b7172e3c0b067b8b6d641ce148d03db9 [revision 2630]
  3869. Author: Martin Storsjö <[email protected]>
  3870. AuthorDate: Thu Sep 3 09:30:43 2015 +0300
  3871. Commit: Henrik Gramner <[email protected]>
  3872. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3873.  
  3874. x86: Share the mbtree_propagate_list macro with aarch64
  3875.  
  3876. This avoids having to duplicate the same code for all architectures
  3877. that implement only the internal part of this function in assembler.
  3878.  
  3879. commit 654901dfca73a21e2bb2366dda79eb413e9bfb66 [revision 2629]
  3880. Author: Martin Storsjö <[email protected]>
  3881. AuthorDate: Wed Sep 2 22:39:51 2015 +0300
  3882. Commit: Henrik Gramner <[email protected]>
  3883. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3884.  
  3885. arm: Implement luma intra deblocking
  3886.  
  3887. checkasm timing Cortex-A7 A8 A9
  3888. deblock_luma_intra[0]_c 5988 4653 4316
  3889. deblock_luma_intra[0]_neon 3103 2170 2128
  3890. deblock_luma_intra[1]_c 7119 5905 5347
  3891. deblock_luma_intra[1]_neon 2068 1381 1412
  3892.  
  3893. This includes extra optimizations by Janne Grunau.
  3894.  
  3895. Timings from a separate build, on Exynos 5422:
  3896.  
  3897. Cortex-A7 A15
  3898. deblock_luma_intra[0]_c 6627 3300
  3899. deblock_luma_intra[0]_neon 3059 1128
  3900. deblock_luma_intra[1]_c 7314 4128
  3901. deblock_luma_intra[1]_neon 2038 720
  3902.  
  3903. commit e2696a60a3e58d92e88e149b63c0b06a066eea9e [revision 2628]
  3904. Author: Martin Storsjö <[email protected]>
  3905. AuthorDate: Mon Aug 31 22:40:31 2015 +0300
  3906. Commit: Henrik Gramner <[email protected]>
  3907. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3908.  
  3909. arm: Implement some neon 8x16c intra predict functions
  3910.  
  3911. checkasm timing Cortex-A7 A8 A9
  3912. intra_predict_8x16c_dct_c 862 540 590
  3913. intra_predict_8x16c_dct_neon 608 511 657
  3914. intra_predict_8x16c_h_c 972 707 719
  3915. intra_predict_8x16c_h_neon 722 656 672
  3916. intra_predict_8x16c_p_c 10183 9819 8655
  3917. intra_predict_8x16c_p_neon 2622 1972 1983
  3918.  
  3919. commit 5db8b6b93aa91079ab785b9b49413625430536fd [revision 2627]
  3920. Author: Martin Storsjö <[email protected]>
  3921. AuthorDate: Fri Aug 28 00:15:01 2015 +0300
  3922. Commit: Henrik Gramner <[email protected]>
  3923. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3924.  
  3925. arm: Implement x264_plane_copy_neon
  3926.  
  3927. checkasm timing Cortex-A7 A8 A9
  3928. plane_copy_c 13124 10925 9106
  3929. plane_copy_neon 7349 5103 8945
  3930.  
  3931. commit 35d32d09e163bb0f2ce60a8e13f9f22125445346 [revision 2626]
  3932. Author: Martin Storsjö <[email protected]>
  3933. AuthorDate: Fri Aug 28 09:40:24 2015 +0300
  3934. Commit: Henrik Gramner <[email protected]>
  3935. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3936.  
  3937. checkasm: arm: Check register clobbering
  3938.  
  3939. Cast the function pointer to a different type signature, to
  3940. be able to use uint64_t as return type (instead of intptr_t) for
  3941. those calls that require it.
  3942.  
  3943. Use two separate functions, depending on whether neon is available.
  3944.  
  3945. commit 9cbdb635a4bd78e6767e735a062c0d9a5766b849 [revision 2625]
  3946. Author: Martin Storsjö <[email protected]>
  3947. AuthorDate: Fri Aug 14 00:00:57 2015 +0300
  3948. Commit: Henrik Gramner <[email protected]>
  3949. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3950.  
  3951. checkasm: Try different widths for ssd_nv12
  3952.  
  3953. To test all codepaths in the aarch64 neon implementation, one at
  3954. the very least needs to test with width 8, 16, 24 and 32.
  3955.  
  3956. commit 39af8c72e618a544baa06ae427fb2b440861abcd [revision 2624]
  3957. Author: Jerome Duval <[email protected]>
  3958. AuthorDate: Fri Jun 13 19:56:27 2014 +0000
  3959. Commit: Henrik Gramner <[email protected]>
  3960. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3961.  
  3962. Haiku support
  3963.  
  3964. Add Haiku as supported platform in configure.
  3965. Haiku has no nice() function, use the platform specific substitute instead.
  3966.  
  3967. commit 59683a97b50b34c6282457a959bb6b3e9e7f8c0d [revision 2623]
  3968. Author: Martin Storsjö <[email protected]>
  3969. AuthorDate: Tue Aug 25 14:38:20 2015 +0300
  3970. Commit: Henrik Gramner <[email protected]>
  3971. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3972.  
  3973. checkasm: aarch64: Check register clobbering
  3974.  
  3975. Disable this on iOS, since it has got a slightly different ABI
  3976. for vararg parameters.
  3977.  
  3978. commit 5c13589be828b524100c787057d6bef77898c657 [revision 2622]
  3979. Author: Martin Storsjö <[email protected]>
  3980. AuthorDate: Tue Aug 25 23:36:45 2015 +0300
  3981. Commit: Henrik Gramner <[email protected]>
  3982. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3983.  
  3984. arm: Implement x284_decimate_score15/16/64_neon
  3985.  
  3986. checkasm timing Cortex-A7 A8 A9
  3987. decimate_score15_c 764 736 535
  3988. decimate_score15_neon 487 494 453
  3989. decimate_score16_c 782 727 553
  3990. decimate_score16_neon 487 494 521
  3991. decimate_score64_c 2361 2597 2011
  3992. decimate_score64_neon 1017 802 785
  3993.  
  3994. commit 3902ae02a0edede5d6c44cb3ee9e24e618c66e6a [revision 2621]
  3995. Author: Martin Storsjö <[email protected]>
  3996. AuthorDate: Tue Aug 25 23:36:44 2015 +0300
  3997. Commit: Henrik Gramner <[email protected]>
  3998. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  3999.  
  4000. arm: Implement chroma intra deblock
  4001.  
  4002. checkasm timing Cortex-A7 A8 A9
  4003. deblock_chroma_420_intra_mbaff_c 1469 1276 1181
  4004. deblock_chroma_420_intra_mbaff_neon 981 717 644
  4005. deblock_chroma_intra[1]_c 2954 2402 2321
  4006. deblock_chroma_intra[1]_neon 947 581 575
  4007. deblock_h_chroma_420_intra_c 2859 2509 2264
  4008. deblock_h_chroma_420_intra_neon 1480 1119 1028
  4009. deblock_h_chroma_422_intra_c 6211 5030 4792
  4010. deblock_h_chroma_422_intra_neon 2894 1990 2077
  4011.  
  4012. commit e8b95e92792d9353277995043757430cf3dc3bf7 [revision 2620]
  4013. Author: Martin Storsjö <[email protected]>
  4014. AuthorDate: Tue Aug 25 14:38:17 2015 +0300
  4015. Commit: Henrik Gramner <[email protected]>
  4016. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4017.  
  4018. arm: Implement x264_pixel_sa8d_satd_16x16_neon
  4019.  
  4020. This requires spilling some registers to the stack,
  4021. contray to the aarch64 version.
  4022.  
  4023. checkasm timing Cortex-A7 A8 A9
  4024. sa8d_satd_16x16_neon 12936 6365 7492
  4025. sa8d_satd_16x16_separate_neon 14841 6605 8324
  4026.  
  4027. commit 6bbaa2758d53d0d6d645142d7d818c960d137a0e [revision 2619]
  4028. Author: Martin Storsjö <[email protected]>
  4029. AuthorDate: Tue Aug 25 14:38:16 2015 +0300
  4030. Commit: Henrik Gramner <[email protected]>
  4031. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4032.  
  4033. arm: Implement x264_deblock_h_chroma_mbaff_neon
  4034.  
  4035. checkasm timing Cortex-A7 A8 A9
  4036. deblock_chroma_420_mbaff_c 1944 1706 1526
  4037. deblock_chroma_420_mbaff_neon 1210 873 865
  4038.  
  4039. commit 3c66591e859045ef79a7131b991a5f20c80ffbb4 [revision 2618]
  4040. Author: Martin Storsjö <[email protected]>
  4041. AuthorDate: Tue Aug 25 14:38:15 2015 +0300
  4042. Commit: Henrik Gramner <[email protected]>
  4043. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4044.  
  4045. arm: Implement x264_deblock_h_chroma_422_neon
  4046.  
  4047. checkasm timing Cortex-A7 A8 A9
  4048. deblock_h_chroma_422_c 6953 6269 5145
  4049. deblock_h_chroma_422_neon 3905 2569 2551
  4050.  
  4051. commit 5265b927b0f2e043dd39cbbbf3909da0862d60e6 [revision 2617]
  4052. Author: Martin Storsjö <[email protected]>
  4053. AuthorDate: Tue Aug 25 14:38:14 2015 +0300
  4054. Commit: Henrik Gramner <[email protected]>
  4055. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4056.  
  4057. arm: Implement integral_init4/8h/v_neon
  4058.  
  4059. checkasm timing Cortex-A7 A8 A9
  4060. integral_init4h_c 10466 8590 6161
  4061. integral_init4h_neon 3021 1494 1800
  4062. integral_init4v_c 16250 13590 13628
  4063. integral_init4v_neon 3473 2073 3291
  4064. integral_init8h_c 10100 8275 5705
  4065. integral_init8h_neon 4403 2344 2751
  4066. integral_init8v_c 6403 4632 4999
  4067. integral_init8v_neon 1184 783 1306
  4068.  
  4069. commit b08403b5593307b919bfe5bfbd743da825326a4c [revision 2616]
  4070. Author: Martin Storsjö <[email protected]>
  4071. AuthorDate: Tue Aug 25 14:38:13 2015 +0300
  4072. Commit: Henrik Gramner <[email protected]>
  4073. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4074.  
  4075. arm: Implement x264_denoise_dct_neon
  4076.  
  4077. checkasm timing Cortex-A7 A8 A9
  4078. denoise_dct_c 6604 5510 5858
  4079. denoise_dct_neon 1774 1139 1614
  4080.  
  4081. commit ceee976bde76a5f4126bfd9d8454f0e601e67204 [revision 2615]
  4082. Author: Martin Storsjö <[email protected]>
  4083. AuthorDate: Tue Aug 25 14:38:12 2015 +0300
  4084. Commit: Henrik Gramner <[email protected]>
  4085. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4086.  
  4087. arm: Add x264_nal_escape_neon
  4088.  
  4089. checkasm timing Cortex-A7 A8 A9
  4090. nal_escape_c 852758 879566 655497
  4091. nal_escape_neon 376831 450678 371673
  4092.  
  4093. commit 8feb733ed1dcb1cc94df3b0e6c98009832ea85cc [revision 2614]
  4094. Author: Martin Storsjö <[email protected]>
  4095. AuthorDate: Tue Aug 25 14:38:11 2015 +0300
  4096. Commit: Henrik Gramner <[email protected]>
  4097. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4098.  
  4099. arm: Add neon versions of vsad, asd8 and ssd_nv12_core
  4100.  
  4101. These are straight translations of the aarch64 versions.
  4102.  
  4103. checkasm timing Cortex-A7 A8 A9
  4104. vsad_c 16234 10984 9850
  4105. vsad_neon 2132 1020 789
  4106.  
  4107. asd8_c 5859 3561 3543
  4108. asd8_neon 1407 1279 1250
  4109.  
  4110. ssd_nv12_c 608096 591072 426285
  4111. ssd_nv12_neon 72752 33549 41347
  4112.  
  4113. commit 42b3b398664349d23b2122ac940417165424542d [revision 2613]
  4114. Author: Martin Storsjö <[email protected]>
  4115. AuthorDate: Tue Aug 25 14:38:10 2015 +0300
  4116. Commit: Henrik Gramner <[email protected]>
  4117. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4118.  
  4119. checkasm: Check the right output range for integral_initXh
  4120.  
  4121. These functions write their output into sum+stride, while we previously
  4122. only checked [0..stride-8] within the sum array.
  4123.  
  4124. This catches the previously broken aarch64 version of these functions.
  4125.  
  4126. Also check up until stride-4 elements for init4h.
  4127.  
  4128. commit 3d86abab097fa26d116112f188458269c6a0415f [revision 2612]
  4129. Author: Janne Grunau <[email protected]>
  4130. AuthorDate: Thu Aug 20 13:55:54 2015 +0200
  4131. Commit: Henrik Gramner <[email protected]>
  4132. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4133.  
  4134. aarch64: Skip deblocking in 264_deblock_h_chroma_422_neon
  4135.  
  4136. If the parameters (alpha, beta, tc0[]) indicated that the deblocking
  4137. should have been skipped, every 2nd chrome line would have deblocked
  4138. anyway.
  4139.  
  4140. deblock_h_chroma_422_neon: 2259 (before)
  4141. deblock_h_chroma_422_neon: 2192 (after)
  4142.  
  4143. commit aec81efd3fe43008551916aa6073eb0732a58210 [revision 2611]
  4144. Author: Janne Grunau <[email protected]>
  4145. AuthorDate: Mon Aug 17 16:39:20 2015 +0200
  4146. Commit: Henrik Gramner <[email protected]>
  4147. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4148.  
  4149. aarch64: Optimize various intra_predict asm functions
  4150.  
  4151. Make them at least as fast as the compiled C version (tested on
  4152. cortex-a53 vs. gcc 4.9.2).
  4153.  
  4154. C NEON (before) NEON (after)
  4155. intra_predict_4x4_dc: 260 335 260
  4156. intra_predict_4x4_dct: 210 265 200
  4157. intra_predict_8x8c_dc: 497 548 493
  4158. intra_predict_8x8c_v: 232 309 179 (arm64)
  4159. intra_predict_8x16c_dc: 795 830 790
  4160.  
  4161. commit b16268ac0826d78455d0d704ea0fc8b1edc6b6bf [revision 2610]
  4162. Author: Janne Grunau <[email protected]>
  4163. AuthorDate: Tue Aug 18 10:25:10 2015 +0200
  4164. Commit: Henrik Gramner <[email protected]>
  4165. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4166.  
  4167. aarch64: Faster intra_predict_4x4_h
  4168.  
  4169. Use multiplication with 0x01010101 for splats.
  4170.  
  4171. On a cortex-a53:
  4172. gcc 4.9.2 llvm 3.6 neon (before) neon (after)
  4173. intra_predict_4x4_h: 162 147 160/155 139/135
  4174.  
  4175. commit f2a6be92e5e42e8ef1daf74f63dbdbc4819d2070 [revision 2609]
  4176. Author: Janne Grunau <[email protected]>
  4177. AuthorDate: Tue Aug 18 10:25:09 2015 +0200
  4178. Commit: Henrik Gramner <[email protected]>
  4179. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4180.  
  4181. aarch64: Fix coeff_level_run* macros with LLVM's assembler
  4182.  
  4183. LLVM's integrated assembler does not treat symbols as integer constants.
  4184.  
  4185. commit 592e92e9a8e47c3f0d0017c8158df5a4830e0bbd [revision 2608]
  4186. Author: Janne Grunau <[email protected]>
  4187. AuthorDate: Tue Aug 18 10:25:08 2015 +0200
  4188. Commit: Henrik Gramner <[email protected]>
  4189. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4190.  
  4191. aarch64: Remove commas LLVM's assembler complains about
  4192.  
  4193. commit 6efb57ada652fd015ec4cacffd09282632bb975b [revision 2607]
  4194. Author: Martin Storsjö <[email protected]>
  4195. AuthorDate: Thu Aug 13 23:59:31 2015 +0300
  4196. Commit: Henrik Gramner <[email protected]>
  4197. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4198.  
  4199. arm: Implement x264_sub8x16_dct_dc_neon
  4200.  
  4201. checkasm timing Cortex-A7 A8 A9
  4202. sub8x16_dct_dc_c 6386 3901 4080
  4203. sub8x16_dct_dc_neon 1491 698 917
  4204.  
  4205. commit 89439b2c604c81e13eb3da9e692d2cdae5a18b53 [revision 2606]
  4206. Author: Martin Storsjö <[email protected]>
  4207. AuthorDate: Thu Aug 13 23:59:28 2015 +0300
  4208. Commit: Henrik Gramner <[email protected]>
  4209. CommitDate: Sun Oct 11 18:44:54 2015 +0200
  4210.  
  4211. arm: Optimize x264_deblock_h_chroma_neon
  4212.  
  4213. Shuffle both chroma components together as a 16 bit unit, and
  4214. don't write the unchanged columns (like in x264_deblock_h_luma_neon
  4215. and in the aarch64 version of the function).
  4216.  
  4217. This causes a minor slowdown for x264_deblock_v_chroma_neon, but
  4218. it is negligible compared to the speedup.
  4219.  
  4220. checkasm timing Cortex-A7 A8 A9
  4221. deblock_chroma[1]_c 4817 4057 3601
  4222. deblock_chroma[1]_neon 1249 716 817 (before)
  4223. deblock_chroma[1]_neon 1249 766 845 (after)
  4224.  
  4225. deblock_h_chroma_420_c 3699 3275 2830
  4226. deblock_h_chroma_420_neon 2068 1414 1400 (before)
  4227. deblock_h_chroma_420_neon 1838 1355 1291 (after)
  4228.  
  4229. commit ff71457d71c5c11ed825d848677cab09c7639012 [revision 2605]
  4230. Author: Martin Storsjö <[email protected]>
  4231. AuthorDate: Thu Aug 13 23:59:27 2015 +0300
  4232. Commit: Henrik Gramner <[email protected]>
  4233. CommitDate: Sun Oct 11 18:44:53 2015 +0200
  4234.  
  4235. aarch64: Remove leftover commented out code
  4236.  
  4237. commit ef6034812162fc8b51bfd5e87387f405d1cc30cb [revision 2604]
  4238. Author: Martin Storsjö <[email protected]>
  4239. AuthorDate: Thu Aug 13 23:59:26 2015 +0300
  4240. Commit: Henrik Gramner <[email protected]>
  4241. CommitDate: Sun Oct 11 18:44:53 2015 +0200
  4242.  
  4243. aarch64: Simplify the decimate_score functions
  4244.  
  4245. After doing a left shift by the number of bits returned by clz,
  4246. only bits set to zero can be shifted out, so if the register
  4247. was nonzero to start with (which is checked), it can't become
  4248. zero here.
  4249.  
  4250. commit d2b04a26b26d02c41ffb05cf1a605dafe9e6fa59 [revision 2603]
  4251. Author: Martin Storsjö <[email protected]>
  4252. AuthorDate: Thu Aug 13 23:59:25 2015 +0300
  4253. Commit: Henrik Gramner <[email protected]>
  4254. CommitDate: Sun Oct 11 18:44:53 2015 +0200
  4255.  
  4256. arm: Use aligned loads in x264_coeff_last15_neon
  4257.  
  4258. After subtracting 2, the pointer will be aligned.
  4259.  
  4260. checkasm timing Cortex-A7 A8 A9
  4261. coeff_last15_c 423 375 230
  4262. coeff_last15_neon 350 420 404 (before)
  4263. coeff_last15_neon 350 400 394 (after)
  4264.  
  4265. commit 3f89a6bbee061cb0361770cf5b8495448515a011 [revision 2602]
  4266. Author: Martin Storsjö <[email protected]>
  4267. AuthorDate: Thu Aug 13 23:59:24 2015 +0300
  4268. Commit: Henrik Gramner <[email protected]>
  4269. CommitDate: Sun Oct 11 18:44:53 2015 +0200
  4270.  
  4271. arm: Simplify x264_predict_8x8c_p_neon
  4272.  
  4273. This gets rid of a few unnecessary (and confusing) steps in
  4274. calculating the increment to i00.
  4275.  
  4276. checkasm timing Cortex-A7 A8 A9
  4277. intra_predict_8x8c_p_c 5525 4732 4755
  4278. intra_predict_8x8c_p_neon 1719 1140 1262 (before)
  4279. intra_predict_8x8c_p_neon 1663 1142 1255 (after)
  4280.  
  4281. commit a0cd7d38acb6c31973228ab207e18344920e0aa3 [revision 2601]
  4282. Author: Vittorio Giovara <[email protected]>
  4283. AuthorDate: Tue Sep 15 15:40:14 2015 +0200
  4284. Commit: Henrik Gramner <[email protected]>
  4285. CommitDate: Sun Oct 11 18:44:53 2015 +0200
  4286.  
  4287. lavf: Use the prefixed name for pixel format enum
  4288.  
  4289. commit 63555e696a997ff795798d3357d770f8ab373cd9 [revision 2600]
  4290. Author: Janne Grunau <[email protected]>
  4291. AuthorDate: Thu Sep 3 00:21:58 2015 +0200
  4292. Commit: Henrik Gramner <[email protected]>
  4293. CommitDate: Thu Sep 3 16:29:15 2015 +0200
  4294.  
  4295. aarch64: fix x264_mbtree_propagate_cost_neon
  4296.  
  4297. The branch conditon caused the loop to execute one time more than
  4298. intended. Detected by a memory corruption on arm with the 1 to 1 port of
  4299. the function.
  4300.  
  4301. commit 5c4728d8dd82ba46901824470db1609ae0f2521d [revision 2599]
  4302. Author: Martin Storsjö <[email protected]>
  4303. AuthorDate: Thu Aug 13 23:59:22 2015 +0300
  4304. Commit: Henrik Gramner <[email protected]>
  4305. CommitDate: Thu Aug 27 19:56:58 2015 +0200
  4306.  
  4307. aarch64: Fix integral_init4/8h_neon
  4308.  
  4309. The stride is the number of uint16_t elements and thus needs
  4310. to be shifted.
  4311.  
  4312. This issue had slipped unnoticed since checkasm didn't actually
  4313. verify the output of these functions.
  4314.  
  4315. commit 67076513267907b5601828ae6864cc063c8c7548 [revision 2598]
  4316. Author: Henrik Gramner <[email protected]>
  4317. AuthorDate: Thu Aug 27 19:53:00 2015 +0200
  4318. Commit: Henrik Gramner <[email protected]>
  4319. CommitDate: Thu Aug 27 19:56:58 2015 +0200
  4320.  
  4321. x86: Fix integral_init4/8h_avx2
  4322.  
  4323. The AVX2 implementation was using the wrong offsets. It went undetected due to
  4324. the checkasm test being incorrect.
  4325.  
  4326. commit e86f3a1993234e8f26050c243aa253651200fa6b [revision 2597]
  4327. Author: Mark Webster <[email protected]>
  4328. AuthorDate: Wed Aug 5 04:28:17 2015 +0100
  4329. Commit: Anton Mitrofanov <[email protected]>
  4330. CommitDate: Wed Aug 19 01:24:30 2015 +0300
  4331.  
  4332. Simplify inclusion of x264.h in C++ projects
  4333.  
  4334. Name all structs to support forward declarations.
  4335. Add a conditional extern "C" wrapper in x264.h itself instead of having to
  4336. specify it in every location where it's included.
  4337.  
  4338. commit 401941cc7099b322864600b62104940542497e7a [revision 2596]
  4339. Author: Henrik Gramner <[email protected]>
  4340. AuthorDate: Sun Aug 16 21:59:26 2015 +0200
  4341. Commit: Anton Mitrofanov <[email protected]>
  4342. CommitDate: Wed Aug 19 01:24:29 2015 +0300
  4343.  
  4344. checkasm: Properly save rdx/edx in checkasm_call() on x86
  4345.  
  4346. If the return value doesn't fit in a single register rdx/edx can in some
  4347. cases be used in addition to rax/eax.
  4348.  
  4349. Doesn't affect any of the existing checkasm tests but it's more correct
  4350. behavior and it might be useful in the future.
  4351.  
  4352. commit 3dff8af3033a9e81d7966c5749fd361ce421467a [revision 2595]
  4353. Author: Henrik Gramner <[email protected]>
  4354. AuthorDate: Tue Aug 11 17:19:35 2015 +0200
  4355. Commit: Anton Mitrofanov <[email protected]>
  4356. CommitDate: Wed Aug 19 01:24:27 2015 +0300
  4357.  
  4358. x86: Enable SSE2 by default on x86-32
  4359.  
  4360. It makes more sense to tune the defaults to benefit the vast majority of users.
  4361.  
  4362. Anyone still using a Pentium III for video encoding is of course free to
  4363. explicitly set different flags when compiling.
  4364.  
  4365. commit 51d8aa09b777dc2969deaa954d5f6af9836c02ba [revision 2594]
  4366. Author: Henrik Gramner <[email protected]>
  4367. AuthorDate: Mon Aug 10 22:30:21 2015 +0200
  4368. Commit: Anton Mitrofanov <[email protected]>
  4369. CommitDate: Wed Aug 19 01:24:26 2015 +0300
  4370.  
  4371. msvs/icl: Improve default CFLAGS
  4372.  
  4373. Use -fp:fast as a substitute for -ffast-math.
  4374. Increase warning level from -W0 to -W1 (the default setting).
  4375. Disable -GS (stack cookies) on MSVS. It's disabled by default on ICL.
  4376.  
  4377. commit 7edaf4b966aaee098ff301436f8d2b33a6fe5983 [revision 2593]
  4378. Author: Henrik Gramner <[email protected]>
  4379. AuthorDate: Wed Aug 12 22:23:31 2015 +0200
  4380. Commit: Anton Mitrofanov <[email protected]>
  4381. CommitDate: Wed Aug 19 01:00:26 2015 +0300
  4382.  
  4383. Use a relative $SRCPATH for out-of-tree builds
  4384.  
  4385. Fixes out-of-tree MSVS builds on Cygwin.
  4386.  
  4387. commit e7b4b863dc2555ed835569c400d3a30f7ddc15ff [revision 2592]
  4388. Author: Henrik Gramner <[email protected]>
  4389. AuthorDate: Sat Aug 8 22:26:38 2015 +0200
  4390. Commit: Anton Mitrofanov <[email protected]>
  4391. CommitDate: Wed Aug 19 01:00:24 2015 +0300
  4392.  
  4393. cygwin: Enable MSVS support
  4394.  
  4395. `cl -showIncludes` creates absolute Windows paths for some files, attempt
  4396. to convert those to Unix paths.
  4397.  
  4398. Use relative paths for dependencies located in or below the working directory
  4399. in order to mimic the behavior of gcc and to make the paths more readable.
  4400.  
  4401. Make the dependency generation script a bit more robust in general.
  4402.  
  4403. commit 817a4414b98e8a511c626932e7d433388bc96507 [revision 2591]
  4404. Author: Henrik Gramner <[email protected]>
  4405. AuthorDate: Sat Aug 8 18:34:21 2015 +0200
  4406. Commit: Anton Mitrofanov <[email protected]>
  4407. CommitDate: Wed Aug 19 01:00:23 2015 +0300
  4408.  
  4409. cltostr.sh: Minor fixes
  4410.  
  4411. commit 1a3d963441eaad25972763423d60158f597c5f65 [revision 2590]
  4412. Author: Henrik Gramner <[email protected]>
  4413. AuthorDate: Sat Aug 8 12:21:54 2015 +0200
  4414. Commit: Anton Mitrofanov <[email protected]>
  4415. CommitDate: Wed Aug 19 01:00:22 2015 +0300
  4416.  
  4417. Simplify version.sh
  4418.  
  4419. Also remove some non-POSIX syntax and improve robustness.
  4420.  
  4421. As a bonus the script now runs about 2-3 times faster.
  4422.  
  4423. `git rev-list --count` could be used to simplify things even further,
  4424. but that functionality was added in git 1.7.2 so keep `wc -l` for now
  4425. to maintain compatibility with older git versions.
  4426.  
  4427. commit f7f6af76ef22e812ef330e2839488e83dd553836 [revision 2589]
  4428. Author: 장영훈 <[email protected]>
  4429. AuthorDate: Fri Aug 7 14:43:24 2015 +0900
  4430. Commit: Anton Mitrofanov <[email protected]>
  4431. CommitDate: Wed Aug 19 01:00:21 2015 +0300
  4432.  
  4433. msvs: Fix cl detection in non-English environments
  4434.  
  4435. commit e1a55bbbff2b4460ceb843f163e349fed7d32969 [revision 2588]
  4436. Author: Henrik Gramner <[email protected]>
  4437. AuthorDate: Mon Aug 3 21:05:11 2015 +0200
  4438. Commit: Anton Mitrofanov <[email protected]>
  4439. CommitDate: Wed Aug 19 01:00:20 2015 +0300
  4440.  
  4441. x86inc: Sync minor changes from ffmpeg/libav
  4442.  
  4443. commit 36f537b141da076032fd11f1745bb62d466dd7bf [revision 2587]
  4444. Author: Henrik Gramner <[email protected]>
  4445. AuthorDate: Wed Jul 29 19:30:52 2015 +0200
  4446. Commit: Anton Mitrofanov <[email protected]>
  4447. CommitDate: Wed Aug 19 01:00:19 2015 +0300
  4448.  
  4449. matroska: Add comments for the remaining element names
  4450.  
  4451. commit f04062e6380cbe10453dab33a3575c373e63ff9b [revision 2586]
  4452. Author: Henrik Gramner <[email protected]>
  4453. AuthorDate: Wed Jul 29 19:30:41 2015 +0200
  4454. Commit: Anton Mitrofanov <[email protected]>
  4455. CommitDate: Wed Aug 19 01:00:18 2015 +0300
  4456.  
  4457. Silence various static analyzer warnings
  4458.  
  4459. Those are false positives, but it doesn't hurt to get rid of them.
  4460.  
  4461. commit b1cbf7ebe4a192bbc25cc910cb2910a34992f807 [revision 2585]
  4462. Author: Henrik Gramner <[email protected]>
  4463. AuthorDate: Sun Jul 26 23:13:29 2015 +0200
  4464. Commit: Anton Mitrofanov <[email protected]>
  4465. CommitDate: Wed Aug 19 01:00:16 2015 +0300
  4466.  
  4467. mingw: Enable the tsaware linker flag
  4468.  
  4469. Avoids an irrelevant compatibility layer in Terminal Services environments.
  4470.  
  4471. https://msdn.microsoft.com/en-us/library/cc834995.aspx
  4472.  
  4473. commit 8a1ff031ecd4b423fc373540b9b68cdf97602bbf [revision 2584]
  4474. Author: Henrik Gramner <[email protected]>
  4475. AuthorDate: Sun Jul 26 23:13:26 2015 +0200
  4476. Commit: Anton Mitrofanov <[email protected]>
  4477. CommitDate: Wed Aug 19 01:00:15 2015 +0300
  4478.  
  4479. msvs: Don't redefine snprintf for VS2015
  4480.  
  4481. Visual Studio 2015 has a proper snprintf implementation.
  4482.  
  4483. commit aa9d22927c0264c08c11c9e72294fc651a155b3e [revision 2583]
  4484. Author: Henrik Gramner <[email protected]>
  4485. AuthorDate: Sun Jul 26 23:13:19 2015 +0200
  4486. Commit: Anton Mitrofanov <[email protected]>
  4487. CommitDate: Wed Aug 19 01:00:14 2015 +0300
  4488.  
  4489. msvs: Prefer link.exe from the same directory as cl.exe
  4490.  
  4491. /usr/bin/link from coreutils may be located before the MSVS linker in $PATH
  4492. which causes linking to fail due to using the wrong binary.
  4493.  
  4494. commit ca8bd68063d74227d917f34fd50942265f9a106c [revision 2582]
  4495. Author: Henrik Gramner <[email protected]>
  4496. AuthorDate: Mon Jul 27 00:10:00 2015 +0200
  4497. Commit: Anton Mitrofanov <[email protected]>
  4498. CommitDate: Wed Aug 19 01:00:13 2015 +0300
  4499.  
  4500. frame_dump: check fseek() return value
  4501.  
  4502. commit 53b3b747e22f53204f6efb5106ab4a5a8eb57626 [revision 2581]
  4503. Author: Henrik Gramner <[email protected]>
  4504. AuthorDate: Mon Jul 27 00:08:38 2015 +0200
  4505. Commit: Anton Mitrofanov <[email protected]>
  4506. CommitDate: Wed Aug 19 00:55:29 2015 +0300
  4507.  
  4508. x264_vfprintf: use va_copy
  4509.  
  4510. It's undefined behavior to use the same va_list twice.
  4511.  
  4512. This most likely didn't cause any issues in practice since the string would
  4513. have to be larger than 4 KiB to trigger the fallback path.
  4514.  
  4515. Use workaround for ICL as it doesn't define va_copy even for C99.
  4516.  
  4517. commit 59e7ded846a832125cb533aadff9895487771ea7 [revision 2580]
  4518. Author: Henrik Gramner <[email protected]>
  4519. AuthorDate: Mon Jul 27 00:08:31 2015 +0200
  4520. Commit: Henrik Gramner <[email protected]>
  4521. CommitDate: Mon Jul 27 00:08:31 2015 +0200
  4522.  
  4523. param_parse: Fix framerate rounding issues
  4524.  
  4525. commit 73ae2d11d472d0eb3b7c218dc1659db32f649b14 [revision 2579]
  4526. Author: Marcin Juszkiewicz <[email protected]>
  4527. AuthorDate: Mon Jun 1 11:24:45 2015 +0200
  4528. Commit: Henrik Gramner <[email protected]>
  4529. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4530.  
  4531. aarch64: Remove broken CFLAGS in configure
  4532.  
  4533. GCC doesn't have an "-arch" switch, but works when that entire line is removed.
  4534.  
  4535. commit cc002bd545b008b1cdc7c6d7cc0c616ba125d4d5 [revision 2578]
  4536. Author: Rong Yan <[email protected]>
  4537. AuthorDate: Mon Jul 20 03:34:20 2015 -0500
  4538. Commit: Henrik Gramner <[email protected]>
  4539. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4540.  
  4541. ppc: Add little-endian PowerPC support
  4542.  
  4543. commit 145f3a6275802a649b8dedb49bb0e054caf31717 [revision 2577]
  4544. Author: Rishikesh More <[email protected]>
  4545. AuthorDate: Thu Jun 18 17:48:46 2015 +0530
  4546. Commit: Henrik Gramner <[email protected]>
  4547. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4548.  
  4549. mips: MSA quant optimizations
  4550.  
  4551. Signed-off-by: Rishikesh More <[email protected]>
  4552.  
  4553. commit 16395d2b6f827b076612eb5b70711b79621da67e [revision 2576]
  4554. Author: Rishikesh More <[email protected]>
  4555. AuthorDate: Thu Jun 18 17:48:45 2015 +0530
  4556. Commit: Henrik Gramner <[email protected]>
  4557. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4558.  
  4559. mips: MSA predict optimizations
  4560.  
  4561. Signed-off-by: Rishikesh More <[email protected]>
  4562.  
  4563. commit 204e1a60237e0b3168ccbdb2905c9af8188b90ee [revision 2575]
  4564. Author: Rishikesh More <[email protected]>
  4565. AuthorDate: Thu Jun 18 17:48:44 2015 +0530
  4566. Commit: Henrik Gramner <[email protected]>
  4567. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4568.  
  4569. mips: MSA pixel optimizations
  4570.  
  4571. Signed-off-by: Rishikesh More <[email protected]>
  4572.  
  4573. commit 3ce6430eb11839c69d606c59c0f8c31ce0b6dd17 [revision 2574]
  4574. Author: Rishikesh More <[email protected]>
  4575. AuthorDate: Thu Jun 18 17:48:43 2015 +0530
  4576. Commit: Henrik Gramner <[email protected]>
  4577. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4578.  
  4579. mips: MSA deblock optimizations
  4580.  
  4581. Signed-off-by: Rishikesh More <[email protected]>
  4582.  
  4583. commit 57618eead025eaf654226add94689d6d2999ccf6 [revision 2573]
  4584. Author: Rishikesh More <[email protected]>
  4585. AuthorDate: Thu Jun 18 17:48:42 2015 +0530
  4586. Commit: Henrik Gramner <[email protected]>
  4587. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4588.  
  4589. mips: MSA dct optimizations
  4590.  
  4591. Signed-off-by: Rishikesh More <[email protected]>
  4592.  
  4593. commit 4ebb23aaf4f46b7a04aa8aefa3c08e7b6493de4c [revision 2572]
  4594. Author: Rishikesh More <[email protected]>
  4595. AuthorDate: Thu Jun 18 17:48:40 2015 +0530
  4596. Commit: Henrik Gramner <[email protected]>
  4597. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4598.  
  4599. mips: MSA mc optimizations
  4600.  
  4601. Signed-off-by: Rishikesh More <[email protected]>
  4602.  
  4603. commit cd19444d3f9915a5a33a95e308bc8021d7e62afe [revision 2571]
  4604. Author: Rishikesh More <[email protected]>
  4605. AuthorDate: Thu Jun 18 17:48:38 2015 +0530
  4606. Commit: Henrik Gramner <[email protected]>
  4607. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4608.  
  4609. mips: Common MSA macros
  4610.  
  4611. Add macros for load/store, slide, shift, transpose and basic arithmetic
  4612. operations required by subsequent patches.
  4613.  
  4614. Signed-off-by: Rishikesh More <[email protected]>
  4615.  
  4616. commit 72b82bd98a99b1d75322b70a74365547382ce062 [revision 2570]
  4617. Author: Rishikesh More <[email protected]>
  4618. AuthorDate: Tue May 12 19:38:09 2015 +0530
  4619. Commit: Henrik Gramner <[email protected]>
  4620. CommitDate: Sat Jul 25 22:52:55 2015 +0200
  4621.  
  4622. mips: Add MSA support to checkasm
  4623.  
  4624. Signed-off-by: Rishikesh More <[email protected]>
  4625.  
  4626. commit ce0757d9d2778e349a7c2f6445b6aa75d8765c30 [revision 2569]
  4627. Author: Kaustubh Raste <[email protected]>
  4628. AuthorDate: Fri Apr 17 17:38:58 2015 +0530
  4629. Commit: Henrik Gramner <[email protected]>
  4630. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4631.  
  4632. mips: Initial MSA support
  4633.  
  4634. MSA is the MIPS SIMD Architecture.
  4635.  
  4636. Add X264_CPU_MSA define.
  4637. Update configure to detect MIPS platform and set flags.
  4638. CPU-specific gcc options are expected through --extra-cflags.
  4639.  
  4640. Sample command line for mips32r5:
  4641. ./configure --host=mipsel-linux-gnu --cross-prefix=<TOOLCHAIN>/mips-mti-linux-gnu-
  4642. --extra-cflags="-EL -mips32r5 -msched-weight -mload-store-pairs"
  4643.  
  4644. Signed-off-by: Kaustubh Raste <[email protected]>
  4645.  
  4646. commit 9140ee1fb39bd4a4ccace28091398e8a96704f07 [revision 2568]
  4647. Author: Anton Mitrofanov <[email protected]>
  4648. AuthorDate: Fri Jul 17 00:22:29 2015 +0300
  4649. Commit: Henrik Gramner <[email protected]>
  4650. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4651.  
  4652. Limit autodetection of threads number according to the source height
  4653.  
  4654. commit aeaed2d07b5b43437bb640e1f987d42a6fab03b9 [revision 2567]
  4655. Author: Anton Mitrofanov <[email protected]>
  4656. AuthorDate: Thu Jul 16 19:04:59 2015 +0300
  4657. Commit: Henrik Gramner <[email protected]>
  4658. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4659.  
  4660. Fine-tune of frame's size predictors at ratecontrol start
  4661.  
  4662. This is attempt to improve VBV at start of video with a lot of threads which
  4663. delay feedback for predictors.
  4664.  
  4665. commit aa275158641e94203003157947d43ff4cc685068 [revision 2566]
  4666. Author: Anton Mitrofanov <[email protected]>
  4667. AuthorDate: Thu Jul 16 16:15:56 2015 +0300
  4668. Commit: Henrik Gramner <[email protected]>
  4669. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4670.  
  4671. Use forced frame types in slicetype analysis
  4672.  
  4673. This should improve MBTree and VBV when a lot of forced frame types are used.
  4674.  
  4675. commit a83edfa053f60ad0c8a164f31e7492a680eef361 [revision 2565]
  4676. Author: Henrik Gramner <[email protected]>
  4677. AuthorDate: Mon Dec 1 22:05:42 2014 +0100
  4678. Commit: Henrik Gramner <[email protected]>
  4679. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4680.  
  4681. x86: SSSE3 and AVX2 implementations of plane_copy_swap
  4682.  
  4683. For NV21 input.
  4684.  
  4685. commit 627f891c571cacb51deb5e211b23c309b14a6587 [revision 2564]
  4686. Author: Yu Xiaolei <[email protected]>
  4687. AuthorDate: Fri Jun 6 16:05:27 2014 +0800
  4688. Commit: Henrik Gramner <[email protected]>
  4689. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4690.  
  4691. NV21 input support
  4692.  
  4693. Eliminates an extra copy when encoding Android camera preview images.
  4694.  
  4695. Checkasm test by Janne Grunau.
  4696. ARM assembly with improvements from Janne Grunau.
  4697.  
  4698. commit 6ee94dc898dc029553e308f1e76891ccefb3f0a7 [revision 2563]
  4699. Author: Henrik Gramner <[email protected]>
  4700. AuthorDate: Tue Jun 23 17:00:47 2015 +0200
  4701. Commit: Henrik Gramner <[email protected]>
  4702. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4703.  
  4704. deblock: Write combining
  4705.  
  4706. commit 08a9c51919f4edbd6e484155e5521a92a0800651 [revision 2562]
  4707. Author: Henrik Gramner <[email protected]>
  4708. AuthorDate: Tue Jun 23 14:59:59 2015 +0200
  4709. Commit: Henrik Gramner <[email protected]>
  4710. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4711.  
  4712. Get rid of some tabs and trailing whitespaces
  4713.  
  4714. commit b568a256b9bc6c500d7b1ffe4b9c3311ee5ff337 [revision 2561]
  4715. Author: Henrik Gramner <[email protected]>
  4716. AuthorDate: Sat May 23 19:44:16 2015 +0200
  4717. Commit: Henrik Gramner <[email protected]>
  4718. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4719.  
  4720. x86: Experimental nasm support
  4721.  
  4722. Enables the use of nasm as an alternative to yasm.
  4723.  
  4724. Note that nasm cannot assemble x264 with PIC enabled since it currently doesn't
  4725. support [symbol-$$] addressing which is used extensively by x264's PIC code.
  4726. This includes all 64-bit Windows and 64-bit OS X builds, even non-shared.
  4727.  
  4728. For the above reason nasm is currently intentionally not auto-detected, instead
  4729. the assembler must be explicitly specified using "AS=nasm ./configure".
  4730.  
  4731. Also drop -O2 from ASFLAGS since it's simply ignored anyway.
  4732.  
  4733. commit d14e38c059c9a2aecc82477b99d56ef74eb731ec [revision 2560]
  4734. Author: Timothy Gu <[email protected]>
  4735. AuthorDate: Tue May 26 19:12:42 2015 +0200
  4736. Commit: Henrik Gramner <[email protected]>
  4737. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4738.  
  4739. x86inc: Prevent warnings when using `struc` and `endstruc`
  4740.  
  4741. struc and endstruc attempts to revert to the previous section state set by
  4742. the SECTION macro.
  4743.  
  4744. Use the primitive [SECTION] directive instead of the SECTION macro for the
  4745. .note.GNU-stack section to prevent it from being emitted again during endstruc.
  4746.  
  4747. commit 353b1f888c34081e94727a1ffa0e4920e2cfe8a9 [revision 2559]
  4748. Author: Henrik Gramner <[email protected]>
  4749. AuthorDate: Wed May 27 21:38:14 2015 +0200
  4750. Commit: Henrik Gramner <[email protected]>
  4751. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4752.  
  4753. x86inc: Drop SECTION_TEXT macro
  4754.  
  4755. The .text section is already 16-byte aligned by default on all supported
  4756. platforms so `SECTION_TEXT` isn't any different from `SECTION .text`.
  4757.  
  4758. commit b615f82e45c88b7915c5571ad09fa65a0b6130d7 [revision 2558]
  4759. Author: Henrik Gramner <[email protected]>
  4760. AuthorDate: Sat May 23 13:38:05 2015 +0200
  4761. Commit: Henrik Gramner <[email protected]>
  4762. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4763.  
  4764. x86inc: Disable vpbroadcastq workaround in newer yasm versions
  4765.  
  4766. The bug was fixed in 1.3.0, so only perform the workaround in earlier versions.
  4767.  
  4768. commit 8f834d6ccc054d8c32d84310664dc07abac553ec [revision 2557]
  4769. Author: Henrik Gramner <[email protected]>
  4770. AuthorDate: Sun May 24 22:57:00 2015 +0200
  4771. Commit: Henrik Gramner <[email protected]>
  4772. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4773.  
  4774. Prefer Unicode versions of Windows API calls
  4775.  
  4776. Just for consistency, doesn't affect behavior.
  4777.  
  4778. commit 3f8c8eb1758d0fa890538eba6f5e699c93dc1304 [revision 2556]
  4779. Author: Henrik Gramner <[email protected]>
  4780. AuthorDate: Sun May 24 23:21:20 2015 +0200
  4781. Commit: Henrik Gramner <[email protected]>
  4782. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4783.  
  4784. Get rid of fPIC warnings when compiling a shared library on Windows
  4785.  
  4786. PIC is always enabled when compiling for Windows so gcc complains when using
  4787. -fPIC since it doesn't do anything.
  4788.  
  4789. commit 0c21480fa2fdee345a3049e2169624dc6fc2acfc [revision 2555]
  4790. Author: Henrik Gramner <[email protected]>
  4791. AuthorDate: Sat Jul 25 22:42:59 2015 +0200
  4792. Commit: Henrik Gramner <[email protected]>
  4793. CommitDate: Sat Jul 25 22:52:54 2015 +0200
  4794.  
  4795. matroska: Write the correct DocTypeVersion when using frame-packing
  4796.  
  4797. The StereoMode element is only valid with DocTypeVersion 3 or higher.
  4798.  
  4799. commit 791d265281af1d022a72ba9e003a987e97da5c0d [revision 2554]
  4800. Author: Anton Mitrofanov <[email protected]>
  4801. AuthorDate: Sat Jul 25 00:21:52 2015 +0300
  4802. Commit: Anton Mitrofanov <[email protected]>
  4803. CommitDate: Sat Jul 25 00:43:53 2015 +0300
  4804.  
  4805. dump_yuv: Fix file handle leak
  4806.  
  4807. commit d6aa586b2f83eeb776744c2e97a8ce9e1181c59b [revision 2553]
  4808. Author: Anton Mitrofanov <[email protected]>
  4809. AuthorDate: Sat Jul 25 00:20:47 2015 +0300
  4810. Commit: Anton Mitrofanov <[email protected]>
  4811. CommitDate: Sat Jul 25 00:43:38 2015 +0300
  4812.  
  4813. mp4: Fix file handle leak
  4814.  
  4815. commit 942e4e4530d0909c2b580be88acd18d1e5fa4fa8 [revision 2552]
  4816. Author: Henrik Gramner <[email protected]>
  4817. AuthorDate: Wed Jun 24 00:40:45 2015 +0200
  4818. Commit: Anton Mitrofanov <[email protected]>
  4819. CommitDate: Sat Jul 25 00:39:38 2015 +0300
  4820.  
  4821. flv: Check fseek() and fwrite() return values
  4822.  
  4823. commit 250d5b0e13045f6a1ebfeb379933b5c5daa9cf41 [revision 2551]
  4824. Author: Henrik Gramner <[email protected]>
  4825. AuthorDate: Wed Jun 24 00:22:56 2015 +0200
  4826. Commit: Anton Mitrofanov <[email protected]>
  4827. CommitDate: Sat Jul 25 00:39:37 2015 +0300
  4828.  
  4829. flv: Fix memory and file handle leaks
  4830.  
  4831. commit 3533520655ef095ef009af9b6b27a20b45fd13ee [revision 2550]
  4832. Author: Henrik Gramner <[email protected]>
  4833. AuthorDate: Wed Jun 24 01:23:35 2015 +0200
  4834. Commit: Anton Mitrofanov <[email protected]>
  4835. CommitDate: Sat Jul 25 00:39:36 2015 +0300
  4836.  
  4837. avs: Fix file handle leak
  4838.  
  4839. commit df152a77e1b17065aecb40c9a2a28d5953887ac9 [revision 2549]
  4840. Author: Henrik Gramner <[email protected]>
  4841. AuthorDate: Tue Jun 23 13:38:02 2015 +0200
  4842. Commit: Anton Mitrofanov <[email protected]>
  4843. CommitDate: Sat Jul 25 00:39:35 2015 +0300
  4844.  
  4845. matroska: Fix memory leak
  4846.  
  4847. commit 6d5249977f5d62f6e167a062bdd94d8546eca1f7 [revision 2548]
  4848. Author: Henrik Gramner <[email protected]>
  4849. AuthorDate: Tue Jun 23 13:24:29 2015 +0200
  4850. Commit: Anton Mitrofanov <[email protected]>
  4851. CommitDate: Sat Jul 25 00:39:34 2015 +0300
  4852.  
  4853. rdo: Fix potential CAVLC overflow issues
  4854.  
  4855. commit 936e8da1a4f9d0431b181d0877bb1602d4de9441 [revision 2547]
  4856. Author: Henrik Gramner <[email protected]>
  4857. AuthorDate: Tue Jun 23 22:08:35 2015 +0200
  4858. Commit: Anton Mitrofanov <[email protected]>
  4859. CommitDate: Sat Jul 25 00:39:31 2015 +0300
  4860.  
  4861. slurp_file: Various minor bug fixes
  4862.  
  4863. * Fix unsigned <= 0 check.
  4864. * Add additional size sanity check on 32-bit systems.
  4865. * Don't read uninitialized data if fread() fails.
  4866.  
  4867. commit d302526d5b97818f588b86f408f910924790242e [revision 2546]
  4868. Author: Henrik Gramner <[email protected]>
  4869. AuthorDate: Tue Jun 23 22:47:53 2015 +0200
  4870. Commit: Anton Mitrofanov <[email protected]>
  4871. CommitDate: Sat Jul 25 00:39:30 2015 +0300
  4872.  
  4873. param_parse: Check strdup() return value
  4874.  
  4875. commit 94e476d80b9635508907893c97e8f8d9f0bc9ddf [revision 2545]
  4876. Author: Henrik Gramner <[email protected]>
  4877. AuthorDate: Tue Jun 23 15:38:16 2015 +0200
  4878. Commit: Anton Mitrofanov <[email protected]>
  4879. CommitDate: Fri Jul 17 00:41:27 2015 +0300
  4880.  
  4881. param_parse: Fix memory leak
  4882.  
  4883. commit 45856b9787eab95434d66b4bc2e18819483f0e43 [revision 2544]
  4884. Author: Anton Mitrofanov <[email protected]>
  4885. AuthorDate: Fri Jun 19 16:01:12 2015 +0300
  4886. Commit: Anton Mitrofanov <[email protected]>
  4887. CommitDate: Fri Jul 17 00:41:26 2015 +0300
  4888.  
  4889. Add FreeBSD's stdint.h header guard to allowed list
  4890.  
  4891. Patch written by Koop Mast <[email protected]>
  4892.  
  4893. commit 35cf1a2cbf253e43cab7747eb903a3b844bd42c1 [revision 2543]
  4894. Author: Henrik Gramner <[email protected]>
  4895. AuthorDate: Fri May 22 19:23:33 2015 +0200
  4896. Commit: Anton Mitrofanov <[email protected]>
  4897. CommitDate: Fri Jul 17 00:41:25 2015 +0300
  4898.  
  4899. x86: Prevent overread of src in plane_copy_interleave
  4900.  
  4901. Could only occur in 4:2:2 with height == 1.
  4902.  
  4903. Also enable asm for inputs with different U/V strides as long as the strides
  4904. have identical signs.
  4905.  
  4906. commit 003414a4b3724f0972e4507dfd1432dd442d2228 [revision 2542]
  4907. Author: Anton Mitrofanov <[email protected]>
  4908. AuthorDate: Wed May 20 23:10:20 2015 +0300
  4909. Commit: Anton Mitrofanov <[email protected]>
  4910. CommitDate: Fri Jul 17 00:41:24 2015 +0300
  4911.  
  4912. checkasm: Fix incorrect memcmp size for ARM architecture
  4913.  
  4914. commit e08fdc81018489217f4bafe7321a3baf372fac1f [revision 2541]
  4915. Author: Anton Mitrofanov <[email protected]>
  4916. AuthorDate: Sun Apr 26 20:51:05 2015 +0300
  4917. Commit: Anton Mitrofanov <[email protected]>
  4918. CommitDate: Fri Jul 17 00:39:41 2015 +0300
  4919.  
  4920. Fix possible use of uninitialized MVs in lookahead analysis for B-frames
  4921.  
  4922. commit 0b0210857ef13214f12861dec672006455a556d6 [revision 2540]
  4923. Author: Anton Mitrofanov <[email protected]>
  4924. AuthorDate: Tue Apr 21 23:08:19 2015 +0300
  4925. Commit: Henrik Gramner <[email protected]>
  4926. CommitDate: Sun May 24 11:34:40 2015 +0200
  4927.  
  4928. Catch incorrect usage of libx264 API for delayed frames flushing
  4929.  
  4930. commit 3a6bd39a650b47572743c2d2ea2fd7c214053fb2 [revision 2539]
  4931. Author: Anton Mitrofanov <[email protected]>
  4932. AuthorDate: Sat Mar 7 23:00:09 2015 +0300
  4933. Commit: Henrik Gramner <[email protected]>
  4934. CommitDate: Sun May 24 11:34:40 2015 +0200
  4935.  
  4936. Fix detection of system libx264 configuration
  4937.  
  4938. commit 121396c71b4907ca82301d1a529795d98daab5f8 [revision 2538]
  4939. Author: Anton Mitrofanov <[email protected]>
  4940. AuthorDate: Mon Feb 23 14:23:18 2015 +0300
  4941. Commit: Anton Mitrofanov <[email protected]>
  4942. CommitDate: Mon Feb 23 20:10:21 2015 +0300
  4943.  
  4944. Cosmetic changes
  4945.  
  4946. commit 8e71b432e5dbe835fa4516064f6841a03c79b183 [revision 2537]
  4947. Author: Anton Mitrofanov <[email protected]>
  4948. AuthorDate: Wed Dec 31 02:15:05 2014 +0300
  4949. Commit: Anton Mitrofanov <[email protected]>
  4950. CommitDate: Mon Feb 23 20:09:33 2015 +0300
  4951.  
  4952. Update configure for auto detection of system libx264 configuration
  4953.  
  4954. commit 0f84192e88d6adc4512f6f320a50a09b4608634c [revision 2536]
  4955. Author: Anton Mitrofanov <[email protected]>
  4956. AuthorDate: Tue Feb 3 14:51:28 2015 +0300
  4957. Commit: Anton Mitrofanov <[email protected]>
  4958. CommitDate: Mon Feb 23 20:09:32 2015 +0300
  4959.  
  4960. Add tile format frame packing value
  4961.  
  4962. Defined in 2014-02 edition.
  4963.  
  4964. commit f08b1c6b8e186ff5a931e9a80e8923e42efff0e4 [revision 2535]
  4965. Author: Anton Mitrofanov <[email protected]>
  4966. AuthorDate: Tue Feb 3 13:39:14 2015 +0300
  4967. Commit: Anton Mitrofanov <[email protected]>
  4968. CommitDate: Mon Feb 23 20:09:30 2015 +0300
  4969.  
  4970. Stricter validation of crop-rect values
  4971.  
  4972. commit 196cb9ab52af9370fc66a474ffc4a52a75dc5eb4 [revision 2534]
  4973. Author: Vittorio Giovara <[email protected]>
  4974. AuthorDate: Tue Jan 20 16:15:56 2015 +0000
  4975. Commit: Anton Mitrofanov <[email protected]>
  4976. CommitDate: Mon Feb 23 20:09:29 2015 +0300
  4977.  
  4978. Add mono frame packing value
  4979.  
  4980. Defined in 2013-04 edition.
  4981.  
  4982. commit c8a773ebfca148ef04f5a60d42cbd7336af0baf6 [revision 2533]
  4983. Author: Vittorio Giovara <[email protected]>
  4984. AuthorDate: Tue Jan 20 15:57:41 2015 +0000
  4985. Commit: Anton Mitrofanov <[email protected]>
  4986. CommitDate: Mon Feb 23 13:34:52 2015 +0300
  4987.  
  4988. Validate frame packing value instead of clipping
  4989.  
  4990. commit a95584945dd9ce3acc66c6cd8f6796bc4404d40d [revision 2532]
  4991. Author: Christophe Gisquet <[email protected]>
  4992. AuthorDate: Tue Feb 3 20:40:41 2015 +0100
  4993. Commit: Anton Mitrofanov <[email protected]>
  4994. CommitDate: Mon Feb 23 13:34:51 2015 +0300
  4995.  
  4996. x86inc: Correctly warn on use of SSE2 instructions in SSE functions
  4997.  
  4998. SSE2 instructions that are XMM-implementations of pre-existing MMX/MMX2
  4999. instructions did not issue warnings when used in SSE functions. Handle
  5000. it by also checking the register type when such instructions are used.
  5001.  
  5002. commit 23d4434de9ab5ef32ebb03401d971b8579a65fc6 [revision 2531]
  5003. Author: Christophe Gisquet <[email protected]>
  5004. AuthorDate: Tue Feb 3 18:02:30 2015 +0100
  5005. Commit: Anton Mitrofanov <[email protected]>
  5006. CommitDate: Mon Feb 23 13:34:50 2015 +0300
  5007.  
  5008. x86inc: Fix instantiation of YMM registers
  5009.  
  5010. commit 4c75f3d729aaf3bcb00edf789c71f09495374bdf [revision 2530]
  5011. Author: Vittorio Giovara <[email protected]>
  5012. AuthorDate: Tue Jan 20 16:28:54 2015 +0000
  5013. Commit: Anton Mitrofanov <[email protected]>
  5014. CommitDate: Mon Feb 23 13:34:49 2015 +0300
  5015.  
  5016. matroska: Correctly write display width and height in stereo mode
  5017.  
  5018. According to the specifications, when stereo mode is set, these values
  5019. represent the single view size.
  5020.  
  5021. commit c3ba2a8c595b1bb36da55b82f7f4046471349d0e [revision 2529]
  5022. Author: Kieran Kunhya <[email protected]>
  5023. AuthorDate: Tue Jan 20 09:38:00 2015 -0600
  5024. Commit: Anton Mitrofanov <[email protected]>
  5025. CommitDate: Mon Feb 23 13:34:48 2015 +0300
  5026.  
  5027. Use POC type 0 for AVC-Intra
  5028.  
  5029. Based on a patch from Capella Systems
  5030.  
  5031. commit b77cc09b9252d70f78726f2472391b63948d9895 [revision 2528]
  5032. Author: Anton Mitrofanov <[email protected]>
  5033. AuthorDate: Sat Jan 3 15:46:19 2015 +0300
  5034. Commit: Anton Mitrofanov <[email protected]>
  5035. CommitDate: Mon Feb 23 13:34:47 2015 +0300
  5036.  
  5037. Fix ARCH variable name conflict with BSD ports (bsd.port.mk) read-only variable
  5038.  
  5039. commit 6e769846626f9185b59f3967e8b4ebe11497d878 [revision 2527]
  5040. Author: Anton Mitrofanov <[email protected]>
  5041. AuthorDate: Sat Dec 27 20:35:39 2014 +0300
  5042. Commit: Anton Mitrofanov <[email protected]>
  5043. CommitDate: Mon Feb 23 13:34:46 2015 +0300
  5044.  
  5045. Fix negative percentages in final stats output
  5046.  
  5047. They were caused by integer overflow when encoding long UHD video.
  5048.  
  5049. commit d7ccd89f1bea53c8c524f8e6eb963d57defb6813 [revision 2526]
  5050. Author: Anton Mitrofanov <[email protected]>
  5051. AuthorDate: Sat Jan 3 23:35:23 2015 +0300
  5052. Commit: Anton Mitrofanov <[email protected]>
  5053. CommitDate: Mon Feb 23 13:34:44 2015 +0300
  5054.  
  5055. Bump dates to 2015
  5056.  
  5057. commit 40bb56814e56ed342040bdbf30258aab39ee9e89 [revision 2525]
  5058. Author: Anton Mitrofanov <[email protected]>
  5059. AuthorDate: Mon Dec 15 18:49:23 2014 +0300
  5060. Commit: Anton Mitrofanov <[email protected]>
  5061. CommitDate: Sat Dec 20 19:29:44 2014 +0300
  5062.  
  5063. x86: Update intel compiler cpu dispatcher override for new versions of ICC/ICL
  5064.  
  5065. commit d72a85b549acd981a8dae3dc5b71920ab2aeea4f [revision 2524]
  5066. Author: Anton Mitrofanov <[email protected]>
  5067. AuthorDate: Tue Sep 6 21:53:29 2011 +0400
  5068. Commit: Anton Mitrofanov <[email protected]>
  5069. CommitDate: Sat Dec 20 19:29:31 2014 +0300
  5070.  
  5071. New AQ mode: auto-variance AQ with bias to dark scenes
  5072.  
  5073. Also known as --aq-mode 3 or auto-variance AQ modification.
  5074.  
  5075. commit f4a455a43df3088bae5208dcc98b8f6214fdce7d [revision 2523]
  5076. Author: Anton Mitrofanov <[email protected]>
  5077. AuthorDate: Wed Aug 29 03:02:27 2012 +0400
  5078. Commit: Anton Mitrofanov <[email protected]>
  5079. CommitDate: Sat Dec 20 19:27:56 2014 +0300
  5080.  
  5081. Improve HRD conformance
  5082.  
  5083. commit fa3549b5f2478f39cbcbd14d2e956e59f70d18eb [revision 2522]
  5084. Author: Henrik Gramner <[email protected]>
  5085. AuthorDate: Fri Nov 28 23:24:56 2014 +0100
  5086. Commit: Anton Mitrofanov <[email protected]>
  5087. CommitDate: Sat Dec 20 19:27:54 2014 +0300
  5088.  
  5089. x86: SSE and AVX implementations of plane_copy
  5090.  
  5091. Also remove the MMX2 implementation and fix src overread for height == 1.
  5092.  
  5093. commit 8797e0f8d416aadb91d359f144e4e7855071870a [revision 2521]
  5094. Author: Anton Mitrofanov <[email protected]>
  5095. AuthorDate: Mon Sep 29 23:26:19 2014 +0400
  5096. Commit: Anton Mitrofanov <[email protected]>
  5097. CommitDate: Tue Dec 16 20:40:13 2014 +0300
  5098.  
  5099. Update to the latest version of gas-preprocessor.pl from http://git.libav.org/?p=gas-preprocessor.git
  5100.  
  5101. Contributions by Janne Grunau, Martin Storsjo, Mans Rullgard, David Conrad, Martin Aumuller and others
  5102.  
  5103. commit 59b9c252cfa6242c7fa6424a463e51913996fe6a [revision 2520]
  5104. Author: Janne Grunau <[email protected]>
  5105. AuthorDate: Wed Nov 19 00:33:55 2014 +0100
  5106. Commit: Anton Mitrofanov <[email protected]>
  5107. CommitDate: Tue Dec 16 20:40:12 2014 +0300
  5108.  
  5109. aarch64: cabac_encode_{decision,bypass,terminal}_asm
  5110.  
  5111. benchmarks on a Nexus 9 (nvidia denver):
  5112. 101.3 cycles in x264_cabac_encode_decision_c, 67105369 runs, 3495 skips
  5113. 97.3 cycles in x264_cabac_encode_decision_asm, 67105493 runs, 3371 skips
  5114. 132.8 cycles in x264_cabac_encode_terminal_c, 1046950 runs, 1626 skips
  5115. 116.1 cycles in x264_cabac_encode_terminal_asm, 1048424 runs, 152 skips
  5116. 92.4 cycles in x264_cabac_encode_bypass_c, 16776192 runs, 1024 skips
  5117. 89.6 cycles in x264_cabac_encode_bypass_asm, 16776453 runs, 763 skips
  5118.  
  5119. Cycle counts are not as stable as one would like. The dynamic code
  5120. optimisation seems to produce different results for small chnages in a
  5121. binary. Repeated runs with the same binary produce stable results
  5122. though (ignoring the first run).
  5123.  
  5124. commit a6ec424939a4d3a59e4ec1e3999cb37e4314408e [revision 2519]
  5125. Author: Janne Grunau <[email protected]>
  5126. AuthorDate: Thu Nov 6 09:20:17 2014 +0100
  5127. Commit: Anton Mitrofanov <[email protected]>
  5128. CommitDate: Tue Dec 16 20:40:11 2014 +0300
  5129.  
  5130. checkasm: add cycle counter read for aarch64
  5131.  
  5132. Needs kernel support since user space access to the cycle counter is not
  5133. allowed on all available AArch64 systems (Android 5 and iOS).
  5134.  
  5135. commit fa7e9d3d082327ceeacfaf85da6cde4c50fb4e5b [revision 2518]
  5136. Author: Janne Grunau <[email protected]>
  5137. AuthorDate: Wed Nov 5 11:35:13 2014 +0100
  5138. Commit: Anton Mitrofanov <[email protected]>
  5139. CommitDate: Tue Dec 16 20:40:10 2014 +0300
  5140.  
  5141. aarch64: nal_escape_neon
  5142.  
  5143. 3-4 times faster.
  5144.  
  5145. commit f13573e490d9f18bbcb10409fb09ec25e477035e [revision 2517]
  5146. Author: Janne Grunau <[email protected]>
  5147. AuthorDate: Fri Oct 31 14:49:04 2014 +0100
  5148. Commit: Anton Mitrofanov <[email protected]>
  5149. CommitDate: Tue Dec 16 20:40:09 2014 +0300
  5150.  
  5151. aarch64: {plane_copy,memcpy_aligned,memzero_aligned}_neon
  5152.  
  5153. 2-3 times faster than C.
  5154.  
  5155. commit 8d655b63b4f7bc021ad038ea64b7c4de9d0ef74b [revision 2516]
  5156. Author: Janne Grunau <[email protected]>
  5157. AuthorDate: Wed Oct 29 18:17:48 2014 +0100
  5158. Commit: Anton Mitrofanov <[email protected]>
  5159. CommitDate: Tue Dec 16 20:40:08 2014 +0300
  5160.  
  5161. aarch64: x264_mbtree_propagate_{cost,list}_neon
  5162.  
  5163. x264_mbtree_propagate_cost_neon is ~7 times faster.
  5164. x264_mbtree_propagate_list_neon is 33% faster.
  5165.  
  5166. commit 4d400a6ec67f17ae3b17876b0318b956b6d5c856 [revision 2515]
  5167. Author: Janne Grunau <[email protected]>
  5168. AuthorDate: Tue Oct 21 15:18:49 2014 +0200
  5169. Commit: Anton Mitrofanov <[email protected]>
  5170. CommitDate: Tue Dec 16 20:40:07 2014 +0300
  5171.  
  5172. aarch64: x264_denoise_dct_neon
  5173.  
  5174. 3.5 times faster.
  5175.  
  5176. commit 4e8ac132cc2feff5786d12c90fd62cf97979bae1 [revision 2514]
  5177. Author: Janne Grunau <[email protected]>
  5178. AuthorDate: Mon Oct 20 13:12:14 2014 +0200
  5179. Commit: Anton Mitrofanov <[email protected]>
  5180. CommitDate: Tue Dec 16 20:40:06 2014 +0300
  5181.  
  5182. aarch64: x264_coeff_level_run{4,8,15,16}
  5183.  
  5184. All functions ~33% faster.
  5185.  
  5186. commit dd7666742d5a1a7af076fb388c6adf1b10dcdb3e [revision 2513]
  5187. Author: Janne Grunau <[email protected]>
  5188. AuthorDate: Tue Oct 14 19:20:52 2014 +0200
  5189. Commit: Anton Mitrofanov <[email protected]>
  5190. CommitDate: Tue Dec 16 20:40:05 2014 +0300
  5191.  
  5192. aarch64: NEON asm for intra luma deblocking
  5193.  
  5194. deblock_luma_intra[0]_neon is 2 times fastes,
  5195. deblock_luma_intra[1]_neon is ~4 times faster.
  5196.  
  5197. commit 0122fd230cbf7351845dd354d5ee883d741222ef [revision 2512]
  5198. Author: Janne Grunau <[email protected]>
  5199. AuthorDate: Mon Oct 13 17:29:22 2014 +0200
  5200. Commit: Anton Mitrofanov <[email protected]>
  5201. CommitDate: Tue Dec 16 20:40:04 2014 +0300
  5202.  
  5203. aarch64: x264_deblock_h_chroma_422_neon
  5204.  
  5205. deblock_h_chroma_422 2.5 times faster
  5206.  
  5207. commit 44cb1dcdbdaafeddd98d2ebe3d02408bc380713e [revision 2511]
  5208. Author: Janne Grunau <[email protected]>
  5209. AuthorDate: Mon Oct 13 12:43:50 2014 +0200
  5210. Commit: Anton Mitrofanov <[email protected]>
  5211. CommitDate: Tue Dec 16 20:40:03 2014 +0300
  5212.  
  5213. aarch64: x264_deblock_h_chroma_mbaff_neon
  5214.  
  5215. deblock_chroma_420_mbaff_neon 2 times faster
  5216.  
  5217. commit f2e439d113ae86a0a1ef8215d4d4111892aed3f7 [revision 2510]
  5218. Author: Janne Grunau <[email protected]>
  5219. AuthorDate: Fri Oct 10 10:29:15 2014 +0200
  5220. Commit: Anton Mitrofanov <[email protected]>
  5221. CommitDate: Tue Dec 16 20:40:02 2014 +0300
  5222.  
  5223. aarch64: NEON asm for intra chroma deblocking
  5224.  
  5225. deblock_h_chroma_420_intra, deblock_h_chroma_422_intra and
  5226. x264_deblock_h_chroma_intra_mbaff_neon are ~3 times faster.
  5227. deblock_chroma_intra[1] is ~4 times faster than C.
  5228.  
  5229. commit ce6c94c0bef3350e9546302aae5909404b056fdb [revision 2509]
  5230. Author: Janne Grunau <[email protected]>
  5231. AuthorDate: Tue Sep 2 10:27:22 2014 +0200
  5232. Commit: Anton Mitrofanov <[email protected]>
  5233. CommitDate: Tue Dec 16 20:40:01 2014 +0300
  5234.  
  5235. aarch64: add myself as author to aarch64/mc.h
  5236.  
  5237. commit be7e5fa6eee2731abdb1b41bc2a4c1a29e672747 [revision 2508]
  5238. Author: Janne Grunau <[email protected]>
  5239. AuthorDate: Thu Aug 14 14:22:50 2014 +0100
  5240. Commit: Anton Mitrofanov <[email protected]>
  5241. CommitDate: Tue Dec 16 20:40:00 2014 +0300
  5242.  
  5243. aarch64: NEON asm for integral init
  5244.  
  5245. integral_init4h_neon and integral_init8h_neon are 3-4 times faster than
  5246. C. integral_init8v_neon is 6 times faster and integral_init4v_neon is 10
  5247. times faster.
  5248.  
  5249. commit eb1d35725e542968c4a6480c157db40570477a95 [revision 2507]
  5250. Author: Janne Grunau <[email protected]>
  5251. AuthorDate: Wed Aug 13 13:30:53 2014 +0100
  5252. Commit: Anton Mitrofanov <[email protected]>
  5253. CommitDate: Tue Dec 16 20:39:59 2014 +0300
  5254.  
  5255. aarch64: NEON asm for 8x16c intra prediction
  5256.  
  5257. Between 10% and 40% faster than C.
  5258.  
  5259. commit 40d5db342b7f5198db9826a51f31e454bd208596 [revision 2506]
  5260. Author: Janne Grunau <[email protected]>
  5261. AuthorDate: Tue Aug 12 17:26:10 2014 +0200
  5262. Commit: Anton Mitrofanov <[email protected]>
  5263. CommitDate: Tue Dec 16 20:39:58 2014 +0300
  5264.  
  5265. aarch64: NEON asm for decimate_score
  5266.  
  5267. decimate_score15 and 16 are 60% faster, decimate_score64 is 4 times
  5268. faster than C.
  5269.  
  5270. commit 45e1ebf88a1c3bf37e1326ce621a9b735d155885 [revision 2505]
  5271. Author: Janne Grunau <[email protected]>
  5272. AuthorDate: Fri Aug 8 11:19:35 2014 +0100
  5273. Commit: Anton Mitrofanov <[email protected]>
  5274. CommitDate: Tue Dec 16 20:39:57 2014 +0300
  5275.  
  5276. aarch64: implement x264_sub8x16_dct_dc_neon
  5277.  
  5278. 4 times faster than C.
  5279.  
  5280. commit 90f0b5c1c881f345c9da15bc482055f2a92f8ceb [revision 2504]
  5281. Author: Janne Grunau <[email protected]>
  5282. AuthorDate: Thu Aug 7 19:46:07 2014 +0200
  5283. Commit: Anton Mitrofanov <[email protected]>
  5284. CommitDate: Tue Dec 16 20:39:56 2014 +0300
  5285.  
  5286. aarch64: implement x264_pixel_asd8_neon
  5287.  
  5288. 7 times faster than C.
  5289.  
  5290. commit f8f8d13d5978b13fc831e041e52aa617550bbdf3 [revision 2503]
  5291. Author: Janne Grunau <[email protected]>
  5292. AuthorDate: Thu Aug 7 16:49:12 2014 +0200
  5293. Commit: Anton Mitrofanov <[email protected]>
  5294. CommitDate: Tue Dec 16 20:39:55 2014 +0300
  5295.  
  5296. aarch64: NEON asm for 4x16 sad, satd and ssd
  5297.  
  5298. pixel_sad_4x16_neon: 33% faster than C
  5299. pixel_satd_4x16_neon: 5 times faster
  5300. pixel_ssd_4x16_neon: 4 times faster
  5301.  
  5302. commit 35b91f2410dcf4fc5191dd85ccda7a42eb01eae8 [revision 2502]
  5303. Author: Janne Grunau <[email protected]>
  5304. AuthorDate: Wed Jul 30 15:48:25 2014 +0100
  5305. Commit: Anton Mitrofanov <[email protected]>
  5306. CommitDate: Tue Dec 16 20:39:54 2014 +0300
  5307.  
  5308. aarch64: implement x264_pixel_ssd_nv12_core_neon
  5309.  
  5310. 13 times faster than C.
  5311.  
  5312. commit 99a1ca1f1a62d51e47d1ac2c92ee9c3bf3b5712b [revision 2501]
  5313. Author: Janne Grunau <[email protected]>
  5314. AuthorDate: Tue Jul 29 18:26:11 2014 +0100
  5315. Commit: Anton Mitrofanov <[email protected]>
  5316. CommitDate: Tue Dec 16 20:39:52 2014 +0300
  5317.  
  5318. aarch64: implement x264_pixel_vsad_neon
  5319.  
  5320. 35 times faster than C.
  5321.  
  5322. commit 6c1632493e5afac8be1e1693377dab27f4704a1d [revision 2500]
  5323. Author: Janne Grunau <[email protected]>
  5324. AuthorDate: Tue Jul 29 11:06:24 2014 +0100
  5325. Commit: Anton Mitrofanov <[email protected]>
  5326. CommitDate: Tue Dec 16 20:39:51 2014 +0300
  5327.  
  5328. aarch64: NEON asm for missing x264_zigzag_* functions
  5329.  
  5330. zigzag_scan_4x4_field_neon, zigzag_sub_4x4_field_neon,
  5331. zigzag_sub_4x4ac_field_neon, zigzag_sub_4x4_frame_neon,
  5332. igzag_sub_4x4ac_frame_neon more than 2 times faster
  5333.  
  5334. zigzag_scan_8x8_frame_neon, zigzag_scan_8x8_field_neon,
  5335. zigzag_sub_8x8_field_neon, zigzag_sub_8x8_frame_neon 4-5 times faster
  5336.  
  5337. zigzag_interleave_8x8_cavlc_neon 6 times faster
  5338.  
  5339. commit d040d28514db7d1fbd5c3f06c37a77de14b15e5b [revision 2499]
  5340. Author: Janne Grunau <[email protected]>
  5341. AuthorDate: Fri Jul 25 11:53:17 2014 +0100
  5342. Commit: Anton Mitrofanov <[email protected]>
  5343. CommitDate: Tue Dec 16 20:39:50 2014 +0300
  5344.  
  5345. aarch64: implement x264_pixel_sa8d_satd_16x16_neon
  5346.  
  5347. ~20% faster than calling pixel_sa8d_16x16 and pixel_satd_16x16
  5348. separately.
  5349.  
  5350. commit 91a01d4ca95ee1c621578e118b86d767eab96b3b [revision 2498]
  5351. Author: Janne Grunau <[email protected]>
  5352. AuthorDate: Thu Aug 14 23:13:27 2014 +0200
  5353. Commit: Anton Mitrofanov <[email protected]>
  5354. CommitDate: Tue Dec 16 20:39:49 2014 +0300
  5355.  
  5356. aarch64: optimize x264_predict_8x8c_dc_left_neon
  5357.  
  5358. 25% faster than the previous version.
  5359.  
  5360. commit 8ae4e1cfa3d16451ccf285228d309f6f4940a747 [revision 2497]
  5361. Author: Henrik Gramner <[email protected]>
  5362. AuthorDate: Sat Aug 2 18:26:18 2014 +0200
  5363. Commit: Henrik Gramner <[email protected]>
  5364. CommitDate: Sat Dec 13 00:42:28 2014 +0100
  5365.  
  5366. x86: Make AVX2 also imply FMA3
  5367.  
  5368. All CPUs with AVX2 supports FMA3 (but not the other way around).
  5369.  
  5370. commit 06882793b260824bc578d0530f64e7f30f2a9f39 [revision 2496]
  5371. Author: Anton Mitrofanov <[email protected]>
  5372. AuthorDate: Thu Nov 13 22:52:00 2014 +0300
  5373. Commit: Henrik Gramner <[email protected]>
  5374. CommitDate: Sat Dec 13 00:42:19 2014 +0100
  5375.  
  5376. Simplify libx264 API usage example
  5377.  
  5378. commit 6a301b6ee0ae8c78fb704e1cd86f4e861070f641 [revision 2495]
  5379. Author: Henrik Gramner <[email protected]>
  5380. AuthorDate: Fri Nov 21 23:47:20 2014 +0100
  5381. Commit: Henrik Gramner <[email protected]>
  5382. CommitDate: Sat Dec 13 00:41:39 2014 +0100
  5383.  
  5384. AvxSynth: Remove a bunch of unused cruft
  5385.  
  5386. commit 30140b34b879605cf70cab0634a4a8faef5b6e60 [revision 2494]
  5387. Author: Anton Mitrofanov <[email protected]>
  5388. AuthorDate: Wed Dec 3 22:36:12 2014 +0300
  5389. Commit: Henrik Gramner <[email protected]>
  5390. CommitDate: Sat Dec 13 00:34:15 2014 +0100
  5391.  
  5392. Fix bugs/typos in motion compensation and cache_load
  5393.  
  5394. Didn't affect output due to the incorrect values either not being used in the
  5395. code path or producing equal results compared to the correct values.
  5396.  
  5397. Also deduplicate hpel_ref arrays.
  5398.  
  5399. commit a46820e00ad3c86b80f5830ed92553de474b7d5c [revision 2493]
  5400. Author: Anton Mitrofanov <[email protected]>
  5401. AuthorDate: Sun Nov 30 23:39:28 2014 +0300
  5402. Commit: Anton Mitrofanov <[email protected]>
  5403. CommitDate: Mon Dec 1 23:47:58 2014 +0300
  5404.  
  5405. checkasm: Fix undefined behavior warnings
  5406.  
  5407. commit 4e97ca566fdf6cd36281e26ee68f64993f4751a1 [revision 2492]
  5408. Author: Henrik Gramner <[email protected]>
  5409. AuthorDate: Sat Nov 29 18:47:52 2014 +0100
  5410. Commit: Henrik Gramner <[email protected]>
  5411. CommitDate: Sat Nov 29 19:13:24 2014 +0100
  5412.  
  5413. checkasm: Fix V210 reporting
  5414.  
  5415. It would previously report FAILED if any of the earlier plane_copy tests failed.
  5416.  
  5417. commit 24e4fed388fcb34c33df7c87e7d6758b9ebed40c [revision 2491]
  5418. Author: Anton Mitrofanov <[email protected]>
  5419. AuthorDate: Sun Oct 12 21:01:53 2014 +0400
  5420. Commit: Fiona <[email protected]>
  5421. CommitDate: Fri Oct 17 14:25:43 2014 -0700
  5422.  
  5423. Safety check against malicious high bit-depth input which could cause crash
  5424.  
  5425. commit 9bec6fed6d1b95f9921f22ba21e7398eff50b75e [revision 2490]
  5426. Author: Anton Mitrofanov <[email protected]>
  5427. AuthorDate: Sun Oct 12 20:45:40 2014 +0400
  5428. Commit: Fiona <[email protected]>
  5429. CommitDate: Fri Oct 17 14:25:28 2014 -0700
  5430.  
  5431. libx264 API usage example
  5432.  
  5433. commit 329fe5f6498be7ab337d98ac22c17d379335c854 [revision 2489]
  5434. Author: Henrik Gramner <[email protected]>
  5435. AuthorDate: Fri Oct 17 21:35:42 2014 +0200
  5436. Commit: Fiona <[email protected]>
  5437. CommitDate: Fri Oct 17 13:22:55 2014 -0700
  5438.  
  5439. x86: AVX2 high bit-depth var_16x16
  5440.  
  5441. 40->27 cycles on Haswell.
  5442.  
  5443. commit 4576cfd8c391b27748d6f97f5b621cec4ed8047c [revision 2488]
  5444. Author: Henrik Gramner <[email protected]>
  5445. AuthorDate: Wed Oct 8 22:25:35 2014 +0200
  5446. Commit: Fiona <[email protected]>
  5447. CommitDate: Wed Oct 8 17:17:27 2014 -0700
  5448.  
  5449. checkasm: Serialize read_time() calls on x86
  5450.  
  5451. Improves the accuracy of benchmarks, especially in short functions.
  5452.  
  5453. To quote the Intel 64 and IA-32 Architectures Software Developer's Manual:
  5454. "The RDTSC instruction is not a serializing instruction. It does not necessarily
  5455. wait until all previous instructions have been executed before reading the counter.
  5456. Similarly, subsequent instructions may begin execution before the read operation
  5457. is performed. If software requires RDTSC to be executed only after all previous
  5458. instructions have completed locally, it can either use RDTSCP (if the processor
  5459. supports that instruction) or execute the sequence LFENCE;RDTSC."
  5460.  
  5461. RDTSCP would accomplish the same task, but it's only available since Nehalem.
  5462.  
  5463. This change makes SSE2 a requirement to run checkasm.
  5464.  
  5465. commit b85a74a22f79c8722674c4cfd7cddf5f54c8421d [revision 2487]
  5466. Author: Vittorio Giovara <[email protected]>
  5467. AuthorDate: Mon Sep 29 18:51:30 2014 +0100
  5468. Commit: Fiona Glaser <[email protected]>
  5469. CommitDate: Mon Sep 29 11:55:36 2014 -0700
  5470.  
  5471. Support case-independent string options
  5472.  
  5473. commit 20f116b29e93574e9607d1abf2960f32b5730e52 [revision 2486]
  5474. Author: Anton Mitrofanov <[email protected]>
  5475. AuthorDate: Sat Sep 6 20:44:49 2014 +0400
  5476. Commit: Fiona Glaser <[email protected]>
  5477. CommitDate: Tue Sep 16 12:12:04 2014 -0700
  5478.  
  5479. Shut up gcc -Wuninitialized warnings
  5480.  
  5481. commit 3df1d248dd8a4b0d0dffd149effe2bde38de49aa [revision 2485]
  5482. Author: Anton Mitrofanov <[email protected]>
  5483. AuthorDate: Fri Sep 5 19:43:52 2014 +0400
  5484. Commit: Fiona Glaser <[email protected]>
  5485. CommitDate: Tue Sep 16 12:12:04 2014 -0700
  5486.  
  5487. Shut up clang -Wuninitialized warning
  5488.  
  5489. commit 01204b60367f4959e8393652dd30f0cfba2d2c80 [revision 2484]
  5490. Author: Anton Mitrofanov <[email protected]>
  5491. AuthorDate: Fri Sep 5 19:30:47 2014 +0400
  5492. Commit: Fiona Glaser <[email protected]>
  5493. CommitDate: Tue Sep 16 12:12:03 2014 -0700
  5494.  
  5495. Fix few clang -Wunused-* warnings
  5496.  
  5497. commit 9df377f87702c82a2202d34919c07e32c60b40ae [revision 2483]
  5498. Author: Anton Mitrofanov <[email protected]>
  5499. AuthorDate: Thu Aug 28 20:13:13 2014 +0400
  5500. Commit: Fiona Glaser <[email protected]>
  5501. CommitDate: Tue Sep 16 12:11:42 2014 -0700
  5502.  
  5503. Fix inappropriate instruction use
  5504.  
  5505. commit 73b8686fc22c9247d90963983d406cd7b9131068 [revision 2482]
  5506. Author: Anton Mitrofanov <[email protected]>
  5507. AuthorDate: Thu Aug 28 18:38:53 2014 +0400
  5508. Commit: Fiona Glaser <[email protected]>
  5509. CommitDate: Tue Sep 16 12:11:41 2014 -0700
  5510.  
  5511. x264asm: warn when inappropriate instruction used in function with specified cpuflags
  5512.  
  5513. commit 204a9bd0a1bc507cbd69a77f3318afcb56ede65d [revision 2481]
  5514. Author: Anton Mitrofanov <[email protected]>
  5515. AuthorDate: Tue Sep 2 01:48:00 2014 +0400
  5516. Commit: Fiona Glaser <[email protected]>
  5517. CommitDate: Wed Sep 3 10:33:45 2014 -0700
  5518.  
  5519. Fix VBV with true VFR streams
  5520.  
  5521. commit b36d44c68cddff00c5b6de1e6cb6a86c1af2cbfc [revision 2480]
  5522. Author: Anton Mitrofanov <[email protected]>
  5523. AuthorDate: Mon Sep 1 22:45:00 2014 +0400
  5524. Commit: Fiona Glaser <[email protected]>
  5525. CommitDate: Wed Sep 3 10:33:34 2014 -0700
  5526.  
  5527. Fix VBV
  5528.  
  5529. commit dd79a61e0e354a432907f2d1f7137b27a12dfce7 [revision 2479]
  5530. Author: Anton Mitrofanov <[email protected]>
  5531. AuthorDate: Wed Jul 30 03:03:32 2014 +0400
  5532. Commit: Fiona Glaser <[email protected]>
  5533. CommitDate: Tue Aug 26 09:19:25 2014 -0700
  5534.  
  5535. Update to the current lavf API and fix memory leak when using --seek
  5536.  
  5537. commit 91727d729a4a33a3f21188f838077040740cb353 [revision 2478]
  5538. Author: Henrik Gramner <[email protected]>
  5539. AuthorDate: Tue Aug 5 01:42:55 2014 +0200
  5540. Commit: Fiona Glaser <[email protected]>
  5541. CommitDate: Tue Aug 26 09:19:25 2014 -0700
  5542.  
  5543. x86inc: Make INIT_CPUFLAGS support an arbitrary number of cpuflags
  5544.  
  5545. Previously there was a limit of two cpuflags.
  5546.  
  5547. commit d4317786b8428b00978459f6de3db219f0f6f8e6 [revision 2477]
  5548. Author: Henrik Gramner <[email protected]>
  5549. AuthorDate: Tue Aug 5 01:42:51 2014 +0200
  5550. Commit: Fiona Glaser <[email protected]>
  5551. CommitDate: Tue Aug 26 09:19:24 2014 -0700
  5552.  
  5553. x86: Minor pixel_ssim_end4 improvements
  5554.  
  5555. Reduce the number of vector registers used from 7 to 5.
  5556. Eliminate some moves in the AVX implementation.
  5557. Avoid bypass delays for transitioning between int and float domains.
  5558.  
  5559. commit 98100b88b475227f375d9bcbaea0bac57008accc [revision 2476]
  5560. Author: Henrik Gramner <[email protected]>
  5561. AuthorDate: Tue Aug 5 01:42:47 2014 +0200
  5562. Commit: Fiona Glaser <[email protected]>
  5563. CommitDate: Tue Aug 26 09:19:24 2014 -0700
  5564.  
  5565. x86: Faster quant_4x4x4
  5566.  
  5567. Also drop the MMX version instead of doing a bunch of ifdeffery to support it after this change.
  5568.  
  5569. commit 56fcb444c4c118ff67cf12838d2b2801d7b43407 [revision 2475]
  5570. Author: Anton Mitrofanov <[email protected]>
  5571. AuthorDate: Sun Aug 10 22:46:12 2014 +0400
  5572. Commit: Fiona Glaser <[email protected]>
  5573. CommitDate: Tue Aug 26 09:19:24 2014 -0700
  5574.  
  5575. configure: improve cc_check for clang and ICL to not ignore unknown options
  5576.  
  5577. commit ecb04d08af654a7cfd5b9aa6261bd789de20613a [revision 2474]
  5578. Author: Henrik Gramner <[email protected]>
  5579. AuthorDate: Tue Aug 5 01:42:44 2014 +0200
  5580. Commit: Fiona Glaser <[email protected]>
  5581. CommitDate: Tue Aug 26 09:19:23 2014 -0700
  5582.  
  5583. checkasm: Only call x264_cpu_detect() once
  5584.  
  5585. commit 1343db872b1d7d43dc7fb431a8207efb5ca31e2e [revision 2473]
  5586. Author: Janne Grunau <[email protected]>
  5587. AuthorDate: Fri Jul 18 14:49:10 2014 +0100
  5588. Commit: Fiona Glaser <[email protected]>
  5589. CommitDate: Tue Aug 26 09:19:23 2014 -0700
  5590.  
  5591. aarch64: deblocking NEON asm
  5592.  
  5593. Deblock chroma/luma are based on libav's h264 aarch64 NEON deblocking
  5594. filter which was ported by me from the existing ARM NEON asm. No
  5595. additional persons to ask for a relicense.
  5596.  
  5597. commit 3c1fa5d9b2ea62f05473080313c543b7e795b307 [revision 2472]
  5598. Author: Janne Grunau <[email protected]>
  5599. AuthorDate: Fri Jul 18 09:29:35 2014 +0100
  5600. Commit: Fiona Glaser <[email protected]>
  5601. CommitDate: Tue Aug 26 09:19:23 2014 -0700
  5602.  
  5603. aarch64: intra predition NEON asm
  5604.  
  5605. Ported from the ARM NEON asm.
  5606.  
  5607. commit 556b0e7928d14818454e0c33032754f6323f02e9 [revision 2471]
  5608. Author: Janne Grunau <[email protected]>
  5609. AuthorDate: Thu Jul 17 15:58:44 2014 +0100
  5610. Commit: Fiona Glaser <[email protected]>
  5611. CommitDate: Tue Aug 26 09:19:22 2014 -0700
  5612.  
  5613. aarch64: motion compensation NEON asm
  5614.  
  5615. Ported from the ARM NEON asm.
  5616.  
  5617. commit 6cda439867fcd9e884a10502845fb79fc7ffed69 [revision 2470]
  5618. Author: Janne Grunau <[email protected]>
  5619. AuthorDate: Wed Jul 16 10:03:52 2014 +0100
  5620. Commit: Fiona Glaser <[email protected]>
  5621. CommitDate: Tue Aug 26 09:19:22 2014 -0700
  5622.  
  5623. aarch64: transform and zigzag NEON asm
  5624.  
  5625. Ported from the ARM NEON asm.
  5626.  
  5627. commit db5c504aa06550f8e916157d1dcc657818e84d62 [revision 2469]
  5628. Author: Janne Grunau <[email protected]>
  5629. AuthorDate: Tue Jul 15 12:57:03 2014 +0100
  5630. Commit: Fiona Glaser <[email protected]>
  5631. CommitDate: Tue Aug 26 09:19:22 2014 -0700
  5632.  
  5633. aarch64: quantization and level-run NEON asm
  5634.  
  5635. Ported from the ARM NEON asm.
  5636.  
  5637. commit f4a82a54885f3dad7106a6855eaef50ea085b27e [revision 2468]
  5638. Author: Janne Grunau <[email protected]>
  5639. AuthorDate: Wed Mar 19 13:48:21 2014 +0100
  5640. Commit: Fiona Glaser <[email protected]>
  5641. CommitDate: Tue Aug 26 09:19:21 2014 -0700
  5642.  
  5643. aarch64: pixel metrics NEON asm
  5644.  
  5645. Ported from the ARM NEON asm.
  5646.  
  5647. commit 3e57554ee4db6ade7a2dccaac92cb8116f3a43d6 [revision 2467]
  5648. Author: Janne Grunau <[email protected]>
  5649. AuthorDate: Fri Jul 18 17:44:57 2014 +0200
  5650. Commit: Fiona Glaser <[email protected]>
  5651. CommitDate: Tue Aug 26 09:19:21 2014 -0700
  5652.  
  5653. aarch64: add utility functions for asm
  5654.  
  5655. commit efaf0b88f7c703533ee8857a6a5039cf64bce3a0 [revision 2466]
  5656. Author: Janne Grunau <[email protected]>
  5657. AuthorDate: Wed Mar 19 13:45:17 2014 +0100
  5658. Commit: Fiona Glaser <[email protected]>
  5659. CommitDate: Tue Aug 26 09:19:21 2014 -0700
  5660.  
  5661. aarch64: add armv8 and neon cpu flags and test them
  5662.  
  5663. commit 943128a527d1b98a63017d58cd1fcf53aaffcb6e [revision 2465]
  5664. Author: Janne Grunau <[email protected]>
  5665. AuthorDate: Tue Mar 18 22:10:24 2014 +0100
  5666. Commit: Fiona Glaser <[email protected]>
  5667. CommitDate: Tue Aug 26 09:19:20 2014 -0700
  5668.  
  5669. aarch64: initial build support
  5670.  
  5671. commit ee427b69868d506182f4e22bffdc45e913f255af [revision 2464]
  5672. Author: Janne Grunau <[email protected]>
  5673. AuthorDate: Tue Jul 22 19:28:27 2014 +0200
  5674. Commit: Fiona Glaser <[email protected]>
  5675. CommitDate: Tue Aug 26 09:19:20 2014 -0700
  5676.  
  5677. checkasm: test zigzag_sub_8x8_{frame,field}
  5678.  
  5679. commit 69740fd362ee1c0a2e80d6f4e2724d731a3c951c [revision 2463]
  5680. Author: Janne Grunau <[email protected]>
  5681. AuthorDate: Sun Jul 20 18:29:01 2014 +0200
  5682. Commit: Fiona Glaser <[email protected]>
  5683. CommitDate: Tue Aug 26 09:19:20 2014 -0700
  5684.  
  5685. arm: use long multiplication in mc_weight_w*_neon
  5686.  
  5687. 9-19% faster on a cortex-a9.
  5688.  
  5689. commit 0a05b3f9aa8c524a67119ec5eb6bcc24eb8f2f3b [revision 2462]
  5690. Author: Janne Grunau <[email protected]>
  5691. AuthorDate: Sun Jul 20 18:24:57 2014 +0200
  5692. Commit: Fiona Glaser <[email protected]>
  5693. CommitDate: Tue Aug 26 09:19:20 2014 -0700
  5694.  
  5695. arm: do not use aligned stores in mc_weight_w4_*neon
  5696.  
  5697. mc_weight_w4_*neon is also used for width 2 which does not guarantee
  5698. 4-byte aligned destination. Fixes crashes caused by random memory
  5699. corruption.
  5700.  
  5701. commit c2df1fc65c98e213c444134d5dbbb79d439af4db [revision 2461]
  5702. Author: Janne Grunau <[email protected]>
  5703. AuthorDate: Wed Apr 2 16:31:28 2014 +0200
  5704. Commit: Fiona Glaser <[email protected]>
  5705. CommitDate: Tue Aug 26 09:19:19 2014 -0700
  5706.  
  5707. checkasm: add memory clobber to read_time inline asm
  5708.  
  5709. The memory acts as compiler barrier preventing aggressive reordering
  5710. of read_time calls. gcc 4.8 reorders some of initial read_time calls
  5711. after the second when targeting arm.
  5712.  
  5713. commit d72760401cb0602b8bf86037988e66cdc810681c [revision 2460]
  5714. Author: Janne Grunau <[email protected]>
  5715. AuthorDate: Sun Jul 20 13:32:10 2014 +0200
  5716. Commit: Fiona Glaser <[email protected]>
  5717. CommitDate: Tue Aug 26 09:19:19 2014 -0700
  5718.  
  5719. arm: check if the assembler supports the '.func' directive
  5720.  
  5721. The integrated assembler in llvm trunk (to be released as 3.5) is
  5722. otherwise capable enough to assemble the arm asm correctly.
  5723.  
  5724. commit 9463ec0004f1bddc49c05ed8e38430a4ce1738fb [revision 2459]
  5725. Author: Janne Grunau <[email protected]>
  5726. AuthorDate: Sun Jul 20 13:40:28 2014 +0200
  5727. Commit: Fiona Glaser <[email protected]>
  5728. CommitDate: Tue Aug 26 09:19:19 2014 -0700
  5729.  
  5730. arm/ppc: use $CC as default assembler
  5731.  
  5732. commit feec4a478bfdfb4426268b2ee79bac473b97488c [revision 2458]
  5733. Author: Janne Grunau <[email protected]>
  5734. AuthorDate: Sun Jul 20 13:34:27 2014 +0200
  5735. Commit: Fiona Glaser <[email protected]>
  5736. CommitDate: Tue Aug 26 09:19:18 2014 -0700
  5737.  
  5738. arm: move instructions after '.rept' to separate line
  5739.  
  5740. The gas manual states "Repeat the sequence of lines between the .rept
  5741. directive and the next .endr directive ...". GNU as seems to support
  5742. instructions on the same line as .rept anyway but the integrated
  5743. assembler in llvm trunk (to be released 3.5 in August 2014) does not.
  5744.  
  5745. commit 6e8971021d2a12505cb2ad9ea677dfc8af676919 [revision 2457]
  5746. Author: Janne Grunau <[email protected]>
  5747. AuthorDate: Sun Jul 20 13:08:17 2014 +0200
  5748. Commit: Fiona Glaser <[email protected]>
  5749. CommitDate: Tue Aug 26 09:19:18 2014 -0700
  5750.  
  5751. arm: set .arch/.fpu from asm.S
  5752.  
  5753. commit 716ee56d0b35e512e8e0ae1a3e71f26e65e86be3 [revision 2456]
  5754. Author: Janne Grunau <[email protected]>
  5755. AuthorDate: Sun Jul 20 12:55:53 2014 +0200
  5756. Commit: Fiona Glaser <[email protected]>
  5757. CommitDate: Tue Aug 26 09:19:18 2014 -0700
  5758.  
  5759. arm: do not append CFLAGS to ASFLAGS
  5760.  
  5761. commit 021c0dc6c95c1bc239c9db78a80dd85fc856a4dd [revision 2455]
  5762. Author: Tristan Matthews <[email protected]>
  5763. AuthorDate: Thu Jul 17 00:03:50 2014 -0400
  5764. Commit: Fiona Glaser <[email protected]>
  5765. CommitDate: Tue Aug 26 09:19:17 2014 -0700
  5766.  
  5767. filters: fix sizeof mismatch
  5768.  
  5769. commit 95beb822e61a8d84dba9743f4b20b4c303f26798 [revision 2454]
  5770. Author: Anton Mitrofanov <[email protected]>
  5771. AuthorDate: Thu Jul 31 16:17:32 2014 +0400
  5772. Commit: Fiona Glaser <[email protected]>
  5773. CommitDate: Tue Aug 26 09:19:17 2014 -0700
  5774.  
  5775. Fix memory leak when using select_every filter
  5776.  
  5777. commit ea0ca51e94323318b95bd8b27b7f9438cdcf4d9e [revision 2453]
  5778. Author: Tsukasa OMOTO <[email protected]>
  5779. AuthorDate: Sun Jul 20 22:17:11 2014 +0900
  5780. Commit: Fiona Glaser <[email protected]>
  5781. CommitDate: Sun Jul 20 11:20:34 2014 -0700
  5782.  
  5783. Fix cltostr.sh on OS X
  5784.  
  5785. commit 08d36b3fc975d049aa3786ca34fb0b2f2ba0007c [revision 2452]
  5786. Author: Fiona Glaser <[email protected]>
  5787. AuthorDate: Wed Jul 9 12:21:33 2014 -0700
  5788. Commit: Fiona Glaser <[email protected]>
  5789. CommitDate: Sat Jul 19 20:53:02 2014 -0700
  5790.  
  5791. Check pf_log is set in validate_parameters
  5792.  
  5793. Help remind people to call x264_param_default in case they didn't read the
  5794. documentation.
  5795.  
  5796. commit 9e93d18b7fe7668f8277b5f117d7e39be24c6070 [revision 2451]
  5797. Author: Anton Mitrofanov <[email protected]>
  5798. AuthorDate: Wed Jul 9 17:17:04 2014 +0400
  5799. Commit: Fiona Glaser <[email protected]>
  5800. CommitDate: Sat Jul 19 20:53:01 2014 -0700
  5801.  
  5802. Check malloc during frame dumping
  5803.  
  5804. commit 8a85db879d57537f91a9908be3585512981c08b8 [revision 2450]
  5805. Author: Yusuke Nakamura <[email protected]>
  5806. AuthorDate: Thu Jun 19 05:21:29 2014 +0900
  5807. Commit: Fiona Glaser <[email protected]>
  5808. CommitDate: Sat Jul 19 20:53:01 2014 -0700
  5809.  
  5810. mp4_lsmash: Use new I/O API instead of deprecated one.
  5811.  
  5812. commit f112c0e1cae71eb5b98b4f86f635f235cc7b81cb [revision 2449]
  5813. Author: Anton Mitrofanov <[email protected]>
  5814. AuthorDate: Sun Jun 8 22:19:46 2014 +0400
  5815. Commit: Fiona Glaser <[email protected]>
  5816. CommitDate: Sat Jul 19 20:53:01 2014 -0700
  5817.  
  5818. Remove meaningless use of abs()
  5819.  
  5820. commit 6fbbb5b0c05a1d95cbd6efa7f01808ea87a39dc9 [revision 2448]
  5821. Author: Steven Walters <[email protected]>
  5822. AuthorDate: Sat May 31 10:31:16 2014 -0400
  5823. Commit: Fiona Glaser <[email protected]>
  5824. CommitDate: Sat Jul 19 20:53:00 2014 -0700
  5825.  
  5826. MSVS 2013 Update 2 support
  5827.  
  5828. The first MSVS compiler C99 compliant enough to build x264.
  5829. Use `CC=cl ./configure` to compile with it.
  5830.  
  5831. commit f53af048ed94551734265cf8f9dbe12d211a77fc [revision 2447]
  5832. Author: Diego Biurrun <[email protected]>
  5833. AuthorDate: Tue Apr 15 22:54:08 2014 +0200
  5834. Commit: Fiona Glaser <[email protected]>
  5835. CommitDate: Sat Jul 19 20:53:00 2014 -0700
  5836.  
  5837. configure: Add -Wno-maybe-uninitialized to CFLAGS
  5838.  
  5839. The warnings generated by -Wmaybe-uninitialized are mostly spurious.
  5840.  
  5841. commit cbd8d7b6db1f29929d1ad347e15afe7828ad7055 [revision 2446]
  5842. Author: Diego Biurrun <[email protected]>
  5843. AuthorDate: Wed May 7 13:20:43 2014 +0200
  5844. Commit: Fiona Glaser <[email protected]>
  5845. CommitDate: Sat Jul 19 20:53:00 2014 -0700
  5846.  
  5847. build: Replace cltostr.pl by a shell script
  5848.  
  5849. This avoids a dependency on Perl to build OpenCL support.
  5850.  
  5851. commit d8b6ce7f703d3c9d83dbd4e8ef44cfabc7e2f78e [revision 2445]
  5852. Author: Diego Biurrun <[email protected]>
  5853. AuthorDate: Tue Apr 15 23:02:39 2014 +0200
  5854. Commit: Fiona Glaser <[email protected]>
  5855. CommitDate: Sat Jul 19 20:52:59 2014 -0700
  5856.  
  5857. build: Simplify phony target declaration with wildcards
  5858.  
  5859. Also add etags to list of phony targets.
  5860.  
  5861. commit 2bd932fdf053faace84028a66d8ba9e17d526456 [revision 2444]
  5862. Author: Diego Biurrun <[email protected]>
  5863. AuthorDate: Wed May 7 12:47:37 2014 +0200
  5864. Commit: Fiona Glaser <[email protected]>
  5865. CommitDate: Sat Jul 19 20:52:59 2014 -0700
  5866.  
  5867. configure: Drop workaround for obsolete gcc 4.2 on ARM
  5868.  
  5869. commit 31311f254971e1da51d817cb580fc4fe1f4d5f20 [revision 2443]
  5870. Author: Diego Biurrun <[email protected]>
  5871. AuthorDate: Wed May 7 21:43:15 2014 +0200
  5872. Commit: Fiona Glaser <[email protected]>
  5873. CommitDate: Sat Jul 19 20:52:59 2014 -0700
  5874.  
  5875. build: Add dependencies on x86inc.asm/x86util.asm for all .asm files
  5876.  
  5877. This is a little bit overzealous, but errs on the side of caution.
  5878. Generating full dependency information is also possible, but slightly
  5879. slows down the build as YASM cannot do it as a sideeffect of compilation.
  5880.  
  5881. commit 016831ec7b3a4a7062908243dbde62d7d89b334e [revision 2442]
  5882. Author: Diego Biurrun <[email protected]>
  5883. AuthorDate: Sun Apr 27 21:09:54 2014 +0200
  5884. Commit: Fiona Glaser <[email protected]>
  5885. CommitDate: Sat Jul 19 20:52:58 2014 -0700
  5886.  
  5887. Delete all SPARC optimizations
  5888.  
  5889. SPARC has been obsolete for a long time and makes little sense as a
  5890. H.264 encoding platform.
  5891.  
  5892. Also update authors file.
  5893.  
  5894. commit c7c8eb15923d1888bb87e7642a66b417fab61e76 [revision 2441]
  5895. Author: Diego Biurrun <[email protected]>
  5896. AuthorDate: Wed May 7 12:46:42 2014 +0200
  5897. Commit: Fiona Glaser <[email protected]>
  5898. CommitDate: Sat Jul 19 20:52:58 2014 -0700
  5899.  
  5900. configure: Don't check for libavcore
  5901.  
  5902. libavcore was a never-released bad idea with a short lifespan.
  5903.  
  5904. commit dd5b5d3959e35c122c7709a9823a26b589c950da [revision 2440]
  5905. Author: Diego Biurrun <[email protected]>
  5906. AuthorDate: Sun Apr 27 23:19:04 2014 +0200
  5907. Commit: Fiona Glaser <[email protected]>
  5908. CommitDate: Sat Jul 19 20:52:58 2014 -0700
  5909.  
  5910. build: Set all ASFLAGS from within configure
  5911.  
  5912. This is how all other toolchain flags are handled.
  5913.  
  5914. commit c15f20bd772487d863f01a2813a3ab45b1f11a6b [revision 2439]
  5915. Author: Diego Biurrun <[email protected]>
  5916. AuthorDate: Sun Apr 27 23:23:49 2014 +0200
  5917. Commit: Fiona Glaser <[email protected]>
  5918. CommitDate: Sat Jul 19 20:52:57 2014 -0700
  5919.  
  5920. opencl: Check return value of fread()
  5921.  
  5922. common/opencl.c:138:10: warning: ignoring return value of 'fread', declared with attribute warn_unused_result [-Wunused-result]
  5923.  
  5924. commit af8e768e2bd3b4398bca033998f83b0eb8874914 [revision 2438]
  5925. Author: Fiona Glaser <[email protected]>
  5926. AuthorDate: Sat Jul 19 20:34:22 2014 -0700
  5927. Commit: Fiona Glaser <[email protected]>
  5928. CommitDate: Sat Jul 19 20:52:57 2014 -0700
  5929.  
  5930. Disable i8x8 in lossless
  5931.  
  5932. x264's implementation was slightly incorrect due to a vague spec, so some
  5933. decoders decoded video incorrectly.
  5934.  
  5935. Minimal impact on compression.
  5936.  
  5937. commit 450cf7ae2592ee0cb474bcefedf90c9911605e26 [revision 2437]
  5938. Author: Thomas Mundt <[email protected]>
  5939. AuthorDate: Fri Jun 27 11:12:06 2014 -0700
  5940. Commit: Fiona Glaser <[email protected]>
  5941. CommitDate: Sat Jul 19 20:52:56 2014 -0700
  5942.  
  5943. AVC-Intra: fix compatibility with Avid Transfermanager
  5944.  
  5945. commit 6eb483e4ca23f34a6a8fe09f3f2e9c9f192fd76b [revision 2436]
  5946. Author: Henrik Gramner <[email protected]>
  5947. AuthorDate: Tue Jul 8 21:15:32 2014 +0200
  5948. Commit: Fiona Glaser <[email protected]>
  5949. CommitDate: Sat Jul 19 20:52:56 2014 -0700
  5950.  
  5951. x86: Fix SIGILL in high bit-depth intra_sad_x3_4x4_sse2
  5952.  
  5953. An SSE3 instruction was used in an SSE2 function.
  5954.  
  5955. commit 5e58ce7a8b39ab66c7d6420b85a8e09dd08dfaaf [revision 2435]
  5956. Author: Anton Mitrofanov <[email protected]>
  5957. AuthorDate: Wed Jul 9 17:01:54 2014 +0400
  5958. Commit: Fiona Glaser <[email protected]>
  5959. CommitDate: Sat Jul 19 20:52:56 2014 -0700
  5960.  
  5961. Fix incorrect row predictor addressing
  5962.  
  5963. Somehow managed to not cause things to explode, but was clearly incorrect.
  5964. Might improve VBV in some cases to have this working right.
  5965.  
  5966. commit 3fda920e6f1e4a8f76680c001962542866408114 [revision 2434]
  5967. Author: Anton Mitrofanov <[email protected]>
  5968. AuthorDate: Sat Jun 21 23:52:39 2014 +0400
  5969. Commit: Fiona Glaser <[email protected]>
  5970. CommitDate: Sat Jul 19 20:52:55 2014 -0700
  5971.  
  5972. Fix b-pyramid MMCO remove for frame-packing==5
  5973.  
  5974. commit 92fdb43dd47fbc3368d9d9c7ad940fbe03657bd3 [revision 2433]
  5975. Author: Tal Aloni <[email protected]>
  5976. AuthorDate: Tue Jun 17 15:10:56 2014 -0700
  5977. Commit: Fiona Glaser <[email protected]>
  5978. CommitDate: Sat Jul 19 20:52:55 2014 -0700
  5979.  
  5980. Fix frame-packing==5 with some decoders
  5981.  
  5982. The spec mandates that frame-packing==5 requires the SEI on every frame that
  5983. begins a view sequence (i.e. the input frames L0-R0-L1-R1 have 4 view sequences,
  5984. but if reordered by the encoder to L0-L1-R0-R1 there are now 2 view sequences).
  5985. For simplicity, we write the SEI on every frame.
  5986.  
  5987. This fixes frame-packing==5 3D playback on some decoders (PlayStation 3, Sony
  5988. W8 series, possibly others).
  5989.  
  5990. commit 13d6dfd83af98e472a9e9a8b6abf5c971707a893 [revision 2432]
  5991. Author: Anton Mitrofanov <[email protected]>
  5992. AuthorDate: Thu May 22 13:27:00 2014 +0400
  5993. Commit: Fiona Glaser <[email protected]>
  5994. CommitDate: Fri Jun 20 10:05:55 2014 -0700
  5995.  
  5996. Fix pixel_ssim_end4 asm function for x86_64 systems
  5997.  
  5998. commit a5831aa256b3161f898d2577d2eb8daa838d88d2 [revision 2431]
  5999. Author: James Almer <[email protected]>
  6000. AuthorDate: Wed Apr 9 03:33:06 2014 -0300
  6001. Commit: Fiona Glaser <[email protected]>
  6002. CommitDate: Tue Apr 22 15:37:52 2014 -0700
  6003.  
  6004. x86: XOP pixel_sad_{x3, x4} high bit-depth
  6005.  
  6006. commit 0d989a4ff3298f9e495be452880b5f9bfb441e93 [revision 2430]
  6007. Author: James Almer <[email protected]>
  6008. AuthorDate: Wed Apr 9 03:33:05 2014 -0300
  6009. Commit: Fiona Glaser <[email protected]>
  6010. CommitDate: Tue Apr 22 15:37:51 2014 -0700
  6011.  
  6012. x86: XOP pixel_ssd_nv12_core
  6013.  
  6014. commit 9b77dffab04e3ea242598454282b40800e720353 [revision 2429]
  6015. Author: James Almer <[email protected]>
  6016. AuthorDate: Wed Apr 9 03:33:04 2014 -0300
  6017. Commit: Fiona Glaser <[email protected]>
  6018. CommitDate: Tue Apr 22 15:37:51 2014 -0700
  6019.  
  6020. x86util: XOP optimized HADDD
  6021.  
  6022. commit 1e517399f76b12fe2e73892970fe3aac01a178f8 [revision 2428]
  6023. Author: James Almer <[email protected]>
  6024. AuthorDate: Wed Apr 9 03:33:03 2014 -0300
  6025. Commit: Fiona Glaser <[email protected]>
  6026. CommitDate: Tue Apr 22 15:37:51 2014 -0700
  6027.  
  6028. x86: add missing initialization for high bit-depth sa8d_satd
  6029.  
  6030. commit aa00925abd6f9ab4e20216ae5a5ad79b67756162 [revision 2427]
  6031. Author: James Almer <[email protected]>
  6032. AuthorDate: Sat Apr 5 23:46:31 2014 -0300
  6033. Commit: Fiona Glaser <[email protected]>
  6034. CommitDate: Tue Apr 22 15:37:51 2014 -0700
  6035.  
  6036. x86: add missing initializations for high bit-depth variance
  6037.  
  6038. commit fadc4045f91ca78c046f301cba6065732b5d27ea [revision 2426]
  6039. Author: Janne Grunau <[email protected]>
  6040. AuthorDate: Tue Apr 1 22:11:45 2014 +0200
  6041. Commit: Fiona Glaser <[email protected]>
  6042. CommitDate: Tue Apr 22 15:37:50 2014 -0700
  6043.  
  6044. arm: use the weight_fn_t typedef for mc weight function arrays
  6045.  
  6046. commit 644c396be97c1e6ace144f8be04afab19fb238af [revision 2425]
  6047. Author: Janne Grunau <[email protected]>
  6048. AuthorDate: Tue Apr 1 22:11:44 2014 +0200
  6049. Commit: Fiona Glaser <[email protected]>
  6050. CommitDate: Tue Apr 22 15:37:50 2014 -0700
  6051.  
  6052. arm: correct x264_mc_chroma_neon function declaration
  6053.  
  6054. commit b2e9ca30f1e9ac25df1f592db04ff0d91faf42d4 [revision 2424]
  6055. Author: Janne Grunau <[email protected]>
  6056. AuthorDate: Tue Apr 1 22:11:43 2014 +0200
  6057. Commit: Fiona Glaser <[email protected]>
  6058. CommitDate: Tue Apr 22 15:37:50 2014 -0700
  6059.  
  6060. arm: do not export every asm function
  6061.  
  6062. Based on Libav's libavutil/arm/asm.S. Also prevents having the same
  6063. label twice for every function on systems not defining EXTERN_ASM.
  6064. Clang's integrated assembler does not like it.
  6065.  
  6066. commit ceb1484da34b7492f539b535a930652690372fe5 [revision 2423]
  6067. Author: Janne Grunau <[email protected]>
  6068. AuthorDate: Tue Apr 1 22:11:42 2014 +0200
  6069. Commit: Fiona Glaser <[email protected]>
  6070. CommitDate: Tue Apr 22 15:37:50 2014 -0700
  6071.  
  6072. arm: move all .macro/.endm to column 0
  6073.  
  6074. commit 24ab0e75db887c2b1a412d00878810ed6501061e [revision 2422]
  6075. Author: William Grant <[email protected]>
  6076. AuthorDate: Sun Mar 23 09:21:52 2014 -0700
  6077. Commit: Fiona Glaser <[email protected]>
  6078. CommitDate: Tue Apr 22 15:37:49 2014 -0700
  6079.  
  6080. aarch64: require PIC in shared mode
  6081.  
  6082. commit 435722c9c1870cd54fdb89be39250d492aecb598 [revision 2421]
  6083. Author: Janne Grunau <[email protected]>
  6084. AuthorDate: Sun Mar 16 17:21:58 2014 +0100
  6085. Commit: Fiona Glaser <[email protected]>
  6086. CommitDate: Tue Apr 22 15:37:49 2014 -0700
  6087.  
  6088. arm: x264_coeff_last8_arm
  6089.  
  6090. checkasm --bench on a coretex-a9:
  6091. coeff_last8_c: 173
  6092. coeff_last8_armv6: 151
  6093.  
  6094. 60 instead of 73 cycles in ~130k runs on the same cpu while encoding.
  6095.  
  6096. commit 2e96c571b8c324304b3d4fbb7914143518349213 [revision 2420]
  6097. Author: Janne Grunau <[email protected]>
  6098. AuthorDate: Sat Mar 15 20:09:18 2014 +0100
  6099. Commit: Fiona Glaser <[email protected]>
  6100. CommitDate: Tue Apr 22 15:37:49 2014 -0700
  6101.  
  6102. arm: x264_store_interleave_chroma_neon
  6103.  
  6104. store_interleave_chroma_c: 4036
  6105. store_interleave_chroma_neon: 1043
  6106.  
  6107. commit 1576e51e52148ad1e1d8b5e76562f9eae8d47e6e [revision 2419]
  6108. Author: Janne Grunau <[email protected]>
  6109. AuthorDate: Sat Mar 15 19:55:50 2014 +0100
  6110. Commit: Fiona Glaser <[email protected]>
  6111. CommitDate: Tue Apr 22 15:37:49 2014 -0700
  6112.  
  6113. arm: x264_plane_copy_interleave_neon
  6114.  
  6115. plane_copy_interleave_c: 40285
  6116. plane_copy_interleave_neon: 10137
  6117.  
  6118. commit 0016dec27080e53c794d7f919bd6df6b890d0128 [revision 2418]
  6119. Author: Janne Grunau <[email protected]>
  6120. AuthorDate: Sat Mar 15 19:21:12 2014 +0100
  6121. Commit: Fiona Glaser <[email protected]>
  6122. CommitDate: Tue Apr 22 15:37:49 2014 -0700
  6123.  
  6124. arm: x264_plane_copy_deinterleave_rgb_neon
  6125.  
  6126. plane_copy_deinterleave_rgb_c: 31543
  6127. plane_copy_deinterleave_rgb_neon: 8312
  6128.  
  6129. commit 5e0ca9aa4eab5e2cb4b124774c3ecebbc6f1ae35 [revision 2417]
  6130. Author: Janne Grunau <[email protected]>
  6131. AuthorDate: Sat Mar 15 18:22:49 2014 +0100
  6132. Commit: Fiona Glaser <[email protected]>
  6133. CommitDate: Tue Apr 22 15:37:48 2014 -0700
  6134.  
  6135. arm: load_deinterleave_chroma_f{dec,enc}_neon
  6136.  
  6137. load_deinterleave_chroma_fdec_c: 4055
  6138. load_deinterleave_chroma_fdec_neon: 995
  6139. load_deinterleave_chroma_fenc_c: 4071
  6140. load_deinterleave_chroma_fenc_neon: 992
  6141.  
  6142. commit c9a5ae0d219b6a28adebdb83faf89f291611f57b [revision 2416]
  6143. Author: Janne Grunau <[email protected]>
  6144. AuthorDate: Sat Mar 15 17:22:08 2014 +0100
  6145. Commit: Fiona Glaser <[email protected]>
  6146. CommitDate: Tue Apr 22 15:37:48 2014 -0700
  6147.  
  6148. arm: x264_plane_copy_deinterleave_neon
  6149.  
  6150. plane_copy_deinterleave_c: 42988
  6151. plane_copy_deinterleave_neon: 10184
  6152.  
  6153. commit c570be3ea9f24942c362e1c2402ec7fccbb5c330 [revision 2415]
  6154. Author: Janne Grunau <[email protected]>
  6155. AuthorDate: Sat Mar 15 13:29:41 2014 +0100
  6156. Commit: Fiona Glaser <[email protected]>
  6157. CommitDate: Tue Apr 22 15:37:48 2014 -0700
  6158.  
  6159. arm: implement deblock_strength_neon
  6160.  
  6161. Based on deblock_strength_avx.
  6162.  
  6163. checkasm --bench on a cortex-a9:
  6164. deblock_strength_c: 14611
  6165. deblock_strength_neon: 1848
  6166.  
  6167. commit 2794ba5bb0007e0edf32d5325ca82cbf654f79b0 [revision 2414]
  6168. Author: Janne Grunau <[email protected]>
  6169. AuthorDate: Sat Mar 15 10:51:11 2014 +0100
  6170. Commit: Fiona Glaser <[email protected]>
  6171. CommitDate: Tue Apr 22 15:37:48 2014 -0700
  6172.  
  6173. arm: add missing macro instantiation for x264_pixel_avg_4x16_neon
  6174.  
  6175. checkasm --bench on a cortex-a9:
  6176. avg_4x16_c: 8910
  6177. avg_4x16_neon: 2091
  6178.  
  6179. commit d6002ebace8194d17ee0ba607ff82c4f9075dd2d [revision 2413]
  6180. Author: Janne Grunau <[email protected]>
  6181. AuthorDate: Thu Mar 13 01:02:13 2014 +0100
  6182. Commit: Fiona Glaser <[email protected]>
  6183. CommitDate: Tue Apr 22 15:37:47 2014 -0700
  6184.  
  6185. arm: implement x264_predict_4x4_v_armv6
  6186.  
  6187. Alone probably not worth it but allows use of predict_4x4_dc|h_armv6
  6188. in intra_sad|satd_x3_4x4_neon.
  6189.  
  6190. commit d7e689680023e327de7e052e01e7faee30135799 [revision 2412]
  6191. Author: Roland Stigge <[email protected]>
  6192. AuthorDate: Sun Mar 23 09:29:37 2014 -0700
  6193. Commit: Fiona Glaser <[email protected]>
  6194. CommitDate: Tue Apr 22 15:37:47 2014 -0700
  6195.  
  6196. ppc: fix build on certain PowerPC variants without Altivec
  6197.  
  6198. commit 863ea2a224cf7380c7a6ea9ae531e16b621cc0b7 [revision 2411]
  6199. Author: Anton Mitrofanov <[email protected]>
  6200. AuthorDate: Tue Apr 22 00:58:24 2014 +0400
  6201. Commit: Fiona Glaser <[email protected]>
  6202. CommitDate: Tue Apr 22 15:37:47 2014 -0700
  6203.  
  6204. Only add strip option '-s' for linker flags
  6205.  
  6206. Fixes some build warnings with clang.
  6207.  
  6208. commit 4102614df9a11d66b506fb435132ddd0f88c6f94 [revision 2410]
  6209. Author: Tsukasa OMOTO <[email protected]>
  6210. AuthorDate: Sat Mar 15 16:53:53 2014 +0900
  6211. Commit: Fiona Glaser <[email protected]>
  6212. CommitDate: Mon Apr 21 13:53:44 2014 -0700
  6213.  
  6214. configure: remove an unnecessary option from CFLAGS on OS X
  6215.  
  6216. Fixes Clang 3.4 compilation on OS X.
  6217.  
  6218. commit b3fb718404d6cce9c82987ea2909cda5072d040c [revision 2409]
  6219. Author: Fiona Glaser <[email protected]>
  6220. AuthorDate: Sun Feb 23 10:36:55 2014 -0800
  6221. Commit: Fiona Glaser <[email protected]>
  6222. CommitDate: Thu Mar 13 11:43:17 2014 -0700
  6223.  
  6224. Macroblock tree overhaul/optimization
  6225.  
  6226. Move the second core part of macroblock tree into an assembly function;
  6227. SIMD-optimize roughly half of it (for x86). Roughly ~25-65% faster mbtree,
  6228. depending on content.
  6229.  
  6230. Slightly change how mbtree handles the tradeoff between range and precision
  6231. for propagation.
  6232.  
  6233. Overall a slight (but mostly negligible) effect on SSIM and ~2% faster.
  6234.  
  6235. commit 00a00ccab316de3d50da6a82ba4af44dcb4655ec [revision 2408]
  6236. Author: Janne Grunau <[email protected]>
  6237. AuthorDate: Thu Mar 13 00:05:48 2014 +0100
  6238. Commit: Fiona Glaser <[email protected]>
  6239. CommitDate: Thu Mar 13 11:43:16 2014 -0700
  6240.  
  6241. arm: use available neon functions for intra_sa8d/sad/satd_x3
  6242.  
  6243. 4% faster on main/medium, 15% faster on baseline/superfast on a cortex-a9.
  6244.  
  6245. commit ac8f2e8a4cf21b2026957509bea8865ff7879fb4 [revision 2407]
  6246. Author: Janne Grunau <[email protected]>
  6247. AuthorDate: Wed Mar 12 14:35:31 2014 +0100
  6248. Commit: Fiona Glaser <[email protected]>
  6249. CommitDate: Wed Mar 12 13:17:01 2014 -0700
  6250.  
  6251. arm: implement x264_pixel_var2_8x16_neon
  6252.  
  6253. checkasm --bench on a cortex-a9:
  6254. var2_8x16_c: 5677
  6255. var2_8x16_neon: 1421
  6256.  
  6257. commit 66836125beabdaff561da89ea1e18e566f5d202a [revision 2406]
  6258. Author: Janne Grunau <[email protected]>
  6259. AuthorDate: Wed Mar 12 13:16:00 2014 +0100
  6260. Commit: Fiona Glaser <[email protected]>
  6261. CommitDate: Wed Mar 12 13:17:01 2014 -0700
  6262.  
  6263. arm: implement x264_pixel_var_8x16_neon
  6264.  
  6265. checkasm --bench on a cortex-a9:
  6266. var_8x16_c: 4306
  6267. var_8x16_neon: 791
  6268.  
  6269. commit a90ea34cf264d6b7733c5ffbe6d46882c306b50f [revision 2405]
  6270. Author: Henrik Gramner <[email protected]>
  6271. AuthorDate: Sun Feb 23 15:33:48 2014 +0100
  6272. Commit: Fiona Glaser <[email protected]>
  6273. CommitDate: Tue Mar 11 21:11:05 2014 -0700
  6274.  
  6275. x86: SSE2 and SSSE3 plane_copy_deinterleave_rgb
  6276.  
  6277. About 5.6x faster than C on Haswell.
  6278.  
  6279. commit f032147ca69401165495a36cf7aba5b8c95ecb3b [revision 2404]
  6280. Author: Henrik Gramner <[email protected]>
  6281. AuthorDate: Sun Feb 16 21:24:54 2014 +0100
  6282. Commit: Fiona Glaser <[email protected]>
  6283. CommitDate: Tue Mar 11 21:11:05 2014 -0700
  6284.  
  6285. x86: Minor mbtree_propagate_cost improvements
  6286.  
  6287. Reduce the number of registers used from 7 to 6.
  6288. Reduce the number of vector registers used by the AVX2 implementation from 8 to 7.
  6289. Multiply fps_factor by 1/256 once per frame instead of once per macroblock row.
  6290. Use mova instead of movu for dst since it's guaranteed to be aligned.
  6291. Some cosmetics.
  6292.  
  6293. commit 7c860f075ccd14fb7891d5fc6c9eab1a37ea555d [revision 2403]
  6294. Author: Henrik Gramner <[email protected]>
  6295. AuthorDate: Sun Feb 9 23:58:04 2014 +0100
  6296. Commit: Fiona Glaser <[email protected]>
  6297. CommitDate: Tue Mar 11 21:11:04 2014 -0700
  6298.  
  6299. x86inc: Support arbitrary stack alignments
  6300.  
  6301. If the stack is known to be at least 32-byte aligned we can safely store ymm
  6302. registers on the stack without doing manual alignment.
  6303.  
  6304. Change ALLOC_STACK to always align the stack before allocating stack space for
  6305. consistency. Previously alignment would occur either before or after allocating
  6306. stack space depending on whether manual alignment was required or not.
  6307.  
  6308. commit 039fab9203179f9e790abfd54ae5b2254ef803e7 [revision 2402]
  6309. Author: Anton Mitrofanov <[email protected]>
  6310. AuthorDate: Fri Feb 14 15:53:58 2014 +0400
  6311. Commit: Fiona Glaser <[email protected]>
  6312. CommitDate: Tue Mar 11 21:10:23 2014 -0700
  6313.  
  6314. x86inc: warn if XOP integer FMA instruction emulation is impossible
  6315.  
  6316. Emulation requires a temporary register if arguments 1 and 4 are the same; this
  6317. doesn't obey the semantics of the original instruction, so we can't emulate
  6318. that in x86inc.
  6319.  
  6320. ffmpeg has an x86util emulation for that case; I'll add it if x264's asm ever
  6321. needs it.
  6322.  
  6323. Also add pmacsdql emulation.
  6324.  
  6325. commit 974f2e78e0cb25e06fedbcfef70f80938f22988b [revision 2401]
  6326. Author: Loren Merritt <[email protected]>
  6327. AuthorDate: Sat Mar 1 02:57:56 2014 +0000
  6328. Commit: Fiona Glaser <[email protected]>
  6329. CommitDate: Tue Mar 11 21:10:23 2014 -0700
  6330.  
  6331. x86inc: free up variable name "n" in global namespace
  6332.  
  6333. commit 8596dd36df38d33d402e848035b1bd31edc2c389 [revision 2400]
  6334. Author: Henrik Gramner <[email protected]>
  6335. AuthorDate: Wed Jan 22 19:09:12 2014 +0100
  6336. Commit: Fiona Glaser <[email protected]>
  6337. CommitDate: Tue Mar 11 21:10:22 2014 -0700
  6338.  
  6339. x86: Pass -Worphan-labels to yasm
  6340.  
  6341. Makes it easier to detect typos.
  6342.  
  6343. commit 0bb3b2edb866dd852bb1f5faed88df4bdcf0c16f [revision 2399]
  6344. Author: Steve Lhomme <[email protected]>
  6345. AuthorDate: Sun Feb 16 13:15:09 2014 +0100
  6346. Commit: Fiona Glaser <[email protected]>
  6347. CommitDate: Tue Mar 11 21:10:22 2014 -0700
  6348.  
  6349. Write 3D metadata when outputting Matroska
  6350.  
  6351. For when --frame-packing is set.
  6352.  
  6353. commit f35e3fc26b99e1b3c943c131100fdfa4733fc932 [revision 2398]
  6354. Author: Anton Mitrofanov <[email protected]>
  6355. AuthorDate: Sun Feb 23 16:56:03 2014 +0400
  6356. Commit: Fiona Glaser <[email protected]>
  6357. CommitDate: Tue Mar 11 21:10:22 2014 -0700
  6358.  
  6359. Don't set chroma_loc_info_present_flag for non-4:2:0
  6360.  
  6361. The H.264 spec says it shouldn't be set in these cases.
  6362.  
  6363. commit b7a50c16414631c8ff5e417da51b190c8999027e [revision 2397]
  6364. Author: Fiona Glaser <[email protected]>
  6365. AuthorDate: Mon Mar 10 08:42:50 2014 -0700
  6366. Commit: Fiona Glaser <[email protected]>
  6367. CommitDate: Tue Mar 11 21:10:22 2014 -0700
  6368.  
  6369. x264.h: fix documentation
  6370.  
  6371. The full details of the return values of encoder_encode and encoder_headers
  6372. were mistakenly removed a while ago; re-add them.
  6373.  
  6374. commit de01d8821b59b85a01c8a89e544e0fed6488b958 [revision 2396]
  6375. Author: Anton Mitrofanov <[email protected]>
  6376. AuthorDate: Sun Feb 23 15:52:57 2014 +0400
  6377. Commit: Fiona Glaser <[email protected]>
  6378. CommitDate: Tue Mar 11 21:10:20 2014 -0700
  6379.  
  6380. Fix pointer cast warning for 64-bit builds
  6381.  
  6382. commit 8b821ec19ba9425c120b8986a57ca7c6b9f088ed [revision 2395]
  6383. Author: Anton Mitrofanov <[email protected]>
  6384. AuthorDate: Mon Mar 10 16:48:02 2014 +0400
  6385. Commit: Fiona Glaser <[email protected]>
  6386. CommitDate: Tue Mar 11 14:48:15 2014 -0700
  6387.  
  6388. mbaff: fix mb_field_decoding_flag tracking and simplify allow skip check
  6389.  
  6390. Fixes an issue with too many forced non-skips in mbaff+cavlc, as well as
  6391. non-deterministic output with mbaff+cavlc+sliced-threads.
  6392.  
  6393. commit 850c8c5d6139df82e969d2174eebba69b479aa16 [revision 2394]
  6394. Author: Anton Mitrofanov <[email protected]>
  6395. AuthorDate: Mon Mar 10 03:22:57 2014 +0400
  6396. Commit: Fiona Glaser <[email protected]>
  6397. CommitDate: Tue Mar 11 14:48:14 2014 -0700
  6398.  
  6399. Fix memory overwrite in x264_deblock_h_chroma_mbaff_sse2
  6400.  
  6401. Fixes possible corruption with MBAFF+sliced threads.
  6402.  
  6403. commit 19dddbcff73541ae15f8e57383ff1c6aa907d99d [revision 2393]
  6404. Author: Fiona Glaser <[email protected]>
  6405. AuthorDate: Sun Mar 2 10:09:01 2014 -0800
  6406. Commit: Fiona Glaser <[email protected]>
  6407. CommitDate: Tue Mar 11 14:48:14 2014 -0700
  6408.  
  6409. Fix corruption with CAVLC overflow handling in MBAFF+main profile
  6410.  
  6411. Probably a regression in r2178.
  6412.  
  6413. commit 48dbfa28201950f7e07e96a7d62b2951dd2dbe03 [revision 2392]
  6414. Author: Anton Mitrofanov <[email protected]>
  6415. AuthorDate: Mon Mar 10 21:17:19 2014 +0400
  6416. Commit: Fiona Glaser <[email protected]>
  6417. CommitDate: Tue Mar 11 14:48:05 2014 -0700
  6418.  
  6419. Fix checkasm --bench output when nop_cycles is too large
  6420.  
  6421. commit ee8d5e4b51da99e576b5aea3008e70d1c7ed2372 [revision 2391]
  6422. Author: Anton Mitrofanov <[email protected]>
  6423. AuthorDate: Wed Jan 22 12:54:49 2014 +0400
  6424. Commit: Fiona Glaser <[email protected]>
  6425. CommitDate: Sun Feb 23 19:14:03 2014 -0800
  6426.  
  6427. Really fix quantization factor allocation
  6428.  
  6429. Actually allocate less (instead of just initialize less) and fix comments.
  6430.  
  6431. commit 0d668be8d7525992c1c163c97551ee897e43c177 [revision 2390]
  6432. Author: Yu Xiaolei <[email protected]>
  6433. AuthorDate: Sun Feb 23 04:12:51 2014 -0800
  6434. Commit: Fiona Glaser <[email protected]>
  6435. CommitDate: Sun Feb 23 19:14:00 2014 -0800
  6436.  
  6437. Fix build with Android NDK
  6438.  
  6439. Android NDK does not expose sched_getaffinity.
  6440.  
  6441. commit 42d25196d423626c12794db3f66322c7a3f4375e [revision 2389]
  6442. Author: Loren Merritt <[email protected]>
  6443. AuthorDate: Thu Jan 16 13:34:46 2014 -0800
  6444. Commit: Fiona Glaser <[email protected]>
  6445. CommitDate: Tue Jan 21 13:40:15 2014 -0800
  6446.  
  6447. x86inc: speed up compilation with yasm
  6448.  
  6449. Work around yasm's inefficiency with handling large numbers of variables
  6450. in the global scope.
  6451.  
  6452. commit dd6a303498d1f55c73037ed925a6ece8e28a95bc [revision 2388]
  6453. Author: Kieran Kunhya <[email protected]>
  6454. AuthorDate: Fri Jan 10 23:27:33 2014 +0000
  6455. Commit: Fiona Glaser <[email protected]>
  6456. CommitDate: Tue Jan 21 13:40:14 2014 -0800
  6457.  
  6458. Add support for AVC-Intra Class 200
  6459.  
  6460. commit 41227fa2531d9263e481b80237d2d9ef6f5a450f [revision 2387]
  6461. Author: James Weaver <[email protected]>
  6462. AuthorDate: Tue Jan 7 10:31:58 2014 +0000
  6463. Commit: Fiona Glaser <[email protected]>
  6464. CommitDate: Tue Jan 21 13:40:13 2014 -0800
  6465.  
  6466. v210 input support
  6467.  
  6468. Assembly based on code by Henrik Gramner and Loren Merritt.
  6469.  
  6470. commit e2a9662751180b7dd2fe538913282ee800445445 [revision 2386]
  6471. Author: Fiona Glaser <[email protected]>
  6472. AuthorDate: Tue Jan 21 13:39:33 2014 -0800
  6473. Commit: Fiona Glaser <[email protected]>
  6474. CommitDate: Tue Jan 21 13:40:12 2014 -0800
  6475.  
  6476. Fix quantization factor allocation
  6477.  
  6478. We don't need to wastefully allocate quant tables above QP_MAX_SPEC; they're
  6479. never used.
  6480.  
  6481. commit 8be6600d10a74ca241dbb27e096883ceed7b4082 [revision 2385]
  6482. Author: Henrik Gramner <[email protected]>
  6483. AuthorDate: Wed Jan 8 01:06:56 2014 +0100
  6484. Commit: Fiona Glaser <[email protected]>
  6485. CommitDate: Wed Jan 8 11:15:46 2014 -0800
  6486.  
  6487. Avoid some unneccesary memory loads in macroblock_encode
  6488.  
  6489. commit 807aeaaae7351e4c2c536463e69dacaac218bccb [revision 2384]
  6490. Author: Henrik Gramner <[email protected]>
  6491. AuthorDate: Sun Jan 5 15:25:05 2014 +0100
  6492. Commit: Fiona Glaser <[email protected]>
  6493. CommitDate: Wed Jan 8 11:15:45 2014 -0800
  6494.  
  6495. Bump dates to 2014
  6496.  
  6497. Also update AUTHORS file and my e-mail address in the headers of various files.
  6498.  
  6499. commit 02697d57d987f8d51a5c3ced5e5b81d7137012ee [revision 2383]
  6500. Author: Henrik Gramner <[email protected]>
  6501. AuthorDate: Mon Jan 6 00:18:31 2014 +0100
  6502. Commit: Fiona Glaser <[email protected]>
  6503. CommitDate: Wed Jan 8 11:15:45 2014 -0800
  6504.  
  6505. Remove tools/xyuv.c
  6506.  
  6507. It's an old stand-alone application that isn't relevant to x264.
  6508.  
  6509. commit 7664014b2b490d81a66f2a13138182dfaaf4be06 [revision 2382]
  6510. Author: Anton Mitrofanov <[email protected]>
  6511. AuthorDate: Thu Nov 7 02:37:23 2013 +0400
  6512. Commit: Fiona Glaser <[email protected]>
  6513. CommitDate: Wed Jan 8 11:15:44 2014 -0800
  6514.  
  6515. Use 8x16c wrappers with x86 asm functions for 4:2:2 with high bit depth
  6516.  
  6517. commit 6bc63417e10e135d8cd881495c71be72d322e1d3 [revision 2381]
  6518. Author: Henrik Gramner <[email protected]>
  6519. AuthorDate: Fri Dec 20 22:44:28 2013 +0100
  6520. Commit: Fiona Glaser <[email protected]>
  6521. CommitDate: Wed Jan 8 11:15:44 2014 -0800
  6522.  
  6523. CLI: Avoid redundant 16-bit upconversions in piped raw input
  6524.  
  6525. It's not possible to seek in pipes, so if we want to skip frames we have to read and
  6526. discard unused ones. It's pointless to do bit-depth upconversions in those frames.
  6527.  
  6528. commit 008c56ec467736bc5d3130ff890c618d28aa7511 [revision 2380]
  6529. Author: Anton Mitrofanov <[email protected]>
  6530. AuthorDate: Fri Jan 3 20:06:06 2014 +0400
  6531. Commit: Fiona Glaser <[email protected]>
  6532. CommitDate: Wed Jan 8 11:15:44 2014 -0800
  6533.  
  6534. Fix input support from named pipes in Windows
  6535.  
  6536. commit 91481419e3acc4bb601600cf32e46e7f93ae02ab [revision 2379]
  6537. Author: Steve Clark <[email protected]>
  6538. AuthorDate: Wed Nov 20 21:40:23 2013 +0400
  6539. Commit: Fiona Glaser <[email protected]>
  6540. CommitDate: Wed Jan 8 11:15:36 2014 -0800
  6541.  
  6542. Fix ARM asm compilation with Apple assembler
  6543.  
  6544. commit a2f5d600bf866899db92e2dae40eb9fe46d44ade [revision 2378]
  6545. Author: Anton Mitrofanov <[email protected]>
  6546. AuthorDate: Wed Nov 13 19:24:48 2013 +0400
  6547. Commit: Fiona Glaser <[email protected]>
  6548. CommitDate: Mon Jan 6 12:15:32 2014 -0800
  6549.  
  6550. Fix uninitialized variable
  6551.  
  6552. Caused if the timebase is not specified in stats file. Found by Clang.
  6553.  
  6554. commit 95d196ef2edde109cfb32f4baa9b0adc67e842e1 [revision 2377]
  6555. Author: Anton Mitrofanov <[email protected]>
  6556. AuthorDate: Sun Oct 27 19:27:23 2013 +0400
  6557. Commit: Fiona Glaser <[email protected]>
  6558. CommitDate: Wed Oct 30 13:02:28 2013 -0700
  6559.  
  6560. Remove --visualize option.
  6561.  
  6562. It probably wasn't used or maintained for last few years.
  6563.  
  6564. commit 09c7010e3d13e66a241c0529b36ae3f7e1664ff4 [revision 2376]
  6565. Author: Anton Mitrofanov <[email protected]>
  6566. AuthorDate: Tue Oct 15 12:32:25 2013 +0400
  6567. Commit: Fiona Glaser <[email protected]>
  6568. CommitDate: Wed Oct 30 13:02:28 2013 -0700
  6569.  
  6570. Add L-SMASH support as preferable alternative for MP4-muxing
  6571.  
  6572. commit c9f2bceb1f37aeaf6b7ed730f0fd210ef8725cab [revision 2375]
  6573. Author: Kieran Kunhya <[email protected]>
  6574. AuthorDate: Sat Sep 21 19:16:12 2013 +0100
  6575. Commit: Fiona Glaser <[email protected]>
  6576. CommitDate: Wed Oct 30 13:02:27 2013 -0700
  6577.  
  6578. Add AVC-Intra 1080p50/60 Class 100 parameters
  6579.  
  6580. Also add some compatibility fixes.
  6581.  
  6582. commit c084f6c029f016cf2024a2fc511825e82fb95865 [revision 2374]
  6583. Author: Fiona Glaser <[email protected]>
  6584. AuthorDate: Mon Sep 9 12:37:59 2013 -0700
  6585. Commit: Fiona Glaser <[email protected]>
  6586. CommitDate: Wed Oct 30 13:02:27 2013 -0700
  6587.  
  6588. Add --filler option
  6589.  
  6590. Allows generation of hard-CBR streams without using NAL HRD.
  6591. Useful if you want to be able to reconfigure the bitrate (which you can't do
  6592. with NAL HRD on).
  6593.  
  6594. commit 350b214c5abe7e82618ac46a14f23b7ab543045e [revision 2373]
  6595. Author: Anton Mitrofanov <[email protected]>
  6596. AuthorDate: Sun Oct 27 15:22:51 2013 +0400
  6597. Commit: Fiona Glaser <[email protected]>
  6598. CommitDate: Wed Oct 30 13:02:27 2013 -0700
  6599.  
  6600. Make x264_encoder_reconfig more threadsafe
  6601.  
  6602. Do the reconfig when the next frame's encode begins.
  6603. Fixes some rare crashes with frame-threading and encoder_reconfig.
  6604.  
  6605. commit 77cc44feea75106fae6d3113f6babbbe8cffba87 [revision 2372]
  6606. Author: Fiona Glaser <[email protected]>
  6607. AuthorDate: Thu Oct 24 17:19:00 2013 -0700
  6608. Commit: Fiona Glaser <[email protected]>
  6609. CommitDate: Fri Oct 25 10:35:38 2013 -0700
  6610.  
  6611. chroma-me: take shortcut in BI analysis
  6612.  
  6613. ~100 cycles faster with subme>=9
  6614.  
  6615. commit 7634f8c6047e9e12036778a8dc8d4cd4b06eebcb [revision 2371]
  6616. Author: Fiona Glaser <[email protected]>
  6617. AuthorDate: Thu Oct 24 14:44:43 2013 -0700
  6618. Commit: Fiona Glaser <[email protected]>
  6619. CommitDate: Fri Oct 25 10:35:38 2013 -0700
  6620.  
  6621. CRF-max: don't warn if VBV underflow occurs
  6622.  
  6623. Only warn if underflow occurs for reasons other than CRF-max, as CRF-max
  6624. implies that VBV underflow is desired by the user.
  6625.  
  6626. commit 4b68633dc375fc372f160a3ae669a32e519b285a [revision 2370]
  6627. Author: Henrik Gramner <[email protected]>
  6628. AuthorDate: Fri Oct 18 22:43:36 2013 +0200
  6629. Commit: Fiona Glaser <[email protected]>
  6630. CommitDate: Fri Oct 25 10:35:37 2013 -0700
  6631.  
  6632. x86inc: Make ym# behave the same way as xm#
  6633.  
  6634. This makes more sense for future implementations of templates with zmm registers.
  6635.  
  6636. commit b54422a858809f39c00fac46207bfa8ad16cdb28 [revision 2369]
  6637. Author: Henrik Gramner <[email protected]>
  6638. AuthorDate: Fri Oct 18 22:21:38 2013 +0200
  6639. Commit: Fiona Glaser <[email protected]>
  6640. CommitDate: Fri Oct 25 10:35:36 2013 -0700
  6641.  
  6642. Use calloc instead of malloc + memset
  6643.  
  6644. commit 8b58a4ce52047b00f5892a9cdd92f9695a50a933 [revision 2368]
  6645. Author: Henrik Gramner <[email protected]>
  6646. AuthorDate: Thu Oct 10 16:54:12 2013 +0200
  6647. Commit: Fiona Glaser <[email protected]>
  6648. CommitDate: Fri Oct 25 10:35:35 2013 -0700
  6649.  
  6650. Replace gf_malloc with regular malloc in mp4 muxer
  6651.  
  6652. It was used as a workaround for a bug that only existed in the GPAC repository
  6653. for a few weeks back in 2010. There's no reason to keep it anymore.
  6654.  
  6655. commit 05f04384a10cb673abea7749cd319971c0017769 [revision 2367]
  6656. Author: Anton Mitrofanov <[email protected]>
  6657. AuthorDate: Tue Oct 8 23:20:40 2013 +0400
  6658. Commit: Fiona Glaser <[email protected]>
  6659. CommitDate: Fri Oct 25 10:35:35 2013 -0700
  6660.  
  6661. Update to current libav/ffmpeg API
  6662.  
  6663. commit b7b6029f0e121b87fd96595b15d0c40fcd1b3bf1 [revision 2366]
  6664. Author: Rafaël Carré <[email protected]>
  6665. AuthorDate: Fri Oct 25 07:12:24 2013 -0700
  6666. Commit: Fiona Glaser <[email protected]>
  6667. CommitDate: Fri Oct 25 10:35:28 2013 -0700
  6668.  
  6669. version.sh: change to use /bin/sh
  6670.  
  6671. commit c3c73f13bb9ee60ccf40f85dbc11c91efac9d1e2 [revision 2365]
  6672. Author: Sean McGovern <[email protected]>
  6673. AuthorDate: Wed Sep 4 14:15:00 2013 -0700
  6674. Commit: Fiona Glaser <[email protected]>
  6675. CommitDate: Thu Oct 24 12:15:58 2013 -0700
  6676.  
  6677. configure: don't generate a git version number if .git isn't present
  6678.  
  6679. commit 12f9d499905199427a0196743c2cde56642d6d99 [revision 2364]
  6680. Author: Martin Storsjo <[email protected]>
  6681. AuthorDate: Tue Sep 3 14:56:18 2013 -0700
  6682. Commit: Fiona Glaser <[email protected]>
  6683. CommitDate: Thu Oct 24 12:15:58 2013 -0700
  6684.  
  6685. configure: include dependency libs in the Libs pkg-config
  6686.  
  6687. If only a static library is built, the user of the library that just
  6688. tries to link to the lib using the flags provided by pkg-config
  6689. might not know that only a static lib exists and that he'd have to
  6690. pass --static to pkg-config to get the internal dependencies to
  6691. be able to link the library.
  6692.  
  6693. For a shared build, the internal dependencies are kept in Libs.private
  6694. as before.
  6695.  
  6696. This matches how libav's pkg-config files are generated.
  6697.  
  6698. commit 03450be799dea03a83dad4dc833ef8ddd7f36b62 [revision 2363]
  6699. Author: Anton Mitrofanov <[email protected]>
  6700. AuthorDate: Fri Oct 18 00:38:06 2013 +0400
  6701. Commit: Fiona Glaser <[email protected]>
  6702. CommitDate: Thu Oct 24 12:15:57 2013 -0700
  6703.  
  6704. Fix compilation in case of HAVE_LOG2F check fails spuriously
  6705.  
  6706. commit 266fdfcd4809afb018e45ab959d4a56a42712c88 [revision 2362]
  6707. Author: Anton Mitrofanov <[email protected]>
  6708. AuthorDate: Sat Oct 12 12:01:57 2013 +0400
  6709. Commit: Fiona Glaser <[email protected]>
  6710. CommitDate: Thu Oct 24 12:15:57 2013 -0700
  6711.  
  6712. Fix compilation of shared library for Windows with original MinGW toolchain
  6713.  
  6714. commit 50a0c33b9b5fa57d0a129b7441a6af55f7a08005 [revision 2361]
  6715. Author: Anton Mitrofanov <[email protected]>
  6716. AuthorDate: Tue Oct 8 23:32:37 2013 +0400
  6717. Commit: Fiona Glaser <[email protected]>
  6718. CommitDate: Thu Oct 24 12:15:57 2013 -0700
  6719.  
  6720. Fix possible crashes in resize and crop filters with high bitdepth input
  6721.  
  6722. commit 5b272b22d8f7511a4abece5a23ad25282bedaea8 [revision 2360]
  6723. Author: Tim Mooney <[email protected]>
  6724. AuthorDate: Tue Sep 3 13:43:50 2013 -0700
  6725. Commit: Fiona Glaser <[email protected]>
  6726. CommitDate: Tue Sep 3 15:55:19 2013 -0700
  6727.  
  6728. Fix INSTALL in configure for Solaris systems
  6729.  
  6730. commit 2fd292391a4d41b9fc65ee652b4663fdd9f8107e [revision 2359]
  6731. Author: Henrik Gramner <[email protected]>
  6732. AuthorDate: Wed Aug 28 00:50:31 2013 +0200
  6733. Commit: Fiona Glaser <[email protected]>
  6734. CommitDate: Tue Aug 27 16:05:38 2013 -0700
  6735.  
  6736. Workaround for FFMS indexing bug
  6737.  
  6738. If FFMS_ReadIndex is used with an empty index file it gets stuck in an infinite loop instead of returning NULL
  6739. like it's supposed to do on failure. Explicitly check if the file is empty before calling it as a workaround.
  6740.  
  6741. commit 5bcff2a62c050376ca54c5e5040d0529c89eb9f2 [revision 2358]
  6742. Author: Anton Mitrofanov <[email protected]>
  6743. AuthorDate: Mon Aug 26 21:20:31 2013 +0400
  6744. Commit: Fiona Glaser <[email protected]>
  6745. CommitDate: Mon Aug 26 10:48:02 2013 -0700
  6746.  
  6747. Fix masked access violation in KERNEL32
  6748.  
  6749. Caused crashes under gdb in Windows and might cause other unknown problems.
  6750.  
  6751. commit 098b686e6397d5bb6b3a5c03cd918aa88216909f [revision 2357]
  6752. Author: Hiroki Taniura <[email protected]>
  6753. AuthorDate: Sun Aug 25 01:18:57 2013 +0900
  6754. Commit: Fiona Glaser <[email protected]>
  6755. CommitDate: Sat Aug 24 09:42:47 2013 -0700
  6756.  
  6757. Fix GPAC support on Windows
  6758.  
  6759. commit fa3cac516cb71b8ece09cedbfd0ce631ca8a2a4c [revision 2356]
  6760. Author: Henrik Gramner <[email protected]>
  6761. AuthorDate: Sun Aug 11 19:50:42 2013 +0200
  6762. Commit: Fiona Glaser <[email protected]>
  6763. CommitDate: Fri Aug 23 14:04:14 2013 -0700
  6764.  
  6765. Windows Unicode support
  6766.  
  6767. Windows, unlike most other operating systems, uses UTF-16 for Unicode strings while x264 is designed for UTF-8.
  6768.  
  6769. This patch does the following in order to handle things like Unicode filenames:
  6770. * Keep strings internally as UTF-8.
  6771. * Retrieve the CLI command line as UTF-16 and convert it to UTF-8.
  6772. * Always use Unicode versions of Windows API functions and convert strings to UTF-16 when calling them.
  6773. * Attempt to use legacy 8.3 short filenames for external libraries without Unicode support.
  6774.  
  6775. commit 9b94896b3735052cabb52d081de3b50020a077cb [revision 2355]
  6776. Author: Kieran Kunhya <[email protected]>
  6777. AuthorDate: Sat Jul 20 18:47:59 2013 +0100
  6778. Commit: Fiona Glaser <[email protected]>
  6779. CommitDate: Fri Aug 23 14:04:13 2013 -0700
  6780.  
  6781. AVC-Intra support
  6782.  
  6783. This format has been reverse engineered and x264's output has almost exactly
  6784. the same bitstream as Panasonic cameras and encoders produce. It therefore does
  6785. not comply with SMPTE RP2027 since Panasonic themselves do not comply with
  6786. their own specification. It has been tested in Avid, Premiere, Edius and
  6787. Quantel.
  6788.  
  6789. Parts of this patch were written by Fiona Glaser and some reverse
  6790. engineering was done by Joseph Artsimovich.
  6791.  
  6792. commit fa1e2b746d95575b5c5b8e49fcfcad3ded9a5420 [revision 2354]
  6793. Author: Henrik Gramner <[email protected]>
  6794. AuthorDate: Mon Jul 8 12:06:42 2013 -0700
  6795. Commit: Fiona Glaser <[email protected]>
  6796. CommitDate: Fri Aug 23 14:04:12 2013 -0700
  6797.  
  6798. Transparent hugepage support
  6799.  
  6800. Combine frame and mb data mallocs into a single large malloc.
  6801. Additionally, on Linux systems with hugepage support, ask for hugepages on
  6802. large mallocs.
  6803.  
  6804. This gives a small performance improvement (~0.2-0.9%) on systems without
  6805. hugepage support, as well as a small memory footprint reduction.
  6806.  
  6807. On recent Linux kernels with hugepage support enabled (set to madvise or
  6808. always), it improves performance up to 4% at the cost of about 7-12% more
  6809. memory usage on typical settings..
  6810.  
  6811. It may help even more on Haswell and other recent CPUs with improved 2MB page
  6812. support in hardware.
  6813.  
  6814. commit e33aac9aba5c6b9c867b92f14c7722152680a61a [revision 2353]
  6815. Author: Henrik Gramner <[email protected]>
  6816. AuthorDate: Fri Jul 5 21:15:54 2013 +0200
  6817. Commit: Fiona Glaser <[email protected]>
  6818. CommitDate: Fri Aug 23 14:04:12 2013 -0700
  6819.  
  6820. x86: SSSE3 implementation of pixel_sad_x3 and pixel_sad_x4
  6821.  
  6822. commit 4becc3e9e031c4207698846369aac2bef1480d15 [revision 2352]
  6823. Author: Henrik Gramner <[email protected]>
  6824. AuthorDate: Fri Jul 5 21:15:49 2013 +0200
  6825. Commit: Fiona Glaser <[email protected]>
  6826. CommitDate: Fri Aug 23 14:04:12 2013 -0700
  6827.  
  6828. x86: Faster AVX2 pixel_sad_x3 and pixel_sad_x4
  6829.  
  6830. commit 401edc3ab08f95777d495b38030e2108d7d3f0b4 [revision 2351]
  6831. Author: Diogo Franco <[email protected]>
  6832. AuthorDate: Tue Jul 23 22:17:44 2013 -0300
  6833. Commit: Fiona Glaser <[email protected]>
  6834. CommitDate: Fri Aug 23 14:04:11 2013 -0700
  6835.  
  6836. configure: Support cygwin64
  6837.  
  6838. commit adc99d17d8c1fbc164fae8319b40d7c45f30314e [revision 2350]
  6839. Author: Derek Buitenhuis <[email protected]>
  6840. AuthorDate: Fri Aug 9 13:39:27 2013 -0400
  6841. Commit: Fiona Glaser <[email protected]>
  6842. CommitDate: Fri Aug 23 14:04:11 2013 -0700
  6843.  
  6844. x86inc: Check for __OUTPUT_FORMAT__ having a value of "x64"
  6845.  
  6846. This is also a valid value for WIN64.
  6847.  
  6848. commit 1430b04988c3bb344e104c974ed3aa825035c0ec [revision 2349]
  6849. Author: Anton Mitrofanov <[email protected]>
  6850. AuthorDate: Tue Jul 23 14:11:50 2013 -0700
  6851. Commit: Fiona Glaser <[email protected]>
  6852. CommitDate: Fri Aug 23 14:04:10 2013 -0700
  6853.  
  6854. Fix cases in which intra refresh allowed prediction from disallowed pixels
  6855.  
  6856. commit a6c396f0fe01f453de115ba0d8c4aa26138aa6b4 [revision 2348]
  6857. Author: Anton Mitrofanov <[email protected]>
  6858. AuthorDate: Wed Aug 7 01:56:34 2013 +0400
  6859. Commit: Fiona Glaser <[email protected]>
  6860. CommitDate: Fri Aug 23 14:04:10 2013 -0700
  6861.  
  6862. Fix a few minor bugs found with a static analyzer
  6863.  
  6864. commit 2d66c7c2471801aa946517226739e9150f6c1948 [revision 2347]
  6865. Author: Fiona Glaser <[email protected]>
  6866. AuthorDate: Fri Jul 12 16:07:35 2013 -0700
  6867. Commit: Fiona Glaser <[email protected]>
  6868. CommitDate: Fri Aug 23 14:04:09 2013 -0700
  6869.  
  6870. Fix AVX2 detection bug with "limit CPUID" enabled in BIOS
  6871.  
  6872. commit ff41804efd4caec120fc9e1b90ad226035f75aaa [revision 2346]
  6873. Author: Henrik Gramner <[email protected]>
  6874. AuthorDate: Fri Jul 5 21:15:43 2013 +0200
  6875. Commit: Fiona Glaser <[email protected]>
  6876. CommitDate: Fri Jul 5 13:10:06 2013 -0700
  6877.  
  6878. x86: Remove X264_CPU_SSE_MISALIGN functions
  6879.  
  6880. Prevents a crash if the misaligned exception mask bit is cleared for some reason.
  6881.  
  6882. Misaligned SSE functions are only used on AMD Phenom CPUs and the benefit is miniscule.
  6883. They also require modifying the MXCSR control register and by removing those functions
  6884. we can get rid of that complexity altogether.
  6885.  
  6886. VEX-encoded instructions also supports unaligned memory operands. I tried adding AVX
  6887. implementations of all removed functions but there were no performance improvements on
  6888. Ivy Bridge. pixel_sad_x3 and pixel_sad_x4 had significant code size reductions though
  6889. so I kept them and added some minor cosmetics fixes and tweaks.
  6890.  
  6891. commit 01087fdbf2042095cb36458fd5c5efab3f4b3a37 [revision 2345]
  6892. Author: Fiona Glaser <[email protected]>
  6893. AuthorDate: Thu Jun 20 15:51:39 2013 -0700
  6894. Commit: Fiona Glaser <[email protected]>
  6895. CommitDate: Tue Jul 2 17:13:10 2013 -0700
  6896.  
  6897. Tweak i16x16-delta-quant-avoidance code
  6898.  
  6899. Don't omit the delta quant if it'd raise the quantizer to do so; this fixes
  6900. a rare flickering issue caused by deblocking.
  6901.  
  6902. commit bfa2f0c44cec2e41fbd7566edb55e405f6c5a49d [revision 2344]
  6903. Author: Fiona Glaser <[email protected]>
  6904. AuthorDate: Sun Jun 9 09:06:27 2013 -0700
  6905. Commit: Fiona Glaser <[email protected]>
  6906. CommitDate: Tue Jul 2 17:13:09 2013 -0700
  6907.  
  6908. x86: faster AVX2 iDCT, AVX deblock_luma_h, deblock_luma_h_intra
  6909.  
  6910. commit 397f60e7f23e2c6ec2cb9b168ebb75cc42983dd7 [revision 2343]
  6911. Author: Lucien <[email protected]>
  6912. AuthorDate: Mon Jun 17 18:28:09 2013 +0000
  6913. Commit: Fiona Glaser <[email protected]>
  6914. CommitDate: Tue Jul 2 17:13:09 2013 -0700
  6915.  
  6916. Add new color primaries, transfer characteristics, matrix coefficients
  6917.  
  6918. commit fa215fc9d77d131595e8b1eda0fc4e9da62c1f94 [revision 2342]
  6919. Author: Fiona Glaser <[email protected]>
  6920. AuthorDate: Fri May 31 17:01:29 2013 -0700
  6921. Commit: Fiona Glaser <[email protected]>
  6922. CommitDate: Tue Jul 2 17:13:08 2013 -0700
  6923.  
  6924. Add "--stitchable" option for segmented encoding
  6925.  
  6926. Stops x264 from attempting to optimize global stream headers, ensuring that
  6927. different segments of a video will have identical headers when used with
  6928. identical encoding settings.
  6929.  
  6930. commit 9143d5ad966a3864597009ba1f1befe87328ec61 [revision 2341]
  6931. Author: Fiona Glaser <[email protected]>
  6932. AuthorDate: Thu Jun 27 08:29:06 2013 -0700
  6933. Commit: Fiona Glaser <[email protected]>
  6934. CommitDate: Tue Jul 2 17:13:08 2013 -0700
  6935.  
  6936. Interface: if vbv-maxrate < bitrate, set bitrate = vbv-maxrate
  6937.  
  6938. This probably makes more sense to the user than setting vbv-maxrate = bitrate,
  6939. as before.
  6940.  
  6941. commit 83d35c7bc4332e4dd27ba7b8baf96f8743c52a8b [revision 2340]
  6942. Author: Anton Mitrofanov <[email protected]>
  6943. AuthorDate: Tue May 28 05:02:42 2013 -0700
  6944. Commit: Fiona Glaser <[email protected]>
  6945. CommitDate: Tue Jul 2 17:13:08 2013 -0700
  6946.  
  6947. OpenCL cosmetics
  6948.  
  6949. commit ffc3ad4945da69f3caa2b40e4eed715a9a8d9526 [revision 2339]
  6950. Author: Anton Mitrofanov <[email protected]>
  6951. AuthorDate: Tue Jun 18 00:16:33 2013 +0400
  6952. Commit: Fiona Glaser <[email protected]>
  6953. CommitDate: Tue Jul 2 17:13:07 2013 -0700
  6954.  
  6955. Fix possible crash when writing very large filler NALUs
  6956.  
  6957. Bitstream-reallocation function didn't handle the case of filler.
  6958.  
  6959. commit 25ef3f5fdbfca0f9a5ff8a97b8475e7f8b4c9202 [revision 2338]
  6960. Author: Loren Merritt <[email protected]>
  6961. AuthorDate: Mon Jun 17 11:27:09 2013 -0700
  6962. Commit: Fiona Glaser <[email protected]>
  6963. CommitDate: Tue Jul 2 17:13:07 2013 -0700
  6964.  
  6965. Fix build with PIC on some systems
  6966.  
  6967. commit c41b629d4831cde47a8c0cde435041cc3b996d85 [revision 2337]
  6968. Author: Henrik Gramner <[email protected]>
  6969. AuthorDate: Sun Jun 2 18:41:17 2013 +0200
  6970. Commit: Fiona Glaser <[email protected]>
  6971. CommitDate: Tue Jul 2 17:13:06 2013 -0700
  6972.  
  6973. Fix potential misaligment crash in AVX2 denoise_dct
  6974.  
  6975. commit e32d9c21339cbb021d6c9ad5897bfde09dcdb63a [revision 2336]
  6976. Author: Anton Mitrofanov <[email protected]>
  6977. AuthorDate: Tue May 28 01:48:15 2013 +0400
  6978. Commit: Fiona Glaser <[email protected]>
  6979. CommitDate: Tue May 28 05:01:55 2013 -0700
  6980.  
  6981. Fix building with compilers without inline asm support
  6982.  
  6983. Also fix crash in high bit depth builds compiled with unaligned stack.
  6984.  
  6985. commit 3b8e924639ac67a4beb0ebe9b9663de03cdce84d [revision 2335]
  6986. Author: Anton Mitrofanov <[email protected]>
  6987. AuthorDate: Wed May 22 22:43:59 2013 +0400
  6988. Commit: Fiona Glaser <[email protected]>
  6989. CommitDate: Wed May 22 12:41:30 2013 -0700
  6990.  
  6991. Fix compilation with OpenCL on MacOS X
  6992.  
  6993. Also fix crash in the case of OpenCL error during encoding.
  6994.  
  6995. commit 3aa9a67b6d62bcf11ee69397647230700a32044b [revision 2334]
  6996. Author: Anton Mitrofanov <[email protected]>
  6997. AuthorDate: Mon May 6 22:51:11 2013 +0400
  6998. Commit: Fiona Glaser <[email protected]>
  6999. CommitDate: Mon May 20 12:25:24 2013 -0700
  7000.  
  7001. OpenCL support improvement/refactoring
  7002.  
  7003. Autoload the OpenCL library so that it's not required to run an openCL-enabled
  7004. build of x264.
  7005.  
  7006. Update X264_BUILD, which should have been changed with the first patch.
  7007.  
  7008. commit 0b2c3d35c168011e73300da5fdc690e00a8238e0 [revision 2333]
  7009. Author: Fiona Glaser <[email protected]>
  7010. AuthorDate: Thu May 16 13:51:37 2013 -0700
  7011. Commit: Fiona Glaser <[email protected]>
  7012. CommitDate: Mon May 20 12:25:24 2013 -0700
  7013.  
  7014. x86: shave a few instructions off AVX deblock
  7015.  
  7016. commit e7cb328580c3e1bd7604a64f40abf3e03c474771 [revision 2332]
  7017. Author: Henrik Gramner <[email protected]>
  7018. AuthorDate: Tue May 14 18:57:40 2013 +0200
  7019. Commit: Fiona Glaser <[email protected]>
  7020. CommitDate: Mon May 20 12:25:23 2013 -0700
  7021.  
  7022. x86: AVX2 dequant_4x4_dc
  7023.  
  7024. commit edf31ed3577f35e7ed3934dd74be474f9d22384a [revision 2331]
  7025. Author: Henrik Gramner <[email protected]>
  7026. AuthorDate: Tue May 14 18:53:12 2013 +0200
  7027. Commit: Fiona Glaser <[email protected]>
  7028. CommitDate: Mon May 20 12:25:23 2013 -0700
  7029.  
  7030. x86: AVX2 high bit-depth dequant
  7031.  
  7032. commit bc88d1bb331ee061c38bea80f7a54a76797c31d0 [revision 2330]
  7033. Author: Fiona Glaser <[email protected]>
  7034. AuthorDate: Thu May 9 17:20:05 2013 -0700
  7035. Commit: Fiona Glaser <[email protected]>
  7036. CommitDate: Mon May 20 12:25:22 2013 -0700
  7037.  
  7038. x86-64: 64-bit variant of AVX2 hpel_filter
  7039.  
  7040. ~5% faster than 32-bit.
  7041.  
  7042. commit 89f067b7cacecf413569e84c6c973c23f67b1ad3 [revision 2329]
  7043. Author: Henrik Gramner <[email protected]>
  7044. AuthorDate: Mon May 6 18:41:24 2013 +0200
  7045. Commit: Fiona Glaser <[email protected]>
  7046. CommitDate: Mon May 20 12:25:22 2013 -0700
  7047.  
  7048. x86: AVX2 high bit-depth denoise_dct
  7049.  
  7050. 28->15 cycles
  7051.  
  7052. Also reorder instructions to use fewer registers, 3 cycles faster on Ivy Bridge with 64-bit Windows.
  7053.  
  7054. commit 481e4cdb52989e4b514a2f4345870a19c5c0ae92 [revision 2328]
  7055. Author: Henrik Gramner <[email protected]>
  7056. AuthorDate: Sat May 4 18:48:58 2013 +0200
  7057. Commit: Fiona Glaser <[email protected]>
  7058. CommitDate: Mon May 20 12:25:22 2013 -0700
  7059.  
  7060. x86: AVX2 high bit-depth quant
  7061.  
  7062. quant_4x4: 13->6 cycles
  7063. quant_4x4_dc: 14->8 cycles
  7064. quant_8x8: 47->24 cycles
  7065. quant_4x4x4: 48->25 cycles
  7066.  
  7067. commit 02aa1368da5c222c8833724abccddd8f02630fc6 [revision 2327]
  7068. Author: Fiona Glaser <[email protected]>
  7069. AuthorDate: Wed May 1 14:32:11 2013 -0700
  7070. Commit: Fiona Glaser <[email protected]>
  7071. CommitDate: Mon May 20 12:25:21 2013 -0700
  7072.  
  7073. x86: AVX2 add16x16_idct_dc
  7074.  
  7075. 27 -> 19 cycles
  7076.  
  7077. commit 0c00c2c7882de130184e02cf1861599aedb425e8 [revision 2326]
  7078. Author: Fiona Glaser <[email protected]>
  7079. AuthorDate: Mon Apr 29 16:16:54 2013 -0700
  7080. Commit: Fiona Glaser <[email protected]>
  7081. CommitDate: Mon May 20 12:25:21 2013 -0700
  7082.  
  7083. x86: faster AVX2 quant_4x4x4
  7084.  
  7085. 10->9 cycles
  7086.  
  7087. commit af6647e0e7d647c660003f65b78b4f1a0b186ec2 [revision 2325]
  7088. Author: Fiona Glaser <[email protected]>
  7089. AuthorDate: Sat Apr 27 21:03:32 2013 -0700
  7090. Commit: Fiona Glaser <[email protected]>
  7091. CommitDate: Mon May 20 12:25:20 2013 -0700
  7092.  
  7093. x86: AVX2 intra_sad_x3_8x8c
  7094.  
  7095. 30->22 cycles
  7096.  
  7097. commit f114746df6ce6a1bcacf46c62b696cc309ab4527 [revision 2324]
  7098. Author: Henrik Gramner <[email protected]>
  7099. AuthorDate: Sun Apr 28 11:11:03 2013 +0200
  7100. Commit: Fiona Glaser <[email protected]>
  7101. CommitDate: Mon May 20 12:25:20 2013 -0700
  7102.  
  7103. x86: AVX2 high bit-depth intra_sad_x3_8x8
  7104.  
  7105. 43->24 cycles
  7106.  
  7107. commit 8e4f045f815a59ca3d6398ff4ddae7af44766dc8 [revision 2323]
  7108. Author: Fiona Glaser <[email protected]>
  7109. AuthorDate: Wed Apr 24 14:22:15 2013 -0700
  7110. Commit: Fiona Glaser <[email protected]>
  7111. CommitDate: Mon May 20 12:25:20 2013 -0700
  7112.  
  7113. x86: AVX2 deblock strength
  7114.  
  7115. 30->18 cycles
  7116.  
  7117. commit 594dd84cb85e616f4e260f7fdef6ce5a34360ac7 [revision 2322]
  7118. Author: Henrik Gramner <[email protected]>
  7119. AuthorDate: Wed May 1 17:42:48 2013 +0200
  7120. Commit: Fiona Glaser <[email protected]>
  7121. CommitDate: Mon May 20 12:25:19 2013 -0700
  7122.  
  7123. x86: Faster high bit-depth intra_sad_x3_4x4
  7124.  
  7125. 20->16 cycles on Ivy Bridge
  7126.  
  7127. commit a8384178bd917576469da040923976cb531be38c [revision 2321]
  7128. Author: Fiona Glaser <[email protected]>
  7129. AuthorDate: Tue Apr 30 17:36:46 2013 -0700
  7130. Commit: Fiona Glaser <[email protected]>
  7131. CommitDate: Mon May 20 12:25:19 2013 -0700
  7132.  
  7133. x86: faster SSSE3 hpel
  7134.  
  7135. ~7% faster using the pmulhrsw trick from mc_chroma.
  7136.  
  7137. commit 1f5a32c2459ed6f42d9c150d008e3471d61af3ee [revision 2320]
  7138. Author: Fiona Glaser <[email protected]>
  7139. AuthorDate: Mon Apr 29 14:22:23 2013 -0700
  7140. Commit: Fiona Glaser <[email protected]>
  7141. CommitDate: Mon May 20 12:25:18 2013 -0700
  7142.  
  7143. x86-64: faster SSSE3 trellis
  7144.  
  7145. ~2% faster trellis.
  7146.  
  7147. commit 7cbb27f0ce5ea3e756c628ac606f65d7de57f285 [revision 2319]
  7148. Author: Fiona Glaser <[email protected]>
  7149. AuthorDate: Thu May 2 17:10:26 2013 -0700
  7150. Commit: Fiona Glaser <[email protected]>
  7151. CommitDate: Mon May 20 12:25:18 2013 -0700
  7152.  
  7153. x86: 32-byte align the stack if possible
  7154.  
  7155. Avoids the need for manual 32 byte array alignment on compilers that support
  7156. -mpreferred-stack-boundary.
  7157.  
  7158. commit 30c91f62906ce08b5d227002b38ebd64f1291fae [revision 2318]
  7159. Author: Henrik Gramner <[email protected]>
  7160. AuthorDate: Sat May 11 23:39:09 2013 +0200
  7161. Commit: Fiona Glaser <[email protected]>
  7162. CommitDate: Mon May 20 12:25:12 2013 -0700
  7163.  
  7164. x86inc: Utilize the shadow space on 64-bit Windows
  7165.  
  7166. Store XMM6 and XMM7 in the shadow space in functions that clobbers them.
  7167. This way we don't have to adjust the stack pointer as often,
  7168. reducing the number of instructions as well as code size.
  7169.  
  7170. commit 33c352673900bd1b362bb2fe0284e999fccd633d [revision 2317]
  7171. Author: Henrik Gramner <[email protected]>
  7172. AuthorDate: Fri May 3 23:06:10 2013 +0200
  7173. Commit: Fiona Glaser <[email protected]>
  7174. CommitDate: Fri May 17 09:26:49 2013 -0700
  7175.  
  7176. x86: Don't use explicitly aligned versions of SAD on AVX CPUs
  7177.  
  7178. On modern CPUs movdqu isn't slower than movdqa when used on aligned data and using the same code in both cases saves cache.
  7179.  
  7180. This was already done for the high bit-depth AVX2 implementation but the aligned version still exists as dead code so remove that.
  7181.  
  7182. commit 16d037211f1dd032288e25ab74d93a569fd93d6c [revision 2316]
  7183. Author: Henrik Gramner <[email protected]>
  7184. AuthorDate: Fri May 3 20:18:03 2013 +0200
  7185. Commit: Fiona Glaser <[email protected]>
  7186. CommitDate: Fri May 17 09:26:49 2013 -0700
  7187.  
  7188. x86: Add missing initializations for high bit-depth sad_aligned
  7189.  
  7190. commit 25e219ad2565e52a6962eb1e16cf19f3482e655b [revision 2315]
  7191. Author: Fiona Glaser <[email protected]>
  7192. AuthorDate: Mon May 13 16:52:18 2013 -0700
  7193. Commit: Fiona Glaser <[email protected]>
  7194. CommitDate: Fri May 17 09:26:49 2013 -0700
  7195.  
  7196. x86: add Jaguar CPU detection
  7197.  
  7198. commit c1e37099627b1dc2f15b295aa4c2eedd431a6dba [revision 2314]
  7199. Author: Henrik Gramner <[email protected]>
  7200. AuthorDate: Tue May 7 17:21:03 2013 +0200
  7201. Commit: Fiona Glaser <[email protected]>
  7202. CommitDate: Fri May 17 09:26:48 2013 -0700
  7203.  
  7204. x86inc: Remove .rodata kludges
  7205.  
  7206. The Mach-O bug was fixed in yasm 0.8.0 and we don't support versions that old.
  7207.  
  7208. a.out was superseded by ELF on sane systems a few decades ago.
  7209.  
  7210. commit 5444e95a5c9ee866625b1122a19dbae6bf044008 [revision 2313]
  7211. Author: Henrik Gramner <[email protected]>
  7212. AuthorDate: Sat May 4 16:21:32 2013 +0200
  7213. Commit: Fiona Glaser <[email protected]>
  7214. CommitDate: Fri May 17 09:26:48 2013 -0700
  7215.  
  7216. checkasm: Use 64-bit cycle counters
  7217.  
  7218. Prevents overflows that can occur in some cases.
  7219.  
  7220. commit 0e000e7a763c9bb5c14257bad365144025013fc9 [revision 2312]
  7221. Author: Henrik Gramner <[email protected]>
  7222. AuthorDate: Fri May 10 13:55:32 2013 +0200
  7223. Commit: Fiona Glaser <[email protected]>
  7224. CommitDate: Fri May 17 09:26:45 2013 -0700
  7225.  
  7226. checkasm: Fix stack alignment bug
  7227.  
  7228. commit 3ba0fb847b1a14f9db5f3dabe209eee2d4edc91d [revision 2311]
  7229. Author: Fiona Glaser <[email protected]>
  7230. AuthorDate: Wed May 8 10:48:41 2013 -0700
  7231. Commit: Fiona Glaser <[email protected]>
  7232. CommitDate: Wed May 15 13:04:45 2013 -0700
  7233.  
  7234. Fix invalid memcpy in sliced-threads
  7235.  
  7236. Likely didn't actually break in practice, but memcpy with src==dst
  7237. is incorrect.
  7238.  
  7239. commit 7f3606572957b63f1169bc793ed55bccdb549d56 [revision 2310]
  7240. Author: Fiona Glaser <[email protected]>
  7241. AuthorDate: Mon Apr 29 12:14:01 2013 -0700
  7242. Commit: Fiona Glaser <[email protected]>
  7243. CommitDate: Mon Apr 29 12:29:38 2013 -0700
  7244.  
  7245. Fix two bugs in slice-min-mbs and slices-max
  7246.  
  7247. Slices-max broke slice-max-size when slice-max wasn't used.
  7248. Slice-min-mbs broke in rare cases near the end of a threadslice.
  7249.  
  7250. commit 67d6f602018d0fc1cb05cd6240e4fe1c2646169f [revision 2309]
  7251. Author: Fiona Glaser <[email protected]>
  7252. AuthorDate: Thu Apr 4 18:00:23 2013 -0700
  7253. Commit: Fiona Glaser <[email protected]>
  7254. CommitDate: Tue Apr 23 14:36:39 2013 -0700
  7255.  
  7256. x86: SSSE3 LUT-based faster coeff_level_run
  7257.  
  7258. ~2x faster coeff_level_run.
  7259. Faster CAVLC encoding: {1%,2%,7%} overall with {superfast,medium,slower}.
  7260. Uses the same pshufb LUT abuse trick as in the previous ads_mvs patch.
  7261.  
  7262. commit c17d12f83381913650d84004815c20a1f7092144 [revision 2308]
  7263. Author: Fiona Glaser <[email protected]>
  7264. AuthorDate: Mon Mar 25 14:03:37 2013 -0700
  7265. Commit: Fiona Glaser <[email protected]>
  7266. CommitDate: Tue Apr 23 14:36:39 2013 -0700
  7267.  
  7268. x86-64: BMI2 cabac_residual functions
  7269.  
  7270. commit 40316f836d42cb5aee8de5ae6b4a5e417d8446f8 [revision 2307]
  7271. Author: Fiona Glaser <[email protected]>
  7272. AuthorDate: Wed Mar 20 15:08:35 2013 -0700
  7273. Commit: Fiona Glaser <[email protected]>
  7274. CommitDate: Tue Apr 23 14:36:38 2013 -0700
  7275.  
  7276. x86: SSSE3 ads_mvs
  7277.  
  7278. ~55% faster ads in benchasm, ~15-30% in real encoding.
  7279. ~4% faster "placebo" preset overall.
  7280.  
  7281. commit 03396f82bd1a709aa83d15de0affd0c4c5bd621d [revision 2306]
  7282. Author: Henrik Gramner <[email protected]>
  7283. AuthorDate: Tue Apr 16 23:27:53 2013 +0200
  7284. Commit: Fiona Glaser <[email protected]>
  7285. CommitDate: Tue Apr 23 14:36:37 2013 -0700
  7286.  
  7287. x86: AVX2 pixel_ssd_nv12_core
  7288.  
  7289. commit dc05aebbc51b64b6cf3cfa95a1fbb20f6ffe94c6 [revision 2305]
  7290. Author: Henrik Gramner <[email protected]>
  7291. AuthorDate: Tue Apr 16 23:27:50 2013 +0200
  7292. Commit: Fiona Glaser <[email protected]>
  7293. CommitDate: Tue Apr 23 14:36:37 2013 -0700
  7294.  
  7295. x86: AVX2 high bit-depth pixel_ssd
  7296.  
  7297. commit f49c2eba352a9087301dfc3c3de902ab083bd9e9 [revision 2304]
  7298. Author: Henrik Gramner <[email protected]>
  7299. AuthorDate: Tue Apr 16 23:27:46 2013 +0200
  7300. Commit: Fiona Glaser <[email protected]>
  7301. CommitDate: Tue Apr 23 14:36:36 2013 -0700
  7302.  
  7303. x86: AVX2 high bit-depth pixel_sad_x3/pixel_sad_x4
  7304.  
  7305. Also reduce the number of xmm registers used by sse2/ssse3 pixel_sad_x3.
  7306.  
  7307. commit 0e69048d4f9664f1293c5eed0604522c67adaff5 [revision 2303]
  7308. Author: Henrik Gramner <[email protected]>
  7309. AuthorDate: Tue Apr 16 23:27:43 2013 +0200
  7310. Commit: Fiona Glaser <[email protected]>
  7311. CommitDate: Tue Apr 23 14:36:35 2013 -0700
  7312.  
  7313. x86: AVX2 high bit-depth vsad
  7314.  
  7315. commit 9f885c112d6566388d472da68ada0301ce330311 [revision 2302]
  7316. Author: Henrik Gramner <[email protected]>
  7317. AuthorDate: Tue Apr 16 23:27:39 2013 +0200
  7318. Commit: Fiona Glaser <[email protected]>
  7319. CommitDate: Tue Apr 23 14:36:35 2013 -0700
  7320.  
  7321. x86: AVX2 high bit-depth pixel_sad
  7322.  
  7323. Also use loops instead of duplicating code; reduces code size by ~10kB with
  7324. negligible effect on performance.
  7325.  
  7326. commit 295f83af2afa93073d7810ab96b1d8d889a53ed2 [revision 2301]
  7327. Author: Henrik Gramner <[email protected]>
  7328. AuthorDate: Tue Apr 16 23:27:35 2013 +0200
  7329. Commit: Fiona Glaser <[email protected]>
  7330. CommitDate: Tue Apr 23 14:36:34 2013 -0700
  7331.  
  7332. x86: AVX2 high_bit_depth pixel_avg2, get_ref, mc_copy_w16, mc_luma
  7333.  
  7334. Also reduce the number of xmm registers used by mc_copy_* to avoid
  7335. saving and restoring xmm6 and xmm7 on 64-bit Windows.
  7336.  
  7337. commit e7a46b6536ab3ea4806f585b771b6cbb261031d1 [revision 2300]
  7338. Author: Henrik Gramner <[email protected]>
  7339. AuthorDate: Tue Apr 16 23:27:32 2013 +0200
  7340. Commit: Fiona Glaser <[email protected]>
  7341. CommitDate: Tue Apr 23 14:36:33 2013 -0700
  7342.  
  7343. x86: AVX2 nal_escape
  7344.  
  7345. Also rewrite the entire function to be faster and drop the AVX version which is no longer useful.
  7346.  
  7347. commit 547a6573af56afe8d551201245775c6ba179e781 [revision 2299]
  7348. Author: Henrik Gramner <[email protected]>
  7349. AuthorDate: Tue Apr 16 23:27:29 2013 +0200
  7350. Commit: Fiona Glaser <[email protected]>
  7351. CommitDate: Tue Apr 23 14:36:32 2013 -0700
  7352.  
  7353. x86: AVX memzero_aligned
  7354.  
  7355. commit 0f776f63daf47eac9b69ef77aaf7c9c16213cba9 [revision 2298]
  7356. Author: Henrik Gramner <[email protected]>
  7357. AuthorDate: Tue Apr 16 23:27:25 2013 +0200
  7358. Commit: Fiona Glaser <[email protected]>
  7359. CommitDate: Tue Apr 23 14:36:31 2013 -0700
  7360.  
  7361. x86: AVX2 predict_16x16_dc
  7362.  
  7363. commit 97ad171ae33c51f48e6214abdf7c978e4dd5d2d1 [revision 2297]
  7364. Author: Henrik Gramner <[email protected]>
  7365. AuthorDate: Tue Apr 16 23:27:22 2013 +0200
  7366. Commit: Fiona Glaser <[email protected]>
  7367. CommitDate: Tue Apr 23 14:36:31 2013 -0700
  7368.  
  7369. x86: AVX2 predict_8x8c_p/predict_8x16c_p
  7370.  
  7371. commit 8ecdeb2709b4b7095237330e68e9a76ea8060a2f [revision 2296]
  7372. Author: Henrik Gramner <[email protected]>
  7373. AuthorDate: Tue Apr 16 23:27:18 2013 +0200
  7374. Commit: Fiona Glaser <[email protected]>
  7375. CommitDate: Tue Apr 23 14:36:30 2013 -0700
  7376.  
  7377. x86: AVX2 predict_16x16_p
  7378.  
  7379. Also fix the AVX implementation to correctly use the SSSE3 inline asm
  7380. instead of SSE2.
  7381.  
  7382. commit f3d521da8163bb9a381284ef0b5c949b8a5c9f9c [revision 2295]
  7383. Author: Henrik Gramner <[email protected]>
  7384. AuthorDate: Tue Apr 16 23:27:14 2013 +0200
  7385. Commit: Fiona Glaser <[email protected]>
  7386. CommitDate: Tue Apr 23 14:36:29 2013 -0700
  7387.  
  7388. x86: AVX high bit-depth predict_16x16_v
  7389.  
  7390. Also restructure some code to reduce code size of various functions,
  7391. especially in high bit-depth.
  7392.  
  7393. commit fa40b44f339501917e7a7c003ab826bf3e7b6a10 [revision 2294]
  7394. Author: Henrik Gramner <[email protected]>
  7395. AuthorDate: Tue Apr 16 23:27:08 2013 +0200
  7396. Commit: Fiona Glaser <[email protected]>
  7397. CommitDate: Tue Apr 23 14:36:29 2013 -0700
  7398.  
  7399. x86: AVX2 high bit-depth predict_4x4_h
  7400.  
  7401. commit 7908dc632330b6028ab7dae42834e2098e628b24 [revision 2293]
  7402. Author: Henrik Gramner <[email protected]>
  7403. AuthorDate: Tue Apr 16 23:27:04 2013 +0200
  7404. Commit: Fiona Glaser <[email protected]>
  7405. CommitDate: Tue Apr 23 14:36:28 2013 -0700
  7406.  
  7407. x86: AVX2 high bit-depth predict_16x16_h
  7408.  
  7409. commit 51708c3e193438439aaeaf31c377b070ca403e0e [revision 2292]
  7410. Author: Henrik Gramner <[email protected]>
  7411. AuthorDate: Tue Apr 16 23:27:00 2013 +0200
  7412. Commit: Fiona Glaser <[email protected]>
  7413. CommitDate: Tue Apr 23 14:36:27 2013 -0700
  7414.  
  7415. x86: AVX2 high bit-depth predict_8x8c_h/predict_8x16c_h
  7416.  
  7417. commit 184c50554ae95aa60edd3fa309ca8013e00a8648 [revision 2291]
  7418. Author: Henrik Gramner <[email protected]>
  7419. AuthorDate: Tue Apr 16 23:26:47 2013 +0200
  7420. Commit: Fiona Glaser <[email protected]>
  7421. CommitDate: Tue Apr 23 14:36:26 2013 -0700
  7422.  
  7423. x86util: Support ymm registers in HADD macros
  7424.  
  7425. commit 0ea5be852e97d8cfdf04e384a8a78210f87c2dc0 [revision 2290]
  7426. Author: Fiona Glaser <[email protected]>
  7427. AuthorDate: Tue Feb 26 16:26:34 2013 -0800
  7428. Commit: Fiona Glaser <[email protected]>
  7429. CommitDate: Tue Apr 23 14:36:25 2013 -0700
  7430.  
  7431. x86: more AVX2 framework, AVX2 functions, plus some existing asm tweaks
  7432.  
  7433. AVX2 functions:
  7434. mc_chroma
  7435. intra_sad_x3_16x16
  7436. last64
  7437. ads
  7438. hpel
  7439. dct4
  7440. idct4
  7441. sub16x16_dct8
  7442. quant_4x4x4
  7443. quant_4x4
  7444. quant_4x4_dc
  7445. quant_8x8
  7446. SAD_X3/X4
  7447. SATD
  7448. var
  7449. var2
  7450. SSD
  7451. zigzag interleave
  7452. weightp
  7453. weightb
  7454. intra_sad_8x8_x9
  7455. decimate
  7456. integral
  7457. hadamard_ac
  7458. sa8d_satd
  7459. sa8d
  7460. lowres_init
  7461. denoise
  7462.  
  7463. commit 19e1a2bbf2d1aaa15ea2d2c118b0236ff64b4bd1 [revision 2289]
  7464. Author: Loren Merritt <[email protected]>
  7465. AuthorDate: Mon Feb 25 21:16:45 2013 +0000
  7466. Commit: Fiona Glaser <[email protected]>
  7467. CommitDate: Tue Apr 23 14:36:24 2013 -0700
  7468.  
  7469. x86inc: create xm# and ym#, analagous to m#
  7470.  
  7471. For when we want to mix simd sizes within one function.
  7472.  
  7473. commit 3a8dfb2bc62be21215b6f7d47c53c5a912878656 [revision 2288]
  7474. Author: Fiona Glaser <[email protected]>
  7475. AuthorDate: Fri Apr 5 16:08:35 2013 -0700
  7476. Commit: Fiona Glaser <[email protected]>
  7477. CommitDate: Tue Apr 23 14:36:24 2013 -0700
  7478.  
  7479. x86inc: fix AVX emulation of cmp(p|s)(s|d)
  7480.  
  7481. commit a3f5c7326c0aa707ccbd5a938a0b65581888b549 [revision 2287]
  7482. Author: Fiona Glaser <[email protected]>
  7483. AuthorDate: Tue Feb 5 17:15:00 2013 -0800
  7484. Commit: Fiona Glaser <[email protected]>
  7485. CommitDate: Tue Apr 23 14:36:23 2013 -0700
  7486.  
  7487. x86-64: cabac_block_residual assembly
  7488.  
  7489. RDO: ~20% faster than C
  7490. Bitstream: ~50% faster than C
  7491. 1-2% faster overall, highest on preset superfast/fast/medium.
  7492.  
  7493. commit f49a1b2ef6d95d8f0f186df0fc3bfe38414e264f [revision 2286]
  7494. Author: Steve Borho <[email protected]>
  7495. AuthorDate: Thu Feb 21 12:48:40 2013 -0600
  7496. Commit: Fiona Glaser <[email protected]>
  7497. CommitDate: Tue Apr 23 14:36:20 2013 -0700
  7498.  
  7499. OpenCL lookahead
  7500.  
  7501. OpenCL support is compiled in by default, but must be enabled at runtime by an
  7502. --opencl command line flag. Compiling OpenCL support requires perl. To avoid
  7503. the perl requirement use: configure --disable-opencl.
  7504.  
  7505. When enabled, the lookahead thread is mostly off-loaded to an OpenCL capable GPU
  7506. device. Lowres intra cost prediction, lowres motion search (including subpel)
  7507. and bidir cost predictions are all done on the GPU. MB-tree and final slice
  7508. decisions are still done by the CPU. Presets which do not use a threaded
  7509. lookahead will not use OpenCL at all (superfast, ultrafast).
  7510.  
  7511. Because of data dependencies, the GPU must use an iterative motion search which
  7512. performs more total work than the CPU would do, so this is not work efficient
  7513. or power efficient. But if there are spare GPU cycles to spare, it can often
  7514. speed up the encode. Output quality when OpenCL lookahead is enabled is often
  7515. very slightly worse in quality than the CPU quality (because of the same data
  7516. dependencies).
  7517.  
  7518. x264 must compile its OpenCL kernels for your device before running them, and in
  7519. order to avoid doing this every run it caches the compiled kernel binary in a
  7520. file named x264_lookahead.clbin (--opencl-clbin FNAME to override). The cache
  7521. file will be ignored if the device, driver, or OpenCL source are changed.
  7522.  
  7523. x264 will use the first GPU device which supports the required cl_image
  7524. features required by its kernels. Most modern discrete GPUs and all AMD
  7525. integrated GPUs will work. Intel integrated GPUs (up to IvyBridge) do not
  7526. support those necessary features. Use --opencl-device N to specify a number of
  7527. capable GPUs to skip during device detection.
  7528.  
  7529. Switchable graphics environments (e.g. AMD Enduro) are currently not supported,
  7530. as some have bugs in their OpenCL drivers that cause output to be silently
  7531. incorrect.
  7532.  
  7533. Developed by MulticoreWare with support from AMD and Telestream.
  7534.  
  7535. commit 2d0c47a50622ec59ade303cf150c21b8910a2bce [revision 2285]
  7536. Author: Fiona Glaser <[email protected]>
  7537. AuthorDate: Mon Mar 4 15:19:47 2013 -0800
  7538. Commit: Fiona Glaser <[email protected]>
  7539. CommitDate: Tue Apr 23 11:47:30 2013 -0700
  7540.  
  7541. weightp: improve scale/offset search, chroma
  7542.  
  7543. Rescale the scale factor if the offset clips. This makes weightp more effective
  7544. in fades to/from white (and an other situation that requires big offsets).
  7545.  
  7546. Search more than 1 scale factor and more than 1 offset, depending on --subme.
  7547.  
  7548. Try to find the optimal chroma denominator instead of hardcoding it.
  7549.  
  7550. Overall improvement: a few percent in fade-heavy clips, such as a sample from
  7551. Avatar: TLA.
  7552.  
  7553. commit 732e4f7e8b9ab6d214cbcf059445b4712709faa4 [revision 2284]
  7554. Author: Fiona Glaser <[email protected]>
  7555. AuthorDate: Tue Feb 19 13:48:44 2013 -0800
  7556. Commit: Fiona Glaser <[email protected]>
  7557. CommitDate: Tue Apr 23 11:47:29 2013 -0700
  7558.  
  7559. Add slices-max feature
  7560.  
  7561. The H.264 spec technically has limits on the number of slices per frame. x264
  7562. normally ignores this, since most use-cases that require large numbers of
  7563. slices prefer it to. However, certain decoders may break with extremely large
  7564. numbers of slices, as can occur with some slice-max-size/mbs settings.
  7565.  
  7566. When set, x264 will refuse to create any slices beyond the maximum number,
  7567. even if slice-max-size/mbs requires otherwise.
  7568.  
  7569. commit fdfffa3058cb590765dbb34afa5706755dcb5319 [revision 2283]
  7570. Author: Fiona Glaser <[email protected]>
  7571. AuthorDate: Thu Feb 14 17:22:02 2013 -0800
  7572. Commit: Fiona Glaser <[email protected]>
  7573. CommitDate: Tue Apr 23 11:47:29 2013 -0700
  7574.  
  7575. Add slice-min-mbs feature
  7576.  
  7577. Works in conjunction with slice-max-mbs and/or slice-max-size to avoid overly
  7578. small slices.
  7579. Useful with certain decoders that barf on extremely small slices.
  7580.  
  7581. If slice-min-mbs would be violated as a result of slice-max-size, x264 will
  7582. exceed slice-max-size and print a warning.
  7583.  
  7584. commit 8a3a41de9e5f54cb6e7b5c69486e50471a5c022d [revision 2282]
  7585. Author: Anton Mitrofanov <[email protected]>
  7586. AuthorDate: Tue Mar 26 18:56:21 2013 +0400
  7587. Commit: Fiona Glaser <[email protected]>
  7588. CommitDate: Tue Apr 23 11:47:28 2013 -0700
  7589.  
  7590. Disable mbtree asm with cpu-independent option
  7591.  
  7592. Results vary between versions because of different rounding results.
  7593.  
  7594. commit bf52bab4e5607d7f3d98b3999a13cb8149aeef1c [revision 2281]
  7595. Author: Anton Mitrofanov <[email protected]>
  7596. AuthorDate: Tue Mar 26 18:30:00 2013 +0400
  7597. Commit: Fiona Glaser <[email protected]>
  7598. CommitDate: Tue Apr 23 11:47:27 2013 -0700
  7599.  
  7600. Show "avs: no" --disable-avs option instead of empty string
  7601.  
  7602. commit e74287e93b0ee7afb384624f60dc440b736fec6b [revision 2280]
  7603. Author: Tim Walker <[email protected]>
  7604. AuthorDate: Tue Mar 19 23:42:43 2013 +0100
  7605. Commit: Fiona Glaser <[email protected]>
  7606. CommitDate: Tue Apr 23 11:47:27 2013 -0700
  7607.  
  7608. lavf input: don't use deprecated AVStream fields
  7609.  
  7610. Fixes building against newer libavcodecs from the Libav project.
  7611.  
  7612. commit aa73459b710f4c08b654d69573c22fd2fc2a99f8 [revision 2279]
  7613. Author: Anton Mitrofanov <[email protected]>
  7614. AuthorDate: Tue Mar 26 19:54:36 2013 +0400
  7615. Commit: Fiona Glaser <[email protected]>
  7616. CommitDate: Tue Apr 23 11:47:26 2013 -0700
  7617.  
  7618. Fix y4m input with C420paldv colorspace
  7619.  
  7620. commit 42c500af62fbe09e7a55ecd47fc72331fbe4ae02 [revision 2278]
  7621. Author: Fiona Glaser <[email protected]>
  7622. AuthorDate: Sat Mar 2 01:22:29 2013 -0800
  7623. Commit: Fiona Glaser <[email protected]>
  7624. CommitDate: Tue Apr 23 11:47:25 2013 -0700
  7625.  
  7626. x86: correctly check stack alignment for Atom hadamard_ac
  7627.  
  7628. Regression in r2265 (only affected compilers with broken stack alignment,
  7629. like ICL on win32).
  7630.  
  7631. commit bed18d0e4545e7528bf585a1a3c7fbc05ddbafa4 [revision 2277]
  7632. Author: Loren Merritt <[email protected]>
  7633. AuthorDate: Mon Feb 25 21:23:55 2013 +0000
  7634. Commit: Fiona Glaser <[email protected]>
  7635. CommitDate: Tue Apr 23 11:47:24 2013 -0700
  7636.  
  7637. x86inc: fix some corner cases of SWAP
  7638.  
  7639. SWAP with >=3 named (rather than numbered) args
  7640. PERMUTE followed by SWAP with 2 named args
  7641. used to produce the wrong permutation
  7642.  
  7643. commit 3cdaca1ac2f6022b1affcd24eff397a03b59fce3 [revision 2276]
  7644. Author: Fiona Glaser <[email protected]>
  7645. AuthorDate: Wed Feb 27 13:30:22 2013 -0800
  7646. Commit: Fiona Glaser <[email protected]>
  7647. CommitDate: Tue Apr 23 11:47:22 2013 -0700
  7648.  
  7649. Fix array overreads that caused miscompilation in gcc 4.8
  7650.  
  7651. commit 37033444036210ddab75c3ec5b9b5c2a5abb9d52 [revision 2275]
  7652. Author: Fiona Glaser <[email protected]>
  7653. AuthorDate: Thu Feb 28 13:32:37 2013 -0800
  7654. Commit: Fiona Glaser <[email protected]>
  7655. CommitDate: Sat Apr 13 10:36:24 2013 -0700
  7656.  
  7657. Fix undefined behavior in x264_ratecontrol_mb
  7658.  
  7659. commit cb4547aefb624105b622368aad62c947f89cc4b1 [revision 2274]
  7660. Author: Stefan Groenroos <[email protected]>
  7661. AuthorDate: Fri Mar 1 22:35:34 2013 +0200
  7662. Commit: Fiona Glaser <[email protected]>
  7663. CommitDate: Fri Mar 1 13:21:59 2013 -0800
  7664.  
  7665. ARM: Fix bug in x264_quant_4x4x4_neon
  7666.  
  7667. Regression in r2273.
  7668.  
  7669. commit 3a8baa0ec68c50db3194ed778d0e744d6311cda3 [revision 2273]
  7670. Author: Stefan Groenroos <[email protected]>
  7671. AuthorDate: Mon Feb 25 23:43:09 2013 +0200
  7672. Commit: Fiona Glaser <[email protected]>
  7673. CommitDate: Tue Feb 26 15:13:17 2013 -0800
  7674.  
  7675. ARM: update NEON mc_chroma to work with NV12 and re-enable it
  7676.  
  7677. Up to 10-15% faster overall.
  7678.  
  7679. commit 215f2beeadb2ade3a318b397f25b8a6ad3a761d1 [revision 2272]
  7680. Author: Fiona Glaser <[email protected]>
  7681. AuthorDate: Thu Feb 14 15:00:48 2013 -0800
  7682. Commit: Fiona Glaser <[email protected]>
  7683. CommitDate: Tue Feb 26 15:12:35 2013 -0800
  7684.  
  7685. CABAC/CAVLC: use the new bit-iterating macro here too
  7686.  
  7687. commit 993c81e94eebaacddbbfcec665831d07d89490b7 [revision 2271]
  7688. Author: Fiona Glaser <[email protected]>
  7689. AuthorDate: Fri Feb 8 15:34:38 2013 -0800
  7690. Commit: Fiona Glaser <[email protected]>
  7691. CommitDate: Mon Feb 25 23:22:56 2013 -0800
  7692.  
  7693. quant_4x4x4: quant one 8x8 block at a time
  7694.  
  7695. This reduces overhead and lets us use less branchy code for zigzag, dequant,
  7696. decimate, and so on.
  7697. Reorganize and optimize a lot of macroblock_encode using this new function.
  7698. ~1-2% faster overall.
  7699.  
  7700. Includes NEON and x86 versions of the new function.
  7701. Using larger merged functions like this will also make wider SIMD, like
  7702. AVX2, more effective.
  7703.  
  7704. commit 5ee1d03a8b86915d98b165d067dce377df3a87ba [revision 2270]
  7705. Author: Stephen Hutchinson <[email protected]>
  7706. AuthorDate: Tue Feb 12 21:55:43 2013 -0500
  7707. Commit: Fiona Glaser <[email protected]>
  7708. CommitDate: Mon Feb 25 23:22:56 2013 -0800
  7709.  
  7710. Add AvxSynth support to the AviSynth input module.
  7711.  
  7712. Uses dlopen to load AvxSynth on Linux and OS X.
  7713.  
  7714. Allows the use of --demuxer avs for AvxSynth, though the only source filter it
  7715. can currently use is FFMS2.
  7716.  
  7717. Add a local copy of avxsynth_c.h and its dependent headers in extras/ so that
  7718. users don't need to actually have AvxSynth development headers installed to
  7719. enable support for it (mirroring the AviSynth behavior).
  7720.  
  7721. Based on a patch by 0x09 ([email protected])
  7722.  
  7723. commit 7b1301e946218cfe6e072fea03702754ee0cc8a6 [revision 2269]
  7724. Author: Fiona Glaser <[email protected]>
  7725. AuthorDate: Fri Feb 8 00:13:15 2013 -0800
  7726. Commit: Fiona Glaser <[email protected]>
  7727. CommitDate: Mon Feb 25 23:22:55 2013 -0800
  7728.  
  7729. Eliminate some branchiness in ME/analysis
  7730.  
  7731. Faster, fewer branch mispredictions.
  7732.  
  7733. commit 7de9a9aa4bc06843dd7d8afe6bc42c02e27b6b73 [revision 2268]
  7734. Author: Fiona Glaser <[email protected]>
  7735. AuthorDate: Wed Feb 6 16:55:39 2013 -0800
  7736. Commit: Fiona Glaser <[email protected]>
  7737. CommitDate: Mon Feb 25 23:22:55 2013 -0800
  7738.  
  7739. Fix some store forwarding stalls
  7740. There's quite a few others, but most of them don't help to fix or there's no
  7741. easy way to avoid them.
  7742.  
  7743. commit 68a6268bae989c55a02b7e86b169bd1a02793a95 [revision 2267]
  7744. Author: Fiona Glaser <[email protected]>
  7745. AuthorDate: Tue Feb 5 01:23:23 2013 -0800
  7746. Commit: Fiona Glaser <[email protected]>
  7747. CommitDate: Mon Feb 25 23:22:55 2013 -0800
  7748.  
  7749. x86: faster AVX satd/sa8d/sa8d_satd/hadamard_ac
  7750.  
  7751. Use Conroe-style movddup in AVX transforms; both Sandy Bridge and Bulldozer
  7752. do movddup in the load unit, so it's totally free this way.
  7753.  
  7754. On Sandy Bridge:
  7755. ~6% faster sa8d_satd
  7756. ~5% faster hadamard_ac
  7757. ~9% faster 32-bit satd
  7758. ~2% faster sa8d
  7759.  
  7760. commit 5d60b9c9ad794a666d0cfe0dd9d66d5b9f58e033 [revision 2266]
  7761. Author: Fiona Glaser <[email protected]>
  7762. AuthorDate: Sat Feb 2 12:37:08 2013 -0800
  7763. Commit: Fiona Glaser <[email protected]>
  7764. CommitDate: Mon Feb 25 23:22:53 2013 -0800
  7765.  
  7766. x86: detect Bobcat, improve Atom optimizations, reorganize flags
  7767.  
  7768. The Bobcat has a 64-bit SIMD unit reminiscent of the Athlon 64; detect this
  7769. and apply the appropriate flags.
  7770.  
  7771. It also has an extremely slow palignr instruction; create a flag for this to
  7772. avoid massive penalties on palignr-heavy functions.
  7773.  
  7774. Improve Atom function selection and document exactly what the SLOW_ATOM flag
  7775. covers.
  7776.  
  7777. Add Atom-optimized SATD/SA8D/hadamard_ac functions: simply combine the ssse3
  7778. optimizations with the sse2 algorithm to avoid pmaddubsw, which is slow on
  7779. Atom along with other SIMD multiplies.
  7780.  
  7781. Drop TBM detection; it'll probably never be useful for x264.
  7782.  
  7783. Invert FastShuffle to SlowShuffle; it only ever applied to one CPU (Conroe).
  7784.  
  7785. Detect CMOV, to fail more gracefully when run on a chip with MMX2 but no CMOV.
  7786.  
  7787. commit 75d927053ef5546eb011ff5a5ac19152dd4e3c63 [revision 2265]
  7788. Author: Oskar Arvidsson <[email protected]>
  7789. AuthorDate: Sat Jan 19 01:47:09 2013 +0100
  7790. Commit: Fiona Glaser <[email protected]>
  7791. CommitDate: Mon Feb 25 16:27:43 2013 -0800
  7792.  
  7793. x86: combined SA8D/SATD dsp function
  7794.  
  7795. Speedup is most apparent for 8-bit (~30%), but gives some improvements
  7796. for 10-bit too (~12%).
  7797. 64-bit only for now.
  7798.  
  7799. commit 790c648d939240808659228f57a22633fc59d6d8 [revision 2264]
  7800. Author: Oskar Arvidsson <[email protected]>
  7801. AuthorDate: Tue Jan 29 23:44:32 2013 +0100
  7802. Commit: Fiona Glaser <[email protected]>
  7803. CommitDate: Mon Feb 25 16:27:13 2013 -0800
  7804.  
  7805. x86: port SSE2+ SATD functions to high bit depth
  7806.  
  7807. Makes SATD 20-50% faster across all partition sizes but 4x4.
  7808.  
  7809. commit 93bf1248f7409958818b281e3e6ecca75ddb8d86 [revision 2263]
  7810. Author: Oskar Arvidsson <[email protected]>
  7811. AuthorDate: Wed Feb 6 02:07:53 2013 +0100
  7812. Commit: Fiona Glaser <[email protected]>
  7813. CommitDate: Mon Feb 25 12:16:51 2013 -0800
  7814.  
  7815. x86: faster high bit depth ssd
  7816.  
  7817. About 15% faster on average.
  7818.  
  7819. commit 6371c3a527a337c7521912990c89d0474288e105 [revision 2262]
  7820. Author: Fiona Glaser <[email protected]>
  7821. AuthorDate: Fri Jan 18 22:55:46 2013 -0800
  7822. Commit: Fiona Glaser <[email protected]>
  7823. CommitDate: Mon Feb 25 12:14:35 2013 -0800
  7824.  
  7825. x86: optimize and clean up predictor checking
  7826. Branchlessly handle elimination of candidates in MMX roundclip asm.
  7827. Add a new asm function, similar to roundclip, except without the round part.
  7828. Optimize and organize the C code, and make both subme>=3 and subme<3 consistent.
  7829. Add lots of explanatory comments and try to make things a little more understandable.
  7830. ~5-10% faster with subme>=3, ~15-20% faster with subme<3.
  7831.  
  7832. commit 004640653ded52f447ffdb71a45b334dc8e6f3d1 [revision 2261]
  7833. Author: Fiona Glaser <[email protected]>
  7834. AuthorDate: Tue Jan 22 12:31:55 2013 -0800
  7835. Commit: Fiona Glaser <[email protected]>
  7836. CommitDate: Mon Feb 25 12:14:35 2013 -0800
  7837.  
  7838. Fix two bugs in predictor checking
  7839. pmv wasn't checked properly in some cases, as well as zero vector.
  7840. Output-changing portion of the following patch.
  7841.  
  7842. commit d2a9d25429b6843874865a37a5b4f6b401d89abc [revision 2260]
  7843. Author: Fiona Glaser <[email protected]>
  7844. AuthorDate: Thu Jan 10 13:15:52 2013 -0800
  7845. Commit: Fiona Glaser <[email protected]>
  7846. CommitDate: Mon Feb 25 12:14:34 2013 -0800
  7847.  
  7848. Improve lookahead-threads auto selection
  7849. Smarter decision to improve fast-first-pass performance in 2-pass encodes.
  7850. Dramatically improves CPU utilization on multi-core systems.
  7851.  
  7852. Tested on a quad-core Ivy Bridge (12 threads, 1080p):
  7853. Fast first pass:
  7854. veryfast: ~7% faster
  7855. faster: ~11% faster
  7856. fast/medium: ~15% faster
  7857. slow/slower: ~42% faster
  7858. veryslow: ~55% faster
  7859. CRF/1-pass:
  7860. veryfast: ~9% faster
  7861. (all others remained the same)
  7862.  
  7863. commit 5a764328bdeba650d99fc8db47275708cce79521 [revision 2259]
  7864. Author: Henrik Gramner <[email protected]>
  7865. AuthorDate: Sun Jan 27 23:01:59 2013 +0100
  7866. Commit: Fiona Glaser <[email protected]>
  7867. CommitDate: Mon Feb 25 12:14:34 2013 -0800
  7868.  
  7869. x86: Use SSE instead of SSE2 for copying data
  7870.  
  7871. Reduces code size because movaps/movups is one byte shorter than movdqa/movdqu.
  7872. Also merge MMX and SSE versions of memcpy_aligned into a single macro.
  7873.  
  7874. commit c3983b811f42ae5e4bc4f9c1c919f8e548fc76e3 [revision 2258]
  7875. Author: Henrik Gramner <[email protected]>
  7876. AuthorDate: Sun Jan 13 18:27:08 2013 +0100
  7877. Commit: Fiona Glaser <[email protected]>
  7878. CommitDate: Mon Feb 25 12:14:34 2013 -0800
  7879.  
  7880. 64-bit cabac optimizations
  7881.  
  7882. ~4% faster PIC
  7883.  
  7884. WIN64:
  7885. ~3% faster and 16 byte shorter cabac_encode_bypass
  7886. ~8% faster cabac_encode_terminal
  7887. Benchmarked on Ivy Bridge
  7888.  
  7889. UNIX64:
  7890. One instruction less in cabac_encode_bypass
  7891.  
  7892. commit f6e0d28ae1bccbda43d95200162f7035661fe1e4 [revision 2257]
  7893. Author: Mike Gorchak <[email protected]>
  7894. AuthorDate: Sat Feb 2 23:35:00 2013 -0800
  7895. Commit: Fiona Glaser <[email protected]>
  7896. CommitDate: Mon Feb 25 12:14:33 2013 -0800
  7897.  
  7898. configure: add QNX support
  7899.  
  7900. commit 5e0fca86444840752eaedbdc5ebfe4ac0b3a0053 [revision 2256]
  7901. Author: Henrik Gramner <[email protected]>
  7902. AuthorDate: Sun Jan 20 19:35:06 2013 +0100
  7903. Commit: Fiona Glaser <[email protected]>
  7904. CommitDate: Mon Feb 25 12:14:33 2013 -0800
  7905.  
  7906. Windows: Enable DEP and ASLR
  7907.  
  7908. commit 5ec5c78920914a88da415c57904fa01c99deeb7b [revision 2255]
  7909. Author: Henrik Gramner <[email protected]>
  7910. AuthorDate: Thu Jan 17 19:17:24 2013 +0100
  7911. Commit: Fiona Glaser <[email protected]>
  7912. CommitDate: Mon Feb 25 12:14:33 2013 -0800
  7913.  
  7914. x86inc: Set ELF hidden visibility for global constants
  7915.  
  7916. commit fd2c4a06c3a4eb02fc1375de782bc5d36eb1d744 [revision 2254]
  7917. Author: Diego Biurrun <[email protected]>
  7918. AuthorDate: Thu Jan 17 11:18:31 2013 +0100
  7919. Commit: Fiona Glaser <[email protected]>
  7920. CommitDate: Mon Feb 25 12:14:32 2013 -0800
  7921.  
  7922. x86inc: Add cvisible macro for C functions with public prefix
  7923.  
  7924. This allows defining externally visible library symbols.
  7925.  
  7926. Signed-off-by: Diego Biurrun <[email protected]>
  7927.  
  7928. commit faf3dbe616c8339590409c9aa25777fa76c987a6 [revision 2253]
  7929. Author: Diego Biurrun <[email protected]>
  7930. AuthorDate: Thu Jan 17 11:30:37 2013 -0800
  7931. Commit: Fiona Glaser <[email protected]>
  7932. CommitDate: Mon Feb 25 12:14:32 2013 -0800
  7933.  
  7934. x86inc: rename program_name to private_prefix
  7935. Synced from libav.
  7936. The new name is more descriptive and will allow defining a separate public
  7937. prefix for externally visible library symbols.
  7938.  
  7939. commit 32695340b0e93e3cc7edd1b5e7db064d94cd3701 [revision 2252]
  7940. Author: Fiona Glaser <[email protected]>
  7941. AuthorDate: Mon Jan 14 05:35:30 2013 -0800
  7942. Commit: Fiona Glaser <[email protected]>
  7943. CommitDate: Mon Feb 25 12:14:32 2013 -0800
  7944.  
  7945. x264.h: improve x264_encoder_reconfig documentation
  7946.  
  7947. commit 6a82e49370e46914ab479d57548508ccf29da6e5 [revision 2251]
  7948. Author: Henrik Gramner <[email protected]>
  7949. AuthorDate: Sat Feb 16 19:36:50 2013 +0100
  7950. Commit: Fiona Glaser <[email protected]>
  7951. CommitDate: Mon Feb 25 12:14:31 2013 -0800
  7952.  
  7953. Cosmetics: stricter definition of parameterless functions
  7954.  
  7955. commit b671762973a162705ceacf924a29999cdc6d35d2 [revision 2250]
  7956. Author: Neil <[email protected]>
  7957. AuthorDate: Mon Jan 28 10:47:38 2013 +0800
  7958. Commit: Fiona Glaser <[email protected]>
  7959. CommitDate: Mon Feb 25 12:14:31 2013 -0800
  7960.  
  7961. Update "Install and compile x264" in doc/regression_test.txt
  7962.  
  7963. commit 43ff8f1681c1cca997ca916508723abea85d0fa2 [revision 2249]
  7964. Author: Anton Mitrofanov <[email protected]>
  7965. AuthorDate: Thu Jan 24 12:11:26 2013 +0400
  7966. Commit: Fiona Glaser <[email protected]>
  7967. CommitDate: Mon Feb 25 12:14:31 2013 -0800
  7968.  
  7969. Fix possible non-determinism with mbtree + open-gop + sync-lookahead
  7970.  
  7971. Code assumed keyframe analysis would only pull one frame off the list; this
  7972. isn't true with open-gop.
  7973.  
  7974. commit c2c2a95708685156a643e920b497d48597e0267c [revision 2248]
  7975. Author: Anton Mitrofanov <[email protected]>
  7976. AuthorDate: Mon Feb 25 19:28:19 2013 +0400
  7977. Commit: Fiona Glaser <[email protected]>
  7978. CommitDate: Mon Feb 25 12:14:30 2013 -0800
  7979.  
  7980. x86: don't use the red zone on win64
  7981.  
  7982. commit 5743b19a8264415ab3ed443abd2fefd81a038d6a [revision 2247]
  7983. Author: Fiona Glaser <[email protected]>
  7984. AuthorDate: Sun Feb 10 16:12:34 2013 -0800
  7985. Commit: Fiona Glaser <[email protected]>
  7986. CommitDate: Mon Feb 25 12:14:21 2013 -0800
  7987.  
  7988. x86-64: fix trellis asm with interlacing
  7989.  
  7990. Regression in r2145.
  7991. Assembly assumed array was [2][64] when it was actually [2][63].
  7992. Tiny (~0.1%) compression improvement.
  7993.  
  7994. commit 9475e6ac48af90e526611b5f11a2690fa077b0ba [revision 2246]
  7995. Author: Ronald S. Bultje <[email protected]>
  7996. AuthorDate: Wed Jan 30 09:48:14 2013 -0800
  7997. Commit: Fiona Glaser <[email protected]>
  7998. CommitDate: Mon Feb 25 12:14:04 2013 -0800
  7999.  
  8000. x86-32: use simple nop codes for <= sse
  8001.  
  8002. The "CentaurHauls family 6 model 9 stepping 8" family of CPUs (flags:
  8003. fpu vme de pse tsc msr cx8 sep mtrr pge mov pat mmx fxsr sse up rng
  8004. rng_en ace ace_en) SIGILLs on long nop codes.
  8005.  
  8006. commit 732b072ae236b57cabdbc3b31cd7b482d1f9f9ff [revision 2245]
  8007. Author: Loren Merritt <[email protected]>
  8008. AuthorDate: Tue Jan 8 21:30:57 2013 +0000
  8009. Commit: Fiona Glaser <[email protected]>
  8010. CommitDate: Tue Jan 8 16:01:32 2013 -0800
  8011.  
  8012. Bump dates to 2013
  8013.  
  8014. commit f2b4f29c636d5e5c223650c5b22bd8089adfcab9 [revision 2244]
  8015. Author: Henrik Gramner <[email protected]>
  8016. AuthorDate: Mon Dec 17 21:54:00 2012 +0100
  8017. Commit: Fiona Glaser <[email protected]>
  8018. CommitDate: Tue Jan 8 16:01:32 2013 -0800
  8019.  
  8020. x86inc: Drop tzcnt workaround
  8021.  
  8022. It is no longer needed now that we've bumped the version requirement of yasm to 1.2.0.
  8023.  
  8024. commit ccda1ba4d8d902945c68aa25ec20867055d1b079 [revision 2243]
  8025. Author: Fiona Glaser <[email protected]>
  8026. AuthorDate: Mon Nov 12 10:28:53 2012 -0800
  8027. Commit: Fiona Glaser <[email protected]>
  8028. CommitDate: Tue Jan 8 16:01:32 2013 -0800
  8029.  
  8030. AVX2/FMA3 version of mbtree_propagate
  8031. First AVX2 function for testing.
  8032. Bump yasm version to 1.2.0 for AVX2 support.
  8033.  
  8034. commit 8a9608bbbdf77ceb3ee537271549111468175a2b [revision 2242]
  8035. Author: Henrik Gramner <[email protected]>
  8036. AuthorDate: Tue Dec 11 16:05:34 2012 +0100
  8037. Commit: Fiona Glaser <[email protected]>
  8038. CommitDate: Tue Jan 8 16:01:30 2013 -0800
  8039.  
  8040. x86inc: Use VEX-encoded instructions in AVX functions
  8041. Automatically use VEX-encoding in AVX/AVX2/XOP/FMA3/FMA4 functions for all instructions that exists in a VEX-encoded version.
  8042. This change makes it easier to extend existing code to use AVX2.
  8043. Also add support for AVX emulation of a few instructions that were missing before.
  8044.  
  8045. commit 4cf272851a9d24aacdf664f27a87ebdbfb50e6c2 [revision 2241]
  8046. Author: Loren Merritt <[email protected]>
  8047. AuthorDate: Sun Dec 2 15:56:30 2012 +0000
  8048. Commit: Fiona Glaser <[email protected]>
  8049. CommitDate: Tue Jan 8 13:52:27 2013 -0800
  8050.  
  8051. x86inc: activate REP_RET automatically
  8052. Now RET checks whether it immediately follows a branch, so the programmer dosen't have to keep track of that condition.
  8053. REP_RET is still needed manually when it's a branch target, but that's much rarer.
  8054. The implementation involves lots of spurious labels, but that's ok because we strip them.
  8055.  
  8056. commit b073e870d135ac27cd97d624330abf0f1fb1ed41 [revision 2240]
  8057. Author: Ronald S. Bultje <[email protected]>
  8058. AuthorDate: Thu Dec 6 15:40:13 2012 -0800
  8059. Commit: Fiona Glaser <[email protected]>
  8060. CommitDate: Tue Jan 8 13:52:21 2013 -0800
  8061.  
  8062. x86inc: support stack mem allocation and re-alignment in PROLOGUE
  8063. Use this in 8-bit loopfilter functions so they can be used if
  8064. there is no aligned stack (e.g. x86-32 MSVC or ICC 10.x).
  8065.  
  8066. commit 9d5ec55b34a4d4f2e044fbc67e2e12a59ea27d2a [revision 2239]
  8067. Author: Henrik Gramner <[email protected]>
  8068. AuthorDate: Mon Dec 17 22:15:02 2012 +0100
  8069. Commit: Fiona Glaser <[email protected]>
  8070. CommitDate: Tue Jan 8 13:30:50 2013 -0800
  8071.  
  8072. Update config.guess and config.sub
  8073.  
  8074. commit 8eddd52b6d5d638709c5c8278c420eac68a8dde1 [revision 2238]
  8075. Author: Anton Mitrofanov <[email protected]>
  8076. AuthorDate: Tue Jan 8 13:29:49 2013 -0800
  8077. Commit: Fiona Glaser <[email protected]>
  8078. CommitDate: Tue Jan 8 13:30:50 2013 -0800
  8079.  
  8080. Fix crash if the first frame is forced to a non-keyframe
  8081. This is obviously bad user input, but x264 shouldn't crash if it happens.
  8082.  
  8083. commit 05c1646333f567aa3de5f7669693b15ee667825d [revision 2237]
  8084. Author: Bernhard Rosenkränzer <[email protected]>
  8085. AuthorDate: Sun Dec 30 12:18:00 2012 -0800
  8086. Commit: Fiona Glaser <[email protected]>
  8087. CommitDate: Tue Jan 8 11:55:22 2013 -0800
  8088.  
  8089. Fix build on ARM with binutils >= 2.23.51.0.6
  8090. GAS doesn't seem to like spaces in vld1 anymore, so remove those.
  8091.  
  8092. commit 23829dd2b2c909855481f46cc884b3c25d92c2d7 [revision 2236]
  8093. Author: Anton Mitrofanov <[email protected]>
  8094. AuthorDate: Fri Nov 23 18:26:53 2012 +0400
  8095. Commit: Fiona Glaser <[email protected]>
  8096. CommitDate: Wed Dec 12 12:16:26 2012 -0800
  8097.  
  8098. Fix pthread_join emulation on win32 and BeOS
  8099. Doesn't actually affect x264, but it's more correct.
  8100.  
  8101. commit 042fdd3e6a0e271f62a108da2a1a244dee936045 [revision 2235]
  8102. Author: Fiona Glaser <[email protected]>
  8103. AuthorDate: Tue Nov 27 07:50:51 2012 -0800
  8104. Commit: Fiona Glaser <[email protected]>
  8105. CommitDate: Thu Dec 6 15:52:23 2012 -0800
  8106.  
  8107. Fix typo in r2222
  8108. Slightly wrong numbers in level table.
  8109.  
  8110. commit cd71765c0ba574bb573e75396ef3c6a5c4c00469 [revision 2234]
  8111. Author: Sergio Basto <[email protected]>
  8112. AuthorDate: Thu Nov 22 18:02:50 2012 -0800
  8113. Commit: Fiona Glaser <[email protected]>
  8114. CommitDate: Thu Dec 6 15:52:23 2012 -0800
  8115.  
  8116. configure: fix gpac detection with -Wp,-D_FORTIFY_SOURCE=2
  8117.  
  8118. commit 12458a23d1374836fecbed381dfe55513b5ba119 [revision 2233]
  8119. Author: Sean McGovern <[email protected]>
  8120. AuthorDate: Thu Nov 22 18:01:16 2012 -0800
  8121. Commit: Fiona Glaser <[email protected]>
  8122. CommitDate: Thu Dec 6 15:52:22 2012 -0800
  8123.  
  8124. Solaris: use sysconf to get processor count
  8125. Solaris responds correctly to the same value as Cygwin, so let's use that.
  8126.  
  8127. commit 0db80bee2765676c2e0e4be21afc2ace900a606c [revision 2232]
  8128. Author: Anton Khirnov <[email protected]>
  8129. AuthorDate: Tue Nov 13 21:01:24 2012 +0100
  8130. Commit: Fiona Glaser <[email protected]>
  8131. CommitDate: Mon Nov 19 00:39:08 2012 -0800
  8132.  
  8133. lavf input: allocate AVFrame correctly
  8134. Allocate AVFrames correctly with avcodec_alloc_frame().
  8135. This caused crashes with newer libavcodecs that try to free frame extradata.
  8136.  
  8137. commit 144b79159ad20954a7faec1023451a630a65aea1 [revision 2231]
  8138. Author: Anton Mitrofanov <[email protected]>
  8139. AuthorDate: Sun Nov 11 03:44:02 2012 +0400
  8140. Commit: Fiona Glaser <[email protected]>
  8141. CommitDate: Mon Nov 12 11:07:13 2012 -0800
  8142.  
  8143. Fix crash when using libx264.dll compiled with ICL for X86_64
  8144.  
  8145. commit bfed708c5358a2b4ef65923fb0683cefa9184e6f [revision 2230]
  8146. Author: Anton Mitrofanov <[email protected]>
  8147. AuthorDate: Fri Nov 9 02:31:10 2012 +0400
  8148. Commit: Fiona Glaser <[email protected]>
  8149. CommitDate: Thu Nov 8 14:42:01 2012 -0800
  8150.  
  8151. Fix possible issues with out-of-spec QP values
  8152. Fixes a possible regression in r2228.
  8153.  
  8154. commit 1580a74e339c59cd856100076d8cf46f2d7247b0 [revision 2229]
  8155. Author: Fiona Glaser <[email protected]>
  8156. AuthorDate: Wed Sep 26 13:49:02 2012 -0700
  8157. Commit: Fiona Glaser <[email protected]>
  8158. CommitDate: Wed Nov 7 14:52:35 2012 -0800
  8159.  
  8160. Attempt to optimize PPS pic_init_qp in 2-pass mode
  8161. Small compression improvement; up to ~0.5% in extreme cases.
  8162. Helps more with small slice sizes (tiny resolutions or slice-max-size).
  8163. Note that this changes the 2-pass stats file format.
  8164.  
  8165. commit b304a7cad10a85d487fa09e7c33e81c6945186b2 [revision 2228]
  8166. Author: Fiona Glaser <[email protected]>
  8167. AuthorDate: Wed Sep 26 13:05:00 2012 -0700
  8168. Commit: Fiona Glaser <[email protected]>
  8169. CommitDate: Wed Nov 7 14:52:34 2012 -0800
  8170.  
  8171. Improve slice header QP selection
  8172. Use the first macroblock of each slice instead of the last of the previous.
  8173. Lets us pick a reasonable initial QP for the first slice too.
  8174. Slightly improved compression.
  8175.  
  8176. commit 0d5f6fbae9f6c4dbba25571a5d8c643b192606b1 [revision 2227]
  8177. Author: Fiona Glaser <[email protected]>
  8178. AuthorDate: Thu Oct 11 13:27:48 2012 -0700
  8179. Commit: Fiona Glaser <[email protected]>
  8180. CommitDate: Wed Nov 7 14:52:34 2012 -0800
  8181.  
  8182. Update level dpb size calculation to match newer H.264 spec
  8183. Doesn't actually change encoding behavior, but makes it more correct.
  8184. Warning messages should now be accurate at higher bit depths and non-4:2:0.
  8185. Technically, since it redefines x264_level_t, this is an API version increment.
  8186.  
  8187. commit cc61a4b4d0838b6d5f4cdaf88a0b6d06a12b6d3e [revision 2226]
  8188. Author: Jan Ekström <[email protected]>
  8189. AuthorDate: Sun Oct 7 21:12:05 2012 +0300
  8190. Commit: Fiona Glaser <[email protected]>
  8191. CommitDate: Wed Nov 7 14:52:34 2012 -0800
  8192.  
  8193. Add support for the ffmpeg/vapoursynth high bit depth y4m extensions
  8194.  
  8195. commit 5d85879921481ef104766657deda4ef8ea4351ec [revision 2225]
  8196. Author: Diego Biurrun <[email protected]>
  8197. AuthorDate: Tue Nov 6 14:48:56 2012 +0100
  8198. Commit: Fiona Glaser <[email protected]>
  8199. CommitDate: Wed Nov 7 14:51:05 2012 -0800
  8200.  
  8201. x86inc: Rename 3dnow2 to 3dnowext
  8202. The name "3dnowext" is more common than "3dnow2". Doesn't affect x264.
  8203.  
  8204. commit 00cc16001b35a71ce2329e02bff6e316201cf700 [revision 2224]
  8205. Author: Diego Biurrun <[email protected]>
  8206. AuthorDate: Wed Oct 31 12:23:54 2012 -0700
  8207. Commit: Fiona Glaser <[email protected]>
  8208. CommitDate: Wed Nov 7 14:48:38 2012 -0800
  8209.  
  8210. x86inc: only define program_name if the macro is unset.
  8211. This allows overriding the value from outside the file.
  8212. This can be useful if x86inc.asm is used outside of x264.
  8213.  
  8214. commit 3f516c5238d0c536ea03c8e5334d231facf9f31b [revision 2223]
  8215. Author: David Wolstencroft <[email protected]>
  8216. AuthorDate: Mon Oct 29 09:07:39 2012 -0700
  8217. Commit: Fiona Glaser <[email protected]>
  8218. CommitDate: Wed Nov 7 14:48:37 2012 -0800
  8219.  
  8220. Disable ARM NEON MRC CPU test for Apple devices
  8221. The Apple A6 CPU doesn't support performance counters, so this test caused a crash.
  8222.  
  8223. commit ac2d7c08452186703424dcc6933524e95b652479 [revision 2222]
  8224. Author: Fiona Glaser <[email protected]>
  8225. AuthorDate: Tue Nov 6 12:03:20 2012 -0800
  8226. Commit: Fiona Glaser <[email protected]>
  8227. CommitDate: Wed Nov 7 14:48:37 2012 -0800
  8228.  
  8229. Fix crash with no-scenecut + mbtree
  8230.  
  8231. commit 480bbc9067da7cce3400cf3988bf5fdfa4d9fa3f [revision 2221]
  8232. Author: Anton Mitrofanov <[email protected]>
  8233. AuthorDate: Fri Oct 12 23:43:40 2012 +0400
  8234. Commit: Fiona Glaser <[email protected]>
  8235. CommitDate: Wed Nov 7 14:48:37 2012 -0800
  8236.  
  8237. Fix reconfiguring to crf=0
  8238. Lossless mode can't currently be enabled mid-stream.
  8239.  
  8240. commit 21ba91ae6c361e4ce49ee65e61cc582b1af648ba [revision 2220]
  8241. Author: Derek Buitenhuis <[email protected]>
  8242. AuthorDate: Mon Sep 17 11:09:20 2012 -0700
  8243. Commit: Fiona Glaser <[email protected]>
  8244. CommitDate: Wed Nov 7 14:48:35 2012 -0800
  8245.  
  8246. Fix ALIGNED_ARRAY_EMU macros on ICL
  8247. ICL's preprocessor doesn't handle it correctly.
  8248. This fix is similar to libav's fix in 0db2d9.
  8249.  
  8250. commit 96577475981d979d151626aae61ef317dc54df67 [revision 2219]
  8251. Author: Jason Martens <[email protected]>
  8252. AuthorDate: Thu Sep 13 11:20:40 2012 -0700
  8253. Commit: Fiona Glaser <[email protected]>
  8254. CommitDate: Wed Sep 26 14:14:02 2012 -0700
  8255.  
  8256. Fix use of deprecated av_close_input_file call
  8257.  
  8258. commit 02217bd2c31feda7aaca813f104c155fe09428b8 [revision 2218]
  8259. Author: Brad Smith <[email protected]>
  8260. AuthorDate: Wed Sep 26 14:13:27 2012 -0700
  8261. Commit: Fiona Glaser <[email protected]>
  8262. CommitDate: Wed Sep 26 14:14:00 2012 -0700
  8263.  
  8264. Fix pkg-config for dynamic vs static linking
  8265.  
  8266. commit e8e8b9a44ffa9b5f585582375515140ea22985d3 [revision 2217]
  8267. Author: Brad Smith <[email protected]>
  8268. AuthorDate: Mon Sep 10 17:52:04 2012 -0700
  8269. Commit: Fiona Glaser <[email protected]>
  8270. CommitDate: Mon Sep 10 17:52:04 2012 -0700
  8271.  
  8272. Set libm in the configure script if the OS has libm
  8273. Prerequisite for another configure patch after this.
  8274. Idea copied from libpthread.
  8275.  
  8276. commit 8980dd8afbfeeb6bcaa17b97aad0b3c24207665e [revision 2216]
  8277. Author: Fiona Glaser <[email protected]>
  8278. AuthorDate: Thu Aug 16 13:40:32 2012 -0700
  8279. Commit: Fiona Glaser <[email protected]>
  8280. CommitDate: Wed Sep 5 11:59:47 2012 -0700
  8281.  
  8282. Enhance mb_info: add mb_info_update
  8283. This feature lets the callee know which decoded macroblocks have changed.
  8284.  
  8285. commit 033df0a8c719f991ab0e0bb0788bd4f08e8b91d7 [revision 2215]
  8286. Author: Fiona Glaser <[email protected]>
  8287. AuthorDate: Thu Aug 16 13:01:17 2012 -0700
  8288. Commit: Fiona Glaser <[email protected]>
  8289. CommitDate: Wed Sep 5 11:59:46 2012 -0700
  8290.  
  8291. Fix mb_info_free with sliced threads
  8292. x264 would free mb_info before it was completely done using it.
  8293.  
  8294. commit f93b7865a96248621af078363d5b59691cbcd8aa [revision 2214]
  8295. Author: Fiona Glaser <[email protected]>
  8296. AuthorDate: Tue Aug 7 12:43:26 2012 -0700
  8297. Commit: Fiona Glaser <[email protected]>
  8298. CommitDate: Wed Sep 5 11:59:46 2012 -0700
  8299.  
  8300. Enhance nalu_process
  8301. Add the input frame opaque pointer to the arguments.
  8302. This makes it easier to use with multiple simultaneous x264 encodes.
  8303.  
  8304. commit 05089a37bf55a4134d9ffd014fdae729804a4e7a [revision 2213]
  8305. Author: Fiona Glaser <[email protected]>
  8306. AuthorDate: Mon Aug 6 14:55:35 2012 -0700
  8307. Commit: Fiona Glaser <[email protected]>
  8308. CommitDate: Wed Sep 5 11:59:46 2012 -0700
  8309.  
  8310. Improve mb_info constant mb optimization
  8311. Allow fast skipping even if the pskip MV isn't zero.
  8312.  
  8313. commit cc5dcedc3b45d8e7390e2e914bb37f3fa92f6acd [revision 2212]
  8314. Author: Fiona Glaser <[email protected]>
  8315. AuthorDate: Mon Jul 30 12:58:34 2012 -0700
  8316. Commit: Fiona Glaser <[email protected]>
  8317. CommitDate: Wed Sep 5 11:59:45 2012 -0700
  8318.  
  8319. Export the average effective CRF of each frame
  8320. Useful to judge the resulting quality of a frame when VBV is enabled.
  8321.  
  8322. commit f8fd6412a94f5f4f0eb5f8a6c0fb2062daebfab8 [revision 2211]
  8323. Author: Brad Smith <[email protected]>
  8324. AuthorDate: Mon Aug 20 23:58:19 2012 -0700
  8325. Commit: Fiona Glaser <[email protected]>
  8326. CommitDate: Wed Sep 5 11:59:45 2012 -0700
  8327.  
  8328. Remove special-casing for OpenBSD pthread handling
  8329. Previously it was policy to use -pthread, but OpenBSD now recommends -lpthread.
  8330. its been libpthread anyway and policy has changed to stop using -pthread.
  8331.  
  8332. commit ed56837e3c56bfb880fac2e4e0025d81d6a7186b [revision 2210]
  8333. Author: Ronald S. Bultje <[email protected]>
  8334. AuthorDate: Thu Jul 26 18:01:49 2012 -0700
  8335. Commit: Fiona Glaser <[email protected]>
  8336. CommitDate: Thu Jul 26 18:01:49 2012 -0700
  8337.  
  8338. x86inc: automatically insert vzeroupper for YMM functions
  8339. Backported from libav.
  8340.  
  8341. commit cbb90707e443f0da2521bda1b98cab5705451b8f [revision 2209]
  8342. Author: Kieran Kunhya <[email protected]>
  8343. AuthorDate: Tue Jul 24 08:47:45 2012 -0700
  8344. Commit: Fiona Glaser <[email protected]>
  8345. CommitDate: Thu Jul 26 07:42:05 2012 -0700
  8346.  
  8347. Free user supplied data when deleting a frame
  8348. This eliminates a memory leak when calling x264_encoder_close.
  8349.  
  8350. commit 3d03b6190c7af7b941fa746c3dff3b17e5115380 [revision 2208]
  8351. Author: Fiona Glaser <[email protected]>
  8352. AuthorDate: Wed Jul 18 08:33:41 2012 -0700
  8353. Commit: Fiona Glaser <[email protected]>
  8354. CommitDate: Wed Jul 18 09:15:58 2012 -0700
  8355.  
  8356. Revert r2204
  8357. People don't seem to like this so I'm just going to get rid of it.
  8358.  
  8359. commit 2ec694181f8ba3eb1c4153e6b955d399d6448c25 [revision 2207]
  8360. Author: Fiona Glaser <[email protected]>
  8361. AuthorDate: Tue Jul 10 14:10:44 2012 -0700
  8362. Commit: Fiona Glaser <[email protected]>
  8363. CommitDate: Tue Jul 17 12:19:14 2012 -0700
  8364.  
  8365. Faster predictor checking with subme<3
  8366. Fix a typo that made an early-skip less effective.
  8367. Avoid a relatively unpredictable branch.
  8368. Slightly changed output due to the typo-fix.
  8369. ~50 cycles faster on Core i7.
  8370.  
  8371. commit d026397b0bf4c87e96b19c9fff7f43be6c4d9def [revision 2206]
  8372. Author: Fiona Glaser <[email protected]>
  8373. AuthorDate: Mon Jun 25 18:01:29 2012 -0700
  8374. Commit: Fiona Glaser <[email protected]>
  8375. CommitDate: Tue Jul 17 12:19:13 2012 -0700
  8376.  
  8377. Try 8x8 transform analysis even when sub8x8 partitions are present
  8378. Turn off the sub8x8 partitions, try it, and turn them back on if it didn't help.
  8379. Small compression improvement with p4x4 on (~0.1-0.5%).
  8380. Also update related comments.
  8381.  
  8382. commit dea5d7a54b5ba948ed71d74e0264a2191bcd9815 [revision 2205]
  8383. Author: Fiona Glaser <[email protected]>
  8384. AuthorDate: Fri Jun 8 18:19:59 2012 -0700
  8385. Commit: Fiona Glaser <[email protected]>
  8386. CommitDate: Tue Jul 17 12:19:13 2012 -0700
  8387.  
  8388. Support changing resolutions between passes with macroblock-tree
  8389. Implement a basic separable bilinear filter to rescale the quantizer offsets.
  8390. Structure inspired by swscale, but floating-point instead of fixed-point.
  8391. Not as optimized as it could be, but it's quite fast already.
  8392.  
  8393. Example compression penalties on a 720p video game recording:
  8394. First pass with 720p and second as 480p: ~-1.5% (vs. same res)
  8395. First pass with 480p and second as 720p: ~-3% (vs. same res)
  8396.  
  8397. commit 498af9c559b8da986544e93f898df02fc9e224b3 [revision 2204]
  8398. Author: Alexander Prikhodko <[email protected]>
  8399. AuthorDate: Tue Jun 12 20:21:35 2012 +0300
  8400. Commit: Fiona Glaser <[email protected]>
  8401. CommitDate: Tue Jul 17 12:19:12 2012 -0700
  8402.  
  8403. Print elapsed time in encoding progress indicator
  8404.  
  8405. commit bcd1a7070dc5224d591731dfdbabcbbaee0bb984 [revision 2203]
  8406. Author: Anton Mitrofanov <[email protected]>
  8407. AuthorDate: Sat Jun 2 21:27:50 2012 +0400
  8408. Commit: Fiona Glaser <[email protected]>
  8409. CommitDate: Tue Jul 17 12:19:05 2012 -0700
  8410.  
  8411. Cap ratecontrol predictor parameters
  8412. Limits VBV mispredictions after long periods of relatively constant video.
  8413.  
  8414. commit 5754ea2db5223b458bd48f0130c13000e3dec15c [revision 2202]
  8415. Author: Loren Merritt <[email protected]>
  8416. AuthorDate: Tue Jul 3 14:38:04 2012 -0700
  8417. Commit: Fiona Glaser <[email protected]>
  8418. CommitDate: Tue Jul 3 14:39:51 2012 -0700
  8419.  
  8420. x86inc: import patches from libav
  8421. Allow manual invocation of WIN64_SPILL_XMM even under INIT_MMX
  8422. SSE version of mova is movaps rather than movdqa.
  8423. YMM version of movnta.
  8424. Add mp size for named arguments.
  8425. Fix DEFINE_ARGS when used outside of a cglobal.
  8426. Define a few more cpuflags.
  8427. 3-argument wrappers for a few more instructions.
  8428.  
  8429. commit 5e3aaf1a49b173df916a384942c8089dd5bd8a22 [revision 2201]
  8430. Author: Anton Mitrofanov <[email protected]>
  8431. AuthorDate: Fri Jun 22 22:02:24 2012 +0400
  8432. Commit: Fiona Glaser <[email protected]>
  8433. CommitDate: Tue Jul 3 13:10:02 2012 -0700
  8434.  
  8435. Fix crash with --fps 0
  8436. Fix some integer overflows and check input parameters better.
  8437. Also fix incorrect type specifiers for demuxer info printing.
  8438.  
  8439. commit df700eae5d5ce5732f80df9ce81e6d3fe99ef56a [revision 2200]
  8440. Author: Fiona Glaser <[email protected]>
  8441. AuthorDate: Tue May 8 15:42:56 2012 -0700
  8442. Commit: Fiona Glaser <[email protected]>
  8443. CommitDate: Fri May 18 16:15:14 2012 -0700
  8444.  
  8445. Threaded lookahead
  8446.  
  8447. Split each lookahead frame analysis call into multiple threads. Has a small
  8448. impact on quality, but does not seem to be consistently any worse.
  8449.  
  8450. This helps alleviate bottlenecks with many cores and frame threads. In many
  8451. case, this massively increases performance on many-core systems. For example,
  8452. over 100% faster 1080p encoding with --preset veryfast on a 12-core i7 system.
  8453. Realtime 1080p30 at --preset slow should now be feasible on real systems.
  8454.  
  8455. For sliced-threads, this patch should be faster regardless of settings (~10%).
  8456.  
  8457. By default, lookahead threads are 1/6 of regular threads. This isn't exacting,
  8458. but it seems to work well for all presets on real systems. With sliced-threads,
  8459. it's the same as the number of encoding threads.
  8460.  
  8461. commit 7cfe43cc7fb5474a87f02da96ebb850cdf83d73b [revision 2199]
  8462. Author: Anton Mitrofanov <[email protected]>
  8463. AuthorDate: Fri May 4 17:18:12 2012 +0400
  8464. Commit: Fiona Glaser <[email protected]>
  8465. CommitDate: Mon May 14 18:53:19 2012 -0700
  8466.  
  8467. Add support for RGB formats in bit-depth conversion filter
  8468.  
  8469. commit 44d2f0885cd95201b67ed54bab88e91f4ba1556e [revision 2198]
  8470. Author: Anton Mitrofanov <[email protected]>
  8471. AuthorDate: Sat May 12 13:57:49 2012 +0400
  8472. Commit: Fiona Glaser <[email protected]>
  8473. CommitDate: Mon May 14 18:53:18 2012 -0700
  8474.  
  8475. Fix some bugs in mb_info code
  8476.  
  8477. commit 8e57a9a0b5bddfecea5e45345c8c50efb0bac10d [revision 2197]
  8478. Author: Fiona Glaser <[email protected]>
  8479. AuthorDate: Thu Mar 29 14:14:07 2012 -0700
  8480. Commit: Fiona Glaser <[email protected]>
  8481. CommitDate: Tue Apr 24 14:13:17 2012 -0700
  8482.  
  8483. Add mb_info API for signalling constant macroblocks
  8484. Some use-cases of x264 involve encoding video with large constant areas of the frame.
  8485. Sometimes, the caller knows which areas these are, and can tell x264.
  8486. This API lets the caller do this and adds internal tracking of modifications to macroblocks to avoid problems.
  8487. This is really only suitable without B-frames.
  8488. An example use-case would be using x264 for VNC.
  8489.  
  8490. commit 4442eaceb4992098e1e4e30aa13e70bb35d2cae6 [revision 2196]
  8491. Author: Henrik Gramner <[email protected]>
  8492. AuthorDate: Sat Apr 7 00:40:09 2012 +0200
  8493. Commit: Fiona Glaser <[email protected]>
  8494. CommitDate: Mon Apr 23 15:00:47 2012 -0700
  8495.  
  8496. Faster chroma weight cost calculation
  8497.  
  8498. New assembly function with SSE2, SSSE3 and XOP implementations for calculating absolute sum of differences.
  8499.  
  8500. commit e8952dffa3b09700e5b7c5e56edd196f0b80a248 [revision 2195]
  8501. Author: Lucien <[email protected]>
  8502. AuthorDate: Sat Mar 31 13:42:49 2012 +0100
  8503. Commit: Fiona Glaser <[email protected]>
  8504. CommitDate: Mon Apr 23 13:28:47 2012 -0700
  8505.  
  8506. Add Level 5.2 support
  8507.  
  8508. commit 66acbbf6ce6b143cd164d251ceb160870e4ee720 [revision 2194]
  8509. Author: Henrik Gramner <[email protected]>
  8510. AuthorDate: Thu Apr 12 19:14:43 2012 +0200
  8511. Commit: Fiona Glaser <[email protected]>
  8512. CommitDate: Mon Apr 23 13:28:45 2012 -0700
  8513.  
  8514. Eradicate all mention of Extended Profile
  8515. x264 never supported it and never will because nobody uses it.
  8516.  
  8517. commit b0f44f9e106afadaded17009079c2281cb18eb56 [revision 2193]
  8518. Author: Anton Mitrofanov <[email protected]>
  8519. AuthorDate: Tue Apr 3 21:46:52 2012 +0400
  8520. Commit: Fiona Glaser <[email protected]>
  8521. CommitDate: Mon Apr 23 13:28:44 2012 -0700
  8522.  
  8523. Fix disabling of mbtree when using 2pass encoding and zones
  8524.  
  8525. commit ffea9f51f7f7e0a550c9326631a9c6f8c5c885be [revision 2192]
  8526. Author: Alexander Prikhodko <[email protected]>
  8527. AuthorDate: Sat Mar 31 12:06:21 2012 +0300
  8528. Commit: Fiona Glaser <[email protected]>
  8529. CommitDate: Mon Apr 23 13:28:42 2012 -0700
  8530.  
  8531. configure: force select -mXX gcc option for i386/x86-64
  8532. Makes multilib compilation more convenient.
  8533.  
  8534. commit f4aefb3853819adf633c56062d1be77db90819b6 [revision 2191]
  8535. Author: Rafaël Carré <[email protected]>
  8536. AuthorDate: Sun Apr 15 21:20:14 2012 -0400
  8537. Commit: Fiona Glaser <[email protected]>
  8538. CommitDate: Mon Apr 23 13:28:41 2012 -0700
  8539.  
  8540. Update config.guess and config.sub
  8541. Adds support for a bunch of targets, including:
  8542. aarch64 (armv8)
  8543. arm-linux-androideabi
  8544.  
  8545. commit 62d7007d35c5f0829d96b6ecf459f21d27210ef3 [revision 2190]
  8546. Author: Alexander Prikhodko <[email protected]>
  8547. AuthorDate: Sat Mar 31 11:33:41 2012 +0300
  8548. Commit: Fiona Glaser <[email protected]>
  8549. CommitDate: Mon Apr 23 13:28:39 2012 -0700
  8550.  
  8551. configure: correct use of RC variable and add --extra-rcflags
  8552.  
  8553. commit 70877e39a4abb4c24d1978a28202c9bf0dce8b47 [revision 2189]
  8554. Author: Steven Walters <[email protected]>
  8555. AuthorDate: Wed Mar 28 21:15:04 2012 -0400
  8556. Commit: Fiona Glaser <[email protected]>
  8557. CommitDate: Mon Apr 23 13:28:34 2012 -0700
  8558.  
  8559. ICL/MSVS: Fix shared library generation and usage
  8560. MSVS requires exported variables to be declared with the DATA keyword, and requires that imported variables be declared with dllimport.
  8561. This does not fix x264 cli being unable to use a shared library built by ICL however.
  8562.  
  8563. commit 52f7a149ef6c39eb0d7eec7884362ba31a4b05ba [revision 2188]
  8564. Author: Kieran Kunhya <[email protected]>
  8565. AuthorDate: Tue Mar 27 17:38:56 2012 +0100
  8566. Commit: Fiona Glaser <[email protected]>
  8567. CommitDate: Tue Mar 27 09:43:57 2012 -0700
  8568.  
  8569. Fix intra-refresh + hrd
  8570.  
  8571. commit fff12b1b7d8ce5cc9cfcfac09f089bae06cac6d5 [revision 2187]
  8572. Author: Anton Mitrofanov <[email protected]>
  8573. AuthorDate: Sun Mar 25 17:34:24 2012 +0400
  8574. Commit: Fiona Glaser <[email protected]>
  8575. CommitDate: Sun Mar 25 09:56:43 2012 -0700
  8576.  
  8577. Fix frame input colorspace check
  8578.  
  8579. commit 065fec2704f3c8c6f3f3f5b0fad6870a078ba48c [revision 2186]
  8580. Author: Fiona Glaser <[email protected]>
  8581. AuthorDate: Thu Mar 22 13:56:50 2012 -0700
  8582. Commit: Fiona Glaser <[email protected]>
  8583. CommitDate: Thu Mar 22 13:56:50 2012 -0700
  8584.  
  8585. Fix comment in deblock.c
  8586. The code does, in fact, handle CAVLC+8x8dct correctly already.
  8587.  
  8588. commit bca412764eb198433ca45abd097368e5154c7fbb [revision 2185]
  8589. Author: Fiona Glaser <[email protected]>
  8590. AuthorDate: Tue Mar 13 14:37:26 2012 -0700
  8591. Commit: Fiona Glaser <[email protected]>
  8592. CommitDate: Wed Mar 14 00:53:18 2012 -0700
  8593.  
  8594. Fix sliced-threads ratecontrol bug
  8595. Was using qp instead of qscale; could cause NANs (not to mention less accurate results).
  8596.  
  8597. commit e046ba72a4230fdd6c7907ebf7ae235edb98faf2 [revision 2184]
  8598. Author: Anton Mitrofanov <[email protected]>
  8599. AuthorDate: Sun Mar 11 23:08:18 2012 -0700
  8600. Commit: Fiona Glaser <[email protected]>
  8601. CommitDate: Sun Mar 11 23:10:23 2012 -0700
  8602.  
  8603. Fix clobbering of mutex/cvs
  8604. Regression in r2183.
  8605. Bizarrely seemed to work on many platforms, but crashed on win64 and may have been slower.
  8606. Only affected sliced threads during encoding, but could cause crashes on x264 encoder close even without sliced threads.
  8607.  
  8608. commit a155572ed547a3627ef00ca70ab804ff452147cd [revision 2183]
  8609. Author: Fiona Glaser <[email protected]>
  8610. AuthorDate: Fri Feb 24 13:34:39 2012 -0800
  8611. Commit: Fiona Glaser <[email protected]>
  8612. CommitDate: Tue Mar 6 18:11:32 2012 -0800
  8613.  
  8614. Sliced-threads: do hpel and deblock after returning
  8615. Lowers encoding latency around 14% in sliced threads mode with preset superfast.
  8616. Additionally, even if there is no waiting time between frames, this improves parallelism, because hpel+deblock are done during the (singlethreaded) lookahead.
  8617. For ease of debugging, dump-yuv forces all of the threads to wait and finish instead of setting b_full_recon.
  8618.  
  8619. commit 90408ecab16a06ceaa181ff2e495b8f1a9d170fa [revision 2182]
  8620. Author: Fiona Glaser <[email protected]>
  8621. AuthorDate: Fri Feb 24 13:16:52 2012 -0800
  8622. Commit: Fiona Glaser <[email protected]>
  8623. CommitDate: Tue Mar 6 17:13:26 2012 -0800
  8624.  
  8625. Add full-recon API option
  8626. Fully reconstruct frames even without dump-yuv.
  8627.  
  8628. commit 5b2c62aec269be7d0b1ff62df09660369f4e20e0 [revision 2181]
  8629. Author: Fiona Glaser <[email protected]>
  8630. AuthorDate: Wed Feb 22 13:33:36 2012 -0800
  8631. Commit: Fiona Glaser <[email protected]>
  8632. CommitDate: Tue Mar 6 17:13:23 2012 -0800
  8633.  
  8634. x86inc: switch to amdnops
  8635. Recent AMD CPUs' instruction decoders choke horribly on extremely long nops (i.e. with 4 prefixes).
  8636. Won't affect much, since we don't use ALIGN much.
  8637.  
  8638. commit 42db5e6f8f704a2b0a9edf5d9cd4a17d80e5b816 [revision 2180]
  8639. Author: Fiona Glaser <[email protected]>
  8640. AuthorDate: Tue Feb 14 16:54:03 2012 -0800
  8641. Commit: Fiona Glaser <[email protected]>
  8642. CommitDate: Tue Mar 6 17:13:21 2012 -0800
  8643.  
  8644. BMI1 decimate functions
  8645. Intel was nice enough to make tzcnt equal to "rep bsf", which is backwards-compatible.
  8646. This means we don't actually have to add new functions to make it work.
  8647.  
  8648. commit 92b0bd9665860d7b48f313d6fd72a583ecb01ddf [revision 2179]
  8649. Author: Fiona Glaser <[email protected]>
  8650. AuthorDate: Tue Feb 14 15:07:10 2012 -0800
  8651. Commit: Fiona Glaser <[email protected]>
  8652. CommitDate: Tue Mar 6 17:13:18 2012 -0800
  8653.  
  8654. Minor asm changes
  8655.  
  8656. commit 2535ba17b2598f4155955857c12d52a377a75517 [revision 2178]
  8657. Author: Fiona Glaser <[email protected]>
  8658. AuthorDate: Thu Feb 9 14:23:52 2012 -0800
  8659. Commit: Fiona Glaser <[email protected]>
  8660. CommitDate: Tue Mar 6 17:13:15 2012 -0800
  8661.  
  8662. Add row-reencoding support to VBV for improved accuracy
  8663. Extremely accurate, possibly 100% so (I can't get it to fail even with difficult VBVs).
  8664. Does not yet support rows split on slice boundaries (occurs often with slice-max-size/mbs).
  8665. Still inaccurate with sliced threads, but better than before.
  8666.  
  8667. commit bc473ddfd2f5925715d2895da666e214ebf04c84 [revision 2177]
  8668. Author: Fiona Glaser <[email protected]>
  8669. AuthorDate: Thu Feb 9 12:38:44 2012 -0800
  8670. Commit: Fiona Glaser <[email protected]>
  8671. CommitDate: Tue Mar 6 17:12:41 2012 -0800
  8672.  
  8673. Abstract bitstream backup/restore functions
  8674. Required for row re-encoding.
  8675.  
  8676. commit 48e8e52e740fdc7ddca792d4afe240a213f66df5 [revision 2176]
  8677. Author: Anton Mitrofanov <[email protected]>
  8678. AuthorDate: Thu Feb 9 15:27:53 2012 -0800
  8679. Commit: Fiona Glaser <[email protected]>
  8680. CommitDate: Tue Mar 6 17:10:30 2012 -0800
  8681.  
  8682. Add an small per-MB cost penalty for lowres
  8683. Helps avoid VBV predictors going nuts with very low-cost MBs.
  8684. One particular case this fixes is zero-cost MBs: adaptive quantization decreases the QP a lot, but (before this patch), no cost penalty gets factored in for this, because anything times zero is zero.
  8685.  
  8686. commit 1b31a10c7c3210d5eb14d522aaa0cfbe0e7a25e8 [revision 2175]
  8687. Author: Fiona Glaser <[email protected]>
  8688. AuthorDate: Mon Feb 13 18:31:51 2012 -0800
  8689. Commit: Fiona Glaser <[email protected]>
  8690. CommitDate: Tue Mar 6 10:38:07 2012 -0800
  8691.  
  8692. Remove explicit run calculation from coeff_level_run
  8693. Not necessary with the CAVLC lookup table for zero run codes.
  8694.  
  8695. commit 9da19fbee621ca5b052891b3c010f8bc89b2ba93 [revision 2174]
  8696. Author: Fiona Glaser <[email protected]>
  8697. AuthorDate: Mon Feb 13 13:20:06 2012 -0800
  8698. Commit: Fiona Glaser <[email protected]>
  8699. CommitDate: Tue Mar 6 10:38:05 2012 -0800
  8700.  
  8701. Export PSNR/SSIM in x264 API
  8702.  
  8703. commit 3a5f2fe30aeb5314b74f83b1960e9a40776347e9 [revision 2173]
  8704. Author: Ronald S. Bultje <[email protected]>
  8705. AuthorDate: Wed Feb 8 13:10:31 2012 -0800
  8706. Commit: Fiona Glaser <[email protected]>
  8707. CommitDate: Tue Mar 6 10:38:02 2012 -0800
  8708.  
  8709. x86inc: support yasm -f win64
  8710. Not necessary for x264, as -m amd64 already does the right thing, but used by external users of x86inc.
  8711.  
  8712. commit 3131a19cabcdca221ce4cd61a3cff68d99f1a517 [revision 2172]
  8713. Author: Henrik Gramner <[email protected]>
  8714. AuthorDate: Wed Feb 1 23:52:48 2012 +0100
  8715. Commit: Fiona Glaser <[email protected]>
  8716. CommitDate: Tue Mar 6 10:37:53 2012 -0800
  8717.  
  8718. Fix incorrect zero-extension assumptions in x86_64 asm
  8719. Some x264 asm assumed that the high 32 bits of registers containing "int" values would be zero.
  8720. This is almost always the case, and it seems to work with gcc, but it is *not* guaranteed by the ABI.
  8721. As a result, it breaks with some other compilers, like Clang, that take advantage of this in optimizations.
  8722. Accordingly, fix all x86 code by using intptr_t instead of int or using movsxd where neccessary.
  8723. Also add checkasm hack to detect when assembly functions incorrectly assumes that 32-bit integers are zero-extended to 64-bit.
  8724.  
  8725. commit d52d0b1e6a9323911818c2a89764f6827974e0f7 [revision 2171]
  8726. Author: Fiona Glaser <[email protected]>
  8727. AuthorDate: Thu Feb 23 09:11:23 2012 -0800
  8728. Commit: Fiona Glaser <[email protected]>
  8729. CommitDate: Tue Mar 6 08:51:13 2012 -0800
  8730.  
  8731. Fix possible alignment crash when linking from MSVC
  8732. x264_cavlc_init needs to be stack-aligned now.
  8733.  
  8734. commit 0a369502ab83c32ccebdb1888e6981ef872baaf0 [revision 2170]
  8735. Author: Anton Mitrofanov <[email protected]>
  8736. AuthorDate: Tue Feb 21 12:58:22 2012 -0800
  8737. Commit: Fiona Glaser <[email protected]>
  8738. CommitDate: Tue Mar 6 08:51:10 2012 -0800
  8739.  
  8740. Fix rare overflow in 10-bit intra_satd_x3_16x16 asm
  8741.  
  8742. commit 38a26cdfc54ffd60c90651f3b96490d772e6dd73 [revision 2169]
  8743. Author: Steven Walters <[email protected]>
  8744. AuthorDate: Sat Feb 11 22:56:43 2012 -0500
  8745. Commit: Fiona Glaser <[email protected]>
  8746. CommitDate: Tue Mar 6 08:51:07 2012 -0800
  8747.  
  8748. ICL: fix out of tree building and resource file usage on Windows
  8749.  
  8750. commit 10e1ba55803970ecd240f2057e7dfe0c22fc8efb [revision 2168]
  8751. Author: Oka Motofumi <[email protected]>
  8752. AuthorDate: Mon Feb 6 06:07:34 2012 +0900
  8753. Commit: Fiona Glaser <[email protected]>
  8754. CommitDate: Tue Mar 6 08:51:04 2012 -0800
  8755.  
  8756. Add error handling for out-of-tree build
  8757.  
  8758. commit 0fc5acc6e6c038f6380f614e4dc4e1893b716b7e [revision 2167]
  8759. Author: Anton Mitrofanov <[email protected]>
  8760. AuthorDate: Tue Mar 6 17:34:02 2012 +0400
  8761. Commit: Fiona Glaser <[email protected]>
  8762. CommitDate: Tue Mar 6 08:51:01 2012 -0800
  8763.  
  8764. Fix RGB colorspace input
  8765. BGR/BGRA input was correct.
  8766.  
  8767. commit 282c3cfb22f4ab526d96678249ccdc7f16531811 [revision 2166]
  8768. Author: Fiona Glaser <[email protected]>
  8769. AuthorDate: Mon Feb 13 16:40:32 2012 -0800
  8770. Commit: Fiona Glaser <[email protected]>
  8771. CommitDate: Wed Feb 15 14:23:19 2012 -0800
  8772.  
  8773. Fix interlaced + extremal slice-max-size
  8774. Broke if the first macroblock in the slice exceeded the set slice-max-size.
  8775.  
  8776. commit a37a42450cdc31393dae56aed5a726a42fd540d6 [revision 2165]
  8777. Author: Henrik Gramner <[email protected]>
  8778. AuthorDate: Sun Feb 5 20:43:09 2012 +0100
  8779. Commit: Fiona Glaser <[email protected]>
  8780. CommitDate: Sun Feb 5 11:57:48 2012 -0800
  8781.  
  8782. Fix regression in r2141
  8783. Broke register preservation in x264_cpu_cpuid and x264_cpu_xgetbv.
  8784. Did not cause any problems.
  8785.  
  8786. commit ae289e6f03b76afa8736806e683349e8e59fcc93 [revision 2164]
  8787. Author: Fiona Glaser <[email protected]>
  8788. AuthorDate: Thu Jan 19 14:56:54 2012 -0800
  8789. Commit: Fiona Glaser <[email protected]>
  8790. CommitDate: Sat Feb 4 07:18:38 2012 -0800
  8791.  
  8792. TBM, AVX2, FMA3, BMI1, and BMI2 CPU detection support
  8793. TBM and BMI1 are supported by Trinity/Piledriver.
  8794. The others (and BMI1) will probably appear in Intel's upcoming Haswell.
  8795. Also update x86inc with AVX2 stuff.
  8796.  
  8797. commit e0581e0878c1995b215c51691af6bdf7a386946f [revision 2163]
  8798. Author: Loren Merritt <[email protected]>
  8799. AuthorDate: Fri Feb 3 06:27:18 2012 +0000
  8800. Commit: Fiona Glaser <[email protected]>
  8801. CommitDate: Sat Feb 4 07:18:36 2012 -0800
  8802.  
  8803. x86inc: add TAIL_CALL macro to abstract a common asm idiom
  8804.  
  8805. commit 04c38190c60658d544801718fc38fa3f745381d9 [revision 2162]
  8806. Author: Fiona Glaser <[email protected]>
  8807. AuthorDate: Wed Jan 25 16:44:38 2012 -0800
  8808. Commit: Fiona Glaser <[email protected]>
  8809. CommitDate: Sat Feb 4 07:18:34 2012 -0800
  8810.  
  8811. Minor asm optimizations/cleanup
  8812.  
  8813. commit 6d7c5efcf6f8751f768177bf828973a7bd4fdcf6 [revision 2161]
  8814. Author: Fiona Glaser <[email protected]>
  8815. AuthorDate: Tue Jan 24 19:03:58 2012 -0800
  8816. Commit: Fiona Glaser <[email protected]>
  8817. CommitDate: Sat Feb 4 07:18:31 2012 -0800
  8818.  
  8819. Clean up and optimize weightp, plus enable SSSE3 weight on SB/BDZ
  8820. Also remove unused AVX cruft.
  8821.  
  8822. commit 047175e610d3d5360f69e4f8168ff6fbafda2465 [revision 2160]
  8823. Author: Fiona Glaser <[email protected]>
  8824. AuthorDate: Mon Jan 23 18:57:58 2012 -0800
  8825. Commit: Fiona Glaser <[email protected]>
  8826. CommitDate: Sat Feb 4 07:18:29 2012 -0800
  8827.  
  8828. XOP frame_init_lowres
  8829. Covers both 8-bit and 16-bit, ~5-10% faster on Bulldozer.
  8830.  
  8831. commit abc88d60b5e0d803d6d4f0a5d9ece7dd0bdde0f1 [revision 2159]
  8832. Author: Fiona Glaser <[email protected]>
  8833. AuthorDate: Tue Jan 17 15:25:10 2012 -0800
  8834. Commit: Fiona Glaser <[email protected]>
  8835. CommitDate: Sat Feb 4 07:18:26 2012 -0800
  8836.  
  8837. XOP 8x8 zigzags
  8838. Field: 35(mmx) ->16(xop) cycles
  8839. Frame: 32(ssse3)->20(xop) cycles
  8840.  
  8841. commit aa47955a0ec65218e8bb967d36689069baca5fd1 [revision 2158]
  8842. Author: Fiona Glaser <[email protected]>
  8843. AuthorDate: Mon Jan 23 15:09:38 2012 -0800
  8844. Commit: Fiona Glaser <[email protected]>
  8845. CommitDate: Sat Feb 4 07:18:24 2012 -0800
  8846.  
  8847. AVX 32-bit hpel_filter_h
  8848. Faster on Sandy Bridge.
  8849. Also add details on unsuccessful optimizations in these functions.
  8850.  
  8851. commit d7407cf81816fff7ab32ceb2398575724e8cc737 [revision 2157]
  8852. Author: Fiona Glaser <[email protected]>
  8853. AuthorDate: Fri Jan 27 16:29:30 2012 -0800
  8854. Commit: Fiona Glaser <[email protected]>
  8855. CommitDate: Sat Feb 4 07:18:21 2012 -0800
  8856.  
  8857. x86inc: add high halfword register support
  8858. Might be useful in a few cases.
  8859.  
  8860. commit acabceb6530d1858bcd009b055e217c75344c442 [revision 2156]
  8861. Author: Ronald S. Bultje <[email protected]>
  8862. AuthorDate: Wed Jan 25 13:53:59 2012 +0800
  8863. Commit: Fiona Glaser <[email protected]>
  8864. CommitDate: Sat Feb 4 07:18:19 2012 -0800
  8865.  
  8866. Change %ifdef directives to %if directives in *.asm files
  8867. This allows combining multiple conditionals in a single statement.
  8868.  
  8869. commit 82d8cdde567b1c1e8d2046bbb831d0daafe8213b [revision 2155]
  8870. Author: Anton Mitrofanov <[email protected]>
  8871. AuthorDate: Sun Jan 22 22:13:52 2012 +0400
  8872. Commit: Fiona Glaser <[email protected]>
  8873. CommitDate: Sat Feb 4 07:18:16 2012 -0800
  8874.  
  8875. Use TV range algorithm for bit-depth conversions
  8876. Such sources are more common, so better to be correct for the common case.
  8877. This also produces less error for the case of full range than the previous algorithm produced for the case of TV range.
  8878.  
  8879. commit 27a7b05b8330d0756e5e3f6669282561030f54fa [revision 2154]
  8880. Author: Hii <[email protected]>
  8881. AuthorDate: Wed Jan 25 16:29:22 2012 +0800
  8882. Commit: Fiona Glaser <[email protected]>
  8883. CommitDate: Sat Feb 4 07:18:13 2012 -0800
  8884.  
  8885. Bump dates to 2012
  8886.  
  8887. commit 762f677e095a40e1927086bb08799c01e05c2ee4 [revision 2153]
  8888. Author: Henrik Gramner <[email protected]>
  8889. AuthorDate: Sat Jan 28 21:38:27 2012 +0100
  8890. Commit: Fiona Glaser <[email protected]>
  8891. CommitDate: Sat Feb 4 07:18:10 2012 -0800
  8892.  
  8893. Add Windows resource file
  8894. Displays version info in Windows Explorer.
  8895.  
  8896. commit 545b41caa5903ebcb5d9336a59f9bf5a50a45037 [revision 2152]
  8897. Author: Sergey Radionov <[email protected]>
  8898. AuthorDate: Mon Jan 16 13:22:44 2012 -0800
  8899. Commit: Fiona Glaser <[email protected]>
  8900. CommitDate: Sat Feb 4 07:18:02 2012 -0800
  8901.  
  8902. Fix win32 pthread_cond_signal
  8903. Isn't used by x264 currently, so didn't cause a problem.
  8904. Fix backported from libav.
  8905.  
  8906. commit 697a11e8ecb1376cddd4a8d4f4fa693e41c1987e [revision 2151]
  8907. Author: Mans Rullgard <[email protected]>
  8908. AuthorDate: Wed Feb 1 15:55:25 2012 -0800
  8909. Commit: Fiona Glaser <[email protected]>
  8910. CommitDate: Sat Feb 4 07:16:31 2012 -0800
  8911.  
  8912. ARM: align asm functions to 4 bytes.
  8913. Some linkers apparently fail to correctly align ARM functions when mixing with Thumb code.
  8914.  
  8915. commit f59b310fd87b643b59d6e109e49fdf9d0a04ce91 [revision 2150]
  8916. Author: Anton Mitrofanov <[email protected]>
  8917. AuthorDate: Sun Jan 22 13:00:23 2012 +0400
  8918. Commit: Fiona Glaser <[email protected]>
  8919. CommitDate: Sat Feb 4 07:16:28 2012 -0800
  8920.  
  8921. Fix normalization of colorspace when input is packed YUV 4:2:2
  8922.  
  8923. commit 9fb055856a617f5ddca15a0c5745ff1c1486ad9a [revision 2149]
  8924. Author: Fiona Glaser <[email protected]>
  8925. AuthorDate: Sat Jan 21 12:54:40 2012 -0800
  8926. Commit: Fiona Glaser <[email protected]>
  8927. CommitDate: Sat Feb 4 07:16:25 2012 -0800
  8928.  
  8929. Force keyint-min 1 with Blu-ray
  8930. Fixes an issue with referencing across I-frames that's prohibited in Blu-ray for some godforsaken reason.
  8931.  
  8932. commit 77cfcb6acf648da00eb4ddb52bcb7006bc64a61a [revision 2148]
  8933. Author: Oka Motofumi <[email protected]>
  8934. AuthorDate: Sun Jan 29 20:34:41 2012 +0900
  8935. Commit: Fiona Glaser <[email protected]>
  8936. CommitDate: Sat Feb 4 07:16:23 2012 -0800
  8937.  
  8938. Fix crash in --demuxer y4m with unsupported colorspace
  8939.  
  8940. commit 30829c0c7e6bbf40d1b3ed5fcb5a45d85407978f [revision 2147]
  8941. Author: Anton Mitrofanov <[email protected]>
  8942. AuthorDate: Mon Jan 16 14:02:53 2012 -0800
  8943. Commit: Fiona Glaser <[email protected]>
  8944. CommitDate: Sat Feb 4 07:16:19 2012 -0800
  8945.  
  8946. Fix overread/possible crash with intra refresh + VBV
  8947.  
  8948. commit 26c8303472b837e301d789ba569eae01955cf7f6 [revision 2146]
  8949. Author: Loren Merritt <[email protected]>
  8950. AuthorDate: Wed Jan 18 15:47:07 2012 -0800
  8951. Commit: Fiona Glaser <[email protected]>
  8952. CommitDate: Wed Jan 18 15:48:34 2012 -0800
  8953.  
  8954. Fix trellis 2 + subme >= 8
  8955. Trellis didn't return a boolean value as it was supposed to.
  8956. Regression in r2143-5.
  8957.  
  8958. commit 7d804baf3bca6ad33e18ccd0a838274214a8a7a0 [revision 2145]
  8959. Author: Loren Merritt <[email protected]>
  8960. AuthorDate: Fri Jan 6 15:53:29 2012 +0000
  8961. Commit: Fiona Glaser <[email protected]>
  8962. CommitDate: Sun Jan 15 14:27:37 2012 -0800
  8963.  
  8964. CABAC trellis opts part 4: x86_64 asm
  8965. Another 20% faster.
  8966. 18k->12k codesize.
  8967.  
  8968. This patch series may have a large impact on encoding speed.
  8969. For example, 24% faster at --preset slower --crf 23 with 720p parkjoy.
  8970. Overall speed increase is proportional to the cost of trellis (which is proportional to bitrate, and much more with --trellis 2).
  8971.  
  8972. commit dd354db4db2f26e63ed36eb790052c6794e5a684 [revision 2144]
  8973. Author: Loren Merritt <[email protected]>
  8974. AuthorDate: Fri Jan 6 15:53:04 2012 +0000
  8975. Commit: Fiona Glaser <[email protected]>
  8976. CommitDate: Sun Jan 15 14:27:34 2012 -0800
  8977.  
  8978. CABAC trellis opts part 3: make some arrays non-static
  8979.  
  8980. commit 4abcf60a04e358b87da284f3a5fac3e2949b6de1 [revision 2143]
  8981. Author: Loren Merritt <[email protected]>
  8982. AuthorDate: Thu Dec 22 17:56:06 2011 +0000
  8983. Commit: Fiona Glaser <[email protected]>
  8984. CommitDate: Sun Jan 15 14:27:32 2012 -0800
  8985.  
  8986. CABAC trellis opts part 2: C optimizations
  8987.  
  8988. Hoist the branch on coef value out of the loop over node contexts.
  8989. Special cases for each possible coef value (0,1,n).
  8990. Special case for dc-only blocks.
  8991. Template the main loop for two common subsets of nodes, to avoid a bunch of branches about which nodes are live.
  8992. Use the nonupdating version of cabac_size_decision in more cases, and omit those bins from the node struct.
  8993. CABAC offsets are now compile-time constants.
  8994. Change TRELLIS_SCORE_MAX from a specific constant to anything negative, which is cheaper to test.
  8995. Remove dct_weight2_zigzag[], since trellis has to lookup zigzag[] anyway.
  8996.  
  8997. 60% faster on x86_64.
  8998. 25k->18k codesize.
  8999.  
  9000. commit 253cd7baefb7f5d101725034b2c37afacc012305 [revision 2142]
  9001. Author: Loren Merritt <[email protected]>
  9002. AuthorDate: Thu Dec 22 17:55:06 2011 +0000
  9003. Commit: Fiona Glaser <[email protected]>
  9004. CommitDate: Sun Jan 15 14:27:29 2012 -0800
  9005.  
  9006. CABAC trellis opts part 1: minor change in output
  9007. Due to different tie-break order.
  9008.  
  9009. commit 0d7a9100d12c618acea3f01b8bb9cc306f475b47 [revision 2141]
  9010. Author: Henrik Gramner <[email protected]>
  9011. AuthorDate: Sun Jan 8 04:14:10 2012 +0100
  9012. Commit: Fiona Glaser <[email protected]>
  9013. CommitDate: Sun Jan 15 14:27:13 2012 -0800
  9014.  
  9015. x86inc improvements for 64-bit
  9016.  
  9017. Add support for all x86-64 registers
  9018. Prefer caller-saved register over callee-saved on WIN64
  9019. Support up to 15 function arguments
  9020.  
  9021. commit 8a6a062e11d4074c081d076408cb0bd6def1af8e [revision 2140]
  9022. Author: Ilia Valiakhmetov <[email protected]>
  9023. AuthorDate: Sun Jan 15 04:47:58 2012 -0600
  9024. Commit: Fiona Glaser <[email protected]>
  9025. CommitDate: Sun Jan 15 13:08:03 2012 -0800
  9026.  
  9027. High bit depth SSE2/AVX add8x8_idct8 and add16x16_idct8
  9028. From Google Code-In.
  9029.  
  9030. commit a35fd4194dd7004abe6f66679496beded405515a [revision 2139]
  9031. Author: Edward Wang <[email protected]>
  9032. AuthorDate: Wed Jan 4 15:35:54 2012 -0800
  9033. Commit: Fiona Glaser <[email protected]>
  9034. CommitDate: Sun Jan 15 13:04:39 2012 -0800
  9035.  
  9036. MMX/SSE2/AVX predict_8x16_p, high bit depth fdct8
  9037. From Google Code-In.
  9038.  
  9039. commit 9301bbd39fb0a49b1e986f9a7c29685439686de4 [revision 2138]
  9040. Author: Fiona Glaser <[email protected]>
  9041. AuthorDate: Thu Dec 22 14:03:15 2011 -0800
  9042. Commit: Fiona Glaser <[email protected]>
  9043. CommitDate: Sun Jan 15 13:04:36 2012 -0800
  9044.  
  9045. XOP 8-bit fDCT
  9046. Use integer MAC for one of the SUMSUB passes. About a dozen cycles faster for 16x16.
  9047.  
  9048. commit c83edc0427e78c58683af99b80e0234c77b3e41a [revision 2137]
  9049. Author: Cristian Militaru <[email protected]>
  9050. AuthorDate: Wed Jan 4 12:38:08 2012 -0800
  9051. Commit: Fiona Glaser <[email protected]>
  9052. CommitDate: Sun Jan 15 13:04:33 2012 -0800
  9053.  
  9054. High bit depth intra_sad_x3_4x4
  9055. From Google Code-In.
  9056.  
  9057. commit 9c0fa2d63f549a44f869562cffa9c041a32ae41d [revision 2136]
  9058. Author: Fiona Glaser <[email protected]>
  9059. AuthorDate: Thu Dec 8 13:45:41 2011 -0800
  9060. Commit: Fiona Glaser <[email protected]>
  9061. CommitDate: Sun Jan 15 13:04:30 2012 -0800
  9062.  
  9063. Use a large LUT for CAVLC zero-run bit codes
  9064. Helps the most with trellis and RD, but also helps with bitstream writing.
  9065. Seems at worst neutral even in the extreme case of a CPU with small L2 cache (e.g. ARM Cortex A8).
  9066.  
  9067. commit de7aed78cd2f70017f3c479d8f8dc32d52cee607 [revision 2135]
  9068. Author: Matt Habel <[email protected]>
  9069. AuthorDate: Fri Dec 16 23:16:09 2011 -0800
  9070. Commit: Fiona Glaser <[email protected]>
  9071. CommitDate: Sun Jan 15 13:04:28 2012 -0800
  9072.  
  9073. High bit depth intra_sad_x3_8x8, intra_satd_x3_4x4/8x8c/16x16
  9074. Also add an ACCUM macro to handle accumulator-induced add-or-swap more concisely.
  9075.  
  9076. commit d9dee734a9af1788461def43321f19be6a3d2d72 [revision 2134]
  9077. Author: Shitiz Garg <[email protected]>
  9078. AuthorDate: Sat Dec 3 15:34:57 2011 -0800
  9079. Commit: Fiona Glaser <[email protected]>
  9080. CommitDate: Sun Jan 15 13:04:25 2012 -0800
  9081.  
  9082. MMX 10-bit predict_8x8c_h and predict_8x16c_h
  9083. From Google Code-In.
  9084.  
  9085. commit 7496fc4aeaaaf5b470b1eb0f73ce8ea71d0116f2 [revision 2133]
  9086. Author: Aaron Schmitz <[email protected]>
  9087. AuthorDate: Wed Nov 30 00:15:45 2011 -0600
  9088. Commit: Fiona Glaser <[email protected]>
  9089. CommitDate: Sun Jan 15 13:04:23 2012 -0800
  9090.  
  9091. Some MBAFF x86 assembly functions.
  9092. deblock_chroma_420_mbaff, plus 422/422_intra_mbaff implemented using existing functions.
  9093. From Google Code-In.
  9094.  
  9095. commit b8d7b8acb48b45afbfd7efb5baac79475682684a [revision 2132]
  9096. Author: George Stephanos <[email protected]>
  9097. AuthorDate: Thu Dec 1 16:53:45 2011 -0800
  9098. Commit: Fiona Glaser <[email protected]>
  9099. CommitDate: Sun Jan 15 13:04:20 2012 -0800
  9100.  
  9101. More ARM NEON assembly functions
  9102. predict_8x8_v, predict_4x4_dc_top, predict_8x8_ddl, predict_8x8_ddr, predict_8x8_vl, predict_8x8_vr, predict_8x8_hd, predict_8x8_hu.
  9103. From Google Code-In.
  9104.  
  9105. commit e269ca55e5244280afd0347c1088083cf7043d48 [revision 2131]
  9106. Author: Ilia <[email protected]>
  9107. AuthorDate: Mon Nov 28 05:20:09 2011 -0800
  9108. Commit: Fiona Glaser <[email protected]>
  9109. CommitDate: Sun Jan 15 13:04:15 2012 -0800
  9110.  
  9111. More 4:2:2 asm functions
  9112. High bit depth version of deblock_h_chroma_422.
  9113. Regular and high bit depth versions of deblock_h_chroma_intra_422.
  9114. High bit depth pixel_vsad.
  9115. SSE2 high bit depth and MMX 8-bit predict_8x8_vl.
  9116. Our first GCI patch this year!
  9117.  
  9118. commit 5d66c5011488539f99ceafdb47b0856a8e9dae0b [revision 2130]
  9119. Author: Henrik Gramner <[email protected]>
  9120. AuthorDate: Thu Dec 8 16:14:35 2011 +0100
  9121. Commit: Fiona Glaser <[email protected]>
  9122. CommitDate: Sun Jan 15 13:04:12 2012 -0800
  9123.  
  9124. SSE2 and SSSE3 versions of sub8x16_dct_dc
  9125. Also slightly faster sub8x8_dct_dc
  9126.  
  9127. commit 3ea6a8b22e0aa89e3749e9c95edfeaad9d341b7e [revision 2129]
  9128. Author: Steven Walters <[email protected]>
  9129. AuthorDate: Mon Dec 5 08:46:34 2011 -0500
  9130. Commit: Fiona Glaser <[email protected]>
  9131. CommitDate: Sun Jan 15 13:03:54 2012 -0800
  9132.  
  9133. Resize filter updates
  9134. Use AVPixFmtDescriptors to pick the most compatible x264 csp for any pixel format.
  9135. Fix deprecated use of av_set_int.
  9136. Now requires libavutil >= 51.19.0
  9137.  
  9138. commit f71d047d0bc129eb9f4724e023bf888a9124338b [revision 2128]
  9139. Author: Oka Motofumi <[email protected]>
  9140. AuthorDate: Thu Jan 5 14:23:50 2012 -0800
  9141. Commit: Fiona Glaser <[email protected]>
  9142. CommitDate: Thu Jan 12 12:07:56 2012 -0800
  9143.  
  9144. Add out-of-tree build support
  9145.  
  9146. commit 5539220e5afc641a6747c6d95f41e5efbe5858e1 [revision 2127]
  9147. Author: Anton Mitrofanov <[email protected]>
  9148. AuthorDate: Fri Dec 16 18:17:00 2011 +0400
  9149. Commit: Fiona Glaser <[email protected]>
  9150. CommitDate: Thu Jan 12 12:07:53 2012 -0800
  9151.  
  9152. Limit SSIM to 100db
  9153. Avoids floating point error for infinite SSIM (lossless).
  9154.  
  9155. commit 13c236172f0ff40ca149a2e862498457cd32ccb9 [revision 2126]
  9156. Author: Reynaldo H. Verdejo Pinochet <[email protected]>
  9157. AuthorDate: Wed Jan 4 13:16:12 2012 -0300
  9158. Commit: Fiona Glaser <[email protected]>
  9159. CommitDate: Thu Jan 12 12:04:45 2012 -0800
  9160.  
  9161. Fix wrong conditional inclusion of inttypes.h
  9162. inttypes.h is required by encoder/ratecontrol.c for SCNxxx macros, and HAVE_STDINT_H does not imply having inttypes.h.
  9163. stdint.h is a subset of inttypes.h, but this isn't enough for x264.
  9164. This change fixes building x264 with Android's toolchain.
  9165.  
  9166. commit 2df9d45db64110854e6da6a2037d6c432c5463fe [revision 2125]
  9167. Author: Anton Mitrofanov <[email protected]>
  9168. AuthorDate: Wed Dec 21 11:08:56 2011 +0400
  9169. Commit: Fiona Glaser <[email protected]>
  9170. CommitDate: Thu Jan 12 12:04:21 2012 -0800
  9171.  
  9172. Fix crash with sliced threads and input height <= 112
  9173.  
  9174. commit e3d311813f3931133962f7ab8ee2305d231df83d [revision 2124]
  9175. Author: Phillip Blucas <[email protected]>
  9176. AuthorDate: Mon Dec 19 17:43:41 2011 -0600
  9177. Commit: Fiona Glaser <[email protected]>
  9178. CommitDate: Mon Dec 19 16:01:51 2011 -0800
  9179.  
  9180. Fix loading custom 8x8 chroma quant matrices in 4:4:4
  9181.  
  9182. commit 9fd7ccb2b635276d019e137844c693b525f92244 [revision 2123]
  9183. Author: Anton Mitrofanov <[email protected]>
  9184. AuthorDate: Fri Dec 16 01:48:07 2011 +0400
  9185. Commit: Fiona Glaser <[email protected]>
  9186. CommitDate: Mon Dec 19 13:25:57 2011 -0800
  9187.  
  9188. Fix PCM cost overflow
  9189.  
  9190. commit 1d70d0e56003b762439ad4b5d8e72729b51516ae [revision 2122]
  9191. Author: Anton Mitrofanov <[email protected]>
  9192. AuthorDate: Fri Dec 9 01:54:22 2011 +0400
  9193. Commit: Fiona Glaser <[email protected]>
  9194. CommitDate: Thu Dec 8 14:00:43 2011 -0800
  9195.  
  9196. Fix overflow in 8-bit x86 vsad asm function
  9197.  
  9198. commit b6ce6c64c17071804676435da9b1c07b902857e3 [revision 2121]
  9199. Author: Anton Mitrofanov <[email protected]>
  9200. AuthorDate: Wed Dec 7 19:14:52 2011 +0400
  9201. Commit: Fiona Glaser <[email protected]>
  9202. CommitDate: Wed Dec 7 16:37:50 2011 -0800
  9203.  
  9204. Fix crash in --fullhelp when compiled against recent ffmpeg
  9205. Don't assume all pixel formats have a description.
  9206.  
  9207. commit c3ba63bbe83bd20d06a64cfecd6b878e8f49bc13 [revision 2120]
  9208. Author: Fiona Glaser <[email protected]>
  9209. AuthorDate: Tue Dec 6 14:39:21 2011 -0800
  9210. Commit: Fiona Glaser <[email protected]>
  9211. CommitDate: Tue Dec 6 14:44:46 2011 -0800
  9212.  
  9213. Fix regression in r2118
  9214. Broke trellis with i16x16 macroblocks.
  9215.  
  9216. commit 9dc2391576b35acb55c04773049a0b817f306969 [revision 2119]
  9217. Author: Fiona Glaser <[email protected]>
  9218. AuthorDate: Wed Nov 30 13:02:12 2011 -0800
  9219. Commit: Fiona Glaser <[email protected]>
  9220. CommitDate: Thu Dec 1 12:27:53 2011 -0800
  9221.  
  9222. Modify MBAFF chroma deblock functions to handle U/V at the same time
  9223. Allows for more convenient asm implementations.
  9224.  
  9225. commit d0bf649fcc1a79da12e220c4364aeca6045dfbed [revision 2118]
  9226. Author: Fiona Glaser <[email protected]>
  9227. AuthorDate: Thu Nov 10 16:16:13 2011 -0800
  9228. Commit: Fiona Glaser <[email protected]>
  9229. CommitDate: Thu Dec 1 12:27:50 2011 -0800
  9230.  
  9231. CABAC trellis optimizations: use SIMD quant
  9232. Significant speed increase, minor change in output due to rounding.
  9233.  
  9234. commit 6767f967831048669e45e65681f37011483b4fa0 [revision 2117]
  9235. Author: Steven Walters <[email protected]>
  9236. AuthorDate: Sun Nov 6 09:48:30 2011 -0800
  9237. Commit: Fiona Glaser <[email protected]>
  9238. CommitDate: Thu Dec 1 12:27:47 2011 -0800
  9239.  
  9240. YUV range detection and support for x264CLI
  9241. Two new options: --input-range and --range.
  9242. --input-range forces the range of the input in case of misdetection; auto by default.
  9243. -- range sets the range of the output; x264cli will convert if necessary, TV by default.
  9244. --fullrange is now removed as a CLI option (but the libx264 API is unchanged).
  9245.  
  9246. commit f9a4c4d9828c1cc60135d0301981ea71fd90f6ca [revision 2116]
  9247. Author: Kieran Kunhya <[email protected]>
  9248. AuthorDate: Fri Nov 4 20:09:13 2011 +0000
  9249. Commit: Fiona Glaser <[email protected]>
  9250. CommitDate: Thu Dec 1 12:27:45 2011 -0800
  9251.  
  9252. Pass through user data
  9253.  
  9254. commit 1c774e936a315fdfb92a35c402b351a1c542a13a [revision 2115]
  9255. Author: Fiona Glaser <[email protected]>
  9256. AuthorDate: Thu Oct 27 14:05:56 2011 -0700
  9257. Commit: Fiona Glaser <[email protected]>
  9258. CommitDate: Thu Dec 1 12:27:43 2011 -0800
  9259.  
  9260. Remove unpredictable branch in CABAC dqp
  9261.  
  9262. commit f3a7517cb9b06a580623cbea0f140be534b99877 [revision 2114]
  9263. Author: Loren Merritt <[email protected]>
  9264. AuthorDate: Sun Oct 23 23:15:11 2011 +0000
  9265. Commit: Fiona Glaser <[email protected]>
  9266. CommitDate: Thu Dec 1 12:27:40 2011 -0800
  9267.  
  9268. x86inc: AVX symmetry optimization
  9269. 3-arg AVX ops with a memory arg can only have it in src2,
  9270. whereas SSE emulation of 3-arg prefers to have it in src1 (i.e. the move).
  9271. So, if the op is symmetric and the wrong one is memory, swap them.
  9272. Eliminates redundant moves in some cases when using 3-operand without AVX with memory arguments.
  9273. Also fix movss and movsd in some cases, and flag shufps correctly as float.
  9274.  
  9275. commit 5ebbcd8748ae8d8b184db5a8f9b46a9ad865f0ae [revision 2113]
  9276. Author: Anton Mitrofanov <[email protected]>
  9277. AuthorDate: Tue Nov 29 13:45:13 2011 -0800
  9278. Commit: Fiona Glaser <[email protected]>
  9279. CommitDate: Thu Dec 1 12:27:33 2011 -0800
  9280.  
  9281. checkasm: shut up gcc warnings, fix some naming of functions in results
  9282.  
  9283. commit 561f71ebf741370075b970fb9d31a593cf47782f [revision 2112]
  9284. Author: Mans Rullgard <[email protected]>
  9285. AuthorDate: Mon Nov 28 16:29:12 2011 -0800
  9286. Commit: Fiona Glaser <[email protected]>
  9287. CommitDate: Wed Nov 30 00:29:21 2011 -0800
  9288.  
  9289. checkasm: fix build on ARM
  9290. Because of how ALIGNED_ARRAY_16 is defined on ARM, array initialisers cannot be used here. Use memset() instead.
  9291.  
  9292. commit 24bf90abde21e77c574f2bd43e38a3222c3183ef [revision 2111]
  9293. Author: Anton Mitrofanov <[email protected]>
  9294. AuthorDate: Sat Nov 12 01:31:49 2011 +0400
  9295. Commit: Fiona Glaser <[email protected]>
  9296. CommitDate: Mon Nov 28 15:02:28 2011 -0800
  9297.  
  9298. Improve makefile rules
  9299. Remove the need for "make clean" after most reconfigures.
  9300.  
  9301. commit 87b23e25eee0c04bb47957445e7cf941a7d8b980 [revision 2110]
  9302. Author: Anton Mitrofanov <[email protected]>
  9303. AuthorDate: Sat Nov 12 00:47:48 2011 +0400
  9304. Commit: Fiona Glaser <[email protected]>
  9305. CommitDate: Fri Nov 11 14:40:15 2011 -0800
  9306.  
  9307. Mark some local functions as static, cosmetics
  9308.  
  9309. commit 2ecbcd73d60d2f749696b39627c91e28a396538b [revision 2109]
  9310. Author: Anton Mitrofanov <[email protected]>
  9311. AuthorDate: Fri Nov 11 23:19:02 2011 +0400
  9312. Commit: Fiona Glaser <[email protected]>
  9313. CommitDate: Fri Nov 11 14:40:12 2011 -0800
  9314.  
  9315. Fix crash if timecode file opening fails
  9316.  
  9317. commit f1387840b98560ae34aea9ca09d55984812ad50b [revision 2108]
  9318. Author: Fabian Greffrath <[email protected]>
  9319. AuthorDate: Fri Nov 11 13:25:43 2011 -0800
  9320. Commit: Fiona Glaser <[email protected]>
  9321. CommitDate: Fri Nov 11 13:41:26 2011 -0800
  9322.  
  9323. Configure: force PIC for shared build on PARISC and MIPS
  9324.  
  9325. commit e5063ab30bcb79f94774b6d9ce91b098ade01d6d [revision 2107]
  9326. Author: Anton Mitrofanov <[email protected]>
  9327. AuthorDate: Sat Oct 22 19:41:07 2011 +0400
  9328. Commit: Fiona Glaser <[email protected]>
  9329. CommitDate: Sun Oct 23 14:51:33 2011 -0700
  9330.  
  9331. Improve yasm version check
  9332. Previous check allowed certain earlier versions that weren't fully compatible.
  9333.  
  9334. commit 12104b22820b38b4976e83a6ee00dcb59ed959f1 [revision 2106]
  9335. Author: Fiona Glaser <[email protected]>
  9336. AuthorDate: Tue Oct 18 14:30:26 2011 -0700
  9337. Commit: Fiona Glaser <[email protected]>
  9338. CommitDate: Fri Oct 21 17:22:59 2011 -0700
  9339.  
  9340. Add fenc prefetching to adaptive quant
  9341. Many fewer cache misses, faster adaptive quant.
  9342.  
  9343. commit 9bbfc30284469a70374a75fecfa322c4740dc2b7 [revision 2105]
  9344. Author: Fiona Glaser <[email protected]>
  9345. AuthorDate: Tue Oct 18 14:14:03 2011 -0700
  9346. Commit: Fiona Glaser <[email protected]>
  9347. CommitDate: Fri Oct 21 17:22:56 2011 -0700
  9348.  
  9349. Split prefetch_fenc between colorspaces
  9350. Add 4:2:2 version.
  9351.  
  9352. commit b63a73da3add660358a4bad1a590c2d4ed466dc4 [revision 2104]
  9353. Author: Fiona Glaser <[email protected]>
  9354. AuthorDate: Tue Oct 11 17:04:32 2011 -0700
  9355. Commit: Fiona Glaser <[email protected]>
  9356. CommitDate: Fri Oct 21 17:22:52 2011 -0700
  9357.  
  9358. Some more 4:2:2 x86 asm
  9359. coeff_last8, coeff_level_run8, var2_8x16, predict_8x16c_dc, satd_4x16, intra_mbcmp_8x16c_x3, deblock_h_chroma_422
  9360.  
  9361. commit 50aaf8d84ac6fc78794b98cfe6a25440a09fbb82 [revision 2103]
  9362. Author: Loren Merritt <[email protected]>
  9363. AuthorDate: Tue Oct 11 18:12:43 2011 +0000
  9364. Commit: Fiona Glaser <[email protected]>
  9365. CommitDate: Fri Oct 21 17:22:49 2011 -0700
  9366.  
  9367. Remove obsolete versions of intra_mbcmp_x3
  9368. intra_mbcmp_x3 is unnecessary if x9 exists (SSSE3 and onwards).
  9369.  
  9370. commit 1111780d8e392455870898bacae30a413ae98464 [revision 2102]
  9371. Author: Loren Merritt <[email protected]>
  9372. AuthorDate: Mon Oct 10 05:42:36 2011 +0000
  9373. Commit: Fiona Glaser <[email protected]>
  9374. CommitDate: Fri Oct 21 17:22:45 2011 -0700
  9375.  
  9376. SSSE3/SSE4/AVX 9-way fully merged i8x8 analysis (sa8d_x9)
  9377. x86_64 only for now, due to register requirements (like sa8d_x3).
  9378.  
  9379. i8x8 analysis cycles (per partition):
  9380. penryn sandybridge bulldozer
  9381. 616->600 482->374 418->356 preset=faster
  9382. 892->632 725->387 598->373 preset=medium
  9383. 948->650 789->409 673->383 preset=slower
  9384.  
  9385. commit 422979198e492d5068034a3a5b1e4991af2b63a1 [revision 2101]
  9386. Author: Fiona Glaser <[email protected]>
  9387. AuthorDate: Fri Sep 30 19:09:19 2011 -0700
  9388. Commit: Fiona Glaser <[email protected]>
  9389. CommitDate: Fri Oct 21 17:22:41 2011 -0700
  9390.  
  9391. SSSE3/SSE4/AVX 9-way fully merged i8x8 analysis (sad_x9)
  9392. ~3 times faster than current analysis, plus (like intra_sad_x9_4x4) analyzes all modes without shortcuts.
  9393.  
  9394. commit da66eef02e8d9cb57c52aeecb7371b9968747c2b [revision 2100]
  9395. Author: Loren Merritt <[email protected]>
  9396. AuthorDate: Wed Oct 5 13:29:21 2011 -0700
  9397. Commit: Fiona Glaser <[email protected]>
  9398. CommitDate: Fri Oct 21 17:22:38 2011 -0700
  9399.  
  9400. Merge i4x4 prediction with intra_mbcmp_x9_4x4
  9401. Avoids a redundant prediction after analysis.
  9402.  
  9403. commit 9f027f4f3f9b03b5dabe081a12ca1b80c20ffc18 [revision 2099]
  9404. Author: Fiona Glaser <[email protected]>
  9405. AuthorDate: Wed Oct 5 13:17:31 2011 -0700
  9406. Commit: Fiona Glaser <[email protected]>
  9407. CommitDate: Fri Oct 21 17:22:34 2011 -0700
  9408.  
  9409. Inline i4x4/i8x8 encode into intra analysis
  9410. Larger code size, but faster.
  9411.  
  9412. commit a5a6d0eeadbba6ae3232f620345762aebca240ab [revision 2098]
  9413. Author: Fiona Glaser <[email protected]>
  9414. AuthorDate: Wed Sep 21 17:12:10 2011 -0700
  9415. Commit: Fiona Glaser <[email protected]>
  9416. CommitDate: Fri Oct 21 17:22:31 2011 -0700
  9417.  
  9418. Initial XOP and FMA4 support on AMD Bulldozer
  9419. ~10% faster Hadamard functions (SATD/SA8D/hadamard_ac) plus other improvements.
  9420.  
  9421. commit e73b85b56437827f881d1406e11d2cca4bbe5583 [revision 2097]
  9422. Author: Mans Rullgard <[email protected]>
  9423. AuthorDate: Tue Sep 27 21:14:14 2011 +0400
  9424. Commit: Fiona Glaser <[email protected]>
  9425. CommitDate: Fri Oct 21 17:22:28 2011 -0700
  9426.  
  9427. ARM: update NEON chroma deblock functions to NV12 pixel format
  9428.  
  9429. commit 9c356e2558948714bdbb991a9f9cb9a3e1f0121b [revision 2096]
  9430. Author: Sean McGovern <[email protected]>
  9431. AuthorDate: Mon Oct 17 12:45:15 2011 -0700
  9432. Commit: Fiona Glaser <[email protected]>
  9433. CommitDate: Fri Oct 21 17:22:24 2011 -0700
  9434.  
  9435. Add /usr/lib/{64/}values-xpg6.o to $LDFLAGS on Solaris
  9436. This is required for POSIX.1-2001 compliance.
  9437.  
  9438. commit 6c50ab569d95ebb07e5fb437a38d646bf607c74b [revision 2095]
  9439. Author: Sean McGovern <[email protected]>
  9440. AuthorDate: Mon Oct 17 12:44:03 2011 -0700
  9441. Commit: Fiona Glaser <[email protected]>
  9442. CommitDate: Fri Oct 21 17:22:21 2011 -0700
  9443.  
  9444. Fix linker test for -Bsymbolic
  9445. The Solaris linker only accepts -Bsymbolic for objects compiled in dynamic mode (i.e. shared objects), so pass -shared to gcc.
  9446. Additionally, for x86_32 unresolved textrels cause a linker error so mark the .text section as 'impure'.
  9447.  
  9448. commit 421c38f22c7bdaf2981b2ffb72332c40cadd7332 [revision 2094]
  9449. Author: Sean McGovern <[email protected]>
  9450. AuthorDate: Mon Oct 17 12:43:28 2011 -0700
  9451. Commit: Fiona Glaser <[email protected]>
  9452. CommitDate: Fri Oct 21 17:22:18 2011 -0700
  9453.  
  9454. Add $SOFLAGS to exported SOFLAGS make variable
  9455.  
  9456. commit dd713cae59c062440b046fe75d60af83d049de3c [revision 2093]
  9457. Author: Henrik Gramner <[email protected]>
  9458. AuthorDate: Sat Sep 24 15:56:08 2011 +0200
  9459. Commit: Fiona Glaser <[email protected]>
  9460. CommitDate: Fri Oct 21 17:22:14 2011 -0700
  9461.  
  9462. Allow setting a chroma format at compile time
  9463. Gives a slight speed increase and significant binary size reduction when only one chroma format is needed.
  9464.  
  9465. commit 68f6db44035e8f9d4d00a73e5703eb1d7ff8d619 [revision 2092]
  9466. Author: Harfe Leier <[email protected]>
  9467. AuthorDate: Fri Sep 30 12:49:33 2011 -0700
  9468. Commit: Fiona Glaser <[email protected]>
  9469. CommitDate: Fri Oct 21 17:22:11 2011 -0700
  9470.  
  9471. Improve profile help
  9472. List high422/high444 profiles, and don't show non-high-bit-depth profiles in high bit depth builds.
  9473.  
  9474. commit 675110a687459cc03685489470bbc730580a793b [revision 2091]
  9475. Author: Yusuke Nakamura <[email protected]>
  9476. AuthorDate: Thu Oct 20 03:09:51 2011 +0900
  9477. Commit: Fiona Glaser <[email protected]>
  9478. CommitDate: Fri Oct 21 17:22:01 2011 -0700
  9479.  
  9480. Fix infinite loop parsing TDecimate Mode 3 timecode v1 files
  9481.  
  9482. commit 2ec99b3b94f986b456de1525087ee85b6fa79091 [revision 2090]
  9483. Author: Fiona Glaser <[email protected]>
  9484. AuthorDate: Mon Oct 10 17:44:31 2011 -0700
  9485. Commit: Fiona Glaser <[email protected]>
  9486. CommitDate: Fri Oct 14 19:03:23 2011 -0700
  9487.  
  9488. Fix some integer overflows/signedness errors found by IOC
  9489. The only real bug here is in slicetype.c, which may or may not affect real encodes.
  9490.  
  9491. commit ae1288c43780ed9be60b59dd556d5f85df7498e2 [revision 2089]
  9492. Author: Fiona Glaser <[email protected]>
  9493. AuthorDate: Wed Oct 12 09:16:32 2011 -0700
  9494. Commit: Fiona Glaser <[email protected]>
  9495. CommitDate: Wed Oct 12 09:17:58 2011 -0700
  9496.  
  9497. Fix pixel_var2 with 4:2:2 encoding
  9498. Might have caused artifacts or suboptimal chroma compression.
  9499.  
  9500. commit 9ac39f6078659f4f5cf548460dec940a04fd52c8 [revision 2088]
  9501. Author: Anton Mitrofanov <[email protected]>
  9502. AuthorDate: Sun Oct 9 19:14:16 2011 +0400
  9503. Commit: Fiona Glaser <[email protected]>
  9504. CommitDate: Sun Oct 9 12:40:46 2011 -0700
  9505.  
  9506. Fix chroma intra analysis in 4:4:4 lossless mode
  9507.  
  9508. commit 294df95060118de1d605ce20fcf490cdb4f4d14c [revision 2087]
  9509. Author: Anton Mitrofanov <[email protected]>
  9510. AuthorDate: Sun Oct 9 01:13:29 2011 +0400
  9511. Commit: Fiona Glaser <[email protected]>
  9512. CommitDate: Sat Oct 8 18:02:02 2011 -0700
  9513.  
  9514. Fix use of uninitialized MVs in sub8x8 RDO
  9515.  
  9516. commit 3ff2feee5a176ec8012c313e4a9e2b3611f29614 [revision 2086]
  9517. Author: Fabian Greffrath <[email protected]>
  9518. AuthorDate: Fri Oct 7 19:04:17 2011 -0700
  9519. Commit: Fiona Glaser <[email protected]>
  9520. CommitDate: Fri Oct 7 19:05:36 2011 -0700
  9521.  
  9522. Fix detection of Alpha CPU arch on alphaev67
  9523.  
  9524. commit 2701440c515a9a8aee1c87d7c06c98e43c3d813f [revision 2085]
  9525. Author: Fiona Glaser <[email protected]>
  9526. AuthorDate: Wed Sep 14 14:53:04 2011 -0700
  9527. Commit: Fiona Glaser <[email protected]>
  9528. CommitDate: Wed Sep 21 12:25:28 2011 -0700
  9529.  
  9530. Optimize x86 asm for Intel macro-op fusion
  9531. That is, place all loop counter tests right before their conditional jumps.
  9532.  
  9533. commit 2d481bc0ee053634fe46c0df2cbc646733dd137d [revision 2084]
  9534. Author: Fiona Glaser <[email protected]>
  9535. AuthorDate: Mon Sep 12 11:51:23 2011 -0700
  9536. Commit: Fiona Glaser <[email protected]>
  9537. CommitDate: Wed Sep 21 12:25:02 2011 -0700
  9538.  
  9539. CAVLC: clean up and restructure
  9540. Somewhat faster CAVLC and RD bit-counting.
  9541.  
  9542. commit da768d95d5d63f1eac77a35731079ce02aaa125c [revision 2083]
  9543. Author: Fiona Glaser <[email protected]>
  9544. AuthorDate: Thu Sep 8 17:27:02 2011 -0700
  9545. Commit: Fiona Glaser <[email protected]>
  9546. CommitDate: Wed Sep 21 09:54:49 2011 -0700
  9547.  
  9548. CABAC: clean up and restructure
  9549. Somewhat faster CABAC and RD bit-counting.
  9550.  
  9551. commit 389b401a99f2f33b41db7d74904b3ff7509d79e5 [revision 2082]
  9552. Author: Fiona Glaser <[email protected]>
  9553. AuthorDate: Sun Sep 4 11:31:29 2011 +0200
  9554. Commit: Fiona Glaser <[email protected]>
  9555. CommitDate: Wed Sep 21 09:54:47 2011 -0700
  9556.  
  9557. Some initial 4:2:2 x86 asm
  9558.  
  9559. commit 5b0cb86f27ba0c5433c404bed51c06a5124dfb49 [revision 2081]
  9560. Author: Henrik Gramner <[email protected]>
  9561. AuthorDate: Fri Aug 26 15:57:04 2011 +0200
  9562. Commit: Fiona Glaser <[email protected]>
  9563. CommitDate: Wed Sep 21 09:54:44 2011 -0700
  9564.  
  9565. 4:2:2 encoding support
  9566.  
  9567. commit 3d82e875d06b9d1e15ca2baa16b1bd9640500972 [revision 2080]
  9568. Author: Loren Merritt <[email protected]>
  9569. AuthorDate: Mon Aug 15 18:18:55 2011 +0000
  9570. Commit: Fiona Glaser <[email protected]>
  9571. CommitDate: Wed Sep 21 09:54:41 2011 -0700
  9572.  
  9573. SSSE3/SSE4 9-way fully merged i4x4 analysis (sad/satd_x9)
  9574.  
  9575. i4x4 analysis cycles (per partition):
  9576. penryn sandybridge
  9577. 184-> 75 157-> 54 preset=superfast (sad)
  9578. 281->165 225->124 preset=faster (satd with early termination)
  9579. 332->165 263->124 preset=medium
  9580. 379->165 297->124 preset=slower (satd without early termination)
  9581.  
  9582. This is the first code in x264 that intentionally produces different behavior
  9583. on different cpus: satd_x9 is implemented only on ssse3+ and checks all intra
  9584. directions, whereas the old code (on fast presets) may early terminate after
  9585. checking only some of them. There is no systematic difference on slow presets,
  9586. though they still occasionally disagree about tiebreaks.
  9587.  
  9588. For ease of debugging, add an option "--cpu-independent" to disable satd_x9
  9589. and any analogous future code.
  9590.  
  9591. commit e184ff26233198932d9b77aa7feed6a49095f136 [revision 2079]
  9592. Author: Loren Merritt <[email protected]>
  9593. AuthorDate: Mon Aug 15 17:43:42 2011 +0000
  9594. Commit: Fiona Glaser <[email protected]>
  9595. CommitDate: Wed Sep 21 09:54:39 2011 -0700
  9596.  
  9597. Faster intra_mbcmp_x3 for versions without dedicated asm
  9598. Select asm subroutines more intelligently in the wrapper functions.
  9599.  
  9600. commit d94edd734304c16265f28dd11e8a2029cbdc5b7f [revision 2078]
  9601. Author: Loren Merritt <[email protected]>
  9602. AuthorDate: Sat Aug 13 19:01:22 2011 +0000
  9603. Commit: Fiona Glaser <[email protected]>
  9604. CommitDate: Wed Sep 21 09:54:36 2011 -0700
  9605.  
  9606. Optimize x86 intra_predict_4x4 and 8x8
  9607.  
  9608. High bit depth Penryn, Sandybridge cycles:
  9609. 4x4_ddl: 11->10, 9-> 8
  9610. 4x4_ddr: 15->13, 12->11
  9611. 4x4_hd: , 15->12
  9612. 4x4_hu: , 14->13
  9613. 4x4_vr: 15->14, 14->12
  9614. 8x8_ddl: 32->19, 19->14
  9615. 8x8_ddr: 42->19, 21->14
  9616. 8x8_hd: , 15->13
  9617. 8x8_hu: 21->17, 16->12
  9618. 8x8_vr: 33->19,
  9619.  
  9620. 8-bit Penryn, Sandybridge cycles:
  9621. 4x4_ddr: 24->15,
  9622. 4x4_hd: 24->16,
  9623. 4x4_hu: 23->15,
  9624. 4x4_vr: 23->16,
  9625. 4x4_vl: 10-> 9,
  9626. 8x8_ddl: 23->15,
  9627. 8x8_hd: , 17->14
  9628. 8x8_hu: , 15->14
  9629. 8x8_vr: 20->16, 17->13
  9630.  
  9631. commit 37b2d963b262d2880271f313a17fceeee27a3de8 [revision 2077]
  9632. Author: Loren Merritt <[email protected]>
  9633. AuthorDate: Sat Aug 13 06:44:28 2011 +0000
  9634. Commit: Fiona Glaser <[email protected]>
  9635. CommitDate: Wed Sep 21 09:54:34 2011 -0700
  9636.  
  9637. Use realistic alignment for intra pred benchmarks in checkasm
  9638.  
  9639. commit 10ef9590e33d209a937fcb3f5ca1be66fb481a17 [revision 2076]
  9640. Author: Yusuke Nakamura <[email protected]>
  9641. AuthorDate: Wed Sep 21 01:15:38 2011 +0900
  9642. Commit: Fiona Glaser <[email protected]>
  9643. CommitDate: Wed Sep 21 09:54:31 2011 -0700
  9644.  
  9645. Fix frame packing SEI with --frame-packing 0
  9646. According to the spec, when frame_packing_arrangement_type is equal to 0, quincunx_sampling_flag shall be equal to 1.
  9647.  
  9648. commit cb648060484f081eba39480a26791a8e0d605989 [revision 2075]
  9649. Author: Oka Motofumi <[email protected]>
  9650. AuthorDate: Mon Sep 5 11:50:37 2011 +0900
  9651. Commit: Fiona Glaser <[email protected]>
  9652. CommitDate: Fri Sep 9 14:41:33 2011 -0700
  9653.  
  9654. Fix install/uninstall shared libs if SYS is WINDOWS/CYGWIN
  9655.  
  9656. commit d2452266ccf4bd9552d7ac94b5600b416d757d34 [revision 2074]
  9657. Author: Reinhard Tartler <[email protected]>
  9658. AuthorDate: Wed Aug 10 00:16:46 2011 -0700
  9659. Commit: Fiona Glaser <[email protected]>
  9660. CommitDate: Wed Aug 24 13:38:17 2011 -0700
  9661.  
  9662. Add Hurd support to configure
  9663.  
  9664. commit 75de7be6d5e7b0e1fc0febf087be65e91c00b80b [revision 2073]
  9665. Author: Loren Merritt <[email protected]>
  9666. AuthorDate: Sat Aug 13 00:39:35 2011 +0000
  9667. Commit: Fiona Glaser <[email protected]>
  9668. CommitDate: Wed Aug 24 13:38:15 2011 -0700
  9669.  
  9670. Optimize x86 intra_satd_x3_*
  9671. ~7% faster.
  9672.  
  9673. commit b597966bfa8a481489e5af93eb25988456c51a5d [revision 2072]
  9674. Author: Loren Merritt <[email protected]>
  9675. AuthorDate: Fri Aug 12 19:13:07 2011 +0000
  9676. Commit: Fiona Glaser <[email protected]>
  9677. CommitDate: Wed Aug 24 13:38:12 2011 -0700
  9678.  
  9679. Optimize x86 intra_sa8d_x3_8x8
  9680. ~40% faster.
  9681. Also some other minor asm cosmetics.
  9682.  
  9683. commit f3fc0c4485aa3ed1774bce462ad3fb92faec114b [revision 2071]
  9684. Author: Loren Merritt <[email protected]>
  9685. AuthorDate: Fri Aug 12 02:15:46 2011 +0000
  9686. Commit: Fiona Glaser <[email protected]>
  9687. CommitDate: Wed Aug 24 13:37:51 2011 -0700
  9688.  
  9689. Scale interlaced refs/mvs for mvr predictors
  9690. Slightly improves compression and fixes a Valgrind error.
  9691.  
  9692. commit ebc334f8d1d2752b9bc2c56e457fffc123ffddee [revision 2070]
  9693. Author: Loren Merritt <[email protected]>
  9694. AuthorDate: Thu Aug 11 15:03:12 2011 +0000
  9695. Commit: Fiona Glaser <[email protected]>
  9696. CommitDate: Wed Aug 24 10:23:22 2011 -0700
  9697.  
  9698. Optimize predict_8x8_filter and incidentally remove a valgrind false-positive
  9699.  
  9700. commit 94493149bbc251d0ce4ceee85a9faa5ad8837a04 [revision 2069]
  9701. Author: Anton Mitrofanov <[email protected]>
  9702. AuthorDate: Mon Aug 15 12:22:18 2011 +0400
  9703. Commit: Fiona Glaser <[email protected]>
  9704. CommitDate: Wed Aug 24 10:23:19 2011 -0700
  9705.  
  9706. Don't override flat SSE2 dequant functions with non-flat AVX ones
  9707. Slightly faster.
  9708.  
  9709. commit 25a8bb9461bf7b0c75e7fadc8d104dbdc61bed5c [revision 2068]
  9710. Author: Loren Merritt <[email protected]>
  9711. AuthorDate: Mon Aug 8 13:40:53 2011 +0000
  9712. Commit: Fiona Glaser <[email protected]>
  9713. CommitDate: Wed Aug 24 10:23:17 2011 -0700
  9714.  
  9715. Shut up some valgrind false-positives
  9716.  
  9717. commit ede9651875846116bdb2d3d84e3630beada7e21d [revision 2067]
  9718. Author: Fiona Glaser <[email protected]>
  9719. AuthorDate: Tue Aug 16 13:02:24 2011 -0700
  9720. Commit: Fiona Glaser <[email protected]>
  9721. CommitDate: Wed Aug 24 10:23:14 2011 -0700
  9722.  
  9723. Avoid some unnecessary allocations with B-frames/CABAC off
  9724.  
  9725. commit 17f16d161e386457f7eaa96866550c497af681d5 [revision 2066]
  9726. Author: Fiona Glaser <[email protected]>
  9727. AuthorDate: Mon Aug 22 17:07:03 2011 -0700
  9728. Commit: Fiona Glaser <[email protected]>
  9729. CommitDate: Wed Aug 24 10:23:12 2011 -0700
  9730.  
  9731. Fix typo in p8x8 RD analysis
  9732. Passed wrong idx to trellis.
  9733.  
  9734. commit 5a22495a2a857b9fcd5825595422c78f0223a417 [revision 2065]
  9735. Author: Anton Mitrofanov <[email protected]>
  9736. AuthorDate: Sun Aug 21 02:44:45 2011 +0400
  9737. Commit: Fiona Glaser <[email protected]>
  9738. CommitDate: Wed Aug 24 10:23:09 2011 -0700
  9739.  
  9740. Fix invalid memory accesses in x86 lowres_init when width <= 16
  9741.  
  9742. commit 8b72a9e4c87bbdfa1b87609fa9cde9bf61440383 [revision 2064]
  9743. Author: Anton Mitrofanov <[email protected]>
  9744. AuthorDate: Mon Aug 15 12:03:09 2011 +0400
  9745. Commit: Fiona Glaser <[email protected]>
  9746. CommitDate: Wed Aug 24 10:23:00 2011 -0700
  9747.  
  9748. Fix intermediate conversion for YUVJ* pixfmts with 4:4:4 encoding
  9749.  
  9750. commit cec1f4039fb6f4bf1c5ef97648b94e489400e5bc [revision 2063]
  9751. Author: Henrik Gramner <[email protected]>
  9752. AuthorDate: Sun Aug 14 13:39:29 2011 +0200
  9753. Commit: Fiona Glaser <[email protected]>
  9754. CommitDate: Wed Aug 24 10:22:54 2011 -0700
  9755.  
  9756. Fix pic_out returned by x264_encoder_encode with 4:4:4
  9757.  
  9758. commit eaa858d33b9dcb6e526b01cc39d0268d4ae6d8c0 [revision 2062]
  9759. Author: Loren Merritt <[email protected]>
  9760. AuthorDate: Thu Aug 11 22:12:26 2011 +0000
  9761. Commit: Fiona Glaser <[email protected]>
  9762. CommitDate: Wed Aug 24 10:22:51 2011 -0700
  9763.  
  9764. Fix zeroing of mvr predictors in bskip blocks
  9765.  
  9766. commit 29e318fd26bd3a2e689801aeb9ff84d9e6c1d25f [revision 2061]
  9767. Author: Loren Merritt <[email protected]>
  9768. AuthorDate: Thu Aug 11 01:33:13 2011 +0000
  9769. Commit: Fiona Glaser <[email protected]>
  9770. CommitDate: Wed Aug 24 10:22:49 2011 -0700
  9771.  
  9772. Fix: chroma planes for weightp analysis were not initted if U early-terminates and V doesn't.
  9773.  
  9774. commit af0d8d8588e9eed4c4895747fcb7485dd0210bcf [revision 2060]
  9775. Author: Henrik Gramner <[email protected]>
  9776. AuthorDate: Wed Aug 10 20:25:07 2011 +0200
  9777. Commit: Fiona Glaser <[email protected]>
  9778. CommitDate: Wed Aug 24 10:22:46 2011 -0700
  9779.  
  9780. Expand borders before chroma weightp analysis
  9781. Prevents mc from using uninitialized source pixels.
  9782.  
  9783. commit cfcce49df42848f601cb05086d1ef89c23675397 [revision 2059]
  9784. Author: Henrik Gramner <[email protected]>
  9785. AuthorDate: Wed Aug 10 19:29:14 2011 +0200
  9786. Commit: Fiona Glaser <[email protected]>
  9787. CommitDate: Wed Aug 24 10:22:43 2011 -0700
  9788.  
  9789. Another 4:4:4 chroma weightp bug fix
  9790.  
  9791. commit 51821635e8dccf877c3521830a8a5598c2bc408b [revision 2058]
  9792. Author: Fiona Glaser <[email protected]>
  9793. AuthorDate: Wed Aug 10 00:17:26 2011 -0700
  9794. Commit: Fiona Glaser <[email protected]>
  9795. CommitDate: Wed Aug 24 10:22:20 2011 -0700
  9796.  
  9797. Fix typo in help
  9798.  
  9799. commit 3817e54a3aeaa387206f78d5eaee98339dd7d93b [revision 2057]
  9800. Author: Fiona Glaser <[email protected]>
  9801. AuthorDate: Sat Aug 6 10:45:47 2011 -0700
  9802. Commit: Fiona Glaser <[email protected]>
  9803. CommitDate: Tue Aug 9 17:21:44 2011 -0700
  9804.  
  9805. Improve support for varying resolution between passes
  9806. Should give much better quality, but still doesn't support MB-tree yet.
  9807. Also check for the same interlaced options between passes.
  9808. Various minor ratecontrol cosmetics.
  9809.  
  9810. commit 9b9a13a98b98385884b7ac25710305ad431c62e4 [revision 2056]
  9811. Author: Loren Merritt <[email protected]>
  9812. AuthorDate: Sun Aug 7 22:57:27 2011 +0000
  9813. Commit: Fiona Glaser <[email protected]>
  9814. CommitDate: Tue Aug 9 17:21:41 2011 -0700
  9815.  
  9816. asm cosmetics: base-4 constants for shuffles
  9817.  
  9818. commit 7e60fcd7af513e48d912dfce21026420698ed6ba [revision 2055]
  9819. Author: Loren Merritt <[email protected]>
  9820. AuthorDate: Wed Aug 3 14:58:50 2011 +0000
  9821. Commit: Fiona Glaser <[email protected]>
  9822. CommitDate: Tue Aug 9 17:21:33 2011 -0700
  9823.  
  9824. Enable some existing asm functions that were missing function pointers
  9825. pixel_ads1_avx, predict_8x8_hd_avxx
  9826. High bit depth mc_copy_w8_sse2, denoise_dct_avx, prefetch_fenc/ref, and several pixel*sse4.
  9827.  
  9828. commit 52f287e84a9965f652221f535a3298c7ce0846b9 [revision 2054]
  9829. Author: Loren Merritt <[email protected]>
  9830. AuthorDate: Wed Aug 3 14:57:06 2011 +0000
  9831. Commit: Fiona Glaser <[email protected]>
  9832. CommitDate: Tue Aug 9 14:17:42 2011 -0700
  9833.  
  9834. Remove some unused, broken, and/or useless functions
  9835. Unused frame_sort.
  9836. Unused x86_64 dequant_4x4dc_mmx2, predict_8x8_vr_mmx2.
  9837. Unused and broken high_depth integral_init*h_sse4, optimize_chroma_*, dequant_flat_*, sub8x8_dct_dc_*, zigzag_sub_*.
  9838. Useless high_depth dequant_sse4, dequant_dc_sse4.
  9839.  
  9840. commit 309ddabbb3fba9ba0a2ae4c23470ec539d052374 [revision 2053]
  9841. Author: Loren Merritt <[email protected]>
  9842. AuthorDate: Wed Aug 3 14:56:27 2011 +0000
  9843. Commit: Fiona Glaser <[email protected]>
  9844. CommitDate: Tue Aug 9 14:17:39 2011 -0700
  9845.  
  9846. asm cosmetics: merge all the variants of ABS macros
  9847.  
  9848. commit 1921c6824e37bdf5a8436a6cbe36b0d3a8c376b3 [revision 2052]
  9849. Author: Loren Merritt <[email protected]>
  9850. AuthorDate: Wed Aug 3 14:53:29 2011 +0000
  9851. Commit: Fiona Glaser <[email protected]>
  9852. CommitDate: Tue Aug 9 14:17:37 2011 -0700
  9853.  
  9854. asm cosmetics part 2
  9855. These changes were split out of the cpuflags commit because they change the output executable.
  9856.  
  9857. commit f85be1cdbe8d9244c0465df13ed58215a8c673cc [revision 2051]
  9858. Author: Loren Merritt <[email protected]>
  9859. AuthorDate: Wed Aug 3 14:46:41 2011 +0000
  9860. Commit: Fiona Glaser <[email protected]>
  9861. CommitDate: Tue Aug 9 14:17:34 2011 -0700
  9862.  
  9863. asm cosmetics: INIT_MMX/XMM/YMM now support a cpuflags argument
  9864.  
  9865. Reduces the number of macro args that need to be passed around.
  9866. Allows multiple implementations of a given macro (e.g. PALIGNR) to check
  9867. cpuflags at the location where the macro is defined, instead of having
  9868. to select implementations by %define at toplevel.
  9869. Remove INIT_AVX, as it's replaced by "INIT_XMM avx".
  9870.  
  9871. This commit does not change the stripped executable.
  9872.  
  9873. commit 67336688cdc0c47468cef4e6f8cf57ffd010b56e [revision 2050]
  9874. Author: Loren Merritt <[email protected]>
  9875. AuthorDate: Wed Aug 3 14:43:34 2011 +0000
  9876. Commit: Fiona Glaser <[email protected]>
  9877. CommitDate: Tue Aug 9 14:17:32 2011 -0700
  9878.  
  9879. Import x86inc.asm patches from libav
  9880.  
  9881. commit 189c30d390d08b2b3d3007acd0a106a4e0cd17b2 [revision 2049]
  9882. Author: Loren Merritt <[email protected]>
  9883. AuthorDate: Wed Aug 3 14:42:12 2011 +0000
  9884. Commit: Fiona Glaser <[email protected]>
  9885. CommitDate: Tue Aug 9 14:17:30 2011 -0700
  9886.  
  9887. Cosmetics: s/mmxext/mmx2/
  9888.  
  9889. commit b37de18947348199bdc392b38e979f619978126e [revision 2048]
  9890. Author: Henrik Gramner <[email protected]>
  9891. AuthorDate: Sun Aug 7 11:58:36 2011 +0200
  9892. Commit: Fiona Glaser <[email protected]>
  9893. CommitDate: Tue Aug 9 14:17:13 2011 -0700
  9894.  
  9895. Fix two bugs in 4:4:4 chroma weightp analysis
  9896. Caused slightly worse compression.
  9897.  
  9898. commit 014f9c8e3fa202f13f926ac037c3a8db494522ea [revision 2047]
  9899. Author: Loren Merritt <[email protected]>
  9900. AuthorDate: Wed Aug 3 14:40:01 2011 +0000
  9901. Commit: Fiona Glaser <[email protected]>
  9902. CommitDate: Fri Aug 5 16:33:27 2011 -0700
  9903.  
  9904. Fix "--asm avx"
  9905. Previously required "--asm sse2fast,fastshuffle,sse4.2,avx".
  9906.  
  9907. commit 3674cf4fd338a7894883a0172ec6bde61eac6c25 [revision 2046]
  9908. Author: Anton Mitrofanov <[email protected]>
  9909. AuthorDate: Fri Aug 5 15:59:20 2011 +0400
  9910. Commit: Fiona Glaser <[email protected]>
  9911. CommitDate: Fri Aug 5 14:42:49 2011 -0700
  9912.  
  9913. Re-add support for glibc <2.6, which doesn't have CPU_COUNT
  9914.  
  9915. commit 1dd4b85fc700db5ec4380e20c2d73882808b3763 [revision 2045]
  9916. Author: Yasuhiro Ikeda <[email protected]>
  9917. AuthorDate: Tue Aug 2 08:59:15 2011 +0900
  9918. Commit: Fiona Glaser <[email protected]>
  9919. CommitDate: Fri Aug 5 14:42:22 2011 -0700
  9920.  
  9921. Avoid using deprecated libavformat functions
  9922. Replace av_find_stream_info with avformat_find_stream_info.
  9923. Now requires libavformat 53.3.0 or newer.
  9924.  
  9925. commit 191b68df93e7ad4096c6aa4df4120dcb0e83dded [revision 2044]
  9926. Author: Henrik Gramner <[email protected]>
  9927. AuthorDate: Wed Jul 27 02:23:12 2011 +0200
  9928. Commit: Fiona Glaser <[email protected]>
  9929. CommitDate: Fri Jul 29 12:09:35 2011 -0700
  9930.  
  9931. Use assembly versions of some deblocking functions in MBAFF
  9932.  
  9933. commit 459ac481e85833550470d231ae4749a138146614 [revision 2043]
  9934. Author: Anton Mitrofanov <[email protected]>
  9935. AuthorDate: Thu Jul 28 00:26:27 2011 +0400
  9936. Commit: Fiona Glaser <[email protected]>
  9937. CommitDate: Fri Jul 29 12:09:33 2011 -0700
  9938.  
  9939. Move X264_VERSION / X264_POINTVER from config.h to x264_config.h
  9940. This makes them available to external programs as part of the public API.
  9941.  
  9942. commit 95f03f9e89c04b29aa4b5ad57fa4869899eedb4c [revision 2042]
  9943. Author: Henrik Gramner <[email protected]>
  9944. AuthorDate: Fri Jul 29 20:15:52 2011 +0200
  9945. Commit: Fiona Glaser <[email protected]>
  9946. CommitDate: Fri Jul 29 12:09:30 2011 -0700
  9947.  
  9948. Fix padding bug in x264_expand_border_mbpair
  9949.  
  9950. commit eee242c1a64db0c4975eaf9add82565502882293 [revision 2041]
  9951. Author: Yusuke Nakamura <[email protected]>
  9952. AuthorDate: Fri Jul 29 23:39:26 2011 +0900
  9953. Commit: Fiona Glaser <[email protected]>
  9954. CommitDate: Fri Jul 29 12:07:33 2011 -0700
  9955.  
  9956. Timecode parsing: Add missing initialization
  9957. Fix crash when failed to parse timecode file before malloc pts.
  9958. Fix detection of user timebase considered to be exceeding H.264 maximum.
  9959.  
  9960. commit e1ec7c8ae8d865165c802a69387e4d41cb004e43 [revision 2040]
  9961. Author: Anton Mitrofanov <[email protected]>
  9962. AuthorDate: Thu Jul 28 13:37:24 2011 +0400
  9963. Commit: Fiona Glaser <[email protected]>
  9964. CommitDate: Fri Jul 29 12:07:30 2011 -0700
  9965.  
  9966. Fix crash with high bitdepth 4:2:0 input
  9967.  
  9968. commit 10dc5bb27739fd112f5b94ffb9419fa8781c5bbe [revision 2039]
  9969. Author: Daniel Kang <[email protected]>
  9970. AuthorDate: Tue Jul 26 21:57:39 2011 -0400
  9971. Commit: Fiona Glaser <[email protected]>
  9972. CommitDate: Fri Jul 29 12:07:27 2011 -0700
  9973.  
  9974. x86 asm cosmetics
  9975. Use FDEC_STRIDEB where appropriate.
  9976.  
  9977. commit bbfbacc9d3fa89cd922f33feb3924b67fdf31f7b [revision 2038]
  9978. Author: Fiona Glaser <[email protected]>
  9979. AuthorDate: Tue Jul 26 07:40:23 2011 -0700
  9980. Commit: Fiona Glaser <[email protected]>
  9981. CommitDate: Tue Jul 26 07:41:11 2011 -0700
  9982.  
  9983. Fix a bug in lossless sub-8x8 RD
  9984. Caused crashes in rare cases with lossless encoding. Regression in 4:4:4.
  9985.  
  9986. commit 10474f5af22f3b2444a4301252175657b6fb1514 [revision 2037]
  9987. Author: Fiona Glaser <[email protected]>
  9988. AuthorDate: Mon Jul 18 23:10:30 2011 -0700
  9989. Commit: Fiona Glaser <[email protected]>
  9990. CommitDate: Fri Jul 22 15:49:41 2011 -0700
  9991.  
  9992. Improved p8x4/4x8 search decision
  9993. Use the same thresholding as for p16x8/8x16.
  9994. Does p8x4/4x8 search more often, for a small compression improvement.
  9995.  
  9996. commit 4a88ee1c649d92bbdbbf128e22d547e9b833f00c [revision 2036]
  9997. Author: Dan Larkin <[email protected]>
  9998. AuthorDate: Wed Jul 13 12:45:23 2011 -0500
  9999. Commit: Fiona Glaser <[email protected]>
  10000. CommitDate: Fri Jul 22 15:49:39 2011 -0700
  10001.  
  10002. Add --subme 11, which disables all early terminations in analysis
  10003. Necessary for a future trellis mode decision/motion estimation patch.
  10004. Also add the slowest presets to the regression test.
  10005.  
  10006. commit 330c8fdaccd63383ba6f7f1ccf787a5f1b89d09b [revision 2035]
  10007. Author: Dan Larkin <[email protected]>
  10008. AuthorDate: Wed Jul 13 11:33:48 2011 -0500
  10009. Commit: Fiona Glaser <[email protected]>
  10010. CommitDate: Fri Jul 22 15:49:36 2011 -0700
  10011.  
  10012. Some trivial changes to RD thresholds
  10013. The output-changing portion of the next patch.
  10014.  
  10015. commit b5e21b60fe4422c85b9f27eda6f45d7a5517e160 [revision 2034]
  10016. Author: Anton Mitrofanov <[email protected]>
  10017. AuthorDate: Wed Jul 20 22:54:43 2011 +0400
  10018. Commit: Fiona Glaser <[email protected]>
  10019. CommitDate: Fri Jul 22 15:49:33 2011 -0700
  10020.  
  10021. Allow setting a wider range of chroma QP offsets
  10022. This allows use of the full range of chroma QP offsets, even in combination with the automatic psy-based adjustments.
  10023.  
  10024. commit 1f285bd40b45dfa97fadc86f912a19c54563fa77 [revision 2033]
  10025. Author: Fiona Glaser <[email protected]>
  10026. AuthorDate: Fri Jul 15 13:24:38 2011 -0700
  10027. Commit: Fiona Glaser <[email protected]>
  10028. CommitDate: Fri Jul 22 15:49:23 2011 -0700
  10029.  
  10030. Optimize macroblock_deblock_strength, add more early terminations
  10031.  
  10032. commit 695bac1d7e66ead90952e333abeab0176ea7221d [revision 2032]
  10033. Author: Fiona Glaser <[email protected]>
  10034. AuthorDate: Thu Jul 14 18:23:44 2011 -0700
  10035. Commit: Fiona Glaser <[email protected]>
  10036. CommitDate: Fri Jul 22 04:03:25 2011 -0700
  10037.  
  10038. Function-pointerify MBAFF deblocking functions
  10039.  
  10040. commit 75466d2e4fff1aeba7e64a1947e8beea3f1235ff [revision 2031]
  10041. Author: Fiona Glaser <[email protected]>
  10042. AuthorDate: Thu Jul 14 14:04:11 2011 -0700
  10043. Commit: Fiona Glaser <[email protected]>
  10044. CommitDate: Fri Jul 22 04:03:22 2011 -0700
  10045.  
  10046. Clean up MBAFF deblocking code
  10047.  
  10048. commit 8ae69dbc7ec37e157a3890c21ec4904973e800f9 [revision 2030]
  10049. Author: Fiona Glaser <[email protected]>
  10050. AuthorDate: Tue Jul 12 17:27:18 2011 -0700
  10051. Commit: Fiona Glaser <[email protected]>
  10052. CommitDate: Fri Jul 22 04:03:19 2011 -0700
  10053.  
  10054. Optimize frame_deblock_row
  10055.  
  10056. commit 44269ed290f1a5457c24b6e2992bc65e92a70ac4 [revision 2029]
  10057. Author: Henrik Gramner <[email protected]>
  10058. AuthorDate: Wed Jul 20 22:30:59 2011 +0200
  10059. Commit: Fiona Glaser <[email protected]>
  10060. CommitDate: Fri Jul 22 04:03:16 2011 -0700
  10061.  
  10062. Shrink two arrays
  10063.  
  10064. commit aea1565f5f5d793935b10cd6081bf8dbe9513db5 [revision 2028]
  10065. Author: Anton Mitrofanov <[email protected]>
  10066. AuthorDate: Mon Jul 18 15:20:05 2011 +0400
  10067. Commit: Fiona Glaser <[email protected]>
  10068. CommitDate: Fri Jul 22 04:03:14 2011 -0700
  10069.  
  10070. Add support for the new (4:4:4) colorspaces to x264_picture_alloc
  10071.  
  10072. commit e93cfd6adcdd246372a38f2598590c0ab7c69b7d [revision 2027]
  10073. Author: Anton Mitrofanov <[email protected]>
  10074. AuthorDate: Wed Jul 20 18:06:41 2011 +0400
  10075. Commit: Fiona Glaser <[email protected]>
  10076. CommitDate: Fri Jul 22 04:03:10 2011 -0700
  10077.  
  10078. Various cosmetics
  10079.  
  10080. commit 3ef68d34b477bfd7410267eecbeaa8ebb44bccc4 [revision 2026]
  10081. Author: Yasuhiro Ikeda <[email protected]>
  10082. AuthorDate: Tue Jul 12 23:41:42 2011 +0900
  10083. Commit: Fiona Glaser <[email protected]>
  10084. CommitDate: Fri Jul 22 04:03:00 2011 -0700
  10085.  
  10086. Improve configure help
  10087.  
  10088. commit 9dd3e96e9420fac3cb00d44eab75450c630fe231 [revision 2025]
  10089. Author: Yasuhiro Ikeda <[email protected]>
  10090. AuthorDate: Tue Jul 12 14:46:29 2011 +0900
  10091. Commit: Fiona Glaser <[email protected]>
  10092. CommitDate: Wed Jul 20 19:52:04 2011 -0700
  10093.  
  10094. Use $optarg for some configure options
  10095.  
  10096. commit f7e6610ba12319d68833526676b16879aaff415c [revision 2024]
  10097. Author: Rafaël Carré <[email protected]>
  10098. AuthorDate: Thu Jul 14 18:51:43 2011 -0700
  10099. Commit: Fiona Glaser <[email protected]>
  10100. CommitDate: Wed Jul 20 19:48:25 2011 -0700
  10101.  
  10102. Linux x264_cpu_num_processors(): use glibc macros
  10103. The cpu_set_t structure is considered opaque.
  10104. Also handle sched_getaffinity() error case if "cpusetsize is smaller than the size of the affinity mask used by the kernel."
  10105.  
  10106. commit 670d81811866e9e5045d25c5def5ba2b9f06d2ac [revision 2023]
  10107. Author: Anton Mitrofanov <[email protected]>
  10108. AuthorDate: Thu Jul 14 17:02:43 2011 +0400
  10109. Commit: Fiona Glaser <[email protected]>
  10110. CommitDate: Mon Jul 18 09:16:35 2011 -0700
  10111.  
  10112. Fix spurious "stream properties changed" with --seek option on some inputs
  10113.  
  10114. commit aa50e72e7c723927325d031ab47b24e069dde4e3 [revision 2022]
  10115. Author: Anton Mitrofanov <[email protected]>
  10116. AuthorDate: Fri Jul 15 15:06:37 2011 +0400
  10117. Commit: Fiona Glaser <[email protected]>
  10118. CommitDate: Mon Jul 18 09:16:32 2011 -0700
  10119.  
  10120. Fix use of deprecated libavcodec functions
  10121. Replace avcodec_open with avcodec_open2. Now requires libavcodec 53.6.0 or newer.
  10122.  
  10123. commit 67c796a37233e66239226bacd74f038281d43095 [revision 2021]
  10124. Author: Kieran Kunhya <[email protected]>
  10125. AuthorDate: Wed Jul 13 20:25:40 2011 +0100
  10126. Commit: Fiona Glaser <[email protected]>
  10127. CommitDate: Thu Jul 14 16:37:57 2011 -0700
  10128.  
  10129. Fix nalu_process callback with HRD
  10130.  
  10131. commit bb784df93d92fb28f67a7998faed0da425b25623 [revision 2020]
  10132. Author: Anton Mitrofanov <[email protected]>
  10133. AuthorDate: Wed Jul 13 15:55:38 2011 +0400
  10134. Commit: Fiona Glaser <[email protected]>
  10135. CommitDate: Thu Jul 14 16:37:54 2011 -0700
  10136.  
  10137. Fix incorrect chroma swap for some input pixfmts
  10138.  
  10139. Problem occurred if pixfmt of lavf/ffms input was PIX_FMT_RGB24 or PIX_FMT_YUV444P.
  10140.  
  10141. commit ad1c2c8e383cb0f23ba8a0ba2ae211ad9f5eba62 [revision 2019]
  10142. Author: Anton Mitrofanov <[email protected]>
  10143. AuthorDate: Tue Jun 28 21:39:09 2011 +0400
  10144. Commit: Fiona Glaser <[email protected]>
  10145. CommitDate: Sat Jul 9 21:15:59 2011 -0700
  10146.  
  10147. Fix resize filter crash with YUVJ* input pixfmt
  10148.  
  10149. commit ce55ae08a6aad516e6aa2ed58fd93a2adf39a997 [revision 2018]
  10150. Author: xvidfan <[email protected]>
  10151. AuthorDate: Wed Jun 22 18:46:14 2011 -0700
  10152. Commit: Fiona Glaser <[email protected]>
  10153. CommitDate: Sat Jul 9 21:15:56 2011 -0700
  10154.  
  10155. RGB encoding support
  10156. Much less efficient than YUV444, but easy to support using the YUV444 framework.
  10157.  
  10158. commit a93e4c4a75d05e7bf379cb9a39caad57f615eeb0 [revision 2017]
  10159. Author: Fiona Glaser <[email protected]>
  10160. AuthorDate: Wed Jun 22 03:32:53 2011 -0700
  10161. Commit: Fiona Glaser <[email protected]>
  10162. CommitDate: Sat Jul 9 21:15:52 2011 -0700
  10163.  
  10164. 4:4:4 encoding support
  10165.  
  10166. commit 323469e393af71dedd357763883232a293c3ab02 [revision 2016]
  10167. Author: Fiona Glaser <[email protected]>
  10168. AuthorDate: Mon Jun 20 16:20:21 2011 -0700
  10169. Commit: Fiona Glaser <[email protected]>
  10170. CommitDate: Sat Jul 9 21:15:48 2011 -0700
  10171.  
  10172. Properly weight slice header lambda in chroma weightp analysis
  10173.  
  10174. commit ae61d0c3c236140b6a7ee4ae5f691cf8191b2282 [revision 2015]
  10175. Author: Daniel Kang <[email protected]>
  10176. AuthorDate: Sun Jul 3 17:32:00 2011 -0400
  10177. Commit: Fiona Glaser <[email protected]>
  10178. CommitDate: Sat Jul 9 21:15:45 2011 -0700
  10179.  
  10180. Better x86 high bit depth predict_8x8c_p
  10181. Avoid the need to check for corner cases by reordering arithmetic.
  10182. Also make a minor optimization to high bit depth predict_16x16_p.
  10183.  
  10184. commit a8e1be77d59ff3e5ba565b6ee133a1b2364a2dfa [revision 2014]
  10185. Author: Fiona Glaser <[email protected]>
  10186. AuthorDate: Thu Jun 23 11:54:42 2011 -0700
  10187. Commit: Fiona Glaser <[email protected]>
  10188. CommitDate: Sat Jul 9 21:15:42 2011 -0700
  10189.  
  10190. Eliminate extra layer of indirection for sps/pps references
  10191. Also remove poc type 1 support (it didn't work anyways) to reduce sps size.
  10192.  
  10193. commit 8ade503619aff45e5be0ee544d8ab8c867eb5720 [revision 2013]
  10194. Author: Fiona Glaser <[email protected]>
  10195. AuthorDate: Sat Jul 9 19:21:00 2011 -0700
  10196. Commit: Fiona Glaser <[email protected]>
  10197. CommitDate: Sat Jul 9 21:15:31 2011 -0700
  10198.  
  10199. Fix SSIM calculation with sliced threads
  10200.  
  10201. commit 03bf7da697967bb8ed0b014e8623532b58051240 [revision 2012]
  10202. Author: Anton Mitrofanov <[email protected]>
  10203. AuthorDate: Sat Jul 9 23:57:44 2011 +0400
  10204. Commit: Fiona Glaser <[email protected]>
  10205. CommitDate: Sat Jul 9 19:40:01 2011 -0700
  10206.  
  10207. Avoid possible NaNs in B-frame output stats
  10208.  
  10209. commit defbf3f4d26d348bf07ec91588a304b59588d96e [revision 2011]
  10210. Author: Rémi Denis-Courmont <[email protected]>
  10211. AuthorDate: Thu Jun 30 14:07:43 2011 -0700
  10212. Commit: Fiona Glaser <[email protected]>
  10213. CommitDate: Sat Jul 9 19:39:58 2011 -0700
  10214.  
  10215. ARM: do not override the toolchain default for FPU ABI
  10216.  
  10217. commit fb629fcf1d280778f50db39f6c1038158321cc3c [revision 2010]
  10218. Author: Steven Walters <[email protected]>
  10219. AuthorDate: Thu Jun 23 20:29:01 2011 -0400
  10220. Commit: Fiona Glaser <[email protected]>
  10221. CommitDate: Sat Jul 9 19:39:54 2011 -0700
  10222.  
  10223. Fix link errors with libswscale/libavutil as shared libraries
  10224.  
  10225. commit e825c625999ddc0a27fc6c5cc0b39f198c22b021 [revision 2009]
  10226. Author: Steven Walters <[email protected]>
  10227. AuthorDate: Sat Jun 18 14:12:34 2011 -0400
  10228. Commit: Fiona Glaser <[email protected]>
  10229. CommitDate: Sat Jul 9 19:39:41 2011 -0700
  10230.  
  10231. Fix deprecation in libavformat usage
  10232. Replace av_open_input_file with avformat_open_input. Now requires libavformat 53.2.0 or newer.
  10233.  
  10234. commit d89c1b43816f05e43a836d38764d74d499e82a80 [revision 2008]
  10235. Author: Anton Mitrofanov <[email protected]>
  10236. AuthorDate: Thu Jun 9 01:34:14 2011 +0400
  10237. Commit: Fiona Glaser <[email protected]>
  10238. CommitDate: Tue Jun 14 19:43:21 2011 -0700
  10239.  
  10240. Fix various issues with VBV+threads
  10241. Eliminate the race condition with interframe row predictors and threads.
  10242. Recalculate frame_size_estimated at the end of a frame, for improved update_vbv_plan.
  10243. Some cosmetics.
  10244.  
  10245. commit ed3b10eb9cffcc346b5a070ce47f5a2beaf9efb6 [revision 2007]
  10246. Author: Anton Mitrofanov <[email protected]>
  10247. AuthorDate: Mon Jun 6 13:54:44 2011 +0400
  10248. Commit: Fiona Glaser <[email protected]>
  10249. CommitDate: Tue Jun 14 19:43:18 2011 -0700
  10250.  
  10251. Fix MBAFF row VBV ratecontrol
  10252. Reverts most of r1984 and implements a much simpler solution.
  10253.  
  10254. commit d091d0e6038e770ada1a856c601c401ba729d083 [revision 2006]
  10255. Author: Fiona Glaser <[email protected]>
  10256. AuthorDate: Mon May 23 17:01:02 2011 -0700
  10257. Commit: Fiona Glaser <[email protected]>
  10258. CommitDate: Tue Jun 14 19:43:16 2011 -0700
  10259.  
  10260. Make ratecontrol_mb less slow
  10261.  
  10262. commit 63eb8bc9b48564f777e98dd2528c07cff09184b1 [revision 2005]
  10263. Author: Steven Walters <[email protected]>
  10264. AuthorDate: Thu Jun 2 21:23:04 2011 -0400
  10265. Commit: Fiona Glaser <[email protected]>
  10266. CommitDate: Tue Jun 14 19:43:13 2011 -0700
  10267.  
  10268. Resize filter updates
  10269. Fix use of deprecated sws_getContext.
  10270. Fix uses of sws_format_name.
  10271. Fix stream change warning not occurring on the first resolution change.
  10272. Drop cpu detection, as it is now performed internally by swscale.
  10273. Update swscale version requirements.
  10274.  
  10275. commit d2e8686121a0418f466a0d79ef6a5367e944f940 [revision 2004]
  10276. Author: Fiona Glaser <[email protected]>
  10277. AuthorDate: Tue May 17 14:50:51 2011 -0700
  10278. Commit: Fiona Glaser <[email protected]>
  10279. CommitDate: Tue Jun 14 19:43:11 2011 -0700
  10280.  
  10281. AVX mbtree_propagate
  10282. Up to ~20-30% faster than SSE2 on Sandy Bridge.
  10283.  
  10284. commit 6d2b51a32bbaabee1a8762adb204d035d590331b [revision 2003]
  10285. Author: Fiona Glaser <[email protected]>
  10286. AuthorDate: Tue Jun 14 10:26:56 2011 -0700
  10287. Commit: Fiona Glaser <[email protected]>
  10288. CommitDate: Tue Jun 14 19:43:09 2011 -0700
  10289.  
  10290. Use -vsync 0 with ffmpeg regression test
  10291.  
  10292. commit 06fbd9df654cd2b7a025c12b3a7d4b3fb3386e23 [revision 2002]
  10293. Author: Henrik Gramner <[email protected]>
  10294. AuthorDate: Sat May 21 19:04:46 2011 +0200
  10295. Commit: Fiona Glaser <[email protected]>
  10296. CommitDate: Tue Jun 14 19:43:07 2011 -0700
  10297.  
  10298. Inline emms instructions on x86 if possible
  10299.  
  10300. commit f7c6d308f38b3193dbb7bd9f427252e296dfcbfe [revision 2001]
  10301. Author: Fiona Glaser <[email protected]>
  10302. AuthorDate: Tue Jun 14 09:35:03 2011 -0700
  10303. Commit: Fiona Glaser <[email protected]>
  10304. CommitDate: Tue Jun 14 19:42:40 2011 -0700
  10305.  
  10306. Make left_index_table const
  10307. Should allow for some missed compiler optimizations in macroblock_cache_load.
  10308.  
  10309. commit ca7852e211b5a270a8e4752526378a898f669017 [revision 2000]
  10310. Author: Hii <[email protected]>
  10311. AuthorDate: Tue May 24 08:31:17 2011 +0800
  10312. Commit: Fiona Glaser <[email protected]>
  10313. CommitDate: Sun Jun 12 19:48:50 2011 -0700
  10314.  
  10315. Make --profile main/baseline force off CQMfile
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!