log

age author description
Wed, 04 Mar 2015 08:30:49 -0600 Steve Borho frameencoder: use a timed wait to avoid a help-wanted flag race hazard
Mon, 02 Mar 2015 19:54:42 -0600 Steve Borho encoder: divide init/start and stop/delete to prevent race hazards
Mon, 02 Mar 2015 19:24:06 -0600 Steve Borho slicetype: clear m_helpWanted with input lock held to prevent race
Mon, 02 Mar 2015 18:52:34 -0600 Steve Borho me: split initialization to avoid harmless race hazard
Wed, 04 Mar 2015 10:35:55 -0600 Steve Borho weightb: avoid signed left-shifts (which cause -ftrapv exceptions)
Wed, 04 Mar 2015 07:20:07 -0600 Steve Borho primitives: ensure x265_cpu_cpuid() clears eax even when built without asm
Wed, 04 Mar 2015 07:11:02 -0600 Steve Borho frameencoder: nit
Wed, 04 Mar 2015 08:28:48 -0600 Steve Borho threading: nit
Fri, 27 Feb 2015 15:14:20 -0600 Steve Borho encoder: protect against logging inf stats if no RDO was performed
Wed, 04 Mar 2015 08:32:04 -0600 Steve Borho cmake: more general support for clang/gcc -fsanitize options
Mon, 02 Mar 2015 16:55:25 -0600 Steve Borho cmake: add -fstack-protector to checked builds, if available
Mon, 02 Mar 2015 10:04:37 -0600 Steve Borho cmake: ignore older versions of libnuma
Fri, 27 Feb 2015 11:46:09 +0530 Gopu Govindaswamy analysis: Share only if best reference is valid
Thu, 26 Feb 2015 15:26:39 -0600 Steve Borho asm: use grouping macros to simplify avx2 function sets
Thu, 26 Feb 2015 15:25:48 -0600 Steve Borho Backed out changeset: 8ba297f59e48
Thu, 26 Feb 2015 12:22:33 -0600 Steve Borho asm: introduce ALL_LUMA_CU_S for more house-cleaning
Thu, 26 Feb 2015 12:18:25 -0600 Steve Borho asm: fix cut-paste errors
Thu, 26 Feb 2015 12:16:20 -0600 Steve Borho asm: move more avx functions to their section, setup luma cu pointer
Thu, 26 Feb 2015 12:09:37 -0600 Steve Borho asm: move avx function assignments into the correct section
Thu, 26 Feb 2015 11:29:15 -0600 Steve Borho test: tabs to spaces
Thu, 26 Feb 2015 11:25:55 -0600 Steve Borho merge with avx2 optimization branch
Thu, 26 Feb 2015 19:11:18 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_8
Thu, 26 Feb 2015 19:00:49 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_28, correct ymm register count
Thu, 26 Feb 2015 18:55:45 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_28
Thu, 26 Feb 2015 17:55:31 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_7
Thu, 26 Feb 2015 17:20:36 +0530 Praveen Tiwari asm-avx2: fix intra_pred_ang8_29 mismatch issue
Thu, 26 Feb 2015 17:13:08 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_29, improved 428.68c -> 311.40c over sse4 asm code
Thu, 26 Feb 2015 16:24:18 +0530 Praveen Tiwari alasm-avx2: align pw_1024 global constant
Thu, 26 Feb 2015 15:39:32 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_6, improved 528.33c -> 412.07c over sse4 asm code
Thu, 26 Feb 2015 15:06:42 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_30, improved 237.03c -> 149.91c
Thu, 26 Feb 2015 13:18:32 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_5
Thu, 26 Feb 2015 12:46:54 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_31
Wed, 25 Feb 2015 20:17:05 +0530 Rajesh Paulraj asm-ssse3: filter_p2s 8bpp code for [4xN],[8xN],[16xN],[32xN]
Wed, 25 Feb 2015 17:39:51 +0530 Divya Manivannan asm-avx2: filter_vpp[8x16], filter_vps[8x16]: improve 579c->496c, 516c->466c
Wed, 25 Feb 2015 15:04:04 +0530 Divya Manivannan asm-avx2: filter_vpp[16x8], filter_vps[16x8]: improve 567c->530c, 543c->420c
Wed, 25 Feb 2015 14:24:55 +0530 Divya Manivannan asm-avx2: filter_vpp[4x8], filter_vps[4x8]: improve 414c->289c, 353c->250c
Thu, 26 Feb 2015 10:34:07 +0530 Praveen Tiwari Merge with public
Wed, 25 Feb 2015 16:25:43 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_4
Wed, 25 Feb 2015 16:02:10 +0530 Praveen Tiwari asm-avx2: cleanup redundant instruction
Wed, 25 Feb 2015 15:53:31 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_32, improved 435.21c -> 323.25c sse4 asm code
Wed, 25 Feb 2015 14:19:56 +0530 Praveen Tiwari asm-avx2: intra_pred_ang8_33, improved 265.79c -> 185.43c over sse4 asm code
Wed, 25 Feb 2015 13:46:58 +0530 Praveen Tiwari avx2-asm: intra_pred_ang8_3, improved 340.01c -> 280.04c over sse4 asm code
Wed, 25 Feb 2015 10:05:40 +0530 Praveen Tiwari Merge with public
Tue, 24 Feb 2015 17:26:37 +0530 Divya Manivannan asm-avx2: filter_vps[4x4]: improve 201c->156c
Tue, 24 Feb 2015 17:11:03 +0530 Rajesh Paulraj asm-avx2: getResidual, fix 32-bit build error
Tue, 24 Feb 2015 09:25:06 +0530 Divya Manivannan fix: missing constant in filter_vpp[2x4] and filter_vps[2x4]
Tue, 24 Feb 2015 15:34:32 +0530 Praveen Tiwari Merge with public
Mon, 23 Feb 2015 15:28:01 +0530 Divya Manivannan asm-avx2: filter_vpp[8x4], filter_vps[8x4]: improve 213c->161c, 195c->175c
Mon, 23 Feb 2015 14:44:28 +0530 Divya Manivannan asm-avx2: filter_vpp[2x4], filter_vps[2x4]: improve 235c->196c, 214c->174c
Mon, 23 Feb 2015 10:22:39 +0530 Praveen Tiwari Merge with public
Fri, 20 Feb 2015 16:48:00 +0530 Sumalatha Polureddy asm: avx2 code for pixel_add_ps[64x64] - 152x
Fri, 20 Feb 2015 17:55:46 +0530 Divya Manivannan asm: change in loading constant
Fri, 20 Feb 2015 14:20:27 +0530 Sumalatha Polureddy asm: avx2 code for pixel_add_ps[32x32] - 80x
Fri, 20 Feb 2015 13:39:51 +0530 Sumalatha Polureddy asm: avx2 code for pixel_add_ps[16x16] - 55x
Fri, 20 Feb 2015 12:39:44 +0530 Rajesh Paulraj asm-avx2: calcResidual code for 8bbpp and 16bpp
Fri, 20 Feb 2015 10:59:34 +0530 Divya Manivannan fix: error in filter_vps[8x8] assembly pointer
Fri, 20 Feb 2015 10:01:23 +0530 Praveen Tiwari Merge with public
Thu, 19 Feb 2015 17:09:04 +0530 Divya Manivannan asm-avx2: filter_vps[32x32]: improve 4113c->2471c
Thu, 19 Feb 2015 16:04:18 +0530 Divya Manivannan asm-avx2: filter_vps[16x16]: improve 978c->790c
Thu, 19 Feb 2015 15:18:12 +0530 Divya Manivannan asm-avx2: filter_vps[8x8]: improve 296c->263c
Thu, 19 Feb 2015 10:01:48 +0530 Praveen Tiwari Merge with public
Wed, 18 Feb 2015 17:03:02 +0530 Divya Manivannan asm-avx2: luma_vsp[12x16, 4x16], luma_vss[12x16, 4x16] for 16bpp
Wed, 18 Feb 2015 16:05:33 +0530 Divya Manivannan asm-avx2: luma_vsp[4x8, 16x12], luma_vss[4x8, 16x12] for 16bpp
Wed, 18 Feb 2015 15:25:58 +0530 Divya Manivannan asm-avx2: luma_vsp[8x4, 16x4, 32x24], luma_vss[8x4, 16x4, 32x24] for 16bpp
Wed, 18 Feb 2015 15:08:06 +0530 Divya Manivannan asm-avx2: luma_vsp[16x8, 32x8], luma_vss[16x8, 32x8] for 16bpp
Wed, 18 Feb 2015 14:54:02 +0530 Divya Manivannan asm-avx2: luma_vsp[8x16, 8x32], luma_vss[8x16, 8x32] for 16bpp
Wed, 18 Feb 2015 14:27:37 +0530 Divya Manivannan asm-avx2: 16bpp: luma_vsp[64xN, 32x32, 32x64, 48x64, 24x32, 16x32, 16x64]
Wed, 18 Feb 2015 14:00:36 +0530 Divya Manivannan asm-avx2: luma_vsp[16x16, 32x16, 64x16], luma_vss[16x16, 32x16, 64x16] for 16bpp
Wed, 18 Feb 2015 13:36:17 +0530 Divya Manivannan asm-avx2: luma_vsp[4x4, 8x8], luma_vss[4x4, 8x8] for 16bpp
Wed, 18 Feb 2015 11:32:52 +0530 Sumalatha Polureddy asm: avx2 assembly code for 8bpp avg_pixel_pp[16xN]
Wed, 18 Feb 2015 10:31:48 +0530 Praveen Tiwari Merge with public
Tue, 17 Feb 2015 17:46:00 +0530 Divya Manivannan asm-avx2: luma_vsp[32x24], luma_vss[32x24]: improve 13898c->7904c, 10802c->7850c
Tue, 17 Feb 2015 17:25:52 +0530 Divya Manivannan asm-avx2: luma_vsp[16x8, 32x8], luma_vss[16x8, 32x8]: 2507c->1535c, 4655c->2954c,
Tue, 17 Feb 2015 16:46:20 +0530 Divya Manivannan asm-avx2: luma_vsp[16x4], luma_vss[16x4]: improve 1368c->955c, 1194c->963c
Tue, 17 Feb 2015 16:21:03 +0530 Divya Manivannan asm-avx2: luma_vsp[16x12], luma_vss[16x12]: improve 3579c->1988c, 2968c->1992c
Tue, 17 Feb 2015 15:45:07 +0530 Sumalatha Polureddy asm: avx2 assembly code for 8bpp avg_pixel_pp[64xN]
Tue, 17 Feb 2015 15:34:48 +0530 Divya Manivannan asm-avx2: luma_vsp[12x16], luma_vss[12x16]: improve 3753c->2097c, 2957c->1950c
Tue, 17 Feb 2015 15:19:01 +0530 Divya Manivannan asm-avx2: luma_vsp[64xN, 48x64, 24x32, 32x32, 32x64, 16x32, 16x64]
Tue, 17 Feb 2015 14:35:38 +0530 Praveen Tiwari asm-avx2: cpy2Dto1D_shr[8x8, 16x16, 32x32] 16bpp asm code integration
Tue, 17 Feb 2015 14:19:40 +0530 Praveen Tiwari asm-cpy2Dto1D_shr: nits
Tue, 17 Feb 2015 14:15:43 +0530 Sumalatha Polureddy asm: avx2 assembly code for 8bpp avg_pixel_pp[32xN]
Tue, 17 Feb 2015 13:59:34 +0530 Divya Manivannan asm-avx2: luma_vsp[16x16, 32x16, 64x16], luma_vss[16x16, 32x16, 64x16]
Tue, 17 Feb 2015 12:24:04 +0530 Divya Manivannan asm-avx2: luma_vsp[8x4], luma_vss[8x4]: improve 837c->500c, 790c->476c
Tue, 17 Feb 2015 14:11:09 +0530 Praveen Tiwari asm-avx2: cpy2Dto1D_shr_32
Tue, 17 Feb 2015 13:47:54 +0530 Praveen Tiwari asm-avx2: cpy2Dto1D_shr_16
Tue, 17 Feb 2015 12:19:55 +0530 Praveen Tiwari asm-avx2: cpy2Dto1D_shr_8
Tue, 17 Feb 2015 11:59:10 +0530 Sumalatha Polureddy asm: avx assembly code for 10bpp satd[4x32] - 26x
Tue, 17 Feb 2015 11:48:16 +0530 Divya Manivannan asm-avx2: luma_vsp[8x16, 8x32], luma_vss[8x16, 8x32]: 6808c->1310c, 4869c->2505c,
Tue, 17 Feb 2015 10:40:58 +0530 Divya Manivannan asm-avx2: luma_vsp[8x8], luma_vss[8x8]: improve 1422c->763c, 1238c->755c
Tue, 17 Feb 2015 10:08:56 +0530 Praveen Tiwari Merge with public
Mon, 16 Feb 2015 16:59:08 +0530 Divya Manivannan asm-avx2: luma_vsp[4x16], luma_vss[4x16]: improve 1499c->890c, 1282c->725c
Mon, 16 Feb 2015 16:55:54 +0530 Sumalatha Polureddy asm: avx assembly code for 10bpp satd[12x32] - 24x
Mon, 16 Feb 2015 16:25:54 +0530 Sumalatha Polureddy asm: avx assembly code for 10bpp satd[16x24]-21x,satd[32x48]-24x,satd[24x64]-22x,satd[8x64]-22x,satd[8x12]-24x
Mon, 16 Feb 2015 16:08:22 +0530 Divya Manivannan asm-avx2: luma_vsp[4x8], luma_vss[4x8]: improve 835c->457c, 776c->393c
Mon, 16 Feb 2015 12:01:02 +0530 Divya Manivannan asm-avx2: luma_vsp[4x4], luma_vss[4x4]: improve 548c->275c, 510c->240c
Mon, 16 Feb 2015 10:09:11 +0530 Praveen Tiwari Merge with public
Fri, 13 Feb 2015 18:28:52 +0530 Divya Manivannan asm-avx2: luma_vpp[12x16], luma_vps[12x16]: improve 3466c->2182c, 3275c->2057c
Fri, 13 Feb 2015 18:07:46 +0530 Divya Manivannan asm-avx2: luma_vpp[4x16], luma_vps[4x16]: improve 1510c->756c, 1295c->728c
Fri, 13 Feb 2015 17:24:51 +0530 Sumalatha Polureddy asm-avx2: pixel_average_pp[32x32] for 8bpp
Fri, 13 Feb 2015 16:52:55 +0530 Divya Manivannan asm-avx2: luma_vpp[4x8], luma_vps[4x8]: improve 875c->439c, 794c->413c
Fri, 13 Feb 2015 15:48:33 +0530 Divya Manivannan asm-avx2: luma_vpp[16x12], luma_vps[16x12]: improve 3446c->2193c, 3235c->2130c
Fri, 13 Feb 2015 15:21:51 +0530 Divya Manivannan asm-avx2: luma_vpp[8x4], luma_vps[8x4]: improve 815c->520c, 776c->504c
Fri, 13 Feb 2015 15:05:18 +0530 Divya Manivannan asm-avx2: luma_vpp[16x4], luma_vps[16x4]: improve 1350c->1070c, 1240c->974c
Fri, 13 Feb 2015 16:34:40 +0530 Sumalatha Polureddy asm: avx assembly code for satd[4x32] -- 23.4x
Fri, 13 Feb 2015 16:29:37 +0530 Sumalatha Polureddy asm: avx assembly code for satd[12x32] -- 25x
Fri, 13 Feb 2015 18:26:02 +0530 Praveen Tiwari asm-avx2: cpy2Dto1D_shl[8x8, 16x16, 32x32] 16bpp
Fri, 13 Feb 2015 17:59:33 +0530 Praveen Tiwari asm-avx2: cpy2Dto1D_shl_32
Fri, 13 Feb 2015 17:33:58 +0530 Praveen Tiwari asm-avx2: cpy2Dto1D_shl_16
Fri, 13 Feb 2015 16:32:26 +0530 Praveen Tiwari avx2-asm: cpy2Dto1D_shl_8
Fri, 13 Feb 2015 12:41:14 +0530 Divya Manivannan asm-avx2: luma_vpp[32x24], luma_vps[32x24]: improve 12928c->8439c, 11985c->8082c
Fri, 13 Feb 2015 12:04:46 +0530 Divya Manivannan asm-avx2: luma_vpp[32x8, 16x8]: improve 4507c->3043c, 2464c->1558c
Fri, 13 Feb 2015 11:46:48 +0530 Praveen Tiwari asm-avx2: copy_cnt[8x8, 16x16, 32x32] 16bpp
Fri, 13 Feb 2015 11:19:20 +0530 Praveen Tiwari asm-SSE4: copy_cnt [4x4, 8x8, 16x16, 32x32] 16bpp
Fri, 13 Feb 2015 10:25:06 +0530 Praveen Tiwari Merge with public
Thu, 12 Feb 2015 18:04:04 +0530 praveen Tiwari asm-avx2: blockcopy_sp, 16xN, 32xN, 64x64 16bpp
Thu, 12 Feb 2015 17:57:52 +0530 Divya Manivannan asm-avx2: luma_vpp[8x16, 8x32]: improve 2455c->1421c, 4644c->2829c
Thu, 12 Feb 2015 17:40:17 +0530 praveen Tiwari asm-avx2: blockcopy_ps, 16xN, 32xN, 64x64 16bpp
Thu, 12 Feb 2015 17:27:50 +0530 Divya Manivannan asm-avx2: luma_vpp[32x16, 64x16]: improve 8693c->5612c, 17070c->10779c
Thu, 12 Feb 2015 17:04:57 +0530 praveen Tiwari asm-avx2: blockcopy_ss, 16xN, 32xN, 64x64 16bpp
Thu, 12 Feb 2015 16:55:40 +0530 Divya Manivannan asm-avx2: luma_vpp and luma_vps for 16x32, 16x64, 24x32, 32x32, 32x64, 48x64,