[x265] [PATCH 6 of 6] asm: 10bpp sse4 code for saoCuOrgB0, improved 173346c-23127c over C code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434973065 -19800 # Mon Jun 22 17:07:45 2015 +0530 # Node ID f282fc4d8915ec712f3915f387fc6018481fd467 # Parent a946a0178f57e65c02e6be2d9c6485c58658fe20 asm: 10bpp sse4 code for saoCuOrgB0, improved 173346c

[x265] [PATCH 5 of 6] asm: 10bpp sse4 code for saoCuOrgE3

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434969412 -19800 # Mon Jun 22 16:06:52 2015 +0530 # Node ID a946a0178f57e65c02e6be2d9c6485c58658fe20 # Parent f85c15cc0e1d70e63182b03e294c2778f598143d asm: 10bpp sse4 code for saoCuOrgE3 Performance

[x265] [PATCH 0 of 6 ] SAO SSE4 asm code for HIGH_BIT_DEPTH

SAO_EO_08.97x974.03 8740.81 SAO_EO_110.18x 492.67 5017.42 SAO_EO_1_2Rows 11.21x 900.82 10095.86 SAO_EO_2[0] 6.27x207.22 1298.92 SAO_EO_2[1] 8.92x555.20 4949.69 SAO_EO_3[0] 4.97x236.72 1177.29

[x265] [PATCH 2 of 6] asm: 10bpp sse4 code for saoCuOrgE1, improved 5017c-470c, over C code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434948494 -19800 # Mon Jun 22 10:18:14 2015 +0530 # Node ID 1c02df66f093b5b1bacc8a1bbf9be2ef81591ad5 # Parent a94e9a1f0fde08e060a9b52e3353ce2f242d9257 asm: 10bpp sse4 code for saoCuOrgE1, improved 5017c-470c

[x265] [PATCH 2 of 2] asm: avx2 code for dequant_scaling, improved 11097c-6860c, 38% over SSE4

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434602517 -19800 # Thu Jun 18 10:11:57 2015 +0530 # Node ID 2b69ad2ac441d0782fef529d5c412edd43a0093e # Parent b977f03d9f0fb0811facc9faf926668a031b3105 asm: avx2 code for dequant_scaling, improved 11097c-6860c

[x265] [PATCH 1 of 2] asm: dequant_scaling asm code, improved 12668c-11097c, 12% over intrinsic

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434543335 -19800 # Wed Jun 17 17:45:35 2015 +0530 # Node ID b977f03d9f0fb0811facc9faf926668a031b3105 # Parent d6c32960b5df5b150569d03eb985d9772e494d13 asm: dequant_scaling asm code, improved 12668c-11097c, 12

Re: [x265] [PATCH] asm: avx2 code for pixel_avg[16xN, 32xN] for 10 bpp

On Wed, Jun 17, 2015 at 12:41 PM, sumala...@multicorewareinc.com wrote: # HG changeset patch # User Sumalatha Polureddysumala...@multicorewareinc.com # Date 1434524988 -19800 # Wed Jun 17 12:39:48 2015 +0530 # Node ID 8774c63a4047fa5e54679237aeb88518740efa8d # Parent

[x265] [PATCH] fix issue #144 10-bit x265 hangs from 1.7+170-4948aeae8a18 on Win7 64-bit

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434531332 -19800 # Wed Jun 17 14:25:32 2015 +0530 # Node ID 2641b93d208f56135a93f012f7eda12e8734ebea # Parent 6a0d45aa41260a58971bac62b8f466d9d85c5fa3 fix issue #144 10-bit x265 hangs from 1.7+170-4948aeae8a18

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 16 20

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434360122 -19800 # Mon Jun 15 14:52:02 2015 +0530 # Node ID b9b8a90d0156ae14a61d79030387b10ad551a75f # Parent 9796fe79ac615fce0ed9e8b503ca5069525df625 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 16 20

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 15 21

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434344695 -19800 # Mon Jun 15 10:34:55 2015 +0530 # Node ID 9796fe79ac615fce0ed9e8b503ca5069525df625 # Parent 534ba74791442449ff7d66d5c0bd3ad208af7be4 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 15 21

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 14 22

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434342817 -19800 # Mon Jun 15 10:03:37 2015 +0530 # Node ID 534ba74791442449ff7d66d5c0bd3ad208af7be4 # Parent b3e5163c5525c25fd91f9357677a61b7b18431a9 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 14 22

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 17 19

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434434239 -19800 # Tue Jun 16 11:27:19 2015 +0530 # Node ID 1b0e0ea4238d15a99604a2ebccfa2b0a50d7cff6 # Parent b9b8a90d0156ae14a61d79030387b10ad551a75f asm: 10bpp avx2 code for intra_pred_ang32x32 mode 17 19

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 13 23

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434088811 -19800 # Fri Jun 12 11:30:11 2015 +0530 # Node ID b3e5163c5525c25fd91f9357677a61b7b18431a9 # Parent 200179f64c5b3f611f304758375113bf1cf61c66 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 13 23

[x265] [PATCH 3 of 6] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 15 21

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434450530 -19800 # Tue Jun 16 15:58:50 2015 +0530 # Node ID 3a4442aa079729f676f3e830d225ec4895772dda # Parent fb42b9668b661c7b3f7c207cd9c6172ea5e56715 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 15 21

[x265] [PATCH 4 of 6] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 16 20

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434450593 -19800 # Tue Jun 16 15:59:53 2015 +0530 # Node ID 0501b1c89275c12eb3aa39e2ae300c6ee337a7f0 # Parent 3a4442aa079729f676f3e830d225ec4895772dda asm: 10bpp avx2 code for intra_pred_ang32x32 mode 16 20

[x265] [PATCH 1 of 6] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 13 23

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434450408 -19800 # Tue Jun 16 15:56:48 2015 +0530 # Node ID 273f5f9d858f15be75bc02ba736b6cb35a1c6ff6 # Parent be0ed447922cc81e809d296e75424bb71822aea7 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 13 23

[x265] [PATCH 2 of 6] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 14 22

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434450504 -19800 # Tue Jun 16 15:58:24 2015 +0530 # Node ID fb42b9668b661c7b3f7c207cd9c6172ea5e56715 # Parent 273f5f9d858f15be75bc02ba736b6cb35a1c6ff6 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 14 22

[x265] [PATCH 6 of 6] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 18, improved 1331c-884c, 31%

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 143006 -19800 # Tue Jun 16 14:10:06 2015 +0530 # Node ID a0579532c68b00b2e6a4de667082c4095e4696cf # Parent 55b27d2b3c2863bab07d872679bdb427aea6a78c asm: 10bpp avx2 code for intra_pred_ang32x32 mode 18

[x265] [PATCH 5 of 6] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 17 19

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434450625 -19800 # Tue Jun 16 16:00:25 2015 +0530 # Node ID 55b27d2b3c2863bab07d872679bdb427aea6a78c # Parent 0501b1c89275c12eb3aa39e2ae300c6ee337a7f0 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 17 19

[x265] [PATCH] fix issue #143 x265 is slow when it is build with GCC 5.1

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434433503 -19800 # Tue Jun 16 11:15:03 2015 +0530 # Node ID dfcefd23cf3d2df0d7ffb1dc4fadfede8ae0e33c # Parent 604eda76ccdefdbd9002baae85f1bbfd94b5acd5 fix issue #143 x265 is slow when it is build with GCC 5.1

[x265] [PATCH] fix issue #143 x265 is slow when it is build with GCC 5.1

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434432354 -19800 # Tue Jun 16 10:55:54 2015 +0530 # Node ID 1dd1efa10db3b2a6aa014f4f605d8a3e3c586a7b # Parent 604eda76ccdefdbd9002baae85f1bbfd94b5acd5 fix issue #143 x265 is slow when it is build with GCC 5.1

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 12 24

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1434019907 -19800 # Thu Jun 11 16:21:47 2015 +0530 # Node ID 98065d5141d80130899e88e31cd136e2ca7e47dd # Parent 091940c84ef56d1373d8aa476e8d3f1618436567 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 12 24

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 8 28

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1433933746 -19800 # Wed Jun 10 16:25:46 2015 +0530 # Node ID 6fad2ccf68ce2419290b238fd57c49f64fc62d3e # Parent 8e973542ac0e8c458c18a979a48f3b6cc6b3896a asm: 10bpp avx2 code for intra_pred_ang32x32 mode 8 28

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 7 29

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1433930339 -19800 # Wed Jun 10 15:28:59 2015 +0530 # Node ID 8e973542ac0e8c458c18a979a48f3b6cc6b3896a # Parent 598183ef071dcf88c3a5d48b9d2d6a5063b7b400 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 7 29

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 9, 10, 26 27

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1433937427 -19800 # Wed Jun 10 17:27:07 2015 +0530 # Node ID ba355a7921b4a6adfe1bea6b57699489655b41f8 # Parent 6fad2ccf68ce2419290b238fd57c49f64fc62d3e asm: 10bpp avx2 code for intra_pred_ang32x32 mode 9,10,26

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 6 30

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1433920584 -19800 # Wed Jun 10 12:46:24 2015 +0530 # Node ID 598183ef071dcf88c3a5d48b9d2d6a5063b7b400 # Parent 2188ac83f8f59c9e68bc90c540f3951b74f2a32a asm: 10bpp avx2 code for intra_pred_ang32x32 mode 6 30

[x265] [PATCH] asm: 10bpp avx2 code for intra_pred_ang32x32 mode 5 31

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1433914676 -19800 # Wed Jun 10 11:07:56 2015 +0530 # Node ID 2188ac83f8f59c9e68bc90c540f3951b74f2a32a # Parent 6245476add8f0562e3ccb657f572ff94fe96adf0 asm: 10bpp avx2 code for intra_pred_ang32x32 mode 5 31

[x265] [PATCH] asm: avx2 code for intra_pred_ang32x32 mode 2, 3, 33 34

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1433833289 -19800 # Tue Jun 09 12:31:29 2015 +0530 # Node ID 05df60c805d3f423db573885eb9f27b17dbc12a7 # Parent b252468dde7ffca57da27575388d95ce538945d2 asm: avx2 code for intra_pred_ang32x32 mode 2,3,33 34

[x265] [PATCH] asm: avx2 code for intra_pred_ang32x32 mode 4 32, improved 8709c-4618c, 47%

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1433849586 -19800 # Tue Jun 09 17:03:06 2015 +0530 # Node ID 7cfd60ed1696d523444aa3cfbf9827c34471b25e # Parent 05df60c805d3f423db573885eb9f27b17dbc12a7 asm: avx2 code for intra_pred_ang32x32 mode 4 32

[x265] [PATCH 6 of 6] asm: avx2 code for intra_pred_ang16x16 mode 12 to 24

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432794060 -19800 # Thu May 28 11:51:00 2015 +0530 # Node ID 9b000dc7dea71d4eda3cb6c4bbf9a680f4afadae # Parent 4884b5e59bbebd648fef97bbafd330b70786befe asm: avx2 code for intra_pred_ang16x16 mode 12 to 24

[x265] [PATCH 3 of 6] asm: avx2 code for intra_pred_ang16x16 mode 5, 6, 30 31

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432792569 -19800 # Thu May 28 11:26:09 2015 +0530 # Node ID 9b83bec91d48517119661cd5ee596115d6f17ae1 # Parent 6689bbd6680214c06299e9a59591e9a4923c3401 asm: avx2 code for intra_pred_ang16x16 mode 5,6,30 31

[x265] [PATCH 0 of 6 ] 10bpp avx2 asm code for intra_pred_ang16x16 for all modes 2 to 34

Performance improvement over SSE: intra_ang_16x16[ 2] 356c- 192c, 46% intra_ang_16x16[ 3] 1830c-1080c, 40% intra_ang_16x16[ 4] 1927c-1219c, 36% intra_ang_16x16[ 5] 1918c-1139c, 40% intra_ang_16x16[ 6] 1938c-1142c, 41% intra_ang_16x16[ 7] 1851c-1206c, 35% intra_ang_16x16[

[x265] [PATCH 4 of 6] asm: avx2 code for intra_pred_ang16x16 mode 7, 8, 28 29

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432792824 -19800 # Thu May 28 11:30:24 2015 +0530 # Node ID d5f08c2b2b765f64b10ca847b12bc7a715a528b7 # Parent 9b83bec91d48517119661cd5ee596115d6f17ae1 asm: avx2 code for intra_pred_ang16x16 mode 7,8,28 29

[x265] [PATCH 1 of 6] asm: avx2 code for intra_pred_ang16x16 mode 2 and 34

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432791818 -19800 # Thu May 28 11:13:38 2015 +0530 # Node ID 1af2413e19a29f28872386ff97b041360745daea # Parent 18939c0e321f08207fa0a383939bc44485773013 asm: avx2 code for intra_pred_ang16x16 mode 2 and 34

[x265] [PATCH] asm: avx2 code for sad_48x64, improved 9413c-3789c

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432798447 -19800 # Thu May 28 13:04:07 2015 +0530 # Node ID 399b47e28145c9fe3a1004cf1d5239b7298242ee # Parent 9b000dc7dea71d4eda3cb6c4bbf9a680f4afadae asm: avx2 code for sad_48x64, improved 9413c-3789c diff

[x265] [PATCH] asm: removed duplicate constants

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432102991 -19800 # Wed May 20 11:53:11 2015 +0530 # Node ID 5244b9a0d9a20262c99801a42e346e0b3e07b315 # Parent cdf14fea15a846f2deca436a8e057711607f41bf asm: removed duplicate constants diff -r cdf14fea15a8 -r

Re: [x265] [PATCH] asm: removed some duplicate constants and moved others into const-a.asm

please ignore this patch, I have sent new patch with modified commit message. thanks. On Wed, May 20, 2015 at 11:16 AM, dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432099930 -19800 # Wed May 20 11:02:10 2015

[x265] [PATCH] asm: removed duplicate constants in intrapred8.asm 8bpp, these constants are already defined into const-a.asm

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432106560 -19800 # Wed May 20 12:52:40 2015 +0530 # Node ID e451fccbee625fae6e337892ec7ca6c4e8779490 # Parent 55945e7be95e8fa3006afe0c37894a045daccd4a asm: removed duplicate constants in intrapred8.asm 8bpp

Re: [x265] [PATCH] asm: removed duplicate constants

, dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432102991 -19800 # Wed May 20 11:53:11 2015 +0530 # Node ID 5244b9a0d9a20262c99801a42e346e0b3e07b315 # Parent cdf14fea15a846f2deca436a8e057711607f41bf asm: removed

[x265] [PATCH] asm: removed some duplicate constants in intrapred16.asm 16bpp

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432106399 -19800 # Wed May 20 12:49:59 2015 +0530 # Node ID 55945e7be95e8fa3006afe0c37894a045daccd4a # Parent 9b31a8a7bd57efededcc3884eec09f649394 asm: removed some duplicate constants in intrapred16.asm

[x265] [PATCH] asm: removed some duplicate constants and moved others into const-a.asm

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 143202 -19800 # Tue May 19 15:18:08 2015 +0530 # Node ID b44cdf8dc08c77e84b8707992cd0006bbf23d864 # Parent ac32faec79be9c6a60d267086b4563bd884537c0 asm: removed some duplicate constants and moved others

Re: [x265] [PATCH] asm: removed some duplicate constants and moved others into const-a.asm

Ok. I will resend this patch on latest tip. On Tue, May 19, 2015 at 8:47 PM, Steve Borho st...@borho.org wrote: On 05/19, dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 143202 -19800 # Tue May 19 15:18

[x265] [PATCH] asm: removed some duplicate constants and moved others into const-a.asm

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432099930 -19800 # Wed May 20 11:02:10 2015 +0530 # Node ID cdf14fea15a846f2deca436a8e057711607f41bf # Parent 9b31a8a7bd57efededcc3884eec09f649394 asm: removed some duplicate constants and moved others

[x265] [PATCH] asm: removed duplicate and redundant constants

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1432037856 -19800 # Tue May 19 17:47:36 2015 +0530 # Node ID e6fc4b6f16b32debf4a252b47ad6fc9c82364188 # Parent b44cdf8dc08c77e84b8707992cd0006bbf23d864 asm: removed duplicate and redundant constants diff -r

[x265] [PATCH] asm: addAvg high_bit_depth avx2 asm for chroma sizes width = 8, reused code from luma

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431603734 -19800 # Thu May 14 17:12:14 2015 +0530 # Node ID c237e9a1fd58fb95d6cab19385fb971989907739 # Parent cd88d1705bcb952e2dd2e0f753a510a86909560f asm: addAvg high_bit_depth avx2 asm for chroma sizes width

[x265] [PATCH] asm: addAvg avx2 code for high_bit_depth sizes = 8, improved over ~45% than previous code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431582271 -19800 # Thu May 14 11:14:31 2015 +0530 # Node ID 722ec5cd93ab072a182f0d94ad53ce4e8ad34f94 # Parent 479087422e29a672d6e9bc8d0cd2a65649d71fe2 asm: addAvg avx2 code for high_bit_depth sizes = 8

[x265] [PATCH] asm: intra_pred_dc32 high_bit_depth code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431408860 -19800 # Tue May 12 11:04:20 2015 +0530 # Node ID 5bf9096a4d93fd8c47e9c81e3dafba5d37391114 # Parent f2081ef64fd27dfd3a5bec92ee1a835a74061761 asm: intra_pred_dc32 high_bit_depth code AVX2

[x265] [PATCH] asm: avx2 code for high_bit_depth intra_planar_16x16

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431322630 -19800 # Mon May 11 11:07:10 2015 +0530 # Node ID e112801421b324350f7de388aef19a9f582b4c2b # Parent ac2832c459edc3e1417d6fb62f89203e23484ec8 asm: avx2 code for high_bit_depth intra_planar_16x16 AVX2

[x265] [PATCH] asm: avx2 code for high_bit_depth intra_dc_16x16

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431409071 -19800 # Tue May 12 11:07:51 2015 +0530 # Node ID ac2832c459edc3e1417d6fb62f89203e23484ec8 # Parent 5bf9096a4d93fd8c47e9c81e3dafba5d37391114 asm: avx2 code for high_bit_depth intra_dc_16x16 AVX2

[x265] [PATCH] asm: avx2 code for high_bit_depth intra_pred_planar32x32

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431409427 -19800 # Tue May 12 11:13:47 2015 +0530 # Node ID 4109cf92731a8a6cfe35019d205476e8719d4c67 # Parent e112801421b324350f7de388aef19a9f582b4c2b asm: avx2 code for high_bit_depth intra_pred_planar32x32

[x265] [PATCH] asm: avx2 code for high_bit_depth psyCost_pp_4x4, reduced 400c-250c

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431344809 -19800 # Mon May 11 17:16:49 2015 +0530 # Node ID b64b19125ced7e0f2de57e190e5e14be274e0d7e # Parent 4109cf92731a8a6cfe35019d205476e8719d4c67 asm: avx2 code for high_bit_depth psyCost_pp_4x4, reduced

[x265] [PATCH] asm: enable psyCost_ss avx2 for HIGH_BIT_DEPTH

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1431412445 -19800 # Tue May 12 12:04:05 2015 +0530 # Node ID b8b6c45dce1ee48b212ffa5b86e5e3c3932e4f9a # Parent b64b19125ced7e0f2de57e190e5e14be274e0d7e asm: enable psyCost_ss avx2 for HIGH_BIT_DEPTH diff -r

[x265] [PATCH 2 of 5] asm: avx2 code for high_bit_depth satd_16xN, improved over ~50% than previous asm

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1430988601 -19800 # Thu May 07 14:20:01 2015 +0530 # Node ID 0ea3afb105ecb96db2fe5a2f8a6e19d718a794b4 # Parent 948636c0bbabd45320b451834471d0976cce947b asm: avx2 code for high_bit_depth satd_16xN, improved over

[x265] [PATCH 0 of 5 ] satd AVX2 asm code for HIGH_BIT_DEPTH

this patch series add satd avx2 asm code for following sizes in high_bit_depth: 16x4, 16x8, 16x12, 16x16, 16x32, 16x64 32x8, 32x16, 32x24, 32x32, 32x64 64x16, 64x32, 64x48, 64x64 and 48x64 ___ x265-devel mailing list x265-devel@videolan.org

[x265] [PATCH 5 of 5] asm: avx2 code for high_bit_depth satd_48x64, improved over ~50% than previous asm

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1430990919 -19800 # Thu May 07 14:58:39 2015 +0530 # Node ID e50526d3101e7eecfde8204e0801a7ba335f8f33 # Parent 443ee8d99edf8c3a11f0c1a62f1dd210ee797048 asm: avx2 code for high_bit_depth satd_48x64, improved

[x265] [PATCH 1 of 5] asm: avx2 code for high_bit_depth satd_16x8

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1430986298 -19800 # Thu May 07 13:41:38 2015 +0530 # Node ID 948636c0bbabd45320b451834471d0976cce947b # Parent 7a1fd70739410f874f8ccd51ad688a29364d5e72 asm: avx2 code for high_bit_depth satd_16x8 AVX2: satd

[x265] [PATCH 3 of 5] asm: avx2 code for high_bit_depth satd_32xN, improved over ~50% than previous asm

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1430989736 -19800 # Thu May 07 14:38:56 2015 +0530 # Node ID 9fc3850626b8d2f2b7e28c2ddf70dd043c0fc165 # Parent 0ea3afb105ecb96db2fe5a2f8a6e19d718a794b4 asm: avx2 code for high_bit_depth satd_32xN, improved over

[x265] [PATCH] asm: intra_pred_dc32 high_bit_depth code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1430744823 -19800 # Mon May 04 18:37:03 2015 +0530 # Node ID 9ca2b6bcb92969ddefe75d04f5a6c2caf4fd994b # Parent 4cf55e54fe3ec33f540b7678b02de34074c0527b asm: intra_pred_dc32 high_bit_depth code AVX2

[x265] [PATCH] asm: avx2 code for satd_16xN, improved over ~50% than SSE code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429681758 -19800 # Wed Apr 22 11:19:18 2015 +0530 # Node ID 75344c17ad4ca1b98f8c767e5cb5c0daf51ae5e8 # Parent 86268e498680951069c48b681eef830b0aa37873 asm: avx2 code for satd_16xN, improved over ~50% than SSE

[x265] [PATCH] asm: avx2 code for satd_48x64 and 64xN, improved over ~50% than SSE

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429682259 -19800 # Wed Apr 22 11:27:39 2015 +0530 # Node ID a6010d20ffe96b165af958d5f76b647d7c62dab5 # Parent 75344c17ad4ca1b98f8c767e5cb5c0daf51ae5e8 asm: avx2 code for satd_48x64 and 64xN, improved over ~50

Re: [x265] [PATCH] asm: new optimized algorithm for satd, improved ~30% over previous algorithm

Ok. I will fix the issue and re-submit these patches. On Sat, Apr 18, 2015 at 12:16 AM, Steve Borho st...@borho.org wrote: On 04/16, dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429164512 -19800 # Thu Apr

[x265] [PATCH] asm: new optimized algorithm for satd, improved ~30% over previous algorithm

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429164512 -19800 # Thu Apr 16 11:38:32 2015 +0530 # Node ID 507135d8bcdcb496783c49b4b0304b961a68c253 # Parent f9c0e1f233cc15ccce4eb96adef11583af082f33 asm: new optimized algorithm for satd, improved ~30% over

[x265] [PATCH] asm: avx2 code for satd_48x64 and 64xN, improved over ~100% than SSE

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429173485 -19800 # Thu Apr 16 14:08:05 2015 +0530 # Node ID 04e7526a8bde9e46867f5c4cfb63b98409c7fb44 # Parent ebca2a0d3ab905b62c346d5d0b23d50c618d5827 asm: avx2 code for satd_48x64 and 64xN, improved over ~100

[x265] [PATCH] asm: avx2 code for satd_16xN, improved over ~70% than SSE code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429167173 -19800 # Thu Apr 16 12:22:53 2015 +0530 # Node ID ebca2a0d3ab905b62c346d5d0b23d50c618d5827 # Parent 507135d8bcdcb496783c49b4b0304b961a68c253 asm: avx2 code for satd_16xN, improved over ~70% than SSE

Re: [x265] Fwd: [PATCH] asm: avx2 code for satd_48x64 and 64xN, improved over ~100% than SSE

] [PATCH] asm: avx2 code for satd_48x64 and 64xN, improved over ~100% than SSE To: x265-devel@videolan.org # HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429173485 -19800 # Thu Apr 16 14:08:05 2015 +0530 # Node ID 04e7526a8bde9e46867f5c4cfb63b98409c7fb44

[x265] [PATCH] asm: avx2 code for satd_32xN

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1429001011 -19800 # Tue Apr 14 14:13:31 2015 +0530 # Node ID 5644a5b24ce03290de3a5bb0fc4d49cc00a19ae2 # Parent dd456de98c239b86e29bf349881854a699056240 asm: avx2 code for satd_32xN AVX2: satd[ 32x8

[x265] [PATCH] asm: avx2 code for intra_planar_32x32

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1428041153 -19800 # Fri Apr 03 11:35:53 2015 +0530 # Node ID cef7834897bc0d53981e5dfe8790bc207deb7346 # Parent d9deeef2cdd8f33d543090459f2f8b3e14be296e asm: avx2 code for intra_planar_32x32 AVX2

[x265] [PATCH] asm: avx2 code for intra_dc_32x32

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1428055909 -19800 # Fri Apr 03 15:41:49 2015 +0530 # Node ID 83f44b5a99a1157683d63a18d05297a58437e7a3 # Parent cef7834897bc0d53981e5dfe8790bc207deb7346 asm: avx2 code for intra_dc_32x32 AVX2: intra_dc_32x32[f

[x265] [PATCH] asm: avx2 code for idst4x4

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1427803839 -19800 # Tue Mar 31 17:40:39 2015 +0530 # Node ID cfc64c1861c282a2aa8dec3f0be127eaaf121469 # Parent ffa14b40f0fff3f6f22fe273458f2a4c83c50acf asm: avx2 code for idst4x4 AVX2: idst4x4 7.03x

[x265] [PATCH] asm: avx2 code for dst4x4

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1427803043 -19800 # Tue Mar 31 17:27:23 2015 +0530 # Node ID ffa14b40f0fff3f6f22fe273458f2a4c83c50acf # Parent ac85c775620f1dcb0df056874633cbf916098bd2 asm: avx2 code for dst4x4 AVX2: dst4x4

[x265] [PATCH] replace for loops with memcpy

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1427870017 -19800 # Wed Apr 01 12:03:37 2015 +0530 # Node ID 95c1c565c38373b3edf651f96c4d5f50ed0e01d2 # Parent cfc64c1861c282a2aa8dec3f0be127eaaf121469 replace for loops with memcpy diff -r cfc64c1861c2 -r

[x265] [PATCH] asm: replace movd+pshufd+vinserti128 with vpbroadcastd instruction

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1427695056 -19800 # Mon Mar 30 11:27:36 2015 +0530 # Node ID 4d6848d8a53ba22dec61d5dd61d096a1accfbd2e # Parent 22a312799bb033d40a66fc83a1ac7af192ce2420 asm: replace movd+pshufd+vinserti128 with vpbroadcastd

[x265] [PATCH] asm: psyCost_ss avx2 code for all sizes(4x4, 8x8, 16x16, 32x32, 64x64)

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1427268038 -19800 # Wed Mar 25 12:50:38 2015 +0530 # Node ID af5172cb915de56d0eb57fcfea3b41361c1f76b1 # Parent e637273e2ae6391d43a037d9a4298a8be8e178ad asm: psyCost_ss avx2 code for all sizes

[x265] [PATCH] asm: psyCost_pp avx2 code for BLOCK(8x8, 16x16, 32x32, 64x64)

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1427100052 -19800 # Mon Mar 23 14:10:52 2015 +0530 # Node ID c6d268fedc36710dc77301c45816246a09f10ad7 # Parent 7b66c36ed9effc864d262f2673f4ed941d550365 asm: psyCost_pp avx2 code for BLOCK(8x8,16x16,32x32,64x64

[x265] [PATCH] asm: psyCost_pp avx2 code for BLOCK_4x4

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1427122202 -19800 # Mon Mar 23 20:20:02 2015 +0530 # Node ID 32c2596f4716520fd52d1685b5e2be176b5bc08c # Parent c6d268fedc36710dc77301c45816246a09f10ad7 asm: psyCost_pp avx2 code for BLOCK_4x4 AVX2: psy_cost_pp

[x265] [PATCH] cmake: fix policy CMP0054 warning for cmake version 3.0

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1426854479 -19800 # Fri Mar 20 17:57:59 2015 +0530 # Node ID c78dcd61610f48e1c6106d1ff4fbfb87b2e437b7 # Parent ee667a53aeb475c5d14a75f1d537111e5654ff79 cmake: fix policy CMP0054 warning for cmake version 3.0

[x265] [PATCH] asm: addAvg avx2 code for chroma sizes width = 8, reused code from luma

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1426745611 -19800 # Thu Mar 19 11:43:31 2015 +0530 # Node ID d2b99b5edfde84bd4ad22daaca6f87662d46c2df # Parent 2807d9a5a494de78340ab6d09867205b6676330b asm: addAvg avx2 code for chroma sizes width = 8, reused

[x265] [PATCH] asm: addAvg avx2 code for luma width = 8

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1426744326 -19800 # Thu Mar 19 11:22:06 2015 +0530 # Node ID 2807d9a5a494de78340ab6d09867205b6676330b # Parent cbfa66e0b50cc2393ccbcf6471406504c6c06011 asm: addAvg avx2 code for luma width = 8 AVX2 improvement

Re: [x265] [PATCH 12 of 29] intrapred: updated HIGH_BIT_DEPTH asm and unit test code for intra_planar 4x4, 8x8, 16x16 and 32x32

its part of logic used for intra_planar new asm code, by this new code is 30-50% faster than previous asm On Tue, Jan 13, 2015 at 2:16 PM, chen chenm...@163.com wrote: At 2015-01-13 15:11:20,dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar G dnyanesh

Re: [x265] [PATCH 00 of 29 ] intra refactoring: intra prediction with modified dc, planar and angular prediction functions

I have made patch to remove _new suffix and also looking into Min's suggestions. On Tue, Jan 13, 2015 at 2:54 PM, Steve Borho st...@borho.org wrote: On 01/13, Ashok Kumar Mishra wrote: The patch series is intended for intra optimization. In HEVC implementation, the left neighbouring

Re: [x265] [PATCH 06 of 29] intra_pred_dc_new: updated asm and unit test code

ok Min, I will modify the code accordingly. On Tue, Jan 13, 2015 at 1:59 PM, chen chenm...@163.com wrote: At 2015-01-13 15:11:14,dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1418718739 -19800 # Tue Dec 16 14

[x265] [PATCH 26 of 29] intrapred: remove deprecated all_angs_pred and planar_pred functions and unit test code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1421125486 -19800 # Tue Jan 13 10:34:46 2015 +0530 # Node ID cefe5fd25fe6435445ba1606e39075c4d439c97f # Parent d5e831e36e2acee2c27e1f187351a4d536325c25 intrapred: remove deprecated all_angs_pred and planar_pred

[x265] [PATCH 27 of 29] intrapred: updated asm code for intra_ang_4x4 mode 2 to 34

# HG changeset patch # User Aasaipriya aasaipr...@multicorewareinc.com # Date 1420780470 -19800 # Fri Jan 09 10:44:30 2015 +0530 # Node ID 36a918aff53652e0a5d47e9527a42ec1c673b1e7 # Parent cefe5fd25fe6435445ba1606e39075c4d439c97f intrapred: updated asm code for intra_ang_4x4 mode 2 to 34

[x265] [PATCH 16 of 29] intrapred: updated asm and unit test code for intra_ang_16x16 mode 2 to 34

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1420027241 -19800 # Wed Dec 31 17:30:41 2014 +0530 # Node ID fb5880617c03c89128cdf728c9ca5f0b6bb0837b # Parent 38933694dc0880453bf34a8faee783ba0633e5cf intrapred: updated asm and unit test code

[x265] [PATCH 28 of 29] 16bpp: updated asm code for intrapred_ang_4x4 mode 2 to 34

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1420781569 -19800 # Fri Jan 09 11:02:49 2015 +0530 # Node ID 72b600d94c2a7a446d5dce118e54871dbc463187 # Parent 36a918aff53652e0a5d47e9527a42ec1c673b1e7 16bpp: updated asm code for intrapred_ang_4x4 mode 2 to 34

[x265] [PATCH 29 of 29] intrapred: remove deprecated intrapred angular functions

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1421048129 -19800 # Mon Jan 12 13:05:29 2015 +0530 # Node ID 9ff0b1b684eaea47b204e2bb4c7c987d3a10def0 # Parent 72b600d94c2a7a446d5dce118e54871dbc463187 intrapred: remove deprecated intrapred angular functions

[x265] [PATCH 10 of 29] intra_pred_dc: fix 32-build error, invalid register used

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1418905745 -19800 # Thu Dec 18 17:59:05 2014 +0530 # Node ID b0ffbe3c3bb33eaa898f10bb3ff4086c65605565 # Parent 723d7d1622b8de39d5221f0b8ff82a2a922e7d8b intra_pred_dc: fix 32-build error, invalid register used

[x265] [PATCH 17 of 29] intrapred: updated asm and unit test code for intra_ang_32x32 mode 2 to 34

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1420199144 -19800 # Fri Jan 02 17:15:44 2015 +0530 # Node ID 359749070949b037a78f475d85835dbc0a9c48dd # Parent fb5880617c03c89128cdf728c9ca5f0b6bb0837b intrapred: updated asm and unit test code

[x265] [PATCH 13 of 29] intrapred: cleanup intra_pred_dc and intra_pred_planar older code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1419400410 -19800 # Wed Dec 24 11:23:30 2014 +0530 # Node ID 558439d703fdc46fe47ce315c33fe982f8269a79 # Parent 44a17e7764a38817869d4a58d6764b14841eb2ec intrapred: cleanup intra_pred_dc and intra_pred_planar

[x265] [PATCH 15 of 29] intrapred: fix 32-bit build crash due to stack broken for intra_ang8x8

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1420026015 -19800 # Wed Dec 31 17:10:15 2014 +0530 # Node ID 38933694dc0880453bf34a8faee783ba0633e5cf # Parent 9aa0b1574481b9ef18824f33837d9b1f0e554d8c intrapred: fix 32-bit build crash due to stack broken

[x265] [PATCH 22 of 29] intrapred: updated asm code for all_ang_pred_8x8

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1420634196 -19800 # Wed Jan 07 18:06:36 2015 +0530 # Node ID 652226606b04407d602108192878a7e355d17ed0 # Parent 53dbcfde6422f1e576d63046398b6d4a8c28a437 intrapred: updated asm code for all_ang_pred_8x8 diff -r

[x265] [PATCH 12 of 29] intrapred: updated HIGH_BIT_DEPTH asm and unit test code for intra_planar 4x4, 8x8, 16x16 and 32x32

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1419342091 -19800 # Tue Dec 23 19:11:31 2014 +0530 # Node ID 44a17e7764a38817869d4a58d6764b14841eb2ec # Parent cbeae41ef5e91685ea7126883c269538dace424d intrapred: updated HIGH_BIT_DEPTH asm and unit test code

[x265] [PATCH 06 of 29] intra_pred_dc_new: updated asm and unit test code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1418718739 -19800 # Tue Dec 16 14:02:19 2014 +0530 # Node ID 70b4e0c84320df0b7443e5aea6e110c1bf483684 # Parent f4daa8744d08b569ae652737c4506b397dfb55cb intra_pred_dc_new: updated asm and unit test code diff -r

[x265] [PATCH 07 of 29] high_bit_depth: intra_pred_dc_new updated asm and unit test code

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1418725213 -19800 # Tue Dec 16 15:50:13 2014 +0530 # Node ID 49cb8574e8f4453f700a52dc8a47fadc966a9c0b # Parent 70b4e0c84320df0b7443e5aea6e110c1bf483684 high_bit_depth: intra_pred_dc_new updated asm and unit

[x265] [PATCH 04 of 29] scale1D_128to64: cleanup depricated primitive

# HG changeset patch # User Praveen Tiwari # Date 1421045417 -19800 # Mon Jan 12 12:20:17 2015 +0530 # Node ID 3caab705cfdf1d1d5549f0986bd9496d5fc606e1 # Parent f3fd0d075bba2320ed8cc93df055b144b5a1b88e scale1D_128to64: cleanup depricated primitive diff -r f3fd0d075bba -r 3caab705cfdf

[x265] [PATCH 01 of 29] intra refactoring: removed storing left neighbouring pixels in 144 bytes stride - new

# HG changeset patch # User Ashok Kumar Mishraas...@multicorewareinc.com # Date 1421054362 -19800 # Mon Jan 12 14:49:22 2015 +0530 # Node ID 9ec15ed0fe2a56eaf0c45954b750cac7293ffcd9 # Parent 50a2071500dc4b813edb357c298867931bbf42a1 intra refactoring: removed storing left neighbouring pixels

[x265] [PATCH 03 of 29] scale1D_128to64_new primitive: ASM code and unit test code

# HG changeset patch # User Praveen Tiwari # Date 1421044413 -19800 # Mon Jan 12 12:03:33 2015 +0530 # Node ID f3fd0d075bba2320ed8cc93df055b144b5a1b88e # Parent cf29bf7824491d35e20df5249810ff9a1520d3e3 scale1D_128to64_new primitive: ASM code and unit test code diff -r cf29bf782449 -r

[x265] [PATCH 08 of 29] intra_pred_dc: cleanup, remove unused functions

# HG changeset patch # User Dnyaneshwar G dnyanesh...@multicorewareinc.com # Date 1421047125 -19800 # Mon Jan 12 12:48:45 2015 +0530 # Node ID c238ea33fa51521639404aeb65864a7e4492dcef # Parent 49cb8574e8f4453f700a52dc8a47fadc966a9c0b intra_pred_dc: cleanup, remove unused functions diff -r

[x265] [PATCH 05 of 29] scale1D_128to64_new: cleanup

# HG changeset patch # User Praveen Tiwari # Date 1421046277 -19800 # Mon Jan 12 12:34:37 2015 +0530 # Node ID f4daa8744d08b569ae652737c4506b397dfb55cb # Parent 3caab705cfdf1d1d5549f0986bd9496d5fc606e1 scale1D_128to64_new: cleanup This patch cleanup the new suffix to match with an existing

[x265] [PATCH 00 of 29 ] intra refactoring: intra prediction with modified dc, planar and angular prediction functions

___ x265-devel mailing list x265-devel@videolan.org https://mailman.videolan.org/listinfo/x265-devel

<    1   2   3   4   >