[x265] [PATCH] asm: avx2 assembly code for dct16

2014-09-08 Thread yuvaraj
# HG changeset patch # User Yuvaraj Venkateshyuva...@multicorewareinc.com # Date 1410159827 -19800 # Mon Sep 08 12:33:47 2014 +0530 # Node ID f097f381abe941bca59584a56d861e53dabbd033 # Parent 8cbfec8d6b4d293a5e7f32e8fef46700b9f1cf6a asm: avx2 assembly code for dct16 diff -r 8cbfec8d6b4d -r

Re: [x265] [PATCH] asm: avx2 assembly code for dct16

2014-09-08 Thread Steve Borho
On 09/08, yuva...@multicorewareinc.com wrote: # HG changeset patch # User Yuvaraj Venkateshyuva...@multicorewareinc.com # Date 1410159827 -19800 # Mon Sep 08 12:33:47 2014 +0530 # Node ID f097f381abe941bca59584a56d861e53dabbd033 # Parent 8cbfec8d6b4d293a5e7f32e8fef46700b9f1cf6a asm:

Re: [x265] [PATCH] x86inc.asm: fix vpbroadcastd bug on Mac platform

2014-09-08 Thread Satoshi Nakagawa
At changeset 27364e9, wrong version pushd. Correct patch is Min's later one. --- a/source/common/x86/x86inc.asm Fri Sep 05 17:36:18 2014 -0700 +++ b/source/common/x86/x86inc.asm Sun Sep 07 15:24:49 2014 +0900 @@ -1489,6 +1489,6 @@ movd %1 %+ xmm, %2 vpbroadcastd

Re: [x265] [PATCH] x86inc.asm: fix vpbroadcastd bug on Mac platform

2014-09-08 Thread Steve Borho
On 09/08, Satoshi Nakagawa wrote: At changeset 27364e9, wrong version pushd. Correct patch is Min's later one. thanks, Satoshi. --- a/source/common/x86/x86inc.asmFri Sep 05 17:36:18 2014 -0700 +++ b/source/common/x86/x86inc.asmSun Sep 07 15:24:49 2014 +0900 @@ -1489,6 +1489,6 @@

Re: [x265] [PATCH] frameencoder: remove second encodeCU() pass over CTUs when SAO is disabled

2014-09-08 Thread Deepthi Nandakumar
On Sat, Sep 6, 2014 at 10:08 PM, Steve Borho st...@borho.org wrote: # HG changeset patch # User Steve Borho st...@borho.org # Date 1409932577 -7200 # Fri Sep 05 17:56:17 2014 +0200 # Node ID 07d69bce1760a28be1b1ee1821dfeb3335602422 # Parent 795878af39730deb24e2ee0e585c625084bb031b

[x265] [PATCH] Analysis: compressIntraCU clean up

2014-09-08 Thread ashok
# HG changeset patch # User Ashok Kumar Mishraas...@multicorewareinc.com # Date 1410162700 -19800 # Mon Sep 08 13:21:40 2014 +0530 # Node ID d0ed8682837d422b15515b3c7c013e4191a48d81 # Parent 8cbfec8d6b4d293a5e7f32e8fef46700b9f1cf6a Analysis: compressIntraCU clean up diff -r 8cbfec8d6b4d -r

[x265] [PATCH] analysis: modified compressInterCU_rd5_6() with CU-specific information

2014-09-08 Thread ashok
# HG changeset patch # User Ashok Kumar Mishraas...@multicorewareinc.com # Date 1410173640 -19800 # Mon Sep 08 16:24:00 2014 +0530 # Node ID 311bd7611ef58bcc36bac80378fd36d3c41fbf36 # Parent d0ed8682837d422b15515b3c7c013e4191a48d81 analysis: modified compressInterCU_rd5_6() with CU-specific

Re: [x265] [PATCH] asm: avx2 assembly code for dct16

2014-09-08 Thread chen
At 2014-09-08 19:22:04,yuva...@multicorewareinc.com wrote: # HG changeset patch # User Yuvaraj venkateshyuva...@multicorewareinc.com # Date 1410175257 -19800 # Mon Sep 08 16:50:57 2014 +0530 # Node ID 74efa467c779ada694ecb9b765b60278aec36950 # Parent

[x265] [PATCH 4 of 4] asm: avx2 version of quant, improve 16.6k cycles - 8.4k cycles

2014-09-08 Thread Min Chen
# HG changeset patch # User Min Chen chenm...@163.com # Date 1410230374 25200 # Node ID 2a36ebe3ee8c48f6ef3d87f3770a64c5bae70953 # Parent 88dd9f5b8b82cc6cc6689979034c4a0635fef473 asm: avx2 version of quant, improve 16.6k cycles - 8.4k cycles diff -r 88dd9f5b8b82 -r 2a36ebe3ee8c

[x265] [PATCH 2 of 4] testbench(quant): the qBits value must be more than or equal to 8

2014-09-08 Thread Min Chen
# HG changeset patch # User Min Chen chenm...@163.com # Date 1410230336 25200 # Node ID 0f1f42504d2d9fdfdf062033252958bd60d682b4 # Parent b00181813b3c2cdd4b13bacfd9600a4626b90baf testbench(quant): the qBits value must be more than or equal to 8 diff -r b00181813b3c -r 0f1f42504d2d

[x265] [PATCH 3 of 4] asm: improve quant by replace variant shift to fixed shift, 19k cycles - 16.6k cycles

2014-09-08 Thread Min Chen
# HG changeset patch # User Min Chen chenm...@163.com # Date 1410230354 25200 # Node ID 88dd9f5b8b82cc6cc6689979034c4a0635fef473 # Parent 0f1f42504d2d9fdfdf062033252958bd60d682b4 asm: improve quant by replace variant shift to fixed shift, 19k cycles - 16.6k cycles diff -r 0f1f42504d2d -r

[x265] [PATCH 1 of 4] testbench(quant): the Round value must be less than (2 ^ qbits)

2014-09-08 Thread Min Chen
# HG changeset patch # User Min Chen chenm...@163.com # Date 1410230321 25200 # Node ID b00181813b3c2cdd4b13bacfd9600a4626b90baf # Parent 033299d2bd00921ffaa403bb404d44694845b2d0 testbench(quant): the Round value must be less than (2 ^ qbits) diff -r 033299d2bd00 -r b00181813b3c

[x265] [PATCH] dpb: select best TMVP candidate from among all of the reference frames

2014-09-08 Thread gopu
# HG changeset patch # User Gopu Govindaswamy g...@multicorewareinc.com # Date 1410236940 -19800 # Tue Sep 09 09:59:00 2014 +0530 # Node ID 88b0307c69c39c1dc9829922d6fcf37d828bd63d # Parent b5f81a83940396c241d33c8f15a5ef48de9cd62e dpb: select best TMVP candidate from among all of the

[x265] [PATCH] copy_cnt replaced align load with unaligned load to avoid code crash, we are not sure about alignment of dst buffer

2014-09-08 Thread praveen
# HG changeset patch # User Praveen Tiwari # Date 1410239785 -19800 # Node ID 91e5893dc024ccfcdec04cef5e8832c80632ee0a # Parent c4fb044c901b8a996f79797be29c894aaf3dcf83 copy_cnt replaced align load with unaligned load to avoid code crash, we are not sure about alignment of dst buffer diff -r