# HG changeset patch
# User Kavitha Sampath kavi...@multicorewareinc.com
# Date 1439450035 -19800
# Thu Aug 13 12:43:55 2015 +0530
# Node ID ac3966a43ba77b609d0e209df80f3a3aac234c08
# Parent bc5a7c2ac38b06d2a232b983f10bc0394d252ad7
stats: fix use of uninitialized pixels in maximum and
Hi Min,
This still generates wrong code for case ymm, xmm (as %ifidni %2,xm will be
false always).
How about the below code ?
%macro vpbroadcastd 2-3 ;; incresed one argument for case ymm, xmm
%ifid %3 ; case vpbroadcastd ymm, ymm, xmm
vpbroadcastd %1, %3
%elifid
# HG changeset patch
# User Deepthi Nandakumar deep...@multicorewareinc.com
# Date 1439460179 -19800
# Thu Aug 13 15:32:59 2015 +0530
# Node ID 9d6ab81073afab38821da2f0b3f9c31fd77df93a
# Parent bc5a7c2ac38b06d2a232b983f10bc0394d252ad7
quantOffsets: move allocation to frame::create
diff -r
# HG changeset patch
# User Kavitha Sampath kavi...@multicorewareinc.com
# Date 1439458769 -19800
# Thu Aug 13 15:09:29 2015 +0530
# Node ID 27226b1c447f4df9cde5b7688f34fb8b7a86e193
# Parent bc5a7c2ac38b06d2a232b983f10bc0394d252ad7
stats: fix use of uninitialized pixels in maximum and
yes deepthi
I will update the change in copyFromPic.
On Thu, Aug 13, 2015 at 4:08 PM, Deepthi Nandakumar
deep...@multicorewareinc.com wrote:
Kavitha,
Can you please pull this neat optimization of Min's into
picyuv::copyFromPicture?
On Thu, Aug 13, 2015 at 8:52 AM, Min Chen
# HG changeset patch
# User Dnyaneshwar G dnyanesh...@multicorewareinc.com
# Date 1439366905 -19800
# Wed Aug 12 13:38:25 2015 +0530
# Node ID 643a001494a42e65366cfa3e468cc0858955095f
# Parent 07110baa95f1d53c8100929b16eafba3b16138d6
asm: AVX2 asm for intra_ang_32 mode 6, improved over 48%
# HG changeset patch
# User Dnyaneshwar G dnyanesh...@multicorewareinc.com
# Date 1439373105 -19800
# Wed Aug 12 15:21:45 2015 +0530
# Node ID c12d411014f68affea550ee640e26ba61f51e509
# Parent 643a001494a42e65366cfa3e468cc0858955095f
asm: AVX2 asm for intra_ang_32 mode 7, improved over 40%
# HG changeset patch
# User Dnyaneshwar G dnyanesh...@multicorewareinc.com
# Date 1439297628 -19800
# Tue Aug 11 18:23:48 2015 +0530
# Node ID 07110baa95f1d53c8100929b16eafba3b16138d6
# Parent bc5a7c2ac38b06d2a232b983f10bc0394d252ad7
asm: AVX2 asm for intra_ang_32 mode 5, improved over 48%
It is right place, we can make a new asm for it, thanks
At 2015-08-13 18:32:15,Deepthi Nandakumar deep...@multicorewareinc.com
wrote:
Min, Kavitha's patch moved maxLumaLevel into copyFromPicture, which is the
right place for this, since maxLumaLevel is calculated from input pic.
On Thu,
On 08/14, Deepthi Nandakumar wrote:
I did a double take on the if (padx) part as well, then figured it works
since padx is guaranteed to be non-zero (because of padx++).
Min, can you do a quick profile on whether it is better to do this outside
the padx loop, for the whole frame, or inside
I did a double take on the if (padx) part as well, then figured it works
since padx is guaranteed to be non-zero (because of padx++).
Min, can you do a quick profile on whether it is better to do this outside
the padx loop, for the whole frame, or inside the padx loop, one row at a
time?
On Thu,
On 08/14, Ximing Cheng wrote:
Hi, the source code in lookahead cuTree function has to calculate the
averageDuration by a while loop each time, why not use double
averageDuration = (double)m_param-fpsDenom / m_param-fpsNum; in the
initial function of lookahead? Thanks!
The cu-tree code was
# HG changeset patch
# User Steve Borho st...@borho.org
# Date 1439523373 -19800
# Fri Aug 14 09:06:13 2015 +0530
# Node ID 25ddcc4f9f0ce7822373a74484ad237bc04e2b78
# Parent bc5a7c2ac38b06d2a232b983f10bc0394d252ad7
getopt: remove double underscores from prototype arguments (fixes #162)
It
# HG changeset patch
# User Steve Borho st...@borho.org
# Date 1439526505 -19800
# Fri Aug 14 09:58:25 2015 +0530
# Branch stable
# Node ID 1b8ff853ad41b3a033b3c98213e9e51f37401b3e
# Parent 567f9b1263a27b9384f10662c169615e62b8e2b3
doc: fix typos and improve readability of threading chapter
Hi, the source code in lookahead cuTree function has to calculate the
averageDuration by a while loop each time, why not use double
averageDuration = (double)m_param-fpsDenom / m_param-fpsNum; in the
initial function of lookahead? Thanks!
___
x265-devel
# HG changeset patch
# User Deepthi Nandakumar deep...@multicorewareinc.com
# Date 1439530515 -19800
# Fri Aug 14 11:05:15 2015 +0530
# Node ID b4ef42c9a18922a48f44799b3d9f8d3e2f6912f2
# Parent 409d6f7461223eddc17db7a45747b2b6450ff10c
picyuv: cleanup, remove always on padx, pady conditionals
16 matches
Mail list logo