[x265] [PATCH] sse: fix data type for sse functions

# HG changeset patch # User Divya Manivannan # Date 1442392516 -19800 # Wed Sep 16 14:05:16 2015 +0530 # Node ID ee39c10fd573444710aca40ec3bd572aac8b3cd0 # Parent 365f7ed4d89628d49cd6af8d81d4edc01f73ffad sse: fix data type for sse functions diff -r 365f7ed4d896

Re: [x265] CU decoding timing

You can also find HEVC decoding sources in the libavcodec project as part of ffmpeg. https://github.com/FFmpeg/FFmpeg/tree/master/libavcodec (hevc*.* etc.) Am 16.09.2015, 12:53 Uhr, schrieb Tom Vaughan : Hi Roshantha, x265 is an HEVC encoder library.

Re: [x265] CU decoding timing

Hi Roshantha, x265 is an HEVC encoder library. There is no such thing as an x265 decoder. There is an open-source HEVC decoder called OpenHEVC. https://github.com/OpenHEVC/openHEVC Best regards, Tom *From:* x265-devel [mailto:x265-devel-boun...@videolan.org] *On Behalf Of *Roshantha

Re: [x265] How can I enable the AVX2 version of DCT and IDCT?

Hi Ximing, If your machine (and OS also) supports AVX2 instruction set then you are already compiling and using AVX2 version of DCT functions. x265 automatically detects & sets all assembly primitives to highest/latest available instruction sets. When you run encoder, you can check command

Re: [x265] How can I enable the AVX2 version of DCT and IDCT?

Thanks! On 9/16/15, Dnyaneshwar Gorade wrote: > Hi Ximing, > > If your machine (and OS also) supports AVX2 instruction set then you are > already compiling and using AVX2 version of DCT functions. > x265 automatically detects & sets all assembly primitives to

[x265] How can I enable the AVX2 version of DCT and IDCT?

I read the source code of the /source/common/vec/dct-sse3.cpp and I found the comments said "Note: We have AVX2 assembly for these functions, but since AVX2 is still somewhat rare on end-user PCs we still compile and link these SSE3 intrinsic SIMD functions". But now both my PC and server support

Re: [x265] [PATCH] sse: fix data type for sse functions

Are you found dynamic range overflow in 8bpp and 10bpp mode? At 2015-09-16 17:11:29,"Divya Manivannan" wrote: ># HG changeset patch ># User Divya Manivannan ># Date 1442392516 -19800 ># Wed Sep 16 14:05:16 2015 +0530 ># Node ID

[x265] [PATCH] fix Main12 satd overflow bug up to SSE4, (fixes #180)

# HG changeset patch # User Min Chen # Date 1442427532 18000 # Node ID 125e4475fc1eb973beac2d00af5acefc441aed71 # Parent ef2ffe88b5f278f2e76f760be61fe29ec50d090f fix Main12 satd overflow bug up to SSE4, (fixes #180) --- source/common/x86/asm-primitives.cpp |2 +-

[x265] [PATCH 2 of 2] fix Main12 satd overflow bug up to SSE4, (fixes #180)

# HG changeset patch # User Min Chen # Date 1442423261 18000 # Node ID 61769ca2358186b36c2498b3eaa8aeded95af6fd # Parent e35d3260b5138abf09bf79808bf55104606c4626 fix Main12 satd overflow bug up to SSE4, (fixes #180) --- source/common/x86/asm-primitives.cpp | 106 ++-

[x265] [PATCH 1 of 2] fix SSE_PP intermedia result overflow in Main12, (fixes #180)

# HG changeset patch # User Min Chen # Date 1442423260 18000 # Node ID e35d3260b5138abf09bf79808bf55104606c4626 # Parent dd77e2f94ec941da1483011e8c382d41ed2afade fix SSE_PP intermedia result overflow in Main12, (fixes #180) --- source/encoder/rdcost.h |4 ++--

Re: [x265] [PATCH 2 of 2] fix Main12 satd overflow bug up to SSE4, (fixes #180)

Please ignore this patch, I forgot cleanup debug code, I send it again. At 2015-09-17 02:14:30,"Min Chen" wrote: ># HG changeset patch ># User Min Chen ># Date 1442423261 18000 ># Node ID 61769ca2358186b36c2498b3eaa8aeded95af6fd ># Parent

Re: [x265] [PATCH] sse: fix data type for sse functions

On 09/16, Divya Manivannan wrote: > # HG changeset patch > # User Divya Manivannan > # Date 1442392516 -19800 > # Wed Sep 16 14:05:16 2015 +0530 > # Node ID ee39c10fd573444710aca40ec3bd572aac8b3cd0 > # Parent 365f7ed4d89628d49cd6af8d81d4edc01f73ffad > sse: fix

Re: [x265] [PATCH] bug: Making windows cpu mask 64-bits

On 09/15, Pradeep Ramachandran wrote: > # HG changeset patch > # User Pradeep Ramachandran > # Date 1442332345 -19800 > # Tue Sep 15 21:22:25 2015 +0530 > # Node ID a103f446053e9992b06e4b522e9fcab62d47ca68 > # Parent 365f7ed4d89628d49cd6af8d81d4edc01f73ffad >

Re: [x265] CU decoding timing

thank you Tom. On 15 September 2015 at 22:33, Roshantha Mendis wrote: > Hi, > > Is the a x265 decoder source available for public ? I only found the > encoder. > > If it is available for public/research use, Please could you also let me > know which function I need to profile

[x265] [PATCH 1 of 2 Update] fix SSE_PP intermedia result overflow in Main12, (fixes #180)

# HG changeset patch # User Min Chen # Date 1442442319 18000 # Node ID b585c3b6834dc3e98b78be5c750047a5e988926a # Parent bc09a5c2aa76a1b9928104a6cbd7cf0c5bc8b00b fix SSE_PP intermedia result overflow in Main12, (fixes #180) --- source/encoder/rdcost.h |4 ++--

[x265] [PATCH 2 of 2 Update] fix Main12 satd overflow bug up to SSE4, (fixes #180)

# HG changeset patch # User Min Chen # Date 1442442321 18000 # Node ID 7ad1738840cb0153b59f952da160f8e9e08dc40a # Parent b585c3b6834dc3e98b78be5c750047a5e988926a fix Main12 satd overflow bug up to SSE4, (fixes #180) --- source/common/x86/asm-primitives.cpp |2 +-