profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/TrianglesPCT/events. GitMemory does not store any data, but only uses NGINX to cache data for a period of time. The idea behind GitMemory is simply to give users a better reading experience.

TrianglesPCT/ALEX 0

A library for building an in-memory, Adaptive Learned indEX

TrianglesPCT/assimp 0

Official Open Asset Import Library Repository. Loads 40+ 3D file formats into one unified and clean data structure.

TrianglesPCT/backward-cpp 0

A beautiful stack trace pretty printer for C++

TrianglesPCT/Beast 0

HTTP and WebSocket built on Boost.Asio in C++11

TrianglesPCT/boost 0

http://svn.boost.org/svn/boost/trunk, http://svn.boost.org/svn/boost/branches/release

TrianglesPCT/bullet3 0

Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects, robotics, machine learning etc.

TrianglesPCT/Chlorine 0

Dead Simple OpenCL (Compute)

TrianglesPCT/CompactNSearch 0

A C++ library to compute neighborhood information for point clouds within a fixed radius. Suitable for many applications, e.g. neighborhood search for SPH fluid simulations.

TrianglesPCT/cuda_voxelizer 0

CUDA voxelizer

pull request commentfacebook/zstd

Enable SSE2 compression path to work on MSVC

All it is doing is normalizing MSVC to target the same code path as GCC/Clang when SSE2 is set as a requirement for your compile target. Not adding any new code..

The defined(__SSE2__) check doesn't work on MSVC because that define doesn't exist even if SSE2 is enabled.

If you want specific performance comparison I can do that might be few days until I have time for that.

TrianglesPCT

comment created time in 2 months

pull request commentfacebook/zstd

Enable SSE2 compression path to work on MSVC

Thanks @TrianglesPCT ,

did you attempt to measure performance differences provided by your PR ?

I removed the AVX2 stuff for a later PR

TrianglesPCT

comment created time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 3723bdd949576570ee1af944f185230befa0a868

Add files via upload avx2...

view details

push time in 2 months

create barnchTrianglesPCT/zstd

branch : AVX2

created branch time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha bee0ef56475eb567b7c801812a1a5f8215b4ffd9

Update zstd_lazy.c It put the changes back when I tried to make a separate pull request, i don't understand githubs interface at all.

view details

push time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha d688ab1e0cfcbe5a894f07bab4033978d99bebd3

Add files via upload AVX2

view details

push time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha bb1cdd8c63046e66de63cf76448868bbc1dc6b72

Update zstd_lazy.c add space

view details

push time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha a62856bf65f381eb2f99d056005b4b39cb7c8725

Update zstd_lazy.c Remove the AVX2 part

view details

push time in 2 months

Pull request review commentfacebook/zstd

Enable SSE2 compression path to work on MSVC, and add AVX2 match find

 static ZSTD_VecMask ZSTD_Vec256_cmpMask8(ZSTD_Vec256 x, ZSTD_Vec256 y) {   sndMask = ZSTD_Vec128_cmpMask8(x.snd, y.snd);   return fstMask | (sndMask << 16); }+#else//AVX2 +typedef struct {+	__m256i v;+} ZSTD_Vec256;++static ZSTD_Vec256 ZSTD_Vec256_read(const void* const ptr) {+	ZSTD_Vec256 v;+    v.v = _mm256_loadu_si256((const __m256i*)ptr);+	return v;+}++static ZSTD_Vec256 ZSTD_Vec256_set8(BYTE val) {+	ZSTD_Vec256 v;+    v.v = _mm256_set1_epi8((char)val);+	return v;+}++static ZSTD_VecMask ZSTD_Vec256_cmpMask8(ZSTD_Vec256 x, ZSTD_Vec256 y) {+    return (ZSTD_VecMask)_mm256_movemask_epi8(_mm256_cmpeq_epi8(x.v, y.v));+}++#endif

Yea I'll remove this and submit it seperate

TrianglesPCT

comment created time in 2 months

PullRequestReviewEvent

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 8f7ea1afeba1f3762e99413424df95b9faf4d2d8

Update zstd_lazy.c Switch to other comment style

view details

push time in 2 months

PullRequestReviewEvent

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 0e071214b5721f3415611d07d33469f8026c3bb0

Update zstd_lazy.c switch to unaligned load as I don't know if buffer will always be aligned to 32 bytes, and compilers aside from MSVC might actually use aligned loads

view details

push time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 69ac124b1209ccf12c8c5969f0d7a2124ccbe554

Update zstd_lazy.c

view details

push time in 2 months

PullRequestReviewEvent
PullRequestReviewEvent

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 0b9f4bb0ff1f313bea9e9166f693ec64a3a6a43e

Update zstd_lazy.c use 8bit

view details

push time in 2 months

PR opened facebook/zstd

Enable SSE2 compression path to work on MSVC, and add AVX2 match find

The compile time detection uses the SSE2 predefined macro, but this does not exist on MSVC--resulting it that compiler always using scalar path.

On MSVC you can detect SSE support by checking for _M_AMD64, as when targeting x64 it requires SSE2 as a baseline, this patch adds this check.

This also adds a AVX2 match finder when AVX2 is predefined.

+25 -3

0 comment

2 changed files

pr created time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 77d54eb3b3116bf9606426730de818f33907aec3

Add files via upload

view details

push time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 52f44bb365337054d30a8e0edf83dd7c612b4d32

Add files via upload msvc

view details

push time in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 25bda9053add6218a58d88c5b8119afa63165231

Add files via upload msvc suport avx2 path

view details

push time in 2 months

fork TrianglesPCT/zstd

Zstandard - Fast real-time compression algorithm

http://www.zstd.net

fork in 2 months

push eventTrianglesPCT/zstd

TrianglesPCT

commit sha 03f2b0d3ad828d7ef99d7a0fbfcd157286918340

Add files via upload The __SSE2__ predfine doe snot exist on MSVC, so the checks to see if supported never work, resulting it always using scalar path. MSVC does support __AVX__ so I added this so at least we can get some perf if AVX is available.

view details

push time in 2 months

startedgroundswellaudio/swl-variant

started time in 3 months