Commits

Branch selector

 master

User selector

fwessels

Datepicker

All time

Commit History

Commits on Aug 23, 2024

• Make ARM SVE code vector length agnostic (#285)

  Show description for 67157af
  fwessels

  authored
  67157af

Commits on Jul 16, 2024

• Fix build for Graviton4 processor (#283)

  Show description for 3412d52
  fwessels

  authored
  3412d52

Commits on Jun 17, 2024

• Support for ARM SVE and better performance for NEON (#281)

  Show description for 7202dfc
  fwessels

  authored
  7202dfc

Commits on May 19, 2020

• Minor refactor for arm NEON version using macros (#147)

  Show description for 01b307e
  fwessels

  authored
  01b307e

Commits on May 15, 2020

• Updated performance numbers for Graviton2 on ARM (#146)

  fwessels

  authored
  6fbce20

Commits on May 14, 2020

• Small optimization to parallal82 for AVX512 by reducing the number of VSHUFI64X2 instructions in the core loop (#143)

  fwessels

  authored
  27f8a7b

• Use proper NEON assembly instructions for ARM (#144)

  Show description for 2475ea7
  fwessels

  authored
  2475ea7

Commits on May 13, 2020

• Take vshufi64x2 out of main loop and initialize upfront (for parallel 81 only) (#139)

  fwessels

  authored
  d6d9fba

• Faster arm64 implementation that does not use PMULL instruction (#140)

  Show description for d5afb5f
  fwessels

  authored
  d5afb5f

Commits on May 7, 2020

• Better test coverage for AVX512 (parallel version) (#134)

  fwessels

  authored
  2f8e50e

Commits on May 6, 2020

• Avx512 parallel81 (#131)

  Show description for 1b9e129
  fwessels

  authored
  1b9e129

Commits on May 3, 2020

• Refactor AVX512 code to use Go assembly instructions. (#121)

  Show description for 0b98f53
  fwessels

  authored
  0b98f53

Commits on Feb 10, 2019

• AVX512 accelerated version resulting in a 4x speed improvement over AVX2 (#91)

  Show description for 79aee05
  

  fwessels

  authored and

  klauspost

  committed
  79aee05

Commits on Dec 18, 2018

• Feature/ppc support (#88)

  Show description for 8885f3a
  

  fwessels

  authored and

  klauspost

  committed
  8885f3a

Commits on Nov 18, 2017

• Use AVX2 SIMD assembly instructions in favor of BYTE sequences. (#73)

  Show description for 3610933
  

  fwessels

  authored and

  klauspost

  committed
  3610933

Commits on Aug 26, 2017

• Add NEON support for ARM64 (#62)

  Show description for 7b88f42
  

  fwessels

  authored and

  klauspost

  committed
  7b88f42

Commits on Jul 20, 2017

• Add ReconstructData interface method (#57)

  Show description for 0de37d7
  

  fwessels

  authored and

  klauspost

  committed
  0de37d7

Commits on Jul 6, 2016

• Add generated byte assembler using asm2plan9s

  Show description for 467733e
  

  fwessels

  authored and

  klauspost

  committed
  467733e

• Removed unnecessary JMP instruction

  fwessels

  committed
  d400006