Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 27, 2018

• MIPS: Implement ImplSaturatingRoundingMultiplyByPOT<,,-1>

  

  Alexey Frunze

  authored and

  bjacob

  committed
  3fb5c17

  Copy full SHA for 3fb5c17

Commits on Nov 13, 2018

• support encoding sdot as well as udot

  bjacob

  committed
  18e4cc0

  Copy full SHA for 18e4cc0

Commits on Nov 6, 2018

• MIPS: Implement RoundingDivideByPOT().

  

  Alexey Frunze

  authored and

  bjacob

  committed
  75c19af

  Copy full SHA for 75c19af

• MIPS: Improve saturation.

  

  Alexey Frunze

  authored and

  bjacob

  committed
  6db6c2a

  Copy full SHA for 6db6c2a

Commits on Oct 19, 2018

• Fix google#136 (test_blocking_counter)

  

  Alexey Frunze

  authored and

  bjacob

  committed
  8416bab

  Copy full SHA for 8416bab

Commits on Oct 16, 2018

• Import the new kernel contributed by ARM using the new dot-product

  Show description for 7d6d123
  bjacob

  committed
  7d6d123

  Copy full SHA for 7d6d123

Commits on Oct 10, 2018

• tiny fix to encode.py

  bjacob

  committed
  3f87bab

  Copy full SHA for 3f87bab

• further improvements on encode.py

  bjacob

  committed
  f183d88

  Copy full SHA for f183d88

• Minor touchups on standalone programs

  bjacob

  committed
  b1d3c06

  Copy full SHA for b1d3c06

Commits on Oct 9, 2018

• Add a tool to encode udot instructions in asm code so we compile

  Show description for f3c2b90
  bjacob

  committed
  f3c2b90

  Copy full SHA for f3c2b90

Commits on Sep 28, 2018

• make udot kernels build on current toolchains by manually encoding the udot instruction, and add another udot kernel, "_narrow"

  bjacob

  committed
  2c33e33

  Copy full SHA for 2c33e33

Commits on Sep 12, 2018

• Fix a use-after-free issue with the TLS data in the profiler,

  Show description for 7c5f9be
  bjacob

  committed
  7c5f9be

  Copy full SHA for 7c5f9be

• Fix a regression of the profiler introduced in commit

  Show description for 3aad165
  bjacob

  committed
  3aad165

  Copy full SHA for 3aad165

Commits on Sep 4, 2018

• fix comment

  bjacob

  committed
  719139c

  Copy full SHA for 719139c

• Fix undefined behavior in the overflow case in ShiftLeft.

  Show description for d877b64
  bjacob

  committed
  d877b64

  Copy full SHA for d877b64

• lint

  bjacob

  committed
  4f09871

  Copy full SHA for 4f09871

• Fixing missing header. (Patch by Ben Vanik)

  bjacob

  committed
  1080dba

  Copy full SHA for 1080dba

Commits on Aug 30, 2018

• Make gemmlowp multithreading correct:

  Show description for fa27313
  bjacob

  committed
  fa27313

  Copy full SHA for fa27313

Commits on Aug 22, 2018

• Added math header to fixed point for round() declaration

  

  petewarden

  authored and

  bjacob

  committed
  dea287e

  Copy full SHA for dea287e

• Separated platform detection into a new header, so fixed point routines can use the right architecture after previous removal of common.h header

  

  petewarden

  authored and

  bjacob

  committed
  a47fd48

  Copy full SHA for a47fd48

Commits on Aug 21, 2018

• Removed fixed point dependency on wider gemmlowp headers, and added explicit casting for platforms where ints are 16-bits, to help with microcontroller portability

  

  petewarden

  authored and

  bjacob

  committed
  e88547b

  Copy full SHA for e88547b

Commits on Aug 14, 2018

• fix warnings about unused parameters

  

  lee-bin

  authored and

  bjacob

  committed
  07eb865

  Copy full SHA for 07eb865

Commits on Jul 4, 2018

• [MSVC] Enable SSE4 when AVX is turned on

  

  rongjiecomputer

  authored and

  bjacob

  committed
  8724b85

  Copy full SHA for 8724b85

• [Bazel] Don't set -lpthread for Windows

  

  rongjiecomputer

  authored and

  bjacob

  committed
  cd685fe

  Copy full SHA for cd685fe

Commits on Jun 11, 2018

• add missing inline keywords to fix linkage issues

  bjacob

  authored
  6c8f5d4

  Copy full SHA for 6c8f5d4

Commits on Jun 9, 2018

• Add 'cc' and 'memory' to the clobbers list of NEON_64bit_GEMM_Int7Operands_AccumEightWithin16Bits and NEON_64bit_GEMM_Int425Operands kernels since they are being modified.

  

  mariecwhite

  authored and

  bjacob

  committed
  697ceea

  Copy full SHA for 697ceea

Commits on Apr 16, 2018

• Add malloc.h for Windows

  Show description for 38ebac7
  

  yongtang

  authored and

  bjacob

  committed
  38ebac7

  Copy full SHA for 38ebac7

• Fix compilation error in Windows

  Show description for fc0ebb1
  

  yongtang

  authored and

  bjacob

  committed
  fc0ebb1

  Copy full SHA for fc0ebb1

Commits on Apr 10, 2018

• add Richart Winterton to CONTRIBUTORS, clang-format and convert to unix line endings

  bjacob

  committed
  85179c4

  Copy full SHA for 85179c4

• Avx2optimizations (google#122)

  Show description for d74760e
  

  rrwinterton

  authored and

  bjacob

  committed
  d74760e

  Copy full SHA for d74760e

Commits on Apr 5, 2018

• Add profiling support to benchmark_all_sizes

  bjacob

  committed
  ad03972

  Copy full SHA for ad03972

Commits on Mar 30, 2018

• Merge pull request google#132 from jalexstark/fixedpoint-constexpr

  Show description for e08d0f6
  bjacob

  authored
  e08d0f6

  Copy full SHA for e08d0f6

Commits on Mar 29, 2018

• Adjust use of static const / static constexpr in fixedpoint.

  jalexstark

  committed
  c033219

  Copy full SHA for c033219

Commits on Mar 28, 2018

• Merge pull request google#131 from MIPS/msa_width_major

  Show description for 87510e5
  bjacob

  authored
  87510e5

  Copy full SHA for 87510e5

• MIPS: Improve the kernel by switching to width-major.

  

  Alexey Frunze

  authored and

  Alexey Frunze

  committed
  c318e3f

  Copy full SHA for c318e3f

Pagination

Previous

Next