Code

Merge for v2#77

Code

Merged

gquintin merged 200 commits intomasteragenium-scale/nsimd:masterfrom

spmdagenium-scale/nsimd:spmdCopy head branch name to clipboard

Dec 11, 2020

Commits

Commits on Dec 20, 2019

• Start tet1d module

  

  Lénaïc Bagnères

  committed
  3c47a93

  Copy full SHA for 3c47a93

Commits on Jan 6, 2020

• Update tet1d module

  

  Lénaïc Bagnères

  committed
  6e8cf68

  Copy full SHA for 6e8cf68

Commits on Jan 14, 2020

• Add CUDA support for tet1d module

  

  Lénaïc Bagnères

  committed
  480e6a8

  Copy full SHA for 480e6a8

Commits on Jan 17, 2020

• Add scalar input support in tet1d module

  

  Lénaïc Bagnères

  committed
  f32c5e7

  Copy full SHA for f32c5e7

Commits on Jan 20, 2020

• simplify command line arguments for hatch.py

  

  gquintin

  committed
  0d9a2ee

  Copy full SHA for 0d9a2ee

• Add support for generation of modules

  

  gquintin

  committed
  9be62f9

  Copy full SHA for 9be62f9

• Make tet1d a module into the new system

  

  gquintin

  committed
  afc76ac

  Copy full SHA for afc76ac

• Merge branch 'master' into tet_1d

  

  gquintin

  committed
  0816949

  Copy full SHA for 0816949

Commits on Jan 23, 2020

• Add scalar version of some function from libm + reinterpret

  

  gquintin

  committed
  85daa16

  Copy full SHA for 85daa16

• Fix refactoring

  

  gquintin

  committed
  0ccb815

  Copy full SHA for 0ccb815

Commits on Jan 29, 2020

• Before merging master

  

  gquintin

  committed
  d80a8f6

  Copy full SHA for d80a8f6

• Merge branch 'master' into tet_1d

  

  gquintin

  committed
  edc79bd

  Copy full SHA for edc79bd

• Fixes after merge

  

  gquintin

  committed
  20ca2f7

  Copy full SHA for 20ca2f7

Commits on Feb 19, 2020

• For backup

  

  gquintin

  committed
  d0f7c68

  Copy full SHA for d0f7c68

• Fix CUDA

  

  gquintin

  committed
  99f095d

  Copy full SHA for 99f095d

Commits on Feb 20, 2020

• Add forgotten files

  

  gquintin

  committed
  66e775c

  Copy full SHA for 66e775c

Commits on Mar 5, 2020

• Working => backup

  

  gquintin

  committed
  d5e973c

  Copy full SHA for d5e973c

Commits on Mar 6, 2020

• Fixes

  

  gquintin

  committed
  69160ec

  Copy full SHA for 69160ec

Commits on Mar 13, 2020

• All tests are passing f16 included

  

  gquintin

  committed
  c378928

  Copy full SHA for c378928

Commits on Mar 15, 2020

• For backup

  

  gquintin

  committed
  7c26ab2

  Copy full SHA for 7c26ab2

Commits on Mar 17, 2020

• COVID-19

  

  gquintin

  committed
  46acd2f

  Copy full SHA for 46acd2f

Commits on Mar 18, 2020

• Fixes

  

  gquintin

  committed
  32e25bc

  Copy full SHA for 32e25bc

Commits on Mar 19, 2020

• Fixes

  

  gquintin

  committed
  d8bd09a

  Copy full SHA for d8bd09a

Commits on Mar 23, 2020

• Fixes: all test compile with nvcc

  

  gquintin

  committed
  a9a6490

  Copy full SHA for a9a6490

Commits on Mar 29, 2020

• ROCm support, addition on f32 and f16 are compiling

  

  gquintin

  committed
  1d5c842

  Copy full SHA for 1d5c842

Commits on Mar 30, 2020

• TET1D tests are compiling with both nvcc and hipcc

  

  gquintin

  committed
  0e33332

  Copy full SHA for 0e33332

Commits on Apr 1, 2020

• Merge CUDA and ROCm when code is the same

  

  gquintin

  committed
  f960e51

  Copy full SHA for f960e51

• Forgot files

  

  gquintin

  committed
  7832e1b

  Copy full SHA for 7832e1b

• Merge branch 'master' into tet_1d

  

  gquintin

  committed
  3ad694e

  Copy full SHA for 3ad694e

• Now we can list generated files

  

  gquintin

  committed
  e6c7c09

  Copy full SHA for e6c7c09

• Forgot to merge nsimd.h

  

  gquintin

  committed
  8882b39

  Copy full SHA for 8882b39

Commits on Apr 4, 2020

• Forgot to push

  

  gquintin

  committed
  ec4d0d3

  Copy full SHA for ec4d0d3

Commits on Apr 6, 2020

• Update .gitignore with the new file generated by the tet1d module.

  

  Paul Gannay

  committed
  7b07e48

  Copy full SHA for 7b07e48

• Return allocated arrays for tests

  

  Paul Gannay

  committed
  58bbd4b

  Copy full SHA for 58bbd4b

• Increase the minimum size of the tests array

  

  Paul Gannay

  committed
  d294984

  Copy full SHA for d294984

Commits on Apr 7, 2020

• Fix segfault

  

  gquintin

  committed
  0b537ad

  Copy full SHA for 0b537ad

• Fix segfault

  

  gquintin

  committed
  a7db980

  Copy full SHA for a7db980

Commits on Apr 12, 2020

• Add mask[oz]_load[zu] and mask_store[au] operators for CPU

  

  gquintin

  committed
  8125024

  Copy full SHA for 8125024

Commits on Apr 13, 2020

• For backup

  

  gquintin

  committed
  113f303

  Copy full SHA for 113f303

Commits on Apr 14, 2020

• For backup

  

  gquintin

  committed
  82a48a1

  Copy full SHA for 82a48a1

Commits on Apr 16, 2020

• Fix for SSE

  

  gquintin

  committed
  f4c0987

  Copy full SHA for f4c0987

• Fix fma for C89

  

  gquintin

  committed
  d93e130

  Copy full SHA for d93e130

• Remove warning from GCC when using long long in C98 and C++98

  

  gquintin

  committed
  6b37c3b

  Copy full SHA for 6b37c3b

• Fix warnings for C98 and C++98 and AVX512

  

  Guillaume Quintin

  committed
  d174b63

  Copy full SHA for d174b63

• Add set1l, iota, mask_for_loop_tail for ARM

  

  Guillaume Quintin

  committed
  74e48fe

  Copy full SHA for 74e48fe

Commits on Apr 18, 2020

• Before merging master

  

  Guillaume Quintin

  committed
  53f31f6

  Copy full SHA for 53f31f6

• Merge branch 'master' into spmd

  

  Guillaume Quintin

  committed
  ec793e8

  Copy full SHA for ec793e8

• Fix ARM mask[oz]_load[au]

  

  Guillaume Quintin

  committed
  47be0f9

  Copy full SHA for 47be0f9

• Fixes for ARM SVE

  

  Guillaume Quintin

  committed
  eb854f9

  Copy full SHA for eb854f9

• Fix warning when using __f16's

  

  Guillaume Quintin

  committed
  a834608

  Copy full SHA for a834608

Commits on Apr 20, 2020

• Add alignment-templated masked loads/stores

  

  Guillaume Quintin

  committed
  901140f

  Copy full SHA for 901140f

Commits on Apr 22, 2020

• Rewrite friendly_but_not_optimized stuff

  

  Guillaume Quintin

  committed
  b1b65de

  Copy full SHA for b1b65de

• Forgot file

  

  Guillaume Quintin

  committed
  f46e44c

  Copy full SHA for f46e44c

• Fix ARM

  

  Guillaume Quintin

  committed
  f150f43

  Copy full SHA for f150f43

• Fix ARM

  

  Guillaume Quintin

  committed
  82219ff

  Copy full SHA for 82219ff

Commits on Apr 23, 2020

• Cosmetic

  

  Guillaume Quintin

  committed
  7da09ed

  Copy full SHA for 7da09ed

Commits on Apr 25, 2020

• Backup

  

  gquintin

  committed
  ab2430d

  Copy full SHA for ab2430d

Commits on Apr 30, 2020

• Backup

  

  gquintin

  committed
  b4a5321

  Copy full SHA for b4a5321

Commits on May 6, 2020

• Backup

  

  gquintin

  committed
  27ae566

  Copy full SHA for 27ae566

Commits on May 8, 2020

• Backup

  

  gquintin

  committed
  a4a4c5f

  Copy full SHA for a4a4c5f

• Forgot file

  

  gquintin

  committed
  41bb6ee

  Copy full SHA for 41bb6ee

Commits on May 9, 2020

• For backup

  

  gquintin

  committed
  125751b

  Copy full SHA for 125751b

• For backup

  

  gquintin

  committed
  a8854e7

  Copy full SHA for a8854e7

Commits on May 16, 2020

• Refactoring of documentation

  

  gquintin

  committed
  c3a7358

  Copy full SHA for c3a7358

Commits on May 17, 2020

• Add build.nsconfig + fix warning in fixed_point exp

  

  gquintin

  committed
  91f3c31

  Copy full SHA for 91f3c31

• Fix warning in SPMd module

  

  gquintin

  committed
  576fe1f

  Copy full SHA for 576fe1f

Commits on May 19, 2020

• Add forgotten file

  

  gquintin

  committed
  4a13a61

  Copy full SHA for 4a13a61

Commits on May 21, 2020

• Fixes for CUDA

  

  gquintin

  committed
  f5202c7

  Copy full SHA for f5202c7

• Fixes for CPU

  

  gquintin

  committed
  8d128ba

  Copy full SHA for 8d128ba

Commits on May 22, 2020

• Fixes

  

  gquintin

  committed
  3cf0a18

  Copy full SHA for 3cf0a18

Commits on May 28, 2020

• Add gather/scatter for cpu and x86

  

  gquintin

  committed
  0afe30c

  Copy full SHA for 0afe30c

• Add gather/scatter for arm (not tested yet)

  

  gquintin

  committed
  0bcb4f9

  Copy full SHA for 0bcb4f9

Commits on May 29, 2020

• Fix gather/scatter for arm

  

  Guillaume Quintin

  committed
  ac3387a

  Copy full SHA for ac3387a

• Deactivate tet1d module

  

  Guillaume Quintin

  committed
  dfe3571

  Copy full SHA for dfe3571

• Cleanup

  

  Guillaume Quintin

  committed
  91b68ef

  Copy full SHA for 91b68ef

Commits on Jun 2, 2020

• Add scripts for building

  

  Guillaume Quintin

  committed
  57a7a29

  Copy full SHA for 57a7a29

Commits on Jun 3, 2020

• Fix setup and build script for Linux

  

  gquintin

  committed
  f9773af

  Copy full SHA for f9773af

• Changing computer

  

  gquintin

  committed
  c8b1d2e

  Copy full SHA for c8b1d2e

Commits on Jun 4, 2020

• Backup

  

  gquintin

  committed
  fd0bc60

  Copy full SHA for fd0bc60

Commits on Jun 5, 2020

• Fix script/setup.sh

  

  gquintin

  committed
  f9ce017

  Copy full SHA for f9ce017

• Fixes for fixed size SVE

  

  Guillaume Quintin

  committed
  a8229ac

  Copy full SHA for a8229ac

Commits on Jun 6, 2020

• Fix Windows scripts

  

  gquintin

  committed
  46a254a

  Copy full SHA for 46a254a

Commits on Jun 7, 2020

• Fix scripts for Linux

  

  gquintin

  committed
  7470b10

  Copy full SHA for 7470b10

Commits on Jun 8, 2020

• Fix Makefile.nix for md2html

  

  gquintin

  committed
  8a01c6d

  Copy full SHA for 8a01c6d

• Fix Makefile.win for md2html

  

  gquintin

  committed
  747e978

  Copy full SHA for 747e978

• Fix generation of documentation

  

  gquintin

  committed
  891cf19

  Copy full SHA for 891cf19

Commits on Jun 12, 2020

• Add mask scatter for cpu

  

  gquintin

  committed
  4b3907f

  Copy full SHA for 4b3907f

Commits on Jun 13, 2020

• Add mask_scatter for x86

  

  gquintin

  committed
  808ecf1

  Copy full SHA for 808ecf1

• Forgot a file

  

  gquintin

  committed
  0ef6e02

  Copy full SHA for 0ef6e02

• Add mask_scatter for arm

  

  Guillaume Quintin

  committed
  7dfdc10

  Copy full SHA for 7dfdc10

Commits on Jun 15, 2020

• Add masked gather for cpu

  

  gquintin

  committed
  9261960

  Copy full SHA for 9261960

Commits on Jun 16, 2020

• Add masked gather for x86

  

  gquintin

  committed
  fec28da

  Copy full SHA for fec28da

• Add masked gather for arm

  

  Guillaume Quintin

  committed
  a5391bd

  Copy full SHA for a5391bd

• Fix masked gather for f16's

  

  Guillaume Quintin

  committed
  d68aec6

  Copy full SHA for d68aec6

• Adapt SVE typedefs to new GCC 10

  

  Guillaume Quintin

  committed
  42aeac6

  Copy full SHA for 42aeac6

Commits on Jun 17, 2020

• Fixes for x86

  

  gquintin

  committed
  19ab895

  Copy full SHA for 19ab895

• Fix tet1d tests for CUDA

  

  gquintin

  committed
  dd1cd4a

  Copy full SHA for dd1cd4a

Commits on Jun 18, 2020

• Fixes for HIP

  

  gquintin

  committed
  fe21e8d

  Copy full SHA for fe21e8d

• Fix warning fr ROCm/HIP

  

  gquintin

  committed
  1ee8081

  Copy full SHA for 1ee8081

Commits on Jun 19, 2020

• Various fixes

  

  Guillaume Quintin

  committed
  7d1628f

  Copy full SHA for 7d1628f

• Fix tests for rec11, rec8, rsqrt11 and rsqrt8

  

  Guillaume Quintin

  committed
  b9e8377

  Copy full SHA for b9e8377

Commits on Jun 20, 2020

• Fix rec11, rec8, rsqrt11, rsqrt8 tests

  

  Guillaume Quintin

  committed
  7982cc7

  Copy full SHA for 7982cc7

• Improve gather/scatter for neon128 and aarch64

  

  Guillaume Quintin

  committed
  127ff98

  Copy full SHA for 127ff98

Commits on Jun 24, 2020

• Add gather_linear + scatter_linear and remove masked gather and scatter

  

  Guillaume Quintin

  committed
  86abc6f

  Copy full SHA for 86abc6f

• Add linear gather + scatter

  

  Guillaume Quintin

  committed
  0eb97d3

  Copy full SHA for 0eb97d3

Commits on Jun 25, 2020

• Fix gather_linear for neon128 + aarch64

  

  Guillaume Quintin

  committed
  4a4b58f

  Copy full SHA for 4a4b58f

• Improve gather on aarch64 + neon128

  

  Guillaume Quintin

  committed
  de5c778

  Copy full SHA for de5c778

Commits on Jun 29, 2020

• Add documentation for module TET1d

  

  gquintin

  committed
  2ef6154

  Copy full SHA for 2ef6154

• Update README

  

  gquintin

  committed
  9b96ee9

  Copy full SHA for 9b96ee9

• Add documentation for module TET1d

  

  gquintin

  committed
  48a3e13

  Copy full SHA for 48a3e13

Commits on Jun 30, 2020

• Improve README with nsconfig stuff

  

  gquintin

  committed
  f321ea5

  Copy full SHA for f321ea5

• Improve README

  

  gquintin

  committed
  79cb600

  Copy full SHA for 79cb600

• Improve README

  

  gquintin

  committed
  0ca1cb1

  Copy full SHA for 0ca1cb1

• Improve README

  

  gquintin

  committed
  f34c03f

  Copy full SHA for f34c03f

Commits on Jul 1, 2020

• Improve README

  

  gquintin

  committed
  45a824b

  Copy full SHA for 45a824b

Commits on Jul 2, 2020

• Fix warning for armclang

  

  Guillaume Quintin

  committed
  7fe1b71

  Copy full SHA for 7fe1b71

Commits on Jul 13, 2020

• Fix warning when compiling with Clang and C++98/03

  

  Guillaume Quintin

  committed
  a9c7f4b

  Copy full SHA for a9c7f4b

Commits on Jul 14, 2020

• Fix generation of benches

  

  gquintin

  committed
  08820bb

  Copy full SHA for 08820bb

Commits on Jul 17, 2020

• For backup

  

  c

  committed
  7b3e748

  Copy full SHA for 7b3e748

Commits on Jul 18, 2020

• First version (not finished yet)

  

  c

  committed
  2d36e78

  Copy full SHA for 2d36e78

Commits on Jul 21, 2020

• Add support for non closed operators

  

  c

  committed
  c0f1bdd

  Copy full SHA for c0f1bdd

Commits on Jul 26, 2020

• Improve doc

  

  gquintin

  committed
  d105b1a

  Copy full SHA for d105b1a

Commits on Aug 1, 2020

• Improve documentation

  

  gquintin

  committed
  e73920c

  Copy full SHA for e73920c

• More fixes

  

  gquintin

  committed
  2c3da76

  Copy full SHA for 2c3da76

Commits on Aug 2, 2020

• Fix broken link in README

  

  gquintin

  committed
  01aa9ef

  Copy full SHA for 01aa9ef

Commits on Aug 3, 2020

• Add CONTRIBUTING.md

  

  gquintin

  committed
  c9d667f

  Copy full SHA for c9d667f

Commits on Aug 6, 2020

• Improve documentation

  

  gquintin

  committed
  339a7f1

  Copy full SHA for 339a7f1

• Improve documentation

  

  gquintin

  committed
  fcc7b21

  Copy full SHA for fcc7b21

• Improve documentation + simplify scoped_aligned_mem_for

  

  gquintin

  committed
  de049b1

  Copy full SHA for de049b1

Commits on Aug 7, 2020

• Fix scoped_aligned_mem

  

  gquintin

  committed
  7609ac0

  Copy full SHA for 7609ac0

• Fixed errors in nsimd.h

  

  Adrien Arnaud

  committed
  b6b0b94

  Copy full SHA for b6b0b94

Commits on Aug 11, 2020

• Improve documentation

  

  gquintin

  committed
  b4404cc

  Copy full SHA for b4404cc

Commits on Aug 12, 2020

• Improve documentation

  

  gquintin

  committed
  a3fcd97

  Copy full SHA for a3fcd97

Commits on Aug 13, 2020

• Improve documentation

  

  gquintin

  committed
  946651c

  Copy full SHA for 946651c

• Replace some print left by common.myprint

  

  gquintin

  committed
  72418bd

  Copy full SHA for 72418bd

Commits on Aug 24, 2020

• Fixed multiple declarations

  

  Adrien Arnaud

  committed
  fa0625d

  Copy full SHA for fa0625d

• Let benches generate despite the new function set1l

  

  Paul Gannay

  committed
  0b4f299

  Copy full SHA for 0b4f299

• Add a module offering a vectorized random generator

  

  Paul Gannay

  committed
  58f1854

  Copy full SHA for 58f1854

Commits on Aug 26, 2020

• Only generate rand module if flags passed from hatch are correct

  

  Paul Gannay

  committed
  f77a5d1

  Copy full SHA for f77a5d1

Commits on Sep 14, 2020

• Removed F-strings

  

  aarnaud

  committed
  211428c

  Copy full SHA for 211428c

Commits on Sep 17, 2020

• Fix build.nsconfig

  rodolphe-c

  authored
  28a5c9e

  Copy full SHA for 28a5c9e

Commits on Sep 21, 2020

• Fix generation of rand module

  

  gquintin

  committed
  4c4df4a

  Copy full SHA for 4c4df4a

• Building the library does not require C++14 anymore, C++98 is more than sufficient

  

  gquintin

  committed
  0fcfa9a

  Copy full SHA for 0fcfa9a

• Update README

  

  gquintin

  committed
  1dfb24d

  Copy full SHA for 1dfb24d

• Update README

  

  gquintin

  committed
  8f27c7d

  Copy full SHA for 8f27c7d

Commits on Sep 29, 2020

• Setup.sh clone nstools using the same protocol as nsimd

  

  gquintin

  committed
  e468fb6

  Copy full SHA for e468fb6

• Add possibility to ignore tests/benches/...

  

  gquintin

  committed
  1d1d00d

  Copy full SHA for 1d1d00d

Commits on Oct 12, 2020

• Add C++20 concepts to nsimd.h

  

  gquintin

  committed
  1ad8ed2

  Copy full SHA for 1ad8ed2

• Add C++20 concepts to cxx_adv_api.hpp

  

  gquintin

  committed
  f317ebb

  Copy full SHA for f317ebb

• Add C++20 concepts to Python-generated functions

  

  gquintin

  committed
  3624827

  Copy full SHA for 3624827

Commits on Oct 13, 2020

• Fix C++20 concepts

  

  gquintin

  committed
  fdc1ef9

  Copy full SHA for fdc1ef9

Commits on Oct 14, 2020

• Prepare support for oneAPI

  

  gquintin

  committed
  e55a7ea

  Copy full SHA for e55a7ea

• Add C++20 concepts doc

  

  gquintin

  committed
  f7d75d9

  Copy full SHA for f7d75d9

• Modify the rand module to allow generation with python 3.5 and earlier

  

  Paul Gannay

  committed
  c0bdf3b

  Copy full SHA for c0bdf3b

Commits on Oct 21, 2020

• Improve doc + rename module rand --> random

  

  gquintin

  committed
  b46f330

  Copy full SHA for b46f330

• Fix menu of doc of random module

  

  gquintin

  committed
  a07141e

  Copy full SHA for a07141e

Commits on Oct 23, 2020

• Fix availability of scoped_mem...

  

  gquintin

  committed
  ad45c85

  Copy full SHA for ad45c85

Commits on Oct 27, 2020

• Fix tests to_pack*

  

  gquintin

  committed
  eecde37

  Copy full SHA for eecde37

Commits on Nov 10, 2020

• Tests are dependant of the SIMD architecture

  

  gquintin

  committed
  a2f31c1

  Copy full SHA for a2f31c1

Commits on Nov 13, 2020

• Improvements for Intel + Fixes for KNL

  

  Guillaume Quintin

  committed
  a3eaf08

  Copy full SHA for a3eaf08

Commits on Nov 15, 2020

• More fixes for KNL and C89

  

  Guillaume Quintin

  committed
  ad1e727

  Copy full SHA for ad1e727

Commits on Nov 22, 2020

• More fixes

  

  gquintin

  committed
  917c3cc

  Copy full SHA for 917c3cc

Commits on Nov 27, 2020

• Fix fms/fnms for aarch64

  

  Guillaume Quintin

  committed
  28e6344

  Copy full SHA for 28e6344

• Fixes for SVE

  

  Guillaume Quintin

  committed
  60dd572

  Copy full SHA for 60dd572

• Fix warning whe compiling for 32-bits targets

  

  Guillaume Quintin

  committed
  c6ca5a7

  Copy full SHA for c6ca5a7

Commits on Nov 29, 2020

• Cleaning in tests generation

  

  Guillaume Quintin

  committed
  e59e4e1

  Copy full SHA for e59e4e1

• Fix ULP bounds for some operators

  

  Guillaume Quintin

  committed
  0fe0ef2

  Copy full SHA for 0fe0ef2

Commits on Nov 30, 2020

• Almost all tests are passing on 32-bits platform

  

  Guillaume Quintin

  committed
  415eca2

  Copy full SHA for 415eca2

• No more warning for 32-bits compilations

  

  Guillaume Quintin

  committed
  30faa8a

  Copy full SHA for 30faa8a

• Forgot a file

  

  Guillaume Quintin

  committed
  2835544

  Copy full SHA for 2835544

• Fix last errors in philox

  

  gquintin

  committed
  f683f27

  Copy full SHA for f683f27

• First version of quick'n'dirty CI

  

  gquintin

  committed
  6fb3f46

  Copy full SHA for 6fb3f46

Commits on Dec 2, 2020

• Fix warnings

  

  gquintin

  committed
  72870bb

  Copy full SHA for 72870bb

Commits on Dec 3, 2020

• Fix more warnings

  

  gquintin

  committed
  98395f4

  Copy full SHA for 98395f4

• Fix Pyhon generation for module/random

  

  gquintin

  committed
  0b1fbea

  Copy full SHA for 0b1fbea

• Fix fnms for SSE2 and SSE42

  

  gquintin

  committed
  0b99216

  Copy full SHA for 0b99216

• Try again to fix warnings for GCC

  

  gquintin

  committed
  5c5d72c

  Copy full SHA for 5c5d72c

Commits on Dec 5, 2020

• Fix warnings for Clang

  

  gquintin

  committed
  c25e77f

  Copy full SHA for c25e77f

Commits on Dec 6, 2020

• Add variable to compile for a given CUDA GPU

  

  gquintin

  committed
  8176137

  Copy full SHA for 8176137

• Fix warnings for ROCm/HIP

  

  gquintin

  committed
  def6d79

  Copy full SHA for def6d79

Commits on Dec 7, 2020

• Fix CUDA f16 implementation

  

  gquintin

  committed
  a9c3bdd

  Copy full SHA for a9c3bdd

• Fix CUDA f16 implementation

  

  gquintin

  committed
  75388b0

  Copy full SHA for 75388b0

• Fix CUDA f16 implementation

  

  gquintin

  committed
  37d1689

  Copy full SHA for 37d1689

• Reduce size of arrays for GPU testing

  

  gquintin

  committed
  f620ba9

  Copy full SHA for f620ba9

• Reduce size of arrays for GPU testing

  

  gquintin

  committed
  34f5bed

  Copy full SHA for 34f5bed

Commits on Dec 8, 2020

• Compile .so with nvcc and hipcc for binary compatibility

  

  gquintin

  committed
  cd895c6

  Copy full SHA for cd895c6

• Fix build.nsconfig

  

  gquintin

  committed
  6a49e6e

  Copy full SHA for 6a49e6e

• Fix build.nsconfig

  

  gquintin

  committed
  2aee251

  Copy full SHA for 2aee251

• Fix build.nsconfig

  

  gquintin

  committed
  a398f38

  Copy full SHA for a398f38

• Fix build.nsconfig

  

  gquintin

  committed
  0bcb59e

  Copy full SHA for 0bcb59e

Commits on Dec 9, 2020

• Improve CI script + add static in NSIMD_INLINE

  

  gquintin

  committed
  536b622

  Copy full SHA for 536b622

• Fix build.nsconfig for HIP

  

  gquintin

  committed
  2afdc1d

  Copy full SHA for 2afdc1d

• Last fixes

  

  gquintin

  committed
  e40e522

  Copy full SHA for e40e522

• Fix issue: __popcnt64 not available in 32-bits mode

  

  gquintin

  committed
  6ddd56d

  Copy full SHA for 6ddd56d

• Fix DLL specifier of *logulps*

  

  gquintin

  committed
  1b01abb

  Copy full SHA for 1b01abb

• Fix MSVC 32-bits related issues

  

  gquintin

  committed
  2b9059d

  Copy full SHA for 2b9059d

Commits on Dec 11, 2020

• Cosmetic

  

  gquintin

  committed
  3fbaea3

  Copy full SHA for 3fbaea3

• Add __vectorcall for MSVC 32-bits

  

  gquintin

  committed
  018b487

  Copy full SHA for 018b487

• Update .gitignore

  

  gquintin

  committed
  d6f06e2

  Copy full SHA for d6f06e2

• Merge branch 'master' into spmd

  

  gquintin

  authored
  5f2045b

  Copy full SHA for 5f2045b