Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Update to version 0.8.0 #70

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 71 commits into from
Jun 28, 2016
Merged

Update to version 0.8.0 #70

merged 71 commits into from
Jun 28, 2016

Conversation

@CNugteren
Copy link
Owner

@CNugteren CNugteren commented Jun 28, 2016

Version 0.8.0

  • Added support for half-precision floating-point (fp16) in the library
  • Made it possible to compile the performance tests (clients) separately from the correctness tests
  • Made a reference BLAS and head-to-head performance comparison optional in the clients
  • Increased the verbosity of the "-verbose" option in the correctness tests
  • Refactored the host code for better compilation times and fewer lines of code
  • Added Appveyor continuous integration and increased coverage of the Travis builds
  • Improved the API documentation
  • Various minor fixes and enhancements
  • Added tuned parameters for various devices (see README)
  • Added half-precision routines:
    • Level-1: HSWAP/HSCAL/HCOPY/HAXPY/HDOT/HNRM2/HASUM/HSUM/iHAMAX/iHMAX/iHMIN
    • Level-2: HGEMV/HGBMV/HHEMV/HHBMV/HHPMV/HSYMV/HSBMV/HSPMV/HTRMV/HTBMV/HTPMV/HGER/HSYR/HSPR/HSYR2/HSPR2
    • Level-3: HGEMM/HSYMM/HSYRK/HSYR2K/HTRMM
  • Added non-BLAS routines:
    • SOMATCOPY/DOMATCOPY/COMATCOPY/ZOMATCOPY/HOMATCOPY (matrix copy, scaling, and/or transpose)

CNugteren added 30 commits May 12, 2016 19:56
@CNugteren
Initial changes in preparation for half-precision fp16 support
f2ba758
@CNugteren
Initial experimental version of the half-precision HAXPY routine
120c31a
@CNugteren
Set kernel arguments for AXPY as constant memory buffers, making it p…
5e1b2e0 
…ossible to transfer half-precision values as well
@CNugteren
Added header with conversions from and to half-precision floating-point
4b6bdd8
@CNugteren
Added an example of using the half-precision HAXPY routine
591e343
@CNugteren
Prepared GEMM and supporting kernels and tuners for half-precision su…
af2ac62 
…pport
@CNugteren
Added half precision tuning results for supporting kernels (pad, copy…
7a3b695 
…, transpose, padtranspose)
@CNugteren
Prepared the changelog for the next release
182d2cf
@CNugteren
Merged in latest changes from 0.7.1 release
489c5d7
@CNugteren
Added half-precision support for all level 1 routines
f70ded3
@CNugteren
Added first tuning results for the half-precision xDOT kernels
3c9e63c
@CNugteren
Added level-1 half-precision routines HSWAP/HSCAL/HCOPY/HAXPY/HDOT/HN…
803aaf3 
…RM2/HASUM/HSUM/iHAMAX/iHMAX/iHMIN
@CNugteren
Prepared the GEMV kernels and tuner for half-precision support
88551b4
@CNugteren
Added first tuning results for the half-precision xGEMV kernels
b6268d0
@CNugteren
Added level-2 half-precision routines HGEMV/HGBMV/HHEMV/HHBMV/HHPMV/H…
95b828d 
…SYMV/HSBMV/HSPMV/HTRMV/HTBMV/HTPMV
@CNugteren
Prepared the GER kernels and tuner for half-precision support
c8ff3f1
@CNugteren
Fixed tuning results for half-precision; added first results for the …
f0cb3fd 
…xGER kernels
@CNugteren
Added level-2 half-precision routines HGER/HSYR/HSPR/HSYR2/HSPR2
3e9a07f
@CNugteren
Updated README with information on half-precision support
ae7d705
@CNugteren
Added proper argument handling and displaying for half-precision data…
ac15750 
…-types
@CNugteren
Added level-3 half-precision routines HGEMM/HSYMM/HSYRK/HSYR2K/HTRMM
9f87455
@CNugteren
Added possibility to run the performance client with half-precision
4612ff3
@CNugteren
Added half-precision tests for the CBLAS reference through conversion…
b487d4d 
… to single-precison
@CNugteren
Added half-precision tests for the clBLAS reference through conversio…
03182f9 
…n to single-precision
@CNugteren
Merge branch 'half_precision' into development
61105e3
@CNugteren
Separated the performance tests (clients) from the correctness tests …
305bf16 
…in CMake
@CNugteren
Increased the verbosity of the -verbose option in the correctness tests
f6b2cd9
@CNugteren
Made use of CMake's built-in unit testing, allowing all tests to be r…
983df6a 
…un using 'make test'
@CNugteren
Added tuning parameters for 'GRID K520' and 'HD Graphics Skylake ULT …
137d1d8 
…GT2'
@CNugteren
Added return value to the test binaries (0: success, 1: failure), all…
e561e3f 
…owing it to work under CTest properly
CNugteren added 29 commits June 17, 2016 11:41
@CNugteren
Moved the ErrorIn function from the Routine class to the utilities he…
520e28e 
…ader
@CNugteren
Moved the RunKernel and PadCopyTransposeMatrix functions out of the R…
98a95c8 
…outine class
@CNugteren
Removed the interface to the cache functions from the Routine class, …
536b7fe 
…calls them directly now
@CNugteren
Removed the precision argument from the routines in favor of a single…
f9947b4 
… templated function
@CNugteren
Removed the template from the Routine base-class
7b4c0e1
@CNugteren
Clean-up of the routine class, moved RunKernel to the routine/common …
bacb5d2 
…file
@CNugteren
Moved all headers into the source tree, changed headers to .hpp exten…
f726fbd 
…sion
@CNugteren
Renamed all C++ source files to .cpp to match the .hpp extension better
6120345
@CNugteren
Merge pull request #69 from CNugteren/refactoring
395a0ef 
Refactoring of the Routine class and file-renaming
@CNugteren
Minor fix to the database script
eab8d3c
@CNugteren
Added tuning results for 'Intel(R) HD Graphics Haswell Ultrabook GT2 …
66908ef 
…Mobile' (thanks to OursDesCavernes)
@CNugteren
Increased the verbosity of the '-verbose' option for the correctness …
8f7131b 
…tests, now printing when a library is called
@CNugteren
Changed the symbol for error-code skipped tests to distinguish from s…
fdfbc9a 
…uccesfull error-code checks in the correctness tests
@CNugteren
Added fp16 to the alltuners target
ca386f9
@CNugteren
Moved the performance graph scripts to the 'scripts' subfolder
69beca9
@CNugteren
Increased coverage of Travis CI automatic builds
5f88863
@CNugteren
Added Appveyor Windows CI support
7eeb790
@CNugteren
Fixed a bug in the Appveyor script
dac9945
@CNugteren
Added vcvarsall to AppVeyor and added AppVeyor icons to README
5557a6a
@CNugteren
Fixes for the AppVeyor Windows build
76b20cf
@CNugteren
Updated the README in various places
9171f1c
@CNugteren
Made it possible to build the OMATCOPY test and client in case only c…
2c031f3 
…lBLAS is present
@CNugteren
Made it possible to build the clients and tests on Windows using Visu…
871b576 
…al Studio
@CNugteren
Added configuration for AppVeyor to keep the results of the builds as…
7c6bb6e 
… an 'artifact'
@CNugteren
Fixes bug in AppVeyor with install directory
88014e3
@CNugteren
Fixes bug in AppVeyor with install directory (2)
743da1b
@CNugteren
Changed the AppVeyor buildscript to use nmake instead of 'cmake --build'
a003cc2
@CNugteren
Changed the AppVeyor buildscript to use nmake instead of 'cmake --bui…
33dddd3 
…ld' (2)
@CNugteren
Updated to version 0.8.0
577f0ee
@CNugteren CNugteren merged commit 7c13bac into master Jun 28, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@CNugteren @psyhtest