[clfft] branch master updated (801830a -> 3a2ae95)
Ghislain Vaillant
ghisvail-guest at moszumanska.debian.org
Thu Jul 30 18:06:26 UTC 2015
This is an automated email from the git hooks/post-receive script.
ghisvail-guest pushed a change to branch master
in repository clfft.
from 801830a Merge branch 'develop'
new 19fe8b0 bumping develop to 2.5
new 79dc76b merging internal fork with performance optimizations for large sizes and other fixes
new 6ab7f55 adding the binary caching feature to store plans on disk
new ffab22e fixing incorrect transpose call bug that caused seg faults
new 2c0bb93 fixing the upper transform size limit restriction to single dimension maximum
new 37b1f44 Install scripts/perf to share/clFFT on non WIN32 systems
new ae84584 Merge pull request #52 from marbre/develop
new f3cbd72 Rename Client to clFFT-client
new a5b54d7 travis: Exec. clFFT-client (was Client before)
new ecf5d65 Merge pull request #53 from marbre/client
new fba3b2b minor doc fixes
new 74dcadb properly use the rpath on mac
new 5d30d17 Merge pull request #59 from glehmann/rpath
new 8df69d5 don't install in lib64 on mac
new 67d1085 Merge pull request #61 from glehmann/do-not-install-in-lib64-on-mac
new 48fb485 fix LoadSharedLibrary for mac
new ecc3462 Merge pull request #63 from glehmann/maclib
new 18b7c43 build google test with an external project
new 38e83ae Merge pull request #64 from glehmann/gtest-external
new 88bf9f3 Merge branch 'develop' of https://github.com/clMathLibraries/clFFT into develop
new 930644c Revert "Merge branch 'develop' of https://github.com/clMathLibraries/clFFT into develop"
new 9fceb38 Revert "minor doc fixes"
new f87503f doc fixes to correct formulas
new c03a95b fixing previous bad commits that clobbered files
new 403641d fixing doxy file with latest doxywizrd tool configuration
new 2a3c219 use the default install prefix on unixes
new d421eea Merge pull request #62 from glehmann/use-default-install-prefix-on-unixes
new c2eedd8 examples framework
new 1644f85 2d fft example
new e172b3e 3d fft example
new e10f0c2 Merge pull request #70 from 9prady9/examples
new c7bac74 install clFFT cmake configuration and version file
new 69a6538 Manual merge of cmake-config branch
new 2c8099b Merge branch 'glehmann-cmake-config' into develop
new 8c5c721 fix gtest build with msvc and ninja
new df56952 Merge pull request #69 from glehmann/gtest-msvc-fix
new 578e695 fixing VS12 builds
new 8fc96b3 Added a device selection feature for the client app.
new 92edbee Merge branch 'devsel_pub' of https://github.com/PhilSee/clFFT into ps-develop
new 533c3c3 cleaning up the selection and printing of CL devices
new bd174d4 dont run test when only print info is needed
new 1c14c41 minor fix, cleaning up unnecessary variables
new 42a90d2 Fixing a clang compilation problem for MacOSX. This closes #65
new 2a53335 syncing upstream repo and local fork
new 024fb77 making transpose support for non pow2 sizes, just starting
new f417ea9 fixing transpose logic bugs, still whole plan debug needed
new 997be17 fix
new 782a87b fixing horizontal out transpose bug
new c59848c adding feature list to readme for the recent updates
new 30626c6 fix for the -lrt link issue with ubuntu/travisci
new 5f17da9 Merge branch 'develop' of https://github.com/clMathLibraries/clFFT into develop
new 9dbaf8a syntax highlight SHA
new 0cd5871 markdown fix
new 85a851d BUILD FIX: for Linux systems when using static libraries
new d9e26a2 fix static build with msvc
new 8eacf68 Merge branch 'static-exports' of https://github.com/glehmann/clFFT into static_build_fix
new 0ffd52f Fixing static build condition in CMakeLists
new 06d584c Change BUILD_STATIC to CLFFT_BUILD_STATIC
new 653b173 Change BUILD_STATIC to CLFFT_BUILD_STATIC in statTimer
new 1c14692 Rename CMake varible BUILD_SHARED_LIBRARY to BUILD_SHARED_LIBS
new aa2056b Using BUILD_SHARED_LIBS as default configuration for library type
new 61b1527 define CLFFT_STATIC in clFFT.version.h
new 637c128 let cmake decide to build the lib as STATIC or SHARED based on BUILD_SHARED_LIBS value
new 1597f0f Merge pull request #2 from glehmann/arrayfire-static_build_fix
new 3d9122e Merge pull request #75 from arrayfire/static_build_fix
new b38f31c fix for multi-gpu issue that caused plans to be non-unique w.r.t device
new ba61721 Merge branch 'develop' of https://github.com/clMathLibraries/clFFT into develop
new d8292e6 adding scalar type support
new 2e14455 extending the size support of some 1D break-downs
new b34393b 2D logic update for real
new 6aa6a93 updating 2D real forward plan and transform
new c25c4b3 fixing event leak bug
new a9fbc30 updating incorrect tmpbufsize
new 30fc76b fixing planar RC format
new b03c26b clarifying logic of transposes w.r.t transOutHorizontal
new c96280f fixing work size calculation bugs
new 05f24ca updating 3D planner logic real forward
new 0d7e198 updating the transform 3D logic real forward
new a2e4b0b cleaning up real backward; updating the C2R logic and temp memory use
new e69f8ee fixing logic to allocate C2R temp buffer once
new 157b0c9 updating 2D real backward
new 23e811f updating 3D real backward
new 6ae0bbc fixing ditance condition to enable transpose optimizations
new 56a46c8 placing pieces of extracted protype gen code
new 8460bd0 updating 1d real with prototype algorithm
new 862cb2c fixing dev issues
new 797715d fixing rc issues and adding minor features
new 3c4ebf6 fixing more issues
new 49fbb5b fixing imaginary part sign issues
new 7196631 adding a real large 1d test
new e7d49b0 fixing some more bugs
new 77f42f7 fixing failure due to sub-64 length and tranpose not supported
new 3ed5368 fixing the planar failure
new f48b9b6 build gtest only when BUILD_TEST is on
new 4d3730c Merge branch 'gtest' of https://github.com/glehmann/clFFT into gbb
new 59b532a Merge branch 'gbb' into develop
new 8b6c24b keeping the message after merging pull request
new 8a144eb don't force the usage of libc++ on mac - it is not the default c++ lib on 10.8 and older
new 1a1be10 Merge branch 'mac-libc++' of https://github.com/glehmann/clFFT into glehmann-mac-libc++
new 9d57ee6 Merge branch 'develop' of https://github.com/clMathLibraries/clFFT into develop
new 40517b2 fixing RC bugs that caused 2d/3d fails
new c130641 merging from develop branch and bumping release version
new 7098959 changing from doc to docs
new 4d2ab24 adding performance numbers
new 25515a5 adding release notes
new 1dd01b6 updated performance measurement scripts and instruction manual
new 3a2ae95 updated performance measurement scripts and instruction manual
The 107 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "adds" were already present in the repository and have only
been added to this reference.
Summary of changes:
.gitignore | 3 +
.travis.yml | 14 +-
README.md | 10 +
ReleaseNotes.txt | 54 +
doc/clFFT.doxy | 1894 ---------
docs/clFFT.doxy | 2383 +++++++++++
.../clFFT_2.6.0/FirePro_W9100/C2C_1D_double.csv | 25 +
.../clFFT_2.6.0/FirePro_W9100/C2C_1D_single.csv | 25 +
.../clFFT_2.6.0/FirePro_W9100/C2C_2D_double.csv | 13 +
.../clFFT_2.6.0/FirePro_W9100/C2C_2D_single.csv | 13 +
.../clFFT_2.6.0/FirePro_W9100/C2C_3D_double.csv | 9 +
.../clFFT_2.6.0/FirePro_W9100/C2C_3D_single.csv | 9 +
.../clFFT_2.6.0/FirePro_W9100/R2C_1D_double.csv | 23 +
.../clFFT_2.6.0/FirePro_W9100/R2C_1D_single.csv | 25 +
.../clFFT_2.6.0/FirePro_W9100/R2C_2D_double.csv | 13 +
.../clFFT_2.6.0/FirePro_W9100/R2C_2D_single.csv | 13 +
.../clFFT_2.6.0/FirePro_W9100/R2C_3D_double.csv | 9 +
.../clFFT_2.6.0/FirePro_W9100/R2C_3D_single.csv | 9 +
.../cuFFT_7.0/Tesla_K40/C2C_1D_double.csv | 25 +
.../cuFFT_7.0/Tesla_K40/C2C_1D_single.csv | 25 +
.../cuFFT_7.0/Tesla_K40/C2C_2D_double.csv | 13 +
.../cuFFT_7.0/Tesla_K40/C2C_2D_single.csv | 13 +
.../cuFFT_7.0/Tesla_K40/C2C_3D_double.csv | 9 +
.../cuFFT_7.0/Tesla_K40/C2C_3D_single.csv | 9 +
.../cuFFT_7.0/Tesla_K40/R2C_1D_double.csv | 25 +
.../cuFFT_7.0/Tesla_K40/R2C_1D_single.csv | 25 +
.../cuFFT_7.0/Tesla_K40/R2C_2D_double.csv | 13 +
.../cuFFT_7.0/Tesla_K40/R2C_2D_single.csv | 13 +
.../cuFFT_7.0/Tesla_K40/R2C_3D_double.csv | 9 +
.../cuFFT_7.0/Tesla_K40/R2C_3D_single.csv | 9 +
{doc => docs}/realfft_1dlen.jpg | Bin
{doc => docs}/realfft_ex_n7.jpg | Bin
{doc => docs}/realfft_ex_n8.jpg | Bin
{doc => docs}/realfft_expl_01.jpg | Bin
{doc => docs}/realfft_expl_02.jpg | Bin
{doc => docs}/realfft_expl_03.jpg | Bin
{doc => docs}/realfft_expl_04.jpg | Bin
{doc => docs}/realfft_expl_05.jpg | Bin
{doc => docs}/realfft_expl_06.jpg | Bin
{doc => docs}/realfft_expl_07.jpg | Bin
{doc => docs}/realfft_expl_08.jpg | Bin
{doc => docs}/realfft_fwdinv.jpg | Bin
src/CMakeLists.txt | 69 +-
src/clFFTConfig.cmake.in | 3 +
src/clFFTConfigVersion.cmake.in | 46 +
src/client/CMakeLists.txt | 22 +-
src/client/client.cpp | 67 +-
src/client/client.h | 41 +
src/client/openCL.misc.cpp | 863 ++--
src/client/openCL.misc.h | 43 +-
src/examples/CMakeLists.txt | 52 +
src/examples/fft1d.c | 136 +
src/examples/fft2d.c | 145 +
src/examples/fft3d.c | 154 +
src/gtest.cmake | 79 +
src/include/clFFT.h | 12 +-
src/include/clFFT.version.h.in | 2 +
src/include/sharedLibrary.h | 16 +-
src/library/CMakeLists.txt | 19 +-
src/library/accessors.cpp | 70 +-
src/library/action.cpp | 777 ++++
src/library/action.h | 257 ++
src/library/fft_binary_lookup.cpp | 688 ++++
src/library/fft_binary_lookup.h | 276 ++
src/library/generator.copy.cpp | 457 ++-
src/library/generator.h | 9 +-
src/library/generator.stockham.cpp | 1128 +++--
src/library/generator.stockham.h | 278 +-
src/library/generator.transpose.gcn.cpp | 1020 +++++
...rator.transpose.h => generator.transpose.gcn.h} | 0
....transpose.cpp => generator.transpose.vliw.cpp} | 167 +-
...ator.transpose.h => generator.transpose.vliw.h} | 0
src/library/lifetime.cpp | 2 +
src/library/mainpage.h | 10 +-
src/library/md5sum.c | 312 ++
src/library/md5sum.h | 50 +
src/library/plan.cpp | 4319 +++++++++++---------
src/library/plan.h | 342 +-
src/library/private.h | 5 +-
src/library/repo.cpp | 55 +-
src/library/repo.h | 77 +-
src/library/transform.cpp | 1378 +++----
src/scripts/perf/CMakeLists.txt | 7 +-
src/scripts/perf/fftPerformanceTesting.py | 37 +-
src/scripts/perf/manual.pdf | Bin 0 -> 180095 bytes
src/scripts/perf/measurePerformance.py | 327 +-
src/scripts/perf/plotPerformance.py | 40 +-
src/statTimer/CMakeLists.txt | 7 +-
src/statTimer/statisticalTimer.extern.h | 12 +-
src/tests/CMakeLists.txt | 17 +-
src/tests/accuracy_test_pow2.cpp | 115 +
src/tests/buffer.h | 2 +-
src/tests/cl_transform.h | 7 +-
src/tests/gtest_main.cpp | 34 +-
src/tests/test_constants.cpp | 5 +-
src/tests/test_constants.h | 5 +-
src/tests/unit_test.cpp | 5 +-
97 files changed, 12659 insertions(+), 6139 deletions(-)
create mode 100644 ReleaseNotes.txt
delete mode 100644 doc/clFFT.doxy
create mode 100644 docs/clFFT.doxy
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/C2C_1D_double.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/C2C_1D_single.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/C2C_2D_double.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/C2C_2D_single.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/C2C_3D_double.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/C2C_3D_single.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/R2C_1D_double.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/R2C_1D_single.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/R2C_2D_double.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/R2C_2D_single.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/R2C_3D_double.csv
create mode 100644 docs/performance/clFFT_2.6.0/FirePro_W9100/R2C_3D_single.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/C2C_1D_double.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/C2C_1D_single.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/C2C_2D_double.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/C2C_2D_single.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/C2C_3D_double.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/C2C_3D_single.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/R2C_1D_double.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/R2C_1D_single.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/R2C_2D_double.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/R2C_2D_single.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/R2C_3D_double.csv
create mode 100644 docs/performance/cuFFT_7.0/Tesla_K40/R2C_3D_single.csv
rename {doc => docs}/realfft_1dlen.jpg (100%)
rename {doc => docs}/realfft_ex_n7.jpg (100%)
rename {doc => docs}/realfft_ex_n8.jpg (100%)
rename {doc => docs}/realfft_expl_01.jpg (100%)
rename {doc => docs}/realfft_expl_02.jpg (100%)
rename {doc => docs}/realfft_expl_03.jpg (100%)
rename {doc => docs}/realfft_expl_04.jpg (100%)
rename {doc => docs}/realfft_expl_05.jpg (100%)
rename {doc => docs}/realfft_expl_06.jpg (100%)
rename {doc => docs}/realfft_expl_07.jpg (100%)
rename {doc => docs}/realfft_expl_08.jpg (100%)
rename {doc => docs}/realfft_fwdinv.jpg (100%)
create mode 100644 src/clFFTConfig.cmake.in
create mode 100644 src/clFFTConfigVersion.cmake.in
create mode 100644 src/examples/CMakeLists.txt
create mode 100644 src/examples/fft1d.c
create mode 100644 src/examples/fft2d.c
create mode 100644 src/examples/fft3d.c
create mode 100644 src/gtest.cmake
create mode 100644 src/library/action.cpp
create mode 100644 src/library/action.h
create mode 100644 src/library/fft_binary_lookup.cpp
create mode 100644 src/library/fft_binary_lookup.h
create mode 100644 src/library/generator.transpose.gcn.cpp
copy src/library/{generator.transpose.h => generator.transpose.gcn.h} (100%)
rename src/library/{generator.transpose.cpp => generator.transpose.vliw.cpp} (81%)
rename src/library/{generator.transpose.h => generator.transpose.vliw.h} (100%)
create mode 100644 src/library/md5sum.c
create mode 100644 src/library/md5sum.h
create mode 100644 src/scripts/perf/manual.pdf
--
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/clfft.git
More information about the debian-science-commits
mailing list