Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | update product bench | Gael Guennebaud | 2009-11-06 |
| | |||
* | switched to BenchUtil.h | Mark Borgerding | 2009-10-30 |
| | |||
* | properly implement BenchTimer on POSIX | Benoit Jacob | 2009-10-29 |
| | | | | (may require a platform check for the clock name on non-linux platforms) | ||
* | added inline to many functions | Mark Borgerding | 2009-10-22 |
| | |||
* | added Eigen::FFT and | Mark Borgerding | 2009-10-20 |
|\ | | | | | | | Eigen::Complex | ||
* | | Added Windows support to the BenchTimer. | Hauke Heibel | 2009-10-20 |
| | | |||
| * | merged eigen2_for_fft into eigen2 mainline | Mark Borgerding | 2009-10-20 |
|/| | |||
* | | extend the sparse matrix assembly benchmark | Gael Guennebaud | 2009-10-07 |
| | | |||
* | | implement __gnuc_forget_about_setZero_its_over_now | Gael Guennebaud | 2009-09-18 |
| | | |||
* | | add a "rot" benchmark in BTL | Gael Guennebaud | 2009-08-15 |
| | | |||
* | | add ger and lu with partial pivoting in BTL | Gael Guennebaud | 2009-08-04 |
| | | |||
* | | * Bye bye MultiplierBase, extend a bit AnyMatrixBase to allow =, +=, and -= | Gael Guennebaud | 2009-08-03 |
| | | | | | | | | * This probably makes ReturnByValue needless | ||
* | | synch with main branch | Gael Guennebaud | 2009-07-28 |
|\ \ | |||
* | | | compilation fixes in BTL | Gael Guennebaud | 2009-07-28 |
| | | | |||
| * | | re-implement stableNorm using a homemade blocky and | Gael Guennebaud | 2009-07-17 |
| | | | | | | | | | | | | vectorization friendly algorithm (slow if no vectorization) | ||
| * | | add other stable norm impl. in the benchmark | Gael Guennebaud | 2009-07-16 |
| | | | |||
| * | | bugfix in blueNorm | Gael Guennebaud | 2009-07-16 |
| | | | |||
| * | | add a benchmark for the different norms | Gael Guennebaud | 2009-07-16 |
|/ / | |||
* | | add a partial LU bench in BTL | Gael Guennebaud | 2009-06-04 |
| | | |||
| * | more work on ei_fftw_impl | Mark Borgerding | 2009-05-31 |
| | | |||
| * | added real-optimized inverse FFT (NFFT must be multiple of 4) | Mark Borgerding | 2009-05-25 |
| | | |||
| * | changed name from simple_fft_traits to ei_kissfft_impl | Mark Borgerding | 2009-05-25 |
| | | |||
| * | added FFT inverse complex-to-scalar interface (not yet optimized) | Mark Borgerding | 2009-05-23 |
| | | |||
| * | started real optimization, added benchmark for FFT | Mark Borgerding | 2009-05-23 |
| | | |||
* | | remove sentence "Eigen itself is part of the KDE project." | Benoit Jacob | 2009-05-22 |
|/ | | | | it never made very precise sense. but now does it still make any? | ||
* | various BTL updates (disable Cholesky for MTL, add new plot settings, | Gael Guennebaud | 2009-03-09 |
| | | | | etc) | ||
* | forgot to include a file in previous commit | Gael Guennebaud | 2009-03-09 |
| | |||
* | add an option to bench eigen without GCC's auto vec (might conflict with | Gael Guennebaud | 2009-03-09 |
| | | | | Eigen's auto vec) | ||
* | BTL: add syr2 action | Gael Guennebaud | 2009-03-05 |
| | |||
* | BTL: - patch from Victor (add ACML support) | Gael Guennebaud | 2009-03-05 |
| | | | | - fix overflow issues | ||
* | various update of of BTL | Gael Guennebaud | 2009-03-04 |
| | |||
* | add symv bench | Gael Guennebaud | 2009-02-20 |
| | |||
* | s/cholesky/llt in precompiled lib and BTL | Gael Guennebaud | 2009-02-06 |
| | |||
* | add bench_reverse, draft of a reverse vectorization for AltiVec, make | Gael Guennebaud | 2009-02-06 |
| | | | | global Scaling function static | ||
* | update cholesky benchmark | Gael Guennebaud | 2009-02-03 |
| | |||
* | various updates in the (still messy) sparse benchmarks | Gael Guennebaud | 2009-01-28 |
| | |||
* | * complete the support of QVector via a QtAlignedMalloc header | Gael Guennebaud | 2009-01-20 |
| | | | | * add a unit test for QVector which shows the issue with QVector::fill | ||
* | bugfix in Map by Keir Mierle | Gael Guennebaud | 2009-01-18 |
| | |||
* | add ublas vector of vector in sparse setter bench | Gael Guennebaud | 2009-01-17 |
| | |||
* | add a sparse setter bench | Gael Guennebaud | 2009-01-17 |
| | |||
* | add a sparse * dense_vector bench | Gael Guennebaud | 2009-01-14 |
| | |||
* | updating FindEigen2.cmake for proper search order | James Richard Tyrer | 2009-01-11 |
| | |||
* | the big memory changes. the most important changes are: | Benoit Jacob | 2009-01-08 |
| | | | | | | | ei_aligned_malloc now really behaves like a malloc (untyped, doesn't call ctor) ei_aligned_new is the typed variant calling ctor EIGEN_MAKE_ALIGNED_OPERATOR_NEW now takes the class name as parameter | ||
* | * require CMake 2.6.2 everywhere, Alexander Neundorf says it'd make it | Benoit Jacob | 2009-01-04 |
| | | | | | | easier to have a uniform requirement in kdesupport for when he makes fixes. * add eigen versioning macros | ||
* | * the 4th template param of Matrix is now Options. One bit for storage | Benoit Jacob | 2009-01-04 |
| | | | | | | | | | | | | | | | | order, one bit for enabling/disabling auto-alignment. If you want to disable, do: Matrix<float,4,1,Matrix_DontAlign> The Matrix_ prefix is the only way I can see to avoid ambiguity/pollution. The old RowMajor, ColMajor constants are deprecated, remain for now. * this prompted several improvements in matrix_storage. ei_aligned_array renamed to ei_matrix_array and moved there. The %16==0 tests are now much more centralized in 1 place there. * unalignedassert test: updated * update FindEigen2.cmake from KDElibs * determinant test: use VERIFY_IS_APPROX to fix false positives; add testing of 1 big matrix | ||
* | * make WithAlignedOperatorNew always align even when vectorization is disabled | Benoit Jacob | 2008-12-30 |
| | | | | * make ei_aligned_malloc and ei_aligned_free honor custom operator new and delete | ||
* | * the Upper->UpperTriangular change | Benoit Jacob | 2008-12-20 |
| | | | | * finally get ei_add_test right | ||
* | various MSVC fixes in BTL | Gael Guennebaud | 2008-12-19 |
| | |||
* | improved MSVC support in cmake files (SSE) | Gael Guennebaud | 2008-12-18 |
| | |||
* | Update e-mail address | Benoit Jacob | 2008-11-24 |
| |