Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | merge with default branch | 2014-03-28 | |
|\ | |||
* | | Enable vectorization of gemv for PacketSize>4 through unaligned loads (still ↵ | 2014-03-28 | |
| | | | | | | | | better than no vectorization) | ||
* | | Fixed compilation error when FMA instructions are enabled. | 2014-03-27 | |
| | | |||
* | | Silenced "unused variable" warnings when compiling with FMA. | 2014-03-27 | |
| | | |||
* | | Vectorized the packing of a col-major matrix used as the right hand side ↵ | 2014-03-27 | |
| | | | | | | | | argument in a matrix-matrix product when AVX instructions are used. No vectorization takes place when SSE instructions are used, however this doesn't seem to impact performance. | ||
* | | Vectorized the packing of a row-major matrix used as the left hand side ↵ | 2014-03-27 | |
| | | | | | | | | argument in a matrix-matrix product. | ||
| * | Fix warning | 2014-03-27 | |
| | | |||
* | | Made sure that the version of gemm_pack_rhs specialized for row major ↵ | 2014-03-26 | |
| | | | | | | | | matrices is vectorized when nr == 2*PacketSize (which is the case for SSE when compiling in 64bit mode). | ||
* | | Merged latest updates from the parent branch | 2014-03-26 | |
|\ \ | |||
| | * | Update gebp kernel to process a panle of 4 columns at once for the remaining ↵ | 2014-03-26 | |
| | | | | | | | | | | | | ones. | ||
| | * | Remove remaining bits of the dead working buffer | 2014-03-26 | |
| |/ | |||
| * | Implement new 1 packet x 8 gebp kernel | 2014-03-26 | |
| | | |||
* | | Merged latest changes from the parent | 2014-03-18 | |
|\ \ | |||
* \ \ | Merged latest changes from the main trunk | 2014-02-24 | |
|\ \ \ | |||
| * \ \ | Merged eigen/eigen into default | 2014-02-24 | |
| |\ \ \ | |||
* | | | | | Added support for FMA instructions | 2014-02-24 | |
|/ / / / | |||
* | | | | Reverted the definition of the EIGEN_ALIGN to its former meaning (i.e. a ↵ | 2014-02-18 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | boolean) Created a new EIGEN_ALIGN_BYTES define to encode how the data should be aligned Fixed a few remaining alignment issues exposed when the Eigen code is compiled with avx enabled. Created a new EIGEN_ALIGN_DEFAULT define, which is set to the minimum alignment value required for the chosen instruction set. Use this value instead of EIGEN_ALIGN32 to preserve the existing alignment on SSE/Altivec/Neon. | ||
* | | | | Added support for AVX to Eigen. | 2014-01-29 | |
| | | | | |||
| | | * | Improved the efficiency if the block-panel matrix multiplication code: the ↵ | 2014-01-02 | |
| | |/ | | | | | | | | | | change reduces the pressure on the L1 cache by removing the calls to gebp_traits::unpackRhs(). Instead the packetization of the rhs blocks is done on the fly in gebp_traits::loadRhs(). This adds numerous calls to pset1<ResPacket> (since we're packetizing on the fly in the inner loop) but this is more than compensated by the fact that we're decreasing the memory transfers by a factor RhsPacketSize. | ||
| | * | Use vectorization when packing row-major rhs matrices. (bug #717) | 2013-12-17 | |
| | | | |||
| | * | Fix selfadjoint_matrix_vector_product for complex with packet size > 2 ↵ | 2014-03-07 | |
| |/ | | | | | | | (e.g., AVX) | ||
| * | Implement bug #317: use a template function call to suppress unused variable ↵ | 2014-02-24 | |
| | | | | | | | | warnings. This also fix the issue of the previous changeset in a much nicer way. | ||
| * | Workaround clang ABI change with unsed arguments (ugly fix) | 2014-02-24 | |
|/ | |||
* | Remove useless register keyword, and optimize predux_min/max for SSE4 | 2014-01-25 | |
| | |||
* | merge with default branch | 2013-11-05 | |
|\ | |||
* | | Use aligned loads in Matrix-Vector product where possible. Fixes bug #689 | 2013-10-29 | |
| | | |||
* | | Fix bug #359: fix AlignedBit flag of CoeffBasedProduct thus enabling the ↵ | 2013-10-28 | |
| | | | | | | | | vectorization of more matrix products | ||
* | | Add explanations of the logic behind the matrix-vector products | 2013-08-21 | |
| | | |||
| * | merge with main branch | 2013-07-17 | |
| |\ | |/ |/| | |||
* | | Fix bug #314: move remaining math functions from internal to numext namespace | 2013-06-10 | |
| | | |||
| * | merge with default branch | 2013-04-19 | |
| |\ | |/ |/| | |||
* | | Fix bug #482: pass scalar value by const reference (it remained a few cases) | 2013-04-12 | |
| | | |||
* | | Clean source code and unit tests with respect to -Wunused-local-typedefs | 2013-04-10 | |
| | | |||
* | | Fix a couple of int versus Index issues. | 2013-04-09 | |
| | | |||
* | | Allow multiplication like binary operators to be applied on type couples ↵ | 2013-03-20 | |
| | | | | | | | | supported by scalar_product_traits | ||
* | | Fix "routine is both "inline" and "noinline"" warnings | 2013-02-28 | |
| | | |||
* | | bug #482: pass scalar arguments by const references. Still remains a few ↵ | 2013-02-25 | |
| | | | | | | | | cases that might affect the ABI (see the bug entry) | ||
* | | Fix bug #496: generalize internal rank1_update implementation to accept ↵ | 2013-02-24 | |
| | | | | | | | | | | | | uplo(A) += v * w and make A.triangularView() += v * w uses it. Update unit tests and blas interface respectively. | ||
* | | Fix bug #551: compilation issue when using EIGEN_DEFAULT_DENSE_INDEX_TYPE | 2013-02-09 | |
| | | |||
| * | Add support for NVCC5: most of the Core and part of LU are callable from ↵ | 2013-02-07 | |
|/ | | | | | | CUDA code. Still a lot to do. | ||
* | multiple warnings for unused variable | 2012-09-07 | |
| | |||
* | Fix out-of-range memory access in GEMV (the memory was not used for the ↵ | 2012-08-30 | |
| | | | | | | | computation, only to assemble unaligned packets from aligned packet loads) (transplanted from 221f54698c2f6690da8c0f44c1e31e55118dedab ) | ||
* | fix bug #495: remove too aggressive EIGEN_FLATTEN_ATTRIB attribute | 2012-08-02 | |
| | | | | (after some benchmarking, it was not useful anymore) | ||
* | fix various regressions with MKL support | 2012-07-28 | |
| | |||
* | Automatic relicensing to MPL2 using Keirs script. Manual fixup follows. | 2012-07-13 | |
| | |||
* | remove dynamic allocation for fixed size object and triangular matrix-matrix ↵ | 2012-06-26 | |
| | | | | products | ||
* | fix compilation issue with MKL's backend | 2012-06-21 | |
| | |||
* | bug #466: better fix for the race condition: this new patch add an ↵ | 2012-06-14 | |
| | | | | | | | initParallel() function which must be called at the initialization time of any multi-threaded application calling Eigen from multiple threads. | ||
* | avoid dynamic allocation for fixed size triangular solving | 2012-06-12 | |
| | |||
* | Fix bug #466: race condition destected by helgrind in manage_caching_sizes. | 2012-06-08 | |
| | | | | After all, the solution based on threadprivate is not that costly. |