[ Thread Index | More lists.tuxfamily.org/eigen-commits Archives ]

87 message(s), the most recent arriving on Wed 01 April at 01:53 GMT.

• [eigen-commits] commit/eigen: 3 new changesets Bitbucket

• [eigen-commits] commit/eigen: ggael: Add SSE vectorization of Quaternion::conjugate. Significant speed-up when combined with products like q1*q2.conjugate() Bitbucket
• [eigen-commits] commit/eigen: bjacob: HalfPacket also needed to be disabled for double, on ARMv8. Bitbucket
• [eigen-commits] commit/eigen: bjacob: Improve analyze-blocking-sizes, and in particular give it a evaluate-defaults tool Bitbucket

• [eigen-commits] commit/eigen: bjacob: Work around an ICE in Clang 3.5 in the iOS toolchain with double NEON intrinsics. Bitbucket
• [eigen-commits] commit/eigen: bjacob: must also disable complex<double> when disabling double vectorization Bitbucket
• [eigen-commits] commit/eigen: mglisse: New scoring functor to select the pivot. Bitbucket
• [eigen-commits] commit/eigen: bjacob: Add a benchmark-default-sizes action to benchmark-blocking-sizes.cpp Bitbucket
• [eigen-commits] commit/eigen: bjacob: Add missing copyright notices Bitbucket
• [eigen-commits] commit/eigen: benoitsteiner: Fixed compilation error when compiling with gcc4.7 Bitbucket
• [eigen-commits] commit/eigen: bjacob: Fix asm comments in 1px1 kernel Bitbucket

• [eigen-commits] commit/eigen: 5 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: Merged in blechta/eigen/fix-cg-zero-guess (pull request #100) Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: blechta: Really use zero guess in ConjugateGradients::solve as documented Bitbucket
• [eigen-commits] commit/eigen: bjacob: Complete the tool to analyze the efficiency of default sizes. Bitbucket
• [eigen-commits] commit/eigen: bjacob: output to cout, not cerr, the actual results Bitbucket
• [eigen-commits] commit/eigen: ggael: Bug 824: improve accuracy of Quaternion::angularDistance using atan2 instead of acos. Bitbucket
• [eigen-commits] commit/eigen: ggael: Bug 824: improve accuracy of Quaternion::angularDistance using atan2 instead of acos. Bitbucket

• [eigen-commits] commit/eigen: ggael: Product optimization: implement a dynamic loop-swapping startegy to improve memory accesses to the destination matrix in the case of K-rank-update like products, i.e., for products of the kind: "large x small" * "small x large" Bitbucket
• [eigen-commits] commit/eigen: ggael: Rename LSCG to LeastSquaresConjugateGradient Bitbucket
• [eigen-commits] commit/eigen: bjacob: Make benchmark-blocking-sizes detect changes to clock speed and be resilient to that. Bitbucket

• [eigen-commits] commit/eigen: ggael: Improve product kernel: replace the previous dynamic loop swaping strategy by a more general one: Bitbucket
• [eigen-commits] commit/eigen: ggael: Update gemm performance monitoring tool: Bitbucket
• [eigen-commits] commit/eigen: ggael: Improve blocking heuristic: if the lhs fit within L1, then block on the rhs in L1 (allows to keep packed rhs in L1) Bitbucket
• [eigen-commits] commit/eigen: ggael: update gemm changeset list Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: Bug 969: workaround abiguous calls to Ref using enable_if. Bitbucket
• [eigen-commits] commit/eigen: ggael: Bug 969: workaround abiguous calls to Ref using enable_if. Bitbucket
• [eigen-commits] commit/eigen: ggael: Avoid undeflow when blocking size are tuned manually. Bitbucket

• [eigen-commits] commit/eigen: 3 new changesets Bitbucket
• [eigen-commits] commit/eigen: bjacob: deserialization assumed benchmarks in same order, but we shuffle them. Bitbucket

• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: Add unit tests for CG and sparse-LLT for long int as storage-index Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket

• [eigen-commits] commit/eigen: bjacob: Add --only-cubic-sizes option to analyze-blocking-sizes tool Bitbucket

• [eigen-commits] commit/eigen: ggael: Bug 980: fix taking a row (resp. column) of a column-major (resp. row-major) sparse matrix and add missing coeff/coeffRef members. Bitbucket
• [eigen-commits] commit/eigen: bjacob: Add support for dumping blocking sizes tables Bitbucket
• [eigen-commits] commit/eigen: 7 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: Bug 949: add static assertion for incompatible scalar types in dense end-user decompositions. Bitbucket
• [eigen-commits] commit/eigen: bjacob: organize a little our default cache sizes, and use a saner default L1 outside of x86 (10% faster on Nexus 5) Bitbucket
• [eigen-commits] commit/eigen: bjacob: update perf_monitoring/gemm/changesets.txt Bitbucket

• [eigen-commits] commit/eigen: 5 new changesets Bitbucket
• [eigen-commits] commit/eigen: bjacob: Fix bug in case where EIGEN_TEST_SPECIFIC_BLOCKING_SIZE is defined but false Bitbucket

• [eigen-commits] commit/eigen: bjacob: fix compilation with GCC 4.8 Bitbucket
• [eigen-commits] commit/eigen: bjacob: Update Nexus 5 lookup table from combining now 2 runs of the benchmark, using the analyze-blocking-sizes partition tool. Gives better worst-case performance. Bitbucket
• [eigen-commits] commit/eigen: bjacob: fix bug in maxsize calculation, which would cause products of size > 2048 to address the lookup table out of bounds Bitbucket
• [eigen-commits] commit/eigen: bjacob: Include stdint.h. Not going for cstdint because it is a C++11 addition. Needed for uint16_t at least, in lookup-table code. Bitbucket
• [eigen-commits] commit/eigen: 4 new changesets Bitbucket
• [eigen-commits] commit/eigen: bjacob: Similar to cset 9b27294a8186, also in 2px4 kernel: actual_panel_rows computation should always be resilient to parameters not consistent with the known L1 cache size, see comment Bitbucket

• [eigen-commits] commit/eigen: bjacob: use unsigned short instead of uint16_t which doesn't exist in c++98 Bitbucket
• [eigen-commits] commit/eigen: benoitsteiner: Changed the way lvalue operations are declared in TensorBase: this fixes constness isses that prevented some expressions mixing lvalues and rvalues from compiling. Bitbucket

• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix comparison warning Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix MSVC compilation: aligned type must be passed by reference Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix random unit test for 32bits systems. Bitbucket

• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: fix loadMarket wrt Index versus int Bitbucket
• [eigen-commits] commit/eigen: ggael: Update approx. minimum ordering method to push and keep structural empty diagonal elements to the bottom-right part of the matrix Bitbucket

• [eigen-commits] commit/eigen: ggael: Avoid SVD: consider denormalized small numbers as zero when computing the rank of the matrix Bitbucket

• [eigen-commits] commit/eigen: benoitsteiner: Fixed the CUDA packet primitives Bitbucket

• [eigen-commits] commit/eigen: benoitsteiner: Cleaned up the TensorDevice code a little bit. Bitbucket
• [eigen-commits] commit/eigen: benoitsteiner: Fixed the vectorized implementation of the Tensor select() method Bitbucket
• [eigen-commits] commit/eigen: benoitsteiner: Avoid making an unecessary copy of the tensor expression when evaluating it on a GPU device Bitbucket

• [eigen-commits] commit/eigen: 7 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix transpose versus adjoint. Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: chtz: Merged in d_hood/eigen/sparse-tutorial-doc-fix (pull request #107) Bitbucket
• [eigen-commits] commit/eigen: 3 new changesets Bitbucket

• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: fix stupid warning with old GCC Bitbucket

• [eigen-commits] commit/eigen: chtz: Bug 982: Make sure numext::maxi and numext::mini are called correctly, in case Scalar expressions return expression templates. Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: 3 new changesets Bitbucket

• [eigen-commits] commit/eigen: ggael: Fix regression introduced in e9a0c8c Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix sign of SuperLU::determinant Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix computeProductBlockingSizes with m==0, and add respective unit test. Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: benoitsteiner: Added documentation for the convolution operation Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: 2 new changesets Bitbucket
• [eigen-commits] commit/eigen: ggael: Add row/column-wise reverseInPlace feature. Bitbucket
• [eigen-commits] commit/eigen: ggael: Remove most of the dynamic memory allocations that occured in D&C SVD. Still remains the calls to JacobiSVD and UpperBidiagonalization. Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix bug 987: wrong alignement guess in diagonal product. Bitbucket
• [eigen-commits] commit/eigen: ggael: Fix bug 987: wrong alignement guess in diagonal product. Bitbucket