From 8fbe0e4699b4c03dd62b266371f23b103319ec36 Mon Sep 17 00:00:00 2001 From: Gael Guennebaud Date: Wed, 4 Dec 2019 10:57:07 +0100 Subject: Update old links to bitbucket to point to gitlab.com --- Eigen/src/Core/arch/SSE/MathFunctions.h | 2 +- README.md | 4 +--- doc/DenseDecompositionBenchmark.dox | 2 +- 3 files changed, 3 insertions(+), 5 deletions(-) diff --git a/Eigen/src/Core/arch/SSE/MathFunctions.h b/Eigen/src/Core/arch/SSE/MathFunctions.h index 85255ad23..92c1eecc7 100644 --- a/Eigen/src/Core/arch/SSE/MathFunctions.h +++ b/Eigen/src/Core/arch/SSE/MathFunctions.h @@ -168,7 +168,7 @@ double sqrt(const double &x) { #if EIGEN_COMP_GNUC_STRICT // This works around a GCC bug generating poor code for _mm_sqrt_pd - // See https://bitbucket.org/eigen/eigen/commits/14f468dba4d350d7c19c9b93072e19f7b3df563b + // See https://gitlab.com/libeigen/eigen/commit/8dca9f97e38970 return internal::pfirst(internal::Packet2d(__builtin_ia32_sqrtsd(_mm_set_sd(x)))); #else return internal::pfirst(internal::Packet2d(_mm_sqrt_pd(_mm_set_sd(x)))); diff --git a/README.md b/README.md index 99c9e2933..9b40e9ed4 100644 --- a/README.md +++ b/README.md @@ -2,6 +2,4 @@ For more information go to http://eigen.tuxfamily.org/. -For ***pull request*** please only use the official repository at https://bitbucket.org/eigen/eigen. - -For ***bug reports*** and ***feature requests*** go to http://eigen.tuxfamily.org/bz. +For ***pull request***, ***bug reports***, and ***feature requests***, go to https://gitlab.com/libeigen/eigen. diff --git a/doc/DenseDecompositionBenchmark.dox b/doc/DenseDecompositionBenchmark.dox index 7be9c70cd..8f9570b7a 100644 --- a/doc/DenseDecompositionBenchmark.dox +++ b/doc/DenseDecompositionBenchmark.dox @@ -35,7 +35,7 @@ Timings are in \b milliseconds, and factors are relative to the LLT decompositio + For large problem sizes, only the decomposition implementing a cache-friendly blocking strategy scale well. Those include LLT, PartialPivLU, HouseholderQR, and BDCSVD. This explain why for a 4k x 4k matrix, HouseholderQR is faster than LDLT. In the future, LDLT and ColPivHouseholderQR will also implement blocking strategies. + CompleteOrthogonalDecomposition is based on ColPivHouseholderQR and they thus achieve the same level of performance. -The above table has been generated by the bench/dense_solvers.cpp file, feel-free to hack it to generate a table matching your hardware, compiler, and favorite problem sizes. +The above table has been generated by the bench/dense_solvers.cpp file, feel-free to hack it to generate a table matching your hardware, compiler, and favorite problem sizes. */ -- cgit v1.2.3