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

109 message(s), the most recent arriving on Tue 24 February at 23:32 GMT.

• [eigen-commits] commit/eigen: 2 new changesets commits-noreply

• [eigen-commits] commit/eigen: hauke: Fixes #89. commits-noreply
• [eigen-commits] commit/eigen: bjacob: Patch by 'Wolf' from the issue tracker: commits-noreply

• [eigen-commits] commit/eigen: hauke: Fixed 32bit builds. commits-noreply
• [eigen-commits] commit/eigen: hauke: Fixed a warning. commits-noreply
• [eigen-commits] commit/eigen: hauke: Silenced type conversion warnings. commits-noreply
• [eigen-commits] commit/eigen: hauke: BenchTimer backport (clock_gettime & QueryPerformanceCounter). commits-noreply

• [eigen-commits] commit/eigen: hauke: Prevent temporaries for reductions. commits-noreply
• [eigen-commits] commit/eigen: ggael: optiization: make hybrid small/large outer products use the unrolled path commits-noreply
• [eigen-commits] commit/eigen: ggael: * resurected Flagged from Eigen2Support commits-noreply
• [eigen-commits] commit/eigen: ggael: make sure the correct diagoanl() function is called in trace() commits-noreply

• [eigen-commits] commit/eigen: ggael: make noalias works for coefficient based products commits-noreply

• [eigen-commits] commit/eigen: 7 new changesets commits-noreply
• [eigen-commits] commit/eigen: ggael: fix one useless temp & copy commits-noreply
• [eigen-commits] commit/eigen: hauke: Added debug only unit test for nesting ops - just run ./check nesting. commits-noreply
• [eigen-commits] commit/eigen: hauke: Fixed paste&copy error. commits-noreply
• [eigen-commits] commit/eigen: hauke: Added a non-diagonal product nesting test. commits-noreply
• [eigen-commits] commit/eigen: ggael: For the record, here is a solution for (a*b).diagonal, at the cost of extra copies commits-noreply
• [eigen-commits] commit/eigen: ggael: * as promised, remove the "optimization" for Product::diagonal() commits-noreply
• [eigen-commits] commit/eigen: ggael: document lazyProduct commits-noreply
• [eigen-commits] commit/eigen: hauke: Now variadic macro related warnings should be supressed as well under Linux. commits-noreply
• [eigen-commits] commit/eigen: 3 new changesets commits-noreply
• [eigen-commits] commit/eigen: 2 new changesets commits-noreply

• [eigen-commits] commit/eigen: ggael: * move dummy_precision and epsilon to NumTraits commits-noreply
• [eigen-commits] commit/eigen: ggael: add highest and lowest functions to NumTraits commits-noreply
• [eigen-commits] commit/eigen: ggael: Improved patch from Manuel Yguel: commits-noreply
• [eigen-commits] commit/eigen: myguel: Updated tests for enhanced AlignedBox with volume, diagonal and better handling of integral types. commits-noreply
• [eigen-commits] commit/eigen: ggael: finally here is a simple solution making (a*b).diagonal() even faster than a.lazyProduct(b).diagonal() !! commits-noreply
• [eigen-commits] commit/eigen: hauke: Fixed warning. commits-noreply
• [eigen-commits] commit/eigen: jitseniesen: unsupported/Eigen/AlignedVector3: dummy_precision is now in NumTraits commits-noreply

• [eigen-commits] commit/eigen: 3 new changesets commits-noreply
• [eigen-commits] commit/eigen: ggael: fix compilation (cwise and epsilon) commits-noreply
• [eigen-commits] commit/eigen: 2 new changesets commits-noreply
• [eigen-commits] commit/eigen: hauke: Here is the proper fix. commits-noreply
• [eigen-commits] commit/eigen: ptroja: std:: namespace fixup for more restricive compilers such as QNX's QCC commits-noreply
• [eigen-commits] commit/eigen: ptroja: std:: namespace fixup for more restricive compilers such as QNX's QCC commits-noreply

• [eigen-commits] commit/eigen: bjacob: work around brain dead ICC commits-noreply
• [eigen-commits] commit/eigen: 2 new changesets commits-noreply
• [eigen-commits] commit/eigen: ggael: fix a couple of ICE with gcc 4.0.1 commits-noreply
• [eigen-commits] commit/eigen: bjacob: Piotr's patch was missing many occurences of size_t. So, commits-noreply
• [eigen-commits] commit/eigen: bjacob: Piotr's patch was missing many occurences of size_t. So, commits-noreply
• [eigen-commits] commit/eigen: 2 new changesets commits-noreply

• [eigen-commits] commit/eigen: 2 new changesets commits-noreply

• [eigen-commits] commit/eigen: hauke: Workaround for compounds affected by #94. commits-noreply
• [eigen-commits] commit/eigen: 3 new changesets commits-noreply
• [eigen-commits] commit/eigen: ggael: avoid 2 redundant calls to resize commits-noreply
• [eigen-commits] commit/eigen: ggael: not all versions of gcc support -Wno-variadic-macros commits-noreply
• [eigen-commits] commit/eigen: jitseniesen: Use ReturnByValue to return result of ei_matrix_exponential() . commits-noreply

• [eigen-commits] commit/eigen: jitseniesen: Use ReturnByValue to return result of ei_matrix_function(), ... commits-noreply

• [eigen-commits] commit/eigen: 8 new changesets commits-noreply
• [eigen-commits] commit/eigen: jitseniesen: Update matrix_exponential test after API change in ei_matrix_function commits-noreply

• [eigen-commits] commit/eigen: ggael: fix stupid enum values commits-noreply

• [eigen-commits] commit/eigen: 2 new changesets commits-noreply
• [eigen-commits] commit/eigen: hauke: CMake cleanup. commits-noreply
• [eigen-commits] commit/eigen: ggael: add missing return value commits-noreply

• [eigen-commits] commit/eigen: ggael: compilation fix in ldlt() for non matrix types commits-noreply
• [eigen-commits] commit/eigen: ggael: compilation fix in ldlt() for non matrix types commits-noreply
• [eigen-commits] commit/eigen: orzel: NonLinearOptimization : clean 'mode' handling from the old minpack code : commits-noreply
• [eigen-commits] commit/eigen: 3 new changesets commits-noreply
• [eigen-commits] commit/eigen: hauke: Added missing precision/eps functions to AutoDiffScalar. commits-noreply
• [eigen-commits] commit/eigen: hauke: Added IsRowMajor enum to DenseBase. commits-noreply
• [eigen-commits] commit/eigen: orzel: fix compilation commits-noreply

• [eigen-commits] commit/eigen: ggael: extend the bench timer to allow benchmarking of parallel code, commits-noreply
• [eigen-commits] commit/eigen: hauke: Added getRealTime() for windows. commits-noreply
• [eigen-commits] commit/eigen: hauke: ups commits-noreply
• [eigen-commits] commit/eigen: 2 new changesets commits-noreply
• [eigen-commits] commit/eigen: 3 new changesets commits-noreply
• [eigen-commits] commit/eigen: orzel: provide default values for CXX, remove duplicate define commits-noreply

• [eigen-commits] commit/eigen: 2 new changesets commits-noreply
• [eigen-commits] commit/eigen: hauke: Attempt to fix PGI compilation issue. commits-noreply
• [eigen-commits] commit/eigen: 3 new changesets commits-noreply
• [eigen-commits] commit/eigen: bjacob: * FullPivLU: replace "remaining==0" termination condition (from Golub) by a fuzzy compare commits-noreply
• [eigen-commits] commit/eigen: bjacob: Further LU test improvements. I'm not aware of any test failures anymore, not even with huge numbers of repetitions. commits-noreply
• [eigen-commits] commit/eigen: bjacob: LDLT: commits-noreply
• [eigen-commits] commit/eigen: bjacob: actually, this is not even meant to be a termination criterion. so the proper fix is this. commits-noreply
• [eigen-commits] commit/eigen: bjacob: precision improvement. Wtf were we using sqrt(precision) for the cutoff? Replaced by precision*biggest. commits-noreply
• [eigen-commits] commit/eigen: bjacob: really fix the LDLt, at the expense of letting isPositiveDefinite() always return true (it was fundamentally broken anyway, especially as in 2.0 we don't even pivot at all). commits-noreply

• [eigen-commits] commit/eigen: ggael: * fix LDLT's default ctor use commits-noreply
• [eigen-commits] commit/eigen: ggael: make testsuite aware of EIGEN_CTEST_ARGS commits-noreply
• [eigen-commits] commit/eigen: ggael: add reconstructedMatrix() to LLT, and LUs commits-noreply
• [eigen-commits] commit/eigen: orzel: Provide "eigen" defines to decide which instruction set is used commits-noreply
• [eigen-commits] commit/eigen: orzel: provide a static method to describe which SIMD instructions are used commits-noreply

• [eigen-commits] commit/eigen: orzel: oops, fix typo commits-noreply
• [eigen-commits] commit/eigen: bjacob: * move the 's': InstructionsSet ---> InstructionSets commits-noreply
• [eigen-commits] commit/eigen: ggael: optimize inverse permutations commits-noreply
• [eigen-commits] commit/eigen: ggael: detect and implement inplace permutations commits-noreply
• [eigen-commits] commit/eigen: ggael: update matrix product selection rules for 1xSmallxLarge and the transposed case commits-noreply
• [eigen-commits] commit/eigen: jitseniesen: ComplexSchur: compute shift more stably, introduce exceptional shifts. commits-noreply

• [eigen-commits] commit/eigen: jitseniesen: ComplexSchur: fix bug introduced in my previous commit. commits-noreply
• [eigen-commits] commit/eigen: nerbonne : Fixed perf problems for vector subtraction: inlining wasn't always happening when necessary. commits-noreply

• [eigen-commits] commit/eigen: 10 new changesets commits-noreply
• [eigen-commits] commit/eigen: bjacob: add examples commits-noreply
• [eigen-commits] commit/eigen: bjacob: Remove the dot product's separate implementation and use cwiseProduct.sum instead. commits-noreply
• [eigen-commits] commit/eigen: orzel: * define COMPARE(, ), which prints expected/actual results in case of failure commits-noreply
• [eigen-commits] commit/eigen: bjacob: dot: handle the rowvector.dot(colvector) case where one needs to transpose. commits-noreply
• [eigen-commits] commit/eigen: 2 new changesets commits-noreply
• [eigen-commits] commit/eigen: orzel: Use a specialization of test_is_equal() instead of defining COMPARE() commits-noreply
• [eigen-commits] commit/eigen: bjacob: add missing inline keyword, thanks to Eamon. commits-noreply
• [eigen-commits] commit/eigen: bjacob: add ei_posix_memalign_realloc commits-noreply

• [eigen-commits] commit/eigen: bjacob: Only include <iosfwd> unless either EIGEN_DEBUG_ASSIGN is defined or we're in eigen2 support mode commits-noreply
• [eigen-commits] commit/eigen: orzel: misc cleaning commits-noreply
• [eigen-commits] commit/eigen: bjacob: comment out cerr's commits-noreply
• [eigen-commits] commit/eigen: hauke: Hide some internal stuff from the docs. commits-noreply
• [eigen-commits] commit/eigen: hauke: Added a generic reallocation implementation based on ei_aligned_malloc/_free. commits-noreply
• [eigen-commits] commit/eigen: bjacob: * fix ei_handmade_aligned_realloc (was calling realloc on wrong ptr) commits-noreply
• [eigen-commits] commit/eigen: bjacob: on 64-bit systems, glibc's malloc returns 16-byte aligned ptrs, and we now take advantage of that. commits-noreply
• [eigen-commits] commit/eigen: bjacob: forgot defined(...) commits-noreply
• [eigen-commits] commit/eigen: bjacob: remove the hack to make the static assertion on types actually show up. commits-noreply
• [eigen-commits] commit/eigen: bjacob: clarify EIGEN_DONT_ALIGN docs wrt versions commits-noreply