Permalink
Commits on Feb 26, 2017
Commits on Feb 23, 2017
  1. Merge pull request #221 from timmoon10/blas1like_openmp

    Parallelizing several BLAS 1 (-like) routines with OpenMP
    poulson committed on GitHub Feb 23, 2017
Commits on Feb 22, 2017
  1. Parallelizing some routines with OpenMP simd pragma.

    This method is recommended to avoid modulo and division in collapse pragma. However, the collapse pragma is retained in EntrywiseMap, IndexDependentFill, and IndexDependentMap since an input function with conditional branching could reduce the effectiveness of SIMD vectorization.
    timmoon10 committed Feb 22, 2017
Commits on Feb 19, 2017
  1. Pullying secular fix into master

    poulson committed Feb 19, 2017
Commits on Feb 18, 2017
Commits on Feb 14, 2017
  1. Forcing zero shifts in some instances in the bidiagonal QR algorithm …

    …even if the tolerance is relative to the maximum singular value
    poulson committed Feb 14, 2017
Commits on Feb 4, 2017
  1. Update SparseInvCov.cpp

    Passing scalars by reference rather than by value in matrix functions in `SparseInvCov`. Thanks to @haimav for noticing this issue.
    poulson committed on GitHub Feb 4, 2017
Commits on Jan 29, 2017
  1. Merge pull request #217 from AidanGG/master

    Fixed default grid height calculation.
    poulson committed on GitHub Jan 29, 2017
Commits on Jan 16, 2017
  1. Merge pull request #216 from AidanGG/master

    Overloaded operator* for Ranges.
    poulson committed on GitHub Jan 16, 2017
Commits on Jan 4, 2017
  1. Fixing several issues with the MPS readers. Several tweaks to the IPM…

    …s are staged but not yet consistent enough to commit (and allow for solving DFL001, TUFF, PILOT, and KB2 in sparse double-precision).
    poulson committed Jan 4, 2017
Commits on Dec 31, 2016
  1. Merge pull request #214 from elemental/refactor_ipm

    Several bugs have now been fixed, an MPS format reader has been introduced, and the stability of the affine LP IPM is now improved via scaling similar to the direct case. The Interior Point Methods still deserve significantly more refactoring and rescaling attention.
    poulson committed on GitHub Dec 31, 2016
  2. Improving stability of affine LP solver by introducing a scaling simi…

    …lar to the direct case. This could likely be improved with a dynamic rescaling based upon the norms of (x,s) and (y,z) at the previous iteration.
    poulson committed Dec 31, 2016
  3. Fixing bugs in the SparseMatrix implementations of the variant 2 Dant…

    …zig selector and affine QP initialization (which effected both the affine QP and LP solvers)
    poulson committed Dec 31, 2016
  4. Extended MPS reader

    poulson committed Dec 31, 2016
  5. Making the MPS reader independent of the matrix format

    poulson committed Dec 31, 2016
Commits on Dec 30, 2016
  1. The MPS reader now appears to work for sequential dense matrices

    poulson committed Dec 30, 2016
  2. Beginning to add support for an MPS reader

    poulson committed Dec 30, 2016
Commits on Dec 29, 2016
  1. Further encapsulating sparse-direct solver helper routines

    poulson committed Dec 29, 2016
Commits on Dec 28, 2016
  1. Moving ChangeFrontType and LDL into [Dist]SparseLDLFactorization

    poulson committed Dec 28, 2016
  2. Further encapsulating sparse-direct solver functionality

    poulson committed Dec 28, 2016
  3. Fixing typo from the last commit

    poulson committed Dec 28, 2016
  4. Adding iterative refinement implementation into [Dist]SparseLDLFactor…

    …ization
    poulson committed Dec 28, 2016
  5. Further incorporating [Dist]SparseLDLFactorization

    poulson committed Dec 28, 2016
Commits on Dec 27, 2016
Commits on Dec 26, 2016