diff options
| author | B. Watson <yalhcru@gmail.com> | 2022-03-12 14:38:22 -0500 |
|---|---|---|
| committer | B. Watson <yalhcru@gmail.com> | 2022-03-12 14:38:22 -0500 |
| commit | 4e008c23b97953c54bc99e50f6b50144c8b83431 (patch) | |
| tree | 54b965faeaf6bdd1a12cee17d44983b1e6e083f8 /academic | |
| parent | 3f5005bf95bb5527ecdd748f2e8482925633aacb (diff) | |
academic/spqr: Wrap README at 72 columns.
Signed-off-by: B. Watson <yalhcru@gmail.com>
Diffstat (limited to 'academic')
| -rw-r--r-- | academic/spqr/README | 22 |
1 files changed, 13 insertions, 9 deletions
diff --git a/academic/spqr/README b/academic/spqr/README index 051977e0e6634..8bb80aef4f371 100644 --- a/academic/spqr/README +++ b/academic/spqr/README @@ -1,18 +1,22 @@ SuiteSparseQR is an implementation of the multifrontal sparse QR -factorization method. Parallelism is exploited both in the BLAS and -across different frontal matrices using Intel's Threading Building Blocks, -a shared-memory programming model for modern multicore +factorization method. Parallelism is exploited both in the BLAS and +across different frontal matrices using Intel's Threading Building +Blocks, a shared-memory programming model for modern multicore architectures. It can obtain a substantial fraction of the theoretical peak performance of a multicore computer. This package is part of SuiteSparse. -This requires a BLAS/LAPACK implementation. Choose one of these package sets: +This requires a BLAS/LAPACK implementation. Choose one of these package +sets: + * OpenBLAS (includes both a BLAS and a LAPACK implementation) * atlas (includes both a BLAS and a LAPACK implementation) * blas, lapack (the Netlib reference implementations) -If more than one set is installed (assuming there are no packaging conflicts) -then the auto-detection will use the first implementation from this list that -it finds. If in doubt, choose the Netlib reference implementations; other -packages that require a BLAS or LAPACK implementation may not build if they -are not configured to detect/use alternate implementations. + +If more than one set is installed (assuming there are no packaging +conflicts) then the auto-detection will use the first implementation +from this list that it finds. If in doubt, choose the Netlib reference +implementations; other packages that require a BLAS or LAPACK +implementation may not build if they are not configured to detect/use +alternate implementations. |