Issues

• global ID for nodes is wrong

  SCOREC/msi#26 · Filter by author seegyoung opened on Sep 28, 2018
  3

• We directly use PETSc APIs, we should instead use LAS, with a macro-templated backend set to PETSc during compilation.

  enhancement
  SCOREC/msi#24 · Filter by author wrtobin opened on Sep 12, 2018
  

• Use an MSI_SCALAR type instead of assuming double

  SCOREC/msi#22 · Filter by author wrtobin opened on Aug 30, 2018
  

• consider supporting compile time definition of integer type

  enhancement
  SCOREC/msi#11 · Filter by author cwsmith opened on Jun 2, 2018

• The dev and master branches currently rely on apfFieldData.h which is a non-public core header.

  invalid
  SCOREC/msi#8 · Filter by author wrtobin opened on May 16, 2018
  2

• CMake configuration cleanup

  enhancement
  SCOREC/msi#7 · Filter by author wrtobin opened on May 16, 2018 · 0.1
  1

  

• Don't have separate matrix_solve and matrix_mult classes, have a single class that manages a parallel or serial matrix, which both support multiplication and solving.

  enhancementlinear algebra
  SCOREC/msi#5 · Filter by author wrtobin opened on Apr 18, 2018 · 0.2
  1

  

• Replace linear algebra api using LAS for a backend so we have access to multiple linear algebra systems without rewriting any msi code.

  enhancementlinear algebra
  SCOREC/msi#4 · Filter by author wrtobin opened on Apr 18, 2018 · 0.2