This is a fork of the GPU/CPU sparse, symmetric indefinite direct
linear solver SSIDS that is available as part of the open-source 
sparse parallel robust algorithms library SPRAL:

  https://github.com/ralna/spral
  https://ralna.github.io/spral/

Since the SPRAL GPU code no longer works, the GALAHAD fork is only 
for CPUs. The fork also adds that following functionality

 *  multiple real precision support
 *  both 32 and 64 bit integer support
 *  native nested-dissection ordering rather than requiring
    an external version of MeTiS
 *  Python and Julia support as well as C
 *  adoption of GALAHAD coding style where possible

Jari Fowkes, Nick Gould and Alexis Montoison
for GALAHAD productions
August 14th 2025
