Release v2.1.0
Release Notes
- Change usage of queues in MAGMA
- Add MKL GPU support for the dense format
- Implement write/read functions for distributed2d
- Support OpenMP GPU offload on iris nodes at ANL
- Accelerated GPU offload bml_add_ellpack and bml_multiply_ellpack
- Implement diagonalization, norm for distributed2d
- Add trace, transpose functions for distributed2d
- Initial implementation of Cannon's algorithm
- Implement first distributed functionalities