Version 1.0.0 is now released and is available from the Files section.

This release add support for single precision and upper triangular matrices.

It additionally offer some minor perofmrance tweaks and bug fixes (including the NaN fix for the L^T case) and support for incx<0.

v0.9.0 of the ASEArch blas have been released! Get it from the Files Section.

This provides the source code for a much imporved triangular solve written in CUDA.