Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.
The LU substitution is buggy for dense matrices - the triangular solvers are fine when used standalone, but the full lu-substitution yields wrong results
The compute kernel had a bug with the backward substitution, which is now fixed. Corrected kernel is included in ViennaCL 1.0.1.