Hi Jeff,

>     Ok, this is on fairly short notice considering the changes required.

    I recommend to start with copying the CUSP sources and migrate it
    over to VECCUDA by replacing any use of cusp::array1d to a raw CUDA
    handle. Operations from CUSP should be replaced by CUBLAS calls.


It's hard to imagine any performance benefit from this unless CUSP
sucks. What am I missing?

This is not about performance, but about providing the ability for users to 'implant' their own memory buffers. CUSP doesn't allow it (which was the initial point of this thread).