Triangular Arrays in julia?

[nalimilan]

@ChrisRackauckas

Is it still going to be much better to use BLAS? For square lower triangular matrices you have less than half of the operations to do, and with @polly being added, could the difference between BLAS and not-BLAS be around 2x?

Even if there are only about half operations to do, that 1/2 is a constant factor: the cost is still increasing with the size of the matrix. So that doesn’t really make a difference from standard arrays, except for changing the threshold above which calling BLAS is beneficial.

@ChrisRackauckas

(Is there a good way to switch between the BLAS and generic operations to do a quick test what the difference is like?)

For matrix multiplication, you can call Base.LinAlg.generic_matmatmul directly. I’m not sure for what operations and types BLAS is called, and for which pure Julia code is used instead.