You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi @ivan-pi yep pretty much, but including support to BLAS and using meta-programming to simplify the code base as well. I didn't found a alternative to MKL's blas95, so here I am.
It would be nice to stdlib have some LAPACK, FFTW and BLAS integration of some sort to speed up some handy high level interface like eig, fft and etc. However, this brings that question about external dependencies. I think we should take insights from other communities like Julia, Boost/C++ and SciPy about it, but since Fortran has a science-driven community it wouldn't hurt to depend on such widely used and popular libraries.
Ok, so the missing generic interface to BLAS is one of the issues.
Without doubt, in stdlib we will need to agree on some form of LAPACK, FFTW, and BLAS integration. Having generic interfaces available would definitely simplify the programming.
I should start with BLAS.
Other references:
Intel MKL
oneAPI BLAS
oneAPI LAPACK
The text was updated successfully, but these errors were encountered: