Ilhan Polat

Thanks Ralf! I covered quite some distance after your comment and @matthew-brett's tooling. So the issue was 2-fold. Our `dev.py` wasn't doing the right thing which I'll send the PR...

For the interested here are the differences over the dependencies of OpenBLAS Working one ```bash C:\Users\ilhan> dlldiag trace C:\Users\ilhan\AppData\Local\Programs\Python\Python310\Lib\site-packages\scipy\.libs\libopenblas.dll DLL Diagnostic Tools version 0.0.18 Copyright (c) 2019-2021 Adam Rehn Parsing...

On top of what @lucascolley mentioned, NumPy does not use the safe norm from LAPACK but does its own unscaled dot product hence bypasses the safety guards. If you don't...

Should we fix 2-norm working with axis? I can cook up something in C or Cython and `nrm2` is not complicated. Based on what we discussed over kron deprecation, not...

It doesn't have to use the BLAS code as all based on Anderson's work. We can write our own native version of nrm2 in C or Cython. Not sure if...

Overall judging by the discussions oat the array api, it seems to me the linalg parts are not well thought out.

@ev-br Is this all we need from FITPACK or there will be more to be translated from it? I can't make heads or tails with this one in terms of...

> There will be more @ilhan :-).  > You mean the python translation and its C work routines? Ah no, I mean the subject matter and the unfortunate Fortran...

> 2D stuff has several tricks on top, which are still to be deciphered, in a sequel not this PR. Most likely, it's something akin to what people do to...

> Meanwhile, need a hand with LAPACK stuff in the lbfgsb PR? I won't say no to any help with Fortran 😅 Just give me some more time to make...