Hi,

I just ran into what I think is a bug in atlas 3.3.13-1 which I have 
reported to the ATLAS group. I observed it in the real (single and 
double) SVD bi-diagonalization routines (which seg faulted on some 
specific examples). The code worked flawlessly with vanilla BLAS. That 
is why I think the bug is in the atlas BLAS.

I thought I would report this here too, so that this version of atlas is 
not moved into the stable tree prematurely.

Thanks to all the fink-developers for their great work!

--shiv--


_______________________________________________
Fink-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-users

Reply via email to