I don't want to *only* support OpenBlas and hardcode that into build scripts instead of ATLAS. For one, it doesn't support ARM (=not at all) or AMD Bulldozer (=probably not that fast on the newest AMD chip). There are also bugs, so I would rather implement a more flexible setup than jump ship completely and then find that it crashes&burns on some ancient Itanium box that happens to be on the "Sage supported platforms" list.
For the record, the https://github.com/xianyi/OpenBLAS/blob/develop/TargetList.txt. Can, in theory, be compiled on OSX. Not sure how it compares to Apple's own sauce. Having a wrapper library would also give us much more control to deal with bugs in BLAS implementations, e.g. one could use ATLAS and override a buggy function with the reference BLAS. The downside is of course added complexity. It is also technically slower, but I don't think in a measurable way. The most performance-critical people will always go for static linking and non-PIC binaries but that's not really an option for a general-purpose system. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sage-devel?hl=en.
