The single biggest improvement in my mind would be made to MPIR. And I know the MPIR devels would absolutely love to have them contribute.
Jason and Brian have been working very hard on simplifying MPIR so that it can be more easily modified. There are some instances in flint that would benefit. If he gets in contact with me I could discuss this. The overall affect on Sage would not be quite as great. It would only affect certain types of polynomials on Sparc. But it certainly needs to be done. Pari, ECL and MPFR would all directly benefit from speeding up MPIR. I think of the three, MPFR is the one most likely to have assembly code. I believe ATLAS does autotuning, so it would speed up linbox automatically. There is some assembly code in NTL I think (or could be). However most of the relevant functionality is not used any more. There is a possibility for speeding up M4RI, in theory. However, it has been difficult to beat the C compiler on that project. Bill. On 5 Aug, 09:17, David Kirkby <david.kir...@onetel.net> wrote: > On 5 August 2010 09:00, David Kirkby <david.kir...@onetel.net> wrote: > > > Following a talk I gave at the London OpenSolaris User Group (LOSUG), > > I had some discussions with someone who knows SPARC assembler, and > > would be willing to devote some time to addling SPARC assembly > > support. The question I have, is where would his efforts be best > > directed? Pari, Flint, MPIR, MPFR, ECL etc might all be candidates for > > optimised 64-bit SPARC assembly code, but I'd like to know who to > > suggests he contacts > > > Dave > > I guess a related question is what developers would most welcome > someone with SPARC assembly language skills to contribute to their > project? > > Dave -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
