gmp-devel Digest, Vol 1, Issue 86

Torbjorn Granlund tege at
Mon Sep 15 13:45:39 CEST 2003

Paul.Zimmermann at (Paul Zimmermann) writes:

     To accomodate vector machines, we might want to put mpn_mul_n,
     mpn_sqr_n, mpn_mul, and the various karatsuba and toom routines
     in a single file.  Or, we might want to move mpn_sqr_n to
     Any informed opinions?
  Perhaps we could put all squaring functions in sqr_n.c, while still
  including them in mul_n.c...
Yes, that's an alternative too.  But if we put all the functions
on one file, including the highest-lever mpn_mul, we will get the
best inlining.

  This could be replaced by mpn_cmp (ap + dn, bp + dn, an - dn), isn't it?

Yes, that'd be even shorter.


More information about the gmp-devel mailing list