On Thursday 05 January 2012 03:42:28 Bill Hart wrote: > So sumdiff_n would be defined for all x86_64 machines? >
sumdiff is already defined on all machines , although only fast if HAVE_NATIVE , the general sumdiff used is an old one from gmp ment to be used for fft's , it just splits it into L1 sized chunks , I seem to think that it was never very effective and probably slower than a separate add/sub for most values > What about addsub_n? in mpir-2.4/2.5 only atom,netburst and sandybridge dont have it > > Bill. > > On 4 January 2012 23:23, jason <ja...@njkfrudils.plus.com> wrote: > > I had forgotten this but mpir-2.5.0 does have a sumdiff for core2 :) > > > > On Dec 27 2011, 5:27 pm, Bill Hart <goodwillh...@googlemail.com> > > wrote: > >> In my FFT I make use of mpn_sumdiff_n and mpn_addsub_n. It seems these > >> are not exported even though there are generic C versions. > >> > >> Also, I see there is no sumdiff_n.as on core2 style machines. Is it > >> possible to include mpn_sumdiff_n.c in the library on such machines so > >> that it is included unconditionally for all machines? > >> > >> Is there a reason to not have an assembly optimised version for core2? > >> > >> Bill. > > > > -- > > You received this message because you are subscribed to the Google Groups > > "mpir-devel" group. > > To post to this group, send email to mpir-devel@googlegroups.com. > > To unsubscribe from this group, send email to > > mpir-devel+unsubscr...@googlegroups.com. > > For more options, visit this group at > > http://groups.google.com/group/mpir-devel?hl=en. > > > > -- You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-devel@googlegroups.com. To unsubscribe from this group, send email to mpir-devel+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en.
