On Saturday 06 June 2009 17:43:07 Fredrik Johansson wrote: > On Sat, Jun 6, 2009 at 4:11 PM, Jason Moxham<ja...@njkfrudils.plus.com> wrote: > > Looking at the existing toom code there are a lot of > > mpn_addmul_1(....,2^k) > > we should change this to > > mpn_addlsh_n(....,k) > > and write asm functions for this , for K8 this would be a macro to call > > addmul_1 , and for core2,K10 a new function > > Various other combinations can be done faster than we do at the moment , > > we just need to choose the most common or the easiest. > > On a related note, I would find an mpz_add(sub)lsh_n function extremely > useful, for addition of floating-point numbers with different magnitude / > exponent. Doing this by zero padding (possibly also forcing reallocation) > is a pretty big waste of electricity. > > Fredrik
Yeah , I use the same for testing floating point and multiplication mod 2^k+-1 mpz_addmul_2exp() mpz_add_2exp() mpz_mulmod_2exp() mpz_mulmod_2expm1() ....etc Most were just fillers to make code simpler to read , I think only one or two were efficient. mulmod_2expm1 was one , it was faster that the fft in gmp-4.2.1! Jason > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---