On Monday, February 17, 2014 4:40:30 PM UTC+1, Bill Hart wrote: > > As far as I can tell, on Windows, GMP simply moves registers about before > and after calling the *nix code. They use some m4 macros for this I think. > So we can potentially do better by having dedicated Windows 64 code. On the > other hand, register renaming is essentially free on modern processors. > > Indeed, basically the difference between the x86_64 and the win64 abis is the number of registers used to pass arguments to functions. So GMP just add some black magic before and after the assembly routines to be able to use the same core routines.
-- You received this message because you are subscribed to the Google Groups "mpir-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to mpir-devel+unsubscr...@googlegroups.com. To post to this group, send email to mpir-devel@googlegroups.com. Visit this group at http://groups.google.com/group/mpir-devel. For more options, visit https://groups.google.com/groups/opt_out.
