Yep, I just did a benchmark and on K8 I see that is true. So yeah, the question is, how on earth did it misdetect that model. The code looks totally solid and correct to me.
Bill. 2009/4/16 Jason Moxham <ja...@njkfrudils.plus.com>: > > On Thursday 16 April 2009 01:39:07 Bill Hart wrote: >> Wait a minute. Is that really true. Shouldn't x86_64 use the assembly >> code in the x86_64 directory, which should be considerably better than >> GMP 4.2.1? > > yes , it uses asm code in x86_64 dir , which is the same as 4.2.1 , pretty > much. > >> >> Bill. >> >> 2009/4/16 Jason Moxham <ja...@njkfrudils.plus.com>: >> > On Thursday 16 April 2009 01:15:17 Jeff Gilchrist wrote: >> >> On Wed, Apr 15, 2009 at 2:33 PM, Bill Hart <goodwillh...@googlemail.com> >> > >> > wrote: >> >> > As no issues with the MPIR 1.1.0 release candidate were found after >> >> > extensive testing, we are making release candidate 1 the final >> >> > release. >> >> >> >> Here are some benchmarks. With K8, MPIR is very competitive with the >> >> new GMP. For some reason my core2 builds are strange. The MPIR 1.1 >> >> library is much slower than 1.0. Config in 1.0 and 1.1 is detecting >> >> the system as: x86_64-unknown-linux-gnu so trying to use K8 code with >> >> it. I'm not sure why there is such a speed difference between the 1.0 >> >> and 1.1 K8 code on a Core 2 though. >> > >> > The detection code is a little bit different than before. If config,guess >> > recognizes your cpu then it reports k8-*-* or penryn-*-* etc . If it >> > doesn't recognize the exact cpu then it will just report x86-*-* or >> > x86_64-*-* whatever it can detect , and the resultant build will use the >> > lowest common denominator , which in your case was x86_64 so you get a >> > build which runs like the old gmp-4.2.4 , safe but slow. >> > The real question is why it hasn't detected your cpu properly? >> > >> > can you do a cat /proc/cpuinfo >> > >> > thanks >> > Jason >> > >> >> MPIR Benchmarks >> >> =============== >> >> mpz_mul(t, r, r); >> >> mpz_sub_ui(t, t, 2); >> >> mpz_mod(r, t, p); >> >> >> >> Intel Xeon E5405 @ 2.0GHz (Family 6, Model 23, Stepping 6) >> >> Linux 64bit, gcc version 4.1.2 >> >> 83338 bits (25088 digits) >> >> Library Time1 Time2 >> >> ---------- --------- --------- >> >> GMP 4.3.0 2m34.857s 2m34.835s >> >> MPIR 1.0.0 3m17.198s 3m17.288s >> >> GMP 4.2.4 5m38.156s 5m38.173s >> >> MPIR 1.1.0 5m45.308s 5m45.236s >> >> >> >> >> >> AMD Opteron 248 @ 2.2GHz (Family 15, Model 37, Stepping 1) >> >> Linux 64bit, gcc version 3.4.6 >> >> 83338 bits (25088 digits) >> >> Library Time1 Time2 >> >> ---------- --------- --------- >> >> GMP 4.3.0 1m35.091s 1m34.815s >> >> MPIR 1.1.0 1m36.480s 1m35.923s >> >> MPIR 1.0.0 1m37.501s 1m36.857s >> >> GMP 4.2.4 2m56.352s 2m55.702s >> >> >> >> Jeff. >> >> > > > > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
