Hmm, that one is a mystery. I compiled for atom on my machine and it works fine.
The code doesn't do anything different at 51 limbs.... Can I just check, it segfaulted *before* you removed mpn/x86_64/atom/ sqr_basecase.as and you were doing ./speed -s 1-100 -t 1 mpn_sqr_basecase when it segfaulted? If so, I really don't have an explanation for that. The only odd thing in the assembly source is db 0x26. I'll have to look up which instruction that is. But surely atom supports it. Bill. On 17 February 2014 22:29, Bill Hart <goodwillh...@googlemail.com> wrote: > Oh! Maybe it only works up to 50 limbs. I'll check. That's ok anyway. > > Bill. > > > On 17 February 2014 22:28, Jean-Pierre Flori <jpfl...@gmail.com> wrote: > >> >> >> On Monday, February 17, 2014 10:26:48 PM UTC+1, Bill Hart wrote: >>> >>> That is more of a worry. I assume it is using mpn/x86_64/atom/ >>> sqr_basecase.as? >>> >> Yes. >> >>> >>> I have no idea why it should segfault only one time out of 50. That's >>> just strange. >>> >>> i was not very clear. >> I did not mean it segfaulted after 50 complete runs. >> I meant after the 50th test of one run: >> >> 49 ... >> 50 ... >> Segfault! >> >> > -- 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.
