Begin forwarded message:
Date: Fri, 22 Nov 2013 03:59:01 -0800 From: Paul Lutus <lut...@gmail.com> To: Erik Massop <e.mas...@hccnet.nl> Subject: Re: [sage-devel] Re: Sage 5.12 fails to build Atlas, 64-bit SuSE 13.1 On 11/22/2013 03:08 AM, Erik Massop wrote: > On Thu, 21 Nov 2013 16:28:52 -0800 (PST) > lutusp <lut...@gmail.com> wrote: > >> On Thursday, November 21, 2013 1:23:14 PM UTC-8, lutusp wrote: >>> matime: Assertion `fscanf(fp, "%lf", &mflop[i])' failed. > ... >> Addendum: Sage version 5.11 successfully compiled on the same machine with >> no changes. > This looks like the bug [1] in glib 2.18, in which scanf fails to read > "0e+00". See the thread "ATLAS build fails on Core i7-3770" on this > list. > > Was 5.11 built using the same glibc version? Yes -- same machine, everything the same. On seeing the error, I immediately replaced the 5.12 TAR source archive with the 5.11 and compiled again, successfully. I just checked, and my glibc version is 2.18. So I have no idea why the prior version successfully compiled (assuming the Atlas library code is unchanged). BTW for those who don't know, to check the glibc version, simply run the library binary: # /lib/libc.so.6 GNU C Library (GNU libc) stable release version 2.18 (git ), by Roland McGrath et al. (...) -- Paul Lutus http://www.arachnoid.com -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/groups/opt_out.
