Bug#828717: Acknowledgement (gmp-ecm: 7.0 regresses on 32 bit platforms)

Martin Pitt mpitt at debian.org
Mon Jun 27 07:20:11 UTC 2016


Hello again,

sorry, this evaded me before, here it is:

| GMP-ECM 7.0.1 [configured with GMP 6.1.0] [P-1]
| Input number is 2^1009-1 (304 digits)
| Using B1=1000, B2=17248, polynomial x^1, x0=1688984659
| Step 1 took 4ms
| Step 2 took 16ms
| ********** Factor found in step 2: 3454817
| Found prime factor of 7 digits: 3454817
| Composite cofactor (2^1009-1)/3454817 has 298 digits
| GMP-ECM 7.0.1 [configured with GMP 6.1.0] [P-1]
| Input number is (2^1009-1)/3454817 (298 digits)
| Using B1=5000, B2=9972-1389888, polynomial x^1, x0=4144063670
| Step 1 took 20ms
| ecm: ../pm1fs2.c:1828: poly_from_sets_V: Assertion `(__builtin_constant_p (1UL) && (1UL) == 0 ? ((F[deg])->_mp_size < 0 ? -1 : (F[deg])->_mp_size > 0) : __gmpz_cmp_ui (F[deg],1UL)) == 0' failed.
| Aborted (core dumped)
| ############### ERROR ###############
| Expected return code 6 but got 134
| make: *** [longcheck+test+pm1+-redc] Error 1
| Makefile:17: recipe for target 'longcheck+test+pm1+-redc' failed

Same error on armhf and i386, and this does not happen on 64 bit.

I can reproduce this in sid with autopkgtest's schroot runner, so it's
not specific to Ubuntu or LXC/LXD/QEMU:

  autopkgtest gmp-ecm -- schroot sid-i386

Thanks,

Martin
-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)



More information about the debian-science-maintainers mailing list