Bug#807492: libmath-bigint-gmp-perl: FTBFS on 32 bit platforms

[Dominic Hargreaves]

Source: libmath-bigint-gmp-perl
Version: 1.46-1
Severity: serious
Justification: FTBFS on platforms where it built before
Forwarded: https://rt.cpan.org/Public/Bug/Display.html?id=103517

The new release, 1.46, introduced a regression on 32 bit platforms
where perl is compiled with 64 bit support. You can see that it has
failed to build on armel, armhf, i386, mips, mipsel, hppa, kfreebsd-i386
as a result:

https://buildd.debian.org/status/package.php?p=libmath-bigint-gmp-perl&suite=unstable

There is discussion about this problem at the CPAN bug above.
Since this collides with the impending perl 5.22 transition, I
checked and it appears that the impact of removing this would not be
huge; so although I will add it as a blocker of the transition, if it
isn't fixed in the next few days it'll probably get removed.

Cheers,
Dominic.