* Manuel A. Fernandez Montecelo <manuel.montez...@gmail.com>, 2012-02-04, 23:26:
libogre-dev is marked as "Multi-Arch: same", but contents of
/usr/include/OGRE/OgreBuildSettings.h varies with architecture
endiannes. Please see the attached diff.
So I should remove "Multi-Arch: same" line from the control and that's
it,
Yes, that's the simplest way to fix it.
or what's the way to fix it?
If you are feeling adventurous, you might want to keep M-A:same and make
the header file identical across architectures, e.g. by exploiting the
fact that GCC defines the __BYTE_ORDER__ macro.
Also, is there any way for maintainers to discover this in advance,
before uploading packages?
Nothing bulletproof.
What I did for my packages, was to build the package for two
architectures (i386, amd64), unpack the resulting .deb (with dpkg
--raw-extract) and then run diff against unpackaged directories. But of
course this procedure wouldn't detect anything for libogre-dev, as i386
and amd64 are both little-endian.
That said, the existence of *.h.in files in the source package could
have raised your suspicion. If header files (or any other files, for
what it's worth...) are generated at build time, there's a chance that
their contents will be architecture-dependent.
Or after uploading, a kind of check like lintian or piuparts? I haven't
seen any error related to this in the PTS interface and the reachable
services from it.
A while ago I announced existences of a tracker for problems like
that[0] (which I'm use to file bugs when I'm bored :P).
It's not connected to the PTS, though.
[0] http://lists.debian.org/debian-devel/2011/11/msg00508.html
--
Jakub Wilk
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
