[Debichem-devel] Bug#670054: Bug#670054: rdkit: FTFBS on powerpc / s390x / sparc (test suite failures)

Greg Landrum greg.landrum at gmail.com
Mon Apr 23 03:55:28 UTC 2012


This is almost certainly a problem with byte-ordering and the RDKit's
method of handling serialization.

I'm sure it's not that tough to fix, and I will happily do the work,
but I don't have access to a big-endian machine to do the development
work and testing. Is there a straightforward solution to this?

-greg

On Sun, Apr 22, 2012 at 6:14 PM, Adam D. Barratt
<adam at adam-barratt.org.uk> wrote:
> Source: rdkit
> Version: 201112-1
> Severity: serious
> Tags: wheezy sid
>
> Hi,
>
> rdkit FTBFS on powerpc, s390x and sparc due to test suite failures.
> From the powerpc build log:
>
> 88% tests passed, 9 tests failed out of 76
>
> Total Test time (real) = 113.88 sec
>
> The following tests FAILED:
>          2 - testDataStructs (OTHER_FAULT)
>          4 - pyDiscreteValueVect (Failed)
>          5 - pySparseIntVect (Failed)
>          7 - testGrid (OTHER_FAULT)
>         22 - graphmoltestPickler (OTHER_FAULT)
>         42 - pyMolDescriptors (Failed)
>         66 - pyCMIM (Failed)
>         67 - pyRanker (Failed)
>         69 - pyFeatures (Failed)
> Errors while running CTest
> make[2]: *** [test] Error 8
> make[2]: Leaving directory `/build/buildd-rdkit_201112-1-powerpc-PES4nC/rdkit-201112/obj-powerpc-linux-gnu'
> dh_auto_test: make -j1 test ARGS+=-j1 returned exit code 2
> make[1]: *** [override_dh_auto_test] Error 29
> make[1]: Leaving directory `/build/buildd-rdkit_201112-1-powerpc-PES4nC/rdkit-201112'
> make: *** [build] Error 2
> dpkg-buildpackage: error: debian/rules build gave error exit status 2
>
> Full logs available via
> https://buildd.debian.org/status/package.php?p=rdkit
>
> Regards,
>
> Adam
>
>
>
>
> _______________________________________________
> Debichem-devel mailing list
> Debichem-devel at lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debichem-devel





More information about the Debichem-devel mailing list