[Debian-med-packaging] Bug#890117: bedtools FTBFS on big endian: Tools failing: bamtobed bamtofastq coverage genomecov groupby intersect jaccard map merge multicov negativecontrol

Charles Plessy plessy at debian.org
Mon Feb 11 11:18:03 GMT 2019


Le Wed, Dec 19, 2018 at 08:45:10AM +0100, Andreas Tille a écrit :
> Control: tags -1 help
> Control: tags -1 upstream
> Control: forwarded -1 https://github.com/arq5x/bedtools2/issues/676

Hi Andreas and everybody,

I bisected the build failure on the sparc64 porterbox kyoto.d.o, and
somewhat unsurprisingly, it pinpointed the following commmit, between
version 2.26.0 and 2.27.0:

    commit 93c11d1f01d6692bac05b3335850ef50947493ca
    Cause make test to exit with non-zero status in the event of a test failure.

I looked at the sparc64 logs for version 2.26.0 (wich did not fail to
build from source) and indeed, I see test failures that did not cause
a makefile errors, and that are not visible from the i386 build log.

I have not checked other platforms but my conclusion is that big-endian
versions have been buggy for some time wihtout any user noticing it.
Actually, I am not sure we have evidence that bedtools ever worked
properly on big-endian machines.

Altogether, this calls for removing bedtools from big endian architectures 
on Debian at the moment, until Upstream or somebody else steps up to do
the porting work.

I will fill a request for removal.

Have a nice day,

Charles

-- 
Charles Plessy
Debian Med packaging team,
http://www.debian.org/devel/debian-med
Akano, Uruma, Okinawa, Japan



More information about the Debian-med-packaging mailing list