[Debian-med-packaging] Bug#806675: HMMER 3.1b2-2

[Eddy, Sean]

Travis, the issue is that they want ARM and PPC64le support, which we currently don't offer (but may in the future -- Andreas, it is on our roadmap, but dependent on $$ and staff). We currently only support Intel-compatible and PPC-compatible platforms, because of our dependence on SSE2 and Altivec/VMX vector instructions.

Sean

> On Dec 3, 2015, at 9:40 AM, Andreas Tille <andreas at an3as.eu> wrote:
> 
> Hi Travis,
> 
> On Wed, Dec 02, 2015 at 02:12:25PM -0700, Travis Wheeler wrote:
>> Andreas -
>> 
>> I notice that you recently posted a changelog "hmmer 3.1b2-2 source
>> package in Ubuntu", in which you indicate a plan to "restrict
>> architectures to Intel architectures".
>> (https://launchpad.net/ubuntu/+source/hmmer/3.1b2-2)
> 
> To be correct:  I never ever posted anything on Ubuntu nor am I familiar
> with launchpad at all.  I'm a Debian developer working inside the Debian
> Med team that cares for packages in the field of biology and medicine
> inside Debian.  I have no idea at all what might become propagated to
> some Ubuntu channels where you somehow recived this news.
> 
>> I'm one of the HMMER developers.
> 
> Great. :-)
> 
>> I'm not familiar enough with the
>> launchpad infrastructure to ask questions within its framework, so I'm
>> e-mailing you directly.
> 
> Since you indicated that you would have directed your question to a
> public forum I keep the relevant bug log in CC[1] (simply "answer to
> all recipients" to do the same).
> 
>> May I ask why you're suggesting a restricted
>> architecture? HMMER is written to work across many architectures,
>> including AMD.
> 
> AMD remains supported.  I just forwarded the problem described in the
> Debian bug #806675 to Sean Eddy who was mentioned as contact address
> for HMMER and I have quotet his response in the packaging metadata[2].
> I'd be delighted to hear that HMMER would also be able to run on
> specifially arm64.
> 
>> Your post mentions bug 806675, but that doesn't seem to
>> be related to HMMER.
> 
> I would be delighted to hear that you would be able to solve the issue
> described in the Debian bug[1] and if so it would be very easy to drop
> the architecture restriction again.
> 
> Kind regards and thanks a lot for your input
> 
>     Andreas.
> 
> [1] https://bugs.debian.org/806675
> [2] https://anonscm.debian.org/viewvc/debian-med/trunk/packages/hmmer/trunk/debian/README.source?view=markup
> 
> 
> -- 
> http://fam-tille.de