Strange crash with the MSSM

[Benjamin Fuks]

Dear friends,

I am computing the following process with madgraph (default cards, model, etc.):
  import model mssm
  generate p p > x1+ x1- QCD=99 QED=2
  add process p p > x1+ x1- j QCD=99 QED=2
  add process p p > x1+ x1- j j QCD=99 QED=2

The generated output cannot however compile (see the error message below). I have tried bopth on linux and macos and the problem persists.

Do you have an idea about what's going on and how to fix it? Thanks in advance!

Cheers,

Benj

*******

     gfortran -O -w -ffixed-line-length-132 -w -o gensym symmetry.o setcuts.o cuts.o cluster.o myamp.o genps.o initcluster.o setscales.o reweight.o get_color.o idenparts.o matrix1.o matrix10.o matrix11.o matrix12.o matrix13.o matrix14.o matrix15.o matrix16.o matrix17.o matrix18.o matrix19.o matrix2.o matrix20.o matrix21.o matrix22.o matrix23.o matrix24.o matrix25.o matrix26.o matrix27.o matrix28.o matrix29.o matrix3.o matrix30.o matrix31.o matrix32.o matrix33.o matrix34.o matrix35.o matrix36.o matrix37.o matrix38.o matrix39.o matrix4.o matrix40.o matrix41.o matrix42.o matrix43.o matrix44.o matrix45.o matrix46.o matrix47.o matrix48.o matrix49.o matrix5.o matrix50.o matrix51.o matrix52.o matrix53.o matrix54.o matrix55.o matrix56.o matrix57.o matrix58.o matrix59.o matrix6.o matrix60.o matrix7.o matrix8.o matrix9.o -L../../lib/ -ldhelas -ldsample -lmodel -lgeneric -lpdf -lcernlib
     Undefined symbols for architecture x86_64:
       "_ffv1c1p0_3_", referenced from:
           _matrix13_ in matrix13.o
           _matrix14_ in matrix14.o
           _matrix15_ in matrix15.o
           _matrix16_ in matrix16.o
           _matrix17_ in matrix17.o
           _matrix18_ in matrix18.o
           _matrix19_ in matrix19.o
           ...
     ld: symbol(s) not found for architecture x86_64
     collect2: error: ld returned 1 exit status
     make: *** [gensym] Error 1

[Benjamin Fuks (fuks)]

I forgot to say: I am using the latest v1.5.11 version of MadGraph

[Olivier Mattelaer (olivier-mattelaer)]

Hi Benj,

Looks like a problem of order in the aloha call. This seems created by a conflict of order between the tag
asking for the massless propagator in ALOHA and the one asking for the conjugate routine.

Could you confirm that you have in the Source/DHELAS directory,
a routine which is ffv1p0c1_3?
If so this is just a name clash between the way aloha set the name routine and the way madgraph does.

Cheers,

Olivier

PS@Valentin: This proves my point that we need a unique routine/convention for setting the name of the function.

On Aug 7, 2013, at 2:44 AM, Benjamin Fuks <email address hidden> wrote:

> I forgot to say: I am using the latest v1.5.11 version of MadGraph
>
> --
> You received this bug notification because you are a member of MadTeam,
> which is subscribed to MadGraph5.
> https://bugs.launchpad.net/bugs/1209113
>
> Title:
> Strange crash with the MSSM
>
> Status in The MadGraph Matrix Element Generator version 5:
> New
>
> Bug description:
> Dear friends,
>
> I am computing the following process with madgraph (default cards, model, etc.):
> import model mssm
> generate p p > x1+ x1- QCD=99 QED=2
> add process p p > x1+ x1- j QCD=99 QED=2
> add process p p > x1+ x1- j j QCD=99 QED=2
>
> The generated output cannot however compile (see the error message
> below). I have tried bopth on linux and macos and the problem
> persists.
>
> Do you have an idea about what's going on and how to fix it? Thanks in
> advance!
>
> Cheers,
>
> Benj
>
>
> *******
>
> gfortran -O -w -ffixed-line-length-132 -w -o gensym symmetry.o setcuts.o cuts.o cluster.o myamp.o genps.o initcluster.o setscales.o reweight.o get_color.o idenparts.o matrix1.o matrix10.o matrix11.o matrix12.o matrix13.o matrix14.o matrix15.o matrix16.o matrix17.o matrix18.o matrix19.o matrix2.o matrix20.o matrix21.o matrix22.o matrix23.o matrix24.o matrix25.o matrix26.o matrix27.o matrix28.o matrix29.o matrix3.o matrix30.o matrix31.o matrix32.o matrix33.o matrix34.o matrix35.o matrix36.o matrix37.o matrix38.o matrix39.o matrix4.o matrix40.o matrix41.o matrix42.o matrix43.o matrix44.o matrix45.o matrix46.o matrix47.o matrix48.o matrix49.o matrix5.o matrix50.o matrix51.o matrix52.o matrix53.o matrix54.o matrix55.o matrix56.o matrix57.o matrix58.o matrix59.o matrix6.o matrix60.o matrix7.o matrix8.o matrix9.o -L../../lib/ -ldhelas -ldsample -lmodel -lgeneric -lpdf -lcernlib
> Undefined symbols for architecture x86_64:
> "_ffv1c1p0_3_", referenced from:
> _matrix13_ in matrix13.o
> _matrix14_ in matrix14.o
> _matrix15_ in matrix15.o
> _matrix16_ in matrix16.o
> _matrix17_ in matrix17.o
> _matrix18_ in matrix18.o
> _matrix19_ in matrix19.o
> ...
> ld: symbol(s) not found for architecture x86_64
> collect2: error: ld returned 1 exit status
> make: *** [gensym] Error 1
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/madgraph5/+bug/1209113/+subscriptions

[Benjamin Fuks (fuks)]

Hi Olivier,

The routine is indeed there.

Cheers,

Benj

[Olivier Mattelaer (olivier-mattelaer)]

Thanks Benj,

Sorry for the delay (was rushing on a grant proposal)

I've just fix the code, this is corrected in version 1.5.12 which should be public very soon.
I still have to add an unit-test to ensure that this will not happen again and I push the fix.

Cheers,

Olivier