GSI Forum: Event Generators

Home » PANDA » PandaRoot » Event Generators » FTF generator

Show: Today's Messages :: Polls :: Message Navigator

FTF generator [message #17958]

Wed, 04 March 2015 16:27

Aida Galoyan
Messages: 79
Registered: May 2007
Location: Dubna

continuous participant

From: *ikp.kfa-juelich.de

Hi Ralf,

why do you replace files with "c" extension
(Test30Materials.c, Test30HadronProduction.c and so on)
on the files *.cxx .
After your changing, FTFmake is not working. It crashes.

If I return *.c, FTFmake works and FTFGen exe-file is created.

Aida

Report message to a moderator

Re: FTF generator [message #17959 is a reply to message #17958]

Wed, 04 March 2015 17:09

Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt

first-grade participant

From: *gsi.de

Hello Aida.

The issue is, that with the *.c extension our CMake based compilation fails. When modifying the standard_FORTRAN.mk to read the *.cxx files, it will try to compile PndFtfDirect, which is not desirable as all settings from CMake are not available in the standalone version.

What I did now (please update): There is the executable <build>/bin/FTFGen in the standard CMake compilation.

Another solution will be to manually specify the relevant .cxx files in the standard_FORTRAN.mk. I'm not sure how to do that correctly.

Kind regards.
Ralf

Report message to a moderator

Re: FTF generator [message #17960 is a reply to message #17959]

Wed, 04 March 2015 17:52

Aida Galoyan
Messages: 79
Registered: May 2007
Location: Dubna

continuous participant

From: *ikp.kfa-juelich.de

Hello Ralf,

CMake based compilation can work with file.cc .
So, I assume, if you replace PndFtfDirect.cxx on PndFtfDirect.cc
CMAke will compile.

In this case, I will perform corresponding changes (*.c on *.cxx) in *.mk files,
to work FTFmake.

I think, it is solution for work of both: standalone FTFmake and FTFdirect.

Aida

Report message to a moderator

Re: FTF generator [message #17961 is a reply to message #17960]

Wed, 04 March 2015 19:47

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: *ip71.fastwebnet.it

The last change break the build, please fix it.
In any case PndFtfDirect should have the priority over the compiled version. If the bin does not compile with .cxx it means that the makefile needs some changed, most probably by substituting .c with .cxx.

Report message to a moderator

Re: FTF generator [message #17962 is a reply to message #17961]

Thu, 05 March 2015 11:09

Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt

first-grade participant

From: *gsi.de

StefanoSpataro wrote on Wed, 04 March 2015 19:47

The last change break the build, please fix it.

It's OK for me - both old and new fairsoft. Freshly checked out.

StefanoSpataro wrote on Wed, 04 March 2015 19:47

If the bin does not compile with .cxx it means that the makefile needs some changed, most probably by substituting .c with .cxx.

Exactly, however there is the issue that it then tries to compile also FtfDirect (also a.cxx). I'll play around with positioning FtfDirect.

Ralf

Report message to a moderator

Re: FTF generator [message #17963 is a reply to message #17961]

Thu, 05 March 2015 11:26

Aida Galoyan
Messages: 79
Registered: May 2007
Location: Dubna

continuous participant

From: *ikp.kfa-juelich.de

Hello Ralf and Stefano,

I don't understand, why PndFtfDirect.cxx
is directly implemented in /FtfEvtGen directory.
I created this directory for standalone FTF calculations.
For example, earlier, PndDpmDirect.cxx is implemented in /pgenerators directory.

How I understand, when we run FTFmake,
PndFtfDirect.cxx is conflicted with other files in the directory,
for example, with main.cc.

I think, the solution will be folowing.
If you move PndFtfDirect.cxx to /pgenerators directory,
similar PndDpmDirect.cxx, I will perform corresponding changes:
to remove *.c on *.cxx in needed makefiles.

If you have better solution, I will be glad.
In any case, it is needed to have normal working FTFmake.

Aida

Report message to a moderator

Re: FTF generator [message #17964 is a reply to message #17963]

Thu, 05 March 2015 12:26

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: 91.252.24*

This could make sense, and since it worked for dpm it should work also for ftf.

Report message to a moderator

Re: FTF generator [message #17965 is a reply to message #17958]

Thu, 05 March 2015 13:19

Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt

first-grade participant

From: *gsi.de

Hello.

The binary (FTFGen) is automatically compiled by Pandaroot and is installed in your build/bin directory, next to DPMGen.

I am able to put FtfDirect into the PGen library (now that I sorted out some lib issues).

However, I have trouble to compile with the standalone makefile. The Problem is that the include files are not found. Which environment setup should one have (e.g. Geant4?) to compile Ftf?

Lastly: Is it really necessary to have the standalone compilation? I would agree for a package _outside_ of our svn repository.

Cheers!
Ralf

Report message to a moderator

Re: FTF generator [message #17966 is a reply to message #17958]

Thu, 05 March 2015 14:32

Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt

first-grade participant

From: *gsi.de

Hello Aida,

it should be OK for you now.

Ralf

Report message to a moderator

Re: FTF generator [message #17967 is a reply to message #17966]

Thu, 05 March 2015 16:52

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: 37.227.2*

From the dashboard the build is still crashing.

Report message to a moderator

Re: FTF generator [message #17972 is a reply to message #17967]

Fri, 06 March 2015 15:05

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: 124.16.101*

Now maybe it is fixed in Linux, but in Mac I can see the following:

Linking CXX executable ../../bin/FTFGen
Undefined symbols for architecture x86_64:
  "G4Evaporation::SetGEMChannel()", referenced from:
      _main in main.cc.o
  "G4Evaporation::SetDefaultChannel()", referenced from:
      _main in main.cc.o
  "G4Evaporation::SetCombinedChannel()", referenced from:
      _main in main.cc.o
  "G4Evaporation::G4Evaporation()", referenced from:
      _main in main.cc.o
  "G4ProductionCuts::G4ProductionCuts()", referenced from:
      _main in main.cc.o
  "G4BGGPionElasticXS::G4BGGPionElasticXS(G4ParticleDefinition const*)", referenced from:
      _main in main.cc.o
  "G4ChipsComponentXS::G4ChipsComponentXS()", referenced from:
      _main in main.cc.o
  "G4PreCompoundModel::UseNGB()", referenced from:
      _main in main.cc.o
  "G4PreCompoundModel::UseSCO()", referenced from:
      _main in main.cc.o
  "G4PreCompoundModel::UseSICB()", referenced from:
      _main in main.cc.o
  "G4PreCompoundModel::SetOPTxs(int)", referenced from:
      _main in main.cc.o
  "G4PreCompoundModel::UseCEMtr()", referenced from:
      _main in main.cc.o
  "G4PreCompoundModel::G4PreCompoundModel(G4ExcitationHandler*)", referenced from:
      _main in main.cc.o
  "G4ExcitationHandler::SetParameters()", referenced from:
      G4ExcitationHandler::SetOPTxs(int) in main.cc.o
      G4ExcitationHandler::UseSICB() in main.cc.o
  "G4ExcitationHandler::SetEvaporation(G4VEvaporation*)", referenced from:
      _main in main.cc.o
  "G4ExcitationHandler::SetMinEForMultiFrag(double)", referenced from:
      _main in main.cc.o
  "G4ExcitationHandler::SetMaxAandZForFermiBreakUp(int, int)", referenced from:
      _main in main.cc.o
  "G4ExcitationHandler::G4ExcitationHandler()", referenced from:
      _main in main.cc.o
  "G4BGGPionInelasticXS::G4BGGPionInelasticXS(G4ParticleDefinition const*)", referenced from:
      _main in main.cc.o
  "G4MaterialCutsCouple::G4MaterialCutsCouple(G4Material const*, G4ProductionCuts*)", referenced from:
      _main in main.cc.o
  "G4BGGNucleonElasticXS::G4BGGNucleonElasticXS(G4ParticleDefinition const*)", referenced from:
      _main in main.cc.o
  "G4HadronCrossSections::GetInelasticCrossSection(G4DynamicParticle const*, int, int)", referenced from:
      _main in main.cc.o
  "G4HadronCrossSections::Instance()", referenced from:
      _main in main.cc.o
  "G4ProductionCutsTable::GetProductionCutsTable()", referenced from:
      _main in main.cc.o
  "G4HadronElasticDataSet::G4HadronElasticDataSet(G4String const&)", referenced from:
      _main in main.cc.o
  "G4IonsShenCrossSection::G4IonsShenCrossSection()", referenced from:
      _main in main.cc.o
  "G4TripathiCrossSection::G4TripathiCrossSection()", referenced from:
      _main in main.cc.o
  "G4VCrossSectionDataSet::ComputeCrossSection(G4DynamicParticle const*, G4Element const*, G4Material const*)", referenced from:
      G4VCrossSectionDataSet::GetCrossSection(G4DynamicParticle const*, G4Element const*, G4Material const*) in main.cc.o
  "G4BGGNucleonInelasticXS::G4BGGNucleonInelasticXS(G4ParticleDefinition const*)", referenced from:
      _main in main.cc.o
  "G4PiNuclearCrossSection::G4PiNuclearCrossSection()", referenced from:
      _main in main.cc.o
  "G4HadronInelasticDataSet::G4HadronInelasticDataSet(G4String const&)", referenced from:
      _main in main.cc.o
  "G4TripathiLightCrossSection::G4TripathiLightCrossSection()", referenced from:
      _main in main.cc.o
  "G4ComponentAntiNuclNuclearXS::G4ComponentAntiNuclNuclearXS()", referenced from:
      _main in main.cc.o
  "G4ProtonInelasticCrossSection::G4ProtonInelasticCrossSection()", referenced from:
      _main in main.cc.o
  "G4NeutronInelasticCrossSection::G4NeutronInelasticCrossSection()", referenced from:
      _main in main.cc.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [bin/FTFGen] Error 1
make[1]: *** [pgenerators/FtfEvtGen/CMakeFiles/FtfEvtGen_exe.dir/all] Error 2
make: *** [all] Error 2

There is still something missing, since before (27139) it was compiling.

[Updated on: Fri, 06 March 2015 15:07]

Report message to a moderator

Re: FTF generator [message #17987 is a reply to message #17958]

Mon, 09 March 2015 11:22

Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt

first-grade participant

From: *gsi.de

Hi Stefano,

I have no access to a mac with a running pandaroot. My first guess is missing lib dependencies, so I added some in Pgen. Maybe, stefano, you could provide me a sandbox account to look at the issue?

Cheers
Ralf

Report message to a moderator

Re: FTF generator [message #17990 is a reply to message #17972]

Mon, 09 March 2015 16:52

Aida Galoyan
Messages: 79
Registered: May 2007
Location: Dubna

continuous participant

From: *ikp.kfa-juelich.de

Hi Stefano,

In GSI Linux, FTF works in new version of PandaRoot.
Just now, I have installed and checked.

In Mac, to overcome errors
please, add in /FtfEvtGen/common.mk
in line
LDLIBS += $(PANDALIBS) -L$(GEANT4_LIBRARY_DIR) -lG4clhep -lG4global -lG4geometry -lG4physicslists $(GLIBS)

-lG4processes

before $(GLIBS)

Good luck
Aida

Report message to a moderator

Re: FTF generator [message #17991 is a reply to message #17987]

Tue, 10 March 2015 02:18

StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino

first-grade participant

From: 121.195.113*

Dear all,
a G4processes was missing in the CMakeLists.txt, in target_link_libraries for "Darwin". Now it is fixed.

[Updated on: Tue, 10 March 2015 02:27]

Report message to a moderator

Previous Topic:	Ftf direct crashes during initialisation
Next Topic:	EMFFGen as a pi + pi - generator

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Dec 06 23:51:01 CET 2024

Total time taken to generate the page: 0.00728 seconds