[petsc-dev] Module missing error with petsc-dev.

[Vijay S. Mahadevan]

Ah I see. I just figured out that "gamgf.c" is the fortran interface
for the gamg preconditioner because I run it correctly. Maybe the
fortran stubs for the source might have to be updated.

On Tue, Sep 27, 2011 at 8:56 PM, Vijay S. Mahadevan <vijay.m at gmail.com> wrote:
> The configure went through correctly but I had errors during make.
>
> ERROR IN Cxx COMPILE ******************************
>      gamgf.c: In function ‘void pcgamgsetproceqlim_(PC, PetscInt*, int*)’:
> gamgf.c:54:30: error: ‘PCGAMGSetProcEqLim’ was not declared in this scope
> gamgf.c: In function ‘void pcgamgavoidrepartitioning_(PC, PetscBool*, int*)’:
> gamgf.c:58:30: error: ‘PCGAMGAvoidRepartitioning’ was not declared in this scope
> gamgf.c: In function ‘void pcgamgsetsolvertype_(PC, char*, PetscInt*, int*)’:
> gamgf.c:62:35: error: ‘PCGAMGSetSolverType’ was not declared in this scope
>
>        Popping language Cxx
>        Moving gamgf.o to
> /Users/mahadevan/source/petsc-dev/arch-darwin-cxx-debug/lib/libpetsc-obj/gamgf.o
>
> Strangely enough, it produced the dynamic library that I can link
> against correctly. I found that libpetsc-obj directory did not have an
> object file "gamgf" and so nothing got moved. I assume there will be a
> runtime error if I try to use gamg since the PC type has not been
> registered ?
>
> If you want, I can send you the make.log but the above errors are the
> only offending ones I found. Also, should it not quit after a compiler
> error or is this the correct behavior ? It just seems a little
> misleading.
>
> Vijay
>
> On Tue, Sep 27, 2011 at 8:30 PM, Vijay S. Mahadevan <vijay.m at gmail.com> wrote:
>> Satish, I just pulled again after looking at your email and it updated
>> a few changes. Now everything goes through correctly. Thanks,
>>
>> Vijay
>>
>> On Tue, Sep 27, 2011 at 8:25 PM, Satish Balay <balay at mcs.anl.gov> wrote:
>>> I don't see this error. My cofigure went thorugh fine [on linux]
>>>
>>> Satish
>>>
>>> ----------
>>> Configure Options: --configModules=PETSc.Configure --optionsModule=PETSc.compilerOptions --with-mpi-dir=/home/balay/soft/linux64/mpich2-1.1 --with-shared-libraries=1 --with-clanguage=cxx PETSC_ARCH=asterix64-cxx-mumps --download-parmetis=1 --download-blacs=1 --download-scalapack=1 --download-mumps=1 --download-hypre=1 --with-valgrind=1
>>>
>>>
>>> On Tue, 27 Sep 2011, Vijay S. Mahadevan wrote:
>>>
>>>> I just pulled the dev version (source+Buildsystem) and the reconfigure
>>>> yields errors regarding FIAT module. I do not even have the module as
>>>> part of the configure options. Here's the command.
>>>>
>>>> ./configure --download-blacs=1 --download-hypre=1 --download-mumps=1
>>>> --download-parmetis=1 --download-scalapack=1
>>>> --known-mpi-shared-libraries=1
>>>> --with-blas-lapack-lib=/System/Library/Frameworks/vecLib.framework/vecLib
>>>> --with-cc=mpicc --with-clanguage=C++ --with-fc=mpif90 --with-mpi=1
>>>> --with-shared-libraries=1 --with-valgrind=1 -with-cxx=mpicxx
>>>> COPTFLAGS="-O3 -ftree-vectorize" CXXOPTFLAGS="-O3 -ftree-vectorize"
>>>> FOPTFLAGS="-O3 -ftree-vectorize" PETSC_ARCH=arch-darwin-cxx-debug
>>>> ===============================================================================
>>>>              Configuring PETSc to compile on your system
>>>> ===============================================================================
>>>> *******************************************************************************
>>>>                      UNABLE to FIND MODULE for ./configure
>>>> -------------------------------------------------------------------------------
>>>> No module named Fiat
>>>> *******************************************************************************
>>>>
>>>>
>>>>   File "./config/configure.py", line 269, in petsc_configure
>>>>     framework =
>>>> config.framework.Framework(['--configModules=PETSc.Configure','--optionsModule=PETSc.compilerOptions']+sys.argv[1:],
>>>> loadArgDB = 0)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/BuildSystem/config/framework.py",
>>>> line 110, in __init__
>>>>     self.createChildren()
>>>>   File "/Users/mahadevan/source/petsc-dev/config/BuildSystem/config/framework.py",
>>>> line 316, in createChildren
>>>>     self.getChild(moduleName)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/BuildSystem/config/framework.py",
>>>> line 301, in getChild
>>>>     config.setupDependencies(self)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/PETSc/Configure.py",
>>>> line 61, in setupDependencies
>>>>     utilityObj                    =
>>>> self.framework.require('PETSc.'+d+'.'+utilityName, self)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/BuildSystem/config/framework.py",
>>>> line 321, in require
>>>>     config = self.getChild(moduleName, keywordArgs)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/BuildSystem/config/framework.py",
>>>> line 301, in getChild
>>>>     config.setupDependencies(self)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/PETSc/packages/FFC.py",
>>>> line 11, in setupDependencies
>>>>     self.fiat = self.framework.require('config.packages.Fiat', self)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/BuildSystem/config/framework.py",
>>>> line 321, in require
>>>>     config = self.getChild(moduleName, keywordArgs)
>>>>   File "/Users/mahadevan/source/petsc-dev/config/BuildSystem/config/framework.py",
>>>> line 274, in getChild
>>>>     type   = __import__(moduleName, globals(), locals(),
>>>> ['Configure']).Configure
>>>>
>>>> The log file does not have anything other than push-pop languages. I
>>>> didn't see any issues when I pulled the source. Ideas ?
>>>>
>>>> Vijay
>>>>
>>>
>>>
>>
>