August 2014 - petsc-dev - mailman3.cels.anl.gov

DMCreateDomainDecomposition_DA error in next
by Barry Smith 29 Aug '14

29 Aug '14

--- > [0]PETSC ERROR: --------------------- Error Message -------------------------------------------------------------- > [0]PETSC ERROR: No support for this operation for this object type > [0]PETSC ERROR: Retturn of subinner and local ISes not yet supported > [0]PETSC ERROR: See http://www.mcs.anl.gov/petsc/documentation/faq.html for trouble shooting. > [0]PETSC ERROR: Petsc Development GIT revision: v3.5.1-1980-ga473ade GIT Date: 2014-08-28 14:40:22 -0500 > [0]PETSC ERROR: ./ex16 on a arch-freebsd-cxx-cmplx-pkgs-dbg named wii. by balay Fri Aug 29 06:57:33 2014 > [0]PETSC ERROR: Configure options --with-cc=gcc46 --with-cxx=g++46 --with-fc=gfortran LIBS=-lexecinfo --with-clanguage=cxx --with-scalar-type=complex --useThreads=0 --download-mpich=1 --download-cmake=1 --download-metis=1 --download-parmetis=1 --download-triangle=1 --download-scalapack=1 --download-mumps=1 --download-hdf5 --with-no-output -PETSC_ARCH=arch-freebsd-cxx-cmplx-pkgs-dbg -PETSC_DIR=/usr/home/balay/petsc.clone-3 > [0]PETSC ERROR: #1 DMCreateDomainDecomposition_DA() line 425 in /usr/home/balay/petsc.clone-3/src/dm/impls/da/dadd.c > [0]PETSC ERROR: #2 DMCreateDomainDecomposition() line 1357 in /usr/home/balay/petsc.clone-3/src/dm/interface/dm.c > [0]PETSC ERROR: #3 SNESSetUp_NASM() line 114 in /usr/home/balay/petsc.clone-3/src/snes/impls/nasm/nasm.c > [0]PETSC ERROR: #4 SNESSetUp() line 2616 in /usr/home/balay/petsc.clone-3/src/snes/interface/snes.c > [0]PETSC ERROR: #5 SNESSolve() line 3731 in /usr/home/balay/petsc.clone-3/src/snes/interface/snes.c > [0]PETSC ERROR: #6 SNESApplyNPC() line 48 in /usr/home/balay/petsc.clone-3/src/snes/interface/snespc.c > [0]PETSC ERROR: #7 SNESSolve_NEWTONLS() line 163 in /usr/home/balay/petsc.clone-3/src/snes/impls/ls/ls.c > [0]PETSC ERROR: #8 SNESSolve() line 3743 in /usr/home/balay/petsc.clone-3/src/snes/interface/snes.c > [0]PETSC ERROR: #9 main() line 175 in /home/balay/petsc.clone-3/src/snes/examples/tutorials/ex16.c > [0]PETSC ERROR: ----------------End of Error Message -------send entire error message to petsc-maint(a)mcs.anl.gov---------- > application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 > [cli_0]: aborting job: > application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 > > ==================================================================================

1 0

Re: [petsc-dev] --download-numpy should it still exist?
by Barry Smith 27 Aug '14

27 Aug '14

I will remove that option. If Apple has numpy then everyone has it :-( Barry On Aug 27, 2014, at 10:21 PM, Jed Brown <jed(a)jedbrown.org> wrote: > Barry Smith <bsmith(a)mcs.anl.gov> writes: > >> It doesn’t work for me. Should it still exist? Does the tar ball just need to be updated? > > petsc4py and mpi4py want numpy, but actual Python users should have it > installed already. If we are installing it, we need to travel 7 years > forward in time.

1 0

Re: [petsc-dev] --download-numpy should it still exist?
by Jed Brown 27 Aug '14

27 Aug '14

Barry Smith <bsmith(a)mcs.anl.gov> writes: > It doesn’t work for me. Should it still exist? Does the tar ball just need to be updated? petsc4py and mpi4py want numpy, but actual Python users should have it installed already. If we are installing it, we need to travel 7 years forward in time.

1 0

--download-numpy should it still exist?
by Barry Smith 27 Aug '14

27 Aug '14

It doesn’t work for me. Should it still exist? Does the tar ball just need to be updated? Error running setup.py on numpy: Could not execute "sudo mkdir -p /Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4 && cd /Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4 && sudo python setup.py install --prefix=/Users/joe": non-existing path in 'numpy/distutils': 'site.cfg' F2PY Version 2_4422Running from numpy source directory. Traceback (most recent call last): File "setup.py", line 89, in <module> setup_package() File "setup.py", line 82, in setup_package configuration=configuration ) File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/core.py", line 145, in setup config = configuration() File "setup.py", line 49, in configuration config.add_subpackage('numpy') File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/misc_util.py", line 834, in add_subpackage caller_level = 2) File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/misc_util.py", line 817, in get_subpackage caller_level = caller_level + 1) File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/misc_util.py", line 764, in _get_configuration_from_setup_py config = setup_module.configuration(*args) File "numpy/setup.py", line 9, in configuration config.add_subpackage('core') File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/misc_util.py", line 834, in add_subpackage caller_level = 2) File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/misc_util.py", line 817, in get_subpackage caller_level = caller_level + 1) File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/misc_util.py", line 764, in _get_configuration_from_setup_py config = setup_module.configuration(*args) File "numpy/core/setup.py", line 246, in configuration blas_info = get_info('blas_opt',0) File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/system_info.py", line 267, in get_info return cl().get_info(notfound_action) File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/system_info.py", line 366, in __init__ self.parse_config_files() File "/Users/barrysmith/Src/petsc/arch-basic/externalpackages/numpy-1.0.4/numpy/distutils/system_info.py", line 374, in parse_config_files self.cp.add_section(self.section) File "/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/ConfigParser.py", line 261, in add_section raise ValueError, 'Invalid section name: %s' % section ValueError: Invalid section name: DEFAULT

1 0

Re: [petsc-dev] what the hey is FFC
by Jed Brown 27 Aug '14

27 Aug '14

Barry Smith <bsmith(a)mcs.anl.gov> writes: > =============================================================================== Trying to download http://www.fenics.org/pub/software/ffc/v0.3/ffc-0.3.3.tar.gz for FFC =============================================================================== ******************************************************************************* > UNABLE to CONFIGURE with GIVEN OPTIONS (see configure.log for details): > ------------------------------------------------------------------------------- > file could not be opened successfully > Downloaded package FFC from: http://www.fenics.org/pub/software/ffc/v0.3/ffc-0.3.3.tar.gz is not a tarball. > [or installed python cannot process compressed files] > > fenics.org doesn’t seem around any more. It was renamed to fenicsproject.org many years ago. > And what is the business of having a tar ball but needing no install? > Surely it must at least need to be put some where? It is a Python code-generation package. With Sieve gone, I'm pretty sure it can be deleted. > Can FFC.py be removed? > > > class Configure(PETSc.package.NewPackage): > def __init__(self, framework): > PETSc.package.NewPackage.__init__(self, framework) > self.download = ['http://www.fenics.org/pub/software/ffc/v0.3/ffc-0.3.3.tar.gz'] > return > > def setupDependencies(self, framework): > PETSc.package.NewPackage.setupDependencies(self, framework) > self.fiat = self.framework.require('config.packages.fiat', self) > self.deps = [self.fiat] > return > > def Install(self): > import sys > ffcDir = self.getDir() > # We could make a check of the md5 of the current configure framework > self.logPrintBox('FFC needs no installation') > self.framework.actions.addArgument('FFC', 'Install', 'Installed FFC into '+ffcDir) > return ffcDir

1 0

what the hey is FFC
by Barry Smith 27 Aug '14

27 Aug '14

=============================================================================== Trying to download http://www.fenics.org/pub/software/ffc/v0.3/ffc-0.3.3.tar.gz for FFC =============================================================================== ******************************************************************************* UNABLE to CONFIGURE with GIVEN OPTIONS (see configure.log for details): ------------------------------------------------------------------------------- file could not be opened successfully Downloaded package FFC from: http://www.fenics.org/pub/software/ffc/v0.3/ffc-0.3.3.tar.gz is not a tarball. [or installed python cannot process compressed files] fenics.org doesn’t seem around any more. And what is the business of having a tar ball but needing no install? Surely it must at least need to be put some where? Can FFC.py be removed? class Configure(PETSc.package.NewPackage): def __init__(self, framework): PETSc.package.NewPackage.__init__(self, framework) self.download = ['http://www.fenics.org/pub/software/ffc/v0.3/ffc-0.3.3.tar.gz'] return def setupDependencies(self, framework): PETSc.package.NewPackage.setupDependencies(self, framework) self.fiat = self.framework.require('config.packages.fiat', self) self.deps = [self.fiat] return def Install(self): import sys ffcDir = self.getDir() # We could make a check of the md5 of the current configure framework self.logPrintBox('FFC needs no installation') self.framework.actions.addArgument('FFC', 'Install', 'Installed FFC into '+ffcDir) return ffcDir

1 0

Re: [petsc-dev] please don't merge broken examples into master!
by Satish Balay 25 Aug '14

25 Aug '14

Previously this was listed in TESTEXAMPLES_C [so it was run] but Barry fixed it [so now is listed in TESTEXAMPLES_C_NOCOMPLEX]. https://bitbucket.org/petsc/petsc/commits/bca1009e69057d2aef5bf0ca910e7b330… Satish On Sun, 24 Aug 2014, hong(a)aspiritech.org wrote: > petsc/src/ksp/ksp/examples/tests/ex8.c is a real-only example. I put it under > > TESTEXAMPLES_C_NOCOMPLEX = ex8.PETSc runex8 runex8_2 ex8.rm > ... > > [0]PETSC ERROR: Spectrum slicing with MUMPS is not available for complex scalars > > Why it is executed in a complex build? > > Hong > > > > On Sat, Aug 23, 2014 at 5:30 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > > > http://ftp.mcs.anl.gov/pub/petsc/nightlylogs/archive/2014/08/22/examples_ma… > > > > >

1 0

Re: [petsc-dev] please don't merge broken examples into master!
by hong＠aspiritech.org 24 Aug '14

24 Aug '14

petsc/src/ksp/ksp/examples/tests/ex8.c is a real-only example. I put it under TESTEXAMPLES_C_NOCOMPLEX = ex8.PETSc runex8 runex8_2 ex8.rm ... > [0]PETSC ERROR: Spectrum slicing with MUMPS is not available for complex scalars Why it is executed in a complex build? Hong On Sat, Aug 23, 2014 at 5:30 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > > http://ftp.mcs.anl.gov/pub/petsc/nightlylogs/archive/2014/08/22/examples_ma… > >

1 0

please don't merge broken examples into master!
by Barry Smith 23 Aug '14

23 Aug '14

http://ftp.mcs.anl.gov/pub/petsc/nightlylogs/archive/2014/08/22/examples_ma…

1 0

Re: [petsc-dev] [petsc-users] gamg failure with petsc-dev
by Mark Adams 22 Aug '14

22 Aug '14

On Thu, Jul 31, 2014 at 10:44 AM, Stephan Kramer <s.kramer(a)imperial.ac.uk> wrote: > An update on this. We have been using mark/gamg-zerod branch and it fixes > for us the issue with zero diagonals in the coarsened operators (making the > sor smoother, but bizarrely not the jacobi smoother fail). We in fact have > some cases where cheby+jacobi does not converge (indefinite pc), but > cheby+sor (with the mark/gamg-zerod branch) works well, so we'd be very > much interested in getting this (or something similar) merged in master. Jed: this has been a pull request for 4+ months (#162).

1 0