ffc team mailing list archive

[ffc@xxxxxxxxxx (FFC)]

One or more new changesets pushed to the primary FFC repository.
A short summary of the last three changesets is included below.

changeset:   752:8d168402b9a1faa8e0efa736cbc4bfacf281471a
tag:         tip
user:        "Anders Logg <logg@xxxxxxxxx>"
date:        Mon May 07 10:39:35 2007 +0200
files:       TODO src/bin/ffc src/ffc/common/constants.py src/ffc/compiler/codegeneration/common/evaluatebasis.py src/ffc/compiler/codegeneration/common/utils.py src/ffc/compiler/codegeneration/tensor/tensorgenerator.py src/ffc/compiler/compiler.py src/ffc/compiler/format/ufcformat.py src/test/regression/reference/Elasticity.h src/test/regression/reference/EnergyNorm.h src/test/regression/reference/Heat.h src/test/regression/reference/Mass.h src/test/regression/reference/MixedMixedElement.h src/test/regression/reference/MixedPoisson.h src/test/regression/reference/NavierStokes.h src/test/regression/reference/NeumannProblem.h src/test/regression/reference/Optimization.h src/test/regression/reference/P5tet.h src/test/regression/reference/P5tri.h src/test/regression/reference/Poisson.h src/test/regression/reference/PoissonDG.h src/test/regression/reference/PoissonSystem.h src/test/regression/reference/Projection.h src/test/regression/reference/Stokes.h src/test/regression/reference/TensorWeightedPoisson.h src/test/regression/test.py
description:
Add compiler flag -fprecision=n where n is the number of significant
digits that should be used in the output. FFC_EPSILON has been removed
and replaced by an entry in the format dictionary (and is set to 10*1e-n).

Regression test is now run with -fprecision=10 (which should make sure
it runs on various platforms/Python versions etc with identical results).

Default precision is now back to n=15.


changeset:   751:1d621f25c8235a27e986ae883bd9d3cda18e7081
parent:      748:d89ee324481c80ff64151f79044ae9127a20aeb0
parent:      750:0fce41e14c77907fcb0a7222dd2434c006e8acd0
user:        "Anders Logg <logg@xxxxxxxxx>"
date:        Mon May 07 00:52:13 2007 +0200
files:       src/ffc/compiler/codegeneration/quadrature/quadraturegenerator.py src/ffc/compiler/format/ufcformat.py
description:
merge


changeset:   750:0fce41e14c77907fcb0a7222dd2434c006e8acd0
user:        "Kristian Oelgaard <k.b.oelgaard@xxxxxxxxxx>"
date:        Wed May 02 14:28:30 2007 +0200
files:       src/ffc/compiler/codegeneration/common/dofmap.py src/ffc/compiler/codegeneration/quadrature/quadraturegenerator.py src/ffc/compiler/format/codesnippets.py src/ffc/compiler/format/removeunused.py src/ffc/compiler/format/ufcformat.py src/ffc/compiler/representation/quadrature/elementtensor.py src/ffc/compiler/representation/quadrature/monomialtabulation.py
description:
Changed removedunused.py so that 'double' and 'const double' array declarations are not removed.


----------------------------------------------------------------------
For more details, visit http://www.fenics.org/hg/ffc