Replaced all uses of complex Xfer class with C++11 "move" constructors and
assignment operators. Removed the now redundant Xfer class.
This substantial changes improves consistency between OpenFOAM and the C++11 STL
containers and algorithms, reduces memory allocation and copy overhead when
returning containers from functions and simplifies maintenance of the core
libraries significantly.
Using the new field mapper framework it is now possible to create specialised
mappers rather than creating a fatter and fatter interface in the base mapper.
This approach is far more extensible, comprehensible and maintainable.
The writeEntry form is now defined and used consistently throughout OpenFOAM
making it easier to use and extend, particularly to support binary IO of complex
dictionary entries.
The dynamic code functionality has been generalised so that the names of
the code entries in the specifying dictionary can be set by the caller.
This means that functions which utilise dynamic code but use different
entry names (e.g., codedFunctionObject uses codeExecute, codeEnd,
etc..., instead of code) now function correctly. The differently named
entries now form part of the library hash, and re-building triggers
appropriately as they are modified.
The construction of some patch fields has been corrected so that the
patchType setting always propagates on mapping, IO, clone, etc...
Dictionary and mapping-based patch field constructors now call the
corresponding constructor from the base class, regardless of whether
dictionary settings or mapping are actually needed.
A "mappingRequired" flag has been added to some of the base constructors
in order to prevent unecessary mapping of field data and retain the
previous level of optimisation.
Resolves bug report https://bugs.openfoam.org/view.php?id=3144
Registration occurs when the temporary field is transferred to a non-temporary
field via a constructor or if explicitly transferred to the database via the
regIOobject "store" methods.
Now for transient simulations "Final" solver settings are required for ALL
equations providing consistency between the solution of velocity, energy,
composition and radiation properties.
However "Final" relaxation factors are no longer required for fields or
equations and if not present the standard value for the variable will be
applied. Given that relaxation factors other than 1 are rarely required for
transient runs and hence the same for all iterations including the final one
this approach provide simpler input while still providing the flexibility to
specify a different value for the final iteration if required. For steady cases
it is usual to execute just 1 outer iteration per time-step for which the
standard relaxation factors are appropriate, and if more than one iteration is
executed it is common to use the same factors for both. In the unlikely event
of requiring different relaxation factors for the final iteration this is still
possible to specify via the now optional "Final" specification.
This was not being used and had not been kept up to date. It also didn't
actually do dimension checking, so the name was misleading, and the
functionality added on top of SubField was minimal.
Zeroing a dimensioned field can now be achieved by assignment to the
zero type. This prevents the clutter associated with constructing an
appropriate dimensioned type, or having to use multiply-equals-zero as a
workaround.
MULES and CMULES have been extended so that the limits can be supplied
as fields. These arguments are templated so that zeroField, oneField or
UniformField<scalar> can be used in place of a scalar value with no
additional overhead. The flux argument has been removed from the
unlimited CMULES correct functions in order to make this templating
possible.
An additional form of limit sum has also been added to MULES. This
limits the flux sum by ofsetting in proportion to the phase fraction,
rather than by reducing the magnitude of the fluxes with the same sign
as the imbalance. The new procedure makes it possible to limit the flux
sum in the presence of constraints without encountering a divide by
zero.
In early versions of OpenFOAM the scalar limits were simple macro replacements and the
names were capitalized to indicate this. The scalar limits are now static
constants which is a huge improvement on the use of macros and for consistency
the names have been changed to camel-case to indicate this and improve
readability of the code:
GREAT -> great
ROOTGREAT -> rootGreat
VGREAT -> vGreat
ROOTVGREAT -> rootVGreat
SMALL -> small
ROOTSMALL -> rootSmall
VSMALL -> vSmall
ROOTVSMALL -> rootVSmall
The original capitalized are still currently supported but their use is
deprecated.
When an OpenFOAM simulation runs in parallel, the data for decomposed fields and
mesh(es) has historically been stored in multiple files within separate
directories for each processor. Processor directories are named 'processorN',
where N is the processor number.
This commit introduces an alternative "collated" file format where the data for
each decomposed field (and mesh) is collated into a single file, which is
written and read on the master processor. The files are stored in a single
directory named 'processors'.
The new format produces significantly fewer files - one per field, instead of N
per field. For large parallel cases, this avoids the restriction on the number
of open files imposed by the operating system limits.
The file writing can be threaded allowing the simulation to continue running
while the data is being written to file. NFS (Network File System) is not
needed when using the the collated format and additionally, there is an option
to run without NFS with the original uncollated approach, known as
"masterUncollated".
The controls for the file handling are in the OptimisationSwitches of
etc/controlDict:
OptimisationSwitches
{
...
//- Parallel IO file handler
// uncollated (default), collated or masterUncollated
fileHandler uncollated;
//- collated: thread buffer size for queued file writes.
// If set to 0 or not sufficient for the file size threading is not used.
// Default: 2e9
maxThreadFileBufferSize 2e9;
//- masterUncollated: non-blocking buffer size.
// If the file exceeds this buffer size scheduled transfer is used.
// Default: 2e9
maxMasterFileBufferSize 2e9;
}
When using the collated file handling, memory is allocated for the data in the
thread. maxThreadFileBufferSize sets the maximum size of memory in bytes that
is allocated. If the data exceeds this size, the write does not use threading.
When using the masterUncollated file handling, non-blocking MPI communication
requires a sufficiently large memory buffer on the master node.
maxMasterFileBufferSize sets the maximum size in bytes of the buffer. If the
data exceeds this size, the system uses scheduled communication.
The installation defaults for the fileHandler choice, maxThreadFileBufferSize
and maxMasterFileBufferSize (set in etc/controlDict) can be over-ridden within
the case controlDict file, like other parameters. Additionally the fileHandler
can be set by:
- the "-fileHandler" command line argument;
- a FOAM_FILEHANDLER environment variable.
A foamFormatConvert utility allows users to convert files between the collated
and uncollated formats, e.g.
mpirun -np 2 foamFormatConvert -parallel -fileHandler uncollated
An example case demonstrating the file handling methods is provided in:
$FOAM_TUTORIALS/IO/fileHandling
The work was undertaken by Mattijs Janssens, in collaboration with Henry Weller.
for consistency with reactingTwoPhaseEulerFoam and to ensure correct operation
of models requiring formal boundedness of phase-fractions.
Resolves bug-report https://bugs.openfoam.org/view.php?id=2589
"pos" now returns 1 if the argument is greater than 0, otherwise it returns 0.
This is consistent with the common mathematical definition of the "pos" function:
https://en.wikipedia.org/wiki/Sign_(mathematics)
However the previous implementation in which 1 was also returned for a 0
argument is useful in many situations so the "pos0" has been added which returns
1 if the argument is greater or equal to 0. Additionally the "neg0" has been
added which returns 1 if if the argument is less than or equal to 0.
vectorField or vector2DField from scalarField components. To do this
properly and have it work for field-type combinations would require some
new field function macros.
Now the interFoam and compressibleInterFoam families of solvers use the same
alphaEqn formulation and supporting all of the MULES options without
code-duplication.
The semi-implicit MULES support allows running with significantly larger
time-steps but this does reduce the interface sharpness.
This supports the abstraction of the set of fields from the field code
generation macros making it easier to change the set of fields supported
by OpenFOAM. This functionality is demonstrated in the updated
fvPatchFields macros and will be applied to the rest of the field code
generation macros in the future.
