Source code for firedrake.slate.slac.tsfc_driver

import collections

from functools import partial

from firedrake.slate.slac.utils import RemoveRestrictions
from firedrake.tsfc_interface import compile_form as tsfc_compile

from tsfc.ufl_utils import extract_firedrake_constants

from ufl.algorithms.map_integrands import map_integrand_dags
from ufl import Form


ContextKernel = collections.namedtuple("ContextKernel",
                                       ["tensor",
                                        "coefficients",
                                        "constants",
                                        "original_integral_type",
                                        "tsfc_kernels"])
ContextKernel.__doc__ = """\
A bundled object containing TSFC subkernels corresponding to a
particular integral type.

:param tensor: The terminal Slate tensor corresponding to the
               list of TSFC assembly kernels.
:param coefficients: The local coefficients of the tensor contained
                     in the integrands (arguments for TSFC subkernels).
:param constants: The local constants of the tensor contained in the integrands.
:param original_integral_type: The unmodified measure type
                               of the form integrals.
:param tsfc_kernels: A list of local tensor assembly kernels
                     provided by TSFC."""


[docs] def compile_terminal_form(tensor, prefix, *, tsfc_parameters=None): """Compiles the TSFC form associated with a Slate :class:`~.Tensor` object. This function will return a :class:`ContextKernel` which stores information about the original tensor, integral types and the corresponding TSFC kernels. :arg tensor: A Slate `~.Tensor`. :arg prefix: An optional `string` indicating the prefix for the subkernel. :arg tsfc_parameters: An optional `dict` of parameters to provide TSFC. Returns: A `ContextKernel` containing all relevant information. """ assert tensor.terminal, ( "Only terminal tensors have forms associated with them!" ) # Sets a default name for the subkernel prefix. mapper = RemoveRestrictions() integrals = map(partial(map_integrand_dags, mapper), tensor.form.integrals()) transformed_integrals = transform_integrals(integrals) cxt_kernels = [] assert prefix is not None for orig_it_type, integrals in transformed_integrals.items(): subkernel_prefix = prefix + "%s_to_" % orig_it_type form = Form(integrals) kernels = tsfc_compile(form, subkernel_prefix, parameters=tsfc_parameters, split=False, diagonal=tensor.diagonal) if kernels: cxt_k = ContextKernel(tensor=tensor, coefficients=form.coefficients(), constants=extract_firedrake_constants(form), original_integral_type=orig_it_type, tsfc_kernels=kernels) cxt_kernels.append(cxt_k) cxt_kernels = tuple(cxt_kernels) return cxt_kernels
[docs] def transform_integrals(integrals): """Generates a mapping of the form: ``{original_integral_type: transformed_integrals}`` where the original_integral_type is the pre-transformed integral type. The transformed_integrals are an iterable of `ufl.Integral`s with the appropriately modified type. For example, an `interior_facet` integral will become an `exterior_facet` integral. """ transformed_integrals = collections.OrderedDict() for it in integrals: it_type = it.integral_type() if it_type == "cell" or it_type.startswith("exterior_facet"): # No need to reconstruct cell or exterior facet integrals transformed_integrals.setdefault(it_type, list()).append(it) elif it_type == "interior_facet": new_it = it.reconstruct(integral_type="exterior_facet") transformed_integrals.setdefault(it_type, list()).append(new_it) elif it_type == "interior_facet_vert": new_it = it.reconstruct(integral_type="exterior_facet_vert") transformed_integrals.setdefault(it_type, list()).append(new_it) elif it_type == "interior_facet_horiz": # Separate into "top" and "bottom" top_it = it.reconstruct(integral_type="exterior_facet_top") bottom_it = it.reconstruct(integral_type="exterior_facet_bottom") it_top = it_type + "_top" it_btm = it_type + "_bottom" transformed_integrals.setdefault(it_top, list()).append(top_it) transformed_integrals.setdefault(it_btm, list()).append(bottom_it) else: raise ValueError("Integral type: %s not recognized!" % it_type) return transformed_integrals