From 8881feb64a94669512c0e641ca2745202913f0f9 Mon Sep 17 00:00:00 2001
From: Dominic Kempf <dominic.kempf@iwr.uni-heidelberg.de>
Date: Fri, 31 Mar 2017 14:02:24 +0200
Subject: [PATCH] Cleanup stage 1 calling scope
---
python/dune/perftool/sumfact/basis.py | 99 ++++-----------------
python/dune/perftool/sumfact/realization.py | 7 +-
python/dune/perftool/sumfact/sumfact.py | 4 +-
3 files changed, 25 insertions(+), 85 deletions(-)
diff --git a/python/dune/perftool/sumfact/basis.py b/python/dune/perftool/sumfact/basis.py
index 5ed69496..892cfd12 100644
--- a/python/dune/perftool/sumfact/basis.py
+++ b/python/dune/perftool/sumfact/basis.py
@@ -6,6 +6,7 @@ multiplication with the test function is part of the sum factorization kernel.
from dune.perftool.generation import (backend,
domain,
+ get_backend,
get_counted_variable,
get_counter,
get_global_context_value,
@@ -65,15 +66,19 @@ def pymbolic_coefficient_gradient(element, restriction, component, coeff_func, v
shape_impl = ('arr',) * rank
temporary_variable(name, shape=shape, shape_impl=shape_impl)
- dim = world_dimension()
+ # Whether direct indexing into the output is possible. This happens
+ # if the positioning within a SIMD vectors coincides with the index!
+ direct_indexing_is_possible = True
+
buffers = []
- for i in range(dim):
+ for i in range(world_dimension()):
# Construct the matrix sequence for this sum factorization
a_matrices = construct_amatrix_sequence(derivative=i,
facedir=get_facedir(restriction),
facemod=get_facemod(restriction),
)
+ # The sum factorization kernel object gathering all relevant information
sf = SumfactKernel(a_matrices=a_matrices,
restriction=restriction,
preferred_position=i,
@@ -85,43 +90,12 @@ def pymbolic_coefficient_gradient(element, restriction, component, coeff_func, v
from dune.perftool.sumfact.vectorization import attach_vectorization_info
sf = attach_vectorization_info(sf)
- # Extract again, for compatibility
- # TODO away!
- a_matrices = sf.a_matrices
- buf = sf.buffer
- inp = sf.input
- index = sf.index
- padding = sf.padding
-
-# if buf is None:
-# buf = get_counted_variable("buffer")
-# if inp is None:
-# inp = get_counted_variable("input")
-#
-# # Initialize the buffer for the sum fact kernel
-# shape = (product(mat.cols for mat in a_matrices),)
-# if index is not None:
-# shape = shape + (4,)
-# inp = initialize_buffer(buf).get_temporary(shape=shape,
-# name=inp,
-# )
-# insn_dep = frozenset({Writes(inp)})
-
- if get_option('fastdg'):
- # Name of direct input, shape and globalarg is set in sum_factorization_kernel
- direct_input = coeff_func(restriction)
- else:
- direct_input = None
- # Setup the input!
- #setup_theta(inp, element, restriction, component, index, coeff_func)
-
# Add a sum factorization kernel that implements the
# evaluation of the gradients of basis functions at quadrature
# points (stage 1)
from dune.perftool.sumfact.realization import realize_sum_factorization_kernel
var, insn_dep = realize_sum_factorization_kernel(sf,
- outshape=tuple(mat.rows for mat in a_matrices if mat.face is None),
- direct_input=direct_input,
+ outshape=tuple(mat.rows for mat in sf.a_matrices if mat.face is None),
)
buffers.append(var)
@@ -129,17 +103,15 @@ def pymbolic_coefficient_gradient(element, restriction, component, coeff_func, v
# Check whether we want to return early with something that has the indexing
# already handled! This happens with vectorization when the index coincides
# with the position in the vector register.
- if index:
+ if direct_indexing_is_possible:
assert len(visitor.indices) == 1
return maybe_wrap_subscript(var, tuple(prim.Variable(i) for i in quadrature_inames()) + visitor.indices), None
# TODO this should be quite conditional!!!
for i, buf in enumerate(buffers):
# Write solution from sumfactorization to gradient variable
- from pymbolic.primitives import Subscript, Variable
- from dune.perftool.generation import get_backend
- assignee = Subscript(Variable(name), i)
- expression = Subscript(buf, tuple(Variable(i) for i in quadrature_inames()))
+ assignee = prim.Subscript(prim.Variable(name), i)
+ expression = prim.Subscript(buf, tuple(prim.Variable(i) for i in quadrature_inames()))
instruction(assignee=assignee,
expression=expression,
forced_iname_deps=frozenset(get_backend("quad_inames")()),
@@ -151,9 +123,6 @@ def pymbolic_coefficient_gradient(element, restriction, component, coeff_func, v
@kernel_cached
def pymbolic_coefficient(element, restriction, component, coeff_func, visitor):
- # Get geometric dimension
- dim = world_dimension()
-
# Construct the matrix sequence for this sum factorization
a_matrices = construct_amatrix_sequence(facedir=get_facedir(restriction),
facemod=get_facemod(restriction),)
@@ -169,49 +138,15 @@ def pymbolic_coefficient(element, restriction, component, coeff_func, visitor):
from dune.perftool.sumfact.vectorization import attach_vectorization_info
sf = attach_vectorization_info(sf)
- # Extract again, for compatibility
- # TODO away!
- a_matrices = sf.a_matrices
- buf = sf.buffer
- inp = sf.input
- index = sf.index
- padding = sf.padding
-#
-# if buf is None:
-# buf = get_counted_variable("buffer")
-# if inp is None:
-# inp = get_counted_variable("input")
-#
-# # Flip flop buffers for sumfactorization
-# shape = (product(mat.cols for mat in a_matrices),)
-# if index is not None:
-# shape = shape + (4,)
-# initialize_buffer(buf).get_temporary(shape=shape,
-# name=inp,
-# )
-
- if get_option('fastdg'):
- # Name of direct input, shape and globalarg is set in sum_factorization_kernel
- direct_input = coeff_func(restriction)
- else:
- direct_input = None
- # Setup the input!
- # setup_theta(inp, element, restriction, component, index, coeff_func)
-
# Add a sum factorization kernel that implements the evaluation of
# the basis functions at quadrature points (stage 1)
- if not get_global_context_value("dry_run", False):
- from dune.perftool.sumfact.realization import realize_sum_factorization_kernel
- var, _ = realize_sum_factorization_kernel(sf,
-# insn_dep=frozenset({Writes(inp)}),
- outshape=tuple(mat.rows for mat in a_matrices if mat.face is None),
- direct_input=direct_input,
- )
- else:
- var = sf
+ from dune.perftool.sumfact.realization import realize_sum_factorization_kernel
+ var, _ = realize_sum_factorization_kernel(sf,
+ outshape=tuple(mat.rows for mat in sf.a_matrices if mat.face is None),
+ )
- if index:
- index = (index,)
+ if sf.index:
+ index = (sf.index,)
else:
index = ()
diff --git a/python/dune/perftool/sumfact/realization.py b/python/dune/perftool/sumfact/realization.py
index 2d789802..0ed1641e 100644
--- a/python/dune/perftool/sumfact/realization.py
+++ b/python/dune/perftool/sumfact/realization.py
@@ -69,7 +69,7 @@ def _realize_input(sf, insn_dep):
@generator_factory(item_tags=("sumfactkernel",),
context_tags=("kernel",),
cache_key_generator=lambda s, **kw: s.cache_key)
-def _realize_sum_factorization_kernel(sf, insn_dep=frozenset(), outshape=None, direct_input=None, direct_output=None):
+def _realize_sum_factorization_kernel(sf, insn_dep=frozenset(), outshape=None, direct_output=None):
# Unify the insn_dep parameter to be a frozenset
if isinstance(insn_dep, str):
insn_dep = frozenset({insn_dep})
@@ -91,6 +91,11 @@ def _realize_sum_factorization_kernel(sf, insn_dep=frozenset(), outshape=None, d
if sf.input:
insn_dep = insn_dep.union(frozenset({lp.match.Writes(sf.input)}))
+ # Construct the direct_input for the FastDG case
+ direct_input = None
+ if get_option('fastdg') and sf.stage == 1:
+ direct_input = sf.coeff_func(sf.restriction)
+
# Prepare some dim_tags/shapes for later use
ftags = ",".join(["f"] * sf.length)
novec_ftags = ftags
diff --git a/python/dune/perftool/sumfact/sumfact.py b/python/dune/perftool/sumfact/sumfact.py
index 50ac3c02..6106246c 100644
--- a/python/dune/perftool/sumfact/sumfact.py
+++ b/python/dune/perftool/sumfact/sumfact.py
@@ -172,7 +172,7 @@ def generate_accumulation_instruction(visitor, accterm, measure, subdomain_id):
inp = get_counted_variable("input")
# Initialize a base storage for this buffer and get a temporay pointing to it
- shape = tuple(mat.cols for mat in a_matrices if mat.face is None)
+ shape = tuple(mat.cols for mat in sf.a_matrices if mat.face is None)
dim_tags = ",".join(['f'] * local_dimension())
if index is not None:
shape = shape + (4,)
@@ -242,7 +242,7 @@ def generate_accumulation_instruction(visitor, accterm, measure, subdomain_id):
within_inames=frozenset(visitor.inames))})
inames = tuple(accum_iname((accterm.argument.restriction, restriction), mat.rows, i)
- for i, mat in enumerate(a_matrices))
+ for i, mat in enumerate(sf.a_matrices))
# Collect the lfs and lfs indices for the accumulate call
test_lfs = determine_accumulation_space(accterm.argument.expr, 0, measure)
--
GitLab