diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 9d960a6f4eb920f920fb7398c1816e447e97e05e..6ee07ae45d90f7fa66cf6d8ca7e219d5cde2017b 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -53,6 +53,19 @@ Python 3.5 POCL:
except:
- tags
+Python 3.6 POCL:
+ script:
+ - export PY_EXE=python3.6
+ - export PYOPENCL_TEST=portable
+ - export EXTRA_INSTALL="numpy mako"
+ - curl -L -O -k https://gitlab.tiker.net/inducer/ci-support/raw/master/build-and-test-py-project.sh
+ - ". ./build-and-test-py-project.sh"
+ tags:
+ - python3.6
+ - pocl
+ except:
+ - tags
+
Documentation:
script:
- EXTRA_INSTALL="numpy mako"
diff --git a/examples/curve-pot.py b/examples/curve-pot.py
index 9d8678f5e6d962c5656fa572c06d48b3a167a932..f42122088dbb21c12b589804566618fd35e1ad7c 100644
--- a/examples/curve-pot.py
+++ b/examples/curve-pot.py
@@ -229,6 +229,7 @@ def draw_pot_figure(aspect_ratio,
else:
# {{{ 3D plots
+
plotval_vol = vol_pot.real
plotval_c = curve_pot.real
@@ -264,7 +265,8 @@ def draw_pot_figure(aspect_ratio,
if __name__ == "__main__":
draw_pot_figure(aspect_ratio=1, nsrc=100, novsmp=100, helmholtz_k=(15+4j)*0.3,
what_operator="D", what_operator_lpot="D", force_center_side=1)
- pt.savefig("eigvals-ext-nsrc100-novsmp100.pdf")
+
+# pt.savefig("eigvals-ext-nsrc100-novsmp100.pdf")
#pt.clf()
#draw_pot_figure(aspect_ratio=1, nsrc=100, novsmp=100, helmholtz_k=0,
# what_operator="D", what_operator_lpot="D", force_center_side=-1)
diff --git a/sumpy/cse.py b/sumpy/cse.py
index 3318fba023eccbc7673fab31db5d31b0cfeb4035..573c78c93961c8fe0e69c9650c4df7d10133e98b 100644
--- a/sumpy/cse.py
+++ b/sumpy/cse.py
@@ -187,11 +187,32 @@ class FuncArgTracker(object):
from collections import defaultdict
count_map = defaultdict(lambda: 0)
- for arg in argset:
- for func_i in self.arg_to_funcset[arg]:
+ # Sorted by size to make best use of the performance hack below.
+ funcsets = sorted((self.arg_to_funcset[arg] for arg in argset), key=len)
+
+ for funcset in funcsets[:-threshold+1]:
+ for func_i in funcset:
if func_i >= min_func_i:
count_map[func_i] += 1
+ for i, funcset in enumerate(funcsets[-threshold+1:]):
+ # When looking at the tail end of the funcsets list, items below
+ # this threshold in the count_map don't have to be considered
+ # because they can't possibly be in the output.
+ count_map_threshold = i + 1
+
+ # We pick the smaller of the two containers to iterate over to
+ # reduce the number of items we have to look at.
+ (smaller_funcs_container,
+ larger_funcs_container) = sorted([funcset, count_map], key=len)
+
+ for func_i in smaller_funcs_container:
+ if count_map[func_i] < count_map_threshold:
+ continue
+
+ if func_i in larger_funcs_container:
+ count_map[func_i] += 1
+
return dict(
(k, v) for k, v in count_map.items()
if v >= threshold)
@@ -275,14 +296,14 @@ def match_common_args(func_class, funcs, opt_subs):
from sumpy.tools import OrderedSet
for i in range(len(funcs)):
- common_arg_candidates = arg_tracker.get_common_arg_candidates(
+ common_arg_candidates_counts = arg_tracker.get_common_arg_candidates(
arg_tracker.func_to_argset[i], i + 1, threshold=2)
# Sort the candidates in order of match size.
# This makes us try combining smaller matches first.
common_arg_candidates = OrderedSet(sorted(
- common_arg_candidates.keys(),
- key=lambda k: (common_arg_candidates[k], k)))
+ common_arg_candidates_counts.keys(),
+ key=lambda k: (common_arg_candidates_counts[k], k)))
while common_arg_candidates:
j = common_arg_candidates.pop(last=False)
diff --git a/sumpy/kernel.py b/sumpy/kernel.py
index 0121cf6bcd198e0c04821d06b7110020e8428cc7..e29ac5167051e5d781a9fdb331d97fb3fef84df2 100644
--- a/sumpy/kernel.py
+++ b/sumpy/kernel.py
@@ -27,7 +27,6 @@ from six.moves import range, zip
import loopy as lp
import numpy as np
from pymbolic.mapper import IdentityMapper, CSECachingMapperMixin
-import sumpy.symbolic as sym
from sumpy.symbolic import pymbolic_real_norm_2
from pymbolic.primitives import make_sym_vector
from pymbolic import var
@@ -51,6 +50,7 @@ PDE kernels
.. autoclass:: LaplaceKernel
.. autoclass:: HelmholtzKernel
.. autoclass:: StokesletKernel
+.. autoclass:: StressletKernel
Derivatives
-----------
@@ -264,11 +264,10 @@ class ExpressionKernel(Kernel):
if self.dim != len(dist_vec):
raise ValueError("dist_vec length does not match expected dimension")
- import sumpy.symbolic as sym
- expr = expr.subs(dict(
- (sym.Symbol("d%d" % i), dist_vec_i)
+ expr = expr.subs([
+ ("d%d" % i, dist_vec_i)
for i, dist_vec_i in enumerate(dist_vec)
- ))
+ ])
return expr
@@ -429,7 +428,7 @@ class StokesletKernel(ExpressionKernel):
init_arg_names = ("dim", "icomp", "jcomp", "viscosity_mu_name")
def __init__(self, dim, icomp, jcomp, viscosity_mu_name="mu"):
- """
+ r"""
:arg viscosity_mu_name: The argument name to use for
dynamic viscosity :math:`\mu` the then generating functions to
evaluate this kernel.
@@ -494,12 +493,11 @@ class StokesletKernel(ExpressionKernel):
class StressletKernel(ExpressionKernel):
- init_arg_names = ("dim", "icomp", "jcomp", "viscosity_mu_name",
- "stresslet_vector_name")
+ init_arg_names = ("dim", "icomp", "jcomp", "kcomp", "viscosity_mu_name")
- def __init__(self, dim=None, icomp=None, jcomp=None, viscosity_mu_name="mu",
- stresslet_vector_name="stresslet_vec"):
- """
+ def __init__(self, dim=None, icomp=None, jcomp=None, kcomp=None,
+ viscosity_mu_name="mu"):
+ r"""
:arg viscosity_mu_name: The argument name to use for
dynamic viscosity :math:`\mu` the then generating functions to
evaluate this kernel.
@@ -508,23 +506,17 @@ class StressletKernel(ExpressionKernel):
if dim == 2:
d = make_sym_vector("d", dim)
- n = make_sym_vector(stresslet_vector_name, dim)
r = pymbolic_real_norm_2(d)
expr = (
- sum(n[axis]*d[axis] for axis in range(dim))
- *
- d[icomp]*d[jcomp]/r**4
+ d[icomp]*d[jcomp]*d[kcomp]/r**4
)
scaling = 1/(var("pi"))
elif dim == 3:
d = make_sym_vector("d", dim)
- n = make_sym_vector(stresslet_vector_name, dim)
r = pymbolic_real_norm_2(d)
expr = (
- sum(n[axis]*d[axis] for axis in range(dim))
- *
- d[icomp]*d[jcomp]/r**5
+ d[icomp]*d[jcomp]*d[kcomp]/r**5
)
scaling = -3/(4*var("pi"))
@@ -535,9 +527,9 @@ class StressletKernel(ExpressionKernel):
raise RuntimeError("unsupported dimensionality")
self.viscosity_mu_name = viscosity_mu_name
- self.stresslet_vector_name = stresslet_vector_name
self.icomp = icomp
self.jcomp = jcomp
+ self.kcomp = kcomp
ExpressionKernel.__init__(
self,
@@ -547,42 +539,26 @@ class StressletKernel(ExpressionKernel):
is_complex_valued=False)
def __getinitargs__(self):
- return (self._dim, self.icomp, self.jcomp, self.viscosity_mu_name,
- self.stresslet_vector_name)
+ return (self._dim, self.icomp, self.jcomp, self.kcomp,
+ self.viscosity_mu_name)
def update_persistent_hash(self, key_hash, key_builder):
key_hash.update(type(self).__name__.encode())
key_builder.rec(key_hash, (
- self.dim, self.icomp, self.jcomp, self.viscosity_mu_name,
- self.stresslet_vector_name))
+ self.dim, self.icomp, self.jcomp, self.kcomp,
+ self.viscosity_mu_name))
def __repr__(self):
- return "StressletKnl%dD_%d%d[%s]" % (self.dim, self.icomp, self.jcomp,
- self.stresslet_vector_name)
+ return "StressletKnl%dD_%d%d%d" % (self.dim, self.icomp, self.jcomp,
+ self.kcomp)
def get_args(self):
return [
KernelArgument(
loopy_arg=lp.ValueArg(self.viscosity_mu_name, np.float64),
- ),
- KernelArgument(
- loopy_arg=lp.GlobalArg(self.stresslet_vector_name,
- None,
- shape=(self.dim, "nsources"),
- dim_tags="sep,C"))
+ )
]
- def get_code_tranformer(self):
- from sumpy.codegen import VectorComponentRewriter
- vcr = VectorComponentRewriter([self.stresslet_vector_name])
- from pymbolic.primitives import Variable
- via = _VectorIndexAdder(self.stresslet_vector_name, (Variable("isrc"),))
-
- def transform(expr):
- return via(vcr(expr))
-
- return transform
-
mapper_method = "map_stresslet_kernel"
# }}}
@@ -739,12 +715,12 @@ class DirectionalTargetDerivative(DirectionalDerivative):
dim = len(bvec)
assert dim == self.dim
- from sumpy.symbolic import make_sym_vector
- dir_vec = make_sym_vector(self.dir_vec_name, dim)
+ from sumpy.symbolic import make_sympy_vector
+ dir_vec = make_sympy_vector(self.dir_vec_name, dim)
# bvec = tgt-center
- return sym.Add(*tuple(dir_vec[axis]*expr.diff(bvec[axis])
- for axis in range(dim)))
+ return sum(dir_vec[axis]*expr.diff(bvec[axis])
+ for axis in range(dim))
mapper_method = "map_directional_target_derivative"
@@ -770,12 +746,12 @@ class DirectionalSourceDerivative(DirectionalDerivative):
dimensions = len(avec)
assert dimensions == self.dim
- from sumpy.symbolic import make_sym_vector
- dir_vec = make_sym_vector(self.dir_vec_name, dimensions)
+ from sumpy.symbolic import make_sympy_vector
+ dir_vec = make_sympy_vector(self.dir_vec_name, dimensions)
# avec = center-src -> minus sign from chain rule
- return sym.Add(*tuple(-dir_vec[axis]*expr.diff(avec[axis])
- for axis in range(dimensions)))
+ return sum(-dir_vec[axis]*expr.diff(avec[axis])
+ for axis in range(dimensions))
mapper_method = "map_directional_source_derivative"
@@ -909,8 +885,8 @@ class KernelDimensionSetter(KernelIdentityMapper):
return StressletKernel(self.dim,
kernel.icomp,
kernel.jcomp,
- viscosity_mu_name=kernel.viscosity_mu_name,
- stresslet_vector_name=kernel.stresslet_vector_name)
+ kernel.kcomp,
+ viscosity_mu_name=kernel.viscosity_mu_name)
# }}}