Commit fbf355ec authored by Martin Bauer's avatar Martin Bauer
Browse files

pystencils: CUDA kernels can now also iterate over slice

parent 882c84e6
......@@ -5,6 +5,7 @@ import pycuda.driver as cuda
import pycuda.autoinit
from pystencils.astnodes import Conditional, Block
from pystencils.slicing import normalizeSlice
BLOCK_IDX = list(sp.symbols("blockIdx.x blockIdx.y blockIdx.z"))
THREAD_IDX = list(sp.symbols("threadIdx.x threadIdx.y threadIdx.z"))
......@@ -15,8 +16,7 @@ class AbstractIndexing(abc.ABCMeta('ABC', (object,), {})):
Abstract base class for all Indexing classes. An Indexing class defines how a multidimensional
field is mapped to CUDA's block and grid system. It calculates indices based on CUDA's thread and block indices
and computes the number of blocks and threads a kernel is started with. The Indexing class is created with
a pystencils field and the number of ghost layers as well as further optional parameters that must have default
a pystencils field, a slice to iterate over, and further optional parameters that must have default values.
......@@ -55,12 +55,12 @@ class AbstractIndexing(abc.ABCMeta('ABC', (object,), {})):
class BlockIndexing(AbstractIndexing):
"""Generic indexing scheme that maps sub-blocks of an array to CUDA blocks."""
def __init__(self, field, ghostLayers, blockSize=(256, 8, 1), permuteBlockSizeDependentOnLayout=True):
def __init__(self, field, iterationSlice=None,
blockSize=(256, 8, 1), permuteBlockSizeDependentOnLayout=True):
:param field: pystencils field (common to all Indexing classes)
:param ghostLayers: number of ghost layers (common to all Indexing classes)
:param blockSize: size of a CUDA block
:param iterationSlice: slice that defines rectangular subarea which is iterated over
:param permuteBlockSizeDependentOnLayout: if True the blockSize is permuted such that the fastest coordinate
gets the largest amount of threads
......@@ -73,20 +73,21 @@ class BlockIndexing(AbstractIndexing):
blockSize = self.limitBlockSizeToDeviceMaximum(blockSize)
self._blockSize = blockSize
self._coordinates = [blockIndex * bs + threadIndex + gl[0]
for blockIndex, bs, threadIndex, gl in zip(BLOCK_IDX, blockSize, THREAD_IDX, ghostLayers)]
self._iterationSlice = iterationSlice
offsets = _getStartFromSlice(self._iterationSlice)
self._coordinates = [blockIndex * bs + threadIndex + off
for blockIndex, bs, threadIndex, off in zip(BLOCK_IDX, blockSize, THREAD_IDX, offsets)]
self._coordinates = self._coordinates[:field.spatialDimensions]
self._ghostLayers = ghostLayers
def coordinates(self):
return self._coordinates
def getCallParameters(self, arrShape):
dim = len(self._coordinates)
arrShape = [s - (gl[0] + gl[1]) for s, gl in zip(arrShape[:dim], self._ghostLayers)]
grid = tuple(math.ceil(length / blockSize) for length, blockSize in zip(arrShape, self._blockSize))
widths = [end - start for start, end in zip(_getStartFromSlice(self._iterationSlice),
_getEndFromSlice(self._iterationSlice, arrShape))]
grid = tuple(math.ceil(length / blockSize) for length, blockSize in zip(widths, self._blockSize))
extendBs = (1,) * (3 - len(self._blockSize))
extendGr = (1,) * (3 - len(grid))
return {'block': self._blockSize + extendBs,
......@@ -95,8 +96,8 @@ class BlockIndexing(AbstractIndexing):
def guard(self, kernelContent, arrShape):
dim = len(self._coordinates)
arrShape = arrShape[:dim]
conditions = [c < shapeComponent - gl[1]
for c, shapeComponent, gl in zip(self._coordinates, arrShape, self._ghostLayers)]
conditions = [c < end
for c, end in zip(self._coordinates, _getEndFromSlice(self._iterationSlice, arrShape))]
condition = conditions[0]
for c in conditions[1:]:
condition = sp.And(condition, c)
......@@ -178,7 +179,7 @@ class LineIndexing(AbstractIndexing):
maximum amount of threads allowed in a CUDA block (which depends on device).
def __init__(self, field, ghostLayers):
def __init__(self, field, iterationSlice=None):
availableIndices = [THREAD_IDX[0]] + BLOCK_IDX
if field.spatialDimensions > 4:
raise NotImplementedError("This indexing scheme supports at most 4 spatial dimensions")
......@@ -188,24 +189,51 @@ class LineIndexing(AbstractIndexing):
fastestCoordinate = field.layout[-1]
coordinates[0], coordinates[fastestCoordinate] = coordinates[fastestCoordinate], coordinates[0]
self._coordiantesNoGhostLayer = coordinates
self._coordinates = [i + gl[0] for i, gl in zip(coordinates, ghostLayers)]
self._ghostLayers = ghostLayers
self._coordinates = coordinates
self._iterationSlice = iterationSlice
def coordinates(self):
return self._coordinates
return [i + offset for i, offset in zip(self._coordinates, _getStartFromSlice(self._iterationSlice))]
def getCallParameters(self, arrShape):
widths = [end - start for start, end in zip(_getStartFromSlice(self._iterationSlice),
_getEndFromSlice(self._iterationSlice, arrShape))]
def getShapeOfCudaIdx(cudaIdx):
if cudaIdx not in self._coordiantesNoGhostLayer:
if cudaIdx not in self._coordinates:
return 1
idx = self._coordiantesNoGhostLayer.index(cudaIdx)
return arrShape[idx] - (self._ghostLayers[idx][0] + self._ghostLayers[idx][1])
idx = self._coordinates.index(cudaIdx)
return widths[idx]
return {'block': tuple([getShapeOfCudaIdx(idx) for idx in THREAD_IDX]),
'grid': tuple([getShapeOfCudaIdx(idx) for idx in BLOCK_IDX])}
def guard(self, kernelContent, arrShape):
return kernelContent
# -------------------------------------- Helper functions --------------------------------------------------------------
def _getStartFromSlice(iterationSlice):
res = []
for sliceComponent in iterationSlice:
if type(sliceComponent) is slice:
res.append(sliceComponent.start if sliceComponent.start is not None else 0)
assert isinstance(sliceComponent, int)
return res
def _getEndFromSlice(iterationSlice, arrShape):
iterSlice = normalizeSlice(iterationSlice, arrShape)
res = []
for sliceComponent in iterSlice:
if type(sliceComponent) is slice:
assert isinstance(sliceComponent, int)
res.append(sliceComponent + 1)
return res
\ No newline at end of file
......@@ -5,8 +5,8 @@ from pystencils.types import TypedSymbol, BasicType, StructType
from pystencils import Field
def createCUDAKernel(listOfEquations, functionName="kernel", typeForSymbol=None, indexingCreator=BlockIndexing,
def createCUDAKernel(listOfEquations, functionName="kernel", typeForSymbol=None, indexingCreator=BlockIndexing,
iterationSlice=None, ghostLayers=None):
fieldsRead, fieldsWritten, assignments = typeAllEquations(listOfEquations, typeForSymbol)
allFields = fieldsRead.union(fieldsWritten)
readOnlyFields = set([ for f in fieldsRead - fieldsWritten])
......@@ -16,13 +16,22 @@ def createCUDAKernel(listOfEquations, functionName="kernel", typeForSymbol=None,
commonShape = getCommonShape(allFields)
if ghostLayers is None:
requiredGhostLayers = max([fa.requiredGhostLayers for fa in fieldAccesses])
ghostLayers = [(requiredGhostLayers, requiredGhostLayers)] * len(commonShape)
if isinstance(ghostLayers, int):
ghostLayers = [(ghostLayers, ghostLayers)] * len(commonShape)
indexing = indexingCreator(field=list(fieldsRead)[0], ghostLayers=ghostLayers)
if iterationSlice is None:
# determine iteration slice from ghost layers
if ghostLayers is None:
# determine required number of ghost layers from field access
requiredGhostLayers = max([fa.requiredGhostLayers for fa in fieldAccesses])
ghostLayers = [(requiredGhostLayers, requiredGhostLayers)] * len(commonShape)
iterationSlice = []
if isinstance(ghostLayers, int):
for i in range(len(commonShape)):
iterationSlice.append(slice(ghostLayers[i], -ghostLayers[i]))
for i in range(len(commonShape)):
iterationSlice.append(slice(ghostLayers[i][0], -ghostLayers[i][1]))
indexing = indexingCreator(field=list(fieldsRead)[0], iterationSlice=iterationSlice)
block = Block(assignments)
block = indexing.guard(block, commonShape)
......@@ -71,7 +80,7 @@ def createdIndexedCUDAKernel(listOfEquations, indexFields, functionName="kernel"
coordinateTypedSymbols = [eq.lhs for eq in coordinateSymbolAssignments]
idxField = list(indexFields)[0]
indexing = indexingCreator(field=idxField, ghostLayers=[(0, 0)] * len(idxField.shape))
indexing = indexingCreator(field=idxField, iterationSlice=[slice(None, None, None)] * len(idxField.spatialShape))
functionBody = Block(coordinateSymbolAssignments + assignments)
functionBody = indexing.guard(functionBody, getCommonShape(indexFields))
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment