288 lines
12 KiB
C++
288 lines
12 KiB
C++
/*******************************************************************************
|
|
* Copyright (c) 2015-2018 Skymind, Inc.
|
|
*
|
|
* This program and the accompanying materials are made available under the
|
|
* terms of the Apache License, Version 2.0 which is available at
|
|
* https://www.apache.org/licenses/LICENSE-2.0.
|
|
*
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
|
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
|
* License for the specific language governing permissions and limitations
|
|
* under the License.
|
|
*
|
|
* SPDX-License-Identifier: Apache-2.0
|
|
******************************************************************************/
|
|
|
|
//
|
|
// @author raver119@gmail.com
|
|
// @author Yurii Shyrma (iuriish@yahoo.com)
|
|
//
|
|
|
|
#include <types/types.h>
|
|
#include <system/op_boilerplate.h>
|
|
#include <loops/reduce_same.h>
|
|
#include <loops/legacy_ops.h>
|
|
#include <helpers/OmpLaunchHelper.h>
|
|
#include <chrono>
|
|
#include <helpers/Loops.h>
|
|
#include <helpers/ConstantTadHelper.h>
|
|
|
|
using namespace simdOps;
|
|
|
|
namespace functions {
|
|
namespace reduce {
|
|
template <typename X>
|
|
template <typename OpType>
|
|
void _CUDA_H ReduceSameFunction<X>::execScalar(void *vx,
|
|
Nd4jLong *xShapeInfo,
|
|
void *vextraParams,
|
|
void *vz,
|
|
Nd4jLong *zShapeInfo) {
|
|
auto x = reinterpret_cast<X *>(vx);
|
|
auto z = reinterpret_cast<X *>(vz);
|
|
auto extraParams = reinterpret_cast<X *>(vextraParams);
|
|
|
|
const auto length = shape::length(xShapeInfo);
|
|
const auto xEws = shape::elementWiseStride(xShapeInfo);
|
|
const int rank = shape::rank(xShapeInfo);
|
|
|
|
if (shape::isEmpty(xShapeInfo)) {
|
|
z[0] = OpType::startingValue(x);
|
|
return;
|
|
}
|
|
|
|
if(sd::ArrayOptions::arrayType(xShapeInfo) == sd::ArrayType::EMPTY) {
|
|
if(sd::ArrayOptions::arrayType(zShapeInfo) == sd::ArrayType::EMPTY)
|
|
return;
|
|
const auto startingVal = OpType::startingValue(x);
|
|
|
|
for (Nd4jLong i = 0; i < length; i++)
|
|
z[i] = startingVal;
|
|
return;
|
|
}
|
|
|
|
if (xEws >= 1) {
|
|
z[0] = execScalar<OpType>(x, xEws, length, extraParams);
|
|
}
|
|
else {
|
|
auto startingValue = OpType::startingValue(x);
|
|
uint xShapeInfoCast[MAX_RANK];
|
|
const bool canCastX = sd::DataTypeUtils::castShapeInfo(xShapeInfo, xShapeInfoCast);
|
|
int maxThreads = sd::math::nd4j_min<int>(64, sd::Environment::getInstance()->maxMasterThreads());
|
|
X intermediate[64];
|
|
|
|
PRAGMA_OMP_SIMD
|
|
for (auto e = 0; e < maxThreads; e++)
|
|
intermediate[e] = OpType::startingValue(x);
|
|
|
|
auto func = PRAGMA_THREADS_FOR {
|
|
for (auto i = start; i < stop; i++)
|
|
intermediate[thread_id] = OpType::update(intermediate[thread_id], OpType::op(x[shape::indexOffset(i, xShapeInfo, xShapeInfoCast, canCastX)], extraParams), extraParams);
|
|
};
|
|
|
|
maxThreads = sd::Threads::parallel_for(func, 0, length, 1, maxThreads);
|
|
|
|
// merge results
|
|
for (int e = 1; e < maxThreads; e++)
|
|
intermediate[0] = OpType::update(intermediate[0], intermediate[e], extraParams);
|
|
|
|
// write out results
|
|
z[0] = OpType::postProcess(intermediate[0], length, extraParams);
|
|
}
|
|
}
|
|
|
|
|
|
template <typename X>
|
|
template <typename OpType>
|
|
X _CUDA_H ReduceSameFunction<X>::execScalar(void *vx,
|
|
Nd4jLong *xShapeInfo,
|
|
void *vextraParams) {
|
|
auto x = reinterpret_cast<X *>(vx);
|
|
auto extraParams = reinterpret_cast<X *>(vextraParams);
|
|
|
|
const Nd4jLong length = shape::length(xShapeInfo);
|
|
const auto xEws = shape::elementWiseStride(xShapeInfo);
|
|
|
|
if (xEws >= 1) {
|
|
return execScalar<OpType>(x, xEws, length, extraParams);
|
|
} else {
|
|
auto startingValue = OpType::startingValue(x);
|
|
uint xShapeInfoCast[MAX_RANK];
|
|
bool canCastX = sd::DataTypeUtils::castShapeInfo(xShapeInfo, xShapeInfoCast);
|
|
|
|
for (Nd4jLong i = 0; i < length; i++)
|
|
startingValue = OpType::update(startingValue, OpType::op(x[shape::indexOffset(i, xShapeInfo, xShapeInfoCast, canCastX)], extraParams), extraParams);
|
|
|
|
return OpType::postProcess(startingValue, length, extraParams);
|
|
}
|
|
}
|
|
|
|
template <typename X>
|
|
X ReduceSameFunction<X>::execScalar(const int opNum,
|
|
void *x,
|
|
Nd4jLong *xShapeInfo,
|
|
void *extraParams) {
|
|
RETURNING_DISPATCH_BY_OPNUM_T(execScalar, PARAMS(x, xShapeInfo, extraParams), REDUCE_SAME_OPS);
|
|
}
|
|
|
|
template <typename X>
|
|
void ReduceSameFunction<X>::execScalar(const int opNum,
|
|
void *x,
|
|
Nd4jLong *xShapeInfo,
|
|
void *extraParams,
|
|
void *z,
|
|
Nd4jLong *zShapeInfo) {
|
|
DISPATCH_BY_OPNUM_T(execScalar, PARAMS(x, xShapeInfo, extraParams, z, zShapeInfo), REDUCE_SAME_OPS);
|
|
}
|
|
|
|
template <typename X>
|
|
void ReduceSameFunction<X>::exec(const int opNum,
|
|
void *x,
|
|
Nd4jLong *xShapeInfo,
|
|
void *extraParams,
|
|
void *z,
|
|
Nd4jLong *zShapeInfo,
|
|
int *dimension,
|
|
int dimensionLength,
|
|
Nd4jLong *tadShapeInfo,
|
|
Nd4jLong *tadOffset, int64_t start, int64_t stop) {
|
|
DISPATCH_BY_OPNUM_T(exec, PARAMS(x,
|
|
xShapeInfo,
|
|
extraParams,
|
|
z,
|
|
zShapeInfo,
|
|
dimension,
|
|
dimensionLength,
|
|
tadShapeInfo,
|
|
tadOffset, start, stop),
|
|
REDUCE_SAME_OPS);
|
|
}
|
|
|
|
template <typename X>
|
|
template <typename OpType>
|
|
void _CUDA_H ReduceSameFunction<X>::exec(void *vx,
|
|
Nd4jLong *xShapeInfo,
|
|
void *vextraParams,
|
|
void *vz,
|
|
Nd4jLong *zShapeInfo,
|
|
int *dimension,
|
|
int dimensionLength,
|
|
Nd4jLong *tadShapeInfo,
|
|
Nd4jLong *tadOffset, int64_t start, int64_t stop) {
|
|
|
|
auto x = reinterpret_cast<X *>(vx);
|
|
auto z = reinterpret_cast<X *>(vz);
|
|
auto extraParams = reinterpret_cast<X *>(vextraParams);
|
|
|
|
auto zLength = shape::length(zShapeInfo);
|
|
|
|
if(sd::ArrayOptions::arrayType(xShapeInfo) == sd::ArrayType::EMPTY) {
|
|
if(sd::ArrayOptions::arrayType(zShapeInfo) == sd::ArrayType::EMPTY)
|
|
return;
|
|
const auto startingVal = OpType::startingValue(x);
|
|
|
|
for (Nd4jLong i = 0; i < zLength; i++)
|
|
z[i] = startingVal;
|
|
return;
|
|
}
|
|
|
|
//pre squeezed: this is for keeping the pointer to the original
|
|
//shape information for tad offset
|
|
//the squeezed information doesn't render the right strides for
|
|
//tad offset
|
|
// || tad.wholeThing
|
|
if (zLength == 1 || dimension == nullptr || dimensionLength == shape::rank(xShapeInfo)) {
|
|
z[0] = execScalar<OpType>(x, xShapeInfo, extraParams);
|
|
return;
|
|
}
|
|
|
|
if (OpType::requiresSpecialAccumulation) {
|
|
OpType::execSpecial(x, xShapeInfo, extraParams, z, zShapeInfo, dimension, dimensionLength, tadShapeInfo, tadOffset);
|
|
return;
|
|
}
|
|
|
|
auto tadOnlyShapeInfo = tadShapeInfo;
|
|
auto tadOffsets = tadOffset;
|
|
|
|
if (tadOnlyShapeInfo == nullptr || tadOffsets == nullptr) {
|
|
if (dimensionLength < 1)
|
|
return;
|
|
|
|
auto tadPack = sd::ConstantTadHelper::getInstance()->tadForDimensions(xShapeInfo, dimension, dimensionLength);
|
|
tadOnlyShapeInfo = tadPack.primaryShapeInfo();
|
|
tadOffsets = tadPack.primaryOffsets();
|
|
}
|
|
|
|
#ifdef INLINE_LOOPS
|
|
sd::ReductionLoops<X,X,X>::template loopReduce<OpType>(x, xShapeInfo, z, zShapeInfo, tadOnlyShapeInfo, tadOffsets, extraParams, start, stop);
|
|
#else
|
|
sd::ReductionSameLoops<X>::template innerloopReduce<OpType>(x, xShapeInfo, z, zShapeInfo, tadOnlyShapeInfo, tadOffsets, extraParams, start, stop);
|
|
#endif
|
|
}
|
|
|
|
|
|
template <typename X>
|
|
template<typename OpType>
|
|
void _CUDA_H ReduceSameFunction<X>::exec(void *x,
|
|
Nd4jLong *xShapeInfo,
|
|
void *extraParams,
|
|
void *vz,
|
|
Nd4jLong *zShapeInfo) {
|
|
// FIXME: wtf???
|
|
auto z = reinterpret_cast<X*>(vz);
|
|
z[0] = execScalar<OpType>(x, xShapeInfo, extraParams);
|
|
}
|
|
|
|
template <typename X>
|
|
template <typename OpType>
|
|
X _CUDA_H ReduceSameFunction<X>::execScalar(void *vx, Nd4jLong xEws, Nd4jLong length, void *vextraParams) {
|
|
|
|
auto x = reinterpret_cast<X *>(vx);
|
|
auto extraParams = reinterpret_cast<X *>(vextraParams);
|
|
int maxThreads = sd::math::nd4j_min<int>(64, sd::Environment::getInstance()->maxMasterThreads());
|
|
X intermediate[64];
|
|
|
|
PRAGMA_OMP_SIMD
|
|
for (auto e = 0; e < maxThreads; e++)
|
|
intermediate[e] = OpType::startingValue(x);
|
|
|
|
#ifdef _OPENMP
|
|
|
|
if (xEws == 1) {
|
|
PRAGMA_OMP_PARALLEL_FOR_THREADS(maxThreads)
|
|
for (Nd4jLong i = 0; i < length; i++)
|
|
intermediate[omp_get_thread_num()] = OpType::update(intermediate[omp_get_thread_num()], OpType::op(x[i], extraParams), extraParams);
|
|
} else {
|
|
PRAGMA_OMP_PARALLEL_FOR_THREADS(maxThreads)
|
|
for (Nd4jLong i = 0; i < length; i++)
|
|
intermediate[omp_get_thread_num()] = OpType::update(intermediate[omp_get_thread_num()], OpType::op(x[i * xEws], extraParams), extraParams);
|
|
}
|
|
|
|
#else
|
|
auto func = PRAGMA_THREADS_FOR {
|
|
if (xEws == 1) {
|
|
for (auto i = start; i < stop; i++)
|
|
intermediate[thread_id] = OpType::update(intermediate[thread_id], OpType::op(x[i], extraParams), extraParams);
|
|
} else {
|
|
for (auto i = start; i < stop; i++)
|
|
intermediate[thread_id] = OpType::update(intermediate[thread_id], OpType::op(x[i * xEws], extraParams), extraParams);
|
|
}
|
|
};
|
|
|
|
maxThreads = sd::Threads::parallel_for(func, 0, length, 1, maxThreads);
|
|
|
|
#endif
|
|
|
|
// merge results
|
|
for (int e = 1; e < maxThreads; e++)
|
|
intermediate[0] = OpType::update(intermediate[0], intermediate[e], extraParams);
|
|
|
|
// return result
|
|
return OpType::postProcess(intermediate[0], length, extraParams);
|
|
}
|
|
|
|
|
|
BUILD_SINGLE_TEMPLATE(template class ND4J_EXPORT ReduceSameFunction, , LIBND4J_TYPES);
|
|
}
|
|
} |