p47068915
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

optimize gather,copy indices before parallel launch

This commit is contained in:
chenjianping 2020-08-27 22:21:56 +08:00
parent 86c39d34ad
commit b7c7140bb1
2 changed files with 18 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
mindspore/lite/src/runtime/kernel/arm/fp32/gather.cc
View File

@ -37,6 +37,10 @@ int GatherCPUKernel::Init() {
return ReSize();
}
GatherCPUKernel::~GatherCPUKernel() {
context_->allocator->Free(indices_data_);
}
int GatherCPUKernel::ReSize() { return RET_OK; }
int GatherCPUKernel::DoGather(int task_id) {
@ -45,7 +49,6 @@ int GatherCPUKernel::DoGather(int task_id) {
auto out_tensor = out_tensors_.at(0);
auto input_ptr = reinterpret_cast<float *>(input_tensor->Data());
auto indices_ptr = reinterpret_cast<float *>(indices_tensor->Data());
auto output_ptr = reinterpret_cast<float *>(out_tensor->Data());
auto input_int32 = reinterpret_cast<int32_t *>(input_tensor->Data());
@ -57,13 +60,6 @@ int GatherCPUKernel::DoGather(int task_id) {
auto axis = (reinterpret_cast<GatherParameter *>(op_parameter_))->axis_;
const int limit = in_shape[axis];
for (int i = 0; i < indices_element_size; ++i) {
indices_data_[i] = static_cast<int>(indices_ptr[i]);
if (indices_data_[i] >= limit) {
MS_LOG(ERROR) << " indice data: " << indices_data_[i] << " is not in [ 0, " << limit - 1 << " ]";
return RET_ERROR;
}
}
int outer_size = 1, inner_size = 1;
for (int i = 0; i < axis; ++i) {
@ -106,12 +102,23 @@ int GatherCPUKernel::Run() {
}
auto indices_tensor = in_tensors_.at(1);
indices_data_ = reinterpret_cast<int *>(context_->allocator->Malloc(indices_tensor->ElementsNum() * sizeof(int)));
indices_data_ = reinterpret_cast<int *>(context_->allocator->Malloc(indices_tensor->Size()));
if (indices_data_ == nullptr) {
MS_LOG(ERROR) << "Memory allocation failed";
context_->allocator->Free(indices_data_);
return RET_ERROR;
}
auto in_shape = in_tensors_.at(0)->shape();
int indices_element_size = indices_tensor->ElementsNum();
auto axis = (reinterpret_cast<GatherParameter *>(op_parameter_))->axis_;;
auto indices_ptr = reinterpret_cast<float *>(indices_tensor->Data());
const int limit = in_shape[axis];
for (int i = 0; i < indices_element_size; ++i) {
indices_data_[i] = static_cast<int>(indices_ptr[i]);
if (indices_data_[i] >= limit) {
MS_LOG(ERROR) << " indice data: " << indices_data_[i] << " is not in [ 0, " << limit - 1 << " ]";
return RET_ERROR;
}
}
int error_code = ParallelLaunch(THREAD_POOL_DEFAULT, GatherRun, this, op_parameter_->thread_num_);
if (error_code != RET_OK) {
MS_LOG(ERROR) << "Gather function error error_code[" << error_code << "]";

2
mindspore/lite/src/runtime/kernel/arm/fp32/gather.h
View File

@ -28,7 +28,7 @@ class GatherCPUKernel : public LiteKernel {
const std::vector<lite::tensor::Tensor *> &outputs, const lite::Context *ctx,
const mindspore::lite::PrimitiveC *primitive)
: LiteKernel(parameter, inputs, outputs, ctx, primitive) {}
~GatherCPUKernel() override = default;
~GatherCPUKernel() override;
int Init() override;
int ReSize() override;