IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[CPU][ARM] Correct execution of the Reorder via Transpose (#20018)

Browse Source
This commit is contained in:
Egor Duplenskii 2023-10-02 13:24:43 +02:00 committed by GitHub
parent 7452656e5a
commit ae00b3e2ab
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 16 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/plugins/intel_cpu/src/nodes/executors/common/ref_transpose.hpp
View File

@ -37,4 +37,4 @@ public:
};
} // namespace intel_cpu
} // namespace ov
} // namespace ov

20
src/plugins/intel_cpu/src/nodes/reorder.cpp
View File

@ -119,8 +119,8 @@ void Reorder::executeDynamicImpl(dnnl::stream strm) {
#if defined(OV_CPU_ARM_ENABLE_FP16)
void Reorder::prepareReorderAsTranspose(MemoryDescPtr parentDesc, MemoryDescPtr childDesc) {
auto getOrder = [](const MemoryDesc& lhs, const MemoryDesc& rhs) -> std::pair<std::vector<size_t>, std::vector<size_t>> {
const auto& in = lhs.getShape().getStaticDims();
auto getOrderAndBlockedDims = [](const MemoryDesc& lhs, const MemoryDesc& rhs) -> std::pair<std::vector<size_t>, std::vector<size_t>> {
const auto& in = lhs.as<BlockedMemoryDesc>()->getBlockDims();
const auto rank = lhs.getShape().getRank();
if (lhs.hasLayoutType(LayoutType::ncsp) && rhs.hasLayoutType(LayoutType::nspc)) {
@ -142,17 +142,17 @@ void Reorder::prepareReorderAsTranspose(MemoryDescPtr parentDesc, MemoryDescPtr
}
};
auto order = getOrder(*parentDesc, *childDesc);
auto order = getOrderAndBlockedDims(*parentDesc, *childDesc);
const auto& transposeOrder = order.first;
const auto& transposedDims = order.second;
const auto& transposedBlockDims = order.second;
auto transposedDesc = std::make_shared<CpuBlockedMemoryDesc>(parentDesc->getPrecision(), Shape{transposedDims});
auto transposedDesc = std::make_shared<CpuBlockedMemoryDesc>(parentDesc->getPrecision(), Shape{transposedBlockDims});
TransposeParams transposeParams;
transposeParams.permuteParams.src_block_dims = parentDesc->as<BlockedMemoryDesc>()->getBlockDims();
transposeParams.permuteParams.src_block_order = parentDesc->as<BlockedMemoryDesc>()->getOrder();
transposeParams.permuteParams.dst_block_dims = transposedDesc->as<BlockedMemoryDesc>()->getBlockDims();
transposeParams.permuteParams.dst_block_order = transposedDesc->as<BlockedMemoryDesc>()->getOrder();
transposeParams.permuteParams.dst_block_dims = transposedBlockDims;
transposeParams.permuteParams.dst_block_order = transposeParams.permuteParams.src_block_order;
transposeParams.permuteParams.order = transposeOrder;
transposeParams.permuteParams.data_size = parentDesc->getPrecision().size();
@ -163,9 +163,9 @@ void Reorder::prepareReorderAsTranspose(MemoryDescPtr parentDesc, MemoryDescPtr
transpose_context);
dnnl::primitive_attr attr;
transposeExecutor = factory->makeExecutor(transposeParams,
{parentDesc},
{transposedDesc},
attr);
{parentDesc},
{transposedDesc},
attr);
getSelectedPrimitiveDescriptor()->setImplementationType(transposeExecutor->getImplType());
return;
}

10
src/plugins/intel_cpu/src/nodes/transpose.cpp
View File

@ -176,11 +176,11 @@ void Transpose::prepareParams() {
auto builder = [&srcDesc, &dstDesc, this](const PermuteParams& key) -> std::shared_ptr<TransposeExecutor> {
dnnl::primitive_attr attr;
auto selectedPD = getSelectedPrimitiveDescriptor();
auto jitExec = selectedPD->getExecutorFactoryAs<TransposeExecutorFactory>()->makeExecutor(transposeParams,
{srcDesc},
{dstDesc},
attr);
return jitExec;
auto executor = selectedPD->getExecutorFactoryAs<TransposeExecutorFactory>()->makeExecutor(transposeParams,
{srcDesc},
{dstDesc},
attr);
return executor;
};
auto cache = context->getParamsCache();