#include "op_plugin/AclOpsInterface.h"
#include "op_plugin/OpApiInterface.h"
#include "op_plugin/utils/op_api_common.h"
#include "op_plugin/utils/AdvancedIndex.h"
#include "torch_npu/csrc/aten/mirror/NPUMemoryOverlap.h"
namespace op_api {
using npu_preparation = at_npu::native::OpPreparation;
namespace {
void check_no_overlap(const at::Tensor& a, const at::Tensor& b)
{
const auto overlap_status = at_npu::native::get_overlap_status(a, b);
TORCH_CHECK(overlap_status != at_npu::native::MemOverlapStatus::PARTIAL &&
overlap_status != at_npu::native::MemOverlapStatus::FULL,
"unsupported operation: some elements of the input tensor and "
"the written-to tensor refer to a single memory location. "
"Please clone() the tensor before performing the operation.",
OPS_ERROR(ErrCode::NOT_SUPPORT));
}
}
at::Tensor index_put(
const at::Tensor& self,
const c10::List<c10::optional<at::Tensor>>& indices,
const at::Tensor& value,
bool accumulate) {
DO_COMPATIBILITY(aclnnIndexPutImpl, acl_op::index_put(self, indices, value, accumulate));
return self.clone(at::MemoryFormat::Contiguous).index_put_(indices, value, accumulate);
}
at::Tensor& index_put_(
at::Tensor& self,
const c10::List<c10::optional<at::Tensor>>& indices,
const at::Tensor& value,
const bool accumulate) {
DO_COMPATIBILITY(aclnnIndexPutImpl, acl_op::index_put_(self, indices, value, accumulate));
return at::_index_put_impl_(self, indices, value, accumulate, false);
}
at::Tensor& _index_put_impl_(
at::Tensor& self,
const c10::List<c10::optional<at::Tensor>>& indices,
const at::Tensor& value,
const bool accumulate,
const bool unsafe) {
DO_COMPATIBILITY(aclnnIndexPutImpl, acl_op::_index_put_impl_(self, indices, value, accumulate, unsafe));
if (self.device().type() == at::kCPU) {
return at::native::_index_put_impl_(self, indices, value, accumulate, unsafe);
}
if (at_npu::native::has_internal_overlap(self) == at_npu::native::MemOverlap::YES) {
TORCH_WARN(
"Use of index_put_ on expanded tensors is deprecated. "
"Please clone() the tensor before performing this operation. "
"This also applies to advanced indexing e.g. tensor[indices] = tensor");
}
check_no_overlap(self, value);
for (const c10::optional<at::Tensor>& index : indices) {
if (index.has_value()) {
check_no_overlap(self, *index);
}
}
bool needCast = op_plugin::AdvanceIndex::checkIndexTensorTypes(indices);
auto indices_after = op_plugin::AdvanceIndex::npu_expand_tensors(self, indices, needCast, true);
std::vector<at::Tensor> all_defined_indices;
at::SmallVector<int64_t, op_infer::N> zeroSize = {0};
at::Tensor emptyTensor = npu_preparation::apply_tensor_without_format(self, zeroSize);
c10::List<c10::optional<at::Tensor>> indices_expand_list;
for (at::Tensor index_opt : indices_after) {
indices_expand_list.push_back(index_opt);
}
auto info = op_plugin::AdvanceIndex::make_info(self, indices_expand_list);
TORCH_CHECK(op_plugin::AdvanceIndex::is_expandable_to(value.sizes(), info.src.sizes()),
"shape mismatch: value tensor of shape ", value.sizes(),
" cannot be broadcast to indexing result of shape ", info.src.sizes(),
OPS_ERROR(ErrCode::PARAM));
for (int i = 0; i < static_cast<int>(indices_after.size()); i++) {
if (indices_after[i].defined()) {
all_defined_indices.emplace_back(indices_after[i]);
continue;
}
all_defined_indices.emplace_back(emptyTensor);
}
for (auto &all_defined_indice : all_defined_indices) {
if (all_defined_indice.device() != self.device()) {
all_defined_indice = all_defined_indice.to(self.device());
}
}
at::TensorList indices_tensor_list = all_defined_indices;
if (self.numel() != 0 && value.numel() != 0) {
EXEC_NPU_CMD(aclnnIndexPutImpl, self, indices_tensor_list, value, accumulate, unsafe);
}
return self;
}
}