Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Simplify size op impl #45808

Merged
merged 4 commits into from
Sep 9, 2022
Merged

Simplify size op impl #45808

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d85ceb5
simplify size op
chenwhql Sep 6, 2022
88f0a51
trans to cuda manuly
chenwhql Sep 7, 2022
66a61d2
fix copy error
chenwhql Sep 7, 2022
639d115
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
chenwhql Sep 8, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
32 changes: 0 additions & 32 deletions paddle/phi/kernels/cpu/size_kernel.cc
View file
Open in desktop

This file was deleted.

31 changes: 0 additions & 31 deletions paddle/phi/kernels/gpu/size_kernel.cu
View file
Open in desktop

This file was deleted.

33 changes: 19 additions & 14 deletions paddle/phi/kernels/impl/size_kernel_impl.h → paddle/phi/kernels/size_kernel.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,28 +12,33 @@
// See the License for the specific language governing permissions and
// limitations under the License.

#pragma once
#include "paddle/phi/kernels/size_kernel.h"

#include "paddle/phi/core/kernel_registry.h"
#include "paddle/phi/core/tensor_utils.h"

namespace phi {

template <typename T, typename Context>
template <typename Context>
void SizeKernel(const Context& ctx,
const DenseTensor& input,
DenseTensor* out) {
auto place = ctx.GetPlace();
auto out_data = ctx.template Alloc<int64_t>(out);
auto cpu_place = phi::CPUPlace();
if (place == cpu_place) {
out_data[0] = input.numel();
} else {
DenseTensor cpu_tensor;
cpu_tensor.Resize(out->dims());
auto cpu_data = ctx.template HostAlloc<int64_t>(&cpu_tensor);
cpu_data[0] = input.numel();
phi::Copy(ctx, cpu_tensor, place, false, out);
}
auto* out_data = ctx.template HostAlloc<int64_t>(out);
out_data[0] = input.numel();
}

} // namespace phi

PD_REGISTER_GENERAL_KERNEL(
size, CPU, ALL_LAYOUT, phi::SizeKernel<phi::CPUContext>, ALL_DTYPE) {
kernel->OutputAt(0).SetDataType(phi::DataType::INT64);
}

#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
PD_REGISTER_GENERAL_KERNEL(
size, GPU, ALL_LAYOUT, phi::SizeKernel<phi::GPUContext>, ALL_DTYPE) {
kernel->OutputAt(0)
.SetBackend(phi::Backend::CPU)
.SetDataType(phi::DataType::INT64);
}
#endif
2 changes: 1 addition & 1 deletion paddle/phi/kernels/size_kernel.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@

namespace phi {

template <typename T, typename Context>
template <typename Context>
void SizeKernel(const Context& ctx, const DenseTensor& input, DenseTensor* out);

} // namespace phi
3 changes: 3 additions & 0 deletions python/paddle/distributed/collective.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1136,6 +1136,9 @@ def all_gather_object(object_list, obj, group=None):
), "all_gather_object doesn't support static graph mode."

tensor, len_of_tensor = _convert_object_to_tensor(obj)
if paddle.get_device() != "cpu":
len_of_tensor = len_of_tensor._copy_to(
paddle.framework._current_expected_place(), False)

# gather len_of_tensor from all ranks
list_len_of_tensor = []
Expand Down