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

Reenable the distributed checkpointing test #8424

Merged
merged 1 commit into from
Dec 2, 2024
Merged

Reenable the distributed checkpointing test #8424

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
3 changes: 1 addition & 2 deletions test/tpu/run_tests.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,8 +8,7 @@ python3 test/pjrt/test_collective_ops_tpu.py
python3 test/spmd/test_mp_input_sharding.py
python3 test/spmd/test_xla_sharding.py
python3 test/spmd/test_xla_virtual_device.py
# TODO(JackCaoG): to reenable
# python3 test/spmd/test_xla_distributed_checkpoint.py
python3 test/spmd/test_xla_distributed_checkpoint.py
python3 test/spmd/test_train_spmd_linear_model.py
python3 test/spmd/test_xla_spmd_python_api_interaction.py
python3 test/spmd/test_xla_auto_sharding.py
Expand Down
12 changes: 11 additions & 1 deletion torch_xla/csrc/tensor.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -562,7 +562,17 @@ void XLATensor::UpdateFromTensor(at::Tensor tensor, bool sync) {
at::Tensor coyped_tensor = torch::lazy::CopyTensor(tensor, dtype());
SetTensorData(coyped_tensor);
data()->handle = nullptr;
data()->sharding = nullptr;
// if shape is different,
if (data()->sharding) {
auto coyped_tensor_dims = XlaHelpers::I64List(coyped_tensor.sizes());
auto sharding_dims = data()->sharding->shape.dimensions();
if (coyped_tensor_dims != sharding_dims) {
// sharding shape from origional tensor is different from the new cpu
// tensor, we need to clear the sharding here.
ClearShardingSpec();
}
}
// ClearShardingSpec();
AssignIrValue(torch::lazy::Value());
if (data()->view != nullptr) {
torch::lazy::Value ir_value = GetIrValueForTensor(coyped_tensor, device);
Expand Down