Supported allreduce sparse

ROCm / pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

http://pytorch.org

Other

219 stars 50 forks source link

Supported allreduce sparse #1409

Closed BLOrange-AMD closed 2 months ago

BLOrange-AMD commented 2 months ago

This PR is to fix https://ontrack-internal.amd.com/browse/SWDEV-447053 - test/distributed/test_distributed_spawn.py::TestDistBackendWithSpawn::test_sparse_all_reduce_sum_cuda with gloo backend. Rebuilt PyTorch after changes and unit test runs fine.

Reference: https://github.com/pytorch/pytorch/pull/111485