Sriram Kumar a19b667bca [ROCm] Update CUDAPluggableAllocator.h (#1984) (#150010) ... Altering the flag to use the correct streamType in CUDAPluggableAllocator class for ROCm gpu. The flag TORCH_HIP_VERSION does not work for ROCm as intended. This flag is replaced with USE_ROCM. This is impacting Distributed Fused Adam in Rocm/APEX when using nccl_ub feature. This has been tested with rocm/apex. See PR https://github.com/ROCm/apex/pull/184 Pull Request resolved: https://github.com/pytorch/pytorch/pull/150010 Approved by: https://github.com/jeffdaily		2025-04-01 16:49:03 +00:00
..
shared	[codemod] Fix unused-value issue in caffe2/aten/src/ATen/cuda/detail/CUDAHooks.cpp +4 (#147555)	2025-03-01 19:46:13 +00:00
comm.cpp
comm.h
CUDAPluggableAllocator.cpp	[GPU Snapshot] Add Clear History Flag (#149352)	2025-03-19 21:44:20 +00:00
CUDAPluggableAllocator.h	[ROCm] Update CUDAPluggableAllocator.h (#1984) (#150010)	2025-04-01 16:49:03 +00:00
device_set.h
Event.cpp
Event.h
GdsFile.cpp
GdsFile.h
Graph.cpp	Revert "Implement cuda graphs implementation of torch.cond and torch.while_loop (#140979)"	2025-02-13 18:04:26 +00:00
memory_snapshot.cpp	[GPU Snapshot] Add Clear History Flag (#149352)	2025-03-19 21:44:20 +00:00
memory_snapshot.h	[GPU Snapshot] Add Clear History Flag (#149352)	2025-03-19 21:44:20 +00:00
MemPool.cpp
Module.cpp	[GPU Snapshot] Add Clear History Flag (#149352)	2025-03-19 21:44:20 +00:00
Module.h
nccl.cpp	Revert "[Environment Variable][7/N] Use thread-safe getenv functions (#140211)"	2025-02-03 22:04:28 +00:00
nccl.h
python_comm.cpp
python_comm.h
python_nccl.cpp	Fix minor typo in python_nccl (#148088)	2025-02-28 00:47:09 +00:00
python_nccl.h
Stream.cpp
Stream.h
Tensor.cpp
THCP.h
utils.cpp	[BE] Add missing throw of std::runtime_error in scrc/cuda/utils.cpp (#144962)	2025-01-16 17:35:39 +00:00