fduwjj bf5c7bf06d [FR] Fix the bug in FR script (e.g., checking all ranks dump check) (#134383) ...

We somehow convert the rank to string which makes the ranks check fail. This fix now convert them all to int.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/134383
Approved by: https://github.com/c-p-i-o

2024-08-26 08:21:14 +00:00

..

_composable

[FSDP2] Resolved strided sharding todo in clipping tests (#134152)

2024-08-22 12:45:13 +00:00

_shard

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

_tensor

Revert "[dtensor][MTPG] make sharding prop lru cache not shared among threads (#134294)"

2024-08-25 11:16:04 +00:00

_tools

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

algorithms

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

bin

Add test owner to distributed files starting with test_ (#66797)

2021-10-19 10:55:20 -07:00

checkpoint

[DCP] Fix meta tensor loading (#133256)

2024-08-14 21:07:11 +00:00

elastic

Allow mp.start_processes to create processes in parallel (#133707)

2024-08-23 17:11:20 +00:00

flight_recorder

[FR] Fix the bug in FR script (e.g., checking all ranks dump check) (#134383)

2024-08-26 08:21:14 +00:00

fsdp

[Test] Add SkipIfRocm to test_grad_acc_cpu_offload (#132975)

2024-08-16 18:26:20 +00:00

launcher

Validate logs are created by default (#128522)

2024-06-17 20:07:13 +00:00

nn/jit

[BE][Easy] replace import pathlib with from pathlib import Path (#129426)

2024-06-30 01:36:07 +00:00

optim

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

pipelining

[PP] Add ZeroBubble schedule (#133467)

2024-08-22 13:32:15 +00:00

rpc

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

tensor/parallel

Enable DTensor sharding propagation of native_layer_norm_backward to more fully accommodate optional args (#133502)

2024-08-24 05:49:54 +00:00

argparse_util_test.py

[skip ci] set more tests with owners for distributed and elastic (#67583)

2021-11-01 12:26:03 -07:00

test_c10d_common.py

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

test_c10d_functional_native.py

[inductor]Add DtypeView to avoid memory leak and unnecessary kernel generations (#128883)

2024-07-23 17:31:39 +00:00

test_c10d_gloo.py

[BE][Easy] enable ruff rule PIE790: unnecessary pass statement (#133200)

2024-08-15 15:50:19 +00:00

test_c10d_logger.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_c10d_nccl.py

Killswitch based rollout for flight recorder (#133237)

2024-08-20 04:27:55 +00:00

test_c10d_object_collectives.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_c10d_ops_nccl.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_c10d_pypg.py

Enable UFMT on all of test/distributed (#123539)

2024-04-17 06:46:02 +00:00

test_c10d_spawn_gloo.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_c10d_spawn_nccl.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_c10d_spawn_ucc.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_c10d_spawn.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_c10d_ucc.py

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

test_collective_utils.py

Enable UFMT on all of test/distributed (#123539)

2024-04-17 06:46:02 +00:00

test_compute_comm_reordering.py

[Traceable FSDP2][Inductor] Create grouped nodes for FSDP2 all-gather code block and reduce-scatter code block (after Buffer/Operation split) (#131510)

2024-07-27 08:39:58 +00:00

test_control_collectives.py

[c10d] add a simple test to demonstrate the user usage of collectives (#127665)

2024-06-05 04:32:11 +00:00

test_data_parallel.py

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

test_device_mesh.py

[DeviceMesh] Allow _flatten() to take in an optional mesh_dim_name (#134048)

2024-08-25 10:36:01 +00:00

test_distributed_spawn.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_dynamo_distributed.py

Make FX Graph Cache work with distributed training (#133374)

2024-08-16 18:51:14 +00:00

test_fake_pg.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_functional_api.py

Only make wait_tensor as a side_effect op (#132341)

2024-08-02 01:24:40 +00:00

test_inductor_collectives.py

Modify ExternKernelAlloc with NoneLayout to not assign its result to anything (#129188)

2024-06-22 02:57:44 +00:00

test_launcher.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_multi_threaded_pg.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_nccl.py

[BE][Easy][10/19] enforce style for empty lines in import segments in test/d*/ (#129761)

2024-07-17 16:57:39 +00:00

test_pg_wrapper.py

[torch/distributed] Check gloo availability when doing isinstance(pg,… (#124233)

2024-04-19 04:07:00 +00:00

test_store.py

Add None return type to init -- tests rest (#132376)

2024-08-01 15:44:51 +00:00

test_symmetric_memory.py

[SymmetricMemory] introduce multicast support, multimem_all_reduce_ and multimem_one_shot_all_reduce (#133424)

2024-08-23 20:09:20 +00:00