Add sharding tests to multigpu-test.sh (pytorch#77708)

Summary: These tests were being skipped since they don't run on multigpu
jobs.
Pull Request resolved: pytorch#77708
Approved by: https://github.com/wanchaol

.jenkins/pytorch/multigpu-test.sh

@@ -28,4 +28,27 @@ time python test/run_test.py --verbose -i distributed/test_c10d_spawn_nccl
 time python test/run_test.py --verbose -i distributed/test_store
 time python test/run_test.py --verbose -i distributed/test_pg_wrapper
 time python test/run_test.py --verbose -i distributed/rpc/cuda/test_tensorpipe_agent
+time python test/run_test.py --verbose -i distributed/_shard/checkpoint/test_checkpoint
+time python test/run_test.py --verbose -i distributed/_shard/checkpoint/test_file_system_checkpoint
+time python test/run_test.py --verbose -i distributed/_shard/sharding_spec/test_sharding_spec
+time python test/run_test.py --verbose -i distributed/_shard/sharding_plan/test_sharding_plan
+time python test/run_test.py --verbose -i distributed/_shard/sharding_spec/test_sharding_spec
+time python test/run_test.py --verbose -i distributed/_shard/sharding_plan/test_sharding_plan
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/test_megatron_prototype
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/test_sharded_tensor
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/test_sharded_tensor_reshard
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_chunk
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_elementwise_ops
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_embedding
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_embedding_bag
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_binary_cmp
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_init
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_linear
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_math_ops
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_matrix_ops
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/ops/test_softmax
+time python test/run_test.py --verbose -i distributed/_shard/sharding_spec/test_sharding_spec
+time python test/run_test.py --verbose -i distributed/_shard/sharded_optim/test_sharded_optim
+time python test/run_test.py --verbose -i distributed/_shard/test_partial_tensor
+time python test/run_test.py --verbose -i distributed/_shard/test_replicated_tensor
 assert_git_not_dirty

test/run_test.py

@@ -201,6 +201,8 @@ def skip_test_p(name: str) -> bool:
     "distributed/pipeline/sync/test_worker",
     "distributed/elastic/agent/server/test/api_test",
     "distributed/elastic/multiprocessing/api_test",
+    "distributed/_shard/checkpoint/test_checkpoint"
+    "distributed/_shard/checkpoint/test_file_system_checkpoint"
     "distributed/_shard/sharding_spec/test_sharding_spec",
     "distributed/_shard/sharding_plan/test_sharding_plan",
     "distributed/_shard/sharded_tensor/test_megatron_prototype",
@@ -217,7 +219,6 @@ def skip_test_p(name: str) -> bool:
     "distributed/_shard/sharded_tensor/ops/test_matrix_ops",
     "distributed/_shard/sharded_tensor/ops/test_softmax",
     "distributed/_shard/sharded_tensor/ops/test_tensor_ops",
-    "distributed/_shard/sharding_spec/test_sharding_spec",
     "distributed/_shard/sharded_optim/test_sharded_optim",
     "distributed/_shard/test_partial_tensor",
     "distributed/_shard/test_replicated_tensor",
@@ -228,6 +229,8 @@ def skip_test_p(name: str) -> bool:
     "distributed/rpc/test_faulty_agent",
     "distributed/rpc/test_tensorpipe_agent",
     "distributed/rpc/cuda/test_tensorpipe_agent",
+    "distributed/_shard/checkpoint/test_checkpoint"
+    "distributed/_shard/checkpoint/test_file_system_checkpoint"
     "distributed/_shard/sharding_spec/test_sharding_spec",
     "distributed/_shard/sharding_plan/test_sharding_plan",
     "distributed/_shard/sharded_tensor/test_megatron_prototype",
@@ -244,7 +247,6 @@ def skip_test_p(name: str) -> bool:
     "distributed/_shard/sharded_tensor/ops/test_matrix_ops",
     "distributed/_shard/sharded_tensor/ops/test_softmax",
     "distributed/_shard/sharded_tensor/ops/test_tensor_ops",
-    "distributed/_shard/sharding_spec/test_sharding_spec",
     "distributed/_shard/sharded_optim/test_sharded_optim",
     "distributed/_shard/test_partial_tensor",
     "distributed/_shard/test_replicated_tensor",