Upgrade distributed test to g4dn instances (T4 GPUs) #137161
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/137161
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit f6f6620 with merge base a1899b5 ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
[ghstack-poisoned]
seemethere
approved these changes
Oct 9, 2024
There was a problem hiding this comment.
LGTM, nit: since these nodes are 2x as expensive it might be good to move the experimental split build duplicate tests to periodic.yml.
WDYT @PaliC
[ghstack-poisoned]
kwen2501
changed the title
kwen2501
added
the
keep-going
[ghstack-poisoned]
|
@pytorchbot merge -f "Previously tested; just a rebase to swap stack order to unblock the other PR" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
pytorchmergebot
pushed a commit
that referenced
this pull request
Oct 10, 2024
This PR contains multiple fixes for issue #135279: ## First part: Moves the GPU guard (`cudaSetDevice`) before the `currentStreamCaptureStatusMayInitCtx` call. As its name suggests, it May Init Ctx. ## Second part: Even with the above fix, additional contexts are still observed during Work object destruction, e.g. ``` work = dist.all_reduce(tensor, async_op=True) time.sleep(5) <-- no additional context yet del work <-- additional context shows up ``` ### Debug process Chasing it down to destruction of a `Future` object -- a member variable of `Work`. Then further down to the following member of `Future`: ``` std::vector<c10::Event> events_; ``` When the `events_` are destroyed, we hit the road down to: https://github.com/pytorch/pytorch/blob/1f3a79379012b408e0375e81fe9205dcba5e34ba/c10/cuda/impl/CUDAGuardImpl.h#L106-L121 When there is no "preset" CUDA context (**which is the case for python garbage collector**), line 112: `c10::cuda::GetDevice(&orig_device)` will set `orig_device` to 0. Then, at line 120, `c10::cuda::SetDevice(orig_device)` will "officially" set the context to device 0 -- **that's where rank 1, 2, ... can create extra context on device 0!** ### Solution This PR adds an explicit destructor to `Future`. In this destructor, destroy each event with a device guard. ## Test Added test_extra_cuda_context, implemented via - `pynvml` (if available), or - memory consumption check. `python test/distributed/test_c10d_nccl.py -k test_extra_cuda_context` Pull Request resolved: #135273 Approved by: https://github.com/fduwjj, https://github.com/wconstab, https://github.com/eqy ghstack dependencies: #137161
|
@pytorchbot revert -m "broken tests on trunk" -c "nosignal" |
|
@pytorchbot successfully started a revert job. Check the current status here. |
pytorchmergebot
added a commit
that referenced
this pull request
Oct 10, 2024
This reverts commit cdd8fa9. Reverted #135273 on behalf of https://github.com/PaliC due to broken tests on trunk ([comment](#137161 (comment)))
pytorchmergebot
added a commit
that referenced
this pull request
Oct 10, 2024
This reverts commit b6a64dc. Reverted #137161 on behalf of https://github.com/PaliC due to broken tests on trunk ([comment](#137161 (comment)))
|
@kwen2501 your PR has been successfully reverted. |
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…dering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
cc XilunWu H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…ompute-comm reordering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…dering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
cc XilunWu H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…ompute-comm reordering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…dering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
cc XilunWu H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…ompute-comm reordering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…dering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
cc XilunWu H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…ompute-comm reordering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
yf225
added a commit
that referenced
this pull request
Oct 20, 2024
…dering tests to test_inductor_distributed" `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
pytorchmergebot
pushed a commit
that referenced
this pull request
Oct 20, 2024
…s to test_inductor_distributed (#138178) `test_replicate_with_compiler.py` and `test_fully_shard_compile.py` requires bf16, so needs to be run within test_inductor_distributed job (which uses A10G (SM80) and has bf16 support). This allows us to migrate distributed jobs to T4 machines in #137161, as the compiled distributed jobs are the only blocking ones now. Pull Request resolved: #138178 Approved by: https://github.com/xmfan, https://github.com/fduwjj, https://github.com/fegin, https://github.com/kwen2501
cc XilunWu H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
|
@pytorchbot merge -f "unrelated failure" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
pytorchmergebot
pushed a commit
that referenced
this pull request
Oct 21, 2024
This PR contains multiple fixes for issue #135279: ## First part: Moves the GPU guard (`cudaSetDevice`) before the `currentStreamCaptureStatusMayInitCtx` call. As its name suggests, it May Init Ctx. ## Second part: Even with the above fix, additional contexts are still observed during Work object destruction, e.g. ``` work = dist.all_reduce(tensor, async_op=True) time.sleep(5) <-- no additional context yet del work <-- additional context shows up ``` ### Debug process Chasing it down to destruction of a `Future` object -- a member variable of `Work`. Then further down to the following member of `Future`: ``` std::vector<c10::Event> events_; ``` When the `events_` are destroyed, we hit the road down to: https://github.com/pytorch/pytorch/blob/1f3a79379012b408e0375e81fe9205dcba5e34ba/c10/cuda/impl/CUDAGuardImpl.h#L106-L121 When there is no "preset" CUDA context (**which is the case for python garbage collector**), line 112: `c10::cuda::GetDevice(&orig_device)` will set `orig_device` to 0. Then, at line 120, `c10::cuda::SetDevice(orig_device)` will "officially" set the context to device 0 -- **that's where rank 1, 2, ... can create extra context on device 0!** ### Solution This PR adds an explicit destructor to `Future`. In this destructor, destroy each event with a device guard. ## Test Added test_extra_cuda_context, implemented via - `pynvml` (if available), or - memory consumption check. `python test/distributed/test_c10d_nccl.py -k test_extra_cuda_context` Pull Request resolved: #135273 Approved by: https://github.com/fduwjj, https://github.com/wconstab, https://github.com/eqy ghstack dependencies: #137161 Co-authored-by: Will Feng <yf225@cornell.edu>
pytorchmergebot
pushed a commit
that referenced
this pull request
Oct 25, 2024
Following change in #137161 , bumping world size for some test suites. Pull Request resolved: #138846 Approved by: https://github.com/fduwjj
kwen2501
added a commit
that referenced
this pull request
Oct 25, 2024
Following change in #137161 , bumping world size for some test suites. cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
pytorchmergebot
pushed a commit
that referenced
this pull request
Oct 25, 2024
Following change in #137161 , bumping world size for some test suites. Pull Request resolved: #138846 Approved by: https://github.com/fduwjj
pytorchmergebot
pushed a commit
that referenced
this pull request
Apr 29, 2025
If there aren't any GPUs the WORLD_SIZE would be zero which does not work. So skip those backends completely in that case. Fix after #137161 It might make sense to still run the (CPU-) part of the tests by using something like `world_size = max(3, gpu_count)` or `num_gpus if num_gpus else 3` instead of skipping them all Pull Request resolved: #150764 Approved by: https://github.com/kwen2501
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
async_op=Truecollective if underallow_inflight_collective_as_graph_input_ctx()context manager #137763cc @XilunWu @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @c-p-i-o