[SymmMem] Barrier on team instead of world #163298
Closed
+107
−72
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/163298
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit ce9d19b with merge base 4840a1a ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
ciflow/h100-symm-mem
oncall: distributed
This was referenced Sep 18, 2025
fegin
approved these changes
Sep 19, 2025
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
|
Starting merge as part of PR stack under #162680 |
pytorchmergebot
pushed a commit
that referenced
this pull request
Sep 19, 2025
Problem: Without MemPool it looks like nvshmem backend never deallocates memory. Cause: Handles in `symm_mems_` (a map) keeps reference to memory allocations. Solution: - Remove reference to allocation from handles -- the reference is never used anyway. - Use `unique_ptr` instead of `shared_ptr` to wrap allocation to ensure single ownership. Pull Request resolved: #162680 Approved by: https://github.com/ezyang ghstack dependencies: #163298
|
@pytorchbot cherry-pick --onto release/2.9 -c critical |
pytorchbot
pushed a commit
that referenced
this pull request
Sep 19, 2025
Problem: Without MemPool it looks like nvshmem backend never deallocates memory. Cause: Handles in `symm_mems_` (a map) keeps reference to memory allocations. Solution: - Remove reference to allocation from handles -- the reference is never used anyway. - Use `unique_ptr` instead of `shared_ptr` to wrap allocation to ensure single ownership. Pull Request resolved: #162680 Approved by: https://github.com/ezyang ghstack dependencies: #163298 (cherry picked from commit 7130b17)
pytorchbot
pushed a commit
that referenced
this pull request
Sep 19, 2025
As titled. Avoiding a potential hang when running dispatch and combine in subgroups. The rest is just re-arrange of the tests to create a sub-group test class. (no substantial change) Pull Request resolved: #163298 Approved by: https://github.com/fegin (cherry picked from commit f8fb437)
Cherry picking #163298The cherry pick PR is at #163376 and it is recommended to link a critical cherry pick PR with an issue. The following tracker issues are updated: Details for Dev Infra teamRaised by workflow job |
mansiag05
pushed a commit
to mansiag05/pytorch
that referenced
this pull request
Sep 22, 2025
As titled. Avoiding a potential hang when running dispatch and combine in subgroups. The rest is just re-arrange of the tests to create a sub-group test class. (no substantial change) Pull Request resolved: pytorch#163298 Approved by: https://github.com/fegin
mansiag05
pushed a commit
to mansiag05/pytorch
that referenced
this pull request
Sep 22, 2025
Problem: Without MemPool it looks like nvshmem backend never deallocates memory. Cause: Handles in `symm_mems_` (a map) keeps reference to memory allocations. Solution: - Remove reference to allocation from handles -- the reference is never used anyway. - Use `unique_ptr` instead of `shared_ptr` to wrap allocation to ensure single ownership. Pull Request resolved: pytorch#162680 Approved by: https://github.com/ezyang ghstack dependencies: pytorch#163298
cleonard530
pushed a commit
to cleonard530/pytorch
that referenced
this pull request
Sep 22, 2025
As titled. Avoiding a potential hang when running dispatch and combine in subgroups. The rest is just re-arrange of the tests to create a sub-group test class. (no substantial change) Pull Request resolved: pytorch#163298 Approved by: https://github.com/fegin
cleonard530
pushed a commit
to cleonard530/pytorch
that referenced
this pull request
Sep 22, 2025
Problem: Without MemPool it looks like nvshmem backend never deallocates memory. Cause: Handles in `symm_mems_` (a map) keeps reference to memory allocations. Solution: - Remove reference to allocation from handles -- the reference is never used anyway. - Use `unique_ptr` instead of `shared_ptr` to wrap allocation to ensure single ownership. Pull Request resolved: pytorch#162680 Approved by: https://github.com/ezyang ghstack dependencies: pytorch#163298
dsashidh
pushed a commit
to dsashidh/pytorch
that referenced
this pull request
Sep 26, 2025
As titled. Avoiding a potential hang when running dispatch and combine in subgroups. The rest is just re-arrange of the tests to create a sub-group test class. (no substantial change) Pull Request resolved: pytorch#163298 Approved by: https://github.com/fegin
dsashidh
pushed a commit
to dsashidh/pytorch
that referenced
this pull request
Sep 26, 2025
Problem: Without MemPool it looks like nvshmem backend never deallocates memory. Cause: Handles in `symm_mems_` (a map) keeps reference to memory allocations. Solution: - Remove reference to allocation from handles -- the reference is never used anyway. - Use `unique_ptr` instead of `shared_ptr` to wrap allocation to ensure single ownership. Pull Request resolved: pytorch#162680 Approved by: https://github.com/ezyang ghstack dependencies: pytorch#163298
Camyll
pushed a commit
that referenced
this pull request
Sep 26, 2025
[SymmMem] Fix memory allocation hold-up (#162680) Problem: Without MemPool it looks like nvshmem backend never deallocates memory. Cause: Handles in `symm_mems_` (a map) keeps reference to memory allocations. Solution: - Remove reference to allocation from handles -- the reference is never used anyway. - Use `unique_ptr` instead of `shared_ptr` to wrap allocation to ensure single ownership. Pull Request resolved: #162680 Approved by: https://github.com/ezyang ghstack dependencies: #163298 (cherry picked from commit 7130b17) Co-authored-by: Ke Wen <kw2501@meta.com>
Camyll
pushed a commit
that referenced
this pull request
Sep 26, 2025
[SymmMem] Barrier on team instead of world (#163298) As titled. Avoiding a potential hang when running dispatch and combine in subgroups. The rest is just re-arrange of the tests to create a sub-group test class. (no substantial change) Pull Request resolved: #163298 Approved by: https://github.com/fegin (cherry picked from commit f8fb437) Co-authored-by: Ke Wen <kw2501@meta.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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):
num_active_allocationsand dealloc checks #162681As titled. Avoiding a potential hang when running dispatch and combine in subgroups.
The rest is just re-arrange of the tests to create a sub-group test class. (no substantial change)
cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta @ezyang @msaroufim @dcci