[SymmMem] Better tuning of A2AV based on accurate node boundary#162003
Closed
kwen2501 wants to merge 3 commits intogh/kwen2501/222/basefrom
Closed
[SymmMem] Better tuning of A2AV based on accurate node boundary#162003kwen2501 wants to merge 3 commits intogh/kwen2501/222/basefrom
kwen2501 wants to merge 3 commits intogh/kwen2501/222/basefrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/162003
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 1 PendingAs of commit 5396dfd with merge base 7a83cf4 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Sep 2, 2025
Skylion007
reviewed
Sep 3, 2025
ngimel
approved these changes
Sep 9, 2025
| // Check user setting first | ||
| int num_blocks = c10d::symmetric_memory::getenv_nblocks(); | ||
| if (num_blocks > 0) { // set by user | ||
| return num_blocks; |
Collaborator
There was a problem hiding this comment.
I think you want to treat it as max_blocks, and not just return
Collaborator
Author
|
@pytorchbot merge |
Collaborator
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 |
fduwjj
approved these changes
Sep 9, 2025
| return num_blocks; | ||
| } | ||
|
|
||
| } // namespace c10d::symmetric_memory No newline at end of file |
Contributor
There was a problem hiding this comment.
you want to make linter happy here.
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…rch#162003) Use `world_within_direct_access()` to distinguish intra- vs inter- node. Previously we assumed a fixed node size of 8, which is not true for NVL72. Also added env var `TORCH_SYMMMEM_NBLOCKS` for control. Pull Request resolved: pytorch#162003 Approved by: https://github.com/ngimel, https://github.com/fduwjj
mansiag05
pushed a commit
to mansiag05/pytorch
that referenced
this pull request
Sep 22, 2025
…rch#162003) Use `world_within_direct_access()` to distinguish intra- vs inter- node. Previously we assumed a fixed node size of 8, which is not true for NVL72. Also added env var `TORCH_SYMMMEM_NBLOCKS` for control. Pull Request resolved: pytorch#162003 Approved by: https://github.com/ngimel, https://github.com/fduwjj
cleonard530
pushed a commit
to cleonard530/pytorch
that referenced
this pull request
Sep 22, 2025
…rch#162003) Use `world_within_direct_access()` to distinguish intra- vs inter- node. Previously we assumed a fixed node size of 8, which is not true for NVL72. Also added env var `TORCH_SYMMMEM_NBLOCKS` for control. Pull Request resolved: pytorch#162003 Approved by: https://github.com/ngimel, https://github.com/fduwjj
dsashidh
pushed a commit
to dsashidh/pytorch
that referenced
this pull request
Sep 26, 2025
…rch#162003) Use `world_within_direct_access()` to distinguish intra- vs inter- node. Previously we assumed a fixed node size of 8, which is not true for NVL72. Also added env var `TORCH_SYMMMEM_NBLOCKS` for control. Pull Request resolved: pytorch#162003 Approved by: https://github.com/ngimel, https://github.com/fduwjj
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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):
Use
world_within_direct_access()to distinguish intra- vs inter- node.Previously we assumed a fixed node size of 8, which is not true for NVL72.
Also added env var
TORCH_SYMMMEM_NBLOCKSfor control.cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta @ezyang @msaroufim