[pytorch/profiler] Profiler NCCL metadata can now contain collective Input and Ouput Tensor addrs #139837
Conversation
…nd addr hints for Tensor values. Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead on custom silicon. Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/139837
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ❌ 2 New Failures, 1 Unrelated FailureAs of commit b3e850c with merge base ffb9790 ( NEW FAILURES - The following jobs have failed:
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. |
| This pull request was exported from Phabricator. Differential Revision: D64413699 |
| @pytorchbot label "topic: not user facing" |
…start and end addr hints for Tensor values." Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead on custom silicon. Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/) cc robieta chaekit guotuofeng guyang3532 dzhulgakov davidberard98 briancoutinho sraikund16 [ghstack-poisoned]
…nd addr hints for Tensor values. Pull Request resolved: #139837 Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead on custom silicon. ghstack-source-id: 252025337 @exported-using-ghexport Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/)
| This pull request was exported from Phabricator. Differential Revision: D64413699 |
sanrise changed the title 
sraikund16 removed the oncall: profiler …start and end addr hints for Tensor values." Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead. Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/) cc robieta chaekit guotuofeng guyang3532 dzhulgakov davidberard98 briancoutinho sraikund16 [ghstack-poisoned]
| This pull request was exported from Phabricator. Differential Revision: D64413699 |
…start and end addr hints for Tensor values." Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead. Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/) cc robieta chaekit guotuofeng guyang3532 dzhulgakov davidberard98 briancoutinho sraikund16 [ghstack-poisoned]
| This pull request was exported from Phabricator. Differential Revision: D64413699 |
sanrise changed the title …collective Input and Ouput Tensor addrs" Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead. Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/) cc robieta chaekit guotuofeng guyang3532 dzhulgakov davidberard98 briancoutinho sraikund16 [ghstack-poisoned]
| This pull request was exported from Phabricator. Differential Revision: D64413699 |
…collective Input and Ouput Tensor addrs" Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead. Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/) cc robieta chaekit guotuofeng guyang3532 dzhulgakov davidberard98 briancoutinho sraikund16 [ghstack-poisoned]
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
14 similar comments
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
| @pytorchbot merge -i (Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally) |
Merge failedReason: This PR has internal changes and must be landed via Phabricator! Please try reimporting/rexporting the PR! Details for Dev Infra teamRaised by workflow job |
| This diff was reverted due to a LOG statement, I am grafting a fix and will submit a new PR. |
…Input and Ouput Tensor addrs (pytorch#139837) Studying memory access patterns is the primary use cases. Internal: The data may be used to find the % of operators that may cause alignment related overhead. Differential Revision: [D64413699](https://our.internmc.facebook.com/intern/diff/D64413699/) Pull Request resolved: pytorch#139837 Approved by: https://github.com/sraikund16
…lective Input and Ouput Tensor addrs (pytorch#139837)" This reverts commit 3e277eb. Reverted pytorch#139837 on behalf of https://github.com/facebook-github-bot due to Diff reverted internally ([comment](pytorch#139837 (comment)))
Stack from ghstack (oldest at bottom):
Studying memory access patterns is the primary use cases.
Internal: The data may be used to find the % of operators that may cause alignment related overhead.
Differential Revision: D64413699
cc @robieta @chaekit @guotuofeng @guyang3532 @dzhulgakov @davidberard98 @briancoutinho @sraikund16