Refactor CUDAAllocatorConfig to reuse AcceleratorAllocatorConfig #150312
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/150312
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: ✅ You can merge normally! (3 Unrelated Failures)As of commit df9befb with merge base bb67660 ( UNSTABLE - The following jobs are marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Mar 31, 2025
yangw-dev
pushed a commit
that referenced
this pull request
Aug 1, 2025
…fig (#150312)" This reverts commit dfacf11. Reverted #150312 on behalf of https://github.com/guangyey due to Static initialization order issue impact the downstream repo ([comment](#150312 (comment)))
Collaborator
|
Starting merge as part of PR stack under #156175 |
3 similar comments
Collaborator
|
Starting merge as part of PR stack under #156175 |
Collaborator
|
Starting merge as part of PR stack under #156175 |
<
8000
!-- no margin wins, so we check it last and use its value if true. -->
Collaborator
|
Starting merge as part of PR stack under #156175 |
pytorchmergebot
pushed a commit
that referenced
this pull request
Aug 5, 2025
# Motivation As @ScottTodd identified in this [comment](#150312 (comment)), using STL containers like `std::string` and `std::unordered_set` at static init time can cause static initialization order issues. This PR is based on and modified from his original PR: #159607. I’m stacking this PR here to help facilitate the landing and validation process. Co-authored-by: @ScottTodd Pull Request resolved: #159629 Approved by: https://github.com/ScottTodd, https://github.com/albanD
pytorchmergebot
pushed a commit
that referenced
this pull request
Aug 5, 2025
…llocatorConfig instead (#156165) Pull Request resolved: #156165 Approved by: https://github.com/albanD ghstack dependencies: #159629, #150312
pytorchmergebot
pushed a commit
that referenced
this pull request
Aug 5, 2025
# Motivation This PR moves the implementation of `torch.cuda.memory._set_allocator_settings` to `torch._C._accelerator_setAllocatorSettings`. Since the original API was intended as a temporary/internal utility, I am not exposing the new function as a public API. Pull Request resolved: #156175 Approved by: https://github.com/albanD ghstack dependencies: #159629, #150312, #156165
joshuuuasu
added a commit
to joshuuuasu/pytorch
that referenced
this pull request
Aug 19, 2025
…onfig (pytorch#150312)" Summary: reverting this diff since it caused S551328. Please see D80217492 for dertails. Test Plan: NA Rollback Plan: Differential Revision: D80553588
joshuuuasu
added a commit
to joshuuuasu/pytorch
that referenced
this pull request
Aug 20, 2025
…onfig (pytorch#150312)" (pytorch#161002) Summary: Pull Request resolved: pytorch#161002 reverting this diff since it caused S551328. Please see D80217492 for dertails. Test Plan: NA Rollback Plan: Reviewed By: sayitmemory, jingsh Differential Revision: D80553588
pytorchmergebot
pushed a commit
that referenced
this pull request
Aug 26, 2025
…onfig (#150312)" (#161002) Summary: reverting this diff since it caused S551328. Please see D80217492 for dertails. Test Plan: NA Rollback Plan: Differential Revision: D80553588 Pull Request resolved: #161002 Approved by: https://github.com/jingsh, https://github.com/izaitsevfb
pytorchmergebot
added a commit
that referenced
this pull request
Aug 27, 2025
…locatorConfig (#150312)" (#161002)" This reverts commit a03cc53. Reverted #161002 on behalf of https://github.com/guangyey due to This PR breaks CI TestCudaMallocAsync::test_allocator_settings ([comment](#161002 (comment)))
guangyey
added a commit
that referenced
this pull request
Aug 27, 2025
guangyey
added a commit
that referenced
this pull request
Aug 27, 2025
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
# Motivation As @ScottTodd identified in this [comment](pytorch#150312 (comment)), using STL containers like `std::string` and `std::unordered_set` at static init time can cause static initialization order issues. This PR is based on and modified from his original PR: pytorch#159607. I’m stacking this PR here to help facilitate the landing and validation process. Co-authored-by: @ScottTodd Pull Request resolved: pytorch#159629 Approved by: https://github.com/ScottTodd, https://github.com/albanD
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…orch#150312) # Motivation Refactor `CUDAAllocatorConfig` to reuse `AcceleratorAllocatorConfig` and `ConfigTokenizer`. We would deprecate those option that overleap with `AcceleratorAllocatorConfig` in the following PR and keep them only for BC. Pull Request resolved: pytorch#150312 Approved by: https://github.com/albanD ghstack dependencies: pytorch#159629
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…llocatorConfig instead (pytorch#156165) Pull Request resolved: pytorch#156165 Approved by: https://github.com/albanD ghstack dependencies: pytorch#159629, pytorch#150312
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…6175) # Motivation This PR moves the implementation of `torch.cuda.memory._set_allocator_settings` to `torch._C._accelerator_setAllocatorSettings`. Since the original API was intended as a tempora 5D99 ry/internal utility, I am not exposing the new function as a public API. Pull Request resolved: pytorch#156175 Approved by: https://github.com/albanD ghstack dependencies: pytorch#159629, pytorch#150312, pytorch#156165
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…onfig (pytorch#150312)" (pytorch#161002) Summary: reverting this diff since it caused S551328. Please see D80217492 for dertails. Test Plan: NA Rollback Plan: Differential Revision: D80553588 Pull Request resolved: pytorch#161002 Approved by: https://github.com/jingsh, https://github.com/izaitsevfb
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…locatorConfig (pytorch#150312)" (pytorch#161002)" This reverts commit a03cc53. Reverted pytorch#161002 on behalf of https://github.com/guangyey due to This PR breaks CI TestCudaMallocAsync::test_allocator_settings ([comment](pytorch#161002 (comment)))
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…fig (pytorch#150312)" (pytorch#161628) This reverts commit ae1a706. Pull Request resolved: pytorch#161628 Approved by: https://github.com/atalman ghstack dependencies: pytorch#161625, pytorch#161626, pytorch#161627
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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):
Motivation
Refactor
CUDAAllocatorConfigto reuseAcceleratorAllocatorConfigandConfigTokenizer. We would deprecate those option that overleap withAcceleratorAllocatorConfigin the following PR and keep them only for BC.cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k