[inductor] dont reuse buffers if it affects peak (#145883) #159530
+615
−2
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/159530
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 40 PendingAs of commit 8aa57d3 with merge base 2507ae6 ( UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
v0i0
added a commit
that referenced
this pull request
Jul 30, 2025
|
@pytorchbot label "topic: not user facing" |
eellison
reviewed
Jul 31, 2025
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Jul 31, 2025
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Jul 31, 2025
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Aug 2, 2025
v0i0
added a commit
that referenced
this pull request
Aug 5, 2025
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Aug 7, 2025
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Aug 7, 2025
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Aug 7, 2025
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Aug 7, 2025
eellison
reviewed
Aug 11, 2025
|
|
||
| # Initialize tree and lazy arrays | ||
| self.tree = [identity_element] * self.size | ||
| self.lazy: list[Optional[T]] = [None] * self.size |
There was a problem hiding this comment.
nit: describe what the lazy array will do ?
| self._build(values, right_child, mid + 1, end) | ||
|
|
||
| # Update current node with summary of children | ||
| self.tree[node] = self.summary_op(self.tree[left_child], self.tree[right_child]) |
There was a problem hiding this comment.
Nice to use the identity element instead of having to handle special cases.
There was a problem hiding this comment.
oh i meant, i think this is what the code is already doing, right.
|
Dashboard run here: https://github.com/pytorch/pytorch/actions/runs/16895400336 |
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben [ghstack-poisoned]
v0i0
added a commit
that referenced
this pull request
Aug 12, 2025
eellison
reviewed
Aug 12, 2025
torch/_inductor/codegen/wrapper.py
Outdated
|
|
||
| self.overall_peak_memory, peak_by_scheduler_node = estimate_peak_memory( | ||
| V.graph.scheduler.nodes, | ||
| {}, |
There was a problem hiding this comment.
Sorry, one last thing, names_to_freeable_bufs is important for the backward when we need to know which activations will deallocate in order to have an accurate memory estimate
v0i0
added a commit
that referenced
this pull request
Aug 12, 2025
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):
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben