functionalization: remove some unnecessary view_copies in inplace views#77713
Closed
bdhirsh wants to merge 17 commits intogh/bdhirsh/235/basefrom
Closed
functionalization: remove some unnecessary view_copies in inplace views#77713bdhirsh wants to merge 17 commits intogh/bdhirsh/235/basefrom
bdhirsh wants to merge 17 commits intogh/bdhirsh/235/basefrom
Conversation
[ghstack-poisoned]
Contributor
🔗 Helpful links
✅ No Failures (0 Pending)As of commit fe9268e (more details on the Dr. CI page): Expand to see more💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Please report bugs/suggestions to the (internal) Dr. CI Users group. |
This was referenced May 18, 2022
…inplace views" [ghstack-poisoned]
…inplace views" [ghstack-poisoned]
…inplace views" [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
ezyang
approved these changes
May 18, 2022
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
Collaborator
Author
|
@pytorchbot merge |
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
Collaborator
|
Merge failed due to Matched rule superuser, but PR #77132 has not been reviewed yet |
…inplace views" The original implementation of inplace_view ops was a bit inefficient, because we would re-play the **entire** stack of views off of the base. We only actually have to run the newly added view meta. I mostly noticed this because `resize_()` can act as an inplace_view op, which I handle in the next PR. [ghstack-poisoned]
facebook-github-bot
pushed a commit
that referenced
this pull request
May 31, 2022
…ws (#77713) Summary: Pull Request resolved: #77713 Approved by: https://github.com/ezyang Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/e9c54ae1c2357651b3377f0cbbdf6d1ac649f6d1 Reviewed By: seemethere Differential Revision: D36783117 Pulled By: bdhirsh fbshipit-source-id: 138b2fe0475e9d0046afb9e85624208570bf9dd8
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
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.
The original implementation of inplace_view ops was a bit inefficient, because we would re-play the entire stack of views off of the base. We only actually have to run the newly added view meta.
I mostly noticed this because
resize_()can act as an inplace_view op, which I handle in the next PR.Stack from ghstack: