[Inductor] short-term fix for needs_fixed_stride_order silent incorrectness #133452
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
…ctness This is a short-term fix for #128084, for the purposes of a 2.4.1. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/133452
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 04a6738 with merge base cd565bc ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Aug 14, 2024
…ctness This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test ghstack-source-id: 8d19392 Pull Request resolved: #133452
eellison
reviewed
Aug 15, 2024
torch/_inductor/lowering.py
Outdated
| new_args = [] | ||
| new_kwargs = {} | ||
| schema_args, schema_kwargs = torch._library.utils.schema_args_kwargs(schema) | ||
| for arg, fx_arg, schema_arg in zip(args, fx_node.args, schema_args): |
There was a problem hiding this comment.
There is no contract that the fx node at this point is using fx_node.args instead of fx_node.kwargs. What I would expect to see here is torch.fx.operator_schemas.normalize_function and then iterate over the schema by argname
There was a problem hiding this comment.
Sounds good. I added this to the list of things we might want to tighten restrictions for (#133250) -- it is the case that after AOTDispatcher (args, kwargs) are normalized, but I guess the inductor passes can do whatever they want
There was a problem hiding this comment.
By the way... the code as previously written assumes that fx_node.{args, kwargs} are zip-able with {args, kwargs}. Are you saying that's incorrect too?
There was a problem hiding this comment.
I updated the code to handle the case where the schema_args/schema_kwargs can be different from args/kwargs. The way constrain_to_fx_strides is called (and from how the code was written previously) it looks like fx_node.{args, kwargs} being zip-able with {args, kwargs} is a reasonable assumption (but if we think that's a bug I can fix it in a follow-up -- I want this PR to go in without changing too much since it needs to be cherry-picked)
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
Closed
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
eellison
approved these changes
Aug 16, 2024
…ent incorrectness" This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test [ghstack-poisoned]
|
@pytorchbot merge -f "unrelated failure" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
zou3519
added a commit
that referenced
this pull request
Aug 19, 2024
…ctness (#133452) This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test Pull Request resolved: #133452 Approved by: https://github.com/eellison
zou3519
added a commit
that referenced
this pull request
Aug 19, 2024
…ctness (#133452) This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test Pull Request resolved: #133452 Approved by: https://github.com/eellison
atalman
pushed a commit
that referenced
this pull request
Aug 21, 2024
…ctness (#133452) (#133888) This is a low-risk short-term fix for #128084, for the purposes of 2.4.1. The actual fix for that issue is more risky and we'll target 2.5. needs_fixed_stride_order is silently incorrect with args that are mutable because it creates clones of those args, writes into them, and doesn't update the original args. This PR makes it so that needs_fixed_stride_order doesn't apply to inputs that are being mutated. This PR doesn't completely fix the problem, but it makes it less incorrect: most of the time the input already has the correct strides but inductor fails to recognize it, and in those cases writing directly to the input is fine. Test Plan: - new test Pull Request resolved: #133452 Approved by: https://github.com/eellison
atalman
added a commit
to atalman/pytorch
that referenced
this pull request
Aug 22, 2024
… incorrectness (pytorch#133452) (pytorch#133888)" This reverts commit 346e0f6.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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):
This is a low-risk short-term fix for
#128084, for the purposes of
2.4.1. The actual fix for that issue is more risky and we'll target 2.5.
needs_fixed_stride_order is silently incorrect with args that are
mutable because it creates clones of those args, writes into them, and
doesn't update the original args.
This PR makes it so that needs_fixed_stride_order doesn't apply to
inputs that are being mutated.
This PR doesn't completely fix the problem, but it makes it less
incorrect: most of the time the input already has the correct strides
but inductor fails to recognize it, and in those cases writing directly
to the input is fine.
Test Plan:
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang