Fix: transactional translation validation insertion. #107523
Closed
Conversation
This file contains 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
This PR fixes transactional behavior of translation validation insertion. Previously, this transactional behavior was implemented by removing the FX node if any issues occurred until the end of `evaluate_expr`. However, since we cache FX nodes, we might end up removing something that wasn't inserted in the same function call. **Solution:** when creating an FX node for `call_function`, we also return whether this is a fresh FX node or not. Then, we can appropriately handle each case. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/107523
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 4ca97ff with merge base 36141de ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
ezyang
approved these changes
Aug 21, 2023
pytorchmergebot
pushed a commit
that referenced
this pull request
Aug 22, 2023
This PR stops `SymNode` from mutating (i.e. simplifying) its expression. Instead, the
simplification (without mutation) is deferred to the `SymNode.maybe_as_int` method.
```python
- FakeTensor(size=(s0,), ...)
- FakeTensor(size=(s1, s2, s3), ...)
- Eq(s0, s1 + s2 + s3)
- FakeTensor(size=(s0,), ...)
- FakeTensor(size=(s1, s2, s3), ...)
```
In summary, this PR:
- Replaces `SymNode._expr` by `SymNode.expr`, removing the old property function
- This makes it so `SymNode` instances never update their expression
- Creates `SymNode.simplified_expr()` method for actually calling `ShapeEnv.replace` on
its expression. Note that this doesn't updates `SymNode.expr`
- Changes how `tensor.size()` gets converted to its Python `torch.Size` type
- Instead of calling `SymInt::maybe_as_int()` method, we create a new
`SymInt::is_symbolic()` method for checking whether it is actually a symbolic value
- This is needed so that when we call `tensor.size()` in the Python side, the returned
sequence is faithful to the actual data, instead of possibly simplifying it and
returning an integer
- 2 files needs this modification:
- _torch/csrc/Size.cpp_: for handling `torch.Tensor.size` Python calls
- _torch/csrc/utils/pybind.cpp_: for handling `symint.cast()` C++ calls
Pull Request resolved: #107492
Approved by: https://github.com/ezyang
ghstack dependencies: #107523
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):
ValidationException. #107493SymNodeexpression. #107492This PR fixes transactional behavior of translation validation insertion.
Previously, this transactional behavior was implemented by removing the FX node if any
issues occurred until the end of
evaluate_expr. However, since we cache FX nodes, wemight end up removing something that wasn't inserted in the same function call.
Solution: when creating an FX node for
call_function, we also return whether this isa fresh FX node or not. Then, we can appropriately handle each case.