fix(core): Add MSVC patch for 64-bit integer comparison #102327
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
| @@ -0,0 +1,35 @@ | |||
| // Platform-specific fix for MSVC 64-bit integer comparison bug in minimum/maximum functors. | |||
There was a problem hiding this comment.
Start with the license header
Comment on lines
+13
to
+33
| #if defined(_MSC_VER) | ||
| // Specialize maximum for int64_t on MSVC to avoid 32-bit truncation. | ||
| template <> | ||
| struct maximum<int64_t> { | ||
| EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int64_t operator()(const int64_t& x, const int64_t& y) const { | ||
| // Use explicit cast to ensure 64-bit comparison | ||
| return (static_cast<uint64_t>(x) > static_cast<uint64_t>(y)) ? x : y; | ||
| } | ||
| }; | ||
|
|
||
| template <> | ||
| struct minimum<int64_t> { | ||
| EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int64_t operator()(const int64_t& x, const int64_t& y) const { | ||
| // Use explicit cast to ensure 64-bit comparison | ||
| return (static_cast<uint64_t>(x) < static_cast<uint64_t>(y)) ? x : y; | ||
| } | ||
| }; | ||
| #endif | ||
|
|
||
| } // namespace functor | ||
| } // namespace tensorflow |
There was a problem hiding this comment.
Why not add these directly to cwise_ops.h?
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Outdated suggestions cannot be applied.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.
This PR resolves a platform-specific bug where
int64_tcomparisons could produce incorrect results on MSVC (Windows builds) due to potential 32-bit truncation.The Problem: Operations like
tf.minimumandtf.maximumwere not behaving correctly for 64-bit integers in some cases on Windows.The Fix: This change introduces a patch that specializes the
minimumandmaximumfunctors forint64_ton MSVC, ensuring a true 64-bit comparison is always performed. This improves numerical stability and correctness on Windows.