[MIGraphX EP] Bugfix for toVector Function causing rollover with int64 inputs #26403
+6
−5
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
…4_t inputs * Bug Fix: large integers incorrectly converted to -1 in IsUnsupportedOpMode() function - prevent casting of int64 to int32 in IsUnsupportedOpMode() function for "Slice" operator * remove testing code * refactor to use toVector() function
Contributor
Author
|
Used to fix models in benchmarking in ROCm - ROCm#188 |
Contributor
Author
|
cc @devang-ml @tianleiwu @cloudhan can we get this looked at/reviewed? This is a bit of an insidious bug since it will clamp addresses on the slice operator due to the casting done in the toVector call. This won't break but will cause roll overs which will produce undefined behavior in our slice operator on the MIGraphX side. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
Refactors the toVector function which causes narrowing of the datatype for vector inputs such as in the slice operator
Description
Removes the casting done for toVector as that causes rollovers due to some Onnx Operators using int64_t inputs rather than int32_t. This case causes us to actually incorrectly represent axis, start and end values in the slice operator
Motivation and Context
Fixes a bug since this causes some of our Onnx operators to not operate in the full correct data range for input vectors