Correct narrowing cast of inputs to threshold datatype.#1553
Merged
Conversation
Collaborator
|
The issue that was revealed by the unit test was addressed in the compiler code. If the MinimizeWeightBitWidth() transform results in WT < WI, we ensure that WT gets expanded to retain the ability to represent all values in [thresh_min-1:thresh_max] to be compatible with the comparison scheme that is done in the RTL Thresholding layer. |
…d values are zero
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
After refactoring RoundAndClipThresholds, we observed cases where the threshold datatype was minimized below the input datatype when the threshold values themselves did not require a larger type. This exposed a bug in the RTL thresholding implementation.
A temporary compiler-side workaround was added to prevent the threshold datatype from being smaller than the input datatype (a67009c).
This PR fixes the issue in the RTL code, allowing the original optimization (threshold dtype < input dtype) to be safely re-enabled.