Fix RTL thresholding runtime weights address space mismatch - #1685
Open
auphelia wants to merge 1 commit into
Open
Fix RTL thresholding runtime weights address space mismatch#1685auphelia wants to merge 1 commit into
auphelia wants to merge 1 commit into
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes runtime writeable weights for RTL thresholding when numSteps < 2^outputBits (e.g., TFC w2a2 with 2 thresholds and INT2 output).
Problem: make_weight_file() calculated width_padded using roundup(n_steps, 2^output_bits), but RTL allocates address space using clog2(numSteps). This caused a 2x address overflow for configurations like TFC w2a2 (numSteps=2, output_bits=2), where Python generated 4 entries per channel but RTL only had space for 2.
Fix: Changed width_padded calculation to 2^clog2(numSteps) to match RTL address allocation.
Test: Added test_runtime_thresholds_tfc_like covering TFC-like configurations (ch=16/64/784, pe=4/8/49, numSteps=2/3).