[hardware] 🐛 Fix false illegal on unaligned widening reduction - #478
Open
emiliengnr wants to merge 1 commit into
Open
[hardware] 🐛 Fix false illegal on unaligned widening reduction#478emiliengnr wants to merge 1 commit into
emiliengnr 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Hello, here is a pull request for a bug I found.
Problem
A widening reduction
vwredsum.vs/vwredsumu.vswhose destination is not alignedto the widened LMUL is legal RVV, but ara's dispatcher rejects it as illegal.
The register-alignment check keys on
ara_req.emul(the widened LMUL) and appliesit to vs1, vd, and vs2. But a widening reduction's vd and vs1 are single 2*SEW
elements (EMUL=1, no group-alignment constraint), and vs2 is the source vector at
the original
vlmul. So a legal unaligned-vd reduction tripsillegal_insn, theop is never issued, no completion returns, and the hart spins on the accelerator
handshake until reset.
Fix
For
VWREDSUM/VWREDSUMU, check only vs2 against the originalvlmul; drop thealignment check on the single-element vd and vs1. Other ops are unchanged.
Changelog
Fixed
Checklist