Skip to content

[Sim] Improve cleanup logic in ProceduralizeSim#10469

Merged
fzi-hielscher merged 1 commit into
llvm:mainfrom
fzi-hielscher:sparse-scc-simproc
May 15, 2026
Merged

[Sim] Improve cleanup logic in ProceduralizeSim#10469
fzi-hielscher merged 1 commit into
llvm:mainfrom
fzi-hielscher:sparse-scc-simproc

Conversation

@fzi-hielscher
Copy link
Copy Markdown
Contributor

Replace the convergence loop for pruning leftover operations with a reverse-topologically-ordered erasure. Emit a warning for operations that are not dead outside of the TriggeredOp.

CC @nanjo712

Replace the convergence loop for pruning leftover operations with a
reverse-topologically-ordered erasure. Emit a warning for operations
that are not dead outside of the TriggeredOp.
@circt-bot
Copy link
Copy Markdown

circt-bot Bot commented May 14, 2026

Results of circt-tests run for 36aa670 compared to results for 20e590b: no change to test results.

@nanjo712
Copy link
Copy Markdown
Contributor

Thanks for the CC! I took a look, and this looks reasonable to me.

Using the SparseOpSCC-based reverse-topological cleanup seems cleaner than the previous convergence loop, especially for pruning the formatting DAG after proceduralization.

I don’t have any concerns from my side.

@fzi-hielscher fzi-hielscher merged commit bc8f154 into llvm:main May 15, 2026
6 checks passed
@fzi-hielscher fzi-hielscher deleted the sparse-scc-simproc branch May 15, 2026 21:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants