You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Mar 17, 2026. It is now read-only.
Here is a concise summary of the major changes in the LLVM IR diff:
Control Flow Simplification and Branch Target Updates: Multiple br instructions had their destination basic blocks updated (e.g., %66 → %63, %414 → %407) to reflect restructured control flow, often due to loop optimizations or critical edge elimination.
Phi Node Restructuring and Reduction: Several phi nodes were simplified—redundant incoming values removed, unnecessary phi operands eliminated (e.g., replacing multi-operand phis with fewer or constant operands), and some phis moved or split across newly introduced loop exit blocks (e.g., .loopexit, .loopexit132). This reflects improved SSA construction after loop transformations.
Loop Exit and Critical Edge Optimization: New loop-exit blocks (e.g., %63, %._crit_edge, %._crit_edge.i.i.i) were introduced or repurposed, often replacing .critedge or .preheader-style blocks. Phi nodes were updated to feed from these new exits, improving loop-carried dependency modeling.
Redundant Comparison and Logic Elimination: Several patterns involving chained or/and of comparisons (e.g., %or.cond = or i1 %a, %b; br i1 %or.cond) were replaced with direct branching on one operand (e.g., br i1 %a), eliminating unnecessary boolean logic and simplifying control flow.
Floating-Point Computation Reordering and FMA Use: In casadi and image-rs, floating-point sequences were refactored: fdiv + fmul combinations were replaced with fmuladd, and redundant phi-based intermediate values were removed (e.g., merging %.034.lcssa and %316 into a single fmul). This improves numerical stability and enables hardware FMA instruction selection.
These changes collectively indicate aggressive loop optimization, control-flow simplification, and arithmetic strength reduction—consistent with high-level passes like LoopRotate, LoopSimplify, SLPVectorizer, and InstCombine.
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 freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Labels
None yet
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.
Link: llvm/llvm-project#183692
Requested by: @imkiva