llvm / llvm/llvm-project

Missed optimization: fold fcmp (fmul x, C), 0.0 to fcmp x, 0.0 when C > 0.0

Open
#186,553 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

llvm:instcombine missed-optimization
Dominant language
LLVM
Stars
40.5k
Forks
18.7k
PR merge metrics
PR metrics pending

Description

define i1 @src(double %arg0) {
  %v0 = fmul double %arg0, 1.000000e+06
  %v1 = fcmp ule double %v0, 0.000000e+00
  ret i1 %v1
}

define i1 @tgt(double %arg0) {
  %v1 = fcmp ule double %arg0, 0.000000e+00
  ret i1 %v1
}

godbolt: https://godbolt.org/z/PrYYMnf1o
alive2: https://alive2.llvm.org/ce/z/Dfj9iw

Pattern found in: https://github.com/dtcxzyw/llvm-opt-benchmark/blob/main/bench/faiss/optimized/sysinfo.ll

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the LLVM IR reproducer and validate the proposed transform using the linked Godbolt and Alive2 examples. Trace the optimization responsible for folding the fcmp/fmul pattern, using the benchmark reference at faiss/optimized/sysinfo.ll as context. Done means the transformation is proven valid and protected by a regression test.

Written by the indexing model from the issue text.

Assessment

Domain
compilers
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.