forked from OSchip/llvm-project
5dc4e7c2b9
scalarizeBinop currently folds vec_bo((inselt VecC0, V0, Index), (inselt VecC1, V1, Index)) -> inselt(vec_bo(VecC0, VecC1), scl_bo(V0,V1), Index) This patch extends this to account for cases where one of the vec_bo operands is already all-constant and performs similar cost checks to determine if the scalar binop with a constant still makes sense: vec_bo((inselt VecC0, V0, Index), VecC1) -> inselt(vec_bo(VecC0, VecC1), scl_bo(V0,extractelt(V1,Index)), Index) Fixes PR42174 Differential Revision: https://reviews.llvm.org/D80885 |
||
|---|---|---|
| .. | ||
| addsub.ll | ||
| horiz-math.ll | ||
| lit.local.cfg | ||
| scalarization.ll | ||
| shuffle.ll | ||
| vdiv.ll | ||
| vector-reductions-expanded.ll | ||
| vector-reductions.ll | ||