| Issue |
166887
|
| Summary |
Missed Optimization: Fold sdiv exact(((trunc x)*24 - 24), 24) into sext(trunc x - 1)
|
| Labels |
new issue
|
| Assignees |
|
| Reporter |
Leo0506
|
```
define i32 @src(i32 %arg0) {
%v0 = trunc nuw nsw i32 %arg0 to i8
%v1 = mul nuw i8 %v0, 24
%v2 = add i8 %v1, -24
%v3 = sdiv exact i8 %v2, 24
%v4 = sext i8 %v3 to i32
ret i32 %v4
}
define i32 @tgt(i32 %arg0) {
%v0 = trunc nuw nsw i32 %arg0 to i8
%v1 = add i8 %v0, -1
%v2 = sext i8 %v1 to i32
ret i32 %v2
}
```
alive2: https://alive2.llvm.org/ce/z/TCcCGr
godbolt: https://godbolt.org/z/o67Toxc9c
Pattern found in: https://github.com/dtcxzyw/llvm-opt-benchmark/blob/main/bench/nix/optimized/indirect.ll
![]()
_______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
