| Issue |
120764
|
| Summary |
[VectorCombine] foldShuffleOfShuffles - failure to merge nested shuffles with common operands
|
| Labels |
missed-optimization,
llvm:transforms
|
| Assignees |
|
| Reporter |
RKSimon
|
```ll
define <4 x double> @add_v4f64_u123(<4 x double> %a, <4 x double> %b) {
%1 = shufflevector <4 x double> %b, <4 x double> %a, <4 x i32> <i32 poison, i32 0, i32 6, i32 poison>
%2 = shufflevector <4 x double> %b, <4 x double> %a, <4 x i32> <i32 poison, i32 1, i32 7, i32 poison>
%3 = shufflevector <4 x double> %1, <4 x double> %b, <4 x i32> <i32 poison, i32 1, i32 2, i32 6>
%4 = shufflevector <4 x double> %2, <4 x double> %b, <4 x i32> <i32 poison, i32 1, i32 2, i32 7>
%result = fadd <4 x double> %3, %4
ret <4 x double> %result
}
```
All shuffles only have 2 operands, so we should be able to fold to:
```ll
define <4 x double> @add_v4f64_u123(<4 x double> %a, <4 x double> %b) {
%1 = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 poison, i32 4, i32 2, i32 6>
%2 = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 poison, i32 5, i32 3, i32 7>
%result = fadd <4 x double> %1, %2
ret <4 x double> %result
}
```
Pulled out of #34072
![]()
_______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
