| Issue |
124255
|
| Summary |
LLVM ERROR: Don't know how to widen the operands for `INSERT_SUBVECTOR`
|
| Labels |
backend:X86,
crash-on-valid,
llvm:SelectionDAG
|
| Assignees |
abhishek-kaushik22
|
| Reporter |
abhishek-kaushik22
|
```llvm
define <8 x i32> @foo(<8 x i32> %a, <2 x i32> %b) {
%result = tail call <8 x i32> @llvm.vector.insert.v8i32.v2i32(<8 x i32> %a, <2 x i32> %b, i64 0)
ret <8 x i32> %result
}
```
https://godbolt.org/z/x6qnfPM7W
We get a fatal error in `DAGTypeLegalizer::WidenVecOp_INSERT_SUBVECTOR` for this seemingly simple code.
I'm not sure how this could be fixed. Can we just replace the `INSERT_SUBVECTOR` with a series of `INSERT_VECTOR_ELT` nodes?
There's also #102016 for AArch that also looks like the same problem as this one.
![]()
_______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
