[Mlir-commits] [mlir] [mlir][ArmSVE] Add convert_to/from_svbool ops (PR #68586)

[Cullen Rhodes]

================
@@ -66,6 +68,77 @@ using ScalableMaskedDivFOpLowering =
     OneToOneConvertToLLVMPattern<ScalableMaskedDivFOp,
                                  ScalableMaskedDivFIntrOp>;
 
+namespace {
+
+/// Unrolls a conversion to/from equivalent vector types, to allow using a
+/// conversion intrinsic that only supports 1-D vector types.
----------------
c-rhodes wrote:

this is doing two things and the op doesn't map 1-1 with intrinsics. Based on previous feedback I've received and general observations, I wonder if the unrolling should be done as a separate transform and this simply maps rank 1 `arm_sve.convert_to_svbool` / `arm_sve.convert_from_svbool` to intrinsics?

It wouldn't have to be done as part of this patch, but something to consider.



https://github.com/llvm/llvm-project/pull/68586