[llvm] [AMDGPU] Split wide integer dpp8 intrinsic calls (PR #113500)

[Matt Arsenault via llvm-commits]

================
@@ -2257,6 +2260,38 @@ bool AMDGPUCodeGenPrepareImpl::visitSqrt(IntrinsicInst &Sqrt) {
   return true;
 }
 
+// Split unsupported wide integer calls.
+bool AMDGPUCodeGenPrepareImpl::visitMovDppIntrinsic(IntrinsicInst &I) {
----------------
arsenm wrote:

This is an optimization pass and cannot be relied on for lowering. This should be handled in codegen proper like the other dpp intrinsics 

https://github.com/llvm/llvm-project/pull/113500