[llvm-branch-commits] [llvm] [AMDGPU][SDAG] DAGCombine PTRADD -> disjoint OR (PR #146075)

[Fabian Ritter via llvm-branch-commits]

================
@@ -2766,6 +2766,33 @@ SDValue DAGCombiner::visitPTRADD(SDNode *N) {
     }
   }
 
+  // Transform (ptradd a, b) -> (or disjoint a, b) if it is equivalent and if
+  // that transformation can't block an offset folding at any use of the ptradd.
+  // This should be done late, after legalization, so that it doesn't block
+  // other ptradd combines that could enable more offset folding.
+  if (LegalOperations && DAG.haveNoCommonBitsSet(N0, N1)) {
+    bool TransformCanBreakAddrMode = false;
+    if (auto *C = dyn_cast<ConstantSDNode>(N1)) {
----------------
ritter-x2a wrote:

It's not only handling the constant case, that should be clearer now, in the updated version.

https://github.com/llvm/llvm-project/pull/146075