[llvm] [SVE] Wide active lane mask (PR #76514)

Momchil Velikov via llvm-commits llvm-commits at lists.llvm.org
Tue Jan 9 09:25:54 PST 2024


================
@@ -19791,15 +19792,63 @@ static SDValue performIntrinsicCombine(SDNode *N,
           EVT::getVectorVT(*DAG.getContext(), PromVT.getVectorElementType(),
                            VT.getVectorElementCount());
 
-      Res = DAG.getNode(ISD::INTRINSIC_WO_CHAIN, DL, WhileVT, ID,
-                        N->getOperand(1), N->getOperand(2));
+      SDValue Res = DAG.getNode(ISD::INTRINSIC_WO_CHAIN, DL, WhileVT, ID,
+                                N->getOperand(1), N->getOperand(2));
       Res = DAG.getNode(ISD::SIGN_EXTEND, DL, PromVT, Res);
       Res = DAG.getNode(ISD::EXTRACT_SUBVECTOR, DL, ExtVT, Res,
                         DAG.getConstant(0, DL, MVT::i64));
       Res = DAG.getNode(ISD::TRUNCATE, DL, VT, Res);
+
+      return Res;
     }
-    return Res;
+
+    if (!Subtarget->hasSVE2p1() && !Subtarget->hasSME2())
+      return SDValue();
+
+    if (!N->hasNUsesOfValue(2, 0))
+      return SDValue();
+
+    auto It = N->use_begin();
+    SDNode *Lo = *It++;
----------------
momchil-velikov wrote:

Then in this case, without making changes, the first extract will be lower half of the vector (`Lo`) and the second will be the higher part of the vector (`Hi`), which means that we will not do swaps, isn't it?

(but still will work correctly if for some reason the uses come in a different order).

https://github.com/llvm/llvm-project/pull/76514


More information about the llvm-commits mailing list