[llvm] [X86] combineEXTRACT_SUBVECTOR - fold extract_subvector(subv_broadcast_load(ptr),0) -> load(ptr) (PR #126523)

[Simon Pilgrim via llvm-commits]

================
@@ -58477,10 +58477,26 @@ static SDValue combineEXTRACT_SUBVECTOR(SDNode *N, SelectionDAG &DAG,
                       DAG.isSplatValue(InVec, /*AllowUndefs*/ false)))
     return extractSubVector(InVec, 0, DAG, DL, SizeInBits);
 
-  // If we're extracting a broadcasted subvector, just use the lowest subvector.
-  if (IdxVal != 0 && InVec.getOpcode() == X86ISD::SUBV_BROADCAST_LOAD &&
-      cast<MemIntrinsicSDNode>(InVec)->getMemoryVT() == VT)
-    return extractSubVector(InVec, 0, DAG, DL, SizeInBits);
+  // Check if we're extracting a whole broadcasted subvector.
+  if (InVec.getOpcode() == X86ISD::SUBV_BROADCAST_LOAD) {
+    auto *MemIntr = cast<MemIntrinsicSDNode>(InVec);
+    EVT MemVT = MemIntr->getMemoryVT();
+    if (MemVT == VT) {
+      // Just use the lowest subvector.
+      if (IdxVal != 0)
+        return extractSubVector(InVec, 0, DAG, DL, SizeInBits);
+      // If this is the only use, we can replace with a regular load (this may
+      // have been missed by SimplifyDemandedVectorElts due to extra uses of the
+      // memory chain).
----------------
RKSimon wrote:

The actual vector load size is the same as for the broadcastf128 - and we already do this in other places - so it should be OK?

https://github.com/llvm/llvm-project/pull/126523