[AArch64] Fold CSEL x, x, cc -> x

This can come up in rare situations, where a csel is created with
identical operands. These can be folded simply to the original value,
allowing the csel to be removed and further simplification to happen.

This patch also removes FCSEL as it is unused, not being produced
anywhere or lowered to anything.

Differential Revision: https://reviews.llvm.org/D101687

Author: davemgreen

llvm/lib/Target/AArch64/AArch64ISelLowering.cpp

@@ -1806,7 +1806,6 @@ const char *AArch64TargetLowering::getTargetNodeName(unsigned Opcode) const {
     MAKE_CASE(AArch64ISD::RET_FLAG)
     MAKE_CASE(AArch64ISD::BRCOND)
     MAKE_CASE(AArch64ISD::CSEL)
-    MAKE_CASE(AArch64ISD::FCSEL)
     MAKE_CASE(AArch64ISD::CSINV)
     MAKE_CASE(AArch64ISD::CSNEG)
     MAKE_CASE(AArch64ISD::CSINC)
@@ -15156,6 +15155,17 @@ static SDValue performBRCONDCombine(SDNode *N,
   return SDValue();
 }
 
+// Optimize CSEL instructions
+static SDValue performCSELCombine(SDNode *N,
+                                  TargetLowering::DAGCombinerInfo &DCI,
+                                  SelectionDAG &DAG) {
+  // CSEL x, x, cc -> x
+  if (N->getOperand(0) == N->getOperand(1))
+    return N->getOperand(0);
+
+  return performCONDCombine(N, DCI, DAG, 2, 3);
+}
+
 // Optimize some simple tbz/tbnz cases.  Returns the new operand and bit to test
 // as well as whether the test should be inverted.  This code is required to
 // catch these cases (as opposed to standard dag combines) because
@@ -15949,7 +15959,7 @@ SDValue AArch64TargetLowering::PerformDAGCombine(SDNode *N,
   case AArch64ISD::TBZ:
     return performTBZCombine(N, DCI, DAG);
   case AArch64ISD::CSEL:
-    return performCONDCombine(N, DCI, DAG, 2, 3);
+    return performCSELCombine(N, DCI, DAG);
   case AArch64ISD::DUP:
     return performPostLD1Combine(N, DCI, false);
   case AArch64ISD::NVCAST:

llvm/lib/Target/AArch64/AArch64ISelLowering.h

@@ -66,7 +66,6 @@ enum NodeType : unsigned {
   RET_FLAG, // Return with a flag operand. Operand 0 is the chain operand.
   BRCOND,   // Conditional branch instruction; "b.cond".
   CSEL,
-  FCSEL, // Conditional move instruction.
   CSINV, // Conditional select invert.
   CSNEG, // Conditional select negate.
   CSINC, // Conditional select increment.

llvm/test/CodeGen/AArch64/srem-seteq.ll

@@ -269,10 +269,7 @@ define i32 @test_srem_int_min(i32 %X) nounwind {
 define i32 @test_srem_allones(i32 %X) nounwind {
 ; CHECK-LABEL: test_srem_allones:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    cmp w0, #0 // =0
-; CHECK-NEXT:    csel w8, w0, w0, lt
-; CHECK-NEXT:    cmp w0, w8
-; CHECK-NEXT:    cset w0, eq
+; CHECK-NEXT:    mov w0, #1
 ; CHECK-NEXT:    ret
   %srem = srem i32 %X, 4294967295
   %cmp = icmp eq i32 %srem, 0