it turns out that when ".with.overflow" intrinsics were added to the X86
backend that they were all implemented except umul.  This one fell back
to the default implementation that did a hi/lo multiply and compared the
top.  Fix this to check the overflow flag that the 'mul' instruction
sets, so we can avoid an explicit test.  Now we compile:

void *func(long count) {
      return new int[count];
}

into:

__Z4funcl:                              ## @_Z4funcl
	movl	$4, %ecx                ## encoding: [0xb9,0x04,0x00,0x00,0x00]
	movq	%rdi, %rax              ## encoding: [0x48,0x89,0xf8]
	mulq	%rcx                    ## encoding: [0x48,0xf7,0xe1]
	seto	%cl                     ## encoding: [0x0f,0x90,0xc1]
	testb	%cl, %cl                ## encoding: [0x84,0xc9]
	movq	$-1, %rdi               ## encoding: [0x48,0xc7,0xc7,0xff,0xff,0xff,0xff]
	cmoveq	%rax, %rdi              ## encoding: [0x48,0x0f,0x44,0xf8]
	jmp	__Znam                  ## TAILCALL

instead of:

__Z4funcl:                              ## @_Z4funcl
	movl	$4, %ecx                ## encoding: [0xb9,0x04,0x00,0x00,0x00]
	movq	%rdi, %rax              ## encoding: [0x48,0x89,0xf8]
	mulq	%rcx                    ## encoding: [0x48,0xf7,0xe1]
	testq	%rdx, %rdx              ## encoding: [0x48,0x85,0xd2]
	movq	$-1, %rdi               ## encoding: [0x48,0xc7,0xc7,0xff,0xff,0xff,0xff]
	cmoveq	%rax, %rdi              ## encoding: [0x48,0x0f,0x44,0xf8]
	jmp	__Znam                  ## TAILCALL

Other than the silly seto+test, this is using the o bit directly, so it's going in the right
direction.




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@120935 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Target/X86/X86ISelDAGToDAG.cpp b/lib/Target/X86/X86ISelDAGToDAG.cpp
index 0226278..1311dba 100644
--- a/lib/Target/X86/X86ISelDAGToDAG.cpp
+++ b/lib/Target/X86/X86ISelDAGToDAG.cpp
@@ -1600,7 +1600,32 @@
       return RetVal;
     break;
   }
-
+  case X86ISD::UMUL: {
+    SDValue N0 = Node->getOperand(0);
+    SDValue N1 = Node->getOperand(1);
+    
+    unsigned LoReg, HiReg;
+    switch (NVT.getSimpleVT().SimpleTy) {
+    default: llvm_unreachable("Unsupported VT!");
+    case MVT::i8:  LoReg = X86::AL;  HiReg = X86::AH;  Opc = X86::MUL8r; break;
+    case MVT::i16: LoReg = X86::AX;  HiReg = X86::DX;  Opc = X86::MUL16r; break;
+    case MVT::i32: LoReg = X86::EAX; HiReg = X86::EDX; Opc = X86::MUL32r; break;
+    case MVT::i64: LoReg = X86::RAX; HiReg = X86::RDX; Opc = X86::MUL64r; break;
+    }
+    
+    SDValue InFlag = CurDAG->getCopyToReg(CurDAG->getEntryNode(), dl, LoReg,
+                                          N0, SDValue()).getValue(1);
+    
+    SDVTList VTs = CurDAG->getVTList(NVT, NVT, MVT::i32);
+    SDValue Ops[] = {N1, InFlag};
+    SDNode *CNode = CurDAG->getMachineNode(Opc, dl, VTs, Ops, 2);
+    
+    ReplaceUses(SDValue(Node, 0), SDValue(CNode, 0));
+    ReplaceUses(SDValue(Node, 1), SDValue(CNode, 1));
+    ReplaceUses(SDValue(Node, 2), SDValue(CNode, 2));
+    return NULL;
+  }
+      
   case ISD::SMUL_LOHI:
   case ISD::UMUL_LOHI: {
     SDValue N0 = Node->getOperand(0);
@@ -1653,11 +1678,12 @@
         CurDAG->getMachineNode(MOpc, dl, MVT::Other, MVT::Flag, Ops,
                                array_lengthof(Ops));
       InFlag = SDValue(CNode, 1);
+
       // Update the chain.
       ReplaceUses(N1.getValue(1), SDValue(CNode, 0));
     } else {
-      InFlag =
-        SDValue(CurDAG->getMachineNode(Opc, dl, MVT::Flag, N1, InFlag), 0);
+      SDNode *CNode = CurDAG->getMachineNode(Opc, dl, MVT::Flag, N1, InFlag);
+      InFlag = SDValue(CNode, 0);
     }
 
     // Prevent use of AH in a REX instruction by referencing AX instead.
@@ -1696,7 +1722,7 @@
       ReplaceUses(SDValue(Node, 1), Result);
       DEBUG(dbgs() << "=> "; Result.getNode()->dump(CurDAG); dbgs() << '\n');
     }
-
+    
     return NULL;
   }