X86 supports i8/i16 overflow ops (except i8 multiplies), we should

generate them.

Now we compile:

define zeroext i8 @X(i8 signext %a, i8 signext %b) nounwind ssp {
entry:
  %0 = tail call %0 @llvm.sadd.with.overflow.i8(i8 %a, i8 %b)
  %cmp = extractvalue %0 %0, 1
  br i1 %cmp, label %if.then, label %if.end

into:

_X:                                     ## @X
## BB#0:                                ## %entry
	subl	$12, %esp
	movb	16(%esp), %al
	addb	20(%esp), %al
	jo	LBB0_2

Before we were generating:

_X:                                     ## @X
## BB#0:                                ## %entry
	pushl	%ebp
	movl	%esp, %ebp
	subl	$8, %esp
	movb	12(%ebp), %al
	testb	%al, %al
	setge	%cl
	movb	8(%ebp), %dl
	testb	%dl, %dl
	setge	%ah
	cmpb	%cl, %ah
	sete	%cl
	addb	%al, %dl
	testb	%dl, %dl
	setge	%al
	cmpb	%al, %ah
	setne	%al
	andb	%cl, %al
	testb	%al, %al
	jne	LBB0_2

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122186 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index 1e8b27b8d10419b27160a503192abb323bc6bf40..06c41a0a450bf0230cdbd49bcbd8b2e2bb83be48 100644 (file)
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -89,6 +89,7 @@ X86TargetLowering::X86TargetLowering(X86TargetMachine &TM)
   TD = getTargetData();
 
   // Set up the TargetLowering object.
+  static MVT IntVTs[] = { MVT::i8, MVT::i16, MVT::i32, MVT::i64 };
 
   // X86 is weird, it always uses i8 for shift amounts and setcc results.
   setShiftAmountType(MVT::i8);
@@ -826,9 +827,8 @@ X86TargetLowering::X86TargetLowering(X86TargetMachine &TM)
     }
   }
 
-  if (Subtarget->hasSSE42()) {
+  if (Subtarget->hasSSE42())
     setOperationAction(ISD::VSETCC,             MVT::v2i64, Custom);
-  }
 
   if (!UseSoftFloat && Subtarget->hasAVX()) {
     addRegisterClass(MVT::v8f32, X86::VR256RegisterClass);
@@ -942,28 +942,27 @@ X86TargetLowering::X86TargetLowering(X86TargetMachine &TM)
   // We want to custom lower some of our intrinsics.
   setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::Other, Custom);
 
-  // Add/Sub/Mul with overflow operations are custom lowered.
-  setOperationAction(ISD::SADDO, MVT::i32, Custom);
-  setOperationAction(ISD::UADDO, MVT::i32, Custom);
-  setOperationAction(ISD::SSUBO, MVT::i32, Custom);
-  setOperationAction(ISD::USUBO, MVT::i32, Custom);
-  setOperationAction(ISD::SMULO, MVT::i32, Custom);
-  setOperationAction(ISD::UMULO, MVT::i32, Custom);
-
+    
   // Only custom-lower 64-bit SADDO and friends on 64-bit because we don't
   // handle type legalization for these operations here.
   //
   // FIXME: We really should do custom legalization for addition and
   // subtraction on x86-32 once PR3203 is fixed.  We really can't do much better
   // than generic legalization for 64-bit multiplication-with-overflow, though.
-  if (Subtarget->is64Bit()) {
-    setOperationAction(ISD::SADDO, MVT::i64, Custom);
-    setOperationAction(ISD::UADDO, MVT::i64, Custom);
-    setOperationAction(ISD::SSUBO, MVT::i64, Custom);
-    setOperationAction(ISD::USUBO, MVT::i64, Custom);
-    setOperationAction(ISD::SMULO, MVT::i64, Custom);
-    setOperationAction(ISD::UMULO, MVT::i64, Custom);
+  for (unsigned i = 0, e = 3+Subtarget->is64Bit(); i != e; ++i) {
+    // Add/Sub/Mul with overflow operations are custom lowered.
+    MVT VT = IntVTs[i];
+    setOperationAction(ISD::SADDO, VT, Custom);
+    setOperationAction(ISD::UADDO, VT, Custom);
+    setOperationAction(ISD::SSUBO, VT, Custom);
+    setOperationAction(ISD::USUBO, VT, Custom);
+    setOperationAction(ISD::SMULO, VT, Custom);
+    setOperationAction(ISD::UMULO, VT, Custom);
   }
+    
+  // There are no 8-bit 3-address imul/mul instructions
+  setOperationAction(ISD::SMULO, MVT::i8, Expand);
+  setOperationAction(ISD::UMULO, MVT::i8, Expand);
 
   if (!Subtarget->is64Bit()) {
     // These libcalls are not available in 32-bit.

diff --git a/test/Transforms/InstCombine/intrinsics.ll b/test/Transforms/InstCombine/intrinsics.ll
index ed9c7b3840ab569056af2af0f7f490346c0928d5..50e7f1f7c92d35cda65f558870a7d783c3faeb77 100644 (file)
--- a/test/Transforms/InstCombine/intrinsics.ll
+++ b/test/Transforms/InstCombine/intrinsics.ll
@@ -80,6 +80,15 @@ define i1 @uaddtest6(i8 %A, i8 %B) {
 ; CHECK-NEXT: ret i1 %z
 }
 
+define i8 @uaddtest7(i8 %A, i8 %B) {
+  %x = call %overflow.result @llvm.uadd.with.overflow.i8(i8 %A, i8 %B)
+  %z = extractvalue %overflow.result %x, 0
+  ret i8 %z
+; CHECK: @uaddtest7
+; CHECK-NEXT: %z = add i8 %A, %B
+; CHECK-NEXT: ret i8 %z
+}
+
 
 define i8 @umultest1(i8 %A, i1* %overflowPtr) {
   %x = call %overflow.result @llvm.umul.with.overflow.i8(i8 0, i8 %A)