Fix bug when memcmp had size equal three

Author: igogo-x86

llvm/lib/CodeGen/ExpandMemCmp.cpp

@@ -648,7 +648,7 @@ Value *MemCmpExpansion::getMemCmpOneBlock() {
 
   // The i8 and i16 cases don't need compares. We zext the loaded values and
   // subtract them to get the suitable negative, zero, or positive i32 result.
-  if (Size < 4) {
+  if (Size == 1 || Size == 2) {
     const LoadPair Loads = getLoadPair(LoadSizeType, BSwapSizeType,
                                        Builder.getInt32Ty(), /*Offset*/ 0);
     return Builder.CreateSub(Loads.Lhs, Loads.Rhs);

llvm/test/CodeGen/AArch64/memcmp.ll

@@ -160,6 +160,9 @@ define i32 @length3(ptr %X, ptr %Y) nounwind {
 ; CHECK-NEXT:    orr w9, w11, w10, lsl #16
 ; CHECK-NEXT:    rev w8, w8
 ; CHECK-NEXT:    rev w9, w9
+; CHECK-NEXT:    cmp w8, w9
+; CHECK-NEXT:    cset w8, hi
+; CHECK-NEXT:    cset w9, lo
 ; CHECK-NEXT:    sub w0, w8, w9
 ; CHECK-NEXT:    ret
   %m = tail call i32 @memcmp(ptr %X, ptr %Y, i64 3) nounwind

llvm/test/Transforms/ExpandMemCmp/AArch64/memcmp.ll

@@ -44,8 +44,12 @@ define i32 @cmp3(ptr nocapture readonly %x, ptr nocapture readonly %y)  {
 ; CHECK-NEXT:    [[TMP4:%.*]] = zext i24 [[TMP2]] to i32
 ; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.bswap.i32(i32 [[TMP3]])
 ; CHECK-NEXT:    [[TMP6:%.*]] = call i32 @llvm.bswap.i32(i32 [[TMP4]])
-; CHECK-NEXT:    [[TMP7:%.*]] = sub i32 [[TMP5]], [[TMP6]]
-; CHECK-NEXT:    ret i32 [[TMP7]]
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp ugt i32 [[TMP5]], [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = icmp ult i32 [[TMP5]], [[TMP6]]
+; CHECK-NEXT:    [[TMP9:%.*]] = zext i1 [[TMP7]] to i32
+; CHECK-NEXT:    [[TMP10:%.*]] = zext i1 [[TMP8]] to i32
+; CHECK-NEXT:    [[TMP11:%.*]] = sub i32 [[TMP9]], [[TMP10]]
+; CHECK-NEXT:    ret i32 [[TMP11]]
 ;
   %call = tail call i32 @memcmp(ptr %x, ptr %y, i64 3)
   ret i32 %call