[turbofan] Add support for overflow add/sub to the MachineOperatorReducer.

src/base/bits-unittest.cc

Index: src/base/bits-unittest.cc
diff --git a/src/base/bits-unittest.cc b/src/base/bits-unittest.cc
index e06fd097ead23487ad9084eb91558e02171eb8e8..06c1183586732d9e59abafbe3827eb2c2764c203 100644
--- a/src/base/bits-unittest.cc
+++ b/src/base/bits-unittest.cc
@@ -2,6 +2,8 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
+#include <limits>
+
 #include "src/base/bits.h"
 #include "src/base/macros.h"
 #include "testing/gtest-support.h"
@@ -119,6 +121,47 @@ TEST(Bits, RotateRight64) {
   EXPECT_EQ(V8_UINT64_C(0x8000000000000000), RotateRight64(1, 1));
 }
 
+
+TEST(Bits, SignedAddOverflow32) {
+  int32_t val = 0;
+  EXPECT_FALSE(SignedAddOverflow32(0, 0, &val));
+  EXPECT_EQ(0, val);
+  EXPECT_TRUE(
+      SignedAddOverflow32(std::numeric_limits<int32_t>::max(), 1, &val));
+  EXPECT_EQ(std::numeric_limits<int32_t>::min(), val);
+  EXPECT_TRUE(
+      SignedAddOverflow32(std::numeric_limits<int32_t>::min(), -1, &val));
+  EXPECT_EQ(std::numeric_limits<int32_t>::max(), val);
+  EXPECT_TRUE(SignedAddOverflow32(std::numeric_limits<int32_t>::max(),
+                                  std::numeric_limits<int32_t>::max(), &val));
+  EXPECT_EQ(-2, val);
+  TRACED_FORRANGE(int32_t, i, 1, 50) {
+    TRACED_FORRANGE(int32_t, j, 1, i) {
+      EXPECT_FALSE(SignedAddOverflow32(i, j, &val));
+      EXPECT_EQ(i + j, val);
+    }
+  }
+}
+
+
+TEST(Bits, SignedSubOverflow32) {
+  int32_t val = 0;
+  EXPECT_FALSE(SignedSubOverflow32(0, 0, &val));
+  EXPECT_EQ(0, val);
+  EXPECT_TRUE(
+      SignedSubOverflow32(std::numeric_limits<int32_t>::min(), 1, &val));
+  EXPECT_EQ(std::numeric_limits<int32_t>::max(), val);
+  EXPECT_TRUE(
+      SignedSubOverflow32(std::numeric_limits<int32_t>::max(), -1, &val));
+  EXPECT_EQ(std::numeric_limits<int32_t>::min(), val);
+  TRACED_FORRANGE(int32_t, i, 1, 50) {
+    TRACED_FORRANGE(int32_t, j, 1, i) {
+      EXPECT_FALSE(SignedSubOverflow32(i, j, &val));
+      EXPECT_EQ(i - j, val);
+    }
+  }
+}
+
 }  // namespace bits
 }  // namespace base
 }  // namespace v8