[arm][arm64] Improve CountTrailingZero implementation.

src/arm64/utils-arm64.h

 // Copyright 2013 the V8 project authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #ifndef V8_ARM64_UTILS_ARM64_H_
 #define V8_ARM64_UTILS_ARM64_H_
 
 #include <cmath>
 
 #include "src/arm64/constants-arm64.h"
@@ skipping 37 matching lines @@
 // Bit counting.
 int CountLeadingZeros(uint64_t value, int width);
 int CountLeadingSignBits(int64_t value, int width);
 int CountTrailingZeros(uint64_t value, int width);
 int CountSetBits(uint64_t value, int width);
 uint64_t LargestPowerOf2Divisor(uint64_t value);
 int MaskToBit(uint64_t mask);
 
 
 template <typename T>
-T ReverseBits(T value) {
-  DCHECK((sizeof(value) == 1) || (sizeof(value) == 2) || (sizeof(value) == 4) ||
-         (sizeof(value) == 8));
-  T result = 0;
-  for (unsigned i = 0; i < (sizeof(value) * 8); i++) {
-    result = (result << 1) | (value & 1);
-    value >>= 1;
-  }
-  return result;
-}
-
-
-template <typename T>
 T ReverseBytes(T value, int block_bytes_log2) {
   DCHECK((sizeof(value) == 4) || (sizeof(value) == 8));
   DCHECK((1U << block_bytes_log2) <= sizeof(value));
   // Split the 64-bit value into an 8-bit array, where b[0] is the least
   // significant byte, and b[7] is the most significant.
   uint8_t bytes[8];
   uint64_t mask = 0xff00000000000000;
   for (int i = 7; i >= 0; i--) {
     bytes[i] = (static_cast<uint64_t>(value) & mask) >> (i * 8);
     mask >>= 8;
@@ skipping 61 matching lines @@
 
 
 inline float FusedMultiplyAdd(float op1, float op2, float a) {
   return fmaf(op1, op2, a);
 }
 
 }  // namespace internal
 }  // namespace v8
 
 #endif  // V8_ARM64_UTILS_ARM64_H_