[scanner] Add fast path for converting utf8 that's ascii to utf16

Change-Id: I386261bc012d0d4556b0f9417b2880cc7f44bf2c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1647697
Commit-Queue: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62029}

src/parsing/scanner-character-streams.cc

@@ -590,7 +590,8 @@ void Utf8ExternalStreamingStream::FillBufferFromCurrentChunk() {
     }
   }
 
-  while (cursor < end && output_cursor + 1 < buffer_start_ + kBufferSize) {
+  const uint16_t* max_buffer_end = buffer_start_ + kBufferSize;
+  while (cursor < end && output_cursor + 1 < max_buffer_end) {
     unibrow::uchar t =
         unibrow::Utf8::ValueOfIncremental(&cursor, &state, &incomplete_char);
     if (V8_LIKELY(t <= unibrow::Utf16::kMaxNonSurrogateCharCode)) {
@@ -601,6 +602,18 @@ void Utf8ExternalStreamingStream::FillBufferFromCurrentChunk() {
       *(output_cursor++) = unibrow::Utf16::LeadSurrogate(t);
       *(output_cursor++) = unibrow::Utf16::TrailSurrogate(t);
     }
+    // Fast path for ascii sequences.
+    size_t remaining = end - cursor;
+    size_t max_buffer = max_buffer_end - output_cursor;
+    int max_length = static_cast<int>(Min(remaining, max_buffer));
+    DCHECK_EQ(state, unibrow::Utf8::State::kAccept);
+    const uint8_t* read_end = cursor + max_length;
+    for (; cursor < read_end; cursor++) {
+      uint8_t c = *cursor;
+      DCHECK_EQ(unibrow::Utf8::kMaxOneByteChar, 0x7F);
+      if (c > unibrow::Utf8::kMaxOneByteChar) break;
+      *(output_cursor++) = c;
+    }
   }
 
   current_.pos.bytes = chunk.start.bytes + (cursor - chunk.data);