8279833: Loop optimization issue in String.encodeUTF8_UTF16

Reviewed-by: rriggs
Backport-of: c3d0a94040

src/java.base/share/classes/java/lang/String.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1994, 2021, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1994, 2022, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -1284,14 +1284,17 @@ public final class String
         int sp = 0;
         int sl = val.length >> 1;
         byte[] dst = new byte[sl * 3];
-        char c;
+        while (sp < sl) {
-        while (sp < sl && (c = StringUTF16.getChar(val, sp)) < '\u0080') {
             // ascii fast loop;
+            char c = StringUTF16.getChar(val, sp);
+            if (c >= '\u0080') {
+                break;
+            }
             dst[dp++] = (byte)c;
             sp++;
         }
         while (sp < sl) {
-            c = StringUTF16.getChar(val, sp++);
+            char c = StringUTF16.getChar(val, sp++);
             if (c < 0x80) {
                 dst[dp++] = (byte)c;
             } else if (c < 0x800) {

test/micro/org/openjdk/bench/java/lang/StringEncode.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2021, 2022, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -30,59 +30,97 @@ import java.util.concurrent.TimeUnit;
 
 @BenchmarkMode(Mode.AverageTime)
 @OutputTimeUnit(TimeUnit.NANOSECONDS)
-@Fork(value = 3, jvmArgs = "-Xmx1g")
+@Fork(value = 3)
 @Warmup(iterations = 5, time = 2)
 @Measurement(iterations = 5, time = 3)
 @State(Scope.Thread)
 public class StringEncode {
 
-    @BenchmarkMode(Mode.AverageTime)
+    @Param({"US-ASCII", "ISO-8859-1", "UTF-8", "MS932", "ISO-8859-6"})
-    @OutputTimeUnit(TimeUnit.NANOSECONDS)
+    private String charsetName;
-    @Fork(value = 3, jvmArgs = "-Xmx1g")
+    private Charset charset;
-    @Warmup(iterations = 5, time = 2)
+    private String asciiString;
-    @Measurement(iterations = 5, time = 2)
+    private String utf16String;
-    @State(Scope.Thread)
+    private String longUtf16String;
-    public static class WithCharset {
+    private String longUtf16StartString;
-
-        @Param({"US-ASCII", "ISO-8859-1", "UTF-8", "MS932", "ISO-8859-6"})
-        private String charsetName;
-
-        private Charset charset;
-        private String asciiString;
-        private String utf16String;
-
-        @Setup
-        public void setup() {
-            charset = Charset.forName(charsetName);
-            asciiString = "ascii string";
-            utf16String = "UTF-\uFF11\uFF16 string";
-        }
-
-        @Benchmark
-        public void encodeCharsetName(Blackhole bh) throws Exception {
-            bh.consume(asciiString.getBytes(charsetName));
-            bh.consume(utf16String.getBytes(charsetName));
-        }
-
-        @Benchmark
-        public void encodeCharset(Blackhole bh) throws Exception {
-            bh.consume(asciiString.getBytes(charset));
-            bh.consume(utf16String.getBytes(charset));
-        }
-    }
-
-    private String asciiDefaultString;
-    private String utf16DefaultString;
 
     @Setup
     public void setup() {
-        asciiDefaultString = "ascii string";
+        charset = Charset.forName(charsetName);
-        utf16DefaultString = "UTF-\uFF11\uFF16 string";
+        asciiString = "ascii string";
+        utf16String = "UTF-\uFF11\uFF16 string";
+        longUtf16String = """
+                 Lorem ipsum dolor sit amet, consectetur adipiscing elit. Aliquam ac sem eu
+                 urna egestas placerat. Etiam finibus ipsum nulla, non mattis dolor cursus a.
+                 Nulla nec nisl consectetur, lacinia neque id, accumsan ante. Curabitur et
+                 sapien in magna porta ultricies. Sed vel pellentesque nibh. Pellentesque dictum
+                 dignissim diam eu ultricies. Class aptent taciti sociosqu ad litora torquent
+                 per conubia nostra, per inceptos himenaeos. Suspendisse erat diam, fringilla
+                 sed massa sed, posuere viverra orci. Suspendisse tempor libero non gravida
+                 efficitur. Vivamus lacinia risus non orci viverra, at consectetur odio laoreet.
+                 Suspendisse potenti.
+
+                 Phasellus vel nisi iaculis, accumsan quam sed, bibendum eros. Sed venenatis
+                 nulla tortor, et eleifend urna sodales id. Nullam tempus ac metus sit amet
+                 sollicitudin. Nam sed ex diam. Praesent vitae eros et neque condimentum
+                 consectetur eget non tortor. Praesent bibendum vel felis nec dignissim.
+                 Maecenas a enim diam. Suspendisse quis ligula at nisi accumsan lacinia id
+                 hendrerit sapien. Donec aliquam mattis lectus eu ultrices. Duis eu nisl
+                 euismod, blandit mauris vel, placerat urna. Etiam malesuada enim purus,
+                 tristique mollis odio blandit quis. Vivamus posuere.
+                 \uFF11
+                """;
+        longUtf16StartString = """
+                 \uFF11
+                 Lorem ipsum dolor sit amet, consectetur adipiscing elit. Aliquam ac sem eu
+                 urna egestas placerat. Etiam finibus ipsum nulla, non mattis dolor cursus a.
+                 Nulla nec nisl consectetur, lacinia neque id, accumsan ante. Curabitur et
+                 sapien in magna porta ultricies. Sed vel pellentesque nibh. Pellentesque dictum
+                 dignissim diam eu ultricies. Class aptent taciti sociosqu ad litora torquent
+                 per conubia nostra, per inceptos himenaeos. Suspendisse erat diam, fringilla
+                 sed massa sed, posuere viverra orci. Suspendisse tempor libero non gravida
+                 efficitur. Vivamus lacinia risus non orci viverra, at consectetur odio laoreet.
+                 Suspendisse potenti.
+
+                 Phasellus vel nisi iaculis, accumsan quam sed, bibendum eros. Sed venenatis
+                 nulla tortor, et eleifend urna sodales id. Nullam tempus ac metus sit amet
+                 sollicitudin. Nam sed ex diam. Praesent vitae eros et neque condimentum
+                 consectetur eget non tortor. Praesent bibendum vel felis nec dignissim.
+                 Maecenas a enim diam. Suspendisse quis ligula at nisi accumsan lacinia id
+                 hendrerit sapien. Donec aliquam mattis lectus eu ultrices. Duis eu nisl
+                 euismod, blandit mauris vel, placerat urna. Etiam malesuada enim purus,
+                 tristique mollis odio blandit quis. Vivamus posuere.
+                """;
     }
 
     @Benchmark
-    public void encodeDefault(Blackhole bh) throws Exception {
+    public byte[] encodeAsciiCharsetName() throws Exception {
-        bh.consume(asciiDefaultString.getBytes());
+        return asciiString.getBytes(charset);
-        bh.consume(utf16DefaultString.getBytes());
+    }
+
+    @Benchmark
+    public byte[] encodeAscii() throws Exception {
+        return asciiString.getBytes(charset);
+    }
+
+    @Benchmark
+    public void encodeMix(Blackhole bh) throws Exception {
+        bh.consume(asciiString.getBytes(charset));
+        bh.consume(utf16String.getBytes(charset));
+    }
+
+    @Benchmark
+    public byte[] encodeUTF16LongEnd() throws Exception {
+        return longUtf16String.getBytes(charset);
+    }
+
+    @Benchmark
+    public byte[] encodeUTF16LongStart() throws Exception {
+        return longUtf16StartString.getBytes(charset);
+    }
+
+    @Benchmark
+    public byte[] encodeUTF16() throws Exception {
+        return utf16String.getBytes(charset);
     }
 }