RFR: 8321283: Reuse StringLatin1::equals in regionMatches

[David Schlosnagle]

On Sat, 2 Dec 2023 16:56:22 GMT, Francesco Nigro <duke at openjdk.org> wrote:

> This improvement has been found on https://github.com/vert-x3/vertx-web/pull/2526.
> 
> It can potentially affect the existing ArraysSupport.mismatch caller code-path performance ie requires investigation.

src/java.base/share/classes/java/lang/String.java line 2184:

> 2182:         byte[] tv = value;
> 2183:         byte[] ov = other.value;
> 2184:         if (coder == otherCoder) {

Is this change needed as `otherCoder` is not reused?

Suggestion:

        byte[] tv = value;
        byte[] ov = other.value;
        byte coder = coder();
        if (coder == other.coder()) {

src/java.base/share/classes/java/lang/String.java line 2186:

> 2184:         if (coder == otherCoder) {
> 2185:             if (ooffset == 0 && toffset == 0 && len == (tv.length >> coder) && ov.length == tv.length) {
> 2186:                 return StringLatin1.equals(tv, ov);

A few questions on this:

- Should this also handle `UTF-16`?
- Could we save a comparison by `OR`ing offsets?
- The `equals` method will do the `tv` and `ov` length comparisons themselves, could we remove that check here or is the method call overhead too large?
- Do you happen to have results for `make test TEST="micro:java.lang.StringComparisons"`?

Suggestion:

            if ((ooffset | toffset) == 0 && len == (tv.length >> coder)) {
                return coder == LATIN1
                  ? StringLatin1.equals(tv, ov)
                  : StringUTF16.equals(tv, ov);
            }

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/16933#discussion_r1412902721
PR Review Comment: https://git.openjdk.org/jdk/pull/16933#discussion_r1412887365