RFR: 8349176: Speed up Integer/Long.toString via StringConcatHelper::newArray [v3]

[Chen Liang]

On Fri, 2 May 2025 03:55:24 GMT, Shaojin Wen <swen at openjdk.org> wrote:

>> The byte[] allocated in Integer/Long.toString is fully filled, so we can use StringConcatHelper::newArray to create byte[] to improve performance.
>
> Shaojin Wen has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains eight commits:
> 
>  - Merge remote-tracking branch 'upstream/master' into allocate_un_init_202501
>    
>    # Conflicts:
>    #	src/java.base/share/classes/java/lang/Integer.java
>    #	src/java.base/share/classes/java/lang/Long.java
>  - use StringConcatHelper.newArray
>  - simplify
>  - use Unsafe.allocateUninitializedArray
>  - revert StringConcatHelper newArray change
>  - copyright
>  - remove duplicate check
>  - allocateUninitializedArray

src/java.base/share/classes/java/lang/Integer.java line 439:

> 437:             return new String(buf, LATIN1);
> 438:         } else {
> 439:             byte[] buf = StringConcatHelper.newArray(size << 1);

@RogerRiggs I recommended StringConcatHelper because that was used in StringLatin1.

For StringUTF16, it has its own newBytesFor:
Suggestion:

            byte[] buf = StringUTF16.newBytesFor(size);

Or we can factor out a `StringLatin1.newBytesFor` too and encapsulate `newArray` usage.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/23353#discussion_r2071969862