Add benchmark for String.indexOf(String)

[NigelYiboYu]

Add benchmark for String.indexOf(String) API

Signed-off-by: Nigel Yu yunigel@ca.ibm.com

[NigelYiboYu]

Updated the benchmark to use a human-readable string as its default source string, based on which sub-strings are generated. This 64K character long source string is preprocessed to contain only compressible characters.

[fjeremic]

@piyush286 / @smlambert do you might know how typically gets assigned to be a committer for these PRs? It would still be valuable to have them merged.

[piyush286]

I usually request a review from @smlambert, who can throw more light on this.

[smlambert]

Did not see this as active, so it fell off my radar, apologies.

I can review, and would merge once I run a Grinder (assuming it will get exercised as part of this target: https://github.com/AdoptOpenJDK/openjdk-tests/blob/master/perf/bumbleBench/playlist.xml#L539).

[piyush286]

This is a new test case StringIndexOfStringBench so we'll need to add it to our BumbleBench playlist file. I can add it and test this change once we're done with Quarkus testing, unless someone else gets to it before that.

The above test case is for StringIndexOfBench: https://github.com/AdoptOpenJDK/bumblebench/blob/master/net/adoptopenjdk/bumblebench/string/StringIndexOfBench.java https://github.com/AdoptOpenJDK/openjdk-tests/blob/dc1569c25d402cd3c4936f95ed49bcf5786b1ef9/perf/bumbleBench/playlist.xml#L540

[smlambert]

Run locally, we can/should create a new test target for it next week.

[fjeremic]

Thanks everyone!