RFR: 8327791: Optimization for new BigDecimal(String) [v2]

[Claes Redestad]

On Mon, 11 Mar 2024 14:15:38 GMT, Chen Liang <liach at openjdk.org> wrote:

> Looks great to me. Sorry for the pings, but we may need @rgiulietti to verify the math correctness and @cl4es to comment on whether having these 2 separate code paths or trying to extract a common part is the better approach.

I'd love to see less code duplication. It'd certainly be possible to refactor to one internal constructor `BigDecimal(CharSequence)` that the `String`-based constructors just delegates to, and then have the `BigDecimal(char[])` variants call with some wrapping (e.g. `this(CharBuffers.wrap(in, offset, len), mc)`). 

Such indirection might cost a few cycles per operation for the `char[]`-based constructors due the need to wrap the `char[]`, but I wonder how performance sensitive the `BigDecimal(char[], int, int)` paths really are once `BigDecimal(String)` takes another path.

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

PR Comment: https://git.openjdk.org/jdk/pull/18177#issuecomment-1988606133