On Sun, 5 Jan 2025 00:12:03 GMT, Shaojin Wen <s...@openjdk.org> wrote:

>  2. The leading zero scenario is not common, and there is no need to optimize 
> this scenario.

Another approach would be to limit the “fast” path to the case where the 
overall string length is 9 or less. That would allow overflow-free calculation 
without the extra  bytecode weight of skipping leading zeros, possibly still 
with simpler code.

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

PR Comment: https://git.openjdk.org/jdk/pull/22919#issuecomment-2571479762

Reply via email to