On Mon, 3 Mar 2025 18:32:42 GMT, Artur Barashev <abaras...@openjdk.org> wrote:
>> Loosen the check; the modified test is sufficient to demonstrate
>> sub-millisecond precision.
>
> test/jdk/sun/security/krb5/MicroTime.java line 49:
>
>> 47: // Before this change, KerberosTime was implemented in
>> milliseconds.
>> 48: // Now there should be more.
>> 49: if (count < 1001) {
>
> Why 1001 and not 1000?
Nevermind that, 1001 is needed to make sure we have a precision at least better
than 1 millisecond. So 1001 unique timestamps out of of 1M gives us just that.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/23867#discussion_r1978028835
