On Fri, 27 Feb 2026 14:41:54 GMT, Kevin Walls <[email protected]> wrote:
>> I agree it's a bit convoluted, I think it's meant to be an args.length == 0
>> check. In practice it never hits anyway because even with no explicit args
>> the test runner adds a bunch. Even with vm.flagless I get a few. I could add
>> the assert but feels a bit misleading if we don't have any case that will
>> actually exercise it.
>
> OK sure yes. Some of these older tests are showing their age. If you have
> time, could we just make it do:
> if (args.length > 0) {
> vmOptions = args;
> } else {
> return ...
> }
> ..so it's just trivial to read. Looks like a trap right now, again an
> existing problem. 8-)
Fine by me. I could also just remove it, like I said it never triggers anyway,
and it seems a bit strange to explicitly handle it. If it does happen, we'll
get an implicit null check later on when we try to iterate
`rm.getInputArguments()`, which is arguably better.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/29793#discussion_r2864701106
