On Wed, 2 Sep 2026 20:51:42 GMT, Nir Lisker <[email protected]> wrote:

>> oh, you are asking whether it fails without synchronization - no it does 
>> not, the starts should align just right for it to fail, really.
>> 
>> I don't think we should test the concurrency contract of the `HashMap` 
>> though.   Do you have a specific suggestion?
>
> I mean the `synchronized` blocks. You're testing concurrent access. If I 
> remove those blocks, I expect the test to fail.
> 
> The parameter values I use depend on what produces a failure reliably.

Well, what does this test test then? Your points 1 and 5 are about lack of 
thread safety. A test that they are resolved should fail before the fix and not 
fail after, unless you think that it's trivial, and then there's no need for a 
test.

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

PR Review Comment: https://git.openjdk.org/jfx/pull/2197#discussion_r3918662678

Reply via email to