RFR: 7902989: jcstress: Add samples for high-level race conditions [v5]

[Vladimir Sitnikov]

On Thu, 1 Jul 2021 14:14:14 GMT, Michael Mirwaldt <github.com+6693355+mmirwaldt at openjdk.org> wrote:

>> jcstress-samples/src/main/java/org/openjdk/jcstress/samples/high/race/condition/RaceCondition_01_ReadModifyWriteSequence.java line 45:
>> 
>>> 43: @Outcome(id = {"250, 150, 150"}, expect = ACCEPTABLE, desc = "Actor2 considered actor1's result and wrote its right result")
>>> 44: @Outcome(id = {"250, 100, 250", "250, 150, 250"}, expect = ACCEPTABLE_INTERESTING, desc = "Actor1 ignored actor2's result and wrote its wrong result")
>>> 45: @Outcome(id = {"250, 100, 100", "150, 100, 100"}, expect = ACCEPTABLE_INTERESTING, desc = "Actor2 ignored actor1's result and wrote its wrong result")
>> 
>> Please explain how `"250, 150, 250"` and `"150, 100, 100"` are acceptable.
>> 
>> I think those outcomes are forbidden.
>
> Why should they be forbidden?
> I think they are undesirable but allowed because the lack of the appropiate synchronisation causes these result within read-modify-write-sequences.

>Why should they be forbidden?

> 17.4.4. Synchronization Order: https://docs.oracle.com/javase/specs/jls/se7/html/jls-17.html#jls-17.4.4
> A write to a volatile variable v (§8.3.1.4) synchronizes-with all subsequent reads of v by any thread (where "subsequent" is defined according to the synchronization order).

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

PR: https://git.openjdk.java.net/jcstress/pull/88