RFR: 7902982: jcstress: Add samples for some mutex algorithms [v3]

[Aleksey Shipilev]

On Tue, 29 Jun 2021 14:05:31 GMT, Michael Mirwaldt <github.com+6693355+mmirwaldt at openjdk.org> wrote:

>> Ah thanks for that answer. I have not got such deep insights into the "secondary effects" of volatile.
>> I have removed it from all 4 samples.
>> Just another question: Is it theoretically possible that we get false positives without volatile?
>> I mean taken1 and taken2 are updated by their actors but the other actor might not see it.
>> Then both actors might enter the critical section at the same time without any actor noticing it in the results.
>> Of course, I know there is no free lunch.
>
> What about adding 
> "In case you make taken1 and taken2 volatiles, you might accidentally re-create IRIW case out of these taken1, taken2, and a couple of other fields (see https://github.com/openjdk/jcstress/blob/df83b446f187ae0b0fa31fa54decb59db9e955da/tests-custom/src/main/java/org/openjdk/jcstress/tests/volatiles/VolatileIRIWTest.java )" as a comment into Mutex_01_NoAlgorithm? I guess many other users might wonder why taken1 and taken2 are not volatile?

Just do this: https://github.com/openjdk/jcstress/pull/85#discussion_r660633790

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

PR: https://git.openjdk.java.net/jcstress/pull/85