RFR: CODETOOLS-7903021: jcstress: Dining Philosophers problem sample [v2]

[Michael Mirwaldt]

On Mon, 6 Sep 2021 15:43:46 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> 
> I think 13 minutes per sample is a bit too much. I think we can cut down the time significantly by only assuming 3 philosophers, not 4.

I have reduced the number of actors from 4 to 3.

> 
> I also believe that the sample should actually demonstrate several **architectural** solutions: resource hierarchy, arbitrator, etc. See Wikipedia entry for the possible solutions. Doing the same resource hierarchy solution with different styles of locks does not serve a purpose for this sample, I think.

I have reduced this sample to the resource hierarchy solution.
However, I don't fully understand the other 2 proposed solutions, the arbitrator and William Stallings's "Limiting the number of diners in the table", in the wikipedia article yet but I will somehow.

I have decided not to implement the Chandy/Misra solution with agents. It sounds much too complicated for me and I don't think JCStress will ever support the ideas of agents.

> My point is to use the minimum number of actors that still makes the test viable. For Dining Philosophers, it seems three threads are enough. We shall address the reader-writer thread problem, when we come to that.

Ok. Let's see how we can enable the following samples when they come. I sometimes think too many steps ahead ;-)

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

PR: https://git.openjdk.java.net/jcstress/pull/93