RFR: 8347047: Cleanup action passed to MemorySegment::reinterpret keeps old segment alive

[Per Minborg]

On Tue, 7 Jan 2025 19:38:23 GMT, Brent Christian <bchristi at openjdk.org> wrote:

>> This PR proposes to eliminate the capturing of `this` in the cleanup action of `AbstractMemorySegment::reinterpretInternal`.
>
> As a matter of style, I believe there is something to be said for writing out cleaning actions as static nested classes, as one can/must include explicit fields for all object and values that the `Runnable` object must reference. But it's more verbose, of course.
> 
> Also, some conversions of finalizer to Cleaner have included tests to ensure that objects become unreachable as expected. (A relatively simple technique is to add the objects in question into a WeakHashMap, and ensure they are removed as expected.) [GssContextCleanup.java](https://github.com/openjdk/jdk/blob/master/test/jdk/sun/security/jgss/GssContextCleanup.java) and [GssNameCleanup.java](https://github.com/openjdk/jdk/blob/master/test/jdk/sun/security/jgss/GssNameCleanup.java) from [JDK-8284490](https://bugs.openjdk.org/browse/JDK-8284490) are examples. I don't know if it's practical to construct such a test in this case.
> 
> Anyway, just a couple things to think about.

Thanks for the tips @bchristi-git : https://bugs.openjdk.org/browse/JDK-8347249

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

PR Comment: https://git.openjdk.org/jdk/pull/22943#issuecomment-2577114779