RFR: 8322149: ConcurrentHashMap copy constructor should not transfer from old table on presizing [v2]

[Volker Simonis]

On Thu, 11 Jan 2024 19:21:00 GMT, Joshua Cao <duke at openjdk.org> wrote:

>>> We don't need to compute max() here. [tryPresize()](https://github.com/openjdk/jdk/blob/8a4dc79e1a40e7115e2971af81623b6b0368f41c/src/java.base/share/classes/java/util/concurrent/ConcurrentHashMap.java#L2397) does that already.
>> 
>> The code you reference is only executed if `table == null` but in this case we know it is not `null` because we only call `tryPresize()` if `table != null`.
>> 
>> In general, it's hard fro me to understand what `tryPresize(int size)` is doing and if its `size` argument is supposed to contain the additional number of elements which will be inserted into the hash map or if it is a hint for the new total size of the hash map. Can you explain?
>
>> tryPresize(int size) is doing and if its size argument is supposed to contain the additional number of elements which will be inserted into the hash map or if it is a hint for the new total size of the hash map
> 
> Argument `size` for `tryPresize()` is a hint for the **new total size** of the hash map. If the size is too small compared to the current size of the map, there will be no resizing.

Thanks. In that case, calling `tryPresize()` with `size < this.size()` doesn't make sense and we should call it with `this.size() + m.size()`.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/17116#discussion_r1450063503