Re: Are there real use cases with the Java access modes?

On 7/20/21 8:27 PM, Gregg Wonderly via Concurrency-interest wrote:

> I am go to once again cease here because I appear to be pushing
> buttons with my conversation mode and I am really not trying to make
> people made or defensive.  At some level these are hard discussions
> to have, I get that, but I just really sense to much defensive
> posturing and so I really don’t believe that the discussion can
> happen without creating even more walls to have to figure out how to
> talk through.

Gregg, really. Please have some respect.

People are giving you time and explaining their reasons to you. We are
not "defensively posturing". In my case, it's sheer amazement that
something like hosting a value into a register, one of the most common
and powerful optimizations we have, is in some way weird.

It's like this: operations in a CPU act on registers. There is near-
zero cost for accessing them, so the basic arithmetic and logical
operations can deliver their results in a single cycle. Memory, on the
other hand, even cache memory, has a latency of four or five cycles,
and costs more energy too. So this optimization is a really big deal
on the billions (probably?) of computers running Java.

I'll try an analogy. Imagine someone came up with a way to make
internal combustion engines run 10% better, consume less fuel, and
produce less emissions. (It's not hard to imagine: it happens.) Then
someone else comes along and says it's terribly complicated, and Bono
in the repair shop on the corner will not be able to maintain it, so
we shouldn't do it.

This is, almost exactly, the same as your argument.

-- 
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671

_______________________________________________
Concurrency-interest mailing list
[email protected]
http://cs.oswego.edu/mailman/listinfo/concurrency-interest