[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: patch 620
From: |
Anthony Green |
Subject: |
Re: patch 620 |
Date: |
02 Dec 2002 22:18:30 -0800 |
On Thu, 2002-11-28 at 05:37, Stephen Crawley wrote:
> Here are a couple of theoretical reasons why interning (in general)
> could have negative performance implications, for Kissme at least.
>
> * I believe that Classpath's String interning uses a weak hashmap.
> Unfortunately, Kissme does not support weak references yet. Instead
> it treats weak references as ordinary references. This means that
> Kissme cannot garbage collect any String that has been interned.
I see. Well, in practice there's going to be a real limit to the number
of Locale's a program will ever see - so I don't see this as a real
problem.
> * You may be aware that the Kissme VM includes an implementation of
> orthogonal persistence. In one possible configuration of Kissme
> O-P, static fields of classes get automatically saved to the
> persistent store. This would include String's interning data
> structures. Thus interned Strings would last "for ever" in the
> store.
I think the same comment applies here. This doesn't really seem like a
bad thing.
> Note: I'm not saying "don't intern these locale Strings". If there is
> a demonstrable performance benefit, go for it.
If anywhere there was a use for string interning, I think this is it.
We're talking about up to three inline pointer comparisons -vs- up to
three expensive String comparison function calls on each Locale
comparison, and there's a real and practical limit to the number of
strings we'd be interning.
Tom - what would you like to see in order to accept this patch?
AG
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Re: patch 620,
Anthony Green <=