RE: [cp-patches] Some VMStackWalker cleanups/simplifications

[Jeroen Frijters]

Mark Wielaard wrote:
> I think that is a little strong.

I apologize if I come across too strong, it's just that I really care
about doing the right thing (a little too much sometimes), so when
someone proposes a IMNSHO bad idea, it sometimes affects me a little too
much.

> I think it makes the interface a lot
> easier, robust and hopefully more efficient to implement.

I have so many different problems with it that it is hard to discuss it
for me. Let me start by saying that I'm pretty much convinced that
passing the caller class isn't actually solving any problem, so I would
appreciate it if you could try to explain exactly what problem it is
that you are addressing. Second, I object to the notion that this makes
the interface easier and more robust, because it allows a caller to pass
in arbitrary classes, what is the defined behavior in such a case?

> > Of course it adds overhead, especially with a 1.4 compiler.
> 
> Aha. OK, I see where you are comming from. I assumed SomeClass.class
> would be compiled efficiently, and I think at least jikes 
> gets it right by caching the result, but I see gcj doesn't do that and

> actually calls Class.forName() each time (eep).

That's not really what I meant. My main concern is that it is very hard
to statically determine the class that is passed and thus it's hard to
optimize it away. With a 1.5 compiler at least that would be easier.

> So I should actually change the patch
> so that it really is a fixed parameter at each callsite by
> precalculating it. Will do that.

Please don't.

Regards,
Jeroen