Re: Alpha vs. new ABI

[Alexandre Petit-Bianco <apbianco at cygnus dot com>]

Jeff Sturm writes:

> Below is what I ended up with.  

I came up with something similar yesterday. It tested OK on x86. It
was my intent to test it on alpha. I just came up with an Alpha build,
but it looks like you just beat me.

> I'm not very pleased with it because of the side effects of
> java_array_type_length() relying on TYPE_MAX_VALUE to remember the
> length an array was allocated with.  

Yes, through building the index for the array which is a range
comprising min and max values -- I guess it's an OK way to store the
length.

> (Since java_array_type_length is only a minor optimization, it
> could go away, but for static array generation depending on it.
> Ick.)

I agree. The optimization is gone right of now. It's a minor one, but
we should keep it -- or at least try to re-install it, it's been gone
for a while as far as I can tell.

I was going to verify things on alpha, but after a successful build,
my alpha build breaks again -- I don't understand what's happening.
But if you report things are working, then it's fine. I'm trying a
build on Linux/PPC, hopefully it'll build a little more reliably than
on Alpha. If results are coming out OK, I say we should check your
patch in.

./A