harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Hindess <mark.hind...@googlemail.com>
Subject Re: [classlib] Patch for H-3148 (OOM when using native memory which is out of GC-control)
Date Sat, 14 Apr 2007 19:14:56 GMT

On 13 April 2007 at 16:53, "Nathan Beyer" <ndbeyer@apache.org> wrote:
> I don't have a particular desire to grab this one, just want to keep
> this moving.
>
> I guess a better question would be, has anyone else tried out this
> patch? Good? Bad? No difference?

I took a (very) quick look a day or so ago.  The code looked good.
(Though I'd probably have refactored the platform-specific native code
to hysysinfo.c which has similar code - I'm happy to do this later when
I'm less busy with porting to classlib to Aix.)

However, I couldn't actually get the regression test to fail whether I
applied the rest of the patch or not so it was hard to trivially check
if it had really improved things.  (This could be fixed later too I
suppose.)

Regards,
 Mark.


> On 4/13/07, Tim Ellison <t.p.ellison@gmail.com> wrote:
> > Nathan Beyer wrote:
> > > It looks like Tim took assignment of this issue. Tim, are you working
> > > on this one?
> >
> > If you have time to take it then please go ahead and steal it from me.
> > It may be a few days before I get to study it properly (I guess my main
> > concern would be performance impact of a memory check on each alloc).
> >
> > Regards,
> > Tim
> >
> 



Mime
View raw message