This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
RE: Cygwin and hash synchronization?
- From: "Boehm, Hans" <hans_boehm at hp dot com>
- To: "'Adam Megacz'" <gcj at lists dot megacz dot com>, java at gcc dot gnu dot org
- Date: Thu, 30 May 2002 10:56:02 -0700
- Subject: RE: Cygwin and hash synchronization?
Is there a real patch implementing this? I think the approach outlined in the message is correct, but the correctness argument is subtle and requires some documentation. In particular, we may end up registering a finalizer using the truncated object address. But that's OK because finalizers on static data are ignored (since they wouldn't ever run anyway). And we assume that only static data is misaligned.
Hans
> -----Original Message-----
> From: Adam Megacz [mailto:gcj@lists.megacz.com]
> Sent: Thursday, May 30, 2002 9:47 AM
> To: java@gcc.gnu.org
> Subject: Re: Cygwin and hash synchronization?
>
>
>
> Bryce McKinlay <bryce@waitaki.otago.ac.nz> writes:
> > So, does hash synchronization work on cygwin or not?
>
> Not yet.
>
>
> > I seem to remember someone saying the cygwin binutils doesn't
> > support alignment of static data correctly, which would be
> > consistent with this report:
>
> Yes, I've confirmed/reported this before.
>
>
> > And what about mingw? Does the linker for mingw not suffer from this
> > problem?
>
> Cygwin is just mingw with an extra DLL linked in automatically. They
> use the same binutils.
>
>
> > Basically I just want to know if this patch should be applied,
>
> Go ahead. Hans has an easy fix for this which I have already tested
> and verified.
>
> http://gcc.gnu.org/ml/java/2002-03/msg00038.html
>
> - a
>
>
> --
> Sick of HTML user interfaces?
> www.xwt.org
>