On Tue, 13 Feb 2001, Chris Gilbert wrote:

> On Tuesday 13 February 2001  1:43 am, Ben Harris wrote:
> > On Sun, 11 Feb 2001, Chris Gilbert wrote:
> > > As requested:
> > >  ktruss ./sh /dev/null
> > >                                                    = 0
> > >    218 ktruss   emul(netbsd)
> > >    218 ktruss   execve("./sh", 0xefbfdc2c, 0xefbfdc38) JUSTRETURN
> > >    218 sh       emul(netbsd)
> > > SIGILL SIG_DFL
> > >    218 sh       "sh.core"
> >
> > Having finally got myself into a position to test this on arm26, I seem to
> > be getting much the same problem.  I'll hunt it down when I've got
> > time.  Maybe I should get core dumps working on arm26 while I'm there.
> 
> That's a good sign, means we're like for like on bugs and core dumping.  
> Could this be something to do with the structure padding thing you mentioned 
> in another thread?

No, because my previous arm26 toolchain used the same padding.  Similarly,
it's probably not a bug in the kernel's ELF support, because that hasn't
changed.  I suspect it's a problem in GCC (my old one was egcs, this is
2.95.2) or the linker script.  Anyway, I'll try to work out what's going
on tonight.

-- 
Ben Harris                                                   <bjh21@netbsd.org>
Portmaster, NetBSD/arm26               <URL:http://www.netbsd.org/Ports/arm26/>