This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: compile question about dll of windows
- From: Danny Smith <danny_r_smith_2001 at yahoo dot co dot nz>
- To: Andrew Haley <aph at redhat dot com>, Anthony Green <green at redhat dot com>
- Cc: gnustuff at thisiscool dot com, Danny Smith <danny_r_smith_2001 at yahoo dot co dot nz>, java at gcc dot gnu dot org
- Date: Sat, 12 Apr 2003 07:20:21 +1000 (EST)
- Subject: Re: compile question about dll of windows
--- Andrew Haley <aph at redhat dot com> wrote: > Anthony Green writes:
> > On Fri, 2003-04-11 at 00:14, Mohan Embar wrote:
> > > I suspect that when Anthony originally did this,
> > > there might have been less object files and he didn't run into any
> command-line
> > > limitations.
> >
> > I'm sorry, but I haven't been following this thread very closely. I did
> > my experiments on Linux, and tested the DLL under wine. Aren't the
> > command-line limitations unique to Windows hosted tools?
>
> What I don't understand is why gcj has such a limitation and ld doesn't.
>
Probably a limitation in CRT spawn which is used in driver to pass
argumentments to next process. I belive spawn family of functions are wrappers
for CreateProcess. See execute in gcc.c and pexecute in libiberty.
There is/was another memory problem in ld. Even when ld is invoked directly,
processing large numbers of exports is ___slow___ and sometimes exhaisteed
virtual memory.
Danny
> Andrew.
http://mobile.yahoo.com.au - Yahoo! Mobile
- Check & compose your email via SMS on your Telstra or Vodafone mobile.