Re: RFA: ia64 portion of libunwind patch

[Marcel Moolenaar <marcel@xcllnt.net>]

On Fri, Oct 24, 2003 at 11:56:25AM -0700, Kevin Buettner wrote:
> >
> > >>+#ifdef HAVE_LIBUNWIND_IA64_H
> > >>+
> > >>+# ifndef __NR_getunwind
> > >>+#  define __NR_getunwind	1215
> > >>+# endif
> > > 
> > > Is this part still needed?
> > 
> > Not if we include <sys/syscall.h>
> 
> I would prefer to see <sys/syscall.h> included -- but not in
> ia64-tdep.c.  This include and the code for ia64_getunwind() will have
> to go in ia64-linux-nat.c.  Presumably, if a remote target wanted to
> use the unwind library, there'd need to be some remote protocol
> modifications.

Isn't the unwind library linked into gdb and using callbacks for
memory reads and obtaining register values so that we can use the
existing protocol commands to extract information from the inferior?

Hmmm... this raises the question how the unwind library knows about
dynamic unwind descriptors in remote targets...

> Also, apparently, there'll be some FreeBSD patches for IA-64 coming
> down the pike.  FreeBSD, if it chooses to use the libunwind library,
> will need to implement its own version of ia64_getunwind().

Yes, I intend to have an unwind library on FreeBSD. I'm happy to see
that support is being added. Standardizing on David's unwind library
is possible now that it has the MIT license. Unfortunately it depends
on many GNU build tools (autoconf, automake, libtool etc). This may
be a reason for me to use an API compatible alternative as I expect it
to be a PITA to import libunwind into FreeBSD and have it play nice
with the BSD make infrastructure. Also, my first attempts to port it
to FreeBSD weren't as successful as I hoped it would be. There's a
certain amount of Linuxism in it that didn't (near) trivially map to
FreeBSDisms. Anyway: I'll be talking to David when the time comes :-)

-- 
 Marcel Moolenaar	  USPA: A-39004		 marcel@xcllnt.net