Re: separated debuginfo patch

[graydon hoare <graydon at redhat dot com>]

Daniel Jacobowitz <drow@mvista.com> writes:

> On Fri, Jul 18, 2003 at 09:42:18AM -0400, Andrew Cagney wrote:
> > Philippe,
> > >
> > >[trying to avoid crc'ing the separate debug file]
> > >
> > >I need to know how GDB guys want I deal with the gdb part, for now
> > >gdb.diff just remove (#if 0) all duplicated code from bfd and use
> > >bfd_follow_gnu_debuglink() to retrieve the debug info file. Is it
> > >ok to remove this code or must I update the duplicated code according
> > >to the change in bfd ? 
> > 
> > I just wonder if it should eventually be made more transparent?
> >    bfd_openr (file, FOLLOW_DEBUG_LINK).
> > Doing things like:
> > 	objdump --follow-debug-link
> > would then become possible.  Regardless, it makes sense to put the 
> > algorighm in BFD.
> 
> It should.  I talked Graydon into trying this at the time he submitted
> the BFD part - it turned out to be a world of trouble given BFDs
> current data structure, and we bailed out.

yeah, we gave up on "transparent" access in bfd_openr due to the
ugliness of merging symbols from separate bfds; nonetheless the
requisite (distinct) debuglink-following function was added upstream.

http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/bfd/opncls.c.diff?r1=1.13&r2=1.14&cvsroot=src&f=h

-graydon