Re: [rfa] Add bfd_runtime

[Andrew Cagney <cagney at gnu dot org>]

> Andrew Cagney <cagney@gnu.org> writes:
>
>
> > > Let me see if I've got this:
> > >
> > > objfile - iterate over sections
> > > corefile - iterate over segments
> > > archive - ???
>
>
> archive - iterate over objfiles

(I wondered)

But what of this:

> For runtime, bfd_map_over_sections needs to do something different again create a list of sections using offset information obtained from the segment table.
>
> That can either be done in two stages; reverse map in-memory segments to on-disk image, open pseudo on-disk image as an object file; or a single direct stage where the "sections" describe the in memory offsets.
>
> The latter, which I think is an operation unlike any of the above, is what I'm trying to implement.

Do you concure that the operation is unlike any of the above?

> Daniel Jacobowitz <drow@false.org> writes:
>
>
> > > > > > > So how would you solve this problem?  Given a memory access method and
> > > > > > > a starting offset, construct a bfd containing a list of sections
> > > > > > > constructed using both the segment and section information in the
> > > > > > > inferior?
> > >
> > > > > 
> > > > > I would write a new BFD target vector; e.g., elf32-i386-runtime.
> >
> > > Ugh, is that really necessary?  It would mean architecture-specific
> > > code to support this generic ELF concept, and I don't see any useful
> > > hooks in the elf backend vector anyway...

Yep, ulgh.

> I'm not arguing for architecture-specific code.  The code should be
> written in a generic way to support any architecture, and
> elf32-i386-runtime would just be a modification of elf32-i386 which
> called this new generic code to dig up the runtime information.

How?  Without adding to BFD's existing macro #include hell?  This is 
needed for all elf object formats.

> My argument is just that we should have a different target vector for
> this type of thing.  I'm basing this on Andrew's description of what
> is needed, and on his arguments for introducing bfd_runtime as a
> parallel to bfd_object, et. al.

Andrew