This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [patch/rfc;rfa:symtab] Deprecate inside_entry_file
Andrew Cagney writes:
> > Andrew Cagney <ac131313@redhat.com> writes:
> >
> >> Before thinking about committing this I'd like to see comments on the
> >> symtab side.
> >
> >
> > The changes in mdebugread.c, dwarfread.c, dwarf2read.c, dbxread.c, and
> > coffread.c are just renamings. Was there something more specific that
> > you wanted reviewed?
>
> I wrote:
>
> > This patch deprecates both the function inside_entry_file() and the object file variables that support it (I don't want there to be any confusion over this :-).
>
> Are the symtab maintainers ok with this?
>
> Andrew
>
In particular note that the function:
symfile.c:init_entry_point_info() which sets up all the
objfile->ei.deprecated_entry_file* is seriously screwed up for PIE
files anyway.
In such a case we need to figure out where the program got loaded,
just like a shlib, the main executable is not marked as EXEC in the
elf header anymore, and lots of other fun stuff. Anyway, the way the
entry file/entry point stuff is handled now won't work for PIE.
I am currently working on PIE support, BTW. I'll post something as
soon as I have some coherent story.
elena