This is the mail archive of the
mailing list for the binutils project.
Re: [PATCH] BFD: Handle copying of absolute section symbols
On Mon, Apr 16, 2012 at 01:00:04PM +0100, Maciej W. Rozycki wrote:
> On Wed, 11 Apr 2012, Alan Modra wrote:
> > > I have tracked it down to ignore_section_sym in bfd/elf.c that considers
> > > the section the symbol is associated with not to be owned by the output
> > > BFD. Arguably the special absolute section should be treated as "owned"
> > > by any BFD, because any symbols defined there are really detached from any
> > > input section and the only sensible action that can be done about them
> > > here is to copy them verbatim.
> > I agree that absolute section syms should be kept, but wonder whether
> > we're treating global section syms correctly. They probably ought to
> > be kept too.
> Excuse me, but I fail to see where we exclude global section syms --
> there's nothing in ignore_section_sym or elf_map_symbols that looks to me
> like eliminating them. For the avoidance of doubt -- does your your
> answer count as approval for my change?
You had this:
63: 0000000000000001 0 SECTION GLOBAL DEFAULT ABS _DYNAMIC_LINKING
and found it was being dropped.
Your solution was to keep ABS symbols. An alternative solution for
your particular problem is to ensure global symbols are kept. I'm
saying we should do both.
Australia Development Lab, IBM