This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: [PATCH 3/6] Add Visium support to bfd
- From: Eric Botcazou <ebotcazou at adacore dot com>
- To: binutils at sourceware dot org
- Cc: Alan Modra <amodra at gmail dot com>
- Date: Fri, 05 Dec 2014 11:20:39 +0100
- Subject: Re: [PATCH 3/6] Add Visium support to bfd
- Authentication-results: sourceware.org; auth=none
- References: <cover dot 1417646850 dot git dot ebotcazou at gcc dot gnu dot org> <ef3c05f1f9cfd35d7b047b2cd53f867bc6bbeeca dot 1417646850 dot git dot ebotcazou at gcc dot gnu dot org> <20141204004526 dot GU20383 at bubble dot grove dot modra dot org>
> Lose "set_" in the name, you're not setting anything here.
Renamed into visium_parity_bit.
> Also p ^= might save some future trouble, since bfd_vma can be 64-bit.
Right, I agree that the ^= idiom is clearly superior here. Also changed in
the assembler (visium_update_parity_bit).
> > + sym_hashes_end =
> > + sym_hashes + symtab_hdr->sh_size / sizeof (Elf32_External_Sym);
> > + if (!elf_bad_symtab (abfd))
> > + sym_hashes_end -= symtab_hdr->sh_info;
>
> You won't ever have elf_bad_symtab set and in any case, sym_hashes_end
> looks to be unused.
Removed.
> > + There is some attempt to make this function usable for many
> > architectures, + both USE_REL and USE_RELA ['twould be nice if such a
> > critter existed], + if only to serve as a learning tool.
>
> Really?
No. :-) Head comment ditched (and ATTRIBUTE_UNUSED removed from output_bfd).
Thanks for the review.
--
Eric Botcazou