This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
RE: [patch] bfd - PE/COFF updates and clarifications
- From: "Phil Lello" <phil dot lello at homecall dot co dot uk>
- To: "'Nick Clifton'" <nickc at redhat dot com>
- Cc: <binutils at sourceware dot org>
- Date: Wed, 15 Nov 2006 03:14:48 -0000
- Subject: RE: [patch] bfd - PE/COFF updates and clarifications
Hi Nick,
> >> 2006-09-01 Phil Lello <phil.lello@homecall.co.uk>
> >> * peXXigen.c: Updates for PE/COFF V8.0, and clarification
> >> (dir_names): Added CLR Runtime Header to dir_names[].
> >> (_bfd_XX_print_private_bfd_data_common): Added
> EFI_ROM and XBOX
> >> subsystem names
> >> (_bfd_XXi_swap_aouthdr_in, _bfd_XXi_swap_aouthdr_out)
> >> (pe_print_idata, pe_print_edata)
> >> (_bfd_XX_bfd_copy_private_bfd_data_common)
> >> (_bfd_XXi_final_link_postscript): Use #DEFINEs for index into
> >> DataDirectory
>
> >> 2006-09-05 Phil Lello <phil.lello@homecall.co.uk>
> >> * pe.h: Added defines for IMAGE_SUBSYSTEM_EFI_ROM and
> >> IMAGE_SUBSYSTEM_XBOX
> >> * internal.h: Added defines for PE directory entry types.
> >> NB: in internal.h because
> IMAGE_NUMBEROF_DIRECTORY_ENTRYIES is in
> >> pe.h
>
> I have approved and applied your patch, except for one small part:
>
> >> @@ -1214,7 +1214,8 @@ pe_print_idata (bfd * abfd, void * vfile
> >> ft_idx = first_thunk - adj;
> >> ft_allocated = 0;
> >>
> >> - if (first_thunk != hint_addr)
> >> + // PCL HACK
> >> + if (first_thunk != first_thunk/*first_thunk != hint_addr*/)
> >> {
> >> /* Find the section which contains the first thunk. */
> >> for (ft_section = abfd->sections;
>
> It is not clear to me whether this change was intentional or
> just some debugging that was left in. Either way it is
> unsuitable. If you really do want to disable the test you
> should use "if (0)" and add a comment explaining why the test
> and the body of the if statement have been suppressed.
>
Thanks for that. This last change (to pe_print_idata) was indeed
left-over debugging, and shouldn't have been in my patch. I will
make sure I check more carefully in future.
It looks like although bfd/peXXigen.c has been updated with the
other changes, the ChangeLog entry (bfd/ChangeLog) hasn't made it
to CVS yet.
Thanks,
Phil Lello