Re: pef.c:bfd_pef_object_p breaks other targets

[Klee Dienes <klee at apple dot com>]

Sorry about that.  We have local patches to bfd_check_format_matches() 
that save and restore the target-specific parts of the bfd before 
calling each of the recognizers, and I neglected to check the 
ambiguous-match code on a version without this patch installed.

I've changed pef_object_p, pef_xlib_object_p, and sym_object_p to only 
trample abfd->tdata if the targets pass the magic-number check (in the 
case of pef, the first 8 bytes must be "Joy!peff"; for xsym, the first 
32 bytes must match "\013Version 3.x" followed by 20 nulls, both of 
which seem pretty safe).

2002-11-10  Klee Dienes  <kdienes@apple.com>

        * pef.c: Move version check out of bfd_pef_scan and into
        bfd_pef_object_p.  Only modify tdata if we have already 
recognized
        the file correctly.  Ditto for bfd_pef_xlib_scan and
        bfd_pef_xlib_object_p.
        * xsym.c (bfd_sym_scan): New function.  Does the real work of
        parsing an already-recognized xSYM file (moved from
        bfd_sym_object_p).
        (bfd_sym_object_p): Only modify tdata if we have already
        recognized the file correctly.
        * xsym.h (bfd_sym_scan): Add prototype.

Attachment: pefsym.txt
Description: Text document

On Thursday, November 7, 2002, at 06:13 PM, Alan Modra wrote:

> You can't just willy nilly assign to abfd->tdata.  It's a union and
> may contain valid tdata for some other format.  See elf_object_p
> for what needs to be done.