RE: [Gcl-devel] BFD related issues - belated reply.

[Mike Thomas]

Hi Aurelian.

I'm gradually catching up on emails from the past couple of months and came
across this one between Camm and yourself.

> 1> Hmmm. Might still be useful, no? Your opinion? The other route is
> > to dig into the internals of why relocate_section works on platforms
> > where bfd_get_relocated_section_contents does not, and put in the
> > patches to fix the latter. Though it is the 'old' way, I think
> > binutils upstream would still accept such patches. The question is,
> > which is less work?

> Hmmm. Would it be possible for me to have remote access to such an
> architecture where bfd_get_relocated_section_contents does not work ?

Probably not what you wanted to hear, but Windows GCL has a problem with
bfd_get_relocated_section_contents().

Are you interested in helping me?

There is a file called "bfdtest.c" in the top level of our CVS repository to
demonstrate - eg:

gcc bfdtest.c -o bfdtest.exe -lbfd -liberty
./bfdtest ./unixport/raw_pre_gcl.exe gmp3/version.o


.....
DOING SOMETHING WITH THE HASHED SYMBOLS

u = 32, v = 7
u loop q[0]->name = version.c
u loop q[1]->name = .text
u loop q[2]->name = .data
u loop q[3]->name = .bss
u loop q[4]->name = .stab
u loop q[5]->name = .stabstr
u loop q[6]->name = ___gmp_version


DOING RELOCATIONS

s->name .text, s->flags = 237
About to get reloc section contents
obj_bfd = 3d4c88, section .text, s->output_section = 8d95f4, q = 22fef4
.... BARF HERE ...


Cheers

Mike Thomas.