[llvm-commits] [patch][MC] Change the way we dump section_data

[Daniel Dunbar]

Sure, LGTM.

Ideally we would replace both of these with more sane dumping tools.
FWIW, the original motivation behind macho-dump was because I wanted a
pure representation of every byte in the object file, since we were
shooting for bitwise equivalence with Darwin 'as'. Now that we have
achieved that, it would definitely make sense to move to more readable
output formats -- especially if we get a real llvm-objdump.

Thanks,
 - Daniel

On Fri, Sep 10, 2010 at 11:38 AM, Rafael Espindola <espindola at google.com> wrote:
> While working on the LEB128 patch I found the output of elf-dump and
> macho-dump for section_data hard to read. The problem is that "%r"
> works best for mostly ascii data. Since that is normally the case, we
> get things like "\xc88\xf4". Note that the second '8' is the ascii
> char '8'. With this patch we get "c838f4" with a space every 4 bytes.
> IMHO it makes the output a lot easier to decode.
>
> Cheers,
> --
> Rafael Ávila de Espíndola
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
>