Re: PATCH: Re: Partial symbol export vs --export-dynamic

[Jean-Francois Panisset <panisset at discreet dot com>]

 <20010617180729.A24660@lucon.org>"H . J . Lu" writes
>On Sun, Jun 17, 2001 at 03:48:02AM -0400, Jean-Francois Panisset wrote:
>> >
>> >If you know something about bfd, it shouldn't be too hard to implement.
>> >Personally, I think it is a useful feature. I may do it myself if we all
>> >agree it is a good feature to have and how it should work.
>> >
>> >
>> >H.J.
>> 
>
>It turned out it was quite easy to do. Here is a patch. I just expanded
>the symbol versioning to executables. You can use
>
># gcc -Wl,-export-dynamic,--version-script,version.map ...
>
>You can use the version-script to control what you want to export. It
>supports regular expression, C++ symbol, Java, ... among other things.
>
>
>H.J.
>----
>2001-06-17  H.J. Lu  <hjl@gnu.org>
>
>	* elflink.h (elf_info_failed): Add a new field, verdefs.
>	(NAME(bfd_elf,size_dynamic_sections): Pass verdefs to
>	elf_export_symbol.
>	(elf_export_symbol): Check eif->verdefs to decide if a symbol
>	should be exported.

Well, it seems the patch works fine. I have created the following
version script:

VERS_1.0 {
global:
api_func_1;
api_func_2;
...
local:
*;
};

and that seems to do the trick: only the API symbols are visible
to plugins, my binary size doesn't grow by 5MB, and my symbol conflict
seems to be gone.

Thanks you very much!
JF