[Gcl-devel] plt.c and Mach-O ABI

[Aurelien Chanudet]

Just a quick note. Nothing really worth Camm's attention here as I 
understand that the plt trick is a temporary solution.

In the Mach-O ABI, the gcc prepends a leading underscore to any C 
symbol name (however, symbols declared in raw assembler don't get this 
leading underscore prepended). The real symbol name is the one with the 
leading underscore (i.e. the mangled one). This means for instance that 
it's not possible to use nm output to declare C function prototypes 
without further parsing.

Here is what I get on Debian :

address@hidden:/tmp$ uname -a
Linux aljunied 2.4.18-newpmac #1 Thu Mar 14 22:44:49 EST 2002 ppc 
unknown
address@hidden:/tmp$ cat toto.c
#include <stdio.h>
int main ()
{
printf ("hello, world!\n");
exit (0);
return 0;
}
address@hidden:/tmp$ gcc -c toto.c
address@hidden:/tmp$ nm toto.o
         U exit
00000000 t gcc2_compiled.
00000000 T main
         U printf
address@hidden:/tmp$ exit
logout


And here is what I get on Darwin :

-bash2.05b /tmp $ uname -a
Darwin sabine.local 7.2.0 Darwin Kernel Version 7.2.0: Thu Dec 11 
16:20:23 PST 2003; root:xnu/xnu-517.3.7.obj~1/RELEASE_PPC  Power 
Macintosh powerpc
-bash2.05b /tmp $ cat toto.c
#include <stdio.h>
int main ()
{
printf ("hello, world!\n");
exit (0);
return 0;
}
-bash2.05b /tmp $ gcc -c toto.c
-bash2.05b /tmp $ nm toto.o
         U _exit
00000000 T _main
         U _printf
         U dyld_stub_binding_helper


Note the difference in the way symbol names are mangled. I've tried to 
unravel the logic in plt.c, but I cannot compile it as such. Anyway, 
I'm not sure I need this file plt.o due to substantial differences in 
the ELF and Mach-O linking semantics.

Aurelien