Re: missing __init_array_start, __init_array_end in __libc_csu_init

[Michael Schmitz <schmitz@mail.biophys.uni-duesseldorf.de>]

> This probably doesn't have any relevance, however it seems that fortran
> segfaults when building r-base, except on akire, which may be the only
> buildd still running a 2.4 kernel. I probably need to validate this with
> aranym.

hobbes and washi also run 2.4.

> Unfortunately, -pie is still a problem.

Looking back over the glibc build logs, it seems -pie never worked, at
last not since 2005. The m68k sysdeps/elf/start.S does not have a separate
#if SHARED section as seen in the hppa patch (and in the i386 start.S)
- does this matter?

csu/elf-init.c says:

/* These magic symbols are provided by the linker.  */
extern void (*__preinit_array_start []) (int, char **, char **)
  attribute_hidden;
extern void (*__preinit_array_end []) (int, char **, char **)
  attribute_hidden;
extern void (*__init_array_start []) (int, char **, char **)
  attribute_hidden;
extern void (*__init_array_end []) (int, char **, char **)
  attribute_hidden;
extern void (*__fini_array_start []) (void) attribute_hidden;
extern void (*__fini_array_end []) (void) attribute_hidden;

so it would appear our binutils are to blame? Or gcc?

	Michael