Hello! On Fri, Oct 10, 2008 at 10:56:34AM +0200, I wrote: > On Wed, Oct 08, 2008 at 11:53:57AM +0200, I wrote: > > First, the check for gcc_cv_libc_provides_ssp is not complete, as has > > already pointed out (with patches!) before, but is still not fixed on > > trunk. Let me revisit that: in configure.ac it is being checked for > > ``case "$target" in *-*-linux*)'' which should rather match ``*-*-linux* > > | *-*-*-gnu* | *-*-gnu*'' to catch all GNU/Linux, GNU/Hurd, GNU/k*BSD > > systems. (This is being patched in the Debian GCC package for some time > > already.) > > Even though the NATIVE_SYSTEM_HEADER_DIR issue is still being discussed, > in either way we'll need this patch installed. Any objections? > > 2008-10-10 Thomas Schwinge <[EMAIL PROTECTED]> > > * configure.ac (gcc_cv_libc_provides_ssp): Also consider GNU/kFreeBSD, > GNU/kNetBSD and GNU/Hurd systems.
Here is a better variant. Paolo (you had already acked the previous version), but is this OK to install into trunk (see the additional Hurd-only change)? Index: configure.ac =================================================================== --- configure.ac (Revision 141837) +++ configure.ac (Arbeitskopie) @@ -3474,7 +3474,7 @@ gcc_cv_libc_provides_ssp, [gcc_cv_libc_provides_ssp=no case "$target" in - *-*-linux*) + *-*-linux* | *-*-kfreebsd*-gnu | *-*-knetbsd*-gnu) if test x$host != x$target || test "x$TARGET_SYSTEM_ROOT" != x; then if test "x$with_sysroot" = x; then glibc_header_dir="${exec_prefix}/${target_noncanonical}/sys-include" @@ -3510,6 +3510,12 @@ fi fi ;; + *-*-gnu*) + # Avoid complicated tests (see + # <http://gcc.gnu.org/ml/gcc/2008-10/msg00130.html>) and for now + # simply assert that glibc does provide this, which is true for all + # realistically usable configurations. + gcc_cv_libc_provides_ssp=yes;; *-*-darwin*) AC_CHECK_FUNC(__stack_chk_fail,[gcc_cv_libc_provides_ssp=yes], [echo "no __stack_chk_fail on this target"]) Regards, Thomas
signature.asc
Description: Digital signature