Re: What is different about the way Apple builds GCC ?
site_archiver@lists.apple.com Delivered-To: darwin-dev@lists.apple.com On Sep 23, 2005, at 6:34 AM, Sam Hart wrote: ld: Undefined symbols: ___guard ___stack_smash_handler ~ boyd Boyd Waters National Radio Astronomy Observatory Socorro, New Mexico _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.appl... This built fine the fsf-way and install and work correctly but would fail when building the Apple way with : I have built GCC compilers with SSP (stack-smashing protection) on 32- bit, x86 Gentoo Linux; this often required patches to the C library glibc. And I worked to compile a number of GCC 4.0 CVS builds last Spring, as I wanted the gfortran... but I haven't done anything recently. I didn't know that SSP was enabled under Darwin, particularly Darwin PPC. The patch from reference 4 of your email turns SSP on by default, I think (normally via --enable-stack-protector argument to ./ configure, but this seems to turn it on by default). I see also that GCC 4.1 stage 2 include a "Reimplementation of IBM's Pro Police Stack Detector" but I can't find any discussion of such. In any event, GCC CVS HEAD seems to be rather unstable at the moment on PPC Darwin? Or is that unfair? This email sent to site_archiver@lists.apple.com
participants (1)
-
Boyd Waters