Andy Dougherty schrieb: > On Mon, 29 Nov 2010, Eric Brine wrote: > >> On Fri, 26 Nov 2010, perlbug @ plan9 . de wrote:> so... please >> please please make -fstack-protector configurable somehow, >> > better yet, don't override user-specified flags and/or improve >> the tets >> > for platform support. >> If Perl only supplements, then -fno-stack-protector should work, right? Does >> it? > > Yes, good call. Configure even actually already contains code to > explicitly deal with this situation. (Thanks, Nicholas!) Explicitly > adding -fno-stack-protector to ccflags will cause Configure to not add > -fstack-protector. This will fix the immediate problem. > > Still, I agree that a test file that made this happen automatically would > be even better. I just don't know what such a test might look like. And just to add to the mix: Using -fstack-protector twice, because it's in CCFLAGS and LDFLAGS, because we call the compiler and linker seperately sometimes (GNUMakefile vs. ExtUtils::Embed), will cause a gcc crash. So if I fold the compiler and linker flags together with the typical compile+link command, I explicitly have to remove -fstack-protector from LDFLAGS; in B::C cc_harness. $ perl -V:ccflags -V:ldflags ccflags='-DPERL_USE_SAFE_PUTENV -U__STRICT_ANSI__ -g3 -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include'; ldflags=' -Wl,--enable-auto-import -Wl,--export-all-symbols -Wl,--enable-auto-image-base -fstack-protector -L/usr/local/lib'; -- Reini Urban http://phpwiki.org/ http://murbreak.at/Thread Previous | Thread Next