FYI, while doing pre-release testing for coreutils-7.4, I noticed a single gnulib test failure on Fedora rawhide. It looks like a bug in <string.h>'s memchr macro. Reported here:
http://bugzilla.redhat.com/499689
Quick summary:
cat > k.c <<\EOF
#include <string.h>
int main() { return !!(memchr (0, 'a', 0)); }
EOF
gcc -O k.c; ./a.out
Segmentation fault
[Exit 139 (SIGSEGV)]
It's not a problem on Fedora 10.
