Looks like the change to how /proc/kallsyms displays addresses was introduced in:
commit d97106ab53f812910a62d18afb9dbe882819c1ba
Author: Linus Torvalds <email address hidden>
Date: Sat Jan 3 11:46:17 2009 -0800
Make %p print '(null)' for NULL pointers
Upstream fixed it in 3a129cc2151425e5aeb69aeb25fbc994ec738137 but was not originally submitted to linux-stable (that has been rectified).
It could be worked around in the qrt script, but I'm not convinced that's the right thing to do here.
It should be noted that the behavior that the script is testing for, kernel addresses in /proc/kallsyms not being visible to unprivileged processes, is correct here. So there's no security impact from this, and it should not block promoting kernels, from the security team's perspective.
Looks like the change to how /proc/kallsyms displays addresses was introduced in:
commit d97106ab53f8129 10a62d18afb9dbe 882819c1ba
Author: Linus Torvalds <email address hidden>
Date: Sat Jan 3 11:46:17 2009 -0800
Make %p print '(null)' for NULL pointers
Upstream fixed it in 3a129cc2151425e 5aeb69aeb25fbc9 94ec738137 but was not originally submitted to linux-stable (that has been rectified).
It could be worked around in the qrt script, but I'm not convinced that's the right thing to do here.
It should be noted that the behavior that the script is testing for, kernel addresses in /proc/kallsyms not being visible to unprivileged processes, is correct here. So there's no security impact from this, and it should not block promoting kernels, from the security team's perspective.