Re: Oops caused by race between livepatch and ftrace

From: Joe Lawrence
Date: Mon May 20 2019 - 16:49:11 EST

[ fixed jeyu's email address ]

On 5/20/19 3:49 PM, Johannes Erdfelt wrote:
[ ... snip ... ]

I have put together a test case that can reproduce the crash using
KVM. The tarball includes a minimal kernel and initramfs, along with
a script to run qemu and the .config used to build the kernel. By
default it will attempt to reproduce by loading multiple livepatches
at the same time. Passing 'test=ftrace' to the script will attempt to
reproduce by racing with ftrace.

My test setup reproduces the race and oops more reliably by loading
multiple livepatches at the same time than with the ftrace method. It's
not 100% reproducible, so the test case may need to be run multiple

It can be found here (not attached because of its size):

Hi Johannes,

This is cool way to distribute the repro kernel, modules, etc!

These two testing scenarios might be interesting to add to our selftests suite. Can you post or add the source(s) to livepatch-test<n>.ko to the tarball?

The simple patch of extending the module_mutex lock over the entirety
of klp_init_object_loaded fixes it from the livepatch side. This
mostly works because set_all_modules_text_{rw,ro} acquires module_mutex
as well, but it still leaves a hole in the ftrace code. A lock should
probably be held over the entirety of remapping the text sections RW.

This is complicated by the fact that remapping the text section in
ftrace is handled by arch specific code. I'm not sure what a good
solution to this is yet.

A lock or some kind of referencing count.. I'll let other folks comment on that side of the bug report.


-- Joe