On Fri, Feb 19, 2021 at 02:36:07PM -0500, Steven Rostedt wrote:
> On Thu, 18 Feb 2021 23:29:23 +0900
> Masami Hiramatsu <mhiramat@xxxxxxxxxx> wrote:
>
> > Commit 36dadef23fcc ("kprobes: Init kprobes in early_initcall")
> > moved the kprobe setup in early_initcall(), which includes kprobe
> > jump optimization.
> > The kprobes jump optimizer involves synchronize_rcu_tasks() which
> > depends on the ksoftirqd and rcu_spawn_tasks_*(). However, since
> > those are setup in core_initcall(), kprobes jump optimizer can not
> > run at the early_initcall().
> >
> > To avoid this issue, make the kprobe optimization disabled in the
> > early_initcall() and enables it in subsys_initcall().
> >
> > Note that non-optimized kprobes is still available after
> > early_initcall(). Only jump optimization is delayed.
> >
> > Fixes: 36dadef23fcc ("kprobes: Init kprobes in early_initcall")
> > Reported-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
> > Signed-off-by: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
> > Cc: stable@xxxxxxxxxxxxxxx
> > ---
>
> I pulled this into my queue to be tested, and when that completes
> (hopefully without failure), I'll add this to my pull request for the
> current merge window (which I still need to send).
Thank you, Steve!
Could you please add the following Reported-by tags?
Reported-by: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
Reported-by: Uladzislau Rezki <urezki@xxxxxxxxx>
Sebastian first noticed the problem, and Uladzislau figured out
how softirqs were involved.
Thanx, Paul
