Re: [PATCH] uprobes: Use synchronize_rcu() not synchronize_sched()

From: Steven Rostedt
Date: Fri Aug 10 2018 - 09:44:18 EST

[ Removing Jovi as his email is bouncing ]

On Fri, 10 Aug 2018 15:36:08 +0200
Oleg Nesterov <oleg@xxxxxxxxxx> wrote:

> > > Can't we change uprobe_trace_func() and uretprobe_trace_func() to use
> > > rcu_read_lock_sched() instead? It is more cheap.
> >
> > Is it? rcu_read_lock_sched() is a preempt_disable(),
> which is just raw_cpu_inc()
> > where
> > rcu_read_lock() may just be a task counter increment.
> and __rcu_read_unlock() is more heavy.
> OK, I agree, this doesn't really matter.

Are you OK with this patch? I set it for stable and plan on pushing it
with the patches for the upcoming merge window. If you are OK, mind
giving me an Acked or Reviewed-by?

> > > Hmm. probe_event_enable() does list_del + kfree on failure, this doesn't
> > > look right... Not only because kfree() can race with list_for_each_entry_rcu(),
> > > we should not put the 1st link on list until uprobe_buffer_enable().
> > >
> > > Does the patch below make sense or I am confused?
> >
> > I guess the question is, if it isn't enabled, are there any users or
> > even past users still running.
> Note that uprobe_register() is not "atomic".
> To simplify, suppose we have 2 tasks T1 and T2 running the probed binary.
> So we are going to do install_breakpoint(T1->mm) + install_breakpoint(T2->mm).
> If the 2nd install_breakpoint() fails for any reason, _register() will do
> remove_breakpoint(T1->mm) and return the error.
> However, T1 can hit this bp right after install_breakpoint(T1->mm), so it
> can call uprobe_trace_func() before list_del(&link->list).
> OK, even if I am right this is mostly theoretical.

Even if it is theoretical, we should make sure it can't happen. But
this is unrelated to the current patch, and if we should fix this, then
it can be a separate patch. I don't think your change hurts, and even
if it can't technically happen, it may let us sleep better at night.
Want to send a formal patch to make this change?

-- Steve