Re: [RFC PATCH 2/2] livepatch: Clear relocation targets on a module removal

[Josh Poimboeuf]

On Thu, Sep 05, 2019 at 02:16:51PM +0200, Miroslav Benes wrote:
> > > > A full demo would require packaging up replacement .ko's with a livepatch, as
> > > > well as "blacklisting" those deprecated .kos, etc.  But that's all I had time
> > > > to cook up last week before our holiday weekend here.
> > > 
> > > Frankly, I'm not sure about this approach. I'm kind of torn. The current 
> > > solution is far from ideal, but I'm not excited about the other options 
> > > either. It seems like the choice is basically between "general but 
> > > technically complicated fragile solution with nontrivial maintenance 
> > > burden", or "something safer and maybe cleaner, but limiting for 
> > > users/distros". Of course it depends on whether the limitation is even 
> > > real and how big it is. Unfortunately we cannot quantify it much and that 
> > > is probably why our opinions (in the email thread) differ.
> > 
> > How would this option be "limiting for users/distros"?  If the packaging
> > part of the solution is done correctly then I don't see how it would be
> > limiting.
> 
> I'll try to explain my worries.
> 
> Blacklisting first. Yes, I agree that it would make things a lot simpler, 
> but I am afraid it would not fly at SUSE. Petr meanwhile explained 
> elsewhere, but I don't think we can limit our customers that much. We 
> perceive live patching as a product as much transparent as possible and as 
> less intrusive as possible. One thing is to forbid to remove a module, the 
> other is to forbid its loading.
> 
> We could warn the admin. Something like "there is a fix for a module foo, 
> which is not loaded currently. It will not be patched and the system will 
> be still vulnerable if you load the module unless a new fixed version is 
> provided."

No.  We just distribute the new .ko with the livepatch.  It should be
transparent to the user.

> Yes, we can distribute the new version of .ko with a livepatch. What is 
> the reason for blacklisting then? I don't probably understand, but either 
> a module is loaded and we can patch it (without late module patching), or 
> it is not and we could replace .ko on disk.

I think the blacklisting is a failsafe to prevent the old module from
accidentally getting loaded after patching.

> Now, I don't think that replacing .ko on disk is a good idea. We've 
> already discussed it. It would lead to a maintenance/packaging problem, 
> because you never know which version of the module is loaded in the 
> system. The state space grows rather rapidly there.

What exactly are your concerns?

Either the old version of the module is loaded, and it's livepatched; or
the new version of the module is loaded, and it's not livepatched.

Anyway that could be reported to the user somehow, e.g. report
srcversion in sysfs.

-- 
Josh