linux-kernel - Re: [PATCH RFC] Make call_srcu() available during very early boot

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Fri, 17 Aug 2018 09:31:43 -0700
From:   "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To:     Tejun Heo <tj@...nel.org>
Cc:     Steven Rostedt <rostedt@...dmis.org>, joel@...lfernandes.org,
        mathieu.desnoyers@...icios.com, peterz@...radead.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH RFC] Make call_srcu() available during very early boot

On Fri, Aug 17, 2018 at 09:08:42AM -0700, Tejun Heo wrote:
> Hello,
> 
> On Tue, Aug 14, 2018 at 10:06:18AM -0700, Paul E. McKenney wrote:
> > > Is there a way we could make a union, or reuse one of the other fields,
> > > as we know that synchronize_srcu() can't be used yet (and if it is,
> > > either warn, or just make it a nop). And when we call srcu_init() and
> > > remove the srcu_struct from the list, we can then initialize whatever
> > > we used as the temporary boot up list field.
> > 
> > I will take a look.  If nothing else, I could union it with the
> > struct work_struct, since it cannot be used that early anyway.  ;-)
> > 
> > Or I could just use the work_struct that is already inside the struct
> > work_struct.  Tejun, would you be OK with that?
> 
> Hmm... not super against it given how specialized the whole thing is
> but maybe just making it a union is cleaner?

The problem with making it be a union is that I cannnot determine how
to initialize it due to the possibility that someone will initialize it
early, but not use it until much later.  For example, if I initialize my
side of the union, but they don't actually use it until after rcu_init()
is invoked, I will end up passing a bogus work_struct to you.

In theory, I could straighten all this out at rcu_init() time, but that
would require linking all srcu_struct instances initialized at that
point, including the compile-time-initialized instances.  This can of
course be done, but not simply.

Or am I missing a trick here somewhere?

							Thanx, Paul