linux-kernel - Re: [PATCH 2/3] kunit: Add kunit_move_action_to_top_or

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

Date:   Mon, 25 Sep 2023 11:59:02 +0200
From:   Maxime Ripard <mripard@...nel.org>
To:     David Gow <davidgow@...gle.com>
Cc:     Arthur Grillo <arthurgrillo@...eup.net>,
        David Airlie <airlied@...il.com>,
        Daniel Vetter <daniel@...ll.ch>,
        Javier Martinez Canillas <javierm@...hat.com>,
        Brendan Higgins <brendan.higgins@...ux.dev>,
        tales.aparecida@...il.com, andrealmeid@...eup.net,
        mairacanal@...eup.net, dri-devel@...ts.freedesktop.org,
        linux-kernel@...r.kernel.org, kunit-dev@...glegroups.com
Subject: Re: [PATCH 2/3] kunit: Add kunit_move_action_to_top_or_reset() to
 reorder actions

Hi David,

On Fri, Sep 22, 2023 at 04:00:21PM +0800, David Gow wrote:
> On Wed, 20 Sept 2023 at 14:12, Arthur Grillo <arthurgrillo@...eup.net> wrote:
> >
> > On Kunit, if we allocate a resource A and B on this order, with its
> > deferred actions to free them. The resource stack would be something
> > like this:
> >
> >          +---------+
> >          | free(B) |
> >          +---------+
> >          |   ...   |
> >          +---------+
> >          | free(A) |
> >          +---------+
> >
> > If the deferred action of A accesses B, this would cause a
> > use-after-free bug. To solve that, we need a way to change the order
> > of actions.
> >
> > Create a function to move an action to the top of the resource stack,
> > as shown in the diagram below.
> >
> >          +---------+    +---------+
> >          | free(B) |    | free(A) |
> >          +---------+    +---------+
> >          |   ...   | -> | free(B) |
> >          +---------+    +---------+
> >          | free(A) |    |   ...   |
> >          +---------+    +---------+
> >
> > Signed-off-by: Arthur Grillo <arthurgrillo@...eup.net>
> > ---
> 
> Thanks. This is a really interesting patch: my hope was that something
> like this wouldn't be necessary, as in most cases freeing things in
> the reverse order to which they were created is the right thing to do.
> 
> It looks like, from the comments on patch 3, this may no longer be
> necessary? Is that so?

Yeah, it's no longer necessary

Maxime