linux-hardening - Re: [PATCH v3] batman-adv: Add flex array to struct batadv_tvlv_tt

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]

Message-ID: <202405060928.272B235@keescook>
Date: Mon, 6 May 2024 09:29:42 -0700
From: Kees Cook <keescook@...omium.org>
To: Sven Eckelmann <sven@...fation.org>
Cc: Erick Archer <erick.archer@...look.com>,
	Marek Lindner <mareklindner@...mailbox.ch>,
	Simon Wunderlich <sw@...onwunderlich.de>,
	Antonio Quartulli <a@...table.cc>,
	"David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	Nathan Chancellor <nathan@...nel.org>,
	Nick Desaulniers <ndesaulniers@...gle.com>,
	Bill Wendling <morbo@...gle.com>,
	Justin Stitt <justinstitt@...gle.com>,
	"Gustavo A. R. Silva" <gustavoars@...nel.org>,
	b.a.t.m.a.n@...ts.open-mesh.org, linux-kernel@...r.kernel.org,
	netdev@...r.kernel.org, linux-hardening@...r.kernel.org,
	llvm@...ts.linux.dev,
	Alexander Lobakin <aleksander.lobakin@...el.com>
Subject: Re: [PATCH v3] batman-adv: Add flex array to struct
 batadv_tvlv_tt_data

On Sun, May 05, 2024 at 05:22:10PM +0200, Sven Eckelmann wrote:
> On Saturday, 4 May 2024 19:08:39 CEST Erick Archer wrote:
> [...]
> > > Thanks for the updates. But I can't accept this at the moment because 
> > > __counted_by_be is used in an uapi header without it being defined
> > > include/uapi/linux/stddef.h (and this file is also not included in this 
> > > header).
> > > 
> > > See commit c8248faf3ca2 ("Compiler Attributes: counted_by: Adjust name and 
> > > identifier expansion") as an example for the similar __counted_by macro.
> > 
> > If I understand correctly, the following changes are also needed because
> > the annotated struct is defined in a "uapi" header. Sorry if it's a stupid
> > question, but I'm new to these topics.
> 
> No, it is absolutely no stupid question.
> 
> > diff --git a/include/uapi/linux/batadv_packet.h b/include/uapi/linux/batadv_packet.h
> > index 6e25753015df..41f39d7661c9 100644
> > --- a/include/uapi/linux/batadv_packet.h
> > +++ b/include/uapi/linux/batadv_packet.h
> > @@ -9,6 +9,7 @@
> > 
> >  #include <asm/byteorder.h>
> >  #include <linux/if_ether.h>
> > +#include <linux/stddef.h>
> >  #include <linux/types.h>
> > 
> >  /**
> 
> This must definitely go into your "original" patch
> 
> > diff --git a/include/uapi/linux/stddef.h b/include/uapi/linux/stddef.h
> > index 2ec6f35cda32..58154117d9b0 100644
> > --- a/include/uapi/linux/stddef.h
> > +++ b/include/uapi/linux/stddef.h
> > @@ -55,4 +55,12 @@
> >  #define __counted_by(m)
> >  #endif
> > 
> > +#ifndef __counted_by_le
> > +#define __counted_by_le(m)
> > +#endif
> 
> If you want to add this (for completeness) then please put it in an extra 
> patch. It is simply not used by batman-adv and I would not be able to find any 
> justification why it should be part of the batman-adv patch.
> 
> > +
> > +#ifndef __counted_by_be
> > +#define __counted_by_be(m)
> > +#endif
> > +
> 
> This part can be either:
> 
> * in the batman-adv patch
> * or together with the __counted_by_le change in an additional patch which is 
>   "in front" of the batman-adv patch (in the patch series).
> 
> From my perspective, it is for you to decide - but of course, other 
> maintainers might have a different opinion about it.
> 
> >  #endif /* _UAPI_LINUX_STDDEF_H */
> > 
> > If this is the right path, can these changes be merged into a
> > single patch or is it better to add a previous patch to define
> > __counted_by{le,be}?
> 
> I don't have a perfect answer here. See the comments above. The file 
> include/uapi/linux/stddef.h doesn't have a specific maintainer (according to 
> ./scripts/get_maintainer.pl) - so it should be fine to get modified through 
> the net-next tree.
> 
> But maybe Kees Cook has a different opinion about it. At least there are a lot 
> of Signed-off-bys for this file by Kees.

FWIW, I'm also fine with the UAPI going in via netdev. It's the most
likely place to use the be/le variants. Is netdev still open for
patches? Whatever the path, we should get it into this coming merge
window so we can use it elsewhere too if we need it.

-- 
Kees Cook