Re: [PATCH v2] overflow: Change DEFINE_FLEX to take __counted_by member

From: Kees Cook <keescook@chromium.org>
To: Simon Horman <horms@kernel.org>
Cc: Przemek Kitszel <przemyslaw.kitszel@intel.com>,
	Jesse Brandeburg <jesse.brandeburg@intel.com>,
	Tony Nguyen <anthony.l.nguyen@intel.com>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	"Gustavo A. R. Silva" <gustavoars@kernel.org>,
	intel-wired-lan@lists.osuosl.org, netdev@vger.kernel.org,
	linux-hardening@vger.kernel.org,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] overflow: Change DEFINE_FLEX to take __counted_by member
Date: Sat, 9 Mar 2024 12:32:45 -0800	[thread overview]
Message-ID: <202403091230.ACF639521@keescook> (raw)
In-Reply-To: <20240308202018.GC603911@kernel.org>

On Fri, Mar 08, 2024 at 08:20:18PM +0000, Simon Horman wrote:
> On Wed, Mar 06, 2024 at 03:51:36PM -0800, Kees Cook wrote:
> > The norm should be flexible array structures with __counted_by
> > annotations, so DEFINE_FLEX() is updated to expect that. Rename
> > the non-annotated version to DEFINE_RAW_FLEX(), and update the
> > few existing users.
> > 
> > Signed-off-by: Kees Cook <keescook@chromium.org>
> 
> Hi Kees,
> 
> I'm unclear what this is based on, as it doesn't appear to apply
> cleanly to net-next or the dev-queue branch of the iwl-next tree.
> But I manually applied it to the latter and ran some checks.

It was based on v6.8-rc2, but it no longer applies cleanly to iwl-next:
https://lore.kernel.org/linux-next/20240307162958.02ec485c@canb.auug.org.au/

Is this something iwl-next can take for the v6.9 merge window? I can
send a rebased patch if that helps?

> > @@ -396,9 +396,9 @@ static inline size_t __must_check size_sub(size_t minuend, size_t subtrahend)
> >   * @name: Name for a variable to define.
> >   * @member: Name of the array member.
> >   * @count: Number of elements in the array; must be compile-time const.
> > - * @initializer: initializer expression (could be empty for no init).
> > + * @initializer...: initializer expression (could be empty for no init).
> 
> Curiously kernel-doc --none seems happier without the line above changed.

I've fixed this up too:
https://lore.kernel.org/linux-next/202403071124.36DC2B617A@keescook/

-- 
Kees Cook