Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Message-ID: <1308b92b9592f6a3331b199658e306714c8c9cac.camel@gmail.com>
Subject: Re: [PATCH] net: netlink: fix error check in
 genl_family_rcv_msg_doit
From:   Pavel Skripkin <paskripkin@gmail.com>
To:     Johannes Berg <johannes@sipsolutions.net>,
        "davem@davemloft.net" <davem@davemloft.net>,
        "kuba@kernel.org" <kuba@kernel.org>
Cc:     "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Date:   Sat, 03 Apr 2021 19:33:24 +0300
In-Reply-To: <b8a83042f83af92e87550085175da5c1d95cc4b0.camel@sipsolutions.net>
References: <20210403151312.31796-1-paskripkin@gmail.com>
         <b8a83042f83af92e87550085175da5c1d95cc4b0.camel@sipsolutions.net>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.38.4 
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi!

On Sat, 2021-04-03 at 18:26 +0200, Johannes Berg wrote:
> On Sat, 2021-04-03 at 15:13 +0000, Pavel Skripkin wrote:
> > genl_family_rcv_msg_attrs_parse() can return NULL
> > pointer:
> > 
> >         if (!ops->maxattr)
> >                 return NULL;
> > 
> > But this condition doesn't cause an error in
> > genl_family_rcv_msg_doit
> 
> And I'm almost certain that in fact it shouldn't cause an error!
> 
> If the family doesn't set maxattr then it doesn't want to have
> generic
> netlink doing the parsing, but still it should be possible to call
> the
> ops. Look at fs/dlm/netlink.c for example, it doesn't even have
> attributes. You're breaking it with this patch.
> 
> Also, the (NULL) pointer is not actually _used_ anywhere, so why
> would
> it matter?
> 

Oh, I see now. I thought, it could cause a NULL ptr deference in some
cases, because some ->doit() functions accessing info.attrs directly.
Now I understand the point, sorry for my misunderstanding the
situation.

> johannes
> 

With regards,
Pavel Skripkin