Re: [Linux-kernel-mentees] [PATCH v4 net] rose: Fix Null pointer dereference in rose_send_frame()

From: Anmol Karn <anmol.karan123@gmail.com>
To: Jakub Kicinski <kuba@kernel.org>
Cc: ralf@linux-mips.org, davem@davemloft.net, saeed@kernel.org,
	gregkh@linuxfoundation.org, netdev@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-hams@vger.kernel.org,
	linux-kernel-mentees@lists.linuxfoundation.org,
	syzkaller-bugs@googlegroups.com,
	syzbot+a1c743815982d9496393@syzkaller.appspotmail.com
Subject: Re: [Linux-kernel-mentees] [PATCH v4 net] rose: Fix Null pointer dereference in rose_send_frame()
Date: Sun, 15 Nov 2020 17:14:48 +0530	[thread overview]
Message-ID: <20201115114448.GA40574@Thinkpad> (raw)
In-Reply-To: <20201114111838.03b933af@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>

On Sat, Nov 14, 2020 at 11:18:38AM -0800, Jakub Kicinski wrote:
> On Wed, 11 Nov 2020 22:29:54 +0530 Anmol Karn wrote:
> > rose_send_frame() dereferences `neigh->dev` when called from
> > rose_transmit_clear_request(), and the first occurrence of the
> > `neigh` is in rose_loopback_timer() as `rose_loopback_neigh`,
> > and it is initialized in rose_add_loopback_neigh() as NULL.
> > i.e when `rose_loopback_neigh` used in rose_loopback_timer()
> > its `->dev` was still NULL and rose_loopback_timer() was calling
> > rose_rx_call_request() without checking for NULL.
> > 
> > - net/rose/rose_link.c
> > This bug seems to get triggered in this line:
> > 
> > rose_call = (ax25_address *)neigh->dev->dev_addr;
> > 
> > Fix it by adding NULL checking for `rose_loopback_neigh->dev`
> > in rose_loopback_timer().
> > 
> > Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> > Reported-by: syzbot+a1c743815982d9496393@syzkaller.appspotmail.com
> > Tested-by: syzbot+a1c743815982d9496393@syzkaller.appspotmail.com
> > Link: https://syzkaller.appspot.com/bug?id=9d2a7ca8c7f2e4b682c97578dfa3f236258300b3
> > Signed-off-by: Anmol Karn <anmol.karan123@gmail.com>
> 
> > diff --git a/net/rose/rose_loopback.c b/net/rose/rose_loopback.c
> > index 7b094275ea8b..6a71b6947d92 100644
> > --- a/net/rose/rose_loopback.c
> > +++ b/net/rose/rose_loopback.c
> > @@ -96,10 +96,12 @@ static void rose_loopback_timer(struct timer_list *unused)
> >  		}
> > 
> >  		if (frametype == ROSE_CALL_REQUEST) {
> > -			if ((dev = rose_dev_get(dest)) != NULL) {
> > +			dev = rose_dev_get(dest);
> > +			if (rose_loopback_neigh->dev && dev) {
> >  				if (rose_rx_call_request(skb, dev, rose_loopback_neigh, lci_o) == 0)
> >  					kfree_skb(skb);
> >  			} else {
> > +				dev_put(dev);
> >  				kfree_skb(skb);
> >  			}
> >  		} else {
> 
> This is still not correct. With this code dev_put() could be called with
> NULL, which would cause a crash.
> 
> There is also a dev_put() missing if rose_rx_call_request() returns 0.
> 
> I think that this is the correct code:
> 
> diff --git a/net/rose/rose_loopback.c b/net/rose/rose_loopback.c
> index 7b094275ea8b..ff252ef73592 100644
> --- a/net/rose/rose_loopback.c
> +++ b/net/rose/rose_loopback.c
> @@ -96,11 +96,22 @@ static void rose_loopback_timer(struct timer_list *unused)
>  		}
>  
>  		if (frametype == ROSE_CALL_REQUEST) {
> -			if ((dev = rose_dev_get(dest)) != NULL) {
> -				if (rose_rx_call_request(skb, dev, rose_loopback_neigh, lci_o) == 0)
> -					kfree_skb(skb);
> -			} else {
> +			if (!rose_loopback_neigh->dev) {
>  				kfree_skb(skb);
> +				continue;
> +			}
> +
> +			dev = rose_dev_get(dest);
> +			if (!dev) {
> +				kfree_skb(skb);
> +				continue;
> +			}
> +
> +			if (rose_rx_call_request(skb, dev, rose_loopback_neigh,
> +						 lci_o) == 0) {
> +				dev_put(dev);
> +				kfree_skb(skb);
>  			}
>  		} else {
>  			kfree_skb(skb);
> 
> Please test this and resubmit it if it works.

Sure sir, I will test it and resend, if it works.

Thanks,
Anmol