[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CANq1E4Rh60iS6WRq6xypV_ahG5N=uErL_hMnVcncc=DTSPvdUw@mail.gmail.com>
Date: Fri, 13 Nov 2015 15:43:31 +0100
From: David Herrmann <dh.herrmann@...il.com>
To: Dan Carpenter <dan.carpenter@...cle.com>
Cc: "David S. Miller" <davem@...emloft.net>,
Thomas Graf <tgraf@...g.ch>,
Herbert Xu <herbert@...dor.apana.org.au>,
Daniel Borkmann <daniel@...earbox.net>,
Johannes Berg <johannes.berg@...el.com>,
Al Viro <viro@...iv.linux.org.uk>,
Nicolas Dichtel <nicolas.dichtel@...nd.com>,
Florian Westphal <fw@...len.de>,
netdev <netdev@...r.kernel.org>,
"kernel-janitors@...r.kernel.org" <kernel-janitors@...r.kernel.org>
Subject: Re: [patch] netlink: fix a limit in NETLINK_LIST_MEMBERSHIPS
Hi
On Fri, Nov 13, 2015 at 3:20 PM, Dan Carpenter <dan.carpenter@...cle.com> wrote:
> This condition doesn't work when len is smaller than expected and not a
> multiple of 4. In that situation "len - pos" is negative and type
> promoted to a high unsigned value and we do not break out of the loop.
> It causes the program calling it to crash.
Could you give an example how this can happen? The loop-invariant
should be "len >= pos", as such, this shouldn't happen. "pos" starts
out as 0, "len" is guaranteed to be >=0. "pos" is only incremented by
4, if "len - pos >= 4".
What am I missing?
Thanks
David
> Fixes: b42be38b2778 ('netlink: add API to retrieve all group memberships')
> Signed-off-by: Dan Carpenter <dan.carpenter@...cle.com>
>
> diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
> index 59651af..76a8466 100644
> --- a/net/netlink/af_netlink.c
> +++ b/net/netlink/af_netlink.c
> @@ -2373,7 +2373,7 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname,
> err = 0;
> netlink_lock_table();
> for (pos = 0; pos * 8 < nlk->ngroups; pos += sizeof(u32)) {
> - if (len - pos < sizeof(u32))
> + if (len < pos + sizeof(u32))
> break;
>
> idx = pos / sizeof(unsigned long);
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists