linux-kernel - Re: [PATCH] net: ipv6: fix invalid alloclen in __ip6_append

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: <66463e26-8564-9f58-ce41-9a2843891d1a@kernel.org>
Date:   Tue, 8 Mar 2022 11:18:18 -0700
From:   David Ahern <dsahern@...nel.org>
To:     Tadeusz Struk <tadeusz.struk@...aro.org>,
        David Laight <David.Laight@...LAB.COM>,
        "davem@...emloft.net" <davem@...emloft.net>
Cc:     Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
        Jakub Kicinski <kuba@...nel.org>,
        Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Andrii Nakryiko <andrii@...nel.org>,
        Martin KaFai Lau <kafai@...com>,
        Song Liu <songliubraving@...com>, Yonghong Song <yhs@...com>,
        John Fastabend <john.fastabend@...il.com>,
        KP Singh <kpsingh@...nel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "bpf@...r.kernel.org" <bpf@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "stable@...r.kernel.org" <stable@...r.kernel.org>,
        "syzbot+e223cf47ec8ae183f2a0@...kaller.appspotmail.com" 
        <syzbot+e223cf47ec8ae183f2a0@...kaller.appspotmail.com>
Subject: Re: [PATCH] net: ipv6: fix invalid alloclen in __ip6_append_data

On 3/8/22 8:43 AM, Tadeusz Struk wrote:
> Hi David,
> On 3/7/22 18:58, David Laight wrote:
>>> diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
>>> index 4788f6b37053..622345af323e 100644
>>> --- a/net/ipv6/ip6_output.c
>>> +++ b/net/ipv6/ip6_output.c
>>> @@ -1629,6 +1629,13 @@ static int __ip6_append_data(struct sock *sk,
>>>                   err = -EINVAL;
>>>                   goto error;
>>>               }
>>> +            if (unlikely(alloclen < fraglen)) {
>>> +                if (printk_ratelimit())
>>> +                    pr_warn("%s: wrong alloclen: %d, fraglen: %d",
>>> +                        __func__, alloclen, fraglen);
>>> +                alloclen = fraglen;
>>> +            }
>>> +
>> Except that is a valid case, see a few lines higher:
>>
>>                 alloclen = min_t(int, fraglen, MAX_HEADER);
>>                 pagedlen = fraglen - alloclen;
>>
>> You need to report the input values that cause the problem later on.
> 
> OK, but in this case it falls into the first if block:
> https://elixir.bootlin.com/linux/v5.17-rc7/source/net/ipv6/ip6_output.c#L1606
> 
> where alloclen is assigned the value of mtu.
> The values in this case are just before the alloc_skb() are:
> 
> alloclen = 1480
> alloc_extra = 136
> datalen = 64095
> fragheaderlen = 1480
> fraglen = 65575
> transhdrlen = 0
> mtu = 1480
> 

Does this solve the problem (whitespace damaged on paste, but it is just
a code move and removing fraglen getting set twice):

diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index e69fac576970..59f036241f1b 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1589,6 +1589,15 @@ static int __ip6_append_data(struct sock *sk,

                        if (datalen > (cork->length <= mtu &&
!(cork->flags & IPCORK_ALLFRAG) ? mtu : maxfraglen) - fragheaderlen)
                                datalen = maxfraglen - fragheaderlen -
rt->dst.trailer_len;
+
+                       if (datalen != length + fraggap) {
+                               /*
+                                * this is not the last fragment, the
trailer
+                                * space is regarded as data space.
+                                */
+                               datalen += rt->dst.trailer_len;
+                       }
+
                        fraglen = datalen + fragheaderlen;
                        pagedlen = 0;

@@ -1615,16 +1624,6 @@ static int __ip6_append_data(struct sock *sk,
                        }
                        alloclen += alloc_extra;

-                       if (datalen != length + fraggap) {
-                               /*
-                                * this is not the last fragment, the
trailer
-                                * space is regarded as data space.
-                                */
-                               datalen += rt->dst.trailer_len;
-                       }
-
-                       fraglen = datalen + fragheaderlen;
-
                        copy = datalen - transhdrlen - fraggap - pagedlen;
                        if (copy < 0) {
                                err = -EINVAL;