netdev - Re: memory leak in do_ipv6

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: <1448977016.25582.18.camel@edumazet-glaptop2.roam.corp.google.com>
Date:	Tue, 01 Dec 2015 05:36:56 -0800
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Dmitry Vyukov <dvyukov@...gle.com>
Cc:	"David S. Miller" <davem@...emloft.net>,
	Alexey Kuznetsov <kuznet@....inr.ac.ru>,
	James Morris <jmorris@...ei.org>,
	Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
	Patrick McHardy <kaber@...sh.net>,
	netdev <netdev@...r.kernel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Vlad Yasevich <vyasevich@...il.com>,
	Neil Horman <nhorman@...driver.com>,
	linux-sctp@...r.kernel.org, syzkaller <syzkaller@...glegroups.com>,
	Kostya Serebryany <kcc@...gle.com>,
	Alexander Potapenko <glider@...gle.com>,
	Sasha Levin <sasha.levin@...cle.com>,
	Eric Dumazet <edumazet@...gle.com>
Subject: Re: memory leak in do_ipv6_setsockopt

On Tue, 2015-12-01 at 13:27 +0100, Dmitry Vyukov wrote:
> Hello,
> 
> The following program causes a memory leak :
> 
> // autogenerated by syzkaller (http://github.com/google/syzkaller)
> #include <syscall.h>
> #include <sys/types.h>
> #include <sys/socket.h>
> #include <string.h>
> #include <stdint.h>
> #include <linux/in.h>
> #include <linux/socket.h>
> 
> int main()
> {
>         long r1 = syscall(SYS_socket, PF_INET6,
> SOCK_STREAM|SOCK_CLOEXEC|SOCK_NONBLOCK, IPPROTO_SCTP);
>         const char *opt = "\x15\x53\x5e\x2d\x97\xab\xe1";
>         long r3 = syscall(SYS_setsockopt, r1, 0x29ul, 0x6ul, opt, 0x7ul);
>         return 0;
> }
> 
> 
> unreferenced object 0xffff880039a55260 (size 64):
>   comm "executor", pid 11746, jiffies 4298984475 (age 16.078s)
>   hex dump (first 32 bytes):
>     2f 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  /...............
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>   backtrace:
>     [<     inline     >] kmalloc include/linux/slab.h:463
>     [<ffffffff848a2f5f>] sock_kmalloc+0x7f/0xc0 net/core/sock.c:1774
>     [<ffffffff84e5bea0>] do_ipv6_setsockopt.isra.7+0x15d0/0x2830
> net/ipv6/ipv6_sockglue.c:483
>     [<ffffffff84e5d19b>] ipv6_setsockopt+0x9b/0x140 net/ipv6/ipv6_sockglue.c:885
>     [<ffffffff8544616c>] sctp_setsockopt+0x15c/0x36c0 net/sctp/socket.c:3702
>     [<ffffffff848a2035>] sock_common_setsockopt+0x95/0xd0 net/core/sock.c:2645
>     [<     inline     >] SYSC_setsockopt net/socket.c:1757
>     [<ffffffff8489f1d8>] SyS_setsockopt+0x158/0x240 net/socket.c:1736
> 
> 
> I confirmed that running this program in a loop steadily increases
> number of objects in kmalloc-64 slab. The leak does not happen with
> IPPROTO_TCP, so probably it is sctp-related.

Thanks for the report.

Probably fixed by :

diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 897c01c029ca..8079ecd8465d 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -7375,6 +7375,12 @@ struct proto sctp_prot = {
 
 #if IS_ENABLED(CONFIG_IPV6)
 
+static void sctp_v6_destroy_sock(struct sock *sk)
+{
+	sctp_destroy_sock(sk);
+	inet6_destroy_sock(sk);
+}
+
 struct proto sctpv6_prot = {
 	.name		= "SCTPv6",
 	.owner		= THIS_MODULE,
@@ -7384,7 +7390,7 @@ struct proto sctpv6_prot = {
 	.accept		= sctp_accept,
 	.ioctl		= sctp_ioctl,
 	.init		= sctp_init_sock,
-	.destroy	= sctp_destroy_sock,
+	.destroy	= sctp_v6_destroy_sock,
 	.shutdown	= sctp_shutdown,
 	.setsockopt	= sctp_setsockopt,
 	.getsockopt	= sctp_getsockopt,


--
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