[pve-devel] [PATCH kernel 2/3] fix CVE-2017-6074: local root in dccp module

Fabian Grünbichler f.gruenbichler at proxmox.com
Thu Feb 23 18:53:17 CET 2017


Signed-off-by: Fabian Grünbichler <f.gruenbichler at proxmox.com>
---
 Makefile                                           |  1 +
 ...reeing-skb-too-early-for-IPV6_RECVPKTINFO.patch | 58 ++++++++++++++++++++++
 2 files changed, 59 insertions(+)
 create mode 100644 CVE-2017-6074-dccp-fix-freeing-skb-too-early-for-IPV6_RECVPKTINFO.patch

diff --git a/Makefile b/Makefile
index e7edb76..6a0228e 100644
--- a/Makefile
+++ b/Makefile
@@ -269,6 +269,7 @@ ${KERNEL_SRC}/README ${KERNEL_CFG_ORG}: ${KERNELSRCTAR}
 	cd ${KERNEL_SRC}; patch -p1 < ../0002-Revert-mm-oom-rework-oom-detection.patch
 	cd ${KERNEL_SRC}; patch -p1 < ../CVE-2017-2583-KVM-x86-fix-emulation-of-MOV-SS-null-selector.patch
 	cd ${KERNEL_SRC}; patch -p1 < ../CVE-2017-2596-kvm-page-reference-leakage-in-handle_vmon.patch
+	cd ${KERNEL_SRC}; patch -p1 < ../CVE-2017-6074-dccp-fix-freeing-skb-too-early-for-IPV6_RECVPKTINFO.patch
 	sed -i ${KERNEL_SRC}/Makefile -e 's/^EXTRAVERSION.*$$/EXTRAVERSION=${EXTRAVERSION}/'
 	touch $@
 
diff --git a/CVE-2017-6074-dccp-fix-freeing-skb-too-early-for-IPV6_RECVPKTINFO.patch b/CVE-2017-6074-dccp-fix-freeing-skb-too-early-for-IPV6_RECVPKTINFO.patch
new file mode 100644
index 0000000..202aa59
--- /dev/null
+++ b/CVE-2017-6074-dccp-fix-freeing-skb-too-early-for-IPV6_RECVPKTINFO.patch
@@ -0,0 +1,58 @@
+From 4feb04ade2a559e0e76a0215d259c5673e5b1cbd Mon Sep 17 00:00:00 2001
+From: Andrey Konovalov <andreyknvl at google.com>
+Date: Thu, 16 Feb 2017 17:22:46 +0100
+Subject: [PATCH] dccp: fix freeing skb too early for IPV6_RECVPKTINFO
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+In the current DCCP implementation an skb for a DCCP_PKT_REQUEST packet
+is forcibly freed via __kfree_skb in dccp_rcv_state_process if
+dccp_v6_conn_request successfully returns.
+
+However, if IPV6_RECVPKTINFO is set on a socket, the address of the skb
+is saved to ireq->pktopts and the ref count for skb is incremented in
+dccp_v6_conn_request, so skb is still in use. Nevertheless, it gets freed
+in dccp_rcv_state_process.
+
+Fix by calling consume_skb instead of doing goto discard and therefore
+calling __kfree_skb.
+
+Similar fixes for TCP:
+
+fb7e2399ec17f1004c0e0ccfd17439f8759ede01 [TCP]: skb is unexpectedly freed.
+0aea76d35c9651d55bbaf746e7914e5f9ae5a25d tcp: SYN packets are now
+simply consumed
+
+Signed-off-by: Andrey Konovalov <andreyknvl at google.com>
+Acked-by: Eric Dumazet <edumazet at google.com>
+Signed-off-by: David S. Miller <davem at davemloft.net>
+
+CVE-2017-6074
+BugLink: http://bugs.launchpad.net/bugs/1665935
+
+(cherry-picked from 5edabca9d4cff7f1f2b68f0bac55ef99d9798ba4 davem)
+Signed-off-by: Stefan Bader <stefan.bader at canonical.com>
+
+Signed-off-by: Fabian Grünbichler <f.gruenbichler at proxmox.com>
+---
+ net/dccp/input.c | 3 ++-
+ 1 file changed, 2 insertions(+), 1 deletion(-)
+
+diff --git a/net/dccp/input.c b/net/dccp/input.c
+index 3bd14e8..dbe2573 100644
+--- a/net/dccp/input.c
++++ b/net/dccp/input.c
+@@ -606,7 +606,8 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
+ 			if (inet_csk(sk)->icsk_af_ops->conn_request(sk,
+ 								    skb) < 0)
+ 				return 1;
+-			goto discard;
++			consume_skb(skb);
++			return 0;
+ 		}
+ 		if (dh->dccph_type == DCCP_PKT_RESET)
+ 			goto discard;
+-- 
+2.1.4
+
-- 
2.1.4





More information about the pve-devel mailing list