rpms
/
kernel
2
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Fix CVE-2017-17712 (rhbz 1526427 1526933)

This commit is contained in:
Justin M. Forbes 2017-12-18 08:17:57 -06:00
parent 2ca4e0f07d
commit 18f6ac2854
2 changed files with 85 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
kernel.spec
View File

@ -670,6 +670,9 @@ Patch627: qxl-fixes.patch
# rhbz 1462175
Patch628: HID-rmi-Check-that-a-device-is-a-RMI-device-before-c.patch
# CVE-2017-17712 rhbz 1526427 1526933
Patch629: net-ipv4-fix-for-a-race-condition-in-raw_sendmsg.patch
# END OF PATCH DEFINITIONS
%endif
@ -2226,6 +2229,7 @@ fi
%changelog
* Mon Dec 18 2017 Justin M. Forbes <jforbes@fedoraproject.org> - 4.14.7-300
- Linux v4.14.7
- Fix CVE-2017-17712 (rhbz 1526427 1526933)
* Thu Dec 14 2017 Jeremy Cline <jeremy@jcline.org> - 4.14.6-300
- Linux v4.14.6

81
net-ipv4-fix-for-a-race-condition-in-raw_sendmsg.patch Normal file
View File

@ -0,0 +1,81 @@
From patchwork Sun Dec 10 03:50:58 2017
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Subject: net: ipv4: fix for a race condition in raw_sendmsg
X-Patchwork-Submitter: simo.ghannam@gmail.com
X-Patchwork-Id: 846641
X-Patchwork-Delegate: davem@davemloft.net
Message-Id: <5a2caf2e.4ce61c0a.5017a.575f@mx.google.com>
To: netdev@vger.kernel.org
Cc: Mohamed Ghannam <simo.ghannam@gmail.com>
Date: Sun, 10 Dec 2017 03:50:58 +0000
From: simo.ghannam@gmail.com
List-Id: <netdev.vger.kernel.org>
From: Mohamed Ghannam <simo.ghannam@gmail.com>
inet->hdrincl is racy, and could lead to uninitialized stack pointer
usage, so its value should be read only once.
Signed-off-by: Mohamed Ghannam <simo.ghannam@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
---
net/ipv4/raw.c | 15 ++++++++++-----
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 33b70bfd1122..125c1eab3eaa 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -513,11 +513,16 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
int err;
struct ip_options_data opt_copy;
struct raw_frag_vec rfv;
+ int hdrincl;
err = -EMSGSIZE;
if (len > 0xFFFF)
goto out;
+ /* hdrincl should be READ_ONCE(inet->hdrincl)
+ * but READ_ONCE() doesn't work with bit fields
+ */
+ hdrincl = inet->hdrincl;
/*
* Check the flags.
*/
@@ -593,7 +598,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
/* Linux does not mangle headers on raw sockets,
* so that IP options + IP_HDRINCL is non-sense.
*/
- if (inet->hdrincl)
+ if (hdrincl)
goto done;
if (ipc.opt->opt.srr) {
if (!daddr)
@@ -615,12 +620,12 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
flowi4_init_output(&fl4, ipc.oif, sk->sk_mark, tos,
RT_SCOPE_UNIVERSE,
- inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol,
+ hdrincl ? IPPROTO_RAW : sk->sk_protocol,
inet_sk_flowi_flags(sk) |
- (inet->hdrincl ? FLOWI_FLAG_KNOWN_NH : 0),
+ (hdrincl ? FLOWI_FLAG_KNOWN_NH : 0),
daddr, saddr, 0, 0, sk->sk_uid);
- if (!inet->hdrincl) {
+ if (!hdrincl) {
rfv.msg = msg;
rfv.hlen = 0;
@@ -645,7 +650,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
goto do_confirm;
back_from_confirm:
- if (inet->hdrincl)
+ if (hdrincl)
err = raw_send_hdrinc(sk, &fl4, msg, len,
&rt, msg->msg_flags, &ipc.sockc);