ipv4: Fix potential Spectre v1 vulnerability

vr.vifi is indirectly controlled by user-space, hence leading to a potential exploitation of the Spectre variant 1 vulnerability.

This issue was detected with the help of Smatch:

net/ipv4/ipmr.c:1616 ipmr_ioctl() warn: potential spectre issue 'mrt->vif_table' [r] (local cap) net/ipv4/ipmr.c:1690 ipmr_compat_ioctl() warn: potential spectre issue 'mrt->vif_table' [r] (local cap)

Fix this by sanitizing vr.vifi before using it to index mrt->vif_table'

Notice that given that speculation windows are large, the policy is to kill the speculation on the first load and not worry if it can be completed with a dependent load/store [1].

[1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2

---
 net/ipv4/ipmr.c | 4 ++++
 1 file changed, 4 insertions(+)

index ea04e38f56e9..8c03fb05d3e6 100644
@@ -69,6 +69,8 @@
 #include <net/nexthop.h>
 #include <net/switchdev.h>
 
+#include <linux/nospec.h>
+
 struct ipmr_rule {
 	struct fib_rule		common;
 };

@@ -1612,6 +1614,7 @@ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg)
 			return -EFAULT;
 		if (vr.vifi >= mrt->maxvif)
 			return -EINVAL;
+		vr.vifi = array_index_nospec(vr.vifi, mrt->maxvif);
 		read_lock(&mrt_lock);
 		vif = &mrt->vif_table[vr.vifi];
 		if (VIF_EXISTS(mrt, vr.vifi)) {

@@ -1686,6 +1689,7 @@ int ipmr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg)
 			return -EFAULT;
 		if (vr.vifi >= mrt->maxvif)
 			return -EINVAL;
+		vr.vifi = array_index_nospec(vr.vifi, mrt->maxvif);
 		read_lock(&mrt_lock);
 		vif = &mrt->vif_table[vr.vifi];
 		if (VIF_EXISTS(mrt, vr.vifi)) {

2.17.1

From: "Gustavo A. R. Silva" <[email protected]>
Date: Mon, 10 Dec 2018 12:41:24 -0600

Applied and queued up for -stable.

Please do not CC: stable for networking bug fixes, thank you.

OK. I've got it.

Thanks
--
Gustavo