pull request (net): ipsec 2020-03-27
Subsystems (4)
Steffen Klassert
maintainer1) Handle NETDEV_UNREGISTER for xfrm device to handle asynchronous unregister events cleanly. From Raed Salem.
2) Fix vti6 tunnel inter address family TX through bpf_redirect(). From Nicolas Dichtel.
3) Fix lenght check in verify_sec_ctx_len() to avoid a slab-out-of-bounds. From Xin Long.
4) Add a missing verify_sec_ctx_len check in xfrm_add_acquire to avoid a possible out-of-bounds to access. From Xin Long.
5) Use built-in RCU list checking of hlist_for_each_entry_rcu to silence false lockdep warning in __xfrm6_tunnel_spi_lookup when CONFIG_PROVE_RCU_LIST is enabled. From Madhuparna Bhowmik.
6) Fix a panic on esp offload when crypto is done asynchronously. From Xin Long.
7) Fix a skb memory leak in an error path of vti6_rcv. From Torsten Hilbrich.
8) Fix a race that can lead to a doulbe free in xfrm_policy_timer. From Xin Long.
Please pull or let me know if there are problems.
Thanks!
The following changes since commit a444ad1432c5a0fb3bd43fc9ac39fb88b1fb141e:
Merge branch 'netdevsim-fix-several-bugs-in-netdevsim-module' (2020-02-03 15:38:50 -0800)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec.git master
for you to fetch changes up to 4c59406ed00379c8663f8663d82b2537467ce9d7:
xfrm: policy: Fix doulbe free in xfrm_policy_timer (2020-03-24 06:56:54 +0100)
----------------------------------------------------------------
Madhuparna Bhowmik (1):
ipv6: xfrm6_tunnel.c: Use built-in RCU list checking
Nicolas Dichtel (1):
vti[6]: fix packet tx through bpf_redirect() in XinY cases
Raed Salem (1):
xfrm: handle NETDEV_UNREGISTER for xfrm device
Torsten Hilbrich (1):
vti6: Fix memory leak of skb if input policy check fails
Xin Long (3):
xfrm: fix uctx len check in verify_sec_ctx_len
xfrm: add the missing verify_sec_ctx_len check in xfrm_add_acquire
esp: remove the skb from the chain when it's enqueued in cryptd_wq
YueHaibing (1):
xfrm: policy: Fix doulbe free in xfrm_policy_timer
net/ipv4/Kconfig | 1 + net/ipv4/ip_vti.c | 38 ++++++++++++++++++++++++++++++-------- net/ipv6/ip6_vti.c | 34 ++++++++++++++++++++++++++-------- net/ipv6/xfrm6_tunnel.c | 2 +- net/xfrm/xfrm_device.c | 9 +++++---- net/xfrm/xfrm_policy.c | 2 ++ net/xfrm/xfrm_user.c | 6 +++++- 7 files changed, 70 insertions(+), 22 deletions(-)
Steffen Klassert
maintainerPATCH 1/8 xfrm: handle NETDEV_UNREGISTER for xfrm device
From: Raed Salem <[email protected]>
This patch to handle the asynchronous unregister device event so the device IPsec offload resources could be cleanly released.
--- net/xfrm/xfrm_device.c | 1 + 1 file changed, 1 insertion(+)
net/xfrm/xfrm_device.c +1 -0
@@ -383,6 +383,7 @@ static int xfrm_dev_event(struct notifier_block *this, unsigned long event, voidreturn xfrm_dev_feat_change (dev);case NETDEV_DOWN : +case NETDEV_UNREGISTER :return xfrm_dev_down (dev); }return NOTIFY_DONE ;
Signature · 1 line
2.17.1
Steffen Klassert
maintainerPATCH 2/8 vti[6]: fix packet tx through bpf_redirect() in XinY cases
In reply to Steffen KlassertFrom: Nicolas Dichtel <[email protected]>
I forgot the 4in6/6in4 cases in my previous patch. Let's fix them.
--- net/ipv4/Kconfig | 1 + net/ipv4/ip_vti.c | 38 ++++++++++++++++++++++++++++++-------- net/ipv6/ip6_vti.c | 32 +++++++++++++++++++++++++------- 3 files changed, 56 insertions(+), 15 deletions(-)
net/ipv4/Kconfig +1 -0
@@ -303,6 +303,7 @@ config SYN_COOKIES config NET_IPVTI tristate "Virtual (secure) IP: tunneling" + depends on IPV6 || IPV6=n select INET_TUNNEL select NET_IP_TUNNEL select XFRM
net/ipv4/ip_vti.c +30 -8
@@ -187,17 +187,39 @@ static netdev_tx_t vti_xmit(struct sk_buff *skb, struct net_device *dev,int mtu;if (!dst) { -struct rtable *rt; - - fl->u.ip4.flowi4_oif = dev->ifindex; - fl->u.ip4.flowi4_flags |=FLOWI_FLAG_ANYSRC ; - rt =__ip_route_output_key (dev_net (dev), &fl->u.ip4); -if (IS_ERR (rt)) { +switch (skb->protocol) { +case htons (ETH_P_IP ): { +struct rtable *rt; + + fl->u.ip4.flowi4_oif = dev->ifindex; + fl->u.ip4.flowi4_flags |=FLOWI_FLAG_ANYSRC ; + rt =__ip_route_output_key (dev_net (dev), &fl->u.ip4); +if (IS_ERR (rt)) { + dev->stats.tx_carrier_errors++; + goto tx_error_icmp; + } + dst = &rt->dst; +skb_dst_set (skb, dst); +break ; + } +#if IS_ENABLED (CONFIG_IPV6 ) +case htons (ETH_P_IPV6 ): + fl->u.ip6.flowi6_oif = dev->ifindex; + fl->u.ip6.flowi6_flags |=FLOWI_FLAG_ANYSRC ; + dst =ip6_route_output (dev_net (dev),NULL , &fl->u.ip6); +if (dst->error) { +dst_release (dst); + dst =NULL ; + dev->stats.tx_carrier_errors++; + goto tx_error_icmp; + } +skb_dst_set (skb, dst); +break ; +#endif +default : dev->stats.tx_carrier_errors++; goto tx_error_icmp; } - dst = &rt->dst; -skb_dst_set (skb, dst); }dst_hold (dst);
net/ipv6/ip6_vti.c +25 -7
@@ -450,15 +450,33 @@ vti6_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl)int mtu;if (!dst) { - fl->u.ip6.flowi6_oif = dev->ifindex; - fl->u.ip6.flowi6_flags |=FLOWI_FLAG_ANYSRC ; - dst =ip6_route_output (dev_net (dev),NULL , &fl->u.ip6); -if (dst->error) { -dst_release (dst); - dst =NULL ; +switch (skb->protocol) { +case htons (ETH_P_IP ): { +struct rtable *rt; + + fl->u.ip4.flowi4_oif = dev->ifindex; + fl->u.ip4.flowi4_flags |=FLOWI_FLAG_ANYSRC ; + rt =__ip_route_output_key (dev_net (dev), &fl->u.ip4); +if (IS_ERR (rt)) + goto tx_err_link_failure; + dst = &rt->dst; +skb_dst_set (skb, dst); +break ; + } +case htons (ETH_P_IPV6 ): + fl->u.ip6.flowi6_oif = dev->ifindex; + fl->u.ip6.flowi6_flags |=FLOWI_FLAG_ANYSRC ; + dst =ip6_route_output (dev_net (dev),NULL , &fl->u.ip6); +if (dst->error) { +dst_release (dst); + dst =NULL ; + goto tx_err_link_failure; + } +skb_dst_set (skb, dst); +break ; +default : goto tx_err_link_failure; } -skb_dst_set (skb, dst); }dst_hold (dst);
Signature · 1 line
2.17.1
Steffen Klassert
maintainerPATCH 3/8 xfrm: fix uctx len check in verify_sec_ctx_len
In reply to Steffen KlassertFrom: Xin Long <[email protected]>
It's not sufficient to do 'uctx->len != (sizeof(struct xfrm_user_sec_ctx) + uctx->ctx_len)' check only, as uctx->len may be greater than nla_len(rt), in which case it will cause slab-out-of-bounds when accessing uctx->ctx_str later.
This patch is to fix it by return -EINVAL when uctx->len > nla_len(rt).
--- net/xfrm/xfrm_user.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/xfrm/xfrm_user.c +2 -1
@@ -110,7 +110,8 @@ static inline int verify_sec_ctx_len(struct nlattr **attrs)return 0 ; uctx =nla_data (rt); -if (uctx->len != (sizeof (struct xfrm_user_sec_ctx ) + uctx->ctx_len)) +if (uctx->len >nla_len (rt) || + uctx->len != (sizeof (struct xfrm_user_sec_ctx ) + uctx->ctx_len))return -EINVAL ;return 0 ;
Signature · 1 line
2.17.1
Steffen Klassert
maintainerPATCH 4/8 xfrm: add the missing verify_sec_ctx_len check in xfrm_add_acquire
In reply to Steffen KlassertFrom: Xin Long <[email protected]>
Without doing verify_sec_ctx_len() check in xfrm_add_acquire(), it may be out-of-bounds to access uctx->ctx_str with uctx->ctx_len, as noticed by syz:
BUG: KASAN: slab-out-of-bounds in selinux_xfrm_alloc_user+0x237/0x430 Read of size 768 at addr ffff8880123be9b4 by task syz-executor.1/11650 Call Trace: dump_stack+0xe8/0x16e print_address_description.cold.3+0x9/0x23b kasan_report.cold.4+0x64/0x95 memcpy+0x1f/0x50 selinux_xfrm_alloc_user+0x237/0x430 security_xfrm_policy_alloc+0x5c/0xb0 xfrm_policy_construct+0x2b1/0x650 xfrm_add_acquire+0x21d/0xa10 xfrm_user_rcv_msg+0x431/0x6f0 netlink_rcv_skb+0x15a/0x410 xfrm_netlink_rcv+0x6d/0x90 netlink_unicast+0x50e/0x6a0 netlink_sendmsg+0x8ae/0xd40 sock_sendmsg+0x133/0x170 ___sys_sendmsg+0x834/0x9a0 __sys_sendmsg+0x100/0x1e0 do_syscall_64+0xe5/0x660 entry_SYSCALL_64_after_hwframe+0x6a/0xdf
So fix it by adding the missing verify_sec_ctx_len check there.
--- net/xfrm/xfrm_user.c | 3 +++ 1 file changed, 3 insertions(+)
net/xfrm/xfrm_user.c +3 -0
@@ -2274,6 +2274,9 @@ static int xfrm_add_acquire(struct sk_buff *skb, struct nlmsghdr *nlh,xfrm_mark_get (attrs, &mark); err =verify_newpolicy_info (&ua->policy); +if (err) + goto free_state; + err =verify_sec_ctx_len (attrs);if (err) goto free_state;
Signature · 1 line
2.17.1
Steffen Klassert
maintainerPATCH 5/8 ipv6: xfrm6_tunnel.c: Use built-in RCU list checking
In reply to Steffen KlassertFrom: Madhuparna Bhowmik <[email protected]>
hlist_for_each_entry_rcu() has built-in RCU and lock checking.
Pass cond argument to list_for_each_entry_rcu() to silence false lockdep warning when CONFIG_PROVE_RCU_LIST is enabled by default.
--- net/ipv6/xfrm6_tunnel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/xfrm6_tunnel.c +1 -1
@@ -78,7 +78,7 @@ static struct xfrm6_tunnel_spi *__xfrm6_tunnel_spi_lookup(struct net *net, consthlist_for_each_entry_rcu (x6spi, &xfrm6_tn->spi_byaddr[xfrm6_tunnel_spi_hash_byaddr (saddr)], - list_byaddr) { + list_byaddr,lockdep_is_held (&xfrm6_tunnel_spi_lock)) {if (xfrm6_addr_equal (&x6spi->addr, saddr))return x6spi; }
Signature · 1 line
2.17.1
Steffen Klassert
maintainerPATCH 6/8 esp: remove the skb from the chain when it's enqueued in cryptd_wq
In reply to Steffen KlassertFrom: Xin Long <[email protected]>
Xiumei found a panic in esp offload:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000020 RIP: 0010:esp_output_done+0x101/0x160 [esp4] Call Trace: ? esp_output+0x180/0x180 [esp4] cryptd_aead_crypt+0x4c/0x90 cryptd_queue_worker+0x6e/0xa0 process_one_work+0x1a7/0x3b0 worker_thread+0x30/0x390 ? create_worker+0x1a0/0x1a0 kthread+0x112/0x130 ? kthread_flush_work_fn+0x10/0x10 ret_from_fork+0x35/0x40
It was caused by that skb secpath is used in esp_output_done() after it's been released elsewhere.
The tx path for esp offload is:
__dev_queue_xmit()->
validate_xmit_skb_list()->
validate_xmit_xfrm()->
esp_xmit()->
esp_output_tail()->
aead_request_set_callback(esp_output_done) <--[1]
crypto_aead_encrypt() <--[2]In [1], .callback is set, and in [2] it will trigger the worker schedule, later on a kernel thread will call .callback(esp_output_done), as the call trace shows.
But in validate_xmit_xfrm():
skb_list_walk_safe(skb, skb2, nskb) {
...
err = x->type_offload->xmit(x, skb2, esp_features); [esp_xmit]
...
}When the err is -EINPROGRESS, which means this skb2 will be enqueued and later gets encrypted and sent out by .callback later in a kernel thread, skb2 should be removed fromt skb chain. Otherwise, it will get processed again outside validate_xmit_xfrm(), which could release skb secpath, and cause the panic above.
This patch is to remove the skb from the chain when it's enqueued in cryptd_wq. While at it, remove the unnecessary 'if (!skb)' check.
--- net/xfrm/xfrm_device.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
net/xfrm/xfrm_device.c +4 -4
@@ -78,8 +78,8 @@ struct sk_buff *validate_xmit_xfrm(struct sk_buff *skb, netdev_features_t featurint err;unsigned long flags;struct xfrm_state *x; -struct sk_buff *skb2, *nskb;struct softnet_data *sd; +struct sk_buff *skb2, *nskb, *pskb =NULL ;netdev_features_t esp_features = features;struct xfrm_offload *xo =xfrm_offload (skb);struct sec_path *sp;
@@ -168,14 +168,14 @@ struct sk_buff *validate_xmit_xfrm(struct sk_buff *skb, netdev_features_t featur }else {if (skb == skb2) skb = nskb; - -if (!skb) -return NULL ; +else + pskb->next = nskb;continue ; }skb_push (skb2, skb2->data -skb_mac_header (skb2)); + pskb = skb2; }return skb;
Signature · 1 line
2.17.1
Steffen Klassert
maintainerPATCH 7/8 vti6: Fix memory leak of skb if input policy check fails
In reply to Steffen KlassertFrom: Torsten Hilbrich <[email protected]>
The vti6_rcv function performs some tests on the retrieved tunnel including checking the IP protocol, the XFRM input policy, the source and destination address.
In all but one places the skb is released in the error case. When the input policy check fails the network packet is leaked.
Using the same goto-label discard in this case to fix this problem.
--- net/ipv6/ip6_vti.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/ip6_vti.c +1 -1
@@ -311,7 +311,7 @@ static int vti6_rcv(struct sk_buff *skb)if (!xfrm6_policy_check (NULL ,XFRM_POLICY_IN , skb)) {rcu_read_unlock (); -return 0 ; + goto discard; } ipv6h =ipv6_hdr (skb);
Signature · 1 line
2.17.1
Steffen Klassert
maintainerPATCH 8/8 xfrm: policy: Fix doulbe free in xfrm_policy_timer
In reply to Steffen KlassertFrom: YueHaibing <[email protected]>
After xfrm_add_policy add a policy, its ref is 2, then
xfrm_policy_timer
read_lock
xp->walk.dead is 0
....
mod_timer()xfrm_policy_kill
policy->walk.dead = 1
....
del_timer(&policy->timer)
xfrm_pol_put //ref is 1
xfrm_pol_put //ref is 0
xfrm_policy_destroy
call_rcu
xfrm_pol_hold //ref is 1
read_unlock xfrm_pol_put //ref is 0
xfrm_policy_destroy
call_rcu
xfrm_policy_destroy is called twice, which may leads to double free.
Call Trace: RIP: 0010:refcount_warn_saturate+0x161/0x210 ... xfrm_policy_timer+0x522/0x600 call_timer_fn+0x1b3/0x5e0 ? __xfrm_decode_session+0x2990/0x2990 ? msleep+0xb0/0xb0 ? _raw_spin_unlock_irq+0x24/0x40 ? __xfrm_decode_session+0x2990/0x2990 ? __xfrm_decode_session+0x2990/0x2990 run_timer_softirq+0x5c5/0x10e0
Fix this by use write_lock_bh in xfrm_policy_kill.
--- net/xfrm/xfrm_policy.c | 2 ++ 1 file changed, 2 insertions(+)
net/xfrm/xfrm_policy.c +2 -0
@@ -434,7 +434,9 @@ EXPORT_SYMBOL(xfrm_policy_destroy);static void xfrm_policy_kill (struct xfrm_policy *policy) { +write_lock_bh (&policy->lock); policy->walk.dead =1 ; +write_unlock_bh (&policy->lock);atomic_inc (&policy->genid);
Signature · 1 line
2.17.1
David Miller
maintainerFrom: Steffen Klassert <[email protected]>
Date: Fri, 27 Mar 2020 09:09:59 +0100
Quoted 19 lines
1) Handle
NETDEV_UNREGISTERfor xfrm device to handle asynchronous unregister events cleanly. From Raed Salem.2) Fix vti6 tunnel inter address family TX through
bpf_redirect(). From Nicolas Dichtel.3) Fix lenght check in
verify_sec_ctx_len()to avoid a slab-out-of-bounds. From Xin Long.4) Add a missing
verify_sec_ctx_lencheck inxfrm_add_acquireto avoid a possible out-of-bounds to access. From Xin Long.5) Use built-in RCU list checking of
hlist_for_each_entry_rcuto silence false lockdep warning in__xfrm6_tunnel_spi_lookupwhenCONFIG_PROVE_RCU_LISTis enabled. From Madhuparna Bhowmik.6) Fix a panic on esp offload when crypto is done asynchronously. From Xin Long.
Quoted 7 lines
7) Fix a skb memory leak in an error path of
vti6_rcv. From Torsten Hilbrich.8) Fix a race that can lead to a doulbe free in
xfrm_policy_timer. From Xin Long.Please pull or let me know if there are problems.
Pulled, thanks Steffen.