Skip to contentSkip to conversation

pull request (net): ipsec 2026-06-22

PATCH Applied Steffen Klassert 2 participants 9 messages Jun 22 – Jun 23

Subsystems (3)
lore
Share & tools
lore

Steffen Klassert

author maintainer
#0 lore raw

cover pull request (net): ipsec 2026-06-22

1) xfrm: use compat translator only for u64 alignment mismatch Gate the XFRM_USER_COMPAT translator on COMPAT_FOR_U64_ALIGNMENT so 32-bit compat tasks on arches whose 32-bit ABI already matches the native 64-bit layout are no longer rejected with -EOPNOTSUPP. From Sanman Pradhan.

2) net: af_key: initialize alg_key_len for IPComp states Initialize the alg_key_len to 0 in the IPComp branch of pfkey_msg2xfrm_state() so an uninitialized value cannot drive xfrm_alg_len() into a slab-out-of-bounds kmemdup during XFRM_MSG_MIGRATE. From Zijing Yin.

3) xfrm: Fix dev use-after-free in xfrm async resumption Stash the original skb->dev and extend the RCU critical section across xfrm_rcv_cb() and transport_finish() to prevent a tunnel-device UAF and original-device refcount leak when a callback replaces skb->dev. From Dong Chenchen.

4) xfrm: Fix xfrm state cache insertion race
Move the state-validity check inside xfrm_state_lock in the input state cache insertion path so a state cannot be killed between the check and the insert. From Herbert Xu.

5) xfrm: annotate data-races around xfrm_policy_count[] and xfrm_policy_default[] Add READ_ONCE()/WRITE_ONCE() annotations on xfrm_policy_count
and xfrm_policy_default to silence the KCSAN data race reported on net->xfrm.policy_count. From Eric Dumazet.

6) espintcp: use sk_msg_free_partial to fix partial send Replace the manual skmsg accounting in espintcp with sk_msg_free_partial() so the skmsg stays consistent on every iteration and the partial-send accounting bugs go away. From Sabrina Dubroca.

7) xfrm: validate selector family and prefixlen during match Reject mismatched address families in xfrm_selector_match() and bound prefixlen in addr4_match()/addr_match() to prevent the shift-out-of-bounds syzbot reported when an AF_UNSPEC selector with a large prefixlen is matched against an IPv4 flow. From Eric Dumazet.

Please pull or let me know if there are problems.

Thanks!

The following changes since commit 9bf10032894f429b3e221de63cf95a8544511a90:

Merge branch 'tipc-fix-netlink-gate-and-receive-path-bugs' (2026-06-11 16:01:19 -0700)

are available in the Git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec.git tags/ipsec-2026-06-22

for you to fetch changes up to 40f0b1047918539f0b0f795ac65e35336b4c2c78:

xfrm: validate selector family and prefixlen during match (2026-06-17 11:17:27 +0200)

----------------------------------------------------------------
ipsec-2026-06-22

----------------------------------------------------------------
Dong Chenchen (1):
xfrm: Fix dev use-after-free in xfrm async resumption

Eric Dumazet (2):
xfrm: annotate data-races around xfrm_policy_count[] and xfrm_policy_default[]
xfrm: validate selector family and prefixlen during match

Herbert Xu (1):
xfrm: Fix xfrm state cache insertion race

Sabrina Dubroca (1):
espintcp: use sk_msg_free_partial to fix partial send

Sanman Pradhan (1):
xfrm: use compat translator only for u64 alignment mismatch

Zijing Yin (1):
net: af_key: initialize alg_key_len for IPComp states

 include/net/xfrm.h     | 15 +++++++++++----
 net/ipv4/xfrm4_input.c |  2 --
 net/ipv6/xfrm6_input.c |  2 --
 net/key/af_key.c       |  1 +
 net/xfrm/espintcp.c    | 34 +++++++---------------------------
 net/xfrm/xfrm_input.c  | 29 ++++++++++++++++-------------
 net/xfrm/xfrm_policy.c | 27 +++++++++++++++------------
 net/xfrm/xfrm_state.c  | 23 +++++++++++++++--------
 net/xfrm/xfrm_user.c   | 20 ++++++++++----------
 9 files changed, 75 insertions(+), 78 deletions(-)
author maintainer
#1 lore raw

PATCH 1/7 xfrm: use compat translator only for u64 alignment mismatch

From: Sanman Pradhan <[email protected]>

The XFRM compat layer (CONFIG_XFRM_USER_COMPAT) translates 32-bit xfrm netlink and setsockopt messages into the native 64-bit layout. It is only needed on architectures where the 32-bit and 64-bit ABIs disagree on u64 alignment, which the kernel encodes as COMPAT_FOR_U64_ALIGNMENT.

That symbol is defined only by arch/x86. XFRM_USER_COMPAT depends on it, so the translator can never be built on any other architecture, including arm64, which still provides a 32-bit compat ABI (CONFIG_COMPAT) for AArch32 EL0 userspace. On arm64 the AArch32 EABI already aligns u64 to 8 bytes, identical to the AArch64 ABI, so no translation is required and the native code path is correct for 32-bit tasks.

However, xfrm_user_rcv_msg() and xfrm_user_policy() gate on in_compat_syscall() alone and then call xfrm_get_translator(), which returns NULL when no translator is registered. On arm64 that is always the case, so every xfrm netlink message and the XFRM_POLICY setsockopt issued by a 32-bit task returns -EOPNOTSUPP. A 32-bit userspace process on arm64 (and on any other arch with CONFIG_COMPAT but without COMPAT_FOR_U64_ALIGNMENT) therefore cannot configure XFRM state or policy through the XFRM_USER netlink API, and cannot use the XFRM_POLICY setsockopt path, because both fail before reaching the native parser.

The translator series replaced the blanket compat rejection with a translator lookup. That made the path usable on x86 when the translator is available, but left architectures that cannot build the translator permanently rejected even when their compat layout already matches the native layout. Let those architectures use the native parser instead.

Gate the translator requirement on COMPAT_FOR_U64_ALIGNMENT instead of on in_compat_syscall() alone. Gating on the ABI property rather than on CONFIG_XFRM_USER_COMPAT is deliberate: on x86 with IA32_EMULATION=y but XFRM_USER_COMPAT=n, a 32-bit task must still be rejected rather than routed through the native parser, which would misread genuinely 4-byte-aligned x86-32 messages. COMPAT_FOR_U64_ALIGNMENT is the ABI property that makes the XFRM translator mandatory.

Only the receive/input direction needs the guard. The send, dump and notification paths already call the translator as "if (xtr) { ... }" with no error on NULL, so on arches without a translator they no-op and the kernel emits native 64-bit-layout messages, which is what an AArch32 task expects.

Tested on Juniper SRX hardware: with the fix, 32-bit IPsec userspace netlink and XFRM_POLICY setsockopt operations that previously failed with -EOPNOTSUPP now succeed; x86 behaviour is unchanged by inspection.

Fixes 5106f4a8acff ("xfrm/compat: Add 32=>64-bit messages translator")
Fixes 96392ee5a13b ("xfrm/compat: Translate 32-bit user_policy from sockptr")
---
 net/xfrm/xfrm_state.c | 2 +-
 net/xfrm/xfrm_user.c  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
net/xfrm/xfrm_state.c +1 -1
index 589c3b6e4679..d8457ceaf28c 100644
@@ -2976,7 +2976,7 @@ int xfrm_user_policy(struct sock *sk, int optname, sockptr_t optval, int optlen)
 	if (IS_ERR(data))
 		return PTR_ERR(data);
 
-	if (in_compat_syscall()) {
+	if (IS_ENABLED(CONFIG_COMPAT_FOR_U64_ALIGNMENT) && in_compat_syscall()) {
 		struct xfrm_translator *xtr = xfrm_get_translator();
 
 		if (!xtr) {
net/xfrm/xfrm_user.c +1 -1
index 71a4b7278eba..3b1cf29bc402 100644
@@ -3472,7 +3472,7 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (!netlink_net_capable(skb, CAP_NET_ADMIN))
 		return -EPERM;
 
-	if (in_compat_syscall()) {
+	if (IS_ENABLED(CONFIG_COMPAT_FOR_U64_ALIGNMENT) && in_compat_syscall()) {
 		struct xfrm_translator *xtr = xfrm_get_translator();
 
 		if (!xtr)
Signature · 1 line
2.43.0
#2 lore raw

Re: [PATCH 1/7] xfrm: use compat translator only for u64 alignment mismatch

This series was applied to netdev/net.git (main) by Steffen Klassert <[email protected]>:

Hello:

This series was applied to netdev/net.git (main) by Steffen Klassert <[email protected]>:

Quoted 8 lines · On Mon, 22 Jun 2026 09:57:03 +0200 you wrote:

From: Sanman Pradhan <[email protected]>

The XFRM compat layer (CONFIG_XFRM_USER_COMPAT) translates 32-bit xfrm netlink and setsockopt messages into the native 64-bit layout. It is only needed on architectures where the 32-bit and 64-bit ABIs disagree on u64 alignment, which the kernel encodes as COMPAT_FOR_U64_ALIGNMENT.

[...]

Here is the summary with links:
- [1/7] xfrm: use compat translator only for u64 alignment mismatch

    https://git.kernel.org/netdev/net/c/355fbcbdc253
  - [2/7] net: af_key: initialize alg_key_len for IPComp states
    https://git.kernel.org/netdev/net/c/d129c3177d7b
  - [3/7] xfrm: Fix dev use-after-free in xfrm async resumption
    https://git.kernel.org/netdev/net/c/8045c0df98d4
  - [4/7] xfrm: Fix xfrm state cache insertion race
    https://git.kernel.org/netdev/net/c/ddd3d0132920
  - [5/7] xfrm: annotate data-races around xfrm_policy_count[] and xfrm_policy_default[]
    https://git.kernel.org/netdev/net/c/68de007d5ac9
  - [6/7] espintcp: use sk_msg_free_partial to fix partial send
    https://git.kernel.org/netdev/net/c/007800408002
  - [7/7] xfrm: validate selector family and prefixlen during match
    https://git.kernel.org/netdev/net/c/40f0b1047918

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
author maintainer
#3 lore raw

PATCH 2/7 net: af_key: initialize alg_key_len for IPComp states

In reply to Steffen Klassert

From: Zijing Yin <[email protected]>

pfkey_msg2xfrm_state() handles the IPComp (SADB_X_SATYPE_IPCOMP) case by allocating x->calg and copying only the algorithm name:

	x->calg = kmalloc_obj(*x->calg);
	if (!x->calg) {
		err = -ENOMEM;
		goto out;
	}
	strcpy(x->calg->alg_name, a->name);
	x->props.calgo = sa->sadb_sa_encrypt;

Unlike the authentication (x->aalg) and encryption (x->ealg) branches of the same function, the compression branch never initializes calg->alg_key_len. IPComp carries no key and the allocation only reserves sizeof(struct xfrm_algo) (i.e. no room for a key), so the field is left containing uninitialized slab data.

calg->alg_key_len is later used as a length by xfrm_algo_clone() when an IPComp state is cloned during XFRM_MSG_MIGRATE:

	xfrm_state_migrate()
	  xfrm_state_clone_and_setup()
	    x->calg = xfrm_algo_clone(orig->calg);
	      kmemdup(orig, xfrm_alg_len(orig));

where xfrm_alg_len() returns sizeof(*alg) + (alg_key_len + 7) / 8. With a non-zero garbage alg_key_len, kmemdup() reads past the end of the 68-byte calg object. Adding an IPComp SA via PF_KEY and then migrating it triggers (net-next, KASAN, init_on_alloc=0):

  BUG: KASAN: slab-out-of-bounds in kmemdup_noprof+0x44/0x60
  Read of size 4164 at addr ff11000025a74980 by task diag2/9287
  CPU: 3 UID: 0 PID: 9287 Comm: diag2 7.1.0-rc6-g903db046d557 #1
  Call Trace:
   <TASK>
   dump_stack_lvl+0x10e/0x1f0
   print_report+0xf7/0x600
   kasan_report+0xe4/0x120
   kasan_check_range+0x105/0x1b0
   __asan_memcpy+0x23/0x60
   kmemdup_noprof+0x44/0x60
   xfrm_state_migrate+0x70a/0x1da0
   xfrm_migrate+0x753/0x18a0
   xfrm_do_migrate+0xb47/0xf10
   xfrm_user_rcv_msg+0x411/0xb50
   netlink_rcv_skb+0x158/0x420
   xfrm_netlink_rcv+0x71/0x90
   netlink_unicast+0x584/0x850
   netlink_sendmsg+0x8b0/0xdc0
   ____sys_sendmsg+0x9f7/0xb90
   ___sys_sendmsg+0x134/0x1d0
   __sys_sendmsg+0x16d/0x220
   do_syscall_64+0x116/0x7d0
   entry_SYSCALL_64_after_hwframe+0x77/0x7f
   </TASK>

  Allocated by task 9287:
   kasan_save_stack+0x33/0x60
   kasan_save_track+0x14/0x30
   __kasan_kmalloc+0xaa/0xb0
   pfkey_add+0x2652/0x2ea0
   pfkey_process+0x6d0/0x830
   pfkey_sendmsg+0x42c/0x850
   __sys_sendto+0x461/0x4b0
   __x64_sys_sendto+0xe0/0x1c0
   do_syscall_64+0x116/0x7d0
   entry_SYSCALL_64_after_hwframe+0x77/0x7f

  The buggy address belongs to the object at ff11000025a74980
   which belongs to the cache kmalloc-96 of size 96
  The buggy address is located 0 bytes inside of
   allocated 68-byte region [ff11000025a74980, ff11000025a749c4)

Depending on the uninitialized value the same field can instead request an oversized kmemdup() allocation and make the migration clone fail.

The XFRM netlink path is not affected: verify_one_alg() rejects an XFRMA_ALG_COMP attribute shorter than xfrm_alg_len(), so a calg added via XFRM_MSG_NEWSA is always self-consistent.

Initialize calg->alg_key_len to 0, matching the aalg/ealg branches.

Fixes 80c9abaabf42 ("[XFRM]: Extension for dynamic update of endpoint address(es)")
---
 net/key/af_key.c | 1 +
 1 file changed, 1 insertion(+)
net/key/af_key.c +1 -0
index 9cffeef18cd9..3216f897a305 100644
@@ -1218,6 +1218,7 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct net *net,
 				goto out;
 			}
 			strcpy(x->calg->alg_name, a->name);
+			x->calg->alg_key_len = 0;
 			x->props.calgo = sa->sadb_sa_encrypt;
 		} else {
 			int keysize = 0;
Signature · 1 line
2.43.0
author maintainer
#4 lore raw

PATCH 3/7 xfrm: Fix dev use-after-free in xfrm async resumption

In reply to Steffen Klassert

From: Dong Chenchen <[email protected]>

xfrm async resumption hold skb->dev refcnt until after transport_finish. However, xfrm_rcv_cb may modify skb->dev to tunnel dev without taking device reference, such as vti_rcv_cb. The subsequent async resumption will decrement the tunnel device's reference count, which lead to uaf of tunnel dev and refcnt leak of orig dev as below:

unregister_netdevice: waiting for vti1 to become free. Usage count = -2

Stash the original skb->dev to fix refcnt imbalance. The new skb->dev set by xfrm_rcv_cb can race with device teardown. Extend rcu protection over xfrm_rcv_cb and transport_finish to prevent races.

Fixes 1c428b038400 ("xfrm: hold dev ref until after transport_finish NF_HOOK")
---
 net/ipv4/xfrm4_input.c |  2 --
 net/ipv6/xfrm6_input.c |  2 --
 net/xfrm/xfrm_input.c  | 29 ++++++++++++++++-------------
 3 files changed, 16 insertions(+), 17 deletions(-)
net/ipv4/xfrm4_input.c +0 -2
index c2eac844bcdb..f6f2a8ef3f88 100644
@@ -76,8 +76,6 @@ int xfrm4_transport_finish(struct sk_buff *skb, int async)
 	NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING,
 		dev_net(dev), NULL, skb, dev, NULL,
 		xfrm4_rcv_encap_finish);
-	if (async)
-		dev_put(dev);
 	return 0;
 }
 
net/ipv6/xfrm6_input.c +0 -2
index 699a001ac166..89d0443b5307 100644
@@ -71,8 +71,6 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async)
 	NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING,
 		dev_net(dev), NULL, skb, dev, NULL,
 		xfrm6_transport_finish2);
-	if (async)
-		dev_put(dev);
 	return 0;
 }
 
net/xfrm/xfrm_input.c +16 -13
index e4c2cd24936d..eecab337bd0a 100644
@@ -467,6 +467,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 {
 	const struct xfrm_state_afinfo *afinfo;
 	struct net *net = dev_net(skb->dev);
+	struct net_device *dev = skb->dev;
 	int err;
 	__be32 seq;
 	__be32 seq_hi;
@@ -493,7 +494,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 					       LINUX_MIB_XFRMINSTATEINVALID);
 
 			if (encap_type == -1)
-				dev_put(skb->dev);
+				dev_put(dev);
 			goto drop;
 		}
 
@@ -655,16 +656,16 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 
 		if (!crypto_done) {
 			spin_unlock(&x->lock);
-			dev_hold(skb->dev);
+			dev_hold(dev);
 
 			nexthdr = x->type->input(x, skb);
 			if (nexthdr == -EINPROGRESS) {
 				if (async)
-					dev_put(skb->dev);
+					dev_put(dev);
 				return 0;
 			}
 
-			dev_put(skb->dev);
+			dev_put(dev);
 			spin_lock(&x->lock);
 		}
 resume:
@@ -699,7 +700,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		err = xfrm_inner_mode_input(x, skb);
 		if (err == -EINPROGRESS) {
 			if (async)
-				dev_put(skb->dev);
+				dev_put(dev);
 			return 0;
 		} else if (err) {
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEMODEERROR);
@@ -726,9 +727,12 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		crypto_done = false;
 	} while (!err);
 
+	rcu_read_lock();
 	err = xfrm_rcv_cb(skb, family, x->type->proto, 0);
-	if (err)
+	if (err) {
+		rcu_read_unlock();
 		goto drop;
+	}
 
 	nf_reset_ct(skb);
 
@@ -739,8 +743,9 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		if (skb_valid_dst(skb))
 			skb_dst_drop(skb);
 		if (async)
-			dev_put(skb->dev);
+			dev_put(dev);
 		gro_cells_receive(&gro_cells, skb);
+		rcu_read_unlock();
 		return 0;
 	} else {
 		xo = xfrm_offload(skb);
@@ -748,23 +753,21 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 			xfrm_gro = xo->flags & XFRM_GRO;
 
 		err = -EAFNOSUPPORT;
-		rcu_read_lock();
 		afinfo = xfrm_state_afinfo_get_rcu(x->props.family);
 		if (likely(afinfo))
 			err = afinfo->transport_finish(skb, xfrm_gro || async);
-		rcu_read_unlock();
 		if (xfrm_gro) {
 			sp = skb_sec_path(skb);
 			if (sp)
 				sp->olen = 0;
 			if (skb_valid_dst(skb))
 				skb_dst_drop(skb);
-			if (async)
-				dev_put(skb->dev);
 			gro_cells_receive(&gro_cells, skb);
-			return err;
 		}
 
+		if (async)
+			dev_put(dev);
+		rcu_read_unlock();
 		return err;
 	}
 
@@ -772,7 +775,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	spin_unlock(&x->lock);
 drop:
 	if (async)
-		dev_put(skb->dev);
+		dev_put(dev);
 	xfrm_rcv_cb(skb, family, x && x->type ? x->type->proto : nexthdr, -1);
 	kfree_skb(skb);
 	return 0;
Signature · 1 line
2.43.0
author maintainer
#5 lore raw

PATCH 4/7 xfrm: Fix xfrm state cache insertion race

In reply to Steffen Klassert

From: Herbert Xu <[email protected]>

The xfrm input state cache insertion code checks the validity of the state before acquiring the global xfrm_state_lock. Thus it's possible for someone else to kill the state after it passed the validity check, and then the insertion will add the dead state to the cache.

Fix this by moving the validity check inside the lock.

This entire function is called on the input path, where BH must be off (e.g., the caller of this function xfrm_input acquires its spinlocks without disabling BH).

So there is no need to disable BH here or take the RCU read lock. Remove both and replace them with an assertion that trips if BH is accidentally enabled on some future calling path.

Fixes 81a331a0e72d ("xfrm: Add an inbound percpu state cache.")
---
 net/xfrm/xfrm_state.c | 21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)
net/xfrm/xfrm_state.c +14 -7
index d8457ceaf28c..9e87f7028201 100644
@@ -1207,9 +1207,11 @@ struct xfrm_state *xfrm_input_state_lookup(struct net *net, u32 mark,
 	struct hlist_head *state_cache_input;
 	struct xfrm_state *x = NULL;
 
+	/* BH is always disabled on the input path. */
+	lockdep_assert_in_softirq();
+
 	state_cache_input = raw_cpu_ptr(net->xfrm.state_cache_input);
 
-	rcu_read_lock();
 	hlist_for_each_entry_rcu(x, state_cache_input, state_cache_input) {
 		if (x->props.family != family ||
 		    x->id.spi       != spi ||
@@ -1227,20 +1229,25 @@ struct xfrm_state *xfrm_input_state_lookup(struct net *net, u32 mark,
 	xfrm_hash_ptrs_get(net, &state_ptrs);
 
 	x = __xfrm_state_lookup(&state_ptrs, mark, daddr, spi, proto, family);
-
-	if (x && x->km.state == XFRM_STATE_VALID) {
-		spin_lock_bh(&net->xfrm.xfrm_state_lock);
-		if (hlist_unhashed(&x->state_cache_input)) {
+	if (x) {
+		spin_lock(&net->xfrm.xfrm_state_lock);
+		if (x->km.state != XFRM_STATE_VALID) {
+			/*
+			 * The state is about to be destroyed.
+			 *
+			 * Don't add it to the cache but still
+			 * return it to the caller.
+			 */
+		} else if (hlist_unhashed(&x->state_cache_input)) {
 			hlist_add_head_rcu(&x->state_cache_input, state_cache_input);
 		} else {
 			hlist_del_rcu(&x->state_cache_input);
 			hlist_add_head_rcu(&x->state_cache_input, state_cache_input);
 		}
-		spin_unlock_bh(&net->xfrm.xfrm_state_lock);
+		spin_unlock(&net->xfrm.xfrm_state_lock);
 	}
 
 out:
-	rcu_read_unlock();
 	return x;
 }
 EXPORT_SYMBOL(xfrm_input_state_lookup);
Signature · 1 line
2.43.0
author maintainer
#6 lore raw

PATCH 5/7 xfrm: annotate data-races around xfrm_policy_count[] and xfrm_policy_default[]

In reply to Steffen Klassert

From: Eric Dumazet <[email protected]>

KCSAN reported a data race involving net->xfrm.policy_count access.

Add missing READ_ONCE()/WRITE_ONCE() annotations on xfrm_policy_count and xfrm_policy_default.

Fixes 2518c7c2b3d7 ("[XFRM]: Hash policies when non-prefixed.")
Reported-by [email protected]
---
 include/net/xfrm.h     |  8 ++++----
 net/xfrm/xfrm_policy.c | 24 ++++++++++++------------
 net/xfrm/xfrm_user.c   | 18 +++++++++---------
 3 files changed, 25 insertions(+), 25 deletions(-)
include/net/xfrm.h +4 -4
index 874409127e29..35a743129329 100644
@@ -1250,8 +1250,8 @@ int __xfrm_policy_check(struct sock *, int dir, struct sk_buff *skb,
 static inline bool __xfrm_check_nopolicy(struct net *net, struct sk_buff *skb,
 					 int dir)
 {
-	if (!net->xfrm.policy_count[dir] && !secpath_exists(skb))
-		return net->xfrm.policy_default[dir] == XFRM_USERPOLICY_ACCEPT;
+	if (!READ_ONCE(net->xfrm.policy_count[dir]) && !secpath_exists(skb))
+		return READ_ONCE(net->xfrm.policy_default[dir]) == XFRM_USERPOLICY_ACCEPT;
 
 	return false;
 }
@@ -1351,8 +1351,8 @@ static inline int xfrm_route_forward(struct sk_buff *skb, unsigned short family)
 {
 	struct net *net = dev_net(skb->dev);
 
-	if (!net->xfrm.policy_count[XFRM_POLICY_OUT] &&
-	    net->xfrm.policy_default[XFRM_POLICY_OUT] == XFRM_USERPOLICY_ACCEPT)
+	if (!READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_OUT]) &&
+	    READ_ONCE(net->xfrm.policy_default[XFRM_POLICY_OUT]) == XFRM_USERPOLICY_ACCEPT)
 		return true;
 
 	return (skb_dst(skb)->flags & DST_NOXFRM) ||
net/xfrm/xfrm_policy.c +12 -12
index 959544425692..1f4afd580105 100644
@@ -685,7 +685,7 @@ static void xfrm_byidx_resize(struct net *net)
 
 static inline int xfrm_bydst_should_resize(struct net *net, int dir, int *total)
 {
-	unsigned int cnt = net->xfrm.policy_count[dir];
+	unsigned int cnt = READ_ONCE(net->xfrm.policy_count[dir]);
 	unsigned int hmask = net->xfrm.policy_bydst[dir].hmask;
 
 	if (total)
@@ -711,12 +711,12 @@ static inline int xfrm_byidx_should_resize(struct net *net, int total)
 
 void xfrm_spd_getinfo(struct net *net, struct xfrmk_spdinfo *si)
 {
-	si->incnt = net->xfrm.policy_count[XFRM_POLICY_IN];
-	si->outcnt = net->xfrm.policy_count[XFRM_POLICY_OUT];
-	si->fwdcnt = net->xfrm.policy_count[XFRM_POLICY_FWD];
-	si->inscnt = net->xfrm.policy_count[XFRM_POLICY_IN+XFRM_POLICY_MAX];
-	si->outscnt = net->xfrm.policy_count[XFRM_POLICY_OUT+XFRM_POLICY_MAX];
-	si->fwdscnt = net->xfrm.policy_count[XFRM_POLICY_FWD+XFRM_POLICY_MAX];
+	si->incnt = READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_IN]);
+	si->outcnt = READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_OUT]);
+	si->fwdcnt = READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_FWD]);
+	si->inscnt = READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_IN+XFRM_POLICY_MAX]);
+	si->outscnt = READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_OUT+XFRM_POLICY_MAX]);
+	si->fwdscnt = READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_FWD+XFRM_POLICY_MAX]);
 	si->spdhcnt = net->xfrm.policy_idx_hmask;
 	si->spdhmcnt = xfrm_policy_hashmax;
 }
@@ -2318,7 +2318,7 @@ static void __xfrm_policy_link(struct xfrm_policy *pol, int dir)
 	}
 
 	list_add(&pol->walk.all, &net->xfrm.policy_all);
-	net->xfrm.policy_count[dir]++;
+	WRITE_ONCE(net->xfrm.policy_count[dir], net->xfrm.policy_count[dir] + 1);
 	xfrm_pol_hold(pol);
 }
 
@@ -2337,7 +2337,7 @@ static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol,
 	}
 
 	list_del_init(&pol->walk.all);
-	net->xfrm.policy_count[dir]--;
+	WRITE_ONCE(net->xfrm.policy_count[dir], net->xfrm.policy_count[dir] - 1);
 
 	return pol;
 }
@@ -3222,7 +3222,7 @@ struct dst_entry *xfrm_lookup_with_ifid(struct net *net,
 
 		/* To accelerate a bit...  */
 		if (!if_id && ((dst_orig->flags & DST_NOXFRM) ||
-			       !net->xfrm.policy_count[XFRM_POLICY_OUT]))
+			       !READ_ONCE(net->xfrm.policy_count[XFRM_POLICY_OUT])))
 			goto nopol;
 
 		xdst = xfrm_bundle_lookup(net, fl, family, dir, &xflo, if_id);
@@ -3296,7 +3296,7 @@ struct dst_entry *xfrm_lookup_with_ifid(struct net *net,
 
 nopol:
 	if ((!dst_orig->dev || !(dst_orig->dev->flags & IFF_LOOPBACK)) &&
-	    net->xfrm.policy_default[dir] == XFRM_USERPOLICY_BLOCK) {
+	    READ_ONCE(net->xfrm.policy_default[dir]) == XFRM_USERPOLICY_BLOCK) {
 		err = -EPERM;
 		goto error;
 	}
@@ -3750,7 +3750,7 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		const bool is_crypto_offload = sp &&
 			(xfrm_input_state(skb)->xso.type == XFRM_DEV_OFFLOAD_CRYPTO);
 
-		if (net->xfrm.policy_default[dir] == XFRM_USERPOLICY_BLOCK) {
+		if (READ_ONCE(net->xfrm.policy_default[dir]) == XFRM_USERPOLICY_BLOCK) {
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOPOLS);
 			return 0;
 		}
net/xfrm/xfrm_user.c +9 -9
index 3b1cf29bc402..61eb5de33b87 100644
@@ -2485,9 +2485,9 @@ static int xfrm_notify_userpolicy(struct net *net)
 	}
 
 	up = nlmsg_data(nlh);
-	up->in = net->xfrm.policy_default[XFRM_POLICY_IN];
-	up->fwd = net->xfrm.policy_default[XFRM_POLICY_FWD];
-	up->out = net->xfrm.policy_default[XFRM_POLICY_OUT];
+	up->in = READ_ONCE(net->xfrm.policy_default[XFRM_POLICY_IN]);
+	up->fwd = READ_ONCE(net->xfrm.policy_default[XFRM_POLICY_FWD]);
+	up->out = READ_ONCE(net->xfrm.policy_default[XFRM_POLICY_OUT]);
 
 	nlmsg_end(skb, nlh);
 
@@ -2511,13 +2511,13 @@ static int xfrm_set_default(struct sk_buff *skb, struct nlmsghdr *nlh,
 	struct xfrm_userpolicy_default *up = nlmsg_data(nlh);
 
 	if (xfrm_userpolicy_is_valid(up->in))
-		net->xfrm.policy_default[XFRM_POLICY_IN] = up->in;
+		WRITE_ONCE(net->xfrm.policy_default[XFRM_POLICY_IN], up->in);
 
 	if (xfrm_userpolicy_is_valid(up->fwd))
-		net->xfrm.policy_default[XFRM_POLICY_FWD] = up->fwd;
+		WRITE_ONCE(net->xfrm.policy_default[XFRM_POLICY_FWD], up->fwd);
 
 	if (xfrm_userpolicy_is_valid(up->out))
-		net->xfrm.policy_default[XFRM_POLICY_OUT] = up->out;
+		WRITE_ONCE(net->xfrm.policy_default[XFRM_POLICY_OUT], up->out);
 
 	rt_genid_bump_all(net);
 
@@ -2547,9 +2547,9 @@ static int xfrm_get_default(struct sk_buff *skb, struct nlmsghdr *nlh,
 	}
 
 	r_up = nlmsg_data(r_nlh);
-	r_up->in = net->xfrm.policy_default[XFRM_POLICY_IN];
-	r_up->fwd = net->xfrm.policy_default[XFRM_POLICY_FWD];
-	r_up->out = net->xfrm.policy_default[XFRM_POLICY_OUT];
+	r_up->in = READ_ONCE(net->xfrm.policy_default[XFRM_POLICY_IN]);
+	r_up->fwd = READ_ONCE(net->xfrm.policy_default[XFRM_POLICY_FWD]);
+	r_up->out = READ_ONCE(net->xfrm.policy_default[XFRM_POLICY_OUT]);
 	nlmsg_end(r_skb, r_nlh);
 
 	return nlmsg_unicast(xfrm_net_nlsk(net, skb), r_skb, portid);
Signature · 1 line
2.43.0
author maintainer
#7 lore raw

PATCH 6/7 espintcp: use sk_msg_free_partial to fix partial send

In reply to Steffen Klassert

From: Sabrina Dubroca <[email protected]>

sk_msg_free_partial() ensures consistency of the skmsg at every iteration, without having to manually handle uncharges and offsets. This simplifies the code, and fixes some bugs in skmsg accounting when we don't send the full contents.

Fixes e27cca96cd68 ("xfrm: add espintcp (RFC 8229)")
---
 net/xfrm/espintcp.c | 34 +++++++---------------------------
 1 file changed, 7 insertions(+), 27 deletions(-)
net/xfrm/espintcp.c +7 -27
index d9035546375e..374e1b964438 100644
@@ -212,43 +212,23 @@ static int espintcp_sendskmsg_locked(struct sock *sk,
 	struct sk_msg *skmsg = &emsg->skmsg;
 	bool more = flags & MSG_MORE;
 	struct scatterlist *sg;
-	int done = 0;
 	int ret;
 
-	sg = &skmsg->sg.data[skmsg->sg.start];
 	do {
 		struct bio_vec bvec;
-		size_t size = sg->length - emsg->offset;
-		int offset = sg->offset + emsg->offset;
-		struct page *p;
-
-		emsg->offset = 0;
 
+		sg = &skmsg->sg.data[skmsg->sg.start];
 		if (sg_is_last(sg) && !more)
 			msghdr.msg_flags &= ~MSG_MORE;
 
-		p = sg_page(sg);
-retry:
-		bvec_set_page(&bvec, p, size, offset);
-		iov_iter_bvec(&msghdr.msg_iter, ITER_SOURCE, &bvec, 1, size);
-		ret = tcp_sendmsg_locked(sk, &msghdr, size);
-		if (ret < 0) {
-			emsg->offset = offset - sg->offset;
-			skmsg->sg.start += done;
+		bvec_set_page(&bvec, sg_page(sg), sg->length, sg->offset);
+		iov_iter_bvec(&msghdr.msg_iter, ITER_SOURCE, &bvec, 1, sg->length);
+		ret = tcp_sendmsg_locked(sk, &msghdr, sg->length);
+		if (ret < 0)
 			return ret;
-		}
-
-		if (ret != size) {
-			offset += ret;
-			size -= ret;
-			goto retry;
-		}
 
-		done++;
-		put_page(p);
-		sk_mem_uncharge(sk, sg->length);
-		sg = sg_next(sg);
-	} while (sg);
+		sk_msg_free_partial(sk, skmsg, ret);
+	} while (skmsg->sg.size);
 
 	memset(emsg, 0, sizeof(*emsg));
 
Signature · 1 line
2.43.0
author maintainer
#8 lore raw

PATCH 7/7 xfrm: validate selector family and prefixlen during match

In reply to Steffen Klassert

From: Eric Dumazet <[email protected]>

syzbot reported a shift-out-of-bounds in xfrm_selector_match() due to AF_UNSPEC selector with large prefixlen (e.g. 128) matched against IPv4 flow (when XFRM_STATE_AF_UNSPEC is set).

Fix this by:

- Rejecting mismatched families in xfrm_selector_match.
- Returning false in addr4_match if prefixlen > 32.
- Returning false in addr_match if prefixlen > 128 (prevents overflow).

Fixes 3f0ab59e6537 ("xfrm: validate new SA's prefixlen using SA family when sel.family is unset")
Reported-by [email protected]
---
 include/net/xfrm.h     | 7 +++++++
 net/xfrm/xfrm_policy.c | 3 +++
 2 files changed, 10 insertions(+)
include/net/xfrm.h +7 -0
index 35a743129329..f8c909b0f0c3 100644
@@ -943,6 +943,9 @@ static inline bool addr_match(const void *token1, const void *token2,
 	unsigned int pdw;
 	unsigned int pbi;
 
+	if (prefixlen > 128)
+		return false;
+
 	pdw = prefixlen >> 5;	  /* num of whole u32 in prefix */
 	pbi = prefixlen &  0x1f;  /* num of bits in incomplete u32 in prefix */
 
@@ -967,6 +970,10 @@ static inline bool addr4_match(__be32 a1, __be32 a2, u8 prefixlen)
 	/* C99 6.5.7 (3): u32 << 32 is undefined behaviour */
 	if (sizeof(long) == 4 && prefixlen == 0)
 		return true;
+
+	if (prefixlen > 32)
+		return false;
+
 	return !((a1 ^ a2) & htonl(~0UL << (32 - prefixlen)));
 }
 
net/xfrm/xfrm_policy.c +3 -0
index 1f4afd580105..639934f30016 100644
@@ -242,6 +242,9 @@ __xfrm6_selector_match(const struct xfrm_selector *sel, const struct flowi *fl)
 bool xfrm_selector_match(const struct xfrm_selector *sel, const struct flowi *fl,
 			 unsigned short family)
 {
+	if (family != sel->family && sel->family != AF_UNSPEC)
+		return false;
+
 	switch (family) {
 	case AF_INET:
 		return __xfrm4_selector_match(sel, fl);
Signature · 1 line
2.43.0