Skip to contentSkip to conversation

xfrm: Introduce direction attribute for SA

PATCH Superseded Antony Antony 4 participants 9 messages Apr 21, 2024 – Apr 23, 2024

Subsystems (3)

v11 of 14 Superseded Latest is v14 →

lore
Share & tools
lore
#0 lore raw

v11 cover xfrm: Introduce direction attribute for SA

Hi,

Inspired by the upcoming IP-TFS patch set, and confusions experienced in the past due to lack of direction attribute on SAs, add a new direction "dir" attribute. It aims to streamline the SA configuration process and enhance the clarity of existing SA attributes.

This patch set introduces the 'dir' attribute to SA, aka xfrm_state, ('in' for input or 'out' for output). Alsp add validations of existing direction-specific SA attributes during configuration and in the data path lookup.

This change would not affect any existing use case or way of configuring SA. You will notice improvements when the new 'dir' attribute is set.

Antony Antony (4):
xfrm: Add Direction to the SA in or out
xfrm: Add dir validation to "out" data path lookup
xfrm: Add dir validation to "in" data path lookup
xfrm: Restrict SA direction attribute to specific netlink message types

 include/net/xfrm.h        |   1 +
 include/uapi/linux/snmp.h |   2 +
 include/uapi/linux/xfrm.h |   6 ++
 net/ipv6/xfrm6_input.c    |   7 ++
 net/xfrm/xfrm_compat.c    |   7 +-
 net/xfrm/xfrm_device.c    |   6 ++
 net/xfrm/xfrm_input.c     |  11 +++
 net/xfrm/xfrm_policy.c    |   6 ++
 net/xfrm/xfrm_proc.c      |   2 +
 net/xfrm/xfrm_replay.c    |   3 +-
 net/xfrm/xfrm_state.c     |   5 ++
 net/xfrm/xfrm_user.c      | 147 ++++++++++++++++++++++++++++++++++++--
 12 files changed, 194 insertions(+), 9 deletions(-)

--
2.30.2

#1 lore raw

PATCH v11 1/4 xfrm: Add Direction to the SA in or out

This patch introduces the 'dir' attribute, 'in' or 'out', to the xfrm_state, SA, enhancing usability by delineating the scope of values based on direction. An input SA will restrict values pertinent to input, effectively segregating them from output-related values.
And an output SA will restrict attributes for output. This change aims to streamline the configuration process and improve the overall consistency of SA attributes during configuration.

This feature sets the groundwork for future patches, including the upcoming IP-TFS patch.

v10->v11:
- set dir out for acquire state
- allow output state with ESN to set replay-window 0
- change XFRMA_SA_DIR using NLA_POLICY_RANGE()
- replace verify_sa_dir() with refactored code.

v9->v10:
- add more direction specific validations XFRM_STATE_NOPMTUDISC, XFRM_SA_XFLAG_DONT_ENCAP_DSCP XFRMA_MTIMER_THRESH
- refactor validations into a fuction.
- add dir to ALLOCSPI to support strongSwan updating SPI to "in" state

v8->v9:
- add validation XFRM_STATE_ICMP not allowed on OUT SA.

v7->v8:
- add extra validation check on replay window and seq
- XFRM_MSG_UPDSA old and new SA should match "dir"

v6->v7:
- add replay-window check non-esn 0 and ESN 1.
- remove :XFRMA_SA_DIR only allowed with HW OFFLOAD

v5->v6:
- XFRMA_SA_DIR only allowed with HW OFFLOAD

v4->v5:
- add details to commit message

v3->v4:
- improve HW OFFLOAD DIR check add the other direction

v2->v3:
- delete redundant XFRM_SA_DIR_USE
- use u8 for "dir"
- fix HW OFFLOAD DIR check

v1->v2:
- use .strict_start_type in struct nla_policy xfrma_policy
- delete redundant XFRM_SA_DIR_MAX enum

---
 include/net/xfrm.h        |   1 +
 include/uapi/linux/xfrm.h |   6 ++
 net/xfrm/xfrm_compat.c    |   7 ++-
 net/xfrm/xfrm_device.c    |   6 ++
 net/xfrm/xfrm_replay.c    |   3 +-
 net/xfrm/xfrm_state.c     |   5 ++
 net/xfrm/xfrm_user.c      | 125 ++++++++++++++++++++++++++++++++++++--
 7 files changed, 144 insertions(+), 9 deletions(-)
include/net/xfrm.h +1 -0
index 57c743b7e4fe..7c9be06f8302 100644
@@ -291,6 +291,7 @@ struct xfrm_state {
 	/* Private data of this transformer, format is opaque,
 	 * interpreted by xfrm_type methods. */
 	void			*data;
+	u8			dir;
 };
 
 static inline struct net *xs_net(struct xfrm_state *x)
include/uapi/linux/xfrm.h +6 -0
index 6a77328be114..18ceaba8486e 100644
@@ -141,6 +141,11 @@ enum {
 	XFRM_POLICY_MAX	= 3
 };
 
+enum xfrm_sa_dir {
+	XFRM_SA_DIR_IN	= 1,
+	XFRM_SA_DIR_OUT = 2
+};
+
 enum {
 	XFRM_SHARE_ANY,		/* No limitations */
 	XFRM_SHARE_SESSION,	/* For this session only */
@@ -315,6 +320,7 @@ enum xfrm_attr_type_t {
 	XFRMA_SET_MARK_MASK,	/* __u32 */
 	XFRMA_IF_ID,		/* __u32 */
 	XFRMA_MTIMER_THRESH,	/* __u32 in seconds for input SA */
+	XFRMA_SA_DIR,		/* __u8 */
 	__XFRMA_MAX
 
 #define XFRMA_OUTPUT_MARK XFRMA_SET_MARK	/* Compatibility */
net/xfrm/xfrm_compat.c +5 -2
index 655fe4ff8621..703d4172c7d7 100644
@@ -98,6 +98,7 @@ static const int compat_msg_min[XFRM_NR_MSGTYPES] = {
 };
 
 static const struct nla_policy compat_policy[XFRMA_MAX+1] = {
+	[XFRMA_UNSPEC]          = { .strict_start_type = XFRMA_SA_DIR },
 	[XFRMA_SA]		= { .len = XMSGSIZE(compat_xfrm_usersa_info)},
 	[XFRMA_POLICY]		= { .len = XMSGSIZE(compat_xfrm_userpolicy_info)},
 	[XFRMA_LASTUSED]	= { .type = NLA_U64},
@@ -129,6 +130,7 @@ static const struct nla_policy compat_policy[XFRMA_MAX+1] = {
 	[XFRMA_SET_MARK_MASK]	= { .type = NLA_U32 },
 	[XFRMA_IF_ID]		= { .type = NLA_U32 },
 	[XFRMA_MTIMER_THRESH]	= { .type = NLA_U32 },
+	[XFRMA_SA_DIR]          = NLA_POLICY_RANGE(NLA_U8, XFRM_SA_DIR_IN, XFRM_SA_DIR_OUT),
 };
 
 static struct nlmsghdr *xfrm_nlmsg_put_compat(struct sk_buff *skb,
@@ -277,9 +279,10 @@ static int xfrm_xlate64_attr(struct sk_buff *dst, const struct nlattr *src)
 	case XFRMA_SET_MARK_MASK:
 	case XFRMA_IF_ID:
 	case XFRMA_MTIMER_THRESH:
+	case XFRMA_SA_DIR:
 		return xfrm_nla_cpy(dst, src, nla_len(src));
 	default:
-		BUILD_BUG_ON(XFRMA_MAX != XFRMA_MTIMER_THRESH);
+		BUILD_BUG_ON(XFRMA_MAX != XFRMA_SA_DIR);
 		pr_warn_once("unsupported nla_type %d\n", src->nla_type);
 		return -EOPNOTSUPP;
 	}
@@ -434,7 +437,7 @@ static int xfrm_xlate32_attr(void *dst, const struct nlattr *nla,
 	int err;
 
 	if (type > XFRMA_MAX) {
-		BUILD_BUG_ON(XFRMA_MAX != XFRMA_MTIMER_THRESH);
+		BUILD_BUG_ON(XFRMA_MAX != XFRMA_SA_DIR);
 		NL_SET_ERR_MSG(extack, "Bad attribute");
 		return -EOPNOTSUPP;
 	}
net/xfrm/xfrm_device.c +6 -0
index 6346690d5c69..2455a76a1cff 100644
@@ -253,6 +253,12 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x,
 		return -EINVAL;
 	}
 
+	if ((xuo->flags & XFRM_OFFLOAD_INBOUND && x->dir == XFRM_SA_DIR_OUT) ||
+	    (!(xuo->flags & XFRM_OFFLOAD_INBOUND) && x->dir == XFRM_SA_DIR_IN)) {
+		NL_SET_ERR_MSG(extack, "Mismatched SA and offload direction");
+		return -EINVAL;
+	}
+
 	is_packet_offload = xuo->flags & XFRM_OFFLOAD_PACKET;
 
 	/* We don't yet support UDP encapsulation and TFC padding. */
net/xfrm/xfrm_replay.c +2 -1
index ce56d659c55a..bc56c6305725 100644
@@ -778,7 +778,8 @@ int xfrm_init_replay(struct xfrm_state *x, struct netlink_ext_ack *extack)
 		}
 
 		if (x->props.flags & XFRM_STATE_ESN) {
-			if (replay_esn->replay_window == 0) {
+			if (replay_esn->replay_window == 0 &&
+			    (!x->dir || x->dir == XFRM_SA_DIR_IN)) {
 				NL_SET_ERR_MSG(extack, "ESN replay window must be > 0");
 				return -EINVAL;
 			}
net/xfrm/xfrm_state.c +5 -0
index 0c306473a79d..c8c5fc47c431 100644
@@ -1292,6 +1292,7 @@ xfrm_state_find(const xfrm_address_t *daddr, const xfrm_address_t *saddr,
 		if (km_query(x, tmpl, pol) == 0) {
 			spin_lock_bh(&net->xfrm.xfrm_state_lock);
 			x->km.state = XFRM_STATE_ACQ;
+			x->dir = XFRM_SA_DIR_OUT;
 			list_add(&x->km.all, &net->xfrm.state_all);
 			XFRM_STATE_INSERT(bydst, &x->bydst,
 					  net->xfrm.state_bydst + h,
@@ -1744,6 +1745,7 @@ static struct xfrm_state *xfrm_state_clone(struct xfrm_state *orig,
 	x->lastused = orig->lastused;
 	x->new_mapping = 0;
 	x->new_mapping_sport = 0;
+	x->dir = orig->dir;
 
 	return x;
 
@@ -1857,6 +1859,9 @@ int xfrm_state_update(struct xfrm_state *x)
 	if (!x1)
 		goto out;
 
+	if (x1->dir != x->dir)
+		goto out;
+
 	if (xfrm_state_kern(x1)) {
 		to_put = x1;
 		err = -EEXIST;
net/xfrm/xfrm_user.c +119 -6

216 lines not on this page · raw message

--
2.30.2

#2 lore raw

PATCH v11 2/4 xfrm: Add dir validation to "out" data path lookup

In reply to Antony Antony

Introduces validation for the x->dir attribute within the XFRM output data lookup path. If the configured direction does not match the expected direction, output, increment the XfrmOutDirError counter and drop the packet to ensure data integrity and correct flow handling.

grep -vw 0 /proc/net/xfrm_stat
XfrmOutPolError 1
XfrmOutStateDirError 1

v10 -> 11
- rename error s/XfrmOutDirError/XfrmOutStateDirError/
- fix possible dereferencing of x reported by Smatch

---
 include/uapi/linux/snmp.h | 1 +
 net/xfrm/xfrm_policy.c    | 6 ++++++
 net/xfrm/xfrm_proc.c      | 1 +
 3 files changed, 8 insertions(+)
include/uapi/linux/snmp.h +1 -0
index a0819c6a5988..23792b8412bd 100644
@@ -337,6 +337,7 @@ enum
 	LINUX_MIB_XFRMFWDHDRERROR,		/* XfrmFwdHdrError*/
 	LINUX_MIB_XFRMOUTSTATEINVALID,		/* XfrmOutStateInvalid */
 	LINUX_MIB_XFRMACQUIREERROR,		/* XfrmAcquireError */
+	LINUX_MIB_XFRMOUTSTATEDIRERROR,		/* XfrmOutStateDirError */
 	__LINUX_MIB_XFRMMAX
 };
 
net/xfrm/xfrm_policy.c +6 -0
index 6affe5cd85d8..298b3a9eb48d 100644
@@ -2489,6 +2489,12 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, const struct flowi *fl,
 
 		x = xfrm_state_find(remote, local, fl, tmpl, policy, &error,
 				    family, policy->if_id);
+		if (x && x->dir && x->dir != XFRM_SA_DIR_OUT) {
+			XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTSTATEDIRERROR);
+			xfrm_state_put(x);
+			error = -EINVAL;
+			goto fail;
+		}
 
 		if (x && x->km.state == XFRM_STATE_VALID) {
 			xfrm[nx++] = x;
net/xfrm/xfrm_proc.c +1 -0
index 5f9bf8e5c933..98606f1078f7 100644
@@ -41,6 +41,7 @@ static const struct snmp_mib xfrm_mib_list[] = {
 	SNMP_MIB_ITEM("XfrmFwdHdrError", LINUX_MIB_XFRMFWDHDRERROR),
 	SNMP_MIB_ITEM("XfrmOutStateInvalid", LINUX_MIB_XFRMOUTSTATEINVALID),
 	SNMP_MIB_ITEM("XfrmAcquireError", LINUX_MIB_XFRMACQUIREERROR),
+	SNMP_MIB_ITEM("XfrmOutStateDirError", LINUX_MIB_XFRMOUTSTATEDIRERROR),
 	SNMP_MIB_SENTINEL
 };
 

--
2.30.2

#5 lore raw

PATCH v11 3/4 xfrm: Add dir validation to "in" data path lookup

In reply to Antony Antony

Introduces validation for the x->dir attribute within the XFRM input data lookup path. If the configured direction does not match the expected direction, input, increment the XfrmInStateDirError counter and drop the packet to ensure data integrity and correct flow handling.

grep -vw 0 /proc/net/xfrm_stat
XfrmInStateDirError 1

v10->v11
- rename error s/XfrmInDirError/XfrmInStateDirError/

---
 include/uapi/linux/snmp.h |  1 +
 net/ipv6/xfrm6_input.c    |  7 +++++++
 net/xfrm/xfrm_input.c     | 11 +++++++++++
 net/xfrm/xfrm_proc.c      |  1 +
 4 files changed, 20 insertions(+)
include/uapi/linux/snmp.h +1 -0
index 23792b8412bd..adf5fd78dd50 100644
@@ -338,6 +338,7 @@ enum
 	LINUX_MIB_XFRMOUTSTATEINVALID,		/* XfrmOutStateInvalid */
 	LINUX_MIB_XFRMACQUIREERROR,		/* XfrmAcquireError */
 	LINUX_MIB_XFRMOUTSTATEDIRERROR,		/* XfrmOutStateDirError */
+	LINUX_MIB_XFRMINSTATEDIRERROR,		/* XfrmInStateDirError */
 	__LINUX_MIB_XFRMMAX
 };
 
net/ipv6/xfrm6_input.c +7 -0
index a17d783dc7c0..a3c50b00d6c1 100644
@@ -279,6 +279,13 @@ int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
 		if (!x)
 			continue;
 
+		if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) {
+			XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR);
+			xfrm_state_put(x);
+			x = NULL;
+			continue;
+		}
+
 		spin_lock(&x->lock);
 
 		if ((!i || (x->props.flags & XFRM_STATE_WILDRECV)) &&
net/xfrm/xfrm_input.c +11 -0
index 161f535c8b94..71b42de6e3c9 100644
@@ -466,6 +466,11 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	if (encap_type < 0 || (xo && xo->flags & XFRM_GRO)) {
 		x = xfrm_input_state(skb);
 
+		if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) {
+			XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR);
+			goto drop;
+		}
+
 		if (unlikely(x->km.state != XFRM_STATE_VALID)) {
 			if (x->km.state == XFRM_STATE_ACQ)
 				XFRM_INC_STATS(net, LINUX_MIB_XFRMACQUIREERROR);
@@ -571,6 +576,12 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 			goto drop;
 		}
 
+		if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) {
+			XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR);
+			xfrm_state_put(x);
+			goto drop;
+		}
+
 		skb->mark = xfrm_smark_get(skb->mark, x);
 
 		sp->xvec[sp->len++] = x;
net/xfrm/xfrm_proc.c +1 -0
index 98606f1078f7..eeb984be03a7 100644
@@ -42,6 +42,7 @@ static const struct snmp_mib xfrm_mib_list[] = {
 	SNMP_MIB_ITEM("XfrmOutStateInvalid", LINUX_MIB_XFRMOUTSTATEINVALID),
 	SNMP_MIB_ITEM("XfrmAcquireError", LINUX_MIB_XFRMACQUIREERROR),
 	SNMP_MIB_ITEM("XfrmOutStateDirError", LINUX_MIB_XFRMOUTSTATEDIRERROR),
+	SNMP_MIB_ITEM("XfrmInStateDirError", LINUX_MIB_XFRMINSTATEDIRERROR),
 	SNMP_MIB_SENTINEL
 };
 

--
2.30.2

#6 lore raw

PATCH v11 4/4 xfrm: Restrict SA direction attribute to specific netlink message types

In reply to Antony Antony

Reject the usage of the SA_DIR attribute in xfrm netlink messages when it's not applicable. This ensures that SA_DIR is only accepted for certain message types (NEWSA, UPDSA, and ALLOCSPI)

---
 net/xfrm/xfrm_user.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)
net/xfrm/xfrm_user.c +22 -0
index 54a0d750be4b..be977e96338b 100644
@@ -3200,6 +3200,24 @@ static const struct xfrm_link {
 	[XFRM_MSG_GETDEFAULT  - XFRM_MSG_BASE] = { .doit = xfrm_get_default   },
 };
 
+static int xfrm_reject_unused(int type, struct nlattr **attrs,
+			      struct netlink_ext_ack *extack)
+{
+	if (attrs[XFRMA_SA_DIR]) {
+		switch (type) {
+		case XFRM_MSG_NEWSA:
+		case XFRM_MSG_UPDSA:
+		case XFRM_MSG_ALLOCSPI:
+			break;
+		default:
+			NL_SET_ERR_MSG(extack, "Invalid attribute SA_DIR");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
 static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 			     struct netlink_ext_ack *extack)
 {
@@ -3259,6 +3277,10 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (err < 0)
 		goto err;
 
+	err = xfrm_reject_unused((type + XFRM_MSG_BASE), attrs, extack);
+	if (err < 0)
+		goto err;
+
 	if (link->doit == NULL) {
 		err = -EINVAL;
 		goto err;

--
2.30.2

#3 lore raw

Re: [PATCH ipsec-next v11 2/4] xfrm: Add dir validation to "out" data path lookup

In reply to Antony Antony

net/xfrm/xfrm_proc.c static const struct snmp_mib xfrm_mib_list[] = { · line 44

Quoted 9 lines · Antony Antony
net/xfrm/xfrm_proc.c +1 -0
index 5f9bf8e5c933..98606f1078f7 100644
@@ -41,6 +41,7 @@ static const struct snmp_mib xfrm_mib_list[] = {
 	SNMP_MIB_ITEM("XfrmFwdHdrError", LINUX_MIB_XFRMFWDHDRERROR),
 	SNMP_MIB_ITEM("XfrmOutStateInvalid", LINUX_MIB_XFRMOUTSTATEINVALID),
 	SNMP_MIB_ITEM("XfrmAcquireError", LINUX_MIB_XFRMACQUIREERROR),
+	SNMP_MIB_ITEM("XfrmOutStateDirError", LINUX_MIB_XFRMOUTSTATEDIRERROR),

This needs a corresponding entry in Documentation/networking/xfrm_proc.rst

Signature · 1 line
Sabrina
maintainer
#7 lore raw
In reply to Antony Antony
Quoted 7 lines · On Mon, 22 Apr 2024 00:23:58 +0200 Antony Antony wrote:

Hi,

Inspired by the upcoming IP-TFS patch set, and confusions experienced in the past due to lack of direction attribute on SAs, add a new direction "dir" attribute. It aims to streamline the SA configuration process and enhance the clarity of existing SA attributes.

Quoted 7 lines

This patch set introduces the 'dir' attribute to SA, aka xfrm_state, ('in' for input or 'out' for output). Alsp add validations of existing direction-specific SA attributes during configuration and in the data path lookup.

This change would not affect any existing use case or way of configuring SA. You will notice improvements when the new 'dir' attribute is set.

This breaks the xfrm_policy.sh selftests.

Signature · 1 line
pw-bot: au
#8 lore raw
Quoted 13 lines · Jakub Kicinski via Devel
Quoted 12 lines · On Mon, 22 Apr 2024 00:23:58 +0200 Antony Antony wrote:

Hi,

Inspired by the upcoming IP-TFS patch set, and confusions experienced in the past due to lack of direction attribute on SAs, add a new direction "dir" attribute. It aims to streamline the SA configuration process and enhance the clarity of existing SA attributes.

This patch set introduces the 'dir' attribute to SA, aka xfrm_state, ('in' for input or 'out' for output). Alsp add validations of existing direction-specific SA attributes during configuration and in the data path lookup.

Quoted 4 lines
Quoted 2 lines

This change would not affect any existing use case or way of configuring SA. You will notice improvements when the new 'dir' attribute is set.

This breaks the xfrm_policy.sh selftests.

I mised spd lookup use diffrent nla_policy array than xfrma_policy. I fixed it in v12.

Quoted 3 lines
Signature · 2 lines
pw-bot: au
-- 
#4 lore raw

Re: [devel-ipsec] [PATCH ipsec-next v11 2/4] xfrm: Add dir validation to "out" data path lookup

In reply to Sabrina Dubroca
Quoted 12 lines · Sabrina Dubroca via Devel

net/xfrm/xfrm_proc.c static const struct snmp_mib xfrm_mib_list[] = { · line 44

Quoted 9 lines · Antony Antony
net/xfrm/xfrm_proc.c +1 -0
index 5f9bf8e5c933..98606f1078f7 100644
@@ -41,6 +41,7 @@ static const struct snmp_mib xfrm_mib_list[] = {
 	SNMP_MIB_ITEM("XfrmFwdHdrError", LINUX_MIB_XFRMFWDHDRERROR),
 	SNMP_MIB_ITEM("XfrmOutStateInvalid", LINUX_MIB_XFRMOUTSTATEINVALID),
 	SNMP_MIB_ITEM("XfrmAcquireError", LINUX_MIB_XFRMACQUIREERROR),
+	SNMP_MIB_ITEM("XfrmOutStateDirError", LINUX_MIB_XFRMOUTSTATEDIRERROR),

This needs a corresponding entry in Documentation/networking/xfrm_proc.rst

fixed. I added both in and out error to it; v12.

-antony