Skip to contentSkip to conversation

ip: support for xfrm interfaces

PATCH Superseded Matt Ellison 2 participants 4 messages Jan 17, 2019 – Jan 21, 2019

Subsystems (2)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 ip: support for xfrm interfaces

Interfaces take a 'if_id' which is an interface id which can be set on an xfrm policy as its interface lookup key (XFRMA_IF_ID).

---
 ip/Makefile                             |  2 +-
 ip/iplink.c                             |  3 +-
 ip/link_xfrm.c                          | 79 +++++++++++++++++++++++++
 man/man8/ip-link.8.in                   | 27 ++++++++-
 testsuite/tests/ip/link/add_type_xfrm.t | 32 ++++++++++
 5 files changed, 140 insertions(+), 3 deletions(-)
 create mode 100644 ip/link_xfrm.c
 create mode 100755 testsuite/tests/ip/link/add_type_xfrm.t
ip/Makefile +1 -1
index a88f9366..7ce6e91a 100644
@@ -5,7 +5,7 @@ IPOBJ=ip.o ipaddress.o ipaddrlabel.o iproute.o iprule.o ipnetns.o \
     ipxfrm.o xfrm_state.o xfrm_policy.o xfrm_monitor.o iplink_dummy.o \
     iplink_ifb.o iplink_nlmon.o iplink_team.o iplink_vcan.o iplink_vxcan.o \
     iplink_vlan.o link_veth.o link_gre.o iplink_can.o iplink_xdp.o \
-    iplink_macvlan.o ipl2tp.o link_vti.o link_vti6.o \
+    iplink_macvlan.o ipl2tp.o link_vti.o link_vti6.o link_xfrm.o \
     iplink_vxlan.o tcp_metrics.o iplink_ipoib.o ipnetconf.o link_ip6tnl.o \
     link_iptnl.o link_gre6.o iplink_bond.o iplink_bond_slave.o iplink_hsr.o \
     iplink_bridge.o iplink_bridge_slave.o ipfou.o iplink_ipvlan.o \
ip/iplink.c +2 -1
index b5519201..f61e570a 100644
@@ -121,7 +121,8 @@ void iplink_usage(void)
 			"          bridge | bond | team | ipoib | ip6tnl | ipip | sit | vxlan |\n"
 			"          gre | gretap | erspan | ip6gre | ip6gretap | ip6erspan |\n"
 			"          vti | nlmon | team_slave | bond_slave | bridge_slave |\n"
-			"          ipvlan | ipvtap | geneve | vrf | macsec | netdevsim | rmnet }\n");
+			"          ipvlan | ipvtap | geneve | vrf | macsec | netdevsim | rmnet |\n"
+			"          xfrm }\n");
 	}
 	exit(-1);
 }
ip/link_xfrm.c new file +79 -0

79 lines not on this page · raw message

man/man8/ip-link.8.in +26 -1
index 73d37c19..53504657 100644
@@ -221,7 +221,8 @@ ip-link \- network device configuration
 .BR vrf " |"
 .BR macsec " |"
 .BR netdevsim " |"
-.BR rmnet " ]"
+.BR rmnet " |"
+.BR xfrm " ]"
 
 .ti -8
 .IR ETYPE " := [ " TYPE " |"
@@ -350,6 +351,9 @@ Link types:
 .sp
 .BR rmnet
 - Qualcomm rmnet device
+.sp
+.BR xfrm
+- Virtual xfrm interface
 .in -8
 
 .TP
@@ -1704,6 +1708,27 @@ the following additional arguments are supported:
 
 .in -8
 
+.TP
+XFRM Type Support
+For a link of type
+.I XFRM
+the following additional arguments are supported:
+
+.BI "ip link add " DEVICE " type xfrm dev " PHYS_DEV " [ if_id " IF_ID " ]"
+
+.in +8
+.sp
+.BI dev " PHYS_DEV "
+- specifies the underlying physical interface from which transform traffic is sent and received.
+
+.sp
+.BI if_id " IF-ID "
+- specifies the hexadecimal lookup key used to send traffic to and from specific xfrm
+policies. Policies must be configured with the same key. If not set, the key defaults to
+0 and will match any policies which similarly do not have a lookup key configuration.
+
+.in -8
+
 .SS ip link delete - delete virtual link
 
 .TP
testsuite/tests/ip/link/add_type_xfrm.t new file +32 -0
new file mode 100755
index 00000000..78ce28e0
@@ -0,0 +1,32 @@
+#!/bin/sh
+
+. lib/generic.sh
+
+ts_log "[Testing Add XFRM Interface, With IF-ID]"
+
+PHYS_DEV="lo"
+NEW_DEV="$(rand_dev)"
+IF_ID="0xf"
+
+ts_ip "$0" "Add $NEW_DEV xfrm interface"    link add dev $NEW_DEV type xfrm dev $PHYS_DEV if_id $IF_ID
+
+ts_ip "$0" "Show $NEW_DEV xfrm interface"   -d link show dev $NEW_DEV
+test_on "$NEW_DEV"
+test_on "if_id $IF_ID"
+
+ts_ip "$0" "Del $NEW_DEV xfrm interface"   link del dev $NEW_DEV
+
+
+ts_log "[Testing Add XFRM Interface, No IF-ID]"
+
+PHYS_DEV="lo"
+NEW_DEV="$(rand_dev)"
+IF_ID="0xf"
+
+ts_ip "$0" "Add $NEW_DEV xfrm interface"    link add dev $NEW_DEV type xfrm dev $PHYS_DEV
+
+ts_ip "$0" "Show $NEW_DEV xfrm interface"   -d link show dev $NEW_DEV
+test_on "$NEW_DEV"
+test_on_not "if_id $IF_ID"
+
+ts_ip "$0" "Del $NEW_DEV xfrm interface"   link del dev $NEW_DEV
Signature · 20 lines
2.20.1


-- 



Please be advised that this email may contain confidential information. 
If you are not the intended recipient, please notify us by email by 
replying to the sender and delete this message. The sender disclaims that 
the content of this email constitutes an offer to enter into, or the 
acceptance of, any agreement; provided that the foregoing does not 
invalidate the binding effect of any digital or other electronic 
reproduction of a manual signature that is included in any attachment.


 
<https://twitter.com/arroyo_networks>   
<https://www.linkedin.com/company/arroyo-networks>   
<https://www.github.com/ArroyoNetworks>
#1 lore raw
Quoted 14 lines · Matt Ellison
ip/link_xfrm.c new file +14 -0
+static int xfrm_parse_opt(struct link_util *lu, int argc, char **argv,
+			  struct nlmsghdr *n)
+{
+	unsigned int link = 0;
+	__u32 if_id = 0;
+
+	while (argc > 0) {
+		if (!matches(*argv, "dev")) {
+			NEXT_ARG();
+			link = ll_name_to_index(*argv);
+			if (!link)
+				exit(nodev(*argv));
+		} else if (!matches(*argv, "if_id")) {
+			NEXT_ARG();

ip/link_xfrm.c line 46

Quoted 10 lines
ip/link_xfrm.c new file +10 -0
+			if (get_u32(&if_id, *argv, 0))
+				invarg("if_id", *argv);
+		} else {
+			xfrm_print_help(lu, argc, argv, stderr);
+			return -1;
+		}
+		argc--; argv++;
+	}
+
+	addattr32(n, 1024, IFLA_XFRM_IF_ID, if_id);

You always add IF_ID even if not set by user. The kernel code does not appear to require it so why pass a default value?

ip/link_xfrm.c line 53

Quoted 7 lines
ip/link_xfrm.c new file +7 -0
+
+	if (link) {
+		addattr32(n, 1024, IFLA_XFRM_LINK, link);
+	} else {
+		fprintf(stderr, "must specify physical device\n");
+		return -1;
+	}

The kernel code does appear to require this parameter, so why have this requirement in iproute2?

Quoted 6 lines
ip/link_xfrm.c new file +6 -0
+
+	return 0;
+}
+
+static void xfrm_print_opt(struct link_util *lu, FILE *f, struct rtattr *tb[])
+{

ip/link_xfrm.c line 69

Quoted 10 lines
ip/link_xfrm.c new file +10 -0
+
+	if (!tb)
+		return;
+
+	if (tb[IFLA_XFRM_IF_ID]) {
+		__u32 id = rta_getattr_u32(tb[IFLA_XFRM_IF_ID]);
+
+		print_0xhex(PRINT_ANY, "if_id", "if_id %#llx ", id);
+
+	}

What about IFLA_XFRM_LINK?

Quoted 29 lines
testsuite/tests/ip/link/add_type_xfrm.t new file +22 -0
new file mode 100755
index 00000000..78ce28e0
@@ -0,0 +1,32 @@
+#!/bin/sh
+
+. lib/generic.sh
+
+ts_log "[Testing Add XFRM Interface, With IF-ID]"
+
+PHYS_DEV="lo"
+NEW_DEV="$(rand_dev)"
+IF_ID="0xf"
+
+ts_ip "$0" "Add $NEW_DEV xfrm interface"    link add dev $NEW_DEV type xfrm dev $PHYS_DEV if_id $IF_ID
+
+ts_ip "$0" "Show $NEW_DEV xfrm interface"   -d link show dev $NEW_DEV
+test_on "$NEW_DEV"
+test_on "if_id $IF_ID"
+
+ts_ip "$0" "Del $NEW_DEV xfrm interface"   link del dev $NEW_DEV
+
+
+ts_log "[Testing Add XFRM Interface, No IF-ID]"
+
+PHYS_DEV="lo"

testsuite/tests/ip/link/add_type_xfrm.t line 32

Quoted 10 lines
testsuite/tests/ip/link/add_type_xfrm.t new file +10 -0
+NEW_DEV="$(rand_dev)"
+IF_ID="0xf"
+
+ts_ip "$0" "Add $NEW_DEV xfrm interface"    link add dev $NEW_DEV type xfrm dev $PHYS_DEV
+
+ts_ip "$0" "Show $NEW_DEV xfrm interface"   -d link show dev $NEW_DEV
+test_on "$NEW_DEV"
+test_on_not "if_id $IF_ID"
+
+ts_ip "$0" "Del $NEW_DEV xfrm interface"   link del dev $NEW_DEV

Thanks for adding test cases!

#2 lore raw
Quoted 2 lines · On Mon, 21 Jan 2019 09:14:52 -0700 David Ahern <[email protected]> wrote:

You always add IF_ID even if not set by user. The kernel code does not appear to require it so why pass a default value?

0 (the default) is a valid IF_ID, so setting an interface with a non-zero IF_ID back to 0 is possible. I think the better solution would be to check for existing values so an "ip link set" doesn't try and automatically use 0 if unsepcified.

Quoted 2 lines

The kernel code does appear to require this parameter, so why have this requirement in iproute2?

Looks to be required, without the check I get:
RTNETLINK answers: No such device

Quoted 1 line

What about IFLA_XFRM_LINK?

It shows up as the parent interface when you ip link show:

13: xfrm2@wlan0: <NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000

    link/none 48:e2:44:f6:77:51 brd ff:ff:ff:ff:ff:ff

But I can print it again if you think I should.

Signature · 13 lines
Please be advised that this email may contain confidential information. 
If you are not the intended recipient, please notify us by email by 
replying to the sender and delete this message. The sender disclaims that 
the content of this email constitutes an offer to enter into, or the 
acceptance of, any agreement; provided that the foregoing does not 
invalidate the binding effect of any digital or other electronic 
reproduction of a manual signature that is included in any attachment.


 
<https://twitter.com/arroyo_networks>   
<https://www.linkedin.com/company/arroyo-networks>   
<https://www.github.com/ArroyoNetworks>
#3 lore raw
Quoted 8 lines · Matt Ellison
Quoted 2 lines · On Mon, 21 Jan 2019 09:14:52 -0700 David Ahern <[email protected]> wrote:

You always add IF_ID even if not set by user. The kernel code does not appear to require it so why pass a default value?

0 (the default) is a valid IF_ID, so setting an interface with a non-zero IF_ID back to 0 is possible. I think the better solution would be to check for existing values so an "ip link set" doesn't try and automatically use 0 if unsepcified.

I would expect IF_ID to only be added to the request if passed on the command line. That should handle the set link case.

Quoted 7 lines
Quoted 3 lines

The kernel code does appear to require this parameter, so why have this requirement in iproute2?

Looks to be required, without the check I get:
RTNETLINK answers: No such device

Quoted 9 lines
Quoted 1 line

What about IFLA_XFRM_LINK?

It shows up as the parent interface when you ip link show:

13: xfrm2@wlan0: <NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000

    link/none 48:e2:44:f6:77:51 brd ff:ff:ff:ff:ff:ff

But I can print it again if you think I should.

ok, so IFLA_XFRM_LINK duplicates IFLA_LINK. no need to print twice.