[PATCH iproute2-next v3 0/2] f_flower: match on the number of vlan tags

[PATCH iproute2-next v3 0/2] f_flower: match on the number of vlan tags

[Boris Sukholitko]

[-- Attachment #1: Type: text/plain, Size: 1654 bytes --]

Hi,

Our customers in the fiber telecom world have network configurations
where they would like to control their traffic according to the number
of tags appearing in the packet.

For example, TR247 GPON conformance test suite specification mostly
talks about untagged, single, double tagged packets and gives lax
guidelines on the vlan protocol vs. number of vlan tags.

This is different from the common IT networks where 802.1Q and 802.1ad
protocols are usually describe single and double tagged packet. GPON
configurations that we work with have arbitrary mix the above protocols
and number of vlan tags in the packet.

The following patch series implement number of vlans flower filter. They
add num_of_vlans flower filter as an alternative to vlan ethtype protocol
matching. The end result is that the following command becomes possible:

tc filter add dev eth1 ingress flower \
  num_of_vlans 1 vlan_prio 5 action drop

Also, from our logs, we have redirect rules such that:

tc filter add dev $GPON ingress flower num_of_vlans $N \
     action mirred egress redirect dev $DEV

where N can range from 0 to 3 and $DEV is the function of $N.

Also there are rules setting skb mark based on the number of vlans:

tc filter add dev $GPON ingress flower num_of_vlans $N vlan_prio \
    $P action skbedit mark $M

Thanks,
Boris.

- v3: rebased to the latest iproute2-next
- v2: add missing f_flower subject prefix

Boris Sukholitko (2):
  f_flower: Add num of vlans parameter
  f_flower: Check args with num_of_vlans

 tc/f_flower.c | 57 ++++++++++++++++++++++++++++++++++++---------------
 1 file changed, 41 insertions(+), 16 deletions(-)

-- 
2.29.2


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4221 bytes --]

[PATCH iproute2-next v3 1/2] f_flower: Add num of vlans parameter

[Boris Sukholitko]

[-- Attachment #1: Type: text/plain, Size: 2385 bytes --]

Our customers in the fiber telecom world have network configurations
where they would like to control their traffic according to the number
of tags appearing in the packet.

For example, TR247 GPON conformance test suite specification mostly
talks about untagged, single, double tagged packets and gives lax
guidelines on the vlan protocol vs. number of vlan tags.

This is different from the common IT networks where 802.1Q and 802.1ad
protocols are usually describe single and double tagged packet. GPON
configurations that we work with have arbitrary mix the above protocols
and number of vlan tags in the packet.

This patch adds num_of_vlans flower key and associated print and parse
routines. The following command becomes possible:

tc filter add dev eth1 ingress flower num_of_vlans 1 action drop

Signed-off-by: Boris Sukholitko <boris.sukholitko@broadcom.com>
---
 tc/f_flower.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/tc/f_flower.c b/tc/f_flower.c
index 686cf121..25ffd295 100644
--- a/tc/f_flower.c
+++ b/tc/f_flower.c
@@ -48,6 +48,7 @@ static void explain(void)
 		"\n"
 		"Where: MATCH-LIST := [ MATCH-LIST ] MATCH\n"
 		"       MATCH      := {	indev DEV-NAME |\n"
+		"			num_of_vlans VLANS_COUNT |\n"
 		"			vlan_id VID |\n"
 		"			vlan_prio PRIORITY |\n"
 		"			vlan_ethtype [ ipv4 | ipv6 | ETH-TYPE ] |\n"
@@ -1525,6 +1526,17 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 			if (check_ifname(*argv))
 				invarg("\"indev\" not a valid ifname", *argv);
 			addattrstrz(n, MAX_MSG, TCA_FLOWER_INDEV, *argv);
+		} else if (matches(*argv, "num_of_vlans") == 0) {
+			__u8 num_of_vlans;
+
+			NEXT_ARG();
+			ret = get_u8(&num_of_vlans, *argv, 10);
+			if (ret < 0) {
+				fprintf(stderr, "Illegal \"num_of_vlans\"\n");
+				return -1;
+			}
+			addattr8(n, MAX_MSG,
+				 TCA_FLOWER_KEY_NUM_OF_VLANS, num_of_vlans);
 		} else if (matches(*argv, "vlan_id") == 0) {
 			__u16 vid;
 
@@ -2694,6 +2706,14 @@ static int flower_print_opt(struct filter_util *qu, FILE *f,
 
 	open_json_object("keys");
 
+	if (tb[TCA_FLOWER_KEY_NUM_OF_VLANS]) {
+		struct rtattr *attr = tb[TCA_FLOWER_KEY_NUM_OF_VLANS];
+
+		print_nl();
+		print_uint(PRINT_ANY, "num_of_vlans", "  num_of_vlans %d",
+			   rta_getattr_u8(attr));
+	}
+
 	if (tb[TCA_FLOWER_KEY_VLAN_ID]) {
 		struct rtattr *attr = tb[TCA_FLOWER_KEY_VLAN_ID];
 
-- 
2.29.2


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4221 bytes --]

[PATCH iproute2-next v3 2/2] f_flower: Check args with num_of_vlans

[Boris Sukholitko]

[-- Attachment #1: Type: text/plain, Size: 4870 bytes --]

Having more than one vlan allows matching on the vlan tag parameters.
This patch changes vlan key validation to take number of vlan tags into
account.

Signed-off-by: Boris Sukholitko <boris.sukholitko@broadcom.com>
---
 tc/f_flower.c | 41 +++++++++++++++++++++++------------------
 1 file changed, 23 insertions(+), 18 deletions(-)

diff --git a/tc/f_flower.c b/tc/f_flower.c
index 25ffd295..fbb7042f 100644
--- a/tc/f_flower.c
+++ b/tc/f_flower.c
@@ -160,21 +160,23 @@ err:
 	return err;
 }
 
-static bool eth_type_vlan(__be16 ethertype)
+static bool eth_type_vlan(__be16 ethertype, bool good_num_of_vlans)
 {
 	return ethertype == htons(ETH_P_8021Q) ||
-	       ethertype == htons(ETH_P_8021AD);
+	       ethertype == htons(ETH_P_8021AD) ||
+	       good_num_of_vlans;
 }
 
 static int flower_parse_vlan_eth_type(char *str, __be16 eth_type, int type,
 				      __be16 *p_vlan_eth_type,
-				      struct nlmsghdr *n)
+				      struct nlmsghdr *n, bool good_num_of_vlans)
 {
 	__be16 vlan_eth_type;
 
-	if (!eth_type_vlan(eth_type)) {
-		fprintf(stderr, "Can't set \"%s\" if ethertype isn't 802.1Q or 802.1AD\n",
-			type == TCA_FLOWER_KEY_VLAN_ETH_TYPE ? "vlan_ethtype" : "cvlan_ethtype");
+	if (!eth_type_vlan(eth_type, good_num_of_vlans)) {
+		fprintf(stderr, "Can't set \"%s\" if ethertype isn't 802.1Q or 802.1AD and num_of_vlans %s\n",
+			type == TCA_FLOWER_KEY_VLAN_ETH_TYPE ? "vlan_ethtype" : "cvlan_ethtype",
+			type == TCA_FLOWER_KEY_VLAN_ETH_TYPE ? "is 0" : "less than 2");
 		return -1;
 	}
 
@@ -1425,6 +1427,7 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 	__be16 tc_proto = TC_H_MIN(t->tcm_info);
 	__be16 eth_type = tc_proto;
 	__be16 vlan_ethtype = 0;
+	__u8 num_of_vlans = 0;
 	__u8 ip_proto = 0xff;
 	__u32 flags = 0;
 	__u32 mtf = 0;
@@ -1527,8 +1530,6 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 				invarg("\"indev\" not a valid ifname", *argv);
 			addattrstrz(n, MAX_MSG, TCA_FLOWER_INDEV, *argv);
 		} else if (matches(*argv, "num_of_vlans") == 0) {
-			__u8 num_of_vlans;
-
 			NEXT_ARG();
 			ret = get_u8(&num_of_vlans, *argv, 10);
 			if (ret < 0) {
@@ -1541,8 +1542,9 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 			__u16 vid;
 
 			NEXT_ARG();
-			if (!eth_type_vlan(tc_proto)) {
-				fprintf(stderr, "Can't set \"vlan_id\" if ethertype isn't 802.1Q or 802.1AD\n");
+			if (!eth_type_vlan(tc_proto, num_of_vlans > 0)) {
+				fprintf(stderr, "Can't set \"vlan_id\" if ethertype isn't 802.1Q or 802.1AD"
+						" and num_of_vlans is 0\n");
 				return -1;
 			}
 			ret = get_u16(&vid, *argv, 10);
@@ -1555,8 +1557,9 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 			__u8 vlan_prio;
 
 			NEXT_ARG();
-			if (!eth_type_vlan(tc_proto)) {
-				fprintf(stderr, "Can't set \"vlan_prio\" if ethertype isn't 802.1Q or 802.1AD\n");
+			if (!eth_type_vlan(tc_proto, num_of_vlans > 0)) {
+				fprintf(stderr, "Can't set \"vlan_prio\" if ethertype isn't 802.1Q or 802.1AD"
+						" and num_of_vlans is 0\n");
 				return -1;
 			}
 			ret = get_u8(&vlan_prio, *argv, 10);
@@ -1570,7 +1573,7 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 			NEXT_ARG();
 			ret = flower_parse_vlan_eth_type(*argv, eth_type,
 						 TCA_FLOWER_KEY_VLAN_ETH_TYPE,
-						 &vlan_ethtype, n);
+						 &vlan_ethtype, n, num_of_vlans > 0);
 			if (ret < 0)
 				return -1;
 			/* get new ethtype for later parsing  */
@@ -1579,8 +1582,9 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 			__u16 vid;
 
 			NEXT_ARG();
-			if (!eth_type_vlan(vlan_ethtype)) {
-				fprintf(stderr, "Can't set \"cvlan_id\" if inner vlan ethertype isn't 802.1Q or 802.1AD\n");
+			if (!eth_type_vlan(vlan_ethtype, num_of_vlans > 1)) {
+				fprintf(stderr, "Can't set \"cvlan_id\" if inner vlan ethertype isn't 802.1Q or 802.1AD"
+						" and num_of_vlans is less than 2\n");
 				return -1;
 			}
 			ret = get_u16(&vid, *argv, 10);
@@ -1593,8 +1597,9 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 			__u8 cvlan_prio;
 
 			NEXT_ARG();
-			if (!eth_type_vlan(vlan_ethtype)) {
-				fprintf(stderr, "Can't set \"cvlan_prio\" if inner vlan ethertype isn't 802.1Q or 802.1AD\n");
+			if (!eth_type_vlan(vlan_ethtype, num_of_vlans > 1)) {
+				fprintf(stderr, "Can't set \"cvlan_prio\" if inner vlan ethertype isn't 802.1Q or 802.1AD"
+						" and num_of_vlans is less than 2\n");
 				return -1;
 			}
 			ret = get_u8(&cvlan_prio, *argv, 10);
@@ -1609,7 +1614,7 @@ static int flower_parse_opt(struct filter_util *qu, char *handle,
 			/* get new ethtype for later parsing */
 			ret = flower_parse_vlan_eth_type(*argv, vlan_ethtype,
 						 TCA_FLOWER_KEY_CVLAN_ETH_TYPE,
-						 &eth_type, n);
+						 &eth_type, n, num_of_vlans > 1);
 			if (ret < 0)
 				return -1;
 		} else if (matches(*argv, "mpls") == 0) {
-- 
2.29.2


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4221 bytes --]

Re: [PATCH iproute2-next v3 0/2] f_flower: match on the number of vlan tags

[Stephen Hemminger]

On Tue, 26 Apr 2022 12:14:15 +0300
Boris Sukholitko <boris.sukholitko@broadcom.com> wrote:

> Hi,
> 
> Our customers in the fiber telecom world have network configurations
> where they would like to control their traffic according to the number
> of tags appearing in the packet.
> 
> For example, TR247 GPON conformance test suite specification mostly
> talks about untagged, single, double tagged packets and gives lax
> guidelines on the vlan protocol vs. number of vlan tags.
> 
> This is different from the common IT networks where 802.1Q and 802.1ad
> protocols are usually describe single and double tagged packet. GPON
> configurations that we work with have arbitrary mix the above protocols
> and number of vlan tags in the packet.
> 
> The following patch series implement number of vlans flower filter. They
> add num_of_vlans flower filter as an alternative to vlan ethtype protocol
> matching. The end result is that the following command becomes possible:
> 
> tc filter add dev eth1 ingress flower \
>   num_of_vlans 1 vlan_prio 5 action drop
> 
> Also, from our logs, we have redirect rules such that:
> 
> tc filter add dev $GPON ingress flower num_of_vlans $N \
>      action mirred egress redirect dev $DEV
> 
> where N can range from 0 to 3 and $DEV is the function of $N.
> 
> Also there are rules setting skb mark based on the number of vlans:
> 
> tc filter add dev $GPON ingress flower num_of_vlans $N vlan_prio \
>     $P action skbedit mark $M
> 
> Thanks,
> Boris.
> 
> - v3: rebased to the latest iproute2-next
> - v2: add missing f_flower subject prefix
> 
> Boris Sukholitko (2):
>   f_flower: Add num of vlans parameter
>   f_flower: Check args with num_of_vlans
> 
>  tc/f_flower.c | 57 ++++++++++++++++++++++++++++++++++++---------------
>  1 file changed, 41 insertions(+), 16 deletions(-)

Can you do this with BPF? instead of kernel change?

Re: [PATCH iproute2-next v3 0/2] f_flower: match on the number of vlan tags

[Boris Sukholitko]

[-- Attachment #1: Type: text/plain, Size: 2325 bytes --]

Hi Stephen,

On Tue, Apr 26, 2022 at 08:11:42AM -0700, Stephen Hemminger wrote:
> On Tue, 26 Apr 2022 12:14:15 +0300
> Boris Sukholitko <boris.sukholitko@broadcom.com> wrote:
> 
> > Hi,
> > 
> > Our customers in the fiber telecom world have network configurations
> > where they would like to control their traffic according to the number
> > of tags appearing in the packet.
> > 
> > For example, TR247 GPON conformance test suite specification mostly
> > talks about untagged, single, double tagged packets and gives lax
> > guidelines on the vlan protocol vs. number of vlan tags.
> > 
> > This is different from the common IT networks where 802.1Q and 802.1ad
> > protocols are usually describe single and double tagged packet. GPON
> > configurations that we work with have arbitrary mix the above protocols
> > and number of vlan tags in the packet.
> > 
> > The following patch series implement number of vlans flower filter. They
> > add num_of_vlans flower filter as an alternative to vlan ethtype protocol
> > matching. The end result is that the following command becomes possible:
> > 
> > tc filter add dev eth1 ingress flower \
> >   num_of_vlans 1 vlan_prio 5 action drop
> > 
> > Also, from our logs, we have redirect rules such that:
> > 
> > tc filter add dev $GPON ingress flower num_of_vlans $N \
> >      action mirred egress redirect dev $DEV
> > 
> > where N can range from 0 to 3 and $DEV is the function of $N.
> > 
> > Also there are rules setting skb mark based on the number of vlans:
> > 
> > tc filter add dev $GPON ingress flower num_of_vlans $N vlan_prio \
> >     $P action skbedit mark $M
> > 
> > Thanks,
> > Boris.
> > 
> > - v3: rebased to the latest iproute2-next
> > - v2: add missing f_flower subject prefix
> > 
> > Boris Sukholitko (2):
> >   f_flower: Add num of vlans parameter
> >   f_flower: Check args with num_of_vlans
> > 
> >  tc/f_flower.c | 57 ++++++++++++++++++++++++++++++++++++---------------
> >  1 file changed, 41 insertions(+), 16 deletions(-)
> 
> Can you do this with BPF? instead of kernel change?

You may have missed my reply to this question at:

https://lore.kernel.org/netdev/20220412104514.GB27480@noodle/

There is also Jamal's reply further at the thread:

https://lore.kernel.org/netdev/b2c83f63-a2e9-92a2-f262-3aae3491dfc3@mojatatu.com/

Thanks,
Boris.

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4221 bytes --]

Re: [PATCH iproute2-next v3 0/2] f_flower: match on the number of vlan tags

[Stephen Hemminger]

On Wed, 27 Apr 2022 17:32:00 +0300
Boris Sukholitko <boris.sukholitko@broadcom.com> wrote:

> Hi Stephen,
> 
> On Tue, Apr 26, 2022 at 08:11:42AM -0700, Stephen Hemminger wrote:
> > On Tue, 26 Apr 2022 12:14:15 +0300
> > Boris Sukholitko <boris.sukholitko@broadcom.com> wrote:
> >   
> > > Hi,
> > > 
> > > Our customers in the fiber telecom world have network configurations
> > > where they would like to control their traffic according to the number
> > > of tags appearing in the packet.
> > > 
> > > For example, TR247 GPON conformance test suite specification mostly
> > > talks about untagged, single, double tagged packets and gives lax
> > > guidelines on the vlan protocol vs. number of vlan tags.
> > > 
> > > This is different from the common IT networks where 802.1Q and 802.1ad
> > > protocols are usually describe single and double tagged packet. GPON
> > > configurations that we work with have arbitrary mix the above protocols
> > > and number of vlan tags in the packet.
> > > 
> > > The following patch series implement number of vlans flower filter. They
> > > add num_of_vlans flower filter as an alternative to vlan ethtype protocol
> > > matching. The end result is that the following command becomes possible:
> > > 
> > > tc filter add dev eth1 ingress flower \
> > >   num_of_vlans 1 vlan_prio 5 action drop
> > > 
> > > Also, from our logs, we have redirect rules such that:
> > > 
> > > tc filter add dev $GPON ingress flower num_of_vlans $N \
> > >      action mirred egress redirect dev $DEV
> > > 
> > > where N can range from 0 to 3 and $DEV is the function of $N.
> > > 
> > > Also there are rules setting skb mark based on the number of vlans:
> > > 
> > > tc filter add dev $GPON ingress flower num_of_vlans $N vlan_prio \
> > >     $P action skbedit mark $M
> > > 
> > > Thanks,
> > > Boris.
> > > 
> > > - v3: rebased to the latest iproute2-next
> > > - v2: add missing f_flower subject prefix
> > > 
> > > Boris Sukholitko (2):
> > >   f_flower: Add num of vlans parameter
> > >   f_flower: Check args with num_of_vlans
> > > 
> > >  tc/f_flower.c | 57 ++++++++++++++++++++++++++++++++++++---------------
> > >  1 file changed, 41 insertions(+), 16 deletions(-)  
> > 
> > Can you do this with BPF? instead of kernel change?  
> 
> You may have missed my reply to this question at:
> 
> https://lore.kernel.org/netdev/20220412104514.GB27480@noodle/
> 
> There is also Jamal's reply further at the thread:
> 
> https://lore.kernel.org/netdev/b2c83f63-a2e9-92a2-f262-3aae3491dfc3@mojatatu.com/
> 
> Thanks,
> Boris.

Thanks, there is a tradeoff here, if you add more logic to the kernel, it
impacts every user and creates long term technical debt. Your use case seemed
quite specific to single use case.

But also, it is an example of something where kernel already has the state
information and it might be hard to get with BPF.

Surprised that the people who do BPF at scale did not chime in on this discussion.

Re: [PATCH iproute2-next v3 0/2] f_flower: match on the number of vlan tags

[David Ahern]

On 4/26/22 3:14 AM, Boris Sukholitko wrote:
>  tc/f_flower.c | 57 ++++++++++++++++++++++++++++++++++++---------------
>  1 file changed, 41 insertions(+), 16 deletions(-)
> 

man page update is missing and I noticed after I pushed. Please send a
patch for the man page.

Re: [PATCH iproute2-next v3 0/2] f_flower: match on the number of vlan tags

[Boris Sukholitko]

[-- Attachment #1: Type: text/plain, Size: 407 bytes --]

On Wed, Apr 27, 2022 at 08:20:28PM -0600, David Ahern wrote:
> On 4/26/22 3:14 AM, Boris Sukholitko wrote:
> >  tc/f_flower.c | 57 ++++++++++++++++++++++++++++++++++++---------------
> >  1 file changed, 41 insertions(+), 16 deletions(-)
> > 
> 
> man page update is missing and I noticed after I pushed. Please send a
> patch for the man page.

Sorry for the mess. The patch has been sent.

Thanks,
Boris.

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4221 bytes --]