Merge branch 'act_pedit-relative-offset'

Amir Vadai says:

====================
net/sched: act_pedit: Use offset relative to conventional network headers

Some FW/HW parser APIs are such that they need to get the specific header type (e.g
IPV4 or IPV6, TCP or UDP) and not only the networking level (e.g network or transport).

Enhancing the UAPI to allow for specifying that, would allow the same flows to be
set into both SW and HW.

This patchset also makes pedit more robust. Currently fields offset is specified
by offset relative to the ip header, while using negative offsets for
MAC layer fields.

This series enables the user to set offset relative to the relevant header.

Usage example:
$ tc filter add dev enp0s9 protocol ip parent ffff: \
   flower \
     ip_proto tcp \
    dst_port 80 \
   action \
       pedit munge ip ttl add 0xff \
       pedit munge tcp dport set 8080 \
     pipe action mirred egress redirect dev veth0

Will forward traffic destined to tcp dport 80, while modifying the
destination port to 8080, and decreasing the ttl by one.

I've uploaded a draft for the userspace [2] to make it easier to review and
test the patchset.

[1] - http://patchwork.ozlabs.org/patch/700909/
[2] - git: https://bitbucket.org/av42/iproute2.git
      branch: pedit

Patchset was tested and applied on top of upstream commit bd092ad ("Merge
branch 'remove-__napi_complete_done'")

Thanks,
Amir

Changes since V2:
- Instead of reusing unused bits in existing uapi fields, using new netlink
	attributes for the new information. This way new/old user space and new/old
	kernel can live together without having misunderstandings.

Changes since V1:
- No changes - V1 was sent and didn't make it for 4.10.
- You asked me [1] why did I use specific header names instead of layers (L2,
	L3...), and I explained that it is on purpose, this extra information is
	planned to be used by hardware drivers to offload the action.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/skbuff.h

@@ -2184,6 +2184,11 @@ static inline unsigned char *skb_mac_header(const struct sk_buff *skb)
 	return skb->head + skb->mac_header;
 }
 
+static inline int skb_mac_offset(const struct sk_buff *skb)
+{
+	return skb_mac_header(skb) - skb->data;
+}
+
 static inline int skb_mac_header_was_set(const struct sk_buff *skb)
 {
 	return skb->mac_header != (typeof(skb->mac_header))~0U;

include/net/tc_act/tc_pedit.h

@@ -3,11 +3,17 @@
 
 #include <net/act_api.h>
 
+struct tcf_pedit_key_ex {
+	enum pedit_header_type htype;
+	enum pedit_cmd cmd;
+};
+
 struct tcf_pedit {
 	struct tc_action	common;
 	unsigned char		tcfp_nkeys;
 	unsigned char		tcfp_flags;
 	struct tc_pedit_key	*tcfp_keys;
+	struct tcf_pedit_key_ex	*tcfp_keys_ex;
 };
 #define to_pedit(a) ((struct tcf_pedit *)a)
 

include/uapi/linux/tc_act/tc_pedit.h

@@ -11,10 +11,41 @@ enum {
 	TCA_PEDIT_TM,
 	TCA_PEDIT_PARMS,
 	TCA_PEDIT_PAD,
+	TCA_PEDIT_PARMS_EX,
+	TCA_PEDIT_KEYS_EX,
+	TCA_PEDIT_KEY_EX,
 	__TCA_PEDIT_MAX
 };
 #define TCA_PEDIT_MAX (__TCA_PEDIT_MAX - 1)
 
+enum {
+	TCA_PEDIT_KEY_EX_HTYPE = 1,
+	TCA_PEDIT_KEY_EX_CMD = 2,
+	__TCA_PEDIT_KEY_EX_MAX
+};
+#define TCA_PEDIT_KEY_EX_MAX (__TCA_PEDIT_KEY_EX_MAX - 1)
+
+ /* TCA_PEDIT_KEY_EX_HDR_TYPE_NETWROK is a special case for legacy users. It
+  * means no specific header type - offset is relative to the network layer
+  */
+enum pedit_header_type {
+	TCA_PEDIT_KEY_EX_HDR_TYPE_NETWORK = 0,
+	TCA_PEDIT_KEY_EX_HDR_TYPE_ETH = 1,
+	TCA_PEDIT_KEY_EX_HDR_TYPE_IP4 = 2,
+	TCA_PEDIT_KEY_EX_HDR_TYPE_IP6 = 3,
+	TCA_PEDIT_KEY_EX_HDR_TYPE_TCP = 4,
+	TCA_PEDIT_KEY_EX_HDR_TYPE_UDP = 5,
+	__PEDIT_HDR_TYPE_MAX,
+};
+#define TCA_PEDIT_HDR_TYPE_MAX (__PEDIT_HDR_TYPE_MAX - 1)
+
+enum pedit_cmd {
+	TCA_PEDIT_KEY_EX_CMD_SET = 0,
+	TCA_PEDIT_KEY_EX_CMD_ADD = 1,
+	__PEDIT_CMD_MAX,
+};
+#define TCA_PEDIT_CMD_MAX (__PEDIT_CMD_MAX - 1)
+
 struct tc_pedit_key {
 	__u32           mask;  /* AND */
 	__u32           val;   /*XOR */