Commit 001b98c9 authored by Paolo Abeni's avatar Paolo Abeni

Merge branch 'net-ipv6-ioam6-introduce-tunsrc'

Justin Iurman says:

====================
net: ipv6: ioam6: introduce tunsrc

This series introduces a new feature called "tunsrc" (just like seg6
already does).

v3:
- address Jakub's comments

v2:
- add links to performance result figures (see patch#2 description)
- move the ipv6_addr_any() check out of the datapath
====================

Link: https://patch.msgid.link/20240817131818.11834-1-justin.iurman@uliege.beSigned-off-by: default avatarPaolo Abeni <pabeni@redhat.com>
parents f32c821a 273f8c14
......@@ -50,6 +50,12 @@ enum {
IOAM6_IPTUNNEL_FREQ_K, /* u32 */
IOAM6_IPTUNNEL_FREQ_N, /* u32 */
/* Tunnel src address.
* For encap,auto modes.
* Optional (automatic if not provided).
*/
IOAM6_IPTUNNEL_SRC, /* struct in6_addr */
__IOAM6_IPTUNNEL_MAX,
};
......
......@@ -42,8 +42,10 @@ struct ioam6_lwt {
struct ioam6_lwt_freq freq;
atomic_t pkt_cnt;
u8 mode;
bool has_tunsrc;
struct in6_addr tunsrc;
struct in6_addr tundst;
struct ioam6_lwt_encap tuninfo;
struct ioam6_lwt_encap tuninfo;
};
static const struct netlink_range_validation freq_range = {
......@@ -72,8 +74,10 @@ static const struct nla_policy ioam6_iptunnel_policy[IOAM6_IPTUNNEL_MAX + 1] = {
[IOAM6_IPTUNNEL_MODE] = NLA_POLICY_RANGE(NLA_U8,
IOAM6_IPTUNNEL_MODE_MIN,
IOAM6_IPTUNNEL_MODE_MAX),
[IOAM6_IPTUNNEL_SRC] = NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),
[IOAM6_IPTUNNEL_DST] = NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),
[IOAM6_IPTUNNEL_TRACE] = NLA_POLICY_EXACT_LEN(sizeof(struct ioam6_trace_hdr)),
[IOAM6_IPTUNNEL_TRACE] = NLA_POLICY_EXACT_LEN(
sizeof(struct ioam6_trace_hdr)),
};
static bool ioam6_validate_trace_hdr(struct ioam6_trace_hdr *trace)
......@@ -143,6 +147,11 @@ static int ioam6_build_state(struct net *net, struct nlattr *nla,
else
mode = nla_get_u8(tb[IOAM6_IPTUNNEL_MODE]);
if (tb[IOAM6_IPTUNNEL_SRC] && mode == IOAM6_IPTUNNEL_MODE_INLINE) {
NL_SET_ERR_MSG(extack, "no tunnel src expected with this mode");
return -EINVAL;
}
if (!tb[IOAM6_IPTUNNEL_DST] && mode != IOAM6_IPTUNNEL_MODE_INLINE) {
NL_SET_ERR_MSG(extack, "this mode needs a tunnel destination");
return -EINVAL;
......@@ -167,16 +176,29 @@ static int ioam6_build_state(struct net *net, struct nlattr *nla,
ilwt = ioam6_lwt_state(lwt);
err = dst_cache_init(&ilwt->cache, GFP_ATOMIC);
if (err) {
kfree(lwt);
return err;
}
if (err)
goto free_lwt;
atomic_set(&ilwt->pkt_cnt, 0);
ilwt->freq.k = freq_k;
ilwt->freq.n = freq_n;
ilwt->mode = mode;
if (!tb[IOAM6_IPTUNNEL_SRC]) {
ilwt->has_tunsrc = false;
} else {
ilwt->has_tunsrc = true;
ilwt->tunsrc = nla_get_in6_addr(tb[IOAM6_IPTUNNEL_SRC]);
if (ipv6_addr_any(&ilwt->tunsrc)) {
NL_SET_ERR_MSG_ATTR(extack, tb[IOAM6_IPTUNNEL_SRC],
"invalid tunnel source address");
err = -EINVAL;
goto free_cache;
}
}
if (tb[IOAM6_IPTUNNEL_DST])
ilwt->tundst = nla_get_in6_addr(tb[IOAM6_IPTUNNEL_DST]);
......@@ -201,6 +223,11 @@ static int ioam6_build_state(struct net *net, struct nlattr *nla,
*ts = lwt;
return 0;
free_cache:
dst_cache_destroy(&ilwt->cache);
free_lwt:
kfree(lwt);
return err;
}
static int ioam6_do_fill(struct net *net, struct sk_buff *skb)
......@@ -256,6 +283,8 @@ static int ioam6_do_inline(struct net *net, struct sk_buff *skb,
static int ioam6_do_encap(struct net *net, struct sk_buff *skb,
struct ioam6_lwt_encap *tuninfo,
bool has_tunsrc,
struct in6_addr *tunsrc,
struct in6_addr *tundst)
{
struct dst_entry *dst = skb_dst(skb);
......@@ -285,8 +314,12 @@ static int ioam6_do_encap(struct net *net, struct sk_buff *skb,
hdr->nexthdr = NEXTHDR_HOP;
hdr->payload_len = cpu_to_be16(skb->len - sizeof(*hdr));
hdr->daddr = *tundst;
ipv6_dev_get_saddr(net, dst->dev, &hdr->daddr,
IPV6_PREFER_SRC_PUBLIC, &hdr->saddr);
if (has_tunsrc)
memcpy(&hdr->saddr, tunsrc, sizeof(*tunsrc));
else
ipv6_dev_get_saddr(net, dst->dev, &hdr->daddr,
IPV6_PREFER_SRC_PUBLIC, &hdr->saddr);
skb_postpush_rcsum(skb, hdr, len);
......@@ -328,7 +361,9 @@ static int ioam6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
case IOAM6_IPTUNNEL_MODE_ENCAP:
do_encap:
/* Encapsulation (ip6ip6) */
err = ioam6_do_encap(net, skb, &ilwt->tuninfo, &ilwt->tundst);
err = ioam6_do_encap(net, skb, &ilwt->tuninfo,
ilwt->has_tunsrc, &ilwt->tunsrc,
&ilwt->tundst);
if (unlikely(err))
goto drop;
......@@ -414,6 +449,13 @@ static int ioam6_fill_encap_info(struct sk_buff *skb,
goto ret;
if (ilwt->mode != IOAM6_IPTUNNEL_MODE_INLINE) {
if (ilwt->has_tunsrc) {
err = nla_put_in6_addr(skb, IOAM6_IPTUNNEL_SRC,
&ilwt->tunsrc);
if (err)
goto ret;
}
err = nla_put_in6_addr(skb, IOAM6_IPTUNNEL_DST, &ilwt->tundst);
if (err)
goto ret;
......@@ -435,8 +477,12 @@ static int ioam6_encap_nlsize(struct lwtunnel_state *lwtstate)
nla_total_size(sizeof(ilwt->mode)) +
nla_total_size(sizeof(ilwt->tuninfo.traceh));
if (ilwt->mode != IOAM6_IPTUNNEL_MODE_INLINE)
if (ilwt->mode != IOAM6_IPTUNNEL_MODE_INLINE) {
if (ilwt->has_tunsrc)
nlsize += nla_total_size(sizeof(ilwt->tunsrc));
nlsize += nla_total_size(sizeof(ilwt->tundst));
}
return nlsize;
}
......@@ -451,17 +497,21 @@ static int ioam6_encap_cmp(struct lwtunnel_state *a, struct lwtunnel_state *b)
return (ilwt_a->freq.k != ilwt_b->freq.k ||
ilwt_a->freq.n != ilwt_b->freq.n ||
ilwt_a->mode != ilwt_b->mode ||
ilwt_a->has_tunsrc != ilwt_b->has_tunsrc ||
(ilwt_a->mode != IOAM6_IPTUNNEL_MODE_INLINE &&
!ipv6_addr_equal(&ilwt_a->tundst, &ilwt_b->tundst)) ||
(ilwt_a->mode != IOAM6_IPTUNNEL_MODE_INLINE &&
ilwt_a->has_tunsrc &&
!ipv6_addr_equal(&ilwt_a->tunsrc, &ilwt_b->tunsrc)) ||
trace_a->namespace_id != trace_b->namespace_id);
}
static const struct lwtunnel_encap_ops ioam6_iptun_ops = {
.build_state = ioam6_build_state,
.destroy_state = ioam6_destroy_state,
.output = ioam6_output,
.output = ioam6_output,
.fill_encap = ioam6_fill_encap_info,
.get_encap_size = ioam6_encap_nlsize,
.get_encap_size = ioam6_encap_nlsize,
.cmp_encap = ioam6_encap_cmp,
.owner = THIS_MODULE,
};
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment