[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <5eaf7a5df51b687f3354d9e065c3358f56b5ad34.1737387719.git.petrm@nvidia.com>
Date: Mon, 20 Jan 2025 16:43:06 +0100
From: Petr Machata <petrm@...dia.com>
To: Stephen Hemminger <stephen@...workplumber.org>, David Ahern
<dsahern@...nel.org>, <netdev@...r.kernel.org>
CC: Petr Machata <petrm@...dia.com>
Subject: [PATCH iproute2-next] ip: vxlan: Support IFLA_VXLAN_RESERVED_BITS
A new attribute, IFLA_VXLAN_RESERVED_BITS, was added in Linux kernel
commit 6c11379b104e ("vxlan: Add an attribute to make VXLAN header
validation configurable") (See the link below for the full patchset).
The payload is a 64-bit binary field that covers the VXLAN header. The set
bits indicate which bits in a VXLAN packet header should be allowed to
carry 1's. Support the new attribute through a CLI keyword "reserved_bits".
Link: https://patch.msgid.link/173378643250.273075.13832548579412179113.git-patchwork-notify@kernel.org
Signed-off-by: Petr Machata <petrm@...dia.com>
---
ip/iplink_vxlan.c | 20 ++++++++++++++++++++
man/man8/ip-link.8.in | 9 +++++++++
2 files changed, 29 insertions(+)
diff --git a/ip/iplink_vxlan.c b/ip/iplink_vxlan.c
index 7781d60b..9649a8eb 100644
--- a/ip/iplink_vxlan.c
+++ b/ip/iplink_vxlan.c
@@ -52,6 +52,7 @@ static void print_explain(FILE *f)
" [ dev PHYS_DEV ]\n"
" [ dstport PORT ]\n"
" [ srcport MIN MAX ]\n"
+ " [ reserved_bits VALUE ]\n"
" [ [no]learning ]\n"
" [ [no]proxy ]\n"
" [ [no]rsc ]\n"
@@ -337,6 +338,17 @@ static int vxlan_parse_opt(struct link_util *lu, int argc, char **argv,
check_duparg(&attrs, IFLA_VXLAN_LOCALBYPASS,
*argv, *argv);
addattr8(n, 1024, IFLA_VXLAN_LOCALBYPASS, 0);
+ } else if (strcmp(*argv, "reserved_bits") == 0) {
+ NEXT_ARG();
+ __be64 bits;
+
+ check_duparg(&attrs, IFLA_VXLAN_RESERVED_BITS,
+ *argv, *argv);
+ if (get_be64(&bits, *argv, 0))
+ invarg("reserved_bits", *argv);
+ addattr_l(n, 1024, IFLA_VXLAN_RESERVED_BITS,
+ &bits, sizeof(bits));
+
} else if (!matches(*argv, "external")) {
check_duparg(&attrs, IFLA_VXLAN_COLLECT_METADATA,
*argv, *argv);
@@ -601,6 +613,14 @@ static void vxlan_print_opt(struct link_util *lu, FILE *f, struct rtattr *tb[])
((maxaddr = rta_getattr_u32(tb[IFLA_VXLAN_LIMIT])) != 0))
print_uint(PRINT_ANY, "limit", "maxaddr %u ", maxaddr);
+ if (tb[IFLA_VXLAN_RESERVED_BITS]) {
+ __be64 reserved_bits =
+ rta_getattr_u64(tb[IFLA_VXLAN_RESERVED_BITS]);
+
+ print_0xhex(PRINT_ANY, "reserved_bits",
+ "reserved_bits %#llx ", ntohll(reserved_bits));
+ }
+
if (tb[IFLA_VXLAN_GBP])
print_null(PRINT_ANY, "gbp", "gbp ", NULL);
if (tb[IFLA_VXLAN_GPE])
diff --git a/man/man8/ip-link.8.in b/man/man8/ip-link.8.in
index 64b5ba21..d0f30556 100644
--- a/man/man8/ip-link.8.in
+++ b/man/man8/ip-link.8.in
@@ -632,6 +632,8 @@ the following additional arguments are supported:
] [
.BI srcport " MIN MAX "
] [
+.BI reserved_bits " VALUE "
+] [
.RB [ no ] learning
] [
.RB [ no ] proxy
@@ -725,6 +727,13 @@ bit is not set.
- specifies the range of port numbers to use as UDP
source ports to communicate to the remote VXLAN tunnel endpoint.
+.sp
+.BI reserved_bits " VALUE "
+- by default the kernel rejects packets that have bits set outside of the fields
+required by the features enabled on the VXLAN netdevice. \fBreserved_bits\fR is
+a 64-bit quantity specifying which bits it should be possible to set in a VXLAN
+header. Each bit set in the value is a tolerated bit set in a packet.
+
.sp
.RB [ no ] learning
- specifies if unknown source link layer addresses and IP addresses
--
2.47.0
Powered by blists - more mailing lists