Revision v1 of 2 in this series.
6 messages, 2 authors, 2017-08-28 · open the first message on its own page
From: David Lebrun <hidden>
Date: 2017-08-28 19:04:56
This patch series adds support for the new L2ENCAP mode for SRv6 encapsulations. David Lebrun (2): iproute: add support for seg6 l2encap mode man: add documentation for seg6 l2encap mode ip/iproute_lwtunnel.c | 39 ++++++++++++++++++++++++++++----------- man/man8/ip-route.8.in | 6 +++++- 2 files changed, 33 insertions(+), 12 deletions(-) -- 2.10.2
From: David Lebrun <hidden>
Date: 2017-08-28 19:04:57
This patch adds support for the L2ENCAP seg6 mode, enabling to encapsulate L2 frames within SRv6 packets. Signed-off-by: David Lebrun <redacted> --- ip/iproute_lwtunnel.c | 39 ++++++++++++++++++++++++++++----------- 1 file changed, 28 insertions(+), 11 deletions(-)
diff --git a/ip/iproute_lwtunnel.c b/ip/iproute_lwtunnel.c
index 14294c6..f125670 100644
--- a/ip/iproute_lwtunnel.c
+++ b/ip/iproute_lwtunnel.c@@ -110,6 +110,30 @@ static void print_srh(FILE *fp, struct ipv6_sr_hdr *srh) } } +static const char *format_seg6mode_type(int mode) +{ + if (mode == SEG6_IPTUN_MODE_ENCAP) + return "encap"; + else if (mode == SEG6_IPTUN_MODE_INLINE) + return "inline"; + else if (mode == SEG6_IPTUN_MODE_L2ENCAP) + return "l2encap"; + + return "<unknown>"; +} + +static int read_seg6mode_type(const char *mode) +{ + if (strcmp(mode, "encap") == 0) + return SEG6_IPTUN_MODE_ENCAP; + else if (strcmp(mode, "inline") == 0) + return SEG6_IPTUN_MODE_INLINE; + else if (strcmp(mode, "l2encap") == 0) + return SEG6_IPTUN_MODE_L2ENCAP; + + return -1; +} + static void print_encap_seg6(FILE *fp, struct rtattr *encap) { struct rtattr *tb[SEG6_IPTUNNEL_MAX+1];
@@ -121,8 +145,7 @@ static void print_encap_seg6(FILE *fp, struct rtattr *encap) return; tuninfo = RTA_DATA(tb[SEG6_IPTUNNEL_SRH]); - fprintf(fp, "mode %s ", - (tuninfo->mode == SEG6_IPTUN_MODE_ENCAP) ? "encap" : "inline"); + fprintf(fp, "mode %s ", format_seg6mode_type(tuninfo->mode)); print_srh(fp, tuninfo->srh); }
@@ -457,11 +480,8 @@ static int parse_encap_seg6(struct rtattr *rta, size_t len, int *argcp, NEXT_ARG(); if (mode_ok++) duparg2("mode", *argv); - if (strcmp(*argv, "encap") == 0) - encap = 1; - else if (strcmp(*argv, "inline") == 0) - encap = 0; - else + encap = read_seg6mode_type(*argv); + if (encap < 0) invarg("\"mode\" value is invalid\n", *argv); } else if (strcmp(*argv, "segs") == 0) { NEXT_ARG();
@@ -490,10 +510,7 @@ static int parse_encap_seg6(struct rtattr *rta, size_t len, int *argcp, tuninfo = malloc(sizeof(*tuninfo) + srhlen); memset(tuninfo, 0, sizeof(*tuninfo) + srhlen); - if (encap) - tuninfo->mode = SEG6_IPTUN_MODE_ENCAP; - else - tuninfo->mode = SEG6_IPTUN_MODE_INLINE; + tuninfo->mode = encap; memcpy(tuninfo->srh, srh, srhlen);
--
2.10.2
From: David Lebrun <hidden>
Date: 2017-08-28 19:04:57
This patch adds documentation for the seg6 L2ENCAP encapsulation mode. Signed-off-by: David Lebrun <redacted> --- man/man8/ip-route.8.in | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/man/man8/ip-route.8.in b/man/man8/ip-route.8.in
index 11dd9d0..803de3b 100644
--- a/man/man8/ip-route.8.in
+++ b/man/man8/ip-route.8.in@@ -214,7 +214,7 @@ throw " | " unreachable " | " prohibit " | " blackhole " | " nat " ]" .IR ENCAP_SEG6 " := " .B seg6 .BR mode " [ " -.BR encap " | " inline " ] " +.BR encap " | " inline " | " l2encap " ] " .B segs .IR SEGMENTS " [ " .B hmac
@@ -750,6 +750,10 @@ is a set of encapsulation attributes specific to the - Encapsulate packet in an outer IPv6 header with SRH .sp +.B mode l2encap +- Encapsulate ingress L2 frame within an outer IPv6 header and SRH +.sp + .I SEGMENTS - List of comma-separated IPv6 addresses .sp
--
2.10.2
From: Stephen Hemminger <stephen@networkplumber.org>
Date: 2017-08-28 19:07:27
On Mon, 28 Aug 2017 20:05:24 +0100 David Lebrun [off-list ref] wrote:
+static const char *format_seg6mode_type(int mode) +{ + if (mode == SEG6_IPTUN_MODE_ENCAP) + return "encap"; + else if (mode == SEG6_IPTUN_MODE_INLINE) + return "inline"; + else if (mode == SEG6_IPTUN_MODE_L2ENCAP) + return "l2encap"; + + return "<unknown>"; +} + +static int read_seg6mode_type(const char *mode) +{ + if (strcmp(mode, "encap") == 0) + return SEG6_IPTUN_MODE_ENCAP; + else if (strcmp(mode, "inline") == 0) + return SEG6_IPTUN_MODE_INLINE; + else if (strcmp(mode, "l2encap") == 0) + return SEG6_IPTUN_MODE_L2ENCAP; + + return -1; +}
Since these values probably will grow over time, it would make sense to have this a name/value table.
From: David Lebrun <hidden>
Date: 2017-08-28 19:11:31
On 08/28/2017 08:07 PM, Stephen Hemminger wrote:
Since these values probably will grow over time, it would make sense to have this a name/value table.
I wasn't sure if it was worth it for 3 values. I do not foresee a large growth either, but I will send a v2 will a table anyway David
From: Stephen Hemminger <stephen@networkplumber.org>
Date: 2017-08-28 19:32:46
On Mon, 28 Aug 2017 20:11:59 +0100 David Lebrun [off-list ref] wrote:
On 08/28/2017 08:07 PM, Stephen Hemminger wrote:quoted
Since these values probably will grow over time, it would make sense to have this a name/value table.I wasn't sure if it was worth it for 3 values. I do not foresee a large growth either, but I will send a v2 will a table anyway David
Either way, not a big worry.