* [PATCH] ip: Support filter links/neighs with no master
@ 2021-08-19 10:45 Lahav Schlesinger
2021-08-24 4:19 ` David Ahern
0 siblings, 1 reply; 4+ messages in thread
From: Lahav Schlesinger @ 2021-08-19 10:45 UTC (permalink / raw)
To: netdev; +Cc: dsahern
Commit d3432bf10f17 ("net: Support filtering interfaces on no master")
in the kernel added support for filtering interfaces/neighbours that
have no master interface.
This patch completes it and adds this support to iproute2:
1. ip link show nomaster
2. ip address show nomaster
3. ip neighbour {show | flush} nomaster
Signed-off-by: Lahav Schlesinger <lschlesinger@drivenets.com>
---
ip/ipaddress.c | 4 +++-
ip/iplink.c | 2 +-
ip/ipneigh.c | 4 +++-
man/man8/ip-address.8.in | 7 ++++++-
man/man8/ip-link.8.in | 7 ++++++-
man/man8/ip-neighbour.8 | 7 ++++++-
6 files changed, 25 insertions(+), 6 deletions(-)
diff --git a/ip/ipaddress.c b/ip/ipaddress.c
index 85534aaf..a5b683f5 100644
--- a/ip/ipaddress.c
+++ b/ip/ipaddress.c
@@ -61,7 +61,7 @@ static void usage(void)
" [ to PREFIX ] [ FLAG-LIST ] [ label LABEL ] [up]\n"
" ip address [ show [ dev IFNAME ] [ scope SCOPE-ID ] [ master DEVICE ]\n"
" [ type TYPE ] [ to PREFIX ] [ FLAG-LIST ]\n"
- " [ label LABEL ] [up] [ vrf NAME ] ]\n"
+ " [ label LABEL ] [up] [ vrf NAME ] [ nomaster ] ]\n"
" ip address {showdump|restore}\n"
"IFADDR := PREFIX | ADDR peer PREFIX\n"
" [ broadcast ADDR ] [ anycast ADDR ]\n"
@@ -2123,6 +2123,8 @@ static int ipaddr_list_flush_or_save(int argc, char **argv, int action)
if (!name_is_vrf(*argv))
invarg("Not a valid VRF name\n", *argv);
filter.master = ifindex;
+ } else if (strcmp(*argv, "nomaster") == 0) {
+ filter.master = -1;
} else if (strcmp(*argv, "type") == 0) {
int soff;
diff --git a/ip/iplink.c b/ip/iplink.c
index 18b2ea25..f017f1f3 100644
--- a/ip/iplink.c
+++ b/ip/iplink.c
@@ -119,7 +119,7 @@ void iplink_usage(void)
" [ protodown_reason PREASON { on | off } ]\n"
" [ gso_max_size BYTES ] | [ gso_max_segs PACKETS ]\n"
"\n"
- " ip link show [ DEVICE | group GROUP ] [up] [master DEV] [vrf NAME] [type TYPE]\n"
+ " ip link show [ DEVICE | group GROUP ] [up] [master DEV] [vrf NAME] [type TYPE] [nomaster]\n"
"\n"
" ip link xstats type TYPE [ ARGS ]\n"
"\n"
diff --git a/ip/ipneigh.c b/ip/ipneigh.c
index 95bde520..b4a2f6df 100644
--- a/ip/ipneigh.c
+++ b/ip/ipneigh.c
@@ -54,7 +54,7 @@ static void usage(void)
" [ dev DEV ] [ router ] [ extern_learn ] [ protocol PROTO ]\n"
"\n"
" ip neigh { show | flush } [ proxy ] [ to PREFIX ] [ dev DEV ] [ nud STATE ]\n"
- " [ vrf NAME ]\n"
+ " [ vrf NAME ] [ nomaster ]\n"
" ip neigh get { ADDR | proxy ADDR } dev DEV\n"
"\n"
"STATE := { delay | failed | incomplete | noarp | none |\n"
@@ -536,6 +536,8 @@ static int do_show_or_flush(int argc, char **argv, int flush)
if (!name_is_vrf(*argv))
invarg("Not a valid VRF name\n", *argv);
filter.master = ifindex;
+ } else if (strcmp(*argv, "nomaster") == 0) {
+ filter.master = -1;
} else if (strcmp(*argv, "unused") == 0) {
filter.unused_only = 1;
} else if (strcmp(*argv, "nud") == 0) {
diff --git a/man/man8/ip-address.8.in b/man/man8/ip-address.8.in
index fe773c91..65f67e06 100644
--- a/man/man8/ip-address.8.in
+++ b/man/man8/ip-address.8.in
@@ -49,7 +49,8 @@ ip-address \- protocol address management
.IR TYPE " ] [ "
.B vrf
.IR NAME " ] [ "
-.BR up " ] ]"
+.BR up " ] ["
+.BR nomaster " ] ]"
.ti -8
.BR "ip address" " { " showdump " | " restore " }"
@@ -340,6 +341,10 @@ output.
.B up
only list running interfaces.
+.TP
+.B nomaster
+only list interfaces with no master.
+
.TP
.BR dynamic " and " permanent
(IPv6 only) only list addresses installed due to stateless
diff --git a/man/man8/ip-link.8.in b/man/man8/ip-link.8.in
index 572bed87..e749f23d 100644
--- a/man/man8/ip-link.8.in
+++ b/man/man8/ip-link.8.in
@@ -179,7 +179,8 @@ ip-link \- network device configuration
.B type
.IR ETYPE " ] ["
.B vrf
-.IR NAME " ]"
+.IR NAME " ] ["
+.BR nomaster " ]"
.ti -8
.B ip link xstats
@@ -2528,6 +2529,10 @@ interface list by comparing it with the relevant attribute in case the kernel
didn't filter already. Therefore any string is accepted, but may lead to empty
output.
+.TP
+.B nomaster
+only show devices with no master
+
.SS ip link xstats - display extended statistics
.TP
diff --git a/man/man8/ip-neighbour.8 b/man/man8/ip-neighbour.8
index a27f9ef8..02862964 100644
--- a/man/man8/ip-neighbour.8
+++ b/man/man8/ip-neighbour.8
@@ -35,7 +35,8 @@ ip-neighbour \- neighbour/arp tables management.
.B nud
.IR STATE " ] [ "
.B vrf
-.IR NAME " ] "
+.IR NAME " ] ["
+.BR nomaster " ]"
.ti -8
.B ip neigh get
@@ -191,6 +192,10 @@ only list the neighbours attached to this device.
.BI vrf " NAME"
only list the neighbours for given VRF.
+.TP
+.BI nomaster
+only list neighbours attached to an interface with no master.
+
.TP
.BI proxy
list neighbour proxies.
--
2.25.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] ip: Support filter links/neighs with no master
2021-08-19 10:45 [PATCH] ip: Support filter links/neighs with no master Lahav Schlesinger
@ 2021-08-24 4:19 ` David Ahern
2021-08-24 14:16 ` Lahav Schlesinger
0 siblings, 1 reply; 4+ messages in thread
From: David Ahern @ 2021-08-24 4:19 UTC (permalink / raw)
To: Lahav Schlesinger, netdev; +Cc: dsahern
On 8/19/21 4:45 AM, Lahav Schlesinger wrote:
> Commit d3432bf10f17 ("net: Support filtering interfaces on no master")
> in the kernel added support for filtering interfaces/neighbours that
> have no master interface.
>
> This patch completes it and adds this support to iproute2:
> 1. ip link show nomaster
> 2. ip address show nomaster
> 3. ip neighbour {show | flush} nomaster
>
> Signed-off-by: Lahav Schlesinger <lschlesinger@drivenets.com>
> ---
> ip/ipaddress.c | 4 +++-
> ip/iplink.c | 2 +-
> ip/ipneigh.c | 4 +++-
> man/man8/ip-address.8.in | 7 ++++++-
> man/man8/ip-link.8.in | 7 ++++++-
> man/man8/ip-neighbour.8 | 7 ++++++-
> 6 files changed, 25 insertions(+), 6 deletions(-)
>
> diff --git a/ip/ipaddress.c b/ip/ipaddress.c
> index 85534aaf..a5b683f5 100644
> --- a/ip/ipaddress.c
> +++ b/ip/ipaddress.c
> @@ -61,7 +61,7 @@ static void usage(void)
> " [ to PREFIX ] [ FLAG-LIST ] [ label LABEL ] [up]\n"
> " ip address [ show [ dev IFNAME ] [ scope SCOPE-ID ] [ master DEVICE ]\n"
move [ nomaster ] to here on a new line to keep the existing line
length, and
> " [ type TYPE ] [ to PREFIX ] [ FLAG-LIST ]\n"
> - " [ label LABEL ] [up] [ vrf NAME ] ]\n"
> + " [ label LABEL ] [up] [ vrf NAME ] [ nomaster ] ]\n"
make this 'novrf' for consistency with existing syntax.
Similarly for the other 2 commands.
> " ip address {showdump|restore}\n"
> "IFADDR := PREFIX | ADDR peer PREFIX\n"
> " [ broadcast ADDR ] [ anycast ADDR ]\n"
> @@ -2123,6 +2123,8 @@ static int ipaddr_list_flush_or_save(int argc, char **argv, int action)
> if (!name_is_vrf(*argv))
> invarg("Not a valid VRF name\n", *argv);
> filter.master = ifindex;
> + } else if (strcmp(*argv, "nomaster") == 0) {
and of course make this a compound check for novrf.
> + filter.master = -1;
> } else if (strcmp(*argv, "type") == 0) {
> int soff;
>
> diff --git a/ip/iplink.c b/ip/iplink.c
> index 18b2ea25..f017f1f3 100644
> --- a/ip/iplink.c
> +++ b/ip/iplink.c
> @@ -119,7 +119,7 @@ void iplink_usage(void)
> " [ protodown_reason PREASON { on | off } ]\n"
> " [ gso_max_size BYTES ] | [ gso_max_segs PACKETS ]\n"
> "\n"
> - " ip link show [ DEVICE | group GROUP ] [up] [master DEV] [vrf NAME] [type TYPE]\n"
> + " ip link show [ DEVICE | group GROUP ] [up] [master DEV] [vrf NAME] [type TYPE] [nomaster]\n"
this line is already too long so add the new options on a new line.
> "\n"
> " ip link xstats type TYPE [ ARGS ]\n"
> "\n"
> diff --git a/ip/ipneigh.c b/ip/ipneigh.c
> index 95bde520..b4a2f6df 100644
> --- a/ip/ipneigh.c
> +++ b/ip/ipneigh.c
> @@ -54,7 +54,7 @@ static void usage(void)
> " [ dev DEV ] [ router ] [ extern_learn ] [ protocol PROTO ]\n"
> "\n"
> " ip neigh { show | flush } [ proxy ] [ to PREFIX ] [ dev DEV ] [ nud STATE ]\n"
> - " [ vrf NAME ]\n"
> + " [ vrf NAME ] [ nomaster ]\n"
> " ip neigh get { ADDR | proxy ADDR } dev DEV\n"
> "\n"
> "STATE := { delay | failed | incomplete | noarp | none |\n"
> @@ -536,6 +536,8 @@ static int do_show_or_flush(int argc, char **argv, int flush)
> if (!name_is_vrf(*argv))
> invarg("Not a valid VRF name\n", *argv);
> filter.master = ifindex;
> + } else if (strcmp(*argv, "nomaster") == 0) {
> + filter.master = -1;
> } else if (strcmp(*argv, "unused") == 0) {
> filter.unused_only = 1;
> } else if (strcmp(*argv, "nud") == 0) {
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] ip: Support filter links/neighs with no master
2021-08-24 4:19 ` David Ahern
@ 2021-08-24 14:16 ` Lahav Schlesinger
2021-08-24 16:16 ` David Ahern
0 siblings, 1 reply; 4+ messages in thread
From: Lahav Schlesinger @ 2021-08-24 14:16 UTC (permalink / raw)
To: David Ahern; +Cc: netdev, dsahern
On Mon, Aug 23, 2021 at 09:19:56PM -0700, David Ahern wrote:
> CAUTION: External E-Mail - Use caution with links and attachments
>
>
> On 8/19/21 4:45 AM, Lahav Schlesinger wrote:
> > Commit d3432bf10f17 ("net: Support filtering interfaces on no master")
> > in the kernel added support for filtering interfaces/neighbours that
> > have no master interface.
> >
> > This patch completes it and adds this support to iproute2:
> > 1. ip link show nomaster
> > 2. ip address show nomaster
> > 3. ip neighbour {show | flush} nomaster
> >
> > Signed-off-by: Lahav Schlesinger <lschlesinger@drivenets.com>
> > ---
> > ip/ipaddress.c | 4 +++-
> > ip/iplink.c | 2 +-
> > ip/ipneigh.c | 4 +++-
> > man/man8/ip-address.8.in | 7 ++++++-
> > man/man8/ip-link.8.in | 7 ++++++-
> > man/man8/ip-neighbour.8 | 7 ++++++-
> > 6 files changed, 25 insertions(+), 6 deletions(-)
> >
> > diff --git a/ip/ipaddress.c b/ip/ipaddress.c
> > index 85534aaf..a5b683f5 100644
> > --- a/ip/ipaddress.c
> > +++ b/ip/ipaddress.c
> > @@ -61,7 +61,7 @@ static void usage(void)
> > " [ to PREFIX ] [ FLAG-LIST ] [ label LABEL ] [up]\n"
> > " ip address [ show [ dev IFNAME ] [ scope SCOPE-ID ] [ master DEVICE ]\n"
>
> move [ nomaster ] to here on a new line to keep the existing line
> length, and
>
> > " [ type TYPE ] [ to PREFIX ] [ FLAG-LIST ]\n"
> > - " [ label LABEL ] [up] [ vrf NAME ] ]\n"
> > + " [ label LABEL ] [up] [ vrf NAME ] [ nomaster ] ]\n"
>
> make this 'novrf' for consistency with existing syntax.
>
> Similarly for the other 2 commands.
I think "nomaster" is more fitting here, because this option only affects
interfaces that have no master at all, so e.g. slaves of a bundle will
not be returned by the "nomaster" option, even if they are in the default VRF.
I'm planning next to add support for the "novrf" option which will indeed
only affect interfaces which are in the default VRF, even if they have a
master.
>
> > " ip address {showdump|restore}\n"
> > "IFADDR := PREFIX | ADDR peer PREFIX\n"
> > " [ broadcast ADDR ] [ anycast ADDR ]\n"
> > @@ -2123,6 +2123,8 @@ static int ipaddr_list_flush_or_save(int argc, char **argv, int action)
> > if (!name_is_vrf(*argv))
> > invarg("Not a valid VRF name\n", *argv);
> > filter.master = ifindex;
> > + } else if (strcmp(*argv, "nomaster") == 0) {
>
> and of course make this a compound check for novrf.
>
> > + filter.master = -1;
> > } else if (strcmp(*argv, "type") == 0) {
> > int soff;
> >
> > diff --git a/ip/iplink.c b/ip/iplink.c
> > index 18b2ea25..f017f1f3 100644
> > --- a/ip/iplink.c
> > +++ b/ip/iplink.c
> > @@ -119,7 +119,7 @@ void iplink_usage(void)
> > " [ protodown_reason PREASON { on | off } ]\n"
> > " [ gso_max_size BYTES ] | [ gso_max_segs PACKETS ]\n"
> > "\n"
> > - " ip link show [ DEVICE | group GROUP ] [up] [master DEV] [vrf NAME] [type TYPE]\n"
> > + " ip link show [ DEVICE | group GROUP ] [up] [master DEV] [vrf NAME] [type TYPE] [nomaster]\n"
>
> this line is already too long so add the new options on a new line.
>
> > "\n"
> > " ip link xstats type TYPE [ ARGS ]\n"
> > "\n"
> > diff --git a/ip/ipneigh.c b/ip/ipneigh.c
> > index 95bde520..b4a2f6df 100644
> > --- a/ip/ipneigh.c
> > +++ b/ip/ipneigh.c
> > @@ -54,7 +54,7 @@ static void usage(void)
> > " [ dev DEV ] [ router ] [ extern_learn ] [ protocol PROTO ]\n"
> > "\n"
> > " ip neigh { show | flush } [ proxy ] [ to PREFIX ] [ dev DEV ] [ nud STATE ]\n"
> > - " [ vrf NAME ]\n"
> > + " [ vrf NAME ] [ nomaster ]\n"
> > " ip neigh get { ADDR | proxy ADDR } dev DEV\n"
> > "\n"
> > "STATE := { delay | failed | incomplete | noarp | none |\n"
> > @@ -536,6 +536,8 @@ static int do_show_or_flush(int argc, char **argv, int flush)
> > if (!name_is_vrf(*argv))
> > invarg("Not a valid VRF name\n", *argv);
> > filter.master = ifindex;
> > + } else if (strcmp(*argv, "nomaster") == 0) {
> > + filter.master = -1;
> > } else if (strcmp(*argv, "unused") == 0) {
> > filter.unused_only = 1;
> > } else if (strcmp(*argv, "nud") == 0) {
>
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] ip: Support filter links/neighs with no master
2021-08-24 14:16 ` Lahav Schlesinger
@ 2021-08-24 16:16 ` David Ahern
0 siblings, 0 replies; 4+ messages in thread
From: David Ahern @ 2021-08-24 16:16 UTC (permalink / raw)
To: Lahav Schlesinger; +Cc: netdev, dsahern
On 8/24/21 8:16 AM, Lahav Schlesinger wrote:
>>> diff --git a/ip/ipaddress.c b/ip/ipaddress.c
>>> index 85534aaf..a5b683f5 100644
>>> --- a/ip/ipaddress.c
>>> +++ b/ip/ipaddress.c
>>> @@ -61,7 +61,7 @@ static void usage(void)
>>> " [ to PREFIX ] [ FLAG-LIST ] [ label LABEL ] [up]\n"
>>> " ip address [ show [ dev IFNAME ] [ scope SCOPE-ID ] [ master DEVICE ]\n"
>>
>> move [ nomaster ] to here on a new line to keep the existing line
>> length, and
>>
>>> " [ type TYPE ] [ to PREFIX ] [ FLAG-LIST ]\n"
>>> - " [ label LABEL ] [up] [ vrf NAME ] ]\n"
>>> + " [ label LABEL ] [up] [ vrf NAME ] [ nomaster ] ]\n"
>>
>> make this 'novrf' for consistency with existing syntax.
>>
>> Similarly for the other 2 commands.
>
> I think "nomaster" is more fitting here, because this option only affects
> interfaces that have no master at all, so e.g. slaves of a bundle will
> not be returned by the "nomaster" option, even if they are in the default VRF.
I was thinking both - 'nomaster' for bridges and 'novrf' for VRFs. Just
friendlier syntax for the latter.
>
> I'm planning next to add support for the "novrf" option which will indeed
> only affect interfaces which are in the default VRF, even if they have a
> master.
ok,
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-08-24 16:17 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-19 10:45 [PATCH] ip: Support filter links/neighs with no master Lahav Schlesinger
2021-08-24 4:19 ` David Ahern
2021-08-24 14:16 ` Lahav Schlesinger
2021-08-24 16:16 ` David Ahern
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.