[PATCH net-next v2] ipv4: igmp: report IFA_TARGET_NETNSID in multicast address dumps

[PATCH net-next v2] ipv4: igmp: report IFA_TARGET_NETNSID in multicast address dumps

[Wenbo Sun]

Commit eb4e17a1d915 ("netlink: support dumping IPv4 multicast addresses")
extended RTM_GETMULTICAST to dump IPv4 multicast addresses by reusing
inet_valid_dump_ifaddr_req(), which already accepts IFA_TARGET_NETNSID
and switches to the target network namespace. However,
inet_fill_ifmcaddr() does not include IFA_TARGET_NETNSID in the dump
replies yet.

In contrast, IPv4 unicast dumps (inet_fill_ifaddr()), IPv6 multicast
dumps (inet6_fill_ifmcaddr()), and AF_PACKET multicast dumps
(dev_mc_fill_addr()) all report IFA_TARGET_NETNSID when a netns id is
specified. Adding the same behavior to IPv4 multicast dumps aligns
reporting across address families and helps userspace consistently
identify target-netns replies.

Emit IFA_TARGET_NETNSID in inet_fill_ifmcaddr() when args->netnsid >= 0,
and initialize fillargs.netnsid = -1 in inet_ifmcaddr_notify() so that
multicast join/leave notifications do not include IFA_TARGET_NETNSID or
exceed their preallocated skb size.

This alignment was identified with the assistance of an LLM-based code
review tool, confirmed in a User-Mode Linux (UML) environment where AF_INET
RTM_GETMULTICAST dumps with IFA_TARGET_NETNSID omitted the attribute while
AF_INET6 included it, and verified in UML that AF_INET multicast dumps
now report IFA_TARGET_NETNSID, regular dumps without netnsid continue to
work, and multicast join/leave notifications succeed without warning.

Signed-off-by: Wenbo Sun <sunwenbo@cmss.chinamobile.com>
---
Changes in v2:
- Target net-next as an enhancement/alignment change and drop the Fixes
  tag, per Hangbin Liu's review.
- Soften and clarify the commit message to reflect feature alignment with
  inet_fill_ifaddr(), inet6_fill_ifmcaddr(), and dev_mc_fill_addr().
- Link to v1: https://lore.kernel.org/r/20261111082222.327963-1-sunwenbo@cmss.chinamobile.com

 net/ipv4/igmp.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 144fca158adc..831d003677ba 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -1517,6 +1517,12 @@ int inet_fill_ifmcaddr(struct sk_buff *skb, struct net_device *dev,
 	ifm->ifa_scope = RT_SCOPE_UNIVERSE;
 	ifm->ifa_index = dev->ifindex;
 
+	if (args->netnsid >= 0 &&
+	    nla_put_s32(skb, IFA_TARGET_NETNSID, args->netnsid)) {
+		nlmsg_cancel(skb, nlh);
+		return -EMSGSIZE;
+	}
+
 	ci.cstamp = (READ_ONCE(im->mca_cstamp) - INITIAL_JIFFIES) * 100UL / HZ;
 	ci.tstamp = ci.cstamp;
 	ci.ifa_prefered = INFINITY_LIFE_TIME;
@@ -1538,6 +1544,7 @@ static void inet_ifmcaddr_notify(struct net_device *dev,
 {
 	struct inet_fill_args fillargs = {
 		.event = event,
+		.netnsid = -1,
 	};
 	struct net *net = dev_net(dev);
 	struct sk_buff *skb;
-- 
2.33.0

Re: [PATCH net-next v2] ipv4: igmp: report IFA_TARGET_NETNSID in multicast address dumps

[Hangbin Liu]

On Thu, Nov 12, 2026 at 04:23:52PM +0800, Wenbo Sun wrote:
> Commit eb4e17a1d915 ("netlink: support dumping IPv4 multicast addresses")
> extended RTM_GETMULTICAST to dump IPv4 multicast addresses by reusing
> inet_valid_dump_ifaddr_req(), which already accepts IFA_TARGET_NETNSID
> and switches to the target network namespace. However,
> inet_fill_ifmcaddr() does not include IFA_TARGET_NETNSID in the dump
> replies yet.
> 
> In contrast, IPv4 unicast dumps (inet_fill_ifaddr()), IPv6 multicast
> dumps (inet6_fill_ifmcaddr()), and AF_PACKET multicast dumps
> (dev_mc_fill_addr()) all report IFA_TARGET_NETNSID when a netns id is
> specified. Adding the same behavior to IPv4 multicast dumps aligns
> reporting across address families and helps userspace consistently
> identify target-netns replies.
> 
> Emit IFA_TARGET_NETNSID in inet_fill_ifmcaddr() when args->netnsid >= 0,
> and initialize fillargs.netnsid = -1 in inet_ifmcaddr_notify() so that
> multicast join/leave notifications do not include IFA_TARGET_NETNSID or
> exceed their preallocated skb size.
> 
> This alignment was identified with the assistance of an LLM-based code
> review tool, confirmed in a User-Mode Linux (UML) environment where AF_INET
> RTM_GETMULTICAST dumps with IFA_TARGET_NETNSID omitted the attribute while
> AF_INET6 included it, and verified in UML that AF_INET multicast dumps
> now report IFA_TARGET_NETNSID, regular dumps without netnsid continue to
> work, and multicast join/leave notifications succeed without warning.
> 
> Signed-off-by: Wenbo Sun <sunwenbo@cmss.chinamobile.com>
> ---
> Changes in v2:
> - Target net-next as an enhancement/alignment change and drop the Fixes
>   tag, per Hangbin Liu's review.
> - Soften and clarify the commit message to reflect feature alignment with
>   inet_fill_ifaddr(), inet6_fill_ifmcaddr(), and dev_mc_fill_addr().
> - Link to v1: https://lore.kernel.org/r/20261111082222.327963-1-sunwenbo@cmss.chinamobile.com
> 
>  net/ipv4/igmp.c | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
> index 144fca158adc..831d003677ba 100644
> --- a/net/ipv4/igmp.c
> +++ b/net/ipv4/igmp.c
> @@ -1517,6 +1517,12 @@ int inet_fill_ifmcaddr(struct sk_buff *skb, struct net_device *dev,
>  	ifm->ifa_scope = RT_SCOPE_UNIVERSE;
>  	ifm->ifa_index = dev->ifindex;
>  
> +	if (args->netnsid >= 0 &&
> +	    nla_put_s32(skb, IFA_TARGET_NETNSID, args->netnsid)) {
> +		nlmsg_cancel(skb, nlh);
> +		return -EMSGSIZE;
> +	}
> +
>  	ci.cstamp = (READ_ONCE(im->mca_cstamp) - INITIAL_JIFFIES) * 100UL / HZ;
>  	ci.tstamp = ci.cstamp;
>  	ci.ifa_prefered = INFINITY_LIFE_TIME;
> @@ -1538,6 +1544,7 @@ static void inet_ifmcaddr_notify(struct net_device *dev,
>  {
>  	struct inet_fill_args fillargs = {
>  		.event = event,
> +		.netnsid = -1,
>  	};
>  	struct net *net = dev_net(dev);
>  	struct sk_buff *skb;
> -- 
> 2.33.0
> 

Reviewed-by: Hangbin Liu <liuhangbin@kylinos.cn>