[PATCH v2] netlink: fixup regression in RTM_GETADDR

[PATCH v2] netlink: fixup regression in RTM_GETADDR

[baloo]

From: Arthur Gautier <baloo@gandi.net>

This commit fixes a regression in AF_INET/RTM_GETADDR and
AF_INET6/RTM_GETADDR.

Before this commit, the kernel would stop dumping addresses once the first
skb was full and end the stream with NLMSG_DONE(-EMSGSIZE). The error
shouldn't be sent back to netlink_dump so the callback is kept alive. The
userspace is expected to call back with a new empty skb.

Changes from V1:
 - The error is not handled in netlink_dump anymore but rather in
   inet_dump_ifaddr and inet6_dump_addr directly as suggested by
   David Ahern.

Fixes: d7e38611b81e ("net/ipv4: Put target net when address dump fails due to bad attributes")
Fixes: 242afaa6968c ("net/ipv6: Put target net when address dump fails due to bad attributes")

Cc: David Ahern <dsahern@gmail.com>
Cc: "David S . Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Signed-off-by: Arthur Gautier <baloo@gandi.net>
---
 net/ipv4/devinet.c  | 2 +-
 net/ipv6/addrconf.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 608a6f4223fb9..fecd0e7672b5d 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1826,7 +1826,7 @@ static int inet_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
 	if (fillargs.netnsid >= 0)
 		put_net(tgt_net);
 
-	return err < 0 ? err : skb->len;
+	return skb->len ? : err;
 }
 
 static void rtmsg_ifa(int event, struct in_ifaddr *ifa, struct nlmsghdr *nlh,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 045597b9a7c05..e3cb53b0ef670 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -5154,7 +5154,7 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb,
 	if (fillargs.netnsid >= 0)
 		put_net(tgt_net);
 
-	return err < 0 ? err : skb->len;
+	return skb->len ? : err;
 }
 
 static int inet6_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
-- 
2.20.1

Re: [PATCH v2] netlink: fixup regression in RTM_GETADDR

[David Ahern]

On 12/30/18 7:10 PM, baloo@gandi.net wrote:
> From: Arthur Gautier <baloo@gandi.net>
> 
> This commit fixes a regression in AF_INET/RTM_GETADDR and
> AF_INET6/RTM_GETADDR.
> 
> Before this commit, the kernel would stop dumping addresses once the first
> skb was full and end the stream with NLMSG_DONE(-EMSGSIZE). The error
> shouldn't be sent back to netlink_dump so the callback is kept alive. The
> userspace is expected to call back with a new empty skb.
> 
> Changes from V1:
>  - The error is not handled in netlink_dump anymore but rather in
>    inet_dump_ifaddr and inet6_dump_addr directly as suggested by
>    David Ahern.
> 
> Fixes: d7e38611b81e ("net/ipv4: Put target net when address dump fails due to bad attributes")
> Fixes: 242afaa6968c ("net/ipv6: Put target net when address dump fails due to bad attributes")
> 
> Cc: David Ahern <dsahern@gmail.com>
> Cc: "David S . Miller" <davem@davemloft.net>
> Cc: netdev@vger.kernel.org
> Signed-off-by: Arthur Gautier <baloo@gandi.net>
> ---
>  net/ipv4/devinet.c  | 2 +-
>  net/ipv6/addrconf.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
> index 608a6f4223fb9..fecd0e7672b5d 100644
> --- a/net/ipv4/devinet.c
> +++ b/net/ipv4/devinet.c
> @@ -1826,7 +1826,7 @@ static int inet_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
>  	if (fillargs.netnsid >= 0)
>  		put_net(tgt_net);
>  
> -	return err < 0 ? err : skb->len;
> +	return skb->len ? : err;
>  }
>  
>  static void rtmsg_ifa(int event, struct in_ifaddr *ifa, struct nlmsghdr *nlh,
> diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
> index 045597b9a7c05..e3cb53b0ef670 100644
> --- a/net/ipv6/addrconf.c
> +++ b/net/ipv6/addrconf.c
> @@ -5154,7 +5154,7 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb,
>  	if (fillargs.netnsid >= 0)
>  		put_net(tgt_net);
>  
> -	return err < 0 ? err : skb->len;
> +	return skb->len ? : err;
>  }
>  
>  static int inet6_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
> 

That looks correct. The only error for in_dev_dump_addr is -EMSGSIZE and
that should not be returned. The other errors happen with skb->len == 0.
Same with the ipv6 version.

Reviewed-by: David Ahern <dsahern@gmail.com>

Re: [PATCH v2] netlink: fixup regression in RTM_GETADDR

[David Miller]

From: baloo@gandi.net
Date: Mon, 31 Dec 2018 02:10:58 +0000

> From: Arthur Gautier <baloo@gandi.net>
> 
> This commit fixes a regression in AF_INET/RTM_GETADDR and
> AF_INET6/RTM_GETADDR.
> 
> Before this commit, the kernel would stop dumping addresses once the first
> skb was full and end the stream with NLMSG_DONE(-EMSGSIZE). The error
> shouldn't be sent back to netlink_dump so the callback is kept alive. The
> userspace is expected to call back with a new empty skb.
> 
> Changes from V1:
>  - The error is not handled in netlink_dump anymore but rather in
>    inet_dump_ifaddr and inet6_dump_addr directly as suggested by
>    David Ahern.
> 
> Fixes: d7e38611b81e ("net/ipv4: Put target net when address dump fails due to bad attributes")
> Fixes: 242afaa6968c ("net/ipv6: Put target net when address dump fails due to bad attributes")
> 
> Cc: David Ahern <dsahern@gmail.com>
> Cc: "David S . Miller" <davem@davemloft.net>
> Cc: netdev@vger.kernel.org
> Signed-off-by: Arthur Gautier <baloo@gandi.net>

Applied and queued up for -stable, thanks!