[PATCH net-next 0/2] ip_gre: add support for i/o_flags update

[PATCH net-next 0/2] ip_gre: add support for i/o_flags update

[Xin Long]

ip_gre is using as many ip_tunnel apis as possible, newlink works
fine as gre would do it's own part in .ndo_init. But when changing
link, ip_tunnel_changelink doesn't even update i/o_flags, and also
the update of these flags would cause some other gre's properties
need to be updated or recalculated.

These two patch are to add i/o_flags update and then do adjustment
on some gre's properties according to the new i/o_flags.

Xin Long (2):
  ip_gre: add the support for i/o_flags update via netlink
  ip_gre: add the support for i/o_flags update via ioctl

 net/ipv4/ip_gre.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 53 insertions(+), 5 deletions(-)

-- 
2.1.0

[PATCH net-next 1/2] ip_gre: add the support for i/o_flags update via netlink

[Xin Long]

Now ip_gre is using ip_tunnel_changelink to update it's properties, but
ip_tunnel_changelink in ip_tunnel doesn't update i/o_flags as a common
function.

o_flags updates would cause that tunnel->tun_hlen / hlen and dev->mtu /
needed_headroom need to be recalculated, and dev->(hw_)features need to
be updated as well.

Therefore, we can't just add the update into ip_tunnel_update called
in ip_tunnel_changelink, and it's also better not to touch ip_tunnel
codes.

This patch updates i/o_flags and calls ipgre_link_update to recalculate
these gre properties after ip_tunnel_changelink does the common update.

Note that since ipgre_link_update doesn't know the lower dev, it will
update gre->hlen, dev->mtu and dev->needed_headroom with the value of
'new tun_hlen - old tun_hlen'. In this way, we can avoid many redundant
codes, unlike ip6_gre.

Reported-by: Jianlin Shi <jishi@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
---
 net/ipv4/ip_gre.c | 39 +++++++++++++++++++++++++++++++++++++--
 1 file changed, 37 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index c105a31..81e1e20 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -773,6 +773,30 @@ static netdev_tx_t gre_tap_xmit(struct sk_buff *skb,
 	return NETDEV_TX_OK;
 }
 
+static void ipgre_link_update(struct net_device *dev, bool set_mtu)
+{
+	struct ip_tunnel *tunnel = netdev_priv(dev);
+	int len;
+
+	len = tunnel->tun_hlen;
+	tunnel->tun_hlen = gre_calc_hlen(tunnel->parms.o_flags);
+	len = tunnel->tun_hlen - len;
+	tunnel->hlen = tunnel->hlen + len;
+
+	dev->needed_headroom = dev->needed_headroom + len;
+	if (set_mtu)
+		dev->mtu = max_t(int, dev->mtu - len, 68);
+
+	if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) {
+		if (!(tunnel->parms.o_flags & TUNNEL_CSUM) ||
+		    tunnel->encap.type == TUNNEL_ENCAP_NONE) {
+			dev->features |= NETIF_F_GSO_SOFTWARE;
+			dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+		}
+		dev->features |= NETIF_F_LLTX;
+	}
+}
+
 static int ipgre_tunnel_ioctl(struct net_device *dev,
 			      struct ifreq *ifr, int cmd)
 {
@@ -1307,9 +1331,9 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
 			    struct netlink_ext_ack *extack)
 {
 	struct ip_tunnel *t = netdev_priv(dev);
-	struct ip_tunnel_parm p;
 	struct ip_tunnel_encap ipencap;
 	__u32 fwmark = t->fwmark;
+	struct ip_tunnel_parm p;
 	int err;
 
 	if (ipgre_netlink_encap_parms(data, &ipencap)) {
@@ -1322,7 +1346,18 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
 	err = ipgre_netlink_parms(dev, data, tb, &p, &fwmark);
 	if (err < 0)
 		return err;
-	return ip_tunnel_changelink(dev, tb, &p, fwmark);
+
+	err = ip_tunnel_changelink(dev, tb, &p, fwmark);
+	if (err < 0)
+		return err;
+
+	t->parms.i_flags = p.i_flags;
+	t->parms.o_flags = p.o_flags;
+
+	if (strcmp(dev->rtnl_link_ops->kind, "erspan"))
+		ipgre_link_update(dev, !tb[IFLA_MTU]);
+
+	return 0;
 }
 
 static size_t ipgre_get_size(const struct net_device *dev)
-- 
2.1.0

Re: [PATCH net-next 1/2] ip_gre: add the support for i/o_flags update via netlink

[William Tu]

On Tue, Nov 7, 2017 at 12:33 AM, Xin Long <lucien.xin@gmail.com> wrote:
> Now ip_gre is using ip_tunnel_changelink to update it's properties, but
> ip_tunnel_changelink in ip_tunnel doesn't update i/o_flags as a common
> function.
>
> o_flags updates would cause that tunnel->tun_hlen / hlen and dev->mtu /
> needed_headroom need to be recalculated, and dev->(hw_)features need to
> be updated as well.
>
> Therefore, we can't just add the update into ip_tunnel_update called
> in ip_tunnel_changelink, and it's also better not to touch ip_tunnel
> codes.
>
> This patch updates i/o_flags and calls ipgre_link_update to recalculate
> these gre properties after ip_tunnel_changelink does the common update.
>
> Note that since ipgre_link_update doesn't know the lower dev, it will
> update gre->hlen, dev->mtu and dev->needed_headroom with the value of
> 'new tun_hlen - old tun_hlen'. In this way, we can avoid many redundant
> codes, unlike ip6_gre.
>
> Reported-by: Jianlin Shi <jishi@redhat.com>
> Signed-off-by: Xin Long <lucien.xin@gmail.com>
> ---
>  net/ipv4/ip_gre.c | 39 +++++++++++++++++++++++++++++++++++++--
>  1 file changed, 37 insertions(+), 2 deletions(-)
>
> diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
> index c105a31..81e1e20 100644
> --- a/net/ipv4/ip_gre.c
> +++ b/net/ipv4/ip_gre.c
> @@ -773,6 +773,30 @@ static netdev_tx_t gre_tap_xmit(struct sk_buff *skb,
>         return NETDEV_TX_OK;
>  }
>
> +static void ipgre_link_update(struct net_device *dev, bool set_mtu)
> +{
> +       struct ip_tunnel *tunnel = netdev_priv(dev);
> +       int len;
> +
> +       len = tunnel->tun_hlen;
> +       tunnel->tun_hlen = gre_calc_hlen(tunnel->parms.o_flags);
> +       len = tunnel->tun_hlen - len;
> +       tunnel->hlen = tunnel->hlen + len;
> +
> +       dev->needed_headroom = dev->needed_headroom + len;
> +       if (set_mtu)
> +               dev->mtu = max_t(int, dev->mtu - len, 68);
> +
> +       if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) {
> +               if (!(tunnel->parms.o_flags & TUNNEL_CSUM) ||
> +                   tunnel->encap.type == TUNNEL_ENCAP_NONE) {
> +                       dev->features |= NETIF_F_GSO_SOFTWARE;
> +                       dev->hw_features |= NETIF_F_GSO_SOFTWARE;
> +               }
> +               dev->features |= NETIF_F_LLTX;
> +       }
> +}
> +
>  static int ipgre_tunnel_ioctl(struct net_device *dev,
>                               struct ifreq *ifr, int cmd)
>  {
> @@ -1307,9 +1331,9 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
>                             struct netlink_ext_ack *extack)
>  {
>         struct ip_tunnel *t = netdev_priv(dev);
> -       struct ip_tunnel_parm p;
>         struct ip_tunnel_encap ipencap;
>         __u32 fwmark = t->fwmark;
> +       struct ip_tunnel_parm p;
>         int err;
>
>         if (ipgre_netlink_encap_parms(data, &ipencap)) {
> @@ -1322,7 +1346,18 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
>         err = ipgre_netlink_parms(dev, data, tb, &p, &fwmark);
>         if (err < 0)
>                 return err;
> -       return ip_tunnel_changelink(dev, tb, &p, fwmark);
> +
> +       err = ip_tunnel_changelink(dev, tb, &p, fwmark);
> +       if (err < 0)
> +               return err;
> +
> +       t->parms.i_flags = p.i_flags;
> +       t->parms.o_flags = p.o_flags;
> +
> +       if (strcmp(dev->rtnl_link_ops->kind, "erspan"))
> +               ipgre_link_update(dev, !tb[IFLA_MTU]);

just to comment: ERSPAN does not need update because its GRE flag is
fixed to SEQ.

Acked-by: William Tu <u9012063@gmail.com>

[PATCH net-next 2/2] ip_gre: add the support for i/o_flags update via ioctl

[Xin Long]

As patch 'ip_gre: add the support for i/o_flags update via netlink'
did for netlink, we also need to do the same job for these update
via ioctl.

This patch is to update i/o_flags and call ipgre_link_update to
recalculate these gre properties after ip_tunnel_ioctl does the
common update.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
---
 net/ipv4/ip_gre.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 81e1e20..bb62391 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -800,17 +800,19 @@ static void ipgre_link_update(struct net_device *dev, bool set_mtu)
 static int ipgre_tunnel_ioctl(struct net_device *dev,
 			      struct ifreq *ifr, int cmd)
 {
-	int err;
 	struct ip_tunnel_parm p;
+	int err;
 
 	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
 		return -EFAULT;
+
 	if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) {
 		if (p.iph.version != 4 || p.iph.protocol != IPPROTO_GRE ||
-		    p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)) ||
-		    ((p.i_flags|p.o_flags)&(GRE_VERSION|GRE_ROUTING)))
+		    p.iph.ihl != 5 || (p.iph.frag_off & htons(~IP_DF)) ||
+		    ((p.i_flags | p.o_flags) & (GRE_VERSION | GRE_ROUTING)))
 			return -EINVAL;
 	}
+
 	p.i_flags = gre_flags_to_tnl_flags(p.i_flags);
 	p.o_flags = gre_flags_to_tnl_flags(p.o_flags);
 
@@ -818,11 +820,22 @@ static int ipgre_tunnel_ioctl(struct net_device *dev,
 	if (err)
 		return err;
 
+	if (cmd == SIOCCHGTUNNEL) {
+		struct ip_tunnel *t = netdev_priv(dev);
+
+		t->parms.i_flags = p.i_flags;
+		t->parms.o_flags = p.o_flags;
+
+		if (strcmp(dev->rtnl_link_ops->kind, "erspan"))
+			ipgre_link_update(dev, true);
+	}
+
 	p.i_flags = gre_tnl_flags_to_gre_flags(p.i_flags);
 	p.o_flags = gre_tnl_flags_to_gre_flags(p.o_flags);
 
 	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
 		return -EFAULT;
+
 	return 0;
 }
 
-- 
2.1.0

Re: [PATCH net-next 2/2] ip_gre: add the support for i/o_flags update via ioctl

[William Tu]

On Tue, Nov 7, 2017 at 12:33 AM, Xin Long <lucien.xin@gmail.com> wrote:
> As patch 'ip_gre: add the support for i/o_flags update via netlink'
> did for netlink, we also need to do the same job for these update
> via ioctl.
>
> This patch is to update i/o_flags and call ipgre_link_update to
> recalculate these gre properties after ip_tunnel_ioctl does the
> common update.
>
> Signed-off-by: Xin Long <lucien.xin@gmail.com>
> ---
LGTM.

Acked-by: William Tu <u9012063@gmail.com>

Re: [PATCH net-next 0/2] ip_gre: add support for i/o_flags update

[David Miller]

From: Xin Long <lucien.xin@gmail.com>
Date: Tue,  7 Nov 2017 16:33:07 +0800

> ip_gre is using as many ip_tunnel apis as possible, newlink works
> fine as gre would do it's own part in .ndo_init. But when changing
> link, ip_tunnel_changelink doesn't even update i/o_flags, and also
> the update of these flags would cause some other gre's properties
> need to be updated or recalculated.
> 
> These two patch are to add i/o_flags update and then do adjustment
> on some gre's properties according to the new i/o_flags.

Series applied, thank you.