From: Simon Horman <simon.horman@netronome•com>
To: pravin shelar <pshelar@ovn•org>
Cc: Linux Kernel Network Developers <netdev@vger•kernel.org>,
ovs dev <dev@openvswitch•org>, Jiri Benc <jbenc@redhat•com>
Subject: Re: [PATCH net-next v10 2/5] openvswitch: set skb protocol and mac_len when receiving on internal device
Date: Thu, 23 Jun 2016 11:04:38 +0900 [thread overview]
Message-ID: <20160623020436.GA28594@vergenet.net> (raw)
In-Reply-To: <CAOrHB_B2mUqUOgwZroRqaYwsHCKd-Vj5nMWMvPxfF=quxhJ2cA@mail.gmail.com>
On Tue, Jun 21, 2016 at 09:30:17AM -0700, pravin shelar wrote:
> On Mon, Jun 20, 2016 at 7:25 PM, Simon Horman
> <simon.horman@netronome•com> wrote:
> > [Cc Jiri Benc]
> >
> > On Sat, Jun 18, 2016 at 06:38:54PM -0700, pravin shelar wrote:
> >> On Thu, Jun 16, 2016 at 10:53 PM, Simon Horman
> >> <simon.horman@netronome•com> wrote:
> >> > On Tue, Jun 07, 2016 at 03:45:27PM -0700, pravin shelar wrote:
> >> >> On Mon, Jun 6, 2016 at 8:08 PM, Simon Horman <simon.horman@netronome•com> wrote:
> >> >> > On Thu, Jun 02, 2016 at 03:01:47PM -0700, pravin shelar wrote:
> >> >> >> On Wed, Jun 1, 2016 at 11:24 PM, Simon Horman
> >> >> >> <simon.horman@netronome•com> wrote:
> >> >> >> > * Set skb protocol based on contents of packet. I have observed this is
> >> >> >> > necessary to get actual protocol of a packet when it is injected into an
> >> >> >> > internal device e.g. by libnet in which case skb protocol will be set to
> >> >> >> > ETH_ALL.
> >> ....
> >> ....
> >> >> > eth_type = eth_type_trans(skb, skb->dev);
> >> >> > skb->mac_len = skb->data - skb_mac_header(skb);
> >> >> > __skb_push(skb, skb->mac_len);
> >> >> >
> >> >> > if (eth_type == htons(ETH_P_8021Q))
> >> >> > skb->mac_len += VLAN_HLEN;
> >> >> >
> >> >> > Perhaps that logic ought to be in a helper used by both internal_dev_xmit()
> >> >> > and netdev_port_receive(). Or somehow centralised in ovs_vport_receive().
> >> >>
> >> >> This does looks bit complex. Can we use other skb metadata like
> >> >> skb_mac_header_was_set()?
> >> >
> >> > Yes, I think that can be made to work if skb->mac_header is unset
> >> > for l3 packets in netdev_port_receive(). The following is an incremental
> >> > patch on the entire series. Is this the kind of thing you had in mind?
> >> >
> >> > diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
> >> > index 86f2cfb19de3..42587d5bf894 100644
> >> > --- a/net/openvswitch/flow.c
> >> > +++ b/net/openvswitch/flow.c
> >> > @@ -729,7 +729,7 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
> >> > key->phy.skb_mark = skb->mark;
> >> > ovs_ct_fill_key(skb, key);
> >> > key->ovs_flow_hash = 0;
> >> > - key->phy.is_layer3 = skb->mac_len == 0;
> >> > + key->phy.is_layer3 = skb_mac_header_was_set(skb) == 0;
> >> > key->recirc_id = 0;
> >> >
> >> > err = key_extract(skb, key);
> >> > diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
> >> > index 484ba529c682..8973d4db509b 100644
> >> > --- a/net/openvswitch/vport-internal_dev.c
> >> > +++ b/net/openvswitch/vport-internal_dev.c
> >> > @@ -50,7 +50,6 @@ static int internal_dev_xmit(struct sk_buff *skb, struct net_device *netdev)
> >> >
> >> > skb->protocol = eth_type_trans(skb, netdev);
> >> > skb_push(skb, ETH_HLEN);
> >> > - skb_reset_mac_len(skb);
> >> >
> >> > len = skb->len;
> >> > rcu_read_lock();
> >> > diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c
> >> > index 3df36df62ee9..4cf3f12ffc99 100644
> >> > --- a/net/openvswitch/vport-netdev.c
> >> > +++ b/net/openvswitch/vport-netdev.c
> >> > @@ -60,22 +60,9 @@ static void netdev_port_receive(struct sk_buff *skb)
> >> > if (vport->dev->type == ARPHRD_ETHER) {
> >> > skb_push(skb, ETH_HLEN);
> >> > skb_postpush_rcsum(skb, skb->data, ETH_HLEN);
> >> > - } else if (vport->dev->type == ARPHRD_NONE) {
> >> > - if (skb->protocol == htons(ETH_P_TEB)) {
> >> > - __be16 eth_type;
> >> > -
> >> > - if (unlikely(skb->len < ETH_HLEN))
> >> > - goto error;
> >> > -
> >> > - eth_type = eth_type_trans(skb, skb->dev);
> >> > - skb->mac_len = skb->data - skb_mac_header(skb);
> >> > - __skb_push(skb, skb->mac_len);
> >> > -
> >> > - if (eth_type == htons(ETH_P_8021Q))
> >> > - skb->mac_len += VLAN_HLEN;
> >> > - } else {
> >> > - skb->mac_len = 0;
> >> > - }
> >> > + } else if (vport->dev->type == ARPHRD_NONE &&
> >> > + skb->protocol != htons(ETH_P_TEB)) {
> >> > + skb->mac_header = (typeof(skb->mac_header))~0U;
> >> > }
> >> >
> >> > ovs_vport_receive(vport, skb, skb_tunnel_info(skb));
> >>
> >> This certainly looks better. I was wondering if we can unset the mac
> >> header offset in L3 tunnel devices itself. So there is no need to have
> >> this check here.
> >
> > I think that might be possible for GRE by modifying the following in
> > __ipgre_rcv().
> >
> > if (tunnel->dev->type != ARPHRD_NONE)
> > skb_pop_mac_header(skb);
> > else
> > skb_reset_mac_header(skb);
> >
> > But I am unsure what side effects this might have on other users of the
> > code.
> >
> I think it is fine with device of type ARPHRD_NONE. metadata tunnel
> devices would be of this type anyways.
>
> > Jiri, do you have any thoughts on this?
I think you are right as IIRC the call to skb_reset_mac_header was
added for this use-case. Its unfortunate that we can't use it in
internal_dev_xmit() because of loosing track of MPLS as you mentioned
earlier. But it does seem that setting mac_header to ~0 works well
in conjunction with updates to OvS posted earlier in this sub-therad.
I have the following working. Jiri, is the ip_gre portion acceptable to you?
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 58d323289872..e6772b6934a3 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -279,7 +279,7 @@ static int __ipgre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi,
if (tunnel->dev->type != ARPHRD_NONE)
skb_pop_mac_header(skb);
else
- skb_reset_mac_header(skb);
+ skb->mac_header = (typeof(skb->mac_header))~0U;
if (tunnel->collect_md) {
__be16 flags;
__be64 tun_id;
diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c
index 4cf3f12ffc99..82b10802abe6 100644
--- a/net/openvswitch/vport-netdev.c
+++ b/net/openvswitch/vport-netdev.c
@@ -60,9 +60,6 @@ static void netdev_port_receive(struct sk_buff *skb)
if (vport->dev->type == ARPHRD_ETHER) {
skb_push(skb, ETH_HLEN);
skb_postpush_rcsum(skb, skb->data, ETH_HLEN);
- } else if (vport->dev->type == ARPHRD_NONE &&
- skb->protocol != htons(ETH_P_TEB)) {
- skb->mac_header = (typeof(skb->mac_header))~0U;
}
ovs_vport_receive(vport, skb, skb_tunnel_info(skb));
next prev parent reply other threads:[~2016-06-23 2:04 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-02 6:24 [PATCH net-next v10 0/5] openvswitch: support for layer 3 encapsulated packets Simon Horman
2016-06-02 6:24 ` [PATCH net-next v10 1/5] net: add skb_vlan_accel helper Simon Horman
2016-06-02 22:01 ` pravin shelar
2016-06-02 6:24 ` [PATCH net-next v10 2/5] openvswitch: set skb protocol and mac_len when receiving on internal device Simon Horman
[not found] ` <1464848686-7656-3-git-send-email-simon.horman-wFxRvT7yatFl57MIdRCFDg@public.gmane.org>
2016-06-02 22:01 ` pravin shelar
2016-06-07 3:08 ` Simon Horman
[not found] ` <20160607030809.GE31696-IxS8c3vjKQDk1uMJSBkQmQ@public.gmane.org>
2016-06-07 22:45 ` pravin shelar
2016-06-17 5:53 ` Simon Horman
[not found] ` <20160617055331.GA24833-IxS8c3vjKQDk1uMJSBkQmQ@public.gmane.org>
2016-06-19 1:38 ` pravin shelar
[not found] ` <CAOrHB_BWfLk8yba2XkF43Hm-czGUZnVWQQ=HW3-vvpFEH7GNpA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-06-21 2:25 ` Simon Horman
[not found] ` <20160621022458.GA28358-IxS8c3vjKQDk1uMJSBkQmQ@public.gmane.org>
2016-06-21 16:30 ` pravin shelar
2016-06-23 2:04 ` Simon Horman [this message]
2016-06-27 9:35 ` Jiri Benc
2016-06-02 6:24 ` [PATCH net-next v10 3/5] openvswitch: add support to push and pop mpls for layer3 packets Simon Horman
2016-06-02 22:02 ` pravin shelar
2016-06-07 2:51 ` Simon Horman
2016-06-07 22:45 ` pravin shelar
2016-06-02 6:24 ` [PATCH net-next v10 4/5] openvswitch: add layer 3 flow/port support Simon Horman
[not found] ` <1464848686-7656-5-git-send-email-simon.horman-wFxRvT7yatFl57MIdRCFDg@public.gmane.org>
2016-06-02 22:02 ` pravin shelar
2016-06-07 2:46 ` Simon Horman
[not found] ` <20160607024609.GC31696-IxS8c3vjKQDk1uMJSBkQmQ@public.gmane.org>
2016-06-07 22:45 ` pravin shelar
2016-06-17 6:53 ` Simon Horman
2016-06-02 6:24 ` [PATCH net-next v10 5/5] openvswitch: use ipgre tunnel rather than gretap tunnel Simon Horman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160623020436.GA28594@vergenet.net \
--to=simon.horman@netronome$(echo .)com \
--cc=dev@openvswitch$(echo .)org \
--cc=jbenc@redhat$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
--cc=pshelar@ovn$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox