| Message ID | 1412294070-11930-4-git-send-email-vyasevic@redhat.com |
|---|---|
| State | Changes Requested, archived |
| Delegated to: | David Miller |
| Headers | show |
On 2014/10/03 8:54, Vladislav Yasevich wrote: > Currently when vlan filtering is turned on on the bridge, the bridge > will drop all traffic untill the user configures the filter. This > isn't very nice for ports that don't care about vlans and just > want untagged traffic. > > A concept of a default_pvid was recently introduced. This patch > adds filtering support for default_pvid. Now, ports that don't > care about vlans and don't define there own filter will belong > to the VLAN of the default_pvid and continue to receive untagged > traffic. > > This filtering can be disabled by setting default_pvid to 0. > > Signed-off-by: Vladislav Yasevich <vyasevic@redhat.com> Acked-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Thanks, Toshiaki Makita -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, Oct 2, 2014 at 4:54 PM, Vladislav Yasevich <vyasevich@gmail.com> wrote: > +static void br_vlan_disable_default_pvid(struct net_bridge *br) > +{ > + struct net_bridge_port *p; > + u16 pvid = br->default_pvid; > + > + /* Disable default_pvid on all ports where it is still > + * configured. > + */ > + This empty line is not necessary. > + if (vlan_default_pvid(br_get_vlan_info(br), pvid)) > + br_vlan_delete(br, pvid); > + > + list_for_each_entry(p, &br->port_list, list) { > + if (vlan_default_pvid(nbp_get_vlan_info(p), pvid)) > + nbp_vlan_delete(p, pvid); > + } > + > + br->default_pvid = 0; > +} > + > +static int __br_vlan_set_default_pvid(struct net_bridge *br, u16 pvid) > +{ > + struct net_bridge_port *p; > + u16 old_pvid; > + int err; > + DECLARE_BITMAP(changed, BR_MAX_PORTS); This bitmap will use 128 bytes on stack, why not using heap? > + > + bitmap_zero(changed, BR_MAX_PORTS); > + > + /* This function runs with filtering turned off so we can > + * remove the old pvid configuration and add the new one after > + * without impacting traffic. > + */ > + > + old_pvid = br->default_pvid; Remove the empty line. [...] > +int nbp_vlan_init(struct net_bridge_port *p) > +{ > + int rc = 0; > + > + if (p->br->default_pvid) { > + rc = nbp_vlan_add(p, p->br->default_pvid, > + BRIDGE_VLAN_INFO_PVID | > + BRIDGE_VLAN_INFO_UNTAGGED); > + } > + > + return rc; > +} 'rc' can be removed. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 10/03/2014 12:41 AM, Cong Wang wrote: > On Thu, Oct 2, 2014 at 4:54 PM, Vladislav Yasevich <vyasevich@gmail.com> wrote: >> +static void br_vlan_disable_default_pvid(struct net_bridge *br) >> +{ >> + struct net_bridge_port *p; >> + u16 pvid = br->default_pvid; >> + >> + /* Disable default_pvid on all ports where it is still >> + * configured. >> + */ >> + > > This empty line is not necessary. > >> + if (vlan_default_pvid(br_get_vlan_info(br), pvid)) >> + br_vlan_delete(br, pvid); >> + >> + list_for_each_entry(p, &br->port_list, list) { >> + if (vlan_default_pvid(nbp_get_vlan_info(p), pvid)) >> + nbp_vlan_delete(p, pvid); >> + } >> + >> + br->default_pvid = 0; >> +} >> + >> +static int __br_vlan_set_default_pvid(struct net_bridge *br, u16 pvid) >> +{ >> + struct net_bridge_port *p; >> + u16 old_pvid; >> + int err; >> + DECLARE_BITMAP(changed, BR_MAX_PORTS); > > > This bitmap will use 128 bytes on stack, why not using heap? > I suppose I wanted to avoid yet another memory allocation failure condition. Is this really going to cause issues? Thanks -vlad >> + >> + bitmap_zero(changed, BR_MAX_PORTS); >> + >> + /* This function runs with filtering turned off so we can >> + * remove the old pvid configuration and add the new one after >> + * without impacting traffic. >> + */ >> + >> + old_pvid = br->default_pvid; > > > Remove the empty line. > > [...] > >> +int nbp_vlan_init(struct net_bridge_port *p) >> +{ >> + int rc = 0; >> + >> + if (p->br->default_pvid) { >> + rc = nbp_vlan_add(p, p->br->default_pvid, >> + BRIDGE_VLAN_INFO_PVID | >> + BRIDGE_VLAN_INFO_UNTAGGED); >> + } >> + >> + return rc; >> +} > > 'rc' can be removed. > -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Oct 3, 2014 at 6:37 AM, Vlad Yasevich <vyasevic@redhat.com> wrote: > On 10/03/2014 12:41 AM, Cong Wang wrote: >> On Thu, Oct 2, 2014 at 4:54 PM, Vladislav Yasevich <vyasevich@gmail.com> wrote: >>> +static int __br_vlan_set_default_pvid(struct net_bridge *br, u16 pvid) >>> +{ >>> + struct net_bridge_port *p; >>> + u16 old_pvid; >>> + int err; >>> + DECLARE_BITMAP(changed, BR_MAX_PORTS); >> >> >> This bitmap will use 128 bytes on stack, why not using heap? >> > > I suppose I wanted to avoid yet another memory allocation failure condition. > Is this really going to cause issues? > Probably not a big issue, but generally stack is more limited than heap in kernel. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 659cac1..ffd379d 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -88,12 +88,17 @@ out: static int br_dev_init(struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); + int err; br->stats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); if (!br->stats) return -ENOMEM; - return 0; + err = br_vlan_init(br); + if (err) + free_percpu(br->stats); + + return err; } static int br_dev_open(struct net_device *dev) @@ -389,5 +394,4 @@ void br_dev_setup(struct net_device *dev) br_netfilter_rtable_init(br); br_stp_timer_init(br); br_multicast_init(br); - br_vlan_init(br); } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 7b7289c..ed307db 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -500,6 +500,9 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) if (br_fdb_insert(br, p, dev->dev_addr, 0)) netdev_err(dev, "failed insert local address bridge forwarding table\n"); + if (nbp_vlan_init(p)) + netdev_err(dev, "failed to initialize vlan filtering on this port\n"); + spin_lock_bh(&br->lock); changed_addr = br_stp_recalculate_bridge_id(br); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index f671561..d8cbaa6 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -605,12 +605,13 @@ bool br_vlan_find(struct net_bridge *br, u16 vid); void br_recalculate_fwd_mask(struct net_bridge *br); int br_vlan_filter_toggle(struct net_bridge *br, unsigned long val); int br_vlan_set_proto(struct net_bridge *br, unsigned long val); -void br_vlan_init(struct net_bridge *br); +int br_vlan_init(struct net_bridge *br); int br_vlan_set_default_pvid(struct net_bridge *br, unsigned long val); int nbp_vlan_add(struct net_bridge_port *port, u16 vid, u16 flags); int nbp_vlan_delete(struct net_bridge_port *port, u16 vid); void nbp_vlan_flush(struct net_bridge_port *port); bool nbp_vlan_find(struct net_bridge_port *port, u16 vid); +int nbp_vlan_init(struct net_bridge_port *port); static inline struct net_port_vlans *br_get_vlan_info( const struct net_bridge *br) @@ -643,6 +644,9 @@ static inline int br_vlan_get_tag(const struct sk_buff *skb, u16 *vid) static inline u16 br_get_pvid(const struct net_port_vlans *v) { + if (!v) + return 0; + smp_rmb(); return v->pvid; } @@ -703,8 +707,9 @@ static inline void br_recalculate_fwd_mask(struct net_bridge *br) { } -static inline void br_vlan_init(struct net_bridge *br) +static inline int br_vlan_init(struct net_bridge *br) { + return 0; } static inline int nbp_vlan_add(struct net_bridge_port *port, u16 vid, u16 flags) @@ -737,6 +742,11 @@ static inline bool nbp_vlan_find(struct net_bridge_port *port, u16 vid) return false; } +static inline int nbp_vlan_init(struct net_bridge_port *port) +{ + return 0; +} + static inline u16 br_vlan_get_tag(const struct sk_buff *skb, u16 *tag) { return 0; diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 12feb8f..182c3ed 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -499,12 +499,112 @@ err_filt: goto unlock; } +static bool vlan_default_pvid(struct net_port_vlans *pv, u16 vid) +{ + return pv && vid == pv->pvid && test_bit(vid, pv->untagged_bitmap); +} + +static void br_vlan_disable_default_pvid(struct net_bridge *br) +{ + struct net_bridge_port *p; + u16 pvid = br->default_pvid; + + /* Disable default_pvid on all ports where it is still + * configured. + */ + + if (vlan_default_pvid(br_get_vlan_info(br), pvid)) + br_vlan_delete(br, pvid); + + list_for_each_entry(p, &br->port_list, list) { + if (vlan_default_pvid(nbp_get_vlan_info(p), pvid)) + nbp_vlan_delete(p, pvid); + } + + br->default_pvid = 0; +} + +static int __br_vlan_set_default_pvid(struct net_bridge *br, u16 pvid) +{ + struct net_bridge_port *p; + u16 old_pvid; + int err; + DECLARE_BITMAP(changed, BR_MAX_PORTS); + + bitmap_zero(changed, BR_MAX_PORTS); + + /* This function runs with filtering turned off so we can + * remove the old pvid configuration and add the new one after + * without impacting traffic. + */ + + old_pvid = br->default_pvid; + + /* If the user has set a different PVID or if the new default pvid + * conflicts with user configuration, do not modify the configuration. + */ + if ((old_pvid && !vlan_default_pvid(br_get_vlan_info(br), old_pvid)) || + br_vlan_find(br, pvid)) + goto do_ports; + + err = br_vlan_add(br, pvid, + BRIDGE_VLAN_INFO_PVID | BRIDGE_VLAN_INFO_UNTAGGED); + if (err) + goto err; + br_vlan_delete(br, old_pvid); + set_bit(0, changed); + +do_ports: + list_for_each_entry(p, &br->port_list, list) { + /* If the user has set a different PVID or if the new default + * pvid conflicts with user configuration, do not modify the + * configuration. + */ + if ((old_pvid && + !vlan_default_pvid(nbp_get_vlan_info(p), old_pvid)) || + nbp_vlan_find(p, pvid)) + continue; + + err = nbp_vlan_add(p, pvid, + BRIDGE_VLAN_INFO_PVID | + BRIDGE_VLAN_INFO_UNTAGGED); + if (err) + goto err_port; + nbp_vlan_delete(p, old_pvid); + set_bit(p->port_no, changed); + } + + br->default_pvid = pvid; + + return 0; + +err_port: + list_for_each_entry_continue_reverse(p, &br->port_list, list) { + if (!test_bit(p->port_no, changed)) + continue; + + nbp_vlan_add(p, old_pvid, + BRIDGE_VLAN_INFO_PVID | + BRIDGE_VLAN_INFO_UNTAGGED); + nbp_vlan_delete(p, pvid); + } + + if (test_bit(0, changed)) { + br_vlan_add(br, old_pvid, + BRIDGE_VLAN_INFO_PVID | + BRIDGE_VLAN_INFO_UNTAGGED); + br_vlan_delete(br, pvid); + } +err: + return err; +} + int br_vlan_set_default_pvid(struct net_bridge *br, unsigned long val) { u16 pvid = val; int err = 0; - if (!val || val >= VLAN_VID_MASK) + if (val >= VLAN_VID_MASK) return -EINVAL; if (!rtnl_trylock()) @@ -520,17 +620,22 @@ int br_vlan_set_default_pvid(struct net_bridge *br, unsigned long val) goto unlock; } - br->default_pvid = vid; + if (!pvid) + br_vlan_disable_default_pvid(br); + else + err = __br_vlan_set_default_pvid(br, pvid); unlock: rtnl_unlock(); return err; } -void br_vlan_init(struct net_bridge *br) +int br_vlan_init(struct net_bridge *br) { br->vlan_proto = htons(ETH_P_8021Q); br->default_pvid = 1; + return br_vlan_add(br, 1, + BRIDGE_VLAN_INFO_PVID | BRIDGE_VLAN_INFO_UNTAGGED); } /* Must be protected by RTNL. @@ -622,3 +727,16 @@ out: rcu_read_unlock(); return found; } + +int nbp_vlan_init(struct net_bridge_port *p) +{ + int rc = 0; + + if (p->br->default_pvid) { + rc = nbp_vlan_add(p, p->br->default_pvid, + BRIDGE_VLAN_INFO_PVID | + BRIDGE_VLAN_INFO_UNTAGGED); + } + + return rc; +}
Currently when vlan filtering is turned on on the bridge, the bridge will drop all traffic untill the user configures the filter. This isn't very nice for ports that don't care about vlans and just want untagged traffic. A concept of a default_pvid was recently introduced. This patch adds filtering support for default_pvid. Now, ports that don't care about vlans and don't define there own filter will belong to the VLAN of the default_pvid and continue to receive untagged traffic. This filtering can be disabled by setting default_pvid to 0. Signed-off-by: Vladislav Yasevich <vyasevic@redhat.com> --- net/bridge/br_device.c | 8 +++- net/bridge/br_if.c | 3 ++ net/bridge/br_private.h | 14 +++++- net/bridge/br_vlan.c | 124 ++++++++++++++++++++++++++++++++++++++++++++++-- 4 files changed, 142 insertions(+), 7 deletions(-)