diff options
| author | Jiri Pirko <jpirko@redhat.com> | 2011-11-11 17:16:48 -0500 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2011-11-13 16:10:10 -0500 |
| commit | 3d249d4ca7d0ed6629a135ea1ea21c72286c0d80 (patch) | |
| tree | e89c8b6bbf4f8d0e4a3428304c26305eca76e4b8 | |
| parent | 5d70b88cd41ef0f2ac0caaab4fd492dd686feee6 (diff) | |
net: introduce ethernet teaming device
This patch introduces new network device called team. It supposes to be
very fast, simple, userspace-driven alternative to existing bonding
driver.
Userspace library called libteam with couple of demo apps is available
here:
https://github.com/jpirko/libteam
Note it's still in its dipers atm.
team<->libteam use generic netlink for communication. That and rtnl
suppose to be the only way to configure team device, no sysfs etc.
Python binding of libteam was recently introduced.
Daemon providing arpmon/miimon active-backup functionality will be
introduced shortly. All what's necessary is already implemented in
kernel team driver.
v7->v8:
- check ndo_ndo_vlan_rx_[add/kill]_vid functions before calling
them.
- use dev_kfree_skb_any() instead of dev_kfree_skb()
v6->v7:
- transmit and receive functions are not checked in hot paths.
That also resolves memory leak on transmit when no port is
present
v5->v6:
- changed couple of _rcu calls to non _rcu ones in non-readers
v4->v5:
- team_change_mtu() uses team->lock while travesing though port
list
- mac address changes are moved completely to jurisdiction of
userspace daemon. This way the daemon can do FOM1, FOM2 and
possibly other weird things with mac addresses.
Only round-robin mode sets up all ports to bond's address then
enslaved.
- Extended Kconfig text
v3->v4:
- remove redundant synchronize_rcu from __team_change_mode()
- revert "set and clear of mode_ops happens per pointer, not per
byte"
- extend comment of function __team_change_mode()
v2->v3:
- team_change_mtu() uses rcu version of list traversal to unwind
- set and clear of mode_ops happens per pointer, not per byte
- port hashlist changed to be embedded into team structure
- error branch in team_port_enter() does cleanup now
- fixed rtln->rtnl
v1->v2:
- modes are made as modules. Makes team more modular and
extendable.
- several commenters' nitpicks found on v1 were fixed
- several other bugs were fixed.
- note I ignored Eric's comment about roundrobin port selector
as Eric's way may be easily implemented as another mode (mode
"random") in future.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
| -rw-r--r-- | Documentation/networking/team.txt | 2 | ||||
| -rw-r--r-- | MAINTAINERS | 7 | ||||
| -rw-r--r-- | drivers/net/Kconfig | 2 | ||||
| -rw-r--r-- | drivers/net/Makefile | 1 | ||||
| -rw-r--r-- | drivers/net/team/Kconfig | 43 | ||||
| -rw-r--r-- | drivers/net/team/Makefile | 7 | ||||
| -rw-r--r-- | drivers/net/team/team.c | 1583 | ||||
| -rw-r--r-- | drivers/net/team/team_mode_activebackup.c | 137 | ||||
| -rw-r--r-- | drivers/net/team/team_mode_roundrobin.c | 107 | ||||
| -rw-r--r-- | include/linux/Kbuild | 1 | ||||
| -rw-r--r-- | include/linux/if.h | 1 | ||||
| -rw-r--r-- | include/linux/if_team.h | 242 |
12 files changed, 2133 insertions, 0 deletions
diff --git a/Documentation/networking/team.txt b/Documentation/networking/team.txt new file mode 100644 index 000000000000..5a013686b9ea --- /dev/null +++ b/Documentation/networking/team.txt | |||
| @@ -0,0 +1,2 @@ | |||
| 1 | Team devices are driven from userspace via libteam library which is here: | ||
| 2 | https://github.com/jpirko/libteam | ||
diff --git a/MAINTAINERS b/MAINTAINERS index 4808256446f2..8d941692c394 100644 --- a/MAINTAINERS +++ b/MAINTAINERS | |||
| @@ -6484,6 +6484,13 @@ W: http://tcp-lp-mod.sourceforge.net/ | |||
| 6484 | S: Maintained | 6484 | S: Maintained |
| 6485 | F: net/ipv4/tcp_lp.c | 6485 | F: net/ipv4/tcp_lp.c |
| 6486 | 6486 | ||
| 6487 | TEAM DRIVER | ||
| 6488 | M: Jiri Pirko <jpirko@redhat.com> | ||
| 6489 | L: netdev@vger.kernel.org | ||
| 6490 | S: Supported | ||
| 6491 | F: drivers/net/team/ | ||
| 6492 | F: include/linux/if_team.h | ||
| 6493 | |||
| 6487 | TEGRA SUPPORT | 6494 | TEGRA SUPPORT |
| 6488 | M: Colin Cross <ccross@android.com> | 6495 | M: Colin Cross <ccross@android.com> |
| 6489 | M: Olof Johansson <olof@lixom.net> | 6496 | M: Olof Johansson <olof@lixom.net> |
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 583f66cd5bbd..b3020bea39e4 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig | |||
| @@ -125,6 +125,8 @@ config IFB | |||
| 125 | 'ifb1' etc. | 125 | 'ifb1' etc. |
| 126 | Look at the iproute2 documentation directory for usage etc | 126 | Look at the iproute2 documentation directory for usage etc |
| 127 | 127 | ||
| 128 | source "drivers/net/team/Kconfig" | ||
| 129 | |||
| 128 | config MACVLAN | 130 | config MACVLAN |
| 129 | tristate "MAC-VLAN support (EXPERIMENTAL)" | 131 | tristate "MAC-VLAN support (EXPERIMENTAL)" |
| 130 | depends on EXPERIMENTAL | 132 | depends on EXPERIMENTAL |
diff --git a/drivers/net/Makefile b/drivers/net/Makefile index fa877cd2b139..4e4ebfe1aa53 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile | |||
| @@ -17,6 +17,7 @@ obj-$(CONFIG_NET) += Space.o loopback.o | |||
| 17 | obj-$(CONFIG_NETCONSOLE) += netconsole.o | 17 | obj-$(CONFIG_NETCONSOLE) += netconsole.o |
| 18 | obj-$(CONFIG_PHYLIB) += phy/ | 18 | obj-$(CONFIG_PHYLIB) += phy/ |
| 19 | obj-$(CONFIG_RIONET) += rionet.o | 19 | obj-$(CONFIG_RIONET) += rionet.o |
| 20 | obj-$(CONFIG_NET_TEAM) += team/ | ||
| 20 | obj-$(CONFIG_TUN) += tun.o | 21 | obj-$(CONFIG_TUN) += tun.o |
| 21 | obj-$(CONFIG_VETH) += veth.o | 22 | obj-$(CONFIG_VETH) += veth.o |
| 22 | obj-$(CONFIG_VIRTIO_NET) += virtio_net.o | 23 | obj-$(CONFIG_VIRTIO_NET) += virtio_net.o |
diff --git a/drivers/net/team/Kconfig b/drivers/net/team/Kconfig new file mode 100644 index 000000000000..248a144033ca --- /dev/null +++ b/drivers/net/team/Kconfig | |||
| @@ -0,0 +1,43 @@ | |||
| 1 | menuconfig NET_TEAM | ||
| 2 | tristate "Ethernet team driver support (EXPERIMENTAL)" | ||
| 3 | depends on EXPERIMENTAL | ||
| 4 | ---help--- | ||
| 5 | This allows one to create virtual interfaces that teams together | ||
| 6 | multiple ethernet devices. | ||
| 7 | |||
| 8 | Team devices can be added using the "ip" command from the | ||
| 9 | iproute2 package: | ||
| 10 | |||
| 11 | "ip link add link [ address MAC ] [ NAME ] type team" | ||
| 12 | |||
| 13 | To compile this driver as a module, choose M here: the module | ||
| 14 | will be called team. | ||
| 15 | |||
| 16 | if NET_TEAM | ||
| 17 | |||
| 18 | config NET_TEAM_MODE_ROUNDROBIN | ||
| 19 | tristate "Round-robin mode support" | ||
| 20 | depends on NET_TEAM | ||
| 21 | ---help--- | ||
| 22 | Basic mode where port used for transmitting packets is selected in | ||
| 23 | round-robin fashion using packet counter. | ||
| 24 | |||
| 25 | All added ports are setup to have bond's mac address. | ||
| 26 | |||
| 27 | To compile this team mode as a module, choose M here: the module | ||
| 28 | will be called team_mode_roundrobin. | ||
| 29 | |||
| 30 | config NET_TEAM_MODE_ACTIVEBACKUP | ||
| 31 | tristate "Active-backup mode support" | ||
| 32 | depends on NET_TEAM | ||
| 33 | ---help--- | ||
| 34 | Only one port is active at a time and the rest of ports are used | ||
| 35 | for backup. | ||
| 36 | |||
| 37 | Mac addresses of ports are not modified. Userspace is responsible | ||
| 38 | to do so. | ||
| 39 | |||
| 40 | To compile this team mode as a module, choose M here: the module | ||
| 41 | will be called team_mode_activebackup. | ||
| 42 | |||
| 43 | endif # NET_TEAM | ||
diff --git a/drivers/net/team/Makefile b/drivers/net/team/Makefile new file mode 100644 index 000000000000..85f2028a87af --- /dev/null +++ b/drivers/net/team/Makefile | |||
| @@ -0,0 +1,7 @@ | |||
| 1 | # | ||
| 2 | # Makefile for the network team driver | ||
| 3 | # | ||
| 4 | |||
| 5 | obj-$(CONFIG_NET_TEAM) += team.o | ||
| 6 | obj-$(CONFIG_NET_TEAM_MODE_ROUNDROBIN) += team_mode_roundrobin.o | ||
| 7 | obj-$(CONFIG_NET_TEAM_MODE_ACTIVEBACKUP) += team_mode_activebackup.o | ||
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c new file mode 100644 index 000000000000..60672bb09960 --- /dev/null +++ b/drivers/net/team/team.c | |||
| @@ -0,0 +1,1583 @@ | |||
| 1 | /* | ||
| 2 | * net/drivers/team/team.c - Network team device driver | ||
| 3 | * Copyright (c) 2011 Jiri Pirko <jpirko@redhat.com> | ||
| 4 | * | ||
| 5 | * This program is free software; you can redistribute it and/or modify | ||
| 6 | * it under the terms of the GNU General Public License as published by | ||
| 7 | * the Free Software Foundation; either version 2 of the License, or | ||
| 8 | * (at your option) any later version. | ||
| 9 | */ | ||
| 10 | |||
| 11 | #include <linux/kernel.h> | ||
| 12 | #include <linux/types.h> | ||
| 13 | #include <linux/module.h> | ||
| 14 | #include <linux/init.h> | ||
| 15 | #include <linux/slab.h> | ||
| 16 | #include <linux/rcupdate.h> | ||
| 17 | #include <linux/errno.h> | ||
| 18 | #include <linux/ctype.h> | ||
| 19 | #include <linux/notifier.h> | ||
| 20 | #include <linux/netdevice.h> | ||
| 21 | #include <linux/if_arp.h> | ||
| 22 | #include <linux/socket.h> | ||
| 23 | #include <linux/etherdevice.h> | ||
| 24 | #include <linux/rtnetlink.h> | ||
| 25 | #include <net/rtnetlink.h> | ||
| 26 | #include <net/genetlink.h> | ||
| 27 | #include <net/netlink.h> | ||
| 28 | #include <linux/if_team.h> | ||
| 29 | |||
| 30 | #define DRV_NAME "team" | ||
| 31 | |||
| 32 | |||
| 33 | /********** | ||
| 34 | * Helpers | ||
| 35 | **********/ | ||
| 36 | |||
| 37 | #define team_port_exists(dev) (dev->priv_flags & IFF_TEAM_PORT) | ||
| 38 | |||
| 39 | static struct team_port *team_port_get_rcu(const struct net_device *dev) | ||
| 40 | { | ||
| 41 | struct team_port *port = rcu_dereference(dev->rx_handler_data); | ||
| 42 | |||
| 43 | return team_port_exists(dev) ? port : NULL; | ||
| 44 | } | ||
| 45 | |||
| 46 | static struct team_port *team_port_get_rtnl(const struct net_device *dev) | ||
| 47 | { | ||
| 48 | struct team_port *port = rtnl_dereference(dev->rx_handler_data); | ||
| 49 | |||
| 50 | return team_port_exists(dev) ? port : NULL; | ||
| 51 | } | ||
| 52 | |||
| 53 | /* | ||
| 54 | * Since the ability to change mac address for open port device is tested in | ||
| 55 | * team_port_add, this function can be called without control of return value | ||
| 56 | */ | ||
| 57 | static int __set_port_mac(struct net_device *port_dev, | ||
| 58 | const unsigned char *dev_addr) | ||
| 59 | { | ||
| 60 | struct sockaddr addr; | ||
| 61 | |||
| 62 | memcpy(addr.sa_data, dev_addr, ETH_ALEN); | ||
| 63 | addr.sa_family = ARPHRD_ETHER; | ||
| 64 | return dev_set_mac_address(port_dev, &addr); | ||
| 65 | } | ||
| 66 | |||
| 67 | int team_port_set_orig_mac(struct team_port *port) | ||
| 68 | { | ||
| 69 | return __set_port_mac(port->dev, port->orig.dev_addr); | ||
| 70 | } | ||
| 71 | |||
| 72 | int team_port_set_team_mac(struct team_port *port) | ||
| 73 | { | ||
| 74 | return __set_port_mac(port->dev, port->team->dev->dev_addr); | ||
| 75 | } | ||
| 76 | EXPORT_SYMBOL(team_port_set_team_mac); | ||
| 77 | |||
| 78 | |||
| 79 | /******************* | ||
| 80 | * Options handling | ||
| 81 | *******************/ | ||
| 82 | |||
| 83 | void team_options_register(struct team *team, struct team_option *option, | ||
| 84 | size_t option_count) | ||
| 85 | { | ||
| 86 | int i; | ||
| 87 | |||
| 88 | for (i = 0; i < option_count; i++, option++) | ||
| 89 | list_add_tail(&option->list, &team->option_list); | ||
| 90 | } | ||
