user/sven/linux.git/net/switchdev, branch v4.9.69

switchdev: Execute bridge ndos only for bridge ports

2016-10-19T14:58:04Z

We recently got the following warning after setting up a vlan device on top of an offloaded bridge and executing 'bridge link': WARNING: CPU: 0 PID: 18566 at drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c:81 mlxsw_sp_port_orig_get.part.9+0x55/0x70 [mlxsw_spectrum] [...] CPU: 0 PID: 18566 Comm: bridge Not tainted 4.8.0-rc7 #1 Hardware name: Mellanox Technologies Ltd. Mellanox switch/Mellanox switch, BIOS 4.6.5 05/21/2015 0000000000000286 00000000e64ab94f ffff880406e6f8f0 ffffffff8135eaa3 0000000000000000 0000000000000000 ffff880406e6f930 ffffffff8108c43b 0000005106e6f988 ffff8803df398840 ffff880403c60108 ffff880406e6f990 Call Trace: [] dump_stack+0x63/0x90 [] __warn+0xcb/0xf0 [] warn_slowpath_null+0x1d/0x20 [] mlxsw_sp_port_orig_get.part.9+0x55/0x70 [mlxsw_spectrum] [] mlxsw_sp_port_attr_get+0xa5/0xb0 [mlxsw_spectrum] [] switchdev_port_attr_get+0x4f/0x140 [] switchdev_port_attr_get+0x100/0x140 [] switchdev_port_attr_get+0x100/0x140 [] switchdev_port_bridge_getlink+0x5b/0xc0 [] ? switchdev_port_fdb_dump+0x90/0x90 [] rtnl_bridge_getlink+0xe7/0x190 [] netlink_dump+0x122/0x290 [] __netlink_dump_start+0x15f/0x190 [] ? rtnl_bridge_dellink+0x230/0x230 [] rtnetlink_rcv_msg+0x1a6/0x220 [] ? __kmalloc_node_track_caller+0x208/0x2c0 [] ? rtnl_bridge_dellink+0x230/0x230 [] ? rtnl_newlink+0x890/0x890 [] netlink_rcv_skb+0xa4/0xc0 [] rtnetlink_rcv+0x28/0x30 [] netlink_unicast+0x18c/0x240 [] netlink_sendmsg+0x2fb/0x3a0 [] sock_sendmsg+0x38/0x50 [] SYSC_sendto+0x101/0x190 [] ? __sys_recvmsg+0x51/0x90 [] SyS_sendto+0xe/0x10 [] entry_SYSCALL_64_fastpath+0x1a/0xa4 The problem is that the 8021q module propagates the call to ndo_bridge_getlink() via switchdev ops, but the switch driver doesn't recognize the netdev, as it's not offloaded. While we can ignore calls being made to non-bridge ports inside the driver, a better fix would be to push this check up to the switchdev layer. Note that these ndos can be called for non-bridged netdev, but this only happens in certain PF drivers which don't call the corresponding switchdev functions anyway. Fixes: 99f44bb3527b ("mlxsw: spectrum: Enable L3 interfaces on top of bridge devices") Signed-off-by: Ido Schimmel Reported-by: Tamir Winetroub Tested-by: Tamir Winetroub Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller

switchdev: remove FIB offload infrastructure

2016-09-28T08:48:00Z

Since this is now taken care of by FIB notifier, remove the code, with all unused dependencies. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller

fib: introduce FIB info offload flag helpers

2016-09-28T08:48:00Z

These helpers are to be used in case someone offloads the FIB entry. The result is that if the entry is offloaded to at least one device, the offload flag is set. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller

rtnetlink: fdb dump: optimize by saving last interface markers

2016-09-01T23:56:15Z

fdb dumps spanning multiple skb's currently restart from the first interface again for every skb. This results in unnecessary iterations on the already visited interfaces and their fdb entries. In large scale setups, we have seen this to slow down fdb dumps considerably. On a system with 30k macs we see fdb dumps spanning across more than 300 skbs. To fix the problem, this patch replaces the existing single fdb marker with three markers: netdev hash entries, netdevs and fdb index to continue where we left off instead of restarting from the first netdev. This is consistent with link dumps. In the process of fixing the performance issue, this patch also re-implements fix done by commit 472681d57a5d ("net: ndo_fdb_dump should report -EMSGSIZE to rtnl_fdb_dump") (with an internal fix from Wilson Kok) in the following ways: - change ndo_fdb_dump handlers to return error code instead of the last fdb index - use cb->args strictly for dump frag markers and not error codes. This is consistent with other dump functions. Below results were taken on a system with 1000 netdevs and 35085 fdb entries: before patch: $time bridge fdb show | wc -l 15065 real 1m11.791s user 0m0.070s sys 1m8.395s (existing code does not return all macs) after patch: $time bridge fdb show | wc -l 35085 real 0m2.017s user 0m0.113s sys 0m1.942s Signed-off-by: Roopa Prabhu Signed-off-by: Wilson Kok Signed-off-by: David S. Miller

bridge: switchdev: Add forward mark support for stacked devices

2016-08-26T20:13:36Z

switchdev_port_fwd_mark_set() is used to set the 'offload_fwd_mark' of port netdevs so that packets being flooded by the device won't be flooded twice. It works by assigning a unique identifier (the ifindex of the first bridge port) to bridge ports sharing the same parent ID. This prevents packets from being flooded twice by the same switch, but will flood packets through bridge ports belonging to a different switch. This method is problematic when stacked devices are taken into account, such as VLANs. In such cases, a physical port netdev can have upper devices being members in two different bridges, thus requiring two different 'offload_fwd_mark's to be configured on the port netdev, which is impossible. The main problem is that packet and netdev marking is performed at the physical netdev level, whereas flooding occurs between bridge ports, which are not necessarily port netdevs. Instead, packet and netdev marking should really be done in the bridge driver with the switch driver only telling it which packets it already forwarded. The bridge driver will mark such packets using the mark assigned to the ingress bridge port and will prevent the packet from being forwarded through any bridge port sharing the same mark (i.e. having the same parent ID). Remove the current switchdev 'offload_fwd_mark' implementation and instead implement the proposed method. In addition, make rocker - the sole user of the mark - use the proposed method. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller

switchdev: Support parent ID comparison for stacked devices

2016-08-26T20:13:36Z

switchdev_port_same_parent_id() currently expects port netdevs, but we need it to support stacked devices in the next patch, so drop the NO_RECURSE flag. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller

switchdev: Put export declaration in the right place

2016-08-15T20:37:50Z

Move exporting of switchdev_port_same_parent_id to be right below it and not elsewhere. Signed-off-by: Or Gerlitz Reported-by: Ido Schimmel Signed-off-by: David S. Miller

net/switchdev: Export the same parent ID service function

2016-07-14T20:34:29Z

This helper serves to know if two switchdev port netdevices belong to the same HW ASIC, e.g to figure out if forwarding offload is possible between them. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller

switchdev: pass pointer to fib_info instead of copy

2016-05-17T17:58:49Z

The problem is that fib_info->nh is [0] so the struct fib_info allocation size depends on number of nexthops. If we just copy fib_info, we do not copy the nexthops info and driver accesses memory which is not ours. Given the fact that fib4 does not defer operations and therefore it does not need copy, just pass the pointer down to drivers as it was done before. Fixes: 850d0cbc91 ("switchdev: remove pointers from switchdev objects") Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller

net: switchdev: Drop EXPERIMENTAL from description

2016-05-14T21:06:37Z

Switchdev has been around for quite a while now, putting "EXPERIMENTAL" in the description is no longer accurate, drop it. Signed-off-by: Florian Fainelli Acked-by: Jiri Pirko Signed-off-by: David S. Miller