2 ===============================================
3 XFRM device - offloading the IPsec computations
4 ===============================================
5 Shannon Nelson <shannon.nelson@oracle.com>
11 IPsec is a useful feature for securing network traffic, but the
12 computational cost is high: a 10Gbps link can easily be brought down
13 to under 1Gbps, depending on the traffic and link configuration.
14 Luckily, there are NICs that offer a hardware based IPsec offload which
15 can radically increase throughput and decrease CPU utilization. The XFRM
16 Device interface allows NIC drivers to offer to the stack access to the
19 Userland access to the offload is typically through a system such as
20 libreswan or KAME/raccoon, but the iproute2 'ip xfrm' command set can
21 be handy when experimenting. An example command might look something
24 ip x s add proto esp dst 14.0.0.70 src 14.0.0.52 spi 0x07 mode transport \
25 reqid 0x07 replay-window 32 \
26 aead 'rfc4106(gcm(aes))' 0x44434241343332312423222114131211f4f3f2f1 128 \
27 sel src 14.0.0.52/24 dst 14.0.0.70/24 proto tcp \
28 offload dev eth4 dir in
30 Yes, that's ugly, but that's what shell scripts and/or libreswan are for.
34 Callbacks to implement
35 ======================
37 /* from include/linux/netdevice.h */
39 int (*xdo_dev_state_add) (struct xfrm_state *x);
40 void (*xdo_dev_state_delete) (struct xfrm_state *x);
41 void (*xdo_dev_state_free) (struct xfrm_state *x);
42 bool (*xdo_dev_offload_ok) (struct sk_buff *skb,
43 struct xfrm_state *x);
44 void (*xdo_dev_state_advance_esn) (struct xfrm_state *x);
47 The NIC driver offering ipsec offload will need to implement these
48 callbacks to make the offload available to the network stack's
49 XFRM subsytem. Additionally, the feature bits NETIF_F_HW_ESP and
50 NETIF_F_HW_ESP_TX_CSUM will signal the availability of the offload.
57 At probe time and before the call to register_netdev(), the driver should
58 set up local data structures and XFRM callbacks, and set the feature bits.
59 The XFRM code's listener will finish the setup on NETDEV_REGISTER.
61 adapter->netdev->xfrmdev_ops = &ixgbe_xfrmdev_ops;
62 adapter->netdev->features |= NETIF_F_HW_ESP;
63 adapter->netdev->hw_enc_features |= NETIF_F_HW_ESP;
65 When new SAs are set up with a request for "offload" feature, the
66 driver's xdo_dev_state_add() will be given the new SA to be offloaded
67 and an indication of whether it is for Rx or Tx. The driver should
68 - verify the algorithm is supported for offloads
69 - store the SA information (key, salt, target-ip, protocol, etc)
70 - enable the HW offload of the SA
72 The driver can also set an offload_handle in the SA, an opaque void pointer
73 that can be used to convey context into the fast-path offload requests.
75 xs->xso.offload_handle = context;
78 When the network stack is preparing an IPsec packet for an SA that has
79 been setup for offload, it first calls into xdo_dev_offload_ok() with
80 the skb and the intended offload state to ask the driver if the offload
81 will serviceable. This can check the packet information to be sure the
82 offload can be supported (e.g. IPv4 or IPv6, no IPv4 options, etc) and
83 return true of false to signify its support.
85 When ready to send, the driver needs to inspect the Tx packet for the
86 offload information, including the opaque context, and set up the packet
89 xs = xfrm_input_state(skb);
90 context = xs->xso.offload_handle;
93 The stack has already inserted the appropriate IPsec headers in the
94 packet data, the offload just needs to do the encryption and fix up the
98 When a packet is received and the HW has indicated that it offloaded a
99 decryption, the driver needs to add a reference to the decoded SA into
100 the packet's skb. At this point the data should be decrypted but the
101 IPsec headers are still in the packet data; they are removed later up
102 the stack in xfrm_input().
104 find and hold the SA that was used to the Rx skb
105 get spi, protocol, and destination IP from packet headers
106 xs = find xs from (spi, protocol, dest_IP)
109 store the state information into the skb
110 skb->sp = secpath_dup(skb->sp);
111 skb->sp->xvec[skb->sp->len++] = xs;
114 indicate the success and/or error status of the offload
115 xo = xfrm_offload(skb);
116 xo->flags = CRYPTO_DONE;
117 xo->status = crypto_status;
119 hand the packet to napi_gro_receive() as usual
121 In ESN mode, xdo_dev_state_advance_esn() is called from xfrm_replay_advance_esn().
122 Driver will check packet seq number and update HW ESN state machine if needed.
124 When the SA is removed by the user, the driver's xdo_dev_state_delete()
125 is asked to disable the offload. Later, xdo_dev_state_free() is called
126 from a garbage collection routine after all reference counts to the state
127 have been removed and any remaining resources can be cleared for the
128 offload state. How these are used by the driver will depend on specific
131 As a netdev is set to DOWN the XFRM stack's netdev listener will call
132 xdo_dev_state_delete() and xdo_dev_state_free() on any remaining offloaded