diff options
| author |   Neale Ranns <nranns@cisco.com> | 2019-04-02 08:13:33 +0000 |
|---|---|---|
| committer |   Damjan Marion <dmarion@me.com> | 2019-12-17 17:57:24 +0000 |
| commit | f62a8c013c6e22c012b9d7df2ef463a6370cf1ce (patch) | |
| tree | 0448d4b53340b84e1a8aca87ec57133c503026da /src/vnet/ipsec/ipsec_format.c | |
| parent | f2bde7ac51123a0a46334b4ec55e2aceae031db7 (diff) | |
ipsec: bind an SA to a worker
the sequence number increment and the anti-replay window
checks must be atomic. Given the vector nature of VPP we
can't simply use atomic increments for sequence numbers,
since a vector on thread 1 with lower sequence numbers could
be 'overtaken' by packets on thread 2 with higher sequence
numbers.
The anti-replay logic requires a critical section, not just
atomics, and we don't want that.
So when the SA see the first packet it is bound to that worker
all subsequent packets, that arrive on a different worker,
are subject to a handoff.
Type: feature
Change-Id: Ia20a8645fb50622ea6235ab015a537f033d531a4
Signed-off-by: Neale Ranns <nranns@cisco.com>
Diffstat (limited to 'src/vnet/ipsec/ipsec_format.c')
| -rw-r--r-- | src/vnet/ipsec/ipsec_format.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/src/vnet/ipsec/ipsec_format.c b/src/vnet/ipsec/ipsec_format.c index e6e828931dd..8b9d8c40be1 100644 --- a/src/vnet/ipsec/ipsec_format.c +++ b/src/vnet/ipsec/ipsec_format.c @@ -292,6 +292,8 @@ format_ipsec_sa (u8 * s, va_list * args) s = format (s, "\n locks %d", sa->node.fn_locks); s = format (s, "\n salt 0x%x", clib_net_to_host_u32 (sa->salt)); + s = format (s, "\n thread-indices [encrypt:%d decrypt:%d]", + sa->encrypt_thread_index, sa->decrypt_thread_index); s = format (s, "\n seq %u seq-hi %u", sa->seq, sa->seq_hi); s = format (s, "\n last-seq %u last-seq-hi %u window %U", sa->last_seq, sa->last_seq_hi, |