diff options
| author | Eric Dumazet <edumazet@google.com> | 2015-02-06 12:59:01 -0800 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2015-02-08 16:53:57 -0800 |
| commit | 567e4b79731c352a17d73c483959f795d3593e03 (patch) | |
| tree | 4af65c205a8b65cfc5fd7b42e7b8750728230616 /drivers/net/tun.c | |
| parent | 096a4cfa5807aa89c78ce12309c0b1c10cf88184 (diff) | |
net: rfs: add hash collision detection
Receive Flow Steering is a nice solution but suffers from
hash collisions when a mix of connected and unconnected traffic
is received on the host, when flow hash table is populated.
Also, clearing flow in inet_release() makes RFS not very good
for short lived flows, as many packets can follow close().
(FIN , ACK packets, ...)
This patch extends the information stored into global hash table
to not only include cpu number, but upper part of the hash value.
I use a 32bit value, and dynamically split it in two parts.
For host with less than 64 possible cpus, this gives 6 bits for the
cpu number, and 26 (32-6) bits for the upper part of the hash.
Since hash bucket selection use low order bits of the hash, we have
a full hash match, if /proc/sys/net/core/rps_sock_flow_entries is big
enough.
If the hash found in flow table does not match, we fallback to RPS (if
it is enabled for the rxqueue).
This means that a packet for an non connected flow can avoid the
IPI through a unrelated/victim CPU.
This also means we no longer have to clear the table at socket
close time, and this helps short lived flows performance.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'drivers/net/tun.c')
| -rw-r--r-- | drivers/net/tun.c | 5 |
1 files changed, 1 insertions, 4 deletions
diff --git a/drivers/net/tun.c b/drivers/net/tun.c index ad7d3d5f3ee5..857dca47bf80 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -256,7 +256,6 @@ static void tun_flow_delete(struct tun_struct *tun, struct tun_flow_entry *e) { tun_debug(KERN_INFO, tun, "delete flow: hash %u index %u\n", e->rxhash, e->queue_index); - sock_rps_reset_flow_hash(e->rps_rxhash); hlist_del_rcu(&e->hash_link); kfree_rcu(e, rcu); --tun->flow_count; @@ -373,10 +372,8 @@ unlock: */ static inline void tun_flow_save_rps_rxhash(struct tun_flow_entry *e, u32 hash) { - if (unlikely(e->rps_rxhash != hash)) { - sock_rps_reset_flow_hash(e->rps_rxhash); + if (unlikely(e->rps_rxhash != hash)) e->rps_rxhash = hash; - } } /* We try to identify a flow through its rxhash first. The reason that |