diff options
| author | Eric Dumazet <edumazet@google.com> | 2013-01-10 12:36:42 +0000 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2013-01-10 14:58:13 -0800 |
| commit | 1def9238d4aa2146924994aa4b7dc861f03b9362 (patch) | |
| tree | 8b529029fae8f8a4575b088dbe76e8b84a1c9ccd | |
| parent | 3d55b323709661df34d93e4cdcc5337620e34dfd (diff) | |
| download | blackbird-op-linux-1def9238d4aa2146924994aa4b7dc861f03b9362.tar.gz blackbird-op-linux-1def9238d4aa2146924994aa4b7dc861f03b9362.zip | |
net_sched: more precise pkt_len computation
One long standing problem with TSO/GSO/GRO packets is that skb->len
doesn't represent a precise amount of bytes on wire.
Headers are only accounted for the first segment.
For TCP, thats typically 66 bytes per 1448 bytes segment missing,
an error of 4.5 % for normal MSS value.
As consequences :
1) TBF/CBQ/HTB/NETEM/... can send more bytes than the assigned limits.
2) Device stats are slightly under estimated as well.
Fix this by taking account of headers in qdisc_skb_cb(skb)->pkt_len
computation.
Packet schedulers should use qdisc pkt_len instead of skb->len for their
bandwidth limitations, and TSO enabled devices drivers could use pkt_len
if their statistics are not hardware assisted, and if they don't scratch
skb->cb[] first word.
Both egress and ingress paths work, thanks to commit fda55eca5a
(net: introduce skb_transport_header_was_set()) : If GRO built
a GSO packet, it also set the transport header for us.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Paolo Valente <paolo.valente@unimore.it>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
| -rw-r--r-- | net/core/dev.c | 22 |
1 files changed, 21 insertions, 1 deletions
diff --git a/net/core/dev.c b/net/core/dev.c index 594830e83589..4794cae84939 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2532,6 +2532,26 @@ struct netdev_queue *netdev_pick_tx(struct net_device *dev, return netdev_get_tx_queue(dev, queue_index); } +static void qdisc_pkt_len_init(struct sk_buff *skb) +{ + const struct skb_shared_info *shinfo = skb_shinfo(skb); + + qdisc_skb_cb(skb)->pkt_len = skb->len; + + /* To get more precise estimation of bytes sent on wire, + * we add to pkt_len the headers size of all segments + */ + if (shinfo->gso_size) { + unsigned int hdr_len = skb_transport_offset(skb); + + if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) + hdr_len += tcp_hdrlen(skb); + else + hdr_len += sizeof(struct udphdr); + qdisc_skb_cb(skb)->pkt_len += (shinfo->gso_segs - 1) * hdr_len; + } +} + static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, struct net_device *dev, struct netdev_queue *txq) @@ -2540,7 +2560,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, bool contended; int rc; - qdisc_skb_cb(skb)->pkt_len = skb->len; + qdisc_pkt_len_init(skb); qdisc_calculate_pkt_len(skb, q); /* * Heuristic to force contended enqueues to serialize on a |
