linux/net/core
Jerry Chu bf5a755f5e net-gre-gro: Add GRE support to the GRO stack
This patch built on top of Commit 299603e837
("net-gro: Prepare GRO stack for the upcoming tunneling support") to add
the support of the standard GRE (RFC1701/RFC2784/RFC2890) to the GRO
stack. It also serves as an example for supporting other encapsulation
protocols in the GRO stack in the future.

The patch supports version 0 and all the flags (key, csum, seq#) but
will flush any pkt with the S (seq#) flag. This is because the S flag
is not support by GSO, and a GRO pkt may end up in the forwarding path,
thus requiring GSO support to break it up correctly.

Currently the "packet_offload" structure only contains L3 (ETH_P_IP/
ETH_P_IPV6) GRO offload support so the encapped pkts are limited to
IP pkts (i.e., w/o L2 hdr). But support for other protocol type can
be easily added, so is the support for GRE variations like NVGRE.

The patch also support csum offload. Specifically if the csum flag is on
and the h/w is capable of checksumming the payload (CHECKSUM_COMPLETE),
the code will take advantage of the csum computed by the h/w when
validating the GRE csum.

Note that commit 60769a5dcd "ipv4: gre:
add GRO capability" already introduces GRO capability to IPv4 GRE
tunnels, using the gro_cells infrastructure. But GRO is done after
GRE hdr has been removed (i.e., decapped). The following patch applies
GRO when pkts first come in (before hitting the GRE tunnel code). There
is some performance advantage for applying GRO as early as possible.
Also this approach is transparent to other subsystem like Open vSwitch
where GRE decap is handled outside of the IP stack hence making it
harder for the gro_cells stuff to apply. On the other hand, some NICs
are still not capable of hashing on the inner hdr of a GRE pkt (RSS).
In that case the GRO processing of pkts from the same remote host will
all happen on the same CPU and the performance may be suboptimal.

I'm including some rough preliminary performance numbers below. Note
that the performance will be highly dependent on traffic load, mix as
usual. Moreover it also depends on NIC offload features hence the
following is by no means a comprehesive study. Local testing and tuning
will be needed to decide the best setting.

All tests spawned 50 copies of netperf TCP_STREAM and ran for 30 secs.
(super_netperf 50 -H 192.168.1.18 -l 30)

An IP GRE tunnel with only the key flag on (e.g., ip tunnel add gre1
mode gre local 10.246.17.18 remote 10.246.17.17 ttl 255 key 123)
is configured.

The GRO support for pkts AFTER decap are controlled through the device
feature of the GRE device (e.g., ethtool -K gre1 gro on/off).

1.1 ethtool -K gre1 gro off; ethtool -K eth0 gro off
thruput: 9.16Gbps
CPU utilization: 19%

1.2 ethtool -K gre1 gro on; ethtool -K eth0 gro off
thruput: 5.9Gbps
CPU utilization: 15%

1.3 ethtool -K gre1 gro off; ethtool -K eth0 gro on
thruput: 9.26Gbps
CPU utilization: 12-13%

1.4 ethtool -K gre1 gro on; ethtool -K eth0 gro on
thruput: 9.26Gbps
CPU utilization: 10%

The following tests were performed on a different NIC that is capable of
csum offload. I.e., the h/w is capable of computing IP payload csum
(CHECKSUM_COMPLETE).

2.1 ethtool -K gre1 gro on (hence will use gro_cells)

2.1.1 ethtool -K eth0 gro off; csum offload disabled
thruput: 8.53Gbps
CPU utilization: 9%

2.1.2 ethtool -K eth0 gro off; csum offload enabled
thruput: 8.97Gbps
CPU utilization: 7-8%

2.1.3 ethtool -K eth0 gro on; csum offload disabled
thruput: 8.83Gbps
CPU utilization: 5-6%

2.1.4 ethtool -K eth0 gro on; csum offload enabled
thruput: 8.98Gbps
CPU utilization: 5%

2.2 ethtool -K gre1 gro off

2.2.1 ethtool -K eth0 gro off; csum offload disabled
thruput: 5.93Gbps
CPU utilization: 9%

2.2.2 ethtool -K eth0 gro off; csum offload enabled
thruput: 5.62Gbps
CPU utilization: 8%

2.2.3 ethtool -K eth0 gro on; csum offload disabled
thruput: 7.69Gbps
CPU utilization: 8%

2.2.4 ethtool -K eth0 gro on; csum offload enabled
thruput: 8.96Gbps
CPU utilization: 5-6%

Signed-off-by: H.K. Jerry Chu <hkchu@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-07 16:21:31 -05:00
..
datagram.c net, datagram: fix the incorrect comment in zerocopy_sg_from_iovec() 2013-10-29 00:19:04 -04:00
dev_addr_lists.c net: remove dead code for add/del multiple 2013-12-17 15:14:04 -05:00
dev_ioctl.c net_tstamp: Add SIOCGHWTSTAMP ioctl to match SIOCSHWTSTAMP 2013-11-19 19:07:21 +00:00
dev.c net-gre-gro: Add GRE support to the GRO stack 2014-01-07 16:21:31 -05:00
drop_monitor.c net: drop_monitor: fix the value of maxattr 2013-12-09 21:10:38 -05:00
dst.c net: pass info struct via netdevice notifier 2013-05-28 13:11:01 -07:00
ethtool.c net: Add layer 2 hardware acceleration operations for macvlan devices 2013-11-07 19:11:41 -05:00
fib_rules.c net: Fix "ip rule delete table 256" 2013-11-08 14:53:10 -05:00
filter.c net: fix unsafe set_memory_rw from softirq 2013-10-07 15:16:45 -04:00
flow_dissector.c net: Change skb_get_rxhash to skb_get_hash 2013-12-17 16:36:21 -05:00
flow.c net: delete __cpuinit usage from all net files 2013-07-14 19:36:58 -04:00
gen_estimator.c net_sched: add 64bit rate estimators 2013-06-11 02:51:03 -07:00
gen_stats.c net_sched: add 64bit rate estimators 2013-06-11 02:51:03 -07:00
iovec.c net: rework recvmsg handler msg_name and msg_namelen logic 2013-11-20 21:52:30 -05:00
link_watch.c net: make all team port device link events urgent 2013-06-13 02:31:41 -07:00
Makefile net: netprio: rename config to be more consistent with cgroup configs 2014-01-03 23:41:42 +01:00
neighbour.c Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2014-01-06 17:37:45 -05:00
net_namespace.c userns: Kill nsown_capable it makes the wrong thing easy 2013-08-30 23:44:11 -07:00
net-procfs.c rps: selective flow shedding during softnet overflow 2013-05-20 13:48:04 -07:00
net-sysfs.c net: netdev_kobject_init: annotate with __init 2014-01-05 20:27:54 -05:00
net-sysfs.h net: netdev_kobject_init: annotate with __init 2014-01-05 20:27:54 -05:00
net-traces.c net: Add export.h for EXPORT_SYMBOL/THIS_MODULE to non-modules 2011-10-31 19:30:30 -04:00
netclassid_cgroup.c net: net_cls: move cgroupfs classid handling into core 2014-01-03 23:41:41 +01:00
netevent.c net: Add export.h for EXPORT_SYMBOL/THIS_MODULE to non-modules 2011-10-31 19:30:30 -04:00
netpoll.c netpoll: Fix missing TXQ unlock and and OOPS. 2014-01-02 19:50:52 -05:00
netprio_cgroup.c net: more spelling fixes 2013-12-10 21:57:11 -05:00
pktgen.c {pktgen, xfrm} Update IPv4 header total len and checksum after tranformation 2013-12-01 20:33:52 -05:00
request_sock.c tcp: fix a panic on UP machines in reqsk_fastopen_remove 2013-01-14 18:10:05 -05:00
rtnetlink.c netlink: cleanup rntl_af_register 2014-01-01 23:42:19 -05:00
scm.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace 2013-09-07 14:35:32 -07:00
secure_seq.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2013-10-23 16:49:34 -04:00
skbuff.c Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jesse/openvswitch 2014-01-06 19:48:38 -05:00
sock_diag.c sock_diag: fix filter code sent to userspace 2013-06-10 22:23:32 -07:00
sock.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next 2014-01-05 20:18:50 -05:00
stream.c net: add sk_stream_is_writeable() helper 2013-07-24 17:54:48 -07:00
sysctl_net_core.c rps: NUMA flow limit allocations 2013-12-19 19:00:07 -05:00
timestamping.c net: Add export.h for EXPORT_SYMBOL/THIS_MODULE to non-modules 2011-10-31 19:30:30 -04:00
user_dma.c net: Add export.h for EXPORT_SYMBOL/THIS_MODULE to non-modules 2011-10-31 19:30:30 -04:00
utils.c net: make net_get_random_once irq safe 2013-10-25 19:03:39 -04:00