Make sure the sw_flow_key structure and valid mask boundaries are always
machine word aligned. Optimize the flow compare and mask operations
using machine word size operations. This patch improves throughput on
average by 15% when CPU is the bottleneck of forwarding packets.

This patch is inspired by ideas and code from a patch submitted by Peter
Klausler titled "replace memcmp() with specialized comparator".
However, The original patch only optimizes for architectures
support unaligned machine word access. This patch optimizes for all
architectures.
Signed-off-by: NAndy Zhou <azhou@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

Key_end is a better name describing the ending boundary than key_len.
Rename those variables to make it less confusing.
Signed-off-by: NAndy Zhou <azhou@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

This patch adds support for rewriting SCTP src,dst ports similar to the
functionality already available for TCP/UDP.

Rewriting SCTP ports is expensive due to double-recalculation of the
SCTP checksums; this is performed to ensure that packets traversing OVS
with invalid checksums will continue to the destination with any
checksum corruption intact.
Reviewed-by: NSimon Horman <horms@verge.net.au>
Signed-off-by: NJoe Stringer <joe@wand.net.nz>
Signed-off-by: NBen Pfaff <blp@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

Add wildcarded flow support in kernel datapath.

Wildcarded flow can improve OVS flow set up performance by avoid sending
matching new flows to the user space program. The exact performance boost
will largely dependent on wildcarded flow hit rate.

In case all new flows hits wildcard flows, the flow set up rate is
within 5% of that of linux bridge module.

Pravin has made significant contributions to this patch. Including API
clean ups and bug fixes.
Signed-off-by: NPravin B Shelar <pshelar@nicira.com>
Signed-off-by: NAndy Zhou <azhou@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

Cc: Jesse Gross <jesse@nicira.com>
Cc: Pravin B Shelar <pshelar@nicira.com>
Signed-off-by: NCong Wang <amwang@redhat.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

Signed-off-by: NJustin Pettit <jpettit@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

Link upper device properly. That will make IFLA_MASTER filled up.
Set the master to port 0 of the datapath under which the port belongs.
Signed-off-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NJesse Gross <jesse@nicira.com>

Flow table destroy is done in rcu call-back context.  Therefore
there is no need to use rcu variant of hlist_del().
Signed-off-by: NPravin B Shelar <pshelar@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

RCUfy dp-dump operation which is already read-only. This
makes all ovs dump operations lockless.
Signed-off-by: NPravin B Shelar <pshelar@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

Flow dump operation is read-only operation.  There is no need to
take ovs-lock.  Following patch use rcu-lock for dumping flows.
Signed-off-by: NPravin B Shelar <pshelar@nicira.com>
Signed-off-by: NJesse Gross <jesse@nicira.com>

We can simply use the %pISc format specifier that was recently added
and thus remove some code that distinguishes between IPv4 and IPv6.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The tcp_probe currently only supports analysis of IPv4 connections.
Therefore, it would be nice to have IPv6 supported as well. Since we
have the recently added %pISpc specifier that is IPv4/IPv6 generic,
build related sockaddress structures from the flow information and
pass this to our format string. Tested with SSH and HTTP sessions
on IPv4 and IPv6.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patches fixes a rather unproblematic function signature mismatch
as the const specifier was missing for the th variable; and next to
that it adds a build-time assertion so that future function signature
mismatches for kprobes will not end badly, similarly as commit 22222997
("net: sctp: add build check for sctp_sf_eat_sack_6_2/jsctp_sf_eat_sack")
did it for SCTP.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It is helpful to sometimes know the TCP window sizes of an established
socket e.g. to confirm that window scaling is working or to tweak the
window size to improve high-latency connections, etc etc. Currently the
TCP snooper only exports the send window size, but not the receive window
size. Therefore, also add the receive window size to the end of the
output line.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The stack currently detects reordering and avoid spurious
retransmission very well. However the throughput is sub-optimal under
high reordering because cwnd is increased only if the data is deliverd
in order. I.e., FLAG_DATA_ACKED check in tcp_ack().  The more packet
are reordered the worse the throughput is.

Therefore when reordering is proven high, cwnd should advance whenever
the data is delivered regardless of its ordering. If reordering is low,
conservatively advance cwnd only on ordered deliveries in Open state,
and retain cwnd in Disordered state (RFC5681).

Using netperf on a qdisc setup of 20Mbps BW and random RTT from 45ms
to 55ms (for reordering effect). This change increases TCP throughput
by 20 - 25% to near bottleneck BW.

A special case is the stretched ACK with new SACK and/or ECE mark.
For example, a receiver may receive an out of order or ECN packet with
unacked data buffered because of LRO or delayed ACK. The principle on
such an ACK is to advance cwnd on the cummulative acked part first,
then reduce cwnd in tcp_fastretrans_alert().
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Instead of hard-coding length values, use a define to make it clear
where those lengths come from.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the functions mld_gq_start_timer(), mld_ifc_start_timer(),
and mld_dad_start_timer(), rather use unsigned long than int
as we operate only on unsigned values anyway. This seems more
appropriate as there is no good reason to do type conversions
to int, that could lead to future errors.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use proper API functions to calculate jiffies from milliseconds and
not the crude method of dividing HZ by a value. This ensures more
accurate values even in the case of strange HZ values. While at it,
also simplify code in the mlh2 case by using max().
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When an Xin6 tunnel is set up, we check other netdevices to inherit the link-
local address. If none is available, the interface will not have any link-local
address. RFC4862 expects that each interface has a link local address.

Now than this kind of tunnels supports x-netns, it's easy to fall in this case
(by creating the tunnel in a netns where ethernet interfaces stand and then
moving it to a other netns where no ethernet interface is available).

RFC4291, Appendix A suggests two methods: the first is the one currently
implemented, the second is to generate a unique identifier, so that we can
always generate the link-local address. Let's use eth_random_addr() to generate
this interface indentifier.

I remove completly the previous method, hence for the whole life of the
interface, the link-local address remains the same (previously, it depends on
which ethernet interfaces were up when the tunnel interface was set up).
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit df8372ca.

These changes are buggy and make unintended semantic changes
to ip6_tnl_add_linklocal().
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We need to move the derefernce after the IS_ERR() check.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Acked-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As discussed last year [1], there is no compelling reason
to limit IPv4 MTU to 0xFFF0, while real limit is 0xFFFF

[1] : http://marc.info/?l=linux-netdev&m=135607247609434&w=2

Willem raised this issue again because some of our internal
regression tests broke after lo mtu being set to 65536.

IP_MTU reports 0xFFF0, and the test attempts to send a RAW datagram of
mtu + 1 bytes, expecting the send() to fail, but it does not.

Alexey raised interesting points about TCP MSS, that should be addressed
in follow-up patches in TCP stack if needed, as someone could also set
an odd mtu anyway.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The nocache-argument was used in tcp_v4_send_synack as an argument to
inet_csk_route_req. However, since ba3f7f04 (ipv4: Kill
FLOWI_FLAG_RT_NOCACHE and associated code.) this is no more used.

This patch removes the unsued argument from tcp_v4_send_synack.
Signed-off-by: NChristoph Paasch <christoph.paasch@uclouvain.be>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ERROR: code indent should use tabs where possible: fix 2.
ERROR: do not use assignment in if condition: fix 5.
Signed-off-by: NDing Tianhong <dingtianhong@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Just follow the Joe Perches's opinion, it is a better way to fix the
style errors.
Suggested-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDing Tianhong <dingtianhong@huawei.com>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Handle context based address when an unspecified address is given.
For other context based address we print a warning and drop the packet
because we don't support it right now.
Signed-off-by: NAlexander Aring <alex.aring@gmail.com>
Reviewed-by: NWerner Almesberger <werner@almesberger.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch drops the pre and postcount calculation from the
lowpan_uncompress_addr function.We use instead a switch/case
over address_mode value. The original implementation has several
bugs in this function and it was hard to decrypt how it works.
To make it maintainable and fix these bugs this patch basically
reimplements lowpan_uncompress_addr from scratch.

A list of bugs we found in the current implementation:

1) Properly support uncompression of short-address based IPv6 addresses
   (instead of basically copying garbage)

2) Fix use and uncompression of long-addresses based IPv6 addresses

3) Add missing ff:fe00 in the case of SAM/DAM = 2 and M = 0
Signed-off-by: NAlexander Aring <alex.aring@gmail.com>
Reviewed-by: NWerner Almesberger <werner@almesberger.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add function to uncompress multicast address.
This function split the uncompress function for a multicast address
in a seperate function.

To uncompress a multicast address is different than a other
non-multicasts addresses according to rfc6282.
Signed-off-by: NAlexander Aring <alex.aring@gmail.com>
Reviewed-by: NWerner Almesberger <werner@almesberger.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds a helper function to parse the ipv6 header to a
6lowpan header in stream.

This function checks first if we can pull data with a specific
length from a skb. If this seems to be okay, we copy skb data to
a destination pointer and run skb_pull.
Signed-off-by: NAlexander Aring <alex.aring@gmail.com>
Reviewed-by: NWerner Almesberger <werner@almesberger.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a new 6lowpan fragment is received, a skbuff is allocated for
the reassembled packet. However when a 6lowpan packet compresses
link-local addresses based on link-layer addresses, the processing
function relies on the skb mac control block to find the related
link-layer address.

This patch copies the control block from the first fragment into
the newly allocated skb to keep a trace of the link-layer addresses
in case of a link-local compressed address.

Edit: small changes on comment issue
Signed-off-by: NDavid Hauweele <david@hauweele.net>
Signed-off-by: NAlexander Aring <alex.aring@gmail.com>
Reviewed-by: NWerner Almesberger <werner@almesberger.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch simplify the handling to set fields inside of struct ipv6hdr
to zero. Instead of setting some memory regions with memset to zero we
initialize the whole ipv6hdr to zero.

This is a simplification for parsing the 6lowpan header for the upcomming
patches.
Signed-off-by: NAlexander Aring <alex.aring@gmail.com>
Reviewed-by: NWerner Almesberger <werner@almesberger.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Following patch adds vxlan vport type for openvswitch using
vxlan api. So now there is vxlan dependency for openvswitch.

CC: Jesse Gross <jesse@nicira.com>
Signed-off-by: NPravin B Shelar <pshelar@nicira.com>
Acked-by: NJesse Gross <jesse@nicira.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch removes a comment in xfrm_input() which became irrelevant
due to commit 2774c131, "xfrm: Handle blackhole route creation via afinfo".
That commit removed returning -EREMOTE in the xfrm_lookup() method when the
packet should be discarded  and also removed the correspoinding -EREMOTE
handlers. This was replaced by calling the make_blackhole() method. Therefore
the comment about -EREMOTE is not relevant anymore.
Signed-off-by: NRami Rosen <ramirose@gmail.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

xfrm_state timer should be independent of system clock change,
so switch to CLOCK_BOOTTIME base which is not only monotonic but
also counting suspend time.

Thus issue reported in commit: 9e0d57fd
("xfrm: SAD entries do not expire correctly after suspend-resume")
could ALSO be avoided.

v2: Use CLOCK_BOOTTIME to count suspend time, but still monotonic.
Signed-off-by: NFan Du <fan.du@windriver.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

Following patch stores struct netlink_callback in netlink_sock
to avoid allocating and freeing it on every netlink dump msg.
Only one dump operation is allowed for a given socket at a time
therefore we can safely convert cb pointer to cb struct inside
netlink_sock.
Signed-off-by: NPravin B Shelar <pshelar@nicira.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

UIDs are printed in the proc_fs as signed int, whereas
they are unsigned int.
Signed-off-by: NFrancesco Fusco <ffusco@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit 56b765b7 ("htb: improved accuracy at high rates")
broke the "linklayer atm" handling.

 tc class add ... htb rate X ceil Y linklayer atm

The linklayer setting is implemented by modifying the rate table
which is send to the kernel.  No direct parameter were
transferred to the kernel indicating the linklayer setting.

The commit 56b765b7 ("htb: improved accuracy at high rates")
removed the use of the rate table system.

To keep compatible with older iproute2 utils, this patch detects
the linklayer by parsing the rate table.  It also supports future
versions of iproute2 to send this linklayer parameter to the
kernel directly. This is done by using the __reserved field in
struct tc_ratespec, to convey the choosen linklayer option, but
only using the lower 4 bits of this field.

Linklayer detection is limited to speeds below 100Mbit/s, because
at high rates the rtab is gets too inaccurate, so bad that
several fields contain the same values, this resembling the ATM
detect.  Fields even start to contain "0" time to send, e.g. at
1000Mbit/s sending a 96 bytes packet cost "0", thus the rtab have
been more broken than we first realized.
Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We know that "dev" is a valid pointer at this point, so we can remove
the test and clean up a little.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch allows to switch the netns when packet is encapsulated or
decapsulated. In other word, the encapsulated packet is received in a netns,
where the lookup is done to find the tunnel. Once the tunnel is found, the
packet is decapsulated and injecting into the corresponding interface which
stands to another netns.

When one of the two netns is removed, the tunnel is destroyed.
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch allows to switch the netns when packet is encapsulated or
decapsulated. In other word, the encapsulated packet is received in a netns,
where the lookup is done to find the tunnel. Once the tunnel is found, the
packet is decapsulated and injecting into the corresponding interface which
stands to another netns.

When one of the two netns is removed, the tunnel is destroyed.
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>