This is additional to the
commit ea1627c2 ("tcp: minor optimizations around tcp_hdr() usage").
At this point, skb->data is same with tcp_hdr() as tcp header has not
been pulled yet. So use the less expensive one to get the tcp header.

Remove the third parameter of tcp_rcv_established() and put it into
the function body.

Furthermore, the local variables are listed as a reverse christmas tree :)

Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: NYafang Shao <laoar.shao@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add support for IFA_RT_PRIORITY to ipv6 addresses.

If the metric is changed on an existing address then the new route
is inserted before removing the old one. Since the metric is one
of the route keys, the prefix route can not be atomically replaced.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add support for IFA_RT_PRIORITY to ipv4 addresses.

If the metric is changed on an existing address then the new route
is inserted before removing the old one. Since the metric is one
of the route keys, the prefix route can not be replaced.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Move config parameters for adding an ipv6 address to a struct. struct
names stem from inet6_rtm_newaddr which is the modern handler for
adding an address.

Start the conversion to ifa6_config with ipv6_add_addr. This is an argument
move only; no functional change intended. Mapping of variable changes:

    addr      -->  cfg->pfx
    peer_addr -->  cfg->peer_pfx
    pfxlen    -->  cfg->plen
    flags     -->  cfg->ifa_flags

scope, valid_lft, prefered_lft have the same names within cfg
(with corrected spelling).
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

MQ doesn't hold any statistics on its own, however, statistic
from offloads are requested starting from the root, hence MQ
will read the old values for its sums.  Call into the drivers,
because of the additive nature of the stats drivers are aware
of how much "pending updates" they have to children of the MQ.
Since MQ reset its stats on every dump we can simply offset
the stats, predicting how stats of offloaded children will
change.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

mq offload is trivial, we just need to let the device know
that the root qdisc is mq.  Alternative approach would be
to export qdisc_lookup() and make drivers check the root
type themselves, but notification via ndo_setup_tc is more
in line with other qdiscs.

Note that mq doesn't hold any stats on it's own, it just
adds up stats of its children.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

AFAICT struct gnet_stats_queue.qlen is not used in Qdiscs.
It may, however, be useful for offloads to report HW queue
length there.  Add that value to the result of qdisc_qlen_sum().
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The struct Qdisc has a lot of holes, especially after commit
a53851e2 ("net: sched: explicit locking in gso_cpu fallback"),
which as a side effect, moved the fields just after 'busylock'
on a new cacheline.

Since both 'padded' and 'refcnt' are not updated frequently, and
there is a hole before 'gso_skb', we can move such fields there,
saving a cacheline without any performance side effect.

Before this commit:

pahole -C Qdisc net/sche/sch_generic.o
	# ...
        /* size: 384, cachelines: 6, members: 25 */
        /* sum members: 236, holes: 3, sum holes: 92 */
        /* padding: 56 */

After this commit:
pahole -C Qdisc net/sche/sch_generic.o
	# ...
	/* size: 320, cachelines: 5, members: 25 */
	/* sum members: 236, holes: 2, sum holes: 28 */
	/* padding: 56 */
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The net_failover driver provides an automated failover mechanism via APIs
to create and destroy a failover master netdev and manages a primary and
standby slave netdevs that get registered via the generic failover
infrastructure.

The failover netdev acts a master device and controls 2 slave devices. The
original paravirtual interface gets registered as 'standby' slave netdev and
a passthru/vf device with the same MAC gets registered as 'primary' slave
netdev. Both 'standby' and 'failover' netdevs are associated with the same
'pci' device. The user accesses the network interface via 'failover' netdev.
The 'failover' netdev chooses 'primary' netdev as default for transmits when
it is available with link up and running.

This can be used by paravirtual drivers to enable an alternate low latency
datapath. It also enables hypervisor controlled live migration of a VM with
direct attached VF by failing over to the paravirtual datapath when the VF
is unplugged.
Signed-off-by: NSridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The failover module provides a generic interface for paravirtual drivers
to register a netdev and a set of ops with a failover instance. The ops
are used as event handlers that get called to handle netdev register/
unregister/link change/name change events on slave pci ethernet devices
with the same mac address as the failover netdev.

This enables paravirtual drivers to use a VF as an accelerated low latency
datapath. It also allows migration of VMs with direct attached VFs by
failing over to the paravirtual datapath when the VF is unplugged.
Signed-off-by: NSridhar Samudrala <sridhar.samudrala@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 05f0fe6b ("RCU, workqueue: Implement rcu_work") introduces
new API's for dispatching work in a RCU callback. Now we can just
switch to the new API's for tc filters. This could get rid of a lot
of code.

Cc: Tejun Heo <tj@kernel.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When sending an xdp_frame through xdp_do_redirect call, then error
cases can happen where the xdp_frame needs to be dropped, and
returning an -errno code isn't sufficient/possible any-longer
(e.g. for cpumap case). This is already fully supported, by simply
calling xdp_return_frame.

This patch is an optimization, which provides xdp_return_frame_rx_napi,
which is a faster variant for these error cases.  It take advantage of
the protection provided by XDP RX running under NAPI protection.

This change is mostly relevant for drivers using the page_pool
allocator as it can take advantage of this. (Tested with mlx5).
Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

In this patch, we add dcbnl buffer attribute to allow user
change the NIC's buffer configuration such as priority
to buffer mapping and buffer size of individual buffer.

This attribute combined with pfc attribute allows advanced user to
fine tune the qos setting for specific priority queue. For example,
user can give dedicated buffer for one or more priorities or user
can give large buffer to certain priorities.

The dcb buffer configuration will be controlled by lldptool.
lldptool -T -i eth2 -V BUFFER prio 0,2,5,7,1,2,3,6
  maps priorities 0,1,2,3,4,5,6,7 to receive buffer 0,2,5,7,1,2,3,6
lldptool -T -i eth2 -V BUFFER size 87296,87296,0,87296,0,0,0,0
  sets receive buffer size for buffer 0,1,2,3,4,5,6,7 respectively

After discussion on mailing list with Jakub, Jiri, Ido and John, we agreed to
choose dcbnl over devlink interface since this feature is intended to set
port attributes which are governed by the netdev instance of that port, where
devlink API is more suitable for global ASIC configurations.

We present an use case scenario where dcbnl buffer attribute configured
by advance user helps reduce the latency of messages of different sizes.

Scenarios description:
On ConnectX-5, we run latency sensitive traffic with
small/medium message sizes ranging from 64B to 256KB and bandwidth sensitive
traffic with large messages sizes 512KB and 1MB. We group small, medium,
and large message sizes to their own pfc enables priorities as follow.
  Priorities 1 & 2 (64B, 256B and 1KB)
  Priorities 3 & 4 (4KB, 8KB, 16KB, 64KB, 128KB and 256KB)
  Priorities 5 & 6 (512KB and 1MB)

By default, ConnectX-5 maps all pfc enabled priorities to a single
lossless fixed buffer size of 50% of total available buffer space. The
other 50% is assigned to lossy buffer. Using dcbnl buffer attribute,
we create three equal size lossless buffers. Each buffer has 25% of total
available buffer space. Thus, the lossy buffer size reduces to 25%. Priority
to lossless  buffer mappings are set as follow.
  Priorities 1 & 2 on lossless buffer #1
  Priorities 3 & 4 on lossless buffer #2
  Priorities 5 & 6 on lossless buffer #3

We observe improvements in latency for small and medium message sizes
as follows. Please note that the large message sizes bandwidth performance is
reduced but the total bandwidth remains the same.
  256B message size (42 % latency reduction)
  4K message size (21% latency reduction)
  64K message size (16% latency reduction)

CC: Ido Schimmel <idosch@idosch.org>
CC: Jakub Kicinski <jakub.kicinski@netronome.com>
CC: Jiri Pirko <jiri@resnulli.us>
CC: Or Gerlitz <gerlitz.or@gmail.com>
CC: Parav Pandit <parav@mellanox.com>
CC: Aron Silverton <aron.silverton@oracle.com>
Signed-off-by: NHuy Nguyen <huyn@mellanox.com>
Reviewed-by: NParav Pandit <parav@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

The BPF seg6local hook should be powerful enough to enable users to
implement most of the use-cases one could think of. After some thinking,
we figured out that the following actions should be possible on a SRv6
packet, requiring 3 specific helpers :
    - bpf_lwt_seg6_store_bytes: Modify non-sensitive fields of the SRH
    - bpf_lwt_seg6_adjust_srh: Allow to grow or shrink a SRH
                               (to add/delete TLVs)
    - bpf_lwt_seg6_action: Apply some SRv6 network programming actions
                           (specifically End.X, End.T, End.B6 and
                            End.B6.Encap)

The specifications of these helpers are provided in the patch (see
include/uapi/linux/bpf.h).

The non-sensitive fields of the SRH are the following : flags, tag and
TLVs. The other fields can not be modified, to maintain the SRH
integrity. Flags, tag and TLVs can easily be modified as their validity
can be checked afterwards via seg6_validate_srh. It is not allowed to
modify the segments directly. If one wants to add segments on the path,
he should stack a new SRH using the End.B6 action via
bpf_lwt_seg6_action.

Growing, shrinking or editing TLVs via the helpers will flag the SRH as
invalid, and it will have to be re-validated before re-entering the IPv6
layer. This flag is stored in a per-CPU buffer, along with the current
header length in bytes.

Storing the SRH len in bytes in the control block is mandatory when using
bpf_lwt_seg6_adjust_srh. The Header Ext. Length field contains the SRH
len rounded to 8 bytes (a padding TLV can be inserted to ensure the 8-bytes
boundary). When adding/deleting TLVs within the BPF program, the SRH may
temporary be in an invalid state where its length cannot be rounded to 8
bytes without remainder, hence the need to store the length in bytes
separately. The caller of the BPF program can then ensure that the SRH's
final length is valid using this value. Again, a final SRH modified by a
BPF program which doesn’t respect the 8-bytes boundary will be discarded
as it will be considered as invalid.

Finally, a fourth helper is provided, bpf_lwt_push_encap, which is
available from the LWT BPF IN hook, but not from the seg6local BPF one.
This helper allows to encapsulate a Segment Routing Header (either with
a new outer IPv6 header, or by inlining it directly in the existing IPv6
header) into a non-SRv6 packet. This helper is required if we want to
offer the possibility to dynamically encapsulate a SRH for non-SRv6 packet,
as the BPF seg6local hook only works on traffic already containing a SRH.
This is the BPF equivalent of the seg6 LWT infrastructure, which achieves
the same purpose but with a static SRH per route.

These helpers require CONFIG_IPV6=y (and not =m).
Signed-off-by: NMathieu Xhonneux <m.xhonneux@gmail.com>
Acked-by: NDavid Lebrun <dlebrun@google.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

The function lookup_nexthop is essential to implement most of the seg6local
actions. As we want to provide a BPF helper allowing to apply some of these
actions on the packet being processed, the helper should be able to call
this function, hence the need to make it public.

Moreover, if one argument is incorrect or if the next hop can not be found,
an error should be returned by the BPF helper so the BPF program can adapt
its processing of the packet (return an error, properly force the drop,
...). This patch hence makes this function return dst->error to indicate a
possible error.
Signed-off-by: NMathieu Xhonneux <m.xhonneux@gmail.com>
Acked-by: NDavid Lebrun <dlebrun@google.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

include/net/seg6.h cannot be included in a source file if CONFIG_IPV6 is
not enabled:
   include/net/seg6.h: In function 'seg6_pernet':
>> include/net/seg6.h:52:14: error: 'struct net' has no member named
                                        'ipv6'; did you mean 'ipv4'?
     return net->ipv6.seg6_data;
                 ^~~~
                 ipv4

This commit makes seg6_pernet return NULL if IPv6 is not compiled, hence
allowing seg6.h to be included regardless of the configuration.
Signed-off-by: NMathieu Xhonneux <m.xhonneux@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

This is a followup to fib rules sport, dport and ipproto
match support. Only supports tcp, udp and icmp for ipproto.
Used by fib rule self tests.
Signed-off-by: NRoopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The struct nft_af_info was removed.
Signed-off-by: NTaehee Yoo <ap420073@gmail.com>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Use NL80211_CMD_UPDATE_CONNECT_PARAMS to update new ERP information,
Association IEs and the Authentication type to driver / firmware which
will be used in subsequent roamings.
Signed-off-by: NVidyullatha Kanchanapally <vidyullatha@codeaurora.org>
[arend: extended fils-sk kernel doc and added check in wiphy_register()]
Reviewed-by: NJithu Jance <jithu.jance@broadcom.com>
Reviewed-by: NEylon Pedinovsky <eylon.pedinovsky@broadcom.com>
Signed-off-by: NArend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

In case of FILS shared key offload the parameters can change
upon roaming of which user-space needs to be notified.
Reviewed-by: NJithu Jance <jithu.jance@broadcom.com>
Reviewed-by: NEylon Pedinovsky <eylon.pedinovsky@broadcom.com>
Signed-off-by: NArend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

Put FILS related parameters into their own struct definition so
it can be reused for roam events in subsequent change.
Reviewed-by: NJithu Jance <jithu.jance@broadcom.com>
Reviewed-by: NEylon Pedinovsky <eylon.pedinovsky@broadcom.com>
Signed-off-by: NArend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

There are specific cases, such as SAE authentication exchange, that
might require long duration to complete. For such cases, add support
for indicating to the driver the required duration of the prepare_tx()
operation, so the driver would still be able to complete the frame
exchange.

Currently, indicate the duration only for SAE authentication exchange,
as SAE authentication can take up to 2000 msec (as defined in IEEE
P802.11-REVmd D1.0 p. 3504).

As the patch modified the prepare_tx() callback API, also modify
the relevant code in iwlwifi.
Signed-off-by: NIlan Peer <ilan.peer@intel.com>
Signed-off-by: NLuca Coelho <luciano.coelho@intel.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

Move decode_session() and parse_nat_setup_hook() indirections to struct
nf_nat_hook structure.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Currently the packet rewrite and instantiation of nat NULL bindings
happens from the protocol specific nat backend.

Invocation occurs either via ip(6)table_nat or the nf_tables nat chain type.

Invocation looks like this (simplified):
NF_HOOK()
   |
   `---iptable_nat
	 |
	 `---> nf_nat_l3proto_ipv4 -> nf_nat_packet
	               |
          new packet? pass skb though iptables nat chain
                       |
		       `---> iptable_nat: ipt_do_table

In nft case, this looks the same (nft_chain_nat_ipv4 instead of
iptable_nat).

This is a problem for two reasons:
1. Can't use iptables nat and nf_tables nat at the same time,
   as the first user adds a nat binding (nf_nat_l3proto_ipv4 adds a
   NULL binding if do_table() did not find a matching nat rule so we
   can detect post-nat tuple collisions).
2. If you use e.g. nft_masq, snat, redir, etc. uses must also register
   an empty base chain so that the nat core gets called fro NF_HOOK()
   to do the reverse translation, which is neither obvious nor user
   friendly.

After this change, the base hook gets registered not from iptable_nat or
nftables nat hooks, but from the l3 nat core.

iptables/nft nat base hooks get registered with the nat core instead:

NF_HOOK()
   |
   `---> nf_nat_l3proto_ipv4 -> nf_nat_packet
		|
         new packet? pass skb through iptables/nftables nat chains
                |
		+-> iptables_nat: ipt_do_table
	        +-> nft nat chain x
	        `-> nft nat chain y

The nat core deals with null bindings and reverse translation.
When no mapping exists, it calls the registered nat lookup hooks until
one creates a new mapping.
If both iptables and nftables nat hooks exist, the first matching
one is used (i.e., higher priority wins).

Also, nft users do not need to create empty nat hooks anymore,
nat core always registers the base hooks that take care of reverse/reply
translation.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

This adds the infrastructure to register nat hooks with the nat core
instead of the netfilter core.

nat hooks are used to configure nat bindings.  Such hooks are registered
from ip(6)table_nat or by the nftables core when a nat chain is added.

After next patch, nat hooks will be registered with nf_nat instead of
netfilter core.  This allows to use many nat lookup functions at the
same time while doing the real packet rewrite (nat transformation) in
one place.

This change doesn't convert the intended users yet to ease review.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Will be used in followup patch when nat types no longer
use nf_register_net_hook() but will instead register with the nat core.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Copy-pasted, both l3 helpers almost use same code here.
Split out the common part into an 'inet' helper.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Bring consistency to ipv6 route replace and append semantics.

Remove rt6_qualify_for_ecmp which is just guess work. It fails in 2 cases:
1. can not replace a route with a reject route. Existing code appends
   a new route instead of replacing the existing one.

2. can not have a multipath route where a leg uses a dev only nexthop

Existing use cases affected by this change:
1. adding a route with existing prefix and metric using NLM_F_CREATE
   without NLM_F_APPEND or NLM_F_EXCL (ie., what iproute2 calls
   'prepend'). Existing code auto-determines that the new nexthop can
   be appended to an existing route to create a multipath route. This
   change breaks that by requiring the APPEND flag for the new route
   to be added to an existing one. Instead the prepend just adds another
   route entry.

2. route replace. Existing code replaces first matching multipath route
   if new route is multipath capable and fallback to first matching
   non-ECMP route (reject or dev only route) in case one isn't available.
   New behavior replaces first matching route. (Thanks to Ido for spotting
   this one)

Note: Newer iproute2 is needed to display multipath routes with a dev-only
      nexthop. This is due to a bug in iproute2 and parsing nexthops.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now sctp uses inet_dgram_connect as its proto_ops .connect, and the flags
param can't be passed into its proto .connect where this flags is really
needed.

sctp works around it by getting flags from socket file in __sctp_connect.
It works for connecting from userspace, as inherently the user sock has
socket file and it passes f_flags as the flags param into the proto_ops
.connect.

However, the sock created by sock_create_kern doesn't have a socket file,
and it passes the flags (like O_NONBLOCK) by using the flags param in
kernel_connect, which calls proto_ops .connect later.

So to fix it, this patch defines a new proto_ops .connect for sctp,
sctp_inet_connect, which calls __sctp_connect() directly with this
flags param. After this, the sctp's proto .connect can be removed.

Note that sctp_inet_connect doesn't need to do some checks that are not
needed for sctp, which makes thing better than with inet_dgram_connect.
Suggested-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Acked-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Reviewed-by: NMichal Kubecek <mkubecek@suse.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add the kernel-doc missed earlier.

Fixes: 52539ca8 ("cfg80211: Expose TXQ stats and parameters to userspace")
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

Determine path MTU from a FIB lookup result. Logic is based on
ip6_dst_mtu_forward plus lookup of nexthop exception.

Add ip6_dst_mtu_forward to ipv6_stubs to handle access by core
bpf code.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

Determine path MTU from a FIB lookup result. Logic is a distillation of
ip_dst_mtu_maybe_forward.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

Before the patch, the erspan BSO bit (Bad/Short/Oversized) is not
handled.  BSO has 4 possible values:
  00 --> Good frame with no error, or unknown integrity
  11 --> Payload is a Bad Frame with CRC or Alignment Error
  01 --> Payload is a Short Frame
  10 --> Payload is an Oversized Frame

Based the short/oversized definitions in RFC1757, the patch sets
the bso bit based on the mirrored packet's size.
Reported-by: NXiaoyan Jin <xiaoyanj@vmware.com>
Signed-off-by: NWilliam Tu <u9012063@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Each driver implements physical port name generation by itself. However
as devlink has all needed info, it can easily do the job for all its
users. So implement this helper in devlink.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Devlink ports can have specific flavour according to the purpose of use.
This patch extend attrs_set so the driver can say which flavour port
has. Initial flavours are:
physical, cpu, dsa
User can query this to see right away what is the purpose of each port.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Change existing setter for split port information into more generic
attrs setter. Alongside with that, allow to set port number and subport
number for split ports.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This per netns sysctl allows for TCP SACK compression fine-tuning.

This limits number of SACK that can be compressed.
Using 0 disables SACK compression.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This per netns sysctl allows for TCP SACK compression fine-tuning.

Its default value is 1,000,000, or 1 ms to meet TSO autosizing period.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When TCP receives an out-of-order packet, it immediately sends
a SACK packet, generating network load but also forcing the
receiver to send 1-MSS pathological packets, increasing its
RTX queue length/depth, and thus processing time.

Wifi networks suffer from this aggressive behavior, but generally
speaking, all these SACK packets add fuel to the fire when networks
are under congestion.

This patch adds a high resolution timer and tp->compressed_ack counter.

Instead of sending a SACK, we program this timer with a small delay,
based on RTT and capped to 1 ms :

	delay = min ( 5 % of RTT, 1 ms)

If subsequent SACKs need to be sent while the timer has not yet
expired, we simply increment tp->compressed_ack.

When timer expires, a SACK is sent with the latest information.
Whenever an ACK is sent (if data is sent, or if in-order
data is received) timer is canceled.

Note that tcp_sack_new_ofo_skb() is able to force a SACK to be sent
if the sack blocks need to be shuffled, even if the timer has not
expired.

A new SNMP counter is added in the following patch.

Two other patches add sysctls to allow changing the 1,000,000 and 44
values that this commit hard-coded.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Acked-by: NYuchung Cheng <ycheng@google.com>
Acked-by: NToke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Socket can not disappear under us.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>