This patch adds proper XDP_TX action support. For each Tx ring, an
additional XDP Tx ring is allocated and setup. This version does the
DMA mapping in the fast-path, which will penalize performance for
IOMMU enabled systems. Further, debugfs support is not wired up for
the XDP Tx rings.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This commit adds basic XDP support for i40e derived NICs. All XDP
actions will end up in XDP_DROP.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Saeed Mahameed says:

====================
mlx5-updates-2017-06-20 (mlx5 IPoIB updates)

This series includes updates to mlx5 IPoIB netdevice driver (mlx5i),

1. We move ipoib files into separate directory, to allow it to grow
   separately in its own space
2. Remove HW update carrier logic from IPoIB and VF representors profiles.
3. Add basic ethtool support. (Rings options/statistics and driver info).
4. Change MTU support.
5. Xmit path statistics reporting.
6. add PTP support.

For the new ethtool ops, PTP (ioctl) and change_mtu ndos in IPoIB, we didn't add new
implementation or new logic, we only reused those callbacks from the already existing
mlx5e (ethernet netdevice profile) and exposed them in IPoIB netdevice/ethtool ops.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Julian Wiedmann says:

====================
s390/net updates, part 2 (v2)

thanks for the feedback. Here's an updated patchset that honours
the reverse christmas tree and drops the __packed attribute. Please apply.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

When a s390 guest runs on a z/VM host that's part of a SSI cluster,
it can be migrated to a different host. In this case, the MAC address
it originally obtained on the old host may be re-assigned to a new
guest. This would result in address conflicts between the two guests.

When running as z/VM guest, use the diag26c MAC Service to obtain
a hypervisor-managed MAC address. The MAC Service is SSI-aware, and
won't re-assign the address after the guest is migrated to a new host.

This patch adds support for the z/VM MAC Service on L2 devices.
Signed-off-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Acked-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Implement support for the hypervisor diagnose 0x26c
('Access Certain System Information').
It passes a request buffer and a subfunction code, and receives
a response buffer and a return code.

Also add the scaffolding for the 'MAC Services' subfunction.
It may be used by network devices to obtain a hypervisor-managed
MAC address.
Signed-off-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

There's two spots in qeth_send_packet() where we don't accurately
account for transmitted packing buffers in qeth's performance
statistics:

1) when flushing the current buffer due to insufficient size,
   and the next buffer is not EMPTY, we need to account for that
   flushed buffer.
2) when synchronizing with the TX completion code, we reset
   flush_count and thus forget to account for any previously
   flushed buffers.
Reported-by: Nils Hoppmann <niho@de.ibm.com>
Signed-off-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

add ipa return codes for Bridgeport (HiperSockets and OSA) according to
system level design.
Signed-off-by: Kittipon Meesompop <kmeesomp@linux.vnet.ibm.com>
Reviewed-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Reviewed-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

It's a bad thing not to handle errors when updating asoc. The memory
allocation failure in any of the functions called in sctp_assoc_update()
would cause sctp to work unexpectedly.

This patch is to fix it by aborting the asoc and reporting the error when
any of these functions fails.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

local_cork is used to decide if it should uncork asoc outq after processing
some cmds, and it is set when replying or sending msgs. local_cork should
always have the same value with current asoc q->cork in some way.

The thing is when changing to a new asoc by cmd SET_ASOC, local_cork may
not be consistent with the current asoc any more. The cmd seqs can be:

  SCTP_CMD_UPDATE_ASSOC (asoc)
  SCTP_CMD_REPLY (asoc)
  SCTP_CMD_SET_ASOC (new_asoc)
  SCTP_CMD_DELETE_TCB (new_asoc)
  SCTP_CMD_SET_ASOC (asoc)
  SCTP_CMD_REPLY (asoc)

The 1st REPLY makes OLD asoc q->cork and local_cork both are 1, and the cmd
DELETE_TCB clears NEW asoc q->cork and local_cork. After asoc goes back to
OLD asoc, q->cork is still 1 while local_cork is 0. The 2nd REPLY will not
set local_cork because q->cork is already set and it can't be uncorked and
sent out because of this.

To keep local_cork consistent with the current asoc q->cork, this patch is
to uncork the old asoc if local_cork is set before changing to the new one.

Note that the above cmd seqs will be used in the next patch when updating
asoc and handling errors in it.
Suggested-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Patch "call inet_add_protocol after register_pernet_subsys in dccp_v4_init"
fixed a null pointer dereference issue for dccp_ipv4 module.

The same fix is needed for dccp_ipv6 module.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Now dccp_ipv4 works as a kernel module. During loading this module, if
one dccp packet is being recieved after inet_add_protocol but before
register_pernet_subsys in which v4_ctl_sk is initialized, a null pointer
dereference may be triggered because of init_net.dccp.v4_ctl_sk is 0x0.

Jianlin found this issue when the following call trace occurred:

[  171.950177] BUG: unable to handle kernel NULL pointer dereference at 0000000000000110
[  171.951007] IP: [<ffffffffc0558364>] dccp_v4_ctl_send_reset+0xc4/0x220 [dccp_ipv4]
[...]
[  171.984629] Call Trace:
[  171.984859]  <IRQ>
[  171.985061]
[  171.985213]  [<ffffffffc0559a53>] dccp_v4_rcv+0x383/0x3f9 [dccp_ipv4]
[  171.985711]  [<ffffffff815ca054>] ip_local_deliver_finish+0xb4/0x1f0
[  171.986309]  [<ffffffff815ca339>] ip_local_deliver+0x59/0xd0
[  171.986852]  [<ffffffff810cd7a4>] ? update_curr+0x104/0x190
[  171.986956]  [<ffffffff815c9cda>] ip_rcv_finish+0x8a/0x350
[  171.986956]  [<ffffffff815ca666>] ip_rcv+0x2b6/0x410
[  171.986956]  [<ffffffff810c83b4>] ? task_cputime+0x44/0x80
[  171.986956]  [<ffffffff81586f22>] __netif_receive_skb_core+0x572/0x7c0
[  171.986956]  [<ffffffff810d2c51>] ? trigger_load_balance+0x61/0x1e0
[  171.986956]  [<ffffffff81587188>] __netif_receive_skb+0x18/0x60
[  171.986956]  [<ffffffff8158841e>] process_backlog+0xae/0x180
[  171.986956]  [<ffffffff8158799d>] net_rx_action+0x16d/0x380
[  171.986956]  [<ffffffff81090b7f>] __do_softirq+0xef/0x280
[  171.986956]  [<ffffffff816b6a1c>] call_softirq+0x1c/0x30

This patch is to move inet_add_protocol after register_pernet_subsys in
dccp_v4_init, so that v4_ctl_sk is initialized before any incoming dccp
packets are processed.
Reported-by: Jianlin Shi <jishi@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

With -Wformat-truncation, gcc throws the following warning.

Fix this by increasing the size of devname to accommodate 15 character
netdev interface name and description.

Remove length format precision for %s. We can fit entire name.

Also increment the version.

drivers/net/ethernet/cisco/enic/enic_main.c: In function ‘enic_open’:
drivers/net/ethernet/cisco/enic/enic_main.c:1740:15: warning: ‘%u’ directive output may be truncated writing between 1 and 2 bytes into a region of size between 1 and 12 [-Wformat-truncation=]
     "%.11s-rx-%u", netdev->name, i);
               ^~
drivers/net/ethernet/cisco/enic/enic_main.c:1740:5: note: directive argument in the range [0, 16]
     "%.11s-rx-%u", netdev->name, i);
     ^~~~~~~~~~~~~
drivers/net/ethernet/cisco/enic/enic_main.c:1738:4: note: ‘snprintf’ output between 6 and 18 bytes into a destination of size 16
    snprintf(enic->msix[intr].devname,
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
     sizeof(enic->msix[intr].devname),
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
     "%.11s-rx-%u", netdev->name, i);
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Govindarajulu Varadarajan <gvaradar@cisco.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

There is nothing in the IP that prevents us from enabling TSO for IPv6.

Before patch:
ftp fe80::2aa:bbff:fecc:1336%eth0
ftp> get /dev/zero
882512708 bytes received in 00:14 (56.11 MiB/s)

After patch:
ftp fe80::2aa:bbff:fecc:1336%eth0
ftp> get /dev/zero
1203326784 bytes received in 00:12 (94.52 MiB/s)
Signed-off-by: Niklas Cassel <niklas.cassel@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

If the ibmvnic driver is not in the VNIC_OPEN state, return from
ibmvnic_resume callback. If we are not in the VNIC_OPEN state, interrupts
may not be initialized and directly calling the interrupt handler will
cause a crash.
Signed-off-by: John Allen <jallen@linux.vnet.ibm.com>
Reviewed-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Provide link partner advertising information.
Removed testing for gigabit modes, which is useless for a fast ethernet phy.
Signed-off-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

John Crispin says:

====================
net-next: mediatek: various performance improvements

During development we mainly ran testing using iperf doing 1500 byte
tcp frames. It was pointed out recently, that the driver does not perform
very well when using 512 byte udp frames. The biggest problem was that
RPS was not working as no rx queue was being set. fixing this more than
doubled the throughput. Additionally the IRQ mask register is now locked
independently for RX and TX. RX IRQ aggregation is also added. With all
these patches applied we can almost triple the throughput.

While at it we also add PHY status change reporting for GMACs connecting
directly to a PHY.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

The get_rps_cpu() function will not do any RPS on the data flow when no
queue is setup and always use the current cpu where the IRQ was handled
to also handle the backlog. As we only have one physical queue we always
set this to 0 unconditionally.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Originally the driver only utilised the new QDMA engine. The current code
still assumes this is the case when locking the IRQ mask register. Since
RX now runs on the old style PDMA engine we can add a second lock. This
patch reduces the IRQ latency as the TX and RX path no longer need to wait
on each other under heavy load.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

The PDMA engine used for RX allows IRQ aggregation. The patch sets up the
corresponding registers to aggregate 4 IRQs into one. Using aggregation
reduces the load on the core handling to a quarter thus reducing IRQ
latency and increasing RX performance by around 10%.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently PHY status changes are only printed for DSA ports. This patch
adds code to also print status changes for non-fixed links.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Matthias Schiffer says:

====================
vxlan: cleanup and IPv6 link-local support

Running VXLANs over IPv6 link-local addresses allows to use them as a
drop-in replacement for VLANs, avoiding to allocate additional outer IP
addresses to run the VXLAN over.

Since v1, I have added a lot more consistency checks to the address
configuration, making sure address families and scopes match. To simplify
the implementation, I also did some general refactoring of the
configuration handling in the new first patch of the series.

The second patch is more cleanup; is slightly touches OVS code, so that
list is in CC this time, too.

As in v1, the last two patches actually make VXLAN over IPv6 link-local
work, and allow multiple VXLANs with the same VNI and port, as long as
link-local addresses on different interfaces are used. As suggested, I now
store in the flags field if the VXLAN uses link-local addresses or not.

v3 removes log messages as suggested by Roopa Prabhu (as it is very unusual
for errors in netlink requests to be printed to the kernel log.) The commit
message of patch 5 has been extended to add a note about IPv4.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

As link-local addresses are only valid for a single interface, we can allow
to use the same VNI for multiple independent VXLANs, as long as the used
interfaces are distinct. This way, VXLANs can always be used as a drop-in
replacement for VLANs with greater ID space.

This also extends VNI lookup to respect the ifindex when link-local IPv6
addresses are used, so using the same VNI on multiple interfaces can
actually work.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

If VXLAN is run over link-local IPv6 addresses, it is necessary to store
the ifindex in the FDB entries. Otherwise, the used interface is undefined
and unicast communication will most likely fail.

Support for link-local IPv4 addresses should be possible as well, but as
the semantics aren't as well defined as for IPv6, and there doesn't seem to
be much interest in having the support, it's not implemented for now.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

* Multicast addresses are never valid as local address
* Link-local IPv6 unicast addresses may only be used as remote when the
  local address is link-local as well
* Don't allow link-local IPv6 local/remote addresses without interface

We also store in the flags field if link-local addresses are used for the
follow-up patches that actually make VXLAN over link-local IPv6 work.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Address families of source and destination addresses must match, and
changelink operations can't change the address family.

In addition, always use the VXLAN_F_IPV6 to check if a VXLAN device uses
IPv4 or IPv6.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

There is no good reason to keep the flags twice in vxlan_dev and
vxlan_config.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

The vxlan_dev_configure function was mixing validation and application of
the vxlan configuration; this could easily lead to bugs with the changelink
operation, as it was hard to see if the function wcould return an error
after parts of the configuration had already been applied.

This commit splits validation and application out of vxlan_dev_configure as
separate functions to make it clearer where error returns are allowed and
where the vxlan_dev or net_device may be configured. Log messages in these
functions are removed, as it is generally unexpected to find error output
for netlink requests in the kernel log. Userspace should be able to handle
errors based on the error codes returned via netlink just fine.

In addition, some validation and initialization is moved to vxlan_validate
and vxlan_setup respectively to improve grouping of similar settings.

Finally, this also fixes two actual bugs:

* if set, conf->mtu would overwrite dev->mtu in each changelink operation,
  reverting other changes of dev->mtu
* the "if (!conf->dst_port)" branch would never be run, as conf->dst_port
  was set in vxlan_setup before. This caused VXLAN-GPE to use the same
  default port as other VXLAN sockets instead of the intended IANA-assigned
  4790.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

yuan linyu says:

====================
net: more skb_put_[data:zero] related work

yuan linyu (3):
  net: introduce __skb_put_[zero, data, u8]
  net: replace more place to skb_put_[data:zero]
  net: manual clean code which call skb_put_[data:zero]
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: yuan linyu <Linyu.Yuan@alcatel-sbell.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>

spatch file,
@@
expression skb, len, data;
type t;
@@
-memcpy((t *)skb_put(skb, len), data, len);
+skb_put_data(skb, data, len);

@@
identifier p;
expression skb, len, data;
type t;
@@
-p = (t *)memset(skb_put(skb, len), data, len);
+p = skb_put_zero(skb, len);

@@
expression skb, len, data;
type t;
@@
-memcpy((t *)__skb_put(skb, len), data, len);
+__skb_put_data(skb, data, len);

@@
identifier p;
expression skb, len, data;
type t;
@@
-p = (t *)memset(__skb_put(skb, len), data, len);
+p = __skb_put_zero(skb, len);
Signed-off-by: yuan linyu <Linyu.Yuan@alcatel-sbell.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>

follow Johannes Berg, semantic patch file as below,
@@
identifier p, p2;
expression len;
expression skb;
type t, t2;
@@
(
-p = __skb_put(skb, len);
+p = __skb_put_zero(skb, len);
|
-p = (t)__skb_put(skb, len);
+p = __skb_put_zero(skb, len);
)
... when != p
(
p2 = (t2)p;
-memset(p2, 0, len);
|
-memset(p, 0, len);
)

@@
identifier p;
expression len;
expression skb;
type t;
@@
(
-t p = __skb_put(skb, len);
+t p = __skb_put_zero(skb, len);
)
... when != p
(
-memset(p, 0, len);
)

@@
type t, t2;
identifier p, p2;
expression skb;
@@
t *p;
...
(
-p = __skb_put(skb, sizeof(t));
+p = __skb_put_zero(skb, sizeof(t));
|
-p = (t *)__skb_put(skb, sizeof(t));
+p = __skb_put_zero(skb, sizeof(t));
)
... when != p
(
p2 = (t2)p;
-memset(p2, 0, sizeof(*p));
|
-memset(p, 0, sizeof(*p));
)

@@
expression skb, len;
@@
-memset(__skb_put(skb, len), 0, len);
+__skb_put_zero(skb, len);

@@
expression skb, len, data;
@@
-memcpy(__skb_put(skb, len), data, len);
+__skb_put_data(skb, data, len);

@@
expression SKB, C, S;
typedef u8;
identifier fn = {__skb_put};
fresh identifier fn2 = fn ## "_u8";
@@
- *(u8 *)fn(SKB, S) = C;
+ fn2(SKB, C);
Signed-off-by: yuan linyu <Linyu.Yuan@alcatel-sbell.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>

Kill the remaining shift macro in favor of calculating at compile time
its value from the more descriptive mask, which gives us a better
representation of the register layout.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Vivien Didelot says:

====================
net: dsa: Global 2 cosmetics

Similarly to what has been done for the Port and Global 1 registers,
this patch series prefixes and documents the macros of Global 2.

It brings no functional changes except for 1/10 which fixes the IRL init
for 88E6390 family.

Changes in v2: make *_g2_irl_init_all static inline without
NET_DSA_MV88E6XXX_GLOBAL2 and compile test with and without the symbol.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Prefix and document the remaining Global 2 registers macros.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The Marvell 88E6352 family has a Global 2 register dedicated to the
watchdog setup. But the 88E6390 turned it into an indirect table.

Prefix and document that.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Prefix and document the Global 2 Switch MAC registers macros.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Prefix and document the Global 2 EEPROM registers macros.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Prefix and document the Global 2 Cross-chip Port VLAN registers macros.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Prefix and document the Global 2 MGMT registers macros.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>