GitHub/moto-9609/android_kernel_motorola_exynos9610.git
8 years agobatman-adv: Split batadv_iv_ogm_orig_del_if function
Sven Eckelmann [Mon, 22 Feb 2016 21:56:34 +0000 (22:56 +0100)]
batman-adv: Split batadv_iv_ogm_orig_del_if function

batadv_iv_ogm_orig_del_if handles two different buffers bcast_own and
bcast_own_sum which should be resized. The error handling two for
allocating these buffers causes the complexity of this function. This can
be avoided completely when the function is split into a main function
handling the locking, freeing and call of the subfunctions.

The subfunction can then independently handle the resize of the buffers.
This also allows to easily reuse the old buffer (which always is larger) in
case a smaller buffer could not be allocated without increasing the code
complexity.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: Merge batadv_v_ogm_orig_update into batadv_v_ogm_route_update
Simon Wunderlich [Mon, 1 Feb 2016 14:21:38 +0000 (15:21 +0100)]
batman-adv: Merge batadv_v_ogm_orig_update into batadv_v_ogm_route_update

Since batadv_v_ogm_orig_update() was only called from one place and the
calling function became very short, merge these two functions together.

This should also reflect the protocol description of B.A.T.M.A.N. V
better.

Signed-off-by: Simon Wunderlich <simon@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: move and restructure batadv_v_ogm_forward
Simon Wunderlich [Mon, 1 Feb 2016 14:21:37 +0000 (15:21 +0100)]
batman-adv: move and restructure batadv_v_ogm_forward

To match our code better to the protocol description of B.A.T.M.A.N. V,
move batadv_v_ogm_forward() out into batadv_v_ogm_process_per_outif()
and move all checks directly deciding whether the OGM should be
forwarded into batadv_v_ogm_forward().

Signed-off-by: Simon Wunderlich <simon@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: fix debuginfo macro style issue
Simon Wunderlich [Fri, 11 Mar 2016 13:01:11 +0000 (14:01 +0100)]
batman-adv: fix debuginfo macro style issue

Structure initialization within the macros should follow the general
coding style used in the kernel: put the initialization of the first
variable and the closing brace on a separate line.

Reported-by: Antonio Quartulli <a@unstable.cc>
Signed-off-by: Simon Wunderlich <simon.wunderlich@open-mesh.com>
[sven@narfation.org: fix conflicts with current version]
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: Fix function names on new line starting with '*'
Sven Eckelmann [Sat, 5 Mar 2016 14:56:01 +0000 (15:56 +0100)]
batman-adv: Fix function names on new line starting with '*'

Some really long function names in batman-adv require a newline between
return type and the function name. This has lead to some lines starting
with *batadv_...

This * belongs to the return type and thus should be on the same line as
the return type.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: Add kernel-doc for batadv_interface_rx
Sven Eckelmann [Sun, 28 Feb 2016 10:38:50 +0000 (11:38 +0100)]
batman-adv: Add kernel-doc for batadv_interface_rx

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: Fix kerneldoc for batadv_compare_claim
Sven Eckelmann [Mon, 22 Feb 2016 17:55:32 +0000 (18:55 +0100)]
batman-adv: Fix kerneldoc for batadv_compare_claim

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: Fix checkpatch warning about 'unsigned' type
Sven Eckelmann [Wed, 9 Mar 2016 21:22:51 +0000 (22:22 +0100)]
batman-adv: Fix checkpatch warning about 'unsigned' type

checkpatch.pl warns about the use of 'unsigned' as a short form for
'unsigned int'.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: fix wrong names in kerneldoc
Antonio Quartulli [Fri, 11 Mar 2016 15:36:19 +0000 (16:36 +0100)]
batman-adv: fix wrong names in kerneldoc

Signed-off-by: Antonio Quartulli <a@unstable.cc>
[sven@narfation.org: Fix additional names]
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
8 years agobatman-adv: use to_delayed_work
Geliang Tang [Mon, 28 Dec 2015 15:43:37 +0000 (23:43 +0800)]
batman-adv: use to_delayed_work

Use to_delayed_work() instead of open-coding it.

Signed-off-by: Geliang Tang <geliangtang@163.com>
Reviewed-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: use list_for_each_entry_safe
Geliang Tang [Fri, 18 Dec 2015 15:33:31 +0000 (23:33 +0800)]
batman-adv: use list_for_each_entry_safe

Use list_for_each_entry_safe() instead of list_for_each_safe() to
simplify the code.

Signed-off-by: Geliang Tang <geliangtang@163.com>
Acked-by: Antonio Quartulli <a@unstable.cc>
Reviewed-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agobatman-adv: use static string for table headers
Antonio Quartulli [Sat, 12 Mar 2016 09:30:18 +0000 (10:30 +0100)]
batman-adv: use static string for table headers

Use a static string when showing table headers rather then
a nonsense parametric one with fixed arguments.

It is easier to grep and it does not need to be recomputed
at runtime each time.

Reported-by: Joe Perches <joe@perches.com>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
[sven@narfation.org: fix conflicts with current version]
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
8 years agobatman-adv: Start new development cycle
Simon Wunderlich [Fri, 15 Apr 2016 10:14:39 +0000 (12:14 +0200)]
batman-adv: Start new development cycle

Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agoMAINTAINERS: Add BATMAN ADVANCED documentation files
Sven Eckelmann [Fri, 1 Apr 2016 13:17:36 +0000 (15:17 +0200)]
MAINTAINERS: Add BATMAN ADVANCED documentation files

The sysfs ABI documentation files and the batman-adv.txt are maintained by
the BATMAN ADVANCED maintainers and patches for them should therefore be
sent to them.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agoMAINTAINERS: Mark BATMAN ADVANCED mailing list as moderated
Sven Eckelmann [Fri, 1 Apr 2016 13:17:35 +0000 (15:17 +0200)]
MAINTAINERS: Mark BATMAN ADVANCED mailing list as moderated

The mailing list of b.a.t.m.a.n@lists.open-mesh.org is moderated for
non-subscribers and non-whitelisted addresses. Such mails will be delayed
but the sender will not be informed about the moderation.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
8 years agonet: ethernet: fec_mpc52xx: move to new ethtool api {get|set}_link_ksettings
Philippe Reynes [Sun, 1 May 2016 15:08:11 +0000 (17:08 +0200)]
net: ethernet: fec_mpc52xx: move to new ethtool api {get|set}_link_ksettings

The ethtool api {get|set}_settings is deprecated.
We move the fec_mpc52xx driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: ethernet: fs-enet: move to new ethtool api {get|set}_link_ksettings
Philippe Reynes [Sun, 1 May 2016 15:08:10 +0000 (17:08 +0200)]
net: ethernet: fs-enet: move to new ethtool api {get|set}_link_ksettings

The ethtool api {get|set}_settings is deprecated.
We move the fs-enet driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: ethernet: ucc: move to new ethtool api {get|set}_link_ksettings
Philippe Reynes [Sun, 1 May 2016 15:08:09 +0000 (17:08 +0200)]
net: ethernet: ucc: move to new ethtool api {get|set}_link_ksettings

The ethtool api {get|set}_settings is deprecated.
We move the ucc driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: ethernet: gianfar: move to new ethtool api {get|set}_link_ksettings
Philippe Reynes [Sun, 1 May 2016 15:08:08 +0000 (17:08 +0200)]
net: ethernet: gianfar: move to new ethtool api {get|set}_link_ksettings

The ethtool api {get|set}_settings is deprecated.
We move the gianfar driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoVSOCK: constify vsock_transport structure
Julia Lawall [Sun, 1 May 2016 12:49:15 +0000 (14:49 +0200)]
VSOCK: constify vsock_transport structure

The vsock_transport structure is never modified, so declare it as const.

Done with the help of Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: constify xgene_cle_ops structure
Julia Lawall [Sun, 1 May 2016 12:36:28 +0000 (14:36 +0200)]
drivers: net: xgene: constify xgene_cle_ops structure

The xgene_cle_ops structure is never modified, so declare it as const.

Done with the help of Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Acked-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agofq_codel: add batch ability to fq_codel_drop()
Eric Dumazet [Sun, 1 May 2016 23:47:26 +0000 (16:47 -0700)]
fq_codel: add batch ability to fq_codel_drop()

In presence of inelastic flows and stress, we can call
fq_codel_drop() for every packet entering fq_codel qdisc.

fq_codel_drop() is quite expensive, as it does a linear scan
of 4 KB of memory to find a fat flow.
Once found, it drops the oldest packet of this flow.

Instead of dropping a single packet, try to drop 50% of the backlog
of this fat flow, with a configurable limit of 64 packets per round.

TCA_FQ_CODEL_DROP_BATCH_SIZE is the new attribute to make this
limit configurable.

With this strategy the 4 KB search is amortized to a single cache line
per drop [1], so fq_codel_drop() no longer appears at the top of kernel
profile in presence of few inelastic flows.

[1] Assuming a 64byte cache line, and 1024 buckets

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dave Taht <dave.taht@gmail.com>
Cc: Jonathan Morton <chromatix99@gmail.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Dave Taht
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoravb: Remove rx buffer ALIGN
Kazuya Mizuguchi [Sun, 1 May 2016 15:19:51 +0000 (00:19 +0900)]
ravb: Remove rx buffer ALIGN

Aligning the reception data size is not required.

Signed-off-by: Kazuya Mizuguchi <kazuya.mizuguchi.ks@renesas.com>
Signed-off-by: Yoshihiro Kaneko <ykaneko0929@gmail.com>
Tested-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge tag 'wireless-drivers-next-for-davem-2016-05-02' of git://git.kernel.org/pub...
David S. Miller [Tue, 3 May 2016 04:35:16 +0000 (00:35 -0400)]
Merge tag 'wireless-drivers-next-for-davem-2016-05-02' of git://git./linux/kernel/git/kvalo/wireless-drivers-next

Kalle Valo says:

====================
wireless-drivers patches for 4.7

Major changes:

brcmfmac

* add support for nl80211 BSS_SELECT feature

mwifiex

* add platform specific wakeup interrupt support

ath10k

* implement set_tsf() for 10.2.4 branch
* remove rare MSI range support
* remove deprecated firmware API 1 support

ath9k

* add module parameter to invert LED polarity

wcn36xx

* fixes to get the driver properly working on Dragonboard 410c
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: relax expensive skb_unclone() in iptunnel_handle_offloads()
Eric Dumazet [Sat, 30 Apr 2016 17:19:29 +0000 (10:19 -0700)]
net: relax expensive skb_unclone() in iptunnel_handle_offloads()

Locally generated TCP GSO packets having to go through a GRE/SIT/IPIP
tunnel have to go through an expensive skb_unclone()

Reallocating skb->head is a lot of work.

Test should really check if a 'real clone' of the packet was done.

TCP does not care if the original gso_type is changed while the packet
travels in the stack.

This adds skb_header_unclone() which is a variant of skb_clone()
using skb_header_cloned() check instead of skb_cloned().

This variant can probably be used from other points.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonetdevice: shrink size of struct netdev_queue
Florian Westphal [Tue, 3 May 2016 01:29:09 +0000 (03:29 +0200)]
netdevice: shrink size of struct netdev_queue

- trans_timeout is incremented when tx queue timed out (tx watchdog).
- tx_maxrate is set via sysfs

Moving tx_maxrate to read-mostly part shrinks the struct by 64 bytes.
While at it, also move trans_timeout (it is out-of-place in the
'write-mostly' part).

Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'bridge-per-vlan-stats'
David S. Miller [Tue, 3 May 2016 02:27:06 +0000 (22:27 -0400)]
Merge branch 'bridge-per-vlan-stats'

Nikolay Aleksandrov says:

====================
bridge: per-vlan stats

This set adds support for bridge per-vlan statistics.
In order to be able to dump statistics for many vlans we need a way to
continue dumping after reaching maximum size, thus patches 01 and 02 extend
the new stats API with a per-device extended link stats attribute and
callback which can save its local state and continue where it left off
afterwards. I considered using the already existing "fill_xstats" callback
but it gets confusing since we need to separate the linkinfo dump from the
new stats api dump and adding a flag/argument to do that just looks messy.
I don't think the rtnl_link_ops size is an issue, so adding these seemed
like the cleaner approach.

Patches 03 and 04 add the stats support and netlink dump support
respectively. The stats accounting is controlled via a bridge option which
is default off, thus the performance impact is kept minimal.
I've tested this set with both old and modified iproute2, kmemleak on and
some traffic stress tests while adding/removing vlans and ports.

v3:
 - drop the RCU pvid patch and remove one pointer fetch as requested
 - make stats accounting optional with default to off, the option is in the
   same cache line as vlan_proto and vlan_enabled, so it is already fetched
   before the fast path check thus the performance impact is minimal, this
   also allows us to avoid one vlan lookup and return early when using pvid
 - rebased and retested

v2:
 - Improve the error checking, rename lidx to prividx and save the current
   idx user instead of restricting it to one in patch 01
 - squash patch 02 into 01 and remove the restriction
 - add callback descriptions, improve the size calculation and change the
   xstats message structure to have an embedding level per rtnl link type
   so we can avoid one call to get the link type (and thus filter on it)
   and also each link type can now have any number of private attributes
   inside
 - fix a problem where the vlan stats are not dumped if the bridge has 0
   vlans on it but has vlans on the ports, add bridge link type private
   attributes and also add paddings for future extensions to avoid at least
   a few netlink attributes and improve struct alignment
 - drop the is_skb_forwardable argument constifying patch as it's not
   needed anymore, but it's a nice cleanup which I'll send separately
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobridge: netlink: export per-vlan stats
Nikolay Aleksandrov [Sat, 30 Apr 2016 08:25:29 +0000 (10:25 +0200)]
bridge: netlink: export per-vlan stats

Add a new LINK_XSTATS_TYPE_BRIDGE attribute and implement the
RTM_GETSTATS callbacks for IFLA_STATS_LINK_XSTATS (fill_linkxstats and
get_linkxstats_size) in order to export the per-vlan stats.
The paddings were added because soon these fields will be needed for
per-port per-vlan stats (or something else if someone beats me to it) so
avoiding at least a few more netlink attributes.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobridge: vlan: learn to count
Nikolay Aleksandrov [Sat, 30 Apr 2016 08:25:28 +0000 (10:25 +0200)]
bridge: vlan: learn to count

Add support for per-VLAN Tx/Rx statistics. Every global vlan context gets
allocated a per-cpu stats which is then set in each per-port vlan context
for quick access. The br_allowed_ingress() common function is used to
account for Rx packets and the br_handle_vlan() common function is used
to account for Tx packets. Stats accounting is performed only if the
bridge-wide vlan_stats_enabled option is set either via sysfs or netlink.
A struct hole between vlan_enabled and vlan_proto is used for the new
option so it is in the same cache line. Currently it is binary (on/off)
but it is intentionally restricted to exactly 0 and 1 since other values
will be used in the future for different purposes (e.g. per-port stats).

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: rtnetlink: add linkxstats callbacks and attribute
Nikolay Aleksandrov [Sat, 30 Apr 2016 08:25:27 +0000 (10:25 +0200)]
net: rtnetlink: add linkxstats callbacks and attribute

Add callbacks to calculate the size and fill link extended statistics
which can be split into multiple messages and are dumped via the new
rtnl stats API (RTM_GETSTATS) with the IFLA_STATS_LINK_XSTATS attribute.
Also add that attribute to the idx mask check since it is expected to
be able to save state and resume dumping (e.g. future bridge per-vlan
stats will be dumped via this attribute and callbacks).
Each link type should nest its private attributes under the per-link type
attribute. This allows to have any number of separated private attributes
and to avoid one call to get the dev link type.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: rtnetlink: allow rtnl_fill_statsinfo to save private state counter
Nikolay Aleksandrov [Sat, 30 Apr 2016 08:25:26 +0000 (10:25 +0200)]
net: rtnetlink: allow rtnl_fill_statsinfo to save private state counter

The new prividx argument allows the current dumping device to save a
private state counter which would enable it to continue dumping from
where it left off. And the idxattr is used to save the current idx user
so multiple prividx using attributes can be requested at the same time
as suggested by Roopa Prabhu.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'ipv6-tunnel-cleanups'
David S. Miller [Mon, 2 May 2016 23:23:32 +0000 (19:23 -0400)]
Merge branch 'ipv6-tunnel-cleanups'

Tom Herbert says:

====================
net: Cleanup IPv6 ip tunnels

The IPv6 tunnel code is very different from IPv4 code. There is a lot
of redundancy with the IPv4 code, particularly in the GRE tunneling.

This patch set cleans up the tunnel code to make the IPv6 code look
more like the IPv4 code and use common functions between the two
stacks where possible.

This work should make it easier to maintain and extend the IPv6 ip
tunnels.

Items in this patch set:
  - Cleanup IPv6 tunnel receive path (ip6_tnl_rcv). Includes using
    gro_cells and exporting ip6_tnl_rcv so the ip6_gre can call it
  - Move GRE functions to common header file (tx functions) or
    gre_demux.c (rx functions like gre_parse_header)
  - Call common GRE functions from IPv6 GRE
  - Create ip6_tnl_xmit (to be like ip_tunnel_xmit)

Tested:
  Ran super_netperf tests for TCP_RR and TCP_STREAM for:
    - IPv4 over gre, gretap, gre6, gre6tap
    - IPv6 over gre, gretap, gre6, gre6tap
    - ipip
    - ip6ip6
    - ipip/gue
    - IPv6 over gre/gue
    - IPv4 over gre/gue
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agogre6: Cleanup GREv6 transmit path, call common GRE functions
Tom Herbert [Sat, 30 Apr 2016 00:12:21 +0000 (17:12 -0700)]
gre6: Cleanup GREv6 transmit path, call common GRE functions

Changes in GREv6 transmit path:
  - Call gre_checksum, remove gre6_checksum
  - Rename ip6gre_xmit2 to __gre6_xmit
  - Call gre_build_header utility function
  - Call ip6_tnl_xmit common function
  - Call ip6_tnl_change_mtu, eliminate ip6gre_tunnel_change_mtu

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv6: Generic tunnel cleanup
Tom Herbert [Sat, 30 Apr 2016 00:12:20 +0000 (17:12 -0700)]
ipv6: Generic tunnel cleanup

A few generic changes to generalize tunnels in IPv6:
  - Export ip6_tnl_change_mtu so that it can be called by ip6_gre
  - Add tun_hlen to ip6_tnl structure.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agogre: Create common functions for transmit
Tom Herbert [Sat, 30 Apr 2016 00:12:19 +0000 (17:12 -0700)]
gre: Create common functions for transmit

Create common functions for both IPv4 and IPv6 GRE in transmit. These
are put into gre.h.

Common functions are for:
  - GRE checksum calculation. Move gre_checksum to gre.h.
  - Building a GRE header. Move GRE build_header and rename
    gre_build_header.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv6: Create ip6_tnl_xmit
Tom Herbert [Sat, 30 Apr 2016 00:12:18 +0000 (17:12 -0700)]
ipv6: Create ip6_tnl_xmit

This patch renames ip6_tnl_xmit2 to ip6_tnl_xmit and exports it. Other
users like GRE will be able to call this. The original ip6_tnl_xmit
function is renamed to ip6_tnl_start_xmit (this is an ndo_start_xmit
function).

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agogre6: Cleanup GREv6 receive path, call common GRE functions
Tom Herbert [Sat, 30 Apr 2016 00:12:17 +0000 (17:12 -0700)]
gre6: Cleanup GREv6 receive path, call common GRE functions

- Create gre_rcv function. This calls gre_parse_header and ip6gre_rcv.
  - Call ip6_tnl_rcv. Doing this and using gre_parse_header eliminates
    most of the code in ip6gre_rcv.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agogre: Move utility functions to common headers
Tom Herbert [Sat, 30 Apr 2016 00:12:16 +0000 (17:12 -0700)]
gre: Move utility functions to common headers

Several of the GRE functions defined in net/ipv4/ip_gre.c are usable
for IPv6 GRE implementation (that is they are protocol agnostic).

These include:
  - GRE flag handling functions are move to gre.h
  - GRE build_header is moved to gre.h and renamed gre_build_header
  - parse_gre_header is moved to gre_demux.c and renamed gre_parse_header
  - iptunnel_pull_header is taken out of gre_parse_header. This is now
    done by caller. The header length is returned from gre_parse_header
    in an int* argument.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv6: Cleanup IPv6 tunnel receive path
Tom Herbert [Sat, 30 Apr 2016 00:12:15 +0000 (17:12 -0700)]
ipv6: Cleanup IPv6 tunnel receive path

Some basic changes to make IPv6 tunnel receive path look more like
IPv4 path:
  - Make ip6_tnl_rcv non-static so that GREv6 and others can call it
  - Make ip6_tnl_rcv look like ip_tunnel_rcv
  - Switch to gro_cells_receive
  - Make ip6_tnl_rcv non-static and export it

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'tcp-preempt'
David S. Miller [Mon, 2 May 2016 21:02:26 +0000 (17:02 -0400)]
Merge branch 'tcp-preempt'

Eric Dumazet says:

====================
net: make TCP preemptible

Most of TCP stack assumed it was running from BH handler.

This is great for most things, as TCP behavior is very sensitive
to scheduling artifacts.

However, the prequeue and backlog processing are problematic,
as they need to be flushed with BH being blocked.

To cope with modern needs, TCP sockets have big sk_rcvbuf values,
in the order of 16 MB, and soon 32 MB.
This means that backlog can hold thousands of packets, and things
like TCP coalescing or collapsing on this amount of packets can
lead to insane latency spikes, since BH are blocked for too long.

It is time to make UDP/TCP stacks preemptible.

Note that fast path still runs from BH handler.

v2: Added "tcp: make tcp_sendmsg() aware of socket backlog"
    to reduce latency problems of large sends.

v3: Fixed a typo in tcp_cdg.c
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agotcp: make tcp_sendmsg() aware of socket backlog
Eric Dumazet [Fri, 29 Apr 2016 21:16:53 +0000 (14:16 -0700)]
tcp: make tcp_sendmsg() aware of socket backlog

Large sendmsg()/write() hold socket lock for the duration of the call,
unless sk->sk_sndbuf limit is hit. This is bad because incoming packets
are parked into socket backlog for a long time.
Critical decisions like fast retransmit might be delayed.
Receivers have to maintain a big out of order queue with additional cpu
overhead, and also possible stalls in TX once windows are full.

Bidirectional flows are particularly hurt since the backlog can become
quite big if the copy from user space triggers IO (page faults)

Some applications learnt to use sendmsg() (or sendmmsg()) with small
chunks to avoid this issue.

Kernel should know better, right ?

Add a generic sk_flush_backlog() helper and use it right
before a new skb is allocated. Typically we put 64KB of payload
per skb (unless MSG_EOR is requested) and checking socket backlog
every 64KB gives good results.

As a matter of fact, tests with TSO/GSO disabled give very nice
results, as we manage to keep a small write queue and smaller
perceived rtt.

Note that sk_flush_backlog() maintains socket ownership,
so is not equivalent to a {release_sock(sk); lock_sock(sk);},
to ensure implicit atomicity rules that sendmsg() was
giving to (possibly buggy) applications.

In this simple implementation, I chose to not call tcp_release_cb(),
but we might consider this later.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: do not block BH while processing socket backlog
Eric Dumazet [Fri, 29 Apr 2016 21:16:52 +0000 (14:16 -0700)]
net: do not block BH while processing socket backlog

Socket backlog processing is a major latency source.

With current TCP socket sk_rcvbuf limits, I have sampled __release_sock()
holding cpu for more than 5 ms, and packets being dropped by the NIC
once ring buffer is filled.

All users are now ready to be called from process context,
we can unblock BH and let interrupts be serviced faster.

cond_resched_softirq() could be removed, as it has no more user.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosctp: prepare for socket backlog behavior change
Eric Dumazet [Fri, 29 Apr 2016 21:16:51 +0000 (14:16 -0700)]
sctp: prepare for socket backlog behavior change

sctp_inq_push() will soon be called without BH being blocked
when generic socket code flushes the socket backlog.

It is very possible SCTP can be converted to not rely on BH,
but this needs to be done by SCTP experts.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoudp: prepare for non BH masking at backlog processing
Eric Dumazet [Fri, 29 Apr 2016 21:16:50 +0000 (14:16 -0700)]
udp: prepare for non BH masking at backlog processing

UDP uses the generic socket backlog code, and this will soon
be changed to not disable BH when protocol is called back.

We need to use appropriate SNMP accessors.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodccp: do not assume DCCP code is non preemptible
Eric Dumazet [Fri, 29 Apr 2016 21:16:49 +0000 (14:16 -0700)]
dccp: do not assume DCCP code is non preemptible

DCCP uses the generic backlog code, and this will soon
be changed to not disable BH when protocol is called back.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agotcp: do not block bh during prequeue processing
Eric Dumazet [Fri, 29 Apr 2016 21:16:48 +0000 (14:16 -0700)]
tcp: do not block bh during prequeue processing

AFAIK, nothing in current TCP stack absolutely wants BH
being disabled once socket is owned by a thread running in
process context.

As mentioned in my prior patch ("tcp: give prequeue mode some care"),
processing a batch of packets might take time, better not block BH
at all.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agotcp: do not assume TCP code is non preemptible
Eric Dumazet [Fri, 29 Apr 2016 21:16:47 +0000 (14:16 -0700)]
tcp: do not assume TCP code is non preemptible

We want to to make TCP stack preemptible, as draining prequeue
and backlog queues can take lot of time.

Many SNMP updates were assuming that BH (and preemption) was disabled.

Need to convert some __NET_INC_STATS() calls to NET_INC_STATS()
and some __TCP_INC_STATS() to TCP_INC_STATS()

Before using this_cpu_ptr(net->ipv4.tcp_sk) in tcp_v4_send_reset()
and tcp_v4_send_ack(), we add an explicit preempt disabled section.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'xgene-channel-number'
David S. Miller [Mon, 2 May 2016 20:47:55 +0000 (16:47 -0400)]
Merge branch 'xgene-channel-number'

Iyappan Subramanian says:

====================
drivers: net: xgene: fix: Get channel number from device binding

This patch set adds 'channel' property to get ethernet to CPU channel number,
thus decoupling the Linux driver from static resource selection.

v2: Address review comments from v1
- removed irq reference from Linux driver
- added 'channel' property to get ethernet to CPU channel number

v1:
- Initial version
====================

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodtb: xgene: Add channel property
Iyappan Subramanian [Fri, 29 Apr 2016 18:10:15 +0000 (11:10 -0700)]
dtb: xgene: Add channel property

Added 'channel' property, describing ethernet to CPU channel number.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoDocumentation: dtb: xgene: Add channel property
Iyappan Subramanian [Fri, 29 Apr 2016 18:10:14 +0000 (11:10 -0700)]
Documentation: dtb: xgene: Add channel property

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: Get channel number from device binding
Iyappan Subramanian [Fri, 29 Apr 2016 18:10:13 +0000 (11:10 -0700)]
drivers: net: xgene: Get channel number from device binding

This patch gets ethernet to CPU channel (prefetch buffer number) from
the newly added 'channel' property, thus decoupling Linux driver from
resource management.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'qed-selftests'
David S. Miller [Mon, 2 May 2016 04:16:45 +0000 (00:16 -0400)]
Merge branch 'qed-selftests'

Sudarsana Reddy Kalluru says:

====================
qed/qede: ethtool selftests support.

This series adds the driver support for following selftests:
1. Register test
2. Memory test
3. Clock test
4. Interrupt test
5. Internal loopback test
Patch (1) adds the qed driver infrastructure for selftests. Patches (2) and
(3) add qede driver support for ethtool selftests.

Please consider applying this series to "net-next".
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoqede: add implementation for internal loopback test.
Sudarsana Reddy Kalluru [Fri, 29 Apr 2016 00:20:54 +0000 (20:20 -0400)]
qede: add implementation for internal loopback test.

This patch adds the qede implementation for internal loopback test.

Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoqede: add support for selftests.
Sudarsana Reddy Kalluru [Fri, 29 Apr 2016 00:20:53 +0000 (20:20 -0400)]
qede: add support for selftests.

This patch adds the qede ethtool support for the following tests:
- interrupt test
- memory test
- register test
- clock test

Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoqed: add infrastructure for device self tests.
Sudarsana Reddy Kalluru [Fri, 29 Apr 2016 00:20:52 +0000 (20:20 -0400)]
qed: add infrastructure for device self tests.

This patch adds the functionality and APIs needed for selftests.
It adds the ability to configure the link-mode which is required for the
implementation of loopback tests. It adds the APIs for clock test,
register test, interrupt test and memory test.

Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: dsa: mv88e6xxx: replace ds with ps where possible
Andrew Lunn [Fri, 29 Apr 2016 01:24:06 +0000 (21:24 -0400)]
net: dsa: mv88e6xxx: replace ds with ps where possible

The dsa_switch structure ds is actually needed in very few places,
mostly during setup of the switch. The private structure ps is however
needed nearly everywhere. Pass ps, not ds internally.

[vd: rebased Andrew's patch.]

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next...
David S. Miller [Mon, 2 May 2016 03:38:49 +0000 (23:38 -0400)]
Merge branch '40GbE' of git://git./linux/kernel/git/jkirsher/next-queue

Jeff Kirsher says:

====================
40GbE Intel Wired LAN Driver Updates 2016-05-01

This series contains updates to i40e and i40evf.

The theme of this series is code reduction, with several code cleanups in
this series.  Starting with Neerav's removal of the code that implemented
the HMC AQ APIs and calls, since they are now obsolete and not supported
by firmware.

Anjali changes the default of VFs to make sure they are not trusted or
privileged until its explicitly set for trust through the new NDO op
interface.  Also limited the number of MAC and VLAN addresses a VF can
add if it is untrusted/privileged.

Carolyn syncs the VF code for the changes made to the PF for the RSS
hash tuple settings, which ends up cleaning up much of the existing code.

Jesse cleans up compiler warnings which were found with gcc's W=2 option.
Then removed duplicate code, especially since only one copy was actually
being used.

Jacob addresses an issue which was found when testing GCC 6's which
happens to produce new warnings when you left shift a signed value
beyond the storage sizeof the type.  The converts i40e & i40evf to use
the BIT() macro more consistently.

Alex actually bucks the trend of code removal by adding support for
both drivers to use GSO_PARTIAL so that segmentation of frames with
checksums enabled in outer headers is supported.  Fortunately it does
not take much to add this support!
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosctp: signal sk_data_ready earlier on data chunks reception
Marcelo Ricardo Leitner [Fri, 29 Apr 2016 17:17:08 +0000 (14:17 -0300)]
sctp: signal sk_data_ready earlier on data chunks reception

Dave Miller pointed out that fb586f25300f ("sctp: delay calls to
sk_data_ready() as much as possible") may insert latency specially if
the receiving application is running on another CPU and that it would be
better if we signalled as early as possible.

This patch thus basically inverts the logic on fb586f25300f and signals
it as early as possible, similar to what we had before.

Fixes: fb586f25300f ("sctp: delay calls to sk_data_ready() as much as possible")
Reported-by: Dave Miller <davem@davemloft.net>
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomdio_bus: Fix MDIO bus scanning in __mdiobus_register()
Marek Vasut [Mon, 2 May 2016 00:47:31 +0000 (02:47 +0200)]
mdio_bus: Fix MDIO bus scanning in __mdiobus_register()

Since commit b74766a0a0fe ("phylib: don't return NULL
from get_phy_device()") in linux-next, phy_get_device() will return
ERR_PTR(-ENODEV) instead of NULL if the PHY device ID is all ones.

This causes problem with stmmac driver and likely some other drivers
which call mdiobus_register(). I triggered this bug on SoCFPGA MCVEVK
board with linux-next 20160427 and 20160428. In case of the stmmac, if
there is no PHY node specified in the DT for the stmmac block, the stmmac
driver ( drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c function
stmmac_mdio_register() ) will call mdiobus_register() , which will
register the MDIO bus and probe for the PHY.

The mdiobus_register() resp. __mdiobus_register() iterates over all of
the addresses on the MDIO bus and calls mdiobus_scan() for each of them,
which invokes get_phy_device(). Before the aforementioned patch, the
mdiobus_scan() would return NULL if no PHY was found on a given address
and mdiobus_register() would continue and try the next PHY address. Now,
mdiobus_scan() returns ERR_PTR(-ENODEV), which is caught by the
'if (IS_ERR(phydev))' condition and the loop exits immediately if the
PHY address does not contain PHY.

Repair this by explicitly checking for the ERR_PTR(-ENODEV) and if this
error comes around, continue with the next PHY address.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: David S. Miller <davem@davemloft.net>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoi40e/i40evf: Add support for GSO partial with UDP_TUNNEL_CSUM and GRE_CSUM
Alexander Duyck [Thu, 14 Apr 2016 21:19:25 +0000 (17:19 -0400)]
i40e/i40evf: Add support for GSO partial with UDP_TUNNEL_CSUM and GRE_CSUM

This patch makes it so that i40e and i40evf can use GSO_PARTIAL to support
segmentation for frames with checksums enabled in outer headers.  As a
result we can now send data over these types of tunnels at over 20Gb/s
versus the 12Gb/s that was previously possible on my system.

The advantage with the i40e parts is that this offload is mostly
transparent as the hardware still deals with the inner and/or outer IPv4
headers so the IP ID is still incrementing for both when this offload is
performed.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40evf: make use of BIT() macro to avoid signed left shift
Jacob Keller [Wed, 13 Apr 2016 23:08:27 +0000 (16:08 -0700)]
i40evf: make use of BIT() macro to avoid signed left shift

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: make use of BIT() macro to prevent left shift of signed values
Jacob Keller [Wed, 13 Apr 2016 23:08:26 +0000 (16:08 -0700)]
i40e: make use of BIT() macro to prevent left shift of signed values

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e/i40evf: fix I40E_MASK signed shift overflow warnings
Jacob Keller [Wed, 13 Apr 2016 23:08:25 +0000 (16:08 -0700)]
i40e/i40evf: fix I40E_MASK signed shift overflow warnings

GCC 6 has a new warning which will display when you attempt to left
shift a signed value beyond the storage size of the type. I40E_MASK
generates a mask value for 32bit registers. Properly typecast the mask
value and place the values in parenthesis to prevent macro expansion
issues.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e/i40evf : Bump driver version from 1.5.5 to 1.5.10
Harshitha Ramamurthy [Wed, 13 Apr 2016 10:08:32 +0000 (03:08 -0700)]
i40e/i40evf : Bump driver version from 1.5.5 to 1.5.10

Signed-off-by: Harshitha Ramamurthy <harshitha.ramamurthy@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Update device ids for X722
Catherine Sullivan [Wed, 13 Apr 2016 10:08:31 +0000 (03:08 -0700)]
i40e: Update device ids for X722

Add a device ID for X722.

Change-Id: I574f2345ab341de98a6a1c212d0603af853e48b0
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Drop extra copy of function
Jesse Brandeburg [Wed, 13 Apr 2016 10:08:30 +0000 (03:08 -0700)]
i40e: Drop extra copy of function

i40e_release_rx_desc was in two files, but was only used
and needed in txrx.c.  Get rid of the extra copy.

Change-Id: I86e18239aa03531fc198b6c052847475084a9200
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Use consistent type for vf_id
Jesse Brandeburg [Wed, 13 Apr 2016 10:08:29 +0000 (03:08 -0700)]
i40e: Use consistent type for vf_id

The driver was all over the place using signed or unsigned types
for vf_id, when it should always be signed.

This fixes warnings of type unsafe comparisons from gcc with W=2.

Change-Id: I2cb681f83d0f68ca124d2e4131e4ac0d9f8a6b22
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: PTP - avoid aggregate return warnings
Jesse Brandeburg [Wed, 13 Apr 2016 10:08:28 +0000 (03:08 -0700)]
i40e: PTP - avoid aggregate return warnings

Aggregate return warnings are when struct types are returned
and must be copied to the lvalue with a struct copy by the compiler.

This fixes warnings of type aggregate-return from gcc with W=2.

Change-Id: I896b1bf514544bf0faeb458869d79914b9f1b168
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Fix uninitialized variable
Catherine Sullivan [Wed, 13 Apr 2016 10:08:27 +0000 (03:08 -0700)]
i40e: Fix uninitialized variable

We have an uninitialized variable warning for valid_len for one case in
validate_vf_mesg. To fix this, just initialize it to 0 at the top of the
function and remove all of the now redundant assignments to 0 in the
individual cases.

Change-Id: Iacbd97f4c521ed8d662eef803a598d8707708cfd
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40evf: RSS Hash Option parameters
Carolyn Wyborny [Wed, 13 Apr 2016 10:08:26 +0000 (03:08 -0700)]
i40evf: RSS Hash Option parameters

This patch syncs the VF code for the changes made to the PF for the RSS
hash tuple settings.  Since the VF still cannot change the RSS hash
settings, change the code to make this clear to the user.  Previously,
the default settings were returned in this function. However, the
default can be changed by the PF so this does not make sense anymore.

Change-Id: I085eaf005fc7978b440d2a1bf2b2dd7cadaff39b
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Remove HMC AQ API implementation
Neerav Parikh [Sun, 1 May 2016 21:29:53 +0000 (14:29 -0700)]
i40e: Remove HMC AQ API implementation

Remove the code that implements the HMC AQ APIs and call these APIs.
This is done because these are obsolete APIs and are not supported
by firmware.

Change-ID: I5d771d8f37c3e16e7b0a972ff9b27e75aa2d05d4
Signed-off-by: Neerav Parikh <neerav.parikh@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Prevent falling to promiscuous if the VF is not trusted
Anjali Singhai Jain [Wed, 13 Apr 2016 10:08:23 +0000 (03:08 -0700)]
i40e: Prevent falling to promiscuous if the VF is not trusted

With this change a non trusted VF can never fall to promiscuous
mode when there is no room for a MAC/VLAN filter.

Change-Id: I8a155aa25c0bcdc6093414920c9ade4ee0bd20e8
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Limit the number of MAC and VLAN addresses that can be added for VFs
Anjali Singhai Jain [Wed, 13 Apr 2016 10:08:22 +0000 (03:08 -0700)]
i40e: Limit the number of MAC and VLAN addresses that can be added for VFs

If the VF is privileged/trusted it can do as it may please including
but not limited to hogging resources and playing unfair.
But if the VF is not privileged/trusted it still can add some number
(8) of MAC and VLAN addresses.
Other restrictions with respect to Port VLAN and normal VLAN still apply
to not privileged/trusted VF.

Change-Id: I3a9529201b184c8873e1ad2e300aff468c9e6296
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agotipc: set 'active' state correctly for first established link
Jon Paul Maloy [Fri, 29 Apr 2016 00:16:08 +0000 (20:16 -0400)]
tipc: set 'active' state correctly for first established link

When we are displaying statistics for the first link established between
two peers, it will always be presented as STANDBY although it in reality
is ACTIVE.

This happens because we forget to set the 'active' flag in the link
instance at the moment it is established. Although this is a bug, it only
has impact on the presentation view of the link, not on its actual
functionality.

Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoof: of_mdio: Check if MDIO bus controller is available
Florian Fainelli [Thu, 28 Apr 2016 23:24:12 +0000 (16:24 -0700)]
of: of_mdio: Check if MDIO bus controller is available

Add a check whether the 'struct device_node' pointer passed to
of_mdiobus_register() is an available (aka enabled) node in the Device
Tree.

Rationale for doing this are cases where an Ethernet MAC provides a MDIO
bus controller and node, and an additional Ethernet MAC might be
connecting its PHY/switches to that first MDIO bus controller, while
still embedding one internally which is therefore marked as "disabled".

Instead of sprinkling checks like these in callers of
of_mdiobus_register(), do this in a central location.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoi40e: Change the default for VFs to be not privileged
Anjali Singhai Jain [Wed, 13 Apr 2016 10:08:21 +0000 (03:08 -0700)]
i40e: Change the default for VFs to be not privileged

Make sure a VF is not trusted/privileged until its explicitly
set for trust through the new NDO op interface.

Change-Id: I476385c290d2b4901d8fceb29de43546accdc499
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoMerge branch 'mlx5-aRFS'
David S. Miller [Fri, 29 Apr 2016 20:29:12 +0000 (16:29 -0400)]
Merge branch 'mlx5-aRFS'

Saeed Mahameed says:

====================
Mellanox 100G mlx5 ethernet aRFS support

This series adds accelerated RFS support for the mlx5e driver.
I have added one patch non-related to aRFS that fixes the rtnl_lock
warning mlx5 driver been getting since b7aade15485a ('vxlan: break dependency with netdev drivers')

aRFS support in details:

A direct TIR per RQ is now required in order to have the essential building blocks
for aRFS.  Today the driver has one direct TIR that forwards traffic to RQ[0] (core 0),
and one indirect TIR for RSS indirection table.  For that we've added one direct TIR
per RQ, e.g.: TIR[i] -> RQ[i] (core i).

Publicize Modify flow rule destination and reveal it in flow steering API, to have the
ability to dynamically modify the destination TIR(core) for aRFS rules from the
ethernet driver.

Initializing CPU reverse mapping to notify upper layer on internal receive queue cpu
mappings.

Some design refactoring for mlx5e ethernet driver flow tables and flow steering API.
Now the caller of create_flow_table can choose the level of the flow table, this way
we will create the mlx5e flow tables in a reversed order and connect them as we go,
we create flow table[i+1] before flow table[i] to be able to set flow table[i + 1] as
a destination of flow table[i] once flow table[i] is created.
also we have split the main flow table in the following manner:
    - From before: RX packet had to visit two flow tables until it is delivered to its receive queue:
        RX packet -> vlan filter flow table -> main flow table.
        > vlan filter will check the packet vlan field is allowed.
        > main flow will check if the dest mac is allowed and will check the l3/l4 headers to
        retrieve the RSS hash for steering the packet into its final receive queue.

    - Now main flow table is split into l2 dst mac steering table and ttc (traffic type classifier) table:
        RX packet -> vlan filter -> l2 table -> ttc table
        > vlan filter - same as before
        > L2 filter - filter packets according their destination mac address
        > ttc table - classify packet headers for RSS steering
            - L3/L4 classification rules to steer the packet according to thier headers hash
            - in case of none of the rules applies the packet is steered to RQ[0]

After the above refactoring all left to-do is to create aRFS flow table which will manage
aRFS steering rules to forward traffic to the desired RQ (core) and just connect the ttc
table rules destinations to aRFS flow table.

aRFS flow table in case of a miss will deliver the traffic to the core where the original
ttc hash would have chosen.

TTC table is not initialized and enabled until the user explicitly asks to, i.e. setting the NETIF_F_NTUPLE
to ON.  This way there is no need for ttc table to forward traffic to aRFS table unless required.
When setting back to OFF aRFS flow table is disabled and disconnected.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Enabling aRFS mechanism
Maor Gottlieb [Thu, 28 Apr 2016 22:36:42 +0000 (01:36 +0300)]
net/mlx5e: Enabling aRFS mechanism

Accelerated RFS requires that ntuple filtering is enabled via
ethtool and driver supports ndo_rx_flow_steer.
When the ntuple filtering is enabled, we modify the l3_l4 ttc
rules to point on the aRFS flow tables and when the filtering
is disabled, we modify the l3_l4 ttc rules to point on the RSS
TIRs.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Add accelerated RFS support
Maor Gottlieb [Thu, 28 Apr 2016 22:36:41 +0000 (01:36 +0300)]
net/mlx5e: Add accelerated RFS support

Implement ndo_rx_flow_steer ndo.
A new flow steering rule will be composed from the
skb 4-tuple and added to the hardware aRFS flow table.

Each rule is stored in an internal hash table, if such
skb 4-tuple rule already exists we update the corresponding
hardware steering rule with the new destination.

For garbage collection rps_may_expire_flow will be
invoked for a limited amount of old rules upon any
ndo_rx_flow_steer invocation.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Create aRFS flow tables
Maor Gottlieb [Thu, 28 Apr 2016 22:36:40 +0000 (01:36 +0300)]
net/mlx5e: Create aRFS flow tables

Create the following four flow tables for aRFS usage:
1. IPv4 TCP - filtering 4-tuple of IPv4 TCP packets.
2. IPv6 TCP - filtering 4-tuple of IPv6 TCP packets.
3. IPv4 UDP - filtering 4-tuple of IPv4 UDP packets.
4. IPv6 UDP - filtering 4-tuple of IPv6 UDP packets.

Each flow table has two flow groups: one for the 4-tuple
filtering (full match)  and the other contains * rule for miss rule.

Full match rule means a hit for aRFS and packet will be forwarded
to the dedicated RQ/Core, miss rule packets will be forwarded to
default RSS hashing.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5: Initializing CPU reverse mapping
Maor Gottlieb [Thu, 28 Apr 2016 22:36:39 +0000 (01:36 +0300)]
net/mlx5: Initializing CPU reverse mapping

Allocating CPU rmap and add entry for each IRQ.
CPU rmap is used in aRFS to get the RX queue number
of the RX completion interrupts.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Split the main flow steering table
Maor Gottlieb [Thu, 28 Apr 2016 22:36:38 +0000 (01:36 +0300)]
net/mlx5e: Split the main flow steering table

Currently, the main flow table is used for two purposes:
One is to do mac filtering and the other is to classify
the packet l3-l4 header in order to steer the packet to
the right RSS TIR.

This design is very complex, for each configured mac address we
have to add eleven rules (rule for each traffic type), the same if the
device is put to promiscuous/allmulti mode.
This scheme isn't scalable for future features like aRFS.

In order to simplify it, the main flow table is split to two flow
tables:
1. l2 table - filter the packet dmac address, if there is a match
we forward to the ttc flow table.

2. TTC (Traffic Type Classifier) table - classify the traffic
type of the packet and steer the packet to the right TIR.

In this new design, when new mac address is added, the driver adds
only one flow rule instead of eleven.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Refactor mlx5e flow steering structs
Maor Gottlieb [Thu, 28 Apr 2016 22:36:37 +0000 (01:36 +0300)]
net/mlx5e: Refactor mlx5e flow steering structs

Slightly refactor and re-order the flow steering structs,
tables and data-bases for better self-containment and
flexibility to add more future steering phases
(tables/rules/data bases) e.g: aRFS.

Changes:
1. Move the vlan DB and address DB into their table structs.
2. Rename steering table structs to unique format: mlx5e_*_table,
e.g: mlx5e_vlan_table.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5: Support different attributes for priorities in namespace
Maor Gottlieb [Thu, 28 Apr 2016 22:36:36 +0000 (01:36 +0300)]
net/mlx5: Support different attributes for priorities in namespace

Currently, namespace could be initialized only
with priorities with the same attributes.
Add support to initialize namespace with priorities
with different attributes(e.g. different number of levels).

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5: Add user chosen levels when allocating flow tables
Maor Gottlieb [Thu, 28 Apr 2016 22:36:35 +0000 (01:36 +0300)]
net/mlx5: Add user chosen levels when allocating flow tables

Currently, consumers of the flow steering infrastructure can't
choose their own flow table levels and are limited to one
flow table per level. This just waste levels.
Instead, we introduce here the possibility to use multiple
flow tables in a level. The user is free to connect these
flow tables, while following the rule (FTEs in FT of level x
could only point to FTs of level y where y > x).

In addition this patch switch the order of the create/destroy
flow tables of the NIC(vlan and main).

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5: Set number of allowed levels in priority
Maor Gottlieb [Thu, 28 Apr 2016 22:36:34 +0000 (01:36 +0300)]
net/mlx5: Set number of allowed levels in priority

Refactors the flow steering namespace creation,
by changing the name num_fts to num_levels.
When new flow table is created, the driver assign new level
to this flow table therefore the meaning is equivalent.
Since downstream patches will introduce the ability to create more
than one flow table per level, the name num_fts is no
longer accurate.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5: Introduce modify flow rule destination
Maor Gottlieb [Thu, 28 Apr 2016 22:36:33 +0000 (01:36 +0300)]
net/mlx5: Introduce modify flow rule destination

This API is used for modifying the flow rule destination.
This is needed for modifying the pointed flow table by the
traffic type classifier rules to point on the aRFS tables.

Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Direct TIR per RQ
Tariq Toukan [Thu, 28 Apr 2016 22:36:32 +0000 (01:36 +0300)]
net/mlx5e: Direct TIR per RQ

Introduce new TIRs for direct access per RQ.
Now we have 2 available kinds of TIRs:
- indirect TIR per traffic type, each points to one RQT (RSS RQT)
          same as before.
- New direct TIR per RQ, each points to RQT with a size of one
          that forwards packets to that RQ only.

Driver will open max channels (num cores) direct TIRs by default,
they will be filled with the actual RQs once channels are allocated.

Needed for downstream aRFS and ethtool direct steering functionalities.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Call vxlan_get_rx_port() with rtnl lock
Matthew Finlay [Thu, 28 Apr 2016 22:36:31 +0000 (01:36 +0300)]
net/mlx5e: Call vxlan_get_rx_port() with rtnl lock

Hold the rtnl lock when calling vxlan_get_rx_port().

Fixes: b7aade15485a ("vxlan: break dependency with netdev drivers")
Signed-off-by: Matthew Finlay <matt@mellanox.com>
Reported-by: Alexander Duyck <alexander.duyck@gmail.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'enc28j60-small-improvements'
David S. Miller [Fri, 29 Apr 2016 20:23:03 +0000 (16:23 -0400)]
Merge branch 'enc28j60-small-improvements'

Michael Heimpold says:

====================
net: ethernet: enc28j60: small improvements

This series of two patches adds the following improvements to the driver:

1) Rework the central SPI read function so that it is compatible with
   SPI masters which only support half duplex transfers.

2) Add a device tree binding for the driver.

Changelog:

v3: * renamed and improved binding documentation as
      suggested by Rob Herring

v2: * took care of Arnd Bergmann's review comments
      - allow to specify MAC address via DT
      - unconditionally define DT id table
    * increased the driver version minor number
    * driver author's email address bounces, removed from address list

v1: * Initial submission
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: ethernet: enc28j60: add device tree support
Michael Heimpold [Thu, 28 Apr 2016 20:06:15 +0000 (22:06 +0200)]
net: ethernet: enc28j60: add device tree support

The following patch adds the required match table for device tree support
(and while at, fix the indent). It's also possible to specify the
MAC address in the DT blob.

Also add the corresponding binding documentation file.

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: ethernet: enc28j60: support half-duplex SPI controllers
Michael Heimpold [Thu, 28 Apr 2016 20:06:14 +0000 (22:06 +0200)]
net: ethernet: enc28j60: support half-duplex SPI controllers

The current spi_read_buf function fails on SPI host masters which
are only half-duplex capable. Splitting the Tx and Rx part solves
this issue.

Tested on Raspberry Pi (full duplex) and I2SE Duckbill (half duplex).

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: constify is_skb_forwardable's arguments
Nikolay Aleksandrov [Thu, 28 Apr 2016 15:59:28 +0000 (17:59 +0200)]
net: constify is_skb_forwardable's arguments

is_skb_forwardable is not supposed to change anything so constify its
arguments

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'ppp-rtnetlink'
David S. Miller [Fri, 29 Apr 2016 20:09:45 +0000 (16:09 -0400)]
Merge branch 'ppp-rtnetlink'

Guillaume Nault says:

====================
ppp: add rtnetlink support

PPP devices lack the ability to be customised at creation time. In
particular they can't be created in a given netns or with a particular
name. Moving or renaming the device after creation is possible, but
creates undesirable transient effects on servers where PPP devices are
constantly created and removed, as users connect and disconnect.
Implementing rtnetlink support solves this problem.

The rtnetlink handlers implemented in this series are minimal, and can
only replace the PPPIOCNEWUNIT ioctl. The rest of PPP ioctls remains
necessary for any other operation on channels and units.
It is perfectly possible to mix PPP devices created by rtnl
and by ioctl(PPPIOCNEWUNIT). Devices will behave in the same way.

mutex_trylock() is used to resolve the locking issue wrt. locking
dependency between rtnl_lock() and ppp_mutex (see ppp_nl_newlink() in
patch #2).

A user visible difference brought by this series is that old PPP
interfaces (those created with ioctl(PPPIOCNEWUNIT)), can now be
removed by "ip link del", just like new rtnl based PPP devices.

Changes since v3:
  - Rebase on net-next.
  - Not an RFC anymore.

Changes since v2:
  - Define ->rtnl_link_ops for ioctl based PPP devices, so they can
    handle rtnl messages just like rtnl based ones (suggested by
    Stephen Hemminger).
  - Move back to original lock ordering between ppp_mutex and rtnl_lock
    to simplify patch series. Handle lock inversion issue using
    mutex_trylock() (suggested by Stephen Hemminger).
  - Do file descriptor lookup directly in ppp_nl_newlink(), to simplify
    ppp_dev_configure().

Changes since v1:
  - Rebase on net-next.
  - Invert locking order wrt. ppp_mutex and rtnl_lock and protect
    file->private_data with ppp_mutex.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoppp: add rtnetlink device creation support
Guillaume Nault [Thu, 28 Apr 2016 15:55:30 +0000 (17:55 +0200)]
ppp: add rtnetlink device creation support

Define PPP device handler for use with rtnetlink.
The only PPP specific attribute is IFLA_PPP_DEV_FD. It is mandatory and
contains the file descriptor of the associated /dev/ppp instance (the
file descriptor which would have been used for ioctl(PPPIOCNEWUNIT) in
the ioctl-based API). The PPP device is removed when this file
descriptor is released (same behaviour as with ioctl based PPP
devices).

PPP devices created with the rtnetlink API behave like the ones created
with ioctl(PPPIOCNEWUNIT). In particular existing ioctls work the same
way, no matter how the PPP device was created.
The rtnl callbacks are also assigned to ioctl based PPP devices. This
way, rtnl messages have the same effect on any PPP devices.
The immediate effect is that all PPP devices, even ioctl-based
ones, can now be removed with "ip link del".

A minor difference still exists between ioctl and rtnl based PPP
interfaces: in the device name, the number following the "ppp" prefix
corresponds to the PPP unit number for ioctl based devices, while it is
just an unrelated incrementing index for rtnl ones.

Signed-off-by: Guillaume Nault <g.nault@alphalink.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoppp: define reusable device creation functions
Guillaume Nault [Thu, 28 Apr 2016 15:55:28 +0000 (17:55 +0200)]
ppp: define reusable device creation functions

Move PPP device initialisation and registration out of
ppp_create_interface().
This prepares code for device registration with rtnetlink.

While there, simplify the prototype of ppp_create_interface():

  * Since ppp_dev_configure() takes care of setting file->private_data,
    there's no need to return a ppp structure to ppp_unattached_ioctl()
    anymore.

  * The unit parameter is made read/write so that ppp_create_interface()
    can tell which unit number has been assigned.

Signed-off-by: Guillaume Nault <g.nault@alphalink.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: ethernet: stmmac: update MDIO support for GMAC4
Alexandre TORGUE [Thu, 28 Apr 2016 13:56:45 +0000 (15:56 +0200)]
net: ethernet: stmmac: update MDIO support for GMAC4

On new GMAC4 IP, MAC_MDIO_address register has been updated, and bitmaps
changed. This patch takes into account those changes.

Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agovxlan: fix initialization with custom link parameters
Jiri Benc [Thu, 28 Apr 2016 14:36:30 +0000 (16:36 +0200)]
vxlan: fix initialization with custom link parameters

Commit 0c867c9bf84c ("vxlan: move Ethernet initialization to a separate
function") changed initialization order and as an unintended result, when the
user specifies additional link parameters (such as IFLA_ADDRESS) while
creating vxlan interface, those are overwritten by vxlan_ether_setup later.

It's necessary to call ether_setup from withing the ->setup callback. That
way, the correct parameters are set by rtnl_create_link later. This is done
also for VXLAN-GPE, as we don't know the interface type yet at that point,
and changed to the correct interface type later.

Fixes: 0c867c9bf84c ("vxlan: move Ethernet initialization to a separate function")
Reported-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Jiri Benc <jbenc@redhat.com>
Tested-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'samples-bpf-user-experience'
David S. Miller [Fri, 29 Apr 2016 18:26:32 +0000 (14:26 -0400)]
Merge branch 'samples-bpf-user-experience'

Jesper Dangaard Brouer says:

====================
samples/bpf: Improve user experience

It is a steep learning curve getting started with using the eBPF
examples in samples/bpf/.  There are several dependencies, and
specific versions of these dependencies.  Invoking make in the correct
manor is also slightly obscure.

This patchset cleanup, document and hopefully improves the first time
user experience with the eBPF samples directory by auto-detecting
certain scenarios.

V4:
 - Address Naveen's nitpicks
 - Handle/fail if extra args are passed in LLC or CLANG (David Laight)

V3:
 - Add Alexei's ACKs
 - Remove README paragraph about LLVM experimental BPF target
   as it only existed between LLVM version 3.6 to 3.7.

V2:
 - Adjusted recommend minimum versions to 3.7.1
 - Included clang build instructions
 - New patch adding CLANG variable and validation of command
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosamples/bpf: like LLC also verify and allow redefining CLANG command
Jesper Dangaard Brouer [Thu, 28 Apr 2016 12:21:14 +0000 (14:21 +0200)]
samples/bpf: like LLC also verify and allow redefining CLANG command

Users are likely to manually compile both LLVM 'llc' and 'clang'
tools.  Thus, also allow redefining CLANG and verify command exist.

Makefile implementation wise, the target that verify the command have
been generalized.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>