openwrt/staging/blogic.git
8 years agoipv4: fib: Replay events when registering FIB notifier
Ido Schimmel [Sat, 3 Dec 2016 15:45:07 +0000 (16:45 +0100)]
ipv4: fib: Replay events when registering FIB notifier

Commit b90eb7549499 ("fib: introduce FIB notification infrastructure")
introduced a new notification chain to notify listeners (f.e., switchdev
drivers) about addition and deletion of routes.

However, upon registration to the chain the FIB tables can already be
populated, which means potential listeners will have an incomplete view
of the tables.

Solve that by dumping the FIB tables and replaying the events to the
passed notification block. The dump itself is done using RCU in order
not to starve consumers that need RTNL to make progress.

The integrity of the dump is ensured by reading the FIB change sequence
counter before and after the dump under RTNL. This allows us to avoid
the problematic situation in which the dumping process sends a ENTRY_ADD
notification following ENTRY_DEL generated by another process holding
RTNL.

Callers of the registration function may pass a callback that is
executed in case the dump was inconsistent with current FIB tables.

The number of retries until a consistent dump is achieved is set to a
fixed number to prevent callers from looping for long periods of time.
In case current limit proves to be problematic in the future, it can be
easily converted to be configurable using a sysctl.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv4: fib: Allow for consistent FIB dumping
Ido Schimmel [Sat, 3 Dec 2016 15:45:06 +0000 (16:45 +0100)]
ipv4: fib: Allow for consistent FIB dumping

The next patch will enable listeners of the FIB notification chain to
request a dump of the FIB tables. However, since RTNL isn't taken during
the dump, it's possible for the FIB tables to change mid-dump, which
will result in inconsistency between the listener's table and the
kernel's.

Allow listeners to know about changes that occurred mid-dump, by adding
a change sequence counter to each net namespace. The counter is
incremented just before a notification is sent in the FIB chain.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv4: fib: Convert FIB notification chain to be atomic
Ido Schimmel [Sat, 3 Dec 2016 15:45:05 +0000 (16:45 +0100)]
ipv4: fib: Convert FIB notification chain to be atomic

In order not to hold RTNL for long periods of time we're going to dump
the FIB tables using RCU.

Convert the FIB notification chain to be atomic, as we can't block in
RCU critical sections.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agorocker: Register FIB notifier before creating ports
Ido Schimmel [Sat, 3 Dec 2016 15:45:04 +0000 (16:45 +0100)]
rocker: Register FIB notifier before creating ports

We can miss FIB notifications sent between the time the ports were
created and the FIB notification block registered.

Instead of receiving these notifications only when they are replayed for
the FIB notification block during registration, just register the
notification block before the ports are created.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agorocker: Implement FIB offload in deferred work
Ido Schimmel [Sat, 3 Dec 2016 15:45:03 +0000 (16:45 +0100)]
rocker: Implement FIB offload in deferred work

Convert rocker to offload FIBs in deferred work in a similar fashion to
mlxsw, which was converted in the previous commits.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agorocker: Create an ordered workqueue for FIB offload
Ido Schimmel [Sat, 3 Dec 2016 15:45:02 +0000 (16:45 +0100)]
rocker: Create an ordered workqueue for FIB offload

As explained in the previous commits, we need to process FIB entries
addition / deletion events in FIFO order or otherwise we can have a
mismatch between the kernel's FIB table and the device's.

Create an ordered workqueue for rocker to which these work items will be
submitted to.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomlxsw: spectrum_router: Implement FIB offload in deferred work
Ido Schimmel [Sat, 3 Dec 2016 15:45:01 +0000 (16:45 +0100)]
mlxsw: spectrum_router: Implement FIB offload in deferred work

FIB offload is currently done in process context with RTNL held, but
we're about to dump the FIB tables in RCU critical section, so we can no
longer sleep.

Instead, defer the operation to process context using deferred work. Make
sure fib info isn't freed while the work is queued by taking a reference
on it and releasing it after the operation is done.

Deferring the operation is valid because the upper layers always assume
the operation was successful. If it's not, then the driver-specific
abort mechanism is called and all routed traffic is directed to slow
path.

The work items are submitted to an ordered workqueue to prevent a
mismatch between the kernel's FIB table and the device's.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomlxsw: core: Create an ordered workqueue for FIB offload
Ido Schimmel [Sat, 3 Dec 2016 15:45:00 +0000 (16:45 +0100)]
mlxsw: core: Create an ordered workqueue for FIB offload

We're going to start processing FIB entries addition / deletion events
in deferred work. These work items must be processed in the order they
were submitted or otherwise we can have differences between the kernel's
FIB table and the device's.

Solve this by creating an ordered workqueue to which these work items
will be submitted to. Note that we can't simply convert the current
workqueue to be ordered, as EMADs re-transmissions are also processed in
deferred work.

Later on, we can migrate other work items to this workqueue, such as FDB
notification processing and nexthop resolution, since they all take the
same lock anyway.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv4: fib: Add fib_info_hold() helper
Ido Schimmel [Sat, 3 Dec 2016 15:44:59 +0000 (16:44 +0100)]
ipv4: fib: Add fib_info_hold() helper

As explained in the previous commit, modules are going to need to take a
reference on fib info and then drop it using fib_info_put().

Add the fib_info_hold() helper to make the code more readable and also
symmetric with fib_info_put().

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Suggested-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv4: fib: Export free_fib_info()
Ido Schimmel [Sat, 3 Dec 2016 15:44:58 +0000 (16:44 +0100)]
ipv4: fib: Export free_fib_info()

The FIB notification chain is going to be converted to an atomic chain,
which means switchdev drivers will have to offload FIB entries in
deferred work, as hardware operations entail sleeping.

However, while the work is queued fib info might be freed, so a
reference must be taken. To release the reference (and potentially free
the fib info) fib_info_put() will be called, which in turn calls
free_fib_info().

Export free_fib_info() so that modules will be able to invoke
fib_info_put().

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoact_mirred: fix a typo in get_dev
WANG Cong [Sat, 3 Dec 2016 18:36:01 +0000 (10:36 -0800)]
act_mirred: fix a typo in get_dev

Fixes: 255cb30425c0 ("net/sched: act_mirred: Add new tc_action_ops get_dev()")
Cc: Hadar Hen Zion <hadarh@mellanox.com>
Cc: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Eric Dumazet <edumazet@google.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 [Sun, 4 Dec 2016 00:10:48 +0000 (19:10 -0500)]
Merge branch '40GbE' of git://git./linux/kernel/git/jkirsher/next-queue

Jeff Kirsher says:

====================
40GbE Intel Wired LAN Driver Updates 2016-12-02

This series contains updates to i40e and i40evf only.

Alex provides changes so that we are much more robust about defining what
we can and cannot offload in i40e and i40evf by doing additional checks
other than L4 tunnel header length.

Jake provides several fixes/changes, first cleaning up a label that is
unnecessary, as well as cleaned up the use of a "magic number".  Clarified
the code by separating the global private flags and the regular private
flags per interface into two arrays, so that future additions will not
produce duplication and buggy code.  Adds additional checks to protect
against NULL values for msix_entries and q_vectors pointers.

Michal adds Clause22 method for accessing registers for some external
PHYs.

Piotr adds additional protocol support for the admin queue discover
capabilities function.

Tushar Dave fixes a panic seen on SPARC, where writel() should not be
used to write directly to a memory address but only to a memory mapped
I/O address otherwise it causes data access exceptions.

Joe Perches separates out a section of code into its own function, to
help reduce i40evf_reset_task() a bit.

Alan fixes an issue by checking for NULL before dereferencing msix_entries
and returning early in the case where it is NULL within the i40evf_close()
code path.

Henry provides code cleanup to remove unreachable and redundant sections
of code.  Fixed up an issue where new NICs were not identifying "unknown
PHYs" correctly.

Harshitha fixes a issue where the ethtool "Supported Link" modes list
backplane interfaces on X722 devices for 10 GbE with SFP+ and Cortina
retimer, where these interfaces should not be visible to the user since
they cannot use them.

Carolyn changes an X722 informational message so that it only appears
when extra messages are desired.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agotcp: fix the missing avr32 SOF_TIMESTAMPING_OPT_STATS
Yuchung Cheng [Sat, 3 Dec 2016 22:46:22 +0000 (14:46 -0800)]
tcp: fix the missing avr32 SOF_TIMESTAMPING_OPT_STATS

The commit of SOF_TIMESTAMPING_OPT_STATS didn't include the
new header for avr32, causing build to break. The patch fixes it.

Fixes: 1c885808e456 ("tcp: SOF_TIMESTAMPING_OPT_STATS option for SO_TIMESTAMPING")
Reported-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoudp: be less conservative with sock rmem accounting
Paolo Abeni [Fri, 2 Dec 2016 16:35:49 +0000 (17:35 +0100)]
udp: be less conservative with sock rmem accounting

Before commit 850cbaddb52d ("udp: use it's own memory accounting
schema"), the udp protocol allowed sk_rmem_alloc to grow beyond
the rcvbuf by the whole current packet's truesize. After said commit
we allow sk_rmem_alloc to exceed the rcvbuf only if the receive queue
is empty. As reported by Jesper this cause a performance regression
for some (small) values of rcvbuf.

This commit is intended to fix the regression restoring the old
handling of the rcvbuf limit.

Reported-by: Jesper Dangaard Brouer <brouer@redhat.com>
Fixes: 850cbaddb52d ("udp: use it's own memory accounting schema")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet_sched: gen_estimator: account for timer drifts
Eric Dumazet [Fri, 2 Dec 2016 16:11:00 +0000 (08:11 -0800)]
net_sched: gen_estimator: account for timer drifts

Under heavy stress, timer used in estimators tend to slowly be delayed
by a few jiffies, leading to inaccuracies.

Lets remember what was the last scheduled jiffies so that we get more
precise estimations, without having to add a multiply/divide in the loop
to account for the drifts.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosfc: remove EFX_BUG_ON_PARANOID, use EFX_WARN_ON_[ONCE_]PARANOID instead
Edward Cree [Fri, 2 Dec 2016 15:51:33 +0000 (15:51 +0000)]
sfc: remove EFX_BUG_ON_PARANOID, use EFX_WARN_ON_[ONCE_]PARANOID instead

Logically, EFX_BUG_ON_PARANOID can never be correct.  For, BUG_ON should
 only be used if it is not possible to continue without potential harm;
 and since the non-DEBUG driver will continue regardless (as the BUG_ON is
 compiled out), clearly the BUG_ON cannot be needed in the DEBUG driver.
So, replace every EFX_BUG_ON_PARANOID with either an EFX_WARN_ON_PARANOID
 or the newly defined EFX_WARN_ON_ONCE_PARANOID.

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'samples-bpf-automated-cgroup-tests'
David S. Miller [Sat, 3 Dec 2016 21:08:01 +0000 (16:08 -0500)]
Merge branch 'samples-bpf-automated-cgroup-tests'

Sargun Dhillon says:

====================
samples, bpf: Refactor; Add automated tests for cgroups

These two patches are around refactoring out some old, reusable code from the
existing test_current_task_under_cgroup_user test, and adding a new, automated
test.

There is some generic cgroupsv2 setup & cleanup code, given that most
environment still don't have it setup by default. With this code, we're able
to pretty easily add an automated test for future cgroupsv2 functionality.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosamples, bpf: Add automated test for cgroup filter attachments
Sargun Dhillon [Fri, 2 Dec 2016 10:42:32 +0000 (02:42 -0800)]
samples, bpf: Add automated test for cgroup filter attachments

This patch adds the sample program test_cgrp2_attach2. This program is
similar to test_cgrp2_attach, but it performs automated testing of the
cgroupv2 BPF attached filters. It runs the following checks:
* Simple filter attachment
* Application of filters to child cgroups
* Overriding filters on child cgroups
* Checking that this still works when the parent filter is removed

The filters that are used here are simply allow all / deny all filters, so
it isn't checking the actual functionality of the filters, but rather
the behaviour  around detachment / attachment. If net_cls is enabled,
this test will fail.

Signed-off-by: Sargun Dhillon <sargun@sargun.me>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosamples, bpf: Refactor test_current_task_under_cgroup - separate out helpers
Sargun Dhillon [Fri, 2 Dec 2016 10:42:18 +0000 (02:42 -0800)]
samples, bpf: Refactor test_current_task_under_cgroup - separate out helpers

This patch modifies test_current_task_under_cgroup_user. The test has
several helpers around creating a temporary environment for cgroup
testing, and moving the current task around cgroups. This set of
helpers can then be used in other tests.

Signed-off-by: Sargun Dhillon <sargun@sargun.me>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosamples/bpf: silence compiler warnings
Alexei Starovoitov [Fri, 2 Dec 2016 02:31:12 +0000 (18:31 -0800)]
samples/bpf: silence compiler warnings

silence some of the clang compiler warnings like:
include/linux/fs.h:2693:9: warning: comparison of unsigned enum expression < 0 is always false
arch/x86/include/asm/processor.h:491:30: warning: taking address of packed member 'sp0' of class or structure 'x86_hw_tss' may result in an unaligned pointer value
include/linux/cgroup-defs.h:326:16: warning: field 'cgrp' with variable sized type 'struct cgroup' not at the end of a struct or class is a GNU extension
since they add too much noise to samples/bpf/ build.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonetns: fix net_generic() "id - 1" bloat
Alexey Dobriyan [Fri, 2 Dec 2016 01:21:32 +0000 (04:21 +0300)]
netns: fix net_generic() "id - 1" bloat

net_generic() function is both a) inline and b) used ~600 times.

It has the following code inside

...
ptr = ng->ptr[id - 1];
...

"id" is never compile time constant so compiler is forced to subtract 1.
And those decrements or LEA [r32 - 1] instructions add up.

We also start id'ing from 1 to catch bugs where pernet sybsystem id
is not initialized and 0. This is quite pointless idea (nothing will
work or immediate interference with first registered subsystem) in
general but it hints what needs to be done for code size reduction.

Namely, overlaying allocation of pointer array and fixed part of
structure in the beginning and using usual base-0 addressing.

Ids are just cookies, their exact values do not matter, so lets start
with 3 on x86_64.

Code size savings (oh boy): -4.2 KB

As usual, ignore the initial compiler stupidity part of the table.

add/remove: 0/0 grow/shrink: 12/670 up/down: 89/-4297 (-4208)
function                                     old     new   delta
tipc_nametbl_insert_publ                    1250    1270     +20
nlmclnt_lookup_host                          686     703     +17
nfsd4_encode_fattr                          5930    5941     +11
nfs_get_client                              1050    1061     +11
register_pernet_operations                   333     342      +9
tcf_mirred_init                              843     849      +6
tcf_bpf_init                                1143    1149      +6
gss_setup_upcall                             990     994      +4
idmap_name_to_id                             432     434      +2
ops_init                                     274     275      +1
nfsd_inject_forget_client                    259     260      +1
nfs4_alloc_client                            612     613      +1
tunnel_key_walker                            164     163      -1

...

tipc_bcbase_select_primary                   392     360     -32
mac80211_hwsim_new_radio                    2808    2767     -41
ipip6_tunnel_ioctl                          2228    2186     -42
tipc_bcast_rcv                               715     672     -43
tipc_link_build_proto_msg                   1140    1089     -51
nfsd4_lock                                  3851    3796     -55
tipc_mon_rcv                                1012     956     -56
Total: Before=156643951, After=156639743, chg -0.00%

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonetns: add dummy struct inside "struct net_generic"
Alexey Dobriyan [Fri, 2 Dec 2016 01:12:58 +0000 (04:12 +0300)]
netns: add dummy struct inside "struct net_generic"

This is precursor to fixing "[id - 1]" bloat inside net_generic().

Name "s" is chosen to complement name "u" often used for dummy unions.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonetns: publish net_generic correctly
Alexey Dobriyan [Fri, 2 Dec 2016 01:11:34 +0000 (04:11 +0300)]
netns: publish net_generic correctly

Publishing net_generic pointer is done with silly mistake: new array is
published BEFORE setting freshly acquired pernet subsystem pointer.

memcpy
rcu_assign_pointer
kfree_rcu
ng->ptr[id - 1] = data;

This bug was introduced with commit dec827d174d7f76c457238800183ca864a639365
("[NETNS]: The generic per-net pointers.") in the glorious days of
chopping networking stack into containers proper 8.5 years ago (whee...)

How it didn't trigger for so long?
Well, you need quite specific set of conditions:

*) race window opens once per pernet subsystem addition
   (read: modprobe or boot)

*) not every pernet subsystem is eligible (need ->id and ->size)

*) not every pernet subsystem is vulnerable (need incorrect or absense
   of ordering of register_pernet_sybsys() and actually using net_generic())

*) to hide the bug even more, default is to preallocate 13 pointers which
   is actually quite a lot. You need IPv6, netfilter, bridging etc together
   loaded to trigger reallocation in the first place. Trimmed down
   config are OK.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonetlink: 2-clause nla_ok()
Alexey Dobriyan [Fri, 2 Dec 2016 00:59:06 +0000 (03:59 +0300)]
netlink: 2-clause nla_ok()

nla_ok() consists of 3 clauses:

1) int rem >= (int)sizeof(struct nlattr)

2) u16 nla_len >= sizeof(struct nlattr)

3) u16 nla_len <= int rem

The statement is that clause (1) is redundant.

What it does is ensuring that "rem" is a positive number,
so that in clause (3) positive number will be compared to positive number
with no problems.

However, "u16" fully fits into "int" and integers do not change value
when upcasting even to signed type. Negative integers will be rejected
by clause (3) just fine. Small positive integers will be rejected
by transitivity of comparison operator.

NOTE: all of the above DOES NOT apply to nlmsg_ok() where ->nlmsg_len is
u32(!), so 3 clauses AND A CAST TO INT are necessary.

Obligatory space savings report: -1.6 KB

$ ./scripts/bloat-o-meter ../vmlinux-000* ../vmlinux-001*
add/remove: 0/0 grow/shrink: 3/63 up/down: 35/-1692 (-1657)
function                                     old     new   delta
validate_scan_freqs                          142     155     +13
tcf_em_tree_validate                         867     879     +12
dcbnl_ieee_del                               328     338     +10
netlbl_cipsov4_add_common.isra               218     215      -3
...
ovs_nla_put_actions                          888     806     -82
netlbl_cipsov4_add_std                      1648    1566     -82
nl80211_parse_sched_scan                    2889    2780    -109
ip_tun_from_nlattr                          3086    2945    -141

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostaging: wilc1000: use reset to set mac header
Zhang Shengju [Fri, 2 Dec 2016 01:51:07 +0000 (09:51 +0800)]
staging: wilc1000: use reset to set mac header

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoiwlwifi: use reset to set transport header
Zhang Shengju [Fri, 2 Dec 2016 01:51:06 +0000 (09:51 +0800)]
iwlwifi: use reset to set transport header

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomlx4: use reset to set mac header
Zhang Shengju [Fri, 2 Dec 2016 01:51:05 +0000 (09:51 +0800)]
mlx4: use reset to set mac header

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobnx2x: use reset to set network header
Zhang Shengju [Fri, 2 Dec 2016 01:51:04 +0000 (09:51 +0800)]
bnx2x: use reset to set network header

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoqede: use reset to set network header
Zhang Shengju [Fri, 2 Dec 2016 01:51:03 +0000 (09:51 +0800)]
qede: use reset to set network header

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Acked-by: Yuval Mintz <Yuval.Mintz@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'xgene-jumbo-and-pause-frame'
David S. Miller [Sat, 3 Dec 2016 20:46:51 +0000 (15:46 -0500)]
Merge branch 'xgene-jumbo-and-pause-frame'

Iyappan Subramanian says:

====================
drivers: net: xgene: Add Jumbo and Pause frame support

This patch set adds,

1. Jumbo frame support
2. Pause frame based flow control

and fixes RSS for non-TCP/UDP packets.
====================

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
8 years agodrivers: net: xgene: ethtool: Add get/set_pauseparam
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:44 +0000 (16:41 -0800)]
drivers: net: xgene: ethtool: Add get/set_pauseparam

This patch adds get_pauseparam and set_pauseparam functions.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Quan Nguyen <qnguyen@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: Add flow control initialization
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:43 +0000 (16:41 -0800)]
drivers: net: xgene: Add flow control initialization

This patch adds flow control/pause frame initialization and
advertising capabilities.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Quan Nguyen <qnguyen@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: Add flow control configuration
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:42 +0000 (16:41 -0800)]
drivers: net: xgene: Add flow control configuration

This patch adds functions to configure mac, when flow control
and pause frame settings change.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Quan Nguyen <qnguyen@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: fix: RSS for non-TCP/UDP
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:41 +0000 (16:41 -0800)]
drivers: net: xgene: fix: RSS for non-TCP/UDP

This patch fixes RSS feature, for non-TCP/UDP packets.

Signed-off-by: Khuong Dinh <kdinh@apm.com>
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: Add change_mtu function
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:40 +0000 (16:41 -0800)]
drivers: net: xgene: Add change_mtu function

This patch implements ndo_change_mtu() callback function that
enables mtu change.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Quan Nguyen <qnguyen@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: Add support for Jumbo frame
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:39 +0000 (16:41 -0800)]
drivers: net: xgene: Add support for Jumbo frame

This patch adds support for jumbo frame, by allocating
additional buffer (page) pool and configuring the hardware.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Quan Nguyen <qnguyen@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: Configure classifier with pagepool
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:38 +0000 (16:41 -0800)]
drivers: net: xgene: Configure classifier with pagepool

This patch configures classifier with the pagepool information.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Quan Nguyen <qnguyen@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodrivers: net: xgene: Add helper function
Iyappan Subramanian [Fri, 2 Dec 2016 00:41:37 +0000 (16:41 -0800)]
drivers: net: xgene: Add helper function

This is a prepartion patch and adds xgene_enet_get_fpsel() helper
function to get buffer pool number.

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Quan Nguyen <qnguyen@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: ethernet: ti: davinci_cpdma: add missing EXPORTs
Paul Gortmaker [Thu, 1 Dec 2016 20:25:28 +0000 (15:25 -0500)]
net: ethernet: ti: davinci_cpdma: add missing EXPORTs

As of commit 8f32b90981dcdb355516fb95953133f8d4e6b11d
("net: ethernet: ti: davinci_cpdma: add set rate for a channel") the
ARM allmodconfig builds would fail modpost with:

ERROR: "cpdma_chan_set_weight" [drivers/net/ethernet/ti/ti_cpsw.ko] undefined!
ERROR: "cpdma_chan_get_rate" [drivers/net/ethernet/ti/ti_cpsw.ko] undefined!
ERROR: "cpdma_chan_get_min_rate" [drivers/net/ethernet/ti/ti_cpsw.ko] undefined!
ERROR: "cpdma_chan_set_rate" [drivers/net/ethernet/ti/ti_cpsw.ko] undefined!

Since these weren't declared as static, it is assumed they were
meant to be shared outside the file, and that modular build testing
was simply overlooked.

Fixes: 8f32b90981dc ("net: ethernet: ti: davinci_cpdma: add set rate for a channel")
Cc: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Cc: Mugunthan V N <mugunthanvnm@ti.com>
Cc: Grygorii Strashko <grygorii.strashko@ti.com>
Cc: linux-omap@vger.kernel.org
Cc: netdev@vger.kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge tag 'linux-can-next-for-4.10-20161201' of git://git.kernel.org/pub/scm/linux...
David S. Miller [Sat, 3 Dec 2016 20:26:30 +0000 (15:26 -0500)]
Merge tag 'linux-can-next-for-4.10-20161201' of git://git./linux/kernel/git/mkl/linux-can-next

Marc Kleine-Budde says:

====================
pull-request: can-next 2016-12-01

this is a pull request of 4 patches for net-next/master.

There are two patches by Chris Paterson for the rcar_can and rcar_canfd
device tree binding documentation. And a patch by Geert Uytterhoeven
that corrects the order of interrupt specifiers.

The fourth patch by Colin Ian King fixes a spelling error in the
kvaser_usb driver.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: stmmac: unify mdio functions
LABBE Corentin [Thu, 1 Dec 2016 15:19:41 +0000 (16:19 +0100)]
net: stmmac: unify mdio functions

stmmac_mdio_{read|write} and stmmac_mdio_{read|write}_gmac4 are not
enought different for being split.
The only differences between thoses two functions are shift/mask for
addr/reg/clk_csr.

This patch introduce a per platform set of variable for setting thoses
shift/mask and unify mdio read and write functions.

Signed-off-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: stmmac: avoid Camelcase naming
LABBE Corentin [Thu, 1 Dec 2016 15:19:40 +0000 (16:19 +0100)]
net: stmmac: avoid Camelcase naming

This patch simply rename regValue to value, like it was named in other
mdio functions.

Signed-off-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Acked-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoirda: w83977af_ir: fix damaged whitespace
Arnd Bergmann [Mon, 28 Nov 2016 14:19:43 +0000 (15:19 +0100)]
irda: w83977af_ir: fix damaged whitespace

As David Miller pointed out for for the previous patch, the whitespace
in some functions looks rather odd. This was caused by commit 6329da5f258a
("obsolete config in kernel source: USE_INTERNAL_TIMER"), which removed
some conditions but did not reindent the code.

This fixes the indentation in the file and removes extraneous whitespace
at the end of the lines and before tabs.

There are many other minor coding style problems in the driver, but I'm
not touching those here.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: cleanup documenation, make it match reality
Pavel Machek [Thu, 1 Dec 2016 10:32:18 +0000 (11:32 +0100)]
stmmac: cleanup documenation, make it match reality

Fix english in documentation, make documentation match reality, remove
options that were removed from code.

Signed-off-by: Pavel Machek <pavel@denx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
David S. Miller [Sat, 3 Dec 2016 16:46:54 +0000 (11:46 -0500)]
Merge git://git./linux/kernel/git/davem/net

Couple conflicts resolved here:

1) In the MACB driver, a bug fix to properly initialize the
   RX tail pointer properly overlapped with some changes
   to support variable sized rings.

2) In XGBE we had a "CONFIG_PM" --> "CONFIG_PM_SLEEP" fix
   overlapping with a reorganization of the driver to support
   ACPI, OF, as well as PCI variants of the chip.

3) In 'net' we had several probe error path bug fixes to the
   stmmac driver, meanwhile a lot of this code was cleaned up
   and reorganized in 'net-next'.

4) The cls_flower classifier obtained a helper function in
   'net-next' called __fl_delete() and this overlapped with
   Daniel Borkamann's bug fix to use RCU for object destruction
   in 'net'.  It also overlapped with Jiri's change to guard
   the rhashtable_remove_fast() call with a check against
   tc_skip_sw().

5) In mlx4, a revert bug fix in 'net' overlapped with some
   unrelated changes in 'net-next'.

6) In geneve, a stale header pointer after pskb_expand_head()
   bug fix in 'net' overlapped with a large reorganization of
   the same code in 'net-next'.  Since the 'net-next' code no
   longer had the bug in question, there was nothing to do
   other than to simply take the 'net-next' hunks.

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoi40e: change message to only appear when extra debug info is wanted
Carolyn Wyborny [Tue, 8 Nov 2016 21:05:12 +0000 (13:05 -0800)]
i40e: change message to only appear when extra debug info is wanted

This patch changes an X722 informational message so that it only
appears when extra messages are desired.  Without this patch,
on X722 devices, this message appears at load, potentially causing
unnecessary alarm.

Change-ID: I94f7aae15dc5b2723cc9728c630c72538a3e670e
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/i40evf: replace for memcpy with single memcpy call in ethtool
Jacob Keller [Tue, 8 Nov 2016 21:05:11 +0000 (13:05 -0800)]
i40e/i40evf: replace for memcpy with single memcpy call in ethtool

memcpy replaced with single memcpy call in ethtool.

Change-ID: I3f5bef6bcc593412c56592c6459784db41575a0a
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: set broadcast promiscuous mode for each active VLAN
Jacob Keller [Tue, 8 Nov 2016 21:05:10 +0000 (13:05 -0800)]
i40e: set broadcast promiscuous mode for each active VLAN

A previous workaround added to ensure receipt of all broadcast frames
incorrectly set the broadcast promiscuous mode unconditionally
regardless of active VLAN status.

Replace this partial workaround with a complete solution that sets the
broadcast promiscuous filters in i40e_sync_vsi_filters. This new method
sets the promiscuous mode based on when broadcast filters are added or
removed.

I40E_VLAN_ANY will request a broadcast filter for all VLANs, (as we're
in untagged mode) while a broadcast filter on a specific VLAN will only
request broadcast for that VLAN.

Thus, we restore addition of broadcast filter to the array, but we add
special handling for these such that they enable the broadcast
promiscuous mode instead of being sent as regular filters.

The end result is that we will correctly receive all broadcast packets
(even those with a *source* address equal to the broadcast address) but
will not receive packets for which we don't have an active VLAN filter.

Change-ID: I7d0585c5cec1a5bf55bf533b42e5e817d5db6a2d
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: Fix for ethtool Supported link modes
Harshitha Ramamurthy [Tue, 8 Nov 2016 21:05:09 +0000 (13:05 -0800)]
i40e: Fix for ethtool Supported link modes

This patch fixes the problem where the ethtool Supported link
modes list backplane interfaces on X722 devices for 10GbE with
SFP+ and Cortina retimer. This patch fixes the problem by setting
and using a flag for this particular device since the backplane
interface is only between the internal PHY and the retimer and it
should not be seen by the user as they cannot use it.
Without this patch, the user wrongly thinks that backplane interfaces
are supported on their device when they actually are not.

Change-ID: I3882bc2928431d48a2db03a51a713a1f681a79e9
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 agoi40evf: protect against NULL msix_entries and q_vectors pointers
Jacob Keller [Tue, 8 Nov 2016 21:05:08 +0000 (13:05 -0800)]
i40evf: protect against NULL msix_entries and q_vectors pointers

Update the functions which free msix_entries and q_vectors so that they
are safe against NULL values. This allows calling code to not care
whether these have already been freed when disabling and freeing them.

Change-ID: I31bfd1c0da18023d971b618edc6fb049721f3298
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: Pass unknown PHY type for unknown PHYs
Henry Tieman [Tue, 8 Nov 2016 21:05:07 +0000 (13:05 -0800)]
i40e: Pass unknown PHY type for unknown PHYs

The PHY type value for unrecognized PHYs and cables was changed
based on firmware version number. Newer hardware use lower firmware
version numbers and this was causing some PHYs to be identified
as type 0x16 instead of 0xe (unknown).

Without this patch, newer card will incorrectly identify unknown
PHYs and cables.

This change adds hardware type to the check for firmware version
so the PHY type is reported correctly.

Change-ID: I0723cbfd263c76fc73ff1a5275d1639051376c9a
Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Remove unreachable code
Henry Tieman [Tue, 8 Nov 2016 21:05:06 +0000 (13:05 -0800)]
i40e: Remove unreachable code

The code at the end of i40e_read_phy_register_clause22() contained
unreachable code and redundant control statements.

This change removes the unreachable code. And deletes the redundant
goto statement and if statement.

Change-ID: I713032b1585396f40f903cbcfdea987abd874400
Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40evf: check for msix_entries null dereference
Alan Brady [Tue, 8 Nov 2016 21:05:05 +0000 (13:05 -0800)]
i40evf: check for msix_entries null dereference

It is possible for msix_entries to be freed by a previous suspend/remove
before a VF is closed.  This patch fixes the issue by checking for NULL
before dereferencing msix_entries and returning early in the case where
it is NULL within the i40evf_close code path.  Without this patch it is
possible to trigger a kernel panic through NULL dereference.

Change-ID: I92a2746e82533a889e25f91578eac9abd0388ae2
Signed-off-by: Alan Brady <alan.brady@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40evf: Move some i40evf_reset_task code to separate function
Joe Perches [Tue, 1 Nov 2016 22:35:14 +0000 (15:35 -0700)]
i40evf: Move some i40evf_reset_task code to separate function

The i40evf_reset_task function is a couple hundred lines and it has
a separable block that disables VF.  Move that block to a new
i40evf_disable_vf function to shorten i40evf_reset_task a bit.

Signed-off-by: Joe Perches <joe@perches.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: fix panic on SPARC while changing num of desc
Tushar Dave [Wed, 26 Oct 2016 17:49:27 +0000 (10:49 -0700)]
i40e: fix panic on SPARC while changing num of desc

On SPARC, writel() should not be used to write directly to memory
address but only to memory mapped I/O address otherwise it causes
data access exception.

Commit 147e81ec75689 ("i40e: Test memory before ethtool alloc
succeeds") introduced a code that uses memory address to fake the HW
tail address and attempt to write to that address using writel()
causes kernel panic on SPARC. The issue is reproduced while changing
number of descriptors using ethtool.

This change resolves the panic by using HW read-only memory mapped
I/O register to fake HW tail address instead memory address.

e.g.
> ethtool -G eth2 tx 2048 rx 2048
i40e 0000:03:00.2 eth2: Changing Tx descriptor count from 512 to 2048.
i40e 0000:03:00.2 eth2: Changing Rx descriptor count from 512 to 2048
sun4v_data_access_exception: ADDR[fff8001f9734a000] CTX[0000]
TYPE[0004], going.
              \|/ ____ \|/
              "@'/ .. \`@"
              /_| \__/ |_\
                 \__U_/
ethtool(3273): Dax [#1]
CPU: 9 PID: 3273 Comm: ethtool Tainted: G            E
4.8.0-linux-net_temp+ #7
task: fff8001f96d7a660 task.stack: fff8001f97348000
TSTATE: 0000009911001601 TPC: 00000000103189e4 TNPC: 00000000103189e8 Y:
00000000    Tainted: G            E
TPC: <i40e_alloc_rx_buffers+0x124/0x260 [i40e]>
g0: fff8001f4eb64000 g1: 00000000000007ff g2: fff8001f9734b92c g3:
00203e0000000000
g4: fff8001f96d7a660 g5: fff8001fa6704000 g6: fff8001f97348000 g7:
0000000000000001
o0: 0006000046706928 o1: 00000000db3e2000 o2: fff8001f00000000 o3:
0000000000002000
o4: 0000000000002000 o5: 0000000000000001 sp: fff8001f9734afc1 ret_pc:
0000000010318a64
RPC: <i40e_alloc_rx_buffers+0x1a4/0x260 [i40e]>
l0: fff8001f4e8bffe0 l1: fff8001f4e8cffe0 l2: 00000000000007ff l3:
00000000ff000000
l4: 0000000000ff0000 l5: 000000000000ff00 l6: 0000000000cda6a8 l7:
0000000000e822f0
i0: fff8001f96380000 i1: 0000000000000000 i2: 00203edb00000000 i3:
0006000046706928
i4: 0000000002086320 i5: 0000000000e82370 i6: fff8001f9734b071 i7:
00000000103062d4
I7: <i40e_set_ringparam+0x3b4/0x540 [i40e]>
Call Trace:
 [00000000103062d4] i40e_set_ringparam+0x3b4/0x540 [i40e]
 [000000000094e2f8] dev_ethtool+0x898/0xbe0
 [0000000000965570] dev_ioctl+0x250/0x300
 [0000000000923800] sock_do_ioctl+0x40/0x60
 [000000000092427c] sock_ioctl+0x7c/0x280
 [00000000005ef040] vfs_ioctl+0x20/0x60
 [00000000005ef5d4] do_vfs_ioctl+0x194/0x4c0
 [00000000005ef974] SyS_ioctl+0x74/0xa0
 [0000000000406214] linux_sparc_syscall+0x34/0x44
Disabling lock debugging due to kernel taint
Caller[00000000103062d4]: i40e_set_ringparam+0x3b4/0x540 [i40e]
Caller[000000000094e2f8]: dev_ethtool+0x898/0xbe0
Caller[0000000000965570]: dev_ioctl+0x250/0x300
Caller[0000000000923800]: sock_do_ioctl+0x40/0x60
Caller[000000000092427c]: sock_ioctl+0x7c/0x280
Caller[00000000005ef040]: vfs_ioctl+0x20/0x60
Caller[00000000005ef5d4]: do_vfs_ioctl+0x194/0x4c0
Caller[00000000005ef974]: SyS_ioctl+0x74/0xa0
Caller[0000000000406214]: linux_sparc_syscall+0x34/0x44
Caller[0000000000107154]: 0x107154
Instruction DUMP: e43620c8
 e436204a  c45e2038
<c2a083a082102000
 81cfe008  90086001
 82102000  81cfe008

Kernel panic - not syncing: Fatal exception

Signed-off-by: Tushar Dave <tushar.n.dave@oracle.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Add protocols over MCTP to i40e_aq_discover_capabilities
Piotr Raczynski [Tue, 25 Oct 2016 23:08:53 +0000 (16:08 -0700)]
i40e: Add protocols over MCTP to i40e_aq_discover_capabilities

Add logical_id to I40E_AQ_CAP_ID_MNG_MODE capability starting from major
version 2.

Change-ID: Idb29214b172ea5c70cbd45a99e6745c0215af7e4
Signed-off-by: Piotr Raczynski <piotr.raczynski@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: fix trivial typo in naming of i40e_sync_filters_subtask
Jacob Keller [Tue, 25 Oct 2016 23:08:52 +0000 (16:08 -0700)]
i40e: fix trivial typo in naming of i40e_sync_filters_subtask

A comment incorrectly referred to i40e_vsi_sync_filters_subtask which
does not actually exist. Reference the correct function instead.

Change-ID: I6bd805c605741ffb6fe34377259bb0d597edfafd
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: Add Clause22 implementation
Michal Kosiarz [Tue, 25 Oct 2016 23:08:51 +0000 (16:08 -0700)]
i40e: Add Clause22 implementation

Some external PHYs require Clause22 method for accessing registers.
This patch also adds some defines to support blink led on devices using
10CBaseT PHY.

Change-ID: I868a4326911900f6c89e7e522fda4968b0825f14
Signed-off-by: Michal Kosiarz <michal.kosiarz@intel.com>
Signed-off-by: Matt Jared <matthew.a.jared@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: avoid duplicate private flags definitions
Jacob Keller [Tue, 25 Oct 2016 23:08:50 +0000 (16:08 -0700)]
i40e: avoid duplicate private flags definitions

Separate the global private flags and the regular private flags per
interface into two arrays. Future additions of private flags will not
need to be duplicated which may lead to buggy code. Also rename
"i40e_priv_flags_strings_gl" to "i40e_gl_priv_flags_strings" for
clarity, as it reads more naturally.

Change-ID: I68caef3c9954eb7da342d7f9d20f2873186f2758
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: remove second check of VLAN_N_VID in i40e_vlan_rx_add_vid
Jacob Keller [Tue, 25 Oct 2016 23:08:49 +0000 (16:08 -0700)]
i40e: remove second check of VLAN_N_VID in i40e_vlan_rx_add_vid

Replace a check of magic number 4095 with VLAN_N_VID. This
makes it obvious that a later check against VLAN_N_VID is
always true and can be removed.

Change-ID: I28998f127a61a529480ce63d8a07e266f6c63b7b
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: remove error_param_int label from i40e_vc_config_promiscuous_mode_msg
Jacob Keller [Tue, 25 Oct 2016 23:08:48 +0000 (16:08 -0700)]
i40e: remove error_param_int label from i40e_vc_config_promiscuous_mode_msg

This label is unnecessary, as are jumping to a block that checks aq_ret
and then immediately skipping it and returning. So just jump straight to
the error_param and remove this unnecessary label.

Also use goto error_param even in the last check for style consistency.

Change-ID: If487c7d10c4048e37c594e5eca167693aaed45f6
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 agoi40evf: Be much more verbose about what we can and cannot offload
Alexander Duyck [Tue, 25 Oct 2016 23:08:47 +0000 (16:08 -0700)]
i40evf: Be much more verbose about what we can and cannot offload

This change makes it so that we are much more robust about defining what we
can and cannot offload.  Previously we were performing no checks.  This
should bring us up to parity with the i40e PF driver.

In addition the device only supports GSO as long as the MSS is 64 or
greater.  We were not checking this so an MSS less than that was resulting
in Tx hangs.

Change-ID: If533553ec92fc6ba694eab6ac81fdaf3004f3592
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
8 years agoi40e: Be much more verbose about what we can and cannot offload
Alexander Duyck [Tue, 25 Oct 2016 23:08:46 +0000 (16:08 -0700)]
i40e: Be much more verbose about what we can and cannot offload

This change makes it so that we are much more robust about defining what we
can and cannot offload.  Previously we were just checking for the L4 tunnel
header length, however there are other fields we should be verifying as
there are multiple scenarios in which we cannot perform hardware offloads.

In addition the device only supports GSO as long as the MSS is 64 or
greater.  We were not checking this so an MSS less than that was resulting
in Tx hangs.

Change-ID: I5e2fd5f3075c73601b4b36327b771c64fcb6c31b
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
8 years agoMerge tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm...
Linus Torvalds [Fri, 2 Dec 2016 21:34:37 +0000 (13:34 -0800)]
Merge tag 'fixes-for-linus' of git://git./linux/kernel/git/arm/arm-soc

Pull ARM SoC fixes from Arnd Bergmann:
 "This should be the last set of bugfixes for arm-soc in v4.9. None of
  these are critical regressions, but it would be nice to still get them
  merged.

   - On the Juno platform, the idle latency was described wrong, leading
     to suboptimal cpuidle tuning.

   - Also on the same platform, PCI I/O space was set up incorrectly and
     could not work.

   - On the sti platform, a syntactically incorrect DT entry caused
     warnings.

   - The newly added 'gr8' platform has somewhat confusing file names,
     which we rename for consistency"

* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  arm64: dts: juno: fix cluster sleep state entry latency on all SoC versions
  arm64: dts: juno: Correct PCI IO window
  ARM: dts: STiH407-family: fix i2c nodes
  ARM: gr8: Rename the DTSI and relevant DTS

8 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Fri, 2 Dec 2016 19:45:27 +0000 (11:45 -0800)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) Lots more phydev and probe error path leaks in various drivers by
    Johan Hovold.

 2) Fix race in packet_set_ring(), from Philip Pettersson.

 3) Use after free in dccp_invalid_packet(), from Eric Dumazet.

 4) Signnedness overflow in SO_{SND,RCV}BUFFORCE, also from Eric
    Dumazet.

 5) When tunneling between ipv4 and ipv6 we can be left with the wrong
    skb->protocol value as we enter the IPSEC engine and this causes all
    kinds of problems. Set it before the output path does any
    dst_output() calls, from Eli Cooper.

 6) bcmgenet uses wrong device struct pointer in DMA API calls, fix from
    Florian Fainelli.

 7) Various netfilter nat bug fixes from FLorian Westphal.

 8) Fix memory leak in ipvlan_link_new(), from Gao Feng.

 9) Locking fixes, particularly wrt. socket lookups, in l2tp from
    Guillaume Nault.

10) Avoid invoking rhash teardowns in atomic context by moving netlink
    cb->done() dump completion from a worker thread. Fix from Herbert
    Xu.

11) Buffer refcount problems in tun and macvtap on errors, from Jason
    Wang.

12) We don't set Kconfig symbol DEFAULT_TCP_CONG properly when the user
    selects BBR. Fix from Julian Wollrath.

13) Fix deadlock in transmit path on altera TSE driver, from Lino
    Sanfilippo.

14) Fix unbalanced reference counting in dsa_switch_tree, from Nikita
    Yushchenko.

15) tc_tunnel_key needs to be properly exported to userspace via uapi,
    fix from Roi Dayan.

16) rds_tcp_init_net() doesn't unregister notifier in error path, fix
    from Sowmini Varadhan.

17) Stale packet header pointer access after pskb_expand_head() in
    genenve driver, fix from Sabrina Dubroca.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (103 commits)
  net: avoid signed overflows for SO_{SND|RCV}BUFFORCE
  geneve: avoid use-after-free of skb->data
  tipc: check minimum bearer MTU
  net: renesas: ravb: unintialized return value
  sh_eth: remove unchecked interrupts for RZ/A1
  net: bcmgenet: Utilize correct struct device for all DMA operations
  NET: usb: qmi_wwan: add support for Telit LE922A PID 0x1040
  cdc_ether: Fix handling connection notification
  ip6_offload: check segs for NULL in ipv6_gso_segment.
  RDS: TCP: unregister_netdevice_notifier() in error path of rds_tcp_init_net
  Revert: "ip6_tunnel: Update skb->protocol to ETH_P_IPV6 in ip6_tnl_xmit()"
  ipv6: Set skb->protocol properly for local output
  ipv4: Set skb->protocol properly for local output
  packet: fix race condition in packet_set_ring
  net: ethernet: altera: TSE: do not use tx queue lock in tx completion handler
  net: ethernet: altera: TSE: Remove unneeded dma sync for tx buffers
  net: ethernet: stmmac: fix of-node and fixed-link-phydev leaks
  net: ethernet: stmmac: platform: fix outdated function header
  net: ethernet: stmmac: dwmac-meson8b: fix probe error path
  net: ethernet: stmmac: dwmac-generic: fix probe error path
  ...

8 years agonet: avoid signed overflows for SO_{SND|RCV}BUFFORCE
Eric Dumazet [Fri, 2 Dec 2016 17:44:53 +0000 (09:44 -0800)]
net: avoid signed overflows for SO_{SND|RCV}BUFFORCE

CAP_NET_ADMIN users should not be allowed to set negative
sk_sndbuf or sk_rcvbuf values, as it can lead to various memory
corruptions, crashes, OOM...

Note that before commit 82981930125a ("net: cleanups in
sock_setsockopt()"), the bug was even more serious, since SO_SNDBUF
and SO_RCVBUF were vulnerable.

This needs to be backported to all known linux kernels.

Again, many thanks to syzkaller team for discovering this gem.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agogeneve: avoid use-after-free of skb->data
Sabrina Dubroca [Fri, 2 Dec 2016 15:49:29 +0000 (16:49 +0100)]
geneve: avoid use-after-free of skb->data

geneve{,6}_build_skb can end up doing a pskb_expand_head(), which
makes the ip_hdr(skb) reference we stashed earlier stale. Since it's
only needed as an argument to ip_tunnel_ecn_encap(), move this
directly in the function call.

Fixes: 08399efc6319 ("geneve: ensure ECN info is handled properly in all tx/rx paths")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agotipc: check minimum bearer MTU
Michal Kubeček [Fri, 2 Dec 2016 08:33:41 +0000 (09:33 +0100)]
tipc: check minimum bearer MTU

Qian Zhang (张谦) reported a potential socket buffer overflow in
tipc_msg_build() which is also known as CVE-2016-8632: due to
insufficient checks, a buffer overflow can occur if MTU is too short for
even tipc headers. As anyone can set device MTU in a user/net namespace,
this issue can be abused by a regular user.

As agreed in the discussion on Ben Hutchings' original patch, we should
check the MTU at the moment a bearer is attached rather than for each
processed packet. We also need to repeat the check when bearer MTU is
adjusted to new device MTU. UDP case also needs a check to avoid
overflow when calculating bearer MTU.

Fixes: b97bf3fd8f6a ("[TIPC] Initial merge")
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reported-by: Qian Zhang (张谦) <zhangqian-c@360.cn>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge tag 'linux-can-fixes-for-4.9-20161201' of git://git.kernel.org/pub/scm/linux...
David S. Miller [Fri, 2 Dec 2016 19:02:13 +0000 (14:02 -0500)]
Merge tag 'linux-can-fixes-for-4.9-20161201' of git://git./linux/kernel/git/mkl/linux-can

Marc Kleine-Budde says:

====================
pull-request: can 2016-12-02

this is a pull request for net/master.

There are two patches by Stephane Grosjean, who adds support for the new
PCAN-USB X6 USB interface to the pcan_usb driver.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: renesas: ravb: unintialized return value
Dan Carpenter [Thu, 1 Dec 2016 20:57:44 +0000 (23:57 +0300)]
net: renesas: ravb: unintialized return value

We want to set the other "err" variable here so that we can return it
later.  My version of GCC misses this issue but I caught it with a
static checker.

Fixes: 9f70eb339f52 ("net: ethernet: renesas: ravb: fix fixed-link phydev leaks")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Johan Hovold <johan@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge tag 'wireless-drivers-next-for-davem-2016-12-01' of git://git.kernel.org/pub...
David S. Miller [Fri, 2 Dec 2016 18:58:10 +0000 (13:58 -0500)]
Merge tag 'wireless-drivers-next-for-davem-2016-12-01' of git://git./linux/kernel/git/kvalo/wireless-drivers-next

Kalle Valo says:

====================
wireless-drivers-next patches for 4.10

Major changes:

rsi

* filter rx frames
* configure tx power
* make it possible to select antenna
* support 802.11d

brcmfmac

* cleanup of scheduled scan code
* support for bcm43341 chipset with different chip id
* support rev6 of PCIe device interface

ath10k

* add spectral scan support for QCA6174 and QCA9377 families
* show used tx bitrate with 10.4 firmware

wil6210

* add power save mode support
* add abort scan functionality
* add support settings retry limit for short frames

bcma

* add Dell Inspiron 3148
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosh_eth: remove unchecked interrupts for RZ/A1
Chris Brandt [Thu, 1 Dec 2016 18:32:14 +0000 (13:32 -0500)]
sh_eth: remove unchecked interrupts for RZ/A1

When streaming a lot of data and the RZ/A1 can't keep up, some status bits
will get set that are not being checked or cleared which cause the
following messages and the Ethernet driver to stop working. This
patch fixes that issue.

irq 21: nobody cared (try booting with the "irqpoll" option)
handlers:
[<c036b71c>] sh_eth_interrupt
Disabling IRQ #21

Fixes: db893473d313a4ad ("sh_eth: Add support for r7s72100")
Signed-off-by: Chris Brandt <chris.brandt@renesas.com>
Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: bcmgenet: Utilize correct struct device for all DMA operations
Florian Fainelli [Thu, 1 Dec 2016 17:45:45 +0000 (09:45 -0800)]
net: bcmgenet: Utilize correct struct device for all DMA operations

__bcmgenet_tx_reclaim() and bcmgenet_free_rx_buffers() are not using the
same struct device during unmap that was used for the map operation,
which makes DMA-API debugging warn about it. Fix this by always using
&priv->pdev->dev throughout the driver, using an identical device
reference for all map/unmap calls.

Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'mvneta-64bit'
David S. Miller [Fri, 2 Dec 2016 18:52:02 +0000 (13:52 -0500)]
Merge branch 'mvneta-64bit'

Gregory CLEMENT says:

====================
Support Armada 37xx SoC (ARMv8 64-bits) in mvneta driver

The Armada 37xx is a new ARMv8 SoC from Marvell using same network
controller as the older Armada 370/38x/XP SoCs. This series adapts the
driver in order to be able to use it on this new SoC. The main changes
are:

- 64-bits support: the first patches allow using the driver on a 64-bit
  architecture.

- MBUS support: the mbus configuration is different on Armada 37xx
  from the older SoCs.

- per cpu interrupt: Armada 37xx do not support per cpu interrupt for
  the NETA IP, the non-per-CPU behavior was added back.

The first patch is an optimization in the rx path in swbm mode.
The second patch remove unnecessary allocation for HWBM.
The first item is solved by patches 4 and 5.
The 2 last items are solved by patch 6.
In patch 7 the dt support is added.

Beside Armada 37xx, this series have been again tested on Armada XP
and Armada 38x (with Hardware Buffer Management and with Software
Buffer Management).

This is the 6th version of the series:
- 1st version:
http://lists.infradead.org/pipermail/linux-arm-kernel/2016-November/469588.html

- 2nd version:
http://lists.infradead.org/pipermail/linux-arm-kernel/2016-November/470476.html

- 3rd version:
http://lists.infradead.org/pipermail/linux-arm-kernel/2016-November/470901.html

- 4th version:
http://lists.infradead.org/pipermail/linux-arm-kernel/2016-November/471039.html

- 5th version:
http://lists.infradead.org/pipermail/linux-arm-kernel/2016-November/471478.html

Changelog:
v5 -> v6:
 - Added Tested-by from  Marcin Wojtas on the series
 - Added Reviewed-by from Jisheng Zhang on patch 3
 - Fix eth1 phy mode for Armada 3720 DB board on patch 7

v4 -> v5:
 - remove unnecessary cast in patch 3

v3 -> v4:
 - Adding new patch: "net: mvneta: do not allocate buffer in rxq init
   with HWBM"

 - Simplify the HWBM case in patch 3 as suggested by Marcin

v2 -> v3:
 - Adding patch 1 "Optimize rx path for small frame"

 - Fix the kbuild error by moving the "phys_addr += pp->rx_offset_correction;"
  line from patch 2 to patch 3 where rx_offset_correction is introduced.

 - Move the memory allocation of the buf_virt_addr of the rxq to be
   called by the probe function in order to avoid a memory leak.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoARM64: dts: marvell: Add network support for Armada 3700
Gregory CLEMENT [Thu, 1 Dec 2016 17:03:10 +0000 (18:03 +0100)]
ARM64: dts: marvell: Add network support for Armada 3700

Add neta nodes for network support both in device tree for the SoC and
the board.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mvneta: Add network support for Armada 3700 SoC
Marcin Wojtas [Thu, 1 Dec 2016 17:03:09 +0000 (18:03 +0100)]
net: mvneta: Add network support for Armada 3700 SoC

Armada 3700 is a new ARMv8 SoC from Marvell using same network controller
as older Armada 370/38x/XP. There are however some differences that
needed taking into account when adding support for it:

* open default MBUS window to 4GB of DRAM - Armada 3700 SoC's Mbus
  configuration for network controller has to be done on two levels:
  global and per-port. The first one is inherited from the
  bootloader. The latter can be opened in a default way, leaving
  arbitration to the bus controller.  Hence filled mbus_dram_target_info
  structure is not needed

* make per-CPU operation optional - Recent patches adding RSS and XPS
  support for Armada 38x/XP enabled per-CPU operation of the controller
  by default. Contrary to older SoC's Armada 3700 SoC's network
  controller is not capable of per-CPU processing due to interrupt lines'
  connectivity.  This patch restores non-per-CPU operation, which is now
  optional and depends on neta_armada3700 flag value in mvneta_port
  structure. In order not to complicate the code, separate interrupt
  subroutine is implemented.

For now, on the Armada 3700, RSS is disabled as the current
implementation depend on the per cpu interrupts.

[gregory.clement@free-electrons.com: extract from a larger patch, replace
some ifdef and port to net-next for v4.10]

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mvneta: Only disable mvneta_bm for 64-bits
Gregory CLEMENT [Thu, 1 Dec 2016 17:03:08 +0000 (18:03 +0100)]
net: mvneta: Only disable mvneta_bm for 64-bits

Actually only the mvneta_bm support is not 64-bits compatible.
The mvneta code itself can run on 64-bits architecture.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mvneta: Convert to be 64 bits compatible
Marcin Wojtas [Thu, 1 Dec 2016 17:03:07 +0000 (18:03 +0100)]
net: mvneta: Convert to be 64 bits compatible

Prepare the mvneta driver in order to be usable on the 64 bits platform
such as the Armada 3700.

[gregory.clement@free-electrons.com]: this patch was extract from a larger
one to ease review and maintenance.

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mvneta: Use cacheable memory to store the rx buffer virtual address
Gregory CLEMENT [Thu, 1 Dec 2016 17:03:06 +0000 (18:03 +0100)]
net: mvneta: Use cacheable memory to store the rx buffer virtual address

Until now the virtual address of the received buffer were stored in the
cookie field of the rx descriptor. However, this field is 32-bits only
which prevents to use the driver on a 64-bits architecture.

With this patch the virtual address is stored in an array not shared with
the hardware (no more need to use the DMA API). Thanks to this, it is
possible to use cache contrary to the access of the rx descriptor member.

The change is done in the swbm path only because the hwbm uses the cookie
field, this also means that currently the hwbm is not usable in 64-bits.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Reviewed-by: Jisheng Zhang <jszhang@marvell.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mvneta: Do not allocate buffer in rxq init with HWBM
Gregory CLEMENT [Thu, 1 Dec 2016 17:03:05 +0000 (18:03 +0100)]
net: mvneta: Do not allocate buffer in rxq init with HWBM

For HWBM all buffers are allocated in mvneta_bm_construct() and in runtime
they are put into descriptors by hardware. There is no need to fill them
at this point.

Suggested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mvneta: Optimize rx path for small frame
Gregory CLEMENT [Thu, 1 Dec 2016 17:03:04 +0000 (18:03 +0100)]
net: mvneta: Optimize rx path for small frame

For small frame reuse the phys_addr variable instead of accessing the
uncacheable value in the rx descriptor.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoFix up a couple of field names in the CREDITS file
Linus Torvalds [Fri, 2 Dec 2016 18:48:50 +0000 (10:48 -0800)]
Fix up a couple of field names in the CREDITS file

Ozgur Karatas reported that the very first entry in the CREDITS file had
the wrong tag for name (M: instead of N: - it happened when moving the
entry from the MAINTAINERS file, where 'M:' stands for "Maintainer").

And when I went looking, I found a couple of other cases of wrong
tagging too.

Reported-by: Ozgur Karatas <mueddib@yandex.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
8 years agoMerge branch 'bpf-support-for-sockets'
David S. Miller [Fri, 2 Dec 2016 18:46:21 +0000 (13:46 -0500)]
Merge branch 'bpf-support-for-sockets'

David Ahern says:

====================
net: Add bpf support for sockets

The recently added VRF support in Linux leverages the bind-to-device
API for programs to specify an L3 domain for a socket. While
SO_BINDTODEVICE has been around for ages, not every ipv4/ipv6 capable
program has support for it. Even for those programs that do support it,
the API requires processes to be started as root (CAP_NET_RAW) which
is not desirable from a general security perspective.

This patch set leverages Daniel Mack's work to attach bpf programs to
a cgroup to provide a capability to set sk_bound_dev_if for all
AF_INET{6} sockets opened by a process in a cgroup when the sockets
are allocated.

For example:
 1. configure vrf (e.g., using ifupdown2)
        auto eth0
        iface eth0 inet dhcp
            vrf mgmt

        auto mgmt
        iface mgmt
            vrf-table auto

 2. configure cgroup
        mount -t cgroup2 none /tmp/cgroupv2
        mkdir /tmp/cgroupv2/mgmt
        test_cgrp2_sock /tmp/cgroupv2/mgmt 15

 3. set shell into cgroup (e.g., can be done at login using pam)
        echo $$ >> /tmp/cgroupv2/mgmt/cgroup.procs

At this point all commands run in the shell (e.g, apt) have sockets
automatically bound to the VRF (see output of ss -ap 'dev == <vrf>'),
including processes not running as root.

This capability enables running any program in a VRF context and is key
to deploying Management VRF, a fundamental configuration for networking
gear, with any Linux OS installation.

This patchset also exports the socket family, type and protocol as
read-only allowing bpf filters to deny a process in a cgroup the ability
to open specific types of AF_INET or AF_INET6 sockets.

v7
- comments from Alexei

v6
- add export of socket family, type and protocol
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosamples/bpf: add userspace example for prohibiting sockets
David Ahern [Thu, 1 Dec 2016 16:48:08 +0000 (08:48 -0800)]
samples/bpf: add userspace example for prohibiting sockets

Add examples preventing a process in a cgroup from opening a socket
based family, protocol and type.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosamples/bpf: Update bpf loader for cgroup section names
David Ahern [Thu, 1 Dec 2016 16:48:07 +0000 (08:48 -0800)]
samples/bpf: Update bpf loader for cgroup section names

Add support for section names starting with cgroup/skb and cgroup/sock.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobpf: Add support for reading socket family, type, protocol
David Ahern [Thu, 1 Dec 2016 16:48:06 +0000 (08:48 -0800)]
bpf: Add support for reading socket family, type, protocol

Add socket family, type and protocol to bpf_sock allowing bpf programs
read-only access.

Add __sk_flags_offset[0] to struct sock before the bitfield to
programmtically determine the offset of the unsigned int containing
protocol and type.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agosamples: bpf: add userspace example for modifying sk_bound_dev_if
David Ahern [Thu, 1 Dec 2016 16:48:05 +0000 (08:48 -0800)]
samples: bpf: add userspace example for modifying sk_bound_dev_if

Add a simple program to demonstrate the ability to attach a bpf program
to a cgroup that sets sk_bound_dev_if for AF_INET{6} sockets when they
are created.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobpf: Add new cgroup attach type to enable sock modifications
David Ahern [Thu, 1 Dec 2016 16:48:04 +0000 (08:48 -0800)]
bpf: Add new cgroup attach type to enable sock modifications

Add new cgroup based program type, BPF_PROG_TYPE_CGROUP_SOCK. Similar to
BPF_PROG_TYPE_CGROUP_SKB programs can be attached to a cgroup and run
any time a process in the cgroup opens an AF_INET or AF_INET6 socket.
Currently only sk_bound_dev_if is exported to userspace for modification
by a bpf program.

This allows a cgroup to be configured such that AF_INET{6} sockets opened
by processes are automatically bound to a specific device. In turn, this
enables the running of programs that do not support SO_BINDTODEVICE in a
specific VRF context / L3 domain.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobpf: Refactor cgroups code in prep for new type
David Ahern [Thu, 1 Dec 2016 16:48:03 +0000 (08:48 -0800)]
bpf: Refactor cgroups code in prep for new type

Code move and rename only; no functional change intended.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoNET: usb: qmi_wwan: add support for Telit LE922A PID 0x1040
Daniele Palmas [Thu, 1 Dec 2016 15:52:05 +0000 (16:52 +0100)]
NET: usb: qmi_wwan: add support for Telit LE922A PID 0x1040

This patch adds support for PID 0x1040 of Telit LE922A.

The qmi adapter requires to have DTR set for proper working,
so QMI_WWAN_QUIRK_DTR has been enabled.

Signed-off-by: Daniele Palmas <dnlplm@gmail.com>
Acked-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agocdc_ether: Fix handling connection notification
Kristian Evensen [Thu, 1 Dec 2016 13:23:17 +0000 (14:23 +0100)]
cdc_ether: Fix handling connection notification

Commit bfe9b9d2df66 ("cdc_ether: Improve ZTE MF823/831/910 handling")
introduced a work-around in usbnet_cdc_status() for devices that exported
cdc carrier on twice on connect. Before the commit, this behavior caused
the link state to be incorrect. It was assumed that all CDC Ethernet
devices would either export this behavior, or send one off and then one on
notification (which seems to be the default behavior).

Unfortunately, it turns out multiple devices sends a connection
notification multiple times per second (via an interrupt), even when
connection state does not change. This has been observed with several
different USB LAN dongles (at least), for example 13b1:0041 (Linksys).
After bfe9b9d2df66, the link state has been set as down and then up for
each notification. This has caused a flood of Netlink NEWLINK messages and
syslog to be flooded with messages similar to:

cdc_ether 2-1:2.0 eth1: kevent 12 may have been dropped

This commit fixes the behavior by reverting usbnet_cdc_status() to how it
was before bfe9b9d2df66. The work-around has been moved to a separate
status-function which is only called when a known, affect device is
detected.

v1->v2:

* Do not open-code netif_carrier_ok() (thanks Henning Schild).
* Call netif_carrier_off() instead of usb_link_change(). This prevents
calling schedule_work() twice without giving the work queue a chance to be
processed (thanks Bjørn Mork).

Fixes: bfe9b9d2df66 ("cdc_ether: Improve ZTE MF823/831/910 handling")
Reported-by: Henning Schild <henning.schild@siemens.com>
Signed-off-by: Kristian Evensen <kristian.evensen@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoip6_offload: check segs for NULL in ipv6_gso_segment.
Artem Savkov [Thu, 1 Dec 2016 13:06:04 +0000 (14:06 +0100)]
ip6_offload: check segs for NULL in ipv6_gso_segment.

segs needs to be checked for being NULL in ipv6_gso_segment() before calling
skb_shinfo(segs), otherwise kernel can run into a NULL-pointer dereference:

[   97.811262] BUG: unable to handle kernel NULL pointer dereference at 00000000000000cc
[   97.819112] IP: [<ffffffff816e52f9>] ipv6_gso_segment+0x119/0x2f0
[   97.825214] PGD 0 [   97.827047]
[   97.828540] Oops: 0000 [#1] SMP
[   97.831678] Modules linked in: vhost_net vhost macvtap macvlan nfsv3 rpcsec_gss_krb5
nfsv4 dns_resolver nfs fscache xt_CHECKSUM iptable_mangle ipt_MASQUERADE nf_nat_masquerade_ipv4
iptable_nat nf_nat_ipv4 nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack
ipt_REJECT nf_reject_ipv4 tun ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter
bridge stp llc snd_hda_codec_realtek snd_hda_codec_hdmi snd_hda_codec_generic snd_hda_intel
snd_hda_codec edac_mce_amd snd_hda_core edac_core snd_hwdep kvm_amd snd_seq kvm snd_seq_device
snd_pcm irqbypass snd_timer ppdev parport_serial snd parport_pc k10temp pcspkr soundcore parport
sp5100_tco shpchp sg wmi i2c_piix4 acpi_cpufreq nfsd auth_rpcgss nfs_acl lockd grace sunrpc
ip_tables xfs libcrc32c sr_mod cdrom sd_mod ata_generic pata_acpi amdkfd amd_iommu_v2 radeon
broadcom bcm_phy_lib i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops
ttm ahci serio_raw tg3 firewire_ohci libahci pata_atiixp drm ptp libata firewire_core pps_core
i2c_core crc_itu_t fjes dm_mirror dm_region_hash dm_log dm_mod
[   97.927721] CPU: 1 PID: 3504 Comm: vhost-3495 Not tainted 4.9.0-7.el7.test.x86_64 #1
[   97.935457] Hardware name: AMD Snook/Snook, BIOS ESK0726A 07/26/2010
[   97.941806] task: ffff880129a1c080 task.stack: ffffc90001bcc000
[   97.947720] RIP: 0010:[<ffffffff816e52f9>]  [<ffffffff816e52f9>] ipv6_gso_segment+0x119/0x2f0
[   97.956251] RSP: 0018:ffff88012fc43a10  EFLAGS: 00010207
[   97.961557] RAX: 0000000000000000 RBX: ffff8801292c8700 RCX: 0000000000000594
[   97.968687] RDX: 0000000000000593 RSI: ffff880129a846c0 RDI: 0000000000240000
[   97.975814] RBP: ffff88012fc43a68 R08: ffff880129a8404e R09: 0000000000000000
[   97.982942] R10: 0000000000000000 R11: ffff880129a84076 R12: 00000020002949b3
[   97.990070] R13: ffff88012a580000 R14: 0000000000000000 R15: ffff88012a580000
[   97.997198] FS:  0000000000000000(0000) GS:ffff88012fc40000(0000) knlGS:0000000000000000
[   98.005280] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   98.011021] CR2: 00000000000000cc CR3: 0000000126c5d000 CR4: 00000000000006e0
[   98.018149] Stack:
[   98.020157]  00000000ffffffff ffff88012fc43ac8 ffffffffa017ad0a 000000000000000e
[   98.027584]  0000001300000000 0000000077d59998 ffff8801292c8700 00000020002949b3
[   98.035010]  ffff88012a580000 0000000000000000 ffff88012a580000 ffff88012fc43a98
[   98.042437] Call Trace:
[   98.044879]  <IRQ> [   98.046803]  [<ffffffffa017ad0a>] ? tg3_start_xmit+0x84a/0xd60 [tg3]
[   98.053156]  [<ffffffff815eeee0>] skb_mac_gso_segment+0xb0/0x130
[   98.059158]  [<ffffffff815eefd3>] __skb_gso_segment+0x73/0x110
[   98.064985]  [<ffffffff815ef40d>] validate_xmit_skb+0x12d/0x2b0
[   98.070899]  [<ffffffff815ef5d2>] validate_xmit_skb_list+0x42/0x70
[   98.077073]  [<ffffffff81618560>] sch_direct_xmit+0xd0/0x1b0
[   98.082726]  [<ffffffff815efd86>] __dev_queue_xmit+0x486/0x690
[   98.088554]  [<ffffffff8135c135>] ? cpumask_next_and+0x35/0x50
[   98.094380]  [<ffffffff815effa0>] dev_queue_xmit+0x10/0x20
[   98.099863]  [<ffffffffa09ce057>] br_dev_queue_push_xmit+0xa7/0x170 [bridge]
[   98.106907]  [<ffffffffa09ce161>] br_forward_finish+0x41/0xc0 [bridge]
[   98.113430]  [<ffffffff81627cf2>] ? nf_iterate+0x52/0x60
[   98.118735]  [<ffffffff81627d6b>] ? nf_hook_slow+0x6b/0xc0
[   98.124216]  [<ffffffffa09ce32c>] __br_forward+0x14c/0x1e0 [bridge]
[   98.130480]  [<ffffffffa09ce120>] ? br_dev_queue_push_xmit+0x170/0x170 [bridge]
[   98.137785]  [<ffffffffa09ce4bd>] br_forward+0x9d/0xb0 [bridge]
[   98.143701]  [<ffffffffa09cfbb7>] br_handle_frame_finish+0x267/0x560 [bridge]
[   98.150834]  [<ffffffffa09d0064>] br_handle_frame+0x174/0x2f0 [bridge]
[   98.157355]  [<ffffffff8102fb89>] ? sched_clock+0x9/0x10
[   98.162662]  [<ffffffff810b63b2>] ? sched_clock_cpu+0x72/0xa0
[   98.168403]  [<ffffffff815eccf5>] __netif_receive_skb_core+0x1e5/0xa20
[   98.174926]  [<ffffffff813659f9>] ? timerqueue_add+0x59/0xb0
[   98.180580]  [<ffffffff815ed548>] __netif_receive_skb+0x18/0x60
[   98.186494]  [<ffffffff815ee625>] process_backlog+0x95/0x140
[   98.192145]  [<ffffffff815edccd>] net_rx_action+0x16d/0x380
[   98.197713]  [<ffffffff8170cff1>] __do_softirq+0xd1/0x283
[   98.203106]  [<ffffffff8170b2bc>] do_softirq_own_stack+0x1c/0x30
[   98.209107]  <EOI> [   98.211029]  [<ffffffff8108a5c0>] do_softirq+0x50/0x60
[   98.216166]  [<ffffffff815ec853>] netif_rx_ni+0x33/0x80
[   98.221386]  [<ffffffffa09eeff7>] tun_get_user+0x487/0x7f0 [tun]
[   98.227388]  [<ffffffffa09ef3ab>] tun_sendmsg+0x4b/0x60 [tun]
[   98.233129]  [<ffffffffa0b68932>] handle_tx+0x282/0x540 [vhost_net]
[   98.239392]  [<ffffffffa0b68c25>] handle_tx_kick+0x15/0x20 [vhost_net]
[   98.245916]  [<ffffffffa0abacfe>] vhost_worker+0x9e/0xf0 [vhost]
[   98.251919]  [<ffffffffa0abac60>] ? vhost_umem_alloc+0x40/0x40 [vhost]
[   98.258440]  [<ffffffff81003a47>] ? do_syscall_64+0x67/0x180
[   98.264094]  [<ffffffff810a44d9>] kthread+0xd9/0xf0
[   98.268965]  [<ffffffff810a4400>] ? kthread_park+0x60/0x60
[   98.274444]  [<ffffffff8170a4d5>] ret_from_fork+0x25/0x30
[   98.279836] Code: 8b 93 d8 00 00 00 48 2b 93 d0 00 00 00 4c 89 e6 48 89 df 66 89 93 c2 00 00 00 ff 10 48 3d 00 f0 ff ff 49 89 c2 0f 87 52 01 00 00 <41> 8b 92 cc 00 00 00 48 8b 80 d0 00 00 00 44 0f b7 74 10 06 66
[   98.299425] RIP  [<ffffffff816e52f9>] ipv6_gso_segment+0x119/0x2f0
[   98.305612]  RSP <ffff88012fc43a10>
[   98.309094] CR2: 00000000000000cc
[   98.312406] ---[ end trace 726a2c7a2d2d78d0 ]---

Signed-off-by: Artem Savkov <asavkov@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomlx4: fix use-after-free in mlx4_en_fold_software_stats()
Eric Dumazet [Thu, 1 Dec 2016 13:02:06 +0000 (05:02 -0800)]
mlx4: fix use-after-free in mlx4_en_fold_software_stats()

My recent commit to get more precise rx/tx counters in ndo_get_stats64()
can lead to crashes at device dismantle, as Jesper found out.

We must prevent mlx4_en_fold_software_stats() trying to access
tx/rx rings if they are deleted.

Fix this by adding a test against priv->port_up in
mlx4_en_fold_software_stats()

Calling mlx4_en_fold_software_stats() from mlx4_en_stop_port()
allows us to eventually broadcast the latest/current counters to
rtnetlink monitors.

Fixes: 40931b85113d ("mlx4: give precise rx/tx bytes/packets counters")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-and-bisected-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Cc: Saeed Mahameed <saeedm@dev.mellanox.co.il>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: thunderx: Fix transmit queue timeout issue
Sunil Goutham [Thu, 1 Dec 2016 12:54:28 +0000 (18:24 +0530)]
net: thunderx: Fix transmit queue timeout issue

Transmit queue timeout issue is seen in two cases
- Due to a race condition btw setting stop_queue at xmit()
  and checking for stopped_queue in NAPI poll routine, at times
  transmission from a SQ comes to a halt. This is fixed
  by using barriers and also added a check for SQ free descriptors,
  incase SQ is stopped and there are only CQE_RX i.e no CQE_TX.
- Contrary to an assumption, a HW errata where HW doesn't stop transmission
  even though there are not enough CQEs available for a CQE_TX is
  not fixed in T88 pass 2.x. This results in a Qset error with
  'CQ_WR_FULL' stalling transmission. This is fixed by adjusting
  RXQ's  RED levels for CQ level such that there is always enough
  space left for CQE_TXs.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoRDS: TCP: unregister_netdevice_notifier() in error path of rds_tcp_init_net
Sowmini Varadhan [Thu, 1 Dec 2016 12:44:43 +0000 (04:44 -0800)]
RDS: TCP: unregister_netdevice_notifier() in error path of rds_tcp_init_net

If some error is encountered in rds_tcp_init_net, make sure to
unregister_netdevice_notifier(), else we could trigger a panic
later on, when the modprobe from a netns fails.

Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'offloading-tc-rules-hw'
David S. Miller [Fri, 2 Dec 2016 18:28:38 +0000 (13:28 -0500)]
Merge branch 'offloading-tc-rules-hw'

Hadar Hen Zion says:

====================
Offloading tc rules using underline Hardware device

This series adds flower classifier support in offloading tc rules when the
Software ingress device is different from the Hardware ingress device,
such as when dealing with IP tunnels

The first two patches are a small fixes to flower, checking the skip_hw flag
wasn't set before calling the Hardware offloading functions which will try to
offload the rule.

The next two patches are infrastructure patches, a preparation for the fourth
patch which is adding support in flower to offload rules when the ingress
device is not a Hardware device and therefore can't offload.
In this case ndo_setup_tc is called with the mirred (egress) device.

The last three patchs are adding mlx5e support to offload rules using the new
"egress_device" flag.

Thanks,
Hadar

Changes from v0:
- check if CONFIG_NET_CLS_ACT is defined befor calling tc_action_ops get_dev()
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Support adding ingress tc rule when egress device flag is set
Hadar Hen Zion [Thu, 1 Dec 2016 12:06:40 +0000 (14:06 +0200)]
net/mlx5e: Support adding ingress tc rule when egress device flag is set

When ndo_setup_tc is called with an egress_dev flag set, it means that
the ndo call was executed on the mirred action (egress) device and not
on the ingress device.

In order to support this kind of ndo_setup_tc call, and insert the
correct decap rule to the hardware, the uplink device on the same eswitch
should be found.

Currently, we use this resolution between the mirred device and the
uplink on the same eswitch to offload vxlan shared device decap rules.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Save the represntor netdevice as part of the representor
Hadar Hen Zion [Thu, 1 Dec 2016 12:06:39 +0000 (14:06 +0200)]
net/mlx5e: Save the represntor netdevice as part of the representor

Replace the representor private data to a net_device pointer holding the
representor netdevice, instead of void pointer holding mlx5e_priv.

It will be used by a new eswitch service function, returning the uplink representor
netdevice.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Bring back representor's ndos that were accidentally removed
Hadar Hen Zion [Thu, 1 Dec 2016 12:06:38 +0000 (14:06 +0200)]
net/mlx5e: Bring back representor's ndos that were accidentally removed

The VF Representor udp tunnel ndo entries were removed by mistake,
return them.

Fixes: 370bad0f9a52 ('net/mlx5e: Support HW (offloaded) and SW counters for SRIOV switchdev mode')
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/sched: cls_flower: Add offload support using egress Hardware device
Hadar Hen Zion [Thu, 1 Dec 2016 12:06:37 +0000 (14:06 +0200)]
net/sched: cls_flower: Add offload support using egress Hardware device

In order to support hardware offloading when the device given by the tc
rule is different from the Hardware underline device, extract the mirred
(egress) device from the tc action when a filter is added, using the new
tc_action_ops, get_dev().

Flower caches the information about the mirred device and use it for
calling ndo_setup_tc in filter change, update stats and delete.

Calling ndo_setup_tc of the mirred (egress) device instead of the
ingress device will allow a resolution between the software ingress
device and the underline hardware device.

The resolution will take place inside the offloading driver using
'egress_device' flag added to tc_to_netdev struct which is provided to
the offloading driver.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>