sfrench/cifs-2.6.git
22 months agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf
David S. Miller [Tue, 6 Nov 2018 01:19:25 +0000 (17:19 -0800)]
Merge git://git./pub/scm/linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

The following patchset contains the first batch of Netfilter fixes for
your net tree:

1) Fix splat with IPv6 defragmenting locally generated fragments,
   from Florian Westphal.

2) Fix Incorrect check for missing attribute in nft_osf.

3) Missing INT_MIN & INT_MAX definition for netfilter bridge uapi
   header, from Jiri Slaby.

4) Revert map lookup in nft_numgen, this is already possible with
   the existing infrastructure without this extension.

5) Fix wrong listing of set reference counter, make counter
   synchronous again, from Stefano Brivio.

6) Fix CIDR 0 in hash:net,port,net, from Eric Westbrook.

7) Fix allocation failure with large set, use kvcalloc().
   From Andrey Ryabinin.

8) No need to disable BH when fetch ip set comment, patch from
   Jozsef Kadlecsik.

9) Sanity check for valid sysfs entry in xt_IDLETIMER, from
   Taehee Yoo.

10) Fix suspicious rcu usage via ip_set() macro at netlink dump,
    from Jozsef Kadlecsik.

11) Fix setting default timeout via nfnetlink_cttimeout, this
    comes with preparation patch to add nf_{tcp,udp,...}_pernet()
    helper.

12) Allow ebtables table nat to be of filter type via nft_compat.
    From Florian Westphal.

13) Incorrect calculation of next bucket in early_drop, do no bump
    hash value, update bucket counter instead. From Vasily Khoruzhick.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: alx: make alx_drv_name static
Rasmus Villemoes [Mon, 5 Nov 2018 17:52:21 +0000 (18:52 +0100)]
net: alx: make alx_drv_name static

alx_drv_name is not used outside main.c, so there's no reason for it to
have external linkage.

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: bpfilter: fix iptables failure if bpfilter_umh is disabled
Taehee Yoo [Mon, 5 Nov 2018 13:31:41 +0000 (22:31 +0900)]
net: bpfilter: fix iptables failure if bpfilter_umh is disabled

When iptables command is executed, ip_{set/get}sockopt() try to upload
bpfilter.ko if bpfilter is enabled. if it couldn't find bpfilter.ko,
command is failed.
bpfilter.ko is generated if CONFIG_BPFILTER_UMH is enabled.
ip_{set/get}sockopt() only checks CONFIG_BPFILTER.
So that if CONFIG_BPFILTER is enabled and CONFIG_BPFILTER_UMH is disabled,
iptables command is always failed.

test config:
   CONFIG_BPFILTER=y
   # CONFIG_BPFILTER_UMH is not set

test command:
   %iptables -L
   iptables: No chain/target/match by that name.

Fixes: d2ba09c17a06 ("net: add skeleton of bpfilter kernel module")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agosock_diag: fix autoloading of the raw_diag module
Andrei Vagin [Mon, 5 Nov 2018 06:37:15 +0000 (22:37 -0800)]
sock_diag: fix autoloading of the raw_diag module

IPPROTO_RAW isn't registred as an inet protocol, so
inet_protos[protocol] is always NULL for it.

Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Xin Long <lucien.xin@gmail.com>
Fixes: bf2ae2e4bf93 ("sock_diag: request _diag module only when the family or proto has been registered")
Signed-off-by: Andrei Vagin <avagin@gmail.com>
Reviewed-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: core: netpoll: Enable netconsole IPv6 link local address
Matwey V. Kornilov [Fri, 2 Nov 2018 18:19:36 +0000 (21:19 +0300)]
net: core: netpoll: Enable netconsole IPv6 link local address

There is no reason to discard using source link local address when
remote netconsole IPv6 address is set to be link local one.

The patch allows administrators to use IPv6 netconsole without
explicitly configuring source address:

    netconsole=@/,@fe80::5054:ff:fe2f:6012/

Signed-off-by: Matwey V. Kornilov <matwey@sai.msu.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoipv6: properly check return value in inet6_dump_all()
Alexey Kodanev [Fri, 2 Nov 2018 16:11:05 +0000 (19:11 +0300)]
ipv6: properly check return value in inet6_dump_all()

Make sure we call fib6_dump_end() if it happens that skb->len
is zero. rtnl_dump_all() can reset cb->args on the next loop
iteration there.

Fixes: 08e814c9e8eb ("net/ipv6: Bail early if user only wants cloned entries")
Fixes: ae677bbb4441 ("net: Don't return invalid table id error when dumping all families")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agortnetlink: restore handling of dumpit return value in rtnl_dump_all()
Alexey Kodanev [Fri, 2 Nov 2018 16:11:04 +0000 (19:11 +0300)]
rtnetlink: restore handling of dumpit return value in rtnl_dump_all()

For non-zero return from dumpit() we should break the loop
in rtnl_dump_all() and return the result. Otherwise, e.g.,
we could get the memory leak in inet6_dump_fib() [1]. The
pointer to the allocated struct fib6_walker there (saved
in cb->args) can be lost, reset on the next iteration.

Fix it by partially restoring the previous behavior before
commit c63586dc9b3e ("net: rtnl_dump_all needs to propagate
error from dumpit function"). The returned error from
dumpit() is still passed further.

[1]:
unreferenced object 0xffff88001322a200 (size 96):
  comm "sshd", pid 1484, jiffies 4296032768 (age 1432.542s)
  hex dump (first 32 bytes):
    00 01 00 00 00 00 ad de 00 02 00 00 00 00 ad de  ................
    18 09 41 36 00 88 ff ff 18 09 41 36 00 88 ff ff  ..A6......A6....
  backtrace:
    [<0000000095846b39>] kmem_cache_alloc_trace+0x151/0x220
    [<000000007d12709f>] inet6_dump_fib+0x68d/0x940
    [<000000002775a316>] rtnl_dump_all+0x1d9/0x2d0
    [<00000000d7cd302b>] netlink_dump+0x945/0x11a0
    [<000000002f43485f>] __netlink_dump_start+0x55d/0x800
    [<00000000f76bbeec>] rtnetlink_rcv_msg+0x4fa/0xa00
    [<000000009b5761f3>] netlink_rcv_skb+0x29c/0x420
    [<0000000087a1dae1>] rtnetlink_rcv+0x15/0x20
    [<00000000691b703b>] netlink_unicast+0x4e3/0x6c0
    [<00000000b5be0204>] netlink_sendmsg+0x7f2/0xba0
    [<0000000096d2aa60>] sock_sendmsg+0xba/0xf0
    [<000000008c1b786f>] __sys_sendto+0x1e4/0x330
    [<0000000019587b3f>] __x64_sys_sendto+0xe1/0x1a0
    [<00000000071f4d56>] do_syscall_64+0x9f/0x300
    [<000000002737577f>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
    [<0000000057587684>] 0xffffffffffffffff

Fixes: c63586dc9b3e ("net: rtnl_dump_all needs to propagate error from dumpit function")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet/ipv6: Move anycast init/cleanup functions out of CONFIG_PROC_FS
Jeff Barnhill [Mon, 5 Nov 2018 20:36:45 +0000 (20:36 +0000)]
net/ipv6: Move anycast init/cleanup functions out of CONFIG_PROC_FS

Move the anycast.c init and cleanup functions which were inadvertently
added inside the CONFIG_PROC_FS definition.

Fixes: 2384d02520ff ("net/ipv6: Add anycast addresses to a global hashtable")
Signed-off-by: Jeff Barnhill <0xeffeff@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agobonding/802.3ad: fix link_failure_count tracking
Jarod Wilson [Sun, 4 Nov 2018 19:59:46 +0000 (14:59 -0500)]
bonding/802.3ad: fix link_failure_count tracking

Commit 4d2c0cda07448ea6980f00102dc3964eb25e241c set slave->link to
BOND_LINK_DOWN for 802.3ad bonds whenever invalid speed/duplex values
were read, to fix a problem with slaves getting into weird states, but
in the process, broke tracking of link failures, as going straight to
BOND_LINK_DOWN when a link is indeed down (cable pulled, switch rebooted)
means we broke out of bond_miimon_inspect()'s BOND_LINK_DOWN case because
!link_state was already true, we never incremented commit, and never got
a chance to call bond_miimon_commit(), where slave->link_failure_count
would be incremented. I believe the simple fix here is to mark the slave
as BOND_LINK_FAIL, and let bond_miimon_inspect() transition the link from
_FAIL to either _UP or _DOWN, and in the latter case, we now get proper
incrementing of link_failure_count again.

Fixes: 4d2c0cda0744 ("bonding: speed/duplex update at NETDEV_UP event")
CC: Mahesh Bandewar <maheshb@google.com>
CC: David S. Miller <davem@davemloft.net>
CC: netdev@vger.kernel.org
CC: stable@vger.kernel.org
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: phy: realtek: fix RTL8201F sysfs name
Holger Hoffstätte [Sun, 4 Nov 2018 18:02:42 +0000 (19:02 +0100)]
net: phy: realtek: fix RTL8201F sysfs name

Since 4.19 the following error in sysfs has appeared when using the
r8169 NIC driver:

$cd /sys/module/realtek/drivers
$ls -l
ls: cannot access 'mdio_bus:RTL8201F 10/100Mbps Ethernet': No such file or directory
[..garbled dir entries follow..]

Apparently the forward slash in "10/100Mbps Ethernet" is interpreted
as directory separator that leads nowhere, and was introduced in commit
513588dd44b ("net: phy: realtek: add RTL8201F phy-id and functions").

Fix this by removing the offending slash in the driver name.

Other drivers in net/phy seem to have the same problem, but I cannot
test/verify them.

Fixes: 513588dd44b ("net: phy: realtek: add RTL8201F phy-id and functions")
Signed-off-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agosctp: define SCTP_SS_DEFAULT for Stream schedulers
Xin Long [Sat, 3 Nov 2018 06:01:31 +0000 (14:01 +0800)]
sctp: define SCTP_SS_DEFAULT for Stream schedulers

According to rfc8260#section-4.3.2, SCTP_SS_DEFAULT is required to
defined as SCTP_SS_FCFS or SCTP_SS_RR.

SCTP_SS_FCFS is used for SCTP_SS_DEFAULT's value in this patch.

Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations")
Reported-by: Jianwen Ji <jiji@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agosctp: fix strchange_flags name for Stream Change Event
Xin Long [Sat, 3 Nov 2018 05:59:45 +0000 (13:59 +0800)]
sctp: fix strchange_flags name for Stream Change Event

As defined in rfc6525#section-6.1.3, SCTP_STREAM_CHANGE_DENIED
and SCTP_STREAM_CHANGE_FAILED should be used instead of
SCTP_ASSOC_CHANGE_DENIED and SCTP_ASSOC_CHANGE_FAILED.

To keep the compatibility, fix it by adding two macros.

Fixes: b444153fb5a6 ("sctp: add support for generating add stream change event notification")
Reported-by: Jianwen Ji <jiji@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
David S. Miller [Sun, 4 Nov 2018 02:33:35 +0000 (19:33 -0700)]
Merge git://git./pub/scm/linux/kernel/git/bpf/bpf

Daniel Borkmann says:

====================
pull-request: bpf 2018-11-03

The following pull-request contains BPF updates for your *net* tree.

The main changes are:

1) Fix BPF prog kallsyms and bpf_prog_get_info_by_fd() jited address export
   to not use page start but actual start instead to work correctly with
   profiling e.g. finding hot instructions from stack traces. Also fix latter
   with regards to dumping address and jited len for !multi prog, from Song.

2) Fix bpf_prog_get_info_by_fd() for !root to zero info.nr_jited_func_lens
   instead of leaving the user provided length, from Daniel.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agomlxsw: spectrum: Fix IP2ME CPU policer configuration
Shalom Toledo [Fri, 2 Nov 2018 19:49:15 +0000 (19:49 +0000)]
mlxsw: spectrum: Fix IP2ME CPU policer configuration

The CPU policer used to police packets being trapped via a local route
(IP2ME) was incorrectly configured to police based on bytes per second
instead of packets per second.

Change the policer to police based on packets per second and avoid
packet loss under certain circumstances.

Fixes: 9148e7cf73ce ("mlxsw: spectrum: Add policers for trap groups")
Signed-off-by: Shalom Toledo <shalomt@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoopenvswitch: fix linking without CONFIG_NF_CONNTRACK_LABELS
Arnd Bergmann [Fri, 2 Nov 2018 15:36:55 +0000 (16:36 +0100)]
openvswitch: fix linking without CONFIG_NF_CONNTRACK_LABELS

When CONFIG_CC_OPTIMIZE_FOR_DEBUGGING is enabled, the compiler
fails to optimize out a dead code path, which leads to a link failure:

net/openvswitch/conntrack.o: In function `ovs_ct_set_labels':
conntrack.c:(.text+0x2e60): undefined reference to `nf_connlabels_replace'

In this configuration, we can take a shortcut, and completely
remove the contrack label code. This may also help the regular
optimization.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoqed: fix link config error handling
Arnd Bergmann [Fri, 2 Nov 2018 15:36:24 +0000 (16:36 +0100)]
qed: fix link config error handling

gcc-8 notices that qed_mcp_get_transceiver_data() may fail to
return a result to the caller:

drivers/net/ethernet/qlogic/qed/qed_mcp.c: In function 'qed_mcp_trans_speed_mask':
drivers/net/ethernet/qlogic/qed/qed_mcp.c:1955:2: error: 'transceiver_type' may be used uninitialized in this function [-Werror=maybe-uninitialized]

When an error is returned by qed_mcp_get_transceiver_data(), we
should propagate that to the caller of qed_mcp_trans_speed_mask()
rather than continuing with uninitialized data.

Fixes: c56a8be7e7aa ("qed: Add supported link and advertise link to display in ethtool.")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: Fix for out-of-bounds access when setting pfc back pressure
Yunsheng Lin [Fri, 2 Nov 2018 09:47:48 +0000 (17:47 +0800)]
net: hns3: Fix for out-of-bounds access when setting pfc back pressure

The vport should be initialized to hdev->vport for each bp group,
otherwise it will cause out-of-bounds access and bp setting not
correct problem.

[   35.254124] BUG: KASAN: slab-out-of-bounds in hclge_pause_setup_hw+0x2a0/0x3f8 [hclge]
[   35.254126] Read of size 2 at addr ffff803b6651581a by task kworker/0:1/14

[   35.254132] CPU: 0 PID: 14 Comm: kworker/0:1 Not tainted 4.19.0-rc7-hulk+ #85
[   35.254133] Hardware name: Huawei D06/D06, BIOS Hisilicon D06 UEFI RC0 - B052 (V0.52) 09/14/2018
[   35.254141] Workqueue: events work_for_cpu_fn
[   35.254144] Call trace:
[   35.254147]  dump_backtrace+0x0/0x2f0
[   35.254149]  show_stack+0x24/0x30
[   35.254154]  dump_stack+0x110/0x184
[   35.254157]  print_address_description+0x168/0x2b0
[   35.254160]  kasan_report+0x184/0x310
[   35.254162]  __asan_load2+0x7c/0xa0
[   35.254170]  hclge_pause_setup_hw+0x2a0/0x3f8 [hclge]
[   35.254177]  hclge_tm_init_hw+0x794/0x9f0 [hclge]
[   35.254184]  hclge_tm_schd_init+0x48/0x58 [hclge]
[   35.254191]  hclge_init_ae_dev+0x778/0x1168 [hclge]
[   35.254196]  hnae3_register_ae_dev+0x14c/0x298 [hnae3]
[   35.254206]  hns3_probe+0x88/0xa8 [hns3]
[   35.254210]  local_pci_probe+0x7c/0xf0
[   35.254212]  work_for_cpu_fn+0x34/0x50
[   35.254214]  process_one_work+0x4d4/0xa38
[   35.254216]  worker_thread+0x55c/0x8d8
[   35.254219]  kthread+0x1b0/0x1b8
[   35.254222]  ret_from_fork+0x10/0x1c

[   35.254224] The buggy address belongs to the page:
[   35.254228] page:ffff7e00ed994400 count:1 mapcount:0 mapping:0000000000000000 index:0x0 compound_mapcount: 0
[   35.273835] flags: 0xfffff8000008000(head)
[   35.282007] raw: 0fffff8000008000 dead000000000100 dead000000000200 0000000000000000
[   35.282010] raw: 0000000000000000 0000000000000000 00000001ffffffff 0000000000000000
[   35.282012] page dumped because: kasan: bad access detected

[   35.282014] Memory state around the buggy address:
[   35.282017]  ffff803b66515700: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe
[   35.282019]  ffff803b66515780: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe
[   35.282021] >ffff803b66515800: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe
[   35.282022]                             ^
[   35.282024]  ffff803b66515880: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe
[   35.282026]  ffff803b66515900: fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe fe
[   35.282028] ==================================================================
[   35.282029] Disabling lock debugging due to kernel taint
[   35.282747] hclge driver initialization finished.

Fixes: 67bf2541f4b9 ("net: hns3: Fixes the back pressure setting when sriov is enabled")
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge branch 'net-bql-better-deal-with-GSO'
David S. Miller [Sat, 3 Nov 2018 22:40:01 +0000 (15:40 -0700)]
Merge branch 'net-bql-better-deal-with-GSO'

Eric Dumazet says:

====================
net: bql: better deal with GSO

While BQL bulk dequeue works well for TSO packets, it is
not very efficient as soon as GSO is involved.

On a GSO only workload (UDP or TCP), this patch series
can save about 8 % of cpu cycles on a 40Gbit mlx4 NIC,
by keeping optimal batching, and avoiding expensive
doorbells, qdisc requeues and reschedules.

This patch series :

- Add __netdev_tx_sent_queue() so that drivers
  can implement efficient BQL and xmit_more support.

- Implement a work around in dev_hard_start_xmit()
  for drivers not using __netdev_tx_sent_queue()

- changes mlx4 to use __netdev_tx_sent_queue()

v2: Tariq and Willem feedback addressed.
    added __netdev_tx_sent_queue() (Willem suggestion)
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet/mlx4_en: use __netdev_tx_sent_queue()
Eric Dumazet [Wed, 31 Oct 2018 15:39:14 +0000 (08:39 -0700)]
net/mlx4_en: use __netdev_tx_sent_queue()

doorbell only depends on xmit_more and netif_tx_queue_stopped()

Using __netdev_tx_sent_queue() avoids messing with BQL stop flag,
and is more generic.

This patch increases performance on GSO workload by keeping
doorbells to the minimum required.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: do not abort bulk send on BQL status
Eric Dumazet [Wed, 31 Oct 2018 15:39:13 +0000 (08:39 -0700)]
net: do not abort bulk send on BQL status

Before calling dev_hard_start_xmit(), upper layers tried
to cook optimal skb list based on BQL budget.

Problem is that GSO packets can end up comsuming more than
the BQL budget.

Breaking the loop is not useful, since requeued packets
are ahead of any packets still in the qdisc.

It is also more expensive, since next TX completion will
push these packets later, while skbs are not in cpu caches.

It is also a behavior difference with TSO packets, that can
break the BQL limit by a large amount.

Note that drivers should use __netdev_tx_sent_queue()
in order to have optimal xmit_more support, and avoid
useless atomic operations as shown in the following patch.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: bql: add __netdev_tx_sent_queue()
Eric Dumazet [Wed, 31 Oct 2018 15:39:12 +0000 (08:39 -0700)]
net: bql: add __netdev_tx_sent_queue()

When qdisc_run() tries to use BQL budget to bulk-dequeue a batch
of packets, GSO can later transform this list in another list
of skbs, and each skb is sent to device ndo_start_xmit(),
one at a time, with skb->xmit_more being set to one but
for last skb.

Problem is that very often, BQL limit is hit in the middle of
the packet train, forcing dev_hard_start_xmit() to stop the
bulk send and requeue the end of the list.

BQL role is to avoid head of line blocking, making sure
a qdisc can deliver high priority packets before low priority ones.

But there is no way requeued packets can be bypassed by fresh
packets in the qdisc.

Aborting the bulk send increases TX softirqs, and hot cache
lines (after skb_segment()) are wasted.

Note that for TSO packets, we never split a packet in the middle
because of BQL limit being hit.

Drivers should be able to update BQL counters without
flipping/caring about BQL status, if the current skb
has xmit_more set.

Upper layers are ultimately responsible to stop sending another
packet train when BQL limit is hit.

Code template in a driver might look like the following :

send_doorbell = __netdev_tx_sent_queue(tx_queue, nr_bytes, skb->xmit_more);

Note that __netdev_tx_sent_queue() use is not mandatory,
since following patch will change dev_hard_start_xmit()
to not care about BQL status.

But it is highly recommended so that xmit_more full benefits
can be reached (less doorbells sent, and less atomic operations as well)

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge branch 's390-qeth-fixes'
David S. Miller [Sat, 3 Nov 2018 17:44:06 +0000 (10:44 -0700)]
Merge branch 's390-qeth-fixes'

Julian Wiedmann says:

====================
s390/qeth: fixes 2018-11-02

please apply one round of qeth fixes for -net.

Patch 1 is rather large and removes a use-after-free hazard from many of our
debug trace entries.
Patch 2 is yet another fix-up for the L3 subdriver's new IP address management
code.
Patch 3 and 4 resolve some fallout from the recent changes wrt how/when qeth
allocates its net_device.
Patch 5 makes sure we don't set reserved bits when building HW commands from
user-provided data.
And finally, patch 6 allows ethtool to play nice with new HW.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agos390/qeth: report 25Gbit link speed
Julian Wiedmann [Fri, 2 Nov 2018 18:04:13 +0000 (19:04 +0100)]
s390/qeth: report 25Gbit link speed

This adds the various identifiers for 25Gbit cards, and wires them up
into sysfs and ethtool.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agos390/qeth: sanitize ARP requests
Julian Wiedmann [Fri, 2 Nov 2018 18:04:12 +0000 (19:04 +0100)]
s390/qeth: sanitize ARP requests

The ARP_{ADD,REMOVE}_ENTRY cmd structs contain reserved fields.
Introduce a common helper that doesn't raw-copy the user-provided data
into the cmd, but only sets those fields that are strictly needed for
the command.

This also sets the correct command length for ARP_REMOVE_ENTRY.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agos390/qeth: fix initial operstate
Julian Wiedmann [Fri, 2 Nov 2018 18:04:11 +0000 (19:04 +0100)]
s390/qeth: fix initial operstate

Setting the carrier 'on' for an unregistered netdevice doesn't update
its operstate. Fix this by delaying the update until the netdevice has
been registered.

Fixes: 91cc98f51e3d ("s390/qeth: remove duplicated carrier state tracking")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agos390/qeth: unregister netdevice only when registered
Julian Wiedmann [Fri, 2 Nov 2018 18:04:10 +0000 (19:04 +0100)]
s390/qeth: unregister netdevice only when registered

qeth only registers its netdevice when the qeth device is first set
online. Thus a device that has never been set online will trigger
a WARN ("network todo 'hsi%d' but state 0") in unregister_netdev() when
removed.

Fix this by protecting the unregister step, just like we already protect
against repeated registering of the netdevice.

Fixes: d3d1b205e89f ("s390/qeth: allocate netdevice early")
Reported-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agos390/qeth: fix HiperSockets sniffer
Julian Wiedmann [Fri, 2 Nov 2018 18:04:09 +0000 (19:04 +0100)]
s390/qeth: fix HiperSockets sniffer

Sniffing mode for L3 HiperSockets requires that no IP addresses are
registered with the HW. The preferred way to achieve this is for
userspace to delete all the IPs on the interface. But qeth is expected
to also tolerate a configuration where that is not the case, by skipping
the IP registration when in sniffer mode.
Since commit 5f78e29ceebf ("qeth: optimize IP handling in rx_mode callback")
reworked the IP registration logic in the L3 subdriver, this no longer
works. When the qeth device is set online, qeth_l3_recover_ip() now
unconditionally registers all unicast addresses from our internal
IP table.

While we could fix this particular problem by skipping
qeth_l3_recover_ip() on a sniffer device, the more future-proof change
is to skip the IP address registration at the lowest level. This way we
a) catch any future code path that attempts to register an IP address
   without considering the sniffer scenario, and
b) continue to build up our internal IP table, so that if sniffer mode
   is switched off later we can operate just like normal.

Fixes: 5f78e29ceebf ("qeth: optimize IP handling in rx_mode callback")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agos390/qeth: sanitize strings in debug messages
Julian Wiedmann [Fri, 2 Nov 2018 18:04:08 +0000 (19:04 +0100)]
s390/qeth: sanitize strings in debug messages

As Documentation/s390/s390dbf.txt states quite clearly, using any
pointer in sprinf-formatted s390dbf debug entries is dangerous.
The pointers are dereferenced whenever the trace file is read from.
So if the referenced data has a shorter life-time than the trace file,
any read operation can result in a use-after-free.

So rip out all hazardous use of indirect data, and replace any usage of
dev_name() and such by the Bus ID number.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonetfilter: conntrack: fix calculation of next bucket number in early_drop
Vasily Khoruzhick [Thu, 25 Oct 2018 19:15:43 +0000 (12:15 -0700)]
netfilter: conntrack: fix calculation of next bucket number in early_drop

If there's no entry to drop in bucket that corresponds to the hash,
early_drop() should look for it in other buckets. But since it increments
hash instead of bucket number, it actually looks in the same bucket 8
times: hsize is 16k by default (14 bits) and hash is 32-bit value, so
reciprocal_scale(hash, hsize) returns the same value for hash..hash+7 in
most cases.

Fix it by increasing bucket number instead of hash and rename _hash
to bucket to avoid future confusion.

Fixes: 3e86638e9a0b ("netfilter: conntrack: consider ct netns in early_drop logic")
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Vasily Khoruzhick <vasilykh@arista.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: nft_compat: ebtables 'nat' table is normal chain type
Florian Westphal [Fri, 2 Nov 2018 10:33:37 +0000 (11:33 +0100)]
netfilter: nft_compat: ebtables 'nat' table is normal chain type

Unlike ip(6)tables, the ebtables nat table has no special properties.
This bug causes 'ebtables -A' to fail when using a target such as
'snat' (ebt_snat target sets ".table = "nat"').  Targets that have
no table restrictions work fine.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: nfnetlink_cttimeout: pass default timeout policy to obj_to_nlattr
Pablo Neira Ayuso [Thu, 1 Nov 2018 23:14:00 +0000 (00:14 +0100)]
netfilter: nfnetlink_cttimeout: pass default timeout policy to obj_to_nlattr

Otherwise, we hit a NULL pointer deference since handlers always assume
default timeout policy is passed.

  netlink: 24 bytes leftover after parsing attributes in process `syz-executor2'.
  kasan: CONFIG_KASAN_INLINE enabled
  kasan: GPF could be caused by NULL-ptr deref or user memory access
  general protection fault: 0000 [#1] PREEMPT SMP KASAN
  CPU: 0 PID: 9575 Comm: syz-executor1 Not tainted 4.19.0+ #312
  Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
  RIP: 0010:icmp_timeout_obj_to_nlattr+0x77/0x170 net/netfilter/nf_conntrack_proto_icmp.c:297

Fixes: c779e849608a ("netfilter: conntrack: remove get_timeout() indirection")
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: conntrack: add nf_{tcp,udp,sctp,icmp,dccp,icmpv6,generic}_pernet()
Pablo Neira Ayuso [Thu, 1 Nov 2018 23:11:34 +0000 (00:11 +0100)]
netfilter: conntrack: add nf_{tcp,udp,sctp,icmp,dccp,icmpv6,generic}_pernet()

Expose these functions to access conntrack protocol tracker netns area,
nfnetlink_cttimeout needs this.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: ipset: Fix calling ip_set() macro at dumping
Jozsef Kadlecsik [Tue, 30 Oct 2018 21:43:42 +0000 (22:43 +0100)]
netfilter: ipset: Fix calling ip_set() macro at dumping

The ip_set() macro is called when either ip_set_ref_lock held only
or no lock/nfnl mutex is held at dumping. Take this into account
properly. Also, use Pablo's suggestion to use rcu_dereference_raw(),
the ref_netlink protects the set.

Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: xt_IDLETIMER: add sysfs filename checking routine
Taehee Yoo [Sat, 20 Oct 2018 15:00:08 +0000 (00:00 +0900)]
netfilter: xt_IDLETIMER: add sysfs filename checking routine

When IDLETIMER rule is added, sysfs file is created under
/sys/class/xt_idletimer/timers/
But some label name shouldn't be used.
".", "..", "power", "uevent", "subsystem", etc...
So that sysfs filename checking routine is needed.

test commands:
   %iptables -I INPUT -j IDLETIMER --timeout 1 --label "power"

splat looks like:
[95765.423132] sysfs: cannot create duplicate filename '/devices/virtual/xt_idletimer/timers/power'
[95765.433418] CPU: 0 PID: 8446 Comm: iptables Not tainted 4.19.0-rc6+ #20
[95765.449755] Call Trace:
[95765.449755]  dump_stack+0xc9/0x16b
[95765.449755]  ? show_regs_print_info+0x5/0x5
[95765.449755]  sysfs_warn_dup+0x74/0x90
[95765.449755]  sysfs_add_file_mode_ns+0x352/0x500
[95765.449755]  sysfs_create_file_ns+0x179/0x270
[95765.449755]  ? sysfs_add_file_mode_ns+0x500/0x500
[95765.449755]  ? idletimer_tg_checkentry+0x3e5/0xb1b [xt_IDLETIMER]
[95765.449755]  ? rcu_read_lock_sched_held+0x114/0x130
[95765.449755]  ? __kmalloc_track_caller+0x211/0x2b0
[95765.449755]  ? memcpy+0x34/0x50
[95765.449755]  idletimer_tg_checkentry+0x4e2/0xb1b [xt_IDLETIMER]
[ ... ]

Fixes: 0902b469bd25 ("netfilter: xtables: idletimer target implementation")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: ipset: Correct rcu_dereference() call in ip_set_put_comment()
Jozsef Kadlecsik [Fri, 19 Oct 2018 17:35:19 +0000 (19:35 +0200)]
netfilter: ipset: Correct rcu_dereference() call in ip_set_put_comment()

The function is called when rcu_read_lock() is held and not
when rcu_read_lock_bh() is held.

Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agoMerge branch 'net-timeout-fixes-for-GENET-and-SYSTEMPORT'
David S. Miller [Sat, 3 Nov 2018 07:03:40 +0000 (00:03 -0700)]
Merge branch 'net-timeout-fixes-for-GENET-and-SYSTEMPORT'

Florian Fainelli says:

====================
net: timeout fixes for GENET and SYSTEMPORT

This patch series fixes occasional transmit timeout around the time
the system goes into suspend. GENET and SYSTEMPORT have nearly the same
logic in that regard and were both affected in the same way.

Please queue up for stable, thanks!
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: systemport: Protect stop from timeout
Florian Fainelli [Thu, 1 Nov 2018 22:55:38 +0000 (15:55 -0700)]
net: systemport: Protect stop from timeout

A timing hazard exists when the network interface is stopped that
allows a watchdog timeout to be processed by a separate core in
parallel. This creates the potential for the timeout handler to
wake the queues while the driver is shutting down, or access
registers after their clocks have been removed.

The more common case is that the watchdog timeout will produce a
warning message which doesn't lead to a crash. The chances of this
are greatly increased by the fact that bcm_sysport_netif_stop stops
the transmit queues which can easily precipitate a watchdog time-
out because of stale trans_start data in the queues.

This commit corrects the behavior by ensuring that the watchdog
timeout is disabled before enterring bcm_sysport_netif_stop. There
are currently only two users of the bcm_sysport_netif_stop function:
close and suspend.

The close case already handles the issue by exiting the RUNNING
state before invoking the driver close service.

The suspend case now performs the netif_device_detach to exit the
PRESENT state before the call to bcm_sysport_netif_stop rather than
after it.

These behaviors prevent any future scheduling of the driver timeout
service during the window. The netif_tx_stop_all_queues function
in bcm_sysport_netif_stop is replaced with netif_tx_disable to ensure
synchronization with any transmit or timeout threads that may
already be executing on other cores.

For symmetry, the netif_device_attach call upon resume is moved to
after the call to bcm_sysport_netif_start. Since it wakes the transmit
queues it is not necessary to invoke netif_tx_start_all_queues from
bcm_sysport_netif_start so it is moved into the driver open service.

Fixes: 40755a0fce17 ("net: systemport: add suspend and resume support")
Fixes: 80105befdb4b ("net: systemport: add Broadcom SYSTEMPORT Ethernet MAC driver")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: bcmgenet: protect stop from timeout
Doug Berger [Thu, 1 Nov 2018 22:55:37 +0000 (15:55 -0700)]
net: bcmgenet: protect stop from timeout

A timing hazard exists when the network interface is stopped that
allows a watchdog timeout to be processed by a separate core in
parallel. This creates the potential for the timeout handler to
wake the queues while the driver is shutting down, or access
registers after their clocks have been removed.

The more common case is that the watchdog timeout will produce a
warning message which doesn't lead to a crash. The chances of this
are greatly increased by the fact that bcmgenet_netif_stop stops
the transmit queues which can easily precipitate a watchdog time-
out because of stale trans_start data in the queues.

This commit corrects the behavior by ensuring that the watchdog
timeout is disabled before enterring bcmgenet_netif_stop. There
are currently only two users of the bcmgenet_netif_stop function:
close and suspend.

The close case already handles the issue by exiting the RUNNING
state before invoking the driver close service.

The suspend case now performs the netif_device_detach to exit the
PRESENT state before the call to bcmgenet_netif_stop rather than
after it.

These behaviors prevent any future scheduling of the driver timeout
service during the window. The netif_tx_stop_all_queues function
in bcmgenet_netif_stop is replaced with netif_tx_disable to ensure
synchronization with any transmit or timeout threads that may
already be executing on other cores.

For symmetry, the netif_device_attach call upon resume is moved to
after the call to bcmgenet_netif_start. Since it wakes the transmit
queues it is not necessary to invoke netif_tx_start_all_queues from
bcmgenet_netif_start so it is moved into the driver open service.

Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file")
Signed-off-by: Doug Berger <opendmb@gmail.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agorxrpc: Fix lockup due to no error backoff after ack transmit error
David Howells [Thu, 1 Nov 2018 13:39:53 +0000 (13:39 +0000)]
rxrpc: Fix lockup due to no error backoff after ack transmit error

If the network becomes (partially) unavailable, say by disabling IPv6, the
background ACK transmission routine can get itself into a tizzy by
proposing immediate ACK retransmission.  Since we're in the call event
processor, that happens immediately without returning to the workqueue
manager.

The condition should clear after a while when either the network comes back
or the call times out.

Fix this by:

 (1) When re-proposing an ACK on failed Tx, don't schedule it immediately.
     This will allow a certain amount of time to elapse before we try
     again.

 (2) Enforce a return to the workqueue manager after a certain number of
     iterations of the call processing loop.

 (3) Add a backoff delay that increases the delay on deferred ACKs by a
     jiffy per failed transmission to a limit of HZ.  The backoff delay is
     cleared on a successful return from kernel_sendmsg().

 (4) Cancel calls immediately if the opening sendmsg fails.  The layer
     above can arrange retransmission or rotate to another server.

Fixes: 248f219cb8bc ("rxrpc: Rewrite the data and ack handling code")
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: dsa: microchip: initialize mutex before use
Tristram Ha [Sat, 3 Nov 2018 02:23:41 +0000 (19:23 -0700)]
net: dsa: microchip: initialize mutex before use

Initialize mutex before use.  Avoid kernel complaint when
CONFIG_DEBUG_LOCK_ALLOC is enabled.

Fixes: b987e98e50ab90e5 ("dsa: add DSA switch driver for Microchip KSZ9477")
Signed-off-by: Tristram Ha <Tristram.Ha@microchip.com>
Reviewed-by: Pavel Machek <pavel@ucw.cz>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet/ipv6: Add anycast addresses to a global hashtable
Jeff Barnhill [Fri, 2 Nov 2018 20:23:57 +0000 (20:23 +0000)]
net/ipv6: Add anycast addresses to a global hashtable

icmp6_send() function is expensive on systems with a large number of
interfaces. Every time it’s called, it has to verify that the source
address does not correspond to an existing anycast address by looping
through every device and every anycast address on the device.  This can
result in significant delays for a CPU when there are a large number of
neighbors and ND timers are frequently timing out and calling
neigh_invalidate().

Add anycast addresses to a global hashtable to allow quick searching for
matching anycast addresses.  This is based on inet6_addr_lst in addrconf.c.

Signed-off-by: Jeff Barnhill <0xeffeff@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agousbnet: smsc95xx: disable carrier check while suspending
Frieder Schrempf [Wed, 31 Oct 2018 21:52:19 +0000 (22:52 +0100)]
usbnet: smsc95xx: disable carrier check while suspending

We need to make sure, that the carrier check polling is disabled
while suspending. Otherwise we can end up with usbnet_read_cmd()
being issued when only usbnet_read_cmd_nopm() is allowed. If this
happens, read operations lock up.

Fixes: d69d169493 ("usbnet: smsc95xx: fix link detection for disabled autonegotiation")
Signed-off-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Reviewed-by: Raghuram Chary J <RaghuramChary.Jallipalli@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: document skb parameter in function 'skb_gso_size_check'
Mathieu Malaterre [Wed, 31 Oct 2018 12:16:58 +0000 (13:16 +0100)]
net: document skb parameter in function 'skb_gso_size_check'

Remove kernel-doc warning:

  net/core/skbuff.c:4953: warning: Function parameter or member 'skb' not described in 'skb_gso_size_check'

Signed-off-by: Mathieu Malaterre <malat@debian.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agobpf: fix bpf_prog_get_info_by_fd to return 0 func_lens for unpriv
Daniel Borkmann [Fri, 2 Nov 2018 10:35:46 +0000 (11:35 +0100)]
bpf: fix bpf_prog_get_info_by_fd to return 0 func_lens for unpriv

While dbecd7388476 ("bpf: get kernel symbol addresses via syscall")
zeroed info.nr_jited_ksyms in bpf_prog_get_info_by_fd() for queries
from unprivileged users, commit 815581c11cc2 ("bpf: get JITed image
lengths of functions via syscall") forgot about doing so and therefore
returns the #elems of the user set up buffer which is incorrect. It
also needs to indicate a info.nr_jited_func_lens of zero.

Fixes: 815581c11cc2 ("bpf: get JITed image lengths of functions via syscall")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Sandipan Das <sandipan@linux.vnet.ibm.com>
Cc: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
22 months agoMerge branch 'bpf-accurate-prog-addr'
Daniel Borkmann [Fri, 2 Nov 2018 20:39:02 +0000 (21:39 +0100)]
Merge branch 'bpf-accurate-prog-addr'

Song Liu says:

====================
Changes v1 -> v2:
1. Added main program length to bpf_prog_info->jited_fun_lens (3/3).
2. Updated commit message of 1/3 and 2/3 with more background about the
   address masking, and why it is still save after the changes.
3. Replace "ulong" with "unsigned long".

This set improves bpf program address showed in /proc/kallsyms and in
bpf_prog_info. First, real program address is showed instead of page
address. Second, when there is no subprogram, bpf_prog_info->jited_ksyms
and bpf_prog_info->jited_fun_lens returns the main prog address and
length.
====================

Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
22 months agobpf: show main program address and length in bpf_prog_info
Song Liu [Fri, 2 Nov 2018 17:16:17 +0000 (10:16 -0700)]
bpf: show main program address and length in bpf_prog_info

Currently, when there is no subprog (prog->aux->func_cnt == 0),
bpf_prog_info does not return any jited_ksyms or jited_func_lens. This
patch adds main program address (prog->bpf_func) and main program
length (prog->jited_len) to bpf_prog_info.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
22 months agobpf: show real jited address in bpf_prog_info->jited_ksyms
Song Liu [Fri, 2 Nov 2018 17:16:16 +0000 (10:16 -0700)]
bpf: show real jited address in bpf_prog_info->jited_ksyms

Currently, jited_ksyms in bpf_prog_info shows page addresses of jited
bpf program. The main reason here is to not expose randomized start
address. However, this is not ideal for detailed profiling (find hot
instructions from stack traces). This patch replaces the page address
with real prog start address.

This change is OK because bpf_prog_get_info_by_fd() is only available
to root.

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
22 months agobpf: show real jited prog address in /proc/kallsyms
Song Liu [Fri, 2 Nov 2018 17:16:15 +0000 (10:16 -0700)]
bpf: show real jited prog address in /proc/kallsyms

Currently, /proc/kallsyms shows page address of jited bpf program. The
main reason here is to not expose randomized start address. However,
This is not ideal for detailed profiling (find hot instructions from
stack traces). This patch replaces the page address with real prog start
address.

This change is OK because these addresses are still protected by sysctl
kptr_restrict (see kallsyms_show_value()), and only programs loaded by
root are added to kallsyms (see bpf_prog_kallsyms_add()).

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
22 months agonet: drop skb on failure in ip_check_defrag()
Cong Wang [Thu, 1 Nov 2018 19:02:37 +0000 (12:02 -0700)]
net: drop skb on failure in ip_check_defrag()

Most callers of pskb_trim_rcsum() simply drop the skb when
it fails, however, ip_check_defrag() still continues to pass
the skb up to stack. This is suspicious.

In ip_check_defrag(), after we learn the skb is an IP fragment,
passing the skb to callers makes no sense, because callers expect
fragments are defrag'ed on success. So, dropping the skb when we
can't defrag it is reasonable.

Note, prior to commit 88078d98d1bb, this is not a big problem as
checksum will be fixed up anyway. After it, the checksum is not
correct on failure.

Found this during code review.

Fixes: 88078d98d1bb ("net: pskb_trim_rcsum() and CHECKSUM_COMPLETE are friends")
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge branch 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa...
Linus Torvalds [Thu, 1 Nov 2018 16:31:19 +0000 (09:31 -0700)]
Merge branch 'i2c/for-next' of git://git./linux/kernel/git/wsa/linux

Pull i2c fixes from Wolfram Sang:
 "I2C has a core bugfix & cleanup as well as an ID addition and
  MAINTAINERS update for you"

* 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  MAINTAINERS: add maintainer for IMX LPI2C driver
  dt-bindings: i2c: i2c-imx-lpi2c: add imx8qxp compatible string
  i2c: Clear client->irq in i2c_device_remove
  i2c: Remove unnecessary call to irq_find_mapping

22 months agoMerge branch 'for-4.20' of git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu
Linus Torvalds [Thu, 1 Nov 2018 16:27:57 +0000 (09:27 -0700)]
Merge branch 'for-4.20' of git://git./linux/kernel/git/dennis/percpu

Pull percpu fixes from Dennis Zhou:
 "Two small things for v4.20.

  The first fixes a clang uninitialized variable warning for arm64 in
  the default path calls BUILD_BUG(). The second removes an unnecessary
  unlikely() in a WARN_ON() use"

* 'for-4.20' of git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu:
  arm64: percpu: Initialize ret in the default case
  mm: percpu: remove unnecessary unlikely()

22 months agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Thu, 1 Nov 2018 16:16:01 +0000 (09:16 -0700)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) BPF verifier fixes from Daniel Borkmann.

 2) HNS driver fixes from Huazhong Tan.

 3) FDB only works for ethernet devices, reject attempts to install FDB
    rules for others. From Ido Schimmel.

 4) Fix spectre V1 in vhost, from Jason Wang.

 5) Don't pass on-stack object to irq_set_affinity_hint() in mvpp2
    driver, from Marc Zyngier.

 6) Fix mlx5e checksum handling when RXFCS is enabled, from Eric
    Dumazet.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (49 commits)
  openvswitch: Fix push/pop ethernet validation
  net: stmmac: Fix stmmac_mdio_reset() when building stmmac as modules
  bpf: test make sure to run unpriv test cases in test_verifier
  bpf: add various test cases to test_verifier
  bpf: don't set id on after map lookup with ptr_to_map_val return
  bpf: fix partial copy of map_ptr when dst is scalar
  libbpf: Fix compile error in libbpf_attach_type_by_name
  kselftests/bpf: use ping6 as the default ipv6 ping binary if it exists
  selftests: mlxsw: qos_mc_aware: Add a test for UC awareness
  selftests: mlxsw: qos_mc_aware: Tweak for min shaper
  mlxsw: spectrum: Set minimum shaper on MC TCs
  mlxsw: reg: QEEC: Add minimum shaper fields
  net: hns3: bugfix for rtnl_lock's range in the hclgevf_reset()
  net: hns3: bugfix for rtnl_lock's range in the hclge_reset()
  net: hns3: bugfix for handling mailbox while the command queue reinitialized
  net: hns3: fix incorrect return value/type of some functions
  net: hns3: bugfix for hclge_mdio_write and hclge_mdio_read
  net: hns3: bugfix for is_valid_csq_clean_head()
  net: hns3: remove unnecessary queue reset in the hns3_uninit_all_ring()
  net: hns3: bugfix for the initialization of command queue's spin lock
  ...

22 months agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc
Linus Torvalds [Thu, 1 Nov 2018 16:07:04 +0000 (09:07 -0700)]
Merge git://git./linux/kernel/git/davem/sparc

Pull sparc fixes from David Miller:
 "Two small fixes"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
  sparc64: Wire up compat getpeername and getsockname.
  sparc64: Remvoe set_fs() from perf_callchain_user().

22 months agoMerge tag 'csky-for-linus-4.20-fixup-dtb' of https://github.com/c-sky/csky-linux
Linus Torvalds [Thu, 1 Nov 2018 16:04:30 +0000 (09:04 -0700)]
Merge tag 'csky-for-linus-4.20-fixup-dtb' of https://github.com/c-sky/csky-linux

Pull csky dtb fixups from Guo Ren:
 "These fix the csky dtb Kbuild to follow the new Devicetree dtb build
  rules"

* tag 'csky-for-linus-4.20-fixup-dtb' of https://github.com/c-sky/csky-linux:
  csky: use common dtb build rules
  csky: remove builtin-dtb Kbuild

22 months agoMerge tag 'platform-drivers-x86-v4.20-1' of git://git.infradead.org/linux-platform...
Linus Torvalds [Thu, 1 Nov 2018 15:42:21 +0000 (08:42 -0700)]
Merge tag 'platform-drivers-x86-v4.20-1' of git://git.infradead.org/linux-platform-drivers-x86

Pull x86 platform driver updates from Darren Hart:

 - Move the Dell dcdbas and dell_rbu drivers into platform/drivers/x86
   as they are closely coupled with other drivers in this location.

 - Improve _init* usage for acerhdf and fix some usage issues with
   messages and module parameters.

 - Simplify asus-wmi by calling ACPI/WMI methods directly, eliminating
   workqueue overhead, eliminate double reporting of keyboard backlight.

 - Fix wake from USB failure on Bay Trail devices (intel_int0002_vgpio).

 - Notify intel_telemetry users when IPC1 device is not enabled.

 - Update various drivers with new laptop model IDs.

 - Update several intel drivers to use SPDX identifers and order headers
   alphabetically.

* tag 'platform-drivers-x86-v4.20-1' of git://git.infradead.org/linux-platform-drivers-x86: (64 commits)
  HID: asus: only support backlight when it's not driven by WMI
  platform/x86: asus-wmi: export function for evaluating WMI methods
  platform/x86: asus-wmi: Only notify kbd LED hw_change by fn-key pressed
  platform/x86: wmi: declare device_type structure as constant
  platform/x86: ideapad: Add Y530-15ICH to no_hw_rfkill
  platform/x86: Add Intel AtomISP2 dummy / power-management driver
  platform/x86: touchscreen_dmi: Add min-x and min-y settings for various models
  platform/x86: touchscreen_dmi: Add info for the Onda V80 Plus v3 tablet
  platform/x86: touchscreen_dmi: Add info for the Trekstor Primetab T13B tablet
  platform/x86: intel_telemetry: Get rid of custom macro
  platform/x86: intel_telemetry: report debugfs failure
  MAINTAINERS: intel_telemetry: Update maintainers info
  platform/x86: Add LG Gram laptop special features driver
  platform/x86: asus-wmi: Simplify the keyboard brightness updating process
  platform/x86: touchscreen_dmi: Add info for the Trekstor Primebook C11 convertible
  platform/x86: mlx-platform: Properly use mlxplat_mlxcpld_msn201x_items
  MAINTAINERS: intel_pmc_core: Update MAINTAINERS
  firmware: dcdbas: include linux/io.h
  platform/x86: intel-wmi-thunderbolt: Add dynamic debugging
  platform/x86: intel-wmi-thunderbolt: Convert to use SPDX identifier
  ...

22 months agocsky: use common dtb build rules
Guo Ren [Tue, 30 Oct 2018 10:46:47 +0000 (18:46 +0800)]
csky: use common dtb build rules

Remove the Kbuild rules in arch/csky and use common dtb build rules.

This modification is based on:

commit 37c8a5fafa3b ("kbuild: consolidate Devicetree dtb build rules")

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
22 months agocsky: remove builtin-dtb Kbuild
Guo Ren [Tue, 30 Oct 2018 10:42:32 +0000 (18:42 +0800)]
csky: remove builtin-dtb Kbuild

Remove the builtin-dtb implementation in arch/csky.

Signed-off-by: Guo Ren <ren_guo@c-sky.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
22 months agoopenvswitch: Fix push/pop ethernet validation
Jaime Caamaño Ruiz [Wed, 31 Oct 2018 17:52:03 +0000 (18:52 +0100)]
openvswitch: Fix push/pop ethernet validation

When there are both pop and push ethernet header actions among the
actions to be applied to a packet, an unexpected EINVAL (Invalid
argument) error is obtained. This is due to mac_proto not being reset
correctly when those actions are validated.

Reported-at:
https://mail.openvswitch.org/pipermail/ovs-discuss/2018-October/047554.html
Fixes: 91820da6ae85 ("openvswitch: add Ethernet push and pop actions")
Signed-off-by: Jaime Caamaño Ruiz <jcaamano@suse.com>
Tested-by: Greg Rose <gvrose8192@gmail.com>
Reviewed-by: Greg Rose <gvrose8192@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: stmmac: Fix stmmac_mdio_reset() when building stmmac as modules
Niklas Cassel [Wed, 31 Oct 2018 15:08:10 +0000 (16:08 +0100)]
net: stmmac: Fix stmmac_mdio_reset() when building stmmac as modules

When building stmmac, it is only possible to select CONFIG_DWMAC_GENERIC,
or any of the glue drivers, when CONFIG_STMMAC_PLATFORM is set.
The only exception is CONFIG_STMMAC_PCI.

When calling of_mdiobus_register(), it will call our ->reset()
callback, which is set to stmmac_mdio_reset().

Most of the code in stmmac_mdio_reset() is protected by a
"#if defined(CONFIG_STMMAC_PLATFORM)", which will evaluate
to false when CONFIG_STMMAC_PLATFORM=m.

Because of this, the phy reset gpio will only be pulled when
stmmac is built as built-in, but not when built as modules.

Fix this by using "#if IS_ENABLED()" instead of "#if defined()".

Signed-off-by: Niklas Cassel <niklas.cassel@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agosparc64: Wire up compat getpeername and getsockname.
David S. Miller [Thu, 1 Nov 2018 01:30:21 +0000 (18:30 -0700)]
sparc64: Wire up compat getpeername and getsockname.

Fixes: 8b30ca73b7cc ("sparc: Add all necessary direct socket system calls.")
Reported-by: Joseph Myers <joseph@codesourcery.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge branch '10GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net...
David S. Miller [Thu, 1 Nov 2018 01:21:37 +0000 (18:21 -0700)]
Merge branch '10GbE' of git://git./linux/kernel/git/jkirsher/net-queue

Jeff Kirsher says:

====================
Intel Wired LAN Driver Updates 2018-10-31

This series contains a various collection of fixes.

Miroslav Lichvar from Red Hat or should I say IBM now?  Updates the PHC
timecounter interval for igb so that it gets updated at least once
every 550 seconds.

Ngai-Mint provides a fix for fm10k to prevent a soft lockup or system
crash by adding a new condition to determine if the SM mailbox is in the
correct state before proceeding.

Jake provides several fm10k fixes, first one marks complier aborts as
non-fatal since on some platforms trigger machine check errors when the
compile aborts.  Added missing device ids to the in-kernel driver.  Due
to the recent fixes, bumped the driver version.

I (Jeff Kirsher) fixed a XFRM_ALGO dependency for both ixgbe and
ixgbevf.  This fix was based on the original work from Arnd Bergmann,
which only fixed ixgbe.

Mitch provides a fix for i40e/avf to update the status codes, which
resolves an issue between a mis-match between i40e and the iavf driver,
which also supports the ice LAN driver.

Radoslaw fixes the ixgbe where the driver is logging a message about
spoofed packets detected when the VF is re-started with a different MAC
address.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
David S. Miller [Thu, 1 Nov 2018 00:34:08 +0000 (17:34 -0700)]
Merge git://git./pub/scm/linux/kernel/git/bpf/bpf

Daniel Borkmann says:

====================
pull-request: bpf 2018-11-01

The following pull-request contains BPF updates for your *net* tree.

The main changes are:

1) Fix tcp_bpf_recvmsg() to return -EAGAIN instead of 0 in non-blocking
   case when no data is available yet, from John.

2) Fix a compilation error in libbpf_attach_type_by_name() when compiled
   with clang 3.8, from Andrey.

3) Fix a partial copy of map pointer on scalar alu and remove id
   generation for RET_PTR_TO_MAP_VALUE return types, from Daniel.

4) Add unlimited memlock limit for kernel selftest's flow_dissector_load
   program, from Yonghong.

5) Fix ping for some BPF shell based kselftests where distro does not
   ship "ping -6" anymore, from Li.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge branch 'verifier-fixes'
Alexei Starovoitov [Wed, 31 Oct 2018 23:53:18 +0000 (16:53 -0700)]
Merge branch 'verifier-fixes'

Daniel Borkmann says:

====================
The series contains two fixes in BPF core and test cases. For details
please see individual patches. Thanks!
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
22 months agobpf: test make sure to run unpriv test cases in test_verifier
Daniel Borkmann [Wed, 31 Oct 2018 23:05:55 +0000 (00:05 +0100)]
bpf: test make sure to run unpriv test cases in test_verifier

Right now unprivileged tests are never executed as a BPF test run,
only loaded. Allow for running them as well so that we can check
the outcome and probe for regressions.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
22 months agobpf: add various test cases to test_verifier
Daniel Borkmann [Wed, 31 Oct 2018 23:05:54 +0000 (00:05 +0100)]
bpf: add various test cases to test_verifier

Add some more map related test cases to test_verifier kselftest
to improve test coverage. Summary: 1012 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
22 months agobpf: don't set id on after map lookup with ptr_to_map_val return
Daniel Borkmann [Wed, 31 Oct 2018 23:05:53 +0000 (00:05 +0100)]
bpf: don't set id on after map lookup with ptr_to_map_val return

In the verifier there is no such semantics where registers with
PTR_TO_MAP_VALUE type have an id assigned to them. This is only
used in PTR_TO_MAP_VALUE_OR_NULL and later on nullified once the
test against NULL has been pattern matched and type transformed
into PTR_TO_MAP_VALUE.

Fixes: 3e6a4b3e0289 ("bpf/verifier: introduce BPF_PTR_TO_MAP_VALUE")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Roman Gushchin <guro@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
22 months agobpf: fix partial copy of map_ptr when dst is scalar
Daniel Borkmann [Wed, 31 Oct 2018 23:05:52 +0000 (00:05 +0100)]
bpf: fix partial copy of map_ptr when dst is scalar

ALU operations on pointers such as scalar_reg += map_value_ptr are
handled in adjust_ptr_min_max_vals(). Problem is however that map_ptr
and range in the register state share a union, so transferring state
through dst_reg->range = ptr_reg->range is just buggy as any new
map_ptr in the dst_reg is then truncated (or null) for subsequent
checks. Fix this by adding a raw member and use it for copying state
over to dst_reg.

Fixes: f1174f77b50c ("bpf/verifier: rework value tracking")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Edward Cree <ecree@solarflare.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
22 months agoMerge tag 'tag-chrome-platform-for-v4.20' of git://git.kernel.org/pub/scm/linux/kerne...
Linus Torvalds [Wed, 31 Oct 2018 23:47:55 +0000 (16:47 -0700)]
Merge tag 'tag-chrome-platform-for-v4.20' of git://git./linux/kernel/git/bleung/chrome-platform

Pull chrome-platform updates from Benson Leung:

 - Move mfd/cros_ec_lpc* includes to drivers/platform from mfd

 - Adding a new interrupt path for cros_ec_lpc

* tag 'tag-chrome-platform-for-v4.20' of git://git.kernel.org/pub/scm/linux/kernel/git/bleung/chrome-platform:
  platform/chrome: chromeos_tbmc - Remove unneeded const
  platform/chrome: Add a new interrupt path for cros_ec_lpc
  mfd: cros_ec: Fix and improve kerneldoc comments.
  platform/chrome: Move mfd/cros_ec_lpc* includes to drivers/platform.

22 months agoMAINTAINERS: add maintainer for IMX LPI2C driver
A.s. Dong [Tue, 30 Oct 2018 15:28:14 +0000 (15:28 +0000)]
MAINTAINERS: add maintainer for IMX LPI2C driver

The LPI2C is used in IMX7ULP/MX8 SoCs.

Signed-off-by: Dong Aisheng <aisheng.dong@nxp.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
22 months agodt-bindings: i2c: i2c-imx-lpi2c: add imx8qxp compatible string
A.s. Dong [Tue, 30 Oct 2018 15:28:11 +0000 (15:28 +0000)]
dt-bindings: i2c: i2c-imx-lpi2c: add imx8qxp compatible string

Add imx8qxp compatible string

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Dong Aisheng <aisheng.dong@nxp.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
22 months agoi2c: Clear client->irq in i2c_device_remove
Charles Keepax [Fri, 19 Oct 2018 08:59:58 +0000 (09:59 +0100)]
i2c: Clear client->irq in i2c_device_remove

The IRQ will be mapped in i2c_device_probe only if client->irq is zero and
i2c_device_remove does not clear this. When rebinding an I2C device,
whos IRQ provider has also been rebound this means that an IRQ mapping
will never be created, causing the I2C device to fail to acquire its
IRQ. Fix this issue by clearing client->irq in i2c_device_remove,
forcing i2c_device_probe to lookup the mapping again.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
22 months agoi2c: Remove unnecessary call to irq_find_mapping
Charles Keepax [Fri, 19 Oct 2018 08:59:57 +0000 (09:59 +0100)]
i2c: Remove unnecessary call to irq_find_mapping

irq_create_mapping calls irq_find_mapping internally and will use the
found mapping if one exists, so there is no need to manually call this
from i2c_smbus_host_notify_to_irq.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
22 months agonetfilter: ipset: fix ip_set_list allocation failure
Andrey Ryabinin [Mon, 22 Oct 2018 21:30:40 +0000 (23:30 +0200)]
netfilter: ipset: fix ip_set_list allocation failure

ip_set_create() and ip_set_net_init() attempt to allocate physically
contiguous memory for ip_set_list. If memory is fragmented, the
allocations could easily fail:

        vzctl: page allocation failure: order:7, mode:0xc0d0

        Call Trace:
         dump_stack+0x19/0x1b
         warn_alloc_failed+0x110/0x180
         __alloc_pages_nodemask+0x7bf/0xc60
         alloc_pages_current+0x98/0x110
         kmalloc_order+0x18/0x40
         kmalloc_order_trace+0x26/0xa0
         __kmalloc+0x279/0x290
         ip_set_net_init+0x4b/0x90 [ip_set]
         ops_init+0x3b/0xb0
         setup_net+0xbb/0x170
         copy_net_ns+0xf1/0x1c0
         create_new_namespaces+0xf9/0x180
         copy_namespaces+0x8e/0xd0
         copy_process+0xb61/0x1a00
         do_fork+0x91/0x320

Use kvcalloc() to fallback to 0-order allocations if high order
page isn't available.

Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: ipset: actually allow allowable CIDR 0 in hash:net,port,net
Eric Westbrook [Tue, 28 Aug 2018 21:14:42 +0000 (15:14 -0600)]
netfilter: ipset: actually allow allowable CIDR 0 in hash:net,port,net

Allow /0 as advertised for hash:net,port,net sets.

For "hash:net,port,net", ipset(8) says that "either subnet
is permitted to be a /0 should you wish to match port
between all destinations."

Make that statement true.

Before:

    # ipset create cidrzero hash:net,port,net
    # ipset add cidrzero 0.0.0.0/0,12345,0.0.0.0/0
    ipset v6.34: The value of the CIDR parameter of the IP address is invalid

    # ipset create cidrzero6 hash:net,port,net family inet6
    # ipset add cidrzero6 ::/0,12345,::/0
    ipset v6.34: The value of the CIDR parameter of the IP address is invalid

After:

    # ipset create cidrzero hash:net,port,net
    # ipset add cidrzero 0.0.0.0/0,12345,0.0.0.0/0
    # ipset test cidrzero 192.168.205.129,12345,172.16.205.129
    192.168.205.129,tcp:12345,172.16.205.129 is in set cidrzero.

    # ipset create cidrzero6 hash:net,port,net family inet6
    # ipset add cidrzero6 ::/0,12345,::/0
    # ipset test cidrzero6 fe80::1,12345,ff00::1
    fe80::1,tcp:12345,ff00::1 is in set cidrzero6.

See also:

  https://bugzilla.kernel.org/show_bug.cgi?id=200897
  https://github.com/ewestbrook/linux/commit/df7ff6efb0934ab6acc11f003ff1a7580d6c1d9c

Signed-off-by: Eric Westbrook <linux@westbrook.io>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agonetfilter: ipset: list:set: Decrease refcount synchronously on deletion and replace
Stefano Brivio [Sat, 14 Jul 2018 19:59:43 +0000 (21:59 +0200)]
netfilter: ipset: list:set: Decrease refcount synchronously on deletion and replace

Commit 45040978c899 ("netfilter: ipset: Fix set:list type crash
when flush/dump set in parallel") postponed decreasing set
reference counters to the RCU callback.

An 'ipset del' command can terminate before the RCU grace period
is elapsed, and if sets are listed before then, the reference
counter shown in userspace will be wrong:

 # ipset create h hash:ip; ipset create l list:set; ipset add l
 # ipset del l h; ipset list h
 Name: h
 Type: hash:ip
 Revision: 4
 Header: family inet hashsize 1024 maxelem 65536
 Size in memory: 88
 References: 1
 Number of entries: 0
 Members:
 # sleep 1; ipset list h
 Name: h
 Type: hash:ip
 Revision: 4
 Header: family inet hashsize 1024 maxelem 65536
 Size in memory: 88
 References: 0
 Number of entries: 0
 Members:

Fix this by making the reference count update synchronous again.

As a result, when sets are listed, ip_set_name_byindex() might
now fetch a set whose reference count is already zero. Instead
of relying on the reference count to protect against concurrent
set renaming, grab ip_set_ref_lock as reader and copy the name,
while holding the same lock in ip_set_rename() as writer
instead.

Reported-by: Li Shuang <shuali@redhat.com>
Fixes: 45040978c899 ("netfilter: ipset: Fix set:list type crash when flush/dump set in parallel")
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
22 months agoMerge tag 'riscv-for-linus-4.20-mw2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 31 Oct 2018 23:20:28 +0000 (16:20 -0700)]
Merge tag 'riscv-for-linus-4.20-mw2' of git://git./linux/kernel/git/palmer/riscv-linux

Pull more RISC-V updates from Palmer Dabbelt:
 "This contains the follow-on patches I'd like to target for the 4.20
  merge window. I'm being somewhat conservative here, as while there are
  a few patches on the mailing list that were posted early in the merge
  window I'd like to let those bake for another round -- this was a
  fairly big release as far as RISC-V is concerened, and we need to walk
  before we can run.

  As far as the patches that made it go:

   - A patch to ignore offline CPUs when calculating AT_HWCAP. This
     should fix GDB on the HiFive unleashed, which has an embedded core
     for hart 0 which is exposed to Linux as an offline CPU.

   - A move of EM_RISCV to elf-em.h, which is where it should have been
     to begin with.

   - I've also removed the 64-bit divide routines. I know I'm not really
     playing by my own rules here because I posted the patches this
     morning, but since they shouldn't be in the kernel I think it's
     better to err on the side of going too fast here.

  I don't anticipate any more patch sets for the merge window"

* tag 'riscv-for-linus-4.20-mw2' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux:
  Move EM_RISCV into elf-em.h
  RISC-V: properly determine hardware caps
  Revert "lib: Add umoddi3 and udivmoddi4 of GCC library routines"
  Revert "RISC-V: Select GENERIC_LIB_UMODDI3 on RV32"

22 months agoMerge branch 'for-linus-4.20-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Wed, 31 Oct 2018 22:46:16 +0000 (15:46 -0700)]
Merge branch 'for-linus-4.20-rc1' of git://git./linux/kernel/git/rw/uml

Pull UML updates from Richard Weinberger:

 - removal of old and dead code

 - a bug fix for our tty driver

 - other minor cleanups across the code base

* 'for-linus-4.20-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml:
  um: Make line/tty semantics use true write IRQ
  um: trap: fix spelling mistake, EACCESS -> EACCES
  um: Don't hardcode path as it is architecture dependent
  um: NULL check before kfree is not needed
  um: remove unused AIO code
  um: Give start_idle_thread() a return code
  um: Remove update_debugregs()
  um: Drop own definition of PTRACE_SYSEMU/_SINGLESTEP

22 months agoMerge tag 'for-linus' of git://linux-c6x.org/git/projects/linux-c6x-upstreaming
Linus Torvalds [Wed, 31 Oct 2018 22:39:25 +0000 (15:39 -0700)]
Merge tag 'for-linus' of git://linux-c6x.org/git/projects/linux-c6x-upstreaming

Pull c6x update from Mark Salter.

* tag 'for-linus' of git://linux-c6x.org/git/projects/linux-c6x-upstreaming:
  c6x: switch to NO_BOOTMEM

22 months agolibbpf: Fix compile error in libbpf_attach_type_by_name
Andrey Ignatov [Wed, 31 Oct 2018 19:57:18 +0000 (12:57 -0700)]
libbpf: Fix compile error in libbpf_attach_type_by_name

Arnaldo Carvalho de Melo reported build error in libbpf when clang
version 3.8.1-24 (tags/RELEASE_381/final) is used:

libbpf.c:2201:36: error: comparison of constant -22 with expression of
type 'const enum bpf_attach_type' is always false
[-Werror,-Wtautological-constant-out-of-range-compare]
                if (section_names[i].attach_type == -EINVAL)
                    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^  ~~~~~~~
1 error generated.

Fix the error by keeping "is_attachable" property of a program in a
separate struct field instead of trying to use attach_type itself.

Fixes: 956b620fcf0b ("libbpf: Introduce libbpf_attach_type_by_name")
Reported-by: Arnaldo Carvalho de Melo <acme@kernel.org>
Signed-off-by: Andrey Ignatov <rdna@fb.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
22 months agokselftests/bpf: use ping6 as the default ipv6 ping binary if it exists
Li Zhijian [Tue, 30 Oct 2018 03:15:55 +0000 (11:15 +0800)]
kselftests/bpf: use ping6 as the default ipv6 ping binary if it exists

ping binary on some distros doesn't support "ping -6" anymore.

Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
22 months agoMerge tag 'fuse-update-4.20' of git://git.kernel.org/pub/scm/linux/kernel/git/mszered...
Linus Torvalds [Wed, 31 Oct 2018 21:50:02 +0000 (14:50 -0700)]
Merge tag 'fuse-update-4.20' of git://git./linux/kernel/git/mszeredi/fuse

Pull fuse updates from Miklos Szeredi:
 "As well as the usual bug fixes, this adds the following new features:

   - cached readdir and readlink

   - max I/O size increased from 128k to 1M

   - improved performance and scalability of request queues

   - copy_file_range support

  The only non-fuse bits are trivial cleanups of macros in
  <linux/bitops.h>"

* tag 'fuse-update-4.20' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse: (31 commits)
  fuse: enable caching of symlinks
  fuse: only invalidate atime in direct read
  fuse: don't need GETATTR after every READ
  fuse: allow fine grained attr cache invaldation
  bitops: protect variables in bit_clear_unless() macro
  bitops: protect variables in set_mask_bits() macro
  fuse: realloc page array
  fuse: add max_pages to init_out
  fuse: allocate page array more efficiently
  fuse: reduce size of struct fuse_inode
  fuse: use iversion for readdir cache verification
  fuse: use mtime for readdir cache verification
  fuse: add readdir cache version
  fuse: allow using readdir cache
  fuse: allow caching readdir
  fuse: extract fuse_emit() helper
  fuse: add FOPEN_CACHE_DIR
  fuse: split out readdir.c
  fuse: Use hash table to link processing request
  fuse: kill req->intr_unique
  ...

22 months agoMerge tag 'ceph-for-4.20-rc1' of git://github.com/ceph/ceph-client
Linus Torvalds [Wed, 31 Oct 2018 21:42:31 +0000 (14:42 -0700)]
Merge tag 'ceph-for-4.20-rc1' of git://github.com/ceph/ceph-client

Pull ceph updates from Ilya Dryomov:
 "The highlights are:

   - a series that fixes some old memory allocation issues in libceph
     (myself). We no longer allocate memory in places where allocation
     failures cannot be handled and BUG when the allocation fails.

   - support for copy_file_range() syscall (Luis Henriques). If size and
     alignment conditions are met, it leverages RADOS copy-from
     operation. Otherwise, a local copy is performed.

   - a patch that reduces memory requirement of ceph_sync_read() from
     the size of the entire read to the size of one object (Zheng Yan).

   - fallocate() syscall is now restricted to FALLOC_FL_PUNCH_HOLE (Luis
     Henriques)"

* tag 'ceph-for-4.20-rc1' of git://github.com/ceph/ceph-client: (25 commits)
  ceph: new mount option to disable usage of copy-from op
  ceph: support copy_file_range file operation
  libceph: support the RADOS copy-from operation
  ceph: add non-blocking parameter to ceph_try_get_caps()
  libceph: check reply num_data_items in setup_request_data()
  libceph: preallocate message data items
  libceph, rbd, ceph: move ceph_osdc_alloc_messages() calls
  libceph: introduce alloc_watch_request()
  libceph: assign cookies in linger_submit()
  libceph: enable fallback to ceph_msg_new() in ceph_msgpool_get()
  ceph: num_ops is off by one in ceph_aio_retry_work()
  libceph: no need to call osd_req_opcode_valid() in osd_req_encode_op()
  ceph: set timeout conditionally in __cap_delay_requeue
  libceph: don't consume a ref on pagelist in ceph_msg_data_add_pagelist()
  libceph: introduce ceph_pagelist_alloc()
  libceph: osd_req_op_cls_init() doesn't need to take opcode
  libceph: bump CEPH_MSG_MAX_DATA_LEN
  ceph: only allow punch hole mode in fallocate
  ceph: refactor ceph_sync_read()
  ceph: check if LOOKUPNAME request was aborted when filling trace
  ...

22 months agoMerge branch 'mlxsw-Enable-minimum-shaper-on-MC-TCs'
David S. Miller [Wed, 31 Oct 2018 19:56:59 +0000 (12:56 -0700)]
Merge branch 'mlxsw-Enable-minimum-shaper-on-MC-TCs'

Ido Schimmel says:

====================
mlxsw: Enable minimum shaper on MC TCs

Petr says:

An MC-aware mode was introduced in commit 7b8195306694 ("mlxsw:
spectrum: Configure MC-aware mode on mlxsw ports"). In MC-aware mode,
BUM traffic gets a special treatment by being assigned to a separate set
of traffic classes 8..15. Pairs of TCs 0 and 8, 1 and 9, etc., are then
configured to strictly prioritize the lower-numbered ones. The intention
is to prevent BUM traffic from flooding the switch and push out all UC
traffic, which would otherwise happen, and instead give UC traffic
precedence.

However strictly prioritizing UC traffic has the effect that UC overload
pushes out all BUM traffic, such as legitimate ARP queries. These
packets are kept in queues for a while, but under sustained UC overload,
their lifetime eventually expires and these packets are dropped. That is
detrimental to network performance as well.

In this patchset, MC TCs (8..15) are configured with minimum shaper of
200Mbps (a minimum permitted value) to allow a trickle of necessary
control traffic to get through.

First in patch #1, the QEEC register is extended with fields necessary
to configure the minimum shaper.

In patch #2, minimum shaper is enabled on TCs 8..15.

In patches #3 and #4, first the MC-awareness test is tweaked to support
the minimum shaper, and then a new test is introduced to test that MC
traffic behaves well under UC overload.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoselftests: mlxsw: qos_mc_aware: Add a test for UC awareness
Petr Machata [Wed, 31 Oct 2018 09:56:46 +0000 (09:56 +0000)]
selftests: mlxsw: qos_mc_aware: Add a test for UC awareness

In a previous patch, mlxsw was updated to configure a minimum bandwidth
allowance on MC TCs. Test that this indeed fixes the problem of UC
traffic overload pushing out all MC traffic.

Fixes: b5638d46c90a ("selftests: mlxsw: Add a test for UC behavior under MC flood")
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoselftests: mlxsw: qos_mc_aware: Tweak for min shaper
Petr Machata [Wed, 31 Oct 2018 09:56:45 +0000 (09:56 +0000)]
selftests: mlxsw: qos_mc_aware: Tweak for min shaper

Since the minimum shaper is now being enabled for MC TCs, it's
unreasonable to expect no UC traffic loss. Minimal min shaper value is
200Mbps, which is 20% of the 1Gbps that this test configures on egress.
To cover for glitches, tolerate up to 25% UC degradation under MC
overload.

Fixes: b5638d46c90a ("selftests: mlxsw: Add a test for UC behavior under MC flood")
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agomlxsw: spectrum: Set minimum shaper on MC TCs
Petr Machata [Wed, 31 Oct 2018 09:56:44 +0000 (09:56 +0000)]
mlxsw: spectrum: Set minimum shaper on MC TCs

An MC-aware mode was introduced in commit 7b8195306694 ("mlxsw:
spectrum: Configure MC-aware mode on mlxsw ports"). In MC-aware mode,
BUM traffic gets a special treatment by being assigned to a separate set
of traffic classes 8..15. Pairs of TCs 0 and 8, 1 and 9, etc., are then
configured to strictly prioritize the lower-numbered ones. The intention
is to prevent BUM traffic from flooding the switch and push out all UC
traffic, which would otherwise happen, and instead give UC traffic
precedence.

However strictly prioritizing UC traffic has the effect that UC overload
pushes out all BUM traffic, such as legitimate ARP queries. These
packets are kept in queues for a while, but under sustained UC overload,
their lifetime eventually expires and these packets are dropped. That is
detrimental to network performance as well.

Therefore configure the MC TCs (8..15) with minimum shaper of 200Mbps (a
minimum permitted value) to allow a trickle of necessary control traffic
to get through.

Fixes: 7b8195306694 ("mlxsw: spectrum: Configure MC-aware mode on mlxsw ports")
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agomlxsw: reg: QEEC: Add minimum shaper fields
Petr Machata [Wed, 31 Oct 2018 09:56:42 +0000 (09:56 +0000)]
mlxsw: reg: QEEC: Add minimum shaper fields

Add QEEC.mise (minimum shaper enable) and QEEC.min_shaper_rate to enable
configuration of minimum shaper.

Increase the QEEC length to 0x20 as well: that's the length that the
register has had for a long time now, but with the configurations that
mlxsw typically exercises, the firmware tolerated 0x1C-sized packets.
With mise=true however, FW rejects packets unless they have the full
required length.

Fixes: b9b7cee40579 ("mlxsw: reg: Add QoS ETS Element Configuration register")
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agoMerge branch 'hns3-fixes'
David S. Miller [Wed, 31 Oct 2018 19:42:39 +0000 (12:42 -0700)]
Merge branch 'hns3-fixes'

Huazhong Tan says:

====================
Bugfix for the HNS3 driver

This patch series include bugfix for the HNS3 ethernet
controller driver.

Change log:
V4->V5:
Fixes comments from Joe Perches & Sergei Shtylyov
V3->V4:
Fixes comments from Sergei Shtylyov
V2->V3:
Fixes comments from Sergei Shtylyov
V1->V2:
Fixes the compilation break reported by kbuild test robot
http://patchwork.ozlabs.org/patch/989818/
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for rtnl_lock's range in the hclgevf_reset()
Huazhong Tan [Tue, 30 Oct 2018 13:50:53 +0000 (21:50 +0800)]
net: hns3: bugfix for rtnl_lock's range in the hclgevf_reset()

Since hclgevf_reset_wait() is used to wait for the hardware to complete
the reset, it is not necessary to hold the rtnl_lock during
hclgevf_reset_wait(). So this patch releases the lock for the duration
of hclgevf_reset_wait().

Fixes: 6988eb2a9b77 ("net: hns3: Add support to reset the enet/ring mgmt layer")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for rtnl_lock's range in the hclge_reset()
Huazhong Tan [Tue, 30 Oct 2018 13:50:52 +0000 (21:50 +0800)]
net: hns3: bugfix for rtnl_lock's range in the hclge_reset()

Since hclge_reset_wait() is used to wait for the hardware to complete
the reset, it is not necessary to hold the rtnl_lock during
hclge_reset_wait(). So this patch releases the lock for the duration
of hclge_reset_wait().

Fixes: 6d4fab39533f ("net: hns3: Reset net device with rtnl_lock")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for handling mailbox while the command queue reinitialized
Huazhong Tan [Tue, 30 Oct 2018 13:50:51 +0000 (21:50 +0800)]
net: hns3: bugfix for handling mailbox while the command queue reinitialized

In a multi-core machine, the mailbox service and reset service
will be executed at the same time. The reset service will re-initialize
the command queue, before that, the mailbox handler can only get some
invalid messages.

The HCLGE_STATE_CMD_DISABLE flag means that the command queue is not
available and needs to be reinitialized. Therefore, when the mailbox
handler recognizes this flag, it should not process the command.

Fixes: dde1a86e93ca ("net: hns3: Add mailbox support to PF driver")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: fix incorrect return value/type of some functions
Huazhong Tan [Tue, 30 Oct 2018 13:50:50 +0000 (21:50 +0800)]
net: hns3: fix incorrect return value/type of some functions

There are some functions that, when they fail to send the command,
need to return the corresponding error value to its caller.

Fixes: 46a3df9f9718 ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support")
Fixes: 681ec3999b3d ("net: hns3: fix for vlan table lost problem when resetting")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for hclge_mdio_write and hclge_mdio_read
Huazhong Tan [Tue, 30 Oct 2018 13:50:49 +0000 (21:50 +0800)]
net: hns3: bugfix for hclge_mdio_write and hclge_mdio_read

When there is a PHY, the driver needs to complete some operations through
MDIO during reset reinitialization, so HCLGE_STATE_CMD_DISABLE is more
suitable than HCLGE_STATE_RST_HANDLING to prevent the MDIO operation from
being sent during the hardware reset.

Fixes: b50ae26c57cb ("net: hns3: never send command queue message to IMP when reset)
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for is_valid_csq_clean_head()
Huazhong Tan [Tue, 30 Oct 2018 13:50:48 +0000 (21:50 +0800)]
net: hns3: bugfix for is_valid_csq_clean_head()

The HEAD pointer of the hardware command queue maybe equal to the command
queue's next_to_use in the driver, so that does not belong to the invalid
HEAD pointer, since the hardware may not process the command in time,
causing the HEAD pointer to be too late to update. The variables' name
in this function is unreadable, so give them a more readable one.

Fixes: 3ff504908f95 ("net: hns3: fix a dead loop in hclge_cmd_csq_clean")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: remove unnecessary queue reset in the hns3_uninit_all_ring()
Huazhong Tan [Tue, 30 Oct 2018 13:50:47 +0000 (21:50 +0800)]
net: hns3: remove unnecessary queue reset in the hns3_uninit_all_ring()

It is not necessary to reset the queue in the hns3_uninit_all_ring(),
since the queue is stopped in the down operation, and will be reset
in the up operation. And the judgment of the HCLGE_STATE_RST_HANDLING
flag in the hclge_reset_tqp() is not correct, because we need to reset
tqp during pf reset, otherwise it may cause queue not being reset to
working state problem.

Fixes: 76ad4f0ee747 ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for the initialization of command queue's spin lock
Huazhong Tan [Tue, 30 Oct 2018 13:50:46 +0000 (21:50 +0800)]
net: hns3: bugfix for the initialization of command queue's spin lock

The spin lock of the command queue only need to be initialized once
when the driver initializes the command queue. It is not necessary to
initialize the spin lock when resetting. At the same time, the
modification of the queue member should be performed after acquiring
the lock.

Fixes: 3efb960f056d ("net: hns3: Refactor the initialization of command queue")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for reporting unknown vector0 interrupt repeatly problem
Huazhong Tan [Tue, 30 Oct 2018 13:50:45 +0000 (21:50 +0800)]
net: hns3: bugfix for reporting unknown vector0 interrupt repeatly problem

The current driver supports handling two vector0 interrupts, reset and
mailbox. When the hardware reports an interrupt of another type of
interrupt source, if the driver does not process the interrupt, but
enables the interrupt, the hardware will repeatedly report the unknown
interrupt.

Therefore, the driver enables the vector0 interrupt after clearing the
known type of interrupt source. Other conditions are not enabled.

Fixes: cd8c5c269b1d ("net: hns3: Fix for hclge_reset running repeatly problem")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: bugfix for buffer not free problem during resetting
Huazhong Tan [Tue, 30 Oct 2018 13:50:44 +0000 (21:50 +0800)]
net: hns3: bugfix for buffer not free problem during resetting

When hns3_get_ring_config()/hns3_queue_to_ring()/
hns3_get_vector_ring_chain() failed during resetting, the allocated
memory has not been freed before these three functions return. So
this patch adds error handler in these functions to fix it.

Fixes: 76ad4f0ee747 ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet: hns3: add error handler for hns3_nic_init_vector_data()
Huazhong Tan [Tue, 30 Oct 2018 13:50:43 +0000 (21:50 +0800)]
net: hns3: add error handler for hns3_nic_init_vector_data()

When hns3_nic_init_vector_data() fails to map ring to vector,
it should cancel the netif_napi_add() that has been successfully
done and then exits.

Fixes: 76ad4f0ee747 ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
22 months agonet/mlx5e: fix csum adjustments caused by RXFCS
Eric Dumazet [Tue, 30 Oct 2018 07:57:25 +0000 (00:57 -0700)]
net/mlx5e: fix csum adjustments caused by RXFCS

As shown by Dmitris, we need to use csum_block_add() instead of csum_add()
when adding the FCS contribution to skb csum.

Before 4.18 (more exactly commit 88078d98d1bb "net: pskb_trim_rcsum()
and CHECKSUM_COMPLETE are friends"), the whole skb csum was thrown away,
so RXFCS changes were ignored.

Then before commit d55bef5059dd ("net: fix pskb_trim_rcsum_slow() with
odd trim offset") both mlx5 and pskb_trim_rcsum_slow() bugs were canceling
each other.

Now we fixed pskb_trim_rcsum_slow() we need to fix mlx5.

Note that this patch also rewrites mlx5e_get_fcs() to :

- Use skb_header_pointer() instead of reinventing it.
- Use __get_unaligned_cpu32() to avoid possible non aligned accesses
  as Dmitris pointed out.

Fixes: 902a545904c7 ("net/mlx5e: When RXFCS is set, add FCS data into checksum calculation")
Reported-by: Paweł Staszewski <pstaszewski@itcare.pl>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Eran Ben Elisha <eranbe@mellanox.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
Cc: Dimitris Michailidis <dmichail@google.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Paweł Staszewski <pstaszewski@itcare.pl>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Tested-By: Maria Pasechnik <mariap@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>