This patch should help to improve the performance of the mlx4 and mlx5 on a
number of architectures.  For example, on x86 the dma_wmb/rmb equates out
to a barrer() call as the architecture is already strong ordered, and on
PowerPC the call works out to a lwsync which is significantly less expensive
than the sync call that was being used for wmb.

I placed the new barriers between any spots that seemed to be trying to
order memory/memory reads or writes, if there are any spots that involved
MMIO I left the existing wmb in place as the new barriers cannot order
transactions between coherent and non-coherent memories.

v2: Reduced the replacments to just the spots where I could clearly
    identify the usage pattern.

Cc: Amir Vadai <amirv@mellanox.com>
Cc: Ido Shamay <idos@mellanox.com>
Cc: Eli Cohen <eli@mellanox.com>
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Update the Chelsio Ethernet drivers to use the dma_rmb/wmb calls instead of
the full barriers in order to improve performance.

Cc: Santosh Raspatur <santosh@chelsio.com>
Cc: Hariprasad S <hariprasad@chelsio.com>
Cc: Casey Leedom <leedom@chelsio.com>
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We need to clean up vxlan despite vxlan_igmp_leave() fails.

This fixes the following kernel warning:

 WARNING: CPU: 0 PID: 6 at lib/debugobjects.c:263 debug_print_object+0x7c/0x8d()
 ODEBUG: free active (active state 0) object type: timer_list hint: vxlan_cleanup+0x0/0xd0
 CPU: 0 PID: 6 Comm: kworker/u8:0 Not tainted 4.0.0-rc7+ #953
 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
 Workqueue: netns cleanup_net
  0000000000000009 ffff88011955f948 ffffffff81a25f5a 00000000253f253e
  ffff88011955f998 ffff88011955f988 ffffffff8107608e 0000000000000000
  ffffffff814deba2 ffff8800d4e94000 ffffffff82254c30 ffffffff81fbe455
 Call Trace:
  [<ffffffff81a25f5a>] dump_stack+0x4c/0x65
  [<ffffffff8107608e>] warn_slowpath_common+0x9c/0xb6
  [<ffffffff814deba2>] ? debug_print_object+0x7c/0x8d
  [<ffffffff81076116>] warn_slowpath_fmt+0x46/0x48
  [<ffffffff814deba2>] debug_print_object+0x7c/0x8d
  [<ffffffff81666bf1>] ? vxlan_fdb_destroy+0x5b/0x5b
  [<ffffffff814dee02>] __debug_check_no_obj_freed+0xc3/0x15f
  [<ffffffff814df728>] debug_check_no_obj_freed+0x12/0x16
  [<ffffffff8117ae4e>] slab_free_hook+0x64/0x6c
  [<ffffffff8114deaa>] ? kvfree+0x31/0x33
  [<ffffffff8117dc66>] kfree+0x101/0x1ac
  [<ffffffff8114deaa>] kvfree+0x31/0x33
  [<ffffffff817d4137>] netdev_freemem+0x18/0x1a
  [<ffffffff817e8b52>] netdev_release+0x2e/0x32
  [<ffffffff815b4163>] device_release+0x5a/0x92
  [<ffffffff814bd4dd>] kobject_cleanup+0x49/0x5e
  [<ffffffff814bd3ff>] kobject_put+0x45/0x49
  [<ffffffff817d3fc1>] netdev_run_todo+0x26f/0x283
  [<ffffffff817d4873>] ? rollback_registered_many+0x20f/0x23b
  [<ffffffff817e0c80>] rtnl_unlock+0xe/0x10
  [<ffffffff817d4af0>] default_device_exit_batch+0x12a/0x139
  [<ffffffff810aadfa>] ? wait_woken+0x8f/0x8f
  [<ffffffff817c8e14>] ops_exit_list+0x2b/0x57
  [<ffffffff817c9b21>] cleanup_net+0x154/0x1e7
  [<ffffffff8108b05d>] process_one_work+0x255/0x4ad
  [<ffffffff8108af69>] ? process_one_work+0x161/0x4ad
  [<ffffffff8108b4b1>] worker_thread+0x1cd/0x2ab
  [<ffffffff8108b2e4>] ? process_scheduled_works+0x2f/0x2f
  [<ffffffff81090686>] kthread+0xd4/0xdc
  [<ffffffff8109eca3>] ? local_clock+0x19/0x22
  [<ffffffff810905b2>] ? __kthread_parkme+0x83/0x83
  [<ffffffff81a31c48>] ret_from_fork+0x58/0x90
  [<ffffffff810905b2>] ? __kthread_parkme+0x83/0x83

For the long-term, we should handle NETDEV_{UP,DOWN} event
from the lower device of a tunnel device.

Fixes: 56ef9c90 ("vxlan: Move socket initialization to within rtnl scope")
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

FastOpen requests are not like other regular request sockets.

They do not yet use rsk_timer : tcp_fastopen_queue_check()
simply manually removes one expired request from fastopenq->rskq_rst
list.

Therefore, tcp_check_req() must not call mod_timer_pending(),
otherwise we crash because rsk_timer was not initialized.

Fixes: fa76ce73 ("inet: get rid of central tcp/dccp listener timer")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Samuel Ortiz says:

====================
NFC: 4.1 pull request

This is the NFC pull request for 4.1.

This is a shorter one than usual, as the Intel Field Peak NFC
driver could not make it in time.

We have:

- A new driver for NXP NCI based chipsets, like e.g. the NPC100 or
  the PN7150. It currently only supports an i2c physical layer, but
  could easily be extended to work on top of e.g. SPI.
  This driver also includes support for user space triggered firmware
  updates.

- A few minor st21nfc[ab] fixes, cleanups, and comments improvements.

- A pn533 error return fix.

- A few NFC related logs formatting cleanups.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commits:

d92916f7 ("sfc: Own header for nic-specific sriov functions,")
25672dba ("sfc: Enable VF's via a write to the sysfs file
 sriov_numvfs")

As they break the build with SRIOV disabled and there is no
easy way to fix it the way things are arranged.
Reported-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

More recent GCC warns about two kinds of switch statement uses:

1) Switching on an enumeration, but not having an explicit case
   statement for all members of the enumeration.  To show the
   compiler this is intentional, we simply add a default case
   with nothing more than a break statement.

2) Switching on a boolean value.  I think this warning is dumb
   but nevertheless you get it wholesale with -Wswitch.

This patch cures all such warnings in netfilter.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
Acked-by: NPablo Neira Ayuso <pablo@netfilter.org>

Nicolas Dichtel says:

====================
selinux: add some missing nlmsg commands

It's not a critical issue, thus the patches are based on net-next.

Patches are splitted because the 'Fixes' tag is not the same for all
commands.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

These commands are missing.

Fixes: 28d8909b ("[XFRM]: Export SAD info.")
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This command is missing.

Fixes: ecfd6b18 ("[XFRM]: Export SPD info")
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This new command is missing.

Fixes: 880a6fab ("xfrm: configure policy hash table thresholds by netlink")
Reported-by: NChristophe Gouault <christophe.gouault@6wind.com>
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This new command is missing.

Fixes: 9a963454 ("netns: notify netns id events")
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

These new commands are missing.

Fixes: 0c7aecd4 ("netns: add rtnl cmd to add and get peer netns ids")
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The DWMAC block on certain SoCs (such as IMG Pistachio) have a second
clock which must be enabled in order to access the peripheral's
register interface, so add support for requesting and enabling an
optional "pclk".
Signed-off-by: NAndrew Bresticker <abrestic@chromium.org>
Cc: James Hartley <james.hartley@imgtec.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 79b16aad
("udp_tunnel: Pass UDP socket down through udp_tunnel{, 6}_xmit_skb()")
introduce 'sk' but we already have one inner 'sk'.

Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Vitaly Kuznetsov says:

====================
hv_netvsc: linearize SKBs bigger than MAX_PAGE_BUFFER_COUNT-2 pages

This patch series fixes the same issue which was fixed in Xen with commit
97a6d1bb ("xen-netfront: Fix handling packets on
compound pages with skb_linearize").

It is relatively easy to create a packet which is small in size but occupies
more than 30 (MAX_PAGE_BUFFER_COUNT-2) pages. Here is a kernel-mode reproducer
which tries sending a packet with only 34 bytes of payload (but on 34 pages)
and fails:

static int __init sendfb_init(void)
{
	struct socket *sock;
	int i, ret;
	struct sockaddr_in in4_addr = { 0 };
	struct page *pages[17];
	unsigned long flags;

	ret = sock_create_kern(AF_INET, SOCK_STREAM, IPPROTO_TCP, &sock);
	if (ret) {
		pr_err("failed to create socket: %d!\n", ret);
		return ret;
	}

	in4_addr.sin_family = AF_INET;
	/* www.google.com, 74.125.133.99 */
	in4_addr.sin_addr.s_addr = cpu_to_be32(0x4a7d8563);
	in4_addr.sin_port = cpu_to_be16(80);

	ret = sock->ops->connect(sock, (struct sockaddr *)&in4_addr, sizeof(in4_addr), 0);
	if (ret) {
		pr_err("failed to connect: %d!\n", ret);
		return ret;
	}

	/* We can send up to 17 frags */
	flags = MSG_MORE;
	for (i = 0; i < 17; i++) {
		if (i == 16)
			flags = MSG_EOR;
		pages[i] = alloc_pages(GFP_KERNEL | __GFP_COMP, 1);
		if (!pages[i]) {
			pr_err("out of memory!");
			goto free_pages;
		}
		sock->ops->sendpage(sock, pages[i], PAGE_SIZE -1, 2, flags);
	}

free_pages:
	for (; i > 0; i--)
		__free_pages(pages[i - 1], 1);

	printk("sendfb_init: test done\n");
        return -1;
}

module_init(sendfb_init);

MODULE_LICENSE("GPL");

A try to load such module results in multiple
'kernel: hv_netvsc vmbus_15 eth0: Packet too big: 100' messages as all retries
fail as well. It should also be possible to trigger the issue from userspace, I
expect e.g. NFS under heavy load to get stuck sometimes.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In netvsc_start_xmit() we can handle packets which are scattered around not
more than MAX_PAGE_BUFFER_COUNT-2 pages. It is, however, easy to create a
packet which is not big in size but occupies more pages (e.g. if it uses frags
on compound pages boundaries). When we drop such packet it cases sender to try
resending it but in most cases it will try resending the same packet which will
also get dropped, this will cause the particular connection to stick. To solve
the issue we can try linearizing skb.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

... which validly uses dev_kfree_skb_any() instead of dev_kfree_skb().

Setting ret to -EFAULT and -ENOMEM have no real meaning here (we need to set
it to anything but -EAGAIN) as we drop the packet and return NETDEV_TX_OK
anyway.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Shradha Shah says:

====================
sfc: Nic specific sriov functions, netdev_ops and sriov_configure

First two patches among the series of patches to support SRIOV on EF10.

First patch declares nic specific sriov functions in nic specific headers,
creates only one instance of the netdev_ops, removes sriov functionality
from Falcon code.

Second patch adds support for sriov_configure.

The Virtual Functions can be enabled but they do not bind to the SFC
driver just yet.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds support for the use of sriov_configure on EF10
to enable Virtual Functions while the driver is loaded.
Signed-off-by: NShradha Shah <sshah@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

sfc: Own header for nic-specific sriov functions, single instance of netdev_ops and sriov removed from Falcon code

By putting all the efx_{siena,ef10}_sriov_* declarations in
{siena,ef10}_sriov.h, ensure they cannot be called from nic-generic code.
Also fixes up an instance of this, where mcdi.c was calling
efx_siena_sriov_flr.

The single instance of netdev_ops should call general high level
functions that can then call something adapter specific in efx_nic_type.
We should only do adapter specialisation via efx_nic_type.

Removal of sriov functionality from the Falcon code means that tests
are needed for the presence of some callbacks.
Signed-off-by: NShradha Shah <sshah@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Alexander Duyck says:

====================
Replace wmb()/rmb() with dma_wmb()/dma_rmb() where appropriate

This is a start of a side project cleaning up the drivers that can make use
of the dma_wmb and dma_rmb calls.  The general idea is to start removing
the unnecessary wmb/rmb calls from a number of drivers and to make use of
the lighter weight dma_wmb/dma_rmb calls as this should allow for an
overall improvement in performance as each barrier can cost a significant
number of cycles and on architectures such as x86 this is unnecessary.

These changes are what I would consider low hanging fruit.  The likelihood
of the changes introducing an error should be low since the use of the
barriers in these cases are fairly obvious.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change replaces calls to rmb with dma_rmb in the case where we want to
order all follow-on descriptor reads after the check for the descriptor
status bit.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Acked-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change updates several spots where a wmb was being used to instead use
a dma_wmb to flush out writes before updating the control portion of the
descriptor.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch goes through and replaces wmb/rmb with dma_wmb/dma_rmb in cases
where the barrier is being used to order writes or reads to just memory and
doesn't involve any programmed I/O.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

bond_3ad_bind_slave() calls ad_initialize_port() and then immediately
assigns correct values making some of that initialization unnecessary.
Signed-off-by: NMahesh Bandewar <maheshb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch breaks the rich assignments into it's own statements
and removes some duplicate code where admin-key, & oper-key are
updated.
Signed-off-by: NMahesh Bandewar <maheshb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes: 79b16aad ("udp_tunnel: Pass UDP socket down through udp_tunnel{, 6}_xmit_skb().")
Reported-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The socket parameter might legally be NULL, thus sock_net is sometimes
causing a NULL pointer dereference. Using net_device pointer in dst_entry
is more reliable.

Fixes: b6a7719a ("ipv4: hash net ptr into fragmentation bucket selection")
Reported-by: NRick Jones <rick.jones2@hp.com>
Cc: Rick Jones <rick.jones2@hp.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NSheng Yong <shengyong1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In the two places changed, we now use netvsc_xmit_completion() which properly
frees hv_netvsc_packet in or not in skb headroom.
Signed-off-by: NHaiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: NK. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The sum of RNDIS msg and PPI struct sizes is used in multiple places, so we define
a macro for them.
Signed-off-by: NHaiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: NK. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NHaiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: NK. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fast Open has been using an experimental option with a magic number
(RFC6994). This patch makes the client by default use the RFC7413
option (34) to get and send Fast Open cookies.  This patch makes
the client solicit cookies from a given server first with the
RFC7413 option. If that fails to elicit a cookie, then it tries
the RFC6994 experimental option. If that also fails, it uses the
RFC7413 option on all subsequent connect attempts.  If the server
returns a Fast Open cookie then the client caches the form of the
option that successfully elicited a cookie, and uses that form on
later connects when it presents that cookie.

The idea is to gradually obsolete the use of experimental options as
the servers and clients upgrade, while keeping the interoperability
meanwhile.
Signed-off-by: NDaniel Lee <Longinus00@gmail.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fast Open has been using the experimental option with a magic number
(RFC6994) to request and grant Fast Open cookies. This patch enables
the server to support the official IANA option 34 in RFC7413 in
addition.

The change has passed all existing Fast Open tests with both
old and new options at Google.
Signed-off-by: NDaniel Lee <Longinus00@gmail.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Also move 'group' description to match the order of the net_device structure.

Fixes: 7a66bbc9 ("net: remove iflink field from struct net_device")
Reported-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Nicolas Dichtel says:

====================
netns: enhance netlink interface for nsid

The first patch is a small cleanup. The second patch implements notifications
for netns id events. And the last one allows to dump existing netns id from
userland.

iproute2 patches are available, I can send them on demand.

v2: drop the first patch (the fix is now in net-next)
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Which this patch, it's possible to dump the list of ids allocated for peer
netns.
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With this patch, netns ids that are created and deleted are advertised into the
group RTNLGRP_NSID.

Because callers of rtnl_net_notifyid() already know the id of the peer, there is
no need to call __peernet2id() in rtnl_net_fill().
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>