Signed-off-by: Eric Dumazet <edumazet@google.com>

mlx4 exclusively uses order-2 allocations in RX path, which are
likely to fail under memory pressure.

We therefore drop frames more than needed.

This patch tries order-3, order-2, order-1 and finally order-0
allocations to keep good performance, yet allow allocations if/when
memory gets fragmented.

By using larger pages, and avoiding unnecessary get_page()/put_page()
on compound pages, this patch improves performance as well, lowering
false sharing on struct page.

Also use GFP_KERNEL allocations in initialization path, as allocating 12
MB (390 order-3 pages) can easily fail with GFP_ATOMIC.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Amir Vadai <amirv@mellanox.com>
Acked-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Ben Hutchings says:

====================
1. Make EEH recovery work when using legacy interrupts, from Alexandre
   Rames.

2. Enable accelerated RFS for VLAN-tagged flows, from Andy Lutomirski.

3. Improve performance for non-TCP (and particularly UDP) traffic, which
   regressed in 3.10 when we switched to always allocating paged RX
   buffers.  Partly by Jon Cooper.

4. Some minor bug fixes to IOMMU detection, timestamping capabilities,
   and IRQ cleanup on the probe failure path.

I've dropped the RX skb cache, which improved some benchmarks but
perhaps needs some reworking to be more generally useful.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Since commit 82dc3c63
"net: introduce NAPI_POLL_WEIGHT" network drivers receive a warning
when they use napi weight higher than NAPI_POLL_WEIGHT. This patch
reduces QETH_NAPI_WEIGHT from 128 to 64 (NAPI_POLL_WEIGHT).
Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When the initial MTU size is changed prior to any activity on the device
(e.g. by attaching a z/VM vNIC already configured in Linux to a guestLAN),
we call dev_kfree_skb_irq(NULL) which results in a kernel panic.
Adding a proper check for NULL pointers to address this issue.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Signed-off-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
Reviewed-by: Ursula Braun <braunu@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

blkt settings (or LAN idle settings) for an OSA Express card
determine when and how often an OSA Express card tells the
operating system about new incoming packets. The semantic of
these settings has changed starting with OSA Express3. Currently
the qeth standard settings apply to OSA Express2 and older
generations of OSA Express cards, while new generations of OSA
Express cards require extra coding of their reasonable default.

To cover future OSA Express generations the qeth default standard
blkt setting is now the desired setting for OSA generations
starting with OSA Express3, while the fixed set of older OSA
Express cards receives its blkt settings explicitly.
Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
Reviewed-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Increase the default MTU for real OSA devices in layer 2 mode
to 1500 Bytes for increased compatibility.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Signed-off-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
Reviewed-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

If someone is interested to dump something they may consider to use
print_hex_dump() or print_hex_dump_bytes() kernel helpers.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch solves several sparse issues as well as an unneeded semicolon
found via coccinelle.
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

commit f88c91dd ("ipv6: statically link
register_inet6addr_notifier()" added following sparse warnings :

net/ipv6/addrconf_core.c:83:5: warning: symbol
'register_inet6addr_notifier' was not declared. Should it be static?
net/ipv6/addrconf_core.c:89:5: warning: symbol
'unregister_inet6addr_notifier' was not declared. Should it be static?
net/ipv6/addrconf_core.c:95:5: warning: symbol
'inet6addr_notifier_call_chain' was not declared. Should it be static?
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

struct tcp_fastopen_context has a field named tfm, which is a pointer
to a crypto_cipher structure.

It currently has a __rcu annotation, which is not needed at all.

tcp_fastopen_ctx is the pointer fetched by rcu_dereference(), but once
we have a pointer to current tcp_fastopen_context, we do not use/need
rcu_dereference() to access tfm.

This fixes a lot of sparse errors like the following :

net/ipv4/tcp_fastopen.c:21:31: warning: incorrect type in argument 1 (different address spaces)
net/ipv4/tcp_fastopen.c:21:31:    expected struct crypto_cipher *tfm
net/ipv4/tcp_fastopen.c:21:31:    got struct crypto_cipher [noderef] <asn:4>*tfm
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jerry Chu <hkchu@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently, there is no good possibility to debug netlink traffic that
is being exchanged between kernel and user space. Therefore, this patch
implements a netlink virtual device, so that netlink messages will be
made visible to PF_PACKET sockets. Once there was an approach with a
similar idea [1], but it got forgotten somehow.

I think it makes most sense to accept the "overhead" of an extra netlink
net device over implementing the same functionality from PF_PACKET
sockets once again into netlink sockets. We have BPF filters that can
already be easily applied which even have netlink extensions, we have
RX_RING zero-copy between kernel- and user space that can be reused,
and much more features. So instead of re-implementing all of this, we
simply pass the skb to a given PF_PACKET socket for further analysis.

Another nice benefit that comes from that is that no code needs to be
changed in user space packet analyzers (maybe adding a dissector, but
not more), thus out of the box, we can already capture pcap files of
netlink traffic to debug/troubleshoot netlink problems.

Also thanks goes to Thomas Graf, Flavio Leitner, Jesper Dangaard Brouer.

 [1] http://marc.info/?l=linux-netdev&m=113813401516110Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Similarly to the networking receive path with ptype_all taps, we add
the possibility to register netdevices that are for ARPHRD_NETLINK to
the netlink subsystem, so that those can be used for netlink analyzers
resp. debuggers. We do not offer a direct callback function as out-of-tree
modules could do crap with it. Instead, a netdevice must be registered
properly and only receives a clone, managed by the netlink layer. Symbols
are exported as GPL-only.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This small patch adds the definition of ARPHRD_NETLINK which can for
example be used by netlink monitoring devices as device type. So that
sockaddr_ll can pick it up and based on that choose the correct packet
dissector.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This restores commits:

c573972c
1a590434
da2e2c21

which initially accidently went into 'net', were
reverted there, and then properly placed into 'net-next'.
But the next net --> net-next merge accidently wiped them
out again.
Reported-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The device::iommu_group field may be set even if no IOMMU is in use.
iommu_present() is still a better indicator, although it doesn't tell
us whether *our* device is affected.
Reported-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

The lifetime of an irq_cpu_rmap is odd: we have to allocate it before
installing IRQ handlers and free it before removing the IRQ handlers.
As a result of this asymmetry, it was omitted from some failure paths.

On another failure path, we could try to remove IRQ handlers we
had not yet installed.

Move the irq_cpu_rmap allocation and freeing alongside IRQ handler
installation and removal, in efx_nic_{init,fini}_interrupts().
Count the number of IRQ handlers successfully installed and only
remove those on the failure path.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

GRO can handle non-TCP packets and pass them up without coalescing,
but it has to do some extra work to parse the packet which we can
bypass using the hardware parse result.  (This condition yields a
false negative for TCP/IPv6 packets received by Falcon, but its
performance is already poor in that case due to lack of checksum
offload.)
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

This will be useful for shortcutting some software packet parsing.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

As far as I know, the hardware doesn't support matching on both IP
fields and vlan tag, but it can at least match on the IP fields.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

The kernel can generate software receive timestamps and we should
report those for all ports regardless of hardware capabilities.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

This allows the SKB to hold the headers without reallocation more often.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

PCI legacy interrupts are level-triggered, and we cannot mask them up
on an isolated device.  Instead, disable the IRQ at the controller
until we have recovered.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>

This fixes an issue caused by submit 78c3bcc5
`bnx2x: Improve PF behaviour toward VF', which made the bnx2x driver fail
compilation when PCI_IOV is not set.
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Callers of skb_seq_read() are currently forced to call skb_abort_seq_read()
even when consuming all the data because the last call to skb_seq_read (the
one that returns 0 to indicate the end) fails to unmap the last fragment page.

With this patch callers will be allowed to traverse the SKB data by calling
skb_prepare_seq_read() once and repeatedly calling skb_seq_read() as originally
intended (and documented in the original commit 677e90ed), that is, only call
skb_abort_seq_read() if the sequential read is actually aborted.
Signed-off-by: Wedson Almeida Filho <wedsonaf@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

John W. Linville says:

====================
I would guess that this is the last big wireless pull request before
the 3.11 merge window...

Regarding the mac80211 bits, Johannes says:

"I have a number of mesh fixes and improvements from Colleen, Jacob,
Ashok and Thomas, powersave fixes in mac80211 from Alex, improved
management-TX from Antonio, and a few various things, including locking
fixes, from others and myself. Overall though, nothing really stands
out."

As for the iwlwifi bits, Johannes says:

"Emmanuel contributed two AP mode fixes, removed an unused field, fixed a
comment and added a warning for something that shouldn't happen in
practice, and I removed the declaration of a function that doesn't even
exist and cleaned up a small include."

"This time I have a number of cleanups, a small fix from Emmanuel and two
performance improvements that combined reduce our driver's CPU
utilisation as much as 75% in high TX-throughput scenarios."

"These two patches fix two issues with using rfkill randomly during
traffic, which would then cause our driver to stop working and not be
able to recover at all."

Regarding the ath6kl bits, Kalle says:

"Here are few simple patches for ath6kl. We have a suspend crash fix for
USB from Shafi, use of mac_pton(), a compiler warning fix and a fix for
module initialisation error path."

Kalle also sends the biggest single item of note, the new ath10k
driver for Qualcomm Atheros 802.11ac CQA98xx devices.

Included is an NFC pull, of which Samuel says:

"These are the pending NFC patches for the 3.11 merge window.

It contains the pending fixes that were on nfc-fixes (nfc-fixes-3.10-2),
along with a few more for the pn544 and pn533 drivers, the LLCP
disconnection path and an LLCP memory leak.

Highlights for this one are:

- An initial secure element API. NFC chipsets can carry an embedded
  secure element or get access to the SIM one. In both cases they
  control the secure elements and this API provides a way to discover,
  enable and disable the available SEs. It also exports that to
  userspace in order for SE focused middleware to actually do something
  with them (e.g. payments).

- NCI over SPI support. SPI is the most complex NCI specified transport
  layer and we now have support for it in the kernel. The next step will
  be to implement drivers for NCI chipsets using this transport like
  e.g. bcm2079x.

- NFC p2p hardware simulation driver. We now have an nfcsim driver that
  is mostly a loopback device between 2 NFC interfaces. It also
  implements the rest of the NFC core API like polling and target
  detection. This driver, with neard running on top of it, allows us to
  completely test the LLCP, SNEP and Handover implementation without
  physical hardware.

- A Firmware update netlink API. Most (All ?) HCI chipsets have a
  special firmware update mode where applications can push a new
  firmware that will be flashed. We now have a netlink API for providing
  that mode to e.g. nfctool."

On top of all that, there are a variety of updates to brcmfmac,
iwlegacy, rtlwifi, wil6210, and the TI wl12xx drivers.  As usual,
the bcma and ssb busses get a little love as well, as do a handful
of others here and there.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

There is a typo here, "i" vs "j", so we would crash on module_exit().
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Tunnel constants can be used in generic code but in these cases
the inline functions in ip_tunnels.h cause compilation problems
if CONFIG_INET is not set.

CC: Pravin Shelar <pshelar@nicira.com>
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Jesse Gross <jesse@nicira.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

This bug was introduced by commit aa310701
(openvswitch: Add gre tunnel support.)
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Cc: Stephen Hemminger <stephen@networkplumber.org>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netif_alloc_netdev_queues() uses kcalloc() to allocate memory
for the "struct netdev_queue *_tx" array.

For large number of tx queues, kcalloc() might fail, so this
patch does a fallback to vzalloc().

As vmalloc() adds overhead on a critical network path, add __GFP_REPEAT
to kzalloc() flags to do this fallback only when really needed.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Yuval Mintz says:

====================
This patch series mostly revolves around improving SR-IOV implementation
(Better PF-VF relation, sanity checks and timings), as well as including
a patch correcting the (outward) advertisement of 20G capabilities.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Don't claim 20G is supported if the speed is unsupported by the phys
(reflected by various ethtools and ndos).
Signed-off-by: Yaniv Rosner <yanivr@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Wait 100ms for FLR to complete in parallel over all VFs instead of serializing
the waits (which can amount to several seconds with 64 VFs).
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

If iproute2 VF callbacks are invoked before PF is loaded,
abort gracefully.
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

If PF is unloaded with loaded VFs, signal towards VFs so they can detect
this gracefully.
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
----
 drivers/net/ethernet/broadcom/bnx2x/bnx2x.h       |  2 ++
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c   |  3 +++
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c  | 23 +++++++++++++++++++---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c | 24 ++++++++++++++++++++---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h |  2 ++
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c  | 12 +++++++++++-
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h  |  5 ++++-
 7 files changed, 63 insertions(+), 8 deletions(-)
Signed-off-by: David S. Miller <davem@davemloft.net>

If we mod with VSOCK_HASH_SIZE -1, we get 0, 1, .... 249.  Actually, we
have vsock_bind_table[0 ... 250] and vsock_connected_table[0 .. 250].
In this case the last entry will never be used.

We should mod with VSOCK_HASH_SIZE instead.
Signed-off-by: Asias He <asias@redhat.com>
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Asias He <asias@redhat.com>
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vmci_transport_recv_dgram_cb always return VMCI_SUCESS even if we fail
to allocate skb, return VMCI_ERROR_NO_MEM instead.
Signed-off-by: Asias He <asias@redhat.com>
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This peace of code is called three times, let's have a helper for it.
Signed-off-by: Asias He <asias@redhat.com>
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>