With increasing receive window sizes, but speed of light not improved
that much, out of order queue can contain a huge number of skbs, waiting
to be moved to receive_queue when missing packets can fill the holes.

Some devices happen to use fat skbs (truesize of 4096 + sizeof(struct
sk_buff)) to store regular (MTU <= 1500) frames. This makes highly
probable sk_rmem_alloc hits sk_rcvbuf limit, which can be 4Mbytes in
many cases.

When limit is hit, tcp stack calls tcp_collapse_ofo_queue(), a true
latency killer and cpu cache blower.

Doing the coalescing attempt each time we add a frame in ofo queue
permits to keep memory use tight and in many cases avoid the
tcp_collapse() thing later.

Tested on various wireless setups (b43, ath9k, ...) known to use big skb
truesize, this patch removed the "packets collapsed in receive queue due
to low socket buffer" I had before.

This also reduced average memory used by tcp sockets.

With help from Neal Cardwell.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: H.K. Jerry Chu <hkchu@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

I found recently that the arp_process function which handles all of our received
arp frames, is using IPV4_DEVCONF_ALL macro to check the state of the arp_process
flag.  This seems wrong, as it implies that either none or all of the network
interfaces accept gratuitous arps.  This patch corrects that, allowing
per-interface arp_accept configuration to deviate from the all setting.  Note
this also brings us into line with the way the arp_filter setting is handled
during arp_process execution.

Tested this myself on my home network, and confirmed it works as expected.
Signed-off-by: NNeil Horman <nhorman@tuxdriver.com>
CC: "David S. Miller" <davem@davemloft.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change adds a memory barrier to the byte queue limit code to address a
possible race as has been seen in the past with the
netif_stop_queue/netif_wake_queue logic.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We are seeing dev_watchdog hangs on several drivers.  I suspect this is due
to the __QUEUE_STATE_STACK_XOFF bit being set prior to a reset for link
change, and then not being cleared by netdev_tx_reset_queue.  This change
corrects that.

In addition we were seeing dev_watchdog hangs on igb after running the
ethtool tests.  We found this to be due to the fact that the ethtool test
runs the same logic as ndo_start_xmit, but we were never clearing the XOFF
flag since the loopback test in ethtool does not do byte queue accounting.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Receive background scan period as part of connect
command and pass the same to the driver.
Signed-off-by: NBala Shanmugam <bkamatch@qca.qualcomm.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The following 4 functions:
  move_addr_to_kernel
  move_addr_to_user
  verify_iovec
  verify_compat_iovec
are always effectively called with a sockaddr_storage.

Make this explicit by changing their signature.

This removes a large number of casts from sockaddr_storage to sockaddr.
Signed-off-by: NMaciej Żenczykowski <maze@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Specifically use it in napi_disable_pending(), napi_schedule_prep(),
napi_reschedule(), netif_tx_queue_stopped(), netif_queue_stopped(),
netif_xmit_stopped(), netif_xmit_frozen_or_stopped(), netif_running(),
__netif_subqueue_stopped(), netif_subqueue_stopped(),
netif_is_multiquue(), netif_carrier_ok(), netif_dormant(),
netif_oper_up(), netif_device_present(), __netif_tx_trylock(),
net_gso_ok(), skb_gso_ok(), netif_needs_gso(), and
netif_is_bond_slave().
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In particular do this for skb_is_nonlinear(), skb_is_gso(), and
skb_is_gso_v6().
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch allows you to attach the timeout policy via the
CT target, it adds a new revision of the target to ensure
backward compatibility. Moreover, it also contains the glue
code to stick the timeout object defined via nfnetlink_cttimeout
to the given flow.

Example usage (it requires installing the nfct tool and
libnetfilter_cttimeout):

1) create the timeout policy:

 nfct timeout add tcp-policy0 inet tcp \
	established 1000 close 10 time_wait 10 last_ack 10

2) attach the timeout policy to the packet:

 iptables -I PREROUTING -t raw -p tcp -j CT --timeout tcp-policy0

You have to install the following user-space software:

a) libnetfilter_cttimeout:
   git://git.netfilter.org/libnetfilter_cttimeout

b) nfct:
   git://git.netfilter.org/nfct

You also have to get iptables with -j CT --timeout support.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

This patch adds the infrastructure to add fine timeout tuning
over nfnetlink. Now you can use the NFNL_SUBSYS_CTNETLINK_TIMEOUT
subsystem to create/delete/dump timeout objects that contain some
specific timeout policy for one flow.

The follow up patches will allow you attach timeout policy object
to conntrack via the CT target and the conntrack extension
infrastructure.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

This patch moves the retransmission and unacknowledged timeouts
to the tcp_timeouts array. This change is required by follow-up
patches.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

These two headers are not required anymore, they have been
replaced by xt_SAME.h and xt_realm.h.

Florian Westphal pointed out this.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Florian Westphal <fw@strlen.de>
Signed-off-by: WANG Cong <xiyou.wangcong@gmail.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

ipt_LOG and ip6_LOG have a lot of common code, merge them
to reduce duplicate code.
Signed-off-by: NRichard Weinberger <richard@nod.at>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

This patch allows you to set expectfn which is specifically used
by the NAT side of most of the existing conntrack helpers.

I have added a symbol map that uses a string as key to look up for
the function that is attached to the expectation object. This is
the best solution I came out with to solve this issue.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

This patch adds the missing bits to create expectations that
are created in NAT setups.

This patch allows you to set the expectation class.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Timed out entries were still matched till the garbage collector
purged them out. The fix is verified in the testsuite.
Signed-off-by: NJozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

The "nomatch" keyword and option is added to the hash:*net* types,
by which one can add exception entries to sets. Example:

        ipset create test hash:net
        ipset add test 192.168.0/24
        ipset add test 192.168.0/30 nomatch

In this case the IP addresses from 192.168.0/24 except 192.168.0/30
match the elements of the set.
Signed-off-by: NJozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

If the set is full, the SET target cannot add more elements.
Log warning so that the admin got notified about it.
Signed-off-by: NJozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

iptables's libxt_SET.c depends on these.
Signed-off-by: NJan Engelhardt <jengelh@medozas.de>
Signed-off-by: NJozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

ipset is actually using NFPROTO values rather than AF (xt_set passes
that along).
Signed-off-by: NJan Engelhardt <jengelh@medozas.de>
Signed-off-by: NJozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The SET_PORT functions are implemented in port.c, which is part
of mlx4_core, these functions are exported. The functions are in use by
the mlx4_en module (were originally part of mlx4_en).
Their declaration remained in mlx4_en module, moving the declaration to the right location.
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For user space to know if a device is up or down.
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Wireless Broadcom chips can have either their SPROM data stored
on either external SPROM or on-chip OTP memory. Both are accessed
through the same register space. This patch adds support for the
on-chip OTP memory.

Tested with:
BCM43224 OTP and SPROM
BCM4331 SPROM
BCM4313 OTP

This patch is in response to linux-wireless thread [1].

[1] http://article.gmane.org/gmane.linux.kernel.wireless.general/85426Tested-by: NSaul St. John <saul.stjohn@gmail.com>
Tested-by: NRafal Milecki <zajec5@gmail.com>
Tested-by: NHauke Mehrtens <hauke@hauke-m.de>
Cc: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: NArend van Spriel <arend@broadcom.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Add the signal strength (in dBm only for now) to
frames that are received via nl80211's various
frame APIs.
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Acked-by: NKalle Valo <kvalo@qca.qualcomm.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

When moving tasks from old memcg (with move_charge_at_immigrate on new
memcg), followed by removal of old memcg, hit General Protection Fault in
mem_cgroup_lru_del_list() (called from release_pages called from
free_pages_and_swap_cache from tlb_flush_mmu from tlb_finish_mmu from
exit_mmap from mmput from exit_mm from do_exit).

Somewhat reproducible, takes a few hours: the old struct mem_cgroup has
been freed and poisoned by SLAB_DEBUG, but mem_cgroup_lru_del_list() is
still trying to update its stats, and take page off lru before freeing.

A task, or a charge, or a page on lru: each secures a memcg against
removal.  In this case, the last task has been moved out of the old memcg,
and it is exiting: anonymous pages are uncharged one by one from the
memcg, as they are zapped from its pagetables, so the charge gets down to
0; but the pages themselves are queued in an mmu_gather for freeing.

Most of those pages will be on lru (and force_empty is careful to
lru_add_drain_all, to add pages from pagevec to lru first), but not
necessarily all: perhaps some have been isolated for page reclaim, perhaps
some isolated for other reasons.  So, force_empty may find no task, no
charge and no page on lru, and let the removal proceed.

There would still be no problem if these pages were immediately freed; but
typically (and the put_page_testzero protocol demands it) they have to be
added back to lru before they are found freeable, then removed from lru
and freed.  We don't see the issue when adding, because the
mem_cgroup_iter() loops keep their own reference to the memcg being
scanned; but when it comes to mem_cgroup_lru_del_list().

I believe this was not an issue in v3.2: there, PageCgroupAcctLRU and
PageCgroupUsed flags were used (like a trick with mirrors) to deflect view
of pc->mem_cgroup to the stable root_mem_cgroup when neither set.
38c5d72f ("memcg: simplify LRU handling by new rule") mercifully
removed those convolutions, but left this General Protection Fault.

But it's surprisingly easy to restore the old behaviour: just check
PageCgroupUsed in mem_cgroup_lru_add_list() (which decides on which lruvec
to add), and reset pc to root_mem_cgroup if page is uncharged.  A risky
change?  just going back to how it worked before; testing, and an audit of
uses of pc->mem_cgroup, show no problem.

And there's a nice bonus: with mem_cgroup_lru_add_list() itself making
sure that an uncharged page goes to root lru, mem_cgroup_reset_owner() no
longer has any purpose, and we can safely revert 4e5f01c2 ("memcg:
clear pc->mem_cgroup if necessary").

Calling update_page_reclaim_stat() after add_page_to_lru_list() in swap.c
is not strictly necessary: the lru_lock there, with RCU before memcg
structures are freed, makes mem_cgroup_get_reclaim_stat_from_page safe
without that; but it seems cleaner to rely on one dependency less.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Previously it was (ab)used by utrace.  Then it was wrongly used by the
scheduler code.

Currently it is not used, kill it before it finds the new erroneous user.
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now that CLONE_VFORK is killable, coredump_wait() no longer needs
complete_vfork_done().  zap_threads() should find and kill all tasks with
the same ->mm, this includes our parent if ->vfork_done is set.

mm_release() becomes the only caller, unexport complete_vfork_done().
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make vfork() killable.

Change do_fork(CLONE_VFORK) to do wait_for_completion_killable().  If it
fails we do not return to the user-mode and never touch the memory shared
with our child.

However, in this case we should clear child->vfork_done before return, we
use task_lock() in do_fork()->wait_for_vfork_done() and
complete_vfork_done() to serialize with each other.

Note: now that we use task_lock() we don't really need completion, we
could turn task->vfork_done into "task_struct *wake_up_me" but this needs
some complications.

NOTE: this and the next patches do not affect in-kernel users of
CLONE_VFORK, kernel threads run with all signals ignored including
SIGKILL/SIGSTOP.

However this is obviously the user-visible change.  Not only a fatal
signal can kill the vforking parent, a sub-thread can do execve or
exit_group() and kill the thread sleeping in vfork().
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

No functional changes.

Move the clear-and-complete-vfork_done code into the new trivial helper,
complete_vfork_done().
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Acked-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Since commit 04c6862c ("kmsg_dump: add kmsg_dump() calls to the
reboot, halt, poweroff and emergency_restart paths"), kmsg_dump() gets
run on normal paths including poweroff and reboot.

This is less than ideal given pstore implementations that can only
represent single backtraces, since a reboot may overwrite a stored oops
before it's been picked up by userspace.  In addition, some pstore
backends may have low performance and provide a significant delay in
reboot as a result.

This patch adds a printk.always_kmsg_dump kernel parameter (which can also
be changed from userspace).  Without it, the code will only be run on
failure paths rather than on normal paths.  The option can be enabled in
environments where there's a desire to attempt to audit whether or not a
reboot was cleanly requested or not.
Signed-off-by: NMatthew Garrett <mjg@redhat.com>
Acked-by: NSeiji Aguchi <seiji.aguchi@hds.com>
Cc: Seiji Aguchi <seiji.aguchi@hds.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Marco Stornelli <marco.stornelli@gmail.com>
Cc: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Don Zickus <dzickus@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch adds an attribute, NL80211_ATTR_INACTIVITY_TIMEOUT,
to set the inactivity timeout which can be used to remove the
station in AP mode. This can be passed in NL80211_CMD_START_AP
and used by the drivers which have AP MLME in firmware but
don't support get_station() properly. To disable inactivity
timer in userspace, wpa_s for example, there is a new flag,
NL80211_FEATURE_INACTIVITY_TIMER, in nl80211_feature_flags
through which drivers can register their capability to use
the inactivity timeout to free the stations.
Signed-off-by: NVasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com>
Acked-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Some drivers use internal netdev stats member to store part of their
stats, yet advertize ndo_get_stats64() to implement some 64bit fields.

Allow them to use netdev_stats_to_stats64() helper to make the copy of
netdev stats before they compute their 64bit counters.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Mesh peer links are established only if average rssi of the peer
candidate satisfies the threshold. This is not in 802.11s specification
but was requested by David Fulgham, an open80211s user. This is a way to avoid
marginal peer links with stations that are barely within range.

This patch adds a new mesh configuration parameter, mesh_rssi_threshold. This
feature is supported only for hardwares that report signal in dBm.
Signed-off-by: NAshok Nagarajan <ashok@cozybit.com>
Signed-off-by: NJavier Cardona <javier@cozybit.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

On SoCs the sprom is stored in the nvram in a special partition on the
flash chip. The nvram contains the sprom for the main bus, but
sometimes also for a pci devices using bcma. This patch makes it
possible for the arch code to register a function to fetch the needed
sprom from the nvram and provide it to the bcma code.
Signed-off-by: NHauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This function is needed by the bcm47xx arch code to get the number of
the ieee80211 core.
Signed-off-by: NHauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This patch extends the sprom struct to contain all sprom attributes
found in sprom version 1 to 9. This was done accordingly to the open
source part of the Broadcom SDK.
Signed-off-by: NHauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This member contains the country code encoded with two chars
Signed-off-by: NHauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

On sprom version 4 and 5 there are 4 values for pa_2g, pa_5gl, pa_5g
and pa_5gh, for sprom version 8 and 9 there are only 3. Make the per
path sprom store also work for older sprom versions.
Signed-off-by: NHauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>