commit fd245a4a (net_sched: move TCQ_F_THROTTLED flag)
added a race.

qdisc_watchdog() is run from softirq, so special care should be taken or
we can lose one state transition (THROTTLED/RUNNING)

Prior to fd245a4a, we were manipulating q->flags (qdisc->flags &=
~TCQ_F_THROTTLED;) and this manipulation could only race with
qdisc_warn_nonwc().

Since we want to avoid atomic ops in qdisc fast path - it was the
meaning of commit 37112105 (QDISC_STATE_RUNNING dont need atomic
bit ops) - fix is to move THROTTLE bit into 'state' field, this one
being manipulated with SMP and IRQ safe operations.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Request_mem_region should be used with release_mem_region, not
release_resource.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression x,E;
@@
*x = request_mem_region(...)
... when != release_mem_region(x)
    when != x = E
* release_resource(x);
// </smpl>
Signed-off-by: NJulia Lawall <julia@diku.dk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Request_mem_region should be used with release_mem_region, not
release_resource.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression x,E;
@@
*x = request_mem_region(...)
... when != release_mem_region(x)
    when != x = E
* release_resource(x);
// </smpl>
Signed-off-by: NJulia Lawall <julia@diku.dk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This prevents possible race between bond_enslave and bond_handle_frame
as reported by Nicolas by moving rx_handler register/unregister.
slave->bond is added to hold pointer to master bonding sructure. That
way dev->master is no longer used in bond_handler_frame.
Also, this removes "BUG: scheduling while atomic" message
Reported-by: NNicolas de Pesloüan <nicolas.2p.debian@gmail.com>
Signed-off-by: NJiri Pirko <jpirko@redhat.com>
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Tested-by: NNicolas de Pesloüan <nicolas.2p.debian@free.fr>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Rmmod myri10ge crash at free_netdev() -> netif_napi_del(), because napi
structures are already deallocated. To fix call netif_napi_del() before
kfree() at myri10ge_free_slices().

Cc: stable@kernel.org
Signed-off-by: NStanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Doorbell is used according to usage of BlueFlame.
For Blue Flame to work in Ethernet mode QP number should have 0
at bits 6,7.
Allocating range of QPs accordingly.
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Do not allow a kernel consumer to allocate a UAR to serve for blue flame if the
number of available UARs gets below MLX4_NUM_RESERVED_UARS (currently 8). This
will allow userspace apps to open a device file and run things like
ibv_devinfo.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add mlx4_bitmap_avail() to give the number of available resources. We want to
use this as a hint to whether to allocate a resources or not. This patch is
introduced to be used with allocation blue flame registers.
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Using blue flame can improve latency by allowing the HW to more efficiently
access the WQE. This patch presents two functions that are used to allocate or
release HW resources for using blue flame; the caller need to supply a struct
mlx4_bf object when allocating resources. Consumers that make use of this API
should post doorbells to the UAR object pointed by the initialized struct
mlx4_bf;
Signed-off-by: NEli Cohen <eli@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The mlx4_en module now uses the new steering mechanism.
The RX packets are now steered through the MCG table instead
of Mac table for unicast, and default entry for multicast.
The feature is enabled through INIT_HCA
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For Ethernet mode only,
When we want to register QP as promiscuous, it must be added to all the
existing steering entries and also to the default one.
The promiscuous QP might also be on of "real" QPs,
which means we need to monitor every entry to avoid duplicates and ensure
we close an entry when all it has is promiscuous QPs.
Same mechanism both for unicast and multicast.
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The same packet steering mechanism would be used both for IB and Ethernet,
Both multicasts and unicasts.
This commit prepares the general infrastructure for this.
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

HW revision is derived from device ID and rev id.
Signed-off-by: NEugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The driver queries the FW for WOL support.
Ethtool get/set_wol is implemented accordingly.
Only magic packets are supported at the time.
Signed-off-by: NIgor Yarovinsky <igory@mellanox.co.il>
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Each RX ring will have its own interrupt vector, and TX rings will share one
(we mostly use polling for TX completions).
The vectors are assigned first time device is opened, and its name includes
the interface name and ring number.
Signed-off-by: NMarkuze Alex <markuze@mellanox.co.il>
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adding a pool of MSI-X vectors and EQs that can be used explicitly by mlx4_core
customers (mlx4_ib, mlx4_en). The consumers will assign their own names to the
interrupt vectors. Those vectors are not opened at mlx4 device initialization,
opened by demand.
Changed the max number of possible EQs according to the new scheme, no longer relies on
on number of cores.
The new functionality is exposed through mlx4_assign_eq() and mlx4_release_eq().
Customers that do not use the new API will get completion vectors as before.
Signed-off-by: NMarkuze Alex <markuze@mellanox.co.il>
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Instead of reseting the module parameters each ifup or mtu change,
they are being set once at device initialization
Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NYevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit 2c8cec5c (Cache learned PMTU information in inetpeer) added
an extra inet_putpeer() call in ip_rt_update_pmtu().

This results in various problems, since we can free one inetpeer, while
it is still in use.

Ref: http://www.spinics.net/lists/netdev/msg159121.htmlReported-by: NAlexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In commit 9435eb1c
("ipv4: Implement __ip_dev_find using new interface address hash.")
we reimplemented __ip_dev_find() so that it doesn't have to
do a full FIB table lookup.

Instead, it consults a hash table of addresses configured to
interfaces.

This works identically to the old code in all except one case,
and that is for loopback subnets.

The old code would match the loopback device for any IP address
that falls within a subnet configured to the loopback device.

Handle this corner case by doing the FIB lookup.

We could implement this via inet_addr_onlink() but:

1) Someone could configure many addresses to loopback and
   inet_addr_onlink() is a simple list traversal.

2) We know the old code works.
Reported-by: NJulian Anastasov <ja@ssi.bg>
Acked-by: NStephen Hemminger <shemminger@vyatta.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

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

In the current undo logic, cwnd is moderated after it was restored
to the value prior entering fast-recovery. It was moderated first
in tcp_try_undo_recovery then again in tcp_complete_cwr.

Since the undo indicates recovery was false, these moderations
are not necessary. If the undo is triggered when most of the
outstanding data have been acknowledged, the (restored) cwnd is
falsely pulled down to a small value.

This patch removes these cwnd moderations if cwnd is undone
  a) during fast-recovery
	b) by receiving DSACKs past fast-recovery
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The ipv6_dev_get_saddr() is currently called with an uninitialized
destination address. Although in tests it usually seemed to nevertheless
always fetch the right source address, there seems to be a possible race
condition.

Therefore this commit changes this, first setting the destination
address and only after that fetching the source address.
Reported-by: NJan Beulich <JBeulich@novell.com>
Signed-off-by: NLinus Lüssing <linus.luessing@web.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With recent changes to the driver(switch to new cpdma layer),
the support for buffer descriptor address translation logic
is broken. This affects platforms where the physical address of
the descriptors as seen by the DMA engine is different from the
physical address.

Original Patch adding translation logic support:
Commit: ad021ae8Signed-off-by: NSriramakrishnan A G <srk@ti.com>
Tested-By: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This avoids explicit cast to avoid 'discards qualifiers'
compiler warning in a netfilter patch that i've been working on.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If SR-IOV is enabled by firmware, even if it is not enabled in the PCI
capability, TX pushes using write-combining may be corrupted.

We want to know whether it is enabled before mapping the NIC
registers, and even if PCI extended capabilities are not accessible.
Therefore, we look for the MSI capability, which is removed if SR-IOV
is enabled.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Optimize the calling of fib_add_ifaddr for all
secondary addresses after the promoted one to start from
their place, not from the new place of the promoted
secondary. It will save some CPU cycles because we
are sure the promoted secondary was first for the subnet
and all next secondaries do not change their place.
Signed-off-by: NJulian Anastasov <ja@ssi.bg>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The secondary address promotion relies on fib_sync_down_addr
to remove all routes created for the secondary addresses when
the old primary address is deleted. It does not happen for cases
when the primary address is also in another subnet. Fix that
by deleting local and broadcast routes for all secondaries while
they are on device list and by faking that all addresses from
this subnet are to be deleted. It relies on fib_del_ifaddr being
able to ignore the IPs from the concerned subnet while checking
for duplication.
Signed-off-by: NJulian Anastasov <ja@ssi.bg>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Alex Sidorenko reported for problems with local
routes left after IP addresses are deleted. It happens
when same IPs are used in more than one subnet for the
device.

	Fix fib_del_ifaddr to restrict the checks for duplicate
local and broadcast addresses only to the IFAs that use
our primary IFA or another primary IFA with same address.
And we expect the prefsrc to be matched when the routes
are deleted because it is possible they to differ only by
prefsrc. This patch prevents local and broadcast routes
to be leaked until their primary IP is deleted finally
from the box.

	As the secondary address promotion needs to delete
the routes for all secondaries that used the old primary IFA,
add option to ignore these secondaries from the checks and
to assume they are already deleted, so that we can safely
delete the route while these IFAs are still on the device list.
Reported-by: NAlex Sidorenko <alexandre.sidorenko@hp.com>
Signed-off-by: NJulian Anastasov <ja@ssi.bg>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

fib_table_delete forgets to match the routes by prefsrc.
Callers can specify known IP in fc_prefsrc and we should remove
the exact route. This is needed for cases when same local or
broadcast addresses are used in different subnets and the
routes differ only in prefsrc. All callers that do not provide
fc_prefsrc will ignore the route prefsrc as before and will
delete the first occurence. That is how the ip route del default
magic works.

	Current callers are:

- ip_rt_ioctl where rtentry_to_fib_config provides fc_prefsrc only
when the provided device name matches IP label with colon.

- inet_rtm_delroute where RTA_PREFSRC is optional too

- fib_magic which deals with routes when deleting addresses
and where the fc_prefsrc is always set with the primary IP
for the concerned IFA.
Signed-off-by: NJulian Anastasov <ja@ssi.bg>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The set_multicast operation performs asynchronous writes to the
device, with some addresses pointing to the stack. Bad things may
happen, and this is trapped CONFIG_DMA_API_DEBUG:

[    5.237762] WARNING: at /build/buildd/linux-linaro-omap-2.6.38/lib/dma-debug.c:867 check_for_stack+0xd4/0x100()
[    5.237792] ehci-omap ehci-omap.0: DMA-API: device driver maps memory fromstack [addr=d9c77dec]
[    5.237792] Modules linked in: smsc95xx(+) usbnet twl6030_usb twl4030_pwrbutton leds_gpio omap_wdt omap2_mcspi
[    5.237854] [<c006d618>] (unwind_backtrace+0x0/0xf8) from [<c00a6a14>] (warn_slowpath_common+0x54/0x64)
[    5.237884] [<c00a6a14>] (warn_slowpath_common+0x54/0x64) from [<c00a6ab8>] (warn_slowpath_fmt+0x30/0x40)
[    5.237915] [<c00a6ab8>] (warn_slowpath_fmt+0x30/0x40) from [<c034e9d8>] (check_for_stack+0xd4/0x100)
[    5.237915] [<c034e9d8>] (check_for_stack+0xd4/0x100) from [<c034fea8>] (debug_dma_map_page+0xb4/0xdc)
[    5.237976] [<c034fea8>] (debug_dma_map_page+0xb4/0xdc) from [<c04242f0>] (map_urb_for_dma+0x26c/0x304)
[    5.237976] [<c04242f0>] (map_urb_for_dma+0x26c/0x304) from [<c0424594>] (usb_hcd_submit_urb+0x78/0x19c)
[    5.238037] [<c0424594>] (usb_hcd_submit_urb+0x78/0x19c) from [<bf049c5c>] (smsc95xx_write_reg_async+0xb4/0x130 [smsc95xx])
[    5.238067] [<bf049c5c>] (smsc95xx_write_reg_async+0xb4/0x130 [smsc95xx]) from [<bf049dd4>] (smsc95xx_set_multicast+0xfc/0x148 [smsc95xx])
[    5.238098] [<bf049dd4>] (smsc95xx_set_multicast+0xfc/0x148 [smsc95xx]) from [<bf04a118>] (smsc95xx_reset+0x2f8/0x68c [smsc95xx])
[    5.238128] [<bf04a118>] (smsc95xx_reset+0x2f8/0x68c [smsc95xx]) from [<bf04a8cc>] (smsc95xx_bind+0xcc/0x188 [smsc95xx])
[    5.238159] [<bf04a8cc>] (smsc95xx_bind+0xcc/0x188 [smsc95xx]) from [<bf03ef1c>] (usbnet_probe+0x204/0x4c4 [usbnet])
[    5.238220] [<bf03ef1c>] (usbnet_probe+0x204/0x4c4 [usbnet]) from [<c0429078>] (usb_probe_interface+0xe4/0x1c4)
[    5.238250] [<c0429078>] (usb_probe_interface+0xe4/0x1c4) from [<c03a8770>] (really_probe+0x64/0x160)
[    5.238250] [<c03a8770>] (really_probe+0x64/0x160) from [<c03a8a30>] (driver_probe_device+0x48/0x60)
[    5.238281] [<c03a8a30>] (driver_probe_device+0x48/0x60) from [<c03a8ad4>] (__driver_attach+0x8c/0x90)
[    5.238311] [<c03a8ad4>] (__driver_attach+0x8c/0x90) from [<c03a7b24>] (bus_for_each_dev+0x50/0x7c)
[    5.238311] [<c03a7b24>] (bus_for_each_dev+0x50/0x7c) from [<c03a82ec>] (bus_add_driver+0x190/0x250)
[    5.238311] [<c03a82ec>] (bus_add_driver+0x190/0x250) from [<c03a8cf8>] (driver_register+0x78/0x13c)
[    5.238433] [<c03a8cf8>] (driver_register+0x78/0x13c) from [<c0428040>] (usb_register_driver+0x78/0x13c)
[    5.238464] [<c0428040>] (usb_register_driver+0x78/0x13c) from [<c005b680>] (do_one_initcall+0x34/0x188)
[    5.238494] [<c005b680>] (do_one_initcall+0x34/0x188) from [<c00e11f0>] (sys_init_module+0xb0/0x1c0)
[    5.238525] [<c00e11f0>] (sys_init_module+0xb0/0x1c0) from [<c0065c40>] (ret_fast_syscall+0x0/0x30)

Move the two offenders to the private structure which is kmalloc-ed,
and thus safe.
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>
Cc: Steve Glendinning <steve.glendinning@smsc.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Implement compatibility with new hw_features for dev_disable_lro().
This is a transition path - dev_disable_lro() should be later
integrated into netdev_fix_features() after all drivers are converted.
Signed-off-by: NMichał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As part of the work to make IPVS network namespace aware
__ip_vs_app_mutex was replaced by a per-namespace lock,
ipvs->app_mutex. ipvs->app_key is also supplied for debugging purposes.

Unfortunately this implementation results in ipvs->app_key residing
in non-static storage which at the very least causes a lockdep warning.

This patch takes the rather heavy-handed approach of reinstating
__ip_vs_app_mutex which will cover access to the ipvs->list_head
of all network namespaces.

[   12.610000] IPVS: Creating netns size=2456 id=0
[   12.630000] IPVS: Registered protocols (TCP, UDP, SCTP, AH, ESP)
[   12.640000] BUG: key ffff880003bbf1a0 not in .data!
[   12.640000] ------------[ cut here ]------------
[   12.640000] WARNING: at kernel/lockdep.c:2701 lockdep_init_map+0x37b/0x570()
[   12.640000] Hardware name: Bochs
[   12.640000] Pid: 1, comm: swapper Tainted: G        W 2.6.38-kexec-06330-g69b7efe-dirty #122
[   12.650000] Call Trace:
[   12.650000]  [<ffffffff8102e685>] warn_slowpath_common+0x75/0xb0
[   12.650000]  [<ffffffff8102e6d5>] warn_slowpath_null+0x15/0x20
[   12.650000]  [<ffffffff8105967b>] lockdep_init_map+0x37b/0x570
[   12.650000]  [<ffffffff8105829d>] ? trace_hardirqs_on+0xd/0x10
[   12.650000]  [<ffffffff81055ad8>] debug_mutex_init+0x38/0x50
[   12.650000]  [<ffffffff8104bc4c>] __mutex_init+0x5c/0x70
[   12.650000]  [<ffffffff81685ee7>] __ip_vs_app_init+0x64/0x86
[   12.660000]  [<ffffffff81685a3b>] ? ip_vs_init+0x0/0xff
[   12.660000]  [<ffffffff811b1c33>] T.620+0x43/0x170
[   12.660000]  [<ffffffff811b1e9a>] ? register_pernet_subsys+0x1a/0x40
[   12.660000]  [<ffffffff81685a3b>] ? ip_vs_init+0x0/0xff
[   12.660000]  [<ffffffff81685a3b>] ? ip_vs_init+0x0/0xff
[   12.660000]  [<ffffffff811b1db7>] register_pernet_operations+0x57/0xb0
[   12.660000]  [<ffffffff81685a3b>] ? ip_vs_init+0x0/0xff
[   12.670000]  [<ffffffff811b1ea9>] register_pernet_subsys+0x29/0x40
[   12.670000]  [<ffffffff81685f19>] ip_vs_app_init+0x10/0x12
[   12.670000]  [<ffffffff81685a87>] ip_vs_init+0x4c/0xff
[   12.670000]  [<ffffffff8166562c>] do_one_initcall+0x7a/0x12e
[   12.670000]  [<ffffffff8166583e>] kernel_init+0x13e/0x1c2
[   12.670000]  [<ffffffff8128c134>] kernel_thread_helper+0x4/0x10
[   12.670000]  [<ffffffff8128ad40>] ? restore_args+0x0/0x30
[   12.680000]  [<ffffffff81665700>] ? kernel_init+0x0/0x1c2
[   12.680000]  [<ffffffff8128c130>] ? kernel_thread_helper+0x0/0x1global0
Signed-off-by: NSimon Horman <horms@verge.net.au>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Julian Anastasov <ja@ssi.bg>
Cc: Hans Schillstrom <hans@schillstrom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Reported-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Cc: Simon Horman <horms@verge.net.au>
Cc: Julian Anastasov <ja@ssi.bg>
Acked-by: NSimon Horman <horms@verge.net.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 44540960 "veth: move loopback logic to common location" introduced
a bug in the packet counters.  I don't understand why that happened as it
is not explained in the comments and the mut check in dev_forward_skb
retains the assumption that skb->len is the total length of the packet.

I just measured this emperically by setting up a veth pair between two
noop network namespaces setting and attempting a telnet connection between
the two.  I saw three packets in each direction and the byte counters were
exactly 14*3 = 42 bytes high in each direction.  I got the actual
packet lengths with tcpdump.

So remove the extra ETH_HLEN from the veth byte count totals.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>