GitHub/MotorolaMobilityLLC/kernel-slsi.git
10 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
David S. Miller [Wed, 6 Aug 2014 01:46:26 +0000 (18:46 -0700)]
Merge git://git./linux/kernel/git/davem/net

Conflicts:
drivers/net/Makefile
net/ipv6/sysctl_net_ipv6.c

Two ipv6_table_template[] additions overlap, so the index
of the ipv6_table[x] assignments needed to be adjusted.

In the drivers/net/Makefile case, we've gotten rid of the
garbage whereby we had to list every single USB networking
driver in the top-level Makefile, there is just one
"USB_NETWORKING" that guards everything.

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agocxgb4 : Disable recursive mailbox commands when enabling vi
Anish Bhatt [Tue, 5 Aug 2014 23:05:23 +0000 (16:05 -0700)]
cxgb4 : Disable recursive mailbox commands when enabling vi

Enabling a Virtual Interface can result in an interrupt during the processing
 of the VI Enable command and, in some paths, result in an attempt to issue
another command in the interrupt context, eventually crashing the system. Thus,
 we disable interrupts during the course of the VI Enable command and ensure
enable doesn't sleep.

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: Casey Leedom <leedom@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: reduce USB network driver config options.
Francois Romieu [Tue, 5 Aug 2014 21:10:52 +0000 (23:10 +0200)]
net: reduce USB network driver config options.

USB network drivers are already handled in drivers/net/usb/Kconfig.
Let's save the maintenance burden of dependencies in drivers/net/Makefile.

The newly introduced USB_NET_DRIVERS umbrella config option defaults
to 'y' so as to minimize the changes of behavior.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agotg3: Modify tg3_tso_bug() to handle multiple TX rings
Prashant Sreedharan [Tue, 5 Aug 2014 23:02:02 +0000 (16:02 -0700)]
tg3: Modify tg3_tso_bug() to handle multiple TX rings

tg3_tso_bug() was originally designed to handle only HW TX ring 0, Commit
d3f6f3a1d818410c17445bce4f4caab52eb102f1 ("tg3: Prevent page allocation failure
during TSO workaround") changed the driver logic to use tg3_tso_bug() for all
HW TX rings that are enabled. This patch fixes the regression by modifying
tg3_tso_bug() to handle multiple HW TX rings.

Signed-off-by: Prashant Sreedharan <prashant@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'amd-xgbe'
David S. Miller [Tue, 5 Aug 2014 23:47:07 +0000 (16:47 -0700)]
Merge branch 'amd-xgbe'

Tom Lendacky says:

====================
amd-xgbe: AMD XGBE driver update 2014-08-05

The following series of patches includes fixes/updates to the driver.

- Use dma_set_mask_and_coherent to set the DMA mask
- Move the phy connect/disconnect logic to allow for module unloading

Changes in V2:
- Check the return value of the dma_set_mask_and_coherent call
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoamd-xgbe: Perform phy connect/disconnect at dev open/stop
Lendacky, Thomas [Tue, 5 Aug 2014 18:30:44 +0000 (13:30 -0500)]
amd-xgbe: Perform phy connect/disconnect at dev open/stop

A change added to the mdiobus/phy api added a module_get/module_put
during phy connect/disconnect processing. Currently, the driver
performs a phy connect during module probe and a phy disconnect during
module remove. With the addition of the module_get during phy connect
the amd-xgbe module use count is incremented and can no longer be
unloaded.

Move the phy connect/disconnect from the driver probe/remove functions
to the net_device_ops ndo_open/ndo_stop functions.  This allows the
module use count to be decremented when the device(s) are brought down
and allows the module to be unloaded.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoamd-xgbe: Use dma_set_mask_and_coherent to set DMA mask
Lendacky, Thomas [Tue, 5 Aug 2014 18:30:38 +0000 (13:30 -0500)]
amd-xgbe: Use dma_set_mask_and_coherent to set DMA mask

Use the dma_set_mask_and_coherent function to set the DMA mask rather
than setting the DMA mask fields directly.  This was originally done
to work around a bug in the arm64 DMA support when RAM started above
the 4GB boundary which has since been fixed.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: sun4i-emac: fix memory leak on bad packet
Marc Zyngier [Tue, 5 Aug 2014 15:44:39 +0000 (16:44 +0100)]
net: sun4i-emac: fix memory leak on bad packet

Upon reception of a new frame, the emac driver checks for a number
of error conditions, and flag the packet as "bad" if any of these
are present. It then allocates a skb unconditionally, but only uses
it if the packet is "good". On the error path, the skb is just forgotten,
and the system leaks memory.

The piece of junk I have on my desk seems to encounter such error
frequently enough so that the box goes OOM after a couple of days,
which makes me grumpy.

Fix this by moving the allocation on the "good_packet" path (and
convert it to netdev_alloc_skb while we're at it).

Tested on a random Allwinner A20 board.

Cc: Stefan Roese <sr@denx.de>
Cc: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: <stable@vger.kernel.org> # 3.11+
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agosctp: fix possible seqlock seadlock in sctp_packet_transmit()
Eric Dumazet [Tue, 5 Aug 2014 14:49:52 +0000 (16:49 +0200)]
sctp: fix possible seqlock seadlock in sctp_packet_transmit()

Dave reported following splat, caused by improper use of
IP_INC_STATS_BH() in process context.

BUG: using __this_cpu_add() in preemptible [00000000] code: trinity-c117/14551
caller is __this_cpu_preempt_check+0x13/0x20
CPU: 3 PID: 14551 Comm: trinity-c117 Not tainted 3.16.0+ #33
 ffffffff9ec898f0 0000000047ea7e23 ffff88022d32f7f0 ffffffff9e7ee207
 0000000000000003 ffff88022d32f818 ffffffff9e397eaa ffff88023ee70b40
 ffff88022d32f970 ffff8801c026d580 ffff88022d32f828 ffffffff9e397ee3
Call Trace:
 [<ffffffff9e7ee207>] dump_stack+0x4e/0x7a
 [<ffffffff9e397eaa>] check_preemption_disabled+0xfa/0x100
 [<ffffffff9e397ee3>] __this_cpu_preempt_check+0x13/0x20
 [<ffffffffc0839872>] sctp_packet_transmit+0x692/0x710 [sctp]
 [<ffffffffc082a7f2>] sctp_outq_flush+0x2a2/0xc30 [sctp]
 [<ffffffff9e0d985c>] ? mark_held_locks+0x7c/0xb0
 [<ffffffff9e7f8c6d>] ? _raw_spin_unlock_irqrestore+0x5d/0x80
 [<ffffffffc082b99a>] sctp_outq_uncork+0x1a/0x20 [sctp]
 [<ffffffffc081e112>] sctp_cmd_interpreter.isra.23+0x1142/0x13f0 [sctp]
 [<ffffffffc081c86b>] sctp_do_sm+0xdb/0x330 [sctp]
 [<ffffffff9e0b8f1b>] ? preempt_count_sub+0xab/0x100
 [<ffffffffc083b350>] ? sctp_cname+0x70/0x70 [sctp]
 [<ffffffffc08389ca>] sctp_primitive_ASSOCIATE+0x3a/0x50 [sctp]
 [<ffffffffc083358f>] sctp_sendmsg+0x88f/0xe30 [sctp]
 [<ffffffff9e0d673a>] ? lock_release_holdtime.part.28+0x9a/0x160
 [<ffffffff9e0d62ce>] ? put_lock_stats.isra.27+0xe/0x30
 [<ffffffff9e73b624>] inet_sendmsg+0x104/0x220
 [<ffffffff9e73b525>] ? inet_sendmsg+0x5/0x220
 [<ffffffff9e68ac4e>] sock_sendmsg+0x9e/0xe0
 [<ffffffff9e1c0c09>] ? might_fault+0xb9/0xc0
 [<ffffffff9e1c0bae>] ? might_fault+0x5e/0xc0
 [<ffffffff9e68b234>] SYSC_sendto+0x124/0x1c0
 [<ffffffff9e0136b0>] ? syscall_trace_enter+0x250/0x330
 [<ffffffff9e68c3ce>] SyS_sendto+0xe/0x10
 [<ffffffff9e7f9be4>] tracesys+0xdd/0xe2

This is a followup of commits f1d8cba61c3c4b ("inet: fix possible
seqlock deadlocks") and 7f88c6b23afbd315 ("ipv6: fix possible seqlock
deadlock in ip6_finish_output2")

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Reported-by: Dave Jones <davej@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoRevert "net: phy: Set the driver when registering an MDIO bus device"
Fabio Estevam [Tue, 5 Aug 2014 11:13:42 +0000 (08:13 -0300)]
Revert "net: phy: Set the driver when registering an MDIO bus device"

Commit a71e3c37960ce5f9 ("net: phy: Set the driver when registering an MDIO bus
device") caused the following regression on the fec driver:

root@imx6qsabresd:~# echo mem > /sys/power/state
PM: Syncing filesystems ... done.
Freezing user space processes ... (elapsed 0.003 seconds) done.
Freezing remaining freezable tasks ... (elapsed 0.002 seconds) done.
Unable to handle kernel NULL pointer dereference at virtual address 0000002c
pgd = bcd14000
[0000002c] *pgd=4d9e0831, *pte=00000000, *ppte=00000000
Internal error: Oops: 17 [#1] SMP ARM
Modules linked in:
CPU: 0 PID: 617 Comm: sh Not tainted 3.16.0 #17
task: bc0c4e00 ti: bceb6000 task.ti: bceb6000
PC is at fec_suspend+0x10/0x70
LR is at dpm_run_callback.isra.7+0x34/0x6c
pc : [<803f8a98>]    lr : [<80361f44>]    psr: 600f0013
sp : bceb7d70  ip : bceb7d88  fp : bceb7d84
r10: 8091523c  r9 : 00000000  r8 : bd88f478
r7 : 803f8a88  r6 : 81165988  r5 : 00000000  r4 : 00000000
r3 : 00000000  r2 : 00000000  r1 : bd88f478  r0 : bd88f478
Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
Control: 10c5387d  Table: 4cd1404a  DAC: 00000015
Process sh (pid: 617, stack limit = 0xbceb6240)
Stack: (0xbceb7d70 to 0xbceb8000)
....

The problem with the original commit is explained by Russell King:

"It has the effect (as can be seen from the oops) of attaching the MDIO bus
device (itself is a bus-less device) to the platform driver, which means
that if the platform driver supports power management, it will be called
to power manage the MDIO bus device.

Moreover, drivers do not expect to be called for power management
operations for devices which they haven't probed, and certainly not for
devices which aren't part of the same bus that the driver is registered
against."

This reverts commit a71e3c37960ce5f9c6a519bc1215e3ba9fa83e75.

Cc: <stable@vger.kernel.org> #3.16
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agocxgb4vf: Turn off SGE RX/TX Callback Timers and interrupts in PCI shutdown routine
Hariprasad Shenai [Tue, 5 Aug 2014 10:11:28 +0000 (15:41 +0530)]
cxgb4vf: Turn off SGE RX/TX Callback Timers and interrupts in PCI shutdown routine

Need to turn off  SGE RX/TX Callback Timers & interrupt in cxgb4vf PCI Shutdown
routine in order to prevent crashes during reboot/poweroff when traffic is
running.

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge tag 'batman-adv-for-davem' of git://git.open-mesh.org/linux-merge
David S. Miller [Tue, 5 Aug 2014 23:39:47 +0000 (16:39 -0700)]
Merge tag 'batman-adv-for-davem' of git://git.open-mesh.org/linux-merge

Antonio Quartulli says:

====================
pull request: batman-adv 2014-08-05

this is a pull request intended for net-next/linux-3.17 (yeah..it's really
late).

Patches 1, 2 and 4 are really minor changes:
- kmalloc_array is substituted to kmalloc when possible (as suggested by
  checkpatch);
- net_ratelimited() is now used properly and the "suppressed" message is not
  printed anymore if not needed;
- the internal version number has been increased to reflect our current version.

Patch 3 instead is introducing a change in the metric computation function
by changing the penalty applied at each mesh hop from 15/255 (~6%) to
30/255 (~11%). This change is introduced by Simon Wunderlich after having
observed a performance improvement in several networks when using the new value.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoteam: Simplify return path of team_newlink
Toshiaki Makita [Tue, 5 Aug 2014 06:58:50 +0000 (15:58 +0900)]
team: Simplify return path of team_newlink

The variable "err" is not necessary.
Return register_netdevice() directly.

Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agobridge: Update outdated comment on promiscuous mode
Toshiaki Makita [Tue, 5 Aug 2014 06:57:15 +0000 (15:57 +0900)]
bridge: Update outdated comment on promiscuous mode

Now bridge ports can be non-promiscuous, vlan_vid_add() is no longer an
unnecessary operation.

Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'net-timestamp-next'
David S. Miller [Tue, 5 Aug 2014 23:36:01 +0000 (16:36 -0700)]
Merge branch 'net-timestamp-next'

Willem de Bruijn says:

====================
net-timestamp: new tx tstamps and tcp

Extend socket tx timestamping:
- allow multiple types of software timestamps aside from send (1)
- add software timestamp on enter packet scheduling (4)
- add software timestamp for TCP (5)
- add software timestamp for TCP on ACK (6)

The sk_flags option space is nearly exhausted. Also move the
many timestamp options to a new sk->sk_tstamps (2).

To disambiguate data when tstamps may arrive out of order,
optionally return a sequential ID assigned at send (3).

Extend Linux tx timestamping to monitoring of latency
incurred within the kernel stack and to protocols embedded in TCP.
Complex kernel setups may have multiple layers of queueing, including
multiple instances of packet scheduling, and many classes per layer.
Many applications embed discrete payloads into TCP bytestreams for
reliability, flow control, etcetera. Detecting application tail
latency in such scenarios relies on identifying the exact queue
responsible if on the host, or the network latency if otherwise.

Changelog:
v4->v5
  - define SCM_TSTAMP_SND == 0, for legacy behavior
  - add TCP tstamps without changing the generated byte stream
    - modify GSO and ACK to find offset: slightly more complex
      than previous invariant that it is the last byte
  - consistent naming of packet scheduling
    - rename SCM_TSTAMP_ENQ to SCM_TSTAMP_SCHED
  - add unique key in ee_data
  - add id field in ee_info to disambiguate tstamps
    - optional, only on new flag SOF_TIMESTAMPING_OPT_ID
    - for bytestream, in bytes

v3->v4
  - (v3 review comment) removed skb->mark packet identification (*A)
  - (v3 review comment) fixed indentation
  - tcp: fixed poll() to return POLLERR on non-zero queue
  - rebased to work without syststamp
  - comments: removed all traces of MSG_TSTAMP_.. (*B)

v2->v3
  - extend the SO_TIMESTAMPING API, instead of defining a new one.
  - add protocol independent support to correlate tstamps with data,
    based on returning skb->mark.
  - removed no-payload optimization and documentation (for now):

    I have a follow-on patch that reintroduces MSG_TSTAMP along with a
    new socket option SOF_TIMESTAMPING_OPT_ONFLAG. This is equivalent
    to sequence setsockopt(<enable>); send(..); setsockopt(<disable>),
    but avoids the need to define a MSG_TSTAMP_<TYPE> for each type.

    I will leave these three patches as follow-on, as this patchset is
    large enough as is.

v1->v2
  - expand timestamping (existing and new) to SOCK_RAW and ping sockets
  - rename sock_errqueue_timestamping to scm_timestamping
  - change timestamp data format: do not add fields to scm_timestamping.
      Doing so could break legacy applications. Instead, communicate
      through an existing, but unused, field in the error message.
  - rename SOF_.._OPT_TX_NO_PAYLOAD to shorter SOF_.._OPT_TSONLY
  - move msg_tstamp test app out of patchset and to github
      git://github.com/wdebruij/kerneltools.git
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet-timestamp: ACK timestamp for bytestreams
Willem de Bruijn [Tue, 5 Aug 2014 02:11:50 +0000 (22:11 -0400)]
net-timestamp: ACK timestamp for bytestreams

Add SOF_TIMESTAMPING_TX_ACK, a request for a tstamp when the last byte
in the send() call is acknowledged. It implements the feature for TCP.

The timestamp is generated when the TCP socket cumulative ACK is moved
beyond the tracked seqno for the first time. The feature ignores SACK
and FACK, because those acknowledge the specific byte, but not
necessarily the entire contents of the buffer up to that byte.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet-timestamp: TCP timestamping
Willem de Bruijn [Tue, 5 Aug 2014 02:11:49 +0000 (22:11 -0400)]
net-timestamp: TCP timestamping

TCP timestamping extends SO_TIMESTAMPING to bytestreams.

Bytestreams do not have a 1:1 relationship between send() buffers and
network packets. The feature interprets a send call on a bytestream as
a request for a timestamp for the last byte in that send() buffer.

The choice corresponds to a request for a timestamp when all bytes in
the buffer have been sent. That assumption depends on in-order kernel
transmission. This is the common case. That said, it is possible to
construct a traffic shaping tree that would result in reordering.
The guarantee is strong, then, but not ironclad.

This implementation supports send and sendpages (splice). GSO replaces
one large packet with multiple smaller packets. This patch also copies
the option into the correct smaller packet.

This patch does not yet support timestamping on data in an initial TCP
Fast Open SYN, because that takes a very different data path.

If ID generation in ee_data is enabled, bytestream timestamps return a
byte offset, instead of the packet counter for datagrams.

The implementation supports a single timestamp per packet. It silenty
replaces requests for previous timestamps. To avoid missing tstamps,
flush the tcp queue by disabling Nagle, cork and autocork. Missing
tstamps can be detected by offset when the ee_data ID is enabled.

Implementation details:

- On GSO, the timestamping code can be included in the main loop. I
moved it into its own loop to reduce the impact on the common case
to a single branch.

- To avoid leaking the absolute seqno to userspace, the offset
returned in ee_data must always be relative. It is an offset between
an skb and sk field. The first is always set (also for GSO & ACK).
The second must also never be uninitialized. Only allow the ID
option on sockets in the ESTABLISHED state, for which the seqno
is available. Never reset it to zero (instead, move it to the
current seqno when reenabling the option).

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet-timestamp: SCHED timestamp on entering packet scheduler
Willem de Bruijn [Tue, 5 Aug 2014 02:11:48 +0000 (22:11 -0400)]
net-timestamp: SCHED timestamp on entering packet scheduler

Kernel transmit latency is often incurred in the packet scheduler.
Introduce a new timestamp on transmission just before entering the
scheduler. When data travels through multiple devices (bonding,
tunneling, ...) each device will export an individual timestamp.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet-timestamp: add key to disambiguate concurrent datagrams
Willem de Bruijn [Tue, 5 Aug 2014 02:11:47 +0000 (22:11 -0400)]
net-timestamp: add key to disambiguate concurrent datagrams

Datagrams timestamped on transmission can coexist in the kernel stack
and be reordered in packet scheduling. When reading looped datagrams
from the socket error queue it is not always possible to unique
correlate looped data with original send() call (for application
level retransmits). Even if possible, it may be expensive and complex,
requiring packet inspection.

Introduce a data-independent ID mechanism to associate timestamps with
send calls. Pass an ID alongside the timestamp in field ee_data of
sock_extended_err.

The ID is a simple 32 bit unsigned int that is associated with the
socket and incremented on each send() call for which software tx
timestamp generation is enabled.

The feature is enabled only if SOF_TIMESTAMPING_OPT_ID is set, to
avoid changing ee_data for existing applications that expect it 0.
The counter is reset each time the flag is reenabled. Reenabling
does not change the ID of already submitted data. It is possible
to receive out of order IDs if the timestamp stream is not quiesced
first.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet-timestamp: move timestamp flags out of sk_flags
Willem de Bruijn [Tue, 5 Aug 2014 02:11:46 +0000 (22:11 -0400)]
net-timestamp: move timestamp flags out of sk_flags

sk_flags is reaching its limit. New timestamping options will not fit.
Move all of them into a new field sk->sk_tsflags.

Added benefit is that this removes boilerplate code to convert between
SOF_TIMESTAMPING_.. and SOCK_TIMESTAMPING_.. in getsockopt/setsockopt.

SOCK_TIMESTAMPING_RX_SOFTWARE is also used to toggle the receive
timestamp logic (netstamp_needed). That can be simplified and this
last key removed, but will leave that for a separate patch.

Signed-off-by: Willem de Bruijn <willemb@google.com>
----

The u16 in sock can be moved into a 16-bit hole below sk_gso_max_segs,
though that scatters tstamp fields throughout the struct.
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet-timestamp: extend SCM_TIMESTAMPING ancillary data struct
Willem de Bruijn [Tue, 5 Aug 2014 02:11:45 +0000 (22:11 -0400)]
net-timestamp: extend SCM_TIMESTAMPING ancillary data struct

Applications that request kernel tx timestamps with SO_TIMESTAMPING
read timestamps as recvmsg() ancillary data. The response is defined
implicitly as timespec[3].

1) define struct scm_timestamping explicitly and

2) add support for new tstamp types. On tx, scm_timestamping always
   accompanies a sock_extended_err. Define previously unused field
   ee_info to signal the type of ts[0]. Introduce SCM_TSTAMP_SND to
   define the existing behavior.

The reception path is not modified. On rx, no struct similar to
sock_extended_err is passed along with SCM_TIMESTAMPING.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agocxgb4i : Move stray CPL definitions to cxgb4 driver
Anish Bhatt [Mon, 4 Aug 2014 23:17:51 +0000 (16:17 -0700)]
cxgb4i : Move stray CPL definitions to cxgb4 driver

These belong to the t4 msg header, will ensure there is no accidental code
duplication in the future

Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agotcp: reduce spurious retransmits due to transient SACK reneging
Neal Cardwell [Mon, 4 Aug 2014 23:12:29 +0000 (19:12 -0400)]
tcp: reduce spurious retransmits due to transient SACK reneging

This commit reduces spurious retransmits due to apparent SACK reneging
by only reacting to SACK reneging that persists for a short delay.

When a sequence space hole at snd_una is filled, some TCP receivers
send a series of ACKs as they apparently scan their out-of-order queue
and cumulatively ACK all the packets that have now been consecutiveyly
received. This is essentially misbehavior B in "Misbehaviors in TCP
SACK generation" ACM SIGCOMM Computer Communication Review, April
2011, so we suspect that this is from several common OSes (Windows
2000, Windows Server 2003, Windows XP). However, this issue has also
been seen in other cases, e.g. the netdev thread "TCP being hoodwinked
into spurious retransmissions by lack of timestamps?" from March 2014,
where the receiver was thought to be a BSD box.

Since snd_una would temporarily be adjacent to a previously SACKed
range in these scenarios, this receiver behavior triggered the Linux
SACK reneging code path in the sender. This led the sender to clear
the SACK scoreboard, enter CA_Loss, and spuriously retransmit
(potentially) every packet from the entire write queue at line rate
just a few milliseconds before the ACK for each packet arrives at the
sender.

To avoid such situations, now when a sender sees apparent reneging it
does not yet retransmit, but rather adjusts the RTO timer to give the
receiver a little time (max(RTT/2, 10ms)) to send us some more ACKs
that will restore sanity to the SACK scoreboard. If the reneging
persists until this RTO then, as before, we clear the SACK scoreboard
and enter CA_Loss.

A 10ms delay tolerates a receiver sending such a stream of ACKs at
56Kbit/sec. And to allow for receivers with slower or more congested
paths, we wait for at least RTT/2.

We validated the resulting max(RTT/2, 10ms) delay formula with a mix
of North American and South American Google web server traffic, and
found that for ACKs displaying transient reneging:

 (1) 90% of inter-ACK delays were less than 10ms
 (2) 99% of inter-ACK delays were less than RTT/2

In tests on Google web servers this commit reduced reneging events by
75%-90% (as measured by the TcpExtTCPSACKReneging counter), without
any measurable impact on latency for user HTTP and SPDY requests.

Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'qlcnic'
David S. Miller [Tue, 5 Aug 2014 23:23:45 +0000 (16:23 -0700)]
Merge branch 'qlcnic'

Rajesh Borundia says:

====================
qlcnic: Bug fixes

The patch series contains following bug fixes.

* Aggregating tx stats in adapter variable was resulting
  in increase of stats when user runs ifconfig command
  and no traffic is running. Instead aggregate tx stats
  in local variable and then assign it to adapter struct
  variable.
* Set_driver_version was called after registering netdev
  which was resulting in a race between FLR in open
  handler and set_driver_version command as open handler
  can be called simulatneously on another cpu even if probe
  is not complete. So call this command before registering
  netdev.
* dcbnl_ops should be initialized before registering netdev
  as they are referenced in open handler.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoqlcnic: Initialize dcbnl_ops before register_netdev
Rajesh Borundia [Mon, 4 Aug 2014 15:51:18 +0000 (11:51 -0400)]
qlcnic: Initialize dcbnl_ops before register_netdev

o Initialization of dcbnl_ops after register netdev may result in
  dcbnl_ops not getting set before it is being accessed from open.
  So, moving it before register_netdev.

Signed-off-by: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoqlcnic: Set driver version before registering netdev
Rajesh Borundia [Mon, 4 Aug 2014 15:51:17 +0000 (11:51 -0400)]
qlcnic: Set driver version before registering netdev

o Earlier, set_drv_version was getting called after register_netdev.
  This was resulting in a race between set_drv_version and FLR called
  from open(). Moving set_drv_version before register_netdev avoids
  the race.

o Log response code in error message on CDRP failure.

Signed-off-by: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoqlcnic: Fix update of ethtool stats.
Rajesh Borundia [Mon, 4 Aug 2014 15:51:16 +0000 (11:51 -0400)]
qlcnic: Fix update of ethtool stats.

o Aggregating tx stats in adapter variable was resulting in
  an increase in stats even after no traffic was run and
  user runs ifconfig/ethtool command.
o qlcnic_update_stats used to accumulate stats in adapter
  struct at each function call, instead accumulate tx stats
  in local variable and then assign it to adapter structure.

Reported-by: Holger Kiehl <holger.kiehl@dwd.de>
Signed-off-by: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge tag 'batman-adv-fix-for-davem' of git://git.open-mesh.org/linux-merge
David S. Miller [Tue, 5 Aug 2014 23:20:16 +0000 (16:20 -0700)]
Merge tag 'batman-adv-fix-for-davem' of git://git.open-mesh.org/linux-merge

Antonio Quartulli says:

====================
pull request net: batman-adv 2014-08-04

this is a pull request intended for net.

It just contains a patch by Sven Eckelmann that fixes the
reception of out-of-order fragments. As explained in the
commit message, the issue was due to a wrong assumption
about hlist_for_each_entry() in batadv_frag_insert_packet().
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'xen-netback-next'
David S. Miller [Tue, 5 Aug 2014 23:04:55 +0000 (16:04 -0700)]
Merge branch 'xen-netback-next'

Zoltan Kiss says:

====================
xen-netback: Changes around carrier handling

This series starts using carrier off as a way to purge packets when the guest is
not able (or willing) to receive them. It is a much faster way to get rid of
packets waiting for an overwhelmed guest.
The first patch changes current netback code where it relies currently on
netif_carrier_ok.
The second turns off the carrier if the guest times out on a queue, and only
turn it on again if that queue (or queues) resurrects.
====================

Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoxen-netback: Turn off the carrier if the guest is not able to receive
Zoltan Kiss [Mon, 4 Aug 2014 15:20:58 +0000 (16:20 +0100)]
xen-netback: Turn off the carrier if the guest is not able to receive

Currently when the guest is not able to receive more packets, qdisc layer starts
a timer, and when it goes off, qdisc is started again to deliver a packet again.
This is a very slow way to drain the queues, consumes unnecessary resources and
slows down other guests shutdown.
This patch change the behaviour by turning the carrier off when that timer
fires, so all the packets are freed up which were stucked waiting for that vif.
Instead of the rx_queue_purge bool it uses the VIF_STATUS_RX_PURGE_EVENT bit to
signal the thread that either the timeout happened or an RX interrupt arrived,
so the thread can check what it should do. It also disables NAPI, so the guest
can't transmit, but leaves the interrupts on, so it can resurrect.
Only the queues which brought down the interface can enable it again, the bit
QUEUE_STATUS_RX_STALLED makes sure of that.

Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: xen-devel@lists.xenproject.org
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoxen-netback: Using a new state bit instead of carrier
Zoltan Kiss [Mon, 4 Aug 2014 15:20:57 +0000 (16:20 +0100)]
xen-netback: Using a new state bit instead of carrier

This patch introduces a new state bit VIF_STATUS_CONNECTED to track whether the
vif is in a connected state. Using carrier will not work with the next patch
in this series, which aims to turn the carrier temporarily off if the guest
doesn't seem to be able to receive packets.

Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: xen-devel@lists.xenproject.org
v2:
- rename the bitshift type to "enum state_bit_shift" here, not in the next patch
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge tag 'master-2014-07-31' of git://git.kernel.org/pub/scm/linux/kernel/git/linvil...
David S. Miller [Tue, 5 Aug 2014 20:18:20 +0000 (13:18 -0700)]
Merge tag 'master-2014-07-31' of git://git./linux/kernel/git/linville/wireless-next

Conflicts:
net/6lowpan/iphc.c

Minor conflicts in iphc.c were changes overlapping with some
style cleanups.

John W. Linville says:

====================
Please pull this last(?) batch of wireless change intended for the
3.17 stream...

For the NFC bits, Samuel says:

"This is a rather quiet one, we have:

- A new driver from ST Microelectronics for their NCI ST21NFCB,
  including device tree  support.

- p2p support for the ST21NFCA driver

- A few fixes an enhancements for the NFC digital laye"

For the Atheros bits, Kalle says:

"Michal and Janusz did some important RX aggregation fixes, basically we
were missing RX reordering altogether. The 10.1 firmware doesn't support
Ad-Hoc mode and Michal fixed ath10k so that it doesn't advertise Ad-Hoc
support with that firmware. Also he implemented a workaround for a KVM
issue."

For the Bluetooth bits, Gustavo and Johan say:

"To quote Gustavo from his previous request:

'Some last minute fixes for -next. We have a fix for a use after free in
RFCOMM, another fix to an issue with ADV_DIRECT_IND and one for ADV_IND with
auto-connection handling.  Last, we added support for reading the codec and
MWS setting for controllers that support these features.'

Additionally there are fixes to LE scanning, an update to conform to the 4.1
core specification as well as fixes for tracking the page scan state. All
of these fixes are important for 3.17."

And,

"We've got:

- 6lowpan fixes/cleanups
- A couple crash fixes, one for the Marvell HCI driver and another in LE SMP.
- Fix for an incorrect connected state check
- Fix for the bondable requirement during pairing (an issue which had
  crept in because of using "pairable" when in fact the actual meaning
  was "bondable" (these have different meanings in Bluetooth)"

Along with those are some late-breaking hardware support patches in
brcmfmac and b43 as well as a stray ath9k patch.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet/usb/hso: Add support for Option GTM671WFS
Ricardo Ribalda [Mon, 4 Aug 2014 09:11:49 +0000 (11:11 +0200)]
net/usb/hso: Add support for Option GTM671WFS

After this patch:

[   32.985530] hso: drivers/net/usb/hso.c: Option Wireless
[   33.000452] hso 2-1.4:1.7: Not our interface
[   33.001849] usbcore: registered new interface driver hso

root@qt5022:~# ls /dev/ttyHS*
/dev/ttyHS0  /dev/ttyHS1  /dev/ttyHS2  /dev/ttyHS3  /dev/ttyHS4
/dev/ttyHS5

root@qt5022:~# lsusb -d 0af0: -vvv

Bus 002 Device 003: ID 0af0:9200 Option
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.00
  bDeviceClass          255 Vendor Specific Class
  bDeviceSubClass       255 Vendor Specific Subclass
  bDeviceProtocol       255 Vendor Specific Protocol
  bMaxPacketSize0        64
  idVendor           0x0af0 Option
  idProduct          0x9200
  bcdDevice            0.00
  iManufacturer           3 Option N.V.
  iProduct                2 Globetrotter HSUPA Modem
  iSerial                 0
  bNumConfigurations      1
  Configuration Descriptor:
    bLength                 9
    bDescriptorType         2
    wTotalLength          200
    bNumInterfaces          8
    bConfigurationValue     1
    iConfiguration          1 Option Configuration
    bmAttributes         0xe0
      Self Powered
      Remote Wakeup
    MaxPower              100mA
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        0
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x81  EP 1 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x01  EP 1 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x82  EP 2 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x02  EP 2 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        2
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        3
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x84  EP 4 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x04  EP 4 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        4
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x85  EP 5 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x05  EP 5 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        5
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x06  EP 6 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x86  EP 6 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        6
      bAlternateSetting       0
      bNumEndpoints           3
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x87  EP 7 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               5
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x88  EP 8 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x07  EP 7 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval              32
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        7
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass         8 Mass Storage
      bInterfaceSubClass      6 SCSI
      bInterfaceProtocol     80 Bulk-Only
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x08  EP 8 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval               1
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x89  EP 9 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval               1
Device Qualifier (for other device speed):
  bLength                10
  bDescriptorType         6
  bcdUSB               2.00
  bDeviceClass          255 Vendor Specific Class
  bDeviceSubClass       255 Vendor Specific Subclass
  bDeviceProtocol       255 Vendor Specific Protocol
  bMaxPacketSize0        64
  bNumConfigurations      1
Device Status:     0x0001
  Self Powered

Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Acked-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: smc911x: fix %d confusingly prefixed with 0x in format string
Hans Wennborg [Mon, 4 Aug 2014 00:19:46 +0000 (17:19 -0700)]
net: smc911x: fix %d confusingly prefixed with 0x in format string

Signed-off-by: Hans Wennborg <hans@hanshq.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agodrivers: atm: fix %d confusingly prefixed with 0x in format strings
Hans Wennborg [Mon, 4 Aug 2014 00:18:20 +0000 (17:18 -0700)]
drivers: atm: fix %d confusingly prefixed with 0x in format strings

Signed-off-by: Hans Wennborg <hans@hanshq.net>
Acked-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agobatman-adv: Start new development cycle
Simon Wunderlich [Mon, 21 Jul 2014 14:05:35 +0000 (16:05 +0200)]
batman-adv: Start new development cycle

Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
10 years agobatman-adv: increase default hop penalty
Simon Wunderlich [Tue, 17 Jun 2014 10:16:03 +0000 (12:16 +0200)]
batman-adv: increase default hop penalty

The default hop penalty is currently set to 15, which is applied like
that for multi interface devices (e.g. dual band APs). Single band
devices will still use an effective penalty of 30 (hop penalty + wifi
penalty).

After receiving reports of too long paths in mesh networks with dual
band APs which were fixed by increasing the hop penalty, we'd like to
suggest to increase that default value in the default setting as well.
We've evaluated that increase in a handful of medium sized mesh
networks (5-20 nodes) with single and dual band devices, with changes
for the better (shorter routes, higher throughput) or no change at all.

This patch changes the hop penalty to 30, which will give an effective
penalty of 60 on single band devices (hop penalty + wifi penalty).

Signed-off-by: Simon Wunderlich <simon@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
10 years agobatman-adv: remove unnecessary logspam
André Gaul [Tue, 10 Jun 2014 15:50:31 +0000 (17:50 +0200)]
batman-adv: remove unnecessary logspam

This patch removes unnecessary logspam which resulted from superfluous
calls to net_ratelimit(). With the supplied patch, net_ratelimit() is
called after the loglevel has been checked.

Signed-off-by: André Gaul <gaul@web-yard.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
10 years agobatman-adv: Fix out-of-order fragmentation support
Sven Eckelmann [Mon, 26 May 2014 15:21:39 +0000 (17:21 +0200)]
batman-adv: Fix out-of-order fragmentation support

batadv_frag_insert_packet was unable to handle out-of-order packets because it
dropped them directly. This is caused by the way the fragmentation lists is
checked for the correct place to insert a fragmentation entry.

The fragmentation code keeps the fragments in lists. The fragmentation entries
are kept in descending order of sequence number. The list is traversed and each
entry is compared with the new fragment. If the current entry has a smaller
sequence number than the new fragment then the new one has to be inserted
before the current entry. This ensures that the list is still in descending
order.

An out-of-order packet with a smaller sequence number than all entries in the
list still has to be added to the end of the list. The used hlist has no
information about the last entry in the list inside hlist_head and thus the
last entry has to be calculated differently. Currently the code assumes that
the iterator variable of hlist_for_each_entry can be used for this purpose
after the hlist_for_each_entry finished. This is obviously wrong because the
iterator variable is always NULL when the list was completely traversed.

Instead the information about the last entry has to be stored in a different
variable.

This problem was introduced in 610bfc6bc99bc83680d190ebc69359a05fc7f605
("batman-adv: Receive fragmented packets and merge").

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
10 years agonetlink: fix lockdep splats
Eric Dumazet [Tue, 5 Aug 2014 05:50:07 +0000 (07:50 +0200)]
netlink: fix lockdep splats

With netlink_lookup() conversion to RCU, we need to use appropriate
rcu dereference in netlink_seq_socket_idx() & netlink_seq_next()

Reported-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Fixes: e341694e3eb57fc ("netlink: Convert netlink_lookup() to use RCU protected hash table")
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agotcp: md5: remove unneeded check in tcp_v4_parse_md5_keys
Dmitry Popov [Sun, 3 Aug 2014 18:45:19 +0000 (22:45 +0400)]
tcp: md5: remove unneeded check in tcp_v4_parse_md5_keys

tcpm_key is an array inside struct tcp_md5sig, there is no need to check it
against NULL.

Signed-off-by: Dmitry Popov <ixaphire@qrator.net>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoDrivers: net-next: hyperv: Increase the size of the sendbuf region
KY Srinivasan [Sat, 2 Aug 2014 17:42:02 +0000 (10:42 -0700)]
Drivers: net-next: hyperv: Increase the size of the sendbuf region

Intel did some benchmarking on our network throughput when Linux on Hyper-V
is as used as a gateway. This fix gave us almost a 1 Gbps additional throughput
on about 5Gbps base throughput we hadi, prior to increasing the sendbuf size.
The sendbuf mechanism is a copy based transport that we have which is clearly
more optimal than the copy-free page flipping mechanism (for small packets).
In the forwarding scenario, we deal only with MTU sized packets,
and increasing the size of the senbuf area gave us the additional performance.
For what it is worth, Windows guests on Hyper-V, I am told use similar sendbuf
size as well.

The exact value of sendbuf I think is less important than the fact that it needs
to be larger than what Linux can allocate as physically contiguous memory.
Thus the change over to allocating via vmalloc().

We currently allocate 16MB receive buffer and we use vmalloc there for allocation.
Also the low level channel code has already been modified to deal with physically
dis-contiguous memory in the ringbuffer setup.

Based on experimentation Intel did, they say there was some improvement in throughput
as the sendbuf size was increased up to 16MB and there was no effect on throughput
beyond 16MB. Thus I have chosen 16MB here.

Increasing the sendbuf value makes a material difference in small packet handling

In this version of the patch, based on David's feedback, I have added
additional details in the commit log.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: remove spurious zd1201 rule.
Francois Romieu [Mon, 4 Aug 2014 21:47:44 +0000 (23:47 +0200)]
net: remove spurious zd1201 rule.

Leftover from 5c601d0c942f5aaf7f3cff7e08f61047d70a964e ("wireless: move
zd1201 where it belongs").

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: phy: spi_ks8995: Introduce the use of devm_kzalloc
Himangi Saraogi [Sun, 3 Aug 2014 08:09:43 +0000 (13:39 +0530)]
net: phy: spi_ks8995: Introduce the use of devm_kzalloc

This patch introduces the use of devm_kzalloc and does away with the
kfrees in the probe and remove functions. Also, a label is removed.

Signed-off-by: Himangi Saraogi <himangi774@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agocxgb4: only free allocated fls
Hariprasad Shenai [Mon, 4 Aug 2014 11:31:30 +0000 (17:01 +0530)]
cxgb4: only free allocated fls

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agobridge: remove a useless comment
Michael S. Tsirkin [Mon, 4 Aug 2014 13:22:40 +0000 (15:22 +0200)]
bridge: remove a useless comment

commit 6cbdceeb1cb12c7d620161925a8c3e81daadb2e4
    bridge: Dump vlan information from a bridge port
introduced a comment in an attempt to explain the
code logic. The comment is unfinished so it confuses more
than it explains, remove it.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agobatman-adv: prefer kmalloc_array to kmalloc when possible
Antonio Quartulli [Sat, 24 May 2014 04:43:47 +0000 (06:43 +0200)]
batman-adv: prefer kmalloc_array to kmalloc when possible

Reported by checkpatch with the following warning:
WARNING: Prefer kmalloc_array over kmalloc with multiply

Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
10 years agor8152: add missing Makefile rule
JF Le Fillatre [Fri, 1 Aug 2014 23:26:40 +0000 (01:26 +0200)]
r8152: add missing Makefile rule

Add missing Makefile rule for r8152 driver

In the current kernel the r8152 driver is *never* built because of a missing rule in drivers/net/Makefile, despite being selected as built-in or module in the .config file. There is no error message or warning to indicate that the driver isn't built. This change adds the rule and lets the driver build.

Tested as built-in and module for 3.15.8.

Signed-off by: JF Le Fillatre <jflf-kernel@gmx.com>

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agocxgb4i : remove spurious use of rcu
Anish Bhatt [Sat, 2 Aug 2014 22:50:44 +0000 (15:50 -0700)]
cxgb4i : remove spurious use of rcu

As pointed out by the intel guys, there is no need to hold rcu read lock in
cxgbi_inet6addr_handler(), this patch removes it.

Fixes: 759a0cc5a3e1 ("cxgb4i: Add ipv6 code to driver, call into libcxgbi ipv6 api")
Signed-off-by: Anish Bhatt <anish@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'concurrent_hash_tables'
David S. Miller [Sun, 3 Aug 2014 02:49:47 +0000 (19:49 -0700)]
Merge branch 'concurrent_hash_tables'

Thomas Graf says:

====================
Lockless netlink_lookup() with new concurrent hash table

Netlink sockets are maintained in a hash table to allow efficient lookup
via the port ID for unicast messages. However, lookups currently require
a read lock to be taken. This series adds a new generic, resizable,
scalable, concurrent hash table based on the paper referenced in the first
patch. It then makes use of the new data type to implement lockless
netlink_lookup().

Patch 3/3 to convert nft_hash is included for reference but should be
merged via the netfilter tree. Inclusion in this series is to provide
context for the suggested API.

Against net-next since the initial user of the new hash table is in net/

Changes:
v4-v5:
 - use GFP_KERNEL to alloc Netlink buckets as suggested by Nikolay
   Aleksandrov
 - free nft hash element on removal as spotted by Nikolay Aleksandrov
   and Patrick McHardy
v3-v4:
 - fixed wrong shift assignment placement as spotted by Nikolay Aleksandrov
 - reverted default size of nft_hash to 4 as requested by Patrick McHardy,
   default size for other hash tables remains at 64 if no hint is given
 - fixed copyright as requested by Patrick McHardy
v2-v3:
 - fixed typo in nft_hash_destroy() when passing rhashtable handle
v1-v2:
 - fixed traversal off-by-one as spotted by Tobias Klauser
 - removed unlikely() from BUG_ON() as spotted by Josh Triplett
 - new 3rd patch to convert nft_hash to rhashtable
 - make rhashtable_insert() return void
 - nl_sk_hash_lock must be a mutex
 - fixed wrong name of rht_shrink_below_30()
 - exported symbols rht_grow_above_75() and rht_shrink_below_30()
 - allow table freeing with RCU callback
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonftables: Convert nft_hash to use generic rhashtable
Thomas Graf [Sat, 2 Aug 2014 09:47:46 +0000 (11:47 +0200)]
nftables: Convert nft_hash to use generic rhashtable

The sizing of the hash table and the practice of requiring a lookup
to retrieve the pprev to be stored in the element cookie before the
deletion of an entry is left intact.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Patrick McHardy <kaber@trash.net>
Reviewed-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonetlink: Convert netlink_lookup() to use RCU protected hash table
Thomas Graf [Sat, 2 Aug 2014 09:47:45 +0000 (11:47 +0200)]
netlink: Convert netlink_lookup() to use RCU protected hash table

Heavy Netlink users such as Open vSwitch spend a considerable amount of
time in netlink_lookup() due to the read-lock on nl_table_lock. Use of
RCU relieves the lock contention.

Makes use of the new resizable hash table to avoid locking on the
lookup.

The hash table will grow if entries exceeds 75% of table size up to a
total table size of 64K. It will automatically shrink if usage falls
below 30%.

Also splits nl_table_lock into a separate mutex to protect hash table
mutations and allow synchronize_rcu() to sleep while waiting for readers
during expansion and shrinking.

Before:
   9.16%  kpktgend_0  [openvswitch]      [k] masked_flow_lookup
   6.42%  kpktgend_0  [pktgen]           [k] mod_cur_headers
   6.26%  kpktgend_0  [pktgen]           [k] pktgen_thread_worker
   6.23%  kpktgend_0  [kernel.kallsyms]  [k] memset
   4.79%  kpktgend_0  [kernel.kallsyms]  [k] netlink_lookup
   4.37%  kpktgend_0  [kernel.kallsyms]  [k] memcpy
   3.60%  kpktgend_0  [openvswitch]      [k] ovs_flow_extract
   2.69%  kpktgend_0  [kernel.kallsyms]  [k] jhash2

After:
  15.26%  kpktgend_0  [openvswitch]      [k] masked_flow_lookup
   8.12%  kpktgend_0  [pktgen]           [k] pktgen_thread_worker
   7.92%  kpktgend_0  [pktgen]           [k] mod_cur_headers
   5.11%  kpktgend_0  [kernel.kallsyms]  [k] memset
   4.11%  kpktgend_0  [openvswitch]      [k] ovs_flow_extract
   4.06%  kpktgend_0  [kernel.kallsyms]  [k] _raw_spin_lock
   3.90%  kpktgend_0  [kernel.kallsyms]  [k] jhash2
   [...]
   0.67%  kpktgend_0  [kernel.kallsyms]  [k] netlink_lookup

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Reviewed-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agolib: Resizable, Scalable, Concurrent Hash Table
Thomas Graf [Sat, 2 Aug 2014 09:47:44 +0000 (11:47 +0200)]
lib: Resizable, Scalable, Concurrent Hash Table

Generic implementation of a resizable, scalable, concurrent hash table
based on [0]. The implementation supports both, fixed size keys specified
via an offset and length, or arbitrary keys via own hash and compare
functions.

Lookups are lockless and protected as RCU read side critical sections.
Automatic growing/shrinking based on user configurable watermarks is
available while allowing concurrent lookups to take place.

Objects to be hashed must include a struct rhash_head. The reason for not
using the existing struct hlist_head is that the expansion and shrinking
will have two buckets point to a single entry which would lead in obscure
reverse chaining behaviour.

Code includes a boot selftest if CONFIG_TEST_RHASHTABLE is defined.

[0] https://www.usenix.org/legacy/event/atc11/tech/final_files/Triplett.pdf

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Reviewed-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'intel-next'
David S. Miller [Sun, 3 Aug 2014 02:41:24 +0000 (19:41 -0700)]
Merge branch 'intel-next'

Aaron Brown says:

====================
Intel Wired LAN Driver Updates

This series contains updates to the i40e and i40evf drivers.

Vasu adds FCOE support, build options and a documentation pointer to i40e.

Shannon exposes a Firmware API request used to do register writes on the
driver's behalf and disables local loopback on VMDQ VSI in order to stop the
VEB from echoing the VMDQ packets back at the VSI.

Ashish corrects the vf_id offset for virtchnl messages in the case of multiple
PFs, removes support for vf unicast promiscuos mode to disallow VFs from
receiving traffic intended for another VF, updates the vfr_stat state check to
handle the existing and future mechanism and adds an adapter state check to
prevent re-arming the watchdog timer after i40evf_remove has been called and
the timer has been deleted.

Serey fixes an issue where a guest OS would panic when removing the vf driver
while the device is being reset due to an attempt to clean a non initialized
mac_filter_list.

Akeem makes a minor comment change.

Jessie changes an instance of sprintf to snprintf that was missed when the
driver was converted to use snprintf everywhere.

Mitch plugs a few memory leaks.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40evf: Fixed guest OS panic when removing vf driver
Serey Kong [Fri, 1 Aug 2014 20:27:15 +0000 (13:27 -0700)]
i40evf: Fixed guest OS panic when removing vf driver

Removing VF driver during device still in reset caused guest OS panic.

in the i40evf_remove(), we're trying to clean mac_filter_list which has
not been initialized since the device is still stuck at the reset.
The change is to initialize the filter_list before setting any task.

Change-ID: I8b59df7384416c7e6f2d264b598f447e1c2c92b0
Signed-off-by: Serey Kong <serey.kong@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40evf: fix memory leak on unused interfaces
Mitch Williams [Fri, 1 Aug 2014 20:27:14 +0000 (13:27 -0700)]
i40evf: fix memory leak on unused interfaces

If the driver is loaded and then unloaded before the interface is
brought up, then it will allocate a MAC filter entry and never free it.
To fix this, on unload, run through the mac filter list and free all the
entries. We also do this during reset recovery when the driver cannot
contact the PF and needs to shut down completely.

Change-ID: I15fabd67eb4a1bfc57605a7db60d0b5d819839db
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40evf: don't leak queue vectors
Mitch Williams [Fri, 1 Aug 2014 20:27:13 +0000 (13:27 -0700)]
i40evf: don't leak queue vectors

Fix a memory leak. Driver was allocating memory for queue vectors on
init but not freeing them on shutdown. These need to be freed at two
different times: during module unload, and during reset recovery when
the driver cannot contact the PF driver and needs to give up.

Change-ID: I7c1d0157a776e960d4da432dfe309035aad7c670
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40evf: do not re-arm watchdog after remove
Ashish Shah [Fri, 1 Aug 2014 20:27:12 +0000 (13:27 -0700)]
i40evf: do not re-arm watchdog after remove

Add in an adapter state check to prevent re-arming watchdog timer after
i40evf_remove has been called and timer has been deleted.

Change-ID: I636ba7c6322be8cbf053231959f90c0a2d8d803a
Signed-off-by: Ashish Shah <ashish.n.shah@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40evf: future-proof vfr_stat state check
Ashish Shah [Fri, 1 Aug 2014 20:27:11 +0000 (13:27 -0700)]
i40evf: future-proof vfr_stat state check

Previously defined state I40E_VFR_VFACTIVE uses bit 1 which is now set to
"reserved."  Update the state checks to also include I40E_VFR_COMPLETED.
This change will allow the VF to work with both existing and future PFs.

Change-ID: Ifd1d34f79f3b0ffd6d2550ee4dadc55825ff52f8
Signed-off-by: Ashish Shah <ashish.n.shah@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: remove support for vf unicast promiscuous mode
Ashish Shah [Fri, 1 Aug 2014 20:27:10 +0000 (13:27 -0700)]
i40e: remove support for vf unicast promiscuous mode

Remove the ability of a VF to set unicast promiscuous mode.
Considered to be a security risk to allow VFs to receive traffic
intended for other VFs so don't allow it, simply ignore the flag.

Also fix it to send the correct seid to aq for multicast promiscuous set.

Change-ID: Icb9c49a281a8e9d3aeebf991ef1533ac82b84b14
Signed-off-by: Ashish Shah <ashish.n.shah@intel.com>
Tested-by: Jim Young <jamesx.m.young@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: Minor comment changes
Akeem G Abodunrin [Fri, 1 Aug 2014 20:27:09 +0000 (13:27 -0700)]
i40e: Minor comment changes

Fixes comment for reset reason

Change-ID: I6fda4fa292255e6eb0f874502b4d38d722149b10
Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
Tested-by: Jim Young <jamesx.m.young@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40evf: fix scan warning on sprintf
Jesse Brandeburg [Fri, 1 Aug 2014 20:27:08 +0000 (13:27 -0700)]
i40evf: fix scan warning on sprintf

The driver was converted to use snprintf everywhere but this one function.
Just use snprintf, instead of sprintf.

Also a small spelling correction in a comment.

Change-ID: I59d45f94a52754c7b4cd6034df9a61d8132b7f77
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: disable local loopback on vmdq vsi
Shannon Nelson [Fri, 1 Aug 2014 20:27:07 +0000 (13:27 -0700)]
i40e: disable local loopback on vmdq vsi

The local loopback should only be enabled for VSIs that are supporting
cascaded VEBs or VEPA setups.  This is not the case here, and we need
to stop the VEB from echoing the VMDQ VSI packets back at the VSI.

Change-ID: I9dfb6ac79db24d04360d7efde62d81e20abc5090
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Jim Young <jamesx.m.young@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: use correct vf_id offset for virtchnl message
Ashish Shah [Fri, 1 Aug 2014 20:27:06 +0000 (13:27 -0700)]
i40e: use correct vf_id offset for virtchnl message

The vf_id needs to be offset by the vf_base_id from hw function capabilities
for the case of multiple PFs.

Change-ID: I20ca8621f98e9cdf98649380b8eeaa35db52677c
Signed-off-by: Ashish Shah <ashish.n.shah@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: expose debug_write_register request
Shannon Nelson [Fri, 1 Aug 2014 20:27:05 +0000 (13:27 -0700)]
i40e: expose debug_write_register request

Now that the HW registers are no longer in debug mode and many are
locked down for writes, we need to expose the Firmware API request
used to do writes on the driver's behalf.

Change-ID: I09a05c4dc9ea0b24c00193faac34d7799eaa8496
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Jim Young <jamesx.m.young>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: adds FCoE to build and updates its documentation
Vasu Dev [Fri, 1 Aug 2014 20:27:04 +0000 (13:27 -0700)]
i40e: adds FCoE to build and updates its documentation

Adds newly added FCoE files to the build but only if FCoE module is configured.

Also, updates i40e document for added FCoE support.

Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Jack Morgan<jack.morgan@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: Adds FCoE related code to i40e core driver
Vasu Dev [Fri, 1 Aug 2014 20:27:03 +0000 (13:27 -0700)]
i40e: Adds FCoE related code to i40e core driver

Adds FCoE specific code to existing i40e core driver to:-

1. have separate FCoE VSI with additional FCoE queues pairs.
2. have FCoE related hash defines.
3. have additional FCoE related stats code.
4. export and then re-use existing functions required by FCoE build.

Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Jack Morgan<jack.morgan@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoi40e: adds FCoE code to the i40e driver
Vasu Dev [Fri, 1 Aug 2014 20:27:02 +0000 (13:27 -0700)]
i40e: adds FCoE code to the i40e driver

This patch adds FCoE ( Fibre Channel Over Ethernet ) code for
Intel XL710 adapters. This patch is limited to only new FCoE
offloads code in newly added files by this patch and then
following patches in the series modifies rest of the existing
driver to enable FCoE with i40e driver.

Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Jack Morgan<jack.morgan@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'amd-xgbe-next'
David S. Miller [Sun, 3 Aug 2014 02:29:59 +0000 (19:29 -0700)]
Merge branch 'amd-xgbe-next'

Tom Lendacky says:

====================
amd-xgbe: AMD XGBE driver update 2014-08-01

The following series of patches includes minor fixes/updates to the
driver.

- Remove some uses of spinlock around ethtool/phylib areas
- Update Rx/Tx ready check logic
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoamd-xgbe-phy: Allow more time for Rx/Tx to become ready
Lendacky, Thomas [Fri, 1 Aug 2014 16:56:36 +0000 (11:56 -0500)]
amd-xgbe-phy: Allow more time for Rx/Tx to become ready

The current time range waiting for Rx/Tx to become ready can sometimes
be too short if a connection is not present.  Increase the number of
retries and the sleep to give a bit more time. Also, change level of
the message issued from _err to _dbg if Rx/Tx do not become ready
since the underlying logic will function as if no link is established
and retry eventually.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoamd-xgbe: Remove unnecessary spinlocks
Lendacky, Thomas [Fri, 1 Aug 2014 16:56:29 +0000 (11:56 -0500)]
amd-xgbe: Remove unnecessary spinlocks

Remove the spinlocks around the ethtool get and set settings
functions and within the link adjustment callback routine.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf
David S. Miller [Sat, 2 Aug 2014 23:43:04 +0000 (16:43 -0700)]
Merge git://git./pub/scm/linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
Netfilter/IPVS fixes for net

The following patchset contains Netfilter/IPVS fixes for your net tree,
they are:

1) Maintain all DSCP and ECN bits for IPv6 tun forwarding. This
   resolves an inconsistency between IPv4 and IPv6 behaviour.
   Patch from Alex Gartrell via Simon Horman.

2) Fix unnoticeable blink in xt_LED when the led-always-blink option is
   used, from Jiri Prchal.

3) Add missing return in nft_del_setelem(), otherwise this results in a
   double call of nft_data_uninit() in the nf_tables code, from Thomas Graf.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: dnet: Use managed interfaces
Himangi Saraogi [Fri, 1 Aug 2014 15:30:03 +0000 (21:00 +0530)]
net: dnet: Use managed interfaces

This patch introduces the use of managed interfaces like
devm_ioremap_resource and does away with the calls to free the
allocated memory in the probe and remove functions. Also, some
labels and variable are done away with. This fixes a bug as there
was a missing release_mem_region in the remove function.

Signed-off-by: Himangi Saraogi <himangi774@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: ks8851-ml: Use devm_ioremap_resource
Himangi Saraogi [Fri, 1 Aug 2014 15:29:13 +0000 (20:59 +0530)]
net: ks8851-ml: Use devm_ioremap_resource

This patch introduces the use of devm_ioremap_resource, devm_kmalloc and
does away with the functions to free the allocated memory in the probe
and remove functions. Also, some labels are done away with. A bug is
fixed as two regions are allocated in the probe function, but only one
is freed in the remove function.

Signed-off-by: Himangi Saraogi <himangi774@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agocirrus: cs89x0: Use managed interfaces
Himangi Saraogi [Fri, 1 Aug 2014 15:28:19 +0000 (20:58 +0530)]
cirrus: cs89x0: Use managed interfaces

This patch introduces the use of managed interfaces like
devm_ioremap_resource and does away with the functions to free the
allocated memory in the probe and remove functions. Also, many labels
are done away with. The field size in no longer needed and is hence
removed from the struct net_local.

Signed-off-by: Himangi Saraogi <himangi774@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: sh_eth: Add r8a7794 support
Hisashi Nakamura [Fri, 1 Aug 2014 15:03:00 +0000 (17:03 +0200)]
net: sh_eth: Add r8a7794 support

Signed-off-by: Hisashi Nakamura <hisashi.nakamura.ak@renesas.com>
[uli: added bindings documentation]
Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoipv6: data of fwmark_reflect sysctl needs to be updated on netns construction
Hannes Frederic Sowa [Fri, 1 Aug 2014 12:41:10 +0000 (14:41 +0200)]
ipv6: data of fwmark_reflect sysctl needs to be updated on netns construction

Fixes: e110861f86094cd ("net: add a sysctl to reflect the fwmark on replies")
Cc: Lorenzo Colitti <lorenzo@google.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'be2net-next'
David S. Miller [Sat, 2 Aug 2014 22:59:22 +0000 (15:59 -0700)]
Merge branch 'be2net-next'

Sathya Perla says:

====================
be2net: patch set

Patch 1 fixes a regression caused by a previous commit on net-next.
Old versions of BE3 FW may not support cmds to re-provision (and hence
optimize) resources/queues in SR-IOV config. Do not treat this FW cmd
failure as fatal and fail the function initialization. Instead, just
enable SR-IOV with the resources provided by the FW.

Patch 2 ignores a VF mac address setting if the new mac is already active
on the VF.

Patch 3 adds support to delete a FW-dump via ethtool on Lancer adapters.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agobe2net: support deleting FW dump via ethtool (only for Lancer)
Kalesh AP [Fri, 1 Aug 2014 12:17:32 +0000 (17:47 +0530)]
be2net: support deleting FW dump via ethtool (only for Lancer)

This patch adds support to delete an existing FW-dump in Lancer via ethtool.
Initiating a new dump is not allowed if a FW dump is already present in the
adapter. The existing dump has to be first explicitly deleted.

Signed-off-by: Kalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agobe2net: ignore VF mac address setting for the same mac
Vasundhara Volam [Fri, 1 Aug 2014 12:17:31 +0000 (17:47 +0530)]
be2net: ignore VF mac address setting for the same mac

ndo_set_vf_mac() call may be issued for a mac-addr that is already
active on a VF. If so, silently ignore the request.

Signed-off-by: Vasundhara Volam <vasundhara.volam@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agobe2net: ignore get/set profile FW cmd failures
Sathya Perla [Fri, 1 Aug 2014 12:17:30 +0000 (17:47 +0530)]
be2net: ignore get/set profile FW cmd failures

Old versions of BE3 FW may not support cmds to re-provision (and hence
optimize) resources/queues in SR-IOV config. Do not treat this FW cmd
failure as fatal and fail the function initialization. Instead, just
enable SR-IOV with the resources provided by the FW.

Prior to the "create optimal number of queues on SR-IOV config" patch
such failures were ignored.
Fixes: bec84e6b2 ("create optimal number of queues on SR-IOV config")

Reported-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agocdc_subset: deal with a device that needs reset for timeout
Oliver Neukum [Fri, 1 Aug 2014 12:01:51 +0000 (14:01 +0200)]
cdc_subset: deal with a device that needs reset for timeout

This device needs to be reset to recover from a timeout.
Unfortunately this can be handled only at the level of
the subdrivers.

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'inet-frags-next'
David S. Miller [Sat, 2 Aug 2014 22:31:40 +0000 (15:31 -0700)]
Merge branch 'inet-frags-next'

Nikolay Aleksandrov says:

====================
inet: frags: cleanup and kmem_cache use

This patchset does a couple of small cleanups in patches 1-5 and then in
patch 06 it introduces the use of kmem_cache for allocation/freeing of
inet_frag_queue+header objects.

v2: Broke up patch 02 into 3 patches as David suggested

Here are the results of a couple of netperf runs:
netperf options: -l 30 -I95,5 -i 15,10 -m 64k

- 10 gig before the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.155.1 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket  Message  Elapsed      Messages
Size    Size     Time         Okay Errors   Throughput
bytes   bytes    secs            #      #   10^6bits/sec

212992   64000   30.00      442466      0    7551.39
212992           30.00      439130           7494.45

- 10 gig after the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.155.1 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket  Message  Elapsed      Messages
Size    Size     Time         Okay Errors   Throughput
bytes   bytes    secs            #      #   10^6bits/sec

212992   64000   30.00      458846      0    7830.94
212992           30.00      457575           7809.25

- Virtio before the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.144.1 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket  Message  Elapsed      Messages
Size    Size     Time         Okay Errors   Throughput
bytes   bytes    secs            #      #   10^6bits/sec

212992   64000   30.00      735000      0    12543.96
212992           30.00      560322           9562.79

- Virtio after the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.144.1 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket  Message  Elapsed      Messages
Size    Size     Time         Okay Errors   Throughput
bytes   bytes    secs            #      #   10^6bits/sec

212992   64000   30.00      731729      0    12488.14
212992           30.00      647241           11046.21
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoinet: frags: use kmem_cache for inet_frag_queue
Nikolay Aleksandrov [Fri, 1 Aug 2014 10:29:48 +0000 (12:29 +0200)]
inet: frags: use kmem_cache for inet_frag_queue

Use kmem_cache to allocate/free inet_frag_queue objects since they're
all the same size per inet_frags user and are alloced/freed in high volumes
thus making it a perfect case for kmem_cache.

Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoinet: frags: use INET_FRAG_EVICTED to prevent icmp messages
Nikolay Aleksandrov [Fri, 1 Aug 2014 10:29:47 +0000 (12:29 +0200)]
inet: frags: use INET_FRAG_EVICTED to prevent icmp messages

Now that we have INET_FRAG_EVICTED we might as well use it to stop
sending icmp messages in the "frag_expire" functions instead of
stripping INET_FRAG_FIRST_IN from their flags when evicting.
Also fix the comment style in ip6_expire_frag_queue().

Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoinet: frags: fix function declaration alignments in inet_fragment
Nikolay Aleksandrov [Fri, 1 Aug 2014 10:29:46 +0000 (12:29 +0200)]
inet: frags: fix function declaration alignments in inet_fragment

Fix a couple of functions' declaration alignments.

Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoinet: frags: enum the flag definitions and add descriptions
Nikolay Aleksandrov [Fri, 1 Aug 2014 10:29:45 +0000 (12:29 +0200)]
inet: frags: enum the flag definitions and add descriptions

Move the flags to an enum definion, swap FIRST_IN/LAST_IN to be in increasing
order and add comments explaining each flag and the inet_frag_queue struct
members.

Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoinet: frags: rename last_in to flags
Nikolay Aleksandrov [Fri, 1 Aug 2014 10:29:44 +0000 (12:29 +0200)]
inet: frags: rename last_in to flags

The last_in field has been used to store various flags different from
first/last frag in so give it a more descriptive name: flags.

Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoinet: frags: use INC_STATS_BH in the ipv6 reassembly code
Nikolay Aleksandrov [Fri, 1 Aug 2014 10:29:43 +0000 (12:29 +0200)]
inet: frags: use INC_STATS_BH in the ipv6 reassembly code

Softirqs are already disabled so no need to do it again, thus let's be
consistent and use the IP6_INC_STATS_BH variant.

Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agodm9000: NULL dereferences on error in probe()
Dan Carpenter [Fri, 1 Aug 2014 08:23:57 +0000 (11:23 +0300)]
dm9000: NULL dereferences on error in probe()

The dm9000_release_board() function is called with NULL ->data_req and
->addr_req pointers if dm9000_probe() fails.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoipv4: remove nested rcu_read_lock/unlock
Duan Jiong [Fri, 1 Aug 2014 06:00:36 +0000 (14:00 +0800)]
ipv4: remove nested rcu_read_lock/unlock

ip_local_deliver_finish() already have a rcu_read_lock/unlock, so
the rcu_read_lock/unlock is unnecessary.

See the stack below:
ip_local_deliver_finish
|
|
->icmp_rcv
|
|
->icmp_socket_deliver

Suggested-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Duan Jiong <duanj.fnst@cn.fujitsu.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoiovec: make sure the caller actually wants anything in memcpy_fromiovecend
Sasha Levin [Fri, 1 Aug 2014 03:00:35 +0000 (23:00 -0400)]
iovec: make sure the caller actually wants anything in memcpy_fromiovecend

Check for cases when the caller requests 0 bytes instead of running off
and dereferencing potentially invalid iovecs.

Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'filter-next'
David S. Miller [Sat, 2 Aug 2014 22:04:10 +0000 (15:04 -0700)]
Merge branch 'filter-next'

Alexei Starovoitov says:

====================
net: filter: split sk_filter into socket and bpf, cleanup names

The main goal of the series is to split 'struct sk_filter' into socket and
bpf parts and cleanup names in the following way:
- everything that deals with sockets keeps 'sk_*' prefix
- everything that is pure BPF is changed to 'bpf_*' prefix

split 'struct sk_filter' into
struct sk_filter {
atomic_t        refcnt;
struct rcu_head rcu;
struct bpf_prog *prog;
};
and
struct bpf_prog {
        u32                     jited:1,
                                len:31;
        struct sock_fprog_kern  *orig_prog;
        unsigned int            (*bpf_func)(const struct sk_buff *skb,
                                            const struct bpf_insn *filter);
        union {
                struct sock_filter      insns[0];
                struct bpf_insn         insnsi[0];
                struct work_struct      work;
        };
};
so that 'struct bpf_prog' can be used independent of sockets and cleans up
'unattached' bpf use cases:
isdn, ppp, team, seccomp, ptp, xt_bpf, cls_bpf, test_bpf
which don't need refcnt/rcu fields.

It's a follow up to the rcu cleanup started by Pablo in
commit 34c5bd66e5 ("net: filter: don't release unattached filter through call_rcu()")

Patch 1 - cleans up socket memory charging and makes it possible for functions
  sk(bpf)_migrate_filter(), sk(bpf)_prepare_filter() to be socket independent
Patches 2-4 - trivial renames
Patch 5 - sk_filter split and renames of related sk_*() functions
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: filter: split 'struct sk_filter' into socket and bpf parts
Alexei Starovoitov [Thu, 31 Jul 2014 03:34:16 +0000 (20:34 -0700)]
net: filter: split 'struct sk_filter' into socket and bpf parts

clean up names related to socket filtering and bpf in the following way:
- everything that deals with sockets keeps 'sk_*' prefix
- everything that is pure BPF is changed to 'bpf_*' prefix

split 'struct sk_filter' into
struct sk_filter {
atomic_t        refcnt;
struct rcu_head rcu;
struct bpf_prog *prog;
};
and
struct bpf_prog {
        u32                     jited:1,
                                len:31;
        struct sock_fprog_kern  *orig_prog;
        unsigned int            (*bpf_func)(const struct sk_buff *skb,
                                            const struct bpf_insn *filter);
        union {
                struct sock_filter      insns[0];
                struct bpf_insn         insnsi[0];
                struct work_struct      work;
        };
};
so that 'struct bpf_prog' can be used independent of sockets and cleans up
'unattached' bpf use cases

split SK_RUN_FILTER macro into:
    SK_RUN_FILTER to be used with 'struct sk_filter *' and
    BPF_PROG_RUN to be used with 'struct bpf_prog *'

__sk_filter_release(struct sk_filter *) gains
__bpf_prog_release(struct bpf_prog *) helper function

also perform related renames for the functions that work
with 'struct bpf_prog *', since they're on the same lines:

sk_filter_size -> bpf_prog_size
sk_filter_select_runtime -> bpf_prog_select_runtime
sk_filter_free -> bpf_prog_free
sk_unattached_filter_create -> bpf_prog_create
sk_unattached_filter_destroy -> bpf_prog_destroy
sk_store_orig_filter -> bpf_prog_store_orig_filter
sk_release_orig_filter -> bpf_release_orig_filter
__sk_migrate_filter -> bpf_migrate_filter
__sk_prepare_filter -> bpf_prepare_filter

API for attaching classic BPF to a socket stays the same:
sk_attach_filter(prog, struct sock *)/sk_detach_filter(struct sock *)
and SK_RUN_FILTER(struct sk_filter *, ctx) to execute a program
which is used by sockets, tun, af_packet

API for 'unattached' BPF programs becomes:
bpf_prog_create(struct bpf_prog **)/bpf_prog_destroy(struct bpf_prog *)
and BPF_PROG_RUN(struct bpf_prog *, ctx) to execute a program
which is used by isdn, ppp, team, seccomp, ptp, xt_bpf, cls_bpf, test_bpf

Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: filter: rename sk_convert_filter() -> bpf_convert_filter()
Alexei Starovoitov [Thu, 31 Jul 2014 03:34:15 +0000 (20:34 -0700)]
net: filter: rename sk_convert_filter() -> bpf_convert_filter()

to indicate that this function is converting classic BPF into eBPF
and not related to sockets

Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: filter: rename sk_chk_filter() -> bpf_check_classic()
Alexei Starovoitov [Thu, 31 Jul 2014 03:34:14 +0000 (20:34 -0700)]
net: filter: rename sk_chk_filter() -> bpf_check_classic()

trivial rename to indicate that this functions performs classic BPF checking

Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: filter: rename sk_filter_proglen -> bpf_classic_proglen
Alexei Starovoitov [Thu, 31 Jul 2014 03:34:13 +0000 (20:34 -0700)]
net: filter: rename sk_filter_proglen -> bpf_classic_proglen

trivial rename to better match semantics of macro

Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: filter: simplify socket charging
Alexei Starovoitov [Thu, 31 Jul 2014 03:34:12 +0000 (20:34 -0700)]
net: filter: simplify socket charging

attaching bpf program to a socket involves multiple socket memory arithmetic,
since size of 'sk_filter' is changing when classic BPF is converted to eBPF.
Also common path of program creation has to deal with two ways of freeing
the memory.

Simplify the code by delaying socket charging until program is ready and
its size is known

Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonetfilter: nf_tables: Avoid duplicate call to nft_data_uninit() for same key
Thomas Graf [Fri, 1 Aug 2014 15:25:38 +0000 (17:25 +0200)]
netfilter: nf_tables: Avoid duplicate call to nft_data_uninit() for same key

nft_del_setelem() currently calls nft_data_uninit() twice on the same
key. Once to release the key which is guaranteed to be NFT_DATA_VALUE
and a second time in the error path to which it falls through.

The second call has been harmless so far though because the type
passed is always NFT_DATA_VALUE which is currently a no-op.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
10 years agonet: use inet6_iif instead of IP6CB()->iif
Duan Jiong [Fri, 1 Aug 2014 01:52:58 +0000 (09:52 +0800)]
net: use inet6_iif instead of IP6CB()->iif

Signed-off-by: Duan Jiong <duanj.fnst@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>