GitHub/LineageOS/android_kernel_motorola_exynos9610.git
14 years agotbf: stop wanton destruction of children (v2)
stephen hemminger [Fri, 14 May 2010 14:38:59 +0000 (14:38 +0000)]
tbf: stop wanton destruction of children (v2)

Several netem users use TBF for rate control. But every time the parameters
of TBF are changed it destroys the child qdisc, requiring reconfigation.
Better to just keep child qdisc and just notify it of changed limit.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: Remove unnecessary semicolons after switch statements
Joe Perches [Fri, 14 May 2010 10:58:26 +0000 (10:58 +0000)]
net: Remove unnecessary semicolons after switch statements

Also added an explicit break; to avoid
a fallthrough in net/ipv4/tcp_input.c

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoX25: Remove bkl in sockopts
andrew hendry [Sun, 16 May 2010 23:00:35 +0000 (23:00 +0000)]
X25: Remove bkl in sockopts

Removes the BKL in x25 setsock and getsockopts.

Signed-off-by: Andrew Hendry <andrew.hendry@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoX25: Move accept approve flag to bitfield
andrew hendry [Sun, 16 May 2010 23:00:27 +0000 (23:00 +0000)]
X25: Move accept approve flag to bitfield

Moves the x25 accept approve flag from char into bitfield.

Signed-off-by: Andrew Hendry <andrew.hendry@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoX25: Move interrupt flag to bitfield
andrew hendry [Sun, 16 May 2010 23:00:02 +0000 (23:00 +0000)]
X25: Move interrupt flag to bitfield

Moves the x25 interrupt flag from char into bitfield.

Signed-off-by: Andrew Hendry <andrew.hendry@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoX25: Move qbit flag to bitfield
andrew hendry [Sun, 16 May 2010 22:59:41 +0000 (22:59 +0000)]
X25: Move qbit flag to bitfield

Moves the X25 q bit flag from char into a bitfield to allow BKL cleanup.

Signed-off-by: Andrew Hendry <andrew.hendry@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobnx2x: avoid TX timeout when stopping device
Stanislaw Gruszka [Tue, 18 May 2010 00:35:38 +0000 (17:35 -0700)]
bnx2x: avoid TX timeout when stopping device

When stop device call netif_carrier_off() just after disabling TX queue to
avoid possibility of netdev watchdog warning and ->ndo_tx_timeout() invocation.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Acked-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobnx2: Use netif_carrier_off() to prevent timeout.
Michael Chan [Tue, 18 May 2010 00:34:43 +0000 (17:34 -0700)]
bnx2: Use netif_carrier_off() to prevent timeout.

Based on original patch from Stanislaw Gruszka <sgruszka@redhat.com>.

Using netif_carrier_off() is better than updating all the ->trans_start
on all the tx queues.

netif_carrier_off() needs to be called after bnx2_disable_int_sync()
to guarantee no race conditions with the serdes timers that can
modify the carrier state.

If the chip or phy is reset, carrier will turn back on when we get the
link interrupt.  If there is no reset, we need to turn carrier back on
in bnx2_netif_start().  Again, the phy_lock prevents race conditions with
the serdes timers.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobnx2: Update 5709 MIPS firmware and version to 2.0.15.
Michael Chan [Tue, 18 May 2010 00:33:31 +0000 (17:33 -0700)]
bnx2: Update 5709 MIPS firmware and version to 2.0.15.

New firmware fixes a performance regression on small packets.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobnx2: Fix register printouts during NETEV_WATCHDOG.
Eddie Wai [Tue, 18 May 2010 00:32:56 +0000 (17:32 -0700)]
bnx2: Fix register printouts during NETEV_WATCHDOG.

Dump the correct MCP registers and add EMAC_RX_STATUS register during
NETDEV_WATCHDOG for debugging.

Signed-off-by: Eddie Wai <waie@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: mark device state fail
Amit Kumar Salecha [Mon, 17 May 2010 01:22:14 +0000 (01:22 +0000)]
qlcnic: mark device state fail

Device state need to be mark as FAILED, if fail to start firmware.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: remove unused register
Amit Kumar Salecha [Mon, 17 May 2010 01:22:13 +0000 (01:22 +0000)]
qlcnic: remove unused register

Removing register defines which are not used by Qlgoic CNA device.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: fix internal loopback test
Sucheta Chakraborty [Mon, 17 May 2010 01:22:12 +0000 (01:22 +0000)]
qlcnic: fix internal loopback test

Reset/set DEV_UP bit during allocation and deallocation of resources.

Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: module param for firmware load option
Amit Kumar Salecha [Mon, 17 May 2010 01:22:11 +0000 (01:22 +0000)]
qlcnic: module param for firmware load option

By default fw is loaded from flash, user can
change this priority using load_fw_file module param.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: fix rx bytes statistics
Sucheta Chakraborty [Mon, 17 May 2010 01:22:10 +0000 (01:22 +0000)]
qlcnic: fix rx bytes statistics

Added lrobytes to it.

Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: change adapter name display
Sucheta Chakraborty [Mon, 17 May 2010 01:22:09 +0000 (01:22 +0000)]
qlcnic: change adapter name display

Append mac address to adapter name.

Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: fix memory leaks
Anirban Chakraborty [Mon, 17 May 2010 01:22:08 +0000 (01:22 +0000)]
qlcnic: fix memory leaks

Fixes memory leak in error path when memory allocation
for adapter data structures fails.

Signed-off-by: Anirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: check for mii_bus platform data
Sonic Zhang [Mon, 10 May 2010 05:39:14 +0000 (05:39 +0000)]
netdev: bfin_mac: check for mii_bus platform data

If the platform data for the mii_bus is missing, gracefully error out
rather than deference NULL pointers.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: handle timeouts with the MDIO registers gracefully
Mike Frysinger [Mon, 10 May 2010 05:39:13 +0000 (05:39 +0000)]
netdev: bfin_mac: handle timeouts with the MDIO registers gracefully

Have the low level MDIO functions pass back up timeout information so we
don't waste time polling them multiple times when there is a problem, and
so we don't let higher layers think the device is available when it isn't.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: use promiscuous flag for promiscuous mode
Sonic Zhang [Mon, 10 May 2010 05:39:12 +0000 (05:39 +0000)]
netdev: bfin_mac: use promiscuous flag for promiscuous mode

Rather than using the Receive All Frames (RAF) bit to enable promiscuous
mode, use the Promiscuous (PR) bit.  This lowers overhead at runtime as
we let the hardware process the packets that should actually be checked.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: add support for wake-on-lan magic packets
Michael Hennerich [Mon, 10 May 2010 05:39:11 +0000 (05:39 +0000)]
netdev: bfin_mac: add support for wake-on-lan magic packets

Note that WOL works only in PM Suspend Standby Mode (Sleep Mode).

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: clear RXCKS if hardware generated checksum is not enabled
Sonic Zhang [Mon, 10 May 2010 05:39:10 +0000 (05:39 +0000)]
netdev: bfin_mac: clear RXCKS if hardware generated checksum is not enabled

Otherwise we might be get a setting mismatch from a previous module or
bootloader and what the driver currently expects.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: deduce Ethernet FCS from hardware IP payload checksum
Sonic Zhang [Mon, 10 May 2010 05:39:09 +0000 (05:39 +0000)]
netdev: bfin_mac: deduce Ethernet FCS from hardware IP payload checksum

IP checksum is based on 16-bit one's complement algorithm, so to deduce a
value from checksum is equal to add its complement.

Unfortunately, the Blackfin on-chip MAC checksum logic only works when the
IP packet has a header length of 20 bytes.  This is true for most IPv4
packets, but not for IPv6 packets or IPv4 packets which use header options.
So only use the hardware checksum when appropriate.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Jon Kowal <jon.kowal@dspecialists.de>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: invalid data cache only once for each new rx skb buffer
Sonic Zhang [Mon, 10 May 2010 05:39:08 +0000 (05:39 +0000)]
netdev: bfin_mac: invalid data cache only once for each new rx skb buffer

The skb buffer isn't actually used until we finish transferring and pass
it up to higher layers, so only invalidate the range once before we start
receiving actual data.  This also avoids the problem with data invalidating
on Blackfin systems -- there is no invalidate-only, just invalidate+flush.
So when running in writeback mode, there is the small (but not uncommon)
possibility of the flush overwriting valid DMA-ed data from the cache.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: handler RX status errors
Peter Meerwald [Tue, 18 May 2010 00:20:50 +0000 (17:20 -0700)]
netdev: bfin_mac: handler RX status errors

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetdev: bfin_mac: add support for IEEE 1588 PTP
Barry Song [Tue, 18 May 2010 00:19:40 +0000 (17:19 -0700)]
netdev: bfin_mac: add support for IEEE 1588 PTP

Newer on-chip MAC peripherals support IEEE 1588 PTP in the hardware, so
extend the driver to support this functionality.

Signed-off-by: Barry Song <barry.song@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: No dst refcounting in ip_queue_xmit()
Eric Dumazet [Mon, 10 May 2010 11:31:49 +0000 (11:31 +0000)]
net: No dst refcounting in ip_queue_xmit()

TCP outgoing packets can avoid two atomic ops, and dirtying
of previously higly contended cache line using new refdst
infrastructure.

Note 1: loopback device excluded because of !IFF_XMIT_DST_RELEASE
Note 2: UDP packets dsts are built before ip_queue_xmit().

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: Use ip_route_input_noref() in input path
Eric Dumazet [Mon, 10 May 2010 11:33:06 +0000 (11:33 +0000)]
net: Use ip_route_input_noref() in input path

Use ip_route_input_noref() in ip fast path, to avoid two atomic ops per
incoming packet.

Note: loopback is excluded from this optimization in ip_rcv_finish()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: implements ip_route_input_noref()
Eric Dumazet [Mon, 10 May 2010 11:32:55 +0000 (11:32 +0000)]
net: implements ip_route_input_noref()

ip_route_input() is the version returning a refcounted dst, while
ip_route_input_noref() returns a non refcounted one.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: add a noref bit on skb dst
Eric Dumazet [Tue, 11 May 2010 23:19:48 +0000 (23:19 +0000)]
net: add a noref bit on skb dst

Use low order bit of skb->_skb_dst to tell dst is not refcounted.

Change _skb_dst to _skb_refdst to make sure all uses are catched.

skb_dst() returns the dst, regardless of noref bit set or not, but
with a lockdep check to make sure a noref dst is not given if current
user is not rcu protected.

New skb_dst_set_noref() helper to set an notrefcounted dst on a skb.
(with lockdep check)

skb_dst_drop() drops a reference only if skb dst was refcounted.

skb_dst_force() helper is used to force a refcount on dst, when skb
is queued and not anymore RCU protected.

Use skb_dst_force() in __sk_add_backlog(), __dev_xmit_skb() if
!IFF_XMIT_DST_RELEASE or skb enqueued on qdisc queue, in
sock_queue_rcv_skb(), in __nf_queue().

Use skb_dst_force() in dev_requeue_skb().

Note: dst_use_noref() still dirties dst, we might transform it
later to do one dirtying per jiffies.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agorps: avoid one atomic in enqueue_to_backlog
Eric Dumazet [Thu, 6 May 2010 23:51:21 +0000 (23:51 +0000)]
rps: avoid one atomic in enqueue_to_backlog

If CONFIG_SMP=y, then we own a queue spinlock, we can avoid the atomic
test_and_set_bit() from napi_schedule_prep().

We now have same number of atomic ops per netif_rx() calls than with
pre-RPS kernel.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodrivers/net/usb/asix.c: Fix unaligned accesses
Neil Jones [Tue, 18 May 2010 00:18:28 +0000 (17:18 -0700)]
drivers/net/usb/asix.c: Fix unaligned accesses

Using this driver can cause unaligned accesses in the IP layer
This has been fixed by aligning the skb data correctly using the
spare room left over by the 4 byte header inserted between packets
by the device.

Signed-off-by: Neil Jones <NeilJay@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoibmveth: Add suspend/resume support
Brian King [Fri, 7 May 2010 08:56:08 +0000 (08:56 +0000)]
ibmveth: Add suspend/resume support

Adds support for resuming from suspend for IBM virtual ethernet devices.
We may have lost an interrupt over the suspend, so we just kick the
interrupt handler to process anything that is outstanding.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoipv6 addrlabel: permit deletion of labels assigned to removed dev
Florian Westphal [Fri, 7 May 2010 11:31:33 +0000 (11:31 +0000)]
ipv6 addrlabel: permit deletion of labels assigned to removed dev

as addrlabels with an interface index are left alone when the
interface gets removed this results in addrlabels that can no
longer be removed.

Restrict validation of index to adding new addrlabels.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
David S. Miller [Mon, 17 May 2010 05:26:58 +0000 (22:26 -0700)]
Merge branch 'master' of /linux/kernel/git/davem/net-2.6

Conflicts:
include/linux/if_link.h

14 years agortnetlink: make SR-IOV VF interface symmetric
Chris Wright [Sun, 16 May 2010 08:05:45 +0000 (01:05 -0700)]
rtnetlink: make SR-IOV VF interface symmetric

Now we have a set of nested attributes:

  IFLA_VFINFO_LIST (NESTED)
    IFLA_VF_INFO (NESTED)
      IFLA_VF_MAC
      IFLA_VF_VLAN
      IFLA_VF_TX_RATE

This allows a single set to operate on multiple attributes if desired.
Among other things, it means a dump can be replayed to set state.

The current interface has yet to be released, so this seems like
something to consider for 2.6.34.

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqeth: synchronize configuration interface
Frank Blaschka [Tue, 11 May 2010 19:34:47 +0000 (19:34 +0000)]
qeth: synchronize configuration interface

Synchronize access to the drivers configuration interface.
Also do not allow configuration changes during online/offline
transition.

Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqeth: new message if OLM limit is reached
Ursula Braun [Tue, 11 May 2010 19:34:46 +0000 (19:34 +0000)]
qeth: new message if OLM limit is reached

z/OS may activate Optimized Latency Mode (OLM) for a connection
through an OSA Express3 adapter, which reduces the number of
allowed concurrent connections, if adapter is used in shared mode.
Create a meaningful message, if activation of an OSA-connection fails
due to an active OLM-connection on the shared OSA-adapter.

Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqeth: exploit HW TX checksumming
Frank Blaschka [Tue, 11 May 2010 19:34:45 +0000 (19:34 +0000)]
qeth: exploit HW TX checksumming

OSA supports HW TX checksumming in layer 3 mode. Enable this
feature and remove software fallback used for TSO. Cleanup
checksum bits to indicate OSA can do checksumming only for
IPv4 TCP and UDP.

Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosctp: delete active ICMP proto unreachable timer when free transport
Wei Yongjun [Sun, 9 May 2010 16:56:07 +0000 (16:56 +0000)]
sctp: delete active ICMP proto unreachable timer when free transport

transport may be free before ICMP proto unreachable timer expire, so
we should delete active ICMP proto unreachable timer when transport
is going away.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Acked-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: congestion notifications are not dropped packets
Eric Dumazet [Mon, 10 May 2010 04:51:02 +0000 (04:51 +0000)]
net: congestion notifications are not dropped packets

vlan/macvlan start_xmit() can inform caller of congestion with
NET_XMIT_CN return value. This doesnt mean packet was dropped.
Increment normal stat counters instead of tx_dropped.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: Introduce sk_route_nocaps
Eric Dumazet [Sun, 16 May 2010 07:36:33 +0000 (00:36 -0700)]
net: Introduce sk_route_nocaps

TCP-MD5 sessions have intermittent failures, when route cache is
invalidated. ip_queue_xmit() has to find a new route, calls
sk_setup_caps(sk, &rt->u.dst), destroying the

sk->sk_route_caps &= ~NETIF_F_GSO_MASK

that MD5 desperately try to make all over its way (from
tcp_transmit_skb() for example)

So we send few bad packets, and everything is fine when
tcp_transmit_skb() is called again for this socket.

Since ip_queue_xmit() is at a lower level than TCP-MD5, I chose to use a
socket field, sk_route_nocaps, containing bits to mask on sk_route_caps.

Reported-by: Bhaskar Dutta <bhaskie@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agotcp: fix MD5 (RFC2385) support
Eric Dumazet [Sun, 16 May 2010 07:34:04 +0000 (00:34 -0700)]
tcp: fix MD5 (RFC2385) support

TCP MD5 support uses percpu data for temporary storage. It currently
disables preemption so that same storage cannot be reclaimed by another
thread on same cpu.

We also have to make sure a softirq handler wont try to use also same
context. Various bug reports demonstrated corruptions.

Fix is to disable preemption and BH.

Reported-by: Bhaskar Dutta <bhaskie@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: Consistent skb timestamping
Eric Dumazet [Sun, 16 May 2010 06:57:10 +0000 (23:57 -0700)]
net: Consistent skb timestamping

With RPS inclusion, skb timestamping is not consistent in RX path.

If netif_receive_skb() is used, its deferred after RPS dispatch.

If netif_rx() is used, its done before RPS dispatch.

This can give strange tcpdump timestamps results.

I think timestamping should be done as soon as possible in the receive
path, to get meaningful values (ie timestamps taken at the time packet
was delivered by NIC driver to our stack), even if NAPI already can
defer timestamping a bit (RPS can help to reduce the gap)

Tom Herbert prefer to sample timestamps after RPS dispatch. In case
sampling is expensive (HPET/acpi_pm on x86), this makes sense.

Let admins switch from one mode to another, using a new
sysctl, /proc/sys/net/core/netdev_tstamp_prequeue

Its default value (1), means timestamps are taken as soon as possible,
before backlog queueing, giving accurate timestamps.

Setting a 0 value permits to sample timestamps when processing backlog,
after RPS dispatch, to lower the load of the pre-RPS cpu.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoxfrm: fix policy unreferencing on larval drop
Timo Teras [Sun, 16 May 2010 06:49:26 +0000 (23:49 -0700)]
xfrm: fix policy unreferencing on larval drop

I mistakenly had the error path to use num_pols to decide how
many policies we need to drop (cruft from earlier patch set
version which did not handle socket policies right).

This is wrong since normally we do not keep explicit references
(instead we hold reference to the cache entry which holds references
to policies). drop_pols is set to num_pols if we are holding the
references, so use that. Otherwise we eventually BUG_ON inside
xfrm_policy_destroy due to premature policy deletion.

Signed-off-by: Timo Teras <timo.teras@iki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: adjust handle_macvlan to pass port struct to hook
Jiri Pirko [Thu, 6 May 2010 01:33:53 +0000 (01:33 +0000)]
net: adjust handle_macvlan to pass port struct to hook

Now there's null check here and also again in the hook. Looking at bridge bits
which are simmilar, port structure is rcu_dereferenced right away in
handle_bridge and passed to hook. Looks nicer.

Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoskge: use the DMA state API instead of the pci equivalents
FUJITA Tomonori [Tue, 27 Apr 2010 14:57:04 +0000 (14:57 +0000)]
skge: use the DMA state API instead of the pci equivalents

This replace the PCI DMA state API (include/linux/pci-dma.h) with the
DMA equivalents since the PCI DMA state API will be obsolete.

No functional change.

For further information about the background:

http://marc.info/?l=linux-netdev&m=127037540020276&w=2

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: reserve ports for applications using fixed port numbers
Amerigo Wang [Wed, 5 May 2010 00:27:06 +0000 (00:27 +0000)]
net: reserve ports for applications using fixed port numbers

(Dropped the infiniband part, because Tetsuo modified the related code,
I will send a separate patch for it once this is accepted.)

This patch introduces /proc/sys/net/ipv4/ip_local_reserved_ports which
allows users to reserve ports for third-party applications.

The reserved ports will not be used by automatic port assignments
(e.g. when calling connect() or bind() with port number 0). Explicit
port allocation behavior is unchanged.

Signed-off-by: Octavian Purdila <opurdila@ixiacom.com>
Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosysctl: add proc_do_large_bitmap
Octavian Purdila [Wed, 5 May 2010 00:26:55 +0000 (00:26 +0000)]
sysctl: add proc_do_large_bitmap

The new function can be used to read/write large bitmaps via /proc. A
comma separated range format is used for compact output and input
(e.g. 1,3-4,10-10).

Writing into the file will first reset the bitmap then update it
based on the given input.

Signed-off-by: Octavian Purdila <opurdila@ixiacom.com>
Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosysctl: refactor integer handling proc code
Amerigo Wang [Wed, 5 May 2010 00:26:45 +0000 (00:26 +0000)]
sysctl: refactor integer handling proc code

(Based on Octavian's work, and I modified a lot.)

As we are about to add another integer handling proc function a little
bit of cleanup is in order: add a few helper functions to improve code
readability and decrease code duplication.

In the process a bug is also fixed: if the user specifies a number
with more then 20 digits it will be interpreted as two integers
(e.g. 10000...13 will be interpreted as 100.... and 13).

Behavior for EFAULT handling was changed as well. Previous to this
patch, when an EFAULT error occurred in the middle of a write
operation, although some of the elements were set, that was not
acknowledged to the user (by shorting the write and returning the
number of bytes accepted). EFAULT is now treated just like any other
errors by acknowledging the amount of bytes accepted.

Signed-off-by: Octavian Purdila <opurdila@ixiacom.com>
Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/inaky/wimax
David S. Miller [Sun, 16 May 2010 06:14:16 +0000 (23:14 -0700)]
Merge branch 'master' of git://git./linux/kernel/git/inaky/wimax

14 years agobridge: update sysfs link names if port device names have changed
Simon Arlott [Mon, 10 May 2010 09:31:11 +0000 (09:31 +0000)]
bridge: update sysfs link names if port device names have changed

Links for each port are created in sysfs using the device
name, but this could be changed after being added to the
bridge.

As well as being unable to remove interfaces after this
occurs (because userspace tools don't recognise the new
name, and the kernel won't recognise the old name), adding
another interface with the old name to the bridge will
cause an error trying to create the sysfs link.

This fixes the problem by listening for NETDEV_CHANGENAME
notifications and renaming the link.

https://bugzilla.kernel.org/show_bug.cgi?id=12743

Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobridge: change console message interface
stephen hemminger [Mon, 10 May 2010 09:31:09 +0000 (09:31 +0000)]
bridge: change console message interface

Use one set of macro's for all bridge messages.

Note: can't use netdev_XXX macro's because bridge is purely
virtual and has no device parent.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agobridge: netpoll cleanup
stephen hemminger [Mon, 10 May 2010 09:31:08 +0000 (09:31 +0000)]
bridge: netpoll cleanup

Move code around so that the ifdef for NETPOLL_CONTROLLER don't have to
show up in main code path. The control functions should be in helpers
that are only compiled if needed.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agorndis_host: Poll status channel before control channel
Ben Hutchings [Sun, 16 May 2010 06:03:29 +0000 (23:03 -0700)]
rndis_host: Poll status channel before control channel

Some RNDIS devices don't respond on the control channel until polled
on the status channel.  In particular, this was reported to be the
case for the 2Wire HomePortal 1000SW.

This is roughly based on a patch by John Carr <john.carr@unrouted.co.uk>
which is reported to be needed for use with some Windows Mobile devices
and which is currently applied by Mandriva.

Reported-by: Mark Glassberg <vzeeaxwl@myfairpoint.net>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Tested-by: Mark Glassberg <vzeeaxwl@myfairpoint.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetfilter: xt_TEE depends on NF_CONNTRACK
Randy Dunlap [Fri, 14 May 2010 20:52:30 +0000 (13:52 -0700)]
netfilter: xt_TEE depends on NF_CONNTRACK

Fix xt_TEE build for the case of NF_CONNTRACK=m and
NETFILTER_XT_TARGET_TEE=y:

xt_TEE.c:(.text+0x6df5c): undefined reference to `nf_conntrack_untracked'
4x

Built with all 4 m/y combinations.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge branch 'net-2.6' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
David S. Miller [Fri, 14 May 2010 10:42:49 +0000 (03:42 -0700)]
Merge branch 'net-2.6' of git://git./linux/kernel/git/mst/vhost

14 years agoqlcnic: add idc debug registers
Sucheta Chakraborty [Thu, 13 May 2010 03:07:50 +0000 (03:07 +0000)]
qlcnic: add idc debug registers

When ever driver changes the device state, it should write
pci-func number and timestamp in debug registers.

Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: support quisce mode
Sucheta Chakraborty [Thu, 13 May 2010 03:07:49 +0000 (03:07 +0000)]
qlcnic: support quisce mode

Device can go to quiescent state, during which drivers
should refrain from using the device.

Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: check device class
Sucheta Chakraborty [Thu, 13 May 2010 03:07:48 +0000 (03:07 +0000)]
qlcnic: check device class

pci-func class can be other than ethernet in Qlogic CNA device.

Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: check IDC version
Sucheta Chakraborty [Thu, 13 May 2010 03:07:47 +0000 (03:07 +0000)]
qlcnic: check IDC version

Warn user if IDC version mismatch with different class of drivers.

Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: cleanup unused code
Amit Kumar Salecha [Thu, 13 May 2010 03:07:46 +0000 (03:07 +0000)]
qlcnic: cleanup unused code

LRO ring, cut-thru mode and specific fw version are not
valid to Qlogic CNA device.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: cleanup dma mask setting
Amit Kumar Salecha [Fri, 14 May 2010 10:07:46 +0000 (03:07 -0700)]
qlcnic: cleanup dma mask setting

Device support 64 bit dma mask.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: fix caching window register
Amit Kumar Salecha [Thu, 13 May 2010 03:07:44 +0000 (03:07 +0000)]
qlcnic: fix caching window register

o Window register is not per pci-func, so caching can
result in expected result.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: remove obsolete register
Amit Kumar Salecha [Thu, 13 May 2010 03:07:43 +0000 (03:07 +0000)]
qlcnic: remove obsolete register

MSI_MODE, CAPABILITIES_FW and SCRATCHPAD registers are obsolete.
Driver should not use them.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoqlcnic: fix context cleanup
Amit Kumar Salecha [Thu, 13 May 2010 03:07:42 +0000 (03:07 +0000)]
qlcnic: fix context cleanup

Before going for recovery, every pci-func should check fw state,
irrespective of device state. This to avoid unnecssary sending
of command for ctx destroy.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: version 1.28
stephen hemminger [Thu, 13 May 2010 06:12:53 +0000 (06:12 +0000)]
sky2: version 1.28

Version 1.28

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: Avoid allocating memory in sky2_resume
Mike McCormack [Thu, 13 May 2010 06:12:52 +0000 (06:12 +0000)]
sky2: Avoid allocating memory in sky2_resume

Allocating memory can fail, and since we have the memory we need
in sky2_resume when sky2_suspend is called, just stop the hardware
without freeing the memory it's using.

This avoids the possibility of failing because we can't allocate
memory in sky2_resume(), and allows sharing code with sky2_restart().

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: Refactor down/up code out of sky2_restart()
Mike McCormack [Thu, 13 May 2010 06:12:51 +0000 (06:12 +0000)]
sky2: Refactor down/up code out of sky2_restart()

Code to bring down all sky2 interfaces and bring it up
again can be reused in sky2_suspend and sky2_resume.

Factor the code to bring the interfaces down into
sky2_all_down and the up code into sky2_all_up.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: Shut off interrupts before NAPI
Mike McCormack [Thu, 13 May 2010 06:12:50 +0000 (06:12 +0000)]
sky2: Shut off interrupts before NAPI

Interrupts should be masked, then synchronized, and
finally NAPI should be disabled.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: Avoid race in sky2_change_mtu
Mike McCormack [Thu, 13 May 2010 06:12:49 +0000 (06:12 +0000)]
sky2: Avoid race in sky2_change_mtu

netif_stop_queue does not ensure all in-progress transmits are complete,
 so use netif_tx_disable() instead.

Secondly, make sure NAPI polls are disabled before stopping the tx queue,
 otherwise sky2_status_intr might trigger a TX queue wakeup between when
 we stop the queue and NAPI is disabled.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosky2: Restore multicast after restart
Mike McCormack [Thu, 13 May 2010 06:12:48 +0000 (06:12 +0000)]
sky2: Restore multicast after restart

Multicast settings will be lost on reset, so restore them.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgb and e1000: Use new function for copybreak tests
Joe Perches [Thu, 13 May 2010 15:26:17 +0000 (15:26 +0000)]
ixgb and e1000: Use new function for copybreak tests

There appears to be an off-by-1 defect in the maximum packet size
copied when copybreak is speified in these modules.

The copybreak module params are specified as:
"Maximum size of packet that is copied to a new buffer on receive"

The tests are changed from "< copybreak" to "<= copybreak"
and moved into new static functions for readability.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000: cleanup unused parameters
Jesse Brandeburg [Thu, 13 May 2010 15:25:56 +0000 (15:25 +0000)]
e1000: cleanup unused parameters

During the cleanup pass after the removal of e1000e hardware from e1000 some
parameters were missed.  Remove them because it is just dead code.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000: fix WARN_ON with mac-vlan
Jesse Brandeburg [Thu, 13 May 2010 15:25:33 +0000 (15:25 +0000)]
e1000: fix WARN_ON with mac-vlan

When adding more than 14 mac-vlan adapters on e1000 the driver
would fire a WARN_ON when adding the 15th.  The WARN_ON in this
case is completely un-necessary, as the code below the WARN_ON is
directly handling the value the WARN_ON triggered on.

CC: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodrivers/net: Remove unnecessary returns from void function()s
Joe Perches [Fri, 14 May 2010 07:19:28 +0000 (00:19 -0700)]
drivers/net: Remove unnecessary returns from void function()s

This patch removes from drivers/net/ all the unnecessary
return; statements that precede the last closing brace of
void functions.

It does not remove the returns that are immediately
preceded by a label as gcc doesn't like that.

It also does not remove null void functions with return.

Done via:
$ grep -rP --include=*.[ch] -l "return;\n}" net/ | \
  xargs perl -i -e 'local $/ ; while (<>) { s/\n[ \t\n]+return;\n}/\n}/g; print; }'

with some cleanups by hand.

Compile tested x86 allmodconfig only.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: Refactor common code between 82598 & 82599 to accommodate new hardware
Mallikarjuna R Chilakala [Thu, 13 May 2010 17:33:41 +0000 (17:33 +0000)]
ixgbe: Refactor common code between 82598 & 82599 to accommodate new hardware

Some of the following MAC functions are moved from 82598 & 82599 specific
hardware files to common.[ch] to accommodate new silicon changes. Also
fixed some white space issues
 * get_san_mac_addr, check_link, set_vmdq, clear_vmdq, clear_vfta,
 * set_vfta, fc_enable, init_uta_tables

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: Use bool flag to see if the packet unmapping is delayed in HWRSC
Mallikarjuna R Chilakala [Thu, 13 May 2010 17:33:21 +0000 (17:33 +0000)]
ixgbe: Use bool flag to see if the packet unmapping is delayed in HWRSC

We can't use zero magic "bad" value to check if IXGBE_RSC_CB(skb)->dma
is valid. It is only valid in x86/arm/m68k/alpha architectures and in
spark, powerPC and other architectures it should be ~0. As per
Benjamin Herrenschmidt feedback use a bool flag to decide if
the packet unmapping is delayed in hardware RSC till EOP is reached

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: fix setting of promisc mode when using mac-vlans
Emil Tantilov [Thu, 13 May 2010 17:33:00 +0000 (17:33 +0000)]
ixgbe: fix setting of promisc mode when using mac-vlans

Introduce uc_set_promisc flag to fix enabling of promisc mode
when exceeding the number of supported RAR entries.

Issue discovered by Ben Greear when using mac-vlans.

Reported-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/kaber/nf-next-2.6
David S. Miller [Thu, 13 May 2010 21:14:10 +0000 (14:14 -0700)]
Merge branch 'master' of git://git./linux/kernel/git/kaber/nf-next-2.6

14 years agonetfilter: remove unnecessary returns from void function()s
Joe Perches [Thu, 13 May 2010 13:16:27 +0000 (15:16 +0200)]
netfilter: remove unnecessary returns from void function()s

This patch removes from net/ netfilter files
all the unnecessary return; statements that precede the
last closing brace of void functions.

It does not remove the returns that are immediately
preceded by a label as gcc doesn't like that.

Done via:
$ grep -rP --include=*.[ch] -l "return;\n}" net/ | \
  xargs perl -i -e 'local $/ ; while (<>) { s/\n[ \t\n]+return;\n}/\n}/g; print; }'

Signed-off-by: Joe Perches <joe@perches.com>
[Patrick: changed to keep return statements in otherwise empty function bodies]
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: cleanup printk messages
Stephen Hemminger [Thu, 13 May 2010 13:02:08 +0000 (15:02 +0200)]
netfilter: cleanup printk messages

Make sure all printk messages have a severity level.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: change NF_ASSERT to WARN_ON
Stephen Hemminger [Thu, 13 May 2010 13:00:20 +0000 (15:00 +0200)]
netfilter: change NF_ASSERT to WARN_ON

Change netfilter asserts to standard WARN_ON. This has the
benefit of backtrace info and also causes netfilter errors
to show up on kerneloops.org.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agonetfilter: bridge-netfilter: fix crash in br_nf_forward_finish()
Bart De Schuymer [Thu, 13 May 2010 12:55:34 +0000 (14:55 +0200)]
netfilter: bridge-netfilter: fix crash in br_nf_forward_finish()

[ 4593.956206] BUG: unable to handle kernel NULL pointer dereference at 0000000000000018
[ 4593.956219] IP: [<ffffffffa03357a4>] br_nf_forward_finish+0x154/0x170 [bridge]
[ 4593.956232] PGD 195ece067 PUD 1ba005067 PMD 0
[ 4593.956241] Oops: 0000 [#1] SMP
[ 4593.956248] last sysfs file:
/sys/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/device:08/ATK0110:00/hwmon/hwmon0/temp2_label
[ 4593.956253] CPU 3
...
[ 4593.956380] Pid: 29512, comm: kvm Not tainted 2.6.34-rc7-net #195 P6T DELUXE/System Product Name
[ 4593.956384] RIP: 0010:[<ffffffffa03357a4>]  [<ffffffffa03357a4>] br_nf_forward_finish+0x154/0x170 [bridge]
[ 4593.956395] RSP: 0018:ffff880001e63b78  EFLAGS: 00010246
[ 4593.956399] RAX: 0000000000000608 RBX: ffff880057181700 RCX: ffff8801b813d000
[ 4593.956402] RDX: 0000000000000000 RSI: 0000000000000002 RDI: ffff880057181700
[ 4593.956406] RBP: ffff880001e63ba8 R08: ffff8801b9d97000 R09: ffffffffa0335650
[ 4593.956410] R10: 0000000000000000 R11: 0000000000000000 R12: ffff8801b813d000
[ 4593.956413] R13: ffffffff81ab3940 R14: ffff880057181700 R15: 0000000000000002
[ 4593.956418] FS:  00007fc40d380710(0000) GS:ffff880001e60000(0000) knlGS:0000000000000000
[ 4593.956422] CS:  0010 DS: 002b ES: 002b CR0: 000000008005003b
[ 4593.956426] CR2: 0000000000000018 CR3: 00000001ba1d7000 CR4: 00000000000026e0
[ 4593.956429] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 4593.956433] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 4593.956437] Process kvm (pid: 29512, threadinfo ffff8801ba566000, task ffff8801b8003870)
[ 4593.956441] Stack:
[ 4593.956443]  0000000100000020 ffff880001e63ba0 ffff880001e63ba0 ffff880057181700
[ 4593.956451] <0> ffffffffa0335650 ffffffff81ab3940 ffff880001e63bd8 ffffffffa03350e6
[ 4593.956462] <0> ffff880001e63c40 000000000000024d ffff880057181700 0000000080000000
[ 4593.956474] Call Trace:
[ 4593.956478]  <IRQ>
[ 4593.956488]  [<ffffffffa0335650>] ? br_nf_forward_finish+0x0/0x170 [bridge]
[ 4593.956496]  [<ffffffffa03350e6>] NF_HOOK_THRESH+0x56/0x60 [bridge]
[ 4593.956504]  [<ffffffffa0335282>] br_nf_forward_arp+0x112/0x120 [bridge]
[ 4593.956511]  [<ffffffff813f7184>] nf_iterate+0x64/0xa0
[ 4593.956519]  [<ffffffffa032f920>] ? br_forward_finish+0x0/0x60 [bridge]
[ 4593.956524]  [<ffffffff813f722c>] nf_hook_slow+0x6c/0x100
[ 4593.956531]  [<ffffffffa032f920>] ? br_forward_finish+0x0/0x60 [bridge]
[ 4593.956538]  [<ffffffffa032f800>] ? __br_forward+0x0/0xc0 [bridge]
[ 4593.956545]  [<ffffffffa032f86d>] __br_forward+0x6d/0xc0 [bridge]
[ 4593.956550]  [<ffffffff813c5d8e>] ? skb_clone+0x3e/0x70
[ 4593.956557]  [<ffffffffa032f462>] deliver_clone+0x32/0x60 [bridge]
[ 4593.956564]  [<ffffffffa032f6b6>] br_flood+0xa6/0xe0 [bridge]
[ 4593.956571]  [<ffffffffa032f800>] ? __br_forward+0x0/0xc0 [bridge]

Don't call nf_bridge_update_protocol() for ARP traffic as skb->nf_bridge isn't
used in the ARP case.

Reported-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Bart De Schuymer <bdschuym@pandora.be>
Signed-off-by: Patrick McHardy <kaber@trash.net>
14 years agocxgb4: report GRO stats with ethtool -S
Dimitris Michailidis [Mon, 10 May 2010 15:58:09 +0000 (15:58 +0000)]
cxgb4: report GRO stats with ethtool -S

Signed-off-by: Dimitris Michailidis <dm@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocxgb4: report the PCIe link speed
Dimitris Michailidis [Mon, 10 May 2010 15:58:08 +0000 (15:58 +0000)]
cxgb4: report the PCIe link speed

Report the PCIe link speed (2.5 or 5 Gbps).

Signed-off-by: Dimitris Michailidis <dm@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocxgb4: configure HW VLAN extraction through FW
Dimitris Michailidis [Mon, 10 May 2010 15:58:07 +0000 (15:58 +0000)]
cxgb4: configure HW VLAN extraction through FW

HW VLAN extraction needs to be configured through FW to work correctly in
virtualization environments.  Remove the direct register manipulation and
rely on FW.

Signed-off-by: Dimitris Michailidis <dm@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: add PCI device id to enable support for 82567V-4
Bruce Allan [Mon, 10 May 2010 15:02:52 +0000 (15:02 +0000)]
e1000e: add PCI device id to enable support for 82567V-4

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: Fix/cleanup PHY reset code for ICHx/PCHx
Bruce Allan [Mon, 10 May 2010 15:02:32 +0000 (15:02 +0000)]
e1000e: Fix/cleanup PHY reset code for ICHx/PCHx

i) Fixes a bug where e1000_sw_lcd_config_ich8lan() was calling
e1000_lan_init_done_ich8lan() to poll the STATUS.LAN_INIT_DONE bit to
make sure the MAC had completed the PHY configuration.  However,
e1000_lan_init_done_ich8lan() had already been called in one of the two
places where PHY reset occurs for ICHx/PCHx parts, which caused the second
call to busy-wait for 150 msec because the LAN_INIT_DONE bit had already
been checked and cleared.

ii) Cleanup the two separate PHY reset code paths, i.e. the full-chip reset
in e1000_reset_hw_ich8lan() and the PHY-only reset in
e1000_phy_hw_reset_ich8lan().  There was duplicate code in both paths to be
performed post-reset that are now combined into one new function -
e1000_post_phy_reset_ich8lan().  This cleanup also included moving the
clearing of the PHY Reset Asserted bit in the STATUS register (now done for
all ICH/PCH parts) and the check for the indication from h/w that basic
configuration has completed back to where it previously was in
e1000_get_cfg_done_ich8lan().

iii) Corrected a few comments

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: move settting of flow control refresh timer to link setup code
Bruce Allan [Mon, 10 May 2010 15:02:12 +0000 (15:02 +0000)]
e1000e: move settting of flow control refresh timer to link setup code

The flow control refresh timer value needs to be saved off so that it can
be programmed into the approrpiate register when applicable but without a
reset, e.g. when changing flow control parameters via ethtool.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: fix checks for manageability enabled and management pass-through
Bruce Allan [Mon, 10 May 2010 15:01:51 +0000 (15:01 +0000)]
e1000e: fix checks for manageability enabled and management pass-through

The mac->arc_subsystem was being incorrectly used to flag whether or not
manageability was enabled when it should only be used to state whether the
ARC (Host interface) subsystem is available on a particular MAC _and_ only
valid when any manageability is enabled. The ARC subsystem is currently
only available on 80003es2lan and 82573 parts supported by the driver.

A new flag, has_fwsm, is introduced to be used when checking if
manageability is enabled but only on parts that acutally have an FWSM
register. While the above parts have an FWSM register, there are other
parts that have FWSM but do not have support for the ARC subsystem,
namely 82571/2 and ICHx/PCH.

And then there are parts that have manageability, but do not have either
FWSM register or support for the ARC subsystem - these are 82574 and 82583.

For 80003es2lan, 82571/2/3 and ICH/PCH parts, this patch makes no
functional changes, it only corrects the usage of the manageability flags.
For 82574 and 82583, it fixes the incorrect accesses of the non-existent
FWSM register and ARC subsystem as well as corrects the check for
management pass-through.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: Incorrect function pointer set for force_speed_duplex on 82577
Bruce Allan [Mon, 10 May 2010 15:01:30 +0000 (15:01 +0000)]
e1000e: Incorrect function pointer set for force_speed_duplex on 82577

The force_speed_duplex function pointer was incorrectly set.  Instead of
calling the 82577-specific version it was calling the m88 version which,
among other incorrect things, reset the PHY causing autonegotiation to be
re-enabled in the PHY resulting in the link defaulting to half-duplex.
The 82577-specific force_speed_duplex function also had an issue where
it disabled Auto-MDI-X which caused the link to not come up.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: Cleanup e1000_sw_lcd_config_ich8lan()
Bruce Allan [Mon, 10 May 2010 15:01:10 +0000 (15:01 +0000)]
e1000e: Cleanup e1000_sw_lcd_config_ich8lan()

After every reset all ICH/PCH parts call this function which acquires the
swflag, performs a workaround on applicable parts and releases the swflag.
There is no reason for parts for which this workaround is not applicable
to acquire and release the swflag so the function should just return
without doing anything for these parts.  This also provides for the
indentation of most of the function contents to be shifted left cleaning up
the code.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: Remove EN_MAC_ADDR_FILTER check from enable_mng_pass_thru check
Bruce Allan [Mon, 10 May 2010 15:00:50 +0000 (15:00 +0000)]
e1000e: Remove EN_MAC_ADDR_FILTER check from enable_mng_pass_thru check

Patch addresses issues when manageability passthrough is enabled, but the
MAC_ADDR_FILTER bit is not set in the MANC register.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: cleanup multiple common exit points
Bruce Allan [Mon, 10 May 2010 15:00:31 +0000 (15:00 +0000)]
e1000e: cleanup multiple common exit points

...in e1000_update_nvm_checksum_ich8lan().

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: s/w initiated LSC MSI-X interrupts not generated; no transmit
Bruce Allan [Mon, 10 May 2010 15:00:10 +0000 (15:00 +0000)]
e1000e: s/w initiated LSC MSI-X interrupts not generated; no transmit

In MSI-X mode when an IMPI SoL session was active (i.e. the PHY reset was
blocked), the LSC interrupt generated by s/w to start the watchdog which
started the transmitter was not getting fired by the hardware because bit
24 (the 'other' cause bit) also needed to be set.  Without an active SoL
session, the PHY was reset which caused the h/w to fire the LSC interrupt.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: initialize manageability (IPMI) pass-through in 82574/82583
Bruce Allan [Mon, 10 May 2010 14:59:51 +0000 (14:59 +0000)]
e1000e: initialize manageability (IPMI) pass-through in 82574/82583

82574/82583 uses different registers/bits to setup manageability filters
than all other parts supported by e1000e; set them accordingly for IPMI
pass-through.  Rename the function to better reflect what it does.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: bad state after running ethtool diagnostics with AMT enabled
Bruce Allan [Mon, 10 May 2010 14:59:31 +0000 (14:59 +0000)]
e1000e: bad state after running ethtool diagnostics with AMT enabled

When running ethtool online diagnostics with no open interface, there is a
short period of time where the driver relinquishes control of the adapter
during which time AMT (manageability firmware) can put the adapter into an
unknown state resulting in such things as link test failure, hardware hang,
reporting an incorrect link speed, etc.  Resetting the adapter during an
open() resolves this by putting the adapter into a quiescent state.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: use static params to save stack space (part 2)
Bruce Allan [Mon, 10 May 2010 14:59:10 +0000 (14:59 +0000)]
e1000e: use static params to save stack space (part 2)

A couple stack cleanups missed in an earlier patch from Jesse.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/kaber/ipmr-2.6
David S. Miller [Thu, 13 May 2010 06:30:45 +0000 (23:30 -0700)]
Merge branch 'master' of git://git./linux/kernel/git/kaber/ipmr-2.6