gro: Do not merge paged packets into frag_list

Bigger is not always better :)

It was easy to continue to merged packets into frag_list after the
page array is full.  However, this turns out to be worse than LRO
because frag_list is a much less efficient form of storage than the
page array.  So we're better off stopping the merge and starting
a new entry with an empty page array.

In future we can optimise this further by doing frag_list merging
but making sure that we continue to fill in the page array.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

Unfortunately simplicity isn't always the best.  The fraginfo
interface turned out to be suboptimal.  The problem was quite
obvious.  For every packet, we have to copy the headers from
the frags structure into skb->head, even though for 99% of the
packets this part is immediately thrown away after the merge.

LRO didn't have this problem because it directly read the headers
from the frags structure.

This patch attempts to address this by creating an interface
that allows GRO to access the headers in the first frag without
having to copy it.  Because all drivers that use frags place the
headers in the first frag this optimisation should be enough.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently VLAN still has a bit of common code handling the aftermath
of GRO that's shared with the common path.  This patch moves them
into shared helpers to reduce code duplication.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

We now only TX hash on pre-computed SKB properties.

The thinking is:

1) High performance routing and firewalling setups will
   have a multiqueue capable card used for receive, and
   therefore would have RX queue recordings made into
   the SKB which can be used for the TX side hash.

2) Locally generated packets will have an attached socket
   and thus a valid sk->sk_hash to make use of.
Signed-off-by: David S. Miller <davem@davemloft.net>

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

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

The idea is that drivers which implement multiqueue RX
pre-seed the SKB by recording the RX queue selected by
the hardware.

If such a seed is found on TX, we'll use that to select
the outgoing TX queue.

This helps get more consistent load balancing on router
and firewall loads.
Signed-off-by: David S. Miller <davem@davemloft.net>

Freeze activity when notified that the underlying chip
is getting reset on a EEH event or fatal error.
Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Set up a notification mechanism to inform upper layer modules
(iWARP, iSCSI) of a chip reset due to an EEH event or a fatal error.
Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

smsc911x_set_multicast_list currently performs the only non-atomic
read-modify-write of INT_EN.  This patch permanently enables the
RXSTOP_INT interrupt, and changes the ISR to only conditionally run the
multicast filter workaround code.
Signed-off-by: Steve Glendinning <steve.glendinning@smsc.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Convert the smc91x driver to use struct net_device_ops.
Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

Alexey Dobriyan pointed that using PDE_NET outside the
proc code is plain bogus (thanks Alexey!). Fix it.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

PDE_NET usage in driver code is a sign and, indeed, switching
to seq_open_net/seq_release_net saves code and fixes bogus things, like
user triggerabble BUG_ON(!net) after maybe_get_net, and NULLifying ->private.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Incoming packets and sockets are already gone.
The netdevice notifier is unregistered under the RTNL lock
There remains a race with the rtnetlink handlers unregistration, but it
is a generic RTNL issue that was already present before this change.
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

And support for the KX/KX4 mezzanine card.  Device id 0x10B6.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Load times for NICs that use i2c to communicate with the phy were taking
~4.5 sec per port.  This fix first checks to see if the link is already
up before calling get_link_capabilities, since if it is we don't need
query the phy for link state.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of keeping candidate tunnel device from all categories,
keep only one candidate with best score. This optimizes stack
usage and speeds up exit code.
Signed-off-by: Timo Teras <timo.teras@iki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>

* master.kernel.org:/home/rmk/linux-2.6-arm: (22 commits)
  [ARM] fix section-based ioremap
  [NET] am79c961a: fix spin_lock usage
  [ARM] omap: usb: thou shalt not provide empty release functions
  [ARM] omap: watchdog: allow OMAP watchdog driver on OMAP34xx platforms
  [ARM] 5369/1: omap mmc: Add new omap hsmmc controller for 2430 and 34xx, v3
  [ARM] clkdev: fix clock matching
  [ARM] 5370/1: at91: fix rm9200 watchdog
  [ARM] 5368/1: arch/arm/mach-davinci/usb.c buildfix
  [ARM] 5365/1: s3cmci: Use new include path of dma.h
  [ARM] fix StrongARM-11x0 page copy implementation
  [ARM] omap: ensure OMAP drivers pass a struct device to clk_get()
  ARM: OMAP: Fix compile for h3 MMC
  ARM: OMAP: Remove unused platform devices, v3
  ARM: OMAP: Fix ASoC by enabling writes to XCCR and RCCR McBSP registers, v3
  ARM: OMAP: Fix OSK ASoC by registering I2C board info for tlvaic23
  ARM: OMAP: remove duplicated #include's
  ARM: OMAP: Fix DMA CCR programming for request line > 63, v3
  ARM: OMAP: Fix gpio.c compile on 15xx with CONFIG_DEBUGFS
  ARM: OMAP: Fix compile for beagle
  ARM: OMAP: Fix gpio by switching to generic gpio calls, v2
  ...

* 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6:
  i2c: Warn on deprecated binding model use
  eeprom: More consistent symbol names
  eeprom: Move 93cx6 eeprom driver to /drivers/misc/eeprom
  spi: Move at25 (for SPI eeproms) to /drivers/misc/eeprom
  i2c: Move old eeprom driver to /drivers/misc/eeprom
  i2c: Move at24 to drivers/misc/eeprom
  i2c: Quilt tree has moved
  i2c: Delete many unused adapter IDs
  i2c: Delete 10 unused driver IDs

* git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild-fixes:
  kbuild: fix kbuild.txt typos
  kbuild: print usage with no arguments in scripts/config
  Revert "kbuild: strip generated symbols from *.ko"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (92 commits)
  gianfar: Revive VLAN support
  vlan: Export symbols as non GPL symbols.
  bnx2x: tx_has_work should not wait for FW
  netxen: reduce memory footprint
  netxen: fix vlan tso/checksum offload
  net: Fix linux/if_frad.h's suitability for userspace.
  net: Move config NET_NS to from net/Kconfig to init/Kconfig
  isdn: Fix missing ifdef in isdn_ppp
  networking: document "nc" in addition to "netcat" in netconsole.txt
  e1000e: workaround hw errata
  af_key: initialize xfrm encap_oa
  virtio_net: Fix MAX_PACKET_LEN to support 802.1Q VLANs
  lcs: fix compilation for !CONFIG_IP_MULTICAST
  rtl8187: Add termination packet to prevent stall
  iwlwifi: fix rs_get_rate WARN_ON()
  p54usb: fix packet loss with first generation devices
  sctp: Fix another socket race during accept/peeloff
  sctp: Properly timestamp outgoing data chunks for rtx purposes
  sctp: Correctly start rtx timer on new packet transmissions.
  sctp: Fix crc32c calculations on big-endian arhes.
  ...

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  sparc64: Fix DAX handling via userspace access from kernel.
  sparc64: Annotate sparc64 specific syscalls with SYSCALL_DEFINEx()
  [CVE-2009-0029] sparc: Enable syscall wrappers for 64-bit
  sparc64: Initialize FHC/CLOCK LED platform_device 'id' field correctly.
  sparc64: fix modpost failure
  sparc64: fix readout of cpu/fpu type

commit 77ecaf2d ("gianfar: Fix VLAN
HW feature related frame/buffer size calculation") wrongly removed
priv->vlgrp assignment, and now priv->vlgrp is always NULL.

This patch fixes the issue, plus fixes following sparse warning
introduced by the same commit:
gianfar.c:1406:13: warning: context imbalance in 'gfar_vlan_rx_register' - wrong count at exit

gfar_vlan_rx_register() checks for "if (old_grp == grp)" and tries
to return w/o dropping the lock.

According to net/8021q/vlan.c VLAN core issues rx_register() callback:
1. In register_vlan_dev() only on a newly created group;
2. In unregister_vlan_dev() only if the group becomes empty.

Thus the check in the gianfar driver isn't needed.
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

In previous kernels, any kernel module could get access to the
'real-device' and the VLAN-ID for a particular VLAN.  In more recent
kernels, the code was restructured such that this is hard to do
without accessing private .h files for any module that cannot use
GPL-only symbols.

Attached is a patch to once again allow non-GPL modules the ability to
access the real-device and VLAN id for VLANs.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

The current tx_has_work waited until all packets sent by the driver
are marked as completed by the FW. This is too greedy and it causes
the bnx2x_poll to spin in vain. The driver should only check that all
packets FW already completed are freed - only in unload flow the
driver should make sure that transmit queue is empty
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

o reduce rx ring size from 8192 to 4096.
o cut down old huge lro buffers.
Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

o set netdev->vlan_features appropriately.
o fix tso descriptor initialization for vlan case.
Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The userspace interfaces are protected by CONFIG_* ifdefs
and that of course can't work.

Reported by Jaswinder Singh Rajput.
Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>

Make NET_NS available underneath the generic Namespaces config option
since all of the other namespace options are there.
Signed-off-by: Matt Helsley <matthltc@us.ibm.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The following patch fixes a warning caused by a missing ifdef in
isdn_ppp.c. A function was defined, but never used if CONFIG_IPPP_FILTER
was not defined.

The warning was: 'get_filter' defined but not used
Patch is against 2.6.28.1
Signed-off-by: Daniele Venzano <venza@brownhat.org>
Acked-by: Karsten Keil <kkeil@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

Let the kernel developers know that i2c_attach_client() and
i2c_detach_client() are deprecated and should no longer be used.
Drivers using these should be converted to the standard device
driver binding model (probe and remove methods.)
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Ben Dooks <ben-linux@fluff.org>

Now that all EEPROM drivers live in the same place, let's harmonize
their symbol names.

Also fix eeprom's dependencies, it definitely needs sysfs, and is no
longer experimental after many years in the kernel tree.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Wolfram Sang <w.sang@pengutronix.de>
Cc: David Brownell <dbrownell@users.sourceforge.net>

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Jean Delvare <khali@linux-fr.org>