[TCP]: Remove dead code in init_sequence

This removes two redundancies:

1) The test (skb->protocol == htons(ETH_P_IPV6) in tcp_v6_init_sequence()
   is always true, due to
* tcp_v6_conn_request() is the only function calling this one
* tcp_v6_conn_request() redirects all skb's with ETH_P_IP protocol to
  tcp_v4_conn_request() [ cf. top of tcp_v6_conn_request()]

2) The first argument, `struct sock *sk' of tcp_v{4,6}_init_sequence() is
   never used.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[DCCP]: Support for partial checksums (RFC 4340, sec. 9.2)

This patch does the following:
  a) introduces variable-length checksums as specified in [RFC 4340, sec. 9.2]
  b) provides necessary socket options and documentation as to how to use them
  c) basic support and infrastructure for the Minimum Checksum Coverage feature
     [RFC 4340, sec. 9.2.1]: acceptability tests, user notification and user
     interface

In addition, it

 (1) fixes two bugs in the DCCPv4 checksum computation:
  * pseudo-header used checksum_len instead of skb->len
* incorrect checksum coverage calculation based on dccph_x
 (2) removes dccp_v4_verify_checksum() since it reduplicates code of the
     checksum computation; code calling this function is updated accordingly.
 (3) now uses skb_checksum(), which is safer than checksum_partial() if the
     sk_buff has is a non-linear buffer (has pages attached to it).
 (4) fixes an outstanding TODO item:
        * If P.CsCov is too large for the packet size, drop packet and return.

The code has been tested with applications, the latest version of tcpdump now
comes with support for partial DCCP checksums.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[IPV6]: Per-interface statistics support.

For IP MIB (RFC4293).

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>

[IPV6]: Introduce ip6_dst_idev() to get inet6_dev{} stored in dst_entry{}.

Otherwise, we will see a lot of casts...

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>

[IPV6] ROUTE: Use &rt->u.dst instead of cast.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>

[IPV6] ROUTE: Use macros to format /proc/net/ipv6_route.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>

[DCCP]: Update code comments for Step 2/3

Sorts out the comments for processing steps 2,3 in section 8.5 of RFC 4340.
All comments have been updated against this document, and the reference to step
2 has been made consistent throughout the files.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCP]: tidy up dccp_v{4,6}_conn_request

This is a code simplification to remove reduplicated code
by concentrating and abstracting shared code.

Detailed Changes:

[DCCP]: Fix logfile overflow

This patch fixes data being spewed into the logs continually. As the
code stood if there was a large queue and long delays timeo would go
down to zero and never get reset.

This fixes it by resetting timeo. Put constant into header as well.

Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCP]: Fix DCCP Probe Typo

Fixes a typo in Kconfig, patch is by Ian McDonald and is re-sent from
http://www.mail-archive.com/dccp@vger.kernel.org/msg00579.html

Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCPv6]: remove forward declarations in ipv6.c

This does the same for ipv6.c as the preceding one does for ipv4.c: Only the
inet_connection_sock_af_ops forward declarations remain, since at least
dccp_ipv6_mapped has a circular dependency to dccp_v6_request_recv_sock.

No code change, merely re-ordering.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCPv4]: remove forward declarations in ipv4.c

This relates to Arnaldo's announcement in
http://www.mail-archive.com/dccp@vger.kernel.org/msg00604.html

Originally this had been part of the Oops fix and is a revised variant of
http://www.mail-archive.com/dccp@vger.kernel.org/msg00598.html

No code change, merely reshuffling, with the particular objective of
having all request_sock_ops close(r) together for more clarity.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCP]: calling dccp_v{4,6}_reqsk_send_ack is a BUG

This patch removes two functions, the send_ack functions of request_sock,
which are not called/used by the DCCP code. It is correct that these
functions are not called, below is a justification why calling these
functions (on a passive socket in the LISTEN/RESPOND state) would mean
a DCCP protocol violation.

A) Background: using request_sock in TCP:

[DCCP] timewait: Remove leftover extern declarations

Gerrit Renker noticed dccp_tw_deschedule and submitted a patch with a FIXME,
but as he suggests in the same patch the best thing is to just ditch this
declaration, while doing that also noticed that tcp_tw_count is as well not
defined anywhere, so ditch it too.

Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCP]: Simplify jump labels in dccp_v{4,6}_rcv

This is a code simplification and was singled out from the
DCCPv6 Oops patch on
http://www.mail-archive.com/dccp@vger.kernel.org/msg00600.html

It mainly makes the code consistent between ipv{4,6}.c for the functions
        dccp_v4_rcv
        dccp_v6_rcv
and removes the do_time_wait label to simplify code somewhat.

Commiter note: fixed up a compile problem, trivial.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCP]: Combine allocating & zeroing header space on skb

This is a code simplification:
it combines three often recurring operations into one inline function,

        * allocate `len' bytes header space in skb
        * fill these `len' bytes with zeroes
        * cast the start of this header space as dccp_hdr

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCPv6]: Add a FIXME for missing IPV6_PKTOPTIONS

This refers to the possible memory leak pointed out in
http://www.mail-archive.com/dccp@vger.kernel.org/msg00574.html,
fixed by David Miller in
http://www.mail-archive.com/netdev@vger.kernel.org/msg24881.html

and adds a FIXME to point out where code is missing.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[DCCP]: set safe upper bound for option length

This is a re-send from
http://www.mail-archive.com/dccp@vger.kernel.org/msg00553.html

It is the same patch as before, but I have built in Arnaldo's suggestions
pointed out in that posting.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[TCP]: Don't set SKB owner in tcp_transmit_skb().

The data itself is already charged to the SKB, doing
the skb_set_owner_w() just generates a lot of noise and
extra atomics we don't really need.

Lmbench improvements on lat_tcp are minimal:

before:
TCP latency using localhost: 23.2701 microseconds
TCP latency using localhost: 23.1994 microseconds
TCP latency using localhost: 23.2257 microseconds

after:
TCP latency using localhost: 22.8380 microseconds
TCP latency using localhost: 22.9465 microseconds
TCP latency using localhost: 22.8462 microseconds

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

[NET] ip-sysctl.txt: Alphabetize.

Rearrange TCP entries in alpha order.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

[TCP]: Allow autoloading of congestion control via setsockopt.

If user has permision to load modules, then autoload then attempt
autoload of TCP congestion module.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

[TCP]: Restrict congestion control choices.

Allow normal users to only choose among a restricted set of congestion
control choices.  The default is reno and what ever has been configured
as default. But the policy can be changed by administrator at any time.

For example, to allow any choice:
    cp /proc/sys/net/ipv4/tcp_available_congestion_control \
       /proc/sys/net/ipv4/tcp_allowed_congestion_control

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

[TCP]: Add tcp_available_congestion_control sysctl.

Create /proc/sys/net/ipv4/tcp_available_congestion_control
that reflects currently available TCP choices.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

[SCTP]: Fix warning

An alternate solution would be to make the digest a pointer, allocate
it in sctp_endpoint_init() and free it in sctp_endpoint_destroy().

I guess I should have originally done it this way...

  CC [M]  net/sctp/sm_make_chunk.o
net/sctp/sm_make_chunk.c: In function 'sctp_unpack_cookie':
net/sctp/sm_make_chunk.c:1358: warning: initialization discards qualifiers from pointer target type

The reason is that sctp_unpack_cookie() takes a const struct
sctp_endpoint and modifies the digest in it (digest being embedded in
the struct, not a pointer).  Make digest a pointer to fix this
warning.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Acked-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6] tcp: Fix typo _read_mostly --> __read_mostly.

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

[DCCP]: Fix typo _read_mostly --> __read_mostly.

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

[NET]: Size listen hash tables using backlog hint

We currently allocate a fixed size (TCP_SYNQ_HSIZE=512) slots hash table for
each LISTEN socket, regardless of various parameters (listen backlog for
example)

On x86_64, this means order-1 allocations (might fail), even for 'small'
sockets, expecting few connections. On the contrary, a huge server wanting a
backlog of 50000 is slowed down a bit because of this fixed limit.

This patch makes the sizing of listen hash table a dynamic parameter,
depending of :
- net.core.somaxconn tunable (default is 128)
- net.ipv4.tcp_max_syn_backlog tunable (default : 256, 1024 or 128)
- backlog value given by user application  (2nd parameter of listen())

For large allocations (bigger than PAGE_SIZE), we use vmalloc() instead of
kmalloc().

We still limit memory allocation with the two existing tunables (somaxconn &
tcp_max_syn_backlog). So for standard setups, this patch actually reduce RAM
usage.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

[PKT_SCHED]: Make sch_fifo.o available when CONFIG_NET_SCHED is not set.

Based on patch by Patrick McHardy.

Add a new option, NET_SCH_FIFO, which provides a simple fifo qdisc
without requiring CONFIG_NET_SCHED.

The d80211 stack needs a generic fifo qdisc for WME.  At present it
uses net/d80211/fifo_qdisc.c which is functionally equivalent to
sch_fifo.c.  This patch will allow the d80211 stack to remove
net/d80211/fifo_qdisc.c and use sch_fifo.c instead.

Signed-off-by: David Kimdon <david.kimdon@devicescape.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NET] rules: Add support to invert selectors

Introduces a new flag FIB_RULE_INVERT causing rules to apply
if the specified selector doesn't match.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NET] rules: Share common attribute validation policy

Move the attribute policy for the non-specific attributes into
net/fib_rules.h and include it in the respective protocols.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NET] rules: Protocol independant mark selector

Move mark selector currently implemented per protocol into
the protocol independant part.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV4] nl_fib_lookup: Rename fl_fwmark to fl_mark

For the sake of consistency.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NET]: Rethink mark field in struct flowi

Now that all protocols have been made aware of the mark
field it can be moved out of the union thus simplyfing
its usage.

The config options in the IPv4/IPv6/DECnet subsystems
to enable respectively disable mark based routing only
obfuscate the code with ifdefs, the cost for the
additional comparison in the flow key is insignificant,
and most distributions have all these options enabled
by default anyway. Therefore it makes sense to remove
the config options and enable mark based routing by
default.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NET]: Turn nfmark into generic mark

nfmark is being used in various subsystems and has become
the defacto mark field for all kinds of packets. Therefore
it makes sense to rename it to `mark' and remove the
dependency on CONFIG_NETFILTER.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

[DECNET]: Don't clear memory twice.

When dn_neigh.c was converted from kmalloc to kzalloc in commit
0da974f4f303a6842516b764507e3c0a03f41e5a it was missed that
dn_neigh_seq_open was actually clearing the allocation twice was
missed.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

[XFRM]: uninline xfrm_selector_match()

Six callsites, huge.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

[BLUETOOTH] lockdep: annotate sk_lock nesting in AF_BLUETOOTH

=============================================
[ INFO: possible recursive locking detected ]
2.6.18-1.2726.fc6 #1

SELinux: Fix SA selection semantics

Fix the selection of an SA for an outgoing packet to be at the same
context as the originating socket/flow. This eliminates the SELinux
policy's ability to use/sendto SAs with contexts other than the socket's.

With this patch applied, the SELinux policy will require one or more of the
following for a socket to be able to communicate with/without SAs:

1. To enable a socket to communicate without using labeled-IPSec SAs:

allow socket_t unlabeled_t:association { sendto recvfrom }

2. To enable a socket to communicate with labeled-IPSec SAs:

allow socket_t self:association { sendto };
allow socket_t peer_sa_t:association { recvfrom };

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: James Morris <jmorris@namei.org>

SELinux: Return correct context for SO_PEERSEC

Fix SO_PEERSEC for tcp sockets to return the security context of
the peer (as represented by the SA from the peer) as opposed to the
SA used by the local/source socket.

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: James Morris <jmorris@namei.org>

SELinux: Various xfrm labeling fixes

Since the upstreaming of the mlsxfrm modification a few months back,
testing has resulted in the identification of the following issues/bugs that
are resolved in this patch set.

1. Fix the security context used in the IKE negotiation to be the context
   of the socket as opposed to the context of the SPD rule.

2. Fix SO_PEERSEC for tcp sockets to return the security context of
   the peer as opposed to the source.

3. Fix the selection of an SA for an outgoing packet to be at the same
   context as the originating socket/flow.

The following would be the result of applying this patchset:

- SO_PEERSEC will now correctly return the peer's context.

- IKE deamons will receive the context of the source socket/flow
  as opposed to the SPD rule's context so that the negotiated SA
  will be at the same context as the source socket/flow.

- The SELinux policy will require one or more of the
  following for a socket to be able to communicate with/without SAs:

  1. To enable a socket to communicate without using labeled-IPSec SAs:

     allow socket_t unlabeled_t:association { sendto recvfrom }

  2. To enable a socket to communicate with labeled-IPSec SAs:

     allow socket_t self:association { sendto };
     allow socket_t peer_sa_t:association { recvfrom };

This Patch: Pass correct security context to IKE for use in negotiation

Fix the security context passed to IKE for use in negotiation to be the
context of the socket as opposed to the context of the SPD rule so that
the SA carries the label of the originating socket/flow.

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: James Morris <jmorris@namei.org>

[BLUETOOTH] rfcomm endianness bug: param_mask is little-endian on the wire

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[BLUETOOTH]: rfcomm endianness annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[BLUETOOTH]: bnep endianness annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[BLUETOOTH] bnep endianness bug: filtering by packet type

<= and => don't work well on net-endian...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: ip6_output annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NETFILTER]: trivial annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[AF_PACKET]: annotate

Weirdness: the third argument of socket() is net-endian
here.  Oh, well - it's documented in packet(7).

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[LLC]: anotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: annotate inet6_csk_search_req()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: flowlabels are net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[INET]: annotate inet_ecn.h

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NET]: annotate dsfield.h

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[XFRM]: annotate ->new_mapping()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[AF_KEY]: annotate

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV4]: encapsulation annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[SUNRPC]: annotate hash_ip()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: annotate ipv6 mcast

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: annotate struct frag_hdr

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: annotate icmpv6 headers

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: 'info' argument of ipv6 ->err_handler() is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[XFRM]: misc annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6]: annotate inet6_hashtables

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[NET]: ipconfig and nfsroot annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[TIPC]: endianness annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

[IPV6] NDISC: Calculate packet length correctly for allocation.

MAX_HEADER does not include the ipv6 header length in it,
so we need to add it in explicitly.

With help from YOSHIFUJI Hideaki.

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

Merge branch 'upstream-linus' of /linux/kernel/git/jgarzik/netdev-2.6

* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6: (118 commits)
  [netdrvr] skge: build fix
  [PATCH] NetXen: driver cleanup, removed unnecessary __iomem type casts
  [PATCH] PHY: Add support for configuring the PHY connection interface
  [PATCH] chelesio: transmit locking (plus bug fix).
  [PATCH] chelsio: statistics improvement
  [PATCH] chelsio: add MSI support
  [PATCH] chelsio: use standard CRC routines
  [PATCH] chelsio: cleanup pm3393 code
  [PATCH] chelsio: add 1G swcixw aupport
  [PATCH] chelsio: add support for other 10G boards
  [PATCH] chelsio: remove unused mutex
  [PATCH] chelsio: use kzalloc
  [PATCH] chelsio: whitespace fixes
  [PATCH] amd8111e use standard CRC lib
  [PATCH] sky2: msi enhancements.
  [PATCH] sky2: kfree_skb_any needed
  [PATCH] sky2: fixes for Yukon EC_U chip revisions
  [PATCH] sky2: add Dlink 560SX id
  [PATCH] sky2: receive error handling fix
  [PATCH] skge: don't clear MC state on link down
  ...

Merge branch 'for-linus' of /linux/kernel/git/drzeus/mmc

* 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/drzeus/mmc:
  mmc: correct request error handling
  mmc: Flush block queue when removing card
  mmc: sdhci high speed support
  mmc: Support for high speed SD cards
  mmc: Fix mmc_delay() function
  mmc: Add support for mmc v4 wide-bus modes
  [PATCH] mmc: Add support for mmc v4 high speed mode
  trivial change for mmc/Kconfig: MMC_PXA does not mean only PXA255
  Make general code cleanups
  Add MMC_CAP_{MULTIWRITE,BYTEBLOCK} flags
  Platform device error handling cleanup
  Move register definitions away from the header file
  Change OMAP_MMC_{READ,WRITE} macros to use the host pointer
  Replace base with virt_base and phys_base
  mmc: constify mmc_host_ops vectors
  mmc: remove kernel_thread()

Merge branch 'release' of /linux/kernel/git/lenb/linux-acpi-2.6

* 'release' of master.kernel.org:/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6:
  Revert "ACPI: SCI interrupt source override"

[netdrvr] skge: build fix

Signed-off-by: Jeff Garzik <jeff@garzik.org>

Revert "ACPI: SCI interrupt source override"

This reverts commit 281ea49b0c294649a6de47a6f8fbe5611137726b,
which broke ACPI Interrupt source overrides that move
the SCI from one IRQ in PIC mode to another in IOAPIC mode.

If the SCI shared an interrupt line with another device,
this would result in a "irq 18: nobody cared" type failure.

http://bugzilla.kernel.org/show_bug.cgi?id=7601

Signed-off-by: Len Brown <len.brown@intel.com>

[PATCH] NetXen: driver cleanup, removed unnecessary __iomem type casts

Signed-off-by: Amit S. Kale <amitkale@netxen.com>
 netxen_nic.h         |   38 ++++++++------------------------------
 netxen_nic_ethtool.c |    5 ++---
 netxen_nic_hw.c      |   12 +++++-------
 netxen_nic_main.c    |    8 +++-----
 4 files changed, 18 insertions(+), 45 deletions(-)
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] PHY: Add support for configuring the PHY connection interface

Most PHYs connect to an ethernet controller over a GMII or MII
interface.  However, a growing number are connected over
different interfaces, such as RGMII or SGMII.

The ethernet driver will tell the PHY what type of connection it
is by setting it manually, or passing it in through phy_connect
(or phy_attach).

Changes include:
* Updates to documentation
* Updates to PHY Lib consumers
* Changes to PHY Lib to add interface support
* Some minor changes to whitespace in phy.h
* gianfar driver now detects interface and passes appropriate
  value to PHY Lib
Signed-off-by: Andrew Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelesio: transmit locking (plus bug fix).

If transmit lock is contended on, then push return code back
and retry at higher level.

Bugfix: If buffer is reallocated because of lack of headroom
and the send is blocked, then drop packet. This is necessary
because caller would end up requeuing a freed skb.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: statistics improvement

Cleanup statistics management:
 * Get rid of duplicate or unused statistics
 * Convert high volume stats to per-cpu and 64 bit

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: add MSI support

Using MSI can avoid sharing IRQ and associated overhead.
Tested on PCI-X.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: use standard CRC routines

Replace driver crc calculation with existing library.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: cleanup pm3393 code

Replace macro with function for updating RMON values

Cleanups:
* remove unused enum's
* Fix comment format

Signed-off-by: Stephen HEmminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: add 1G swcixw aupport

Add support for 1G versions of Chelsio devices.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: add support for other 10G boards

Add support for other versions of the 10G Chelsio boards.
This is basically a port of the vendor driver with the
TOE features removed.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: remove unused mutex

This mutex is unused in current (non TOE) code.

Signed-off-by: Stephen Hemminger<shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: use kzalloc

Use kzalloc in several places.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: whitespace fixes

Fix indentation and blank/tab issues.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] amd8111e use standard CRC lib

I noticed this driver (and several others) reinvent their own copy of the
existing CRC library. Don't have the hardware, but tested by extracting
code and comparing result.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] sky2: msi enhancements.

If using Message Signaled Interrupts (MSI) then the IRQ will never
be shared. Don't call pci_disable_msi() unless using MSI.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] sky2: kfree_skb_any needed

It is possible for the sky2 driver NAPI poll routine to be called with
IRQ's disabled if netpoll is trying to make space in the tx queue. This
is an obscure path, but if it happens, the kfree_skb needs to happen
via softirq. Calling kfree_skb with IRQ's disabled is a not allowed.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] sky2: fixes for Yukon EC_U chip revisions

Update workarounds for 88E803X based on the latest SysKonnect vendor
driver version (8.41). Tested on EC_U rev A1, only.
These up the receive performance.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] sky2: add Dlink 560SX id

Add new PCI ID for DLink 560SX.
This from the latest SysKonnect vendor driver (version 8.41).

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] sky2: receive error handling fix

If sky2 detects out of memory, or gets a bad frame, it reuses the same receive
buffer, but forgets to poke the hardware. This could lead to the receiver
getting stuck if there were lots of errors.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] skge: don't clear MC state on link down

I would rather fix Andy's problem by not clearing
multicast information on link down.

Also, add code to restore multicast state after ethtool phy reset.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] mv643xx_eth: fix unbalanced parentheses in macros

Signed-off-by: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

drivers/net/tulip/dmfe: support basic carrier detection

Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] netdev: don't allow register_netdev with blank name

This bit of old backwards compatibility cruft can be removed in 2.6.20.
If there is still an device that calls register_netdev()
with a zero or blank name, it will get -EINVAL from register_netdevice().

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] netxen: uses PCI

drivers/built-in.o: In function `netxen_nic_remove':
netxen_nic_main.c:(.text+0x31b4d): undefined reference to `pci_disable_msi'
netxen_nic_main.c:(.text+0x31b8e): undefined reference to `pci_release_regions'
drivers/built-in.o: In function `netxen_init_module':
netxen_nic_main.c:(.init.text+0x3f17): undefined reference to `pci_module_init'
make: *** [.tmp_vmlinux1] Error 1

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: use kzalloc

Use kzalloc() in chelsio driver.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: use __netif_rx_schedule_prep

The chelsio driver can use __netif_rx_schedule_prep instead of it's own
test_and_set inline.  Applies after the previous 4 patches.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: free_netdev

Network devices need to be free'd with free_netdev() not kfree()
otherwise the kernel will panic if an application has /sys/class/net/ethX/value
open and reads it.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: procectomy

Complete removal of proc stuff from chelsio. The orignal driver had a debug
proc interface, but not all the code got removed.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: whitespace cleanup

Whitespace cleanups.  Replace leading spaces with tabs and fix indentation

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] chelsio: remove leftover code

The chelsio network driver has some extra ifdef's that got in because the
driver was originally based on code that worked on 2.4 as well as 2.6.
This patch removes the dead code.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

[PATCH] Adapt ucc_geth driver to use new of_platform_device support

Include of_platform header, and use
new of_[un]register_platform_driver() fns.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>