The recent fbdev changes broke something quite fundamental on ppc, which
is the way offb works. It's a "fallback" driver that is to be used when
no other driver picked up the video, or when "forced" via the video=ofonly
command line option.

The recent changes completely broke that (which is a pretty important
behaviour on ppc since we still have some models regulary with video
cards that don't work properly with the kernel drivers, like some nVidias
or some older stuffs).

This patch moves offb to the end of the Makefile, so at least the behaviour
of taking over as a "fallback" is restored (the current kernel will have
offb take control of the framebuffer before any fbdev has a chance to do
it, which breaks everything).

Apparently, the entire support for "video=ofonly" was removed though,
that NEEDS to be restored in some way, though i'm not yet sure what is
the best path to that yet, I have to look more deeply at the new code.

 - create in_lock_functions() to match in_sched_functions(). Export it
   for use in oprofile.
 - use char __lock_text_start[] instead of long __lock_text_start when
   declaring linker symbols. Rusty fixed a number of these a while ago
   based on advice from rth.
 - Move __preempt_*_lock into kernel/spinlock.c and make it inline. This
   means locks are only one deep.
 - Make in_sched_functions() check in_lock_functions()
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Andries made a declaration static, and this clashed with Al who
changed it for byteorder annotation. Fix it all up.

That one was slightly messier than ext2 - several data structures had
been used both in little-endian (on-disk) and host-endian (in-core)
situations.  They had to be split, the rest was trivial annotations. 

Gets UDF sparse-clean with -Wbitwise.  Note that here we just annotate -
bug fixes from the original version of that patch had already been
merged.
Signed-off-by: Al Viro <viro@parcelfarce.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

AFFS partially annotated - some fields of on-disk structures are never
used and while I'm reasonably sure that they ought to be big-endian, I'd
rather leave them alone for now; annotating them won't change the amount
of noise since nothing in the kernel ever accesses them. 
Signed-off-by: Al Viro <viro@parcelfarce.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Signed-off-by: Al Viro <viro@parcelfarce.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Signed-off-by: Al Viro <viro@parcelfarce.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This adds the types and annotates conversion functions.  I've converted
the ...p() versions to inlines; AFAICS, everything's still happy...
Signed-off-by: Al Viro <viro@parcelfarce.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

In highmem end_io handling, we need to iterate over the completed bio
from 0, not bio->bi_idx. If not we leak N-1 pages for any bio with N
pages where N > 1.
Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

into ppc970.osdl.org:/home/torvalds/v2.6/linux

into kernel.bkbits.net:/home/davem/net-2.6

We divide by the wrong MSS.
Signed-off-by: David S. Miller <davem@davemloft.net>

Unused outside of tcp_output.c
Signed-off-by: David S. Miller <davem@davemloft.net>

Looks like a bug in the cleanup patch :)

From: Andrew Morton <akpm@osdl.org>

Shows how many people are testing ext2.  Let's fix up that whitespace
also. 
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

into ppc970.osdl.org:/home/torvalds/v2.6/linux

into agluck-lia64.sc.intel.com:/data/home/aegl/BK/linux-ia64-release-2.6.9

into kernel.bkbits.net:/home/davem/net-2.6

into agluck-lia64.sc.intel.com:/data/home/aegl/BK/linux-ia64-release-2.6.9

Signed-off-by: Andreas Schwab <schwab@suse.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>

into agluck-lia64.sc.intel.com:/data/home/aegl/BK/linux-ia64-release-2.6.9

On multi-node ia64 system, SysRq-M seems to dump wrong memory info.
(Since I don't have such a large machine, I don't confirm it)
It should reset counters every iteration each node in show_mem().
Jesse Barnes: "This looks good".
Signed-off-by: Akinobu Mita <mita@yacht.ocn.ne.jp>
Signed-off-by: Tony Luck <tony.luck@intel.com>

Jon Smirl's patch to export roms via sysfs exposes a bug in the ia64
PCI code; we're not assigning a parent to the ROM resource. Patch
supplied by Matthew Wilcox
Signed-off-by: Tony Luck <tony.luck@intel.com>

 
Patch (against 2.6.8.1) removes unnecessary min/max macros and changes
calls to use kernel.h macros instead.
Signed-off-by: Michael Veeck <michael.veeck@gmx.net>
Signed-off-by: Maximilian Attems <janitor@sternwelten.at>
Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

Signed-off-by: davidm@hpl.hp.com
Signed-off-by: Tony Luck <tony.luck@intel.com>

This patch adds ECN encapsulation/decapsulation to ip6_tunnel.c.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

While doing the ECN patch I discovered that IPsec on IPv6 wasn't
encapsulating the ECN correctly.  This patch fixes that.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

Optimize away a branch and clean up the logic.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

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

Add create_proc_ide_interfaces() call to sgiioc4_ide_setup_pci_device()
so /proc/ide gets populated properly.
Signed-off-by: Erik Jacobson <erikj@sgi.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@elka.pw.edu.pl>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Signed-off-by: Peter Buckingham <peter@pantasys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

into ppc970.osdl.org:/home/torvalds/v2.6/linux

I want to change my vote :-) The problem isn't a spurious extra "#else"
(which Paul's patch removed) ...  it is a missing "#endif"

Here's a patch that puts the #else back, adds the #endif, and fixes the
whitespace to make this nested mess of pre-processor noise a bit more
legible. 
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

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

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

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

This patch brings the IP ECN handling up-to-date with repsect to
RFC 3168.  Mostly this means treating ECT(1) in the same way as
ECT(0).
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
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>

New version of the NETIF_F_LLTX for network devices patch. 

This allows network drivers to set the NETIF_F_LLTX flag
and then do their own locking in start_queue_xmit. 
This lowers locking overhead in this critical path. 

The drivers can use try lock if they want and return -1
when the lock wasn't grabbed. In this case the packet
will be requeued. For better compatibility this is only
done for drivers with LLTX set, others don't give a special
meaning to -1.

Most of the modern drivers who have a lock around hard_start_xmit
can just set this flag. It may be a good idea to convert the spin
lock there to a try lock. The only thing that should be audited
is that they do enough locking in the set_multicast_list function
too, and not also rely on xmit_lock here.

Now doesn't move any code around and does things with gotos instead.

The loop printk is also still there even for NETIF_F_LLTX

For drivers that don't set the new flag nothing changes.
Signed-off-by: David S. Miller <davem@davemloft.net>

Here's the optimization: if we know the remaining data exceeds the
mtu, we do not need to fill up full of it.

skb_prev:
                mtu
+----------+--+-+
|          |  | |
+----------+--+-+
           ^  ^
skb_prev->len |
              maxfraglen

appending data:
           +--------+
           |        |
           +--------+
           --------->
                  length

In this case, we know we need more fragment(s).
So, let's fill up to maxfraglen (instead of mtu) 
to avoid needless copy in the next loop.
Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>