From: Rusty Russell <rusty@au1.ibm.com>

Vatsa spotted this: you can remove a module while it's being initialized,
and that will be bad.  Hole was opened when I dropped the sem around the
init routine (which can probe for other modules).

From: Nick Piggin <piggin@cyberone.com.au>,
      Rusty Russell <rusty@rustcorp.com.au>

Prevents a race where sys_sched_setaffinity can race with
sched_migrate_task and cause sched_migrate_task to restore an invalid
cpu mask.

- Make it static

- It is larger and has two callsites: uninline it.

From: Jesper Juhl <juhl-lkml@dif.dk>

I'm compiling 2.6.1-rc1-mm1 with "-W -Wall" to look for potential problems
and minor stuff to clean up.

One of the things that enabling the extra warnings turn up is errors about
the placement of the inline keyword.

Fix warnings and build errors under EXT2_DEBUG.

From: Pavel Machek <pavel@ucw.cz>

Softcursor was broken for half of 2.5 series.  This fixes it by first
hiding cursor _then_ hiding softcursor.  Very simple mistake...

From: Pavel Machek <pavel@ucw.cz>

This is obsolete x86-64 code.

- Add missing locking around s_next_generation increment

- Correctly set the initial value of s_next_generation.

There is no locking around the increment of this per-filesystem counter.
Create a new lock, just for this.

We've cached EXT2_SB(sb) in local variable `sbi'.  Use it.

From: Pavel Machek <pavel@ucw.cz>

This kills outdated docs, and adds some helpful docs about video issues
with suspend.

From: Pavel Machek <pavel@ucw.cz>

wakeup.S includes some rather nasty, and unneccessary debugging code.  (It
used to try to flush caches/tlbs; now its totally useless).

From: "H. Peter Anvin" <hpa@zytor.com>

MAINTAINERS update

From: "Bryan O'Sullivan" <bos@serpentine.com>

For compound pages, page_count needs to be sure to reference the head page.

This affects code that plays tricks with memory mappings into userspace,
which would mostly involve video drivers.

From: "Nakajima, Jun" <jun.nakajima@intel.com>

Restore the HT detection algorithm.  Make the processor package mapping
subarch-specific so that it can reflect the APIC ID info provided by BIOS
if required.

into home.osdl.org:/home/torvalds/v2.5/linux

From: Luiz Fernando Capitulino <lcapitulino@prefeitura.sp.gov.br>,
      Randy.Dunlap <rddunlap@osdl.org>

into home.osdl.org:/home/torvalds/v2.5/linux

into home.osdl.org:/home/torvalds/v2.5/linux

If we detect an error late in this function, local variable `interpreter' can
be NULL.  Don't try to fput() it in that case.

3c59x has a tunable `rx_copybreak'.  But it's declared const, so recent gcc's
rub it out altogether, and modprobe says `3c59x: falsely claims to have
parameter rx_copybreak'.

This way, the "static char buf" is defined only once instead
of once per file.

From: NeilBrown <neilb@cse.unsw.edu.au>

From: Greg Banks <gnb@melbourne.sgi.com>,

see comment in code.

From: NeilBrown <neilb@cse.unsw.edu.au>

Based on code from Terje Malmedal <terje.malmedal@usit.uio.no>

From: NeilBrown <neilb@cse.unsw.edu.au>

follow_up doesn't check that the passed dentry is at the root of the mounted
file system, so we have to do that, to avoid calling follow_up too many
times.

From: NeilBrown <neilb@cse.unsw.edu.au>

We weren't converting the -E* error code to an nfserr_* error code as we
should.

From: NeilBrown <neilb@cse.unsw.edu.au>

nfsd_setuser was not called until after filehandle lookup was done, so
filehandle lookup would use the wrong userid, and so the permission checking
in nfsd_acceptable would be bogus.

From: John Levon <levon@movementarian.org>

Please apply the below and accept patches directly from Philippe for
OProfile in the future, as necessary.

(Philippe is the other main developer of OProfile)

From: Tom Rini <trini@kernel.crashing.org>

Skip debug sections of modules - taking 12 minutes to load the ipv6 module
is a bug.  :)

From: Mike Christie <michaelc@cs.wisc.edu>,
      Jens Axboe <axboe@suse.de>

It's cleaner and more correct to look at req->rl to determine whether this
request got from the block layer requests lists instead of using req->q.
It's handy to always have req->q available, to lookup the queue from the
request.

From: Gerd Knorr <kraxel@bytesex.org>

This patch adds a experimental driver for Conexant 2388x based TV cards.

DESC
cx88-video compile fixes for older gcc
EDESC

Fix some gcc-3.xisms.

From: Gerd Knorr <kraxel@bytesex.org>

This is a saa7134 driver update.  Changes:

* add infrared remote support.

* add support for more TV cards.

* misc minor fixes.

This patch depends on the ir-input patch.

From: Gerd Knorr <kraxel@bytesex.org>

This patch adds linux input layer based support for infrared remote
  controls.  It adds two new modules:

* ir-kbd-i2c supports i2c-based IR receivers.

* ir-kbd-gpio supports IR receivers connected to the bt878 gpio pins.

This patch depends on the ir-input patch and the bttv driver update.

From: Gerd Knorr <kraxel@bytesex.org>

This is a i2c tuner module update.  Changes:

* adds support for more tunes (mt2050, atsc, ...)

* some reorganization, uses function pointers to branch to different
  functions for different tuner types.

From: Gerd Knorr <kraxel@bytesex.org>

This is a collection of misc i2c tv helper module fixes.

From: Gerd Knorr <kraxel@bytesex.org>

This patch adds a module with some helper functions to handle infrared
remotes using the linux input layer.  It doesn't do any useful stuff alone,
but the saa7134 and bttv drivers will use that to support the remotes shipped
with some TV cards.

From: Gerd Knorr <kraxel@bytesex.org>

bttv driver update.  Changes:

(1) Added a new, sysfs-based interface to allow access to the bt878 gpio
    lines from other kernel modules (new bttv-gpio.c file), which fixes alot
    of problems the old interface has (which still is in bttv-if.c, to be
    removed in 2.7).

(2) moved the i2c code from bttv-if.c to bttv-i2c.c.

(3) created a new "struct bttv_core" and moved some of the entries from
    "struct bttv" to the new one.  This is needed for (1) and makes the patch
    pretty big, althrough there is no actual code changes.

(4) first cut for suspend support (S1 works for me).

(5) usual batch of new/updated tv card list entries.

(6) minor fixes.

From: Gerd Knorr <kraxel@bytesex.org>

This patch contains video-buf fixes, mainly remove videobuf_lock() and
videobuf_unlock() functions.  They are not needed as get_user_pages() locks
down the pages anyway.

From: Gerd Knorr <kraxel@bytesex.org>

This patch is a v4l2 update.  Changes:

* added new ioctls (VIDIOC_G_PRIORITY, VIDIOC_S_PRIORITY) to give v4l apps
  priorities for tuning and other controls.  Main purpose is that backgrounds
  applictions like nxtvepg which are fishing informations from /dev/vbi
  (teletext, epg, ...) can run at lower priority than interactive tv apps and
  thus the user isn't annonyed with unexpected channel switches.

* add a set of helper functions to handle priorities to the v4l2-common
  module.

* minor fixes in the v4l1-compat module.

* minor header file fixes.

From: Gerd Knorr <kraxel@bytesex.org>

This patch fixes some v4l2 ioctl #defines which have wrong _IO* macros.  It
also adds a function which maps the old numbers to to new ones to maintain
binary backward compatibility.