Agreeing with Ben H's recommendation, I reduce the max-hw-segments
number to 128, to match max-phys-segments number.  This shouldn't
affect performance because 128*64K is far above the max transfer size
for most current IDE devices.  Even 128*4K is OK.

This switches alpha to use drivers/Kconfig instead of including most of
files included by drivers/Kconfig, thus enabling previously omitted i2c
(already used in some drivers available on alpha) and telephony drivers.

Because drivers/message/fusion/Kconfig was included only conditionally
("if PCI"), this patch also changes CONFIG_FUSION to depend on PCI.

(these changes were consulted with Christoph Hellwig in "i2c on alpha
- used but not available in 2.6.3" thread on LKML)

into ppc970.osdl.org:/home/torvalds/v2.5/linux

into ppc970.osdl.org:/home/torvalds/v2.5/linux

arm-26, ppc, sparc, sparc64 and sh have per-arch definitions of
__cacheline_aligned that are identical to default.  And yes, removal is
safe - all users of __cacheline_aligned actually pull linux/cache.h in.

We put ____cacheline_aligned things in their own section, simply
because we waste less space that way.  Otherwise we end up padding
innocent variables to the next cacheline to get the required
alignment.

There's no reason not to do this in modules, too.

	Removes gratitious ifdefs - get_nodes() is only called from __init
and by itself, check_space() is only called by get_nodes(), so they can
be made unconditionally __init.

into ppc970.osdl.org:/home/torvalds/v2.5/linux

Use KERN_DEBUG in all debugging printk() output.

This patch originally from the kernel janitors.

Fix interrupt handler return types to be irqreturn_t.

Remove non-existent config option from defconfig.

I goofed in my last patch to this code..
It reported 1 less CPU than it should have. Doh.

into ppc970.osdl.org:/home/torvalds/v2.5/linux

From: Russell King <rmk@arm.linux.org.uk>

Export scaling_available_frequencies on the remaining
x86 freq_table-based cpufreq drivers. powernow-k7, acpi and speedstep-centrino
are already queued.
                                                                                                       
Please note that this cannot be done in the cpufreq core as the cpufreq core
tries very hard to be independent of the freq_table helpers. 

From: Kronos <kronos at kronoz.cjb.net>
 
The following patch make powernow-k7.c export supported frequencies via
sysfs. I'm trying to write a scaling deamon and I need to know them.

From: Herbert Xu <herbert@gondor.apana.org.au>

update_screen() requires the console semaphore.

net/wanrouter/wanmain.c:195: error: parse error before "static"

From: Sam Ravnborg <sam@ravnborg.org>

List all entries in arch/$(ARCH)/configs/*_defconfig when doing 'make
help'.

Results in output like this (ppc64 as example):

  g5_defconfig             - Build for g5
  pSeries_defconfig        - Build for pSeries

The implementation is generic and enables this for all users of _defconfig.

From: linux@MichaelGeng.de (Michael Geng)

Add new driver for Teletext decoder SAA5246A from Philips.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99
initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99
initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99
initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing GNU-style initializers to C99 initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99
initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99
initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99
initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99
initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing the GNU-style initializers to C99 initializers.

From: "Art Haas" <ahaas@airmail.net>

Here's a small patch changing a few GNU-style initializers to C99
initiailzers.

From: Dave Jones <davej@redhat.com>

Now that the syscalls.h stuff is in mainline, that clears the way for this
I guess?

From: "Art Haas" <ahaas@airmail.net>

This patch converts the file to use C99 initializers.  The file had this
patch applied before, but a subsequent patch reversed the change and
readded the GNU-style initializer.  This patch brings back the C99 one.

From: David Woodhouse <dwmw2@infradead.org>,
      Dave Jones <davej@redhat.com>

Add a missing unlock.

From: Jakub Bogusz <qboosh@pld-linux.org>

This patch removes using of floating point operations in tdfxfb - they
are really not needed here (these consts are used only in substitutions
and comparisons with integers) are cause unresolved symbols on some
archs - e.g. on alpha:

*** Warning: "__ltdf2" [drivers/video/tdfxfb.ko] undefined!
*** Warning: "__adddf3" [drivers/video/tdfxfb.ko] undefined!
*** Warning: "__floatsidf" [drivers/video/tdfxfb.ko] undefined!

From: Marcelo Tosatti <marcelo.tosatti@cyclades.com>

The callout code has been removed long ago from 2.6.  I believe it is safe
to remove some of its unused defines.

From: Herbert Xu <herbert@gondor.apana.org.au>

RADEON is selecting I2C_ALGOBIT but as kconfig currently does not propagate
selects up the dependency chain (that is according to Roman), this can
leave I2C as m while I2C_ALGOBIT is y.

early printk documentation fix

From: Manuel Estrada Sainz <ranty@ranty.pantax.net>

Based on patch and suggestions from Dmitry Torokhov

- Don't remove attributes, they should be gone automatically.