[ARM] 3047/1: SMDK2440 - add framebuffer platform data

Patch from Ben Dooks

Add platform data for framebuffer for the
onboard LCD module

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3046/1: BAST - add framebuffer platform data

Patch from Ben Dooks

Add framebuffer platform data

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3045/2: S3C2410 - change init for lcd platform data

Patch from Ben Dooks

Change set_s3c2410fb_info to s3c2410_fb_set_platdata
and use kmalloc() for the copy of the information it
is passed.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2908/1: add register definitions for ixp2000 msf

Patch from Lennert Buytenhek

We already had a mapping for the msf, but we didn't have any
register definitions for it yet.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3044/1: Fix sparse warnings about incompatible pointer types for register defined in pxa-regs.h

Patch from Ian Campbell

The sparse warning initially surfaced in sound/arm/pxa2xx-ac97.c
because it was using u32 * variables to hold the unsigned long *
register addresses.

I submitted an ALSA patch for this http://thread.gmane.org/gmane.linux.alsa.devel/27804 issue and it was suggested that it might be preferable to change the register
definitions to use u32.

Most other subarches seem to use u32 for their register type, at least
the ones which use a __REG macro (like the PXA) do. Nico indicated in
the thread above that he wouldn't mind this patch.

Changing the type required fixes for opposite warnings in the pxa2xx usb
gadget code but that was the only new warning introduced on defconfig
or lubbock, mainstone and our own PXA255 boards.

Signed-off-by: Ian Campbell <icampbell@arcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3034/1: S3C2410 - fix size of devices in devs.c

Patch from Ben Dooks

From: Guillaume GOURAT <guillaume.gourat@nexvision.fr>

A number of devices have an extra byte on the
end of their areas due to mis-calculating the
.end field of their resources

Signed-off-by: Guillaume GOURAT <guillaume.gourat@nexvision.fr>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3033/1: S3C2410 - add generic gpio_cfgpin options

Patch from Ben Dooks

Add generic values for the parameters to the
s3c2410_gpio_cfgpin() function, so that a caller
does not need to know the exact constant for
the specified pin.

This is very useful for the case where a driver
is passed a gpio pin number and needs to reconfigure
the pin's function.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2930/1: optimized sha1 implementation for ARM

Patch from Nicolas Pitre

Here's an ARM assembly SHA1 implementation to replace the default C
version. It is approximately 50% faster than the generic C version. On
an XScale processor running at 400MHz:
generic C version: 9.8 MB/s
my version: 14.5 MB/s
This code is useful to quite a few callers in the tree:
crypto/sha1.c: sha_transform(sctx->state, sctx->buffer, temp);
crypto/sha1.c: sha_transform(sctx->state, &data[i], temp);
drivers/char/random.c: sha_transform(buf, (__u8 *)r->pool+i, buf + 5);
drivers/char/random.c: sha_transform(buf, (__u8 *)data, buf + 5);
net/ipv4/syncookies.c: sha_transform(tmp + 16, (__u8 *)tmp, tmp + 16 + 5);

Signed-off-by: Nicolas Pitre <nico@cam.org>
Seems to work fine on big-endian as well.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3017/1: Add support for 36-bit addresses to create_mapping()

Patch from Deepak Saxena

This patch adds support for 36-bit static mapped I/O. While there
are no platforms in the tree ATM that use it, it has been tested
tested on the IXP2350 NPU and I would like to get the support for
that chipset upstream one piece at a time. There are also other
Intel chipset ports in development that are waiting on this to go
upstream.

The patch replaces the print formats for physical addresses with
%016llx which will create a bit extraneous output on 32-bit systems,
but I think that is cleaner than having #ifdefs, specially since
users will only see the output in error cases.

Depends on 3016/1.

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3016/1: Replace map_desc.physical with map_desc.pfn

Patch from Deepak Saxena

Convert map_desc.physical to map_desc.pfn. This allows us to add
support for 36-bit addressed physical devices in the static maps
without having to resort to u64 variables.

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3001/1: Replace map_desc.physical with map_desc.pfn: Integrator

Patch from Deepak Saxena

Integrator map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3000/1: Replace map_desc.physical with map_desc.pfn: iMX

Patch from Deepak Saxena

iMX map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2999/1: Replace map_desc.physical with map_desc.pfn: H720x

Patch from Deepak Saxena

H720x map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2998/1: Replace map_desc.physical with map_desc.pfn: Footbridge

Patch from Deepak Saxena

Footbridge map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2997/1: Replace map_desc.physical with map_desc.pfn: EBSA110

Patch from Deepak Saxena

EBSA110 map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2996/1: Replace map_desc.physical with map_desc.pfn: EPXA10DB

Patch from Deepak Saxena

EPXA10DB mapd_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2995/1: Replace map_desc.physical with map_desc.pfn: Versatile

Patch from Deepak Saxena

Versatile map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2994/1: Replace map_desc.physical with map_desc.pfn: CLPS7500

Patch from Deepak Saxena

CLSP7500 map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2993/1: Replace map_desc.physical with map_desc.pfn: CLPS711x

Patch from Deepak Saxena

CLSP711x map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2992/1: Replace map_desc.physical with map_desc.pfn: SA1100

Patch from Deepak Saxena

SA1100 map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2991/1: Replace map_desc.physical with map_desc.pfn: Shark

Patch from Deepak Saxena

Shark map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2990/1: Replace map_desc.physical with map_desc.pfn: S3C2410

Patch from Deepak Saxena

S3C2410 map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2989/1: Replace map_desc.physical with map_desc.pfn: RiscPC

Patch from Deepak Saxena

RiscPC map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2988/1: Replace map_desc.physical with map_desc.pfn: PXA

Patch from Deepak Saxena

PXA map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2987/1: Replace map_desc.physical with map_desc.pfn: OMAP

Patch from Deepak Saxena

OMAP map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2986/1: Replace map_desc.physical with map_desc.pfn: LH7A40x

Patch from Deepak Saxena

LH7A40x map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2985/1: Replace map_desc.physical with map_desc.pfn: IXP4xx

Patch from Deepak Saxena

IXP4xx map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2984/1: Replace map_desc.physical with map_desc.pfn: IXP2000

Patch from Deepak Saxena

IXP2000 map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2983/1: Replace map_desc.physical with map_desc.pfn: IOP3xx

Patch from Deepak Saxena

IOP3xx map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2982/1: Replace map_desc.physical with map_desc.pfn: aaec2000

Patch from Deepak Saxena

aaec2000 map_desc.pfn conversion

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 4/4 Combine oprofile common and init code

There is nothing special about having the init code separate from
the common code, so combine the two.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3/4 Rename common oprofile code

The common oprofile code assumes the name "PMU" (from Intel's
performance management unit).  This is misleading when we
start adding oprofile support for other machine types which
don't use the same terminology.  Call it op_arm_* instead of
pmu_*.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2/4 Fix oprofile suspend/resume

The oprofile suspend/resume was missing locking.  If we failed
to start oprofile on resume, we still reported that it was
enabled.  Instead, disable oprofile on error.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 1/4 Move oprofile driver model code

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] Re-jig bootmem initialisation

Make ARM independent of the way bootmem operates internally.  We
now map each node as we initialise it, and place the bootmem bitmap
inside each node, rather than all in the first node.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] Fix sparse warnings

Fix sparse warnings in arch/arm/kernel/module.c,
arch/arm/mm/consistent.c, drivers/pcmcia/sa1111_generic.c,
and platform support files.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3/4: Remove asm/hardware.h from SA1100 io.h

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 3/4: Remove asm/hardware.h from Versatile and Integrator io.h

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 2/4: Remove asm/hardware.h from asm-arm/arch-ebsa110/io.h

EBSA110 only requires hardware.h to be included for a couple of
files.  Move the include there.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[ARM] 1/4: Move include of asm/hardware.h to asm-arm/arch-*/io.h

Including asm/hardware.h into asm/io.h can cause #define clashes
between platform specific definitions and driver local definitions.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

Linux v2.6.14

"Better late than never"

Merge /pub/scm/linux/kernel/git/acme/net-2.6

[PATCH] cpufreq: SMP fix for conservative governor

Don't try to access not-present CPUs.  Conservative governor will always
oops on SMP without this fix.

Fixes http://bugzilla.kernel.org/show_bug.cgi?id=4781

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Revert "x86-64: Avoid unnecessary double bouncing for swiotlb"

Commit id 6142891a0c0209c91aa4a98f725de0d6e2ed4918

Andi Kleen reports that it seems to break things for some people,
and since it's purely a small optimization, revert it for now.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[TCP]: Clear stale pred_flags when snd_wnd changes

This bug is responsible for causing the infamous "Treason uncloaked"
messages that's been popping up everywhere since the printk was added.
It has usually been blamed on foreign operating systems.  However,
some of those reports implicate Linux as both systems are running
Linux or the TCP connection is going across the loopback interface.

In fact, there really is a bug in the Linux TCP header prediction code
that's been there since at least 2.1.8.  This bug was tracked down with
help from Dale Blount.

The effect of this bug ranges from harmless "Treason uncloaked"
messages to hung/aborted TCP connections.  The details of the bug
and fix is as follows.

When snd_wnd is updated, we only update pred_flags if
tcp_fast_path_check succeeds.  When it fails (for example,
when our rcvbuf is used up), we will leave pred_flags with
an out-of-date snd_wnd value.

When the out-of-date pred_flags happens to match the next incoming
packet we will again hit the fast path and use the current snd_wnd
which will be wrong.

In the case of the treason messages, it just happens that the snd_wnd
cached in pred_flags is zero while tp->snd_wnd is non-zero.  Therefore
when a zero-window packet comes in we incorrectly conclude that the
window is non-zero.

In fact if the peer continues to send us zero-window pure ACKs we
will continue making the same mistake.  It's only when the peer
transmits a zero-window packet with data attached that we get a
chance to snap out of it.  This is what triggers the treason
message at the next retransmit timeout.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[PATCH] Yet more posix-cpu-timer fixes

This just makes sure that a thread's expiry times can't get reset after
it clears them in do_exit.

This is what allowed us to re-introduce the stricter BUG_ON() check in
a362f463a6d316d14daed0f817e151835ce97ff7.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Revert "remove false BUG_ON() from run_posix_cpu_timers()"

This reverts commit 3de463c7d9d58f8cf3395268230cb20a4c15bffa.

Roland has another patch that allows us to leave the BUG_ON() in place
by just making sure that the condition it tests for really is always
true.

That goes in next.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] Fix cpu timers expiration time

There's a silly off-by-one error in the code that updates the expiration
of posix CPU timers, causing them to not be properly updated when they
hit exactly on their expiration time (which should be the normal case).

This causes them to then fire immediately again, and only _then_ get
properly updated.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Merge master.kernel.org:/home/rmk/linux-2.6-serial

Merge master.kernel.org:/home/rmk/linux-2.6-arm

posix cpu timers: fix timer ordering

Pointed out by Oleg Nesterov, who has been walking over the code
forwards and backwards.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] fix radeon_cp_init_ring_buffer()

I've seen similar failure on alpha.

Obviously, someone forgot to convert sg->handle stuff for
PCI gart case.

Signed-off-by: Dave Airlie <airlied@linux.ie>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] svcsock timestamp fix

Convert nanoseconds to microseconds correctly.

Spotted by Steve Dickson <SteveD@redhat.com>

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] Fix HFS+ to free up the space when a file is deleted.

fsck_hfs reveals lots of temporary files accumulating in the hidden
directory "\000\000\000HFS+ Private Data".  According to the HFS+
documentation these are files which are unlinked while in use.  However,
there may be a bug in the Linux hfsplus implementation which causes this to
happen even when the files are not in use.  It looks like the "opencnt"
field is never initialized as (I think) it should be in hfsplus_read_inode.
 This means that a file can appear to be still in use when in fact it has
been closed.  This patch seems to fix it for me.

Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] kill massive wireless-related log spam

Although this message is having the intended effect of causing wireless
driver maintainers to upgrade their code, I never should have merged this
patch in its present form.  Leading to tons of bug reports and unhappy
users.

Some wireless apps poll for statistics regularly, which leads to a printk()
every single time they ask for stats.  That's a little bit _too_ much of a
reminder that the driver is using an old API.

Change this to printing out the message once, per kernel boot.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] export cpu_online_map

With CONFIG_SMP=n:

*** Warning: "cpu_online_map" [drivers/firmware/dcdbas.ko] undefined!

due to set_cpus_allowed().

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] ppc64: Fix wrong register mapping in mpic driver

The mpic interrupt controller driver (used on G5 and early pSeries among
others) has a bug where it doesn't get the right virtual address for the
timer registers.  It causes the driver to poke at the MMIO space of
whatever has been mapped just next to it (ouch !) when initializing and
causes boot failures on some IBM machines.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] NUMA: broken per cpu pageset counters

The NUMA counters in struct per_cpu_pageset (linux/mmzone.h) are never
cleared today.  This works ok for CPU 0 on NUMA machines because
boot_pageset[] is already zero, but for other CPU:s this results in
uninitialized counters.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] md: make sure mdthreads will always respond to kthread_stop

There are still a couple of cases where md threads (the resync/recovery
thread) is not interruptible since the change to use kthreads.  All places
there it tests "signal_pending", it should also test kthread_should_stop,
as with this patch.

Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[ARM] 3032/1: sparse: complains about generic_fls() prototype in asm-arm/bitops.h

Patch from Ian Campbell

Sparse complains about the definition of generic_fls in asm-arm/bitops.h:
  CHECK   /home/icampbell/devel/kernel/2.6/arch/arm/mach-pxa/viper.c
include2/asm/bitops.h:350:34: error: marked inline, but without a definition

The definition is unnecessary since linux/bitops.h defines generic_fls before including asm/bitops.h and asm/bitops.h should not be included directly. There are still some places where asm/bitops.h is directly included, but I think that code should be fixed. I was a little wary of the patch for this reason but lubbock, mainstone and assabet all build OK and so do my in house boards...

ARM is the only arch with the generic_fls prototype in this way.

Signed-off-by: Ian Campbell <icampbell@arcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

PCI: be more verbose about resource quirks

When reserving an PCI quirk, note that in the kernel bootup messages.

Also, parse the strange PIIX4 device resources - they should get their
own PCI resource quirks, but for now just print out what it finds to
verify that the code does the right thing.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[LLC]: Strip RIF flag from source MAC address

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

[TR]: Preserve RIF flag even for 2 byte RIF fields.

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

[IPV6]: Fix refcnt of struct ip6_flowlabel

Signed-off-by: Yan Zheng <yanzheng@21cn.com>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[PATCH] qlogic lockup fix

If qla2x00_probe_one()'s call to qla2x00_iospace_config() fails, we call
qla2x00_free_device() to clean up.  But because ha->dpc_pid hasn't been set
yet, qla2x00_free_device() tries to stop a kernel thread which hasn't started
yet.  It does wait_for_completion() against an uninitialised completion struct
and the kernel hangs up.

Fix it by initialising ha->dpc_pid a bit earlier.

Cc: Andrew Vasquez <andrew.vasquez@qlogic.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] alpha: atomic dependency fix

My alpha build is exploding because asm/atomic.h now needs smb_mb(), which is
over in the (not included) system.h.

I fear what will happen if I include system.h into atomic.h, so let's put the
barriers into their own header file.

Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[ARM] fix sharp zaurus c-3000 compile failure without CONFIG_FB_PXA

This fixes compile problem when CONFIG_FB_PXA is not set.

  LD      .tmp_vmlinux1
arch/arm/mach-pxa/built-in.o(.text+0x1d74): In function
`spitz_get_hsync_len':
: undefined reference to `pxafb_get_hsync_time'
make: *** [.tmp_vmlinux1] Error 1
3.46user 0.46system 5.10 (0m5.106s) elapsed 77.01%CPU

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[SERIAL] new hp diva console port

Add the new ID 0x132a and configure the new PCI Diva console port.  This
device supports only 1 single console UART.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[SERIAL] support the Exsys EX-4055 4S four-port card

Tested by Wolfgang Denk with this device:

    00:0f.0 Network controller: PLX Technology, Inc. PCI <-> IOBus Bridge (rev 01)
        Subsystem: Exsys EX-4055 4S(16C550) RS-232
        Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B-
        Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Interrupt: pin A routed to IRQ 10
        Region 0: Memory at 80100000 (32-bit, non-prefetchable) [size=128]
        Region 1: I/O ports at 7080 [size=128]
        Region 2: I/O ports at 7400 [size=32]

    00:0f.0 Class 0280: 10b5:9050 (rev 01)
        Subsystem: d84d:4055

Results with this patch:

    Serial: 8250/16550 driver $Revision: 1.90 $ 32 ports, IRQ sharing enabled
    ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
    PCI: Found IRQ 10 for device 0000:00:0f.0
    ttyS4 at I/O 0x7400 (irq = 10) is a 16550A
    ttyS5 at I/O 0x7408 (irq = 10) is a 16550A
    ttyS6 at I/O 0x7410 (irq = 10) is a 16550A
    ttyS7 at I/O 0x7418 (irq = 10) is a 16550A

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

[PATCH] Return the line length via sysfs for fbdev

This small patch returns the stride/line length of the framebuffer via
sysfs.

Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] ALSA: Fix Oops of suspend/resume with generic drivers

The patch fixes Oops from sound drivers using generic platform device
but have no suspend/resume callbacks.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] uml: fix compile failure for TT mode

Without this patch, uml compile fails with:

  LD      .tmp_vmlinux1
arch/um/kernel/built-in.o: In function `config_gdb_cb':
arch/um/kernel/tt/gdb.c:129: undefined reference to `TASK_EXTERN_PID'

Tested on i386, but fix needed on x86_64 too AFAICS.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] posix-timers: fix posix_cpu_timer_set() vs run_posix_cpu_timers() race

This might be harmless, but looks like a race from code inspection (I
was unable to trigger it).  I must admit, I don't understand why we
can't return TIMER_RETRY after 'spin_unlock(&p->sighand->siglock)'
without doing bump_cpu_timer(), but this is what original code does.

posix_cpu_timer_set:

read_lock(&tasklist_lock);

spin_lock(&p->sighand->siglock);
list_del_init(&timer->it.cpu.entry);
spin_unlock(&p->sighand->siglock);

We are probaly deleting the timer from run_posix_cpu_timers's 'firing'
local list_head while run_posix_cpu_timers() does list_for_each_safe.

Various bad things can happen, for example we can just delete this timer
so that list_for_each() will not notice it and run_posix_cpu_timers()
will not reset '->firing' flag. In that case,

....

if (timer->it.cpu.firing) {
read_unlock(&tasklist_lock);
timer->it.cpu.firing = -1;
return TIMER_RETRY;
}

sys_timer_settime() goes to 'retry:', calls posix_cpu_timer_set() again,
it returns TIMER_RETRY ...

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] posix-timers: exit path cleanup

No need to rebalance when task exited

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] posix-timers: remove false BUG_ON() from run_posix_cpu_timers()

do_exit() clears ->it_##clock##_expires, but nothing prevents
another cpu to attach the timer to exiting process after that.

After exit_notify() does 'write_unlock_irq(&tasklist_lock)' and
before do_exit() calls 'schedule() local timer interrupt can find
tsk->exit_state != 0. If that state was EXIT_DEAD (or another cpu
does sys_wait4) interrupted task has ->signal == NULL.

At this moment exiting task has no pending cpu timers, they were cleaned
up in __exit_signal()->posix_cpu_timers_exit{,_group}(), so we can just
return from irq.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] posix-timers: fix cleanup_timers() and run_posix_cpu_timers() races

1. cleanup_timers() sets timer->task = NULL under tasklist + ->sighand locks.
   That means that this code in posix_cpu_timer_del() and posix_cpu_timer_set()

    lock_timer(timer);
if (timer->task == NULL)
return;
read_lock(tasklist);
put_task_struct(timer->task)

   is racy. With this patch timer->task modified and accounted only under
   timer->it_lock. Sadly, this means that dead task_struct won't be freed
   until timer deleted or armed.

2. run_posix_cpu_timers() collects expired timers into local list under
   tasklist + ->sighand again. That means that posix_cpu_timer_del()
   should check timer->it.cpu.firing under these locks too.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Merge /linux/kernel/git/jejb/scsi-for-linus-2.6

[PATCH] ib: mthca: Always re-arm EQs in mthca_tavor_interrupt()

We should always re-arm an event queue's interrupt in
mthca_tavor_interrupt() if the corresponding bit is set in the event cause
register (ECR), even if we didn't find any entries in the EQ.  If we don't,
then there's a window where we miss an EQ entry and then get stuck because
we don't get another EQ event.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] inotify/idr leak fix

Fix a bug which was reported and diagnosed by
Stefan Jones <stefan.jones@churchillrandoms.co.uk>

IDR trees include a cache of idr_layer objects.  There's no way to destroy
this cache, so when we discard an overall idr tree we end up leaking some
memory.

Add and use idr_destroy() for this.  v9fs and infiniband also need to use
idr_destroy() to avoid leaks.

Or, we make the cache global, like radix_tree_preload().  Which is probably
better.  Later.

Cc: Eric Van Hensbergen <ericvh@ericvh.myip.org>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: Robert Love <rml@novell.com>
Cc: John McCutchan <ttb@tentacle.dhs.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] Kconfig: saa7134-dvb should not select cx22702

On 2005-05-01, Gerd Knorr sent in a patch to add cx22702 to cx88-dvb:

 [PATCH] dvb: cx22702 frontend driver update
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=9990d744bea7d28e83c420e2c9d524c7a8a2d136

...but as we can see, the Kconfig portion of his patch was incorrectly
applied to saa7134-dvb instead of cx88-dvb.

On 2005-06-24, Adrian bunk fixed cx88-dvb:

 [PATCH] VIDEO_CX88_DVB must select DVB_CX22702
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=d6988588e13616587aa879c2e0bd7cd811705e5d

...but we never removed the original patch from Gerd.

This patch sets things straight:

saa7134-dvb should not select cx22702

Signed-off-by: Michael Krufky <mkrufky@m1k.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] SELinux: handle sel_make_bools() failure in selinuxfs

This patch fixes error handling in sel_make_bools(), where currently we'd
get a memory leak via security_get_bools() and try to kfree() the wrong
pointer if called again.

Signed-off-by: James Morris <jmorris@namei.org>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] selinux: Fix NULL deref in policydb_destroy

This patch fixes a possible NULL dereference in policydb_destroy, where
p->type_attr_map can be NULL if policydb_destroy is called to clean up a
partially loaded policy upon an error during policy load.  Please apply.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] aio syscalls are not checked by lsm

Another case of missing call to security_file_permission: aio functions
(namely, io_submit) does not check credentials with security modules.

Below is the simple patch to the problem.  It seems that it is enough to
check for rights at the request submission time.

Signed-off-by: Kostik Belousov <kostikbel@gmail.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] kernel-parameters cleanup

Fix typos & trailing whitespace.
Add blank lines in a few places.
Remove "AM53C974=" option:  driver does not exist.
Restrict to < 80 columns in most places (but don't split formatted
  command-line arguments).
Add a few option arguments for completeness.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

cardbus: limit IO windows to 256 bytes

That's what we've always historically done, and bigger windows seem to
confuse some cardbus bridges. Or something.

Alan reports that this makes the ThinkPad 600x series work properly
again: the 4kB IO window for some reason made IDE DMA not work, which
makes IDE painfully slow even if it works after DMA timeouts.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Merge /pub/scm/linux/kernel/git/acme/net-2.6

Posix timers: limit number of timers firing at once

Bursty timers aren't good for anybody, very much including latency for
other programs when we trigger lots of timers in interrupt context.  So
set a random limit, after which we'll handle the rest on the next timer
tick.

Noted by Oleg Nesterov <oleg@tv-sign.ru>

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[NEIGH] Fix timer leak in neigh_changeaddr

neigh_changeaddr attempts to delete neighbour timers without setting
nud_state.  This doesn't work because the timer may have already fired
when we acquire the write lock in neigh_changeaddr.  The result is that
the timer may keep firing for quite a while until the entry reaches
NEIGH_FAILED.

It should be setting the nud_state straight away so that if the timer
has already fired it can simply exit once we relinquish the lock.

In fact, this whole function is simply duplicating the logic in
neigh_ifdown which in turn is already doing the right thing when
it comes to deleting timers and setting nud_state.

So all we have to do is take that code out and put it into a common
function and make both neigh_changeaddr and neigh_ifdown call it.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[NEIGH] Fix add_timer race in neigh_add_timer

neigh_add_timer cannot use add_timer unconditionally.  The reason is that
by the time it has obtained the write lock someone else (e.g., neigh_update)
could have already added a new timer.

So it should only use mod_timer and deal with its return value accordingly.

This bug would have led to rare neighbour cache entry leaks.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[NEIGH] Print stack trace in neigh_add_timer

Stack traces are very helpful in determining the exact nature of a bug.
So let's print a stack trace when the timer is added twice.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[PATCH] alpha: additional smp barriers

As stated in Documentation/atomic_ops.txt, atomic functions
returning values must have the memory barriers both before and after
the operation.

Thanks to DaveM for pointing that out.

Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[AX.25]: Fix signed char bug

On architectures where the char type defaults to unsigned some of the
arithmetic in the AX.25 stack to fail, resulting in some packets being dropped
on receive.

Credits for tracking this down and the original patch to
Bob Brose N0QBJ <linuxhams@n0qbj-11.ampr.org>.

Signed-off-by: Ralf Baechle DL5RB <ralf@linux-mips.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[SK_BUFF]: ipvs_property field must be copied

IPVS used flag NFC_IPVS_PROPERTY in nfcache but as now nfcache was removed the
new flag 'ipvs_property' still needs to be copied. This patch should be
included in 2.6.14.

Further comments from Harald Welte:

Sorry, seems like the bug was introduced by me.

Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

[PATCH] typo fix in last cpufreq powernow patch

Not sure how it slipped by, but here's a trivial typo fix for powernow.

Signed-off-by: Chris Wright <chrisw@osdl.org>
[ It's "nurter" backwards.. Maybe we have a hillbilly The Shining fan? ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] Call exit_itimers from do_exit, not __exit_signal

When I originally moved exit_itimers into __exit_signal, that was the only
place where we could reliably know it was the last thread in the group
dying, without races.  Since then we've gotten the signal_struct.live
counter, and do_exit can reliably do group-wide cleanup work.

This patch moves the call to do_exit, where it's made without locks.  This
avoids the deadlock issues that the old __exit_signal code's comment talks
about, and the one that Oleg found recently with process CPU timers.

[ This replaces e03d13e985d48ac4885382c9e3b1510c78bd047f, which is why
  it was just reverted. ]

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Revert "Fix cpu timers exit deadlock and races"

Revert commit e03d13e985d48ac4885382c9e3b1510c78bd047f, to be replaced
by a much nicer fix from Roland.

[PATCH] cpufreq: fix pending powernow timer stuck condition

AMD recently discovered that on some hardware, there is a race condition
possible when a C-state change request goes onto the bus at the same
time as a P-state change request.

Both requests happen, but the southbridge hardware only acknowledges the
C-state change.  The PowerNow! driver is then stuck in a loop, waiting
for the P-state change acknowledgement.  The driver eventually times
out, but can no longer perform P-state changes.

It turns out the solution is to resend the P-state change, which the
southbridge will acknowledge normally.

Thanks to Johannes Winkelmann for reporting this and testing the fix.

Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

[PATCH] ppc64: Fix typo bug in iSeries hash code

This fixes a stupid typo bug in the iSeries hash table code.

When we place a hash PTE in the secondary bucket, instead of setting the
SECONDARY flag bit, as we should, we (redundantly) set the VALID flag.

This was introduced with the patch abolishing bitfields from the hash
table code.  Mea culpa, oops.  It hasn't been noticed until now because
in practice we don't hit the secondary bucket terribly often.

Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Merge master.kernel.org:/home/rmk/linux-2.6-arm

Merge /pub/scm/linux/kernel/git/acme/net-2.6