into nuts.davemloft.net:/disk1/BK/net-2.6

The vendor packet type 0xff is not handled at the moment. This patch
corrects this behaviour and also adapts the security filter for this
packet type.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

When the selecting of the alternate setting for the SCO audio support
fails, the hci_usb_disconnect() will dereference a NULL pointer. To
avoid this, the isoc_iface variable must be set before releasing the
interface and unset afterwards.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

The BPA 100/105 devices from Digianswer identify themself with the USB
class code for H:2 devices, but they are using quite a different host
transport protocol and so ignore them.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

into nuts.davemloft.net:/disk1/BK/net-2.6

Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>

into nuts.davemloft.net:/disk1/BK/net-2.6

Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

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

- [DVB] add legacy DishNetwork support to dvb core and stv0299, thanks to Jeremy Hall
- [DVB] mt352: major cleanup, support DVICO FusionHDTV DVB-T, thanks to Christopher Pascoe
Signed-off-by: Michael Hunold <hunold@linuxtv.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

- [DVB] dibusb: update documentation for the dibusb DVB driver
- [DVB] dibusb: major overhaul of the driver, including adding new vendor and product ids from clones
Signed-off-by: Michael Hunold <hunold@linuxtv.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

- [DVB] add new driver for the Terratec CinergyT2/qanu USB2 DVB-T receiver, thanks to Daniel Mack and Holger Waechtler
Signed-off-by: Michael Hunold <hunold@linuxtv.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

- [DVB] ttusb_dec: add new 2000T type model number
- [DVB] dvb_ca_en50221: properly zero out private pointer
- [DVB] dvb-bt8xx: properly use a mutex when starting/stopping the dma engine, remove card list, it's not need anymore
- [DVB] bt878: fix hex <=> dec typo (missing 0x prefix), mark remove bt878_remove() function __devexit_p
- [DVB] dvb-core: add dvb_net_debug module parameter, remove some dead code
Signed-off-by: Michael Hunold <hunold@linuxtv.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

- [DVB] av7110: switch from stupid DEB_xx() debug macros to saner dprintk() style debugging
- [DVB] av7110: fix dvb-ttpci ca write() polling
- [DVB] budget: switch from stupid DEB_xx() debug macros to saner dprintk() style debugging
Signed-off-by: Michael Hunold <hunold@linuxtv.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

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

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

All ARM binutils versions post 2.11.90 contains an extra "feature" which
interferes with the kernel in various ways - extra "mapping symbols"
in the ELF symbol table '$a', '$t' and '$d'.  This causes two problems:

1. Since '$a' symbols have the same value as function names, this
   causes anything which uses the kallsyms infrastructure to report
   wrong values.
2. programs which parse System.map do not expect symbols to start with
   '$'.
Signed-off-by: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>

===== kernel/module.c 1.120 vs edited =====

Here's a trivial patch fixing a small spelling error in init/Kconfig in
the description of MODULE_SRCVERSION_ALL
Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>

The following set of patches is based loosely on the patches that
Jean-Christophe Dubois came up with for 2.6.7.  Where as the original
patches added a number of casts to unsigned char, I went the route of
making the chars be explicitly signed.  I honestly don't know which
route is better to go down.  Doing this is the bulk of the patch.  Out
of the rest of the odds 'n ends is that on Solaris, Elf32_Word is a
ulong, which means all of the printf's are unhappy (uint format, ulong
arg) for most of the typedefs.
Signed-off-by: Tom Rini <trini@kernel.crashing.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>

Moving logic to scripts/gen_initramfs_list.sh make a nice cleanup in
usr/Makefile.
A new initramfs image will be generated if the initramfs_list file changes.
This patch also fixes the bug with make O=..
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>

Also remove unused MAX_DMA define.
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Should prevent bugs like the recent piix one in the future.
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

This patch adds pci_get_legacy_ide_irq() to the PCI layer for dealing
with PCI IDE chipsets that use the "legacy mode" IRQ routing, thus
violating the normal PCI routing.  The generic implementation provides
IRQ numbers 14 and 15, and it adds a ppc64 specific one with platform
callbacks so that a plaform can provide different IRQ numbers for "legacy"
IDE.  I only fixed the amd7xxx.c driver for now, I expect people using
this interface will slowly fix their drivers (I will fix via soon as I'll
need it too, and maybe a few others).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

From: Arjan van de Ven <arjan@fenrus.demon.nl>

ide-iops.c exports a few functions that either have no users or have no
users AND shouldn't be used by drivers; the patch below unexports them.

-EXPORT_SYMBOL(SELECT_INTERRUPT);
-EXPORT_SYMBOL(SELECT_MASK);
-EXPORT_SYMBOL(QUIRK_LIST);
-EXPORT_SYMBOL(ata_vlb_sync);

only used in the core ide code

-EXPORT_SYMBOL(ata_input_data);
-EXPORT_SYMBOL(ata_output_data);

drivers should (and do) use the hwif-> function pointer variant of these,
these functions are internal and used for setting the default hwif->
function pointers only. No need to export, in fact exporting is only asking
for accidents
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

into mars.ravnborg.org:/home/sam/bk/to-linus

From: Krzysztof Chmielewski <k.chmielewski@neostrada.pl>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Undecoded slave fixup is a oneliner patch to ide-probe to
recognize both of my Maxtor drives that appear to have the same
serial number, D3000000.

Signed-off-by: tabris@tabris.net
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Add missing ->data_phase assignment.
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

This patch removes netlink_sock_nr which is only used by a printk
statement in af_netlink.c.  Even that's only there if
NETLINK_REFCNT_DEBUG is defined.

If we were really looking for netlink refcnt bugs there are probably
better places to do that anyway.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

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

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

Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>

Optimize away the unconditional write to debug registers on signal delivery
path.  This is already done on x86_64.

We only need to write to dr7 if there is a breakpoint to re-enable, and 
MOVDR is a serializing instruction, which is expensive.  Getting rid of 
it gets a 33% faster signal delivery path (at least on Xeon - I didn't
test other CPUs, so your gain may vary).

[ Editors note: it's likely only that slow on Netburst.  Serializing is
  not that expensive, but it is likely that writing to %db7 invalidates
  the trace cache, which explains why it's so slow on Xeon - it's not
  just the op itself, it has to re-populate the cache all the time.
					--- Linus ]


Measured delta TSC for three paths on a 2.4GHz Xeon.

1) With unconditional write to dr7 :  800-1000 cycles
2) With conditional write to dr7   :  84-112 cycles
3) With unlikely write to dr7      :  84 cycles

Performance test using divzero microbenchmark (3 million divide by zeros):

With unconditional write:
   7.445 real / 6.136 system
   7.529 real / 6.482 system
   7.541 real / 5.974 system
   7.546 real / 6.217 system
   7.445 real / 6.167 system

With unlikely write:
   5.779 real / 4.518 system
   5.783 real / 4.591 system
   5.552 real / 4.569 system
   5.790 real / 4.528 system
   5.554 real / 4.382 system

That's about a 33% speedup - more than I expected; apparently getting rid
of the serializing instruction makes the do_signal path much faster.

Zachary Amsden (zach@vmware.com)

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

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

Clean up a really messy cross reference where a macro cls_set_class
in pkt_sched.h would point to a function tcf_set_class in cls_api.c
that uses a inlined function __cls_set_class in pkt_sched.h again.

Make tcf_set_class be cls_set_class since it was never used and
inline it as well.
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

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

tc_classiy has grown too big to be inlined, move it to sch_api.c
and export it.
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

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