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

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

Bingo...  edd_devices[] was too short.  When we keep more than 6
signatures, it overruns the end.  Also, I rewrote edd_num_devices to be
clearer about its goal. 

This patch is necessary even after the last edd.S patch was reverted.

Thanks to Christian Kujau for testing and narrowing it down enough to
debug this. 

It still doesn't explain why Christian's BIOS reports more devices than
he has, that's still up in the air, so don't re-apply the edd.S patch
just reverted. 
Signed-off-by: Matt Domsch <Matt_Domsch@dell.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

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

Remove static flash device location definitions since they're now
obsolete.  Additionally, dynamically allocate driver private data
structures.

Use partitioning information passed in via the platform device
instead.  This means that individual machine files in
arch/arm/mach-sa1100 can contain their partitioning information.

- combine sa1100_locate_partitions with sa1100_mtd_probe.
- set device driver data in the probe method, and use this to get
  at our driver private information in the other methods.

This prepares the SA1100 flash driver to accept platform data via
the driver model.

This removes a bunch of SA1100 machine support files which have been
unable to be built for a while, due to missing out on updates.

Warning of removal had been posted to the ARM kernel mailing list,
and this cset represents those which did not attract any interest.

This adds platform_device-based partition and map information for
badge4, cerf, collie, h3600, hackkit and shannon SA1100 machine
types.

Patch from Ben Dooks

S3C2410/S3C2440 RTC driver using arch/arm/common/rtctime.c as
a base. Supports alarm, and periodic IRQ from 1 to 64Hz

Signed-off-by: Ben Dooks
Signed-off-by: Russell King

Patch from Dave Jiang

Remove nolonger used file from arch/arm/mach-iop3xx/mm-321.c

Signed-off-by: Dave Jiang
Signed-off-by: Russell King

Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Add the name of the bus and the driver to the hotplug event for
/sys/devices/*. With this addition, userspace knows what it can
expect from sysfs to show up, instead of waiting for a timeout
for devices without a bus.

  ACTION=add
  DEVPATH=/devices/pci0000:00/0000:00:1d.1/usb3/3-1
  SUBSYSTEM=usb
  SEQNUM=978
  PHYSDEVBUS=usb
  PHYSDEVDRIVER=usb
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Add the name of the device's driver to the hotplug environment of class
and block devices.

  ACTION=add
  DEVPATH=/block/sda
  SUBSYSTEM=block
  SEQNUM=986
  PHYSDEVPATH=/devices/pci0000:00/0000:00:1d.1/usb3/3-1/3-1:1.0/host0/target0:0:0/0:0:0:0
  PHYSDEVBUS=scsi
  PHYSDEVDRIVER=sd
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

o Do not release existing directory if the new directory happens to be a
  duplicate directory. Thanks to Kay Sievers for the testcase.
Signed-off-by: Maneesh Soni <maneesh@in.ibm.com>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

I am trying to fix the current ACPI container.ko and processor.ko to use
kobject_hotplug() for notification. For this I would be requiring the
KOBJ_ONLINE definitions to be added to kobject_action.
Signed-off-by: Anil Keshavamurty <anil.s.keshavamurthy@intel.com>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

On Nov 3, 2004, at 3:42 PM, Greg KH wrote:

|On Tue, Nov 02, 2004 at 10:03:34AM -0600, Maneesh Soni wrote:
||On Tue, Nov 02, 2004 at 02:46:58AM -0600, Milton Miller wrote:
|||sysfs_new_dirent returns ERR_PTR(-ENOMEM) if kmalloc fails but the callers
|||were expecting NULL.
||
||Thanks for spotting this. But as you said, I will prefer to change the callee.
||How about this patch?
..
||-		return -ENOMEM;
||+		return NULL;
|
|Actually, this needs to be a 0, not NULL, otherwise the compiler
|complains with a warning.  I've fixed it up and applied it.
|
|thanks,
|
|greg k-h

I wondered why greg thought the type was wrong.   After it was merged I
realized that the wrong function was changed.  Here's an attempt to fix
both errors.
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

This patch removes two fields from "struct device" that are duplicated
in "struct dev_pm_info":  power_state (which should probably vanish)
and "saved_state".  There were only two "real" uses of saved_state;
both are now switched over to use dev_pm_info.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

The more relaxed memory ordering od the __raw_{write,read}b broke things
on ppc in the recent rivafb cleanups.

[ Uglyness alert: only the 8-bit accesses are more strongly serialized. 
  The 16- and 32-bit ones have always been relaxed. Gaah. ]
Signed-off-by: Guido Guenther <agx@sigxcpu.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

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

Patch from Nicolas Pitre

The declaration of generic_fls recently changed from
extern inline to static inline.

Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King

Patch from Ben Dooks

Add register definitions for channel hardware
sources, and update for s3c2440 extra bits.

(should be applied after #2226/1 and #2227/1)

Signed-off-by: Ben Dooks
Signed-off-by: Russell King

Patch from Ben Dooks

Added EXPORT_SYMBOL() to appropriate routines

(should be applied after Patch 2226/1)

Signed-off-by: Ben Dooks
Signed-off-by: Russell King

Patch from Ben Dooks

Add sysfs class and devices for s3c24xx-dma code, and
add simple pm suspend code to ensure all channels are
stopped before suspending

Signed-off-by: Ben Dooks
Signed-off-by: Russell King

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

The patch below fixes a problem with shutting down 802.3ad bonds on the 2.6
kernel.  Taking the interface down or removing the module causes a stack
dump if spinlock debugging is enabled.  This patch was generated from the
2.6.9 kernel.

This patch has been peer reviewed by our Linux software engineering team,
and the fix has been verified by our test labs.
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>

This patch undoes a change that we believe will impact performance
adversely, by creating possibly too long a delay between servicing
completions.  The comment pretty much explains it.  We need to call both
cleanup routines each pass through the loop, this time we have a comment
explaining why.
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>

Obviously pci_enable_device should be called after pci_restore_state.
Signed-off-by: Li Shaohua <shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>

This fixes a bug where netpoll can dereference a null ifa_list pointer when
not supplied an IP address at module load and the interface is up but no IP
is configured.

Bonus: unrelated netif_running cleanup

Signed-off by: Jeff Moyer <jmoyer@redhat.com>
Signed-off by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>

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

It seems to result in unexplained memory corruption. Matt is working on it.

Cset exclude: Matt_Domsch@dell.com[torvalds]|ChangeSet|20041020153622|50713