Oh, duh.  The race is obvious.  Sorry for the confusion there.

The BUG_ON's were useful for debugging, since they trigger on a lot of
errors, but they _also_ trigger on some unlikely (but valid) races.

So just remove them - just fall through to the regular exit code after
core-dumping (which does everything right).
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

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

Doing access control checks with rq_lock held can cause deadlock when
audit messages are created (via printk or audit infrastructure) which
trigger a wakeup and deadlock, as noted by both SELinux and SubDomain
folks.  This patch will let the security checks happen w/out lock held,
then re-sample the p->policy in case it was raced. 

Originally from John Johansen <johansen@immunix.com>, reworked by me.
AFAIK, this version drew no objections from Ingo or Andrea. 

From: John Johansen <johansen@immunix.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

* writes to PCI config space are non-functional since 2.4.21
* reads of full PCI config space are allowed for normal users
* I'm not aware of any applications using this interface
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

* make __ide_dma_verbose() void and drop "__" prefix
* ide_dma_verbose() is always available now
* use it instead of ide_hwif_t->ide_dma_verbose
* sgiioc4.c version reported wrong mode
* icside.c version repeated info given by ->ide_dma_check()
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

* add hwif->sg_mapped flag
* add idescsi_map_sg() converting scsi_cmd->sg into
  hwif->sg_table (this removes need for rq->bio)
* remove code (de)allocating rq->bio
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

ide_build_sglist() can be now used for REQ_DRIVE_TASKFILE requests.
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

* make Etrax ide.c, icside.c and ide-dma.c use ide_map_sg()
* use one sg for REQ_DRIVE_TASKFILE requests in ide-dma.c
  (no reason for 128 sectors per sg limit)
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

Just use ide_dma_[raw_]build_sglist() from ide-dma.c.
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

This marks sets change_bits to all ones when bringing up a hub, since
not all hubs seem to send change events for devices that were plugged in
when the hub was reset. 

David Miller confirms this fixes his boot-time lost keyboard/mouse
problem

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

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

into kroah.com:/home/greg/linux/BK/usb-2.6

This updates debug messages, declaring that labels can be constant strings
and changing some old-school dbg() calls to driver model dev_dbg().
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

The recent patch to scrub out ep0 state earlier (to get rid of some of
the enumeration problems that started with about 2.6.6) requires EHCI
to handle endpoint_disable() calls in a slightly different context.

This makes those calls work when an endpoint's QH may still be on the
async schedule, rather than already unlinked.  (The QH stays on the async
schedule for a few milliseconds after it's empty, since it's routine to
issue another request almost immediately.)
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

This has minor PCI tweaks for the EHCI driver:

  - If the (nonfunctional) AMD 8111 controller shows up, ignore it.
    (Patch from Matt Dharm, with minor coding style tweaks).

  - Delete some code that interprets an EHCI capability code; it
    should be a PCI capability instead.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

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

New and Improved! "Woozy Numbat!"

Pick up the phone in the next two minutes and phone in your
order, and we'll throw in a second copy of Woozy Numbat,
ABSOLUTELY FREE! You too could start a Numbat farm!

lib/kobject_uevent.c:39: warning: `action_to_string' defined but not used
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

From: Keshavamurthy Anil S <anil.s.keshavamurthy@intel.com>

Remove cpu_run_sbin_hotplug() - use kobject_hotplug() instead.
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Make kobject_hotplug() work even if the kobject's kset doesn't implement any
hotplug_ops.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

kobject_set_name takes a printf style argument list. There are many
callers that pass only one string, if this string contained a '%' character
than bad things would happen.  The fix is simple.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Here is a better fix (thanks Greg) that allows long names for character
device objects.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Signed-off-by: cw@f00f.org
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

into kroah.com:/home/greg/linux/BK/driver-2.6

arch/sh/tools/machgen.sh is no longer used as it has been replaced by
arch/sh/tools/gen-mach-types
Signed-off-by: Tom Rini <trini@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Uses msleep() instead of schedule_timeout() to guarantee the task delays
as expected.  Two of the changes are reworks of previous msleep() calls
which unnecessarily added a jiffy to the parameter.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

there was an uneeded packed attribute for a data structure.
Signed-off-by: Oliver Neukum <oliver@neukum.name>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

kaweth used its own synchronisation superseded by usb_unlink_urb().
Signed-off-by: Oliver Neukum <oliver@neukum.name>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

This tweaks the USB input driver support, notably fixing a botched
dependency that makes all the USB drivers appear strangely in Kconfig.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

The following compile error seems to come from Linus' tree:


<--  snip  -->

...
  CC      drivers/usb/serial/bus.o
  CC      drivers/usb/serial/console.o
drivers/usb/serial/console.c: In function `usb_console_write':
drivers/usb/serial/console.c:221: warning: passing arg 3 of pointer to function makes integer from pointer without a cast
drivers/usb/serial/console.c:221: error: too many arguments to function
drivers/usb/serial/console.c:223: warning: passing arg 3 of `usb_serial_generic_write' makes integer from pointer without a cast
drivers/usb/serial/console.c:223: error: too many arguments to function `usb_serial_generic_write'
make[3]: *** [drivers/usb/serial/console.o] Error 1

<--  snip  -->


This was caused by the changed "write" in usb_serial_device_type.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

This patch implements the Windows scheme for USB device initialization.
It also incorporates the change recently posted by David to scrub the
endpoint state following a SET-ADDRESS.  Other noteworthy changes:

	There are two new module parameters to control whether the
	old scheme or the new one is used first and whether the other
	scheme is tried if the first one fails.  Default settings are
	to use the new scheme only.

	hub_set_address() returns 0 immediately if the device is already
	in the USB_STATE_ADDRESS state.

	On the first attempt to read the device descriptor the code
	uses a short 1-second timeout.  This ought to help prevent
	full-speed devices with an 8- or 16-byte maxpacket from
	slowing the procedure down by NAKing the unexpectedly early
	status stage of the transfer.

	For debugging, the ep0 maxpacket value is printed.  It might
	be a good idea to validate it rather than just believing the
	device -- although I haven't heard of any device providing
	an incorrect value other than 0.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Description: Use msleep_interruptible() instead of schedule_timeout()
so that the task delays as expected.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Maximilian Attems <janitor@sternwelten.at>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Description: Use msleep_interruptible() instead of schedule_timeout() to
guarantee the task delays as expected.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Maximilian Attems <janitor@sternwelten.at>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

.. broken by the rename.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This didn't show up in my "allmodconfig" tests, because the
console requires things to be built-in, not modules.

Posix timers preallocate siqueue structures during timer creation
and keep them for reuse.  This allocation happens in user context
with no locks held, however it's designated as an atomic allocation.
Loosen this restriction, and while we're at it let's do a bit of code
consolidation so signal sending uses same __sigqueue_alloc() helper.
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

missing #include (and placement cleanup)
Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Lindent security/security.c.
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Registering a security module can be a noisy operation, esp. when it
retries registration with the primary module.  Eliminate some noise, and
distinguish the return values for register_security so a module can tell
the difference between failure modes.
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>