you cannot use usb_bulk_msg() on buffers on the stack.

the whiteheat driver in two places does DMA to the stack by usb_bulk_msg().

There's a properly assigned vendor/product ID pair for this; use it.
Use new boolean CONFIG_* symbols.  For PXA/IXP, use the first two
endpoints.

Autoconfigure ep0 maxpacket size, and simplify configuration
for device power consumption.  Use new boolean CONFIG_* symbols.
SuperH UDC support.

Use new boolean CONFIG_* symbols.

This updates and simplifies the kernel config for the gadget
drivers.  It also adds build support for three new drivers:
two controllers (goku_udc, pxa2xx_udc) and one gadget driver
(file_storage).

(From Alan Stern.)

The File-backed Storage Gadget provides support for the USB Mass Storage
protocol.  It causes a gadget to appear to the host as a USB disk drive,
using a regular file or a block device to provide the necessary storage
somewhat like the "loop" device.  The driver supports up to 8 Logical
Units, appearing to the host as multiple disks, each configurable
independently for read-write or read-only access.  The settings for the
Logical Units are available through sysfs attribute files.

The driver has a testing version, selectable by a kernel configuration
option.  The test version includes many additional module parameters,
allowing the USB protocol, subclass, vendor, product, and release values
to be set.  It also has an option to emulate removable media, and the
sysfs interface allows the backing files and the read-only settings to be
switched on-the-fly.  In both versions several logging and debugging
levels are available.  They provide a nice way of seeing exactly what
commands a USB host is sending to the gadget.

A variety of fixes:

    - Resolves some problems with DMA chaining.  It should stream
      a lot better now; but not all the funky cases are handled yet.

    - Now "use_dma_chaining" is a module parameter not a #define.
      It also defaults to false.  Some gadget drivers will work fine
      with this enabled, getting some IRQ reduction and increased
      I/O parallelism (given deep I/O queues); others won't.

    - Handle isochronous transfers (from Mark Huang at Broadcom)

    - Some of the chiprev 0100 workarounds weren't quite right.
      Neither were buffer allocations on dma-incoherent systems.

    - Handle bulk endpoint halts better, for file_storage gadget driver.

    - Handle the hardware device status bits better:
        * selfpowered by default, clearable with the API;
        * remote wakeup disabled by default, host must enable it.

    - Fix a shutdown problem seen in some statically linked configs.

Needed to use the new file_storage gadget without disabling DMA.

Add back the old i82489DX bits to use timer scaling for the old
non-integrated APIC setup. 

It's possible these bits don't need to be set on i82489DXs,
but not having this HW for testing I elected to maintain
the old behaviour on these old machines.

We used to write fields that were marked RESERVED and that
are apparently some old stale timer base. Stop doing that.

Verified with Mikael Pettersson, and confirmed to fix ACPI
boot-time lockups for a few people. 

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

ppc64 doesn't use drivers/Kconfig (it should) so it needs to include i2c by
hand.

From: Anton Blanchard <anton@samba.org>

We were missing the sched_balance_exec call.  Could explain some NUMA
scheduling weirdness we were seeing.

From: "H. Peter Anvin" <hpa@zytor.com>

This is purely a doc patch saying RAID-6 support is available in
mdadm-1.5.0 and the patch is no longer necessary.

From: "H. Peter Anvin" <hpa@zytor.com>

As expected, when it hit mainline I started getting real bug reports... 
the attached patch does the following:

- Fixes a few x86-64 specific bugs;

- Removes MMX and SSE-1 from x86-64 (if we have x86-64 we have SSE-2);

- Slightly astracts the integer code to make it easier to add
  architecture-specific optimizations later (e.g.  IA64 once gcc gets
  better IA64 intrinsics support);

- Resurrects the user-space testbench, and makes it not output the known
  false positive of the D+Q case (D+Q is equivalent to a RAID-5 recovery,
  so I didn't implement it in the user-space testbench.)

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

RFC 2461 requires that the source address of Neighbor Discovery messages
is an address assigned to the sending interface.

Duplicate Address Detection should also be interface specific. We don't,
for example, want a node to DoS itself just because it has two interfaces
on the same link and both happen to listen to the same multicast group. If
there is a true duplicate on the link, the interface doing DAD will notice
it anyway.

The attached patch adds a 'strict' parameter to ip6_chk_addr() and
ip6_get_ifaddr() to allow link-local protocols like ND and DAD to do
strict address checks even on addresses with greater scope than
link-local.

> Paul Mackerras:
>   o sort exception tables

And as more proof that Paul is leaving us ppc32 folks, *sniff*, the
following is needed for PPC32 to compile:

Here's a novel idea, check the CPU has machine check capabilities
before we start polling registers.

I was wondering why my VIA C3 was starting this.  Who knows it may solve
some of the random crashes I saw there.

This updates / adds documentation for the bttv, saa7134 and cx88
video4linux drivers.

trivial one: uses __init and thus needs linux/init.h

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

       we take the appropriate action. (throw the byte away, reset
       byte counter, return NAK if acking, and complain).

       psmouse_rate= option.

into kernel.bkbits.net:/home/davem/sparc-2.6

Do not pass a NULL mm to mmdrop().

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