To handle control messaging in contexts other than the
IRQ handler's setup() callback, a bit of automagic needed
to be removed; and some minor bugs fixed.

This is the addition needed to make my previous patch for the UHCI
controller work properly with PM suspend/resume.

Paul:

Okay, I think this patch ought to do the trick.  I modified the PM
suspend/resume code so that on buggy motherboards like yours the suspend
routine really does a reset, while on normal motherboards the resume
routine really does a resume.  I haven't tried that part out because,
truth to tell, I'm a little scared of doing an APM/ACPI suspend.  Not long
ago I walked away from my computer for about a half-hour, leaving 2.5.69
running.  When I got back the screen was blank and the machine was totally
non-responsive.

I changed the delays in reset_hc() to use schedule_timeout() rather than
wait_ms(), which should make it more friendly.

Finally, I put the USBCMD_FGR back into wakeup_hc().  The reason for it is
now evident: a wakeup might be the result of a system-initiated event
as opposed to something requested by a device.

I think I'd better work on my plausible deniability...

This resolves some problems in the endpoint disable
patch I sent recently:

  - wasn't un-halting some endpoints
  - was clobbering ep0 setup in some cases.

into kroah.com:/home/greg/linux/BK/gregkh-2.5

into kernel.bkbits.net:/home/davem/net-2.5

into home.transmeta.com:/home/torvalds/v2.5/linux

into home.transmeta.com:/home/torvalds/v2.5/linux

into tetrachloride.(none):/mnt/raid/src/kernel/2.5/agpgart

The calling task must have a valid reclaim_state when running page
reclaim.  But I had forgotten about shrink_all_memory().

into kroah.com:/home/linux/linux/BK/pci-2.5

Move the PCI core to start using these, enabling proper reference counting
on struct pci_dev.

into home.transmeta.com:/home/torvalds/v2.5/linux

From Jeff Wiedemeier and Richard Henderson

into nuts.ninka.net:/home/davem/src/BK/net-2.5

into tetrachloride.(none):/mnt/raid/src/kernel/2.5/agpgart

Adjusts the IPTables targets to handle non-linear packets.
ipt_ULOG untested.

Adjusts the IPTables matches to handle non-linear packets.
Untested: ipt_ah and ipt_esp.

Adjusts the IPTables core to handle non-linear packets.  Extensions
done separately in next patch.  Also, comments about when to set
"me" field in struct ipt_match, ipt_table and ipt_target are
corrected.

into kroah.com:/home/greg/linux/BK/gregkh-2.5