This eliminates lock contention among IRQs as well as the need to
disable IRQs around idr_find, because there are no IRQ writers.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

- ehca_cq.nr_events is made an atomic_t, eliminating a lot of locking.
- The CQ is removed from the CQ idr first now to make sure no more
  completions are scheduled on that CQ. The "wait for all completions to
  end" code becomes much simpler this way.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

- In ehca_process_eq(), we're IRQ safe throughout the whole function, so we
  don't need another _irqsave in the middle of flight.

- take_over_work() is only called by comp_pool_callback(), so it can move
  into the same #ifdef block.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Since nonboot CPUs are now disabled after tasks and devices have been
frozen and the CPU hotplug infrastructure is used for this purpose, we need
special CPU hotplug notifications that will help the CPU-hotplug-aware
subsystems distinguish normal CPU hotplug events from CPU hotplug events
related to a system-wide suspend or resume operation in progress.  This
patch introduces such notifications and causes them to be used during
suspend and resume transitions.  It also changes all of the
CPU-hotplug-aware subsystems to take these notifications into consideration
(for now they are handled in the same way as the corresponding "normal"
ones).

[oleg@tv-sign.ru: cleanups]
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

eHCA scaling code must not depend on register_cpu_notifier() if
CONFIG_HOTPLUG_CPU is not set, so put all related code into #ifdefs.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

This patch fixes two issues reported by Roland Dreier and Christoph Hellwig:

- Mismatched sync/locking between completion handler and destroy cq We
  introduced a counter nr_events per cq to track number of irq events
  seen. This counter is incremented when an event queue entry is seen
  and decremented after completion handler has been called regardless
  if scaling code is active or not. Note that nr_callbacks tracks
  number of events assigned to a cpu and both counters can potentially
  diverge.

  The sync between running completion handler and destroy cq is done
  by using the global spin lock ehca_cq_idr_lock.

- Replace yield by wait_event on the counter above to become zero.
Signed-off-by: Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Allow users to en/disable scaling code when loading ib_ehca module,
rather than requiring the module to be rebuilt to change the setting.
Signed-off-by: Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Fix a race condition in find_next_cpu_online() and some other locking
issues in ehca scaling code.
Signed-off-by: Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Rework ehca interrupt handling to avoid/reduce missed irq events.
Signed-off-by: Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Percpu data is not freed on module unloading.

Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Christoph Raisch <raisch@de.ibm.com>
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

The lock is taken with _irqsave and hence must be released with
_irqrestore on all paths.

Signed-off-by Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Here is a patch for ehca to use proper flag, ie. GFP_ATOMIC
resp. GFP_KERNEL, when calling get_zeroed_page() to prevent "Bug:
scheduling while atomic...". This error does not cause a kernel panic
but makes ipoib un-usable afterwards.  It is reproducible on
2.6.20-rc4 if one does ifconfig down during a flood ping test.  I have
not observed this error in earlier releases incl. 2.6.20-rc1.

This error occurs when a qp event/irq is received and ehca event
handler allocates a control block/page to obtain HCA error data block.
Use of GFP_ATOMIC when in interrupt context prevents this issue.

Signed-off-by Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Assure 4K alignment for firmware control blocks in 64K page mode,
because kzalloc()'s result address might not be 4K aligned if 64K
pages are enabled. Thus, we introduce wrappers called
ehca_{alloc,free}_fw_ctrlblock(), which use a slab cache for objects
with 4K length and 4K alignment in order to alloc/free firmware
control blocks in 64K page mode. In 4K page mode those wrappers just
are defines of get_zeroed_page() and free_page().
Signed-off-by: Hoang-Nam Nguyen <hnguyen@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Maintain a per-CPU global "struct pt_regs *" variable which can be used instead
of passing regs around manually through all ~1800 interrupt handlers in the
Linux kernel.

The regs pointer is used in few places, but it potentially costs both stack
space and code to pass it around.  On the FRV arch, removing the regs parameter
from all the genirq function results in a 20% speed up of the IRQ exit path
(ie: from leaving timer_interrupt() to leaving do_IRQ()).

Where appropriate, an arch may override the generic storage facility and do
something different with the variable.  On FRV, for instance, the address is
maintained in GR28 at all times inside the kernel as part of general exception
handling.

Having looked over the code, it appears that the parameter may be handed down
through up to twenty or so layers of functions.  Consider a USB character
device attached to a USB hub, attached to a USB controller that posts its
interrupts through a cascaded auxiliary interrupt controller.  A character
device driver may want to pass regs to the sysrq handler through the input
layer which adds another few layers of parameter passing.

I've build this code with allyesconfig for x86_64 and i386.  I've runtested the
main part of the code on FRV and i386, though I can't test most of the drivers.
I've also done partial conversion for powerpc and MIPS - these at least compile
with minimal configurations.

This will affect all archs.  Mostly the changes should be relatively easy.
Take do_IRQ(), store the regs pointer at the beginning, saving the old one:

	struct pt_regs *old_regs = set_irq_regs(regs);

And put the old one back at the end:

	set_irq_regs(old_regs);

Don't pass regs through to generic_handle_irq() or __do_IRQ().

In timer_interrupt(), this sort of change will be necessary:

	-	update_process_times(user_mode(regs));
	-	profile_tick(CPU_PROFILING, regs);
	+	update_process_times(user_mode(get_irq_regs()));
	+	profile_tick(CPU_PROFILING);

I'd like to move update_process_times()'s use of get_irq_regs() into itself,
except that i386, alone of the archs, uses something other than user_mode().

Some notes on the interrupt handling in the drivers:

 (*) input_dev() is now gone entirely.  The regs pointer is no longer stored in
     the input_dev struct.

 (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking.  It does
     something different depending on whether it's been supplied with a regs
     pointer or not.

 (*) Various IRQ handler function pointers have been moved to type
     irq_handler_t.
Signed-Off-By: David Howells <dhowells@redhat.com>
(cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)

Add a driver for IBM GX bus InfiniBand adapters, which are usable with
some pSeries/System p systems.

Signed-off-by: Heiko J Schick <schickhj.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>