The runtime expansion of nr_irqs does not take into account that
bitmap_find_next_zero_area() returns "start" + size in case the search
for an matching zero area fails. That results in a start value which
can be completely off and is not covered by the following
expand_nr_irqs() and possibly outside of the absolute limit. But we
use it without further checking.

Use IRQ_BITMAP_BITS as the limit for the bitmap search and expand
nr_irqs when the start bit is beyond nr_irqs. So start is always
pointing to the correct area in the bitmap. nr_irqs is just the limit
for irq enumerations, not the real limit for the irq space.

[ tglx: Let irq_expand_nr_irqs() take the new upper end so we do not
  	expand nr_irqs more than necessary. Made changelog readable ]
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
LKML-Reference: <4D6014F9.8040605@kernel.org>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Most of the managing functions get the irq descriptor and lock it -
either with or without buslock. Instead of open coding this over and
over provide a common function to do that.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Force the usage of wrappers by another nasty CPP substitution.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Keep status in sync until all abusers are fixed.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

The irq_desc.status field will either go away or renamed to
settings. Anyway we need to maintain compatibility to avoid breaking
the world and some more. While moving bits into the core, I need to
avoid that I use any of the still existing IRQ_ bits in the core code
by typos. So that file will hold the inline wrappers and some nasty
CPP tricks to break the build when typoed.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

We face more and more the requirement to expand nr_irqs at
runtime. The reason are irq expanders which can not be detected in the
early boot stage. So we speculate nr_irqs to have enough room. Further
Xen needs extra irq numbers and we really want to avoid adding more
"detection" code into the early boot. There is no real good reason why
we need to limit nr_irqs at early boot.

Allow the allocation code to expand nr_irqs. We have already 8k extra
number space in the allocation bitmap, so lets use it.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Lars-Peter Clausen pointed out:

   I stumbled upon this while looking through the existing archs using
   SPARSE_IRQ.  Even with SPARSE_IRQ the NR_IRQS is still the upper
   limit for the number of IRQs.

   Both PXA and MMP set NR_IRQS to IRQ_BOARD_START, with
   IRQ_BOARD_START being the number of IRQs used by the core.

   In various machine files the nr_irqs field of the ARM machine
   defintion struct is then set to "IRQ_BOARD_START + NR_BOARD_IRQS".

   As a result "nr_irqs" will greater then NR_IRQS which then again
   causes the "allocated_irqs" bitmap in the core irq code to be
   accessed beyond its size overwriting unrelated data.

The core code really misses a sanity check there.

This went unnoticed so far as by chance the compiler/linker places
data behind that bitmap which gets initialized later on those affected
platforms.

So the obvious fix would be to add a sanity check in early_irq_init()
and break all affected platforms. Though that check wants to be
backported to stable as well, which will require to fix all known
problematic platforms and probably some more yet not known ones as
well. Lots of churn.

A way simpler solution is to allocate a slightly larger bitmap and
avoid the whole churn w/o breaking anything. Add a few warnings when
an arch returns utter crap.
Reported-by: NLars-Peter Clausen <lars@metafoo.de>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Cc: stable@kernel.org # .37
Cc: Haojian Zhuang <haojian.zhuang@marvell.com>
Cc: Eric Miao <eric.y.miao@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>

CONFIG_KSTAT_IRQS_ONDEMAND does not exist. It's not worth to implement
it. Use sparse irqs if you care about memory consumption of the
interrupt layer.

Found by undertaker: http://vamos.informatik.uni-erlangen.de/trac/undertakerSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

Use modern per_cpu API to increment {soft|hard}irq counters, and use
per_cpu allocation for (struct irq_desc)->kstats_irq instead of an array.

This gives better SMP/NUMA locality and saves few instructions per irq.

With small nr_cpuids values (8 for example), kstats_irq was a small array
(less than L1_CACHE_BYTES), potentially source of false sharing.

In the !CONFIG_SPARSE_IRQ case, remove the huge, NUMA/cache unfriendly
kstat_irqs_all[NR_IRQS][NR_CPUS] array.

Note: we still populate kstats_irq for all possible irqs in
early_irq_init().  We probably could use on-demand allocations.  (Code
included in alloc_descs()).  Problem is not all IRQS are used with a prior
alloc_descs() call.

kstat_irqs_this_cpu() is not used anymore, remove it.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: NChristoph Lameter <cl@linux.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In /proc/stat, the number of per-IRQ event is shown by making a sum each
irq's events on all cpus.  But we can make use of kstat_irqs().

kstat_irqs() do the same calculation, If !CONFIG_GENERIC_HARDIRQ,
it's not a big cost. (Both of the number of cpus and irqs are small.)

If a system is very big and CONFIG_GENERIC_HARDIRQ, it does

	for_each_irq()
		for_each_cpu()
			- look up a radix tree
			- read desc->irq_stat[cpu]
This seems not efficient. This patch adds kstat_irqs() for
CONFIG_GENRIC_HARDIRQ and change the calculation as

	for_each_irq()
		look up radix tree
		for_each_cpu()
			- read desc->irq_stat[cpu]

This reduces cost.

A test on (4096cpusp, 256 nodes, 4592 irqs) host (by Jack Steiner)

%time cat /proc/stat > /dev/null

Before Patch:	 2.459 sec
After Patch :	  .561 sec

[akpm@linux-foundation.org: unexport kstat_irqs, coding-style tweaks]
[akpm@linux-foundation.org: fix unused variable 'per_irq_sum']
Signed-off-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Tested-by: NJack Steiner <steiner@sgi.com>
Acked-by: NJack Steiner <steiner@sgi.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This option can be set to verify the full conversion to the new chip
functions. Fix the fallout of the patch rework, so the core code
compiles and works with it.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

The allocator functions are now called outside of preempt disabled
regions. Switch to GFP_KERNEL.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

No callers from atomic regions. 
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

The move_irq_desc() function was only used due to the problem that the
allocator did not free the old descriptors. So the descriptors had to
be moved in create_irq_nr(). That's history.

The code would have never been able to move active interrupt
descriptors on affinity settings. That can be done in a completely
different way w/o all this horror.

Remove all of it.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

Use the cleanup functions of the dynamic allocator. No need to have
separate implementations.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

This function should have not been there in the first place.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

sparse irq sets up NR_IRQS_LEGACY irq descriptors and archs then go
ahead and allocate more.

Use the unused return value of arch_probe_nr_irqs() to let the
architecture return the number of early allocations. Fix up all users.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

Make irq_to_desc_alloc_node() a wrapper around the new allocator.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

Mark a range of interrupts as allocated. In the SPARSE_IRQ=n case we
need this to update the bitmap for the legacy irqs so the enumerator
via irq_get_next_irq() works.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Use the allocator bitmap to lookup active interrupts.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

/proc/irq never removes any entries, but when irq descriptors can be
freed for real this is necessary. Otherwise we'd reference a freed
descriptor in /proc/irq/N
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

The current sparse_irq allocator has several short comings due to
failures in the design or the lack of it:

 - Requires iteration over the number of active irqs to find a free slot
   (Some architectures have grown their own workarounds for this)
 - Removal of entries is not possible
 - Racy between create_irq_nr and destroy_irq (plugged by horrible
   callbacks)
 - Migration of active irq descriptors is not possible
 - No bulk allocation of irq ranges
 - Sprinkeled irq_desc references all over the place outside of kernel/irq/
   (The previous chip functions series is addressing this issue)

Implement a sane allocator which fixes the above short comings (though
migration of active descriptors needs a full tree wide cleanup of the
direct and mostly unlocked access to irq_desc).

The new allocator still uses a radix_tree, but uses a bitmap for
keeping track of allocated irq numbers. That allows:

 - Fast lookup of a free slot
 - Allows the removal of descriptors
 - Prevents the create/destroy race
 - Bulk allocation of consecutive irq ranges
 - Basic design is ready for migration of life descriptors after
   further cleanups

The bitmap is also used in the SPARSE_IRQ=n case for lookup and
raceless (de)allocation of irq numbers. So it removes the requirement
for looping through the descriptor array to find slots.

Right now it uses sparse_irq_lock to protect the bitmap and the radix
tree, but after cleaning up all users we should be able convert that
to a mutex and to switch the radix_tree and decriptor allocations to
GFP_KERNEL.

[ Folded in a bugfix from Yinghai Lu ]
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

Arch code sets it's own irq_desc.status flags right after boot and for
dynamically allocated interrupts. That might involve iterating over a
huge array.

Allow ARCH_IRQ_INIT_FLAGS to set separate flags aside of IRQ_DISABLED
which is the default.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

The statistics accessor is only used by proc/stats and
show_interrupts(). Both are compiled in.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

early_init_irq_lock_class() is called way before anything touches the
irq descriptors. In case of SPARSE_IRQ=y this is a NOP operation
because the radix tree is empty at this point. For the SPARSE_IRQ=n
case it's sufficient to set the lock class in early_init_irq(). 
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>

kernel/irq/handle.c has become a dumpground for random code in random
order. Split out the irq descriptor management and the dummy irq_chip
implementation into separate files. Cleanup the include maze while at
it.

No code change.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NIngo Molnar <mingo@elte.hu>