The cache reaper currently tries to free all alien caches and all remote
per cpu pages in each pass of cache_reap.  For a machines with large number
of nodes (such as Altix) this may lead to sporadic delays of around ~10ms.
Interrupts are disabled while reclaiming creating unacceptable delays.

This patch changes that behavior by adding a per cpu reap_node variable.
Instead of attempting to free all caches, we free only one alien cache and
the per cpu pages from one remote node.  That reduces the time spend in
cache_reap.  However, doing so will lengthen the time it takes to
completely drain all remote per cpu pagesets and all alien caches.  The
time needed will grow with the number of nodes in the system.  All caches
are drained when they overflow their respective capacity.  So the drawback
here is only that a bit of memory may be wasted for awhile longer.

Details:

1. Rename drain_remote_pages to drain_node_pages to allow the specification
   of the node to drain of pcp pages.

2. Add additional functions init_reap_node, next_reap_node for NUMA
   that manage a per cpu reap_node counter.

3. Add a reap_alien function that reaps only from the current reap_node.

For us this seems to be a critical issue.  Holdoffs of an average of ~7ms
cause some HPC benchmarks to slow down significantly.  F.e.  NAS parallel
slows down dramatically.  NAS parallel has a 12-16 seconds runtime w/o rotor
compared to 5.8 secs with the rotor patches.  It gets down to 5.05 secs with
the additional interrupt holdoff reductions.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

We require that all archs implement atomic_cmpxchg(), for the generic
version of atomic_add_unless().
Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This fixes a typo introduced with 90f0094d.
Signed-off-by: Horst Hummel <horst.hummel@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

When pages are onlined, not only zone->present_pages but also
pgdat->node_present_pages should be refreshed.

This parameter is used to show information at
/sys/device/system/node/nodeX/meminfo via si_meminfo_node().

So, it shows strange value for MemUsed which is calculated
(node_present_pages - all zones free pages).
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

EDAC is still causing a few problems and the code is relatively green.  Mark
it as experimental until thing settle down.

Also, provide some documentation pointers in Kconfig help.
Signed-off-by: Tim Small <tim@buttersideup.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Currently the code tries up to spin_retry times to grab a lock using the cs
instruction.  The cs instruction has exclusive access to a memory region
and therefore invalidates the appropiate cache line of all other cpus.  If
there is contention on a lock this leads to cache line trashing.  This can
be avoided if we first check wether a cs instruction is likely to succeed
before the instruction gets actually executed.
Signed-off-by: Christian Ehrhardt <ehrhardt@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The kobject_put() can free the memory at *cmd, but cmd->lock points to a
persistent lock that is not freed with cmd.
Signed-off-by: Max Asbock <masbock@us.ibm.com>
Cc: Vernon Mauery <vernux@us.ibm.com>
Cc: Srihari Vijayaraghavan <sriharivijayaraghavan@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

If the process has already set PF_MALLOC and is already using
current->reclaim_state then do not try to reclaim memory from the zone.
This is set by kswapd and/or synchrononous global reclaim which will not
take it lightly if we zap the reclaim_state.
Signed-off-by: Christoph Lameter <clameter@sig.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

- wrong test for 'is this a BARRIER bio'
- not freeing on all possible paths.
- using r1_bio after freeing it.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

/usr/src/ctest/git/kernel/mm/rmap.c: In function `page_referenced_one':
/usr/src/ctest/git/kernel/mm/rmap.c:354: warning: implicit declaration of function `rwsem_is_locked'
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

arch/sh/Kconfig shouldn't source non-existing Kconfig files.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Fix some bugs in mtd/jffs2 on 64bit platform.

The MEMGETBADBLOCK/MEMSETBADBLOCK ioctl are not listed in compat_ioctl.h.

And some variables in jffs2 are declared as uint32_t but used to hold
size_t values.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Fix a lockup which was introduced during the conversion to the generic IRQ
framework.

Cc: Richard Henderson <rth@twiddle.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

ERR_PTR() is supposed to be passed a negative value.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

smi_data_buf_free() references dcdbas_pdev when calling
dma_free_coherent().  In dcdbas_exit(), smi_data_buf_free() is called after
platform_device_unregister(dcdbas_pdev).

This patch moves platform_device_unregister(dcdbas_pdev) after
smi_data_buf_free() in dcdbas_exit().
Signed-off-by: Doug Warzecha <Douglas_Warzecha@dell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Remove two early-development BUG_ONs from page_add_file_rmap.

The pfn_valid test (originally useful for checking that nobody passed an
artificial struct page) comes too late, since we already have the struct
page.

The PageAnon test (useful when anon was first distinguished from file rmap)
prevents ->nopage implementations from reusing ->mapping, which would
otherwise be available.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

    
Some people still haven't heared that fp in the kernel is forbidden.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

    
This is gluing the build of cirrusfb but really the mess that would need
cleaning and fixing is <video/vga.h> and <linux/vt_buffer.h> ...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

    
Found by make buildcheck.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

    
At times gcc will place bits of __exit functions into .rodata.  If
compiled into the kernle itself we used to discard .exit.text - but
not the bits left in .rodata.  While harmless this did at times result
in a large number of warnings.  So until gcc fixes this, discard
.exit.text at runtime.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

    
In case a particular system doesn't support highmem the runtime checks
will ensure nothing bad is going to happen.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

    
What:   Support for NEC DDB5074 and DDB5476 evaluation boards.
When:   June 2006
Why:    Board specific code doesn't build anymore since ~2.6.0 and no
        users have complained indicating there is no more need for these
        boards.  This should really be considered a last call.
Who:    Ralf Baechle <ralf@linux-mips.org>

* git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc-merge:
  powerpc: Fix various syscall/signal/swapcontext bugs
  [PATCH] powerpc: incorrect rmo_top handling in prom_init
  [PATCH] powerpc: Fix incorrect pud_ERROR() message
  [PATCH] powerpc: Expose SMT and L1 icache snoop userland features
  [PATCH] powerpc: Fix windfarm_pm112 not starting all control loops
  [PATCH] powerpc: Fix old g5 issues with windfarm
  powerpc32: Fix timebase synchronization on 32-bit powermacs
  powerpc: Turn off verbose debug output in powermac platform functions
  powerpc: Fix might-sleep warning in program check exception handler

The low level PCI DMA mapping functions should handle it in most cases.

This should fix problems with depleting the DMA zone early. The old
code used precious GFP_DMA memory in many cases where it was not needed.
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

ATI chipsets tend to generate double timer interrupts for the local APIC
timer when both the 8254 and the IO-APIC timer pins are enabled.  This is
because they route it to both and the result is anded together and the CPU
ends up processing it twice.

This patch changes check_timer to disable the 8254 routing for interrupt 0.

I think it would be safe on all chipsets actually (i tested it on a couple
and it worked everywhere) and Windows seems to do it in a similar way, but
to be conservative this patch only enables this mode on ATI (and adds
options to enable/disable too)

Ported over from a similar x86-64 change.

I reused the ACPI earlyquirk infrastructure for the ATI bridge check, but
tweaked it a bit to work even without ACPI.

Inspired by a patch from Chuck Ebbert, but redone.

Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
  [NET] compat ifconf: fix limits

* master.kernel.org:/pub/scm/linux/kernel/git/mchehab/v4l-dvb:
  Merge branch 'work-fixes'
  V4L/DVB (3413): Typos grab bag of the month
  V4L/DVB (3403): Workaround to fix initialization for Nexus CA
  Merge branch 'work-fixes'
  V4L/DVB (3395): Fixed Pinnacle 300i DVB-T support
  V4L/DVB (3399): ELSA EX-VISION 500TV: fix incorrect PCI subsystem ID
  V4L/DVB (3382): Fix stv0297 for qam128 on tt c1500 (saa7146)
  V4L/DVB (3300a): Removing personal email from DVB maintainers
  V4L/DVB (3385): Dvb: fix __init/__exit section references in av7110 driver
  V4L/DVB (3378): Restore power on defaults of tda9887 after tda8290 probe
  V4L/DVB (3354): Fix maximum for the saturation and contrast controls.
  V4L/DVB (3352): Cxusb: fix lgdt3303 naming
  V4L/DVB (3348): Fixed saa7134 ALSA initialization with multiple cards
  V4L/DVB (3347): Pinnacle PCTV 40i: add filtered Composite2 input
  V4L/DVB (3341): Upstream sync - make 2 structs static
  V4L/DVB (3340): Make a struct static
  V4L/DVB (3337): Drivers/media/dvb/frontends/mt312.c: cleanups
  V4L/DVB (3336): Bt8xx documentation authors fix

A recent change to compat. dev_ifconf() in fs/compat_ioctl.c
causes ifconf data to be truncated 1 entry too early when copying it
to userspace.  The correct amount of data (length) is returned,
but the final entry is empty (zero, not filled in).
The for-loop 'i' check should use <= to allow the final struct
ifreq32 to be copied.  I also used the ifconf-corruption program
in kernel bugzilla #4746 to make sure that this change does not
re-introduce the corruption.
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

While testing kexec and kdump we hit problems where the new kernel would
freeze or instantly reboot.  The easiest way to trigger it was to kexec a
kernel compiled for CONFIG_M586 on an athlon cpu.  Compiling for CONFIG_MK7
instead would work fine.

The patch fixes a few problems with the kexec inline asm.
Signed-off-by: Chris Mason <mason@suse.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Signed-off-by: Matt Mackall <mpm@selenic.com>
Tested-by: Anders K. Pedersen <akp@cohaesio.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

kmem_cache_init() incorrectly assumes that the cache_cache object will fit
in an order 0 allocation.  On very large systems, this is not true.  Change
the code to try larger order allocations if order 0 fails.
Signed-off-by: Jack Steiner <steiner@sgi.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Miscellaneous fixes related to accessing uninitialized variables or memory
that was already freed.
Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Cc: Eric Van Hensbergen <ericvh@ericvh.myip.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The x86_model calculation also applies for family 6. early_cpu_detect
does the right thing, but generic_identify misses.

Signed-off-by: Shaohua Li<shaohua.li@intel.com>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

DASD allows to open a device as soon as gendisk is registered, which means the
device is a fake device (capacity=0) and we do know nothing about blocksize
and partitions at that point of time.  In case the device is opened by
someone, the bdev and inode creation is done with the fake device info and the
following partition detection code is just using the wrong data.

To avoid this modify the DASD state machine to make sure that the open is
rejected until the device analysis is either finished or an unformatted device
was detected.
Signed-off-by: Horst Hummel <horst.hummel@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The message limit on the iucv connect call for the smsg module is too low.
Therefore increase the smsg message limit to 255.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

strnlen_user is supposed to return then length count + 1 if no terminating \0
is found, and it should return 0 on exception.  Found by David Howells
<dhowells@redhat.com>.
Signed-off-by: Gerald Schaefer <geraldsc@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-By: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

I have benchmarked this on an x86_64 NUMA system and see no significant
performance difference on kernbench.  Tested on both x86_64 and powerpc.

The way we do file struct accounting is not very suitable for batched
freeing.  For scalability reasons, file accounting was
constructor/destructor based.  This meant that nr_files was decremented
only when the object was removed from the slab cache.  This is susceptible
to slab fragmentation.  With RCU based file structure, consequent batched
freeing and a test program like Serge's, we just speed this up and end up
with a very fragmented slab -

llm22:~ # cat /proc/sys/fs/file-nr
587730  0       758844

At the same time, I see only a 2000+ objects in filp cache.  The following
patch I fixes this problem.

This patch changes the file counting by removing the filp_count_lock.
Instead we use a separate percpu counter, nr_files, for now and all
accesses to it are through get_nr_files() api.  In the sysctl handler for
nr_files, we populate files_stat.nr_files before returning to user.

Counting files as an when they are created and destroyed (as opposed to
inside slab) allows us to correctly count open files with RCU.
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>