For each new populated zone of hotadded node, need to update its pagesets
with dynamically allocated per_cpu_pageset struct for all possible CPUs:

    1) Detach zone->pageset from the shared boot_pageset
       at end of __build_all_zonelists().

    2) Use mutex to protect zone->pageset when it's still
       shared in onlined_pages()

Otherwises, multiple zones of different nodes would share same boot strapping
boot_pageset for same CPU, which will finally cause below kernel panic:

  ------------[ cut here ]------------
  kernel BUG at mm/page_alloc.c:1239!
  invalid opcode: 0000 [#1] SMP
  ...
  Call Trace:
   [<ffffffff811300c1>] __alloc_pages_nodemask+0x131/0x7b0
   [<ffffffff81162e67>] alloc_pages_current+0x87/0xd0
   [<ffffffff81128407>] __page_cache_alloc+0x67/0x70
   [<ffffffff811325f0>] __do_page_cache_readahead+0x120/0x260
   [<ffffffff81132751>] ra_submit+0x21/0x30
   [<ffffffff811329c6>] ondemand_readahead+0x166/0x2c0
   [<ffffffff81132ba0>] page_cache_async_readahead+0x80/0xa0
   [<ffffffff8112a0e4>] generic_file_aio_read+0x364/0x670
   [<ffffffff81266cfa>] nfs_file_read+0xca/0x130
   [<ffffffff8117b20a>] do_sync_read+0xfa/0x140
   [<ffffffff8117bf75>] vfs_read+0xb5/0x1a0
   [<ffffffff8117c151>] sys_read+0x51/0x80
   [<ffffffff8103c032>] system_call_fastpath+0x16/0x1b
  RIP  [<ffffffff8112ff13>] get_page_from_freelist+0x883/0x900
   RSP <ffff88000d1e78a8>
  ---[ end trace 4bda28328b9990db ]

[akpm@linux-foundation.org: merge fix]
Signed-off-by: NHaicheng Li <haicheng.li@linux.intel.com>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Reviewed-by: NAndi Kleen <andi.kleen@intel.com>
Reviewed-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Enable users to online CPUs even if the CPUs belongs to a numa node which
doesn't have onlined local memory.

The zonlists(pg_data_t.node_zonelists[]) of a numa node are created either
in system boot/init period, or at the time of local memory online.  For a
numa node without onlined local memory, its zonelists are not initialized
at present.  As a result, any memory allocation operations executed by
CPUs within this node will fail.  In fact, an out-of-memory error is
triggered when attempt to online CPUs before memory comes to online.

This patch tries to create zonelists for such numa nodes, so that the
memory allocation for this node can be fallback'ed to other nodes.

[akpm@linux-foundation.org: remove unneeded export]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: minskey guo<chaohong.guo@intel.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Reimplement stop_machine using cpu_stop.  As cpu stoppers are
guaranteed to be available for all online cpus,
stop_machine_create/destroy() are no longer necessary and removed.

With resource management and synchronization handled by cpu_stop, the
new implementation is much simpler.  Asking the cpu_stop to execute
the stop_cpu() state machine on all online cpus with cpu hotplug
disabled is enough.

stop_machine itself doesn't need to manage any global resources
anymore, so all per-instance information is rolled into struct
stop_machine_data and the mutex and all static data variables are
removed.

The previous implementation created and destroyed RT workqueues as
necessary which made stop_machine() calls highly expensive on very
large machines.  According to Dimitri Sivanich, preventing the dynamic
creation/destruction makes booting faster more than twice on very
large machines.  cpu_stop resources are preallocated for all online
cpus and should have the same effect.
Signed-off-by: NTejun Heo <tj@kernel.org>
Acked-by: NRusty Russell <rusty@rustcorp.com.au>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Dimitri Sivanich <sivanich@sgi.com>

_cpu_down() changes the current task's affinity and then recovers it at
the end. The problems are well known: we can't restore old_allowed if it
was bound to the now-dead-cpu, and we can race with the userspace which
can change cpu-affinity during unplug.

_cpu_down() should not play with current->cpus_allowed at all. Instead,
take_cpu_down() can migrate the caller of _cpu_down() after __cpu_disable()
removes the dying cpu from cpu_online_mask.
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Acked-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100315091023.GA9148@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h

percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: NTejun Heo <tj@kernel.org>
Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>

Additional_cpus is only supported for IA64 now.  X86_64 should not be
included.
Signed-off-by: NChen Gong <gong.chen@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Due to an incorrect line break the output currently contains tabs.
Also remove trailing space.

The actual output that logcheck sent me looked like this:
 Task events/1 (pid = 10) is on cpu 1^I^I^I^I(state = 1, flags = 84208040)

After this patch it becomes:
 Task events/1 (pid = 10) is on cpu 1 (state = 1, flags = 84208040)

Signed-off-by: Frans Pop <elendilplanet.nl>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <201001251456.34996.elendil@planet.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

We moved to migrate on wakeup, which means that sleeping tasks could
still be present on offline cpus. Amend the check to only test running
tasks.
Reported-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Sachin found cpu hotplug test failures on powerpc, which made
the kernel hang on his POWER box.

The problem is that we fail to re-activate a cpu when a
hot-unplug fails. Fix this by moving the de-activation into
_cpu_down after doing the initial checks.

Remove the synchronize_sched() calls and rely on those implied
by rebuilding the sched domains using the new mask.
Reported-by: NSachin Sant <sachinp@in.ibm.com>
Signed-off-by: NXiaotian Feng <dfeng@redhat.com>
Tested-by: NSachin Sant <sachinp@in.ibm.com>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <20091216170517.500272612@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Since (e761b772: cpu hotplug, sched: Introduce cpu_active_map and redo
sched domain managment) we have cpu_active_mask which is suppose to rule
scheduler migration and load-balancing, except it never (fully) did.

The particular problem being solved here is a crash in try_to_wake_up()
where select_task_rq() ends up selecting an offline cpu because
select_task_rq_fair() trusts the sched_domain tree to reflect the
current state of affairs, similarly select_task_rq_rt() trusts the
root_domain.

However, the sched_domains are updated from CPU_DEAD, which is after the
cpu is taken offline and after stop_machine is done. Therefore it can
race perfectly well with code assuming the domains are right.

Cure this by building the domains from cpu_active_mask on
CPU_DOWN_PREPARE.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Limit the number of per cpu calibration messages by only
printing out results for the first cpu to boot.

Also, don't print "CPUx is down" as this is expected, and we
don't need 4096 reminders... ;-)
Signed-off-by: NMike Travis <travis@sgi.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Roland Dreier <rdreier@cisco.com>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Tejun Heo <tj@kernel.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20091118002219.889552000@alcatraz.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Move tboot.h from asm to linux to fix the build errors of intel_txt
patch on non-X86 platforms. Remove the tboot code from generic code
init/main.c and kernel/cpu.c.
Signed-off-by: NShane Wang <shane.wang@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

SDM Vol 3a section titled "MTRR considerations in MP systems" specifies
the need for synchronizing the logical cpu's while initializing/updating
MTRR.

Currently Linux kernel does the synchronization of all cpu's only when
a single MTRR register is programmed/updated. During an AP online
(during boot/cpu-online/resume)  where we initialize all the MTRR/PAT registers,
we don't follow this synchronization algorithm.

This can lead to scenarios where during a dynamic cpu online, that logical cpu
is initializing MTRR/PAT with cache disabled (cr0.cd=1) etc while other logical
HT sibling continue to run (also with cache disabled because of cr0.cd=1
on its sibling).

Starting from Westmere, VMX transitions with cr0.cd=1 don't work properly
(because of some VMX performance optimizations) and the above scenario
(with one logical cpu doing VMX activity and another logical cpu coming online)
can result in system crash.

Fix the MTRR initialization by doing rendezvous of all the cpus. During
boot and resume, we delay the MTRR/PAT init for APs till all the
logical cpu's come online and the rendezvous process at the end of AP's bringup,
will initialize the MTRR/PAT for all AP's.

For dynamic single cpu online, we synchronize all the logical cpus and
do the MTRR/PAT init on the AP that is coming online.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Support for graceful handling of sleep states (S3/S4/S5) after an Intel(R) TXT launch.

Without this patch, attempting to place the system in one of the ACPI sleep
states (S3/S4/S5) will cause the TXT hardware to treat this as an attack and
will cause a system reset, with memory locked.  Not only may the subsequent
memory scrub take some time, but the platform will be unable to enter the
requested power state.

This patch calls back into the tboot so that it may properly and securely clean
up system state and clear the secrets-in-memory flag, after which it will place
the system into the requested sleep state using ACPI information passed by the kernel.

 arch/x86/kernel/smpboot.c     |    2 ++
 drivers/acpi/acpica/hwsleep.c |    3 +++
 kernel/cpu.c                  |    7 ++++++-
 3 files changed, 11 insertions(+), 1 deletion(-)
Signed-off-by: NJoseph Cihula <joseph.cihula@intel.com>
Signed-off-by: NShane Wang <shane.wang@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

SLAB uses get/put_online_cpus() which use a mutex which is itself only
initialized when cpu_hotplug_init() is called.  Currently we hang suring
boot in SLAB due to doing that too late.

Reported by James Bottomley and Sachin Sant (and possibly others).
Debugged by Benjamin Herrenschmidt.

This just removes the dynamic initialization of the data structures, and
replaces it with a static one, avoiding this dependency entirely, and
removing one unnecessary special initcall.
Tested-by: NSachin Sant <sachinp@in.ibm.com>
Tested-by: NJames Bottomley <James.Bottomley@HansenPartnership.com>
Tested-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

cpu_active_map is deprecated in favor of cpu_active_mask, which is
const for safety: we use accessors now (set_cpu_active) is we really
want to make a change.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

disable_nonboot_cpus calls _cpu_down. But _cpu_down requires that the
caller already created the stop_machine workqueue (like cpu_down does).
Otherwise a call to stop_machine will lead to accesses to random memory
regions.

When introducing this new interface (9ea09af3
"stop_machine: introduce stop_machine_create/destroy") I missed the second
call site of _cpu_down.
So add the missing stop_machine_create/destroy calls to disable_nonboot_cpus
as well.

Fixes suspend-to-ram/disk and also this bug:

[  286.547348] BUG: unable to handle kernel paging request at 6b6b6b6b
[  286.548940] IP: [<c0150ca4>] __stop_machine+0x88/0xe3
[  286.550598] Oops: 0002 [#1] SMP
[  286.560580] Pid: 3273, comm: halt Not tainted (2.6.28-06127-g238c6d54
[  286.560580] EIP: is at __stop_machine+0x88/0xe3
[  286.560580] Process halt (pid: 3273, ti=f1a28000 task=f4530f30
[  286.560580] Call Trace:
[  286.560580]  [<c03d04e4>] ? _cpu_down+0x10f/0x234
[  286.560580]  [<c012a57e>] ? disable_nonboot_cpus+0x58/0xdc
[  286.560580]  [<c01360c0>] ? kernel_poweroff+0x22/0x39
[  286.560580]  [<c0136301>] ? sys_reboot+0xde/0x14c
[  286.560580]  [<c01331b2>] ? complete_signal+0x179/0x191
[  286.560580]  [<c0133396>] ? send_signal+0x1cc/0x1e1
[  286.560580]  [<c03de418>] ? _spin_unlock_irqrestore+0x2d/0x3c
[  286.560580]  [<c0133b65>] ? group_send_signal_info+0x58/0x61
[  286.560580]  [<c0133b9e>] ? kill_pid_info+0x30/0x3a
[  286.560580]  [<c0133d49>] ? sys_kill+0x75/0x13a
[  286.560580]  [<c01a06cb>] ? mntput_no_expire+ox1f/0x101
[  286.560580]  [<c019b3b3>] ? dput+0x1e/0x105
[  286.560580]  [<c018ef87>] ?  __fput+0x150/0x158
[  286.560580]  [<c0157abf>] ? audit_syscall_entry+0x137/0x159
[  286.560580]  [<c010329f>] ? sysenter_do_call+0x12/0x34
Reported-and-tested-by: N"Justin P. Mattock" <justinmattock@gmail.com>
Reviewed-by: NPekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Tested-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Introduce stop_machine_create/destroy. With this interface subsystems
that need a non-failing stop_machine environment can create the
stop_machine machine threads before actually calling stop_machine.
When the threads aren't needed anymore they can be killed with
stop_machine_destroy again.

When stop_machine gets called and the threads aren't present they
will be created and destroyed automatically. This restores the old
behaviour of stop_machine.

This patch also converts cpu hotplug to the new interface since it
is special: cpu_down calls __stop_machine instead of stop_machine.
However the kstop threads will only be created when stop_machine
gets called.

Changing the code so that the threads would be created automatically
on __stop_machine is currently not possible: when __stop_machine gets
called we hold cpu_add_remove_lock, which is the same lock that
create_rt_workqueue would take. So the workqueue needs to be created
before the cpu hotplug code locks cpu_add_remove_lock.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

Impact: Reduce kernel stack and memory usage, use new cpumask API.

Use cpumask_var_t for take_cpu_down() stack var, and frozen_cpus.

Note that notify_cpu_starting() can be called before core_initcall
allocates frozen_cpus, but the NULL check is optimized out by gcc for
the CONFIG_CPUMASK_OFFSTACK=n case.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

They're only for use in boot/cpu hotplug code anyway, and this avoids
the use of deprecated cpu_*_map.

Stephen Rothwell points out that gcc 4.2.4 (on powerpc at least)
didn't like the cast away of const anyway:

  include/linux/cpumask.h: In function 'set_cpu_possible':
  include/linux/cpumask.h:1052: warning: passing argument 2 of 'cpumask_set_cpu' discards qualifiers from pointer target type

So this kills two birds with one stone.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

Impact: cleanup

This implements the obsolescent cpu_online_map in terms of
cpu_online_mask, rather than the other way around.  Same for the other
maps.

The documentation comments are also updated to refer to _mask rather
than _map.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NMike Travis <travis@sgi.com>

Impact: cleanup

Each SMP arch defines these themselves.  Move them to a central
location.

Twists:
1) Some archs (m32, parisc, s390) set possible_map to all 1, so we add a
   CONFIG_INIT_ALL_POSSIBLE for this rather than break them.

2) mips and sparc32 '#define cpu_possible_map phys_cpu_present_map'.
   Those archs simply have phys_cpu_present_map replaced everywhere.

3) Alpha defined cpu_possible_map to cpu_present_map; this is tricky
   so I just manipulate them both in sync.

4) IA64, cris and m32r have gratuitous 'extern cpumask_t cpu_possible_map'
   declarations.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Reviewed-by: NGrant Grundler <grundler@parisc-linux.org>
Tested-by: NTony Luck <tony.luck@intel.com>
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: Mike Travis <travis@sgi.com>
Cc: ink@jurassic.park.msu.ru
Cc: rmk@arm.linux.org.uk
Cc: starvik@axis.com
Cc: tony.luck@intel.com
Cc: takata@linux-m32r.org
Cc: ralf@linux-mips.org
Cc: grundler@parisc-linux.org
Cc: paulus@samba.org
Cc: schwidefsky@de.ibm.com
Cc: lethal@linux-sh.org
Cc: wli@holomorphy.com
Cc: davem@davemloft.net
Cc: jdike@addtoit.com
Cc: mingo@redhat.com

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Impact: introduce new APIs

We want to deprecate cpumasks on the stack, as we are headed for
gynormous numbers of CPUs.  Eventually, we want to head towards an
undefined 'struct cpumask' so they can never be declared on stack.

1) New cpumask functions which take pointers instead of copies.
   (cpus_* -> cpumask_*)

2) Several new helpers to reduce requirements for temporary cpumasks
   (cpumask_first_and, cpumask_next_and, cpumask_any_and)

3) Helpers for declaring cpumasks on or offstack for large NR_CPUS
   (cpumask_var_t, alloc_cpumask_var and free_cpumask_var)

4) 'struct cpumask' for explicitness and to mark new-style code.

5) Make iterator functions stop at nr_cpu_ids (a runtime constant),
   not NR_CPUS for time efficiency and for smaller dynamic allocations
   in future.

6) cpumask_copy() so we can allocate less than a full cpumask eventually
   (for alloc_cpumask_var), and so we can eliminate the 'struct cpumask'
   definition eventually.

7) work_on_cpu() helper for doing task on a CPU, rather than saving old
   cpumask for current thread and manipulating it.

8) smp_call_function_many() which is smp_call_function_mask() except
   taking a cpumask pointer.

Note that this patch simply introduces the new functions and leaves
the obsolescent ones in place.  This is to simplify the transition
patches.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Right now, there is no notifier that is called on a new cpu, before the new
cpu begins processing interrupts/softirqs.
Various kernel function would need that notification, e.g. kvm works around
by calling smp_call_function_single(), rcu polls cpu_online_map.

The patch adds a CPU_STARTING notification. It also adds a helper function
that sends the message to all cpu_chain handlers.

Tested on x86-64.
All other archs are untested. Especially on sparc, I'm not sure if I got
it right.
Signed-off-by: NManfred Spraul <manfred@colorfullife.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

When a cpu is taken offline, the CPU_DYING notifiers are called on the
dying cpu. According to <linux/notifiers.h>, the cpu should be "not
running any task, not handling interrupts, soon dead".

For the current implementation, this is not true:
- __cpu_disable can fail. If it fails, then the cpu will remain alive
  and happy.
- At least on x86, __cpu_disable() briefly enables the local interrupts
  to handle any outstanding interrupts.

What about moving CPU_DYING down a few lines, behind the __cpu_disable()
line?
There are only two CPU_DYING handlers in the kernel right now: one in
kvm, one in the scheduler. Both should work with the patch applied
[and: I'm not sure if either one handles a failing __cpu_disable()]

The patch survives simple offlining a cpu. kvm untested due to lack
of a test setup.
Signed-off-By: NManfred Spraul <manfred@colorfullife.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

s390 doesn't support the additional_cpus kernel parameter anymore since a
long time.  So we better update the code and documentation to reflect
that.

Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Mark Langsdorf reported:

> One of my co-workers noticed that the powernow-k8
> driver no longer restarts when a CPU core is
> hot-disabled and then hot-enabled on AMD quad-core
> systems.
>
> The following comands work fine on 2.6.26 and fail
> on 2.6.27-rc1:
>
> echo 0 > /sys/devices/system/cpu/cpu3/online
> echo 1 > /sys/devices/system/cpu/cpu3/online
> find /sys -name cpufreq
>
> For 2.6.26, the find will return a cpufreq
> directory for each processor.  In 2.6.27-rc1,
> the cpu3 directory is missing.
>
> After digging through the code, the following
> logic is failing when the core is hot-enabled
> at runtime.  The code works during the boot
> sequence.
>
>       cpumask_t = current->cpus_allowed;
>       set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
>       if (smp_processor_id() != cpu)
>               return -ENODEV;

So set the CPU active before calling the CPU_ONLINE notifier chain,
there are a handful of notifiers that use set_cpus_allowed().

This fix also solves the problem with x86-microcode. I've sent
alternative patches for microcode, but as this "rely on
set_cpus_allowed_ptr() being workable in cpu-hotplug(CPU_ONLINE, ...)"
assumption seems to be more broad than what we thought, perhaps this fix
should be applied.

With this patch we define that by the moment CPU_ONLINE is being sent,
a 'cpu' is online and ready for tasks to be migrated onto it.
Signed-off-by: NDmitry Adamushko <dmitry.adamushko@gmail.com>
Reported-by: NMark Langsdorf <mark.langsdorf@amd.com>
Tested-by: NMark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Clean up and optimize cpumask_of_cpu(), by sharing all the zero words.

Instead of stupidly generating all possible i=0...NR_CPUS 2^i patterns
creating a huge array of constant bitmasks, realize that the zero words
can be shared.

In other words, on a 64-bit architecture, we only ever need 64 of these
arrays - with a different bit set in one single world (with enough zero
words around it so that we can create any bitmask by just offsetting in
that big array). And then we just put enough zeroes around it that we
can point every single cpumask to be one of those things.

So when we have 4k CPU's, instead of having 4k arrays (of 4k bits each,
with one bit set in each array - 2MB memory total), we have exactly 64
arrays instead, each 8k bits in size (64kB total).

And then we just point cpumask(n) to the right position (which we can
calculate dynamically). Once we have the right arrays, getting
"cpumask(n)" ends up being:

  static inline const cpumask_t *get_cpu_mask(unsigned int cpu)
  {
          const unsigned long *p = cpu_bit_bitmap[1 + cpu % BITS_PER_LONG];
          p -= cpu / BITS_PER_LONG;
          return (const cpumask_t *)p;
  }

This brings other advantages and simplifications as well:

 - we are not wasting memory that is just filled with a single bit in
   various different places

 - we don't need all those games to re-create the arrays in some dense
   format, because they're already going to be dense enough.

if we compile a kernel for up to 4k CPU's, "wasting" that 64kB of memory
is a non-issue (especially since by doing this "overlapping" trick we
probably get better cache behaviour anyway).

[ mingo@elte.hu:

  Converted Linus's mails into a commit. See:

     http://lkml.org/lkml/2008/7/27/156
     http://lkml.org/lkml/2008/7/28/320

  Also applied a family filter - which also has the side-effect of leaving
  out the bits where Linus calls me an idio... Oh, never mind ;-)
]
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Mike Travis <travis@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Instead of a "cpu" arg with magic values NR_CPUS (any cpu) and ~0 (all
cpus), pass a cpumask_t.  Allow NULL for the common case (where we
don't care which CPU the function is run on): temporary cpumask_t's
are usually considered bad for stack space.

This deprecates stop_machine_run, to be removed soon when all the
callers are dead.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

Akinobu points out that if take_cpu_down() succeeds, the cpu must be offline.
Remove the cpu_online() check, and put a BUG_ON().

Quoting Akinobu Mita:
   Actually the cpu_online() check was necessary before appling this
   stop_machine: simplify patch.

   With old __stop_machine_run(), __stop_machine_run() could succeed
   (return !IS_ERR(p) value) even if take_cpu_down() returned non-zero value.
   The return value of take_cpu_down() was obtained through kthread_stop()..
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: "Akinobu Mita" <akinobu.mita@gmail.com>

stop_machine creates a kthread which creates kernel threads.  We can
create those threads directly and simplify things a little.  Some care
must be taken with CPU hotunplug, which has special needs, but that code
seems more robust than it was in the past.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Acked-by: NChristian Borntraeger <borntraeger@de.ibm.com>

fix:

 ERROR: "cpumask_of_cpu_map" [drivers/acpi/processor.ko] undefined!
 ERROR: "cpumask_of_cpu_map" [arch/x86/kernel/microcode.ko] undefined!
 ERROR: "cpumask_of_cpu_map" [arch/x86/kernel/cpu/cpufreq/speedstep-ich.ko] undefined!
 ERROR: "cpumask_of_cpu_map" [arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.ko] undefined!
Signed-off-by: NIngo Molnar <mingo@elte.hu>

  * Create the cpumask_of_cpu_map statically in the init data section
    using NR_CPUS but replace it during boot up with one sized by
    nr_cpu_ids (num possible cpus).
Signed-off-by: NMike Travis <travis@sgi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

If an arch doesn't define cpumask_of_cpu_map, create a generic
statically-initialized one for them.  This allows removal of the buggy
cpumask_of_cpu() macro (&cpumask_of_cpu() gives address of
out-of-scope var).

An arch with NR_CPUS of 4096 probably wants to allocate this itself
based on the actual number of CPUs, since otherwise they're using 2MB
of rodata (1024 cpus means 128k).  That's what
CONFIG_HAVE_CPUMASK_OF_CPU_MAP is for (only x86/64 does so at the
moment).

In future as we support more CPUs, we'll need to resort to a
get_cpu_map()/put_cpu_map() allocation scheme.
Signed-off-by: NMike Travis <travis@sgi.com>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jack Steiner <steiner@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

workqueue_cpu_callback(CPU_DEAD) flushes cwq->thread under
cpu_maps_update_begin().  This means that the multithreaded workqueues
can't use get_online_cpus() due to the possible deadlock, very bad and
very old problem.

Introduce the new state, CPU_POST_DEAD, which is called after
cpu_hotplug_done() but before cpu_maps_update_done().

Change workqueue_cpu_callback() to use CPU_POST_DEAD instead of CPU_DEAD.
This means that create/destroy functions can't rely on get_online_cpus()
any longer and should take cpu_add_remove_lock instead.

[akpm@linux-foundation.org: fix CONFIG_SMP=n]
Signed-off-by: NOleg Nesterov <oleg@tv-sign.ru>
Acked-by: NGautham R Shenoy <ego@in.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Paul Menage <menage@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This goes on top of the cpu_active_map (take 2) patch.

Currently we depend on the stop_machine to provide nescessesary
synchronization for the cpu_active_map updates.
As Dmitry Adamushko pointed this is fragile and is not much clearer
than the previous scheme. In other words we do not want to depend on
the internal stop machine operation here.
So make the synchronization rules clear by doing synchronize_sched()
after clearing out cpu active bit.

Tested on quad-Core2 with:

   while true; do
      for i in 1 2 3; do
        echo 0 > /sys/devices/system/cpu/cpu$i/online
      done
      for i in 1 2 3; do
        echo 1 > /sys/devices/system/cpu/cpu$i/online
      done
   done
and
   stress -c 200

No lockdep, preempt or other complaints.
Signed-off-by: NMax Krasnyansky <maxk@qualcomm.com>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This is based on Linus' idea of creating cpu_active_map that prevents
scheduler load balancer from migrating tasks to the cpu that is going
down.

It allows us to simplify domain management code and avoid unecessary
domain rebuilds during cpu hotplug event handling.

Please ignore the cpusets part for now. It needs some more work in order
to avoid crazy lock nesting. Although I did simplfy and unify domain
reinitialization logic. We now simply call partition_sched_domains() in
all the cases. This means that we're using exact same code paths as in
cpusets case and hence the test below cover cpusets too.
Cpuset changes to make rebuild_sched_domains() callable from various
contexts are in the separate patch (right next after this one).

This not only boots but also easily handles
	while true; do make clean; make -j 8; done
and
	while true; do on-off-cpu 1; done
at the same time.
(on-off-cpu 1 simple does echo 0/1 > /sys/.../cpu1/online thing).

Suprisingly the box (dual-core Core2) is quite usable. In fact I'm typing
this on right now in gnome-terminal and things are moving just fine.

Also this is running with most of the debug features enabled (lockdep,
mutex, etc) no BUG_ONs or lockdep complaints so far.

I believe I addressed all of the Dmitry's comments for original Linus'
version. I changed both fair and rt balancer to mask out non-active cpus.
And replaced cpu_is_offline() with !cpu_active() in the main scheduler
code where it made sense (to me).
Signed-off-by: NMax Krasnyanskiy <maxk@qualcomm.com>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: NGregory Haskins <ghaskins@novell.com>
Cc: dmitry.adamushko@gmail.com
Cc: pj@sgi.com
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The ACPI device node for the cpu has already been unregistered
when acpi_processor_handle_eject is called.
Thus we should offline the cpu and continue, rather than a failure here.
http://bugzilla.kernel.org/show_bug.cgi?id=9772Signed-off-by: NZhang Rui <rui.zhang@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NAndi Kleen <ak@linux.intel.com>

kernel/cpu.c seems a more logical place for those maps since they do not really
have much to do with the scheduler these days.

kernel/cpu.c is now built for the UP kernel too, but it does not affect the size
the kernel sections.

$ size vmlinux

before
   text       data        bss        dec        hex    filename
3313797     307060     310352    3931209     3bfc49    vmlinux

after
   text       data        bss        dec        hex    filename
3313797     307060     310352    3931209     3bfc49    vmlinux
Signed-off-by: NMax Krasnyansky <maxk@qualcomm.com>
Cc: pj@sgi.com
Cc: menage@google.com
Cc: rostedt@goodmis.org
Cc: mingo@elte.hu
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>