Rob Herring has observed that c81611c4 "xen: event channel arrays are
xen_ulong_t and not unsigned long" introduced a compile failure when building
without CONFIG_AEABI:

/tmp/ccJaIZOW.s: Assembler messages:
/tmp/ccJaIZOW.s:831: Error: even register required -- `ldrexd r5,r6,[r4]'

Will Deacon pointed out that this is because OABI does not require even base
registers for 64-bit values. We can avoid this by simply using the existing
atomic64_xchg operation and the same containerof trick as used by the cmpxchg
macros. However since this code is used on memory which is shared with the
hypervisor we require proper atomic instructions and cannot use the generic
atomic64 callbacks (which are based on spinlocks), therefore add a dependency
on !GENERIC_ATOMIC64. Since we already depend on !CPU_V6 there isn't much
downside to this.

While thinking about this we also observed that OABI has different struct
alignment requirements to EABI, which is a problem for hypercall argument
structs which are shared with the hypervisor and which must be in EABI layout.
Since I don't expect people to want to run OABI kernels on Xen depend on
CONFIG_AEABI explicitly too (although it also happens to be enforced by the
!GENERIC_ATOMIC64 requirement too).
Signed-off-by: NIan Campbell <ian.campbell@citrix.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Rob Herring <robherring2@gmail.com>
Acked-by: NStefano Stabellini <Stefano.Stabellini@eu.citrix.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

The git commit d5aaffa9
(cpufreq: handle cpufreq being disabled for all exported function)
tightens the cpufreq API by returning errors when disable_cpufreq()
had been called.

The problem we are hitting is that the module xen-acpi-processor which
uses the ACPI's functions: acpi_processor_register_performance,
acpi_processor_preregister_performance, and acpi_processor_notify_smm
fails at acpi_processor_register_performance with -22.

Note that earlier during bootup in arch/x86/xen/setup.c there is also
an call to cpufreq's API: disable_cpufreq().

This is b/c we want the Linux kernel to parse the ACPI data, but leave
the cpufreq decisions to the hypervisor.

In v3.9 all the checks that d5aaffa9
added are now hit and the calls to cpufreq_register_notifier will now
fail. This means that acpi_processor_ppc_init ends up printing:

"Warning: Processor Platform Limit not supported"

and the acpi_processor_ppc_status is not set.

The repercussions of that is that the call to
acpi_processor_register_performance fails right away at:

	if (!(acpi_processor_ppc_status & PPC_REGISTERED))

and we don't progress any further on parsing and extracting the _P*
objects.

The only reason the Xen code called that function was b/c it was
exported and the only way to gather the P-states. But we can also
just make acpi_processor_get_performance_info be exported and not
use acpi_processor_register_performance. This patch does so.
Acked-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

While shuting down a HVM guest with pci devices passed through we
get this:

pciback 0000:04:00.0: restoring config space at offset 0x4 (was 0x100000, writing 0x100002)
------------[ cut here ]------------
WARNING: at drivers/pci/pci.c:1397 pci_disable_device+0x88/0xa0()
Hardware name: MS-7640
Device pciback
disabling already-disabled device
Modules linked in:
Pid: 53, comm: xenwatch Not tainted 3.9.0-rc1-20130304a+ #1
Call Trace:
 [<ffffffff8106994a>] warn_slowpath_common+0x7a/0xc0
 [<ffffffff81069a31>] warn_slowpath_fmt+0x41/0x50
 [<ffffffff813cf288>] pci_disable_device+0x88/0xa0
 [<ffffffff814554a7>] xen_pcibk_reset_device+0x37/0xd0
 [<ffffffff81454b6f>] ? pcistub_put_pci_dev+0x6f/0x120
 [<ffffffff81454b8d>] pcistub_put_pci_dev+0x8d/0x120
 [<ffffffff814582a9>] __xen_pcibk_release_devices+0x59/0xa0

This fixes the bug.

CC: stable@vger.kernel.org
Reported-and-Tested-by: NSander Eikelenboom <linux@eikelenboom.it>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Adding an include of linux/mm.h resolves this:
	drivers/xen/xenbus/xenbus_client.c: In function ‘xenbus_map_ring_valloc_hvm’:
	drivers/xen/xenbus/xenbus_client.c:532:66: error: implicit declaration of function ‘page_to_section’ [-Werror=implicit-function-declaration]

CC: stable@vger.kernel.org
Signed-off-by: NSteven Noonan <steven@uplinklabs.net>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

There is no hypercall to setup multiple MSI per PCI device.
As such with these two new commits:
-  08261d87
   PCI/MSI: Enable multiple MSIs with pci_enable_msi_block_auto()
- 5ca72c4f
   AHCI: Support multiple MSIs

we would call the PHYSDEVOP_map_pirq 'nvec' times with the same
contents of the PCI device. Sander discovered that we would get
the same PIRQ value 'nvec' times and return said values to the
caller. That of course meant that the device was configured only
with one MSI and AHCI would fail with:

ahci 0000:00:11.0: version 3.0
xen: registering gsi 19 triggering 0 polarity 1
xen: --> pirq=19 -> irq=19 (gsi=19)
(XEN) [2013-02-27 19:43:07] IOAPIC[0]: Set PCI routing entry (6-19 -> 0x99 -> IRQ 19 Mode:1 Active:1)
ahci 0000:00:11.0: AHCI 0001.0200 32 slots 4 ports 6 Gbps 0xf impl SATA mode
ahci 0000:00:11.0: flags: 64bit ncq sntf ilck pm led clo pmp pio slum part
ahci: probe of 0000:00:11.0 failed with error -22

That is b/c in ahci_host_activate the second call to
devm_request_threaded_irq  would return -EINVAL as we passed in
(on the second run) an IRQ that was never initialized.

CC: stable@vger.kernel.org
Reported-and-Tested-by: NSander Eikelenboom <linux@eikelenboom.it>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

The git commit 8eaffa67
(xen/pat: Disable PAT support for now) explains in details why
we want to disable PAT for right now. However that
change was not enough and we should have also disabled
the pat_enabled value. Otherwise we end up with:

mmap-example:3481 map pfn expected mapping type write-back for
[mem 0x00010000-0x00010fff], got uncached-minus
 ------------[ cut here ]------------
WARNING: at /build/buildd/linux-3.8.0/arch/x86/mm/pat.c:774 untrack_pfn+0xb8/0xd0()
mem 0x00010000-0x00010fff], got uncached-minus
------------[ cut here ]------------
WARNING: at /build/buildd/linux-3.8.0/arch/x86/mm/pat.c:774
untrack_pfn+0xb8/0xd0()
...
Pid: 3481, comm: mmap-example Tainted: GF 3.8.0-6-generic #13-Ubuntu
Call Trace:
 [<ffffffff8105879f>] warn_slowpath_common+0x7f/0xc0
 [<ffffffff810587fa>] warn_slowpath_null+0x1a/0x20
 [<ffffffff8104bcc8>] untrack_pfn+0xb8/0xd0
 [<ffffffff81156c1c>] unmap_single_vma+0xac/0x100
 [<ffffffff81157459>] unmap_vmas+0x49/0x90
 [<ffffffff8115f808>] exit_mmap+0x98/0x170
 [<ffffffff810559a4>] mmput+0x64/0x100
 [<ffffffff810560f5>] dup_mm+0x445/0x660
 [<ffffffff81056d9f>] copy_process.part.22+0xa5f/0x1510
 [<ffffffff81057931>] do_fork+0x91/0x350
 [<ffffffff81057c76>] sys_clone+0x16/0x20
 [<ffffffff816ccbf9>] stub_clone+0x69/0x90
 [<ffffffff816cc89d>] ? system_call_fastpath+0x1a/0x1f
---[ end trace 4918cdd0a4c9fea4 ]---

(a similar message shows up if you end up launching 'mcelog')

The call chain is (as analyzed by Liu, Jinsong):
do_fork
  --> copy_process
    --> dup_mm
      --> dup_mmap
       	--> copy_page_range
          --> track_pfn_copy
            --> reserve_pfn_range
              --> line 624: flags != want_flags
It comes from different memory types of page table (_PAGE_CACHE_WB) and MTRR
(_PAGE_CACHE_UC_MINUS).

Stefan Bader dug in this deep and found out that:
"That makes it clearer as this will do

reserve_memtype(...)
--> pat_x_mtrr_type
  --> mtrr_type_lookup
    --> __mtrr_type_lookup

And that can return -1/0xff in case of MTRR not being enabled/initialized. Which
is not the case (given there are no messages for it in dmesg). This is not equal
to MTRR_TYPE_WRBACK and thus becomes _PAGE_CACHE_UC_MINUS.

It looks like the problem starts early in reserve_memtype:

       	if (!pat_enabled) {
                /* This is identical to page table setting without PAT */
                if (new_type) {
                        if (req_type == _PAGE_CACHE_WC)
                                *new_type = _PAGE_CACHE_UC_MINUS;
                        else
                               	*new_type = req_type & _PAGE_CACHE_MASK;
               	}
                return 0;
        }

This would be what we want, that is clearing the PWT and PCD flags from the
supported flags - if pat_enabled is disabled."

This patch does that - disabling PAT.

CC: stable@vger.kernel.org # 3.3 and further
Reported-by: NSander Eikelenboom <linux@eikelenboom.it>
Reported-and-Tested-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reported-and-Tested-by: NStefan Bader <stefan.bader@canonical.com>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Recently at native Rafael did some cleanup for acpi, say, drop
acpi_bus_add, remove unnecessary argument of acpi_bus_scan,
and run acpi_bus_scan under acpi_scan_lock.

This patch does similar cleanup for xen cpu hotplug, removing
redundant logic, and adding lock.
Signed-off-by: NLiu Jinsong <jinsong.liu@intel.com>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Dan Carpenter found current xen memory hotplug logic
has potential issue: at func acpi_memory_get_device()
*mem_device = acpi_driver_data(device);
while the device may be NULL and then dereference.

At native side, Rafael recently updated acpi_memory_get_device(),
dropping acpi_bus_add, adding lock, and avoiding above issue.

This patch updates xen memory hotplug logic accordingly, removing
redundant logic, adding lock, and avoiding dereference.
Signed-off-by: NLiu Jinsong <jinsong.liu@intel.com>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Merge tag 'stable/for-linus-3.9-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen

Pull Xen update from Konrad Rzeszutek Wilk:
 "This has two new ACPI drivers for Xen - a physical CPU offline/online
  and a memory hotplug.  The way this works is that ACPI kicks the
  drivers and they make the appropiate hypercall to the hypervisor to
  tell it that there is a new CPU or memory.  There also some changes to
  the Xen ARM ABIs and couple of fixes.  One particularly nasty bug in
  the Xen PV spinlock code was fixed by Stefan Bader - and has been
  there since the 2.6.32!

  Features:
   - Xen ACPI memory and CPU hotplug drivers - allowing Xen hypervisor
     to be aware of new CPU and new DIMMs
   - Cleanups
  Bug-fixes:
   - Fixes a long-standing bug in the PV spinlock wherein we did not
     kick VCPUs that were in a tight loop.
   - Fixes in the error paths for the event channel machinery"

Fix up a few semantic conflicts with the ACPI interface changes in
drivers/xen/xen-acpi-{cpu,mem}hotplug.c.

* tag 'stable/for-linus-3.9-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
  xen: event channel arrays are xen_ulong_t and not unsigned long
  xen: Send spinlock IPI to all waiters
  xen: introduce xen_remap, use it instead of ioremap
  xen: close evtchn port if binding to irq fails
  xen-evtchn: correct comment and error output
  xen/tmem: Add missing %s in the printk statement.
  xen/acpi: move xen_acpi_get_pxm under CONFIG_XEN_DOM0
  xen/acpi: ACPI cpu hotplug
  xen/acpi: Move xen_acpi_get_pxm to Xen's acpi.h
  xen/stub: driver for CPU hotplug
  xen/acpi: ACPI memory hotplug
  xen/stub: driver for memory hotplug
  xen: implement updated XENMEM_add_to_physmap_range ABI
  xen/smp: Move the common CPU init code a bit to prep for PVH patch.

Pull KVM updates from Marcelo Tosatti:
 "KVM updates for the 3.9 merge window, including x86 real mode
  emulation fixes, stronger memory slot interface restrictions, mmu_lock
  spinlock hold time reduction, improved handling of large page faults
  on shadow, initial APICv HW acceleration support, s390 channel IO
  based virtio, amongst others"

* tag 'kvm-3.9-1' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (143 commits)
  Revert "KVM: MMU: lazily drop large spte"
  x86: pvclock kvm: align allocation size to page size
  KVM: nVMX: Remove redundant get_vmcs12 from nested_vmx_exit_handled_msr
  x86 emulator: fix parity calculation for AAD instruction
  KVM: PPC: BookE: Handle alignment interrupts
  booke: Added DBCR4 SPR number
  KVM: PPC: booke: Allow multiple exception types
  KVM: PPC: booke: use vcpu reference from thread_struct
  KVM: Remove user_alloc from struct kvm_memory_slot
  KVM: VMX: disable apicv by default
  KVM: s390: Fix handling of iscs.
  KVM: MMU: cleanup __direct_map
  KVM: MMU: remove pt_access in mmu_set_spte
  KVM: MMU: cleanup mapping-level
  KVM: MMU: lazily drop large spte
  KVM: VMX: cleanup vmx_set_cr0().
  KVM: VMX: add missing exit names to VMX_EXIT_REASONS array
  KVM: VMX: disable SMEP feature when guest is in non-paging mode
  KVM: Remove duplicate text in api.txt
  Revert "KVM: MMU: split kvm_mmu_free_page"
  ...

Pull signal handling cleanups from Al Viro:
 "This is the first pile; another one will come a bit later and will
  contain SYSCALL_DEFINE-related patches.

   - a bunch of signal-related syscalls (both native and compat)
     unified.

   - a bunch of compat syscalls switched to COMPAT_SYSCALL_DEFINE
     (fixing several potential problems with missing argument
     validation, while we are at it)

   - a lot of now-pointless wrappers killed

   - a couple of architectures (cris and hexagon) forgot to save
     altstack settings into sigframe, even though they used the
     (uninitialized) values in sigreturn; fixed.

   - microblaze fixes for delivery of multiple signals arriving at once

   - saner set of helpers for signal delivery introduced, several
     architectures switched to using those."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/signal: (143 commits)
  x86: convert to ksignal
  sparc: convert to ksignal
  arm: switch to struct ksignal * passing
  alpha: pass k_sigaction and siginfo_t using ksignal pointer
  burying unused conditionals
  make do_sigaltstack() static
  arm64: switch to generic old sigaction() (compat-only)
  arm64: switch to generic compat rt_sigaction()
  arm64: switch compat to generic old sigsuspend
  arm64: switch to generic compat rt_sigqueueinfo()
  arm64: switch to generic compat rt_sigpending()
  arm64: switch to generic compat rt_sigprocmask()
  arm64: switch to generic sigaltstack
  sparc: switch to generic old sigsuspend
  sparc: COMPAT_SYSCALL_DEFINE does all sign-extension as well as SYSCALL_DEFINE
  sparc: kill sign-extending wrappers for native syscalls
  kill sparc32_open()
  sparc: switch to use of generic old sigaction
  sparc: switch sys_compat_rt_sigaction() to COMPAT_SYSCALL_DEFINE
  mips: switch to generic sys_fork() and sys_clone()
  ...

Merge second patch-bomb from Andrew Morton:

 - A little DM fix

 - the MM queue

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (154 commits)
  ksm: allocate roots when needed
  mm: cleanup "swapcache" in do_swap_page
  mm,ksm: swapoff might need to copy
  mm,ksm: FOLL_MIGRATION do migration_entry_wait
  ksm: shrink 32-bit rmap_item back to 32 bytes
  ksm: treat unstable nid like in stable tree
  ksm: add some comments
  tmpfs: fix mempolicy object leaks
  tmpfs: fix use-after-free of mempolicy object
  mm/fadvise.c: drain all pagevecs if POSIX_FADV_DONTNEED fails to discard all pages
  mm: export mmu notifier invalidates
  mm: accelerate mm_populate() treatment of THP pages
  mm: use long type for page counts in mm_populate() and get_user_pages()
  mm: accurately document nr_free_*_pages functions with code comments
  HWPOISON: change order of error_states[]'s elements
  HWPOISON: fix misjudgement of page_action() for errors on mlocked pages
  memcg: stop warning on memcg_propagate_kmem
  net: change type of virtio_chan->p9_max_pages
  vmscan: change type of vm_total_pages to unsigned long
  fs/nfsd: change type of max_delegations, nfsd_drc_max_mem and nfsd_drc_mem_used
  ...

It is a pity to have MAX_NUMNODES+MAX_NUMNODES tree roots statically
allocated, particularly when very few users will ever actually tune
merge_across_nodes 0 to use more than 1+1 of those trees.  Not a big
deal (only 16kB wasted on each machine with CONFIG_MAXSMP), but a pity.

Start off with 1+1 statically allocated, then if merge_across_nodes is
ever tuned, allocate for nr_node_ids+nr_node_ids.  Do not attempt to
free up the extra if it's tuned back, that would be a waste of effort.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Petr Holasek <pholasek@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

I dislike the way in which "swapcache" gets used in do_swap_page():
there is always a page from swapcache there (even if maybe uncached by
the time we lock it), but tests are made according to "swapcache".
Rework that with "page != swapcache", as has been done in unuse_pte().
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Petr Holasek <pholasek@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Before establishing that KSM page migration was the cause of my
WARN_ON_ONCE(page_mapped(page))s, I suspected that they came from the
lack of a ksm_might_need_to_copy() in swapoff's unuse_pte() - which in
many respects is equivalent to faulting in a page.

In fact I've never caught that as the cause: but in theory it does at
least need the KSM_RUN_UNMERGE check in ksm_might_need_to_copy(), to
avoid bringing a KSM page back in when it's not supposed to be.

I intended to copy how it's done in do_swap_page(), but have a strong
aversion to how "swapcache" ends up being used there: rework it with
"page != swapcache".
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Petr Holasek <pholasek@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Izik Eidus <izik.eidus@ravellosystems.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In "ksm: remove old stable nodes more thoroughly" I said that I'd never
seen its WARN_ON_ONCE(page_mapped(page)).  True at the time of writing,
but it soon appeared once I tried fuller tests on the whole series.

It turned out to be due to the KSM page migration itself: unmerge_and_
remove_all_rmap_items() failed to locate and replace all the KSM pages,
because of that hiatus in page migration when old pte has been replaced
by migration entry, but not yet by new pte.  follow_page() finds no page
at that instant, but a KSM page reappears shortly after, without a
fault.

Add FOLL_MIGRATION flag, so follow_page() can do migration_entry_wait()
for KSM's break_cow().  I'd have preferred to avoid another flag, and do
it every time, in case someone else makes the same easy mistake; but did
not find another transgressor (the common get_user_pages() is of course
safe), and cannot be sure that every follow_page() caller is prepared to
sleep - ia64's xencomm_vtop()? Now, THP's wait_split_huge_page() can
already sleep there, since anon_vma locking was changed to mutex, but
maybe that's somehow excluded.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Petr Holasek <pholasek@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Think of struct rmap_item as an extension of struct page (restricted to
MADV_MERGEABLE areas): there may be a lot of them, we need to keep them
small, especially on 32-bit architectures of limited lowmem.

Siting "int nid" after "unsigned int checksum" works nicely on 64-bit,
making no change to its 64-byte struct rmap_item; but bloats the 32-bit
struct rmap_item from (nicely cache-aligned) 32 bytes to 36 bytes, which
rounds up to 40 bytes once allocated from slab.  We'd better avoid that.

Hey, I only just remembered that the anon_vma pointer in struct
rmap_item has no purpose until the rmap_item is hung from a stable tree
node (which has its own nid field); and rmap_item's nid field no purpose
than to say which tree root to tell rb_erase() when unlinking from an
unstable tree.

Double them up in a union.  There's just one place where we set anon_vma
early (when we already hold mmap_sem): now we must remove tree_rmap_item
from its unstable tree there, before overwriting nid.  No need to
spatter BUG()s around: we'd be seeing oopses if this were wrong.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Petr Holasek <pholasek@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

An inconsistency emerged in reviewing the NUMA node changes to KSM: when
meeting a page from the wrong NUMA node in a stable tree, we say that
it's okay for comparisons, but not as a leaf for merging; whereas when
meeting a page from the wrong NUMA node in an unstable tree, we bail out
immediately.

Now, it might be that a wrong NUMA node in an unstable tree is more
likely to correlate with instablility (different content, with rbnode
now misplaced) than page migration; but even so, we are accustomed to
instablility in the unstable tree.

Without strong evidence for which strategy is generally better, I'd
rather be consistent with what's done in the stable tree: accept a page
from the wrong NUMA node for comparison, but not as a leaf for merging.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Petr Holasek <pholasek@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Added slightly more detail to the Documentation of merge_across_nodes, a
few comments in areas indicated by review, and renamed get_ksm_page()'s
argument from "locked" to "lock_it".  No functional change.
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Petr Holasek <pholasek@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix several mempolicy leaks in the tmpfs mount logic.  These leaks are
slow - on the order of one object leaked per mount attempt.

Leak 1 (umount doesn't free mpol allocated in mount):
    while true; do
        mount -t tmpfs -o mpol=interleave,size=100M nodev /mnt
        umount /mnt
    done

Leak 2 (errors parsing remount options will leak mpol):
    mount -t tmpfs -o size=100M nodev /mnt
    while true; do
        mount -o remount,mpol=interleave,size=x /mnt 2> /dev/null
    done
    umount /mnt

Leak 3 (multiple mpol per mount leak mpol):
    while true; do
        mount -t tmpfs -o mpol=interleave,mpol=interleave,size=100M nodev /mnt
        umount /mnt
    done

This patch fixes all of the above.  I could have broken the patch into
three pieces but is seemed easier to review as one.

[akpm@linux-foundation.org: fix handling of mpol_parse_str() errors, per Hugh]
Signed-off-by: NGreg Thelen <gthelen@google.com>
Acked-by: NHugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The tmpfs remount logic preserves filesystem mempolicy if the mpol=M
option is not specified in the remount request.  A new policy can be
specified if mpol=M is given.

Before this patch remounting an mpol bound tmpfs without specifying
mpol= mount option in the remount request would set the filesystem's
mempolicy object to a freed mempolicy object.

To reproduce the problem boot a DEBUG_PAGEALLOC kernel and run:
    # mkdir /tmp/x

    # mount -t tmpfs -o size=100M,mpol=interleave nodev /tmp/x

    # grep /tmp/x /proc/mounts
    nodev /tmp/x tmpfs rw,relatime,size=102400k,mpol=interleave:0-3 0 0

    # mount -o remount,size=200M nodev /tmp/x

    # grep /tmp/x /proc/mounts
    nodev /tmp/x tmpfs rw,relatime,size=204800k,mpol=??? 0 0
        # note ? garbage in mpol=... output above

    # dd if=/dev/zero of=/tmp/x/f count=1
        # panic here

Panic:
    BUG: unable to handle kernel NULL pointer dereference at           (null)
    IP: [<          (null)>]           (null)
    [...]
    Oops: 0010 [#1] SMP DEBUG_PAGEALLOC
    Call Trace:
      mpol_shared_policy_init+0xa5/0x160
      shmem_get_inode+0x209/0x270
      shmem_mknod+0x3e/0xf0
      shmem_create+0x18/0x20
      vfs_create+0xb5/0x130
      do_last+0x9a1/0xea0
      path_openat+0xb3/0x4d0
      do_filp_open+0x42/0xa0
      do_sys_open+0xfe/0x1e0
      compat_sys_open+0x1b/0x20
      cstar_dispatch+0x7/0x1f

Non-debug kernels will not crash immediately because referencing the
dangling mpol will not cause a fault.  Instead the filesystem will
reference a freed mempolicy object, which will cause unpredictable
behavior.

The problem boils down to a dropped mpol reference below if
shmem_parse_options() does not allocate a new mpol:

    config = *sbinfo
    shmem_parse_options(data, &config, true)
    mpol_put(sbinfo->mpol)
    sbinfo->mpol = config.mpol  /* BUG: saves unreferenced mpol */

This patch avoids the crash by not releasing the mempolicy if
shmem_parse_options() doesn't create a new mpol.

How far back does this issue go? I see it in both 2.6.36 and 3.3.  I did
not look back further.
Signed-off-by: NGreg Thelen <gthelen@google.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Rob van der Heij reported the following (paraphrased) on private mail.

	The scenario is that I want to avoid backups to fill up the page
	cache and purge stuff that is more likely to be used again (this is
	with s390x Linux on z/VM, so I don't give it as much memory that
	we don't care anymore). So I have something with LD_PRELOAD that
	intercepts the close() call (from tar, in this case) and issues
	a posix_fadvise() just before closing the file.

	This mostly works, except for small files (less than 14 pages)
	that remains in page cache after the face.

Unfortunately Rob has not had a chance to test this exact patch but the
test program below should be reproducing the problem he described.

The issue is the per-cpu pagevecs for LRU additions.  If the pages are
added by one CPU but fadvise() is called on another then the pages
remain resident as the invalidate_mapping_pages() only drains the local
pagevecs via its call to pagevec_release().  The user-visible effect is
that a program that uses fadvise() properly is not obeyed.

A possible fix for this is to put the necessary smarts into
invalidate_mapping_pages() to globally drain the LRU pagevecs if a
pagevec page could not be discarded.  The downside with this is that an
inode cache shrink would send a global IPI and memory pressure
potentially causing global IPI storms is very undesirable.

Instead, this patch adds a check during fadvise(POSIX_FADV_DONTNEED) to
check if invalidate_mapping_pages() discarded all the requested pages.
If a subset of pages are discarded it drains the LRU pagevecs and tries
again.  If the second attempt fails, it assumes it is due to the pages
being mapped, locked or dirty and does not care.  With this patch, an
application using fadvise() correctly will be obeyed but there is a
downside that a malicious application can force the kernel to send
global IPIs and increase overhead.

If accepted, I would like this to be considered as a -stable candidate.
It's not an urgent issue but it's a system call that is not working as
advertised which is weak.

The following test program demonstrates the problem.  It should never
report that pages are still resident but will without this patch.  It
assumes that CPU 0 and 1 exist.

int main() {
	int fd;
	int pagesize = getpagesize();
	ssize_t written = 0, expected;
	char *buf;
	unsigned char *vec;
	int resident, i;
	cpu_set_t set;

	/* Prepare a buffer for writing */
	expected = FILESIZE_PAGES * pagesize;
	buf = malloc(expected + 1);
	if (buf == NULL) {
		printf("ENOMEM\n");
		exit(EXIT_FAILURE);
	}
	buf[expected] = 0;
	memset(buf, 'a', expected);

	/* Prepare the mincore vec */
	vec = malloc(FILESIZE_PAGES);
	if (vec == NULL) {
		printf("ENOMEM\n");
		exit(EXIT_FAILURE);
	}

	/* Bind ourselves to CPU 0 */
	CPU_ZERO(&set);
	CPU_SET(0, &set);
	if (sched_setaffinity(getpid(), sizeof(set), &set) == -1) {
		perror("sched_setaffinity");
		exit(EXIT_FAILURE);
	}

	/* open file, unlink and write buffer */
	fd = open("fadvise-test-file", O_CREAT|O_EXCL|O_RDWR);
	if (fd == -1) {
		perror("open");
		exit(EXIT_FAILURE);
	}
	unlink("fadvise-test-file");
	while (written < expected) {
		ssize_t this_write;
		this_write = write(fd, buf + written, expected - written);

		if (this_write == -1) {
			perror("write");
			exit(EXIT_FAILURE);
		}

		written += this_write;
	}
	free(buf);

	/*
	 * Force ourselves to another CPU. If fadvise only flushes the local
	 * CPUs pagevecs then the fadvise will fail to discard all file pages
	 */
	CPU_ZERO(&set);
	CPU_SET(1, &set);
	if (sched_setaffinity(getpid(), sizeof(set), &set) == -1) {
		perror("sched_setaffinity");
		exit(EXIT_FAILURE);
	}

	/* sync and fadvise to discard the page cache */
	fsync(fd);
	if (posix_fadvise(fd, 0, expected, POSIX_FADV_DONTNEED) == -1) {
		perror("posix_fadvise");
		exit(EXIT_FAILURE);
	}

	/* map the file and use mincore to see which parts of it are resident */
	buf = mmap(NULL, expected, PROT_READ, MAP_SHARED, fd, 0);
	if (buf == NULL) {
		perror("mmap");
		exit(EXIT_FAILURE);
	}
	if (mincore(buf, expected, vec) == -1) {
		perror("mincore");
		exit(EXIT_FAILURE);
	}

	/* Check residency */
	for (i = 0, resident = 0; i < FILESIZE_PAGES; i++) {
		if (vec[i])
			resident++;
	}
	if (resident != 0) {
		printf("Nr unexpected pages resident: %d\n", resident);
		exit(EXIT_FAILURE);
	}

	munmap(buf, expected);
	close(fd);
	free(vec);
	exit(EXIT_SUCCESS);
}
Signed-off-by: NMel Gorman <mgorman@suse.de>
Reported-by: NRob van der Heij <rvdheij@gmail.com>
Tested-by: NRob van der Heij <rvdheij@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We at SGI have a need to address some very high physical address ranges
with our GRU (global reference unit), sometimes across partitioned
machine boundaries and sometimes with larger addresses than the cpu
supports.  We do this with the aid of our own 'extended vma' module
which mimics the vma.  When something (either unmap or exit) frees an
'extended vma' we use the mmu notifiers to clean them up.

We had been able to mimic the functions
__mmu_notifier_invalidate_range_start() and
__mmu_notifier_invalidate_range_end() by locking the per-mm lock and
walking the per-mm notifier list.  But with the change to a global srcu
lock (static in mmu_notifier.c) we can no longer do that.  Our module has
no access to that lock.

So we request that these two functions be exported.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Acked-by: NRobin Holt <holt@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This change adds a follow_page_mask function which is equivalent to
follow_page, but with an extra page_mask argument.

follow_page_mask sets *page_mask to HPAGE_PMD_NR - 1 when it encounters
a THP page, and to 0 in other cases.

__get_user_pages() makes use of this in order to accelerate populating
THP ranges - that is, when both the pages and vmas arrays are NULL, we
don't need to iterate HPAGE_PMD_NR times to cover a single THP page (and
we also avoid taking mm->page_table_lock that many times).
Signed-off-by: NMichel Lespinasse <walken@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use long type for page counts in mm_populate() so as to avoid integer
overflow when running the following test code:

int main(void) {
  void *p = mmap(NULL, 0x100000000000, PROT_READ,
                 MAP_PRIVATE | MAP_ANON, -1, 0);
  printf("p: %p\n", p);
  mlockall(MCL_CURRENT);
  printf("done\n");
  return 0;
}
Signed-off-by: NMichel Lespinasse <walken@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

nr_free_zone_pages(), nr_free_buffer_pages() and nr_free_pagecache_pages()
are horribly badly named, so accurately document them with code comments
in case of the misuse of them.

[akpm@linux-foundation.org: tweak comments]
Reviewed-by: NRandy Dunlap <rdunlap@infradead.org>
Signed-off-by: NZhang Yanfei <zhangyanfei@cn.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

error_states[] has two separate states "unevictable LRU page" and
"mlocked LRU page", and the former one has the higher priority now.  But
because of that the latter one is rarely chosen because pages with
PageMlocked highly likely have PG_unevictable set.  On the other hand,
PG_unevictable without PageMlocked is common for ramfs or SHM_LOCKed
shared memory, so reversing the priority of these two states helps us
clearly distinguish them.
Signed-off-by: NNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Chen Gong <gong.chen@linux.intel.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

memory_failure() can't handle memory errors on mlocked pages correctly,
because page_action() judges such errors as ones on "unknown pages"
instead of ones on "unevictable LRU page" or "mlocked LRU page".  In
order to determine page_state page_action() checks page flags at the
timing of the judgement, but such page flags are not the same with those
just after memory_failure() is called, because memory_failure() does
unmapping of the error pages before doing page_action().  This unmapping
changes the page state, especially page_remove_rmap() (called from
try_to_unmap_one()) clears PG_mlocked, so page_action() can't catch
mlocked pages after that.

With this patch, we store the page flag of the error page before doing
unmap, and (only) if the first check with page flags at the time decided
the error page is unknown, we do the second check with the stored page
flag.  This implementation doesn't change error handling for the page
types for which the first check can determine the page state correctly.

[akpm@linux-foundation.org: tweak comments]
Signed-off-by: NNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Chen Gong <gong.chen@linux.intel.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Whilst I run the risk of a flogging for disloyalty to the Lord of Sealand,
I do have CONFIG_MEMCG=y CONFIG_MEMCG_KMEM not set, and grow tired of the
"mm/memcontrol.c:4972:12: warning: `memcg_propagate_kmem' defined but not
used [-Wunused-function]" seen in 3.8-rc: move the #ifdef outwards.
Signed-off-by: NHugh Dickins <hughd@google.com>
Acked-by: NMichal Hocko <mhocko@suse.cz>
Cc: Glauber Costa <glommer@parallels.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This member of struct virtio_chan is calculated from nr_free_buffer_pages
so change its type to unsigned long in case of overflow.
Signed-off-by: NZhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Cc: Ron Minnich <rminnich@sandia.gov>
Cc: Latchesar Ionkov <lucho@ionkov.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This variable is calculated from nr_free_pagecache_pages so
change its type to unsigned long.
Signed-off-by: NZhang Yanfei <zhangyanfei@cn.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The three variables are calculated from nr_free_buffer_pages so change
their types to unsigned long in case of overflow.
Signed-off-by: NZhang Yanfei <zhangyanfei@cn.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

max_buffer_heads is calculated from nr_free_buffer_pages(), so change
its type to unsigned long in case of overflow.
Signed-off-by: NZhang Yanfei <zhangyanfei@cn.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now the function nr_free_buffer_pages returns unsigned long, so use %ld
to print its return value.
Signed-off-by: NZhang Yanfei <zhangyanfei@cn.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently, the amount of RAM that functions nr_free_*_pages return is
held in unsigned int.  But in machines with big memory (exceeding 16TB),
the amount may be incorrect because of overflow, so fix it.
Signed-off-by: NZhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Simon Horman <horms@verge.net.au>
Cc: Julian Anastasov <ja@ssi.bg>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Cc: Ron Minnich <rminnich@sandia.gov>
Cc: Latchesar Ionkov <lucho@ionkov.net>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We should encourage all memcg controller initialization independent on a
specific mem_cgroup to be done here rather than exploit css_alloc
callback and assume that nothing happens before root cgroup is created.
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <htejun@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

memcg_stock are currently initialized during the root cgroup allocation
which is OK but it pointlessly pollutes memcg allocation code with
something that can be called when the memcg subsystem is initialized by
mem_cgroup_init along with other controller specific parts.

This patch wraps the current memcg_stock initialization code into a
helper calls it from the controller subsystem initialization code.
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <htejun@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Per-node-zone soft limit tree is currently initialized when the root
cgroup is created which is OK but it pointlessly pollutes memcg
allocation code with something that can be called when the memcg
subsystem is initialized by mem_cgroup_init along with other controller
specific parts.

While we are at it let's make mem_cgroup_soft_limit_tree_init void
because it doesn't make much sense to report memory failure because if
we fail to allocate memory that early during the boot then we are
screwed anyway (this saves some code).
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <htejun@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Recently, Luigi reported there are lots of free swap space when OOM
happens.  It's easily reproduced on zram-over-swap, where many instance
of memory hogs are running and laptop_mode is enabled.  He said there
was no problem when he disabled laptop_mode.  The problem when I
investigate problem is following as.

Assumption for easy explanation: There are no page cache page in system
because they all are already reclaimed.

1. try_to_free_pages disable may_writepage when laptop_mode is enabled.
2. shrink_inactive_list isolates victim pages from inactive anon lru list.
3. shrink_page_list adds them to swapcache via add_to_swap but it doesn't
   pageout because sc->may_writepage is 0 so the page is rotated back into
   inactive anon lru list. The add_to_swap made the page Dirty by SetPageDirty.
4. 3 couldn't reclaim any pages so do_try_to_free_pages increase priority and
   retry reclaim with higher priority.
5. shrink_inactlive_list try to isolate victim pages from inactive anon lru list
   but got failed because it try to isolate pages with ISOLATE_CLEAN mode but
   inactive anon lru list is full of dirty pages by 3 so it just returns
   without  any reclaim progress.
6. do_try_to_free_pages doesn't set may_writepage due to zero total_scanned.
   Because sc->nr_scanned is increased by shrink_page_list but we don't call
   shrink_page_list in 5 due to short of isolated pages.

Above loop is continued until OOM happens.

The problem didn't happen before [1] was merged because old logic's
isolatation in shrink_inactive_list was successful and tried to call
shrink_page_list to pageout them but it still ends up failed to page out
by may_writepage.  But important point is that sc->nr_scanned was
increased although we couldn't swap out them so do_try_to_free_pages
could set may_writepages.

Since commit f80c0673 ("mm: zone_reclaim: make isolate_lru_page()
filter-aware") was introduced, it's not a good idea any more to depends
on only the number of scanned pages for setting may_writepage.  So this
patch adds new trigger point of setting may_writepage as below
DEF_PRIOIRTY - 2 which is used to show the significant memory pressure
in VM so it's good fit for our purpose which would be better to lose
power saving or clickety rather than OOM killing.
Signed-off-by: NMinchan Kim <minchan@kernel.org>
Reported-by: NLuigi Semenzato <semenzato@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make a sweep through mm/ and convert code that uses -1 directly to using
the more appropriate NUMA_NO_NODE.
Signed-off-by: NDavid Rientjes <rientjes@google.com>
Reviewed-by: NYasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>