Currently we are keeping faulted page locked throughout whole __do_fault
call (except for page_mkwrite code path) after calling file system's fault
code.  If we do early COW, we allocate a new page which has to be charged
for a memcg (mem_cgroup_newpage_charge).

This function, however, might block for unbounded amount of time if memcg
oom killer is disabled or fork-bomb is running because the only way out of
the OOM situation is either an external event or OOM-situation fix.

In the end we are keeping the faulted page locked and blocking other
processes from faulting it in which is not good at all because we are
basically punishing potentially an unrelated process for OOM condition in
a different group (I have seen stuck system because of ld-2.11.1.so being
locked).

We can do test easily.

 % cgcreate -g memory:A
 % cgset -r memory.limit_in_bytes=64M A
 % cgset -r memory.memsw.limit_in_bytes=64M A
 % cd kernel_dir; cgexec -g memory:A make -j

Then, the whole system will live-locked until you kill 'make -j'
by hands (or push reboot...) This is because some important page in a
a shared library are locked.

Considering again, the new page is not necessary to be allocated
with lock_page() held. And usual page allocation may dive into
long memory reclaim loop with holding lock_page() and can cause
very long latency.

There are 3 ways.
  1. do allocation/charge before lock_page()
     Pros. - simple and can handle page allocation in the same manner.
             This will reduce holding time of lock_page() in general.
     Cons. - we do page allocation even if ->fault() returns error.

  2. do charge after unlock_page(). Even if charge fails, it's just OOM.
     Pros. - no impact to non-memcg path.
     Cons. - implemenation requires special cares of LRU and we need to modify
             page_add_new_anon_rmap()...

  3. do unlock->charge->lock again method.
     Pros. - no impact to non-memcg path.
     Cons. - This may kill LOCK_PAGE_RETRY optimization. We need to release
             lock and get it again...

This patch moves "charge" and memory allocation for COW page
before lock_page(). Then, we can avoid scanning LRU with holding
a lock on a page and latency under lock_page() will be reduced.

Then, above livelock disappears.

[akpm@linux-foundation.org: fix code layout]
Signed-off-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reported-by: NLutz Vieweg <lvml@5t9.de>
Original-idea-by: NMichal Hocko <mhocko@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Ying Han <yinghan@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

ZAP_BLOCK_SIZE became unused in the preemptible-mmu_gather work ("mm:
Remove i_mmap_lock lockbreak").  So zap it.

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

__tlb_remove_page() switches to a new batch page, but still checks space
in the old batch.  This check always fails, and causes a forced tlb flush.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

You would expect to find vmtruncate_range() next to vmtruncate() in
mm/truncate.c: move it there.
Signed-off-by: NHugh Dickins <hughd@google.com>
Acked-by: NChristoph Hellwig <hch@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Running a ktest.pl test, I hit the following bug on x86_32:

  ------------[ cut here ]------------
  WARNING: at arch/x86/mm/highmem_32.c:81 __kunmap_atomic+0x64/0xc1()
   Hardware name:
  Modules linked in:
  Pid: 93, comm: sh Not tainted 2.6.39-test+ #1
  Call Trace:
   [<c04450da>] warn_slowpath_common+0x7c/0x91
   [<c042f5df>] ? __kunmap_atomic+0x64/0xc1
   [<c042f5df>] ? __kunmap_atomic+0x64/0xc1^M
   [<c0445111>] warn_slowpath_null+0x22/0x24
   [<c042f5df>] __kunmap_atomic+0x64/0xc1
   [<c04d4a22>] unmap_vmas+0x43a/0x4e0
   [<c04d9065>] exit_mmap+0x91/0xd2
   [<c0443057>] mmput+0x43/0xad
   [<c0448358>] exit_mm+0x111/0x119
   [<c044855f>] do_exit+0x1ff/0x5fa
   [<c0454ea2>] ? set_current_blocked+0x3c/0x40
   [<c0454f24>] ? sigprocmask+0x7e/0x8e
   [<c0448b55>] do_group_exit+0x65/0x88
   [<c0448b90>] sys_exit_group+0x18/0x1c
   [<c0c3915f>] sysenter_do_call+0x12/0x38
  ---[ end trace 8055f74ea3c0eb62 ]---

Running a ktest.pl git bisect, found the culprit: commit e303297e
("mm: extended batches for generic mmu_gather")

But although this was the commit triggering the bug, it was not the one
originally responsible for the bug.  That was commit d16dfc55 ("mm:
mmu_gather rework").

The code in zap_pte_range() has something that looks like the following:

	pte =  pte_offset_map_lock(mm, pmd, addr, &ptl);
	do {
		[...]
	} while (pte++, addr += PAGE_SIZE, addr != end);
	pte_unmap_unlock(pte - 1, ptl);

The pte starts off pointing at the first element in the page table
directory that was returned by the pte_offset_map_lock().  When it's done
with the page, pte will be pointing to anything between the next entry and
the first entry of the next page inclusive.  By doing a pte - 1, this puts
the pte back onto the original page, which is all that pte_unmap_unlock()
needs.

In most archs (64 bit), this is not an issue as the pte is ignored in the
pte_unmap_unlock().  But on 32 bit archs, where things may be kmapped, it
is essential that the pte passed to pte_unmap_unlock() resides on the same
page that was given by pte_offest_map_lock().

The problem came in d16dfc55 ("mm: mmu_gather rework") where it introduced
a "break;" from the while loop.  This alone did not seem to easily trigger
the bug.  But the modifications made by e303297e caused that "break;" to
be hit on the first iteration, before the pte++.

The pte not being incremented will now cause pte_unmap_unlock(pte - 1) to
be pointing to the previous page.  This will cause the wrong page to be
unmapped, and also trigger the warning above.

The simple solution is to just save the pointer given by
pte_offset_map_lock() and use it in the unlock.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix new kernel-doc warnings in mm/memory.c:

  Warning(mm/memory.c:1327): No description found for parameter 'tlb'
  Warning(mm/memory.c:1327): Excess function parameter 'tlbp' description in 'unmap_vmas'
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Two new stats in per-memcg memory.stat which tracks the number of page
faults and number of major page faults.

  "pgfault"
  "pgmajfault"

They are different from "pgpgin"/"pgpgout" stat which count number of
pages charged/discharged to the cgroup and have no meaning of reading/
writing page to disk.

It is valuable to track the two stats for both measuring application's
performance as well as the efficiency of the kernel page reclaim path.
Counting pagefaults per process is useful, but we also need the aggregated
value since processes are monitored and controlled in cgroup basis in
memcg.

Functional test: check the total number of pgfault/pgmajfault of all
memcgs and compare with global vmstat value:

  $ cat /proc/vmstat | grep fault
  pgfault 1070751
  pgmajfault 553

  $ cat /dev/cgroup/memory.stat | grep fault
  pgfault 1071138
  pgmajfault 553
  total_pgfault 1071142
  total_pgmajfault 553

  $ cat /dev/cgroup/A/memory.stat | grep fault
  pgfault 199
  pgmajfault 0
  total_pgfault 199
  total_pgmajfault 0

Performance test: run page fault test(pft) wit 16 thread on faulting in
15G anon pages in 16G container.  There is no regression noticed on the
"flt/cpu/s"

Sample output from pft:

  TAG pft:anon-sys-default:
    Gb  Thr CLine   User     System     Wall    flt/cpu/s fault/wsec
    15   16   1     0.67s   233.41s    14.76s   16798.546 266356.260

  +-------------------------------------------------------------------------+
      N           Min           Max        Median           Avg        Stddev
  x  10     16682.962     17344.027     16913.524     16928.812      166.5362
  +  10     16695.568     16923.896     16820.604     16824.652     84.816568
  No difference proven at 95.0% confidence

[akpm@linux-foundation.org: fix build]
[hughd@google.com: shmem fix]
Signed-off-by: NYing Han <yinghan@google.com>
Acked-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reviewed-by: NMinchan Kim <minchan.kim@gmail.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: NBalbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-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 type of vma->vm_flags is 'unsigned long'. Neither 'int' nor
'unsigned int'. This patch fixes such misuse.
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
[ Changed to use a typedef - we'll extend it to cover more cases
  later, since there has been discussion about making it a 64-bit
  type..                      - Linus ]
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Some of these functions have grown beyond inline sanity, move them
out-of-line.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Requested-by: NAndrew Morton <akpm@linux-foundation.org>
Requested-by: NHugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Straightforward conversion of i_mmap_lock to a mutex.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Miller <davem@davemloft.net>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Tony Luck <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Hugh says:
 "The only significant loser, I think, would be page reclaim (when
  concurrent with truncation): could spin for a long time waiting for
  the i_mmap_mutex it expects would soon be dropped? "

Counter points:
 - cpu contention makes the spin stop (need_resched())
 - zap pages should be freeing pages at a higher rate than reclaim
   ever can

I think the simplification of the truncate code is definitely worth it.

Effectively reverts: 2aa15890 ("mm: prevent concurrent
unmap_mapping_range() on the same inode") and takes out the code that
caused its problem.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Miller <davem@davemloft.net>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Instead of using a single batch (the small on-stack, or an allocated
page), try and extend the batch every time it runs out and only flush once
either the extend fails or we're done.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Requested-by: NNick Piggin <npiggin@kernel.dk>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Miller <davem@davemloft.net>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In case other architectures require RCU freed page-tables to implement
gup_fast() and software filled hashes and similar things, provide the
means to do so by moving the logic into generic code.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Requested-by: NDavid Miller <davem@davemloft.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Tony Luck <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Rework the existing mmu_gather infrastructure.

The direct purpose of these patches was to allow preemptible mmu_gather,
but even without that I think these patches provide an improvement to the
status quo.

The first 9 patches rework the mmu_gather infrastructure.  For review
purpose I've split them into generic and per-arch patches with the last of
those a generic cleanup.

The next patch provides generic RCU page-table freeing, and the followup
is a patch converting s390 to use this.  I've also got 4 patches from
DaveM lined up (not included in this series) that uses this to implement
gup_fast() for sparc64.

Then there is one patch that extends the generic mmu_gather batching.

After that follow the mm preemptibility patches, these make part of the mm
a lot more preemptible.  It converts i_mmap_lock and anon_vma->lock to
mutexes which together with the mmu_gather rework makes mmu_gather
preemptible as well.

Making i_mmap_lock a mutex also enables a clean-up of the truncate code.

This also allows for preemptible mmu_notifiers, something that XPMEM I
think wants.

Furthermore, it removes the new and universially detested unmap_mutex.

This patch:

Remove the first obstacle towards a fully preemptible mmu_gather.

The current scheme assumes mmu_gather is always done with preemption
disabled and uses per-cpu storage for the page batches.  Change this to
try and allocate a page for batching and in case of failure, use a small
on-stack array to make some progress.

Preemptible mmu_gather is desired in general and usable once i_mmap_lock
becomes a mutex.  Doing it before the mutex conversion saves us from
having to rework the code by moving the mmu_gather bits inside the
pte_lock.

Also avoid flushing the tlb batches from under the pte lock, this is
useful even without the i_mmap_lock conversion as it significantly reduces
pte lock hold times.

[akpm@linux-foundation.org: fix comment tpyo]
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Miller <davem@davemloft.net>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Tony Luck <tony.luck@intel.com>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: NHugh Dickins <hughd@google.com>
Acked-by: NMel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently we have expand_upwards exported while expand_downwards is
accessible only via expand_stack or expand_stack_downwards.

check_stack_guard_page is a nice example of the asymmetry.  It uses
expand_stack for VM_GROWSDOWN while expand_upwards is called for
VM_GROWSUP case.

Let's clean this up by exporting both functions and make those names
consistent.  Let's use expand_{upwards,downwards} because expanding
doesn't always involve stack manipulation (an example is
ia64_do_page_fault which uses expand_upwards for registers backing store
expansion).  expand_downwards has to be defined for both
CONFIG_STACK_GROWS{UP,DOWN} because get_arg_page calls the downwards
version in the early process initialization phase for growsup
configuration.
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Linux kernel excludes guard page when performing mlock on a VMA with
down-growing stack. However, some architectures have up-growing stack
and locking the guard page should be excluded in this case too.

This patch fixes lvm2 on PA-RISC (and possibly other architectures with
up-growing stack). lvm2 calculates number of used pages when locking and
when unlocking and reports an internal error if the numbers mismatch.

[ Patch changed fairly extensively to also fix /proc/<pid>/maps for the
  grows-up case, and to move things around a bit to clean it all up and
  share the infrstructure with the /proc bits.

  Tested on ia64 that has both grow-up and grow-down segments  - Linus ]
Signed-off-by: NMikulas Patocka <mikulas@artax.karlin.mff.cuni.cz>
Tested-by: NTony Luck <tony.luck@gmail.com>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The logic in __get_user_pages() used to skip the stack guard page lookup
whenever the caller wasn't interested in seeing what the actual page
was.  But Michel Lespinasse points out that there are cases where we
don't care about the physical page itself (so 'pages' may be NULL), but
do want to make sure a page is mapped into the virtual address space.

So using the existence of the "pages" array as an indication of whether
to look up the guard page or not isn't actually so great, and we really
should just use the FOLL_MLOCK bit.  But because that bit was only set
for the VM_LOCKED case (and not all vma's necessarily have it, even for
mlock()), we couldn't do that originally.

Fix that by moving the VM_LOCKED check deeper into the call-chain, which
actually simplifies many things.  Now mlock() gets simpler, and we can
also check for FOLL_MLOCK in __get_user_pages() and the code ends up
much more straightforward.
Reported-and-reviewed-by: NMichel Lespinasse <walken@google.com>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

With transparent hugepage support, handle_mm_fault() has to be careful
that a normal PMD has been established before handling a PTE fault.  To
achieve this, it used __pte_alloc() directly instead of pte_alloc_map as
pte_alloc_map is unsafe to run against a huge PMD.  pte_offset_map() is
called once it is known the PMD is safe.

pte_alloc_map() is smart enough to check if a PTE is already present
before calling __pte_alloc but this check was lost.  As a consequence,
PTEs may be allocated unnecessarily and the page table lock taken.  Thi
useless PTE does get cleaned up but it's a performance hit which is
visible in page_test from aim9.

This patch simply re-adds the check normally done by pte_alloc_map to
check if the PTE needs to be allocated before taking the page table lock.
The effect is noticable in page_test from aim9.

  AIM9
                  2.6.38-vanilla 2.6.38-checkptenone
  creat-clo      446.10 ( 0.00%)   424.47 (-5.10%)
  page_test       38.10 ( 0.00%)    42.04 ( 9.37%)
  brk_test        52.45 ( 0.00%)    51.57 (-1.71%)
  exec_test      382.00 ( 0.00%)   456.90 (16.39%)
  fork_test       60.11 ( 0.00%)    67.79 (11.34%)
  MMTests Statistics: duration
  Total Elapsed Time (seconds)                611.90    612.22

(While this affects 2.6.38, it is a performance rather than a functional
bug and normally outside the rules -stable.  While the big performance
differences are to a microbench, the difference in fork and exec
performance may be significant enough that -stable wants to consider the
patch)
Reported-by: NRaz Ben Yehuda <raziebe@gmail.com>
Signed-off-by: NMel Gorman <mgorman@suse.de>
Reviewed-by: NRik van Riel <riel@redhat.com>
Reviewed-by: NAndrea Arcangeli <aarcange@redhat.com>
Reviewed-by: NMinchan Kim <minchan.kim@gmail.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: <stable@kernel.org>		[2.6.38.x]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In __access_remote_vm() we need to check that we have found the right
vma, not the following vma before we try to access it.  Otherwise we
might call the vma's access routine with an address which does not fall
inside the vma.

It was discovered on a current kernel but with an unreleased driver,
from memory it was strace leading to a kernel bad access, but it
obviously depends on what the access implementation does.

Looking at other access implementations I only see:

  $ git grep -A 5 vm_operations|grep access
  arch/powerpc/platforms/cell/spufs/file.c-	.access = spufs_mem_mmap_access,
  arch/x86/pci/i386.c-	.access = generic_access_phys,
  drivers/char/mem.c-	.access = generic_access_phys
  fs/sysfs/bin.c-	.access		= bin_access,

The spufs one looks like it might behave badly given the wrong vma, it
assumes vma->vm_file->private_data is a spu_context, and looks like it
would probably blow up pretty quickly if it wasn't.

generic_access_phys() only uses the vma to check vm_flags and get the
mm, and then walks page tables using the address.  So it should bail on
the vm_flags check, or at worst let you access some other VM_IO mapping.

And bin_access() just proxies to another access implementation.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Reviewed-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 53a7706d ("mlock: do not hold mmap_sem for extended periods
of time") changed mlock() to care about the exact number of pages that
__get_user_pages() had brought it.  Before, it would only care about
errors.

And that doesn't work, because we also handled one page specially in
__mlock_vma_pages_range(), namely the stack guard page.  So when that
case was handled, the number of pages that the function returned was off
by one.  In particular, it could be zero, and then the caller would end
up not making any progress at all.

Rather than try to fix up that off-by-one error for the mlock case
specially, this just moves the logic to handle the stack guard page
into__get_user_pages() itself, thus making all the counts come out
right automatically.
Reported-by: NRobert Święcki <robert@swiecki.net>
Cc: Hugh Dickins <hughd@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix mm/memory.c incorrect kernel-doc function notation:

  Warning(mm/memory.c:3718): Cannot understand  * @access_remote_vm - access another process' address space
   on line 3718 - I thought it was a doc line
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Remove initialization of vaiable in caller of memory cgroup function.
Actually, it's return value of memcg function but it's initialized in
caller.

Some memory cgroup uses following style to bring the result of start
function to the end function for avoiding races.

   mem_cgroup_start_A(&(*ptr))
   /* Something very complicated can happen here. */
   mem_cgroup_end_A(*ptr)

In some calls, *ptr should be initialized to NULL be caller.  But it's
ugly.  This patch fixes that *ptr is initialized by _start function.
Signed-off-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NDaisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Provide an alternative to access_process_vm that allows the caller to obtain a
reference to the supplied mm_struct.
Signed-off-by: NStephen Wilson <wilsons@start.ca>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Introduce an internal helper __access_remote_vm and base access_process_vm on
top of it.  This new method may be called with a NULL task_struct if page fault
accounting is not desired.  This code will be shared with a new address space
accessor that is independent of task_struct.
Signed-off-by: NStephen Wilson <wilsons@start.ca>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

We now check if a requested user page overlaps a gate vma using the supplied mm
instead of the supplied task.  The given task is now used solely for accounting
purposes and may be NULL.
Signed-off-by: NStephen Wilson <wilsons@start.ca>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Now that gate vma's are referenced with respect to a particular mm and not a
particular task it only makes sense to propagate the change to this predicate as
well.
Signed-off-by: NStephen Wilson <wilsons@start.ca>
Reviewed-by: NMichel Lespinasse <walken@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Morally, the question of whether an address lies in a gate vma should be asked
with respect to an mm, not a particular task.  Moreover, dropping the dependency
on task_struct will help make existing and future operations on mm's more
flexible and convenient.
Signed-off-by: NStephen Wilson <wilsons@start.ca>
Reviewed-by: NMichel Lespinasse <walken@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Morally, the presence of a gate vma is more an attribute of a particular mm than
a particular task.  Moreover, dropping the dependency on task_struct will help
make both existing and future operations on mm's more flexible and convenient.
Signed-off-by: NStephen Wilson <wilsons@start.ca>
Reviewed-by: NMichel Lespinasse <walken@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

GUP user may want to try to acquire a reference to a page if it is already
in memory, but not if IO, to bring it in, is needed.  For example KVM may
tell vcpu to schedule another guest process if current one is trying to
access swapped out page.  Meanwhile, the page will be swapped in and the
guest process, that depends on it, will be able to run again.

This patch adds FAULT_FLAG_RETRY_NOWAIT (suggested by Linus) and
FOLL_NOWAIT follow_page flags.  FAULT_FLAG_RETRY_NOWAIT, when used in
conjunction with VM_FAULT_ALLOW_RETRY, indicates to handle_mm_fault that
it shouldn't drop mmap_sem and wait on a page, but return VM_FAULT_RETRY
instead.

[akpm@linux-foundation.org: improve FOLL_NOWAIT comment]
Signed-off-by: NGleb Natapov <gleb@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Hugh Dickins <hughd@google.com>
Acked-by: NRik van Riel <riel@redhat.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make __get_user_pages return -EHWPOISON for HWPOISON page only if
FOLL_HWPOISON is specified.  With this patch, the interested callers
can distinguish HWPOISON pages from general FAULT pages, while other
callers will still get -EFAULT for all these pages, so the user space
interface need not to be changed.

This feature is needed by KVM, where UCR MCE should be relayed to
guest for HWPOISON page, while instruction emulation and MMIO will be
tried for general FAULT page.

The idea comes from Andrew Morton.
Signed-off-by: NHuang Ying <ying.huang@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

In most cases, get_user_pages and get_user_pages_fast should be used
to pin user pages in memory.  But sometimes, some special flags except
FOLL_GET, FOLL_WRITE and FOLL_FORCE are needed, for example in
following patch, KVM needs FOLL_HWPOISON.  To support these users,
__get_user_pages is exported directly.

There are some symbol name conflicts in infiniband driver, fixed them too.
Signed-off-by: NHuang Ying <ying.huang@intel.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Michel Lespinasse <walken@google.com>
CC: Roland Dreier <roland@kernel.org>
CC: Ralph Campbell <infinipath@qlogic.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Michael Leun reported that running parallel opens on a fuse filesystem
can trigger a "kernel BUG at mm/truncate.c:475"

Gurudas Pai reported the same bug on NFS.

The reason is, unmap_mapping_range() is not prepared for more than
one concurrent invocation per inode.  For example:

  thread1: going through a big range, stops in the middle of a vma and
     stores the restart address in vm_truncate_count.

  thread2: comes in with a small (e.g. single page) unmap request on
     the same vma, somewhere before restart_address, finds that the
     vma was already unmapped up to the restart address and happily
     returns without doing anything.

Another scenario would be two big unmap requests, both having to
restart the unmapping and each one setting vm_truncate_count to its
own value.  This could go on forever without any of them being able to
finish.

Truncate and hole punching already serialize with i_mutex.  Other
callers of unmap_mapping_range() do not, and it's difficult to get
i_mutex protection for all callers.  In particular ->d_revalidate(),
which calls invalidate_inode_pages2_range() in fuse, may be called
with or without i_mutex.

This patch adds a new mutex to 'struct address_space' to prevent
running multiple concurrent unmap_mapping_range() on the same mapping.

[ We'll hopefully get rid of all this with the upcoming mm
  preemptibility series by Peter Zijlstra, the "mm: Remove i_mmap_mutex
  lockbreak" patch in particular.  But that is for 2.6.39 ]
Signed-off-by: NMiklos Szeredi <mszeredi@suse.cz>
Reported-by: NMichael Leun <lkml20101129@newton.leun.net>
Reported-by: NGurudas Pai <gurudas.pai@oracle.com>
Tested-by: NGurudas Pai <gurudas.pai@oracle.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

do_file_page and do_no_page don't exist anymore, but some comments
still refers them. The patch fixes them by replacing them with
existing ones.
Signed-off-by: NRyota Ozaki <ozaki.ryota@gmail.com>
Acked-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: NMinchan Kim <minchan.kim@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

If the page is going to be written to, __do_page needs to break COW.

However, the old page (before breaking COW) was never mapped mapped into
the current pte (__do_fault is only called when the pte is not present),
so vmscan can't have marked the old page as PageMlocked due to being
mapped in __do_fault's VMA.  Therefore, __do_fault() does not need to
worry about clearing PageMlocked() on the old page.
Signed-off-by: NMichel Lespinasse <walken@google.com>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

vmscan can lazily find pages that are mapped within VM_LOCKED vmas, and
set the PageMlocked bit on these pages, transfering them onto the
unevictable list.  When do_wp_page() breaks COW within a VM_LOCKED vma,
it may need to clear PageMlocked on the old page and set it on the new
page instead.

This change fixes an issue where do_wp_page() was clearing PageMlocked
on the old page while the pte was still pointing to it (as well as
rmap).  Therefore, we were not protected against vmscan immediately
transfering the old page back onto the unevictable list.  This could
cause pages to get stranded there forever.

I propose to move the corresponding code to the end of do_wp_page(),
after the pte (and rmap) have been pointed to the new page.
Additionally, we can use munlock_vma_page() instead of
clear_page_mlock(), so that the old page stays mlocked if there are
still other VM_LOCKED vmas mapping it.
Signed-off-by: NMichel Lespinasse <walken@google.com>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add debug checks for invariants that if broken could lead to mapcount vs
page_mapcount debug checks to trigger later in split_huge_page.
Signed-off-by: NAndrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

No pmd_trans_huge should ever materialize in migration ptes areas, because
we split the hugepage before migration ptes are instantiated.
Signed-off-by: NAndrea Arcangeli <aarcange@redhat.com>
Acked-by: NRik van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

pte_trans_huge must not leak in certain vmas like the mmio special pfn or
filebacked mappings.
Signed-off-by: NAndrea Arcangeli <aarcange@redhat.com>
Acked-by: NRik van Riel <riel@redhat.com>
Acked-by: NMel Gorman <mel@csn.ul.ie>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If you configure THP in addition to HUGETLB_PAGE on x86_32 without PAE,
the p?d-folding works out that munlock_vma_pages_range() can crash to
follow_page()'s pud_huge() BUG_ON(flags & FOLL_GET): it needs the same
VM_HUGETLB check already there on the pmd_huge() line.  Conveniently,
openSUSE provides a "blogd" which tests this out at startup!
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Lately I've been working to make KVM use hugepages transparently without
the usual restrictions of hugetlbfs.  Some of the restrictions I'd like to
see removed:

1) hugepages have to be swappable or the guest physical memory remains
   locked in RAM and can't be paged out to swap

2) if a hugepage allocation fails, regular pages should be allocated
   instead and mixed in the same vma without any failure and without
   userland noticing

3) if some task quits and more hugepages become available in the
   buddy, guest physical memory backed by regular pages should be
   relocated on hugepages automatically in regions under
   madvise(MADV_HUGEPAGE) (ideally event driven by waking up the
   kernel deamon if the order=HPAGE_PMD_SHIFT-PAGE_SHIFT list becomes
   not null)

4) avoidance of reservation and maximization of use of hugepages whenever
   possible. Reservation (needed to avoid runtime fatal faliures) may be ok for
   1 machine with 1 database with 1 database cache with 1 database cache size
   known at boot time. It's definitely not feasible with a virtualization
   hypervisor usage like RHEV-H that runs an unknown number of virtual machines
   with an unknown size of each virtual machine with an unknown amount of
   pagecache that could be potentially useful in the host for guest not using
   O_DIRECT (aka cache=off).

hugepages in the virtualization hypervisor (and also in the guest!) are
much more important than in a regular host not using virtualization,
becasue with NPT/EPT they decrease the tlb-miss cacheline accesses from 24
to 19 in case only the hypervisor uses transparent hugepages, and they
decrease the tlb-miss cacheline accesses from 19 to 15 in case both the
linux hypervisor and the linux guest both uses this patch (though the
guest will limit the addition speedup to anonymous regions only for
now...).  Even more important is that the tlb miss handler is much slower
on a NPT/EPT guest than for a regular shadow paging or no-virtualization
scenario.  So maximizing the amount of virtual memory cached by the TLB
pays off significantly more with NPT/EPT than without (even if there would
be no significant speedup in the tlb-miss runtime).

The first (and more tedious) part of this work requires allowing the VM to
handle anonymous hugepages mixed with regular pages transparently on
regular anonymous vmas.  This is what this patch tries to achieve in the
least intrusive possible way.  We want hugepages and hugetlb to be used in
a way so that all applications can benefit without changes (as usual we
leverage the KVM virtualization design: by improving the Linux VM at
large, KVM gets the performance boost too).

The most important design choice is: always fallback to 4k allocation if
the hugepage allocation fails!  This is the _very_ opposite of some large
pagecache patches that failed with -EIO back then if a 64k (or similar)
allocation failed...

Second important decision (to reduce the impact of the feature on the
existing pagetable handling code) is that at any time we can split an
hugepage into 512 regular pages and it has to be done with an operation
that can't fail.  This way the reliability of the swapping isn't decreased
(no need to allocate memory when we are short on memory to swap) and it's
trivial to plug a split_huge_page* one-liner where needed without
polluting the VM.  Over time we can teach mprotect, mremap and friends to
handle pmd_trans_huge natively without calling split_huge_page*.  The fact
it can't fail isn't just for swap: if split_huge_page would return -ENOMEM
(instead of the current void) we'd need to rollback the mprotect from the
middle of it (ideally including undoing the split_vma) which would be a
big change and in the very wrong direction (it'd likely be simpler not to
call split_huge_page at all and to teach mprotect and friends to handle
hugepages instead of rolling them back from the middle).  In short the
very value of split_huge_page is that it can't fail.

The collapsing and madvise(MADV_HUGEPAGE) part will remain separated and
incremental and it'll just be an "harmless" addition later if this initial
part is agreed upon.  It also should be noted that locking-wise replacing
regular pages with hugepages is going to be very easy if compared to what
I'm doing below in split_huge_page, as it will only happen when
page_count(page) matches page_mapcount(page) if we can take the PG_lock
and mmap_sem in write mode.  collapse_huge_page will be a "best effort"
that (unlike split_huge_page) can fail at the minimal sign of trouble and
we can try again later.  collapse_huge_page will be similar to how KSM
works and the madvise(MADV_HUGEPAGE) will work similar to
madvise(MADV_MERGEABLE).

The default I like is that transparent hugepages are used at page fault
time.  This can be changed with
/sys/kernel/mm/transparent_hugepage/enabled.  The control knob can be set
to three values "always", "madvise", "never" which mean respectively that
hugepages are always used, or only inside madvise(MADV_HUGEPAGE) regions,
or never used.  /sys/kernel/mm/transparent_hugepage/defrag instead
controls if the hugepage allocation should defrag memory aggressively
"always", only inside "madvise" regions, or "never".

The pmd_trans_splitting/pmd_trans_huge locking is very solid.  The
put_page (from get_user_page users that can't use mmu notifier like
O_DIRECT) that runs against a __split_huge_page_refcount instead was a
pain to serialize in a way that would result always in a coherent page
count for both tail and head.  I think my locking solution with a
compound_lock taken only after the page_first is valid and is still a
PageHead should be safe but it surely needs review from SMP race point of
view.  In short there is no current existing way to serialize the O_DIRECT
final put_page against split_huge_page_refcount so I had to invent a new
one (O_DIRECT loses knowledge on the mapping status by the time gup_fast
returns so...).  And I didn't want to impact all gup/gup_fast users for
now, maybe if we change the gup interface substantially we can avoid this
locking, I admit I didn't think too much about it because changing the gup
unpinning interface would be invasive.

If we ignored O_DIRECT we could stick to the existing compound refcounting
code, by simply adding a get_user_pages_fast_flags(foll_flags) where KVM
(and any other mmu notifier user) would call it without FOLL_GET (and if
FOLL_GET isn't set we'd just BUG_ON if nobody registered itself in the
current task mmu notifier list yet).  But O_DIRECT is fundamental for
decent performance of virtualized I/O on fast storage so we can't avoid it
to solve the race of put_page against split_huge_page_refcount to achieve
a complete hugepage feature for KVM.

Swap and oom works fine (well just like with regular pages ;).  MMU
notifier is handled transparently too, with the exception of the young bit
on the pmd, that didn't have a range check but I think KVM will be fine
because the whole point of hugepages is that EPT/NPT will also use a huge
pmd when they notice gup returns pages with PageCompound set, so they
won't care of a range and there's just the pmd young bit to check in that
case.

NOTE: in some cases if the L2 cache is small, this may slowdown and waste
memory during COWs because 4M of memory are accessed in a single fault
instead of 8k (the payoff is that after COW the program can run faster).
So we might want to switch the copy_huge_page (and clear_huge_page too) to
not temporal stores.  I also extensively researched ways to avoid this
cache trashing with a full prefault logic that would cow in 8k/16k/32k/64k
up to 1M (I can send those patches that fully implemented prefault) but I
concluded they're not worth it and they add an huge additional complexity
and they remove all tlb benefits until the full hugepage has been faulted
in, to save a little bit of memory and some cache during app startup, but
they still don't improve substantially the cache-trashing during startup
if the prefault happens in >4k chunks.  One reason is that those 4k pte
entries copied are still mapped on a perfectly cache-colored hugepage, so
the trashing is the worst one can generate in those copies (cow of 4k page
copies aren't so well colored so they trashes less, but again this results
in software running faster after the page fault).  Those prefault patches
allowed things like a pte where post-cow pages were local 4k regular anon
pages and the not-yet-cowed pte entries were pointing in the middle of
some hugepage mapped read-only.  If it doesn't payoff substantially with
todays hardware it will payoff even less in the future with larger l2
caches, and the prefault logic would blot the VM a lot.  If one is
emebdded transparent_hugepage can be disabled during boot with sysfs or
with the boot commandline parameter transparent_hugepage=0 (or
transparent_hugepage=2 to restrict hugepages inside madvise regions) that
will ensure not a single hugepage is allocated at boot time.  It is simple
enough to just disable transparent hugepage globally and let transparent
hugepages be allocated selectively by applications in the MADV_HUGEPAGE
region (both at page fault time, and if enabled with the
collapse_huge_page too through the kernel daemon).

This patch supports only hugepages mapped in the pmd, archs that have
smaller hugepages will not fit in this patch alone.  Also some archs like
power have certain tlb limits that prevents mixing different page size in
the same regions so they will not fit in this framework that requires
"graceful fallback" to basic PAGE_SIZE in case of physical memory
fragmentation.  hugetlbfs remains a perfect fit for those because its
software limits happen to match the hardware limits.  hugetlbfs also
remains a perfect fit for hugepage sizes like 1GByte that cannot be hoped
to be found not fragmented after a certain system uptime and that would be
very expensive to defragment with relocation, so requiring reservation.
hugetlbfs is the "reservation way", the point of transparent hugepages is
not to have any reservation at all and maximizing the use of cache and
hugepages at all times automatically.

Some performance result:

vmx andrea # LD_PRELOAD=/usr/lib64/libhugetlbfs.so HUGETLB_MORECORE=yes HUGETLB_PATH=/mnt/huge/ ./largep
ages3
memset page fault 1566023
memset tlb miss 453854
memset second tlb miss 453321
random access tlb miss 41635
random access second tlb miss 41658
vmx andrea # LD_PRELOAD=/usr/lib64/libhugetlbfs.so HUGETLB_MORECORE=yes HUGETLB_PATH=/mnt/huge/ ./largepages3
memset page fault 1566471
memset tlb miss 453375
memset second tlb miss 453320
random access tlb miss 41636
random access second tlb miss 41637
vmx andrea # ./largepages3
memset page fault 1566642
memset tlb miss 453417
memset second tlb miss 453313
random access tlb miss 41630
random access second tlb miss 41647
vmx andrea # ./largepages3
memset page fault 1566872
memset tlb miss 453418
memset second tlb miss 453315
random access tlb miss 41618
random access second tlb miss 41659
vmx andrea # echo 0 > /proc/sys/vm/transparent_hugepage
vmx andrea # ./largepages3
memset page fault 2182476
memset tlb miss 460305
memset second tlb miss 460179
random access tlb miss 44483
random access second tlb miss 44186
vmx andrea # ./largepages3
memset page fault 2182791
memset tlb miss 460742
memset second tlb miss 459962
random access tlb miss 43981
random access second tlb miss 43988

============
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/time.h>

#define SIZE (3UL*1024*1024*1024)

int main()
{
	char *p = malloc(SIZE), *p2;
	struct timeval before, after;

	gettimeofday(&before, NULL);
	memset(p, 0, SIZE);
	gettimeofday(&after, NULL);
	printf("memset page fault %Lu\n",
	       (after.tv_sec-before.tv_sec)*1000000UL +
	       after.tv_usec-before.tv_usec);

	gettimeofday(&before, NULL);
	memset(p, 0, SIZE);
	gettimeofday(&after, NULL);
	printf("memset tlb miss %Lu\n",
	       (after.tv_sec-before.tv_sec)*1000000UL +
	       after.tv_usec-before.tv_usec);

	gettimeofday(&before, NULL);
	memset(p, 0, SIZE);
	gettimeofday(&after, NULL);
	printf("memset second tlb miss %Lu\n",
	       (after.tv_sec-before.tv_sec)*1000000UL +
	       after.tv_usec-before.tv_usec);

	gettimeofday(&before, NULL);
	for (p2 = p; p2 < p+SIZE; p2 += 4096)
		*p2 = 0;
	gettimeofday(&after, NULL);
	printf("random access tlb miss %Lu\n",
	       (after.tv_sec-before.tv_sec)*1000000UL +
	       after.tv_usec-before.tv_usec);

	gettimeofday(&before, NULL);
	for (p2 = p; p2 < p+SIZE; p2 += 4096)
		*p2 = 0;
	gettimeofday(&after, NULL);
	printf("random access second tlb miss %Lu\n",
	       (after.tv_sec-before.tv_sec)*1000000UL +
	       after.tv_usec-before.tv_usec);

	return 0;
}
============
Signed-off-by: NAndrea Arcangeli <aarcange@redhat.com>
Acked-by: NRik van Riel <riel@redhat.com>
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>