linux.git/mm/rmap.c, branch v3.9

mm/rmap: rename anon_vma_unlock() => anon_vma_unlock_write()

2013-02-24T01:50:17+00:00

The comment in commit 4fc3f1d66b1e ("mm/rmap, migration: Make
rmap_walk_anon() and try_to_unmap_anon() more scalable") says:

| Rename anon_vma_[un]lock() => anon_vma_[un]lock_write(),
| to make it clearer that it's an exclusive write-lock in
| that case - suggested by Rik van Riel.

But that commit renames only anon_vma_lock()

Signed-off-by: Konstantin Khlebnikov 
Cc: Ingo Molnar 
Reviewed-by: Rik van Riel 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

s390/mm: implement software dirty bits

2013-02-14T14:55:23+00:00

The s390 architecture is unique in respect to dirty page detection,
it uses the change bit in the per-page storage key to track page
modifications. All other architectures track dirty bits by means
of page table entries. This property of s390 has caused numerous
problems in the past, e.g. see git commit ef5d437f71afdf4a
"mm: fix XFS oops due to dirty pages without buffers on s390".

To avoid future issues in regard to per-page dirty bits convert
s390 to a fault based software dirty bit detection mechanism. All
user page table entries which are marked as clean will be hardware
read-only, even if the pte is supposed to be writable. A write by
the user process will trigger a protection fault which will cause
the user pte to be marked as dirty and the hardware read-only bit
is removed.

With this change the dirty bit in the storage key is irrelevant
for Linux as a host, but the storage key is still required for
KVM guests. The effect is that page_test_and_clear_dirty and the
related code can be removed. The referenced bit in the storage
key is still used by the page_test_and_clear_young primitive to
provide page age information.

For page cache pages of mappings with mapping_cap_account_dirty
there will not be any change in behavior as the dirty bit tracking
already uses read-only ptes to control the amount of dirty pages.
Only for swap cache pages and pages of mappings without
mapping_cap_account_dirty there can be additional protection faults.
To avoid an excessive number of additional faults the mk_pte
primitive checks for PageDirty if the pgprot value allows for writes
and pre-dirties the pte. That avoids all additional faults for
tmpfs and shmem pages until these pages are added to the swap cache.

Signed-off-by: Martin Schwidefsky

Merge tag 'balancenuma-v11' of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma

2012-12-16T23:18:08+00:00

Pull Automatic NUMA Balancing bare-bones from Mel Gorman:
 "There are three implementations for NUMA balancing, this tree
  (balancenuma), numacore which has been developed in tip/master and
  autonuma which is in aa.git.

  In almost all respects balancenuma is the dumbest of the three because
  its main impact is on the VM side with no attempt to be smart about
  scheduling.  In the interest of getting the ball rolling, it would be
  desirable to see this much merged for 3.8 with the view to building
  scheduler smarts on top and adapting the VM where required for 3.9.

  The most recent set of comparisons available from different people are

    mel:    https://lkml.org/lkml/2012/12/9/108
    mingo:  https://lkml.org/lkml/2012/12/7/331
    tglx:   https://lkml.org/lkml/2012/12/10/437
    srikar: https://lkml.org/lkml/2012/12/10/397

  The results are a mixed bag.  In my own tests, balancenuma does
  reasonably well.  It's dumb as rocks and does not regress against
  mainline.  On the other hand, Ingo's tests shows that balancenuma is
  incapable of converging for this workloads driven by perf which is bad
  but is potentially explained by the lack of scheduler smarts.  Thomas'
  results show balancenuma improves on mainline but falls far short of
  numacore or autonuma.  Srikar's results indicate we all suffer on a
  large machine with imbalanced node sizes.

  My own testing showed that recent numacore results have improved
  dramatically, particularly in the last week but not universally.
  We've butted heads heavily on system CPU usage and high levels of
  migration even when it shows that overall performance is better.
  There are also cases where it regresses.  Of interest is that for
  specjbb in some configurations it will regress for lower numbers of
  warehouses and show gains for higher numbers which is not reported by
  the tool by default and sometimes missed in treports.  Recently I
  reported for numacore that the JVM was crashing with
  NullPointerExceptions but currently it's unclear what the source of
  this problem is.  Initially I thought it was in how numacore batch
  handles PTEs but I'm no longer think this is the case.  It's possible
  numacore is just able to trigger it due to higher rates of migration.

  These reports were quite late in the cycle so I/we would like to start
  with this tree as it contains much of the code we can agree on and has
  not changed significantly over the last 2-3 weeks."

* tag 'balancenuma-v11' of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux-balancenuma: (50 commits)
  mm/rmap, migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable
  mm/rmap: Convert the struct anon_vma::mutex to an rwsem
  mm: migrate: Account a transhuge page properly when rate limiting
  mm: numa: Account for failed allocations and isolations as migration failures
  mm: numa: Add THP migration for the NUMA working set scanning fault case build fix
  mm: numa: Add THP migration for the NUMA working set scanning fault case.
  mm: sched: numa: Delay PTE scanning until a task is scheduled on a new node
  mm: sched: numa: Control enabling and disabling of NUMA balancing if !SCHED_DEBUG
  mm: sched: numa: Control enabling and disabling of NUMA balancing
  mm: sched: Adapt the scanning rate if a NUMA hinting fault does not migrate
  mm: numa: Use a two-stage filter to restrict pages being migrated for unlikely task<->node relationships
  mm: numa: migrate: Set last_nid on newly allocated page
  mm: numa: split_huge_page: Transfer last_nid on tail page
  mm: numa: Introduce last_nid to the page frame
  sched: numa: Slowly increase the scanning period as NUMA faults are handled
  mm: numa: Rate limit setting of pte_numa if node is saturated
  mm: numa: Rate limit the amount of memory that is migrated between nodes
  mm: numa: Structures for Migrate On Fault per NUMA migration rate limiting
  mm: numa: Migrate pages handled during a pmd_numa hinting fault
  mm: numa: Migrate on reference policy
  ...

hwpoison, hugetlbfs: fix RSS-counter warning

2012-12-13T01:38:35+00:00

Memory error handling on hugepages can break a RSS counter, which emits a
message like "Bad rss-counter state mm:ffff88040abecac0 idx:1 val:-1".
This is because PageAnon returns true for hugepage (this behavior is
necessary for reverse mapping to work on hugetlbfs).

[akpm@linux-foundation.org: clean up code layout]
Signed-off-by: Naoya Horiguchi 
Cc: Andi Kleen 
Cc: Tony Luck 
Cc: Wu Fengguang 
Cc: Aneesh Kumar K.V 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: add comment on storage key dirty bit semantics

2012-12-12T01:22:23+00:00

Add comments that dirty bit in storage key gets set whenever page content
is changed.  Hopefully if someone will use this function, he'll have a
look at one of the two places where we comment on this.

Signed-off-by: Jan Kara 
Cc: Martin Schwidefsky 
Cc: Mel Gorman 
Cc: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: introduce mm_find_pmd()

2012-12-12T01:22:22+00:00

Several place need to find the pmd by(mm_struct, address), so introduce a
function to simplify it.

[akpm@linux-foundation.org: fix warning]
Signed-off-by: Bob Liu 
Cc: Andrea Arcangeli 
Cc: Michal Hocko 
Cc: Minchan Kim 
Cc: Ni zhan Chen 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm/rmap, migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-11T14:43:00+00:00

rmap_walk_anon() and try_to_unmap_anon() appears to be too
careful about locking the anon vma: while it needs protection
against anon vma list modifications, it does not need exclusive
access to the list itself.

Transforming this exclusive lock to a read-locked rwsem removes
a global lock from the hot path of page-migration intense
threaded workloads which can cause pathological performance like
this:

    96.43%        process 0  [kernel.kallsyms]  [k] perf_trace_sched_switch
                  |
                  --- perf_trace_sched_switch
                      __schedule
                      schedule
                      schedule_preempt_disabled
                      __mutex_lock_common.isra.6
                      __mutex_lock_slowpath
                      mutex_lock
                     |
                     |--50.61%-- rmap_walk
                     |          move_to_new_page
                     |          migrate_pages
                     |          migrate_misplaced_page
                     |          __do_numa_page.isra.69
                     |          handle_pte_fault
                     |          handle_mm_fault
                     |          __do_page_fault
                     |          do_page_fault
                     |          page_fault
                     |          __memset_sse2
                     |          |
                     |           --100.00%-- worker_thread
                     |                     |
                     |                      --100.00%-- start_thread
                     |
                      --49.39%-- page_lock_anon_vma
                                try_to_unmap_anon
                                try_to_unmap
                                migrate_pages
                                migrate_misplaced_page
                                __do_numa_page.isra.69
                                handle_pte_fault
                                handle_mm_fault
                                __do_page_fault
                                do_page_fault
                                page_fault
                                __memset_sse2
                                |
                                 --100.00%-- worker_thread
                                           start_thread

With this change applied the profile is now nicely flat
and there's no anon-vma related scheduling/blocking.

Rename anon_vma_[un]lock() => anon_vma_[un]lock_write(),
to make it clearer that it's an exclusive write-lock in
that case - suggested by Rik van Riel.

Suggested-by: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Paul Turner 
Cc: Lee Schermerhorn 
Cc: Christoph Lameter 
Cc: Rik van Riel 
Cc: Mel Gorman 
Cc: Andrea Arcangeli 
Cc: Johannes Weiner 
Cc: Hugh Dickins 
Signed-off-by: Ingo Molnar 
Signed-off-by: Mel Gorman

mm/rmap: Convert the struct anon_vma::mutex to an rwsem

2012-12-11T14:43:00+00:00

Convert the struct anon_vma::mutex to an rwsem, which will help
in solving a page-migration scalability problem. (Addressed in
a separate patch.)

The conversion is simple and straightforward: in every case
where we mutex_lock()ed we'll now down_write().

Suggested-by: Linus Torvalds 
Reviewed-by: Rik van Riel 
Cc: Peter Zijlstra 
Cc: Paul Turner 
Cc: Lee Schermerhorn 
Cc: Christoph Lameter 
Cc: Mel Gorman 
Cc: Andrea Arcangeli 
Cc: Johannes Weiner 
Cc: Hugh Dickins 
Signed-off-by: Ingo Molnar 
Signed-off-by: Mel Gorman

mm: fix XFS oops due to dirty pages without buffers on s390

2012-10-25T21:37:52+00:00

On s390 any write to a page (even from kernel itself) sets architecture
specific page dirty bit.  Thus when a page is written to via buffered
write, HW dirty bit gets set and when we later map and unmap the page,
page_remove_rmap() finds the dirty bit and calls set_page_dirty().

Dirtying of a page which shouldn't be dirty can cause all sorts of
problems to filesystems.  The bug we observed in practice is that
buffers from the page get freed, so when the page gets later marked as
dirty and writeback writes it, XFS crashes due to an assertion
BUG_ON(!PagePrivate(page)) in page_buffers() called from
xfs_count_page_state().

Similar problem can also happen when zero_user_segment() call from
xfs_vm_writepage() (or block_write_full_page() for that matter) set the
hardware dirty bit during writeback, later buffers get freed, and then
page unmapped.

Fix the issue by ignoring s390 HW dirty bit for page cache pages of
mappings with mapping_cap_account_dirty().  This is safe because for
such mappings when a page gets marked as writeable in PTE it is also
marked dirty in do_wp_page() or do_page_fault().  When the dirty bit is
cleared by clear_page_dirty_for_io(), the page gets writeprotected in
page_mkclean().  So pagecache page is writeable if and only if it is
dirty.

Thanks to Hugh Dickins for pointing out mapping has to have
mapping_cap_account_dirty() for things to work and proposing a cleaned
up variant of the patch.

The patch has survived about two hours of running fsx-linux on tmpfs
while heavily swapping and several days of running on out build machines
where the original problem was triggered.

Signed-off-by: Jan Kara 
Cc: Martin Schwidefsky 
Cc: Mel Gorman 
Cc: Hugh Dickins 
Cc: Heiko Carstens 
Cc: 		[3.0+]
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: move all mmu notifier invocations to be done outside the PT lock

2012-10-09T07:22:58+00:00

In order to allow sleeping during mmu notifier calls, we need to avoid
invoking them under the page table spinlock.  This patch solves the
problem by calling invalidate_page notification after releasing the lock
(but before freeing the page itself), or by wrapping the page invalidation
with calls to invalidate_range_begin and invalidate_range_end.

To prevent accidental changes to the invalidate_range_end arguments after
the call to invalidate_range_begin, the patch introduces a convention of
saving the arguments in consistently named locals:

	unsigned long mmun_start;	/* For mmu_notifiers */
	unsigned long mmun_end;	/* For mmu_notifiers */

	...

	mmun_start = ...
	mmun_end = ...
	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);

	...

	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);

The patch changes code to use this convention for all calls to
mmu_notifier_invalidate_range_start/end, except those where the calls are
close enough so that anyone who glances at the code can see the values
aren't changing.

This patchset is a preliminary step towards on-demand paging design to be
added to the RDMA stack.

Why do we want on-demand paging for Infiniband?

  Applications register memory with an RDMA adapter using system calls,
  and subsequently post IO operations that refer to the corresponding
  virtual addresses directly to HW.  Until now, this was achieved by
  pinning the memory during the registration calls.  The goal of on demand
  paging is to avoid pinning the pages of registered memory regions (MRs).
   This will allow users the same flexibility they get when swapping any
  other part of their processes address spaces.  Instead of requiring the
  entire MR to fit in physical memory, we can allow the MR to be larger,
  and only fit the current working set in physical memory.

Why should anyone care?  What problems are users currently experiencing?

  This can make programming with RDMA much simpler.  Today, developers
  that are working with more data than their RAM can hold need either to
  deregister and reregister memory regions throughout their process's
  life, or keep a single memory region and copy the data to it.  On demand
  paging will allow these developers to register a single MR at the
  beginning of their process's life, and let the operating system manage
  which pages needs to be fetched at a given time.  In the future, we
  might be able to provide a single memory access key for each process
  that would provide the entire process's address as one large memory
  region, and the developers wouldn't need to register memory regions at
  all.

Is there any prospect that any other subsystems will utilise these
infrastructural changes?  If so, which and how, etc?

  As for other subsystems, I understand that XPMEM wanted to sleep in
  MMU notifiers, as Christoph Lameter wrote at
  http://lkml.indiana.edu/hypermail/linux/kernel/0802.1/0460.html and
  perhaps Andrea knows about other use cases.

  Scheduling in mmu notifications is required since we need to sync the
  hardware with the secondary page tables change.  A TLB flush of an IO
  device is inherently slower than a CPU TLB flush, so our design works by
  sending the invalidation request to the device, and waiting for an
  interrupt before exiting the mmu notifier handler.

Avi said:

  kvm may be a buyer.  kvm::mmu_lock, which serializes guest page
  faults, also protects long operations such as destroying large ranges.
  It would be good to convert it into a spinlock, but as it is used inside
  mmu notifiers, this cannot be done.

  (there are alternatives, such as keeping the spinlock and using a
  generation counter to do the teardown in O(1), which is what the "may"
  is doing up there).

[akpm@linux-foundation.orgpossible speed tweak in hugetlb_cow(), cleanups]
Signed-off-by: Andrea Arcangeli 
Signed-off-by: Sagi Grimberg 
Signed-off-by: Haggai Eran 
Cc: Peter Zijlstra 
Cc: Xiao Guangrong 
Cc: Or Gerlitz 
Cc: Haggai Eran 
Cc: Shachar Raindel 
Cc: Liran Liss 
Cc: Christoph Lameter 
Cc: Avi Kivity 
Cc: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds