linux-toradex.git/mm/mremap.c, branch v3.10.84

mm, thp: close race between mremap() and split_huge_page()

2014-06-07T20:25:31+00:00

commit dd18dbc2d42af75fffa60c77e0f02220bc329829 upstream.

It's critical for split_huge_page() (and migration) to catch and freeze
all PMDs on rmap walk.  It gets tricky if there's concurrent fork() or
mremap() since usually we copy/move page table entries on dup_mm() or
move_page_tables() without rmap lock taken.  To get it work we rely on
rmap walk order to not miss any entry.  We expect to see destination VMA
after source one to work correctly.

But after switching rmap implementation to interval tree it's not always
possible to preserve expected walk order.

It works fine for dup_mm() since new VMA has the same vma_start_pgoff()
/ vma_last_pgoff() and explicitly insert dst VMA after src one with
vma_interval_tree_insert_after().

But on move_vma() destination VMA can be merged into adjacent one and as
result shifted left in interval tree.  Fortunately, we can detect the
situation and prevent race with rmap walk by moving page table entries
under rmap lock.  See commit 38a76013ad80.

Problem is that we miss the lock when we move transhuge PMD.  Most
likely this bug caused the crash[1].

[1] http://thread.gmane.org/gmane.linux.kernel.mm/96473

Fixes: 108d6642ad81 ("mm anon rmap: remove anon_vma_moveto_tail")

Signed-off-by: Kirill A. Shutemov 
Reviewed-by: Andrea Arcangeli 
Cc: Rik van Riel 
Acked-by: Michel Lespinasse 
Cc: Dave Jones 
Cc: David Miller 
Acked-by: Johannes Weiner 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

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

mm: use mm_populate() for mremap() of VM_LOCKED vmas

2013-02-24T01:50:11+00:00

Signed-off-by: Michel Lespinasse 
Acked-by: Rik van Riel 
Tested-by: Andy Lutomirski 
Cc: Greg Ungerer 
Cc: David Howells 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

sched: Move sched.h sysctl bits into separate header

2013-02-07T19:50:54+00:00

Move the sysctl-related bits from include/linux/sched.h into
a new file: include/linux/sched/sysctl.h. Then update source
files requiring access to those bits by including the new
header file.

Signed-off-by: Clark Williams 
Cc: Peter Zijlstra 
Cc: Steven Rostedt 
Link: http://lkml.kernel.org/r/20130207094659.06dced96@riff.lan
Signed-off-by: Ingo Molnar

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
  ...

thp: change split_huge_page_pmd() interface

2012-12-13T01:38:31+00:00

Pass vma instead of mm and add address parameter.

In most cases we already have vma on the stack. We provides
split_huge_page_pmd_mm() for few cases when we have mm, but not vma.

This change is preparation to huge zero pmd splitting implementation.

Signed-off-by: Kirill A. Shutemov 
Cc: Andrea Arcangeli 
Cc: Andi Kleen 
Cc: "H. Peter Anvin" 
Cc: Mel Gorman 
Cc: David Rientjes 
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: 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

mm: avoid taking rmap locks in move_ptes()

2012-10-09T07:22:42+00:00

During mremap(), the destination VMA is generally placed after the
original vma in rmap traversal order: in move_vma(), we always have
new_pgoff >= vma->vm_pgoff, and as a result new_vma->vm_pgoff >=
vma->vm_pgoff unless vma_merge() merged the new vma with an adjacent one.

When the destination VMA is placed after the original in rmap traversal
order, we can avoid taking the rmap locks in move_ptes().

Essentially, this reintroduces the optimization that had been disabled in
"mm anon rmap: remove anon_vma_moveto_tail".  The difference is that we
don't try to impose the rmap traversal order; instead we just rely on
things being in the desired order in the common case and fall back to
taking locks in the uncommon case.  Also we skip the i_mmap_mutex in
addition to the anon_vma lock: in both cases, the vmas are traversed in
increasing vm_pgoff order with ties resolved in tree insertion order.

Signed-off-by: Michel Lespinasse 
Cc: Andrea Arcangeli 
Cc: Rik van Riel 
Cc: Peter Zijlstra 
Cc: Daniel Santos 
Cc: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm anon rmap: remove anon_vma_moveto_tail

2012-10-09T07:22:41+00:00

mremap() had a clever optimization where move_ptes() did not take the
anon_vma lock to avoid a race with anon rmap users such as page migration.
 Instead, the avc's were ordered in such a way that the origin vma was
always visited by rmap before the destination.  This ordering and the use
of page table locks rmap usage safe.  However, we want to replace the use
of linked lists in anon rmap with an interval tree, and this will make it
harder to impose such ordering as the interval tree will always be sorted
by the avc->vma->vm_pgoff value.  For now, let's replace the
anon_vma_moveto_tail() ordering function with proper anon_vma locking in
move_ptes().  Once we have the anon interval tree in place, we will
re-introduce an optimization to avoid taking these locks in the most
common cases.

Signed-off-by: Michel Lespinasse 
Cc: Andrea Arcangeli 
Cc: Rik van Riel 
Cc: Peter Zijlstra 
Cc: Daniel Santos 
Cc: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds