The purge_lock spinlock causes high latencies with non RT kernel.  This
has been reported multiple times on lkml [1] [2] and affects
applications like audio.

This patch replaces it with a mutex to allow preemption while holding
the lock.

Thanks to Joel Fernandes for the detailed report and analysis as well as
an earlier attempt at fixing this issue.

[1] http://lists.openwall.net/linux-kernel/2016/03/23/29
[2] https://lkml.org/lkml/2016/10/9/59

Link: http://lkml.kernel.org/r/1479474236-4139-10-git-send-email-hch@lst.deSigned-off-by: NChristoph Hellwig <hch@lst.de>
Tested-by: NJisheng Zhang <jszhang@marvell.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We will take a sleeping lock in later in this series, so this adds the
proper safeguards.

Link: http://lkml.kernel.org/r/1479474236-4139-9-git-send-email-hch@lst.deSigned-off-by: NChristoph Hellwig <hch@lst.de>
Tested-by: NJisheng Zhang <jszhang@marvell.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

vfree() is going to use sleeping lock.  free_ldt_struct() may be called
with disabled preemption, therefore we must use vfree_atomic() here.

E.g. call trace:
	vfree()
	free_ldt_struct()
	destroy_context_ldt()
	__mmdrop()
	finish_task_switch()
	schedule_tail()
	ret_from_fork()

Link: http://lkml.kernel.org/r/1479474236-4139-7-git-send-email-hch@lst.deSigned-off-by: NAndrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Jisheng Zhang <jszhang@marvell.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

vfree() is going to use sleeping lock.  Thread stack freed in atomic
context, therefore we must use vfree_atomic() here.

Link: http://lkml.kernel.org/r/1479474236-4139-6-git-send-email-hch@lst.deSigned-off-by: NAndrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Jisheng Zhang <jszhang@marvell.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We are going to use sleeping lock for freeing vmap.  However some
vfree() users want to free memory from atomic (but not from interrupt)
context.  For this we add vfree_atomic() - deferred variation of vfree()
which can be used in any atomic context (except NMIs).

[akpm@linux-foundation.org: tweak comment grammar]
[aryabinin@virtuozzo.com: use raw_cpu_ptr() instead of this_cpu_ptr()]
  Link: http://lkml.kernel.org/r/1481553981-3856-1-git-send-email-aryabinin@virtuozzo.com
Link: http://lkml.kernel.org/r/1479474236-4139-5-git-send-email-hch@lst.deSigned-off-by: NAndrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Jisheng Zhang <jszhang@marvell.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Move the purge_lock synchronization to the callers, move the call to
purge_fragmented_blocks_allcpus at the beginning of the function to the
callers that need it, move the force_flush behavior to the caller that
needs it, and pass start and end by value instead of by reference.

No change in behavior.

Link: http://lkml.kernel.org/r/1479474236-4139-4-git-send-email-hch@lst.deSigned-off-by: NChristoph Hellwig <hch@lst.de>
Tested-by: NJisheng Zhang <jszhang@marvell.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Just inline it into the only caller.

Link: http://lkml.kernel.org/r/1479474236-4139-3-git-send-email-hch@lst.deSigned-off-by: NChristoph Hellwig <hch@lst.de>
Tested-by: NJisheng Zhang <jszhang@marvell.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Patch series "reduce latency in __purge_vmap_area_lazy", v2.

This patch (of 10):

Sort out the long lock hold times in __purge_vmap_area_lazy.  It is
based on a patch from Joel.

Inline free_unmap_vmap_area_noflush() it into the only caller.

Link: http://lkml.kernel.org/r/1479474236-4139-2-git-send-email-hch@lst.deSigned-off-by: NChristoph Hellwig <hch@lst.de>
Tested-by: NJisheng Zhang <jszhang@marvell.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Dias <joaodias@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Since commit 59dc76b0 ("mm: vmscan: reduce size of inactive file
list") the size of the active file list is no longer limited to half of
memory.  Increase the shadow node limit accordingly to avoid throwing
out shadow entries that might still result in eligible refaults.

The exact size of the active list now depends on the overall size of the
page cache, but converges toward taking up most of the space:

In mm/vmscan.c::inactive_list_is_low(),

 * total     target    max
 * memory    ratio     inactive
 * -------------------------------------
 *   10MB       1         5MB
 *  100MB       1        50MB
 *    1GB       3       250MB
 *   10GB      10       0.9GB
 *  100GB      31         3GB
 *    1TB     101        10GB
 *   10TB     320        32GB

It would be possible to apply the same precise ratios when determining
the limit for radix tree nodes containing shadow entries, but since it
is merely an approximation of the oldest refault distances in the wild
and the code also makes assumptions about the node population density,
keep it simple and always target the full cache size.

While at it, clarify the comment and the formula for memory footprint.

Link: http://lkml.kernel.org/r/20161117214701.29000-1-hannes@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Shadow entries in the page cache used to be accounted behind the radix
tree implementation's back in the upper bits of node->count, and the
radix tree code extending a single-entry tree with a shadow entry in
root->rnode would corrupt that counter.  As a result, we could not put
shadow entries at index 0 if the tree didn't have any other entries, and
that means no refault detection for any single-page file.

Now that the shadow entries are tracked natively in the radix tree's
exceptional counter, this is no longer necessary.  Extending and
shrinking the tree from and to single entries in root->rnode now does
the right thing when the entry is exceptional, remove that limitation.

Link: http://lkml.kernel.org/r/20161117193244.GF23430@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently, we track the shadow entries in the page cache in the upper
bits of the radix_tree_node->count, behind the back of the radix tree
implementation.  Because the radix tree code has no awareness of them,
we rely on random subtleties throughout the implementation (such as the
node->count != 1 check in the shrinking code, which is meant to exclude
multi-entry nodes but also happens to skip nodes with only one shadow
entry, as that's accounted in the upper bits).  This is error prone and
has, in fact, caused the bug fixed in d3798ae8 ("mm: filemap: don't
plant shadow entries without radix tree node").

To remove these subtleties, this patch moves shadow entry tracking from
the upper bits of node->count to the existing counter for exceptional
entries.  node->count goes back to being a simple counter of valid
entries in the tree node and can be shrunk to a single byte.

This vastly simplifies the page cache code.  All accounting happens
natively inside the radix tree implementation, and maintaining the LRU
linkage of shadow nodes is consolidated into a single function in the
workingset code that is called for leaf nodes affected by a change in
the page cache tree.

This also removes the last user of the __radix_delete_node() return
value.  Eliminate it.

Link: http://lkml.kernel.org/r/20161117193211.GE23430@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Support handing __radix_tree_replace() a callback that gets invoked for
all leaf nodes that change or get freed as a result of the slot
replacement, to assist users tracking nodes with node->private_list.

This prepares for putting page cache shadow entries into the radix tree
root again and drastically simplifying the shadow tracking.

Link: http://lkml.kernel.org/r/20161117193134.GD23430@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Suggested-by: NJan Kara <jack@suse.cz>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Page cache shadow entry handling will be a lot simpler when it can use a
single generic replacement function for pages, shadow entries, and
emptying slots.

Make __radix_tree_replace() properly account insertions and deletions in
node->count and garbage collect nodes as they become empty.  Then
re-implement radix_tree_delete() on top of it.

Link: http://lkml.kernel.org/r/20161117193058.GC23430@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The bug in khugepaged fixed earlier in this series shows that radix tree
slot replacement is fragile; and it will become more so when not only
NULL<->!NULL transitions need to be caught but transitions from and to
exceptional entries as well.  We need checks.

Re-implement radix_tree_replace_slot() on top of the sanity-checked
__radix_tree_replace().  This requires existing callers to also pass the
radix tree root, but it'll warn us when somebody replaces slots with
contents that need proper accounting (transitions between NULL entries,
real entries, exceptional entries) and where a replacement through the
slot pointer would corrupt the radix tree node counts.

Link: http://lkml.kernel.org/r/20161117193021.GB23430@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Suggested-by: NJan Kara <jack@suse.cz>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The way the page cache is sneaking shadow entries of evicted pages into
the radix tree past the node entry accounting and tracking them manually
in the upper bits of node->count is fraught with problems.

These shadow entries are marked in the tree as exceptional entries,
which are a native concept to the radix tree.  Maintain an explicit
counter of exceptional entries in the radix tree node.  Subsequent
patches will switch shadow entry tracking over to that counter.

DAX and shmem are the other users of exceptional entries.  Since slot
replacements that change the entry type from regular to exceptional must
now be accounted, introduce a __radix_tree_replace() function that does
replacement and accounting, and switch DAX and shmem over.

The increase in radix tree node size is temporary.  A followup patch
switches the shadow tracking to this new scheme and we'll no longer need
the upper bits in node->count and shrink that back to one byte.

Link: http://lkml.kernel.org/r/20161117192945.GA23430@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

When the shadow page shrinker tries to reclaim a radix tree node but
finds it in an unexpected state - it should contain no pages, and
non-zero shadow entries - there is no need to kill the executing task or
even the entire system.  Warn about the invalid state, then leave that
tree node be.  Simply don't put it back on the shadow LRU for future
reclaim and move on.

Link: http://lkml.kernel.org/r/20161117191138.22769-4-hannes@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The radix tree counts valid entries in each tree node.  Entries stored
in the tree cannot be removed by simpling storing NULL in the slot or
the internal counters will be off and the node never gets freed again.

When collapsing a shmem page fails, restore the holes that were filled
with radix_tree_insert() with a proper radix tree deletion.

Fixes: f3f0e1d2 ("khugepaged: add support of collapse for tmpfs/shmem pages")
Link: http://lkml.kernel.org/r/20161117191138.22769-3-hannes@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reported-by: NJan Kara <jack@suse.cz>
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Patch series "mm: workingset: radix tree subtleties & single-page file
refaults", v3.

This is another revision of the radix tree / workingset patches based on
feedback from Jan and Kirill.

This is a follow-up to d3798ae8 ("mm: filemap: don't plant shadow
entries without radix tree node").  That patch fixed an issue that was
caused mainly by the page cache sneaking special shadow page entries
into the radix tree and relying on subtleties in the radix tree code to
make that work.  The fix also had to stop tracking refaults for
single-page files because shadow pages stored as direct pointers in
radix_tree_root->rnode weren't properly handled during tree extension.

These patches make the radix tree code explicitely support and track
such special entries, to eliminate the subtleties and to restore the
thrash detection for single-page files.

This patch (of 9):

When a radix tree iteration drops the tree lock, another thread might
swoop in and free the node holding the current slot.  The iteration
needs to do another tree lookup from the current index to continue.

[kirill.shutemov@linux.intel.com: re-lookup for replacement]
Fixes: f3f0e1d2 ("khugepaged: add support of collapse for tmpfs/shmem pages")
Link: http://lkml.kernel.org/r/20161117191138.22769-2-hannes@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <mawilcox@linuxonhyperv.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Move the 4-byte `capabilities' field next to other 4-byte things.
Shrinks sizeof(backing_dev_info) by 8 bytes on x86_64.
Reviewed-by: NJens Axboe <axboe@kernel.dk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We ran into a funky issue, where someone doing 256K buffered reads saw
128K requests at the device level.  Turns out it is read-ahead capping
the request size, since we use 128K as the default setting.  This
doesn't make a lot of sense - if someone is issuing 256K reads, they
should see 256K reads, regardless of the read-ahead setting, if the
underlying device can support a 256K read in a single command.

This patch introduces a bdi hint, io_pages.  This is the soft max IO
size for the lower level, I've hooked it up to the bdev settings here.
Read-ahead is modified to issue the maximum of the user request size,
and the read-ahead max size, but capped to the max request size on the
device side.  The latter is done to avoid reading ahead too much, if the
application asks for a huge read.  With this patch, the kernel behaves
like the application expects.

Link: http://lkml.kernel.org/r/1479498073-8657-1-git-send-email-axboe@fb.comSigned-off-by: NJens Axboe <axboe@fb.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>

Compiling shmem.c with SHMEM and TRANSAPRENT_HUGE_PAGECACHE enabled
raises warnings on two unused functions when CONFIG_TMPFS and
CONFIG_SYSFS are both disabled:

  mm/shmem.c:390:20: warning: `shmem_format_huge' defined but not used [-Wunused-function]
   static const char *shmem_format_huge(int huge)
                      ^~~~~~~~~~~~~~~~~
  mm/shmem.c:373:12: warning: `shmem_parse_huge' defined but not used [-Wunused-function]
   static int shmem_parse_huge(const char *str)
               ^~~~~~~~~~~~~~~~

A conditional compilation on tmpfs or sysfs removes the warnings.

Link: http://lkml.kernel.org/r/20161118055749.11313-1-jeremy.lefaure@lse.epita.frSigned-off-by: NJérémy Lefaure <jeremy.lefaure@lse.epita.fr>
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>

b_more_io non-empty check is already preceded by an opposite check.

Link: http://lkml.kernel.org/r/1478591249-30641-1-git-send-email-tahsin@google.comSigned-off-by: NTahsin Erdogan <tahsin@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Unlike THP, hugetlb pages are represented by one entry in the
radix-tree.

[akpm@linux-foundation.org: tweak comment]
Link: http://lkml.kernel.org/r/20161110163640.126124-1-kirill.shutemov@linux.intel.comSigned-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The header uses types and definitions from the linux/init.h as well as
linux/types.h headers without explicitly including them.  This causes a
failure to compile if they are not implicitly pulled in by includers.

Link: http://lkml.kernel.org/r/20161115133235.13387-1-thierry.reding@gmail.comSigned-off-by: NThierry Reding <treding@nvidia.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Mark dax vmas as not migratable to exclude them from task_numa_work().
This is especially relevant for device-dax which wants to ensure
predictable access latency and not incur periodic faults.

[akpm@linux-foundation.org: add comment]
Link: http://lkml.kernel.org/r/147892450132.22062.16875659431109209179.stgit@dwillia2-desk3.amr.corp.intel.comSigned-off-by: NDan Williams <dan.j.williams@intel.com>
Reported-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Having code for the pkey_mprotect, pkey_alloc and pkey_free system calls
makes only sense if ARCH_HAS_PKEYS is selected.  If not selected these
system calls will always return -ENOSPC or -EINVAL.

To simplify things and have less code generate the pkey system call code
only if ARCH_HAS_PKEYS is selected.

For architectures which have already wired up the system calls, but do
not select ARCH_HAS_PKEYS this will result in less generated code and a
different return code: the three system calls will now always return
-ENOSYS, using the cond_syscall mechanism.

For architectures which have not wired up the system calls less
unreachable code will be generated.

Link: http://lkml.kernel.org/r/20161114111251.70084-1-heiko.carstens@de.ibm.comSigned-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: NDave Hansen <dave.hansen@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Summarize the "hotpluggable" property of dt memory nodes.

Link: http://lkml.kernel.org/r/1479160961-25840-6-git-send-email-arbab@linux.vnet.ibm.comSigned-off-by: NReza Arbab <arbab@linux.vnet.ibm.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Alistair Popple <apopple@au1.ibm.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Stewart Smith <stewart@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

When movable nodes are enabled, any node containing only hotpluggable
memory is made movable at boot time.

On x86, hotpluggable memory is discovered by parsing the ACPI SRAT,
making corresponding calls to memblock_mark_hotplug().

If we introduce a dt property to describe memory as hotpluggable,
configs supporting early fdt may then also do this marking and use
movable nodes.

Link: http://lkml.kernel.org/r/1479160961-25840-5-git-send-email-arbab@linux.vnet.ibm.comSigned-off-by: NReza Arbab <arbab@linux.vnet.ibm.com>
Tested-by: NBalbir Singh <bsingharora@gmail.com>
Acked-by: NBalbir Singh <bsingharora@gmail.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Alistair Popple <apopple@au1.ibm.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Stewart Smith <stewart@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

To support movable memory nodes (CONFIG_MOVABLE_NODE), at least one of
the following must be true:

1. This config has the capability to identify movable nodes at boot.
   Right now, only x86 can do this.

2. Our config supports memory hotplug, which means that a movable node
   can be created by hotplugging all of its memory into ZONE_MOVABLE.

Fix the Kconfig definition of CONFIG_MOVABLE_NODE, which currently
recognizes (1), but not (2).

Link: http://lkml.kernel.org/r/1479160961-25840-4-git-send-email-arbab@linux.vnet.ibm.comSigned-off-by: NReza Arbab <arbab@linux.vnet.ibm.com>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: NBalbir Singh <bsingharora@gmail.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Alistair Popple <apopple@au1.ibm.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Stewart Smith <stewart@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In commit c5320926 ("mem-hotplug: introduce movable_node boot
option"), the memblock allocation direction is changed to bottom-up and
then back to top-down like this:

1. memblock_set_bottom_up(true), called by cmdline_parse_movable_node().
2. memblock_set_bottom_up(false), called by x86's numa_init().

Even though (1) occurs in generic mm code, it is wrapped by #ifdef
CONFIG_MOVABLE_NODE, which depends on X86_64.

This means that when we extend CONFIG_MOVABLE_NODE to non-x86 arches,
things will be unbalanced.  (1) will happen for them, but (2) will not.

This toggle was added in the first place because x86 has a delay between
adding memblocks and marking them as hotpluggable.  Since other arches
do this marking either immediately or not at all, they do not require
the bottom-up toggle.

So, resolve things by moving (1) from cmdline_parse_movable_node() to
x86's setup_arch(), immediately after the movable_node parameter has
been parsed.

Link: http://lkml.kernel.org/r/1479160961-25840-3-git-send-email-arbab@linux.vnet.ibm.comSigned-off-by: NReza Arbab <arbab@linux.vnet.ibm.com>
Acked-by: NBalbir Singh <bsingharora@gmail.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Alistair Popple <apopple@au1.ibm.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Stewart Smith <stewart@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Patch series "enable movable nodes on non-x86 configs", v7.

This patchset allows more configs to make use of movable nodes.  When
CONFIG_MOVABLE_NODE is selected, there are two ways to introduce such
nodes into the system:

1. Discover movable nodes at boot. Currently this is only possible on
   x86, but we will enable configs supporting fdt to do the same.

2. Hotplug and online all of a node's memory using online_movable. This
   is already possible on any config supporting memory hotplug, not
   just x86, but the Kconfig doesn't say so. We will fix that.

We'll also remove some cruft on power which would prevent (2).

This patch (of 5):

Remove the check which prevents us from hotplugging into an empty node.

The original commit b226e462 ("[PATCH] powerpc: don't add memory to
empty node/zone"), states that this was intended to be a temporary measure.
It is a workaround for an oops which no longer occurs.

Link: http://lkml.kernel.org/r/1479160961-25840-2-git-send-email-arbab@linux.vnet.ibm.comSigned-off-by: NReza Arbab <arbab@linux.vnet.ibm.com>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: NBalbir Singh <bsingharora@gmail.com>
Acked-by: NMichael Ellerman <mpe@ellerman.id.au>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Alistair Popple <apopple@au1.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Stewart Smith <stewart@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The MPOL_F_STATIC_NODES and MPOL_F_RELATIVE_NODES flags are irrelevant
when setting them for MPOL_LOCAL NUMA memory policy via set_mempolicy or
mbind.

Return the "invalid argument" from set_mempolicy and mbind whenever any
of these flags is passed along with MPOL_LOCAL.

It is consistent with MPOL_PREFERRED passed with empty nodemask.

It slightly shortens the execution time in paths where these flags are
used e.g.  when trying to rebind the NUMA nodes for changes in cgroups
cpuset mems (mpol_rebind_preferred()) or when just printing the mempolicy
structure (/proc/PID/numa_maps).  Isolated tests done.

Link: http://lkml.kernel.org/r/20161027163037.4089-1-kwapulinski.piotr@gmail.comSigned-off-by: NPiotr Kwapulinski <kwapulinski.piotr@gmail.com>
Acked-by: NDavid Rientjes <rientjes@google.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Liang Chen <liangchen.linux@gmail.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Nathan Zimmer <nzimmer@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In the previous round of get_user_pages* changes comments attached to
__get_user_pages_unlocked() and get_user_pages_unlocked() were rendered
incorrect, this patch corrects them.

In addition the get_user_pages_unlocked() comment seems to have already
been outdated as it referred to tsk, mm parameters which were removed in
c12d2da5 ("mm/gup: Remove the macro overload API migration helpers from
the get_user*() APIs"), this patch fixes this also.

Link: http://lkml.kernel.org/r/20161025233435.5338-1-lstoakes@gmail.comSigned-off-by: NLorenzo Stoakes <lstoakes@gmail.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now that we check for page size change early in the loop, we can
partially revert e9d55e15 ("mm: change the interface for
__tlb_remove_page").

This simplies the code much, by removing the need to track the last
address with which we adjusted the range.  We also go back to the older
way of filling the mmu_gather array, ie, we add an entry and then check
whether the gather batch is full.

Link: http://lkml.kernel.org/r/20161026084839.27299-6-aneesh.kumar@linux.vnet.ibm.comSigned-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

With commit e77b0852 ("mm/mmu_gather: track page size with mmu
gather and force flush if page size change") we added the ability to
force a tlb flush when the page size change in a mmu_gather loop.  We
did that by checking for a page size change every time we added a page
to mmu_gather for lazy flush/remove.  We can improve that by moving the
page size change check early and not doing it every time we add a page.

This also helps us to do tlb flush when invalidating a range covering
dax mapping.  Wrt dax mapping we don't have a backing struct page and
hence we don't call tlb_remove_page, which earlier forced the tlb flush
on page size change.  Moving the page size change check earlier means we
will do the same even for dax mapping.

We also avoid doing this check on architecture other than powerpc.

In a later patch we will remove page size check from tlb_remove_page().

Link: http://lkml.kernel.org/r/20161026084839.27299-5-aneesh.kumar@linux.vnet.ibm.comSigned-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This add tlb_remove_hugetlb_entry similar to tlb_remove_pmd_tlb_entry.

Link: http://lkml.kernel.org/r/20161026084839.27299-4-aneesh.kumar@linux.vnet.ibm.comSigned-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We use __tlb_adjust_range to update range convered by mmu_gather struct.
We later use the 'start' and 'end' to do a mmu_notifier_invalidate_range
in tlb_flush_mmu_tlbonly().  Update the 'end' correctly in
__tlb_adjust_range so that we call mmu_notifier_invalidate_range with
the correct range values.

Wrt tlbflush, this should not have any impact, because a flush with
correct start address will flush tlb mapping for the range.

Also add comment w.r.t updating the range when we free pagetable pages.
For now we don't support a range based page table cache flush.

Link: http://lkml.kernel.org/r/20161026084839.27299-3-aneesh.kumar@linux.vnet.ibm.comSigned-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We are removing a pmd hugepage here.  Use the correct page size.

Link: http://lkml.kernel.org/r/20161026084839.27299-2-aneesh.kumar@linux.vnet.ibm.comSigned-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

After enabling -Wmaybe-uninitialized warnings, we get a false-postive
warning for shmem:

  mm/shmem.c: In function `shmem_getpage_gfp':
  include/linux/spinlock.h:332:21: error: `info' may be used uninitialized in this function [-Werror=maybe-uninitialized]

This can be easily avoided, since the correct 'info' pointer is known at
the time we first enter the function, so we can simply move the
initialization up.  Moving it before the first label avoids the warning
and lets us remove two later initializations.

Note that the function is so hard to read that it not only confuses the
compiler, but also most readers and without this patch it could\ easily
break if one of the 'goto's changed.

Link: https://www.spinics.net/lists/kernel/msg2368133.html
Link: http://lkml.kernel.org/r/20161024205725.786455-1-arnd@arndb.deSigned-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Since commit bda807d4 ("mm: migrate: support non-lru movable page
migration") isolate_migratepages_block) can isolate !PageLRU pages which
would acct_isolated account as NR_ISOLATED_*.  Accounting these non-lru
pages NR_ISOLATED_{ANON,FILE} doesn't make any sense and it can misguide
heuristics based on those counters such as pgdat_reclaimable_pages resp.
too_many_isolated which would lead to unexpected stalls during the
direct reclaim without any good reason.  Note that
__alloc_contig_migrate_range can isolate a lot of pages at once.

On mobile devices such as 512M ram android Phone, it may use a big zram
swap.  In some cases zram(zsmalloc) uses too many non-lru but
migratedable pages, such as:

      MemTotal: 468148 kB
      Normal free:5620kB
      Free swap:4736kB
      Total swap:409596kB
      ZRAM: 164616kB(zsmalloc non-lru pages)
      active_anon:60700kB
      inactive_anon:60744kB
      active_file:34420kB
      inactive_file:37532kB

Fix this by only accounting lru pages to NR_ISOLATED_* in
isolate_migratepages_block right after they were isolated and we still
know they were on LRU.  Drop acct_isolated because it is called after
the fact and we've lost that information.  Batching per-cpu counter
doesn't make much improvement anyway.  Also make sure that we uncharge
only LRU pages when putting them back on the LRU in
putback_movable_pages resp.  when unmap_and_move migrates the page.

[mhocko@suse.com: replace acct_isolated() with direct counting]
Fixes: bda807d4 ("mm: migrate: support non-lru movable page migration")
Link: http://lkml.kernel.org/r/20161019080240.9682-1-mhocko@kernel.orgSigned-off-by: NMing Ling <ming.ling@spreadtrum.com>
Signed-off-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NMinchan Kim <minchan@kernel.org>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <js1304@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>