Tail page refcounting is utterly complicated and painful to support.

It uses ->_mapcount on tail pages to store how many times this page is
pinned.  get_page() bumps ->_mapcount on tail page in addition to
->_count on head.  This information is required by split_huge_page() to
be able to distribute pins from head of compound page to tails during
the split.

We will need ->_mapcount to account PTE mappings of subpages of the
compound page.  We eliminate need in current meaning of ->_mapcount in
tail pages by forbidding split entirely if the page is pinned.

The only user of tail page refcounting is THP which is marked BROKEN for
now.

Let's drop all this mess.  It makes get_page() and put_page() much
simpler.
Signed-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Tested-by: NSasha Levin <sasha.levin@oracle.com>
Tested-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Acked-by: NJerome Marchand <jmarchan@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Steve Capper <steve.capper@linaro.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

For a pte entry we will have _PAGE_PTE set. Our pte page
address have a minimum alignment requirement of HUGEPD_SHIFT_MASK + 1.
We use the lower 7 bits to indicate hugepd. ie.

For pmd and pgd we can find:
1) _PAGE_PTE set pte -> indicate PTE
2) bits [2..6] non zero -> indicate hugepd.
   They also encode the size. We skip bit 1 (_PAGE_PRESENT).
3) othewise pointer to next table.
Acked-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

W.r.t hugetlb, we support two format for pmd. With book3s_64 and
64K linux page size, we can have pte at the pmd level. Hence we
don't need to support hugepd there. For everything else hugepd
is supported and pmd_huge is (0).
Acked-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

We need to properly identify whether a hugepage is an explicit or
a transparent hugepage in follow_huge_addr(). We used to depend
on hugepage shift argument to do that. But in some case that can
result in wrong results. For ex:

On finding a transparent hugepage we set hugepage shift to PMD_SHIFT.
But we can end up clearing the thp pte, via pmdp_huge_get_and_clear.
We do prevent reusing the pfn page via the usage of
kick_all_cpus_sync(). But that happens after we updated the pte to 0.
Hence in follow_huge_addr() we can find hugepage shift set, but transparent
huge page check fail for a thp pte.

NOTE: We fixed a variant of this race against thp split in commit
691e95fd
("powerpc/mm/thp: Make page table walk safe against thp split/collapse")

Without this patch, we may hit the BUG_ON(flags & FOLL_GET) in
follow_page_mask occasionally.

In the long term, we may want to switch ppc64 64k page size config to
enable CONFIG_ARCH_WANT_GENERAL_HUGETLB
Reported-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

After commit e2b3d202
("powerpc: Switch 16GB and 16MB explicit hugepages to a
different page table format"), we don't need to support
is_hugepd() for 64K page size.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Back in the olden days we added support for using 64K pages to map the
SPU (Synergistic Processing Unit) local store on Cell, when the main
kernel was using 4K pages.

This was useful at the time because distros were using 4K pages, but
using 64K pages on the SPUs could reduce TLB pressure there.

However these days the number of Cell users is approaching zero, and
supporting this option adds unpleasant complexity to the memory
management code.

So drop the option, CONFIG_SPU_FS_64K_LS, and all related code.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Acked-by: NJeremy Kerr <jk@ozlabs.org>

Currently we have many duplicates in definitions of huge_pmd_unshare.  In
all architectures this function just returns 0 when
CONFIG_ARCH_WANT_HUGE_PMD_SHARE is N.

This patch puts the default implementation in mm/hugetlb.c and lets these
architectures use the common code.
Signed-off-by: NZhang Zhen <zhenzhang.zhang@huawei.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Chris Metcalf <cmetcalf@ezchip.com>
Cc: David Rientjes <rientjes@google.com>
Cc: James Yang <James.Yang@freescale.com>
Cc: Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The hugetlbpage.o is obj-y (always built in).  It will never
be modular, so using module_init as an alias for __initcall is
somewhat misleading.

Fix this up now, so that we can relocate module_init from
init.h into module.h in the future.  If we don't do this, we'd
have to add module.h to obviously non-modular code, and that
would be a worse thing.

Note that direct use of __initcall is discouraged, vs. one
of the priority categorized subgroups.  As __initcall gets
mapped onto device_initcall, our use of arch_initcall (which
makes sense for arch code) will thus change this registration
from level 6-device to level 3-arch (i.e. slightly earlier).
However no observable impact of that small difference has
been observed during testing, or is expected.

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: NPaul Gortmaker <paul.gortmaker@windriver.com>

This adds an extra argument onto parse_params() to be used
as a way to make the unused callback a bit more useful and
generic by allowing the caller to pass on a data structure
of its choice. An example use case is to allow us to easily
make module parameters for every module which we will do
next.

@ parse @
identifier name, args, params, num, level_min, level_max;
identifier unknown, param, val, doing;
type s16;
@@
 extern char *parse_args(const char *name,
 			 char *args,
 			 const struct kernel_param *params,
 			 unsigned num,
 			 s16 level_min,
 			 s16 level_max,
+			 void *arg,
 			 int (*unknown)(char *param, char *val,
					const char *doing
+					, void *arg
					));

@ parse_mod @
identifier name, args, params, num, level_min, level_max;
identifier unknown, param, val, doing;
type s16;
@@
 char *parse_args(const char *name,
 			 char *args,
 			 const struct kernel_param *params,
 			 unsigned num,
 			 s16 level_min,
 			 s16 level_max,
+			 void *arg,
 			 int (*unknown)(char *param, char *val,
					const char *doing
+					, void *arg
					))
{
	...
}

@ parse_args_found @
expression R, E1, E2, E3, E4, E5, E6;
identifier func;
@@

(
	R =
	parse_args(E1, E2, E3, E4, E5, E6,
+		   NULL,
		   func);
|
	R =
	parse_args(E1, E2, E3, E4, E5, E6,
+		   NULL,
		   &func);
|
	R =
	parse_args(E1, E2, E3, E4, E5, E6,
+		   NULL,
		   NULL);
|
	parse_args(E1, E2, E3, E4, E5, E6,
+		   NULL,
		   func);
|
	parse_args(E1, E2, E3, E4, E5, E6,
+		   NULL,
		   &func);
|
	parse_args(E1, E2, E3, E4, E5, E6,
+		   NULL,
		   NULL);
)

@ parse_args_unused depends on parse_args_found @
identifier parse_args_found.func;
@@

int func(char *param, char *val, const char *unused
+		 , void *arg
		 )
{
	...
}

@ mod_unused depends on parse_args_found @
identifier parse_args_found.func;
expression A1, A2, A3;
@@

-	func(A1, A2, A3);
+	func(A1, A2, A3, NULL);

Generated-by: Coccinelle SmPL
Cc: cocci@systeme.lip6.fr
Cc: Tejun Heo <tj@kernel.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Felipe Contreras <felipe.contreras@gmail.com>
Cc: Ewan Milne <emilne@redhat.com>
Cc: Jean Delvare <jdelvare@suse.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: linux-kernel@vger.kernel.org
Reviewed-by: NTejun Heo <tj@kernel.org>
Acked-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NLuis R. Rodriguez <mcgrof@suse.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

We need to check whether pte is present in follow_huge_addr() and
properly return NULL if mapping is not present. Also use READ_ONCE
when dereferencing pte_t address.

Without this patch, we may wrongly return a zero pfn page in
follow_huge_addr().
Reviewed-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

For THP that is marked trans splitting, we return the pte.
This require the callers to handle the pmd_trans_splitting scenario,
if they care. All the current callers are either looking at pfn or
write_ok, hence we don't need to update them.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

We can disable a THP split or a hugepage collapse by disabling irq.
We do send IPI to all the cpus in the early part of split/collapse,
and disabling local irq ensure we don't make progress with
split/collapse. If the THP is getting split we return NULL from
find_linux_pte_or_hugepte(). For all the current callers it should be ok.
We need to be careful if we want to use returned pte_t pointer outside
the irq disabled region. W.r.t to THP split, the pfn remains the same,
but then a hugepage collapse will result in a pfn change. There are
few steps we can take to avoid a hugepage collapse.One way is to take page
reference inside the irq disable region. Other option is to take
mmap_sem so that a parallel collapse will not happen. We can also
disable collapse by taking pmd_lock. Another method used by kvm
subsystem is to check whether we had a mmu_notifer update in between
using mmu_notifier_retry().
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Commit dc6c9a35 ("mm: account pmd page tables to the process")
added a counter that is incremented whenever a PMD is allocated and
decremented whenever a PMD is freed.  For hugepages on PPC, common code
is used to allocated PMDs, but arch-specific code is used to free PMDs.

This results in kernel output such as "BUG: non-zero nr_pmds on freeing
mm: 1" when using hugepages.

Update the PPC hugepage PMD freeing code to decrement the count, just
as the above commit did for free_pmd_range().

Fixes: dc6c9a35 ("mm: account pmd page tables to the process")
Signed-off-by: NScott Wood <scottwood@freescale.com>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org # 4.0.x

Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Fix the 32-bit code also]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Currently we have many duplicates in definitions around
follow_huge_addr(), follow_huge_pmd(), and follow_huge_pud(), so this
patch tries to remove the m.  The basic idea is to put the default
implementation for these functions in mm/hugetlb.c as weak symbols
(regardless of CONFIG_ARCH_WANT_GENERAL_HUGETL B), and to implement
arch-specific code only when the arch needs it.

For follow_huge_addr(), only powerpc and ia64 have their own
implementation, and in all other architectures this function just returns
ERR_PTR(-EINVAL).  So this patch sets returning ERR_PTR(-EINVAL) as
default.

As for follow_huge_(pmd|pud)(), if (pmd|pud)_huge() is implemented to
always return 0 in your architecture (like in ia64 or sparc,) it's never
called (the callsite is optimized away) no matter how implemented it is.
So in such architectures, we don't need arch-specific implementation.

In some architecture (like mips, s390 and tile,) their current
arch-specific follow_huge_(pmd|pud)() are effectively identical with the
common code, so this patch lets these architecture use the common code.

One exception is metag, where pmd_huge() could return non-zero but it
expects follow_huge_pmd() to always return NULL.  This means that we need
arch-specific implementation which returns NULL.  This behavior looks
strange to me (because non-zero pmd_huge() implies that the architecture
supports PMD-based hugepage, so follow_huge_pmd() can/should return some
relevant value,) but that's beyond this cleanup patch, so let's keep it.

Justification of non-trivial changes:
- in s390, follow_huge_pmd() checks !MACHINE_HAS_HPAGE at first, and this
  patch removes the check. This is OK because we can assume MACHINE_HAS_HPAGE
  is true when follow_huge_pmd() can be called (note that pmd_huge() has
  the same check and always returns 0 for !MACHINE_HAS_HPAGE.)
- in s390 and mips, we use HPAGE_MASK instead of PMD_MASK as done in common
  code. This patch forces these archs use PMD_MASK, but it's OK because
  they are identical in both archs.
  In s390, both of HPAGE_SHIFT and PMD_SHIFT are 20.
  In mips, HPAGE_SHIFT is defined as (PAGE_SHIFT + PAGE_SHIFT - 3) and
  PMD_SHIFT is define as (PAGE_SHIFT + PAGE_SHIFT + PTE_ORDER - 3), but
  PTE_ORDER is always 0, so these are identical.
Signed-off-by: NNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Nishanth Aravamudan <nacc@linux.vnet.ibm.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Steve Capper <steve.capper@linaro.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

ACCESS_ONCE does not work reliably on non-scalar types. For
example gcc 4.6 and 4.7 might remove the volatile tag for such
accesses during the SRA (scalar replacement of aggregates) step
(https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145)

Change the ppc/hugetlbfs code to replace ACCESS_ONCE with READ_ONCE.
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>

Limit the number of gigantic hugepages specified by the
hugepages= parameter to MAX_NUMBER_GPAGES.
Signed-off-by: NJames Yang <James.Yang@freescale.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Although we are now selecting NO_BOOTMEM, we still have some traces of
bootmem lying around. That is because even with NO_BOOTMEM there is
still a shim that converts bootmem calls into memblock calls, but
ultimately we want to remove all traces of bootmem.

Most of the patch is conversions from alloc_bootmem() to
memblock_virt_alloc(). In general a call such as:

  p = (struct foo *)alloc_bootmem(x);

Becomes:

  p = memblock_virt_alloc(x, 0);

We don't need the cast because memblock_virt_alloc() returns a void *.
The alignment value of zero tells memblock to use the default alignment,
which is SMP_CACHE_BYTES, the same value alloc_bootmem() uses.

We remove a number of NULL checks on the result of
memblock_virt_alloc(). That is because memblock_virt_alloc() will panic
if it can't allocate, in exactly the same way as alloc_bootmem(), so the
NULL checks are and always have been redundant.

The memory returned by memblock_virt_alloc() is already zeroed, so we
remove several memsets of the result of memblock_virt_alloc().

Finally we convert a few uses of __alloc_bootmem(x, y, MAX_DMA_ADDRESS)
to just plain memblock_virt_alloc(). We don't use memblock_alloc_base()
because MAX_DMA_ADDRESS is ~0ul on powerpc, so limiting the allocation
to that is pointless, 16XB ought to be enough for anyone.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

On architectures with hardware broadcasting of TLB invalidation messages
, it makes sense to reduce the range of the mmu_gather structure when
unmapping page ranges based on the dirty address information passed to
tlb_remove_tlb_entry.

arm64 already does this by directly manipulating the start/end fields
of the gather structure, but this confuses the generic code which
does not expect these fields to change and can end up calculating
invalid, negative ranges when forcing a flush in zap_pte_range.

This patch moves the minimal range calculation out of the arm64 code
and into the generic implementation, simplifying zap_pte_range in the
process (which no longer needs to care about start/end, since they will
point to the appropriate ranges already). With the range being tracked
by core code, the need_flush flag is dropped in favour of checking that
the end of the range has actually been set.

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Michal Simek <monstr@monstr.eu>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This patch switch the ppc arch to use the generic RCU based
gup implementation.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This patch add documentation and missing accessors.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Now bootmem is gone from powerpc we can remove comments mentioning it.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Tested-by: NEmil Medve <Emilian.Medve@Freescale.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This still has not been merged and now powerpc is the only arch that does
not have this change. Sorry about missing linuxppc-dev before.

V2->V2
  - Fix up to work against 3.18-rc1

__get_cpu_var() is used for multiple purposes in the kernel source. One of
them is address calculation via the form &__get_cpu_var(x).  This calculates
the address for the instance of the percpu variable of the current processor
based on an offset.

Other use cases are for storing and retrieving data from the current
processors percpu area.  __get_cpu_var() can be used as an lvalue when
writing data or on the right side of an assignment.

__get_cpu_var() is defined as :

__get_cpu_var() always only does an address determination. However, store
and retrieve operations could use a segment prefix (or global register on
other platforms) to avoid the address calculation.

this_cpu_write() and this_cpu_read() can directly take an offset into a
percpu area and use optimized assembly code to read and write per cpu
variables.

This patch converts __get_cpu_var into either an explicit address
calculation using this_cpu_ptr() or into a use of this_cpu operations that
use the offset.  Thereby address calculations are avoided and less registers
are used when code is generated.

At the end of the patch set all uses of __get_cpu_var have been removed so
the macro is removed too.

The patch set includes passes over all arches as well. Once these operations
are used throughout then specialized macros can be defined in non -x86
arches as well in order to optimize per cpu access by f.e.  using a global
register that may be set to the per cpu base.

Transformations done to __get_cpu_var()

1. Determine the address of the percpu instance of the current processor.

	DEFINE_PER_CPU(int, y);
	int *x = &__get_cpu_var(y);

    Converts to

	int *x = this_cpu_ptr(&y);

2. Same as #1 but this time an array structure is involved.

	DEFINE_PER_CPU(int, y[20]);
	int *x = __get_cpu_var(y);

    Converts to

	int *x = this_cpu_ptr(y);

3. Retrieve the content of the current processors instance of a per cpu
variable.

	DEFINE_PER_CPU(int, y);
	int x = __get_cpu_var(y)

   Converts to

	int x = __this_cpu_read(y);

4. Retrieve the content of a percpu struct

	DEFINE_PER_CPU(struct mystruct, y);
	struct mystruct x = __get_cpu_var(y);

   Converts to

	memcpy(&x, this_cpu_ptr(&y), sizeof(x));

5. Assignment to a per cpu variable

	DEFINE_PER_CPU(int, y)
	__get_cpu_var(y) = x;

   Converts to

	__this_cpu_write(y, x);

6. Increment/Decrement etc of a per cpu variable

	DEFINE_PER_CPU(int, y);
	__get_cpu_var(y)++

   Converts to

	__this_cpu_inc(y)

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
CC: Paul Mackerras <paulus@samba.org>
Signed-off-by: NChristoph Lameter <cl@linux.com>
[mpe: Fix build errors caused by set/or_softirq_pending(), and rework
      assignment in __set_breakpoint() to use memcpy().]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This reverts commit 5828f666 due to
build failure after merging with pending powerpc changes.

Link: http://lkml.kernel.org/g/20140827142243.6277eaff@canb.auug.org.auSigned-off-by: NTejun Heo <tj@kernel.org>
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>

__get_cpu_var() is used for multiple purposes in the kernel source. One of
them is address calculation via the form &__get_cpu_var(x).  This calculates
the address for the instance of the percpu variable of the current processor
based on an offset.

Other use cases are for storing and retrieving data from the current
processors percpu area.  __get_cpu_var() can be used as an lvalue when
writing data or on the right side of an assignment.

__get_cpu_var() is defined as :

#define __get_cpu_var(var) (*this_cpu_ptr(&(var)))

__get_cpu_var() always only does an address determination. However, store
and retrieve operations could use a segment prefix (or global register on
other platforms) to avoid the address calculation.

this_cpu_write() and this_cpu_read() can directly take an offset into a
percpu area and use optimized assembly code to read and write per cpu
variables.

This patch converts __get_cpu_var into either an explicit address
calculation using this_cpu_ptr() or into a use of this_cpu operations that
use the offset.  Thereby address calculations are avoided and less registers
are used when code is generated.

At the end of the patch set all uses of __get_cpu_var have been removed so
the macro is removed too.

The patch set includes passes over all arches as well. Once these operations
are used throughout then specialized macros can be defined in non -x86
arches as well in order to optimize per cpu access by f.e.  using a global
register that may be set to the per cpu base.

Transformations done to __get_cpu_var()

1. Determine the address of the percpu instance of the current processor.

	DEFINE_PER_CPU(int, y);
	int *x = &__get_cpu_var(y);

    Converts to

	int *x = this_cpu_ptr(&y);

2. Same as #1 but this time an array structure is involved.

	DEFINE_PER_CPU(int, y[20]);
	int *x = __get_cpu_var(y);

    Converts to

	int *x = this_cpu_ptr(y);

3. Retrieve the content of the current processors instance of a per cpu
variable.

	DEFINE_PER_CPU(int, y);
	int x = __get_cpu_var(y)

   Converts to

	int x = __this_cpu_read(y);

4. Retrieve the content of a percpu struct

	DEFINE_PER_CPU(struct mystruct, y);
	struct mystruct x = __get_cpu_var(y);

   Converts to

	memcpy(&x, this_cpu_ptr(&y), sizeof(x));

5. Assignment to a per cpu variable

	DEFINE_PER_CPU(int, y)
	__get_cpu_var(y) = x;

   Converts to

	__this_cpu_write(y, x);

6. Increment/Decrement etc of a per cpu variable

	DEFINE_PER_CPU(int, y);
	__get_cpu_var(y)++

   Converts to

	__this_cpu_inc(y)

tj: Folded a fix patch.
    http://lkml.kernel.org/g/alpine.DEB.2.11.1408172143020.9652@gentwo.org

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
CC: Paul Mackerras <paulus@samba.org>
Signed-off-by: NChristoph Lameter <cl@linux.com>
Signed-off-by: NTejun Heo <tj@kernel.org>

Currently hugepage migration is available for all archs which support
pmd-level hugepage, but testing is done only for x86_64 and there're
bugs for other archs.  So to avoid breaking such archs, this patch
limits the availability strictly to x86_64 until developers of other
archs get interested in enabling this feature.

Simply disabling hugepage migration on non-x86_64 archs is not enough to
fix the reported problem where sys_move_pages() hits the BUG_ON() in
follow_page(FOLL_GET), so let's fix this by checking if hugepage
migration is supported in vma_migratable().
Signed-off-by: NNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reported-by: NMichael Ellerman <mpe@ellerman.id.au>
Tested-by: NMichael Ellerman <mpe@ellerman.id.au>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: David Miller <davem@davemloft.net>
Cc: <stable@vger.kernel.org>	[3.12+]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Replace __get_cpu_var safely with get_cpu_var to avoid
the following call trace:

[ 7253.637591] BUG: using smp_processor_id() in preemptible [00000000 00000000]
code: hugemmap01/9048
[ 7253.637601] caller is free_hugepd_range.constprop.25+0x88/0x1a8
[ 7253.637605] CPU: 1 PID: 9048 Comm: hugemmap01 Not tainted 3.10.20-rt14+ #114
[ 7253.637606] Call Trace:
[ 7253.637617] [cb049d80] [c0007ea4] show_stack+0x4c/0x168 (unreliable)
[ 7253.637624] [cb049dc0] [c031c674] debug_smp_processor_id+0x114/0x134
[ 7253.637628] [cb049de0] [c0016d28] free_hugepd_range.constprop.25+0x88/0x1a8
[ 7253.637632] [cb049e00] [c001711c] hugetlb_free_pgd_range+0x6c/0x168
[ 7253.637639] [cb049e40] [c0117408] free_pgtables+0x12c/0x150
[ 7253.637646] [cb049e70] [c011ce38] unmap_region+0xa0/0x11c
[ 7253.637671] [cb049ef0] [c011f03c] do_munmap+0x224/0x3bc
[ 7253.637676] [cb049f20] [c011f2e0] vm_munmap+0x38/0x5c
[ 7253.637682] [cb049f40] [c000ef88] ret_from_syscall+0x0/0x3c
[ 7253.637686] --- Exception: c01 at 0xff16004

Signed-off-by: Tiejun Chen<tiejun.chen@windriver.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

The callers of free_pgd_range() and hugetlb_free_pgd_range() don't hold
page table locks.  The comments seems to be obsolete, so let's remove
them.
Signed-off-by: NNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently hugepage migration works well only for pmd-based hugepages
(mainly due to lack of testing,) so we had better not enable migration of
other levels of hugepages until we are ready for it.

Some users of hugepage migration (mbind, move_pages, and migrate_pages) do
page table walk and check pud/pmd_huge() there, so they are safe.  But the
other users (softoffline and memory hotremove) don't do this, so without
this patch they can try to migrate unexpected types of hugepages.

To prevent this, we introduce hugepage_migration_support() as an
architecture dependent check of whether hugepage are implemented on a pmd
basis or not.  And on some architecture multiple sizes of hugepages are
available, so hugepage_migration_support() also checks hugepage size.
Signed-off-by: NNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Hillf Danton <dhillf@gmail.com>
Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Rik van Riel <riel@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use the already existing interface huge_page_shift instead of h->order +
PAGE_SHIFT.
Signed-off-by: NWanpeng Li <liwanp@linux.vnet.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reviewed-by: NMichal Hocko <mhocko@suse.cz>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

GCC is very likely to read the pagetables just once and cache them in
the local stack or in a register, but it is can also decide to re-read
the pagetables. The problem is that the pagetable in those places can
change from under gcc.

With THP/hugetlbfs the pmd (and pgd for hugetlbfs giga pages) can
change under gup_fast. The pages won't be freed untill we finish
gup fast because we have irq disabled and we free these pages via
rcu callback.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

We need to have irqs disabled to handle all the possible parallel update for
linux page table without holding locks.

Events that we are intersted in while walking page tables are
1) Page fault
2) umap
3) THP split
4) THP collapse

A) local_irq_disabled:
------------------------
1) page fault:
A none to valid transition via page fault is not an issue because we
would either see a none or valid. If it is none, we would error out
the page table walk. We may need to use on stack values when checking for
type of page table elements, because if we do

if (!is_hugepd()) {
    if (!pmd_none() {
       if (pmd_bad() {

We could take that bad condition because the pmd got converted to a hugepd
after the !is_hugepd check via a hugetlb fault.

The right way would be to check for pmd_none higher up or use on stack value.

2) A valid to none conversion via unmap:
We can safely walk the upper level table, because we don't remove the the
page table entries until rcu grace period. So even if we followed a
wrong pointer we still have the pointer valid till the grace period.

A PTE pointer returned need to be atomically checked for _PAGE_PRESENT and
 _PAGE_BUSY. A valid pointer returned could becoming none later. To prevent
pte_clear we take _PAGE_BUSY.

3) THP split:
A valid transparent hugepage is converted to nomal page. Before we split we
do pmd_splitting_flush, which sets the hugepage PTE to _PAGE_SPLITTING
So when walking page table we need to check for pmd_trans_splitting and
handle that. The pte returned should also need to be checked for
_PAGE_SPLITTING before setting _PAGE_BUSY similar to _PAGE_PRESENT. We save
the value of PTE on stack and check for the flag in the local pte value.
If we don't have the value set we can safely operate on the local pte value
and we atomicaly set _PAGE_BUSY.

4) THP collapse:
A normal page gets converted to hugepage. In the collapse path, we
mark the pmd none early (pmdp_clear_flush). With irq disabled, if we
are aleady walking page table we would see the pmd_none and won't continue.
If we see a valid PMD, we should still check for _PAGE_PRESENT before
setting _PAGE_BUSY, to make sure we didn't collapse the PTE to a Huge PTE.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Replace find_linux_pte with find_linux_pte_or_hugepte and explicitly
document why we don't need to handle transparent hugepages at callsites.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Reviewed-by: NDavid Gibson <dwg@au1.ibm.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

We will use this in the later patch for handling THP pages
Reviewed-by: NDavid Gibson <dwg@au1.ibm.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Book3E uses the hugepd at PMD level and don't encode pte directly
at the pmd level. So it will find the lower bits of pmd set
and the pmd_bad check throws error. Infact the current code
will never take the free_hugepd_range call at all because it will
clear the pmd if it find a hugepd pointer. Fix this by clearing
bad pmd only if it is not a hugepd pointer.

This is regression introduced by e2b3d202
"powerpc: Switch 16GB and 16MB explicit hugepages to a different page table format"
Reported-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

We will be switching PMD_SHIFT to 24 bits to facilitate THP impmenetation.
With PMD_SHIFT set to 24, we now have 16MB huge pages allocated at PGD level.
That means with 32 bit process we cannot allocate normal pages at
all, because we cover the entire address space with one pgd entry. Fix this
by switching to a new page table format for hugepages. With the new page table
format for 16GB and 16MB hugepages we won't allocate hugepage directory. Instead
we encode the PTE information directly at the directory level. This forces 16MB
hugepage at PMD level. This will also make the page take walk much simpler later
when we add the THP support.

With the new table format we have 4 cases for pgds and pmds:
(1) invalid (all zeroes)
(2) pointer to next table, as normal; bottom 6 bits == 0
(3) leaf pte for huge page, bottom two bits != 00
(4) hugepd pointer, bottom two bits == 00, next 4 bits indicate size of table
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Change the hugepage directory format so that we can have leaf ptes directly
at page directory avoiding the allocation of hugepage directory.

With the new table format we have 3 cases for pgds and pmds:
(1) invalid (all zeroes)
(2) pointer to next table, as normal; bottom 6 bits == 0
(4) hugepd pointer, bottom two bits == 00, next 4 bits indicate size of table

Instead of storing shift value in hugepd pointer we use mmu_psize_def index
so that we can fit all the supported hugepage size in 4 bits
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

As all other architectures have been converted to use vm_unmapped_area(),
we are about to retire the free_area_cache.

This change simply removes the use of that cache in
slice_get_unmapped_area(), which will most certainly have a
performance cost. Next one will convert that function to use the
vm_unmapped_area() infrastructure and regain the performance.
Signed-off-by: NMichel Lespinasse <walken@google.com>
Acked-by: NRik van Riel <riel@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Commit 9fb48c74

    "params: add 3rd arg to option handler callback signature"

added an extra arg to the function, but didn't catch all the use
cases needing it, causing this compile fail in mpc85xx_defconfig:

 arch/powerpc/mm/hugetlbpage.c:316:4: error: passing argument 7 of
 'parse_args' from incompatible pointer type [-Werror]

 include/linux/moduleparam.h:317:12: note: expected
	 'int (*)(char *, char *, const char *)' but argument is of type
	 'int (*)(char *, char *)'

This function has no need to printk out the "doing" value, so
just add the arg as an "unused".

Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jim Cromie <jim.cromie@gmail.com>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Becky Bruce <beckyb@kernel.crashing.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>