Radix guests do normally invalidate process-scoped translations when a
new pid is allocated but migrated guests do not invalidate these so
migrated guests crash sometime, especially easy to reproduce with
migration happening within first 10 seconds after the guest boot start
on the same machine.

This adds the "Invalidate process-scoped translations" flush to fix
radix guests migration.

Fixes: 2ee13be3 ("KVM: PPC: Book3S HV: Update kvmppc_set_arch_compat() for ISA v3.00")
Cc: stable@vger.kernel.org # v4.10+
Signed-off-by: NAlexey Kardashevskiy <aik@ozlabs.ru>
Tested-by: NLaurent Vivier <lvivier@redhat.com>
Tested-by: NDaniel Henrique Barboza <danielhb@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

commit ac29c640 ("powerpc/mm: Replace _PAGE_USER with
_PAGE_PRIVILEGED") introduced _PAGE_PRIVILEGED for BOOK3S/64

This patch generalises _PAGE_PRIVILEGED for all CPUs, allowing
to have either _PAGE_PRIVILEGED or _PAGE_USER or both.

PPC_8xx has a _PAGE_SHARED flag which is set for and only for
all non user pages. Lets rename it _PAGE_PRIVILEGED to remove
confusion as it has nothing to do with Linux shared pages.

On BookE, there's a _PAGE_BAP_SR which has to be set for kernel
pages: defining _PAGE_PRIVILEGED as _PAGE_BAP_SR will make
this generic
Signed-off-by: NChristophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Most callers users of free_hot_cold_page claim the pages being released
are cache hot.  The exception is the page reclaim paths where it is
likely that enough pages will be freed in the near future that the
per-cpu lists are going to be recycled and the cache hotness information
is lost.  As no one really cares about the hotness of pages being
released to the allocator, just ditch the parameter.

The APIs are renamed to indicate that it's no longer about hot/cold
pages.  It should also be less confusing as there are subtle differences
between them.  __free_pages drops a reference and frees a page when the
refcount reaches zero.  free_hot_cold_page handled pages whose refcount
was already zero which is non-obvious from the name.  free_unref_page
should be more obvious.

No performance impact is expected as the overhead is marginal.  The
parameter is removed simply because it is a bit stupid to have a useless
parameter copied everywhere.

[mgorman@techsingularity.net: add pages to head, not tail]
  Link: http://lkml.kernel.org/r/20171019154321.qtpzaeftoyyw4iey@techsingularity.net
Link: http://lkml.kernel.org/r/20171018075952.10627-8-mgorman@techsingularity.netSigned-off-by: NMel Gorman <mgorman@techsingularity.net>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

CONFIG_PPC_STD_MMU_64 indicates support for the "standard" powerpc MMU
on 64-bit CPUs. The "standard" MMU refers to the hash page table MMU
found in "server" processors, from IBM mainly.

Currently CONFIG_PPC_STD_MMU_64 is == CONFIG_PPC_BOOK3S_64. While it's
annoying to have two symbols that always have the same value, it's not
quite annoying enough to bother removing one.

However with the arrival of Power9, we now have the situation where
CONFIG_PPC_STD_MMU_64 is enabled, but the kernel is running using the
Radix MMU - *not* the "standard" MMU. So it is now actively confusing
to use it, because it implies that code is disabled or inactive when
the Radix MMU is in use, however that is not necessarily true.

So s/CONFIG_PPC_STD_MMU_64/CONFIG_PPC_BOOK3S_64/, and do some minor
formatting updates of some of the affected lines.

This will be a pain for backports, but c'est la vie.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Currently KERN_IO_START is defined as:

 #define KERN_IO_START  (KERN_VIRT_START + (KERN_VIRT_SIZE >> 1))

Although it looks like a constant, both the components are actually
variables, to allow us to have a different value between Radix and
Hash with a single kernel.

However that still requires both Radix and Hash to place the kernel IO
region at the same location relative to the start and end of the
kernel virtual region (namely 1/2 way through it), and we'd like to
change that.

So split KERN_IO_START out into its own variable, and initialise it
for Radix and Hash. In the medium term we should be able to
reconsolidate this, by doing a more involved rearrangement of the
location of the regions.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: NBalbir Singh <bsingharora@gmail.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Currently even with STRICT_KERNEL_RWX we leave the __init text marked
executable after init, which is bad.

Add a hook to mark it NX (no-execute) before we free it, and implement
it for radix and hash.

Note that we use __init_end as the end address, not _einittext,
because overlaps_kernel_text() uses __init_end, because there are
additional executable sections other than .init.text between
__init_begin and __init_end.

Tested on radix and hash with:

  0:mon> p $__init_begin
  *** 400 exception occurred

Fixes: 1e0fc9d1 ("powerpc/Kconfig: Enable STRICT_KERNEL_RWX for some configs")
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

The Radix linear mapping code (create_physical_mapping()) tries to use
the largest page size it can at each step. Currently the only reason
it steps down to a smaller page size is if the start addr is
unaligned (never happens in practice), or the end of memory is not
aligned to a huge page boundary.

To support STRICT_RWX we need to break the mapping at __init_begin,
so that the text and rodata prior to that can be marked R_X and the
regular pages after can be marked RW.

Having done that we can now implement mark_rodata_ro() for Radix,
knowing that we won't need to split any mappings.
Signed-off-by: NBalbir Singh <bsingharora@gmail.com>
[mpe: Split down to PAGE_SIZE, not 2MB, rewrite change log]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

With hash we update the bolted pte to mark it read-only. We rely
on the MMU_FTR_KERNEL_RO to generate the correct permissions
for read-only text. The radix implementation just prints a warning
in this implementation
Signed-off-by: NBalbir Singh <bsingharora@gmail.com>
[mpe: Make the warning louder when we don't have MMU_FTR_KERNEL_RO]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Add support for the devmap bit on PTEs and PMDs for PPC64 Book3S.  This
is used to differentiate device backed memory from transparent huge
pages since they are handled in more or less the same manner by the core
mm code.

Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NOliver O'Halloran <oohall@gmail.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Add a trace point for tlbie(l) (Translation Lookaside Buffer Invalidate
Entry (Local)) instructions.

The tlbie instruction has changed over the years, so not all versions
accept the same operands. Use the ISA v3 field operands because they are
the most verbose, we may change them in future.

Example output:

  qemu-system-ppc-5371  [016]  1412.369519: tlbie:
  	tlbie with lpid 0, local 1, rb=67bd8900174c11c1, rs=0, ric=0 prs=0 r=0
Signed-off-by: NBalbir Singh <bsingharora@gmail.com>
[mpe: Add some missing trace_tlbie()s, reword change log]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Introduce a helper pgtable_gfp_flags() which
just returns the current gfp flags and adds
__GFP_ACCOUNT to account for page table allocation.
The generic helper is added to include/asm/pgalloc.h
and has two variants - WARNING ugly bits ahead

1. If the header is included from a module, no check
for mm == &init_mm is done, since init_mm is not
exported
2. For kernel includes, the check is done and required
see (3e79ec7d arch: x86: charge page tables to kmemcg)

The fundamental assumption is that no module should be
doing pgd/pud/pmd and pte alloc's on behalf of init_mm
directly.

NOTE: This adds an overhead to pmd/pud/pgd allocations
similar to x86.  The other alternative was to implement
pmd_alloc_kernel/pud_alloc_kernel and pgd_alloc_kernel
with their offset variants.

For 4k page size, pte_alloc_one no longer calls
pte_alloc_one_kernel.
Signed-off-by: NBalbir Singh <bsingharora@gmail.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

When changing a partition table entry on POWER9, we do a particular
form of the tlbie instruction which flushes all TLBs and caches of
the partition table for a given logical partition ID (LPID).
This instruction has a field in the instruction word, labelled R
(radix), which should be 1 if the partition was previously a radix
partition and 0 if it was a HPT partition.  This implements that
logic.
Signed-off-by: NPaul Mackerras <paulus@ozlabs.org>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

POWER9 contains an off core mmu called the nest mmu (NMMU). This is
used by other hardware units on the chip to translate virtual
addresses into real addresses. The unit attempting an address
translation provides the majority of the context required for the
translation request except for the base address of the partition table
(ie. the PTCR) which needs to be programmed into the NMMU.

This patch adds a call to OPAL to set the PTCR for the nest mmu in
opal_init().
Signed-off-by: NAlistair Popple <alistair@popple.id.au>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

POWER9 requires the host to set up a partition table, which is a
table in memory indexed by logical partition ID (LPID) which
contains the pointers to page tables and process tables for the
host and each guest.

This factors out the initialization of the partition table into
a single function.  This code was previously duplicated between
hash_utils_64.c and pgtable-radix.c.

This provides a function for setting a partition table entry,
which is used in early MMU initialization, and will be used by
KVM whenever a guest is created.  This function includes a tlbie
instruction which will flush all TLB entries for the LPID and
all caches of the partition table entry for the LPID, across the
system.

This also moves a call to memblock_set_current_limit(), which was
in radix_init_partition_table(), but has nothing to do with the
partition table.  By analogy with the similar code for hash, the
call gets moved to near the end of radix__early_init_mmu().  It
now gets called when running as a guest, whereas previously it
would only be called if the kernel is running as the host.
Signed-off-by: NPaul Mackerras <paulus@ozlabs.org>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This is the third version of the patchset previously sent [1].  I have
basically only rebased it on top of 4.7-rc1 tree and dropped "dm: get
rid of superfluous gfp flags" which went through dm tree.  I am sending
it now because it is tree wide and chances for conflicts are reduced
considerably when we want to target rc2.  I plan to send the next step
and rename the flag and move to a better semantic later during this
release cycle so we will have a new semantic ready for 4.8 merge window
hopefully.

Motivation:

While working on something unrelated I've checked the current usage of
__GFP_REPEAT in the tree.  It seems that a majority of the usage is and
always has been bogus because __GFP_REPEAT has always been about costly
high order allocations while we are using it for order-0 or very small
orders very often.  It seems that a big pile of them is just a
copy&paste when a code has been adopted from one arch to another.

I think it makes some sense to get rid of them because they are just
making the semantic more unclear.  Please note that GFP_REPEAT is
documented as

* __GFP_REPEAT: Try hard to allocate the memory, but the allocation attempt

* _might_ fail.  This depends upon the particular VM implementation.
  while !costly requests have basically nofail semantic.  So one could
  reasonably expect that order-0 request with __GFP_REPEAT will not loop
  for ever.  This is not implemented right now though.

I would like to move on with __GFP_REPEAT and define a better semantic
for it.

  $ git grep __GFP_REPEAT origin/master | wc -l
  111
  $ git grep __GFP_REPEAT | wc -l
  36

So we are down to the third after this patch series.  The remaining
places really seem to be relying on __GFP_REPEAT due to large allocation
requests.  This still needs some double checking which I will do later
after all the simple ones are sorted out.

I am touching a lot of arch specific code here and I hope I got it right
but as a matter of fact I even didn't compile test for some archs as I
do not have cross compiler for them.  Patches should be quite trivial to
review for stupid compile mistakes though.  The tricky parts are usually
hidden by macro definitions and thats where I would appreciate help from
arch maintainers.

[1] http://lkml.kernel.org/r/1461849846-27209-1-git-send-email-mhocko@kernel.org

This patch (of 19):

__GFP_REPEAT has a rather weak semantic but since it has been introduced
around 2.6.12 it has been ignored for low order allocations.  Yet we
have the full kernel tree with its usage for apparently order-0
allocations.  This is really confusing because __GFP_REPEAT is
explicitly documented to allow allocation failures which is a weaker
semantic than the current order-0 has (basically nofail).

Let's simply drop __GFP_REPEAT from those places.  This would allow to
identify place which really need allocator to retry harder and formulate
a more specific semantic for what the flag is supposed to do actually.

Link: http://lkml.kernel.org/r/1464599699-30131-2-git-send-email-mhocko@kernel.orgSigned-off-by: NMichal Hocko <mhocko@suse.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Chen Liqin <liqin.linux@gmail.com>
Cc: Chris Metcalf <cmetcalf@mellanox.com> [for tile]
Cc: Guan Xuetao <gxt@mprc.pku.edu.cn>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Helge Deller <deller@gmx.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: John Crispin <blogic@openwrt.org>
Cc: Lennox Wu <lennox.wu@gmail.com>
Cc: Ley Foon Tan <lftan@altera.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Rich Felker <dalias@libc.org>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Only code movement in this patch. No functionality change.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

In this patch we make the number of pte fragments per level 4 page table
page a variable. Radix level 4 table size is 256 bytes and hence we can
have 256 fragments per level 4 page. We don't update the fragment count
in this patch. We need to do performance measurements to find the right
value for fragment count.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

The vmalloc range differs between hash and radix config. Hence make
VMALLOC_START and related constants a variable which will be runtime
initialized depending on whether hash or radix mode is active.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Fix missing init of ioremap_bot in pgtable_64.c for ppc64e]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This patch switches 4K Linux page size config to use pte_t * type
instead of struct page * for pgtable_t. This simplifies the code a lot
and helps in consolidating both 64K and 4K page allocator routines. The
changes should not have any impact, because we already store physical
address in the upper level page table tree and that implies we already
do struct page * to physical address conversion.

One change to note here is we move the pgtable_page_dtor() call for
nohash to pte_fragment_free_mm(). The nohash related change is due to
the related changes in pgtable_64.c.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Only code cleanup. No functionality change.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Radix and hash MMU models support different page table sizes. Make
the #defines a variable so that existing code can work with variable
sizes.

Slice related code is only used by hash, so use hash constants there. We
will replicate some of the boundary conditions with resepct to TASK_SIZE
using radix values too. Right now we do boundary condition check using
hash constants.

Swapper pgdir size is initialized in asm code. We select the max pgd
size to keep it simple. For now we select hash pgdir. When adding radix
we will switch that to radix pgdir which is 64K.

BUILD_BUG_ON check which is removed is already done in hugepage_init()
using MAYBE_BUILD_BUG_ON().
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This helps to make following hash only pte bits easier.

We have kept _PAGE_CHG_MASK, _HPAGE_CHG_MASK and _PAGE_PROT_BITS as it
is in this patch eventhough they use hash specific bits. Using them in
radix as it is should be ok, because with radix we expect those bit
positions to be zero.

Only renames in this patch, no change in functionality.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This patch reduces the number of #ifdefs in C code and will also help in
adding radix changes later. Only code movement in this patch.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Propagate copyrights and update GPL text]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

PowerISA 3.0 adds a parition table indexed by LPID. Parition table
allows us to specify the MMU model that will be used for guest and host
translation.

This patch adds support with SLB based hash model (UPRT = 0). What is
required with this model is to support the new hash page table entry
format and also setup partition table such that we use hash table for
address translation.

We don't have segment table support yet.

In order to make sure we don't load KVM module on Power9 (since we don't
have kvm support yet) this patch also disables KVM on Power9.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

The radix variant is going to require a flush_pmd_tlb_range(). With
flush_pmd_tlb_range() added, pmdp_clear_flush_young() is the same as the
generic version. So drop the powerpc specific variant.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

PowerISA 3.0 introduces two pte bits with the below meaning for radix:
  00 -> Normal Memory
  01 -> Strong Access Order (SAO)
  10 -> Non idempotent I/O (Cache inhibited and guarded)
  11 -> Tolerant I/O (Cache inhibited)

We drop the existing WIMG bits in the Linux page table in favour of the
above constants. We loose _PAGE_WRITETHRU with this conversion. We only
use writethru via pgprot_cached_wthru() which is used by
fbdev/controlfb.c which is Apple control display and also PPC32.

With respect to _PAGE_COHERENCE, we have been marking hpte always
coherent for some time now. htab_convert_pte_flags() always added
HPTE_R_M.

NOTE: KVM changes need closer review.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Use a helper instead of open coding with constants. A later patch will
drop the WIMG bits and use PowerISA 3.0 defines.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

PTE_RPN_SHIFT is actually page size dependent. Even though PowerISA 3.0
expects only the lower 12 bits to be zero, we will always find the pages
to be PAGE_SHIFT aligned. In case of hash config, this also allows us to
use the additional 3 bits to track pte specific information. We need
to make sure we use these bits only for hash specific pte flags.

For both 4K and 64K config, pte now can hold 57 bits address.

Inorder to keep things simpler, drop PTE_RPN_SHIFT and PTE_RPN_SIZE and
specify the 57 bit detail explicitly.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

_PAGE_PRIVILEGED means the page can be accessed only by the kernel. This
is done to keep pte bits similar to PowerISA 3.0 Radix PTE format. User
pages are now marked by clearing _PAGE_PRIVILEGED bit.

Previously we allowed the kernel to have a privileged page in the lower
address range (USER_REGION). With this patch such access is denied.

We also prevent a kernel access to a non-privileged page in higher
address range (ie, REGION_ID != 0).

Both the above access scenarios should never happen.

Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Jeremy Kerr <jk@ozlabs.org>
Cc: Frederic Barrat <fbarrat@linux.vnet.ibm.com>
Acked-by: NIan Munsie <imunsie@au1.ibm.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This splits the _PAGE_RW bit into _PAGE_READ and _PAGE_WRITE. It also
removes the dependency on _PAGE_USER for implying read only. Few things
to note here is that, we have read implied with write and execute
permission. Hence we should always find _PAGE_READ set on hash pte
fault.

We still can't switch PROT_NONE to !(_PAGE_RWX). Auto numa depends on
marking a prot none pte _PAGE_WRITE. (For more details look at
b191f9b1 "mm: numa: preserve PTE write permissions across a NUMA
hinting fault")

Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Jeremy Kerr <jk@ozlabs.org>
Cc: Frederic Barrat <fbarrat@linux.vnet.ibm.com>
Acked-by: NIan Munsie <imunsie@au1.ibm.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Traditionally Power server machines have used the Hashed Page Table MMU
mode. In this mode Linux manages its own tree of nested page tables,
aka. "the Linux page tables", which are not used by the hardware
directly, and software loads translations into the hash page table for
use by the hardware.

Power ISA 3.0 defines a new MMU mode, known as Radix Tree Translation,
where the hardware can directly operate on the Linux page tables.
However the hardware requires that the page tables be in big endian
format.

To accommodate this, switch the pgtable types to __be64 and add
appropriate endian conversions.

Because we will be supporting a single kernel binary that boots using
either radix or hash mode, we always store the Linux page tables big
endian, even in hash mode where they are not actually used by the
hardware.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Fix sparse errors, flesh out change log]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

pmd_hugepage_update() is inside #ifdef CONFIG_TRANSPARENT_HUGEPAGE. THP
can only be enabled if PPC_BOOK3S_64=y && PPC_64K_PAGES=y, aka. hash64.

On hash64 we always define PTE_ATOMIC_UPDATES to 1, meaning the #ifdef
in pmd_hugepage_update() is unnecessary, so drop it.

That is also the only use of PTE_ATOMIC_UPDATES in any of the hash code,
meaning we no longer need to #define it at all in the hash headers.

Note it's still #defined and used in the nohash code.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

The success of CMA allocation largely depends on the success of
migration and key factor of it is page reference count.  Until now, page
reference is manipulated by direct calling atomic functions so we cannot
follow up who and where manipulate it.  Then, it is hard to find actual
reason of CMA allocation failure.  CMA allocation should be guaranteed
to succeed so finding offending place is really important.

In this patch, call sites where page reference is manipulated are
converted to introduced wrapper function.  This is preparation step to
add tracepoint to each page reference manipulation function.  With this
facility, we can easily find reason of CMA allocation failure.  There is
no functional change in this patch.

In addition, this patch also converts reference read sites.  It will
help a second step that renames page._count to something else and
prevents later attempt to direct access to it (Suggested by Andrew).
Signed-off-by: NJoonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: NMichal Nazarewicz <mina86@mina86.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

With next generation power processor, we are having a new mmu model
[1] that require us to maintain a different linux page table format.

Inorder to support both current and future ppc64 systems with a single
kernel we need to make sure kernel can select between different page
table format at runtime. With the new MMU (radix MMU) added, we will
have two different pmd hugepage size 16MB for hash model and 2MB for
Radix model. Hence make HPAGE_PMD related values as a variable.

Actual conversion of HPAGE_PMD to a variable for ppc64 happens in a
followup patch.

[1] http://ibm.biz/power-isa3 (Needs registration).
Signed-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This changes _PAGE_PRESENT for 64-bit Book 3S processors from 0x2 to
0x8000_0000_0000_0000, because that is where PowerISA v3.0 CPUs in
radix mode will expect to find it.
Signed-off-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This frees up bits 57-63 in the Linux PTE on 64-bit Book 3S machines.
In the 4k page case, this is done just by reducing the size of the
RPN field to 39 bits, giving 51-bit real addresses.  In the 64k page
case, we had 10 unused bits in the middle of the PTE, so this moves
the RPN field down 10 bits to make use of those unused bits.  This
means the RPN field is now 3 bits larger at 37 bits, giving 53-bit
real addresses in the normal case, or 49-bit real addresses for the
special 4k PFN case.

We are doing this in order to be able to move some other PTE bits
into the positions where PowerISA V3.0 processors will expect to
find them in radix-tree mode.  Ultimately we will be able to move
the RPN field to lower bit positions and make it larger.
Signed-off-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

With ppc64 we use the deposited pgtable_t to store the hash pte slot
information. We should not withdraw the deposited pgtable_t without
marking the pmd none. This ensure that low level hash fault handling
will skip this huge pte and we will handle them at upper levels.

Recent change to pmd splitting changed the above in order to handle the
race between pmd split and exit_mmap. The race is explained below.

Consider following race:

		CPU0				CPU1
shrink_page_list()
  add_to_swap()
    split_huge_page_to_list()
      __split_huge_pmd_locked()
        pmdp_huge_clear_flush_notify()
	// pmd_none() == true
					exit_mmap()
					  unmap_vmas()
					    zap_pmd_range()
					      // no action on pmd since pmd_none() == true
	pmd_populate()

As result the THP will not be freed. The leak is detected by check_mm():

	BUG: Bad rss-counter state mm:ffff880058d2e580 idx:1 val:512

The above required us to not mark pmd none during a pmd split.

The fix for ppc is to clear the huge pte of _PAGE_USER, so that low
level fault handling code skip this pte. At higher level we do take ptl
lock. That should serialze us against the pmd split. Once the lock is
acquired we do check the pmd again using pmd_same. That should always
return false for us and hence we should retry the access. We do the
pmd_same check in all case after taking plt with
THP (do_huge_pmd_wp_page, do_huge_pmd_numa_page and
huge_pmd_set_accessed)

Also make sure we wait for irq disable section in other cpus to finish
before flipping a huge pte entry with a regular pmd entry. Code paths
like find_linux_pte_or_hugepte depend on irq disable to get
a stable pte_t pointer. A parallel thp split need to make sure we
don't convert a pmd pte to a regular pmd entry without waiting for the
irq disable section to finish.

Fixes: eef1b3ba ("thp: implement split_huge_pmd()")
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

With new refcounting we don't need to mark PMDs splitting.  Let's drop
code to handle this.

pmdp_splitting_flush() is not needed too: on splitting PMD we will do
pmdp_clear_flush() + set_pte_at().  pmdp_clear_flush() will do IPI as
needed for fast_gup.
Signed-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Tested-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
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>