For addresses above 512TB we allocate additional mmu contexts. To make
it all easy, addresses above 512TB are handled with IR/DR=1 and with
stack frame setup.

The mmu_context_t is also updated to track the new extended_ids. To
support upto 4PB we need a total 8 contexts.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Minor formatting tweaks and comment wording, switch BUG to WARN
      in get_ea_context().]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

While the implementation of the "slices" address space allows
a significant amount of high slices, it limits the number of
low slices to 16 due to the use of a single u64 low_slices_psize
element in struct mm_context_t

On the 8xx, the minimum slice size is the size of the area
covered by a single PMD entry, ie 4M in 4K pages mode and 64M in
16K pages mode. This means we could have at least 64 slices.

In order to override this limitation, this patch switches the
handling of low_slices_psize to char array as done already for
high_slices_psize.
Signed-off-by: NChristophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Radix keeps no meaningful state in addr_limit, so remove it from radix
code and rename to slb_addr_limit to make it clear it applies to hash
only.
Signed-off-by: NNicholas Piggin <npiggin@gmail.com>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

In commit e6f81a92 ("powerpc/mm/hash: Support 68 bit VA") the
masking is folded into ASM_VSID_SCRAMBLE but the comment about masking
is removed only from the firt use of ASM_VSID_SCRAMBLE.
Signed-off-by: NMichal Suchanek <msuchanek@suse.de>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

On 64-bit book3s, with the hash MMU, we currently define the kernel
virtual space (vmalloc, ioremap etc.), to be 16T in size. This is a
leftover from pre v3.7 when our user VM was also 16T.

Of that 16T we split it 50/50, with half used for PCI IO and ioremap
and the other 8T for vmalloc.

We never bothered to make it any bigger because 8T of vmalloc ought to
be enough for anybody. But it turns out that's not true, the per cpu
allocator wants large amounts of vmalloc space, not to make large
allocations, but to allow a large stride between allocations, because
we use pcpu_embed_first_chunk().

With a bit of juggling we can increase the entire kernel virtual space
to 64T. The only real complication is the check of the address in the
SLB miss handler, see the comment in the code.

Although we could continue to split virtual space 50/50 as we do now,
no one seems to be running out of PCI IO or ioremap space. So instead
keep that as 8T, and use the remaining 56T for vmalloc.

In future we should be able to increase the kernel virtual space to
512T, the code already supports that, it just needs testing on older
hardware.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>

There is a comment in slb_allocate() referring to the load of
paca->vmalloc_sllp, but it's several lines prior in the assembly.
We're about to change this code, and we want to add another comment,
so move the comment immediately prior to the instruction it's talking
about.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

As for slb_miss_realmode(), rename slb_allocate_realmode() to avoid
confusion over whether it runs in real or virtual mode - it runs in
both.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Reviewed-by: NNicholas Piggin <npiggin@gmail.com>

One fewer registers clobbered by this function means the SLB miss
handler can save one fewer.
Signed-off-by: NNicholas Piggin <npiggin@gmail.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

We optmize the slice page size array copy to paca by copying only the
range based on addr_limit. This will require us to not look at page size
array beyond addr_limit in PACA on slb fault. To enable that copy task
size to paca which will be used during slb fault.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Rename from task_size to addr_limit, consolidate #ifdefs]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Inorder to support large effective address range (512TB), we want to
increase the virtual address bits to 68. But we do have platforms like
p4 and p5 that can only do 65 bit VA. We support those platforms by
limiting context bits on them to 16.

The protovsid -> vsid conversion is verified to work with both 65 and 68
bit va values. I also documented the restrictions in a table format as
part of code comments.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Currently we use the top 4 context ids (0x7fffc-0x7ffff) for the kernel.
Kernel VSIDs are built using these top context values and effective the
segement ID. In subsequent patches we want to increase the max effective
address to 512TB. We will achieve that by increasing the effective
segment IDs there by increasing virtual address range.

We will be switching to a 68bit virtual address in the following patch.
But platforms like Power4 and Power5 only support a 65 bit virtual
address. We will handle that by limiting the context bits to 16 instead
of 19 on those platforms. That means the max context id will have a
different value on different platforms.

So that we don't have to deal with the kernel context ids changing
between different platforms, move the kernel context ids down to use
context ids 1-4.

We can't use segment 0 of context-id 0, because that maps to VSID 0,
which we want to keep as invalid, so we avoid context-id 0 entirely.
Similarly we can't use the last segment of the maximum context, so we
avoid it too.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Switch from 0-3 to 1-4 so VSID=0 remains invalid]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

We can't sensibly take a trap at this point. So, blacklist these
symbols.
Reported-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NNaveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

slb_finish_load and slb_finish_load_1T are both only used within
slb_low.S, so make them local symbols.

This makes the code a little clearer, as it's more obvious neither is
intended to be an entry point from arbitrary other code, only the uses
in this file.

It also prevents them being used with kprobes and other tracing tools,
which is good because we're not able to safely take traps at these
locations, so making them local symbols avoids us needing to blacklist
them.
Signed-off-by: NNaveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Currently, if userspace or the kernel accesses a completely bogus address,
for example with any of bits 46-59 set, we first take an SLB miss interrupt,
install a corresponding SLB entry with VSID 0, retry the instruction, then
take a DSI/ISI interrupt because there is no HPT entry mapping the address.
However, by the time of the second interrupt, the Come-From Address Register
(CFAR) has been overwritten by the rfid instruction at the end of the SLB
miss interrupt handler.  Since bogus accesses can often be caused by a
function return after the stack has been overwritten, the CFAR value would
be very useful as it could indicate which function it was whose return had
led to the bogus address.

This patch adds code to create a full exception frame in the SLB miss handler
in the case of a bogus address, rather than inserting an SLB entry with a
zero VSID field.  Then we call a new slb_miss_bad_addr() function in C code,
which delivers a signal for a user access or creates an oops for a kernel
access.  In the latter case the oops message will show the CFAR value at the
time of the access.

In the case of the radix MMU, a segment miss interrupt indicates an access
outside the ranges mapped by the page tables.  Previously this was handled
by the code for an unrecoverable SLB miss (one with MSR[RI] = 0), which is
not really correct.  With this patch, we now handle these interrupts with
slb_miss_bad_addr(), which is much more consistent.
Signed-off-by: NPaul Mackerras <paulus@ozlabs.org>
Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

In commit c60ac569 ("powerpc: Update kernel VSID range", 2013-03-13)
we lost a check on the region number (the top four bits of the effective
address) for addresses below PAGE_OFFSET.  That commit replaced a check
that the top 18 bits were all zero with a check that bits 46 - 59 were
zero (performed for all addresses, not just user addresses).

This means that userspace can access an address like 0x1000_0xxx_xxxx_xxxx
and we will insert a valid SLB entry for it.  The VSID used will be the
same as if the top 4 bits were 0, but the page size will be some random
value obtained by indexing beyond the end of the mm_ctx_high_slices_psize
array in the paca.  If that page size is the same as would be used for
region 0, then userspace just has an alias of the region 0 space.  If the
page size is different, then no HPTE will be found for the access, and
the process will get a SIGSEGV (since hash_page_mm() will refuse to create
a HPTE for the bogus address).

The access beyond the end of the mm_ctx_high_slices_psize can be at most
5.5MB past the array, and so will be in RAM somewhere.  Since the access
is a load performed in real mode, it won't fault or crash the kernel.
At most this bug could perhaps leak a little bit of information about
blocks of 32 bytes of memory located at offsets of i * 512kB past the
paca->mm_ctx_high_slices_psize array, for 1 <= i <= 11.

Fixes: c60ac569 ("powerpc: Update kernel VSID range")
Cc: stable@vger.kernel.org # v3.9+
Signed-off-by: NPaul Mackerras <paulus@ozlabs.org>
Reviewed-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>

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>

We have a bunch of SLB related code in the tree which is there to handle
dynamic VSIDs - but currently it's all disabled at compile time. The
comments say "Keep that around for when we re-implement dynamic VSIDs".

But that was over 10 years ago (commit 3c726f8d ("[PATCH] ppc64:
support 64k pages")). The chance that it would still work unchanged is
minimal, and in the meantime it's confusing to folks browsing/grepping
the code. If we ever want to re-instate it, it's in the git history.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Acked-by: NBalbir Singh <bsingharora@gmail.com>

In case of extending the eaddr in future, use this macro
PGTABLE_EADDR_SIZE to ease the maintenance of the code.
Signed-off-by: NLiu Ping Fan <pingfank@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

The MMU hashtable and SLB branch patching code uses function
pointers for the update sites. This creates a difference between
ABIv1 and ABIv2 because we don't have function descriptors on
ABIv2.

Get rid of the function pointer and just point at the update
sites directly. This works on both ABIs.
Signed-off-by: NAnton Blanchard <anton@samba.org>

Now we use ESID_BITS of kernel address to build proto vsid. So rename
USER_ESIT_BITS to ESID_BITS
Acked-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
CC: <stable@vger.kernel.org> [v3.8]

This patch change the kernel VSID range so that we limit VSID_BITS to 37.
This enables us to support 64TB with 65 bit VA (37+28). Without this patch
we have boot hangs on platforms that only support 65 bit VA.

With this patch we now have proto vsid generated as below:

We first generate a 37-bit "proto-VSID". Proto-VSIDs are generated
from mmu context id and effective segment id of the address.

For user processes max context id is limited to ((1ul << 19) - 5)
for kernel space, we use the top 4 context ids to map address as below
0x7fffc -  [ 0xc000000000000000 - 0xc0003fffffffffff ]
0x7fffd -  [ 0xd000000000000000 - 0xd0003fffffffffff ]
0x7fffe -  [ 0xe000000000000000 - 0xe0003fffffffffff ]
0x7ffff -  [ 0xf000000000000000 - 0xf0003fffffffffff ]
Acked-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Tested-by: NGeoff Levand <geoff@infradead.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
CC: <stable@vger.kernel.org> [v3.8]

Increase max addressable range to 64TB. This is not tested on
real hardware yet.
Reviewed-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

With larger vsid we need to track more bits of ESID in slb cache
for slb invalidate.
Reviewed-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

ASM_VSID_SCRAMBLE can leave non-zero bits in the high 28 bits of the result
for 256MB segment (40 bits for 1T segment). Properly mask them before using
the values in slbmte
Reviewed-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

This patch makes the high psizes mask as an unsigned char array
so that we can have more than 16TB. Currently we support upto
64TB
Reviewed-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

This removes the various bits of assembly in the kernel entry,
exception handling and SLB management code that were specific
to running under the legacy iSeries hypervisor which is no
longer supported.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Some of the 64bit PPC CPU features are MMU-related, so this patch moves
them to MMU_FTR_ bits.  All cpu_has_feature()-style tests are moved to
mmu_has_feature(), and seven feature bits are freed as a result.
Signed-off-by: NMatt Evans <matt@ozlabs.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

On pSeries, we always force the IO space to be mapped using 4K
pages even with a 64K base page size to cope with some limitations
in the HV interface to some devices.

However, the SLB miss handler code to discriminate between vmalloc
and ioremap space uses a CPU feature section such that the code
is nop'ed out when the processor support large pages non-cachable
mappings.

Thus, we end up always using the ioremap page size for vmalloc
segments on such processors, causing a discrepency between the
segment and the hash table, and thus a hang continously hashing
the page.

It works for the first segment of the vmalloc space since that
segment is "bolted" in by C code correctly, and thankfully we
almost never use the vmalloc space beyond the first segment,
but the new percpu code made the bug happen.

This fixes it by removing the feature section from the assembly,
we now always do the comparison between vmalloc and ioremap.

Signed-off-by; Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

This changes vmemmap to use a different region (region 0xf) of the
address space, and to configure the page size of that region
dynamically at boot.

The problem with the current approach of always using 16M pages is that
it's not well suited to machines that have small amounts of memory such
as small partitions on pseries, or PS3's.

In fact, on the PS3, failure to allocate the 16M page backing vmmemmap
tends to prevent hotplugging the HV's "additional" memory, thus limiting
the available memory even more, from my experience down to something
like 80M total, which makes it really not very useable.

The logic used by my match to choose the vmemmap page size is:

 - If 16M pages are available and there's 1G or more RAM at boot,
   use that size.
 - Else if 64K pages are available, use that
 - Else use 4K pages

I've tested on a POWER6 (16M pages) and on an iSeries POWER3 (4K pages)
and it seems to work fine.

Note that I intend to change the way we organize the kernel regions &
SLBs so the actual region will change from 0xf back to something else at
one point, as I simplify the SLB miss handler, but that will be for a
later patch.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Currently we hardwire the number of SLBs to 64, but PAPR says we
should use the ibm,slb-size property to obtain the number of SLB
entries.  This uses this property instead of assuming 64.  If no
property is found, we assume 64 entries as before.

This soft patches the SLB handler, so it shouldn't change performance
at all.
Signed-off-by: NMichael Neuling <mikey@neuling.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This makes the kernel use 1TB segments for all kernel mappings and for
user addresses of 1TB and above, on machines which support them
(currently POWER5+, POWER6 and PA6T).

We detect that the machine supports 1TB segments by looking at the
ibm,processor-segment-sizes property in the device tree.

We don't currently use 1TB segments for user addresses < 1T, since
that would effectively prevent 32-bit processes from using huge pages
unless we also had a way to revert to using 256MB segments.  That
would be possible but would involve extra complications (such as
keeping track of which segment size was used when HPTEs were inserted)
and is not addressed here.

Parts of this patch were originally written by Ben Herrenschmidt.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The basic issue is to be able to do what hugetlbfs does but with
different page sizes for some other special filesystems; more
specifically, my need is:

 - Huge pages

 - SPE local store mappings using 64K pages on a 4K base page size
kernel on Cell

 - Some special 4K segments in 64K-page kernels for mapping a dodgy
type of powerpc-specific infiniband hardware that requires 4K MMU
mappings for various reasons I won't explain here.

The main issues are:

 - To maintain/keep track of the page size per "segment" (as we can
only have one page size per segment on powerpc, which are 256MB
divisions of the address space).

 - To make sure special mappings stay within their allotted
"segments" (including MAP_FIXED crap)

 - To make sure everybody else doesn't mmap/brk/grow_stack into a
"segment" that is used for a special mapping

Some of the necessary mechanisms to handle that were present in the
hugetlbfs code, but mostly in ways not suitable for anything else.

The patch relies on some changes to the generic get_unmapped_area()
that just got merged.  It still hijacks hugetlb callbacks here or
there as the generic code hasn't been entirely cleaned up yet but
that shouldn't be a problem.

So what is a slice ?  Well, I re-used the mechanism used formerly by our
hugetlbfs implementation which divides the address space in
"meta-segments" which I called "slices".  The division is done using
256MB slices below 4G, and 1T slices above.  Thus the address space is
divided currently into 16 "low" slices and 16 "high" slices.  (Special
case: high slice 0 is the area between 4G and 1T).

Doing so simplifies significantly the tracking of segments and avoids
having to keep track of all the 256MB segments in the address space.

While I used the "concepts" of hugetlbfs, I mostly re-implemented
everything in a more generic way and "ported" hugetlbfs to it.

Slices can have an associated page size, which is encoded in the mmu
context and used by the SLB miss handler to set the segment sizes.  The
hash code currently doesn't care, it has a specific check for hugepages,
though I might add a mechanism to provide per-slice hash mapping
functions in the future.

The slice code provide a pair of "generic" get_unmapped_area() (bottomup
and topdown) functions that should work with any slice size.  There is
some trickiness here so I would appreciate people to have a look at the
implementation of these and let me know if I got something wrong.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

and use it an all the obvious places in assembler code.
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>

Signed-off-by: NJörn Engel <joern@wohnheim.fh-wedel.de>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

Some POWER5+ machines can do 64k hardware pages for normal memory but
not for cache-inhibited pages.  This patch lets us use 64k hardware
pages for most user processes on such machines (assuming the kernel
has been configured with CONFIG_PPC_64K_PAGES=y).  User processes
start out using 64k pages and get switched to 4k pages if they use any
non-cacheable mappings.

With this, we use 64k pages for the vmalloc region and 4k pages for
the imalloc region.  If anything creates a non-cacheable mapping in
the vmalloc region, the vmalloc region will get switched to 4k pages.
I don't know of any driver other than the DRM that would do this,
though, and these machines don't have AGP.

When a region gets switched from 64k pages to 4k pages, we do not have
to clear out all the 64k HPTEs from the hash table immediately.  We
use the _PAGE_COMBO bit in the Linux PTE to indicate whether the page
was hashed in as a 64k page or a set of 4k pages.  If hash_page is
trying to insert a 4k page for a Linux PTE and it sees that it has
already been inserted as a 64k page, it first invalidates the 64k HPTE
before inserting the 4k HPTE.  The hash invalidation routines also use
the _PAGE_COMBO bit, to determine whether to look for a 64k HPTE or a
set of 4k HPTEs to remove.  With those two changes, we can tolerate a
mix of 4k and 64k HPTEs in the hash table, and they will all get
removed when the address space is torn down.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This patch removes all self references and fixes references to files
in the now defunct arch/ppc64 tree.  I think this accomplises
everything wanted, though there might be a few references I missed.
Signed-off-by: NJon Mason <jdmason@us.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This patch separates usage of KERNELBASE and PAGE_OFFSET. I haven't
looked at any of the PPC32 code, if we ever want to support Kdump on
PPC we'll have to do another audit, ditto for iSeries.

This patch makes PAGE_OFFSET the constant, it'll always be 0xC * 1
gazillion for 64-bit.

To get a physical address from a virtual one you subtract PAGE_OFFSET,
_not_ KERNELBASE.

KERNELBASE is the virtual address of the start of the kernel, it's
often the same as PAGE_OFFSET, but _might not be_.

If you want to know something's offset from the start of the kernel
you should subtract KERNELBASE.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This patch, however, should be applied on top of the 64k-page-size patch to
fix some problems with hugepage (some pre-existing, another introduced by
this patch).

The patch fixes a bug in the SLB miss handler for hugepages on ppc64
introduced by the dynamic hugepage patch (commit id
c594adad) due to a misunderstanding of the
srd instruction's behaviour (mea culpa).  The problem arises when a 64-bit
process maps some hugepages in the low 4GB of the address space (unusual).
In this case, as well as the 256M segment in question being marked for
hugepages, other segments at 32G intervals will be incorrectly marked for
hugepages.

In the process, this patch tweaks the semantics of the hugepage bitmaps to
be more sensible.  Previously, an address below 4G was marked for hugepages
if the appropriate segment bit in the "low areas" bitmask was set *or* if
the low bit in the "high areas" bitmap was set (which would mark all
addresses below 1TB for hugepage).  With this patch, any given address is
governed by a single bitmap.  Addresses below 4GB are marked for hugepage
if and only if their bit is set in the "low areas" bitmap (256M
granularity).  Addresses between 4GB and 1TB are marked for hugepage iff
the low bit in the "high areas" bitmap is set.  Higher addresses are marked
for hugepage iff their bit in the "high areas" bitmap is set (1TB
granularity).

To avoid conflicts, this patch must be applied on top of BenH's pending
patch for 64k base page size [0].  As such, this patch also addresses a
hugepage problem introduced by that patch.  That patch allows hugepages of
1MB in size on hardware which supports it, however, that won't work when
using 4k pages (4 level pagetable), because in that case hugepage PTEs are
stored at the PMD level, and each PMD entry maps 2MB.  This patch simply
disallows hugepages in that case (we can do something cleverer to re-enable
them some other day).

Built, booted, and a handful of hugepage related tests passed on POWER5
LPAR (both ARCH=powerpc and ARCH=ppc64).

[0] http://gate.crashing.org/~benh/ppc64-64k-pages.diffSigned-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Adds a new CONFIG_PPC_64K_PAGES which, when enabled, changes the kernel
base page size to 64K.  The resulting kernel still boots on any
hardware.  On current machines with 4K pages support only, the kernel
will maintain 16 "subpages" for each 64K page transparently.

Note that while real 64K capable HW has been tested, the current patch
will not enable it yet as such hardware is not released yet, and I'm
still verifying with the firmware architects the proper to get the
information from the newer hypervisors.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>