The 16G huge pages have to be reserved in the HMC prior to boot.  The
location of the pages are placed in the device tree.  This patch adds code
to scan the device tree during very early boot and save these page
locations until hugetlbfs is ready for them.
Acked-by: NAdam Litke <agl@us.ibm.com>
Signed-off-by: NJon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

At present, if we have a kernel with a 64kB page size, and some
process maps something that has to be mapped with 4kB pages (such as a
cache-inhibited mapping on POWER5+, or the eHCA infiniband queue-pair
pages), we change the process to use 4kB pages everywhere.  This hurts
the performance of HPC programs that access eHCA from userspace.

With this patch, the kernel will only demote the slice(s) containing
the eHCA or cache-inhibited mappings, leaving the remaining slices
able to use 64kB hardware pages.

This also changes the slice_get_unmapped_area code so that it is
willing to place a 64k-page mapping into (or across) a 4k-page slice
if there is no better alternative, i.e. if the program specified
MAP_FIXED or if there is not sufficient space available in slices that
are either empty or already have 64k-page mappings in them.
Signed-off-by: NPaul Mackerras <paulus@samba.org>
Acked-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>

... instead of having extern declarations in a .c file.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Make two vmemmap helpers static in init_64.c
Make stab variables static in stab.c
Make psize defs static in hash_utils_64.c
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This eliminates a warning in builds that don't define
CONFIG_MEMORY_HOTPLUG.
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

If the platform doesn't support hpte_removebolted(), gracefully
return failure rather than success.
Signed-off-by: NBadari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

On pSeries, the hypervisor doesn't let us map in the eHEA ethernet
adapter using 64k pages, and thus the ehea driver will fail if 64k
pages are configured.  This works around the problem by always
using 4k pages for ioremap on pSeries (but not on other platforms).
A better fix would be to check whether the partition could ever
have an eHEA adapter, and only force 4k pages if it could, but this
will do for 2.6.25.

This is based on an earlier patch by Tony Breeds.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

My recent hack to allocate the hash table under 1GB on cell was poorly
tested, *cough*. It turns out on blades with large amounts of memory we
fail to allocate the hash table at all. This is because RTAS has been
instantiated just below 768MB, and 0-x MB are used by the kernel,
leaving no areas that are both large enough and also naturally-aligned.

For the cell IOMMU hack the page tables must be under 2GB, so use that
as the limit instead. This has been tested on real hardware and boots
happily.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

For memory remove, we need to clean up htab mappings for the
section of the memory we are removing.

This implements support for removing htab bolted mappings for pSeries
logical partitions.  Other sub-archs may need to implement similar
functionality for hotplug memory remove to work on them.
Signed-off-by: NBadari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In order to support the fixed IOMMU mapping (in a subsequent patch),
we need the hash table to be inside the IOMMUs DMA window.  This is
usually 2G, but let's make sure the hash table is under 1G as that
will satisfy the IOMMU requirements and also means the hash table will
be on node 0.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Using 64k pages on 64-bit PowerPC systems makes life difficult for
emulators that are trying to emulate an ISA, such as x86, which use a
smaller page size, since the emulator can no longer use the MMU and
the normal system calls for controlling page protections.  Of course,
the emulator can emulate the MMU by checking and possibly remapping
the address for each memory access in software, but that is pretty
slow.

This provides a facility for such programs to control the access
permissions on individual 4k sub-pages of 64k pages.  The idea is
that the emulator supplies an array of protection masks to apply to a
specified range of virtual addresses.  These masks are applied at the
level where hardware PTEs are inserted into the hardware page table
based on the Linux PTEs, so the Linux PTEs are not affected.  Note
that this new mechanism does not allow any access that would otherwise
be prohibited; it can only prohibit accesses that would otherwise be
allowed.  This new facility is only available on 64-bit PowerPC and
only when the kernel is configured for 64k pages.

The masks are supplied using a new subpage_prot system call, which
takes a starting virtual address and length, and a pointer to an array
of protection masks in memory.  The array has a 32-bit word per 64k
page to be protected; each 32-bit word consists of 16 2-bit fields,
for which 0 allows any access (that is otherwise allowed), 1 prevents
write accesses, and 2 or 3 prevent any access.

Implicit in this is that the regions of the address space that are
protected are switched to use 4k hardware pages rather than 64k
hardware pages (on machines with hardware 64k page support).  In fact
the whole process is switched to use 4k hardware pages when the
subpage_prot system call is used, but this could be improved in future
to switch only the affected segments.

The subpage protection bits are stored in a 3 level tree akin to the
page table tree.  The top level of this tree is stored in a structure
that is appended to the top level of the page table tree, i.e., the
pgd array.  Since it will often only be 32-bit addresses (below 4GB)
that are protected, the pointers to the first four bottom level pages
are also stored in this structure (each bottom level page contains the
protection bits for 1GB of address space), so the protection bits for
addresses below 4GB can be accessed with one fewer loads than those
for higher addresses.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This adds the hugepagesz boot-time parameter for ppc64.  It lets one
pick the size for huge pages.  The choices available are 64K and 16M
when the base page size is 4k.  It defaults to 16M (previously the
only only choice) if nothing or an invalid choice is specified.

Tested 64K huge pages successfully with the libhugetlbfs 1.2.
Signed-off-by: NJon Tollefson <kniht@linux.vnet.ibm.com>
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 fixes the error
	error: implicit declaration of function "udbg_printf"

We have a few spots where we reference udbg_printf() without #including
udbg.h.  These are within #ifdef DEBUG blocks, so unnoticed until we do
a #define DEBUG or #define DEBUG_LOW nearby.
Signed-off-by: NWill Schmidt <will_schmidt@vnet.ibm.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When demoting a process to use 4K HW pages (instead of 64K), which
happens under various circumstances such as doing cache inhibited
mappings on machines that do not support 64K CI pages, the assembly
hash code calls back into the C function flush_hash_page().  This
function prototype was recently changed to accomodate for 1T segments
but the assembly call site was not updated, causing applications that
do demotion to hang.  In addition, when updating the per-CPU PACA for
the new sizes, we didn't properly update the slice "map", thus causing
the SLB miss code to re-insert segments for the wrong size.

This fixes both and adds a warning comment next to the C
implementation to try to avoid problems next time someone changes it.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

PA6T has a bug where the slbie instruction does not honor the large
segment bit.  As a result, we have to always use slbia when switching
context.

We don't have to worry about changing the slbie's during fault processing,
since they should never be replacing one VSID with another using the
same ESID.  I.e. there's no risk for inserting duplicate entries due to a
failed slbie of the old entry.  So as long as we clear it out on context
switch we should be fine.
Signed-off-by: NOlof Johansson <olof@lixom.net>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Buglet in the 1TB detection makes it return after checking the first
property word, even if it's not a match.
Signed-off-by: NOlof Johansson <olof@lixom.net>
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>

This removes some of the #ifdefs from .c files.
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This removes several duplicate includes from arch/powerpc/.
Signed-off-by: NJesper Juhl <jesper.juhl@gmail.com>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

A similar fix to netfilter from Eric Dumazet inspired me to
look around a bit by using some grep/sed stuff as looking for
this kind of bugs seemed easy to automate.  This is one of them
I found where it looks like this semicolon is not valid.
Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

On a machine with hardware 64kB pages and a kernel configured for a
64kB base page size, we need to change the vmalloc segment from 64kB
pages to 4kB pages if some driver creates a non-cacheable mapping in
the vmalloc area.  However, we never updated with SLB shadow buffer.
This fixes it.  Thanks to paulus for finding this.

Also added some write barriers to ensure the shadow buffer contents
are always consistent.
Signed-off-by: NMichael Neuling <mikey@neuling.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The config symbol for SPE support is called CONFIG_SPU_BASE, not
CONFIG_SPE_BASE.
Signed-off-by: NGeert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Using typedefs to rename structure types if frowned on by CodingStyle.
However, we do so for the hash PTE structure on both ppc32 (where it's
called "PTE") and ppc64 (where it's called "hpte_t").  On ppc32 we
also have such a typedef for the BATs ("BAT").

This removes this unhelpful use of typedefs, in the process
bringing ppc32 and ppc64 closer together, by using the name "struct
hash_pte" in both cases.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Fix up comment on two #endifs to match their #ifs.
Signed-off-by: NJon Tollefson <kniht@linux.vnet.ibm.com>
----

 hash_utils_64.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This adds the ability for a kernel compiled with 4K page size
to have special slices containing 64K pages and hash the right type
of hash PTEs.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
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>

The code for demoting segments to 4K had some issues, like for example,
when using _PAGE_4K_PFN flag, the first CPU to hit it would do the
demotion, but other CPUs hitting the same page wouldn't properly flush
their SLBs if mmu_ci_restriction isn't set.  There are also potential
issues with hash_preload not handling _PAGE_4K_PFN.  All of these are
non issues on current hardware but might bite us in the future.

This patch thus fixes it by:

 - Taking the test comparing the mm and current CPU context page
sizes to decide to flush SLBs out of the mmu_ci_restrictions test
since that can also be triggered by _PAGE_4K_PFN pages

 - Due to the above being done all the time, demote_segment_4k
doesn't need update the context and flush the SLB

 - demote_segment_4k can be static and doesn't need an EXPORT_SYMBOL

 - Making hash_preload ignore anything that has either _PAGE_4K_PFN
or _PAGE_NO_CACHE set, thus avoiding duplication of the complicated
logic in hash_page() (and possibly making hash_preload a little bit
faster for the normal case).
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Fixes:

BUG: spinlock bad magic on CPU#0, swapper/0
 lock: c00000000064ec30, .magic: 00000000, .owner: <none>/-1, .owner_cpu: 0
Call Trace:
[c00000000062b980] [c00000000000f920] .show_stack+0x6c/0x1a0 (unreliable)
[c00000000062ba20] [c0000000001c2b40] .spin_bug+0xb0/0xd4
[c00000000062bab0] [c0000000001c2ed0] ._raw_spin_lock+0x44/0x184
[c00000000062bb50] [c0000000003a42b4] ._spin_lock+0x10/0x24
[c00000000062bbd0] [c00000000002b4dc] .kernel_map_pages+0x198/0x278
[c00000000062bc90] [c000000000079720] .free_hot_cold_page+0x124/0x418
[c00000000062bd70] [c000000000530278] .free_all_bootmem_core+0x14c/0x224
[c00000000062be50] [c00000000052a178] .mem_init+0x68/0x170
[c00000000062bee0] [c00000000051d874] .start_kernel+0x2a0/0x37c
[c00000000062bf90] [c0000000000084c8] .start_here_common+0x54/0x8c
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Here's an implementation of DEBUG_PAGEALLOC for 64 bits powerpc.
It applies on top of the 32 bits patch.

Unlike Anton's previous attempt, I'm not using updatepp. I'm removing
the hash entries from the bolted mapping (using a map in RAM of all the
slots). Expensive but it doesn't really matter, does it ? :-)

Memory hot-added doesn't benefit from this unless it's added at an
address that is below end_of_DRAM() as calculated at boot time.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

 arch/powerpc/Kconfig.debug      |    2
 arch/powerpc/mm/hash_utils_64.c |   84 ++++++++++++++++++++++++++++++++++++++--
 2 files changed, 82 insertions(+), 4 deletions(-)
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Some drivers have resources that they want to be able to map into
userspace that are 4k in size.  On a kernel configured with 64k pages
we currently end up mapping the 4k we want plus another 60k of
physical address space, which could contain anything.  This can
introduce security problems, for example in the case of an infiniband
adaptor where the other 60k could contain registers that some other
program is using for its communications.

This patch adds a new function, remap_4k_pfn, which drivers can use to
map a single 4k page to userspace regardless of whether the kernel is
using a 4k or a 64k page size.  Like remap_pfn_range, it would
typically be called in a driver's mmap function.  It only maps a
single 4k page, which on a 64k page kernel appears replicated 16 times
throughout a 64k page.  On a 4k page kernel it reduces to a call to
remap_pfn_range.

The way this works on a 64k kernel is that a new bit, _PAGE_4K_PFN,
gets set on the linux PTE.  This alters the way that __hash_page_4K
computes the real address to put in the HPTE.  The RPN field of the
linux PTE becomes the 4k RPN directly rather than being interpreted as
a 64k RPN.  Since the RPN field is 32 bits, this means that physical
addresses being mapped with remap_4k_pfn have to be below 2^44,
i.e. 0x100000000000.

The patch also factors out the code in arch/powerpc/mm/hash_utils_64.c
that deals with demoting a process to use 4k pages into one function
that gets called in the various different places where we need to do
that.  There were some discrepancies between exactly what was done in
the various places, such as a call to spu_flush_all_slbs in one case
but not in others.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The SPU code doesn't properly invalidate SPUs SLBs when necessary,
for example when changing a segment size from the hugetlbfs code. In
addition, it saves and restores the SLB content on context switches
which makes it harder to properly handle those invalidations.

This patch removes the saving & restoring for now, something more
efficient might be found later on. It also adds a spu_flush_all_slbs(mm)
that can be used by the core mm code to flush the SLBs of all SPEs that
are running a given mm at the time of the flush.

In order to do that, it adds a spinlock to the list of all SPEs and move
some bits & pieces from spufs to spu_base.c
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Remove CPU_FTR_16M_PAGE from the cupfeatures mask at runtime on iSeries.
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

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

With the ppc_md htab pointers setup earlier, we can use ppc_md.hpte_insert
in htab_bolt_mapping(), rather than deciding which version to call by hand.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Initialise the ppc_md htab callbacks earlier, in the probe routines. This
allows us to call htab_finish_init() from htab_initialize(), and makes it
private to hash_utils_64.c. Move htab_finish_init() and make_bl() above
htab_initialize() to avoid forward declarations.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

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>

No need to always print page orders.
Signed-off-by: NOlof Johansson <olof@lixom.net>
Signed-off-by: NPaul Mackerras <paulus@samba.org>