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: Michael Neuling <mikey@neuling.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The config symbol for SPE support is called CONFIG_SPU_BASE, not
CONFIG_SPE_BASE.
Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul 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: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Fix up comment on two #endifs to match their #ifs.
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
----
hash_utils_64.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
Signed-off-by: Paul 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: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul 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: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul 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: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul 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: Benjamin 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: Paul 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: Paul 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: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Remove CPU_FTR_16M_PAGE from the cupfeatures mask at runtime on iSeries.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
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: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul 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: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul 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: Paul Mackerras <paulus@samba.org>
This removes statically assigned platform numbers and reworks the
powerpc platform probe code to use a better mechanism. With this,
board support files can simply declare a new machine type with a
macro, and implement a probe() function that uses the flattened
device-tree to detect if they apply for a given machine.
We now have a machine_is() macro that replaces the comparisons of
_machine with the various PLATFORM_* constants. This commit also
changes various drivers to use the new macro instead of looking at
_machine.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
It has been decreed that platform numbers are evil, so as a step in that
direction, replace platform_is_lpar() with a FW_FEATURE_LPAR bit.
Currently FW_FEATURE_LPAR really means i/pSeries LPAR, in the future we might
have to clean that up if we need to be more specific about what LPAR actually
means. But that's another patch ...
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
htab_bolt_mapping() takes a vstart and pstart parameter, but all but one of
its callers actually pass it vstart and vstart. Luckily before it passes
paddr (calculated from paddr) to the hpte_insert routines it calls
virt_to_abs() (aka. __pa()) on the address, so there isn't actually a bug.
map_io_page() however does pass pstart properly, so currently it's broken
AFAICT because we're calling __pa(paddr) which will get us something very
large. Presumably no one's calling map_io_page() in the right context.
Anyway, change htab_bolt_mapping() callers to properly pass pstart, and then
use it properly in htab_bolt_mapping(), ie. don't call __pa() on it again.
Booted on p5 LPAR, iSeries and Power3.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Before the merge I updated create_pte_mapping() to work for iSeries, by
calling iSeries_hpte_bolt_or_insert. (4c55130b2a)
Later we changed iSeries_hpte_insert to cope with the bolting case, and called
that instead from create_pte_mapping() (which was renamed to htab_bolt_mapping)
(3c726f8dee).
Unfortunately that change introduced a subtle bug, where we pass an absolute
address to iSeries_hpte_insert() where it expects a physical address. This
leads to us calling phys_to_abs() twice on the physical address, which is
seriously bogus.
This only causes a problem if the absolute address from the first translation
can be looked up again in the chunk_map, which depends on the size and layout
of memory. I've seen it fail on one box, but not others.
The minimal fix is to pass the physical address to iSeries_hpte_insert(). For
2.6.17 we should make phys_to_abs() BUG if we try to double-translate an
address.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
For kexec we need to know the size of the MMU hash table.
Currently we calculate the size once in the htab code, and then twice more in
the kexec code, once using htab_hash_mask and once using ppc64_pft_size.
On some machines the ppc64_pft_size calculation is broken because
ppc64_pft_size is not set.
So we need to fix the second calculation, but better still we should just
calculate the size once and use it everywhere else.
Tested on Power5 LPAR, Power4 non-LPAR and Power3.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Currently most callers of lmb_alloc() don't check if it worked or not, if it
ever does weird bad things will probably happen. The few callers who do check
just panic or BUG_ON.
So make lmb_alloc() panic internally, to catch bugs at the source. The few
callers who did check the result no longer need to.
The only caller that did anything interesting with the return result was
careful_allocation(). For it we create __lmb_alloc_base() which _doesn't_ panic
automatically, a little messy, but passable.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul 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: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This is the current version of the spu file system, used
for driving SPEs on the Cell Broadband Engine.
This release is almost identical to the version for the
2.6.14 kernel posted earlier, which is available as part
of the Cell BE Linux distribution from
http://www.bsc.es/projects/deepcomputing/linuxoncell/.
The first patch provides all the interfaces for running
spu application, but does not have any support for
debugging SPU tasks or for scheduling. Both these
functionalities are added in the subsequent patches.
See Documentation/filesystems/spufs.txt on how to use
spufs.
Signed-off-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Sonny has noticed hotplug CPU on ppc64 is broken in 2.6.15-*. One of the
problems is that htab_initialize_secondary is called when a cpu is being
brought up, but it is marked __init.
Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
On most powerpc CPUs, the dcache and icache are not coherent so
between writing and executing a page, the caches must be flushed.
Userspace programs assume pages given to them by the kernel are icache
clean, so we must do this flush between the kernel clearing a page and
it being mapped into userspace for execute. We were not doing this
for hugepages, this patch corrects the situation.
We use the same lazy mechanism as we use for normal pages, delaying
the flush until userspace actually attempts to execute from the page
in question.
Tested on G5.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch merges platform codes. systemcfg->platform is no longer used,
systemcfg use in general is deprecated as much as possible (and renamed
_systemcfg before it gets completely moved elsewhere in a future patch),
_machine is now used on ppc64 along as ppc32. Platform codes aren't gone
yet but we are getting a step closer. A bunch of asm code in head[_64].S
is also turned into C code.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Add the create_section_mapping() routine to create hptes for memory
sections dynamically added after system boot.
Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
For some stupid reason I can't explain (brown paper bag is at hand), I
removed the check pfn_valid() in the code that does the icache/dcache
coherency on POWER4 and later. That causes us to eventually try to
access non existing struct page when hashing in IO pages.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul 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
c594adad56) 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.diff
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The ancient ppcdebug/PPCDBG mechanism is now only used in two places.
First, in the hash setup code, one of the bits allows the size of the
hash table to be reduced by a factor of 8 - which would be better
accomplished with a command line option for that purpose. The other
was a bunch of bus walking related messages in the iSeries code, which
would seem to be insufficient reason to keep the mechanism.
This patch removes the last traces of this mechanism.
Built and booted on iSeries and pSeries POWER5 LPAR (ARCH=powerpc).
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.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: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We weren't computing the size of the hash table correctly on iSeries
because the relevant code in prom.c was #ifdef CONFIG_PPC_PSERIES.
This moves the code to hash_utils_64.c, makes it unconditional, and
cleans it up a bit.
Signed-off-by: Paul Mackerras <paulus@samba.org>
This moves the remaining files in arch/ppc64/mm to arch/powerpc/mm,
and arranges that we use them when compiling with ARCH=ppc64.
Signed-off-by: Paul Mackerras <paulus@samba.org>
