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: Matt Evans <matt@ozlabs.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Commit a4fe3ce769 introduced a new
get_user_pages() path for hugepages on powerpc. Unfortunately, there
is a bug in it's loop logic, which can cause it to overrun the end of
the intended region. This came about by copying the logic from the
normal page path, which assumes the address and end parameters have
been pagesize aligned at the top-level. Since they're not *hugepage*
size aligned, the simplistic logic could step over the end of the gup
region without triggering the loop end condition.
This patch fixes the bug by using the technique that the normal page
path uses in levels above the lowest to truncate the ending address to
something we know we'll match with.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The hugepage arch code provides a number of hook functions/macros
which mirror the functionality of various normal page pte access
functions. Various changes in the normal page accessors (in
particular BenH's recent changes to the handling of lazy icache
flushing and PAGE_EXEC) have caused the hugepage versions to get out
of sync with the originals. In some cases, this is a bug, at least on
some MMU types.
One of the reasons that some hooks were not identical to the normal
page versions, is that the fact we're dealing with a hugepage needed
to be passed down do use the correct dcache-icache flush function.
This patch makes the main flush_dcache_icache_page() function hugepage
aware (by checking for the PageCompound flag). That in turn means we
can make set_huge_pte_at() just a call to set_pte_at() bringing it
back into sync. As a bonus, this lets us remove the
hash_huge_page_do_lazy_icache() function, replacing it with a call to
the hash_page_do_lazy_icache() function it was based on.
Some other hugepage pte access hooks - huge_ptep_get_and_clear() and
huge_ptep_clear_flush() - are not so easily unified, but this patch at
least brings them back into sync with the current versions of the
corresponding normal page functions.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This patch separates the parts of hugetlbpage.c which are inherently
specific to the hash MMU into a new hugelbpage-hash64.c file.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This patch simplifies the logic used to initialize hugepages on
powerpc. The somewhat oddly named set_huge_psize() is renamed to
add_huge_page_size() and now does all necessary verification of
whether it's given a valid hugepage sizes (instead of just some) and
instantiates the generic hstate structure (but no more).
hugetlbpage_init() now steps through the available pagesizes, checks
if they're valid for hugepages by calling add_huge_page_size() and
initializes the kmem_caches for the hugepage pagetables. This means
we can now eliminate the mmu_huge_psizes array, since we no longer
need to pass the sizing information for the pagetable caches from
set_huge_psize() into hugetlbpage_init()
Determination of the default huge page size is also moved from the
hash code into the general hugepage code.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Currently each available hugepage size uses a slightly different
pagetable layout: that is, the bottem level table of pointers to
hugepages is a different size, and may branch off from the normal page
tables at a different level. Every hugepage aware path that needs to
walk the pagetables must therefore look up the hugepage size from the
slice info first, and work out the correct way to walk the pagetables
accordingly. Future hardware is likely to add more possible hugepage
sizes, more layout options and more mess.
This patch, therefore reworks the handling of hugepage pagetables to
reduce this complexity. In the new scheme, instead of having to
consult the slice mask, pagetable walking code can check a flag in the
PGD/PUD/PMD entries to see where to branch off to hugepage pagetables,
and the entry also contains the information (eseentially hugepage
shift) necessary to then interpret that table without recourse to the
slice mask. This scheme can be extended neatly to handle multiple
levels of self-describing "special" hugepage pagetables, although for
now we assume only one level exists.
This approach means that only the pagetable allocation path needs to
know how the pagetables should be set out. All other (hugepage)
pagetable walking paths can just interpret the structure as they go.
There already was a flag bit in PGD/PUD/PMD entries for hugepage
directory pointers, but it was only used for debug. We alter that
flag bit to instead be a 0 in the MSB to indicate a hugepage pagetable
pointer (normally it would be 1 since the pointer lies in the linear
mapping). This means that asm pagetable walking can test for (and
punt on) hugepage pointers with the same test that checks for
unpopulated page directory entries (beq becomes bge), since hugepage
pointers will always be positive, and normal pointers always negative.
While we're at it, we get rid of the confusing (and grep defeating)
#defining of hugepte_shift to be the same thing as mmu_huge_psizes.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Currently we have a fair bit of rather fiddly code to manage the
various kmem_caches used to store page tables of various levels. We
generally have two caches holding some combination of PGD, PUD and PMD
tables, plus several more for the special hugepage pagetables.
This patch cleans this all up by taking a different approach. Rather
than the caches being designated as for PUDs or for hugeptes for 16M
pages, the caches are simply allocated to be a specific size. Thus
sharing of caches between different types/levels of pagetables happens
naturally. The pagetable size, where needed, is passed around encoded
in the same way as {PGD,PUD,PMD}_INDEX_SIZE; that is n where the
pagetable contains 2^n pointers.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Currently, hpte_need_flush() only correctly flushes the given address
for normal pages. Callers for hugepages are required to mask the
address themselves.
But hpte_need_flush() already looks up the page sizes for its own
reasons, so this is a rather silly imposition on the callers. This
patch alters it to mask based on the pagesize it has looked up itself,
and removes the awkward masking code in the hugepage caller.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This adds the PTE and pgtable format definitions, along with changes
to the kernel memory map and other definitions related to implementing
support for 64-bit Book3E. This also shields some asm-offset bits that
are currently only relevant on 32-bit
We also move the definition of the "linux" page size constants to
the common mmu.h file and add a few sizes that are relevant to
embedded processors.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
mm: Pass virtual address to [__]p{te,ud,md}_free_tlb()
Upcoming paches to support the new 64-bit "BookE" powerpc architecture
will need to have the virtual address corresponding to PTE page when
freeing it, due to the way the HW table walker works.
Basically, the TLB can be loaded with "large" pages that cover the whole
virtual space (well, sort-of, half of it actually) represented by a PTE
page, and which contain an "indirect" bit indicating that this TLB entry
RPN points to an array of PTEs from which the TLB can then create direct
entries. Thus, in order to invalidate those when PTE pages are deleted,
we need the virtual address to pass to tlbilx or tlbivax instructions.
The old trick of sticking it somewhere in the PTE page struct page sucks
too much, the address is almost readily available in all call sites and
almost everybody implemets these as macros, so we may as well add the
argument everywhere. I added it to the pmd and pud variants for consistency.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: David Howells <dhowells@redhat.com> [MN10300 & FRV]
Acked-by: Nick Piggin <npiggin@suse.de>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> [s390]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The KernelPageSize entry in /proc/pid/smaps is the pagesize used by the
kernel to back a VMA. This matches the size used by the MMU in the
majority of cases. However, one counter-example occurs on PPC64 kernels
whereby a kernel using 64K as a base pagesize may still use 4K pages for
the MMU on older processor. To distinguish, this patch reports
MMUPageSize as the pagesize used by the MMU in /proc/pid/smaps.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: "KOSAKI Motohiro" <kosaki.motohiro@jp.fujitsu.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It looks like most of the hugetlb code is doing the correct thing if
hugepages are not supported, but the mmap code is not. If we get into
the mmap code when hugepages are not supported, such as in an LPAR
which is running Active Memory Sharing, we can oops the kernel. This
fixes the oops being seen in this path.
oops: Kernel access of bad area, sig: 11 [#1]
SMP NR_CPUS=1024 NUMA pSeries
Modules linked in: nfs(N) lockd(N) nfs_acl(N) sunrpc(N) ipv6(N) fuse(N) loop(N)
dm_mod(N) sg(N) ibmveth(N) sd_mod(N) crc_t10dif(N) ibmvscsic(N)
scsi_transport_srp(N) scsi_tgt(N) scsi_mod(N)
Supported: No
NIP: c000000000038d60 LR: c00000000003945c CTR: c0000000000393f0
REGS: c000000077e7b830 TRAP: 0300 Tainted: G
(2.6.27.5-bz50170-2-ppc64)
MSR: 8000000000009032 <EE,ME,IR,DR> CR: 44000448 XER: 20000001
DAR: c000002000af90a8, DSISR: 0000000040000000
TASK = c00000007c1b8600[4019] 'hugemmap01' THREAD: c000000077e78000 CPU: 6
GPR00: 0000001fffffffe0 c000000077e7bab0 c0000000009a4e78 0000000000000000
GPR04: 0000000000010000 0000000000000001 00000000ffffffff 0000000000000001
GPR08: 0000000000000000 c000000000af90c8 0000000000000001 0000000000000000
GPR12: 000000000000003f c000000000a73880 0000000000000000 0000000000000000
GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000010000
GPR20: 0000000000000000 0000000000000003 0000000000010000 0000000000000001
GPR24: 0000000000000003 0000000000000000 0000000000000001 ffffffffffffffb5
GPR28: c000000077ca2e80 0000000000000000 c00000000092af78 0000000000010000
NIP [c000000000038d60] .slice_get_unmapped_area+0x6c/0x4e0
LR [c00000000003945c] .hugetlb_get_unmapped_area+0x6c/0x80
Call Trace:
[c000000077e7bbc0] [c00000000003945c] .hugetlb_get_unmapped_area+0x6c/0x80
[c000000077e7bc30] [c000000000107e30] .get_unmapped_area+0x64/0xd8
[c000000077e7bcb0] [c00000000010b140] .do_mmap_pgoff+0x140/0x420
[c000000077e7bd80] [c00000000000bf5c] .sys_mmap+0xc4/0x140
[c000000077e7be30] [c0000000000086b4] syscall_exit+0x0/0x40
Instruction dump:
fac1ffb0 fae1ffb8 fb01ffc0 fb21ffc8 fb41ffd0 fb61ffd8 fb81ffe0 fbc1fff0
fbe1fff8 f821fef1 f8c10158 f8e10160 <7d49002e> f9010168 e92d01b0 eb4902b0
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
called only from __init, calls __init. Incidentally, it ought to be static
in file.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andrew Morton suggested that using a macro that makes an array
reference look like a function call makes it harder to understand the
code.
This therefore removes the huge_pgtable_cache(psize) macro and
replaces its uses with pgtable_cache[HUGE_PGTABLE_INDEX(psize)].
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
There is a small bug in the handling of 16G hugepages recently added
to the kernel. This doesn't cause a crash or other user-visible
problems, but it does mean that more levels of pagetable are allocated
than makes sense for 16G pages. The hugepage pagetables for the 16G
pages are allocated much lower in the pagetable tree than they should
be, with the intervening levels allocated with full pmd and pud pages
which will only ever have one entry filled in.
This corrects this problem, at the same time cleaning up the handling
of which level 64k versus 16M hugepage pagetables are allocated at.
The new way of formatting the tests should be more robust against
changes in pagetable structure, or any newly added hugepage sizes.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The 64K SPU local store mapping feature is incompatible with the
64K huge pages support due to the inability of some parts of
the memory management to differenciate between them while they
use a different page table format.
For now, disable 64K huge pages when CONFIG_SPU_FS_64K_LS,
in the long run, this can be fixed by making this feature use
the hugetlb page table format.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Kmem cache passed to constructor is only needed for constructors that are
themselves multiplexeres. Nobody uses this "feature", nor does anybody uses
passed kmem cache in non-trivial way, so pass only pointer to object.
Non-trivial places are:
arch/powerpc/mm/init_64.c
arch/powerpc/mm/hugetlbpage.c
This is flag day, yes.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Jon Tollefson <kniht@linux.vnet.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Matt Mackall <mpm@selenic.com>
[akpm@linux-foundation.org: fix arch/powerpc/mm/hugetlbpage.c]
[akpm@linux-foundation.org: fix mm/slab.c]
[akpm@linux-foundation.org: fix ubifs]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Instead of using the variable mmu_huge_psize to keep track of the huge
page size we use an array of MMU_PAGE_* values. For each supported huge
page size we need to know the hugepte_shift value and have a
pgtable_cache. The hstate or an mmu_huge_psizes index is passed to
functions so that they know which huge page size they should use.
The hugepage sizes 16M and 64K are setup(if available on the hardware) so
that they don't have to be set on the boot cmd line in order to use them.
The number of 16G pages have to be specified at boot-time though (e.g.
hugepagesz=16G hugepages=5).
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The huge page size is defined for 16G pages. If a hugepagesz of 16G is
specified at boot-time then it becomes the huge page size instead of the
default 16M.
The change in pgtable-64K.h is to the macro pte_iterate_hashed_subpages to
make the increment to va (the 1 being shifted) be a long so that it is not
shifted to 0. Otherwise it would create an infinite loop when the shift
value is for a 16G page (when base page size is 64K).
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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: Adam Litke <agl@us.ibm.com>
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The 16G page locations have been saved during early boot in an array. The
alloc_bootmem_huge_page() function adds a page from here to the
huge_boot_pages list.
Acked-by: Adam Litke <agl@us.ibm.com>
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Straight forward extensions for huge pages located in the PUD instead of
PMDs.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The goal of this patchset is to support multiple hugetlb page sizes. This
is achieved by introducing a new struct hstate structure, which
encapsulates the important hugetlb state and constants (eg. huge page
size, number of huge pages currently allocated, etc).
The hstate structure is then passed around the code which requires these
fields, they will do the right thing regardless of the exact hstate they
are operating on.
This patch adds the hstate structure, with a single global instance of it
(default_hstate), and does the basic work of converting hugetlb to use the
hstate.
Future patches will add more hstate structures to allow for different
hugetlbfs mounts to have different page sizes.
[akpm@linux-foundation.org: coding-style fixes]
Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The double indirection here is not needed anywhere and hence (at least)
confusing.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: "David S. Miller" <davem@davemloft.net>
Acked-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Cc: Jon Tollefson <kniht@linux.vnet.ibm.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This frees a PTE bit when using 64K pages on ppc64. This is done
by getting rid of the separate _PAGE_HASHPTE bit. Instead, we just test
if any of the 16 sub-page bits is set. For non-combo pages (ie. real
64K pages), we set SUB0 and the location encoding in that field.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul 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: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Slab constructors currently have a flags parameter that is never used. And
the order of the arguments is opposite to other slab functions. The object
pointer is placed before the kmem_cache pointer.
Convert
ctor(void *object, struct kmem_cache *s, unsigned long flags)
to
ctor(struct kmem_cache *s, void *object)
throughout the kernel
[akpm@linux-foundation.org: coupla fixes]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.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: Paul Mackerras <paulus@samba.org>
This removes several duplicate includes from arch/powerpc/.
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Slab destructors were no longer supported after Christoph's
c59def9f22 change. They've been
BUGs for both slab and slub, and slob never supported them
either.
This rips out support for the dtor pointer from kmem_cache_create()
completely and fixes up every single callsite in the kernel (there were
about 224, not including the slab allocator definitions themselves,
or the documentation references).
Signed-off-by: Paul Mundt <lethal@linux-sh.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>
Remove includes of <linux/smp_lock.h> where it is not used/needed.
Suggested by Al Viro.
Builds cleanly on x86_64, i386, alpha, ia64, powerpc, sparc,
sparc64, and arm (all 59 defconfigs).
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The current get_unmapped_area code calls the f_ops->get_unmapped_area or the
arch one (via the mm) only when MAP_FIXED is not passed. That makes it
impossible for archs to impose proper constraints on regions of the virtual
address space. To work around that, get_unmapped_area() then calls some
hugetlbfs specific hacks.
This cause several problems, among others:
- It makes it impossible for a driver or filesystem to do the same thing
that hugetlbfs does (for example, to allow a driver to use larger page sizes
to map external hardware) if that requires applying a constraint on the
addresses (constraining that mapping in certain regions and other mappings
out of those regions).
- Some archs like arm, mips, sparc, sparc64, sh and sh64 already want
MAP_FIXED to be passed down in order to deal with aliasing issues. The code
is there to handle it... but is never called.
This series of patches moves the logic to handle MAP_FIXED down to the various
arch/driver get_unmapped_area() implementations, and then changes the generic
code to always call them. The hugetlbfs hacks then disappear from the generic
code.
Since I need to do some special 64K pages mappings for SPEs on cell, I need to
work around the first problem at least. I have further patches thus
implementing a "slices" layer that handles multiple page sizes through slices
of the address space for use by hugetlbfs, the SPE code, and possibly others,
but it requires that serie of patches first/
There is still a potential (but not practical) issue due to the fact that
filesystems/drivers implemeting g_u_a will effectively bypass all arch checks.
This is not an issue in practice as the only filesystems/drivers using that
hook are doing so for arch specific purposes in the first place.
There is also a problem with mremap that will completely bypass all arch
checks. I'll try to address that separately, I'm not 100% certain yet how,
possibly by making it not work when the vma has a file whose f_ops has a
get_unmapped_area callback, and by making it use is_hugepage_only_range()
before expanding into a new area.
Also, I want to turn is_hugepage_only_range() into a more generic
is_normal_page_range() as that's really what it will end up meaning when used
in stack grow, brk grow and mremap.
None of the above "issues" however are introduced by this patch, they are
already there, so I think the patch can go ini for 2.6.22.
This patch:
Handle MAP_FIXED in powerpc's arch_get_unmapped_area() in all 3
implementations of it.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: William Irwin <bill.irwin@oracle.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: David Howells <dhowells@redhat.com>
Cc: Andi Kleen <ak@suse.de>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Grant Grundler <grundler@parisc-linux.org>
Cc: Matthew Wilcox <willy@debian.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Adam Litke <agl@us.ibm.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It is not necessary to tell the slab allocators to align to a cacheline
if an explicit alignment was already specified. It is rather confusing
to specify multiple alignments.
Make sure that the call sites only use one form of alignment.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch was recently posted to lkml and acked by Pekka.
The flag SLAB_MUST_HWCACHE_ALIGN is
1. Never checked by SLAB at all.
2. A duplicate of SLAB_HWCACHE_ALIGN for SLUB
3. Fulfills the role of SLAB_HWCACHE_ALIGN for SLOB.
The only remaining use is in sparc64 and ppc64 and their use there
reflects some earlier role that the slab flag once may have had. If
its specified then SLAB_HWCACHE_ALIGN is also specified.
The flag is confusing, inconsistent and has no purpose.
Remove it.
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The current tlb flush code on powerpc 64 bits has a subtle race since we
lost the page table lock due to the possible faulting in of new PTEs
after a previous one has been removed but before the corresponding hash
entry has been evicted, which can leads to all sort of fatal problems.
This patch reworks the batch code completely. It doesn't use the mmu_gather
stuff anymore. Instead, we use the lazy mmu hooks that were added by the
paravirt code. They have the nice property that the enter/leave lazy mmu
mode pair is always fully contained by the PTE lock for a given range
of PTEs. Thus we can guarantee that all batches are flushed on a given
CPU before it drops that lock.
We also generalize batching for any PTE update that require a flush.
Batching is now enabled on a CPU by arch_enter_lazy_mmu_mode() and
disabled by arch_leave_lazy_mmu_mode(). The code epects that this is
always contained within a PTE lock section so no preemption can happen
and no PTE insertion in that range from another CPU. When batching
is enabled on a CPU, every PTE updates that need a hash flush will
use the batch for that flush.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
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>
This patch changes handling return value of ppc_md.hpte_insert() into
the same way as __hash_page_*().
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The powerpc specific version of hugetlb_get_unmapped_area() makes some
unwarranted assumptions about what checks have been made to its
parameters by its callers. This will lead to a BUG_ON() if a 32-bit
process attempts to make a hugepage mapping which extends above
TASK_SIZE (4GB).
I'm not sure if these assumptions came about because they were valid
with earlier versions of the get_unmapped_area() path, or if it was
always broken. Nonetheless this patch fixes the logic, and removes
the crash.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Replace all uses of kmem_cache_t with struct kmem_cache.
The patch was generated using the following script:
#!/bin/sh
#
# Replace one string by another in all the kernel sources.
#
set -e
for file in `find * -name "*.c" -o -name "*.h"|xargs grep -l $1`; do
quilt add $file
sed -e "1,\$s/$1/$2/g" $file >/tmp/$$
mv /tmp/$$ $file
quilt refresh
done
The script was run like this
sh replace kmem_cache_t "struct kmem_cache"
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Following up with the work on shared page table done by Dave McCracken. This
set of patch target shared page table for hugetlb memory only.
The shared page table is particular useful in the situation of large number of
independent processes sharing large shared memory segments. In the normal
page case, the amount of memory saved from process' page table is quite
significant. For hugetlb, the saving on page table memory is not the primary
objective (as hugetlb itself already cuts down page table overhead
significantly), instead, the purpose of using shared page table on hugetlb is
to allow faster TLB refill and smaller cache pollution upon TLB miss.
With PT sharing, pte entries are shared among hundreds of processes, the cache
consumption used by all the page table is smaller and in return, application
gets much higher cache hit ratio. One other effect is that cache hit ratio
with hardware page walker hitting on pte in cache will be higher and this
helps to reduce tlb miss latency. These two effects contribute to higher
application performance.
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: Dave McCracken <dmccr@us.ibm.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Gibson <david@gibson.dropbear.id.au>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
(David:)
If hugetlbfs_file_mmap() returns a failure to do_mmap_pgoff() - for example,
because the given file offset is not hugepage aligned - then do_mmap_pgoff
will go to the unmap_and_free_vma backout path.
But at this stage the vma hasn't been marked as hugepage, and the backout path
will call unmap_region() on it. That will eventually call down to the
non-hugepage version of unmap_page_range(). On ppc64, at least, that will
cause serious problems if there are any existing hugepage pagetable entries in
the vicinity - for example if there are any other hugepage mappings under the
same PUD. unmap_page_range() will trigger a bad_pud() on the hugepage pud
entries. I suspect this will also cause bad problems on ia64, though I don't
have a machine to test it on.
(Hugh:)
prepare_hugepage_range() should check file offset alignment when it checks
virtual address and length, to stop MAP_FIXED with a bad huge offset from
unmapping before it fails further down. PowerPC should apply the same
prepare_hugepage_range alignment checks as ia64 and all the others do.
Then none of the alignment checks in hugetlbfs_file_mmap are required (nor
is the check for too small a mapping); but even so, move up setting of
VM_HUGETLB and add a comment to warn of what David Gibson discovered - if
hugetlbfs_file_mmap fails before setting it, do_mmap_pgoff's unmap_region
when unwinding from error will go the non-huge way, which may cause bad
behaviour on architectures (powerpc and ia64) which segregate their huge
mappings into a separate region of the address space.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: "David S. Miller" <davem@davemloft.net>
Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Checking source for other get_paca()->field preemption dangers found that
open_high_hpage_areas does a structure copy into its paca while preemption
is enabled: unsafe however gcc accomplishes it. Just remove that copy:
it's done safely afterwards by on_each_cpu, as in open_low_hpage_areas.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
On Tue, 2006-08-15 at 08:22 -0700, Dave Hansen wrote:
> kernel BUG in cache_free_debugcheck at mm/slab.c:2748!
Alright, this one is only triggered when slab debugging is enabled. The
slabs are assumed to be aligned on a HUGEPTE_TABLE_SIZE boundary. The free
path makes use of this assumption and uses the lowest nibble to pass around
an index into an array of kmem_cache pointers. With slab debugging turned
on, the slab is still aligned, but the "working" object pointer is not.
This would break the assumption above that a full nibble is available for
the PGF_CACHENUM_MASK.
The following patch reduces PGF_CACHENUM_MASK to cover only the two least
significant bits, which is enough to cover the current number of 4 pgtable
cache types. Then use this constant to mask out the appropriate part of
the huge pte pointer.
Signed-off-by: Adam Litke <agl@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
At present, ARCH=powerpc kernels can waste considerable space in
pagetables when making large hugepage mappings. Hugepage PTEs go in
PMD pages, but each PMD page maps 256M and so contains only 16
hugepage PTEs (128 bytes of data), but takes up a 1024 byte
allocation. With CONFIG_PPC_64K_PAGES enabled (64k base page size),
the situation is worse. Now hugepage PTEs are at the PTE page level
(also mapping 256M), so we store 16 hugepage PTEs in a 64k allocation.
The PowerPC MMU already means that any 256M region is either all
hugepage, or all normal pages. Thus, with some care, we can use a
different allocation for the hugepage PTE tables and only allocate the
128 bytes necessary.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Quite a long time back, prepare_hugepage_range() replaced
is_aligned_hugepage_range() as the callback from mm/mmap.c to arch code to
verify if an address range is suitable for a hugepage mapping.
is_aligned_hugepage_range() stuck around, but only to implement
prepare_hugepage_range() on archs which didn't implement their own.
Most archs (everything except ia64 and powerpc) used the same
implementation of is_aligned_hugepage_range(). On powerpc, which
implements its own prepare_hugepage_range(), the custom version was never
used.
In addition, "is_aligned_hugepage_range()" was a bad name, because it
suggests it returns true iff the given range is a good hugepage range,
whereas in fact it returns 0-or-error (so the sense is reversed).
This patch cleans up by abolishing is_aligned_hugepage_range(). Instead
prepare_hugepage_range() is defined directly. Most archs use the default
version, which simply checks the given region is aligned to the size of a
hugepage. ia64 and powerpc define custom versions. The ia64 one simply
checks that the range is in the correct address space region in addition to
being suitably aligned. The powerpc version (just as previously) checks
for suitable addresses, and if necessary performs low-level MMU frobbing to
set up new areas for use by hugepages.
No libhugetlbfs testsuite regressions on ppc64 (POWER5 LPAR).
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
