tlb_finish_mmu used to batch zap_pte_range's update of mm rss, which may be
worthwhile if the mm is contended, and would reduce atomic operations if the
counts were atomic.  Let zap_pte_range now batch its updates to file_rss and
anon_rss, per page-table in case we drop the lock outside; and copy_pte_range
batch them too.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I was lazy when we added anon_rss, and chose to change as few places as
possible.  So currently each anonymous page has to be counted twice, in rss
and in anon_rss.  Which won't be so good if those are atomic counts in some
configurations.

Change that around: keep file_rss and anon_rss separately, and add them
together (with get_mm_rss macro) when the total is needed - reading two
atomics is much cheaper than updating two atomics.  And update anon_rss
upfront, typically in memory.c, not tucked away in page_add_anon_rmap.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

zap_pte_range has been counting the pages it frees in tlb->freed, then
tlb_finish_mmu has used that to update the mm's rss.  That got stranger when I
added anon_rss, yet updated it by a different route; and stranger when rss and
anon_rss became mm_counters with special access macros.  And it would no
longer be viable if we're relying on page_table_lock to stabilize the
mm_counter, but calling tlb_finish_mmu outside that lock.

Remove the mmu_gather's freed field, let tlb_finish_mmu stick to its own
business, just decrement the rss mm_counter in zap_pte_range (yes, there was
some point to batching the update, and a subsequent patch restores that).  And
forget the anal paranoia of first reading the counter to avoid going negative
- if rss does go negative, just fix that bug.

Remove the mmu_gather's flushes and avoided_flushes from arm and arm26: no use
was being made of them.  But arm26 alone was actually using the freed, in the
way some others use need_flush: give it a need_flush.  arm26 seems to prefer
spaces to tabs here: respect that.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

tlb_is_full_mm?  What does that mean?  The TLB is full?  No, it means that the
mm's last user has gone and the whole mm is being torn down.  And it's an
inline function because sparc64 uses a different (slightly better)
"tlb_frozen" name for the flag others call "fullmm".

And now the ptep_get_and_clear_full macro used in zap_pte_range refers
directly to tlb->fullmm, which would be wrong for sparc64.  Rather than
correct that, I'd prefer to scrap tlb_is_full_mm altogether, and change
sparc64 to just use the same poor name as everyone else - is that okay?
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Speeding up mremap's moving of ptes has never been a priority, but the locking
will get more complicated shortly, and is already too baroque.

Scrap the current one-by-one moving, do an extent at a time: curtailed by end
of src and dst pmds (have to use PMD_SIZE: the way pmd_addr_end gets elided
doesn't match this usage), and by latency considerations.

One nice property of the old method is lost: it never allocated a page table
unless absolutely necessary, so you could free empty page tables by mremapping
to and fro.  Whereas this way, it allocates a dst table wherever there was a
src table.  I keep diving in to reinstate the old behaviour, then come out
preferring not to clutter how it now is.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Impose a little more consistency on the page fault handlers do_wp_page,
do_swap_page, do_anonymous_page, do_no_page, do_file_page: why not pass their
arguments in the same order, called the same names?

break_cow is all very well, but what it did was inlined elsewhere: easier to
compare if it's brought back into do_wp_page.

do_file_page's fallback to do_no_page dates from a time when we were testing
pte_file by using it wherever possible: currently it's peculiar to nonlinear
vmas, so just check that.  BUG_ON if not?  Better not, it's probably page
table corruption, so just show the pte: hmm, there's a pte_ERROR macro, let's
use that for do_wp_page's invalid pfn too.

Hah!  Someone in the ppc64 world noticed pte_ERROR was unused so removed it:
restored (and say "pud" not "pmd" in its pud_ERROR).
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

exit_mmap resets various mm_struct fields, but the mm is well on its way out,
and none of those fields matter by this point.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Divide remove_vm_struct into two parts: first anon_vma_unlink plus
unlink_file_vma, to unlink the vma from the list and tree by which rmap or
vmtruncate might find it; then remove_vma to close, fput and free.

The intention here is to do the anon_vma_unlink and unlink_file_vma earlier,
in free_pgtables before freeing any page tables: so we can be sure that any
page tables traversed by rmap and vmtruncate are stable (and other, ordinary
cases are stabilized by holding mmap_sem).

This will be crucial to traversing pgd,pud,pmd without page_table_lock.  But
testing the split-out patch showed that lifting the page_table_lock is
symbiotically necessary to make this change - the lock ordering is wrong to
move those unlinks into free_pgtables while it's under ptlock.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

unmap_vma doesn't amount to much, let's put it inside unmap_vma_list.  Except
it doesn't unmap anything, unmap_region just did the unmapping: rename it to
remove_vma_list.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The original vm_stat_account has fallen into disuse, with only one user, and
only one user of vm_stat_unaccount.  It's easier to keep track if we convert
them all to __vm_stat_account, then free it from its __shackles.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

do_anonymous_page's pte_wrprotect causes some confusion: in such a case,
vm_page_prot must already be forcing COW, so must omit write permission, and
so the pte_wrprotect is redundant.  Replace it by a comment to that effect,
and reword the comment on unuse_pte which also caused confusion.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

zap_pte_range already avoids wasting time to mark_page_accessed on anon pages:
it can also skip anon set_page_dirty - the page only needs to be marked dirty
if shared with another mm, but that will say pte_dirty too.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Use latency breaking in msync_pte_range like that in copy_pte_range, instead
of the ugly CONFIG_PREEMPT filemap_msync alternatives.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

My latency breaking in copy_pte_range didn't work as intended: instead of
checking at regularish intervals, after the first interval it checked every
time around the loop, too impatient to be preempted.  Fix that.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch adds some stack dumps if the slab logic is processing slab
blocks from the wrong node.  This is necessary in order to detect
situations as encountered by Petr.
Signed-off-by: NChristoph Lameter <clameter@sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Martin Hicks' page cache reclaim patch added the 'may_swap' flag to the
scan_control struct; and modified shrink_list() not to add anon pages to
the swap cache if may_swap is not asserted.

Ref:  http://marc.theaimsgroup.com/?l=linux-mm&m=111461480725322&w=4

However, further down, if the page is mapped, shrink_list() calls
try_to_unmap() which will call try_to_unmap_one() via try_to_unmap_anon ().
 try_to_unmap_one() will BUG_ON() an anon page that is NOT in the swap
cache.  Martin says he never encountered this path in his testing, but
agrees that it might happen.

This patch modifies shrink_list() to skip anon pages that are not already
in the swap cache when !may_swap, rather than just not adding them to the
cache.
Signed-off-by: NLee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This is not problem actually, but sync_page_range() is using for exported
function to filesystems.

The msync_xxx is more readable at least to me.
Signed-off-by: NOGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Acked-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Most of them can never be triggered and were only for development.
Signed-off-by: N"Andi Kleen" <ak@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The NUMA policy code predated nodemask_t so it used open coded bitmaps.
Convert everything to nodemask_t.  Big patch, but shouldn't have any actual
behaviour changes (except I removed one unnecessary check against
node_online_map and one unnecessary BUG_ON)
Signed-off-by: N"Andi Kleen" <ak@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Set the low water mark for hot pages in pcp to zero.

(akpm: for the life of me I cannot remember why we created pcp->low.  Neither
can Martin and the changelog is silent.  Maybe it was just a brainfart, but I
have this feeling that there was a reason.  If not, we should remove the
fields completely.  We'll see.)
Signed-off-by: NRohit Seth <rohit.seth@intel.com>
Cc: <linux-mm@kvack.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Increase the page allocator's per-cpu magazines from 1/4MB to 1/2MB.

Over 100+ runs for a workload, the difference in mean is about 2%.  The best
results for both are almost same.  Though the max variation in results with
1/2MB is only 2.2%, whereas with 1/4MB it is 12%.
Signed-off-by: NRohit Seth <rohit.seth@intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

It turns out that the original swap token implementation, by Song Jiang, only
enforced the swap token while the task holding the token is handling a page
fault.  This patch approximates that, without adding an additional flag to the
mm_struct, by checking whether the mm->mmap_sem is held for reading, like the
page fault code does.

This patch has the effect of automatically, and gradually, disabling the
enforcement of the swap token when there is little or no paging going on, and
"turning up" the intensity of the swap token code the more the task holding
the token is thrashing.

Thanks to Song Jiang for pointing out this aspect of the token based thrashing
control concept.

The new code shows a slight degradation over the old swap token code, but
still a big win over running without the swap token.

2.6.12+ swap token disabled

$ for i in `seq 10` ; do /usr/bin/time ./qsbench -n 30000000 -p 3 ; done
101.74user 23.13system 8:26.91elapsed 24%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (38597major+430315minor)pagefaults 0swaps
101.98user 24.91system 8:03.06elapsed 26%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (33939major+430457minor)pagefaults 0swaps
101.93user 22.12system 7:34.90elapsed 27%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (33166major+421267minor)pagefaults 0swaps
101.82user 22.38system 8:31.40elapsed 24%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (39338major+433262minor)pagefaults 0swaps

2.6.12+ swap token enabled, timeout 300 seconds

$ for i in `seq 4` ; do /usr/bin/time ./qsbench -n 30000000 -p 3 ; done
102.58user 16.08system 3:41.44elapsed 53%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (19707major+285786minor)pagefaults 0swaps
102.07user 19.56system 4:00.64elapsed 50%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (19012major+299259minor)pagefaults 0swaps
102.64user 18.25system 4:07.31elapsed 48%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (21990major+304831minor)pagefaults 0swaps
101.39user 19.41system 5:15.81elapsed 38%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (24850major+323321minor)pagefaults 0swaps

2.6.12+ with new swap token code, timeout 300 seconds

$ for i in `seq 4` ; do /usr/bin/time ./qsbench -n 30000000 -p 3 ; done
101.87user 24.66system 5:53.20elapsed 35%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (26848major+363497minor)pagefaults 0swaps
102.83user 19.95system 4:17.25elapsed 47%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (19946major+305722minor)pagefaults 0swaps
102.09user 19.46system 5:12.57elapsed 38%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (25461major+334994minor)pagefaults 0swaps
101.67user 20.61system 4:52.97elapsed 41%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (22190major+329508minor)pagefaults 0swaps
Signed-off-by: NRik Van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch adds

vmalloc_node(size, node)	-> Allocate necessary memory on the specified node

and

get_vm_area_node(size, flags, node)

and the other functions that it depends on.
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

zone handling, mapping->flags handling
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Beginning of gfp_t annotations:

 - -Wbitwise added to CHECKFLAGS
 - old __bitwise renamed to __bitwise__
 - __bitwise defined to either __bitwise__ or nothing, depending on
   __CHECK_ENDIAN__ being defined
 - gfp_t switched from __nocast to __bitwise__
 - force cast to gfp_t added to __GFP_... constants
 - new helper - gfp_zone(); extracts zone bits out of gfp_t value and casts
   the result to int
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The NUMA counters in struct per_cpu_pageset (linux/mmzone.h) are never
cleared today.  This works ok for CPU 0 on NUMA machines because
boot_pageset[] is already zero, but for other CPU:s this results in
uninitialized counters.
Signed-off-by: NMagnus Damm <magnus@valinux.co.jp>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This reverts commit 3359b54c and
replaces it with a cleaner version that is purely based on page table
operations, so that the synchronization between inode size and hugetlb
mappings becomes moot.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This introduces a limit parameter to the core bootmem allocator; The new
parameter indicates that physical memory allocated by the bootmem
allocator should be within the requested limit.

We also introduce alloc_bootmem_low_pages_limit, alloc_bootmem_node_limit,
alloc_bootmem_low_pages_node_limit apis, but alloc_bootmem_low_pages_limit
is the only api used for swiotlb.

The existing alloc_bootmem_low_pages() api could instead have been
changed and made to pass right limit to the core allocator.  But that
would make the patch more intrusive for 2.6.14, as other arches use
alloc_bootmem_low_pages().  We may be done that post 2.6.14 as a
cleanup.

With this, swiotlb gets memory within 4G for both x86_64 and ia64
arches.
Signed-off-by: NYasunori Goto <y-goto@jp.fujitsu.com>
Cc: Ravikiran G Thirumalai <kiran@scalex86.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

hugetlbfs allows truncation of its files (should it?), but hugetlb.c often
forgets that: crashes and misaccounting ensue.

copy_hugetlb_page_range better grab the src page_table_lock since we don't
want to guess what happens if concurrently truncated.  unmap_hugepage_range
rss accounting must not assume the full range was mapped.  follow_hugetlb_page
must guard with page_table_lock and be prepared to exit early.

Restyle copy_hugetlb_page_range with a for loop like the others there.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The hugetlb pages are currently pre-faulted.  At the time of mmap of
hugepages, we populate the new PTEs.  It is possible that HW has already
cached some of the unused PTEs internally.  These stale entries never
get a chance to be purged in existing control flow.

This patch extends the check in page fault code for hugepages.  Check if
a faulted address falls with in size for the hugetlb file backing it.
We return VM_FAULT_MINOR for these cases (assuming that the arch
specific page-faulting code purges the stale entry for the archs that
need it).
Signed-off-by: NRohit Seth <rohit.seth@intel.com>

[ This is apparently arguably an ia64 port bug. But the code won't
  hurt, and for now it fixes a real problem on some ia64 machines ]
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

As noticed by Nick Piggin, we need to make sure that we check the page
count before we check for PageDirty, since the dirty check is only valid
if the count implies that we're the only possible ones holding the page.

We always did do this, but the code needs a read-memory-barrier to make
sure that the orderign is also honored by the CPU.

(The writer side is ordered due to the atomic decrement and test on the
page count, see the discussion on linux-kernel)
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Refuse to install a page into a mapping if the mapping count is already
ridiculously large.

You probably cannot trigger this on 32-bit architectures, but on a
64-bit setup we should protect against it.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Revert this recent correctness change: Douglas Crosher <dcrosher@scieneer.com>
reported that it broke an existing application, and that madvise() works
without error on anonymous mappings on Solaris.

This means that madvise() will remain non-standards-compliant: we should
return -EBADF for all requests against non-file-backed vma's, but Linux only
does this for MADV_WILLNEED requests.
Signed-off-by: NSuzuki K P <suzuki@in.ibm.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

 - added typedef unsigned int __nocast gfp_t;

 - replaced __nocast uses for gfp flags with gfp_t - it gives exactly
   the same warnings as far as sparse is concerned, doesn't change
   generated code (from gcc point of view we replaced unsigned int with
   typedef) and documents what's going on far better.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

As requested by Thomas Gleixner <tglx@linutronix.de>:

  "5d3d0f77 breaks a couple of ARM
   boards, which depend on the historical bootmem allocation order.
   There is a cleaner solution around to remove the pgdat list
   completely, but this is a topic for post 2.6.14

   Andi signalled ACK already."
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

In kmalloc_node we are checking if the allocation is for the same node when
interrupts are "on".  This may lead to an allocation on another node than
intended.

This patch just shifts the check for the current node in __cache_alloc_node
when interrupts are disabled.
Signed-off-by: NAlok N Kataria <alokk@calsoftinc.com>
Acked-by: NChristoph Lameter <clameter@sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Move the ZERO_PAGE remapping complexity to the move_pte macro in
asm-generic, have it conditionally depend on
__HAVE_ARCH_MULTIPLE_ZERO_PAGE, which gets defined for MIPS.

For architectures without __HAVE_ARCH_MULTIPLE_ZERO_PAGE, move_pte becomes
a noop.

From: Hugh Dickins <hugh@veritas.com>

Fix nasty little bug we've missed in Nick's mremap move ZERO_PAGE patch.
The "pte" at that point may be a swap entry or a pte_file entry: we must
check pte_present before perhaps corrupting such an entry.

Patch below against 2.6.14-rc2-mm1, but the same bug is in 2.6.14-rc2's
mm/mremap.c, and more dangerous there since it's affecting all arches: I
think the safest course is to send Nick's patch and Yoichi's build fix and
this fix (build tested) on to Linus - so only MIPS can be affected.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

As davem points out, this wasn't such a great idea.  There may be some code
which does:

	size = 1024*1024;
	while (kmalloc(size, ...) == 0)
		size /= 2;

which will now explode.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Problem: In some circumstances, bd_claim() is returning the wrong error
code.

If we try to swapon an unused block device that isn't swap formatted, we
get -EINVAL.  But if that same block device is already mounted, we instead
get -EBUSY, even though it still isn't a valid swap device.

This issue came up on the busybox list trying to get the error message
from "swapon -a" right.  If a swap device is already enabled, we get -EBUSY,
and we shouldn't report this as an error.  But we can't distinguish the two
-EBUSY conditions, which are very different errors.

In the code, bd_claim() returns either 0 or -EBUSY, but in this case busy
means "somebody other than sys_swapon has already claimed this", and
_that_ means this block device can't be a valid swap device.  So return
-EINVAL there.
Signed-off-by: NRob Landley <rob@landley.net>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>