linux-toradex.git/include/linux/vmalloc.h, branch v2.6.34-rc3

vmalloc: implement pcpu_get_vm_areas()

2009-08-14T06:00:52+00:00

To directly use spread NUMA memories for percpu units, percpu
allocator will be updated to allow sparsely mapping units in a chunk.
As the distances between units can be very large, this makes
allocating single vmap area for each chunk undesirable.  This patch
implements pcpu_get_vm_areas() and pcpu_free_vm_areas() which
allocates and frees sparse congruent vmap areas.

pcpu_get_vm_areas() take @offsets and @sizes array which define
distances and sizes of vmap areas.  It scans down from the top of
vmalloc area looking for the top-most address which can accomodate all
the areas.  The top-down scan is to avoid interacting with regular
vmallocs which can push up these congruent areas up little by little
ending up wasting address space and page table.

To speed up top-down scan, the highest possible address hint is
maintained.  Although the scan is linear from the hint, given the
usual large holes between memory addresses between NUMA nodes, the
scanning is highly likely to finish after finding the first hole for
the last unit which is scanned first.

Signed-off-by: Tejun Heo 
Cc: Nick Piggin

Merge branch 'tj-percpu' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/misc into core/percpu

2009-02-24T20:52:45+00:00

Conflicts:
	arch/x86/include/asm/pgtable.h

vmalloc: add @align to vm_area_register_early()

2009-02-24T02:57:21+00:00

Impact: allow larger alignment for early vmalloc area allocation

Some early vmalloc users might want larger alignment, for example, for
custom large page mapping.  Add @align to vm_area_register_early().
While at it, drop docbook comment on non-existent @size.

Signed-off-by: Tejun Heo 
Cc: Nick Piggin 
Cc: Ivan Kokshaysky

vmalloc: add un/map_kernel_range_noflush()

2009-02-20T07:29:08+00:00

Impact: two more public map/unmap functions

Implement map_kernel_range_noflush() and unmap_kernel_range_noflush().
These functions respectively map and unmap address range in kernel VM
area but doesn't do any vcache or tlb flushing.  These will be used by
new percpu allocator.

Signed-off-by: Tejun Heo 
Cc: Nick Piggin

vmalloc: implement vm_area_register_early()

2009-02-20T07:29:08+00:00

Impact: allow multiple early vm areas

There are places where kernel VM area needs to be allocated before
vmalloc is initialized.  This is done by allocating static vm_struct,
initializing several fields and linking it to vmlist and later vmalloc
initialization picking up these from vmlist.  This is currently done
manually and if there's more than one such areas, there's no defined
way to arbitrate who gets which address.

This patch implements vm_area_register_early(), which takes vm_area
struct with flags and size initialized, assigns address to it and puts
it on the vmlist.  This way, multiple early vm areas can determine
which addresses they should use.  The only current user - alpha mm
init - is converted to use it.

Signed-off-by: Tejun Heo

vmalloc: add __get_vm_area_caller()

2009-02-18T23:37:53+00:00

We have get_vm_area_caller() and __get_vm_area() but not
__get_vm_area_caller()

On powerpc, I use __get_vm_area() to separate the ranges of addresses
given to vmalloc vs.  ioremap (various good reasons for that) so in order
to be able to implement the new caller tracking in /proc/vmallocinfo, I
need a "_caller" variant of it.

(akpm: needed for ongoing powerpc development, so merge it early)

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Benjamin Herrenschmidt 
Reviewed-by: KOSAKI Motohiro 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: make vread() and vwrite() declaration

2009-01-06T23:59:05+00:00

Sparse output following warnings.

mm/vmalloc.c:1436:6: warning: symbol 'vread' was not declared. Should it be static?
mm/vmalloc.c:1474:6: warning: symbol 'vwrite' was not declared. Should it be static?

However, it is used by /dev/kmem. fixed here.

Signed-off-by: KOSAKI Motohiro 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

proc: move /proc/vmallocinfo to mm/vmalloc.c

2008-10-23T11:48:28+00:00

Signed-off-by: Alexey Dobriyan 
Acked-by: Christoph Lameter

mm: rewrite vmap layer

2008-10-20T15:52:32+00:00

Rewrite the vmap allocator to use rbtrees and lazy tlb flushing, and
provide a fast, scalable percpu frontend for small vmaps (requires a
slightly different API, though).

The biggest problem with vmap is actually vunmap.  Presently this requires
a global kernel TLB flush, which on most architectures is a broadcast IPI
to all CPUs to flush the cache.  This is all done under a global lock.  As
the number of CPUs increases, so will the number of vunmaps a scaled
workload will want to perform, and so will the cost of a global TLB flush.
 This gives terrible quadratic scalability characteristics.

Another problem is that the entire vmap subsystem works under a single
lock.  It is a rwlock, but it is actually taken for write in all the fast
paths, and the read locking would likely never be run concurrently anyway,
so it's just pointless.

This is a rewrite of vmap subsystem to solve those problems.  The existing
vmalloc API is implemented on top of the rewritten subsystem.

The TLB flushing problem is solved by using lazy TLB unmapping.  vmap
addresses do not have to be flushed immediately when they are vunmapped,
because the kernel will not reuse them again (would be a use-after-free)
until they are reallocated.  So the addresses aren't allocated again until
a subsequent TLB flush.  A single TLB flush then can flush multiple
vunmaps from each CPU.

XEN and PAT and such do not like deferred TLB flushing because they can't
always handle multiple aliasing virtual addresses to a physical address.
They now call vm_unmap_aliases() in order to flush any deferred mappings.
That call is very expensive (well, actually not a lot more expensive than
a single vunmap under the old scheme), however it should be OK if not
called too often.

The virtual memory extent information is stored in an rbtree rather than a
linked list to improve the algorithmic scalability.

There is a per-CPU allocator for small vmaps, which amortizes or avoids
global locking.

To use the per-CPU interface, the vm_map_ram / vm_unmap_ram interfaces
must be used in place of vmap and vunmap.  Vmalloc does not use these
interfaces at the moment, so it will not be quite so scalable (although it
will use lazy TLB flushing).

As a quick test of performance, I ran a test that loops in the kernel,
linearly mapping then touching then unmapping 4 pages.  Different numbers
of tests were run in parallel on an 4 core, 2 socket opteron.  Results are
in nanoseconds per map+touch+unmap.

threads           vanilla         vmap rewrite
1                 14700           2900
2                 33600           3000
4                 49500           2800
8                 70631           2900

So with a 8 cores, the rewritten version is already 25x faster.

In a slightly more realistic test (although with an older and less
scalable version of the patch), I ripped the not-very-good vunmap batching
code out of XFS, and implemented the large buffer mapping with vm_map_ram
and vm_unmap_ram...  along with a couple of other tricks, I was able to
speed up a large directory workload by 20x on a 64 CPU system.  I believe
vmap/vunmap is actually sped up a lot more than 20x on such a system, but
I'm running into other locks now.  vmap is pretty well blown off the
profiles.

Before:
1352059 total                                      0.1401
798784 _write_lock                              8320.6667 <- vmlist_lock
529313 default_idle                             1181.5022
 15242 smp_call_function                         15.8771  <- vmap tlb flushing
  2472 __get_vm_area_node                         1.9312  <- vmap
  1762 remove_vm_area                             4.5885  <- vunmap
   316 map_vm_area                                0.2297  <- vmap
   312 kfree                                      0.1950
   300 _spin_lock                                 3.1250
   252 sn_send_IPI_phys                           0.4375  <- tlb flushing
   238 vmap                                       0.8264  <- vmap
   216 find_lock_page                             0.5192
   196 find_next_bit                              0.3603
   136 sn2_send_IPI                               0.2024
   130 pio_phys_write_mmr                         2.0312
   118 unmap_kernel_range                         0.1229

After:
 78406 total                                      0.0081
 40053 default_idle                              89.4040
 33576 ia64_spinlock_contention                 349.7500
  1650 _spin_lock                                17.1875
   319 __reg_op                                   0.5538
   281 _atomic_dec_and_lock                       1.0977
   153 mutex_unlock                               1.5938
   123 iget_locked                                0.1671
   117 xfs_dir_lookup                             0.1662
   117 dput                                       0.1406
   114 xfs_iget_core                              0.0268
    92 xfs_da_hashname                            0.1917
    75 d_alloc                                    0.0670
    68 vmap_page_range                            0.0462 <- vmap
    58 kmem_cache_alloc                           0.0604
    57 memset                                     0.0540
    52 rb_next                                    0.1625
    50 __copy_user                                0.0208
    49 bitmap_find_free_region                    0.2188 <- vmap
    46 ia64_sn_udelay                             0.1106
    45 find_inode_fast                            0.1406
    42 memcmp                                     0.2188
    42 finish_task_switch                         0.1094
    42 __d_lookup                                 0.0410
    40 radix_tree_lookup_slot                     0.1250
    37 _spin_unlock_irqrestore                    0.3854
    36 xfs_bmapi                                  0.0050
    36 kmem_cache_free                            0.0256
    35 xfs_vn_getattr                             0.0322
    34 radix_tree_lookup                          0.1062
    33 __link_path_walk                           0.0035
    31 xfs_da_do_buf                              0.0091
    30 _xfs_buf_find                              0.0204
    28 find_get_page                              0.0875
    27 xfs_iread                                  0.0241
    27 __strncpy_from_user                        0.2812
    26 _xfs_buf_initialize                        0.0406
    24 _xfs_buf_lookup_pages                      0.0179
    24 vunmap_page_range                          0.0250 <- vunmap
    23 find_lock_page                             0.0799
    22 vm_map_ram                                 0.0087 <- vmap
    20 kfree                                      0.0125
    19 put_page                                   0.0330
    18 __kmalloc                                  0.0176
    17 xfs_da_node_lookup_int                     0.0086
    17 _read_lock                                 0.0885
    17 page_waitqueue                             0.0664

vmap has gone from being the top 5 on the profiles and flushing the crap
out of all TLBs, to using less than 1% of kernel time.

[akpm@linux-foundation.org: cleanups, section fix]
[akpm@linux-foundation.org: fix build on alpha]
Signed-off-by: Nick Piggin 
Cc: Jeremy Fitzhardinge 
Cc: Krzysztof Helt 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: VM_flags comment fixes

2008-08-16T23:45:56+00:00

Try to comment away a little of the confusion between mm's vm_area_struct
vm_flags and vmalloc's vm_struct flags: based on an idea by Ulrich Drepper.

Signed-off-by: Hugh Dickins 
Signed-off-by: Linus Torvalds