This patch marks a number of allocations that are either short-lived such as
network buffers or are reclaimable such as inode allocations. When something
like updatedb is called, long-lived and unmovable kernel allocations tend to
be spread throughout the address space which increases fragmentation.
This patch groups these allocations together as much as possible by adding a
new MIGRATE_TYPE. The MIGRATE_RECLAIMABLE type is for allocations that can be
reclaimed on demand, but not moved. i.e. they can be migrated by deleting
them and re-reading the information from elsewhere.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The grouping mechanism has some memory overhead and a more complex allocation
path. This patch allows the strategy to be disabled for small memory systems
or if it is known the workload is suffering because of the strategy. It also
acts to show where the page groupings strategy interacts with the standard
buddy allocator.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Joel Schopp <jschopp@austin.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch adds the core of the fragmentation reduction strategy. It works by
grouping pages together based on their ability to migrate or be reclaimed.
Basically, it works by breaking the list in zone->free_area list into
MIGRATE_TYPES number of lists.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Here is the latest revision of the anti-fragmentation patches. Of particular
note in this version is special treatment of high-order atomic allocations.
Care is taken to group them together and avoid grouping pages of other types
near them. Artifical tests imply that it works. I'm trying to get the
hardware together that would allow setting up of a "real" test. If anyone
already has a setup and test that can trigger the atomic-allocation problem,
I'd appreciate a test of these patches and a report. The second major change
is that these patches will apply cleanly with patches that implement
anti-fragmentation through zones.
kernbench shows effectively no performance difference varying between -0.2%
and +2% on a variety of test machines. Success rates for huge page allocation
are dramatically increased. For example, on a ppc64 machine, the vanilla
kernel was only able to allocate 1% of memory as a hugepage and this was due
to a single hugepage reserved as min_free_kbytes. With these patches applied,
17% was allocatable as superpages. With reclaim-related fixes from Andy
Whitcroft, it was 40% and further reclaim-related improvements should increase
this further.
Changelog Since V28
o Group high-order atomic allocations together
o It is no longer required to set min_free_kbytes to 10% of memory. A value
of 16384 in most cases will be sufficient
o Now applied with zone-based anti-fragmentation
o Fix incorrect VM_BUG_ON within buffered_rmqueue()
o Reorder the stack so later patches do not back out work from earlier patches
o Fix bug were journal pages were being treated as movable
o Bias placement of non-movable pages to lower PFNs
o More agressive clustering of reclaimable pages in reactions to workloads
like updatedb that flood the size of inode caches
Changelog Since V27
o Renamed anti-fragmentation to Page Clustering. Anti-fragmentation was giving
the mistaken impression that it was the 100% solution for high order
allocations. Instead, it greatly increases the chances high-order
allocations will succeed and lays the foundation for defragmentation and
memory hot-remove to work properly
o Redefine page groupings based on ability to migrate or reclaim instead of
basing on reclaimability alone
o Get rid of spurious inits
o Per-cpu lists are no longer split up per-type. Instead the per-cpu list is
searched for a page of the appropriate type
o Added more explanation commentary
o Fix up bug in pageblock code where bitmap was used before being initalised
Changelog Since V26
o Fix double init of lists in setup_pageset
Changelog Since V25
o Fix loop order of for_each_rclmtype_order so that order of loop matches args
o gfpflags_to_rclmtype uses gfp_t instead of unsigned long
o Rename get_pageblock_type() to get_page_rclmtype()
o Fix alignment problem in move_freepages()
o Add mechanism for assigning flags to blocks of pages instead of page->flags
o On fallback, do not examine the preferred list of free pages a second time
The purpose of these patches is to reduce external fragmentation by grouping
pages of related types together. When pages are migrated (or reclaimed under
memory pressure), large contiguous pages will be freed.
This patch works by categorising allocations by their ability to migrate;
Movable - The pages may be moved with the page migration mechanism. These are
generally userspace pages.
Reclaimable - These are allocations for some kernel caches that are
reclaimable or allocations that are known to be very short-lived.
Unmovable - These are pages that are allocated by the kernel that
are not trivially reclaimed. For example, the memory allocated for a
loaded module would be in this category. By default, allocations are
considered to be of this type
HighAtomic - These are high-order allocations belonging to callers that
cannot sleep or perform any IO. In practice, this is restricted to
jumbo frame allocation for network receive. It is assumed that the
allocations are short-lived
Instead of having one MAX_ORDER-sized array of free lists in struct free_area,
there is one for each type of reclaimability. Once a 2^MAX_ORDER block of
pages is split for a type of allocation, it is added to the free-lists for
that type, in effect reserving it. Hence, over time, pages of the different
types can be clustered together.
When the preferred freelists are expired, the largest possible block is taken
from an alternative list. Buddies that are split from that large block are
placed on the preferred allocation-type freelists to mitigate fragmentation.
This implementation gives best-effort for low fragmentation in all zones.
Ideally, min_free_kbytes needs to be set to a value equal to 4 * (1 <<
(MAX_ORDER-1)) pages in most cases. This would be 16384 on x86 and x86_64 for
example.
Our tests show that about 60-70% of physical memory can be allocated on a
desktop after a few days uptime. In benchmarks and stress tests, we are
finding that 80% of memory is available as contiguous blocks at the end of the
test. To compare, a standard kernel was getting < 1% of memory as large pages
on a desktop and about 8-12% of memory as large pages at the end of stress
tests.
Following this email are 12 patches that implement thie page grouping feature.
The first patch introduces a mechanism for storing flags related to a whole
block of pages. Then allocations are split between movable and all other
allocations. Following that are patches to deal with per-cpu pages and make
the mechanism configurable. The next patch moves free pages between lists
when partially allocated blocks are used for pages of another migrate type.
The second last patch groups reclaimable kernel allocations such as inode
caches together. The final patch related to groupings keeps high-order atomic
allocations.
The last two patches are more concerned with control of fragmentation. The
second last patch biases placement of non-movable allocations towards the
start of memory. This is with a view of supporting memory hot-remove of DIMMs
with higher PFNs in the future. The biasing could be enforced a lot heavier
but it would cost. The last patch agressively clusters reclaimable pages like
inode caches together.
The fragmentation reduction strategy needs to track if pages within a block
can be moved or reclaimed so that pages are freed to the appropriate list.
This patch adds a bitmap for flags affecting a whole a MAX_ORDER block of
pages.
In non-SPARSEMEM configurations, the bitmap is stored in the struct zone and
allocated during initialisation. SPARSEMEM statically allocates the bitmap in
a struct mem_section so that bitmaps do not have to be resized during memory
hotadd. This wastes a small amount of memory per unused section (usually
sizeof(unsigned long)) but the complexity of dynamically allocating the memory
is quite high.
Additional credit to Andy Whitcroft who reviewed up an earlier implementation
of the mechanism an suggested how to make it a *lot* cleaner.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
GFP_THISNODE checks that the zone selected is within the pgdat (node) of the
first zone of a nodelist. That only works if the node has memory. A
memoryless node will have its first node on another pgdat (node).
GFP_THISNODE currently will return simply memory on the first pgdat. Thus it
is returning memory on other nodes. GFP_THISNODE should fail if there is no
local memory on a node.
Add a new set of zonelists for each node that only contain the nodes that
belong to the zones itself so that no fallback is possible.
Then modify gfp_type to pickup the right zone based on the presence of
__GFP_THISNODE.
Drop the existing GFP_THISNODE checks from the page_allocators hot path.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Tested-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Acked-by: Bob Picco <bob.picco@hp.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Mel Gorman <mel@skynet.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We have flags to indicate whether a section actually has a valid mem_map
associated with it. This is never set and we rely solely on the present bit
to indicate a section is valid. By definition a section is not valid if it
has no mem_map and there is a window during init where the present bit is set
but there is no mem_map, during which pfn_valid() will return true
incorrectly.
Use the existing SECTION_HAS_MEM_MAP flag to indicate the presence of a valid
mem_map. Switch valid_section{,_nr} and pfn_valid() to this bit. Add a new
present_section{,_nr} and pfn_present() interfaces for those users who care to
know that a section is going to be valid.
[akpm@linux-foundation.org: coding-syle fixes]
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Andi Kleen <ak@suse.de>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The NUMA layer only supports NUMA policies for the highest zone. When
ZONE_MOVABLE is configured with kernelcore=, the the highest zone becomes
ZONE_MOVABLE. The result is that policies are only applied to allocations
like anonymous pages and page cache allocated from ZONE_MOVABLE when the
zone is used.
This patch applies policies to the two highest zones when the highest zone
is ZONE_MOVABLE. As ZONE_MOVABLE consists of pages from the highest "real"
zone, it's always functionally equivalent.
The patch has been tested on a variety of machines both NUMA and non-NUMA
covering x86, x86_64 and ppc64. No abnormal results were seen in
kernbench, tbench, dbench or hackbench. It passes regression tests from
the numactl package with and without kernelcore= once numactl tests are
patched to wait for vmstat counters to update.
akpm: this is the nasty hack to fix NUMA mempolicies in the presence of
ZONE_MOVABLE and kernelcore= in 2.6.23. Christoph says "For .24 either merge
the mobility or get the other solution that Mel is working on. That solution
would only use a single zonelist per node and filter on the fly. That may
help performance and also help to make memory policies work better."
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Tested-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Acked-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The arm26 port has been in a state where it was far from even compiling
for quite some time.
Ian Molton agreed with the removal.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Ian Molton <spyro@f2s.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When we are out of memory of a suitable size we enter reclaim. The current
reclaim algorithm targets pages in LRU order, which is great for fairness at
order-0 but highly unsuitable if you desire pages at higher orders. To get
pages of higher order we must shoot down a very high proportion of memory;
>95% in a lot of cases.
This patch set adds a lumpy reclaim algorithm to the allocator. It targets
groups of pages at the specified order anchored at the end of the active and
inactive lists. This encourages groups of pages at the requested orders to
move from active to inactive, and active to free lists. This behaviour is
only triggered out of direct reclaim when higher order pages have been
requested.
This patch set is particularly effective when utilised with an
anti-fragmentation scheme which groups pages of similar reclaimability
together.
This patch set is based on Peter Zijlstra's lumpy reclaim V2 patch which forms
the foundation. Credit to Mel Gorman for sanitity checking.
Mel said:
The patches have an application with hugepage pool resizing.
When lumpy-reclaim is used used with ZONE_MOVABLE, the hugepages pool can
be resized with greater reliability. Testing on a desktop machine with 2GB
of RAM showed that growing the hugepage pool with ZONE_MOVABLE on it's own
was very slow as the success rate was quite low. Without lumpy-reclaim,
each attempt to grow the pool by 100 pages would yield 1 or 2 hugepages.
With lumpy-reclaim, getting 40 to 70 hugepages on each attempt was typical.
[akpm@osdl.org: ia64 pfn_to_nid fixes and loop cleanup]
[bunk@stusta.de: static declarations for internal functions]
[a.p.zijlstra@chello.nl: initial lumpy V2 implementation]
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The following 8 patches against 2.6.20-mm2 create a zone called ZONE_MOVABLE
that is only usable by allocations that specify both __GFP_HIGHMEM and
__GFP_MOVABLE. This has the effect of keeping all non-movable pages within a
single memory partition while allowing movable allocations to be satisfied
from either partition. The patches may be applied with the list-based
anti-fragmentation patches that groups pages together based on mobility.
The size of the zone is determined by a kernelcore= parameter specified at
boot-time. This specifies how much memory is usable by non-movable
allocations and the remainder is used for ZONE_MOVABLE. Any range of pages
within ZONE_MOVABLE can be released by migrating the pages or by reclaiming.
When selecting a zone to take pages from for ZONE_MOVABLE, there are two
things to consider. First, only memory from the highest populated zone is
used for ZONE_MOVABLE. On the x86, this is probably going to be ZONE_HIGHMEM
but it would be ZONE_DMA on ppc64 or possibly ZONE_DMA32 on x86_64. Second,
the amount of memory usable by the kernel will be spread evenly throughout
NUMA nodes where possible. If the nodes are not of equal size, the amount of
memory usable by the kernel on some nodes may be greater than others.
By default, the zone is not as useful for hugetlb allocations because they are
pinned and non-migratable (currently at least). A sysctl is provided that
allows huge pages to be allocated from that zone. This means that the huge
page pool can be resized to the size of ZONE_MOVABLE during the lifetime of
the system assuming that pages are not mlocked. Despite huge pages being
non-movable, we do not introduce additional external fragmentation of note as
huge pages are always the largest contiguous block we care about.
Credit goes to Andy Whitcroft for catching a large variety of problems during
review of the patches.
This patch creates an additional zone, ZONE_MOVABLE. This zone is only usable
by allocations which specify both __GFP_HIGHMEM and __GFP_MOVABLE. Hot-added
memory continues to be placed in their existing destination as there is no
mechanism to redirect them to a specific zone.
[y-goto@jp.fujitsu.com: Fix section mismatch of memory hotplug related code]
[akpm@linux-foundation.org: various fixes]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make zonelist creation policy selectable from sysctl/boot option v6.
This patch makes NUMA's zonelist (of pgdat) order selectable.
Available order are Default(automatic)/ Node-based / Zone-based.
[Default Order]
The kernel selects Node-based or Zone-based order automatically.
[Node-based Order]
This policy treats the locality of memory as the most important parameter.
Zonelist order is created by each zone's locality. This means lower zones
(ex. ZONE_DMA) can be used before higher zone (ex. ZONE_NORMAL) exhausion.
IOW. ZONE_DMA will be in the middle of zonelist.
current 2.6.21 kernel uses this.
Pros.
* A user can expect local memory as much as possible.
Cons.
* lower zone will be exhansted before higher zone. This may cause OOM_KILL.
Maybe suitable if ZONE_DMA is relatively big and you never see OOM_KILL
because of ZONE_DMA exhaution and you need the best locality.
(example)
assume 2 node NUMA. node(0) has ZONE_DMA/ZONE_NORMAL, node(1) has ZONE_NORMAL.
*node(0)'s memory allocation order:
node(0)'s NORMAL -> node(0)'s DMA -> node(1)'s NORMAL.
*node(1)'s memory allocation order:
node(1)'s NORMAL -> node(0)'s NORMAL -> node(0)'s DMA.
[Zone-based order]
This policy treats the zone type as the most important parameter.
Zonelist order is created by zone-type order. This means lower zone
never be used bofere higher zone exhaustion.
IOW. ZONE_DMA will be always at the tail of zonelist.
Pros.
* OOM_KILL(bacause of lower zone) occurs only if the whole zones are exhausted.
Cons.
* memory locality may not be best.
(example)
assume 2 node NUMA. node(0) has ZONE_DMA/ZONE_NORMAL, node(1) has ZONE_NORMAL.
*node(0)'s memory allocation order:
node(0)'s NORMAL -> node(1)'s NORMAL -> node(0)'s DMA.
*node(1)'s memory allocation order:
node(1)'s NORMAL -> node(0)'s NORMAL -> node(0)'s DMA.
bootoption "numa_zonelist_order=" and proc/sysctl is supporetd.
command:
%echo N > /proc/sys/vm/numa_zonelist_order
Will rebuild zonelist in Node-based order.
command:
%echo Z > /proc/sys/vm/numa_zonelist_order
Will rebuild zonelist in Zone-based order.
Thanks to Lee Schermerhorn, he gives me much help and codes.
[Lee.Schermerhorn@hp.com: add check_highest_zone to build_zonelists_in_zone_order]
[akpm@linux-foundation.org: build fix]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@suse.de>
Cc: "jesse.barnes@intel.com" <jesse.barnes@intel.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently the slab allocators contain callbacks into the page allocator to
perform the draining of pagesets on remote nodes. This requires SLUB to have
a whole subsystem in order to be compatible with SLAB. Moving node draining
out of the slab allocators avoids a section of code in SLUB.
Move the node draining so that is is done when the vm statistics are updated.
At that point we are already touching all the cachelines with the pagesets of
a processor.
Add a expire counter there. If we have to update per zone or global vm
statistics then assume that the pageset will require subsequent draining.
The expire counter will be decremented on each vm stats update pass until it
reaches zero. Then we will drain one batch from the pageset. The draining
will cause vm counter updates which will then cause another expiration until
the pcp is empty. So we will drain a batch every 3 seconds.
Note that remote node draining is a somewhat esoteric feature that is required
on large NUMA systems because otherwise significant portions of system memory
can become trapped in pcp queues. The number of pcp is determined by the
number of processors and nodes in a system. A system with 4 processors and 2
nodes has 8 pcps which is okay. But a system with 1024 processors and 512
nodes has 512k pcps with a high potential for large amount of memory being
caught in them.
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>
Generally we work under the assumption that memory the mem_map array is
contigious and valid out to MAX_ORDER_NR_PAGES block of pages, ie. that if we
have validated any page within this MAX_ORDER_NR_PAGES block we need not check
any other. This is not true when CONFIG_HOLES_IN_ZONE is set and we must
check each and every reference we make from a pfn.
Add a pfn_valid_within() helper which should be used when scanning pages
within a MAX_ORDER_NR_PAGES block when we have already checked the validility
of the block normally with pfn_valid(). This can then be optimised away when
we do not have holes within a MAX_ORDER_NR_PAGES block of pages.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make ZONE_DMA optional in core code.
- ifdef all code for ZONE_DMA and related definitions following the example
for ZONE_DMA32 and ZONE_HIGHMEM.
- Without ZONE_DMA, ZONE_HIGHMEM and ZONE_DMA32 we get to a ZONES_SHIFT of
0.
- Modify the VM statistics to work correctly without a DMA zone.
- Modify slab to not create DMA slabs if there is no ZONE_DMA.
[akpm@osdl.org: cleanup]
[jdike@addtoit.com: build fix]
[apw@shadowen.org: Simplify calculation of the number of bits we need for ZONES_SHIFT]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@suse.de>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Matthew Wilcox <willy@debian.org>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Values are readily available via ZVC per node and global sums.
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 global and per zone counter sums are in arrays of longs. Reorder the ZVCs
so that the most frequently used ZVCs are put into the same cacheline. That
way calculations of the global, node and per zone vm state touches only a
single cacheline. This is mostly important for 64 bit systems were one 128
byte cacheline takes only 8 longs.
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 is again simplifies some of the VM counter calculations through the use
of the ZVC consolidated counters.
[michal.k.k.piotrowski@gmail.com: build fix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The determination of the dirty ratio to determine writeback behavior is
currently based on the number of total pages on the system.
However, not all pages in the system may be dirtied. Thus the ratio is always
too low and can never reach 100%. The ratio may be particularly skewed if
large hugepage allocations, slab allocations or device driver buffers make
large sections of memory not available anymore. In that case we may get into
a situation in which f.e. the background writeback ratio of 40% cannot be
reached anymore which leads to undesired writeback behavior.
This patchset fixes that issue by determining the ratio based on the actual
pages that may potentially be dirty. These are the pages on the active and
the inactive list plus free pages.
The problem with those counts has so far been that it is expensive to
calculate these because counts from multiple nodes and multiple zones will
have to be summed up. This patchset makes these counters ZVC counters. This
means that a current sum per zone, per node and for the whole system is always
available via global variables and not expensive anymore to calculate.
The patchset results in some other good side effects:
- Removal of the various functions that sum up free, active and inactive
page counts
- Cleanup of the functions that display information via the proc filesystem.
This patch:
The use of a ZVC for nr_inactive and nr_active allows a simplification of some
counter operations. More ZVC functionality is used for sums etc in the
following patches.
[akpm@osdl.org: UP build fix]
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>
Fix an oops experienced on the Cell architecture when init-time functions,
early_*(), are called at runtime. It alters the call paths to make sure
that the callers explicitly say whether the call is being made on behalf of
a hotplug even, or happening at boot-time.
It has been compile tested on ppc64, ia64, s390, i386 and x86_64.
Acked-by: Arnd Bergmann <arndb@de.ibm.com>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Acked-by: Andy Whitcroft <apw@shadowen.org>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
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>
- move some file_operations structs into the .rodata section
- move static strings from policy_types[] array into the .rodata section
- fix generic seq_operations usages, so that those structs may be defined
as "const" as well
[akpm@osdl.org: couple of fixes]
Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Rearrange the struct members in the 'struct zonelist_cache' structure, so
as to put the readonly (once initialized) z_to_n[] array first, where it
will come right after the zones[] array in struct zonelist.
This pretty much eliminates the chance that the two frequently written
elements of 'struct zonelist_cache', the fullzones bitmap and last_full_zap
times, will end up on the same cache line as the performance sensitive,
frequently read, never (after init) written zones[] array.
Keeping frequently written data off frequently read cache lines is good for
performance.
Thanks to Rohit Seth for the suggestion.
Signed-off-by: Paul Jackson <pj@sgi.com>
Cc: Rohit Seth <rohitseth@google.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Optimize the critical zonelist scanning for free pages in the kernel memory
allocator by caching the zones that were found to be full recently, and
skipping them.
Remembers the zones in a zonelist that were short of free memory in the
last second. And it stashes a zone-to-node table in the zonelist struct,
to optimize that conversion (minimize its cache footprint.)
Recent changes:
This differs in a significant way from a similar patch that I
posted a week ago. Now, instead of having a nodemask_t of
recently full nodes, I have a bitmask of recently full zones.
This solves a problem that last weeks patch had, which on
systems with multiple zones per node (such as DMA zone) would
take seeing any of these zones full as meaning that all zones
on that node were full.
Also I changed names - from "zonelist faster" to "zonelist cache",
as that seemed to better convey what we're doing here - caching
some of the key zonelist state (for faster access.)
See below for some performance benchmark results. After all that
discussion with David on why I didn't need them, I went and got
some ;). I wanted to verify that I had not hurt the normal case
of memory allocation noticeably. At least for my one little
microbenchmark, I found (1) the normal case wasn't affected, and
(2) workloads that forced scanning across multiple nodes for
memory improved up to 10% fewer System CPU cycles and lower
elapsed clock time ('sys' and 'real'). Good. See details, below.
I didn't have the logic in get_page_from_freelist() for various
full nodes and zone reclaim failures correct. That should be
fixed up now - notice the new goto labels zonelist_scan,
this_zone_full, and try_next_zone, in get_page_from_freelist().
There are two reasons I persued this alternative, over some earlier
proposals that would have focused on optimizing the fake numa
emulation case by caching the last useful zone:
1) Contrary to what I said before, we (SGI, on large ia64 sn2 systems)
have seen real customer loads where the cost to scan the zonelist
was a problem, due to many nodes being full of memory before
we got to a node we could use. Or at least, I think we have.
This was related to me by another engineer, based on experiences
from some time past. So this is not guaranteed. Most likely, though.
The following approach should help such real numa systems just as
much as it helps fake numa systems, or any combination thereof.
2) The effort to distinguish fake from real numa, using node_distance,
so that we could cache a fake numa node and optimize choosing
it over equivalent distance fake nodes, while continuing to
properly scan all real nodes in distance order, was going to
require a nasty blob of zonelist and node distance munging.
The following approach has no new dependency on node distances or
zone sorting.
See comment in the patch below for a description of what it actually does.
Technical details of note (or controversy):
- See the use of "zlc_active" and "did_zlc_setup" below, to delay
adding any work for this new mechanism until we've looked at the
first zone in zonelist. I figured the odds of the first zone
having the memory we needed were high enough that we should just
look there, first, then get fancy only if we need to keep looking.
- Some odd hackery was needed to add items to struct zonelist, while
not tripping up the custom zonelists built by the mm/mempolicy.c
code for MPOL_BIND. My usual wordy comments below explain this.
Search for "MPOL_BIND".
- Some per-node data in the struct zonelist is now modified frequently,
with no locking. Multiple CPU cores on a node could hit and mangle
this data. The theory is that this is just performance hint data,
and the memory allocator will work just fine despite any such mangling.
The fields at risk are the struct 'zonelist_cache' fields 'fullzones'
(a bitmask) and 'last_full_zap' (unsigned long jiffies). It should
all be self correcting after at most a one second delay.
- This still does a linear scan of the same lengths as before. All
I've optimized is making the scan faster, not algorithmically
shorter. It is now able to scan a compact array of 'unsigned
short' in the case of many full nodes, so one cache line should
cover quite a few nodes, rather than each node hitting another
one or two new and distinct cache lines.
- If both Andi and Nick don't find this too complicated, I will be
(pleasantly) flabbergasted.
- I removed the comment claiming we only use one cachline's worth of
zonelist. We seem, at least in the fake numa case, to have put the
lie to that claim.
- I pay no attention to the various watermarks and such in this performance
hint. A node could be marked full for one watermark, and then skipped
over when searching for a page using a different watermark. I think
that's actually quite ok, as it will tend to slightly increase the
spreading of memory over other nodes, away from a memory stressed node.
===============
Performance - some benchmark results and analysis:
This benchmark runs a memory hog program that uses multiple
threads to touch alot of memory as quickly as it can.
Multiple runs were made, touching 12, 38, 64 or 90 GBytes out of
the total 96 GBytes on the system, and using 1, 19, 37, or 55
threads (on a 56 CPU system.) System, user and real (elapsed)
timings were recorded for each run, shown in units of seconds,
in the table below.
Two kernels were tested - 2.6.18-mm3 and the same kernel with
this zonelist caching patch added. The table also shows the
percentage improvement the zonelist caching sys time is over
(lower than) the stock *-mm kernel.
number 2.6.18-mm3 zonelist-cache delta (< 0 good) percent
GBs N ------------ -------------- ---------------- systime
mem threads sys user real sys user real sys user real better
12 1 153 24 177 151 24 176 -2 0 -1 1%
12 19 99 22 8 99 22 8 0 0 0 0%
12 37 111 25 6 112 25 6 1 0 0 -0%
12 55 115 25 5 110 23 5 -5 -2 0 4%
38 1 502 74 576 497 73 570 -5 -1 -6 0%
38 19 426 78 48 373 76 39 -53 -2 -9 12%
38 37 544 83 36 547 82 36 3 -1 0 -0%
38 55 501 77 23 511 80 24 10 3 1 -1%
64 1 917 125 1042 890 124 1014 -27 -1 -28 2%
64 19 1118 138 119 965 141 103 -153 3 -16 13%
64 37 1202 151 94 1136 150 81 -66 -1 -13 5%
64 55 1118 141 61 1072 140 58 -46 -1 -3 4%
90 1 1342 177 1519 1275 174 1450 -67 -3 -69 4%
90 19 2392 199 192 2116 189 176 -276 -10 -16 11%
90 37 3313 238 175 2972 225 145 -341 -13 -30 10%
90 55 1948 210 104 1843 213 100 -105 3 -4 5%
Notes:
1) This test ran a memory hog program that started a specified number N of
threads, and had each thread allocate and touch 1/N'th of
the total memory to be used in the test run in a single loop,
writing a constant word to memory, one store every 4096 bytes.
Watching this test during some earlier trial runs, I would see
each of these threads sit down on one CPU and stay there, for
the remainder of the pass, a different CPU for each thread.
2) The 'real' column is not comparable to the 'sys' or 'user' columns.
The 'real' column is seconds wall clock time elapsed, from beginning
to end of that test pass. The 'sys' and 'user' columns are total
CPU seconds spent on that test pass. For a 19 thread test run,
for example, the sum of 'sys' and 'user' could be up to 19 times the
number of 'real' elapsed wall clock seconds.
3) Tests were run on a fresh, single-user boot, to minimize the amount
of memory already in use at the start of the test, and to minimize
the amount of background activity that might interfere.
4) Tests were done on a 56 CPU, 28 Node system with 96 GBytes of RAM.
5) Notice that the 'real' time gets large for the single thread runs, even
though the measured 'sys' and 'user' times are modest. I'm not sure what
that means - probably something to do with it being slow for one thread to
be accessing memory along ways away. Perhaps the fake numa system, running
ostensibly the same workload, would not show this substantial degradation
of 'real' time for one thread on many nodes -- lets hope not.
6) The high thread count passes (one thread per CPU - on 55 of 56 CPUs)
ran quite efficiently, as one might expect. Each pair of threads needed
to allocate and touch the memory on the node the two threads shared, a
pleasantly parallizable workload.
7) The intermediate thread count passes, when asking for alot of memory forcing
them to go to a few neighboring nodes, improved the most with this zonelist
caching patch.
Conclusions:
* This zonelist cache patch probably makes little difference one way or the
other for most workloads on real numa hardware, if those workloads avoid
heavy off node allocations.
* For memory intensive workloads requiring substantial off-node allocations
on real numa hardware, this patch improves both kernel and elapsed timings
up to ten per-cent.
* For fake numa systems, I'm optimistic, but will have to leave that up to
Rohit Seth to actually test (once I get him a 2.6.18 backport.)
Signed-off-by: Paul Jackson <pj@sgi.com>
Cc: Rohit Seth <rohitseth@google.com>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Cc: David Rientjes <rientjes@cs.washington.edu>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The temp_priority field in zone is racy, as we can walk through a reclaim
path, and just before we copy it into prev_priority, it can be overwritten
(say with DEF_PRIORITY) by another reclaimer.
The same bug is contained in both try_to_free_pages and balance_pgdat, but
it is fixed slightly differently. In balance_pgdat, we keep a separate
priority record per zone in a local array. In try_to_free_pages there is
no need to do this, as the priority level is the same for all zones that we
reclaim from.
Impact of this bug is that temp_priority is copied into prev_priority, and
setting this artificially high causes reclaimers to set distress
artificially low. They then fail to reclaim mapped pages, when they are,
in fact, under severe memory pressure (their priority may be as low as 0).
This causes the OOM killer to fire incorrectly.
From: Andrew Morton <akpm@osdl.org>
__zone_reclaim() isn't modifying zone->prev_priority. But zone->prev_priority
is used in the decision whether or not to bring mapped pages onto the inactive
list. Hence there's a risk here that __zone_reclaim() will fail because
zone->prev_priority ir large (ie: low urgency) and lots of mapped pages end up
stuck on the active list.
Fix that up by decreasing (ie making more urgent) zone->prev_priority as
__zone_reclaim() scans the zone's pages.
This bug perhaps explains why ZONE_RECLAIM_PRIORITY was created. It should be
possible to remove that now, and to just start out at DEF_PRIORITY?
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Reintroduce NODES_SPAN_OTHER_NODES for powerpc
Revert "[PATCH] Remove SPAN_OTHER_NODES config definition"
This reverts commit f62859bb68.
Revert "[PATCH] mm: remove arch independent NODES_SPAN_OTHER_NODES"
This reverts commit a94b3ab7ea.
Also update the comments to indicate that this is still required
and where its used.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Mike Kravetz <kravetz@us.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Will Schmidt <will_schmidt@vnet.ibm.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add the node in order to optimize zone_to_nid.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This moves the definition of struct page from mm.h to its own header file
page-struct.h. This is a prereq to fix SetPageUptodate which is broken on
s390:
#define SetPageUptodate(_page)
do {
struct page *__page = (_page);
if (!test_and_set_bit(PG_uptodate, &__page->flags))
page_test_and_clear_dirty(_page);
} while (0)
_page gets used twice in this macro which can cause subtle bugs. Using
__page for the page_test_and_clear_dirty call doesn't work since it causes
yet another problem with the page_test_and_clear_dirty macro as well.
In order to avoid all these problems caused by macros it seems to be a good
idea to get rid of them and convert them to static inline functions.
Because of header file include order it's necessary to have a seperate
header file for the struct page definition.
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The VM is supposed to minimise the number of pages which get written off the
LRU (for IO scheduling efficiency, and for high reclaim-success rates). But
we don't actually have a clear way of showing how true this is.
So add `nr_vmscan_write' to /proc/vmstat and /proc/zoneinfo - the number of
pages which have been written by the vm scanner in this zone and globally.
Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
At a basic level, architectures define structures to record where active
ranges of page frames are located. Once located, the code to calculate zone
sizes and holes in each architecture is very similar. Some of this zone and
hole sizing code is difficult to read for no good reason. This set of patches
eliminates the similar-looking architecture-specific code.
The patches introduce a mechanism where architectures register where the
active ranges of page frames are with add_active_range(). When all areas have
been discovered, free_area_init_nodes() is called to initialise the pgdat and
zones. The zone sizes and holes are then calculated in an architecture
independent manner.
Patch 1 introduces the mechanism for registering and initialising PFN ranges
Patch 2 changes ppc to use the mechanism - 139 arch-specific LOC removed
Patch 3 changes x86 to use the mechanism - 136 arch-specific LOC removed
Patch 4 changes x86_64 to use the mechanism - 74 arch-specific LOC removed
Patch 5 changes ia64 to use the mechanism - 52 arch-specific LOC removed
Patch 6 accounts for mem_map as a memory hole as the pages are not reclaimable.
It adjusts the watermarks slightly
Tony Luck has successfully tested for ia64 on Itanium with tiger_defconfig,
gensparse_defconfig and defconfig. Bob Picco has also tested and debugged on
IA64. Jack Steiner successfully boot tested on a mammoth SGI IA64-based
machine. These were on patches against 2.6.17-rc1 and release 3 of these
patches but there have been no ia64-changes since release 3.
There are differences in the zone sizes for x86_64 as the arch-specific code
for x86_64 accounts the kernel image and the starting mem_maps as memory holes
but the architecture-independent code accounts the memory as present.
The big benefit of this set of patches is a sizable reduction of
architecture-specific code, some of which is very hairy. There should be a
greater reduction when other architectures use the same mechanisms for zone
and hole sizing but I lack the hardware to test on.
Additional credit;
Dave Hansen for the initial suggestion and comments on early patches
Andy Whitcroft for reviewing early versions and catching numerous
errors
Tony Luck for testing and debugging on IA64
Bob Picco for fixing bugs related to pfn registration, reviewing a
number of patch revisions, providing a number of suggestions
on future direction and testing heavily
Jack Steiner and Robin Holt for testing on IA64 and clarifying
issues related to memory holes
Yasunori for testing on IA64
Andi Kleen for reviewing and feeding back about x86_64
Christian Kujau for providing valuable information related to ACPI
problems on x86_64 and testing potential fixes
This patch:
Define the structure to represent an active range of page frames within a node
in an architecture independent manner. Architectures are expected to register
active ranges of PFNs using add_active_range(nid, start_pfn, end_pfn) and call
free_area_init_nodes() passing the PFNs of the end of each zone.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently one can enable slab reclaim by setting an explicit option in
/proc/sys/vm/zone_reclaim_mode. Slab reclaim is then used as a final
option if the freeing of unmapped file backed pages is not enough to free
enough pages to allow a local allocation.
However, that means that the slab can grow excessively and that most memory
of a node may be used by slabs. We have had a case where a machine with
46GB of memory was using 40-42GB for slab. Zone reclaim was effective in
dealing with pagecache pages. However, slab reclaim was only done during
global reclaim (which is a bit rare on NUMA systems).
This patch implements slab reclaim during zone reclaim. Zone reclaim
occurs if there is a danger of an off node allocation. At that point we
1. Shrink the per node page cache if the number of pagecache
pages is more than min_unmapped_ratio percent of pages in a zone.
2. Shrink the slab cache if the number of the nodes reclaimable slab pages
(patch depends on earlier one that implements that counter)
are more than min_slab_ratio (a new /proc/sys/vm tunable).
The shrinking of the slab cache is a bit problematic since it is not node
specific. So we simply calculate what point in the slab we want to reach
(current per node slab use minus the number of pages that neeed to be
allocated) and then repeately run the global reclaim until that is
unsuccessful or we have reached the limit. I hope we will have zone based
slab reclaim at some point which will make that easier.
The default for the min_slab_ratio is 5%
Also remove the slab option from /proc/sys/vm/zone_reclaim_mode.
[akpm@osdl.org: cleanups]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove the atomic counter for slab_reclaim_pages and replace the counter
and NR_SLAB with two ZVC counter that account for unreclaimable and
reclaimable slab pages: NR_SLAB_RECLAIMABLE and NR_SLAB_UNRECLAIMABLE.
Change the check in vmscan.c to refer to to NR_SLAB_RECLAIMABLE. The
intend seems to be to check for slab pages that could be freed.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
*_pages is a better description of the role of the variable.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I wonder why we need this bitmask indexing into zone->node_zonelists[]?
We always start with the highest zone and then include all lower zones
if we build zonelists.
Are there really cases where we need allocation from ZONE_DMA or
ZONE_HIGHMEM but not ZONE_NORMAL? It seems that the current implementation
of highest_zone() makes that already impossible.
If we go linear on the index then gfp_zone() == highest_zone() and a lot
of definitions fall by the wayside.
We can now revert back to the use of gfp_zone() in mempolicy.c ;-)
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make ZONE_HIGHMEM optional
- ifdef out code and definitions related to CONFIG_HIGHMEM
- __GFP_HIGHMEM falls back to normal allocations if there is no
ZONE_HIGHMEM
- GFP_ZONEMASK becomes 0x01 if there is no DMA32 and no HIGHMEM
zone.
[jdike@addtoit.com: build fix]
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make ZONE_DMA32 optional
- Add #ifdefs around ZONE_DMA32 specific code and definitions.
- Add CONFIG_ZONE_DMA32 config option and use that for x86_64
that alone needs this zone.
- Remove the use of CONFIG_DMA_IS_DMA32 and CONFIG_DMA_IS_NORMAL
for ia64 and fix up the way per node ZVCs are calculated.
- Fall back to prior GFP_ZONEMASK of 0x03 if there is no
DMA32 zone.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use enum for zones and reformat zones dependent information
Add comments explaning the use of zones and add a zones_t type for zone
numbers.
Line up information that will be #ifdefd by the following patches.
[akpm@osdl.org: comment cleanups]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The ZVC counter update threshold is currently set to a fixed value of 32.
This patch sets up the threshold depending on the number of processors and
the sizes of the zones in the system.
With the current threshold of 32, I was able to observe slight contention
when more than 130-140 processors concurrently updated the counters. The
contention vanished when I either increased the threshold to 64 or used
Andrew's idea of overstepping the interval (see ZVC overstep patch).
However, we saw contention again at 220-230 processors. So we need higher
values for larger systems.
But the current default is already a bit of an overkill for smaller
systems. Some systems have tiny zones where precision matters. For
example i386 and x86_64 have 16M DMA zones and either 900M ZONE_NORMAL or
ZONE_DMA32. These are even present on SMP and NUMA systems.
The patch here sets up a threshold based on the number of processors in the
system and the size of the zone that these counters are used for. The
threshold should grow logarithmically, so we use fls() as an easy
approximation.
Results of tests on a system with 1024 processors (4TB RAM)
The following output is from a test allocating 1GB of memory concurrently
on each processor (Forking the process. So contention on mmap_sem and the
pte locks is not a factor):
X MIN
TYPE: CPUS WALL WALL SYS USER TOTCPU
fork 1 0.552 0.552 0.540 0.012 0.552
fork 4 0.552 0.548 2.164 0.036 2.200
fork 16 0.564 0.548 8.812 0.164 8.976
fork 128 0.580 0.572 72.204 1.208 73.412
fork 256 1.300 0.660 310.400 2.160 312.560
fork 512 3.512 0.696 1526.836 4.816 1531.652
fork 1020 20.024 0.700 17243.176 6.688 17249.863
So a threshold of 32 is fine up to 128 processors. At 256 processors contention
becomes a factor.
Overstepping the counter (earlier patch) improves the numbers a bit:
fork 4 0.552 0.548 2.164 0.040 2.204
fork 16 0.552 0.548 8.640 0.148 8.788
fork 128 0.556 0.548 69.676 0.956 70.632
fork 256 0.876 0.636 212.468 2.108 214.576
fork 512 2.276 0.672 997.324 4.260 1001.584
fork 1020 13.564 0.680 11586.436 6.088 11592.523
Still contention at 512 and 1020. Contention at 1020 is down by a third.
256 still has a slight bit of contention.
After this patch the counter threshold will be set to 125 which reduces
contention significantly:
fork 128 0.560 0.548 69.776 0.932 70.708
fork 256 0.636 0.556 143.460 2.036 145.496
fork 512 0.640 0.548 284.244 4.236 288.480
fork 1020 1.500 0.588 1326.152 8.892 1335.044
[akpm@osdl.org: !SMP build fix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It turns out that it is advantageous to leave a small portion of unmapped file
backed pages if all of a zone's pages (or almost all pages) are allocated and
so the page allocator has to go off-node.
This allows recently used file I/O buffers to stay on the node and
reduces the times that zone reclaim is invoked if file I/O occurs
when we run out of memory in a zone.
The problem is that zone reclaim runs too frequently when the page cache is
used for file I/O (read write and therefore unmapped pages!) alone and we have
almost all pages of the zone allocated. Zone reclaim may remove 32 unmapped
pages. File I/O will use these pages for the next read/write requests and the
unmapped pages increase. After the zone has filled up again zone reclaim will
remove it again after only 32 pages. This cycle is too inefficient and there
are potentially too many zone reclaim cycles.
With the 1% boundary we may still remove all unmapped pages for file I/O in
zone reclaim pass. However. it will take a large number of read and writes
to get back to 1% again where we trigger zone reclaim again.
The zone reclaim 2.6.16/17 does not show this behavior because we have a 30
second timeout.
[akpm@osdl.org: rename the /proc file and the variable]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The numa statistics are really event counters. But they are per node and
so we have had special treatment for these counters through additional
fields on the pcp structure. We can now use the per zone nature of the
zoned VM counters to realize these.
This will shrink the size of the pcp structure on NUMA systems. We will
have some room to add additional per zone counters that will all still fit
in the same cacheline.
Bits Prior pcp size Size after patch We can add
------------------------------------------------------------------
64 128 bytes (16 words) 80 bytes (10 words) 48
32 76 bytes (19 words) 56 bytes (14 words) 8 (64 byte cacheline)
72 (128 byte)
Remove the special statistics for numa and replace them with zoned vm
counters. This has the side effect that global sums of these events now
show up in /proc/vmstat.
Also take the opportunity to move the zone_statistics() function from
page_alloc.c into vmstat.c.
Discussions:
V2 http://marc.theaimsgroup.com/?t=115048227000002&r=1&w=2
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Conversion of nr_bounce to a per zone counter
nr_bounce is only used for proc output. So it could be left as an event
counter. However, the event counters may not be accurate and nr_bounce is
categorizing types of pages in a zone. So we really need this to also be a
per zone counter.
[akpm@osdl.org: bugfix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Conversion of nr_unstable to a per zone counter
We need to do some special modifications to the nfs code since there are
multiple cases of disposition and we need to have a page ref for proper
accounting.
This converts the last critical page state of the VM and therefore we need to
remove several functions that were depending on GET_PAGE_STATE_LAST in order
to make the kernel compile again. We are only left with event type counters
in page state.
[akpm@osdl.org: bugfixes]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Conversion of nr_writeback to per zone counter.
This removes the last page_state counter from arch/i386/mm/pgtable.c so we
drop the page_state from there.
[akpm@osdl.org: bugfix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This makes nr_dirty a per zone counter. Looping over all processors is
avoided during writeback state determination.
The counter aggregation for nr_dirty had to be undone in the NFS layer since
we summed up the page counts from multiple zones. Someone more familiar with
NFS should probably review what I have done.
[akpm@osdl.org: bugfix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Conversion of nr_page_table_pages to a per zone counter
[akpm@osdl.org: bugfix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- Allows reclaim to access counter without looping over processor counts.
- Allows accurate statistics on how many pages are used in a zone by
the slab. This may become useful to balance slab allocations over
various zones.
[akpm@osdl.org: bugfix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The zone_reclaim_interval was necessary because we were not able to determine
how many unmapped pages exist in a zone. Therefore we had to scan in
intervals to figure out if any pages were unmapped.
With the zoned counters and NR_ANON_PAGES we now know the number of pagecache
pages and the number of mapped pages in a zone. So we can simply skip the
reclaim if there is an insufficient number of unmapped pages. We use
SWAP_CLUSTER_MAX as the boundary.
Drop all support for /proc/sys/vm/zone_reclaim_interval.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The current NR_FILE_MAPPED is used by zone reclaim and the dirty load
calculation as the number of mapped pagecache pages. However, that is not
true. NR_FILE_MAPPED includes the mapped anonymous pages. This patch
separates those and therefore allows an accurate tracking of the anonymous
pages per zone.
It then becomes possible to determine the number of unmapped pages per zone
and we can avoid scanning for unmapped pages if there are none.
Also it may now be possible to determine the mapped/unmapped ratio in
get_dirty_limit. Isnt the number of anonymous pages irrelevant in that
calculation?
Note that this will change the meaning of the number of mapped pages reported
in /proc/vmstat /proc/meminfo and in the per node statistics. This may affect
user space tools that monitor these counters! NR_FILE_MAPPED works like
NR_FILE_DIRTY. It is only valid for pagecache pages.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently a single atomic variable is used to establish the size of the page
cache in the whole machine. The zoned VM counters have the same method of
implementation as the nr_pagecache code but also allow the determination of
the pagecache size per zone.
Remove the special implementation for nr_pagecache and make it a zoned counter
named NR_FILE_PAGES.
Updates of the page cache counters are always performed with interrupts off.
We can therefore use the __ variant here.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
nr_mapped is important because it allows a determination of how many pages of
a zone are not mapped, which would allow a more efficient means of determining
when we need to reclaim memory in a zone.
We take the nr_mapped field out of the page state structure and define a new
per zone counter named NR_FILE_MAPPED (the anonymous pages will be split off
from NR_MAPPED in the next patch).
We replace the use of nr_mapped in various kernel locations. This avoids the
looping over all processors in try_to_free_pages(), writeback, reclaim (swap +
zone reclaim).
[akpm@osdl.org: bugfix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Per zone counter infrastructure
The counters that we currently have for the VM are split per processor. The
processor however has not much to do with the zone these pages belong to. We
cannot tell f.e. how many ZONE_DMA pages are dirty.
So we are blind to potentially inbalances in the usage of memory in various
zones. F.e. in a NUMA system we cannot tell how many pages are dirty on a
particular node. If we knew then we could put measures into the VM to balance
the use of memory between different zones and different nodes in a NUMA
system. For example it would be possible to limit the dirty pages per node so
that fast local memory is kept available even if a process is dirtying huge
amounts of pages.
Another example is zone reclaim. We do not know how many unmapped pages exist
per zone. So we just have to try to reclaim. If it is not working then we
pause and try again later. It would be better if we knew when it makes sense
to reclaim unmapped pages from a zone. This patchset allows the determination
of the number of unmapped pages per zone. We can remove the zone reclaim
interval with the counters introduced here.
Futhermore the ability to have various usage statistics available will allow
the development of new NUMA balancing algorithms that may be able to improve
the decision making in the scheduler of when to move a process to another node
and hopefully will also enable automatic page migration through a user space
program that can analyse the memory load distribution and then rebalance
memory use in order to increase performance.
The counter framework here implements differential counters for each processor
in struct zone. The differential counters are consolidated when a threshold
is exceeded (like done in the current implementation for nr_pageache), when
slab reaping occurs or when a consolidation function is called.
Consolidation uses atomic operations and accumulates counters per zone in the
zone structure and also globally in the vm_stat array. VM functions can
access the counts by simply indexing a global or zone specific array.
The arrangement of counters in an array also simplifies processing when output
has to be generated for /proc/*.
Counters can be updated by calling inc/dec_zone_page_state or
_inc/dec_zone_page_state analogous to *_page_state. The second group of
functions can be called if it is known that interrupts are disabled.
Special optimized increment and decrement functions are provided. These can
avoid certain checks and use increment or decrement instructions that an
architecture may provide.
We also add a new CONFIG_DMA_IS_NORMAL that signifies that an architecture can
do DMA to all memory and therefore ZONE_NORMAL will not be populated. This is
only currently set for IA64 SGI SN2 and currently only affects
node_page_state(). In the best case node_page_state can be reduced to
retrieving a single counter for the one zone on the node.
[akpm@osdl.org: cleanups]
[akpm@osdl.org: export vm_stat[] for filesystems]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Record the node id as we mark sections for instantiation. Use this nid
during instantiation to direct allocations.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Mike Kravetz <kravetz@us.ibm.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Bob Picco <bob.picco@hp.com>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Martin Bligh <mbligh@google.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mel Gorman