Mel Gorman
599d0c954f
mm, vmscan: move LRU lists to node ...
This moves the LRU lists from the zone to the node and related data such
as counters, tracing, congestion tracking and writeback tracking.
Unfortunately, due to reclaim and compaction retry logic, it is
necessary to account for the number of LRU pages on both zone and node
logic. Most reclaim logic is based on the node counters but the retry
logic uses the zone counters which do not distinguish inactive and
active sizes. It would be possible to leave the LRU counters on a
per-zone basis but it's a heavier calculation across multiple cache
lines that is much more frequent than the retry checks.
Other than the LRU counters, this is mostly a mechanical patch but note
that it introduces a number of anomalies. For example, the scans are
per-zone but using per-node counters. We also mark a node as congested
when a zone is congested. This causes weird problems that are fixed
later but is easier to review.
In the event that there is excessive overhead on 32-bit systems due to
the nodes being on LRU then there are two potential solutions
1. Long-term isolation of highmem pages when reclaim is lowmem
When pages are skipped, they are immediately added back onto the LRU
list. If lowmem reclaim persisted for long periods of time, the same
highmem pages get continually scanned. The idea would be that lowmem
keeps those pages on a separate list until a reclaim for highmem pages
arrives that splices the highmem pages back onto the LRU. It potentially
could be implemented similar to the UNEVICTABLE list.
That would reduce the skip rate with the potential corner case is that
highmem pages have to be scanned and reclaimed to free lowmem slab pages.
2. Linear scan lowmem pages if the initial LRU shrink fails
This will break LRU ordering but may be preferable and faster during
memory pressure than skipping LRU pages.
Link: http://lkml.kernel.org/r/1467970510-21195-4-git-send-email-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-28 16:07:41 -07:00
wireless-drivers-next patches for 4.8
2016-07-25 11:09:19 -07:00
dmaengine: hsu: Export hsu_dma_get_status()
2016-06-25 14:30:42 -07:00
extcon: adc-jack: add suspend/resume support
2016-07-02 14:31:34 +09:00
HSI: core: switch port event notifier from atomic to blocking
2016-06-28 00:38:53 +02:00
iio: st_sensors: harden interrupt handling
2016-07-02 20:40:15 +01:00
Input: of_touchscreen - add support for inverted / swapped axes
2016-07-15 14:50:53 -07:00
irqchip/gicv3-its: Implement two-level(indirect) device table support
2016-06-13 11:53:53 +01:00
Merge remote-tracking branches 'regulator/topic/qcom-spmi', 'regulator/topic/rn5t618', 'regulator/topic/tps65218' and 'regulator/topic/twl' into regulator-next
2016-07-20 18:02:08 +01:00
net/mlx4_en: break out tx_desc write into separate function
2016-07-19 21:46:33 -07:00
net/mlx5e: Query minimum required header copy during xmit
2016-07-25 17:53:40 -07:00
netfilter: x_tables: speed up jump target validation
2016-07-18 21:35:23 +02:00
netfilter: Convert FWINV<[foo]> macros and uses to NF_INVF
2016-07-03 10:55:07 +02:00
phy: Add set_mode callback
2016-07-04 17:19:22 +05:30
Merge branch 'i2c/for-4.8' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
2016-07-27 14:19:25 -07:00
max8903: cleans up confusing relationship between dc_valid, dok and dcm.
2016-06-28 20:05:41 +02:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
2016-06-30 05:03:36 -04:00
Merge remote-tracking branches 'regulator/topic/fixed', 'regulator/topic/headers', 'regulator/topic/lp837x', 'regulator/topic/max8973' and 'regulator/topic/mt6323' into regulator-next
2016-07-20 18:02:03 +01:00
spi: Split bus and I/O locking
2016-07-25 11:47:52 +01:00
rpc: share one xps between all backchannels
2016-06-15 10:32:25 -04:00
Merge 4.7-rc6 into usb-next
2016-07-04 08:19:21 -07:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
2016-07-27 12:03:20 -07:00
alarmtimer: Fix comments describing structure fields
2016-06-20 12:47:09 -07:00
ata: define ATA_PROT_* in terms of ATA_PROT_FLAG_*
2016-07-18 20:55:38 -04:00
ath9k: Allow configuration of LED polarity in platform data.
2016-07-08 17:01:14 +03:00
locking/atomic: Introduce inc/dec variants for the atomic_fetch_$op() API
2016-07-07 09:16:20 +02:00
Merge branch 'stable-4.7' of git://git.infradead.org/users/pcmoore/audit
2016-06-29 15:18:47 -07:00
mm, vmscan: move LRU lists to node
2016-07-28 16:07:41 -07:00
mm: fix build warnings in <linux/compaction.h>
2016-07-26 16:19:19 -07:00
Merge branch 'for-4.8/drivers' of git://git.kernel.dk/linux-block
2016-07-26 15:37:51 -07:00
bitmap: bitmap_equal memcmp optimization
2016-06-13 15:58:21 +02:00
blk-mq: Introduce blk_mq_reinit_tagset
2016-07-08 08:38:49 -06:00
Merge branch 'for-4.8/drivers' of git://git.kernel.dk/linux-block
2016-07-26 15:37:51 -07:00
Merge branch 'for-4.8/drivers' of git://git.kernel.dk/linux-block
2016-07-26 15:37:51 -07:00
bpf, events: fix offset in skb copy handler
2016-07-25 10:34:11 -07:00
Merge branch 'for-4.8/core' of git://git.kernel.dk/linux-block
2016-07-26 15:03:07 -07:00
[media] cec-funcs.h: add missing 'reply' for short audio descriptor
2016-07-08 18:35:42 -03:00
[media] cec: split the timestamp into an rx and tx timestamp
2016-07-13 07:59:30 -03:00
cgroup: Add cgroup_get_from_fd
2016-07-01 16:30:38 -04:00
clk: Add missing clk_get_sys() stub
2016-06-28 10:22:02 +02:00
clocksources: Switch back to the clksrc table
2016-06-28 10:19:35 +02:00
mm: fix build warnings in <linux/compaction.h>
2016-07-26 16:19:19 -07:00
locking/barriers: Move smp_cond_load_acquire() to asm-generic/barrier.h
2016-06-14 11:55:14 +02:00
tty: vt, remove consw->con_bmove
2016-06-25 09:04:48 -07:00
tty: vt, remove unused vc_deccolm
2016-06-25 09:04:48 -07:00
x86/entry: Avoid interrupt flag save and restore
2016-07-10 13:33:02 +02:00
cpufreq: add cpufreq_driver_resolve_freq()
2016-07-21 14:46:08 +02:00
cpuidle: introduce CPU_PM_CPU_IDLE_ENTER macro for ARM{32, 64}
2016-07-21 23:29:38 +02:00
crypto: skcipher - Remove top-level givcipher interface
2016-07-18 17:35:46 +08:00
dax: remote unused fault wrappers
2016-07-26 16:19:19 -07:00
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
2016-06-20 10:41:51 -07:00
debugobjects.h: fix trivial kernel doc warning
2016-07-26 16:19:19 -07:00
dm: add infrastructure for DAX support
2016-07-20 23:49:49 -04:00
drbd: correctly handle failed crypto_alloc_hash
2016-06-13 21:43:08 -06:00
drbd: when receiving P_TRIM, zero-out partial unaligned chunks
2016-06-13 21:43:05 -06:00
drbd: allow larger max_discard_sectors
2016-06-13 21:43:05 -06:00
efi: Reorganize the GUID table to make it easier to read
2016-07-08 15:21:37 +02:00
block: do not merge requests without consulting with io scheduler
2016-07-20 21:35:12 -06:00
etherdevice.h & bridge: netfilter: Add and use ether_addr_equal_masked
2016-07-01 16:37:06 +02:00
fs: move struct iomap from exportfs.h to a separate header
2016-06-21 09:22:39 +10:00
extcon: Fix the wrong description about extcon_set/get_cable_state_()
2016-06-29 19:50:43 +09:00
staging/android: bring struct sync_pt back
2016-06-17 21:17:40 -07:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
2016-07-24 00:53:32 -04:00
mm, frontswap: convert frontswap_enabled to static key
2016-07-26 16:19:19 -07:00
Merge branch 'akpm' (patches from Andrew)
2016-07-26 19:55:54 -07:00
block: remove ->driverfs_dev
2016-06-27 12:26:08 -07:00
mm: charge/uncharge kmemcg from generic page allocator paths
2016-07-26 16:19:19 -07:00
thp: extract khugepaged from mm/huge_memory.c
2016-07-26 16:19:19 -07:00
i2c: i2c-smbus: drop useless stubs
2016-07-22 09:07:02 +02:00
Revert "i2c: core: Add function for finding the bus speed from ACPI"
2016-07-19 05:57:23 +02:00
ipv6: RFC 4884 partial support for SIT/GRE tunnels
2016-06-18 22:11:39 -07:00
mac80211: Encrypt "Group addressed privacy" action frames
2016-06-30 12:06:20 +02:00
ieee802154: add ieee802154_skb_src_pan helper
2016-07-08 13:23:12 +02:00
fix up initial thread stack pointer vs thread_info confusion
2016-06-24 17:07:33 -07:00
Input: fix a double word "is is" in include/linux/input.h
2016-07-13 17:27:24 -07:00
genirq: Add a helper to spread an affinity mask for MSI/MSI-X vectors
2016-07-04 12:25:14 +02:00
fs: iomap based fiemap implementation
2016-06-21 09:38:45 +10:00
ipmi: remove trydefaults parameter and default init
2016-07-27 10:24:38 -05:00
Merge branch 'irq/for-block' into irq/core
2016-07-04 12:26:05 +02:00
Merge branch 'irq/for-block' into irq/core
2016-07-04 12:26:05 +02:00
isa: Dummy isa_register_driver should return error code
2016-06-17 20:47:11 -07:00
jbd2: track more dependencies on transaction commit
2016-06-30 11:40:54 -04:00
locking/static_key: Fix concurrent static_key_slow_inc()
2016-06-24 08:23:16 +02:00
mm: mempool: kasan: don't poot mempool objects in quarantine
2016-06-24 17:23:52 -07:00
rcu: sysctl: Panic on RCU Stall
2016-06-15 16:00:05 -07:00
xen: add steal_clock support on x86
2016-07-06 10:34:48 +01:00
thp: extract khugepaged from mm/huge_memory.c
2016-07-26 16:19:19 -07:00
mm: migrate: support non-lru movable page migration
2016-07-26 16:19:19 -07:00
leds: LED driver for TI LP3952 6-Channel Color LED
2016-07-18 08:46:09 +02:00
leds: pca9532: Add device tree support
2016-06-20 09:43:33 +02:00
leds: Add no-op gpio_led_register_device when LED subsystem is disabled
2016-06-27 08:58:42 +02:00
ata: define ATA_PROT_* in terms of ATA_PROT_FLAG_*
2016-07-18 20:55:38 -04:00
lightnvm: make ppa_list const in nvm_set_rqd_list
2016-07-07 08:51:52 -06:00
hlist: Add hlist_is_singular_node() helper
2016-07-07 10:35:07 +02:00
include/linux/memblock.h: Clean up code for several trivial details
2016-07-26 16:19:19 -07:00
mm, vmscan: move LRU lists to node
2016-07-28 16:07:41 -07:00
memory-hotplug: more general validation of zone during online
2016-07-26 16:19:19 -07:00
net: phy: micrel: Add KSZ8041FTL fiber mode support
2016-07-15 11:53:02 -07:00
mm: fix build warnings in <linux/compaction.h>
2016-07-26 16:19:19 -07:00
mm, vmstat: add infrastructure for per-node vmstats
2016-07-28 16:07:41 -07:00
mm, vmscan: move LRU lists to node
2016-07-28 16:07:41 -07:00
mm, vmscan: move lru_lock to the node
2016-07-28 16:07:41 -07:00
include/linux/mmdebug.h: add VM_WARN which maps to WARN()
2016-07-26 16:19:19 -07:00
mm, vmscan: move LRU lists to node
2016-07-28 16:07:41 -07:00
staging: fsl-mc: add support for device table matching
2016-06-26 17:14:03 -07:00
crypto: rsa - Generate fixed-length output
2016-07-01 23:45:18 +08:00
net: ipmr/ip6mr: add support for keeping an entry age
2016-07-16 20:19:43 -07:00
net: ipmr/ip6mr: add support for keeping an entry age
2016-07-16 20:19:43 -07:00
genirq/msi: Make use of affinity aware allocations
2016-07-04 12:25:14 +02:00
tun: switch to use skb array for tx
2016-07-01 05:32:17 -04:00
net/sched: Add match-all classifier hw offloading.
2016-07-24 23:11:59 -07:00
Revert "perf/x86/intel, watchdog: Switch NMI watchdog to ref cycles on x86"
2016-07-10 20:58:36 +02:00
nvme-rdma.h: Add includes for nvme rdma_cm negotiation
2016-07-08 08:38:49 -06:00
nvme.h: Add keep-alive opcode and identify controller attribute
2016-07-05 11:28:18 -06:00
nvmem: Declare nvmem_cell_read() consistently
2016-06-25 07:42:55 -07:00
Merge branch 'clockevents/4.8' of http://git.linaro.org/people/daniel.lezcano/linux into timers/core
2016-07-07 15:41:13 +02:00
FDT: Add a helper to get the subnode by given name
2016-07-06 10:34:46 +01:00
of_mdio: Abstract a general interface for phy connect
2016-07-16 21:32:58 -07:00
Linux 4.7-rc6
2016-07-08 18:14:03 -03:00
mm, oom: fortify task_will_free_mem()
2016-07-28 16:07:41 -07:00
page-flags: relax policy for PG_mappedtodisk and PG_reclaim
2016-07-26 16:19:19 -07:00
mm/page_owner: use stackdepot to store stacktrace
2016-07-26 16:19:19 -07:00
mm/page_owner: introduce split_page_owner and replace manual handling
2016-07-26 16:19:19 -07:00
mm, memcg: use consistent gfp flags during readahead
2016-07-26 16:19:19 -07:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
2016-07-27 12:03:20 -07:00
PM / clk: Add support for adding a specific clock from device-tree
2016-06-28 00:42:10 +02:00
PM / Domains: Convert pm_genpd_init() to return an error code
2016-06-29 02:15:19 +02:00
posix_acl: de-union a_refcount and a_rcu
2016-07-11 13:48:02 -04:00
power_supply: fix return value of get_property
2016-07-01 22:44:34 +02:00
printk: Make the printk*once() variants return a value
2016-07-08 11:33:18 +02:00
ptr_ring: support resizing multiple queues
2016-07-01 05:32:17 -04:00
pwm: Fix pwm_apply_args()
2016-06-23 18:37:05 +02:00
radix-tree: implement radix_tree_maybe_preload_order()
2016-07-26 16:19:19 -07:00
Introduce rb_replace_node_rcu()
2016-07-06 10:51:14 +01:00
Introduce rb_replace_node_rcu()
2016-07-06 10:51:14 +01:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
2016-07-27 12:03:20 -07:00
Merge remote-tracking branches 'regmap/topic/bulk', 'regmap/topic/i2c', 'regmap/topic/iopoll', 'regmap/topic/irq' and 'regmap/topic/maintainers' into regmap-next
2016-07-15 13:44:47 +01:00
reset: TRIVIAL: Add line break at same place for similar APIs
2016-06-29 21:24:53 +02:00
rmap: support file thp
2016-07-26 16:19:19 -07:00
net_sched: add the ability to defer skb freeing
2016-06-15 14:08:34 -07:00
rxrpc: Fix exclusive connection handling
2016-06-22 09:10:00 +01:00
mm, oom_reaper: do not attempt to reap a task more than twice
2016-07-28 16:07:41 -07:00
tty/serial/8250: use mctrl_gpio helpers
2016-06-25 14:30:42 -07:00
earlycon: mark earlycon code as __used iif the caller is built-in
2016-06-25 14:31:04 -07:00
x86/sfi: Enable enumeration of SD devices
2016-07-13 09:24:51 +02:00
shmem: split huge pages beyond i_size under memory pressure
2016-07-26 16:19:19 -07:00
skb_array: add wrappers for resizing
2016-07-01 05:32:17 -04:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
2016-07-06 10:35:22 -07:00
mm: faster kmalloc_array(), kcalloc()
2016-07-26 16:19:19 -07:00
mm: reorganize SLAB freelist randomization
2016-07-26 16:19:19 -07:00
mm: SLUB freelist randomization
2016-07-26 16:19:19 -07:00
sock_diag: do not broadcast raw socket destruction
2016-06-28 09:08:51 -04:00
locking/spinlock, arch: Update and fix spin_unlock_wait() implementations
2016-06-14 11:55:15 +02:00
drivers: net: stmmac: add port selection programming
2016-06-28 08:54:23 -04:00
PM / sleep: Make pm_prepare_console() return void
2016-06-15 01:26:04 +02:00
mm, vmscan: move LRU lists to node
2016-07-28 16:07:41 -07:00
drivers: misc: ti-st: Use int instead of fuzzy char for callback status
2016-07-17 19:59:26 +02:00
time: Add time64_to_tm()
2016-06-20 12:47:15 -07:00
timers: Remove set_timer_slack() leftovers
2016-07-07 10:35:09 +02:00
torture: Break online and offline functions out of torture_onoff()
2016-06-14 16:02:16 -07:00
mm: introduce fault_env
2016-07-26 16:19:19 -07:00
mm, vmscan: move LRU lists to node
2016-07-28 16:07:41 -07:00
mm, vmscan: move LRU lists to node
2016-07-28 16:07:41 -07:00
tty: vt, ignore PIO_UNIMAPCLR param
2016-06-25 09:04:48 -07:00
sched/cputime: Reorganize vtime native irqtime accounting headers
2016-07-14 10:42:35 +02:00
fs/fs-writeback.c: add a new writeback list for sync
2016-07-26 16:19:19 -07:00
