linux/kernel
Mel Gorman 5e1f0f098b mm, compaction: capture a page under direct compaction
Compaction is inherently race-prone as a suitable page freed during
compaction can be allocated by any parallel task.  This patch uses a
capture_control structure to isolate a page immediately when it is freed
by a direct compactor in the slow path of the page allocator.  The
intent is to avoid redundant scanning.

                                     5.0.0-rc1              5.0.0-rc1
                               selective-v3r17          capture-v3r19
Amean     fault-both-1         0.00 (   0.00%)        0.00 *   0.00%*
Amean     fault-both-3      2582.11 (   0.00%)     2563.68 (   0.71%)
Amean     fault-both-5      4500.26 (   0.00%)     4233.52 (   5.93%)
Amean     fault-both-7      5819.53 (   0.00%)     6333.65 (  -8.83%)
Amean     fault-both-12     9321.18 (   0.00%)     9759.38 (  -4.70%)
Amean     fault-both-18     9782.76 (   0.00%)    10338.76 (  -5.68%)
Amean     fault-both-24    15272.81 (   0.00%)    13379.55 *  12.40%*
Amean     fault-both-30    15121.34 (   0.00%)    16158.25 (  -6.86%)
Amean     fault-both-32    18466.67 (   0.00%)    18971.21 (  -2.73%)

Latency is only moderately affected but the devil is in the details.  A
closer examination indicates that base page fault latency is reduced but
latency of huge pages is increased as it takes creater care to succeed.
Part of the "problem" is that allocation success rates are close to 100%
even when under pressure and compaction gets harder

                                5.0.0-rc1              5.0.0-rc1
                          selective-v3r17          capture-v3r19
Percentage huge-3        96.70 (   0.00%)       98.23 (   1.58%)
Percentage huge-5        96.99 (   0.00%)       95.30 (  -1.75%)
Percentage huge-7        94.19 (   0.00%)       97.24 (   3.24%)
Percentage huge-12       94.95 (   0.00%)       97.35 (   2.53%)
Percentage huge-18       96.74 (   0.00%)       97.30 (   0.58%)
Percentage huge-24       97.07 (   0.00%)       97.55 (   0.50%)
Percentage huge-30       95.69 (   0.00%)       98.50 (   2.95%)
Percentage huge-32       96.70 (   0.00%)       99.27 (   2.65%)

And scan rates are reduced as expected by 6% for the migration scanner
and 29% for the free scanner indicating that there is less redundant
work.

Compaction migrate scanned    20815362    19573286
Compaction free scanned       16352612    11510663

[mgorman@techsingularity.net: remove redundant check]
  Link: http://lkml.kernel.org/r/20190201143853.GH9565@techsingularity.net
Link: http://lkml.kernel.org/r/20190118175136.31341-23-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: David Rientjes <rientjes@google.com>
Cc: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-03-05 21:07:17 -08:00
..
bpf Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next 2019-03-04 10:14:31 -08:00
cgroup Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2019-03-05 08:26:13 -08:00
configs kvm_config: add CONFIG_VIRTIO_MENU 2018-10-24 20:55:56 -04:00
debug kdb: use bool for binary state indicators 2018-12-30 08:31:52 +00:00
dma swiotlb: clear io_tlb_start and io_tlb_end in swiotlb_exit 2019-01-16 09:59:17 -05:00
events perf/core: Fix impossible ring-buffer sizes warning 2019-02-13 08:05:02 +01:00
gcov
irq genirq/irqdesc: Fix double increment in alloc_descs() 2019-01-18 00:43:09 +01:00
livepatch livepatch: Replace synchronize_sched() with synchronize_rcu() 2018-12-01 12:38:50 -08:00
locking futex: Handle early deadlock return correctly 2019-02-08 13:00:36 +01:00
power PM/Hibernate: exclude all PageOffline() pages 2019-03-05 21:07:14 -08:00
printk Remove 'type' argument from access_ok() function 2019-01-03 18:57:57 -08:00
rcu rcutorture: Don't do busted forward-progress testing 2018-12-01 12:45:42 -08:00
sched mm, compaction: capture a page under direct compaction 2019-03-05 21:07:17 -08:00
time posix-cpu-timers: Unbreak timer rearming 2019-01-15 16:34:37 +01:00
trace Two more tracing fixes 2019-02-18 09:40:16 -08:00
.gitignore
acct.c
async.c
audit_fsnotify.c audit: minimize our use of audit_log_format() 2018-11-26 18:40:00 -05:00
audit_tree.c audit: minimize our use of audit_log_format() 2018-11-26 18:40:00 -05:00
audit_watch.c audit: minimize our use of audit_log_format() 2018-11-26 18:40:00 -05:00
audit.c audit: remove duplicated include from audit.c 2018-12-14 12:09:30 -05:00
audit.h audit: use current whenever possible 2018-11-26 18:41:21 -05:00
auditfilter.c
auditsc.c audit: use current whenever possible 2018-11-26 18:41:21 -05:00
backtracetest.c
bounds.c kbuild: fix kernel/bounds.c 'W=1' warning 2018-10-31 08:54:14 -07:00
capability.c
compat.c make 'user_access_begin()' do 'access_ok()' 2019-01-04 12:56:09 -08:00
configs.c
context_tracking.c
cpu_pm.c
cpu.c cpu/hotplug: Fix "SMT disabled by BIOS" detection for KVM 2019-01-30 19:27:00 +01:00
crash_core.c kexec: export PG_offline to VMCOREINFO 2019-03-05 21:07:14 -08:00
crash_dump.c
cred.c cred: export get_task_cred(). 2018-12-19 13:52:44 -05:00
delayacct.c delayacct: track delays from thrashing cache pages 2018-10-26 16:26:32 -07:00
dma.c
elfcore.c
exec_domain.c
exit.c kernel/exit.c: release ptraced tasks before zap_pid_ns_processes 2019-02-01 15:46:23 -08:00
extable.c
fail_function.c kernel/fail_function.c: remove meaningless null pointer check before debugfs_remove_recursive 2018-10-31 08:54:12 -07:00
fork.c Merge branch 'akpm' (patches from Andrew) 2019-01-08 18:58:29 -08:00
freezer.c PM / reboot: Eliminate race between reboot and suspend 2018-08-06 12:35:20 +02:00
futex.c futex: Handle early deadlock return correctly 2019-02-08 13:00:36 +01:00
groups.c
hung_task.c kernel/hung_task.c: break RCU locks based on jiffies 2019-01-04 13:13:45 -08:00
iomem.c
irq_work.c
jump_label.c jump_label: move 'asm goto' support test to Kconfig 2019-01-06 09:46:51 +09:00
kallsyms.c kallsyms: reduce size a little on 64-bit 2018-09-10 22:54:33 +09:00
kcmp.c
Kconfig.freezer
Kconfig.hz
Kconfig.locks bpf: introduce bpf_spin_lock 2019-02-01 20:55:38 +01:00
Kconfig.preempt kconfig: warn no new line at end of file 2018-12-15 17:44:35 +09:00
kcov.c kernel/kcov.c: mark write_comp_data() as notrace 2019-01-04 13:13:47 -08:00
kexec_core.c mm: convert totalram_pages and totalhigh_pages variables to atomic 2018-12-28 12:11:47 -08:00
kexec_file.c kexec_file: kexec_walk_memblock() only walks a dedicated region at kdump 2018-12-06 14:38:50 +00:00
kexec_internal.h
kexec.c kexec: add call to LSM hook in original kexec_load syscall 2018-07-16 12:31:57 -07:00
kmod.c
kprobes.c Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2018-12-26 14:45:18 -08:00
ksysfs.c
kthread.c mm: replace all open encodings for NUMA_NO_NODE 2019-03-05 21:07:14 -08:00
latencytop.c
Makefile kbuild: change filechk to surround the given command with { } 2019-01-06 09:46:51 +09:00
memremap.c mm/hmm: fix memremap.h, move dev_page_fault_t callback to hmm 2018-12-28 12:11:52 -08:00
module_signing.c modsign: use all trusted keys to verify module signature 2018-11-07 14:41:41 +01:00
module-internal.h
module.c jump_label: move 'asm goto' support test to Kconfig 2019-01-06 09:46:51 +09:00
notifier.c
nsproxy.c
padata.c padata: clean an indentation issue, remove extraneous space 2018-11-16 14:11:04 +08:00
panic.c kernel/sysctl: add panic_print into sysctl 2019-01-04 13:13:47 -08:00
params.c
pid_namespace.c signal: Use group_send_sig_info to kill all processes in a pid namespace 2018-09-16 16:08:25 +02:00
pid.c Fix failure path in alloc_pid() 2018-12-28 12:42:30 -08:00
profile.c mm: remove include/linux/bootmem.h 2018-10-31 08:54:16 -07:00
ptrace.c Remove 'type' argument from access_ok() function 2019-01-03 18:57:57 -08:00
range.c
reboot.c kernel/reboot.c: export pm_power_off_prepare 2018-09-11 16:13:24 +01:00
relay.c relay: check return of create_buf_file() properly 2019-01-31 14:01:48 +01:00
resource.c kernel, resource: check for IORESOURCE_SYSRAM in release_mem_region_adjustable 2018-12-28 12:11:49 -08:00
rseq.c Remove 'type' argument from access_ok() function 2019-01-03 18:57:57 -08:00
seccomp.c seccomp, bpf: disable preemption before calling into bpf prog 2019-02-22 00:14:19 +01:00
signal.c signal: Restore the stop PTRACE_EVENT_EXIT 2019-02-13 08:31:41 -06:00
smp.c cpu/hotplug: Fix "SMT disabled by BIOS" detection for KVM 2019-01-30 19:27:00 +01:00
smpboot.c
smpboot.h
softirq.c Merge branch 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2018-10-25 11:43:47 -07:00
stackleak.c stackleak: Mark stackleak_track_stack() as notrace 2018-12-05 19:31:44 -08:00
stacktrace.c
stop_machine.c Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2018-08-13 11:25:07 -07:00
sys_ni.c ipc: introduce ksys_ipc()/compat_ksys_ipc() for s390 2019-01-18 09:33:18 +01:00
sys.c kernel/sys.c: Clarify that UNAME26 does not generate unique versions anymore 2019-01-14 10:38:03 +12:00
sysctl_binary.c kernel/sysctl: add panic_print into sysctl 2019-01-04 13:13:47 -08:00
sysctl.c mm: remove sysctl_extfrag_handler() 2019-03-05 21:07:15 -08:00
task_work.c
taskstats.c
test_kprobes.c
torture.c torture: Remove unnecessary "ret" variables 2018-12-01 12:45:35 -08:00
tracepoint.c tracing: Replace synchronize_sched() and call_rcu_sched() 2018-11-27 09:21:41 -08:00
tsacct.c
ucount.c
uid16.c
uid16.h
umh.c umh: add exit routine for UMH process 2019-01-11 18:05:40 -08:00
up.c smp,cpumask: introduce on_each_cpu_cond_mask 2018-10-09 16:51:11 +02:00
user_namespace.c userns: also map extents in the reverse map to kernel IDs 2018-11-07 23:51:16 -06:00
user-return-notifier.c
user.c userns: use irqsave variant of refcount_dec_and_lock() 2018-08-22 10:52:47 -07:00
utsname_sysctl.c sys: don't hold uts_sem while accessing userspace memory 2018-08-11 02:05:53 -05:00
utsname.c
watchdog_hld.c watchdog: Mark watchdog touch functions as notrace 2018-08-30 12:56:40 +02:00
watchdog.c watchdog: Mark watchdog touch functions as notrace 2018-08-30 12:56:40 +02:00
workqueue_internal.h psi: fix aggregation idle shut-off 2019-02-01 15:46:23 -08:00
workqueue.c psi: fix aggregation idle shut-off 2019-02-01 15:46:23 -08:00