[lkp] [mm] 089e25e502: -5.4% will-it-scale.per_process_ops
FYI, we noticed the below changes on
https://git.kernel.org/pub/scm/linux/kernel/git/andrea/aa.git master
commit 089e25e5022ed5a1bb821f0c7ca6b48781ecc122 ("mm: gup: make
get_user_pages_fast and __get_user_pages_fast latency conscious")
=
tbox_group/testcase/rootfs/kconfig/compiler/test:
lkp-sb03/will-it-scale/debian-x86_64-2015-02-07.cgz/x86_64-rhel/gcc-4.9/futex2
commit:
226c833d8e769b4bb1900eaffd3efb1cbb95f820
089e25e5022ed5a1bb821f0c7ca6b48781ecc122
226c833d8e769b4b 089e25e5022ed5a1bb821f0c7c
--
%stddev %change %stddev
\ |\
4287684 ± 0% -5.4%4055429 ± 0% will-it-scale.per_process_ops
1432530 ± 0% -3.9%1377184 ± 0% will-it-scale.per_thread_ops
0.60 ± 0% -1.8% 0.59 ± 0% will-it-scale.scalability
142663 ± 18%+171.2% 386864 ± 82% cpuidle.C3-SNB.time
220.25 ± 8% +49.0% 328.25 ± 14% cpuidle.C3-SNB.usage
24244125 ± 41% -91.9%1963754 ±108% cpuidle.POLL.time
663.50 ± 12% -77.8% 147.50 ±134% cpuidle.POLL.usage
344.25 ± 12% -29.6% 242.25 ± 17% slabinfo.kmem_cache.active_objs
344.25 ± 12% -29.6% 242.25 ± 17% slabinfo.kmem_cache.num_objs
474.00 ± 11% -27.0% 346.00 ± 15%
slabinfo.kmem_cache_node.active_objs
496.00 ± 10% -25.8% 368.00 ± 14% slabinfo.kmem_cache_node.num_objs
1.50 ± 33%+150.0% 3.75 ± 22%
sched_debug.cfs_rq[20]:/.nr_spread_over
7.00 ± 58%+285.7% 27.00 ± 35% sched_debug.cfs_rq[25]:/.load
151.50 ± 23% +34.8% 204.25 ± 1% sched_debug.cfs_rq[25]:/.util_avg
-764780 ±-13% -18.2%-625478 ± -1% sched_debug.cfs_rq[31]:/.spread0
12046 ± 8% -15.5% 10180 ± 14% sched_debug.cpu#1.nr_switches
7057 ± 8% -33.8% 4669 ± 23% sched_debug.cpu#10.ttwu_count
24588 ± 17% -31.8% 16770 ± 16% sched_debug.cpu#11.nr_switches
-2.50 ±-60%-120.0% 0.50 ±223%
sched_debug.cpu#11.nr_uninterruptible
25784 ± 16% -29.9% 18064 ± 16% sched_debug.cpu#11.sched_count
1536 ± 15%+124.0% 3441 ± 49% sched_debug.cpu#13.ttwu_local
-4.00 ± 0%-106.2% 0.25 ±1479%
sched_debug.cpu#14.nr_uninterruptible
2563 ± 42% -50.0% 1281 ± 12% sched_debug.cpu#16.nr_switches
2760 ± 39% -47.0% 1463 ± 10% sched_debug.cpu#16.sched_count
3.25 ±107%-192.3% -3.00 ±-54%
sched_debug.cpu#2.nr_uninterruptible
2635 ± 14% -27.0% 1923 ± 10% sched_debug.cpu#2.sched_goidle
957.25 ± 14% +44.7% 1385 ± 29% sched_debug.cpu#21.nr_switches
1140 ± 11% +37.3% 1566 ± 25% sched_debug.cpu#21.sched_count
3293 ± 64% -49.9% 1648 ± 74% sched_debug.cpu#23.nr_switches
4654 ± 43% -60.8% 1824 ± 66% sched_debug.cpu#23.sched_count
1595 ± 25% +79.0% 2856 ± 25% sched_debug.cpu#25.curr->pid
7.00 ± 58%+285.7% 27.00 ± 35% sched_debug.cpu#25.load
2.25 ±184%-355.6% -5.75 ±-31%
sched_debug.cpu#3.nr_uninterruptible
1.25 ±153%+320.0% 5.25 ± 15%
sched_debug.cpu#31.nr_uninterruptible
24696 ± 21% -36.6% 15662 ± 29% sched_debug.cpu#9.nr_switches
11398 ± 24% -43.2% 6474 ± 30% sched_debug.cpu#9.sched_goidle
lkp-sb03: Sandy Bridge-EP
Memory: 64G
To reproduce:
git clone
git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Ying Huang
---
LKP_SERVER: inn
LKP_CGI_PORT: 80
LKP_CIFS_PORT: 139
testcase: will-it-scale
default-monitors:
wait: activate-monitor
kmsg:
uptime:
iostat:
vmstat:
numa-numastat:
numa-vmstat:
numa-meminfo:
proc-vmstat:
proc-stat:
interval: 10
meminfo:
slabinfo:
interrupts:
lock_stat:
latency_stats:
softirqs:
bdi_dev_mapping:
diskstats:
nfsstat:
cpuidle:
cpufreq-stats:
turbostat:
pmeter:
sched_debug:
interval: 60
cpufreq_governor:
default-watchdogs:
oom-killer:
watchdog:
commit: b58730ac9d33c9b2a59117f4b3f3d83368ff34f1
model: Sandy Bridge-EP
memory: 64G
hdd_partitions: "/dev/disk/by-id/ata-WDC_WD10EARS-00Y5B1_WD-WCAV5F059499-part3"
swap_partitions:
rootfs_partition:
"/dev/disk/by-id/ata-WDC_WD10EARS-00Y5B1_WD-WCAV5F059499-part4"
category: benchmark
perf-profile:
freq: 800
will-it-scale:
test: futex2
queue: cyclic
testbox: lkp-sb03
tbox_group: lkp-sb03
kconfig: x86_64-rhel
enqueue_time:
[lkp] [mm] 089e25e502: -5.4% will-it-scale.per_process_ops
FYI, we noticed the below changes on
https://git.kernel.org/pub/scm/linux/kernel/git/andrea/aa.git master
commit 089e25e5022ed5a1bb821f0c7ca6b48781ecc122 ("mm: gup: make
get_user_pages_fast and __get_user_pages_fast latency conscious")
=
tbox_group/testcase/rootfs/kconfig/compiler/test:
lkp-sb03/will-it-scale/debian-x86_64-2015-02-07.cgz/x86_64-rhel/gcc-4.9/futex2
commit:
226c833d8e769b4bb1900eaffd3efb1cbb95f820
089e25e5022ed5a1bb821f0c7ca6b48781ecc122
226c833d8e769b4b 089e25e5022ed5a1bb821f0c7c
--
%stddev %change %stddev
\ |\
4287684 ± 0% -5.4%4055429 ± 0% will-it-scale.per_process_ops
1432530 ± 0% -3.9%1377184 ± 0% will-it-scale.per_thread_ops
0.60 ± 0% -1.8% 0.59 ± 0% will-it-scale.scalability
142663 ± 18%+171.2% 386864 ± 82% cpuidle.C3-SNB.time
220.25 ± 8% +49.0% 328.25 ± 14% cpuidle.C3-SNB.usage
24244125 ± 41% -91.9%1963754 ±108% cpuidle.POLL.time
663.50 ± 12% -77.8% 147.50 ±134% cpuidle.POLL.usage
344.25 ± 12% -29.6% 242.25 ± 17% slabinfo.kmem_cache.active_objs
344.25 ± 12% -29.6% 242.25 ± 17% slabinfo.kmem_cache.num_objs
474.00 ± 11% -27.0% 346.00 ± 15%
slabinfo.kmem_cache_node.active_objs
496.00 ± 10% -25.8% 368.00 ± 14% slabinfo.kmem_cache_node.num_objs
1.50 ± 33%+150.0% 3.75 ± 22%
sched_debug.cfs_rq[20]:/.nr_spread_over
7.00 ± 58%+285.7% 27.00 ± 35% sched_debug.cfs_rq[25]:/.load
151.50 ± 23% +34.8% 204.25 ± 1% sched_debug.cfs_rq[25]:/.util_avg
-764780 ±-13% -18.2%-625478 ± -1% sched_debug.cfs_rq[31]:/.spread0
12046 ± 8% -15.5% 10180 ± 14% sched_debug.cpu#1.nr_switches
7057 ± 8% -33.8% 4669 ± 23% sched_debug.cpu#10.ttwu_count
24588 ± 17% -31.8% 16770 ± 16% sched_debug.cpu#11.nr_switches
-2.50 ±-60%-120.0% 0.50 ±223%
sched_debug.cpu#11.nr_uninterruptible
25784 ± 16% -29.9% 18064 ± 16% sched_debug.cpu#11.sched_count
1536 ± 15%+124.0% 3441 ± 49% sched_debug.cpu#13.ttwu_local
-4.00 ± 0%-106.2% 0.25 ±1479%
sched_debug.cpu#14.nr_uninterruptible
2563 ± 42% -50.0% 1281 ± 12% sched_debug.cpu#16.nr_switches
2760 ± 39% -47.0% 1463 ± 10% sched_debug.cpu#16.sched_count
3.25 ±107%-192.3% -3.00 ±-54%
sched_debug.cpu#2.nr_uninterruptible
2635 ± 14% -27.0% 1923 ± 10% sched_debug.cpu#2.sched_goidle
957.25 ± 14% +44.7% 1385 ± 29% sched_debug.cpu#21.nr_switches
1140 ± 11% +37.3% 1566 ± 25% sched_debug.cpu#21.sched_count
3293 ± 64% -49.9% 1648 ± 74% sched_debug.cpu#23.nr_switches
4654 ± 43% -60.8% 1824 ± 66% sched_debug.cpu#23.sched_count
1595 ± 25% +79.0% 2856 ± 25% sched_debug.cpu#25.curr->pid
7.00 ± 58%+285.7% 27.00 ± 35% sched_debug.cpu#25.load
2.25 ±184%-355.6% -5.75 ±-31%
sched_debug.cpu#3.nr_uninterruptible
1.25 ±153%+320.0% 5.25 ± 15%
sched_debug.cpu#31.nr_uninterruptible
24696 ± 21% -36.6% 15662 ± 29% sched_debug.cpu#9.nr_switches
11398 ± 24% -43.2% 6474 ± 30% sched_debug.cpu#9.sched_goidle
lkp-sb03: Sandy Bridge-EP
Memory: 64G
To reproduce:
git clone
git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
Thanks,
Ying Huang
---
LKP_SERVER: inn
LKP_CGI_PORT: 80
LKP_CIFS_PORT: 139
testcase: will-it-scale
default-monitors:
wait: activate-monitor
kmsg:
uptime:
iostat:
vmstat:
numa-numastat:
numa-vmstat:
numa-meminfo:
proc-vmstat:
proc-stat:
interval: 10
meminfo:
slabinfo:
interrupts:
lock_stat:
latency_stats:
softirqs:
bdi_dev_mapping:
diskstats:
nfsstat:
cpuidle:
cpufreq-stats:
turbostat:
pmeter:
sched_debug:
interval: 60
cpufreq_governor:
default-watchdogs:
oom-killer:
watchdog:
commit: b58730ac9d33c9b2a59117f4b3f3d83368ff34f1
model: Sandy Bridge-EP
memory: 64G
hdd_partitions: "/dev/disk/by-id/ata-WDC_WD10EARS-00Y5B1_WD-WCAV5F059499-part3"
swap_partitions:
rootfs_partition:
"/dev/disk/by-id/ata-WDC_WD10EARS-00Y5B1_WD-WCAV5F059499-part4"
category: benchmark
perf-profile:
freq: 800
will-it-scale:
test: futex2
queue: cyclic
testbox: lkp-sb03
tbox_group: lkp-sb03
kconfig: x86_64-rhel
enqueue_time:
