All of lore.kernel.org
 help / color / mirror / Atom feed
* [lkp-robot] [mm, page_alloc] d34b0733b4: will-it-scale.per_thread_ops -4% regression
@ 2017-05-12  4:47 kernel test robot
  2017-05-12  9:24 ` Mel Gorman
  0 siblings, 1 reply; 3+ messages in thread
From: kernel test robot @ 2017-05-12  4:47 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 5592 bytes --]


Greeting,

We noticed a -4% regression of will-it-scale.per_thread_ops due to commit:

commit: d34b0733b452ca3ef1dee36436dab08b8aa6a85c ("Revert "mm, page_alloc: only use per-cpu allocator for irq-safe requests"")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

in testcase: will-it-scale
on test machine: 256 threads Phi with 96G memory
with following parameters:

	nr_task: 100%
	mode: thread
	test: page_fault2
	cpufreq_governor: performance

test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale



Details are as below:
-------------------------------------------------------------------------------------------------->


To reproduce:

        git clone https://github.com/01org/lkp-tests.git
        cd lkp-tests
        bin/lkp install job.yaml  # job file is attached in this email
        bin/lkp run     job.yaml

testcase/path_params/tbox_group/run: will-it-scale/100%-thread-page_fault2-performance/lkp-knl-f1

f61143c45077df4f  d34b0733b452ca3ef1dee36436  
----------------  --------------------------  
         %stddev      change         %stddev
             \          |                \  
      2363              -4%       2278        will-it-scale.per_thread_ops
      9383              -3%       9069        will-it-scale.time.percent_of_cpu_this_job_got
 1.845e+08              -3%  1.782e+08        will-it-scale.time.minor_page_faults
     29880              -4%      28713        will-it-scale.time.system_time
   2085066 ±  4%        -5%    1981442        interrupts.CAL:Function_call_interrupts
          1:4          -25%            :4     stderr./lkp/lkp/src/monitors/wrapper:#:/lkp/lkp/src/monitors/wrapper:cannot_create/tmp/lkp/fifo-kmsg:Interrupted_system_call
 1.852e+08              -3%  1.789e+08        perf-stat.page-faults
 1.852e+08              -3%  1.789e+08        perf-stat.minor-faults
 6.039e+10              -3%  5.832e+10        perf-stat.cache-references
 1.136e+10              -4%  1.091e+10        perf-stat.iTLB-load-misses
 4.297e+13              -4%  4.122e+13        perf-stat.cpu-cycles
 4.623e+12              -4%  4.431e+12        perf-stat.instructions
 4.624e+12              -4%  4.428e+12        perf-stat.iTLB-loads
 1.171e+10              -4%   1.12e+10        perf-stat.cache-misses
 1.658e+12              -4%  1.583e+12        perf-stat.branch-instructions
      0.83              -7%       0.77        perf-stat.branch-miss-rate%
 1.369e+10             -11%  1.219e+10        perf-stat.branch-misses



                           will-it-scale.per_thread_ops

  2380 ++-------------------------------------------------------------------+
       |      .*..*                   .*..*.       .*..*.               .*..|
  2360 *+. .*.     +  .*..    .*.. .*.      *.. .*.      *..*..*.*..*..*    *
       |  *         *.    *.*.    *            *                            |
  2340 ++                                                                   |
       |                                                                    |
  2320 ++                                                                   |
       |                                       O                            |
  2300 ++                                                                   |
       |                                                                    |
  2280 O+   O     O                              O                          |
       |  O    O          O O  O  O    O    O       O                       |
  2260 ++           O  O            O     O            O                    |
       |                                                                    |
  2240 ++-------------------------------------------------------------------+


                           will-it-scale.time.system_time

  30400 ++------------------------------------------------------------------+
  30200 *+.                     *.           *.                             |
        |  *                   :  *..       +  *..             *..*         |
  30000 ++  +          .*.     :           +             .*. ..    +  .*.*..*
  29800 ++   *..*.*..*.   *.. :      *.*..*       *.*..*.   *       *.      |
  29600 ++                   *                                              |
  29400 ++                                                                  |
        |                                                                   |
  29200 ++                                                                  |
  29000 ++                                                                  |
  28800 ++   O                       O O  O  O    O O                       |
  28600 ++      O O       O                    O                            |
        O  O         O       O  O O                    O                    |
  28400 ++                                                                  |
  28200 ++--------------O---------------------------------------------------+

  [*] bisect-good sample
  [O] bisect-bad  sample


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,
Xiaolong

[-- Attachment #2: job-script.ksh --]
[-- Type: text/plain, Size: 6784 bytes --]

#!/bin/sh

export_top_env()
{
	export suite='will-it-scale'
	export testcase='will-it-scale'
	export category='benchmark'
	export nr_task=256
	export job_origin='/lkp/lkp/.src-20170426-224010/allot/cyclic:linux-devel:devel-hourly/lkp-knl-f1/will-it-scale-100.yaml'
	export queue='bisect'
	export testbox='lkp-knl-f1'
	export tbox_group='lkp-knl-f1'
	export submit_id='59127b640b9a9371627261dc'
	export job_file='/lkp/scheduled/lkp-knl-f1/will-it-scale-100%-thread-page_fault2-performance-debian-x86_64-2016-08-31.cgz-d34b0733b452ca3ef1dee36436dab08b8aa6a85c-20170510-94562-xjut9e-0.yaml'
	export id='467c63929f24d489ec9bbeab516c8528cccfc7df'
	export model='Phi'
	export nr_cpu=256
	export memory='96G'
	export nr_ssd_partitions=
	export swap_partitions=
	export rootfs_partition=
	export commit='d34b0733b452ca3ef1dee36436dab08b8aa6a85c'
	export kconfig='x86_64-kexec'
	export compiler='gcc-6'
	export rootfs='debian-x86_64-2016-08-31.cgz'
	export enqueue_time='2017-05-10 10:31:01 +0800'
	export _id='59127b640b9a9371627261dc'
	export _rt='/result/will-it-scale/100%-thread-page_fault2-performance/lkp-knl-f1/debian-x86_64-2016-08-31.cgz/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c'
	export user='lkp'
	export head_commit='ae6dbf83c88ae6678b80897d4e37921da02e6696'
	export base_commit='5a7ad1146caa895ad718a534399e38bd2ba721b7'
	export branch='linux-devel/devel-hourly-2017043014'
	export result_root='/result/will-it-scale/100%-thread-page_fault2-performance/lkp-knl-f1/debian-x86_64-2016-08-31.cgz/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/0'
	export LKP_SERVER='inn'
	export max_uptime=1500
	export initrd='/osimage/debian/debian-x86_64-2016-08-31.cgz'
	export bootloader_append='root=/dev/ram0
user=lkp
job=/lkp/scheduled/lkp-knl-f1/will-it-scale-100%-thread-page_fault2-performance-debian-x86_64-2016-08-31.cgz-d34b0733b452ca3ef1dee36436dab08b8aa6a85c-20170510-94562-xjut9e-0.yaml
ARCH=x86_64
kconfig=x86_64-kexec
branch=linux-devel/devel-hourly-2017043014
commit=d34b0733b452ca3ef1dee36436dab08b8aa6a85c
BOOT_IMAGE=/pkg/linux/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/vmlinuz-4.11.0-rc7-00043-gd34b073
max_uptime=1500
RESULT_ROOT=/result/will-it-scale/100%-thread-page_fault2-performance/lkp-knl-f1/debian-x86_64-2016-08-31.cgz/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/0
LKP_SERVER=inn
debug
apic=debug
sysrq_always_enabled
rcupdate.rcu_cpu_stall_timeout=100
net.ifnames=0
printk.devkmsg=on
panic=-1
softlockup_panic=1
nmi_watchdog=panic
oops=panic
load_ramdisk=2
prompt_ramdisk=0
drbd.minor_count=8
systemd.log_level=err
ignore_loglevel
earlyprintk=ttyS0,115200
console=ttyS0,115200
console=tty0
vga=normal
rw'
	export lkp_initrd='/lkp/lkp/lkp-x86_64.cgz'
	export modules_initrd='/pkg/linux/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/modules.cgz'
	export bm_initrd='/osimage/deps/debian-x86_64-2016-08-31.cgz/lkp_2017-05-01.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/rsync-rootfs_2016-11-15.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/run-ipconfig_2016-11-15.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/will-it-scale_2016-11-15.cgz,/osimage/pkg/debian-x86_64-2016-08-31.cgz/will-it-scale-x86_64_2016-09-02.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/iostat_2016-11-15.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/turbostat_2016-11-15.cgz,/osimage/pkg/debian-x86_64-2016-08-31.cgz/turbostat-x86_64_2016-09-02.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/perf_2017-04-24.cgz,/osimage/pkg/debian-x86_64-2016-08-31.cgz/perf-x86_64_2017-05-08.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/hw_2016-11-15.cgz'
	export site='inn'
	export LKP_CGI_PORT=80
	export LKP_CIFS_PORT=139
	export repeat_to=2
	export kernel='/pkg/linux/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/vmlinuz-4.11.0-rc7-00043-gd34b073'
	export dequeue_time='2017-05-10 10:36:39 +0800'
	export job_initrd='/lkp/scheduled/lkp-knl-f1/will-it-scale-100%-thread-page_fault2-performance-debian-x86_64-2016-08-31.cgz-d34b0733b452ca3ef1dee36436dab08b8aa6a85c-20170510-94562-xjut9e-0.cgz'

	[ -n "$LKP_SRC" ] ||
	export LKP_SRC=/lkp/${user:-lkp}/src
}

run_job()
{
	echo $$ > $TMP/run-job.pid

	. $LKP_SRC/lib/http.sh
	. $LKP_SRC/lib/job.sh
	. $LKP_SRC/lib/env.sh

	export_top_env

	run_setup $LKP_SRC/setup/cpufreq_governor 'performance'

	run_monitor $LKP_SRC/monitors/wrapper kmsg
	run_monitor $LKP_SRC/monitors/wrapper iostat
	run_monitor $LKP_SRC/monitors/wrapper heartbeat
	run_monitor $LKP_SRC/monitors/wrapper vmstat
	run_monitor $LKP_SRC/monitors/wrapper numa-numastat
	run_monitor $LKP_SRC/monitors/wrapper numa-vmstat
	run_monitor $LKP_SRC/monitors/wrapper numa-meminfo
	run_monitor $LKP_SRC/monitors/wrapper proc-vmstat
	run_monitor $LKP_SRC/monitors/wrapper proc-stat
	run_monitor $LKP_SRC/monitors/wrapper meminfo
	run_monitor $LKP_SRC/monitors/wrapper slabinfo
	run_monitor $LKP_SRC/monitors/wrapper interrupts
	run_monitor $LKP_SRC/monitors/wrapper lock_stat
	run_monitor $LKP_SRC/monitors/wrapper latency_stats
	run_monitor $LKP_SRC/monitors/wrapper softirqs
	run_monitor $LKP_SRC/monitors/one-shot/wrapper bdi_dev_mapping
	run_monitor $LKP_SRC/monitors/wrapper diskstats
	run_monitor $LKP_SRC/monitors/wrapper nfsstat
	run_monitor $LKP_SRC/monitors/wrapper cpuidle
	run_monitor $LKP_SRC/monitors/wrapper cpufreq-stats
	run_monitor $LKP_SRC/monitors/wrapper turbostat
	run_monitor $LKP_SRC/monitors/wrapper sched_debug
	run_monitor $LKP_SRC/monitors/wrapper perf-stat
	run_monitor $LKP_SRC/monitors/wrapper mpstat
	run_monitor $LKP_SRC/monitors/no-stdout/wrapper perf-profile
	run_monitor $LKP_SRC/monitors/wrapper oom-killer
	run_monitor $LKP_SRC/monitors/plain/watchdog
	run_monitor $LKP_SRC/monitors/wrapper nfs-hang

	run_test mode='thread' test='page_fault2' $LKP_SRC/tests/wrapper will-it-scale
}

extract_stats()
{
	$LKP_SRC/stats/wrapper will-it-scale
	$LKP_SRC/stats/wrapper kmsg
	$LKP_SRC/stats/wrapper iostat
	$LKP_SRC/stats/wrapper vmstat
	$LKP_SRC/stats/wrapper numa-numastat
	$LKP_SRC/stats/wrapper numa-vmstat
	$LKP_SRC/stats/wrapper numa-meminfo
	$LKP_SRC/stats/wrapper proc-vmstat
	$LKP_SRC/stats/wrapper meminfo
	$LKP_SRC/stats/wrapper slabinfo
	$LKP_SRC/stats/wrapper interrupts
	$LKP_SRC/stats/wrapper lock_stat
	$LKP_SRC/stats/wrapper latency_stats
	$LKP_SRC/stats/wrapper softirqs
	$LKP_SRC/stats/wrapper diskstats
	$LKP_SRC/stats/wrapper nfsstat
	$LKP_SRC/stats/wrapper cpuidle
	$LKP_SRC/stats/wrapper turbostat
	$LKP_SRC/stats/wrapper sched_debug
	$LKP_SRC/stats/wrapper perf-stat
	$LKP_SRC/stats/wrapper mpstat
	$LKP_SRC/stats/wrapper perf-profile

	$LKP_SRC/stats/wrapper time will-it-scale.time
	$LKP_SRC/stats/wrapper time
	$LKP_SRC/stats/wrapper dmesg
	$LKP_SRC/stats/wrapper kmsg
	$LKP_SRC/stats/wrapper stderr
	$LKP_SRC/stats/wrapper last_state
}

"$@"

[-- Attachment #3: job.yaml --]
[-- Type: text/plain, Size: 4419 bytes --]

---

#! jobs/will-it-scale-100.yaml
suite: will-it-scale
testcase: will-it-scale
category: benchmark
nr_task: 100%
will-it-scale:
  mode: thread
  test: page_fault2
job_origin: "/lkp/lkp/.src-20170426-224010/allot/cyclic:linux-devel:devel-hourly/lkp-knl-f1/will-it-scale-100.yaml"

#! queue options
queue: bisect
testbox: lkp-knl-f1
tbox_group: lkp-knl-f1
submit_id: 59127b640b9a9371627261dc
job_file: "/lkp/scheduled/lkp-knl-f1/will-it-scale-100%-thread-page_fault2-performance-debian-x86_64-2016-08-31.cgz-d34b0733b452ca3ef1dee36436dab08b8aa6a85c-20170510-94562-xjut9e-0.yaml"
id: 467c63929f24d489ec9bbeab516c8528cccfc7df

#! hosts/lkp-knl-f1
model: Phi
nr_cpu: 256
memory: 96G
nr_ssd_partitions: 
swap_partitions: 
rootfs_partition: 

#! include/category/benchmark
kmsg: 
iostat: 
heartbeat: 
vmstat: 
numa-numastat: 
numa-vmstat: 
numa-meminfo: 
proc-vmstat: 
proc-stat: 
meminfo: 
slabinfo: 
interrupts: 
lock_stat: 
latency_stats: 
softirqs: 
bdi_dev_mapping: 
diskstats: 
nfsstat: 
cpuidle: 
cpufreq-stats: 
turbostat: 
sched_debug: 
perf-stat: 
mpstat: 
perf-profile: 

#! include/category/ALL
cpufreq_governor: performance

#! include/queue/cyclic
commit: d34b0733b452ca3ef1dee36436dab08b8aa6a85c

#! include/testbox/lkp-knl-f1
kconfig: x86_64-kexec

#! default params
compiler: gcc-6
rootfs: debian-x86_64-2016-08-31.cgz
enqueue_time: 2017-05-10 10:31:01.173541316 +08:00
_id: 59127b640b9a9371627261dc
_rt: "/result/will-it-scale/100%-thread-page_fault2-performance/lkp-knl-f1/debian-x86_64-2016-08-31.cgz/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c"

#! schedule options
user: lkp
head_commit: ae6dbf83c88ae6678b80897d4e37921da02e6696
base_commit: 5a7ad1146caa895ad718a534399e38bd2ba721b7
branch: linux-devel/devel-hourly-2017043014
result_root: "/result/will-it-scale/100%-thread-page_fault2-performance/lkp-knl-f1/debian-x86_64-2016-08-31.cgz/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/0"
LKP_SERVER: inn
max_uptime: 1500
initrd: "/osimage/debian/debian-x86_64-2016-08-31.cgz"
bootloader_append:
- root=/dev/ram0
- user=lkp
- job=/lkp/scheduled/lkp-knl-f1/will-it-scale-100%-thread-page_fault2-performance-debian-x86_64-2016-08-31.cgz-d34b0733b452ca3ef1dee36436dab08b8aa6a85c-20170510-94562-xjut9e-0.yaml
- ARCH=x86_64
- kconfig=x86_64-kexec
- branch=linux-devel/devel-hourly-2017043014
- commit=d34b0733b452ca3ef1dee36436dab08b8aa6a85c
- BOOT_IMAGE=/pkg/linux/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/vmlinuz-4.11.0-rc7-00043-gd34b073
- max_uptime=1500
- RESULT_ROOT=/result/will-it-scale/100%-thread-page_fault2-performance/lkp-knl-f1/debian-x86_64-2016-08-31.cgz/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/0
- LKP_SERVER=inn
- debug
- apic=debug
- sysrq_always_enabled
- rcupdate.rcu_cpu_stall_timeout=100
- net.ifnames=0
- printk.devkmsg=on
- panic=-1
- softlockup_panic=1
- nmi_watchdog=panic
- oops=panic
- load_ramdisk=2
- prompt_ramdisk=0
- drbd.minor_count=8
- systemd.log_level=err
- ignore_loglevel
- earlyprintk=ttyS0,115200
- console=ttyS0,115200
- console=tty0
- vga=normal
- rw
lkp_initrd: "/lkp/lkp/lkp-x86_64.cgz"
modules_initrd: "/pkg/linux/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/modules.cgz"
bm_initrd: "/osimage/deps/debian-x86_64-2016-08-31.cgz/lkp_2017-05-01.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/rsync-rootfs_2016-11-15.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/run-ipconfig_2016-11-15.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/will-it-scale_2016-11-15.cgz,/osimage/pkg/debian-x86_64-2016-08-31.cgz/will-it-scale-x86_64_2016-09-02.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/iostat_2016-11-15.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/turbostat_2016-11-15.cgz,/osimage/pkg/debian-x86_64-2016-08-31.cgz/turbostat-x86_64_2016-09-02.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/perf_2017-04-24.cgz,/osimage/pkg/debian-x86_64-2016-08-31.cgz/perf-x86_64_2017-05-08.cgz,/osimage/deps/debian-x86_64-2016-08-31.cgz/hw_2016-11-15.cgz"
site: inn

#! /lkp/lkp/.src-20170430-210822/include/site/inn
LKP_CGI_PORT: 80
LKP_CIFS_PORT: 139
oom-killer: 
watchdog: 
nfs-hang: 

#! runtime status
repeat_to: 2

#! user overrides
kernel: "/pkg/linux/x86_64-kexec/gcc-6/d34b0733b452ca3ef1dee36436dab08b8aa6a85c/vmlinuz-4.11.0-rc7-00043-gd34b073"
dequeue_time: 2017-05-10 10:36:39.311087432 +08:00

#! /lkp/lkp/.src-20170509-205611/include/site/inn
job_state: finished
loadavg: '111.23'

[-- Attachment #4: reproduce.ksh --]
[-- Type: text/plain, Size: 142 bytes --]


for file in /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
do
	echo performance > $file
done

./runtest.py page_fault2 295 thread 256

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [lkp-robot] [mm, page_alloc] d34b0733b4: will-it-scale.per_thread_ops -4% regression
  2017-05-12  4:47 [lkp-robot] [mm, page_alloc] d34b0733b4: will-it-scale.per_thread_ops -4% regression kernel test robot
@ 2017-05-12  9:24 ` Mel Gorman
  2017-05-14  5:48   ` Ye Xiaolong
  0 siblings, 1 reply; 3+ messages in thread
From: Mel Gorman @ 2017-05-12  9:24 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 1155 bytes --]

On Fri, May 12, 2017 at 12:47:23PM +0800, kernel test robot wrote:
> 
> Greeting,
> 
> We noticed a -4% regression of will-it-scale.per_thread_ops due to commit:
> 
> commit: d34b0733b452ca3ef1dee36436dab08b8aa6a85c ("Revert "mm, page_alloc: only use per-cpu allocator for irq-safe requests"")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
> 
> in testcase: will-it-scale
> on test machine: 256 threads Phi with 96G memory
> with following parameters:
> 
> 	nr_task: 100%
> 	mode: thread
> 	test: page_fault2
> 	cpufreq_governor: performance
> 
> test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
> test-url: https://github.com/antonblanchard/will-it-scale
> 

Regrettably, this is a regression that cannot be avoided. The revert removes
an optimisation that has consequences for network-intensive workloads.
The whole idea needs to be revisited and the revert was the safest
option for 4.11.

-- 
Mel Gorman
SUSE Labs

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [lkp-robot] [mm, page_alloc] d34b0733b4: will-it-scale.per_thread_ops -4% regression
  2017-05-12  9:24 ` Mel Gorman
@ 2017-05-14  5:48   ` Ye Xiaolong
  0 siblings, 0 replies; 3+ messages in thread
From: Ye Xiaolong @ 2017-05-14  5:48 UTC (permalink / raw)
  To: lkp

[-- Attachment #1: Type: text/plain, Size: 1257 bytes --]

On 05/12, Mel Gorman wrote:
>On Fri, May 12, 2017 at 12:47:23PM +0800, kernel test robot wrote:
>> 
>> Greeting,
>> 
>> We noticed a -4% regression of will-it-scale.per_thread_ops due to commit:
>> 
>> commit: d34b0733b452ca3ef1dee36436dab08b8aa6a85c ("Revert "mm, page_alloc: only use per-cpu allocator for irq-safe requests"")
>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
>> 
>> in testcase: will-it-scale
>> on test machine: 256 threads Phi with 96G memory
>> with following parameters:
>> 
>> 	nr_task: 100%
>> 	mode: thread
>> 	test: page_fault2
>> 	cpufreq_governor: performance
>> 
>> test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
>> test-url: https://github.com/antonblanchard/will-it-scale
>> 
>
>Regrettably, this is a regression that cannot be avoided. The revert removes
>an optimisation that has consequences for network-intensive workloads.
>The whole idea needs to be revisited and the revert was the safest
>option for 4.11.

Thanks for the explanation.

Xiaolong
>
>-- 
>Mel Gorman
>SUSE Labs

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2017-05-14  5:48 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-05-12  4:47 [lkp-robot] [mm, page_alloc] d34b0733b4: will-it-scale.per_thread_ops -4% regression kernel test robot
2017-05-12  9:24 ` Mel Gorman
2017-05-14  5:48   ` Ye Xiaolong

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.