Skip to contentSkip to conversation

17cdf560f27: +4.0% boot-meminfo.VmallocUsed

LKP Huang Ying 3 participants 4 messages Apr 2, 2015 – Apr 3, 2015

lore
Share & tools
lore

Huang Ying

author
#0 lore raw

FYI, we noticed the below changes on

git://git.kernel.org/pub/scm/linux/kernel/git/vireshk/linux.git timer/alloc_tvec_statically commit 17cdf560f2727f687ab159707d0aa591f8a2f82d ("timer: Allocate per-cpu tvec_base's statically")

testbox/testcase/testparams: lkp-sbx04/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

         %stddev     %change         %stddev
             \          |                \  
    410996 ±  0%      +4.0%     427412 ±  0%  boot-meminfo.VmallocUsed

testbox/testcase/testparams: wsm/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

     85712 ±  0%      +2.4%      87772 ±  0%  boot-meminfo.VmallocUsed

testbox/testcase/testparams: lkp-snb01/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

    447001 ±  0%      -1.0%     442558 ±  0%  boot-slabinfo.num_objs
    125616 ±  5%     -13.5%     108720 ±  4%  boot-meminfo.DirectMap4k
    347196 ±  0%      +4.7%     363612 ±  0%  boot-meminfo.VmallocUsed

lkp-sbx04: Sandy Bridge-EX
Memory: 64G

wsm: Westmere
Memory: 6G

lkp-snb01: Sandy Bridge-EP
Memory: 32G

boot-meminfo.VmallocUsed

90000 O+-O--O--O-O--O--O--O--O--O-O--O--O--O--O--O-O--O--O--O--O--O-O--O--O

        *..*..*..*.*..*..*..*..*       *..*..*..*       *..*..*..*          |
  80000 ++                     :       :        :       :                   |
  70000 ++                     :       :        :       :                   |
        |                       :     :          :     :                    |
  60000 ++                      :     :          :     :                    |
  50000 ++                      :     :          :     :                    |
        |                       :     :          :     :                    |
  40000 ++                       :   :            :   :                     |
  30000 ++                       :   :            :   :                     |
        |                        :   :            :   :                     |
  20000 ++                       :   :            :   :                     |
  10000 ++                        : :              : :                      |
        |                         : :              : :                      |

0 ++------------------------*-*--------------*-*----------------------+

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

To reproduce:

apt-get install ruby
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/setup-local job.yaml # the job file attached in this email
bin/run-local 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 mailing list
LKP(a)linux.intel.com

---
testcase: boot
default-monitors:
default-watchdogs:
watch-oom:
watchdog:
cpufreq_governor: performance
model: Westmere
memory: 6G
nr_hdd_partitions: 1
hdd_partitions:
swap_partitions:
rootfs_partition:
netconsole_port: 6667
boot-slabinfo:
boot-meminfo:
memmap:
boot-memory:
boot-time:
kernel-size:
sleep: 1
branch: linux-devel/devel-hourly-2015040111
commit: 4a5cba20d58c8acccab671b13c83165be9cba0fb
repeat_to: 1
testbox: wsm
tbox_group: wsm
kconfig: x86_64-rhel
enqueue_time: 2015-04-01 13:38:08.480648100 +08:00
kernel: "/kernel/x86_64-rhel/4a5cba20d58c8acccab671b13c83165be9cba0fb/vmlinuz-4.0.0-rc6-03709-g4a5cba2"
user: lkp
queue: boot
rootfs: debian-x86_64-2015-02-07.cgz
result_root: "/result/wsm/boot/performance-1/debian-x86_64-2015-02-07.cgz/x86_64-rhel/4a5cba20d58c8acccab671b13c83165be9cba0fb/0"
job_file: "/lkp/scheduled/wsm/boot_boot-performance-1-x86_64-rhel-4a5cba20d58c8acccab671b13c83165be9cba0fb-0-20150401-18244-52f40d.yaml"
dequeue_time: 2015-04-01 13:38:26.975590830 +08:00
nr_cpu: "$(nproc)"
max_uptime: 3600
modules_initrd: "/kernel/x86_64-rhel/4a5cba20d58c8acccab671b13c83165be9cba0fb/modules.cgz"
job_state: finished
loadavg: 0.15 0.04 0.02 1/189 631
start_time: '1427866643'
end_time: '1427866644'
version: "/lkp/lkp/.src-20150331-175520"

echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu1/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu10/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu11/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu2/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu3/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu4/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu5/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu6/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu7/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu8/cpufreq/scaling_governor echo performance > /sys/devices/system/cpu/cpu9/cpufreq/scaling_governor

#1 lore raw
Quoted 20 lines · Huang Ying

FYI, we noticed the below changes on

git://git.kernel.org/pub/scm/linux/kernel/git/vireshk/linux.git timer/alloc_tvec_statically commit 17cdf560f2727f687ab159707d0aa591f8a2f82d ("timer: Allocate per-cpu tvec_base's statically")

testbox/testcase/testparams: lkp-sbx04/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

         %stddev     %change         %stddev
             \          |                \  
    410996 ±  0%      +4.0%     427412 ±  0%  boot-meminfo.VmallocUsed

testbox/testcase/testparams: wsm/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

     85712 ±  0%      +2.4%      87772 ±  0%  boot-meminfo.VmallocUsed
Quoted 8 lines

testbox/testcase/testparams: lkp-snb01/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

    447001 ±  0%      -1.0%     442558 ±  0%  boot-slabinfo.num_objs
    125616 ±  5%     -13.5%     108720 ±  4%  boot-meminfo.DirectMap4k
    347196 ±  0%      +4.7%     363612 ±  0%  boot-meminfo.VmallocUsed

well, duh.. it changes a dynamic allocation into a static per-cpu one. What did you expect?

#2 lore raw
Quoted 32 lines · Peter Zijlstra
Quoted 3 lines · Huang Ying

FYI, we noticed the below changes on

git://git.kernel.org/pub/scm/linux/kernel/git/vireshk/linux.git

timer/alloc_tvec_statically

Quoted 1 line

commit 17cdf560f2727f687ab159707d0aa591f8a2f82d ("timer: Allocate

per-cpu tvec_base's statically")

Quoted 16 lines

testbox/testcase/testparams: lkp-sbx04/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

         %stddev     %change         %stddev
             \          |                \
    410996 ±  0%      +4.0%     427412 ±  0%  boot-meminfo.VmallocUsed

testbox/testcase/testparams: wsm/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

     85712 ±  0%      +2.4%      87772 ±  0%  boot-meminfo.VmallocUsed
Quoted 8 lines

testbox/testcase/testparams: lkp-snb01/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

    447001 ±  0%      -1.0%     442558 ±  0%  boot-slabinfo.num_objs
    125616 ±  5%     -13.5%     108720 ±  4%  boot-meminfo.DirectMap4k
    347196 ±  0%      +4.7%     363612 ±  0%  boot-meminfo.VmallocUsed
Quoted 3 lines

well, duh.. it changes a dynamic allocation into a static per-cpu one. What did you expect?

Just FYI. This may be the expected behavior. If so, you have a third party test report.

Best Regards,
Huang, Yin

#3 lore raw
Quoted 34 lines · huang ying
Quoted 32 lines · Peter Zijlstra
Quoted 3 lines · Huang Ying

FYI, we noticed the below changes on

git://git.kernel.org/pub/scm/linux/kernel/git/vireshk/linux.git

timer/alloc_tvec_statically

Quoted 1 line

commit 17cdf560f2727f687ab159707d0aa591f8a2f82d ("timer: Allocate

per-cpu tvec_base's statically")

Quoted 24 lines

testbox/testcase/testparams: lkp-sbx04/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

         %stddev     %change         %stddev
             \          |                \
    410996 ±  0%      +4.0%     427412 ±  0%  boot-meminfo.VmallocUsed

testbox/testcase/testparams: wsm/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

     85712 ±  0%      +2.4%      87772 ±  0%  boot-meminfo.VmallocUsed

testbox/testcase/testparams: lkp-snb01/boot/performance-1

c5e77f5216abdd1d 17cdf560f2727f687ab159707d
---------------- --------------------------

    447001 ±  0%      -1.0%     442558 ±  0%  boot-slabinfo.num_objs
    125616 ±  5%     -13.5%     108720 ±  4%  boot-meminfo.DirectMap4k
    347196 ±  0%      +4.7%     363612 ±  0%  boot-meminfo.VmallocUsed
Quoted 6 lines
Quoted 3 lines

well, duh.. it changes a dynamic allocation into a static per-cpu one. What did you expect?

Just FYI. This may be the expected behavior. If so, you have a third party test report.

Ah ok. I thought you thought it was a bug/regression.

Thanks