Greeting,
FYI, we noticed a -15.4% regression of will-it-scale.per_thread_ops due to commit:
commit: fa4e6b1ad57df096ddcf091fece3d9babfe90048 ("[patch 100/212] memcg: enable accounting for pollfd and select bits arrays")
url: https://github.com/0day-ci/linux/commits/Andrew-Morton/ia64-fix-typo-in-a-comment/20210903-065028
in testcase: will-it-scale
on test machine: 144 threads 4 sockets Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz with 512G memory
with following parameters:
nr_task: 50%
mode: thread
test: poll2
cpufreq_governor: performance
ucode: 0x16
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
If you fix the issue, kindly add following tag
Reported-by: kernel test robot <[email protected]>
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
bin/lkp run generated-yaml-file
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase/ucode:
gcc-9/performance/x86_64-rhel-8.3/thread/50%/debian-10.4-x86_64-20200603.cgz/lkp-hsw-4ex1/poll2/will-it-scale/0x16
commit:
ddae5955e0 ("memcg: enable accounting for mnt_cache entries")
fa4e6b1ad5 ("memcg: enable accounting for pollfd and select bits arrays")
ddae5955e09fc376 fa4e6b1ad57df096ddcf091fece
---------------- ---------------------------
%stddev %change %stddev
\ | \
20037899 -15.4% 16955694 ? 3% will-it-scale.72.threads
278303 -15.4% 235495 ? 3% will-it-scale.per_thread_ops
20037899 -15.4% 16955694 ? 3% will-it-scale.workload
1583653 ? 2% +47.6% 2338047 ? 8% cpuidle..usage
0.01 ? 2% +0.0 0.02 ? 8% mpstat.cpu.all.soft%
2568 ?188% +336.5% 11212 ? 44% numa-vmstat.node0.numa_other
76315 +3.5% 78953 vmstat.system.in
2953 ? 15% +39.4% 4118 ? 6% slabinfo.kmalloc-cg-1k.active_objs
2953 ? 15% +39.4% 4118 ? 6% slabinfo.kmalloc-cg-1k.num_objs
50719 ? 3% -21.6% 39775 ? 10% turbostat.C1E
1403174 +55.9% 2187482 ? 8% turbostat.C6
645.68 -2.6% 628.84 turbostat.PkgWatt
61545 ?178% +175.6% 169623 ? 78% interrupts.CPU110.LOC:Local_timer_interrupts
8916 ?147% +1620.7% 153419 ? 96% interrupts.CPU134.LOC:Local_timer_interrupts
108655 ?124% +105.1% 222831 ? 49% interrupts.CPU23.LOC:Local_timer_interrupts
611.50 ? 62% +65.4% 1011 ? 63% interrupts.CPU25.CAL:Function_call_interrupts
56545 ?189% +205.4% 172674 ? 74% interrupts.CPU25.LOC:Local_timer_interrupts
1652 ? 13% -48.5% 850.67 ? 63% interrupts.CPU52.CAL:Function_call_interrupts
346.83 +72.7% 598.83 ? 57% interrupts.CPU94.CAL:Function_call_interrupts
1279 ? 8% +83.7% 2350 ? 31% interrupts.RES:Rescheduling_interrupts
10413 ? 9% -15.1% 8842 ? 11% softirqs.CPU0.RCU
10111 ? 18% -34.4% 6636 ? 19% softirqs.CPU10.RCU
9749 ? 8% -35.8% 6263 ? 32% softirqs.CPU22.RCU
10009 ? 7% -59.3% 4069 ? 69% softirqs.CPU30.RCU
61.83 ? 38% +3560.9% 2263 ?210% softirqs.CPU74.TIMER
754991 ? 2% -23.4% 578181 ? 5% softirqs.RCU
727783 +16.3% 846764 ? 3% softirqs.SCHED
40315 +29.1% 52046 ? 4% softirqs.TIMER
5492 +14.7% 6298 ? 4% syscalls.sys_mmap.med
1.083e+08 ? 6% +1.3e+08 2.414e+08 ? 12% syscalls.sys_mmap.noise.100%
1.806e+08 ? 3% +1.3e+08 3.115e+08 ? 9% syscalls.sys_mmap.noise.2%
1.715e+08 ? 3% +1.3e+08 2.981e+08 ? 10% syscalls.sys_mmap.noise.25%
1.801e+08 ? 3% +1.3e+08 3.111e+08 ? 9% syscalls.sys_mmap.noise.5%
1.4e+08 ? 4% +1.3e+08 2.737e+08 ? 11% syscalls.sys_mmap.noise.50%
1.181e+08 ? 6% +1.4e+08 2.571e+08 ? 12% syscalls.sys_mmap.noise.75%
5921 +11.7% 6612 ? 3% syscalls.sys_openat.med
5133257 ? 10% +4.5e+08 4.563e+08 ? 38% syscalls.sys_poll.noise.100%
7282972 ? 31% +4.5e+08 4.599e+08 ? 37% syscalls.sys_poll.noise.2%
6705445 ? 30% +4.5e+08 4.593e+08 ? 37% syscalls.sys_poll.noise.25%
6828883 ? 32% +4.5e+08 4.596e+08 ? 37% syscalls.sys_poll.noise.5%
6600419 ? 29% +4.5e+08 4.592e+08 ? 37% syscalls.sys_poll.noise.50%
5686935 ? 13% +4.5e+08 4.585e+08 ? 37% syscalls.sys_poll.noise.75%
36152 ? 26% +56.4% 56553 ? 12% syscalls.sys_write.max
85796438 ? 15% +5.5e+07 1.407e+08 ? 20% syscalls.sys_write.noise.100%
1.024e+08 ? 14% +4.3e+07 1.451e+08 ? 19% syscalls.sys_write.noise.75%
will-it-scale.per_thread_ops
300000 +------------------------------------------------------------------+
| ++.+.+.+.++.+.+.+.+.++.+.+.+.++.+.+.+.++.+.+.+.+.++.+.+.+.++.+.|
250000 |-+ : O O O |
| O : OO O O O O O OO O O O OO |
| :O O O O O O O O O O O O |
200000 |:+: O O |
|: : O |
150000 |:+: |
|: : |
100000 |:+: |
|: : |
| : |
50000 |-: |
| : |
0 +------------------------------------------------------------------+
[*] 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.
---
0DAY/LKP+ Test Infrastructure Open Source Technology Center
https://lists.01.org/hyperkitty/list/[email protected] Intel Corporation
Thanks,
Oliver Sang