Sorry, you need to enable JavaScript to visit this website.

Linux Kernel Performance

Linux development evolves rapidly. The performance and scalability of the OS kernel has been a key part of its success. However, discussions have appeared on LKML (Linux Kernel Mailing List) regarding large performance regression between kernel versions. These discussions underscore the need for a systematic and disciplined way to characterize, improve, and test Linux kernel performance. Our goal is to work with the Linux community to further enhance the Linux kernel with consistent performance increases (avoiding degradations) across releases. The information available on this site gives community members better information about what 0-Day and LKP (Linux Kernel Performance) are doing to preserve performance integrity of the kernel.

0-Day CI Linux Kernel Performance Report (v5.8)

BY Rong Chen ON Aug 18, 2020
  1. Introduction

0-Day CI is an automated Linux kernel test service that provides comprehensive test coverage of the Linux kernel. It covers kernel build, static analysis, boot, functional, performance and power tests. This report shows the recent observations of kernel performance status on IA platform based on the test results from 0-Day CI service. It is structured in the following manner:

  • Section 2, merged regressions and improvements in v5.8 release candidates

  • Section 3, test parameter description

  • Section 4, captured regressions and improvements by shift-left testing during developers’ and maintainers’ tree during v5.8 release cycle

  • Section 5, performance comparison among different kernel releases

  • Section 6, test machine list

 

  1. Test parameters

Here are the descriptions for each parameter/field used in the tests. 

 

Classification

Name

Description

General

runtime

Run the test case within a certain time period (seconds or minutes)

 

nr_task

If it is an integer, which means the number of processes/threads (to run the workload) of this job. Default is 1.

If it is a percentage, e.g. 200% means the number of processes/threads is double of cpu number

 

nr_threads

Alias of nr_task

 

iterations

Number to repeat this job

 

test_size

Test disk size or memory size

 

set_nic_irq_affinity

Set NIC interrupt affinity

 

disable_latency_stats

Latency_stats may introduce too much noise if there are too many context switches, allow to disable it

 

transparent_hugepage

Set transparent hugepage policy (/sys/kernel/mm/transparent_hugepage)

 

boot_params:bp1_memmap

Boot parameters of memmap

 

disk:nr_pmem

number of pmem partitions used by test

 

swap:priority

Priority means the  priority  of  the  swap device. priority is a value between -1 and 32767, the default is -1 and higher priority with higher value. 

Test Machine

model

Name of Intel processor microarchitecture

 

brand

Brand name of cpu

 

cpu_number

Number of cpu

 

memory

Size of memory


 

  1. Linux Kernel v5.8 Release Test

The 5.8 release of the Linux kernel was on August 2, 2020. Linus has released the 5.8 kernel. "So I considered making an rc8 all the way to the last minute, but decided it's not just worth waiting another week when there aren't any big looming worries around." Headline features in this release include: branch target identification and shadow call stacks for the arm64 architecture, the BPF iterator mechanism, inline encryption support in the block layer, the CAP_PERFMON and CAP_BPF capabilities, a generalized kernel event-notification subsystem, the KCSAN data-race detector, and more. As always, see the KernelNewbies 5.8 page for more information.

 

0-Day CI monitored the release closely to trace down the performance status on IA platform. 0-Day observed 13 regressions and 9 improvements during feature development phase for v5.8. We will share more detailed information together with correlated patches that led to the results. Note that the assessment is limited by the test coverage 0-Day has now. The list is summarized in the observation summary section.

    1. Observation Summary

0-Day CI observed 13 regressions and 9 improvements during the feature development phase for v5.8, which is in the time frame from v5.8-rc1 to v5.8 release. 

Test Indicator

Report

Test Scenario

Test Machine

Development Base

Status

aim7.jobs-per-min

[LKP] [xfs] a5949d3fae: -33.6% regression

disk: 1BRD_48G

fs: xfs

test: sync_disk_rw

load: 600

cpufreq_governor: performance

lkp-ivb-2ep1

v5.7-rc4

merged at v5.8-rc1, no response from author yet

aim7.jobs-per-min

[LKP] [f2fs] ca7f76e680: 41.5% improvement

disk: 4BRD_12G

md: RAID0

fs: f2fs

test: sync_disk_rw

load: 100

cpufreq_governor: performance

lkp-ivb-2ep1

v5.6

merged at v5.8-rc1

aim7.jobs-per-min

[LKP] [ext4, jbd2] 7b97d868b7: 21.9% improvement

disk: 4BRD_12G

md: RAID0

fs: ext4

test: disk_cp

load: 3000

cpufreq_governor: performance

lkp-skl-2sp7

v5.7-rc4

merged at v5.8-rc2

aim7.jobs-per-min

[LKP] [f2fs] ca7f76e680: -84.0% regression

disk: 1BRD_48G

fs: f2fs

test: sync_disk_rw

load: 200

cpufreq_governor: performance

lkp-cpx-4s1

v5.6

merged at v5.8-rc1, and the fix commit 79963d967b has been merged at v5.9-rc1

reaim.jobs_per_min

[LKP] [btrfs] e7a79811d0: 15.0% improvement

runtime: 300s

nr_task: 100

disk: 1HDD

fs: btrfs

test: disk

cpufreq_governor: performance

lkp-ivb-d04

v5.7-rc7

merged at v5.8-rc3

reaim.jobs_per_min

[LKP] [sched/cfs] e21cf43406: 2.5% improvement

runtime: 300s

nr_task: 100%

test: fserver

cpufreq_governor: performance

lkp-csl-2sp5

v5.8-rc2

merged at v5.8-rc3

stress-ng.af-alg.ops_per_sec

[LKP] [crypto] beeb460cd1: 17454.7% improvement

nr_threads: 100%

disk: 1HDD

testtime: 1s

class: cpu

cpufreq_governor: performance

lkp-bdw-ep6

v5.7-rc1

merged at v5.8-rc1

stress-ng.icache.ops_per_sec

[LKP] [ima] 8eb613c0b8: -84.2% regression

nr_threads: 100%

disk: 1HDD

testtime: 30s

class: cpu-cache

cpufreq_governor: performance

lkp-csl-2sp5

v5.7-rc2

It was merged at v5.8-rc1, and the fix commit 4235b1a4ef has been merged at v5.8-rc1

stress-ng.vm-splice.ops_per_sec

[LKP] [gup] 17839856fd: 2158.6% improvement

nr_threads: 100%

disk: 1HDD

testtime: 30s

class: pipe

cpufreq_governor: performance

lkp-csl-2sp5

v5.7

merged at v5.8-rc1

unixbench.score

[LKP] [perf/x86/rapl] 16accae3d9: -4.1% regression

runtime: 300s

nr_task: 30%

test: context1

cpufreq_governor: performance

lkp-csl-2ap2

v5.7

merged at v5.8-rc3, no response from author yet

vm-scalability.throughput

[LKP] [thp] 3917c80280: 10.8% improvement

runtime: 300s

size: 8T

test: anon-cow-seq

cpufreq_governor: performance

lkp-cfl-e1

v5.7

merged at v5.8-rc1

vm-scalability.throughput

[LKP] [mm] 1431d4d11a: -11.5% regression

runtime: 300s

test: lru-file-mmap-read

cpufreq_governor: performance

lkp-bdw-ex2

v5.7

merged at v5.8-rc1, the author is working on it

vm-scalability.throughput

[LKP] [thp] 3917c80280: -48.0% regression

runtime: 300s

size: 8T

test: anon-cow-seq-mt

cpufreq_governor: performance

lkp-cfl-e1

v5.7

merged at v5.8-rc1, the commit causes improvement too

will-it-scale.per_process_ops

[LKP] [x86/mce] 1de08dccd3: -14.1% regression

nr_task: 100%

mode: process

test: malloc1

cpufreq_governor: performance

lkp-knm01

v5.7-rc1

merged at v5.8-rc1, the author doubted the regression

will-it-scale.per_process_ops

[LKP] [bpf] 6890896bd7: -5.2% regression

nr_task: 16

mode: process

test: mmap2

cpufreq_governor: performance

lkp-ivb-d02

v5.7-rc2

merged at v5.8-rc1, no response from author yet

will-it-scale.per_process_ops

[LKP] [mm] 974b9b2c68: -1.8% regression

nr_task: 100%

mode: process

test: brk1

cpufreq_governor: performance

lkp-skl-fpga01

v5.7

merged at v5.8-rc1, no response from author yet

will-it-scale.per_process_ops

[LKP] [tracing/probe] 3aa8fdc37d: 3.6% improvement

nr_task: 16

mode: process

test: poll2

cpufreq_governor: performance

lkp-csl-2ap2

v5.8-rc1

merged at v5.8-rc2

will-it-scale.per_process_ops

[LKP] [mm/memcontrol.c] 3a98990ae2: 17.6% improvement

nr_task: 50%

mode: process

test: page_fault3

cpufreq_governor: performance

lkp-csl-2ap3

v5.8-rc2

merged at v5.8-rc3

will-it-scale.per_process_ops

[LKP] [x86/entry/common] 8f159f1dfa: -6.4% regression

nr_task: 100%

mode: process

test: lseek1

cpufreq_governor: performance

lkp-cpx-4s1

v5.7

merged at v5.8-rc1, no response from author yet

will-it-scale.per_thread_ops

[LKP] [rcu] 276c410448: -12.3% regression

nr_task: 100%

mode: thread

test: page_fault3

cpufreq_governor: performance

lkp-knm01

v5.7-rc2

merged at v5.8-rc1, the author doesn’t believe the regression

will-it-scale.per_thread_ops

[LKP] [sched/core] 2ebb177175: -3.7% regression

nr_task: 100%

mode: thread

test: pthread_mutex1

cpufreq_governor: performance

lkp-ivb-d01

v5.7-rc6

merged at v5.8-rc1, the author think it’s acceptable

will-it-scale.per_thread_ops

[LKP] [fs] b6509f6a8c: -12.6% regression

nr_task: 100%

mode: thread

test: eventfd1

cpufreq_governor: performance

lkp-csl-2ap4

v5.8-rc3

merged at v5.8-rc4, no response from author yet

    1. aim7.jobs-per-min

aim7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of a multiuser system. 

      1. scenario: sync_disk_rw test on xfs

 

Commit a5949d3fae was reported to have -33.6% regression of aim7.jobs-per-min when comparing to v5.7-rc4. It was merged to mainline at v5.8-rc1.

 

Correlated commits

a5949d3fae

xfs: force writes to delalloc regions to unwritten

branch

linux/master

report

[LKP] [xfs] a5949d3fae: -33.6% regression

test scenario

disk: 1BRD_48G

fs: xfs

test: sync_disk_rw

load: 600

cpufreq_governor: performance

test machine

lkp-ivb-2ep1

status

merged at v5.8-rc1, no response from author yet

 

    1. reaim.jobs_per_min

reaim updates and improves the existing Open Source AIM 7 benchmark. aim7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of a multiuser system.

 

    1. scenario: disk test on btrfs

 

Commit e7a79811d0 was reported to have 15.0% improvement of reaim.jobs_per_min when comparing to v5.7-rc7. It was merged to mainline at v5.8-rc3.

 

Correlated commits

e7a79811d0

btrfs: check if a log root exists before locking the log_mutex on unlink

branch

linux/master

report

[LKP] [btrfs] e7a79811d0: 15.0% improvement

test scenario

runtime: 300s

nr_task: 100

disk: 1HDD

fs: btrfs

test: disk

cpufreq_governor: performance

test machine

lkp-csl-2sp7

status

merged at v5.8-rc3

 

  1. will-it-scale.per_thread_ops

Will-it-scale takes a test case and runs it from 1 through to n parallel copies to see if the test case will scale. It builds both process and threads based tests in order to see any differences between the two.

    1. Scenario: thread eventfd1 test

 

 

Commit b6509f6a8c was reported to have -12.6% regression of will-it-scale.per_thread_ops when comparing to v5.8-rc3. It was merged to mainline at v5.8-rc4.

 

Correlated commits

b6509f6a8c

Revert "fs: Do not check if there is a fsnotify watcher on pseudo inodes"

branch

linux/master

report

[LKP] [fs] b6509f6a8c: -12.6% regression

test scenario

nr_task: 100%

mode: thread

test: eventfd1

cpufreq_governor: performance

test machine

lkp-csl-2ap4

status

merged at v5.8-rc4, no response from author yet


 

  1. Shift-Left Testing

Beyond testing trees in the upstream kernel, 0-Day CI also tests developers’ and maintainers’ trees, which can catch issues earlier and reduce wider impact. We call it “shift-left” testing. During the v5.8 release cycle, 0-Day CI had reported 18 major performance regressions and 14 major improvements by doing shift-left testing. We will share more detailed information together with possible code changes that led to this result for some of these, though the assessment is limited by the test coverage we have now. The whole list is summarized in the report summary section.

    1. Report Summary

0-Day CI had reported 18 performance regressions and 14 improvements by doing shift-left testing on developer and maintainer repos.

 

Test Indicator

Mail

Test Scenario

Test Machine

Status

aim7.jobs-per-min

[LKP] [cpufreq] d83f959b5e: -30.2% regression

disk: 1BRD_48G

fs: xfs

test: disk_src

load: 3000

cpufreq_governor: performance

lkp-skl-2sp7

currently not merged, no response from author yet

aim7.jobs-per-min

[LKP] [xfs] 976f0d2155: -32.4% regression

disk: 1BRD_48G

fs: xfs

test: disk_rw

load: 3000 

cpufreq_governor: performance

lkp-csl-2ap2

currently not merged, the commit causes improvement too

aim7.jobs-per-min

[LKP] [mm/swap] d8a3620566: -22.8% regression

disk: 4BRD_12G

md: RAID1 

fs: ext4

test: disk_rd

load: 9000

cpufreq_governor: performance

lkp-skl-2sp7

currently not merged, no response from author yet

aim7.jobs-per-min

[LKP] [HAX sched/core] 56afc8316c: -20.6% regression

disk: 4BRD_12G

md: RAID1

fs: ext4

test: sync_disk_rw

load: 600

cpufreq_governor: performance

lkp-cpx-4s1

currently not merged, no response from author yet

fio.read_iops

[LKP] [mm/lru] ea4ebd4126: -10.0% regression

disk: 2pmem

fs: btrfs

runtime: 200s

nr_task: 50%

time_based: tb

rw: read

bs: 4k

ioengine: sync

test_size: 100G cpufreq_governor: performance

lkp-csl-2sp6

currently not merged, the author is working on it

fio.write_iops

[LKP] [btrfs] b5fa368b72: 71.4% improvement

runtime: 300s

disk: 1SSD

fs: btrfs

nr_task: 100%

test_size: 128G

rw: write

bs: 4k

ioengine: sync

cpufreq_governor: performance

fs2: nfsv4

lkp-csl-2ap1

currently not merged

fsmark.files_per_sec

[LKP] [nfsd4] 94415b06eb: -24.6% regression

iterations: 1x

nr_threads: 1t

disk: 1BRD_32G

fs: xfs

fs2: nfsv4

filesize: 4K

test_size: 4G

sync_method: fsyncBeforeClose

nr_files_per_directory: 1fpd

cpufreq_governor: performance

lkp-cpx-4s1

currently not merged, no response from author yet

fsmark.files_per_sec

[LKP] [block] 54529aac98: -13.6% regression

iterations: 1x

nr_threads: 1t

disk: 1BRD_32G

fs: btrfs

filesize: 4K

test_size: 4G

sync_method: fsyncBeforeClose

nr_files_per_directory: 1fpd

cpufreq_governor: performance

lkp-cpx-4s1

currently not merged, no response from author yet

iostat.md0.wkB/s

[LKP] [xfs] 976f0d2155: 141.1% improvement

disk: 4BRD_12G

md: RAID1

fs: xfs

test: disk_wrt

load: 3000

cpufreq_governor: performance

lkp-skl-2sp7

currently not merged

netperf.Throughput_tps

[LKP] [sctp] cf8d0e3e58: -100.0% regression

ip: ipv4

runtime: 300s

nr_threads: 25%

cluster: cs-localhost

test: SCTP_RR

cpufreq_governor: performance

lkp-cfl-e1

currently not merged, no response from author yet

netperf.Throughput_tps

[LKP] ae04f172df: -8.2% regression

ip: ipv4

runtime: 300s

nr_threads: 16

cluster: cs-localhost

test: TCP_CRR

cpufreq_governor: performance

lkp-csl-2ap3

currently not merged, the author has fixed it

reaim.jobs_per_min

[LKP] [btrfs] 7b04b5c2d3: 20.7% improvement

runtime: 300s

nr_task: 100

disk: 1HDD

fs: btrfs

test: disk

cpufreq_governor: performance

lkp-csl-2sp7

currently not merged

stress-ng.ioport.ops_per_sec

[LKP] [x86/entry] 4ab57d31b1: -100.0% regression

nr_threads: 100%

disk: 1HDD

testtime: 30s

class: cpu

cpufreq_governor: performance

lkp-csl-2sp6

currently not merged, no response from author yet

stress-ng.remap.ops_per_sec

[LKP] [mm/swap] 8e007273a3: -76.3% regression

nr_threads: 100%

disk: 1HDD

testtime: 10s

class: memory

cpufreq_governor: performance

lkp-csl-2sp5

currently not merged, the author is working on it

stress-ng.splice.ops_per_sec

[LKP] [fs] 140402bab8: -100.0% regression

nr_threads: 100%

disk: 1HDD

testtime: 30s

class: pipe

cpufreq_governor: performance

lkp-csl-2sp6

currently not merged, Linus thinks it can be fixed

stress-ng.timerfd.ops_per_sec

[LKP] [printk] 96d5c61b1f: 8.3% improvement

nr_threads: 100%

disk: 1HDD

testtime: 30s

class: interrupt

cpufreq_governor: performance

lkp-csl-2sp6

currently not merged

unixbench.score

[LKP] [fs] 936e92b615: 32.3% improvement

runtime: 300s

nr_task: 30%

test: syscall

cpufreq_governor: performance

lkp-csl-2ap3

currently not merged

vm-scalability.throughput

[LKP] [fs] 420772c473: 41.7% improvement

runtime: 300s

test: lru-file-readtwice

cpufreq_governor: performance

lkp-cfl-e1

currently not merged

vm-scalability.throughput

[LKP] [sched/fair] 0b9730e694: 7.7% improvement

runtime: 300s

size: 8T

test: anon-w-seq-mt

cpufreq_governor: performance

lkp-skl-fpga01

currently not merged

vm-scalability.throughput

[LKP] [sched/fair] 0b9730e694: -7.9% regression

runtime: 300s

size: 512G

test: anon-w-rand-hugetlb

cpufreq_governor: performance

lkp-csl-2sp6

currently not merged, no response from author yet

vm-scalability.throughput

[LKP] [mm/swap] 8e007273a3: 68.0% improvement

runtime: 300s

test: lru-file-readtwice

cpufreq_governor: performance

lkp-csl-2ap4

currently not merged

vm-scalability.throughput

[LKP] [hugetlbfs] 878308e2e0: 1.2% improvement

runtime: 300s

size: 8T

test: anon-cow-seq-hugetlb

cpufreq_governor: performance

lkp-csl-2sp6

currently not merged

will-it-scale.per_process_ops

[LKP] [groups] 67467ae141: 4.3% improvement

nr_task: 100%

mode: process

test: poll2

cpufreq_governor: performance

lkp-skl-fpga01

currently not merged

will-it-scale.per_process_ops

[LKP] [mm] 4e2c82a409: 1894.6% improvement

nr_task: 100%

mode: process

test: mmap1

cpufreq_governor: performance

lkp-skl-fpga01

currently not merged

will-it-scale.per_process_ops

[LKP] [bpf] af7ec13833: -2.5% regression

nr_task: 16

mode: process

test: mmap1

cpufreq_governor: performance

lkp-csl-2ap2

currently not merged, it’s a function align problem

will-it-scale.per_process_ops

[LKP] [perf] e17d43b93e: 5.8% improvement

nr_task: 50%

mode: process

test: signal1

cpufreq_governor: performance

lkp-csl-2ap3

currently not merged

will-it-scale.per_process_ops

[LKP] [fsnotify] c738fbabb0: -9.5% regression

nr_task: 16

mode: process

test: open1

cpufreq_governor: performance

lkp-csl-2ap2

currently not merged, the author is working on it

will-it-scale.per_process_ops

[LKP] [fsnotify] 71d734103e: 4.2% improvement

nr_task: 50%

mode: process

test: eventfd1

cpufreq_governor: performance

lkp-csl-2ap3

currently not merged

will-it-scale.per_process_ops

[LKP] [perf/x86/intel/lbr] 47125db27e: 229.0% improvement

nr_task: 50%

mode: process

test: context_switch1

cpufreq_governor: performance

lkp-csl-2ap3

currently not merged

will-it-scale.per_thread_ops

[LKP] [fs] 0587211dff: 4.7% improvement

nr_task: 100%

mode: thread

test: unix1

cpufreq_governor: performance

lkp-skl-fpga01

currently not merged

will-it-scale.per_thread_ops

[LKP] [sched_pair_cpu] e9f2fb8893: -67.5% regression

nr_task: 100%

mode: thread

test: context_switch1

cpufreq_governor: performance

lkp-cpx-4s1

currently not merged, no response from author yet

    1. aim7.jobs-per-min

aim7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of a multiuser system. 

 

      1. scenario: sync_disk_rw test on ext4

 

Commit 56afc8316c was reported to have -20.6% regression of aim7.jobs-per-min when comparing to v5.8-rc4.

 

Correlated commits

56afc8316c

HAX sched/core: Paper over the ttwu() race

branch

linux-review/Chris-Wilson/HAX-sched-core-Paper-over-the-ttwu-race/20200721-013235

report

[LKP] [HAX sched/core] 56afc8316c: -20.6% regression

test scenario

disk: 4BRD_12G

md: RAID1

fs: ext4

test: sync_disk_rw

load: 600

cpufreq_governor: performance

test machine

lkp-cpx-4s1

status

currently not merged, no response from author yet

 

    1. vm-scalability.throughput

vm-scalability exercises functions and regions of the mm subsystem of the Linux kernel. We tested on multiple machines such as HSW EP server, during which we reported improvement on one test scenario.

 

      1. scenario: lru-file-readtwice test

 

 

Commit 8e007273a3 was reported to have 68.0% improvement of vm-scalability.throughput when comparing to v5.8-rc3.

 

Correlated commits

8e007273a3

mm/swap: serialize memcg changes during pagevec_lru_move_fn

branch

alexshi/58lru

report

[LKP] [mm/swap] 8e007273a3: 68.0% improvement

test scenario

runtime: 300s

test: lru-file-readtwice

cpufreq_governor: performance

test machine

lkp-csl-2ap4

status

currently not merged

 

    1. fio.write_iops

Fio is a tool that will spawn a number of threads or processes doing a particular type of I/O action as specified by the user.

    1. scenario: write on btrfs

 

Commit b5fa368b72 was reported to have 71.4% improvement of fio.write_iops when comparing to v5.8-rc6.

 

Correlated commits

b5fa368b72

btrfs: don't balance btree inode pages from buffered write path

branch

linux-next/master

report

[LKP] [btrfs] b5fa368b72: 71.4% improvement

test scenario

runtime: 300s

disk: 1SSD

fs: btrfs

nr_task: 100%

test_size: 128G

rw: write

bs: 4k

ioengine: sync

cpufreq_governor: performance

fs2: nfsv4

test machine

lkp-csl-2ap1

status

currently not merged

 

  1. will-it-scale.per_process_ops

Will-it-scale takes a test case and runs it from 1 through to n parallel copies to see if the test case will scale. It builds both process and threads based tests in order to see any differences between the two.

    1. Scenario: process open1 test

 

 

Commit c738fbabb0 was reported to have -9.5% regression of will-it-scale.per_process_ops when comparing to v5.8-rc4.

 

Correlated commits

c738fbabb0

fsnotify: fold fsnotify() call into fsnotify_parent()

branch

linux-next/master

report

[LKP] [firmware] 88a5883981: -66.8% regression

test scenario

nr_task: 16

mode: process

test: open1

cpufreq_governor: performance

test machine

lkp-csl-2ap2

status

currently not merged

 

  1. Latest Release Performance Comparing

 

This session gives some information about the performance difference among different kernel releases, especially between v5.8 and v5.7. There are 50+ performance benchmarks running in 0-Day CI, and we selected 9 benchmarks which historically showed the most regressions/improvements reported by 0-Day CI. Some typical configuration/parameters are used to run the test. For some of the regressions from the comparison, 0-Day did not successfully bisect it thus no related report sent out during the release development period, but it is still worth checking. The root cause to cause the regressions won’t be covered in this session. 

 

In the following figures, the value on the Y-axis is the relative performance number. We used the v5.6 data as the base (performance number is 100).

    1. test suite: vm-scalability

vm-scalability exercises functions and regions of the mm subsystem of the Linux kernel. Below 2 tests show the typical test results. 

 

vm-scalability Test 1

vm-scalability Test 2

 

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                           

 

vm-scalability Test 1 

vm-scalability Test 2

test machine

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

runtime

300s

300s

size

512G

No requirement

vm-scalability test parameter

test case: anon-cow-rand-mt

test case: lru-file-readonce

performance summary

vm-scalability.throughput on kernel v5.8 has -12.12% regression when comparing to v5.7

vm-scalability.throughput on kernel v5.8 has 14.14% improvement when comparing to v5.7

 

    1. test suite: will-it-scale

Will-it-scale takes a test case and runs it from 1 through to n parallel copies to see if the test case will scale. It builds both process and threads based tests in order to see any differences between the two.

 

will-it-scale  Test 1

Will-it-scale Test 2

will-it-scale  Test 3

will-it-scale  Test 4

 

Here are the parameters and performance test summary for above tests:                                                                                                                                                                                                                          

 

will-it-scale Test 1 

will-it-scale Test 2 

will-it-scale Test 3 

will-it-scale Test 4 

test machine

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

nr_task

50%

50%

16

100%

will-it-scale test parameter

mode: process

test case: context_switch1

mode: thread

test: futex4

mode: thread

test: eventfd1

mode: process

test: malloc1

summary

will-it-scale.per_process_ops on kernel v5.8 has -61.72% regression when comparing to v5.7

will-it-scale.per_process_ops on kernel v5.8 has -61.4% regression when comparing to v5.7

will-it-scale.per_thread_ops on kernel v5.8 has -7.32% regression when comparing to v5.7

will-it-scale.per_process_ops on kernel v5.8 has 9.55% improvement when comparing to v5.7

 

    1. test suite: unixbench

UnixBench is a system benchmark to provide a basic indicator of the performance of a Unix-like system.

 

Unixbench Test 1

Unixbench Test 2

 

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                           

 

Unixbench Test 1 

Unixbench Test 2

test machine

model: Coffee Lake

brand: Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz

cpu_number: 16

memory: 32G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

runtime

300s

300s

nr_task

1

30%

unixbench test parameter

test case: fsdisk

test case: context1

performance summary

unixbench.score on kernel v5.8 has -18.09% regression when comparing to v5.7

unixbench.score on kernel v5.8 has -7.1% regression when comparing to v5.7

 

    1. test suite: reaim

reaim updates and improves the existing Open Source AIM 7 benchmark. aim7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of a multiuser system.

 

reaim Test 1                             

reaim Test 2

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                                      

 

reaim Test 1 

eaim Test 2

test machine

model: Cascade Lake

brand: Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz

cpu_number: 96

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz

cpu_number: 96

memory: 256G

runtime

300s

300s

nr_task

100%

100

disk

No requirement

1HDD

fs

No requirement

btrfs

reaim test parameter

test case: fserver

test: disk

performance  summary

pigz.throughput on kernel v5.8 is almost the same as that in v5.7

reaim.jobs_per_min on kernel v5.8 has 39.76% improvement when comparing to v5.7

                                                                                                                                                                                                     

    1. test suite: pigz

pigz, which stands for Parallel Implementation of GZip, is a fully functional replacement for gzip that exploits multiple processors and multiple cores to the hilt when compressing data.

pigz Test 1 


 

 

Here are the test configuration and performance test summary for above tests:          

 

 

pigz Test 1

test machine

model: Ivy Bridge

brand: Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz

cpu_number: 4

memory: 8G

nr_threads

25%

pigz Test parameter

blocksize: 512K

performance  summary

pigz.throughput on kernel v5.8 is almost the same as that in v5.7

            

  1. test suite: netperf

Netperf is a benchmark that can be used to measure the performance of many different types of networking. It provides tests for both unidirectional throughput, and end-to-end latency.

 

netperf Test 1

netperf Test 2

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                                          

 

netperf Test 1 

netperf Test 2 

test machine

model: Coffee Lake

brand: Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz

cpu_number: 16

memory: 32G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

disable_latency_stats

1

1

set_nic_irq_affinity

1

1

runtime

300s

300s

nr_threads

1

16

ip

ipv4

ipv4

netperf test parameter

test case: UDP_STREAM

test case: TCP_CRR

performance  summary

netperf.Throughput_Mbps on kernel v5.8 has -6.46% regression when comparing to v5.7

netperf.Throughput_tps on kernel v5.8 has 11.21% improvement when comparing to v5.7

 

    1. test suite: hackbench

Hackbench is both a benchmark and a stress test for the Linux kernel scheduler. It's  main job  is  to  create a specified number of pairs of schedulable entities (either threads or traditional processes) which communicate via either sockets or pipes and time how long  it takes for each pair to send data back and forth.

hackbench Test 1

hackbench Test 2

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                           

 

hackbench Test 1 

hackbench Test 2 

test machine

model: Cascade Lake

brand: Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz

cpu_number: 96

memory: 192G

model: Cascade Lake

brand: Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz

cpu_number: 96

memory: 192G

disable_latency_stats

1

1

nr_task

50%

1600%

unixbench test parameter

iterations: 10

mode: threads

ipc: pipe

iterations: 10

mode: process

ipc: pipe

performance summary

hackbench.throughput on kernel v5.8 has -10.37% regression when comparing to v5.7

hackbench.throughput on kernel v5.8 has -6.72% regression when comparing to v5.7

            

    1. test suite: fio

Fio was originally written to save me the hassle of writing special test case programs when I wanted to test a specific workload, either for performance reasons or to find/reproduce a bug.

fio Test 1 

  

fio Test 2

Here are the test configuration and performance test summary for above tests:                                                                                                                                                                                                                          

 

fio Test 1

fio Test 2

test machine

model: Cascade Lake

brand: Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz

cpu_number: 96

memory: 256G

model: Cascade Lake

brand: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz

cpu_number: 192

memory: 192G

runtime

200s

20m

file system

ext4

ext4

disk

2pmem

1SSD

boot_params

bp1_memmap: 104G!8G

bp2_memmap: 104G!132G

No requirement

nr_task

50%

100%

time_based

tb

No requirement

fio test parameter

fio-setup-basic:

  rw: randread

  bs: 2M

  ioengine: mmap

  test_size: 200G

fio-setup-basic:

  rw: randwrite

  bs: 4k

  ioengine: sync

  test_size: 32G

performance  summary

fio.read_bw_MBps on kernel v5.8 has -10.94% regression when comparing to v5.7

fio.write_bw_MBps on kernel v5.8 has 8.68% improvement when comparing to v5.7

 

    1. test suite: ebizzy

ebizzy is designed to generate a workload resembling common web application server workloads. It is highly threaded, has a large in-memory working set, and allocates and deallocates memory frequently.

ebizzy Test 1

 

 

Here are the test configuration and performance test summary for above test:                                                                                                                                                                                                                          

 

ebizzy Test 1

test machine

model: Cooper Lake

brand: 

cpu_number: 192

memory: 128G

transparent_hugepage

thp_enabled: always

thp_defrag: always

nr_threads

200%

iterations

100x

ebizzy test parameter

duration: 10s

performance  summary

ebizzy.throughput on kernel v5.8 has -0.91% regression when comparing to v5.7


 

  1. Test Machines

    1. IVB Desktop

model

Ivy Bridge

brand

Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz

cpu number

8

memory

16G

 

model

Ivy Bridge

brand

Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz

cpu number

4

memory

8G

 

    1. SKL SP

model

Skylake

brand

Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz

cpu number

80

memory

64G

 

    1. BDW EP

model

Broadwell-EP

brand

Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz

cpu number

88

memory

128G

 

    1. HSW EP

model

Haswell-EP

brand

Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz

cpu number

72

memory

128G

 

    1. IVB EP

model

Ivy Bridge-EP

brand

Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz

cpu number

40

memory

384G

 

model

Ivytown Ivy Bridge-EP

brand

Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz

cpu number

48

memory

64G

 

    1. HSX EX

model

Brickland Haswell-EX

brand

Intel(R) Xeon(R) CPU E7-8890 v3 @ 2.50GHz

cpu number

144

memory

512G