Private bug reported:
With the CONFIG_NO_HZ_FULL option enabled, visible rt performance improvement
observed for both cyclictest and oslat.
RT kernel version - linux-image-5.11.0-26-realtime_5.11.0-26.27_amd64.deb.
Would like to know if we could have this option enabled by default.
And appreciate for the reason if not.
detailed results:
# taskset -c 0-10 ./cyclictest -m -a 1-10 -t 10 -d 0 -h 15 -p 95
--mainaffinity=0
WARN: distance is ignored and set to 0, if histogram enabled
/dev/cpu_dma_latency set to 0us
policy: fifo: loadavg: 0.75 0.66 0.60 1/933 3320
T: 0 ( 3291) P:95 I:1000 C:3344936 Min: 2 Act: 3 Avg: 3 Max: 9
T: 1 ( 3292) P:95 I:1000 C:3344936 Min: 2 Act: 3 Avg: 3 Max: 8
T: 2 ( 3293) P:95 I:1000 C:3344937 Min: 2 Act: 3 Avg: 3 Max: 7
T: 3 ( 3294) P:95 I:1000 C:3344937 Min: 2 Act: 3 Avg: 3 Max: 8
T: 4 ( 3295) P:95 I:1000 C:3344937 Min: 2 Act: 3 Avg: 3 Max: 8
T: 5 ( 3296) P:95 I:1000 C:3344937 Min: 2 Act: 3 Avg: 3 Max: 8
T: 6 ( 3297) P:95 I:1000 C:3344937 Min: 3 Act: 3 Avg: 3 Max: 8
T: 7 ( 3298) P:95 I:1000 C:3344937 Min: 3 Act: 3 Avg: 3 Max: 9
T: 8 ( 3299) P:95 I:1000 C:3344937 Min: 2 Act: 3 Avg: 3 Max: 8
T: 9 ( 3300) P:95 I:1000 C:3344937 Min: 2 Act: 3 Avg: 3 Max: 9
^C# Histogram
000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000
000001 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000
000002 000010 000001 000002 000003 000003 000015 000000 000000 000001 000021
000003 3271164 3336407 3330173 3329589 3336373 3340246 3321190 3322566 3330415
3341556
000004 073455 007552 014534 012150 007633 004469 022749 022036 013813 002408
000005 000210 000705 000223 001543 000818 000152 000953 000232 000558 000334
000006 000021 000049 000008 001239 000054 000040 000039 000017 000023 000508
000007 000054 000162 000002 000392 000052 000017 000007 000085 000097 000108
000008 000024 000064 000000 000025 000008 000002 000003 000004 000034 000005
000009 000002 000000 000000 000000 000000 000000 000000 000001 000000 000001
000010 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000
000011 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000
000012 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000
000013 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000
000014 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000
Total: 003344940 003344940 003344942 003344941 003344941 003344941 003344941
003344941 003344941 003344941
Min Latencies: 00002 00002 00002 00002 00002 00002 00003 00003 00002 00002
Avg Latencies: 00003 00003 00003 00003 00003 00003 00003 00003 00003 00003
Max Latencies: 00009 00008 00007 00008 00008 00008 00008 00009 00008 00009
Histogram Overflows: 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000
Histogram Overflow at cycle number:
Thread 0:
Thread 1:
Thread 2:
Thread 3:
Thread 4:
Thread 5:
Thread 6:
Thread 7:
Thread 8:
Thread 9:
# ./oslat -c 1-10 -m 50M -D 10m
oslat V 2.20
Total runtime: 600 seconds
Thread priority: default
CPU list: 1-10
CPU for main thread: 0
Workload: no
Workload mem: 0 (KiB)
Preheat cores: 10
Pre-heat for 1 seconds...
Test starts...
Test completed.
Core: 1 2 3 4 5 6 7 8 9 10
Counter Freq: 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 (Mhz)
001 (us): 21077201149 21077201144 21077201147 21077201140 21077201142
21077201140 21077201143 21077201138 21077201148 21077201138
002 (us): 0 0 0 0 0 0 0 0 0 0
003 (us): 0 0 0 0 0 0 0 0 0 0
004 (us): 0 0 0 0 0 0 0 0 0 0
005 (us): 0 0 0 0 0 0 0 0 0 0
006 (us): 0 0 0 0 0 0 0 0 0 0
007 (us): 0 0 0 0 0 0 0 0 0 0
008 (us): 0 0 0 0 0 0 0 0 0 0
009 (us): 0 0 0 0 0 0 0 0 0 0
010 (us): 0 0 0 0 0 0 0 0 0 0
011 (us): 0 0 0 0 0 0 0 0 0 0
012 (us): 0 0 0 0 0 0 0 0 0 0
013 (us): 0 0 0 0 0 0 0 0 0 0
014 (us): 0 0 0 0 0 0 0 0 0 0
015 (us): 0 0 0 0 0 0 0 0 0 0
016 (us): 0 0 0 0 0 0 0 0 0 0
017 (us): 0 0 0 0 0 0 0 0 0 0
018 (us): 0 0 0 0 0 0 0 0 0 0
019 (us): 0 0 0 0 0 0 0 0 0 0
020 (us): 0 0 0 0 0 0 0 0 0 0
021 (us): 0 0 0 0 0 0 0 0 0 0
022 (us): 0 0 0 0 0 0 0 0 0 0
023 (us): 0 0 0 0 0 0 0 0 0 0
024 (us): 0 0 0 0 0 0 0 0 0 0
025 (us): 0 0 0 0 0 0 0 0 0 0
026 (us): 0 0 0 0 0 0 0 0 0 0
027 (us): 0 0 0 0 0 0 0 0 0 0
028 (us): 0 0 0 0 0 0 0 0 0 0
029 (us): 0 0 0 0 0 0 0 0 0 0
030 (us): 0 0 0 0 0 0 0 0 0 0
031 (us): 0 0 0 0 0 0 0 0 0 0
032 (us): 0 0 0 0 0 0 0 0 0 0 (including overflows)
Minimum: 1 1 1 1 1 1 1 1 1 1 (us)
Average: 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 (us)
Maximum: 1 1 1 1 1 1 1 1 1 1 (us)
Max-Min: 0 0 0 0 0 0 0 0 0 0 (us)
Duration: 599.792 599.792 599.792 599.792 599.792 599.792 599.792 599.792
599.792 599.792 (sec)
** Affects: ubuntu
Importance: Undecided
Status: New
** Information type changed from Public to Private
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1948369
Title:
Enable nohz_full by default
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+bug/1948369/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
