From: Dave Taht <dave.taht@gmail.com>
To: Cake List <cake@lists.bufferbloat.net>
Subject: [Cake] armbian and linux 4.19 and cake
Date: Mon, 21 Jan 2019 16:40:04 -0800 [thread overview]
Message-ID: <CAA93jw4aPCKOmaGmqyG+aKU=PaXoUgtm_ds6qdipXA9jCBrgvA@mail.gmail.com> (raw)
I have a whole bunch of hackerboards gradually getting more and more
useful with ever more current kernels. I noticed armbian had updated
to 4.19 on the nano-pi hw today.... which is a quad core a53.
it bottlenecks in softirqd 1 no matter what I do (device driver
writers should try and think about more cores...)
The default is pfifo fast on a single mq. Ironically they build cake,
but not fq_codel
Ping (ms) ICMP : 8.23 8.36 ms 350
Ping (ms) UDP BE : 666.96 8.61 ms 343
Ping (ms) UDP BK : 664.27 8.82 ms 322
Ping (ms) UDP EF : 667.07 8.59 ms 332
Ping (ms) avg : 666.10 8.63 ms 352
TCP download BE : 199.67 200.87 Mbits/s 298
TCP download BK : 200.05 201.43 Mbits/s 299
TCP download CS5 : 207.60 206.68 Mbits/s 298
TCP download EF : 202.81 201.90 Mbits/s 298
TCP download avg : 202.53 202.77 Mbits/s 300
TCP download sum : 810.13 811.07 Mbits/s 300
TCP totals : 1121.50 1121.82 Mbits/s 300
TCP upload BE : 75.07 74.55 Mbits/s 253
TCP upload BK : 79.43 79.88 Mbits/s 263
TCP upload CS5 : 78.00 78.31 Mbits/s 271
TCP upload EF : 78.87 77.78 Mbits/s 270
TCP upload avg : 77.84 77.43 Mbits/s 300
TCP upload sum : 311.37 309.70 Mbits/s 300
And with cake...
Starting rrul test. Expected run time: 70 seconds.
Data file written to ./rrul-2019-01-21T163453.465592.cake.flent.gz.
Summary of rrul test run 'cake' (at 2019-01-22 00:34:53.465592):
avg median # data pts
Ping (ms) ICMP : 19.16 22.10 ms 349
Ping (ms) UDP BE : 574.54 18.24 ms 161
Ping (ms) UDP BK : 468.37 17.77 ms 124
Ping (ms) UDP EF : 576.97 17.40 ms 145
Ping (ms) avg : 539.96 21.22 ms 352
TCP download BE : 361.03 353.08 Mbits/s 300
TCP download BK : 42.61 41.67 Mbits/s 300
TCP download CS5 : 297.66 309.86 Mbits/s 300
TCP download EF : 156.38 161.39 Mbits/s 300
TCP download avg : 214.42 215.40 Mbits/s 300
TCP download sum : 857.68 861.60 Mbits/s 300
TCP totals : 927.55 919.47 Mbits/s 300
TCP upload BE : 18.12 17.87 Mbits/s 157
TCP upload BK : 17.45 17.78 Mbits/s 164
TCP upload CS5 : 17.03 16.45 Mbits/s 177
TCP upload EF : 17.27 14.72 Mbits/s 184
TCP upload avg : 17.47 14.63 Mbits/s 300
TCP upload sum : 69.87 58.51 Mbits/s 300
--
Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740
reply other threads:[~2019-01-22 0:40 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://lists.bufferbloat.net/postorius/lists/cake.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAA93jw4aPCKOmaGmqyG+aKU=PaXoUgtm_ds6qdipXA9jCBrgvA@mail.gmail.com' \
--to=dave.taht@gmail.com \
--cc=cake@lists.bufferbloat.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox