Date: Tue, 7 Jul 2020 21:52:14 -0700 From: Mark Millard <marklmi@yahoo.com> To: Oleksandr Tymoshenko <gonzo@bluezbox.com> Cc: Peter Jeremy <peter@rulingia.com>, freebsd-arm <freebsd-arm@freebsd.org> Subject: Re: freebsd-arm Digest, Vol 740, Issue 7 (Rock64 Ethernet testing) Message-ID: <CC52258E-513F-438F-A634-21FFA35CEA45@yahoo.com> In-Reply-To: <20200708021226.GA77884@bluezbox.com> References: <mailman.75.1593950402.45034.freebsd-arm@freebsd.org> <1731fbded28.10a3342f0357159.8148813293316485882@fkardame.com> <20200706204707.GA94158@bluezbox.com> <0A2E974E-39D3-46C8-8791-3BD914EBE7E9@yahoo.com> <0C77695E-A9D0-410A-B105-5B69823E17E2@yahoo.com> <EB92D6CC-940A-4429-A257-7D17955B8379@yahoo.com> <EFA17CEC-A2A2-4B0D-B63F-DB4E98CB2672@yahoo.com> <3289DA3D-03FB-43BD-9A6A-956AC0A03B59@yahoo.com> <20200708021226.GA77884@bluezbox.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2020-Jul-7, at 19:12, Oleksandr Tymoshenko <gonzo at bluezbox.com> = wrote: > Mark Millard (marklmi at yahoo.com) wrote: >> Any chance that the delays (or other parameters) depend >> on the operating temperature(s) of some parts? >>=20 >> If yes, then some of the following about the Rock64 V2 >> context that I have access to might be relevant to >> explaining my already reported V2 results (not much >> for Retr): >>=20 >> A) The Rock64 has a "case" that is really just a top >> and a bottom with posts: open on all 4 sides. >>=20 >> B) It has a fan blowing down on the board from the >> top. >>=20 >> C) It has a heat sink on the SOC, which the fan blows >> on directly. >>=20 >> D) It has a heat sink on the RAM, which the fan also >> blows on directly. >>=20 >> (I've not dealt with a more modern non-debug kernel >> build yet. It still may be some time before I deal >> with that.) >=20 > Temperature is not likely to be a factor in the delay values. > Rock64 V2 has a known issue with Gigabit ethernet stability: >=20 > https://forum.pine64.org/showthread.php?tid=3D7545 > https://sanisimov.com/2019/08/fixing-rock64-v2-gigabit-ethernet/ >=20 > Althought judging by description it's more like an almost complete > network lock-up and not performance degradation. >=20 > I received another board with RK3328 today and will investigate > the issue further. Okay. Looks like I should have copied iperf3 output from the server side as well: somewhat different information. The output was still available from the earlier runs so here it is . . . The modern debug-kernel runs: Accepted connection from 192.168.1.109, port 47111 [ 5] local 192.168.1.122 port 5201 connected to 192.168.1.109 port = 17015 [ ID] Interval Transfer Bitrate [ 5] 0.00-1.00 sec 17.5 MBytes 147 Mbits/sec =20 [ 5] 1.00-2.00 sec 45.3 MBytes 380 Mbits/sec =20 [ 5] 2.00-3.00 sec 44.9 MBytes 376 Mbits/sec =20 [ 5] 3.00-4.00 sec 45.2 MBytes 379 Mbits/sec =20 [ 5] 4.00-5.00 sec 44.9 MBytes 377 Mbits/sec =20 [ 5] 5.00-6.00 sec 45.1 MBytes 379 Mbits/sec =20 [ 5] 6.00-7.00 sec 44.5 MBytes 373 Mbits/sec =20 [ 5] 7.00-8.00 sec 45.0 MBytes 378 Mbits/sec =20 [ 5] 8.00-9.00 sec 44.9 MBytes 377 Mbits/sec =20 [ 5] 9.00-10.00 sec 44.5 MBytes 373 Mbits/sec =20 [ 5] 10.00-10.62 sec 27.9 MBytes 379 Mbits/sec =20 - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate [ 5] 0.00-10.62 sec 450 MBytes 355 Mbits/sec = receiver Accepted connection from 192.168.1.109, port 22375 [ 5] local 192.168.1.122 port 5201 connected to 192.168.1.109 port = 54738 [ ID] Interval Transfer Bitrate Retr Cwnd [ 5] 0.00-1.00 sec 24.7 MBytes 207 Mbits/sec 0 265 KBytes = =20 [ 5] 1.00-2.00 sec 61.6 MBytes 517 Mbits/sec 4 211 KBytes = =20 [ 5] 2.00-3.00 sec 61.4 MBytes 515 Mbits/sec 1 352 KBytes = =20 [ 5] 3.00-4.00 sec 61.3 MBytes 514 Mbits/sec 4 269 KBytes = =20 [ 5] 4.00-5.00 sec 61.4 MBytes 515 Mbits/sec 2 355 KBytes = =20 [ 5] 5.00-6.00 sec 61.3 MBytes 514 Mbits/sec 3 304 KBytes = =20 [ 5] 6.00-7.00 sec 61.3 MBytes 514 Mbits/sec 2 327 KBytes = =20 [ 5] 7.00-8.00 sec 61.4 MBytes 515 Mbits/sec 5 278 KBytes = =20 [ 5] 8.00-9.00 sec 61.4 MBytes 515 Mbits/sec 2 393 KBytes = =20 [ 5] 9.00-10.00 sec 61.4 MBytes 515 Mbits/sec 3 284 KBytes = =20 [ 5] 10.00-10.61 sec 37.3 MBytes 514 Mbits/sec 2 282 KBytes = =20 - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate Retr [ 5] 0.00-10.61 sec 614 MBytes 486 Mbits/sec 28 = sender (So a fairly consistent Retr rate.) The non-debug head -r360311 kernel runs: Accepted connection from 192.168.1.109, port 46431 [ 5] local 192.168.1.122 port 5201 connected to 192.168.1.109 port = 39541 [ ID] Interval Transfer Bitrate [ 5] 0.00-1.00 sec 50.3 MBytes 422 Mbits/sec =20 [ 5] 1.00-2.00 sec 72.7 MBytes 610 Mbits/sec =20 [ 5] 2.00-3.00 sec 72.9 MBytes 611 Mbits/sec =20 [ 5] 3.00-4.00 sec 72.8 MBytes 611 Mbits/sec =20 [ 5] 4.00-5.00 sec 72.9 MBytes 611 Mbits/sec =20 [ 5] 5.00-6.00 sec 72.9 MBytes 611 Mbits/sec =20 [ 5] 6.00-7.00 sec 72.8 MBytes 611 Mbits/sec =20 [ 5] 7.00-8.00 sec 72.8 MBytes 610 Mbits/sec =20 [ 5] 8.00-9.00 sec 72.8 MBytes 610 Mbits/sec =20 [ 5] 9.00-10.00 sec 72.8 MBytes 610 Mbits/sec =20 [ 5] 10.00-10.32 sec 23.4 MBytes 612 Mbits/sec =20 - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate [ 5] 0.00-10.32 sec 729 MBytes 593 Mbits/sec = receiver Accepted connection from 192.168.1.109, port 40223 [ 5] local 192.168.1.122 port 5201 connected to 192.168.1.109 port = 50696 [ ID] Interval Transfer Bitrate Retr Cwnd [ 5] 0.00-1.00 sec 78.5 MBytes 659 Mbits/sec 0 480 KBytes = =20 [ 5] 1.00-2.00 sec 113 MBytes 945 Mbits/sec 0 747 KBytes = =20 [ 5] 2.00-3.00 sec 112 MBytes 941 Mbits/sec 0 940 KBytes = =20 [ 5] 3.00-4.00 sec 84.5 MBytes 709 Mbits/sec 52 368 KBytes = =20 [ 5] 4.00-5.00 sec 112 MBytes 941 Mbits/sec 0 681 KBytes = =20 [ 5] 5.00-6.00 sec 112 MBytes 941 Mbits/sec 0 889 KBytes = =20 [ 5] 6.00-7.00 sec 112 MBytes 942 Mbits/sec 0 1.03 MBytes = =20 [ 5] 7.00-8.00 sec 112 MBytes 942 Mbits/sec 0 1.17 MBytes = =20 [ 5] 8.00-9.00 sec 112 MBytes 941 Mbits/sec 0 125 KBytes = =20 [ 5] 9.00-10.00 sec 112 MBytes 941 Mbits/sec 3 586 KBytes = =20 [ 5] 10.00-10.31 sec 34.7 MBytes 942 Mbits/sec 0 667 KBytes = =20 - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate Retr [ 5] 0.00-10.31 sec 1.07 GBytes 892 Mbits/sec 55 = sender So: Larger total Retr than the modern debug kernel case but not a fairly consistent rate of Retr values. I've still not dealt with updating to a modern non-debug environment to test it. =3D=3D=3D Mark Millard marklmi at yahoo.com ( dsl-only.net went away in early 2018-Mar)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CC52258E-513F-438F-A634-21FFA35CEA45>