From: Rich Brown <richb.hanover@gmail.com>
To: Sebastian Moeller <moeller0@gmx.de>
Cc: cerowrt-devel <cerowrt-devel@lists.bufferbloat.net>
Subject: Re: [Cerowrt-devel] Better results from CeroWrt 3.10.28-16
Date: Sun, 2 Mar 2014 20:38:01 -0500 [thread overview]
Message-ID: <FBB005D1-0180-482B-B72F-F49AECF51F80@gmail.com> (raw)
In-Reply-To: <EE50AA38-54D6-4ABB-81DE-D6CDA106BC76@gmx.de>
Sebastian,
> Currently there is no default auto tuning, the user needs to put the word auto into the "target" fields in the queue discipline tab in SQM (which is only exposed after selecting "show advanced configuration" and "show dangerous configuration", not very user friendly, at least it does not say 'Beware of the Leopard'...). So I am sure Rich would have noted that. I think we need a bit of testing of the auto-tuning feature and then make auto the default for ingest and egress. But for the current users I did not want to change the default without some testing (my track record of breaking things slightly comes to mind ;) )
Once we get the calculation correct, I would vote for the default being "auto" (that is, the ingress and egress strings would be pre-configure to "auto"). It would be the right thing for virtually anyone, but people could change it in the name of research.
While I'm lobbying, I would still like to see the second item in the Link Layer choices be "Ethernet with overhead" (instead of simply "Ethernet') Doing this makes it easier to explain:
"None" means none. CeroWrt does not make any per-packet adjustments at all.
"Ethernet with overhead" means that CeroWrt assumes that each packet has an additional overhead of N bytes that needs to be taken into account.
"ATM: ..." means that CeroWrt accounts for the incredibly arcane (stupid?) vagaries of the ATM 48-in-53 byte framing.
Thanks.
Rich
next prev parent reply other threads:[~2014-03-03 1:38 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-02 21:18 Rich Brown
2014-03-02 21:41 ` Dave Taht
2014-03-02 21:48 ` Sebastian Moeller
2014-03-03 1:38 ` Rich Brown [this message]
2014-03-05 18:52 ` Sebastian Moeller
2014-03-02 21:51 ` Aaron Wood
2014-03-02 22:19 ` Dave Taht
2014-03-03 1:27 ` Rich Brown
2014-03-03 1:41 ` Dave Taht
2014-03-03 1:43 ` Dave Taht
2014-03-03 1:49 ` Rich Brown
2014-03-03 2:00 ` Dave Taht
2014-03-03 2:28 ` Rich Brown
2014-03-03 8:51 ` Dave Taht
2014-03-03 8:53 ` Dave Taht
2014-03-03 13:35 ` Rich Brown
2014-03-03 13:43 ` Toke Høiland-Jørgensen
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/cerowrt-devel.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=FBB005D1-0180-482B-B72F-F49AECF51F80@gmail.com \
--to=richb.hanover@gmail.com \
--cc=cerowrt-devel@lists.bufferbloat.net \
--cc=moeller0@gmx.de \
/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