Development issues regarding the cerowrt test router project
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: cerowrt-devel@lists.bufferbloat.net
Subject: Re: [Cerowrt-devel] cerowrt-3.6.9-3 test release
Date: Fri, 7 Dec 2012 09:20:21 +0100	[thread overview]
Message-ID: <CAA93jw7Gk+eQpB+UZ+W=taG55MxRF8Hr3g5aNO5UapTt8Yq+AQ@mail.gmail.com> (raw)
In-Reply-To: <CAA93jw7yG=bg2YMVonzWsSH+gCGkompt9ULMaGVcUd650C0YXA@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 2262 bytes --]

Probably the biggest bug in this first attempt: opkg was not included.
This is because the opkg swalker had produced had support for package
signing (in cerowrt 3.3.x), and I didn't port the patches forward...
and then I forgot to include the original opkg in this build!

Felix had reviewed the package signing stuff and felt that it was too
large to use. Certainly part of the opkg signing stuff could be shrunk
(it brings in an unnecessary library), but cutting it as far down as
felix wanted is going to be hard, and worse, doing so will require a
major audit as to the actual security of a non-gpg based signing
system. I'm in a quandary here. Personally I'd prefer to leverage gpg
for package signing, and leverage openssl, which we use anyway.

Also, some new unaligned exception traps have surfaced with ipv6. They
are not bad - only about 1200 on a 60 second rrul test with 24.5/5.5
ceroshaper, but enough to dramatically affect ipv6 latencies, and
affect fq_codel. (see attached)

Aside from that, ad-hoc mode is broken on wifi on either my test box
(iwl x86), or the ar71xx hardware - an interface comes up, associates
with another cell, then fails to transmit packets. I'll be bringing up
another cerowrt box to isolate this today.

I also get utterly dismal performance from same laptop to the router
over wifi, I really am hoping it's the !@#! iwl card, not the
ath9k....

the rrul test (12 netperfs) CAN indeed be run directly to a wndr3800,
but it pays to disable netserver startup from xinetd and run it as a
standalone daemon. It eats an enormous amount of cpu and should only
be used when trying stuff at 100Mbit or less...

Haven't tried the new (BFQ) block I/O scheduler (which might help on
the xinetd startup issue)

On Thu, Dec 6, 2012 at 2:50 PM, Dave Taht <dave.taht@gmail.com> wrote:
> Under absolutely no circumstances try this on your default home
> gateway. ipv6 support needs surgery, other tools are in flux...
>
> http://snapon.lab.bufferbloat.net/~cero2/cerowrt/3.6/3.6.9-3/
>
> --
> Dave Täht
>
> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html



-- 
Dave Täht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html

[-- Attachment #2: 6vs4.ps --]
[-- Type: application/postscript, Size: 132958 bytes --]

  reply	other threads:[~2012-12-07  8:20 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-12-06 13:50 Dave Taht
2012-12-07  8:20 ` Dave Taht [this message]
2012-12-08 17:53   ` Robert Bradley

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='CAA93jw7Gk+eQpB+UZ+W=taG55MxRF8Hr3g5aNO5UapTt8Yq+AQ@mail.gmail.com' \
    --to=dave.taht@gmail.com \
    --cc=cerowrt-devel@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