[Make-wifi-fast] [Bloat] [Cake] New board that looks interesting

Fri Dec 18 18:48:46 EST 2020

I have, finally.  It's been running for a week or so, now.

OpenWRT was an _adventure_.  The board is UEFI, not standard bios.  And
while it will merrily boot OpenWRT's non-uefi images off of USB, it won't
boot the non-UEFI setup from the internal storage (I'm using the eMMC).  So
_that_ was fun (and I made some dumb mistakes that were especially fun to
correct.

But it's running OpenWRT 19.07 (and a UEFI bootloader before grub that's
from ToT OpenWRT).

Anyway, I have cake running, 950Mbps ingress and 35Mbps egress (modem is
provisioned at 1.3G ingress, and a bit over 35Mbps egress).  fq_codel was
defaulted, in multi-queue mode.  While I'm using cake on ingress, my local
link hasn't been hitting the limiter very often:

                Tin 0
  thresh        950Mbit
  target          1.5ms
  interval       30.0ms
  pk_delay         22us
  av_delay          9us
  sp_delay          2us
  backlog            0b
  pkts        243608193
  bytes    250748364896
  way_inds     13167720
  way_miss      1245030
  way_cols            0
  drops            1075
  marks             101
  ack_drop            0
  sp_flows            0
  bk_flows            1
  un_flows            0
  max_len         69876
  quantum          1514

Given that most of the hosts that I interact with are only about 10-15ms
away, I'm probably going to change the interval target to better match that.

Interestingly, while it has a pair of multiqueue NICs (i211s), the igbe
driver isn't configuring them for RSS.  Both output queues are being used,
but not the ingress queues:

wan interface:
     tx_queue_0_packets: 56635989
     tx_queue_1_packets: 39777210
     rx_queue_0_packets: 243646072
     rx_queue_1_packets: 0

lan interface:
     tx_queue_0_packets: 85047897
     tx_queue_1_packets: 162004500
     rx_queue_0_packets: 111174855
     rx_queue_1_packets: 0

Since I have housemates that don't appreciate me messing with the network
during their meetings, I haven't gotten around to poking more deeply at
that (or at experimenting with running cake on two ingress queues).

That being said, I bench-tested this before I put it into operation and was
able to see 940Mbps of iperf goodput through cake and NAT...  Took all of a
core, though (and that core was still cold and therefore potentially able
to boost to 2.5GHz).  I haven't determined how long it will take to
thermally throttle, and if bandwidth suffers as a result.

Pretty happy with it so far, though.

On Sun, Apr 26, 2020 at 7:46 PM Dave Taht <dave.taht at gmail.com> wrote:

> anyone got around to hacking on this board yet?
>
> On Sat, Apr 4, 2020 at 9:27 AM Aaron Wood <woody77 at gmail.com> wrote:
> >
> > The comparison of chipset performance link (to OpemWRT forums) that went
> out had this chip, the J4105 as the fastest.  Able to do a gigabit with
> cake (nearly able to do it in both directions).
> >
> > I think this has replaced the apu2 as the board I’m going with as my
> edge router.
> >
> > On Sat, Apr 4, 2020 at 9:10 AM Dave Taht <dave.taht at gmail.com> wrote:
> >>
> >> Historically I've found the "Celeron" chips rather weak, but it's just
> >> a brand. I haven't the foggiest idea how well this variant will
> >> perform.
> >>
> >> The intel ethernet chips are best of breed in linux, however. It's
> >> been my hope that the 211 variant with the timed networking support
> >> would show up in the field (sch_etx) so we could fiddle with that,
> >> (the apu2s aren't using that version) but I cannot for the life of me
> >> remember the right keywords to look it up at the moment. this feature
> >> lets you program when a packet emerges from the driver and is sort of
> >> a whole new ballgame when it comes to scheduling - there hasn't been
> >> an aqm designed for it, and you can do fq by playing tricks with the
> >> sent timestamp.
> >>
> >> All the other features look rather nice on this board.
> >>
> >> On Sat, Apr 4, 2020 at 7:47 AM David P. Reed <dpreed at deepplum.com>
> wrote:
> >> >
> >> > Thanks! I ordered one just now. In my experience, this company does
> rather neat stuff. Their XMOS based microphone array (ReSpeaker) is really
> useful. What's the state of play in Linux/OpenWRT for Intel 9560
> capabilities regarding AQM?
> >> >
> >> > On Saturday, April 4, 2020 12:12am, "Aaron Wood" <woody77 at gmail.com>
> said:
> >> >
> >> > > _______________________________________________
> >> > > Cake mailing list
> >> > > Cake at lists.bufferbloat.net
> >> > > https://lists.bufferbloat.net/listinfo/cake
> >> > > https://www.seeedstudio.com/ODYSSEY-X86J4105800-p-4445.html
> >> > >
> >> > > quad-core Celeron J4105 1.5-2.5 GHz x64
> >> > > 8GB Ram
> >> > > 2x i211t intel ethernet controllers
> >> > > intel 9560 802.11ac (wave2) wifi/bluetooth chipset
> >> > > intel built-in graphics
> >> > > onboard ARM Cortex-M0 and RPi & Arduino headers
> >> > > m.2 and PCIe adapters
> >> > > <$200
> >> > >
> >> >
> >> >
> >> > _______________________________________________
> >> > Bloat mailing list
> >> > Bloat at lists.bufferbloat.net
> >> > https://lists.bufferbloat.net/listinfo/bloat
> >>
> >>
> >>
> >> --
> >> Make Music, Not War
> >>
> >> Dave Täht
> >> CTO, TekLibre, LLC
> >> http://www.teklibre.com
> >> Tel: 1-831-435-0729
> >
> > --
> > - Sent from my iPhone.
>
>
>
> --
> Make Music, Not War
>
> Dave Täht
> CTO, TekLibre, LLC
> http://www.teklibre.com
> Tel: 1-831-435-0729
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/make-wifi-fast/attachments/20201218/2c5052ab/attachment.html>