[Cerowrt-devel] Fwd: [OpenWrt-Devel] OpenWrt release name

[Dave Taht]

On Sat, Apr 4, 2015 at 2:56 PM, Felix Fietkau <nbd at openwrt.org> wrote:
> On 2015-04-04 23:50, Dave Taht wrote:
>> On Sat, Apr 4, 2015 at 12:49 PM, Felix Fietkau <nbd at openwrt.org> wrote:
>>> -------- Forwarded Message --------
>>> Subject: [OpenWrt-Devel] OpenWrt release name
>>> Date: Sat, 04 Apr 2015 20:45:56 +0200
>>> From: Felix Fietkau <nbd at openwrt.org>
>>> To: OpenWrt Development List <openwrt-devel at lists.openwrt.org>
>>>
>>> Hey OpenWrt community,
>>>
>>> We're still working hard on preparing the upcoming Chaos Calmer release,
>>> and the work on the first release candidate is almost done.
>>
>> I hope that the fixes for dnsmasq-dnssec and updated babel make it before
>> the final, but I have to say the nightlies have been superb of late. Great
>> going!

I note I tested the linksys AC1900 the other night. It worked quite
well. Will argue the mvneta driver needs BQL.

> Anything ready to merge yet?

C) babeld-1.6.0 is stablizing. After/during that...

work is needed to remove the current split between babels and babeld,
and I would love full procd support, but  don't know how to get there
from here.

B) dnsmasq has had several nasty bugs exposed in the last few weeks.

1) all versions with dnssec support break now that cloudflare is
signing a domain with
a CNTRL-Cxxx.cloudflare.org

This among other things, breaks www.ietf.org (starting the week that
ietf met, sigh)

It turns out a CNTRL-C in DNS is legal (WTF?), but dnsmasq dnssec
fails to validate due to prematurely not liking this domain.

We havent found anyone at cloudflare with IQ points yet that can fix
it, and this is so obviously a misconfiguration on their part. Even
then, the right answer is to allow that domain, and filter out the
evil octet in the logger where it belongs.

Fixed in dnsmasq-2.73rc3 and later. This fix is a candidate for a
backport into BB.

2) fixed: issues with breakage with DS records

This fix is also a candidate for BB.

3) There has been a long standing nearly impossible to reproduce
runaway on native ipv6
     in particular, that takes days to induce. I am ramping up a huge
number of boxes
     to try and pound it flat enough to find. Maybe it is gone
already. Dunno. This week,
     hopefully.

C) I have no intent to get cake into chaos calmer anytime, and
probably not in time for the release at all.

Too much work remains. Found a bug in either the scheduler or codel
part yesterday in particular, but lots more analysis and tuning left
to do.

> By the way, can you still reproduce the
> performance issues triggered by having hostapd running?

My last test had the hostapd issue still existing (80mbit/sec
forwarding speed loss).

It was against  r45228. (I have frozen against that while other issues
got resolved)

Documented here. Gave you the strace already.

http://comments.gmane.org/gmane.comp.embedded.cerowrt.devel/4163

80mbit max forwarding speed loss shown here:
http://snapon.lab.bufferbloat.net/~d/xmit_more_tcp_upload.svg

On yer side:

Getting anywhere with per-sta queuing on the ath9k?

Minstrel-blues? (been trying to get ahold of thomas for a month now...)

>
>>> We've decided to include the community in picking the name of the next
>>> major release after CC. After some internal discussions, we came up with
>>> the following two options:
>>>
>>> == Dirty Diamond ==
>>> 2 cl     Rum
>>> 2 cl     Vodka
>>> 2 cl     Tequila Silver
>>> 2 cl     Campari
>>> 2 cl     Curaçao Blue
>>> 2 cl     Lime syrup
>>> 2 1/2 cl Orange juice
>>>          Coca Cola
>>>
>>> == Dark Destroyer ==
>>> 2 oz     Malibu coconut rum
>>> 2 oz     Southern Comfort peach liqueur
>>> 3 oz     Lemonade
>>
>> I have had both of these drinks in my life... and left the bar wobbly
>> both times and incapable of coding.
>>
>> Request from a californian:  When you hit a double "G" release can you
>> try a "Green Goddess"?
>>
>> https://www.stickyguide.com/dispensaries/the-apothecarium/products/green-goddess-full-sun
>>
>> "Skunk #1 with a touch of sweet leaf Indica. She's a real gem; sparkly
>> light emerald flowers with bright orange hairs and aroma of ruby red
>> grapefruit and rose. This is a real toe-tapper. Expect a rushing head
>> effect. Good for going on a scavenger hunt, or embracing your OCD."
>>
>> Much less incapacitating.
> :)
>
> - Felix



-- 
Dave Täht
Let's make wifi fast, less jittery and reliable again!

https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb