[Cerowrt-devel] breakage

[Cerowrt-devel] breakage

[Dave Taht]

as is kind of usual, me submitting patches upstream to openwrt breaks
downstream cerowrt when they are accepted.

I'm happy that several were accepted, too brain-dead to find and fix
all the conflicts induced as yet.

-- 
Dave Täht

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

Re: [Cerowrt-devel] breakage

[Outback Dingo]

On Sat, Oct 13, 2012 at 12:32 PM, Dave Taht <dave.taht@gmail.com> wrote:
> as is kind of usual, me submitting patches upstream to openwrt breaks
> downstream cerowrt when they are accepted.
>
> I'm happy that several were accepted, too brain-dead to find and fix
> all the conflicts induced as yet.

hahahaha, do you have a list of whats been accepted?

>
> --
> Dave Täht
>
> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel

Re: [Cerowrt-devel] breakage

[Dave Taht]

If I did, I'd be done now. It looks mostly to be:

iproute2 moved (I had patches to that): fixed
fq_codel and ecn fixes with changed patch names: fixed
package signing: stomped on thoroughly, don't know what to do with it yet
random number generation: I'm happy the 3.5 rng stuff was backported
into 3.3.8, however that disabled IRQF_RANDOM
and it's unclear to me what level of new driver support is needed to
make the new rng stuff work better.

I care about the random  number stuff a lot, (the previous situation
re real entropy and WPA was horrible) but I don't know enough about it
to tell if additional support is needed in the ath9k and ar71xx and
elsewhere to truly enable the new rng stuff. ?

I thought openwrt was freezing....

On Sat, Oct 13, 2012 at 9:36 AM, Outback Dingo <outbackdingo@gmail.com> wrote:
> On Sat, Oct 13, 2012 at 12:32 PM, Dave Taht <dave.taht@gmail.com> wrote:
>> as is kind of usual, me submitting patches upstream to openwrt breaks
>> downstream cerowrt when they are accepted.
>>
>> I'm happy that several were accepted, too brain-dead to find and fix
>> all the conflicts induced as yet.
>
> hahahaha, do you have a list of whats been accepted?
>
>>
>> --
>> Dave Täht
>>
>> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
>> _______________________________________________
>> Cerowrt-devel mailing list
>> Cerowrt-devel@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Dave Täht

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

Re: [Cerowrt-devel] breakage

[Dave Taht]

I lied, I had updated iproute2 to 3.5, and nuked the fq_codel backport, the move
in the package dir reverted it to 3.3.x...

still working on it, brain cells depleting rapidly...


On Sat, Oct 13, 2012 at 9:42 AM, Dave Taht <dave.taht@gmail.com> wrote:
> If I did, I'd be done now. It looks mostly to be:
>
> iproute2 moved (I had patches to that): fixed
> fq_codel and ecn fixes with changed patch names: fixed
> package signing: stomped on thoroughly, don't know what to do with it yet
> random number generation: I'm happy the 3.5 rng stuff was backported
> into 3.3.8, however that disabled IRQF_RANDOM
> and it's unclear to me what level of new driver support is needed to
> make the new rng stuff work better.
>
> I care about the random  number stuff a lot, (the previous situation
> re real entropy and WPA was horrible) but I don't know enough about it
> to tell if additional support is needed in the ath9k and ar71xx and
> elsewhere to truly enable the new rng stuff. ?
>
> I thought openwrt was freezing....
>
> On Sat, Oct 13, 2012 at 9:36 AM, Outback Dingo <outbackdingo@gmail.com> wrote:
>> On Sat, Oct 13, 2012 at 12:32 PM, Dave Taht <dave.taht@gmail.com> wrote:
>>> as is kind of usual, me submitting patches upstream to openwrt breaks
>>> downstream cerowrt when they are accepted.
>>>
>>> I'm happy that several were accepted, too brain-dead to find and fix
>>> all the conflicts induced as yet.
>>
>> hahahaha, do you have a list of whats been accepted?
>>
>>>
>>> --
>>> Dave Täht
>>>
>>> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
>>> _______________________________________________
>>> Cerowrt-devel mailing list
>>> Cerowrt-devel@lists.bufferbloat.net
>>> https://lists.bufferbloat.net/listinfo/cerowrt-devel
>
>
>
> --
> 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

Re: [Cerowrt-devel] breakage

[Dave Taht]

and the breakage is post-freeze behavior.

I had basically planned to rebase against the frozen openwrt tree for
at least a while, and maintain a 3.3.8 version through the end of the
year.

However we lack resources to really manage a freeze at all, and my
inclination is to keep moving forward as fast as possible onto 3.6 and
later. Got a decision point coming on whether to try for a stable long
term (for small values of long term) inclined version, or not.
Opinions?

"sugarland" was almost good enough to declare stable and long
term-ish. Certainly seems suitable as a default home gateway box.

So I'm fixing up the merge I just did in the ubnt tree I have, but
holding off on fiddling directly with cero's trees.


On Sat, Oct 13, 2012 at 9:43 AM, Dave Taht <dave.taht@gmail.com> wrote:
> I lied, I had updated iproute2 to 3.5, and nuked the fq_codel backport, the move
> in the package dir reverted it to 3.3.x...
>
> still working on it, brain cells depleting rapidly...
>
>
> On Sat, Oct 13, 2012 at 9:42 AM, Dave Taht <dave.taht@gmail.com> wrote:
>> If I did, I'd be done now. It looks mostly to be:
>>
>> iproute2 moved (I had patches to that): fixed
>> fq_codel and ecn fixes with changed patch names: fixed
>> package signing: stomped on thoroughly, don't know what to do with it yet
>> random number generation: I'm happy the 3.5 rng stuff was backported
>> into 3.3.8, however that disabled IRQF_RANDOM
>> and it's unclear to me what level of new driver support is needed to
>> make the new rng stuff work better.
>>
>> I care about the random  number stuff a lot, (the previous situation
>> re real entropy and WPA was horrible) but I don't know enough about it
>> to tell if additional support is needed in the ath9k and ar71xx and
>> elsewhere to truly enable the new rng stuff. ?
>>
>> I thought openwrt was freezing....
>>
>> On Sat, Oct 13, 2012 at 9:36 AM, Outback Dingo <outbackdingo@gmail.com> wrote:
>>> On Sat, Oct 13, 2012 at 12:32 PM, Dave Taht <dave.taht@gmail.com> wrote:
>>>> as is kind of usual, me submitting patches upstream to openwrt breaks
>>>> downstream cerowrt when they are accepted.
>>>>
>>>> I'm happy that several were accepted, too brain-dead to find and fix
>>>> all the conflicts induced as yet.
>>>
>>> hahahaha, do you have a list of whats been accepted?
>>>
>>>>
>>>> --
>>>> Dave Täht
>>>>
>>>> Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
>>>> _______________________________________________
>>>> Cerowrt-devel mailing list
>>>> Cerowrt-devel@lists.bufferbloat.net
>>>> https://lists.bufferbloat.net/listinfo/cerowrt-devel
>>
>>
>>
>> --
>> 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



-- 
Dave Täht

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