rpmfusion:kernel modules decision: kmod

Thorsten Leemhuis fedora at leemhuis.info
Mon Sep 29 20:42:10 CEST 2008


On 28.09.2008 21:57, KH KH wrote:
> 2008/9/28 Thorsten Leemhuis <fedora at leemhuis.info>:
>> On 28.09.2008 09:24, Hans de Goede wrote:
>>> Thorsten will begin building all kmod's from livna in rpmfusion today.
>> Didn't get around to it.

Same today, sorry :-/

>> Some work for day job, some other stuff for RPM
>> Fusion, a little bit real life and building kmods for livna consumed all the
>> time... I'll start with it tomorrow (hopefully).
> I will be online tommorrow, so maybe I can help.
> At least the nvidia's shouldn't be imported.. the parallele
> installable version need to be discussed with the nvidia-kmod
> maintainers (firewing1 and me). For now it is still raw. (we need to
> have tools to handle the switch and fallbacks).

Firewing1 told me to just import everything just as it is. The other 
changes that can be done later.

That, btw, is IMHO the only way forward afaics, as the livna infra is 
more then damaged a bit right now -- thus I'd really want to close livna 
as soon as possible.

> i can renew the madwifi snapshot. i will probably set it as
> dead.package if it doesn't works with kernels 2.6.27 (assuming ath5k
> now work better except maybe for master mode with hostapd). (I will
> check this tommorrow)

Not that it matters much, but I'd prefer if we could keep madwifi for a 
little while longer if possible without to much hassle.

But yeah, if ath5k really does all the major things the users need then 
it might be time to drop madwifi.

> Then i can review some new kmod

Thanks.

> and help importing (mock test if needed) the older (livna) ones.

I think it might be easier for everyone if I work through that mess; 
especially as https://bugzilla.redhat.com/show_bug.cgi?id=464613 made 
things a bit more complicated in the past few days. :-/

> I will start submitting merge review of some freshrpms package i'm
> tracking for some time now (dvdrip perl dependencies and cinelerra
> patches upstreamed).

Just wondering: Why do we need merge reviews for those? Wasn't the rule 
we had: All those packages that were in just one of the three repos 
(dribble, freshrpms, livna) were automatically kind of approved without 
further review? But whatever: if you have enhanced packages in your repo 
then it'll be best for all if they get merged.

CU
knurd


More information about the rpmfusion-developers mailing list