[01:49] <Nafallo> zZzZ
[05:12] <forbin117> Hi, I have a question about saa7134 support, and wether I need to compile a new kernel to get it or not?
[11:13] <jes-o-mat> kylem: the backported patch from Cedric Schieli does not work for my machine :/
[11:50] <Mithrandir> zul: xen-source ftbfs on i386
[01:05] <zul> Mithrandir: already have a fix 
[02:21] <Nafallo> zul_: I don't think xen-image-2.6.19-1-generic-amd64 is for "Pentium Pro/Celeron/Pentium II/Pentium III/Pentium 4 machines" to be honest :-)
[02:31] <Nafallo> morning BenC :-)
[02:31] <BenC> good morning
[03:18] <zul_> Nafallo: i know will be fixed in the next upload
[04:41] <jes-o-mat> kylem: any news on the kernel image?
[04:42] <kylem> ... i told you, i have no idea what patch will fix it, you're going to have to experiment. i can't just take a whole new driver for -security...
[04:46] <kylem> i can guess and respin you one this afternoon if you'd like... but it would be more helpful if it can be narrowed down.
[04:48] <jes-o-mat> kylem: I'm gonna try to work out some things first
[05:05] <jes-o-mat> kylem: using the suggested patches from Cedric Schieli I get the following rejects: http://jesusch.de/~jesusch/tmp/rejects/
[05:06] <jes-o-mat> I think his work was based on an earlier version, that did not include your changes you made on the repo ~6 days ago
[06:21] <jes-o-mat> hi olifante :)
[08:32] <fabbione> kylem: ping?
[09:08] <kylem> fabbione, yo
[09:08] <kylem> fabbione, sorry, was eating lunch
[09:09] <fabbione> kylem: no problem.. can you please investigate why once again .20 the hpt366 module is missing from sparc udebs?
[09:12] <kylem> in a bit, sure.
[09:14] <fabbione> and can you make sure is there again by the next upload?
[09:15] <kylem> no, ben would only be able to say that.
[09:20] <fabbione> i could just upload it...
[09:20] <fabbione> bu then it would go async with git
[09:46] <zul> whee...going to hockey game tonight
[10:08] <kylem> fabbione, ok. i'll push it to our shared git repository and ping ben.
[10:09] <fabbione> kylem: thanks..
[10:09] <kylem> will turn up in ide-modules
[10:09] <fabbione> danke
[10:09] <kylem> fabbione, it would be helpful to have someone run automated diffs of /lib/modules between all 5 arches
[10:10] <fabbione> kylem: script it.. there are mirrors of archive on roockery or chinstrap
[10:10] <fabbione> i used to do that at my time
[10:10] <kylem> heh. i have enough work to do atm.
[10:10] <lifeless> hmm, no benC
[10:10] <kylem> lifeless, he's travelling atm.
[10:10] <lifeless> ah
[10:10] <kylem> lifeless, (to london for the osdl wireless summit)
[10:10] <lifeless> so remember that ipsets discussion at the start of the week ?
[10:11] <kylem> yes.
[10:12] <lifeless> turns out that it isn't able to be built fully externally without lots of gymnastics
[10:12] <lifeless> it needs an iptables rebuild with a makefile option changed, and a new kernel header installed
[10:12] <kylem> eh?
[10:14] <kylem> i don't see why needing a kernel header installed means it needs patching... sounds like they aren't trying hard enough. ;P
[10:14] <lifeless> hmm
[10:14] <lifeless> perhaps I'm using the wrong terms :)
[10:14] <lifeless> to get ipsets working you need to:
[10:14] <lifeless> build the module [can do trivially with kernel headers installed] 
[10:15] <lifeless> install the modules extra kernel header 
[10:15] <lifeless> patch the iptables deb source trivially and rebuild
[10:15] <kylem> oh.
[10:15] <kylem> ok. my bad, i thought you meant we needed to patch the kernel.
[10:16] <lifeless> FSVO patch. We do need to install an extra header into the space the kernel headers go
[10:16] <lifeless> anyhow, this means that ipsets will need its own copy of iptables built differently, unless we make ipsets be on by default
[10:17] <kylem> couple choices...
[10:17] <kylem> could build iptables in ipsets and divert it...
[10:17] <kylem> or could just make it the default as you said.
[10:18] <lifeless> so I'm here to whine and plead for ipsets to be the default :)
[10:18] <kylem> i dunno if i agree with that, not that my voice is really worth much.
[10:19] <Mithrandir> what's ipsets?  Like iptables, only rewritten?
[10:19] <lifeless> Mithrandir: its an iptables extension for handling very large or very dynamic groups of ip addresses
[10:19] <lifeless> kylem: well as one of the 2 kernel dudes, your voice counts heaps
[10:19] <lifeless> kylem: so if you say 'build another iptables binary and divery' -> will do so
[10:19] <kylem> not really...
[10:20] <kylem> my problem with it is, it seems to have been written a long time ago... and still not merged.
[10:20] <kylem> so clearly either something is wrong with it... or... something..
[10:20] <lifeless> kylem: ok, I'll get Jerub to ask around about that aspect
[10:21] <lifeless> Mithrandir: with ipsets you create a match on set inclusion/exclusion, then add and remove ips from the set
[10:21] <kylem> lifeless, otoh it doesn't seem like it could make a big impact, my concerns are mostly from the maintainability/security aspect.
[10:22] <lifeless> thats fine, I'll ask. I have to run now, thanks for the chat
[10:23] <kylem> ok. cheers.
[11:54] <[g2] > is there much kernel development on the 2.17 (edgy) kernel or is it mostly Feisty (2.6.20) ?
[12:02] <crimsun> [g2] : no.
[12:02] <crimsun> (for the former, that is. For the latter, yes.)