/srv/irclogs.ubuntu.com/2005/06/21/#ubuntu-kernel.txt

Micksasuspend-to-ram is working like a charm01:21
Micksaand all it right with the world once more01:21
zuljbailey: dont start hacking on the kernel? sheesh thats what i did :)01:58
jbaileyI didn't say don't start there.  I said it's possible not the best place to start.02:18
zulhehe..02:18
jbaileyAnd really, to start by doing kernel hacking requires a whole lot of drive.02:26
jbaileySomeone with it will do it regardless of what I say.02:26
jbailey(Esp as I don't mean to be discouraging.  It is a daunting place to start on its own without my help)02:26
zuloh i know...02:28
zulits a rough place to start if you are just starting out02:28
=== lamont__ [~lamont@rover3.mmjgroup.com] has joined #ubuntu-kernel
=== lamont kicks ipsec, hard.
lamontnow to figure out if I really care about the bug05:50
lamontnew oo.o2 and gcc-3.3 sigh05:51
lamontand gcc-3.405:51
lamontthat explains why it's taking forever. :=(05:51
dilingerjbailey: ping05:58
=== lamont could swear there was a way to have iptables look at the unencapsulated ipsec packet
fabbionemorning06:18
=== TMM [~hp@c51471f2c.cable.wanadoo.nl] has joined #ubuntu-kernel
=== dilinger [dilinger@mouth.voxel.net] has joined #ubuntu-kernel
=== amu [~amu@80.237.210.60] has joined #ubuntu-kernel
=== crimsun [~crimsun@crimsun.silver.supporter.pdpc] has joined #ubuntu-kernel
=== chmj [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-kernel
chmjfabbione, ping 09:52
fabbionepong09:53
chmjfabbione, is it possible to merge the last patch in kernel-debian--pre1,3--2.6.1 1.93 to the new playground ?09:55
fabbionechmj: i think so yes.. but that change needs to be reverted in that branch09:57
chmjok, I don't know my way around baz that much09:58
fabbionechmj: ok i will try to clean up10:01
jbaileydilinger: pong =)11:40
fabbionehey jbailey 11:45
jbaileyg'm Fabio11:46
=== doko_ [~doko___@dsl-084-059-033-037.arcor-ip.net] has joined #ubuntu-kernel
zulmorning01:17
fabbionehi zul01:18
zulhow is the external-modules going?01:19
fabbionei didn't work on it yet01:19
zulah01:19
fabbionedoing other stuff atm01:19
fabbionelooking at the new kernel-package and kernel-wedge01:20
chmjzul, where are the external-modules packages ? 01:49
chmjif there are any packages 01:49
fabbionethere are no packages01:51
jbaileyfabbione: *poke*01:58
fabbionejbailey: ?01:59
jbaileyI think in ppc32 mode this thing has gotten farther.01:59
jbaileyIt's got a bunch of XMLTO lines on the screen now.01:59
fabbionejbailey: why am i not surprised? :)01:59
fabbioneyeah that's it01:59
fabbionei think01:59
fabbionelet it finish01:59
jbaileyWill do.01:59
fabbionei will deal with it soon.. i think i know what it is01:59
jbaileyI think I'm going to have a nap.  I woke up crazy early this morning because of the cats.02:00
fabbionei noticed :)02:00
fabbionethanks a lot02:00
jbailey=)02:00
fabbionespeaking of which.. a nap is a really good idea02:18
fabbionezul: i did upload the new kernel-package02:19
chmju have cats ?02:19
fabbioneso you can build again02:19
fabbionechmj: no... i have a wife :P02:19
chmjwho has cats I assume ? 02:19
fabbionechmj: i am not following you :)02:20
fabbioneanwyay.. later02:20
chmjlater 02:20
=== lamont__ [~lamont@15.238.5.160] has joined #ubuntu-kernel
jbaileyfabbione: Build finished fine.04:12
dilingerjbailey: i was just curious what you thought regarding cdbs2 testing framework stuff04:14
jbaileydilinger: When I read it, it made my brain hurt, so I went to bed. =)04:15
jbaileyI'll take a look again today.04:16
dilingerjbailey: writing it made my brain hurt a little.  i should probably try and make it a bit more coherent or something ;)04:16
jbaileyI wrote the original testsuite, and it's basically the one from cdbs1 ported up.04:16
jbaileyI agree that it would be nice to make it a bit more generic.04:16
lamont__fabbione: ping04:19
jbaileyI do occasionally run the testsuite by hand while I'm hacking, largely with debugs turned on so that I can see what's happenning.04:19
dilingerjbailey: yea, basically the abi checking stuff fits right in w/ this..04:20
jbaileyRight.04:20
jbaileyI need to learn a bit more about how that's happening.04:20
jbaileyBut if we could unify it with stuff that we're developping collectively, then that would be lovely.04:20
zulfabbione: ping04:43
lamont__"BTW, nobody should be using transport mode over the Internet anyway."04:54
=== lamont__ wonders why not...
jbaileycontext?05:16
dilingeripsec?05:21
=== dilinger makes wild guesses
lamont__ipsec05:27
fabbionepong05:27
lamont__vs tunnel mode05:27
fabbionelamont__: dunno why.. but you can use tunnel mode between 2 hosts exactly as you use transport05:30
fabbionelamont__: i am pretty sure that what he sais is ok, given he is one of the Ipsex developer :)05:30
fabbionezul: pong?05:30
lamont__fabbione: ok05:30
zulfabbione: i made some changes my arch you can sync if you want05:32
fabbionezul: ok thanks05:32
lamont__fabbione: what I'd really like is a comprehensive tutorial on the subject. :-)05:35
fabbionelamont__: i remember i read all the RFC's 5 years ago... and i met who wrote them. you really don't want to go there.. do you? ;)05:37
jbaileylamont__: Step 1: Install checkpoint Step 2: Slit wrists.05:37
lamont__fabbione: all the stuff from 6 months ago is completely incorrect05:38
fabbionelamont__: probably the linux implementation sucks....05:38
fabbionei did use fbsd+racoon against an ericsson ipsec implementation at that time05:38
fabbioneand it was working05:39
dilingerheh05:39
dilingeripsex?05:39
fabbionebut i didn't even attempt firewalling05:39
fabbionedilinger: yeah.. isn't ipsec sexy??05:39
lamont__fabbione: that's the thing - none of the ipsec howto's address firewalling05:39
fabbionei even wwrote ipsex several times in my test reports for ericsson...05:39
fabbionenobody bothered to correct them :)05:39
dilingerhehe, nice05:39
lamont__Jun 14 09:39:57 localhost racoon: NOTIFY: the packet is retransmitted by 10.17.72.128[500] . 05:40
=== lamont__ grumbles
fabbionelamont__: right, because ipsec and firewalling are 2 separate issues...05:40
fabbionelamont__: one to protect data/connection , one to protect the host05:40
lamont__right.05:42
lamont__many of the ipsec tutorials tell you that to make ipsec work, you need iptables -A INPUT -p esp ACCEPT05:43
lamont__which isn't necessarily so05:43
=== gahan [gahan@coyote.fsh.hu] has joined #ubuntu-kernel
=== gahan [gahan@coyote.fsh.hu] has left #ubuntu-kernel []
lamont__what is /dev/vcsN ?06:27
zulvirtual console screen06:28
lamont__ah, ok06:28
=== Seveas [~seveas@seveas.demon.nl] has joined #ubuntu-kernel
=== thom [~thom@amnesiac.heapspace.net] has joined #ubuntu-kernel
fabbionehey thombot06:57
fabbioneok06:59
thomyo07:01
lamont__fabbione: note that you're making it harder and harder to build 2.6.12 on hoary. </grumble>07:06
thomfabbione: dude, ipw2100 is at 1.1.0 which is apparently the version we have07:21
zulfor breezy?07:26
fabbionelamont__: you are not even supposed to build 2.6.12 on hoary :)07:27
fabbionethom: is it???07:27
lamont__fabbione: yeah, and?07:27
fabbioneipw2100            | 1.1.0                      | ok     | 12/04/2005   | http://ipw2100.sourceforge.net/07:27
fabbioneoh yeah indeed07:27
fabbionelamont__: i know.. but there are a bunch of assholes sitting on #ubuntu-toolchain that keeps breking things... and i have to put pieces back together again07:28
fabbione;)07:28
lamont__gcc-3.4 (>= 3.4.4-0ubuntu6) [powerpc i386]  | gcc-3.4 (= 3.4.3-9ub07:33
lamont__untu4) [i386] , gcc-3.4 [!powerpc !i386] 07:33
=== lamont__ sighs - that just looks, well, wrong
fabbionei didn't do it :)07:34
lamont__that's my addition in between the | and the ,07:34
lamont__and makes hoary happy again07:34
zuli hate sql07:39
lamont__zul: I've never met anyone who claimed otherwise. 07:43
zulgood so im not alone07:46
=== RaverDK [~Ask@212.130.38.6] has joined #ubuntu-kernel
thombleah. EIP is at ieee8011_wx_get_scan08:44
lamont__fabbione: the ones doing the breaking are all here, too. :-)08:50
lamont__cache hit                             6509:05
lamont__cache miss                         1764209:05
lamont__grumble09:05
thomfabbione: right, this ipw2100 problem is a regression from 2.6.1009:06
jbaileyShould the kernel be part of ubuntu-minimal?09:25
fabbionethom: it's an upstream problem. ipw2100 is using an old version of ieee$randomnumbers09:27
fabbionei remember very well that from when i did the merge09:27
fabbioneso we need to wait upstream to sync09:27
fabbioneboth ipw2100 and 2200 have this ieee code duplication09:27
fabbioneand they often go out of sync09:28
fabbionelamont__: that was the point of my comment :P09:28
thomfabbione: grah, k09:29
lamont__6000 more misses, 20 more hits.  woot!09:42
zulfabbione: new inotify10:22
=== Seveas [~seveas@seveas.demon.nl] has joined #ubuntu-kernel
jbaileydilinger: Around?11:19
dilingeryep11:20
jbaileydilinger: cdbs2 thoughts for you.  Once we're done this testsuite and some basic docs.  What all do you think there is left before we can include it for some basic usage?  I know that the debhelper and autotools need a bunch of variables populated and help things written for them.  Is there other just basic functionality that's missing before people can at least start peeking at it for basic apps?11:21
dilingeryea, the main thing that's missing/broken is per-module pass stuff11:22
dilingerright now, if you have CDBS_MODULES := autotools, foo_CDBS_MODULES := autotools debhelper, debhelper gets run for both11:22
dilingeri think that's a bug i introduced, with the way i'm keeping track of stuff11:23
dilingerthat's the main thing that's remaining11:23
jbaileyNice. =)11:23
jbaileyI haven't dug through everything yet that you've done.11:23
dilingerafter that, we just start porting modules11:23
jbaileyMy initial targets for modules are just autotools, python, ant, gnome and kde.11:23
jbaileyThose are ones that I feel confident writing testscripts for.11:24
dilingerthe framework is there, debian/rules help will spit out all variables, hooks, etc; i'm not too concerned about the test suite stuff, but if we decide to make it built in (w/ the way i'd prefer, test-specific hooks), it wouldn't be too hard to add in there11:24
jbaileyI'm hoping to be more hardcore about features needing testcases for this so that we don't have the kind of fuckage debian had from robert in the last few months of cdbs.11:24
dilingerjbailey: that would be nice11:24
dilingeri'd certainly appreciate it if you pushed for the test stuff11:25
dilingerbecause i can pretty much guarantee how it'll go if i do it (first, i'll say i just want to get stuff working, so i won't concentrate on the test suite.  after stuff is working, i'll say things need enhancements before i can get to the test stuff.  finally, i'll say it's time for the test suite, and i'll get distracted and work on some other project that's bugging me)11:26
=== dilinger is very much motivated by itches

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!