[02:28] <mjg59> BenC: Hmm. When I end up rsyncing a huge pack file, should I be doing something else?
[02:29] <infinity> Get more bandwidth?
[02:29] <mjg59> Ah, it's just finished
[02:33] <mjg59> fatal: Entry 'drivers/video/imacfb.c' would be overwritten by merge. Cannot merge.
[02:33] <mjg59> Mm? How do I avoid that?
[02:52] <fbond|away> well, at least git is ideologically superior to other RCS's
[11:01] <Mithrandir> BenC: unionfs bug.  When casper does an mv /root/sbin/readahead-list /root/sbin/readahead-list.disabled, /sbin suddenly turns into mode 0700.
[02:07] <zul> heylo
[02:57] <Mithrandir> BenC: did you see my unionfs bug above?
[03:05] <janimo> BenC: can you please revert the Radeon PCI id patch I suggested and it's in .17? While it makes 3D DRI work on said card it also sometimes causes X 99% cpu usage
[03:06] <janimo> and sometmes locks on resume
[03:06] <janimo> probbaly more that the ID pacth is needed (mesa,drm module) etc fixes which
[03:06] <janimo> are too much for dapper
[03:06] <janimo> sorry for not testing more thorougly before asking inclusion
[03:30] <BenC> Mithrandir: yeah
[03:30] <BenC> janimo: please send me an email so I don't forget
[03:31] <janimo> BenC: ok, there are otther radeon IDs which seem to behave similarly and have bugs in malone.those may need to be reverted too 
[03:31] <janimo> unless there's a way to blacklist them and not enable dri on them within X
[03:31] <janimo> I'll send the mail
[03:54] <Mithrandir> BenC: running Ubuntu kernels on auric and vore? :-P
[03:55] <BenC> of course :)
[04:20] <infinity> BenC: Seriously?
[04:21] <infinity> BenC: Did neuro have a fit about that?
[04:21] <Keybuk> Debian has run our Kernels for ages
[04:21] <Keybuk> like, 2 years
[04:22] <BenC> the sparc boxes were always running a custom kernel that I built
[04:22] <BenC> so using an ubuntu kernel really isn't much different :)
[04:23] <infinity> Fair enough.  The hppa buildds are in a similar boat.
[04:23] <infinity> But that's more because lamont's only got time to focus on one kernel at a time.
[04:24] <infinity> (Actually the m68k buildd mostly run custom kernels by me too... I often wonder how many of the SCC arch kernels in Debian can do much more than run the installer)
[04:24] <BenC> I don't know what is all in the debian 2.6.15 sparc kernel, but all I know is that it's only a 9Meg image, and our sparc kernel is 14megs
[04:24] <infinity> s/buildd/buildds/
[04:24] <BenC> I'm a little leary of that
[04:25] <infinity> BenC: Aren't you still, technically, a Debian sparc porter?  Perhaps you should look at their kernels and make them suck less?
[04:25] <infinity> (And yes, I should do the same for Debian arches I care about, but I've always been a custom kernel guy on Debian...)
[04:25] <infinity> Never ran a distribution kernel until I started running Ubuntu on my laptop.
[04:30] <BenC> same here
[04:30] <BenC> I always ran custom kernels till I became the ubuntu-kernel-guy :)
[04:30] <zul> heylo
[04:31] <BenC> I've been really disinterested in messing with debian stuff...I dropped my debian-sparc authorti a long time ago
[04:32] <BenC> my notification icon is telling me to reboot, and I'm not one to argue with widgets, so brb
[04:45] <mxpxpod> BenC: are you still using bcm43xx on your laptop?
[04:45] <BenC> yeah
[04:45] <mxpxpod> do you still coax it into 11mbs?
[04:46] <mxpxpod> mb/s
[04:47] <BenC> I've been using 24M for awhile
[04:47] <BenC> haven't tried 54M again
[04:48] <mxpxpod> how about network-manager?
[04:48] <BenC> haven't tried that
[04:49] <mxpxpod> ok
[05:00] <zul> BenC: kyle did a presentation to my local lug on eficent use of git, ill put it up somewhere if you want
[05:25] <janimo> zul, I'm interested in that too
[05:25] <janimo> all the tutorials I read so far are scary
[05:25] <janimo> if it were specifically written for the occasional ubuntu kernel contributor it would be even better
[05:30] <crimsun> the link to the tutorial on kernel.org/git and wiki.u.c/KernelGitGuide are good starters
[05:31] <crimsun> it's probably far easier if you set up a push server ; I have to use git-format-patch
[05:33] <janimo> crimsun, I find the commands very low level
[05:33] <janimo> I canot set upa push srever for bandwidth reasons
[05:34] <janimo> I could not find simple recepies like : cherrypick patch from given repo
[05:34] <janimo> those commit names are ugly. reminds me of tla frustrations of 2 years ago :)
[05:34] <crimsun> janimo: have you tried any of cogito, StGIT, etc.?
[05:35] <janimo> cogito I tried briefly
[05:35] <janimo> stgit no
[05:37] <janimo> crimsun, do you have 5 minutes now to tell me how you use git with alsa?
[05:37] <crimsun> sure
[05:37] <janimo> what I have now is git cloned ubuntu tree and ocasionally pull
[05:37] <janimo> and beneath same working dir git bracnhes
[05:38] <janimo> but from now on not much :)
[05:38] <janimo> do you keep all branches under the same directory?
[05:38] <crimsun> I generally don't worry about cwd at all
[05:38] <janimo> and swicth with gt checkout
[05:38] <crimsun> yep
[05:38] <crimsun> git branch foo && git checkout foo
[05:39] <janimo> git checkout -b foo
[05:39] <janimo> :)
[05:39] <janimo> how do you pick pacthes from upstream
[05:39] <janimo> do you give the SHA ids?
[05:39] <crimsun> oh, I can't do that for ALSA
[05:40] <crimsun> for the HDA work in particular, I have to hand-merge
[05:40] <janimo> :(
[05:40] <janimo> is there some command line query on remote branches
[05:40] <janimo> say I want to know diff between ubuntu tree and dri tree for file radeon_dri.c
[05:41] <janimo> right now I go to gitweb on kernel.org
[05:41] <janimo> that's kind of slow
[05:41] <crimsun> hmm, why not just pull in the dri tree as a branch?
[05:42] <janimo> hmm, I should try that . I was not sure how many upstream branches I can have locally
[05:42] <janimo> I mean orig/masters
[05:42] <janimo> but since they all should use the same objects underneath it may make sense
[05:43] <janimo> ok so you do have alsa git pulled as a branch too?
[05:43] <crimsun> no, but that would probably make a lot of sense, eh? :)
[05:43] <janimo> :)
[05:43] <janimo> related: I assume you test your patches by building the kernel
[05:44] <janimo> how did you make it not take forever
[05:44] <janimo> I touched config/i386/xx.config.disabled
[05:44] <janimo> for all but one 
[05:44] <janimo> but still seemed to build for hours and i had to stop
[05:44] <crimsun> I just rm the ones I don't need
[05:44] <janimo> last time I rm-d it did not make debs
[05:44] <janimo> cannot remember the errors
[05:45] <crimsun> oh, that's right, I remember why I don't have alsa git as a local branch
[05:45] <janimo> so just rm, no need to touch makefiles or rules?
[05:45] <crimsun> all devel work is done in cvs, and jaroslav pushes to git only occasionally
[05:45] <janimo> same for dri
[05:45] <janimo> so you actually take from cvs now?
[05:45] <crimsun> it's just much faster for me to trawl cvsweb and apply what's necessary
[05:46] <janimo> yeah
[05:46] <crimsun> nope, I don't touch Makefiles or rules
[05:46] <janimo> there must be a faster way that's why I think the tutorials are not telling us the whole story
[05:46] <crimsun> perhaps zul has pointers :)
[05:46] <janimo> yeah, I hope :)
[05:47] <janimo> do you also use a smaller config that makeallconfig I assume?
[05:47] <crimsun> nope, I use stock
[05:47] <janimo> oh, doesn;t it take ages to build?
[05:47] <crimsun> I have to make sure it at least builds on 686
[05:47] <crimsun> oh yes, it takes a good long while
[05:48] <janimo> you make debs with debuild or just plain make the kernel
[05:48] <crimsun> debs
[05:48] <janimo> debuild binary
[05:48] <janimo> ?
[05:49] <janimo> or full with clean and all that?
[05:49] <crimsun> just binary
[05:49] <janimo> thanks, I'll have to try the removal of config stuff
[05:49] <janimo> althoug it seems the .disabled suffix is the one checked for
[10:34] <TMM> hi all!
[10:34] <TMM> to repost my question from both -devel and -server :
[10:35] <TMM> what would me the chances of getting iscsi support into the -server kernel before dapper releases? slim? extremely small? near-zero? :) it pretty much doesn't touch any files, it just adds a couple
[10:36] <TMM> I only today noticed it isn't in there, and, we (our company) wants to use more ubuntu, and it really should be there (r) I am working on packaging the userspace tools, and, I am using the suse sles9 patchesset as a basis, I can confirm that I am using these very same drivers in a couple of VERY large environments, and they have shown no sign of any breakage
[10:36] <TMM> a seperate -source file in universe would be pretty bad imho, even under control of m-a
[10:42] <TMM> I am willing to do any heavy-lifting that might be required btw
[10:52] <TMM> I really need to go sleep now, if someone can help me, or wants more information or whatever please email me at hein-pieter.van.braam@ictivity.nl I basically know how to do the packaging, and have the equipment to test it all too. 
[10:52] <TMM> thanks
[11:27] <BenC> TMM: still around?