[12:19] <BenC> kylem: I'm running into the same sort of crap in the rt2x00 drivers :/
[12:19] <BenC> The scan work queue has this check at the top of the handler
[12:20] <BenC>         if (unlikely(!rt2x00dev->scan))
[12:20] <BenC>                 return;
[12:20] <BenC> the work_struct is rt2x00dev->scan->work
[12:20] <BenC> so not only is it unlikely, it's seems to be just plain buggy if scan == NULL
[12:20] <kylem> indeed
[12:20] <BenC> because it only gets freed at the bottom of the handler
[12:21] <kylem> the wireless code is pretty much crap afaics.
[12:33] <BenC> we need to start a save-the-code foundation so drivers like this stand a chance of actually making it into the kernel proper
[12:35] <kylem> agreed
[12:35] <BenC> they're waiting for when a driver worth going into the kernel actually uses it
[12:38] <BenC> I think it's in the same git tree as dscape
[12:39] <kylem> hmm. ah well.
[12:40] <BenC> man, I know what I want to do to fix this rt2x00 the right way, but it's just to much damn work
[12:41] <kylem> if it's this bad forward porting, i wonder how bad backporting is going to be... 
[12:42] <zul> hey
[12:43] <BenC> fuck this, I'm doing it the right way
[12:43] <mjg59> kylem: It's maintained in wireless-dev
[12:44] <kylem> i should ping linville and find out wtf is up
[12:52] <BenC> I think dscape is on the wireless summit agenda
[12:53] <kylem> is there a url for this yet?
[12:58] <BenC> kylem: Don't think so, #wireless on the other network
[01:03] <kylem> k.
[02:55] <derekS> BenC: ping
[02:59] <derekS> I did whatyou said about take out quiet and splash, went to take a photo but the text wasn't readible. it was all gray dotted boxes and stuff
[02:59] <derekS> this happened before in the edgy release cycle (never dapper or earlier for me)
[02:59] <derekS> but fixed after *some* update
[04:27] <BenC> almost there
[04:31] <kkubasik> Hey, I'm still having issues with the bcm4
[04:31] <kkubasik> 3xx
[04:31] <kkubasik> driver in the latest feisty kernel
[04:31] <kkubasik> 0c:00.0 Network controller: Broadcom Corporation BCM4310 UART (rev 01)
[04:32] <BenC> wait till 2.6.20, which might upload this evening
[04:32] <kkubasik> I put the results of a modprobe bcm43xx (from dmesg) up in a pastebin
[04:32] <kkubasik> o, alright, sounds cool
[04:32] <BenC> usually works better if you put it in a bug report
[04:33] <BenC> IRC is too interactive for taking in bugs
[04:33] <kkubasik> just wanted to make sure it wasn't user error before filing a super-hardware specific bug report
[04:33] <BenC> what's the actual error?
[04:33] <BenC> do you have firmware installed?
[04:33] <BenC> did it work in edgy, and start failing in feisty?
[04:36] <kkubasik> it never worked
[04:36] <kkubasik> there's a bug already filed against edgy
[04:36] <kkubasik> the BCM4310 just doesn't work under linux
[04:37] <kkubasik> ndiswrapper has an irq conflict
[04:37] <kkubasik> [377600.228000]  ieee80211_crypt: registered algorithm 'NULL'
[04:37] <kkubasik> [377600.232000]  ieee80211: 802.11 data/management/control stack, git-1.1.13
[04:37] <kkubasik> [377600.232000]  ieee80211: Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>
[04:37] <kkubasik> [377600.280000]  bcm43xx driver
[04:37] <kkubasik> is the closest thing to an error
[04:39] <kkubasik> and yeah, I have all the firmware
[04:39] <kkubasik> /usr/share/bcm43xx-fwcutter/install_bcm43xx_firmware.sh
[04:39] <kkubasik> ran with no errors
[05:07] <derekS> BenC: almost there?
[05:28] <derekS> it still doesn't work, and i cna't take a pic
[05:51] <derekS> BenC: I can't get to a gui to add a bug in launchpad.... but I managed to get these.... do these help? http://slovin.info/image3.jpg http://slovin.info/image4.jpg
[05:53] <BenC> derekS: Both of those are 403
[05:55] <derekS> benc: lemme try to fix it
[05:58] <derekS> ahh permissions issues :)
[05:58] <derekS> all fixed
[05:59] <BenC> derekS: There's nothing wrong there other than maybe X isn't setup right, or it's broken somehow
[06:00] <BenC> the system booted successfully (as noted by the login prompt)
[06:00] <BenC> so there's no kernel error there
[06:00] <derekS> BenC: yeah, thats a previous kernel, lemme give you the present one, i am a moron
[06:00] <derekS> hold up
[06:00] <derekS> that wasn't the error i was describing before
[06:04] <BenC> infinity: ping
[06:10] <infinity> BenC: pong
[06:10] <BenC> infinity: Remember the lrm bug where it seemed like a newly installed one wouldn't have correctly run depmod on first boot?
[06:11] <infinity> Yep.
[06:11] <infinity> Did you just find it?
[06:11] <BenC> I just experienced it on 2.6.20 lrm over two reboots, and then it went away the first time I ran lrm-manager by hand :/
[06:12] <derekS> BenC: http://slovin.info/image6.jpg and http://slovin.info/image7.jpg these were taken with splash and quiet turned on.... i am getting you a screen shot of the unreadible text i get with it turned on
[06:12] <infinity> I love how much sense that bug doesn't make.
[06:12] <infinity> BenC: Is there any hope of you reproducing it and leaving the system/chroot broken?
[06:13] <BenC> I have one other machine that uses lrm so I'll give it a try there
[06:13] <BenC> derekS: I'm pretty sure I know what that one is, give 2.6.20 a try when it's available, like within a few days
[06:13] <derekS> BenC: http://slovin.info/image8.jpg it freezes printing "text" like this
[06:13] <derekS> BenC: ok... i can wait :)
[06:14] <BenC> that garbage I don't know about, but who knows
[06:14] <derekS> any idea why there is this nasty grub bug not showing text right?
[06:14] <derekS> it happened in edgy for a while, then it magically wnet away :)
[06:14] <derekS> BenC: any idea when you are going to try 2.6.20?
[06:14] <BenC> you talking about that last pic?
[06:15] <derekS> yea
[06:15] <BenC> if that last pic happened after the softlockup, it has nothing to do with grub
[06:15] <derekS> no the last pic is when quiet and splash are turned off
[06:15] <derekS> seperate boot
[06:15] <BenC> it's still not grub
[06:15] <derekS> after the softlockup nothing really happens (for as long as i have left it)
[06:15] <derekS> BenC: what would it be?
[06:15] <BenC> usplash
[06:16] <derekS> ahh, prob wrong settings?
[06:43] <derekS> BenC: so when do you predict .20 will be out?
[06:43] <BenC> derekS: I said within a few days
[06:43] <derekS> perfect thank you
[07:37] <bronson> My build fails at the end: failure: cannot open upload file ../linux-vserver-image-kdump_2.6.17.1-11.1_i386.deb for reading: No such file or directory
[07:39] <bronson> That's from dpkg-genchanges.
[07:40] <bronson> It's right; that .deb doesn't exist.
[07:40] <bronson> Does anyone know what would generate it?
[07:52] <bronson> Ah, I think it's because I changed the flavor...
[07:52] <bronson> Yep, that's it.  flavours = vserver-386 but crashdump = 386
[07:53] <bronson> spin spin spin...
[07:55] <bronson> That's interesting...  in debian/control it says: Kernel Crash Dump is only available on i386, x86_64(amd64) and PowerPC64 systems.
[07:57] <bronson> But as far as I can tell, a kdump kernel is only generated for 386.  There's no debian/config/amd64/crashdump or debian/config/powerpc/crashdump.
[07:58] <bronson> Is one of these wrong?  Or is my analysis wrong?  :)
[08:32] <dade`> BenC: ping
[11:03] <thom> mjg59: so, when i hotdock, i get all the usb ports and sound and video and shit, but no cdrom drive from the ultrabay
[11:54] <mjg59> thom: Hm
[11:54] <thom> works fine if i reboot, tho
[11:55] <mjg59> Bay events are harder
[11:55] <thom> uh, in that it sees the ultrabay, so it's hotdock specific and not something larger
[11:55] <thom> right
[12:38] <kkubasik> is the ubuntu-2.6.git repo somewhat synced against mainline kernel devel, or does it just catch up when there's a release
[12:39] <kkubasik> scratch that dumb question
[12:39] <kkubasik> git log
[02:59] <kylem> morning.
[03:01] <zul> morning
[03:54] <zul> yay doom
[04:05] <zul> BenC: you are free i have someone else testing for me
[04:05] <kylem> temporary doom? :)
[04:06] <BenC> zul: I'm still going to test toay
[04:06] <zul> sure
[04:07] <_MMA_> BenC. I installed Feisty on my laptop. I installed the linux-lowlatency meta. It installs the kernel but didnt update GRUB. Should I file a bug?
[04:08] <kylem> BenC, woo.
[04:08] <zul> BenC: you are nad your freaking bear :)
[04:08] <BenC> _MMA_: It probably updates grub, but -generic would come first...go to the grub menu and select it manually
[04:10] <_MMA_> Yea. I looked after I booted. It wasnt listed in menu.lst.
[04:10] <BenC> _MMA_: That's not likely the lowlatency kernel's fault...try running "sudo update-grub" manually and see
[04:10] <_MMA_> I had to add it manually.
[04:10] <_MMA_> Hmm..
[04:11] <_MMA_>  Can I remove my entry and still run "sudo update-grub"?
[04:12] <_MMA_> Ok. I removed my entry then "sudo update-grub". That worked.
[04:14] <BenC> _MMA_: your entry would get overwritten anyway, unless you added it below the "add your entries here" line
[04:16] <_MMA_> k
[04:17] <kylem> or more to the point, queuing three different ones...
[04:18] <BenC> kylem: Look in debian/bin/bens-build-scripts/
[04:19] <BenC> I use that for all my local machines, and for the dev machines like davis and faure
[04:19] <kylem> oooh. sweet.
[04:19] <BenC> kylem: There's no docs, and it may be hard to understand, but they work
[04:19] <BenC> kylem All the build-* scripts are local
[04:20] <BenC> edit build-common for the machines and settings
[04:20] <BenC> do_build and do_clean go on the remotes
[04:20] <kylem> sweet. this should cut a good chunk of time off my turn around.
[04:21] <BenC> kylem: Basically the remotes have a git repo, and "build-clean foo" pushes to the remote, and totally blows away any changes and checks out a fresh tree
[04:21] <BenC> kylem: My machines here all have /org/<dist>-<arch>/ubuntu-2.6
[04:22] <BenC> I have amd64 and i386 on the same machine, and it builds both in parallel
[04:22] <kylem> urgh. it totally didn't occur to me that i could push into a porter machine from my laptop.
[04:22] <BenC> build-monitor -v foo bar ...
[04:22] <BenC> that is the icing on the cake, so you can watch how the build is going
[04:23] <BenC> the builds occur in a screen, so you can always login and attach to it
[04:23] <kylem> sweet, thanks!
[04:23] <BenC> np
[04:59] <zul> BenC: now you have a fan club..
[04:59] <BenC> zul: I expect you to be president and take in club dues, %90 of which goes toward my hardware fund :)
[05:00] <zul> sweet!
[05:00] <zul> i dont get a cut though
[05:00] <BenC> %10 is for you
[05:00] <zul> even better
[05:00] <zul> it can go to diapers
[05:26] <fabbione> BenC: 
[05:27] <fabbione>    * x86: Revert patch that allows disabling hyper-threading.
[05:27] <fabbione>      - GIT-SHA a5443e2694749510381b4c54b95f5a3ee2656dcf
[05:27] <fabbione> why is that?
[05:27] <BenC> fabbione: Two reasons. 1) It wasn't even enabled in edgy
[05:27] <BenC> 2) Linus, and others said it was overkill...the SSL vulnerability has been worked around in openssl, and the fact is, it's only theoretical
[05:28] <fabbione> fair enough
[05:33] <fabbione> BenC: i cannot find the git commit to commit to fix futex_atomic_cmpxchg_inatomic in edgy-security
[05:33] <fabbione> kylem: also to you ^^
[05:33] <kylem> on powerpc or sparc?
[05:33] <fabbione> sparc
[05:34] <fabbione> not the implementation.. but the fix of the implementation
[05:34] <BenC> fabbione: I'm pretty sure I pulled that stuff in
[05:34] <fabbione> i did send you the patch via email
[05:34] <fabbione> i didn't put it in git
[05:34] <kylem> it's probably in -updates
[05:34] <BenC> we don't have anything in edgy-updates yet
[05:34] <fabbione> please don't tell me that you forgot about it
[05:34] <BenC> fabbione: I remember you asking me about it before, and I checked and did a git-cherry-pick or something
[05:35] <BenC> as we were talking about it
[05:35] <BenC> fabbione: Did you just check the changelog, or did you check to see if the patch is actually there?
[05:35] <fabbione> i am doing a git pull now
[05:35] <BenC> please do, I am almost 100% positive I grabbed that fix
[05:36] <BenC> If not, it's a big fuck on my part, and we'll do another -security upload with that before the end of the week
[05:37] <kylem> the only stuff i put in -security was the CVE fixes... it was probably my fault, i probably confused yours with the powerpc one.
[05:42] <fabbione> * refs/heads/origin: fast forward to branch 'master' of rsync://rsync.kernel.org/pub/scm/linux/kernel/git/bcollins/ubuntu-edgy
[05:42] <fabbione> i have this tree here
[05:43] <fabbione> the patch is in that tree
[05:43] <fabbione> commit 76e1615aec30973bea2b2261243b247c136d2d7e
[05:43] <BenC> sweet, than it's in edgy security
[05:43] <fabbione> kylem: can you check if it is also in your tree?
[05:43] <BenC> *then
[05:43] <fabbione> or the one that has been uploaded
[05:43] <kylem> i have, it's not.
[05:43] <BenC> fabbione: I did the upload from the git repo you pulled from
[05:44] <fabbione> you two don't agree
[05:44] <kylem> commit 6e57a3a89785692bd8d012d80f5ee210ab8e0b68
[05:44] <kylem> Author: David S. Miller <davem@sunset.davemloft.net>
[05:44] <kylem> Date:   Tue Mar 28 01:00:08 2006 -0800
[05:44] <kylem>     [SPARC64] : Implement futex_atomic_cmpxchg_inatomic().
[05:44] <kylem>     Signed-off-by: David S. Miller <davem@davemloft.net>
[05:44] <kylem> ok, maybe it is, but i didn't put it there.
[05:44] <fabbione>   * [SPARC64] : Fix futex_atomic_cmpxchg_inatomic implementation.
[05:44] <fabbione>     Upstream GIT-SHA: c7fed9d75074f7c243ec8ff2c55d04de2839a6f6
[05:44] <fabbione>     - Malone #68266
[05:45] <kylem> maybe my tree is out of date.
[05:45] <fabbione> this one is also part of CVE-2006-5648
[05:45] <fabbione> so ok.. i guess we need to define a better way of handling this stuff
[05:45] <fabbione> who is going to coordinate with pitti?
[05:46] <BenC> kylem: Can you pull to see if you get all that stuff?
[05:46] <kylem> yes
[05:46] <BenC> fabbione: We both do
[05:46] <fabbione> ok
[05:48] <fabbione> so.. i don't understand
[05:48] <fabbione> if that commit was there
[05:48] <fabbione> and Ben did the upload
[05:48] <fabbione> from that tree
[05:48] <fabbione> why did it disappeared?
[05:49] <kylem> hold still
[05:49] <BenC> what do you mean disappeared?
[05:49] <fabbione> BenC: kylem said that's not in the tree that has been uploaded?
[05:49] <BenC> no he didn't
[05:49] <BenC> he said it wasn't in his local git
[05:49] <kylem> *i didn't upload*
 i have, it's not.
[05:49] <BenC> I uploaded
[05:49] <fabbione> you all suck :P
[05:50] <zul> i dont
[05:50] <BenC> by all I hope you included yourself :P
[05:50] <fabbione> zul: you more than these 2 guys
[05:50] <fabbione> BenC: well yes of course
[05:50] <BenC> this is the suck channel
[05:50] <kylem> fabbione, my bad, i assumed i had it, but since you already pointed out it was done before i was an employee...
[05:50] <fabbione> #ubuntu-tehsux
[05:50] <fabbione> kylem: it was done even before you were born :P
[05:51] <kylem> oh sod off. :)
[05:51] <fabbione> MUHAHAHA
[05:51] <BenC> kylem: You cannot claim post-employment for problems :P
[05:51] <kylem> lol.
[05:51] <fabbione> ok.. well
[05:51] <kylem> fabbione, according to the tag it was uploaded, so there.
[05:51] <fabbione> then we need at least to notify pitti to correct the USN
[05:52] <fabbione> BenC: stalled on kylem and jbailey to give a .19 running kernel to infinity for the buildd
[05:53] <kylem> BenC, i'm too busy fixing upstream in the time i spend not-working...
[05:53] <fabbione> kylem: not-working? does that mean you have free time? we can fix that :)
[05:53] <kylem> fabbione, no, i don't really.
[05:54] <BenC> free-time for me is time I spend working on Ubuntu for free
[05:54] <fabbione> BenC: or feeding the cows
[05:55] <BenC> gotta feed my security guards
[05:55] <fabbione> BenC: eheheh
[05:56] <fabbione> ok.. i am back feeding my dinoshark
[06:56] <kylem> BenC, heh, nice electric fence.
[06:57] <ivoks> if someone is interested in test machine for marvell ide controler, let me know
[06:57] <ivoks> i'm wiling to provide unlimited access
[06:57] <BenC> ivoks: I have one, and we fixed it already :)
[06:57] <ivoks> oh, great :)
[06:57] <BenC> edgy-security contains the marvell driver
[06:57] <ivoks> since when?
[06:57] <zul> oh looky moo cows
[06:58] <BenC> like today
[06:58] <ivoks> today, up-to-date edgy didn't recognise it
[06:58] <BenC> you may have to update -security and try again
[06:58] <ivoks> eh, i'm on feisty allready :)
[06:58] <BenC> feisty already contains it
[06:59] <ivoks> hm...
[06:59] <ivoks> generic?
[06:59] <BenC> hmm...why is PATA_MARVELL disabled in feisty kernel :/
[06:59] <ivoks> eh :)
[07:00] <BenC> fixing it now
[07:01] <ivoks> great
[07:48] <dade`> well, can't see the battery
[07:48] <dade`> in the gnome-power-manager
[07:48] <dade`> ONLY using ubuntu kernel, the 2.6.19 vanilla works
[07:51] <BenC> dade`: Any ubuntu kernels, or just feisty?
[07:51] <dade`> feisty
[07:52] <dade`> 2.6.19
[07:52] <BenC> try 2.6.20, should be available in a day
[07:52] <dade`> has mactel patches and things like that ?
[07:58] <BenC> yes
[07:58] <BenC> looks like most of the mactel patches got merged in 2.6.20
[07:59] <dade`> ow yea
[08:00] <dade`> maybe suspend will work too
[08:02] <dade`> dade@choo:~$ lsmod |grep bluetooth
[08:02] <dade`> dade@choo:~$ 
[08:02] <dade`> ok ? now i move bluetooth mouse and it works :D
[08:02] <dade`> (but this means I can't turn off bluetooth :( )
[08:05] <BenC> must be working off of usb compat
[08:05] <dade`> yes
[08:06] <dade`> infact looks like a usb mouse
[08:06] <dade`> but uses battery power .
[08:38] <Floris> anyone here?
[08:39] <tepsipakki> umm, the tg3-backport didn't get in the latest dapper-kernel?
[08:40] <tepsipakki> wait a sec
[08:41] <tepsipakki> -27.50 is dated Nov 29 but released today?
[08:41] <Floris> anyone know how long before the nvidia drivers will get updated to 1.0-9742
[08:43] <BenC> And I had the answer all typed out too
[08:43] <tepsipakki> :)
[08:43] <tepsipakki> "when it is ready" ?-)
[08:43] <BenC> No, actually tomorrow :)
[08:43] <tepsipakki> haha
[08:46] <zul> er...so on kernel.org which one should we be pulling from to fix bugreports for dapper,edgy? ;)
[08:46] <BenC> zul: -updates
[08:47] <zul> ah ok
[08:48] <zul> your's or kyles?
[08:48] <BenC> mine
[08:49] <zul> ok because your dapper one is 2 month's old and kyle's is 13 days
[08:49] <zul> according to gitweb
[08:51] <BenC> I'll pull/push
[08:51] <zul> thanks, sorry for being a pain in the ass
[08:52] <markedwards> Can anyone tell me the status of the sky2 driver problems in the Edgy kernel?  It looks like it didn't get updated with the recent kernel update.
[08:53] <BenC> zul: Synced
[08:53] <zul> merci buckets
[08:54] <BenC> markedwards: That will go into edgy-proposed first
[08:54] <markedwards> BenC: Ah, thanks.  So perhaps in the next kernel update?
[08:55] <BenC> definitely the next one, just not sure when that will be :)
[08:55] <BenC> sweetness, 2.6.20 is built everywhere except ia64 so far
[08:55] <markedwards> Got it, thanks.  Just wondering.  I've got a script running to check my eth0 connection every two minutes and reboot it if there's no IP. That will hold me :-)
[09:00] <gnomefreak> BenC: anything you like user to leave on a kernel will not boot issue?
[09:00] <gnomefreak> in a bug
[09:19] <BenC> gnomefreak: Like where it stops booting...let is sit 3 minutes to see if busybox comes up
[09:19] <BenC> those sorts of things
[09:19] <BenC> ask them to boot without quite/splash options too
[09:19] <gnomefreak> well after running initramfs (to see if same issue) now no kernel boots
[09:20] <gnomefreak> hes in a bad spot and im not sure what he did to get that way
[09:20] <gnomefreak> bug #75655
[09:20] <gnomefreak> that is his bug
[09:22] <BenC> gnomefreak: Probably easy fix, I think
[09:22] <BenC> he needs to boot rescue from CD
[09:22] <BenC> I'll put this in the bug report
[09:22] <gnomefreak> k
[09:23] <gnomefreak> we no longer have safe kernel mode in feisty (just FYI)
[09:24] <BenC> We still have rescue mode on the CD right? :)
[09:24] <gnomefreak> i dont know i cant use livecd
[09:25] <gnomefreak> i have to look at alt. cd to see if it has it
[09:25] <gnomefreak> it should
[09:25] <gnomefreak> alt. shouldnt have it iirc
[09:27] <BenC> alt should have a menu option in the installer to mount and chroot into a partition
[09:27] <gnomefreak> ah i will look at it tonight sometime after i get caught up
[09:32] <BenC> Keybuk: ping
[09:32] <Keybuk> BenC: yo
[09:32] <BenC> Keybuk: Any more info on that broadcom firmware?
[09:33] <Keybuk> BenC: the site is fujitsu siemens
[09:34] <Keybuk> their driver licence implies that you could redistribute the entire driver file unmodified if the other user accepts the EULA as well
[09:34] <Keybuk> except the "entire driver file" is 90MB!
[09:34] <BenC> oh, damn
[09:39] <BenC> Keybuk: Here's one that's only 0.26Meg
[09:39] <BenC> http://download.fujitsu-siemens.com/download/ShowDescription.asp?SoftwareGUID=6E4EBBDD-D5F8-440B-B556-D345D134E52E
[09:39] <BenC> Translation of the EULA seems to be the same as you describe
[09:41] <Keybuk> is that the same firmware?
[09:42] <BenC> bcmwl5.sys is the driver
[09:42] <BenC> and it's in there
[09:43] <BenC> hmm, fwcutter doesn't work with the .sys in this driver
[09:43] <BenC> wait, I'm reading the output wrong
[09:44] <BenC> bcm43xx-fwcutter can cut the firmware out of BCMWL564.SYS
[09:44] <BenC> sweet
[09:44] <bronson_> Searching for GRUB installation directory ... found: /boot/grub
[09:44] <bronson_> dpkg: error processing linux-vserver-image-2.6.17-11-vserver-generic (--install):
[09:45] <bronson_>   subprocess post-installation script returned error exit status 128
[09:45] <bronson_> Anyone know offhand what I can check?  update-grub runs without errors.
[09:46] <BenC> no idea
[09:46] <BenC> add -x to the postinst script
[09:46] <BenC> like #!/bin/sh -x
[09:47] <bronson_> ...and then do another build?
[09:47] <bronson_> OK, will do.
[09:48] <BenC> Keybuk: Do you think we'll need to pop up an EULA for this driver?
[09:50] <Keybuk> BenC: we could put it in debian/copyright <g>
[09:50] <BenC> well, we assume people using all the rest of the stuff agree to the GPL without a popup
[09:51] <BenC> s/GPL/whatever license/
[09:51] <Keybuk> yeah
[09:51] <Keybuk> popup on first use?
[09:51] <BenC> "third commits itself expressly"
[09:51] <BenC> would have to
[09:59] <BenC> Keybul: One thing I noticed is that it doesn't say you can't modify it
[10:01] <zul>  /msg BenC you know that queen song "im going slightly mad"?
[10:02] <BenC> zul: Yes :)
[10:02] <zul> see...
[10:02] <zul> maybe i should pay more attention to detail
[10:03] <ajmitch> heh
[10:46] <bronson_> Wait, wtf?  I haven't changed this package and now it installs great.
[10:47] <bronson_> I used to get that postinst exited with error 128 every time.
[10:47] <bronson_> I'm still using the exact .deb that I've always been using and --purging between attempts.
[10:49] <bronson_> Oh good.  Got the error again.
[10:49] <bronson_> Great, I love non-deterministic package installs.  :(
[10:49] <bronson_> All I'm doing is dpkg --purge / dpkg --install cycles. 
[10:56] <Keybuk> BenC: true, but the way copyright works, it has to say you *can*
[10:56] <Keybuk> I'm not even sure that licence as is would get past elmo :p
[10:57] <BenC> hehe
[11:02] <Keybuk> but maybe it just needs a "LA LA LA" <g>
[11:02] <Keybuk> personally, I was all for shipping the firmware, and claiming that "every other download site on the net does, so why can't we?"
[11:02] <Keybuk> and "it's all in good faith, y'r honour"
[12:00] <bronson_> Well, I sure don't get it.  My postinst returns 0 every time (at least, it executes "exit 0" at line 1295 every time), yet sometimes dpkg claims that it returned error 128.
[12:00] <bronson_> Guess I'll ask on ubuntu-devel...  I'm stumped.