[12:27] <elcasey> anyone home? I've got a pretty serious issue with Feisty
[12:28] <jdong> I am accompanying elcasey on this one
[12:28] <elcasey> yay! I have an escort to the danc...never mind
[12:28] <jdong> possible Feisty kernel regression, kernel appears to hardlock when probing his chipset
[12:28] <jdong> (storage chipset)
[12:28] <jdong> hence system doesn't boot
[12:28] <jdong> reproduced with Feisty LiveCD
[12:29] <elcasey> initally I upgraded from Edgy, but LiveCD won't boot, either
[12:29] <jdong> the chipset reportedly worked fine in Edgy
[12:29] <jdong> elcasey: do you know what your storage chipset is?
[12:29] <elcasey> no, lemme find out
[12:31] <elcasey> it's Gigabyte brand
[12:31] <elcasey> the spec sheet just says "GIGABYTE SATA II Controller"
[12:31] <elcasey> so it's not Promise or anything
[12:33] <jdong> elcasey: can you get on a Edgy or something and lspci?
[12:33] <elcasey> I don't have an Edgy liveCD
[12:33] <elcasey> I have full access to all my partitions from Windows, though
[12:35] <jdong> elcasey: do you have any kind of Linux LiveCD?
[12:35] <jdong> (that can boot up)
[12:36] <elcasey> cool, found my Edgy Live
[12:36] <elcasey> i'll reboot and come back...I don't know if my NIC will work from a LiveCD, though, certainly not from Edgy
[12:36] <elcasey> it's a serious process to get this Ralink working
[12:37] <elcasey> wait, I have the hackin9 live distro!
[12:38] <jdong> elcasey: anything that can get you back here, with a mountable hard drive and a lspci command :D
[12:38] <elcasey> anyway, lemme try the two livecds and see if any of them work
[12:39] <elcasey> worst case, I'll install Solaris 10 and try that...I have a spare 10GB partition
[12:49] <elcasey> jdong: I managed to boot 2.6.17-10 in recovery mode
[12:49] <elcasey> it hung at the bootsplash earlier today
[12:49] <elcasey> but I have the lspci info
[12:50] <jdong> elcasey: I can't believe that worked
[12:50] <elcasey> neither can I
[12:50] <jdong> Feisty's not supposed to boot with edgy kernels.
[12:50] <elcasey> for all I know, Feisty will boot now
[12:50] <jdong> hehe hope for it
[12:50] <elcasey> BUT, the hackin9 livecd didn't boot, it got the same error(s) when it tried to load the drives
[12:50] <jdong> wow
[12:50] <jdong> hmm
[12:50] <elcasey> and it uses a 2.6.17-11
[12:50] <jdong> that's real interesting
[12:51] <elcasey> it's also Debian-based...and just like my install with -11, it dumped me to a BusyBox/ash prompt
[12:51] <elcasey> I started to think it was everywhere, but *everything* works in Windows
[12:51] <elcasey> shall I try booting feisty?
[12:52] <jdong> meh give it a shit
[12:52] <jdong> shot*
[12:52] <jdong> ugh
[12:52] <elcasey> heh, 10-4
[12:52] <elcasey> if it fails, I'll go back to XP and send/pastebin you my lspci.log?
[12:53] <elcasey> I just piped the output to a file
[12:59] <elcasey> no dice with either version of 2.6.20-9
[01:00] <jdong> what kind of error do you get booting up?
[01:00] <jdong> that would be useful
[01:00] <jdong> i.e. a recovery mode screencap
[01:01] <elcasey> is there any way to force it to output all the startup stuff a logfile?
[01:01] <elcasey> it doesn't give me errors, it just halts and all my HDDs spin down.
[01:01] <jdong> umm, the boot messages stop somewhere.
[01:02] <jdong> that's what I mean by error
[01:02] <elcasey> It seems like it halts on discovering USB devices, but even if I unplug my USB HDD, it still halts
[01:02] <jdong> no you can't log stuff before the system boots
[01:02] <elcasey> I mean that it's not giving me a *specific* error message or code
[01:02] <elcasey> it just...stops
[01:02] <jdong> elcasey: yeah, screencap WHERE it stops.
[01:03] <elcasey> how? Just PrtScr?
[01:03] <elcasey> well whatever I cap with PrtScr will be gone when I reboot
[01:03] <jdong> digital camera
[01:03] <elcasey> lol
[01:03] <elcasey> ok
[01:03] <jdong> or patience and a keyboard :D
[01:04] <jdong> or Vista wreck ignition
[01:04] <elcasey> I think I'll take some photos
[01:05] <elcasey> I'll try to pause it and see if I catch anything else
[01:05] <elcasey> brb
[01:14] <elcasey> jdong: I booted full-on into 2.6.10, but it's got "issues"
[01:15] <jdong> heh what kind?
[01:16] <elcasey> sda1 (Windows) doesn't mount and Beryl doesn't run at all (not concerned about that, currently)
[01:16] <elcasey> it's just a bit "off"
[01:16] <elcasey> presumably because the new kernel and some package replacements jacked some stuff around
[01:16] <elcasey> but here's my lspci output - http://paste.ubuntu-nl.org/10112/
[01:17] <jdong> well old kernels aren't supposed to work in Edgy
[01:17] <jdong> err feisty
[01:17] <jdong> I am totally shocked that it boots at all
[01:17] <jdong> can you apt-get update and apt-get install ubuntu-desktop and apt-get dist-upgrade?
[01:18] <elcasey> lemme check
[01:18] <elcasey> it's hitting the repos, and they all say "feisty"
[01:20] <elcasey> i'll definitely be doing a fresh install after my get my new HDD, but it'll be nice to have a working Linux system in the meantime (assuming tonight' kernel update makes some progress and/or old kernel keeps working)
[01:20] <elcasey> dist-upgrade works, too
[01:21] <elcasey> it's downloading quite a few packagaes
[01:23] <jdong> elcasey: yikes jmicron chipset
[01:23] <elcasey> jdong: you still need those screenshots?
[01:23] <jdong> BenC can probably enlighten you more on the state of jmicron
[01:23] <jdong> elcasey: but do file a bug report and attach the screenie
[01:24] <elcasey> which screenshot? It doesn't always hang in the same place
[01:25] <jdong> then attach two :)
[01:25] <elcasey> ok...I think I took like five :P
[01:29] <elcasey> jdong: what package should I report the bug in?
[01:29] <jdong> linux-souce-2.6.20
[01:29] <elcasey> thanks
[01:36] <elcasey> jdong: https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/91797
[01:41] <jdong> elcasey: your eyes need to re-train
[01:41] <jdong> look at your 00004.jpg
[01:41] <jdong> look about 8 lines down
[01:41] <elcasey> one sec
[01:42] <jdong> bug 84964
[01:42] <elcasey> ah ha
[01:43] <elcasey> checking out the forum thread now
[01:43] <elcasey> sheesh, not exactly an isolated problem
[01:44] <elcasey> all-SATA + USB = Feisty hates you?
[01:45] <jdong> no
[01:45] <jdong> it's a JMicron controller regression as it seems
[01:46] <jdong> but it's a confirmed critical bug.
[01:46] <elcasey> any way for me to delete/retract my report?
[01:47] <crimsun> reject it
[01:47] <kylem> mark it as a dupe of 84964
[01:47] <jdong> I already did
[01:48] <elcasey> ok, thanks
[09:44] <Mithrandir> BenC: I have an UVF request for git-core 1.5 ; do you have any thoughts on that?
[09:51] <fabbione> Mithrandir: it's probably good to have new git
[12:18] <fforw> hi..
[12:18] <fforw> what would be the best place to report problems with a new kernel?
[12:42] <BenC> fforw: launchpad
[12:45] <fabbione> hey BenC 
[12:48] <zul> hey guys
[01:40] <abogani> BenC: May i disturb you for two very simple and fast questions about my spec (https://wiki.ubuntu.com/RealTime)?
[01:40] <BenC> abogani: Sure
[01:40] <abogani> I want to release all necessary packages to offer a complete solution for RT-Preempt in Ubuntu.
[01:41] <abogani> 1) If i release all source packages of the all binary packages i'm ok with license (in particulary with restrcited-modules), right?  2) I should change name to my packages: What is, in you opinion,  better name? realtime-image (realtime-headers, realtime-source, realtime-restricted-modules etc) or linuxrt-image (linuxrt-headers, linuxrt-source, linuxrt-restricted-modules etc) ?
[01:50] <abogani> BenC: Do you have any suggestions?
[01:50] <BenC> abogani: 1) Yes, 2) Use a specific flavour like -realtime
[01:51] <BenC> your suggestion is probably too confusing for users, plus I doubt you really want to build linuxrt-image-{generic,386,server,server-bigiron,lowlatency}
[01:52] <BenC> So you'd have linux-image-2.6.20-10-realtime
[01:52] <BenC> probably best to base it on the -generic config
[01:53] <BenC> abogani: It would also mean you could do something intelligent like build-dep on linux-source-2.6.20 to build your kernels, just apply a patch for real time, and create the deb's (like what xen source does in universe)
[01:53] <BenC> makes it easier to rebuild against newer kernel source too
[01:56] <abogani> BenC I understand (i already use this flavours like naming), i cloned you git repositary and build my kernel with a AUTOBUILD=1 fakeroot debian/rules binary-debs flavours="realtime" :-)
[01:56] <zul> scarey....since when xen became the model ;)
[01:57] <abogani> BenC: My problem are (naming) source packages that not should conflict with official source packages.
[01:57] <BenC> abogani: If you do as I suggested, you wont have a full linux tree source package
[01:58] <BenC> but if you must do that (and I strongly suggest you don't :) then yeah, linux-rt-source-2.6.20 is a good name, but I wouldn't carry that name to the packages
[02:08] <abogani> BenC: Ok. If i want to follow your suggestion (and thus not release a linux-rt-source-2.6.20) how i set source field in the my kernel package? I set this to linux-source-2.6.20 as like official kernel? People don't expect to find all source in source package (in this case you package don't have RT-preempt source obviously)? I don't break package policy? Sorry Ben i'm a stupid person. :-(
[02:09] <BenC> abogani: Download the xen source package and check it out. You still have a source package, but it does not include the entire linux source tree. It build-depends on linux-source-2.6.20, unpacks it during build, applies the rt patches, and builds a -realtime flavour from it
[02:09] <mrec> BenC: do you directly take the v4l-dvb repository from linuxtv.org for the ubuntu kernel?
[02:09] <BenC> mrec: No, we use stock, unless there some important fixes which we might git-cherrypick from that repo
[02:10] <BenC> In the past mkrufky has collected fixes for us into a repo we've pulled from
[02:10] <abogani> BenC: I will do it, thank you very much!
[02:10] <BenC> abogani: Np
[02:11] <BenC> abogani: If you feel like writing a wiki howto for people who want to do kernel flavours like this, I can link it from the KernelTeam pages
[02:11] <mrec> BenC: regarding my problems with some devs there, did you do anything? seems like the post of another developer calmed done the whole story at least
[02:15] <mrec> the other question I have is regarding the ubuntu installation CD is there any installation CD with a recent kernel out there?
[02:15] <mrec> it simply doesn't detect the SATA drive for that testmachine.
[02:15] <abogani> BenC: Ok
[03:11] <crimsun> BenC: thanks for the merges
[03:17] <BenC> crimsun: Np, thanks for the patches
[03:17] <Mithrandir> BenC: if you could ack or nack a new git-core, that'd be nice.  I don't really mind either way.
[03:17] <BenC> Mithrandir: makes no difference to me, but I can't see it hurting anything
[03:18] <Mithrandir> BenC: ok.
[04:04] <kylem> BenC, git 1.5 changes the format of the .git dir...
[04:05] <Mithrandir> kylem: not by default.
[04:05] <BenC> kylem: does it change existing dir's?
[04:05] <kylem> no.
[04:05] <Mithrandir> at least not by default from how I read the LWN article
[04:05] <kylem> Mithrandir, ah, it does it on hera.
[04:05] <kylem> Mithrandir, which is how i found out about it when some of my scripts stopped working. :)
[04:05] <BenC> kylem: Do we need new git to work with hera repo's?
[04:05] <kylem> no.
[04:06] <kylem> i still say we should have 1.5 in feisty tho.
[04:06] <kylem> might as well be current.
[04:08] <Mithrandir> I've approved the UVFe request, so we should get it
[04:08] <kylem> ok.
[05:07] <zul> right Im off to have a baby see you guys later..
[05:21] <fabbione> it sounds like zul is going to give birth and not his wife....
[05:23] <Lure> fabbione: ;-), probably the first
[06:25] <BenC> crimsun: I have a build failure in one of your patches
[06:25] <crimsun> BenC: patch_analog?
[06:25] <crimsun> or hda_intel, hmm
[06:25] <BenC>         ucontrol->value.integer.value[0]  = ac97->spec.ad18xx.swap_mic_linein;
[06:26] <BenC> swap_mic_linein isn't defined, and neither is ad18xx
[06:26] <crimsun> gah, let me pull that
[06:26] <BenC> crimsun: Want me to push my current tree to have a look?
[06:26] <crimsun> yes please
[06:26] <BenC> planning an upload tonight, so trying to get some test builds done
[06:27] <crimsun> ok
[06:27] <BenC> crimsun: BTW, would sabdfl's sound being broken in -10 (regressed from -9) be fixed with the patches you sent?
[06:27] <crimsun> BenC: possibly, though I don't remember Mark's SSIDs
[06:28] <crimsun> I _think_ he used AD1986A?
[06:30] <crimsun> oh wait, did Matt Z report that one?
[06:30] <crimsun> Maybe I'm looking under the wrong reporter
[06:39] <crimsun> right, bug 69529
[06:47] <BenC> crimsun: pushed to kernel.org
[06:48] <crimsun> BenC: thanks
[06:51] <crimsun> BenC: do you know which hardware sabdfl was referring to for his sound regression in 2.6.20-10.17?
[06:52] <BenC> crimsun: trying to find out
[06:52] <crimsun> thanks
[06:57] <BenC> crimsun: 00:1b.0 Audio device: Intel Corporation 82801G (ICH7 Family) High Definition Audio
[06:58] <crimsun> BenC: can you get the subvendor and subdevice, please?
[06:58] <kylem> it's an X60, iirc
[07:04] <BenC> crimsun: I'm worried that he's gone missing...I'll keep trying
[07:04] <crimsun> BenC: thanks.
[07:11] <BenC> crimsun: How soon do you think before I can at least get the compile fix?
[07:11] <crimsun> just emailed to you
[07:12] <BenC> crimsun: Thanks
[07:12] <crimsun> sorry for the thinko
[07:14] <BenC> np
[07:14] <BenC> crimsun: Do you have difficulty test compiling these patches?
[07:15] <crimsun> BenC: not normally. I was up late. :/
[07:15] <BenC> It's only like once in a blue moon you send me a thinko, so it's not a huge deal, but if you could just do a test SUBDIRS=sound compile it might help
[07:15] <crimsun> sure, I'll use SUBDIRS=sound from now on, thanks
[07:16] <BenC> crimsun: usually I do "mkdir /tmp/build; cat debian/config/i386/config{,.generic} > /tmp/build/.config; make O=/tmp/build oldconfig; make O=/tmp/build prepare; make O=/tmp/build SUBDIRS=foo modules"
[07:17] <BenC> then you can just rm -rf /tmp/build when you're done and not add cruft to your src tree
[07:17] <crimsun> noted.
[07:19] <BenC> crimsun: Any chance alsa is considering moving to git still? :)
[07:20] <crimsun> BenC: not likely
[07:21] <BenC> crimsun: Would make it so much easier for us to just pull their tree :/
[07:21] <BenC> wonder if we can make a strong enough case for it that they might do it...I would think the effort they spend getting the patches to Linus would have been enough
[07:22] <crimsun> well, Jaroslav does have linux/kernel/git/perex/alsa.git
[07:22] <crimsun> I think you've pulled from that one before
[07:22] <BenC> is there a branch in there that is stable enough to keep in sync with?
[07:23] <BenC> I think last time i messed with it, I just copied the entire tree instead of actually pulling
[07:25] <crimsun> I don't see one that seems updated
[10:29] <mrec> does anyone here work on the installation cds?
[10:30] <mrec> the default kernel which comes with the cds simply doesn't support the hardware we have here, it would be nice to support it out of the box
[10:31] <BenC> mrec: would be nice to know what hardware that is
[10:48] <mrec> BenC: sorry took a while, I'm booting it and dumping the hardware list
[10:49] <ivoks> i noticed a regreesion with bluetooth since edgy :/
[10:50] <ivoks> like... i can't turn it on :D
[10:52] <BenC> ivoks: I have no bluetooth to test...is it maybe an rfkill thing?
[10:52] <BenC> is it showing up in lsusb?
[10:52] <ivoks> i'm in 2.6.17 now
[10:53] <ivoks> it works here, it's software switch
[10:53] <ivoks> it shows up in lsusb here
[10:54] <ivoks>       bInterfaceSubClass      1 Radio Frequency
[10:54] <ivoks>       bInterfaceProtocol      1 Bluetooth
[10:54] <BenC> ivoks: Let me know if it doesn't show up under feisty
[10:54] <ivoks> i'm rebooting right now :)
[10:58] <ivoks> no, it doesn't show up
[11:10] <mrec> BenC: http://rafb.net/p/nUlDLa61.html
[11:21] <BenC> mrec: That doesn't tell me which hw isn't working :)
[11:22] <mrec> harddisk isn't recognized :)
[11:22] <mrec> sorry
[11:28] <BenC> Assuming this is the ATI IDE?
[11:34] <mrec> yes
[11:35] <BenC> and it worked in edgy?
[11:35] <BenC> Can you send the dmesg from edgy?
[11:36] <mrec> it didn't work
[11:36] <mrec> I didn't go on with it but I'm interested in getting it supported out of the box
[11:39] <mrec> BenC: what do you need of dmesg?
[11:41] <johanbr> mrec: I have a very similar controller in my laptop (possibly identical) and it works fine in Feisty.
[11:41] <mrec> johanbr: the problem is I cannot go on with the installation since the harddisk isn't detected
[11:42] <BenC> mrec: All of it
[11:42] <BenC> mrec: So neither edgy nor feisty works with it?
[11:43] <johanbr> mrec: Oh, I see. My installation is upgraded from Edgy (or maybe even Dapper, don't remember).
[11:44] <mrec> BenC: the installation cd is supposed to be edgy
[11:44] <BenC> mrec: Can you test if feisty works, just so I can see if it's something that can be backported?
[11:46] <mrec> where can I get feisty?
[11:47] <stgraber> cdimage.ubuntu.com/releases/feisty/
[11:49] <mrec> ok grabbing it