[00:08] <jameswf> would someone be so kind as to just tar up their net/bluetooth and email it over
[01:39] <mtaylor> morning lovely people... intrepid doesn't seem to have linux-image-debug-* packages ... is there somewhere I should be looking new, or do I just suck?
[01:40] <trichobezoar> Whoa there now.
[01:42]  * lifeless prods randomly
[01:42] <lifeless> lamont: ^ any idea re: mtaylors query?
[01:43] <lamont> I thought they wound up in universe or something
[01:45] <mtaylor> mtaylor@drizzle-x4450:~$ apt-cache search linux-image-debug
[01:45] <mtaylor> nothing
[01:45] <lamont> they're certainly buildable...
[01:45]  * mtaylor added the ddebs repos and all
[01:45] <lamont> but no, I haven't been following the kernel build output for a while
[01:48] <mtaylor> darn
[01:49]  * mtaylor just needs a vmlinux file for oprofile use :(
[03:46] <mase_work> i hey guys , can those mainline kernel ppa's be installed in intrepid ?
[11:05] <Keybuk> stgraber: hey, around?
[11:12] <Keybuk> smb_tp: around?
[11:12] <Kano> hi, did anybody else notice that 2.6.30 does not boot when a dvb-c card is installed in the system
[11:13] <Kano> i have in every system such a card, a bit bad to have to remove it when i want to test a new kernel...
[11:15] <Kano> budget-av/saa7146_vv driver
[12:35] <smb_tp> Keybuk, back
[13:19] <Kano> hi rtg , i did some testing, on a system with 04:07.0 Multimedia controller [0480]: Philips Semiconductors SAA7146 [1131:7146] (rev 01)
[13:19] <Kano> 2.6.30 does never boot
[13:20] <Kano> when i remove it it does
[13:20] <Keybuk> smb_tp: bumped a bug over to you
[13:20] <Keybuk> backporting the libusual disabling to jaunty for -updates along with your other upload
[13:20] <Kano> it is using normally budget-av driver
[13:21] <rtg> Kano: find the maintainer and send him an email
[13:22] <Kano> i think at this point there is not even the driver loaded when it stops
[13:22] <Kano> the last thing it does is to init alsa hda drivers
[13:26] <smb_tp> Keybuk, Can you drop me the number here? I seem to miss that
[13:27] <Keybuk> smb_tp: bug #364538
[13:27] <smb_tp> Keybuk, Thanks
[13:28] <smb_tp> Keybuk, Poor guy named smb (which is unfortunately not me) ;-)
[13:31] <Keybuk> ahh
[13:31] <Keybuk> lol
[14:20] <apw> smb_tp, if i read this bug right then we are reporting the disk blank when there is no disk at all, that i believe should have failed at the TOC lookup which was done as a raw command
[14:22] <smb_tp> Is that going down to the scsi layer? I see a few drive status and drive id calls from the cdrom driver but have not enabled debugging for below
[14:23] <smb_tp> From there it prints "No DVD+RW" / Unlocking door!
[14:24] <smb_tp> Ok, I just see that is normal exit path
[14:30] <smb_tp> apw, cdrom_id seems to do two ioctls CDROM_DRIVE_STATUS and CDROM_GET_CAPABILITY (which it prints out). Hm, did I miss a switch. That is only generic capability info
[14:32] <apw> smb_tp, yeah but the blank bit coms from a raw status request
[14:32] <apw> via sg
[14:33] <apw> and from the output from their commands it appears to have given its capabilities ok
[14:33] <apw> ie. said it does CD and DVD and the like
[14:33] <smb_tp> Nope, with a disk in it enters CDROM_MEDIA_CHANGED and displays more info. So this drive is ok. Seems you are right
[14:33] <apw> ie. its run the profiles command successfully
[14:33] <apw> which again is a raw command
[14:34] <apw> then it asks for a TOC and then the media info
[14:34] <smb_tp> So the drive itself returns unititialized data
[14:34] <apw> both raw commands, i would have expected the former to have said something implying there is no disk
[14:35] <apw> smb_tp, trying to decode the commands it is using now, well trying to find the spec to see what they contain as answers to see if we are missing anything that tells us its empty
[14:36] <smb_tp> apw, On know I had it somewhere. looking at t10.org
[14:37] <apw> http://ldkelley.com/SCSI2/SCSI2/SCSI2-14.html#14.2.11 is where i am reading
[14:37] <cking> that's useful to know
[14:38]  * cking notes apw is now the scsi cd-rom expert
[14:39]  * apw notes that cking is the expert in all things h/w
[14:39] <apw> and reassigns the bug to him
[14:39] <cking> heh
[14:40] <cking> ..I'm just a bit twiddler, not a SCSI guru
[14:41] <apw> if i am reading the cdrom_id code right its saying "what can you do" and getting the CD/DVD bits back, then saying "give me the TOC from the disk" and its getting something valid enough not to explode, then moves on to saying "tell me about it" and thats going wrong as there really iusn't anything in the drive
[14:41] <apw> which implies to me the TOC result is unexpected in content
[14:41] <smb_tp> Is there any inquiry command to test the drive status ?
[14:42] <cking> ..is this H/W related? What about checking the ISO image in kvm or similar
[14:44] <apw> smb_tp, yeah it did that before hand
[14:44] <smb_tp> So it seems on the drive I am looking at, the exit works. If I eject the disc it won't go further than generic caps
[14:45] <apw> cking, yeah i believe there are a specific couple of cdroms which exhibits this behaviour
[14:45] <apw> smb_tp, yeah think thats right.  it does the toc command and gets an error
[14:45] <apw> (TOC is a raw scsi command in this case)
[14:46] <apw> smb_tp, actually if you look at the bug itself it seems pitti may have picked up something from upstream for cdrom_id to fix this
[14:46] <apw> so all this may be moot
[14:47] <smb_tp> yeah maybe. FYI, the drive here exist with current profile 0x00, assuming no media
[14:47] <apw> which makes sense
[14:47] <apw> just looking at the change from pitti
[15:01] <rtg> cking: do you have a mini-9 ?
[15:02] <cking> rtg: nope
[15:02] <cking> should I?
[15:03] <smb_tp> cking, You are supposed to have a sample of _everything_ ;-)
[15:04] <cking> ..what kind of samples are we talking about? :-)
[15:19] <ikepanhc> rtg: ok, but I have to do it tomorrow, I can not pick others desk for a mini-9 :P
[15:20] <rtg> ikepanhc: ok, I'll ask one of the bug reporters.
[15:22] <ikepanhc> I need to go home, if any test result, I will post on launchpad
[16:23] <amitk> rtg: what is the magic git ref-log incantation you use to compare commits between two branches?
[16:24] <rtg> amitk: I generally just create two log files of commits, e.g., 'git log --pretty=oneline' , then sort and diff for unique log messages
[16:25] <amitk> rtg: aah. I thought you'd gone hi-tech with git foo from apw 
[16:25] <rtg> amitk: there might be a cooler way to do it, but this one I can remember :)
[16:26] <apw> amitk, you can try and use git log --left-right A...B
[16:27] <apw> note that that is three dots
[16:27] <amitk> apw: k, thanks
[16:27]  * cking scribbles down more git magic
[16:28] <apw> the < and > at the front tell you which branch they are on
[18:25] <bryce> apw: heya, I'm putting together a ppa for debugging X freeze issues
[18:25] <bryce> apw: https://edge.launchpad.net/~ubuntu-x-swat/+archive/x-freeze-test
[18:26] <apw> bryce, i saw something zap by me about that.  though i seem to think ihave to run some rather not what we are shipping kernel to get the features, right?
[18:26] <bryce> apw: I think it would simplify things a little for users if the 2.6.30-rc2 kernel was included in that ppa
[18:26] <bryce> apw: that's right
[18:26] <apw> yeah it would help i recon
[18:26] <bryce> apw: currently I'm linking to a kernel ppa for people to manually snag the debs for that kernel
[18:27] <bryce> is there a place I could grab the .dsc, etc. for 2.6.30-rc2 so I could copy it into this ppa?
[18:27] <apw> it never gets made into a source package per-see when making those, as currently ppas don't allow us to have a series of versions
[18:28] <apw> its likely they are deficient in some way for uploading
[18:28] <apw> i can cirtainly try and make one tho for that specific package
[18:29] <bryce> that would be very helpful, thanks!
[18:29] <bryce> yeah I know ppa's are braindead for doing series... I have quite a love/hate relationship with them
[18:30] <bryce> apw: fwiw, with X, I have a similar situation where I need multiple builds of -intel for different bugs, containing different patches; I ended up just creating multiple ppa's, one for each bug
[18:45] <bryce> apw: have you guys thought about mounting debugfs /sys/kernel/debug by default?  (Or is that more of a platform team decision?)
[18:46] <bryce> seems like there's lots of useful info there, that would be nice to make available for testers and bug reporters, at least during the development period
[18:57] <rtg> pgraner: kernel uploaded for LP 364678 "Globetrotter 3g modem not seen"