[01:52] <Mase> hi guys, i noticed that when you install a new ubuntu kernel it automatically runs update-initramfs however when i build my own packages with make-kpkg it does not do this , even with --initrd
[01:52] <Mase> how can i get my packages to run update-initramfs on installation?
[02:00] <JanC> I think it's implemented as an dpkg trigger ?
[02:01] <JanC> or maybe in a post-install script
[02:03] <Womble2> JanC: I believe it's done by the postinst (it is in Debian).
[02:04] <Womble2> Mase: man kernel-img.conf
[02:04] <JanC> hm, it really should be dpkg trigger IMHO (as other packages need update-initramfs too), but it's possible  ;)
[02:04] <Mase> ok will check it out.
[02:10] <Mase> Womble2, JanC i don't have my build box accessable right now, would it be possible to give you a buzz if i run into issues with either approach ?
[02:11] <JanC> I'm not an expert on this, but you can always ask questions in the channel here
[02:12] <JanC> if you hang around, maybe somebody can give a more definite answer too
[02:14] <Mase> i've been trying to find the answer for about a week now, google, kernel mailing lists and asking in this channel a couple of times a day. You guys/girls have been the only ones who have been able to point me in the right direction
[02:15] <JanC> Mase: maybe have a look at the source of the official Ubuntu kernel packages
[02:15] <Womble2> Mase: make-kpkg is being deprecated by the Debian kernel team and probably Ubuntu too. You may find that 'make deb-pkg' works better for you.
[02:16] <Mase> ahh ok. i didn't know that
[02:16] <Mase> i hadn't found any docs which mentioned deb-pkg
[02:16] <Mase> but i will look at it.
[02:16] <Womble2> It's part of the upstream build system
[02:16] <JanC> I think make-kpkg isn't even included in karmic anymore 
[02:17] <Mase> i see. I was / am building a Xen 2.6.31 domU kernel for hardy
[02:17] <Mase> it seems that the newer kernels don't always play well with the older build tools
[02:17] <JanC> Mase: there is some stuff about building kernels on the wiki IIRC
[02:17] <Mase> make-kpkg in hardy seems to treat Xen as as an arch. I got the kernel and modules built and working
[02:18] <Mase> just didn't get the initrd to be make automatically
[02:18] <Mase> but it seems i may have been heading down the wrong path
[02:31] <Darxus> bumping the abi from 2.6.31-11.38 to 2.6.31-12.38 didn't stop abi from complaining.  I tried "touch debian.master/rules.d/control.stub.in" as (sort of) mentioned in https://wiki.kubuntu.org/KernelTeam/KernelMaintenance  - will that satisfy abi?
[02:32] <Darxus> Is there any more accurate documentation on building a kernel package for the archives than that page, for example, mentioning debian.master?
[02:46] <Darxus> I'm asking because I'm looking into https://bugs.launchpad.net/ubuntu/+source/linux/+bug/424927
[02:46] <ubot3> Malone bug 424927 in linux "[needs-packaging] include Brain fuck Scheduler" [Wishlist,Confirmed] 
[04:36] <tj83_> apw, still working?
[04:38] <tj83_> apw if you recall our earlier discussion today about the RTL8187B wireless bug I have located it in LP. https://bugs.launchpad.net/ubuntu/+source/linux/+bug/254438  I have commented on the subject, and also sent mail to that potential contact for working drivers. I will follow up further with if/when I receive a reply.
[04:38] <ubot3> Malone bug 254438 in linux "realtek RTL8187B wireless card does not work on toshiba satellite A210-15J" [Medium,Triaged] 
[05:28] <Darxus> II: Checking modules for generic...previous or current modules file missing!
[11:21]  * apw and smb are doing a quick review of the milestoned kernel bugs for karmic: https://bugs.edge.launchpad.net/ubuntu/karmic/+source/linux
[11:22] <apw> bug #290153
[11:22] <ubot3> Malone bug 290153 in linux "Fails to find boot device in Intel D945Gnt" [High,Confirmed] https://launchpad.net/bugs/290153
[11:23] <apw> this has been improving over the 2.6.31 cycle, with errors occuring, but now booting.  requested testing with karmic final kernels
[11:25] <apw> bug #406466
[11:25] <ubot3> Malone bug 406466 in linux "2.6.31 - Can't see files in CIFS-mounted directories" [High,Confirmed] https://launchpad.net/bugs/406466
[11:25] <apw> some success reported, requesting confirmation and debug as indicated by chuck
[11:26] <apw> bug #407793
[11:26] <ubot3> Malone bug 407793 in xserver-xorg-video-intel "[i865g][Karmic Alpha 3] X corruption and freeze when clicking "Other" on GDM login screen" [High,Triaged] https://launchpad.net/bugs/407793
[11:26] <apw> suggested fixed by an upstream patch which is now in the uploaded kernels, requested testing there
[11:27] <apw> bug #435917
[11:27] <ubot3> Malone bug 435917 in linux "BUG: scheduling while atomic: swapper - using ppp [kernel 2.6.31-10.35-generic]" [High,Triaged] https://launchpad.net/bugs/435917
[11:27] <apw> ppp issue, with so many tty layer fixed arriving in 2.6.31.2 testing on latest karmic requested
[11:28] <apw> bug #100110
[11:28] <ubot3> Malone bug 100110 in acpi "18 seconds ACPI delay while booting due to DSDT" [Undecided,New] https://launchpad.net/bugs/100110
[11:28] <apw> broken DSDT on specific laptop, override nolonger recommended/supported, will produce a test kernel which attempts to detect and sort out these errant sleeps from the kernel side
[11:29] <apw> bug#392692
[11:29] <apw> bug #392692
[11:29] <ubot3> Malone bug 392692 in linux "hibernate, suspend, monitor switch keys don't work in Panasonic CF-Y7 laptop" [Medium,Triaged] https://launchpad.net/bugs/392692
[11:30] <apw> panasonic laptop specific issue, hotkeys not working.   need to confirm the driver is even being loaded, and which specific laptop this is from the ACPI info, requested
[11:30] <apw> bug #395358
[11:30] <ubot3> Malone bug 395358 in linux "thinkpad fn+f5, Asus fn+f2: regression, rfkill toggling in the kernel instead of userspace" [Medium,Triaged] https://launchpad.net/bugs/395358
[11:31] <apw> looks like it may be related to the next bug
[11:31] <apw> bug #397698
[11:31] <ubot3> Malone bug 397698 in linux "isAnyWirelessPoweredOn in state-funcs always returns 1 in karmic" [Medium,In progress] https://launchpad.net/bugs/397698
[11:31] <apw> seems that there is a proposed patch for this, which is claimed migth fix the preceeding bug also.  need to get a test version of this acpi-support package built for testing
[11:32] <apw> bug #404064
[11:32] <ubot3> Malone bug 404064 in linux "KMS error message while intializing modesetting (during boot and resume) -  render error detected, EIR: 0x00000010 [i915]" [Medium,Confirmed] https://launchpad.net/bugs/404064
[11:33] <apw> seems to be a benign message, functionality appears unaffected.  likely we can ignore this error
[11:36] <apw> bug #430694
[11:36] <ubot3> Malone bug 430694 in linux "agpgart-intel not loaded before drm sometimes, causes KMS to fail" [Medium,Confirmed] https://launchpad.net/bugs/430694
[11:36] <apw> new module dependancy is needed, potential fix uploaded
[11:37] <apw> bug #439648
[11:37] <ubot3> Malone bug 439648 in linux "usb_id errors: "unable to access..."" [Medium,Confirmed] https://launchpad.net/bugs/439648
[11:38] <apw> again these appear to be benign, requested confirmation of funcionality
[11:38] <apw> bug #440470
[11:38] <ubot3> Malone bug 440470 in linux "[ubuntu-boot-experience] nvidia boot messages" [Low,Triaged] https://launchpad.net/bugs/440470
[11:38] <apw> these seem to be benign messages, suggested alternative options
[11:55] <apw> Keybuk, bugs about boot messages which are rc script related ... is there a place to assign those to
[11:56] <apw> in this case we only have approximate contents so it is hard to tie them to a specific rc script
[11:56] <apw> sysv-rcinit perhaps ?
[12:09] <tj83> apw, you've been working hard this am.
[12:11] <tj83> bug #254438
[12:11] <ubot3> Malone bug 254438 in linux "realtek RTL8187B wireless card does not work on toshiba satellite A210-15J" [Medium,Triaged] https://launchpad.net/bugs/254438
[12:12] <tj83> very old bug, have nothing as of yet to work with. trying to get apw the pieces.
[12:13] <tj83> apw can we somehow modify the title of the bug? it does not really apply to any one model or brand of computer or even computers specifically as there are usb adapters galore with the chip also. 
[12:55] <smb> tj83, Reading through that bug, I wonder whether it possibly makes sense to close this one sort of fixed (as the card seems at least detected) and open a new one to the fact that performance/stability is poor... 
[13:08] <tj83_> apw, smb rtg  is it possible to request a bug title to be modified? 
[13:09] <apw> bug titles are mutable
[13:09] <smb> tj83_, You can modify a bug title but it won't help cleaning up comments in the report that now might be irrelevant
[13:10] <tj83_> well, i am not the creator of this bug, but i think the title should be modified for 254438 
[13:10] <rtg> tj83_, yes, but I always have to fumble around before I can figure out how to do it. 
[13:11] <tj83_> particularly to remove the specific laptop model and to include that it applies to 8.10, 9.04, and 9.10
[13:11] <apw> bug #254438
[13:11] <ubot3> apw: Error: Could not parse data returned by Malone: The read operation timed out
[13:11]  * apw slaps ubot3 
[13:11] <ogra> :P
[13:11] <apw> bug #254438
[13:11] <tj83_> he he
[13:11] <tj83_> "
[13:11] <tj83_> realtek RTL8187B wireless card does not work on toshiba satellite A210-15J"
[13:12] <tj83_> the title is just horrible. it does infact work, but performance is not acceptable and it covers thousands of laptops and usb adapters across all ubuntu versions after 8.04
[13:12] <apw> bug #254438
[13:12] <ubot3> Malone bug 254438 in linux "realtek RTL8187B wireless card does not work on toshiba satellite A210-15J" [Medium,Triaged] https://launchpad.net/bugs/254438
[13:13] <apw> bug #254438
[13:13]  * apw gets out a larger hammer and smacks ubot3 on the toe
[13:13] <smb> bug #254438
[13:13] <ubot3> Malone bug 254438 in linux "realtek RTL8187B wireless card does not work on toshiba satellite A210-15J" [Medium,Triaged] https://launchpad.net/bugs/254438
[13:14] <apw> cheek, now you cache out of date data do you ubot3 ... you trying to mess with m y head?
[13:14] <smb> Hm i changed that
[13:15] <tj83_> better, TY 
[13:15] <smb> Likely the description should be modified as well
[13:15] <smb> To get a better fitting description there as well
[13:16] <tj83_> smb, yea its all messed up, How can i change?
[13:17] <smb> tj83_, Are you able to change the description? IOW do you see some yellow circle to the right of bug description?
[13:17] <apw> with the pencil inside it, yep thats meant to be a pencil
[13:17] <smb> Somewhere below "does thisbug affect you"
[13:17] <tj83_> smb perfect thanks
[13:18] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bugs?field.searchtext=&orderby=datecreated&field.status%3Alist=NEW&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPLETE_WITHOUT_RESPONSE&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&assignee_option=any&field.assignee=&field.bug_reporter=&field.bug_supervisor=&field.bug_commenter=&field.subscriber=&field.status_upstream-empty-mar
[13:18] <apw> ker=1&field.omit_dupes.used=&field.omit_dupes=on&field.has_patch.used=&field.has_cve.used=&field.tag=regression-potential&field.tags_combinator=ANY&search=Search
[13:18] <apw> yay for launchpad
[13:21]  * apw and smb are reviewing the primary linux regression-proposed list
[13:22] <smb> bug 271258
[13:22] <ubot3> Malone bug 271258 in linux "Acer Orbicam gspca's module fails" [High,Triaged] https://launchpad.net/bugs/271258
[13:22]  * tj83_ is 26 years young today. woot
[13:22] <apw> wipper snapper :)
[13:23] <rtg> apw, thats my line
[13:24] <apw> http://kernel.ubuntu.com/~kernel-ppa/info/kernel-version-map.html
[13:24]  * tj83_ doesnt get to really see the family on the b-day due to work and school.... gee
[13:28] <smb> So basically this also looks like something we should ask to get re-tested with 12-39++
[13:30] <apw> as they are seeing delays on boot, lets get a bootchard
[13:31] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/290153
[13:31] <ubot3> smb: Error: Could not parse data returned by Malone: The read operation timed out
[13:31] <smb> That one we already had on the other review
[13:31] <smb> Fails to find boot device in Intel D945Gnt
[13:31] <apw> bug #396286
[13:31] <ubot3> Malone bug 396286 in linux "2.6.31-generic: kernel panic near the end of initramfs" [High,In progress] https://launchpad.net/bugs/396286
[13:32] <smb> Also from previous review
[13:32] <smb> We asked to try with latest kernel
[13:32] <smb> Oops srry
[13:32] <smb> yeah the other one
[13:32] <smb> Need him to check with the latest
[13:33] <smb> patch to see whether I can get more info on the files affected
[13:33] <apw> as yes the nasty 'b4 appears in your -1 pointer'
[13:33] <smb> yep
[13:33] <apw> bug #397906
[13:33] <ubot3> Malone bug 397906 in linux "blank cd-r not detected [Karmic]" [High,Incomplete] https://launchpad.net/bugs/397906
[13:34] <soren> apw: Is anything in particular holding up bug #416325?
[13:34] <ubot3> Malone bug 416325 in linux "Please reenable CONFIG_GFS2_FS_LOCKING_DLM" [Low,In progress] https://launchpad.net/bugs/416325
[13:35] <smb> soren, only piles of stacking work interrupts. :-P
[13:36] <apw> soren, only it not being on the regessions list and focus being there, is that a regression from jaunty, it likley is and should be o the list
[13:38] <smb> So about that cd-r not detected thing. The last comment is from August, so we should get more recent feedback
[13:38] <apw> yep, asking for the latest kernel to be tested
[13:38] <apw> bug #400484
[13:38] <ubot3> Malone bug 400484 in linux "unable to show the contents of my kernel keyring" [High,In progress] https://launchpad.net/bugs/400484
[13:42] <rtg> soren, https://bugs.edge.launchpad.net/ubuntu/karmic/+source/linux/+bug/416325/comments/1
[13:42] <ubot3> Malone bug 416325 in linux "Please reenable CONFIG_GFS2_FS_LOCKING_DLM" [Low,In progress] 
[13:42] <smb> Hm, seems we are not sure whether this is really a bug or only a behaviour change which was intended
[13:42] <smb> We ask for clarification in the bug
[13:55] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/402982
[13:55] <ubot3> Malone bug 402982 in linux "Kernel mode switching causes frequent screen flickering with intel video card (not present with the same packages and kernel 2.6.30)" [High,In progress] 
[13:56] <smb> apw Seems your last ppa had good results
[13:56] <smb> But as a lot of radeon patches went into the kernel we need to check the latest one to see
[13:57] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/414560
[13:57] <ubot3> Malone bug 414560 in linux "ath9k disassociates/reassociates a lot" [Unknown,Confirmed] 
[13:58] <soren> rtg: Awesome, thank you.
[14:00] <smb> As some success was claimed with some backports modules, need to check the latest kernel and lbm 
[14:01] <apw> bug #430011
[14:01] <ubot3> Malone bug 430011 in linux "Oops randomly with Huawei E220 3G dongle (regression)" [High,Triaged] https://launchpad.net/bugs/430011
[14:03] <smb> usb-serial has gone through huge changes in stable.
[14:03] <smb> So retest is essential
[14:05] <apw> bug #430809
[14:05] <ubot3> Malone bug 430809 in linux "[Dell Latitude D430, iwl3945] Wireless can't be activated after disabling kill switch" [High,Triaged] https://launchpad.net/bugs/430809
[14:05] <apw> this bug is in active development, a patch appears to be being testing and working for testers
[14:07] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/432959
[14:07] <ubot3> Malone bug 432959 in linux "empty barriers not supported by kvm virtio_blk - "end_request: I/O error, dev vda, sector 0" log spam" [High,Triaged] 
[14:15] <Username> hello can you tell me witch kernel ubuntu 9.04 on the downloadable cd has?
[14:22] <apw> thats likely wahts in the -release pocket for jaunty 2.6.28-11.42
[14:22] <apw> seems that these messages are benign
[14:23] <apw> upsteeam talks about removing them, no fix as yet
[14:24] <apw> bug #433904
[14:24] <ubot3> Malone bug 433904 in linux "[karmic] regression: WinFast DTV 2000 H support has disappeared" [High,Triaged] https://launchpad.net/bugs/433904
[14:41] <apw> seems that this used to be fixed via a sauce patch, and is now fixed in mainline 2.6.32-rc1.  need to investigate whether this can be ported to karmic
[14:48] <apw> ok ... mediums ...
[14:53] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/347458
[14:53] <ubot3> Malone bug 347458 in linux "lpia kernel used on alternate CD doesn't load cd-rom drivers" [Medium,Confirmed] 
[14:58] <apw> latest report of this was on karmic alpha-6.  we have consolidates kernel configs i386-generic and lpia so we expect them to behave the same.  needs retesting on latest 
[15:01] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/394354
[15:01] <ubot3> Malone bug 394354 in linux "Xorg update failed with 2.6.31-1" [Medium,Incomplete] 
[15:01] <smb> The original problem might likely be fixed already. The last statement about that is very old
[15:02] <apw> re-prodded for testing results but likely closable
[15:02] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/395058
[15:02] <ubot3> Malone bug 395058 in linux "[karmic] Ubuntu kernel 2.6.31-1-generic fails to boot" [Medium,Triaged] 
[15:04] <smb> Doh! Yeah, something strange in the early alphas. Unfortunately it seemed to not even go beyond grub
[15:05] <apw> thats last tested on -5 so we need a retest before we even try and diagnose that one
[15:06] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/396563
[15:06] <ubot3> Malone bug 396563 in linux "Headphone jack regression in 2.6.31-1" [Medium,Triaged] 
[15:09] <smb> Upstream fix seems to have to gone into 2.6.31-rc9, so should be fixed. Will need to ask for feedback
[15:11] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/399877
[15:11] <ubot3> Malone bug 399877 in linux "Unable to mount 80GB iPod with nautilus" [Medium,Incomplete] 
[15:19] <apw> seems that the original report is that a bad sdX1 partition cannot be mounted, but this really was a bad partition and the kernel was not at fault
[15:21] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/400682
[15:21] <ubot3> Malone bug 400682 in linux "[Karmic stac9227 regression] No sound after upgrade from Jaunty to Karmic" [Medium,In progress] 
[15:22] <smb> It seems there has been some work going on and then nothing more since September.
[15:22] <smb> Maybe time for a little reminder...
[15:24] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/406312
[15:24] <ubot3> Malone bug 406312 in udev "loading/unloading of option driver does not work - breaks use of common usb modems" [Undecided,Invalid] 
[15:25] <smb> As upstream basically rewrote usb-serial, we should ask for testing
[15:26] <dschulz> hi all
[15:26] <smb> Oh, there has been
[15:26] <dschulz> is there any known bug related to sound and switching between ttys?
[15:27] <apw> asked for testng on the usb issue as he has the updates in 31.2.  suspect userspace issue here
[15:27] <smb> dschulz, I have not heard of any, but that is not conclusive
[15:28] <smb> apw, yeah, the crash seems gone
[15:28] <smb> still might be a reference silently hold, so the module is not really removed...
[15:29] <dschulz> look, recently noticed that, while amarok was running (or any other player) and i switched to tty01, sound is muted until i switch back to tty07
[15:30] <dschulz> but when I switch back to tty07, sound is crappy for a while
[15:30] <apw> dschulz, not tried that myself
[15:31] <apw> perhaps the apps are stalled, i can't remember how its meant to work
[15:31] <apw> smb https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/407790
[15:31] <ubot3> Malone bug 407790 in linux "[Regression] No sound with Karmic on HP HDX 9300" [Medium,Triaged] 
[15:31] <smb> I know amarok goes silence when the powersaving kicks in, so maybe the silence is intended...
[15:32] <dschulz> it used to keep running. I tried with mpg123, and sometimes it will never resume
[15:33] <dschulz> i mean, mpg123 will not continue playing after switching back to tty07
[15:33] <apw> that sounds a bit wrong, not in a position to test right now myself
[15:34] <dschulz> to me it looks like a scheduler issue
[15:34] <apw> actually i can test just audio
[15:34] <rtg> apw, do you remember how to get kmake to puke out cpp files ?
[15:35] <apw> rtg don't think i've ever done that
[15:35] <apw> oh i get you yes
[15:35] <apw> rtg you want -E output
[15:35] <rtg> hmm, I'm having interesting preprocessor problems.
[15:35] <apw> i normally compile V=1
[15:35] <apw> that prints the whole command line
[15:35] <apw> then switch that by hand
[15:36] <smb> dschulz, It rather is something special on Kubuntu. Didn't they move now from arts to pulseaudio...
[15:36] <apw> rtg did i make any sense there?
[15:36] <rtg> apw, yeah, but it seems like CPP is already -E
[15:37] <apw> dschulz, i just tested on my gnome install and i could switch to VT-1 for extended perdious with music and skpye working
[15:37] <apw> oh is this something which only uses CPP ?
[15:37] <apw> what is the line if you use V=1 for it?
[15:37] <rtg> apw, it seems the MODULE_IMPORT/MODULE_EXPORT macros don't work on amd64. wtf?
[15:38] <dschulz> apw: strange
[15:38] <apw> oh ... hrm ...
[15:38] <rtg> apw, I only tested i386
[15:39] <apw> rtg, how the heck can those only work on i386
[15:39] <apw> they are soooo simple
[15:39] <smb> So about bug 407790, seems we both have no clear idea. Seems just a case of needed alsa twiddles with input/outputs
[15:39] <ubot3> Malone bug 407790 in linux "[Regression] No sound with Karmic on HP HDX 9300" [Medium,Triaged] https://launchpad.net/bugs/407790
[15:40] <rtg> apw, my thought exactly, but MODULE_IMPORT does not create an external reference.
[15:41] <apw> rtg hrm, actually that might just be the compiler being smart
[15:41] <apw> as actually you don't make a reference as such do you
[15:41] <apw> you don't _use_ the value at all
[15:41] <rtg> apw, so, if the symbol isn't assigned some value, then it won't reference?
[15:42] <apw> if the consuming file doesn't consume the value of the variable its not clear i would expect any reference to get generated
[15:42] <rtg> apw, hmm, sledge hammer tiem
[15:42] <apw> the declaration you emit is an indidation to type and existance but not a reference
[15:42] <apw> you likely need to generate like
[15:43] <apw> int use_name() {
[15:43] <apw> return name;
[15:43] <apw> }
[15:43] <apw> in your import one
[15:44] <apw> and that might need to have __FILE__ in it too to make sure its not clashing
[15:44] <rtg> apw, I'll bet there  is a more elegant way of doing this by implanting a static symbol using linker scripts, for now I'm gonna solve the issue the hard way.
[15:44] <apw> yeah i am sure there is a saner way
[15:45] <apw> why do we not just add module_request('intel-agp') to the i915 init ?
[15:45] <apw> is that not what we mean?
[15:45] <apw> rtg ^^
[15:46] <rtg> apw, turns out that doesn't load the intel-agp dependencies (I don't think)
[15:46] <apw> really, hrm that sucks
[15:46] <rtg> apw, I'll reexamine that concept
[15:46] <apw> Keybuk, if the kernel loads a module via request_module doesn't that really get done via modprobe and should i expect the deps loaded also?
[15:47] <rtg> apw, that only works if there is a root fs
[15:47] <apw> rtg the other option might be to literally join the two together in the module.deps file?
[15:47] <apw> ie. can we somehow just add the deps when we run depmod?
[15:47] <apw> outside the kernel
[15:47] <rtg> wretch!
[15:47] <apw> :)
[15:48] <apw> ok perhaps not
[15:52] <rtg> apw, this does it: #define MODULE_IMPORT(mod_name) extern int sym_link_##mod_name; int func_sym_link_##mod_name(void) {sym_link_##mod_name=1;}; EXPORT_SYMBOL(func_sym_link_##mod_name);
[15:52] <apw> yeah thats what i was aluding too ... VILE
[15:52] <apw> but if it works, ship it :)
[15:52] <rtg> apw, I'm gonna
[15:52] <pgraner> rtg: what is the scheduler bug number?
[15:52] <apw> *shudder*
[15:52] <pgraner> rtg: for deadline over cfq
[15:52] <rtg> then I'm gonna talk to Jon about a better way.
[15:52] <rtg> pgraner, hang on, lookign
[15:53] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/410984
[15:53] <ubot3> Malone bug 410984 in linux "Regression: HP hdx 9300 often hangs on boot with acpi enabled" [Medium,Triaged] 
[15:53] <apw> requesting retesting with latest kernel, and with nousb
[15:53] <rtg> pgraner, bug #381300
[15:53] <ubot3> Malone bug 381300 in linux "elevator=cfq (default) cause starvation" [High,Fix released] https://launchpad.net/bugs/381300
[15:54] <pgraner> rtg: thanks, did this effect network transfers as well?
[15:55] <rtg> pgraner, in as much as processes use block devices. nfs perhaps?
[15:55] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/412704
[15:55] <ubot3> Malone bug 412704 in linux "Random Kernel Panics on Dell Latitude 2100 on Karmic" [Medium,Confirmed] 
[15:56] <apw> this seems to be wireless related, works much better with LBM, don't think we'll fix that one
[15:59] <pgraner> rtg: what scheduler are we shipping with? FAIR_SLEEPERS?
[15:59] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/415632
[15:59] <ubot3> Malone bug 415632 in linux "apparmor not properly handling file deletion on NFS" [Medium,Triaged] 
[16:00] <apw> jjohansen, how is this one ?
[16:00] <rtg> pgraner,  FAIR_SLEEPERS is a feature of CFS (which is disabled), but appeared to have a major impact on CFQ (an I/O scheduler)
[16:00] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/421617
[16:00] <ubot3> Malone bug 421617 in network-manager "USB Dongle Not Detected Since Upgrade to Network Manager 0.8x" [Undecided,Invalid] 
[16:02] <smb> Seems fix released
[16:03] <rtg> pgraner, lemme clarify, CFS is the default process scheduler. one of the features of CFS _was_ FAIR_SLEEPERS, but I've disabled it per upstream cherry-pick.
[16:09] <apw> rtg didn't it come back in a fixed form in .2 ?
[16:10] <apw> must have imagined it
[16:10] <rtg> apw, I don't think so, but I'd better check. I don't remember seeing any sched changes.
[16:10] <apw> must have been mainline -rc2
[16:10] <rtg> the upstream changes were much more then Greg would have taken, IMHO
[16:12] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/426870
[16:12] <ubot3> Malone bug 426870 in linux "BUG: unable to handle kernel NULL pointer dereference at 00000004" [Medium,Triaged] 
[16:14] <smb> No stacktrace, and no description what he was actually doing when the crash happened or whether this happened ever again
[16:15] <smb> Lets close it for now and wait whether it comes again
[16:16] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/430694
[16:16] <ubot3> Malone bug 430694 in linux "agpgart-intel not loaded before drm sometimes, causes KMS to fail" [Medium,Confirmed] 
[16:16] <apw> rtg is solving that one right now ... 
[16:17] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/442463
[16:17] <ubot3> Malone bug 442463 in linux "Regression: frequent popping sound with light indicating that sound is supposed to be muted" [Medium,Triaged] 
[16:20]  * smb has now idea for that one
[16:24] <jjohansen> apw: still working on it
[16:24] <jjohansen> apw: I thought I had it but my fix didn't work :(
[16:24] <smb> Decided to shove that to the ubuntu-audio-team
[16:24] <smb> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/444801
[16:24] <ubot3> Malone bug 444801 in linux "prism 2.5 broke in 2.6.30.x (fixed upstream in 2.6.32)" [Unknown,Fix released] 
[16:24] <apw> deffo audio, assigned to their team
[16:24] <apw> jjohansen, thanks for the update
[16:25] <smb> apw, Seems we got a patch there to try a test kernel
[16:25]  * smb offers to do that
[16:26] <apw> am already building it
[16:26] <smb> wfm :)
[16:28]  * lamont hopes someone other than him will track down 455456
[16:29] <rtg> bug #455456
[16:30] <lamont> meh..  I should remember that 'b-u-g' token
[16:30] <rtg> apw, of what use is the bugbot if it goes deaf?
[16:30] <lamont> rtg: fails to boot winxp pro sp2
[16:30] <lamont> kvm regression somewhere between 2.6.28-11 and 2.6.28-15
[16:30] <apw> bug #455456
[16:30] <rtg> lamont, wasn't that a grub thing?
[16:30] <lamont> I _thought_ I'd filed the bug then, but apparently I fail
[16:30]  * apw slaps ubot3 
[16:30] <lamont> pretty sure winxp pro booting in a kvm machine doesn't use grub
[16:31] <rtg> lamont, ah, I was just remembering a fix that went by from cjwatson yesterday or the day before.
[16:32] <rtg> lamont, ah, thats a _Jaunty_ regression. smb should worry about it.
[16:32] <lamont> rtg: and the reason that I'll probably be running karmic with 2.6.28-11 installed.
[16:32] <lamont> I'd really like to come currnet
[16:32] <smb> rtg, got too much to worry to worry
[16:32] <cjwatson> no such bug as 455456
[16:32] <lamont> I rarely care about the vm, but when I do, rebooting and killing all my windows?  painful
[16:33] <lamont> bug 445456
[16:33] <ubot3> Malone bug 445456 in linux "kvm hangs booting windows XP Pro SP2 or later, since at least 2.6.28-15" [Undecided,New] https://launchpad.net/bugs/445456
[16:33] <lamont> stupid fingers
[16:33] <cjwatson> 2.6.28 -> jaunty, unlikely to have been anything I touched certainly
[16:34] <lamont> yeah - when I looked a the package diff, there was a HUGE diff in kvm/* introduced in a security fix
[16:34] <lamont> so we should blame -security
[16:34] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/395219
[16:34] <ubot3> Malone bug 395219 in linux "selinux kernel panic 2.6.28-13.45" [Medium,Fix released] 
[16:35] <apw> ok this one is fixed as the code is gone
[16:35] <smb> lamont, huge diff sounds a bit strange for security
[16:36] <lamont> smb: well.. huge is a relative term, and there were probably other things between -11 and -15... I didn't narrow it down much at that point - I was about 5 days into fighting with the reinstall thinking windoze had gone sideways before I remembered that the kernel had recently upgraded
[16:38] <apw> 7adfd5c71693b81e995283805b17aa4a2ee0ecd9
[16:38] <apw> smb ^^
[16:42] <smb> lamont, I try to get at that tomorrow
[16:43] <lamont> smb: ta
[16:43] <lamont> if you need an image, or have something for me to test, I'm all yours
[16:44] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/414373
[16:44] <ubot3> Malone bug 414373 in linux "WLAN Ralink 2561/RT61 "Device not ready"" [Low,Incomplete] 
[16:44] <apw> ok this one seems to be a case of two drivers are picking up the device.  asked for infor from the working kernel
[16:45] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/414795
[16:45] <ubot3> Malone bug 414795 in linux "PC beep no longer works in Karmic alpha4" [Medium,Confirmed] 
[16:45] <apw> audio, punt
[16:49] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/417842
[16:50] <ubot3> Malone bug 417842 in linux "WARNING: at /build/buildd/linux-2.6.31/kernel/power/suspend_test.c:52 suspend_test_finish+0x7c/0x80()" [Low,In progress] 
[16:50] <apw> benign message from the kernel on slow resume.  this is not a bug per-see, but we should stop an oops being generated to stop the bugs being reported
[16:51] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/444362
[16:51] <ubot3> Malone bug 444362 in linux "status/debugging text incorrectly appears on boot" [Undecided,New] 
[16:55] <apw> not kernel message, forked to the packages which seem to be making noise
[16:59] <apw> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/401591
[16:59] <ubot3> Malone bug 401591 in linux "brightness dark and not working on acer 5715" [Low,In progress] 
[16:59] <apw> pre-acpi fixes, needds retesting on the latest also
[17:15] <lamont> bug 445588
[17:16] <ubot3> lamont: Error: Could not parse data returned by Malone: HTTP Error 503: Service Unavailable
[17:16] <lamont> brillaint
[17:16] <lamont> bug 445588
[17:17] <lamont> meh.  all your builds are fail guys. :-(  you Build-Depend: cpio without declaring such
[17:26] <rtg> lamont: I just added that a couple of uploads ago
[17:27] <rtg> for Karmic, that ios
[17:27] <rtg> is*
[17:28] <lamont> ah, ok
[17:28] <lamont> I'm playing catchup from a sync on 1 oct
[17:29] <lamont> rtg: I rather suspect that cpio was cruft in the tarball for quite a while
[17:29] <rtg> lamont: I inherited that bit from pre-Hardy days AFAICT
[17:31] <lamont> rtg: in that case, feel free to mark 445588 fix-released.  sorry
[17:32] <lamont> on the bright side, I didn't file 3 bugs
[17:33] <rtg> lamont: yeah, some of the builds started failing for no reason that I could see other then cpio was missing.
[17:37] <rtg> lamont: there is no linux-ports package for Karmic. is it OK to mark that one invalid?
[17:37] <lamont> uh.... there _is_ a linux-ports package in karmic (as of oct 1), so that's probably a "pls remove this package" bug instead?
[17:38] <rtg> or rather, the ports package has been incorporated into the standard linux package
[17:38] <lamont> meh
[17:38] <rtg> lamont: ack
[17:38] <lamont> ./l/linux-ports/2.6.30-2.4/linux-ports_2.6.30-2.4_20091006-0409-i386-failed.gz <-- after that version?
[17:38] <rtg> lamont: yes
[17:38] <lamont> --> invalid
[17:39] <rtg> AFAIK its no longer used
[18:18] <tj83_> apw, rtg you available for a moment?
[18:18] <apw> sup?
[18:19] <tj83_> https://www.securepaste.com/decrypt_data.php?id=169TsI5Y11&password=rtl8187b
[18:19] <tj83_> there are your rtl8187b pieces hopefully i have not tested them as of yet. 
[18:20] <tj83_> bug #254438 
[18:20] <ubot3> Malone bug 254438 in linux "realtek RTL8187B wireless card does not work reliable" [Medium,Triaged] https://launchpad.net/bugs/254438
[18:20] <tj83_> i'm not in a wireless environment atm but as soon as i try them myself let you know the results
[18:22] <tj83_> if it can be put into karmic.... I would love you lol.
[18:22] <tj83_> if it makes it into Lucid, then i can like you a lot still
[18:25] <apw> tj83_, put that link to the source into the bug
[18:25] <tj83_> will do.
[18:33] <tj83_> apw, I just noticed that when i tried to update the description earlier it did not take, i am trying again, when i click on the green check to save it, i get a red frame and it sits without accepting the changes
[18:36] <tj83_> apw, this is what i have tried to post to it as a proceeding addition http://www.pastebin.ca/1601944
[18:41] <tj83_> apw, is it locked of some sort?
[18:43] <apw> is what locked
[18:44] <tj83_> the description i cannot modify it. It allows me to input text then when i save (green check icon) it gets a red frame and a busy animation it never accepts the change. I actually tried to change it this am then again hours later.
[18:45] <tj83_> it accepts comments but not description changes
[18:46] <tj83_> dont want anyone to pass by the bug based on the bad info in the description.
[18:59] <tj83_> got it fixed.... 
[19:14] <Darxus> I hate ABI.
[19:17] <Darxus> On every line in debian.master/rules.d/* that contained "silentoldconfig" I appended "skipabi=true skipmodule=true".  Why am I still getting ABI errors?
[19:27] <apw> Darxus, what does silentoldconfig have to do with ABI?
[19:32] <Darxus> apw: It occurrs on the lines where make is run.
[19:33] <Darxus> The first occurrance of "make" in my build log is:
[19:33] <Darxus> make ARCH=i386 EXTRAVERSION=-11-generic CONFIG_DEBUG_SECTION_MISMATCH=y SUBLEVEL=31 KBUILD_BUILD_VERSION="38" O=/home/darxus/ubuntu/linux/linux-2.6.31/debian/build/build-generic silentoldconfig prepare scripts skipabi=true skipmodule=true
[19:33] <Darxus> Why didn't that skip ABI?
[19:36] <apw> Darxus, the skipabi there is being passed to the kernel makefile, but abi checks are done in the rules file
[19:41] <Darxus> So what do I actually need to do to disable abi and modules checks?
[19:43] <apw> Darxus, set autobuild=1 in the rules.d/0*
[19:43] <apw> AUTOBUILD=1 even
[19:45] <Darxus> apw: That will skip modules checks too?
[19:46] <Darxus> Ah, yes, cool, thanks.
[20:06] <Darxus> Have you recently seen any cases where LINUX_VERSION_CODE is defined as nothing in include/linux/version.h?
[20:07] <Darxus> It was causing me problems until approximately when I traced it to that file.  Now I can't reproduce it.
[20:09] <Darxus> Any opinions of Con Kolivas' new kernel scheduler, BFS?
[20:10] <bucky> it's for an old kernel and you can't use tickless... i don't know if i have an opinion about that
[20:11] <bucky> i think Linus treated Con poorly and Alan Cox as well... that is the only opinion i have about that
[20:11] <ikonia> bucky: stop that ranting in this channel
[20:12] <Darxus> Old kernel?
[20:12] <bucky> yup
[20:12] <Darxus> The current BFS patch is for the kernel version used in Karmic.
[20:13] <Darxus> http://ck.kolivas.org/patches/bfs/2.6.31-sched-bfs-303.patch
[20:13] <Darxus> Which is the current non-rc upstream.
[20:14] <bucky> then ubuntu rewrote it...  btw BFS is a dirty acronym
[20:14] <bucky> naughty
[20:14] <Darxus> Rewrote what?
[20:15] <Darxus> "Dirty" as in it's already used by the Be File System?  That is disappointing.
[20:16] <bucky> linux-2.6.22 http://users.on.net/~ckolivas/kernel/
[20:16] <Darxus> bucky: Yeah that's old, and not what I'm talking about.
[20:16] <Darxus> Con is back.
[20:16] <Darxus> BFS 303 was released two days ago.
[20:17] <Darxus> It was first built in August.
[20:17] <bucky> Darxus, so what's your opinion of it since you know more about it than i do
[20:17] <Darxus> bucky: Curious.  
[20:17] <Darxus> I guess that's not really an opinion.
[20:18] <Darxus> I guess I'm excited.
[20:18] <Darxus> I'm attempting to package it.
[20:18] <bucky> what about apparmor?  is ubuntu going to rewrite that for every kernel update seeing as Novell/SuSE doesn't update it often?
[20:19] <Darxus> I don't know.  I think it's good to avoid stuff that's not in mainline.
[20:19] <jjohansen> bucky: it doesn't need to be rewriiten for every release, and it is the process of being upstreamed
[20:20] <Darxus> But maybe the people maintaining mainline are dumb about some things, and it's worth Ubuntu maintaining deltas on those.
[20:20] <Darxus> The BFS story is all about mainline folks being dumb about what they accept.
[20:20] <Darxus> It's sounds pretty convincing.
[20:21] <jjohansen> there are certainly reasons to maintain things outside of mainline
[20:21] <Darxus> I'm really shocked that linux_2.6.31-11.38.diff.gz is 280 thousand lines.
[20:21] <jjohansen> eg. aufs, is the union file system being used for livecds, it won't go up to mainline
[20:22] <jjohansen> because mainline is waiting for union mounts, which aren't there yet
[20:22] <Darxus> In the case of BFS, I believe the delta would be much easier to maintain if mainline would accept plugable schedulers.
[20:23] <Darxus> And the fact that they have been completely unwilling... seems... odd.
[20:24] <bucky> i think the fact that Linus says the kernel is bloated is a sign that he's getting tired and something needs to be done to streamline the development process, i don't blame him, if i had his money i'd be out laying on a beach somewhere
[20:25] <Darxus> Heh.
[20:25] <Darxus> I'm sure it must be nuts to maintain.
[20:31] <bucky> someday Linus is going to take his kernel and go home, i wonder if there is a clause in it that he can take it back like Hans Reiser had in his contract with his developers
[20:33] <apw> he can't take it home it is released under the GL2
[20:33] <apw> GPL2
[20:34] <Darxus> Hah.  I wasn't aware Reiser had that.
[20:35] <Darxus> I've waited for the kernel package build to fail for three hours so many times :(
[20:36] <Darxus> But I'm strangely optomistic that AUTOBUILD=1 in debian.master/rules.d/0* will make it work this time.
[20:36] <apw> Darxus, you could just build one of the flavours
[20:38] <Darxus> apw: Yeah, but that works.
[20:38] <Darxus> It's a full debuild that fails.
[20:39] <apw> heh
[20:39] <Darxus> linux-image-2.6.31-11-generic version 2.6.31-11.38 patched with bfs would be linux-image-2.6.31-11-generic-bfs version 2.6.31-11.38?
[20:42] <Darxus> apw: My failures have all been ABI and module check failures, because the documentation sucks.
[20:43] <Darxus> Well, except for the one early thing where the version_code wasn't getting defined.  Which I can't replicate now :(
[20:44] <apw> often we would leave the name the same and the version ewould be 2.6.31-11.38bfs1
[20:44] <tj83_> apw,  well.... i am sad to say the drivers given for the rtl8187b do not build under 2.6.31 because of some change in the network driver API. So that would be a kernel hacker's department to pick up from here. I would not know where to start to mend he driver up for the new API
[20:45] <tj83_> apw, but i am going back to my jaunty install, waiting on updates to finish up to test further for proper function.
[20:45] <Darxus> apw: Well, BFS is too new to go in the main kernel, so it would need to be a different package name.  And thanks for the version.
[20:47] <bjf> apw, the lines in d-i/modules/input-modules have a '?' at the end, I thought that meant the module could be built in or a module
[20:47] <apw> yep if its in a PPA you can just append bfsN indicaqting it is based from the original version before, with bfs
[20:48] <rtg> bjf, taht is correct unless none of the modules in that package are built. kernel-wedge considers it an error to have an empty package.
[20:48] <apw> bjf, yep thats the .... what he said
[20:48] <bjf> rtg, that's what I ran into, thanks
[20:49] <bjf> rtg, would have been nice if that were the error message
[20:49] <rtg> bjf, there is a way to add arch and flavour specific exclusions. Look at the master branch for examples.
[20:49] <rtg> bjf, yeah, kernel-wedge is kind of primitive
[20:50] <bjf> rtg, is that something we maintain? (kernel-wedge)
[20:51] <rtg> bjf, nah, I make cjwatson fix it.
[20:52] <rtg> bjf, what takes me days to figure out he can usually whip out in a few minutes. 
[21:39] <Darxus> I'm tempted to write a debian/rules, or maybe make wrapper that records timestamps of output lines, and then during the next build it uses those timestamps to calculate and print percentage complete and eta.
[21:40] <Darxus> Hmm, I could make it a wrapper for anything.
[21:43] <Darxus> Wow, only 14% of the lines in my build log are uniq.
[21:54] <cjwatson> Darxus: there's a 'ts' program in the moreutils package that does the first half of that
[22:09] <Darxus> Thanks.
[23:12] <Darxus> mv: cannot stat `/home/darxus/ubuntu/linux/linux-2.6.31/debian/linux-image-2.6.31-11-a33-generic/lib/modules/2.6.31-11-a33-generic/modules.order': No such file or directory
[23:12] <Darxus> How do I make debuild not fail on that?
[23:12] <Darxus> Another three hours.
[23:13] <Darxus> I'm using AUTOBUILD=1.
[23:14] <Darxus> Where did the "a33" come from?  I've never seen that sequence of characters anywhere in this.