[00:22] infinity: eh, I forget to write tracker bug number... === skaet is now known as skaet__ [01:13] ikepanhc: did you run verify-release-ready ? [01:13] bjf: eh, no. how to use it? [01:14] ikepanhc: it is in the wiki as one of the steps .. one sec. [01:15] https://wiki.ubuntu.com/KernelTeam/StableHandbook/StablePatchApplication [01:15] ikepanhc: ^ [01:15] bjf: thanks.... [01:15] ikepanhc: step #11 [01:16] bjf: ACK... I shall print this page [04:59] hey folks. I'm trying to write a compile a kernel guide. Am I on the right track? ( I still have to figure out the configuration part) http://chogydan.blogspot.com/2012/08/pre-draft-build-instructions.html [05:00] any feedback will be much appreciated. I would really like to write the guide so folks can do it in the proper Ubuntu setup way, whatever [05:18] ChogyDan: https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel [05:20] bjf: I keep getting that link [05:20] or maybe not [05:23] bjf: thanks [05:23] ChogyDan: np [05:39] test [05:48] I filed a bug against the ubuntu-kernel 3.5.x packages. [05:48] I hope I did the right thing [05:49] https://bugs.launchpad.net/ubuntu/+source/policykit-desktop-privileges/+bug/1042540 [05:49] Ubuntu bug 1042540 in policykit-desktop-privileges "Cannot install 3.5.x kernel, virtual-box-guest errors during compile" [Undecided,New] [09:58] why is there another spin of the precise -proposed kernel? [10:01] brendand: i'm currently re-spining it (again) because we need to revert a patch [10:02] brendand: bug #1031630 failed to be verified, and since it didn't came from a stable update, we'll revert it [10:02] Launchpad bug 1031630 in linux "“[drm:pch_irq_handler] *ERROR* PCH poison interrupt” shows after S3 resume when connecting with an external monitor" [High,Fix released] https://launchpad.net/bugs/1031630 [10:03] henrix - thanks! [10:03] brendand: np === yofel_ is now known as yofel [11:23] hey guys I have an issue on a HP G62. When ever the webcam is activated it kills the system. There are no flashing leds, I've tried using ssh and tailing the syslog but as soon as the webcam activates the ssh stops sending and the system is in hard lock up, there isn't a sysrq button on the laptop either :( [11:23] this however stops me installing via ubiquity as it triggers the webcam [11:23] checking the logs currently [11:30] paste.ubuntu.com/1171601/ is all I see in the syslog, I mounted the cd to log where the I started the cheese application. [11:30] 12:19 I'm assuming is the start of the restart after hard powering it down [11:35] if I can provide any other info please let me know what [11:56] * ogra_ looks around for someone with a bit of kernel packaging knowledge ... seems the omap4 packaging needs some slight changes to fix bug 1041607 [11:56] Launchpad bug 1041607 in linux-meta-ti-omap4 "cannot install linux-ti-omap4 3.5.0-209-omap4" [Undecided,New] https://launchpad.net/bugs/1041607 [12:01] ogra_, looking. I'll get back to it in a bit. [12:01] great, thanks [12:59] hi guys, Im trying to write a specific guide for compiling a kernel. Ive been following this page: https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel and there are some errors [13:01] ChogyDan, the 'Building the kernel' part should prolly use dpkg-buildpackage so that uninstalled build dependencies get caught [13:02] rtg: oO that wasn't even what I was talking about. Do you know of a guide that is actually accurate? [13:03] ChogyDan, on the surface of it, I think it looks accurate. maybe you could tell me what specific errors you're having ? [13:04] the part about configuration. chmod a+x debian/scripts/misc/* <---- that doesn't hit /debian/scripts/config-check [13:04] (Im having trouble understanding the configuration stuff in general, which is why Im focusing on it) [13:05] ChogyDan, looks like its already executable, at least if you clone it using git. [13:05] rtg: Im using apt-get [13:06] lemme check... [13:08] rtg: anyway, it is a minor error. The main issue is after I apply a patch set Im working on, Im getting config errors. I still need to work through that [13:09] rtg: thanks for confirming the guide! [13:09] I'll add the one line so that all the files are marked executable [13:12] sounds good [13:46] jjohansen: you here? It looks like CVE-2012-3400 is flapping between needed and pending in the tracker [13:46] sconklin: ** RESERVED ** This candidate has been reserved by an organization or individual that will use it when announcing a new security problem. When the candidate has been publicized, the details for this candidate will be provided. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2012-3400) [14:19] * ogra_ wonders what went wrong with rsalveti's patch to install linux/omap_drv.h [14:19] i cant find it in either 3.5.0-12.12 nor 3.5.0-13.13 [14:20] (assuming its supposed to land in the linux-libc-dev binary package) [14:23] ogra_, 'UBUNTU: [Config] installing omapdrm specific headers for external drivers' ? [14:23] rtg, yeah [14:24] that should install linux/omap_drv.h in linux-libc-dev:armhf [14:24] should be in Ubuntu-3.5.0-12.12 [14:24] but i somehow cant find where it actually ends up [14:24] right, its neither in 12.12 nor in 13.13 [14:24] ogra_, its in the headers package [14:24] which one ? [14:25] ogra_, in each flavour AFAICT [14:25] if it goes into an arch specific headers package that doesnt help, we only install linux-libc-dev:armhf from that build on panda [14:25] all other headers come from linux-ti-omap4 [14:26] well, everything thats called linux-hearders-* at least [14:27] aha [14:27] ubuntu@panda:~$ dpkg -c linux-headers-3.5.0-12_3.5.0-12.12_all.deb |grep omap_drv.h [14:27] -rw-r--r-- root/root 7442 2012-07-21 22:58 ./usr/src/linux-headers-3.5.0-12/drivers/staging/omapdrm/omap_drv.h [14:28] hmm [14:28] dkms seems to search for linux/omap_drv.h [14:28] ogra_, linux-headers-3.5.0-12_3.5.0-12.12_all.deb is an install dependency for the flavour specific headers package [14:28] and very likely under my kernel version (omap4) not under some x86 one [14:29] rtg, well, i assume the prob is that i run a totally differen version on pandas ... [14:29] and dkms will likely look in usr/src/linux-headers-$kver/ [14:29] ogra_, yeah, we may have aconsolidated headers package for omap4 [14:30] a consolidated* [14:30] well, imho that header should really go into linux-libc-dev:armhf [14:31] ogra_, because its a DKMS dependency ? I'm not sure that makes sense. its not used to build glibs or anything else is it ? [14:31] glibc* [14:31] no, its only used to build the video driver [14:31] brendand: just fyi, the precise kernel respin has been cancelled. we've decided to keep the patch i was reverting. [14:32] ogra_, in which case you just need to adjust the include path [14:32] but you know that... [14:32] henrix - hmm, so we should still test 48? [14:33] brendand: yep [14:33] henrix - ok. bugs are being updated to reflect that i guess? [14:34] brendand: yes, still working on that. i can ping you back when that's all done [14:40] brendand: ok, everything should be updated now. let me know if you find any inconsistency. [15:03] Hey guys anyone about who can help with this machine killing webcam I have? I had the same issue in precise but it got fixed, but currently I can't complete the install process which will become more of an issue if the alternate cd goes away [15:11] davmor2, do you know how it got fixed in precise ? [15:11] there is a laptop going though the certificaiton process that has an unusual bug that after an s3 cycle, once you try to shut down, it will only reboot under precise. i tried a few kernels from the mainline build page and the earliest kernel that doesn't have this bug is 3.4.1-quantal. [15:11] rtg: no idea it was broken pre beta when I did a fresh install for beta it worked [15:11] i tried bisecting between 3.4 and 3.4.1 using the precise configuration but NONE of the builds worked. [15:12] ayan, cking is back tomorrow. he'll likely have some suggestions. [15:12] finally, i tried manually budiling with the quantal kernel config and that build didn't have the bug. [15:13] rtg: the annoying thing for me is that I ssh'd into the box prior to running cheese did tail -f /var/log/syslog and nothing the machine is killed so bad that not even ssh works :( [15:14] davmor2, can you collect apport info on it ? you might have a crash in dmesg if its that messed up [15:14] rtg: i compared the two kernel configs and nothing seemed obvious. [15:15] ayan, you tried to build 3.4, or you tried installing 3.4 ? [15:18] rtg: i haven't built 3.4 but i tried one of the precompiled packages bulit for precise and it has the bug. [15:18] rtg: i might be interesting to build a 3.4 with the quantal .config. [15:18] ayan, so the precompiled 3.41. _did_ work, right ? [15:18] 3.4.1 * [15:19] rtg: but i think i've verified that the configuration is the culprit by building 3.4.1 with both. [15:19] rtg: yes, it does work. [15:19] ayan, so the precompiled 3.4.1 likely started with the Precise config. apw - is that right ? [15:20] rtg: will do [15:20] rtg: no, it started with the quantal configh iirc. [15:20] rtg, it started with whatever the -series part of the directory name says it was [15:20] but it would have been the config at the build time [15:20] http://kernel.ubuntu.com/~kernel-ppa/mainline/ [15:21] rtg: v3.4 goes up to -precise and -quantal starts at v3.4.1. [15:21] ayan, how unfortunate :) [15:21] thats helpful of me to change in the middle like that :) [15:22] rtg, perhaps i should kick off a v3.4-quantal to allow comparison [15:22] rtg: yes. :^) this has kept me up for about two days now. [15:22] ayan, I see a 3.4 Precise as well as a 3.4 Quantal. do they both behave the same ? [15:22] wow. somehow i missed the 3.4-quantal. [15:22] apw, there is already a 3.4 Quantal [15:22] i'll give it a whirl now. [15:23] now i am confused, i thought ayan said that v3.4 was precise [15:23] apw, thats the one he tried, but I think he missed seeing the 3.4 Quantal [15:24] ya -- there is a v3.4 quantal in that directory. [15:24] oh how nice of me to make both [15:24] apw, ayan, so if the Quantal config works then we know its not code, rather its a config difference [15:24] rtg sounds resaonable indeed [15:32] rtg: bug #1042809 [15:32] Launchpad bug 1042809 in linux "machine locking webcam" [Undecided,New] https://launchpad.net/bugs/1042809 [15:32] ogra_, wrt bug #1041607, is there any reason that postinst should just always call 'update-initramfs -c -k $version' as opposed to 'update-initramfs -u -k $version' ? [15:32] Launchpad bug 1041607 in linux-ti-omap4 "linux-ti-omap4 generates the initial initrd to late for its own postinst scripts" [High,New] https://launchpad.net/bugs/1041607 [15:37] rtg, apw: 3.4-quantal does not have the bug. [15:38] ayan, well, bisecting between configs is gonna be a real pain in the ass [15:38] ayan, so 3.4-precise has it, and 3.4-quantal does not -- correct ? [15:38] (thought i thought i already confirmed this by testing 3.4.1 with quantal and precise configs. :^P ) [15:38] s/thought/though/ [15:38] apw: correct. [15:38] rtg, i would expect the differences that early to be small [15:39] ayan, so if you extract both of those .configs from /boot and compare them with diff [15:39] apw, true, damn near no-existent. is the tools set different ? [15:39] how bad is it [15:39] apw: i'll take a look now. [15:39] rtg, we may have compiler changes too [15:39] rtg, no both are built in precise chroots i think [15:39] rtg: if you get a chance let me know if there is anything of any use there if not if there is anything I can do to get more info for you [15:40] i wrote a tool to show differences and they were pretty dramatic between a recent quantal config and precise. [15:40] rtg, -u doesnt create the initrd on some arches if there isnt a pre-existing one [15:40] davmor2, actually, I need the crash info _after_ you'ce started the USB camera [15:41] ogra_, which is why I'm thinking just always use '-c' [15:41] rtg, in other news, beyond the fact that we need the omap_drv.h file in linuc-ti-omap4, it seems we also miss http://dev.omapzoom.org/?p=integration/kernel-ubuntu.git;a=commit;h=0428bbdcdb53255b8b42a8ed73ab6ec1d8975054 [15:41] rtg, right [15:41] rtg: no probs I'll see if I'm able to trigger it and trigger apport via terminal or not [15:41] (our omap_drv.h searches for omap_priv.h which is non-existing) [15:45] rtg: Meh nope looks to be stuck at collecting problem info and now the system is unresponsive, I'll see if this finishes or not though [15:45] ogra_, ok. isn't there a bug about the difficulties compiling against omapdrm ? [15:45] not that i know of [15:46] and if so, it would be outdated ... i just uploaded the first (supposedly) working pvr for quantal right now [15:46] rtg, apw: maybe this will be useful: http://pastebin.com/tKdi1PyK [15:46] ogra_, then you should start one so we can track this (please) [15:47] ayan, could i have both of the v3.4-{precise,quantal} configs please [15:47] rtg, will do, but i want to debuig that properly with rsalveti first [15:47] CONFIG_X86_MCE_XEON75XX ? [15:47] apw: sure. [15:47] ogra_, ok. just assign it to me once you've created the bug and explained the issues. [15:48] ok [15:48] rtg: and now it's completely dead :( [15:48] apw, why are there _any_ differences in the final .config file ? its the same Kconfig's [15:49] rtg, it is starting from a different configuration shoved into the same valid options [15:50] rtg, there is cirtainly the possibility of a different result [15:50] apw: http://zinc.canonical.com/~ayan/apw/ [15:51] d'oh. [15:51] sorry. [15:51] uploading config from 3.4-quantal and 3.4-precise. [15:54] davmor2, so this must be the built-in HP Webcam-101 that is crashing, right ? [15:54] rtg: it is yes [15:55] rtg: but it doesn't crash the app as such it's just the system stops responding to anything [15:55] davmor2, which means its crashing the kernel somewhere [15:56] rtg: indeed but I don't get all the flashing led like I would normally from a kernel crash and there is no apport data in /var/crash [15:57] davmor2, that just means something has gone far more wrong then usual. could be a DMA wedging the PCI bus or something. [16:02] apw,rtg: http://zinc.canonical.com/~ayan/apw/ [16:02] the are makred with -quantal and -precise. [16:03] davmor2, it seems like this should behave as a standard UVC USB video device. You say this works fine on Precise, right ? [16:03] sconklin: okay I will look into it [16:03] rtg: yeap and the first alpha too iirc [16:04] davmor2, the first Quantal alpha was a 3.4 based kernel IIRC [16:04] jjohansen: I prepped the meeting report for today without running the normal merge process I do, so no hurry [16:04] davmor2, I've got an HP mini. lemme have a look at it. [16:06] rtg: I can't be certain on the first alpha though. But I know alpha 2 onward I started using alternate as the desktop ubiquity would die on the webcam page [16:11] davmor2, unfortunately mine is a different device, Chicony Electronics. I'm also not finding any crash reports in Google [16:12] rtg: so Joseph just linked my bug to this one https://bugs.launchpad.net/bugs/875297 [16:12] Ubuntu bug 875297 in linux "Ubuntu installer hangs on "Choose a picture" screen. Never recovers. Installation not bootable." [Critical,Incomplete] [16:12] rtg: so it could well be the one driver [16:13] davmor2, can you disable the camera from the BIOS ? [16:13] rtg: let me reboot the box I'll let you know [16:15] rtg: Nope very few hardware settings [16:16] davmor2, drat [16:16] davmor2, so the bug jsalisbury has linked to has a different device. Furthermore its a USB 1.0 device whereas yours is USB 2.0. [16:18] davmor2, in fact, that bug originated with 11.10, so I don't think its the same problem. [16:18] jsalisbury, ^^ [16:18] rtg, ack, I'll unlink it as a dupe [16:18] rtg: so the uvcvideo module is the one I have however [16:19] davmor2, yeah, its kind of a generic video handler. [16:19] USB has a standard for video class devices IIRC [16:20] rtg: any of the devs close to the midlands in the UK? I'm happy to pay a visit if it gets it fixed :) [16:21] davmor2, apw lives in Wimbledon [16:23] rtg: I might be able to make a trip to the new office but I was hoping for more of a trip 30 minutes up the road rather than 2-3 hours away :) [16:23] davmor2, UDS ? [16:24] Nope [16:24] ** [16:24] ** Ubuntu Kernel Team Meeting - Today @ 17:00 UTC - #ubuntu-meeting [16:24] ** [16:24] davmor2, well, maybe you can bribe apw to meet you somewhere. [16:24] rtg: I can take a trip to the office at some point in the near future, I'll sort something out apw [16:25] davmor2, hmmm is this an installation issue or something which appears when installed [16:25] apw: both [16:26] davmor2, and whats the known good/bad kernels at this point [16:26] davmor2, you can work around it for now by using the alternate installer, then black listing the module before you reboot. [16:26] apw: because there is a webcam page in the installer it kills that, but also cheese kills the system on an installed system [16:27] rtg: that's what I did but the devs are looking to lose alternate soon so that won't be an option for much longer [16:27] davmor2, does rtgs work around make it work, or just get rid of your webcam [16:27] davmor2, thats true [16:27] rtg, what module breaks [16:28] uvcvideo (I think) [16:28] apw, his machine wedges so bad that he can't get a stack trace [16:28] apw: alternate works, I can try and disable the webcam module as described in 875297 and see if I can complete the install via live if you want [16:29] uvcvideo.__disable__=true on the kernel command line under F6 in the installer? davmor2 ? [16:29] that sort of thing? [16:30] https://bugs.launchpad.net/bugs/875297 this was remove the module and black list it [16:30] Ubuntu bug 875297 in linux "Ubuntu installer hangs on "Choose a picture" screen. Never recovers. Installation not bootable." [Critical,Incomplete] [16:30] apw: it worked on precise from beta1 on iirc [16:31] apw: and iirc I was able to install alpha 1 I think it died after alpha 1 [16:35] apw: if it help I can start a live install and add openssh-server and probably give you access but to be honest once the webcam page appears for me the system is dead and I couldn't get any info locally via ssh but you might know a lot more cli-fu [17:01] ogasawara, you are about 10% there [17:01] ogasawara, your mumble session went completely to hell [17:01] damn [17:01] I'm just going to write up what we just talked about and circulate it === jsalisbury changed the topic of #ubuntu-kernel to: Home: https://wiki.ubuntu.com/Kernel/ || Ubuntu Kernel Team Meeting - Tues September 4th, 2012 - 17:00 UTC || If you have a question just ask, and do wait around for an answer! === ogra_ is now known as ogra [17:27] * rtg -> lunch [17:32] ;b msm/b msm [17:34] * henrix -> EOD === ogra is now known as ogra_ [20:02] hi good people, is the Quantal kernel in todays builds, or do we have to wait for the cron job to do a daily respin? [20:08] phillw, https://launchpad.net/~ubuntu-x-swat/+archive/q-lts-backport linux-meta-lts-quantal [20:14] herton, were you gonna review the last patch for Lucid CVE-2012-3412 ? [20:14] rtg: ** RESERVED ** This candidate has been reserved by an organization or individual that will use it when announcing a new security problem. When the candidate has been publicized, the details for this candidate will be provided. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2012-3412) [20:16] rtg, it was scarily almost equal to the patch I posted, so I think it's ok, you applied it already to lucid didn't you? [20:17] *to the diff I posted [20:17] herton, no, I have not pushed it to the main repo yet, only my public repo. [20:18] herton, the only change from your patch was to move netif_skb_features() into a header file as a static inline. [20:20] rtg, ah now I see. Yes that's ok, I think we can push it [20:20] herton, ack [20:48] * rtg -> EOD [21:01] ogasawara: was bug 1042903 tested on a 3.5.3 based kernel? looks like it might be fixed by 9830605d4c0 [21:01] Launchpad bug 1042903 in linux "mgag200 driver hangs on HP ProLiant Gen 8 platform" [Medium,Confirmed] https://launchpad.net/bugs/1042903 [21:02] jwi: yes, that test kernel I posted was built on a v3.5.3 based kernel. [21:02] * ogasawara looks at that sha1 [21:15] ogasawara: I meant if the bug was reproduced on 3.5.3 - "Quantal OS" isn't very specific. [21:17] jwi: thanks, I've asked if we could get a quick re-test with the 3.5.0-13.13 kernel that was just uploaded and rebased on v3.5.3. the test kernel I built was based on 3.5.0-13.13 but disabled that experimental driver and previous tests were on a 3.5.0-12.12 or older kernel which was only rebased to v3.5.2 or older. [21:22] ogasawara: thanks - not that disabling experimental drivers for server hw is bad per se, but if there's an easier way to fix it ...;) [21:23] jwi: indeed, thanks for the heads up === BenC__ is now known as BenC