**** BEGIN LOGGING AT Mon Jun 14 02:59:57 2010 Jun 14 07:20:42 cooloney: ping Jun 14 07:22:06 lag: yeah, Jun 14 07:22:21 cooloney: Good morning Jun 14 07:22:36 lag: morning, man Jun 14 07:22:44 Is the OMAP4 repo on our server yet? Jun 14 07:23:35 morning guys! Jun 14 07:23:45 Morning Amitk Jun 14 07:24:17 I heard that we have a newly-rebased tree from TI? Jun 14 07:24:28 lag: yeah, i just synced one from sebjan Jun 14 07:24:53 and he is trying to unify the panda code into his tree Jun 14 07:25:10 Great Jun 14 07:25:13 i will pull from him after that and built for you guys for testing Jun 14 07:25:24 what version is the tree based on cooloney ? Jun 14 07:25:27 What's it' called? Jun 14 07:25:39 lag: do you have other omap4 hardware instead of panda? Jun 14 07:25:50 amitk: oh it is based on TI 2.6.34 integration tree Jun 14 07:26:02 No, just Panda Jun 14 07:26:24 morning Jun 14 07:26:28 I'm guessing it would be called ubuntu/ubuntu-lucid-arm.git Jun 14 07:26:41 Gnome bytter ut F-spot med shotwell Jun 14 07:26:53 Wrong chan, sorry Jun 14 07:27:17 lag: lucid-arm.git? no, the new branch is for maverick Jun 14 07:27:43 I'm guessing it would be called ubuntu/ubuntu-maverick-arm.git Jun 14 07:27:45 ;) Jun 14 07:28:28 why is it a separate tree and not a branch in the maverick git tree? Jun 14 07:28:54 I understood that it's not ready to go into the main repo yet Jun 14 07:29:15 I was chatting with cooloney and ogra about it the other day Jun 14 07:29:50 lag: http://kernel.ubuntu.com/git?p=roc/ubuntu-maverick.git;a=shortlog;h=refs/heads/ti-ubuntu-2.6.34 Jun 14 07:30:01 that is a temp branch in my repo Jun 14 07:30:14 Okay Jun 14 07:30:18 And after? Jun 14 07:30:19 after it is ready, i will push it out for review and merge into maverick Jun 14 07:30:32 amitk ^^ Jun 14 07:30:42 ok Jun 14 09:29:52 cooloney: ping Jun 14 10:01:33 lag: could you please try the kernel from http://people.canonical.com/~roc/kernel/omap4-2.6.34-panda/ on your panda? Jun 14 10:03:00 Was it built from roc/ubuntu-maverick.git Jun 14 10:03:43 lag: yeah, it is Jun 14 10:03:51 lag: i cross built it locally Jun 14 10:04:09 By locally, do you mean on 'your' machine? Jun 14 10:04:28 I am building the same kernel currently on one of the build servers Jun 14 10:04:44 lag: yes, on my machine, since i failed to use sbuild on tyler Jun 14 10:04:58 lag: aha, cool, are you using sbuild? Jun 14 10:05:26 No idea Jun 14 10:05:50 Whatever's specified in the chroot Jun 14 10:06:17 I guess so, as I received an sbuild error this morning when the permissions were messed up Jun 14 10:08:17 lag: yeah, me2, so i switched to my local builder Jun 14 10:09:20 Well once the maverick-armel issue is sorted apw will have it available again within 2 mins Jun 14 10:10:00 lag its not that simple Jun 14 10:10:04 but i am working on it Jun 14 10:10:27 Is it not the same issue as the other build server? Jun 14 10:10:32 lag no Jun 14 10:10:38 You nailed that in moments few Jun 14 10:10:47 as in the machine is now working for you, but not for him Jun 14 10:10:53 Ok, sorry for giving false hope :) Jun 14 10:11:00 Oh, that sucks Jun 14 10:11:07 (for him) Jun 14 10:11:09 there is a collision due to two different chroot types, a normal one, and a union one for hard builds Jun 14 10:11:09 :) Jun 14 10:12:27 I'll not pretend to know a great deal about chroot architecture. I thought they were just different directories each with their own TLD in Jun 14 10:18:49 lag, nope they arn't quite. well they are in the normal case which you use to build, and are union mount overalys in the case of the sbuild use case Jun 14 10:18:58 and by default they use the same names and everything breaks Jun 14 10:19:21 i think i have it sorted out though Jun 14 10:22:01 Good stuff Jun 14 10:24:51 apw: ok, let me try again. Jun 14 10:25:19 cooloney, not fixed yet Jun 14 10:25:33 cooloney, i'll let you know when we are there Jun 14 10:25:33 !Good stuff Jun 14 10:25:34 Factoid 'Good stuff' not found Jun 14 10:25:46 lag, i am waitng on a new chroot building Jun 14 10:25:53 apw: no, failed again. ok, no problem. Jun 14 10:26:09 * lag gigges - "stupid bot" Jun 14 10:26:16 cooloney, yep i know Jun 14 10:26:26 sbuild -d sbuild-maverick-armel PACKAGE*.dsc Jun 14 10:26:43 cooloney, could you try again now, note the sbuild- prefix on the chroot name as above Jun 14 10:29:01 apw: thanks a lot, man. it is running now Jun 14 10:29:22 apw: i guess you created a new chroot named sbuild-maverick-armel? Jun 14 10:30:15 cooloney, yes, as it needs to be a different form from the normal ones, in which changes are persistant Jun 14 10:30:35 i've only make one on that machine and only for maverick Jun 14 10:30:52 now i know it works i'll get with rtg to get them made across the board Jun 14 10:31:12 apw: ok, very nice, thx. Jun 14 10:31:30 cooloney, if you need a different release made let me know Jun 14 10:33:33 apw: one more sbuild-lucid-armel is better. Jun 14 10:33:50 cooloney, you need lucid yes ? Jun 14 10:34:11 apw: yeah, i might use it for build ti-omap and fsl-imx51 Jun 14 10:34:24 cooloney, ok, its building now will let you know when its done Jun 14 10:34:38 apw: currently, i focus on ti-omap4 for M Jun 14 10:34:49 yp Jun 14 10:35:00 apw: thanks a lot. Jun 14 10:35:01 i suspect we'll make them all as they are pretty cheap Jun 14 10:35:42 yeah, i think so, and are those schroot setup scripts in our kteam-tools? Jun 14 10:44:03 cooloney: lag: how are we doing on the bugs in omap support in lucid? Jun 14 10:44:40 can one of you take up the USB OTG bug Jun 14 10:44:44 ? Jun 14 10:46:13 amitk: I am still quite tied up with a suspend-resume bug, but I can take a look for you if you like? Jun 14 10:49:29 lag: please do, assuming you have a beagleboard now Jun 14 10:50:27 I do not :( Jun 14 10:51:00 does either mporier or cooloney have one? Jun 14 10:51:02 Just my lonely panda - no wonder they're going extinct Jun 14 10:51:30 most zoos would love to get a panda :) Jun 14 10:51:49 I couldn't tell you Jun 14 10:55:39 amitk: i'd love to do that. but i don't have beagleboard and panda Jun 14 10:55:49 amitk: what's the USB OTG bug? Jun 14 10:56:02 i saw some fixing in 2.6.34 kernel Jun 14 10:57:55 cooloney: we need to find patches and fix configuration so the USB OTG works in Lucid Jun 14 10:59:26 amitk: ok, any bug on LP? i do love to do that. Jun 14 10:59:58 cooloney: That image works Jun 14 11:00:14 lag: thanks for the testing. Jun 14 11:00:35 But has a heart attack when a monitor is plugged in via HDMI Jun 14 11:02:24 lag: do you think that is the first time you meet that? Jun 14 11:02:27 cooloney: lag: you should bookmark this page -> https://bugs.edge.launchpad.net/ubuntu/+source/linux-ti-omap/ Jun 14 11:02:56 no such thing in .33 kernel? Jun 14 11:03:03 amitk: got it. thx Jun 14 11:06:52 amit: I already have it, thanks Jun 14 11:06:58 cooloney: bug592295 Jun 14 11:07:11 cooloney: bug 592295 Jun 14 11:07:14 Launchpad bug 592295 in linux-ti-omap (Ubuntu) "omapdss DISPC error: SYNC_LOST_DIGIT (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/592295 Jun 14 12:26:22 cooloney: Are you still around? Jun 14 13:39:35 cwillu@lucid-zippy:~$ sudo modprobe usbhid Jun 14 13:39:35 [sudo] password for cwillu: Jun 14 13:39:35 [ 400.146759] usbhid: unknown relocation: 3 Jun 14 13:39:35 FATAL: Error inserting usbhid (/lib/modules/2.6.35-rc3-dl0/kernel/drivers/hid/usbhid/usbhid.ko): Invalid module format Jun 14 13:39:51 that doesn't strike me as the sort of thing that should be possible :p Jun 14 13:47:16 does anyone know the url for gcc 4.4 in Maverick for ARM ? Jun 14 13:47:53 I dug around on launchpad, but only found lp:ubuntu/gcc-4.4 Jun 14 14:02:58 rsavoye, http://ports.ubuntu.com/ubuntu-ports/pool/universe/g/gcc-4.4/ Jun 14 14:08:48 is there a bzr branch for it ? Jun 14 14:09:01 apt-get source it, it'll tell you if there is Jun 14 14:09:06 but yes, probably Jun 14 14:09:17 I wanted to check the level of ARM patches Jun 14 14:10:02 okay, that's a week since rcn was last on, anybody want to come with me on a rescue mission? :p Jun 14 14:10:29 https://launchpad.net/ubuntu/+source/gcc-4.4 Jun 14 14:11:40 that looks like it, thanks Jun 14 14:11:41 rsavoye: afaik, doko uses debian svn for gcc devel, svn.debian.org -- project: gcccvs Jun 14 14:12:14 I wanted to check the code sourcery patches for ARM, which aren't in the Debian sources Jun 14 14:12:41 there are some new patches in gcc trunk for Android I was going to migrate Jun 14 14:13:06 rsavoye: http://svn.debian.org/wsvn/gcccvs/branches/sid/gcc-4.5/debian/patches/#_branches_sid_gcc-4.5_debian_patches_ Jun 14 14:13:39 right, but this patch I don't believe is in debian at all Jun 14 14:13:41 rsavoye: where are CS patches at? Jun 14 14:13:58 got me, but I heard there was one for ARM stuff Jun 14 14:14:24 well, there are tons of fixes and patches for ARM Jun 14 14:14:43 as the CS sources haves fixes for 4.4 based on 4.6 Jun 14 14:15:04 not a huge big deal, I was just doing some Android toolchain hacking and thought maybe I'd check Jun 14 14:15:26 they backport 4.6 development into 4.4 (probably for next release) Jun 14 14:15:38 yes Jun 14 14:15:53 maybe it's not merged in yet Jun 14 14:15:54 well, g'luck! Jun 14 14:16:29 Guess I'll stick to my build of 4.5 for now. Jun 14 14:22:59 where is the soucelist for arm Jun 14 14:29:16 wocao, how do you mean? Jun 14 14:29:44 sources.list? Jun 14 14:29:54 sauce list? Jun 14 14:29:55 :) Jun 14 14:30:02 yum Jun 14 14:30:09 sauce Jun 14 14:35:57 sourcelist for arm platform Jun 14 14:36:00 ? Jun 14 14:36:15 where i can fine Jun 14 14:36:17 where i can find Jun 14 14:37:09 in /etc/apt/ Jun 14 14:41:50 wocao, deb ports.ubuntu.com/ubuntu-ports lucid main universe Jun 14 15:09:13 * cwillu_at_work found rcn Jun 14 15:59:22 lag: what was that ? Jun 14 16:10:01 amitk: Ping Jun 14 16:13:41 Ogra: Ping Jun 14 16:19:36 lag, yep ? Jun 14 16:20:00 If I'm fixing bugs for Panda, which tree should I be using? Jun 14 16:20:14 Things seem a little scattered Jun 14 16:20:24 no idea, cooloney should have created one on the server Jun 14 16:20:37 He has one for Maverick Jun 14 16:20:42 right Jun 14 16:20:45 Hang on Jun 14 16:20:54 panda will be maverick Jun 14 16:21:20 He has one for Maverick in his own area Jun 14 16:21:25 hmm Jun 14 16:21:29 Will panda only be Maverick? Jun 14 16:21:32 then i dont know Jun 14 16:21:35 yes Jun 14 16:21:37 I have a working Lucid kernel? Jun 14 16:21:40 good morning Jun 14 16:21:51 lag, we wont add new kernels to lucid :) Jun 14 16:21:55 its released Jun 14 16:22:17 Okay, but the Maverick one doesn't work on Panda yet? Jun 14 16:22:19 Or does it? Jun 14 16:22:23 no idea Jun 14 16:22:30 afaik we dont have binaries yet Jun 14 16:22:34 I've nearly finished the first set of ARM symbol de-dup patches. I honestly had no idea how much work I had signed up for at UDS. This is tough, tough work Jun 14 16:22:37 so i havent tested anything yet Jun 14 16:22:49 Okay Jun 14 16:23:04 we got the branch on friday afaik Jun 14 16:23:05 So I can't fix Panda bugs yet then, is that what you're saying? Jun 14 16:23:09 I thought it would be easy to separate the functions out .. but it wasn't. Jun 14 16:23:17 ogra : What's new? Jun 14 16:23:41 lag, do we do have bugs for omap4 yet ? Jun 14 16:24:01 since there is no package in the archive yet i doubt we do Jun 14 16:24:13 I have filled one Jun 14 16:24:23 But that was with the Lucid kernel! Jun 14 16:24:34 * lag head explodes Jun 14 16:24:37 's* Jun 14 16:24:42 which was just a build of the plain TI omap4 upstream kernel Jun 14 16:25:01 we dont have the actual kernel yet we will use Jun 14 16:25:18 Okay, I guess that clears things up Jun 14 16:25:23 Phew! Jun 14 16:25:35 the lucid package you use was just a quick build of what was available to get the boards up at all Jun 14 16:25:46 Got you Jun 14 16:26:15 the actual kernel package we will use has to be built by cooloney first, would probably be best to ask if you can help him with that Jun 14 16:26:26 since thats the current blocking factor Jun 14 16:26:45 Okay, I'll have a word with him tomorrow Jun 14 16:27:03 what was your bug about ? Jun 14 16:27:05 * cwillu_at_work continues happily using his working rcn kernels :) Jun 14 16:27:08 clearly a kernel issue ? Jun 14 16:27:17 HDMI Jun 14 16:27:19 Yeah Jun 14 16:27:39 k Jun 14 16:27:42 https://bugs.launchpad.net/ubuntu/+source/linux-ti-omap/+bug/592295 Jun 14 16:27:44 Launchpad bug 592295 in linux-ti-omap (Ubuntu) "omapdss DISPC error: SYNC_LOST_DIGIT (affects: 1) (heat: 6)" [Undecided,New] Jun 14 16:28:05 now, if only I could get firefox to stop crashing when using webworkers Jun 14 16:28:06 When did the Panda board support go into the kernel? Jun 14 16:28:19 Martyn, not yet Jun 14 16:29:18 *nod* The only thing I've seen so far was the proposed patch by David Anders, and it missed the patch window (just like the Smooth-Stone patch missed it) Jun 14 16:29:39 Both then end up on hold until 2.6.36 Jun 14 16:45:23 I'm trying to replicate http://launchpadlibrarian.net/49300098/buildlog_ubuntu-maverick-armel.hdf5_1.8.4-patch1-2_FAILEDTOBUILD.txt.gz in qemu but it keeps segfaulting at the same point in the build processes Jun 14 16:46:03 are qemu-system-arm segfaults the norm? Jun 14 16:56:44 tumbleweed, depneds what kernel you use Jun 14 16:57:37 ogra: the versatile kernel in the repos doesn't work with my qemu Jun 14 16:58:01 so I'm using the one from http://ports.ubuntu.com/ubuntu-ports/dists/lucid/main/installer-armel/current/images/versatile/netboot/ Jun 14 16:59:03 thats the same kernel as the archive kernel Jun 14 16:59:26 but thats alright then, you shouldnt see segfaults Jun 14 17:00:12 ogra: it doesn't seem to be. md5sums don't mathc Jun 14 17:00:32 also 2.8M vs 3M Jun 14 17:00:43 well, debian-installer just pulls the binary out of the package Jun 14 17:00:51 when it builds the netinstall kernel Jun 14 17:01:38 with a HEAD qemu and the archive kernel: http://paste.ubuntu.com/449748/ Jun 14 17:01:52 oh Jun 14 17:02:00 you should mention that you dont use the ubuntu qemu Jun 14 17:02:16 I was getting it with ubuntu qemu first, tried git after that Jun 14 17:02:51 actually haven't tried ubuntu qemu with the archive kernel. does that Jun 14 17:03:12 vmlinuz-2.6.35-2-versatile ????? Jun 14 17:03:18 thats not a lucid kernel Jun 14 17:03:49 ogra: maverick Jun 14 17:04:01 aah, right Jun 14 17:04:04 right, thats untested yet :) Jun 14 17:04:31 theoretically it should work though Jun 14 17:05:00 there were no changes to the versatile branch to my knowledge Jun 14 17:05:10 apart from newer upstream Jun 14 17:06:02 ok. I see no bugs. I'll file one Jun 14 17:06:17 great, thanks Jun 14 17:06:50 anyway, your build log seems to have a SIGILL Jun 14 17:07:23 could be a toolchain issue Jun 14 17:07:34 yeah, I saw that. Tried to replicate, and got sidetracked by this Jun 14 17:07:35 since it happens during linking Jun 14 17:07:40 I tihnk it cgot past that, though Jun 14 17:08:09 are you using a lucid rootfs in your vm image ? Jun 14 17:08:16 maverick Jun 14 17:08:24 ah Jun 14 17:09:45 what's the newest firefox build available for arm? Jun 14 17:10:08 i.e., does anyone have 3.7 builds? Jun 14 17:10:33 i dont think so Jun 14 17:10:46 theer was no firefox upload to maverick yet Jun 14 17:11:00 so the latest in the archive must be the lucid version Jun 14 17:11:44 :/ Jun 14 17:11:51 building firefox isn't fun :*( Jun 14 17:12:47 i dont know if there is any PPA with dailies of trunk Jun 14 17:12:54 asac would know such stuff Jun 14 17:13:36 cwillu_at_work: ogra: https://edge.launchpad.net/~ubuntu-mozilla-daily/+archive/ppa Jun 14 17:13:42 not for armel ;) Jun 14 17:13:52 right, thats what i suspected Jun 14 17:14:02 but firefox 3.7 is there ... which afaik is now branched ... so trunk might be missing temporarily Jun 14 17:14:03 asac, I suppose I could apt-get source it myself :/ Jun 14 17:14:08 cwillu_at_work: yep Jun 14 17:14:23 cwillu_at_work: is there any important reason why we should try a ffox 3.7 build? Jun 14 17:14:34 tahts great, you can tell us about build failures in advance then :) Jun 14 17:14:39 if there is a good reason (like important bug fix landing for arm that needs landing), i spin that in a ppa Jun 14 17:15:28 asac, I can crash a 3.6 build at will by running my app on it? Jun 14 17:15:51 no idea if it's fixed yet, but it's annoying enough to duplicate that I really want it to get fixed by accident :p Jun 14 17:16:05 cwillu_at_work: do you see that problem o intel too? Jun 14 17:16:08 yep Jun 14 17:16:13 then test it with that ppa Jun 14 17:16:23 and it worked on former releases i assume Jun 14 17:16:25 asac, have been, so far so good Jun 14 17:16:25 intel builds are avail there ;) Jun 14 17:16:35 ogra, no, webworkers didn't exist previously Jun 14 17:16:42 ah Jun 14 17:16:49 but they're a pretty big win performance wise in my case Jun 14 17:16:51 ogra: gaaah. wrong terminal. ubuntu qemu boots the maverick kernel fine, sorry Jun 14 17:16:59 phew Jun 14 17:17:13 * ogra wipes the sweat off his forehead Jun 14 17:18:49 tumbleweed, what i would try is to build the package in the same vm inside a lucid chroot, if thta succeeds its most likely a toolchain issue Jun 14 17:20:29 brb supper Jun 14 17:28:12 Martyn: I didn't see patches for either go by on arm-kernel. Where were they posted? (panda and smooth-stone support)? Jun 14 17:31:58 ogra: when I say "boots fine" I mean it says "booting the kernel" and then hangs at 100% CPU Jun 14 18:02:15 * gsnedders found the reason why the BeagleBoard he has wasn't working properly: the NAND is b0rked. Jun 14 22:35:31 ogra, well, I seem to be stable given some code modifications on 3.6.6 from the nightlies Jun 14 22:35:43 on x86 Jun 14 23:20:42 how much space is required to bootstrap ubuntu-minimal on ARM? **** ENDING LOGGING AT Tue Jun 15 02:59:57 2010