[01:55] GrueMaster: from what I saw, it seems that the one at the proposed was the linaro's one, not ubuntu [01:56] the message from martin pit is from 10-26, but the proposed ubuntu kernel is from 11-05 [01:56] I didn't think it would automatically install unless we had the linaro kernel on the original image. [01:58] GrueMaster: ok, I was confused if the user actually tested with the ubuntu one [01:58] as the linaro's kernel got the fix and released some days before the ubuntu one [01:59] but if you say it was tested with the ubuntu kernel, please move it to verification-done and add an extra kernel [01:59] The test process he used was to image an SD card with our released image, update uImage per your instructions on the wiki, then after booting and running through oem-config, he enabled maverick-proposed per the wiki instructions and ran an update. [01:59] because the latest message said you were waiting the tester to confirm the kernel used [02:00] ok then :-) [02:00] should be the correct one [02:00] sorry for the confusion, wasn't clear for me [02:01] Unfortunately I really have no way to test this (other than procedural) and my system is not setup at the moment (office restructuring and I lost a monitor). [02:01] sure, np, you need a xm a3 to be able to test it [02:02] That's why I had someone on #beagle do it. [02:02] np :-) [02:03] I'll be out again, tomorrow is a national holiday here but will be on-line most of the day [02:03] see ya [02:03] That a holiday, dude. You work too hard. [02:04] will be on-line most of the time because I'm going to plan the vacation properly :-) [02:04] I meant "take". [02:04] Ah. [02:04] that will be the next week [02:04] alright, wife is going to kill me, see ya [02:04] Later. === lilstevie|ZNC is now known as lilstevie === hrw|gone is now known as hrw [08:15] morgen [10:01] Hi, i m newbie to this channel. Have a query regarding how can i compile a arm binary on maverick ? [10:02] i have installed arm-linux-gnueabi- package [10:05] silverroots: how you would compile it for your machine? [10:14] hrw, thats actually my Q is..? Can i compile using cross tool chain available in maverick? [10:15] yes, you can [10:15] is there any guide for that? [10:16] normally i m compiling for x86 is using ./autogen && make && make install..but i assume there should be pre things to be done to make it arm binary [10:19] hrw, thanks i guess i got a good link :) [10:20] nope..its not..it has how to for just kernel..not any regular module [10:23] configure --target arm-linux-gnueabi [10:24] hrw, ok will try that now.. [10:30] Are there anyone who happen to have Qt 4.7 embedded compiled against SGX for maverick? [13:32] ogra_ac: bug 675542 [13:32] Launchpad bug 675542 in x-loader-omap4 (Ubuntu) "Fails to apply patches (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/675542 [13:33] hrw, why isnt that properly triaged ? [13:33] please make sure to subscribe ubuntu-armel to bugs you want to be seen by us [13:33] ogra_ac: I just filled it [13:34] ah, yeah, 50sec ago, sorry [13:34] ogra_ac: and I prefer to not subscribe teams to bugs on packages/projects which have living maintainer [13:36] ~curse LP for breaking formatting [13:36] hrw, works fine for me [13:36] (on maverick in a natty chroot) [13:38] fun [13:38] and given that there was no change to the package since maverick release i would wonder why the patch cant apply [13:39] let me check other arm [13:42] crap. on efikasb it works [13:42] ~curse pandaboard [13:43] works on panda too here [13:43] marked invalid [13:43] thanks [13:43] bad things happen in armel world ;( [13:44] qt/kde ftfbs [13:44] blame gcc [13:44] I know [13:44] but there are workarounds [13:44] and nokia for ignoring thumb2 [13:44] yeah [13:45] The qt FTBFS is actually a qt one, not a gcc bug [13:45] at least the thumb=it part of it [13:45] oh ? [13:45] i thought its gccs fault [13:45] ogra_ac: gcc-4.5 dropped workaround [13:45] ah [13:46] sth like that [13:46] well, changed defaults [13:46] but if the defaults carried a workaround, i agree [13:46] qt is armv6 and is UP [13:46] no thumb support, no smp [13:46] and still broken neon detection [13:46] It was a flag to assume that lack of specifier meant thumb specifier, which is ugly and not the upstream default [13:47] yeah [13:49] I hope that one day "mem=1G" will work on my panda [13:50] fix the kernel ;) [13:50] ;D [13:50] Is there a guide where I can use Ubuntu, add some packages, and be able to cross compile to Arm? My current way is to download the CodeSourcery compiler .tar.gz file. [13:50] simosx: wanna cross build ? http://42.pl/u/2u8U [13:51] next thing to blog about... "how to use xdeb in cross builds" [13:53] ogra_ac: implicit-it=thumb wasn't just removed from defaults, it doesn't work at all currently (see 675240). Not sure if that was on purpose or not. [13:53] hrw: thanks :-). It's much easier this way. I need to upgrade to 10.10 to get the ARM GCC. I promise to blog about the proper way. [13:53] ScottK, ouch [13:54] simosx: I have to blog about it too. will replace current link in /topic then [13:55] shit... I forgot that natty+panda == kernel panic on boot [13:55] ogra_ac: Yes. And give Thiago's comment in 673085, I suspect we are just about stuck between a rock and a hard place. [13:55] bug 673085 [13:55] Launchpad bug 673085 in qt4-x11 (Ubuntu) (and 2 other projects) "Qt/KDE fails to build on ARM without implicit-it=thumb (affects: 2) (dups: 2) (heat: 38)" [Undecided,New] https://launchpad.net/bugs/673085 [13:56] ScottK, well, we can always carry a distro patch as workaround for now [13:57] ogra_ac: If we have one that works, yes, but I'd want to understand why upstream wouldn't accept it (I suspect it won't just be cosmetic reasons). [13:57] no, unlikely [13:57] lool, hrw, is the thumb2/multicore issue reported upstream? I didn't spot it in a quick search. [13:59] suihkulokki: no idea, did not tracked [13:59] suihkulokki: but I assume that it is not present in qt bugtracker [13:59] or does nokia care at all [14:00] * ogra_ac guesses nokia will also have to support A9 at some point ;) [14:00] Its hard to care about issues people dont report :) [14:00] hrw: Considering Thiago Maciera is commenting in 673085, I think not caring isn't it. [14:05] shit. booted panda to 460+256MB setup, refetched xloader source and still fail [14:06] now test in tmpfs [14:06] i dont think you have a ram issue there, loolks more liek filesystem corruption to me if a patch doesnt apply [14:06] same [14:07] fresh tmpfs fails [14:07] md5sums of archive are correct [14:11] lool, do you plan to SRU your debootstrap workaround into maverick ? (would that make sense) [14:11] (for the dpkg segfault) [14:21] suihkulokki: Dunno [14:22] suihkulokki: I guess not [14:22] ogra_ac: Is it a problem under maverick? [14:22] lool, if you build natty, yes [14:22] its nattys dpkg that has the issue so building a natty chroot in maverick causes the same issue [14:23] ogra_ac: I wouldn't mind; but I'd rather get the fix merged in Debian first; you're welcome to SRU it too; I don't debootstrap natty from maverick myself these days, albeit I could have to in the future [14:23] but effectively its just a workaround [14:24] Ideally, we'd fix dpkg in natty and this would fix debootstrap from maverick as well [14:24] which i expect to be dropped as soon as dpkg is fixed [14:24] well, i think thats still the plan [14:24] so SRUing back and forth for the workaround probably doesnt make sense [14:33] ogra_ac: Would you have time to reduce the test case in dpkg? [14:33] I'm not sure James is looking at it and I certainly don't have the time to look into it this week === asac_ is now known as asac [14:36] lool, i think james is supposed to do that [14:37] at leqast thats how i understood colin === hrw is now known as hrw|gone === asac_ is now known as asac === hrw|gone is now known as hrw|n900 [16:36] Quick question I want to get a toshiba ac100 and I know that most stuff works out of the box in 10.10, is there anything to stop me getting one and sticking ubuntu on it? [16:37] * fagan is looking to get one [16:37] Last I heard sound and video werent working all that well [16:47] ogra_ac: I'm working on testing the IT patch from michaelh1 for qt4-x11. My plan is to build a fixed Qt locally and then if I can build a package that currently FTBFS due to the lack of implicit-it=thumb, then I'll declare victory and upload it. [16:48] ScottK, awesome [16:48] It's a LOT easier to work on armel stuff once one has some hardware ... [16:48] heh, yeah === mturquet1e is now known as mturquette === j_ack_ is now known as j_ack === hrw|n900 is now known as hrw [21:36] ScottK: what HW you are using for builds? [21:42] Hey all. [21:47] anyone even vaguely awake? [21:55] a bit [21:59] hrw: I've got three Efika MX smarttops (running Karmic) sharing the build using icecc. [22:00] nice [22:02] The trickiest part was arranging to have icecc also distribute the compiler to use so they all build their bits using the same GCC version. [22:02] * ScottK didn't figure out that part - was a team effort. === hrw is now known as hrw|gone [22:41] Anyone awake? I have a question about platform devices... [22:41] specifically, initializing the data that goes into the pdev->dev.platform_device [22:41] er dev.platform_data === JaMa is now known as JaMa|Zzz === Guest89205 is now known as Martyn [22:50] 'ogra, you online? [23:40] I've been told that the pandaboard supports 1080p playback, but that xbmc and mythtv do not work with that yet, but "ubuntu does".. could anyone clarify the usefulness of a pandaboard with ubuntu as a media front end? [23:41] Sp0tter: you may have to wait for a bit for someone to answer [23:41] alrighty [23:41] Sp0tter: the other person to as is robclark [23:41] prpplague: do you know where to purchase a case for the pandaboard? [23:42] or is it just make your own only [23:42] Sp0tter: no case is available currently [23:42] Sp0tter: pretty much [23:42] k [23:42] Sp0tter: there is plans for an accessory board plus case called the Bamboo [23:42] Sp0tter: http://www.elinux.org/Panda_Bamboo [23:42] yea it seems to be evolving quickly [23:43] thats the only reason i'm hesitant to order one [23:43] 1 month wait too