=== bjf[] is now known as bjf[afk] [03:34] * mozzwald is away: sleepytime === bjf__ is now known as bjf === bjf is now known as bjf[afk] [06:24] haha [06:25] lool: look, someone spamming with his away again :D [08:28] Hello all [08:30] I have a doubt in the RootfsFromScratch . anybody there to help me [08:30] ? [08:38] Hi === hrw|gone is now known as hrw [08:47] morning [08:54] morning hrw === XorA|gone is now known as XorA [09:04] zyga: hi [09:05] Hello all [09:05] i have an issue in running ARM/RootfsFromScratch [09:05] you was as neo_ few minutes ago? [09:05] 09:31 < neo_> I have a doubt in the RootfsFromScratch . anybody there to help me [09:05] yes [09:06] so next time do not change nick [09:06] sure [09:06] so what issue? [09:06] i am unable to use the mouse inside the Qemu VNC [09:06] heh, neo is a bad choice of nick anyway :-) [09:06] ha ha :) [09:06] XorA: thats too [09:06] ARMport is not better anyway [09:06] I ll change it soon ;) [09:07] its the equiv these days of countzero in the 90s [09:07] ARMport: never used qemu with vnc [09:07] XorA: mess with the best, die like the rest? [09:07] don't get u ! [09:07] hrw, czesc :-) [09:07] I mean using the command qemu-system-arm -M versatilepb -cpu cortex-a8 -kernel ./vmlinuz -hda arm-rootfs.img -m 256 -append "root=/dev/sda mem=256M devtmpfs.mount=0 rw" [09:08] I tried to run the qemu, [09:08] and opened the emulated ARM image in the VNC viewer [09:08] amusingly "The Plague" didnt really take off as a nick even though the film was awesome [09:08] XorA: i don't think so [09:09] its one of my favourite films, brings back so many memories of 2600 meets [09:09] has any one tried running the ARM image using qemu? [09:10] XorA : its my fav too [09:10] XorA: from 2006 http://www.imdb.com/title/tt0479125/ one? [09:11] hrw: from Hackers [09:11] ah [09:11] http://www.imdb.com/title/tt0113243/ [09:11] long time since last time I watched Hackers [09:12] o.. there was 720p edition [09:13] ..oo00OO I shall have to look for bluray then [09:14] Instead of going to bluray I bought mini hifi with dvd [09:16] I have a ps3 which is awesome for video playback [09:16] and require transcoding of all divx movies [09:16] nope [09:16] plays DivX fine, just not DivX:-) [09:17] what about matroska container? etc? [09:17] friend told me that he has a bit of work with most movies to get them played on his ps3 [09:17] it doesnt understand mkv, but tsmuxer will remux is happilly to mpeg2 ps which it does understand [09:18] ps3-media-server does it on the fly :-) [09:18] its DivX certified, so plays them without issue [09:19] http://ps3mediaserver.blogspot.com/ [09:19] I have to check divx on my hifi [09:19] also divx certified [09:19] the thing to watch out is that xvid != divx [09:19] different set of ASP options [09:20] most players dont care, but a few do [09:20] and one day fight enough to get some movie converted to avhcd (or how it was named) and then I will check my TV capabilities [09:20] warez stuff is normally xvid for obvious reasons === ericm-Zzz is now known as ericm [11:03] sebjan: those images work on lag's panda, thanks a lot, man [11:04] cooloney: I've already emailed sebjan [11:05] Sebjan: What's changed in u-boot lately? Does it now check for system IDs? [11:05] *different system IDs [11:39] ndec: you around? [12:06] lag: I did not have a look, and I did not build myslef the images I provided to you. Will try to get so info about that. [12:08] Thank you sebjan [12:08] lag, the u-boot branch from gitorious.org should be fine, i'll update the ubuntu package soon [12:09] lag, http://www.gitorious.org/pandaboard/u-boot [12:12] ogra: Yes it is. I was just curious as to what the differences were - thirst for knowledge and all that. ;) [12:13] well, the diffs might tell :) [12:17] #define MACH_TYPE_OMAP4_PANDA 2791 [12:17] looks like that one [12:17] from include/asm-arm/mach-types.h [12:23] hate make [12:23] I would if I had the sources for both ;) [12:24] lag, http://www.gitorious.org/pandaboard/u-boot has the source :) [12:25] that was copy pasted from the web ui [12:25] That's the one I do have [12:39] Hello === Guest23625 is now known as JameswStubbs [12:40] Is there an arm version of the Ubuntu Minimal Cd? [12:41] Specifically Karmic ? [12:41] lag: so you can boot with the images I provided to you, but not with the gitorious ones? [12:42] sebjan: I think the ones that were provided for me were the ones used for Blaze [12:42] I don't think they were the gitorious ones [12:43] lag, the ones i provided to you were older panda binaries [12:43] without the sysid in them, they worked on panda and blaze by a matter of luck [12:44] Okay [12:44] there was a change in the panda kernel that now checks for that ID afaik [12:44] Are you going to update your sources? [12:44] yes [12:44] Bump [12:44] Then upload them to your people account? [12:45] no, to maverick :) [12:45] Okay [12:45] https://edge.launchpad.net/ubuntu/+source/u-boot-omap4 [12:46] thats using the sources from omapzoom, predating the panda gitorious tree [12:46] i'll pull in the panda stuff [12:46] What git repos do I need from kernel.ubuntu.com? [12:46] no idea [12:46] aks #ubuntu-kernel :) [12:46] *ask even [12:47] Wont they be the same ones you use? [12:47] Is there an Ubuntu Minimal for the Arm arch? [12:48] JameswStubbs: I don't know the answer for sure, but I wouldn't have thought so [12:48] JameswStubbs: just wait a bit, maybe you will get an answer eventually :) [12:48] JameswStubbs, you can use the server images or a netinstall [12:48] ogra: i dont think a netinstall would be useful :p [12:48] I'll look into the server image for arm thanks ogra [12:49] lilstevie, ?? [12:49] No lilstevie it won't be, can't do a net install when we can't boot :p [12:49] We're porting Ubuntu to iPhone [12:49] ogra: the device we are working on is the iPhone as JameswStubbs just said [12:49] JameswStubbs, oh, then better look at the channel topic for rootfs from scratch [12:49] Debating between using debian armel or ubuntu 9.10 [12:49] use rootstock to create an ubuntu-minimal rootfs [12:50] I'm looking into that aswell :) [12:50] the images we provide are only for supported arches so the server image wont gain you much on an iphone [12:50] JameswStubbs: Is this a project, or are you doing it for a company? [12:50] lag : part of the iDroid project [12:50] lag: its an open source project [12:51] Sounds good. Best of luck with it. [12:51] Thanks, we'll need i [12:51] t [12:51] heh [12:51] * lilstevie is not a fan of having android [12:51] so instead going for something a little more sane :) [12:52] I have to admit, I quite like it [12:52] I have an HTC Desire which I am a fan of [12:52] heh, android is not nice on proprietry devices that have to have drivers written for them, [12:52] Only reason I want it is for GTK Ettercap :) [12:53] I see, and what do you want to do with Ettercap? ;) [12:54] lag: Monitor my own network for educational purposes of course. [12:55] JameswStubbs: Of course. How silly of me! ;) [12:55] lilstevie: https://wiki.ubuntu.com/Specs/ARMServer [12:56] It's for armv7 though, I think we'll need to rootfs [12:56] i think building a rootfs from scratch will be the easiest for our purposes [12:57] Defiantly do you think we'll still be ok using the iDroid kernel [12:57] ? [12:57] yeah [12:57] ogra: is this the right place to go for advice with rootfs? [13:03] Hm, Can you run rootfs from an x86 machine with the intention of building for Arm? [13:10] orga : ? [13:14] * mozzwald is away: hardly working [13:26] JameswStubbs: Have you tried using rootstock? [13:27] JameswStubbs, yes [13:28] if you use the qemu-arm-static package [13:28] you can a) use rootstock to create a tarball and b) just create armel chroots using qemu-debootstrap [13:30] Thanks, I'll be using rootstock on x86 ubuntu, creating an image, using iDroid's kernel then mounting in qemu [13:31] ogra: So you also need initrd on the card to run? [13:31] lag, you *can* use ubuntu without initrd but i wouldnt recommend it other than for testing [13:31] JameswStubbs: That should work. I'm using rootstock on an x86 machine too [13:31] Is the idea behind rootstock to build a Very minimal system and then to build from there? [13:32] well, depends if the droid kernel has all the features ubuntu uses [13:32] JameswStubbs, yes, use -s ubuntu-minimal [13:32] and do everything you want additional on the real HW [13:32] :p Thanks ogra, I was going to leave the --seed option blank :) [13:33] should work as well [13:33] will result in the same :) [13:34] ogra: So, do I just bang the rootfs.tgz and initrd onto the rootfs partition and expect it to work? [13:34] lag: Are you talking about something different to me? (just stopping myself getting confused :p ) [13:34] what features are required from kernel for ubuntu [13:35] JameswStubbs: Yes, I'm doing something slightly different [13:35] Ok [13:35] with rootstock we need to generate an img [13:35] as openiboot for the iphone loads an img for rootfs [13:36] lag, nope [13:36] lag, you put the rootfs into the rootfs partition [13:36] and initrd wherever your bootloader wants it [13:38] lag, for omap thats usually the first partition on the SD and it should be a vfat [13:38] lilstevie: Yeh that's right, but Idroid has multiple images where as our whole distro will be in one image [13:38] We might have to point oib in a different direction to find the info [13:38] JameswStubbs: thats because they are mounted individually [13:38] So we could probably get away with renaming or image system.img [13:38] initrc points to the other images [13:39] It doesnt do a hash check or anything [13:39] Yeh lilstevie [13:39] so we just need our own init [13:40] make sure that your img usues a supported filesystem [13:41] ogra: So the initrd goes in the /boot partition then? [13:41] no [13:41] it goes into the vfat [13:42] and you need to specify it in your bootm command for u-boot [13:42] /dev/sde1 on /media/boot type vfat [13:42] well, you labeled it boot :) [13:42] I didn't [13:42] something did [13:42] The script that cooloney pointed me to did [13:42] :) [13:42] ah [13:43] lilstevie: We wouldn't necesarily need our own init, simply modify iDroids to point to the right images [13:43] its not /boot (and cant be) [13:43] it just smells like /boot :) [13:43] Mm, fresh smell of /boot in the morning [13:43] eh? [13:43] /boot cant be vfat on debian based systems [13:44] dpkg unpacks the kernel there, while unpacking it creates hardlinks of the former files [13:45] Oh okay, it's just the partition label [13:45] It won't actually be /boot, as in /boot/grub [13:45] vfat doesnt support hardlinks [13:45] so /boot cant be vfat [13:45] But yes, I see what you mean [13:46] right [13:46] Perhaps it's a confusing name [13:46] its still your boot-partition [13:46] but you cant mount it as /boot [13:46] Correct [13:46] Which is where the initrc needs to go, correct? [13:47] since our kernels and initrds live in /boot, we have a tool called flash-kernel thats works around this prob [13:47] its usually called buy update-initramfs (which in turn is called by the kernel package when you install it) [13:47] flash-kernel mounts /dev/mmcblk0p1 and runs mkimage for kernel and initrd to create uInitrd and uImage [13:48] (or at least iots supposed to, it doesnt do that for the panda or beagle yet, i will have to write some code still) [13:51] lag, mkimage -A arm -O linux -T ramdisk -C none -a 0x0 -e 0x0 -n "Ubuntu Initrd" -d /path/to/initrd.img-$ver uInitrd [13:51] you will need that [13:51] Ag [13:51] Ah* [13:51] fatload mmc 0:1 0x80000000 uImage [13:52] fatload mmc 0:1 0x81600000 uInitrd [13:52] bootm 0x80000000 0x81600000 [13:52] thats for the u-boot prompt [13:52] and should get you a boot with initrd [13:54] So the uImage and the uInitrd go on the same place on the SD card? [13:56] right, into the vfat [13:57] Yep [13:57] And is the rootfs okay to go in as a tgz? [13:57] Or should I unpack it first? [13:57] Thanks again for your help ogra [13:58] lag, format the partition and unpack [13:58] JameswStubbs, welcome, good luck [13:59] Thought as much [13:59] * lag wondered when they'd put tgz support into the kernel [14:01] heh [14:08] NCommander, bah, bug 583317 will bite us badly [14:08] Launchpad bug 583317 in genext2fs (Ubuntu) "genext2fs creates revision 0 filesystems instead of revision 1 (affects: 1) (heat: 156)" [Undecided,Confirmed] https://launchpad.net/bugs/583317 [14:19] ha ! [14:19] finally [14:19] how silly [14:20] * ogra finally found the reason why resize2fs didnt work in initramfs ... seems it has a call to /etc/mtab hardcoded [14:36] Is the rootfs created by rootstock meant to pull up a terminal? === ericm is now known as ericm-afk [14:37] lag, yep [14:37] Freeing init memory: 192K [14:37] init: ureadahead main process (55) terminated with status 5 [14:37] twl: i2c_write failed to transfer all messages [14:37] i2c_omap i2c_omap.1: controller timed out [14:37] twl: i2c_read failed to transfer all messages [14:37] twl6030_bci: error reading CONTROLLER_CTRL1 [14:37] Then nothing [14:38] whats your console= setting in the cmdline ? [14:38] and what was the rootstock cmdline you used :) [14:39] by default ubuntu has no serial getty running ... you can enable it in rootstock with the --serial switch [14:40] I did that [14:40] Let me have a play [14:40] do you have a monitor attached ? [14:40] i'm pretty sure you have a login prompt on tty1 :) [14:40] I don't have a monitor which supports HDMI [14:40] you should get one :) [14:41] or DVI at least and an adapter [14:41] cheapest lcd with dvi is enough for BB [14:41] True HDMI doesn't work with adaptors [14:42] lag: define 'true hdmi' please [14:42] hrw, the panda has hdmi [14:42] True HDMI sends co-data to control the output settings [14:42] ogra: with audio over hdmi or just hdmi connector? [14:43] lag: you mean HDMI/CI? [14:43] I forget what it's called [14:43] CI or CEC - adapted from SCART connector [14:43] hrw, the current build i have has a headphone jack and two HDMI ports [14:43] I just know that when I was working on set-top boxes some worked with a connector and some didn't [14:43] used to say 'TV - I am here, switch to my hdmi port' [14:43] hrw, but that might change [14:44] Could well be [14:44] Will the Panda work with a connector? [14:44] BB has hdmi connector just because it is smaller then dvi. works fine with hdmi->hdmi cable and with hdmi->hdmi-to-dvi adapters [14:44] so far worked for me with dvi monitors and hdmi tv [14:44] lag, i havent booted mine beyond the uboot prompt yet [14:44] What about Panda? [14:45] ogra: Why so? [14:45] Aren't you working on the kernel? [14:45] while the BB has DVI i think the panda has HDMI [14:45] lag, i'm working on all the rest :) [14:45] Right, so the connector will be no good? [14:45] yeah, likely [14:46] Well I don't want to outlay for a new monitor :S [14:49] * ogra stbs jasper [14:49] *stabs even [15:05] cooloney: ping [15:11] lag: lee, man [15:12] ubuntu@panda:~$ ls / [15:12] bin dev home lost+found mnt proc sbin srv tmp var [15:12] boot etc lib media opt root selinux sys usr [15:12] Good work [15:13] cooloney: --^ [15:13] lag: haha, cong, man [15:14] cong? [15:14] lag: very good news before my sleep. you built your own u-boot to make it boot? [15:14] lag: sorry, congrats [15:14] The u-boot I was trying to use didn't pass the kernel's system ID check [15:14] king cong :) [15:15] (very big congrats) [15:15] ogra: lol [15:15] You mean King Kong? [15:15] pfft [15:15] chars [15:15] lag: hmm, that reminds me a warning message from dmesg [15:15] Oh? [15:16] lag: Uninitialized omap_chip, please fix! [15:16] did you see that in your dmesg? [15:16] Oh yeah, I see it [15:16] Yeah, it's in dmesg [15:16] the kernel code checked omap cpu id or system id something like that [15:16] ubuntu@panda:~$ dmesg | grep -i omap_chip [15:16] Uninitialized omap_chip, please fix! [15:16] and failed [15:17] then print out this [15:17] maybe relatedt [15:17] i will check them later [15:17] lag, thanks for testing, man [15:17] * cooloney gonna sleep now. [15:17] No problem - speak to you tomorrow [15:18] have a nice day, guys === bjf[afk] is now known as bjf [15:32] ogra: any issues with pandaboard or trees? [15:32] orbarron, see above, lag successfully booted his panda [15:32] * orbarron reads... [15:32] i'm currently hogged with image building which i do on a C4 [15:33] not much spare time for omap4 work here [15:34] np... [15:44] lag: pandaboard has DVI + HDMI and HDMI to DVI connection will work with a small hack [15:48] I should have a monitor with HDMI built-in with me tomorrow [15:49] What would the hack be? [15:52] * orbarron dealing with dsi clk issues but for the mean time you can do this --> http://pastebin.com/8LVZYWnx [15:53] orbarron: hi Omar [15:53] hrw: sup... [15:53] asac [16:04] mpoirier_: !! [16:05] asac: good morning. [16:05] asac: ogra tells me you have a beagleboard C4 - is this correct ? [16:06] mpoirier_: why? ;) [16:07] asac: I could use it if you don't need it. [16:07] I need it [16:07] asac, mpoirier is our new omap3 maintainer in the kernel team [16:07] mpoirier_: alread discused with ogra. we gave boards to davidm for distro [16:07] mpoirier, we're waiting for feedback from lool, he should know what happened to the missing boards [16:08] if nothing helps and its urgent i would suggest to buy a board ;) [16:08] according to asac they should all be at davidm's place [16:09] ogra: and davidm is coming back next week ? [16:09] will be faster to wait until next week [16:09] mpoirier_, yes [16:09] asac, and we *have* 7 boards in mobile [16:09] its just a matter where they actually are [16:09] well ... one dev blocked for half a day is probably more expensive than a board ;) [16:10] and we are already pulling in 4 devs into this discussion ;) [16:10] asac, ordering and shipping a beagle takes more than a weekend [16:10] 4 ? [16:10] me you mpoirier_ and lool [16:10] well, no trace of lool :) [16:10] so 3 :P [16:11] but lool will have to read plenty of backlog now ;) [16:12] anyway ... mpoirier_: welcome! [16:22] mpoirier_, GrueMaster is your guineapig ! [16:23] ogra: cool - thanks. [16:24] GrueMaster: can I send you a uImage ? [16:27] Or post it online and I can download it. [16:58] asac: Is the backlog relevant still? [16:58] I read this channel usually, but I wont make the full backlog Im afraid [16:58] hey mpoirier_! [16:59] lool: good morning. [16:59] lool: I need a beableboard C4, do you need yours ? [16:59] lool, it isnt [16:59] lool: what version then ? [16:59] lool, the discussion in the other channel solved itr [16:59] mpoirier_: Mine is a C3 and is personally owned [17:00] *it [17:00] mpoirier_, see PM [17:00] mpoirier_: I understand that if you want to find an idle one, you should ask davidm cause he might have a couple of spare ones [17:00] lool: yes, he's on vacation. [17:00] mpoirier_: If that blocks you in anyway, I highly recommend you talk to your manager: buying it and expensing it might be cheaper than the time you're losing searching for one [17:00] lool, for now GrueMaster will help testing [17:01] mpoirier_: Are you testing kernels? [17:01] mpoirier_: You could also use qemu-maemo :-) [17:01] (/me lols) [17:01] *grin* [17:01] lool: yes, I ping'ed you before GrueMaster got on board. [17:01] lool: I need real Hw [17:01] mpoirier_: So mattman has been making progress on booting up to a framebuffer our regular Ubuntu omap kernels in qemu-maemo + his patches [17:02] lool: I'm after a misbehaving SD card controller. [17:02] mpoirier_: So perhaps you can ask pgraner whether you can expense it? [17:02] lool: yes, that is also in the works. [17:02] lool, davidm shoudl be able to send one out on monday [17:03] Up to you guys, many options on the table [17:03] lool: ogra is correct. [17:03] so it should arrive on wed. on CDN [17:03] *in [17:03] lool: thanks for getting back to me. [17:03] lool: nope [17:04] orga: I'm curious...Is there any stats on how many people actually downloaded the 10.04 Ubuntu image for beagleboard from ubuntu.com ? Who would know this detail [17:04] jayabharath, hmm, no idea, our is team probably, try to ask elmo in #ubuntu-devel or #ubuntu-release [17:04] ok [17:05] orga: Also, I got a bunch of beagle folks a little annoyed with the way Ubunutu CD images.. work... it boots up from SD... overwrite the uboot on nand... messes with the existing SD card... [17:05] (i guess -release is the better channel for that) [17:05] orga: Will check on -release channel [17:05] jayabharath, yeah, that will be fixed in maverick [17:05] have nice day [17:06] jayabharath, i wanted to use a vfat partition but had no time left to implement that, NAND was the quickest way back then [17:06] orga: I see... [17:06] maverick will be quite different in that regard [17:07] orga: gotcha.. wanted to feed this input back to you... [17:07] orga: As you are changing behavior feel free to discuss on #beagle channel too... [17:07] yeah, i knew it wasnt the best to do when i implemented it, if someone asks you again send them to me, i'll pay them beer [17:07] orga: or write up a blog post :) [17:08] ogra: I am sure people a few folks here can use beers ;) [17:08] jayabharath, i will hopefulle meet some of the #beagle folks tomorrow at linuxtag [17:08] * ogra wants to meet koen in person at least [17:08] ogra: most certainly... I belive Koen is planning to be there... [17:08] yeah, he will leave tomorrow evening [17:09] orga: check his pockets he will have a few beagleXMs :) [17:09] thats why i go tomorrow, else i would have gone fri and sat [17:09] heh [17:12] * ogra goes back to his initramfs pain [17:20] NCommander, did you see my ping above btw ? [17:20] NCommander, bug 583317 [17:20] Launchpad bug 583317 in genext2fs (Ubuntu) "genext2fs creates revision 0 filesystems instead of revision 1 (affects: 1) (heat: 156)" [Undecided,Confirmed] https://launchpad.net/bugs/583317 [17:40] yay [17:41] asac, dropping the fsck and instead diong e2fsck during image build fixes jasper :) [17:42] ogra: hackish! [17:43] why ? [17:43] j.k. [17:43] heh [17:43] what is that doing different? [17:43] now i need a solution for that horridly long resize process [17:43] write a script to put stuff on sdcard ;) [17:43] or other media [17:44] the fsck exited with exit 2 due to the fact that the timestamp is out of order [17:44] that tears down the whole initramfs script [17:44] so it never gets to the resize operation [17:44] no, no scripts, the image will be dd'able like all ubuntu images [17:48] (initramfs) df [17:48] Filesystem 1024-blocks Used Available Use% Mounted on [17:48] none 116988 128 116860 0% /dev [17:48] /dev/mmcblk0p2 3767479 400383 3177098 11% /root [17:48] wohoo ! [17:57] Begin: Running /scripts/local-premount ... [17:57] [ 4.288238] mmcblk0: p1 p2 [17:57] Resizing root filesystem please wait, this will take several minutes ... [17:57] * ogra dances [18:03] cool. [18:03] hmm, several ... [18:07] 6 min for 4G [18:10] what system is this on? [18:11] Also, what class SD card? [18:11] ogra: ^^^ [18:12] class4 4G sandisk on the beagle C4 [18:12] growing the root from 450M to 4G [18:12] That will be a bit slow. [18:13] the netbook image will be 1.4G [18:13] Slow I/O, slow SD card. [18:13] i have some hope it will be faster to grow from 1.4 to 4G [18:13] 1.4G for the image? Is that the download size? [18:14] nop [18:14] ok. [18:14] the image will be a .img.bz2 file :) [18:14] will be around 500M [18:14] you need to uncompress and dd it [18:14] We need to keep it below 700M if possible. [18:15] well [18:15] i'll do that if thats possible, but its a preinstalled image [18:15] its not like anything we built before [18:15] I understand. That is going to have to be part of the package selection process. [18:16] so i dont really care if it bumps above 700M [18:16] The 700M limit is mainly for distribution purposes. So people can order a copy of the image on CD. [18:16] who does that ? [18:17] where would that be offered ? [18:17] People with low bandwidth. [18:17] for these 20 ppl we can also send out a dvd [18:17] Doesn't that show up on ubuntu.com? [18:17] heh. [18:17] since that will be a manual effort anyway [18:17] arm images arent offered in the shop [18:18] oh. [18:18] cross your fingers, i might be able to give you a "serial only, commandline only" preinstalled image today [18:18] Then why the fuss over image size in past cycles? [18:18] That would be cool. [18:18] I can always add from there. [18:19] ask david and persia :) [18:19] (assuming I have some console). [18:19] i always insisted its nonsense to bind ourselves to 700M [18:29] GrueMaster, bah, oem-config doesnt start :/ [18:29] oops. [18:29] lol [18:30] heh, i spoke to soon [18:30] it just runs on tty1 [18:30] so you have to jump between serial for booting and tty1 for oem-config with this image, lets see [18:34] hmm, no keyboard might be because it doesnt think there is a console [18:34] * ogra retries [18:34] last round now [18:48] GrueMaster, http://people.canonical.com/~ogra/jasper/ give it another 10min for the upload and you can pull it [18:48] (shoudl be around 132M if its done) [18:48] ok [18:49] so, why hold down the user button? [18:49] GrueMaster, oh, and ignore the OOM message from ureadahead during boot :) i have to talk to Keybuk about that one its no issue (at least it shouldnt be) [18:49] to make it use x.loader from the SD [18:50] ah. [18:50] *x-loader [18:50] i have one old B3 board here that doesnt get along with the old x-loader in NAND and the new u-boot on the SD [18:50] I would have just used the uboot cmdline, but this works. [18:50] so thats the safest option [18:51] * ogra really hopes oem-config will DTRT [18:51] i havent seen it finishing yet, but i have to end my day [18:52] ok. I'll start testing and let you know how it turns out. [18:52] oh, and if you fiddle with kernels, you have to mount mmcblk0p1 manually and generate uImage and uInitrd manually onto that partition [18:52] flash-kernel doesnt know the new image design yet [18:53] I only have 2G & 4G SD cards atm. Will order an 8, 16, & 32 on payday. [18:53] be aware it will all be fiddly and broken atm :) [18:53] ok [18:53] I didn't think it would be release wothy yet. [18:53] its my first manually rolled test [18:53] s/wothy/worthy [18:54] if stuff works its a matter of luck :) [18:55] hmm, but it seems to work [18:56] bah [18:56] jasper is removed but oem-config didnt re-roll the initramfs [18:56] hrm. [18:56] so it will try to resize on next boot again :( [18:57] I didn't think oem-config rerolled the initrd. [18:57] cjwatson said it would [18:58] well, sudo apt-get purge jasper && sudo update-initramfs -u [18:58] Might need an extra parameter. [18:58] and then the mkimage stuff and copying uInitrd to the first partition [18:59] finished downloading, extracting and dd'ng now. [19:00] good luck :) [19:00] * ogra is gone [19:00] g'night. === asac_ is now known as asac [21:19] jcrigby: any luck yet with linkat? [21:20] jcrigby: it's looking to me like a regression in gnulib's linkat replacement, which was only introduced to the coreutils package in the merge of the latest upstream release; but I have no clue why it's not showing up on all archs [21:24] ciao a tutti...c'รจ qualche anima pia che mi potrebbe consigliare qualche fonte per poter installare ubuntu su processore arm? [21:26] sorry...there is someone who can advise me about some source of information for installing ubuntu on arm? [21:32] bullet9mm: http://releases.ubuntu.com/10.04/ provides images for Dove an i.MX51 systems; http://cdimage.ubuntu.com/ports/releases/10.04/release/ has images for OMAP3 systems (e.g., the Beagleboard); https://wiki.ubuntu.com/ARM is the landing page for ARM on Ubuntu information in general [21:32] <- anima pia [21:35] slangasek .....grazie...era proprio quello che stavo cercando... [21:36] you're welcome :) [21:37] bullet9mm: What platform are you interested in specifically? [21:40] slangasek: still working on it, not much progress yet [21:41] I did notice a linkat fix in ltp. Something about not doing the test in a tmpfs mount because you get a different error for one of the tests [21:43] Also once my build did finish the resulting binary seemed to work for me on qemu [21:43] jcrigby: ok. Things I've learned so far: REPLACE_LINKAT is being set but isn't used when linking the test suite; LINK_FOLLOWS_SYMLINK is detected as '0' in the build env, which is the condition that causes the failing assert to be tested; and these two things are related [21:43] jcrigby: oh, interesting - I was testing on tmpfs here as well, maybe I should test on ext3 [21:43] (my test on NFS+gssapi failed miserably :) [21:44] ahh [21:44] it was definitely reproducible for me on tmpfs [21:48] hey slangasek [21:48] * slangasek waves [21:48] kblin: how goes? :) [21:48] fine, busy with work, but then who isn't :) [22:34] sorry for the delay...I'm watching for beagleboard and igepv3 board... [22:34] GrueMaster [22:35] Not sure about the igepv3 board, but we have detailed instructions for 10.04 on beagle at https://wiki.ubuntu.com/ARM/Beagle. [22:38] yes...I've watched it 20 minute ago...cool doc... :) [22:38] *minutes [22:39] Maverick images (10.10) will be live from the start. No installation. (we will still have a netboot install image). [22:44] I've found this... [22:44] http://elinux.org/BeagleBoard [22:44] it's ok? [22:47] I guess. It is more of a general info page for beagleboards. It has a lot of great info, but their Ubuntu Lucid info is a bit weak. [22:49] mmm.. [22:49] It doesn't even mention our live netbook image. [22:49] Other than that minor detail, the info looks good. [22:50] got it...some impression about beagleboard+lucid? [22:52] It works. Main issue is memory limitations on the beagleboard, and there are some apps that have issues, mainly mono based apps. [22:53] Also, see https://wiki.ubuntu.com/ARM/LucidReleaseNotes [22:54] are you talking about the RAM? [22:55] yes. The beagleboard only has 256M. The XM has 512, which should help with performance. [22:56] yes...but it isn't on sale...still... [22:58] ...think wait for the XM version... [22:58] The XM is on sale. Just out of stock (I think). [22:59] is it, now? I heard there were production delays [23:01] yes...I have asked into the beagleboard's irc chan and they responded me that it isn't on sale... [23:02] Ah. beagleboard.org hasn't been updated then. === bjf is now known as bjf[afk] [23:03] @.@ [23:04] then xm are for sale but out of stock? [23:05] *on sale [23:06] jcrigby: same failure on ext3 [23:07] jcrigby: beagleboard, in all cases; if you're in qemu, maybe the x86ness of the env is showing through and leading you astray :/ [23:13] perhaps [23:16] GrueMaster, bullet9mm, afaik the xm isn't available yet [23:17] That's what was said earlier. Any idea as to eta? [23:17] nope [23:17] "soon" [23:18] :( [23:19] I'm disappointed they dropped the sd reader; you can't get good microsd cards in my experience [23:21] Well, Amazon has 2 16G class 10 microSD cards, but they are almost $100 US. [23:21] that doesn't tell you anything useful though [23:21] Too bad they aren't using regular SD instead of microSD. [23:21] you need to know if it's slc flash, and you need to know the wear-levelling, or at least a firmware revision so that you can know that your testing is meaningful [23:25] gerald and ti are saying early-mid july for shipments in quanity... [23:27] nice :) [23:28] ...do you think that xm will able to support kde4? [23:28] ah.. my 'Bx' rand kde 2 years ago... [23:29] it just needs lots of swap, use an external usb drive and don't bother with a slow mmc card... [23:29] 512mb should be sufficient to run it without swap though [23:30] yes...i think too... [23:31] if you have a igepv_2 laying around, just give that a try. (it has 512Mb). (and image the cpu speed bumped from 720 to 1Ghz) [23:32] hmm, there's a 2gb microsd with slc flash :/ [23:33] in fact they are undecided between XM and beagleboard igepv2 ... any advice? [23:34] sorry...I'm undecided [23:34] well the beagleboard and igepv2 are available now.. xm in july... [23:34] bullet9mm, igepv2 is actually intended for industrial/commercial purposes, the beagle doesn't have any such guarentees [23:35] the beagles is only 'guarented' to be a dev board, anything you do beyond that is up to your testing.. ;) [23:35] exactly [23:35] mmm...and igev2 have bluetooth and wifi modules embedded... [23:35] i keep hearing rumors of a igev3.. is that comming out this summer? [23:36] oooo, a 4gb microsd with a manufacturer quoted write endurance of 2 million read/write/erase cycles [23:37] that would make the xm (and the igep) doable for me [23:37] and I can get them through bluestar [23:38] oh, goodie [23:38] * cwillu_at_work does a happy dance [23:38] do they have a written guarntee? i know your going to test it.. (on the bright side, i believe the xm is coming with a micro-sd card with angstrom (not 100%) i'll probally be an el cheapo micro-sd0 [23:38] yep [23:38] yep to the el cheapo, that was [23:38] http://www.atpinc.com/p2-4a.php?sn=00000391 [23:38] they quote that value in the specs sheet [23:39] I'll have to wait until tomorrow to find prices though :/ [23:39] any idea about an lcd touch for igepv2?..the beagle lcd it's ok? [23:40] nope, sorry [23:40] at first sight seems to be... [23:40] O.o [23:41] as in I don't know :p [23:41] http://www.igep.es/images/stories/igepv2_rc_photo2.jpg [23:42] no.. i haven't seen a good lcd solution available for beagles yet.. the interface is their on the cx's.. [23:42] the socket seems the same... [23:42] http://www.liquidware.com/shop/show/BB-BT/BeagleTouch [23:43] that's actually very nice... [23:45] heh: http://www.liquidware.com/shop/show/QCT/QuadCore+Tower+Kit [23:47] niiicee :) [23:47] siiiily :p [23:49] it's a gem...not useful, but a gem... [23:49] :P