[05:48] so, Dove X0 should be good to go WRT default audio settings. I've uploaded both alsa-utils and pulseaudio for that. [09:50] Someone tried the openoffice binary packages? [09:50] Would be nice to confirm the fix and close this awful bug :-) [09:58] lool, did you see my question from yesterday ? [09:59] (why dont we cat the partition/mbr files together butu do complex math and an additional image instead ?) [09:59] *but [10:02] ogra: You mean why don't we release the individual bits? [10:02] instead of an SD card image? [10:02] or why don't we have blobs in the debian-cd/cdimage trees? [10:02] no, in our debian-cd scripts we create an empty image, add an mbr, do complex math to dd the contents into it etc [10:03] dd if=/dev/zero of=mbr.img bs=1 count=512 [10:03] parted -s mbr.img mklabel msdos [10:03] cat $IMAGE >>mbr.img [10:03] etc etc [10:03] why dont we do it like that instead and use the filesizes for the values we need [10:03] way less complex [10:04] ogra: Because you don't know which size your partition is actually going to have [10:05] i.e. append each partition image we create to the full image and "mv mbr.img resulting.img" in the end [10:05] We need to create partitions before we know for sure which size they'll have [10:05] sure i do, at least if i can use byte values like parted does [10:05] well, cat'ing definately works here [10:05] You can specify a partition size in bytes, but its size wont be the one you passed... [10:06] hmm [10:07] you mean because of block sizes ? [10:07] vs byte size [10:07] No [10:08] sector size, and in your case you have additional constraints due to chs addressing [10:08] yes, i wasnt talking about my current case though ... just in general [10:08] (we have other constraints in other cases, for instance the FIS needs to have a size which is a multiple of the flash block size 0x2000 IIRC) [10:09] ogra: I don't think there's any easy case [10:09] right, we could still fill up the images with ueros though [10:09] *zeros [10:09] and then glue them together [10:09] ogra: which is what we do by dd-ing them in place [10:09] though it doesnt make it easier ... [10:10] right, the constraints somehow make my point moot :) [10:10] i'm wondering why i dont see any probs with that here though [10:10] Also, tools like parted or other disk utilities don't like creating partitions which go beyond the device boundaries [10:10] but probably because i dont read to the end of the partition [10:10] so we'd have to create the partitions with the data in place already anyway [10:11] It's just becaue your current file sizes happen to work and you're not trying to write at the end of partitions [10:11] right [10:11] thanks for clearifying [10:12] I think the current approach is proven solid, even if it's a bit complex; I went through great length to put a bunch of sanity checks in place which prevent us from doing anything nasty, and I think it's the right way to do it [10:12] sadly its only solid on the build server, as soon as i run it locally i get size probs [10:12] However you'll have to be creative because parted doesn't seem to allow enforcing the chs geometry [10:12] i'm not sure why [10:12] well, sfdisk does allow it [10:13] so i'll likely resort to fdisk or sfdisk [10:13] I would prefer if we'd keep their usage minimal though [10:13] or worst care just rely on u-boot being in NAND ... [10:14] *case [10:14] for many reasons: the checks are currently written using parted which is the canonical tool for partitioning which we use in other circunstances, I would prefer if we kept the number of tools to a minimum, and this puts higher constraints on the image building host and developer machines building images [10:14] sure, but partwed doesnt give mee the functionallity i need [10:14] i need something that works ... *now* [10:15] so my options are small [10:17] I know, but for instance you could just create one partition with fdisk and use parted for the rest [10:17] that would allow keeping the same logic everywhere else [10:17] Such as partition size checks [10:17] Cause parted can actually read the partition table just fine, just not create it [10:18] Or you could just dd the partition info just like we dd the partition type for redboot fis (non-fs data) [10:18] indeed [10:18] i didnt plan to use fdisk all over the place :) [10:32] ha ! [10:32] that worked [10:33] heh, intresting, the SD i dd the image to boots fine, but my x86 machine cant mount the SD anymore [11:39] morning [11:52] morning === jamie is now known as Guest31141 [13:41] ogra: yes, two reasons why you won't see anything on the display on omap - I'm still merging the DSS2 stack from 2.6.34-rc2 back into our tree and the drivers are modules. [13:41] ah, sad [13:41] i was hoping to have a live image ready for next week [13:41] i also see that compcache only assigns 60M [13:42] i think we should raise that [13:42] ogra: try the kernel on people.canonical.com:~/amitk/ti/zImage [13:43] does that have the modules builtin ? [13:43] ogra: yes, its the first compile after porting the dss2 stack back and compiled-in [13:43] 51 patches [13:44] * ogra downloads [13:44] i hope the modules in initramfs will still work [13:46] ogra: no guarantees, I can start compile for a temp debian package if this improves things [13:46] [ 35.326751] FAT: codepage cp437 not found [13:46] [ 36.601043] FAT: codepage cp437 not found [13:46] over and over [13:47] and nothing on screen [13:47] *sigh* [13:48] I'll compile a debian package [13:48] ogra: or perhaps you could try w/o the initramfs [13:48] well, it should show any output before it goes to initramfs [13:49] but the screen just shuts down [13:49] so *something* in the DSS2 stack is working :) [13:50] mount: mounting /dev/mmcblk0p1 on /cdrom failed: Invalid argument [13:50] modprobe: FATAL: Could not load [ 67.495635] FAT: codepage cp437 not found [13:50] /lib/modules/2.6.33/modules.dep: No such file or directory [13:50] well, it behaves like with the archive kernel [13:51] ogra: that could be missing udebs [13:52] on a live image ? [13:52] hmm, no [13:53] its missing modules that are in your build but not in the casper initramfs from the archive kernel [13:58] ogra: Do you have an initramfs with nls_cp437.ko in it? [13:59] ogra: It might be worth checking whether we put any codepage in the kernel for kernels which are used with FAT images [13:59] lool, apparently not [13:59] well, i do, but amitk's kernel looks for a different modules dir [13:59] ogra: new 500.3 debs on people [14:00] hmm, thanks though it will be tricky to build a livefs with them === JaMa is now known as JaMaOff [14:01] ogra: this is a problem with the casper initramfs, right? [14:02] well, it would also be with any other initramfs [14:02] /lib/modules/2.6.33/modules.dep: No such file or directory [14:02] (initramfs) ls /lib/modules/ [14:02] 2.6.33-500-omap [14:36] amitk, :( [14:36] amitk, http://paste.ubuntu.com/401821/ [14:39] amitk, doesnt look good [14:43] ogra: archive kernel? [14:53] amitk, nope [14:53] http://people.canonical.com/~amitk/ti/linux-image-2.6.33-500-omap_2.6.33-500.3_armel.deb [14:54] do you have unionfs in that package ? [14:55] ogra: yes [14:55] * ogra tries that instead [14:55] urgh [14:56] casper dropped unionfs support [14:56] only unionfs-fuse is supported [14:56] * ogra curses [14:56] we're trouble for M :) [14:56] in trouble [14:56] oh, no, i just didnt see it [14:56] its there [14:57] i thought aufs was about to go upstream [14:58] ogra: aufs was rejected from upstream a year ago [14:58] forever [14:58] upstream is working on a union mount solution, but it isn't ready yet [15:02] ah, still [15:02] i thought that was dropped in favour of including aufs [15:03] mount: mounting unionfs on /root failed: No such device [15:03] unionfs mount failed [15:03] (initramfs) cat /proc/modules |grep union [15:03] (initramfs) [15:03] hmm [15:03] 2.6.33 has now changed API so that aufs doesn't even compile w/o a hack [15:04] i dont have a unionfs module :/ [15:04] (initramfs) ls /lib/modules/2.6.33-500-omap/kernel/fs/ [15:04] configfs nls squashfs nfs_common reiserfs exportfs [15:04] nfs lockd udf jfs isofs xfs [15:04] * ogra checks the chroot he rolled the initrd in [15:05] nope [15:05] hmm, shouldnt that be in the ubuntu subdir ? seems the whole dir is missing [15:06] ah, no, its there [15:06] but only compcache in it [15:06] ogra: sigh, sorry, I forgot to add that commit in this kernel [15:06] it is on another branch of my git tree [15:07] ah [15:09] ogra: so what happens with the archive kernel? [15:09] wrt aufs [15:09] hrm, i didnt have a squashfs ready when i tried it last, one sec, i'll test again [15:14] amitk, same breakage [15:15] bug 417009 [15:15] Launchpad bug 417009 in openoffice.org (Ubuntu Karmic) (and 4 other projects) "all openoffice apps die in 'com::sun::star::ucb::InteractiveAugmentedIOException' on armel in karmic (affects: 1)" [Low,In progress] https://launchpad.net/bugs/417009 [15:15] bug 537356 [15:15] Launchpad bug 537356 in webservice-office-zoho (Ubuntu Lucid) (and 3 other projects) "application menu entries dont do anything (affects: 2)" [High,Triaged] https://launchpad.net/bugs/537356 [15:15] ogra: ok, i'll look at the oops [15:15] ogra: could you file the bug please? [15:17] hrm, the serial console behaves very weird with the archive kernel [15:18] ogra: freezes? [15:19] yeah [15:19] i get the busybox shell but cant type anything [15:19] ogra: huh, never saw that [15:20] and it looks like that http://paste.ubuntu.com/401839/ [15:20] looks like it adds tabs on the go [15:20] I don't see that [15:21] do you have any HW/SW flow control enabled in minicom? [15:21] if so, disable it [15:21] no, and it works fine with your kernel [15:22] no changes in my kernel except new DSS2 patches [15:22] strange [15:31] asac: Hello, Are you there? [15:32] amitk, well, different toolchain [15:37] ogra: huh? [15:37] yours is codesourcery [15:46] bug 519897 [15:46] Launchpad bug 519897 in squid (Ubuntu Lucid) (and 1 other project) "[armel] squid FTBFS: cf_gen Segmentation fault (affects: 1)" [Medium,Fix released] https://launchpad.net/bugs/519897 [15:46] abdalla_: busy atm. ping me in 1h+ [15:47] sorry. [15:47] asac, i think i'll not MIR x-loader and u-boot for now, if we can assume they are preinstalled in NAND and build out images accordingly they can stay in universe for now [15:48] s/out/our/ [15:51] bug 541399 [15:51] ogra: ok [15:51] Launchpad bug 541399 in linux-mvl-dove (Ubuntu) "netboot image fails to boot. (affects: 1)" [Medium,New] https://launchpad.net/bugs/541399 [15:52] bug 537356 [15:52] Launchpad bug 537356 in webservice-office-zoho (Ubuntu Lucid) (and 3 other projects) "application menu entries dont do anything (affects: 2)" [High,Triaged] https://launchpad.net/bugs/537356 [15:52] bug 537083 [15:52] Launchpad bug 537083 in linux-fsl-imx51 (Ubuntu Lucid) (and 1 other project) "Suspend no longer works after updating to 2.6.31-605.9 kernel (affects: 2)" [High,Fix released] https://launchpad.net/bugs/537083 [15:53] bug 537356 [15:53] asac, bug 457878 is actually fixed in git [15:53] Launchpad bug 537356 in webservice-office-zoho (Ubuntu Lucid) (and 3 other projects) "application menu entries dont do anything (affects: 2)" [High,Triaged] https://launchpad.net/bugs/537356 [15:53] Launchpad bug 457878 in linux-fsl-imx51 (Ubuntu Lucid) (and 1 other project) "imx51 on board ethernet plug/unplug events not detected (affects: 2)" [Medium,In progress] https://launchpad.net/bugs/457878 [15:54] (pending upload) [15:54] bug 540477 [15:54] Launchpad bug 540477 in xorg-server (Ubuntu) "X restarted, but no .crash file (dup-of: 540256)" [Undecided,New] https://launchpad.net/bugs/540477 [15:54] Launchpad bug 540256 in upstart (Ubuntu) (and 1 other project) "enter kills X when booting Live CD or w/cryptsetup with plymouth text plugin (affects: 7) (dups: 3)" [Critical,Fix released] https://launchpad.net/bugs/540256 [15:55] bug 515023 [15:55] Launchpad bug 515023 in linux (Ubuntu) "ATA pass-through commands preventing external HDD to be mounted (affects: 10) (dups: 1)" [Undecided,Confirmed] https://launchpad.net/bugs/515023 [15:55] bug 528524 [15:55] Launchpad bug 528524 in totem (Ubuntu Lucid) (and 5 other projects) "Sound not working in all apps on dove (affects: 3)" [High,Invalid] https://launchpad.net/bugs/528524 [16:19] asac, bug 548924 [16:19] Launchpad bug 548924 in linux-ti-omap (Ubuntu Lucid) (and 1 other project) "aufs broken in ti-omap kernel (affects: 1)" [Undecided,New] https://launchpad.net/bugs/548924 [16:19] amitk, want that assigned ? [16:25] ogra: yes please [16:26] done [16:26] amitk, is there a bug about the DSS2 open anywhere ? [16:29] ogra: no, it will go in w/o the bug [16:30] i guessed so, but asac wants something to track status === hrw is now known as hrw|gone === hrw|gone is now known as hrw [18:11] lool, now it works withe the command line from wiki [18:12] lool, with -hdb [18:21] howdy all, so I just obtained and htc tilt 8925 and I am interested in running ubuntu arm on it. suggestions on where I should start? [18:24] awolfson: You can use -hda flag with any filename; are you saying that if you try with -hdb it works, and -hda it doesn't? [18:43] ogra: omap3 allows you to prepare sd card which will flash xloader and uboot into nand on board without them [18:44] thats how I got bug2.0 running from being not programmed at factory [18:46] lool, It did not work for me first time. I will try again === hrw is now known as hrw|gone [19:27] lool, I tried procedure for QEMU from wiki with hda again - it worked that time [19:34] davidm: alive today? [20:28] awolfson: Ok; just bad luck first time I guess [23:31] I got in a tegra board today [23:32] time from getting the board, to booting Karmic (then Lucid..) 1 hour [23:32] can't wait to see the official nVidia installer [23:34] cool Martyn, so what does nVidia give you with the kit? [23:44] rcn-ee : Well, the tegra 250, 1Gb RAM (800MB available after video ram is allocated), WiFi, Bluetooth, 3 USB, 1 debug USB, a serial/debug breakout board, MMC slot, eMMC card slot, SIM card slot, two compact PCI-e (miniPCI) slots, VGA port, HDMI port... [23:45] full SDK available, Windows 6.5 (and supposedly 7) images available, an android image based on android 2.2, and an empty promise to have an Ubuntu image for download [23:45] (you hit the website and find there is no ubuntu image ready yet) [23:46] those specs sound sweet.. they are probally waiting till closer to the release.. i was looking at their dev site, thinking it would be a nice addition to my arm farm.. [23:46] rcn-ee : Apply for one [23:46] The dev boards are one per company, $400 [23:47] useful and quite powerful little thing .. 1Ghz dual core [23:48] ouch, was hoping they'd be less... (my order for a touch book just went thru, so i'll have to wait a month or so) [23:49] first they have to accept your app :) [23:50] true... and they might not particularly like the fact, I'll start benching them against 4 omap boards... [23:51] c*nod* [23:51] have you gotten the omap4430 eval board yet? [23:51] i wish... I've been bugging my TI contacts for the last 6 months for that board.. ;) [23:53] i have one of the early XM proto's so as soon as i get my IGEPv2 installed into my gcc test farm, I'm going to see how far that goes.. (Single Core 800Mhz-1.2Ghz range) [23:57] I got mine three weeks ago, and let me tell you .. the tegra2 runs rings around it [23:59] i bet, dual core, 1Gb of memory and real pheripherals..