**** BEGIN LOGGING AT Fri Apr 16 03:00:02 2010 Apr 16 03:22:50 [ 26.829498] HS USB OTG: no transceiver configured Apr 16 03:22:51 [ 26.834442] musb_hdrc musb_hdrc: musb_init_controller failed with status -19 Apr 16 03:28:39 great.... my asix is broken again. Apr 16 03:28:43 And my musb also is broken. Apr 16 03:29:23 http://pastebin.com/THwXNMEE Apr 16 03:29:31 wow, that's a nice "word". Apr 16 03:29:35 thwxnmee! Apr 16 03:35:03 ah, adding twl4030-usb to initramfs ahead of musb_hdrc fixed it. Apr 16 03:36:54 is there a way to retrieve/store/copy the contents of a flash drive prior to formatting it? Apr 16 03:37:46 basically ive got this awesome $90 netbook that comes with windows CE and an ARM processor but im essentially looking to backup the current OS and replace with something else like ubuntu arm, angstrom, or some other streamlined linux OS.. any general recommendations? Apr 16 03:41:00 You may be able to copy the contents of the MTD stuff using dd. Apr 16 03:41:11 What netbook is this? Apr 16 03:42:28 this is an EPC-702 Apr 16 03:42:53 im not familiar with MTD or dd sorry im a supreme newbie to ARM and this sort of stuff :( im technically savvy just not with this (yet) Apr 16 03:43:10 MTD is "memory technology device" Apr 16 03:43:15 such as NAND flash. Apr 16 03:43:28 gotcha yeah this is a 2gb nand flash utilizing XIP Apr 16 03:45:17 so far i've got "F1 to update the system" at boot, asks for password, its ztk, then it gives me the options to format nand disk, format xip disk, format flash2 disk, update xip, update eboot Apr 16 03:57:28 hmm... any option to boot from an sd card? Apr 16 03:57:46 Or to check what's in the disk? Apr 16 03:58:04 hmm, there's both nand and flash2? Apr 16 03:58:13 xip is the kernel, most likely. Apr 16 03:58:18 eboot... not sure. Apr 16 04:01:19 sounds like the winCE boot loader. Apr 16 04:02:54 styol: its likely someone has to port the linux kernel to that board or device before you'll have any luck running on it Apr 16 04:03:05 http://www.experts-exchange.com/OS/Microsoft_Operating_Systems/Q_25366423.html -- is this you? Apr 16 04:03:27 persia: armeb is dead in Debian and was never an official port. If we need an EABI big endian port, it will be armeb Apr 16 04:03:39 thats not me but thats where i got the password from hehe Apr 16 04:03:53 my friend has an ad blocker and apparently it makes the answers appear, nothing particularly helpful Apr 16 04:04:19 DanaG yeah it has SD reader.. not sure why theres both nand and flash2 Apr 16 04:06:00 hmm, unfortunately, there seems to be very little info on that EPC-701 thingy. Apr 16 04:06:01 how can i find out more information about this board? this is about all i have to go off of CPU: AK7802 and this at boot - http://img.vishun.com/IMG_0759.JPG Apr 16 04:06:06 Not even a spec of what ARM it is. Apr 16 04:06:12 702 not sure if it makes a difference Apr 16 04:06:43 a friend claims that its the same processor that the motorola razr v3m uses Apr 16 04:06:57 http://en.wikipedia.org/wiki/ARM_architecture Apr 16 04:07:03 arm926ej-s Apr 16 04:07:06 search for that. Apr 16 04:07:43 hmm, is that "versatile"? Apr 16 04:08:55 not sure.. im a bit confused ;) Apr 16 04:11:22 somebody more knowledgeable about ARM stuff would have to answer that. Apr 16 04:11:33 interesting... my Cowon S9 has a Telechips 7901 in it. Apr 16 04:11:41 thanks very much for your help DanaG, its much appreciated Apr 16 04:12:54 hmm, you could also ask in #hardware, perhaps. Apr 16 04:13:33 ill give it a whirl for sure Apr 16 04:15:26 styol: its unlikely you'll be able to run Linux on that device. Unlike most other architectures, ARM doesn't have a standized mechanism of bringup, so the early system components through to the kernel must be tailored for each device Apr 16 04:17:01 NCommander ah ok.. hrm yeah it does seem to be something that might be a bit of a stretch.. curious, how do devices that run ubuntu arm normally get ubutu arm installed? Apr 16 04:18:04 found one thing thats promising.. http://www.arm.com/community/software-enablement/linux.php Apr 16 04:20:44 styol: on paper, there are two ways Apr 16 04:21:20 1. A vendor works with Canonical and Canonical adds the support to Ubuntu Apr 16 04:21:57 2. Community members take Ubuntu/ARM and port it themselves (i.e., beagle community). Due to optimizations used in ubuntu, only a small subset of devices can run Ubuntu/ARM sadly Apr 16 04:22:51 yeah that makes sense Apr 16 04:23:25 styol: looking at your device, Ubuntu's specs are too high (your chip only implemented ARMv5 it seems. Ubuntu on AMR needs Ubuntu ARMv7+Thumb2) Apr 16 04:23:45 s/AMR/ARM Apr 16 04:24:11 ARMv5TEJ yeah? Apr 16 04:24:32 how about something like Angstrom, or ttylinux Apr 16 04:24:52 http://www.angstrom-distribution.org/ http://www.minimalinux.org/ttylinux/downloadARM.html Apr 16 04:25:49 styol: well you still need kernel support before that's going to work unfortunately. Apr 16 04:26:05 kernel support by one of those systems? Apr 16 04:26:30 how about http://www.arm.linux.org.uk/ Apr 16 04:26:43 styol: the Linux kernel is a special type of application that sits inbetween the raw hardware, and the userland (aka, the set of applications that the user interacts with it) Apr 16 04:28:07 how can i determine what this has presently? Apr 16 04:28:52 the first arm.com has an image you can download that says the kernel is supported for Apr 16 04:28:55 styol: if its running Windows CE, then its running the Windows CE client :-) Apr 16 04:29:01 ARM926EJ-S (http://infocenter.arm.com/help/topic/com.arm.doc.ddi0184-) Apr 16 04:29:17 styol: its not that simple. Even if the SoC (or processor) is supported, the board as a whole must also be supported Apr 16 04:31:20 ARGH -- stupid asix. Apr 16 04:31:26 It's giving me ZERO packets. Apr 16 04:31:53 wow, and now it randomly came up. Apr 16 04:32:10 nope, it's just seeing its own outgoing arp. Apr 16 04:33:37 I see this looping on usbmon1: Apr 16 04:33:38 7.039795 4.1 -> host USB URB_INTERRUPT Apr 16 04:33:38 7.039825 host -> 4.1 USB URB_INTERRUPT Apr 16 04:36:19 It loops, and never goes anywahere. Apr 16 04:36:34 Even does it with no ethernet cable plugged in. Apr 16 04:37:36 Upon plugging in a different usb-ethernet thingy, I see usb URB_CONTROL Apr 16 04:42:51 I suppose I should ask on the mailing-list about that. Apr 16 08:47:35 Do you know how to control the screen (tty / fb) blanking interval? After a given time the LCD goes blank Apr 16 09:06:26 nosse1: setterm -blank 0 -powerdown 0 ? Apr 16 09:28:43 amitk: thanks Apr 16 11:49:08 hmmm ... so ffox debug build from trunk doesnt have frame/scrollbar issue for me Apr 16 13:02:37 NCommander: hi there Apr 16 13:03:41 hi Apr 16 13:05:57 rcn-ee, meh, your initramfs patch uses bashisms ... Apr 16 13:13:39 asac, can you put bug 563618 on the release meeting agenda ? it hits us hard on beagle Apr 16 13:13:42 Launchpad bug 563618 in util-linux (Ubuntu Lucid) (and 3 other projects) "Ignoring a broken clock results in infinite reboots; not ignoring results in fsck failure; no solution to this problem (affects: 1) (heat: 14)" [High,Triaged] https://launchpad.net/bugs/563618 Apr 16 13:17:23 hi blotr Apr 16 13:19:17 hey hrw Apr 16 13:58:11 ogra: seems the initramfs approach looks quite ok Apr 16 13:58:18 is that tested? Apr 16 13:58:22 not yet Apr 16 13:58:37 i'm still in rootstock tests that eats all my resources Apr 16 14:00:02 asac, also that not all of it, there needs to be a hook to get all the binaries into the initramfs Apr 16 14:00:20 i'll try to improve and test it today Apr 16 14:03:21 A couple of comments... Apr 16 14:04:10 You assume that the argument to root= is a UUID. For systems set up by Ubiquity that's probably not a problem, but embedded devs will often use the device name instead when tweaking the boot config by hand Apr 16 14:04:38 embedded devs will probably not even use an initramfs Apr 16 14:04:56 Does that work these days? I remember it being very unreliable in karmic days Apr 16 14:05:04 but yes, i see your point ... the fix i work on is rather for the current official omap image though Apr 16 14:05:17 it got a lot better Apr 16 14:05:35 no-initramfs works fine, but requires a bit of tweaking to boot (and is less safe as a generic model) Apr 16 14:05:43 and i think the approach is that you *can* run your system without initramfs if youwant in certain cases Apr 16 14:05:49 right Apr 16 14:05:59 it will always stay for the std distro Apr 16 14:05:59 Also, the hwclock backend often doesn't work during kernel porting I remember having a lot of trouble with this earlier during iMX51 support Apr 16 14:06:24 but if people have a selected set of packages and cant install harmful stuff its pretty safe to go without one Apr 16 14:06:45 i'm not sure if fsck even looks at the hwclock Apr 16 14:06:54 it might suffice to use date --set Apr 16 14:07:01 No, but hwclock --hctosys is used to set the soft clock Apr 16 14:07:09 Maybe do both? Apr 16 14:07:29 yeah, i'll discover such stuff when i test :) Apr 16 14:07:32 It's not feasilble to have both online-updates and can't-install-harmful-stuff Apr 16 14:08:02 persia, it is if you use your own archive Apr 16 14:08:03 We can only do hard-to-install-harmful-stuff Apr 16 14:08:14 ogra, anyway, looks like a reasonable workaround Apr 16 14:08:14 ogra: No it isn't, because of DNS spoofing. Apr 16 14:08:17 and restrict to it indeed Apr 16 14:08:22 * dmart still thinks e2fsck is wrong, but that's another discussion... Apr 16 14:09:00 dmart, i agree though i also agree with Keybuk who says it needs to be solved on the actual filesystem layer Apr 16 14:09:15 i guess we need to fix both Apr 16 14:09:30 sure, it needs more thinking about Apr 16 14:09:35 Gets harder for ext3/ext4 than ext2, but yeah. Apr 16 14:10:16 But if we rely on the reliability of the RTC _at all_ for preserving low-level fs consistency, that sounds really scary, especially for dual boot configs where something else may interfere with the clock sometimes Apr 16 14:10:26 lool, writeback hangs too (just reached iso-codes in rootstock this second) Apr 16 14:12:42 dmart: Without reliance on the RTC, how can we know the time? Apr 16 14:13:27 If you need to know the time, you are at risk. I need to understand more about the way the fs works though, in order to understand whether knowledge if the time is actually needed to ensure consistency Apr 16 14:14:11 It's worth considering that the journal has already been replayed before we get into userspace, so if the RTC really mattered, the fs has already been corrupted. Apr 16 14:14:23 I don't know the deep details, but my limited understanding is that the journalling is timestamped. Apr 16 14:15:04 Well, except in the initramfs case, in which case we're in userspace without actually using the filesystems. Apr 16 14:15:23 (or rather, without using on-secondary-storage filesystems) Apr 16 14:15:31 fair point Apr 16 14:15:39 afaict, serial numbers (not timestamps) are used in the on-disk structures in jbd{,2}.h, but I confess I don't really understand how it all works yet... Apr 16 14:15:51 heh, yeah. Apr 16 14:16:20 Would probably benefit from a discussion with the fs devs at some conference. Apr 16 14:16:53 i dont think the journal has been touched until you get to mountall ... and it wont be touched until mountall remounts in RW mode (note that we specify ro on the cmdline by default) Apr 16 14:17:11 ogra: The journal has to have been replayed to be able to mount RO. Apr 16 14:17:23 Otherwise you have an inconsistent filesystem state. Apr 16 14:17:24 Mouting a dirty journalled filesystem ro does replay the journal Apr 16 14:17:24 persia, replayed but not changed Apr 16 14:17:35 right. Apr 16 14:17:40 there wont be changes on disk yet Apr 16 14:17:53 But the issue we have is at fsck time, not at mount time. Apr 16 14:18:05 Well, except for refuse-to-mount-because-of-fs-age Apr 16 14:18:08 Not sure about that -- the kernel prints out "temporarily mounting fs read-write to recover the journal" (or something like that( Apr 16 14:18:12 (but that's just a sanity check) Apr 16 14:18:24 dmart, aww, evil Apr 16 14:18:38 dmart: Ooh, that makes lots of sense, and completely changes the semantics :) Apr 16 14:18:52 maybe that just happens when mounting the root fs... ? dunno Apr 16 14:19:13 easy to test Apr 16 14:20:05 I guess you need to replay or unwind the journal to mount, because the metadata might be inconsistent garbage until you do that Apr 16 14:20:24 ah, yeah Apr 16 14:20:34 That makes sense. So we need to be sure we're somewhere sane. Apr 16 14:20:46 Unless you're using ext2 (it's still garbage, but you just hope the system doesn't explode ;) ) Apr 16 14:21:00 No, if we happen to have a reliable, working, RTC, I expect that having a date in the future is a good indication of likely corruption. Apr 16 14:21:12 s/No/Now/ Apr 16 14:21:54 I think 90% of the times I've triggered that problem it's been the RTC, not the fs, which is corrupt. But it depends on what you're trying to do. "Normal" users shouldn't hit that so much. Apr 16 14:22:51 wow, neither hwclock, nor date nor dumpe2fs are in the initramfs Apr 16 14:23:02 argh Apr 16 14:23:07 i would have expected the latter to be missing Apr 16 14:23:10 settimeofday? Apr 16 14:23:15 but surely not the first two Apr 16 14:23:16 RIght. The tricky bit is to come up with something that works for both end-users and bring-up-developers Apr 16 14:23:36 nah, no issue, i'll write a hook that copies them Apr 16 14:24:05 i just didnt expect date to miss ... Apr 16 14:24:10 nor hwclock Apr 16 14:24:15 Is the "last mount time" recorded in the fs the time the filesystem was last mounted, or the time it was unmounted? Apr 16 14:24:35 both i think Apr 16 14:24:41 Really we would want to set the clock to the last unmount time (except where there was a crash the the fs wasn't unmounted) Apr 16 14:24:42 last mount action Apr 16 14:24:56 hm Apr 16 14:25:17 How is "mount action" defined? Is that mount+unmount? Apr 16 14:25:23 i guess so Apr 16 14:25:31 (i was also guessing above) Apr 16 14:25:41 there is only that value in dumpe2fs Apr 16 14:25:44 That sounds like desired behaviour. Now I wonder if that's how it's implemented. Apr 16 14:25:58 The ext3 superblock has "mount time" and "write time". Dunno their behaviour though Apr 16 14:26:16 yeah, i see the same in dumpe2fs Apr 16 14:26:29 Does anyone know a filesystem expert? Apr 16 14:26:38 i guess the stamp is updated by mount/umount Apr 16 14:26:47 Keybuk is probably closest Apr 16 14:34:41 It could still be worth discussing it at UDS Apr 16 14:35:44 ogra: session name? booting clockless systems with journaling FSes? Apr 16 14:36:00 asac, RTCless Apr 16 14:36:00 BootingWithTimeZero ;) Apr 16 14:36:03 but yes Apr 16 14:36:19 or batteryless :) Apr 16 14:36:22 ok i will add that Apr 16 14:36:22 asac: not clockless please, use non-rtc or something Apr 16 14:36:33 amitk: Booting With Time Zero ;) Apr 16 14:36:33 our prob is that we *have a clock* Apr 16 14:36:44 whatever causes this condition is probably irrelevant ;) Apr 16 14:36:52 dmart: Keybuk suggested a discussion at UDS would be fruitless unless we had the fs developers there (and it may be late for them to make travel plans) Apr 16 14:37:09 persia, i disagree Apr 16 14:37:23 Fair. Just sharing. Apr 16 14:37:26 hi! about gcc-4.4 in karmic, which is current? 4.4.1 (--with-arch=armv6) or 4.4.3 (--with-arch=armv7-a)? Apr 16 14:37:54 while i agree for the superduper proper fix that might be needed it isnt for finding the cleanest solution that doesnt involve filesystems Apr 16 14:38:23 since the filesystem fix might take years Apr 16 14:38:39 zumbi: According to https://launchpad.net/ubuntu/+source/gcc-4.4 4.4.1 and it ought be ARMv6, as the rest of karmic is supposed to be that. Apr 16 14:39:16 uhm.. ok thanks persia Apr 16 14:39:21 ogra: Oh, a session on workarounds could work, as long as everyone understands it's workaounds. Apr 16 14:39:31 right Apr 16 14:39:44 the filesystem fix will likely require a design change Apr 16 14:39:45 so Lucid is armv7-a :) Apr 16 14:39:55 ogra: dmart: let me know if you think you can lead a session on that booting zero time stuff Apr 16 14:40:04 otherwise i will note it down as a blueprint without session Apr 16 14:40:05 asac, i'm happy to Apr 16 14:40:11 zumbi: Mostly. Only ~7000 packages have been recompiled, so there's still bundles that may be ARMv6 or even ARMv5. Apr 16 14:40:22 ogra: but do you have all the details so we can achieve something? Apr 16 14:40:32 asac, though i dont know how many sessions i have assigned at all Apr 16 14:40:49 might be a lot or none ... nobody tells me Apr 16 14:40:55 zumbi: But I'd say that the majority of packages that generate ELF binaries are likely to have been recompiled. Apr 16 14:41:00 ogra: you had two ... iirc Apr 16 14:41:06 asac, well, i know the problem and the possible areas for fixing Apr 16 14:41:09 i avent added them to the table/schedule yet. but will do Apr 16 14:41:39 if you want to drop one of the suggested ones let me know Apr 16 14:43:08 nah, all fine Apr 16 14:43:31 if its not 50 that i dont know about yet there is nothing to discuss :) Apr 16 14:50:25 bug 559295 Apr 16 14:50:28 Launchpad bug 559295 in flash-kernel (Ubuntu Lucid) (and 1 other project) "flash-kernel-installer needs to learn to handle omap (dup-of: 562888)" [High,New] https://launchpad.net/bugs/559295 Apr 16 14:50:29 Launchpad bug 562888 in ubiquity (Ubuntu Lucid) (and 3 other projects) "omap installation failed with unrecoverable error (affects: 2) (dups: 2)" [Medium,Invalid] https://launchpad.net/bugs/562888 Apr 16 14:50:51 bug 559297 Apr 16 14:50:54 Launchpad bug 559297 in flash-kernel (Ubuntu Lucid) (and 1 other project) "flash-kernel needs to learn to handle omap (dup-of: 562888)" [High,New] https://launchpad.net/bugs/559297 Apr 16 14:50:58 bug 559301 Apr 16 14:51:00 Launchpad bug 559301 in partman-uboot (Ubuntu Lucid) (and 1 other project) "partman-uboot needs to handle omap installs (affects: 1) (heat: 8)" [High,New] https://launchpad.net/bugs/559301 Apr 16 14:51:11 bug 563618 Apr 16 14:51:12 Launchpad bug 563618 in util-linux (Ubuntu Lucid) (and 3 other projects) "Ignoring a broken clock results in infinite reboots; not ignoring results in fsck failure; no solution to this problem (affects: 1) (heat: 14)" [High,Triaged] https://launchpad.net/bugs/563618 Apr 16 14:52:00 ogra: what about partman? is that RC still? Apr 16 14:52:12 asac, invalid Apr 16 14:52:21 can yo do that? Apr 16 14:52:37 done Apr 16 14:52:44 ogra: both flash-kernel things too= Apr 16 14:52:45 ? Apr 16 14:52:47 e.g. see above Apr 16 14:52:54 bug 559144 Apr 16 14:52:58 they are duped to the ubiquity one Apr 16 14:52:59 Launchpad bug 559144 in udisks (Ubuntu Lucid) (and 1 other project) "[armel] udisks-part-id crashed with signal 7 in memcpy() (dup-of: 561426)" [High,New] https://launchpad.net/bugs/559144 Apr 16 14:53:00 Launchpad bug 561426 in parted (Ubuntu Lucid) (and 1 other project) "partman dies when trying to detect disks due to kernel error (affects: 1) (dups: 1)" [High,Fix released] https://launchpad.net/bugs/561426 Apr 16 14:53:12 ogra: so all invalid for flash-kernel-* ? Apr 16 14:53:29 bug 541030 Apr 16 14:53:32 Launchpad bug 541030 in linux-ti-omap (Ubuntu Lucid) (and 1 other project) "omap kernel musb/ehci ports not enabled on beagleboard (affects: 1)" [High,Fix released] https://launchpad.net/bugs/541030 Apr 16 14:53:44 ogra: so only omap issue is bug 563618 ? Apr 16 14:53:45 Launchpad bug 563618 in util-linux (Ubuntu Lucid) (and 3 other projects) "Ignoring a broken clock results in infinite reboots; not ignoring results in fsck failure; no solution to this problem (affects: 1) (heat: 14)" [High,Triaged] https://launchpad.net/bugs/563618 Apr 16 14:53:46 asac, nope, the flash-kernel task on bug 562888 is fix released already ... ubiquity was just uploaded Apr 16 14:53:47 Launchpad bug 562888 in ubiquity (Ubuntu Lucid) (and 3 other projects) "omap installation failed with unrecoverable error (affects: 2) (dups: 2)" [Medium,Invalid] https://launchpad.net/bugs/562888 Apr 16 14:53:55 bug 541399 Apr 16 14:54:01 Launchpad bug 541399 in linux-mvl-dove (Ubuntu) (and 1 other project) "netboot image fails to boot. (affects: 1)" [Undecided,Invalid] https://launchpad.net/bugs/541399 Apr 16 14:54:08 hmm Apr 16 14:54:14 so that was fix committed ... now its invalid Apr 16 14:54:15 well Apr 16 14:54:17 asac, only *known* omap issue, i need to have images i can test now Apr 16 14:54:23 bug 514257 Apr 16 14:54:25 Launchpad bug 514257 in xine-lib (Ubuntu Lucid) (and 1 other project) "[arm] needs porting to thumb2 (affects: 1) (heat: 8)" [High,Won't fix] https://launchpad.net/bugs/514257 Apr 16 14:54:30 bug 528524 Apr 16 14:54:34 Launchpad bug 528524 in speex (Ubuntu Lucid) (and 5 other projects) "Sound not working in all apps through pulse on arm (affects: 3) (dups: 1)" [High,Fix released] https://launchpad.net/bugs/528524 Apr 16 14:54:45 ogra: we never had a working image? Apr 16 14:54:46 bug 541399 is a mess Apr 16 14:55:12 bug 528524 Apr 16 14:55:13 asac, no, flash-kernel and ubiquity were missing, we never had one that finished the bootloader part Apr 16 14:55:20 yeah Apr 16 14:55:28 but thats not our release goal afaiui Apr 16 14:55:29 so its not RC Apr 16 14:55:36 oh Apr 16 14:55:36 hmm Apr 16 14:55:38 ogra: ok Apr 16 14:55:41 so bootloader missing? Apr 16 14:55:44 asac, Bug #541399 is fix released for d-i (i did the change and upload) Apr 16 14:55:45 Launchpad bug 541399 in linux-mvl-dove (Ubuntu) (and 1 other project) "netboot image fails to boot. (affects: 1)" [Undecided,Invalid] https://launchpad.net/bugs/541399 Apr 16 14:55:53 the bot shows the wrong task Apr 16 14:55:53 ogra: ok. i dont list released bugs anymore Apr 16 14:55:55 ;) Apr 16 14:56:03 bug 532733 Apr 16 14:56:04 Launchpad bug 532733 in qemu-kvm (Ubuntu Lucid) (and 2 other projects) "apt/dpkg in qemu-system-arm hangs if a big task is installed (affects: 4) (dups: 1) (heat: 32)" [High,Incomplete] https://launchpad.net/bugs/532733 Apr 16 14:56:12 bug 528887 Apr 16 14:56:15 Launchpad bug 528887 in maximus (Ubuntu Lucid) (and 2 other projects) "maximus does not give default focus to newly started apps in combination with efl launcher (affects: 3) (heat: 18)" [Medium,Triaged] https://launchpad.net/bugs/528887 Apr 16 14:56:19 bug 555977 Apr 16 14:56:21 Launchpad bug 555977 in openoffice.org (Ubuntu Lucid) (and 1 other project) "openoffice.org FTBFS on armel (affects: 1)" [High,Fix released] https://launchpad.net/bugs/555977 Apr 16 14:56:23 is ooo build now? Apr 16 14:57:08 ogra: please checkout https://wiki.ubuntu.com/MobileTeam/ReleaseStatus/Lucid Apr 16 14:57:12 plars: ^^ Apr 16 14:57:27 i am cros checking with the buglist now Apr 16 14:58:11 looks fine Apr 16 14:58:30 hmm Apr 16 14:58:32 bug 559065 Apr 16 14:58:33 Launchpad bug 559065 in linux-fsl-imx51 (Ubuntu Lucid) (and 1 other project) "ifconfig eth0 down will cause system hang after fec.c driver update (affects: 1) (heat: 8)" [High,In progress] https://launchpad.net/bugs/559065 Apr 16 14:58:34 whats up with that? Apr 16 14:58:36 mountall+fsck and qemu-kvm hang are my most serious ones Apr 16 14:58:36 ogra: ? Apr 16 14:58:42 wasnt that done? Apr 16 14:58:49 i think the description is nonsense Apr 16 14:58:52 plars, ? Apr 16 14:59:38 well, or it should be fix released Apr 16 14:59:39 ogra: I didn't see that with the newest kernel Apr 16 14:59:44 is that RC? .... what does driver update mean? module reloaded? Apr 16 14:59:46 since the described issue is gone Apr 16 14:59:52 usually we dont reload modules Apr 16 14:59:58 ogra: I think that bryan saw that, and thought it may be related to the suspend/resume issues Apr 16 15:00:14 well, but you dont see the hang anymore Apr 16 15:00:14 but his kernels that I tested, that certainly did not have this bug in it, still would not suspend/resume Apr 16 15:00:30 bug 522858 is new? Apr 16 15:00:31 Launchpad bug 522858 in maximus (Ubuntu Lucid) (and 2 other projects) "Sometimes app windows come up undecorated, unselectable, and not full screen (affects: 1) (heat: 6)" [High,New] https://launchpad.net/bugs/522858 Apr 16 15:00:39 asac: the fec driver code was updated in the kernel Apr 16 15:00:56 plars: but how can that change something? Apr 16 15:01:03 without reloading the kernel/module? Apr 16 15:01:21 asac: because we have a new kernel with the new code in it Apr 16 15:01:23 bug 67544 Apr 16 15:01:27 whats that? Apr 16 15:01:33 asac: Bug 67544 on http://launchpad.net/bugs/67544 is private Apr 16 15:01:38 very old apparently... Apr 16 15:01:38 why is such an old bug targetted for us? Apr 16 15:01:40 67544 ? Apr 16 15:01:45 i will drop that Apr 16 15:01:49 are you missing a digit? Apr 16 15:02:11 no Apr 16 15:02:15 plars: no seems doko gave it to us Apr 16 15:02:40 Please don't drop it. Apr 16 15:02:51 looks like persia has commented on it before Apr 16 15:02:57 It's an old bug, but it's very valid. That bug represents pascal support for armel. Apr 16 15:03:01 asac, doko is also fixing it it seems Apr 16 15:03:10 yeah Apr 16 15:03:10 * persia has been watching that bug since before armel was available in Ubuntu. Apr 16 15:03:29 persia: good. i will call you to give info during release meeting when questions come up ;) Apr 16 15:03:39 so be prepared (in 20 minutes i guess) Apr 16 15:04:00 If I'm still awake sure. Ping me first (I'm getting really close to sleep). Apr 16 15:04:23 Basically, some of us can build it, and lamont can't, and nobody knows why. Apr 16 15:05:01 doko did a first-stage bootstrap that is hoped to ease the situation. Apr 16 15:05:38 Oh, I will be awake. The release meeting seems to have moved with DST. Apr 16 15:07:51 persia: right. as i said: about 20 minutes till we are on :) Apr 16 15:14:30 asac: you said all done for porting to thumb2, but there are still a couple of needs porting to thumb2 bugs that are rc targeted Apr 16 15:14:41 plars: those are not armel tagged then Apr 16 15:14:44 i just looked at the list Apr 16 15:14:48 asac: yes they are Apr 16 15:14:56 https://bugs.edge.launchpad.net/ubuntu/lucid/+bugs?field.searchtext=&orderby=-importance&search=Search&field.status%3Alist=NEW&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPLETE_WITHOUT_RESPONSE&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&assignee_option=any&field.assignee=&field.bug_reporter=&field.bug_supervisor=&field.bug_commenter=&field.subscrib Apr 16 15:14:59 asac: https://bugs.launchpad.net/ubuntu/+source/qt4-x11/+bug/514253 Apr 16 15:15:02 Launchpad bug 514253 in qt4-x11 (Ubuntu Lucid) (and 1 other project) "[arm] needs porting to thumb2 (affects: 1) (heat: 10)" [High,Won't fix] Apr 16 15:15:08 asac: https://bugs.launchpad.net/ubuntu/+source/xine-lib/+bug/514257 Apr 16 15:15:09 plars: thats wont fix Apr 16 15:15:13 Launchpad bug 514257 in xine-lib (Ubuntu Lucid) (and 1 other project) "[arm] needs porting to thumb2 (affects: 1) (heat: 8)" [High,Won't fix] Apr 16 15:15:14 plars: wont fix too Apr 16 15:15:14 ah Apr 16 15:15:20 didn't look in the bug, just did a search Apr 16 15:15:29 must have at least one task that isn't won't fix Apr 16 15:15:30 plars: a search that includes won't fix? ;) Apr 16 15:15:38 plars: lucid task is wont fix Apr 16 15:15:43 i probably left the normal task open Apr 16 15:15:46 yeah Apr 16 15:15:47 so yeah Apr 16 15:16:00 asac: was that intentional? just won'tfix for lucid then? Apr 16 15:16:17 I will untarget them then Apr 16 15:16:24 It's a bug in LP that there's no good way to do that :( Apr 16 15:16:26 bug 528524 Apr 16 15:16:27 Launchpad bug 528524 in speex (Ubuntu Lucid) (and 5 other projects) "Sound not working in all apps through pulse on arm (affects: 3) (dups: 1)" [High,Fix released] https://launchpad.net/bugs/528524 Apr 16 15:16:29 no sense in having them targetted if they are wontfix Apr 16 15:16:32 plars: wont fix is the way to untarget Apr 16 15:16:50 hmm, except they are still milestoned Apr 16 15:16:53 plars: thats the way you get a normal task back Apr 16 15:17:03 plars: but they are wont fix ... so they are hidden from all listss ;) Apr 16 15:17:12 unless you search for invalid+wontfix ;) Apr 16 15:17:21 feel free to change them to match your needs though Apr 16 15:17:38 asac: I just did search for tags including armel and milestoned for 10.04, and I did not include invalid or wontfix Apr 16 15:17:55 plars: right. you didnt search for "lucid" Apr 16 15:18:03 you have to go to bugs.launchpad.net/ubuntu/lucid Apr 16 15:18:07 nd then use advanced search there Apr 16 15:18:09 and Apr 16 15:18:15 yeah, I just started from ubuntu Apr 16 15:18:28 feel free to unmilestone ... but if the not targetted task is milestoned its not RC in the book of release team Apr 16 15:18:46 means: opportunity to fix for that milestone Apr 16 15:18:53 or for personal prioritization Apr 16 15:21:02 asac: there is https://bugs.launchpad.net/ubuntu/+source/linux-mvl-dove/+bug/551491 that ericm milestoned for 10.04, but it's not targeted to lucid, I doubt it will happen in time though Apr 16 15:21:04 Launchpad bug 551491 in linux-mvl-dove (Ubuntu) "BSP update for Marvell Dove kernel LSP 5.1.0 (affects: 1) (heat: 8)" [High,In progress] Apr 16 15:21:27 plars: thats not rc ... Apr 16 15:21:31 SRUs etc. Apr 16 15:21:39 we can target and milestone for :later Apr 16 15:21:44 if we want to ensure that gets in Apr 16 15:22:14 * plars would like to confirm ericm's plans with him first, but that's my assumption Apr 16 15:29:23 ogra@ubuntu:~$ date Apr 16 15:29:23 Sat Jan 1 01:01:14 CET 2000 Apr 16 15:29:23 ogra@ubuntu:~$ Apr 16 15:29:24 has anyone tried any flavor of ubuntu on one of these? http://delstar.net/multi-function-devices/ds700-netbook.html Apr 16 15:29:26 wohoo ! Apr 16 15:29:29 works ! Apr 16 15:31:33 no one? Apr 16 15:41:54 * asac out to run some errands Apr 16 15:43:04 sn9, no kernel for that device and 128MB wont be fun Apr 16 15:43:35 i was thinking MID edition or something Apr 16 15:45:03 ogra what hardware you running on? Apr 16 15:45:26 which generation arm is that, anyway? Apr 16 15:45:29 TheWhiteGhost, frescale imx51, marvell dove and beagle Apr 16 15:45:42 recent ubuntu requires ARMv7 Apr 16 15:46:09 And doesn'T include MID anymore (it ended up being too hard to integrate the libgtk stuff). Apr 16 15:46:09 i don't know whether samsung qualifies Apr 16 15:46:47 persia: because of arm? Apr 16 15:47:07 sn9: Not at all. It was about the nature of the hildon-gtk changes. Apr 16 15:47:45 i mean, MID is gone from arm, or gone altogether? Apr 16 15:47:58 gone altogether. flavours are independent of architectues. Apr 16 15:48:13 anything replacing it? Apr 16 15:49:12 Not right now. There's been a lot of work to get LXDE ready, but it didn't quite make it for lucid. It will probably make it for maverick. Apr 16 15:49:23 ko Apr 16 15:49:26 *ok Apr 16 15:49:39 There's also been a lot of work on the Qt/Plasma front, which might hit maverick or might hit maverick+1, depending. Apr 16 15:50:15 The Mer folk did a tremendous job of integrating later releases of hildon, but we just couldn't merge all that back into Ubuntu because of the libgtk issues. Apr 16 15:50:47 And I don't know of anyone chasing that anywhere to make it work (hildon work seems unfashionable these days) Apr 16 15:51:01 so, basically, for lucid, it's netbook edition or bust, but 128MB is too thin for that, right? Apr 16 15:51:21 Or roll-your-own works. Apr 16 15:51:27 right Apr 16 15:51:36 LXDE should mostly just work, but the integration isn't 100% yet (and there aren't any images) Apr 16 15:52:07 The Qt/Plasma stuff isn't ready at all yet, although there's kubuntu netbook which has some of the same technologies (but has the same issues at 128MB) Apr 16 15:52:22 which generation arm are the samsung cpu's? Apr 16 15:52:31 Depends which samsung cpu :) Apr 16 15:52:38 (but none of them have distro kernels) Apr 16 15:52:56 unfortunately, that page did not say, as you saw Apr 16 15:54:19 Oh, that? I heard that device was ARMv5. Apr 16 15:54:28 I may be mistaken though. Apr 16 15:56:57 most of s3c are armv4t, some are armv5te. s3c6410 is armv6 iirc Apr 16 16:02:13 google seems to suggest arm9 Apr 16 16:03:29 sn9: arm9 can be armv4t or armv5te Apr 16 16:03:50 sn9: there are ARM core numbers and ARM ISA (instruction set arch) numbers Apr 16 16:04:06 arm920t is armv4t, arm926ejs is armv5te Apr 16 16:04:17 S3C2450 Apr 16 16:06:26 arm926ejs Apr 16 16:06:50 so no ubuntu Apr 16 16:06:57 past jaunty Apr 16 16:07:08 yep Apr 16 16:07:14 but Debian may work Apr 16 16:07:40 but getting linux working in it is probably more expensive task then buying something linux supported Apr 16 16:07:45 debian is armv6 and above, i thought Apr 16 16:08:04 no, Debian is armv4t Apr 16 16:08:46 i know for a fact debian does not support armv4 Apr 16 16:09:22 it is hard to find armv4 device today Apr 16 16:09:50 920 is not all gone yet Apr 16 16:10:20 i think gentoo is the only thing that supports armv4 Apr 16 16:10:23 gcc 4.4 can generate EABI binaries for armv4.... Apr 16 16:10:35 sn9: OpenEmbedded distros also supports Apr 16 16:10:40 I have to go Apr 16 16:10:41 ok Apr 16 16:21:30 We tend to recommend folks with older HW run Debian though, just because 90% of bugs found in Debian also apply to Ubuntu (as well as the fixes), and using the same development technologies makes it easy to come back once one has newer hardware. Apr 16 16:24:15 interesting how just-released hardware is "old" Apr 16 16:25:50 sn9: Older processors. You can go buy a brand-new device that implements the i486 instruction set if you want, to show a parallel to ia32. Apr 16 16:25:55 That can't run Ubuntu either. Apr 16 16:26:35 looks like there's a forum thread: http://ubuntuforums.org/printthread.php?t=1414364&pp=75 Apr 16 16:26:40 So the age is about the time-since-processor-design, or even time-since-architecture-specification rather than about time-since-manufacture or time-since-purchase Apr 16 16:52:18 ogra, you still there? Apr 16 16:53:01 dmart, yep Apr 16 16:53:09 seen -devel ? Apr 16 16:53:27 Thanks for offering to lead the BootingWithTimeZero thing --- I was offline for a while (honest!) Apr 16 16:53:53 heh Apr 16 16:54:07 have a look at the backlog in -devel :) Apr 16 16:57:09 ogra: Oh you mean the upstreamed proposal to ignore date issues in e2fsck with broken_system_clock Apr 16 16:57:20 Hmmm, does that mean we don't need to bother? Apr 16 16:57:49 well, probably not for maverick Apr 16 16:58:20 but for lucid ... Apr 16 17:00:37 right... I guess we need to pull together the proposed workarounds and choose one Apr 16 17:11:16 hrm, it doesnt work and i dont get why Apr 16 18:08:52 ogra, just an odd ball idea, have you tried bypassing qemu's image files and just using a real disk? i know it's not the best method in the end, but i just add 2nd sata drive to test... ;) Apr 16 19:02:36 laughs, it actually seemed to get farther, right before a kernel panic: [ 1306.403516] Kernel panic - not syncing: Attempted to kill init! current patch (didn't implement the final save to file conversion) http://pastebin.com/FK3TaViG Apr 16 19:29:41 rcn-ee: hey, did you ever get the spidev patch working? Apr 16 19:30:53 nope, sorry andruk, not yet, it's still on my list for this week.. I'm playing with a hacked up rootstock at to the moment.. Apr 16 20:26:08 ~seen davidm Apr 16 20:26:18 hmm guess no bot in the channel Apr 16 20:28:15 !seen davem Apr 16 20:28:16 I have no seen command Apr 16 20:28:19 :D Apr 16 20:28:49 ahh thanks different prefix Apr 16 20:28:53 ~seen davidm Apr 16 20:28:56 !seen davidm Apr 16 20:28:56 I have no seen command Apr 16 20:29:20 * prpplague checks the bot help Apr 16 20:30:24 bummer Apr 16 22:30:25 NCommander after a couple hours i was finally able to determine which cheap device i will need to try accomplish what im trying to do ;) ping me when you're around, id love to share **** ENDING LOGGING AT Sat Apr 17 02:59:56 2010