=== lilstevie is now known as lilstevie|ZNC | ||
topfs2 | Hi so I'm trying to get ubuntu running on the pandaboard. Problem I have is that I don't ahve a monitor atm and trying to get it working so I can login over terminal so I can start doing stuff to it | 04:22 |
---|---|---|
topfs2 | Setup ssh etc. | 04:22 |
topfs2 | at any rate, I use setenv bootargs console=ttyO2,115200n8 serialtty=ttyO2 rw rootwait mem=463M elevator=noop vram=32M root=/dev/mmcblk0p2 as boot.scr (and some other stuff) | 04:23 |
topfs2 | still I only get kernel output, no login and such | 04:23 |
rsalveti | topfs2: to get login you need to create the ttyO2.conf file at /etc/init/ | 04:23 |
rsalveti | like tty1.conf | 04:23 |
rsalveti | but with getty ttyO2 115200 | 04:23 |
topfs2 | oh ok, will try that | 04:24 |
topfs2 | thanks | 04:24 |
rsalveti | then upstart calls getty with the proper console | 04:24 |
rsalveti | for example, change "exec /sbin/getty -8 38400 tty1" to "exec /sbin/getty 115200 ttyO2" | 04:24 |
topfs2 | awesome, I got a login :D | 04:28 |
topfs2 | thanks rsalveti ! | 04:28 |
topfs2 | you don't happen to know the std user and pass? | 04:28 |
rsalveti | topfs2: there's none, because it calls the oem-config and ask you to create the proper user account | 04:37 |
topfs2 | so no way to login? | 04:38 |
topfs2 | without booting with monitor? | 04:38 |
rsalveti | no, you need to hack up the image | 04:38 |
rsalveti | you can remove the root password request, and then create one | 04:38 |
rsalveti | or get inside with chroot, then you can create it with adduser | 04:38 |
topfs2 | hmm, chroot seems best | 04:39 |
topfs2 | I'll try that | 04:39 |
rsalveti | if you need a minimum rootfs you can also create it with rootstock | 04:39 |
rsalveti | and then use it at the second sd card partition | 04:39 |
rsalveti | then you can also request a user to be created during image build time | 04:39 |
topfs2 | Should that be ok to do with panda aswell? | 04:40 |
rsalveti | topfs2: yup | 04:42 |
topfs2 | haha, hand editing /etc/shadow worked wonders ;) | 04:54 |
topfs2 | thanks for the help | 04:54 |
rsalveti | np | 05:02 |
topfs2 | next step, getting usb network. thats always fun | 05:05 |
=== hrw|gone is now known as hrw | ||
hrw|uds | hi | 14:01 |
hrw|uds | I feel that next week will be interesting for me in arm world | 14:33 |
hrw|uds | kde in natty/armel is behind natty/amd64 so my smartbook has x11 problems... will use pandaboard for local rebuilds | 14:34 |
hrw|uds | hi eric | 14:36 |
lag | rsalveti: Thanks | 14:37 |
lag | rsalveti: I'll speak more with jcastellanos to figure out what's going on when he reappears | 14:38 |
=== rgreening_ is now known as rgreening | ||
=== lilstevie|ZNC is now known as lilstevie | ||
=== hrw is now known as hrw|uds | ||
hrw|uds | where plans for work on efikamx smart[top|book] devices will be synchronized/discussed? irc or ubuntu-mobile ML? | 16:11 |
markos_ | hrw|uds, let's start with irc first and see if we have needs for a specific list or if an existing one can be useful for us | 16:31 |
hrw|uds | wii markos_ | 16:32 |
hrw|uds | ops | 16:32 |
hrw|uds | markos_: how many things needs to be done then just kernel? flash-kernel, initrd stuff, xorg-xserver, gstreamer imx51 codecs? | 16:34 |
markos_ | you want the complete list? :) | 16:35 |
hrw|uds | I think that such one needs to be created at one point | 16:36 |
markos_ | there already is, it's just too long to paste here | 16:37 |
markos_ | but yeah, the ones you mentioned are already included, and high on the list | 16:37 |
hrw|uds | ok | 16:37 |
sveinse | Is pulseaudio working properly on ARM? | 17:51 |
sveinse | I.e. can I consider using it, or should I rather rely on direct access to alsa? | 17:52 |
=== JaMa is now known as JaMa|Off | ||
gean | anybody ? | 18:37 |
Neko | sveinse, it works | 18:57 |
camm | Can someone grant me access to an Ubuntu arm box to port gcl and its dependencies? | 19:01 |
gean | I Wanna port linux to my ARM netbook | 19:26 |
gean | anybody can help ? | 19:26 |
lag | gean: What notebook do you have? | 19:32 |
gean | Chinese One | 19:34 |
gean | ARM926EJ - Anyka AK7802 | 19:35 |
gean | 128 RAM | 19:35 |
gean | 266 Mhz Proc. | 19:35 |
gean | I've done my toolkit ( Binutils + gcc) for cross-compiling | 19:35 |
gean | but lacks the kernel | 19:36 |
cooloney | gean: don't you have any bsp or toolchain from the Anyka? | 19:36 |
lag | gean: You're going to struggle with that much RAM | 19:36 |
gean | ahahaha so much | 19:37 |
cooloney | gean: 128M RAM is not enough for ubuntu, i think | 19:37 |
lag | cooloney: cross-compiling shouldn't be an issue | 19:37 |
lag | RAM and proc specs will cripple you | 19:37 |
gean | i just need kind of blackbox or a lighter box | 19:37 |
cooloney | gean: what's kind of OS in this notebook | 19:38 |
cooloney | originally? | 19:38 |
lag | gean: Google is your friend here | 19:39 |
gean | no need for ubuntu, I'm using it at this desktop .. for the netbook it will be very slow ( if runs ... ) | 19:39 |
lag | gean: People has Linux running on it | 19:39 |
gean | WinCE | 19:39 |
cooloney | gean: yeah, i got it | 19:39 |
cooloney | gean: no kernel = no linux | 19:39 |
gean | Man .. i saw many things .. but officialy there is not any bsp | 19:40 |
gean | for this model .. | 19:40 |
cooloney | gean: so what's the cpu's model number? | 19:40 |
gean | no kernel patches | 19:40 |
gean | cooloney : anyka AK-7802 | 19:41 |
gean | wince says ARM926-AKCHIP | 19:42 |
gean | there is not win .. probably need hack it | 19:43 |
gean | but i never done such of thing ... any help would be appreciated | 19:44 |
Neko | anyone know anything about compcache/ramzswap support in initrd? | 19:53 |
Neko | it doesn't seem to do anything | 19:53 |
Neko | module is loaded by the initrd scripts but it's not set up for some reason | 19:53 |
Neko | (I tried 25 % and 96 M in the file and it never happens) | 19:53 |
ikke-t | getting frustrated with beagleboard and nfsroot | 20:56 |
hrw|uds | ikke-t: bbxm? | 20:56 |
ikke-t | bb c4 | 20:56 |
hrw|uds | auch | 20:56 |
ikke-t | unfortunately | 20:56 |
ikke-t | but with usb eth | 20:57 |
ikke-t | network works | 20:57 |
ikke-t | but now boot is complaining about not getting to lock nfs stuff, try with option '-o nolock' | 20:57 |
ikke-t | well, it doesn't read the option from kernel params | 20:57 |
hrw|uds | statd not running? | 20:57 |
ikke-t | statd on nfs host or bb? | 20:58 |
ikke-t | host I assume | 20:58 |
hrw|uds | iirc you should have portmap + statd + lockd working | 20:58 |
ikke-t | ? | 20:58 |
hrw|uds | do not remember. long time since last nfsroot | 20:58 |
ikke-t | hmm, just changed the host onto debian, maybe something missing from install. | 20:58 |
ikke-t | just installed kernel-nfs and nfs-common | 20:58 |
ikke-t | nfs-kernel-server that is | 20:59 |
* hrw|uds -> sessions | 21:00 | |
ikke-t | apt-cache doesn't find statd | 21:02 |
ikke-t | yahoo, found it from /etc/default/nfs-common | 21:04 |
ikke-t | \o/ | 21:04 |
ikke-t | no-luck, gets stuck at the same point as the earlier server. somewhere after reading fstab, since it complained about typo in fstab on nfsroot | 21:17 |
ikke-t | last message on screen being "Begin: Running /scriptsinit-bottom ...done." | 21:18 |
ikke-t | and happily stays there and doesn't answer even ping. pretty dead looking to me. | 21:18 |
ikke-t | i wonder if this works for somebody? | 21:19 |
ikke-t | beagleboard nfsroot -boot, ubuntu maverick. | 21:19 |
ikke-t | hmmm. kernel hung? echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this | 21:21 |
ikke-t | message | 21:21 |
ikke-t | doesn't sound promising | 21:22 |
hrw|uds | ikke-t: what I would do: remove / from fstab, check. check for statd/lockd/portmap on host and on target | 21:25 |
ikke-t | hrw|uds: statd/portmap running on host, removed / from nfsroot/etc/fstab | 21:29 |
ikke-t | no difference: | 21:29 |
hrw|uds | ;( | 21:30 |
ikke-t | i wonder should ubuntu init honor the console setting from kernel params? | 21:31 |
ikke-t | now it jumps to print on screen instead of serial at some point | 21:31 |
ikke-t | INFO: task bluetoothd:576 blocked for more than 120 seconds. | 21:32 |
ikke-t | stuck. | 21:33 |
acid | hello | 22:03 |
acid | I wrote an algorithm which when run on 486 processor takes 15 seconds, and on ARM takes 15 minutes. Same input data. Anyone has a clue why, or where to start looking? | 22:05 |
cwillu | acid, the algorithm is probably relying on libraries that are implemented ina completely different way, such that something which was cheap on the 486 isn't on arm | 22:08 |
cwillu | post the source on a pastebin if you want any more detail | 22:09 |
acid | cwillu; sec, I will | 22:26 |
acid | cwillu; http://pastebin.com/CBMZZZ1W | 22:30 |
acid | cwillu; this is no the whole thing, just the most expencive part. | 22:30 |
acid | and | 22:31 |
acid | this is actually not my code, I just rewrote some parts of it | 22:31 |
acid | :P | 22:31 |
acid | b/c they were slow... | 22:31 |
topfs2 | rsalveti, Just wanted to say that the kernel image I have doesn't seem to have the parameter. I checked ls /sys/module/smsc95xx/parameters/ and it just gives me turbo_mode. Is there any place youhave your kernel images so I could perhaps make a rootstock of them? | 23:15 |
topfs2 | and for others that might wonder what I'm on about its the http://cdimage.ubuntu.com/ubuntu-netbook/ports/releases/maverick/release/ for omap4 | 23:21 |
topfs2 | Seems like it might be missing the macaddr parameter for the ethernet on panda | 23:21 |
rsalveti | topfs2: weird, it's in our upstream kernel http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-maverick.git;a=blob;f=drivers/net/usb/smsc95xx.c;h=a3c8fe6e639eda5ec4eeb1cedf28e297d6386c37;hb=refs/heads/ti-omap4 | 23:21 |
rsalveti | module_param(macaddr, charp, 0); | 23:21 |
topfs2 | oh awesome your here :) | 23:21 |
topfs2 | yeah its rather weird, perhaps I should try to build a kernel myself and see | 23:22 |
topfs2 | Or is there any daily ones available somewere? | 23:22 |
rsalveti | topfs2: nops, just the one at the archive | 23:23 |
rsalveti | topfs2: if you are using maverick as a host you can quickly cross compile it | 23:23 |
topfs2 | Yup using maverick so I could probably check into that | 23:23 |
topfs2 | Do you know if there is any docs about it? | 23:23 |
topfs2 | but rsalveti it should show up in that ls right? so I'm not doing something wrong? | 23:24 |
rsalveti | topfs2: http://marcin.juszkiewicz.com.pl/2010/10/19/how-to-cross-compile-arm-kernel-under-ubuntu-10-10/ | 23:26 |
rsalveti | topfs2: and yes, you're right, it should be there | 23:26 |
topfs2 | awesome, I'll try compiling one then | 23:26 |
topfs2 | thanks for the help | 23:26 |
tmzt | kernel is easy to compile | 23:26 |
tmzt | just CROSS_COMPILE=toolchain- ARCH=arm | 23:26 |
rsalveti | topfs2: what did you get when you set up the macaddr argument at the kernel cmd line? | 23:27 |
tmzt | though your building kpkgs so hrw's stuff probably helps there | 23:27 |
topfs2 | rsalveti, should it be different if I have or not added it to the kernel commadn line? because I added it then checked the ls when it didn't work | 23:27 |
rsalveti | topfs2: module_param(macaddr, charp, 0); | 23:28 |
rsalveti | the 0 is the perm argument | 23:29 |
rsalveti | @perm: visibility in sysfs | 23:29 |
topfs2 | aha | 23:29 |
topfs2 | So perhaps I just have done it wrong in the kernel command line | 23:29 |
rsalveti | perm sets the visibility in sysfs: 000 means it's | 23:29 |
rsalveti | not there, | 23:29 |
rsalveti | topfs2: probably | 23:29 |
topfs2 | I tried just plain macaddr="foo" and smbc95xx.macaddr= (this one mostly as I wanted to try abit more) | 23:30 |
topfs2 | *smsc95xx | 23:30 |
topfs2 | setenv bootargs console=ttyO2,115200n8 serialtty=ttyO2 rw rootwait mem=463M elevator=noop vram=32M root=/dev/mmcblk0p2 macaddr="00:11:22:33:44:55" | 23:31 |
rsalveti | smsc95xx.macaddr=01:23:45:67:89:AB should work | 23:31 |
rsalveti | try this one | 23:32 |
topfs2 | Oh, without " " | 23:32 |
topfs2 | I'll try that | 23:32 |
topfs2 | Sorry, need to vanish for a quicky, gonna go and buy beer before it closes | 23:35 |
rsalveti | :-) | 23:35 |
tmzt | topfs2: try modinfo if it's buildt asa module | 23:39 |
tmzt | or check /sys/module | 23:40 |
* rsalveti will be out for now, uds party | 23:54 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!