#touchbook IRC Log


IRC Log for 2010-01-31

Timestamps are in GMT/BST.

[0:15] <Sn0wman_Desk> gregoiregentil: I'm not sure if this is even possible but... have you thought about, instead of having the linux kernel control everything, give the user the ability to set a boot flag which deterimines which kernel to load? So... boot -> selection -> reboot -> load OS
[0:15] <Sn0wman_Desk> wait...
[0:15] <Sn0wman_Desk> nm
[0:15] <Sn0wman_Desk> that was a dumb idea because android uses the linux kernel anyway
[0:15] <Sn0wman_Desk> its 2am here so I'm not quite thinking right...
[0:22] <Sn0wman_Desk> gregoiregentil: why is the android portion of the git tree empty?
[0:25] * MMlosh (~MMlosh@2001:470:1f0b:b78:9545:533f:6b11:fc97) has joined #touchbook
[0:34] * Matthias- (~Matthias@2a01:e35:2e79:e590:217:31ff:feb6:acf3) has joined #touchbook
[0:43] <Sn0wman_Desk> has anyone connected to the internet with android?
[0:45] <darrent> no wifi drivers working in the android build yet.
[0:47] <Sn0wman_Desk> isn't that up to the kernel?
[0:47] <Sn0wman_Desk> because if the kernel has it then why wouldn't android recognize it?
[0:54] <darrent> if that is true, then the configuration tool within android is broken, It does not "report" and available wifi device to configure.
[0:55] <Sn0wman_Desk> :_/
[0:55] <Sn0wman_Desk> :-/ *
[0:55] <darrent> it is very new, that could already be resolved in later builds.
[0:56] <Sn0wman_Desk> well I'm hoping to compile 2.0 tomorrow
[0:56] <Sn0wman_Desk> well nighty night guys
[0:56] * Sn0wman_Desk is now away - Reason : ZzzZzzZzz
[0:56] <darrent> personally, I find android an "ugly" interface for the TB...
[0:56] <darrent> Goodnight.
[0:56] <Sn0wman_Desk> well it is just easy to use imo
[0:57] <Sn0wman_Desk> I just wish the keyboard on AIOS or Ubuntu was more... usable...
[0:57] <Sn0wman_Desk> but gn... again...
[0:57] <darrent> ttfn
[0:58] * Josh-Harris (~JoshHarri@ Quit (Quit: Leaving...)
[2:00] * azaghal (~azaghal@ has joined #touchbook
[2:11] <Sn0wman_Desk> anyone know about how the TB starts up?
[2:12] <Sn0wman_Desk> like is X-loader renamed as MLO so that it will be the first thing to start?
[2:12] <Sn0wman_Desk> and then that loads uBoot?
[2:12] <Sn0wman_Desk> PLEASE answer that, because that has been bugging me for the past hour and a half
[2:13] * azaghal (~azaghal@ Quit (Quit: Lost terminal)
[2:15] * Vito89 (~quassel@gw.loccal.net) Quit (Quit: No Ping reply in 90 seconds.)
[2:15] * Vito89 (~quassel@gw.loccal.net) has joined #touchbook
[2:15] <DJWillis> Sn0wman_Desk: MLO is a signed (well hashed) version of X-Loader that then loads the stage 2 bootloader (U-Boot in most cases).
[2:16] <DJWillis> X-Loader is actually a very slimmed down hackup of old U-Boot code to fit in the stage 1 loader space.
[2:16] <Sn0wman_Desk> what do you mean "signed(well hashed)"
[2:16] <Sn0wman_Desk> but doesn't there have to be a MLO because thats what the OMAP ROM looks for on a partition of a SD card?
[2:17] <DJWillis> Sn0wman_Desk: I mean just that, to load X-Loader off the NAND or an SD it has to be hashed with a TI utility (that is now open source). The MLO file is the file 'after' it has been hashed (note: you do not need a hashed file to load it over USB if you booted that way).
[2:18] * Vito89 (~quassel@gw.loccal.net) Quit (Client Quit)
[2:18] <DJWillis> Sn0wman_Desk: that's just what I said - Boot rom checks NAND for MLO then SD, finds and runs suitable MLO (as a stage 1 bootloader), that then loads some stage 2 loader (U-Boot) that then has enough control to load an OS (or even runs apps from it if you want to code on the metal).
[2:19] <Sn0wman_Desk> hashed meaning it has to be an exact application?
[2:19] * Vito89 (~quassel@gw.loccal.net) has joined #touchbook
[2:19] <Corsac> hey there
[2:20] <Sn0wman_Desk> because if you tried to run a different stage1 loader, it wouldn't have the correct hash
[2:20] <Corsac> ok, now it's official, lzma pisses me off
[2:20] <Sn0wman_Desk> hehe
[2:20] <Corsac> (lzma build fails on OE)
[2:20] <Corsac> patches won't apply
[2:21] <Sn0wman_Desk> DJWillis: I'm not understanding the "hashed" part
[2:21] <Sn0wman_Desk> I know what a hash is
[2:22] * azaghal (~azaghal@ has joined #touchbook
[2:22] <DJWillis> Sn0wman_Desk: it's simple, you build X-Loader then 'signed/hashed' with this http://www.google.co.uk/url?sa=t&source=web&ct=res&cd=1&ved=0CAcQFjAA&url=http%3A%2F%2Fbeagleboard.googlecode.com%2Ffiles%2FsignGP.c&ei=illlS73GCor80wTiy-XJBg&usg=AFQjCNEQ4hgiNcZExa4-ENfj8Wla-Pq2YQ&sig2=PTK-KaeMRRcE9Qj1wrkHfg so the bootloader knows it's ok to run the code.
[2:22] * azaghal (~azaghal@ Quit (Client Quit)
[2:23] <DJWillis> Sn0wman_Desk: it's rather a jump to say that counts as 'signed' really ;-)
[2:23] <Sn0wman_Desk> but what is the point of 'signing' it? why can't the bootloader run it anyway?
[2:24] <DJWillis> Sn0wman_Desk: simple, it was not designed that way and other closed OS'es (CE/WinMobile/etc./etc.) do not work want to always let you boot unhashed code.
[2:24] * azaghal (~azaghal@ has joined #touchbook
[2:24] * azaghal (~azaghal@ Quit (Client Quit)
[2:25] <Sn0wman_Desk> it says it writes an "x-load.bin.ift" file when finished... do you just rename it? sorry for all of the questions, I'm just trying to understand how it all works
[2:25] <DJWillis> Sn0wman_Desk: got to go, that is the way the stage 0 bootrom works, it's been that way in the OMAPs for a long time and I expect it will stay that way for compatability.
[2:25] <DJWillis> Sn0wman_Desk: the .ift IS MLO
[2:25] <DJWillis> Got to go out for a bit.
[2:25] * azaghal (~azaghal@ has joined #touchbook
[2:25] <Sn0wman_Desk> so you can just rename it?
[2:26] <DJWillis> Sn0wman_Desk: yep
[2:26] <Sn0wman_Desk> alright, I might talk to you more tomorrow about since you g2g
[2:26] <Corsac> does anyone go to fosdem?
[2:28] <Sn0wman_Desk> DJWillis: one last questions so does that mean that the booting portion of the OMAP is pretty much locked up? Or can you change up the uboot or get xloader to load something different or what?
[2:31] * azaghal (~azaghal@ Quit (Quit: Reconnecting)
[2:31] * azaghal (~azaghal@ has joined #touchbook
[2:35] <DJWillis> Sn0wman_Desk: X-Loader can be 'any' signed app that is small enough, U-Boot V2 with the right options taken out, hell, I even have an 'MLO' somewhere that just pulses an LED on the Pandora and does nothing more.
[2:36] <DJWillis> Sn0wman_Desk: X-Loader is open source (http://git.openpandora.org/cgi-bin/gitweb.cgi?p=pandora-x-loader.git;a=summary is as current as any tree) so you can make it do whatever you want within reason, and no, your not 'forced' to load U-Boot etc. you load whatever you tell X-Loader to load to whatever address you see fit.
[2:38] <DJWillis> Sn0wman_Desk: U-Boot is loaded so it can setup the system to take a Linux kernel, you could right a small little micro bootloader that may do enough to start a kernel image and get it all into something small enough to be 'signed' as an MLO file.
[2:42] <josch> which would be cool given the uboot bloat
[2:45] * _koen_ (~x0115699@nat/ti/x-obaehykipehwivrv) has joined #touchbook
[2:46] <DJWillis> josch: well bloat is a matter of opinion ;), it can be done, I have knocked up micro bootloaders before for ARM's to launch Linux (but never an OMAP in fairness) it should not be hard but by the time you add futures hackers will want your back to U-Boot (or at least something that needs a stage 1 and 2 split) ;-)
[2:46] <DJWillis> Morning _koen_
[2:49] * azaghal (~azaghal@ Quit (Quit: Lost terminal)
[3:02] * azaghal (~azaghal@ has joined #touchbook
[3:02] * azaghal (~azaghal@ Quit (Client Quit)
[3:03] * MikeEvans (~MikeEvans@tandem.force9.co.uk) has joined #touchbook
[3:03] * azaghal (~azaghal@ has joined #touchbook
[3:06] <Corsac> http://paste.debian.net/58177/ _koen_, DJWillis, any idea?
[3:11] * azaghal_ (~azaghal@ has joined #touchbook
[3:13] * kelvinn (~kelvinn@220-244-12-240.static.tpgi.com.au) Quit (Quit: Ex-Chat)
[3:14] * azaghal (~azaghal@ Quit (Ping timeout: 276 seconds)
[3:17] * azaghal_ (~azaghal@ Quit (Ping timeout: 265 seconds)
[3:23] * azaghal (~azaghal@ has joined #touchbook
[3:27] <Corsac> hmhm, seems like a dos2unix problem
[3:45] * Vito89_ (~quassel@gw.loccal.net) has joined #touchbook
[3:52] * _koen_ wonders why some versions of patch break on dos style EOL markers
[4:01] <MikeEvans> _koen_: I've not experienced this (having applied only a few patches in my limited linux time) but I do recall writing various C programs to handle file input. Assuming that one ignores the CR and treats the LF as newline then all is well. There were two pitfalls however...
[4:01] <MikeEvans> both are associated with the end of the file rather than the end of the line per se
[4:02] <Corsac> _koen_: manually running what do_unpack_append() do worked
[4:02] <Corsac> and linux-omap_memory-optimization.patch (in the -pm recipe) is missing
[4:02] <MikeEvans> if your algorithm for reading lines for the file _requires_ line termination then files which do not have CR/LF at the end of the last line can cause problems. Also dos used to be in the habit of putting a CTRL-Z as an EF marker and that caused no end of headaches in some environments
[4:03] <_koen_> Corsac: from which repo are you building?
[4:04] * josch (~josch@vr142.rayconnect.de) Quit (Changing host)
[4:04] * josch (~josch@unaffiliated/josch) has joined #touchbook
[4:22] <Corsac> _koen_: ai's :/
[4:22] <Corsac> but gregoire merged not too long ago iirc
[4:23] * WebZrout (~kvirc@r11dh185.net.upc.cz) has joined #touchbook
[4:23] <Corsac> rah, still fail to build an initramfs-enabled uImage
[4:37] * azaghal_ (~azaghal@ has joined #touchbook
[4:40] * azaghal (~azaghal@ Quit (Ping timeout: 240 seconds)
[4:40] * azaghal_ is now known as azaghal
[4:49] <Corsac> _koen_: any idea what to do? gregoiregentil told me to put the cpio.gz file into the work folder but I'm not exactly sure what to do
[4:49] <Corsac> I put it into work/omap3-touchbook-angstrom-linux-gnueabi/linux-omap-pm-2.6.29-r49-pm2+gitre63cf0710a4fb639d91d3e8b05aa485fbfa381b3/ but it doesn't seem enough
[5:00] * azaghal_ (~azaghal@ has joined #touchbook
[5:04] * azaghal (~azaghal@ Quit (Ping timeout: 272 seconds)
[5:36] * azaghal_ is now known as azaghal
[5:39] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Remote host closed the connection)
[5:39] <_koen_> Corsac: I don't use the AI repo, so no idea
[5:40] <_koen_> all the stuff in the AI repo is too tangled to understand :(
[5:40] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[5:41] <Corsac> _koen_: yeah but in that case, it might not be ai related
[5:41] <Corsac> I just wonder how to make an uImage with initramfs
[5:52] <Corsac> I do have a 2.6M uImage which only has the kernel, I think
[5:52] <Corsac> and I have a 3.3M Angstrom-initramfs-image-glibc-ipk-2009.X-test-20100131-omap3-touchbook.rootfs.squashfs
[5:52] <Corsac> which might be an initramfs for that uImage
[5:52] <Corsac> but I don't know how to mix them
[5:59] <christoph_debian> hm anyone a idea when end-of-december orders will be shipped?
[6:00] <christoph_debian> (and anyone tried debian already on the tb?)
[6:01] <Corsac> christoph_debian: not yet, but in fact it might not be a really good idea
[6:01] <Corsac> considering that debian armel is armv4 while the touchbook has armv7
[6:01] <Corsac> that means that debian won't use all the available power, and runs slower than oe or even ubuntu
[6:01] <Corsac> (which I sad, I know)
[6:01] <Corsac> but there's not much we can do about that
[6:02] <christoph_debian> sure I'd like to measure that whjen I have $hardware and then decide for which os to go
[6:02] <christoph_debian> but running debian on a Core2 while debian being i486 isn't that bad so
[6:02] <christoph_debian> no idea how much it matters on arm
[6:02] <Corsac> that's completely differnt
[6:02] <Corsac> different
[6:03] <christoph_debian> might well be
[6:03] <_koen_> not to mention debian doesn't use the SGX, DSP or neon
[6:04] <_koen_> so you might just as well get some paper and crayons to draw your GUI
[6:06] <Corsac> christoph_debian: basically there's no such thing has genericity when you enter arm world
[6:06] <Corsac> which is painful for debian :/
[6:07] <Corsac> christoph_debian: that being said, it shouldn't be that hard
[6:07] <josch> <= runs debian on beagle without ever complaining about performance....
[6:08] <Corsac> I wanted to test it at some point, but I don't have enough time
[6:08] <_koen_> I wonder why the dpkg folks still havent fixed the bug that dpkg hardcodes all known archs in its source
[6:08] <Corsac> maybe because they wait for patches? :)
[6:08] <josch> since i also run debian on my openmoko freerunner i'm glad it's binaries are still armv4t compatible
[6:08] <christoph_debian> you'll have to rebuild anyway for a new arch ;)
[6:09] * christoph_debian has debian on a sheeva plug armv5t
[6:09] <Corsac> christoph_debian: and tbh I have to admit I'm fine with the ubuntu (which is built for armv5t iirc) install on the TB
[6:10] <christoph_debian> Corsac: 9.10 is v6 and 10.04 is v7
[6:10] <christoph_debian> 9.04 was v5
[6:10] <Corsac> and my goal these days is to manage to run mer decently
[6:10] * christoph_debian toys the idea of a gentoo
[6:10] <Corsac> which needs -pm according to Stskeeps
[6:11] <Corsac> which means I really need to build that !@#$ initramfs
[6:11] * _koen_ wonders why corsac needs an initramfs
[6:11] <MikeEvans> one starts to see the attraction of Gentoo's 'compile it for your target' approach in situations like that
[6:11] <_koen_> my touchbook works fine without initramfs
[6:11] <Corsac> _koen_: because I want to integrate properly with the current infrastructure
[6:12] <Corsac> (with the multi-boot)
[6:12] <Corsac> ok, I could multi-boot using multiple sd cards, good point
[6:12] <Corsac> but I don't know enough the boot stuff to cook that these days
[6:12] <Corsac> so just replacing the uImage and modules would be easier
[6:13] <Corsac> fsvo ???easier???
[6:13] <Corsac> -rw-r--r-- 1 corsac corsac 5,2M 31 janv. 15:11 ./work/omap3-touchbook-angstrom-linux-gnueabi/linux-omap-pm-2.6.29-r49-pm2+gitre63cf0710a4fb639d91d3e8b05aa485fbfa381b3/git/arch/arm/boot/uImage
[6:13] <Corsac> !
[6:13] <Corsac> it /seems/ that that could be an initramfs-enabled uImage
[6:21] * WebZrout (~kvirc@r11dh185.net.upc.cz) Quit (Quit: KVIrc Insomnia 4.0.0, revision: , sources date: 20090520, built on: 2009/06/08 23:52:48 UTC http://www.kvirc.net/)
[6:22] <Corsac> hmhm, ok, doesn't boot
[6:22] <Corsac> would have been to easy
[6:23] <Corsac> well, it boots but doesn't find the stuff to run, I guess
[6:23] <Corsac> I have the AI red splashscreen, but it just sits there
[6:37] * Meizirkki (~Meizirkki@bbwirelessgw2-fee1dc00-2.dhcp.inet.fi) has joined #touchbook
[6:44] * shiznebit (~quassel@ool-18b99096.dyn.optonline.net) has joined #touchbook
[6:55] * reis (~reini@cm133-175.liwest.at) has joined #touchbook
[6:58] <Corsac> gregoiregentil: hmmh, when the red splashscreen with progressbar is displayed, that means kernel has booted and i tries to jump to initramfs?
[6:59] <Corsac> _koen_: for a ???no initramfs??? boot, you just put a rootfs on the second partition?
[6:59] <Corsac> as a squashfs image?
[7:03] * shiznebit_ (~quassel@ool-18b99096.dyn.optonline.net) has joined #touchbook
[7:05] * shiznebit (~quassel@ool-18b99096.dyn.optonline.net) Quit (Ping timeout: 272 seconds)
[7:06] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Quit: Leaving)
[7:14] <Meizirkki> Corsac, I have that kind of setup
[7:14] <Meizirkki> ubuntu rootfs in the second partition
[7:14] <Meizirkki> only 3 partitions, no sqfs
[7:16] <Meizirkki> the red splashscreen stays on the screen till X starts (The progressbar doesn't move though)
[7:18] <_koen_> Corsac: no, as ext2 in the second partition
[7:18] <Corsac> in my case, I rebuilt a -pm kernel which doesn't boot fine (it doesn't run the initramfs, it seems, though I'm not exactly sure what it doesn)
[7:18] <_koen_> and I have a boot.scr file on the first partition that sets the bootargs
[7:19] <Corsac> well, it doesn run something since I have the red splashscreen
[7:19] <Corsac> but then it fails for unknown reason
[7:19] <Corsac> and it sits there, without any way to debug it
[7:19] <Meizirkki> try ctrl-alt-f2 ?
[7:19] <Corsac> I'll try debug that using a custom initramfs which wont run the psplash stuff so I can see what happens
[7:20] <Corsac> Meizirkki: considering the kernel is barely booted, a x shortcuts doesn't have much chance to work :/
[7:21] <Meizirkki> ok
[7:21] <Corsac> I guess the kernel is stuck on something
[7:21] <Corsac> it might have panic() or something
[7:21] <Corsac> but I can't see :)
[7:29] * azaghal_ (~azaghal@ has joined #touchbook
[7:30] * Meizirkki (~Meizirkki@bbwirelessgw2-fee1dc00-2.dhcp.inet.fi) Quit (Quit: L?hd?ss?)
[7:31] * Meizirkki (~Meizirkki@bbwirelessgw2-fee1dc00-2.dhcp.inet.fi) has joined #touchbook
[7:32] * azaghal (~azaghal@ Quit (Ping timeout: 258 seconds)
[7:35] * azaghal_ is now known as azaghal
[7:41] <Corsac> _koen_: is it possible/easy to change bootargs?
[7:50] <_koen_> I use a boot.src
[7:50] <_koen_> boot.scr for that
[8:01] <Corsac> created with a mkimage-x86 stuff?
[8:15] <_koen_> sort of
[8:15] <_koen_> there's an OE recipe to create then in recipes/angstrom
[8:15] <_koen_> with sample scripts in there as well
[8:27] <Corsac> hmhm http://www.exopc.com/
[8:27] <Corsac> win7, 550???
[8:29] <Corsac> (and french)
[8:38] <Corsac> _koen_ I wonder if the bootargs at http://www.alwaysinnovating.com/wiki/index.php/Booting are accurate for 2010.10b though
[8:38] <Corsac> 01 I mean
[8:45] <Corsac> ok, it doesn't care about my psplash=no arg :)
[8:46] * t_s_o (~tso@105.84-49-131.nextgentel.com) has joined #touchbook
[8:47] * Sn0wman_Desk (~IceChat7@c-75-66-187-145.hsd1.tn.comcast.net) Quit (Read error: Connection reset by peer)
[8:47] <Corsac> =false I meant
[8:49] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[9:03] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Quit: Leaving)
[9:24] <gregoiregentil> _koen_: tangled? I need to go to http://translate.google.com/. Don't think that it's a compliment! ;-) lol
[9:25] <gregoiregentil> Indeed! I don't disagree
[9:26] <gregoiregentil> Corsac: still lost?
[9:30] <Corsac> a little
[9:30] <Corsac> gregoiregentil: I managed to built an uImage which seems to contain an initrd, but it won't boot
[9:31] <Corsac> basically what I did was to build the kernel, then build the initrd, then rebuild the kernel
[9:31] <gregoiregentil> OK. Let me help you because you seem to put up-side-down the chat room ;-)
[9:31] <Corsac> :)
[9:31] <gregoiregentil> That's the order. so let me understand what you have done
[9:31] <gregoiregentil> you have built the kernel. Have you cross-compiled it
[9:31] <gregoiregentil> with OE or with your own toolchain?
[9:31] <Corsac> I've restarted from clean room, so I'm at that stage
[9:32] <Corsac> I ran bitbake linux-omap-pm
[9:32] <Corsac> I've adjusted the linux-omap-pm recipe and deconfig to merge initramfs-related stuff
[9:32] <Corsac> from linux-omap recipe and deconfig
[9:32] <Corsac> ok, it just finished
[9:33] <gregoiregentil> to merge initramfs-related stuff?
[9:33] <gregoiregentil> what do you mean?
[9:33] <Corsac> I'll paste the diff
[9:33] <gregoiregentil> You ned to have this:
[9:33] <gregoiregentil> CONFIG_INITRAMFS_SOURCE="initramfs.cpio.gz"
[9:34] <gregoiregentil> that's the only line to modify for INITRAMFS
[9:34] <Corsac> ok, it's present
[9:34] <gregoiregentil> then you need to copy the initramfs to the git folder
[9:34] <gregoiregentil> have you done that?
[9:35] <Corsac> http://paste.debian.net/58235/
[9:35] <Corsac> which git folder?
[9:35] <Corsac> (for the moment I don't yet have an initramfs built, I didn't yet ran bitbake ai-initramfs
[9:35] <gregoiregentil> the linux-omap-pm compilation folder
[9:35] <Corsac> ai-initramfs-image even
[9:36] <Corsac> deploy/glibc/images/omap3-touchbook/ ?
[9:36] <Corsac> or recipe/linux/linux-omap-pm-2.6.29 ?
[9:36] <gregoiregentil> no
[9:37] <gregoiregentil> you need to compile ai-initramfs-image with bitbake ai-initramfs-image
[9:37] <Corsac> ok, running that
[9:37] <gregoiregentil> then you need to do something like:
[9:37] <gregoiregentil> cp /OE/build/tmpdeploy/glibc/images/omap3-touchbook/ai-initramfs-image-omap3-touchbook.cpio.gz /OE/build/tmp/work/omap3-touchbook-angstrom-linux-gnueabi/linux-omap-2.6.29-r49/git/initramfs.cpio.gz
[9:37] <gregoiregentil> then you need to do bitbake -if
[9:37] <gregoiregentil> compile linux-omap
[9:37] <gregoiregentil> to recompile the linux-omap (with take only secondsss)
[9:38] <Corsac> I don't have a git folder there
[9:38] <gregoiregentil> in order to get the intramfs.cpio.gz inside the uImage
[9:38] <Corsac> here it's work/omap3-touchbook-angstrom-linux-gnueabi/linux-omap-pm-2.6.29-r49-pm2+gitre63cf0710a4fb639d91d3e8b05aa485fbfa381b3/
[9:38] <gregoiregentil> Possible. Strange.
[9:38] <Corsac> I have a temp folder inside
[9:38] <gregoiregentil> in any case, the initramfs.cpio.gz needs to be where the .config is
[9:39] <gregoiregentil> quite strange that you don't have a git folder. You may need to ask Koen for this.
[9:39] <gregoiregentil> but you said that you managed to compile linux-omap. Correct?
[9:39] <gregoiregentil> bitbake linux-omap-pm is not failing
[9:39] <gregoiregentil> according to what you said...
[9:39] <gregoiregentil> correct?
[9:40] <Corsac> yes
[9:40] <Corsac> but maybe there's stuff cleaned up
[9:41] <Corsac> ai-initramfs-image-omap3-touchbook.cpio.gz is a symlink, correct?
[9:41] <Corsac> to Angstrom-ai-initramfs-image-glibc-ipk-2009.X-test-20100131-omap3-touchbook.rootfs.cpio.gz in my case
[9:42] <gregoiregentil> Correct
[9:42] <gregoiregentil> Corsac: you need to git pull my OE to get the latest
[9:42] <Corsac> already done
[9:42] <gregoiregentil> there are a few typo in a few places
[9:42] <gregoiregentil> do it now! I have just committed something that you need
[9:42] <gregoiregentil> also, you need to do the following:
[9:43] <Corsac> ok
[9:43] <gregoiregentil> bitbake initramfs-module-ai -c clean && bitbake ai-install-image ai-initramfs-image
[9:43] <gregoiregentil> there is a fix in ons of the intramfs ai script so you need to clean initramfs-module-ai and then recompile ai-initramfs-image
[9:44] <gregoiregentil> also you probably need ai-install-image which is needed if you select one of the other options in the initramfs image
[9:44] <gregoiregentil> in the boot UI I mean
[9:44] <gregoiregentil> OK. Now there is one more question:
[9:44] <gregoiregentil> why are you trying to do that?
[9:45] <Corsac> because Stskeeps said running -pm might be mandatory for mer to run fine with accel
[9:46] <gregoiregentil> I don't believe that...
[9:46] <gregoiregentil> who is Stskeeps?
[9:46] <gregoiregentil> where did you read that?
[9:46] <Corsac> Stskeeps is Mer main developer
[9:47] <gregoiregentil> Also, the linux-omap-pm has been somewhat left behind by Koen. The future of kernel is linux-omap-psp
[9:47] <gregoiregentil> Stskeeps has a Touch Book?
[9:47] * Stskeeps (~chobits@Maemo/community/distmaster/Stskeeps) has joined #touchbook
[9:47] <Corsac> hey Stskeeps :)
[9:47] <Stskeeps> moo
[9:47] <gregoiregentil> The only difference between linux-omap-pm and linux-omap for Touch Book is the PM stuff. So I'm not sure
[9:47] <gregoiregentil> how it can be related to rootfs. but perhaps I'm wrong
[9:48] <gregoiregentil> Stskeeps: could you please help us and clarify
[9:48] <Corsac> gregoiregentil: I think it's related to compatibility with SGX drivers
[9:48] <gregoiregentil> why you need linux-omap-pm?
[9:48] <Corsac> (about mer and acceleration)
[9:48] <gregoiregentil> but SGX drivers is not related to pm or not
[9:49] <Stskeeps> gregoiregentil: long story short, nokia SGX libraries comes with texture from pixmap extension, which is very useful for composition. nokia SGX libs are tightly connected with SGX driver. in nokia SGX libs, power management is enabled and as they're closed, we can't do anything about that, this means that PM needs to be enabled for the kernel driver too, which requires PM
[9:50] <gregoiregentil> OK. Thanks for the clarification. I understand now. One question:
[9:50] <Stskeeps> we have done the same approach on zoom2 and beagle.
[9:50] <Corsac> ok so that means I need to add nokia sgx libs and drivers too, and not used the ones from aios
[9:51] <gregoiregentil> Corsac: I don't think that you can use Nokia SGX driver
[9:51] <gregoiregentil> I think that you need indeed to take SGX userspace libs from Nokia
[9:51] <Stskeeps> gregoiregentil: it's available under EULA through tablets-dev.nokia.com/eula :P
[9:51] <gregoiregentil> and you need to compile SGX driver + linux-omap-pm.
[9:52] <gregoiregentil> Stskeeps: Am I correct?
[9:52] <Stskeeps> right
[9:52] <gregoiregentil> Stskeeps: that's interesting stuff. I was not aware of this.
[9:52] <gregoiregentil> Stskeeps: so does it mean that there is some kind of X acceleration using the SGX?
[9:53] <Stskeeps> gregoiregentil: the window manager does composition using the texture from pixmap method, yes
[9:53] <Stskeeps> which is why n900 is so snappy :)
[9:54] <gregoiregentil> Stskeeps: I start to be interested... does it help for scrolling?
[9:54] <Stskeeps> dunno, but it's very nice to have texture from pixmap - it's what enables compiz and all that stuff on desktop GL
[9:54] <gregoiregentil> How can I use that with xfce4/xfwm4/Fourdan's compositor ?
[9:54] <gregoiregentil> yup. Stskeeps: I pretty understand what you mean
[9:55] <gregoiregentil> two more questions: so you imported the SGX libs from Nokia
[9:55] <gregoiregentil> and you also recompiled
[9:55] <gregoiregentil> the pixmap library with some special stuff from Nokia.
[9:55] <gregoiregentil> Am I correct?
[9:55] <Corsac> gregoiregentil: you cant (about xfwm)
[9:55] <gregoiregentil> Corsac: yes, it's what I'm thinking
[9:55] <Stskeeps> Corsac: sure? clutter does it fine :P
[9:56] <Stskeeps> gregoiregentil: the SGX libs include a new GLES extension - hang on
[9:56] <Corsac> gregoiregentil: xfwm only uses RENDER accel
[9:56] <Corsac> gregoiregentil: it doesn't use 3d accel at all
[9:56] <Stskeeps> ah
[9:56] <gregoiregentil> Stskeeps: so beyond the SGX libraries, what else have you taken from the Nokia SDK?
[9:57] <Stskeeps> gregoiregentil: well, kernel patch
[9:57] <Stskeeps> and that's about it
[9:57] <gregoiregentil> in other words, how do you leverage the new GLES extension?
[9:57] <gregoiregentil> strange. I was thinking that there is some kind of modified X or modified Pixman.
[9:57] <gregoiregentil> how Pixman knows to use the SGX new GLES extension....???
[9:58] <gregoiregentil> You see my point?
[9:58] <Stskeeps> http://omappedia.org/wiki/Maemo_Getting_Started is the project i do in my official capacity together with a guy from TI.
[9:59] <Stskeeps> http://maemo.gitorious.org/+mer/fremantle-hildon-desktop/clutter_0_8-mer/blobs/Mer/clutter/eglx/clutter-eglx-texture-pixmap.c - search for EGL_NOKIA_texture_from_pixmap
[9:59] <Stskeeps> also, their xorg fbdev driver seems to support ordinary compositing
[9:59] <Stskeeps> at least EXA
[10:01] <gregoiregentil> It's funny because it's the road to improve X through SGX, and I have always been told by TI that such path would not really help taking into account the low frequency of SGX stuff...
[10:01] <Stskeeps> dunno, but it works well on zoom2 and beagle :P
[10:03] <Stskeeps> but it would be nice to have a maemo like desktop accelerated on TB.
[10:04] <Stskeeps> which is really the key point
[10:04] <Corsac> yeah, working on it
[10:04] <Corsac> :)
[10:04] <Stskeeps> it could even steal sales from iPad ;p
[10:10] <_koen_> Isn't texture from pixmap already the the last 2 SGX SDKs from TI?
[10:11] <Stskeeps> is it now?
[10:11] <Stskeeps> i would be happy to hear that
[10:12] * _koen_ is used to nokia using obsolete software and hacking it, just look at the maemo5 base system
[10:12] <_koen_> I'm currently too lazy to download the SGX sources and look for it
[10:16] <Corsac> _koen_: and sdk base system is worse :)
[10:43] * Josh-Harris (~JoshHarri@ has joined #touchbook
[10:46] * samham (~d53f9766@gateway/web/freenode/x-okilxpwshmeblalx) has joined #touchbook
[10:49] <samham> Hello, i was about to order my touchbook, but then i noticed this: Ordering today will ensure you a better place in the queue list, but expect your order to be shipped within two months. Is this for tablet and notebook format? I really need a unit ASAP, if it's not available i will have to order a diff dev kit from IGEPv2
[11:10] <Corsac> gregoiregentil: ok, seems that the git/ folder is created during the build, and somehow cleaned up after
[11:10] <gregoiregentil> very strange. it should not...
[11:14] * samham (~d53f9766@gateway/web/freenode/x-okilxpwshmeblalx) Quit (Quit: Page closed)
[11:22] <Corsac> arg, it boots fine but I can't select the ???other??? image :)
[11:22] <Corsac> it boots aios
[11:22] <Meizirkki> Corsac, you have working omap-pm now ?
[11:22] <Corsac> Meizirkki: it seems
[11:23] <Meizirkki> w00tf
[11:23] <Corsac> though I don't know how to test reliabily
[11:23] <Meizirkki> Please upload uImage somewhere for me to download :)
[11:25] <Corsac> wait a sec
[11:26] <Corsac> arg, though I fogot to add PVR=m to defconfig
[11:27] <Corsac> and it panics when trying to boot mer ><
[11:27] * Meizirkki want's to test it anyway
[11:27] <Corsac> you'll have to wait, I have to run
[11:28] <Meizirkki> np, take you time
[11:36] * shiznebit_ (~quassel@ool-18b99096.dyn.optonline.net) Quit (Remote host closed the connection)
[12:10] * Josh-Harris (~JoshHarri@ Quit (Read error: Connection reset by peer)
[12:15] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[12:21] * Josh-Harris (~JoshHarri@ has joined #touchbook
[12:29] * javaJake (~javaJake@unaffiliated/javajake) Quit (Quit: leaving)
[12:30] * javaJake (~javaJake@unaffiliated/javajake) has joined #touchbook
[12:38] * Josh-Harris (~JoshHarri@ Quit (Read error: Connection reset by peer)
[12:47] * Josh-Harris (~JoshHarri@ has joined #touchbook
[13:02] * MMlosh (~MMlosh@2001:470:1f0b:b78:9545:533f:6b11:fc97) Quit (Quit: Bye...)
[13:08] * Meizirkki (~Meizirkki@bbwirelessgw2-fee1dc00-2.dhcp.inet.fi) Quit (Quit: L?hd?ss?)
[13:11] * zelrikriando (~zelrikria@unaffiliated/zelrikriando) Quit (Quit: Lost terminal)
[13:21] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Remote host closed the connection)
[13:21] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[13:22] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Remote host closed the connection)
[13:24] * Sn0wman_Desk (~IceChat7@c-75-66-187-145.hsd1.tn.comcast.net) has joined #touchbook
[13:24] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[13:31] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Remote host closed the connection)
[13:31] * craigevil (~thouartgo@adsl-76-249-22-201.dsl.klmzmi.sbcglobal.net) has joined #touchbook
[13:31] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[13:37] * DREW_ (~chatzilla@cpe-70-112-6-224.austin.res.rr.com) Quit (Ping timeout: 240 seconds)
[13:37] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Remote host closed the connection)
[13:38] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[13:38] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Remote host closed the connection)
[13:38] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[14:15] * MikeEvans (~MikeEvans@tandem.force9.co.uk) has left #touchbook
[14:15] * DREW_ (~chatzilla@cpe-70-112-6-224.austin.res.rr.com) has joined #touchbook
[14:38] * javaJake (~javaJake@unaffiliated/javajake) Quit (Remote host closed the connection)
[14:38] * javaJake (~javaJake@unaffiliated/javajake) has joined #touchbook
[14:58] * minimec (~minimec@195-202-200-93.dclient.hispeed.ch) has joined #touchbook
[15:19] * gregoiregentil (~gregoire@c-98-210-206-210.hsd1.ca.comcast.net) has left #touchbook
[15:19] * gregoiregentil (~gregoire@c-98-210-206-210.hsd1.ca.comcast.net) has joined #touchbook
[15:20] <tommd> Welcome back Gregorie - gone for a whole 5 seconds! Must be busy.
[15:21] <tommd> gregoiregentil: Is anyone working on an applet to provide gui control over ai_daemon? If not I might start reading up on xfce and build one.
[15:22] <Gary13579> ai_daemon?
[15:22] <tommd> I'm not clear on the API (or experienced with making icons) but figure it can't be that bad.
[15:22] <tommd> Gary13579: The thing that performs the rotations.
[15:22] <Gary13579> ah, that should be incredibly simple
[15:23] <tommd> Yeah, I've just not made an XFCE applet before and don't want to start if someone is already doing it.
[15:28] <Corsac> tommd: there already is one, iirc
[15:28] <Corsac> though I can't remember the name
[15:29] <Corsac> can you try ai-menu-ui?
[15:29] <Corsac> it segfaults here, but
[15:29] <gregoiregentil> working here.
[15:29] <gregoiregentil> tommd: There are /usr/bin/ai/desktop/display.py
[15:29] <Corsac> yeah but remember I have a weird kernel
[15:30] <gregoiregentil> which briefly do some stuff on ai-daemon settings
[15:30] <gregoiregentil> Corsac: not surprised....
[15:30] <gregoiregentil> tommd: My recommendation would be perhaps to try to improve display.py. It's in Python, so not too complex.
[15:30] <Corsac> ha yes, display.py is what I remembered
[15:30] <gregoiregentil> An advanced button which shows / hides some advanced configurations would be great I think
[15:31] <gregoiregentil> it's the way I would do it if I put time and energy on this
[15:32] <gregoiregentil> Corsac: Is there an option in dpkg-buildpackage to avoid build the doc packages. It takes ages for mplayer
[15:32] <gregoiregentil> to build the doc package and I don' t care about this
[15:33] <Corsac> gregoiregentil: not a global option at least
[15:33] <Corsac> gregoiregentil: look in debian/rules
[15:34] <Corsac> but you might have to pass something to configure if the docs are built by mplayer build script
[15:35] <tommd> Corsac, gregoiregentil: I see that now, so I guess what I was thinking is a design change. Would it be preferable to have an applet that simply has a menu of five options (left, right, invert, normal, auto) - this is two clicks from the normal desktop vs start->control panel->display->drop-down box -> invert or whatever.
[15:36] <tommd> OTOH, it would be another thing in the panel and we don't want this to look like a late 90s windows box with 15 different applets thinking they are important enough to the user.
[15:36] <Corsac> tommd: you could configure yourself a shortcut?
[15:37] <Corsac> (and isn't the autorotation working?)
[15:37] <tommd> The point to me is once I select a rotation (say, left) I want to be done, not have a "display" window still open. It seems simple stupid but little aspects like this are the difference between a slick experience and something that seems rougher / less fluid.
[15:38] <tommd> Corsac: The autorotation does work, but isn't always desired (laying in bed holding the TB upside-down).
[15:42] <Corsac> you could bind some key shortcuts
[15:42] <tommd> Corsac: ai-menu-ui has always been faulty to me (I thought) tried it just now and it took a minute to work (frozen) but is functional now.
[15:42] <Corsac> gregoiregentil: hmh, weird, the selection choice doesn't work anymore, it'll only select AIOS :)
[15:42] <tommd> very slow
[15:42] <Corsac> (which works, but...)
[15:43] <tommd> Corsac: It just crashed
[15:43] <tommd> Corsac: last message in the terminal from ai-menu-ui was "--> end fill_scrolling"
[15:43] <tommd> is there a --debug option or some such?
[15:44] <Corsac> no idea
[15:44] <Corsac> gregoiregentil: aha, it seems that the display is weirdly calibrated in fact, when I touch bottom left it'll boot ubuntu
[15:44] <tommd> Humm, display issues too - the "always-on-top" onscreen keyboard I had open doesn't play nicely.
[15:45] <Corsac> gregoiregentil: ok, it's completely reverted :)
[15:45] <Corsac> top left is bottom right, top right is bottom left
[15:45] <Corsac> no idea how I managed to do that
[15:46] <Corsac> (well, except using -pm kernel)
[15:50] * carsonb (~carson@69-196-165-93.dsl.teksavvy.com) has joined #touchbook
[15:50] * Josh-Harris (~JoshHarri@ Quit (Read error: Connection reset by peer)
[15:51] * Josh-Harris (~JoshHarri@ has joined #touchbook
[16:07] * craigevil (~thouartgo@adsl-76-249-22-201.dsl.klmzmi.sbcglobal.net) Quit (Quit: leaving)
[16:16] * carsonb (~carson@69-196-165-93.dsl.teksavvy.com) has left #touchbook
[16:17] * Matthias- (~Matthias@2a01:e35:2e79:e590:217:31ff:feb6:acf3) Quit (Remote host closed the connection)
[16:31] * edgar (~edgar@p5DC009FB.dip0.t-ipconnect.de) has joined #touchbook
[16:31] <edgar> tommd ?
[16:40] * snlemons (~snlemons@c-24-60-77-165.hsd1.nh.comcast.net) has joined #touchbook
[16:40] * edgar (~edgar@p5DC009FB.dip0.t-ipconnect.de) Quit (Ping timeout: 265 seconds)
[16:48] * snlemons (~snlemons@c-24-60-77-165.hsd1.nh.comcast.net) has left #touchbook
[16:49] * Stskeeps (~chobits@Maemo/community/distmaster/Stskeeps) Quit (Quit: Reconnecting)
[16:49] * topi` (~topi@host-62-220-235-64.fonet.fi) Quit (Ping timeout: 246 seconds)
[16:50] * t_s_o (~tso@105.84-49-131.nextgentel.com) Quit (Ping timeout: 246 seconds)
[16:51] * t_s_o (~tso@105.84-49-131.nextgentel.com) has joined #touchbook
[16:51] * topi` (~topi@host-62-220-235-64.fonet.fi) has joined #touchbook
[16:55] * shiznebit (~quassel@ool-18b99096.dyn.optonline.net) has joined #touchbook
[17:13] * reis (~reini@cm133-175.liwest.at) Quit (Quit: reis)
[17:25] * mcr (mcr@wlan204.sandelman.ca) Quit (Remote host closed the connection)
[17:25] * mcr (mcr@wlan204.sandelman.ca) has joined #touchbook
[17:35] * darrent (~ai@202-161-10-50.dyn.iinet.net.au) has left #touchbook
[17:41] * ed_ (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[17:42] * ed_ (~Ed@dsl-216-221-38-194.aei.ca) Quit (Client Quit)
[18:04] <tommd> edgar?
[18:16] * minimec (~minimec@195-202-200-93.dclient.hispeed.ch) Quit (Quit: Lost terminal)
[18:16] * t_s_o (~tso@105.84-49-131.nextgentel.com) Quit (Remote host closed the connection)
[18:18] * Vito89_ (~quassel@gw.loccal.net) Quit (Ping timeout: 246 seconds)
[18:33] <gregoiregentil> Corsac: Please send this upstrean: http://git.alwaysinnovating.com/cgit.cgi/ai.openembedded.dev/commit/?id=c191416f78baf6e1aaf99d14915d1e8bec455ac5
[18:43] * craigevil (~thouartgo@adsl-76-249-22-201.dsl.klmzmi.sbcglobal.net) has joined #touchbook
[18:51] <tommd> My top battery appears to no longer be getting charged (or providing power). Is there a common fix, such as fixing a solder joint?
[18:52] <Sn0wman_Desk> same problem
[19:42] <tommd> Humm, moderately confusing. Not willing to test to confirm it right now but the TB seems to power off when I remove the top half... if I turn the top half on when its alone then it will operate fine. Perhaps there is a brief spike in current when the top half is removed that the battery can not fulfill?
[20:16] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Ping timeout: 272 seconds)
[20:20] * DREW_ (~chatzilla@cpe-70-112-6-224.austin.res.rr.com) Quit (Read error: Connection reset by peer)
[20:28] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[20:36] * Sn0wman_ (~IceChat7@c-75-66-187-145.hsd1.tn.comcast.net) has joined #touchbook
[20:36] * Sn0wman_Desk (~IceChat7@c-75-66-187-145.hsd1.tn.comcast.net) Quit (Read error: Connection reset by peer)
[20:39] * bandwidthcrunch (~quassel@ has joined #touchbook
[21:01] <Sn0wman_> what partition are the .squashfs's on?
[21:03] <tommd> 2nd
[21:03] <tommd> hidden files (starting with a ".")
[21:04] * Josh-Harris (~JoshHarri@ Quit (Read error: Connection reset by peer)
[21:05] <Sn0wman_> ah thanks. I remember that now
[21:05] <Sn0wman_> GOD, ls -A needs to be default
[21:09] * shiznebit (~quassel@ool-18b99096.dyn.optonline.net) Quit (Remote host closed the connection)
[21:14] * Josh-Harris (~JoshHarri@ has joined #touchbook
[21:14] <Sn0wman_> if you lose /etc settings, is there a way to get them back? I reinstalled with aptitude with no avail...
[21:21] * Josh-Harris (~JoshHarri@ Quit (Read error: Connection reset by peer)
[21:22] * Josh-Harris (~JoshHarri@ has joined #touchbook
[21:22] * merp (~merp@ has joined #touchbook
[21:36] * edt (~Ed@dsl-216-221-38-194.aei.ca) Quit (Ping timeout: 240 seconds)
[21:49] * edt (~Ed@dsl-216-221-38-194.aei.ca) has joined #touchbook
[22:02] * Josh-Harris (~JoshHarri@ Quit (Read error: Connection reset by peer)
[22:02] * Josh-Harris (~JoshHarri@ has joined #touchbook
[22:06] * tommd (~Thomas_Du@65-102-21-214.ptld.qwest.net) has left #touchbook
[22:42] * reis (~reini@cm133-175.liwest.at) has joined #touchbook
[22:47] <Corsac> gregoiregentil: I can't send a patch which is just a function removal without any explanation :)
[22:47] * alextisserant (~alextisse@c-76-21-41-103.hsd1.ca.comcast.net) has joined #touchbook
[22:48] <gregoiregentil> the explanation is quite straightforward
[22:48] <gregoiregentil> which version do you want? the diplomatic one or what I really think? ;-)
[22:48] <gregoiregentil> the diplomatic one is:
[22:48] <gregoiregentil> you should not do a widget_show after a menu creation
[22:48] <gregoiregentil> what happens is that gtk+ is confused by this
[22:48] <gregoiregentil> because when gtk+ get the very first activate event for showing the submenu
[22:49] <gregoiregentil> it considers that the menu has already been shown. so it doesn' t do anything
[22:49] <gregoiregentil> then the user clicks a second time. gtk+ hides the submenu which is already hidden
[22:50] <gregoiregentil> then the user clicks a third time (at this point, the user is really pissed off especially when this is the 6th submenu that he wants to open)
[22:50] <gregoiregentil> and then gtk+ shows for the first time the submenu. and the user wonders: what's the hell?
[22:53] <Corsac> hmhm, ok
[22:54] <Sn0wman_> did someone have to do heavy editing of something to get android running? because I just compile 0xdroid's android 2.0 and it is pretty messed up
[23:04] <Corsac> gregoiregentil: how comes the patch is against modules/menu/desktop-menu-file.c while the file is called modules/menu/desktop-menu.c?
[23:05] <Corsac> ha no wait
[23:07] <Corsac> gregoiregentil: against which version is the patch?
[23:07] <Corsac> gregoiregentil: 4.4 only?
[23:13] <Corsac> hmhm, seems that the libxfcegui4 patch is only for create_sendto_menu()
[23:14] <bandwidthcrunch> Anyone noticing this while baking ? arm-angstrom-linux-gnueabi-objcopy: Unable to recognise the format of the input file `/OE/build/tmp/work/armv7a-angstrom-linux-gnueabi/lzma-4.65-r1.1/package/usr/bin/lzma'
[23:14] <bandwidthcrunch> ERROR: runstrip: ''arm-angstrom-linux-gnueabi-strip' --remove-section=.comment --remove-section=.note '/OE/build/tmp/work/armv7a-angstrom-linux-gnueabi/lzma-4.65-r1.1/package/usr/bin/lzma'' strip command failed
[23:14] <bandwidthcrunch> ERROR: QA Issue with lzma: Architecture did not match (40 to 3) on /work/armv7a-angstrom-linux-gnueabi/lzma-4.65-r1.1/packages-split/lzma/usr/bin/lzma
[23:14] <bandwidthcrunch> ERROR: QA run found fatal errors. Please consider fixing them.
[23:15] <Corsac> can you run file on /OE/build/tmp/work/armv7a-angstrom-linux-gnueabi/lzma-4.65-r1.1/package/usr/bin/lzma ?
[23:15] <Corsac> I had issues building lzma because of dos2unix problems, but that seems unrelated
[23:15] <bandwidthcrunch> yeah i fixed that by apt getting dos2unix too
[23:16] <bandwidthcrunch> yeah i can run /OE/build/tmp/work/armv7a-angstrom-linux-gnueabi/lzma-4.65-r1.1/package/usr/bin/lzma
[23:17] <bandwidthcrunch> wondering why should the stripping be done by arm-angstrom-linux-gnueabi-strip when the lma is a 386 binary
[23:18] <Corsac> well, why would it be a 386 binary
[23:18] <Corsac> since it's not lzma-native and in armv7a-angstrom-linux-gnueabi folder?
[23:18] <Corsac> there's definitely something wrong
[23:18] <bandwidthcrunch> file /OE/build/tmp/work/armv7a-angstrom-linux-gnueabi/lzma-4.65-r1.1/package/usr/bin/lzma
[23:18] <bandwidthcrunch> /OE/build/tmp/work/armv7a-angstrom-linux-gnueabi/lzma-4.65-r1.1/package/usr/bin/lzma: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.15, not stripped
[23:19] <Corsac> doesnt look very good
[23:20] <bandwidthcrunch> doesnt .. is there a way i can clean all the changes that i might have done to openembedded and rebuild the tree ?
[23:24] * reis (~reini@cm133-175.liwest.at) Quit (Quit: reis)
[23:28] <Corsac> git clean -x -d; git checkout .; git pull
[23:28] <Corsac> which will wipe *all* your modifs
[23:29] <Corsac> and remove the tmp/ folder will remove all the build stuff
[23:34] <bandwidthcrunch> let me take that for a spin
[23:35] * alextisserant (~alextisse@c-76-21-41-103.hsd1.ca.comcast.net) Quit (Quit: alextisserant)
[23:48] <Sn0wman_> when you build the AI-image, what does it make? a filesystem or a .ipk or squashfs or what?
[23:49] <Corsac> I my guess is a squashfs but I never did that
[23:51] <Sn0wman_> what do you do?
[23:52] * Sn0wman_ is now known as Sn0wman_Desk
[23:58] <gregoiregentil> Corsac: it's against ANY version of xfce
[23:59] <gregoiregentil> Corsac: the problem occurs in multiple locations. In xfdesktop/modules/menu for the menu itself
[23:59] <Sn0wman_Desk> Corsac: by "I never did that", do you mean you never built an AI image or you use some other method to build it?
[23:59] <gregoiregentil> but also libxfcegui4 for the right click on the panel on a window title
[23:59] <gregoiregentil> also on the right click on the desktop

These logs were automatically created by TouchBook-LogBot on irc.freenode.net using the Java IRC LogBot.