**** BEGIN LOGGING AT Mon Jun 29 02:59:56 2009
Jun 29 06:54:33 <mckoan> good morning
Jun 29 07:28:05 <cdbot2> * * OE Bug 5277 has been created by chrisfriedt(AT)gmail.com
Jun 29 07:28:07 <cdbot2> * * wmiconfig_svn fails due to 'no GNU_HASH' bug
Jun 29 07:28:09 <cdbot2> * * http://bugs.openembedded.org/show_bug.cgi?id=5277
Jun 29 07:32:46 <vhallac> Hi. Is there a full list of tasks used in openembedded somewhere?
Jun 29 07:33:08 <vhallac> I am trying to track down what do_build does, but can't seem to find it anywhere.
Jun 29 07:38:44 <fraxinath> http://andy.ath.cx/mediawiki/index.php/OpenEmbedded#Toolchain-abfolge
Jun 29 07:41:56 <vhallac> Thanks for the info, but I was lookig for a description of what each task does.
Jun 29 07:42:43 <fraxinath> fetch downloads the source
Jun 29 07:43:15 <fraxinath> actually it's all pretty obvious until package and populate_staging right?
Jun 29 07:43:31 <vhallac> Well, when there is compile and build, things get a bit hairy
Jun 29 07:43:39 <fraxinath> as in how?
Jun 29 07:43:48 <vhallac> compile is obvious... What does build do, then?
Jun 29 07:44:02 <fraxinath> i don't think it really does anything
Jun 29 07:44:09 <fraxinath> it's just the flag stating that it's done i think
Jun 29 07:44:34 <vhallac> Hm. That could explain it. There was only a build definition for a single recipe in the classes.
Jun 29 07:44:44 <vhallac> I thought I was missing something.
Jun 29 07:45:03 <fraxinath> there may be other stages too
Jun 29 07:45:11 <fraxinath> the recipe can define arbitrary ones
Jun 29 07:46:24 <vhallac> Fairly flexible, but confusing for the newcomers. :)
Jun 29 07:47:59 <fraxinath> yeah i had that problem too ^^
Jun 29 07:49:18 <vhallac> Thanks for the info. I may come and give you more headache later on.
Jun 29 07:49:49 <vhallac> Someone told me that u-boot image is not created when doing the rootfs image build. I was trying to track that down
Jun 29 07:50:24 <vhallac> I found the do_deploy being added after do_compile, before do_build, and it all started getting too confusing after that. :)
Jun 29 07:54:47 <fraxinath> aah okay i see ^^
Jun 29 08:16:36 <hrw> morning
Jun 29 08:16:53 <mckoan> hi hrw 
Jun 29 08:19:15 <hrw> vhallac: there are many tasks in OE and some of them are not used by default
Jun 29 08:20:05 <hrw> vhallac: "build" task is default one - other tasks are added as needed to be done before 'build' one
Jun 29 08:23:43 <vhallac> I see. I was checking base.bbclass, and it looks like it is not as simple. It looks like we scatter tasks, some of them with dependencies, and some (like listtasks) stand alone. bitbake then orders them.
Jun 29 08:24:28 <vhallac> I am now chasing what task gets pulled in when I specify something with DEPENDS or RDEPENDS.
Jun 29 08:27:28 <hrw> vhallac: "bitbake -g target" then
Jun 29 08:32:41 <vhallac> Aha. Thanks, it looks like I can track it down now.
Jun 29 08:38:16 <vhallac> So, is there some sort of setting somewhere that says task X of package P1 will pull in task Y of packages in DEPENDS?
Jun 29 08:38:29 <vhallac> Or is it inside bitbake?
Jun 29 08:52:43 <BlindMan> morning all
Jun 29 09:08:56 <pb_> BlindMan: good morning
Jun 29 09:27:57 <Treibholz> is there an official way of reconfiguring a kernel? Or do I have to edit defconfig by hand?
Jun 29 09:34:23 <RP> morning all
Jun 29 09:35:30 <pb_> Treibholz: "make menuconfig"?
Jun 29 09:35:35 <pb_> or bitbake -c menuconfig if you want bitbake to do it
Jun 29 09:40:18 <sgh> When compiling mesa. What is does --enable-glx-tls do ?
Jun 29 09:41:01 <pb_> enables tls support in the glx libraries, I guess.  what does "configure --help" say about it?
Jun 29 09:44:39 <sgh> pb_:   --enable-glx-tls        enable TLS support in GLX [default=disabled]... But what is it.?
Jun 29 09:45:42 <pb_> presumably that's tls as in thread local storage
Jun 29 09:46:26 <pb_> glx is the glue between the opengl api and X.
Jun 29 09:58:26 <sgh> pb_: ah ok .... so tls migh improveve performance I surpose.
Jun 29 10:04:50 <pb_> yeah, I guess that's the idea.  I don't really know what glx would use it for, but generally it's a way of avoiding locking.
Jun 29 10:04:57 <florian> good morning
Jun 29 10:05:29 <pb_> I'm not sure if this means that glx uses threads internally, or if it's just in case the gl api is called from a multithreaded client.
Jun 29 10:17:51 <RP> Hmm, we don't have any syntax to unset a variable do we? :/
Jun 29 10:27:20 <dth> bis später!
Jun 29 10:41:32 <pb_> RP: No, I don't think so.  Of course, you can set it to "" which often has much the same effect, but I don't think there is any way to actually delete one without resorting to python.
Jun 29 10:45:06 <hrw> RP: question I have
Jun 29 10:45:38 <hrw> RP: your recent change  to psplash (2009-05-28) makes it work properly on RGB and BGR screens?
Jun 29 11:09:39 <hrw> Palm SDK is done with MACHINE=qemux86 ;D
Jun 29 11:18:41 <florian> hrw: is it available now?
Jun 29 11:20:48 <ynezz> it leaked, and it's windows only
Jun 29 11:21:16 <ynezz> http://rs44tl2.rapidshare.com/files/249167601/PalmSDK-Win-0.3.4.exe
Jun 29 11:21:50 <BlindMan> palm sdk for webos? :)
Jun 29 11:30:04 <hrw> florian: it leaked
Jun 29 11:30:18 <hrw> ynezz: inside you will find virtualbox image+iso ;D
Jun 29 11:30:28 <florian> :-)
Jun 29 11:30:29 <hrw> I plan to write about it on blog
Jun 29 11:30:38 <florian> that's a cool idea
Jun 29 11:30:55 * florian plans to write a blog entry about linuxtag
Jun 29 11:32:55 <hrw> good idea it is
Jun 29 11:33:08 <hrw> someone know which vesa mode is 'vga=864'?
Jun 29 11:35:17 <hrw> screen is totally decalibrated in vga or xga ;(
Jun 29 11:35:58 <ynezz> http://en.wikipedia.org/wiki/VESA_BIOS_Extensions
Jun 29 11:39:35 <hrw> gracias
Jun 29 11:40:12 <hrw> 1440x900 8bit... strange
Jun 29 12:55:58 <RP> hrw: Yes, psplash should work on different byte orders now
Jun 29 12:56:36 <hrw> great, we need to update then
Jun 29 12:57:06 <hrw> RP: any plans to include greyscale patch from OE? so it will work on 1,2,16bit?
Jun 29 12:58:27 <RP> hrw: I wasn't aware of it...
Jun 29 12:59:01 <hrw> RP: neither do I - just found it
Jun 29 12:59:30 <hrw> would be fun to run psplash on 160x105x1 display ;)
Jun 29 13:00:07 <RP> hrw: For certain definitions of fun ;-)
Jun 29 13:00:57 <hrw> RP: now we use fbprogress which is BUG only app
Jun 29 13:01:48 <hrw>         geometry 160 105 160 105 1
Jun 29 13:02:08 <RP> hrw: I'll take sane patches for psplash...
Jun 29 13:09:09 <hrw> I know
Jun 29 13:09:21 <hrw> RP: I met Dirk during LinuxTag
Jun 29 13:10:56 <BlindMan> oh, there were more than just woglinde from the channel at the lt
Jun 29 13:12:31 <pb_> heh
Jun 29 13:12:37 <hrw> BlindMan: ;D
Jun 29 13:13:11 <hrw> BlindMan: Florian, Crofton, Ph5, woglinde, me, stefan, mickeyl and few others
Jun 29 13:18:30 <BlindMan> ah, well
Jun 29 13:18:39 <BlindMan> they all forgot to put their nicknames on the badges :p
Jun 29 13:19:54 <methril|work> good morning
Jun 29 13:20:12 <hrw> hi methril|work 
Jun 29 13:23:37 <methril|work> hi hrw!!
Jun 29 13:23:51 <methril|work> hrw: you know a lot of "handled" like devices
Jun 29 13:24:32 <methril|work> hrw: did you test the mini2440?
Jun 29 13:24:55 <hrw> methril|work: Florian has mini2440 and micro2440 - I played with them a bit during LinuxTag
Jun 29 13:25:10 <hrw> methril|work: I suggest getting version with WVGA screen
Jun 29 13:25:18 <methril|work> hrw: I was lookinf for something similar with a 7" LCD ...
Jun 29 13:25:33 <hrw> methril|work: device is very nice to play for someone who wants to learn about embedded systems
Jun 29 13:25:38 <methril|work> hrw: o you know more hw like that?
Jun 29 13:25:50 <methril|work> hrw: it's for a prototype at work
Jun 29 13:25:51 <hrw> methril|work: 165€ for wvga version is bargain
Jun 29 13:26:03 <hrw> methril|work: http://friendlyarm.eu/ then shop
Jun 29 13:26:19 <methril|work> hrw: nice, thank you
Jun 29 13:26:29 <hrw> methril|work: s3c2440 on board, ethernet, usb host, usb device, 3 rs232 ports, audio in/out jacks
Jun 29 13:26:41 <hrw> expansion connectors with extra signals
Jun 29 13:26:58 <methril|work> yes, i see it, i'd like to see a board like this for beagle :)
Jun 29 13:27:01 <hrw> there are two versions available: wvga and qvga (35€ difference)
Jun 29 13:27:19 <hrw> we had both at OE stand
Jun 29 13:27:40 <methril|work> what a pitty not been in LinuxTag :(
Jun 29 13:27:49 <florian> these were the two versions of the micro2440 btw
Jun 29 13:28:03 <hrw> florian: ops, my fault then
Jun 29 13:28:30 <florian> the mini2440 is more or less the same but the cpu board is not detatchable but part of the baseboard
Jun 29 13:28:46 <florian> but the mini2440 is smaller than the micro2440 baseboard
Jun 29 13:29:09 <methril|work> florian: what product do you recommend to make a info system?
Jun 29 13:29:59 <florian> methril|work: depends on the required performance, price and guarantees about availability
Jun 29 13:30:30 <methril|work> florian: for java, i think i need better performance
Jun 29 13:31:09 <methril|work> i think the mini feets our needs
Jun 29 13:31:31 <methril|work> because you're able to connect a 7 inch LCD without external interface
Jun 29 13:31:36 <florian> methril|work: maybe yes... ask thebohemian, he should have experience with java on this cpu. 
Jun 29 13:32:17 <methril|work> florian: and other kind of products, with more or less the same cap? I'm not able to find a lot of them
Jun 29 13:33:19 <florian> methril|work: therare some more, but likely to be more expensive
Jun 29 13:34:15 <florian> methril|work: http://garz-fricke.de/jupiter-openframe_en.html?ptab=1
Jun 29 13:34:34 <methril|work> florian: It doesn't matter so much, is for my company ;)
Jun 29 13:34:47 <florian> One example I found recently... should be faster than the mini2440 at least.
Jun 29 13:35:04 <hrw> I start to like few things in WebOS
Jun 29 13:35:31 <methril|work> hrw: where are you testing the WebOS? :)
Jun 29 13:35:42 <methril|work> do you have a palm pre?
Jun 29 13:35:55 <hrw> methril|work: I am running leaked Palm SDK
Jun 29 13:36:44 <methril|work> hrw: where do you find the time?? hehehe
Jun 29 13:36:51 * florian wants to see screensots
Jun 29 13:37:10 <methril|work> florian: i ask here because i see your las't blog post on planet linuxtogo
Jun 29 13:38:35 <florian> This reminds me that I have some topics for my blog and shour start to write today.
Jun 29 13:39:04 <methril|work> hehee, i'm not the only lazy one blogging :)
Jun 29 13:39:59 <florian> btw... while this is on topic here: if someone wants to add his blog to planet.linuxtogo.org I can do this easily.
Jun 29 13:40:26 <methril|work> well, i didn't update it so often
Jun 29 13:41:16 <florian> the cronjob deoesn't care ;)
Jun 29 13:42:38 <methril|work> openblog.methril.net
Jun 29 13:42:45 <methril|work> i'm on planet.openmoko,org
Jun 29 13:43:04 <hrw> florian: http://marcin.juszkiewicz.com.pl/download/pre-shots/
Jun 29 13:43:59 <methril|work> nice!
Jun 29 13:46:00 <florian> interesting... 
Jun 29 13:47:00 <hrw> it is clearly visible that their programmers thinks that WebOS == 320x480px 
Jun 29 13:47:22 <hrw> many apps have hardcoded parts of UI
Jun 29 13:48:33 <broonie> Given that they have full control of the hardware these things are supposed to run on it doesn't seem unreasonable.
Jun 29 13:49:28 <hrw> broonie: sure, but it looks strange in emulator
Jun 29 13:49:45 <hrw> which can be run in any resolution 
Jun 29 13:49:52 <hrw> or even on bare metal x86
Jun 29 13:55:16 <thebohemian> hrw: hardcoding UI sizes .. what year do we have 1995 ?
Jun 29 13:57:27 <hrw> thebohemian: remember that this is Palm...
Jun 29 13:57:33 <hrw> they had 160x160 for ages
Jun 29 14:01:52 <mario-goulart> Gnutoo: any progress on your mips booting process?
Jun 29 14:04:39 <Gnutoo> mario-goulart, hi
Jun 29 14:04:46 <Gnutoo> mario-goulart, no I did others things
Jun 29 14:04:54 <Gnutoo> mario-goulart, and from your side?
Jun 29 14:05:01 <mario-goulart> Gnutoo: ah, ok.
Jun 29 14:05:16 <Gnutoo> mario-goulart, we must coordonate our efforts
Jun 29 14:05:25 <mario-goulart> Gnutoo: I can boot it, but right now I'm stuck on (I guess) uClibc.
Jun 29 14:05:43 <Gnutoo> mario-goulart, ah ld-uclibc that segfault?
Jun 29 14:05:54 <mario-goulart> Gnutoo: it seems that dynamic linked binaries segfault
Jun 29 14:06:04 <Gnutoo> mario-goulart, yes it's the fault of ld-uclibc
Jun 29 14:06:13 <mario-goulart> Gnutoo: yeah, it seems so.
Jun 29 14:06:58 <mario-goulart> Gnutoo: I've applied openwrt patches, but no luck so far.
Jun 29 14:06:59 <Gnutoo> mario-goulart, I didn't succeed in making debug libraries for that particular library...others parts of uclibc are debug-enabled but not this particular part...
Jun 29 14:07:12 <Gnutoo> mario-goulart, did you try with openwrt's CFLAGS?
Jun 29 14:08:23 <mario-goulart> Gnutoo: I haven't tried openwrt's CFLAGS.
Jun 29 14:08:38 <Gnutoo> mario-goulart, ok
Jun 29 14:09:08 <Gnutoo> mario-goulart, else I could try to debug it without the source...it's mips assembly and I learned it long time ago
Jun 29 14:09:36 <mario-goulart> Gnutoo: that looks scary. :-)
Jun 29 14:09:51 <Gnutoo> mario-goulart, lol
Jun 29 14:10:13 <Gnutoo> mario-goulart, the problem would be to follow the execution...I had problems with it
Jun 29 14:12:03 <mario-goulart> Gnutoo: now I'm trying to build an image with some more tools to see if I can get something from the environment.
Jun 29 14:12:09 <Gnutoo> ok
Jun 29 14:12:12 <Gnutoo> I've:
Jun 29 14:12:15 <Gnutoo> *gdbserver
Jun 29 14:12:21 <Gnutoo> *NFS client
Jun 29 14:12:23 <Gnutoo> *portmap
Jun 29 14:12:34 <Gnutoo> else I use the cross tools
Jun 29 14:13:32 <mario-goulart> Gnutoo: right.
Jun 29 14:13:51 <hrw> root access to webos is enabled via ssh... 
Jun 29 14:13:56 <hrw> jumi ;D
Jun 29 14:14:44 <Gnutoo> hrw, you've got a palm pre?
Jun 29 14:14:54 <hrw> no - palm sdk leaked
Jun 29 14:15:13 <Gnutoo> ah ok
Jun 29 14:15:27 <Gnutoo> there are 3 interesting phones that are not openmoko:
Jun 29 14:15:29 <Gnutoo> *palm pre
Jun 29 14:15:36 <Gnutoo> *android g1 unlocked
Jun 29 14:15:38 <mckoan> hrw: where is announced SDK public release? Ithought it was under NDA
Jun 29 14:15:40 <Gnutoo> *exeda
Jun 29 14:15:45 <chouimat> morning
Jun 29 14:16:07 <hrw> mckoan: google for 'palm sdk leaked'
Jun 29 14:18:07 <mckoan> uh leaked aka un-officially released just to tect market response ;-)
Jun 29 14:18:18 <mckoan> s/tect/test
Jun 29 14:18:47 <mckoan> It doesn't work into VmWare winXP client
Jun 29 14:19:11 <mckoan> maybe it's too leaked :-D
Jun 29 14:19:37 <hrw> mckoan: it works in virtualbox
Jun 29 14:19:47 <hrw> mckoan:  http://marcin.juszkiewicz.com.pl/download/pre-shots/ - Palm Pre SDK
Jun 29 14:21:57 * mckoan wonders why a modern SDK should put on the weight of a JVM
Jun 29 14:22:23 <broonie> mckoan: You want it to run as slowly as teh target system for best emulation :)
Jun 29 14:22:46 <mckoan> broonie: LOL
Jun 29 14:29:06 <Gnutoo> mario-goulart, my problem is this one: segfault+disass => No function contains program counter for selected frame.
Jun 29 14:29:49 <Gnutoo> mario-goulart, bt => warning: GDB can't find the start of the function at 0x2aab060f.
Jun 29 14:29:51 <Gnutoo> etc...
Jun 29 14:30:46 <mario-goulart> Gnutoo: hmmm.  Are you also stuck on the execution of dynamic linked binaries?
Jun 29 14:30:55 <Gnutoo> mario-goulart, yes
Jun 29 14:31:15 <mario-goulart> Gnutoo: I thought you were having problems with the kernel.
Jun 29 14:31:24 <Gnutoo> it didn't change since last time since I didn't touch my wrt
Jun 29 14:31:36 <Gnutoo> mario-goulart, that was long time ago...
Jun 29 14:32:16 <mario-goulart> Gnutoo: ah, ok.  I was disconnected since wednesday...
Jun 29 14:32:40 <Gnutoo> mario-goulart, didn't you remember...when I saw you on irc last time..I saw you had the same problem than me with mips+shared libs
Jun 29 14:33:04 <Gnutoo> and then I found out that you were doing the same thing than me...
Jun 29 14:33:33 <mario-goulart> Gnutoo: ah, ok.  Now I remember.
Jun 29 14:33:38 <Gnutoo> :)
Jun 29 14:35:34 <Gnutoo> is bal branch and link?
Jun 29 14:36:19 <kergoth> morning
Jun 29 14:36:33 <pb_> g'day kergoth
Jun 29 14:36:46 <Gnutoo> mario-goulart, http://pastebin.com/m218aa87
Jun 29 14:37:51 <Gnutoo> but strange branch and link should be jal
Jun 29 14:37:53 <Gnutoo> jump and link
Jun 29 14:38:05 <Gnutoo> http://en.wikipedia.org/wiki/MIPS_architecture for the mips assembly 
Jun 29 14:39:02 <Gnutoo> maybe I'll ask on #mips
Jun 29 14:39:18 <Gnutoo> doesn't exist
Jun 29 14:40:23 <Gnutoo> I'll try ##asm
Jun 29 14:43:12 <mario-goulart> Gnutoo: looks like it is branch and link
Jun 29 14:43:23 <mario-goulart> Gnutoo: from http://www.cs.unibo.it/~solmi/teaching/arch_2002-2003/AssemblyLanguageProgDoc.pdf
Jun 29 14:43:28 <Gnutoo> mmm seems the same thing than mov pc to ra
Jun 29 14:43:32 <Gnutoo> ok thanks
Jun 29 14:44:47 <Gnutoo> ok I understand it's realy a move 
Jun 29 14:44:57 <mario-goulart> Gnutoo: just a random guess: couldn't it be a linker bug?
Jun 29 14:45:11 <Gnutoo> what do you mean exactly?
Jun 29 14:45:45 <mario-goulart> Gnutoo: that maybe ld is linking things wrong, so the problem would be binutils, not uClibc.
Jun 29 14:45:54 <Gnutoo> ok
Jun 29 14:46:21 <mario-goulart> Gnutoo: as I said, just a random guess.
Jun 29 14:46:46 <Gnutoo> that would be a good guess since
Jun 29 14:47:10 <Gnutoo> *we didn't tried the openwrt obtimizations but I guess it won't work either with them
Jun 29 14:47:20 <Gnutoo> *you tried patching uclibc
Jun 29 14:47:35 <Gnutoo> maybe we could do the following
Jun 29 14:48:07 <Gnutoo> no bad idea
Jun 29 14:48:23 <Gnutoo> (was trying the openwrt toolchain on the oe rootfs)
Jun 29 14:48:37 <Gnutoo> (compile with openwrt and run inside oe's rootfs)
Jun 29 14:55:03 <Gnutoo> mario-goulart, I bet it's more the undelying libs because:
Jun 29 14:55:16 <Gnutoo> *if I run my hello world in openwrt's rootfs it works
Jun 29 14:56:10 <Gnutoo> mario-goulart, unfortunately I have to go
Jun 29 14:56:12 <mario-goulart> Gnutoo: is hello word dynamically linked?
Jun 29 14:56:16 <Gnutoo> mario-goulart, yes
Jun 29 14:56:34 <mario-goulart> Gnutoo: I have to go in a few minute too.  We can talk later.  Thanks for your help and time.
Jun 29 14:56:35 <Gnutoo> mario-goulart, it work with openwrt's rootfs but not in oe's rootfs
Jun 29 14:56:47 <Gnutoo> mario-goulart, thanks for your cooperation too
Jun 29 14:57:02 <Gnutoo> mario-goulart, just in case do you have my mail?
Jun 29 14:57:38 <mario-goulart> Gnutoo: no, I don't have your e-mail (mine is mario.goulart at gmail).
Jun 29 14:57:53 <Gnutoo> mario-goulart, mine is gnutoo at no-log.org
Jun 29 14:58:07 <mario-goulart> Gnutoo: ok
Jun 29 15:33:03 <mckoan> bbl
Jun 29 15:55:07 <hrw> http://marcin.juszkiewicz.com.pl/2009/06/29/palm-sdk-has-leaked/
Jun 29 16:04:53 <BlindMan> hrw: w00t :)
Jun 29 16:12:21 <florian> re
Jun 29 17:02:30 <hrw> have a nice rest of day
Jun 29 17:22:05 <cdbot2> * * OE Bug 5156 has been RESOLVED (WONTFIX) by quickx(AT)hotmail.com
Jun 29 17:22:07 <cdbot2> * * meta-toolchain doesn't produce any *.pc  files in deployed file
Jun 29 17:22:09 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=5156
Jun 29 17:39:10 <oespirit> ~curse un-auto-tooled packages having hard-coded paths for making recipe writing a nightmare
Jun 29 17:39:11 <ibot> May you be reincarnated as a Windows XP administrator, un-auto-tooled packages having hard-coded paths for making recipe writing a nightmare !
Jun 29 17:45:48 <kergoth> un-autotooled packages are annoying in general.  i hate autotools, but at least it can be worked around in reliable and predictable ways, instead of having to invent new workarounds for each and every damn package..
Jun 29 17:45:49 <kergoth> heh
Jun 29 17:51:35 <oespirit> kergoth: +1
Jun 29 18:00:13 <BusError> is ther a way NOT to build the -dbg packages ? i got a small (console image) tree and it's already 800MB (!)
Jun 29 18:01:04 <kergoth> building them isn't a problem
Jun 29 18:01:15 <kergoth> your real question is what are you including in the image that results in depending on them, no?
Jun 29 18:01:45 <kergoth> but you could always override FILES_${PN}-dbg = "" in your local.conf, if you really want to avoid packaging them
Jun 29 18:01:47 <BusError> example : alsa libs, 7KB. Alsa-lib-dev, 1.3KB --- alsa-lib-dbg 3.2MB
Jun 29 18:01:57 <kergoth> right, so why are you including the debug stuff in your image?
Jun 29 18:02:53 <BusError> well I'm making a 3G accessible, production box, I;m trying to package the whole *.ipk on a SD card for when I go update the box...
Jun 29 18:03:16 <BusError> they are not on the image, but they are package of package-index, so they end up on my card...
Jun 29 18:04:25 <kergoth> if you override the FILES for that package itll make them empty, but still create the packages, iirc they're set to allow empty
Jun 29 18:05:38 <BusError> can I filter files afterward maybe ? is there a "package-index that subset" ?
Jun 29 18:05:51 <BusError> I don't mind having them on the workstation
Jun 29 18:07:23 <oespirit> BusError: if you want to do away with building them altogether, you can just edit PACKAGES variable in bitbake.conf
Jun 29 18:10:25 * BusError goes and have a look
Jun 29 18:14:13 * BusError removes doc, locales and dbg :>
Jun 29 18:21:33 * kergoth thinks we really do need -=
Jun 29 18:22:44 <BusError> went from 759MB to 250 -- just removing the -dbg- :>
Jun 29 18:25:37 <oespirit|zZz> BusError: :) 
Jun 29 18:27:25 <BusError> well, that is enough of a difference I could consider doing remote update here, it's far from insignificant!
Jun 29 18:34:55 <RP> kergoth: Think about the headaches that could introduce with ordering ;-)
Jun 29 18:36:42 <kergoth> well, it's clear that -= would be immediate, the same way += and .= are, so obviously the user would have to consider ordering issues there.  I don't think we'd want an _remove.  It's clear that += and =+ are semantically word based, so a -= that's also word based would make sense
Jun 29 18:49:09 <BET_frogger>       
Jun 29 20:14:01 <hoj> anyone have experience running oprofile?  I see no sample file generated, and must've missed some part of install/setup
Jun 29 20:29:16 <sgh> Hi! ... is it possible to aboid vompiling mesa with --enable-glx-tls without actually modifying the recipe ?
Jun 29 20:31:11 <pb__> Yes, you can override EXTRA_OECONF if you want.
Jun 29 20:37:31 <sgh> pb__: But how do I do it without editing the recipe - if it is possible of cause :)
Jun 29 20:38:16 <kergoth> EXTRA_OECONF_pn-<recipename> = "foo".   standard use of OVERRIDES to conditionally override a variable, see the bitbake manual for details
Jun 29 20:45:12 <sgh> kergoth: thanks ... will try that.
Jun 29 20:54:07 <Gnutoo> I start realizing...I'm doing:
Jun 29 20:55:01 <Gnutoo> gdbserver 168.0.0.139:8022 chroot /mnt/NFS/ /bin/hello
Jun 29 20:55:10 <Gnutoo> how do I follow that with the debugger?
Jun 29 20:55:32 <Gnutoo> do I need to use the second serial port and do the following:
Jun 29 20:56:38 <Gnutoo> tftp-kernel-boot->static-gdbserver that listen on the serial port->(hello world+while loop) non-static
Jun 29 21:30:32 <Gnutoo> I'll try the following setup: chroot->gdbserver-static->hello-wold-dynamic
Jun 29 21:33:13 <Gnutoo> mmm...I broke the compiler
Jun 29 22:28:51 <timtimred> lorem ipsum...
Jun 29 23:18:04 <pb__> dolor sit amet
Jun 30 02:01:08 <NekoXP> if I restarted a build with packaged-staging enabled but didn't set the pstage directory where on earth did it get put?
Jun 30 02:02:28 <NekoXP> can I just copy everything from tmp/deploy/glibc/pstage to my permanent pstage dir and have it pick them up still?
Jun 30 02:13:53 <Tartarus> nope, gotta have it from the start
**** ENDING LOGGING AT Tue Jun 30 02:59:57 2009