**** BEGIN LOGGING AT Tue Sep 25 02:59:58 2012
Sep 25 07:11:01 <afournier> morning!
Sep 25 07:13:42 <afournier> i am trying to build a linux module, but it's looking for a header that is not present in the sysroot, even if it is present in the linux source
Sep 25 07:13:50 <afournier> linux/bounds.h
Sep 25 07:14:03 <afournier> is it normal ?
Sep 25 07:18:47 <JaMa|Zzz> afournier: are you using denzil? it was fixed in master
Sep 25 07:19:43 <afournier> yes i am
Sep 25 07:24:56 <JaMa|Zzz> afournier: then backport those changes if you want
Sep 25 07:25:07 <afournier> yes i will
Sep 25 07:35:45 <mckoan> good morning
Sep 25 08:07:33 <gmc> khem: that I figured, but there's a bunch of stuff i'm gonna have to figure out how to fill in.. such as TCMODE
Sep 25 08:10:51 <bluelightning> morning all
Sep 25 08:17:43 <gmc> morning!
Sep 25 08:28:05 <hrw> hello
Sep 25 08:56:18 <silvio> morning all
Sep 25 08:59:09 <afournier> hi silvio
Sep 25 09:01:03 <silvio> hi afournier 
Sep 25 10:14:05 <pb_> morning all
Sep 25 10:16:21 <bluelightning> hi pb_
Sep 25 10:17:42 <silvio> hi pb
Sep 25 10:34:32 <pb_> hi bluelightning, silvio
Sep 25 12:10:23 <pb_> Does anybody know why gettext-native depends on libxml2-native?
Sep 25 12:58:20 <bluelightning> pb_: nothing suggestive in git blame?
Sep 25 13:45:30 <denisATeukrea> hi JaMa 
Sep 25 13:46:05 <denisATeukrea> what do you think of xinput-calibrator for resistive touchscreens?
Sep 25 13:47:44 <denisATeukrea> like touchscreen converted from tslib to evdev
Sep 25 13:48:32 <denisATeukrea> for instance the gta04 is like that
Sep 25 13:49:04 <denisATeukrea> and some people on a mailing list said it was better with xf86-input-tslib
Sep 25 13:57:29 <pb_> bluelightning: not really.  as far as I can tell it's unnecessary nowadays.
Sep 25 13:57:58 <pb_> that line was added in aae5021101224344a2b1a3af5becf74291fbbfe5, "Add depends to gettext to avoid host contamination".
Sep 25 13:58:29 <pb_> but, subsequently, the configure arguments were changed to --with-included-libxml which makes it use a locally-bundled copy of libxml rather than the installed/sysroot one anyway.
Sep 25 13:59:39 <pb_> the thing that seems even more peculiar is that the target gettext depends on libxml2-native, which doesn't sound like an appropraite fix for "host contamination" in any event.
Sep 25 13:59:43 <denisATeukrea> hmmm I found that:
Sep 25 14:00:02 <denisATeukrea> http://lists.goldelico.com/pipermail/gta04-owner/2012-April/002228.html 
Sep 25 14:00:06 <denisATeukrea> => tslib is better
Sep 25 14:00:56 <bluelightning> pb_: I dislike terse commit messages such as that one </grumble>
Sep 25 14:04:49 <bluelightning> pb_: assuming libxml2-native only provides libraries and no utilities, definitely
Sep 25 14:05:25 <bluelightning> pb_: seems like it ought not to be needed anyway if we're explicitly telling it to use an internal copy of libxml2 as you say
Sep 25 14:16:47 <pb_> yeah
Sep 25 14:25:27 <pb_> I sent email to the list about it.
Sep 25 14:29:29 <bluelightning> yep, just saw that, thanks
Sep 25 14:54:40 <pb_> morning kergoth
Sep 25 14:56:56 <kergoth> hey
Sep 25 14:57:59 * sgw YPTM: Saul is on
Sep 25 15:01:00 <sgw> oops wrong channel
Sep 25 15:01:53 <afournier> i dont understand how sysroots/qemux86/kernel/ is populated, in the kernel.bbclass kernel_do_install there is "cp -fR * $kerneldir" so it should copy everything but include/linux/bounds.h is not present
Sep 25 15:16:08 <pb_> I think that might be one of the files that kernel.bbclass helpfully deletes for you.
Sep 25 15:16:33 <pb_> it's actually in generated/, right?
Sep 25 15:44:29 <hrw> does someone has busybox-only-image recipe?
Sep 25 15:49:26 <bluelightning> hrw: core-image-minimal should be busybox for the shell unless something else depends on bash, if that's what you mean
Sep 25 15:51:31 <kergoth> you can use the VIRTUAL-RUNTIME vars to adjust what goes into core-image-minimal, iirc. e.g. tinylogin vs busybox, busybox vs sysvinit.. but you'd have to add extra bits to install an inittab for it :)
Sep 25 15:51:32 <hrw> bluelightning: but to build it I need libffi
Sep 25 15:51:49 <hrw> trying micro-(base-)image from old OE
Sep 25 15:52:16 <hrw> on AArch64 we do not have libffi yet so no glib-2.0 and lot of stuff
Sep 25 15:53:34 <pb_> yeah, micro-base-image is fairly close to busybox only
Sep 25 15:53:40 <pb_> there's a meta-micro layer which works with oe-core, fwiw
Sep 25 15:53:47 <hrw> cool
Sep 25 15:55:14 <hrw> pb_: grabbed micro-base-image recipe
Sep 25 15:57:43 <bluelightning> hrw: what's the dependency chain there?
Sep 25 15:58:52 <hrw> bluelightning: ?
Sep 25 15:59:08 <hrw> bluelightning: ah, image
Sep 25 15:59:16 <bluelightning> hrw: the layer itself does not pull in libffi directly, so my question is what does?
Sep 25 15:59:28 <hrw> Missing or unbuildable dependency chain was: ['core-image-base', 'packagegroup-base-extended', 'ofono', 'bluez4', 'udev', 'glib-2.0', 'libffi']
Sep 25 15:59:31 <bluelightning> er, s/layer/image/
Sep 25 15:59:45 <kergoth> udev needs glib? sheesh
Sep 25 15:59:52 <kergoth> you should be able to switch to mdev or something, no?
Sep 25 15:59:55 <kergoth> or devtmpfs
Sep 25 15:59:59 <bluelightning> hrw: that's core-image-base, I was suggesting core-image-minimal
Sep 25 16:00:20 <hrw> bluelightning: will tell after kernel/bootwrapper rebuild
Sep 25 16:00:25 <pb_> kergoth: or manage without ofono, heh
Sep 25 16:00:50 * pb_ fights a small pitched battle with boost
Sep 25 16:00:55 <hrw> Missing or unbuildable dependency chain was: ['core-image-minimal', 'packagegroup-core-boot', 'udev', 'glib-2.0', 'libffi']
Sep 25 16:01:28 <bluelightning> right, and if you really want to use core-image-base without those things, you probably want to remove bluetooth and phone from DISTRO_FEATURES
Sep 25 16:01:56 <bluelightning> hrw: ok, so it's udev
Sep 25 16:03:18 <hrw> anyway if all goes fine libffi/aarch64 should be written soon
Sep 25 16:03:51 <pb_> very good
Sep 25 16:03:54 <bluelightning> FYI I think udev itself doesn't actually need glib, only udev-acl
Sep 25 16:04:07 <bluelightning> but it's still a build-time dependency
Sep 25 16:04:15 <pb_> who came up with the name "AArch64" though?  it's one of the ugliest I've seen in I dunno how long.
Sep 25 16:07:02 <hrw> ARM Architecture 64 - probably like IA32/IA64
Sep 25 16:07:22 <hrw> AArgh64 is fine :)
Sep 25 16:09:27 <Crofton> how do I find what libs a library depends on (by inspecting binary)
Sep 25 16:09:56 <kergoth> readelf -d or objdump
Sep 25 16:10:10 <Crofton> kergoth, thanks
Sep 25 16:10:21 <Crofton> I was fishing for the right argument to readelf
Sep 25 16:11:03 <hrw> Crofton: objdump -X bin|grep NEEDED
Sep 25 16:11:36 <hrw> Crofton: objdump -X bin|grep NEEDED
Sep 25 16:11:38 <hrw> ops
Sep 25 16:12:17 <Crofton> -X?
Sep 25 16:12:54 <hrw> s/X/x/ sorry
Sep 25 16:13:06 <Crofton> yeah, just caught that
Sep 25 16:17:18 <pb_> just use "readelf -a", that works for any problem :-)
Sep 25 16:56:06 <silvio_> bb all
Sep 25 17:11:40 <khem> Crofton|work: readelf -d would dump small enough if you want to get DT_NEEDED
Sep 25 17:19:55 <otavio> have someone worked in winbind package for OE?
Sep 25 17:28:20 <bluelightning> otavio: is that separate from samba these days?
Sep 25 17:29:04 <otavio> bluelightning: in oe-core it is disabled; in Debian it is built in a separated binary package but from samba source package.
Sep 25 17:29:20 <bluelightning> otavio: you mean meta-oe ... ?
Sep 25 17:29:29 <otavio> yeah! sorry
Sep 25 17:29:43 <bluelightning> I guess we should just enable it and ensure it's packaged separately
Sep 25 17:30:33 <otavio> bluelightning: this is what I will do; but wanted to avoid to work on that if someone has done it already but did not send the patch ;-)
Sep 25 17:30:49 <bluelightning> otavio: fair enough :)
Sep 25 17:38:01 <Crofton> bluelightning, thansks for find the satellite guys
Sep 25 17:49:41 <bluelightning> Crofton: np, it was totally by accident...
Sep 25 17:50:16 <Crofton> I'm not 100% certain how to deal with them
Sep 25 17:50:46 <Crofton> the name is unfortunate, but it would be good to steer them into working more with us
Sep 25 17:50:57 <bluelightning> yes, it's a tricky one
Sep 25 18:16:00 <florian> re
**** ENDING LOGGING AT Wed Sep 26 02:59:58 2012