| 00:00.00 | Mech | kergoth: but again, you have to go looking thru each .bb to find them |
| 00:00.18 | kergoth | of course you do. every app has its own configure options. |
| 00:00.23 | Mech | kergoth: and if you have a hundred packages, with lots of different runtime configuration choices - thats a lot of files to grep :-P |
| 00:00.27 | kergoth | every app has options that no one else has |
| 00:00.45 | Mech | yes, but some options are more 'global' then others.. |
| 00:00.46 | kergoth | if mysql has a mysql specific configuration option, that doesnt belong anywhere but in the mysql metadata |
| 00:01.12 | kergoth | dood, just shut up. we've been talking about implementing something like USE since days after OE was created 2 years ago. you'll be able to do that, yes. |
| 00:01.20 | kergoth | yes, you'll have that functionality. just wait. |
| 00:01.41 | Mech | no need to be rude |
| 00:01.55 | kergoth | well i get sick of running around in circles. its a waste of your time and mine |
| 00:02.13 | kergoth | a lot of your assertions werent accurate, i corrected them. what you actually want, you'll be able to do in the future. |
| 00:03.00 | Mech | kergoth: I've spent what now - 3 weeks trying to learn the system ? If I'm running around in circles, perhaps it might show the system isn't as easy to learn as people that have been using it for years think... |
| 00:03.23 | kergoth | i've already told you that OE needs documentation |
| 00:03.27 | kergoth | what part of that do you not fucking understand? |
| 00:03.32 | kergoth | no shit. |
| 00:03.35 | kergoth | we know. |
| 00:03.42 | Mech | kergoth: it needs more then documentation.. |
| 00:03.57 | Mech | kergoth: locality of reference is totally screwed.. |
| 00:04.03 | kergoth | i'm sure you'll continue to believe that. good thing i dont give a shit |
| 00:04.16 | Mech | well, this is productive... |
| 00:04.21 | Mech | your right - its perfect |
| 00:04.27 | Mech | so why bother with NG |
| 00:04.34 | kergoth | either get off your ass and help |
| 00:04.36 | kergoth | or stop bitching |
| 00:04.39 | kergoth | complaining isnt helping anyone |
| 00:04.42 | Tmob | easy guys... |
| 00:05.11 | Tmob | Mech, u should just give a suggestion and its upto the devs if they want to do it.. |
| 00:05.17 | Mech | I was trying to help - trying to provide perspective from someone new to the system, and I offered to proof read and edit the new docs |
| 00:05.30 | Mech | what more would you like ? |
| 00:05.58 | kergoth | i'd like you to pay attention |
| 00:06.13 | kergoth | i dont like repeating myself |
| 00:06.30 | Mech | kergoth: ok - whatever.. |
| 00:06.42 | Mech | kergoth: this is counter productive |
| 00:06.54 | Mech | I'll talk to ya all later |
| 00:09.05 | reenoo_ | I'm off to bed. 'night all |
| 00:09.48 | Tmob | does oe use libelf? or some binutils tool to manipulate elf files? |
| 00:10.49 | kergoth | a couple of the bbclasses use readelf |
| 00:11.17 | kergoth | for the automatic runtime dependencies based on NEEDED shared libs |
| 00:11.19 | Tmob | kergoth, hmm ok.. and you parse the readelf output? |
| 00:11.35 | kergoth | the kernel build uses it to figure out deps between the kernel module packages as well |
| 00:11.38 | kergoth | yep |
| 00:11.43 | Tmob | ah i c.. |
| 00:13.17 | Tmob | thundershower warning.. sucks.. gonna head home before it starts to pour down |
| 00:13.24 | Tmob | later guys |
| 00:20.36 | RP | Well, I understand the problem at last. Its a race between w100fb and the backlight driver. The backlight driver is linked through the lcd control chip. If you shut one down before the other... |
| 00:20.54 | florian | good night |
| 00:21.06 | RP | How you fix it, I have no idea without linking the two drivers. Which would be frowned upon... |
| 00:29.08 | *** join/#oe mallum (~mallum@host81-155-188-174.range81-155.btcentralplus.com) |
| 00:52.53 | kergoth | damnit |
| 01:01.25 | Pendalar | I agree |
| 01:01.28 | Pendalar | damn them all |
| 01:01.47 | treke | cool. 101 is finally reopenning tomorrow |
| 01:33.24 | *** join/#oe micropal (ident@139.80-202-82.nextgentel.com) |
| 01:48.54 | *** part/#oe keturn (~kevint@c-24-21-141-200.client.comcast.net) |
| 01:50.06 | *** join/#oe keturn (~kevint@c-24-21-141-200.client.comcast.net) |
| 01:54.36 | *** join/#oe pewk (~pwk@CPE000b6a025b73-CM013439901713.cpe.net.cable.rogers.com) |
| 01:55.57 | *** join/#oe raster (~raster@vagw.valinux.co.jp) |
| 01:59.26 | *** join/#oe W8TVI (~noel@63-85-158-167.dialup.torchlake.com) |
| 02:00.32 | *** join/#oe gnufan (~mhfan@61.191.205.130) |
| 02:11.52 | *** join/#oe andersee (~andersee@codepoet.org) |
| 02:13.09 | *** join/#oe Cwiiis (~cwiiis@user-214-207-151-83.e7even.com) |
| 02:21.14 | *** join/#oe HuntsMan (~hunts@CM-lflo3-118-148.cm.vtr.net) |
| 03:33.01 | kergoth_ | nope |
| 03:33.05 | kergoth_ | er, damn mouse focus |
| 03:47.37 | *** join/#oe hufnus (~slonsiki@host-66-81-198-217.rev.o1.com) |
| 04:31.27 | chouimat|Zzzz | night |
| 04:31.36 | Mech | nite |
| 05:14.05 | *** join/#oe jh__ (~jason@69.145.84.82) |
| 05:19.33 | *** join/#oe bluelightning (~bluelight@203-173-245-158.akl.ihugultra.co.nz) |
| 05:32.41 | *** join/#oe anomaly_ (~pop@dsl-202-173-142-213.sa.westnet.com.au) |
| 05:48.29 | *** join/#oe noclouds (~mhfan@61.191.204.25) |
| 05:51.29 | *** join/#oe greyL (~bsaunder@dialport76.west.eskimo.net) |
| 06:00.29 | *** join/#oe qfh^ (~qfh@ppp-82-135-1-208.mnet-online.de) |
| 06:00.56 | *** join/#oe zecke (~ich@83.169.146.40) |
| 06:57.29 | *** join/#oe zedkatuf (~dfutak@82-32-58-45.cable.ubr08.azte.blueyonder.co.uk) |
| 07:51.32 | *** join/#oe rob_w (~roboo@p83.129.173.247.tisdip.tiscali.de) |
| 07:55.17 | *** join/#oe raster (~raster@vagw.valinux.co.jp) |
| 07:58.47 | *** join/#oe bluelightning (~bluelight@203-173-245-75.akl.ihugultra.co.nz) |
| 08:05.05 | jobi | hi |
| 08:19.31 | *** join/#oe micropal_ (ident@139.80-202-82.nextgentel.com) |
| 08:46.53 | *** join/#oe XorA (~dp@81-178-255-226.dsl.pipex.com) |
| 08:49.40 | *** join/#oe ex_newbie (~kvirc@host-62-245-140-194.customer.m-online.net) |
| 08:50.22 | ex_newbie|work | g'morning |
| 08:51.12 | XorA | morning |
| 09:06.21 | koen | 'morning all |
| 09:07.58 | hrw|work | EHLO #oe |
| 09:09.18 | XorA | STARTTLS hrw|work :-) |
| 09:10.25 | jobi | | /home/jobi/oe/build/conf/tmp/cross/lib/gcc/arm-linux/3.4.3/../../../../arm-linux/bin/ld: cannot find -lgmp |
| 09:10.46 | jobi | got that while building the gpe image for familiar |
| 09:16.02 | *** join/#oe count0 (~jmbreuer@pD951D023.dip.t-dialin.net) |
| 09:18.42 | *** join/#oe mallum (~mallum@host81-155-188-174.range81-155.btcentralplus.com) |
| 09:20.34 | *** join/#oe _Andre (~Andre@cc6918-a.ensch1.ov.home.nl) |
| 09:21.57 | hrw|work | jobi: exactly which package? |
| 09:31.50 | jobi | openswan |
| 09:33.40 | hrw|work | jobi: did it build gmp before? |
| 09:38.10 | *** join/#oe florian_kc (~fuchs@dns.kernelconcepts.de) |
| 09:38.33 | florian_kc | hi all |
| 09:48.39 | RP | morning all |
| 09:52.35 | jobi | hrw|work: yes |
| 09:59.32 | *** join/#oe z0rgl (~gottella@freeway.rd.francetelecom.com) |
| 10:00.13 | hrw|work | hi florian_kc & RP |
| 10:00.18 | z0rgl | hi |
| 10:00.21 | hrw|work | jobi: hmm.. strange.. |
| 10:00.23 | florian_kc | hi hrw|work |
| 10:00.27 | hrw|work | jobi: what machine/distro? |
| 10:00.59 | z0rgl | my build failed with : ""can't read from file '%s' (%s)", ('/.../openembedded/tmp/staging/i686-linux/kernel/kernel-abiversion'" |
| 10:01.04 | hrw|work | jobi: sometime ago it worked for me |
| 10:01.15 | hrw|work | ~oe bugs |
| 10:01.16 | XorA | z0rgl: dbus build? |
| 10:01.38 | hrw|work | z0rgl: if you report bug then include: distro, machine, package name |
| 10:01.44 | XorA | z0rgl: actually forget that, Im not awake yet |
| 10:04.57 | jobi | hrw|work: i386/Debian |
| 10:05.28 | z0rgl | XorA: MACHINE = "native" DISTRO="familiar" package initscripts-1.0-r27 |
| 10:05.51 | jobi | hrw|work: building for h3900/familiar |
| 10:06.11 | XorA | z0rgl: I got that yesterday when I mistyped my distro as a non existant one, I am assuming its because I didnt define a kernel version |
| 10:12.38 | hrw|work | XorA: kernel is defined in machines conf |
| 10:14.25 | XorA | hrw|work: Hmmm, well I defineately caused that exact error by DISTRO="openzaurus-2.5.3" by mistake, so I am ussing there is some info in the real file that is needed otherwise you get that error |
| 10:14.47 | hrw|work | ok |
| 10:16.48 | z0rgl | XorA : just retry a clean compile but I et the same error |
| 10:18.02 | XorA | z0rgl: my OE machine is turned off at moment, is there a familiar-0.[89] file? You could try it with them |
| 10:19.26 | hrw|work | XorA: in debian there is a package "wakeonlan" |
| 10:19.36 | z0rgl | XorA : ? |
| 10:19.58 | XorA | z0rgl: openembedded/conf/famailiar-0.8.conf (or sim,ilar) |
| 10:20.01 | XorA | similar |
| 10:20.17 | XorA | hrw|work: This is gentoo, I only installed 2 days ago |
| 10:22.00 | z0rgl | XorA : I have a openembedded/conf/distro/familiar-0.8.1.conf |
| 10:24.05 | *** join/#oe pb_ (~pb@2002:3e03:42cc:2:20a:5eff:fe00:391e) |
| 10:25.04 | XorA | z0rgl: I would try DISTRO="familiar-0.8.1" then, but Im not a guru of OE |
| 10:27.40 | *** join/#oe tigrux (~tigrux@charybdis-ext.suse.de) |
| 10:30.21 | cinix | ~bbcow |
| 10:30.22 | ibot | hmm... bbcow is a couple patches for bitbake and openembedded to use copy-on-write for the metadata handling. It greatly reduces memory consumption. Get them at http://www.frankengul.org/~seb/cowbb/. |
| 10:46.51 | _Andre | which of the patches on that URL should one install? |
| 10:48.37 | jobi | hrw|work: there was no -L/home/jobi/oe/build/conf/tmp/staging/arm-linux/lib in the linking of openswan |
| 10:49.48 | *** part/#oe _Andre (~Andre@cc6918-a.ensch1.ov.home.nl) |
| 10:49.55 | *** join/#oe _Andre (~Andre@cc6918-a.ensch1.ov.home.nl) |
| 10:52.24 | _Andre | I have a dependency problem when doing `bitbake opie-image`. The problem seems to be that while "libopieui2-1 (>= 1.1.8+cvs-20050113)" is needed, I get "libopieui2_1.1.8+cvs-20050113-r9_arm.ipk" being build. Note the subtle difference (a - instead of an _). |
| 10:52.41 | _Andre | Can anyone give me a clue what might be wrong? |
| 10:54.10 | *** join/#oe Securez (~securez@99.Red-80-24-222.pooles.rima-tde.net) |
| 10:54.12 | Securez | Hello |
| 10:54.28 | Securez | itry to compile sablevm for x86 but when compiling libffi it fails: |
| 10:54.39 | Securez | try to copy a path that is incorrect: |
| 10:54.54 | Securez | in ther bb file of libffi: mv ${D}/${libdir}/gcc/arm-linux/3.4.1/include/libffi/ffitarget.h ${D}/${includedir |
| 10:54.55 | Securez | } |
| 10:55.20 | Securez | but i don't have the arm-linux dir i have i686-oe-linux ones |
| 10:57.02 | Securez | can change this to something generic for all platforms |
| 10:58.18 | *** join/#oe zinga (~arnaudb@186.80-203-227.nextgentel.com) |
| 11:20.24 | *** join/#oe z0rgl (~gottella@freeway.rd.francetelecom.com) [NETSPLIT VICTIM] |
| 11:20.24 | *** join/#oe XorA (~dp@81-178-255-226.dsl.pipex.com) |
| 11:20.24 | *** join/#oe CoreDump|afk (~mhentges@hentges.net) |
| 11:20.24 | *** join/#oe Shdwdrgn (Picard@sourpuss.net) |
| 11:20.24 | *** join/#oe cyn (~cyn@adsl-4-200-252.mem.bellsouth.net) |
| 11:20.25 | *** join/#oe Twiun[aw1y] (~twiun@adorphuye.com) [NETSPLIT VICTIM] |
| 11:20.25 | *** join/#oe exastra (~go@l5.org) |
| 11:20.25 | *** join/#oe AndyCap (~aoy@host-81-191-64-77.bluecom.no) |
| 11:20.25 | *** join/#oe drw|bbl (~drw@c-67-172-218-120.client.comcast.net) |
| 11:20.25 | *** join/#oe treke|wt1 (~ggilbert@tinman.treke.net) [NETSPLIT VICTIM] |
| 11:20.25 | *** join/#oe RP (0@tim.rpsys.net) |
| 11:20.56 | *** join/#oe z0rgl (~gottella@freeway.rd.francetelecom.com) |
| 11:22.23 | RP | pb_: Have you got a minute for a kernel implemenation question? |
| 11:25.28 | robtaylor | hmm, anyone commit access to handhelds.org cvs? |
| 11:26.20 | robtaylor | someone's introduced a c99ism to ipkg, and i'd like to kill it =) seems proper to fix this upstream rather than just in oe. |
| 11:26.57 | hrw|work | robtaylor: what wrong in C99? |
| 11:27.37 | robtaylor | hrw|work: i'm building on a redhat 7.3 box =) |
| 11:28.12 | robtaylor | hrw|work: if ipkg was specified to need gcc 3.x then i wouldnt mind either |
| 11:28.39 | hrw|work | robtaylor: rh 7.3? auๆ! |
| 11:30.35 | robtaylor | hrw|work: yeah. thats what this company build on .. :/ |
| 11:31.03 | hrw|work | I worked incompany where rh 7.2 was the only one allowed to use |
| 11:31.07 | robtaylor | http://pastebin.ca/4168 for the patch |
| 11:31.10 | hrw|work | but we had gcc 3.x on them |
| 11:31.26 | robtaylor | hrw|work: it gets worse, some systems are still done on redhat 6 !! |
| 11:31.45 | hrw|work | ~seen Pigi |
| 11:31.46 | ibot | pigi <~NoOne@host188-27.pool80117.interbusiness.it> was last seen on IRC in channel #oe, 12h 35m 38s ago, saying: 'Nite all'. |
| 11:32.06 | hrw|work | robtaylor: you have to catch Pigi to change |
| 11:32.07 | robtaylor | hrw|work: yeah, this box has *vomit* 2.96 |
| 11:32.11 | robtaylor | hrw|work: hmm |
| 11:32.51 | robtaylor | hrw|work: thanks :) i'll locally patch it for now then, and keep an eye out :) |
| 11:33.11 | hrw|work | robtaylor: check does hh.org have bugzilla working |
| 11:34.15 | pb_ | RP: sure |
| 11:34.44 | pb_ | yes, bugzilla should be working. |
| 11:34.51 | pb_ | file a patch there, and assign it to Pigi. |
| 11:35.41 | robtaylor | pb_: ah, cool |
| 11:36.08 | RP | pb_: I've found my problem with the backlight driver. Its a race problem. I need some way of communicating between modules... |
| 11:36.32 | RP | The problem is the LCD code in w100fb can't suspend until the backlight driver has suspended :-/ |
| 11:37.07 | RP | I could just call the backlight driver from w100fb but then that would break allowing the backlight code as a module... |
| 11:37.26 | pb_ | Ah. Well, the traditional way to organise the ordering would be to arrange for the backlight to be a child device of w100fb, somehow. |
| 11:37.39 | RP | and then there's the fact that w100fb shouldn't really be calling a specific backlight anyhow... |
| 11:38.00 | RP | but until the w100 code gets split from the lcd code, that's just a problem we'll have to live with |
| 11:38.25 | RP | pb_: How do you do child devices with the platform bus? :-/ |
| 11:40.04 | pb_ | RP: I think you can have devices on the platform bus with dev->parent set to some arbitrary other device. |
| 11:40.19 | pb_ | iirc, platform_device_register() checks to see if dev->parent was already set, and leaves it alone if so. |
| 11:40.47 | pb_ | alternatively, you could make a bus abstraction for whatever interface the w100fb code exports to the backlight driver. |
| 11:40.47 | RP | And that would sort out the suspend calls? |
| 11:41.25 | pb_ | Probably. Suspend walks the device tree to determine the order. A parent device is only suspended once all its children are suspended, and must be resumed before any children are resumed. |
| 11:41.36 | pb_ | bbiab |
| 11:52.11 | *** join/#oe kjoe (~tody@bantha.ftk.de) |
| 12:05.22 | *** join/#oe robmv (~robert@201.248.237.158) |
| 12:18.33 | z0rgl | where does the file build/tmp/staging/i686-linux/kernel/kernel-abiversion is supposed to come ?? |
| 12:28.40 | z0rgl | I'm trying to build bootstrap-image but the buil of initscripts failed because of kernel-abiversion |
| 12:32.28 | XorA | z0rgl: out of interest, what platform you supposed to be building for? |
| 12:33.20 | z0rgl | XorA : I try to build a native version to run under vmware |
| 12:33.33 | XorA | z0rgl: ah :-) |
| 12:33.53 | z0rgl | XorA : bad idea ? :~\ |
| 12:34.17 | XorA | z0rgl: no idea, I just wondered why you were booting familiar on i686 |
| 12:41.32 | *** join/#oe micropal (ident@139.80-202-82.nextgentel.com) |
| 12:51.50 | *** join/#oe rob_w (~roboo@p83.129.211.17.tisdip.tiscali.de) |
| 13:00.37 | *** join/#oe emme (~emme@manin.usr.dico.unimi.it) |
| 13:06.30 | *** join/#oe noclouds (~mhfan@61.191.206.171) |
| 13:10.34 | *** join/#oe mrz80 (1000@caledonia.cns.ufl.edu) |
| 13:25.46 | *** join/#oe pb_ (~pb@2002:3e03:42cc:2:20a:5eff:fe00:391e) |
| 13:40.12 | AndyCap | Now I'm going to ask a stupid question. :-P Is there any "stable" release of the openembedded recipies? for instance for familiar 0.8? |
| 13:53.56 | robtaylor | AndyCap: just set DISTRO appropriately in your local.conf |
| 13:55.34 | *** join/#oe reenoo_ (~r@pD952832C.dip.t-dialin.net) |
| 13:55.43 | reenoo_ | afternoon |
| 13:55.45 | AndyCap | just realized that I'm probably doomed to use bk for that. :/ |
| 13:56.34 | *** join/#oe XEN (~XEN@d198-166-22-94.abhsia.telus.net) |
| 14:01.02 | hrw|work | robtaylor: to get exactly familiar 0.8 he has to use openembedded repo from fam 0.8 release time |
| 14:01.20 | robtaylor | hrw|work: he does? that sucks =) |
| 14:02.13 | robtaylor | hrw|work: why? |
| 14:02.47 | AndyCap | robtaylor: because the openembedded recipies are constantly moving forward. :-P (that's the general idea at least) |
| 14:03.25 | robtaylor | AndyCap: but they *are* versioned.. |
| 14:04.11 | robtaylor | woo. i just had a really crazy idea =) |
| 14:04.14 | *** join/#oe raster (~raster@YahooBB219198104088.bbtec.net) |
| 14:04.14 | hrw|work | robtaylor: and from time to time we have to change, upgrade, remove some of them |
| 14:04.26 | AndyCap | robtaylor: gpe-image is only one version though.. |
| 14:04.43 | robtaylor | hrw|work: yeah. i kinda think they shoudln;t be removed if they're used in any exsiteing items in conf/distro |
| 14:05.51 | reenoo_ | the point is that the versioned familiar.conf isn't used by anyone but koen |
| 14:06.02 | hrw|work | robtaylor: some of qpf-fonts was removed due to licence |
| 14:06.11 | robtaylor | hrw|work: uurk :( |
| 14:06.43 | AndyCap | I guess what I am asking for is a openembedded-at-familiar-0.8-rel.tar.bz2 :-P |
| 14:07.06 | robtaylor | i guess koen doesnt tag at realease time? :/ |
| 14:07.44 | chouimat | morning |
| 14:07.57 | AndyCap | robtaylor: I don't know. Sorry to say, but I don't use bk. |
| 14:08.04 | reenoo_ | hey chouimat |
| 14:08.17 | robtaylor | AndyCap: ah, well that will make life harder |
| 14:09.34 | AndyCap | and fails. :-) |
| 14:10.36 | robtaylor | AndyCap: up, yeah top secret ubuntu project ;) |
| 14:11.17 | hrw|work | robtaylor: right - koen doesnt tag |
| 14:11.25 | robtaylor | AndyCap: bazaar is good right now (bazaar.canonical.com). bazaar-ng is basically combing togther the best of tla/bazaar, darcs and bk |
| 14:11.40 | robtaylor | s/combing/combining |
| 14:12.38 | robtaylor | AndyCap: iirc canonical are paying someone to work on it |
| 14:12.59 | *** join/#oe zwi (~zwi@ewa-denver.com) |
| 14:13.11 | AndyCap | robtaylor: Now if they also convince Tortoise to make a windows integration for it, they've got it made. :P |
| 14:13.31 | robtaylor | AndyCap: see bazaar.canonical.com, already on the books =0 |
| 14:14.09 | AndyCap | robtaylor: where? :) |
| 14:15.28 | robtaylor | ok. maybe not tortoise exactly, but.. |
| 14:15.30 | robtaylor | Short list of planned features: |
| 14:15.30 | robtaylor | <PROTECTED> |
| 14:15.30 | robtaylor | <PROTECTED> |
| 14:15.39 | AndyCap | ok |
| 14:16.09 | robtaylor | if they're sane, which they usually are, they'll probably do a devel fork off tortoise |
| 14:16.30 | AndyCap | mm. Tortoise SVN works beautifully for me at least. |
| 14:18.43 | robtaylor | yeah i've used it before, and for something that's windowsy it seems pretty good |
| 14:19.03 | robtaylor | personally, give me commandline any day.. |
| 14:19.23 | AndyCap | robtaylor: well, if you're going to do windows one should do it properly. |
| 14:19.47 | RP | pb_: I think I can see a way to solve this with your parent/child suggestion. One small problem. I'm guessing calling platform_device_regsiter from another _probe function is a bad idea...? |
| 14:19.50 | robtaylor | hmm bluez-utils-2.11 is complaininga about SDP_SUPPORTED_FEATURES not defined.. anyone got a clue? |
| 14:20.04 | AndyCap | robtaylor: and saying Tortoise works for me means that the windows people commit to svn. :-P |
| 14:20.05 | robtaylor | AndyCap: oh i wholeheartedly agree |
| 14:20.27 | robtaylor | just saying for *me* *personally* ;) |
| 14:20.37 | AndyCap | ditto. :) |
| 14:20.52 | XorA | robtaylor: I hit that yesterday, didnt have time to do any looking though |
| 14:21.04 | AndyCap | mozilla and xpdf in a sea of xterms. :-) |
| 14:22.07 | robtaylor | XorA: ah, appears sdp is nw deprecated |
| 14:22.11 | robtaylor | :/ |
| 14:22.15 | robtaylor | *sigh* more work |
| 14:22.45 | reenoo_ | deprecated? |
| 14:23.13 | robtaylor | http://sourceforge.net/mailarchive/forum.php?thread_id=6220961&forum_id=1883 |
| 14:25.02 | AndyCap | service discovery deprecated? Is this how they close the security holes in bluetooth? :-P |
| 14:25.13 | robtaylor | futz, i dont think i'll be able to fix this.. who's maintainng this |
| 14:25.33 | robtaylor | (and no they havn't put a MAINTAINER field in the bb..) |
| 14:26.35 | XorA | AndyCap: sorry, I have really bad memories of working on bluetooth hardware, worst job I ever had |
| 14:26.52 | AndyCap | :> |
| 14:29.36 | robtaylor | meh, but welve been using bluez-utils-2.11 for *aaaages* |
| 14:30.56 | XorA | robtaylor: isnt it bluez-utils-dbus? |
| 14:31.21 | robtaylor | ahahahha |
| 14:31.22 | robtaylor | i know,. |
| 14:31.41 | robtaylor | it's building agains bluez-libs-2.13, rather than 2.11 |
| 14:32.19 | robtaylor | XorA: yep |
| 14:32.44 | robtaylor | can DEPENDS take versions? |
| 14:36.20 | reenoo_ | DEPENDS is matched against PROVIDES |
| 14:37.47 | robtaylor | reenoo_: is that a yes or no? |
| 14:38.08 | pb_ | RP: yes, I think that can lead to deadlock if you are not careful. |
| 14:38.30 | pb_ | you'll probably have to schedule a callback or something to register the child device. |
| 14:38.55 | reenoo_ | robtaylor: grep PROVIDES conf/bitbake.conf |
| 14:39.55 | reenoo_ | or bbread path/to/dependency.bb | grep PROVIDES |
| 14:40.30 | robtaylor | reenoo_: thanks , got it:) |
| 14:45.05 | *** join/#oe bipolar (~bflong@146.145.26.90) |
| 14:46.13 | RP | pb_: What is the best mechanism to schedule callbacks like that? |
| 14:47.37 | pb_ | schedule_work(), probably. |
| 14:47.48 | pb_ | That's what I used to solve a similar problem in the samcop touchscreen driver, anyway. |
| 14:49.52 | RP | That what I was thinking. I just wondered if there was a neater way :) |
| 14:50.12 | jobi | what is the best way to fix the missing -L.../arm-linux/lib I have in the link of openswan, when building gpe-image? |
| 14:50.47 | pb_ | patch openswan to respect ${LDFLAGS}, probably. |
| 14:58.25 | *** join/#oe ade|desk (~adavey@194.200.143.249) |
| 15:02.13 | *** join/#oe prpplague (~dave@mailhost.amltd.com) |
| 15:13.05 | jobi | they are doing ifneq ($(LD_LIBRARY_PATH),) |
| 15:13.05 | jobi | LDFLAGS=-L$(LD_LIBRARY_PATH) |
| 15:13.05 | jobi | endif |
| 15:13.19 | jobi | which erases LDFLAGS from the environment |
| 15:20.03 | rob_w | what else do i need to do to delete all previous attempts of a siongle package build vi bitbake ..so it starts that package from scratch .. -c clean did not fully do its job |
| 15:31.40 | prpplague | anyone using the new dropbear ssh client? |
| 15:33.11 | *** join/#oe robmv (~robert@201.248.237.158) |
| 15:34.09 | RP | Got there. Having the backlight synced to the framebuffer so much nicer visually as well :). Thanks for the pointer pb_ - it really helped! |
| 15:37.27 | pb_ | RP: excellent |
| 15:40.57 | RP | pb_: It raises some other questions about the corgi code though. Really the ssp and scoop interfaces should be shut down last and restarted first. That means adding in more parent pointers all over :-/ |
| 15:42.13 | RP | One alternative would be to use platform_notify in corgi.c, look out for the devices we know and then build a chain maually from there... |
| 15:42.36 | *** join/#oe Disconnect (~dis@dns.spam.rules.cuz.I.gotontheinter.net) |
| 15:42.41 | RP | platform_notify is a call back after devices are registered so we could get parent ids from there... |
| 15:42.52 | RP | morning kergoth |
| 15:43.02 | kergoth | hey |
| 15:43.48 | RP | kergoth: Found the backlight bug. It opens a whole can of worms though. Bascially we need to use the correct device suspend/resume order :-/ |
| 15:45.06 | kergoth | that can be a pain |
| 15:45.42 | RP | Just a question of the best way to do it. For each device we can add workqueue to find the parent address then reigtser the next device |
| 15:45.57 | RP | or we could use the relatively unused platform_notify callback |
| 15:46.14 | *** join/#oe marcansoft (1337@80-29-98-32.adsl.nuria.telefonica-data.net) |
| 15:47.04 | RP | That would involve code like if device xxx has registers, find out the device id and register yyy with id as the parent |
| 15:47.34 | RP | kergoth / pb_: Which is less evil? :-/ |
| 15:47.54 | pb_ | RP: heh, neither of them seems very appealing. |
| 15:48.10 | pb_ | I suppose really what you want is for dev->parent to be replaced by some kind of dev->dependencies[] vector. |
| 15:48.15 | RP | I'm tempted by the latter as it puts all the hacks into once place (corgi.c) but its not ideal... |
| 16:04.50 | *** join/#oe T`|Heli (total@GO.HELI.RI.CMU.EDU) |
| 16:09.17 | RP | This can actually be solved very simply with no dodgy functions and no changes to the drivers! |
| 16:10.18 | RP | Just set .dev->.parent in the platform structure in the first place, then register the drivers in the right order :) |
| 16:11.19 | *** join/#oe jacques (~jacques@c-67-171-198-193.client.comcast.net) |
| 16:13.57 | hrw|work | RP: it will soon be 10 lines long ;) |
| 16:15.29 | XorA | hrw|work: yeah, but youve seen the obfuscated C contest |
| 16:16.51 | RP | hrw|work: Sadly it was just code I added to fix the backlight problem but I did kill a *lot* of code from it during the rewrite :) |
| 16:17.07 | hrw|work | XorA: yes - great code |
| 16:18.01 | RP | I think I could manage an entry with the things I was trying to with waitqueues before when one line of code into the device structure would fix it! :) |
| 16:26.05 | florian_kc | A higher number for BBFILE_PRIORITY means that the collection has highter priority? |
| 16:28.22 | pb_ | yes |
| 16:28.27 | pb_ | RP: ah, very good |
| 16:32.02 | kergoth | damnit, i hate it when i get an email in a character set mutt doesnt seem to comprehend, so i see just one block of random ascii characters instead of what they wrote |
| 16:32.06 | kergoth | heh |
| 16:32.47 | florian_kc | hmm i wonder why bitbake used the wrong .bb then :-/ |
| 16:38.29 | mallum | hmm, what package sets tslib env vars ? |
| 16:39.05 | kergoth | FILES_libts = "${sysconfdir}/ts.conf ${sysconfdir}/profile.d/tslib.sh " |
| 16:39.06 | kergoth | FILES_libts += "${libdir}/*.so.* ${libdir}/ts/*.so*" |
| 16:39.38 | mallum | kergoth, thanks you sir |
| 16:39.44 | kergoth | np |
| 16:40.40 | mallum | oooh some one prettie dup ts_calibrate a little |
| 16:40.53 | kergoth | thatd be zap |
| 16:41.12 | kergoth | he fixed a number of things. made the variance filter actually usable |
| 16:41.41 | *** join/#oe joshin (~joshg@VDSL-130-13-185-112.PHNX.QWEST.NET) |
| 16:42.09 | *** join/#oe joshin1 (~joshg@VDSL-130-13-185-112.PHNX.QWEST.NET) |
| 16:43.27 | florian_kc | Do we have a fix for this pango bug? |
| 16:43.42 | mallum | florian_kc, pango bug ? |
| 16:43.53 | pb_ | florian_kc: I think it's a GTK bug, and no. |
| 16:44.02 | florian_kc | ah |
| 16:44.03 | florian_kc | oh |
| 16:44.14 | pb_ | iirc, gtk 2.4.13 with the new pango works fine |
| 16:44.37 | mallum | p-p-p-p-p-pick up a pango |
| 16:44.40 | florian_kc | Okay... sounds like i shouldn't use gtk 2.6... |
| 16:44.47 | florian_kc | *g* |
| 16:44.50 | pb_ | yeah |
| 16:45.04 | mallum | pb_, is it an arm related thing ? |
| 16:45.10 | pb_ | florian_kc: none of our code actually needs gtk 2.6 right now, so probably best to stick with 2.4 |
| 16:45.12 | pb_ | mallum: I don't know. |
| 16:45.28 | florian_kc | pb_: will do... |
| 16:47.35 | *** join/#oe treke (~ggilbert@tinman.treke.net) |
| 16:52.25 | pb_ | g'day treke |
| 16:53.14 | *** join/#oe joshin (~joshg@VDSL-130-13-185-112.PHNX.QWEST.NET) |
| 16:55.25 | treke | g'day |
| 16:55.29 | *** join/#oe joshin1 (~joshg@VDSL-130-13-185-112.PHNX.QWEST.NET) |
| 16:56.17 | *** part/#oe joshin1 (~joshg@VDSL-130-13-185-112.PHNX.QWEST.NET) |
| 16:56.43 | *** join/#oe hufnus (~slonsiki@host-66-81-182-26.rev.o1.com) |
| 17:04.42 | robtaylor | mook! |
| 17:04.43 | robtaylor | !!! Failed to move ./usr/share/pixmaps to /scratch/openembedded-build/tmp/work/gpe-package-0.0.12-r3/install/gpe-package/./usr/share/pixmaps |
| 17:04.44 | robtaylor | !!! [Errno 39] Directory not empty |
| 17:04.44 | robtaylor | NOTE: Task failed: File population failed |
| 17:04.44 | robtaylor | ERROR: package gpe-package-0.0.12-r3: task do_package failed |
| 17:04.50 | robtaylor | anyone else seen this? |
| 17:05.20 | *** part/#oe ade|desk (~adavey@194.200.143.249) |
| 17:06.30 | *** join/#oe andersee (~andersee@codepoet.org) |
| 17:19.07 | *** join/#oe dkey (~dkey@L0006P27.dipool.highway.telekom.at) |
| 17:21.15 | *** join/#oe lardman (~lardman@138.38.73.178) |
| 17:28.30 | CIA-1 | 03kergoth 07 * r1.2999 10openembedded/ (30 files in 3 dirs): Yank old unionfs versions, clean up the metadata, and add unionfs 1.0.7. |
| 17:28.34 | CIA-1 | 03kergoth 07 * r1.2998 10openembedded/packages/lemon/files/ (snprintf.patch lemon.1 snprintf.patch lemon.1): Add the files for 'lemon' that I forgot to check in yesterday. |
| 17:39.18 | robtaylor | kergoth: any idea why this would have started happening in gpe-package's do_package? http://pastebin.ca/4189 |
| 17:40.13 | CIA-1 | 03kergoth 07 * r1.2998 10/packages/lemon/files/ (snprintf.patch lemon.1): Add the files for 'lemon' that I forgot to check in yesterday. |
| 17:42.27 | *** join/#oe J (~john@cpc3-mfld2-5-1-cust17.nott.cable.ntl.com) |
| 17:42.48 | CIA-1 | 03kergoth 07 * r1.3000 10openembedded/BitKeeper/triggers/ciabot_bk.py: |
| 17:42.48 | CIA-1 | Fix a bug in the CIA notification script that resulted in files being listed |
| 17:42.48 | CIA-1 | as many times as there were revisions in the cset for that file. |
| 17:43.02 | robtaylor | heh :) |
| 17:43.35 | robtaylor | ah well, removing the FILES_${PN} += "${datadir}" from gpe-package_0.0.12.bb seems to fix it :/ |
| 17:44.31 | kergoth | thats odd |
| 17:44.52 | robtaylor | kergoth: yes it is |
| 17:45.03 | florian_kc | robtaylor: you are trying to build an old version of gpe-package. |
| 17:45.20 | robtaylor | florian_kc: building openzaurus-3.2.3 |
| 17:46.01 | robtaylor | we really have grave reproducability problems =) |
| 17:46.40 | pb_ | really? bitbake is almost completely deterministic. |
| 17:46.48 | pb_ | if you build from the same sources, you should get the same output. |
| 17:47.09 | robtaylor | pb_: as ong as you've snapshotted everything |
| 17:47.17 | robtaylor | pb_: nothing gets tagged |
| 17:47.20 | pb_ | sure. that's why releases are tagged. |
| 17:47.32 | pb_ | oh, really? I was sure mickey was tagging the OZ versions. |
| 17:47.38 | robtaylor | bitbake isnt tagged at the same time as releases, is it? |
| 17:47.42 | treke | he's definitly been tagging things |
| 17:48.40 | robtaylor | unless bitbake-1.0 was used to release oz 3.2.3 ? |
| 17:48.49 | pb_ | afaik, it should always be safe to use the latest version of bitbake. We haven't made any non-backwards-compatible changes that I can think of. |
| 17:49.15 | robtaylor | pb_: so this error isnt a bitbake change, its a oe change? |
| 17:49.30 | pb_ | which error? |
| 17:49.41 | robtaylor | pb_: http://pastebin.ca/4189 |
| 17:50.48 | treke | that could just be a matter of changing packages |
| 17:50.56 | treke | since you're using openzaurus-3.5.3 |
| 17:51.07 | treke | which hasn't really been locked down yet |
| 17:51.13 | pb_ | I can't access pastebin.ca at the moment for some reason. |
| 17:51.17 | robtaylor | ah cool, that was just what i was about to ask.. |
| 17:51.21 | robtaylor | pb_: hmm odd |
| 17:51.23 | treke | !!! Failed to move ./usr/share/pixmaps to /scratch/openembedded-build/tmp/work/gpe-package-0.0.12-r3/install/gpe-package/./usr/share/pixmaps |
| 17:51.23 | treke | !!! [Errno 39] Directory not empty |
| 17:51.28 | pb_ | oh, right |
| 17:51.32 | pb_ | yeah, that looks like an oe issue |
| 17:52.04 | pb_ | I don't know exactly what, but most of the code in question is in package.oeclass. |
| 17:52.07 | pb_ | er, bbclass |
| 17:52.34 | robtaylor | pb_: soemthing to do with mkdirhier ? |
| 17:52.53 | robtaylor | changing over to using, or a change in its semantics? |
| 17:53.41 | robtaylor | DEBUG: mkdirhier(/scratch/openembedded-build/tmp/work/gpe-package-0.0.12-r3/install/gpe-package/./usr/share/pixmaps) |
| 17:53.41 | robtaylor | DEBUG: mkdirhier(/scratch/openembedded-build/tmp/work/gpe-package-0.0.12-r3/install/gpe-package/./usr/share/applications) |
| 17:53.41 | robtaylor | DEBUG: created /scratch/openembedded-build/tmp/work/gpe-package-0.0.12-r3/install/gpe-package/./usr/share/applications |
| 17:53.41 | robtaylor | DEBUG: (/scratch/openembedded-build/tmp/work/gpe-package-0.0.12-r3/install/gpe-package/./usr/share) |
| 17:54.00 | robtaylor | !!! Failed to move ./usr/share/pixmaps to /scratch/openembedded-build/tmp/work/gpe-package-0.0.12-r3/install/gpe-package/./usr/share/pixmaps |
| 17:54.25 | robtaylor | anyway. i'll just change the package in oz3.2.3 :) |
| 17:58.48 | treke | kergoth: what was that umsdos alternative? |
| 17:59.12 | *** join/#oe HuntsMan (~hunts@CM-lflo3-118-148.cm.vtr.net) |
| 17:59.54 | *** join/#oe VoodooZ (~VoodooZ@CPE00c0f021712c-CM014110003785.cpe.net.cable.rogers.com) |
| 18:14.10 | *** join/#oe zhanght_ (~zhanght@210.76.124.126) |
| 18:17.47 | *** join/#oe minipanda (~zhanght@210.76.124.126) |
| 18:20.04 | *** join/#oe WideCharacter (~chatzilla@206.186.92.109) |
| 18:20.58 | *** join/#oe jmau_tril (~jmau@pD9E5DCA8.dip.t-dialin.net) |
| 18:29.30 | kergoth | hmm, thats a problem |
| 18:29.48 | kergoth | if you install udev on a device that expects devfs.. and the udev rules default to non-devfs naming, you've got a problem |
| 18:29.51 | kergoth | heh |
| 18:30.32 | kergoth | guess i should adapt the udev .bb to adjust its configs based on the devfs use variable |
| 18:34.51 | *** join/#oe jmau2 (~jmau@pD9E5DCA8.dip.t-dialin.net) |
| 18:35.35 | mallum | kergoth, yo |
| 18:36.00 | mallum | kergoth, could I make a bk -> svn bridge, which just regularly pulled for bk and commited in an svn repo ? |
| 18:36.12 | mallum | kergoth, would that be breaking any bk licenses ? |
| 18:37.44 | *** join/#oe zap (~zap@217.170.93.9) |
| 18:40.05 | florian_kc | how do i point oe_libinstall to a library that hides in .libs? |
| 18:41.48 | reenoo_ | <PROTECTED> |
| 18:41.54 | reenoo_ | for examples |
| 18:46.45 | florian_kc | heh okay :-) |
| 18:46.52 | treke | mallum: That would pretty likely violate the license |
| 18:48.00 | mallum | treke, but why, bk dont own the files in the repo |
| 18:49.00 | treke | you aren't allowed to contribute to their competition |
| 18:49.01 | florian_kc | mallum: that idea sounds cool |
| 18:49.13 | treke | and building a bk -> svn bridge would qualify as working on svn |
| 18:49.24 | kergoth | florian_kc: you dont need to do anything. |
| 18:49.43 | kergoth | florian_kc: point it to the dir above .libs. it looks in .libs automatically, and uses the .la to determine the soname |
| 18:50.02 | kergoth | oe_libinstall -C src .. etc |
| 18:50.29 | florian_kc | kergoth: okay... right, looks i forgot to pint it to the correct directory... |
| 18:50.35 | florian_kc | kergoth: thanks! |
| 18:50.36 | mallum | treke, *hmmfph* |
| 18:51.35 | florian_kc | ~lart bk for its sucking license |
| 18:51.54 | florian_kc | bbl |
| 18:52.47 | kergoth | mallum: read the lkml archives. that sort of thing has come up on multiple occasions |
| 18:52.50 | *** join/#oe pb_ (~pb@2002:5168:d330:1:20b:6aff:fe40:e27f) |
| 18:53.04 | mallum | kergoth, and its always poopood ? |
| 18:53.05 | *** join/#oe J_ (~john@cpc3-mfld2-5-1-cust17.nott.cable.ntl.com) |
| 18:53.23 | CIA-1 | 03kergoth 07 * r1.3001 10openembedded/packages/unionfs/ (3 files in 2 dirs): Fix the unionfs modules to build and install correctly when building for 2.6. |
| 18:54.24 | mallum | treke, so how doe th hh.org cvs work ? |
| 18:54.43 | treke | pb downloads a tarball of the oe source |
| 18:54.45 | kergoth | mallum: treke creates snapshot tarballs. cvs is updated from those |
| 18:54.48 | treke | and checks it into cvs |
| 18:55.14 | mallum | treke, where do you grab the tarball from ? |
| 18:57.08 | treke | I generate them from bitkeeper |
| 18:57.38 | treke | bk export |
| 18:58.42 | mallum | treke, so why could I not bk pull, bk export, untar, svn commit |
| 18:59.16 | treke | there are also file renames to worry about |
| 18:59.19 | treke | and things like that. |
| 18:59.32 | kergoth | its a rather large pain in the ass |
| 18:59.34 | kergoth | ~today |
| 18:59.35 | ibot | I don't know what today is, but it probably sucks. |
| 18:59.44 | treke | in my case I just create a tarball which isn't against the license at all |
| 19:00.00 | treke | once you start getting into making it work with svn you get into a grayer area |
| 19:00.12 | treke | you'd need to email mcvoy and ask him |
| 19:00.23 | kergoth | KV = "${@(bb.data.getVar('KERNEL_VERSION', d, 1) or '').split('.')[1]}" |
| 19:00.23 | kergoth | OBJ = "${@['unionfs.ko', 'unionfs.o'][bb.data.getVar('KV', d, 1) == '4']}" |
| 19:00.23 | kergoth | TARGET = "${@['unionfs2.6', 'unionfs.o'][bb.data.getVar('KV', d, 1) == '4']}" |
| 19:00.25 | kergoth | blech |
| 19:00.26 | kergoth | ugly ugly |
| 19:00.27 | mallum | treke, yeah I think I will |
| 19:00.54 | pb_ | kergoth: ugh |
| 19:01.19 | kergoth | i guess i should probably fix their makefile to accept the object as the target regardless |
| 19:01.26 | kergoth | could at least ditch the seperate TARGET var |
| 19:01.28 | treke | an svn repo would be great if it could be done |
| 19:01.30 | pb_ | good plan |
| 19:01.49 | pb_ | I guess module-base.bbclass ought to know about the .ko/.o thing. |
| 19:02.05 | pb_ | then you could write unionfs.${MODULE_SUFFIX} or some such |
| 19:02.12 | kergoth | i'd love to switch to svn instead of bk in general, but svn just plain cant handle merges back and forth amongst tons of independent branches the way bk can |
| 19:02.21 | kergoth | aha |
| 19:02.22 | treke | but I do remember something about an svn->bk gateway being written by a third party and bm not being to happy |
| 19:02.22 | mallum | treke, I'll mail him cant hurt. I have the bandwidth to set one up and share it if its legal |
| 19:02.24 | kergoth | mickey already added it |
| 19:02.25 | kergoth | KERNEL_OBJECT_SUFFIX |
| 19:02.28 | pb_ | oh, very good |
| 19:02.31 | pb_ | ~praise mickeyl |
| 19:02.33 | ibot | All hail mickeyl! |
| 19:02.44 | treke | he responds quickly :) |
| 19:03.08 | treke | every time I've ever emailed him, I've gotten a response in under 5 minutes |
| 19:04.00 | pb_ | heh |
| 19:04.21 | pb_ | I guess it doesn't take long to type "No" and hit send. |
| 19:04.58 | mallum | hehe |
| 19:05.25 | pb_ | actually, I guess he could use an autoresponder. |
| 19:05.55 | mallum | lm uses didiwiki and likes in a lot, so I may have some hope |
| 19:06.43 | CIA-1 | 03kergoth 07 * r1.3002 10openembedded/packages/unionfs/ (unionfs-modules.inc unionfs-1.0.7/make.patch): Clean up the unionfs modules build to use the KERNEL_OBJECT_SUFFIX variable. |
| 19:07.02 | kergoth | guess i should send them the patch to make their makefile suck a bit less |
| 19:07.09 | kergoth | i dont push things upstream often enough |
| 19:08.59 | treke | upstream should just make a habit of watching oe for patches |
| 19:10.11 | kergoth | hehe |
| 19:10.22 | *** join/#oe ictyl (bradford@diploid.med.unc.edu) |
| 19:10.30 | *** join/#oe micropal_ (ident@139.80-202-82.nextgentel.com) |
| 19:10.31 | kergoth | that said, a lot of our patches suck, and need cleanup before they're ready to go anywhere |
| 19:10.48 | pb_ | yeah, that's always the problem |
| 19:11.55 | pb_ | I keep wanting to push some of our autofoo-related cross compiling patches upstream, but often the patch would end up breaking things in the non-oe case. |
| 19:12.07 | kergoth | indeed |
| 19:12.50 | kergoth | the bits to add the perl script for updating config.guess/config.sub, and the autoreconf modifications, might be worth pushing |
| 19:13.01 | kergoth | the autoconf macro modifications and libtool bits are more questionable |
| 19:20.34 | pb_ | yeah |
| 19:21.03 | kergoth | i still want to patch aclocal to have an option to emit all of the copmonent .m4's that went into aclocal.m4 into a specified directory |
| 19:21.10 | pb_ | by the way, I ran into a problem with our -dev packages the other day. it turns out that paths to ${STAGING_LIBDIR} and the like are leaking into some of the generated .pc files, which isn't so good. |
| 19:21.17 | kergoth | eep |
| 19:21.20 | pb_ | not quite sure how to suppress that, though |
| 19:21.40 | kergoth | that shouldnt ever happen with normal packages, it must be the pkgconfig.bbclass's mangling? |
| 19:22.05 | pb_ | I think it happens when packages try to include the libraries that they depend on in their Libs line. |
| 19:22.29 | kergoth | which is wrong in most cases anyway, but.. |
| 19:22.30 | kergoth | heh |
| 19:22.45 | pb_ | for example, some x-using package might do "Libs: -lmylib @X_LIBS@", where X_LIBS is what configure got from `pkg-config --libs x11` |
| 19:22.54 | kergoth | ah |
| 19:22.58 | kergoth | hmm, thats annoying |
| 19:23.05 | pb_ | yeah |
| 19:23.09 | kergoth | guess we can make pkgconfig.bbclass do a second mangling, in the other direction |
| 19:23.12 | kergoth | heh |
| 19:23.30 | pb_ | heh, yes |
| 19:23.50 | pb_ | trouble is that the .pc files aren't installed under control of pkgconfig.bbclass at the moment. |
| 19:24.17 | pb_ | "make install" does that, so pkgconfig.bbclass would have to go grubbing around in the image/ directory to track them down after the fact. |
| 19:24.22 | kergoth | well, it could mangle the one in the source tree. prepend to do_compile, or insert a task between configure & compile |
| 19:24.31 | pb_ | I guess that wouldn't be too hard, though, since we know where they live. |
| 19:24.55 | kergoth | unless its test apps or something use its own .pc to find itself, but thatd be rather silly |
| 19:24.58 | kergoth | heh |
| 19:25.06 | pb_ | heh, yeah |
| 19:25.30 | pb_ | if anything, I imagine they'd use the -uninstalled version for that, which pkgconfig.bbclass shouldn't be doing anything with. |
| 19:25.40 | kergoth | ah, good point |
| 19:25.45 | pb_ | (I think it does actually stage and mangle it at the moment, which is kind of a bug in its own right, but still.) |
| 19:25.58 | kergoth | lets just do that then. no searching in image/ necessary |
| 19:26.09 | pb_ | righto, sounds good |
| 19:37.50 | *** join/#oe W8TVI (~noel@209-124-42-94.dialup.torchlake.com) |
| 19:41.04 | *** join/#oe jacques (~jacques@c-67-171-198-193.client.comcast.net) |
| 19:51.10 | *** join/#oe darkschneider (~gab@81-208-36-80.fastres.net) |
| 19:58.38 | *** join/#oe jmau (~jmau@pD9E5DCA8.dip.t-dialin.net) |
| 20:06.18 | *** join/#oe chouimat|ibook (~dieu@r2351064.cidc.net) |
| 20:06.25 | *** join/#oe chouimat (~dieu@r2351064.cidc.net) |
| 20:09.44 | *** join/#oe gints (~gints@195.2.115.160) |
| 20:10.00 | *** join/#oe nhorlock (~neil_irc@spc2-ashf1-3-0-cust73.asfd.broadband.ntl.com) |
| 20:11.04 | nhorlock | ~bbcow |
| 20:11.05 | ibot | extra, extra, read all about it, bbcow is a couple patches for bitbake and openembedded to use copy-on-write for the metadata handling. It greatly reduces memory consumption. Get them at http://www.frankengul.org/~seb/cowbb/. |
| 20:11.31 | nhorlock | ~oecow |
| 20:11.32 | ibot | rumour has it, oecow is http://www.frankengul.org/~seb/cow/ |
| 20:11.50 | kergoth | ~forget oecow |
| 20:11.51 | ibot | kergoth: i forgot oecow |
| 20:12.16 | nhorlock | I guess I should too then :-) |
| 20:12.45 | kergoth | oecow was before the core rename. |
| 20:12.48 | kergoth | its obselete |
| 20:13.01 | nhorlock | Lardman (on Zug) pointed me this way to see if it fixed by bitbake problems |
| 20:13.12 | nhorlock | s/by/my |
| 20:15.12 | nhorlock | The url has 3 patch files 9a,9b and packages.bb.cow9a I presume I need all three? |
| 20:19.34 | *** join/#oe zecke (~ich@83.169.146.40) |
| 20:21.31 | kergoth | nhorlock: no. you apply 9b to bitbake and packages.bb to openembedded |
| 20:22.22 | nhorlock | Ahhh OK thanks. |
| 20:28.12 | *** join/#oe hufnus (~slonsiki@host-66-81-176-78.rev.o1.com) |
| 20:29.22 | VoodooZ | kergoth: fyi, I changed the KERNEL_VERSION like you said in initscripts_xxx.bb file and although I got lots of those IOErrors again but it completed succesfully so... |
| 20:31.04 | kergoth | well, the errors were in the expansion of KERNEL_VERSION. by making it openzaurus specific, your KERNEL_VERSION should have defaulted to empty |
| 20:31.20 | kergoth | perhaps a KERNEL_VERSION = "" above the openzaurus one would do |
| 20:31.36 | VoodooZ | I see. I'll try that. |
| 20:31.47 | kergoth | k, thanks |
| 20:31.51 | kergoth | if that takes care of it, i'll check it in |
| 20:33.24 | nhorlock | kergoth: patches applied runnig bitbake now. time should tell if I can get to the end without getting killed. |
| 20:33.33 | VoodooZ | KERNEL_VERSION = "" |
| 20:33.33 | VoodooZ | KERNEL_VERSION_openzaurus = "${@base_read_file('${STAGING_KERNEL_DIR}/kernel-abiversion')}" |
| 20:33.39 | VoodooZ | Those seem to work fine. go ahead |
| 20:33.48 | VoodooZ | Are all those IOErrors normal though? |
| 20:33.57 | kergoth | ? |
| 20:34.03 | kergoth | you're not paying attention |
| 20:34.10 | kergoth | i just told you those errors were caused by KERNEL_VERSION |
| 20:34.14 | kergoth | they should NOT still be happening |
| 20:34.25 | kergoth | paste one of them here |
| 20:35.31 | VoodooZ | NOTE: bb.build.FuncFailed:("can't read from file '%s' (%s)", ('/local/openslug/build/tmp/staging/armeb-linux/kernel/kernel-abiversion', <exceptions.IOError instance at 0x44b27c8c>)) while evaluating: |
| 20:35.31 | VoodooZ | ${@base_read_file('/local/openslug/build/tmp/staging/armeb-linux/kernel/kernel-abiversion')} |
| 20:35.46 | kergoth | yes. base_read_file. it shouldnt be expanding that at all |
| 20:35.50 | kergoth | strange. |
| 20:35.54 | VoodooZ | yep. |
| 20:36.07 | VoodooZ | It doesn't fail for me but I thought I'd let you know. |
| 20:36.09 | kergoth | ah well, i'll check it in, better than nothing |
| 20:36.26 | kergoth | surprised nobody noticed. guess it depended on the build order |
| 20:36.33 | VoodooZ | scary |
| 20:36.35 | kergoth | if the kernel was built, by chance, before initscripts, it wouldve worked |
| 20:36.42 | kergoth | because the abiversion file wouldve been there |
| 20:36.46 | VoodooZ | ic |
| 20:37.01 | VoodooZ | kernel is built right after in my case. |
| 20:37.10 | kergoth | its purely OZ specific. need to find a better way to get those bits ignored for everyone else |
| 20:37.17 | VoodooZ | actually, there's a few packages in between. |
| 20:37.26 | VoodooZ | cool. thanks |
| 20:42.11 | kergoth | VoodooZ: can you test the fix before your kernel build stages? |
| 20:42.21 | kergoth | VoodooZ: http://hyperion.kergoth.com/initscripts_1.0.bb |
| 20:42.26 | kergoth | see if the errors go away |
| 20:43.10 | VoodooZ | you want me replace mine with this? |
| 20:43.28 | VoodooZ | What do you want me to rebuild though? glibc takes forever... |
| 20:44.56 | kergoth | just rebuild initscripts |
| 20:44.59 | kergoth | using that .bb |
| 20:45.01 | VoodooZ | cool |
| 20:45.04 | kergoth | make sure the IOErrors go away |
| 20:45.17 | VoodooZ | cleaning... |
| 20:45.27 | kergoth | k |
| 20:45.28 | VoodooZ | building... |
| 20:45.46 | VoodooZ | still errors |
| 20:46.06 | VoodooZ | keep in mind that my kernel package started compiling but never completed. |
| 20:46.11 | kergoth | run bbread|grep ^DISTRO |
| 20:46.25 | VoodooZ | DISTRO="openslug" |
| 20:46.41 | kergoth | that doesnt make any sense. |
| 20:46.52 | kergoth | the KERNEL_VERSION is inside of a function append that only happens when you're openzaurus |
| 20:46.54 | VoodooZ | the story of my life! |
| 20:46.56 | kergoth | wtf |
| 20:47.01 | kergoth | do_install_append_openzaurus () { |
| 20:47.01 | kergoth | ยปยทยทยทยทยทยทยทcat ${WORKDIR}/checkversion | sed -e "s,VERSION,${@base_read_file('${STAGING_KERNEL_DIR}/kernel-abiversion')}-${DISTRO_VERSION}," > ${D}/etc/init.d/checkversion |
| 20:47.14 | kergoth | and bitbake only expands vars when it has to.. hmm |
| 20:47.38 | kergoth | oh, shit.. thats right.. ugh |
| 20:47.45 | VoodooZ | what? |
| 20:47.48 | kergoth | when bitbake emits a shell script, it emits _all_ functions |
| 20:48.10 | kergoth | i wonder if its trying to emit that openzaurus specific chunk of a function |
| 20:48.12 | pb_ | oh yeah |
| 20:48.13 | pb_ | heh |
| 20:48.34 | kergoth | really, the _append handling code should have deleted the original version |
| 20:48.36 | pb_ | I guess do_install_append_blahblah() is a function in its own right, as well as getting glommed onto the main do_install. |
| 20:48.39 | pb_ | kergoth: yeah |
| 20:49.30 | kergoth | VoodooZ: you just discovered a bitbake bug. congrats |
| 20:49.31 | kergoth | :P |
| 20:49.37 | VoodooZ | yipee!! |
| 20:49.45 | VoodooZ | I should get paid for this! :) |
| 20:49.48 | jacques | VoodooZ is good at finding bugs. |
| 20:49.54 | VoodooZ | damn right I am! |
| 20:50.42 | VoodooZ | :) |
| 20:50.42 | VoodooZ | My mom always said I was only good at breaking stuff! :) |
| 20:51.01 | VoodooZ | Let me know if I can assist. |
| 20:52.56 | nhorlock | kergoth: hehe COW patches have got me further.. bitbake now manages to build the cache :-) thanks |
| 20:54.09 | *** join/#oe oe_newbie (~kvirc@195.234.128.72.cablesurf.de) |
| 20:54.28 | kergoth | nhorlock: you can disable the cache... just itll have to reparse everything every time |
| 20:54.31 | kergoth | heh |
| 20:55.34 | nhorlock | given it took my machine half an hour to build the cache I think I'll stick with it :-) time to upgrade my poor old build machine I guess. |
| 20:55.57 | treke | if it takes a half hour to build the cache, you should probably just delete oe and not proceed further |
| 20:56.26 | nhorlock | :-( |
| 20:57.16 | kergoth | heh, or go buy some RAM |
| 20:57.35 | nhorlock | build server = P2 350Mhz with 192M, it is due a holiday :-) |
| 20:58.13 | hrw|tv | cool machine for oe... |
| 20:59.16 | hrw|tv | nhorlock: disable psyco as it take ram too |
| 20:59.50 | nhorlock | hrw|tv: ... and build overnight :-) |
| 21:00.43 | hrw|tv | overweek rather :) |
| 21:02.25 | kergoth | hehe |
| 21:02.33 | kergoth | good point about psyco, i always forget about that |
| 21:04.12 | hrw|tv | nhorlock: better build glibc using bitbake -b (or it take eons to get packaged) |
| 21:05.48 | nhorlock | thanks I'll remember that. I am thinking I might put this on hold until I can get a better machine. |
| 21:07.07 | kergoth | oe wont build under cygwin without a bit of tender loving care |
| 21:07.10 | kergoth | however, you could use coLinux. |
| 21:07.17 | kergoth | ~colinux |
| 21:07.18 | ibot | it has been said that colinux is at http://www.colinux.org |
| 21:08.07 | nhorlock | I was thinking of wiping it :-), just wait 'til it blue screens again....... |
| 21:08.24 | kergoth | colinux is a linux kernel that runs under windows. no emulation, no vmware |
| 21:08.28 | hrw|tv | or connect hdd from p2 and boot linux |
| 21:08.30 | kergoth | you can run a normal linux distro with it |
| 21:08.41 | kergoth | pretty fast, too |
| 21:09.48 | nhorlock | thanks for the ideas guys, the hdd from the p2 idea is sounding quite good to me. |
| 21:11.27 | nhorlock | beleive it or not it isn't the oldest machine I have here. I have a Pentium 90Mhz running Mandrake 10 :-) It just rsyncs my webservers to give me a local backup. |
| 21:11.37 | hrw|tv | kergoth: about colinux... does it need cygwin xfree to run x11 apps inside colinux? |
| 21:13.24 | kergoth | hrw|tv: i doubt it. never tried using X |
| 21:13.43 | kergoth | right now i'm using an OE built rootfs in it :) |
| 21:13.46 | kergoth | bootstrap, even |
| 21:14.55 | hrw|tv | FAQ: "If you want to use X, you can run a Windows X server such as Cygwin/X, and launch an X session across the virtual network." |
| 21:15.00 | hrw|tv | ':( |
| 21:15.41 | *** join/#oe chouimat|ibook (~dieu@r2351064.cidc.net) |
| 21:16.31 | kergoth | aw, that sucks |
| 21:16.45 | kergoth | still quite nice though |
| 21:16.52 | kergoth | good for people who are stuck on windows |
| 21:16.57 | pb_ | yeah |
| 21:17.16 | pb_ | running x inside colinux would be a complete nightmare. xfree86 would fight windows for the graphics card, and it would all end in tears. |
| 21:17.38 | nhorlock | I'm going get some dinner and dream of building oe. catch you later. thanks for the help |
| 21:18.21 | *** join/#oe Pigi (~NoOne@host188-27.pool80117.interbusiness.it) |
| 21:18.27 | Pigi | Ciao all |
| 21:19.26 | pb_ | ugh, debugging a 25MB DSO on the ipaq is no fun |
| 21:19.29 | pb_ | hi Pigi |
| 21:19.47 | Pigi | what's a DSO ? |
| 21:19.53 | pb_ | shared library |
| 21:20.06 | pb_ | or "dynamic shared object" |
| 21:20.27 | Pigi | Wowowowowo What a nice job :) ( never called this way, only shared object :) ) |
| 21:20.33 | pb_ | or, in this case, libgtk. |
| 21:21.36 | Pigi | Is it possible that the newest build ( gcc 3.4 ) are bigger in size than the oldest one ? |
| 21:21.51 | pb_ | It's possible, but that shouldn't (in general) be the case. |
| 21:22.34 | *** join/#oe rob_w (~roboo@p83.129.211.17.tisdip.tiscali.de) |
| 21:22.50 | Pigi | I have notice that installing kopi/kapi in 0.8 the space needed is 5+megs, while in 0.7 was surely less. |
| 21:36.35 | pb_ | mm, dunno. I guess for C++ it's possible that gcc-3.x code is bigger than gcc-2.x code. |
| 21:36.50 | pb_ | you'd probably need to take that up with the #opie crew. |
| 21:36.52 | hrw|tv | Pigi: kdepim/pi is a beast... |
| 21:37.01 | kergoth | hmm |
| 21:37.02 | hrw|tv | Pigi: its not opie stuff |
| 21:37.09 | hrw|tv | s/pigi/pb |
| 21:37.14 | kergoth | this register in this ts controller that you can read to find out whether the pen is down or not |
| 21:37.19 | kergoth | reports that it isnt, at random, when it is |
| 21:37.20 | kergoth | :\ |
| 21:37.24 | pb_ | doh |
| 21:37.31 | kergoth | spurious press/release events are no fun |
| 21:37.46 | Pigi | noclouds, I know, but it's really easy to sync with kde apps :) |
| 21:37.59 | Pigi | s/noclouds/no |
| 21:47.20 | *** join/#oe dkey| (~dkey@L0005P04.dipool.highway.telekom.at) |
| 21:51.53 | *** part/#oe ictyl (bradford@diploid.med.unc.edu) |
| 21:52.57 | hrw|tv | hmm.. I compiled kdepim/pi with -lsupc++ instead of -lstdc++ and they are bigger but dont want stdc++ |
| 21:53.13 | hrw|tv | so for systems without libstdc++ installed they will be better |
| 21:55.05 | Pigi | supc++ ? |
| 21:56.01 | kergoth | Pigi: supplementary c++ lib. includes some things that are needed if you dont use g++ and/or link in libstdc++ manually |
| 21:56.48 | Pigi | wow. is her in oe ? |
| 21:56.55 | hrw|tv | Pigi: binary get little bigger but dont depend on 3.5MiB libstdc++ |
| 21:56.57 | kergoth | ? |
| 21:57.09 | hrw|tv | Pigi: it's part of gcc |
| 21:57.18 | kergoth | hrw|tv: thats cool. all our qt c++ apps should be linked with gcc and linked to libsupc++ |
| 21:57.24 | Pigi | Ok. apologize. |
| 21:57.57 | hrw|tv | kergoth: dont know does every will link then |
| 21:58.10 | Pigi | never heard about. Should google a bit. |
| 21:58.32 | kergoth | hrw|tv: well, all of our qt c++ apps that dont use stdc++. very very few do, and those that do should be fixed |
| 21:58.35 | pb_ | dammit, now my ipaq is crashed. |
| 21:58.39 | kergoth | doh |
| 21:59.03 | pb_ | evidently loading 20MB of debugging data into gdb was too much for its tiny mind. |
| 21:59.22 | Pigi | memory fault, core..... ehm ... core..... uhm.... Damn, I forgot it !!!! |
| 21:59.45 | pb_ | if only I could get this weird behaviour to happen on my desktop, I'd be all set. |
| 22:00.24 | kergoth | cant you split out the debugging info and load that into the gdb on the desktop end? use gdb server? |
| 22:00.46 | pb_ | gdbserver doesn't play well with shared libraries |
| 22:01.01 | kergoth | ah |
| 22:01.03 | kergoth | thats unfortunate |
| 22:01.14 | pb_ | native gdb does some special stuff involving breakpoints in ld.so to make debugging of DSOs work right. |
| 22:01.27 | pb_ | that's why you need a non-stripped ld.so. |
| 22:01.41 | Pigi | hrw|tv, application linked against libsupc++, are linked statically, or the libsupc++ is needed ? |
| 22:01.56 | Pigi | And in case, how big is libsupc++ ? |
| 22:02.06 | pb_ | I guess I could go to the office and get the 128MB ipaq. |
| 22:02.29 | hrw|tv | Pigi: static |
| 22:03.13 | Pigi | ok, That's was my tought. Could you package those kapi/kopi and put somewhere to give a look ? |
| 22:03.38 | Pigi | or, advice me on how to recompile in oe with libsupc++ ^ |
| 22:03.46 | Pigi | s/^/? |
| 22:04.00 | hrw|tv | Pigi: edit kdepimpi-base.inc and change -lstdc++ -> -lsupc++ and rebuild |
| 22:04.21 | Pigi | ok. ths a lot |
| 22:09.06 | mickeyl | mmmh this is great |
| 22:09.07 | kergoth | hey mickeyl |
| 22:09.11 | mickeyl | hi |
| 22:09.28 | hrw|tv | mickeyl: opie-kbill use stl ;( |
| 22:09.31 | mickeyl | i've just been greeted by a nice full screen logo using ELPP on 2.6 collie |
| 22:09.36 | mickeyl | s/collie/shepherd/ |
| 22:09.40 | mickeyl | this looks wonderful |
| 22:10.01 | mickeyl | hrw|tv: unfortunate. overwrite QMAKE_LINK like we do on the few other occasions where this is needed |
| 22:10.02 | kergoth | ELPP? |
| 22:10.09 | mickeyl | Enhanced Linux Progress Patch |
| 22:10.15 | kergoth | ah |
| 22:10.17 | kergoth | cool |
| 22:10.27 | mickeyl | oz 3.6.0 will rule |
| 22:10.28 | mickeyl | simply rule |
| 22:10.29 | mickeyl | :D |
| 22:10.30 | hrw|tv | mickeyl: I know |
| 22:10.59 | hrw|tv | mickeyl: when we will have 2.6 on collie too... ;) |
| 22:11.02 | mickeyl | hrw|tv: or if it's just a couple of instructions, rewrite? :) |
| 22:11.05 | mickeyl | heh |
| 22:11.09 | mickeyl | need to pester john lenz a bit |
| 22:11.19 | mickeyl | he should look what richard is doing heh |
| 22:11.22 | kergoth | i need to get motivated to actually work on tosa |
| 22:11.24 | kergoth | heh |
| 22:11.33 | mickeyl | kergoth: that'd rock |
| 22:11.41 | mickeyl | shouldn't be that hard with the prework of richard or should it? |
| 22:11.55 | kergoth | there's some substantial differences. |
| 22:12.51 | CosmicPenguin | What does ELPP have over LPP? |
| 22:13.04 | hrw|tv | kergoth: other gfx chip for example? |
| 22:13.10 | mickeyl | CosmicPenguin: lpp maintainer gave up |
| 22:13.22 | mickeyl | new maintainer just calls it elpp |
| 22:14.24 | mickeyl | it looks much less intrusive than bootsplash and friends thoug |
| 22:14.30 | mickeyl | and applied cleanly to our shepherd 2.6.10 |
| 22:14.36 | mickeyl | and it just works :D |
| 22:14.37 | mickeyl | this is why i like it |
| 22:14.46 | hrw|tv | hmm.. I should write again to subapplet author to get info what he thinks about my recent patches |
| 22:15.05 | kergoth | hmm |
| 22:15.08 | kergoth | current uclibc cvs doesnt like me |
| 22:16.49 | *** join/#oe beewoolie (~beewoolie@florence.buici.com) |
| 22:18.51 | hrw|tv | ok. some apps doesnt like supc++ instead of stdc++ - will leave it for other day anyway |
| 22:20.14 | hrw|tv | time to sleep |
| 22:21.03 | kergoth | night hrw|tv |
| 22:21.11 | hrw|tv | its good that simple change to palmtop.bbclass allow to do that replace ;) |
| 22:25.50 | mickeyl | yeah |
| 22:25.53 | mickeyl | g'night hrw|tv |
| 22:25.58 | hrw|tv | bye |
| 22:26.22 | mickeyl | hmm |
| 22:26.37 | mickeyl | who added that bootlogd btw.? apart from the fact that it doesn't work it's insane having that in the default image |
| 22:27.03 | kergoth | bootlogd is quite nice when it works, but yeah, its more useful as a debugging tool than anything else |
| 22:27.28 | mickeyl | ok, we have some more places where we should conditionally apply patches and stuff |
| 22:27.36 | hrw|gone | ChangeSet@1.2648.78.21, 2004-12-04 22:42:39+00:00, pb@stealth.nexus.co.uk |
| 22:27.36 | hrw|gone | <PROTECTED> |
| 22:27.39 | hrw|gone | maybe? |
| 22:27.40 | hrw|gone | bye |
| 22:27.45 | mickeyl | e.g. the rootpw patch for dropbear |
| 22:27.49 | mickeyl | which should vanish in release images |
| 22:27.57 | mickeyl | any idea how to deal with that in a generic manner? |
| 22:28.03 | mickeyl | like DISTRO_TYPE = "debug" |
| 22:28.07 | kergoth | nod |
| 22:28.07 | mickeyl | and DISTRO_TYPE = "release" ? |
| 22:28.08 | kergoth | hmm |
| 22:28.16 | mickeyl | or a better idea? |
| 22:28.52 | mickeyl | please give it a thought, we should do something like that asap |
| 22:28.55 | mickeyl | bbl, wifey calls |
| 22:40.38 | *** join/#oe Timelord (~TL@64.147.65.204) |
| 22:45.35 | kergoth | beewoolie: where can i get at the APEX code? might want to try to get it working on some of the OMAP platforms I have at work. |
| 22:47.10 | beewoolie | kergoth: http://projects.buici.com/apex |
| 22:47.40 | beewoolie | kergoth: it took me about 3 hours to get the ixp42x code written. It isn't debugged, but it ought to be sufficient. |
| 22:48.03 | kergoth | cool |
| 22:55.59 | Pigi | time to sleep |
| 22:56.02 | Pigi | Nite all |
| 22:57.40 | kergoth | hrm |
| 22:57.48 | kergoth | doxygen doesnt want to do macro expansion for me |
| 23:00.00 | *** join/#oe CoreDump|home (~mhentges@hentges.net) |
| 23:01.46 | Pendalar | this has been a fun week |
| 23:02.25 | CIA-1 | 03kergoth * r105 10bitbake-ng/doc/ (Makefile.am doxygen.cfg): Update doxygen config to clean things up a bit, and build the doxygen docs by default. |
| 23:02.37 | Pendalar | car broke down, fight with mother dearest, got reported to family and children services, got investigated by family and children services, having to install carpet on a 30x30 area.. alone.. in under 4 hours |
| 23:03.24 | kergoth | Pendalar: ack |
| 23:03.26 | CIA-1 | 03kergoth * r106 10bitbake-ng/ (6 files in 4 dirs): Updates to the doxygen markup in various files. |
| 23:03.47 | Pendalar | oh and add to that glibc now fails at patcher like it did a last week |
| 23:03.53 | Pendalar | the latter bugs me most |
| 23:04.26 | CIA-1 | 03kergoth * r107 10bitbake-ng/ (bitbake-0.0.pc.in configure.ac): Fix PACKAGE_VERSION per the version m4 changes. |
| 23:05.15 | Pendalar | I know you generally don't do this sort of thing, but could you venture into my system later this weekend and see what's wrong? |
| 23:05.26 | Pendalar | I assume it's some RH9 version of something or other |
| 23:05.55 | Pendalar | but I believe I still meet the min requirements, using mostly the same local.conf as before (other than rid of various oe* vars and updated them to the bb equiv) |
| 23:28.44 | *** join/#oe Ikkakujyu (~kjara@ip68-8-220-23.sd.sd.cox.net) |
| 23:40.57 | CIA-1 | 03pb 07 * r1.3000.1.1 10openembedded/packages/sysvinit/sysvinit_2.86.bb: banish bootlogd to eponymous package |
| 23:42.07 | CosmicPenguin | eponmous? |
| 23:42.53 | pb_ | no, eponymous |
| 23:43.05 | CIA-1 | 03kergoth * r108 10bitbake-ng/ (5 files in 3 dirs): Adjust our doxygen usage. |
| 23:46.48 | pb_ | hrm |
| 23:46.49 | pb_ | <PROTECTED> |
| 23:46.49 | pb_ | new_keyboard_dir = GTK_TEXT_DIR_LTR; |
| 23:46.49 | pb_ | <PROTECTED> |
| 23:46.49 | pb_ | new_keyboard_dir = GTK_TEXT_DIR_RTL; |
| 23:46.55 | pb_ | that doesn't look very good |
| 23:47.25 | pb_ | I bet that's why gtk keeps switching to RTL mode on the ipaq. |
| 23:52.36 | ex_newbie|home | nite folks |