**** BEGIN LOGGING AT Thu Jun 14 02:59:57 2007 Jun 14 04:55:05 03pfalcon 07org.oe.dev * r443650a5... 10/ (3 files in 2 dirs): libmimedir: Fix packaging. Jun 14 05:43:45 morning Jun 14 06:53:39 v8jlene: it sort of builds with your 'fix' Jun 14 06:53:45 but it certainly is strange Jun 14 06:54:34 oxo: Got visitors at the moment... I have what I hope is a proper fix, pushing now (been waiting for monotone to work properly).. be back in an while. Jun 14 06:55:16 03lenehan 07org.oe.dev * r7565aee4... 10/ (3 files in 3 dirs): Jun 14 06:55:16 perl 5.8.8: Fix the regexp for removing /usr/local paths - it was eating the Jun 14 06:55:16 seperating space in some cases. Report as part of #2483. Jun 14 06:55:20 03lenehan 07org.oe.dev * r4a5a6f45... 10/ (4 files in 3 dirs): Jun 14 06:55:20 perl 5.8.8: Directly patch the files that need patching instead of having a Jun 14 06:55:20 makefile within the perl build apply the patches. This was a left over from Jun 14 06:55:20 the previous perl version and makes modifying these patches a lot more Jun 14 06:55:23 difficult than it really needs to be. Jun 14 06:55:24 03lenehan 07org.oe.dev * r47eb1f06... 10/ (1 packages/perl/perl-5.8.8/Makefile.SH.patch): (log message trimmed) Jun 14 06:55:27 perl 5.8.8: Don't set LD_LIBRARY_PATH when running miniperl. This was being Jun 14 06:55:29 set to point at the current perl build to find the shared perl library, Jun 14 06:55:31 however in the OE case we actually want the staged native version not those Jun 14 06:55:33 that belong to the build for the target. The miniperl binary actually has an Jun 14 06:55:35 rpath pointing at the staging dir so no LD_LIBRARY_PATH is required. It Jun 14 06:55:37 seems gentoo has started letting LD_LIBRARY_PATH override rpath which is Jun 14 07:01:24 v8jlene: thank you very much Jun 14 07:15:57 goooooooood morning all Jun 14 07:20:01 morning koen Jun 14 07:20:12 hey mickey_fic Jun 14 07:20:40 mickey_fic: will the svn overlay go away once you have branched? Jun 14 07:21:06 sure. there's no point in keeping it then# Jun 14 07:23:11 mickey_fic: I'm planning to add some moko bits to the clearlooks gtkrc so people can use the rrsreader and packagamanager outside of openmoko-image Jun 14 07:23:30 currecntly it works, but things like the search entry are a bit ugly Jun 14 07:24:25 cool Jun 14 07:24:48 it's a lot faster as well :) Jun 14 07:57:26 oxo: Back now, hopefully that fix will work for you (and not break anything else!). Since the LD_LIBRARY_PATH thing shouldn't be needed at all during the build I removed it completely... it looks like it should resolve the issue but you never know! Jun 14 07:58:00 * oxo will check soon Jun 14 07:58:20 it's now building an angstrom-x11-image which will (hopefully) run on my h3600 Jun 14 08:23:03 I hate it when broken autotools macros cause stuff like this: Jun 14 08:23:04 ui.c:564: warning: implicit declaration of function 'rpl_malloc' Jun 14 08:23:12 morning Jun 14 08:23:19 hey hrw Jun 14 08:25:19 morning all Jun 14 08:25:51 hey ade|desk Jun 14 08:34:58 CIA-3: wb Jun 14 08:35:16 CIA-3: morning :) Jun 14 08:35:53 hi zecke Jun 14 08:36:19 hey zecke Jun 14 08:37:14 03koen 07org.oe.dev * rb24e3d33... 10/ (3 files in 2 dirs): parted: add 1.8.7 Jun 14 08:37:44 ~seen florian Jun 14 08:38:03 florian was last seen on IRC in channel #openmoko, 18h 5m 38s ago, saying: 'sxpert-work: between taiwan and germany rather ;)'. Jun 14 08:38:30 mickey_fic: http://svn.o-hand.com/view/poky?rev=1932&view=rev is interesting for OE Jun 14 08:38:58 mickey_fic: it allows one person to build meta-toolchain and other people won't have to build the crosscompiler and gcc anymore in OE Jun 14 08:39:17 interesting! Jun 14 08:39:20 ;) Jun 14 08:39:36 hats off to hrw and RP, of course Jun 14 08:39:51 OH have also few other interesting things ;) Jun 14 08:40:05 like labquests :) Jun 14 08:40:12 (with holes) Jun 14 08:40:33 hats off to florian too for his work on OEOE OE Jun 14 08:40:40 hats off to florian too for his work on OOE SDK Jun 14 08:41:09 There are some nasty issues exposed by that that external toolchain code :/ Jun 14 08:41:59 Like the pkgdata and pkgmaps directories aren't arch specific and that the sdk class is apparently i686-poky-gnueabi :/ Jun 14 08:42:23 ~curse ATI and DFI Jun 14 08:42:23 May you be reincarnated as a Windows XP administrator, ATI and DFI ! Jun 14 08:42:24 * koen pets his awk script script to convert quilt series into OE SRC_URIs Jun 14 08:42:51 koen: it is :s in vim ;D Jun 14 08:43:13 :s+^$.*$+file://\1;patch=1 \\+g Jun 14 08:43:22 koen: I'm running a build now to see if anything is missing from that to get working images. I suspect we need a few more ipks installed but apart from that it should work... Jun 14 08:43:45 * hrw has a build without ipkg usage - only dpkg+apt-native Jun 14 08:44:00 fsck... Jun 14 08:44:18 RP: I'd like to make meta-toolchain work with debs as well Jun 14 08:44:30 2xserial+1lpt PCI card does not show in lspci.. onboard gigabit ethernet does not work.. argh... Jun 14 08:44:33 brb Jun 14 08:44:38 RP: since that opens up a way for packaged-staging :) Jun 14 08:45:04 koen: I think staging is goiong to need separate packages to be honest :/ Jun 14 08:45:34 even after the recent .la and .pc fixups? Jun 14 08:47:18 koen: Yes, at least to contain the pkgmaps/pkgdata/shlibs stuff Jun 14 08:48:33 heh.. each time when I use X11 on laptop I am getting more angry Jun 14 08:48:35 ah, that kind of different packages Jun 14 08:48:46 it looks much better then on my desktop... fscking ATI Jun 14 08:49:17 koen: If you add them, you might as well generate special staging packages though... Jun 14 08:49:35 ${PN}-stagingdata ? Jun 14 08:50:08 stored in tmp/deploy/ipk/staging/ dir? Jun 14 08:50:39 My goal is to be able to point OE to an online feed and have it Just Work(TM) Jun 14 08:51:19 o yes! Jun 14 08:51:38 koen: Someday :) Jun 14 08:52:09 "hi vendor - here is 5 lines long instruction how to build xserver for your device" Jun 14 08:52:31 hm... Poky instruction is 3 lines... Jun 14 08:52:32 ;) Jun 14 08:52:33 hrw: Threee with poky already ;-) Jun 14 08:52:46 fetch, init, bitbake Jun 14 08:52:54 right :) Jun 14 08:53:57 what mac to set... Jun 14 08:54:38 AB:AD:DE:AD:C0:FF:EE Jun 14 08:54:48 ops.. too long Jun 14 08:55:01 C0:FF:EE:C0:FF:EE is nice ;) Jun 14 08:57:45 * hrw want composite on X11... Jun 14 09:11:32 hi woglinde Jun 14 09:11:41 hi lrg woglinde Jun 14 09:11:55 mickey_fic: how goes openmoko-from-dev build? Jun 14 09:11:59 woglinde: hey Jun 14 09:12:35 hrw: fairly well. i did an angstrom build and although it looked freaky (font/theming messed up), nearly everything worked. Jun 14 09:12:51 hi zecke and mickey Jun 14 09:12:55 i have started pushing our changes into .dev Jun 14 09:12:55 hi hrw Jun 14 09:13:17 mickey_fic: it's exhibiting the problems Thomas was talking about, the openmoko theme isn't selected anywhere Jun 14 09:13:48 hi koen Jun 14 09:13:55 hey woglinde Jun 14 09:14:15 mickey_fic: how you plan to integrate OM bitbake changes? using 1.6 with .dev is going to be dropped and builds with 1.8 are much faster Jun 14 09:14:46 hrw: that depends on the availability of said changes :/ Jun 14 09:14:48 and SRCDATE=now causes parsing to take *hours* Jun 14 09:15:38 * koen mumbles about hiring $random dude to do bitbake work instead of bitbake hackers Jun 14 09:15:39 my record is 47 minutes to parse with openmoko Jun 14 09:16:01 koen: heh, it Jun 14 09:16:12 's not like there are too much free bitbake hackers available Jun 14 09:16:23 all i know are booked Jun 14 09:16:35 you hired o-hand, which employs RP Jun 14 09:16:41 and hrw Jun 14 09:17:02 koen: and we both are busy with !fic projects Jun 14 09:17:05 *g* Jun 14 09:17:22 so FIC needs to pay more :) Jun 14 09:17:33 heh Jun 14 09:18:05 i think our midterm strategy is to emply one dude that does nothing but maintaining OE and bitbake Jun 14 09:18:15 it needs to be someone senior though Jun 14 09:18:23 where shoud that guy come from? Jun 14 09:18:28 we need a training lab... Jun 14 09:18:36 hm Jun 14 09:18:59 * zecke needs his clone army Jun 14 09:19:03 ya Jun 14 09:19:07 get married and raise chidren Jun 14 09:19:19 *hust* Jun 14 09:19:19 but that may take too long for us to benefit Jun 14 09:19:53 plus you loose the existing dev that made the child because of sleepless nights Jun 14 09:20:01 hi hrw, mickey_fic, koen Jun 14 09:21:26 mickey_fic: so where are your childs? Jun 14 09:21:35 mickey_fic: or forks ;) Jun 14 09:22:16 but bevare that they can became zombies.. Jun 14 09:24:09 heh Jun 14 09:24:26 hey lrg Jun 14 09:24:33 I have plans for next generation Jun 14 09:26:18 Sabine and me will discuss this after move is finished Jun 14 09:26:23 i.e. in few weeks Jun 14 09:26:50 mickey_fic: she wants or you both?\ Jun 14 09:27:12 tbh, we don't know Jun 14 09:27:30 OT: what is error -512 from nfs-server while mounting an nfs-root? Jun 14 09:27:53 mickeyl hm uh Jun 14 09:28:24 maybee authentication with hosts.allow Jun 14 09:28:27 -13 is permission denied Jun 14 09:28:31 but 512... Jun 14 09:28:33 its longtime ago Jun 14 09:28:46 I hit this error Jun 14 09:29:36 ah Jun 14 09:29:36 no Jun 14 09:29:49 its the configfile /etc/exports Jun 14 09:29:54 I remeber Jun 14 09:30:28 mickey_fic: I'd like to get something added to bitbake 1.8 that solves your problem. Henryk never replied to my emails though :/ Jun 14 09:31:41 RP: could you please send them again and cc Harald? I make sure he*ll pester him Jun 14 09:32:02 (laforge@openmoko.org) Jun 14 09:32:16 ccing me won't hurt either Jun 14 09:32:25 mickey_fic: ok Jun 14 09:32:27 thanks Jun 14 09:32:43 * koen receives a big-ass ups parcel containing just a flash card Jun 14 09:34:52 koen: so finally you will be able to boot at91sam? Jun 14 09:35:15 yes Jun 14 09:35:58 I need to reboot cable modem and router - probably be back in few minutes Jun 14 09:37:13 * koen fires up windows Jun 14 09:39:33 -I- Send File linux-2.6.19-at91.img at address 0x00000000 Jun 14 09:48:00 aha, uboot is installed on the dataflash card Jun 14 10:03:17 jo laibsch Jun 14 10:03:32 hi woglinde Jun 14 10:03:37 good morning! Jun 14 10:50:00 * * OE Bug 2485 has been created by slapinid@gmail.com Jun 14 10:50:02 * * angstrom-x11-image doesn't build due to libgpevtype package Jun 14 10:50:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2485 Jun 14 10:50:23 ~seen psokolovsky* Jun 14 10:50:24 slapin: i haven't seen 'psokolovsky*' Jun 14 10:50:28 ~seen psokolovsky Jun 14 10:50:29 psokolovsky was last seen on IRC in channel #htc-linux, 4d 10h 53m 15s ago, saying: 'good'. Jun 14 10:50:38 ~seen psokolovsky_ Jun 14 10:50:39 psokolovsky_ was last seen on IRC in channel #htc-linux, 15h 34m 43s ago, saying: 'http://www.flickr.com/photos/atmasphere/375733253/'. Jun 14 10:50:56 ~seen psokolovsky__ Jun 14 10:50:57 psokolovsky__ was last seen on IRC in channel #handhelds, 17d 12h 53m 22s ago, saying: 'dnix: with Linux, anything is possible'. Jun 14 10:52:07 Please somebody look into 2485 Jun 14 11:01:26 Where do you get angstrom-x11-image? Jun 14 11:01:34 Is your OE tree up to date? Jun 14 11:02:42 I don't have that image anymore. What I have is gpe-image. Jun 14 11:03:35 Laibsch: 13:03 hrw@home:~$ ls devel/oe/org.openembedded/packages/angstrom/angstrom-x11-image.bb Jun 14 11:03:48 re Jun 14 11:03:55 laibsch they are the same Jun 14 11:05:04 Oh, now that is not a good decision to move some packages out of images Jun 14 11:05:14 Especially if they are the same. Jun 14 11:05:17 :-/ Jun 14 11:06:27 Laibsch: angstrom-x11-image was in angstrom/ dir always Jun 14 11:07:09 ??? Jun 14 11:07:18 I could have sworn I saw it in images Jun 14 11:07:31 Well, the argument remains the same, basically Jun 14 11:07:44 group alike stuff to reference it easily Jun 14 11:33:09 Laibsch, any comments? Jun 14 11:33:22 what is the correct way to specify the already builded library in staging in a .bb file ? I try to build linux-iscsi as a package but I can't link on -ldb whereas I specify DEPENDS="db". Jun 14 11:40:30 I am talking with a guy who needs a really small rootfs Jun 14 11:42:20 Crofton: define really small Jun 14 11:42:26 4M Jun 14 11:42:30 :) Jun 14 11:42:31 that's big Jun 14 11:42:34 ;) Jun 14 11:42:51 small is <900kb or so ;) Jun 14 11:42:58 I can see figuring out tasks in my future Jun 14 11:43:00 bitbake angstrom-minimal-image and you'll get something that is ~2M-~3M big Jun 14 11:43:12 thanks, I am goign to give that a wack Jun 14 11:43:29 NAiL: My avr32 rootfs is 968kB big :) Jun 14 11:43:31 trying to drum up some OE converts from the davinci crowd Jun 14 11:43:59 he needs thttpd, sshd, dhcpd Jun 14 11:44:57 he found a link to timesys Jun 14 11:45:21 RDEPENDS = "task-boot dropbear thttpd dchpd" should work afaik Jun 14 11:45:35 after you spell dhcpd correctly :) Jun 14 11:45:39 right Jun 14 11:45:42 s/RDEPENDS/PACKAGE_INSTALL Jun 14 11:46:15 hrw: the RDEPENDS step is needed for shlib renaming iirc Jun 14 11:51:52 RDEPENDS = "${PACKAGE_INSTALL}" Jun 14 11:58:18 * koen finds another point to add branding Jun 14 11:58:33 uboot-mkimage -A ${ARCH} -O linux -T kernel -C gzip -a 20008000 -e 20008000 -n "${DISTRO_NAME} Kernel for ${MACHINE}" -d linux.bin.gz ${DEPLOY_DIR_IMAGE}/uImage-${PV}-${PR}-${MACHINE}-${DATETIME}.bin Jun 14 12:00:40 I recently get a lot of non-fatal errors like NOTE: Couldn't find shared library provider for libc.so.6 in my builds. Is this normal and safe to ignore? Jun 14 12:00:52 no, and no Jun 14 12:01:04 it means a image wouldn't have a c library Jun 14 12:01:04 OK Jun 14 12:01:47 "bitbake -c rebuild glibc" ? Jun 14 12:02:22 that should fix it Jun 14 12:02:59 OK, thanks. Jun 14 12:05:47 these guys doing embedded for money do some weird stuff Jun 14 12:06:02 product based on 2.6.10 Jun 14 12:06:08 and enormous patch sets Jun 14 12:06:50 Crofton|home: 2.6.10 + patches is normal thing Jun 14 12:07:53 why? Jun 14 12:08:10 yay! Jun 14 12:08:11 http://rafb.net/p/WtOTLD21.html Jun 14 12:08:42 isn't 2.6.10 MVL 3.1 ? Jun 14 12:09:01 koen: just this, not more output from the kernel? Jun 14 12:09:01 is MV the root of all evil :) Jun 14 12:09:41 schurig: http://rafb.net/p/KppH4O93.html Jun 14 12:09:41 ARMv5TEJ? Does this chip have a java jazelle, or what is the "J" from? Jun 14 12:09:51 j is for jazelle Jun 14 12:10:02 arm926ej-s core Jun 14 12:10:35 koen: oh, you still have floating point emulator compiled in? Jun 14 12:10:57 yeah for people booting hardfpa rootfses Jun 14 12:11:01 koen: another catch: usbmon: debugfs is not available Jun 14 12:12:13 koen: I didn't knew that anybody still uses hard-fp on NEW devices, it's so sloooow Jun 14 12:12:31 schurig: debian/ARM Jun 14 12:12:49 koen: isn't there an debian arm-eabi? Jun 14 12:12:56 not officially Jun 14 12:12:58 I'd just disable nwfpe and oabi-compat Jun 14 12:13:08 also, woudl debian-arm useful on a machine with only 64 MB ram? Jun 14 12:13:41 for people like pavelm that only understand 'apt-get install bash' but not 'ipkg install bash' Jun 14 12:23:00 hrw: Do you still have your directory with sources used for md5sums? Can you check if your version of the source for tomoe gives you e5f6d1d31d03004812d3eebb6a6d2460? Jun 14 12:23:48 hrw: http://sourceforge.jp/projects/tomoe/files/ and my download say 95eca127a1d519ff4e836b3d00d7fa84 Jun 14 12:27:10 hrw: It seems the entries for uim-tomoe-gtk and libtomoe-gtk are also wrong Jun 14 12:27:37 I think I will update checksums.ini with the published md5sums Jun 14 12:38:33 anyone else seeing this Jun 14 12:38:45 WARNING: You should not compile GNU libc without versioning. Not using Jun 14 12:38:51 in glibc-intermediate Jun 14 12:38:54 I just pulled Jun 14 12:40:36 arg than the fatal Jun 14 12:40:44 configure: error: Need linker with .init_array/.fini_array support. Jun 14 12:42:27 I used to see that Jun 14 12:42:36 do you have MACHINE as an env var? Jun 14 12:42:39 until? Jun 14 12:42:41 env | grep MACH Jun 14 12:42:57 MACHINE=omap5912osk Jun 14 12:43:00 :) Jun 14 12:43:18 unset MACHINE ; bitbake binutils-cross -c rebuild Jun 14 12:43:32 I had just nuked tmp Jun 14 12:44:18 * koen discovers a typo Jun 14 12:44:31 * koen needs more coffee Jun 14 12:45:47 does this type have anything to do with me? Jun 14 12:45:59 no Jun 14 12:46:08 It's in code I haven't checked in ye Jun 14 12:46:12 t Jun 14 12:46:22 I'm confused why I have trouble rebuilding after wipin tmp Jun 14 12:49:02 is anyone using TI DaVinci dm6446 with OE? Jun 14 12:49:18 not yet Jun 14 12:49:21 unsetting MACHINE causes failure due to no TARGET_ARCH Jun 14 12:49:24 Jin^eLD, not yet Jun 14 12:49:28 send me a board Jun 14 12:49:45 CPU is basically same as OMAP5912 Jun 14 12:49:46 Crofton: echo MACHINE=omap5912osk > conf/auto.conf Jun 14 12:49:53 Crofton: don't have one myself at this point :P Jun 14 12:50:03 koen, I am getting seriously confused Jun 14 12:50:05 :) Jun 14 12:50:10 neither do I Jun 14 12:50:12 wait, I do Jun 14 12:50:19 heh ;) Jun 14 12:50:23 Lyrtech SFF SDR but it dosn't boot linux yet Jun 14 12:53:10 hi everybody Jun 14 12:53:20 I wonder how difficult it would be to get that board running in OE Jun 14 12:53:35 which board? Jun 14 12:53:36 hi thesing Jun 14 12:53:36 after all it comes with monta vista stuff, and I did not hear one good word about MV... Jun 14 12:54:06 Crofton: the one we were just talking about? :) the da vinci dm6446 Jun 14 12:54:11 I do not thinnk it will be hard to get OE going for DV Jun 14 12:54:19 dm6446 is on more than one board Jun 14 12:54:40 I think DV is more than one board but dm6446 is a specific one? well I may be wrong Jun 14 12:54:43 I have talked with guys that have them and confiirmed CPU is about the smae as omap5912 Jun 14 12:54:51 DV EVM is a board :) Jun 14 12:55:01 aah Jun 14 12:55:02 thesing any advance with usb-gadet for sa1100? Jun 14 12:56:26 woglinde: No. I'm working on sound first. And I even don't have enough time for that :( Jun 14 12:58:12 thesing oh okay Jun 14 13:24:27 hmm Jun 14 13:24:42 I see that OE/bitbake handles dependencies different than 6 months ago Jun 14 13:25:09 it seems that libtododb is not completely built, but needed at some point (during angstrom-X11-image) Jun 14 13:26:16 and thus failed, but with an easy fix (bitbake libtododb, which had only to be installed and packaged) Jun 14 13:31:42 steliosk: the next commit is interesting for you Jun 14 13:31:59 oxo: no, it is just that the bb files for libtodo are broken Jun 14 13:32:05 ah Jun 14 13:32:20 oxo: although that can be easily worked around as you found out Jun 14 13:32:26 yeah Jun 14 13:32:44 oxo: I will see if I can fix it properly Jun 14 13:35:16 zecke: could you add a test that any package with SECTION related to libs defines do_stage? Jun 14 13:36:09 mmcblk0: mmc0:e727 SD256 249088KiB (ro) Jun 14 13:36:09 mmcblk0: p1 p2 Jun 14 13:36:10 VFS: Cannot open root device "mmcblk0p1" or unknown-block(254,1) Jun 14 13:36:19 * koen scratches head Jun 14 13:46:00 koen: Am I wrong to assume that a lib without something defined for do_stage is broken? Jun 14 13:46:17 whatever that something may be Jun 14 13:51:18 generally a library stages stuff Jun 14 13:52:09 It should, no doubt about it. Jun 14 13:52:38 The question is if a bb with SECTION = "libs" but without a do_stage is broken or not. Jun 14 13:53:05 it isn't Jun 14 13:53:27 our sections are bogus for too many recipes Jun 14 14:02:18 koen, setting machine in auto.conf seems to work Jun 14 14:02:30 why is setting machine in env broken? Jun 14 14:02:46 because binutils is broken Jun 14 14:03:01 is this a permanent thing? Jun 14 14:04:09 no Jun 14 14:04:15 :) Jun 14 14:04:15 RP should have a fix for it Jun 14 14:04:28 * koen reads gtk 2.10.13 release notes Jun 14 14:04:30 I was getting to like setting machine in the env Jun 14 14:05:53 it saves reparsing Jun 14 14:06:29 morning Jun 14 14:06:33 !oebug 2485 Jun 14 14:06:35 * * Bug 2485, Status: NEW, Created: 2007-06-14 03:48 Jun 14 14:06:36 * * slapinid(AT)gmail.com: angstrom-x11-image doesn't build due to libgpevtype package Jun 14 14:06:36 * * http://bugs.openembedded.org/show_bug.cgi?id=2485 Jun 14 14:06:40 please Jun 14 14:10:10 hi Jun 14 14:13:33 oxo: Your best bet is to file a bug report. It seems that my initial assessment that libtodo should have a do_stage declaration was not correc.t Jun 14 14:24:21 psokolovsky_, hi Jun 14 14:24:55 psokolovsky_, oebug 2485 please Jun 14 14:25:42 Hi Jun 14 14:25:50 slapin: what you want from it? Jun 14 14:26:13 psokolovsky_, I want to get image finally. Jun 14 14:26:32 slapin: DEPENDS? Jun 14 14:26:44 psokolovsky_, so to be able to go to beach Jun 14 14:26:49 koen, Crofton|home: There is some fix in poky which makes that work, I thought it was merged into OE but it seems its not judging by what you're saying... Jun 14 14:27:04 psokolovsky_, probably Jun 14 14:27:15 I had this problem jsut now Jun 14 14:29:03 psokolovsky_, so everybody have this problem Jun 14 14:34:13 * koen watches angstrom boot on the at91sam9263-ek Jun 14 14:35:35 re Jun 14 14:41:05 slapin: I wonder if you experience a problem similar to oxo. Can you just build libgpevtype separately? Jun 14 14:41:44 Laibsch, I just testing this now, so in a few mins I'll report. Jun 14 14:42:01 * * OE Bug 2485 has been RESOLVED (FIXED) by pmiscml@gmail.com Jun 14 14:42:03 * * angstrom-x11-image doesn't build due to libgpevtype package Jun 14 14:42:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2485 Jun 14 14:50:41 Laibsch, it doesn't build w/o prior build of libtododb, which I build now Jun 14 14:56:26 Laibsch, udev build error :( Jun 14 14:56:53 Laibsch, arm-angstrom-linux-gnueabi-gcc: Internal error: Segmentation fault (program as) Jun 14 14:58:01 oxo built it successfully if built separately Jun 14 14:58:37 I am no doing GPE builds at the moment but it seems some recent changes have rearranged the stacks of "will it build today" ;-) Jun 14 14:59:12 slapin: If you want something on your device now, why not grab one of the official images and update that via ipkg? Jun 14 14:59:41 Laibsch, I encountered udev-0.92 build error Jun 14 14:59:50 you already said that Jun 14 14:59:51 Laibsch, I participate in testing. Jun 14 15:00:01 * * OE Bug 2486 has been created by slapinid@gmail.com Jun 14 15:00:03 * * udev-0.92 build error Jun 14 15:00:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2486 Jun 14 15:00:06 slapin: Well, it seemed like you were frustrated Jun 14 15:00:28 Laibsch, bug 2486 Jun 14 15:01:09 Laibsch, not that much, I'm just want for everything to be ok, it is not personal, you know :) Jun 14 15:02:18 slapin: I think it is better to attach long logs instead of pasting them inline. Jun 14 15:03:08 Laibsch, if you request them - ok but I doubt they are needet for this particular bug... Jun 14 15:04:23 I was talking about the log you already pasted Jun 14 15:04:33 therefore "instead" Jun 14 15:05:34 hrw: http://www.flickr.com/photos/koenkooi/548080684/ Jun 14 15:06:49 Laibsch, attached. Jun 14 15:07:06 Laibsch, compile and configure logs are attached. Jun 14 15:07:45 slapin: instead!! Jun 14 15:07:56 No use attaching them plus pasting them Jun 14 15:09:24 Laibsch, you want new bug? Jun 14 15:11:05 slapin: What? Jun 14 15:11:17 Maybe I should write Jun 14 15:11:42 In the FUTURE, I think it is better to attach long logs INSTEAD of pasting them inline. Jun 14 15:11:49 Better now ;-) Jun 14 15:11:51 ? Jun 14 15:14:13 Laibsch, you don't need to get so overenthisiastic, I got your point for FUTURE from first time, I thought whatever you want NOW. Jun 14 15:17:14 Laibsch, I tried with -j 1 and it seems to work, seems to be parallel issue (udev) Jun 14 15:17:50 Laibsch, I think it should be added to other "unparallelable" packages. Jun 14 15:24:58 koen, how a package which doesn't build with PARALLEL_MAKE="-j3" could be marked so to make unsetting not required? Jun 14 15:25:24 PARALLEL_MAKE=" " Jun 14 15:25:27 in recipe Jun 14 15:29:08 OK, I will make that update Jun 14 15:29:13 slapin, even I know that :B Jun 14 15:30:46 ~lart BGR framebuffers Jun 14 15:30:46 * ibot_ takes out a seltzer bottle and sprays BGR framebuffers in the face. You know, one of those old-school seltzer bottles clowns have? Yeah those. Anyway, consider yourself spritzed Jun 14 15:31:09 ~lart at91-mci Jun 14 15:31:10 * ibot_ follow's at91-mci with a gauntlet and ... scratch ... HUMILIATION Jun 14 15:32:26 Marex I asked how it is encouraged to do, so that's why I asked. Jun 14 15:33:02 Angstrom minimal for OSK is 13M? Jun 14 15:33:37 slapin, its used in other recipes too :b Jun 14 15:33:48 I think I have some figuring to do .... Jun 14 15:34:19 Crofton|home: all modules? Jun 14 15:34:26 well Jun 14 15:34:35 /boot is present with kernel Jun 14 15:34:45 I need to go through it carefully Jun 14 15:34:55 and ask questions Jun 14 15:35:04 and compare with boostrap-image Jun 14 15:35:06