**** BEGIN LOGGING AT Thu Nov 01 02:59:57 2007 Nov 01 03:34:20 build update-alternatives Nov 01 03:34:25 ERROR: No providers of build target update-alternatives (for []) Nov 01 03:34:27 hmm.... Nov 01 07:27:28 hi Nov 01 07:28:05 I need to execute some shell code before and after unpacking (basically because a source package is really messed up) Nov 01 07:28:39 I added a do_unpack_prepend() and do_unpack_append() for that but it looks like bitbake does not like my construction: Nov 01 07:29:16 2:SyntaxError:' mkdir ${WORKDIR}/swt-unpack' Nov 01 07:29:16 0001:def do_unpack(): Nov 01 07:29:16 0002: mkdir ${WORKDIR}/swt-unpack Nov 01 07:29:16 0003: bb.build.exec_func('base_do_unpack', d) Nov 01 07:29:34 any idea what went wrong? Nov 01 07:31:50 hi all Nov 01 09:02:48 03koen 07org.oe.dev * rd6c0da94... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom 2008: prefer 2.6.23 for libc headers Nov 01 09:19:30 Anyone that is building oe on cygwin or knows anyone that is doing this? Nov 01 09:20:04 esben_: it was done once upon a time many years ago, but no-one is that crazy :-) Nov 01 09:42:22 XorA: I might have to be that crazy soon... Nov 01 09:42:42 03koen 07org.oe.dev * re969264b... 10/ (1 packages/linux/linux-bfin_svn.bb): linux-bfin: add compatible host Nov 01 09:43:10 I need to build a gcc4 cross-compiler toolchain for cygwin host and ppc603e target with NPTL support. Nov 01 09:43:59 and it should be easily reproducable for a less skilled software developer, which preempts the option for building it manually from scratch. Nov 01 09:44:46 and crosstool does not seem to be going anywhere with NPTL support for gcc4. Nov 01 09:46:34 morning Nov 01 10:38:18 * steliosk wonders why people like to use a Windows machine, to build a linux toolchain for cross-compiling Nov 01 10:38:44 steliosk: working gameloader Nov 01 10:39:34 steliosk: but mainly because in a lot of companies the IT staff are morons and it would take 2-3 years to get a linux machine comissioned Nov 01 10:42:16 XorA : :) Nov 01 10:45:40 * koen writes a script to remove duplicate mails for his Maildir after thunderbird messed up again Nov 01 10:48:21 heh heh Nov 01 10:49:24 * koen discovers http://debian-multimedia.org/ Nov 01 10:49:35 time to update mythtv Nov 01 11:05:38 esben_, kergoth used to use colinux Nov 01 11:07:09 Crofton|home: that won't really help me a lot. Nov 01 11:07:28 have you tried vmware also? Nov 01 11:08:10 I need a good tool for building a toolchain bundle for host=cygwin Nov 01 11:08:45 and I need to build a similar toolchain bundle for host=linux-x86 Nov 01 11:09:36 esben_: talk to mickeyl, he's own newly appointed SDK and toolchain expert :) Nov 01 11:09:46 * mickeyl coughs Nov 01 11:09:56 you have no idea... Nov 01 11:10:00 *sigh* Nov 01 11:10:02 It doesn't matter if I build the toolchain on linux or cygwin, but as far as I can see, OE is not really prepared for building a true cross-compiler with build!=host!=target Nov 01 11:10:03 I figure mickeyl may actually be interested in that solution Nov 01 11:10:08 * koen hands mickeyl a glass of water Nov 01 11:10:20 * mickeyl rather grabs coffee instead Nov 01 11:10:37 and he is here, great :-) Nov 01 11:10:52 it's like 6 years since I did a build!=host!=target toolchain Nov 01 11:11:20 mickeyl: do you have an idea for how much effort it would require to abuse the current OE toolchain recipes for building a build!=host!=target toolchain ? Nov 01 11:11:48 that would be REALLY cool for other purposes also :-) Nov 01 11:12:02 you mean like building a powerpc->arm cross compiler on x86 ? Nov 01 11:12:47 like setting up an automated buildmachine for maintaining a zillion different pre-compiled cross-compiler tar.gz files available for people that would then not have to figure out how to use OE. Nov 01 11:13:14 mickeyl: exactly (or a cygwin->powerpc on linux-86) Nov 01 11:13:32 aha Nov 01 11:13:52 well, at one point in OE's life this worked Nov 01 11:13:58 i have no idea whether it still does Nov 01 11:14:10 unfortunately our _real_ toolchain experts are kind of missing in action Nov 01 11:15:13 * mickeyl thinks Phil Blundell Nov 01 11:16:02 my gut feeling is that it might be a lot of work to get that feature back Nov 01 11:16:11 and it will probably break for lots of combinations Nov 01 11:18:08 mickeyl: my gut is telling me the same :-( Nov 01 11:18:26 which is why I am considering making OE work under cygwin. Nov 01 11:18:59 esben_: that sounds pretty good to me. i'd like to see that working Nov 01 11:19:11 its a pity LINE never took off Nov 01 11:19:11 is it fair to believe that it might be a faster road to getting a meta-toolchain building for cygwin->powerpc by making OE build on cygwin? Nov 01 11:19:45 esben_: i think so, yes Nov 01 11:19:55 esben_: might also consult steliosk, who i think did some work on cygwin Nov 01 11:20:29 cygwin building is just soooooooo slooooooooowwwwwwwww Nov 01 11:22:34 building gcc_4.1.2.bb for powerpc (efika/angstroem) fails Nov 01 11:24:47 esben_ : If this is for customers tell them to install a linux pc or you will face so many support requests your head will spin Nov 01 11:25:14 s/linux pc/pc with Ubuntu feisty/ Nov 01 11:27:05 * steliosk is happy as the internet speed was increased and prise went down in his connection Nov 01 11:28:08 my modem locked at 11.9Mbit/1.1Mbit Nov 01 11:28:27 * koen is only a week away from going from 1Mbit/s to 20MB/s Nov 01 11:28:33 for 18 Euros+vat a month Nov 01 11:28:46 still adsl so that will get like 4Mb/s Nov 01 11:29:02 (and 20Mb/s, stupid capitals) Nov 01 11:29:52 how much you would pay for it ? Nov 01 11:30:04 * hrw still has 3072/512kbps cable Nov 01 11:30:15 €20/month Nov 01 11:30:44 koen: what about that with your university account -- do we need to adjust all blogs on planets? Nov 01 11:30:48 * hrw wants .nl net offers here Nov 01 11:31:34 mickeyl: yes, but I'd wait with that till the adsl modem goes online (20071106) and I can properly use dyndns Nov 01 11:31:47 ah, right Nov 01 11:31:56 then everyone hammer koens ADSL link :-) Nov 01 11:32:03 heh Nov 01 11:33:10 * XorA lives too far from exchange for super fast ADSL Nov 01 11:33:30 hmm Nov 01 11:33:44 ok, i think it's time to say good bye to mandriva and give ubuntu a chance on my laptop Nov 01 11:33:56 let's see how this works out Nov 01 11:34:04 but dinner beforehand Nov 01 11:34:05 mickeyl: heh heh, didnt you see ubuntu gutsy is destroying laptop harddrives Nov 01 11:34:07 bbiab Nov 01 11:34:11 what??? Nov 01 11:34:18 they fucked up powersaving Nov 01 11:34:24 argh Nov 01 11:34:34 they park/unpark the drivers once every 30 secs or more often Nov 01 11:36:07 re Nov 01 11:36:43 Hi Nov 01 11:37:01 XorA : This is not a bug its a feature.... :) Nov 01 11:37:30 steliosk: save power by killing the HD so it doesnt run anymore :-) Nov 01 11:38:08 no it's the BIOS' fault! Nov 01 11:38:16 Nov 01 11:38:30 * koen RP'ed his ubuntu install Nov 01 11:38:55 build own kernel, don't use ramdisk Nov 01 11:39:35 on desktop I have own kernel no ramdisk but on laptop I use ramdisk (but own kernels too) Nov 01 11:39:54 we should make a desktop version of Angstrom :) Nov 01 11:40:21 Crofton|home: I will not run it on my laptop ;) Nov 01 11:40:39 Angsrtom should be self hosting ... Nov 01 11:40:58 it is Nov 01 11:41:09 or 'it was nearly year ago' Nov 01 11:41:21 steliosk: I'm afraid it is not an option (making the customer switch to Linux) Nov 01 11:42:17 steliosk: the customer is not meant to be using OE, much less OE under cygwin. they just need to use the OE built stand-alone toolchain to build their own application. Nov 01 11:43:03 one developer will have to be able to build new images, but he is equiped with a linux pc (fedora though...) Nov 01 11:45:32 I have a slight problem with busybox. I was using version 1.2.1, but due the missing killall5 function, i had to switch to another version. Version 1.7.2 aswell 1.6.0 builds, but i get a fucked up prompt: "\u@\h:\w$ ". The option "CONFIG_ASH_EXPAND_PRMT" is set.. What am i doing wrong? Nov 01 11:52:10 you build for !angstrom distro? Nov 01 11:55:45 * hrw just pushed busybox 1.7.2 change Nov 01 11:55:55 or rather I am pushing it Nov 01 11:56:03 We're using our own distro... Nov 01 11:57:31 any news on the packaged-staging development, what is the ETA for merging it in to .dev? Nov 01 11:59:11 Reneger: pull/update Nov 01 12:02:03 Crofton|home: Hey there! How is your u-boot work goig? Nov 01 12:02:04 03hrw 07org.oe.dev * r25169276... 10/ (5 files in 4 dirs): busybox 1.7.2: use Angstrom defconfig as default - there are usable fixes in it Nov 01 12:02:05 n Nov 01 12:02:29 russf, read finshing MS, that is going well Nov 01 12:02:40 off to a conference next week Nov 01 12:02:45 hrw: Check. Nov 01 12:03:09 have you resolved your uboot versus kernel decompressor issue? Nov 01 12:03:27 Crofton|home: happy that your MS is geting done.... Nov 01 12:03:39 very happy Nov 01 12:04:05 Crofton|home: I've been working on the decomp issue... Bit of a dead end, but I did find more patches needed for u-boot. Nov 01 12:04:23 there where a some patches on the uboot list recently Nov 01 12:04:40 the 1.2.0 I builg had no NAND mtd support. Nov 01 12:04:47 jsut testing that. Nov 01 12:05:14 I looked for very recent patches on the 2 lists I know of and only found stuff from June. Nov 01 12:05:23 Perhaps you can send a pointer. Nov 01 12:05:57 http://www.nabble.com/-patch-0-3--KB9202-CodeSourcery-ARM-EABI-toolchain-build-fixes-tf4692013.html#a13410606 Nov 01 12:06:27 Crofton|home: looks like kwokbyte submitted changes that were not that tidy, and Wolfgang bounced them. So I'll tidy and resubmit. Nov 01 12:06:51 wolfgang takes things very seriously Nov 01 12:07:26 He reminds me of people that obsess over correct punctuation :) Nov 01 12:07:40 He carries a big cross. Nov 01 12:07:54 the ARM maintainer tends to be asleep at the wheel I think Nov 01 12:08:12 A bit of discipline can cut his load, but in this case it caused a pretty clean patch from getting in. Nov 01 12:08:18 check the arm maintainer git tree to see if he has pushed those patches Nov 01 12:08:27 Has anyone read shuttleworth's recent comments about working together? Nov 01 12:08:45 OK. I don't know where to look for that, tho. Nov 01 12:08:58 russf, where? Nov 01 12:10:32 http://www.denx.de/wiki/UBoot/Custodians Nov 01 12:11:03 I don't see them Nov 01 12:11:39 Crofton|home: http://www.linux-watch.com/news/NS8470376604.html Nov 01 12:11:57 Crofton|home: thx for tips. Nov 01 12:57:37 morning Nov 01 13:35:32 any volunteers for adding alsa-lib .15? Nov 01 13:35:56 I have it in local poky tree iirc Nov 01 13:36:40 hrw: Thanks! That worked! Nov 01 13:37:09 koen: I have them. booting my c7x0 now to test Nov 01 13:38:33 hrw: thanks Nov 01 13:42:25 1.0.14 on device Nov 01 13:43:34 hi Nov 01 13:44:27 I am trying to integrate the openembedded toolchain with scratchbox. but when i execute ~/sb-toolchain-extras/confhelper/create_toolchain_conf.py Nov 01 13:44:51 i am getting the following error "./arm-angstrom-linux-gnueabi/bin/ld: /scratchbox/host_shared/lib/libc.so.6: version `GLIBC_2.4' not found (required by ./arm-angstrom-linux-gnueabi/bin/ld)" Nov 01 13:45:21 can somebody help me.. Nov 01 13:45:36 03henryk 07org.oe.dev * r67a297cf... 10/ (1 packages/vte/vte_0.16.8.bb): vte: Add missing DEPENDS Nov 01 13:48:33 In some scratchbox mailing list, its specifed that the library path will be overriden with the alien binary wrapper.. Nov 01 13:49:54 techindian: it seems our local scratchbox export florian stepped out for lunch Nov 01 13:49:57 ~seen florian Nov 01 13:49:59 florian was last seen on IRC in channel #oe, 22h 21m 33s ago, saying: '?'. Nov 01 13:50:23 :-( Nov 01 13:52:35 koen: alsa 1.0.15 pushed to poky Nov 01 13:53:51 any clues please.. Nov 01 13:54:19 koen: thanks. Nov 01 13:54:27 I am avoiding crapbox^Wsuxxbox^Wscratchbox as much as I can Nov 01 13:54:33 so never used it Nov 01 14:04:11 Crofton: Sigh! Looks like CFG_CMD_NAND is a legacy thing... Wondering why I don't have mdtparts as a u-boot command.... Nov 01 14:25:45 * cbrake tries to convince bitbake to fetch u-boot-mkimage-openmoko-native ... Nov 01 14:30:09 moin Nov 01 14:54:47 Crofton: I feel like Hansel, picking up your breadcrumbs through the u-boot forest... Nov 01 14:55:44 Crofton: Just found the CFG_NAND_LEGACY advice to prevent multiple defs of low level stuff in 32&64 bit Nov 01 14:56:15 russf, are you working with git? Nov 01 14:56:22 not at all. Nov 01 14:57:02 I've been called a git. Never used it. Nov 01 14:57:26 heh Nov 01 14:57:27 Seems like 'm missing part of the picture. Nov 01 14:57:38 you may want to use the latest u-boot from git Nov 01 14:58:49 If things get slow at the conference, I'll try building it from git Nov 01 14:59:06 the conference organizers are good about keeping us covered with wireless Nov 01 14:59:15 which conf? Nov 01 14:59:18 hi crofton Nov 01 15:00:34 kb9202 Nov 01 15:00:38 I can see if it builds Nov 01 15:01:06 I mean which conference... Nov 01 15:01:22 www.sdrforum.org Nov 01 15:01:58 Ah. cool. Your first love! Nov 01 15:02:20 yeah, I have to work :) Nov 01 15:02:38 bbiab Nov 01 15:03:58 ;) Nov 01 15:05:51 Crofton: woohoo. I have my mtdparts command. I may have a 1.2.0 build that will work.... 8-) Nov 01 15:08:13 Crofton: nah! :( "Support for Nand devices not present" ... ho hum... Nov 01 15:08:54 Crofton: and now ls is broken too. Nov 01 15:09:06 * cbrake notes cleaning all traces of git stuff from downloads does wonders for the git fetcher ... Nov 01 15:09:25 how do you checkout a branch at a given date with mtn ? Nov 01 15:09:46 cbrake: any idea ? Nov 01 15:10:01 nud: not sure offhand Nov 01 15:10:16 mtn --db=/stuff/OE.mtn checkout --branch=org.openembedded.dev --date=2007-10-09 <-- this doesn't work Nov 01 15:10:48 it says it doesn't understand the date format, despite it looks as rfc8601 as I could do Nov 01 15:11:56 nud, tried using a --revision ?? Nov 01 15:12:09 or use a magic selector Nov 01 15:12:19 russf: and how do I get the revision that corresponds to a given date ? :-) Nov 01 15:12:55 mtn co ;mtn log ;mtn update -rrevision Nov 01 15:13:07 nud: http://monotone.ca/monotone.html#Selectors Nov 01 15:14:56 koen: thanks! Nov 01 15:15:08 actually -r was to be used instead of --date Nov 01 15:15:54 mtn checkout -r2007-10-09 worked Nov 01 15:45:03 * * OE Bug 3251 has been created by autobuild(AT)openembedded.org Nov 01 15:45:05 * * popt-1.7-r4-do_compile Nov 01 15:45:07 * * http://bugs.openembedded.org/show_bug.cgi?id=3251 Nov 01 15:47:07 mickeyl: new pulseaudio release, doesn't compile for me Nov 01 15:49:21 ooh 0.9.7 Nov 01 15:49:25 coolies Nov 01 15:49:32 will look into that over the weekend Nov 01 15:49:36 perhaps even earlier Nov 01 15:49:43 got a recipe? Nov 01 15:51:18 copy the 0.9.6 one Nov 01 15:51:29 it fails with tls stuff over here Nov 01 15:51:36 ~lart openmoko-devel-image Nov 01 15:51:36 * ibot puts on a hockey mask and jumps out at openmoko-devel-image Nov 01 15:51:55 1.5M free on / Nov 01 15:52:12 ya, getting a bit fat around the hips Nov 01 15:52:26 ~lart 64MB flash Nov 01 15:52:26 * ibot eats 64MB flash and falls over dead Nov 01 15:52:30 mickeyl: gta02 will use jffs2 or yaffs? Nov 01 15:52:40 hrw: unclear at this point of time. Nov 01 15:52:54 we didn't have time to thorougly evaluate both yet Nov 01 15:53:07 gta02 will have 128MB, so both will be an option Nov 01 15:53:18 "yaffs2 write path has more bugs than it has lines of code" according to a kernel hacker Nov 01 15:53:26 128? had to be 256... Nov 01 15:53:33 anyway small Nov 01 15:53:45 could be true, however hxd8 is using 3G of flash w/ yaffs2 and they have no problems Nov 01 16:02:04 03koen 07org.oe.dev * r7fe4c06c... 10/ (4 files in 3 dirs): pty: add Andrews tty forwarder Nov 01 16:04:41 * hrw wants gconf-editor Nov 01 16:06:09 there are 2 or three around iirc Nov 01 16:07:05 I really need to move neo to sd Nov 01 16:07:31 root@a780:/tmp# df -h Nov 01 16:07:32 Filesystem Size Used Available Use% Mounted on Nov 01 16:07:34 :D Nov 01 16:08:04 dev/mmcblk0p2 424.7M 109.7M 293.0M 27% / Nov 01 16:09:15 aaha Nov 01 16:09:24 handy Nov 01 16:09:42 and the only solution on a780 iirc Nov 01 16:10:01 yeah Nov 01 16:10:05 anyway for neo it looks like rootfs on sd will be the only reliable solution too Nov 01 16:10:08 I don't count nfsroot as an option Nov 01 16:10:20 neither do I Nov 01 16:11:12 nfsroot by GPRS :-) Nov 01 16:11:22 my neo just freezed - 1.5M free killed jffs2 Nov 01 16:13:05 * * OE Bug 3252 has been created by autobuild(AT)openembedded.org Nov 01 16:13:07 * * uclibc-0.9.29-r8-do_compile Nov 01 16:13:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3252 Nov 01 16:35:14 mickeyl: heh, the toolchain experts always seem to go MIA. i think originally it was gb2, then pb_.. Nov 01 16:37:59 03hrw 07org.oe.dev * rc7ee1600... 10/ (4 files in 2 dirs): alsa: added 1.0.15 from Poky Nov 01 16:38:05 03hrw 07org.oe.dev * r647167f3... 10/ (1 packages/qte/qte-mt_2.3.10.bb): qte: mark qte-mt 2.3.10 as preferred version to make bitbake a bit more quiet on parse Nov 01 16:40:26 kergoth: heh. right, must be really painful knowledge ;) Nov 01 16:41:29 indeed Nov 01 16:49:45 so they are all called ?b? and dissapear, thats suspicious :-) Nov 01 16:50:32 hehe Nov 01 16:55:24 hmm Nov 01 16:55:38 is openembedded supporting bsd ? Nov 01 16:59:42 no Nov 01 17:01:49 Does anyone know if it's possible to make ipk put installed file size and md5sum on the target db? Nov 01 17:47:01 hi Nov 01 17:47:29 can some please test my patches attached at bug 2348 on glibc und uclibceabi? Nov 01 17:47:58 03koen 07org.oe.dev * r85c8d7e0... 10/ (1 packages/ezx/ezxd_svn.bb): ezxd: compile with -DDEBUG Nov 01 17:48:56 args Nov 01 17:49:01 new alsa-utils Nov 01 18:31:28 Anyone know if there is a current shorewall package? Nov 01 19:13:15 bye Nov 01 19:25:03 If I have two .bb files, 0.8.0.bb and 0.8.1.bb, under what circumstances would it choose to build 0.8.0 ? Nov 01 19:25:36 prefered_version set Nov 01 19:25:51 or DEFAULT_PREFERENCE="-1" in 0.8.1 Nov 01 19:26:54 the latter isn't in 08.1; where do I look for prefered_version to be set? Nov 01 19:27:26 what distro? Nov 01 19:27:33 OpenMoko Nov 01 19:27:37 hmmm Nov 01 19:27:51 what bb file specifically? Nov 01 19:28:07 I'm creating gnash_0.8.1.bb Nov 01 19:28:11 look in the distro Nov 01 19:28:26 and it's *been* building 0.8.1 just fine. Nov 01 19:28:35 "look in the distro"? Nov 01 19:28:46 read the distro file and see if there are any soruce for version lock downs Nov 01 19:29:15 you have have better luck asking in #openmoko also Nov 01 19:29:29 I'm sorry, I don't know what you mean by "read the distro file". Nov 01 19:29:44 conf/distro/openmoko.conf Nov 01 19:33:37 Nope, don't see any mention of gnash. Nov 01 19:34:21 Is there a switch I can give to bitbake to get it to tell me why it's doing what it's doing? Nov 01 19:34:32 I'm not sure Nov 01 19:34:42 it seems like that would be useful :) Nov 01 19:34:51 Indeed. I'll look around. Nov 01 19:35:06 it can make a dependency map it dot Nov 01 19:35:17 reading that may help Nov 01 19:35:37 compiling into a figure gives a silly picture you can not read Nov 01 19:35:57 yeah, and which threatens to crash my machine because it takes so much memory. Nov 01 19:36:14 but try reading the source, it may give insight ... Nov 01 19:36:15 annoyingly, "bitbake -c build gnash" doesn't say which version it's building. Nov 01 19:36:45 try gnash_0.8.1 Nov 01 19:36:57 we met the gnash dude at FOSDEM last year Nov 01 19:37:07 Rob Savoye? Nov 01 19:37:09 apparently autotools is also his fault Nov 01 19:37:11 Yeah Nov 01 19:37:15 heh, yep. Nov 01 19:37:25 interesting fellow Nov 01 19:37:39 yep, that too. Nov 01 19:37:58 good to see even dead heads can lead productive lives :) Nov 01 19:38:57 * Crofton has been a a few dead shows, back in the day Nov 01 19:39:29 crfoton can ypu do me a favor and test my patches for alsa-utils from bug http://bugs.openembedded.org/show_bug.cgi?id=2348 Nov 01 19:39:30 It just says: NOTE: Selecting already-staged gnash (0:0.8.0-r0) to satisfy gnash Nov 01 19:39:51 try rebuild Nov 01 19:39:55 -c rebuild Nov 01 19:41:04 I've done that. It keeps trying to rebuild 0.8.0 Nov 01 19:41:21 woglinde, what are you trying to fix? Nov 01 19:41:33 crofton build under uclibc Nov 01 19:41:35 I do not have any uclibc builds handy Nov 01 19:41:42 with nls Nov 01 19:41:54 under uclibc it compiles Nov 01 19:41:57 the annoying thing is that it *used* to build 0.8.1 Nov 01 19:42:02 but please test it under glibc Nov 01 19:42:05 ok Nov 01 19:42:23 and tell me if it compiles and run Nov 01 19:42:43 I check compile Nov 01 19:42:50 okay Nov 01 19:42:51 I do not have anything I can test sound on Nov 01 19:43:01 hm Nov 01 19:43:32 I will check build though Nov 01 19:44:36 but be aware koen or some one else checked in alsa-utils 1.0.15 Nov 01 19:44:46 patches are against 1.0.14 Nov 01 19:45:08 ok Nov 01 19:45:21 and the automake one did not apply Nov 01 19:45:49 hm Nov 01 19:45:53 uh? Nov 01 19:45:59 let me update Nov 01 19:46:28 are sorry Nov 01 19:46:46 you have to put the automake under packages/alsa-utils/files Nov 01 19:47:07 one is for the .bb the other for the alsa-utils source Nov 01 19:47:39 oops Nov 01 19:48:12 packages/alsa/files Nov 01 19:48:25 I can apply by hand .... Nov 01 19:49:14 ???? Nov 01 19:49:27 something is preferring gnash 0.8.0. How do I find out what it is? Nov 01 19:49:31 the automake patch is one line Nov 01 19:49:40 crofton *g* yes Nov 01 19:49:47 nelson bitbake bla -g Nov 01 19:50:06 creates you two .dot files Nov 01 19:50:20 still waiting on pull/update Nov 01 19:50:23 and bitbake -e shows which version is prefferred Nov 01 19:50:24 then grovel through the .dot files? Nov 01 19:50:46 -s shows which version is preferred. Nov 01 19:50:47 Does anyone know if it's possible to make ipk put installed file size and md5sum on the target db? Nov 01 19:52:03 nelson or try -D Nov 01 19:53:38 -g doesn't tell me why it's choosing 0.8.0 ...it merely tells me that it has done so. Nov 01 19:54:37 nelson you did your own gnash.bb? Nov 01 19:54:53 woglinde: yes, for 0.8.1 Nov 01 19:55:02 hm okay Nov 01 19:55:11 try to grep the conf-dir Nov 01 19:55:33 maybe its pinned in sane revision Nov 01 19:56:07 No mentions on gnash in openembedded/conf/distro/* Nov 01 19:56:23 But somebody *is* preferring 0.8.0 over 0.8.1 Nov 01 19:56:38 -D doesn't tell me why it chose 0.8.0 over 0.8.1 Nov 01 19:56:40 *g* Nov 01 19:56:50 try to move the old .bb aside Nov 01 19:56:53 and try again Nov 01 19:57:02 well, sure that will work, as a work-around. Nov 01 19:57:02 maybee something wrong with your new bb Nov 01 19:57:19 oh, okay, as a diagnostic, sure. Nov 01 19:58:29 It's building 0.8.1 but whinging that it prefers 0.8.0 (for item gnash) Nov 01 19:58:51 lol Nov 01 19:59:04 try to grep the whole oe dir Nov 01 19:59:19 I dont have another idea right now Nov 01 20:02:16 bugger all, pull/update is rebuilding cross Nov 01 20:02:26 woglinde, it will be a lame build test :) Nov 01 20:05:27 croftom okay than I will play now a game Nov 01 20:05:31 ask maybee later Nov 01 20:05:36 bye for now Nov 01 20:14:04 armv5te/alsa-lib_1.0.15-r0_armv5te.ipk Nov 01 20:14:16 that built without problems after hrw's commit Nov 01 20:19:59 can we change CVS_TARBALL_STASH to SRC_TARBALL_STASH in local.conf.sample? Nov 01 20:21:46 alsa-utils_1.0.14 builds with the change for glibc ... Nov 01 21:08:42 nelson, look in conf/distro/include/angstrom-2007-preferred-versions.inc Nov 01 21:09:05 does openmoko.conf use the angstrom source rev files? Nov 01 21:09:17 l8r Nov 01 21:14:02 yes. Nov 01 21:15:36 Crofton: thanks, yes, for some reason, rather than relying on the presence of a .bb (or newer) file, it forces certain versions. That's not a choice I would have made. Nov 01 22:41:51 Crofton: for your GIT builds of u-boot, what tool chain do you use? Nov 01 22:54:37 the one that comes from OE Nov 01 22:54:43 I set the path to use it Nov 01 22:54:56 you may have to set the prefix for u-boot also Nov 01 23:04:36 Crofton|home: ok. I'll do the same. Nov 01 23:06:14 I don't think I should be using the LEGACY_NAND flag. It allows a build, but without some of the funcs I need. I need to get past those udiv errors, possibly by mickeyl's patch or maybe GIT has updated that problem. Nov 01 23:06:38 g'night! Nov 01 23:07:04 look at teh davinci "schmoogie" machine Nov 01 23:07:14 I think it should be close (NAND wise) Nov 01 23:07:22 ok. thanks for that. Nov 01 23:07:30 I think git solved that problem Nov 01 23:08:51 I don't really want the GIT learning curve. I'm getting a bit frustrated with missing patches, and walled gardens of code. I need to get to my app.... Nov 01 23:09:10 Nov 01 23:11:18 Does anyone know if it's possible to make ipk put installed file size and md5sum on the target db? Nov 01 23:12:03 russf, we all whine about git :) Nov 01 23:12:06 just do it Nov 01 23:12:31 If things get slow next week, I'll try doing some builds during talks Nov 01 23:12:41 erk.. ok. But tomorrow... Nov 01 23:12:47 tomorrow I drive Nov 01 23:12:58 I am going to Denver via Charleston SC Nov 01 23:13:06 visiting my father on the way Nov 01 23:14:58 That's a long drive! Nov 01 23:16:49 1,705 mi – about 1 day 2 hours Nov 01 23:17:11 And through some of the most brain-dead states in the known universe..... Nov 01 23:17:45 heh Nov 01 23:17:49 Don't mention Darwin, even if you are talking about a BSD variant. Nov 01 23:17:59 no, Blacksburg to Charleston is only 6 hours Nov 01 23:18:05 I catch a plane in Charleston Nov 01 23:18:10 my wife drives back :) Nov 01 23:18:38 My firend has driven it several times Nov 01 23:18:56 Prolly a nice drive, joking apart. Nov 01 23:19:11 Been to Knoxville, the GSMs. Nice. Nov 01 23:19:14 there is a set of signs in Missouri, one promoting religon as the answer Nov 01 23:19:22 the next adevertises the p0rn shop Nov 01 23:19:39 I am "near" there Nov 01 23:19:53 as i Knoxville is about 3-4 hours from heere Nov 01 23:20:21 driving through Kansas/eastern Colorado is something that should be done once or twice Nov 01 23:21:31 blacksburg, near lynchburg. Nov 01 23:21:52 there are many terrible puns I could make, but I'll save everyone the embarrassment. ;) Nov 01 23:26:29 I once drove 2500 miles on a thxgiving weekend. Spent a very cold and sleepless night freezing in a rental car on Dante's view, overlooking Death Valley. Stupid trick. By morning I was wearing a suitcasefull of clothes. Nov 01 23:26:38 bey. Nov 01 23:26:42 bye Nov 01 23:27:16 gn Nov 01 23:45:38 crofton okay than I will check it in Nov 01 23:46:04 hm oh seems the have fixed upstream Nov 01 23:46:26 args Nov 01 23:48:16 hm Nov 01 23:48:18 okay **** ENDING LOGGING AT Fri Nov 02 02:59:57 2007