**** BEGIN LOGGING AT Sun Feb 17 02:59:57 2008 Feb 17 03:34:11 03pfalcon 07org.oe.dev * r316494a6... 10/ (1 packages/initrdscripts/devimage_0.1.bb): Feb 17 03:34:11 devimage 0.1: Link init script to /sbin/init, so this can be booted as normal rootfs. Feb 17 03:34:11 * In addition to default initramfs booting. Feb 17 03:34:21 03pfalcon 07org.oe.dev * r766dada8... 10/ (3 files in 3 dirs): initramfs-uniboot: Modprobe mtdblock and factor out fatal(). Feb 17 03:34:29 03pfalcon 07org.oe.dev * r461df56d... 10/ (3 files in 3 dirs): Feb 17 03:34:29 initramfs-module-block: Accept additional FSTYPE param, handle mount failure. Feb 17 03:34:29 * FSTYPE useful for FSes not automatically recognized by mount (e.g. jffs2). Feb 17 03:34:34 03pfalcon 07org.oe.dev * rf6ab5be3... 10/ (5 files in 3 dirs): * initramfs-module-bootmenu: Add support for jffs2 booting and "drop to shell" option. Feb 17 03:34:39 03pfalcon 07org.oe.dev * rbfa62d5b... 10/ (7 files in 3 dirs): initramfs-uniboot: FSTYPE -> ROOT_FSTYPE, set from rootfstype= command line param. Feb 17 04:03:42 is there any reason I wouldn;t be able to create an ext2 opie image? Feb 17 04:22:03 Soopaman: logs have all the reasons Feb 17 04:40:33 psok, hmm.. logs don't say much different than what the build errored out with Feb 17 04:41:43 genext2fs: ignoring entry initctl Feb 17 04:41:44 genext2fs: group descr 8. free blocks count == 0 (corrupted fs?) Feb 17 04:43:59 and so far google doesn't seem to pull up anytihng Feb 17 05:37:33 strange Feb 17 05:37:43 built bootstrap-image first, and it works now Feb 17 06:10:55 spoke to soon... Feb 17 06:15:23 Hi Feb 17 08:14:04 * * OE Bug 3801 has been created by  Feb 17 08:14:06 * * at76c503a-modules-0.17-r0-do_compile Feb 17 08:14:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3801 Feb 17 08:19:04 * * OE Bug 3802 has been created by  Feb 17 08:19:06 * * minimo-0.02+cvs20070626-r0-do_unpack Feb 17 08:19:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3802 Feb 17 08:46:22 03koen 07org.oe.dev * r26862525... 10/ (1 packages/at76c503a/at76c503a-modules_0.17.bb): Feb 17 08:46:22 at76c503a: add 0.17 release of at76_usb modules Feb 17 08:46:22 * this works with 2.6.21-hh for the h5000 machine Feb 17 08:46:22 * this driver also needs a firmware package, that will follow later Feb 17 08:46:28 03koen 07org.oe.dev * r6448d171... 10/ (1 conf/machine/h5000.conf): h5000: ship all kernel-modules and at76_usb Feb 17 09:57:39 03Laibsch 07org.oe.dev * r32634096... 10/ (1 packages/busybox/busybox-1.9.1/adduser-longops.patch): Feb 17 09:57:39 busybox: add docu for "push 'em"-status of packages/busybox/busybox-1.9.1/adduser-longops.patch Feb 17 09:57:39 * add lines upstream, status and comment to the patch Feb 17 10:17:58 hi all, i'm not able to find instructions to make OE use an external toolchain (at least gcc) ... Feb 17 10:18:11 i set : Feb 17 10:18:18 ASSUME_PROVIDED += "virtual/${TARGET_PREFIX}gcc-3.4.6" Feb 17 10:18:24 #PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc:gcc-cross" Feb 17 10:18:34 PREFERRED_VERSION_gcc-cross = "3.4.6" Feb 17 10:18:44 note the second is a comment ... Feb 17 10:19:07 but OE continue to build gcc-cross-3.4.6 and use it !!! :( Feb 17 10:38:09 03Laibsch 07org.oe.dev * r1d82d7da... 10/ (1 packages/anki/anki_0.4.3.bb): anki: export PV that should work with bitbake 1.8.8, too. Feb 17 10:38:13 03pfalcon 07org.oe.angstrom-2007.12-stable * r9142ddab... 10/ (1 conf/distro/angstrom-2007.1.conf): Feb 17 10:38:13 angstrom-2007.1.conf: Bump to 2007.12-r5, marks kexec cmdline support for linux-hh. Feb 17 10:38:13 * Corresponding kernel version is 2.6.21-hh20-r13. Feb 17 10:38:13 * Worth noting as kexec cmdline support is not backwards compatible, so Feb 17 10:38:14 when there will be tools-on-critical-path relying on it (like bootloaders), older Feb 17 10:38:16 versions will need to be unconditionally upgraded. Feb 17 11:11:08 gremlin[it]: have a look at distro/openwrt.conf as an example. Feb 17 11:18:53 03pfalcon 07org.oe.angstrom-2007.12-stable * rc85783d7... 10/ (1 classes/image.bbclass conf/bitbake.conf): Feb 17 11:18:53 applied changes from 465f9c73626153a7e195a5b8c08ae66282c3722f Feb 17 11:18:53 through 242687582dbdcc1fdca2fc9af1f32a2b5c591f3c Feb 17 11:18:53 bitbake.conf, image.bbclass: Add and use USERDISTRO var to preserve original DISTRO value. (PaulS, Koen) Feb 17 11:18:57 03pfalcon 07org.oe.angstrom-2007.12-stable * rb0fa5276... 10/ (1 classes/image.bbclass conf/bitbake.conf): disapproval of revision 'c85783d7a75d18143af2aab8988ca7ab1d7a395a' Feb 17 11:19:02 03pfalcon 07org.oe.angstrom-2007.12-stable * r5e9a8eb3... 10/ (4 files in 4 dirs): Feb 17 11:19:02 applied changes from 465f9c73626153a7e195a5b8c08ae66282c3722f Feb 17 11:19:02 through 242687582dbdcc1fdca2fc9af1f32a2b5c591f3c Feb 17 11:19:04 bitbake.conf, image.bbclass: Add and use USERDISTRO var to preserve original DISTRO value. (PaulS, Koen) Feb 17 11:19:06 03pfalcon 07org.oe.angstrom-2007.12-stable * rb792b763... 10/ (17 files in 3 dirs): Feb 17 11:19:08 applied changes from 2233080e8bdaa3d4a3eb8a016e6bb014d7c38e45 Feb 17 11:19:10 through 8707d2256d87ca3844a2e60a4e054ffec5b09af7 Feb 17 11:19:12 conf/machine/: Add jffs2 as default image for embedded boards and vendor machines. (PaulS, Koen) Feb 17 11:19:16 03pfalcon 07org.oe.angstrom-2007.12-stable * rffee8d54... 10/ (1 conf/bitbake.conf): Feb 17 11:19:18 applied changes from 34e43612b25ab2cef55713b9d003cdb82e7cfa25 Feb 17 11:19:20 through 465f9c73626153a7e195a5b8c08ae66282c3722f Feb 17 11:19:21 bitbake.conf: Set default IMAGE_FSTYPES to "tar.gz". (PaulS, Koen) Feb 17 11:19:24 03pfalcon 07org.oe.angstrom-2007.12-stable * r29c21251... 10/ (1 BACKPORTS.txt): BACKPORTS.txt: Note bitbake.conf-related changes applied. **** ENDING LOGGING AT Sun Feb 17 11:45:51 2008 **** BEGIN LOGGING AT Sun Feb 17 11:46:14 2008 Feb 17 12:25:02 RP: When are we going to see first implementations of bitbake in client/server mode? Feb 17 12:25:17 morning all Feb 17 12:28:51 thesing: hi Feb 17 12:29:41 mrdata: did you already fix the sa11x0-rtc bug? Feb 17 12:32:25 thesing: not realy, i comented out the irq_request line for ...rtc_open and the free_irq line for IRQ_RTC1Hz Feb 17 12:33:23 hi all Feb 17 12:33:30 hi all, i'm not able to find instructions to make OE use an external toolchain (at least gcc) ... Feb 17 12:34:11 mrdata: so the problem is the irq. Do you have any further information about this? Feb 17 12:35:35 the error comes while irq_request for IRQ_RTC1Hz and comes not return and runes in a endless loop Feb 17 12:36:42 but this jumpes trough kernel parts, debug path is not complete yet from me Feb 17 12:37:04 gremlin[it]: gremlin[it]: have a look at distro/openwrt.conf as an example. Feb 17 12:54:11 thanks psokolovsky ! Feb 17 12:54:37 ;-) Feb 17 12:57:07 mhhh what i don't understand is ... is i set ASSUME_PROVIDED += "virtual/${TARGET_PREFIX}gcc" have also to setup PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc:gcc-cross" and/or PREFERRED_VERSION_gcc-cross = "3.4.6" ??? Feb 17 13:08:54 psokolovsky i mean what local.conf is need to use openwrt-sdk to be used as reported in file "DISTRO=openwrt-sdk TARGET_ARCH= bitbake " ? Feb 17 13:09:21 psokolovsky i mean what local.conf is need to use openwrt-sdk as reported in file "DISTRO=openwrt-sdk TARGET_ARCH= bitbake " ? Feb 17 13:09:31 gremlin[it]: you have to not have local.conf at all. Feb 17 13:09:49 local.conf considered harmful ;-F Feb 17 13:10:48 heheh ... mhh i'll try ... the bad things is that or bitbake use the most recent gcc it have (4.1.2) or build 3.4.6 (is a my test) but it fail ... Feb 17 13:29:50 Hello Feb 17 13:48:36 Laibsch: Have you looked at bitbake trunk? Feb 17 13:49:06 Laibsch: Its already started to an extent although it needs a lot more work. Its being worked on as and when I have the time/energy... Feb 17 14:05:14 03rpurdie 07org.oe.dev * rd2dfc84f... 10/ (1 classes/kernel.bbclass): classes/kernel.bbclass: Remove usage of non-standard variables (DEPLOY_TO) and random renaming of output. Feb 17 14:05:18 03rpurdie 07org.oe.dev * r56704f75... 10/ (1 packages/initrdscripts/kernel-initramfs_1.0.bb): kernel-initramfs: Remove recursive usage of bitbake as per discussion on the mailing list and agreement by the OE core team. Feb 17 14:27:03 RP: I have not looked at the code. I thought just asking you would be quicker ;-) Feb 17 14:27:14 Is there anything that works yet, that can be tested? Feb 17 14:48:46 03pfalcon 07org.oe.dev * r6bbae187... 10/ (3 files in 2 dirs): Feb 17 14:48:46 kernel-initramfs, initramfs-image-ipk: Remove recipes with recursive bitbake invocations. Feb 17 14:48:46 * As not maintainable on global OE level. Feb 17 14:51:11 hi, i have patch adding new machine, how to get it to mainstream OE? Feb 17 14:51:56 Magon: Submit it to the OE mailing list as a patch Feb 17 14:52:20 ok..to ml Feb 17 14:52:39 i probably need to subscribe for that ml Feb 17 14:53:48 Magon: a better option is to attach it to the bug tracker, anyway Feb 17 14:54:02 !oebug 1 Feb 17 14:54:03 Laibsch: ok..i am just registering at bugzilla Feb 17 14:54:03 * * Bug 1, Status: RESOLVED (FIXED), Created: 2005-05-11 18:09 Feb 17 14:54:05 * * keturn(AT)keturn.net: wpa_supplicant 0.3.8-r0 fails on driver_hermes Feb 17 14:54:05 * * http://bugs.openembedded.org/show_bug.cgi?id=1 Feb 17 14:54:17 Magon: Great Feb 17 14:54:59 Laibsch: how long it can take to make it to OE? Feb 17 14:55:14 anything between 5 minutes to 5 years ;-) Feb 17 14:55:24 ok...that is not too promissing Feb 17 14:55:31 You can get closer to 5 minutes by nagging people from time to time Feb 17 14:55:38 bugzilla is broken! Feb 17 14:55:49 Magon: try .net instead of .org Feb 17 14:55:53 it still wants to connect to bugs.openembeded.net instead of .org Feb 17 14:55:59 * Laibsch thought the DNS issues were fixed Feb 17 14:56:09 ok..which one is new? Feb 17 14:56:34 Magon: try the IP, then 72.233.35.82 Feb 17 14:56:48 i can see main page..but clicking on log in is not working Feb 17 14:56:56 Is it? Feb 17 14:57:02 What problem do you get? Feb 17 14:57:06 cookies enabled? Feb 17 14:57:23 trying to connect to .net and for me just .org is working Feb 17 14:57:34 * Laibsch takes a look Feb 17 14:57:53 Magon: .net seems to point to the correct IP Feb 17 14:58:10 What do you get for "host bugs.openembedded.net"? Feb 17 14:58:12 What do you get for "host bugs.openembedded.org"? Feb 17 14:58:36 bugs.openembedded.net has address 76.74.156.109 Feb 17 14:58:41 bugs.openembedded.org has address 72.233.35.82 Feb 17 14:59:02 what should i get? Feb 17 14:59:05 Magon: subscribing to the ML is still mandatory ;-) Feb 17 14:59:20 psokolovsky: why? Feb 17 14:59:35 "just do it!" ;-) Feb 17 14:59:43 but ok..i will subscribe anyway Feb 17 15:00:00 psokolovsky: is there a way you will review my patch and apply it in short time? Feb 17 15:00:19 or let's try that: subscribing to list may make adding new machine faster ;-) Feb 17 15:00:23 i solved my DNS problem by /etc/hosts entry :-) Feb 17 15:00:35 Magon: which is what I did, too Feb 17 15:00:38 But this is annoying Feb 17 15:00:46 We have had this problem for a long time, now Feb 17 15:00:47 yes..someone should fix it :-) Feb 17 15:00:49 too long Feb 17 15:00:52 Magon: "you" as existing developers - yes, create bug, post to ML Feb 17 15:01:50 psokolovsky: i have no access to do anything other than creating but..i need to convince someone to actually push it :-) Feb 17 15:02:27 Magon: to convince someone to push it, you need to convince that you're going to maintain it Feb 17 15:02:42 assign to ? :-) Feb 17 15:02:55 mickey|flu: please update bugs.openembedded.net to point to 72.233.35.82, you are lagging behind .org which is already correct ;-) Feb 17 15:02:57 to default Feb 17 15:03:03 psokolovsky: i have no problem maintraining it..i am developing for that machine now Feb 17 15:03:54 psokolovsky: Good to see you more active on IRC Feb 17 15:04:30 dunno if I'll be able to keep up with that ;-) Feb 17 15:08:06 http://bugs.openembedded.net/show_bug.cgi?id=3803 Feb 17 15:08:11 morning all Feb 17 15:08:35 now the ml Feb 17 15:09:43 psokolovsky: which ml you suggest? Feb 17 15:10:41 Magon: and where's? linux-zipit2 Feb 17 15:11:04 Magon: oe-devel Feb 17 15:11:07 Magon: if you're interested to have your machine supported in angstrom, then angstrom-distro-devel Feb 17 15:11:41 well, posting to oe-devel about a patch to add a machine makes little sense ;-) Feb 17 15:14:04 * * OE Bug 3803 has been created by magon(AT)dobaledova.net Feb 17 15:14:06 * * New machine Zipit2 Feb 17 15:14:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3803 Feb 17 15:22:43 psokolovsky: sorry..network problems and i forgot about kernel completly Feb 17 15:25:13 Magon_ : Talk to Crofton Feb 17 15:25:47 Magon_ : He had a discussion (as far i can remember) with the guy that builds the zipiit devices Feb 17 15:26:08 hi all ! Feb 17 15:26:32 * steliosk is looking at 5 cm of snow outside his window Feb 17 15:26:53 steliosk: i have been in touch with him Feb 17 15:27:14 magon : Crofton ? Feb 17 15:28:04 steliosk: yes Feb 17 15:28:26 but he hasnt been around lately..or i missed him Feb 17 15:28:30 03Laibsch 07org.oe.dev * rd0092933... 10/ (1 packages/zten/files/zten.patch packages/zten/zten_1.6.2.bb): zten: push patches upstream. Add AUTHOR to bb file. Feb 17 15:30:23 magon : probably. Most OE devs seem to be busy Feb 17 15:30:34 steliosk: as most of the people i know Feb 17 15:30:41 yeah Feb 17 15:31:22 magon : How far did you get with OE/zipit2 ? Feb 17 15:32:03 steliosk: booting for now..someone already tested SDL on it Feb 17 15:32:40 steliosk: keyboard driver is written, wifi is working Feb 17 15:32:50 but no luck with X so far Feb 17 15:37:04 is possible force a package to be build with a specifica compiles (maybe external to OE) ... something like is already possible with kernels ? Feb 17 15:37:21 compiles -> compiler Feb 17 15:39:34 are there any tutorials/help script for installing oe images to compact flash cards? Feb 17 15:40:22 magon : You have both machine and angstrom stuff in one patch. Don't want to make you go through hoops, but it would be better if you could separate them. It would make easier for any dev to push the machine specific part while some angstrom-dev takes a look at the angstrom stuff Feb 17 15:40:35 Soopaman: bitbake a tar.gz rootfs and extract it onto CF, SD, MD, whatever. Feb 17 15:41:05 Soopaman : There was a link at the wiki Feb 17 15:41:31 Soopaman : i think it was for x86/grub Feb 17 15:41:44 steliosk: the angstrom stuff is trivial..just added my machine to few lists of arm-oabi machines Feb 17 15:42:18 maybe I don't get it but why would one need a manual for a simple "tar -xzf" ? Feb 17 15:42:21 steliosk: bigger problem is the kernel patch.... Feb 17 15:42:33 * gremlin[it] think too much machine in oabi :S Feb 17 15:43:15 magon : That was my five Euro cents :) Feb 17 15:43:22 steliosk: yes ok Feb 17 15:43:58 steliosk , there also 1 and 2 euro cents :P Feb 17 15:44:17 heh yeah...haven't seen them for a while now Feb 17 15:45:25 * steliosk is going to clear the snow from his car Feb 17 15:45:31 brb Feb 17 15:46:02 psokolovsky: are you around? Feb 17 15:46:09 yep Feb 17 15:46:43 psokolovsky: do you think it would be ok if i just removed the line about prefered kernel and submited it without kernel? Feb 17 15:47:12 i have problem that kernel is not flashable and there are patches for kernel but they are not nice Feb 17 15:47:27 magon: heh. dunno ;-) Feb 17 15:47:32 coredump, what's the best way to get it to boot hten? Feb 17 15:47:56 Soopaman: depends on the device of course. You only asked about installation ;) Feb 17 15:48:13 lol Feb 17 15:48:25 Some bootloaders can do it OOTB, some don't. Feb 17 15:49:13 k Feb 17 15:55:17 psokolovsky: so now it should be ok http://bugs.openembedded.net/show_bug.cgi?id=3803 Feb 17 15:55:46 * magon finishing subscriptions to lists Feb 17 15:57:00 Well, and how does it build - without a kernel? have you tested that? Feb 17 15:57:06 this is wrong: Feb 17 15:57:07 +DISTRO_SSH_DAEMON = "openssh" Feb 17 15:57:08 psokolovsky: Error: "DISTRO_SSH_DAEMON" is not a valid command. Feb 17 15:57:08 +DISTRO_FEED_CONFIGS += "wireless-tools" Feb 17 15:57:08 +DISTRO_FEED_CONFIGS += "wpa-supplicant" Feb 17 15:57:10 psokolovsky: Error: "DISTRO_FEED_CONFIGS" is not a valid command. Feb 17 15:57:11 psokolovsky: Error: "DISTRO_FEED_CONFIGS" is not a valid command. Feb 17 15:57:38 MACHINE_ESSENTIAL_EXTRA_RDEPENDS += "kernel-modules" - what is this if you don't even build a kernel?? Feb 17 15:58:11 psokolovsky: is should build some generic kernel..or not? Feb 17 15:58:30 magon: likely! and that's wrong! Feb 17 15:58:45 psokolovsky: ok so how should it look like? Feb 17 15:59:03 DISTRO* cannot be in machine Feb 17 15:59:20 with kernel, you need to do something for it to not build at all ;-) Feb 17 15:59:58 good point..DISTRO in machine is insane Feb 17 16:06:06 psokolovsky: do you know how to disable kernel at all ? Feb 17 16:06:50 not sure Feb 17 16:07:34 why is this in angstrom MACHINE_KERNEL_VERSION = "2.6" Feb 17 16:08:38 because angstrom is for 2.6 Feb 17 16:09:10 so it should be DISTRO_KERNEL_VERSION Feb 17 16:09:51 no Feb 17 16:18:44 psokolovsky: without prefered kernel it warns me that i dont have one but seems like it will build Feb 17 16:19:04 ... random one? right. Feb 17 16:19:12 and that's bad ;-) Feb 17 16:19:33 magon: as you see, adding a new machine well is not always easy Feb 17 16:20:22 psokolovsky: psokolovsky seems like it Feb 17 16:20:26 magon: I suggest sending message to ML - now even I suggest oe-devel - and describe why you need to add machine w/o kernel recipe and some detail about machine Feb 17 16:20:32 psokolovsky: i added simply simple "linux" Feb 17 16:21:28 that's hack. for something to go into OE and enjoy all benifits from that (some distros start to support your device without too much effort from your side), everything should be done properly Feb 17 16:21:30 it is probably impossible to have machine without kernel...it needs to build glibc againt something..at least i need kernel headers Feb 17 16:22:23 i can properly add my kernel and add right kernel provider..problem is my kernel is total mess i got from zipit wireless..and it is quite big Feb 17 16:22:31 but it build Feb 17 16:24:57 magon: if it builds, and patches are available by URL, that's good start Feb 17 16:25:35 magon: otherwise, send mail to ML. I'd say, you open a precedent, and it should be discussed wide how to handle it. Feb 17 16:34:51 psokolovsky: problem is it might be available by URL..but i dont know exactly how to write proper file Feb 17 16:35:35 magon: OE tree has hundreds of samples ;-) Feb 17 16:35:49 psokolovsky: ok Feb 17 16:36:17 most of my today work was; look how does it other, try to do the same Feb 17 16:37:44 magon: and that's very right approach! if you just start with that, in a week you'll be pretty good at it. otherwise, it's usually takes couple of hours to figure out what you didn't know before ;-) Feb 17 16:39:06 defconfig probably can stay in tree right? i will just change the linux-zipit2.bb to use patches from svn Feb 17 16:40:15 yes, defconfig preferrably should be in OE tree Feb 17 16:42:42 psokolovsky: and thx for help Feb 17 16:43:59 I'd be glad if Angstrom served as many machs as possible - it was conceived with that aim in mind ;-) Feb 17 16:45:10 psokolovsky: ok..you are angstrom developer? Feb 17 16:45:25 yep Feb 17 16:45:43 so if i do it correctly you may push it in? Feb 17 16:46:56 magon: I will +1 it ;-). another +1 required for stable branch. as for .dev, it's easier, but still should be clean to go. Feb 17 16:48:28 i probably use dev branch..or i completly dont know Feb 17 16:49:23 devel branch is org.openembedded.org Feb 17 16:49:33 err org.openembedded.dev Feb 17 16:49:34 you likely do (use .dev) ;-) Feb 17 16:49:42 yes..i have devel Feb 17 16:50:35 now i have patches in SVN..need to figure out how to download them Feb 17 16:55:20 ok..giving it a shot Feb 17 16:55:34 i hope it will download and use the patches correctly Feb 17 17:09:55 how to set which revision oe will fetch? Feb 17 17:14:36 ok i have it Feb 17 17:15:35 I'm trying to setup local overlay, using bb colections for testing before I submit patches, but I've problem building kernel http://pastebin.com/m1b8bf935 Feb 17 17:16:10 it seems, that on the line number 53, there's additional ", but I'dont know from where it comes Feb 17 17:19:02 if I try to build this recipe under OE it builds fine Feb 17 17:38:18 psokolovsky: i have solved my problem..it downloads patches from svn and applies correctly Feb 17 17:52:26 psokolovsky: http://bugs.openembedded.net/show_bug.cgi?id=3803 Feb 17 17:52:35 psokolovsky: i hope this is correct Feb 17 17:55:08 Laibsch: You can try bitbake trunk and it will mostly work Feb 17 18:10:07 hi Feb 17 18:11:57 hi Feb 17 18:27:50 psokolovsky: are you around? Feb 17 18:41:10 RP: I am using bitbake trunk. What does this split into client/server entail? Feb 17 18:41:39 I can start multiple bitbake processes on one machine and they won't step on one another's toes like they sometimes did in the past? Feb 17 18:41:52 (when two subprocesses tried to do the same thing) Feb 17 18:42:22 Can I "submit" work from one machine on a LAN to another machine? Feb 17 18:42:52 I guess I am not yet clear about what it means (as well as what it does *not* mean)for bitbake to go client/server Feb 17 18:44:00 Laibsch: The trunk code has a UI thread which communicates with a server thread over xmlrpc Feb 17 18:44:12 Oh, GUI? Feb 17 18:44:19 That sounds also nice Feb 17 18:44:32 Any GUI software yet, to watch all those processes? Feb 17 18:44:35 Laibsch: Currently it only supports doing it over localhost but obviously once that works, it can be made inter-machine easily Feb 17 18:44:42 bitbake with GUI ???? Feb 17 18:44:47 Laibsch: No GUI has been written yet Feb 17 18:45:04 Laibsch: The infrastructure to do it is being put into place though Feb 17 18:45:05 magon: yes, it was something that RP wanted to do (eventually), if I am not mistaken Feb 17 18:45:13 RP: Cool Feb 17 18:45:17 Thanks Feb 17 18:45:19 Laibsch: that feels strange Feb 17 18:45:37 magon: I am not sure you understand what will be in the GUI Feb 17 18:45:42 It can make a lot of sense Feb 17 18:45:59 Laibsch: probably just selecting packages and so on? Feb 17 18:46:11 magon: Just take looking for progress of a single task, for example Feb 17 18:46:21 Laibsch: i dont say it is nonsence...just for me OE and bitbake is untamed beast Feb 17 18:46:28 magon: Now those messages are all over the place if you have BBTHREADS Feb 17 18:46:37 One aim I have is a red/green light in my systray on my desktop... Feb 17 18:46:40 Laibsch: i know..that is a bit mess Feb 17 18:47:54 Laibsch: As things stand there are some bugs/problems with the client/server split and I need to spend some time polishing that up. Once that is done I'll turn attention to remote connections and GUIs Feb 17 18:48:22 * RP -> food Feb 17 18:49:06 Laibsch: can you review my patch if it seems good or i still have errors in it? Feb 17 18:49:45 magon: Sorry, I don't think I am the right person for that. Feb 17 18:49:49 ok Feb 17 18:49:52 np Feb 17 18:49:52 RP: Sounds awesome Feb 17 18:49:59 i will wait for psokolovsky Feb 17 18:50:06 RP: I am looking very much forward to it Feb 17 18:53:21 RP: smakelijk eeten, btw Feb 17 18:53:35 Am I correct to assume that I can now have two bitbakes running at the same time and it should not bomb like it sometimes happened in the past? Feb 17 19:02:04 * * OE Bug 3804 has been created by  Feb 17 19:02:06 * * gcc-cross-sdk-4.1.1-r17-do_install Feb 17 19:02:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3804 Feb 17 19:05:39 03mickeyl 07org.oe.dev * rd3bfb633... 10/ (3 files in 3 dirs): openmoko-sound-system2: remove hard-coded fragment size for alsa sink Feb 17 19:05:44 03mickeyl 07org.oe.dev * r049e91a5... 10/ (3 files in 2 dirs): Feb 17 19:05:44 pulseaudio: 0.9.9 works good enough to be the default version now Feb 17 19:05:44 * add pulse to audio user group, so it can run as system daemon eventually Feb 17 19:05:50 03mickeyl 07org.oe.dev * ree7bb5fb... 10/ (1 packages/gstreamer/gst-plugin-pulse_0.9.7.bb): gst-plug-pulse 0.9.7 new default version Feb 17 19:05:56 03mickeyl 07org.oe.dev * rcb01f274... 10/ (1 conf/distro/include/sane-srcrevs.inc): sane-srvrevs.inc: bump gsm0710muxd to r77 Feb 17 19:34:30 hi again, i was reafing FAQ about adding a package to a image, the esaple refer to a file isn't present in stable branch angstrom-console-image.bb ... what's the correct filename ?! Feb 17 19:34:36 concole-image ? Feb 17 19:34:46 console-image ? Feb 17 19:35:42 second question, if i want to 'remove' a package from a image what can i do? Feb 17 19:36:47 there is a smaller image than 'console-image' with just tinylogin, busybox, dropbear ... ? Feb 17 20:13:25 Laibsch: You mean in the same build directory? Feb 17 20:13:39 RP: yes Feb 17 20:14:04 Laibsch: That isn't safe to do at present Feb 17 20:14:09 Otherwise, what real changes can I expect to see? This is the only thing I could think of Feb 17 20:14:50 Laibsch: That isn't the point of the client/server. The point of client/server is to let multiple clients control one server and abstract the control interface enough to let a GUI be written Feb 17 20:17:13 well, since clients does not yet mean clients from different machines on the LAN, I figured this meant clients from the same machine. Which is how I came to assume multiple bitbakes. I guess eventually, this would have to become a safe thing to do before allowing jobs from the LAN Feb 17 20:17:52 I guess we are still at preparatory stage before seeing any differences in functionality or output Feb 17 20:18:28 Laibsch: Not yet. It wouldn't be hard from the current code base though Feb 17 20:18:56 Laibsch: At any one time only one "job" will be active although in theory others will be queued Feb 17 20:19:34 The exact mechanism is to be determined of course ;-) Feb 17 20:21:28 yes, the queueing what I think would be nice Feb 17 20:21:38 Just fire off what you want to do Feb 17 20:21:43 and forget about it ;-) Feb 17 20:21:57 Laibsch: Yes, its still preparatory and the lack of visible difference is actually and aim at present :) Feb 17 20:22:06 a/and/an/ Feb 17 20:22:31 Is SlugOS still alive? Who works on it? Feb 17 20:24:31 RP: Do our glibc patches usually apply to upstream, too? Or are they mostly local to OE, not applicable to glibc at large? Feb 17 20:25:12 I think I'd like to send most of them upstream today if most of them are applicable, but I'd have a hard time to tell Feb 17 20:26:04 Laibsch: To be honest I don't know Feb 17 20:26:11 OK Feb 17 20:26:23 ah yes patches :} Feb 17 20:26:30 I'll try to find another high-prio target Feb 17 20:26:36 Laibsch: I do help a lot of patches upstream, I'm just perhaps quiet about it Feb 17 20:26:37 zecke: Would you know? Feb 17 20:26:56 RP: I know, no offense meant Feb 17 20:27:36 Laibsch: I'm not taking any, I just want you to know that whilst I'm not doing that this weekend, I do believe in what you're doing :) Feb 17 20:28:00 Laibsch: and that I do contrinute in my own way (OH in general does really) Feb 17 20:29:00 RP: I also think I want to continue after tomorrow Feb 17 20:29:20 It is something that people like me can do easily and have a good impact Feb 17 20:29:33 Laibsch: I mussed the question :) Feb 17 20:29:38 the benefit of unificying recipes for example has worn off a bit Feb 17 20:29:57 zecke: WTH is "mussing a question"? Feb 17 20:30:19 Laibsch: s/mussed/missed/ ;-) Feb 17 20:30:31 oh Feb 17 20:30:55 zecke: Do our glibc patches usually apply to upstream, too? Or are they mostly local to OE, not applicable to glibc at large? Feb 17 20:32:02 Laibsch: hehe, I'm tired... Feb 17 20:32:11 Laibsch: oh, glibc Feb 17 20:32:31 Laibsch: have you heard about Ulrich Drapper/Drepper (can't remember)? Feb 17 20:33:43 no Feb 17 20:33:46 who is that? Feb 17 20:34:35 * RP notices palm emails to the bitbake list. I wonder what they're cooking... Feb 17 20:35:19 Laibsch: find out, send a patch upstream :) Feb 17 20:36:12 zecke: glibc not a good target to pick for sending (our) patches? Feb 17 20:36:43 RP, zecke: Which are IYO the most valuable project picks for sending patches upstream from OE? Feb 17 20:36:53 low-hanging, high-impact fruit? Feb 17 20:36:59 Laibsch: hehe, pick some easy ones first Feb 17 20:37:18 Laibsch: Ones where there is a good chance the upstream will apply the patch Feb 17 20:37:20 IOW, where should we have sent patches, but did not? Feb 17 20:37:36 RP: Help me pick one project Feb 17 20:37:40 Laibsch: You can poke dodji on a libcroco patch :) Feb 17 20:37:51 OK Feb 17 20:38:06 Dodji hangs out here Feb 17 20:38:20 I'd like something more core, too Feb 17 20:38:31 Maybe, I'll just pick what I like Feb 17 20:39:26 Laibsch: good as well Feb 17 20:39:29 Laibsch: I know we have a lot of what upstreams consider "crack" for things like gtk :( Feb 17 20:39:57 RP: Do they have a better solution Feb 17 20:40:13 They should also be marked as rejected or unapplicable Feb 17 20:40:26 so we know that those patches don't need to be pushed upstream Feb 17 20:40:57 Laibsch: Some are slowly being beaten into shape, rewritten or dropped by OH... Feb 17 20:41:16 Laibsch: I agree we need to document the patches more Feb 17 20:42:54 Laibsch: The good thing is, we don't change much of the upstream stuff. Mostly for building and packaging purposes Feb 17 20:43:14 Laibsch: sometimes theses are clean and right but upstream will discard them anyway (glibc) Feb 17 20:43:38 Laibsch: or probably sqlite because we use autotools and not the other buildsystem Feb 17 20:43:40 zecke: Well, that's life Feb 17 20:43:44 We cannot force them Feb 17 20:43:56 But we can (and should) try and let them know Feb 17 20:44:01 Laibsch: and on the other spectrum we have build hacks Feb 17 20:44:09 Laibsch: right, informing is good Feb 17 20:44:21 you mean, sqlite is a good candidate? Feb 17 20:44:28 Laibsch: e.g. with libmrss and libnxml I mailed the upstream author Feb 17 20:45:03 Laibsch: sqlite3, bad patch :) Feb 17 20:45:16 Laibsch: we patch an autogenerated file, which is ugly Feb 17 20:45:48 Laibsch: I got most of the xcalibrate changes I made pushed into the respective upstream repos and poky has updates to make it use the latest git versions which don't need patches Feb 17 20:47:58 Laibsch: the mono patches by thesing and henryk look good Feb 17 20:48:21 OK, Feb 17 20:52:22 hi all Feb 17 20:53:05 Laibsch: did you start creating a patch header yet? is there at least one example? Feb 17 20:53:10 how i can create a new image smaller than console-image ... mainly i know some packages i don't need and fail to build :( Feb 17 20:53:31 gremlin[it]: minimal-image? Feb 17 20:53:41 zecke: I considered my initial example good Feb 17 20:53:49 I was unsure about the mail you sent Feb 17 20:54:06 Are you sure we both understand "status:" to be the same? Feb 17 20:54:14 I meant upstream status Feb 17 20:54:40 zecke: Have you seen http://oz.leggewie.org/wip/patches-upstream.txt ? Feb 17 20:55:32 ahhaha it work ... ok i'll try Feb 17 20:55:59 mickey|flu: Nice work on the handout, it reads really well :) Feb 17 20:56:10 mickey|flu: Hope the flu clears up soon :/ Feb 17 20:57:38 Laibsch: yeah, I would extend it though Feb 17 20:57:45 Laibsch: hack and hack-deserves-cleaning Feb 17 20:57:55 Laibsch: actually they are the same ;) Feb 17 20:59:50 Laibsch: I will push something to libmrss, let's see if we agree Feb 17 21:00:57 zecke: the policy proposal is open for input Feb 17 21:01:02 I think that should be clear Feb 17 21:01:24 too bad, the wiki does not work as it should Feb 17 21:04:42 03Laibsch 07org.oe.dev * rd6af6dba... 10/ (6 files in 4 dirs): wxbase: consolidate into wxwidgets project directory. Feb 17 21:20:04 * * OE Bug 3805 has been created by  Feb 17 21:20:06 * * most heavily patched projects in OE Feb 17 21:20:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3805 Feb 17 21:21:42 Laibsch: you have linux in the list 2 times Feb 17 21:22:16 magon: already fixed Feb 17 21:22:20 see the update Feb 17 21:24:26 Laibsch: http://sources.redhat.com/bugzilla/show_bug.cgi?id=3004 for drepper in action Feb 17 21:29:24 zecke: painful... Feb 17 21:29:41 lol Feb 17 21:31:19 we should each reopen that suggesting its a good idea... Feb 17 21:32:25 Getting every OE dev barred from the redhat bugzilla would be different... Feb 17 21:32:55 RP: I have seen/met him at FOSS.IN Feb 17 21:33:25 RP: I just recognized that even on normal systems you might have an issue, if your binutils are not multiarch Feb 17 21:33:51 zecke: He's no better in person I presume? Feb 17 21:35:37 RP: hehe, he seems to be a nice guy Feb 17 21:36:40 RP: I think most people working for Red Hat have some issues (j5 is the exception) Feb 17 21:37:59 zecke: hehe :) Feb 17 21:48:14 eh, that Lua stuff is pain to package Feb 17 21:58:31 hi ant Feb 17 21:58:48 woglinde: hey Feb 17 21:59:45 * ant_ cannot login in x11-gpe images lately...thought was resolved by busybox upgrades but still... Feb 17 22:00:36 * ant_ is talking about angstrom-2008.1 Feb 17 22:04:46 nobody else? worth opening a bug? Feb 17 22:05:18 03freyther 07org.oe.dev * rf84e4db3... 10/ (1 packages/libmrss/files/better-parse-url-r0.patch): packages/libmrss/files/better-parse-url-r0.patch: Document what's up with this patch, hopefully some one is sending it upstream Feb 17 22:05:20 03freyther 07org.oe.dev * r9193ea51... 10/ (3 files in 3 dirs): Feb 17 22:05:20 packages/glibc/zecke-sane-readelf.patch: This was needed because OSX has no readelf command by default Feb 17 22:05:20 and thread support of the compiler was misdetected to the missing readelf. Feb 17 22:05:26 03freyther 07org.oe.dev * r473156ab... 10/ (1 packages/gcc/gcc-4.2.1/zecke-no-host-includes.patch): packages/gcc/gcc-4.2.1/zecke-no-host-includes.patch: Document this patch and explain why it is a hack Feb 17 22:05:31 03freyther 07org.oe.dev * r4b9bf269... 10/ (1 packages/gcc/gcc-4.2.1/zecke-xgcc-cpp.patch): packages/gcc/gcc-4.2.1/zecke-xgcc-cpp.patch: When cross compiling we want to use gcc's cpp for the cross compiled headers Feb 17 22:05:36 03Laibsch 07org.oe.dev * rfce78846... 10/ (6 files in 4 dirs): acpid: push gcc patch upstream and consolidate identical patches. Feb 17 22:32:55 when i have mtn diff output..how to apply it? Feb 17 22:39:51 ok figured out..it is simple diff Feb 17 23:00:08 hmm... what is the best way to cleaning a build? Feb 17 23:00:18 like a "make clean" Feb 17 23:00:32 Soopaman: bitbake -c clean Feb 17 23:04:09 hmm.. i tried that, but now it's just erroring on me Feb 17 23:04:35 Soopaman: If you want to nuke everything, rm -rf /base/of/oe/tmp Feb 17 23:04:57 but wouldn't i have to redo the entire build process again? Feb 17 23:05:39 Soopaman: What happens when you -c clean? Feb 17 23:06:42 hmm... sorry should have evaluated Feb 17 23:06:54 i cleaned, but when i try to bitbake bootstrap-image, it fails Feb 17 23:06:59 03deller 07org.oe.dev * r6cf1636a... 10/ (1 conf/machine/include/zaurus-2.6.inc): zaurus-2.6: add pcmcia modules for Poodle Feb 17 23:07:04 03deller 07org.oe.dev * rbbb903b6... 10/ (1 packages/tasks/task-base.bb): task-base: add crypto_algapi/cryptomgr for task-base-wifi Feb 17 23:07:09 03deller 07org.oe.dev * r816caaf2... 10/ (1 packages/linux/linux-rp-2.6.24/pxafb.patch): linux-rp 2.6.24: remove repeated content from pxafb.patch Feb 17 23:26:40 hrw|gone: are you sure about the change in the pxafb patch? the two lines look similar but in the second line the get_pcd function get two arguments and in the first line only one. Feb 17 23:49:05 * * OE Bug 3311 has been RESOLVED (FIXED) by Feb 17 23:49:07 * * Fix for task-opie-all Feb 17 23:49:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3311 Feb 17 23:55:45 bye Feb 18 00:05:14 03Laibsch 07org.oe.dev * r6b62c710... 10/ (1 packages/tasks/task-opie-all.bb): task-opie-all: drop opie-usermanager. Fixes bug 3311. Feb 18 00:05:18 03Laibsch 07org.oe.dev * r2802c077... 10/ (1 packages/libmrss/files/better-parse-url-r0.patch): libmrss: document better-parse-url-r0.patch being sent upstream. Feb 18 00:42:04 * * OE Bug 2283 has been RESOLVED (FIXED) by Feb 18 00:42:06 * * tshark recipe Feb 18 00:42:08 * * http://bugs.openembedded.org/show_bug.cgi?id=2283 Feb 18 01:06:21 03Laibsch 07org.oe.dev * r285f8edf... 10/ (4 files in 4 dirs): Feb 18 01:06:21 gcc: add upstream comments to three more, otherwise identical patches. Feb 18 01:06:21 * complimentary to 4b9bf269fccb2d8469b9d1ccfe6d73ac060a6b29 Feb 18 01:06:21 * "upstream: n/a" should really have a more verbose explanation Feb 18 01:06:21 is it a hack, is upstream being a d0rk, did you try but failed? Feb 18 01:06:28 03Laibsch 07org.oe.dev * rf37bc671... 10/ (1 packages/wireshark/tshark_0.99.4.bb): Feb 18 01:06:28 tshark: initial commit of tshark, the text-based version of wireshark. closes 2283. Feb 18 01:06:28 * the latest version 0.99.7 fails do_patch, so I left everything at 0.99.4 Feb 18 01:24:04 * * OE Bug 3044 has been RESOLVED (FIXED) by Feb 18 01:24:06 * * djvulibre 3.5.19 not compiling Feb 18 01:24:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3044 Feb 18 01:43:14 03Laibsch 07org.oe.dev * rd042c316... 10/ (1 packages/djvulibre/djvulibre_3.5.19.bb): Feb 18 01:43:14 djvulibre: replace QTDIR with PALMTOPDIR to enable successful compilation. closes 3044. Feb 18 01:43:14 * use sourceforge mirror Feb 18 01:43:14 * reorder fields according to style guide Feb 18 01:43:20 03Laibsch 07org.oe.dev * r747b468a... 10/ (3 files in 2 dirs): djvulibre: update to version 3.5.20. Feb 18 01:57:05 * * OE Bug 2580 has been RESOLVED (FIXED) by Feb 18 01:57:07 * * bison added to sanity.bbclass Feb 18 01:57:09 * * http://bugs.openembedded.org/show_bug.cgi?id=2580 Feb 18 02:20:04 * * OE Bug 2724 has been RESOLVED (FIXED) by Feb 18 02:20:06 * * makeinfo-native depends on util-macros-native Feb 18 02:20:08 * * http://bugs.openembedded.org/show_bug.cgi?id=2724 Feb 18 02:22:04 * * OE Bug 3602 has been RESOLVED (FIXED) by Feb 18 02:22:06 * * tune-strongarm.inc uses mtune=xscale instead of mtune= strongarm1100 Feb 18 02:22:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3602 Feb 18 02:24:04 * * OE Bug 3806 has been created by smallpotato000(AT)gmail.com Feb 18 02:24:06 * * Release 2007.12 c7x0 kernel is buggy Feb 18 02:24:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3806 Feb 18 02:54:02 03Laibsch 07org.oe.dev * r0ec6a2fa... 10/ (8 files in 3 dirs): qclockchange: update to version 0.1b. Closes 2653. Feb 18 02:54:06 03Laibsch 07org.oe.dev * r6accbea5... 10/ (3 files in 2 dirs): Feb 18 02:54:06 icewm: RRECOMENDS ttf-dejavu-sans to make sure some font is installed. partly closes 1506. Feb 18 02:54:06 * if there is a more general solution, let your voice be heard. **** ENDING LOGGING AT Mon Feb 18 02:59:57 2008