**** BEGIN LOGGING AT Sat Jul 07 02:59:56 2007 Jul 07 03:51:26 03mickeyl 07org.oe.dev * rd5b50f42... 10/ (3 files in 2 dirs): etk upgrade to 0.1.0.004 Jul 07 03:51:31 03woglinde2 07org.oe.dev * rf60008d7... 10/ (1 packages/linux/linux/simpad/defconfig): Jul 07 03:51:31 putting ide-support, so we can use cf-pcmcia adapters or Jul 07 03:51:31 pcmic cdroms Jul 07 08:07:34 good morning all Jul 07 08:21:03 ~botmail for woglinde: *NEVER* DEPEND on libiconv directly, but on virtual/libiconv, otherwise you break all glibc based builds out there Jul 07 08:29:26 03koen 07org.oe.dev * r9958e850... 10/ (3 files in 3 dirs): disapproval of revision '66573a06c1ee62e38d70a6c82f8d99a41243cd5c' Jul 07 08:36:05 * koen|inverness heads to the station Jul 07 08:36:07 later all Jul 07 09:16:46 hi Jul 07 09:18:10 does anyone know a decent, up to date userland for the ipaq h3800 Jul 07 09:20:07 morning ! Jul 07 09:23:25 morning Jul 07 09:56:37 morning wvdeschel Jul 07 10:32:54 anyone know what flow control params to use to login to getty on Zaurus? Jul 07 10:33:01 yo koen|train in the wet lands Jul 07 10:41:21 hi all Jul 07 10:55:56 * XorA|gone gets as far as Got Here 4 :-D Jul 07 10:56:23 hi mrdata Jul 07 11:10:39 mickey|zzZZzz, you awake? Jul 07 11:21:42 woowoo, my COnnect+ card is working Jul 07 11:25:02 RP: are you on the hostap mailing lists or anything? Jul 07 12:00:56 psokolovsky, ping Jul 07 12:02:33 pong Jul 07 12:02:35 Hi! Jul 07 12:02:39 hey Jul 07 12:03:03 I just added update-alternatives to taskbar-images (yes, cia kept silence), but now I wonder how do we go about actually installing the proper res? Jul 07 12:03:44 I do have some ideas, but I'm not yet well-acquainted with all the tasks/images/confs stuff, so I thought I'd rather ask you Jul 07 12:04:03 polyonymous: I need to write RFC on that Jul 07 12:04:19 psokolovsky, that makes sense. Jul 07 12:04:22 each machine should specify its standard resolution in its config Jul 07 12:04:27 yes Jul 07 12:04:45 there's runtime suport for that in poky, needs to be merged Jul 07 12:04:51 I'd think we need to specify resolution per machine and we want to have the list of resolution as well. Jul 07 12:04:59 no build-time though, but needs to be modelled on that Jul 07 12:05:30 as funny as it gets, but runtime support seems much easier task for me :) Jul 07 12:06:11 polyonymous: are you sure? ;-) Jul 07 12:06:24 I'm sure it SEEMS much easier :) Jul 07 12:06:37 and still, we can't install resources for each possible resolution, so we must have default Jul 07 12:06:49 But yes, user should be free to install any per his likes Jul 07 12:06:55 psokolovsky, but I can put it the other way - bitbake stuff for now seems a bit harder to me :) Jul 07 12:07:35 polyonymous: nope, its trivially easy. but if I commit 1st instead of IRC, somebody will lart me ;-) Jul 07 12:07:43 yup, I don't oppose having only necessary stuff installed. Jul 07 12:08:19 well, it might be trivial for you, because you've already spent a lot of time digging it. I've been mostly hacking on the package scope. Jul 07 12:08:26 polyonymous, psokolovsky: task-base for opie-image fails because bluez-dtl1-workaround has not been built. I suspect a missing dependency somewhere. Jul 07 12:08:54 Laibsch: yay on the serial cable, I am currently ipkg upgrading via Connect+ card Jul 07 12:09:20 Laibsch, it didn't fail for me. Well, I think psokolovsky knows better, especially in the light of our recent conversation regarding my experience (or lack thereof) with tasks :)) Jul 07 12:09:50 XorA|gone: Congrats! Great work. Jul 07 12:09:56 Laibsch: there's bug in bitbake then. Wanna have fun to find a workaround for it? Jul 07 12:10:16 Laibsch: I have found a fix, probably not a correct fix, but better than a kick in the stones Jul 07 12:10:16 XorA|gone: Nice to know about one more important piece of hardware being supported. Jul 07 12:10:40 polyonymous: I doubt we had conversation on lack of exp ;-I Jul 07 12:10:48 XorA|gone: Well, for me it is quite OK to get things working first and worry about prettifying them later. Jul 07 12:11:02 ok I've setup a scratchbox env with oe, can I install ipkgs built with oe into the scratchbox env? Jul 07 12:11:18 hi tobb Jul 07 12:11:21 psokolovsky, well, I did state my lack of experience in certain areas, although we didn't quite discuss it, as we've had better subjects :) Jul 07 12:11:25 Laibsch: :) Jul 07 12:11:26 Laibsch: its hard to judge what is wrong, driver is getting an irq before it has initialised, and dereferences a NULL pointer Jul 07 12:11:50 psokolovsky: Why is it a bitbake error? Is the dependency declared? Jul 07 12:12:00 I did not start totally from scratch. Jul 07 12:12:15 what's bluez-dtl1-workaround, anyway? Jul 07 12:12:23 tobb: I am sorry but I know nothing about scratchbox Jul 07 12:12:24 Laibsch: yes Jul 07 12:12:33 polyonymous: see .bb Jul 07 12:12:44 tobb: But there are others around and they might be able to help you. Jul 07 12:12:54 * polyonymous sees .bb Jul 07 12:13:36 seems fine to me. Jul 07 12:13:40 psokolovsky: Any recent changes? Might be that I just missed some update Jul 07 12:14:09 brb Jul 07 12:14:47 Laibsch: that dtl1 things is of course my refactor. see commit list. Jul 07 12:18:43 03rwhitby 07org.oe.dev * rf3312a5f... 10/ (3 files in 3 dirs): openssl: Added configure option for gnueabi-armeb Jul 07 12:18:48 03polyonymous 07org.oe.dev * rcec84fe6... 10/ (3 files in 2 dirs): opie-taskbar-images: update-alternatives for pixmaps. Jul 07 12:19:29 openssl should compile for angstrom/BE now Jul 07 12:19:46 (and therefore make openssh available) Jul 07 12:46:06 mickeyl, ping Jul 07 12:49:04 pöng Jul 07 12:49:24 the u-boot patch took care of one issue Jul 07 12:49:29 leaving the raise problem Jul 07 12:49:38 I think raise is in glibc? Jul 07 12:49:52 some kind of error trapping? Jul 07 12:51:15 I'm thinking we could put the stubs in libarm in u-boot Jul 07 12:51:56 03xora 07org.oe.dev * r0d37beef... 10/ (3 files in 3 dirs): Jul 07 12:51:56 linux-rp_2.6.21.bb : add a hack to prevent connect+ card from oopsing Jul 07 12:51:56 on insert. This is caused by an irq arriving before dev->base_addr Jul 07 12:51:56 has been set so the driver ends up doing an inw(NULL) which oopses. Jul 07 12:51:56 have checked for base_addr == 0 in driver and returned from irq early. This Jul 07 12:51:56 prevents oops and doesnt break with other cards in my collection. Jul 07 12:52:06 03polyonymous 07org.oe.dev * rc2b7bf2b... 10/ (4 files in 3 dirs): opie-icon-reload: adjust to the new filesystem layout Jul 07 12:54:33 Crofton: hmm libarm would make sense. Jul 07 12:54:54 I'll give that a try Jul 07 12:54:59 cool, thanks. keep me posted Jul 07 12:55:04 will do Jul 07 12:57:52 03daniel 07org.oe.dev * r96413855... 10/ (1 packages/db/db3_3.2.9.bb): * Change URL of package db3_3.2.9 to working mirror. (Closes #2606) Jul 07 12:58:15 alphaone: welcome on board. Jul 07 12:58:22 mickeyl: Thanks :-) Jul 07 12:58:30 alphaone: == daniel? Jul 07 12:58:36 yeah Jul 07 12:58:43 cool Jul 07 12:58:50 XorA|gone: more daniel == alphaone :-) Jul 07 12:59:33 stefan_schmidt and alphaone have direct OE access now. The rest of the new OpenMoko team will likely push to an own server which gets fed through me to org.openembedded.dev Jul 07 12:59:44 cool Jul 07 13:02:37 mickeyl: BTW been any improvements on the suspend/resume situation on Neo? Jul 07 13:06:49 XorA|gone: i don't think so. there hasn't been any gta01 kernel work lately. it seems to work ok here except that it doesn't wake up on incoming calls Jul 07 13:06:50 (yet) Jul 07 13:07:13 mickeyl: doesnt wake up here if usb is plugged in :-( Jul 07 13:09:34 XorA|gone: that works here on my bv4 Jul 07 13:22:11 mickeyl: I shall have to build an image later and try again then Jul 07 13:22:18 * XorA|gone goes to do some gardening Jul 07 13:22:39 oki Jul 07 13:22:43 l8er Jul 07 13:23:41 alphaone: BTW the style of commits is package.bb|file|directory : message Jul 07 13:24:38 http://www.openembedded.org/wiki/CommitPolicy Jul 07 13:24:46 yo woglinde Jul 07 13:24:46 args Jul 07 13:24:50 XorA|gone: Okay, thanks. I'll keep that in mind Jul 07 13:24:54 I am sucking Jul 07 13:25:08 alphaone: Please don't forget to actually close the bug. Jul 07 13:25:12 welcome aboard Jul 07 13:26:00 woglinde: hi Jul 07 13:26:05 alphaone: welcome to #oe-nag :-) Jul 07 13:26:19 Laibsch: Right. Let me find the tab. Jul 07 13:26:19 woglinde: hi Jul 07 13:26:31 mrdata I broke oe Jul 07 13:26:45 why? Jul 07 13:26:46 !oebug 2606 Jul 07 13:26:48 * * Bug 2606, Status: NEW, Created: 2007-07-05 15:08 Jul 07 13:26:49 * * jpuhlman(AT)mvista.com: Source for db3 archive seems to not be avalible any more. Jul 07 13:26:50 * * http://bugs.openembedded.org/show_bug.cgi?id=2606 Jul 07 13:26:57 mrdata libiconv Jul 07 13:27:01 alphaone: there you go Jul 07 13:27:03 don't worry Jul 07 13:27:08 false dependency Jul 07 13:27:11 stuff like that happens Jul 07 13:27:15 yo Jul 07 13:27:21 thankfully we have enough eyes on nowadays Jul 07 13:27:23 ups Jul 07 13:27:27 s/on/open/ Jul 07 13:27:49 Laibsch: Got it already thanks. The page was open most of yesterday, but there I lacked commit access. Jul 07 13:28:10 mickeyl: time for some ruby dude to write a mtn commit checker for common misakes like libiconv DEPENDS Jul 07 13:28:24 alphaone: In case you are interested in more quick to fix things, look at bugs 2228 and 2194. Jul 07 13:28:41 xora python Jul 07 13:28:58 or change overnight to ruby? Jul 07 13:29:01 woglinde: isnt mtn scripts ruby Jul 07 13:29:08 hm oh Jul 07 13:31:00 psokolovsky: I do think something is wrong with the dtl stuff Jul 07 13:31:14 I don't have enough computing power at the moment to restart from scratch Jul 07 13:31:38 Laibsch, have you tried to build it explicitly? Jul 07 13:31:40 But "bitbake -g opie-image; grep dtl task-depends.dot" does not return anything Jul 07 13:31:49 When it should Jul 07 13:31:54 XorA|gone: lua, rather Jul 07 13:32:00 * * OE Bug 2606 has been RESOLVED (FIXED) by daniel(AT)totalueberwachung.de Jul 07 13:32:02 * *  Source for db3 archive seems to not be avalible any more. Jul 07 13:32:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2606 Jul 07 13:32:04 since the native monotone script language is lua Jul 07 13:32:23 polyonymous: The build runs fine after I have built the dtl thing explicitly Jul 07 13:32:26 XorA|gone: actually a good idea -- adding some static tests to pre-commit Jul 07 13:32:33 * mickeyl notes this idea Jul 07 13:32:41 makes our Q/A guys happy Jul 07 13:32:47 Laibsch, ok. Just wanted to make sure. Jul 07 13:34:10 Laibsch: Okay, thanks. I have enough to do at the moment, but I bookmarked the pages. Jul 07 13:34:36 btw. Jul 07 13:34:46 lm-sensors-app needs some iconv love as well Jul 07 13:34:52 psokolovsky: I am building for collie/angstrom/.dev Jul 07 13:36:03 psokolovsky, where is it mentioned in deps? Jul 07 13:36:47 ah Jul 07 13:36:54 in kernel.bbclass Jul 07 13:37:27 botmail read Jul 07 13:37:38 ~botmail read Jul 07 13:39:03 how can I prevent oe to start from the first tast if I comment out something which is in task 332? I only added a #SRC_URI_append_simpad in qte. do I need to use a command switch or something? Jul 07 13:40:01 woglinde: is it possible to have a simpad 2.4/2.6 switch in qte-mt? Jul 07 13:40:17 woglinde: or should we drop simpad 2.4 support? Jul 07 13:40:28 mr_nice no I will solve it with lambdakalkuel Jul 07 13:40:53 we read the status from bb-variable Jul 07 13:41:06 and then decide on it which file we chosse Jul 07 13:41:16 okay Jul 07 13:41:23 I have to go shopping now Jul 07 13:41:30 woglinde: ok :) Jul 07 13:41:32 * mickeyl prepares for bbq'ing Jul 07 13:41:34 l8er Jul 07 13:42:06 mr_nice make the patch and provide it to me Jul 07 13:42:12 than I will put it in Jul 07 13:44:16 woglinde: hm, my bitbake knowledge is very low Jul 07 13:44:34 woglinde: but I will try it Jul 07 13:53:46 typical I get an electric strimmer out and it starts pissing down Jul 07 14:00:15 03rwhitby 07org.oe.dev * rc3160bba... 10/ (1 packages/gcc/gcc-cross-kernel-3.4.4_csl-arm-2005q3.bb): gcc-cross-kernel: Added specific csl 2005q3 version for compiling a kernel to match the Freecom FSG-3 vendor binary modules. Jul 07 14:00:20 03rwhitby 07org.oe.dev * r252ea2a5... 10/ (1 packages/linux/ixp4xx-kernel.inc): ixp4xx-kernel: Support building for the fsg3 machine Jul 07 14:00:24 03rwhitby 07org.oe.dev * rc4fbf610... 10/ (1 packages/arm-kernel-shim/arm-kernel-shim_1.5.bb): arm-kernel-shim: Support building for the fsg3 machine Jul 07 14:00:30 03rwhitby 07org.oe.dev * rb25bbba0... 10/ (4 files in 3 dirs): ixp4xx-npe: Now builds NPE-C as well as NPE-B ethernet microcode. Jul 07 14:00:37 03rwhitby 07org.oe.dev * ref8fb5f7... 10/ (1 conf/machine/fsg3be.conf): fsg3be: Machine configuration for the Freecom FSG-3 NAS device. Jul 07 14:17:05 what does ?= in a .conf file? example is conf/machine/simpad.conf MACHINE_KERNEL_VERSION ?= "2.4" Jul 07 14:17:59 mr_nice: only set if not set Jul 07 14:18:15 XorA|gone: thx Jul 07 14:42:15 hey koen|inverness Jul 07 14:42:38 hi all Jul 07 14:43:05 koen|inverness: I got that csl gcc-cross-kernel thing to work finally. Jul 07 14:43:16 rwhitby: ah, cool Jul 07 14:43:30 I'm currently running angstrom/be on the freecom fsg3 Jul 07 14:44:07 XorA|gone: it's cool that the GNER trains in scotland have wifi, but I suck for not noticing I was using the wifi in the train next to me :) Jul 07 14:44:53 Laibsch, polyonymous: Yes, I reproduced it, will fix (well, workaround) soon. Jul 07 14:45:22 ;-) Jul 07 14:45:59 koen|inverness: which endianness do you run ixp4xx angstrom on that board you have? Jul 07 14:46:10 rwhitby: BE Jul 07 14:46:33 "boots, busybox prompt, powerdown" Jul 07 14:46:38 that's what I did Jul 07 14:47:38 koen|inverness: I checked in a patch for openssl which then allows you to use openssh for gnueabi-armeb Jul 07 14:48:27 Freecom have chosen ixp4xx BE eabi for the 2.6 kernel version of their firmware for the FSG-3. Jul 07 14:51:35 * koen|inverness vpns Jul 07 14:53:14 hmmm Jul 07 14:53:16 I fear I sent a mail to oe-devel twice Jul 07 14:54:03 koen|inverness: your penance will be use to a windows mobile device as your sole internet access for the next 24 hours Jul 07 14:54:14 s/use to/to use/ Jul 07 14:56:02 * koen|inverness_ only brough a powerbook and a n800 Jul 07 14:56:30 koen|inverness: loaded up the new firmware yet? Jul 07 14:56:49 (or do you run angstrom on the n800?) Jul 07 14:57:13 * dcordes wants an n800, too Jul 07 14:57:25 the n800 was intended to be my angstrom toy, but the 770 kernel fscked up the mmc card, so the n800 is still running the old firmware Jul 07 14:59:37 is this (http://pastebin.ca/607331) ok If I want to check the kernel version and decide to apply a patch or not? Jul 07 15:00:11 polyonymous: that postinst is clever. I actually didn't make it myself if a postinst script can figure out which package it acts on, and get list of files in that package. Jul 07 15:00:32 hi koen sorry, that I wasnt carefully enough Jul 07 15:01:45 woglinde: just please make sure you make it right and re-push again, it's of course useful change. Jul 07 15:02:05 mr_nice that would work but with lambda calculus its one liner Jul 07 15:02:24 woglinde: how to do it with lambda? Jul 07 15:02:42 psokolovsky I didnt remeber that libiconv was virtual after doing last thinks with uclibc Jul 07 15:03:36 polyonymous: Other thing I was thinking of is generic template substitution support for OE Jul 07 15:05:15 rwhitby: my penance is using OWA as mailer right now :( Jul 07 15:05:20 ~lart OWA Jul 07 15:05:20 * ibot_ --purges OWA Jul 07 15:05:26 03tmbinc 07org.oe.dreambox * rb80dbc7e... 10/ (1 packages/dreambox/dreambox-dvb-modules.bb): dreambox-dvb-modules: use wildcard instead of explicit module list Jul 07 15:05:33 03pfalcon 07org.oe.dev * r9e44e4ad... 10/ (1 packages/tasks/task-base.bb): Jul 07 15:05:33 task-base.bb: Add explicit dependency on bluez-dtl1-workaround to task-base-bluetooth. Jul 07 15:05:33 * As a workaround for bitbake's issued with handling DEPENDS on a PACKAGES_DYNAMIC-generated Jul 07 15:05:33 package. Jul 07 15:09:31 SRC_URI_append_simpad = "file://devfs.patch;patch=${@['0','1'][bb.data.getVar('MACHINE_KERNEL_VERSION',d,1)== "2.6"] Jul 07 15:09:41 ups Jul 07 15:09:50 } \ Jul 07 15:10:05 woglinde: no problem, sorry for the blunt dissapprove Jul 07 15:10:14 koen no you it was right Jul 07 15:10:18 no problem for me to fix it Jul 07 15:10:24 and check in again Jul 07 15:14:02 woglinde: thx :) that one real looks quite better Jul 07 15:14:15 let me test it Jul 07 15:14:18 *g* Jul 07 15:14:23 hm the \ is wron Jul 07 15:14:25 g Jul 07 15:14:27 the last one Jul 07 15:15:05 there is a " missing before it Jul 07 15:15:07 or? Jul 07 15:15:20 yes Jul 07 15:15:28 } " Jul 07 15:15:39 "file://devfs.patch;patch=${@['0','1'][bb.data.getVar('MACHINE_KERNEL_VERSION',d,1)== "2.6"]} " Jul 07 15:17:45 ok, can you add it to qte-common? Jul 07 15:18:20 hm Jul 07 15:18:23 first try wron Jul 07 15:18:24 g Jul 07 15:18:47 sure Jul 07 15:18:51 I am working on it Jul 07 15:18:53 koen|inverness_: hahahahahahah Jul 07 15:18:59 without the 2.6 without the "" ? Jul 07 15:19:50 s/without the 2.6/2.6 Jul 07 15:20:09 I realy need to read a python book Jul 07 15:20:24 hm thats not python ist lambda calculus Jul 07 15:20:48 but you hardly remeber which postion is true or false Jul 07 15:28:31 woglinde, mr_nice: If that's all about qte, I suggest dropping that patch at all. there's no need to support legacy stuff like 2.4 if there's 2.6 available. Jul 07 15:29:02 and the whole opie build moves towards machine-independence anyway Jul 07 15:33:19 re Jul 07 15:36:29 psokolovsky hm not all is working on 2.6 right now Jul 07 15:36:44 sound for example Jul 07 15:36:59 hm Jul 07 15:37:05 woglinde: so invest time into 2.6, not patching qte Jul 07 15:39:30 psokolovsky whats about odevice_simpad.cpp can we kill the 2.4 support there? Jul 07 15:40:01 mr_nice: well, I'm concerned with the state and shape of .bb Jul 07 15:40:41 mr_nice: 's. *.cpp is generally upstream matter, you should put it up with them Jul 07 15:41:20 mr_nice: but in OE.dev, there're almost no 2.4 devices left, so maintaining stuff for them simply doesn't scale Jul 07 15:41:20 psokolovsky the general question is killing 2.4 support in opie or not Jul 07 15:42:25 psokolovsky I am myself vote for not supporting 2.4 after 2.6 works well with opie Jul 07 15:43:04 mr_nice: and OPIE in OE is not in perfect shape, and the only way to keep it maintained is to reduce number of configuration. And my aim is to reduce it to 1 (one). Jul 07 15:43:13 mr_nice: nice Jul 07 15:43:31 hm Jul 07 15:43:33 okay Jul 07 15:43:40 than drop the devfs.patch Jul 07 15:44:25 why the hack bb ignores patch=0 Jul 07 15:48:32 woglinde: read patch.bbclass Jul 07 15:48:44 woglinde: OE just checks for patch=, nothing else Jul 07 15:50:05 hm Jul 07 15:50:23 oh okay Jul 07 16:03:08 woglinde: ok, agreed. next time I'll do cleanup on qte-mt, I'll do that. Jul 07 16:03:59 psokolovsky okay Jul 07 16:04:47 re koen Jul 07 16:21:02 psokolovsky, I think there's no need for generic substitution, because the engine is there - sed. But it may make sense to employ more seddings in some classes. Jul 07 16:21:24 psokolovsky, and what exactly do you mean by your previous statement? :) Jul 07 16:48:26 hi Jul 07 16:48:33 hi Jul 07 16:48:43 hi Jul 07 16:48:46 hi Jul 07 16:48:50 :) Jul 07 16:48:52 :) Jul 07 17:21:18 polyonymous: more generic syntax/semantics for example Jul 07 17:21:49 psokolovsky, no, the one before. I was simply unable to fully parse your sentence :) Jul 07 17:22:40 well, nevermind Jul 07 17:22:49 ok :) Jul 07 17:23:17 I've just pushed like 10 commits adjusting stuff installed with opie-image, mostly postinsts/rms Jul 07 17:23:44 actually, I've done grep -rl /opt/Qt /usr :) Jul 07 17:25:35 psokolovsky, do we want to remove /opt/QtPalmtop/lib from glibc's ld.so.conf ? Jul 07 17:26:16 hm Jul 07 17:26:22 sounds like good idea Jul 07 17:26:55 polyonymous: yes Jul 07 17:27:02 I'm only asking because (a) glibc is a big package (I definitely won't bump PR for this, anyway) and (b) it's harmless. Jul 07 17:27:04 Ok. Jul 07 17:54:54 03tmbinc 07org.oe.dreambox * re38c5875... 10/ (1 conf/distro/opendreambox-1.4.conf): opendreambox-1.4: pin down zope,twisted,twisted-web,twisted-web2 versions Jul 07 17:55:00 03woglinde2 07org.oe.dev * rf2cf538e... 10/ (3 files in 3 dirs): Jul 07 17:55:00 Shame on me that I broke oe, now I will be more Jul 07 17:55:00 carefully what is virtual and whats not Jul 07 17:55:00 same update fot obexftp but now depends Jul 07 17:55:00 on virtual/libiconv Jul 07 17:55:05 03woglinde2 07org.oe.dev * r32f3152e... 10/ (3 files in 3 dirs): Jul 07 17:55:05 fixing ntp building for uclibc Jul 07 17:55:07 discussion about the problem is here Jul 07 17:55:09 https://lists.ntp.org/pipermail/bugs/2007-March/005578.html Jul 07 17:55:11 and the sugesstion for the fix is from here Jul 07 17:55:13 http://blackfin.uclinux.org/gf/project/uclinux-dist/scmsvn/?action=browse&path=%2Ftrunk%2Fuser%2Fntp%2Futil%2Ftickadj.c&view=markup Jul 07 17:55:16 03polyonymous 07org.oe.dev * rd94ee58f... 10/ (4 files in 4 dirs): opie-alarm: adjust for the new filesystem layout Jul 07 17:55:24 03polyonymous 07org.oe.dev * r850db344... 10/ (3 files in 2 dirs): opie-brightnessapplet: adjusted to suit new layout Jul 07 17:55:27 03polyonymous 07org.oe.dev * r3fc4150d... 10/ (3 files in 2 dirs): opie-clockapplet: make it ready for the new layout Jul 07 17:55:34 03polyonymous 07org.oe.dev * r39a544cf... 10/ (3 files in 2 dirs): opie-handwriting: welcome to /usr Jul 07 17:55:40 03polyonymous 07org.oe.dev * ra187dbd6... 10/ (3 files in 2 dirs): opie-handwriting: actually bump PR Jul 07 17:55:54 03polyonymous 07org.oe.dev * r5c9e8cec... 10/ (3 files in 2 dirs): opie-irdaapplet: now aware of filesystem layout too Jul 07 17:56:04 03polyonymous 07org.oe.dev * rbaa7958e... 10/ (3 files in 2 dirs): opie-multikey: get the idea of proper filesystem layout Jul 07 17:56:09 03polyonymous 07org.oe.dev * r36d1f032... 10/ (3 files in 2 dirs): opie-multikeyapplet: we're not in /opt anymore Jul 07 17:56:14 03polyonymous 07org.oe.dev * r8e380128... 10/ (3 files in 2 dirs): opie-pcmciaapplet: qcop is in ${bindir} now, isn't it? Jul 07 17:56:21 03polyonymous 07org.oe.dev * r92ffad42... 10/ (3 files in 2 dirs): opie-pickboard: it's time to be aware of the sane filesystem layout Jul 07 17:56:25 wau Jul 07 17:56:26 spam Jul 07 17:56:28 ;) Jul 07 17:56:33 sorry for flooding :)) Jul 07 17:56:36 :)) Jul 07 17:56:38 *g* Jul 07 17:56:38 03polyonymous 07org.oe.dev * r74726bc2... 10/ (3 files in 2 dirs): opie-screenshotapplet: opie is not in /opt anymore Jul 07 17:56:45 btw, whats the purpose of opie-pcmciaapplet ? Jul 07 17:56:48 03polyonymous 07org.oe.dev * r7ec96152... 10/ (3 files in 2 dirs): opie-volumeapplet: adjust to the new filesystem layout Jul 07 17:56:55 03polyonymous 07org.oe.dev * r57e6c5f5... 10/ (3 files in 2 dirs): opie-clipboardapplet: get /usr-ready. Jul 07 17:57:07 03polyonymous 07org.oe.dev * r955a1cc5... 10/ (6 files in 6 dirs): glibc: remove /opt/QtPalmtop/lib from ld.so.conf Jul 07 17:57:08 darkschneider, iirc it doesn't work with recent kernel Jul 07 17:57:11 oops Jul 07 17:57:14 03polyonymous 07org.oe.dev * r3c0a5ec7... 10/ (3 files in 2 dirs): opie-memoryapplet: adjust to the proper prefix Jul 07 17:57:17 Darth_Wader, that was for you :) Jul 07 17:57:22 03polyonymous 07org.oe.dev * rd38a87ce... 10/ (3 files in 2 dirs): opie-batteryapplet: adjust to the sane prefix Jul 07 17:57:46 :) Jul 07 17:57:50 polyonymous oh why? Jul 07 17:58:01 whats wrong with pcmciaaplett? Jul 07 17:58:01 polyonymous: so if it not working, maybe kick it off from opie-image then? Jul 07 17:58:03 woglinde, Or was it not kernel but pcmcia stuff Jul 07 17:58:11 hm? Jul 07 17:58:14 hm right Jul 07 17:58:21 pcmciautils vs. pcmcia-cs Jul 07 17:58:28 yes, exactly. Jul 07 17:58:29 I will look at it Jul 07 17:58:29 "unable to infer tagged configuration" reported when apmd is compiled. Jul 07 17:58:39 http://bugs.openembedded.org/show_bug.cgi?id=2060 Jul 07 18:18:38 according to cia I've just hit my 100th commit :)) Jul 07 18:19:35 grats :) Jul 07 18:19:41 opie-pcmcia-applet uses the obsolete ioctl Jul 07 18:20:03 koehn hm Jul 07 18:20:40 Hello :-) Jul 07 18:21:05 OPcmciaSystem::instance()->socket( socket )->reset(); Jul 07 18:21:19 I think thats abstract enough Jul 07 18:24:26 http://handhelds.org/cgi-bin/cvsweb.cgi/opie/libopie2/opiecore/linux/opcmciasystem.cpp?rev=1.15&content-type=text/x-cvsweb-markup Jul 07 18:24:30 libopie2 Jul 07 18:26:45 //TODO: Revise for pcmciautils Jul 07 18:30:46 Hi everyone. Little problem here : I gave X.org a try on my SL-C1000, and it works like a charm. Except for one thing : the touchscreen (used through evdev) works pretty well, but doesn't generate ButtonPress events. Do you know where this might come from ? Jul 07 18:33:49 Darth_Wader, thanks :) Jul 07 18:35:07 hi florian Jul 07 18:35:20 Darth_Wader, but I'm going for 101th with restartable opie :) Jul 07 18:35:56 seems opie needed sysfs support Jul 07 18:36:07 it needs way more :) Jul 07 18:36:27 dont know if its worth Jul 07 18:36:45 good question.... I would only fix it if it would fix something I need. Jul 07 18:38:13 Darth_Wader, feel free to pull opie-taskbar fix Jul 07 18:44:56 so i`ll rebuild opie-image then and see what will be broken this time :) Jul 07 18:45:45 Darth_Wader, yes. although you can just rebuild opie-taskbar and ipkg upgrade :) Jul 07 18:45:59 but with opie-image you will, of course, get much more fixes. Jul 07 18:46:15 I'm not done with your whole pastebin yet, though :) Jul 07 18:46:30 :) Jul 07 18:47:13 I'm at opie-dvorak now. Jul 07 18:56:39 Crofton: http://freedesktop.org/wiki/GstOpenMAX Jul 07 18:57:24 03polyonymous 07org.oe.dev * rad674e02... 10/ (3 files in 3 dirs): opie-taskbar: restore opie's restartability Jul 07 18:57:28 03polyonymous 07org.oe.dev * rb8f3ceea... 10/ (5 files in 3 dirs): opie-dvorak: adjust post{inst,rm} and rpath to the new layout Jul 07 18:57:33 03polyonymous 07org.oe.dev * r9d3b634e... 10/ (3 files in 2 dirs): opie-lockapplet: get out of /opt Jul 07 18:57:40 03polyonymous 07org.oe.dev * rdddaf7a1... 10/ (3 files in 2 dirs): opie-unikeyboard: get into /usr Jul 07 19:00:48 Bernardo: if you build a kernel, Connect+ should work Jul 07 19:09:35 * koen|inverness pats belly Jul 07 19:09:39 XorA|gone: scotland rocks Jul 07 19:09:43 :))) Jul 07 19:11:48 koen|inverness: you will become a fat bastard at that rate :-) Jul 07 19:13:41 hmm, is there online browser for oe repo? Jul 07 19:13:54 XorA|gone: not if I climp holyrood hill again :) Jul 07 19:18:39 as there is no such, I assume koen means the Crags or Arthurs Seat Jul 07 19:24:12 probably Jul 07 19:24:27 that big thing next to the parliament Jul 07 19:31:35 1/wi 52 Jul 07 19:31:43 argh Jul 07 19:34:12 what's a 1/wi 52? :) Jul 07 19:42:20 typo of /wi 52 in irssi Jul 07 19:42:25 ah :) Jul 07 19:52:23 polyonymous: great work! Jul 07 19:52:36 psokolovsky, :) at least a lot of :) Jul 07 19:59:33 03polyonymous 07org.oe.dev * ra1f836cf... 10/ (3 files in 3 dirs): opie-mediaplayer1-libmadplugin: install codec into $palmtopdir Jul 07 19:59:38 03polyonymous 07org.oe.dev * rda1379d3... 10/ (3 files in 3 dirs): opie-mediaplayer1-libtremorplugin: install into ${palmtopdir} Jul 07 19:59:44 03polyonymous 07org.oe.dev * re2afe354... 10/ (4 files in 3 dirs): opie-mediaplayer1-libwavplugin: go ${palmtopdir} way Jul 07 20:00:50 I'm afraid I'm going to revert these three commits. I'm afraid those weren't really broken :) Jul 07 20:03:00 heh heh Jul 07 20:03:35 I got a ls /opt/QtPalmtop from Darth_Wader and started to eagerly fix stuff. No matter if it's broken or not :) Jul 07 20:04:03 I suppose someone will have to remember to set palmtopdir in sharprom-compatible.conf Jul 07 20:04:19 :) Jul 07 20:04:29 do really people use it? If they do, they're welcome to fix :) Jul 07 20:05:01 polyonymous: I only thought of it because I using cacko at moment Jul 07 20:05:58 XorA|gone, well, if you will build for cacko, I'm sure you won't have any troubles fixing that :) I don't feel like doing even trivial fixes blindly without testing. Jul 07 20:06:39 polyonymous: nah I dont build for it, but there is a sharpromfeed project on l2g Jul 07 20:06:57 so they're going to have fun fixing it :) Jul 07 20:08:20 is there a way to revert certain commit in the workspace? Jul 07 20:08:31 mtn revert Jul 07 20:08:55 XorA|gone, doesn't it revert to your current head? Jul 07 20:08:55 oh sorry misread mtn dissaprove or something like that Jul 07 20:09:11 I think disapprove operates not on the workspace... Jul 07 20:09:19 sounds like mtn diff | patch -R :) Jul 07 20:09:25 polyonymous: dissaprove then mtn revert Jul 07 20:09:46 I don't want to pollute oe-commits even more :) Jul 07 20:10:01 and to each disapprove I want to still add a pr bump. Jul 07 20:10:26 ah then by hand is probably the way then Jul 07 20:10:30 yup Jul 07 20:10:46 I think I don't even need -R in patch if I specify correct diff :) Jul 07 20:11:48 Darth_Wader, can you ipkg search /opt/QtPalmtop/lib/libftplib.so.1.0.0 ? Jul 07 20:12:12 says nothing Jul 07 20:12:30 darkschneider, then it's just a stray file... Jul 07 20:12:35 Darth_Wader, I mean :) Jul 07 20:12:55 hmm.. i`ll look at the package content now Jul 07 20:13:25 It isn't mentioned in any package .list, since your grep gave me the info on that. Jul 07 20:14:12 hmm Jul 07 20:16:19 yes, it was fixed in 20070703 and those file seems to left after libftplib1_1.2.2+cvs20070428-r0_armv5te.ipk Jul 07 20:16:59 ok. now I'll clean up these three commits which also seem to be of no use... Jul 07 20:19:38 btw, the /dev/mtdblock* driver was turned off in kernel config at least for h2200 Jul 07 20:26:00 * * OE Bug 1570 has been REOPENED by utx(AT)penguin.cz Jul 07 20:26:02 * *  ncurses-native-5.4-r8 OE install fails to place libncurses.so. 5 correctly Jul 07 20:26:04 * * http://bugs.openembedded.org/show_bug.cgi?id=1570 Jul 07 20:41:48 I hit the RTLD_SINGLE_THREAD_P problem in glibc-intermediate 2.6; has this been resolved? I just did a pull and mtn update with no success ('openmoko')... trying angstrom 2007.1 now, too. Jul 07 20:43:06 tnb: angstrom uses glibc2.5 Jul 07 20:43:33 03polyonymous 07org.oe.dev * raf468c48... 10/ (8 files in 7 dirs): opie-mediaplayer1-lib{wav,mad,tremor}plugin: revert unneded fixes, except for PR bumps Jul 07 20:43:39 03polyonymous 07org.oe.dev * rb83fe923... 10/ (1 packages/opie-mailapplet/opie-mailapplet_cvs.bb): opie-mailapplet: bump PR to force upgrade to the new layout Jul 07 20:44:12 psokolovsky: ok, I'll try PREFERRED_VERSION for now Jul 07 20:44:39 polyonymous: well, why those empty PR bumps? I bumped SRCDATE for the whole OPIE after making those changes, that's scalable way to propogate them. Jul 07 20:46:30 psokolovsky, well, despite your bump Darth_Wader managed to have wrongly built packages, so I thought it wouldn't hurt. Jul 07 20:47:47 polyonymous: but that's monkey's work. either bump SRCDATE again, or tell users that they should rebuild from scratch. Jul 07 20:48:08 psokolovsky, that's about mailapplet. As for the rest - I just don't want to be these PRs reused. Jul 07 20:48:47 psokolovsky, I think generally you're right, but I don't think these should be reverted :) Again, there was only one really empty pr bump. Jul 07 20:49:32 polyonymous: just trying to save you a bit of time ;-) Jul 07 20:50:03 psokolovsky, I definitely don't go into .bb for the sole purpose of bumping PR normally, but thanks :) Jul 07 20:50:09 instead of bumping 100's of PR's, bump more priority component, for CVS version it's easy. Jul 07 20:50:38 of course. It' just that I didn't exactly go around bumping PRs :) Jul 07 20:50:57 ok Jul 07 20:52:45 morning Jul 07 20:53:00 hi hrw Jul 07 20:53:14 morning Jul 07 20:53:16 XorA|wifi: gratz on kernel fix Jul 07 20:53:38 * hrw -> tv Jul 07 20:54:28 hrw|tv: s/fix/hack/ Jul 07 20:55:11 XorA|gone: I am pulling to check how it is done Jul 07 21:00:30 XorA: sent it to hostap devel ML? Jul 07 21:01:14 hrw|tv: not yet, I was hoping someone was already signed up to save me joining yet another list Jul 07 21:02:17 hrw|tv, XorA: a fix for the sandisk connect plus? Jul 07 21:02:35 Bernardo: yes Jul 07 21:02:39 I was just starting to look at the kernel panic output... Jul 07 21:02:42 Bernardo: s/fix/hack/ Jul 07 21:02:59 Bernardo: the cause is an irq happens before dev structure is filled Jul 07 21:03:04 XorA: send to list and add 'please cc: me as I am not subscribed' Jul 07 21:03:16 has it been pushed in? I'd like to give it a try Jul 07 21:03:18 I do that with my patches Jul 07 21:03:25 Bernardo: yes it is pushed Jul 07 21:03:27 Bernardo: it is - 2.6.21 kernel Jul 07 21:03:36 as my akita seems peculiar with some stuff (namely pxa fb) Jul 07 21:04:20 time to build... :) Jul 07 21:05:52 it's always time to build! :) Jul 07 21:07:13 I just had to blast my whole build dir yesterday due to all these opie changes... Jul 07 21:07:39 Bernardo, and they still keep coming :) But why blast build dir? Jul 07 21:08:19 because iqnotes, kdepimpi and other qpe stuff was installing in /opt/qtpalmtop Jul 07 21:08:37 Bernardo, I'm fixing kdepimpi right now :) Jul 07 21:08:40 and I decided to "sanitize" everything and start from scratch Jul 07 21:08:52 kdepimpi installs into /usr/bin if you start from scratch Jul 07 21:09:03 and works Jul 07 21:09:10 Bernardo, well, yes, but it installs binaries into wrong location. Jul 07 21:09:53 I still have to find where does opie reset the date&time after a reboot, instead of using the rtc clock Jul 07 21:10:11 does it? Jul 07 21:10:32 if I reboot the system using the reset button, it always puts the date and time to that of the kernel build date Jul 07 21:10:49 (I think, it might be the initrd.bin build date) Jul 07 21:11:13 In those cases I have to do a "hwclock --hctosys" Jul 07 21:11:13 I think it's somewhere in initscripts then Jul 07 21:11:19 not opie. Jul 07 21:11:52 possibly... But I think angstrom-x-image doesn't do that - not that I've tested it much Jul 07 21:12:06 strange... Jul 07 21:12:18 well, my opie image doesn't do it either. Jul 07 21:12:28 But! I think it tries to and can't :) Jul 07 21:14:15 I'm almost ready to upgrade my 'production' Z to use /usr... Jul 07 21:15:41 I probably have to clear my /home Jul 07 21:15:52 do a backup of everything, and start from scratch Jul 07 21:16:03 I think it still has stuff from the collie Jul 07 21:16:13 hrw|tv: crap, hostap isnt a public list Jul 07 21:17:02 kdepimpi is done Jul 07 21:19:31 psokolovsky, should we cleanup children of the '/opt -> /usr' bug? Jul 07 21:19:53 psokolovsky, I mean these: http://bugs.openembedded.org/showdependencytree.cgi?id=2254 Jul 07 21:20:15 * XorA finds a bugzilla though, that'll do Jul 07 21:20:33 XorA: the kernel with your patch is r9 ? Jul 07 21:21:10 Bernardo: yes Jul 07 21:21:32 ok, built and ready to flash... :) Jul 07 21:23:09 dont forget to upgrade the modules as well Jul 07 21:23:24 ~fish XorA Jul 07 21:23:25 * ibot_ slaps XorA around with a large trout Jul 07 21:23:45 ~zap lrg_ Jul 07 21:23:45 * ibot_ takes out a cattle prod and gives lrg_ a good jolt. Jul 07 21:24:21 evening ! Jul 07 21:25:22 XorA: lrg_ having fun? Jul 07 21:26:52 when I want to bitbake a root fs for a 2.4 kernel based machine, say, a netbook, which options would I have ? Jul 07 21:27:21 polyonymous: leave to original submitter for verification? ;-) Jul 07 21:27:43 psokolovsky, :))) that should work :) Jul 07 21:27:50 psokolovsky, opie-init doesn't ship qpe.conf :) Jul 07 21:27:59 * * OE Bug 2613 has been created by  Jul 07 21:28:01 * * collie kernel 2.6.21 and modules Jul 07 21:28:03 * * http://bugs.openembedded.org/show_bug.cgi?id=2613 Jul 07 21:28:06 psokolovsky, and I'm not quite sure why Jul 07 21:28:59 but fix is no problem, tho :) Jul 07 21:29:08 hi, all! Jul 07 21:29:13 hi Jul 07 21:29:16 hey, slapin Jul 07 21:30:20 do anybody remembers some userland app to write to kernel memory (I need that to write PXA regs from userland to aid in GPIO research) Jul 07 21:30:25 Laibsch: does that not mean you just forgot to ipkg upgrade the kernels? Jul 07 21:30:33 Laibsch: modules I mean Jul 07 21:30:52 polyonymous: it should and did Jul 07 21:30:56 or put it another way: Could I use Angstrom to build a root fs for a SA1100 based machine with a 2.4 Kernel ? Jul 07 21:30:59 slapin: mmap /dev/kmem Jul 07 21:31:37 psokolovsky, the one in /usr/lib/opie/etc/skel/ ? It does for me after adding FILES_${PN} += ... Jul 07 21:31:47 or would I have to use openzaurus ? Jul 07 21:32:11 jan_js: there's such option, but it's a perversion. assume it doesn't work because nobody uses it because nobody needs it Jul 07 21:32:21 hmm... Jul 07 21:32:26 slapin: devmem2 is canonical Jul 07 21:32:35 Configuring opie-init Jul 07 21:32:35 System startup links for /etc/init.d/opie already exist. Jul 07 21:32:35 *** glibc detected *** which: free(): invalid next size (fast): 0x000d9088 *** Jul 07 21:32:38 wtf? Jul 07 21:32:46 psokolovsky, it it in OE? Jul 07 21:32:58 sure Jul 07 21:33:02 broken update-rc? Jul 07 21:33:50 psokolovsky: As it is done for the 2.4 kernel collie image, right ? Jul 07 21:33:54 XorA: gmane keeps your sanity when you don't want to subscribe to every ML out there. Jul 07 21:34:15 jan_js: yes, sth like that. angstrom-legacy or how it is called Jul 07 21:34:25 hmm.. running postinst manually works just fine... Jul 07 21:34:51 polyonymous: cleaning up 2254 is probably a good idea. I assume the "child bugs" have been fixed. Jul 07 21:35:05 Laibsch, most of them, likely. Jul 07 21:35:10 polyonymous: so, either something thrashes memory (ipkg good at that), or time to put new dimm in ;-) Jul 07 21:35:11 Laibsch: gmane can only post to public lists Jul 07 21:35:12 psokolovsky: I have seen it, but there is nothing in it so far, isn't it ? Jul 07 21:35:25 Laibsch, you still haven't committed our qpobox/nunome work, I think. Jul 07 21:35:36 Laibsch: I posted to your bug Jul 07 21:36:02 jan_js: as Laibsch if he ever used it. Generally, OE is pretty deep in 2.6 stuff Jul 07 21:36:13 XorA: re kernel, likely. I just did "bitbake console-image" and flashed the latest kernel I had and the latest console image I had in my deploy dir. Anything else I need to do? Jul 07 21:36:38 psokolovsky, put new dimm in qemuarm? Sounds like fun :) Maybe that's because my qemuarm lacks swap :) Anyway, psokolovsky are you sure qpe.conf is shipped? Because I have fixed .bb, but I'm not sure if I should fix what ain't broke :) Jul 07 21:37:34 ok, so, Laibsch, can you give me a hint how to use angstrom-legacy ? Jul 07 21:37:42 XorA: re gmane, not true. You can post to members-only list. That is a setting on gmane. There are some lists that don't want any mail from gmane at all. Those are ro, the others usually are rw. Jul 07 21:38:19 Laibsch: erm, gmane disagrees with you, it sent me an email saying I had to subscribe to the list to post Jul 07 21:38:58 Laibsch: I heard you can you give me a hint how to use angstrom-legacy ? Jul 07 21:39:40 XorA: No, gmane fully agrees with me ;-) You must have hit one of those lists explicitly set to ro, it looks like Jul 07 21:39:45 XorA: which list? Jul 07 21:39:55 polyonymous: apparently, you're right, it is not Jul 07 21:40:02 ah ok, then I'm pushing Jul 07 21:40:05 jan_js: No, I don't remember ever using it. What are you trying to do? Jul 07 21:40:37 Laibsch: Build a root filesystem for the venerable Netbook classic Jul 07 21:40:39 Laibsch: why go through all the crap of typing the message without telling me this info, gman is crap arsed waste of time Jul 07 21:41:18 jan_js: Sorry, no clue Jul 07 21:41:41 pushed Jul 07 21:41:54 XorA: Gmane is one of the finest services around. As I said, it keeps my sanity so I don't need to curse so often like you do ;-) Jul 07 21:42:13 Laibsch: I would prefer to use a 2.6 kernel, but some of the hardware is not supported then Jul 07 21:42:36 Laibsch: each to their own Jul 07 21:43:01 btw, I told you up front. You need to read carefully ;-) Jul 07 21:43:25 jan_js: What does not work Jul 07 21:43:26 ? Jul 07 21:44:50 hrw|tv: What do I do with the modules tar.gz? So far, I considered it superfluous because for 2.6.20 it seems it was. Jul 07 21:45:08 Laibsch: Not shure, touchscreen I think, maybee cf card slot. I didn't try it myself. Jul 07 21:45:48 Bernardo: how goes? Jul 07 21:45:55 jan_js: Oh, you build an image successfully but have trouble when running it? Jul 07 21:47:07 polyonymous: The qpobox stuff is not working right. That is why I did not commit it so far. Jul 07 21:47:43 Laibsch: My info comes from OpenPsion mailing list. They say they can boot a 2.6 kernel, but had troble getting the hardware to work. Jul 07 21:47:45 Laibsch, but how can it start working if the wip work isn't in the tree :) Remember, you've been moving files around Jul 07 21:48:43 polyonymous: ? Jul 07 21:48:54 Laibsch: Then again, there is a patched 2.4.27 kernel that runs very well on the Netbook. Jul 07 21:49:26 Laibsch: So my idea was to use a filesystem like the oe from openzaurus collie image. Jul 07 21:49:41 The changes we made are local here. And I did not feel like committin them before they work. Are suggesting to sync nonetheless "early and often" to work from what we have? Jul 07 21:49:43 Laibsch, iirc, you've been rearranging layout, moving stuff from one directory to another in the tree, so I left it up to you to commit stuff in progress and of course I'm not touching these now Jul 07 21:49:46 oe -> one Jul 07 21:49:47 Laibsch: modules*tgz is for devs only Jul 07 21:50:22 hrw|tv: Well, I don't really know how to get the modules onto the collie, I have to admit Jul 07 21:50:24 Laibsch, we don't have a working version to break, do we? iirc the stuff we've had was... well, better than before our intervention :) Jul 07 21:50:35 polyonymous: I think it was working. Jul 07 21:50:46 But maybe only sometime in the past. I don't remember Jul 07 21:50:57 Laibsch, I mean we don't have working stuff _now_. Jul 07 21:50:59 But there probably aren't that many users besides me. Jul 07 21:51:09 polyonymous: To be honest, I don't know. Jul 07 21:51:24 I#ll just commit so we are sync'ed Jul 07 21:51:27 Laibsch: on cf card? Jul 07 21:51:27 Laibsch, anyway, of course, it's your realm, but I do suggest that you commit it. Jul 07 21:51:46 yup. that would make sense, I think. At least it can be fixed wrt filesystem layout then :) Jul 07 21:51:47 hrw|tv: The initrd.bin and zImage are on the CF, yes. Jul 07 21:52:19 hrw|tv: The modules*.tar.gz is still on my server. I'd need to copy it. But I don't know what to do with it Jul 07 21:52:47 polyonymous: Well, the filesystem layout is of course of utmost importance here ;-) Jul 07 21:53:32 Laibsch, :)))) Well, I can't say I remember it all very clear, but for what I remember we've had stuff that was locking opie up and after patching it was at least harmless :) Jul 07 21:54:06 polyonymous: I have to admit my memory is also very shady on this. Jul 07 21:54:08 konqueror-embedded is totally broken now :) Jul 07 21:54:19 polyonymous: good job! Jul 07 21:54:21 ;-) Jul 07 21:54:34 polyonymous: nunome has no local changes here Jul 07 21:54:37 Laibsch: if you plan to reflash collie then rebuild image to have proper images Jul 07 21:54:38 XorA: haven't tested yet. My daughter woke up and decided to have dinner two hours after her normal sleeping time... Jul 07 21:55:12 hrw|tv: I build the image this afternoon Jul 07 21:55:12 just updated and I'm flashing zimage right now Jul 07 21:55:12 Laibsch, well, of course you only need to take care of the stuff you care about. As for nunome, I can commit what I have here, of course, after I check its fs compliance Jul 07 21:55:13 built Jul 07 21:55:23 Bernardo: heh heh, I cant patch kids sorry Jul 07 21:55:39 polyonymous: I care about nunome and qpobox/poboxserver Jul 07 21:55:43 and a few other things Jul 07 21:55:55 Laibsch, surely you do. At least more than I do:) Jul 07 21:56:12 XorA: :) Jul 07 21:56:26 that's my job, I think... :) Jul 07 21:57:18 03polyonymous 07org.oe.dev * rb9a7c56e... 10/ (3 files in 2 dirs): kdepimpi: install binaries into ${bindir} instead of ${palmtopdir}/bin Jul 07 21:57:24 03polyonymous 07org.oe.dev * r206e5a19... 10/ (1 packages/kdepimpi/pwmpi.inc): pwmpi: move binaries from ${palmtopdir}/bin to ${bindir} Jul 07 21:57:29 03polyonymous 07org.oe.dev * r416abd52... 10/ (4 files in 3 dirs): opie-init: initscript cosmetics and fixed packaging wrt qpe.conf Jul 07 21:58:01 polyonymous: kdepimpi libs to libdir? Jul 07 21:58:21 hrw|tv, I'm not sure about it. Jul 07 21:58:31 after all they're mess :) Jul 07 21:58:46 and not used by anything else. Jul 07 22:01:40 they are Jul 07 22:02:11 XorA: it seems to work Jul 07 22:02:21 now I need to configure wpa_supplicant Jul 07 22:02:55 Bernardo: yay Jul 07 22:02:56 hrw|tv, are mess or are used? :) Jul 07 22:04:19 uh Jul 07 22:04:32 XorA: no memory, only wifi, but I expected that Jul 07 22:04:43 run.do_configure.15973: line 227: oe_runconf: command not found... weird Jul 07 22:05:48 ahh Jul 07 22:06:14 Bernardo: yeah memory is disable by me, as both subsystems are recognised, but neither work in that mode Jul 07 22:06:45 I sometimes managed to have both working Jul 07 22:07:01 but it would enter a race most of the times Jul 07 22:07:26 * Bernardo wonders why the wpa_supplicant package doesn't include /etc/wpa_supplicant.conf Jul 07 22:13:06 polyonymous: pobox stuff is being committed. Jul 07 22:13:06 good night, everyone. Jul 07 22:13:14 ok, night. Jul 07 22:14:33 I'm not having much success with wpa_supplicant - seems like it doesn't know TKIP and CCMP Jul 07 22:15:22 Bernardo: relevant modules loaded? Jul 07 22:15:30 mmm Jul 07 22:15:33 let me check Jul 07 22:15:58 only ieee80211_crypt Jul 07 22:16:44 * XorA finishes typing epic adventure summary for WFRP Jul 07 22:18:52 maybe kernel-module-crypt-tkip and ccmp should be added to the default image... Jul 07 22:20:17 me points Bernardo to Jul 07 22:20:21 !oebug 2477 Jul 07 22:20:22 * * Bug 2477, Status: NEW, Created: 2007-06-12 12:19 Jul 07 22:20:23 * * : RFE: Better wifi support Jul 07 22:20:24 * * http://bugs.openembedded.org/show_bug.cgi?id=2477 Jul 07 22:22:37 networking seems to be an area that needs some commercial polish Jul 07 22:22:48 Bernardo: task-base-wifi contain all wifi stuff Jul 07 22:26:49 thanks, I just added myself to the cc Jul 07 22:31:46 well, time to go. Tomorrow I'll try again to get wpa to work Jul 07 22:31:52 'night Jul 07 22:31:57 * Bernardo is away: Ausente por agora. Jul 07 22:53:54 bye Jul 07 22:54:29 nite hrw Jul 07 22:58:27 03polyonymous 07org.oe.dev * reb704a13... 10/ (3 files in 3 dirs): qpf-font-common: fonts aren't in /opt/QtPalmtop anymore Jul 07 23:11:52 does MACHINE_EXTRA_RRECOMMENDS supposed to work with angstrom-comsole-image? Jul 07 23:12:11 slapin yes should Jul 07 23:13:00 woglinde, I'm unable to make it work, I add modules packages with it and they are not added to image :( Jul 07 23:13:21 slapin hm you could do bitbake angstrom-comsole-image -g Jul 07 23:13:36 and look at the depends.dot Jul 07 23:13:40 if it is there Jul 07 23:13:54 woglinde, I can't find these packages in do_rootfs log :( Jul 07 23:14:11 woglinde, thanks, I'll look into that Jul 07 23:19:39 ipkg are more dumb than should Jul 07 23:21:46 * Darth_Wader can`t understand why ipkg is unable to deal with multiple kernels Jul 07 23:22:07 woglinde, they are inside depends.dot, but not in image :( Jul 07 23:22:32 indeed, ipkg upgrad'ing all modules to a different kernel version sucks Jul 07 23:24:45 unneded a.i suxx Jul 07 23:25:23 I think it would be great if it installed new modules while preserving the old ones. Jul 07 23:25:41 the way they are named makes that impossible :\ Jul 07 23:25:53 what if kernel-modules- had a kernel version in thename? Jul 07 23:26:00 yes, what about changing the naming? Jul 07 23:26:52 it shouldn't be kernel-module-blahh_2.6.20-r7.ipk but kernel-module-blahh-2.6.20_2.6.20-r7.ipk but Jul 07 23:27:05 exactly what I meant, but? Jul 07 23:27:19 or was it a copy'n'paste leftover? :) Jul 07 23:27:22 well, I guess there is some valid reason for doing it this way Jul 07 23:27:42 I think this issue at least can be brought up. Jul 07 23:27:56 after all, maybe there WAS a reason :) Jul 07 23:28:02 it was you typing faster than me, and me overlooking the line ;9 Jul 07 23:28:29 no, I meant the second 'but' Jul 07 23:28:36 slapin ah Jul 07 23:28:44 and heck, do I type fast! :) Jul 07 23:29:02 woglinde, any ideas? Jul 07 23:29:19 slapin not yet I am to tired Jul 07 23:29:26 please ask hrw tomorrow Jul 07 23:29:46 woglinde, ok, thanks! Jul 07 23:29:50 polyonymous: doh! that was a copy'n paste problem ideed Jul 07 23:30:20 heh :) Jul 07 23:31:15 * CoreDump|home is tired Jul 07 23:31:19 so am I Jul 07 23:31:34 30 commits a day keep insomnia away, I should think. Jul 07 23:31:54 30?!? Jul 07 23:31:54 indeed Jul 07 23:32:13 cia says "31 messages so far today" Jul 07 23:32:29 oh, not 30 by you, just in general Jul 07 23:32:43 and I've just pushed a few revisions :) Jul 07 23:32:50 no, by me Jul 07 23:32:54 summatusmentis: polyonymous wants to make first spot on koens weekly commit email =) Jul 07 23:32:55 http://cia.vc/stats/author/polyonymous Jul 07 23:33:01 :))))))))))))))) Jul 07 23:33:11 good lord above... Jul 07 23:33:28 not exactly my intention, but indeed it may happen :) Jul 07 23:34:06 but thanks god koens emails aren't sorted by the number of commits (I think), so we don't have this kind of competition :) Jul 07 23:34:46 lol Jul 07 23:36:24 I think one more commit today and I'm gone to bed. Jul 07 23:36:38 Tomorrow I'm going to upgrade my production Z to the /usr opie layout... Jul 07 23:36:56 how does one search for all bugs? Jul 07 23:37:03 summatusmentis, ALL BUGS? Jul 07 23:37:36 lol, I guess... I need something to work on, and mb-applet-wireless has me stumped beyond belief Jul 07 23:37:46 my goal for the summer is to fix one thing Jul 07 23:37:57 night all Jul 07 23:37:58 any one? Jul 07 23:38:02 night, data Jul 07 23:38:24 polyonymous: yeah, I don't really care what, I just want to be useful :-P Jul 07 23:38:31 and it's a good way to learn Jul 07 23:38:34 hm. Jul 07 23:41:55 summatusmentis, http://bugs.openembedded.org/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&short_desc=&product=Openembedded&long_desc_type=substring&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&emailtype1=substring&email1=&emailtype2=substring&email2=&bugidtype=include&bug_id=&chfieldfrom=&chfieldto=Now&chfieldvalue=&cmdtype=doit&order=Reuse+same+sort+as Jul 07 23:41:55 +last+time&field0-0-0=noop&type0-0-0=noop&value0-0-0= ? :) Jul 07 23:42:34 ohmygod... Jul 07 23:42:35 that was a long url :) Jul 07 23:44:03 no kidding Jul 07 23:46:10 oh well, I'm gone, night all Jul 07 23:58:57 03Laibsch 07org.oe.dev * r5076ed26... 10/ (10 files in 5 dirs): Jul 07 23:58:57 qpobox/poboxserver: some unfinished changes Jul 07 23:58:57 * moved qpobox stuff into poboxserver directory, they belong together Jul 07 23:58:57 * tried to move the combo to unicode with polyonymous' help Jul 07 23:58:57 -> unfinished Jul 07 23:59:02 03polyonymous 07org.oe.dev * r14c9d8e8... 10/ (1 packages/konqueror/konqueror-embedded_20070212.bb): konqueror-embedded: fixed and adjusted to the new filesystem layout Jul 07 23:59:10 03polyonymous 07org.oe.dev * rde51552e... 10/ (3 files in 2 dirs): opie-autorotateapplet: go ${bindir} Jul 07 23:59:15 03polyonymous 07org.oe.dev * r4b50a70c... 10/ (5 files in 3 dirs): opie-keyboard: adjust post{inst,rm} and rpath to the new layout Jul 07 23:59:20 03polyonymous 07org.oe.dev * r0ff0f216... 10/ (3 files in 2 dirs): opie-keyview: adjust to the new filesystem layout Jul 07 23:59:25 03polyonymous 07org.oe.dev * r9a0ea092... 10/ (1 packages/opie-tinykate/opie-tinykate.inc): opie-tinykate: removed multiple occurences of package in PACKAGES Jul 08 00:16:00 polyonymous, 8-O what did you do to our beloved opie installed in /opt Jul 08 00:16:07 polyonymous, hi btw Jul 08 00:16:21 he moved it out of /opt Jul 08 00:18:00 Darth_Wader, I noticed ... its written three lines before what I said Jul 08 00:18:35 :) Jul 08 00:19:03 btw, looks like that CIA-3 have some big delay in querying the changelogs.. Jul 08 00:27:59 * * OE Bug 2614 has been created by  Jul 08 00:28:01 * * MACHINE_EXTRA_RRECOMMENDS doesn't work Jul 08 00:28:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2614 **** ENDING LOGGING AT Sun Jul 08 02:59:57 2007