**** BEGIN LOGGING AT Wed Oct 04 02:59:57 2006 Oct 04 03:08:38 hi. folks. Oct 04 05:15:49 Good Morning Oct 04 05:18:45 One question, I'm building "world" right now, and I think that I'm going to be running out of space in a while, so, is deleting my "..tmp/work/*" folders going to bring dependacy problems? Oct 04 05:19:49 Do I need the everything under "work" to build other packages, or will it have no effect? Oct 04 05:24:30 I'll be AFK, please answer my question, if someone of you knows. Thanks Oct 04 05:34:05 morning Oct 04 05:34:13 Hello Oct 04 05:34:19 B_Lizzard: add INHERIT += "rm_work" into local.conf Oct 04 05:34:35 And that will clean up after itself? Oct 04 05:34:43 B_Lizzard: its basically save to remove tmp/work - if something will break then give us info Oct 04 05:34:54 B_Lizzard: rm_work remove WORKDIR after building Oct 04 05:35:09 I'm building world, so CTRL+C is ok, I guess Oct 04 05:35:14 yes Oct 04 05:35:31 Thanks a lot Oct 04 05:35:35 np Oct 04 05:39:22 hi mithro Oct 04 05:39:25 hi ArjanS Oct 04 05:39:28 ar_: hi Oct 04 05:39:31 ~lart my tab Oct 04 05:39:31 * ibot cuts off my tab's head with a halberd that could have been a little bit sharper Oct 04 06:08:14 morning Oct 04 06:08:14 hi cyrilRomain Oct 04 06:08:14 hi hrw Oct 04 06:12:34 hi Oct 04 06:12:53 hi obergix Oct 04 06:13:00 time to pack a bit Oct 04 06:13:51 cu Oct 04 06:18:14 wow , amd plans to use oe a demo platform for their mips chips Oct 04 06:19:28 rob_w: iirc amd doesn't have any mips chips anymore Oct 04 06:19:37 rob_w: they sold it to Raza, right? Oct 04 06:20:56 yes , but they feature the RMI developer site together with raza Oct 04 06:21:05 koen, do you know any raza guy in here ? Oct 04 06:22:16 the bad side is they say the have a oe image , but it says "Coming in July 2006" Oct 04 06:23:18 they merged quite a bit of stuff to .dev a while back Oct 04 06:24:21 rob_w: try asking CosmicPenguin when he's awake Oct 04 06:24:52 ok cool Oct 04 06:27:07 You guys are Rock Stars! Oct 04 06:27:32 RP: 'ERROR: global name 'bb_cache' is not defined while parsing ' for every recipe in my BBFILES :( Oct 04 06:31:56 B_Lizzard, why is that ? Oct 04 06:32:34 Cause OE rules Oct 04 06:32:52 And you guys are the best Oct 04 06:33:26 Just thought some praise would do some good Oct 04 06:33:38 After all, you work hard Oct 04 06:43:32 layout of database /home/bob/oe/current/OE.mtn doesn't match this version of monotone Oct 04 06:43:32 , but i just downloaded a fresh mtn and a fresh OE.db ,, is this ok to migrate ? Oct 04 06:46:41 rob_w: have a look at http://www.openembedded.org/snapshots/ Oct 04 06:47:31 oh i see Oct 04 06:48:09 thanks Oct 04 06:48:20 :) Oct 04 06:50:09 i should finally get to the mailing list ;-) Oct 04 07:28:31 re Oct 04 07:41:55 morning Oct 04 07:54:45 good morning Oct 04 07:54:52 03rpurdie * r618 10bitbake/bin/bitbake: bin/bitbake: fix typo Oct 04 07:54:53 good morning ll Oct 04 07:54:58 s/ll/all Oct 04 07:55:32 hi likewise Oct 04 07:56:11 morning all Oct 04 07:57:00 RP: hi Oct 04 07:57:40 hi RP likewise Oct 04 07:57:45 hi florian_kc Oct 04 07:57:55 all these people sneaking in when Im busy :-) Oct 04 07:58:14 slow start this morning... I am still booting. Oct 04 08:04:27 hi zecke XorA koen|away RP florian_kc Oct 04 08:04:31 hey hrw|work Oct 04 08:04:57 koen: I was thinking about small GPG signing party Oct 04 08:05:08 hrw|work: that could be usefull Oct 04 08:05:16 RP: thanks for fixing bitbake Oct 04 08:05:47 koen: np, just a silly mistake Oct 04 08:07:36 RP: could you bump PV to 1.7.1? Oct 04 08:07:38 koen: why? Oct 04 08:08:00 koen: I think that sooner or later we will have sources mirror which would need uploading files by devels. GPG signing would be needed Oct 04 08:08:45 RP: makes it easier to see which svn version people are running from the buildsummary Oct 04 08:08:59 koen: ok Oct 04 08:11:54 hrw-pcmcia-ids-r5 MERGED Oct 04 08:12:05 http://42.pl/u/miC Oct 04 08:13:22 * koen adds another hack to OE Oct 04 08:14:07 Alan sayeth: "Same update needs to go into drivers/ata/pata_pcmcia" Oct 04 08:14:08 ugh Oct 04 08:14:13 03koen 07org.oe.dev * r5988c415... 10/ (1 packages/gdbm/gdbm_1.8.3.bb): gdbm: add PACKAGES_DYNAMIC to satisfy python Oct 04 08:14:18 03koen 07org.oe.dev * rbc5677e9... 10/ (1 packages/python/python_2.4.3.bb): python: python has been broken long enough with recent bitbakes, so lets do PACKAGES_DYNAMIC="python*". Feel free to properly fix it. Oct 04 08:16:21 koen: what about other python recipes? Oct 04 08:16:23 koen: That looks like total madness? Why do we need to do that? Oct 04 08:16:31 python-sqlite for example? Oct 04 08:16:54 like I said "Feel free to properly fix it" Oct 04 08:17:12 at least I can build python now Oct 04 08:17:15 koen: At least explain the problem... Oct 04 08:18:29 * koen tries to find the bug report Oct 04 08:18:53 RP: 'bitbake python' didn't work because of the python-manifest magic Oct 04 08:19:44 and who's idea was it to make bitbake bugreports hidden for normal users by default? Oct 04 08:20:08 "Only users in all of the selected groups can view this bug: BitBake Developers" Oct 04 08:21:18 koen: bitbake python would work. What might not work is RDEPENDS = "python" Oct 04 08:21:34 koen: and that is because anyone should be using "python-core" Oct 04 08:22:42 hmmm Oct 04 08:22:53 Whether RPROIDES_python-core = "python" should be added is a question best addressed to the python maintainer Oct 04 08:23:06 koen: but PACKAGES_DYANMIC is a really bad idea Oct 04 08:24:42 umbaumba_61.2.bb RDEPENDS = "python-pyraf" - it will not work now Oct 04 08:25:16 03koen 07org.oe.dev * r87e693cc... 10/ (1 packages/python/python_2.4.3.bb): disapproval of revision 'bc5677e9c976a0343cf5804e7e3ef53a9a8fc47b' Oct 04 08:25:26 thx koen Oct 04 08:26:22 koen: The PACKAGES_DYNAMIC in gdbm is also wrong :-/. You'd be safer changing it to an RPROVIDES Oct 04 08:26:44 RP: the (R)DEPENDS in python* is wrong on gdbm Oct 04 08:27:01 Anything using PACKAGES_DYNAMIC without a wildcard in is nearly always wrong Oct 04 08:27:36 koen: If it RDEPENDS on python, yes Oct 04 08:28:07 RP: no, something in python (R)DEPENDS libgdbm3 Oct 04 08:28:25 RP: mickeyl is a bit weak in the shlib renaming department :) Oct 04 08:28:42 python-2.4.3-manifest.inc:RDEPENDS_python-gdbm="python-core libgdbm3" Oct 04 08:28:51 manifest need tweaking Oct 04 08:28:58 right Oct 04 08:29:11 koen: Ah, that. That is a bug in python - s/libgdbm3/gdbm/ Oct 04 08:29:12 mickey|tv said he'd look at that weeks ago Oct 04 08:29:45 koen: Adding other junk to the metadata is not acceptable - we need to fix python Oct 04 08:30:06 koen: Can you revert the other PACKAGES_DYNAMIC change to gdbm please? Oct 04 08:30:45 right, my point was that python needs to get fixed, we've known that for weeks and did nothing Oct 04 08:31:52 breaking other packages because one is broken is never going to be acceptable Oct 04 08:35:16 crap ,, the raza guys do loudly talk about oe but then when you try to compile their 2.6 kernels they require a Mvista toolchain linker Oct 04 08:47:49 rob_w: raza? Oct 04 08:48:00 hrw|work: ping Oct 04 08:48:35 likewise: http://wwwd.amd.com/RMI/EmbeddedDeveloper.nsf/ Oct 04 08:48:48 likewise, the company which bought the alchemy cpu`s from amd Oct 04 08:51:46 WTF? Oct 04 08:51:55 their dev site requires signing an NDA Oct 04 08:54:38 koen, well ... Oct 04 08:54:44 morning all Oct 04 08:55:33 hey lrg Oct 04 08:57:09 hey lrg Oct 04 09:06:44 zecke: pöng Oct 04 09:11:34 interesting: http://www.geekzone.co.nz/content.asp?ContentId=6731 Oct 04 09:16:57 ehh , errror D-Link ?=!? Oct 04 09:19:25 RP: isn't qemuarm using the versatile board now: http://www.o-hand.com/~richard/qemu.html ? Oct 04 09:19:54 rob_w: yes the same dlink as http://gpl-violations.org/news/20060922-dlink-judgement_frankfurt.html Oct 04 09:20:22 yeah .. i surved about this again lately .. Oct 04 09:24:59 koen: yes, those instructions need updating Oct 04 09:28:24 I will push changed ones Oct 04 09:31:04 pushed Oct 04 09:31:40 03hrw 07org.oe.dev * re81ad8bb... 10/ (4 files in 3 dirs): python: fixed manifest to NOT rdepend on libraries - shlibs code will do it in correct way Oct 04 09:31:56 ~lart ssh chains Oct 04 09:31:56 * ibot beats ssh chains severely about the head and shoulders with a rubber chicken Oct 04 09:32:17 ssh -t one -t ssh two -t screen -RDU Oct 04 09:36:28 mickey|tv: any plans to update python-native and/or python to 2.5? Oct 04 09:48:53 koen: got info from celf: the machine which is used for building is not firewalled. thats why it works Oct 04 09:50:54 great Oct 04 09:51:03 so either you have a fascist firewall or no firewall Oct 04 09:51:09 all hail common sense! Oct 04 09:52:32 hrw|work, I m going to try the older omap1 kernel Oct 04 09:55:10 Crofton: ok Oct 04 09:55:41 koen: they have fascist firewalls in other locations - I got detailed email today Oct 04 09:55:48 I hvae lost confidence in my ability to make stuff work :) Oct 04 09:56:03 Crofton: I posted compilefix patch to linux-omap ML Oct 04 09:56:18 koen: also some suggestions to GettingStarted Oct 04 09:56:51 it's a wiki Oct 04 09:56:58 NAiL: did you resolved situation when libx11 cant be build because of lack of X headers? Oct 04 09:57:20 hrw|work, compile for omap kernel? Oct 04 09:57:26 koen: there is a knowledge needed to fix it properly as it is not only typos Oct 04 09:57:47 Crofton: that one for situation when CONFIG_OMAP_DSP is not defined Oct 04 09:57:58 ah Oct 04 09:58:07 I didn't see the patch Oct 04 09:58:13 maybe it is too early for me Oct 04 09:58:30 apparently there is a bunch of stuff broken in gt for th OSK right now ... Oct 04 10:00:19 koen: Now python is fixed (thanks hrw), can we fix gdbm? Oct 04 10:00:30 sure Oct 04 10:00:38 | makekeys.c:34:19: X11/X.h: No such file or directory Oct 04 10:00:40 ;((( Oct 04 10:00:43 like I said: "Feel free to fix it properly" Oct 04 10:01:21 does2.6.12 work with udev? Oct 04 10:02:00 Crofton: depends on the udev version Oct 04 10:02:13 ug Oct 04 10:02:16 092? Oct 04 10:08:17 koen: http://bugs.openembedded.org/show_bug.cgi?id=1286 - can you look? Oct 04 10:09:01 that one is ok Oct 04 10:09:05 the udev rule works on my hx4700 Oct 04 10:09:23 RP: did you since updates to the ts rule from poky yet? Oct 04 10:09:24 then if it is ok now then maybe its time to close it? Oct 04 10:09:36 hrw|work: sure, go ahead Oct 04 10:09:44 hrw|work, now the 2.6.12 kernel fails for me, vmlinux in the worng place? Oct 04 10:10:08 koen: for me you are h2200 maintainer rather then me Oct 04 10:10:15 Crofton: same problem as with 2.6.x-git Oct 04 10:10:28 * Crofton goes insane Oct 04 10:10:29 hrw|work: then it will stay open Oct 04 10:10:55 koen: I will ask in other way. Oct 04 10:10:58 this worked once upon a time Oct 04 10:11:10 koen: does #1286 bug is resolved with current .dev? Oct 04 10:11:21 by any chance does the kernel bbclass move vmlinnux around? Oct 04 10:12:51 Crofton: rather not Oct 04 10:13:15 wow just got a response for a bugfix i submitted 2 years ago Oct 04 10:14:14 mithro: nice Oct 04 10:14:28 hrw|work: tslib will still depend on X in .dev for some machines Oct 04 10:14:54 koen: so someone need to get detect-tsdevice from familiar? Oct 04 10:15:53 [ALSA - utils 0001400]: sys_siglist unavalible in uclibc Oct 04 10:16:14 hrw|work: no Oct 04 10:16:24 hrw|work, I am going to try getting 2.6.16 built from patch at atomide Oct 04 10:17:14 your patch made it to the list :) Oct 04 10:17:55 Crofton: I'm interested in reactions on list ;) Oct 04 10:18:09 hopefully someone fixes it Oct 04 10:18:13 koen: We changed the rule in .dev and I recently changed the rule in poky to follow Oct 04 10:18:48 I need to pull a mickey and disappear for a month to write my thesis ... Oct 04 10:19:12 03xora 07org.oe.dev * rc0c96c93... 10/ (5 files in 4 dirs): Oct 04 10:19:12 conf/machine/smdk2440.conf, packages/linux/linux-smdk2440_2.6.18.bb : Initial Oct 04 10:19:12 support for the Meritech smdk2440/2442 dev boards. Oct 04 10:21:25 hrw|work: #1286 is not fixed in dev but could easily be fixed Oct 04 10:21:45 ok Oct 04 10:22:03 hrw|work: Basically, apply the patch Oct 04 10:23:09 RP: I prefer to not touch such things without testing on device Oct 04 10:31:30 03rpurdie 07org.oe.dev * rce640ed0... 10/ (1 packages/gdbm/gdbm_1.8.3.bb): Oct 04 10:31:30 gdbm: Fix breakage introduced by koen. Breaking X because Y is broken is not Oct 04 10:31:30 acceptable behaviour, neither is just telling other people to fix the problem. Oct 04 10:31:30 This hack is also broken in numerous other ways and is a total abuse of Oct 04 10:31:30 PACKAGES_DYANMIC which is meant for a very specific use only (splitting packages Oct 04 10:31:31 when the packages names are unknown prior to splitting). No, I'm not happy. Oct 04 10:33:11 hrw|work: nope, haven't had the time. It's low prio for me really Oct 04 10:33:20 RP: there are ~10 more commits adding PACKAGES_DYNAMIC in the way I did still in the metadata Oct 04 10:33:39 koen: Do you plan to fix them? Oct 04 10:33:54 NAiL: ok Oct 04 10:34:11 RP: actually no, since they have been in there for ages without anybody telling me they are wrong Oct 04 10:34:33 koen: I hadn't noticed them. I'm not telling you they are wrong Oct 04 10:34:38 I think it's amusing that stuff I've been doing for months is suddenly wrong Oct 04 10:34:43 s/not/now/ Oct 04 10:35:42 as I evidently don't know how PACKAGES_DYNAMIC work, I can't fix it Oct 04 10:35:53 how do I get udev to make ttySAC0? Oct 04 10:36:10 XorA: does it exist in /sys/ at all? Oct 04 10:36:43 koen: Can you at least explain the need for adding all these PACKAGES_DYNAMICs ? Oct 04 10:37:18 task-sdk* Oct 04 10:37:23 hrw|work: I can see stuff for UART 0 on s3c2440 Oct 04 10:38:04 XorA: use udevinfo to get infos Oct 04 10:38:10 I added those after discussing it on IRC, which makes it even more amusing that they are wrong Oct 04 10:38:28 koen: task-sdk* is broken Oct 04 10:45:11 hi Cliff Oct 04 10:45:21 hrw|work: ah hey :) Oct 04 10:45:38 hrw|work: you have my mobile phone number? Oct 04 10:45:44 * Varoudis .. hello world! Oct 04 10:45:51 zecke: you sent it in OEDEM mail Oct 04 10:46:01 koen: Discussing things like that on irc probably isn't the wisest course of action Oct 04 10:46:19 hrw|work: you can either get to the BMCO directly (by public transportation) or call me and we can see where I will pick you up Oct 04 10:46:42 florian_kc: you know where you need to get to on thursday? Oct 04 10:46:52 zecke: +49162xxx317 one stored in phone (and photo added into contact) Oct 04 10:47:25 zecke: did you read your mail yesterday? Oct 04 10:47:40 zecke: I will walk around and be at bmco after 17 like we discussed few weeks ago Oct 04 10:48:20 zecke: yes, sounds like it is easy to find... i'll try to be there around 1800. Any hint where to leave my car? Oct 04 10:49:35 koen: no, my macbook is still broken -> I do not read mails Oct 04 10:49:38 XorA: udevinfo -a -p /sys/class/tty/ttySAC0 will tell you Oct 04 10:49:55 florian_kc: U-Bahnhof Alt-Mariendorf has Park and Rail Oct 04 10:50:18 florian_kc: and it is near the U6, otherwise leave it in front of the BMCO Oct 04 10:51:02 zecke: ok Oct 04 10:51:04 zecke: I land at 21:10 tomorrow at Schoenefeld. Is transport at that time of day going to be problematic? Oct 04 10:51:23 hrw|work: not there :-( Oct 04 10:51:53 XorA: do you have something at all in /sys/class/tty/ (other then tty[0-9])? Oct 04 10:52:08 http://pastebin.ca/190836 Oct 04 10:52:15 stupid vmlinux Oct 04 10:52:25 hrw|work: ttyS* Oct 04 10:52:32 can't find vmlinux in do_install Oct 04 10:52:48 XorA: and each of them is not your UART0? Oct 04 10:52:52 RP: it should be fine. 'S' and metro ('U') should operate until one or even two o' clock Oct 04 10:53:22 hrw|work: ah found it, s3c2410_serial0 Oct 04 10:53:25 ~seen woglinde Oct 04 10:53:38 hm. I need to read somewhere what is worth seeing to have some place to visit tomorrow Oct 04 10:53:43 woglinde was last seen on IRC in channel #oe, 1d 17h 12m 24s ago, saying: 'here only 15'. Oct 04 10:53:59 hmm I need to decide on a place to go tomorrow Oct 04 10:54:12 zecke: ok, thanks. Any idea where people will be by then? Still near bmco or somewhere else? Oct 04 10:54:13 XorA: then udevinfo query it Oct 04 10:54:29 RP: likely somewhere else Oct 04 10:54:50 hrw|work: got that info Oct 04 10:55:06 zecke: I'm just wondering where to head for as tf will probably have my room key and will probably be with you by then :) Oct 04 10:55:45 RP: ah okay. Oct 04 10:56:08 be back in a minute Oct 04 10:56:11 hrw|work: I assume a KERNEL="s3c2410_serial[0-9]" NAME="ttySAC%n" should make the device? Oct 04 10:56:21 XorA: probably Oct 04 10:56:27 hrw|work: cheers for help Oct 04 10:56:31 XorA: add it into rules and then /etc/init.d/udev reload Oct 04 10:57:15 koen: Fixing task-sdk is relatively straightforward - s/libasound2/alsa-lib/ etc. Oct 04 10:59:08 hrw|work: did you ever get the chance to look over the final changes Mike did for the wm97xx touch. He made some more changes to tosa lcd and pm that needed a little work by either youself or Dirk to clean up. Oct 04 10:59:55 re Oct 04 11:00:20 http://www.kneipen-suche.com/berlin-malheur-4778.html Oct 04 11:00:29 * lrg or Xora will be at OEDEM next year :) Oct 04 11:00:58 let us meet there tomorrow Oct 04 11:01:10 lrg or Xora can still make it this year? Oct 04 11:01:21 * zecke really wonders where pb_ is :} Oct 04 11:01:27 zecke: http://rafb.net/paste/results/PrkPqq94.html Oct 04 11:01:29 zecke: a bit busy with support atm :( Oct 04 11:01:44 hrw|work: magic, that worked, thanks for the quick 101 Oct 04 11:01:55 zecke: xora has 6 day to do ac97 driver Oct 04 11:02:04 zecke: i wonder that too... sent him a mail last week but no answer so far. Oct 04 11:02:37 zecke: I got email from pb_ 2 weeks ago. He may be abroad doing some debuging. Oct 04 11:02:58 lrg: oh, that's good news Oct 04 11:03:12 lrg: ah good to know Oct 04 11:03:46 koen: I need to finish some stuff now. punch to mail this evening Oct 04 11:03:55 zecke: We should add this info to the wiki and let people know if that's the final plan :) Oct 04 11:04:06 should I add Samsung uarts to main udev.rules in OE or should I make seperate file with overrides? Oct 04 11:04:11 zecke: should we meet there instead of meeting at the oedem place? Oct 04 11:04:11 Hi, if compiling for different setups, what do I need to separate out into different tmp directories? a) MACHINE, b) DISTRO c) something else? Oct 04 11:04:27 RP: It is quite easy to get to the malheur. From the airport take the bus to "U-Bahnhof Rudow" it departs right in front of the terminal Oct 04 11:05:01 XorA: I would add to main, this board and it's sucessors may prove to be popular Oct 04 11:05:09 RP: leave the U7 at Gneisenau Strasse (and you have luck Rudow is the final station so you can't take the train to a wrong direction) Oct 04 11:05:14 lrg: especially for tytnlinux :-) Oct 04 11:05:24 koen: I have no idea where to get breakfast Oct 04 11:05:45 zecke: that's why it was 2 questions in one :) Oct 04 11:06:01 ~lart Crofton Oct 04 11:06:02 * ibot decapitates Crofton conan the destroyer style Oct 04 11:06:06 ug Oct 04 11:06:32 that is what I get for forgeting to inherit from kernel, then ovveride settings ... Oct 04 11:06:45 zecke: Are tickets for the U-bahn automated machines or real people? Oct 04 11:08:06 RP: Just buy a ticket when you enter the bus. They should speak english Oct 04 11:08:22 zecke: ah, ok :) Oct 04 11:08:23 RP: just get a regular ticket. We have three zones (A,B,C) and if he asks you get one for AB Oct 04 11:08:59 zecke: right, that makes sense. Thanks for the info, its very useful :) Oct 04 11:09:38 * XorA goes for some food Oct 04 11:09:46 Although I've heard the comment "should speak english" before - they invariably won't :) Oct 04 11:09:53 03xora 07org.oe.dev * r520c4478... 10/ (4 files in 4 dirs): Oct 04 11:09:53 udev-092/udev.rules udev-097/udev.rules udev-100/udev.rules : Add rule Oct 04 11:09:53 to create UART devices for s3c24XX CPUs. Oct 04 11:10:35 hrw|work, I built 2.6.16 for the OSK, if it boots, I'll push it Oct 04 11:11:34 hmm, why did is libusb building .... Oct 04 11:17:13 RP: yeah, it might be a different english Oct 04 11:21:25 zecke: None whatsoever is usually what I find once someone's told me they speak English but I'll manage. I could once speak a small amount of German so I'll be better off than I was in Finland :) Oct 04 11:24:46 hehe Oct 04 11:27:42 hey people Oct 04 11:28:06 should speak english? Oct 04 11:29:17 mithro: did you have difficulties buying a ticket for public transportation? Oct 04 11:29:31 in germany? Oct 04 11:29:35 mithro: yes Oct 04 11:29:39 nope Oct 04 11:29:47 RP: and we have vending machines as well :) Oct 04 11:29:52 if i remeber correctly the ticketing computers all had english option Oct 04 11:31:27 morning all Oct 04 11:31:33 hey Dirk Oct 04 11:31:42 hi Dirk Oct 04 11:31:55 hi Liam, Richard Oct 04 11:32:12 zecke: it will be fine :) Oct 04 11:32:32 RP: and www.bvg.de is useful! Oct 04 11:33:12 and even in English :) Oct 04 11:34:35 zecke, i prefer www.vvs.de rather ;) Oct 04 11:35:29 TheCan: yeah, but somehow Stuttgart und Berlin do not cooperate on the level of public transportation :) Oct 04 11:35:35 :D Oct 04 11:45:37 do13: did you get a cahnce to look at the latest touch patches ? Oct 04 11:48:17 lrg: yep. I tried to cleanup the tosa part. But I broke something :-( Oct 04 11:49:09 do13: ok, can you send me the changes/output and I'll take a look. cheers Oct 04 11:51:00 lrg: yep, after work Oct 04 11:51:21 do13: thanks Oct 04 12:15:27 03tmbinc 07org.oe.dreambox * r7e2d6226... 10/ (4 files in 4 dirs): busybox-1.2.1: port nptl patch, fix config Oct 04 12:15:29 03tmbinc 07org.oe.dreambox * r1ae71587... 10/ (6 files in 3 dirs): tuxbox-libtuxtxt: patch to prevent crash on invalid teletext data Oct 04 12:15:37 03tmbinc 07org.oe.dreambox * rdd1252be... 10/ (1 packages/images/dreambox-image.bb): dreambox-image: use busybox instead of tinylogin and less Oct 04 12:16:21 tmbinc_: don't forget to put fixes in .dev as well Oct 04 12:17:25 koen: I think this is one of the things I'm trying to do (dreambox->dev) Oct 04 12:28:04 morning Oct 04 12:32:31 hi Oct 04 12:34:36 03tmbinc 07org.oe.dreambox * r601454d2... 10/ (4 files in 4 dirs): tuxbox-libtuxtxt: move patch to correct location Oct 04 12:36:05 * koen double checks which chargers to pack Oct 04 12:37:39 silly Europeans, can't agree on languages or electrical outlets :) Oct 04 12:38:54 * Crofton is also very glad the electronics he travels with no longer care about voltage Oct 04 12:40:47 Crofton: outlets aren't the problem right now: Oct 04 12:40:52 glibc-2.4 still fails to compile here even after removing tmp-> http://ewi546.ewi.utwente.nl/tinderbox/showlog.pl?machine_id=123&logfile=20061004022812.log. It looks like something is wrong with the compiler that OE uses here. Oct 04 12:41:32 Crofton: the netbook, nokia770, ipaqs and laptop all have a different plug and voltage Oct 04 12:41:44 at least the ipaq can charge over usb Oct 04 12:41:46 heh Oct 04 12:42:16 I didn't I traveled with one charger to rule them, in the land of mordor ,,, Oct 04 12:42:35 only that they all accept about 100-250 VAC 50/60Hz these days Oct 04 12:42:37 wierd initialisation of g-ether is modprobe ohci_hcd; rmmod ohcu_hcd; modprobe g-ether Oct 04 12:44:43 XorA: looks like the gadget driver lacks some initialization calls Oct 04 12:46:34 koen: yeah Oct 04 12:49:35 Crofton: you got libusb built because omap5912osk has usbhost in machine features and openomap distro has it in distro features. I recently changed task-base to add usbutils for machines which has usbhost function. Oct 04 12:49:36 re Oct 04 12:49:58 zecke: to get to you AB section ticket will be ok too? Oct 04 12:50:25 re after meeting btw Oct 04 12:50:42 ok Oct 04 13:05:47 hrw|work: sure. If you want to travel a lot tomorrow. Get an All-Day Ticket Oct 04 13:06:22 03koen 07org.oe.dev * r223327fd... 10/ (1 conf/machine/htcuniversal.conf): htcuniversal: add MACHINE_FEATURES Oct 04 13:07:00 zecke: currently I'm planing with google maps and it looks like I will rather go by foot Oct 04 13:07:24 zecke: any places worth visiting? Oct 04 13:07:26 hrw|work: it is 6 euro Oct 04 13:07:47 03koen 07org.oe.dev * rce63e148... 10/ (1 packages/angstrom/task-angstrom.bb): task-angstrom: decrease overlap with task-base some more Oct 04 13:08:14 hrw|work: I live here, how should I know? I think the obvious tourist places are Oct 04 13:08:14 hrw|work: Potsdamer-Platz, Brandenburger Tor, Reichstag Oct 04 13:08:24 hrw|work: take the long walk from Branderburger Tor to Alexander Platz Oct 04 13:08:56 zecke: hauptbahnhof -> Brandenburger Tor -> Potsdamer-Platz is current plan Oct 04 13:09:19 doesn't sound too bad Oct 04 13:09:40 hrw|work: walking from branderburger tor to alexanderplatz is quite interesting for two reasons Oct 04 13:09:48 google maps is nightmare to print in hybrid Oct 04 13:09:57 zecke: I'm listening Oct 04 13:09:58 hrw|work: a) you see buildings that survived ww2 Oct 04 13:10:45 hrw|work: b) at alexanderplatz you see the buildings build in the GDR (socialist architecture) but you might know this kind of architecture already ;) Oct 04 13:10:45 hrw|work: thats why GPRS was invented :-) Oct 04 13:11:06 hrw|work: but your walk can be interesting Oct 04 13:11:08 XorA: gprs over roaming + zaurus without BT? Oct 04 13:11:41 03koen 07org.oe.dev * rd70f9864... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom 2007.1: prefer hh-pxa cvs for htcuniversal Oct 04 13:11:43 hrw|work: e.g. you pass the Reichstag when coming from Hauptbahnhof, and you pass the "holocaust mahnmal" and you end up at potsdamer platz Oct 04 13:12:00 hrw|work: I have a J2ME program that reads google maps Oct 04 13:12:02 mahnmal? Oct 04 13:12:09 XorA: GIVE me Oct 04 13:12:22 hrw|work: google.com/gmm Oct 04 13:12:46 nokia 770 Oct 04 13:12:55 Crofton: maemo-mapper Oct 04 13:13:12 * koen just install gmm on his motorola and maemo-mapper-aw on his 770 Oct 04 13:13:17 installed* Oct 04 13:14:54 hrw|work: http://www.mgmaps.com/ Oct 04 13:16:44 hrw|work: hrw|work memorial Oct 04 13:17:26 zecke: should we meet there instead of meeting at the oedem place? Oct 04 13:20:24 koen, any idea how the 770 compares speedwise to the present generation zaurus Oct 04 13:20:46 Hi! Oct 04 13:20:50 joshin: no idea Oct 04 13:21:18 psokolovsky: did you get my botmail? Oct 04 13:21:45 koen: Yes. Why is _cvs used? _git should be used instead. Oct 04 13:21:47 Bummer. My 5500 is showing its age and I'm not sure what I want to replace it with. Oct 04 13:22:19 joshin: it only has a 220MHz cpu, but it also has a dsp for multimedia Oct 04 13:23:16 psokolovsky: http://rafb.net/paste/results/hIgJ2j39.html Oct 04 13:23:17 I know. But AMD taught me that mhz !== best performance is not always true. :) Oct 04 13:23:25 koen: that's what I have as built for angstrom snaphost on sep 29: xcalibrate-0.0+git20060814-r0, xcalibrateext-0.0+git20060814-r0 Oct 04 13:23:31 koen: looking Oct 04 13:24:22 POS! Oct 04 13:29:14 03pfalcon 07org.oe.dev * r60e987f7... 10/ (1 packages/xcalibrate/xcalibrate_git.bb): Oct 04 13:29:14 xcalibrate_git: Add latest xcalibrate from X.org's git. Oct 04 13:29:14 * Was forgotten to be added, the same as xcalibrateext_git before ;-(. Oct 04 13:29:26 koen: Soooorryyyy! Running global mtn ls unknown now... ;'-( Oct 04 13:39:23 zecke: http://miejsce.info/zobacz/?667702ebcfb9fda83a09d8140016797d Oct 04 13:41:41 hmm.. will change Oct 04 13:43:18 wikipedia is good thing Oct 04 13:44:47 03pfalcon 07org.oe.dev * r1124357c... 10/ (1 packages/glibc/glibc-package.bbclass): Oct 04 13:44:47 glibc-package.bbclass: Add supersecret -r switch to qemu userspace emulator, Oct 04 13:44:47 which allows to override kernel version report to guest apps (otherwise, Oct 04 13:44:47 host kernel version is used, which causes "Kernel too old" errors in glibc Oct 04 13:44:47 startup code in guess, if host kernel is older than 2.6.14-something). Oct 04 13:44:48 Tested with qemu 0.8.2 (and yes, I couldn't find -r switch in qemu doc, Oct 04 13:44:52 taken from a mail by Paul Brook, qemu co-maintainer). Oct 04 13:45:23 koen, hrw|work: pls have a look at this ^. it "just works", but fyi Oct 04 13:46:08 psokolovsky: after oedem? Oct 04 13:49:10 hrw|work, mean, just have a look at that patch, and note a switch added (though I don't expect any problems). And have a good jorney!! Oct 04 13:52:50 03pfalcon 07org.oe.dev * r5e5057db... 10/ (1 packages/qemu/qemu-native.inc): qemu-native.inc: Do not overwrite EXTRA_OECONF, append to it. Oct 04 13:58:01 zecke: updated Oct 04 14:08:27 psokolovsky: great, thanks for fixing it :) Oct 04 14:11:21 koen: when you be in berlin? Oct 04 14:11:35 tomorrow at 13:00 Oct 04 14:11:43 koen: where? Oct 04 14:11:46 I'll meet you all after dinner at malheur Oct 04 14:11:59 ko Oct 04 14:14:05 hrm Oct 04 14:14:17 * koen gets lost in the u-a changes in poky Oct 04 14:22:19 hrw|work: actually my position is wrong :) Oct 04 14:22:40 isn't that where bmco is? Oct 04 14:22:47 koen: almost :) Oct 04 14:23:17 koen: if you see a huge vodafone sign and "property of vodafone" go lef Oct 04 14:23:25 koen: BMCO is in the building on the left Oct 04 14:23:37 koen: when do you arrive again? Oct 04 14:24:10 zecke: but generally it is correct marked? Oct 04 14:24:17 tomorrow 13:00, but I'm going to sight-see with my girlfriend first :) Oct 04 14:24:22 hrw|work: generally yes Oct 04 14:24:44 koen: iirc you had plans to be in Berlin today... ;D Oct 04 14:24:57 hrw|work: right :) Oct 04 14:25:24 koen: thanks for the raza ptr (earlier today) Oct 04 14:25:53 likewise: np, I still need to poke CosmicPenguin about their diffs (if the have any) Oct 04 14:25:55 koen: and have a nice OEDM Oct 04 14:26:00 For Alchemy? Oct 04 14:26:04 we have nothing Oct 04 14:27:03 * koen ticks off another TODO Oct 04 14:27:37 All the existing drivers are suffering from bitrot, and I'm not sure what Raza has done to update them Oct 04 14:31:27 koen: u-a changes in poky? Oct 04 14:32:30 update-alternative Oct 04 14:33:01 I haven't figured out what happens if people do 'ipkg upgrade' after those changes Oct 04 14:38:43 * koen wonders why bluez-utils 3.4 thinks dbus 0.23 is present Oct 04 14:38:53 I have 0.62 on my host and 0.92 in staging Oct 04 14:40:26 ~lart danbold twice Oct 04 14:40:26 * ibot sends a legion of lawyers after danbold twice's head Oct 04 14:40:55 he seems to have some trouble reading what we are saying Oct 04 14:41:01 "no sram" Oct 04 14:42:45 in Poland word 'sram' can be wrong understood in non-tech environment Oct 04 14:43:00 'bom' as well? Oct 04 14:43:19 koen: There is a circular dependency problem in there too :-( Oct 04 14:43:39 RP: right, but it searches for dbus-glib on the host Oct 04 14:44:04 after I updated libdbus-glib-1-dev in debian it passed configure Oct 04 14:48:29 koen: bom is horizontal mast on sailing ships Oct 04 14:52:20 hi everyone Oct 04 14:52:55 hrw|work: according to bruce willis in 'the last boyscout' it means something else :) Oct 04 14:52:59 hey tomas_f Oct 04 14:53:40 maybe Oct 04 14:53:44 hi tomas_f Oct 04 14:53:57 what time are folk planing to be in the Malheur tomorrow ? Oct 04 14:56:53 hm.. Debian has glibc 2.4 in experimental... Oct 04 14:57:03 tomas_f: I will be with zecke Oct 04 14:58:03 hi tomas_f :) Oct 04 14:58:41 Malheur is some bar/music place, right? Oct 04 15:07:26 cu tomorrow guys Oct 04 15:12:29 hi guys Oct 04 15:13:21 * mickeyl preparing OEDEM "keynote" Oct 04 15:13:32 don't worry, i'll keep it short Oct 04 15:14:02 mickeyl: no powerpoint allowed! Oct 04 15:14:11 hi mickeyl Oct 04 15:14:21 koen: hey, don't be dogmatic Oct 04 15:14:30 me? never :) Oct 04 15:14:33 :D Oct 04 15:14:42 fact is... Oct 04 15:14:44 * koen hides his MS office install Oct 04 15:14:58 with windows xp my battery lasts 9h Oct 04 15:15:03 with linux it lasts 5h Oct 04 15:15:08 which is kind of annoying Oct 04 15:15:28 yo RP Oct 04 15:16:04 'bitbake virtual\kernel' doesn't appear to put a new kernel in my deploy dir. how do i force it to build a new kernel? Oct 04 15:16:17 -c rebuild ? Oct 04 15:16:18 bitbake --help Oct 04 15:16:19 I haven't booted windows on the laptop since I got it. The battery only ever seems to last me a couple of hours but I tend to do things like compiling :-/ Oct 04 15:16:40 mickeyl: linux 'desktop' stuff polls a lot Oct 04 15:17:00 RP: *nod* i guess windows is just better because it really agressively undervolts and underclocks the CPU when idle as well as shutting devices down when possible Oct 04 15:17:40 koen: yeah. only very few things use push-based interfaces, inotify and stuff :/ a lot to improve Oct 04 15:18:42 mickeyl: your pc won't work with the projector anyway :) Oct 04 15:19:05 zecke: eh...why? Oct 04 15:19:12 mickeyl: Do we need to do anything for the OE steering dicussion, except perhaps refine the agenda? Oct 04 15:19:37 RP: no i think that should be it. Oct 04 15:19:43 zecke: Will you be able to report on the status of bitbake QA/unit testing and the status of the c parser? Oct 04 15:19:53 mickeyl: I have a util on my linux desktop that undervolts my cpu when idle Oct 04 15:20:09 'report' yes Oct 04 15:20:16 mickeyl: because you use linux? Oct 04 15:20:27 koen: '-c rebuild' appears to be working. thanks. Oct 04 15:20:40 heh Oct 04 15:20:41 zecke: ah. no no, this is a thinkpad. thinkpads work with projectors both under Win and under Linux :D Oct 04 15:20:42 http://www.gamedev.net/reference/articles/article1176.asp Oct 04 15:20:48 zecke: thanks. That mainly covers the areas I don't know anything about :) Oct 04 15:21:10 QA/Bitbake-C == void :) Oct 04 15:21:48 Do we have a dress-code? Oct 04 15:22:33 zecke: suit and tie Oct 04 15:22:37 * mickeyl chuckles Oct 04 15:22:52 * koen packs his suit Oct 04 15:22:55 just kidding Oct 04 15:22:58 puhlease not :) Oct 04 15:23:06 anyone minds Mexican or Vietnamese food at the evening? Oct 04 15:23:13 * mickeyl wanted to wear his military outfit Oct 04 15:23:22 zecke: sounds good to me Oct 04 15:23:36 zecke: http://www.monsieurvuong.de/mvweb/en/index_main.html ? Oct 04 15:23:39 mickeyl: during the day/launch our options are "Frittenbude" or ordering food Oct 04 15:23:41 koen: yes Oct 04 15:23:46 the picture is scary, but the food looks good Oct 04 15:23:59 koen: he will shake your hand personally Oct 04 15:24:09 koen: I think he is around 60 or such nowadays Oct 04 15:24:17 s/shake/crush/ ? Oct 04 15:24:26 zecke: good. more important to me is... how do we get coffee and tee and water? Oct 04 15:24:31 (for during the sessions) Oct 04 15:24:31 lol Oct 04 15:26:02 Added 14537 packages in 0 days, 0 hours, 3 minutes and 25 seconds Oct 04 15:26:11 First you tee then you pee Oct 04 15:26:13 thats the way it works Oct 04 15:26:29 mickeyl: I will buy some stuff to drink tomorrow Oct 04 15:26:37 mickeyl: and next to the BMCO is a "Plus" market Oct 04 15:26:38 zecke: excellent :) Oct 04 15:26:45 zecke: do we have a cofee machine there? Oct 04 15:26:52 coffee, even Oct 04 15:26:58 * mickeyl should drink less coffee Oct 04 15:27:31 I'm not entirely sure what Vietnamese food consists of but it looks like I'll be finding out... Oct 04 15:28:01 RP: mm - tasty Oct 04 15:28:10 don't fear the fish sauce Oct 04 15:28:22 mickeyl: yes, we have a cofee machine. I do not know how to operate it though Oct 04 15:29:09 zecke: I'll find that out :D Oct 04 15:29:14 RP: heh Oct 04 15:31:00 zecke: what time are you going guys going to be at the Mauhler ? Oct 04 15:31:25 20.00, 21.00 ? Oct 04 15:31:33 ok Oct 04 15:31:41 tomas_f: that's a guess Oct 04 15:31:59 florian arrives at BMCO at about 18h (he said) so we might be there at 19h Oct 04 15:32:21 * tomas_f wonders what BMCO is Oct 04 15:33:02 tomas_f: Broadcast Mobile Convergence. It is a forum where vendors talk about DVB-H Oct 04 15:33:13 tomas_f: we have some office space rented there Oct 04 15:34:30 does Malheur sever food ? Oct 04 15:34:51 tomas_f: they have good pizza, and not so good nachos Oct 04 15:35:35 pizza might do Oct 04 15:36:00 tomas_f: otherwise there are a lot of food places around Oct 04 15:38:52 ~seen Cwiis Oct 04 15:38:53 ~seen Cwiiis Oct 04 15:39:28 i haven't seen 'cwiis', zecke Oct 04 15:39:35 cwiiis was last seen on IRC in channel #oe, 13d 6h 7m 22s ago, saying: '(and using vorbisparse / oggmux before passing it to tremor isn't working either)'. Oct 04 15:47:25 OT: anyone knows how to to read from stdin unbuffered? Oct 04 15:48:57 set terminal to raw Oct 04 15:49:05 set terminal to blocking Oct 04 15:49:10 read one character Oct 04 15:49:16 (IIRC) Oct 04 15:50:30 mickeyl: how? Oct 04 15:50:49 mickeyl: I have 'stdin' which is a FILE*, I can get the fd with fileno but then? Oct 04 15:51:01 see http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/134892 Oct 04 15:51:09 should be translatable to C :) Oct 04 15:52:14 mickeyl: no, I might just use freeze to make it c? Oct 04 15:52:35 mickeyl: but actually no... Oct 04 15:53:30 heh Oct 04 15:54:20 :) Oct 04 15:54:27 mickeyl: I have an encoding issue :) Oct 04 15:58:30 BTW: if any of you could bring a Switch/Hub I would be glad Oct 04 15:58:39 we have enough cables but no unmounted switch Oct 04 15:58:48 zecke: open(), and use O_NONBLOCK? Oct 04 15:58:54 zecke, why not just handle it as a raw data stream? Oct 04 15:59:00 NAbyss: stdin is already opened :) Oct 04 15:59:01 not a file Oct 04 15:59:20 zecke: D'oh, yeah... too early in the morning :) Oct 04 15:59:26 NAbyss: :) Oct 04 15:59:45 I want to write a portable keylogger as I'm struggling with an encoding issue Oct 04 15:59:46 from what i recall, all files are buffered Oct 04 16:00:04 umm ... Oct 04 16:00:13 keystrokes are buffered ... Oct 04 16:00:24 zecke: er... fcntl(), F_SETFL, O_NONBLOCK? Oct 04 16:00:39 NAbyss: I will give that a try Oct 04 16:00:51 its not a programming thing, its hardware design Oct 04 16:00:59 I was just about to say, reopen stdin in NONBLOCK mode Oct 04 16:01:16 zecke: Not sure about portability.. might be worth looking at the source of the ncurses libs if you're ready to bang your head against the wall, see how they handle it? Oct 04 16:01:30 the keylogger stuff i had to build, all you do is keep flushing the keyboard buffer Oct 04 16:02:51 on the PC side ... on the PIC side it was a multiplexing issue Oct 04 16:03:53 thanks :) Oct 04 16:05:42 i am curious, what do you need a keylogger for? Oct 04 16:05:45 can't resolve *.openembedded.org now ? Oct 04 16:07:13 emte: I have encoding issues with sending my msg. Some guy remote claims he doesn't have the issue Oct 04 16:07:40 emte: I want to see what he is actually sending becuase it is not latin1/as it is supposed to be Oct 04 16:08:09 you sure its not a simple iso-utf8 clash? Oct 04 16:08:36 leoncamel, it resolves fine Oct 04 16:08:45 did you hardcode it? Oct 04 16:09:54 emte, my laptop is OK. and the build server here can't resolve monotone.openembedded.org Oct 04 16:10:25 64 bytes from linuxtogo.org (85.214.40.226): icmp_seq=1 ttl=49 time=509 ms Oct 04 16:10:41 slow response, but it seems to be there Oct 04 16:11:00 then again it is across the puddle Oct 04 16:11:19 OE developers: Could you make a folder for me in : "org.openembedded.dev/packages/linux/handhelds-pxa-2.6" directory named htcuniversal. And in that folder place a defconfig file i can mail you or something? Oct 04 16:11:39 emte, OMG. I have to add it to static host Oct 04 16:12:09 goxboxlive, why not add it as a feature in the buglist so someone can dld it? Oct 04 16:12:35 i did, but it seems like koen forgot it :-) Oct 04 16:12:53 use a bigger stick to prod Oct 04 16:12:54 :) Oct 04 16:13:43 anyway off to learn more about op-amp theory (whee) Oct 04 16:14:49 goxboxlive: bugzilla Oct 04 16:14:49 goxboxlive: your patch didn't contain the defconfig Oct 04 16:14:51 emte, what is "op-amp theory" ? Oct 04 16:15:14 koen: I'll upload it. Oct 04 16:17:04 leoncamel: the theory behind operational amplifiers... quite possibly the best analog parts since sliced bread :) Oct 04 16:17:08 leoncamel, do you know what an op-amp is? Oct 04 16:17:51 as for the best analog part ... that depends on aplication Oct 04 16:17:55 emte: but a theoretical op amp is pretty simple... it's those real ones that are a pain..:) Oct 04 16:18:06 they are not much use in high freq circuits Oct 04 16:18:32 emte: unless you need an attenuator :) Oct 04 16:19:11 yeah this is all the normal construction, usage, flaw, stuff Oct 04 16:19:18 anyway bbl Oct 04 16:19:29 emte: have fun Oct 04 16:19:53 emte, cu Oct 04 16:23:41 how do i add files to the bug report? Oct 04 16:23:50 a+ Oct 04 16:32:01 bug comitted. i just pasted in the defconfig content. Oct 04 16:32:04 koen, goxboxlive: so, how's xcalibrate issue? Does fresh mtn pull works finally? Oct 04 16:35:48 psokolovsky, I can't reslove monotone.openembedded.org and I can't connect to it now .. can you pull and update ? Oct 04 16:36:31 psokolovsky: the build completed Oct 04 16:36:59 psokolovsky: I didn't test it yet, I'm trying to get util-linux and busybox to play together nicely Oct 04 16:36:59 psokolovsky : check your mail, i have just sent you a mail. Oct 04 16:37:34 goxboxlive: could you attach it instead of pasting it inline? Oct 04 16:38:17 koen: How? I didnt find any upload file buttons in the bug list. Oct 04 16:38:33 now i found one Oct 04 16:38:43 it wasent there a minute ago Oct 04 16:38:52 yep, it only appears after you opened the bug Oct 04 16:39:30 goxboxlive, did you try latest pull? Oct 04 16:40:12 goxboxlive, you should. xcalibrate-0.0+cvs20060814-r0 - that's not right, fix was committed ~2hrs ago Oct 04 16:41:14 ok, i'll update . txh psokolovsy. Oct 04 16:43:41 koen: btw: PREFERRED_VERSION_handhelds-pxa-2.6 = "2.6.16-hh5" Oct 04 16:43:41 +PREFERRED_VERSION_handhelds-pxa-2.6_htcuniversal = "2.6.17-hh1+cvs${SRCDATE}" Oct 04 16:43:55 koen: what about letting a machine decide what kernel to use? Oct 04 16:44:23 koen: Actually, that's what happens for most machines, and only for handhelds-pxa-2.6 there's special treatment ;-) Oct 04 16:44:39 psokolovsky: the machine gets to choose which kernel to use, the distro which version Oct 04 16:44:55 psokolovsky: if that kernel doesn't work, it's a bug in the distro :) Oct 04 16:45:19 koen: Hm. but kernel version is special in this respect. it is really fully machine-specific Oct 04 16:45:56 good evening Oct 04 16:46:19 koen: but there're no PREFERRED_VERSION for linux-zaurus, netbook, etc ;-) Oct 04 16:46:20 psokolovsky: right, I like locked down versions, less surprises Oct 04 16:46:47 maybe we should make handelds-pxa-2.6_cvs.bb DEFAULT_PREFERENCE="-1" Oct 04 16:47:06 psokolovsky: those kernels don't have a history of brokenness as hh-pxa has :/ Oct 04 16:47:11 koen: but that doesn't work in real world ;-(. the fact that many kernels live in one repo (handhelds.org) doesn't mean they all devloped to the same extent ;-( Oct 04 16:47:44 koen: well, my suggestion is to have ?= for that. Oct 04 16:48:21 psokolovsky: I'd leave that to the machine mentors for the distro Oct 04 16:48:42 any thought on http://ewi546.ewi.utwente.nl/tmp/util-linux.diff ? Oct 04 16:49:32 yes Oct 04 16:49:39 looks good ;) Oct 04 16:49:42 koen: So yes, but how to approach that: either add machine-specific overrides to anstrom.conf, or have there "default", but let override in machine configs. I vote for the second. Oct 04 16:49:43 koen: the second part seems to be missing a closing brace Oct 04 16:49:59 s/second/third/ Oct 04 16:52:23 psokolovsky : xcalibrate where successfully build. thanks. Oct 04 16:52:30 goxboxlive, nice Oct 04 16:54:23 koen: so what do you think? Oct 04 16:54:53 later Oct 04 16:56:07 koen, where can I get the wiki of Moto A780 ? Oct 04 17:02:03 leoncamel: openezx.org Oct 04 17:02:15 psokolovsky: I'd like to have the distro to have a final say Oct 04 17:02:48 koen: with all your duplication removal going on, should we http://en.pastebin.ca/191102 Oct 04 17:03:28 koen: Ok! Oct 04 17:03:49 pH5: sure, although I'd like BOOTSTRAP to go away completely Oct 04 17:04:01 hey guys.. me a newbie. Need some suggestions on how I should go about building an OE distribution Oct 04 17:05:42 anushsh, read the GettingStart document on wiki.openembedded.org Oct 04 17:07:50 koen: it's a first step. where could I put the firmware packages tiinit and acx-firmware? can machine files set task-base-wifi-rrecommends_append? Oct 04 17:08:15 probably not, that would make task-base-wifi machine specific. better add a task-base-hx4700? Oct 04 17:08:15 pH5: RP is the task-base master Oct 04 17:12:19 ~conjure up RP Oct 04 17:13:05 hm, that didn't work Oct 04 17:13:22 pH5: summon? Oct 04 17:13:39 ~summon RP Oct 04 17:13:40 apt takes out 20 clean, identical-looking phones, some extra hands, and pretends to be a telemarketer for a large corporation, so he gets delivered a phonelist containing RP's coordinates Oct 04 17:18:56 03pH5 07org.oe.dev * r1eda1999... 10/ (1 conf/machine/ipaq-pxa270.conf): ipaq-pxa270.conf: first step towards removing BOOTSTRAP_ Oct 04 17:47:43 later Oct 04 18:13:45 zecke: I'm affraid I can't take the switch, I'm one network port short :( Oct 04 18:15:10 koen: we have wlan but we are firewalled anyway :) Oct 04 18:15:23 we can move mtn to https Oct 04 18:15:58 that's 441? Oct 04 18:16:37 443 i think Oct 04 18:16:55 or once and for all we can consider firewalls :) Oct 04 18:17:04 consider 'corporate' firewalls Oct 04 18:23:28 zecke: mtn --db=OE.mtn sync ewi546.ewi.utwente.nl:443 :) Oct 04 18:24:01 :) Oct 04 18:24:18 koen: let us hope this firewall is not filtering on content Oct 04 18:25:12 * koen leaves util-linux and turns on the tv Oct 04 18:26:11 koen|away: should that work now? Oct 04 18:29:29 freyther: How strict is the firewall? Oct 04 18:30:09 RP: the public wlan allows you to do http,https,pop,imap Oct 04 18:30:17 RP: in the private wlan you can do ssh Oct 04 18:30:55 eek Oct 04 18:31:08 ~lart too-restrictive firewalls Oct 04 18:31:08 * ibot squeezes too-restrictive firewalls till too-restrictive firewalls turns blue like papa smurf Oct 04 18:31:10 zecke: so vpn is blocked as well? Oct 04 18:31:38 ah well, if I can ssh, I can reconfig my vpn as needed :) Oct 04 18:31:39 RP: I think OpenVPN worked Oct 04 18:31:41 ssh forwarding is always an option, if ssh is allowed Oct 04 18:32:10 RP: This is why I asked for a switch :) Oct 04 18:33:07 I don't have anything small enough to transport :-/ Oct 04 18:33:27 We could turn something into a wireless access point :) Oct 04 18:34:11 ad-hoc networking Oct 04 18:34:46 or don't code and only talk :) Oct 04 18:34:57 :) Oct 04 18:35:03 We have enough to talk about... Oct 04 18:35:45 for friday have invited some OpenUsability.org folks Oct 04 18:35:58 I've invited... Oct 04 18:36:22 for friday evening to be exact Oct 04 18:37:11 You might want to define friday evening :p Oct 04 18:39:22 ~lart crappy wlan Oct 04 18:39:22 * ibot acting on orders from an unspecified client drags crappy wlan into court suing for $200 million Oct 04 18:42:12 ?! Oct 04 18:42:27 :) Oct 04 18:42:35 RP: Do you really want to know? Oct 04 18:42:52 Probably not :) Oct 04 18:43:08 RP: too late :) Oct 04 18:43:32 RP: I will mail you on where to meet Oct 04 18:43:52 zecke: ok Oct 04 18:43:54 RP: and I'm searching mail addresses Oct 04 18:44:21 zecke: email addresses for oedem people? Oct 04 18:44:32 RP: then I remembered, mickeyl used to call himself mickey and I tried to register his real nick :) Oct 04 18:44:40 RP: and one needs to have the nick to register it :) Oct 04 18:44:50 RP: yes, OEDEM people Oct 04 18:45:13 zecke: I can forward you one of the OEDEM emails if that would help? Oct 04 18:45:37 RP: yes, to freyther@inf.fu-berlin.de please :) Oct 04 18:46:08 zecke: done Oct 04 19:03:12 03pfalcon 07org.oe.dev * r22493c02... 10/ (12 files in 12 dirs): various packages: Add/adjust SECTION. Oct 04 19:25:18 hngr Oct 04 19:25:25 * koen stabs motorola Oct 04 19:25:39 zecke: the a780 refuses to accept your vcard :) Oct 04 19:26:13 which a780? Oct 04 19:27:42 my motorola a780 qt/e crap phone Oct 04 19:41:58 koen: bring your mirrorball Oct 04 19:44:13 done :) Oct 04 19:44:36 and a box of frozen frikandel ;) Oct 04 19:46:41 sorry, no more room Oct 04 19:46:53 20 t-shirts take a lot of room in a small bag :) Oct 04 19:48:02 * RP plays with maemo-mapper on the 770 Oct 04 19:48:09 03koen 07org.oe.dev * r91356d1e... 10/ (3 files in 3 dirs): handhelds-pxa-2.6: add defconfig for htc-universal Oct 04 19:48:42 hi all Oct 04 19:48:50 hey cedricb Oct 04 19:49:00 RP: I did the same today Oct 04 19:49:34 RP: I used this one: http://armin-warda.de/maemo-mapper-1.1-cj-aw-5 Oct 04 19:49:46 that allows me to switch between topo and satelite Oct 04 19:50:22 koen: I have something I installed a while back. I know if I try and change now, it'll break... Oct 04 19:50:34 koen: sorry to bother you again... about the util-linux-mount thingy, I've rebuild task-bootstrap and bootstrap-image, it doesn't included the util-linux-mount stuff... Oct 04 19:50:36 do most touchscreen programs use the compaq touchsreen interface? Oct 04 19:50:43 * koen cached a lot of berlin today Oct 04 19:50:44 Gerrath: no Oct 04 19:50:51 Gerrath: It should die Oct 04 19:50:58 RP so I should not enable it in the kernel :-) Oct 04 19:50:59 Gerrath: die horribly Oct 04 19:51:11 Gerrath: people should use tslib Oct 04 19:52:12 Gerrath: No Oct 04 19:52:33 evdev should be enough, right? Oct 04 19:52:38 koen, ok, I've applied the arm ts patches for the UCB1400, I'm just having some difficaulty getting it all set up. Is there a good example I can work off of? I Oct 04 19:53:11 koen: right Oct 04 19:53:41 Gerrath: most apps use tslib Oct 04 19:53:58 Gerrath: which abstracts from the kernel interface and has a rich set of modules Oct 04 19:54:06 Gerrath: and some basic utils to test your driver Oct 04 19:54:55 zecke, thats what I'm using (ts_calibrate..)'' Oct 04 19:55:57 good :) Oct 04 19:56:07 QtE/QtopiaCore and X use/can use tslib Oct 04 19:56:34 03koen 07org.oe.dev * r98fb56b9... 10/ (1 packages/angstrom/task-angstrom.bb): task-angstrom: move tslib utils to debug-section Oct 04 20:06:19 zecke, Have you used the UCB1400 TS before? Oct 04 20:07:01 is RMK still slacking on the ucb front? Oct 04 20:07:19 I seem to remember he said he'd fix it a few *years* ago Oct 04 20:09:08 koen, I have applied the patches from the arm linux kernel site. It seems to work but I can only get an event device. This only seems to signal when there has been a change in the ts state and not the actual values.. When I used the compaq interface I got actual changing values but of course it is an impossible interface that doesn't work with anything :-/ Oct 04 20:11:05 there are also patches in the linux-gumstix kernel directory of OE for the ucb1400 which are just the arm linux kernel patches. Oct 04 20:11:11 there was a post on L-A-K about ucb1400 recently Oct 04 20:11:26 koen: I've added to task-base and the device is using the real mount now but the root partition is stilled read-only... Oct 04 20:12:14 cedricb: weird Oct 04 20:12:39 I'm assuming the fact that I have compiled in support for the compaq ts that it is not causing me issues. I do not have the compaq ts module loaded. Oct 04 20:13:18 koen: have u got a similar issue with yr angstrom distro? (using busybox 1.2.1?) Oct 04 20:13:33 cedricb: I haven't noticed it yet Oct 04 20:14:58 I'm also assuming that I do want the event interface compiled into the kernel.. Oct 04 20:15:08 so it's something else... RP: do u mind building a poodle bootstrap image from .dev? :-) Oct 04 20:20:13 cedricb: tried any of these: http://www.angstrom-distribution.org/unstable/images/20060905/poodle/ ? Oct 04 20:28:22 koen: same thing, so same problem with oz-unstable Oct 04 20:28:57 the only difference, I was using kernel 2.6.18 Oct 04 20:38:01 koen|away, isnt KERNEL_CCSUFFIX_sh3 = "-3.4.4" a correct expression? gcc_4.1.1.bb seems to ignore it .. Oct 04 20:40:54 koen: let me know next time if you find something... Oct 04 20:40:59 nite all Oct 04 20:41:57 re Oct 04 20:46:05 wpw my bonsai rocks :) Oct 04 20:46:31 luckily hrw does not look at my html2 skills :) Oct 04 20:53:35 I want create ipk files (ipackages) but not have them install into the rootfs image itself. These particular packages I would like to simply put on my own ipk server. How can I do this as a part of my project build (IOW when I do a bitbake bootzipit-image, those ipk files are also built). Oct 04 20:54:07 would I make that a task, and run that task seperately? Oct 04 21:05:50 T0mW: DEPEND on this package somewhere Oct 04 21:06:05 T0mW: but do not add it to RDEPENDS Oct 04 21:06:12 ah, thank you Oct 04 21:06:21 T0mW: run bitbake package-index to generate the 'feed' Oct 04 21:06:36 DEPENDS (just to make sure) Oct 04 21:24:35 somebody wants to try my bonsai? Oct 04 21:24:39 or wants a free copy? Oct 04 21:32:56 * v8jlene is back Oct 04 21:35:30 zecke: uh? Oct 04 21:35:52 florian: a clone of bonsai.mozilla.org Oct 04 21:36:11 florian: for Subversion, written in python, using django Oct 04 21:36:27 florian: a hack within eight hours of time Oct 04 21:36:44 http://127.0.0.1:8000/bonsai/query/description/foo/ <- this queries all commit messages containing foo Oct 04 21:40:59 zecke: hmm... that's nice Oct 04 21:41:21 * florian needs to move gpe to use svn Oct 04 21:43:29 page.mi.fu-berlin.de/~freyther/django_tinderbox_bonsai_1.tar.bz2 should be the src Oct 04 21:48:18 hmmm Oct 04 21:54:41 sorry, my usb wlan adapter is... Oct 04 21:56:27 JustinP: did you have any comments? Oct 04 21:59:11 zecke: no, just the hmmm ;-) Oct 04 21:59:29 zecke: I didn't look at yours at all (the link being for localhost), but the idea seems decent Oct 04 21:59:45 searching the source repository is a good thing to have ;-) Oct 04 22:00:01 perhaps I should try it here at work.....we could use such a thing.... Oct 04 22:00:25 am I right in thinking this is a SVN searching script? Oct 04 22:00:41 JustinP: this is a SVN searching webapp :) Oct 04 22:01:00 ./bonsai_add.py "BitBake all" "svn" "viewsvn" "svn.berlios.de" "http" "svnroot/repos/bitbake" "" Oct 04 22:01:08 this add a Repository Oct 04 22:01:22 and ./bonsai_update.py "BitBake all" will get the log and index Oct 04 22:02:24 hmmm....does this index a viewsvn site then? Oct 04 22:03:11 JustinP: no, for svn it uses the svn python bindings Oct 04 22:03:13 errr...this has something to do with tinderbox? Oct 04 22:03:51 JustinP: tinderbox will integrate bonsai and will be written with django as well. As I'm learning how to use django, I have started with writing bonsai first Oct 04 22:04:08 what are the parameters? I'll attempt to test it on my home site.... Oct 04 22:04:52 what does the "viewsvn" do, for instance Oct 04 22:05:04 good night Oct 04 22:05:32 g'night. see you in a couple of hours :D Oct 04 22:06:56 JustinP: you need to get django first Oct 04 22:07:27 JustinP: and update settings.py to point somewhere else than /home/freyther/tinderbox/templates Oct 04 22:07:34 JustinP: python manage.py syncdb (will create the initial sqlite db) Oct 04 22:08:24 hmmm, ok Oct 04 22:08:24 JustinP: python manage.py reset tinderboxserver (will create the bonsai tables) Oct 04 22:08:24 JustinP: now you can enter which site to index Oct 04 22:08:48 * JustinP emerges django Oct 04 22:09:09 my poor single-core machine..... Oct 04 22:09:26 * JustinP suspends his revdep-rebuild Oct 04 22:09:40 ./bonsai_add.py "NAME " "SCM system (svn for now)" "URL to a ViewVC/ViewSVN site" "URL of the server/path" "PROTOCOL" "PATH TO REPO" "BRANCH e.g. 'trunk/foo'" Oct 04 22:09:56 ./bonsai_update.py "NAME" Oct 04 22:10:30 ok, I'll see if I can get it working :-) Oct 04 22:10:32 thanks Oct 04 22:10:47 of course I don't have viewvc, but I assume it will just make broken links ;-) Oct 04 22:11:05 it is not used at all for now :} Oct 04 22:11:11 ah, good :-) Oct 04 22:11:25 django is a really nice framework, but my HTML 'skills' are lacking Oct 04 22:12:09 and the view code is quite dirty as well (some code duplication already), my queries are inefficient but django really rocks so far Oct 04 22:12:20 adding RSS support sounds easy as well Oct 04 22:14:09 * JustinP has never understood the utility of RSS Oct 04 22:16:06 JustinP: almost pushing information to people Oct 04 22:19:15 good nite Oct 04 22:26:19 that's exactly it....it's *not* push....and I don't see how using a reader is better than just getting this via e-mail.....e-mail *is* push and you keep all of your old messages and can read them when you want. With RSS you have to check it within X hours/days to get all entries and then use the same program to remember read/unread, etc. Oct 04 22:27:54 RSS is flawed because you wan't to be able to read it in two locations? Oct 04 22:27:59 thats strange logic Oct 04 22:29:45 no Oct 04 22:30:00 ugh Oct 04 22:30:05 read the rest, forget the last part Oct 04 22:30:37 why is it that people always attack one point as if it's the only thing that matters? Oct 04 22:30:48 you shouldn't conclude with it Oct 04 22:30:59 The last point is always considered to be the strongest Oct 04 22:31:45 well, people need to read things as a list of points, not as if one is necessarily the most important....depends on what type of writing if the first or last is most important... Oct 04 22:31:49 anwyay Oct 04 22:32:28 03lenehan 07org.oe.dev * r9694cc03... 10/ (6 files in 2 dirs): Oct 04 22:32:28 perl modules: Exclude debugging information from the modules that are Oct 04 22:32:28 currently including it. Oct 04 22:32:32 it's flawed because it acts like push when it's pull and inherently makes you re-check every (x) if you want to catch anything. Why not just use e-mail? Oct 04 22:32:34 03lenehan 07org.oe.dev * r43a00136... 10/ (6 files in 2 dirs): Oct 04 22:32:34 util-linux: Put mkfs.minix and fsck.minix under update-alternatives Oct 04 22:32:34 control so they don't conflict with the busybox versions. Oct 04 22:36:59 I personally like getting my slashdot/VCS commits/etc in my inbox. **** ENDING LOGGING AT Thu Oct 05 02:59:57 2006