**** BEGIN LOGGING AT Fri Jan 07 02:59:58 2011 Jan 07 04:18:40 anyone around for some bitbake questions? Jan 07 04:38:34 Hi friends Jan 07 04:40:21 I m get an error while buildind oe ERROR: Please set the 'PERSISTENT_DIR' or 'CACHE' variable. Jan 07 04:40:42 what is the problem Jan 07 04:51:52 how to set 'PERSISTENT_DIR' or 'CACHE' variable Jan 07 05:54:47 can someone please help? for some reason none of my kernel modules that I specify in my recipe show up in the image it creates Jan 07 05:56:25 lostincake1, what timezone are you in? Jan 07 06:07:07 USA east coast Jan 07 06:07:11 -5/-4 Jan 07 06:08:11 you are about 2hrs early for .eu and 3hrs too late for US :( Jan 07 06:11:59 can anyone help ... I m getting thid error while bitbake ERROR: Please set the 'PERSISTENT_DIR' or 'CACHE' variable. Jan 07 06:13:12 ka6sox: yeah I know... I should ask in the morning Jan 07 06:13:27 ya, Jan 07 06:13:31 sorry lostincake1 Jan 07 06:13:48 sanket, what TZ are you in? Jan 07 06:14:06 Asia pacific Jan 07 06:14:54 sanket, as I was saying to lostincake1 .eu isn't up yet and .us is asleep :( Jan 07 06:15:02 2hrs and .eu will start waking up. Jan 07 06:17:54 I m nt able to get out of that error ERROR: Please set the 'PERSISTENT_DIR' or 'CACHE' variable. Plz help me.... Jan 07 06:19:36 sanket, I work on the infrastructure here...so I can't help much. sorry :( Jan 07 06:29:34 morning Jan 07 06:31:31 morning hrw... Jan 07 06:31:40 nuked 423 BAD users Jan 07 06:31:48 nuked 480 BAD pages Jan 07 06:32:12 locked down wiki Jan 07 06:32:34 using NukeDPL instead of Nuke...easier to find/kill Jan 07 06:33:30 thx Jan 07 06:33:49 I lacked time in last month to take a look at situation Jan 07 06:34:33 hrw, np...I found a mediawiki expert and he and I have been kicking it. Jan 07 07:06:53 sanket: did u change the user id ? Jan 07 07:07:10 and you should never run OE builds using root Jan 07 07:07:50 without using root.... I m geting that error Jan 07 07:09:21 lostincake1: are you writing your own kernel recipe Jan 07 07:09:56 sanket: ok remove the oe tmpdir Jan 07 07:09:58 and redo Jan 07 07:10:12 sanket: and what version of bitbake and OE are u using Jan 07 07:11:48 bitbake version 1.8.18 Jan 07 07:13:13 I guess you didn't setup env in your shell Jan 07 07:14:05 gm Jan 07 07:14:47 eFfeM_work, morning...could you please try to do an edit on the wiki? Jan 07 07:17:41 khem: nope, just trying to extend x11-image to pre-install somethings I want like ti-dsplink and ti-local-power-manager Jan 07 07:20:32 ka6sox: with the .net or .org addr ? Jan 07 07:20:54 NEVER .net...forget that exists :D Jan 07 07:21:24 dah it is in my bookmarks :-( Jan 07 07:22:03 NUKE that... Jan 07 07:22:24 will do Jan 07 07:22:40 edit gives Jan 07 07:22:41 Spam protection filter From Openembedded Jump to: navigation, search The page you wanted to save was blocked by the spam filter. This is probably caused by a link to a blacklisted external site. The following text is what triggered our spam filter: http://akismet.com blacklist error Return to Testing. Jan 07 07:22:45 and I am logged in Jan 07 07:23:14 okay what were you trying to point to? Jan 07 07:23:22 and what was the edit? Jan 07 07:24:02 the edit was on http://wiki.openembedded.net/index.php/Testing#Overview and I added and removed a . after the item starting with 1 Jan 07 07:24:12 I did edit only the section, not the page Jan 07 07:24:30 lostincake1: x11-image should have modules Jan 07 07:24:38 lostincake1: does it have none ? Jan 07 07:25:59 eFfeM_work, yes, that is CORRECT...openembedded.net does NOT exist Jan 07 07:26:01 you are still pointing to it Jan 07 07:26:26 khem, gm, will you pick up readding USE_LDCONFIG for glibc? it was dropped in with your commit 4e6b1bfe289ad346cbd84926a43ad0f93760930d Jan 07 07:26:42 ka6sox: sorry pasted the wrong url Jan 07 07:27:02 k Jan 07 07:27:07 but will recheck Jan 07 07:27:11 that url should fail. Jan 07 07:27:40 khem, or maybe better: give ldconfig its own package Jan 07 07:28:37 khem: there are some modules in modules-beagleboard.tgz, but no ti-dsplink for some reason, it builds the .ipk and sticks it in the local ipk store, so if I boot into my image nad say "opkg install ti-dsplink", it already has the .ipk... but I want it to be installed already Jan 07 07:29:29 ka6sox: , verified: same error Jan 07 07:29:31 url http://wiki.openembedded.org/index.php?title=Testing&action=submit Jan 07 07:29:45 okay jas Jan 07 07:29:46 let me try something Jan 07 07:30:23 strange thing is that http://akismet.com is not in my section and as far as I can see also not on the page Jan 07 07:30:38 right I think I see it. Jan 07 07:30:42 lostincake1: if you have problems with the ti recipes better go to the ti people e.g. in #beagle Jan 07 07:30:45 try again please Jan 07 07:31:31 works :-) Jan 07 07:32:00 okay thanks Jan 07 07:32:11 yw Jan 07 07:32:12 have to fix akismet tommorrow. Jan 07 07:32:30 for now we shouldn't have any more bozo's being able to sign up Jan 07 07:32:47 tomorrow? it is only 11.30 pm for you ! we're used to have you here until 3 am :-) Jan 07 07:33:23 I've already put in my 16hrs today :D Jan 07 07:34:03 I'm just kidding, i hope you noticed that, i really appreciate what you do for the community Jan 07 07:34:14 glad to help Jan 07 07:35:03 okay I need to figure out how to use the ML correctly and see if I can dredge up some volunteers for a booth @ SCALE (www.socallinuxexpo.org) Jan 07 07:37:18 gm Jan 07 07:37:25 gm likewise Jan 07 07:38:06 hi likewise, you're early :-) good intentions for 2011 ? Jan 07 07:39:26 eFfeM_work: yes: do more OE related work :-) Jan 07 07:39:37 ah great Jan 07 07:40:46 likewise: you still have the src material for the OE posters? I'm not sure whether they still exist and can be recovered, otherwise I probably need to make a new set (with your instructions :-) ) Jan 07 07:41:20 eFfeM_work: hmmm Jan 07 07:42:15 eFfeM_work: yes I should have those still Jan 07 07:42:34 eFfeM_work: you still have the availability to that poster printer? Jan 07 07:42:40 likewise: I hope the originals still show up Jan 07 07:42:56 eFfeM_work: the originals are on my Mac Jan 07 07:43:21 eFfeM_work: and printing to that machine worked only from a Mac (I have tried Linux and that other OS). Jan 07 07:43:29 likewise: this was done at the printer at Peter S 's office isn't it? Jan 07 07:43:33 yes Jan 07 07:43:56 that one is still there, guess I need to find some paper and tackle Walter with his mac to do the printing Jan 07 07:44:01 people should just use eglibc Jan 07 07:44:53 khem, not sure what eglibc brings compared to glibc (but not sure about the reverse either); actually sizewize they seem to be on par Jan 07 07:45:18 * eFfeM_work originally assumed that eglibc would be a lot smaller being for embedded but soon found out it wasn't) Jan 07 07:45:28 eFfeM_work: eglibc is more maintained Jan 07 07:45:40 it can be smaller Jan 07 07:46:09 I havennt yet added the logic to make it configurable like uclibc Jan 07 07:46:21 and glibc can not be compiled wth -Os Jan 07 07:46:37 I'd love to learn how to make it smaller, moving to a newer version is probably not too acceptable for some of our products (actually moving to eglibc might not be either) Jan 07 07:46:41 so it definitely has advantage Jan 07 07:47:16 and if I recall correctly uclibc was not good enough if you needed pango or cairo or something like that (forgot details, so maybe it was a different package) Jan 07 07:47:28 well if moving is not acceptable then making it smaller for you is mootpoint Jan 07 07:47:43 btw the uclibc++ recipe seems to work nicely Jan 07 07:48:08 gm khem Jan 07 07:48:16 likewise: hello Jan 07 07:48:32 khem, for new products move can be done, but I need to have some reasoning, otherwise people will just yell at me if there are problems Jan 07 07:48:37 which probably translates to gm-(UTC-6) or so Jan 07 07:48:56 likewise: yes its 11:45pm here Jan 07 07:49:17 hello I m getting this error while executing "bitbake " sqlite3.OperationalError: attempt to write a readonly database Jan 07 07:49:29 eFfeM_work: well what we already have in there itself has advantages Jan 07 07:49:30 eFfeM_work: reasoning -> testing Jan 07 07:49:41 like you dont need qemu anymore to generate locales Jan 07 07:49:45 with eglibc Jan 07 07:49:56 likewise: indeed Jan 07 07:49:58 khem: yes, thanks for that work khem! Jan 07 07:50:06 and -Os will safe you size Jan 07 07:50:36 eFfeM_work: there is a README of some sorts on how to configure eglibc components in the top level dir Jan 07 07:50:53 likewise: will peek at it Jan 07 07:51:00 eFfeM_work: nothing prevents you from modifying that (say, just before configure stage) Jan 07 07:51:11 likewise: yes I have it being done in my local toolchain butnot in OE Jan 07 07:51:26 OE will need knobs to configure them Jan 07 07:51:33 may be DISTRO_FEATURES or somesuch Jan 07 07:51:37 yes, that's probably the hardest part Jan 07 07:51:40 the mapping Jan 07 07:55:55 ERROR: Please set the 'PERSISTENT_DIR' or 'CACHE' variable. Jan 07 07:56:06 how to solve this error Jan 07 07:56:06 sanket: I guess you didn't setup env in your shell Jan 07 07:57:19 JaMa, Do u hv a complete guidelines/website/pdf for openembedded Jan 07 07:58:48 http://wiki.openembedded.org/index.php/Getting_started "Setup the environment" Jan 07 07:58:52 sanket: ^ Jan 07 08:02:24 no e-mail in openembedded-members this year or web archive stopped working on 2010-December? Jan 07 08:03:01 JaMa, not seen any yet Jan 07 08:03:14 ok, thx Jan 07 08:45:36 03Simon Busch  07org.openembedded.dev * r471eef5612 10openembedded.git/conf/machine/ (include/palmpre.inc palmpre.conf): Jan 07 08:45:36 palmpre machine: split into machine and common parts Jan 07 08:45:36 Signed-off-by: Simon Busch Jan 07 08:45:38 03Simon Busch  07org.openembedded.dev * r7adf1b0206 10openembedded.git/conf/machine/palmpre2.conf: Jan 07 08:45:38 machines: introduce new palmpre2 machine Jan 07 08:45:38 The new palmpre2 machine is nearly the same as the already available palmpre machine. The only Jan 07 08:45:38 difference is the internal processor. The palmpre2 uses a omap 3630 and the palmpre a omap 3430 Jan 07 08:45:38 processor. This takes us into the situation to handle two different machines. Jan 07 08:45:38 Signed-off-by: Simon Busch Jan 07 08:45:39 03Simon Busch  07org.openembedded.dev * re5bdccdf86 10openembedded.git/recipes/linux/linux-palmpre/palmpre2/defconfig: Jan 07 08:45:40 linux-palmpre: add defconfig for palmpre2 machine Jan 07 08:45:40 Signed-off-by: Simon Busch Jan 07 09:16:21 good morning Jan 07 09:35:32 eFfeM_work: likewise I put together something quickly to enable eglibc optioning. Its pretty much controlled with DISTRO_FEATURES, default is to enable all features as it is now and if one needs then he can disable certain features. Try it out the patch should be on the ml by now Jan 07 09:37:04 * khem thinks its time to sleep now Jan 07 09:37:07 gn Jan 07 09:42:59 khem gn Jan 07 09:43:26 gn khem Jan 07 09:43:33 khem: and tnx, will have a look at that Jan 07 09:44:08 khem, just read the mail, looks good to me Jan 07 09:45:40 03Chase Maupin  07org.openembedded.dev * rf3930609a8 10openembedded.git/recipes/openssl/ (ocf-linux.inc ocf-linux_20100325.bb): Jan 07 09:45:40 ocf-linux: add recipe to stage ocf linux headers Jan 07 09:45:40 * This recipe is used to stage ocf linux headers in the proper Jan 07 09:45:40 usr/include/crypto directory structure expected by packages Jan 07 09:45:40 such as openssl. Jan 07 09:45:41 Signed-off-by: Chase Maupin Jan 07 09:45:42 Signed-off-by: Koen Kooi Jan 07 10:24:59 how can i change DISTRO_FEED_URI without removing my tmp dir? Jan 07 10:27:16 03Chase Maupin  07org.openembedded.dev * rdd48577386 10openembedded.git/recipes/openssl/openssl_1.0.0c.bb: (log message trimmed) Jan 07 10:27:16 openssl: add support for OCF linux drivers Jan 07 10:27:16 * This patch adds support for openssl to use OCF linux drivers Jan 07 10:27:16 if they are available. Jan 07 10:27:16 * If the drivers are not available then openssl will Jan 07 10:27:16 default back to its own software implementation. Jan 07 10:27:16 * Only enabled for the target side version since it doesn't Jan 07 10:27:35 Heinervdm: isn't this just used in distro-feed-configs? Jan 07 10:27:38 so you'd have to rebuild this Jan 07 10:28:08 mickey|office: ah ok, i thought it's saved somewhere else Jan 07 10:29:10 a grep doesn't reveal anything here, so i think it's just that Jan 07 10:30:04 yes, i tried to grep and haven't found anything too :) Jan 07 10:31:02 fwiw, someone should check whether there's anything that can be taken from the newer angstrom-feed-configs Jan 07 10:34:43 03Koen Kooi  07org.openembedded.dev * r546e301bee 10openembedded.git/conf/distro/include/angstrom-2010-preferred-versions.inc: Jan 07 10:34:43 angstrom next: prefer wpa supplicant 0.6.10 Jan 07 10:34:43 Signed-off-by: Koen Kooi Jan 07 10:34:54 03Koen Kooi  07org.openembedded.dev * r3878f478e3 10openembedded.git/conf/distro/include/angstrom-2010-preferred-versions.inc: Jan 07 10:34:54 angstrom next: prefer dropbear 0.52 Jan 07 10:34:54 Signed-off-by: Koen Kooi Jan 07 10:35:07 Go Koen! Jan 07 10:53:11 concerning Fosdem: what hour does it usually close on sunday evening (to see if I can catch a train sunday evening or not in case I manage to come) ? Jan 07 10:54:32 hi GNUtoo|laptop Jan 07 10:54:39 ericben, hi Jan 07 10:54:45 did you saw my pastebin? Jan 07 10:54:49 happy new year to you :) Jan 07 10:54:59 thanks, happy new year to you too Jan 07 10:55:06 I saw it. So you get the the kernel booting Jan 07 10:55:11 yes Jan 07 10:55:31 now you need to map the cpld to enable the RS232 transceiver and you will get a console Jan 07 10:55:35 I'll try to get serial working when I'll have some time Jan 07 10:55:42 ok Jan 07 10:58:08 ericben, any news about the internship? Jan 07 10:58:26 GNUtoo|laptop: I'm sending you a mail to discuss this Jan 07 10:58:39 ok thanks a lot Jan 07 11:04:39 GNUtoo|laptop: mail sent Jan 07 11:05:04 thanks a lot Jan 07 11:10:24 btw GNUtoo|laptop are you going to Fosdem ? Jan 07 11:10:54 I think so Jan 07 11:11:32 GNUtoo|laptop: do you know when that usually ends on sunday ? late or not ? to know if I need to stay in brussell on sunday night Jan 07 11:12:32 yes it ends on sunday, usually it ends like that Jan 07 11:12:36 there is the last main talk Jan 07 11:12:42 then the lottery etc... Jan 07 11:12:46 about which hour ? Jan 07 11:12:57 I'll look Jan 07 11:13:08 do you stay on sunday evening ? Jan 07 11:13:53 18:00 Jan 07 11:13:55 yes Jan 07 11:14:07 in fact for my 6 hours train trip I think I would need to leave to early so I have to stay one more night. Last thing to do negociate this with at home :-) Jan 07 11:14:17 ok Jan 07 11:18:16 GNUtoo|laptop: for the cpld mapping, you can take inspiration in : http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=arch/arm/plat-mxc/3ds_debugboard.c;h=639c54a0799232a9f2b3f17657154f514ea71417;hb=HEAD which handles the debug board of the FSL's evaluation boards which also has a CPLD. Jan 07 11:18:44 thanks a lot!!!! Jan 07 11:18:52 I also found some resources on free electron Jan 07 11:19:02 now they have stuff on the kernel Jan 07 11:19:12 it was added recently Jan 07 11:19:42 better than the last book on kernel that was reviewed by lwn.... Jan 07 11:19:44 03Lianhao Lu  07master * r01c4bffcd9 10bitbake.git/lib/bb/data.py: (log message trimmed) Jan 07 11:19:45 bitbake/data.py: corrected the output for shell syntax. Jan 07 11:19:45 [BUGID# 645], modify the emit_var() Jan 07 11:19:45 1. Added "#" to the beginning of each line if the comment contains Jan 07 11:19:45 multiple lines. Jan 07 11:19:45 2. Added "\" to the end of each line if the shell variable value Jan 07 11:19:45 contains multiple lines. Jan 07 11:23:56 GNUtoo|laptop: http://www.amazon.com/gp/product/0672329468?ie=UTF8&tag=roblov-20&linkCode=as2&camp=1789&creative=390957&creativeASIN=0672329468 Jan 07 11:23:59 eFfeM_work: the OE FOSDEM 2011 poster is in your email Jan 07 11:24:33 likewise: eFfeM_work: for the OE poster it would be great to have small logos of sponsors of OE Jan 07 11:25:11 ericben, yes this one, it isn't that good according to lwn.net, there are some stuff from the 2.4 kernel when talking about 2.6.xx Jan 07 11:25:16 ericben: three sponsors are currently on there, logo design, poster design and poster printing sponsors. We should discuss other sponsors on the list then., Jan 07 11:25:26 will there be stickers? Pins Jan 07 11:25:29 ericben: which company you work for? Jan 07 11:25:31 ? Jan 07 11:25:38 or other nice to have stuff? Jan 07 11:25:40 likewise: eukrea Jan 07 11:25:41 :) Jan 07 11:25:41 tdebrouw: I have a design for pins Jan 07 11:26:26 I did a design for OpenEmbedded people wearing the new OE logo with their name underneath. I am open for ideas. Jan 07 11:26:32 Let's throw this on the nl. Jan 07 11:26:35 nl->ml Jan 07 11:27:10 pins are nice; the OE logo is round :-) Jan 07 11:27:32 I mean pin buttons. Expect they are so 198x :-( Jan 07 11:27:39 fsf sold pins last year (i believe) Jan 07 11:29:37 I don't believe people buying OE pins, they expect them to be free. Jan 07 11:30:17 shirts would be nice Jan 07 11:30:24 indeed Jan 07 11:31:25 indeed Jan 07 11:31:40 my OE t-shirt is over 4 years old now and has old logo Jan 07 11:32:55 good morning Jan 07 11:34:28 I agree about pins... they are hard to sell compared to the quantities you have to order. Jan 07 11:34:43 np Jan 07 11:34:58 selling t-shirts then? Jan 07 11:35:14 btw, there's an error on the OE site "Warning: Parameter 1 to Language::getMagic() expected to be a reference, value given in /usr/share/mediawiki/includes/StubObject.php on line 58" Jan 07 11:35:17 doesn't look good Jan 07 11:37:24 likewise, all, got the files from likewise, if they need modification, feel free to do so; I am graphically challenged, got straight F's for drawing since kindergarten Jan 07 11:38:37 lubnch Jan 07 11:39:55 ~bon appetit Jan 07 11:47:32 ericben, what kind of test can be done on an embedded board? are the tests hardware(like kernel driver) related, or software related, note that the kernel people rely on user bugreports for testing Jan 07 12:03:02 GNUtoo|laptop: the 3rd edition of that book is good Jan 07 12:03:11 GNUtoo|laptop: it is very recent AFAIK Jan 07 12:03:18 yes but did you read that: Jan 07 12:03:32 http://lwn.net/Articles/419855/ Jan 07 12:04:23 GNUtoo|laptop: no thanks, hmm, too bad Jan 07 12:38:49 Hi all Jan 07 12:38:58 what's the best smartphone today to develop ? Jan 07 12:39:15 angelox_123, hi Jan 07 12:39:22 it depends Jan 07 12:39:40 if it has to be supported by openembedded/SHR/FSO looks here: Jan 07 12:39:54 http://wiki.freesmartphone.org/index.php/HardwareComparison Jan 07 12:41:27 hum...n900 is a good choice Jan 07 12:42:19 yes but modem is incomplete Jan 07 12:42:24 no calls,sms etc...yet Jan 07 12:42:32 but ofono is free software Jan 07 12:42:38 and it has support for the modem Jan 07 12:42:55 good Jan 07 12:43:15 i can't work with old phones no more :( Jan 07 12:53:49 angelox_123, /join #openmoko-cdevel then Jan 07 12:54:01 the ports to newer phone happen there Jan 07 12:54:43 hum...i'll check Jan 07 12:54:45 thanks :) Jan 07 13:22:53 GNUtoo|laptop: back from lunch. For the intership subject, we will discuss in private. For the book, http://www.amazon.fr/Essential-Device-Drivers-Sreekrishnan-Venkateswaran/dp/0132396556/ref=sr_1_2?ie=UTF8&qid=1294406306&sr=8-2 this one is not bad. R Love's one and LDD are not bad also, considering the speed Linux evolves a book can't be up to date. Jan 07 13:23:21 yes but having 2.4 stuff.... Jan 07 13:23:28 free electron material is fairly recent Jan 07 13:23:35 ok Jan 07 13:28:06 FE's things is not a book so it can evolve daily ! Jan 07 13:28:22 *g* Jan 07 13:35:03 hi jama Jan 07 13:35:22 hi woglinde Jan 07 13:36:35 likewise: for T-shirts, main problem is the delay (at least 2 weeks to get a reasonnable price) Jan 07 13:58:12 * mwester can't see his kernel boot in qemu... :( Jan 07 13:59:32 DISTRO=minimal MACHINE=qemuarm bitbake console-image <------- builds fine, boots in qemu-system-arm to the vnc session, but I cannot get any console output. Jan 07 14:00:01 Last words are about decompressing the kernel, then nothing. Jan 07 14:00:09 Anyone have the secret? Jan 07 14:01:51 Seems easy enough -- use the "-nographic" option, and make sure that the -append option includes "console=ttyAMA0"... but no luck -- not a peep from the booting kernel. Jan 07 14:03:07 So strange... one would think the whole point of qemu is to debug things like kernels -- I must be missing something so obvious that not even google can find help for me! :) Jan 07 14:08:35 mwester hm Jan 07 14:08:37 I got it run Jan 07 14:08:44 but dont know the steps athand Jan 07 14:08:47 03Richard Purdie  07master * r6de4f9f585 10bitbake.git/lib/bb/build.py: Jan 07 14:08:48 bitbake/build.py: Fix del_stamp work correctly after recent stamp function changes Jan 07 14:08:48 (From Poky rev: 5da9747) Jan 07 14:08:48 Signed-off-by: Richard Purdie Jan 07 14:09:05 anymore Jan 07 14:10:30 I tested it with klibc stuff Jan 07 14:12:56 Ok, well, I guess that's good news. It means that it does work. It also means that I'm still missing something very basic. :( Jan 07 14:13:20 mwester: look here for some more hints http://www.linuxtogo.org/gowiki/AngstromAndQemu Jan 07 14:14:15 mwester: "console=ttyAMA0". Is it ttyAM0 (one A) ? Jan 07 14:15:16 ant_work: That was the most helpful page that google found for me -- with that I was able to get it to boot to the vnc device. But still no output to the tty from the booting kernel showed up on any of the consoles, stdio nor the virtual ones. Jan 07 14:15:31 ericben: I shall try... one moment. Jan 07 14:16:50 Nope, same problem. As far as I can tell from the booted image, the serial port devices as identified by the kernel are ttyAMA. Jan 07 14:17:31 mwester: which kernel version is it and which qemu version ? Jan 07 14:18:15 kernel is 2.6.35 Jan 07 14:19:19 qemu-system-arm (build by OE) is 0.12.5 Jan 07 14:19:31 qemu-system-arm (Fedora host) is 0.11.0 Jan 07 14:19:55 Neither qemu-system-arm version displays any difference in the console output Jan 07 14:20:32 mwester: is it working with this vmlinuz : http://people.debian.org/~aurel32/qemu/arm/ ? Jan 07 14:21:40 mwester: this one works here with qemu 0.13 (Fedora 14) Jan 07 14:24:16 mwester hm the other problem I faced was input stuff Jan 07 14:24:29 there was serval option not enabled in the oe build kernel Jan 07 14:26:04 ericben: Same problem. Just get the decompressing kernel message on stdout, then nothing further. Jan 07 14:26:48 mwester: with both 0.12.5 & 0.11.0 ? strange Jan 07 14:27:06 mwester: qemu-system-arm -M versatilepb -kernel vmlinuz-2.6.26-2-versatile -nographic -append "console=ttyAMA0" works here Jan 07 14:27:32 woglinde: I noticed some options in the qemu-system-arm built by OE didn't work as described by the docs, so i installed the host version of qemu-system-arm. Both fail to provide any console output after the decompressing kernel message. Jan 07 14:27:36 mwester: I'll launch a build for qemuarm on my machine and let you know Jan 07 14:27:46 :) Thanks. Jan 07 14:29:26 * mwester types ericben's command line literally, and observes (!) a kernel booting (and panicking) Jan 07 14:29:30 Hmmm. Jan 07 14:30:06 maginc inside ;-) Jan 07 14:30:09 magic Jan 07 14:33:46 * mwester is stumped. Jan 07 14:35:43 Ok. I fat-fingered the console= part of the command line that ericben provided -- when I fixed that, it booted and I saw output! Wonderful. Now, when I recall my command line history, and go back to one of the original commands to boot up the minimal image -- now it boots. (!) Jan 07 14:36:25 haha Jan 07 14:37:09 I don't understand everything but that's cool that works :-D Jan 07 14:38:20 Gaah! I ran the script I wrote last evening to start qemu-system-arm. Now it works. Jan 07 14:39:39 Time to reboot - this is just too weird. Thanks for the help! :) Jan 07 14:49:04 hi kergoth Jan 07 14:49:18 hey Jan 07 14:52:42 gm Jan 07 14:58:59 ericben: also, for t-shirts, who is paying upfront? Jan 07 15:01:32 So Jan 07 15:01:38 SHR blacklisted libnl, why? Jan 07 15:01:47 libnl2 Jan 07 15:01:48 I see minimal+console-image failing w/ a libnl problem Jan 07 15:02:20 er, libnl2 failing Jan 07 15:02:37 | route/pktloc.c:39:27: fatal error: pktloc_syntax.h: No such file or directory Jan 07 15:03:03 strange that is created by bison iirc Jan 07 15:03:11 from pktloc_syntax.y Jan 07 15:03:26 that's why I had to upgrade bison-native before adding libnl2-2.0 Jan 07 15:04:18 looks like bison-native is missing in libnl2 DEPENDS, can you check after building bison-native? Jan 07 15:07:06 is built Jan 07 15:07:08 it's a race Jan 07 15:07:18 hm? Jan 07 15:07:21 (that comes from doing a make -j8) Jan 07 15:07:35 hm hm Jan 07 15:07:58 * JaMa didn't notice it with -j3 Jan 07 15:08:10 or not listed? Jan 07 15:08:10 sec Jan 07 15:08:38 http://pastebin.com/e3rwyg9H is first to last refs of pktloc Jan 07 15:08:39 but please add that bison-native, or I'll do it in few days, because I'm leaving now.. Jan 07 15:08:44 enjoy weekend Jan 07 15:08:57 k, later Jan 07 15:09:41 hmm route/pktloc_syntax.c is generated not route/pktloc_syntax.h, strange Jan 07 15:10:12 bye jama Jan 07 15:10:21 i'll see if i can dig up a built log later Jan 07 15:12:11 likewise: a generous sponsor maybe ;-) Jan 07 15:12:58 appart from problems to handle the thing what OE's t-shirts brings to the project ? Jan 07 15:13:29 fewer shirtless developers, presumably Jan 07 15:13:33 ericben: They bring fresh clothing to developers Jan 07 15:13:37 pb__ beat me to it Jan 07 15:13:47 :-) Jan 07 15:13:58 :) Jan 07 15:16:49 Is that (shirtlessness) a serious problem at these sorts of events, then? Jan 07 15:17:11 * XorA really hopes there is never shirtlessness at FOSDEM Jan 07 15:17:27 * mwester only gets to IT-oriented shows, and finds shirts are generally commonplace there. Jan 07 15:20:43 mwester: not generally a serious problem, but that might be only due to the ready supply of t-shirts. Jan 07 15:20:58 who knows what would happen otherwise Jan 07 15:21:04 Years ago, when I worked for a minicomputer company, we got a lot of mileage out of a simple give-away -- it was a bright yellow sticker shaped to fit around and highlight the "reset" button on of our competitor's less-reliable systems. :D Jan 07 15:21:13 heh Jan 07 15:21:51 pb are you going to fosdem again? Jan 07 15:22:07 not this year, no Jan 07 15:22:18 okay Jan 07 15:22:21 And another year we gave out a bright red keycap that fit the vt100 keyboard, labeled "panic". Jan 07 15:22:22 linux conferences need sponsored t-shirts and right guard Jan 07 15:22:26 pb__, do you have a house to live in yet? Jan 07 15:22:27 I am going not too Jan 07 15:22:27 Beats t-shirts. Jan 07 15:22:32 I will be at cebit Jan 07 15:22:33 unfortuneately we miss the second so far Jan 07 15:22:40 Crofton|work: nearly Jan 07 15:22:44 * mwester likes XorA's idea! Jan 07 15:22:52 woglinde, can you get some info on the web site about cebit Jan 07 15:23:04 crofton poke robert Jan 07 15:23:08 k Jan 07 15:23:13 he organize it at well Jan 07 15:23:20 I wasn't sure if he or mickeyl was organizing that Jan 07 15:23:22 ok Jan 07 15:23:36 thanks for the info Jan 07 15:23:39 mickeyl only said he will coming manning the booth Jan 07 15:23:56 ok Jan 07 15:24:02 damn power source Jan 07 15:24:11 cuts down after 3 hours running Jan 07 15:24:27 seems a missdesign in heat Jan 07 15:24:28 you have to feed the hamster more. Jan 07 15:36:04 jo mickeyl Jan 07 15:36:51 hi Jan 07 15:45:11 03Koen Kooi  07org.openembedded.dev * r74d03b8893 10openembedded.git/recipes/wpa-supplicant/ (13 files in 2 dirs): Jan 07 15:45:11 wpa-supplicant 0.7.3: import from yocto Jan 07 15:45:11 Fixed LDFLAGS issue after import Jan 07 15:45:11 Signed-off-by: Koen Kooi Jan 07 15:45:12 03Koen Kooi  07org.openembedded.dev * rd3fe1f74b5 10openembedded.git/conf/distro/include/angstrom-2010-preferred-versions.inc: Jan 07 15:45:12 angstrom next: move to wpa-supplicant 0.7 Jan 07 15:45:13 Signed-off-by: Koen Kooi Jan 07 15:47:29 03Chris Larson  07master * r96ee684001 10bitbake.git/lib/bb/cache.py: Jan 07 15:47:30 cache: don't expand variables for skipped recipes Jan 07 15:47:30 Errors can result from these expansions, but for skipped recipes, we Jan 07 15:47:30 shouldn't care about those failures. This fixes the same issue which Jan 07 15:47:30 Richard Purdie fixed in poky, commit 847b717. Jan 07 15:47:30 Signed-off-by: Chris Larson Jan 07 16:14:10 khem, ping Jan 07 16:34:45 kergoth: So good thing / annoying thing about master is it's making run time multiple provider issues more visible now Jan 07 16:37:49 (ie bug (which i was looking at to fixup qemu*) gives multiple runtime providers for xserver-kdrive, which it doesn't try and build, for minimal+x11-image) Jan 07 16:40:21 Tartarus: hmm, good point. its probably good to hear about them, but we should probably make the error message more useful Jan 07 16:40:42 Tartarus: the underlying change was to move the actual message printing into the UI, since we were already sending it an event about it -- was redundant Jan 07 16:41:10 Tartarus: thoughts on how to improve the message? given you can't specify a runtime preference directly.. Jan 07 16:45:12 Yeah, I dunno exactly Jan 07 16:45:19 But can we do a different message easily? Jan 07 17:03:23 Warning: require_once($IP/extensions/Bad-Behavior/bb2ext.php): failed to open stream: No such file or directory in /etc/mediawiki/LocalSettings.php-wiki on line 218 Fatal error: require_once(): Failed opening required '$IP/extensions/Bad-Behavior/bb2ext.php' (include_path='/var/www/wiki:/var/www/wiki/includes:/var/www/wiki/languages:.:/usr/share/php:/usr/share/pear:/usr/share/mediawiki-extensions/include') in /etc/mediawiki/LocalSettings.php-wiki Jan 07 17:03:25 on line 218 Jan 07 17:03:34 for OE wiki Jan 07 17:04:37 Heinervdm: yeah, I just ran into that as well Jan 07 17:05:03 ok, i just wanted to mention it Jan 07 17:19:52 03Tom Rini  07master * r3da481577f 10openembedded.git/recipes/help2man/ (help2man_1.36.4.bb help2man_1.38.2.bb): Jan 07 17:19:52 help2man: Mangle for /usr/bin/env perl Jan 07 17:19:52 If help2man won't work when part of the program that it is generating Jan 07 17:19:52 pages for uses one version of perl and it's using another, so we need Jan 07 17:19:52 to use /usr/bin/env perl to ensure a match. Jan 07 17:19:53 Signed-off-by: Tom Rini Jan 07 17:24:22 anyone here using pyqt? It appears to want to pull in qt-x11, I'm interested in using it with qt-embedded Jan 07 17:44:48 mwester: there are scripts for qemu booting in contrib/qemu/run-qemu.sh Jan 07 17:45:08 mwester: you say run-qemu.sh Jan 07 17:45:12 hey khem Jan 07 17:45:36 I'm going to post these shortly, but (a) minimal is doing both bluez3 and bluez4 since it has bluez-libs for bluez-libs rather than bluez4 Jan 07 17:45:42 mwester: ofcourse you need to adjust few things in the script to use your paths and images Jan 07 17:46:05 Tartarus: yes we should get rid of bluez3 Jan 07 17:46:08 and we need to switch qemu to using xserver-xorg & co over xserver-kdrive since there's a build time race when both happen and the rest of the X11R7.5 stuff won't work w/ the older kdrive Jan 07 17:46:22 * Tartarus is building all of the x11-images now Jan 07 17:46:25 Tartarus: Agreed Jan 07 17:46:51 Tartarus: you have my Ack if you do Jan 07 17:48:06 1) Fix libtool issue in binutils 2.21 2) Fix gcc 4.5 for SH Jan 07 17:58:56 I've heard of people using bluez3 becuase it does things the later one can't do Jan 07 17:59:05 rare, but I have heard of it Jan 07 17:59:15 i see Jan 07 17:59:25 still, we need to pick one or the other, not mix :) Jan 07 17:59:31 might wnt to check with sakoman and see if he has of it Jan 07 17:59:36 since the overo has working bt hw Jan 07 17:59:48 that is why blacklisting exists Jan 07 18:00:13 distro decision Jan 07 18:00:25 minimal needs to make a decision Jan 07 18:00:54 yes Jan 07 18:00:57 it did Jan 07 18:01:03 just not completely, heh, by accident Jan 07 18:01:36 this sort of thing is why I don't recommend people use minimal Jan 07 18:01:49 bluez4 works fine on overo (as did bluez3) Jan 07 18:02:01 just not at the same time :) Jan 07 18:02:43 so removing one is not the answer Jan 07 18:03:37 at least not until bluez3 is dead (which it's not) Jan 07 18:03:47 no Jan 07 18:04:23 03Melissa Watkins  07org.openembedded.dev * rb330a52353 10openembedded.git/recipes/ti/ (4 files): Jan 07 18:04:23 ti-pru-sw: Update SRC_REV to 23 Jan 07 18:04:23 For the ti-pru-sw-examples, ti-pru-sw-app-loader, ti-pru-sw-edma-driver, Jan 07 18:04:23 ti-pru-sw-edma-library: Jan 07 18:04:23 * Update SRC_REV to 23 Jan 07 18:04:23 Signed-off-by: Melissa Watkins Jan 07 18:04:24 Signed-off-by: Koen Kooi Jan 07 18:04:50 right, not at the same time :-) Jan 07 18:11:55 yes, I know the wiki is broken (in various ways)..its being "fixed" so we can eliminate all the wikispam we have been getting. Jan 07 18:12:23 Hi. Do you know if the phonon-vlc recipe discussed on the mailing list some months ago (http://www.mail-archive.com/openembedded-devel@lists.openembedded.org/msg13581.html) hit the git repository? Jan 07 18:12:44 the nature of the way it works makes it so that you can't "test" things without it breaking *immediately* Jan 07 19:49:27 03Khem Raj  07master * r35cea7a0d4 10openembedded.git/conf/distro/slugos.conf: Jan 07 19:49:27 slugos: Dont install ldconfig Jan 07 19:49:27 Signed-off-by: Khem Raj Jan 07 19:49:27 Acked-by: Leon Woestenberg Jan 07 19:49:27 Acked-by: Frans Meulenbroeks Jan 07 19:49:40 03Khem Raj  07master * r61884ba0b9 10openembedded.git/recipes/binutils/ (binutils-2.21/libtool-2.4-update.patch binutils_2.21.bb): Jan 07 19:49:40 binutils_2.21.bb: Fix the QA staging errors for target binutils Jan 07 19:49:40 * The spurious paths were caused by a -L on commandline which pointed to Jan 07 19:49:40 build dir. So we dont use -L -liberty instead use the Jan 07 19:49:41 libiberty.a directly on commandline effects are same but .la does Jan 07 19:49:41 not have the workdir path in deplibs Jan 07 19:49:42 Signed-off-by: Khem Raj Jan 07 19:53:57 re Jan 07 20:54:45 * kergoth hacks the bitbake vim syntax highlighting Jan 07 21:05:45 mario-goulart, hi, can't you fetch and look? Jan 07 21:06:03 are you under a corporate proxy or something like that? Jan 07 21:22:54 hmm, i should send the tasks-reduction branch patches to the list for review Jan 07 21:24:33 tasks reduction? Jan 07 21:24:57 drops coreutils-native in favor of a wrapper for 'install', drops the stagemanager-native recipe in favor of putting the files in openembedded/bin/, etc Jan 07 21:25:07 reduction in the number of tasks we run :) Jan 07 21:25:19 only a couple changes really Jan 07 22:27:52 03Yu Ke  07master * rce6071c64d 10bitbake.git/lib/bb/fetch/__init__.py: (log message trimmed) Jan 07 22:27:52 fetch: add SRC_URI checksum Jan 07 22:27:52 This patch adds per-recipe SRC_URI checksum verification. Jan 07 22:27:52 - SRC_URI format Jan 07 22:27:52 The format of SRC_URI checksum follow OE definition: Jan 07 22:27:53 1. SRC_URI has single src Jan 07 22:27:53 SRC_URI = "http://some.domain/file.tar.gz" Jan 07 22:27:54 03Richard Purdie  07master * r4318da793a 10bitbake.git/lib/bb/codeparser.py: (log message trimmed) Jan 07 22:27:55 codeparser: add module for parsing shell/python Jan 07 22:27:55 This commit is derived from Chris Larson's checksum work, turned into a Jan 07 22:27:56 standalone piece of code for parsing python and shell functions. Jan 07 22:27:56 The deindent code has been replaced with code to work around indentation Jan 07 22:27:57 for speed. The original NodeVisitor in the ast was replaced with a faster Jan 07 22:27:57 class walk call. Jan 07 22:27:58 03Richard Purdie  07master * rdab657bf8a 10bitbake.git/lib/bb/data_smart.py: Jan 07 22:27:58 data_smart: track variable references Jan 07 22:28:10 data: Add emit_func and generate_dependencies Jan 07 22:28:10 These functions allow generation of dependency data between functions and Jan 07 22:28:11 variables allowing moves to be made towards generating checksums and allowing Jan 07 22:28:11 use of the dependency information in other parts of bitbake. Jan 07 22:28:12 Signed-off-by: Richard Purdie Jan 07 22:28:12 03Richard Purdie  07master * r1d318acb06 10bitbake.git/lib/bb/ (runqueue.py siggen.py): Jan 07 22:28:13 runqueue: pass task hash information to tasks Jan 07 22:28:14 Signed-off-by: Richard Purdie Jan 07 22:28:14 Signed-off-by: Chris Larson Jan 07 22:28:15 03Richard Purdie  07master * rb7b63f6ba8 10bitbake.git/lib/bb/ (build.py runqueue.py ui/knotty.py): Jan 07 22:28:15 Add support for 'noexec' tasks Jan 07 22:28:16 Signed-off-by: Richard Purdie Jan 07 22:28:16 03Kevin Tian  07master * r42b08455b9 10bitbake.git/lib/bb/runqueue.py: (log message trimmed) Jan 07 22:28:17 runqueue: use correct task ID when checking validity of setscene tasks Jan 07 22:28:17 sstate hash validation is done at initialization of RunQueueExecuteScenequeue. Jan 07 22:28:18 However the index of 'valid' list returned from the validation doesn't Jan 07 22:28:18 correspond to setscene task ID. It's just an intermediate namespace between Jan 07 22:28:19 runqueue and sstate hash func. Use it as setscene task ID fully mess the flow. Jan 07 22:28:20 Previously this doesn't cause trouble because all setscene tasks are passed. Commit Jan 07 22:28:20 03Richard Purdie  07master * r9ac2ce32b9 10bitbake.git/lib/bb/runqueue.py: Jan 07 22:28:32 03Richard Purdie  07master * rbbfaf3e1d8 10bitbake.git/lib/bb/parse/ (ast.py parse_py/BBHandler.py): Jan 07 22:28:32 parse: save python functions into the metadata Jan 07 22:28:33 Signed-off-by: Richard Purdie Jan 07 22:32:31 03Chris Larson  07master * r7af84fb26f 10openembedded.git/ (MAINTAINERS recipes/editline/editline_20100424-3.0.bb): Jan 07 22:32:32 editline: add recipe, version 20100424-3.0 Jan 07 22:32:32 This is an autotooled/libtoolized port of the NetBSD libedit. Jan 07 22:32:32 Signed-off-by: Chris Larson Jan 07 22:48:02 03Paul Menzel  07master * rce1f572b4e 10openembedded.git/contrib/testing/testbuilder/README: Jan 07 22:48:02 testbuilder/README: correct grammar and markup in Markdown [1] Jan 07 22:48:02 [1] http://daringfireball.net/projects/markdown/ Jan 07 22:48:02 Signed-off-by: Paul Menzel Jan 07 22:48:02 CC: Yuri Bushmelev Jan 07 22:48:03 Signed-off-by: Khem Raj Jan 07 22:48:06 03Bernhard Reutner-Fischer  07master * raf261ba6ea 10openembedded.git/recipes/xz/ (xz.inc xz_4.999.9beta.bb xz_5.0.0.bb): Jan 07 22:48:06 xz: bump version Jan 07 22:48:06 Signed-off-by: Bernhard Reutner-Fischer Jan 07 22:48:06 Signed-off-by: Khem Raj Jan 07 22:48:08 03Bernhard Reutner-Fischer  07master * rc65ccdbe7c 10openembedded.git/recipes/uclibc/uclibc.inc: Jan 07 22:48:08 uclibc: shared-flat: put crtreloc.o into -dev package Jan 07 22:48:08 Signed-off-by: Bernhard Reutner-Fischer Jan 07 22:48:08 Signed-off-by: Khem Raj Jan 07 22:48:09 03Bernhard Reutner-Fischer  07master * r0281e0eaae 10openembedded.git/docs/usermanual/reference/class_image.xml: Jan 07 22:48:10 class_image: fix typos Jan 07 22:48:10 Signed-off-by: Bernhard Reutner-Fischer Jan 07 22:48:10 Acked-by: Roman I Khimov Jan 07 22:48:28 03Ben Gamari  07master * rd470cfde5b 10openembedded.git/recipes/boost/ (3 files in 2 dirs): Jan 07 22:48:28 boost: Initial commit of 1.45.0 Jan 07 22:48:28 Signed-off-by: Khem Raj Jan 07 22:48:31 03Denis Dydychkin  07master * rd81a0f0623 10openembedded.git/classes/qmake_base.bbclass: Jan 07 22:48:31 qmake_base.bbclass patch to lookup for *.pro within ${S} instead of ${B} Jan 07 22:48:31 qmake_base.bbclass from release-2010.12 has ability to lookup for QMAKE_PROFILES if none specified by recipe. Unfortunately it does that in ${B} directory instead of ${S}, thus it does not work if they differ. Jan 07 22:48:31 Here is suggested patch: Jan 07 22:48:31 Signed-off-by: Khem Raj Jan 07 22:48:33 03Bernhard Reutner-Fischer  07master * rffdf56431d 10openembedded.git/classes/image.bbclass: Jan 07 22:48:34 image.bbclass: rm symlink destination via ln Jan 07 22:48:34 Rather than rm'ing the destination-file manually, do ln -f Jan 07 22:48:34 Signed-off-by: Bernhard Reutner-Fischer Jan 07 22:48:35 Acked-by: Roman I Khimov Jan 07 22:48:35 Signed-off-by: Khem Raj Jan 07 22:48:35 03Bernhard Reutner-Fischer  07master * r649732c37e 10openembedded.git/recipes/glib-2.0/ (glib-2.0-2.26.1/configure-ipv6.patch glib-2.0_2.26.1.bb): Jan 07 22:48:36 glib-2.0: fix compilation for !IPv6 Jan 07 22:48:36 Signed-off-by: Bernhard Reutner-Fischer Jan 07 22:48:37 Signed-off-by: Khem Raj Jan 07 22:48:41 03Ben Gamari  07master * r412ab1b823 10openembedded.git/recipes/python/ (34 files in 3 dirs): Jan 07 22:48:41 python: Initial commit of python 2.7.1 Jan 07 22:48:41 Signed-off-by: Khem Raj Jan 07 23:54:36 Any examples how in OE add new kernel driver? Jan 07 23:54:57 Sould I just copy files under drivers directory or something more complicated required? Jan 08 00:14:53 03Chris Larson  07master * rbf4dfa0099 10openembedded.git/recipes/psmisc/ (psmisc-22.8/libintl-link.patch psmisc_22.8.bb): Jan 08 00:14:53 psmisc: add version 22.8 Jan 08 00:14:53 Signed-off-by: Chris Larson **** ENDING LOGGING AT Sat Jan 08 02:59:57 2011