**** BEGIN LOGGING AT Sun Mar 11 02:59:56 2007 Mar 11 03:50:15 03nail 07org.oe.dev * r32536525... 10/ (6 files in 5 dirs): uboot-utils: Update to include configs specific to fic-gta01 and turbostation Mar 11 04:24:14 RP: ping? Mar 11 04:26:11 RP: Regarding leds on the Turbostation. The stuff appears to be very simple. The PIC is connected to the second UART (I believe 9600n8). Writing single bytes switches leds on/off/blink, buzzes short/long, reboots/powerdowns. Mar 11 05:03:20 JoseJX: Does your buddy the coder know about the bounty? Mar 11 05:51:22 someone be a pal and push 1967 for me please? Mar 11 05:51:27 !oebug 1967 Mar 11 05:51:29 * * Bug 1967, Status: NEW, Created: 2007-03-10 21:48 Mar 11 05:51:30 * * r.farina(AT)adelphia.net: Raw GlueAP Mar 11 05:51:31 * * http://bugs.openembedded.org/show_bug.cgi?id=1967 Mar 11 05:57:52 nobody available to be my monotone pusher? Mar 11 06:02:16 Zero_Chaos: well, it's 6am sunday in europe, so they are probably sleeping in :) Mar 11 06:02:34 hvontres|home: it happens ;-) Mar 11 06:07:47 hvontres|home: No, I wasn't aware there was one Mar 11 06:14:09 JoseJX: : check here: http://wiki.openzaurus.org/FAQ#SD_Card_related_questions Mar 11 06:15:46 I'll forward the link to him Mar 11 06:17:21 Done :p Mar 11 06:20:11 JoseJX: heh... nothing like some cold hard cash for a motivator :) Mar 11 06:20:15 heh Mar 11 06:20:35 * JoseJX is motivated by working hardware :p Mar 11 06:22:37 Anyway, good night Mar 11 07:19:28 Could someone pls help me ? I have openembedded installed, building works. Now I would like to edit code of the packages and run make configure (kernel mainly). What is the standard procedure for doing so ? Mar 11 08:03:39 morning all Mar 11 08:36:00 good morning Mar 11 08:40:00 morning all Mar 11 08:47:07 morn :) Mar 11 08:55:02 anyone knows which libsdl should I be using when building opie (either in dev/angstrom or oz354x/openzaurus)? It seems that the error I get with kdepimpi ( http://pastebin.ca/389751 ) is because I'm building against libsdl-x11 Mar 11 08:56:21 at least from what I've googled Mar 11 08:58:44 Hello, I'm trying to 'bitbake gcc-cross' on newly updated org.oe.dev snapshot, but it seems that bitbake first wants to compile glibc for armv5te, which in turn needs gcc... Mar 11 09:00:01 there ia gcc-initial for that :) Mar 11 09:01:11 Could someone pls help me ? I have openembedded installed, building works. Now I would like to edit code of the packages and run make configure (kernel mainly). What is the standard procedure for doing so ? Mar 11 09:01:48 gremlin[it]: hmm, so I can't bitbake gcc-cross directly? Mar 11 09:03:20 demo_: the thing you will probably need is cross compiler from openembedded (well, and few more things, like headers, etc.:) For kernel it is quite easy, you just need to set CROSS_COMPILE variable properly to crosscompiler in OE Mar 11 09:03:59 mmp: I am able to build the kernel "bitbake virtual/kernel" Mar 11 09:04:21 mmp: I just would like to see some output of configure and make Mar 11 09:05:08 mmp: or should I do it directly, without bitbake ? Mar 11 09:05:23 demo_: I don't know what exactly do you plan:) Mar 11 09:05:47 depends on the ratio of modifying packages and building them using OE:) Mar 11 09:05:49 mmp: grab some cpufreq stuff from hh.org to oz kernel Mar 11 09:07:29 demo_:you can build your kernel outside OE as well... Mar 11 09:07:47 if you are going to do some intensive debugging, outside OE is more straightforward. Mar 11 09:08:22 mmp: I see, CROSS_COMPILE in Makefile ? Mar 11 09:08:36 demo_: but if you need just quick cut&paste of some code and then compile kernel, you can also force OE to do things you want Mar 11 09:09:00 demo_: yes, you need to set it to prefix of your cross compile tools... Mar 11 09:09:41 if I modify stuff in work dir than I can do "bittake -c compile" and it will understand ? Mar 11 09:10:09 demo_: it should Mar 11 09:10:20 :) Mar 11 09:10:32 mmp: what about the make output ? Mar 11 09:10:38 can I see that ? Mar 11 09:10:55 it will output you the problems if they happen and make will return non-zero retval Mar 11 09:11:22 ok, thanx, i'll try that :) Mar 11 09:11:41 but I strongly believe there is some way to force bitbake to do full output:) Mar 11 09:14:01 mmp: great too, very badly documented :) Mar 11 09:14:09 s/too/tool Mar 11 09:15:51 :-) Mar 11 09:16:18 gremlin[it]: I meant, should not glibc depend on gcc-initial then? Mar 11 09:18:42 should be ... Mar 11 09:19:34 demo_ : add -DDDD and you will see all the debug info you want Mar 11 09:20:43 mmp : In order to build a cross compiler you need to do what is called a "bootstraping" Mar 11 09:20:43 Ifaistos: thx :) Mar 11 09:21:22 mmp : first you build a simple one using the tools of the host system Mar 11 09:21:26 Ifaistos: yes, but should not that be done 'automagically' when I invoke just bitbake gcc-cross .. Mar 11 09:21:47 mmp : yes Mar 11 09:22:16 I'll try it one more time... Mar 11 09:22:20 to build gcc-cross you need to build gcc-initial, glibc-initial, glibc-intermediate and then cross Mar 11 09:22:45 this is done automagicaly Mar 11 09:22:47 maybe just bad constellation of stars, moon, sun, some cosmic radiation, ... :) Mar 11 09:23:39 mmp : Well OE has a *very* steep learning curve and documentation is not that helpfull Mar 11 09:24:18 mmp : so get a good/fast machine and prepare your self for a number of rebuilds until you get it right Mar 11 09:24:20 Ifaistos: :-) I've been doing this same with OE earlier and it worked, so I wonder what went wrong... Mar 11 09:24:32 the setup should be OK I guess.. Mar 11 09:24:46 maybe some remains from older builds somewhere... Mar 11 09:24:47 mmp : hmmm things tend to break in .dev Mar 11 09:25:06 mmp : when i dought.... rm -rf tmp Mar 11 09:25:34 yes, I did that, I was more thinking of some forgotten env variables which could break something... Mar 11 09:25:47 mmp : it takes sometime to rebuild the system but it least you make sure your problem is not some "residual evil" from the previus ones Mar 11 09:26:06 :) Mar 11 09:26:40 Ifaistos: actually I need just to get crosscompiler, building rest of distro will come later:) Mar 11 09:27:29 mmp : this can be the hardest part of all.... Mar 11 09:27:50 gcc/glibc combination can kill you Mar 11 09:27:56 :) Mar 11 09:29:54 but definitely there is something wrong - from empty tmp/ I'm getting build sequence like: quilt-native gnu-config-native, m4-native, autoconf-native, automake-native, libtool-native, ipkg-utils-native, pax-utils-native, linux-libc-headers, glibc-2.5 Mar 11 09:30:08 nop this is the correct one Mar 11 09:30:15 for gcc-cross? Mar 11 09:30:19 yes Mar 11 09:30:28 glibc-2.5 seems to need crosscompiler already Mar 11 09:30:46 | checking for arm-angstrom-linux-gnueabi-gcc... ccache arm-angstrom-linux-gnueabi-gcc -march=armv5te -mtune=xscale Mar 11 09:30:46 | checking for suffix of object files... configure: error: cannot compute suffix of object files: cannot compile Mar 11 09:31:16 JustinP: ping Mar 11 09:31:16 you need glibc-interemediate first Mar 11 09:31:49 Ifaistos: so bitbake gcc-cross doesn't do all neccessary bootstrapping things itself? Mar 11 09:32:31 mmp i think it does but to tell you the truth never did a bitbake gcc-cross Mar 11 09:33:13 the corect order should be glibc-initial, gcc-intermediate, glibc-intermediate, gcc-cross Mar 11 09:33:55 nop mistake Mar 11 09:34:30 the corect order should be glibc-initial, gcc-cross-initial glibc-intermediate, gcc-cross Mar 11 09:35:28 Ifaistos: going to try, thanks a lot:) Mar 11 09:35:34 I guess I should file a bug in OE bugzilla... Mar 11 09:35:37 and for arm i think it builds gcc-cross using qemu i believe Mar 11 09:36:09 eek Mar 11 09:36:12 try to build one of the distro's in OE before Mar 11 09:37:15 Ifaistos: do I understand it correctly, that 'gcc-cross' is crosscompiler? Runnable on x86(_64 in this case), producing arm binaries? Mar 11 09:37:32 Why would it be compiled using qemu? Mar 11 09:37:44 cyrilRomain: what gcc are you building and on what host ? Mar 11 09:38:02 mmp: for locale generation if i remember correctly Mar 11 09:38:27 Ifaistos: ahaa, then yes, you're right. Unless locale generation is turned off:) Mar 11 09:38:42 Ifaistos: gcc4.1.1 for both armv5te and i686 Mar 11 09:39:02 Ifaistos: my host is i686 Mar 11 09:39:40 cyrilRomain: i think i found a bug when building gcc 4.1.1 (not cross) Mar 11 09:40:16 Ifaistos: why kind of bug ? Mar 11 09:40:23 s/why/what/ Mar 11 09:41:08 cyrilRomain: gcc needs to build a number of files that are executed by the build machine in order to get the correct configs/includes Mar 11 09:41:35 cyrilRomain: but it seems that in OE is using the targets site files Mar 11 09:42:04 cyrilRomain: if target and host have the same arch/ and similar lenght variables it works Mar 11 09:42:25 cyrilRomain: but if you try to build an x86_64 for example it fails Mar 11 09:43:24 Ifaistos: I see.. there are some mail on the mailing list about x86_64 build iirc. None of them about that ? Mar 11 09:43:57 cyrilRomain: In both cases its wrong, simply on 64bit is shows up Mar 11 09:44:32 !oebug 1951 Mar 11 09:44:34 * * Bug 1951, Status: NEW, Created: 2007-03-07 15:45 Mar 11 09:44:34 * * nail(AT)nslu2-linux.org: gcc fails to build on x86_64 Mar 11 09:44:35 * * http://bugs.openembedded.org/show_bug.cgi?id=1951 Mar 11 09:45:42 Ifaistos: yes its wrong, as wrong as cross compile badness where target build looks into host includes Mar 11 09:54:28 Ifaistos: maybe a test should be added in the qa tests to prevent such bug. Mar 11 10:01:56 cyrilRomain: gcc when host!=build make aut-build.h file by running configure again, the problem is it using the cached values from the site files which contain the target's values Mar 11 10:02:12 auto-build.h Mar 11 10:02:54 cyrilRomain: that's the bug 1951 Mar 11 10:07:40 hey psokolovsky_ Mar 11 10:09:41 psokolovsky_: I did the change in mtnpatch.py to output command in the right order. But I want to see with JustinP if my change with the '-R' option is right (I assumes -R would be to unapply the patch, but I think JustinP wanted it for another purpose) Mar 11 10:10:35 psokolovsky_: I don't use 'patch -R' myself so it can be a bad assumption. Mar 11 10:20:57 cyrilRomain, nice. I really didn't use mtnpatch.py myself (nor the author of it ;-) ), so I don't have direct suggestions ;-) Mar 11 10:21:02 Hi! Mar 11 10:35:39 psokolovsky_: Hi Mar 11 10:35:46 Hi! Mar 11 11:44:39 Ifaistos: seems that cross compiler is built now; thanks:) Mar 11 11:46:25 mmp: good Mar 11 13:01:09 *sigh* Mar 11 13:01:31 why do websites think that having a .br ip means you can read portugese? Mar 11 13:02:29 koen|recife: I'm sure you can read it Mar 11 13:02:41 ok, you have a point :) Mar 11 13:02:52 zecke: and why do so many people know you? Mar 11 13:03:11 nobody knows me, I'm white trash Mar 11 13:03:25 "I don't know any Holger Freyther" "zecke" "yes, I know zecke!!!" Mar 11 13:03:59 daniel from opens(use)(ync) is trying to look like wookey as well ;) Mar 11 13:04:08 hehe Mar 11 13:04:29 koen|recife: congratulate him on the release of OpenSync :) Mar 11 13:06:12 QOTSA seems a bit out of place here, though Mar 11 13:06:54 RP: is ross on irc? Mar 11 13:09:23 koen|recife: this is the reason it is impossible to change the nickname... Mar 11 13:12:36 morning Mar 11 13:12:48 hrw: good morning Mar 11 13:13:04 koen|recife: how goes .br? Mar 11 13:13:34 sunny Mar 11 13:14:21 koen|recife: If you didn't see it earlier, Boi Preto Grill is an awesome 'all you can eat' grill place. ;) Mar 11 13:14:29 too bad the beach has all those "don't feed the sharks" signs Mar 11 13:14:48 CM: I'm moving to porto de galinhas today, so I'll try it next week :) Mar 11 13:15:16 Honestly I've only eaten at the Boi Preto in Sao Paulo, but it was the best meat I've ever had Mar 11 13:15:47 They have 3 resturants, S.P, Salvador and Recife :) Mar 11 13:16:51 the buffet pay-per-kilo thing yesterday was quite nice as well Mar 11 13:17:15 pay-per-kilo? how it works? Mar 11 13:17:33 Their pizzas are good too, and always with roasted garlic to sprinkle on Mar 11 13:17:52 you scoop up a pile of food on your plate and they note the weight on your receipt Mar 11 13:18:03 ik was 26 BRL/kilo Mar 11 13:18:15 meat, cake, fruit, all the same Mar 11 13:18:30 koen|recife: fruit? come on ;) Mar 11 13:18:37 ah Mar 11 13:23:32 koen|recife have u a sec for a devel issue ? Mar 11 13:23:52 sure Mar 11 13:24:47 ~change 26 brl to pln Mar 11 13:24:55 the last gcc in oe (for h3600 oabi) generate same target binary also if host differ ??? (i386 versusn x86_64) ??? Mar 11 13:24:55 26.00 Real (BRL) makes 36.7246 Zloty (PLN) (from http://www.xe.com/) Mar 11 13:25:23 03Marek 07org.oe.dev * r214ab3b1... 10/ (4 files in 2 dirs): hexatrolic, labirynth, zsubhunt: create directory for .desktop file - close #1962, #1963, #1965 Mar 11 13:25:26 03Cyril 07org.oe.dev * rd2b8acc4... 10/ (4 files in 3 dirs): mrxvt: update to 0.5.2 - close #1964 Mar 11 13:25:30 i remember an issue of this kind for "old" 3.x gcc Mar 11 13:27:10 gremlin[it]: the gcc people claim they solved it in gcc4 Mar 11 13:31:39 zecke: django is amazing =) I'm doing some tutorials to discover what can be done and how. Mar 11 13:32:20 03hrw 07org.oe.dev * re860ab75... 10/ (1 packages/rglueap packages/rglueap/rglueap_0.1.bb): rglueap: added 0.1 version - close #1967 (initial recipe written by Rick Farina , cleaned by me) Mar 11 13:32:55 is cause when i try to build angstrom-x11-image i obtain an image bigger than 16MByte ... psokolovsky less than 16MByte ... out setup should be the same (we will check shortly) so maybe the only difference is the host ... Mar 11 13:34:52 gremlin[it]: maybe one of you (or both) have local diff in the metadata? Mar 11 13:35:43 could be ... we will check ... :) Mar 11 13:35:55 thanks, koen, how is brasil ?!?!? Mar 11 13:36:09 hot Mar 11 13:36:24 "armv4" - we do not use it iirc... Mar 11 13:36:59 arm in OE == armv4 iirc Mar 11 13:37:00 koen|recife: ? Mar 11 13:37:03 hi CoreDump|home Mar 11 13:37:04 hi Mar 11 13:37:04 yes Mar 11 13:37:16 koen|recife: then conf/machine req cleanup Mar 11 13:37:18 I'm not sure if gcc even support ARMv3 Mar 11 13:38:03 have u try the meat cooked in spit :P Mar 11 13:38:33 hrw shut'up :P ... (StrongArm) armv4 is my business :) :) :) Mar 11 13:40:13 gremlin[it]: armv4==arm in OE Mar 11 13:40:47 plain armv4 or armv4t ? Mar 11 13:40:57 plain Mar 11 13:41:07 03hrw 07org.oe.dev * r3a51adfd... 10/ (4 files in 2 dirs): ziq, ztappy, win4: create directory for .desktop file - close #1961 Mar 11 13:41:11 03woglinde 07org.oe.dev * rf64f76da... 10/ (4 files in 3 dirs): simpad config cleanup - close #1950 Mar 11 13:41:39 could be nice to make it expicitly armv4 as well, but I have no strong opinions on that Mar 11 13:41:50 write rfc? :D Mar 11 13:42:09 which noone will read? Mar 11 13:42:22 ... Mar 11 13:49:34 I want to remove task-qpe from task-sdk Mar 11 13:50:32 but sorry, i have some confusion ... so what's the issue about oabi vs eabi ??? Mar 11 13:51:45 03rwhitby 07org.oe.dev * r2e364467... 10/ (1 packages/tasks/task-slugos.bb): task-slugos: Changed a lot of packages in the rootfs from RDEPENDS to RRECOMMENDS Mar 11 13:51:53 03rwhitby 07org.oe.dev * rf9f78fea... 10/ (1 packages/tasks/task-mokogateway.bb): task-mokogateway: Added iptables stuff for cdc-ether masquerading Mar 11 13:54:06 gremlin[it]: oabi runs with any kernel. eabi with 2.6.16+ Mar 11 13:54:17 gremlin[it]: and eabi need arm4*t* Mar 11 13:59:09 03hrw 07org.oe.dev * r4ed30587... 10/ (1 packages/tasks/task-sdk-opie.bb packages/tasks/task-sdk.bb): Mar 11 13:59:09 task-sdk/task-sdk-opie: moved OPIE sdk task to separate recipe. Mar 11 13:59:09 OPIE run only on subset of our target machines, X11 run on most. Mar 11 14:03:11 hrw ... ahh ok as i remember eabi need armv4t or better ... so i though all OE was armv4t ... :) Mar 11 14:03:32 gremlin[it]: sa1100 is armv4 Mar 11 14:03:38 it lack t Mar 11 14:04:21 i know ... that's cause i use oabi Mar 11 14:09:44 ok, cu Mar 11 14:11:01 03Bernhard 07org.oe.dev * rccbbf95d... 10/ (3 files in 3 dirs): linux: SIMpad ucb1x00 switches driver code beautification - close #1920 Mar 11 14:12:25 03pH5 07org.oe.dev * rbbeaeb48... 10/ (1 packages/avahi/avahi.inc): avahi: add missing glib-2.0 dependency Mar 11 14:29:16 morning Mar 11 14:46:13 psokolovsky_ Hi! Mar 11 14:56:16 psokolovsky, hi! Mar 11 14:56:47 gremlin[it], Hi! Mar 11 14:57:57 psokolovsky can u send me your local.conf used to build h3600 oabi image and the result of 'mtn diff' on org.oe.dev ??? so meybe we can figure out why i build an image bigger than yours ... Mar 11 15:01:28 psokolovsky_ can u send me your local.conf used to build h3600 oabi image and the result of 'mtn diff' on org.oe.dev ??? so meybe we can figure out why i build an image bigger than yours ... Mar 11 15:01:35 gremlin[it], ok, we can do that Mar 11 15:02:12 gremlin[it], but even more interesting is to compare list of installed packages, and potentially, their sizes Mar 11 15:03:12 sure, first i would check if there is some differeces i our starting configuration ... Mar 11 15:03:30 gremlin[it], so, want my /usr/lib/ipkg/status ? Mar 11 15:03:47 ok send it too Mar 11 15:04:13 gremlin[it], you'd better start from something which directly related to diffs in sizes ;-) Mar 11 15:06:22 sure, i would check in order, 1)starting config files, 2) list of packages build 3) size of some executable and library ... Mar 11 15:07:11 gremlin[it], suggested reverse order, but ymmv ;-) Mar 11 15:08:06 gremlin[it], sent Mar 11 15:08:32 http://ars.userfriendly.org/cartoons/?id=20070311 Mar 11 15:08:42 thnaks .. Mar 11 15:20:36 Why does bitbake think 0.25.5 is greater than 0.26.3? Mar 11 15:21:02 looking at neon package bitbake versioning Mar 11 15:23:09 hillct: it doesn't. Either 0.26.3 has a default_pref of -1 or the version is hardwired somewhere Mar 11 15:24:51 I see the default pref. what does that mean? Mar 11 15:24:58 where can I find docs on it? Mar 11 15:25:37 does that just cause the recipe to be ignored? Mar 11 15:27:29 how should I be trating release number increments? should release number be incremented only when the recipe is updated but not the underlying package version? Mar 11 15:41:03 mr_nice: hi, hi all Mar 11 15:44:47 mr_nice: hi Mar 11 15:47:03 rwhitby: so you can do unattended reflash of the neo device? Mar 11 15:47:25 psokolovsky_ ... 15360 Kbyte !!!!!!!!!! Mar 11 15:48:21 gremlin[it], cool!! now - does it work? ;-) Mar 11 15:49:06 now i flash ... at least we have found the issue !!! Mar 11 15:49:54 gremlin[it], yes, finally ;-) Mar 11 15:50:28 i really would like bitbake will warning if it found a unknown tag/variable :S Mar 11 15:51:49 mr_nice: found bug in driver for mq200 and is solved now Mar 11 15:58:23 mrdata: interesting what was it? Mar 11 15:58:56 hi zecke Mar 11 16:04:44 zecke: hi Mar 11 16:05:09 zecke: error in gc1_reset() Mar 11 16:05:47 zecke: we need GC1 and GC2 Mar 11 16:07:45 zecke: gc2 is in mq200_data.h not defined Mar 11 16:08:18 zecke: i have changed that Mar 11 16:09:44 mrdata: so you added a gc2_reset? Mar 11 16:09:55 zecke: also enable LCD for GC2 and CRT for GC1 Mar 11 16:10:06 zecke: yes, and that was all Mar 11 16:10:19 joys of having documentation :) Mar 11 16:11:10 zecke: not realy true Mar 11 16:12:06 zecke: full debugging of mq200fb.c and find the missing parts Mar 11 16:12:56 zecke: 4 days of hard work Mar 11 16:22:33 psokolovsky_ booting with OE logo and bar ... but after 100% it crach Mar 11 16:22:36 crash Mar 11 16:23:18 gremlin[it], ;-((. crash how? Mar 11 16:23:20 mhhh at serial prompt 'h3600 login:' typing root, got "login: no valid shadow password" :( Mar 11 16:24:02 gremlin[it], as it was before, any password worked - at least for ssh Mar 11 16:24:29 >> psplash.c:76,pspalsh_draw_progress() ..... got cmd QUIT Mar 11 16:24:40 this is show on framebuffer Mar 11 16:25:04 mhh bad ... there is no way to login so :\ :( Mar 11 16:27:20 gremlin[it], any news on usb support? if there was NFS, I could join in with hacking, because I really don't want to deal with flashing ;-) Mar 11 16:28:17 psokolovsky_ hey no ... but i have a good news for me ..i whange work :) ... will work for a mediacal company that use embedded linux :) ... Mar 11 16:28:45 gremlin[it], cool, congratulations! Mar 11 16:29:06 there is no way to write on jffs2 with bootldr ??? Mar 11 16:30:04 gremlin[it], sure, there's. I just hate flashing ;-) Mar 11 16:31:08 mhh ok ... probably will be better for early testing to build an image with an already set root password ... Mar 11 16:35:02 gremlin[it], ah damn, you see, it's that problem: debug distro type has empty password, but too big. release is good size, but has unset password ... Mar 11 16:35:38 mhh ok ... now i know what to check ... Mar 11 16:37:50 mr: hi Mar 11 16:45:09 hi Mar 11 16:48:13 cyrilRomain: pong Mar 11 16:48:38 mr: aka mr_nice hi Mar 11 16:48:45 JustinP: ah, it was to ask you about the use of -R in mtnpatch.py Mar 11 16:49:10 JustinP: I sent a mail with an updated mtnpatch.py Mar 11 16:49:37 but as I don't use patch -R myself, I may have change it wrong Mar 11 16:51:20 cyrilRomain: you patched mtnpatch? Mar 11 16:51:47 JustinP: no, but I sent it by mail for peer-review Mar 11 16:52:23 JustinP: I mean if you could have a look and tell me what you think about it would be nice :) Mar 11 16:54:05 cyrilRomain: I think you sent that to an old list.... Mar 11 16:54:14 cyrilRomain: I did find it, though Mar 11 16:54:59 cyrilRomain: I would rather you'd sent a patch ;-) Mar 11 16:55:11 JustinP: ok I'll do Mar 11 16:55:19 if you want Mar 11 16:56:15 cyrilRomain: it looks ok :-) Mar 11 16:57:03 JustinP: don't you have had issue with drop order ? or maybe do you not use it anymore ? Mar 11 17:04:50 Hello everybody.. Mar 11 17:05:42 Having problem with building librsvg keeps throwing error during QA check after compiling Mar 11 17:06:18 mickeyl, how was skiing? Mar 11 17:06:50 ERROR: Error in executing: /stuff/org.openembedded.dev/packages/librsvg/librsvg_2.16.0.bb Mar 11 17:07:09 ERROR: Exception:exceptions.SystemExit Message:1 Mar 11 17:07:26 Any suggestions? Mar 11 17:15:36 mickeyl: hey Mar 11 17:17:01 we should eliminate the FOSDEM refs from the topic Mar 11 17:17:32 Crofton: or change the date Mar 11 17:17:38 :) Mar 11 17:17:50 hmmm, I need to start watching plane fares :) Mar 11 17:18:46 I can only get tickets for the first weekend in Feb atm Mar 11 17:20:35 $609 for the first weekend in Feb to BRU atm Mar 11 17:35:21 psokolovsky_: I'm unable to build opie-image in angstrom, following your OpieWithAngstrom wiki - it fails to download libqpe-opie with a strange error - http://pastebin.ca/390751 Mar 11 17:36:14 there is no opie-image for angstrom, you can choose between "angstrom-x11-image" and "angstrom-x11-image" Mar 11 17:36:53 Bernardo, that's not OE error, but CVS one. try to do CSV comman manually, and see, if it works Mar 11 17:37:14 any open hand slacker hang out here? Mar 11 17:38:54 I'm wondering if there are any bb files for exmap and exmap-console Mar 11 17:39:07 you're right, it seems it is trying to update cvs without first doing a checkout Mar 11 17:40:34 thanks mickeyl Mar 11 17:40:41 I'm already watching prices :) Mar 11 17:40:59 hey! Mar 11 17:41:02 hehe Mar 11 17:41:29 did you check poky for .bb's for exmap? Mar 11 17:41:31 How was skiing?, We leave for Colorado on Wednesday Mar 11 17:41:35 no Mar 11 17:41:36 I'm also unable to build opie-image in oz354x, but that seems to be due to dbus postinst trying to write to /etc (http://pastebin.ca/390777) Mar 11 17:41:39 I should Mar 11 17:42:32 i didn't ski, but austria was great. while the others were skiing, i worked 8h per day on a contract project while the others were on the lopes Mar 11 17:42:46 but at least we had fun together in the evenings Mar 11 17:43:03 I suspect I will be writing some days Mar 11 17:44:51 poky has a bb file for the exmap-console Mar 11 17:45:07 I'll look at adding it to OE when I have some time Mar 11 17:45:34 I really need to run it to make some measurememtns Mar 11 17:46:20 yeah Mar 11 17:46:24 me too Mar 11 17:49:08 maintainer in bbfiles is bad, correct? Mar 11 17:51:53 ya Mar 11 17:51:54 Crofton: yes; preferred is the MAINTAINERS file instead Mar 11 17:52:06 I'll fix the style guide then Mar 11 17:53:00 done Mar 11 18:28:37 OK, I made a bb file for exmap-console Mar 11 18:28:43 well, copied from poky Mar 11 18:33:27 wb mickey|dinner Mar 11 18:37:33 RP, you the poky guy? Mar 11 18:37:41 Crofton: Yes Mar 11 18:37:55 I'm copying the exmap-console bb files into OE Mar 11 18:38:03 I hope you don't mind Mar 11 18:38:27 Crofton: I don't mind at all, just say where they came from it the commit ;-) Mar 11 18:38:57 pretty sure the message says copy blah from Poky. :) Mar 11 18:39:36 Crofton: thats fine :) Mar 11 18:39:45 I really need to run exmap-console on some stuff Mar 11 18:40:03 I had been using pmap and top to guesstimate memory usage Mar 11 18:40:34 It can be quite useful if you're looking at memory usage althought admittedly, I don't do much of that :) Mar 11 18:41:12 heh Mar 11 18:45:30 Need some help figuring out how to fix librsvg so package QA doesn't error Mar 11 18:46:15 it doesn't like the fact that there is a .debug directory in a non debug package Mar 11 18:47:32 all of the librsvg packages compiled fine.. any suggestions? Mar 11 18:49:01 FILES_python-pyusrp-dbg = "${libdir}/python*/site-packages/.debug" Mar 11 18:49:09 is what I see in one bb file Mar 11 18:49:36 geekoid: Its certainly the FILES_ variables that need adjusting Mar 11 18:51:50 RP: where might I find the files variables? makefile? autoconifg? Mar 11 18:52:16 geekoid: The packages .bb file (or inherited from bitbake.conf Mar 11 18:52:19 geekoid, look in usrp/usrp-xx.bb Mar 11 18:55:22 03crofton 07org.oe.dev * rb0e8a7b1... 10/ (5 files in 3 dirs): Add exmap-console from Poky. Mar 11 18:55:26 03pfalcon 07org.oe.dev * re7740c0d... 10/ (1 packages/qpe-games/ziq_1.1ern.bb): Mar 11 18:55:26 ziq 1.1ern: Create dir for desktop file. Mar 11 18:55:26 * Based on Marek Vasut's patch at #1961. Mar 11 18:59:43 RP, good enough message? Mar 11 18:59:51 Crofton: fine, thanks Mar 11 19:01:19 03nail 07org.oe.dev * reee01f8c... 10/ (10 files in 3 dirs): linux-turbostation: Update to 2.6.20.2, repartition, remove unneccessary patches Mar 11 19:01:36 03nail 07org.oe.dev * r0c29b5c4... 10/ (1 packages/images/foonas-image.bb): foonas-image: update kernel/rootfs size Mar 11 19:03:22 mrdata: hi, verry cool!!!! big thx! I am creating a patch out of your attachment and send it to the oe.bugtracker and the slackpad svn repro. Mar 11 19:03:32 RP & Crofton: thanks found the lines in the .bb file recompiling now with fingers crossed ;) Mar 11 19:08:50 Hmm, able to fix the FILES_variable for librsvg-gtk-dgk but not for gtk-dev Mar 11 19:09:29 original line in bb is: FILES_librsvg-gtk-dev += "${libdir}/gtk-2.0/*" Mar 11 19:10:41 but there are 2 directories with debug files: lib/gtk-2.0/2.10.0/engines/.debug/ & lib/gtk-2.0/2.10.0/loaders/.debug/ Mar 11 19:11:23 So I changes bb to: FILES_librsvg-gtk-dev += "${libdir}/gtk-2.0/2.10.0/*/.debug" but it's no joy. What did I miss? Mar 11 19:23:30 mr_nice: hi, please wait with the patch, i have forgot somthing to change in mq200data.h Mar 11 19:24:12 mr_nice: the bitfields for GC21R are wrong Mar 11 19:25:00 mr_nice: i will change this now and send you then the new file with mail Mar 11 19:26:00 mrdata: ok Mar 11 19:26:11 mrdata: many thx! Mar 11 19:27:34 mrdata: It would be verry nice if you will extend the comment at the top of the files with your name and a short description of what you have changed. Mar 11 19:28:08 mrdata: everybody should know that you did the work Mar 11 19:29:09 mr_nice: the most part for the work was doing by zecke befor ;-) Mar 11 19:29:30 geekoid, not sure what you missed Mar 11 19:30:13 mr_nice: my part only was to find the mistake and solve Mar 11 19:30:39 mrdata: yes but I think it would also good to document the changes Mar 11 19:31:12 mr_nice: okay, i will do my best to document the changes Mar 11 19:31:22 mrdata: big thx! Mar 11 19:31:54 Crofton: Thanks, got me scratching my head. error message @ http://pastebin.ca/390911 Mar 11 19:31:57 mr_nice: must i declare every line, wich was changed Mar 11 19:32:51 Also wondering why I'm the only one having problems with librsvg if the bb file points to the wrong directory :( Mar 11 19:33:25 mrdata: no just a short comment at the top of the file Mar 11 19:33:25 ~ping NAiL Mar 11 19:33:28 pong NAiL Mar 11 19:33:44 mrdata: chanded gc1_reset() or something Mar 11 19:33:57 sorry geekoid I'm not sure how to sovle the problem Mar 11 19:34:22 mr_nice: okay, i will do so Mar 11 19:35:24 mrdata: and I think where the information is from is also important. zecke told me that the 2.4 mq200 driver was not gpl. Mar 11 19:35:30 Added revised bb file to pastebin error for anyone else who has a clue http://pastebin.ca/390915 Mar 11 19:36:05 Crofton: thanks for trying! at least I have one fewer errors! Mar 11 19:37:44 geekoid: well its insane.bbclass basicaly complaining Mar 11 19:38:42 koen : Hi ! Mar 11 19:38:59 mrdata: I hope I will find more time to also do some new kernel stuff for the simpad :) Mar 11 19:39:18 mr_nice: i have only change the driver from zecke ;-) Mar 11 19:40:26 mr_nice: and the names of the registers comes from the datasheet Mar 11 19:40:36 mrdata: it is cool that the 2.6 kernel gets complete. Mar 11 19:41:35 mr_nice: the cs3 stuff is also okay now, i think Mar 11 19:42:02 mrdata: please send it to me :) Mar 11 19:42:07 mr_nice: i must send the files to you Mar 11 19:42:21 * Ifaistos feels happy that gcc 4.1.1 finaly compiled on amd64 :) Mar 11 19:42:21 mrdata: hehe Mar 11 19:42:44 mr_nice: but only raw files for now Mar 11 19:43:39 mrdata: yes, no problem I will create patches out of the files and send a bugrequest to oe (with naming you) and upload them to the slackpad svn Mar 11 19:44:07 mr_nice: many thx! Mar 11 19:44:19 mrdata: I have to thank you ;) Mar 11 19:45:15 good morning all Mar 11 19:45:48 koen|galinhas: hi Mar 11 19:46:39 hey mr_nice Mar 11 19:55:31 koen|galinhas: It seems you cant libe with your OE dose :) Mar 11 19:55:47 koen|galinhas: It seems you cant live without your OE dose :) Mar 11 19:56:50 Apparently indt wants to use OE for a new project and pointed me out as "mr OE" Mar 11 19:56:57 so I'm getting enough OE doses :) Mar 11 19:57:03 indt? Mar 11 19:57:53 http://www.indt.org.br/ Mar 11 19:58:34 interesting Mar 11 20:00:08 koen|galinhas: cool Mar 11 20:05:25 koen|galinhas: btw got gcc 4.1.1 to compile Mar 11 20:05:42 Ifaistos: I saw your comment in bugzilla yesterday Mar 11 20:06:06 but the connection is to flaky to get vpn up and running, so I can't send (signed) e-mail :( Mar 11 20:06:51 koen|galinhas: it tool me the whole day to find it... and at the end all it was required was basically SITE_CONFIG="" Mar 11 20:07:04 that's gcc for you Mar 11 20:08:07 koen|galinhas: what puzzles me know is in how many places the same thing might be happening.... Mar 11 20:08:35 koen|galinhas: i.e mixing target/build configs Mar 11 20:08:48 RP, who did the bb file for exmap-console? Mar 11 20:09:31 zecke : hi Mar 11 20:09:39 hey Mar 11 20:35:29 had enough fun for today. Mar 11 20:35:35 night everyone Mar 11 20:36:07 nite ifaistos Mar 11 20:37:28 gn Mar 11 20:38:02 I'm looking at creating a bunch of sub-packages. I see how to specify individual files for each package but can I specify file lists by regular expression? Mar 11 20:41:18 hillct: grep for do_split_packages :) Mar 11 20:41:42 thanks Mar 11 20:41:59 on a particular page? Mar 11 20:42:05 in OE Mar 11 20:42:09 oh Mar 11 20:42:13 for te class Mar 11 20:42:14 K Mar 11 20:42:15 its python, so I don't know how it work :) Mar 11 20:42:45 * hillct is not a python fan but I understand the appear for some Mar 11 20:43:24 I was hoping it would be detailed in the docs, but I can read python enough to figure it out (I hope) Mar 11 20:43:59 koen|galinhas: You know such things. Do I need to implement a SyncML Client on Server for the OpenMoko device? Mar 11 20:44:20 zecke: daniel is setting next to me :) Mar 11 20:44:27 so syncml is a bad word ;) Mar 11 20:44:40 koen|galinhas: I even browse libsyncml.opensync.org Mar 11 20:45:03 I have no idea how syncml works, sorry Mar 11 20:50:21 koen|galinhas: at one point you mentioned the var I needed to set to mark a set of files as config files. I can't find it now. could you remind me of the variable name? Mar 11 20:50:45 CONFFILES iirc Mar 11 20:52:47 ah Mar 11 20:53:07 thanks. grepping for CONF was a hopeless excercise Mar 11 20:55:33 03nail 07org.oe.dev * r794c9bc6... 10/ (1 conf/distro/foonas.conf): foonas: Add missing $ Mar 11 20:55:38 03nail 07org.oe.dev * re7f88b3d... 10/ (5 files in 4 dirs): gcc: Fix crossbuild issue on x86_64 hosts (4.1.1/4.1.2) Mar 11 20:55:42 03nail 07org.oe.dev * r929600df... 10/ (1 packages/meta/foonas-packages.bb): foonas: Add gcc to feed Mar 11 21:08:37 what package needs to be patched if I want to change /etc/X11/Xserver ? Mar 11 21:09:02 I can't find this file... Mar 11 21:15:04 mr_nice: have send email to you with all changes; i will do some sleep now; bye Mar 11 21:15:17 he mrdata Mar 11 21:15:29 he woglinde Mar 11 21:16:33 woglinde: i hope simpad stuff goes forward now Mar 11 21:17:10 oh, simpad is alive again? Mar 11 21:17:17 cool Mar 11 21:17:27 tmbinc aeh yes Mar 11 21:17:32 2.6.20 runs Mar 11 21:17:36 but some problems Mar 11 21:17:41 no wake up from suspend Mar 11 21:17:49 pcmcia isnt working full speed Mar 11 21:18:12 but phoneme runs on it :) Mar 11 21:18:32 tmbinc: yes, and mq200 stuff is also working again Mar 11 21:18:57 but we need another hackingsession with zecke Mar 11 21:19:37 ah mrdata you made a patch? Mar 11 21:19:50 or is it the one in oe.dev? Mar 11 21:20:01 yes, after 4 days i found the bug Mar 11 21:20:09 ah Mar 11 21:20:11 nice Mar 11 21:20:14 nice. I still have a wince here to replace with linux. Does LCD-dimming work, instead of real standby? Mar 11 21:20:16 where is the patch? Mar 11 21:20:28 tmbinc hm Mar 11 21:20:32 never tested Mar 11 21:20:45 i have send the files to mr_nice Mar 11 21:21:02 hehe Mar 11 21:21:06 I need them too Mar 11 21:21:08 NOW! Mar 11 21:21:16 he will make patchs and send this to oe Mar 11 21:21:41 woglinde: witch email address Mar 11 21:22:02 you dont have websapce? *g* Mar 11 21:22:32 woglinde: not realy in short time Mar 11 21:23:12 woglinde: please give me some minutes Mar 11 21:23:56 mrdata no prob Mar 11 21:23:58 woglinde: hi Mar 11 21:24:02 he mr_nice Mar 11 21:24:24 mrdata: you can upload the stuff on opensimpad.org or on the slackpad svn Mar 11 21:24:27 mr_nice wake up from suspend isnt working here, or made you a new patch? Mar 11 21:24:56 woglinde: no here the same and mrdata tested it as well with the same result Mar 11 21:25:15 hm is something disabling the buttons? Mar 11 21:26:14 woglinde: disabling the buttons? Mar 11 21:27:05 mr_nice simpad should awake if you press the poweron button Mar 11 21:27:56 woglinde: the powerbutton has no driver but it should work without a driver Mar 11 21:29:04 woglinde: the power button is on a different gpio as the other keys. I will write a driver for it. I allready have some code for it but it is not complete. Mar 11 21:30:21 woglinde: but I will give it a other input device than the switches driver because of the autoresponse Mar 11 21:30:42 mr_nice: there is a generic gpio driven keyboard driver, so it should be really easy Mar 11 21:30:44 cyrilRomain|bbl: I had never used it for complicated patch sets Mar 11 21:32:55 mrdata: got your mail thx - creating patches now + uploading your file for woglinde Mar 11 21:32:58 woglinde: email done Mar 11 21:33:06 thanks Mar 11 21:33:24 woglinde Mar 11 21:33:32 woglinde: http://opensimpad.org/images/2/22/Simpad_stuff.tar.gz Mar 11 21:33:45 woglinde: I will upload the patches later Mar 11 21:34:27 mr_nice: the simpad_stuff.tar.gz is not complete i think Mar 11 21:35:02 mr_nice: i have send 2 emails to you, and you must merge Mar 11 21:36:09 Crofton: tomas (aka tf), one of the people at Openedhand. I know it could be better ;-) Mar 11 21:36:28 hi zecke Mar 11 21:36:37 and woglinde Mar 11 21:36:42 mrdata: with mq200.tar.gz ? Mar 11 21:36:46 mr_nice: sorry, should okay Mar 11 21:37:30 hi rp :) Mar 11 21:37:35 mrdata: yes, seems that there are all files included. everything which is not for the video driver is for the cs3 latch driver or? Mar 11 21:38:43 mr_nice: yes, but mq_skeleton.c is missing Mar 11 21:38:43 RP, I think the bb file needs some work Mar 11 21:38:57 when I have time and am happy with it, I'll let you know Mar 11 21:39:12 right noew I am having trouble installing the server on an image Mar 11 21:39:17 mrdata: ok, thx Mar 11 21:39:20 Crofton: tf is kind of learning OE/poky with it ;-) Mar 11 21:39:32 ok Mar 11 21:39:42 I'll keep that in mind :) Mar 11 21:40:34 Crofton: I did make some changes recently to sort a few major things, its on my list of things to revisit. If you have some suggestions, I will pass them on Mar 11 21:40:37 i will realy sleep now, good night all Mar 11 21:40:43 will do Mar 11 21:41:31 mrdata: good night Mar 11 21:58:57 JustinP: do you know if mtn is going to manage 'mtn diff' patch ? Mar 11 22:07:03 anyway, next times I submit a patch to the bug tracker, I'll add a note if it is a 'mtn diff' patch that require mtn operation to effectively apply Mar 11 22:07:24 * cyrilRomain think e.g. about its submitted quilt-0.46 patch that require the files/ drop to make quilt-0.46 build successfully Mar 11 22:08:45 good night Mar 11 22:12:18 good night Mar 11 22:14:52 mr_nice we need to fix this too -> pcmcia: request for exclusive IRQ could not be fulfilled. Mar 11 22:14:56 pcmcia: the driver needs updating to supported shared IRQ lines. Mar 11 22:17:07 woglinde: I will add it on the todo list (http://opensimpad.org/index.php/SIMpad_Kernel_2.6) . Mar 11 22:17:56 03nail 07org.oe.dev * r4743c739... 10/ (5 files in 4 dirs): disapproval of revision 'e7f88b3d1877bd1013495ae0870737c163c6c433' Mar 11 22:18:33 woglinde: here is the mq200 driver as a patch http://slackpad.svn.sourceforge.net/viewvc/slackpad/trunk/SIMpad-Kernel-2.6/patches/ Mar 11 22:19:04 woglinde: I will write a bugrep now. do you need the cs3 latch patch today? Mar 11 22:20:46 mjr_nice mrdata send me all files Mar 11 22:20:52 and I compiled it right now Mar 11 22:20:57 and it works Mar 11 22:21:06 nice to have a good screen again Mar 11 22:25:23 woglinde: yes! Mar 11 22:26:01 another task is to solve the ipc problem for opie Mar 11 22:27:36 woglinde: ipc? I need to complie opie. Mar 11 22:27:51 here is the bugrep 1971 Mar 11 22:30:24 woglinde: would be very nice to have opie on the simpad with 2.6 kernel. gpe is very mem hungry on my pad Mar 11 22:33:39 hm yes Mar 11 22:33:46 gpe needs X needs gtk Mar 11 22:34:01 opie only qt Mar 11 22:36:01 woglinde: yes. Mar 11 22:37:08 woglinde: I am trying to use xynth on simpad. it is on framebuffer and has gtk and sdl libs ported Mar 11 22:49:31 which package contains /etc/X11/Xserver ? Mar 11 22:49:42 03rpurdie * r790 10/ (5 files in 3 dirs): Update ChangeLog for 1.7.x changes, cleanup TODO Mar 11 22:51:49 noctambulus: xserver-common IIRC Mar 11 22:52:10 CoreDump|home: thx, I'll check Mar 11 22:54:06 I think bitbake 1.8.0 is ready to roll... Mar 11 23:05:26 there goes my raid device... :( /dev/md0 giving errors like crazy. bbl Mar 11 23:06:35 Anybody an idea about the problem in the following bug? Mar 11 23:06:40 !oebug 1946 Mar 11 23:06:41 * * Bug 1946, Status: NEW, Created: 2007-03-06 15:00 Mar 11 23:06:43 * * stefan(AT)datenfreihafen.org: Add recipe for fbgrab-viewer Mar 11 23:06:43 * * http://bugs.openembedded.org/show_bug.cgi?id=1946 Mar 11 23:58:05 night all Mar 12 02:20:52 I'm workig on hostap-daemon support for madwifi (in particular, for the IOMega NAS100d). Mar 12 02:21:01 It has an Atheros MiniPCI card. Mar 12 02:21:29 In hostap-daemon, there is a dependency on hostap-modules, which does not make sense for madwifi. Mar 12 02:21:35 indeed Mar 12 02:21:43 So I will override that RDEPENDS. Mar 12 02:21:54 Should I use an _ixp4xx override, or a _slugos override? Mar 12 02:22:26 I'd just remove it... Mar 12 02:22:32 koen|away has always got upset when I use a _slugos override, but how do I know whether other ixp4xx devices supported by OE all use madwifi? Mar 12 02:22:51 won't removing it break existing hostap-daemon users? Mar 12 02:24:08 why? most people are smart enough to have their proper wifi drivers installed, does hostapd need that dep? Mar 12 02:24:42 ok, I will ask hrw (since he's the hostap* maintainer) **** ENDING LOGGING AT Mon Mar 12 02:59:57 2007