**** BEGIN LOGGING AT Sat Sep 02 02:59:56 2006 Sep 02 06:51:00 morning Sep 02 07:11:04 good morning all Sep 02 07:11:18 hello koen Sep 02 07:11:55 what is the "dreambox" branch? Sep 02 07:17:14 morning all Sep 02 07:17:50 * NAiL makes a wild guess that the first hit on google for "linux dreambox" is correct... http://linuxdevices.com/articles/AT7482684956.html Sep 02 07:20:19 yes, PVRs and sat-boxes Sep 02 07:20:38 tmbinc is thee dreambox master over here Sep 02 07:51:54 Is this the right place to ask for some help? Sep 02 07:52:02 Just a pointer Sep 02 07:52:44 don't ask to ask, ask. If someone has teh solutions he/she will speak up Sep 02 07:52:53 Ok, sure Sep 02 07:52:55 Here goes Sep 02 07:53:25 Ok, I've just set up OE for help in the jlime distro Sep 02 07:53:55 So, I've set up configuration as set by jlime Sep 02 07:54:19 I've been trying to build nano (For practice) Sep 02 07:55:24 The toolchain builds nicely, but when it reaches the step to build glibc, it says that "preffered version 2.4 does not exist" and continues with 2.3.5+cvs. Sep 02 07:55:40 that's ok Sep 02 07:55:46 Thing is, 2.3.5+cvs fails at compilation Sep 02 07:55:52 (I think) Sep 02 07:56:03 * koen looks for kristoffer Sep 02 07:56:08 ~seen kristoffer Sep 02 07:56:24 I tried building 2.4 by itself, and everything went well Sep 02 07:56:38 kristoffer was last seen on IRC in channel #oe, 4d 13h 40m 22s ago, saying: 'nevermind, big duh'. Sep 02 07:56:45 :) Sep 02 07:56:48 4 days Sep 02 07:57:23 The .bb file for glibc_2.4 exists Sep 02 07:57:35 Why doesn't bitbake pick it up? Sep 02 07:58:38 I tried different building different targets, deleted the tmp directory a couple of times, and nothins Sep 02 08:00:10 good morning Sep 02 08:00:21 hello Sep 02 08:01:56 Seems like bitbake and the OE repository have a mismatch Sep 02 08:03:17 I think it's more likely the jlime.conf is faulty Sep 02 08:03:37 Hmmmmm Sep 02 08:03:44 I just want to force 2.4 Sep 02 08:04:01 Maybe the line in jlime.conf isn't right... Sep 02 08:05:01 Do I leave the distribution's configuration files in "org.openembedded.dev/conf/distro" ? Sep 02 08:05:33 Or do they have to go along with local.conf Sep 02 08:05:39 B_Lizzard: with bitbake 1.4.x, glibc 2.4 is selected even if bitbake dump "preffered version 2.4 does not exist" (at least with angstrom). What version of bitbake do you use ? Sep 02 08:06:03 I followed the Getting Started guide, so I think 1.6 Sep 02 08:06:10 That branch Sep 02 08:06:22 I'll check Sep 02 08:06:47 Yup, 1.6 Sep 02 08:07:09 Shall I revert to an earlyer one? Sep 02 08:07:44 Ok. I debugged a bit bitbake few weeks ago to see why such message was dump. I didn't go so far but found that a call to an update_data() method after the message dump makes bitbake selecting the glibc2.4 Sep 02 08:07:57 I hope bitbake 1.6 behaves the same Sep 02 08:10:43 ok, I thought that "dreambox" was an imaginary device, not a real one. What is the "oz345fam083" branch? AFAICT, the "oz354x" is merely for a zaurus unit.. Sep 02 08:12:42 what I'm wondering at this point, having failed miserably at a gpe-image, is that should I choose a specific branch and slog my way through with "PREFERRED_xxx" statements to correct for build problems, or just create my own isolated org.opendevelopment.dev branch and never update it until I get things working? Sep 02 08:15:11 ArjanS: ping Sep 02 08:17:16 T0mW: I think you should choose a specific branch and report bugs if you have build issues. And if you can fix bugs, it is even better :) Sep 02 08:18:24 cyrilRomain: ok, I'll try to contact the developers of the h1940 device, which is most similar to this board here, see, what they are using to build with and then go from there. Sep 02 08:18:32 T0mW: working isolated is not very good, because it does not help OE people, and OE people will have difficulties to help you Sep 02 08:18:43 I'm not able to build anything other than a basic bootstrap image Sep 02 08:19:27 while bootstrap is nice, it does leave the LCD + touchscreen out in the cold, heh Sep 02 08:20:01 T0mW: first be sure to choose the right branch. Then you can look at http://bugs.openembedded.org/ , just in case the build issue has already been detected by someone or is being fixed Sep 02 08:20:54 cyrilRomain: do I want a specific branch then, or is the dev branch something that I should use? Sep 02 08:21:10 or should I avoid the dev branch? Sep 02 08:21:16 T0mW: .dev Sep 02 08:21:23 ok Sep 02 08:21:33 then I'll focus on that then, thank you. Sep 02 08:22:07 * cyrilRomain remember that .dev branch built really better than the oz345fam083 few month ago... Sep 02 08:22:30 * cyrilRomain and I'm sure it is still true Sep 02 08:23:17 I'm using bitbake v1.6.1, is there anything more recent that I should be using? Sep 02 08:23:47 OE is just too intriguing to ignore, or give up on. :) Sep 02 08:24:17 T0mW: build from dev and put the PREFERRED_* stuff in your local.conf Sep 02 08:24:23 and file bugs where appropriate Sep 02 08:24:27 T0mW: it is ok to use bitbake v1.6.1, Sep 02 08:24:32 ok Sep 02 08:24:38 * T0mW goes back to work Sep 02 08:33:05 hi koen Sep 02 08:33:50 hey bluelightning Sep 02 08:41:02 T0mW: pong Sep 02 08:41:20 ArjanS: ah, you are doing the h1940? Sep 02 08:41:39 no, I'm just one of the guys helping Sep 02 08:42:01 ok, I'm on OFTP #h1940, shall we take this over there? Sep 02 08:42:20 OFTC rather Sep 02 08:43:36 yep Sep 02 09:54:59 really strange. I have OE on two machine but in a similar chrooted environment. On one machine bitbake 1.6 work perfectly. On the second bitbake 1.4 work well but not bitbake 1.6: I can't do any bitbake tasks because all os.system() call return -1 instead of 0, hence the throw of an exception. 8-( Sep 02 09:56:05 The most strange thingy is that when I manually call the os.system() command, it returns 0 Sep 02 10:09:48 03pH5 07org.oe.dev * r2b731b6c... 10/ (4 files in 3 dirs): Sep 02 10:09:48 xserver-kdrive: w100 accelerated kdrive for people to play with Sep 02 10:09:48 - enable compilation of Xw100 and package it as xserver-kdrive-w100 Sep 02 10:09:48 - thanks to Mardy, RP, sirfred, Spyro and XorA! Sep 02 10:10:52 pH5: does it do accell on w3220 too? Sep 02 10:11:11 pH5: woowoo Sep 02 10:11:48 pH5: cool :) Sep 02 10:11:55 koen: Seems to run stable. Are there any benchmark programs to test this? Sep 02 10:12:06 * koen has no idea Sep 02 10:12:43 morning all btew Sep 02 10:12:56 hey RP Sep 02 10:16:15 but the git version doesn't compile... somehow KDRIVE_LIBS contains "-L -lts" Sep 02 10:18:26 pH5: could change "X server from freedesktop.org, w100 version" to something like "X server from freedesktop.org, supporting the ATI imageon w100 chipset"? Sep 02 10:19:11 koen: sure Sep 02 10:19:49 hmm. the w100 Makefile.am uses KDRIVE_LIBS as make dependencies for Xw100, but git configure.ac appends TSLIB_LIBS to it, which contains the offending -lts because it's generated by pkg-config Sep 02 10:20:38 hello, is there any command to add all 2.6 modules on the system? (I use OZ/gpe on C3100) Sep 02 10:20:47 do I have to change w100/Makefile.am or can TSLIB_LIBS be removed from KDRIVE_LIBS? Sep 02 10:21:19 pH5: change Makefile.am I think, _LIBS shouldnt be dependencies IMO Sep 02 10:22:32 Varoudis|Laptop: 'ipkg install kernel-modules' Sep 02 10:22:41 thnx Sep 02 10:23:05 pH5: copy what fbdev does Sep 02 10:23:20 fbdev has just libfbdev.a in DEPENDENCIES Sep 02 10:23:54 pH5: that sounds right for ati as well, as it uses libfbdev.a for initialisation Sep 02 10:24:09 XorA|gone: great, I'll try that. Sep 02 10:24:27 pH5: sounds like they still cleaning the crap out of kdrive then Sep 02 10:26:08 pH5: should by FBDEV_LIBS VESA_LIBS libw100.a from looks of it Sep 02 10:26:33 pH5: to handle when fbdev and vesa are enabled/disabled Sep 02 10:26:53 XorA|gone: OLPC and maemo use xephyr a lot nowadays Sep 02 10:27:28 koen: X on X? Sep 02 10:27:43 yes Sep 02 10:27:49 koen: Why? Sep 02 10:28:05 xephyr can do composite, damage and different bitdepths as the host xserver Sep 02 10:28:15 xnest on steroids Sep 02 10:28:24 koen: but 2x framebuffer lost in ram Sep 02 10:28:30 instead of 1x Sep 02 10:29:02 XorA|gone: a small price to pay for a nice sandbox Sep 02 10:29:18 koen: sounds like a quick hack instead of a fix to me Sep 02 10:29:32 anyway time for me to offski for shower Sep 02 10:30:04 XorA|gone: Its ust useful for development when you need different screen properties to your host system Sep 02 10:30:26 like everything, it has advanatges and disadvantages Sep 02 10:30:40 I still don't understand... Sep 02 10:30:45 RP: I see that for devel work, but for a shipping system I doubt Sep 02 10:31:13 Has anyone got any problems with building glibc_2.3.5+cvs? Sep 02 10:31:19 XorA|gone: it's only for devel work Sep 02 10:31:23 Anyone else? Sep 02 10:31:32 XorA|gone: the real system will use kdrive/xorg/whatever Sep 02 10:32:59 koen: ah, I misunderstood then, pohew! Sep 02 10:33:37 koen: in the distro file it writes: PREFERRED_VERSION_glibc = "2.4" ... It seems just about right to me... Why doesn't bitbake find glibc 2.4 in the repository? Sep 02 10:34:24 probably because jlime.conf is missing settings for glibc-intermediate et al Sep 02 10:35:07 PREFERRED_VERSION_glibc-intermediate = "2.4" Sep 02 10:35:15 The line below Sep 02 10:35:34 I think it built glibc-intermediate just fine Sep 02 10:36:04 I think the jlime.conf file should be fine... Kristoffer has been using it for some time now Sep 02 10:36:32 It's probably some mismatch.... Sep 02 10:37:54 I can try to revert to an older bitbake release Sep 02 10:38:20 I got the one I use from the 1.6.0 branch yesterday Sep 02 10:39:06 koen, ipkg files kernel-modules shows 0 files! Sep 02 10:39:43 B_Lizzard: and if you 'bitbake -b ../org.openembedded.dev/packages/glibc/glibc_2.4.bb' manually ? Sep 02 10:39:56 It builds just fine Sep 02 10:40:04 Then, I go back to /build Sep 02 10:40:14 do a bitbake nano again Sep 02 10:40:30 and it doesn't realise 2.4 has been already built Sep 02 10:40:38 B_Lizzard: 'go back to /build' ? Sep 02 10:40:51 B_Lizzard: just launch this bitbake command _from_ the build dir Sep 02 10:40:57 yeah, /stuff/build or whatever you call it Sep 02 10:41:01 yep Sep 02 10:41:10 yup Sep 02 10:42:22 Anyways, I launch bitbake again so that it uses the same tmp directory, and it still insists on building 2.3.5+cvs Sep 02 10:43:32 Even though I've already built 2.4 which is recommended Sep 02 10:43:47 B_Lizzard: you built 2.4 from the build dir ? Sep 02 10:43:56 yes Sep 02 10:44:02 hm.. no clue then Sep 02 10:44:24 It just doesn't find 2.4, for some stupid reason Sep 02 10:44:34 I'll try bitbake 1.4.4 Sep 02 10:44:50 B_Lizzard: you can still go back for bitbake 1.4. But I think we should track the bug down Sep 02 10:45:27 If no-one else has this problem, it's probably something I've done Sep 02 10:51:41 koen: to build qemu with gcc-4.1, did you add '--disable-gcc-check' to the qemu bb file ? use ASSUME_PROVIDED to use the host qemu ? Or use a special version ? .. or maybe you don't use qemu ? Sep 02 10:53:18 cyrilRomain: I have only built qemu-native Sep 02 10:53:38 koen: I meant qemu-native Sep 02 10:54:14 koen@bitbake:~/OE/monotone/org.openembedded.dev/packages$ gcc -v Sep 02 10:54:14 Reading specs from /usr/lib/gcc-lib/i486-linux-gnu/3.3.6/specs Sep 02 10:54:14 Configured with: ../src/configure -v --enable-languages=c,c++ --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-gxx-include-dir=/usr/include/c++/3.3 --enable-shared --enable-__cxa_atexit --with-system-zlib --enable-nls --without-included-gettext --enable-clocale=gnu --enable-debug i486-linux-gnuThread model: posixgcc version 3.3.6 (Debian 1:3.3.6-13) Sep 02 10:54:22 where does 4.1 come in? Sep 02 10:56:02 koen: hum ok so you don't have this problem http://ewi546.ewi.utwente.nl/tinderbox/showlog.pl?machine_id=67&logfile=20060902114608.log Sep 02 10:56:31 well, I think I'll locally change the qemu-native bb file then Sep 02 10:56:56 no, since my build machine uses 3.x by default Sep 02 10:57:13 Can anyone here likely help with an nfs-utils cross compile failure? Sep 02 10:57:21 03koen 07org.oe.dev * r343dc70f... 10/ (5 files in 4 dirs): imlib: add 1.9.15 Sep 02 10:57:32 koen: do you know gcc4.x enough not to use it by default ? ;) Sep 02 10:58:08 I rarely build something with the native compiler, so I don't need 4.x Sep 02 11:00:03 http://pastebin.com/781879 if anyone can make head or tail of it Sep 02 11:05:53 blaster8: you use uclibc ? Sep 02 11:06:09 yes, unfortunately ;) Sep 02 11:06:23 blaster8: just use google, you will find that http://busybox.net/lists/uclibc/2006-March/014848.html Sep 02 11:06:48 I've tried google for the error, but didn't come up with that, thanks Sep 02 11:11:12 03koen 07org.oe.dev * r77d11684... 10/ (5 files in 4 dirs): qiv: add 1.9 Sep 02 11:12:38 03pH5 07org.oe.dev * rca845f15... 10/ (4 files in 3 dirs): xserver-kdrive: git build fix, better w100 description Sep 02 11:17:18 03koen 07org.oe.dev * r7881a5c7... 10/ (1 packages/wxbase/wxbase_2.7+cvs.bb): wxbase cvs: fix SRC_URI Sep 02 11:41:26 Two new bugs then, from me - 1377 & 1378 Sep 02 11:41:31 both nice and easy, hopefully Sep 02 11:43:32 * koen needs some coffee Sep 02 11:44:18 Even with bitbake 1.4.x, it fails Sep 02 11:44:21 Dammit Sep 02 11:45:04 B_Lizzard: maybe kristoffer has some local diffs Sep 02 11:45:11 Hmmmm Sep 02 11:45:14 I'll look Sep 02 11:47:15 koen: do I place the jlime-donkey.conf and jornada6xx.conf files along with local.conf? Sep 02 11:47:27 Or do I just leave them where they are? Sep 02 11:47:45 I see no reason why they should be moved Sep 02 11:47:53 who decided to put the xserver-common recipes into the xorg-xserver directory? Sep 02 11:48:29 So ../org.openembedded.dev/conf is ok Sep 02 11:48:31 pH5: probably an oversight of the AMD people Sep 02 11:48:41 ok, I'll put it back into its own dir Sep 02 11:48:45 03koen 07org.oe.dev * ra78ef645... 10/ (1 packages/openntpd/openntpd_3.7p1.bb): openntpd: don't pickup unwanted ssl libs, closes #1378 Sep 02 11:48:55 03koen 07org.oe.dev * r4ce40940... 10/ (3 files in 3 dirs): Sep 02 11:48:55 nfs-utils: fix #1378 and fix RDEPENDS on a kernel-module Sep 02 11:48:55 * NEVER (R)DEPEND on a kernel-module, since those can be built-in, use RRECOMMENDS instead Sep 02 11:52:33 * NAiL looks at koen Sep 02 11:53:02 aren't you supposed to ask the maintainer? :-P Sep 02 12:04:59 03pH5 07org.oe.dev * ra6a2ab84... 10/ (10 files in 4 dirs): xserver-common: move recipes back into their own dir, xserver-common is not part of xorg Sep 02 12:05:05 03pH5 07org.oe.dev * r9eda3dc8... 10/ (9 files in 5 dirs): xserver-common: move patches back into their own dir, xserver-common is not part of xorg Sep 02 12:05:11 03pH5 07org.oe.dev * rb390e39e... 10/ (3 files in 3 dirs): xserver-common: use Xw100 if available Sep 02 12:11:17 koen: I'd prefer the other solution to #1378, simply depending on openssl. Sep 02 12:13:42 why? openssl is huge Sep 02 12:14:07 and this problem only manifests itself if you built openssl before openntpd Sep 02 12:14:55 since a lot of packages already depend on openssl, so it is already present in most cases Sep 02 12:15:10 openssh, openvpn, python, ssmtp, vsftp, wpa-supplicant, tcpdump, mutt, nail, links, hostap, git, cups, cyrus-sasl, bind, centericq, apache, asterisk, ctorrent, ctrlproxy, just to mention a few. Sep 02 12:15:40 And frankly, I trust openssl more than the built-in crypto of openntpd :P Sep 02 12:16:21 openssl being huge is a point though Sep 02 12:16:26 do you know how big? Sep 02 12:17:12 ~700kB Sep 02 12:17:17 daim Sep 02 12:17:27 ok then ;) Sep 02 12:44:08 morning Sep 02 12:44:21 morning? Sep 02 12:44:23 hey chouimat Sep 02 12:44:30 * B_Lizzard checks watch Sep 02 12:44:32 hey chouimat Sep 02 12:44:34 ~ugt Sep 02 12:44:43 i heard ugt is Universial Greeting Time. Created in #mipslinux, it is a rule that states that whenever somebody enters an IRC channel it is always morning, and it is always late when the person leaves. The local time of any other people in the channel, including the greeter, is irrelevant. http://www.total-knowledge.com/~ilya/mips/ugt.html Sep 02 12:45:02 :D Sep 02 12:45:19 I'll keep that in mind Sep 02 12:45:24 B_Lizzard: it's 9am here Sep 02 12:45:45 4 PM here Sep 02 12:46:15 You guys wake up early! Sep 02 12:55:40 Anyways, good night! Sep 02 13:56:22 03nail 07org.oe.dev * r6401c65a... 10/ (19 files in 6 dirs): zd1211: Remove deprecated versions, update to rev83 Sep 02 13:57:04 XorA|gone, koen: I and sirfred meet on #w100 Sep 02 13:57:21 Good luck with this bug, gentlemen: Sep 02 13:57:22 http://bugs.openembedded.org/show_bug.cgi?id=1380 Sep 02 13:57:22 ah, cool Sep 02 13:57:24 but lately only sirfred works Sep 02 13:58:52 Ive begun a new work abroad, and I didn't even took the Zaurus with me; will take it next time I go home :-) Sep 02 13:59:45 Mardy: http://www.openembedded.org/viewmtn/revision.psp?id=2b731b6c8282c7d942d3117af572616c0d9c89db Sep 02 14:00:02 Anyone actually using ImageMagick, btw? Sep 02 14:00:28 I have it installed, but didn't try it yet Sep 02 14:00:36 (imagemagick, that is) Sep 02 14:01:07 I just tried to build it, and it fails when trying to link in zlib from /usr/lib Sep 02 14:01:27 http://bugs.openembedded.org/show_bug.cgi?id=1379 <-- log Sep 02 14:07:22 Mardy: cool Sep 02 14:08:42 NAiL: I can't reproduce that here: NOTE: package imagemagick-6.2.9: completed Sep 02 14:15:23 hmm why is it that all the slug folks are linking to host files, did someone change how slug builds ? Sep 02 14:15:58 hmm? Sep 02 14:16:11 dunno Sep 02 14:16:29 but something is definitely broken when building on x86_64 :-\ Sep 02 14:18:03 hey zecke Sep 02 14:18:36 I still can't figure out why log_check breaks Sep 02 14:19:20 tried a clean rebuild? Sep 02 14:19:26 yup Sep 02 14:19:48 doesn't work with the 3.10-beta sources from svn, nor oe head. Sep 02 14:20:03 hey Sep 02 14:20:18 it builds just fine on nudi (which is etch, IIRC) Sep 02 14:21:00 A test that "cannot possibly fail", fails. Sep 02 14:27:47 why on earth does it match a specific string which isn't present Sep 02 14:28:03 I have absolutely *no* idea Sep 02 14:31:39 * koen wonders why OOo needs Xaw Sep 02 14:35:06 hehe Sep 02 14:35:14 because KDAB ported it to Linux ;) Sep 02 14:35:56 koen: because being a *doze clone it just #include /usr/include/* Sep 02 14:36:36 * zecke takes a nap Sep 02 14:39:03 * koen fixes the emptiness of the fridge Sep 02 15:08:09 NAiL: have the site/ files changed for your host lately? Sep 02 15:08:27 ade|home: no Sep 02 15:20:08 sorted Sep 02 15:20:41 XorA|gone: gprs works now? Sep 02 15:20:44 XorA|gone: you got a good gprs tariff ? Sep 02 15:20:53 ade|home: flatrate Sep 02 15:21:04 ade|home: Tmobile Web and Walk Sep 02 15:21:26 koen: just testing connectivity from my phone Sep 02 15:21:28 the tmobile one that blocks loads of stuff? Sep 02 15:21:37 koen: it does? Sep 02 15:22:10 try making a voip call :) Sep 02 15:24:33 my next trick is to find a proxy for WM5 so Zaurus can get connection Sep 02 15:24:58 * RP should really get a decent phone Sep 02 15:25:06 it doesn't support DUN? Sep 02 15:25:21 you should be able to stick a bt card in the Z and dial in Sep 02 15:25:37 koen: using phone as modem is banned Sep 02 15:25:47 weird Sep 02 15:26:00 koen: so I figure using it as a proxy istn Sep 02 15:26:10 * XorA|gone suspects DUN would work fine Sep 02 15:26:26 koen, I install all the packages from your OOo blog but I get the same error with "configure" Sep 02 15:26:59 Varoudis|Laptop: is strings.h present? Sep 02 15:27:49 its just a pity I have no clue about an omap kernel for the HTC Wizard Sep 02 15:28:16 let me "find" it Sep 02 15:28:32 yeap in /usr/include/ Sep 02 15:29:10 Varoudis|Laptop: what's the error it reports into config.log? Sep 02 15:29:26 checking for strcasecmp() in string.h... no Sep 02 15:29:27 checking for strcasecmp() in strings.h... no Sep 02 15:29:27 configure: error: No case-insensitive string comparison function found. Sep 02 15:29:41 Varoudis|Laptop: that's not config.log Sep 02 15:29:52 config.log includes the exact error message Sep 02 15:30:15 sorry give me a sec Sep 02 15:35:24 koen, http://www.oesf.org/forums/index.php?showtopic=21010&st=15&p=140542&#entry140542 Sep 02 15:37:05 Varoudis|Laptop: that's what I suspected, try making the symlinks like I did on my blog Sep 02 15:37:31 'ln -sf /usr/lib/libstdc\+\+.so.6 /usr/lib/libstdc\+\+.so ; ln -s /lib/libgcc_s.so.1 /lib/libgcc_s.so ; ln /usr/lib/libz.so.1 /usr/lib/libz.so -sf' Sep 02 15:37:50 ok Sep 02 15:40:45 koen, I think its OK now.. Thnx! Sep 02 16:18:32 hi Sep 02 16:18:38 koen: building xaw now Sep 02 16:18:56 someone with poodle/tosa/collie here? Sep 02 16:19:40 hrw, can you add qiv,wxGTK(2.6.0),chmlib,xchm on the feed? Sep 02 16:31:03 Hello ! Sep 02 16:31:13 * NetJammer klots Varoudis|Laptop Sep 02 16:32:41 hello! Sep 02 16:33:01 anyone knows about linux at qtek s100 (pocket pc) ???anything Sep 02 16:45:36 koen: built Sep 02 16:45:46 NetJammer: mda3? mda2? Sep 02 16:45:59 hrw: many thanks! Sep 02 16:46:09 koen: ooo build progressing? Sep 02 16:46:22 yep Sep 02 16:46:36 with xaw it will get a bit further :) Sep 02 16:46:39 ;) Sep 02 16:46:55 ~lart ipkg-link Sep 02 16:46:56 * ibot takes out ipkg-link with the trash Sep 02 16:55:12 koen : u know anything for qtek ? Sep 02 16:55:39 if I can compile a Linux distro (kind a oz,opi etc) ? Sep 02 16:56:19 opie isn't a distro Sep 02 16:57:11 e package? Sep 02 16:57:34 is a project Sep 02 16:58:03 i think that linux on qtek is not ready! Sep 02 16:58:16 skase esy:P Sep 02 16:58:18 heh Sep 02 17:18:31 03Marcin 07org.oe.dev * r269c1d3f... 10/ (1 contrib/feed-browser/index.php): feed-browser: add search functionality Sep 02 17:42:27 03nail 07org.oe.dev * r7e36e79d... 10/ (12 files in 3 dirs): samba: upstream upgrade to 3.0.23c. Add MAINTAINER and LICENCE fields Sep 02 17:47:18 hi Sep 02 17:51:42 koen: use hrw@ instead of realname... Sep 02 17:51:43 hi CoreDump|home Sep 02 17:52:27 CoreDump|home: feed.openzaurus.org - can you test? Sep 02 17:52:37 s/feed/feeds/ Sep 02 17:53:12 * CoreDump|home looks Sep 02 17:54:48 * koen ipkg list | awk '{print $1}' | grep -- -dev | grep libx | grep -v xine | xargs ipkg -d loop install Sep 02 17:57:06 hrw|tv: round-robin is activated for feeds.openzaurus.org Sep 02 17:57:11 thx Sep 02 17:57:22 nice feed-browser =) Sep 02 17:57:34 CoreDump|home: feel free to improve it ;) Sep 02 17:57:38 :) Sep 02 17:58:08 CoreDump|home: can you add your hentges.rom tols into oe? Sep 02 17:58:15 s/tols/tools Sep 02 17:58:37 ARGH Sep 02 17:58:45 ~lart HAMA 8-in-1 card reader Sep 02 17:58:45 * ibot beats HAMA 8-in-1 card reader senseless with a 50lb Unix manual Sep 02 17:58:47 hrw|tv: heh I could, yeah. But some folks would probably bitch over how they work, that they are pretty Zaurus-specific etc Sep 02 17:58:59 CoreDump|home: not in .oz354x Sep 02 17:59:05 true Sep 02 17:59:09 contrib/hentges ? Sep 02 17:59:26 once I get my build env back up Ill have a look ;) Sep 02 17:59:30 koen: rather packages/hentges/ Sep 02 17:59:44 CoreDump|home: some people will bitch no matter what you do Sep 02 18:00:10 +mhentges@oe-akita:~/own_stuff >find . -type f|wc -l Sep 02 18:00:10 744 Sep 02 18:00:12 ;) Sep 02 18:00:57 CoreDump|home: there's one magical phrase you can say to complainers: 'patches accepted' :) Sep 02 18:01:01 Is anyone here compiling GPE (or gconf-dbus) on a Debian box? Sep 02 18:01:12 koen: true Sep 02 18:01:31 CoreDump|home: yes, ewi546 is running debian Sep 02 18:01:38 koen: great! Sep 02 18:01:54 could you give me that installed package list? Sep 02 18:02:08 CoreDump|home: want to test 3.5.4.1/collie images? Sep 02 18:02:09 ~curse gconf-dbus Sep 02 18:02:11 May the fleas of a thousand camels infest your most sensitive regions, gconf-dbus ! Sep 02 18:02:21 what's the preferred debian way to do that? Sep 02 18:02:33 hrw|tv: another time sure, but not today =) Sep 02 18:02:33 dpkg -l Sep 02 18:02:40 CoreDump|home: np. Sep 02 18:02:51 koen: or tar.gz /var/dpkg/whatever it was Sep 02 18:03:06 CoreDump|home: http://ewi546.ewi.utwente.nl/tmp/hrw/3.5.4x-experimental/collie/ Sep 02 18:03:11 CoreDump|home: http://ewi546.ewi.utwente.nl/tmp/dpkg-l.txt Sep 02 18:03:16 =D Sep 02 18:03:21 thanks you two! Sep 02 18:06:26 GRRRRRRRRRRRRR Sep 02 18:06:49 I have every dbus and orbit packages installed in that list but it doesn't help Sep 02 18:07:54 ewi used to lack X, but all the java crap the uni needed on their ipaqs pulled it in Sep 02 18:09:51 going to test 3.5.4.2-rc2 (final rc2) on osa/poodle.. Sep 02 18:14:35 hrw|tv: I added qiv and imlib to .dev today, they may work in .oz354x Sep 02 18:17:24 koen: will try Sep 02 18:20:57 koen: you created qiv.bb? Sep 02 18:21:09 yes Sep 02 18:21:53 wrong depends Sep 02 18:21:56 wrong do_install Sep 02 18:22:05 11 CFLAGS += " -lSM -lICE -lXdmcp" Sep 02 18:22:12 no such libs in depends Sep 02 18:22:21 gtk+ depends on those Sep 02 18:22:24 do_install into not existanse dirs Sep 02 18:22:44 do_install creates those dirs Sep 02 18:22:50 it suprised me as well Sep 02 18:23:02 ERROR: log data follows (/home/hrw/devel/build/3541-feed/tmp/work/arm-linux/qiv-1.9-r0/temp/log.do_install.25689) Sep 02 18:23:05 | install: cannot create regular file `/home/hrw/devel/build/3541-feed/tmp/work/arm-linux/qiv-1.9-r0/image/usr/bin/qiv': No such file or directory Sep 02 18:23:17 shit.. now strip fails Sep 02 18:23:48 bitbake -c clean qiv ; bitbake qiv works over here :( Sep 02 18:24:11 NOTE: package qiv-1.9-r0: task do_build: completed Sep 02 18:24:16 with 3 changes Sep 02 18:24:29 install -d bindir, mandir/man1 Sep 02 18:24:35 and nostripping Sep 02 18:30:00 hrw@bitbake:~/devel$ du -hs sources/ Sep 02 18:30:00 4.7G sources/ Sep 02 18:30:03 au Sep 02 18:30:33 3.3G dl_dir Sep 02 18:30:52 ewi has nice pipe.. rm -rf dl_dir Sep 02 18:40:20 oz 3.5.4.2-rc2/tosa/gpe starting.. Sep 02 18:41:25 gna Sep 02 18:41:27 configure: error: install xmkmf to run this script, include openwin in path Sep 02 18:43:33 argh Sep 02 18:44:06 hrw@bitbake:~$ dpkg -S `which xmkmf` Sep 02 18:44:06 xutils-dev: /usr/bin/xmkmf Sep 02 18:45:20 and/or imake Sep 02 18:45:35 * koen gives up for today and turns on the tv Sep 02 18:46:39 tosa works with 3542rc2 Sep 02 18:47:55 ipkg-make-index moved openzaurus-version into morgue... the only one existing ver.. Sep 02 18:49:50 it moves any version of it... Sep 02 18:55:11 woglinde: hi Sep 02 18:55:15 hi hrw Sep 02 18:56:01 woglinde: do you know where oedem will take place? I still need to find a place to sleep so would like to find something close to oedem rather Sep 02 18:56:38 hrw hm good question Sep 02 18:56:58 zecke's boss is still searching Sep 02 18:57:05 hrw|tv: could you paste bin the shell env (set) on ewi please? Sep 02 18:57:41 moment - flashing poodle with 3542rc2 Sep 02 18:58:03 CoreDump|home: http://ewi546.ewi.utwente.nl/tmp/hrw/t/set.txt Sep 02 18:58:20 thanks! Sep 02 18:58:53 that's the env you compile stuff with? Sep 02 18:59:25 moment Sep 02 18:59:42 CoreDump|home: reload Sep 02 19:00:13 hrw|tv: thanks =) Sep 02 19:00:26 CoreDump|home: your poodle is pxa255? Sep 02 19:00:33 yeah Sep 02 19:00:39 lucky one Sep 02 19:00:48 heh Sep 02 19:00:51 or even: lucky you Sep 02 19:01:17 * CoreDump|home prefers a VGA screen.... Sep 02 19:01:38 for some reasn gconf-dbus breaks because it can't find pkg-config. Sep 02 19:02:12 export PKG_CONFIG=/usr/bin/pkg-config works around the problem but is less than ideal for non-native stuff right? Sep 02 19:02:38 yes Sep 02 19:03:10 :\ Sep 02 19:03:21 poodle boot 3.5.4.2-rc2 Sep 02 19:04:21 anyone know why a bitbake would produce x86 binaries instead of arm ? Sep 02 19:04:25 yay Sep 02 19:04:31 i have a pretty minimal .bb file Sep 02 19:04:41 bb or conf? Sep 02 19:04:46 CoreDump|home: bootsplash need to be redone on all machines Sep 02 19:04:58 why is that? Sep 02 19:04:59 vom: you have 'inherit native'? Sep 02 19:05:09 hrw|tv: i have 'inherit autotools' Sep 02 19:05:13 i dont have that Sep 02 19:05:18 vom: or try to build stuff which has hardcoded gcc instead of $CC Sep 02 19:05:25 do i need both or just inherit native ? Sep 02 19:05:45 im trying to learn the bb syntax by reading other bb files Sep 02 19:05:47 not native Sep 02 19:05:52 and forum posts etc Sep 02 19:06:02 vom: check makefiles and compile logs Sep 02 19:06:05 it completes Sep 02 19:06:11 makes me a .ipk Sep 02 19:06:21 of course strip fails as it's not an arm bin Sep 02 19:06:36 other than that it's "successful" :( Sep 02 19:06:38 check compile logs Sep 02 19:06:45 hrw|tv: okay Sep 02 19:06:48 it probably use gcc instead of $CC Sep 02 19:08:00 hrw|tv: lots of signedness erros Sep 02 19:08:13 vom: warnings not errors Sep 02 19:08:17 guess i need to get it to reverse endian Sep 02 19:08:20 yeah warnings Sep 02 19:08:26 it completes Sep 02 19:09:13 argh... poodle with 32M ram is useless under gpe... 1M free ram Sep 02 19:10:21 so yeah lots of signedness warnings, no errors, it compiles and makes x86 bins Sep 02 19:10:35 vom: it probably use gcc instead of $CC Sep 02 19:10:53 hrw|tv: is there a way i can override that in the .bb or the Makefile of the package ? Sep 02 19:14:47 if it has broken makefiles then you need to patch them Sep 02 19:15:27 hrw|tv: you're right Sep 02 19:15:31 im changing the Makefile Sep 02 19:16:52 remember to submit patch upstream Sep 02 19:20:42 i still think the endianess is gonna kill me Sep 02 19:20:50 well see Sep 02 19:21:03 i made this package on the ipaq cluster some years ago Sep 02 19:21:05 and it worked Sep 02 19:21:10 this was familiar/ipaq etc Sep 02 19:25:06 make: *** Recursive variable `CC' references itself (eventually). Stop. Sep 02 19:25:08 i suck Sep 02 19:25:25 hrw|tv: so CC should I be using ? Sep 02 19:25:28 gcc-cross-arm ? Sep 02 19:25:47 vom: arm-linux-gcc if build for arm Sep 02 19:26:01 vom: show your recipe Sep 02 19:26:05 ~pastebin Sep 02 19:26:07 pastebin is probably a place to paste your stuff without flooding the channel - try http://pastebin.com/ (BROKEN AND SUCKING NUTS), or http://pastebin.ca, or http://channels.debian.net/paste, or for #oe use http://oe.pastebin.com, or http://bzflag.pastebin.ca/, or http://paste.lisp.org/ for the lisp/scheme nuts, or http://paste-it.net Sep 02 19:28:25 http://oe.pastebin.com/782214 Sep 02 19:29:18 tmp/work/nstx-1.1-beta5-1.0-r0/nstx-1.1-beta5/nstxcd: ELF 32-bit LSB executable, ARM, version 1 (ARM) Sep 02 19:29:22 well i got arm bins :) Sep 02 19:29:25 it was ugly Sep 02 19:29:43 now ill see if endianess is gonna kill me Sep 02 19:30:12 vom: s/mkdir -p/install -d Sep 02 19:30:42 vom: improve description, add email into maintainer Sep 02 19:30:54 vom: and ${D}${sbindir} is ok Sep 02 19:30:57 no need for '/' Sep 02 19:31:14 vom: add AUTHOR Sep 02 19:31:19 hrw|tv: ahh, okay Sep 02 19:31:51 re koen Sep 02 19:33:56 hey hrw|tv Sep 02 19:36:37 woglinde: is a place near the korfurstendam close enough? Sep 02 19:42:11 * koen <- Berlin n00b Sep 02 19:42:20 koen: not only you Sep 02 19:42:37 | ./gethost /home/hrw/devel/build/3541-feed/tmp/work/arm-linux/tcsh-6.14.00-r0/tcsh-6.14.00/host.defs >> tc.defs.c Sep 02 19:42:40 ;*( Sep 02 19:43:06 I figured out why tcsh compilation broke on the nbp Sep 02 19:43:28 the tcsh makefile can't handle busybox-sed Sep 02 19:47:02 hrw|tv: http://dominion.kabel.utwente.nl/koen/OE/tcsh-hostdefs-armv5te-linux-softfpa.txt Sep 02 19:47:18 the output of 'root@netbook-pro:/media/loop/build/tcsh-6.14.00# ./gethost host.defs > tcsh-hostdefs-armv5te-linux-softfpa.txt' Sep 02 19:47:30 Im fixing build Sep 02 19:49:50 NOTE: package tcsh-6.14.00-r0: task do_build: completed Sep 02 19:50:02 -rwxr-xr-x root/root 299732 2006-09-02 22:04 ./usr/bin/tcsh Sep 02 19:51:38 cool Sep 02 19:53:11 fsck.. just wrote patch and have to write it again.. forgot to copy.. Sep 02 19:59:09 koen: added into feeds Sep 02 20:21:13 03hrw 07org.oe.oz354x * rd1e2b437... 10/ (5 files in 4 dirs): Sep 02 20:21:13 imlib: added 1.9.15 from .dev Sep 02 20:21:13 applied changes from 2b731b6c8282c7d942d3117af572616c0d9c89db Sep 02 20:21:13 through 343dc70f181d19df8735634dae3396b73b9de348 Sep 02 20:21:18 03hrw 07org.oe.oz354x * red296776... 10/ (5 files in 4 dirs): Sep 02 20:21:18 qiv: added 1.9 from .dev (with do_install changes) Sep 02 20:21:18 applied changes from 343dc70f181d19df8735634dae3396b73b9de348 Sep 02 20:21:20 through 77d1168447e90e85323f9ec845e69ae66dfafff0 Sep 02 20:21:24 03nail 07org.oe.dev * rac412c49... 10/ (9 files in 2 dirs): samba: Remove duplicate files Sep 02 20:21:29 03nail 07org.oe.dev * r229b8b20... 10/ (11 files in 4 dirs): zd1211: Clean up duplicate files Sep 02 20:21:36 03nail 07org.oe.dev * r460db8f3... 10/ (1 packages/samba/samba.inc packages/samba/samba_3.0.23c.bb): samba: Move MAINTAINER/LICENCE to .inc Sep 02 20:21:41 03nail 07org.oe.dev * ra8265370... 10/ (1 packages/initscripts/initscripts-slugos_1.0.bb): initscripts-slugos: Fix S06alignment link removal Sep 02 20:21:47 03rpurdie 07org.oe.dev * r6b5174a9... 10/ (7 files in 2 dirs): qemu-native: Create qemu-native.bbclass so all versions have the gcc 3.x detection code Sep 02 20:22:27 constipation ohoy Sep 02 20:22:34 yep Sep 02 20:23:03 lag lag Sep 02 20:23:28 woglinde: no, I kicked the ciabot script Sep 02 20:23:41 isnt that a lag too? Sep 02 20:23:49 in some meanings Sep 02 20:23:58 more like nail said,a constipation :) Sep 02 20:24:08 hehe have to look up Sep 02 20:24:39 aehm well Sep 02 20:26:00 RP: did you see the bugrepoert on the cairo fixed patch? Sep 02 20:26:35 RP: qemu-native.inc was already in .oz354x Sep 02 20:28:03 quit Sep 02 20:28:06 sorry:) Sep 02 20:28:37 03nail 07org.oe.dev * r4f4d3c19... 10/ (1 packages/lcdproc/lcdproc_0.5.0.bb): lcdproc: Split off driver packages (large). Change MAINTAINER Sep 02 20:30:10 03nail 07org.oe.dev * rff7b9f5e... 10/ (1 packages/lcdproc/lcdproc_0.5.0.bb): lcdproc: Add forgotten --enable-libusb Sep 02 20:58:05 ugh...I know Sep 02 20:58:22 ...that someone added something to bb/oe to just build already built packages.... Sep 02 21:02:37 ? Sep 02 21:17:45 hrw|tv: I know I remember some feature added some months back which would "freeze" the bbfiles somehow... Sep 02 21:18:12 hrw|tv: and IIRC it would allow you to, say, build world and only have the packages built that you had already built Sep 02 21:18:18 hrw|tv: something like "emerge -auv world Sep 02 21:18:23 " for portage Sep 02 21:18:43 I'd like to update everything in my feed but not build all of world, in other words Sep 02 21:18:52 BBMASK? Sep 02 21:19:12 except that's manual.... Sep 02 21:19:26 I could swear I remember someone adding this somewhere...an auto-BBMASK kind of thins... Sep 02 21:28:48 yeah Sep 02 21:28:54 lemme find it Sep 02 21:30:26 it was basically "freeze" Sep 02 21:30:32 slugos used it Sep 02 21:31:11 # automatically generated by bitbake freeze Sep 02 21:31:46 bb freeze / bb unfreeze, IIRC Sep 02 21:32:01 except it fails to build... Sep 02 21:33:08 /home/repvik/slug/debianslug/tmp/work/freeze-1.0-r1/freeze: 42: Syntax error: Bad substitution Sep 02 21:33:52 for d in ${BBPATH//:/ } Sep 02 21:34:05 feel free to fix ;-) Sep 02 21:37:33 cu Sep 02 21:59:19 NAiL: thx. I'll try to look into it.... Sep 02 21:59:40 np Sep 02 22:24:00 http://bugs.openembedded.org/show_bug.cgi?id=1381 Sep 02 22:24:19 anyone interested might want to have a look at these LCDProc improvements Sep 02 22:28:51 03hrw 07org.oe.oz354x * rb0ddda2c... 10/ (5 files in 4 dirs): tcsh: added 6.14.00 (need some love) Sep 02 22:29:04 03hrw 07org.oe.oz354x * rc8c9e7a9... 10/ (3 files in 3 dirs): linux-openzaurus 2.6.17: added HACK which remove some IDs from orinoco_cs to not get it loaded for some cards Sep 02 22:29:04 03hrw 07org.oe.oz354x * r3dffb952... 10/ (3 files in 2 dirs): openzaurus 3.5.4.1/2: switched ZKERNEL_VERSION to MACHINE_KERNEL_VERSION like in .dev Sep 03 02:40:50 03mwester 07org.oe.dev * r0cb2088c... 10/ (1 packages/linux/ixp4xx-kernel_2.6.17.bb): DSM-G600A support added (Artop IDE and Via velocity patches) **** ENDING LOGGING AT Sun Sep 03 02:59:57 2006