--- Log opened Wed May 09 00:00:07 2007 --- Day changed Wed May 09 2007 00:00 < sepp> then rxr will read it and answer - he did the avr32 work ... 00:00 < write_erase> do you have his email ? 00:01 < sepp> better send it to the mailinglist 00:01 < write_erase> ok 00:01 < sepp> gets archived and one can google things later 00:56 -!- idealm [n=ideal@c58-107-18-60.belrs2.nsw.optusnet.com.au] has joined #t2 01:12 -!- mtr_ [n=Michael@kobz-590cbec6.pool.einsundeins.de] has joined #t2 01:30 -!- mtr [n=Michael@kobz-590cb392.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 01:30 -!- mtr_ is now known as mtr 04:52 -!- idealm [n=ideal@c58-107-18-60.belrs2.nsw.optusnet.com.au] has quit [Read error: 60 (Operation timed out)] 07:30 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 08:17 < rene> re 08:17 < rene> moin 08:17 < sepp> moin rene :) 08:19 < Stelz> rehi rene, sepp :) 08:20 < sepp> dobry dien :) 08:20 < Stelz> dobroe utro, sepp :) 08:20 < sepp> hehe 08:23 < rene> hi Stelz 08:23 < rene> hi all 08:23 [Users #t2] 08:23 [@ChanServ] [ emte] [ R4gnar0k] [ Stelz ] [ write_erase] 08:23 [ CIA-9 ] [ LMJ ] [ rene ] [ TobiX ] 08:23 [ Dallur ] [ mtr ] [ sepp ] [ valentin] 08:23 -!- Irssi: #t2: Total of 13 nicks [1 ops, 0 halfops, 0 voices, 12 normal] 08:32 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 60 (Operation timed out)] 08:36 < CIA-9> mtr * r23941 /trunk/package/gnome2/gtk+/gtk+.desc: * updated gtk+ (2.10.11 -> 2.10.12) 08:36 < CIA-9> mtr * r23936 /trunk/package/network/ndiswrapper/ndiswrapper.desc: * updated ndiswrapper (1.42 -> 1.43) 08:36 < CIA-9> mtr * r23939 /trunk/package/gnome2/gtkmm/gtkmm.desc: * updated gtkmm (2.10.9 -> 2.10.10) 08:36 < CIA-9> mtr * r23935 /trunk/package/develop/php/php.desc: * updated php (5.2.1 -> 5.2.2) 08:36 < CIA-9> mtr * r23938 /trunk/package/gnome2/gnumeric/gnumeric.desc: * updated gnumeric (1.7.9 -> 1.7.10) 08:37 < CIA-9> mtr * r23943 /trunk/package/filesystem/udev/ (udev.conf udev.desc): * updated udev (109 -> 110) 08:37 < CIA-9> mtr * r23937 /trunk/package/audio/gnupod/gnupod.desc: * updated gnupod (0.99.1 -> 0.99.2) 08:37 < CIA-9> mtr * r23940 /trunk/package/gnome2/glibmm/glibmm.desc: * updated glibmm (2.12.8 -> 2.12.9) 08:37 < CIA-9> mtr * r23942 /trunk/package/graphic/gnofract4d/gnofract4d.desc: * updated gnofract4d (3.3 -> 3.4) 08:40 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 08:48 < CIA-9> susan * r23944 /trunk/package/base/man-pages/man-pages.desc: * updated man-pages (2.44 -> 2.46) 09:04 < CIA-9> susan * r23945 /trunk/package/gnome2/gnome-games/gnome-games.desc: * updated gnome-games (2.18.1 -> 2.18.1.1) 09:05 < CIA-9> susan * r23946 /trunk/package/develop/git/git.desc: * updated git (1.5.1.1 -> 1.5.1.3) 09:08 -!- idealm [n=ideal@c58-107-18-60.belrs2.nsw.optusnet.com.au] has joined #t2 09:10 -!- You're now known as rxr 09:10 -!- You're now known as rxr 09:34 -!- write_erase [n=olivier@bon13-2-82-237-125-220.fbx.proxad.net] has quit ["Leaving"] 09:37 < CIA-9> rene * r23947 /trunk/package/java/javawm/ (. javawm.cache javawm.desc): * added javawm (1.4.5) - A new, small Java Virtual Machine conforming to the JVM specification 09:41 < CIA-9> rene * r23948 /trunk/package/python/kiwi/ (fix_ProxyFileChooserButton.patch kiwi.desc): * updated kiwi (1.9.8 -> 1.9.14) 10:18 < CIA-9> rene * r23949 /trunk/package/gnome2/pidgin/pidgin.desc: * updated pidgin (1.5.0 -> 2.0.0) 10:19 < Stelz> stupid AOL 10:28 < CIA-9> susan * r23950 /trunk/package/security/opensc/opensc.desc: * updated opensc (0.11.1 -> 0.11.2) 10:29 < CIA-9> susan * r23951 /trunk/package/printing/cups-pdf/cups-pdf.desc: * updated cups-pdf (2.4.5 -> 2.4.6) 10:42 < CIA-9> susan * r23952 /trunk/package/contrib/rrdtool/rrdtool.desc: * updated rrdtool (1.2.21 -> 1.2.23) 10:42 < CIA-9> susan * r23953 /trunk/package/e17/imlib2/imlib2.desc: * updated imlib2 (1.3.0 -> 1.4.0) 10:42 < CIA-9> susan * r23954 /trunk/package/e17/imlib2_loaders/imlib2_loaders.desc: * updated imlib2_loaders (1.3.0 -> 1.4.0) 10:43 < CIA-9> susan * r23955 /trunk/package/base/inotify-tools/inotify-tools.desc: * updated inotify-tools (3.9 -> 3.10) 10:43 < CIA-9> susan * r23956 /trunk/package/develop/boost-jam/boost-jam.desc: * updated boost-jam (3.1.13 -> 3.1.14) 10:43 < CIA-9> susan * r23957 /trunk/package/base/pciutils/pciutils.desc: * updated pciutils (2.2.4 -> 2.2.5) 10:44 < CIA-9> susan * r23958 /trunk/package/network/cfengine/cfengine.desc: * updated cfengine (2.1.22 -> 2.2.0) 10:46 < CIA-9> susan * r23959 /trunk/package/multimedia/xine-ui/xine-ui.desc: * updated xine-ui (0.99.4 -> 0.99.5) 10:46 < CIA-9> susan * r23960 /trunk/package/perl/perl-dbi/perl-dbi.desc: * updated perl-dbi (1.54 -> 1.55) 10:46 < CIA-9> susan * r23961 /trunk/package/base/hdparm/hdparm.desc: * updated hdparm (7.1 -> 7.3) 10:46 < CIA-9> susan * r23962 /trunk/package/perl/perl-devel-stacktrace/perl-devel-stacktrace.desc: * updated perl-devel-stacktrace (1.14 -> 1.15) 10:46 < CIA-9> susan * r23963 /trunk/package/network/pound/pound.desc: * updated pound (2.3 -> 2.3.1) 10:46 < CIA-9> susan * r23964 /trunk/package/graphic/dcraw/dcraw.desc: * updated dcraw (8.70 -> 8.72) 10:47 < CIA-9> susan * r23965 /trunk/package/network/dclib/dclib.desc: * updated dclib (0.3.8 -> 0.3.9) 10:47 < CIA-9> susan * r23966 /trunk/package/network/ipw3945/ipw3945.desc: * updated ipw3945 (1.2.0 -> 1.2.1) 10:48 < CIA-9> susan * r23967 /trunk/package/java/jetty/jetty.desc: * updated jetty (6.1.1 -> 6.1.3) 11:07 < CIA-9> susan * r23968 /trunk/package/develop/autogen/autogen.desc: * updated autogen (5.8.9 -> 5.9.1) 11:13 < CIA-9> rene * r23969 /trunk/package/base/readline/readline.desc: * updated readline (5.2-002 -> 5.2-004) 11:14 < CIA-9> rene * r23970 /trunk/package/shells/bash/bash.desc: * updated bash (3.2-15 -> 3.2-17) 11:14 < CIA-9> rene * r23971 /trunk/package/graphic/asymptote/asymptote.desc: * updated asymptote (1.26 -> 1.27) 11:15 < CIA-9> rene * r23972 /trunk/package/graphic/xfig/ (gcc-4.0.patch mandir.patch xfig.desc): * updated xfig (3.2.4 -> 3.2.5) 11:37 < CIA-9> rene * r23973 /trunk/package/network/zaptel/zaptel.desc: * updated zaptel (1.4.1 -> 1.4.2.1) 11:51 < CIA-9> rene * r23974 /trunk/package/filesystem/cdrkit/cdrkit.desc: * updated cdrkit (1.1.5.1 -> 1.1.6) 11:52 -!- write_erase [n=olivier@81.80.156.36] has joined #t2 11:52 < write_erase> Hi rxr. 11:52 < rxr> hi write_erase 11:52 < write_erase> I heard you were the maintainer of avr32 branch ? 11:52 < rxr> yep 11:55 < write_erase> I try to compile t2 with T2_TRUNK / GENERIC_EMBEDDED / avr32 / AP7000 optimisations / DISABLE NLS , and I get the following error 11:55 < write_erase> on ucLibc : 11:56 < write_erase> (hold on, until I paste you a fresh compilation error message ) 12:02 < write_erase> Sorry, The re-compilation of uClibc is a bit long ..but This is some trouble with usr/include common files 12:02 < rxr> I can wait 12:02 < rxr> I hang around whole days, usually 12:04 < write_erase> rxr, what kind of AVR32 board you have ? 12:04 < rxr> the STK1000 12:04 < rxr> and you ? 12:04 < write_erase> I have NGW100 ... 12:04 < write_erase> :-) 12:04 < rxr> ah, wanted to order one as well just for testing 12:04 < rxr> in which country are you located, and where did you order the board ? 12:04 < write_erase> You have already boot a T2 linux kernel & T2 filesystem ? 12:05 < rxr> yes, sure 12:05 < write_erase> France... local reseller ... . 12:05 < rxr> we even showcasted the thing on CeBIT 12:05 < write_erase> rxr, you are from Atmel ? 12:05 < rxr> nope, I'm from ExactCODE 12:05 < rxr> http://www.t2-project.org/architectures/avr32.html 12:06 < rxr> e> France... local reseller ... . 12:06 < write_erase> Ok , happy to meet you... Hope I can help you on thiis project 12:06 < rxr> http://rene.rebe.name/photos/?p=/Hardware/AVR32/ 12:06 < write_erase> Where are you from ? 12:06 < rxr> Germany, Berlin 12:07 < write_erase> Here is the log : 12:07 < write_erase> http://rafb.net/p/M5nEMm73.html 12:07 < rxr> http://rene.rebe.name/photos/?p=/CeBIT/2007/&index=71 12:08 < write_erase> Nice :-) 12:08 < rxr> hm, what distribution are you building on? 12:08 < rxr> did you change the "flist detection technique" setting in the Config ? 12:09 < write_erase> I'm ordering 640x480 TFT LCD from AXIS x50v PDA.... for my board 12:09 < write_erase> Distro Debian Etch 12:09 < write_erase> flist not changed 12:09 < rxr> hm, ok, strange 12:09 < write_erase> Could you explain me what's wrong ? 12:09 < write_erase> So I can troubleshoot 12:09 < rxr> if that is the only error (that is there is nothing important one hidden while the tail of the file was displayed) 12:10 < rxr> it's just that the build system "thinks" that usr/include is in two packages, 00-dirtree and uclibc 12:10 < CIA-9> rene * r23975 /trunk/package/graphic/transfig/ (multi-line-string-fix.patch use-stdarg-h-gcc33-fix.patch): * removed outdated, non-applying transfig patches 12:10 < rxr> I never saw such an error message 12:11 < write_erase> rxr, how this detection is done ? by diffing the rootfs before and after the make install ? or something like that ? 12:11 < rxr> can you open build/.../var/adm/logs/1-uclibc.err and look if there is another error message in the file? 12:11 < rxr> write_erase: no, not by default:; this diffing is slow and does not work with parallel builds 12:12 < rxr> by default an T2 flist-wrapper is used, it's a .so shared object which is preloaded and wraps open,read,write, etc. 12:12 < write_erase> Ok I understand 12:12 < rxr> that is way faster, and works with parallel build packages 12:12 < rxr> the find and diff is available optionally 12:12 < rxr> but it is seldomly used 12:13 < rxr> if there is no real error ini the log you could rename the 1-uclibc.err into 1-uclibc.log and let the build continue for 12:13 < rxr> now 12:13 < write_erase> Is there some verbose mode available to see what the wrapper 12:13 < write_erase> thinks . 12:13 < rxr> the wrapper could be compiled with debugging enabled 12:13 < rxr> but you need to edit a file for that 12:14 < rxr> I would propose to rather rename the log file if there is no real other error and let the build continue 12:14 < rxr> to get working binaries and a feeling for T2 until you start debugging this 12:17 < write_erase> Is there something special to do with your kernel and UBOOT on the AVR32 device ? 12:17 < write_erase> (when the compilation is done) 12:17 < rxr> well - I'm not sure if the current T2 kernel for the STK1000 will boot on the NGW100 12:17 < rxr> I wanted to order the NFW100 for some weeks now but the resellers in Germany are a bit lagging and annoying ... 12:18 < rxr> the current AVR32 kernel built by T2 is the 2.6.20+AVR32 patches for the STK1000 12:18 < rxr> maybe it works on the NGW100 as well, but I do not know off hand (e.g. different wiring and also some fixed device condig like the LCD and such not present on the NGW100) 12:18 < write_erase> OK, the boards are quite the same .... the STK1000 has more devices, but the core is the same 12:18 < write_erase> :-) 12:18 < rxr> yes, the AP7000 12:19 < rxr> the board comes preloaded with a kernel, right? 12:19 < write_erase> Yeah ... 12:19 < rxr> I would start coping T2 built binaries onto the device and use the board kernel for the bigging 12:19 < rxr> when you gained some experience just replace the kernel at the end when you are more familiar with the boot loader and debugger 12:19 < rxr> one thing to note 12:20 < write_erase> I already have the DHCP/TFTP upgrade method, so no problem if it goes wrong 12:20 < rxr> ah, ok 12:20 < rxr> one thing to note: 12:20 < rxr> atmel had to change the u-boot image ID while they where shipping the boards 12:20 < rxr> this was because they picked the next best free in u-boot when they did the inital hardware over a year ago 12:21 < rxr> and in the meantime this ID was given to I think it was the Blackfin architecture 12:21 < rxr> so they where assigned another official one 12:21 < rxr> the T2 u-boot mkimage currently uses the old one as boards in the field so far used the old one 12:21 < rxr> and would need to reflash the u-boot loader in the flash to recognize the new ID 12:21 < rxr> I guess the NGW100 is new enough to come with the new u-boot using the new ID by default 12:22 < rxr> in that case you need to add a patch to u-boot to use the new ID 12:22 < write_erase> What is this ID used for ? something about the partitions magics ? 12:22 < rxr> not partition but kernel magic 12:22 < write_erase> ah ok 12:23 < rxr> just to prevent to accidently load a kernel for another architecture 12:23 < rxr> ah, no I think T2/trunk already uses the new ID 12:24 < rxr> I have a local downdate to use the old ID on our boards 12:24 < write_erase> I can't upgrade UBOOT, 'cause I fried my JTAG ICE 12:24 < rxr> heh 12:24 < write_erase> I'll check these ID if that don't boot 12:25 < rxr> well - I also did not update our boards, yet 12:25 < rxr> T2/trunk should use the recent ID already 12:25 < rxr> u-boot will complain verbosely when the ID does not match 12:25 < rxr> something about unsupported architecture IIRC 12:27 < write_erase> Has MPLAYER AVR32 optimisations ? 12:29 < rxr> nope, only patched to compile 12:29 < rxr> if you find assembly optimizations I'm happy to add them 12:29 < rxr> due to patent reasons atmel can not open their mplayer patches they did for performance tests 12:30 < rxr> they fear mp3 and co patent lawyers coming around then ... 12:32 < CIA-9> rene * r23976 /trunk/package/multimedia/easytag/easytag.desc: * updated easytag (2.0.2 -> 2.1) 12:40 < rxr> what the fuck: Sch?uble ?bernimmt Schirmherrschaft des LinuxTag 12:40 < rxr> ^- valentin !?!?!!?!!? 12:40 < rxr> http://www.golem.de/0705/52176.html 12:41 < valentin> hm ? Bundestrojaner becomes open source ;-) 12:41 < rxr> somehow my motivation to run a T2 booth there now dropped by about 100% 12:45 < valentin> why ? you went there before when schilly was innenminister... 12:46 < rxr> hm - last time I was on linuxtag was about 2003 or so ... 12:46 < valentin> ic 12:46 < rxr> though Schaeuble is way more stupid and evil than Schilly ... 12:47 < rxr> can't believen somehow this puppy was given some pseudo "schirmherrschafts" position .. 12:51 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 12:51 < valentin> oehm, schirmherrschaft is not "given" 12:52 < rxr> what he takes it ? 12:52 < valentin> he offers it 12:52 < valentin> and it has been accepted 12:53 < rxr> they could have accepted a more competant person 12:53 < rxr> (or none) 12:57 < CIA-9> rene * r23977 /trunk/package/fonts/fontforge/fontforge.desc: * updated fontforge (20070312 -> 20070501) 12:57 < valentin> why ? i'd even appreciate bill gates for that position... by doing so schauble says linux is a good thing. much better than some people in the american administration who claim open source is for communists 12:58 < rxr> did you know microsoft and Novell have a Linux foundationtion thing? 12:58 < rxr> and Dell joined it, now ? 12:59 < valentin> yeah, what ever 13:02 < CIA-9> rene * r23978 /trunk/package/graphic/transfig/ (transfig.conf usr-local.patch): * fixed the new transfig to actually build and install 13:03 < write_erase> rxr, did you meet this one : 13:04 < write_erase> !> avr32-t2-linux-uclibc-gcc -g -O2 -o gzip bits.o crypt.o deflate.o gzip.o inflate.o lzw.o trees.o unlzh.o unlzw.o unpack.o unzip.o util.o zip.o lib/libgzip.a 13:04 < write_erase> !> gzip.o: In function `treat_file': 13:04 < write_erase> !> gzip.c:(.text+0x17d0): relocation truncated to fit: R_AVR32_16N_PCREL against symbol `quiet' defined in .bss section in gzip.o 13:04 < rxr> not that I remeber 13:04 < rxr> though I might not have built gzip 13:04 < rxr> I build some in-house mobile-device target 13:05 < rxr> and gzip is unlikely to be included 13:05 < rxr> O 01-------- 102.040 archiver gzip 1.3.11 / base/tool CROSS DIETLIBC 0 13:05 < rxr> yes, gzip is not included in my avr32 default test-build 13:05 < write_erase> My first attempt was a GENERIC_EMBEDDED without any work on package selection 13:05 < rxr> you can just deselect it for now 13:05 < write_erase> Ok ... I'll remove it 13:06 < rxr> likely the a busybox included version is enough 13:06 < rxr> you have to addd mplayer into your selection as well 13:06 < rxr> the default generic embedded target does not include it 13:06 < rxr> btw. I just wrote some paragraphs about target creation in the currently rewritten T2 handbook 13:06 < rxr> did you had a look into the handbook already 13:08 < write_erase> I read some parts ... 13:09 < write_erase> I'm having a look on package selection - New targets 13:20 < write_erase> rxr, should I alter config/xxxx/packages to remove packages ? 13:22 < rxr> you can either edit target/embedded/config.in or enable the expert config section, as well as custom package selection and then add rules like: 13:22 < rxr> O gzip 13:22 < rxr> X mplayer 13:24 < write_erase> Ok 13:28 < rxr> valentin: heise.de: "Wie auf der Website der Tagungsorganisation betont wird, ist die ?bernahme der Schirmherrschaft durch das Bundesinnenministerium eine Tradition, die vom Linuxtag seit f?nf Jahren gepflegt wird. Eine politische Wertung sei mit der Schirmherrschaft nicht verbunden." 13:48 -!- sepp_ [n=sepp@p85.212.2.105.tisdip.tiscali.de] has joined #t2 13:51 -!- keinek [n=chatzill@201.250.3.69] has joined #t2 13:56 -!- keinek [n=chatzill@201.250.3.69] has quit [Remote closed the connection] 13:59 < write_erase> rxr, there's definitely a problem with uclibc (on 2 differents PC / builds ) Same distro Debian Testing. I have to rename log to err 14:05 -!- sepp [n=sepp@p213.54.53.245.tisdip.tiscali.de] has quit [Read error: 110 (Connection timed out)] 14:14 < rxr> re 14:16 < rxr> ok, I make a note and try to reproduce this 14:29 < CIA-9> rene * r23979 /trunk/package/audio/audacious-plugins/audacious-plugins.desc: * updated audacious-plugins (1.3.3 -> 1.3.4) 15:02 < rxr> heh, adding the Sony HidMD to the Qemu VM made the whole VM very kinda sluggish 15:04 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 15:07 [Users #t2] 15:07 [@ChanServ] [ emte ] [ mtr ] [ sepp_] [ valentin ] 15:07 [ CIA-9 ] [ idealm] [ R4gnar0k] [ Stelz] [ write_erase] 15:07 [ Dallur ] [ LMJ ] [ rxr ] [ TobiX] 15:07 -!- Irssi: #t2: Total of 14 nicks [1 ops, 0 halfops, 0 voices, 13 normal] 15:07 < CIA-9> rene * r23980 /trunk/package/graphic/gnofract4d/install.patch: * removed outdated, non-applying patch from gnofract4d 15:27 < rxr> oh, this Qemu slowness with usb host devices is a known limitation, damn 15:29 < write_erase> rxr, /home/olivier/avr32/t2-trunk/download/mirror/a/acpi-dsdt-initrd-v0.8.4-2.6.21.patch: No such file or directory while compiling the kernel 15:32 < write_erase> Isn't these files downloaded at compilation time ? 15:36 < write_erase> architecture/avr32/package/linux26/9-avr32-fixes.patch is missing too ... ./scripts/Download -cfg system -required is OK 15:38 < rxr> write_erase: when did you checkout your t2 trunk/ ? 15:38 < rxr> just recently? 15:38 < rxr> or a week ago ? 15:38 < write_erase> today 15:38 < write_erase> this morning ? 15:40 < rxr> hm, ok then there still is a bug with the overlaying 15:40 < rxr> that is a generic, new T2 feature - not related to AVR32 as it 15:40 < rxr> I take a quick look 15:48 < rxr> somehow I have the feeling this error does not happen on my side 15:48 < rxr> one more minute for a test-build 15:51 < write_erase> rxr, architecture/avr32/package/linux26/9-avr32-fixes.patch and download/mirror/a/acpi-dsdt-initrd-v0.8.4-2.6.21.patch are not present on my system. Is that a download problem ? or some dependancy error ? 15:52 < rxr> the second one should not be needed 15:52 < rxr> it would be downloaded for normal, e.g. x86 end-user system builds s... 15:52 < rxr> the former should be part of t2 15:52 < rxr> what does this output: 15:53 < rxr> # svn st architecture/avr32/package/linux26/9-avr32-fixes.patch 15:54 < write_erase> returns nothing 15:55 < write_erase> But the file exists 15:55 < rxr> ins't it the one you wrote is missing = 15:55 < rxr> ? 15:56 < write_erase> I"m sorry ... I try to rebuild to get a fresh error 15:58 < rxr> hm - here it built 15:58 < rxr> == 05/09/07 15:56:11 =[1]=> Finished building package linux26. 15:59 < rxr> but when you get these files missing the probably still is a bug hiding in the new package overlay / inherit feature of T2 16:00 < write_erase> cp: cannot stat `arch/avr32/boot/images/uImage': No such file or directory 16:03 < rxr> is that the only error? 16:03 < write_erase> olivier@blackpearl:~/avr32/t2-trunk$ sudo file architecture/avr32/package/linux26/9-avr32-fixes.patch 16:03 < write_erase> architecture/avr32/package/linux26/9-avr32-fixes.patch: ASCII English text 16:04 < write_erase> Apply patch architecture/avr32/package/linux26/9-avr32-fixes.patch ... 16:04 < write_erase> sed: can't read architecture/avr32/package/linux26/9-avr32-fixes.patch: No such file or directory 16:04 < write_erase> There's something strange 16:04 < rxr> do you run the Build-Target with sudo ? 16:04 < write_erase> yes all the time sudo 16:04 < rxr> intersting that it works that well at all ... 16:04 < write_erase> I hate working as root 16:05 < rxr> in the past sudo caused more problems ... 16:05 < rxr> well, anyway it should work 16:05 < write_erase> I can try as full root 16:05 < rxr> no - wait 16:05 < rxr> let's try to really nail this down 16:05 < rxr> T2 is suppost to just work 16:05 < rxr> everything else is a bug 16:05 < write_erase> Do you want I upload the full linux2.6.err ? 16:05 < rxr> you are in the t2-trunk directory and run sudo ./scripts/Build-Target ?? 16:06 < write_erase> olivier@blackpearl:~/avr32/t2-trunk$ sudo ./scripts/Build-Target -cfg myavr 16:08 < rxr> ok 16:08 < rxr> ok - do you have a place to place the full log ? 16:08 < rxr> btw the uImage probaly is not there because you don't have uboot in the generic embedded package selection 16:09 < rxr> so near the O gzip and X mplayer in the pkgsel rules add: X u-boot 16:09 < write_erase> oh ... I'm stupid for uImage... should have guessed 16:11 < write_erase> For testing, I only have : X 00-dirtree X linux-header X linux26 X binutils X gcc X u-boot #X quagga X dropbear 16:11 < write_erase> u-boot just added 16:12 < rxr> ok - maybe I can reproduce the missing acpi file indeed 16:12 < write_erase> Is gcc required for the toolchain ? and linux-header too ? 16:12 * rxr rechecking 16:12 < write_erase> or I can remove them ? 16:12 < rxr> no - they are required for the toolchain 16:12 < rxr> linux-header defines the kernel ABI parts 16:13 < rxr> and the gcc package is used to bootstap a fresh and up-to-date cross compiler in the beginning of the build 16:13 < rxr> this does not mean you have to include them on your sd cards 16:13 < rxr> just needed to build all the other stuff 16:13 < write_erase> ok ... Now I have uboot added , how can make that work ? 16:13 < write_erase> I mean, just recompiling ? and the u-boot will be automagicaly used ? 16:16 < rxr> yes, a rebuild will first build u-boot and then the linux kernel again 16:16 < rxr> (if it still failed before) 16:17 < rxr> hm - I see 2 bugs here 16:17 < rxr> first the build does not abort on missing patch files due to the patch filter of the linux kernel package 16:17 < rxr> and second there are patches from D tags currently not activated 16:17 < rxr> with u-boot your kernel should buildl nontheless 16:18 < rxr> though I'll have to hunt those bugs .... :-( 16:20 < write_erase> You mean 9-avr32-fixes.patch and acpi-dsdt-initrd-v0.8.4-2.6.21.patch.. are not required ? 16:20 < rxr> acpi not 16:21 < write_erase> == 05/09/07 16:18:15 =[1]=> Finished building package linux26 16:21 < write_erase> But without avr32-fixes 16:21 < rxr> the avr32 one should is applied here: 16:21 < rxr> develop/t2-trunk/architecture/avr32/package/linux26/9-avr32-fixes.patch 16:21 < rxr> patching file arch/avr32/kernel/ptrace.c 16:21 < rxr> patching file arch/avr32/mm/cache.c 16:21 < rxr> patching file include/asm-avr32/dma-mapping.h 16:22 < rxr> there where some issues without that 16:22 < rxr> ah yes, CF (compact flash) does not work without hte patch 16:22 < rxr> and possible something strace related neither 16:23 < rxr> are all of the architecture/avr32/package/linux26/*.patch files not beeing applied ? 16:24 < write_erase> Apply patch /home/olivier/avr32/t2-trunk/architecture/avr32/package/linux26/9-avr32-fixes.patch ... 16:24 < write_erase> patching file arch/avr32/kernel/ptrace.c 16:24 < write_erase> Apply patch /home/olivier/avr32/t2-trunk/download/mirror/a/acpi-dsdt-initrd-v0.8.4-2.6.21.patch ... 16:24 < write_erase> sed: can't read /home/olivier/avr32/t2-trunk/download/mirror/a/acpi-dsdt-initrd-v0.8.4-2.6.21.patch: No such file or directory 16:24 < write_erase> Redefining VERSION ... 16:24 < write_erase> It's ok , althought there are 2 attempts for the same patch 16:25 < rxr> so the avr32-fixes is applied correclty, just the acpi one (what should not be tried in the first place) isn't ? 16:25 * rxr grabing a fresh coffee 16:33 < rxr> so - re 16:38 < write_erase> minit.c:19:17: error: fmt.h: No such file or directory 16:38 < rxr> guess it misses some dependency 16:39 < write_erase> It's a long way to compile T2 ... sorry for annoying you 16:39 < write_erase> Doesn't T2 manage dependancies ? 16:39 < rxr> yes, but when a target selects something minimal it's suppost that the target knows what it does 16:41 < write_erase> ok ... Understood 16:41 < rxr> ok, fmt.h usually comes from libowfat 16:41 < rxr> it should be selected in the embedded build 16:41 < rxr> was it build already ? 16:42 < write_erase> Nope 16:42 < rxr> really not ? 16:42 < write_erase> Not present in my package configuration 16:42 < rxr> oh, did you remove it ? 16:44 < rxr> btw. yes we know the scripts/Config is too slow, that is why we already started to rewrite it ... 16:44 < rxr> (just so you do not run away due to the annoyingly slow Config, ... :-))) 16:44 < rxr> write_erase: I now started a generic-embedded-avr32 test-build on my side as well 16:47 < write_erase> great 16:55 < CIA-9> rene * r23981 /trunk/scripts/config-functions.in: 16:55 < CIA-9> * removed indenting spacing from a config-functions.in code-block that 16:55 < CIA-9> had an enclosing conditional untl the lua accelerator was introduced 17:04 < CIA-9> rene * r23982 /trunk/target/embedded/config.in: 17:04 < CIA-9> * removed duplicate pkgconfig selection in the embedded target 17:04 < CIA-9> * improved the embedded target to only offer the dietlibc choice 17:04 < CIA-9> if the dietlibc is actually available (for the CPU architecture) 17:15 -!- write_erase [n=olivier@81.80.156.36] has quit ["Leaving"] 17:49 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 17:49 -!- emte_ [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 18:06 < LMJ> moin moin 18:07 < rxr> moin moin LMJ 18:08 < LMJ> how was your weekend ? 18:08 < rxr> ok nice, every timeslot filled with susan's friends in south germany around the wedding celebration, but fine 18:09 < rxr> in fact in 2 months we'll be down for the next wedding to celebrate ... 18:11 < LMJ> ;) 18:12 < LMJ> at the end of the summer, one of my friend choose me to be his groomsman 18:15 < rxr> :-) at this wedding susan was the maid of honor 18:21 -!- emte_ [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 18:21 -!- emte__ [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 18:25 [Users #t2] 18:25 [@ChanServ] [ emte__] [ mtr ] [ sepp_] [ valentin] 18:25 [ CIA-9 ] [ idealm] [ R4gnar0k] [ Stelz] 18:25 [ Dallur ] [ LMJ ] [ rxr ] [ TobiX] 18:25 -!- Irssi: #t2: Total of 13 nicks [1 ops, 0 halfops, 0 voices, 12 normal] 18:47 -!- write_erase [n=olivier@bon13-2-82-237-125-220.fbx.proxad.net] has joined #t2 18:53 < rxr> hi write_erase 18:54 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2 18:55 < misl> hi folks 18:56 < rxr> hi one 18:58 < misl> hi two ;-) 18:58 < write_erase> hi ... 18:58 < rxr> .oO (the borgs of Star Trek come to mind) 18:58 < write_erase> rxr, I"m adding all dependencies I removed 18:58 < rxr> misl: how are you ? 18:59 < misl> rxr: fine thx 18:59 < misl> how about you 18:59 < misl> how did LinuxTag go ? 19:04 < rxr> it's in the future 19:04 < rxr> beg. of June 19:05 < rxr> if you mean last weekend, we have been on a wedding celebration in south germany :-) 19:07 < write_erase> !> checking build system type... i686-nocross-linux-gnu 19:07 < write_erase> !> checking host system type... Invalid configuration `avr32-t2-linux-uclibc': machine `avr32-t2' not recognized 19:07 < write_erase> !> configure: error: /bin/sh ./support/config.sub avr32-t2-linux-uclibc failed 19:08 < write_erase> While readline compilation . 19:08 < rxr> which package is this ? 19:08 < write_erase> readline 19:09 < R4gnar0k> moin moin 19:09 < rxr> mon moin R4gnar0k 19:09 < R4gnar0k> how's it going? 19:09 < rxr> fine, thanks - had a sort of "relaxing" extended weekend, and you ? 19:10 < R4gnar0k> I wish :) ohwell the normal weekend was relaxing too 19:12 < rxr> write_erase: I have a local tweak to fixup more of those avr32 not recognized in package/base/automake/parse-config 19:12 < rxr> I did not yet commit it as it did not yet pass the 2800 package reference build ... 19:13 < write_erase> ok, I remove readline until you fix it 19:13 < rxr> wauit 19:13 < rxr> wait 19:13 < rxr> you have some program experience and read diffs ? 19:13 < rxr> --- package/base/automake/parse-config (revision 23978) 19:13 < rxr> +++ package/base/automake/parse-config (working copy) 19:13 < rxr> @@ -41,6 +41,7 @@ 19:13 < rxr> } 19:13 < rxr> 19:13 < rxr> # some packages do not include an up-to-date config.{sub,guess} 19:13 < rxr> -if ! atstage toolchain && hasflag AUTOMAKE-QUIRK; then 19:13 < rxr> +if hasflag AUTOMAKE-QUIRK || 19:13 < rxr> + [ $arch = 'avr32' ] || [ $arch = 'bfin' ]; then 19:13 < rxr> hook_add preconf 3 automake_fixup 19:13 < rxr> fi 19:14 < rxr> just perform the tiny edit and give it a go 19:14 < rxr> oh, this must be blackfin even after the latest rename ... 19:14 < rxr> anyway, you just care about avr32 anyway :_) 19:15 < write_erase> build/myavr-7.0-trunk-mediavr-avr32-ap7000-cross/var/adm/packages/readline should not be present 19:15 < write_erase> can I remove them by hand ? 19:15 < rxr> remove what ? 19:15 < write_erase> the file that should not be present 19:15 < write_erase> as i removed readline in my package selection 19:16 < rxr> you you get that on the end of the build I guess? 19:16 < write_erase> sorry "should not be present" is a T2 warning message. 19:16 < write_erase> yes 19:16 < rxr> at the bulid end you could ignore it, it is just indicating that the files should not be part of the final system 19:16 < rxr> but this tiny meta data files, especially in your first-try t2 build will not matter 19:18 < write_erase> The image is located at ..... Done ! 19:18 < write_erase> Great ! 19:20 * rxr presses the thumbs it also works :-))) 19:20 < rxr> I guess you disabled openssl ? 19:20 < rxr> as I just notice it does not build in avr32/generic-embedded 19:21 < write_erase> hehe yes for my 1st attempt 19:22 < write_erase> are the binaries stripped ? 19:24 < write_erase> rxr, What's the next step for booting NGW100 ? 19:29 < rxr> I would test singles binaries on the board first 19:32 < write_erase> Do atmel use the same libc ? 19:32 < CIA-9> rene * r23983 /trunk/package/security/openssl/openssl.conf: 19:32 < CIA-9> * fixed openssl to build the generic C code for all architectures with 19:32 < CIA-9> the associated endianess defines when assembly optimizations are not 19:32 < CIA-9> available (unified mips, arm that already ussing the generic code) 19:32 < rxr> yes, but only because no newer uclibc is available for avr32 19:32 < rxr> usually t2 builds use way newer libraries and compiler than the aging vendor ones 20:02 < rxr> vastly revisited: http://www.t2-project.org/documentation/ 20:02 < rxr> more work tomorrow 20:02 * rxr hading home 20:02 < rxr> write_erase: I'll be back tomorrow morning for your avr32 questions 20:03 < rxr> feel free to write me a mail to rene @ exactcode . de if you like to queue something up 20:03 < rxr> cu all 20:10 -!- d-marc [n=d-marc@HSI-KBW-091-089-002-063.hsi2.kabelbw.de] has joined #t2 20:10 < d-marc> hello 20:14 < Stelz> cu rxr, hi d-marc :) 20:55 -!- d-marc [n=d-marc@HSI-KBW-091-089-002-063.hsi2.kabelbw.de] has quit ["KVIrc 3.2.4 Anomalies http://www.kvirc.net/"] 21:36 -!- emte__ [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 21:37 -!- emte__ [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 22:33 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has quit ["Chatzilla 0.9.75 [Firefox 2.0.0.3/2007030919]"] 22:35 -!- BusError [n=buserror@80.68.82.10] has joined #t2 22:35 < BusError> Hi guys; been recommended this channel/project :D 22:37 < sepp_> hi BusError 22:38 < BusError> When I looked at the website I didn't find a comparison with stuff like 'buildroot' etc... is it a similar goal at all ? 22:38 < BusError> must say I'm wary of buildroot; and I'm looking for a 'stable' alternative :D 22:40 < sepp_> i dont know what buildroot is :/ 22:41 < sepp_> ok read it :) 22:42 < BusError> buildroot is 'the' standard thing for making embedded distros. openembedded and crew use that, as well as many others. 22:43 < sepp_> openembedded was the stuff used for openzaurus and that? 22:43 < BusError> problem is, it's a moving target. no 'freezes' etc, it's impossible to use. You do a 'svn update' to try to get a fix for a problem and it breaks something else :/ 22:43 < write_erase> BusError, There are no code freeze or releases ? 22:44 < BusError> nope, none that I know of anyway.. 22:44 < sepp_> there is a stable and devel t2 version ;) 22:45 < BusError> yesh, I see that, how novel :D 22:45 < sepp_> :) 22:47 < BusError> so, given say a PowerPC405GP, could I build a distro with a 2.4.x, my own sicko patches, and apps etc and have a reasonable way to configure it ? Like which toolchain, which fielsystem format etc ? 22:47 * BusError dreams :D 22:48 < sepp_> 2.4.x kernel? 22:48 < BusError> yup 22:48 < sepp_> you can do that, not very hard 22:49 < BusError> with T2 as it is then ? 22:49 < sepp_> dont know why one would want to use 2.4 but ... not hard 22:49 < BusError> I have a bunch of drivers that I don't want to port for no good reasons :D 22:50 < BusError> FPGA related stuff, DSP interface etc 22:51 < BusError> right now the distro is hardhat; I want to get rid of it 22:52 < sepp_> well, default is udev and afaik that will not work with 2.4 - or was it something else? 22:54 < sepp_> the rest is easy, i use it for that reason, only the apps i want, no 500 filesystems etc 22:54 < sepp_> dont need 20 mail clients and 4 desktop/wms ... 22:55 < BusError> well I do embedded stuff, I don't want /any/ email client or desktop :D 22:55 < sepp_> but the idea is the same 22:58 < sepp_> you choose what you want and build it in a sane way 22:58 < BusError> So T2 is Linux 2.6 only ? 22:59 < sepp_> there is 2.4 but i dont care about embedded stuff or 2.4 so i do not know about that 22:59 < sepp_> in the t2 tree 23:00 < BusError> oh ok 23:00 < write_erase> sepp_, what do you use T2 for ? 23:02 < sepp_> small desktop with e17 and no silly deps for optional gnome/kde features(crap) i dont like 23:03 < write_erase> ok 23:07 < sepp_> something that easily fits onto 1 CD, has all the things i use but nothing more 23:10 < write_erase> You compile for i386 ? 23:10 < sepp_> x86_64 23:20 < CIA-9> sebastian * r23984 /trunk/package/games/wesnoth/wesnoth.desc: * updated wesnoth (1.2.1 -> 1.2.4) 23:26 < CIA-9> sebastian * r23985 /trunk/package/games/sauerbraten/ (. sauerbraten.cache sauerbraten.conf sauerbraten.desc): * added sauerbraten (2007_04_15) - A 3d open source first person shooter 23:35 < sepp_> and sometimes i play stupid games :p 23:49 < write_erase> How can I get the full list of avaliable T2 package 23:50 < R4gnar0k> ls package/* 23:51 < write_erase> oh, no lxdoom .... 23:53 < sepp_> 2842 packages in my tree --- Log closed Thu May 10 00:00:36 2007