**** BEGIN LOGGING AT Wed Jun 08 23:59:56 2005 Jun 09 00:07:31 morning all Jun 09 00:10:41 good morning RP Jun 09 00:11:52 morning RP Jun 09 00:12:41 hi all Jun 09 00:12:50 hey hrw & do13 Jun 09 00:13:00 wifi is nice thing to have Jun 09 00:13:06 :) Jun 09 00:13:51 now I need: terminal which know what is Alt, Ctrl and will give me Alt+1 in irssi over screen over ssh Jun 09 00:13:59 better www browser Jun 09 00:14:05 fixed opie-irc Jun 09 00:14:11 hi koen Jun 09 00:14:36 hrw: esc 1 Jun 09 00:14:43 the cancel button should also work Jun 09 00:14:53 koen: iirc I have Alt key on c7x0.. Jun 09 00:15:03 next to '-' Jun 09 00:15:19 koen: but in opie-console it does not work Jun 09 00:15:37 hrw, koen: The two japanese keys are ctrl+alt. I suspect something is wrong with opie's handling of them though Jun 09 00:15:38 on vt1 also (when did ssh somewhere -t screen -RDU) Jun 09 00:16:04 probably also COLUMNS/ROWS are not set ok on vt1 Jun 09 00:16:57 RP: they work with rxvt Jun 09 00:17:42 hi all Jun 09 00:18:48 ok. opie-irc bugs raported Jun 09 00:19:08 ~spell raported Jun 09 00:19:11 possible spellings for raported: reported sported ported reparteed reputed reporter riposted deported resorted retorted parted rapports spotted repeated potted rapped rapport ratted rooted rotted rapporteur replotted raped rapport's rated reporters roted reports roosted spouted spurted parroted purported pouted report reportedly rioted routed rafted ranted ... Jun 09 00:19:34 ~botsnack Jun 09 00:19:34 :), hrw Jun 09 00:20:05 heh.. and now I again have >10 open bugs in opie Jun 09 00:20:50 I don't suppose anyone here knows anything about the correct behaviour of the floating point code? Jun 09 00:31:50 RP: you mean the nwfpe? Jun 09 00:31:57 don't know anything about that, sorry Jun 09 00:36:11 RP: I've never looked into this. I guess with our softfloat toolchain this doesn't affect us. Jun 09 00:39:46 hi people! can someone explain to me how the version sorting is done in OE? i.e. why '2.6.11-bk1' is considered newer than '2.6.11.11'? Jun 09 00:43:31 d013: Linux won't compile without an emulator so it will affect us Jun 09 00:47:37 why does the mailman ring when I'm in the shower? Jun 09 00:48:16 RP: I just received your evil counterpart: hx4700 Jun 09 00:48:59 RP: what do you think about a new MACHINE, like ipaq-pxa-270? Jun 09 00:59:53 re Jun 09 01:02:21 RP: You can compile the kernel without fp emulation Jun 09 01:05:17 koen: We could probably just fix ipaq-pxa-2.6 to work, or just use c7x0... Jun 09 01:05:37 d013: That's not what the comments in Kconfig say :-/ Jun 09 01:06:26 RP: pxa270 kernels can't run on pxa255 machines Jun 09 01:06:44 RP: I want 1 'magic' kernel for the pxa270 machines Jun 09 01:06:58 and 1 for the pxa25y ones Jun 09 01:07:14 koen: I always compile the kernel seperately. You install it seperately on the hx2750 anyway as we don't touch the mtd yet... Jun 09 01:07:34 koen: Someone should make kernels work on both pxa270 and pxa255... Jun 09 01:07:39 It should be straightforward Jun 09 01:07:40 yeah Jun 09 01:07:57 I keep meaning to do that... Jun 09 01:08:13 RP: Kernel: arch/arm/boot/zImage is ready (15K smaller) Jun 09 01:08:42 RP: and I'm not really OK with running a kernel plastered with OZ propaganda on an ipaq running !OZ Jun 09 01:09:33 koen: I know. I have said I want to rename that kernel. Just a question of to what... Jun 09 01:09:51 linux-RP Jun 09 01:10:02 linux-chav ;) Jun 09 01:10:42 do13: Someone should really fix the comments if that kernel will run... Jun 09 01:11:26 It's untested. Maybe after work I will test this. Jun 09 01:13:57 do13: I'll be interested to know if it works. I might have a try at building one here later on... Jun 09 01:15:17 I have most of a fix for nwfpe. I just need someone to answer some questions on what the correct bahaviour should be under certain conditions... Jun 09 01:16:44 * koen wonders when pocketpc will learn what 'Antialiasing' means Jun 09 01:17:17 koen: pocketpc has that quite a long time already Jun 09 01:18:28 Arjan: it *says* it has it Jun 09 01:18:44 Arjan: the menus still look hidious Jun 09 01:19:20 koen: I never tried it actually, I hate it :) Jun 09 01:35:05 anyone seen a problem with openssh where only root can ssh to a remote server, and non-root gets Host key verification failed. ? Jun 09 01:37:08 jacques: seems a problem with incorrect rights on the files. Jun 09 01:37:25 yeah I'm checking that but haven't found anything Jun 09 01:37:34 jacques: try to delete the offending host key from ~/.ssh/known_hosts if you're sure it's the right machine on the other end? Jun 09 01:38:24 I was using dropbear and it was working other than the dropbear sshd was disconnecting me with "Disconnecting: Protocol error: expected packet type 31, got 82" so I switched to openssh Jun 09 01:38:45 this is between machines on my home network Jun 09 01:38:55 and I have tried removing known_hosts Jun 09 01:39:13 if I don't have it, I get " Host key verification failed." Jun 09 01:39:32 if I copy root's, I get : Jun 09 01:39:35 Permission denied, please try again. Jun 09 01:39:35 Permission denied, please try again. Jun 09 01:39:35 Permission denied (publickey,password,keyboard-interactive). Jun 09 01:40:39 jacques: Did you set the option to reject anything that isn't in the known_hosts ? Jun 09 01:41:48 no, but anyway, it works for root Jun 09 01:42:24 even when I remove root's known_hosts file, it works as expected and generates a new one Jun 09 01:43:17 Your host key is correct ? Can you connect to something else ? Jun 09 01:44:38 as root I can connect to whatever I want Jun 09 01:45:16 Can you do a strace as normal user to check if an open syscall fails ? Jun 09 01:45:46 yeah I know, but I have to find an old strace ipkg or build it native first Jun 09 01:46:00 current oe-built strace does not work on my platform (it used to) Jun 09 01:48:52 jacques: Let see. It tries ~/.ssh/ssh_config, then /etc/ssh/ssh_config. Are the permissions on those files and dir corrects ? Jun 09 01:49:55 jacques: could this be related to your problem: http://www.experts-exchange.com/Security/Linux_Security/Q_20877265.html Jun 09 01:50:56 jacques: Then I suggest to try a ssh -vvv for a connection that fails. Jun 09 01:52:19 ah OK I'd only tried -vv just a sec Jun 09 01:53:31 nothing obvious in the debug info - give me a few minutes to build strace Jun 09 02:03:26 hmm, my native-built strace isn't working either - I had found an olf ipk which worked a couple of days ago - maybe there's something wrong with this version of strace Jun 09 02:04:19 jacques: i think atm that ssh can't open /dev/tty. what are your permissions there? Jun 09 02:04:58 crw-rw--w- 1 root root 5, 0 Jun 7 01:55 /dev/tty Jun 09 02:05:34 yes!!!! Jun 09 02:05:37 that was it! Jun 09 02:05:42 THANKS pH5 Jun 09 02:05:45 np :) Jun 09 02:06:17 apparently dropbear doesn't use /dev/tty Jun 09 02:08:54 03koen 07 * r1.3499.10.1 10openembedded/packages/maemo/ (2 files in 2 dirs): Add hildon-libs, part of the maemo.org suite Jun 09 02:13:38 jacques: what machine you use? Jun 09 02:15:30 hrw|work, nslu2 (armeb) Jun 09 02:15:44 openslug dist currently Jun 09 02:15:50 hrw|work: I thought about a nice to have feature for bitbake : drop to interactive shell if a task failed. Jun 09 02:16:44 AHA!! Jun 09 02:17:08 they broke strace for armeb between 4.5.9 and 4.5.11 Jun 09 02:17:57 linux/arm/syscallent.h: Fix 113 entry (syscall, not vm86old) Jun 09 02:18:04 in 4.5.11 changes is probably it Jun 09 02:18:11 I can try 4.5.10 to be sure Jun 09 02:18:48 too bad oe has been so "efficient" lately removing "old" versions of things Jun 09 02:19:06 only 4.5.11 remains Jun 09 02:20:14 well, I can try 4.5.10 if sourceforge ever responds Jun 09 02:25:37 jacques: it is 2.4 or 2.6 based? (kernel) Jun 09 02:25:56 hrw|work, openslug is 2.6 based - currently at 2.6.11.2 Jun 09 02:26:24 armv5teb arch Jun 09 02:26:28 jacques: update udev 058 to newer Jun 09 02:26:58 hrw|work, you mean for the tty perms problem? Jun 09 02:27:21 I think we may not be using udev :-\ Jun 09 02:27:32 jacques: for /dev/ permissions at all Jun 09 02:27:47 jacques: I had most of /dev/ entries as 660 Jun 09 02:27:54 ah Jun 09 02:33:13 OK, strace 4.5.10 is also broken - so 4.5.9 is the last strace that works for armeb Jun 09 02:34:05 jacques: You could try to remove the offending syscall modification to see if it is the culprit arch brokeness. Jun 09 02:35:16 apparently it was somethingin 4.5.10 Jun 09 02:36:38 and I don't see anything obvious looking at the changes from 4.5.9 to 4.5.10 Jun 09 02:37:19 strace-devel looks like 98% spam Jun 09 02:37:26 (list) Jun 09 02:44:25 I send a mail to the strace-devel list, I am not subscribed so it may bounce Jun 09 02:44:36 and nobody may read it because of the spam problem :-\ Jun 09 02:49:50 jacques: Try to contact roland at redhat dot com (from the changelog). Jun 09 02:50:22 proti, OK I'll send the email to him too Jun 09 03:23:26 morning Jun 09 03:25:42 hi mickeyl Jun 09 03:31:16 morning mickeyl Jun 09 03:32:48 hi mickeyl Jun 09 03:41:12 hi all Jun 09 03:41:34 hi pb_ Jun 09 03:42:42 hi pb_ Jun 09 03:44:09 hi pb_ Jun 09 03:46:45 hi pb_ Jun 09 04:37:46 afternoon all Jun 09 04:37:54 hi ade|desk Jun 09 04:37:59 hey ade|desk Jun 09 04:43:08 afternoon Jun 09 04:43:25 hi reenoo Jun 09 04:44:02 hey bluelightning Jun 09 05:18:55 <[g2]> pb_ around ? Jun 09 05:29:38 03koen 07 * r1.3506 10openembedded/packages/maemo/ (2 files in 2 dirs): Add osso-gnome-vfs-2, port of the maemo.org suite Jun 09 06:00:04 don't suppose I sent my CV for anyone to review back in oct 2003? Jun 09 06:06:59 anyone in here has an opinion about needing an initrd or not when booting from mmc with a kernel with both mmc suppport and ext2 support builtin ? Jun 09 06:11:44 03mickeyl * r245 10bitbake/lib/bb/shell.py: Jun 09 06:11:44 Greatly improve pastebin functionality: Jun 09 06:11:44 The BitBake shell now features a generic output buffer for the last 10 commands. Jun 09 06:11:44 You can dump an overview about the buffered commands with 'buffers' and dump Jun 09 06:11:44 the contents of a buffer with 'buffer '. Likewise, 'pastebin ' uploads Jun 09 06:11:45 the contents of the said buffer. NOTE: 'lasterror' is still available, but has been Jun 09 06:11:47 renamed to 'lastlog'. The old pastebin command is still available with 'pastelog'. Jun 09 06:13:33 * BigAl does the happy dance. Jun 09 06:14:25 BigAl: LCD? Jun 09 06:14:33 mickeyl: Yup! Jun 09 06:14:37 yahoo Jun 09 06:14:42 ~praise BigAl Jun 09 06:14:43 All hail BigAl! Jun 09 06:14:48 BigAl: what made it work? Jun 09 06:14:57 mickeyl: It was that lcdtg tchip that was fucking it up, it wasn't playing nice on the ssp bus. Jun 09 06:15:09 aah Jun 09 06:15:26 I copied the really nasty sharp code over that bit-bangs the bits into it and it worked. Jun 09 06:15:27 BigAl: yay! :) Jun 09 06:15:33 excellent Jun 09 06:15:43 that's a brakethrough Jun 09 06:15:48 eh Jun 09 06:15:50 breakthrough even Jun 09 06:15:53 ~spell breakthrough Jun 09 06:15:55 'breakthrough' may be spelled correctly Jun 09 06:16:26 RP: It looks like you got it working without using the sharp bit banging, how did you manage that? Jun 09 06:16:50 BigAl: There is bit banging code in there - I just cleaned it up Jun 09 06:16:51 don't we still do the bitbaning ? Jun 09 06:16:56 yeah, but elsewhere Jun 09 06:17:00 http://pastebin.com/297667 Jun 09 06:17:06 ~hail mickeyl Jun 09 06:17:07 * ibot bows down to mickeyl and chants, "I'M NOT WORTHY!!" Jun 09 06:17:27 BigAl: I'd be interested to see the changes you need Jun 09 06:17:30 koen: IT WORKS! :)) Jun 09 06:18:26 * koen rigs a minimo build to fail Jun 09 06:18:37 lets see how many browsers I can choke :) Jun 09 06:18:41 hehe Jun 09 06:19:33 the hostname bit is still a bit shakey Jun 09 06:19:37 RP: the bit banging you mean is in sharpsl, right? Jun 09 06:19:42 RP: sharp_params even Jun 09 06:19:55 koen: does your environment not define $HOST ? Jun 09 06:20:10 RP: I needed to add in the bit-banging code from pxa_ssp.c that toggles the gpio lines individually to get the lcdtg chip working, although in the corgi_ssp.c file from you it looks like you just set up some paramaters and use the ssp bus normally. Jun 09 06:20:23 mickeyl: apparently not Jun 09 06:20:27 mickeyl: No, there is some bit banging code in corgi_lcd.c Jun 09 06:20:32 koen: can you check $HOSTNAME please Jun 09 06:20:37 RP: ah right Jun 09 06:20:45 koen@bitbake:~/OE/build$ env | grep HOSTkoen@bitbake:~/OE/build$ Jun 09 06:20:51 nothing whatsoever Jun 09 06:20:58 mickeyl: `hostname` Jun 09 06:21:01 d'oh Jun 09 06:21:12 yeah, i'll probably better just call hostname Jun 09 06:21:13 BigAl: Correct, but there is an extra layer of bit manipulation does still go on in corgi_lcd Jun 09 06:21:35 RP: I'm using that as well, but it was the SSP stuff which was failing. Jun 09 06:21:57 BigAl: You did map the gpio's correctly? Jun 09 06:22:33 BigAl: I'm not entirely sure which bits of code we're talking about - I'd be very interested to see the patch, then perhaps I can comment more Jun 09 06:22:34 RP: I managed to get some data out of the MAX1111, so I thought it was working ok. Jun 09 06:22:57 mickeyl: Are you still interested in bluetooth? Here is a picture of my soldered module http://www.do13.de/openzaurus/pictures/tosa_bluetooth.jpg Jun 09 06:23:16 BigAl: Each ssp device is essentially independent - just because one works, doesn't mean they all will Jun 09 06:23:46 do13: very neat :) Jun 09 06:23:49 RP: I'll double check the GPIO mappings for the lcdtg chip then. Jun 09 06:24:12 BigAl: I'm thinking of the ssp enable/disable pins Jun 09 06:24:30 do13: whoah, that's cool Jun 09 06:25:31 03mickeyl * r246 10bitbake/lib/bb/shell.py: Use socket.gethostname() instead of os.environ["HOST"]. That should work on all systems. Jun 09 06:26:35 do13: looks beyond my capabilities :) Jun 09 06:27:21 No soldering skills? Jun 09 06:28:25 not at all :/ Jun 09 06:39:29 mickeyl: out of interest, any movement on opie/x11 stuff ? Jun 09 06:40:08 night all Jun 09 06:40:20 'night bluelightning Jun 09 06:40:30 'night bluelightning Jun 09 06:45:22 mallum: iirc one person work on it Jun 09 06:45:31 mallum: bero started to port some things to qt4/x11. Jun 09 06:45:47 mallum: there's no real masterplan atm. though Jun 09 06:46:29 When we go qt4, I want to get rid of libqpe. That makes us more flexible Jun 09 06:46:31 less compatible though Jun 09 06:46:33 ... Jun 09 06:47:24 mickeyl: how much code is in libqpe now? maybe someone put some wrappers there and it will be ok? Jun 09 06:47:37 campatible with the 0,1% of the users Jun 09 06:47:40 aargh Jun 09 06:47:43 incompatible Jun 09 06:48:03 hrw|work: O{device,resource} etc Jun 09 06:48:37 hrw|work: http://doc.trolltech.com/qtopia1.6/html/qtopiaclasses.html Jun 09 06:48:47 quite some code Jun 09 06:48:54 mickeyl: okey :) good luck Jun 09 06:49:47 mallum: one last note about that... when I get my Nokia 770 I want to port Wellenreiter to it, so I will definitly go Qt4/X11 to make that happen :) Jun 09 06:51:44 mickeyl: *if* you get if ;) Jun 09 06:51:51 hah Jun 09 06:51:52 s/if/it/ somewhere Jun 09 06:52:06 why shouldn't i get it? Jun 09 06:52:21 not necessarily for the devel price Jun 09 06:52:44 although i really would deserve one Jun 09 06:52:44 heh Jun 09 06:52:57 port wellenreiter to gtk :) Jun 09 06:53:46 that'd be too much of an undetake Jun 09 06:54:01 my complete protocol dissection engine is based on qt utility classes Jun 09 06:54:48 if you want to keep the original gui, it might be worth it to hook wellenreiter up to osso Jun 09 06:55:15 that gives you lots of integration stuff Jun 09 06:55:29 i'm a c++ guy though Jun 09 06:55:38 gtkmm? Jun 09 06:55:39 mickeyl: whats Wellenreiter ? Jun 09 06:55:45 kismet for opie Jun 09 06:55:53 "waverider" Jun 09 06:56:02 mallum: THE Wireless network scanner for Opie. http://www.vanille.de/projects/wellenreiter.html Jun 09 06:56:35 it's much better than kismet though ;) Jun 09 06:56:38 mickeyl: ah cool :) Jun 09 06:57:41 mickeyl: can you look into opiestumbler to check does it will find network without broadcasted ssid? Jun 09 06:58:10 wellenreiter and opie-networksettings are able to find my home network but stumbler not Jun 09 06:58:21 hrw|work: It won't Jun 09 06:58:32 opiestumbler bases on active scan Jun 09 06:58:41 mickeyl: which need ssid? Jun 09 06:58:41 only APs which 'want' to response will be shown Jun 09 06:58:46 ok Jun 09 06:59:05 anyway networksettings found my AP so no problem Jun 09 06:59:16 according to ljp it's fluff Jun 09 06:59:20 i should remove that feature Jun 09 06:59:21 heh Jun 09 06:59:24 and bloat Jun 09 06:59:27 I need to find some time and flash something usefull to that linksys... Jun 09 06:59:37 mickeyl: remove everything !PIM Jun 09 07:00:18 ya Jun 09 07:01:12 my blue-blackbox works but I feel strange when cannot hack into it... Jun 09 07:01:59 since br1 added the kernel there shouldn Jun 09 07:02:08 t be anything hindering us to finish a distro Jun 09 07:02:36 mickeyl: sooner or later I will look into openwrt and cooperate with br1 to get something working Jun 09 07:02:50 there are many noncompatible revs of wrt54g(s) Jun 09 07:03:32 hrw|work: did you add an sd card reader yet? Jun 09 07:04:14 hrw|work: http://kiel.kool.dk/ Jun 09 07:04:42 koen: first I have to get it working under !linksys Jun 09 07:05:14 and that sdmod is only for wrt54g not wrt54gs Jun 09 07:06:33 mickeyl: I was playing with wellenreiter recently - its a very nice piece of software. Can't wait to interface it to my new gps receiever :) Jun 09 07:06:50 RP: thanks :) Jun 09 07:09:04 morning Jun 09 07:10:13 [g2]: hello Jun 09 07:10:16 hi mickeyl Jun 09 07:10:21 greetings pb_ Jun 09 07:10:46 koen: sd card reader is a nice addon. Maybe I can solder bluetooth into my wrt54g:) Jun 09 07:10:56 :) Jun 09 07:11:10 do13: do you have vgaout already? Jun 09 07:11:40 Do you have a small module or chip? Jun 09 07:11:55 do13: I only saw that hack Jun 09 07:12:32 its amazing what people do with those boxes... sdcard, vgaout, sound out (covox), serial ports, nokia 3310lcdfb Jun 09 07:13:25 vga out? Jun 09 07:14:48 http://wiki.openwrt.org/OpenWrtDocs/Customizing Jun 09 07:15:01 http://www.duff.dk/wrt54gs/pics/Complete_VGA_Setup.jpg Jun 09 07:15:15 holy crap Jun 09 07:16:56 heh cool Jun 09 07:17:58 but dont show it to zaurus users... they will start to ask "does it run opie? can I run openoffice on it? can it bootstrap debian?" etc crap Jun 09 07:18:13 can it run pxaxROM ? Jun 09 07:18:18 pdaXROM even Jun 09 07:18:38 can I run debian inside X/qt inside sharp cra^h^h^hrom? Jun 09 07:18:59 which is altbooted from oz 3.5.3 on 5000D Jun 09 07:19:06 anyone know if there's a way to (easily) prevent bitbake from downloading new cvs snapshots every day? I.e. only use the last downloaded snapshot. Jun 09 07:19:17 dmag: CVSDATE Jun 09 07:19:24 dmag: CVSDATE=20050609 Jun 09 07:19:29 cvsdate is your friend :) Jun 09 07:19:34 CVSDATE :P Jun 09 07:19:59 cool, but what happens when it needs a new file (and it's past the CVSDATE?) Jun 09 07:20:07 it won't download it :) Jun 09 07:20:41 * koen hugs bitbake -i Jun 09 07:21:22 so if it's past CVSDATE, and there is *no* previous download, it will fail? Jun 09 07:21:36 no, it will download it Jun 09 07:21:44 for the CVSADTE specified Jun 09 07:22:02 it will use CVS then (instead of CVS snap)? Jun 09 07:22:24 <[g2]> pb_, hey :) Jun 09 07:34:21 will CVSDATE work with svn or do I do something different? Jun 09 07:35:52 yes, it will work Jun 09 07:37:03 <[g2]> pb_, I was looking for an arm assembler guru earlier Jun 09 07:37:22 <[g2]> if I get stuck I may ping you again for a very simple issue Jun 09 07:40:36 [g2]: I'd just ask as someone else might be able to help. I have some knowledge of arm assembler for example... Jun 09 07:46:27 03koen 07 * r1.3507 10openembedded/packages/maemo/osso-thumbnail_0.3.1-1.bb: osso-thumbnail_0.3.1-1.bb: fix packaging snafu, thanks to Florian for spotting it Jun 09 07:49:31 [g2]: righto Jun 09 07:49:46 03koen 07 * r1.3508 10openembedded/packages/maemo/ (hildon-home_0.8.20-2.bb osso-thumbnail_0.3.1-1.bb): Add hildon-home, part of the maemo.org suite, courtesy Florian Boor Jun 09 07:52:00 <[g2]> Ok here goes.... I'm working with ep1220 who is ready to load the mini icache directly via JTAG on the IXP420 Jun 09 07:52:30 <[g2]> our simple assemble program is along the lines of Jun 09 07:52:59 <[g2]> init: Jun 09 07:52:59 <[g2]> mov r1,#0xC800 Jun 09 07:52:59 <[g2]> lsl r1,r1,#16 Jun 09 07:52:59 <[g2]> mov r2,#0x4000 Jun 09 07:53:02 hi Jun 09 07:53:20 <[g2]> and there are some more lines Jun 09 07:54:16 <[g2]> currently as (for the armv5) doesn't like lsl or orr instructions Jun 09 07:54:22 <[g2]> maybe we don't need the #16 Jun 09 07:54:25 <[g2]> just 16 Jun 09 07:54:46 yah, there's no such instruction as "lsl" Jun 09 07:54:51 you want "mov r1, r1, lsl #16" Jun 09 07:55:36 I was just about to say that :) Jun 09 07:56:02 LSL only exists as a standalone instruction in Thumb, and the syntax is a bit different there. Jun 09 07:56:03 <[g2]> ok excellent that works.... one down Jun 09 07:56:24 ORR, though, is an instruction. what's the failing line you have in that case? Jun 09 07:56:24 <[g2]> how about orr r2,r1 Jun 09 07:56:32 ah, no, you need "orr r2, r2, 1" Jun 09 07:56:37 er, "orr r2, r2, r1" Jun 09 07:56:56 "orr r2, r1" would be a valid thumb instruction, but you need all three operands in arm state. Jun 09 07:56:59 <[g2]> ok... so is dest, source1, source2 Jun 09 07:57:09 <[g2]> thx again Jun 09 07:57:19 right Jun 09 07:59:55 <[g2]> ok and is r0 a special register ? Jun 09 08:00:12 no Jun 09 08:00:30 the only really special register is r15, which is the program counter. Jun 09 08:00:44 <[g2]> Error: invalid constant -- `mov r0,#0x1E0F' Jun 09 08:01:02 that constant is too wide to move in one go. You'll need to break it up, or load it from memory. Jun 09 08:01:17 MOV can only move eight bits at a time (though they can be located anywhere in the word). Jun 09 08:01:34 so: mov r0, #0x1e00; orr r0, r0, #0xf Jun 09 08:01:35 or something Jun 09 08:03:08 <[g2]> so it's the first 8 bits after the 0x then Jun 09 08:03:29 <[g2]> and they get aligned based on the zeros Jun 09 08:07:36 more or less, yeah. The actual rule is that MOV can move any 32-bit word so long as it contains a block of 24 consecutive zeros, or a block of 24 consecutive ones, and the block starts on an even-numbered bit. Jun 09 08:08:01 or, to put it another way, you get to choose an 8-bit constant, rotate it by a multiple of 2 bits, and optionally invert the result. Jun 09 08:08:03 <[g2]> pb_, THX a lot! Jun 09 08:08:15 <[g2]> clear as day Jun 09 08:08:22 heh Jun 09 08:08:27 <[g2]> really Jun 09 08:08:42 * [g2] has done too much assember :) Jun 09 08:08:50 <[g2]> just not ARM :) Jun 09 08:08:52 <[g2]> yes Jun 09 08:08:54 <[g2]> yet Jun 09 08:09:02 cool Jun 09 08:09:27 <[g2]> the program fully assembles now Jun 09 08:09:29 <[g2]> THX Jun 09 08:09:57 pb_: Did you have a chance to look that fpa datasheet out? Jun 09 08:13:45 RP: no, it's on the other side of town. I'll have to get it later this afternoon. Jun 09 08:14:00 pb_: ah, ok Jun 09 08:29:07 I think there is hope for the wifi loving zaurus owners: nintendo will install a gazillion hotspots in japan Jun 09 08:29:20 so sharp *might* consider butting wifi in a Z Jun 09 08:29:50 ehm Jun 09 08:29:57 putting in Jun 09 08:32:04 * RP ponders all the potential combinations of sharp and butting in a sentence together :) Jun 09 08:32:51 lol Jun 09 08:42:15 in oe, how do I know what Jun 09 08:42:23 what's going to be built? Jun 09 08:42:46 it builds what you need to build what you want Jun 09 08:42:50 when i build collie, how does it know what packages to build? Jun 09 08:43:06 It's not in handheld-conf, it's not in collie.conf.. Jun 09 08:43:40 (as an aside, it would be nice to be able to see out the dependancy tree..) Jun 09 08:44:00 what do you mean by "build collie"? collie is a machine; it's not something you can build. Jun 09 08:44:09 unless you are sharp, I suppose Jun 09 08:44:52 if you want to see what's going to be built without actually building it, you can use "bitbake --dry-run". Jun 09 08:45:01 good point. how do I build GPE? Jun 09 08:45:16 bitbake gpe-image Jun 09 08:46:05 aha. that helps.. Jun 09 08:47:29 can i dry-run from bitbake -i ? Jun 09 08:48:39 in other words, test a package before i build it without exiting bitbake? Jun 09 08:53:14 I'm not sure. You might have to take that up with shellmeister mickeyl. Jun 09 08:53:35 ok, thanks! Jun 09 09:01:24 cu Jun 09 09:03:50 cu monday Jun 09 09:04:38 bye Jun 09 09:34:35 morning kergoth Jun 09 09:34:43 hey Jun 09 09:34:50 all unpacked? :) Jun 09 09:35:26 just about. still lots of books and stuff to fill my bookshelf with Jun 09 09:37:09 not too bad then. Settled in to the new/old place? Jun 09 09:37:21 yep Jun 09 09:37:33 now need to figure out what to do to establish some cash flow Jun 09 09:37:56 ah, that problem :-/ Jun 09 09:38:49 packed? Jun 09 09:38:50 hi kergoth Jun 09 09:39:33 kergoth: win the lottery and start kergoth technologies inc. Jun 09 09:56:23 hi pb Jun 09 09:57:55 hi woglinde Jun 09 10:04:53 03koen 07 * r1.3512 10openembedded/packages/maemo/ (3 files in 2 dirs): Add libosso-help, part of the maemo.org suite, courtesy Florian Boor Jun 09 10:04:59 03koen 07 * r1.3511 10openembedded/packages/maemo/ (2 files in 2 dirs): Add hildon-control-panel, part of the maemo.org suite, courtesy Florian Boor Jun 09 10:05:04 03koen 07 * r1.3510 10openembedded/packages/maemo/hildon-initscripts_0.8.14-1.bb: Add hildon-initscripts, part of the maemo.org suite, courtesy Florian Boor Jun 09 10:05:09 03koen 07 * r1.3509 10openembedded/packages/maemo/hildon-status-bar_0.8.11-1.bb: Add hildon-status-bar, part of the maemo.org suite, courtesy Florian Boor Jun 09 10:36:18 hey zecke Jun 09 10:45:16 hehe reading arm linux kernel ml... Jun 09 10:45:35 I've 2.4.0-test0-ac1-rmk0 and gcc5 does not compile it Jun 09 10:46:04 heh Jun 09 10:46:05 koen: IIRC, Japan already has total WiFi coverage... o.o Jun 09 10:46:29 zecke lol Jun 09 10:48:02 I wonder when rmk released 2.4.21-rmk1 Jun 09 10:50:04 zecke sometimes its the same on the mips list Jun 09 10:52:32 woglinde: I'll going to the congress tomorrow Jun 09 10:53:13 zecke jo Jun 09 10:53:31 I have to go my basketball club for the financials Jun 09 10:55:16 *need* food Jun 09 11:03:27 03mickeyl * r247 10bitbake/lib/bb/shell.py: Jun 09 11:03:27 add 'shell ' which hands a command over to the shell and dumps the output Jun 09 11:03:27 NOTE: Due to a limitation in the bbshell command processor, 'shell' and 'exec' are Jun 09 11:03:27 currently broken when using more than one parameter. This will be sorted out asap. Jun 09 11:04:57 mickeyl: does pastebin have a sizelimit? Jun 09 11:05:30 koen: not that i know Jun 09 11:05:37 koen: is it getting truncated ? Jun 09 11:05:41 it is Jun 09 11:05:56 bummer. where ~ ? Jun 09 11:05:56 http://pastebin.com/297847 Jun 09 11:06:23 hmm Jun 09 11:06:35 that looks like a limit in the service, not in the command Jun 09 11:07:04 you can work around it by only sending the end of the buffer Jun 09 11:07:16 (where the error usually is) Jun 09 11:07:58 Right. Jun 09 11:08:47 ~apple users Jun 09 11:09:20 ~users Jun 09 11:09:27 so this is what we get for sharing our unpaid volunteer work with you? complaints and accuses? well done, this clearly supports our motivation to continue working on open source projects. Jun 09 11:09:35 "Hey when there will be a Mac/intel Qt4 version.... I'm presuming it only needs a recompile..." Jun 09 11:09:49 ~factinfo users Jun 09 11:09:50 users -- created by mickeyl <~mickey@deneb.tm.informatik.uni-frankfurt.de> at Sun May 15 12:09:43 2005 (25 days); it has been requested 4 times, last by koen, 22s ago. Jun 09 11:09:52 hahaha Jun 09 11:09:53 :) Jun 09 11:11:08 * zecke catched up with emails Jun 09 11:11:16 and continues the quest for food Jun 09 11:11:23 the emails ate it Jun 09 11:11:24 Open Fridge, Look, Close Fridge Jun 09 11:11:30 scratch head Jun 09 11:11:35 Open Fridge... Jun 09 11:11:39 zecke lol Jun 09 11:12:44 * zecke decides for the Choco Chips+Milk Jun 09 11:18:19 zecke: anyway, mono choco chip are better Jun 09 11:19:04 right anyway I will now get my mono looking chips and tune on of the mono-logues in tv Jun 09 11:19:22 :) Jun 09 11:19:31 zecke: are we rich yet? Jun 09 11:19:58 ~pastebin Jun 09 11:19:59 i heard pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.ca Jun 09 11:20:48 is there a list of public pastebins? perhaps we'll find one without a limit Jun 09 11:21:07 * koen doesn't know Jun 09 11:21:26 koen: rich of thoughts Jun 09 11:21:33 mickeyl: google pastepin.ca Jun 09 11:21:36 mickeyl: check the related sites Jun 09 11:23:21 koen: we could sell our souls and offer Java and C# support Jun 09 11:23:32 hmmm Jun 09 11:23:35 tempting Jun 09 11:26:53 woglinde: wow gcc4 has fortran77 support ;) Jun 09 11:27:05 zecke? Jun 09 11:27:08 nopaste at http://www.rafb.net/paste/ doesn't *mention* a limit Jun 09 11:27:22 zecke whtas worn with f77 support? Jun 09 11:31:51 and if we were feeling clever, we could probably figure out some way to turn groups.google into a pastebin. Jun 09 11:32:12 sure, i could paste it into a usenet group but i don't think people'll like that Jun 09 11:32:12 heh Jun 09 11:33:09 a dummy mailinglist Jun 09 11:33:45 * zecke will auto subsribe everyone that replies to list and private.. Jun 09 11:34:09 my evolution had a 'reply to list' option Jun 09 11:34:16 thunderbird doesn't Jun 09 11:35:21 'L' Jun 09 11:36:17 bah ha ha Jun 09 11:36:20 Try the new Soylent Green Energy Drink! Jun 09 11:36:20 "How is it?" Jun 09 11:36:20 "Well, it varies from person to person." Jun 09 11:48:15 mickeyl: hm, bbshell help doesn't fit in 80 columns Jun 09 11:50:23 and readline support went away? Jun 09 12:09:37 <[cc]smart> i've got a feeling cyrus-imapd doesn't package everything: http://pastebin.ca/13831 Jun 09 12:10:17 * koen shares that feeling Jun 09 12:11:55 <[cc]smart> in cyrus-imapd workdir i have image/usr/cyrus/bin which seems to contain the rest of the tsuff Jun 09 12:12:44 [cc]smart: changing the packaging is easy Jun 09 12:12:56 <[cc]smart> once you know how to do it that is Jun 09 12:12:59 <[cc]smart> :) Jun 09 12:13:32 <[cc]smart> well, i don't feel like a python noob, but every time i tried to get how this OE stuff ticks i ended up frustrated Jun 09 12:14:43 [cc]smart: first find out where the cyrus-imap .bb file is Jun 09 12:14:47 <[cc]smart> or better described, i felt like advabce is so slow, you could a lot of nifty things in the same time and didn't find appropriate docs. that may have changed, but when i think about trying to get at OE owadays i feel lik- no, not agaoin Jun 09 12:14:54 <[cc]smart> easy one Jun 09 12:15:08 <[cc]smart> i'm talking about all those variables Jun 09 12:15:24 [cc]smart: we still suck at docs but we're improving slowly Jun 09 12:17:19 <[cc]smart> so i think i can be more efficient in finding some things to help like describing improvements of nt.conf, tickadj setting and the like i formerly did Jun 09 12:18:26 * keturn is unsure about some of these patches. Jun 09 12:18:29 Is br1 still around? Jun 09 12:18:46 ~seen br1 Jun 09 12:18:47 br1 <~br1@pD9E77FCC.dip.t-dialin.net> was last seen on IRC in channel #oe, 31d 3h 27m 2s ago, saying: 'when do i have to use "export VARIABLE = " in .bb files?'. Jun 09 12:19:03 ~seen br1_ Jun 09 12:19:04 br1_ <~br1@pD9E0EF47.dip.t-dialin.net> was last seen on IRC in channel #oe, 189d 23h 8m 15s ago, saying: 'bye, cya'. Jun 09 12:21:32 <[cc]smart> wel, it's time for niteshift. thanks all for BB & OE. cu. Jun 09 12:22:02 [cc]smart: cya Jun 09 12:37:32 http://pastebin.ca/13838 Jun 09 12:37:41 my fucking modem is possessed Jun 09 12:38:23 anyway, now pastebin is using the lucene.net indexer running mono it is getting useable Jun 09 12:39:36 no kidding I wonder where my modem gets that crap from... Jun 09 12:39:54 probably they slimmed the Java VM and installed fortune on it Jun 09 12:41:04 re Jun 09 12:41:16 ~lart c7x0 keyboard Jun 09 12:41:17 * ibot gives c7x0 keyboard a good seeing to Jun 09 12:41:43 zecke: is somebody trying a very stupid phishing attempt on you? Jun 09 12:42:36 RP: can we have pgup/pgdown on shift+cursors? Jun 09 12:42:45 CosmicPenguin: It is neither an sms or email.. Jun 09 12:42:56 heh - strange Jun 09 12:42:57 hrw|husky: just map it Jun 09 12:43:18 CosmicPenguin: I'm having a little problem wih multiplexing... and data connections Jun 09 12:43:35 zecke: probably will. looking around log in irsiiiii via esc+p/n make me mad Jun 09 12:43:43 Ciao all Jun 09 12:43:50 hi Pigi Jun 09 12:43:56 hey Pigi Jun 09 12:44:03 hrw|husky: both Qt/E and kdrive should cope with it Jun 09 12:44:13 evening Jun 09 12:44:28 morning rene Jun 09 12:44:56 hi france Jun 09 12:45:01 hrw|husky: hi Jun 09 12:45:17 hey Marcin Jun 09 12:45:21 hi france Jun 09 12:46:28 france_: I will mail you in next few days to get rw access to opie cvs. iirc i have to send some infos about me and ssh2 public key? Jun 09 12:46:52 * hrw|husky will killl for working Alt in opie-console Jun 09 12:47:04 koen, I was wondering which could be the right "ipkglibdir" for ipkg-native and ipkg in oe Jun 09 12:47:27 I would like to prepare the bb for 0.99.150 but I don't know how to handle this.... Jun 09 12:48:07 Pigi: I have no idea, sorry Jun 09 12:48:26 should ask pb_ Jun 09 12:48:37 Pigi: one question.. -verbose_wget but option verbose-wget - does it is correct? Jun 09 12:48:56 could even be a bug.... Jun 09 12:49:00 lemme se Jun 09 12:49:59 hrw|husky, the commandline option is either verbose-wget and verbose_wget Jun 09 12:50:17 the conf variable inside ipkg is verbose_wget Jun 09 12:50:17 Pigi: ah.. ok then Jun 09 12:50:47 it took me a while to get option name - finally found in sources Jun 09 12:51:01 koen, "ipkglibdir" is the new option I have added to configure to remove the needed of the ipkg.h patch in oe Jun 09 12:51:45 Pigi: please set it to "/usr/lib/ipkg" in the .bb file Jun 09 12:51:46 hrw|husky, I know, but fortunally args.c is one of the better code that is in ipkg Jun 09 12:52:07 pb_ ok. Either for ipkg-native and ipkg ? Jun 09 12:52:15 Pigi: yeah, for both Jun 09 12:52:23 ok. Jun 09 12:52:27 thanks Jun 09 12:52:39 pb_: hello Jun 09 12:52:44 hi france_ Jun 09 12:52:51 pb_ you're welcome Jun 09 12:52:57 france_: hey Jun 09 12:53:09 hrw|husky: and a note from the maintianer. Jun 09 12:53:28 koen: do you think tinderbox would run on mono? Jun 09 12:53:30 pb_ what ${libdir} is supposed to be in oe ? Jun 09 12:53:40 Pigi: check bitbake.conf Jun 09 12:53:48 zecke, thx Jun 09 12:53:58 Pigi: most likel /usr/lib Jun 09 12:53:59 Pigi: it will be /usr/lib for most packages, but -native ones will get a different value. Jun 09 12:54:00 zecke: *everything* runs on mono Jun 09 12:54:08 anyway, perl is bad Jun 09 12:54:12 thx Jun 09 12:54:21 koen: not that is perl.NET Jun 09 12:54:30 koen: anyway you can call the luence indexer from it Jun 09 12:54:40 :) Jun 09 12:54:54 france_: ofcourse Jun 09 12:55:17 zecke: who finally is a maintainer of opie? Jun 09 12:55:38 not me Jun 09 12:55:47 pb_: today I did a chmod o-rx oe_wiki to stop the spammers, since I did not get any response from the oe folks. I suspect that they are not usiing it any more. It would take a lot of work to restore the pages as spammers have been overwritting the pages of spammers. Jun 09 12:55:47 hrw|husky: do you want an account? Jun 09 12:56:20 france_: okay, sounds like a good idea Jun 09 12:56:34 france_: make a tar.bz2, put it somewhere, remove the old stuff from the net? Jun 09 12:56:52 actually I would do it, If I've enough rights to do so Jun 09 12:57:40 zecke: are you part of the oe group? Jun 09 12:57:51 france_: I don't think so Jun 09 12:58:01 zecke: finally yes - had recently some misc patches, maybe willl update Polisssh translations because they are too old etc Jun 09 12:58:35 france_: should I drop a mail to admin@hh.org to give hrw an account? Jun 09 12:58:44 zecke: yup! Jun 09 12:58:53 zecke: assuming you want him to have an account :-D Jun 09 12:58:55 zecke: thx Jun 09 12:59:12 zecke: nope you are not in proj_oe Jun 09 12:59:30 france_: and I should post public sshkey? Jun 09 12:59:55 hrw|husky: yup, so that you can log in vai ssh. Jun 09 13:00:23 include it in the e-mail with the user name that would prefer. Jun 09 13:00:45 france_: did you have time to look over the tinderbox requirements? Jun 09 13:00:47 france_: ofcourse Jun 09 13:01:46 zecke: kergoth, pb, mickeyl, mithro and noda Jun 09 13:02:36 zecke: I would be happy to make the tar.bz2 file for you but changing the permissions, should stop folks from seeing the pages, but I do dislike creating broken links. Jun 09 13:03:10 brb Jun 09 13:03:40 hrw|husky: mail sent Jun 09 13:04:11 ok Jun 09 13:04:45 so i will start another opie-console session and login to home box Jun 09 13:09:19 cya later... Jun 09 13:13:02 hrw|husky: Feel free to improve the c7x0 keymap. It was always a just prove it works step :). I'd just ask you document any keys you add Jun 09 13:13:22 (So I can benefit as well! ;-) Jun 09 13:15:30 hrw|husky: mail recieved from zecke Jun 09 13:17:35 RP: Did you build a kernel without FPE? Jun 09 13:17:50 do13: Not yet. Did you try the one you built? Jun 09 13:18:08 Yes. It works as expected. No problem. Jun 09 13:19:09 do13: nice. Makes me wonder why we bother compiling the emulation in really :) Jun 09 13:20:41 I guess it does allow us to run statically linked hardflaot binaries... Jun 09 13:20:57 Yep Jun 09 13:21:34 do13: I think I can fix nwfpe anyway... Jun 09 13:22:11 is there a way to do "cvs co -d notTheModuleName moduleName" with the cvs fetch method? I don't see one offhand. Jun 09 13:22:50 RP: Is it really necessary to support hardfloat binaries? Jun 09 13:22:52 at the moment bitbake has cvs checkout directories clobbering each other. Jun 09 13:23:36 do13: not really, although familar uses them Jun 09 13:50:29 hi Jun 09 13:50:42 anyone here building their own familiar linux images? Jun 09 13:51:42 jjg_: koen does Jun 09 13:56:47 koen, are there oe meta files available for the familiar build? Jun 09 13:57:01 i guess i should just say oe files? since oe implies the meta files? Jun 09 13:57:29 jjg_, http://oe.handhelds.org/cgi-bin/moin.cgi/GettingStarted Jun 09 13:58:31 Crofton_|laptop, hi Jun 09 14:01:49 jjg_: slightly outdated instructions: http://dominion.kabel.utwente.nl/koen/pda/?page=software&sub=bb Jun 09 14:09:21 thanks koen Jun 09 14:09:33 btw, do you buld for h5550 or simliar? Jun 09 14:09:43 Bye Jun 09 14:10:27 koen, looks like i definitely need to ge the OE repo before building Familiar, that correct? Jun 09 14:11:18 jjg_: http://oe.handhelds.org/cgi-bin/moin.cgi/GettingStarted Jun 09 14:11:24 jjg_: read that Jun 09 14:11:27 reenoo_, yes yes i already am there Jun 09 14:11:31 it does not mention famliar Jun 09 14:12:02 so someone wanting to build for famliar does not know where to start Jun 09 14:12:18 jjg_, lot of us started from that page Jun 09 14:12:21 but the page koen pasted is what i'm looking for Jun 09 14:12:26 I have done Jun 09 14:12:27 Pigi, as have I Jun 09 14:12:51 so what' s the problem/question ? Jun 09 14:13:25 i've seen no mention of Famliair and figured there would be specific documentation for that Jun 09 14:13:44 i'm pulling down the oe repo, now Jun 09 14:14:11 basically the job starts from DISTRO in local.conf Jun 09 14:14:51 am I wrong in thinking there *should* be documentation germane just to familiar? Jun 09 14:15:16 I realize that oe/bitbake are unrelated, but Familiar is a very important project relying on it Jun 09 14:15:24 unrelated to Familiar that is Jun 09 14:16:35 jjg_: yes it is. and? if the familiar guys think they need a seperate doc, which is unlikely, they're welcome to write one. until then, follow the instructions in GettingStarted and set your DISTRO to familiar. thats all you need to do. Jun 09 14:16:50 ok, great Jun 09 14:16:58 maybe that should be in a wiki somewhere Jun 09 14:17:19 or somebody should put that in a chatbot that can cluebat *ME* Jun 09 14:17:22 kergoth: that document is in place if I delete my local.conf *again* Jun 09 14:17:27 welcome back kergoth. Jun 09 14:18:33 hey Jun 09 14:18:36 heh Jun 09 14:19:37 (there used to be a "hire me" statement on that page, which worked out) Jun 09 14:20:04 cu monday Jun 09 14:20:11 later hrw|gone Jun 09 14:21:38 hey kergoth Jun 09 14:22:11 kergoth: hey Jun 09 14:22:15 kergoth: where are you going? Jun 09 14:22:39 ? Jun 09 14:23:09 people were talking about your unpacking, etc Jun 09 14:23:30 subdue: generally, you unpack after you arrive, not before you leave Jun 09 14:23:47 they also talked about his packing Jun 09 14:24:02 earlier today Jun 09 14:24:06 subdue: quit my job at ti, in texas, and moved back to my old townhouse in minnesota. now unemployed Jun 09 14:24:15 got here like a week or two ago Jun 09 14:24:15 oh! Jun 09 14:24:29 wow. ti's pretty impressive Jun 09 14:24:31 only just got my stuff from the movers yesterday, hence the unpacking Jun 09 14:24:40 ti's great, juts the job wasnt a good fit Jun 09 14:24:44 just, even Jun 09 14:24:45 ah Jun 09 14:26:22 03pb 07 * r1.3513 10openembedded/packages/glibc/glibc_cvs.bb: bash glibc_cvs around a bit Jun 09 14:30:27 i believe I have everything configured correctly and am now getting this : ERROR: Unable to parse conf/bitbake.conf Jun 09 14:30:30 any idears? Jun 09 14:30:53 jjg_: BBPATH Jun 09 14:30:55 check BBPATH and BBFILES Jun 09 14:31:05 oh yah, forgot to source my .bashrc Jun 09 14:31:21 that wasn't it Jun 09 14:31:34 BBPATH not set correctly Jun 09 14:31:45 ok Jun 09 14:36:34 03reenoo 07 * r1.3499.11.1 10openembedded/packages/kaffe/ (kaffe-gtk_cvs.bb kaffe-gtk_1.1.5.bb kaffe.inc): Jun 09 14:36:34 kaffe: Jun 09 14:36:34 - kaffe.inc: only add postinst/postrm for ${PN} packages. Jun 09 14:36:34 - kaffe-gtk: add xtst to DEPENDS. fixes bug #65. Jun 09 14:40:19 do i need to export BBFILES var even though it is in local.conf? Jun 09 14:40:30 nope. you dont need to set that in your environment Jun 09 14:40:49 i don't need to create bitbake.conf right? Jun 09 14:40:56 because my path is correct Jun 09 14:41:11 paste your BBPATH here. Jun 09 14:41:29 /home/sisa/oe/build:/home/sisa/oe/openembedded Jun 09 14:41:39 [sisa@SDR-2 .bk]$ echo $BBPATH Jun 09 14:41:39 /home/sisa/oe/build:/home/sisa/oe/openembedded Jun 09 14:41:42 and /home/sisa/oe/openembedded/conf/bitbake.conf exists? Jun 09 14:41:47 no! Jun 09 14:41:48 and BBPATH isnt just set, its _exported_? Jun 09 14:41:53 eh? Jun 09 14:41:53 it is exported Jun 09 14:42:03 conf/bitbake.conf is in the oe tree. Jun 09 14:42:03 yah, i didn't see anything bout creating that file Jun 09 14:42:19 ehm Jun 09 14:42:21 it's in the repo Jun 09 14:42:22 doh :) Jun 09 14:42:33 if you downloaded the repo, the file should be there Jun 09 14:42:37 i see it Jun 09 14:42:38 :) Jun 09 14:42:39 if you have the file and bitbake still cant find it, try doing a rm -rf tmp/cache. Jun 09 14:42:42 i was thinking build/conf/bitbake.conf Jun 09 14:43:31 thats local.conf :) Jun 09 14:43:51 yah, i square on that now. Now just to figure why this thing is bleating about bitbake.conf Jun 09 14:44:22 [sisa@SDR-2 conf]$ ls -l Jun 09 14:44:22 total 72 Jun 09 14:44:22 -r--r--r-- 1 sisa sisa 14756 Jun 9 07:24 bitbake.conf Jun 09 14:44:23 once again, if bitbake cant find it, and BBPATH is correct, exported, and the file exists, try wiping the cache. Jun 09 14:44:33 kergoth, sry...do that now Jun 09 14:44:49 which should be in build dir, no? Jun 09 14:45:01 it shuld be in the directory you're running bitbake in Jun 09 14:45:05 which _should_ be build/, yes Jun 09 14:45:13 tmp isn't yet created Jun 09 14:46:53 * jjg_ slips out his claw hammer and eyes his laptop Jun 09 14:51:30 time to sleep Jun 09 14:51:55 nite all Jun 09 14:52:03 'night Pigi Jun 09 14:52:15 'night all Jun 09 15:00:07 should i have a bitbake.conf in ~/oe/bitbake/conf/bitbake.con and ~/oe/openembedded/conf/bitbake.conf? Jun 09 15:01:28 kergoth, any ideas on how to make bitbake happy? Jun 09 15:02:40 jjg_: i had to copy bitbake.conf from bitbake/conf to openembedded/conf Jun 09 15:02:55 not sure when/if that changed but once i did that it was alright Jun 09 15:04:18 jh_, i think that did it Jun 09 15:04:21 strange Jun 09 15:04:22 a bug? Jun 09 15:05:06 more likely either your bitbake or OE trees wern't up to date Jun 09 15:05:49 CosmicPenguin, interesting. I just got the repo with svn and bk from the website. Jun 09 15:05:59 now it says this : ERROR: no files to build. Jun 09 15:06:13 ERROR: Nothing provides gpe-image Jun 09 15:08:50 did you set BBFILES correctly in local.conf ? Jun 09 15:09:26 ascent, BBFILES := "/home/sisa/oe/openembedded/packages/*/*.bb" Jun 09 15:09:53 i'm going cross eyed from staring so hard at it...it's right though Jun 09 15:09:57 copying bitbake.conf from bitbake to oe is NOT going to work. Jun 09 15:10:01 at all. Jun 09 15:10:04 kergoth, oops Jun 09 15:10:08 well, it didn't complain this time Jun 09 15:10:19 yes, but it isnt going to _do_ anything. Jun 09 15:10:25 so i need to back up and repull the repo Jun 09 15:10:26 ok Jun 09 15:10:33 that bitbake.conf doesnt contain most of the variables that are necessary Jun 09 15:10:36 its minimal Jun 09 15:10:48 i'm pulling the repo again Jun 09 15:11:01 3rd time's a charm ? :) Jun 09 15:13:29 have to say that oe looks like it is gonna be the bomb Jun 09 15:19:45 kergoth, 3 times *is* a charm :) Jun 09 15:19:52 NOTE: Handling BitBake files: - (0211/2417) [ 8 %] Jun 09 15:20:09 1% inspiration 99% perspiration Jun 09 15:21:44 kergoth, aren't you going to pat me on my newbie head? Jun 09 15:22:08 heh, congrats :) Jun 09 15:22:21 awe, tanks onnare Jun 09 15:22:49 i've been using familiar for a couple of years and have always wanted to build my own distro Jun 09 15:22:53 now...WAM Jun 09 15:27:04 ~lart kergoth for talking to the newbie instead of testing scms Jun 09 15:27:05 * ibot whacks kergoth with a giant beaver's tail for talking to the newbie instead of testing scms Jun 09 15:27:12 kergoth, a bug was introduced between 0.8.1 and 0.8.2 wlan driver for H5550 ... how does this relate to the source code that oe is pulling? Jun 09 15:27:40 meaning is the bug going to be in the code that i'm pulling? Jun 09 15:27:58 treke, please don't whack him, he's been very nice to help me out. Jun 09 15:28:10 btw, newbs make the world go round Jun 09 15:30:10 ~lart kergoth to spite jjg Jun 09 15:30:10 * ibot beats kergoth to spite jjg into protomatter with the andromeda galaxy Jun 09 15:30:30 awe man Jun 09 15:31:17 ~lart treke Jun 09 15:31:18 heh Jun 09 15:31:18 * ibot takes large quantities of Krispy Kream donuts and stuffs them one after another down treke's throat until treke puts on 150lbs Jun 09 15:31:21 awe man Jun 09 15:31:30 whoa Jun 09 15:32:44 bitbake gpe-image just failed with this : incorrect classpath: /usr/share/java/libgcj-3.4.2.jar Jun 09 15:32:44 | ---------- Jun 09 15:32:44 | 1. ERROR in /home/sisa/oe/build/tmp/work/gettext-native-0.14.1-r3/gettext-0.14 .1/gettext-runtime/intl-java/gnu/gettext/GettextResource.java (at line 1) Jun 09 15:34:04 treke, wanna help the newb now? Jun 09 15:34:15 why is bitbake complaining about java? Jun 09 15:34:25 no, I don't have the file it was looking for Jun 09 15:36:53 atleast in that directory Jun 09 15:37:01 anyone running bitbake on fc3? Jun 09 15:39:14 it's looking for an older java lib Jun 09 15:43:21 03pb 07 * r1.3513.1.1 10openembedded/ (3 files in 3 dirs): new and updated jlime files, per bug #39 Jun 09 15:44:45 maybe specifying the localpath parameter to the cvs fetch method would do what I need. but it's not exactly clear what that does. Jun 09 15:45:17 ok, so i've gotta problem here...oe is asking for a java version that is older than the one i've got. anyone experienced this? Jun 09 15:45:47 it's looking for libgcj-3.4.2.tar and i've got libgcj-3.4.3.tar Jun 09 15:45:59 should i just symlink it? Jun 09 15:46:07 seems like an oe issue Jun 09 15:49:24 awe man, someone help the newbie Jun 09 15:50:21 that's not an oe issue. Jun 09 15:50:27 something is busted about your gcj installation Jun 09 15:52:14 pb_: Does http://www.rpsys.net/openzaurus/patches/preempt_softfloat-r0.patch look sane to you? Jun 09 15:56:36 03reenoo 07 * r1.3514.1.1 10openembedded/packages/unzip/unzip_552.bb: unzip: use oe_runmake rather than make so that CC is set properly etc. install man pages to ${D}${mandir}. Fixes bug #45. Jun 09 15:58:22 pb_, what makes you think that exactly? because this is a brandnew fc3 install from yesterday and then I ran up2date. Jun 09 15:59:49 RP: yeah, looks okay. You should probably check that it doesn't cause any regressions in the glibc math testsuite. Jun 09 16:00:44 jjg_: because it seems unlikely that anything in OE would be looking for files in /usr/share/java of its own accord. Jun 09 16:01:13 pb_: That may be tricky given my machines are all currently softfloat... Jun 09 16:01:57 pb_, ok Jun 09 16:02:25 RP: that is unfortunate. Jun 09 16:02:30 pb_: thanks for figuring out the reason to the unzip breakage. that should make a few java users pretty happy (to extract .jar files) Jun 09 16:04:09 reenoo_: ah, right. don't you use fastjar or something for that? Jun 09 16:04:43 fixing unzip should be pretty trivial, anyway. I haven't actually looked at the makefile yet, but I imagine it just needs a small tweak. Jun 09 16:05:09 my pty code works... Jun 09 16:05:17 very good Jun 09 16:05:18 but I think my fd is not bidirectional... Jun 09 16:05:32 pb_: well.. it was trivial (see last cset) Jun 09 16:05:49 oh, heh Jun 09 16:05:51 reenoo_: the same might apply for automount? Jun 09 16:05:52 I didn't notice that. Jun 09 16:06:06 reenoo_: very good Jun 09 16:06:07 zecke|hack: dunno. maybe Jun 09 16:07:13 * pb_ bedtime now Jun 09 16:07:14 later all Jun 09 16:07:21 'night pb_ Jun 09 16:07:21 pb_: cya Jun 09 16:07:27 Tomorrow will be fun Jun 09 16:07:27 'night pb_ Jun 09 16:07:42 1.) dentist 2.) Chaos Computer Club camp Jun 09 16:11:32 zecke|hack: actually, no, the autofs .bb inherits autotools which has a proper do_compile() Jun 09 16:11:54 zecke|hack: must be a problem with the makefiles Jun 09 16:12:23 anyone using fc3? Jun 09 16:13:23 :} Jun 09 16:18:22 anyone using *any* fedora version for oe? Jun 09 16:26:31 'night all Jun 09 16:27:24 should i send an email to anyone saying that building gettext on fc3 is broken? Jun 09 16:32:17 'night all Jun 09 16:46:04 ok, turns out that in fc3.../usr/bin/java -> /etc/alternatives/javac which is a bash script with a lib hardcoded !!!! Jun 09 16:58:10 <[g2]> pb_ RP thx for the ARM asm help earlier today Jun 09 16:59:00 <[g2]> The tiny LED program runs from Redboot and tomorrow (later today) it'll get tested with loading directly into the mini i cache on the IXP420 Jun 09 16:59:18 <[g2]> propably when ep1220 wake up :) Jun 09 17:07:35 good nite Jun 09 17:11:53 building libc baby, OH yah Jun 09 17:12:26 i'm gone be tha oe mastuh Jun 09 17:13:00 should I go ahead and remove the recommendation to use BBCOW from the wiki? Jun 09 18:11:01 treke: do you really want to be default owner for all bitbake bugs, or should those get sent to bitbake-dev or something like the oe ones do? Jun 09 19:35:44 can somebody do the following on their /stuff directory and their /sources directory.... du -h --max-depth=0 ... Jun 09 19:36:12 I'm trying to get a handle on how much space I'll need to build... I had a partition of 4.5GB... but that is not enough Jun 09 20:08:02 jumpkick, its ~ 8gb Jun 09 20:19:11 emte: ty Jun 09 21:17:36 cf or sd? Jun 09 22:39:05 <_law_> how can i see if i am using ccache? Jun 09 23:53:08 morning **** ENDING LOGGING AT Thu Jun 09 23:59:57 2005