| 00:00.02 | pb_ | yeah, at least it will remove one source of random lossage in our builds. |
| 00:00.04 | gb2 | i find debian to be more painful *shrug* |
| 00:00.10 | kergoth | heh |
| 00:00.12 | W8TVI | lol |
| 00:00.13 | gb2 | especially the installer and package building |
| 00:00.21 | gb2 | their package build tools change all the time |
| 00:00.22 | kergoth | working for digi, we distributed kernel modules outside the kernel |
| 00:00.29 | gb2 | so you can't move a sid package to build on woody easily |
| 00:00.30 | kergoth | building them against sources that didnt match the running kenrel was rampant |
| 00:00.33 | gb2 | ok.. Enterprise time. |
| 00:00.43 | kergoth | so rh doing that for their glibc kern headers helped preempt those |
| 00:00.50 | kergoth | #error DONT DO THAT |
| 00:00.51 | kergoth | heh |
| 00:00.56 | pb_ | heh |
| 00:01.09 | kergoth | 7 |
| 00:01.25 | pb_ | mm, bedtime for me. |
| 00:09.12 | noda | ibot oz bk |
| 00:09.13 | | hmm... oz bk is bk clone bk://openzaurus.bkbits.net/buildroot buildroot-oz; cd buildroot-oz; bk -r co -q; cp def-configs/collie-opie .config (for zaurus 5000d/5500) .. make oldconfig; make; or see bitkeeper (free download but non-gpl) or see 'oz snapshot' if you either cannot or will not use bitkeeper. |
| 00:11.38 | noda | ACK! There's a mosquito in my glass of water! I almost drank it! AAAAAAARGH |
| 00:11.46 | W8TVI | lol |
| 00:12.15 | W8TVI | noda: http://knerd.com/~kb8sjt/vlog.txt |
| 00:17.22 | *** join/#openembedded andersee (~andersee@codepoet.org) |
| 00:18.30 | treke|home | noda: less talk. more code |
| 00:19.00 | noda | treke|home: Haven't you heard? fdask is coding a new download page :) |
| 00:19.34 | treke|home | noda: No new image for you |
| 00:19.48 | noda | treke|home: Awww, c'mon! :) |
| 00:20.01 | *** join/#openembedded decaff (~kevin@clt74-76-015.carolina.rr.com) |
| 00:20.42 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 00:22.21 | CIA | repository=03packages user=11kergoth@direwolf.ppp.ti.com |
| 00:22.23 | CIA | 07ChangeSet@1.65 Initial initscripts package. |
| 00:35.37 | kergoth`bbl | noda: <pester> rt? </pester> |
| 00:44.18 | treke|home | I say we delay 3.4 one day for every day noda doesnt install rt and write a new download script |
| 00:45.22 | treke|home | kergoth`bbl: hmmm. Odd 3.3-oz1 worked, but 3.3.1pre-oz1 didnt |
| 00:45.33 | kergoth`bbl | weird |
| 00:46.16 | treke|home | ugh |
| 00:46.25 | treke|home | I hate it when I run make mrproper by accident |
| 00:46.30 | kergoth`bbl | hehe |
| 00:46.36 | kergoth`bbl | push your new .oe's |
| 00:47.44 | treke|home | I was looking at a bad example :) |
| 00:47.51 | kergoth`bbl | hehe, what were you looking at? |
| 00:47.56 | treke|home | zsh |
| 00:47.58 | kergoth`bbl | ah |
| 00:48.00 | kergoth`bbl | thats an old one |
| 00:48.17 | treke|home | need to do packages |
| 00:48.18 | kergoth`bbl | whats wrong with it though? |
| 00:48.19 | kergoth`bbl | ah |
| 00:48.28 | kergoth`bbl | i've been leaving the packaging undone |
| 00:48.35 | treke|home | also eed to do do_stage |
| 00:48.38 | kergoth`bbl | which means it creates one package with name PN and FILES * |
| 00:48.55 | treke|home | ah |
| 00:49.31 | treke|home | brb. need to go drop off cleaning at the laundry |
| 00:50.57 | *** join/#openembedded chouimat (~dieu@dsl-157-168.b2b2c.ca) |
| 00:50.57 | *** mode/#openembedded [+o chouimat] by ChanServ |
| 00:51.18 | noda | I'm back. |
| 00:51.34 | noda | I've tested out RT. Need 47 Perl dependencies. |
| 00:51.53 | noda | There's a debian (testing/unstable) package that has RT. |
| 00:57.57 | gb2 | i keep trying to look at zsh as an example too |
| 00:58.50 | kergoth`bbl | its a clean example for the most part |
| 00:58.56 | kergoth`bbl | doesnt make subpackages, just uses the default |
| 00:59.16 | gb2 | yeah, looked simple |
| 00:59.59 | gb2 | the router box i'm building, i'm planning to use zsh on it as the shell |
| 01:00.04 | gb2 | no other shell on it |
| 01:00.07 | kergoth`bbl | cool |
| 01:00.16 | gb2 | since it's nearly as powerful as perl |
| 01:00.25 | kergoth`bbl | course that zsh build fails atm, and ncurses fails also.. but thats inconsequential |
| 01:00.27 | gb2 | and i'll do up all my router scripts using zsh |
| 01:00.31 | gb2 | heh |
| 01:00.33 | gb2 | :P |
| 01:00.40 | noda | kergoth`bbl: Where do you want rt installed? |
| 01:00.54 | gb2 | i'm planning to do it all using uclibc also, so i've got a ways to get to that point if i'm going to use OE for the build system |
| 01:03.22 | treke|home | ncurses worked last night |
| 01:03.35 | kergoth`bbl | noda: heh, hell if i know. we should really dig up somewhere to host our stuff other than sf |
| 01:03.41 | kergoth`bbl | treke|home: yeah i know. |
| 01:04.00 | kergoth`bbl | the failure is elusive, it comes down to the metadata passing issues i found last nigh |
| 01:04.03 | kergoth`bbl | t |
| 01:04.04 | chouimat | noda: rt -> request tracker? |
| 01:04.06 | noda | kergoth`bbl: SF would be impossible I think. |
| 01:04.08 | noda | chouimat: yes. |
| 01:04.14 | noda | WHOA. http://gdesklets.gnomedesktop.org/shot/cornerxmms-0.png |
| 01:04.19 | chouimat | noda: better than bugzilla? |
| 01:04.20 | noda | I'm installing *that* :> |
| 01:04.25 | noda | chouimat: Less... intense. |
| 01:04.32 | kergoth`bbl | we coudl host it on handhelds.org, but its pretty far behind, would be a pain to get operational |
| 01:04.33 | treke|home | noda: That's nifty |
| 01:04.49 | kergoth`bbl | perhaps we should send an email soliciting hosting space for OZ web mirrors? |
| 01:04.59 | noda | handhelds.org is slow, too, isn't it... |
| 01:05.07 | noda | Debian preferred! |
| 01:05.12 | treke|home | not the fastest machine in the world, but it's available |
| 01:05.17 | noda | With Debian, RT can be installed really really easily :) |
| 01:05.24 | noda | I've got machines, but no pipes. |
| 01:06.06 | treke|home | apt-get install request-tracker1 :p |
| 01:06.16 | treke|home | wow. that was easy |
| 01:06.19 | noda | treke|home: 3 |
| 01:06.27 | noda | Not 1, 3 :P |
| 01:06.33 | treke|home | noda: I use stable :p |
| 01:07.41 | noda | apt-get -t unstable install request-tracker3 :P |
| 01:07.45 | treke|home | fuck no |
| 01:07.50 | treke|home | ~lart noda |
| 01:07.54 | noda | Why not? |
| 01:08.09 | treke|home | noda: That'll suck down god knows what from unstable |
| 01:08.21 | noda | Perl 5.8, that's about the only stuff that'll interfere with everything else. |
| 01:08.25 | noda | Better than using RT1 :P |
| 01:09.20 | treke|home | took me the better part of a day to build php 4.3, but I'm upgraded to it now |
| 01:09.38 | kergoth`bbl | apt-src? |
| 01:09.39 | kergoth`bbl | heh |
| 01:10.06 | treke|home | kergoth`bbl: Still have to build the depenencies, and I want it built against as must of woody as possible |
| 01:10.15 | kergoth | ah |
| 01:10.30 | treke|home | takes more work, but I think it's worth it in the end |
| 01:10.33 | kergoth | apt-get source then |
| 01:10.36 | kergoth | heh |
| 01:10.51 | treke|home | yup. I had a build vm at home that I installed the buildtools for stable on |
| 01:11.00 | kergoth | ah |
| 01:11.00 | noda | Why use the source? That's just as unstable as using -t unstable, the dependencies are there for a reason. |
| 01:11.09 | kergoth | noda: because of the shlib deps |
| 01:11.22 | kergoth | noda: oftentimes you can build it against differing versions of libs than are in unstable |
| 01:11.22 | treke|home | noda: Because using -t unstable starts pulling in all kinds of stuff like say glibc 2.3.2 |
| 01:11.23 | noda | Oh, you're talking PHP... thought you meant in general |
| 01:11.26 | kergoth | and have it still function |
| 01:11.29 | kergoth | right |
| 01:11.30 | treke|home | The same goes in general |
| 01:11.34 | kergoth | due to the dh_shlibdeps |
| 01:11.36 | kergoth | or whatever |
| 01:11.45 | kergoth | yep |
| 01:11.48 | noda | kergoth: But if you use apt-get -t unstable install <file>, it'll only upgrade what it needs, and it'll install as much as possible from stable if that's your Default-Release, won't it? |
| 01:11.56 | noda | s/file/package |
| 01:11.57 | kergoth | you arent listening |
| 01:12.06 | kergoth | the packgae will "need" glibc 2.3.2 lib versioned symbols |
| 01:12.10 | kergoth | so "needs" 2.3.2 |
| 01:12.11 | noda | Ah. |
| 01:12.35 | kergoth | which is why source rebuilding works, it bases the deps on what the binary linked against |
| 01:12.43 | kergoth | we need a helper like that for OE |
| 01:12.49 | treke|home | For my sanity, I prefer leaving as much in the same state as debian's security team deals with it |
| 01:13.55 | *** join/#openembedded bipolar_ (~bflong@ctt186159.ceinetworks.com) |
| 01:16.10 | treke|home | good god. that's a lot of perl modules |
| 01:17.11 | kergoth | hehe |
| 01:17.13 | kergoth | yep |
| 01:17.17 | noda | 47 :) |
| 01:17.32 | noda | Not exactly ideal for apt-getting source code and compiling :) |
| 01:17.45 | treke|home | hehe |
| 01:18.08 | treke|home | the easy thing to do is just to upgrade to testing |
| 01:18.11 | noda | Whee, gdesklets are awesome :) |
| 01:18.18 | treke|home | too bad metacity sucks |
| 01:18.23 | noda | It rocks! |
| 01:18.29 | treke|home | bullshit |
| 01:18.32 | noda | So simple, efficient, friendly. |
| 01:18.38 | noda | What *doesn't* it do? :) |
| 01:18.43 | treke|home | noda: Window matching |
| 01:18.48 | noda | Matching? Whassat? |
| 01:18.59 | noda | Not to mention, metacity is not a GNOME requirement, nor does gdesklets require metacity. |
| 01:19.12 | noda | (and there's nothing illegal about running GNOME programs in KDE, anyway) |
| 01:19.15 | treke|home | <window manager> Look here. Evolution is opening. Greg always puts this over here |
| 01:19.24 | treke|home | Who said I liked kde? |
| 01:19.47 | treke|home | <window manager> gee I bet he wants this gaim window on every desktop |
| 01:19.51 | noda | Evolution doesn't have an option to move to a specific location? |
| 01:20.00 | treke|home | That isnt evolutions job |
| 01:20.08 | treke|home | The window manager manages windows |
| 01:20.09 | noda | That is pretty nifty, but I will gladly live without it. |
| 01:20.44 | treke|home | suit yourself |
| 01:20.46 | noda | :) |
| 01:20.52 | noda | Anyway, doesn't sawfish work with GNOME? |
| 01:21.17 | treke|home | don't recall if it ever got proper gnome2 support |
| 01:21.29 | noda | I think it has it. |
| 01:25.45 | treke|home | noda: Havoc pennington was being a bitch about it. Wouldnt include window matching support until someone developed an X standard describing how all apps do it, regardless of the fact that window managers have been doing it for years now fairly damn well :) |
| 01:26.31 | noda | treke|home: I happen to like Havoc, and I certainly agree about this -- freedesktop.org should make a standard. |
| 01:27.23 | treke|home | a freedesktop.org standard doesnt do any good. It means that GNOME apps would support the feature, and that it wouldnt work with most apps out there |
| 01:28.41 | ljp | why oh why isn't networking working.. hmmpft |
| 01:29.33 | chouimat | treke|home: havoc was invited at the kde conference to talk about freedesktop and dbus |
| 01:29.41 | ljp | there's like some missing kung-fu, or something |
| 01:30.35 | noda | treke|home: Both KDE and GNOME support freedesktop standards. |
| 01:30.53 | treke|home | noda: Ok. So GNOME and KDE apps work. What about all those other ones? |
| 01:30.53 | ljp | how come the sharp om give a wifi a eth0 device, and OZ gives it a wlan0 ? |
| 01:30.55 | noda | Seems to me there'd be no good reason *not* to. |
| 01:31.22 | chouimat | noda: i we want linux on the corporate desktop we need to have something |
| 01:31.43 | treke|home | the gtk 1.2 apps. the qt 1.4 apps. the motif apps. mozilla |
| 01:31.49 | treke|home | qt 2 apps |
| 01:31.51 | noda | treke|home: The way I see it, there are 4 camps: 1. GNOME apps; 2. KDE apps; 3. Apps which are based on QT/GTK/whatever, and are kept up-to-date with the latest technologies, and 4. Apps which are either not maintained, or only get bugfixes. |
| 01:32.04 | noda | The only category which doesn't conform is 4. |
| 01:32.16 | noda | Personally, the only program I use in that category is xmms. |
| 01:32.21 | ljp | 5. opie apps |
| 01:32.37 | ljp | opie as desktop! |
| 01:32.48 | chouimat | ljp: ask your bosses :) |
| 01:33.11 | kergoth | ljp: differnet drivers. |
| 01:33.18 | kergoth | ljp: hostap uses wlan0 |
| 01:33.40 | treke|home | noda: Apps in #4 tend to be pretty important when people are still using them, or they'd have switched to a gtk or qt app :) |
| 01:34.02 | ljp | hmm, I dont think dhcp is working correctly, cause I cant get an ip |
| 01:34.12 | treke|home | even if they use crappy widgets sets like say... xforms |
| 01:34.14 | treke|home | ~lart lyx |
| 01:34.16 | kergoth | ah good, looks like easydns supports round robin |
| 01:35.43 | treke|home | noda: I just think it's stupid to not do something that can be reasonably done and wait for a solution that only works with recent applications |
| 01:36.09 | kergoth | heh |
| 01:36.31 | kergoth | A non hackish solution is all well & good, but sometimes you need hacks to hold over until that is feasable. |
| 01:36.39 | treke|home | yup. |
| 01:37.08 | treke|home | rather like my opinion on gcc3 and openzaurus. gcc2 sucks, but not having any third party apps sucks more :) |
| 01:37.15 | kergoth | indeed |
| 01:37.22 | chouimat | and sometimes you need to decide to left some old apps die |
| 01:37.27 | kergoth | yep |
| 01:38.03 | noda | treke|home: I'm an idealist. GNOME is idealist. And so far, it's really accomplishing a lot. |
| 01:38.16 | noda | And the thing I like: For the most part, it works. Very well. |
| 01:38.19 | chouimat | or we still be studying OS/390 cobol in 500 years |
| 01:38.32 | chouimat | gnome sucks |
| 01:38.40 | noda | People criticize the HIG, but apps that conform are for the most part higher quality. They act reliably. |
| 01:39.34 | chouimat | noda: my days of writing GUI applications in C are over |
| 01:39.42 | noda | chouimat: It's not a requirement. |
| 01:39.52 | noda | You can use gnomemm. |
| 01:40.01 | noda | Or Python. gdesklets is written in Python. |
| 01:40.26 | chouimat | noda: I'm sticking with kde and OS X for now and I'm happy |
| 01:40.35 | noda | OSX? Ick :) |
| 01:40.45 | treke|home | really a shame linux and c++ are such a pain in the ass still :) |
| 01:40.58 | noda | How so? g++ -o output_file file.cc |
| 01:40.59 | chouimat | treke|home: yuup |
| 01:41.23 | chouimat | noda: bah :) |
| 01:41.50 | treke|home | noda: little things like "will this app link against this library" |
| 01:42.26 | chouimat | treke|home: or how many dayys it will take to compile with g++ |
| 01:42.33 | noda | ccache... |
| 01:42.43 | noda | Plus, gcc-3.4 will have precompiled headers :) |
| 01:42.44 | treke|home | s/gcc2 app/gcc2 lib/ |
| 01:42.51 | fdask | ccache is such a great timesaver |
| 01:43.42 | ljp | hmm, interesting |
| 01:45.06 | kergoth | ccache rocks |
| 01:48.27 | Soopaman | dynapad? |
| 01:48.30 | Soopaman | what's that |
| 01:48.51 | treke|home | Soopaman: user environment |
| 01:49.03 | Soopaman | url? |
| 01:49.18 | treke|home | google can find it |
| 01:49.23 | Soopaman | k |
| 01:49.35 | chouimat | treke|home: the smalltalk based one? |
| 01:49.41 | treke|home | chouimat: yes |
| 01:49.56 | treke|home | kergoth: guh. I've made a mess :p |
| 01:50.03 | treke|home | a very big mess |
| 01:53.23 | noda | kergoth: In the buildroot, what def-config should I use for a collie which would allow SD? |
| 01:53.55 | noda | Or do I just use kergoth-build and kergoth-image... |
| 01:54.37 | kergoth | collie-opie |
| 01:54.42 | noda | Thanks :) |
| 01:54.53 | noda | That uses gcc-2.94? |
| 01:55.08 | chouimat | http://www.softfield.com/mx7.html |
| 01:55.08 | noda | ibot oz bk |
| 01:55.10 | | somebody said oz bk was bk clone bk://openzaurus.bkbits.net/buildroot buildroot-oz; cd buildroot-oz; bk -r co -q; cp def-configs/collie-opie .config (for zaurus 5000d/5500) .. make oldconfig; make; or see bitkeeper (free download but non-gpl) or see 'oz snapshot' if you either cannot or will not use bitkeeper. |
| 01:56.24 | noda | kergoth: Is there a way to use gcc-2.94 for the kernel and gcc-3.3 for everything else? |
| 01:57.27 | kergoth | heh |
| 01:57.51 | noda | Oh SWEET. These options magically appeared when I put gcc3 in somewhere. Nice :) |
| 01:58.04 | kergoth | eh, those options are always there |
| 01:58.39 | noda | Wait a sec... there's a .config and a output/.config? |
| 01:58.44 | kergoth | yes |
| 01:58.47 | noda | So that .config should have, like, everything enabled? |
| 01:58.50 | kergoth | output/.config is the packages selected to go into the image |
| 01:58.55 | kergoth | .config is whats to be built |
| 01:58.56 | noda | Ah. That's cool. |
| 01:59.05 | kergoth | make image_{old,def,rand,menu,...}config |
| 01:59.10 | kergoth | thats what the _image defconfigs are |
| 01:59.14 | noda | Awesome. |
| 01:59.14 | treke|home | hmmm |
| 01:59.22 | treke|home | wonder if maildir would be quicker over imap |
| 01:59.35 | chouimat | night |
| 01:59.41 | noda | randconfig? |
| 01:59.47 | kergoth | random selection |
| 01:59.51 | kergoth | heheh |
| 01:59.53 | noda | lol |
| 02:00.05 | noda | I don't need to set 'pad bytes' on the image? |
| 02:00.07 | kergoth | i find allyesconfig and allnoconfig useful |
| 02:00.20 | kergoth | if pad is unset it'll override with the default for your device |
| 02:00.25 | kergoth | the value that is |
| 02:00.31 | noda | Sweet. |
| 02:00.41 | noda | Um. I go to 'Packages -->' and the the screen flashes. |
| 02:00.49 | kergoth | then there arent any. |
| 02:00.53 | kergoth | let me guess, image_meuconfig? |
| 02:00.56 | kergoth | it selects IPKS |
| 02:00.57 | noda | Yep. |
| 02:01.00 | kergoth | do you have ipks? |
| 02:01.00 | kergoth | fuck no, you dont |
| 02:01.08 | kergoth | it can only install packages that exist. |
| 02:01.09 | noda | Erm. So I gotta build the whole freakin' thing first? :) |
| 02:01.11 | noda | Okay, that's neat :) |
| 02:01.39 | treke|home | god bless debian |
| 02:01.43 | noda | Hey, 'make image_randconfig' is fun :) |
| 02:01.48 | kergoth | hehe |
| 02:01.51 | kergoth | random selection of packages |
| 02:01.55 | kergoth | maybe itll boot, maybe it wont! |
| 02:01.56 | treke|home | took 5 less than 5 minutes to uninstall all of those perl modules I built :p |
| 02:04.37 | fdask | does OZ support iPaq devices? |
| 02:04.43 | treke|home | yes |
| 02:04.44 | treke|home | some |
| 02:05.37 | kergoth | fdask: oh, one thing.. the versioning is completely different between branches of development |
| 02:05.44 | kergoth | fdask: z/ipaq/simpad/etc |
| 02:05.59 | kergoth | which means the branch and changelog handling is now all insufficient |
| 02:06.00 | kergoth | hehe |
| 02:07.18 | fdask | oh cool! so you'll get that all fixed up, right? |
| 02:07.19 | fdask | :p |
| 02:07.21 | kergoth | so we need a form of logical grouping |
| 02:07.27 | kergoth | openzaurus consists of a certain set of devices |
| 02:07.33 | kergoth | opensimpad consists of another subset |
| 02:07.35 | kergoth | etc |
| 02:07.40 | kergoth | and that grouping is whats versioned |
| 02:07.53 | treke|home | then there is the odd man out, openembedded/ipaq |
| 02:07.53 | treke|home | :p |
| 02:08.04 | kergoth | yeah well, you havent released yet, have you? :P |
| 02:08.24 | treke|home | hehe. just the stuff in experimental |
| 02:08.36 | fdask | all i need to know is like the scope of the OZ page... is opensimpad/openembedded going to be worked into the oz page? |
| 02:08.45 | kergoth | i want one openembedded site |
| 02:08.53 | fdask | ah. |
| 02:08.56 | fdask | ok |
| 02:09.06 | kergoth | with the ability to break down once you've selected your section |
| 02:09.07 | kergoth | or device |
| 02:09.27 | kergoth | eventually anyway. |
| 02:09.34 | fdask | ya |
| 02:09.36 | kergoth | feeling bored? *coughs* |
| 02:09.43 | kergoth | whered noda go |
| 02:09.45 | kergoth | heh |
| 02:09.48 | fdask | he had 2 leave |
| 02:09.55 | fdask | he left me in charge |
| 02:09.56 | fdask | haha |
| 02:10.00 | kergoth | hehe |
| 02:10.05 | kergoth | are you the assistant? |
| 02:10.09 | fdask | yes sir. |
| 02:11.01 | treke|home | I like this guy |
| 02:11.53 | kergoth | we should really use our -devel mailing lists more often |
| 02:12.21 | kergoth | can essentially kill openzaurus-devel soon |
| 02:12.41 | kergoth | hehe |
| 02:12.57 | treke|home | hoping to steal some ideas for the app we're writing at work |
| 02:13.30 | fdask | so how many devices fall under the grand scope? Zaurus, iPaq, Simpad, ....? |
| 02:13.41 | treke|home | fdask: No upwards limit |
| 02:14.00 | kergoth | yep, need to be able to adjust the groupings on demand |
| 02:14.03 | treke|home | Right now? 4 or so major device cateogires |
| 02:14.22 | treke|home | fdask: Inside those categories there may be three or for subcategories |
| 02:14.37 | treke|home | in addition to seperate images |
| 02:14.47 | kergoth | subcats are usually model |
| 02:14.51 | kergoth | or a group of models |
| 02:15.43 | fdask | man... i thought this was going to php work... not sql! :p |
| 02:15.50 | kergoth | hehe |
| 02:16.03 | treke|home | if you can think of a better way to do it than sql, go for it :p |
| 02:16.10 | kergoth | yeah, didnt think about that. going to have to adjust the existing tables and add new ones |
| 02:16.26 | fdask | noda has things set up really nice right now... i was just probing a bit to see if i can update the tables |
| 02:16.46 | fdask | right now for devices there is only ipaq and zaurus listed. wanted to know what i could safely add |
| 02:17.03 | fdask | guess i could check the make targets or something |
| 02:17.15 | treke|home | fdask: Zaurus, Ipaq, SimPad |
| 02:17.23 | kergoth | he does good work. |
| 02:17.27 | kergoth | heh |
| 02:17.28 | kergoth | well |
| 02:17.34 | kergoth | could base it on the dir structure |
| 02:17.40 | kergoth | when we create a new one, a new cat appears |
| 02:18.01 | kergoth | i.e. official/[branch]/[cat]/[subcat]/etc |
| 02:18.17 | kergoth | well, we need a structure that reflects it anyway |
| 02:18.21 | fdask | ya... maybe an admin menu thatd update the db + dir structure |
| 02:18.21 | kergoth | and we need to keep them in sync |
| 02:18.37 | kergoth | so however to manage that, i dont care abuot the implementation detail |
| 02:18.39 | kergoth | heh |
| 02:18.48 | chouimat | anyone have a good book on extreme programming to recommend? |
| 02:18.57 | treke|home | cool. requesttracker3 is done. |
| 02:41.21 | treke|home | kergoth: want an rt login to play with? |
| 02:41.30 | kergoth | yeah thatd be cool |
| 02:41.52 | treke|home | http://treke.net/rt |
| 02:41.57 | treke|home | kergoth/password |
| 02:42.08 | W8TVI | anyone found a good shopping list program? |
| 02:42.20 | treke|home | W8TVI: opie-todo |
| 02:42.28 | W8TVI | no thanks... |
| 02:42.48 | kergoth | treke|home: can you give me access to add/remove groups and stuff? |
| 02:42.52 | W8TVI | I was hoping for something more like handyshoper for the palm |
| 02:42.54 | kergoth | theres no group to submit a bug on :) |
| 02:43.03 | treke|home | kergoth: I think you do |
| 02:43.04 | treke|home | lemme check |
| 02:43.06 | kergoth | k |
| 02:43.31 | treke|home | I've been using this about 1 minute longer than you :p |
| 02:43.41 | kergoth | hehe |
| 02:43.44 | kergoth | looks pretty cool though |
| 02:44.11 | treke|home | try going to groups/new group |
| 02:44.49 | kergoth | <PROTECTED> |
| 02:44.53 | kergoth | group culd not be created |
| 02:44.55 | kergoth | permissiond enied |
| 02:44.59 | kergoth | minus my typos |
| 02:45.00 | kergoth | heheh |
| 02:45.07 | *** join/#openembedded Jumpkick_ (~chatzilla@CPE0080c6f0e7ba-CM014500002511.cpe.net.cable.rogers.com) |
| 02:49.28 | treke|home | ah |
| 02:49.30 | treke|home | <PROTECTED> |
| 02:49.52 | Jumpkick_ | anyone know off hand which opie package has 'quicklauncher' in it? |
| 02:50.11 | treke|home | Jumpkick_: launcher? |
| 02:50.17 | treke|home | opie-taskbar probably |
| 02:50.26 | treke|home | kergoth: try doing things in the OpenEmbedded group |
| 02:50.27 | Jumpkick_ | ahh |
| 02:50.37 | Jumpkick_ | thanks... I'll try reinstalling that one |
| 02:51.25 | kergoth | hmm |
| 02:51.29 | kergoth | how do we submit tickets? |
| 02:51.32 | kergoth | ah, whats a queue? |
| 02:51.33 | kergoth | heh |
| 02:52.00 | treke|home | http://www.bestpractical.com/rt/docs.html |
| 02:52.10 | kergoth | the dropdown next to new ticket is empty here... |
| 02:52.11 | kergoth | hmm |
| 02:52.34 | kergoth | queue could nto be loaded |
| 02:52.40 | kergoth | so yeah, tickets are bound to queues |
| 02:53.04 | treke|home | feel free to poke about with it |
| 02:55.31 | Jumpkick_ | hmmm... is quicklauncher a symlink to qpe? |
| 02:55.43 | treke|home | Jumpkick_: shouldnt be |
| 02:57.12 | Jumpkick_ | oh... cause opie-taskbar 1.01 20030909 contains qpe but no quicklauncher... (my quicklauncher is awol) |
| 02:57.49 | treke|home | quicklaunch may not be enabled in that build |
| 02:58.04 | treke|home | or 1.01 may not have even had quicklaunch |
| 02:58.10 | Jumpkick_ | ooh |
| 02:58.31 | Jumpkick_ | crappy... I shouldn't have upgraded then... dough! :) |
| 02:58.32 | ljp | quicklauncher is an executable. quicklaunch apps are a symlink to quicklauncher |
| 02:59.07 | Jumpkick_ | yeah I know... I have the apps linking to it but the quicklauncher itself dissappeared |
| 02:59.14 | treke|home | ljp: Tell trolltech to not come up with such confusing names and no one will be confused |
| 02:59.20 | Jumpkick_ | I ran out of disk space half way though my upgrade |
| 02:59.26 | treke|home | "FastLoad" "qucklaunder" "quicklaunch" |
| 02:59.30 | Jumpkick_ | and had to clean up ans start over |
| 02:59.33 | kergoth | there we go |
| 02:59.42 | kergoth | had to give us rights within the general queue |
| 02:59.52 | treke|home | hows performance |
| 03:00.02 | treke|home | everything is sluggish here, so i cant tell well |
| 03:00.12 | ljp | well, it used to be quickexec |
| 03:00.32 | ljp | I prefer quicklunch, personally |
| 03:00.47 | ljp | or better yet - quickie-before-lunch |
| 03:00.52 | treke|home | I prefer longlunch |
| 03:01.18 | treke|home | though this week I'd settle for any lunch |
| 03:02.09 | *** join/#openembedded treke|z (~root@lsanca2-ar29-4-41-064-167.lsanca2.elnk.dsl.genuity.net) |
| 03:02.49 | ljp | luches are included here |
| 03:02.56 | ljp | lunch, even |
| 03:03.18 | treke|z | bastard |
| 03:03.24 | Jumpkick_ | lol... I like lunch too :) |
| 03:04.13 | kergoth | treke|home: performance, doesnt seem too bad. heh, these 'scrip' things are cool. can assocate actions to events |
| 03:04.31 | kergoth | hm |
| 03:04.32 | treke|z | cool |
| 03:04.43 | kergoth | i like that you can add custom fields too |
| 03:04.48 | kergoth | for whatever data we need |
| 03:05.01 | treke|z | that machine is only a lowly dual p2 400 |
| 03:05.07 | treke|z | :P |
| 03:06.13 | Jumpkick_ | I think I found it... must have not had opie-quicklauncher reinstalled |
| 03:08.03 | fdask | yikes |
| 03:09.09 | W8TVI | 2.51, 2.30, 2.11 |
| 03:09.49 | Jumpkick_ | yeah that's what I thought |
| 03:09.51 | Jumpkick_ | lol |
| 03:11.50 | W8TVI | and I have been doing installs and running opera |
| 03:13.28 | W8TVI | yes! I found the neocal calulator thats free |
| 03:13.33 | W8TVI | anyone need it? |
| 03:16.31 | kergoth | There is no need to reply to this message right now. Your ticket has been |
| 03:16.35 | kergoth | assigned an ID of [treke.net #2]. |
| 03:16.35 | kergoth | heh, cool |
| 03:17.26 | kergoth | Reply-To: "RT::CorrespondAddress.not.set"@ |
| 03:17.27 | kergoth | hehe |
| 03:17.58 | treke|z | hehe |
| 03:18.11 | absentia | heheh |
| 03:18.20 | treke|z | kinda skimmed over that part of the config file :p |
| 03:20.55 | ljp | k. time to maybe do something here |
| 03:22.03 | kergoth | i like it |
| 03:22.12 | kergoth | seems nicer than bugzilla or mantis |
| 03:22.33 | kergoth | wonder if theres a way to submit bugs without being a user, short of a guest account |
| 03:22.46 | ljp | what bug tracker? |
| 03:22.46 | kergoth | brb |
| 03:23.57 | ljp | nm |
| 03:24.48 | treke|home | there. configured the mail stuff |
| 03:29.01 | kergoth | hmm |
| 03:29.21 | absentia | h0h0h0 |
| 03:30.34 | kergoth | heh, i like how it tracks progress and time worked |
| 03:30.39 | kergoth | spiffy |
| 03:30.59 | *** join/#openembedded stigger (stigger@p076-n1-gos.nlc.net.au) |
| 03:32.17 | kergoth | just buy a c760 |
| 03:32.57 | W8TVI | my mom is looiking at getting a sony nx80 |
| 03:35.16 | treke|home | why would you want the 5500 keyboard |
| 03:35.17 | treke|home | ick |
| 03:45.19 | *** join/#openembedded stigger (stigger@p076-n1-gos.nlc.net.au) |
| 03:46.03 | kergoth | hmm |
| 03:46.07 | kergoth | wonder if rt has any reporting capabilities |
| 03:46.15 | kergoth | like for generating charts of progress and the like |
| 03:46.37 | kergoth | or if not, maybe an external addon or tool |
| 03:52.09 | *** join/#openembedded dL-DeBaan (~debaan@cs6625218-171.austin.rr.com) |
| 03:58.57 | W8TVI | I like the keyboard on the 5500 |
| 04:01.39 | *** join/#openembedded CIA (CIA@cia.utility.freenode) |
| 04:03.35 | kergoth | W8TVI: you obviously havent tried a c7x0 :P |
| 04:07.37 | *** join/#openembedded slowhog (~henryjen@12-210-10-137.client.attbi.com) |
| 04:09.14 | *** join/#openembedded dL-DeBaan_ (~debaan@cs6625218-171.austin.rr.com) |
| 04:15.18 | cyn_ | about how big is buildroot after everythings compiled? |
| 04:15.35 | kergoth | cyn_: few gigs |
| 04:15.39 | cyn_ | like full sources plus output binaries (if you compiled for one device and all the stuff) |
| 04:15.43 | cyn_ | y0w |
| 04:15.47 | kergoth | heh |
| 04:15.51 | cyn_ | I'll be network mounting then |
| 04:16.07 | cyn_ | it's about time I gave it another go |
| 04:17.01 | cyn_ | erm |
| 04:17.08 | cyn_ | kergoth: does it rely on symlinks or anything |
| 04:17.23 | kergoth | buildroot itself, no. |
| 04:17.28 | kergoth | the individual builds? who knows |
| 04:17.30 | cyn_ | the environ im not worried about |
| 04:17.31 | cyn_ | heh |
| 04:17.35 | cyn_ | I'll find out for ya! |
| 04:17.43 | kergoth | heheh |
| 04:18.16 | treke|home | you need symlinks |
| 04:18.22 | treke|home | or else you wont be building any images |
| 04:18.24 | fdask | 2.1G . |
| 04:18.35 | fdask | du -h run in my buildroot dir ^^^ |
| 04:18.41 | treke|home | or any packages |
| 04:18.42 | dL-DeBaan_ | do you need any mirrors? |
| 04:18.45 | cyn_ | treke|home: can you set the output dirs for that stuff |
| 04:18.52 | cyn_ | well obviously you can |
| 04:18.56 | cyn_ | can you do it relatively simply :) |
| 04:19.02 | cyn_ | aka not hand editing everything |
| 04:19.08 | treke|home | cyn_: The contents of the files require symlinks |
| 04:19.54 | treke|home | output/rootfs/tmp is a symlink |
| 04:20.12 | dL-DeBaan_ | kergoth: just wondering how y'all are doing for bandwidth, or if you need more volunteer mirrors... (i've got 70GB/month of bandwidth that's a-wasting) |
| 04:20.26 | treke|home | each library has a couple symlinks attached to it |
| 04:20.27 | cyn_ | dL-DeBaan_: that's some nice bandwidth |
| 04:20.28 | kergoth | dL-DeBaan_: could definately use mirrors. for feeds/images, or web, or both |
| 04:20.38 | cyn_ | treke|home: :( |
| 04:21.11 | kergoth | cyn_: vfat? |
| 04:21.16 | cyn_ | ya |
| 04:22.23 | dL-DeBaan_ | kergoth: k. will you email me at openembedded@aspire-is.com with details of whatever info you need from me, and so i can pass on my info to you? |
| 04:22.31 | kergoth | will do |
| 04:22.36 | kergoth | cyn_: use umsdos? |
| 04:22.38 | dL-DeBaan_ | thx |
| 04:22.39 | cyn_ | could umsdos it I suppose maybe, but it's just gunna be smbmounted |
| 04:22.47 | kergoth | ah |
| 04:23.03 | cyn_ | which wouldn't really work afaik |
| 04:24.31 | *** join/#openembedded treke|z (~root@lsanca2-ar29-4-41-064-167.lsanca2.elnk.dsl.genuity.net) |
| 04:26.37 | *** join/#openembedded W8TVI (~root@1114104797.mi.dial.hexcom.net) |
| 04:32.11 | kergoth | ibot: bleh |
| 04:32.11 | | kergoth: have you tried http://www.tldp.org/ ? |
| 04:39.39 | treke|z | wtf |
| 04:40.12 | treke|z | i misread the clock. it isnt 11:30:p |
| 04:43.11 | ljp | nope. its 14:40 |
| 04:43.40 | W8TVI | damn it... I hate #redhat |
| 04:43.49 | W8TVI | they ignore you most of the time |
| 04:45.43 | treke|z | did someone say something? |
| 04:46.23 | W8TVI | treke, yep... mostly like tha |
| 04:46.23 | W8TVI | t |
| 04:50.13 | ljp | did I hear a fly buzzing? |
| 04:51.51 | W8TVI | ~lart the fly |
| 04:52.04 | W8TVI | heh |
| 04:53.15 | *** join/#openembedded CIA (CIA@cia.utility.freenode) |
| 04:55.11 | *** join/#openembedded heatxsink (~jack@12-234-106-86.client.attbi.com) |
| 04:55.20 | heatxsink | ibot seen TCMAN |
| 04:55.21 | | tcman <~dbehr@adsl-64-170-122-35.dsl.snfc21.pacbell.net> was last seen on IRC in channel #zaurus, 2d 6h 59m 8s ago, saying: 'hi'. |
| 04:55.40 | *** join/#openembedded ulmer (~ulmer@wsip-24-136-46-195.ga.at.cox.net) |
| 04:56.45 | *** join/#openembedded CIA (CIA@cia.utility.freenode) |
| 05:08.47 | W8TVI | hey kergoth |
| 05:08.53 | W8TVI | http://www.hacksrus.com/~recompiler/Projects/Customization.htm |
| 05:24.50 | *** part/#openembedded ulmer (~ulmer@wsip-24-136-46-195.ga.at.cox.net) |
| 05:31.14 | *** join/#openembedded Bala (~Bala@hsdbsk142-165-117-29.sasknet.sk.ca) |
| 05:32.41 | cyn_ | W8TVI: that looks like a fun device |
| 05:33.38 | W8TVI | it is... but it is giving me headaches right now... |
| 05:34.20 | W8TVI | It'll go and ignore all the computers pluged into it for 3 mins (or so) at random... then be fine |
| 05:34.32 | W8TVI | and I cant see any error in the log for it... |
| 05:34.57 | W8TVI | I though it might be that ethernet problem that was in rh7.3 |
| 05:35.13 | W8TVI | but it would give me timeouts in the log about it |
| 06:25.11 | *** join/#openembedded schurig (~schurig@pD951A4E5.dip.t-dialin.net) |
| 06:25.59 | schurig | opps, debian deleted netbase_3.18 ... |
| 06:26.31 | *** join/#openembedded CAMCAM (~camrein@AMontsouris-108-1-17-54.w80-15.abo.wanadoo.fr) |
| 06:28.51 | W8TVI | why dont buzzme work anymore |
| 06:29.27 | ljp | buzzme? |
| 06:29.43 | W8TVI | for kismet |
| 06:31.25 | W8TVI | Error Getting Buzzer Volume: Function not implemented |
| 06:31.25 | W8TVI | buzzme:main: Problems getting volume on device '/dev/sharp_buz'. |
| 06:31.35 | W8TVI | use to work |
| 06:32.02 | W8TVI | anyone happen to be using kismet 2.8.1? |
| 06:32.22 | W8TVI | and could try running buzzme -j |
| 06:34.05 | schurig | hehe, but I found the same package at some Sony site ;-) |
| 07:00.10 | *** join/#openembedded lsmith (~vandal@p213.54.25.128.tisdip.tiscali.de) |
| 07:02.31 | *** part/#openembedded slowhog (~henryjen@12-210-10-137.client.attbi.com) |
| 07:09.05 | W8TVI | how do you make sound on the buzzer? |
| 07:22.40 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 07:28.34 | *** join/#openembedded funkyjoe (~funkyjoe@62.153.88.6) |
| 07:29.16 | funkyjoe | good morning! |
| 07:29.22 | W8TVI | hi |
| 07:29.51 | funkyjoe | what happend to the packages ppp in buildroot-oz? it seems to be broken, any ideas about this? |
| 07:30.41 | W8TVI | night |
| 07:36.32 | funkyjoe | w8tvi: ? good night? ideas about my ppp compile error, patch did not work. w8tvi, callsign, tvi ;-)) |
| 07:48.47 | *** join/#openembedded HCl (~HCl@acrux.student.utwente.nl) |
| 07:55.43 | HCl | ibot, oz buildroot |
| 07:55.43 | | it has been said that oz bk is bk clone bk://openzaurus.bkbits.net/buildroot buildroot-oz; cd buildroot-oz; bk -r co -q; cp def-configs/collie-opie .config (for zaurus 5000d/5500) .. make oldconfig; make; or see bitkeeper (free download but non-gpl) or see 'oz snapshot' if you either cannot or will not use bitkeeper. |
| 08:01.52 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) |
| 08:27.22 | HCl | does the .21 kernel load sdmmc yet? |
| 08:34.00 | *** join/#openembedded lsmith (~vandal@port-212-202-39-108.reverse.qsc.de) |
| 08:36.29 | mickeyl | HCl: machine? |
| 08:36.58 | HCl | 5500 |
| 08:39.36 | mickeyl | HCl: no, the sharp binary module will never function in 2.4.21-cl. Someone has to implement 5x00 support on top of the base work of hh.org |
| 08:39.57 | HCl | igh.. |
| 08:39.59 | HCl | ok.. |
| 08:40.15 | HCl | hh.org? |
| 08:41.00 | HCl | why wont it function in 21? |
| 08:42.36 | mickeyl | too many changes in the block driver layer |
| 08:43.33 | mickeyl | handhelds.org implemented a driver for the ipaqs |
| 08:43.56 | mickeyl | now anyone must implement the necessary glue between the sharp custom chips and the base code. |
| 08:43.56 | HCl | ah... |
| 08:44.02 | mickeyl | that is work for a kernel expert |
| 08:44.06 | mickeyl | which we don't have |
| 08:44.16 | HCl | mmmnn... |
| 08:44.16 | mickeyl | except kergoth, who is busy with other things |
| 08:44.17 | mickeyl | so... |
| 08:44.24 | HCl | sigh. |
| 08:44.24 | mickeyl | on to recruiting kernel people... ); |
| 08:44.45 | HCl | well, i can do some basic things with the kernel.. but.. i dont think i could implement new hardware |
| 08:45.41 | mickeyl | moin Twiun |
| 08:54.37 | *** join/#openembedded stief (~stief@213-136-25-234.adsl.bit.nl) |
| 09:03.39 | *** join/#openembedded SuKoSh| (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 09:15.44 | HCl | where is the homepage of the driver for the ipa;? |
| 09:15.48 | HCl | ipaq* |
| 09:16.24 | yeiazel | ibot: sdmmc |
| 09:16.25 | | yeiazel: I wish you would RTFM. |
| 09:16.29 | yeiazel | ibot: you suck |
| 09:16.29 | | :( |
| 09:16.46 | mickeyl | cvs.handhelds.org |
| 09:19.17 | HCl | nm.. i'll just take your word for it that it exists :s |
| 09:21.32 | HCl | so i guess it'll take a long time before there'll be a decent kernel for 5500..? |
| 09:22.48 | mickeyl | depends. since there are only two or three things missing, it would take an expert only a couple of weeks. |
| 09:23.02 | HCl | hm. |
| 09:23.22 | mickeyl | kergoth finished the most things |
| 09:23.53 | mickeyl | missing is (iirc) sound,apm and sd |
| 09:24.10 | HCl | hm |
| 09:24.19 | HCl | sound is rather vital too.. but sd is most important for me.. |
| 09:30.42 | *** join/#openembedded bisho (~bisho@136.Red-213-97-191.pooles.rima-tde.net) |
| 09:34.37 | mickeyl | HCl: alternatively you can try getting the sharp embedix 2.4.18 for collie to work with OpenZaurus. might be an easier task than finishing 2.4.21-cl |
| 09:35.27 | HCl | well, the sharp version of the kernel wont work with openzaurus |
| 09:35.34 | HCl | it insists on cramfs |
| 09:35.45 | HCl | so i'd have to build a new one |
| 09:35.45 | mickeyl | yes, that's what i mean |
| 09:35.58 | mickeyl | getting it to work as in patching it so it works with OpenZaurus |
| 09:36.01 | HCl | george built a version of embedix for the 5500, but it didnt work |
| 09:36.10 | HCl | where can i get the patches i need? |
| 09:36.20 | mickeyl | you have to write them |
| 09:36.28 | HCl | eh.. aha. |
| 09:36.32 | HCl | i'll just wait x.x |
| 09:38.12 | HCl | the problem with kernels < 2.4.9 is that they leak resources a lot.. |
| 09:42.37 | *** join/#openembedded stigger (stigger@p045-n1-gos.nlc.net.au) |
| 09:45.12 | mickeyl | i know |
| 09:45.31 | mickeyl | i queried for participation over a lot of communication channels |
| 09:45.40 | mickeyl | but nothing happens |
| 09:45.45 | HCl | -.- |
| 09:45.48 | HCl | annoying.. |
| 09:45.52 | mickeyl | yeahj |
| 09:45.57 | *** join/#openembedded Shdwdrgn (Picard@sourpuss.net) |
| 09:45.59 | andersee | mickeyl: btw |
| 09:46.35 | andersee | mickeyl: I just noticed... I forgot to copy a rather important gcc patch from the uclibc buildroot into the uClibc toolchain area. |
| 09:46.53 | mickeyl | andersee: ah, good. could that be the source of my problem? |
| 09:47.17 | andersee | mickeyl: w/o that c++ will be in pretty sorry shape |
| 09:50.35 | mickeyl | andersee: another thing, is your toolchain build to finish without error messsages in between? here and there I spot a "command not found" while its doing its thing, but it finished and it seems I have a complete toolchain. |
| 09:51.01 | andersee | I had no problems with it. |
| 09:51.14 | andersee | The version in the uClibc buildroot worked for me. |
| 09:51.43 | andersee | I built the whole thing for arm last night. Copied it to my netwinder and chrooted in, and everything was working as expected. |
| 09:52.37 | mickeyl | andersee: did you remember it run through without any error? (i mean as in spitting out an error message but continuing nevertheless) |
| 09:53.11 | mickeyl | s/did/do |
| 09:54.26 | andersee | mickeyl: I didn't watch it build. Was busy coding... |
| 09:55.57 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) [NETSPLIT VICTIM] |
| 09:55.57 | *** mode/#openembedded [+bbb *!*@12-227-17-69.client.attbi.com *!*@12-227-27-36.client.attbi.com] *!*@*12-227-27-36.client.attbi.com] by jordan.freenode.net |
| 10:04.06 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) |
| 10:04.21 | mickeyl | andersee: I'm always using the $CVS/toolchain - would I be better off using $CVS/buildroot to build a 3.3.1-uclibc toolchain? |
| 10:04.49 | andersee | depends on what you are planning on doing with it |
| 10:05.01 | andersee | using the toolchain version should work. |
| 10:05.12 | andersee | As I try to keep them closely in sync |
| 10:05.20 | mickeyl | andersee: well... i just want to migrate our OpenZaurus buildroot to using uclibc. And for this I need a working 3.3.1-uclibc toolchain ;) |
| 10:05.35 | mickeyl | the last one worked great up to the C++ stuff |
| 10:05.44 | andersee | last one? |
| 10:06.09 | andersee | anyways |
| 10:06.14 | mickeyl | when I tried last time (several weeks ago), i used a 3.3-uclibc toolchain with uclibc .20 |
| 10:06.27 | mickeyl | which went fine (also compiled libxine) until the C++ apps. |
| 10:06.33 | andersee | The toolchain one now has gcc-810-libstd++-locale.patch, which is necessary for c++ to work |
| 10:06.36 | mickeyl | now this time I don't get that far due to the shadowing thing |
| 10:07.05 | mickeyl | ah great. now just having to get over libxine |
| 10:16.30 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) |
| 10:27.34 | mickeyl | andersee: here come the first errors (or warnings?), right after configuring uclibc |
| 10:28.19 | mickeyl | Build the shared library loader with early debugging support (SUPPORT_LD_DEBUG_EARLY) [N/y/?] n |
| 10:28.19 | mickeyl | Manuel's hidden warnings (UCLIBC_MJN3_ONLY) [N/y/?] n |
| 10:28.19 | mickeyl | make[1]: Leaving directory `/var/compile/gcc-3.3.1/build_arm/uClibc-0.9.21' |
| 10:28.19 | mickeyl | make -C /var/compile/gcc-3.3.1/build_arm/uClibc-0.9.21 headers install_dev; |
| 10:28.33 | mickeyl | bin/sh: line 1: /usr/local/arm/3.3.1uc/bin/arm-uclibc-gcc: No such file or directory |
| 10:28.54 | mickeyl | + CC=/usr/local/arm/3.3.1uc/bin/arm-uclibc-gcc |
| 10:28.54 | mickeyl | + /bin/sh extra/scripts/gen_bits_syscall_h.sh |
| 10:28.54 | mickeyl | extra/scripts/gen_bits_syscall_h.sh: line 18: /usr/local/arm/3.3.1uc/bin/arm-uclibc-gcc: No such file or directory |
| 10:28.54 | mickeyl | extra/scripts/gen_bits_syscall_h.sh: line 15: /usr/local/arm/3.3.1uc/bin/arm-uclibc-gcc: No such file or directory |
| 10:28.54 | mickeyl | + cmp include/bits/sysnum.h include/bits/sysnum.h.new |
| 10:29.01 | mickeyl | but as i've said... the build continues |
| 10:29.17 | mickeyl | so I'm really curious to know if these are really errors or i should ignore them |
| 10:29.23 | chouimat | morning |
| 10:29.31 | mickeyl | hi ChanServ |
| 10:29.32 | mickeyl | argh |
| 10:29.35 | mickeyl | ~lart tab completion |
| 10:29.39 | mickeyl | hi chouimat |
| 10:32.19 | mickeyl | bbl |
| 10:36.42 | andersee | mickeyl|lunch: you do not need to worry about that |
| 10:37.18 | andersee | mickeyl|lunch: If you needed to worry, I would have made it abort, not continue... |
| 10:45.35 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) [NETSPLIT VICTIM] |
| 10:45.35 | *** mode/#openembedded [+bbb *!*@12-227-17-69.client.attbi.com *!*@*12-227-27-36.client.attbi.com *!*@12-227-27-36.client.attbi.com]] by jordan.freenode.net |
| 10:53.05 | *** join/#openembedded mithro (~tim@dsl1-83.gw1.adl1.airnet.com.au) |
| 10:56.08 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) [NETSPLIT VICTIM] |
| 10:56.08 | *** mode/#openembedded [+bbb *!*@12-227-17-69.client.attbi.com *!*@*12-227-27-36.client.attbi.com *!*@12-227-27-36.client.attbi.com]] by jordan.freenode.net |
| 11:07.38 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 11:24.30 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 11:24.30 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) [NETSPLIT VICTIM] |
| 11:24.30 | *** mode/#openembedded [+bbb *!*@12-227-17-69.client.attbi.com *!*@*12-227-27-36.client.attbi.com *!*@12-227-27-36.client.attbi.com]] by jordan.freenode.net |
| 11:35.04 | *** join/#openembedded schurig (~schurig@pD951A64A.dip.t-dialin.net) |
| 11:41.14 | mickeyl | andersee: ok, thanks for reassuring that. the toolchain build is finished, I'll attempt a new OZ build now. |
| 11:41.27 | andersee | mickeyl: k |
| 11:41.38 | andersee | do lemme know if you have any problems |
| 11:54.20 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 11:54.20 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) [NETSPLIT VICTIM] |
| 11:54.20 | *** mode/#openembedded [+bbb *!*@12-227-17-69.client.attbi.com *!*@*12-227-27-36.client.attbi.com *!*@12-227-27-36.client.attbi.com]] by jordan.freenode.net |
| 11:59.49 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 12:09.51 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) [NETSPLIT VICTIM] |
| 12:09.51 | *** mode/#openembedded [+bbb *!*@12-227-17-69.client.attbi.com *!*@*12-227-27-36.client.attbi.com *!*@12-227-27-36.client.attbi.com]] by jordan.freenode.net |
| 12:29.20 | *** join/#openembedded cmartin (~Chris@gw.krypnet.de) |
| 12:42.04 | mickeyl | andersee: bailed out on conftest. |
| 12:42.09 | mickeyl | error is: |
| 12:42.33 | mickeyl | hidden symbol `__udivsi3' in /usr/local/arm/3.3.1uc/lib/gcc-lib/arm-linux/3.3.1/libgcc. |
| 12:42.33 | mickeyl | a(_udivsi3.oS) is referenced by DSO |
| 12:43.13 | andersee | mickeyl: can you try binutils-2.14.90.0.6 and let me know if you still see that? |
| 12:43.43 | mickeyl | andersee: I must rebuild the toolchain for that, right? |
| 12:44.00 | andersee | sadly |
| 12:44.03 | andersee | yes |
| 12:44.13 | mickeyl | andersee: no problem. will do that. |
| 12:46.42 | Twiun | how long does it take? |
| 12:47.15 | mickeyl | Twiun: on this machine here, around one hour |
| 12:48.44 | Twiun | that's not too bad I suppose |
| 12:48.48 | mickeyl | yeah |
| 12:49.20 | Twiun | btw, anyone wanting a stab at the OE blurb, OpenEmbeddedBlurb on the OE wiki is ready for your mangling |
| 12:52.06 | mickeyl | andersee: binutils-906-hjl_libtool_dso.patch does not apply for the latest binutils |
| 12:52.22 | mickeyl | andersee: however it spits out "previously applied", so it may has found its way upstream? |
| 12:52.52 | mickeyl | well not completely obviously |
| 12:53.51 | cyn_ | haha okay, so if I were to smbmount some vfat space from my win box to my laptop, then have a file there that I loopback mounted as type ext2 ... THAT would work to let me compile buildroot - right? |
| 12:54.28 | cyn_ | performance would suffer more than native, god only knows how badly - but it'd work |
| 12:57.02 | *** join/#openembedded bobbyd (~none@217.204.66.43) |
| 12:59.36 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) [NETSPLIT VICTIM] |
| 12:59.36 | *** mode/#openembedded [+bbb *!*@12-227-17-69.client.attbi.com *!*@*12-227-27-36.client.attbi.com *!*@12-227-27-36.client.attbi.com]] by jordan.freenode.net |
| 13:29.56 | *** part/#openembedded bobbyd (~none@217.204.66.43) |
| 13:41.55 | *** join/#openembedded DanWilliams (~DanWillia@proxyle01.ext.ti.com) |
| 13:48.43 | *** join/#openembedded prpplague (~billybob@12.148.134.14) |
| 14:13.20 | *** join/#openembedded sieve_work (~klu5ter@12.148.134.14) |
| 14:34.12 | cyn_ | blah smbfs performance gets pegged when creating lots of tiny files and I get i/o errors during the process, even when grabbing the repository locally then copying it over heh |
| 14:40.05 | *** join/#openembedded stigger (stigger@p027-n1-gos.nlc.net.au) |
| 14:44.49 | *** join/#openembedded TimRiker (~timr@proxyle02.ext.ti.com) |
| 14:46.44 | *** join/#openembedded cmartin (~cmartin@gw.krypnet.de) |
| 14:51.45 | *** join/#openembedded __stigger (stigger@p060-n1-gos.nlc.net.au) |
| 14:51.50 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 14:52.11 | *** join/#openembedded Pixel_ (~chatzilla@212.112.131.169) |
| 14:56.38 | Pixel_ | Q: Which os (FreeBSD, RedHat, Debian...) should i get in order to compile the newest version of OZ for my C750 ? |
| 14:56.57 | yeiazel | OS/2 Warp 3 |
| 14:57.05 | yeiazel | or MS-DOS 5.0 |
| 14:57.06 | Pixel_ | Nah |
| 14:57.12 | Pixel_ | your pulling my leg |
| 14:57.14 | Pixel_ | haha |
| 14:57.19 | gb2 | AmigaOS 1.1 |
| 14:57.39 | Pixel_ | I've got an old Newton 110 with NewtOS, will that do ? |
| 15:01.09 | schurig | Pixel_: debian or gentoo might not be bad |
| 15:03.41 | Pixel_ | Okay |
| 15:03.50 | Twiun | Pixel_: I've got mdk9.1 - no problems |
| 15:03.53 | Pixel_ | What would i need to get afterwards ? |
| 15:04.04 | Pixel_ | Mandrake 9.1 ? |
| 15:04.36 | mickeyl | mdk91 is a good choice |
| 15:04.40 | cmartin | Pixel_: I've used built from MDK 9.1, gentoo and debian without problems |
| 15:05.06 | Pixel_ | So it's more or less a matter of choice ? |
| 15:05.14 | cmartin | yep |
| 15:05.38 | Twiun | wonderful world, eh? |
| 15:07.15 | cmartin | Pixel_: after that you need a Toolchain (cross-compiler), bitkeeper client software and maybe make sure certain libs are installed |
| 15:07.47 | cmartin | Pixel_: oh and cvs client too |
| 15:08.24 | Pixel_ | Hmm okay |
| 15:08.42 | Pixel_ | Thats not included in any os ? |
| 15:09.40 | cmartin | yes some dists have all those - gentoo, mdk and debian has everything you need |
| 15:10.12 | Pixel_ | Wh00t, then i'm getting one of those. |
| 15:10.19 | cmartin | except for the toolchain - you will have to get that from openzaurus.org and unpack it under /usr/local/arm |
| 15:10.57 | Pixel_ | Which is most userfriendly of gentoo and mdk ? |
| 15:11.16 | cmartin | and in order to use the free cersion of bitkeeper you have to register by bitkeeper too |
| 15:11.22 | cmartin | mdk |
| 15:11.40 | cmartin | gentoo is pretty hard-core linux from scratch |
| 15:12.06 | cmartin | if your used to working under M$ then get mdk |
| 15:12.11 | Pixel_ | :D |
| 15:13.44 | cmartin | advantage for mdk is that it is ~98% compatible to RedHat, so you can also use RedHat RPMs if there's none available for mdk - and urpmi is much more comfortable than the basic rpm |
| 15:14.19 | Pixel_ | cool |
| 15:14.55 | Pixel_ | Gonna install it on a WMWare image at work and use it for compiling newest OZ and work relating to my Computer Science courses |
| 15:15.12 | cmartin | but my laptop had problems with the mdk kernel, so i'm using debian at the time - mdk is very user-friendly |
| 15:15.28 | Pixel_ | goodie |
| 15:16.02 | Pixel_ | downloading : 2.4 mb/sec |
| 15:16.08 | Pixel_ | mbyte |
| 15:16.22 | cmartin | that would work fine if you have enough memory ;) |
| 15:18.17 | Pixel_ | hey, they had a max on each connection. Downloading the 3 cd's with 2.2 Mbyte pr connection = 6.6 Megabyte pr. sec - i love my works connection :D |
| 15:18.40 | Twiun | yeah, sounds like a nice thing to have |
| 15:19.03 | Twiun | don't suppose they'd like to volunteer to provide mirroring facilities? |
| 15:19.06 | Pixel_ | Max speed i've ever reached was 11 megabyte pr. sec |
| 15:19.25 | Pixel_ | Nah, they are are the biggest internet portal in denmark ( jubii.dk ) |
| 15:23.07 | cmartin | mickeyl: did you get my message about the chvt problem? |
| 15:24.42 | mickeyl | cmartin: no? via email? |
| 15:27.18 | cmartin | no via "/msg" |
| 15:27.26 | cmartin | i'll repeat |
| 15:28.10 | cmartin | when opie is running and i run chvt 1 on the serial login i get messages on the FB... |
| 15:28.31 | cmartin | when i run chvt 2 i get opie on FB, but... |
| 15:28.50 | *** join/#openembedded TimRiker (~timr@proxyle01.ext.ti.com) |
| 15:29.07 | mickeyl | well - if you complain about a missing display refresh... well... how should that be, that is not implemented. |
| 15:29.17 | cmartin | the screen isn't refreshed, so i have to tap on one of the (non-visible) tabs at the top to get most of the screen refreshed |
| 15:30.03 | mickeyl | we have to live with that for now |
| 15:30.08 | cmartin | well we could write two scripts for changing back and forth - using your trick for resume |
| 15:30.16 | mickeyl | yes that may possible |
| 15:30.16 | cmartin | ;) |
| 15:30.19 | mickeyl | but a hack :) |
| 15:30.31 | Twiun | linux is full of hacks, one more won't hurt |
| 15:30.31 | mickeyl | however, VT support for Opie is currently being written... |
| 15:30.41 | mickeyl | so we will have it in later versions. |
| 15:31.18 | cmartin | yes, i can live with that for now - as long as you know how to get something on the screen it's ok |
| 15:32.06 | cmartin | i could just post those scripts to the forum as a workaround or just manually put them in the next image |
| 15:32.32 | *** join/#openembedded Neo|Work (~neotron@proxyle01.ext.ti.com) |
| 15:32.59 | mickeyl | cmartin: might put them manually into the image, I don't know if the Opie VT support will appear in 1.0.3 or 1.2 |
| 15:35.20 | cmartin | BTW: i made that change to /etc/init.d/opie in buildroot for the kernel messages - i will move the banner script manually in the image, since it looks bad on the reconfiguring screen when modprobe is run after it |
| 15:36.13 | mickeyl | cmartin: okay. the whole situation will be relieved with that VT support, so I wouldn't feel comfortable with doing it for all archs in buildroot., |
| 15:36.23 | *** join/#openembedded Zippo_ (~thomas@p508C33A8.dip.t-dialin.net) |
| 15:36.26 | cmartin | yes |
| 15:37.07 | mickeyl | bbl - wandering home |
| 15:37.14 | cmartin | by |
| 15:37.16 | mickeyl | cu |
| 15:41.29 | kergoth | mickeyl|away: its currently being written? by whom? :) |
| 15:43.11 | kergoth | ibot: thursday |
| 15:43.11 | | Thursday sucks, because it isnt friday (See friday). |
| 15:46.21 | Twiun | hey kergoth |
| 15:46.44 | kergoth | hey Twiun |
| 15:46.46 | kergoth | how goes it? |
| 15:47.16 | Twiun | dreadfully... ugly code at work :/ |
| 15:47.35 | Twiun | btw, added a wiki entry for cyn's blurb, edit at leisure |
| 15:47.43 | Twiun | --> http://www.openembedded.org/phpwiki/index.php/OpenEmbeddedBlurb |
| 15:53.43 | treke | morning |
| 15:53.49 | kergoth | ah, cool |
| 15:53.53 | SuKoShi | hi |
| 15:54.23 | kergoth | Twiun: hmm, maybe i should make sure it works in windows under cygwin to make the blurb sound better. |
| 15:54.26 | kergoth | Twiun: :) |
| 15:57.17 | kergoth | we had remnant patched test defaults when crosscompiling |
| 15:57.25 | kergoth | no need for that if we just fully populate the cache |
| 15:58.14 | prh | kergoth: do you got OE online anywhere yet? |
| 15:58.20 | gb2 | kergoth: package building.. does it properly build root owned stuff in the packages if you build as a different user? |
| 15:58.21 | gb2 | hrm. |
| 15:58.51 | kergoth | prh: openembedded.bkbits.net |
| 15:58.56 | gb2 | doesn't seem to. |
| 15:59.01 | kergoth | gb2: ipk creation stomps perms |
| 15:59.11 | gb2 | rpm creation doesn't |
| 15:59.14 | kergoth | gb2: as does mkfs.{jffs2,cramfs} at image creation time |
| 15:59.15 | kergoth | correct |
| 15:59.18 | kergoth | we need a solution for that |
| 15:59.21 | kergoth | and tarball creation too |
| 15:59.23 | gb2 | %defattr(-,root,root) |
| 15:59.27 | gb2 | in %files for rpm |
| 15:59.27 | kergoth | ah |
| 15:59.34 | kergoth | go ahead and alter the package_rpm class |
| 15:59.37 | kergoth | in oe/bin/classes/ |
| 15:59.45 | prh | eugh more bk ! :O( |
| 15:59.51 | gb2 | yeah, i know that one is an easy fix |
| 15:59.53 | kergoth | send the patch to openembedded.bkbits.net or send me a ssh v1 pub key and i'll give you access to the repos |
| 16:00.08 | gb2 | i was actually interested in the tarball creation. |
| 16:00.25 | kergoth | we need to put ${ARCH} in the tar filename |
| 16:00.29 | gb2 | and i can't think of a simple fix for that |
| 16:00.36 | kergoth | the other day i had x86 and arm versions stomping on one another |
| 16:00.37 | kergoth | hehe |
| 16:00.43 | gb2 | heh |
| 16:01.02 | kergoth | gb2: the ipkg-build perm stomping just passes a parameter to tar |
| 16:01.10 | kergoth | gb2: so just need to do that for the tar creation as well |
| 16:01.26 | gb2 | some generic method for specifying stuff like suid files or specific permissions/ownership on files in packages will become necessary too. |
| 16:01.33 | kergoth | suid is easy |
| 16:01.38 | kergoth | but you're right |
| 16:01.52 | kergoth | suid you just suid it.. the fact that its owned by you doenst matter since we'll stomp that anyway |
| 16:01.59 | gb2 | yeah. |
| 16:02.04 | kergoth | i like the device table |
| 16:02.08 | kergoth | used by mkfs.jffs2 |
| 16:02.14 | kergoth | can create devices or set file perms |
| 16:02.24 | kergoth | need an impl of that for tar and rpm :) |
| 16:02.24 | gb2 | yeah, that sort of solution is probably the way |
| 16:03.29 | gb2 | with rpms, in the %files section, you can do a %attr(perm,owner,group) and the %defattr sets the defaults for files without specific %attr definitions. (the - for perms means use the perms the file actually has). |
| 16:04.07 | gb2 | some table in the .oe file which serves the same purpose is probably the solution. |
| 16:04.24 | kergoth | yeah, but we need a core table too |
| 16:04.27 | gb2 | also allows you to put specials into packages too. |
| 16:04.29 | kergoth | for the image creation anyway |
| 16:04.34 | gb2 | *nod* |
| 16:04.35 | kergoth | for /dev/ initial population |
| 16:04.41 | kergoth | hmm |
| 16:04.49 | kergoth | so rpm is pretty easy |
| 16:04.58 | kergoth | tar/ipk is the real challenge |
| 16:04.59 | kergoth | hmm |
| 16:05.03 | gb2 | tar is probably hard unfortunately. |
| 16:05.13 | kergoth | hehe |
| 16:05.24 | gb2 | it's scary that the easiest way i can think of is LD_PRELOAD=... |
| 16:05.35 | kergoth | lol |
| 16:05.40 | kergoth | how .. disgusting |
| 16:05.45 | gb2 | i'm thinking the debian fakeroot method |
| 16:05.51 | gb2 | which i hate, personally. |
| 16:05.53 | kergoth | i'd rather patch tar to accept a device table |
| 16:06.00 | *** join/#openembedded Harlekin (~max@pD9E591EF.dip.t-dialin.net) |
| 16:06.07 | kergoth | swipe the dev table parsing bits from mkfs.jffs2/mkcramfs |
| 16:06.15 | Harlekin | re |
| 16:06.28 | *** join/#openembedded Neo|Work (~neotron@proxyle01.ext.ti.com) |
| 16:07.01 | gb2 | tar with a --owner= --group= --device-table= (where the owner/group get applied by default to all files, and the device table can apply perms, owner, group, special(chardev,blockdev,etc) to lists of files). |
| 16:07.11 | kergoth | exactly |
| 16:07.15 | kergoth | thatd fit our needs perfectly |
| 16:07.25 | kergoth | hey mickeyl |
| 16:07.56 | mickeyl | hi kergoth |
| 16:09.14 | mickeyl | <kergoth> [17:41] mickeyl|away: its currently being written? by whom? :) |
| 16:09.28 | kergoth | mickeyl: opie VT handling |
| 16:09.30 | mickeyl | I'm taking care about that, because it interests me :D |
| 16:09.33 | kergoth | cool |
| 16:10.33 | mickeyl | btw.,I have a bug in the sysrq on c7x0 - probably my fault. HOME acts as sysrq - instead of ALT-HOME. |
| 16:10.49 | mickeyl | any idea? i though sysrq.c checks for ALT being held? |
| 16:10.54 | mickeyl | s/though/thought |
| 16:10.59 | kergoth | dunno, but home acts as sysrq on collie |
| 16:11.03 | kergoth | heh |
| 16:11.10 | mickeyl | oh - so that's normal. |
| 16:11.19 | mickeyl | but then home isn't working as a normal key anymore |
| 16:11.22 | kergoth | which doesnt matter much, how often does someone hold home+anotherkey? |
| 16:11.27 | kergoth | it does in collie |
| 16:11.30 | mickeyl | it does? |
| 16:11.31 | kergoth | yep |
| 16:11.42 | kergoth | its sysrq, but still sends its normal keycode up |
| 16:11.44 | kergoth | so it seems |
| 16:12.04 | mickeyl | ok, this must be missing in corgi-keyb.c. will see. |
| 16:13.09 | kergoth | heh, his patch was more complex than i would have figured |
| 16:13.13 | kergoth | look at the collie sysrq patch |
| 16:13.38 | mickeyl | have a changeset ID handy? |
| 16:13.59 | kergoth | uh |
| 16:14.02 | kergoth | packages/kernel/sysrq.patch |
| 16:14.09 | mickeyl | *cough* |
| 16:14.10 | mickeyl | ;) |
| 16:14.29 | mickeyl | i thought it was in linux-cl and searched and searched... |
| 16:14.35 | kergoth | yeah, he used a different sysrq xlate table |
| 16:14.43 | kergoth | i never pulled it in, cause it looked unclean |
| 16:14.47 | kergoth | touches keyboard.c directly |
| 16:14.53 | kergoth | hackish at best |
| 16:15.06 | kergoth | so "someone" needs to figure out how that works and find a clean solution |
| 16:15.13 | mickeyl | hehe, yeah. |
| 16:15.20 | treke | any volunteers |
| 16:15.26 | mickeyl | but then again, i can live with home being unfunctional while we're in debugging phase |
| 16:15.32 | treke | * #openembedded takes one step back from kergoth |
| 16:15.50 | kergoth | oh yeah, with all my free time |
| 16:17.41 | *** join/#openembedded bipolar (~bflong@dhcp-103-630-1.msns.sm.ptd.net) |
| 16:18.15 | kergoth | hey bipolar |
| 16:18.21 | bipolar | kergoth: hi |
| 16:18.56 | bipolar | I tried building opie again last night. There are still build errors. I hope they get fixed soon. |
| 16:20.27 | mickeyl | bipolar: what branch and what errors? i must not tell you to give more context, must i? ;-D |
| 16:21.21 | bipolar | mickeyl: I can get them for you. I was not expecting them to be needed. I think this is a known problem. hold on... |
| 16:22.55 | bipolar | BRANCH_1_0 |
| 16:23.01 | bipolar | /usr/local/arm/3.3/lib/gcc-lib/arm-linux/3.3.1/../../../../arm-linux/bin/ld: libaqpkg.so.1.0.0: undefined versioned symbol name std::time_put_w@@GLIBCPP_3.2 |
| 16:23.01 | bipolar | /usr/local/arm/3.3/lib/gcc-lib/arm-linux/3.3.1/../../../../arm-linux/bin/ld: failed to set dynamic section sizes: Bad value |
| 16:23.01 | bipolar | collect2: ld returned 1 exit status |
| 16:23.02 | bipolar | make[3]: *** [/home/bflong/build/buildroot-gcc3/build/opie/plugins/application/libaqpkg.so.1.0.0] Error 1 |
| 16:23.18 | kergoth | bipolar: toolchain problem |
| 16:23.21 | bipolar | I get errors like this unless I turn back the CVSDAE |
| 16:23.31 | kergoth | bipolar: works with 3.3-oz1 instead of 3.3.1-oz1, according to treke |
| 16:23.44 | mickeyl | and works here with 3.3.1-oz1 |
| 16:23.53 | mickeyl | so this is definitly an environment issue, not an opie issue |
| 16:24.02 | bipolar | kergoth: It can't be. I've tried 3.2 3.3 and 3.3.1 |
| 16:24.18 | kergoth | mickeyl: indeed. |
| 16:24.18 | bipolar | And if I turn back the cvsdate, the problem goes away |
| 16:24.24 | kergoth | yes, i heard you the first time. |
| 16:24.31 | kergoth | that doesnt imply that it isnt a toolchain problem |
| 16:24.50 | kergoth | just that the code that _exhibits_ the problem goes away |
| 16:25.25 | bipolar | What could possiblly be wrong with the toolchain? I've gone through this so many times. I've tried so many versions. |
| 16:25.51 | kergoth | tried moving your .ccache out of the way? |
| 16:26.00 | kergoth | heheh |
| 16:26.09 | bipolar | kergoth: does mrproper do that? |
| 16:26.28 | mickeyl | no |
| 16:26.33 | mickeyl | rm -rf $CCACHE_DIR |
| 16:26.47 | mickeyl | or cccache -someCommandLineSwitch |
| 16:27.10 | bipolar | what is ccache, anyway? |
| 16:27.16 | mickeyl | ibot: ccache |
| 16:27.16 | | rumour has it, ccache is a compiler cache that speeds up re-compilations of same sourcecodes where only some have different sourcefiles, see http://ccache.samba.org/ and also look for distcc |
| 16:27.21 | kergoth | you dont use ccache? |
| 16:27.23 | kergoth | dear god. |
| 16:27.26 | kergoth | you dont know what you've been missing |
| 16:27.28 | bipolar | oh.... I'm not using that. |
| 16:27.33 | kergoth | haha |
| 16:27.36 | kergoth | you build feeds |
| 16:27.38 | kergoth | and you dont use it? |
| 16:27.40 | kergoth | ~lart bipolar |
| 16:27.44 | kergoth | ~lart bipolar again |
| 16:27.48 | *** join/#openembedded TimRiker (~timr@proxyle01.ext.ti.com) |
| 16:28.02 | bipolar | at least I don't have to worry about the ccache fudging my builds. :P |
| 16:28.07 | kergoth | heh |
| 16:28.11 | kergoth | its a _rare_ occurance |
| 16:28.15 | kergoth | you have no idea how rare. |
| 16:28.26 | bipolar | Ok... i'll start using it. |
| 16:28.36 | bipolar | But first I need to get past this problem. |
| 16:29.29 | kergoth | grr |
| 16:29.32 | *** join/#openembedded treke (~ggilbert@12.107.12.130) |
| 16:29.37 | bipolar | mickeyl: you said that 3.3.1 worked for you? |
| 16:29.37 | kergoth | the ipaq cluster doesnt have the ssl dev stuff |
| 16:29.44 | kergoth | so i cant complete the ssh configure run to get the test results |
| 16:29.46 | kergoth | how annoying is that |
| 16:29.54 | bipolar | kergoth: the ipaq cluster sucks |
| 16:30.03 | kergoth | compared to what? |
| 16:30.11 | kergoth | i hate native builds, but I _need_ a native arm machine to do this on |
| 16:30.22 | bipolar | kergoth: making punch cards by hand. :) |
| 16:30.48 | kergoth | treke: what db backend does rt use? |
| 16:31.01 | mickeyl | bipolar: yes. i was building a combined 2.95.3/3.3.1pre image without problems |
| 16:31.10 | bipolar | mickeyl: whats your host OS? |
| 16:31.24 | mickeyl | bipolar: mdk91 |
| 16:31.30 | mickeyl | + a lot of recent cooker packages |
| 16:31.44 | bipolar | mickeyl: what do you have the host os set to in buildroot? |
| 16:31.49 | mickeyl | bipolar: other |
| 16:31.52 | kergoth | bipolar: irrelevent |
| 16:31.52 | bipolar | ok. |
| 16:32.00 | kergoth | none of the host specific hacks are even useful anymore |
| 16:32.00 | bipolar | kergoth: that setting does nothing? |
| 16:32.06 | kergoth | now that we build autoconf & automake in host build tools |
| 16:32.18 | kergoth | we used that to call the correct auto* versions in the build machine |
| 16:32.56 | bipolar | What user should own the toolchain? Should any of it be writeable by the user that the build uses? |
| 16:33.08 | kergoth | shouldnt matter |
| 16:33.13 | kergoth | it only writes there if you're building a chain |
| 16:33.18 | bipolar | ok, just checking. |
| 16:35.12 | bipolar | This makes little sence |
| 16:35.19 | bipolar | sense, even |
| 16:35.30 | *** join/#openembedded pb_ (~pb@2002:5160:45ef:0:2e0:7dff:fe74:8b87) |
| 16:35.51 | bipolar | I got the same error when I built from my debain stable server, and from this debian sid box. |
| 16:36.30 | mickeyl | fwiw, i _never_ seen such an error on my box - and i'm juggling around with various toolchains also |
| 16:36.41 | mickeyl | must be a debian speciality |
| 16:39.47 | kergoth | I never ran into problems when i was building on my debian unstable box, but that was a while ago |
| 16:41.44 | bipolar | I wonder if any of this quicklaunch stuff is fudging it up |
| 16:42.16 | kergoth | highly doubtful. |
| 16:42.18 | mickeyl | unlikely |
| 16:42.39 | bipolar | running a bk pull and then I'll rebuild opie |
| 16:42.47 | mickeyl | maybe do a bk clone |
| 16:42.51 | mickeyl | to really clean everything |
| 16:43.05 | bipolar | if opie does not build, I will. |
| 16:43.15 | kergoth | ibot: etticate |
| 16:43.16 | | kergoth: i don't know |
| 16:43.19 | kergoth | ibot: spell etticate |
| 16:43.21 | | possible spellings for etticate: educate eradicate petticoat edict evocate metricate Attica dedicate medicate educator estimate abdicate indicate titivate Attica's educated educates litigate mitigate artiste attacked petticoated ticket Ettie articulate equate rusticate attics evacuate agitate titrate |
| 16:43.26 | Harlekin | mickeyl: do you like quicklaunch by now? |
| 16:43.34 | kergoth | damnit, its not there |
| 16:43.37 | kergoth | how teh fuck do i spell that |
| 16:43.37 | mickeyl | Harlekin: yes - it is great. |
| 16:43.40 | Harlekin | .-) |
| 16:43.53 | pb_ | kergoth: ettiquette? not sure what word you're after. |
| 16:43.54 | Harlekin | mickeyl: especially it also shows some coding errors |
| 16:43.58 | mickeyl | Harlekin: and... i like the new bouncing icon hehe. me and a dozen snapshot users |
| 16:44.11 | mickeyl | Harlekin: so please let it in as an option |
| 16:44.19 | kergoth | pb_: ah, thats what i needed |
| 16:44.21 | kergoth | pb_: hehe. thanks |
| 16:44.24 | fdask | the bouncing icon? like when u click a prog and wait for it 2 load? |
| 16:44.35 | mickeyl | yeah |
| 16:44.35 | Harlekin | mickeyl: ok, maybe we leave that for now, but the bouncing as "turn off option" |
| 16:44.41 | fdask | ya, saw that. its cute |
| 16:44.44 | mickeyl | k |
| 16:44.50 | Harlekin | mickeyl: so as default only hilight |
| 16:44.57 | Harlekin | bouncing eats cpu cycles |
| 16:44.59 | Harlekin | not much |
| 16:45.00 | Harlekin | but still |
| 16:45.04 | kergoth | anyone have the url to the email/ML ettiquette page of david woodhouse's? |
| 16:45.08 | Harlekin | now if only gcc would be a better c++ compiler |
| 16:45.22 | mickeyl | how should it |
| 16:45.31 | mickeyl | gcc is just _not_ a c++ compiler ;) |
| 16:45.57 | mickeyl | no really, i like everything >= 3.2 |
| 16:46.49 | Harlekin | gnu Cxx compiler |
| 16:47.00 | kergoth | <PROTECTED> |
| 16:47.05 | kergoth | thats what i was looking for |
| 16:47.43 | mickeyl | kergoth: ah that's cool. didn't know that. i only knew esr's questions article which is also great |
| 16:48.04 | kergoth | mickeyl: this one is handy for things like zecke's recent email about not cc'ing him directly.. |
| 16:48.11 | kergoth | just send them there |
| 16:48.12 | kergoth | heh |
| 16:48.43 | cyn_ | (email ettiquette) I prefer putting my reply above the original if the original is short and just 1 block - if it's a long message i'll be breaking the original up and answering/refering to each piece as i go |
| 16:49.17 | cyn_ | because if anyone's following the thread they don't want to keep rereading, if anyones new - they can refer to the previous stuff - whereas if i'm addressing specific points it just makes sense to quote them before answering |
| 16:49.23 | mickeyl | umm... |
| 16:49.25 | mickeyl | let the flames begin |
| 16:49.28 | kergoth | most people have page down |
| 16:49.41 | kergoth | if as you say, the section is small, they dont have to go far in the first place |
| 16:49.44 | kergoth | heh |
| 16:49.55 | kergoth | mickeyl: hehe, indeed |
| 16:50.03 | cyn_ | I read email in pine and prefer new stuff at the top personally |
| 16:50.08 | bipolar | is anyone useing distcc to compile buildroot? :) |
| 16:50.11 | kergoth | sometimes i reply at the top out of laziness, but.. |
| 16:50.12 | mickeyl | bipolar: i used to |
| 16:50.20 | kergoth | bipolar: i have on many occasions, but buidlroot doesnt like -j very much |
| 16:50.28 | bipolar | ahh |
| 16:50.30 | kergoth | OE works great with it |
| 16:50.34 | kergoth | since make is only used at the package level |
| 16:50.41 | kergoth | only oen package is built at a time, but that pakcage uses it |
| 16:50.42 | cyn_ | kergoth: my habit may just be a symptom of people lazily quoting entire threads back 10 messages tho |
| 16:50.54 | cyn_ | all too often people do not selectively quote |
| 16:50.56 | kergoth | cyn_: indeed. i always, always snip the hell out of it |
| 16:50.57 | kergoth | right |
| 16:51.00 | kergoth | i hate that. |
| 16:51.08 | kergoth | end up with like 20 pages of crap |
| 16:51.11 | cyn_ | yea |
| 16:51.21 | bipolar | When I make the switch to OE (after I get one more successful build with buildroot) I'll start using distcc too. :) |
| 16:51.27 | mickeyl | boils down to most people are lazy and think their time is more valuable than ours :( |
| 16:51.28 | cyn_ | so 5 messages replying to the same thing are like 200kb each and you need to page down 10 times per message |
| 16:51.36 | kergoth | bipolar: start converting packages to the new system |
| 16:51.40 | kergoth | bipolar: its really easy. |
| 16:51.45 | cyn_ | mickeyl: well, my point was that I prefer to recieve it that way too |
| 16:51.52 | mickeyl | bipolar: its cool, but you have to be _very_ careful to synchronize your toolchains |
| 16:52.05 | kergoth | mickeyl: yeah. which is why at a certain point you refuse to assist without them reading and abiding by david's rules |
| 16:52.07 | mickeyl | cyn_: i was referring to not willing to cut down quotes |
| 16:52.11 | kergoth | mickeyl: which, incidentally, is what he does |
| 16:52.11 | kergoth | hehe |
| 16:52.12 | bipolar | kergoth: That I will. I've got to get one more build out. unstable has langusihed too long. |
| 16:52.14 | mickeyl | hehe |
| 16:52.15 | mickeyl | true |
| 16:52.15 | cyn_ | oh yeah they sux |
| 16:52.34 | bipolar | mickeyl: how about sharing the toolchain via nfs between boxes? |
| 16:52.47 | kergoth | mickeyl: thats what USEFULLPATH was for |
| 16:52.48 | mickeyl | bipolar: that sucks all you would win |
| 16:52.54 | mickeyl | bipolar: time-wise |
| 16:52.54 | kergoth | mickeyl: then you dont need PATH pointing there on the remote machine |
| 16:53.00 | kergoth | mickeyl: it uses the full path to the binaries |
| 16:53.01 | mickeyl | bipolar: or else you'd need a very fast machine |
| 16:53.02 | kergoth | :) |
| 16:53.05 | cyn_ | my favorite is when someone asks you something, you refer them to a very specific page that answers their question, and they reply back "yeah I read that but didn't understand it so asked here" or something |
| 16:53.07 | mickeyl | kergoth: ah i see |
| 16:53.12 | kergoth | echo "USEFULLPATH=yes" >> Vars.local |
| 16:53.21 | cyn_ | generally its like "so why didn't you STFW and learn to comprehend it" |
| 16:53.32 | bipolar | mickeyl: I didn't understand "that sucks all (bad) you would win (good)" ??? |
| 16:53.33 | kergoth | cyn_: yeah, exactly. i like it short and sweet. "Read it again." |
| 16:53.46 | cyn_ | so smb has too much overhead for the using it for buildroot dealie |
| 16:53.48 | mickeyl | bipolar: ok i'll explain |
| 16:53.51 | bipolar | heh |
| 16:53.53 | cyn_ | get to find a nfs server for win |
| 16:54.11 | kergoth | bipolar: a certain amount of overhead exists due to the network transactions |
| 16:54.15 | mickeyl | bipolar: that sucks time and bandwidth (for the compiler to be loaded from net) |
| 16:54.22 | bipolar | oh... ok |
| 16:54.24 | kergoth | yep. that can kill any benefit you would have gained |
| 16:54.26 | mickeyl | bipolar: which is what you wanted to win by distributing theload |
| 16:54.30 | kergoth | if the machine's speed isnt substantial |
| 16:54.30 | bipolar | gotcha |
| 16:54.50 | kergoth | course if you have a gigibit ethernet, you're probably okay, minus latency |
| 16:55.00 | mickeyl | ya, probably |
| 16:55.16 | kergoth | at least sending the code and objects wouldnt take long |
| 16:56.18 | bipolar | 2 dual athlon 1800+ servers with 1GB of ram, + a 700Mhz Duron + 1GHZ athlon. Too bad it doesn't run on windows, I'd have a lot more. ;) |
| 16:56.18 | pb_ | mickeyl: if you were compiling more than one file, the compiler would just sit in page cache on the local host. you wouldn't need to send it over the wire every time. |
| 16:56.49 | bipolar | hmm... I could make a knoppix distcc disk to run in the windows pc's after everyone left. :D |
| 16:57.07 | kergoth | bipolar: hah |
| 16:57.23 | bipolar | SHIT!!! |
| 16:57.26 | *** join/#openembedded icefox (~ben@xltpat1.symbol.com) |
| 16:57.28 | bipolar | I got the exact same error |
| 16:57.35 | icefox | Why doesn't the opie make support -jX? |
| 16:57.51 | bipolar | damn fucking thing... Why doesn't it just work |
| 16:57.57 | kergoth | icefox: not sure. |
| 16:58.05 | kergoth | icefox: probably some packages missing dependencies in config.in |
| 16:58.12 | icefox | bummer |
| 16:58.16 | kergoth | the deps have to be _perfect_ |
| 16:58.21 | kergoth | or itll build something before something it needs |
| 16:58.21 | kergoth | heh |
| 16:59.46 | icefox | opie/include/qpe/qpedecoration_qws.h:25:38: qwsdefaultdecoration_qws.h: No such file or directory |
| 17:00.36 | icefox | What is the difference between libopie and libopie2? |
| 17:00.44 | mickeyl | libopie2 = libopie1 extensions |
| 17:00.49 | mickeyl | totally different purpose |
| 17:00.57 | icefox | why named the same then? |
| 17:01.04 | mickeyl | was thought to replace it eventually, but plans changed |
| 17:01.18 | icefox | yah |
| 17:01.43 | kergoth | hey look, a perfect example of bad list etiquette |
| 17:01.43 | mickeyl | bbl - need to do some house work :( |
| 17:01.46 | kergoth | stupid autorepliers |
| 17:04.19 | bipolar | ok, I'm starting from scratch. I'll use cross-2.95.4-oz-1.tar.bz2 and cross-3.3.1pre-oz1.tar.bz2 since they are the newest. Is there anything wrong with either one of those toolchains? |
| 17:04.43 | cyn_ | haha i hate peoples vacation autoreplies |
| 17:05.21 | icefox | man the branch doesn't even compile |
| 17:05.40 | *** join/#openembedded bisho (~bisho@136.Red-213-97-191.pooles.rima-tde.net) |
| 17:06.10 | icefox | How do you expect to get new developers when stuff doesn't even compile? |
| 17:06.21 | mickeyl|bbl | bipolar: i had all sorts of bad experiences with 2.95.4 |
| 17:06.28 | cyn_ | heh |
| 17:06.44 | mickeyl|bbl | bipolar: 2.95.3-oz1.2 seemed to better for me (for collie) |
| 17:07.17 | mickeyl|bbl | BRANCH_1_0 compiles fine here. |
| 17:07.22 | bipolar | mickeyl|bbl: ok. I'll use that. |
| 17:08.16 | bipolar | icefox: what is not compiling? |
| 17:08.42 | kergoth | icefox: how about asking in #opie, instead of wasting our fucking time? |
| 17:08.44 | icefox | head or branch? |
| 17:08.56 | icefox | good point |
| 17:09.12 | kergoth | aha |
| 17:09.14 | kergoth | stupid ssh |
| 17:09.20 | *** part/#openembedded icefox (~ben@xltpat1.symbol.com) |
| 17:09.21 | kergoth | it doesnt have config.cache items for all the tests |
| 17:10.14 | kergoth | i'll do the crosscompilation properly this time, and submit the patch upstream |
| 17:17.24 | kergoth | hmm |
| 17:17.35 | kergoth | it needs to check the ssl library version, then the header version, and compare |
| 17:17.40 | kergoth | wonder if i should make those config.cache items |
| 17:17.53 | kergoth | or if we should make it only obey the header version if crosscompiling |
| 17:17.55 | kergoth | opinions? |
| 17:18.19 | gb2 | assuming header matches library is probably ok |
| 17:18.44 | kergoth | problem with the cache is we end up assuming that the header & lib havent changed since the previous execution |
| 17:18.48 | kergoth | and thats probabably a Bad Thing |
| 17:19.11 | gb2 | that's what i was thinking |
| 17:19.27 | kergoth | could make it config overridable. --with-ssl-version=0.9.7 or whatever |
| 17:19.39 | kergoth | either instead of or in addition to obeying the header |
| 17:19.51 | gb2 | does OE play config.cache tricks when cross building things? |
| 17:19.56 | kergoth | yep |
| 17:20.04 | kergoth | it uses site/${ARCH}-${OS} |
| 17:20.12 | gb2 | ah |
| 17:20.15 | kergoth | in $OEDIR |
| 17:20.16 | gb2 | wondered what that file was |
| 17:20.19 | kergoth | heheh |
| 17:20.22 | gb2 | i never actually opened it though :P |
| 17:20.35 | gb2 | probably would have guessed if i'd seen its contents |
| 17:20.36 | kergoth | we use the CONFIG_SITE env var, which makes it so you dont have to pass -C to every configure call to get it to obey the results |
| 17:21.09 | gb2 | some of the newer autoconfs make the cache disabled in the default configure invocation |
| 17:21.32 | kergoth | yep |
| 17:21.41 | kergoth | and some versions take slightly differing syntax to enable it |
| 17:21.45 | kergoth | much easier this way |
| 17:22.01 | gb2 | probably because of some of that lib version caching confused people |
| 17:22.07 | gb2 | or similar |
| 17:22.08 | kergoth | hehe |
| 17:22.09 | kergoth | probably |
| 17:22.42 | gb2 | "you must have atleast <-this-> much clue to use the cache" |
| 17:23.13 | kergoth | hehe |
| 17:33.42 | kergoth | hmm |
| 17:34.01 | kergoth | i really dont want to play sed games just to get the ssl version from the header |
| 17:34.14 | kergoth | maybe i'll just add the new --with-ssl-version opt for now |
| 17:34.30 | Harlekin | kergoth: is QMAKESPEC and PLATFORM set somwhere on the way in opie build system? |
| 17:34.36 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-111-101.ppp.tiscali.fr) |
| 17:34.43 | kergoth | Harlekin: from buildroot? |
| 17:34.52 | kergoth | or..? |
| 17:34.53 | Harlekin | kergoth: standalone |
| 17:35.01 | kergoth | read Vars.make |
| 17:35.07 | Harlekin | since it does not work .-) |
| 17:35.16 | kergoth | its always worked fine in the past. |
| 17:35.19 | Harlekin | at least for icefox |
| 17:35.26 | Harlekin | got neither set |
| 17:35.29 | kergoth | PLATFORM wasnt added by me. |
| 17:35.35 | Harlekin | ljp? |
| 17:35.37 | kergoth | somebody fucked with things. |
| 17:35.43 | kergoth | i dont havfe time to look at it right now |
| 17:35.46 | kergoth | when i get home i will |
| 17:35.47 | Harlekin | k |
| 17:41.45 | gb2 | kergoth: grep -i '^.define[[:space:]]*SHLIB_VERSION_NUMBER' /usr/include/openssl/opensslv.h |sed -e 's/.*"\(.*\)"/\1/' |
| 17:41.46 | gb2 | ? |
| 17:42.38 | gb2 | my regexps are probably a little overkillish too |
| 17:44.10 | kergoth | first, you used /usr/include |
| 17:44.10 | gb2 | probably doesn't quite give you what you want though i guess. |
| 17:44.12 | kergoth | ~lart gb2 |
| 17:44.20 | kergoth | dont touch x86 headers when crosscompiling! |
| 17:44.25 | gb2 | um. that was a demonstration. |
| 17:44.36 | kergoth | yeah i know, just couldnt help myself |
| 17:44.39 | kergoth | heheh |
| 17:44.44 | gb2 | i'd assume you'd use ${includedir} or whichever is appropriate given context |
| 17:44.57 | kergoth | includedir isnt either |
| 17:45.04 | kergoth | we need it to check for the header in the includepath of the compiler we're using |
| 17:45.08 | kergoth | wonder if you can obtain th at info |
| 17:45.31 | gb2 | ${openssl_includedir}/opensslv.h |
| 17:45.42 | gb2 | :P |
| 17:46.11 | gb2 | not sure if it's OPENSSL_VERSION_NUMBER that's wanted though.. rather than the shlib version. |
| 17:46.21 | gb2 | openssl really has crappy versioning unfortunately |
| 17:46.28 | kergoth | i just took care of it the easy way |
| 17:46.32 | kergoth | it doenst AC_DEFINE the version |
| 17:46.36 | kergoth | so it doesnt affect _anything_ |
| 17:46.39 | *** join/#openembedded Kerberos (~Kerberos@200.76.152.5) |
| 17:46.43 | kergoth | its only a sanity check to make sure the lib matches the headers |
| 17:46.51 | kergoth | so i just bypassed them during a crosscopmile |
| 17:47.41 | gb2 | ah |
| 17:47.43 | gb2 | ok |
| 17:48.09 | kergoth | easy fix :) |
| 17:48.52 | gb2 | have you used any of the other ssh implementations? |
| 17:49.00 | gb2 | lsh and there's atleast one other |
| 17:49.07 | kergoth | dropbear |
| 17:49.10 | kergoth | but thats server only, not client |
| 17:49.42 | kergoth | so no scp binary and stuff |
| 17:49.46 | gb2 | does it work reasonably well? or were you just noting the other one |
| 17:49.49 | kergoth | i'd like to use it otherwise |
| 17:49.52 | Harlekin | btw, is the scp in oz compiled now without openssl depend? |
| 17:50.10 | kergoth | eh, we build scp from openssh, i'd have to assume it uses openssl |
| 17:50.11 | gb2 | hm. lack of scp is definately a lose yeah |
| 17:50.17 | kergoth | gb2: indeed |
| 17:50.25 | kergoth | maybe lsh's client side + dropbear's server |
| 17:50.26 | gb2 | kergoth: scp shouldn't need openssl |
| 17:50.27 | kergoth | someone should analyze these |
| 17:50.31 | Harlekin | kergoth: pb build it without openssl |
| 17:50.33 | Harlekin | compile option |
| 17:50.37 | kergoth | oh right |
| 17:50.41 | Harlekin | kergoth: so only ssh client itself needs openssl |
| 17:50.43 | kergoth | well i'll do that now |
| 17:50.53 | kergoth | cool |
| 17:51.01 | kergoth | so the scp package is relatively self contained |
| 17:51.08 | gb2 | a lot of projects link in unnecessary stuff like that |
| 17:51.11 | Harlekin | at least pb_ told me |
| 17:51.29 | kergoth | gb2: indeed |
| 17:51.36 | kergoth | gb2: the transition gives us a chance to clean things up |
| 17:51.46 | kergoth | gb2: analyze our build options, clean up cross patches for submission upstream, etc |
| 17:51.59 | Harlekin | so 1 mb more space on device |
| 17:51.59 | gb2 | *nod* |
| 17:52.28 | kergoth | 1mb more space? |
| 17:52.36 | Harlekin | without openssl |
| 17:52.42 | kergoth | ah yeah |
| 17:52.49 | kergoth | it would be nice to have a client though |
| 17:52.49 | pb_ | kergoth: a millibit may not sound like much, but they all add up |
| 17:53.05 | kergoth | heheh, indeed |
| 17:53.25 | Harlekin | kergoth: do you really thing the majority of zaurus users needs a ssh client on the device? |
| 17:53.32 | kergoth | Harlekin: probably not. |
| 17:53.36 | Harlekin | kergoth: who really wants it can still install it |
| 17:53.38 | kergoth | Harlekin: could easily leave it out of the default images |
| 17:53.38 | Harlekin | along with openssl |
| 17:53.39 | kergoth | yep |
| 17:56.03 | kergoth | pb_: i take it you patched the Makefile.in for openssh to remove the explicit hardcoded -lssl on the scp link line? |
| 17:56.11 | pb_ | yep |
| 17:56.15 | kergoth | k |
| 17:56.25 | kergoth | i hate anything hardcoded like that |
| 17:56.32 | kergoth | should be SCP_LIBS or something |
| 17:56.32 | kergoth | heh |
| 17:56.33 | pb_ | it doesn't actually use any functions from libssl. must have been added as a joke. |
| 17:56.35 | kergoth | anyway |
| 17:56.46 | kergoth | odd |
| 18:05.01 | *** join/#openembedded heatx (~John@63.146.78.10) |
| 18:09.34 | Twiun | re |
| 18:10.17 | *** join/#openembedded treke (~ggilbert@12.107.12.130) |
| 18:10.37 | *** join/#openembedded LV|work (~lordvan@62.218.218.165) |
| 18:18.32 | *** join/#openembedded CAMCAM (~camrein@AMontsouris-108-1-17-54.w80-15.abo.wanadoo.fr) |
| 18:18.52 | *** part/#openembedded CAMCAM (~camrein@AMontsouris-108-1-17-54.w80-15.abo.wanadoo.fr) |
| 18:25.08 | mewyn | <PROTECTED> |
| 18:25.34 | bipolar | mewyn: yeah... I saw that in a slashdot comment. ;) |
| 18:25.36 | bipolar | heh |
| 18:25.50 | mewyn | seems about right |
| 18:26.36 | *** part/#openembedded Kerberos (~Kerberos@200.76.152.5) |
| 18:27.31 | mewyn | i'm wondering if they did a character-by-character comparison |
| 18:32.39 | bipolar | I think that the SCO execs can afford the really potent crack. |
| 18:33.12 | bipolar | But sometimes they act like they're hitting the cheap shit. |
| 18:33.12 | mewyn | but not for long |
| 18:34.31 | mewyn | they have to use the cheap shit to make their money last |
| 18:34.51 | mewyn | in alll reality, i find that SCO is just a pawn for Canopy |
| 18:35.22 | *** join/#openembedded lsmith (~vandal@p213.54.25.128.tisdip.tiscali.de) |
| 18:43.17 | bipolar | *** 2.4 kernels no longer build correctly with old versions of binutils. |
| 18:43.17 | bipolar | *** Please upgrade your binutils to 2.9.5. |
| 18:43.26 | bipolar | Sigh |
| 18:43.34 | mickeyl | wtf.. |
| 18:43.40 | mickeyl | is this 2.95.3-oz1.2 ? |
| 18:43.42 | bipolar | I know thats not right. |
| 18:43.45 | bipolar | yes. |
| 18:43.50 | mickeyl | damn |
| 18:43.57 | mickeyl | and kernel 2.4.6 ? |
| 18:44.34 | bipolar | Yes |
| 18:44.54 | bipolar | I'm the most unlucky guy when it comes to this shit. |
| 18:45.07 | bipolar | This is on my debian stable server. |
| 18:45.28 | bipolar | I've done upteen builds on this box, and now this. |
| 18:45.32 | mickeyl | hmm... we've been through stuff like that... but i swear i can't remember how |
| 18:45.24 | mickeyl | anyone? |
| 18:45.46 | kergoth | i dont recommend that chain |
| 18:45.49 | kergoth | use 2.95.3 from erikm |
| 18:46.03 | bipolar | kergoth: where can I get it? |
| 18:46.17 | mickeyl | bipolar: oz.org/official/toolchain/old |
| 18:46.19 | kergoth | official/toolchain/old |
| 18:46.24 | kergoth | -2.95.3.tar.bz2 or whatever |
| 18:46.48 | bipolar | ~lart sf.net |
| 18:48.55 | bipolar | oz.org is down.. go figure. |
| 18:49.07 | bipolar | whats the file name of the toolchain? |
| 18:49.11 | bipolar | I might have it here |
| 18:49.19 | mickeyl | it is also on the armlinux ftp |
| 18:49.39 | mickeyl | cross-2.95.3.tar.bz2 |
| 18:50.05 | kergoth | pub/armlinux/toolchain |
| 18:50.41 | bipolar | I think I have it |
| 18:50.43 | bipolar | # md5sum cross-2.95.3.tar.bz2 |
| 18:50.43 | bipolar | c40277b25adb1bd2bcef0dbaabb65283 cross-2.95.3.tar.bz |
| 18:50.45 | bipolar | # md5sum cross-2.95.3.tar.bz2 |
| 18:50.46 | bipolar | c40277b25adb1bd2bcef0dbaabb65283 cross-2.95.3.tar.bz2 |
| 18:51.14 | bipolar | sorry about the double paste. I don't know why thats happening either. |
| 18:51.59 | mickeyl | bipolar: that's right md5 |
| 18:52.50 | bipolar | mickeyl: cool |
| 18:53.02 | bipolar | I've got more toolchains here then I can count. |
| 18:53.38 | prpplague | bipolar: i have the same problem with bootloaders |
| 18:54.02 | mickeyl | prpplague: wanna port us one? ;) |
| 18:54.26 | prpplague | mickeyl: for what board? |
| 18:54.56 | kergoth | prpplague: any zaurus. or simpad. |
| 18:54.57 | mickeyl | prpplague: you can choose. SA1110-SIMPAD, PXA255-SHEPHERD, SA1110-COLLIE |
| 18:55.14 | kergoth | blob, lab, or u-boot preferably |
| 18:55.25 | mickeyl | u-boot preferably? i though lab preferably |
| 18:55.27 | prpplague | kergoth: i thought the bootloader for the z series was in a custom eprom |
| 18:55.30 | kergoth | .. |
| 18:55.30 | mickeyl | s/though/thought |
| 18:55.35 | kergoth | you misread my statement. |
| 18:55.43 | mickeyl | ah right |
| 18:55.46 | kergoth | (this, that, thatotherthing) is preferred |
| 18:55.49 | kergoth | heh |
| 18:56.06 | mickeyl | prpplague: no, it is in a dedicated and normally write protected flash area |
| 18:56.17 | prpplague | mickeyl: ahh |
| 18:56.23 | mickeyl | (as defined by the mtd-mapping) |
| 18:56.24 | kergoth | and sharp flashes it in in their ospack iamges |
| 18:56.26 | kergoth | idiots |
| 18:56.28 | mickeyl | hah |
| 18:56.29 | mickeyl | yeah |
| 18:56.29 | kergoth | anyway |
| 18:56.34 | kergoth | for collie that is |
| 18:56.35 | kergoth | :) |
| 18:56.41 | kergoth | dunno about the c7x0 |
| 18:56.48 | kergoth | they could have locked mtd0 |
| 18:57.09 | prpplague | mickeyl: i'd have to get the jtag working and iirc sorphin tried for a long time to get that running without docs |
| 18:57.09 | kergoth | the proto units were nor, but i hear it may be eeprom in production |
| 18:57.16 | kergoth | prpplague: i have docs. |
| 18:57.19 | kergoth | prpplague: under nda of course, but.. |
| 18:57.22 | prpplague | kergoth: right |
| 18:57.26 | kergoth | we dont have software for it though |
| 18:57.34 | kergoth | guess we could alter one of hte open ones |
| 18:57.43 | mickeyl | openjtag is nice |
| 18:57.45 | prpplague | kergoth: probably get jflash running |
| 18:57.45 | kergoth | brb, meeting |
| 19:00.41 | bipolar | dodedodedo |
| 19:19.19 | bipolar | Hey... looks like it is possible to run distcc on winduhs via cygwin. |
| 19:19.33 | bipolar | That will be very, very cool |
| 19:19.42 | bipolar | It's build to suit though. |
| 19:20.23 | bipolar | I could use 10 1ghz pc's that just sit here and do nothing after hours to build OE ;) |
| 19:22.57 | mickeyl | sounds interesting |
| 19:23.07 | mickeyl | you'd have a problem with -j though |
| 19:23.14 | mickeyl | that still doesn't work properly |
| 19:23.22 | mickeyl | in our buildroot |
| 19:28.31 | killefiz | bipolar: why not insert a knoppix cd and use that for running distcc? |
| 19:30.45 | *** join/#openembedded TimRiker (~timr@proxyle02.ext.ti.com) |
| 19:31.39 | *** join/#openembedded imk|z (pda@68.170.56.122) |
| 19:35.46 | killefiz | yes ... my buildroot compiled ... no errors |
| 19:35.48 | *** join/#openembedded treke (~ggilbert@12.107.12.130) |
| 19:36.35 | treke | why? |
| 19:36.54 | treke | what possible reason could there be to be happy |
| 19:37.08 | killefiz | my buildroot compiled - after 4 days of fiddling |
| 19:38.22 | treke | hehehe |
| 19:38.37 | treke | "Ze language of engineering is not English, es ist BROKEN English!" |
| 19:39.04 | killefiz | ;) |
| 19:39.06 | kergoth | woo |
| 19:39.11 | kergoth | ssh 3.6.1p2 built in oe |
| 19:39.12 | kergoth | next! |
| 19:39.13 | kergoth | heh |
| 19:39.16 | bipolar | killefiz: I could do that. I'd have to build my own anyway, since I'm building arm-linux from x86-cygwin. |
| 19:39.19 | treke | kergoth: opie :) |
| 19:39.23 | kergoth | nah |
| 19:39.23 | bipolar | kergoth: very cool. |
| 19:39.24 | kergoth | hehe |
| 19:39.30 | kergoth | i think maybe.. say.. modutils would help |
| 19:39.42 | Harlekin | anybody make a recent c7xx BRANCH 1.0.1.1 feed? |
| 19:40.35 | treke | need to get opie to build first |
| 19:44.18 | bipolar | is libpng.org hosted on sf.net by chance? |
| 19:44.28 | treke | could be |
| 19:44.45 | bipolar | 'cause it's down |
| 19:44.55 | bipolar | ~lart sf.net |
| 19:45.56 | treke | it's up |
| 19:46.09 | Harlekin | treke: wheres the problerm |
| 19:46.17 | kergoth | heh |
| 19:46.18 | kergoth | hey pb_ |
| 19:46.23 | treke | Harlekin: My build environment apparently |
| 19:46.26 | kergoth | openssh 3.6.1p2 doesnt link against -lssl at all |
| 19:46.27 | kergoth | anywhere |
| 19:46.30 | kergoth | only -lcrypt |
| 19:46.36 | kergoth | is crypt required for scp? |
| 19:46.54 | kergoth | looks like they cleaned it up a bit in p2 |
| 19:51.30 | treke | bipolar: try now. sf looks like it's back |
| 19:51.37 | bipolar | treke: got it ;) |
| 19:51.58 | kergoth | hmm |
| 19:52.01 | bipolar | god... I am soooo sick |
| 19:52.05 | bipolar | I should not be at work |
| 19:54.47 | *** join/#openembedded SuKoSh| (~sukoshi@dyn-195-242-112-52.ppp.tiscali.fr) |
| 19:57.32 | gb2 | kergoth: scp calls ssh to setup the channel between machines |
| 19:57.54 | kergoth | yeah, i know |
| 19:58.02 | kergoth | i figure the answer is it isnt needed, but i figured i'd check |
| 19:58.21 | gb2 | afaik it shouldn't need anything but the c library |
| 19:59.48 | pb_ | kergoth: no, scp doesn't depend on libcrypto |
| 20:00.45 | kergoth | yep, just saw that |
| 20:00.47 | kergoth | k |
| 20:04.29 | bipolar | ok... buildroot has reached the kernel. So my binuils problem seems to be fixed. |
| 20:05.11 | kergoth | k, pushed the openssh 3.6.1p2 bits. adding modutils |
| 20:05.52 | Twiun | woops, looks like the oe wiki db is ill |
| 20:06.03 | bipolar | kergoth: as soon as I get this last build done, i'm going to help you move packages over to OE, I swear. :) |
| 20:06.15 | kergoth | yeah yeah, thats what they all say |
| 20:06.17 | kergoth | :) |
| 20:06.19 | bipolar | :) |
| 20:06.54 | bipolar | kergoth: are you ready for me to be asking you stupid OE questions like I did when I started with buildroot? ;p |
| 20:07.12 | kergoth | sure, if you're going to write them down somewhere |
| 20:07.15 | kergoth | we need docs :) |
| 20:09.34 | bipolar | kergoth: yeah... I'll do that. Will you be online til late tonight? |
| 20:09.42 | killefiz | :-( make image gives: |
| 20:09.44 | killefiz | Installing alsa-modules into root |
| 20:09.44 | killefiz | pkg_info_preinstall_check: updating arch priority for each package |
| 20:09.44 | killefiz | pkg_info_preinstall_check: update file owner list |
| 20:09.44 | killefiz | Cannot find package alsa-modules. |
| 20:09.47 | kergoth | bipolar: yep |
| 20:09.49 | killefiz | Check the spelling or perhaps run 'ipkg update' |
| 20:09.49 | killefiz | configuring unpacked packages |
| 20:10.10 | bipolar | kergoth: cool. I'm not going anywhere but home anyway since I'm sick |
| 20:10.13 | killefiz | but the ipk is there: ./output/ipks/alsa-modules_2.4.19-rmk6-pxa1-hh19-5_h3600.ipk |
| 20:10.24 | kergoth | is h3600 an 'arch' in output/ipkg.conf? |
| 20:10.44 | killefiz | yes it has arch h3600 100 |
| 20:10.58 | kergoth | dunno then |
| 20:11.31 | bipolar | killefiz: is in in the Packages file? |
| 20:11.58 | killefiz | bipolar: no output/ipks/Packages is empty |
| 20:12.04 | prh | bah libxine wouldnt build |
| 20:12.07 | bipolar | killefiz: thats a problem. :) |
| 20:12.30 | killefiz | any hint on how to solve it? |
| 20:12.36 | kergoth | it generates it |
| 20:12.37 | bipolar | make image_clean; make image |
| 20:12.40 | kergoth | did you not build ipkg-utils? |
| 20:13.11 | killefiz | ipkg-utils is built |
| 20:13.30 | *** join/#openembedded bisho (~bisho@136.Red-213-97-191.pooles.rima-tde.net) |
| 20:13.35 | prh | make[1]: Entering directory `/home/paul/Z/c/buildroot-oz/packages/qte' |
| 20:13.35 | prh | Makefile:38: *** missing separator. Stop. |
| 20:13.45 | killefiz | bipolar: same error |
| 20:13.52 | prh | I'm not a make expert - anyone know of a problem there? |
| 20:14.47 | bipolar | killefiz: not sure. |
| 20:15.52 | bipolar | killefiz: what does "make image_config" do? |
| 20:16.32 | kergoth | the same thing as make config |
| 20:16.33 | killefiz | bipolar: ask a lot of questions ;) |
| 20:16.37 | kergoth | only for package selection in the image, not build |
| 20:16.48 | kergoth | make image_{old,rand,menu,x,...}config |
| 20:16.56 | killefiz | i've used make image_menuconfig |
| 20:17.06 | kergoth | yeah, thats the best miho |
| 20:17.07 | kergoth | er |
| 20:17.07 | kergoth | imho |
| 20:18.27 | bipolar | yeah... image_menuconfig... thats what I meant. :P |
| 20:18.57 | *** join/#openembedded leffe (~leffe@212.75.86.130) |
| 20:19.14 | killefiz | hmm - ../staging/build/bin/ipkg-make-index is empty |
| 20:19.19 | kergoth | heh |
| 20:19.22 | kergoth | thatd be a problem |
| 20:19.22 | killefiz | that could be the reason |
| 20:19.26 | kergoth | clean and rebuild ipkg-utils |
| 20:20.48 | *** join/#openembedded LV|off (~lordvan@62.218.218.165) |
| 20:21.19 | *** join/#openembedded BigBuG (~boeglin@crous-153-040.u-strasbg.fr) |
| 20:22.26 | killefiz | i can use ipkg-make-index.py to create a Packages file manually - is that sufficient |
| 20:22.44 | kergoth | .. |
| 20:22.45 | kergoth | uh |
| 20:22.48 | kergoth | thats what gets used |
| 20:22.51 | kergoth | read output/Makefile |
| 20:23.50 | killefiz | kergoth: no - there is only one occurence of ipkg-make-index in output/Makefile and it doesn't have a .py suffix |
| 20:24.49 | killefiz | seems to work now - there is an unsatisfied dep now but i think i can handle that myself ;) |
| 20:25.27 | kergoth | hmm odd |
| 20:26.47 | prh | hmm do I want libcrypt? |
| 20:33.39 | kergoth | modutils is in |
| 20:34.07 | kergoth | mtd next, then autoconf, automake, then tslib |
| 20:34.21 | kergoth | <PROTECTED> |
| 20:34.24 | kergoth | <PROTECTED> |
| 20:34.27 | kergoth | 15:17 < bipolar> killefiz: what does "make image_config" do? |
| 20:34.29 | kergoth | 15:18 <@kergoth> the same thing as make config |
| 20:34.32 | kergoth | 15:18 < killefiz> bipolar: ask a lot of questions ;) |
| 20:34.34 | kergoth | 15:18 <@kergoth> only for package selection in the image, not build |
| 20:34.35 | Twiun | very amusing page... _not_ tasteful or boss friendly: http://www.artofresistance.org/bush_mosaic/ |
| 20:34.37 | kergoth | 15:18 <@kergoth> make image_{old,rand,menu,x,...}config |
| 20:34.39 | kergoth | 15:18 < killefiz> i've used make image_menuconfig |
| 20:34.42 | kergoth | 15:19 <@kergoth> yeah, thats the best miho |
| 20:34.45 | kergoth | 15:19 <@kergoth> er |
| 20:34.47 | kergoth | 15:19 <@kergoth> imho |
| 20:34.50 | kergoth | 15:20 < bipolar> yeah... image_menuconfig... thats what I meant. :P |
| 20:34.52 | kergoth | 15:20 -!- leffe [~leffe@212.75.86.130] has joined #OPENEMBEDDED |
| 20:34.55 | kergoth | 15:21 < killefiz> hmm - ../staging/build/bin/ipkg-make-index is empty |
| 20:34.57 | kergoth | 15:21 <@kergoth> heh |
| 20:35.00 | kergoth | 15:21 <@kergoth> thatd be a problem |
| 20:35.03 | kergoth | 15:21 < killefiz> that could be the reason |
| 20:35.05 | kergoth | 15:21 <@kergoth> clean and rebuild ipkg-utils |
| 20:35.07 | kergoth | 15:22 -!- LV|off [~lordvan@62.218.218.165] has quit [Excess Flood] |
| 20:35.10 | kergoth | 15:22 -!- LV|off [~lordvan@62.218.218.165] has joined #openembedded |
| 20:35.13 | kergoth | 15:23 -!- BigBuG [~boeglin@crous-153-040.u-strasbg.fr] has joined #openembedded |
| 20:35.16 | kergoth | 15:24 < killefiz> i can use ipkg-make-index.py to create a Packages file |
| 20:35.18 | kergoth | <PROTECTED> |
| 20:35.20 | kergoth | 15:24 <@kergoth> .. |
| 20:35.23 | kergoth | 15:24 <@kergoth> uh |
| 20:35.26 | kergoth | 15:24 <@kergoth> thats what gets used |
| 20:35.28 | kergoth | 15:24 <@kergoth> read output/Makefile |
| 20:35.31 | kergoth | 15:25 < killefiz> kergoth: no - there is only one occurence of ipkg-make-index |
| 20:35.34 | kergoth | <PROTECTED> |
| 20:35.36 | kergoth | ack |
| 20:35.42 | kergoth | damn touchpad |
| 20:36.01 | bipolar | lol |
| 20:36.30 | Twiun | that's why they should be renamed strokepads |
| 20:38.20 | *** join/#openembedded andersee (~andersee@codepoet.org) |
| 20:38.25 | kergoth | hey andersee |
| 20:38.38 | andersee | kergoth: morning |
| 20:38.45 | kergoth | andersee: how goes it? |
| 20:38.55 | andersee | kergoth: ok |
| 20:39.08 | andersee | kergoth: just woke up |
| 20:45.00 | bipolar | wooohoo... finaly started building opie.... |
| 20:56.07 | kergoth | mtd & mtd-buildarch down |
| 20:57.36 | kergoth | ~/coding/projects/user/oe/packages$ ls -l tmp/deploy/tar/|wc -l |
| 20:57.37 | kergoth | <PROTECTED> |
| 20:57.47 | kergoth | 25 down, 2 hundred & something to go |
| 21:08.51 | ljp | <PROTECTED> |
| 21:09.00 | ljp | hmm |
| 21:09.02 | kergoth | ljp: start adding qtopia, damnit. |
| 21:09.23 | ljp | ya, I suppose I should start looking at that |
| 21:09.40 | *** join/#openembedded W8TVI|z (~zic@3639244824.mi.dial.hexcom.net) |
| 21:10.11 | cyn_ | > anyone know whether zaurus has an mmu? |
| 21:10.11 | cyn_ | yes. |
| 21:10.11 | cyn_ | .oO "So, now that I know you know - does it?" |
| 21:10.26 | cyn_ | I'm gunna laugh my ass off if he replies back with that |
| 21:10.32 | Harlekin | kergoth: cool, so now really moving to OE? |
| 21:10.39 | bipolar | :) |
| 21:10.41 | kergoth | Harlekin: yep |
| 21:10.55 | kergoth | cyn_: hehe, i was thinking about that |
| 21:11.03 | Harlekin | kergoth: so also want to move opie to bk? |
| 21:11.22 | kergoth | Harlekin: opie needs to move to bk asap so that we dont lose metadata when we restructure the repos |
| 21:11.30 | kergoth | Harlekin: but i think we should merge the branch into head frist |
| 21:11.32 | kergoth | first rather |
| 21:11.41 | Harlekin | k with me as long as your help docu is done by then .-) |
| 21:11.46 | kergoth | hehe |
| 21:11.58 | Harlekin | let 1.0.2 roll out before that |
| 21:12.07 | cyn_ | mmmm opie |
| 21:12.40 | bipolar | Has anyone seen the msg that xine gives about arm-unknown-linux when it finishes it's configure? |
| 21:12.46 | kergoth | yeah |
| 21:13.08 | bipolar | safe to ignore, I suspect? |
| 21:14.08 | *** join/#openembedded W8TVI (~w8tvi@3639244824.mi.dial.hexcom.net) |
| 21:14.24 | Harlekin | bipolar: yes |
| 21:14.38 | Harlekin | hmm, maybe i should patch that out of it |
| 21:15.02 | W8TVI | hey Harlekin |
| 21:15.10 | W8TVI | want to read something funny? |
| 21:15.52 | W8TVI | http://knerd.com/~kb8sjt/vlog.txthttp://knerd.com/~kb8sjt/vlog.txt |
| 21:16.35 | W8TVI | oops |
| 21:16.37 | W8TVI | http://knerd.com/~kb8sjt/vlog.txt |
| 21:16.40 | ljp | "but we wait with bated breath for when you will actually care to inform us about what you are blathering about." |
| 21:17.36 | W8TVI | ljp: :P |
| 21:19.57 | cyn_ | haha W8TVI |
| 21:20.01 | W8TVI | :) |
| 21:20.27 | W8TVI | truth is funner than fiction |
| 21:21.09 | *** join/#openembedded treke (~ggilbert@12.107.12.130) |
| 21:21.22 | treke | cool |
| 21:21.30 | treke | so dell went and sent me an axim update cd |
| 21:22.31 | kergoth | k libtoolize obeys STAGING_DIR now, so you can use it to prep a tree in buildroot with our versions of the world |
| 21:24.34 | mickeyl | andersee: ping |
| 21:24.45 | andersee | mickeyl: pong |
| 21:25.19 | mickeyl | andersee: i tried the binutils .6, but the error persists. still that shadowed symbol error. |
| 21:26.28 | andersee | that is strange |
| 21:27.42 | treke | my battery went dead |
| 21:27.51 | treke | that's the first that's happened :) |
| 21:29.14 | andersee | treke: wont be the last |
| 21:29.16 | andersee | :) |
| 21:29.29 | treke | andersee: might well be :) |
| 21:29.40 | treke | andersee: ipaq battery only went dead a handful of times :) |
| 21:29.44 | andersee | hehe |
| 21:30.03 | treke | at least I think the battery went dead. APM reads 100% :) |
| 21:30.19 | treke | but the damn thing wouldnt turn on till I had it plugged in for a few minutes |
| 21:30.22 | treke | ~lart lineo |
| 21:30.34 | kergoth | ~lart lineo |
| 21:30.41 | mickeyl | andersee: I patched configure.ac to skip that test only to run into the next problem... : |
| 21:30.50 | mickeyl | usr/local/projects/opensimpad/output/staging/target/lib/libdl.so: undefined reference to `_dl_linux_resolve' |
| 21:30.56 | mickeyl | usr/local/projects/opensimpad/output/staging/target/lib/libdl.so: undefined reference to `_dl_fixup' |
| 21:30.58 | treke | kergoth: the 750 battery monitoring do that also? |
| 21:31.14 | kergoth | treke: all Zs afaik. |
| 21:31.17 | kergoth | idiocy. |
| 21:31.55 | treke | kergoth: I have this feeling that when apm reads 5% it really means 50 :) |
| 21:32.11 | kergoth | heh, woudlnt surprise me |
| 21:32.47 | treke | since 75% seems to last about 20 minutes before it hits 50 |
| 21:32.54 | *** join/#openembedded SuKoShi (~sukoshi@dyn-195-242-112-52.ppp.tiscali.fr) |
| 21:33.39 | pb_ | the ipaq has rather better battery monitoring hardware, I think. |
| 21:33.57 | Harlekin | treke: zaurus has 4 steps .-) |
| 21:34.01 | Harlekin | as the older one |
| 21:34.01 | Harlekin | too |
| 21:34.13 | W8TVI | pb_: the zaurus can tell the battery voltage |
| 21:34.25 | W8TVI | its the sharp software that sucks |
| 21:34.35 | andersee | Harlekin: full, not-full, not-empty, empty |
| 21:34.37 | treke | pb_: was 2.4.6 the one that just assumed you would have 5 hours of life and just divided that up? :) |
| 21:34.59 | pb_ | W8TVI: right, ipaq doesn't rely on voltage measurement for capacity estimate. |
| 21:35.00 | kergoth | bipolar: problems? |
| 21:35.00 | bipolar | bin/ld: libaqpkg.so.1.0.0: undefined versioned symbol name std::time_put_w@@GLIBCPP_3.2 |
| 21:35.07 | bipolar | it's the same fucking error. |
| 21:35.09 | mickeyl | bipolar: sorry |
| 21:35.13 | treke | W8TVI: The ipaq can actually give you a decent time estimate on remaining life |
| 21:35.18 | mickeyl | bipolar: i just got the same one |
| 21:35.18 | treke | bipolar: I have the same problem. mickeyl doesnt |
| 21:35.26 | mickeyl | i have now |
| 21:35.28 | bipolar | treke: what OS? |
| 21:35.30 | mickeyl | after the latest cvs update |
| 21:35.33 | treke | bipolar: Debian unstable |
| 21:35.36 | W8TVI | the zaurus does it the same way that most everyone does it |
| 21:35.40 | mickeyl | treke,bipolar: but i guess Harlekin has found the problem |
| 21:35.41 | kergoth | pb_: what information does it have at its disposal? |
| 21:35.44 | mickeyl | something with the template |
| 21:35.47 | bipolar | mickeyl: you too now? So it's not just me? |
| 21:35.57 | treke | mickeyl: I get it in opieplayer2, oxygen, opierec, and tetrix |
| 21:35.57 | mickeyl | no longer :D |
| 21:36.14 | bipolar | I'm not a crazy loon who has a fucked up toolchain and won't admit it :D |
| 21:36.16 | pb_ | kergoth: the battery monitor can report instantaneous current draw, and also integrate current over time to give accumulated charge. |
| 21:36.17 | bipolar | yay! |
| 21:36.18 | mickeyl | treke: the new launcher scheme i guess |
| 21:36.44 | bipolar | ok... it's time for me to head home. I'll login from thiere. ttyl |
| 21:36.47 | mickeyl | cu |
| 21:36.51 | W8TVI | some one needs to write a better battery monitor... |
| 21:36.52 | pb_ | kergoth: (battery capacity - charge removed since going off AC) / (instantaneous current) = time remaining |
| 21:37.16 | kergoth | spiffy. |
| 21:38.08 | pb_ | it also measures battery temperature to compensate for varying efficiency in hot or cold conditions. |
| 21:38.34 | treke | pb_: I had lots of fun when the ipaq suddenly thought battery capacity was 500mah |
| 21:38.35 | pb_ | treke: yeah, we should teach the software to cope more gracefully with abandoned learning cycles. |
| 21:45.12 | *** join/#openembedded _Psycho_ (~Bob@MTL-HSE-ppp165301.qc.sympatico.ca) |
| 21:46.50 | *** join/#openembedded _Psycho (~Bob@HSE-Montreal-ppp3472126.sympatico.ca) |
| 21:47.42 | kergoth | k, kernel headers is merged into glibc |
| 21:47.53 | kergoth | autoconf & automake next |
| 21:47.56 | gb2 | redhat-ish style one? |
| 21:48.01 | gb2 | (kernel headers) |
| 21:48.18 | kergoth | including them in libc6-dev like debian, but its the glibc source package that gets them |
| 21:48.34 | kergoth | so sorta |
| 21:48.45 | gb2 | still lets you shoot yourself in the foot though? |
| 21:48.58 | kergoth | heh, we always let you shoot yourself in the foot |
| 21:49.02 | kergoth | just hte kinda people we are |
| 21:49.43 | pb_ | heh |
| 21:50.47 | *** join/#openembedded Twiun[af1] (~twiun@adorphuye.com) |
| 21:53.42 | *** join/#openembedded _Psy (~Bob@MTL-ppp-148330.qc.sympatico.ca) |
| 21:55.00 | CIA | repository=03buildroot user=11mickey@allanon.vanille.de |
| 21:55.01 | CIA | 07ChangeSet@1.1318 first attempt at proper VT support for Qt/E |
| 21:55.44 | CIA | repository=03buildroot user=11mickeyl@gandalf.tm.informatik.uni-frankfurt.de |
| 21:57.08 | pb_ | obviously mickey is a man with many identities. |
| 21:57.21 | mickeyl | :D |
| 21:57.41 | W8TVI | whats a good, small command line sound player for the zaurus |
| 21:57.46 | mickeyl | and you wonder why i can't concentrate anymore... |
| 21:58.22 | W8TVI | that can it output to /dev/sharp_buz |
| 21:59.09 | kergoth | uh |
| 21:59.16 | kergoth | OZ doesnt use sharp_buz for anything but compatibility |
| 21:59.21 | kergoth | we have a dsp for that |
| 21:59.47 | W8TVI | well.. whats the device name for the buzzer? |
| 22:00.11 | *** join/#openembedded noda (~noda@modemcable013.119-201-24.mtl.mc.videotron.ca) |
| 22:00.13 | noda | Moo! |
| 22:00.23 | W8TVI | I want to be able to have kismet make a sound when it finds something |
| 22:01.36 | kergoth | hey noda |
| 22:01.42 | W8TVI | m00 noda |
| 22:01.45 | kergoth | W8TVI: uh, the kismet buzzer app can make a buzz using sharp_buz |
| 22:01.51 | kergoth | W8TVI: in the OZ kismet pacckages |
| 22:01.52 | noda | kergoth: I'm having problems with buildroot and qte... I'll reproduce the error and then annoy you :) |
| 22:01.59 | kergoth | fuck that |
| 22:02.00 | W8TVI | kergoth: I tried it |
| 22:02.02 | kergoth | annoy somebody else |
| 22:02.03 | kergoth | I use OE |
| 22:02.09 | noda | kergoth: More fun annoying you :) |
| 22:02.22 | kergoth | noda: we have 26 packages converted so far |
| 22:02.31 | noda | kergoth: To OE? |
| 22:02.38 | kergoth | yes |
| 22:02.45 | noda | kergoth: Maybe once I figure out the old system I'll be able to convert to OE ;) |
| 22:02.46 | W8TVI | brb |
| 22:02.51 | kergoth | you dont need to know the old system |
| 22:02.52 | noda | I've got a 4-day weekend, after all :> |
| 22:02.56 | noda | Starting.... *now* |
| 22:02.56 | W8TVI | kergoth: I'll get you the error message |
| 22:02.57 | *** join/#openembedded bipolar (~bflong@ctt186139.ceinetworks.com) |
| 22:03.04 | bipolar | yo all |
| 22:03.05 | kergoth | i added apache when we never had it in before |
| 22:03.09 | kergoth | heh |
| 22:03.13 | noda | Really? Surprising :) |
| 22:03.24 | kergoth | apache 2.0.47 with ssl and webdav support no less |
| 22:03.27 | kergoth | fun stuff. |
| 22:03.27 | bipolar | ok... back to the fun stuff |
| 22:03.31 | noda | Ooooh :) |
| 22:03.37 | noda | What's webdav, btw? File transfers or something? |
| 22:03.38 | kergoth | noda: 3.5 meg package, compressed |
| 22:03.39 | kergoth | hehe |
| 22:03.47 | noda | WHOA :P |
| 22:03.48 | kergoth | course it has all the docs & man pages in it atm |
| 22:03.49 | noda | NOTE: import patch '/usr/src/bk/buildroot-oz/build/opie/qt/qte234-for-opie091-makefile.patch' as 'qte234-for-opie091-makefile' |
| 22:03.49 | noda | ERROR: patch did not apply, try again with force (-f -i ...) |
| 22:03.49 | noda | Patch had errors |
| 22:03.57 | kergoth | noda: mrproper your opie and rebuild. |
| 22:04.01 | noda | I just did :P |
| 22:04.02 | kergoth | er |
| 22:04.04 | kergoth | and qte |
| 22:04.11 | kergoth | then you did something wrong |
| 22:04.19 | noda | I did make SUBDIRS=packages/qte mrproper; make SUBDIRS=packages/opie mrproper |
| 22:04.19 | noda | lol |
| 22:04.20 | *** join/#openembedded W8TVI|z (~zic@3639244824.mi.dial.hexcom.net) |
| 22:04.22 | noda | I'll do it again :) |
| 22:04.52 | noda | Then 'nice -+20 make', no objections to that? :) |
| 22:05.16 | kergoth | noda: http://openembedded.bkbits.net:8080/packages/anno/content/zsh-4.1.1.oe@1.3?nav=index.html|src/|src/content <-an example of an .oe for an autoconf-based package. |
| 22:05.37 | noda | Holy mother of sweetness! |
| 22:06.12 | kergoth | actually, i should set the default S var to ${WORKDIR}/${P} |
| 22:06.17 | noda | kergoth: And when a new version comes out, you just rename the .oe file? |
| 22:06.19 | kergoth | thatll remove a line from the majority of the .oe's |
| 22:06.22 | kergoth | noda: yep |
| 22:06.25 | kergoth | no |
| 22:06.25 | noda | Awesome |
| 22:06.27 | kergoth | you add a new one actually |
| 22:06.31 | kergoth | we dont lose old versions |
| 22:06.32 | noda | Oh, right. |
| 22:06.34 | noda | Cool |
| 22:06.37 | kergoth | so at any time you can build any version of a thing |
| 22:06.43 | kergoth | at least, once we start mirroring soruces you can |
| 22:06.48 | noda | ...as long as the sources st-- right :)] |
| 22:07.08 | noda | kergoth: I'm getting this qte error again. |
| 22:07.14 | noda | Just mrproper'd qte and opie. |
| 22:07.23 | W8TVI | kergoth: |
| 22:07.23 | noda | This is a clean buildroot checkout from yesterday... |
| 22:07.23 | W8TVI | Error Getting Buzzer Volume: Function not implemented |
| 22:07.24 | W8TVI | buzzme:main: Problems getting volume on device '/dev/sharp_buz'. |
| 22:07.30 | kergoth | noda: what branch are you using? |
| 22:07.34 | kergoth | W8TVI: thats an OZ package? |
| 22:07.37 | noda | kergoth: Erm... |
| 22:07.38 | noda | ibot oz bk |
| 22:07.39 | | oz bk is probably bk clone bk://openzaurus.bkbits.net/buildroot buildroot-oz; cd buildroot-oz; bk -r co -q; cp def-configs/collie-opie .config (for zaurus 5000d/5500) .. make oldconfig; make; or see bitkeeper (free download but non-gpl) or see 'oz snapshot' if you either cannot or will not use bitkeeper. |
| 22:07.39 | W8TVI | yep |
| 22:07.41 | noda | <-- that one :) |
| 22:07.41 | kergoth | thats obviously a stock buzz tool |
| 22:07.43 | kergoth | noda: no. |
| 22:07.47 | kergoth | noda: bk doesnt have branches. |
| 22:07.53 | kergoth | noda: OPIE branch |
| 22:07.53 | noda | Branch of what then? |
| 22:07.56 | kergoth | noda: you select in the menuconfig |
| 22:07.57 | noda | v1_0 |
| 22:07.57 | W8TVI | I've tried all 3 versions |
| 22:08.00 | kergoth | dont use that |
| 22:08.04 | noda | HEAD? |
| 22:08.05 | kergoth | use BRANCH_1_0 |
| 22:08.09 | kergoth | v1_0 is a static tag |
| 22:08.09 | noda | Okay |
| 22:08.15 | kergoth | branch_1_0 is a bugfix branch |
| 22:08.19 | kergoth | with lots of fixes |
| 22:08.21 | noda | I was wondering about that, but I figured v1_0 was the branch :) |
| 22:08.23 | noda | Awesome :) |
| 22:09.05 | W8TVI | does any one have a copy of buzzme that they know that works that they can stick on their webserver for me? |
| 22:09.46 | noda | kergoth: You sure it's not 'branch1_0'? |
| 22:09.49 | noda | cvs [server aborted]: no such tag branch_1_0 |
| 22:10.02 | W8TVI | I tried installing the oz stable kismet and it's buzzme gave me the same error |
| 22:10.16 | W8TVI | same with dragorn's |
| 22:10.21 | kergoth | noda: its all caps. |
| 22:10.26 | noda | Ah, whoops :) |
| 22:10.28 | noda | Thanks :) |
| 22:10.53 | W8TVI | and the one on www.killefiz.de/zaurus/ |
| 22:11.20 | bipolar | so, opie is not building right now do to the error I've been getting? Or is it just not building with gcc3? |
| 22:11.27 | W8TVI | all three gave me that error |
| 22:11.38 | kergoth | noda: treke got an rt instance setup, been playing with it. i like it alot :) |
| 22:11.46 | mickeyl | bipolar: i start to believe it is the toolchain. |
| 22:11.58 | mickeyl | bipolar: it works fine with gcc3.x/x86 |
| 22:12.05 | bipolar | mickeyl: not again. |
| 22:12.06 | W8TVI | kergoth: now what? |
| 22:12.11 | bipolar | mickeyl: *sigh* |
| 22:12.19 | noda | kergoth: Great! :) |
| 22:12.40 | mickeyl | bipolar: well - wish i had better news, but this is the only explanation or is it? how come it is looking for some 3.2 symbol in there |
| 22:12.56 | noda | kergoth: No errors! Woo! :) |
| 22:12.58 | mickeyl | seems something is mixed up there |
| 22:13.21 | mickeyl | bipolar: I'd try with my all brand-new 3.3.1-uclibc toolchain but I'm hanging @ libxine yet |
| 22:13.28 | kergoth | W8TVI: how should i know? |
| 22:13.33 | bipolar | mickeyl: I've been using these 3.x toolchains for a while. These errors are new. I was able to build opie with an older cvsdate and it went fine. |
| 22:13.34 | kergoth | i'm a bit busy doing what i get paid to do atm |
| 22:13.49 | W8TVI | kergoth: just though you'd have an idea |
| 22:13.51 | pb_ | that old excuse |
| 22:14.07 | mickeyl | bipolar: ok. I'm lost there |
| 22:15.17 | mickeyl | argh |
| 22:15.20 | mickeyl | shawn is depressing me |
| 22:15.26 | mickeyl | http://www.zaurus.com/dev/board/index.php?showtopic=5966&st=15&hl= |
| 22:16.24 | pb_ | who's shawn? |
| 22:16.26 | mickeyl | i fear he has a point, though |
| 22:16.30 | mickeyl | shawn gordon |
| 22:16.31 | mickeyl | aka bigboss |
| 22:16.34 | mickeyl | mr. tkd |
| 22:16.35 | mickeyl | tkc even |
| 22:17.06 | pb_ | ah |
| 22:17.13 | pb_ | "Stay with me on this and I'll explain why cloning Qtopia is pointless." |
| 22:17.18 | bipolar | mickeyl: the only thing I can find with google talks about that error coming up when libgcc is not built shared, or can not be found. |
| 22:17.36 | mickeyl | pb_: yeah... i fear he really has a point |
| 22:17.39 | mickeyl | i hate legal stuff |
| 22:17.41 | mickeyl | really |
| 22:18.19 | mickeyl | ljp |
| 22:18.20 | mickeyl | is that right? |
| 22:18.21 | mickeyl | "One of the thingsin the Trolltech commercial license is that you cannot create competing software with their tools. |
| 22:18.30 | mickeyl | is that true? |
| 22:18.36 | pb_ | well, he doesn't seem to be saying anything really new. |
| 22:18.50 | ljp | what? |
| 22:18.56 | mickeyl | hmm it was news for me |
| 22:19.11 | Harlekin | hehe |
| 22:19.16 | mickeyl | how the hell shall we do an independent framework then? |
| 22:19.18 | Harlekin | big boss going down? |
| 22:19.24 | W8TVI | "loosing compatibility with all existing software"? bah! |
| 22:19.25 | kergoth | need a company to back Opie. |
| 22:19.34 | kergoth | to pursue its shipment on devices |
| 22:19.42 | pb_ | mickeyl: well, I must admit, I thought you _were_ using the GPL'd qt/e. |
| 22:19.43 | noda | I know it's OT, but anybody here use GCJ? I'm a newbie to Java and I wanna use gcc :P |
| 22:19.44 | mickeyl | and supporting the $$$ to buy that special licenses ? |
| 22:19.56 | W8TVI | I could use it on my pdas that I'm thinking about making |
| 22:20.01 | ljp | I was told.. you _can_ _write a lgpl library for qtopia/opie, but when you link it with gpl, it essentially is gpl. |
| 22:20.14 | bipolar | kergoth: hp/compaq? :) |
| 22:20.16 | pb_ | yeah, that's the difference between lgpl and gpl. |
| 22:20.27 | mickeyl | pb_: we are of course. but we wanted to change that to get in relationsships with companies |
| 22:20.39 | pb_ | mickeyl: ah, right, I see. |
| 22:20.41 | kergoth | mickeyl: yep. thatd be the only way to kill the gpl infection from qte. |
| 22:20.57 | ljp | harmony-embedded |
| 22:21.12 | Harlekin | hmm |
| 22:21.16 | W8TVI | IBM needs to make a pda that runs linux and uses opie :) |
| 22:21.19 | Harlekin | lets so how long gpl will last anyway |
| 22:21.22 | kergoth | at that point you're essentially another trolltech, but with better ties to the community. |
| 22:21.29 | mickeyl | so that means we can dump that plan unless we either have the $$$ to buy the special license or rewrite Qt/E ? |
| 22:21.32 | Harlekin | it does not hold in court |
| 22:21.47 | Harlekin | in nearly all important countries |
| 22:21.53 | Harlekin | that is one think shawn was right about |
| 22:21.58 | kergoth | mickeyl: or we abandon commercial app support completely |
| 22:22.06 | kergoth | mickeyl: and let gpl infect away |
| 22:22.07 | kergoth | heh |
| 22:22.08 | mickeyl | kergoth: that would suck. |
| 22:22.12 | W8TVI | screw the other countrys |
| 22:22.15 | kergoth | yep, but those are the options |
| 22:22.28 | mickeyl | kergoth: hell it would be nice to make a few bucks out of that much of time we spend on that project |
| 22:22.35 | Harlekin | W8TVI: which others, that where the gpl has a chance? |
| 22:22.47 | ljp | but I think, a company with a commercial license, can link with the commercial qtopia & the lgpg lib and be ok |
| 22:22.50 | W8TVI | no... the us :P |
| 22:22.53 | mickeyl | i rather should get employed for a company which lets me do gpl work then... |
| 22:22.58 | W8TVI | and the EU |
| 22:22.59 | kergoth | mickeyl: yep, but think about how much the licensing will hurt us |
| 22:23.14 | kergoth | mickeyl: we'd have to line up some kickass contracts to compensate |
| 22:23.30 | pb_ | ljp: mmm, but if you use the commercial qtopia you couldn't produce a zero-cost distribution for devices like ipaq, presumably. |
| 22:23.31 | mickeyl | kergoth: hmm... bummer. that wasn't really clear to me. |
| 22:23.45 | mickeyl | did i say i hate legal stuff today yet ? |
| 22:23.59 | bipolar | Question: Why is it trying to find a gcc 3.2 symbol when I'm using gcc 3.3 to build it? |
| 22:24.14 | kergoth | pb_: well, somebody would have to swallow the qte licensing costs |
| 22:24.15 | kergoth | heh |
| 22:24.18 | Harlekin | bipolar: since 3.3 supercedes 3.2 |
| 22:24.36 | pb_ | kergoth: yeah. if it was just a one-off hit that would probably be tolerable, but presumably commercial qt/e also has per unit royalties. |
| 22:24.41 | bipolar | Harlekin: So that symbol should also be in 3.3? |
| 22:24.50 | Harlekin | pb_: nope |
| 22:25.03 | Harlekin | pb_: thats propaganda nils spreads |
| 22:25.05 | bipolar | std::time_put_w@@GLIBCPP_3.2 |
| 22:25.08 | Harlekin | knowingly better |
| 22:25.35 | pb_ | bipolar: GLIBCPP_3.2 is the abi version, not necessarily tied to the tool version. |
| 22:25.37 | W8TVI|z | I want to build a zaurus like pda, and market the thing |
| 22:25.46 | W8TVI|z | better than sharp |
| 22:25.51 | Harlekin | W8TVI: good luck |
| 22:25.51 | Harlekin | .-) |
| 22:26.21 | ljp | I have a vercel prototype here you can use... ;) no one else is using it |
| 22:26.27 | pb_ | heh |
| 22:26.33 | W8TVI|z | heh |
| 22:26.39 | W8TVI|z | vercel? |
| 22:26.45 | kergoth | ljp: are there per unit royalties for the commercial qte? |
| 22:27.06 | ljp | hm, not real sure |
| 22:27.09 | collord | sorry, why does the opie effort need a commercial qte license? |
| 22:27.20 | mickeyl | we are talking about future developments |
| 22:27.22 | mickeyl | not current |
| 22:27.36 | bipolar | ok... further google searching has told me that this error is defenatly something to do with using libgcc |
| 22:27.36 | kergoth | collord: in order to allow closed source apps to not be infected by the gpl qte via our lgpl opie library |
| 22:27.54 | ljp | I think I did hear of something like that, royalities perunit |
| 22:28.11 | W8TVI|z | if I were to do the pda thing, I'd want to use OE/Opie |
| 22:28.31 | kergoth | well, we could charge royalties per unit as well, but we'd have to keep it competitive, which means we'd have to make some serious cash to offset the loss |
| 22:29.07 | pb_ | main problem with royalties isn't so much the cash as the fact that suddenly you have to track all your users and figure out who has a license. |
| 22:29.16 | kergoth | heheh, true that |
| 22:29.46 | collord | please if you do something like this, make it a foundation |
| 22:29.49 | killefiz | treke: are you around? |
| 22:30.11 | pb_ | I doubt the per-unit cost is more than a couple of dollars, and most users would probably cough that up without much of a struggle. But you need to have the machinery in place to deal with that and funnel the payments into trolltech's coffers. |
| 22:30.24 | ljp | less than a dollar |
| 22:30.50 | ljp | most oems wont pay a dollar per, for all software royalties |
| 22:30.57 | pb_ | right |
| 22:30.58 | Harlekin | ljp: but wasnt that qtopia? |
| 22:31.01 | Harlekin | ljp: not qpe |
| 22:31.04 | Harlekin | arg |
| 22:31.06 | Harlekin | ljp: not qte |
| 22:31.21 | W8TVI | are you talking about commercial apps being compiled using an opie lib? |
| 22:31.30 | kergoth | W8TVI: s/compiled/run/ |
| 22:31.33 | W8TVI | I'm not sure I understaned |
| 22:31.38 | collord | there are runtime licenses for qt embedded. |
| 22:31.39 | kergoth | s/?/ legally?/ |
| 22:31.43 | Harlekin | afair there is _no_ per unit cost for using qte |
| 22:31.50 | Harlekin | and we dont really need libqpe |
| 22:31.55 | Harlekin | in future |
| 22:31.57 | kergoth | yep |
| 22:32.01 | kergoth | ljp: can you confirm? |
| 22:32.03 | mickeyl | hmmm |
| 22:32.04 | Harlekin | and we dont want to deploy qtopia on devices |
| 22:32.07 | kergoth | whehter there is or isnt for qte |
| 22:32.13 | collord | http://www.trolltech.com/products/embedded/pricing.html?cid=1 |
| 22:32.35 | W8TVI | I guess I dont understand how running a commercial app on a gpl os is a problem |
| 22:32.38 | Harlekin | collord: yeah, per developer |
| 22:32.43 | mickeyl | that really makes a detailed discussion neccessary if and where we want to steer with that project. |
| 22:32.51 | Harlekin | collord: yeah, per developer / creating closed source app |
| 22:33.06 | Harlekin | arg |
| 22:33.06 | kergoth | Harlekin: look at the bottom, at the runtime license cost |
| 22:33.08 | Harlekin | runtime license |
| 22:33.11 | kergoth | yep |
| 22:33.17 | Harlekin | that is still in place? |
| 22:33.21 | Harlekin | paul said something else |
| 22:33.30 | Harlekin | man that makes 6 ¢ per unit |
| 22:33.35 | collord | W8TVI|z: you're linking against a gpl library |
| 22:33.44 | kergoth | W8TVI: qte is gpl. lgpl libopie links against qte. libopie now gpl infected. commercial app links against either, gpl infected. |
| 22:33.58 | Harlekin | kergoth: or we just wait 1 year until gpl is no more |
| 22:33.59 | Harlekin | .-) |
| 22:34.34 | kergoth | eh, it says # of devices, and cost per license |
| 22:34.37 | W8TVI | tht sounds stupid... |
| 22:34.40 | kergoth | but doenst clarify how many devices are covered per license |
| 22:34.55 | kergoth | just one? if so your calc is wrong Harlekin |
| 22:34.55 | Harlekin | one maybe |
| 22:34.56 | Harlekin | .-) |
| 22:34.58 | Harlekin | yeah |
| 22:35.09 | kergoth | $6 a device would be insanity |
| 22:35.10 | kergoth | heh |
| 22:35.12 | kergoth | imho |
| 22:35.13 | pb_ | what's the licence on qt/x11? |
| 22:35.14 | Harlekin | yeah |
| 22:35.14 | pb_ | heh |
| 22:35.22 | Harlekin | where you get wince for free |
| 22:35.27 | Harlekin | and it really really improved lately |
| 22:36.00 | ljp | $6 per 100 |
| 22:36.06 | kergoth | heh, you know, given OE, companies like montavista dont give quite as much a value add as they used to |
| 22:36.21 | Harlekin | kergoth: you could sell OE platform |
| 22:36.22 | Harlekin | .-) |
| 22:36.26 | kergoth | ljp: then why does it show $3 under 10000? $3 for what? |
| 22:36.30 | Harlekin | and use the money for rolaties ,) |
| 22:36.33 | kergoth | Harlekin: yep, thought about it. |
| 22:36.37 | collord | no way, that's per device |
| 22:36.55 | kergoth | ljp: it doesnt make any sense for that table to be anything but 1 license per device |
| 22:36.58 | kergoth | afaict |
| 22:37.03 | Harlekin | qte on top of wince is fun btw |
| 22:37.09 | ljp | as I said, oems will not pay over a $1.00 per unit for all royalities |
| 22:37.15 | pb_ | kergoth: right. $6 per device isn't completely outrageous in low volume. |
| 22:37.26 | pb_ | ljp: right, but those oems will be making millions of units, not 100. |
| 22:37.34 | kergoth | ah, true that |
| 22:37.41 | pb_ | if you are manufacturing in volumes of 100, you are not in a cost sensitive market. |
| 22:38.23 | ljp | I think its a $6 license covers 100 units |
| 22:38.33 | kergoth | that doesnt seem right ljp, look at the table |
| 22:38.37 | kergoth | or go ask someone for us |
| 22:38.38 | kergoth | :) |
| 22:38.39 | noda | kergoth: Host type 'arm-unknown-linux-gnu' (arm-linux) is currently not supported by xine |
| 22:38.41 | pb_ | ljp: then what for the rest of the table? are they all for qty 100? |
| 22:38.45 | kergoth | noda: harmless |
| 22:38.45 | noda | kergoth: Is that normal/good? |
| 22:38.46 | Harlekin | noda: yeah yeah |
| 22:38.48 | noda | Okay, great :) |
| 22:38.56 | ljp | hmmm |
| 22:39.01 | pb_ | ljp: it would make no sense for the $3 license to cover all 10000 units. |
| 22:39.11 | noda | Just making sure! |
| 22:39.14 | kergoth | right, thered be no point to paying $6 for 100 then |
| 22:39.15 | kergoth | hehe |
| 22:39.23 | pb_ | heh, quite. |
| 22:40.18 | bipolar | we could all switch to GPE.... |
| 22:40.27 | collord | blah |
| 22:40.28 | mickeyl | is that lgpl ? |
| 22:40.42 | bipolar | I beleve so |
| 22:40.47 | mickeyl | sounds good. |
| 22:40.48 | kergoth | theres no way in hell i'm doing gtk coding |
| 22:40.50 | kergoth | no offense |
| 22:40.51 | pb_ | ljp: there are certainly OEMs operating in vertical markets who produce low volume, high ticket products. if your device retails for $1000, which wouldn't be unrealistic in that space, you can probably swallow a $6 royalty. |
| 22:40.54 | kergoth | :) |
| 22:40.55 | collord | what's the point though? you're not distributing commercial apps |
| 22:41.06 | Harlekin | collord: mickey wants to make money |
| 22:41.11 | collord | if someone wants to distribute oe and their commerical apps on a device, let them manage licensing |
| 22:41.12 | pb_ | collord: no, but it's desirable that others be able to do so. |
| 22:41.18 | kergoth | collord: others are. do we prevent them from producing products for our platform? |
| 22:41.20 | Harlekin | question is : how to make money with no market share .-) |
| 22:41.21 | kergoth | right |
| 22:41.25 | kergoth | Harlekin: hehe |
| 22:41.30 | Harlekin | wince |
| 22:41.32 | kergoth | s/share/share yet/ |
| 22:41.32 | Harlekin | palmos |
| 22:41.36 | collord | no, but you let them figure out how to drop in a commercial-ready qte |
| 22:41.37 | Harlekin | both improving |
| 22:41.37 | kergoth | ugh, headache |
| 22:41.53 | pb_ | collord: if they have to replace qte, they are probably not going to use the opie platform. |
| 22:42.13 | kergoth | you want every customer to have to replace qte and install the app? |
| 22:42.15 | kergoth | thats a joke |
| 22:42.24 | collord | no |
| 22:42.25 | kergoth | particularly if hte device has a read only primary partition |
| 22:42.26 | pb_ | heh, and not even a funny one. |
| 22:42.27 | kergoth | for safety |
| 22:42.29 | kergoth | which nearly all do |
| 22:42.40 | bipolar | um... wait a min. |
| 22:42.46 | collord | so it's e.g. tkc that's worried about all this? |
| 22:43.03 | pb_ | collord: no, tkc won't be worried, they have nothing to lose. |
| 22:43.04 | kergoth | collord: its us worried about tkc apps not being able to be used legally on the platform. |
| 22:43.05 | bipolar | why would a developer need to have his app run on only a comercialy licenesed qte? |
| 22:43.08 | kergoth | pb_: exactly |
| 22:43.19 | kergoth | bipolar: his app is closed source |
| 22:43.29 | kergoth | bipolar: your hypothetical developer that is |
| 22:43.31 | collord | but it's the end users that would be the violators |
| 22:43.44 | bipolar | kergoth: what does that mean for the license? |
| 22:43.49 | pb_ | collord: no, it would be those distributing the code who are violating the license. |
| 22:43.49 | kergoth | collord: yep |
| 22:43.51 | *** join/#openembedded masticator (~jason@ool-43512bbd.dyn.optonline.net) |
| 22:43.51 | *** mode/#openembedded [+o masticator] by ChanServ |
| 22:43.52 | kergoth | ah |
| 22:43.54 | kergoth | hey masticator |
| 22:43.57 | masticator | yo. |
| 22:44.05 | kergoth | welcome to #gpllegality |
| 22:44.16 | masticator | lol |
| 22:44.17 | Harlekin | if gpl is "legal" at all |
| 22:44.24 | Harlekin | according to shawn its not at all |
| 22:44.25 | noda | rofl |
| 22:44.37 | masticator | hey |
| 22:44.41 | masticator | softfield updated its web site |
| 22:44.43 | kergoth | well, we know shawn is often full of shit, so it could go either way |
| 22:44.44 | masticator | when did they do that? |
| 22:44.46 | ljp | where does shawn say that? |
| 22:44.52 | bipolar | If a user cant use a Comercial qt app on a GPL'ed qt lib, then how the hell does TKC get away with the comercial qt apps they make? |
| 22:44.55 | Harlekin | ljp: on irc |
| 22:45.08 | ljp | he is still around on irc? |
| 22:45.11 | noda | bipolar: They make it with a non-GPL'd qt lib |
| 22:45.20 | Harlekin | ljp: was a way back |
| 22:45.27 | ljp | ahhh |
| 22:45.31 | kergoth | bipolar: its the users choice to install it onto opie, and its questionable legally |
| 22:45.44 | bipolar | noda: so, then the developer of a comercial app just needs to get his own qt licence. |
| 22:45.48 | Harlekin | as the whole gpl is questionable .-) |
| 22:46.01 | noda | bipolar: Yup. |
| 22:46.09 | kergoth | bipolar: you're failing to see the point. yes, thats true. and as soon as it gets put on a device with a gpl qte, its infected. game over |
| 22:46.10 | bipolar | kergoth: so, the guy who says that it can't be done is doing it? |
| 22:46.17 | kergoth | hmm? |
| 22:46.27 | ljp | mickeyl: you might want to direct questions to sales@trolltech.com |
| 22:46.28 | kergoth | ~fishslap bipolar |
| 22:46.29 | | ACTION slaps bipolar up side the head with a wet fish. |
| 22:46.45 | bipolar | kergoth: TKC sells apps that run on KDE. How could he legaly do that? |
| 22:46.59 | ljp | kde has a special license from TT |
| 22:46.59 | noda | rofl, this is funny: my first assignment in Intro to Computers is to make a web page :P |
| 22:47.00 | Harlekin | special license for kde |
| 22:47.13 | mickeyl | ljp: yeah. |
| 22:47.16 | *** join/#openembedded imk|z (pda@68.170.56.122) |
| 22:47.18 | W8TVI | I dont understand how some software can get "infected" by gpl |
| 22:47.23 | bipolar | I see. So why can't we get that? |
| 22:47.25 | kergoth | W8TVI: then you dont understand the gpl. |
| 22:47.27 | kergoth | W8TVI: read it |
| 22:47.43 | pb_ | kergoth: I don't think end users are in any jeopardy; GPL only applies restrictions to distribution, not to use. You can do whatever you like in the privacy of your own home. |
| 22:47.46 | kergoth | bipolar: you think tt would give us a special license to produce a product that will directly compete with qtopia? |
| 22:47.53 | kergoth | bipolar: you must be kidding. |
| 22:47.56 | W8TVI | does it say in the gpl that if you use any software with a gpled software that the software is now gpl? |
| 22:48.08 | kergoth | W8TVI: not _with_. linked against. |
| 22:48.13 | kergoth | pb_: ah. hmm |
| 22:48.15 | pb_ | W8TVI: very approximately, yes. |
| 22:48.16 | Harlekin | bipolar: TT fears opie like hell |
| 22:48.28 | W8TVI | thats the stupidest thing I have ever heard |
| 22:48.38 | Harlekin | bipolar: why should they support a superior platform when having its own competitior |
| 22:48.48 | pb_ | W8TVI: anything linked with gpl software is potentially a derived work, and must be distributed under gpl terms. |
| 22:48.59 | bipolar | God... |
| 22:49.02 | kergoth | imho TT should drop the qtopia apps, focus on the lib, and work with opie on the app front |
| 22:49.02 | mickeyl | so can we conclude that having a company-friendly opie development library is doomed in the first place? |
| 22:49.05 | kergoth | but hey |
| 22:49.16 | noda | Wait wait wait wait wait... aren't QT apps linked statically by the developer? |
| 22:49.19 | kergoth | mickeyl: short of having a company to push it, probably |
| 22:49.25 | kergoth | noda: not generally. |
| 22:49.36 | noda | But that's the way to get around the legal issues, right? Plain and simple. |
| 22:49.42 | pb_ | noda: the binaries would be huge if every application was carrying qt around with it. |
| 22:49.56 | kergoth | why would they do that? what do they have to gn? |
| 22:49.57 | kergoth | gain |
| 22:49.58 | kergoth | bleh |
| 22:50.02 | pb_ | right |
| 22:50.12 | W8TVI | after gpl dies, some one with a brain should make a new "gpl" like thing... |
| 22:50.14 | noda | pb_: yup. But if they don't, it's illegal.... |
| 22:50.30 | noda | What is a major failing of the GPL that makes it doomed? |
| 22:50.49 | mickeyl | kergoth: too bad. because that what we had in mind definitly was way over a fun-project, especially regarding time-consume :) |
| 22:50.50 | pb_ | W8TVI: in this context, the GPL is doing exactly what it was intended to do. |
| 22:51.06 | mickeyl | s/over/more |
| 22:51.06 | W8TVI | "kill microsoft" |
| 22:51.13 | kergoth | mickeyl: yep |
| 22:51.20 | killefiz | W8TVI: you've obviously never read the gpl - you should change that |
| 22:51.22 | kergoth | W8TVI: get clue. return. rejoin conversation. |
| 22:51.22 | pb_ | W8TVI: it was invented precisely to discourage people from producing closed-source applications, by making it impossible for them to leverage GPL'd product in doing so. |
| 22:51.58 | pb_ | heh, I keep typing "GPE" by mistake and having to back up. |
| 22:52.48 | killefiz | summary: opie has to stay bc to qtopia |
| 22:52.57 | kergoth | killefiz: that isnt whats in question here |
| 22:53.03 | kergoth | not really anyway |
| 22:53.17 | kergoth | s/not really/only indirectly/ |
| 22:53.24 | collord | so the idea is there would be an org to maintain a commercial-okay opie in parallel with the gpl? |
| 22:53.25 | killefiz | hehe |
| 22:53.34 | collord | s/gpl/gpl version/ |
| 22:53.44 | pb_ | noda: it's not illegal for those who produce the applications, but it would be impossible to distribute a device containing the gpl qt/e library and a commercial application without infringing the licence. |
| 22:53.44 | W8TVI | is it posable run an app on opie with out it being linked against something gpl? |
| 22:53.52 | kergoth | collord: there's no need for them to be any different. as long as they have the license |
| 22:53.52 | mickeyl | collord: sounds like an option - at least theoretically |
| 22:54.14 | kergoth | pb_: hmm, so what if one doesnt distribute both the lgpl lib and the commercial application? but only one? |
| 22:54.17 | kergoth | pb_: i.e. our case |
| 22:54.22 | kergoth | heh |
| 22:54.27 | collord | kergoth: "they" being the opie.org or the commercial vendor |
| 22:54.30 | pb_ | kergoth: then, afaik, you have nothing to fear. |
| 22:54.30 | kergoth | W8TVI: qte is gpl. |
| 22:54.35 | kergoth | pb_: hmm |
| 22:54.55 | pb_ | kergoth: the problem would only arise if you wanted commercial vendors to adopt opie as their primary platform, ie displace qtopia. |
| 22:55.23 | kergoth | right, since they couldnt include their own custom closed source apps in their images |
| 22:55.25 | pb_ | or if you wanted to ship closed source applications as part of opie itself, but that seems less likely. |
| 22:55.27 | kergoth | bleh |
| 22:55.37 | pb_ | kergoth: yep, exactly. |
| 22:55.40 | mickeyl | yeah |
| 22:55.45 | W8TVI | ok, is it posable run an app on opie/qte with out it being linked against something gpl? can you make a self contained app that runs on opie/qte? |
| 22:55.55 | kergoth | W8TVI: QTE IN THIS CONTEXT IS GPL |
| 22:55.59 | kergoth | ~lart W8TVI |
| 22:56.06 | kergoth | heh |
| 22:56.21 | W8TVI | I dont CARE that qte is gpl! |
| 22:56.30 | collord | running "on" opie/qte implies that it's linked against it |
| 22:56.31 | kergoth | then you're a fucking idiot. |
| 22:56.33 | pb_ | W8TVI: then, er, why are you participating in this discussion? |
| 22:56.35 | kergoth | running an app is runtime linking |
| 22:56.50 | W8TVI | ok, thats what I was asking |
| 22:57.55 | W8TVI | you guys assumed that I knew that |
| 22:58.25 | pb_ | <kergoth> W8TVI: get clue. return. rejoin conversation. |
| 22:58.35 | kergoth | a certain amount of understanding of what linking is and how the gpl works was a prerequisite to a conversation on consequences of the gpl. |
| 23:00.16 | kergoth | anyway, back to OE |
| 23:00.50 | bipolar | Lets put all this legal stuff on hold. We should just come up with an letter and send it to TT to see what they say about it. |
| 23:01.02 | kergoth | that wont help much |
| 23:01.11 | kergoth | we need to determine exactly what our goals are from the project |
| 23:01.19 | kergoth | and then look at how feasable it is given our circumstances |
| 23:02.12 | bipolar | kergoth: if we base our system on qte, and we want to go comercial, then we need to talk to TT's lawyers about what they expect. |
| 23:02.45 | bipolar | If we don't care about going comercial, and just run gpl apps, then thats another story. |
| 23:02.50 | kergoth | how many if's are you going to have in a statement |
| 23:02.55 | kergoth | lets revamp what i just said |
| 23:02.57 | kergoth | establish goals |
| 23:03.01 | kergoth | act |
| 23:03.04 | kergoth | not vice versa |
| 23:03.07 | bipolar | ok |
| 23:03.09 | kergoth | heh |
| 23:03.17 | bipolar | Do we want to go comercial? |
| 23:03.19 | bipolar | y/n? |
| 23:03.21 | bipolar | :) |
| 23:03.54 | bipolar | personaly, I don't care one way or the other. |
| 23:04.03 | bipolar | ljp: realllllly.... :) |
| 23:04.12 | ljp | :D |
| 23:04.23 | bipolar | ljp: you work for TT? |
| 23:04.25 | W8TVI | no comercial = free software (for the most part) |
| 23:04.59 | kergoth | W8TVI: depends on whether or not pb is right in his opinion on the subject |
| 23:05.07 | kergoth | if he is, it isnt a concern as things stand |
| 23:05.40 | ljp | not really. currently I am consulting TT, until my extended work visa gets ok'd |
| 23:05.51 | bipolar | ljp: I see |
| 23:05.53 | ljp | bipolar: I move to Brisbane over a month ago |
| 23:06.07 | ljp | hence my ip |
| 23:06.15 | bipolar | ljp: so you know TT's number? :) |
| 23:06.28 | kergoth | W8TVI: thank you captain obvious! |
| 23:06.29 | ljp | uhh, ya. I am sitting in TT office. |
| 23:06.37 | W8TVI | any one hear about sharps new 3-D laptop? |
| 23:06.45 | ljp | right next to warwick's office |
| 23:06.52 | bipolar | ljp: heh... All I see is asdl connection in .au |
| 23:06.57 | CaptainObvious | dang |
| 23:07.02 | mickeyl | lets continue our talk on the developer mailing list. everyone of us take the time and write was his personal goals are and where he would see a future for him. then we can discuss further. |
| 23:07.03 | CaptainObvious | -NickServ- This nickname is owned by someone else |
| 23:07.09 | kergoth | hehe |
| 23:07.14 | kergoth | mickeyl: agreed |
| 23:07.29 | kergoth | mickeyl: you want to initiate the thread? |
| 23:07.35 | mickeyl | yes i will |
| 23:07.46 | ljp | and contact sales, for professional answers |
| 23:07.49 | mickeyl | kergoth: i'll write after getting hours of sleep :D |
| 23:07.52 | kergoth | hehe |
| 23:07.53 | mickeyl | good night all |
| 23:07.54 | kergoth | good plan |
| 23:07.56 | kergoth | night |
| 23:07.57 | W8TVI | hmmm |
| 23:08.12 | Captain_Obvious | AAAHHHHHHH!!!!!! |
| 23:08.25 | Captain-Obvious | there we go |
| 23:09.01 | Captain-Obvious | "Sharp Will Sell Laptop Showing 3-D Images" |
| 23:09.12 | Captain-Obvious | how badly are they going to market that? |
| 23:09.44 | gb2 | i read that as "3-D glasses" |
| 23:09.53 | kergoth | gb2: you want to pursue tar device table support or should I? I asked around, but nobody seems to know if such a patch already exists. |
| 23:09.53 | Captain-Obvious | http://apnews.myway.com/article/20030911/D7TG6UGG0.html |
| 23:09.59 | Harlekin | gb2: Captain-Obvious no |
| 23:10.05 | Harlekin | sharp has 3d displays |
| 23:10.10 | gb2 | and imagined cardboard glasses with red/green stuff |
| 23:10.13 | Harlekin | but you get eye cancer after 20 min |
| 23:10.14 | Captain-Obvious | they have it in cellphones |
| 23:10.16 | Harlekin | -) |
| 23:10.18 | Captain-Obvious | lol |
| 23:10.19 | Harlekin | at least i got |
| 23:10.30 | gb2 | kergoth: i'm horribly short on time, but so are you. |
| 23:10.32 | Harlekin | Captain-Obvious: while playing quake3 on a 12" one |
| 23:10.32 | kergoth | you got eye cancer in 20 min? |
| 23:10.35 | kergoth | gb2: hehe, indeed |
| 23:10.41 | gb2 | kergoth: i had the thought though.. |
| 23:10.46 | kergoth | gb2: maybe just something to add to the TODO until "someone" feels like implementing |
| 23:10.48 | Captain-Obvious | "The computer display produces 3-D images by sending a slightly different image to the right eye and the left eye at once by bending them in different angles, according to Sharp. The special screen has applications in architecture, medicine, science and gaming." |
| 23:10.51 | Harlekin | kergoth: i really had problems after that 20 min session |
| 23:10.54 | Captain-Obvious | sounds cool... |
| 23:11.00 | Harlekin | that displays are for japanese only |
| 23:11.01 | Harlekin | i guess |
| 23:11.01 | gb2 | most tars won't have it, so it'll be a tool that'll need to be built. |
| 23:11.06 | Harlekin | other people get sick |
| 23:11.09 | gb2 | and i was reminded of my toolchain rant from before |
| 23:11.18 | kergoth | gb2: ah, yeah, we'll need 'tar' as a host build tool |
| 23:11.26 | kergoth | aka -buildarch |
| 23:11.41 | kergoth | but then, it wouldnt be bad to add gnu tar as an .oe for target anyway |
| 23:11.45 | kergoth | and a -buildarch .oe is like 3 lines |
| 23:11.47 | kergoth | heh |
| 23:11.51 | gb2 | heh |
| 23:12.14 | Captain-Obvious | Harlekin: " |
| 23:12.14 | Captain-Obvious | The Mebius PC-RD3D, billed by Sharp as the world's first 3-D laptop, goes on sale Oct. 27 in Japan and is planned for release later this year in the United States. |
| 23:12.14 | Captain-Obvious | " |
| 23:12.24 | gb2 | i still think it'll be useful to build the buildarch tools and install them into the oe build world in a standard way. |
| 23:12.41 | Harlekin | Captain-Obvious: yeah, i know that display in there |
| 23:12.47 | kergoth | gb2: standard way.. you mean using the packaging tools of the buil dmachine? apt-get/;rpm/whatever? |
| 23:12.52 | Harlekin | Captain-Obvious: i did not say the are not gonna sell it elsewhere |
| 23:13.09 | gb2 | kergoth: i'm likely not going to want to replace my system's tar for example. |
| 23:13.13 | kergoth | I dont like that because it adds too much knowledge of the build machine. this needs to stay completely build distro agnostic |
| 23:13.21 | kergoth | gb2: -buildarch packages install into staging and build a package |
| 23:13.25 | gb2 | where i == some random user |
| 23:13.26 | kergoth | gb2: tmp/staging/build/bin/ |
| 23:13.27 | Captain-Obvious | anyway |
| 23:13.29 | gb2 | ok |
| 23:13.41 | gb2 | that's what i meant, cool. |
| 23:13.43 | kergoth | so okay, we're arguing the same thing at each other |
| 23:13.51 | gb2 | as long as i can have my world where i build gcc and stuff in there too ;) |
| 23:14.33 | kergoth | heheh, just a matter of stuffing CROSSCOMPILEPATH in your conf/local.conf most likely |
| 23:14.48 | kergoth | something like that anywya |
| 23:14.58 | gb2 | will i be able to do paths in there relative to $OEDIR? |
| 23:15.24 | kergoth | CROSSCOMPILEPATH=${STAGING_DIR}/cross |
| 23:15.26 | kergoth | ta da |
| 23:15.30 | gb2 | aha! |
| 23:15.33 | kergoth | yeah, just reference vars in a normal .oe fashion |
| 23:15.37 | gb2 | good good |
| 23:24.43 | gb2 | will oemake properly pick those up as dependencies to be build/installed? |
| 23:24.57 | gb2 | s/build/built/ |
| 23:29.09 | kergoth | pick what up as dependencies? |
| 23:29.12 | kergoth | it obeys DEPENDS |
| 23:29.21 | kergoth | if something is in DEPENDS for a package, that gets built first |
| 23:33.07 | gb2 | hrmm |
| 23:37.12 | kergoth | hmm, make -e is dangerous even with MAKEFLAGS="" |
| 23:38.11 | kergoth | could lose necessary defines |
| 23:38.14 | kergoth | grr. |
| 23:39.35 | gb2 | I guess what i'm really thinking about with tools is best solved by a config system before building "Use existing tools?/Build own tools?" |
| 23:39.46 | kergoth | well, thats an easy one |
| 23:39.50 | kergoth | what you do is, add a new env var |
| 23:39.55 | ljp | make -j30 is also dangerous |
| 23:40.03 | kergoth | that allows you to tell oemake these things are already satisfied, dont try to build them |
| 23:40.06 | kergoth | hehe |
| 23:40.10 | kergoth | then you just say HAVETHINGS=tar |
| 23:40.14 | kergoth | and it knows you've arleady got it |
| 23:40.17 | gb2 | heh |
| 23:40.18 | kergoth | the config system would just set this |
| 23:40.42 | gb2 | yeah. |
| 23:48.03 | gb2 | tools definately seem they need more thought with regard to deps |
| 23:48.22 | kergoth | right now i'm using a different packagename |
| 23:48.24 | kergoth | ipkg-buidlarch |
| 23:48.25 | kergoth | vs ipkg |
| 23:48.39 | kergoth | but itd be nice to be able to express that better, and have the build system automatically switch ARCH to BUILD_ARCH |
| 23:48.42 | kergoth | and shit |
| 23:48.45 | kergoth | i.e. |
| 23:48.56 | kergoth | DEPENDS = ipkg (buildarch) |
| 23:48.58 | kergoth | or some shit |
| 23:49.48 | gb2 | yeah |
| 23:49.56 | gb2 | stuff like the compiler though.. |
| 23:50.31 | gb2 | which in your view of the world, would be passed in via HAVETHINGS= in the most common case. |
| 23:50.58 | gb2 | but in mine, would require building and installing before building the rest of the world |
| 23:53.11 | kergoth | yep |
| 23:54.56 | gb2 | i'll continue to think on it and see if i can come up with a way to really express toolchain build dependencies |
| 23:57.19 | *** join/#openembedded stigger (stigger@p060-n1-gos.nlc.net.au) |
