**** BEGIN LOGGING AT Sat Oct 20 02:59:56 2007 Oct 20 03:22:36 night Oct 20 03:26:18 don't let the code bugs by^Hite Oct 20 03:26:29 hehe Oct 20 07:05:22 hey mickey|tw|shoppi Oct 20 07:05:32 mickey|tw|shoppi: don't forget to buy stinky tofu ;) Oct 20 07:10:51 Henryk: http://mkestner.blogspot.com/2007/10/moko-bindings.html Oct 20 07:18:00 morning all Oct 20 07:19:26 hey steliosk Oct 20 07:23:06 koen : working on moving [456]86 machine kernel to OE's kernel, so no more kittens get killed Oct 20 07:24:06 ...and also avoid comments about "how strange people are" :) Oct 20 07:28:11 It's been a week since I sent the angstrom lockdown file to the mailinglist for review and sofar only pH5 has looked at it Oct 20 07:28:32 which means everybody else thinks locking down is fluff and unnecessary Oct 20 07:28:46 so .dev isn't too fast moving or unstable Oct 20 07:28:57 (or people just like to complain but do nothing) Oct 20 07:41:58 I think you are misinterpreting some things but irc is not the best place to have these kind of discussions Oct 20 07:42:20 * steliosk goes back to work Oct 20 07:43:41 obviously the mailinglist is the wrong place as well Oct 20 07:43:52 * koen goes outside to shout people are idiots Oct 20 08:06:13 03koen 07org.oe.dev * rfb2c0449... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom 2007: activate lock down file, nobody has complained about errors in the file for the past week, or objected to using it. Oct 20 08:18:48 koen: hmm, thanks for the pointer. no point in doing a recipe for it, though Oct 20 08:40:22 Henryk: you boycot non-autotooled projects as well? Oct 20 08:42:57 koen: that's one point :) The other is that the posting explicitly states that every user should copy the binding to their projects and there would be no point in installing it globally. Oct 20 09:35:52 koen: you have to learn in the software world, no-one will ever comment/review a change until it breaks something, eventually you learn to just check changes in wait until people scream, then revert them for discussion :-) Oct 20 09:42:00 :) Oct 20 09:43:44 XorA|gone: look at the rightmost collumn if 'bitbake -s' Oct 20 09:44:05 bitbake bug :( Oct 20 09:46:32 revert bitbake :-D Oct 20 09:50:05 i'm a bit confused by {essential,minimal,bootstrap}-image ... what difference have them ? Oct 20 09:56:39 03koen 07org.oe.dev * r6826902a... 10/ (1 conf/distro/include/angstrom-2007-preferred-versions.inc): angstrom-2007-preferred-versions.inc: remove gcc, kernels and images Oct 20 09:56:45 03ifaistos 07org.oe.dev * r0d6023df... 10/ (4 files in 4 dirs): Oct 20 09:56:45 conf/machine/x86.conf : Switch x86 to use OE's default kernel Oct 20 09:56:45 Clean up machine config Oct 20 09:56:53 03ifaistos 07org.oe.dev * r02176e6b... 10/ (4 files in 4 dirs): conf/machine/i586-generic.conf : Switch i586-generic to use OE's default kernel Oct 20 09:57:00 03ifaistos 07org.oe.dev * r1fb08a07... 10/ (4 files in 4 dirs): conf/machine/i686-generic.conf : Switch i686-generic to use OE's default kernel Oct 20 10:33:45 gm Oct 20 11:35:16 todays climate change update, forecast highs for the next five days Oct 20 11:35:30 21, 27, 26, 19, 18 C Oct 20 11:36:36 Crofton in itlay too ... 8 minimum ... 12 max temperature !!! Oct 20 11:36:57 high or low compared to normal Oct 20 11:37:09 we shoudl being seeing highs aroun 18-9 atm Oct 20 11:38:17 mhhh quile low ... but standard for that month Oct 20 11:39:23 it has been much warmer then normal for the past few weeks Oct 20 11:39:31 prbably be really cold this winter :) Oct 20 13:17:33 good morning Oct 20 13:19:20 what's the proper way to override a PREFERRED_VERSION in the angstrom .conf file? Oct 20 13:20:12 I need to use a more recent version of wpa_supplicant Oct 20 13:20:54 Tried doing it in machine.conf, but no joy Oct 20 13:21:07 hm.. openembedded.org having quite some troubles these days eh? Oct 20 13:22:17 angstrom-2007.1.conf has PREFERRED_VERSION_wpa-supplicant = "0.5.5", that line should be removed imho Oct 20 13:22:35 there is a PREFERRED_VERSION_wpa-supplicant ?= "0.4.9" in angstrom-2007-preferred-versions.inc Oct 20 13:24:13 pH5: yes, I've been removing that line by hand. But that just feels dirty :-) Oct 20 13:26:07 sakoman, koen: looks like cleaning up to me. we have to decide on whether angstrom should use 0.4.9 or 0.5.5 and whether the version selection should be in the .conf or the .inc file and then set it there with ?= Oct 20 13:27:03 pH5: I vote for using ?= BTW, I'm using "0.5.8" Oct 20 13:27:10 koen: btw, angstrom 2008.1 includes the angstrom 2007.1 file and thus does the version lock-down, too. Oct 20 13:27:17 how can I fix that? Oct 20 13:28:30 pH5: probably by making 2008 not including 2007 Oct 20 13:28:52 and we need to add a wpa-supplicant-nodbus 0.5.8 as well Oct 20 13:29:29 koen: so I just copy the rest of 2007.1 into 2008.1 and remove the require? Oct 20 13:30:13 yes Oct 20 13:30:27 "the rest" being "the bits we need" Oct 20 13:32:27 So for now I'll just keep editing angstrom.conf to make wpa ?= :-) Oct 20 13:33:01 that'd be everyting except preferred versions, I assume. maybe we should split that out in separate include files? (feed uris, preferred providers, distro specifics like tarball stash, extra rrecommends etc.) Oct 20 13:35:10 the feed uris should go away since we want to use angstrom-feed-config.bb anyway Oct 20 13:37:41 injected files are evil Oct 20 13:40:41 yay Oct 20 13:52:04 hey mickey|tw Oct 20 13:53:06 hi ho Oct 20 13:59:19 03likewise 07org.oe.dev * rf6d103c9... 10/ (3 files in 3 dirs): mpc8313e-rdb: Added support to build u-boot. Oct 20 14:02:29 koen, i suppose someone is working on openembedded.org ? Oct 20 14:04:45 gilligan_: dunno Oct 20 14:05:35 postgres installation seems to be somewhat broken.. already had problems with bugzilla the other day.. now postgres is on its knees Oct 20 14:06:15 bugzilla is hosted on another continent Oct 20 14:06:38 and with 'broken' you mean 'being DDOS'ed by spammers', right? Oct 20 14:18:01 well it is hard to get a connection yes Oct 20 14:18:11 i'd say it is about time that some anti-spam stuff is installed to protect the wiki Oct 20 14:18:33 some captcha stuff or whatever Oct 20 14:19:34 recaptcha? :) Oct 20 14:31:44 mickey|tw: what do you think about making the Xserver script(s) check fbwidth before setting DPI? Oct 20 14:32:00 mickey|tw: the clamshell zaurusses and neos can run in qvga as well Oct 20 14:43:02 sakoman, your overlay goes directly on top of .dev? Oct 20 15:03:11 Crofton: It's a bb collection, so I guess it is more "along side" Oct 20 15:03:50 Crofton: if you look at the profile file you'll see how I have things set up Oct 20 15:04:35 morning Oct 20 15:15:55 ah Oct 20 15:15:56 ok Oct 20 15:16:24 I am interested in how you work, because you seem to have a good solution for working in || with .dev Oct 20 15:16:37 and we can easily check your local diffs Oct 20 15:17:03 as opposed to poky, where it is "harder" to locate the diffs Oct 20 15:17:23 but for today, I am a kernel hacker :) Oct 20 15:26:35 Did you finish the thesis, Crofton|work_ ? Oct 20 15:26:47 close Oct 20 15:26:54 doing edits now Oct 20 15:27:04 working on stuff for a conference in two weeks now Oct 20 15:27:05 Cool! Oct 20 15:27:20 defend in mid/late November Oct 20 15:27:28 hmm need to get that scheduled Oct 20 15:27:33 You're done then, once you turn in the completed thesis? Oct 20 15:27:52 there is an oral presentation known as the defense Oct 20 15:28:06 then they pick on you and try to get you to write some more Oct 20 15:28:16 I left a section for them to pick on :) Oct 20 15:28:30 hehe! "defensive writing" :) Oct 20 15:28:35 yeah Oct 20 15:28:47 and I didn't feel like dealing with it at the time Oct 20 15:29:25 Crofton: Using multiple collections is working out pretty well Oct 20 15:30:05 good Oct 20 15:30:16 Crofton: The only change I need to make in OE is making wpa-supplicant ?= in angstrom distro.conf Oct 20 15:30:17 it is easy for me to check on your local changes to .dev Oct 20 15:30:42 I something abbout that Oct 20 15:30:57 I have a tgz there to make it easy for scruggs to grab the whole mess Oct 20 15:32:15 when I get some time, I'll go through what you have and see what I can commit Oct 20 15:32:53 but there is conferenece Nov 4-8 I need to get ready for Oct 20 15:33:44 Crofton: no problem -- it will still be there :-) Oct 20 15:33:49 yep Oct 20 15:34:18 that is what I like, you are happy, and there is no pressure on volunteer devs to push stuff for you :) Oct 20 15:34:49 Crofton: and it will be easy for me to remove stuff from my collection as you add the changes to OE Oct 20 15:35:08 So everybody is happy :-) Oct 20 15:35:36 hopefully Oct 20 15:35:53 poky maintains a svn repo of everything Oct 20 15:36:01 well everything we want Oct 20 15:36:05 er they want from OE Oct 20 15:36:24 and we can freely move stuff back and forth, but there is not a good way to find local changes Oct 20 15:36:38 Although there is the poky commits list now, which helps Oct 20 15:37:54 Crofton: http://www.openembedded.org/bonsai/view/scm/6/ Oct 20 15:38:11 even better :) Oct 20 15:45:57 * Crofton is not a fan of git, yet Oct 20 15:55:46 03koen 07org.oe.dev * r223e77ca... 10/ (1 conf/distro/include/sane-srcrevs.inc): sane-srcrevs: bump linu-bfin to get pata support= Oct 20 15:55:51 03koen 07org.oe.dev * rc9606a6d... 10/ (1 conf/distro/include/angstrom-2007-preferred-versions.inc): angstrom-2007-preferred-versions.inc: more fixes Oct 20 16:31:16 * Crofton feels like the mtd web page is out of date ... Oct 20 16:37:33 Crofton: and the stuff that isn't obsolete is incomprihensible Oct 20 16:37:44 heh Oct 20 16:37:48 Crofton: hop on #mtd or something similar Oct 20 16:38:12 * Crofton puts on his asbestos suit :) Oct 20 16:38:36 just like the arm-linux website Oct 20 16:39:05 RMK complains if you don't read it and he complains if you follow outdated advice on that site Oct 20 16:39:24 sometimes I think we are victims of our own success Oct 20 16:39:38 to much work and not enough people to maintain websites .. Oct 20 16:42:49 in the meantime, I can look at gta patches Oct 20 16:52:36 koen: i like that idea. would be sane to assume that the fb is set up in the way you want X. and it would remove a lot of hardcoded sizes in the Xserver script Oct 20 16:57:34 mickey|tw: stefan_schmidt's post about qvga on openmoko-kernel inspired me Oct 20 16:57:43 good Oct 20 16:57:46 * Crofton is annoyed at elitist devs who figure something out, then refuse to pass on the info! Oct 20 16:57:57 * koen switches to wifi Oct 20 17:01:28 Crofton: stop being so unelitist :-D Oct 20 17:01:44 :) Oct 20 17:02:18 gta01 QVGA patches are actually kicking about Oct 20 17:02:37 it's amazing how much of a difference it makes Oct 20 17:02:43 finger scrolling is _smoooooooooooooth_ Oct 20 17:02:51 mickey|away: 1/4 the amount of data, its what I would expect Oct 20 17:04:53 I am hoping the platform NAND driver works Oct 20 17:07:41 mickey|tw: qvga + CFS in 2.6.23 :) Oct 20 17:08:24 cfs? Oct 20 17:09:01 the scheduler i think Oct 20 17:09:30 ah, right Oct 20 17:09:36 completely fair scheduler Oct 20 17:09:50 yes, that goes for a little bit of more visible performance Oct 20 17:13:20 * CoreDump is happy to hear that the software stack is coming along so well Oct 20 17:14:18 CoreDump is back Oct 20 17:14:21 unbelievable Oct 20 17:14:23 hi Oct 20 17:14:40 heh I've been back for a few weeks now ;) heya Oct 20 17:14:52 i have been underway Oct 20 17:15:05 offtopic: IntelĀ®Core2Duo E6320 - E6850 or AMD Athlon X2 BE-2350 - X2 6000+ EE ? Oct 20 17:15:05 I'm usually online when europe is offline Oct 20 17:15:09 you can't answer a phone call with the neo right now Oct 20 17:15:25 koen: I couldn't even establish a call Oct 20 17:15:36 that works Oct 20 17:15:52 mickey|away: In dual core CPUs, Intel beats AMD hands down Oct 20 17:15:56 after killing gsmd a zillion times to recover from mokogsmd screwage Oct 20 17:16:03 CoreDump: k Oct 20 17:16:15 I flashed the qtopia image and it worked Oct 20 17:16:38 err yes, you should've said what you mean with "the software stack" ... Oct 20 17:16:40 At least that way I know that my hardware is not faulty / problematic Oct 20 17:17:28 I was talking about DISTRO + UI ;) Oct 20 17:18:11 mickey|tw: is jserv going to work on gsmd next week? Oct 20 17:18:12 And know I see that this is #oe, not #openmoko Oct 20 17:18:19 koen: i have no idea Oct 20 17:18:48 not seen him yet Oct 20 17:19:15 i plan to meet the gsmd team and tell them my ideas next week Oct 20 17:19:30 but first, more important global structure and platform discussion Oct 20 17:19:34 mickey|tw: did you read the latest mails on gsmd-devel (3 days ago) Oct 20 17:19:40 yes Oct 20 17:19:43 something about patches Oct 20 17:19:45 and merges Oct 20 17:19:51 right? Oct 20 17:19:57 yeah Oct 20 17:20:07 I just realized you responded in that thread Oct 20 17:20:13 yep Oct 20 17:20:29 it's hard for me to get excited before the what-i'm-afraid-is-the-root-of-everything gets cured Oct 20 17:20:49 what's the what-i'm-afraid-is-the-root-of-everything ? Oct 20 17:20:58 the boogey man! Oct 20 17:21:19 we're literally bombing the modem all the time. these old chips need synchronous treating or at least guaranteed minimum time between two consecutive commands Oct 20 17:21:43 iirc qtopia does that Oct 20 17:21:50 which is the reason why it works better on the hardwrae Oct 20 17:21:55 (guessing here) Oct 20 17:22:07 didn't have time to look at their code Oct 20 17:22:37 mickey|tw: http://www.openembedded.org/~koen/Xserver-adaptive.patch Oct 20 17:22:56 does the right thing on my vga neo Oct 20 17:23:48 i like that Oct 20 17:24:16 horribly inefficient when using busybox, but it gets the job done Oct 20 17:25:42 ya Oct 20 17:29:12 just need to drop the default rotation into a file and save most of that file worth of text Oct 20 17:29:53 03koen 07org.oe.dev * r2d6f350b... 10/ (3 files in 3 dirs): xserver-kdrive-common: adapt script to take framebuffer size into account when running on devices that can do both vga and qvga Oct 20 17:29:57 there we go Oct 20 17:30:21 * koen followed XorA|gone's advice about not waiting for people to test stuff :) Oct 20 17:30:59 mickey|tw: http://scap.linuxtogo.org/files/2784326fede9285e3965bcebf1328c4a.png <- better contrast, but it doesn't fit anymore Oct 20 17:31:07 koen: been in software industry since 97, never managed to get anyones interest in a change unless it breaks stuff Oct 20 17:31:44 XorA|gone: that's because you didn't put a git repo online for it Oct 20 17:31:46 koen: why does DPI for spitz/c7x0 not halve at half screen res :-D Oct 20 17:32:06 heh Oct 20 17:32:20 your giving the GTA0? too much lurvins :-) Oct 20 17:32:37 my a780 crapped out Oct 20 17:32:54 the h2200 doesn't boot because the handhelds kernel is broken Oct 20 17:33:03 and I ran out of power for the devboards :) Oct 20 17:33:26 time to linux-rp the 2200? Oct 20 17:34:02 * koen hopes so Oct 20 17:34:35 the hx4700 can be RP'ed quite easily, since it's using the same asic3 as htc-uni Oct 20 17:35:16 hx2000 is already in linux-rp Oct 20 17:35:25 different series Oct 20 17:35:45 h22xx is pxa255 with HAMCOP, hx2xxx is pxa270 with other asics Oct 20 17:36:03 HAMCOP is basically a stripped down s3c24xx, though Oct 20 17:36:33 heh heh Oct 20 17:37:58 XorA|gone: h2200 is a h5500 with a newer companion asic (HAMCOP vs SAMCOP, both s3c2xxx without arm core) Oct 20 17:38:42 koen: sounds like not too much hassle to bring into linux-rp if some C coder had one or two :-) Oct 20 17:39:10 h5550 patches were posted to LAK as part of google SoC Oct 20 17:39:29 by the same dude that maintains the power subsystem Oct 20 17:39:55 Anton Vontzov or something Oct 20 17:48:53 "19-Oct-2007 Netherlands Pending customs inspection." Oct 20 17:48:59 lazy customs bastards Oct 20 17:49:15 what did you order? Oct 20 17:49:31 analog camera from hong-kong Oct 20 17:50:07 hk -> amsterdam was less than a day Oct 20 17:50:20 yet another camera??? Oct 20 17:50:25 yeah Oct 20 17:50:26 dude, you have a collector's problem :) Oct 20 17:50:30 yeah Oct 20 17:50:50 I gained access to the darkroom at the uni :) Oct 20 17:51:07 now I can even make prints and enlargements myself Oct 20 17:52:51 nice3 Oct 20 17:53:39 koen is gonna look like an octopus with all those lenses sticking out Oct 20 17:54:01 * koen reads http://docs.blackfin.uclinux.org/doku.php?id=mplayer Oct 20 18:14:43 "optimized blackfin codecs since April 1, 2007" Oct 20 18:14:53 * koen feels an april fools joke coming up Oct 20 18:16:23 heh Oct 20 18:54:26 03likewise 07org.oe.dev * r411624e6... 10/ (3 files in 2 dirs): libzvbi: Moved from 0.2.24 to 0.2.25 release. Oct 20 19:06:21 * koen is slightly amused by the fact that the blackfin devboard has a better disk than his workstation Oct 20 19:29:45 * koen aarghs Oct 20 19:29:58 why are buildroot users too stupid to use patches? Oct 20 19:30:17 koen swap the disk then ;) Oct 20 19:30:28 * koen sees his hd fullup with stupid forks Oct 20 19:30:45 the blackfin people forked buildroot and then copied all sources they use into svn Oct 20 19:31:28 koen it could be worse at xandros they use patches but the patches are at wrong place and feel like ugly hacks Oct 20 19:31:43 in buildroot the patches are always in the wrong place Oct 20 19:32:12 I can't figure out where they stashed the uclibc config, so I'm going to build it and then see if I can locate .config Oct 20 19:32:14 * chouimat goes implement a custom qt model Oct 20 19:54:12 * Crofton curses nand flash Oct 20 19:57:04 * koen reads up on how to develop negatives in coffee Oct 20 19:58:59 * Crofton observes koen is not joking Oct 20 19:59:52 Crofton: http://translate.google.com/translate?u=http%3A%2F%2Fmedienfrech.wordpress.com%2F2007%2F09%2F30%2Ffilmentwicklung-in-kaffee%2F&langpair=de%7Cen&hl=en&ie=UTF8 Oct 20 20:00:14 plenty of hits for that in english also :) Oct 20 20:53:42 "morning" all Oct 20 20:54:17 gm Oct 20 20:55:25 * RP attempts to make staging layout changes work a little more... Oct 20 21:02:42 gm Oct 20 21:12:27 likewise, do you have a copy of the MV kernel for davinci handy? Oct 20 21:12:53 Crofton: sorry, no, at work I do. You can GIT it though. Oct 20 21:13:02 the MV one? Oct 20 21:13:28 the git version hosted by MV does not appear to have the NAND falsh driver .. Oct 20 21:14:12 Crofton: http://source.mvista.com/git/?p=linux-davinci-2.6.git;a=summary Oct 20 21:14:20 I have that one Oct 20 21:14:53 Crofton: ok, I have a pre-released Davinci kernel with the eval kit, I guess. Oct 20 21:14:58 yeah Oct 20 21:15:11 they have their kernel, but it is based of 2.6.10 I think Oct 20 21:15:19 Crofton: I can see if someone put the CD on our server. Lemme check. Oct 20 21:16:10 thanks Oct 20 21:16:27 Crofton: sorry, nothing in the TI directory. Oct 20 21:16:46 guess I have to wait until Monday to chat some people up Oct 20 21:16:51 Crofton: did you ever play with the DSP side of the chip? Oct 20 21:16:57 not yet Oct 20 21:17:02 A friend of mine is atm Oct 20 21:17:09 I trying to get Linux on it Oct 20 21:17:15 I have git booting Oct 20 21:17:24 and I can mount root over NFS! Oct 20 21:18:08 Crofton: cool, I just got that far with my PowerPC board :-) It's very neat not to have the flash the board on every test cycle :-) Oct 20 21:18:19 yes Oct 20 21:18:40 but I want to flash the board so I can run it without the NFS server Oct 20 21:18:52 Crofton: sure, me too someday :-) Oct 20 21:19:11 but I understand, you need to get the NAND Flash working Oct 20 22:27:05 * * OE Bug 3180 has been created by joshua(AT)willowisp.net Oct 20 22:27:07 * * bitbake gtk+ fails on 20071020 (and back at least a week) due to missing DEPENDS Oct 20 22:27:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3180 Oct 20 22:56:04 * * OE Bug has been RESOLVED by clement.hermann(AT)free.fr Oct 20 22:56:06 * *  usb-gadget is broken ( at least in card reader and file-backed mode) Oct 20 22:56:08 * * http://bugs.openembedded.org/show_bug.cgi?id= Oct 20 23:18:41 phew. Got Ruby *and* the cairo + gtk bindings done. Oct 21 00:32:41 I'm trying to get started with OE and I have mtn 0.36 and it complains: Oct 21 00:32:41 paulproteus@asheesh:~/openembedded $ mtn --version Oct 21 00:32:41 monotone 0.36 (base revision: e4bc808d89e029ce623f9e8f2b10c84006b83fb5) Oct 21 00:32:52 mtn: misuse: database /home/paulproteus/openembedded/OE-this-is-for-mtn-0.30.mtn is laid out according to an old schema Oct 21 00:32:52 mtn: misuse: try 'mtn db migrate' to upgrade Oct 21 00:32:52 mtn: misuse: (this is irreversible; you may want to make a backup copy first) Oct 21 00:33:58 It seems that "mtn --db=OE-this-is-for-mtn-0.30.mtn db migrate " may fix it. (-: Oct 21 00:41:55 migration should work. my debian unstable has 0.36, too Oct 21 00:42:44 (tho I recall my repo was 0.32 not 0.30 ? ) Oct 21 00:45:41 kergoth, Did you wget OE.mtn or instead OE-this-is-for-mtn-0.30.mtn ? Oct 21 00:46:41 OE.mtn Oct 21 00:47:06 did not realize there was a 0.30 version, too Oct 21 00:47:33 I'm reading http://www.openembedded.org/user-manual which seems a little old. Oct 21 00:49:04 Oh, look, http://www.openembedded.org/snapshots/OE-this-is-for-mtn-0.36.mtn.bz2 exists. Oct 21 00:49:08 That'll save me some time. Oct 21 00:49:41 hm, that page has some info I wish I had found by `man bitbake` or so. Oct 21 00:50:13 I'm just getting started; ph5 pointed me there. Oct 21 01:32:33 I guess now I need to set up my cross-compile toolchain, now that I have my OE snapshot checked out. **** ENDING LOGGING AT Sun Oct 21 02:59:56 2007