**** BEGIN LOGGING AT Sun Dec 30 02:59:57 2007 Dec 30 06:57:25 03pfalcon 07org.oe.dev * r6cde30c2... 10/ (6 files in 3 dirs): Dec 30 06:57:25 matchbox-applet-startup-monitor 0.1: Start from Xsession. Dec 30 06:57:25 * Apparently, OpenedHand design it to be started from session, not as a Dec 30 06:57:25 normal panel app via panel manager, and that can't be patched in easily. Dec 30 06:57:25 So, give up on that, and add Xsession.d file instead. Dec 30 09:34:12 good morning Dec 30 09:35:37 pfalcon: a small typo in your last matchbox-applet-startup-monitor commit ? Dec 30 09:36:19 pfalcon: - install -m 0644 ${WORKDIR}/mb-applet-startup-monitor.desktop ${D}${datadir}/applications/ Dec 30 09:36:32 pfalcon: + install -m 0755 ${WORKDIR}/10mb-applet-startup-monitor ${D}${sysconfdir}/X11/Xsession.d/ Dec 30 09:37:40 the last line ? 10mb- Dec 30 09:39:02 pfalcon: there is more... Dec 30 09:39:49 * ant posts here doubtful about yesterday's issues with mail server... Dec 30 09:42:03 pfalcon: http://www.pastebin.org/13437 Dec 30 09:46:14 morning all Dec 30 09:46:21 cyrilRomain: bonjour Dec 30 09:46:31 ant_: hey Dec 30 09:47:09 cyrilRomain: at 3 AM my c7x0 build did finish! Dec 30 09:47:36 ant_: yay =) Dec 30 09:47:45 ant_: =) Dec 30 09:47:57 cyrilRomain: today is a slow day and if young baby allows I'll attack Gentoo... Dec 30 09:48:04 ant_: sadly your logs didn't help Dec 30 09:48:13 DEBUG: set_password(): Entering main-loop with tty [/dev/tty1] Dec 30 09:48:13 New password: Dec 30 09:48:34 CoreDump: hey Dec 30 09:48:39 if it is just sitting there, then altboot doesn't do anything but wait for input, no run-away loop or anything Dec 30 09:48:49 CoreDump: at least the message was longer... Dec 30 09:49:11 CoreDump: (at the beginning is "wrong type") Dec 30 09:49:34 WARNING: Assignment of OUT_TTY failed (not a tty)! Dec 30 09:49:44 CoreDump: exactly Dec 30 09:49:48 (kernel thingy that can be seen on all 2.6 machines) Dec 30 09:50:10 it gets auto-fixed and the message can be ignored Dec 30 09:50:35 CoreDump: in october/november I think this logs weren't there Dec 30 09:51:10 I'll doctor an image for you w/o the /etc/altboot.rc/* scripts Dec 30 09:51:20 maybe a kernel module freezes the Z Dec 30 09:51:44 CoreDump: what's interesting I could not collect the altboot logs on ttyS0 Dec 30 09:52:03 yes, that is to be expected Dec 30 09:52:14 all output is forcibly redirected to /dev/tty1 Dec 30 09:52:17 CoreDump: kernel set at 9600 then second kernel/inittab switch to 115200 Dec 30 09:52:36 CoreDump: I have to patch my CONSOLE= Dec 30 09:52:42 the redirection is ugly but a must-have Dec 30 09:53:01 CoreDump: to have all at 9600 Dec 30 09:53:19 CoreDump: and have a full log Dec 30 09:53:41 well, if the port can so 115200, then the kernel should also be set at 115200, or not? Dec 30 09:54:18 CoreDump: I can try a unique getty on ttyS0... Dec 30 09:54:42 CoreDump: or I'm out because vc ??? Dec 30 09:55:01 CoreDump: (virtual consoles) Dec 30 09:55:14 I doubt you'd get the altboot text in ttyS0 in any case Dec 30 09:55:24 CoreDump: yeah.. Dec 30 09:55:40 can not be helped for now Dec 30 09:55:43 CoreDump: but must be possible, even on a printer... Dec 30 09:56:11 well, of course. I could redirect all messages to ttyS0 Dec 30 09:56:20 I least I _think_ that would work Dec 30 09:56:30 CoreDump: make a debug image please... Dec 30 09:56:33 but then you'd have no text on the screen =) Dec 30 09:56:40 CoreDump: np Dec 30 09:56:58 I shall do so later today and email you the address when you are not around Dec 30 09:57:21 CoreDump: getty -L ttyS0 9600 vt100 Dec 30 09:57:50 that's a job for inittab Dec 30 09:58:04 CoreDump: please set it this way Dec 30 09:58:09 I doubt that a getty would work at the altboot promt Dec 30 09:58:38 I shall try... Dec 30 09:59:30 CoreDump: in fact we miss /etc/ttys /etc/gettitab ....dunno Dec 30 09:59:38 *gettytab Dec 30 10:00:07 usually folks ain't big on a working ttyS0... Dec 30 10:00:21 CoreDump: it is a long time I don't work from serial console Dec 30 10:00:34 CoreDump: lately hacking some routers... Dec 30 10:00:41 I shall get me a cable in 2 or three weeks Dec 30 10:01:08 heh Dec 30 10:01:11 what is binconfig.bbclass for ? Dec 30 10:01:23 CoreDump: ...finally ! Dec 30 10:01:45 ant_: I mainly get it to play with u-boot =) Dec 30 10:02:17 the thing costs about 70 € at trisoft.de... Dec 30 10:03:16 CoreDump: u-boot will be our next step Dec 30 10:03:42 CoreDump: I annoyed hrw so lon... Dec 30 10:03:42 I wish....need to find someone willing to port the old patch to a recent u-boot version Dec 30 10:04:09 the old pdax version can already be compiled and made to work via OE Dec 30 10:04:21 CoreDump: I could not yet look at the code. Is resizing implemented? Dec 30 10:04:56 no idea, really Dec 30 10:05:09 CoreDump: yeah...there are some pdaxrom patches to import Dec 30 10:05:28 CoreDump: yesterday was buzz about overclock Dec 30 10:05:39 CoreDump: got it? Dec 30 10:05:41 just one + EABI fix stolen from OM Dec 30 10:05:50 I saw someone mentioning it Dec 30 10:05:58 CoreDump: 2 lines Dec 30 10:06:10 ? Dec 30 10:06:21 CoreDump: adds two lines Dec 30 10:06:35 err...what does? Dec 30 10:06:40 CoreDump: to the freq array Dec 30 10:06:51 he had a working patch? Dec 30 10:06:52 CoreDump: lemme find the url... Dec 30 10:07:30 I don't think we should ship that in the Angstrom kernels....users frying their CPUs and stuff Dec 30 10:07:53 CoreDump: me too, but a lot of users will ask for it Dec 30 10:08:10 it would have to have sane hard-coded limits Dec 30 10:08:40 CoreDump: freq is scaling Dec 30 10:08:59 pdax didn't give a shit about users (or they would have done the u-boot switch in a more responsible manner), we do... Dec 30 10:09:10 well, I do Dec 30 10:10:16 CoreDump: there was a nice song end '80 (Adiva - Respect) Dec 30 10:10:28 CoreDump: he he Dec 30 10:10:38 =) Dec 30 10:10:56 * CoreDump gets some more coffee Dec 30 10:13:11 CoreDump: http://comomo.shacknet.nu/files/pdaxrom/r121/patches/0302_pxa25x_clockup-r4.patch/view Dec 30 10:13:25 ph5: (thx) Dec 30 10:16:51 ant_: did you rebuild your kernel with this patch applied? Dec 30 11:36:56 CoreDump: not yet but other users did it successfully (see #angstrom) Dec 30 11:47:27 can anyone please try to compile lua or lua-native ? Dec 30 11:52:18 due to OE being statefull, I'm never 100% sure an issue depends on the tmp directory state or not (especially staging dir), and it is really time consuming to always build from an empty tmp dir to be _sure_ of the issue Dec 30 11:52:59 * cyrilRomain think next generation of packaged staging will improve that though Dec 30 11:55:10 * CoreDump bitbakes lua Dec 30 11:57:51 CoreDump: thanks for trying Dec 30 11:58:08 NOTE: package lua-5.0.2-r4: task do_build: completed Dec 30 11:58:10 NOTE: package lua-native-5.0.2: completed Dec 30 11:58:21 np Dec 30 11:58:42 CoreDump: ok thank you, you saved me at least one hour compilation :) Dec 30 11:58:54 * cyrilRomain now has to find what's wrong here Dec 30 11:59:08 you're welcome ;) Dec 30 12:07:30 I should hack up a bitbot !bitbake foo Dec 30 12:07:35 ;) Dec 30 12:12:06 bbl Dec 30 12:18:25 ok it was what I suspected: staging directory not being cleaned Dec 30 12:20:59 hmm, the staging directory is not structured to handle multiple version of a same recipe Dec 30 12:21:46 so it makes recipe bump harder Dec 30 12:25:18 cyrilRomain: You need packaged-staging for that Dec 30 12:25:38 RP: the new one I guess :à Dec 30 12:25:39 : Dec 30 12:25:41 :) Dec 30 12:26:12 RP: does it handles multiple version of a same recipe ? Dec 30 12:27:37 RP: ok at least file are supposed to be cleaned in the new packaged staging, which is _really_ a good thing :) Dec 30 12:28:32 hi all Dec 30 12:28:34 RP: so one can 'bitbake -c clean package_0.1 && bitbake package_0.2' and vice versa :) Dec 30 12:29:01 cyrilRomain: right, and bitbake package_0.2 should be able to remove 0.1 from staging Dec 30 12:29:06 RP: but if someone do 'bitbake package_0.1 && bitbake package_0.2' ? Dec 30 12:29:09 RP: ah ok Dec 30 12:29:24 cyrilRomain: Having two versions of something in staging doesn't really make sense Dec 30 12:30:04 RP: except if you build different distro for example Dec 30 12:30:24 cyrilRomain: but why would you want two versions in staging at once? Dec 30 12:30:36 and building different distros in one build tree is not supported Dec 30 12:30:49 RP: just to save more build time :) Dec 30 12:31:16 RP: but you're right in most case, having one version and some rebuild in rare case is enough :) Dec 30 12:31:20 Building different distros in the same tree is not supported. Dec 30 12:31:26 That is never going to work Dec 30 12:31:27 RP: ah ok Dec 30 12:31:56 RP: conceptually if it would handle multiple version of a recipe in staging it would be possible Dec 30 12:32:24 RP: but it is probably too much work for that (more important thing has to be done) Dec 30 12:33:03 cyrilRomain: No, there are many other reasons different distros in the same tree will not work Dec 30 12:33:43 cyrilRomain: Distros are about policy. Different policies mean they work different ways. Distros can add overrides which don't show up as a package version change for example Dec 30 12:34:42 RP: hmm I see Dec 30 12:36:21 RP: I tried to switch to the new package staging (with a fresh tmp dir) but it failed with freetype. And it is maybe to early to report bug so I switched back to the current one (and my old tmp) Dec 30 12:37:10 packaged staging is a work in progress Dec 30 12:37:36 I had hoped to do something over these holidays but its not looking likely now :( Dec 30 12:38:38 RP: you have apparently did most part of the work though :) Dec 30 12:39:16 cyrilRomain: We have a working proof of concept and it did generate working builds for me... Dec 30 12:39:42 ok thanks. I'll switch back again to packaged staging Dec 30 12:40:21 and will look more deeply about the freetype issue I have Dec 30 12:49:08 Hello, I have a question, not explicitly OE-related, but then again I have no idea where else to ask. Dec 30 12:49:11 It's about psplash Dec 30 12:49:58 How do I move focus (the foreground console thing) back to tty1 after psplash quits? Dec 30 12:50:27 depends on the device Dec 30 12:50:39 psplash-write "QUIT" moves back, and so does chvt 1, but applications still run on tty2, which is not deallocated Dec 30 12:51:08 X11 is configured to run on tty2, yes. That is intended Dec 30 12:51:26 I mean during startup Dec 30 12:51:48 all console boot-messages should be displayed on tty1 Dec 30 12:51:59 I run some ncurses based configuration stuff at every bootup, and psplash messes everything up. Dec 30 12:52:00 tty is usually X11 Dec 30 12:52:06 *tty2 Dec 30 12:52:09 Yeah Dec 30 12:52:17 ahh Dec 30 12:52:21 Before X11 is started, but after psplash quits Dec 30 12:53:10 So the focus is at tty2 (including all keystroke events), but, if chvt 1 is run, some things come up in tty1 Dec 30 12:53:24 Kind of messed up situation, to tell you the truth. :/ Dec 30 12:53:39 your ncuses app is buggy then Dec 30 12:53:53 it should stick to tty1 and _not_ the active tty Dec 30 12:54:03 Nah, made with "dialog" Dec 30 12:54:07 ah Dec 30 12:54:13 echo "blahh" > /dev/tty1 Dec 30 12:54:19 just redirect the output Dec 30 12:54:30 did that in the past as well Dec 30 12:54:58 might have to redirect the input as well Dec 30 12:55:06 That might do, as openvt might, but that's kind of hackish, um, won't work for everyone. Dec 30 12:55:17 Hmmmm Dec 30 12:55:27 tty1 is pretty much the default Dec 30 12:55:41 I dunno how to do that in any other way, sorry Dec 30 12:55:54 Thanks a lot for your help :) Dec 30 12:56:00 I'll see what I can do. Dec 30 12:56:00 np ;) Dec 30 12:56:25 gremlin[it]: hi Dec 30 12:56:59 hi univac Dec 30 12:58:59 B_Lizzard: It does sound like psplash isn't exitting nicely. All I can suggest is to look at something like X and see if psplash is missing something Dec 30 13:00:11 psplash is supposed to move focus back to the initial tty (tty1) when it exits, which I think it does when psplash-write "QUIT" is executed Dec 30 13:00:36 That is the intention, yes Dec 30 13:00:44 Thing is, it seems that startup scripts run whilst psplash was running are run under tty2 Dec 30 13:00:52 I'm not sure this is intended Dec 30 13:01:03 Or if it's a problem with my implementation Dec 30 13:01:42 To tell you the truth, I just too the /etc/init.d/rc and /etc/init.d/rcS scripts from angstrom and used those :) Dec 30 13:01:46 *took Dec 30 13:03:01 psplash runs great, and seems to exit when I run psplash-write "QUIT" in one of my other startup scripts, but the focus remains in tty2 Dec 30 13:05:43 "Your message to Angstrom-distro-devel awaits moderator approval"..so...disagreing with the Angstrom uber lord gets one into moderation. He should have at least the decency to inform one of the reasons. They same reasons would place koen and Paus S on moderation btw, but I somehow doubt they are... Dec 30 13:19:07 CoreDump: You shouldn't be under moderation :( Dec 30 13:19:37 then it's the entire list Dec 30 13:21:12 nm that. I should be used to it ;) Dec 30 13:22:57 CoreDump: I'm not an admin on ltg so I don't know whats going on... Dec 30 13:36:33 Oh for god's sakes psplash is getting on my nerves Dec 30 13:45:54 chvt *should* work in changing what is percieved as the foreground console, but it doesn't! Dec 30 13:47:01 a still running psplash would probably mess up tty1 Dec 30 13:47:16 I exit it Dec 30 13:47:51 With a startup script which runs before my ncurses stuff. Dec 30 13:50:47 The problem seems to be that things that are started up whilst psplash is running run under the tty psplash is runs in, and continue doing so even after psplash stops running. Dec 30 13:51:47 never ran into that one Dec 30 13:51:55 didn't mess with it in ages tho Dec 30 13:51:59 Yeah :/ Dec 30 14:11:48 Oh great, now the output comes up in tty1 but the input is still in tty2 Dec 30 14:11:58 Then the output goes back to tty2 Dec 30 14:12:00 :/ Dec 30 14:28:29 morning mickeyl Dec 30 14:29:17 B_Lizzard: If psplash hasn't exitted yet, the consoles will most likely mess up Dec 30 14:29:30 Thing is, it does exit! Dec 30 14:31:01 hi RP Dec 30 14:36:19 Finally made it work. Dec 30 15:29:39 re Dec 30 15:36:05 thesing: hi Dec 30 16:13:25 hi mrdata Dec 30 16:14:09 mrdata: I read somewhere that you work on u-boot on simpad. Dec 30 16:14:34 mrdata: does it already work? Dec 30 16:19:03 not true, Slims does work on u-boot and we have initial startpoint on sourceforge->slackpad Dec 30 16:20:06 thesing: and the current state is not usefull, but we have also redboot working Dec 30 16:20:54 thesing: i have made tests with collie-ts for simpad and i am not happy with the result Dec 30 16:22:26 thesing: the touchscreen response is not better then the current with ucb1x00-ts and the my own patches Dec 30 16:58:15 mrdata: is the response not better or worse? we should really create one ts-driver that work for both devices. Dec 30 17:06:42 thesing: the response is worse, continue touchscreen presses are not okay Dec 30 17:07:15 thesing: with ts_test i can not draw any long line Dec 30 17:08:15 thesing: but it is possible that i made a mistake while adding changes for simpad Dec 30 17:08:50 hi Dec 30 17:09:34 Crofton|home: Astrid no longer does free breakfast :-) Dec 30 17:10:02 that was the other problem gremlin[it] mentioned Dec 30 17:10:10 the breakfast was the best part Dec 30 17:10:23 well, except for the bar across the street from my room :) Dec 30 17:12:07 mhh what u saw on the bar ? Dec 30 17:12:49 well, about 4 in the morning it would close Dec 30 17:12:58 so there would be a bunch of noise in the street Dec 30 17:13:16 I saw them carry one guy out and leave him in the doorway passed out Dec 30 17:14:17 hahahah :) Dec 30 17:15:46 then he got up and talked to one of the staff and the pulled a liquor botle out and had a few shots Dec 30 17:16:31 * XorA|gone doesnt get on with continental breakfasts Dec 30 17:16:53 you want proper english breakfasts? Dec 30 17:17:10 * Crofton|home thinks XorA|gone does not get along with menus written in "foreign" Dec 30 17:17:10 Crofton|home: I need fried stuff Dec 30 17:40:19 CoreDump: :D Dec 30 17:59:50 * RP is with XorA|gone on breakfasts Dec 30 18:00:36 well, find a hotel in Brussels with fried breakfasts ..... Dec 30 18:01:17 Crofton: You don't find many hotels who manage to cook fried breakfasts properly, even in the UK :/ Dec 30 18:02:24 I had one at a B&B about 12 years ago Dec 30 18:02:38 I did buy some marmite the other day Dec 30 18:03:27 RP: you need a 10 quid a night B&B for a decent heart stopper special :-) Dec 30 18:03:28 The only thing I use marmite for is flavouring gravy and even that not often... Dec 30 18:03:48 * XorA|gone loves marmite Dec 30 18:04:18 XorA|gone: I save the heart stopping for the roadside greasy spoon bacon sandwiches when out on the bike ;-) Dec 30 18:05:27 * RP has a secret knowledge of good bacon sandwich vendors all along the UKs nice roads... Dec 30 18:08:43 RP: yummy :) Dec 30 18:22:16 RP: ok the freetype error I have has nothing to do with packaged staging 2 :) Dec 30 18:23:09 but I'm still looking for the side effect.. maybe because I have libtool-native in ASSUME_PROVIDED ? ... Dec 30 18:23:21 before I haven't Dec 30 18:23:43 cyrilRomain: I don't know what the problems is, let alone what the solution is ;-) Dec 30 18:26:17 the error is http://rafb.net/p/AkvvYX64.html Dec 30 18:27:20 I already found and tested a solution, but not sure it is really _the_ solution Dec 30 18:27:23 http://rafb.net/p/sSC5W092.html Dec 30 18:27:48 i586??? Dec 30 18:28:02 mwester: yes, building a qemu image :) Dec 30 18:28:25 libtool is missing. Dec 30 18:28:39 do you have the dependency, or is it in the wrong place? Dec 30 18:28:41 what is strange is that in my previous environment ${S}/builds/unix/${HOST_SYS}-libtool is correctly generated Dec 30 18:29:04 now it generates ${S}/builds/unix/libtool Dec 30 18:29:23 Sounds like its not OE's libtool being used Dec 30 18:29:43 RP: yes I added libtool-native to ASSUM_PROVIDED Dec 30 18:29:53 cyrilRomain: That would be the problem then Dec 30 18:30:01 yes I suspect that too Dec 30 18:30:30 * cyrilRomain starts another build without libtool-native in ASSUME_PROVIDED Dec 30 18:36:46 * cyrilRomain wonders why freetype needs to build libtool in its source directory Dec 30 18:37:35 cyrilRomain: Its the way libtool works Dec 30 18:38:01 RP: ok Dec 30 18:38:24 * cyrilRomain should better knows tools he build Dec 30 18:48:45 ok, I removed libtool-native from ASSUME_PROVIDED, still fails Dec 30 18:49:00 I also removed automake-native from ASSUME_PROVIDED, it then succeed Dec 30 18:49:16 sounds like ASSUME_PROVIDED has to be used with very caution Dec 30 18:49:56 I also noticed that if the assume provided tool version differ from the OE, some build may fails Dec 30 18:50:21 well, I have to leave now Dec 30 18:50:37 RP: thanks Dec 30 19:15:42 re Dec 30 19:18:05 hi flo_lap Dec 30 19:18:41 pb_: hey Dec 30 20:25:57 Looks like I missed something lately... what is the problem with the avr32 support? Dec 30 20:40:40 flo_lap: politics :( Dec 30 20:42:37 RP: hrmpf... where was this discussed? I missed that somehow... Dec 30 22:24:02 how do you usually solve such rpath issue ? http://rafb.net/p/ADWhm486.html Dec 30 22:24:07 RP, ping Dec 30 22:24:14 Crofton|home: pong Dec 30 22:24:29 did you do the git fetcher? Dec 30 22:24:30 from what I can read here http://wiki.debian.org/RpathIssue there are several way to 'fix' it Dec 30 22:24:43 Crofton|home: Er, yes Dec 30 22:24:49 http://bugs.openembedded.org/show_bug.cgi?id=3410 Dec 30 22:25:26 can I safely '--disable-rpath' ? in EXTRA_OECONF for instance ? Dec 30 22:25:38 Crofton|home: yes, git fetch over http is bust. Its a git problem :( Dec 30 22:25:39 cyrilRomain, not sure, worth a try Dec 30 22:25:48 can we add the work around? Dec 30 22:26:26 hmm, one of my occasional problems is not via http Dec 30 22:26:47 hmm Dec 30 22:26:57 both of my problems are git not http .... Dec 30 22:27:23 Crofton|home: With the same error message about HEAD? Dec 30 22:27:25 I need to rerun some builds and update comments Dec 30 22:27:29 not sure Dec 30 22:27:34 I will need to revisit the problem Dec 30 22:27:53 my old bugs got marked as dupes, not sure if that is 100% correct Dec 30 22:28:07 I'll try building again and update the comments on this report Dec 30 22:28:08 I'm hoping that specific problem goes away with a new version of git, then I'll just tell people to update Dec 30 22:28:14 heh Dec 30 22:28:18 when will that happen? Dec 30 22:28:32 I did complain on the git list about it Dec 30 22:28:41 Not sure what happened after that, I lost track Dec 30 22:28:56 ok, I think I will care more next year Dec 30 22:30:36 I'm running a buidl now that will do some git stuff Dec 30 22:46:26 morning Dec 30 22:46:50 hi rwhitby Dec 30 22:47:01 hi rwhitby Dec 30 22:49:09 RP, I just added my failure message to the bug Dec 30 22:50:20 Crofton|home: A totally different problem Dec 30 22:50:42 I was afraid of that ... Dec 30 22:50:49 Crofton|home: Its trying to pull revision "1" Dec 30 22:50:57 oh crap Dec 30 22:51:01 This will be becasue SRCREV = "1" ? Dec 30 22:51:09 stupid srv rev stuff Dec 30 22:51:15 I forgot all about that Dec 30 22:51:25 You probably want a SRCREV_pn-foobar = "${AUTOREV}" Dec 30 22:51:37 yeah Dec 30 22:51:44 * RP sometimes wishes he'd never got involved with SRCREV... Dec 30 22:51:54 we need a FAQ on that .... Dec 30 22:52:17 we do Dec 30 22:52:22 can I do this in the bb file? Dec 30 22:52:23 RP: heh Dec 30 22:52:32 or better still sanity.bbclass could error out... Dec 30 22:52:41 Crofton|home: You can do it anywhere Dec 30 22:54:28 SRCREV_linux-davinci ?= ${AUTOREV} Dec 30 22:54:28 ? Dec 30 22:57:17 SRCREV_pn-linux... Dec 30 22:57:35 pn-linux is literla? Dec 30 22:58:03 SRCREV_pn-linux-davinci ?= "${AUTOREV}" Dec 30 22:58:13 pn-${PN} is an override Dec 30 22:59:50 so "pn-" is magic? Dec 30 23:01:11 The pn- prefix to whatever the package name is does magic things due to it being an override (in OVERRIDES) Dec 30 23:01:17 thanks RP looks like I can build davinci kernels again Dec 30 23:01:38 CoreDump: Now add the FAQ entry :) Dec 30 23:02:09 * CoreDump hands this honor over to Crofton|home Dec 30 23:02:17 heh Dec 30 23:02:21 I hate writing Dec 30 23:02:35 that is why it took me so long to get a MS degree Dec 30 23:02:44 CoreDump: sorry :) Dec 30 23:02:50 ;) Dec 30 23:03:35 There is also alot of explanation/background I don't know Dec 30 23:03:42 and proper usage Dec 30 23:04:05 although ?= it in bb files so they build from latest is not bad Dec 30 23:04:30 I'd not recommend it in the .bb files in general Dec 30 23:04:44 The problem is AUTOREV means it hits the network. Dec 30 23:04:55 to check for latest rev Dec 30 23:04:59 Hitting the network by default isn't really allowed Dec 30 23:05:01 right Dec 30 23:05:24 specific revs it pulls once and keeps in sources? Dec 30 23:05:31 yes Dec 30 23:05:55 I remember the headaches doing builds when some servers went away for a few hours .... Dec 30 23:05:56 Nobody reads my posts to the dev mailing list, do they? ;-) Dec 30 23:06:05 I read, but I forget Dec 30 23:06:21 it is coming back now Dec 30 23:07:56 Its all a giant can of worms... Dec 30 23:08:07 yeah Dec 30 23:08:20 no one right behavior Dec 30 23:08:40 at least I am the only person I know of buidling davinci kernels Dec 30 23:09:03 when we get some real users we can set a default rev Dec 30 23:09:14 where is the OE FAW Dec 30 23:09:17 FAQ Dec 30 23:09:22 sounds reasonable... Dec 30 23:09:38 does parsing hit the network? Dec 30 23:09:54 if you set it to AUTOREV yes, if not, no Dec 30 23:09:59 what type of machine is this davinci? Dec 30 23:10:16 crap, so anyone that parses the linux-davinci bb fiel hits the network ... Dec 30 23:10:30 davinci-evm of lytech sff sdr board Dec 30 23:10:41 TI processor, ARM and C6$ DSP Dec 30 23:10:44 c^$ Dec 30 23:10:47 c64 Dec 30 23:10:48 Crofton|home: www.openembedded.org/wiki/OeFAQ Dec 30 23:11:09 even the FAQ needs to be easier to find Dec 30 23:11:19 and I cannot edit Dec 30 23:11:26 Crofton|home: login Dec 30 23:11:32 can't Dec 30 23:11:38 and I can't create an accont ... Dec 30 23:11:40 Crofton|home: To find it I hit Documentation then scrolled to the faq... Dec 30 23:12:03 ah it is labled oeFAQ ... Dec 30 23:12:13 I was scanning for first letter F Dec 30 23:12:26 I'll email the list about wiki accounts Dec 30 23:12:55 Crofton|home: just click the Edit tab Dec 30 23:13:05 hmmm Dec 30 23:13:07 ok Dec 30 23:13:13 I was just able to edit it without logging in Dec 30 23:13:15 Crofton|home: no need to login (unless you really want an account and name logged) Dec 30 23:13:18 RP, link was bad :) Dec 30 23:13:36 Crofton|home: sorry :/ Dec 30 23:13:50 * RP should give up with computers for a few days Dec 30 23:13:58 * RP is supposed to be relaxing... Dec 30 23:14:07 heh Dec 30 23:14:15 I should go play half life some more Dec 30 23:15:12 why does parsing hit the network to check the rev? Dec 30 23:15:39 * Crofton|home shoudl read the list archives Dec 30 23:16:02 because it wants to find out what version the recipe provides Dec 30 23:17:08 RP: latest poodle kernel oopses: http://www.pastebin.org/13523 Dec 30 23:17:33 * hvontres|home is now re-building with LED driver disabled Dec 30 23:17:57 hvontres|home: I already know about this? Dec 30 23:18:45 RP: ohh.. current defconfig still has LED enabled tho Dec 30 23:19:09 hvontres|home: yes, I'd hoped to have it fixed by now Dec 30 23:19:27 RP: ok, np Dec 30 23:24:06 RP: www.slug-firmware.net - I've done five releases in five days, and I'm supposed to be relaxing too :-) Dec 30 23:24:51 rwhitby: Nice work :) Dec 30 23:25:19 Lots of things we said we'd do in the 2006 end of year status report, so it had to be done :-) Dec 30 23:25:32 rwhitby: I wish I could point to something as productive... Dec 30 23:25:39 * flo_lap too Dec 30 23:26:00 it was all just finishing stuff that had been at 99% for a long time Dec 30 23:26:17 (plus a lot of other people helping) Dec 30 23:26:45 I have a todo list covered in hundreds of little things... Dec 30 23:26:56 I'm slowly wiping out some of them I guess Dec 30 23:35:56 hvontres|home: I think the fix might be to s/LOCOMO_DEV(led_cdev->dev)/LOCOMO_DEV(led_cdev->dev->parent)/ in the leds-locomo.c file Dec 30 23:36:13 RP: FYI, latest 2.6.23 suspends fine from flash... testing root-on-SD next Dec 30 23:36:30 RP: right after scraping some Oak.... Dec 30 23:36:47 * hvontres|home should be working on a pot rack for the kitchen right now.... Dec 30 23:37:01 ah :) Dec 30 23:37:22 * hvontres|home has his to-do list and a honey-do list on top of that Dec 30 23:37:44 * RP was wonding what that was a euphemism for ;-) Dec 30 23:38:10 "Honey, could you....." Dec 30 23:46:33 RP: where should the suspendkey function (press short for ESC, hold for suspend and wakeup) of collie and poodle implemented? kernel or userspace (zaurusd?)? Dec 30 23:48:41 thesing: I think userspace would be nicer Dec 30 23:49:12 It already should work as wakeup, its just suspend thats the problem... Dec 30 23:51:01 Ok. I will look at the poodle source for the wakeup part. Dec 30 23:51:38 thesing: Its in the locomo keyboard driver, it should just work on collie? Dec 30 23:52:09 thesing: ah, it'll need to be in the power management code though Dec 30 23:52:23 There are probably bits you can reuse and bits you'll need to add Dec 30 23:55:02 Is zaurusd suitable for implementation of the suspend part? Dec 30 23:58:05 thesing: yes Dec 31 00:00:37 Ok. Last question: Should patches to arch/arm/common/sharpsl_pm.c go directly to you or to arm-linux? Dec 31 00:01:00 thesing: me at first or at least cc me Dec 31 00:02:05 Thank you. Dec 31 00:07:29 thesing: thank you, its nice to see someone giving collie some care! :) Dec 31 00:09:26 At least I want to finish the work I started. It would have been a waste of time otherwise. ;) Dec 31 00:09:38 yay make collie work, soon it be my only Z Dec 31 00:09:59 XorA|gone: A sad day :( Dec 31 00:10:36 RP: not sad, it was my first Z and will be my last :-D Dec 31 00:10:45 RP: and its actually physically a different collie Dec 31 00:11:00 good in the sense we can move on to better things :) Dec 31 00:11:09 hvontres|home: That is the correct fix the for the LED btw, will commit shortly and push upstream... Dec 31 00:11:30 XorA|gone: Didn't you have experience in sound drivers? You could write one for collie ;) Dec 31 00:11:44 thesing: which codec is it? Dec 31 00:11:50 codecs? Dec 31 00:12:07 XorA|gone: You just walked into that one :) Dec 31 00:13:13 XorA|gone: an ucb1200 compatible (buzzer) and and another. (will look) Dec 31 00:14:58 XorA|gone: TC35143 and pcm1741. Dec 31 00:15:34 there is also an M62332 involved, but I forgot what this is. Dec 31 00:15:53 Isn't that for the LCD? (bias control?) Dec 31 00:16:52 send PCM to everything and see what buzzes :-D Dec 31 00:17:33 datasheets for all this components are available. Dec 31 00:17:57 and the 2.4 kernel source! :) Dec 31 00:18:02 * RP hides Dec 31 00:18:09 the M62332 seems to be an DAC. Dec 31 00:18:43 hvontres|home: LED fix pushed Dec 31 00:21:16 with the sharp remote control collie would be a nice linux mp3player ;) Dec 31 00:22:09 does remote work on collie? Dec 31 00:22:38 not yet. But its realy easy implementable. Dec 31 00:23:06 * RP never got a zaurus remote... Dec 31 00:23:39 Still I don't often use it as a media player I guess :) Dec 31 00:23:57 just read values from adc. One could likely add support for it to the rc-patch for spitz. Dec 31 00:24:00 * RP hugs his squeezebox Dec 31 00:24:05 RP: there is the OE one sitting on my desk, I havent used it since helping Justin work out the protocol Dec 31 00:25:10 RP: what is a squeezebox? some nice linux toy? Dec 31 00:25:55 moma's got a squeezebox ... Dec 31 00:25:56 thesing: www.slimdevices.com, mine is an old version 1 Dec 31 00:56:45 RP: I want to build an initramfs that only loads a new kernel and executes it (via kexec) do I need a shell in the initramfs or can I can I write I program which executes kexec and put it to /sbin/init (or wherever it has to be) ? Dec 31 00:57:44 03pfalcon 07org.oe.dev * rf75a9fcd... 10/ (2 files in 2 dirs): matchbox-applet-startup-monitor 0.1: Fix reference to renamed file. Dec 31 00:57:49 03pfalcon 07org.oe.dev * r37765fd1... 10/ (3 files in 2 dirs): gpe-nmf: Fix SECTION to "gpe/multimedia". Dec 31 00:57:55 03pfalcon 07org.oe.dev * rf40fbad1... 10/ (4 files in 2 dirs): vlc-gpe: SECTION = "x11/multimedia" Dec 31 00:58:00 03pfalcon 07org.oe.dev * r66515f98... 10/ (1 classes/openmoko-base.bbclass): Dec 31 00:58:00 openmoko-base.bbclass: Use OM-2007.2. Dec 31 00:58:00 * http://svn.openmoko.org/trunk/src/target/OM-2007/ doesn't exist anymore. Dec 31 00:58:05 03pfalcon 07org.oe.dev * rfac66456... 10/ (6 files in 4 dirs): Dec 31 00:58:05 backsaver: Simple and efficient backlight saver daemon. Dec 31 00:58:05 * Here comes the backsaver, ruthless ipaq-sleep killer. Dec 31 00:58:05 * Work in progress. Dec 31 00:58:10 03koen 07org.oe.dev * rba071689... 10/ (1 packages/hwdata/hwdata_0.191.bb): hwdata: fix do_install Dec 31 00:58:16 thesing: You should be able to write a program which does that Dec 31 00:58:17 03koen 07org.oe.dev * r13782a89... 10/ (1 packages/gcc/gcc-package.inc): gcc-package.inc: fix the powerpc 'nof' problem for native gcc as well Dec 31 00:58:24 03mickeyl 07org.oe.dev * rc7dad9e7... 10/ (4 files in 2 dirs): vte add 0.16.9, refactor into .inc file Dec 31 00:58:31 03mickeyl 07org.oe.dev * r9de858c0... 10/ (1 packages/openmoko2/openmoko-terminal2_svn.bb): openmoko-terminal2 svn now needs vala Dec 31 00:58:50 03mickeyl 07org.oe.dev * r1887e059... 10/ (1 packages/openmoko2/openmoko-terminal2_svn.bb): openmoko-terminal2 svn the correct dependency is vala-native, not vala Dec 31 00:59:18 03koen 07org.oe.dev * rb8146248... 10/ (5 files in 3 dirs): webkit-gtk: prepare for autotools support: http://www.unpluggable.com/?p=156 Dec 31 00:59:36 03rpurdie 07org.oe.dev * r50afd6b8... 10/ (3 files in 2 dirs): Add gcc-native-3.4.4 as a possibility to help with qemu issues Dec 31 00:59:47 03rpurdie 07org.oe.dev * refe97ae3... 10/ (1 packages/qemu/qemu-gcc3-check.inc): qemu: Fix .inc file to expand PATH and EXTRA_OECONF and check for gcc-3.4.4 as a priority Dec 31 00:59:58 03rpurdie 07org.oe.dev * r190a3601... 10/ (1 packages/linux/linux-rp_2.6.23.bb): linux-rp-2.6.23: Apply a couple of poodle fixes for sound and serial issues Dec 31 01:00:09 03koen 07org.oe.dev * rb86387e4... 10/ (4 files in 3 dirs): Dec 31 01:00:09 distcc: update avahi patch and initscript Dec 31 01:00:09 * see http://0pointer.de/blog/projects/avahi-distcc.html for details Dec 31 01:00:23 03koen 07org.oe.dev * r80fc38ee... 10/ (7 files in 2 dirs): bluez: update to 3.24 for utils and libs and 0.15 for bluez-gnome Dec 31 01:00:32 03mwester 07org.oe.dev * rf13835dc... 10/ (1 packages/nslu2-binary-only/unslung-rootfs/README): Unslung: Updated README for the 6.10 release. Dec 31 01:00:40 03rwhitby 07org.oe.dev * rac100263... 10/ (4 files in 3 dirs): ixp4xx-npe: Added license file Dec 31 01:01:01 03rwhitby 07org.oe.dev * rd5e89232... 10/ (3 files in 3 dirs): unslung-rootfs: Updated README and NOTES Dec 31 01:01:33 03hrw 07org.oe.dev * rfec9ba8d... 10/ (3 files in 3 dirs): linux-rp 2.6.23: Added Linksys WCF11 to list of HostAP supported cards. Acked by upstream Dec 31 01:01:54 03rpurdie 07org.oe.dev * rfafce7c2... 10/ (1 packages/linux/linux-rp_2.6.23.bb): linux-rp-2.6.23: Fix uri typo Dec 31 01:02:04 03rpurdie 07org.oe.dev * rdfb5c206... 10/ (1 packages/linux/linux-rp_2.6.23.bb): linux-rp-2.6.23: Add LED fix for poodle Dec 31 01:02:15 03rpurdie 07org.oe.dev * ra9808e56... 10/ (1 packages/linux/linux-rp-2.6.23/defconfig-poodle): linux-rp-2.6.23: Enable sound for poodle Dec 31 01:02:26 03koen 07org.oe.angstrom-2007.12-stable * rdbfb231a... 10/ (1 BACKPORTS.txt): BACKPORTS.txt: add a rev for review Dec 31 01:02:40 03pfalcon 07org.oe.angstrom-2007.12-stable * ree59838b... 10/ (6 files in 3 dirs): Dec 31 01:02:40 applied changes from 968e6781dca5ee420f8b36434c88d54cafe77f35 Dec 31 01:02:40 through 6cde30c201c07026e4889c1df2852bb29eed037e Dec 31 01:02:40 matchbox-applet-startup-monitor 0.1: Start from Xsession. Dec 31 01:02:40 * Apparently, OpenedHand design it to be started from session, not as a Dec 31 01:02:40 normal panel app via panel manager, and that can't be patched in easily. Dec 31 01:02:44 So, give up on that, and add Xsession.d file instead. Dec 31 01:02:51 03pfalcon 07org.oe.angstrom-2007.12-stable * r2ae6c15a... 10/ (1 BACKPORTS.txt): BACKPORTS.txt: Note 6cde30c201c07026e4889c1df2852bb29eed037e Dec 31 01:03:01 03pfalcon 07org.oe.angstrom-2007.12-stable * r881010d2... 10/ (2 files in 2 dirs): Dec 31 01:03:01 applied changes from 6cde30c201c07026e4889c1df2852bb29eed037e Dec 31 01:03:01 through f75a9fcd060f03cadb8d0c99ad2649c5fb05293e Dec 31 01:03:01 matchbox-applet-startup-monitor 0.1: Fix reference to renamed file. Dec 31 01:03:08 03pfalcon 07org.oe.angstrom-2007.12-stable * rca3d963d... 10/ (1 BACKPORTS.txt): BACKPORTS.txt: Note f75a9fcd060f03cadb8d0c99ad2649c5fb05293e Dec 31 01:37:58 'night all Dec 31 01:38:17 RP: night. Dec 31 01:38:30 RP: before you go: one more question Dec 31 01:38:51 how to I get the packge name+version in a bb file? Dec 31 01:39:41 RP: night, and thanks for tinkering on poodle again :) Dec 31 01:39:58 * hvontres|home is waiting for the automerger to do it's thing Dec 31 02:03:38 good night. Dec 31 02:47:13 Five new NSLU2 firmware releases now available: http://article.gmane.org/gmane.comp.misc.nslu2.linux/20642 Dec 31 02:47:39 (three of which use OpenEmbedded) **** ENDING LOGGING AT Mon Dec 31 02:59:56 2007