**** BEGIN LOGGING AT Mon Sep 07 02:59:57 2009 Sep 07 04:58:59 mornig,....someone got a hint for a light-console-pop3-client to crosscompile with the OM-tootchain? tried to crosscompile libspopc "http://brouits.free.fr/libspopc/"...(wich works graet on my i386) ....without success x(wich works graet on my i386) ... Sep 07 05:23:00 freesmartphone.org: seba.dos1 framework * r32928fa5a562 /framework/subsystems/opimd/pimd_tasks.py: opimd: Tasks: fix Update Sep 07 07:08:53 07:26 < holtmann> http://www.freesmartphone.org/index.php/Implementations/dbus-hlid Sep 07 07:08:57 07:26 < holtmann> That is btw. another fuckup of FSO. Sep 07 07:08:59 07:26 < holtmann> The GSM mux with two libraries, one extra daemon and D-Bus between that and their GSM daemon is pure stupid. No wonder that this is slow. Sep 07 07:09:02 07:27 < denkenz> yah, but a full telephony stack is just too much for them to bite off Sep 07 07:09:05 07:28 < denkenz> Especially on the shit hardware they had Sep 07 07:13:30 uh oh Sep 07 07:14:08 http://www.freesmartphone.org/index.php/Implementations/dbus-hlid Sep 07 07:14:12 ups Sep 07 08:31:04 hi Ainulindale Sep 07 08:31:08 time for a quick query ? Sep 07 09:07:53 Ainulindale: r u around ? Sep 07 09:14:34 vegyraupe: hello there, welcome! Sep 07 09:15:09 hi rhkfin Sep 07 09:15:18 only here for a few minutes Sep 07 09:15:26 trying to get Ainulindale to respond :) Sep 07 09:15:40 ok :) Sep 07 09:21:47 * vegyraupe seems to be unlucky today :( Sep 07 09:21:58 anyhow later :) Sep 07 10:46:11 *sigh* I feel too stupid for git Sep 07 10:51:04 хбпд 23 Sep 07 10:51:07 blindcoder: welcome to the club ;-) Sep 07 10:52:04 blindcoder: i had to use bzr today, was surprised by it, can't see why can anybody prefer that to git. Sep 07 10:52:28 I still love svn Sep 07 10:52:44 blindcoder: that means you don't understand DVCS :-/ Sep 07 10:52:50 PaulFertser: I've been told to 'git add dir' 'git comit dir' and 'git send-email HEAD^..HEAD' Sep 07 10:53:05 and the last one just gives me an error and the help text Sep 07 10:53:45 dvcs --> you can use it offline, and is *FAST* Sep 07 10:53:59 also, I couldn't find a howto on the shr-project.org site and not on google :( Sep 07 10:54:13 blindcoder: tried man git-send-email? Sep 07 10:54:22 blindcoder: what exactly the error? Sep 07 10:54:36 blindcoder@fortuna:~/SHR/shr-unstable/openembedded/recipes$ git send-email HEAD^..HEAD Sep 07 10:54:39 Skipping HEAD^..HEAD - not found. Sep 07 10:54:41 No patch files specified! Sep 07 10:55:38 after that, the help tetx Sep 07 10:55:43 blindcoder: you're supposed to use "git format-patch HEAD^" first Sep 07 10:55:54 grmpf Sep 07 10:55:57 blindcoder: it'll give you the patch, you'll send it with git send-email or other way you like. Sep 07 10:56:06 * blindcoder mumbles something about svn diff Sep 07 10:56:14 blindcoder: there's git diff as well. Sep 07 10:56:26 blindcoder: but svn is not nearly as convenient to send patches by mail. Sep 07 10:58:26 PaulFertser: if I get it right git-send-email is a script that sends files by mail Sep 07 10:58:54 but I don't want to discuss versioning system semantics, I want to send paches to shr-dev list Sep 07 10:59:41 let me put it this way: I can send the 20 patches per year in 10min with git, after I learned and practiced for 6h. With svn I probably would need 30min Sep 07 10:59:49 blindcoder: there is an excellent book published freely also online too Sep 07 11:00:06 http://progit.org/book/ Sep 07 11:00:29 DocScrutinizer: git has numerous advantages over svn. I used cvs and svn but switched to git. Sep 07 11:00:31 ok, QED Sep 07 11:00:32 khiraly_bk: there's also excellent books with 500 pages on svn. but still, all I need it "how can I get the diff from my fs to the repository" Sep 07 11:01:10 and trying to get this information on the mailinglist over the timeframe of 1 week is frustrating Sep 07 11:01:25 blindcoder: why didn't you use git diff then? Because everybody'd laugh at you, technically it'd be the same nonsense as svn diff. Sep 07 11:01:44 blindcoder: you should have asked here on irc. Sep 07 11:02:03 PaulFertser: because the first answer was that git send-email HEAD^..HEAD Sep 07 11:02:09 PaulFertser: I did a few times :P Sep 07 11:02:20 and today I got an naswer \o/ Sep 07 11:02:50 blindcoder: man, you could have read man page, it takes 5 min. Sep 07 11:03:08 lol Sep 07 11:03:31 blindcoder: i was away over the weekend... Sep 07 11:04:35 man git is 750 lines which includes everything bu at a quick'n'dirty starting point Sep 07 11:04:42 it links to a 'tutorial.html' without a path Sep 07 11:05:03 blindcoder: power tools take time to master, that's life Sep 07 11:05:14 QED Sep 07 11:05:14 just "look at this tutorial[1]" Sep 07 11:05:24 and in the bottom "Fotonote 1: tutorial.html" Sep 07 11:05:32 fscking awesome, really Sep 07 11:05:35 blindcoder: I sended you a linkto a *book*. Which should be suffice for the beginners Sep 07 11:06:10 blindcoder: full ack Sep 07 11:06:37 khiraly_bk: currently waiting for the page to load Sep 07 11:09:34 blindcoder: did "git commit dir" work? your favorite editor opend and you entered a commit message? Sep 07 11:11:17 Heinervdm: yes Sep 07 11:12:07 hmm, then there should be a patch to send... Sep 07 11:13:32 it also[Bblindcoder@fortuna:~/SHR/shr-unstable/openembedded/recipes$ git log brickshooter Sep 07 11:13:35 commit 1f6b600b57a7046e508b58f5648daa495ab29dde Sep 07 11:13:39 git also shows me the commit Sep 07 11:13:58 author me, date half an hour ago Sep 07 11:14:04 ah you commited to brickshooter branch Sep 07 11:14:18 err... I committed the brickshooter directory, no? Sep 07 11:14:25 at least, that's what I wanted to do Sep 07 11:14:46 hmm Sep 07 11:15:47 brickshooteris a directory with 3 files, a .bb, a .desktop and a .patch Sep 07 11:16:02 I can git add brickshooter in the recipes dir, then git comit brickshooter in the recipes dir Sep 07 11:16:15 sorry for typos, connection is very slow here atm :/ Sep 07 11:16:58 u c, git greatly simplifies things especially for small tasks (like single apps) ;-P Sep 07 11:17:35 not that svn is so much better wrt that Sep 07 11:20:45 Guys, is glowfish X500 connector mini-usb compatible? Sep 07 11:39:45 max_posedon1 ? Sep 07 11:40:44 PaulFertser: ?? Sep 07 11:41:09 DocScrutinizer: it looks weird Sep 07 11:41:24 X500 == device? Sep 07 11:41:37 DocScrutinizer: yes Sep 07 11:41:53 glofiish X500 Sep 07 11:42:05 no idea Sep 07 11:42:16 heh Sep 07 11:43:28 *I* wonder if N900 has host mode Sep 07 11:46:20 it's incredible how Nokia fails to list all properties of their devices. MMS, 3D gmeter, usbhost? (they failed to mention that for N810 as well) Sep 07 11:47:01 I think they do it ~on purpose - just to make it hard to compare the devices... Sep 07 11:52:37 Ainulindale, yes Sep 07 11:55:40 max_posedon1: I have some issues with my cross-toolchain, I was hoping you could help me :-) Sep 07 11:55:52 max_posedon1: but I'm recompiling it properly so I'll bug you when I'm done with it Sep 07 11:55:57 I can try Sep 07 11:56:31 max_posedon1: apparently I didn't have binutils properly installed Sep 07 11:56:43 max_posedon1: and what was bugging me was timezonedata trying to run zic to install files Sep 07 11:56:50 and thus not being able to run it properly Sep 07 11:56:57 as it was arm stuff Sep 07 11:57:35 Ainulindale: when will you start addressing SHR users' concerns? :( Sep 07 11:57:41 Well Sep 07 11:57:52 When I'll have time to, and sadly it's not nearly now Sep 07 11:57:56 (h:1 release in a week) Sep 07 11:58:05 which concerns are you talking about though? Sep 07 11:59:15 PaulFertser? Sep 07 11:59:18 Ainulindale: you were asked repeatedely to use a sane shell by default, no matter if it's not truncated busybox or bash. And it really affects usability. Also the x.org driver is still not packaged and used by default despite it's stable for more than half a year. Sep 07 11:59:36 PaulFertser: I never heard anything about shell Sep 07 11:59:56 Ainulindale: probably you have DocScrutinizer in /ignore list? ;) Sep 07 11:59:56 And for the X.org driver I didn't even know it was stable Sep 07 12:00:20 PaulFertser: Did he ask anything to me? Did he create a ticket? Does he represent the whole SHR user base? Sep 07 12:00:25 Ainulindale: it should have been packaged in march even if unstable. Sep 07 12:01:02 Ainulindale: what is the problem with using less restricted busybox config by default? It was spoken about by many users repeatedly. Sep 07 12:01:10 PaulFertser: Well as you may have seen, there are other developers than me, I can't be everywhere, I'm trying to cope up with my work, and personal responsabilities Sep 07 12:01:26 PaulFertser: There's no problem with that IMHO it's just a matter of doing it Sep 07 12:01:36 Ainulindale: when i ask dos1 he says he can't do it, when i ask mrmoku, he's on vacation or slow etc etc. Sep 07 12:01:53 PaulFertser: well then you didn't ask before two weeks ago :-) Sep 07 12:02:02 Ainulindale: i know you have your new cool work... Sep 07 12:02:15 It's not that it's cool or anything, it's time consuming, that's the point Sep 07 12:02:16 Ainulindale: i was asking mrmoku for x.org driver for more than half a year. Sep 07 12:02:37 PaulFertser: Why don't you try to package it? Sep 07 12:03:29 Ainulindale: because i wasn't SHR user until recently and even now i'm focused on low-level gta01 stuff. I too can't be everywhere and do everything, especially given how lazy i am :( Sep 07 12:03:54 PaulFertser: well here's my point as well :-) Sep 07 12:04:01 These days I've been outdated and busy elsewhere Sep 07 12:04:12 Mostly because I received my taxes and have to figure out a way to survive with -20 € a month :-) Sep 07 12:04:21 PaulFertser: do like me, build yourself the shr images :) Sep 07 12:04:21 Ainulindale: but you have some leader skills so you might want to at least try to motivate others to fix things. Sep 07 12:04:35 PaulFertser: Yep it's just a matter of asking, I wasn't aware Sep 07 12:04:46 If people are dumb enough not to create a ticket and raise attention to that ticket, I can't do a thing Sep 07 12:04:59 (Not meaning you're dumb or anything) Sep 07 12:05:11 rtp: in fact i first built shr image before F9C in May and never used it. I was trying to help SHR folks to solve some build problems (and probably did it to some extent). Sep 07 12:05:42 PaulFertser: although I can't understand why noone packaged the driver already Sep 07 12:05:43 Ainulindale: busybox vs bash is a matter of "taste". you can't manage to make everyone happy imho Sep 07 12:05:51 Ainulindale: the problem is that SHR currently lacks OE-experienced active devs. Sep 07 12:05:58 PaulFertser: wrong Sep 07 12:06:01 PaulFertser: asking for /ignore as well? Are you representing anybody? ;-) Sep 07 12:06:05 PaulFertser: we have nytowl, the FSO team, mrmoku & me Sep 07 12:06:18 The latter is a lazy ass busy elsewhere Sep 07 12:06:27 The one before a lazy prolific ass in holidays :-) Sep 07 12:06:38 rtp: not taste, it's a matter of nonsense: why have very limited busybox install by default? Why not at least use sane config? Sep 07 12:06:42 DocScrutinizer: oh Sep 07 12:07:02 DocScrutinizer: it's just because you say how you hate busybox more often than anybody. Sep 07 12:07:17 PaulFertser: the "strange" config of busybox is a different problem. iirc the config comes from oe directly Sep 07 12:07:50 rtp: i know, it's exactly the reason. Sep 07 12:07:53 PaulFertser: you know how I managed to get there (ignore). When it starts about lazyness to be an excuse, you're close Sep 07 12:11:16 PaulFertser: next steps, insult, spreading lies about you, /ignore paul. I'm not talking about how *I* manage things Sep 07 12:42:37 max_posedon1: hmpf, same issue again Sep 07 12:43:03 max_posedon1: http://pastebin.com/m17ea6aa3 Sep 07 12:43:09 max_posedon1: if you see someting wrong here... Sep 07 12:43:13 I'd be glad to know Sep 07 12:55:47 max_posedon1: seems I needed proper env setup for cross-emerge Sep 07 12:55:54 It went through setting CHOST, CTARGET & CBUILD Sep 07 13:17:31 USE crossdev-wrappers Sep 07 13:19:56 oh... Sep 07 13:20:45 MARCH_TUNE is good Sep 07 13:21:24 Ainulindale, add smth there Sep 07 13:23:06 also LDFLAGS brings bugs now Sep 07 13:24:50 CFLAGS also should be simplier, or you will see problems Sep 07 13:25:09 well it seems here it worked properly Sep 07 13:25:18 max_posedon1: although I'm not sure about the behavior if I cross emerge glib Sep 07 13:25:25 (its hard explain from neo, touch me at evening) Sep 07 13:25:43 BUT mtune really must Sep 07 13:26:17 in other case you wont be able catch using wrong compiler Sep 07 13:42:29 Heinervdm: one more question: why are all my changes deleted when I run 'make update'? Sep 07 13:42:57 blindcoder: because it resetst the git dir Sep 07 13:43:27 so how can I keep my changes? Sep 07 13:43:42 nothing serious lost, but annoying Sep 07 13:44:34 i'm putting my own recieps in a different dir and added it in local.conf Sep 07 13:44:58 BBFILES += "${TOPDIR}/mypackages/*/*.bb" Sep 07 13:45:14 ah, okay. guess I'll do that, too Sep 07 13:45:42 for commiting and so you have to move them to the openembedded dir, of course Sep 07 13:46:25 sure Sep 07 13:46:59 but then the whole point of 'git commit' is moot and I can just git diff > file and use mutt :) Sep 07 13:50:10 so the last two times I upgraded shr-u (about 2 weeks ago, and the night before last) I lost GSM afterwards, and had to do an opkg install --force-reinstall fsoabyss in order to get things working again... Sep 07 13:50:16 is that a known problem? Sep 07 14:05:34 Ainulindale: develop instead of browsing facebook! :D Sep 07 14:06:07 I'm doing so Sep 07 14:10:47 Ainulindale: http://patchwork.dev.bearstech.com/project/SHR/list/ Sep 07 14:10:58 Ainulindale: there are few patches left to commit ;x Sep 07 14:12:04 blindcoder: heh git still is a great way to avoid using rsync, and diff ;-) Sep 07 14:12:50 dos1: Did you apply the russian translation? Sep 07 14:14:15 DocScrutinizer: rsync is how I download new version of brickshooter to neo :) Sep 07 14:14:24 DocScrutinizer: 2MB files over a 20kB line is no fun Sep 07 14:17:13 Ainulindale: i applied everything not OE related Sep 07 14:17:25 (me ands soltys) Sep 07 14:17:25 DocScrutinizer: back in the day(TM) before version control, I had just two directories for development: rock-linux-orig and rock-linux-blindy Sep 07 14:17:29 s/ands/and/ Sep 07 14:17:30 dos1 meant: (me and soltys) Sep 07 14:17:41 DocScrutinizer: for changes I'd just diff the two and send the patchfile to clifford :P Sep 07 14:18:06 Ainulindale: but please don't commit this one: http://patchwork.dev.bearstech.com/patch/89/ Sep 07 14:18:13 I read the comments Sep 07 14:18:19 DocScrutinizer: it was so nice and easy :P Sep 07 14:18:20 I won't, the patch is wrong Sep 07 14:18:30 dos1: couldn't you accept the patches then? Sep 07 14:18:37 I thought I gave you the rights to Sep 07 14:18:49 Ainulindale: you didn't informed me :x Sep 07 14:18:57 checking Sep 07 14:18:59 Errr I thought, I'm not sure Sep 07 14:19:20 Ainulindale: no, i can't accept patches Sep 07 14:19:30 Ok I'll fix that now Sep 07 14:19:31 Ainulindale: at least i don't see it in UI Sep 07 14:19:56 dos1: you should be able to now Sep 07 14:20:10 Please check Sep 07 14:21:51 Ainulindale: ok, i can Sep 07 14:21:53 Ainulindale: thanks :) Sep 07 14:22:09 You're welcome Sep 07 14:34:44 Ainulindale: Is there a problem with the trac site? I try to log in and I just get errors about the page not redirecting properly. Sep 07 14:35:23 It's a known trac bug Sep 07 14:35:39 ahh Sep 07 14:35:47 is there away around it? Sep 07 14:37:01 packagekit 0.5.2 released Sep 07 14:37:04 with my patch ;) Sep 07 14:38:55 Toaster`: recreate an account sadly Sep 07 14:39:04 I'll delete your current one Sep 07 14:39:07 ahh, ok Sep 07 14:39:29 Done Sep 07 14:40:26 Can't understand why pkg-config --libs glib-2.0 outputs the -L flag but not the -l Sep 07 14:40:35 Success! Thanks Ainulindale Sep 07 14:40:38 You're welcome Sep 07 14:41:03 Ainulindale: -L/opt/gnome2/lib -lglib-2.0 Sep 07 14:41:10 Ainulindale: that's my output from that command Sep 07 14:41:18 blindcoder: yep and here in my crossdev env Sep 07 14:41:22 I have -L/path only Sep 07 14:48:16 dos1: should i send reciepe for midori 0.1.8? Sep 07 14:48:56 Heinervdm: send ;) Sep 07 14:49:07 bump prefered version too? Sep 07 14:49:59 Heinervdm: Does your recipe have patches for getting rid of the "You are running as a root user" warning? Sep 07 14:50:27 yeah, that wastes valuable pixels :P Sep 07 14:50:29 Toaster`: that's droped somewhere after 0.1.2 Sep 07 14:50:35 ahh, awesome Sep 07 14:51:28 dos1: Any word if a prelim version of eve will be added to the feeds? I had though I'd seen talk of a recipe for it on the mailing list a little while back Sep 07 14:52:38 Toaster`: i tried to build it over night Sep 07 14:52:41 it failed Sep 07 14:52:52 ahh, damn Sep 07 14:53:38 dos1: where does it fail? Sep 07 14:53:51 dos1: for me it builds Sep 07 14:53:52 Heinervdm: +PREFERRED_VERSION_db ?= "4.3.29"file://ua-iphone.patch;patch=1 Sep 07 14:54:02 Heinervdm: is it really what you mean? Sep 07 14:54:03 :P Sep 07 14:54:14 hmm Sep 07 14:54:26 dos1: pico gone wild Sep 07 14:55:02 dos1: will you fix it? Sep 07 14:56:21 Heinervdm: how? :P Sep 07 14:56:29 Heinervdm: i don't have commit access to OE Sep 07 14:56:50 dos1: ok :) Sep 07 14:57:04 what does patch=1 do? auto-apply patch? Sep 07 14:57:33 blindcoder: 1 is the value for -p parameter Sep 07 14:58:21 Heinervdm: and if patch=x is not added it will not be applied? Sep 07 14:58:34 blindcoder: yes Sep 07 14:58:38 nice Sep 07 14:58:45 * blindcoder heads over to his brickshooter .bb file :) Sep 07 15:01:01 one less thing to do manually Sep 07 15:02:30 dos1: can you reject the wrong midori patch in patchwork? Sep 07 15:02:33 Heinervdm: ok Sep 07 15:02:49 Heinervdm: done Sep 07 15:05:11 max_posedon: I can't understand why pkg-config strips off -lglib-2.0 (for instance) Sep 07 15:08:48 hi all , has anyone experience with blown battery in gta01 ? Sep 07 15:17:48 err... blown? as in 'blown up'? Sep 07 15:22:47 i had my old gta01 laying around for 2 weeks, and i openend it to change the battery, but this battery is blown ( not exploded ) up Sep 07 15:23:48 the battery is now round about 30% thicker ... (:-(( Sep 07 15:26:38 pwgen, enquire on #openmoko -- I think someone there noted this problem some months back... Sep 07 15:29:13 hey hey!! wifi works! Sep 07 15:30:58 pwgen: like on every arbitrary electric device, you're supposed to remove bat for extended periods on non-usage Sep 07 15:32:40 pwgen: particularly on 01 it's easy you deplete bat even when device is off supposedly. And LiIon batteries really do not respond well to being depleted completely for any extended period Sep 07 15:33:51 : i think/thought overcharging can blow a Li-ion battery , not deep chargeing .. Sep 07 15:34:23 actually deep discharge is worse than a slight/mild overcharging Sep 07 15:34:34 .. now i see .. Sep 07 15:38:24 especially on overcharge when protective circuitry in bat cuts charging, the cell will "recover" automatically by simple self-discharge. Obviously this isn't applicable for deep discharge even when protective circuit breaks discharging Sep 07 15:44:42 pwgen: are you comfortable with soldering 0402 parts? Sep 07 15:45:44 pwgen: i'd really suggest to avoid using "inflated" battery. Sep 07 15:45:49 PaulFertser: is the Thermistor in bat an 0402? Sep 07 15:46:10 pwgen: either you change one 0402 R or i'll soon cook a patch that'll allow to charge nokia's batteries at 200mA. Sep 07 15:46:35 DocScrutinizer: no, i guess it's better to change one R on the board itself to be fully compatible with nokia's batteries. Sep 07 15:46:36 dos1: using midori, were you able to log into gmail? Sep 07 15:46:50 Toaster`: i wasn't using midori Sep 07 15:46:53 always links Sep 07 15:46:58 heh Sep 07 15:47:00 even on om2007.2 ;) Sep 07 15:47:02 ooh, I see. yes, also a good idea Sep 07 15:47:11 and with links i always could :D Sep 07 15:47:37 losing gta01 bat compatibility, but trading it for a whole world of cheap nokia spare Sep 07 15:47:38 : i removed this ballon .. and unsoldered the "electronic " maybe i will find som Li-ion cell and replace it .. Sep 07 15:47:44 ok. I'll give it a shot Sep 07 15:48:33 pwgen: cells of nokia BL-6C should fit nicely Sep 07 15:48:41 pwgen: take extreme care dealing with batteries, not recommended in general. Sep 07 15:48:49 mwester: Hey :D Sep 07 15:48:53 but is there a solution to prevent my spare battery to get the same destiny ? Sep 07 15:49:03 pwgen: take out when not used Sep 07 15:49:30 pwgen: and ask mickeyl to finally fix powering off the modem on gta01 ;) Sep 07 15:49:40 *G* Sep 07 15:50:31 PaulFertser: you always can do hard shutdown, which will never power down modem Sep 07 15:50:33 mwester: btw, i've finally got a gta01 and i'm going to actively support it, yesterday i was measuring suspend current and already found (with a little help from my friends) a bug :) So gta01 users are not left in vain. Sep 07 15:50:52 DocScrutinizer: sure, but the probability is lower when fso-abyss does the right thing. Sep 07 15:51:06 sure Sep 07 15:51:10 bbl Sep 07 15:51:36 Also the bug prevents restarting frameworkd at all on gta01. Sep 07 15:52:52 can you check whether http://pastebin.ca/1557284 does the trick? Sep 07 15:52:54 pwgen: so if you're brave you can take BL5C cells and exchange that. Or take a thermistor from gta01 and use it in nokia battery. Or just use some 75k resistor instead of 10k and that way you'll get full compatibility with nokia batteries but only partial (200mA max charging current) with gta01. Sep 07 15:54:04 mickeyl: a bit later yes. Thanks :) Sep 07 15:54:15 Or probably rtp will check. Sep 07 15:54:43 no hurry Sep 07 15:54:49 it's been broken for so long now ;) Sep 07 15:55:21 thx i will take a look. but i wanted to use this device in an unsupervised enviroment, where power can completly fail . so this is not a good idea .. Sep 07 15:56:13 hi mickeyl Sep 07 15:56:23 hi pwgen Sep 07 15:56:41 pwgen: how's life? any progress towards upstart? ;) Sep 07 15:58:39 now i have a idea how tho access the upstart-dbus endpoint, but did not tried it out. ( i have the usual problems getting fso/shr to compile ...) Sep 07 16:00:28 really? both org.oe.dev and shr/import seem to compile fine for me these days Sep 07 16:01:09 mickeyl: libfsobasic doesn't compile for me in shr/import Sep 07 16:01:25 how does it bail out? Sep 07 16:01:39 unfortunately the org.oe.dev merger didn't happen Sep 07 16:01:39 do_configure fails Sep 07 16:01:54 pastebin? Sep 07 16:02:22 mickeyl: have to build first Sep 07 16:03:24 dos1: just tried eve r41151 and it builds fine Sep 07 16:04:22 dos1: oh, i've seen, the webkit-efl reciepe is an old one... Sep 07 16:05:44 atm shr-import fails on pax-utils because of a wrong src_uri. Sep 07 16:07:25 mickeyl: http://shr.pastebin.com/m35bf525a Sep 07 16:08:33 well Sep 07 16:08:37 i wouldn't call that bailing out Sep 07 16:08:47 misc-vapi needs to be built beforehand Sep 07 16:11:12 mickeyl: libfsobasiscs depends on misc-vapi, so why isn't it build before? Sep 07 16:11:56 this i don't know Sep 07 16:12:02 hmm Sep 07 16:12:19 are you including fso-autorev.inc? Sep 07 16:12:53 mickeyl: don't know using standard shr enviroment Sep 07 16:15:39 no, it does not include fso-autorev.inc, but shr-autorev-unstable.inc contains nearly the same, expect for SRCREV_pn-vala-dbus-binding-tool-native = "6071da6ad645e1200eb2aee88d254ada348d361b" Sep 07 16:17:15 might want to check the diff between the recipes and classes in org.oe.dev and shr/import Sep 07 16:17:34 especially the vala and freesmartphone/* ones Sep 07 16:19:13 dos1: wsod is happening for me on gta01 sometimes, especially after it spends some considerable time in suspend. Sep 07 16:21:03 there are a lot of differences... Sep 07 16:29:16 Heinervdm: trying to build eve with new webkit now Sep 07 16:29:21 Heinervdm: (and midori too) Sep 07 16:29:36 dos1: ok :) Sep 07 16:30:47 also try arora, but it's Qt based Sep 07 16:31:26 dos1: ping Sep 07 16:31:34 khiraly1: pong Sep 07 16:31:43 will in future shr release this automatic ringtone disabled? Sep 07 16:31:45 tmzt: if you'll provide recipe - why not? :P Sep 07 16:31:46 in rules.yaml file Sep 07 16:31:57 TAsn: ping Sep 07 16:32:12 I dont want to change anything (ie. dont want to require any changes) in rules.yaml. I want to use the stock one Sep 07 16:32:23 khiraly1: well, i think we planned to move that sort of things to ologicd Sep 07 16:32:31 but dunno in what shape is ologicd at the moment Sep 07 16:32:54 for oe? I don't have one but I like it better than midori at least on desktop Sep 07 16:33:12 and midori was quite slow on arm the last time I tried it Sep 07 16:34:43 but if you implement PlaySound request in shr telephony app, then this rule can be removed. (thats the first step in the right direction I think) Sep 07 16:35:12 the whole audio layer is somewhat irritating Sep 07 16:35:39 I have a significant delay on incoming call. The audio starts abuot 1.5sec after the vibration Sep 07 16:35:40 the reason why the dialer is responsible for the ringtone is solely a performance one Sep 07 16:36:12 I'll try with the stock shr apps too Sep 07 16:36:13 a sec Sep 07 16:36:23 with more and more parts moving to C^wVala, this problem will vanish Sep 07 16:36:44 what i DO like, is how fast the screen dimming is Sep 07 16:36:55 when I touch the screen I have backlight immediatly Sep 07 16:37:01 before it was about 1-2sec Sep 07 16:37:10 (ie. on om2009) Sep 07 16:41:10 Weird, gentoo lacks ebuild for miredo :-O Sep 07 16:41:37 Heinervdm: Sep 07 16:41:42 Initialized empty Git repository in /home/shr/shr-unstable/downloads/git/gitorious.org.webkit-efl.webkit-efl.git/.git/ Sep 07 16:41:43 fatal: The remote end hung up unexpectedly Sep 07 16:41:45 fatal: early EOF Sep 07 16:41:46 fatal: index-pack failed Sep 07 16:41:48 :( Sep 07 16:42:12 dos1: no frameworkd nor ophonekitd running, reason : no socket in /var/run/dbus Sep 07 16:42:12 heh Sep 07 16:42:33 tag8tag: thanks for reporting! Sep 07 16:42:41 oh dos1: without ophonekitd running shr settings do not display the battery state Sep 07 16:42:43 probably other images have that too... so not syncing then Sep 07 16:42:50 khiraly1: wtf? Sep 07 16:42:56 khiraly1: it's using plain FSO in battery module Sep 07 16:42:57 it states: for battery display you need ophonekitd running *or* fso Sep 07 16:43:05 dos1: ok, I'll retry Sep 07 16:43:18 maybe yesterday frameworkd was in half crashy state Sep 07 16:43:26 dos1: lite img may have the same issue ? Sep 07 16:43:36 tag8tag: probably, test if you want ;D Sep 07 16:43:50 'k :p Sep 07 16:43:56 khiraly1: it's only using ophonekitd in GSM and Power modules Sep 07 16:44:05 Power settings: Sep 07 16:44:14 khiraly1: power. that's not battery ;P Sep 07 16:44:14 Coulnd't connect to FSO or ophonekitd Sep 07 16:44:24 but fso is running Sep 07 16:44:25 khiraly1: power is auto-dimming and auto-suspending Sep 07 16:44:33 and it's relying on ophonekitd usage interface Sep 07 16:44:54 aham. That maybe somehow related to my other problem: the phone does not suspend by itself Sep 07 16:45:02 I always need to suspend it manually Sep 07 16:45:17 yup Sep 07 16:45:23 ophonekitd is suspending Sep 07 16:45:31 in this one i still don't know why ;P Sep 07 16:45:47 (btw, where can I set, that power button suspends the phone immediatly and do not bring up enlightenment menu?) Sep 07 16:45:56 I tend to click by accident to the shutdown icon Sep 07 16:46:01 it is really counterproductive Sep 07 16:46:03 khiraly1: immediately - oeventsd Sep 07 16:46:10 khiraly1: to not bring syscon - illume wrench Sep 07 16:46:51 khiraly1: btw. how do you do that? O_o Sep 07 16:47:02 i never accidentaly clicked shutdown button ;P Sep 07 16:47:16 PaulFertser: mickeyl: so now we have @POFF for GTA01, but still not for GTA02 despite I suggested to have it there as well? :-/ Sep 07 16:47:32 DocScrutinizer: what makes you think so? Sep 07 16:47:37 DocScrutinizer: that was for both Sep 07 16:47:45 aah, ok Sep 07 16:47:46 dos1: I do want to suspend immediatly, so I just push the screen, and sometime I accidently push the power off button Sep 07 16:47:51 it is not *that* far from suspend Sep 07 16:48:20 doh Sep 07 16:48:23 yet again Sep 07 16:48:28 its f*cking annoying Sep 07 16:48:40 khiraly1: you're strange :P Sep 07 16:48:50 the screen dimmed, so to bring it back i pushed on the screen. Sep 07 16:48:54 mickeyl: with your commit d53fca8d933fcf4622efb344b9f8bd1ee1959a57 on framework GetWakeupTime throws an exception TypeError: 'NoneType' object is not iterable when rtcReadAlarm() returns None Sep 07 16:48:56 and I pushed the power button.... Sep 07 16:49:57 betheg: hmm... i think it works here Sep 07 16:50:03 betheg: and thanks for reminding Sep 07 16:50:25 mickeyl: what do you think about WakeupTimeChanged signal in odeviced? Sep 07 16:50:35 dos1: btw, the new image comes with new kernel too? Sep 07 16:50:42 or kms Sep 07 16:50:44 i already have code for that Sep 07 16:50:54 khiraly1: kms - not. new kernel - yup, as in repo Sep 07 16:51:20 * [Rui] hopes to have a working phone when opkg upgrade finishes :) Sep 07 16:51:51 ok finished rebooting Sep 07 16:53:01 dos1: wrench->system->power settings brings up shr settings Sep 07 16:53:06 betheg: mickeyl: i used to get that error too...but only until i had never used alarms after flashing... Sep 07 16:54:45 (18.46.10) dos1: khiraly1: to not bring syscon - illume wrench Sep 07 16:54:55 dos1: elaborate a bit please Sep 07 16:55:15 betheg: do you have pyrtc installed? Sep 07 16:55:27 khiraly1: wrench -> input -> key bindings Sep 07 16:55:29 ah nevermind Sep 07 16:55:33 let me try to reproduce Sep 07 16:55:44 khiraly1: Execute is power button Sep 07 16:56:03 heh, it does not fit on the screen Sep 07 16:56:09 I dont see the half of the images Sep 07 16:56:11 argh Sep 07 16:56:16 Heinervdm: configure failed Sep 07 16:56:16 dos1: wakeuptimechanged? sounds like a bad concept. AIUI framework should support concurrent wakeuptimes and handle them in a smart way. so basically each waketime is an object with a single owner and can't be changed by anybody else Sep 07 16:56:17 you're right, guys Sep 07 16:56:18 Heinervdm: http://tinderbox.openembedded.net/public/logs/task/2825020.txt Sep 07 16:56:21 let me fix that Sep 07 16:56:31 DocScrutinizer: that's what otimed do Sep 07 16:57:35 DocScrutinizer: well, shr-today has some function to display next RTC alarm Sep 07 16:57:49 hmm, nice Sep 07 16:57:59 DocScrutinizer: and i think listening to signal could be better than refreshing screen every n seconds :P Sep 07 16:58:27 dos1: ok, I deleted the keybindings. Now nothing happens Sep 07 16:58:40 ok, I'll look oeventsd config file Sep 07 16:58:49 hmm, no Sep 07 16:58:59 betheg: catholicon: can you import pyrtc ? Sep 07 16:59:12 ahh, so you are talking about a call to otimed with reply giving all waketime objects, and "changed" means an object vanished or was added, then? Sep 07 16:59:23 rtcReadAlarm() should never return None, but always a tuple Sep 07 16:59:49 aah Sep 07 16:59:59 indeed it returns none, when it fails Sep 07 17:00:30 mickey: after setting up alarm once, i have never got a failure...so i can't really test tht :( Sep 07 17:00:32 DocScrutinizer: no, i want signal in odeviced, which signalizes RTC alarm value in hardware changed Sep 07 17:00:39 <[Rui]> after opkg upgrade, with kernel and kernel modules, etc... I should boot twice, right? Sep 07 17:00:47 dos1: such a list-of-objects-changed signal generally is a good concept Sep 07 17:00:48 DocScrutinizer: nothing more, setting/resetting/other things should be done by atd or otimed Sep 07 17:01:03 with their own signals ;) Sep 07 17:01:39 dos1: seems to me fso still has a poor implementation here Sep 07 17:02:24 getting next RTC alarm is just a special case of getting *all* RTC alarms Sep 07 17:02:43 DocScrutinizer: dunno how FSO/FR handles alarms Sep 07 17:02:47 dos1, half pong :) Sep 07 17:03:06 btw: where is the interface to the pcf50633 alarm management? Sep 07 17:03:08 * TAsn wonders what happened to the TAsn that was here 24/7 Sep 07 17:03:17 dos1: but you know how to get that info from otimed? Sep 07 17:03:18 TAsn: how do we want to play ringtone in future? (khiraly1 ping) Sep 07 17:03:28 what do you mean? Sep 07 17:03:35 who's in charge of that? Sep 07 17:03:43 Is that what you meant? ^ Sep 07 17:03:46 dos1: DocScrutinizer and PaulFertser has some proposal. But definietly application side Sep 07 17:03:46 TAsn: now it's played by oeventsd Sep 07 17:03:59 ok, so who's in charge, that's what I thought. Sep 07 17:04:01 ie. not through fso Sep 07 17:04:32 as I said from the start, I think the same as DocScrutinizer and PaulFertser (if I got what they think correctly) Sep 07 17:04:33 this power button suspend should be in rules.yaml? Sep 07 17:04:35 fso is a framework Sep 07 17:04:44 I dont find it. I disabled the e keybinding for it Sep 07 17:04:47 it should allow me to easily play ringtones Sep 07 17:04:48 i.e Sep 07 17:04:52 "playringtone" Sep 07 17:05:00 but it should let me do it myself Sep 07 17:05:11 khiraly1: no, you have to write it yourself Sep 07 17:05:12 TAsn: it will not be removed (thats mickeyl told me), but it is not a recommended way Sep 07 17:05:30 because application can play it faster Sep 07 17:05:30 khiraly1: exactly Sep 07 17:05:36 khiraly1, I know it's not recommended Sep 07 17:05:45 TAsn: that's why i'm asking Sep 07 17:05:52 mickeyl: sorry i was smokking; yes i can import pyrtc, but you have found the problem already. Sep 07 17:05:53 TAsn: do we have any plans regarding that? Sep 07 17:05:53 khiraly1, really? I thought this is the only downside (I thought fso can do it better) Sep 07 17:06:04 dos1, *we* don't have any plans Sep 07 17:06:11 and *I* haven't thought about it :) Sep 07 17:06:13 yet. Sep 07 17:06:13 dos1: with RingTone in ruley.yaml, the problem is the user *CANT* change the ringtone file path. Sep 07 17:06:19 ok Sep 07 17:06:22 so it cant change ringtone on his phone Sep 07 17:06:27 khiraly1: WTF? Sep 07 17:06:37 khiraly1: he can by profiles Sep 07 17:06:42 khiraly1: but of course, i understand Sep 07 17:06:43 dos1, I think we should start a Sep 07 17:06:49 <[Rui]> dos1: after upgrade, after 1st boot, contacts crashed Sep 07 17:06:56 "general phone architecture discussion" Sep 07 17:07:01 khiraly1: user should be able to select different ringotnes for different contacts Sep 07 17:07:07 i.e what goes where Sep 07 17:07:10 and what does what Sep 07 17:07:15 as soon as mrmoku is back Sep 07 17:07:31 (which is conveniently after my test:) ) Sep 07 17:07:34 (19.06.37) dos1: khiraly1: he can by profiles <-- yeah thats the only way. He should edit the profile file and relaunch(?) the frameworkd to take effect Sep 07 17:07:44 khiraly1: WTF? Sep 07 17:07:50 khiraly1: FUD :P Sep 07 17:07:53 ;) Sep 07 17:07:59 TAsn: hmm, I think that's largely FSO domain, but anyway a good idea Sep 07 17:08:08 <[Rui]> dos1: ok, after all it was only because ophonekitd hadn't started up yet. Sep 07 17:08:11 /etc/freesmartphone/opreferences/conf/phone/default.yaml Sep 07 17:08:12 khiraly1: i even have script which changes ringtone after every call Sep 07 17:08:15 khiraly1: and it works ;P Sep 07 17:08:17 DocScrutinizer, referring to what? :) Sep 07 17:08:26 the discussion? Sep 07 17:08:31 ^^ there is written the ringtone file, and the path is hardcoded (/usr/share/sounds) in frameworkd source Sep 07 17:08:33 khiraly1: by dbus call, without editing any files Sep 07 17:08:38 i do not think it is the right way Sep 07 17:08:41 TAsn: discussion. or more like "whitepaper" Sep 07 17:08:47 khiraly1: and you have even gui for that Sep 07 17:08:49 khiraly1: in shr-settings Sep 07 17:08:52 a discussion followed by a whitepaper ;) Sep 07 17:08:59 dos1: sorry my mistake than Sep 07 17:09:05 btw dos1 Sep 07 17:09:06 freesmartphone.org: mickey framework * rc0ae955a0497 /framework/subsystems/odeviced/kernel26.py: odeviced: kernel26: do not bail out, if we can't query the RTC Sep 07 17:09:16 what's the max volume for ringtones? Sep 07 17:09:25 TAsn: dunno Sep 07 17:09:33 and it takes effect as soon as I change it, right? Sep 07 17:09:43 I really want to make my phone ring lowder Sep 07 17:09:43 TAsn: well. here not :D Sep 07 17:09:51 dos1, I have to reboot? Sep 07 17:10:05 TAsn: here ring-volume in profile doesn't have any influence on played ringtone at all :P Sep 07 17:10:07 I think volume is broken XD Sep 07 17:10:12 DocScrutinizer: ++ ;D Sep 07 17:10:18 I hate you guys :) Sep 07 17:10:19 damn. Sep 07 17:10:28 ;) Sep 07 17:10:34 as it was a concept with gstreamer. gst is dead Sep 07 17:10:36 so I need to change the stereout state file? Sep 07 17:10:41 TAsn: you can use alsactl to immediatly take effect Sep 07 17:10:52 see elmixer.py source file for an idea Sep 07 17:10:57 khiraly1, yeah, exactly what I asked :) Sep 07 17:11:06 volume is not implemented yet Sep 07 17:11:10 the infrastructure is missing Sep 07 17:11:15 khiraly1, no need, I know how state files work, it's just a minor tweak. Sep 07 17:11:18 and it doesn't look like it's coming soon Sep 07 17:11:19 mickeyl, i c. Sep 07 17:11:20 does it? Sep 07 17:11:23 so dos1 fooled us all. Sep 07 17:11:24 os.system( 'alsactl store 0 -f ' + newFile ) Sep 07 17:11:26 ^^ TAsn Sep 07 17:11:33 TAsn: ignore khiraly1. look for softvol control in ALSA ;-) Sep 07 17:11:37 mickeyl: so can i push WakeupTimeChanged? :) Sep 07 17:11:47 DocScrutinizer, I can just edit the state file. Sep 07 17:11:48 dos1: what's that? Sep 07 17:11:49 which I will. Sep 07 17:12:19 mickeyl: i already said :P that's dbus signal in odeviced/RTC Sep 07 17:12:20 ok, my audio understanding is messy, you can ignore me Sep 07 17:12:22 if we had fine granular audio paths, or at least a mapping to the volume control in the scenarios, i'd implement support for it Sep 07 17:12:31 TAsn: sure you can, but you don't want to do that in any app you write for *future* use Sep 07 17:12:36 dos1: didn't see that. Sep 07 17:12:42 mickeyl: :) Sep 07 17:12:47 <[Rui]> ok, calling seems fine... Sep 07 17:12:49 dos1: what's the usecase for it? Sep 07 17:12:49 DocScrutinizer, that is correct. Sep 07 17:12:54 mickeyl: just signal emited after calling SetWakeupTime Sep 07 17:12:55 dos1: http://paste.pocoo.org/show/138541/ Sep 07 17:13:00 dos1, do you remember what we decided about ophonekitd? Sep 07 17:13:11 I think that after my test I'll fix some issues with ophonekitd Sep 07 17:13:18 mickeyl: Sep 07 17:13:20 [18:57] DocScrutinizer: well, shr-today has some function to display next RTC alarm Sep 07 17:13:20 some I promised I'll do weeks ago, and some new ones :) Sep 07 17:13:21 [18:57] hmm, nice Sep 07 17:13:23 [18:57] DocScrutinizer: and i think listening to signal could be better than refreshing screen every n seconds :P Sep 07 17:13:24 dos1: can we incorporate into rules.yaml (even in commented out form). Just others could enable more easily Sep 07 17:13:55 TAsn: i don't remember such old things ;D Sep 07 17:14:00 like maybe ringtone handling. Sep 07 17:14:06 using amixer Sep 07 17:14:07 dos1: don't quote partially please! Sep 07 17:14:16 and with hackish volume control :) Sep 07 17:14:33 DocScrutinizer, but that's the best way to manipulate what other people say for your needs... Sep 07 17:14:42 :) Sep 07 17:14:44 TAsn: grab fsodeviced and add volume mapping for the scenarios Sep 07 17:15:00 err Sep 07 17:15:11 :) Sep 07 17:15:16 was just trying to get stuff moving ;) Sep 07 17:15:22 yeah Sep 07 17:15:34 a good thing would be volume mappings for the existing scenarios Sep 07 17:15:37 hehe :D Sep 07 17:15:39 i'll get you the URI of the new states Sep 07 17:15:44 hehe, anyhow I wish I even had time to git clone it. Sep 07 17:16:04 :) Sep 07 17:16:07 I'm too damn busy. ;| (until 14/9 when mrmoku|away is back and I'll be after my test) Sep 07 17:16:16 DocScrutinizer: well, i'm just too lazy to cut your "hmm, nice". all i wanted to quote were my text ;) Sep 07 17:16:21 dos1: where it comes rules.yaml? From freesmartphone? Then I should send my patch to mickeyl Sep 07 17:16:24 http://paste.pocoo.org/show/138541/ Sep 07 17:16:25 ^^ Sep 07 17:16:37 khiraly1: rules.yaml from SHR? it's in shr-themes.git Sep 07 17:16:39 * soltys wongers if his gta01 will survive opkg upgrade Sep 07 17:16:41 http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/freesmartphone/fso-alsa-data/om-gta02 Sep 07 17:16:54 TAsn: add softvol plugin to ringtone playback! then you have a controll in alsamixer called "ringtone" and you're done Sep 07 17:16:56 thanks. Sep 07 17:17:03 dos1: can we incorporate the above two rules to this file? (even in commented out form) Sep 07 17:17:04 ? Sep 07 17:17:07 http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/freesmartphone/fso-alsa-data/om-gta02/default Sep 07 17:17:09 contains the states Sep 07 17:17:24 http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/freesmartphone/fso-alsa-data/om-gta02/default.conf Sep 07 17:17:27 is the configuration Sep 07 17:17:48 khiraly1: just ping me today, tomorow, day after tomorrow... until i'll commit it :D Sep 07 17:17:50 default.conf should now contain aliases for interesting controls Sep 07 17:17:54 like volume Sep 07 17:17:59 dos1: thank you! Sep 07 17:18:29 mickeyl, thanks. Sep 07 17:18:48 I think the press 2s and the device will shut down is a nice thing. The other one (pressing the power button and it suspends) may be disputed to be default Sep 07 17:18:50 dos1: midori 0.1.8 needs newer webkit-gtk version, i'm having a reciepe for 1.1.4 Sep 07 17:18:51 np. this is still not ideal, but once we have that, we can at least support volume in our audio paths Sep 07 17:18:53 will hopefully look into it soon. Sep 07 17:19:04 as maybe others prefers enlightenment suspend screen Sep 07 17:19:15 dos1: and in the webkit-efl was the tag misspelled Sep 07 17:19:29 thinking about it... Sep 07 17:19:31 it should not be in OE Sep 07 17:19:35 but rather in the FSO git Sep 07 17:19:40 as other distros need that as well Sep 07 17:19:57 Heinervdm: btw, eve will evolve. one of an #e developer promised me if I create a gui for it he will implement it for us;-) Sep 07 17:21:06 anyone knows how the palm pre handles alsa audio? Sep 07 17:21:24 * DocScrutinizer wonders why everybody sticks to scenario concept while I constantly push in a better direction. It's really freaking me I have to come up with a complete new concept, tested and fully implemented, before anybody even starts to debate with me and might consider to switch eventually Sep 07 17:21:30 do they have multiple devices or fine granular scenarios or use a monolithic audio server? Sep 07 17:21:41 khiraly1: nice, but there is still the problem, that im getting git read-tree 0f6ab2ac9d5ba863984f0bd595ab9828e2f7bd0e failed with signal 128, output: fatal: failed to unpack tree object 0f6ab2ac9d5ba863984f0bd595ab9828e2f7bd0e when trying to get latest webkit-efl with bitbake Sep 07 17:22:07 DocScrutinizer: i'm afraid we will use the scenario concept as long as we have an alternative. Sep 07 17:22:19 the moment i have the paths, i will use them Sep 07 17:22:41 s/as long as we have/as long as we don't have/ Sep 07 17:22:57 or multiple devices Sep 07 17:23:01 i don't care, i take everything Sep 07 17:23:10 but it needs to be there :) Sep 07 17:23:29 mickeyl: it's more than simply using pathes. It's about a mutch smoother concept Sep 07 17:23:37 oh even smoother. ok Sep 07 17:23:42 mickeyl: so? :D what about commiting that signal? ;D Sep 07 17:23:44 where is it? Sep 07 17:24:00 dos1: ya, go for it, but update the specs beforehand Sep 07 17:24:09 TAsn: ignore khiraly1. look for softvol control in ALSA ;-) Sep 07 17:24:10 dos1: just pushed your patches ;) Sep 07 17:24:11 dos1: i'll catch up in fsodeviced Sep 07 17:24:12 mickeyl: ok Sep 07 17:24:20 mickeyl: ^ Sep 07 17:24:47 hmm Sep 07 17:24:55 ok, that will introduce even more latency Sep 07 17:25:00 and i doubt it works for GSM Sep 07 17:25:02 no Sep 07 17:25:16 yep, for gms it won't Sep 07 17:25:48 DocScrutinizer, :) Sep 07 17:25:48 but then GSM has a dedicated hw-control Sep 07 17:25:54 ciao for the meanwhile. Sep 07 17:26:59 mickeyl: so it won't need a personal softvol Sep 07 17:27:36 mickeyl: but ringtone might want, as you probably want to level ringtone and mp3 playback independently Sep 07 17:28:00 eventually Sep 07 17:28:10 on Om hardware i have my doubts Sep 07 17:28:23 as this involves additional processing Sep 07 17:28:31 dos1, btw, you are the new mrmoku|away right? :) Sep 07 17:28:32 yeah, if we don't start anytime, then eventually becomes never Sep 07 17:28:47 well Sep 07 17:28:48 TAsn: nooooooo, i won't build! Sep 07 17:28:50 ;D Sep 07 17:28:51 softvol might be nice Sep 07 17:28:58 i find the audio path problem more important Sep 07 17:29:05 so i'd rather start with this one Sep 07 17:29:06 dos1, I'll bug you about a package rebuild soon ;) Sep 07 17:29:18 but this requires the microscenarios Sep 07 17:29:25 and i don't see anyone coming up with these Sep 07 17:29:28 except perhaps you Sep 07 17:30:14 mickeyl: I babbled about alsa hooks to manage "micro-scenarios" for a long time now. I can paste an example in 5min. Sep 07 17:30:27 hmm Sep 07 17:30:35 i don't know whether i recall correctly Sep 07 17:30:37 does this require dmix? Sep 07 17:30:41 but i don't think this convinced me Sep 07 17:30:45 but I'm really facing overload on cranking up all this in a single step Sep 07 17:30:48 sounded like much more work Sep 07 17:31:00 as opposed to identifying a small set of diffs Sep 07 17:31:04 DocScrutinizer: hey, did you get my mail? Sep 07 17:31:25 yes. BluesLee . thanks :-) Sep 07 17:31:37 perhaps you can get Stefan_schmidt to help you on that Sep 07 17:31:48 he got a contract to work on libascenario Sep 07 17:31:51 for some weeks Sep 07 17:32:16 dunno whether Liam would think this helps Sep 07 17:32:25 ooh, that's interesting. Though in my world there is no room for a libascenario Sep 07 17:32:42 DocScrutinizer: why is a "defect" freerunner gps antenna more worth to you than the one you gave me? did you see anything odd with the one i sent you? Sep 07 17:32:52 as scenarios are basically evel Sep 07 17:32:56 yeah, Liam seems to want to push libascenario directly in alsa Sep 07 17:33:05 alsa-libs, that is Sep 07 17:33:11 now would be a chance to stop him and present your concept Sep 07 17:33:16 BluesLee: yes. cable was broken Sep 07 17:34:15 DocScrutinizer: soldering at the contact would have done the job i guess Sep 07 17:34:38 DocScrutinizer: or was it broken somewhere else Sep 07 17:34:56 BluesLee: not exactly. you would have to cut 5mm off the cables end Sep 07 17:35:20 okay, i see Sep 07 17:35:42 DocScrutinizer: i think its easy to break it for people who open the device Sep 07 17:36:13 DocScrutinizer: the gps antenna/cable hangs there immediately Sep 07 17:36:24 yes. that's why it's of worth to me. learn failure modes Sep 07 17:37:02 anyway, have to run. buy some smart tings ;-) Sep 07 17:37:16 okay Sep 07 17:37:30 i will do;-) Sep 07 17:38:48 DocScrutinizer: i got me a "feinmechanikerset" since the antenna was broken and now i have fun opening things like my loud speakers who make noise etc Sep 07 17:41:58 dos1: seems like the buildhost built an old revision of shr-today (a rev from 24h ago), maybe github is laging? Sep 07 17:42:47 Slyon: no Sep 07 17:42:51 Slyon: syncing feed now Sep 07 17:42:52 dos1: now i've send working recieps :) Sep 07 17:43:42 dos1: btw, about wifi. Do you tell me your wifi started to work at school where it never worked with .28 but worked with .24? Sep 07 17:43:55 dos1: but why isn't the rev mentioned in filename not the same as HEAD? Sep 07 17:44:12 PaulFertser: no, it started to work in home, where it usually didn't work with 28 Sep 07 17:44:18 at school it always worked Sep 07 17:44:28 and i checked today, it works at school ;) Sep 07 17:45:08 Heinervdm: http://patchwork.dev.bearstech.com/patch/85/ Sep 07 17:45:11 Heinervdm: what about that one? Sep 07 17:45:23 dos1: the latest change was reverting the patch that appeared only in .29. Sep 07 17:45:38 dos1: that's working with webkit-efl-0.1 Sep 07 17:46:23 PaulFertser: hi, wifi doesnt work for me too ... how i apply you patch, opkg upgrade? Sep 07 17:46:33 BluesLee: yes Sep 07 17:46:44 BluesLee: just upgrade to the latest kernel and modules in SHR repos. Sep 07 17:46:48 Heinervdm: short answer: should that patch be applied when applying also your latest patch? Sep 07 17:46:55 and nothing more? :P Sep 07 17:47:01 BluesLee: it doesn't "fix" wifi, it just reverts the patch that broke it too much. Sep 07 17:47:08 dos1: yes Sep 07 17:47:31 PaulFertser: okay, is see. can i extract the newest modules and flash the kernel? Sep 07 17:47:45 BluesLee: why? SHR does that for you automatically iirc. Sep 07 17:48:11 PaulFertser: a complete opkg upgrade may break something;-) Sep 07 17:48:18 is there some on with the same issue described in http://trac.freesmartphone.org/ticket/474 Sep 07 17:48:25 dos1, ! Sep 07 17:48:26 :) Sep 07 17:48:37 please rebuild hebrew keyboard layout ;) Sep 07 17:48:41 PaulFertser: for instance my personal themes etc .. Sep 07 17:48:44 SHR: tom illume-keyboards * r89e689967d8a /hebrew/Hebrew.kbd: improved the hebrew keyboard so the shift layout now makes some sense Sep 07 17:48:52 BluesLee: then do upgrade at least ar6000 module Sep 07 17:49:01 PaulFertser: okay, will do it Sep 07 17:49:05 TAsn: is it in task-shr-feed? Sep 07 17:50:33 dos1, it's in tasn-is-clueless task iirc :) Sep 07 17:50:36 no idea, really. Sep 07 17:51:37 dos1: ah, now the right rev is built. nice. Sep 07 17:51:40 have to leave, bye Sep 07 17:52:32 PaulFertser: i have a stupid question, i connect via an usb wifi stick on my desktop, wehenver i plug in my neo via usb i loose internet connection there Sep 07 17:52:41 dos1, please please please :) Sep 07 17:52:48 and please tell me when you are done ;) Sep 07 17:53:02 (19.52.32) BluesLee: PaulFertser: i have a stupid question, i connect via an usb wifi stick on my desktop, wehenver i plug in my neo via usb i loose internet connection there Sep 07 17:53:03 TAsn: i'll build it, just let me apply some patches from patchwork before starting build Sep 07 17:53:05 ^^ that is normal Sep 07 17:53:06 ;) Sep 07 17:53:06 ;) Sep 07 17:53:15 dos1, cool, thanks :) Sep 07 17:53:20 NM prefers eth over wifi Sep 07 17:53:20 finally after ~1 year Sep 07 17:53:26 the hebrew keyboard will make sense :) Sep 07 17:53:47 khiraly1: i dont think so, i am sure one can avoid it but i dont know how Sep 07 17:54:02 BluesLee: i guess you need proper metrics set for the interfaces Sep 07 17:54:14 BluesLee: no idea about NM configuration, khiraly1 knows much more Sep 07 17:54:19 about that shit Sep 07 17:54:33 khiraly1: okay, tell me more about that sh... Sep 07 17:54:51 I began to write a tutorial, but got distracted by other things. Sep 07 17:54:58 it needs to be cleaned up Sep 07 17:55:07 in short you need to use keystring plugin of NM Sep 07 17:55:14 there you can finetune the connctions Sep 07 17:55:34 khiraly1: i use knetworkamanager Sep 07 17:55:50 khiraly1: keystring plugin, hmm Sep 07 17:55:54 Seen that?! I installed ipv6 support specifically to watch Star Wars over telnet in colors. But currently it doesn't respond at all via ipv6. :-/ Sep 07 17:55:56 I think it use NetworkManager in the background Sep 07 17:56:11 BluesLee: look at the /etc/NetworkManager dir Sep 07 17:56:39 khiraly1: its present Sep 07 17:56:46 you could also play with the ifupdown plugin, but it has bugs. It prefers always eth over usb Sep 07 17:57:00 TAsn: done :) Sep 07 17:57:16 khiraly1: how do i activate that plugin? Sep 07 17:57:37 but you know what? I gave the pointer, or you can google for yourself, our you wait until I finish the tutorial (I hope i will have time this week for this), or you help the tutorial writing Sep 07 17:58:20 others reported, that setting freerunner's connection via the gui (as static connection) seems reliable (although you cant finetune some pretty awesome thing) Sep 07 17:58:24 dos1, arrigato. synced? Sep 07 17:58:32 Ainulindale, hi! I'm back, if you still have some questions, welcome Sep 07 17:59:30 TAsn: yup Sep 07 18:03:43 bbyt Sep 07 18:53:15 ~spell accelerometer Sep 07 18:53:17 'accelerometer' may be spelled correctly Sep 07 18:53:22 ~spell accellerometer Sep 07 18:53:24 possible spellings for accellerometer: accelerometer accelerometers accelerometer's Sep 07 18:53:30 ~botsnack Sep 07 18:53:30 mickey|bbl: :) Sep 07 19:05:37 freesmartphone.org: seba.dos1 specs * r4ff25fe92212 / (3 files in 3 dirs): org.freesmartphone.Device.RealtimeClock: add WakeupTimeChanged signal Sep 07 19:06:35 dos1: featurerequest (or is this already implemented?) change date with shr-settings Sep 07 19:06:53 bumbl: WONTFIX until calendar widget will be there ;) Sep 07 19:07:01 morphis is working on it Sep 07 19:07:20 dos1: ok Sep 07 19:07:30 because otimed does not want to set the time apparently Sep 07 19:12:15 What the status of the touchscreen filter kernel drivers is with respect to upstream submission? Sep 07 19:12:40 darkstar62: comments received, Nelson is slowly working on them. Sep 07 19:13:10 darkstar62: he's going to post s3c ADC TS driver with really minimal filter implementation (and it has good chances to be accepted) soon. Sep 07 19:13:52 cool. Anything I can do to help? I'm working on a project to port Android to omap850-based devices (http://wing-linux.sf.net) and we're very interested in using the filters for touchscreen calibration. Sep 07 19:14:19 but we'd like to do so in a way that would be accepted upstream Sep 07 19:14:44 darkstar62: i guess if you search through the LAKML archives you'll see what is going on wrt TS filters and so whom to ask about details. Sep 07 19:15:12 darkstar62: may i ask why you want android on your devices instead of proper GNU/Linux system? Sep 07 19:15:28 actually, we're going for both Sep 07 19:15:36 Android as the GSM/UI, with Linux underneath Sep 07 19:15:59 currently we've got Angstrom, although I've been exploring the idea of using Ubuntu as the linux userspace Sep 07 19:17:37 darkstar62: but android is full OS on its own, running on Linux, the kernel, how are you going to make it work in parallel with proper libc implementation, alongside with X etc? And most important, why do you want semi-closed end-user-targeted "opensource" crap on your devices? Sep 07 19:17:41 freesmartphone.org: mickey cornucopia * r8e33c15a3bba /fsodeviced/ (5 files in 3 dirs): fsodeviced: add build skeleton for accelerometer devices and plugins Sep 07 19:18:44 It's not without it's bugs, but for the most part yes, they work beautifully side-by-side. We don't have X running presently, although that wouldn't be hard to do. Sep 07 19:19:18 As for the closed stuff, it seems pretty open to me. And a lot of people really like it, so I figure why not target it? Sep 07 19:19:35 There's nothing stopping us from using other things like GPE, Qtopia, etc. either though Sep 07 19:19:45 darkstar62: it's not hacker friendly, even android devs say it was never targeted for cool guys. Sep 07 19:20:03 I'll agree with that, definitely Sep 07 19:20:23 darkstar62: also android wants some stupid kernel drivers that are never going upstream... Sep 07 19:20:40 true, and I don't think we'd ever get around that Sep 07 19:21:01 but I'm ok with that Sep 07 19:21:06 for me, it's all about choice Sep 07 19:21:31 Android is the most difficult of things to get working on a new device. Once we have Linux, everything else is fairly easy Sep 07 19:21:33 darkstar62: i just don't understand why spend any time on getting it to work... I remember it was half a year only in binary form. Also it looks like their internal repository is still closed, they only do occassional code drops to public etc etc. Sep 07 19:21:38 so if we can get it all, why not? Sep 07 19:22:09 darkstar62: how many java developers are around here? Sep 07 19:22:19 PaulFertser: that may be, and that's their option Sep 07 19:22:22 darkstar62: how many python developers are around here? Sep 07 19:22:28 Whom should we target? Sep 07 19:22:53 whatever you want. I chose Linux and Android Sep 07 19:23:02 darkstar62: that's definetely an option for some google assholes who're ok about collaborating with china goverment, with selling shit on their android market and much more... Sep 07 19:23:30 freesmartphone.org: seba.dos1 framework * r0c4b12c4df42 /framework/subsystems/odeviced/kernel26.py: odeviced: implement WakeupTimeChanged signal in RTC Sep 07 19:23:48 They choose "opensource" not because they care about _our_ freedom but because that way they get more money. I don't like that attitude and can't see why collaborate with them... Sep 07 19:24:12 Every company has to make money somehow. I respect that and have no issues with companies using open source to make money Sep 07 19:24:29 call me a non-purist, but I think the spirit of open source is anyone can use it Sep 07 19:24:48 darkstar62: open source is not always == free software Sep 07 19:24:53 darkstar62: it'd be nice if some devs out there helped to get FSO running on more devices. It already has a decent modem abstraction and is hackable. Sep 07 19:25:03 dos1: exactly what i wanted to say Sep 07 19:25:34 i'm not purist, but i love free software. being only open source is not enough Sep 07 19:25:56 licence of most of Android code is free software. but maitaining Android and market around it isn't free. Sep 07 19:26:14 that's why i really prefer openmoko over android Sep 07 19:26:26 Opensource is marginally better than proprietary soft, sure. But it's the motivation behind it that bothers me. Sep 07 19:26:40 darkstar62: btw, do you remember Opie? Sep 07 19:26:46 yup Sep 07 19:27:07 freesmartphone.org: mickey specs * rb840fdddff70 / (4 files in 3 dirs): add org.freesmartphone.Device.Orientation (tentative, please don't use yet) Sep 07 19:27:09 darkstar62: Qtopia was opensource for quite some time. And folks needed to fork it to make any sense. And everything failed. Sep 07 19:28:30 not sure I understand...Qtopia was forked into Opie? Sep 07 19:28:45 darkstar62: do you remember Qtopia's rule of submitting patches? It was like this: "you give them BSD-licensed patches and if they're merciful and it fits all requirements of all their customers, they'll probably give it as GPL to everybody else later". Sep 07 19:28:57 darkstar62: sure, Opie is Qtopia2.something fork. Sep 07 19:29:05 ah, no I wasn't familiar with that Sep 07 19:29:21 yeah, that's not cool Sep 07 19:29:23 darkstar62: do you know that Openmoko project also tried to fork Qtopia (3.something this time)? And ultimately failed? Sep 07 19:29:37 PaulFertser: gpl is cool but i am more pragmatic too - i don't use linux because of the license and i bought the moko because back then i thought i would get the good experience i had on my desktop on my phone Sep 07 19:29:48 Exactly because "opensource" is evil. We had enough of that :S Sep 07 19:29:54 I see Sep 07 19:30:14 bumbl: do you care about good people? Sep 07 19:30:49 bumbl: do you think it's ethical to accept only BSD patches to be able to hide them or later relicense as GPL (thus taking freedom away, because GPS is more restrictive)? Sep 07 19:30:58 bumbl: would you like your buddy do that to you? Sep 07 19:31:03 That's what i'm talking about. Sep 07 19:31:04 for gstreamer experts: is it possible to start a pipeline synchronously? Sep 07 19:31:22 no Sep 07 19:31:28 even if it was, you don't want that Sep 07 19:31:31 it can take _ages_ Sep 07 19:31:46 better do something meaningful instead of waiting Sep 07 19:32:49 Hi all Sep 07 19:32:50 :D Sep 07 19:33:08 anyone tried to run windows mobile on freerunner :D Sep 07 19:33:19 freesmartphone.org: mickey specs * r28df51280ca9 /html/index.html: link to new orientation API Sep 07 19:33:30 Technological advantages are cool and all and probably matter for some stuff but i personally care about people i met along the way, not about "polished" or even "friendly" software. Sep 07 19:33:41 damn, spent a fortune for some atmel and cheap plastic housings... Well now I'll starve in an automated home ;-) Sep 07 19:34:21 That's why i do as much as i can to stay away from both proprietary and opensource folks, they just doesn't seem to be friendly enough, caring about money and their consumeristic demands too much. Sep 07 19:34:32 DocScrutinizer: woohoo Sep 07 19:34:37 DocScrutinizer: what are you going to automate? Sep 07 19:34:44 DocScrutinizer: garage door? ;) Sep 07 19:34:59 for now flat door Sep 07 19:35:06 and light Sep 07 19:35:25 PaulFertser: if my buddy tells me he wants my patches that way and does not bother when i don't submit any patches because of the license i don't see why we can't be buddies Sep 07 19:35:27 homematic.com Sep 07 19:35:52 cost more than gold, but soooo cute Sep 07 19:36:55 bumbl: because he's spoiling your fun? Your patches don't get upstream -> you have to update them every time a new version is out, also it's not a very pleasant feeling you have something cool but can't share because of license limitations. Sep 07 19:36:55 and as long as his product is superior from a productive point of view i don't see why i shouldn't use it just because of the license Sep 07 19:37:03 hmm, the gmeter in remote control is a little too sensitive. have to check if I can recalibrate to be a little more economic with auto backlight Sep 07 19:38:10 freesmartphone.org: mickey cornucopia * r1e814417cae6 /libfsoframework/fsoframework/ (fsoframework-2.0.vapi interfaces.vala): fsoframework: add service path prefix for org.freesmartphone.Device.Orientation Sep 07 19:40:16 DocScrutinizer: starter kit for 260EUR? oooh Sep 07 19:42:05 PaulFertser: if you want to get serious on hacking the central control unit (which runs linux btw) you don't even start at twice that value :-/ Sep 07 19:42:27 DocScrutinizer: why not go fully DIY solution? Sep 07 19:43:13 (21.35.27) DocScrutinizer: homematic.com <-- why not arduino and freerunner? Sep 07 19:43:25 because I had to care about ME/ID then, don't have the cool bidir AES key RF chips... Sep 07 19:43:59 and can't sell my knowledge to prolls that want a cool home but don't know a shit about coding Sep 07 19:44:26 ME/ID? Sep 07 19:44:38 Mechanical engineering, intellectual ... Sep 07 19:44:45 DocScrutinizer: aha, that starts to make sense. Sep 07 19:44:57 you have used this shortcut just forgot it Sep 07 19:45:45 industrial design == case Sep 07 19:46:32 afaik Sep 07 19:46:33 we had no glossary in OM ;-) Sep 07 19:47:44 hmm, they "industrial design" looks ugly Sep 07 19:47:55 could as well mean intelligent demolition Sep 07 19:55:42 mickey|bbl, i'm just looking into you new specs and i think we should split the orientations into 2 parts or add 2 orientations Sep 07 19:56:23 you can be on the left and on the right side if you are in landscape mode Sep 07 20:24:03 freesmartphone.org: mickey cornucopia * raf8adcff4154 /fsodeviced/src/ (7 files in 6 dirs): fsodeviced: accelerometer: instanciate accelerometer plugin Sep 07 20:24:38 DocScrutinizer2: yupyup, i don't know why the hell it's borked again, probably networking connectivity problems :( Sep 07 20:25:37 PaulFertser_: freenode down Sep 07 20:25:50 I see you Sep 07 20:25:55 the cake is a lie, though... Sep 07 20:26:11 PaulFertser_: bip will need a while to recover probably Sep 07 20:27:57 PaulFertser_: the FSCKNG 450bucks central control unit is DOA :-((( Sep 07 20:29:38 DocScrutinizer2: that's a truly bad surprise :( Sep 07 20:30:09 PaulFertser_: yup, no real fun this night Sep 07 20:30:41 need dinner and a couple of beers to mediate my frustration Sep 07 20:31:06 bbl Sep 07 20:31:14 DocScrutinizer2: see you tomorrow Sep 07 20:31:24 yo, night paul Sep 07 20:32:03 PaulFertser_: maybe have a look after bip Sep 07 20:32:22 might need a kick Sep 07 20:39:25 DocScrutinizer2: what does the central control unit control ? Sep 07 20:40:40 is setting the accelerometer sample rate broken or just here? Sep 07 20:41:38 mickeyl: about your e-mail re: device orientation api Sep 07 20:42:04 mickeyl: should not absoulute directions be used? Sep 07 20:42:29 no Sep 07 20:42:31 ie: +/-Z == faceup/down? Sep 07 20:42:46 then how do you define the difference in landscapes? Sep 07 20:43:12 well, obviously i've been missing one category :) Sep 07 20:43:48 i should either add upsidedown Sep 07 20:43:52 or the portiats Sep 07 20:44:01 portiats? Sep 07 20:44:14 err... portrait Sep 07 20:44:21 normal/inverted Sep 07 20:44:24 right Sep 07 20:44:46 need to add normal/inverted Sep 07 20:45:23 so there will be six possible substrings in the response? Sep 07 20:45:40 yes, but some contradict each other Sep 07 20:45:45 how so? Sep 07 20:45:53 landscape and portrait won't come together Sep 07 20:45:59 normal and inverted also Sep 07 20:46:04 faceup facedown also Sep 07 20:46:32 I'm not sure I'm understanding the possible responses then. Sep 07 20:46:43 portrait normal faceup Sep 07 20:46:49 that's one possible response Sep 07 20:46:51 is it one of those six, or an array of all applicable Sep 07 20:46:54 ahh pl Sep 07 20:46:55 ok Sep 07 20:46:56 hehe. same thoughts ;) Sep 07 20:47:09 it's an array, but in one string Sep 07 20:47:39 I was misunderstanding a bit, sorry Sep 07 20:47:45 np Sep 07 20:47:58 if it is cumbersome we can do it differently Sep 07 20:48:06 i just want to go ahead and start making use of the sensors Sep 07 20:48:54 and preventing apps fighting against each other Sep 07 20:49:03 i thought of [ {left|right}landscape, {up|down}portait} in combination with face[up|down] Sep 07 20:49:04 so that has to be in fso Sep 07 20:49:56 i think that should cover all posiibilities Sep 07 20:50:04 s/sii/ssi/ Sep 07 20:50:04 playya_ meant: i think that should cover all possibilities Sep 07 20:50:07 I would personaly prefer [+-][XYZ] oriented off of accel2 Sep 07 20:51:18 but I'm a cartesian kind of guy :) Sep 07 20:51:46 hehe Sep 07 20:51:48 Toaster`, i think this would be useful for a game like mokomaze, but not if you have to flip your appication Sep 07 20:52:09 +-xyz is what comes out of the input nodes Sep 07 20:52:10 s/appi/appli/ Sep 07 20:52:11 playya_ meant: Toaster`, i think this would be useful for a game like mokomaze, but not if you have to flip your application Sep 07 20:52:19 i need a new keyboard :( Sep 07 20:52:19 if you want raw access, you don't bother with FSO Sep 07 20:52:27 you will just claim access to the Accelerometer resource Sep 07 20:52:29 and start tweaking Sep 07 20:52:37 however if all you want is orientation Sep 07 20:52:40 then you can use FSO Sep 07 20:52:50 (and later on triggering wakeup etc.) Sep 07 20:52:59 what I'm thinking is do the filtering (obviously) but just call out the major change Sep 07 20:53:12 exactly Sep 07 20:53:26 to keep CPU impact low, we need to use a high threshold Sep 07 20:53:31 as in, you went from +X to -Y Sep 07 20:54:27 I have script that polls the accels for 10 data point then re-orients the screen as appropraitely. setting the filter any less than .75G was just noise Sep 07 20:54:36 so, yeah, I agree :) Sep 07 20:56:34 mickeyl, do you understand pvborza's code? Sep 07 20:57:19 never bothered to look Sep 07 20:57:31 unfortunately he didn't show interest to bring that stuff into FSO Sep 07 20:57:39 hence I need to do something more simple Sep 07 20:57:52 i don't feel like learning these algorithms now Sep 07 20:58:54 but before not using the sensors at all, i'd rather add the boring things Sep 07 20:59:02 we can always go fancy later Sep 07 21:01:12 i looked into it but never figured how it works Sep 07 21:02:29 any fso people here? Sep 07 21:03:55 maybe it's because it was his master thesis ;) Sep 07 21:04:02 mickeyl, btw fakeroot 1.9.6 doesn't exist on the debian mirror Sep 07 21:04:36 budfive_: depends... what's up? Sep 07 21:05:20 mickeyl: if I issue org.freesmartphone.GSM.SIM.RetrieveMessagebook do you know where the timezone field comes from? Sep 07 21:05:36 mickeyl: it's wrong for me and causes shr-messages and such to get the date completely wrong Sep 07 21:07:11 mickeyl: specifically, it's not just wrong, but invalid. I have "+2400" Sep 07 21:07:27 iirc this is verbosely taken from the source, i.e. the SIM Sep 07 21:07:36 do you have logs? Sep 07 21:08:03 be back in a bit Sep 07 21:10:54 mickey|bbl: I don't know which logs you're referring to, but I can trivially generate anything you ask for Sep 07 21:15:26 hi all Sep 07 21:18:37 sorry i have a question Sep 07 21:18:53 there is a method to mount jffs2 in rw mode? Sep 07 21:19:37 afaik no Sep 07 21:20:02 afaik? Sep 07 21:21:17 oh Sep 07 21:21:17 ok Sep 07 21:21:20 thanks lot Sep 07 21:21:24 :( damn jffs Sep 07 21:21:25 xD Sep 07 21:44:04 SHR: seba.dos1 opimd-utils * rfa7516cc0a45 /TODO: Update TODO Sep 07 21:44:04 SHR: seba.dos1 opimd-utils * r6f3ad432f4ad / (5 files in 2 dirs): Merge branch 'master' of git@git.shr-project.org:opimd-utils Sep 07 21:45:24 sanvy: huh? how could you ever modify jffs2 if it was only read-only? Sep 07 21:46:02 lindi-, it's this the problem Sep 07 21:46:09 I need to modify a jffs2 Sep 07 21:46:14 but it is readonly Sep 07 21:46:18 sanvy: om2008 modifies jffs2. Sep 07 21:46:24 :\ Sep 07 21:46:27 om2008? Sep 07 21:46:33 om2008 distro Sep 07 21:46:44 how does it modify jffs2? Sep 07 21:46:47 sanvy: it has its root filesystem on jffs2. surely it is able to modify it? Sep 07 21:46:55 kk Sep 07 21:47:09 but if I need to modify jffs2 using my pc? Sep 07 21:48:19 sanvy: mtdram? Sep 07 21:48:21 because if i need to modify using pc is because from moko i can't ^^ Sep 07 21:48:24 mrdram??? Sep 07 21:48:28 mtd Sep 07 21:48:30 mtdram* Sep 07 21:48:36 what's that? Sep 07 21:48:55 a module that creates a ram-based mtd device Sep 07 21:49:03 :\ Sep 07 21:49:23 so if i mount my jffs2 on a mtd device I can modify it? Sep 07 21:49:34 yep Sep 07 21:49:40 :P Sep 07 21:49:56 budfive_: fso emits debug logs (see FsoFramework page in Openmoko wiki), these logs dump communication between modem and fso Sep 07 21:50:05 budfive_: we could find out where the problem is by inspecting these logs Sep 07 21:51:36 mickey|bbl: I've been looking Sep 07 21:51:49 mickey|bbl: decodePDUTime in gsm/convert.py parses the timezone Sep 07 21:52:03 freesmartphone.org: seba.dos1 dos/opimd-tracking * r1541ea8a3bce framework/framework/subsystems/opimd/pimd_messages.py: Sep 07 21:52:03 freesmartphone.org: opimd: Messages: implement AddIncoming dbus method Sep 07 21:52:03 freesmartphone.org: It can be used by apps to "deliver" message to user, without writing special Sep 07 21:52:03 freesmartphone.org: opimd backend/handler for that. Sep 07 21:52:23 freesmartphone.org: mickey cornucopia * r8cee554efe49 /fsodeviced/ (configure.ac data/Makefile.am): fsodeviced: install dbus configuration file Sep 07 21:52:32 mickey|bbl: debug logging will show me the binary content of the sms? Sep 07 21:52:37 thanks lot lindi- Sep 07 21:52:41 router reboot Sep 07 21:52:53 budfive_: hmm, actually, it'd be good to see it in ascii first Sep 07 21:53:00 you can run mickeyterm to talk with the modem Sep 07 21:53:06 while the rest is running Sep 07 21:53:24 mickey|bbl: ok. I'm using shr. So run mickeyterm and send myself an sms? Sep 07 21:53:42 PaulFertser: waaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaahhhhAAAAAAAAAAAAAAAAAHHH Sep 07 21:53:43 no, just read the messagebox Sep 07 21:53:49 let me get you the commands Sep 07 21:55:03 PaulFertser: bip replayed *all* channels *an* *private* *queries* of the last 6 months :-S #\%\<##4\/\/>#4@44# Sep 07 21:55:05 sorry people Sep 07 21:55:09 at+CSCS="GSM" Sep 07 21:55:18 at+cmgf=1 Sep 07 21:56:06 at+cmgl="ALL" Sep 07 21:56:15 then check if the date are ok here Sep 07 21:56:28 mickey|bbl: ok. one sec. having trouble with mickeyterm. Thank you for helping. Sep 07 21:56:29 if not, we need to get the PDUs for analysis Sep 07 21:57:22 bye Sep 07 21:57:24 thanks all Sep 07 21:57:40 mickey|bbl: ok. timestamp string is "09/09/07,09:55:34-32" Sep 07 21:58:28 mickey|bbl: is "-32" the timezone? Sep 07 21:58:49 yes Sep 07 21:58:55 that's mucho bogus Sep 07 21:59:05 or where are you located? Sep 07 21:59:23 mickey|bbl: was this parsed by anything first, or this comes directly from the provider? Sep 07 21:59:26 mickey|bbl: I'm in california Sep 07 21:59:47 IIRC this comes from the modem Sep 07 21:59:48 mickey|bbl: gmt-7 or so Sep 07 22:00:17 mickey|bbl: is "-32" supposed to be in hours? pdu is in units of 15min, right? Sep 07 22:00:19 * DocScrutinizer mmmbblllmmmblll Sep 07 22:00:41 mickey|bbl: so if it IS in 15min chunks, -32== -8hours, which could be correct Sep 07 22:00:53 hmm Sep 07 22:00:54 dunno whom to shoot first. BIP or freenode Sep 07 22:01:38 haaahh, time over gsm, right? Sep 07 22:01:50 no, sms timestamp Sep 07 22:02:04 ooh. almost the same I guess Sep 07 22:02:15 i'm not sure Sep 07 22:02:25 can't look that up now Sep 07 22:02:33 DocScrutinizer: something is causing bogus timezones to be attached to sms messages. This confuses shr-messages to say that the sms came from 1969 Sep 07 22:02:55 mickey|bbl: I can take a look. Can you tell me how to see the raw pdu bytes? Sep 07 22:03:01 budfive_: create a bug @ fso with the PDUs as well Sep 07 22:03:10 at+cscs="UCS2" Sep 07 22:03:12 at+cmgf=0 Sep 07 22:03:15 at+cmgl="ALL" Sep 07 22:03:26 please attach both the PDU and ascii version to the bug Sep 07 22:03:32 one of us will analyse that then Sep 07 22:03:33 mickey|bbl: ok. Which one is fso parsing? Sep 07 22:03:37 anybody traveling to hongkong soon? I have a small package of highly potent explosives to deliver over there... Sep 07 22:03:40 fso is using pdu mode Sep 07 22:03:44 TZ -- lol! Sep 07 22:03:48 mickey|bbl: ok. Thank you again Sep 07 22:03:50 np Sep 07 22:04:12 mickey|bbl: I have a suspicion that it's broken for everybody < GMT Sep 07 22:04:30 could be. i'm just wondering why no one else complained yet Sep 07 22:04:37 it's not as you're the first user in USA Sep 07 22:04:40 fso, that is Sep 07 22:04:44 mickey|bbl: lots of complaints about sms dates being set to 1969 Sep 07 22:04:55 really? you're the first i heard Sep 07 22:04:57 http://shr-project.org/trac/ticket/205 for example Sep 07 22:04:58 where are these people complaining? Sep 07 22:05:20 wtf 69? always thought 1-1-70 is the magic date Sep 07 22:05:26 mickey|bbl: I've seen this forever, but only started to look at it today Sep 07 22:05:40 DocScrutinizer: Some people supposedly see 1970. I see 1969. Don't know. Sep 07 22:06:58 budfive_: prolly those are in negative TZ ;-))) (wrt GMT) Sep 07 22:07:58 s/those/you/ Sep 07 22:07:58 DocScrutinizer meant: budfive_: prolly you are in negative TZ ;-))) (wrt GMT) Sep 07 22:08:04 DocScrutinizer: yep. the relevant code from fso is: sign = (timezone >> 7) * 2 - 1 Sep 07 22:08:04 zone = (timezone & 0x7f) / -4. * sign Sep 07 22:08:40 giggle, rotfl Sep 07 22:09:58 where goes in blofish algo??? XD Sep 07 22:10:07 http://pastebin.ca/1557750 Sep 07 22:10:12 is that expected? Sep 07 22:10:21 hey mickey|bbl Sep 07 22:10:26 bad boy you don't return kudos :-) Sep 07 22:10:33 some months ago the data I got was always 0,1,2 after another Sep 07 22:10:44 now i see the axis being non-uniformly distributed Sep 07 22:10:51 well, age old saying: "the source is the doc" Sep 07 22:10:53 mickey|bbl: if timezone is negative, won't "sign = (timezone >> 7) * 2 - 1" produce -3? Sep 07 22:10:55 Ainulindale: heh, thanks man, i didn't login to ohloh since then Sep 07 22:10:56 mickey|bbl: do you want me to kiss vegyraupe for you tomorrow? :-p Sep 07 22:11:13 Ainulindale: hmm, depends. did he deserve it? and why do you see him anyways? Sep 07 22:11:23 he's in bed with pyneo Sep 07 22:11:27 that gives bad charma :D Sep 07 22:11:28 mickey|bbl: that's it. I should see -8, but I see -24 Sep 07 22:11:32 mickey|bbl: that is the bug Sep 07 22:11:53 mickey|bbl: well for qi hardware Sep 07 22:12:05 budfive_: i didn't write that part of ogsmd. Please open a bug with all your info and one of us will take care Sep 07 22:12:16 so >>2.17 Sep 07 22:12:20 mickey|bbl: ok. I'll even attach a patch :) Sep 07 22:12:28 Ainulindale: well... i'm not yet excited about anything i've seen from them, so... you obviously must know more Sep 07 22:12:35 budfive_: great! :) Sep 07 22:13:26 budfive_: or do you know a better way to divide by 3? ;-) Sep 07 22:13:53 budfive_: than rightshift 2.17 bits Sep 07 22:14:49 DocScrutinizer: :) Sep 07 22:19:28 ooh, thought you'd change withoutna comment from bbl to zzZZzz Sep 07 22:20:31 esp since I spread nonsense unaware of log-dual(3) Sep 07 22:20:47 which isn't 2.17 Sep 07 22:22:48 PaulFertser: I think BIP is a whore. Will abandon it... Sep 07 22:24:45 PaulFertser: gave me lots of awe and shock, and not really any benefit. Sep 07 22:25:53 PaulFertser: even /bip backlog 1 is borked and replays *full* backlog of everything it got Sep 07 22:33:30 hmm Sep 07 22:33:31 286 /* we have a valid sample set? */ Sep 07 22:33:32 287 if (read[0] & LIS302DL_STATUS_XYZDA) { Sep 07 22:33:32 288 input_report_abs(lis->input_dev, ABS_X, mg_per_sample * Sep 07 22:33:32 289 (s8)read[LIS302DL_REG_OUT_X - LIS302DL_REG_STATUS]); Sep 07 22:33:32 290 input_report_abs(lis->input_dev, ABS_Y, mg_per_sample * Sep 07 22:33:33 291 (s8)read[LIS302DL_REG_OUT_Y - LIS302DL_REG_STATUS]); Sep 07 22:33:35 292 input_report_abs(lis->input_dev, ABS_Z, mg_per_sample * Sep 07 22:33:37 293 (s8)read[LIS302DL_REG_OUT_Z - LIS302DL_REG_STATUS]); Sep 07 22:33:45 this suggest that the three axis are always reported in one set Sep 07 22:33:49 i wonder why my output is not confirming that Sep 07 22:34:38 mickeyl: are you trying to write a parser for the accels? Sep 07 22:34:55 sure, i'm implementing the new Orientation API Sep 07 22:35:34 as I recall they're reported one at a time, then a sync signal is sent Sep 07 22:35:54 * Toaster` looks are his parser... Sep 07 22:36:01 well, check http://pastebin.ca/1557750 Sep 07 22:36:30 * mickeyl doublechecks with hexdump Sep 07 22:36:45 is that dumping everything? Sep 07 22:37:21 yes Sep 07 22:37:39 http://pastebin.ca/1557778 is the hexdump Sep 07 22:37:43 in the middle you see an odd set Sep 07 22:37:59 line 16 Sep 07 22:39:00 is that not the sync set? Sep 07 22:40:03 err, line 17 Sep 07 22:40:07 16 sync Sep 07 22:40:11 17 X axis Sep 07 22:40:13 then 18 sync again Sep 07 22:40:18 where are the Y and Z? Sep 07 22:40:38 mickeyl: /sys semantics changed for gmeter. Only reports changed values Sep 07 22:40:48 oooh! Sep 07 22:40:59 ahh, thanks DocScrutinizer2 Sep 07 22:41:00 that's why Sep 07 22:41:05 np Sep 07 22:42:15 absolute vs relative Sep 07 22:42:33 however misleading those names might be Sep 07 22:42:40 mickeyl: http://pastebin.com/m231efce9 Bad ides me giving you references, but that's my rotating script Sep 07 22:43:03 it's adapted from the wiki Sep 07 22:43:08 hehe Sep 07 22:43:08 thanks Sep 07 22:43:13 that looks familiar Sep 07 22:43:19 i guess i've written the first version of that :) Sep 07 22:43:24 heh Sep 07 22:43:36 there are only so many ways to implement it as well :) Sep 07 22:43:54 are you writing this in vala? Sep 07 22:44:16 yes Sep 07 22:44:22 * DocScrutinizer2 wonders how an atmel box for 450 bucks can be DOA >>:-(( Sep 07 22:44:25 it's going to be a new plugin in fsodeviced Sep 07 22:44:33 DocScrutinizer2: :/ Sep 07 22:44:36 FSCK!! Sep 07 22:48:41 err 499 http://www.conrad.de/goto.php?artikel=646262 Sep 07 22:49:07 499 Euro... damn Sep 07 22:49:30 you should expect they can hand solder and check with multimeter each and every solderpoint for that price Sep 07 22:49:43 heh Sep 07 22:50:26 Grmbl, ohloh takes ages to get shr/import Sep 07 22:50:32 Toaster`: it's a fsckng ATMEL(!!!) chip in there. Roundabout the functions we refuse to make use of in FR Sep 07 22:55:14 DocScrutinizer: I chuckle as to hand solder/check each component (something my company does) is very expensive. Not to deminish your problem at all, it's just a matter of scale. and my odd sense of humour. Sep 07 22:57:12 Toaster`: this stupid pastic box has some 64Mb storage, one usb-device and 2 host, 10bT, and a display we need 6 of them to make up for one FR LCM. Still it costs twice the price :-( and comes DOA Sep 07 22:57:42 And this is for your home automation project? Sep 07 22:57:48 yep Sep 07 22:58:25 seems rather extensive Sep 07 22:58:28 they have one killer feature and that's bidir RF with AES Sep 07 22:58:34 ahh ok Sep 07 22:59:14 and it's hard to find a bidir RF-module with a standby of years on battery Sep 07 22:59:37 even unidir without AES Sep 07 22:59:37 heh, so the FR is definitly out then :) Sep 07 23:04:06 plus their devices like that http://www.conrad.de/goto.php?artikel=646284 or that http://www.conrad.de/goto.php?artikel=646337 or that http://www.conrad.de/goto.php?artikel=751501 or that http://www.conrad.de/goto.php?artikel=646469 -- you don't really want to start and try build that on your own Sep 07 23:04:52 but costs more than gold Sep 07 23:05:14 hope to make money out of it eventually Sep 07 23:05:25 once I hacked that stupid system Sep 07 23:06:50 <_phil> know which RF protocol those are based on? Sep 07 23:06:57 U know my businesscard subline "hw and synergy". Bet I van make three times the value out of same hw, by simply exploiting the oportunities given Sep 07 23:07:11 _phil: prop I guess Sep 07 23:07:15 866MHz Sep 07 23:08:52 I planned to root that stupid box this evening. Now it seems I have to go back to ALSA-ACI ;-D Sep 07 23:09:25 ...or switch on TV Sep 07 23:09:54 <_phil> years of battery life and AES sounded like zigbee :) Sep 07 23:10:09 yet it isn't Sep 07 23:10:19 866MHz Sep 07 23:10:39 <_phil> yeah, cc 1100 Sep 07 23:10:56 one of the 2 RF TX I suggested to implement to GTA04 ;-) Sep 07 23:11:49 well, was *too* good for OM Sep 07 23:12:07 <_phil> :D Sep 07 23:17:16 g'night Sep 07 23:21:30 mickey|zzZZzz: night Sep 07 23:22:45 mickey|zzZZzz: night Sep 07 23:30:20 freesmartphone.org: mickey cornucopia * r7c6af6e6afa4 /fsodeviced/src/plugins/ (3 files in 2 dirs): fsodeviced: accelerometer_lis302: read and parse values, call delegate for further processing Sep 07 23:41:04 mickey is a fool (much same like me). Says "n8" then checks in patches Sep 07 23:42:48 /kick mickey ;-) Sep 07 23:59:41 ~opimd-utils Sep 07 23:59:41 well, opimd-utils is not phone suite, it's only set of test scripts for developers. You can use it as phone apps, but without any warranty. Many things are not implemented, *AND THEY PROBABLY WON'T BE*. **** ENDING LOGGING AT Tue Sep 08 02:59:57 2009