**** BEGIN LOGGING AT Sun Nov 22 02:59:57 2009 Nov 22 03:03:19 why are so many programs absent from the SHR package repository ? for example, mg, zile, sox, mpg321, etc.. Nov 22 03:07:10 maybe it is because i'm using the christmas image, and the packages haven't been built for it ? or there really isn't a need for such a variety ? or maybe everyones effort is best used somewhere else than making packages ? Nov 22 03:11:46 gonzoism: some aren't built yet, some are failing to build.. Nov 22 03:12:23 what kind of issues would cause the simpler images to fail to build ? Nov 22 03:13:06 simpler programs, i mean. Nov 22 03:13:53 ie missing patch for gcc-4.4 Nov 22 03:14:43 gonzoism: float arithmetics is a major issue Nov 22 03:14:51 is the patch arm or shr specific Nov 22 03:15:08 ah. i'm guessing arm specific troubles. Nov 22 03:15:10 also libs that don't build but the program needs them Nov 22 03:17:13 so i'm guessing that its problems related to the architecture ? everything is still being ported ? Nov 22 03:18:03 generally if there's a receipe in OE, then it's simple to build the app for OM Nov 22 03:18:30 OE... open ? Nov 22 03:18:40 Open Embedded Nov 22 03:18:58 the distro OM bases on Nov 22 03:20:05 why does everything in suse work ? Nov 22 03:21:36 for example, openoffice Nov 22 03:24:40 well, OpenOffice fpr sure is too heavy for FR Nov 22 03:25:43 and Suse is a rather standard linux, architecture and platform whise Nov 22 03:26:24 ya Nov 22 03:26:47 thanks for all the answers. Nov 22 03:26:55 nevertheeam systempstrless a lot of packages need adaption of their build process as provided by upstream, to build on a distro that's not 100% identical to the u Nov 22 03:27:08 ooops Nov 22 03:27:46 not identical to the ? x86 arch ? Nov 22 03:27:56 nevertheless a lot of packages need adaption of their build process as provided by upstream, to build on a distro that's not 100% identical to the upstream system Nov 22 03:28:34 as SHR is now, can it build itself (for the most part) ? Nov 22 03:28:52 FR has no X86 arch. And it's a different distro than OE, and even more different to Suse Nov 22 03:29:11 right. its arm arch, right ? Nov 22 03:29:13 I heard of people doing that, yeah Nov 22 03:29:24 yes, arm Nov 22 03:29:42 when you say upstream system, what do you mean ? Nov 22 03:30:05 basically the distro on the original author's PC Nov 22 03:30:11 if i'm asking too many questions, you can tell me. :) Nov 22 03:30:23 nah, no problem Nov 22 03:31:47 I also heard of native gentoo builds :-D Nov 22 03:32:18 but generally you don't want to do that, as it'll take ages Nov 22 03:33:42 but I think most apps can be ported with reasonable effort Nov 22 03:34:36 screen size and resolution/orientation one of the major most common issues Nov 22 03:35:31 aaah, and I thought about sox myself yesterday :-) Think it's a must-have Nov 22 03:35:45 didn't check if it's in the repo Nov 22 03:38:03 it wasn't when i checked opkg list | grep sox Nov 22 03:39:24 mrmoku|away: build sox? Nov 22 03:40:53 i didn't notice that i asked the qemu question into #openmoko instead of here. should i ask again here, or are most people here in both channels ? Nov 22 03:42:12 gonzoism: did you have sox in older shr-unstable? Nov 22 03:42:31 i've only had my freerunner 2 days. and i am so happy ! :) Nov 22 03:42:37 its not in task-shr-feed Nov 22 03:42:47 ah.. so probably nobody asked for sox sofar.. Nov 22 03:44:17 gonzoism: you can add request in shr trac Nov 22 03:44:36 JaMae: I did ;-) Nov 22 03:44:45 gonzoism: but to be honest there are bigger problems than missing utils... Nov 22 03:45:11 gonzoism: to it will be pretty down on list probably.. unless you sent patch for OE Nov 22 03:45:45 sox is easy :) ../dev/recipes/sox/sox_13.0.0.bb Nov 22 03:45:50 I guess that's exactly what gonzoism is thinking about right atm Nov 22 03:45:55 i would like to emulate the freerunner in qemu. i think that way, i could do a lot better/more. being that i wouldn't be working on the gsm, wifi, screen, or, any drivers, wouldn't that be a good way to find the library problems you mentioned, or find apps that aren't compiling ? Nov 22 03:46:16 zile and mpg123 too Nov 22 03:46:51 gonzoism: read SHR wiki about how to setup OE environment for cross compile Nov 22 03:47:01 build environment on x86 is much more convenient than qemu Nov 22 03:47:25 and faster than to write patch for qemu to run gta02 rootfs :) Nov 22 03:47:33 thanks ! i like zile. you might consider jed. i don't use jed. but it seems like a good candidate. Nov 22 03:48:19 ok, so if i go with x86, i can what.. chroot into the filesystem ? and start working from there, or should i also use one of the shr kernels ? Nov 22 03:49:00 you cross compile Nov 22 03:49:17 ok Nov 22 03:49:24 your tools are native x86, the resulting binary is arm Nov 22 03:50:18 yep. i did that for openwrt a little. for my bufalo router. and a lot for 4 and 586's Nov 22 03:50:44 or you don't "cross" but simply compile for x86 target, and test the app on your PC Nov 22 03:51:23 so the http://www.shr-project.org/trac/wiki/Building%20SHR instructions. would that be where I should start, (already started) and go from there. Nov 22 03:51:24 ? Nov 22 03:52:03 or is that the wrong direction ? Nov 22 03:52:27 it looks like the right direction, but I haven't started looking through it all, and haven't seen an overview Nov 22 03:53:28 its right direction Nov 22 03:53:52 jamae, i missed the setup OE environment for cross compile above. i see it now. I will read it too. Nov 22 03:55:59 gonzoism: I'd be very interested in sox :-) Nov 22 03:56:03 ok, would it be spinning my wheels to start making as many packages as i can now ? or would that even be counter productive ? would ya'll like to point me into a different direction ? what I see is that with the more tools, the more the OS can do. the less limits there are. Nov 22 03:56:59 every contribution appreciated Nov 22 03:57:58 i'm sure. no doubt in my mind. especially in such a small community. Nov 22 03:59:13 i don't code. so I don't know what else to do. plus (of course,) i want to stick all sorts of stuff on my phone anyways. Nov 22 04:01:28 i'll do packages for a little while. i'll learn a lot more about the OS and its state while i'm doing it, and then after i get the hang of it, i can look up and see if there is anywhere else I can contribute. Nov 22 04:02:41 i think that sounds like a good plan. Nov 22 04:02:56 yes, indeed Nov 22 04:04:03 ok. sox is on somebody elses todo list at the moment ? Nov 22 04:04:26 I dare to say no Nov 22 04:04:41 a good start Nov 22 04:04:48 lol thanks. i agree. Nov 22 04:05:23 ok, so after i do sox, and give what i did to somebody here, what are they going to do with it after checking it out ? Nov 22 04:05:37 i mean, how are devels checking stuff in ? Nov 22 04:06:02 or whatever the process is. what is the process might be a clear way of asking it. Nov 22 04:06:10 git, or svn, or something called trac ? Nov 22 04:07:13 sorry, you need to ask JaMae or mrmoku|away or TAsn or any of the other devels Nov 22 04:07:33 git is all I know to answer here Nov 22 04:08:50 as a lot of devels are in a european TZ, you probably get more response when asking again in ~8h Nov 22 04:09:12 gonzoism: DocScrutinizer-8: sox is already added.. Nov 22 04:09:18 i just did that :) Nov 22 04:09:25 lol Nov 22 04:10:03 JaMae: so you can walk gonzoism thru the process easily :-) Nov 22 04:10:13 yup but its 5:10 AM here.. Nov 22 04:10:20 so I'll go sleep Nov 22 04:10:28 here too ;-P Nov 22 04:10:37 DocScrutinizer-8: night shift? :) Nov 22 04:10:41 DocScrutinizer-8: or insomnia? Nov 22 04:10:51 #2 Nov 22 04:11:03 or more my usual schedule Nov 22 04:11:38 He just lives in the wrong timezone, that's all. Nov 22 04:11:40 :) Nov 22 04:11:43 ok, i'll let it sit there until one sees it, or ask another day. i think ya'll answered all the questions i should have for today. the git question is really me just looking ahead and trying to get lined up for that. Nov 22 04:11:43 thanks Docscrutinizer and jamae for answering all my questions, ya'll were great. very helpful. Nov 22 04:12:16 yw Nov 22 04:12:45 gonzoism:, DocScrutinizer-8: http://cgit.openembedded.net/cgit.cgi/openembedded/commit/?h=shr/merge&id=9558c730cfae22fa50df607bd435ae5ba90a980d Nov 22 04:13:02 gonzoism, DocScrutinizer-8: that's how easy it is if recipe is alredy in OE repo :) Nov 22 04:13:21 gonzoism: for mg you need to prepare recipe (I don't even know what it is..) Nov 22 04:13:57 JaMae: great. if you can give a clue about how to check out from OE and then what to do with the result, then gonzoism has an easy upstart Nov 22 04:14:38 DocScrutinizer-8: its in wiki.. Nov 22 04:14:53 oh, ok :-) Nov 22 04:16:07 gonzoism: read wiki, start building shr-unstable on your desktop, read http://docs.openembedded.org/usermanual/usermanual.html and after you create and test recipe for mg, send it to shr-devel mailing list Nov 22 04:16:30 gonzoism: shr devs will check it on http://patchwork.dev.bearstech.com/project/SHR/list/ Nov 22 04:17:13 gonzoism: and if no changes needed.. will apply to git and all shr users will have your mg available.. Nov 22 04:17:39 that all I'm off Nov 22 04:18:58 night Nov 22 04:49:40 thanks, i will read all that. and maybe i can make some packages tonight or in the next few days. Nov 22 04:52:39 and its insomnia for me. :) Nov 22 04:59:05 :D Nov 22 04:59:11 06 here Nov 22 04:59:13 kek Nov 22 04:59:21 cet Nov 22 04:59:43 things are looking bright now Nov 22 05:11:43 hm, at least i know i'm not the only one Nov 22 05:20:28 only one? Nov 22 05:20:54 only... one! Nov 22 05:20:56 whos awake at 6 in the morning Nov 22 05:20:58 DocScrutinizer-8: got your n900 yet? Nov 22 05:21:21 nah, still 25th scheduled Nov 22 05:21:29 poobags Nov 22 05:22:08 at least they didn't postpone another time ;-D Nov 22 05:22:36 hehehe Nov 22 05:22:42 just in time for xmas Nov 22 05:22:43 :) Nov 22 05:22:59 hmmm, yes Nov 22 05:23:50 seems a couple of days ago they put the official image download site online Nov 22 05:24:13 and 25th still on webshop Nov 22 05:24:37 so I hope this time they mean it for real Nov 22 05:25:27 they have a working sdk for it? Nov 22 05:25:47 raster: they ship without sw for the FM-radio ;-P Nov 22 05:26:04 (sdk) aiui since a year Nov 22 05:26:12 bah Nov 22 05:26:14 fm radio Nov 22 05:26:16 pffft Nov 22 05:26:19 lol Nov 22 05:26:27 fm is dead Nov 22 05:26:38 all u get is crappy commercial stations Nov 22 05:26:41 and bad quality Nov 22 05:26:42 nah, TMC is great Nov 22 05:26:50 digital radio is already out and kicking Nov 22 05:27:02 and... u have streams for good quality raido poo Nov 22 05:27:17 nothing on fm comes close to di.fm or etn.ch Nov 22 05:27:29 sure Nov 22 05:27:53 just was amused they want to avoid the taxes hussle Nov 22 05:28:50 more quality of content Nov 22 05:28:52 asd well as audio Nov 22 05:29:10 but i can see why they want to avoid the tax shit Nov 22 05:29:11 and actually the FM-tx plus rx make another nice channel for communication Nov 22 05:29:17 the world is changing Nov 22 05:29:26 fm is on its way to its grave Nov 22 05:29:41 the internet is taking over prettymuch all data delivery Nov 22 05:29:43 slowly Nov 22 05:29:47 i get sleepy at sunrise most of the time. Nov 22 05:29:50 radio is easy to do - low bitrate Nov 22 05:30:00 video will take a while due to simply massive data to shift Nov 22 05:30:02 but it will happen Nov 22 05:30:09 but.. yeah Nov 22 05:30:11 i agree Nov 22 05:30:14 fm is nice for comms Nov 22 05:30:20 if u had an fm transmitter inside too Nov 22 05:30:32 i'd say it'd make for a nice local-area data channel Nov 22 05:31:31 that would be cool. seems like you would need a big antanea for more than the neighbors house though. Nov 22 05:32:09 not too different to wifi Nov 22 05:33:00 my isp comes from wifi 6 miles away. Nov 22 05:33:13 that needs special antennae :) Nov 22 05:33:18 he came over today and helped me weld up my gate. Nov 22 05:33:38 ya, those antaneas come in handy. Nov 22 05:33:45 yeah Nov 22 05:34:02 but fm could make for a nice wan with the right transmitters Nov 22 05:34:09 so the plug looking thing underneath the battery, what kind of antanea is that ? Nov 22 05:34:11 tho might just not eb worth it Nov 22 05:42:50 its actually a little jack. looks like the wifi antanea plugin on pcmcia cards Nov 22 05:43:01 also looks like the jack on the side of the phone. Nov 22 05:43:10 do yall know what it goes to ? Nov 22 05:43:13 its on the a6 Nov 22 05:43:14 dunno Nov 22 05:43:16 dont have an n900 Nov 22 05:43:23 i mean the freerunner Nov 22 05:43:27 oh the one on the outside on the gta02? Nov 22 05:43:30 i don't have a n900 either. Nov 22 05:43:40 no the one under the battery on the gta02 Nov 22 05:43:41 not sure what thats used for Nov 22 05:43:50 i stopped worrking about the gta02 > 1 years ago Nov 22 05:43:51 :) Nov 22 05:43:55 err worrying Nov 22 05:43:57 i think the one on the side is a gps antanea. Nov 22 05:44:04 lol, why quit worrying ? Nov 22 05:45:29 year Nov 22 05:45:53 yeah... because... there isnt goign to be a successor from om Nov 22 05:45:58 its the end of a line of hw Nov 22 05:46:15 i hadn't read that. people keep saying it, but i can't find it coming from om. Nov 22 05:46:19 the hw was old when it came out (depending on component from maybe 2 to 6 years) Nov 22 05:46:32 ie compared to when it "peakeD" Nov 22 05:46:37 not "rolled out of r&d" Nov 22 05:46:52 om doesnt need to say anything Nov 22 05:46:57 it has let go almsot all hw staff Nov 22 05:46:59 and software staff Nov 22 05:47:19 it was reduced to skeleton staff - pretty much sean + will and then they hired 1 or 2 people to finish off wikireader Nov 22 05:47:32 om has missed the boat Nov 22 05:47:40 gta02 was queezing in near the end Nov 22 05:47:55 if they had post-haste pushed a gtaq03 out with modern hw - they could have been on it Nov 22 05:48:13 the boat has left. the major manufacturers are now pulling along - eg n900 from nokia Nov 22 05:48:33 out of the box its pretty much as close to open as you're going to realistically get Nov 22 05:48:57 the n900 ? Nov 22 05:49:01 (if u are a purist, you'll never be happy, and i suggest you throw out your pc's and lapotps - they have those nasty "closed" bioses and acpi dst's etc.) :) Nov 22 05:49:17 to an extent we are all parmatic to some extent Nov 22 05:49:20 the nokia n900 Nov 22 05:49:28 have u been living under a rock? Nov 22 05:49:30 :) Nov 22 05:49:40 trust me... and thats just the first of many Nov 22 05:49:43 my friend had one. he loved it. Nov 22 05:49:48 nokia have come clean and admitted symbian will go on their smartphones Nov 22 05:49:55 all maemo (linux) in the future Nov 22 05:49:58 (by 2012) Nov 22 05:50:06 had one? Nov 22 05:50:10 it has yet to ship Nov 22 05:50:16 its right now on the way to its first customers Nov 22 05:50:18 he traded it for a macbook pro Nov 22 05:50:27 canada... Nov 22 05:50:50 he had it a long time ago. maybe 6 months. Nov 22 05:50:55 not sure u are talking of the same thing Nov 22 05:51:00 ya Nov 22 05:51:05 n900 is barely out the door Nov 22 05:51:09 i am. i'm familiar with it. Nov 22 05:51:21 he had one... he loved it. Nov 22 05:51:28 regretted trading it later. Nov 22 05:51:32 only way he got one is if he was a maemo dev and got a pre-release "loaner" from nokia Nov 22 05:51:36 canada got them early for some reason. Nov 22 05:51:36 eh? Nov 22 05:51:41 dont u mean something else? Nov 22 05:51:45 not a loaner. he bought it. Nov 22 05:51:54 not available for sale in canada Nov 22 05:51:59 u'd have to buy from the states Nov 22 05:52:05 look it up. they were in canada like 6 months ago or something. Nov 22 05:52:15 i seriously doubt that Nov 22 05:52:18 nokia n900 Nov 22 05:52:22 yes Nov 22 05:52:24 for real. Nov 22 05:52:28 not n97 Nov 22 05:52:29 or n800 Nov 22 05:52:32 or 810 Nov 22 05:52:36 never heard of those. Nov 22 05:52:39 http://maemo.nokia.com/n900/ Nov 22 05:52:41 that one Nov 22 05:52:42 only have heard about the n900 from him. Nov 22 05:52:44 ya. Nov 22 05:52:59 want me to check my logs ? i can grep them. paste a bit for you. Nov 22 05:52:59 that has been availble for pre-order world-wide only until the last few days Nov 22 05:53:08 maybe he got the model wrong Nov 22 05:53:13 he didn't. Nov 22 05:53:15 but it isnt even for sale in canada Nov 22 05:53:21 its a limtied set of countries Nov 22 05:53:52 lol maybe they gave it to him ! (silly joke) Nov 22 05:53:53 http://conversations.nokia.com/2009/11/10/nokia-n900-begins-shipping/ Nov 22 05:54:00 nov 10 - begins shipping Nov 22 05:54:09 (not quite true.. it shipped a bit later to retailers) Nov 22 05:54:15 nokia didnt give them out Nov 22 05:54:25 about 2 months ago they gave them to attendees of the maemo summit Nov 22 05:54:26 develoeprs Nov 22 05:54:29 and only as loaners Nov 22 05:54:34 they expect to ask for them back Nov 22 05:54:41 so i suspect yoiur friend is talking of somethig else Nov 22 05:54:43 not an n900 Nov 22 05:55:14 no man. he sent me pics of it and stuff. i just greped my logs. they aren't on this machine. i'll check gmail. Nov 22 05:55:26 for real. he had a maemo n900. Nov 22 05:56:19 as i said Nov 22 05:56:32 unless it was a prorotype pre-release developer device Nov 22 05:56:35 it wasnt an n900 Nov 22 05:56:45 if it was - that was a loaner from nokia expected to be given back Nov 22 05:56:56 so selling it was a bad move as he sold something he didnt own Nov 22 05:56:57 6 months ago you could make real money with just a fotofootage of N900 Nov 22 05:57:49 heheh Nov 22 05:57:51 http://en.wikipedia.org/wiki/Nokia_N900 Nov 22 05:57:53 he isn't any kind of developer... and he had one like 6 months ago. traded it for 2 mac things. a macbook pro and something else. maybe an iphone or something. Nov 22 05:57:55 History and availability Nov 22 05:57:56 ... Nov 22 05:57:59 yes, that is the one. Nov 22 05:58:03 The next generation Maemo device was announced on 17 September 2008 during a keynote presentation by Dr. Ari Jaaksi of Nokia. New supported features were announced Nov 22 05:58:09 "announced 17th sep" Nov 22 05:58:14 so.. can't be 6 months ago Nov 22 05:58:17 I guess there were something like 5 N900 pre-pre-pre design studies worldwide 6 months ago Nov 22 05:58:25 It will be available in Canada through independent retailers for around C$730 Nov 22 05:58:39 it was black. Nov 22 05:58:45 nokia dont sell it in canda - they just ship to the standard retail outlets Nov 22 05:58:53 they sell it themselevs in the states (and other european countries) Nov 22 05:59:21 i'm telling you man, he had one. no doubt about it. he made a really big deal about it. i didn't know until like 2 days ago to be impressed. Nov 22 05:59:57 probably N97 Nov 22 05:59:59 i seriously don't believe him Nov 22 06:00:08 must have been the 5800 Nov 22 06:00:09 or n97 Nov 22 06:00:11 not an n900 Nov 22 06:00:18 the n97 and n900 look very similar Nov 22 06:00:22 the 5800 too Nov 22 06:00:24 seriously... he said n900 so many times, it is like burned in my brain. Nov 22 06:00:32 i dont believe it Nov 22 06:00:40 because all facts indicate otherwise Nov 22 06:00:45 lol Nov 22 06:00:53 i have friends who work in nokia reserach who cant even get one Nov 22 06:01:07 i just dont believe it Nov 22 06:01:14 he's bullshitting or has no clue Nov 22 06:01:37 well, want me to pull out my backups and check my drives ? not to prove it to you, but to like show you he really had one ? sure, i could forge pidgin logs, but really, he had one. i can show you the stuff he was saying about it. Nov 22 06:01:38 because.... unless he literally WORKS for nokia and designs the n900 or its softyware - he wont jave had it 6 months ago Nov 22 06:01:52 the n97 and 5800 were definitely out 6+ months ago Nov 22 06:01:59 maybe it was 3 months ago, but i'm thinking longer than that. Nov 22 06:02:08 gonzoism: if he SAIS he had one.. doesnt matter Nov 22 06:02:11 not saying he didnt say it Nov 22 06:02:15 saying he's wrong Nov 22 06:02:20 i mentioned it to my brother 2 days ago, that my friend had one, and he said this: Nov 22 06:02:28 he can say n900 until he's blue in the face... i don't believe it Nov 22 06:03:23 +1 Nov 22 06:03:34 and i don't believe it because i know when the first ones got into the wild (outside of nokia) and what conditiosn there were on it and who those ppl were Nov 22 06:03:56 if your friend sold it... he violated those conditions and nokia will be able to sue him for return of it Nov 22 06:03:57 man he showed it to me on webcams and stuff. showing it off like crazy. Nov 22 06:04:06 and that would assume he got is 2 months ago or so Nov 22 06:04:07 not 6 Nov 22 06:04:15 it could be an n97 Nov 22 06:04:16 as i said Nov 22 06:04:21 the n97 and n900 look very similar Nov 22 06:04:22 VERY Nov 22 06:04:24 and he isn't some kind of developer. he just had one. he always just has stuff. i'm sure about it, n900 Nov 22 06:04:57 http://www.ajun.org/wp-content/uploads/2009/08/331e_Nokia-N97-Mini-vs-N900.jpg Nov 22 06:05:08 i seriously doubt it Nov 22 06:05:17 not 6 months ago Nov 22 06:05:24 not 5 or 4 - maybe 2 Nov 22 06:05:26 maaaaybe Nov 22 06:05:36 well, maybe he like wrote n900 on it somehow or something, and then bragged about it, or whatever. but seriously, he wouldn't just make that stuff up. Nov 22 06:05:37 maybe 2 Nov 22 06:05:39 if he's a journalist he may have gotten a review unit Nov 22 06:05:47 i'm real bad with time. Nov 22 06:05:49 in whihc case Nov 22 06:05:54 he cant say he "regrestted getting it" Nov 22 06:06:03 he didnt choose it - it was for review :) Nov 22 06:06:22 get the photos of it Nov 22 06:06:28 and some timestamps Nov 22 06:06:31 and we'll see Nov 22 06:06:50 first journalist samples were some 3..4 months ago, and this was a BIG sory Nov 22 06:06:53 http://dailymobile.se/wp-content/uploads/2009/09/nokia-n900-vs-nokia-n97-mini-01.jpg Nov 22 06:07:00 next to eachother they are very similar Nov 22 06:07:27 DocScrutinizer-8: yeah. and they wouldnt have been phones u buy and then "regret buying it" Nov 22 06:07:35 it'd have been specially "given" to you Nov 22 06:07:50 i'll see if i can find his phone number. we have known each other for like 10 years or something. Nov 22 06:07:51 and likely with a "you must return it in N months" as opposed to sell it because u "regret getting it" Nov 22 06:07:56 i smell a turd here Nov 22 06:08:05 he never regretted buying it Nov 22 06:08:10 as it makes no sense, and if it doesn't make sense.. it very likely isn't true. Nov 22 06:08:11 :) Nov 22 06:08:26 he regretted trading it for a macbook pro and whatever other kind of apple thing he got for it. Nov 22 06:08:27 oh sorry Nov 22 06:08:32 he regretted TRADING it in Nov 22 06:08:33 sorry Nov 22 06:08:34 :) Nov 22 06:08:45 not in. trading it to his friend. Nov 22 06:09:04 ok, first reports were around 26.5. Nov 22 06:09:04 well a trade-in is similar Nov 22 06:09:10 exchange one item for another Nov 22 06:09:14 http://www.areamobile.de/news/11125-huebscher-als-n97-erstes-foto-zeigt-nokia-internet-tablet-n900 Nov 22 06:09:20 i'll call him. see if he answers. he'll likely have pics. Nov 22 06:09:39 sure Nov 22 06:09:47 pics.. with timestamps Nov 22 06:09:48 :) Nov 22 06:10:01 i actually dont know what nokias policy is on review units Nov 22 06:10:01 http://www.mobilecrunch.com/2009/05/24/exclusive-everything-there-is-to-know-about-nokias-next-tablet/ Nov 22 06:10:15 tho in general industry-wise companies dont give review units to the media Nov 22 06:10:18 timestamps like, the date in the picture, a newspaper or something ? that kind of timestamp ? Nov 22 06:10:18 they loan them with a return date Nov 22 06:10:50 oftne then "ingore" the return date as thgey are too busy with gettign the thing out the door and they are happy to keep the goodwill Nov 22 06:11:05 but trading your review unit for a macbook is likely violating the agreement and as i said Nov 22 06:11:08 nokia could come with a big stick Nov 22 06:11:11 cause i can't do a date in the picture. but meta data or whatever, ya, that would be in the picture. Nov 22 06:11:13 is your friend int he media? Nov 22 06:11:28 gonzoism: well exif data with time will be good Nov 22 06:11:32 he wasn't in any kind of agreement. no man. he is just some guy. Nov 22 06:11:37 gonzoism: or somehting in the photo that indicates date Nov 22 06:11:42 k, i'll look up his number and give him a call. Nov 22 06:11:53 well "just some guy" doesnt get an n900 long before its released Nov 22 06:11:57 you're ion the media Nov 22 06:12:02 a developer or a partner Nov 22 06:12:08 you're in some special position Nov 22 06:12:37 and if in that position- you have strings attached. Nov 22 06:12:57 :) Nov 22 06:13:41 yay i could hear the ringtone Nov 22 06:14:47 well, anyway I hope I'll be one of the first customers to receive their N900 next week Nov 22 06:14:53 DocScrutinizer: pong Nov 22 06:15:07 PaulFertser: :-D morning Nov 22 06:15:37 PaulFertser: would you like to review http://wiki.openmoko.org/wiki/Neo_1973_audio_subsystem#ALSA_Channels Nov 22 06:15:55 and some strange name came up in the display, Dina.D? Nov 22 06:15:56 wtf Nov 22 06:16:00 i was testing from skype Nov 22 06:16:05 calling myself Nov 22 06:16:11 ya ! that happened to me a few minutes ago Nov 22 06:16:14 but i could hear the ring tone Nov 22 06:16:20 :D Nov 22 06:16:36 maybe i can use shr-unstable as daily phone now Nov 22 06:17:28 dina.d Nov 22 06:17:40 lol Nov 22 06:18:51 DocScrutinizer: nice morning :) Nov 22 06:18:57 by the way, how do i turn off the ringtone? don't want to wake up all the people around me Nov 22 06:20:27 fredrin: for now I bet giving /dev/null for ringtone file is the easiest way :-P Nov 22 06:20:48 DocScrutinizer: ok, will do Nov 22 06:21:04 i was just to turn down the ringvolume in settings Nov 22 06:21:07 to 0 Nov 22 06:21:08 :D Nov 22 06:21:55 PaulFertser: I for my life couldn't find what's "playback monomixer volume" #46 (iirc) Nov 22 06:21:55 the second time i called from skype now number showed up Nov 22 06:21:59 just as normal :( Nov 22 06:22:30 lol, check your call log and see if you can find dina.d in there :) Nov 22 06:22:46 could not find it there Nov 22 06:23:02 just null Nov 22 06:23:33 anyway i could both send sms and call myself from skype with sound Nov 22 06:23:46 after latest update && upgrade Nov 22 06:23:51 from win7, lol Nov 22 06:24:58 DocScrutinizer-8: i'm not sure the idea to introduce new non-hardware controls with implicit logic behind those is a good one. Probably it's something that belongs to the user simplified mixer apps. Nov 22 06:25:11 hey, upgrade fron win7 to SHR - great ;-P Nov 22 06:25:29 :) Nov 22 06:27:00 PaulFertser: I'd say gsm mic master volume labled as "playback volume" and appearing on the playback tab of alsamixer is an even worse idea Nov 22 06:27:28 what's the problem? Nov 22 06:28:10 the neo needs a good mixer program Nov 22 06:28:47 aha Nov 22 06:28:54 DocScrutinizer-8: i do not see an immediate difference between the playback controls and record controls given the numerous possibilities offered by the internal routing. Nov 22 06:29:22 look at #4 e.g Nov 22 06:29:45 "mono playback volume" Nov 22 06:30:49 err #5 Nov 22 06:30:56 i like 'music volum' + 'mesage volum' + 'ring volum' + 'varius mic volums' + 'phonecall volume' Nov 22 06:31:31 you'll get all that next step of operation alsa sanitizing masterplan Nov 22 06:31:31 :) Nov 22 06:32:01 in fact it's already been planned exactly this way Nov 22 06:32:25 dedicated softvol featured audio devices for each app Nov 22 06:33:39 sweet Nov 22 06:34:14 DocScrutinizer: (playback mono mix) page 39, register 01h, bit 4 Nov 22 06:34:56 thanks! :-) Nov 22 06:35:04 * PaulFertser --> back to his cozy single bed Nov 22 06:35:31 * DocScrutinizer too Nov 22 06:35:34 finally Nov 22 06:36:21 even starts to dawn outside, eeeew Nov 22 06:38:42 true Nov 22 06:38:50 gnite Nov 22 06:41:02 PaulFertser: larsc talked with broonie (iirc) and seems we all decided on a set of aliases in driver wouldn't hurt anybody Nov 22 06:42:05 the separate controls for headset and speaker are quite mandatory Nov 22 06:42:18 no other sane way to handle that Nov 22 06:44:56 and all the power management and DAPM-* foo is absolutely generic and no pure policy of "just export the raw controls" Nov 22 06:50:46 fuck enlightment crash, gah Nov 22 06:53:43 raster i can't find a single thing about it in my logs... i'll find out anything i can next time i see him online. sent him an email. i'm telling you man, he had one. he isn't any kind of developer. Nov 22 06:58:50 gonzoism: ok... get proof. good proof. :) Nov 22 07:00:12 raster: i don't think he was lying to me. but i'll get the date from him, if nothing else. he ought to have pics around somewhere too. i'll get you what i can. mostly the reason i looked so hard was to blow your mind. :) Nov 22 07:03:02 maybe he has a friend in korea at some nokia fab Nov 22 07:03:48 so maybe "got" one of the PVT Nov 22 07:05:15 canada Nov 22 07:05:36 runs resteraunts Nov 22 07:06:08 gonzoism: i give it like 1 in a 100 that its true - 6 months ago, to have one Nov 22 07:06:16 if he did.. get HOW he got it Nov 22 07:06:24 but my bet is.. its not an n90 Nov 22 07:06:26 err n900 Nov 22 07:08:06 you can tell apart easily N97 and N900 when kbd slides out Nov 22 07:08:12 i might be wrong on the date. i am really really bad with dates Nov 22 07:08:41 i don't think he had a kbd. only touchscreen ? but i wasn't too impressed with it. Nov 22 07:10:08 well.. n900 has a kbd Nov 22 07:10:09 really, i didn't give it much thought. i thought the description was awesome and all, and it was linux.... just everybody seems to be talking about them now Nov 22 07:10:13 so.. no kbd... not an n900 Nov 22 07:10:14 :) Nov 22 07:10:56 ah. maybe had a keybd.. like i said, i didn't scrutinize it. he was the only person i had heard talking about one until like 2 days ago. (thurs) Nov 22 07:11:03 (wed.) Nov 22 07:11:16 lol, so that many days ago. heh. Nov 22 07:11:55 i started telling people my freerunner was coming on friday, and suddenly everybody mentions n900 Nov 22 07:12:02 it isn't a phone, is it ? Nov 22 07:12:21 i'll wikipedia it. Nov 22 07:12:57 maybe he just thought it was a n900 cause he kept calling it a tablet. Nov 22 07:13:26 gonzoism: hahahahah Nov 22 07:13:31 dude. need to do your research Nov 22 07:13:31 :) Nov 22 07:13:46 n900 is the next thing after their web tablets (n70,800,810) Nov 22 07:13:50 err n770 Nov 22 07:13:57 and .. it runs maemo - linux (debian) Nov 22 07:14:19 regular xserver and x apps.. but unlike the previous tablets, it's a full smartphone Nov 22 07:14:24 i knew about the maemo part. Nov 22 07:14:32 with 3g/hspa, smaller 3.5" screen (4.3" on the previous tablets) Nov 22 07:14:35 and smaller Nov 22 07:14:48 and thats what as everyone buzzed Nov 22 07:15:04 as its one of the few thngs that might compete with iphone...and unlike android it's building on a full linux system Nov 22 07:15:11 not just a heavily patches and modded kernel Nov 22 07:15:44 yeah, it being real linux was a big deal. Nov 22 07:16:10 yup Nov 22 07:16:11 sounds nice now. kinda wish i hadn't like slighted him about it. i understand why he was excited. Nov 22 07:16:24 theres a lot of developers just not interested in android Nov 22 07:16:45 they dont want to leanr androids complete custom api's and java and have their code run nice and slowly under a non-jit vm Nov 22 07:16:57 the n900 offers the only viable platform for them at the moment Nov 22 07:16:58 i can understand that. Nov 22 07:17:14 and its a good bit of hw Nov 22 07:17:23 does it run shr? Nov 22 07:18:06 ahahah Nov 22 07:18:07 no Nov 22 07:18:31 can we run mameo ? Nov 22 07:18:41 yes Nov 22 07:18:52 look at mer Nov 22 07:19:00 is it "better" than shr ? Nov 22 07:19:14 depends Nov 22 07:19:16 dunno Nov 22 07:19:23 evaluate yourself Nov 22 07:19:54 ya Nov 22 07:22:18 i dont like its ui Nov 22 07:22:20 but thats me Nov 22 07:22:29 i have special views on ui's Nov 22 07:22:29 :) Nov 22 07:28:10 what are they ? Nov 22 07:28:27 i like very simple and efficient. Nov 22 07:28:59 raster is the inventor of bitmapped display ;-) Nov 22 07:29:49 rather bytemapped Nov 22 07:29:58 hahahah Nov 22 07:30:05 gonzoism: i dont like bloated Nov 22 07:30:13 and i dont like ugly Nov 22 07:31:28 and you like console theme even for desktop ;-) Nov 22 07:31:31 I like small becaue I can know everything on it. Nov 22 07:32:10 with 5pt fontsize, on a 26' screen Nov 22 07:32:53 heh Nov 22 07:33:31 i also hate wasteful ui's Nov 22 07:33:38 eg ones that impost stupidly large font sizes on me Nov 22 07:33:43 DocScrutinizer-8: yup Nov 22 07:33:54 what wm do you use on your desktops and laptops ? Nov 22 07:33:58 hehe Nov 22 07:34:03 i like to be able to say "down font size... down!" and down it goes Nov 22 07:34:23 i like my look ad feel to be fleixble so i get what *I* want Nov 22 07:34:28 i use e Nov 22 07:34:31 i like font sizes large enough that i can read from across the room. Nov 22 07:35:11 u need glasses or contacts? Nov 22 07:35:13 i like e. i use fluxbox everywhere. except the neo. I'm thinking about switching to e so i can know it better because it is awesome for the phone. Nov 22 07:35:26 no. got pretty good sight, i think. Nov 22 07:36:05 just, if i am across the room, or very far away, i want to be able to read what the screen says. Nov 22 07:37:20 i got all sorts of beeps set up. :) in case the monitor is facing the opposite direction or the laptop is closed. Nov 22 07:38:29 gonzoism: i dont like to see it Nov 22 07:38:38 i dont like it that everyone in the room is reading my email Nov 22 07:39:26 don't get me started about beeps. I got a strange harpish sound randomly every second hour or @ay, and I dunno which device is playing it - frives me nuts Nov 22 07:39:45 hahahaha Nov 22 07:41:25 browsers are small and stuff because i don't get live feeds off of them. like i do with logs. Nov 22 07:41:43 and anything printing to the console. Nov 22 07:43:57 in conf/local-builds.inc, what does this entry do: SRC_URI_pn-libframeworkd-phonegui-efl = "file:///path/to/shr-apps" ? Nov 22 07:44:26 shr-unstable/local-builds.inc Nov 22 07:44:42 Should I be changing that to something else ? Nov 22 07:45:39 grr.. shr-unstable/conf/local-builds.inc Last little
 box on  http://www.shr-project.org/trac/wiki/Getting%20started%20developing%20SHR
Nov 22 07:48:10  i understand now.  
Nov 22 08:05:33  how do i get rid of /etc/localtime ?   ls: /etc/localtime: Stale NFS file handle
Nov 22 08:14:41  good morning
Nov 22 08:14:58  gonzoism: stale nfs file handle on a local device is a hint for a broken filesystem
Nov 22 08:14:58  good morning
Nov 22 08:15:03  gonzoism: you should fsck
Nov 22 08:15:24  its /
Nov 22 08:15:34  i can do that.  
Nov 22 08:15:41  gonzoism: nand or uSD?
Nov 22 08:15:46  thanks.  uSD
Nov 22 08:15:56  ah, that's easier then :-)
Nov 22 08:15:59  reboot in nand and then fsck.  :)
Nov 22 08:16:07  or use a uSD card reader
Nov 22 08:58:15  lol, how do you unlock it after you lock E ?
Nov 22 08:58:24  no onscreen keyboard :P
Nov 22 09:01:15  AUX key?
Nov 22 09:02:16  dunno, i crashed X and when it didn't respawn, I rebooted
Nov 22 09:03:40  i was ssh'd in
Nov 22 09:07:26  and now it's locked how?
Nov 22 09:11:57  not locked anymore.
Nov 22 09:12:07  i chose lock screen....
Nov 22 09:12:13  rebooted.  fixed now.
Nov 22 09:12:38  how do you save the settings you set, like telling it not to suspend all the time....
Nov 22 09:14:55  not at all, unfortunately :(
Nov 22 09:15:06  but it's on the TODO, I've been told
Nov 22 09:25:12  bitbake -c clean shr-dialer   gives a lot of 404's with wget... why ?
Nov 22 09:33:20  gonzoism: because the mirrors don't have the sources yet
Nov 22 09:33:31  gonzoism: the sources will then be downloaded from the original site
Nov 22 09:38:23  thanks
Nov 22 09:38:31  sleep time for me.
Nov 22 09:56:33  moin
Nov 22 09:58:12  moin
Nov 22 09:58:22  guten morning!
Nov 22 10:00:01  Dzien dobry
Nov 22 10:03:03  zdravstvujtie!
Nov 22 10:18:15  dos1
Nov 22 10:30:10  *yawn*
Nov 22 10:30:13  morning
Nov 22 10:33:30  moin
Nov 22 10:35:13  dobré ráno všem :P
Nov 22 10:35:50  moin mickeyl 
Nov 22 10:37:00  JaMae: that oops is fixed a while ago, but not brought across onto the 2.6.31 branch yet..
Nov 22 10:38:51  Weiss: great, thank you
Nov 22 10:43:24  TAsn: where are the mockups?
Nov 22 10:43:36  oh, i see
Nov 22 10:44:44  oh, i don't see
Nov 22 10:44:45  :X
Nov 22 10:53:17  mickeyl, a warning from fsodeviced: utilities.vala:121: can't open for writing to /sys/devices/s3c-ohci/usb_mode: No such file or directory
Nov 22 10:53:35  i think it should be /sys/devices/platform/s3c-ohci/usb_mode
Nov 22 10:54:45  and this one:  WARNING **: utilities.vala:129: couldn't write all bytes to /sys/bus/platform/drivers/s3c2440-sdi/unbind (4294967295 of 11)
Nov 22 10:54:51  seems to be -1
Nov 22 10:55:01  i'm somewhat at a loss wrt. these pathsh
Nov 22 10:55:08  kernel folks seem to shuffle them around all the tie
Nov 22 10:55:09  time
Nov 22 10:55:30  can someone confirm that .31 will use the same path as the one SHR is actually using?
Nov 22 10:55:41  s/actually/currently/
Nov 22 10:55:41  mickeyl meant: can someone confirm that .31 will use the same path as the one SHR is currently using?
Nov 22 10:57:32  JaMae, afair you are running .31
Nov 22 10:59:30  mickeyl: i can't, sorry
Nov 22 10:59:59  maybe we should just add sysfs_root or sth in the config so we can change it for every kernel
Nov 22 11:00:10  or a list of nodes
Nov 22 11:02:55  or exloit the get_localized_* functions
Nov 22 11:07:49  morning
Nov 22 11:07:59  JaMa: you here?
Nov 22 11:24:10  mickeyl: i don't think it ever was /sys/devices/s3c-ohci/usb_mode
Nov 22 11:24:17  mickeyl: in SHR it's with platform
Nov 22 11:32:15  TAsn: i also have that problem with rejecting call on resume from suspend
Nov 22 11:32:21  dos1: I've only seen  /sys/devices/platform/s3c-ohci/usb_mode /sys/devices/platform/s3c2410-ohci/usb_mode
Nov 22 11:33:34  larsc: IRQ_GLAMO(GLAMO_IRQIDX_2D) seems to have gone away with 2.6.31.. what should I use instead?
Nov 22 11:44:41  freesmartphone.org: 03mickey 07cornucopia * rf61473fd591d 10/fsodatad/ (9 files in 4 dirs): fsodatad: create better iso3361+dialcode file (handcrafted, with the help of Wikipedia...), add fallback zone.tab
Nov 22 11:47:10  freesmartphone.org: 03mickey 07cornucopia * r34fe2fb88cce 10/fsodeviced/src/plugins/openmoko_powercontrol/plugin.vala: fsodeviced: openmoko_powercontrol: fix path to usb_mode
Nov 22 11:48:19  Please build speex in feed..  Should I open a bug?
Nov 22 11:53:26  JaMae, mrmoku: cookie for ya!
Nov 22 11:59:04  TAsn: http://trac.shr-project.org/trac/ticket/728 invalid or wontfix? He did opkg upgrade...
Nov 22 12:03:03  dos1: If I eat cookies before lunch my wife will spank me ;)
Nov 22 12:04:31  bb after lunch
Nov 22 12:05:24  :)
Nov 22 12:28:20  SHR: 03seba.dos1 07shr-wizard * r4fafff16689f 10/python/shr-wizard: python: use gettext
Nov 22 12:31:56  mickey|breakfast: with newer udev, we have disabled DEPRECATED_SYSFS_PATHS... (automatically from linux.inc) would be pain to reenable and Weiss already changed his glamo to accept newer paths too.. so probably some are changed..
Nov 22 12:34:51  dos1: eating cookie.. thanks
Nov 22 12:49:16  JaMa: no problem as long as someone tells me 
Nov 22 12:52:37  Heinervdm: new fltk apps from Benjamin Schieder can go in? iirc you requested some changes..
Nov 22 12:53:20  JaMa: the recipe needs some cleanup
Nov 22 12:54:09  Please build speex in feed..  Should I open a bug?
Nov 22 12:55:12  Heinervdm: ok.. all?
Nov 22 12:55:42  JaMa: no, only the recipes with additional files
Nov 22 12:55:56  ah now I see something
Nov 22 12:56:08  DEPENDS = "fltk gcc" do we need depends like this?
Nov 22 12:56:19  gcc shouldn't be needed
Nov 22 12:56:21  and is wrong
Nov 22 12:56:29  there is a virtual for it
Nov 22 12:56:38  and also PV is bad
Nov 22 12:56:40  PV = "41"
Nov 22 12:56:44  its svn recipe..
Nov 22 12:56:59  yes
Nov 22 12:57:01  blindcoder: ping
Nov 22 12:57:24  JaMa: but you can commit my gpe-sketchbook cookie
Nov 22 12:57:50  gpe-sketchbook crashes without gpe-icons
Nov 22 12:57:52  Heinervdm: I wanted to ask, why we need this one..
Nov 22 12:58:16  Heinervdm: ah so its not workaround for but i created wit update-alternatives for gpe-icons?
Nov 22 12:58:31  s/for but/for bug/
Nov 22 12:58:32  JaMa meant: Heinervdm: ah so its not workaround for bug i created wit update-alternatives for gpe-icons?
Nov 22 12:58:39  i try to figure out why my touch screen is broken. i wrote a short skrip to dump the coords. here's th result for from top left to bottom right: http://pastebin.ca/1682236
Nov 22 12:58:40  JaMa: there is a bug too
Nov 22 12:58:47  anyone sees sth strange in there?
Nov 22 12:59:03  JaMa: because gpe-sketchbook installes files to /usr/share/pixmaps/gpe
Nov 22 12:59:10  or sth like that
Nov 22 12:59:19  Heinervdm: I know.. its broken by design..
Nov 22 12:59:36  JaMa: and for this one has to be shure that gpe-icons is installbed before gpe-sketchbook
Nov 22 12:59:38  Heinervdm: described here http://patchwork.dev.bearstech.com/patch/468/
Nov 22 13:00:56  JaMa: yes that's broken...
Nov 22 13:01:36  Heinervdm: so should I apply that cookie or do you have some idea how to make it work (update-alternatives)?
Nov 22 13:01:56  JaMa: the gpe-icons are required 
Nov 22 13:02:10  ok
Nov 22 13:02:10  independent of your change
Nov 22 13:02:13  playya__: Tried installing tslib-calibrate and run ts_calibrate  ?
Nov 22 13:02:38  TeLLuS: ts_calibrate won't work for evedv
Nov 22 13:02:46  s/evedv/evdev/
Nov 22 13:02:46  Heinervdm meant: TeLLuS: ts_calibrate won't work for evdev
Nov 22 13:02:49  yes
Nov 22 13:02:57  that's my problem
Nov 22 13:03:18  playya__: we should switch to xf86-input-tslib
Nov 22 13:03:30  then it's working again
Nov 22 13:06:14  freesmartphone.org: 03mickey 07cornucopia * rfb3d3feef0ab 10/fsodatad/ (src/lib/mbpi.vala tests/testworld.vala): fsodatad: read zone.tab and augment country information
Nov 22 13:07:00  Heinervdm: what about cookie 500?
Nov 22 13:07:34  JaMa: i think that it's working
Nov 22 13:08:06  JaMa: i tried it locally and it build an image with atd-over-fso instead of atd
Nov 22 13:08:53  I've seen some problem with atd-over-fso conflicts with atd, but atd-over-fso-dbg was picked as replacement
Nov 22 13:09:10  JaMa: ah yes
Nov 22 13:09:24  there is a small change needed
Nov 22 13:09:39  add _${PN} to the variables
Nov 22 13:18:24  freesmartphone.org: 03mickey 07cornucopia * r67159a4c561a 10/fsodatad/ (NEWS data/iso3361+dialcode.txt src/plugins/world/plugin.vala): fsodatad: iso3361+dialcode: add .vc
Nov 22 13:23:52  mickeyl: if you need a 2.6.31 kernel to try... http://build.shr-project.org/tests/mrmoku/2.6.31/images/om-gta02/
Nov 22 13:24:20  mrmoku: how well is taht working?
Nov 22 13:25:15  mickeyl: no idea... just built it :P
Nov 22 13:25:27  mickeyl: but you can switch between kernels
Nov 22 13:25:43  as modules have a different mpath
Nov 22 13:25:44  -m
Nov 22 13:26:22  my qi doesn't like my sd card
Nov 22 13:26:30  mickeyl: ohh :(
Nov 22 13:26:31  no switching without flashing here
Nov 22 13:26:41  anyways, i'll give it a try eventually
Nov 22 13:26:49  what's the state right now?
Nov 22 13:26:54  mickeyl: as my Qi likes my sd card... I will just try it
Nov 22 13:26:55  mrmoku: Could you or someone with access build and add speex in feed again, or should I open a ticket?
Nov 22 13:26:56  are there any critical problems with fso*d
Nov 22 13:27:03  that destroy the SHR experience?
Nov 22 13:27:21  mickeyl: don't know... my experience is not very good right now :|
Nov 22 13:27:30  though I don't know if it's fso... or something else
Nov 22 13:27:38  hmm
Nov 22 13:27:43  -v please
Nov 22 13:27:50  answering a phonecall when phone was suspended is quite impossible
Nov 22 13:28:06  have to debug some more for more -v :)
Nov 22 13:28:10  okies
Nov 22 13:28:22  latest unstable image shows these problems?
Nov 22 13:28:23  mickeyl: btw. I won't change the sound states
Nov 22 13:28:27  mickeyl: yup
Nov 22 13:28:31  why not?
Nov 22 13:28:46  I compared it to the reference statefile from the alsa wiki
Nov 22 13:28:50  on om
Nov 22 13:28:57  and it is exactly that one
Nov 22 13:28:59  heh
Nov 22 13:29:09  so why is that worse than you are used to?
Nov 22 13:29:28  mickeyl: speaker is tooo loud
Nov 22 13:29:31  and mic to low
Nov 22 13:29:36  people don't hear me :|
Nov 22 13:29:57  guess it is important that we teach our new volume sliders to save the value
Nov 22 13:30:04  right
Nov 22 13:30:10  this way everybody can adjust on the fly while being in a call
Nov 22 13:30:14  and on next call all is fine :)
Nov 22 13:30:34  i never seen a slider for mic sensitivity on any phone though
Nov 22 13:30:37  we're special...
Nov 22 13:30:49  yup, as me
Nov 22 13:31:41  mickeyl: somehow FR hw differs very much
Nov 22 13:31:50  guess that is not like that on any other phone too :P
Nov 22 13:32:30  JaMa: hmm... why is the modules pack for 2.6.31 that small?
Nov 22 13:32:34  just 239K
Nov 22 13:34:41  JaMa: basically modules are just sound + vfat
Nov 22 13:34:53  and I like that :-)
Nov 22 13:49:20  mickeyl: 2.6.31 is usable..
Nov 22 13:49:51  cool. when do we switch?
Nov 22 13:49:56  mickeyl: only gsm is not working and I guess its why you want it :)
Nov 22 13:50:32  mickeyl: switch for default shr-u kernel? definetely no, if you haven't ported gsm 
Nov 22 13:50:36  heh
Nov 22 13:50:39  ah when
Nov 22 13:50:41  no idea
Nov 22 13:50:49  when its ready ;)
Nov 22 13:50:59  is there a detailed document that describes the necessary changes for GSM?
Nov 22 13:51:08  hey, got a problem ^^
Nov 22 13:51:45  what may i do, when the FSO cannot connect after entering pin?
Nov 22 13:53:11  mickeyl: I have no idea, PaulFertser?
Nov 22 13:55:41  JaMa: cookie!
Nov 22 13:56:31  dos1: Could you or someone with access build and add speex in feed again, or should I open a ticket?
Nov 22 13:57:05  dos1: nice, tested?
Nov 22 13:57:24  JaMa: it builds nicely
Nov 22 13:57:24  now installing on neo
Nov 22 13:58:32  dos1: I'll wait for your test and then apply
Nov 22 14:02:55 <[Rui]> TAsn, hi, i fixed the build from a fresh git clone and moved all to the 1st page
Nov 22 14:03:42  mickeyl: one thing that changes for 2.6.31 is s/neo1973/gta02/ for sysfs paths in /sys/bus/platform/devices
Nov 22 14:03:54  !logs
Nov 22 14:03:55  Channel logs for #openmoko-cdevel are archived at:
Nov 22 14:03:55  http://hentges.net/tmp/logs/irc/%23openmoko-cdevel
Nov 22 14:03:56  Live-logs are available at
Nov 22 14:03:58  http://hentges.net/tmp/logs/irc/livelogs/%23openmoko-cdevel.livelog
Nov 22 14:04:01  See ?? help-logs for usage instructions
Nov 22 14:04:20 <[Rui]> so the second page can stay for some details or even disappear completely
Nov 22 14:04:35  mrmoku: interesting. hopefully only for the stuff that's really gta02 specific
Nov 22 14:05:18  root@om-gta02 /sys/bus/platform/devices $ ls -d gta02*
Nov 22 14:05:19  gta02-pm-bt.0     gta02-pm-gps.0    gta02-pm-gsm.0    gta02-pm-wlan.0   gta02-vibrator.0
Nov 22 14:06:33  hmm
Nov 22 14:06:35  that's looks bad
Nov 22 14:06:49  so the Neo1973 now has a gta02-vibrator?
Nov 22 14:09:03  JaMa: it works, apply :)
Nov 22 14:09:24  dos1: righto
Nov 22 14:09:38 <[Rui]> could it be that in a gta01 it's called gta01-vibrator?
Nov 22 14:10:06  freesmartphone.org: 03mickey 07specs * r4531526141f6 10/ (3 files in 3 dirs): org.freesmartphone.Data.World: add GetTimezonesForCountryCode() -> a{ss}
Nov 22 14:11:10  Could one of the feed-admins please build and add speex?  bitbake -b shr-unstable/openembedded/recipes/speex/speex_1.2.bb
Nov 22 14:12:43 <[Rui]> brb
Nov 22 14:15:16  TeLLuS: ok, but buildhost is quite busy atm ;p
Nov 22 14:16:03  dos1: you haven't included checksum.ini change.. bad boy!
Nov 22 14:19:02  freesmartphone.org: 03mickey 07libfso-glib * rad0c114c00f6 10/src/freesmartphone-data.vala: regen
Nov 22 14:19:58  it's kind of pointless calling the same device differently 
Nov 22 14:20:16  mickeyl: just fixing the paths in ti_calypso/modem.py seems to fix gsm
Nov 22 14:20:18  so i don't quite understand these renamings
Nov 22 14:20:39  mrmoku: so it's not yet moved to baremetal gpio control?
Nov 22 14:20:45  :)
Nov 22 14:21:15  where's the tree for 2.6.31? i want to see who was responsible for those name changes
Nov 22 14:21:18  mickeyl: did not do any phonecall yet, but it registered and gets my contacts from SIM
Nov 22 14:21:48  mickeyl: http://git.openmoko.org/?p=kernel.git;a=shortlog;h=refs/heads/om-2.6.31
Nov 22 14:21:50  mickeyl: you might consider to just make those paths configurable...
Nov 22 14:21:51  thanks
Nov 22 14:22:04  mmh :/
Nov 22 14:22:40  JaMa: cookie!
Nov 22 14:22:41 * DocScrutinizer-8 hands mickeyl the whip
Nov 22 14:22:54  TeLLuS: speex built, will be in feed in few minutes
Nov 22 14:23:03  dos1: someone needs to build it before opkg upgrade works :P
Nov 22 14:23:15  Lars-Peter Clausen
Nov 22 14:23:20  larsc: ?
Nov 22 14:23:22  JaMa: tangogps? it's already built and synced :D
Nov 22 14:23:31  dos1: Perfectly
Nov 22 14:23:34  ah you're fast :)
Nov 22 14:23:37  larsc: where's the reasoning shuffling all the device names around?
Nov 22 14:23:46  JaMa: maybe another placebo... but 2.6.31 feels even faster while scrolling
Nov 22 14:24:13  mrmoku: do you know that 2.6.31 in shr/merge is not using DRM at all? :)
Nov 22 14:24:24  mrmoku: i have drm patches here if you want to try it..
Nov 22 14:24:35  JaMa: just try yourself... it is scrolling really good :P
Nov 22 14:24:37  mrmoku: but with drm patches it amazingly fast here..
Nov 22 14:25:01  JaMa: shall we switch drm to 2.6.31? or add yet another kernel build?
Nov 22 14:25:07  fast scrolling on neo? dreams comes true?)
Nov 22 14:25:23  dos1: do I need to eat every cookie? I'm not hungry anymore :P
Nov 22 14:25:26  :)
Nov 22 14:25:46  mrmoku: cookie?
Nov 22 14:25:47  :D
Nov 22 14:25:57  dos1: and I'll be fat if you send me cookie every 5 mins..
Nov 22 14:26:13  you can send some cookies to me
Nov 22 14:26:15  i like cookies
Nov 22 14:26:17  ;)
Nov 22 14:26:26  dos1: but you're not allowed to eat them ;)
Nov 22 14:26:38  and that's the problem :(
Nov 22 14:26:48  mrmoku: see contrib/patchwork/, someone in #oe pointed to it and its good for spammers like dos1 :P
Nov 22 14:27:08  I'll eat that one
Nov 22 14:27:26  and then lunch to eat something with less calories
Nov 22 14:28:29  JaMa: hehe, nice script :)
Nov 22 14:28:48  JaMa: you should have told me that back when you did not yet have commit acces :P
Nov 22 14:28:52  +s
Nov 22 14:28:55  mrmoku: ok, buildhost is now yours :P
Nov 22 14:29:13  :)
Nov 22 14:29:30  mrmoku: I haven't known about it back then.. I know about it from yesterday and already used it about 5 times :)
Nov 22 14:30:06  JaMa: now that I tasted some drops of blood.... I really want to try that 2.6.31 drm kernel ;)
Nov 22 14:30:10  mickeyl: you remember that guy who talked about suspected sabotage inside OM sw-development? Maybe the topic wawn't that off ;-)
Nov 22 14:30:11  TAsn: I've removed x11-16 from the wiki ;-)
Nov 22 14:30:27  freesmartphone.org: 03mickey 07cornucopia * ra89aec17ccf1 10/fsodatad/src/plugins/world/plugin.vala: fsodatad: implement org.freesmartphone.Data.World.GetTimezonesForCountryCode()
Nov 22 14:30:45  mrmoku: wait for Weiss to push last oops fix..
Nov 22 14:30:57  well
Nov 22 14:30:59  JaMa: ok
Nov 22 14:30:59  things can be renamed again
Nov 22 14:31:05  mrmoku: without it, Xorg hangs after few mins
Nov 22 14:31:08  JaMa: give it then to me too! :)
Nov 22 14:31:27  JaMa: will try the standard 2.6.31 for a while then
Nov 22 14:31:32  as it already is blazing fast
Nov 22 14:31:45  feels even faster than the old drm one from yesterday
Nov 22 14:31:58  mrmoku, dos1: we should send few non-alcoholic beers to Thomas for fix push :)
Nov 22 14:32:30  JaMa: or maybe the promissed cake...
Nov 22 14:32:46  though I thinks it's mwester who's always saying "don't search for other reasons as long as you can assume stupidity"
Nov 22 14:32:55 * dos1 would like cake if he was Thomas ;D
Nov 22 14:33:34  mrmoku: yes we should definetely send him something..
Nov 22 14:34:04  cake with chocolate glamo on it!
Nov 22 14:34:55 * DocScrutinizer-8 adds a few 2410 and other starnge crunchy chips
Nov 22 14:36:07  http://people.openmoko.org/joerg/unsorted/impressions-of-25C3/impressions-of-25c3/snc00244_jpg.jpeg.html
Nov 22 14:36:09  DocScrutinizer-8: sent a mail to openmoko-kernel
Nov 22 14:37:03  seems I have to dare a glance to that folder of kmail then ;-P
Nov 22 14:37:59  actually it's one of the "clean" folders... So no new mail yet
Nov 22 14:39:10  mickey|away: much too polite ;-D
Nov 22 14:39:34  what for do you have a capslock? :-P
Nov 22 14:40:17 * DocScrutinizer googles to learn what's git-blame - sounds interesting
Nov 22 14:42:30  JaMa, mrmoku: what about splash screen? i heard that still -tux theme is selected, instead of -logo
Nov 22 14:42:51  DocScrutinizer: btw svn have svn praise as well :) as svn blame seems offending some developers :)
Nov 22 14:44:18  TeLLuS, do open a ticket with regard to  speex
Nov 22 14:44:38  u meqn libspeex? for what app?
Nov 22 14:44:42  DocScrutinizer: what exactly are you doing with those chips?
Nov 22 14:45:07  usually  apps should specify their lib dependencies
Nov 22 14:45:08  feed my cat?
Nov 22 14:45:21  DocScrutinizer: pretty expensive cat food :P
Nov 22 14:45:36  nah, not exactly
Nov 22 14:46:26  saved them from the dump - a huge heap of junk in front of OM-office door
Nov 22 14:46:42  mhm
Nov 22 14:47:05  actually I got lots and lots more of them
Nov 22 14:47:14  mrmoku, some reports that ffalarms fails to resume on alarm
Nov 22 14:47:14  TeLLuS: dos1 already send patch for that..
Nov 22 14:47:39 * JaMa reports that no alarm is shown in ffalarms screen but it still wakes him up
Nov 22 14:47:43  DocScrutinizer: if they weren't BGA, I'd say "send some over for me to play with"
Nov 22 14:47:43  so resume on some events seems to be broken...
Nov 22 14:48:37  md2k7: most is BGA, but I also have a kilo of transistors (bc338 or sth)
Nov 22 14:48:57  probably xx.000
Nov 22 14:49:05  DocScrutinizer: if you don't need 'em, tell me :P
Nov 22 14:49:14  JaMa, some say, it doesn´t wake up, but you get the alarm on next manual resume...
Nov 22 14:49:24  as well as some 50..100 other 2+3pin parts
Nov 22 14:49:30  mrmoku: yeah read that (ffalarms)
Nov 22 14:49:32  types of
Nov 22 14:49:58  dos1: we can leave it for unstable... but for testing we should use the logo splash
Nov 22 14:50:06  some only 200 in a bag, some a whole reel
Nov 22 14:50:30  mrmoku: if you want different splashes, then for unstable i would like dontpanic one :P
Nov 22 14:50:41 * md2k7 is running "The Makefile", which is cloning git from omapzoom.org with astonishing 9 KB/s
Nov 22 14:51:14  md2k7: I'd love to share this crap
Nov 22 14:51:33  dos1: I don't care for unstable... just for the more stable images we should use log
Nov 22 14:51:42  we can switch to logo for unstable too
Nov 22 14:52:02  DocScrutinizer: you could also put it into "the bay" to make money with it
Nov 22 14:52:15  md2k7: removing BB_GIT_CLONE will help you
Nov 22 14:52:16  DocScrutinizer: but I'll take transistors at least, if possible^^
Nov 22 14:52:34  won't even yield the package and postage
Nov 22 14:52:39  JaMa: don't I need source from that repo for building?
Nov 22 14:52:42  JaMa: is sound one of the missing things in 2.6.31?
Nov 22 14:53:11  DocScrutinizer: charge that separately
Nov 22 14:53:42  mrmoku: no, probably not autoloaded because module name change?
Nov 22 14:54:08  mrmoku: but I had to add workarround for gcc-4.4 bug in sound module :/
Nov 22 14:54:46  JaMa: even after manually loading the module... no sound
Nov 22 14:54:56  btw. there they dropped gta02 in the module name :P
Nov 22 14:54:57  md2k7: it depends what are you building, but if you're cloning it just because of SRCPV+BB_GIT_CLONE then no
Nov 22 14:55:06  snd_soc_neo1973_wm8753
Nov 22 14:55:43  mrmoku: I always had to restore alsastate after module load
Nov 22 14:56:07  mrmoku: that's a good example of a reasonable naming - the audio has small diffs between gta01 and 02
Nov 22 14:56:28  JaMa: comment out BB_GIT_CLONE_FOR_SRCREV = "1"?
Nov 22 14:56:50  JaMa: aplay: main:608: audio open error: Connection timed out
Nov 22 14:56:56  it's not just that it's silent
Nov 22 14:57:33  md2k7, yes
Nov 22 14:57:41  conection timed out??? WTF?
Nov 22 14:57:59  only disadvantage is that you´ll get different version numbers
Nov 22 14:58:08  DocScrutinizer: the two lines before are:
Nov 22 14:58:08  ALSA lib pcm_direct.c:1047:(snd1_pcm_direct_initialize_slave) unable to start PCM stream
Nov 22 14:58:11  ALSA lib pcm_dmix.c:1020:(snd_pcm_dmix_open) unable to initialize slave
Nov 22 14:58:28  so you can´t mix and match local stuff with pkgs from the feed
Nov 22 14:58:30  mrmoku: btw RP explained why we hit so many gits with BB_GIT_CLONE_FOR_SRCREV + SRCPV..
Nov 22 14:58:44  ml?
Nov 22 14:58:45  RP?
Nov 22 14:58:50  :P
Nov 22 14:59:13  mrmoku: its like having all set to AUTOREV, because without BB_GIT_CLONE_FOR_SRCREV or SRCPV, it finds sane SRCREV and its enough for PV
Nov 22 14:59:38  mrmoku: hmm, tat might be a crappy dmix def in some alsa cfg file (runtime)
Nov 22 15:00:03  mrmoku: but with SRCPV and BB_GIT_CLONE_FOR_SRCREV you need to git clone just to count thouse rev-list changes
Nov 22 15:00:16  DocScrutinizer: asound.conf: http://shr.pastebin.com/mae2d3df
Nov 22 15:00:25  DocScrutinizer: is that crap?
Nov 22 15:00:45  JaMa: what would happen if we turn that off?
Nov 22 15:00:47  it's a stub
Nov 22 15:00:54  mrmoku: and there is my first bitbake patch in bitbake/master and RP worked on 2nd a bit already
Nov 22 15:01:06  the actual dmix def obviously somewhere else
Nov 22 15:01:15  DocScrutinizer: ahh, ok will search
Nov 22 15:01:21  mrmoku: we won't have consistent PV naming between buildhost and local builds
Nov 22 15:01:31  prolly /usr/share/*/*/alsa or sth
Nov 22 15:02:05  mrmoku: but as long as buildhost started with BB_GIT_CLONE I guess it will have always highest PV which is "good"
Nov 22 15:02:52  mrmoku: localbuilders needs to download sqlite cache from build.shr if then want to start with same PVs or even build newer revisions with higher PV than buildhost has in feeds
Nov 22 15:02:54  JaMa: we already tell everybody to turn it off
Nov 22 15:03:00  mrmoku: what's result if you add -vv to the aplay call?
Nov 22 15:03:15  mrmoku: its terrible load on upstream git repos..
Nov 22 15:03:23  DocScrutinizer: same result... just those three lines
Nov 22 15:03:34  mrmoku: as I said before.. I would disable it in makefile for users
Nov 22 15:03:39  yeah I expected this
Nov 22 15:03:47  :-/
Nov 22 15:03:55  mrmoku: and you can stay with it only on buildhost which has all sources already cloned
Nov 22 15:03:55  mrmoku, then the makefile should have it commented out
Nov 22 15:04:05  or is it?
Nov 22 15:04:27  JaMa: well... it is trying to re-get stuff though
Nov 22 15:04:36  spaetz: not yet, will do
Nov 22 15:04:57  mrmoku: failing re-get stuff isn't as bad as cloning all git world :)
Nov 22 15:05:32  mrmoku: but maybe we can advertise our builders as git world mirrors :P
Nov 22 15:05:53  :)
Nov 22 15:06:03  hehe
Nov 22 15:06:53  can´t we get date-based version numbers :-)
Nov 22 15:06:55  SHR: 03mok 07shr-makefile * rf4a6fefb8c99 10/Makefile: Makefile: comment out BB_GIT_CLONE_FOR_SRCREV when setting up local.conf
Nov 22 15:07:17  hehe, dos1 started to build with -b because parsing does not work :|
Nov 22 15:07:31  mrmoku: what is failing?
Nov 22 15:07:54  restarting feed to see what
Nov 22 15:08:55  I'll update navit srcrev as soon as I found which srcrev broke it :)
Nov 22 15:09:05  DocScrutinizer: there is /usr/share/alsa with lots of complicated stuff :P
Nov 22 15:09:29  dos1: you cheater1
Nov 22 15:09:30  !
Nov 22 15:10:50  JaMa: fatal: Not a valid object name 1
Nov 22 15:10:54  for moblin
Nov 22 15:11:04  -menus
Nov 22 15:11:26  and parsing then will fail... after waiting for ages :|
Nov 22 15:12:33  atd-over-fso : http://shr.pastebin.ca/1682353 this is right fix?
Nov 22 15:13:06  mrmoku: ah I guess its after my last bitbake patch update which is dooing self.go again :/
Nov 22 15:13:56  mrmoku: do you have some more verbose output? I have no idea how to fix this properly..
Nov 22 15:15:20  hmm tangogps 0.99 looks great
Nov 22 15:16:14  JaMa: http://shr.pastebin.com/m5fe6a1d7
Nov 22 15:16:17  that is the start
Nov 22 15:17:54  JaMa: scrolling around in tango is a hell lot faster with 2.6.31 too :)
Nov 22 15:18:44  mrmoku: as quickfix you can remove self.go() in git.py         if not os.path.exists(repodir): self.go(None, ud, d)
Nov 22 15:19:01  ok
Nov 22 15:23:12  zoff99: why do you export LC_ALL and then unset it?
Nov 22 15:23:30  zoff99: we should vote about requesting Display resource for navit :)
Nov 22 15:23:50  JaMa: yes true
Nov 22 15:24:04  zoff99: imho half people is using navit in car (where display is needed and battery can be charged)
Nov 22 15:24:39  zoff99: and other half is using it while walking, biking where battery should be saved as much as possible and taping display from time to time is not a problem at all
Nov 22 15:24:47  JaMa: what is that u entered in .bb file: speechd ?
Nov 22 15:24:58  JaMa: why not make an option for that?
Nov 22 15:25:10  zoff99: or we can provide 2 .desktop files with/without Display
Nov 22 15:25:20  pff, make a shr-setting for navit resources :-)
Nov 22 15:25:35  spaetz: you should try 2.6.31... it is real fun :D
Nov 22 15:25:38  yea, why not make an 'shr daemon' which could set preferences for any app
Nov 22 15:25:43  (apart from sound and fso not working)
Nov 22 15:25:58  md2k7: setting that option somewhere in /etc/ seems as much work as editing launcher file, but you're right that /etc/ file won't be rewritten during navit upgrade
Nov 22 15:26:05  Jama: i was thinkg more like navit-theme-bike
Nov 22 15:26:06  Jama: i was thinkg more like navit-theme-car
Nov 22 15:26:08  ??
Nov 22 15:26:09  mrmoku: I got here /usr/share/alsa/pcm/dmix.conf, and that doesn't give a goot clue why the error may throw
Nov 22 15:26:21  mrmoku, sounds cool. i´d love to fix up a bunch of stuff first though
Nov 22 15:26:34  autoload-module stuff, eg
Nov 22 15:26:36  mrmoku: SSH is working?
Nov 22 15:26:38 * JaMa wants neural interface for apps settings..
Nov 22 15:26:41  dos1: yup
Nov 22 15:26:48  who want's to implement that?
Nov 22 15:26:50  but 2.6.31 sounds good
Nov 22 15:26:52  Jama: the you can do "opkg install navit navit-locale-de navit-theme-car"
Nov 22 15:26:57  mrmoku: why is *scrolling* faster with a new *kernel*?
Nov 22 15:27:07  I'll wait for release candidate, no more alpha/beta to my brain
Nov 22 15:27:08  dos1: ssh, x and after fixing paths in ogsmd even gsm
Nov 22 15:27:18  :)
Nov 22 15:27:24  md2k7: asking the wrong one here :)
Nov 22 15:27:27  mrmoku: give it to me! :D
Nov 22 15:27:29  no idea... 
Nov 22 15:27:34  mrmoku: is it that one from tests/kms?
Nov 22 15:27:36  err
Nov 22 15:27:42  s/kms/2.6.31/
Nov 22 15:27:48  dos1: get it yourself :P
Nov 22 15:27:48  dos1: yes..
Nov 22 15:27:48  http://build.shr-project.org/tests/mrmoku/2.6.31/images/om-gta02/
Nov 22 15:28:02  mrmoku: so howzbout efl and gtk lists?
Nov 22 15:28:08 * JaMa is glad that persuaded mrmoku to build alternate kernels :P
Nov 22 15:28:09  downloading ;)
Nov 22 15:28:13  these are the particular ones that annoy me most
Nov 22 15:28:17  md2k7: what do you mean?
Nov 22 15:28:21  speed?
Nov 22 15:28:25  if scrolling speed has improved
Nov 22 15:28:28  md2k7,  graphic driver improvements
Nov 22 15:28:33  well I just have efl lists... and there is a big difference
Nov 22 15:28:36  mrmoku: btw even the terse
Nov 22 15:28:38  zoff99: some people are maybe using both car/bike
Nov 22 15:28:40  pcm.!default {
Nov 22 15:28:42      type plug
Nov 22 15:28:43      slave.pcm "dmix"
Nov 22 15:29:05  is flaky, as it fails when you try to do arecord from default device
Nov 22 15:29:07  zoff99: if you create some nice icon for navit/car and navit/bike than we can provide 2 .desktop files for that :)
Nov 22 15:29:10  spaetz: as, for example, the 'epdfview' scrolling is ultimately smooth, i can't imagine display driver is the problem
Nov 22 15:29:21  zoff99: with medium icons I have planty of space :)
Nov 22 15:29:24  DocScrutinizer: alsamixer tells me Card: neo1973gta02
Nov 22 15:29:26  but i still would do it as different packages
Nov 22 15:29:34  why?
Nov 22 15:29:41  is there an easy way to port my existing contacts,messages from old opim (before current face) to current one?
Nov 22 15:29:43  spaetz: I rather think that the rendering shouldn't be done every frame, but that the parts already rendered should be kept in memory
Nov 22 15:29:45  mrmoku: hmm, ok
Nov 22 15:29:50  zoff99: only difference is one fso resource sofar
Nov 22 15:29:55  at least, that's what i think could make scrolling faster
Nov 22 15:30:43  sorry guys, bbl
Nov 22 15:30:49  DocScrutinizer: ok, thanks
Nov 22 15:31:07  md2k7: I was thinking of doing a little demo of exactly this. the downside is that it's a lot of platform-specific code. the upside is that it'll be REALLY fast
Nov 22 15:33:00  shouldn't frameworkd-conf-shr be renamed to frameworkd-conf-om-gta02?
Nov 22 15:33:25  playya__: why?
Nov 22 15:33:37  it's providing different config for different architectures
Nov 22 15:34:39  md2k7: raster (validly) doesn't like such platform-specific code. but the power is there in the FR to be used if we choose to...
Nov 22 15:34:39  dos1, look at shr-devel
Nov 22 15:34:42  depending on the $MACHINE?
Nov 22 15:34:44  he answered to my mail.
Nov 22 15:35:01  mrmoku: booting 2.6.31 ;)
Nov 22 15:35:22  mrmoku: try "aplay -D hw:0.0" and "aplay -D plughw:0.0..." and -D dmix
Nov 22 15:35:35  mrmoku, did you see JaMa fixed the kernel modules stripping?
Nov 22 15:35:42  btw, mrmoku bumped efl?
Nov 22 15:35:45  Weiss: and I don't like laggy flick lists, to be honest :P I don't care about platform-specific code. it just needs to be included in a 'clean way'
Nov 22 15:35:56  TAsn: not pushed sofar..
Nov 22 15:36:02  JaMa, :(
Nov 22 15:36:09  JaMa, bug them at oe-devel!
Nov 22 15:36:20  TAsn: I did about 5AM :)
Nov 22 15:36:24  mrmoku: also check your "statefiles" (fsck!) have the right DAImode
Nov 22 15:36:47  JaMa, :)
Nov 22 15:36:52  DocScrutinizer-8: I compared the statefile to the reference one from the wiki
Nov 22 15:37:00  (well trac ticket linked from wiki)
Nov 22 15:37:01  TAsn: then I was sleeping till 8AM and guess what? nobody answered between :)
Nov 22 15:37:14  JaMa, I went to sleep at 6am
Nov 22 15:37:18  and woke up at 7am
Nov 22 15:37:21  I sold my car. ;(
Nov 22 15:37:27  another one?
Nov 22 15:37:40  :P
Nov 22 15:37:44  that one.
Nov 22 15:37:46  :)
Nov 22 15:37:55 * JaMa still wants some not so expensive BMW5 (GT)
Nov 22 15:39:27  mrmoku, btw, dos did the shr-wizard utility :)
Nov 22 15:39:37  I'm almost ready to push it to image :)))
Nov 22 15:39:47  mrmoku: (compared) I thought they changed completely
Nov 22 15:39:50  hmm
Nov 22 15:39:56  really seems faster
Nov 22 15:40:05  DocScrutinizer-8: no, just less verbose
Nov 22 15:40:27  wtf means *that*?
Nov 22 15:40:27  DocScrutinizer-8: 1:'PCM Volume':2:0,0
Nov 22 15:40:42  DocScrutinizer-8: ommitting the enums
Nov 22 15:40:52  dos1, stable? no wsod?
Nov 22 15:40:53  mrmoku: feed building?
Nov 22 15:40:56  every control in just one line with the respective values
Nov 22 15:41:06  JaMa: NOTE: Parsing finished. 7055 cached, 487 parsed, 297 skipped, 2 masked.
Nov 22 15:41:13  ERROR: Parsing errors found, exiting...
Nov 22 15:41:23  well, have fun validating it
Nov 22 15:41:35  DocScrutinizer-8: I had fun :P
Nov 22 15:41:48  opened old and new in vim and compared
Nov 22 15:41:51  value per valuer
Nov 22 15:41:52  mrmoku: ah :/
Nov 22 15:41:56  seem fun just started
Nov 22 15:42:24  brb
Nov 22 15:42:24  how to do try {} catch in python? :)
Nov 22 15:42:56  what a utterly useless effort
Nov 22 15:42:57  JaMa, 
Nov 22 15:42:59  try:
Nov 22 15:43:01  except
Nov 22 15:43:09  *except:
Nov 22 15:43:24  TAsn: can you do it in bitbake git.py around self.go?
Nov 22 15:43:48  JaMa, I don't think I get you
Nov 22 15:43:57  give me line numbers you want to ignore errors for.
Nov 22 15:44:03  mmt I'll show you
Nov 22 15:47:36  TAsn: dunno, didn't tested carefully - i'll wait for those new patches
Nov 22 15:48:17  [Rui], 
Nov 22 15:48:25  ***Error***: You must have automake >= 1.7 installed
Nov 22 15:48:26  TAsn: something like this? http://jama.homelinux.org/org.openembedded.shr/bitbake/0003-git-fetcher-catch-exceptions.patch ?
Nov 22 15:48:30  I got 1.11 ;|
Nov 22 15:48:36  mrmoku: try this one ^^
Nov 22 15:48:37 <[Rui]> hi
Nov 22 15:48:53 <[Rui]> TAsn: hms...
Nov 22 15:48:53  JaMa: yes, something like that
Nov 22 15:48:55  JaMa, that looks good.
Nov 22 15:49:11 <[Rui]> TAsn: so do I
Nov 22 15:49:17  weird.
Nov 22 15:49:51  JaMa, looks good :-)
Nov 22 15:50:02 <[Rui]> TAsn: that seems to come from autogen.sh: REQUIRED_AUTOMAKE_VERSION=1.7 exec gnome-autogen.sh $@
Nov 22 15:50:13 <[Rui]> maybe it isn't really needed
Nov 22 15:50:32  error: patch failed: lib/bb/fetch/git.py:162
Nov 22 15:50:32  error: lib/bb/fetch/git.py: patch does not apply
Nov 22 15:50:35  JaMa: ^^
Nov 22 15:50:41  spaetz: not sure if that can catch exceptions from os.run 
Nov 22 15:50:52  mrmoku: did you remove first self.go?
Nov 22 15:50:58  JaMa: I did reset --hard
Nov 22 15:50:58  [Rui], no idea.
Nov 22 15:51:01  mrmoku: as that quickfix before?
Nov 22 15:51:03  and reapplied 1+2+3
Nov 22 15:51:29 <[Rui]> TAsn: I suggest removing it, I will too...
Nov 22 15:51:40  sure, it will throw an exception too
Nov 22 15:51:43  [Rui], doesn't work.
Nov 22 15:51:52  JaMa: after doing the quickfix... I did git reset --hard and reapplied all three patches
Nov 22 15:51:59  dos1: you like the kernel? :)
Nov 22 15:52:11  will just regenerate everything :|
Nov 22 15:52:19  mrmoku: ah ok.. try again.. with 000* from jama../bitbake
Nov 22 15:52:25  maybe it's not amazing fast, but it's faster ;)
Nov 22 15:52:44  JaMa: ok
Nov 22 15:52:46 <[Rui]> TAsn: maybe you need to clone it again (quicker than finding all autogen generated files)
Nov 22 15:53:02  [Rui], clone what?
Nov 22 15:53:13  ahh, actually os.run will not always throw an exception
Nov 22 15:53:31 <[Rui]> TAsn: I've pushing that change to gitorious
Nov 22 15:53:34 <[Rui]> done
Nov 22 15:53:55 <[Rui]> but I did a git clone on another dir and it compiled ok
Nov 22 15:53:57  JaMa: ok, applied... building now
Nov 22 15:54:03  bbiab
Nov 22 15:54:12 <[Rui]> I would like to check it out on the neo, but my build tree is stuck on gcc
Nov 22 15:54:56  spaetz: all I know its http://shr.pastebin.com/m5fe6a1d7 and its from self.go :)
Nov 22 15:55:06 <[Rui]> TAsn: "git clone" again
Nov 22 15:55:12  spaetz: we'll see if it works on buildhost ;)
Nov 22 15:55:17 <[Rui]> permission denied on mmap?? http://pastebin.com/d24e4d9a9
Nov 22 15:55:54  question: how can i turn of "auto-completion"?
Nov 22 15:56:08  [Rui], doesn't work.
Nov 22 15:56:40  [Rui]: cat /proc/sys/vm/mmap_min_addr ?
Nov 22 15:57:28  [Rui]: echo 0 there if you know what you're doing (potentional security problem) but not so bad...
Nov 22 16:02:03  JaMa (et al.): new 2.6.31+DRM pushed. this time it's rebased against the latest om-2.6.31, and it should have everything that the 2.6.29 version had, although I've found a modesetting problem already. should let X run nicely though.. let me know...
Nov 22 16:02:18  ok [Rui] found what's wrong.
Nov 22 16:03:45  Weiss: ok, will do
Nov 22 16:04:16  [Rui], had a missing package
Nov 22 16:05:11  JaMa: tell me when you pushed :D
Nov 22 16:05:16  will build it then
Nov 22 16:05:22  [Rui], sweet, works.
Nov 22 16:05:46  mrmoku: sure
Nov 22 16:06:26  JaMa: From git://git.moblin.org/moblin-menus * branch            master     -> FETCH_HEAD
Nov 22 16:06:30  NOTE: Creating tarball of git repository
Nov 22 16:06:32  fatal: Not a valid object name 1
Nov 22 16:06:35  ERROR: GIT repository for git://git.moblin.org/moblin-menus.git;protocol=git doesn't contain revision 1, cannot be update
Nov 22 16:06:51  JaMa: incidentally, what do you do for TS calibration? at the moment, I do a load of "echo"s into sysfs on starting the X server..
Nov 22 16:08:45  Weiss: I guess we took your echos :)
Nov 22 16:09:23  ah :)
Nov 22 16:09:38  you won't want to be told how I worked those numbers out, then..
Nov 22 16:10:09  Weiss: yes the same
Nov 22 16:10:26  mrmoku: and then it continues?
Nov 22 16:10:36  PaulFertser: ping, do you have working sound on 2.6.31?
Nov 22 16:10:44  JaMa: still parsing :P
Nov 22 16:11:12  mrmoku: but this message is when it failed.. so I guess should be better
Nov 22 16:11:13  Weiss: well, at least they seem to work :P
Nov 22 16:11:38  JaMa: it will finish parsing before it fails, no?
Nov 22 16:11:44  at leas that is my experience :|
Nov 22 16:11:46  +t
Nov 22 16:12:01  [Rui], mind doing another fix for sss?
Nov 22 16:13:17  [Rui], maybe give each "network" an info line of it's own just beneath it (or something)
Nov 22 16:13:37  and Ringing profile should be text outside of the hoversel
Nov 22 16:13:51  mrmoku: if error is fatal, what you pasted is just error I added when exception catched
Nov 22 16:14:01  ahh ok
Nov 22 16:14:06  as the hoversel will start with the correct profile there.
Nov 22 16:14:13  but other than that, super cool. :)
Nov 22 16:14:16  will add it asap :)
Nov 22 16:14:24  (asap = tomorrow evening)
Nov 22 16:14:37 <[Rui]> TAsn: ah... ok
Nov 22 16:14:57  JaMa: it passed parsing
Nov 22 16:14:58  JaMa: looks good
Nov 22 16:15:00  :)
Nov 22 16:15:05  now it's NOTE: Resolving any missing task queue dependencies
Nov 22 16:15:08  :)
Nov 22 16:15:19 <[Rui]> ok
Nov 22 16:15:21  dos1: apps start faster with 2.6.31
Nov 22 16:15:24  a lot faster :)
Nov 22 16:15:33  dos1, git clone git://gitorious.org/sss/sss.git
Nov 22 16:15:35  and build
Nov 22 16:15:41  tell us if you have any comments
Nov 22 16:16:47 <[Rui]> dos1: and I don't mind being beaten with an iron-stick (oh yes please do ;) )
Nov 22 16:18:53  [Rui]: why for you need gconf and gnome-common? :x
Nov 22 16:19:23 <[Rui]> dos1: most likely something inherited from elmdentica (from which I copied the auto* stuff)
Nov 22 16:19:34  Good morning everybody.
Nov 22 16:19:51  are you planning to rebuild a new image with last errors solved?
Nov 22 16:20:09  yup, soon Sharwin :)
Nov 22 16:20:22  ok, great! :)
Nov 22 16:20:30  waiting for it :)
Nov 22 16:20:44  TAsn: hurry up with the wizard :)
Nov 22 16:22:20  ok, dinner time
Nov 22 16:22:22  bbl
Nov 22 16:23:55  [Rui]: add slider for brightness
Nov 22 16:24:07 <[Rui]> dos1: ah good
Nov 22 16:24:23  well, all i need in "quick settings" is - profile, brightness, autodimming and autosuspend
Nov 22 16:24:34 * Weiss wonders if we're going to be ready with 2.6.31 just in time for 2.6.32 to arrive :)
Nov 22 16:24:35  dos1, no.
Nov 22 16:24:44  autodimming and autosuspend
Nov 22 16:24:47  should not be there
Nov 22 16:24:59  they are general configurations
Nov 22 16:25:07  hmm? no
Nov 22 16:25:18  general configuration is in timeouts
Nov 22 16:25:19  i.e you don't change them all the time
Nov 22 16:25:26  and i don't mean timeouts
Nov 22 16:25:32  then?
Nov 22 16:25:36  what do you want to change?
Nov 22 16:25:38  i mean that module which is under "Power" button in shr-settings
Nov 22 16:25:39  dos1: I've just updated and tried to phone myself. I've set an .mp3 file as a ringer but there's no sound at all
Nov 22 16:25:43  with two sliders
Nov 22 16:25:53  autodimming on/off, autosuspend on/off
Nov 22 16:26:01  Q-Master: supposed. use .wav
Nov 22 16:26:04  oh
Nov 22 16:26:07  dos1, good idea.
Nov 22 16:26:09  :)
Nov 22 16:26:21  dos1: it worked till new unstable. 8(
Nov 22 16:26:36  [Rui], that's a good idea ^
Nov 22 16:26:45  TAsn: i had shr-settings invocation for long AUX press some time ago
Nov 22 16:26:50  it was sloooow, but worked :P
Nov 22 16:26:54  hehe
Nov 22 16:26:56  this will be
Nov 22 16:26:58  scary fast :)
Nov 22 16:27:03  pop on click
Nov 22 16:27:36  TAsn: run it:
Nov 22 16:27:38  shr-settings "AUX menu" shr_profile.Profile shr_display.Display shr_pm.Pm
Nov 22 16:27:47 <[Rui]> two toggles for autodimming and auto suspend?
Nov 22 16:28:08  that's what i was using ;)
Nov 22 16:28:11 * [Rui] is adding the individual status areas for networks, will do that next or after he makes his return trip home
Nov 22 16:29:50  ring-tone - Playing: Illusion.wav.mp3
Nov 22 16:29:51  [Rui], cool :)
Nov 22 16:29:51  org.freesmartphone.Device.Audio.PlayerError: Can't play song /usr/share/sounds/Illusion.wav.mp3: File or data corrupt
Nov 22 16:29:58  suxx
Nov 22 16:30:13  Q-Master: supposed...
Nov 22 16:30:16  Do any of you know what's going on with Tichy?
Nov 22 16:30:22  [Rui], no hurry, I won't be able add it until tomorrow night anyway.
Nov 22 16:30:24  Q-Master: libcanberra-alsa supports only wave files
Nov 22 16:30:28  Q-Master: really, convert it to wave
Nov 22 16:30:32  btw [Rui] about profile changing
Nov 22 16:30:46  Q-Master: you'll get improvement in speed
Nov 22 16:30:46  Like, where the public git archive is actually hosted?
Nov 22 16:30:47  [Rui], I like the way dos1 did it in shr-settings, think you can imitate it?
Nov 22 16:30:53  dos1: gstreamer is not supported anymore?
Nov 22 16:31:12  Q-Master: there is some libcanberra-gstreamer, but not tested
Nov 22 16:31:15  dos1, lol at you stitching this settings page :)
Nov 22 16:31:16  you can try to play with it ;;p
Nov 22 16:31:42  Because the bitbake recipe says git.openmoko.org but I don't see it;
Nov 22 16:32:02  so bitbake is failing to bootstrap for me.
Nov 22 16:32:11 <[Rui]> sss comes from shr simple settings because I was not pompous enough to call it when it's just a proposal :)
Nov 22 16:32:25  dos1: trying to
Nov 22 16:32:35  Q-Master: but really, wave is better
Nov 22 16:32:44  Does SHR even depend on tichy?
Nov 22 16:32:47  but much bigger
Nov 22 16:33:07  Q-Master: but who cares? :p
Nov 22 16:33:17  delete default wav ringtone
Nov 22 16:33:18  suxx 8)
Nov 22 16:33:20  and you'll get space ;P
Nov 22 16:33:28  or use sd card
Nov 22 16:33:53  Or can I make this problem go away by just setting SRCPV_pn-tichy?
Nov 22 16:34:49  TAsn: i'm testing bumped python-elementary now
Nov 22 16:34:59  rozzin: remove BB_GIT_CLONE in your Makefile/local.conf
Nov 22 16:35:00  dos1, weepee ;)
Nov 22 16:35:12  dos1, we should bump all of efl though.
Nov 22 16:35:19  yup
Nov 22 16:35:30  with my up to date patch I sent mrmoku|away for unwanted pages in wizard
Nov 22 16:35:30  but the most important issues are not fixed yet...
Nov 22 16:35:45  in elementary
Nov 22 16:36:30  JaMa: What exactly does that do?
Nov 22 16:36:46  rozzin: search logs/ml please..
Nov 22 16:37:00  rozzin: I said that too many times and it cannot be said in one sentence :)
Nov 22 16:37:31  JaMa: er, sorry--can you tell me where the code that cares about it is?
Nov 22 16:37:47  JaMa: I'd actually rather read that than english ;)
Nov 22 16:38:00  Weiss: platform_get_irq(pdev, 0)
Nov 22 16:38:29  larsc: thanks.. actually I'd figured it out by looking at the MMC driver..
Nov 22 16:39:13  DRM/KMS for 2.6.31 is not up and running, apart from a few modesetting and resume bugs
Nov 22 16:39:15  TAsn: ImportError: /usr/lib/python2.6/site-packages/elementary/c_elementary.so: undefined symbol: elm_hoversel_item_del_cb_set
Nov 22 16:39:22  so we have to bump EFL :P
Nov 22 16:39:26  shit. fbreader also crappy built
Nov 22 16:39:31  dos1, probably...
Nov 22 16:39:35  :)
Nov 22 16:39:41  dos1, try to bump locally
Nov 22 16:39:42 <[Rui]> maybe it should be renamed if it gets to go "official" :)
Nov 22 16:39:42 <[Rui]> sss is quite cryptic
Nov 22 16:39:42  and give it a go.
Nov 22 16:39:57  mickey|away: depends on what device names you are reffering to
Nov 22 16:39:57  [Rui], it'll be added to libphone-ui-shr
Nov 22 16:40:05  so it won't have it's own name anyway
Nov 22 16:40:11  well, i don't have local shr build
Nov 22 16:40:18  rozzin: bitbake itself
Nov 22 16:40:24  JaMa: Or at least, why is that the right solution,
Nov 22 16:40:27  did I piss him off? :)
Nov 22 16:40:35  JaMa as opposed to fixing the tichy recipe?
Nov 22 16:40:51  rozzin: because using it on every builder is wrong solution
Nov 22 16:41:03  JaMa: or is the tichy recipe actually correct, and there's something that I don't understand going on?
Nov 22 16:41:41  rozzin: no its tichy problem (wrong SRC_URI) but with that enabled you're git cloning every git source used in whole OE tree
Nov 22 16:41:50  Weiss: nice to hear :)
Nov 22 16:41:57  rozzin: which is terrible load for git servers
Nov 22 16:44:16  JaMa: yes, I suppose it would be. :)
Nov 22 16:47:54  Why does BB_GIT_CLONE even exist?  It seems to me that it's very existence is cause for blacklisting by the public git hosting servers. :(
Nov 22 16:48:08  http://www.hentges.net/irclogs/%23oe/2009/August/20090820_oe.log?lines=all
Nov 22 16:48:41  hi all, is there an easy way to port my existing contacts,messages from old opim (before current face) to current one?
Nov 22 16:48:54  mwester_: I gather that it solves the problem of git revision-strings being unsortable.
Nov 22 16:48:58  catholicon: there is no "old opimd" and "new opimd"
Nov 22 16:49:23  rozzin: I gather that, but it seems that as a solution, it causes many times more pain than it actually solves. 
Nov 22 16:49:38  Is that an SHR "feature" or was that originated by the OE team?
Nov 22 16:49:40  dos1: i tried copying the db file from the other partition (unflashed)....but opimd says that it can't read contacts,messages
Nov 22 16:50:08  dos1: i don't have the exact error right now...but can paste here in 5 minutes...would you need it?
Nov 22 16:50:15  well
Nov 22 16:50:22  how were you using opimd before?
Nov 22 16:50:26  it was unsupported at all
Nov 22 16:50:37  mwester_: it creates consistent PVs across multipe builders
Nov 22 16:50:39  dos1: litephone...!!
Nov 22 16:51:02  well
Nov 22 16:51:05  JaMa: I saw that claim in the log that I'm reading...
Nov 22 16:51:18  catholicon: there wasn't any change recently in opimd
Nov 22 16:51:31  JaMa: I'm going to have to read some more before I understand that as meaning something that I can believe....
Nov 22 16:51:36  mwester_: i think is OM thing
Nov 22 16:51:49  catholicon: you can try to look at old db file and new with sqlite3 tool
Nov 22 16:51:53  and compare structure
Nov 22 16:51:59  dos1: k...just a sec then....i would reboot to get the exact error
Nov 22 16:52:08  dos1: i checked the schema and they match....
Nov 22 16:52:16  catholicon: types too?
Nov 22 16:52:23  mwester_: without it are sortable git revisions solved by incremental counter every time revision change for your build
Nov 22 16:52:28  dos1: i would have to recheck
Nov 22 16:52:36  larsc: any objections to anything in http://git.bitwiz.org.uk/?p=kernel.git;a=commitdiff;h=05aa5f2eb91d646212caa13cb84ff18221bd37ef ?
Nov 22 16:52:45  dos1: jsut a sec
Nov 22 16:53:08  mwester_: that way if you're building more often than other builders than feeds from different builders cannot be used on same device
Nov 22 16:53:14  JaMa: I understand that.
Nov 22 16:53:21  mwester_: because one builder can return higher PV for older revision
Nov 22 16:53:25  I don't agree that it is a viable solution
Nov 22 16:53:32  I am well aware of the git problems with OE.
Nov 22 16:53:44  But you can't go killing every other developer just to solve that poblem.
Nov 22 16:53:46  ffs. :)
Nov 22 16:53:51  JaMa: does that assume that the `multiple builders' are all assumed to be building the same (unmodified) source from the same upstream?
Nov 22 16:53:52  git sucks.
Nov 22 16:53:59  git needs to be fixed
Nov 22 16:54:21  but to have OE go download 15,000 git repos just to parse bb files is so wrong.
Nov 22 16:54:39  mwester_: I'm not using it, I did not enable it in shr makefile by default, I persuaded mrmoku to disable it :)
Nov 22 16:54:39  Please tell me that it only downloads the ones in the dependency tree...
Nov 22 16:54:59  no its not..
Nov 22 16:55:12  git-ls-remote doesn't do anything useful, here?
Nov 22 16:55:17  rozzin: no
Nov 22 16:55:28  rozzin: you need git rev-list $your_rev | wc -l
Nov 22 16:55:35  Then it is a massively broken misfeature, and needs to be removed from OE (if it's actually in OE).  I'll go check to see if it's in OE or just in the OM version of it.
Nov 22 16:55:36  rozzin: in your git cloned dir
Nov 22 16:55:52  JaMa: I see.
Nov 22 16:56:19  mwester_: its OE, with SRCPV (only in shr/merge, martin_jansa/srcpv) it's done while parsing tree
Nov 22 16:56:24  Weiss: glamo_enable_irq: We have enable_irq for that and glamo_clear_irq shouldn't be needed either
Nov 22 16:56:32  mwester_: because SRCPV needs those build numbers to create PV
Nov 22 16:56:52  mwester_: without SRCPV it was needed only in recipes with AUTOREV used
Nov 22 16:57:20  Thanks.  Then I shall vote "no" when it comes time to merge the SRCPV stuff into OE.
Nov 22 16:57:53  you shall vote for disabling BB_GIT_CLONE
Nov 22 16:58:04  Man.
Nov 22 16:58:12  SRCPV isn't that bad guy
Nov 22 16:58:12  Weiss: and i'm not sure if the core driver should dictate what region to use for the cmdq.
Nov 22 16:58:39  dos1: yup....even the types ... http://pastebin.com/d5e3fa096
Nov 22 16:59:28  dos1: would you need the exact error i get from newly flashed shr?
Nov 22 17:00:13  mrmoku, bored enough to bump efl?
Nov 22 17:00:40  mrmoku: can you add xf86-input-tslib to the feed?
Nov 22 17:00:48  JaMa: BB_GIT_CLONE_FOR_SRCREV *is* in the default configuration....
Nov 22 17:00:51  don't forget to use my drop-unwanted-wizard-pages patch, illume-flow.patch and edje.patch can be dropped though (if they still exist)
Nov 22 17:01:00  JaMa: the problem should be solved some other way; perhaps an OE Builder-to-Builder network protocol so they can communicate among themselves to agree on the naming/numbering -- that would be the right way to solve what is an OE problem, rather than to brutalize all the git servers out there, not to mention massively excess disk space consumption for the builders.
Nov 22 17:01:04  rozzin: no
Nov 22 17:01:13  rozzin: only was in shr Makefile before..
Nov 22 17:01:48  In the long run, git is clearly broken -- it needs to provide the necessary info remotely without a clone - so a patch needs to be developed and submitted upstream.
Nov 22 17:01:49  mwester_: agreed, but SRCPV creates sortable versions WITHOUT using any git clone
Nov 22 17:02:01  Now I'm confused.
Nov 22 17:02:05  Anyhow, you'll hopefully won't see me until tomorrow night.
Nov 22 17:02:06  ciao.
Nov 22 17:02:12  You just told me that the git cloning was required for SRCPV.
Nov 22 17:02:19  Now you say it doesn't.
Nov 22 17:02:28  JaMa: Unless it's changed in the past 2 days, then yes it is.
Nov 22 17:02:42  mwester_: I said that SRCPV expose problem in BB_GIT_CLONE_FOR_SRCREV a bit more
Nov 22 17:02:58  mwester_: but SRCPV doesn't do/need any "git clone"!
Nov 22 17:03:05  JaMa: because that's how long I've been waiting for my SHR environment to bootstrap from Makefile.
Nov 22 17:03:10  rozzin: its changed today
Nov 22 17:03:18  JaMa Ah.
Nov 22 17:03:41  rozzin: won't change your existing config though
Nov 22 17:03:42  16:06:57 < CIA-86> SHR: mok shr-makefile * rf4a6fefb8c99 /Makefile: Makefile: comment out BB_GIT_CLONE_FOR_SRCREV when setting up local.conf
Nov 22 17:03:56  JaMa: A hearty thank-you to you, then.! :)
Nov 22 17:04:19  JaMa: now I don't have to eat my feet off anymore! :)
Nov 22 17:04:58  larsc: ah.. that irq_chip stuff tells it how to do it with a generic method?
Nov 22 17:04:59  mwester_: AUTOREV+BB_GIT_CLONE_FOR_SRCREV did git clone while parsing OE tree, without SRCPV
Nov 22 17:05:16  Only for AUTOREV recipes.
Nov 22 17:05:23  Which was tolerable.
Nov 22 17:05:25  mwester_: yes
Nov 22 17:05:41  mwester_: debatably....
Nov 22 17:06:04  Weiss: yes
Nov 22 17:06:13  mwester_: but you cannot switch between AUTOREV and sane-srcrev without PE bump 
Nov 22 17:06:39  So?  Lots of digits in a 32-bit int.
Nov 22 17:06:42 * rozzin is biased, though--has yet to see any of this work.... :\
Nov 22 17:06:42  mwester_: which is worse than disabling BB_GIT_CLONE_FOR_SRCREV, isn't it?
Nov 22 17:06:46  No
Nov 22 17:06:51  Not at all.
Nov 22 17:07:06  mwester_: digits for what?
Nov 22 17:07:07  If you switch from dev to released upstreams, a major bump is to be expected.
Nov 22 17:07:19  digits for PE
Nov 22 17:07:31  You won't run out if you have to bump once in a while.
Nov 22 17:07:54 * JaMa doesn't like package names like navit4343_0.2.0.ipk
Nov 22 17:08:08  And it puts the problem where it belongs -- in the distro doing the switch between AUTOREV and fixed release, rather than beating to death upstream git repos.
Nov 22 17:08:34 * mwester_ doesn't like package names like navit1_urty894798738957tu8905637896739870938609845903856038u-gitr934593486769807800398.ipkh
Nov 22 17:08:50  mwester_: I don't want to argument with you...
Nov 22 17:08:54  It's a good thing we have a package manager so none of that matters.
Nov 22 17:09:11  JaMa: I have a non-arguing question, about this....
Nov 22 17:09:15  mwester_: you need to type opkg install navit4343, no?
Nov 22 17:09:35  larsc: ah, ok.. core acks the IRQ as well. much cleaner..
Nov 22 17:10:04  Then argue with me when you push it -- or put together an email when you do so that will completely and thoroughly explain what your commits do because I cannot and will not support anythign that downloads git repos gratuitously, AUTREV excepted.
Nov 22 17:10:26  JaMa: no, you type opkg install navit  -- the PE disappers.
Nov 22 17:10:39  freesmartphone.org: 03mickey 07cornucopia * r0635be45a904 10/fsotimed/src/plugins/ (3 files in 3 dirs): fsotimed: handle alphanumerical time zone report and adjust /etc/timezone accordingly
Nov 22 17:10:40  The PE is just part of the version number.
Nov 22 17:10:52  freesmartphone.org: 03mickey 07cornucopia * ra2d2cb8b2675 10/docs/frameworkd.conf.sample: update frameworkd.conf.sample
Nov 22 17:11:32  mwester_: do you need BB_GIT_CLONE?
Nov 22 17:11:53  It's not about "I" -- it's about the distros that may choose to abuse it.
Nov 22 17:12:21  "I" have no control over what SHR or OM or any other distro sets up.
Nov 22 17:13:19  mwester_: do you control how many distros include insane-srcrevs?
Nov 22 17:13:35  Historically, I trusted that a handful of recipes under active development used AUTOREV and cloned git repos; that's ok.  If this change will clone every git repo for every recipe using git, regardless of AUTOREV or if it is even in the dependency graph, I cannot accept that.
Nov 22 17:14:17  mwester_: have you read wbout LOCALCOUNT override in SRCPV migration thread?
Nov 22 17:14:20  JaMa: No I don't control that.  I can edit their configs in OE and commit changes, but that would get my commit rights removed rather quickly, don't you think? :D
Nov 22 17:14:37  It's a little frustrating that even AUTOREV requires a git clone for things that aren't in the dependency-graph....
Nov 22 17:14:51  mwester_: probably :)
Nov 22 17:15:37  mwester_: missusing AUTOREV exists as well as missusing BB_GIT_CLONE imho
Nov 22 17:15:55  JaMa: I'm trying to encourage you or someone to summarize the whole mess -- so far I've been told snippets, and told to read dozens of emails arguing fragments -- a summary of the whole situation that would change the AUTOREV thing is desirable, otherwise I simply have to vote no. 
Nov 22 17:16:02  mwester_: we should put "sleep 60" to bitbake if BB_GIT_CLONE enabled with fat warning
Nov 22 17:16:11  Silly.
Nov 22 17:16:21  mwester_: I ended up going through and assigning revs to all of the neuros packages, just to get past them when parsing, because I obviously don't use them on my FreeRunner....
Nov 22 17:17:02  rozzin: wait until those managing those servers add a block for the OE users who keep blasting their servers with git clones!
Nov 22 17:17:05  mwester_: but I'm not confident in my ability to track down all of the other packages that I don't use.
Nov 22 17:17:06  mwester_: changing format of PV from 0.1+gitrHash to 0.2+gitrNNN+hash is SILLY
Nov 22 17:17:36  changing the gauge of a railroad track is silly too -- but what has either to do with this situation?
Nov 22 17:18:12  mwester_: SRCPV is solving that
Nov 22 17:18:18  Nobody has clearly summarized the impact of this misfeature; I shall cease discussion and wait for such an email -- until then, NO is the vote on all SRCPV stuff.  Sorry.
Nov 22 17:18:38  mwester_: Actually,
Nov 22 17:18:56  mwester: I was starting to wonder if I was running into that problem already,
Nov 22 17:18:59  The present status with AUTOREV is barely tolerable; no more changes that make OE worse when it comes to parsing.
Nov 22 17:19:37  mwester_: why is AUTOREV tolerable and BB_GIT_CLONE is not pain in ass?
Nov 22 17:19:41  mwester_: because some of the servers hosting only packages that I don't use (like the neuros ones) were apparently just not responding to me.
Nov 22 17:20:12  mwester_: sorry I still dont see why not ban BB_GIT_CLONE
Nov 22 17:20:51  JaMa: did you see my question about this, earlier?
Nov 22 17:21:00  JaMa: ??  Clearly because AUTOREV affects only the recipes that are specifically listed as AUTOREV.  BB_GIT_CLONE, as you have described, clones every git repo for every recipe, AUTOREV or not, regardless of if they are even part of the dependency graph.
Nov 22 17:21:08  mwester_: LOCALCOUNT varibles were created to do similar job as GIT_CLONE manually without git clone
Nov 22 17:21:32  mwester_: LOCALCOUNT varibles can stop missusing PR in PV just to make bumped SRCREVS upgradeable
Nov 22 17:21:38  Great, then an email clearly describing that (and proposing the remove of BB_GIT_CLONE) would be great!
Nov 22 17:22:04  mwester_: ok, that's what I wanted to hear..
Nov 22 17:22:36  ...?
Nov 22 17:22:48  rozzin: no.. haven't seen?
Nov 22 17:23:08 * rozzin finds it...:
Nov 22 17:23:36  JaMa: does that assume that the `multiple builders' are al
Nov 22 17:23:38  l assumed to be building the same (unmodified) source from the same upstream?
Nov 22 17:24:07  er, rather:
Nov 22 17:24:08  JaMa: does that assume that the `multiple builders' are all assumed to be building the same (unmodified) source from the same upstream?
Nov 22 17:24:21  rozzin: yes
Nov 22 17:24:54  rozzin: but not in all case you need feeds from multiple builders shared on one target
Nov 22 17:26:00  JaMa: because, if there are any differences in the git history that a builder is using, then the revision-count is not a legitimate UUID; which is why git uses hashes....
Nov 22 17:27:08  JaMa: sorry, I'm having trouble parsing that last response....
Nov 22 17:27:12  JaMa: ?
Nov 22 17:28:21  if you have always increasing number with every git hash change (you can even use older revision in sane-srcrev intentionally)
Nov 22 17:28:38  than you have always upgradable path between versions
Nov 22 17:29:01  and you don't need revision-count (rev-list | wc -l in git clone)
Nov 22 17:29:18  that's what SRCPV do
Nov 22 17:29:52  just increments some BUILDNUMBER (or LOCALCOUNT) every time you build something with changed revision
Nov 22 17:30:13  builders are happy and users are happy with that
Nov 22 17:30:20  right, but revision-count is what SRCPV boils down to when BB_GIT_CLONE_FOR_SRCREV is enabled, right?
Nov 22 17:30:42  problem is when you have multiple builders across world and they all are increasing BUILDNUMBER when revision changes
Nov 22 17:30:50  JaMa: OK, I understand what you meant by the `you need multiple feeds' comment.
Nov 22 17:31:43  failing packages for task-shr-feed: http://shr.pastebin.com/m256cd34d
Nov 22 17:31:55  JaMa: will turn it off then on the buildhost too
Nov 22 17:31:56  because then one builder did build few more changed revisions and have PV=0.1+gitr21+hashABCD and another builder have PV=0.1+gitr18+hashABCD
Nov 22 17:32:04  JaMa: has someone posted a whitepaper on this somewhere?
Nov 22 17:32:21  JaMa: ... so that I don't have to make you run through this for me? :)
Nov 22 17:32:48  rozzin: probably no.. but I'll probably have to write that mail mwester_ requested...
Nov 22 17:33:36  mrmoku: same task-shr-feed here 
Nov 22 17:34:19  JaMa: well, someone will. This seems like a big enough (or opaque enough) issue that /someone/ should write up a whitepaper on it.
Nov 22 17:35:05  So...
Nov 22 17:35:40  hopefully my bad english won't require dozens of following e-mails to describe it in better way :)
Nov 22 17:35:42  hey folks. Phoneui is quite unstable on my freerunner. Is this known or should I start reporting bugs?
Nov 22 17:36:17  JaMa: I just started afresh with `wget .../Makefile'; I'll see how that goes. I'll see if I can just wait for the whitepaper, instead of pestering people on IRC about this.
Nov 22 17:36:57  JaMa: if you have difficulty writing clearly, then I guess that's what we have wikis for.
Nov 22 17:37:43  JaMa: thank you very much for making the effort to document this, though.
Nov 22 17:39:57  JaMa: and I will completely remove it from the makefile... not just comment it now ;)
Nov 22 17:41:20  SHR: 03mok 07shr-makefile * r72a80aded494 10/Makefile: Makefile: completely eliminate BB_GIT_CLONE_FOR_SRCREV
Nov 22 17:42:21  JaMa: I wonder if it would be possible and interesting to have some SRCPV_PREFIX var
Nov 22 17:42:45  this way one could determine, which feed from which buildhost has preference
Nov 22 17:42:58  and buildhosts with same preference would have to sync cache
Nov 22 17:44:41  JaMa: when you're drafting that whitepaper, if you could include descriptions of what happens with BB_GIT_CLONE_FOR_SRCREV both when an entire `git clone' is necessary but *also* what happens when there's already a local git clone present (that may or may not be out of date), that would be great.
Nov 22 17:47:18  JaMa: because that's been on my mind during this ordeal: `what's it going to be like if and when I do actually get bootstrapped?'
Nov 22 17:47:30  TAsn: ping
Nov 22 17:47:59  mrmoku, pong.
Nov 22 17:48:05  (though make it quick. please)
Nov 22 17:48:24  TAsn: anything we want in the image I'm going to build now? like the wizard?
Nov 22 17:48:34  JaMa: `do I have to be able to `git pull' from every host just to parse the bb recipes *every* time I build?'
Nov 22 17:48:34  or is it not yet ready?
Nov 22 17:48:46  TAsn: just add your wizard skip patch?
Nov 22 17:48:55  mrmoku, First of all, yes, add this patch.
Nov 22 17:48:57  second of all.
Nov 22 17:49:11  please include JaMa's fix for modules-strip
Nov 22 17:49:25  and EFL bump :P
Nov 22 17:49:25  even if it's not committed upstream yet, it's very important.
Nov 22 17:49:27  JaMa: or, in alternate phrasing, `am I always going to be unable to build if *any* git host is down?'
Nov 22 17:49:35  mrmoku, that goes without saying :)
Nov 22 17:49:40  :)
Nov 22 17:49:43  hm.. I think that's it.
Nov 22 17:50:02  JaMa: what do you think... could you put your kernel module patch into shr/merge? until oe.dev gets it?
Nov 22 17:50:30  TAsn: last question
Nov 22 17:50:36  (I'll probably finish the wizard tomorrow though, so be advised you'll have to build in image tomorrow as well ;p)
Nov 22 17:50:38  ?
Nov 22 17:50:40  is your patch needed when we have our own wizard too?
Nov 22 17:50:51  yes.
Nov 22 17:50:55  ok :)
Nov 22 17:51:04  and yes... will build an image tomorrow too :)
Nov 22 17:51:07  this patch drops parts of the .e wizard
Nov 22 17:51:08  have fun studying now :P
Nov 22 17:51:17  though keeps some of the 
Nov 22 17:51:19  +m
Nov 22 17:51:35  shr-wizard that dos built will be called from a module I'll add to the e wizard
Nov 22 17:52:16  mrmoku, ciao and thanks.
Nov 22 17:52:23  :)
Nov 22 17:52:26  let me know when you are done
Nov 22 17:52:29  I'll opkg upgrade
Nov 22 17:52:31  yup
Nov 22 18:03:00  what package contains python base64 enc?
Nov 22 18:12:02  Q-Master: it's one of the depencies of python-email
Nov 22 18:13:01  hm... and where is make???
Nov 22 18:21:01  so how usable is .31 now?
Nov 22 18:21:24  hehe
Nov 22 18:21:43  Contacts shows data from pim wrong
Nov 22 18:28:02  numbers detecting also doesnt work
Nov 22 18:28:29  but phonelog shows everything ok
Nov 22 18:29:46  Q-Master: about numbers detecting - do you have correct phoneutils settings?
Nov 22 18:30:03  yes
Nov 22 18:30:21  thety're same as in prev shr distro
Nov 22 18:30:26  but...
Nov 22 18:30:46  ok
Nov 22 18:31:01  here it doesn't work too, but i thought i have something broken
Nov 22 18:31:06  as i play with this image a lot ;P
Nov 22 18:31:10  so i'll debug it soon
Nov 22 18:32:00  what should be the correct phone field names?
Nov 22 18:32:05  I use CellPhone
Nov 22 18:32:13  but not Cell Phone
Nov 22 18:32:14  oh
Nov 22 18:32:24  "Cell phone" is correct
Nov 22 18:32:28  ok
Nov 22 18:32:44 * Q-Master need fixing the script
Nov 22 18:32:55  and name resolving won't work with those fields yet
Nov 22 18:33:01  i have to work on it... :X
Nov 22 18:33:12  ok
Nov 22 18:33:21  it worked lately
Nov 22 18:33:30  really?
Nov 22 18:33:32  have you broke everything?
Nov 22 18:33:52  maybe i just don't remember that i implemented it already? :P
Nov 22 18:33:53  I meant opimd-utils with patch
Nov 22 18:34:35  hi
Nov 22 18:36:34  freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * rcfd94f6f02b4 10framework/framework/subsystems/opimd/docs/TODO: opimd: update TODO
Nov 22 18:36:41  Q-Master: i meant opimd ;P
Nov 22 18:36:50  ah
Nov 22 18:36:56  yep. it suxx. 8)
Nov 22 18:37:58  dos1: and what about email? how should it be named?
Nov 22 18:39:28  Q-Master: "E-mail" and "Work e-mail"
Nov 22 18:39:35  (and similar)
Nov 22 18:39:39  ok
Nov 22 18:40:58  dos1: and what about Home cell phone?
Nov 22 18:41:18  dos1: or just abandon sucha things?
Nov 22 18:41:37  "Home cell phone" should work too ;P
Nov 22 18:45:45  hi guys, now i have my display on neo completly flickering, and neo is very slow
Nov 22 18:46:02  how can i check what is the cause? i believe its SD card somehow
Nov 22 19:04:36  ( what about email? how should it be named?)  bloodyTelNum  <1>POTS <2>GSM <3>SMS  generic phonenumber   
Nov 22 19:05:36  ~lart hardcoded name2properties
Nov 22 19:05:36 * apt squishes hardcoded name2properties like a bug
Nov 22 19:05:57  mwester: really long description sent
Nov 22 19:09:26   fsck-ng telnum2 bloodyTelNum 
Nov 22 19:09:36  
Nov 22 19:14:51  dos1: wonder if you read my last msgs to you!!
Nov 22 19:14:53    <1>POTS  fsck-ng telnum2 plain old telefone system number 
Nov 22 19:15:06  repeat:: dos1: yup....even the types ... http://pastebin.com/d5e3fa096
Nov 22 19:15:13  repeat:: dos1: would you need the exact error i get from newly flashed shr?
Nov 22 19:15:36  catholicon: didn't get them before :x
Nov 22 19:15:59  catholicon: but not now, i'm "quite afk"
Nov 22 19:16:32  dos1: thats ok...can you just give me a hint as to where can i start probing into....
Nov 22 19:16:50  dos1: i have no cross-compilation setup...so python is all i've got 
Nov 22 19:29:40 <[Rui]> wife is driving, we're half way.
Nov 22 19:30:53 <[Rui]> so... anyone knows if the call sound problem is solved or has a workaround?
Nov 22 19:35:04  rui: as for now I get it mrmoku is facing an ALSA error of unclear cause
Nov 22 19:35:34  [Rui], DocScrutinizer-8: mrmoku's problem is with 2.6.31 kernel :P
Nov 22 19:35:38  DocScrutinizer-8, [Rui]: my problems are with 2.6.31
Nov 22 19:35:43  ahh :)
Nov 22 19:35:46  ;D
Nov 22 19:35:46  possibly related to new "statefiles"
Nov 22 19:35:51  DocScrutinizer-8: no
Nov 22 19:35:59  with "official" kernel it's everything ok
Nov 22 19:36:00  DocScrutinizer-8: it works all fine with older kernel
Nov 22 19:36:29  can I use Makefile.am without configure.am|in somehow? Ie for building only some subdirectory without checkout of other sources?
Nov 22 19:36:35  Hi just briefly here- Someone should write a mail to shr-devel with the BB_CLONE info too
Nov 22 19:36:47  if nobody else does, I can do it tomorrow afternnoon
Nov 22 19:36:48  sorry have to confess I have no way to get a clue what kernel is supposed to ship with current SHR
Nov 22 19:36:48 <[Rui]> cool, hope it gets fuxed as well
Nov 22 19:36:51  JaMa: nice mail
Nov 22 19:37:00 * JaMa is finished writting emails for this year or 2
Nov 22 19:37:07  :P
Nov 22 19:37:27  DocScrutinizer-8: with SHR it's still shipped andy-tracking HEAD, for ages and that probably won't change very soon
Nov 22 19:38:13 <[Rui]> but I meant hopefully be able to make and receive calls with current shr-u
Nov 22 19:38:36  [Rui]: just upgrade, all problems should be fixed
Nov 22 19:38:45  lol
Nov 22 19:38:58  from remaining, AFAIR only hoversels in python apps and rejecting call when resuming from suspend
Nov 22 19:39:14  first one is going to be fixed soon, second has still unknown reason :P
Nov 22 19:39:29 <[Rui]> dos1, ok, I will try when I get home :)
Nov 22 19:40:20  aah, rejecting call during resume on inbound call. A minor problem indeed
Nov 22 19:40:21  JaMa: I don't quite understand your Makefile.am question.
Nov 22 19:40:52  rozzin: http://navit.svn.sourceforge.net/viewvc/navit/trunk/navit/navit/xpm/
Nov 22 19:41:12  rozzin: this directory is only rasterizing svgs to pngs for navit
Nov 22 19:41:25  rozzin: I would like to split navit and navit-icons to separate packages
Nov 22 19:41:46  dos1: locally building 43860 now
Nov 22 19:41:48  rozzin: because navit-icons are updated seldom and navit source very often
Nov 22 19:41:50  that should fix hoversels, no?
Nov 22 19:41:58  mrmoku: yup
Nov 22 19:42:10  rozzin: also navit-icons package is huge and takes lots of time to upgrade on small devices like neo
Nov 22 19:42:10  JaMa: where do I put SHR EFL_SRCREV?
Nov 22 19:42:14  JaMa: to be released as separate tarbals?
Nov 22 19:42:16  JaMa: in shr.conf?
Nov 22 19:42:32  rozzin: no released just built in OE environment
Nov 22 19:42:53  mrmoku: hmm or preferred-shr-versions?
Nov 22 19:43:06  JaMa: you're talking about an upstream change in navit, though?
Nov 22 19:43:13  mrmoku: or ask koen if he agrees to bump it globally..
Nov 22 19:43:20  rozzin: no
Nov 22 19:43:37  rozzin: it would be easy to change in upstream
Nov 22 19:43:50  JaMa: probably prefer the first one for now :P
Nov 22 19:44:03  rozzin: I just created navit-icons_svn.bb where I checkout just that subdirectory
Nov 22 19:44:33  rozzin: and will remove that folder from Makefile, one directory above for navit_svn.bb
Nov 22 19:44:50  JaMa: Ah, I see.
Nov 22 19:45:00  rozzin: that way I can build navit.ipk without icons (small easily upgradable and often upgraded)
Nov 22 19:45:45  and navit-icons.ipk huge and updated only when there is new icon (about 3 months old is last change now..)
Nov 22 19:46:02  JaMa: got it.
Nov 22 19:47:22  JaMa: hm....
Nov 22 19:48:12  JaMa: Well, automake just outputs `.in' templates; you need a configure script or something to process them.
Nov 22 19:48:28  rozzin: I can download whole svn tree, build all, pack only xpm ..
Nov 22 19:48:38  rozzin: but I wanted some cleaner sollution
Nov 22 19:50:53  mrmoku, !
Nov 22 19:50:58  JaMa: I gather that just splitting into different filesets into different ipk pckages in the bb recipe isn't sufficient for you..., because you'd still be producing `natit-icons' ipks with updated versions (but identical contents)?
Nov 22 19:51:07  mrmoku, 43860 is not enough, we need at least 43886
Nov 22 19:51:19  er, ...`navit-icons'...
Nov 22 19:51:54  oh
Nov 22 19:52:01  rozzin: no.. if I build it twice on buildhost 
Nov 22 19:52:10  rozzin: I mean if I create 2 independent recipes
Nov 22 19:52:20  rozzin: which are building the same (whole navit tree)
Nov 22 19:52:33  rozzin: but than packing different fileset after rebuild
Nov 22 19:52:50  JaMa: oh, I guess you could do that too....
Nov 22 19:52:58  TAsn: ahh
Nov 22 19:53:03  JaMa: I see why you don't like that thoug.
Nov 22 19:53:22  JaMa: so you want to just `svn co .../icons' and build that.
Nov 22 19:53:35  mrmoku, sorry for not telling you in advance
Nov 22 19:53:43  I assumed you were going to bump to lastest ;)
Nov 22 19:53:47  latest*
Nov 22 19:54:08  JaMa: Mm. navit is one of these packages using AUTOREV?
Nov 22 19:54:42  rozzin: not now.. but was
Nov 22 19:54:59  rozzin: exactly
Nov 22 19:55:04  TAsn: 43898 then?
Nov 22 19:55:21  TAsn: I took the 43860 from your patch :P
Nov 22 19:56:07  mrmoku, oh, no, that's just what I patched against :P not when my last one got committed :)
Nov 22 19:56:16  JaMa: Mmph. What version are you going to put on the `navit-icons' package, if it doesn't follow the same release-cycle as the navit-bin package?
Nov 22 19:56:17  mrmoku, (rev) yeah, why not. 
Nov 22 19:56:28  TAsn: maybe wait for two more commits? ;)
Nov 22 19:56:40  mrmoku, hehe quick, send patches ;p
Nov 22 19:56:42  43900 would be nice :P
Nov 22 19:57:02  mrmoku, mark in your calender, we'll build again on 44000 ;p
Nov 22 19:57:11  :)
Nov 22 19:58:03  TAsn: could you make use of a raphael?
Nov 22 19:58:41  TAsn: i'm somewhat thinking htclinux is not going anywhere, so i might as well donate mine...
Nov 22 19:58:41  rozzin: its not a big issue.. navit will depend on newes navit-icons provided
Nov 22 19:58:57  hmm
Nov 22 19:59:04  i might confuse you with someone else
Nov 22 19:59:05  sorry
Nov 22 19:59:13  mickeyl, I don't have even the slightest idea what's raphael is. ;]
Nov 22 19:59:14  rozzin: so ie navit-0.1.0+svnr2752 now and navit-icons-0.1.0+svnr2618
Nov 22 19:59:18  so yes, you probably confused me.
Nov 22 19:59:30  sorry, the other 4-letter nickname starting w/ t
Nov 22 19:59:31  with someone else
Nov 22 19:59:31  tmzt: ping
Nov 22 19:59:37  :)
Nov 22 19:59:56  :)
Nov 22 19:59:58  ciao.
Nov 22 20:00:10  mrmoku, again, let me know when everything is there ;)
Nov 22 20:00:15  (opkg upgrade)
Nov 22 20:00:17  TAsn: sure :)
Nov 22 20:00:24 * mrmoku is waiting for two commits ;)
Nov 22 20:00:33  the chaos starts to make me feel at home :-P
Nov 22 20:00:46  mrmoku, lol.
Nov 22 20:00:59  JaMa: and then the bb maintainer for navit-icons.bb will just pay attention and decide when to say `time for a new icons release, since they actually changed something this time?'
Nov 22 20:01:19  rozzin: yes
Nov 22 20:01:27  JaMa: OK. Got it.
Nov 22 20:02:00  clearly stated
Nov 22 20:02:05  JaMa: Sorry for all of the meta-questions before I could answer your question...
Nov 22 20:02:11  publically discussed and agreed upon
Nov 22 20:02:17  goal and deadlines
Nov 22 20:02:25  and releases
Nov 22 20:02:28  ;-P
Nov 22 20:02:39  JaMa: The Makefile.am in the xmp/ dir obviously won't work standalone,
Nov 22 20:03:01  JaMa: since it references something from $(top_srcdir)
Nov 22 20:03:57  rozzin: yes.. maybe I have an idea :)
Nov 22 20:04:03  JaMa: so, you'd have to either patch that Makefile.am or provide a copy of Makefile.inc in the same xpm/ dir, because that'd become the new top_srcdir.
Nov 22 20:04:04  rozzin: just call do_configure on whole tree
Nov 22 20:04:10  why are the new shr-unstable image so huge? shr-lite used to be 60-70MB
Nov 22 20:04:20  rozzin: then cd to xpm directory and call do_compile only there
Nov 22 20:04:26  JaMa: And you'd need some sort of confiugre script.
Nov 22 20:04:37  Flyser, A bug in OE that meant striping modules was broken
Nov 22 20:04:43  and probably a bit more OE issues.
Nov 22 20:04:46 <[Rui]> Flyser, tha4 is the full image
Nov 22 20:04:50  JaMa: Something like that might work.
Nov 22 20:05:03  [Rui]: ?
Nov 22 20:05:19 <[Rui]> or th light one?
Nov 22 20:05:40  TAsn: okay ... :-/ sucks, if you have a low-bandwith connection ...
Nov 22 20:05:56  Flyser, it sucks even if you don't.
Nov 22 20:06:00  yeah^^
Nov 22 20:06:04  though todays image will probably consist a fix
Nov 22 20:06:06  mrmoku, right? ^
Nov 22 20:06:13 <[Rui]> sorry for typos   i am on a on screen display keyb8ard and the road is bumby
Nov 22 20:06:51 <[Rui]> TAsn, gooe news, tgen!
Nov 22 20:07:54  JaMa: can you just split the files into a separate ipk via `FILES_' vars in navit.inc, and then conditional include "navit-icons" in the PACKAGES list?
Nov 22 20:07:59  [Rui], We got a fix (jama fixed it) it's just a matter of waiting for OE to accept the patch vs applying on our branch and building from there.
Nov 22 20:08:17  So this only affects modules and not all the other binaries? so there is no slowdown?
Nov 22 20:08:27  TAsn: fix for what?
Nov 22 20:08:34  ahh the modules
Nov 22 20:08:38  JaMa: define all of the `FILES_' vars in navit.inc, and define PACKAGES in navit_0.x.bb?
Nov 22 20:09:11  TAsn: as I said before.. already pushed..
Nov 22 20:09:14  JaMa: That scheme seems like it would work to implement what you want, no?
Nov 22 20:09:22  mickeyl: small OE policy question... when I add a patch to e-wm which affects only SHR (via shr override) - do I have to go via the maintainer and ml?
Nov 22 20:09:26  TAsn: I pushed first one module_strip fix to oe.dev
Nov 22 20:09:46  TAsn: and second modules.tgz with stripped and modules-dbg.tgz with unstriped just for us in shr/merge
Nov 22 20:10:02  JaMa, oh, sweet! :)
Nov 22 20:10:03  rozzin: in same recipe?
Nov 22 20:10:07  Flyser, here's your answer.
Nov 22 20:10:11  ^
Nov 22 20:10:13  ^^
Nov 22 20:10:20  JaMa: pardon?
Nov 22 20:10:26  rozzin: no.. because in same recipe I cannot specify which SRCREV to use for navit-icons
Nov 22 20:10:40  mrmoku, concerning your question, if yes, that'd be amazingly stupid. :)
Nov 22 20:10:56  rozzin: mmt I'll do it my way and show you a patch :)
Nov 22 20:11:17  JaMa: different recipes.
Nov 22 20:11:24  TAsn: :P
Nov 22 20:11:53  you should have the freedom to f*** up your distro.
Nov 22 20:12:05  mrmoku: obviously different people have different ideas of OE's policies, but given this case I'd say go ahead.
Nov 22 20:12:27  (not mentioning that i used to be the EFL maintainer before someone dragged it away)
Nov 22 20:12:31  mickeyl: yeah fine... especially because he did not respond to another mail regarding illume keyboards :|
Nov 22 20:13:17  btw, is it possible via easy way remove numbers and simple-keyboard illume layouts?
Nov 22 20:13:24  JaMa: e.g.: navit_0.1.0.bb uses PACKAGES="navit-bin navit-icons", navit_0.1.1.bb uses PACKAGES="navit-bin", navit.inc defines FILES_navit-bin and FILES_navit-icons; does that work?
Nov 22 20:13:36  max_posedon: that was the mail he did not respond to :P
Nov 22 20:13:40 <[Rui]> max_posedon, rm them
Nov 22 20:13:49  max_posedon: we want to have the standard keyboards in extra packages
Nov 22 20:13:57  so that one can just easily remove them
Nov 22 20:14:08  (or someday we might want to include others as default)
Nov 22 20:14:31  TAsn: btw. there is quite some activity in the illume rewrite :)
Nov 22 20:14:39  mrmoku, I know.
Nov 22 20:14:47  I talked with raster about it.
Nov 22 20:14:59  he said he'll try to remember to give us an heads up :)
Nov 22 20:15:00  [Rui], where and what I should remove?
Nov 22 20:15:00  rozzin: for releases it could work (you want them in sync navit.ipk and navit-icons.ipk)
Nov 22 20:15:05  I'm registered to e-devels anyway
Nov 22 20:15:14  so I hope I'll see something before.
Nov 22 20:15:27  rozzin: but then there is no need to split it for releases, I'm splitting only _svn recipes
Nov 22 20:15:34  mrmoku, anyhow, how's keyboard spliting doing?
Nov 22 20:15:43  TAsn: bad
Nov 22 20:16:05 <[Rui]> max_posedon, you'll hqve to search them, I do not know from heart and i cant easily check that now
Nov 22 20:16:08  JaMa: would it be ok to downgrade PR for e-wm when updating EFL_SRCREV in the same push?
Nov 22 20:16:17  rozzin: in releases will be only CONFLICTS_${PN} = "navit-icons" to remove svn icons from rootfs before installing release version
Nov 22 20:16:21 <[Rui]> almost home... brb
Nov 22 20:16:25  max_posedon: /usr/lib/enlightenment/modules/illume/keyboards
Nov 22 20:16:29  I think
Nov 22 20:16:45  mrmoku: global PR?
Nov 22 20:16:54  mrmoku: you cannot do that, no?
Nov 22 20:16:55  mrmoku, why bad?
Nov 22 20:17:03  JaMa: e-wm PR... to be in sync with org.oe.dev
Nov 22 20:17:20  merging conflicts because we have r9 and dev r7
Nov 22 20:17:31  mrmoku, the patch I sent you removes them. adding packages for keyboards is easy, and adding those to rdepends is also easy.
Nov 22 20:17:36  JaMa: well, I guess you'd need navit_0.1.0+svn.bb (without icons) and navit_0.1.0.bb (with icons) :)
Nov 22 20:17:59  TAsn: I hoped to get an answer and not have to do it via shr overrides
Nov 22 20:18:15  TAsn: but I guess I will have to do that
Nov 22 20:18:23  TAsn: not for todays image
Nov 22 20:18:28  mrmoku, they are just dead?
Nov 22 20:18:35  who?
Nov 22 20:18:38  rozzin: yes.. that's what I'm doing right now :)
Nov 22 20:18:51  rozzin: remove icons only in navit_svn.bb
Nov 22 20:19:02  rozzin: depend on navit-icons only in navit_svn.bb
Nov 22 20:19:18  rozzin: and block navit-icons from all other released navit recipes
Nov 22 20:20:07  mrmoku, oe-devel
Nov 22 20:21:23  JaMa: hmm. Something like that, yeah.
Nov 22 20:23:41  JaMa: want me to sync the PR? :)
Nov 22 20:23:45  (for e-wm)
Nov 22 20:23:50 * mrmoku wants to commit :P
Nov 22 20:24:46  mrmoku: ahh its lower in shr/merge..
Nov 22 20:24:59  mrmoku: I would send it in another patch just to shr/merge repo
Nov 22 20:25:38  mrmoku: and do we have all shr patches in e-wm already in oe.dev?
Nov 22 20:25:43  JaMa: no it's lower in oe
Nov 22 20:25:46  .dev
Nov 22 20:26:30  JaMa: just thought because I bump EFL_SRCREV in the same time that might be a good moment
Nov 22 20:26:33  mrmoku: ah then you need to bump PR globaly and increas EFL rev locally :) I see
Nov 22 20:27:25  JaMa: what do you mean with bump PR globaly?
Nov 22 20:27:35  not SHR specific?
Nov 22 20:27:40  mrmoku: I dont understand I think :)
Nov 22 20:27:52  mrmoku: you wanted to lower PR in e-wm.bb recipe right?
Nov 22 20:27:57  yup
Nov 22 20:28:04  <<<<<<< HEAD
Nov 22 20:28:04  PR = "r9"
Nov 22 20:28:04  =======
Nov 22 20:28:04  PR = "r7"
Nov 22 20:28:07  >>>>>>> origin/org.openembedded.dev
Nov 22 20:28:10  that's my conflict :)
Nov 22 20:28:21  to same level as oe.dev is, because you can, because you bumped EFL_SRCREV
Nov 22 20:28:28  yup
Nov 22 20:28:30  exactly
Nov 22 20:28:47  yes its ok.. if you do it only in shr/merge branch :)
Nov 22 20:28:53  sure :)
Nov 22 20:29:06  thats what I didn't understand :)
Nov 22 20:29:09  will hit builders of other distros from shr/merge branch though
Nov 22 20:29:20  as the bump for EFL_SRCREV is shr specific
Nov 22 20:29:21  hmm
Nov 22 20:29:28  problem?
Nov 22 20:29:41  I guess there are no non-shr builders from shr/merge :)
Nov 22 20:29:44  don't think so...
Nov 22 20:29:45  :)
Nov 22 20:30:06  mrmoku, it won't affect them until next bump anyway
Nov 22 20:30:06  mrmoku, thanks, removing was exactly what I wanted
Nov 22 20:30:15  mrmoku: btw what about pushing those patches from patchwork to same mergeable branch?
Nov 22 20:30:19  as they'll already have up to date files in they repos.
Nov 22 20:30:34  mrmoku: not because I'm author there :P
Nov 22 20:30:34  JaMa: your patchwork bomb?
Nov 22 20:31:08  mrmoku: yes.. not its difficult to track diff between oe.dev+srcpv+xorg-7.5
Nov 22 20:31:14  s/not/now/
Nov 22 20:31:14  JaMa meant: mrmoku: yes.. now its difficult to track diff between oe.dev+srcpv+xorg-7.5
Nov 22 20:31:24  JaMa: let me start to build the image and I will look at what I understand
Nov 22 20:31:39  oki thanks
Nov 22 20:31:41  JaMa: and you're right... there is more patches for e-wm in shr/merge
Nov 22 20:33:31  yes if we start branch from oe.dev. merge just srcpv and xorg-7.5 then rev-list could work to see which commits weren't pushed to oe.dev
Nov 22 20:57:52  JaMa: the navit one I can close at superseded?
Nov 22 20:57:59  s/at/as/
Nov 22 20:57:59  mrmoku meant: JaMa: the navit one I can close as superseded?
Nov 22 20:58:41  mrmoku: yes..
Nov 22 20:58:49  good one less :P
Nov 22 21:02:26  mrmoku, how's the build doing? :)
Nov 22 21:04:10  TAsn: building ;)
Nov 22 21:04:16  (at least I hope so)
Nov 22 21:04:19  let me check
Nov 22 21:04:20  :P
Nov 22 21:04:44  building ecore right now :)
Nov 22 21:04:57  NOTE: Running task 2663 of 7454
Nov 22 21:05:03  for what it's worth...
Nov 22 21:05:35  shr-feed?
Nov 22 21:06:28  playya__: no images
Nov 22 21:06:43  playya__: we bumped EFL_SRCREV :D
Nov 22 21:07:07  hmpf. maybe i should update OE before building the feed :P
Nov 22 21:07:38  do you use AUTOREV for FSO?
Nov 22 21:12:45  playya__: no
Nov 22 21:13:28  playya__: but you can just include the fso-autorev.inc thing in your local.conf?
Nov 22 21:14:09  i added them to some other include
Nov 22 21:14:34  JaMa: hanging on the first one I'm looking at... netbase
Nov 22 21:14:46  we change allow-hotplug to auto for om-gta01/2 there
Nov 22 21:14:55  as allow-hotplug did not work
Nov 22 21:15:10  which might be because we're using busybox ifup/down :P
Nov 22 21:15:33  don't know if it is ok to change that...
Nov 22 21:16:24  JaMa: I think I will split that one into two... one that adds the shr specific files and should be ok
Nov 22 21:16:45  and the other one changing stuff for om-gta01/02 and send that to the list first
Nov 22 21:18:09  mrmoku: ok
Nov 22 21:18:22  shr-settings
Nov 22 21:18:30  ImportError: /usr/lib/python2.6/site-packages/elementary/c_elementary.so: undefined symbol: elm_hoversel_item_del_cb_set
Nov 22 21:18:37  mrmoku: tested 2.6.31?
Nov 22 21:18:41  mrmoku: with drm?
Nov 22 21:22:30  JaMa: ahh, you pushed that?
Nov 22 21:22:34  did not see that :P
Nov 22 21:22:36  not yet
Nov 22 21:22:53  max_posedon: fix is building... we bumped EFL
Nov 22 21:22:59  ah, ok
Nov 22 21:23:36  mrmoku: yes pushed and build localy
Nov 22 21:23:46  mrmoku: haven't tested yet.. still building navits :)
Nov 22 21:24:33  JaMa: if I would only get sound to work on it :|
Nov 22 21:24:48  JaMa: it feels sooo much faster
Nov 22 21:24:51  even without drm
Nov 22 21:24:54  mrmoku: I'll recheck, but iirc I had sound last time i tested
Nov 22 21:24:55  apps start faster :)
Nov 22 21:25:19 * JaMa is a bit stupid today (lack of sleep)
Nov 22 21:25:23  mrmoku: i think sound on .31 should work but all statefiles are incompatible, you might want to try the ones from the OpenWRT.
Nov 22 21:25:54  bisecting which navit revision, looking which one broke it..
Nov 22 21:26:13  first cut was in rev2755 which didn't work
Nov 22 21:26:29  now I iterated to 2754! and it still works
Nov 22 21:26:53  so I checked history and first revision I tried was 2775 not 2755 :)
Nov 22 21:27:51  http://trac.navit-project.org/ticket/506 this just said its broken in svnr2776 :/
Nov 22 21:27:58  PaulFertser: but state files would not explain aplay erroring out, no?
Nov 22 21:28:16  mrmoku: if some wrong dai stuff is used, they might.
Nov 22 21:28:36  mrmoku: but sorry, i think i missed your error description.
Nov 22 21:29:22  root@om-gta02 ~ $ aplay /usr/share/sounds/notify_message.wav
Nov 22 21:29:22  ALSA lib pcm_direct.c:1047:(snd1_pcm_direct_initialize_slave) unable to start PCM stream
Nov 22 21:29:25  ALSA lib pcm_dmix.c:1020:(snd_pcm_dmix_open) unable to initialize slave
Nov 22 21:29:27  aplay: main:608: audio open error: Connection timed out
Nov 22 21:30:38  mrmoku: anything of interest in dmesg? aplay -l output?
Nov 22 21:32:02  PaulFertser: http://shr.pastebin.com/m4c8f39f7 (aplay -l)
Nov 22 21:32:19  PaulFertser: ohhh... hmm
Nov 22 21:32:24  No device for DAI s3c24xx-i2s
Nov 22 21:32:27  might be the one?
Nov 22 21:32:45  mrmoku: i think it appears later in the sequence and gets associated.
Nov 22 21:32:57  asoc: WM8753 HiFi <-> s3c24xx-i2s mapping ok
Nov 22 21:32:59  ok
Nov 22 21:33:28  asoc: WM8753 Voice <-> Bluetooth mapping ok
Nov 22 21:33:28  mapped channel 10 to 2
Nov 22 21:33:33  that's all from dmesg
Nov 22 21:33:56  mrmoku: probably /etc/asoundrc or ~/.asoundrc interfering?
Nov 22 21:34:56  PaulFertser: neither of them exist
Nov 22 21:35:05  PaulFertser: I just have /etc/asound.conf
Nov 22 21:35:11  (and asound.state)
Nov 22 21:35:58  mrmoku: asound.conf, yes. I'd try without it.
Nov 22 21:37:05  PaulFertser: hmm changes behaviour
Nov 22 21:37:05  Playing WAVE '/usr/share/sounds/notify_message.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Mono
Nov 22 21:37:09  aplay: pcm_write:1528: write error: Connection timed out
Nov 22 21:37:21  and in dmesg another
Nov 22 21:37:26  mapped channel 10 to 2
Nov 22 21:37:47  mrmoku: what is DAI mode currently?
Nov 22 21:37:54  PaulFertser: how do I check?
Nov 22 21:38:01  mrmoku: with alsamixer
Nov 22 21:39:06  PaulFertser: DAI Mode [DAI 0]
Nov 22 21:40:19  mrmoku: looks reasonable. I think i must try myself to say for sure what's going on. But i'm going to sleep soon, so sorry, not today :(
Nov 22 21:40:32  PaulFertser: no problem, thanks :)
Nov 22 21:40:43  PaulFertser: just that it feels so much faster the new kernel :P
Nov 22 21:41:45  PaulFertser: you you have numbers from that x11perf glamo benchmark from your gta01? :)
Nov 22 21:43:32 * JaMa bumbles something about warmer food from microwave, when microwave is really on..
Nov 22 21:43:38  JaMa: glamo benchmark and gta01 are mutually exclusive
Nov 22 21:44:09  PaulFertser: but you know what I mean, don't you?
Nov 22 21:45:09  mrmoku, did you remember to bump kernel PR?
Nov 22 21:45:18  PaulFertser: http://pastebin.ca/1678180
Nov 22 21:45:31  JaMa: i didn't test gta01 yes.
Nov 22 21:45:33  yte
Nov 22 21:45:39  mrmoku, (just making sure)
Nov 22 21:45:44 * PaulFertser should go sleep
Nov 22 21:46:17  PaulFertser: ah, ok.. 
Nov 22 21:46:34  TAsn: me?
Nov 22 21:46:42  mrmoku, yes.
Nov 22 21:46:55  no
Nov 22 21:46:59  before building the new image and feed :|
Nov 22 21:47:06  so kernel modules won't get stripped :(
Nov 22 21:47:26  if you can stop the build, bump PR and start building the image again :P
Nov 22 21:47:34  TAsn: http://bugs.openembedded.org/show_bug.cgi?id=1182
Nov 22 21:47:44  I'm happy I thought about it.
Nov 22 21:48:00  TAsn: I'll retest my modules first.. w8
Nov 22 21:48:13  TAsn: hmm... probably not even needed P
Nov 22 21:48:43  mrmoku, why not?
Nov 22 21:48:52  JaMa: without BB_CLONE_FOR_SRCREV... does it update localcount for the kernel?
Nov 22 21:48:56  +1?
Nov 22 21:49:14  yes
Nov 22 21:49:18  TAsn: see :)
Nov 22 21:49:25  mrmoku: but revision didn't changed..
Nov 22 21:49:37  hmm
Nov 22 21:49:39  s/d../../
Nov 22 21:49:39  JaMa meant: mrmoku: but revision didn't change..
Nov 22 21:49:39  mrmoku, see :P ?
Nov 22 21:49:46  ;)
Nov 22 21:49:48  I see!
Nov 22 21:50:01 * mrmoku is blind
Nov 22 21:50:03  :P
Nov 22 21:50:19  JaMa: it just won't build it, right?
Nov 22 21:50:30  TAsn: I need to recheck if that fixed function was called soon enough..
Nov 22 21:50:31  mrmoku, please if you can, bump PR for shr kernel and rebulid image. :)
Nov 22 21:50:46  JaMa, it'll be there for packages
Nov 22 21:50:47  TAsn: I see stripped modules in .tgz, but not sure if .ipk have them..
Nov 22 21:50:51  and those make the image
Nov 22 21:51:00  TAsn: btw. do you read e-devel?
Nov 22 21:51:02  JaMa, ipk is after the tgz, isn't it?
Nov 22 21:51:06  mrmoku, yes.
Nov 22 21:51:12  TAsn: I'll just check if updated kernel-module.ipk are stripped in my build
Nov 22 21:51:14  TAsn: no
Nov 22 21:51:28  TAsn: python-elementary callback breakage :P
Nov 22 21:51:35  mostly "mark all as read"
Nov 22 21:51:43  mrmoku, not now :P
Nov 22 21:52:04  I'm really studying, just came to make sure you are getting us stripped modules :)
Nov 22 21:52:34  TAsn: guess what
Nov 22 21:52:35  r43900 | billiob | 2009-11-22 22:25:25 +0100 (Sun, 22 Nov 2009) | 1 line
Nov 22 21:52:36  JaMa, let me (and more importantly mrmoku) know.
Nov 22 21:52:39  pyelm: rewrite callbacks handling
Nov 22 21:52:44  :)
Nov 22 21:52:50  mrmoku, hehe. ;P
Nov 22 21:52:52  good thing I did not wait for it ;)
Nov 22 21:53:06  good thing?
Nov 22 21:53:20  TAsn: bad news.. /lib/modules/2.6.31/kernel/sound/core/snd.ko: ELF 32-bit LSB relocatable, ARM, version 1 (SYSV), not stripped
Nov 22 21:53:30  JaMa, :|
Nov 22 21:53:31  because we would have to fix all pyelm callbacks :P
Nov 22 21:53:35  dos1 would not like it
Nov 22 21:53:40  oh. ;)
Nov 22 21:53:47  but now they work?
Nov 22 21:53:48  JaMa: do you want to bump the kernel?
Nov 22 21:53:58  TAsn: hoversels should work with what we have now
Nov 22 21:54:08  mrmoku, no need, as it's broken (kernel)
Nov 22 21:54:10  mrmoku, cool. ;P
Nov 22 21:54:15  mrmoku, please send an email to shr-devel
Nov 22 21:54:21  mrmoku: you can, but only difference will be modules-dbg.tgz and modules.tgz
Nov 22 21:54:22  asking all the python devs
Nov 22 21:54:30  TAsn: you dan download stripped tgz :)
Nov 22 21:54:32  can
Nov 22 21:54:35  to prepare themselves to a bump in the far future. ;)
Nov 22 21:54:41  JaMa: ahh, so your modstripping is not yet perfeect?
Nov 22 21:54:42  JaMa, I don't care about myself
Nov 22 21:54:45  I'll apply better fix..
Nov 22 21:54:45  I care about the users. :|
Nov 22 21:55:01  you're such a nice dev.. :P
Nov 22 21:55:08  and users shoudl not d/l modules.tgz
Nov 22 21:55:33  TAsn: I know, I'll push similar patch as that oebug I pasted here..
Nov 22 21:55:57  JaMa, nah, sounds fine with me ;P
Nov 22 21:55:59  so push it
Nov 22 21:56:05  force mrmoku to bump kernel PR
Nov 22 21:56:19  and force him to rebuild image
Nov 22 21:56:25  and then we'll be set :P
Nov 22 21:56:58  JaMa, do you by any chance remember the date of the commit that broke module_stripping? (you gave me a link to it the other day)
Nov 22 21:57:04  TAsn: nothing is set
Nov 22 21:57:09  e-wm failed ;)
Nov 22 21:57:15  sweet!
Nov 22 21:57:17  :P
Nov 22 21:57:22  mrmoku, do-patch? 
Nov 22 21:57:28  we have a patch that got upstream
Nov 22 21:57:29  yup
Nov 22 21:57:32  TAsn: its in commit message of my patch
Nov 22 21:57:38  mrmoku, what patch?
Nov 22 21:57:39  file://illume-flow.patch
Nov 22 21:57:46  mrmoku, I told you it got upstream!
Nov 22 21:57:51  TAsn: but it broke module_stripping which was used too late anyways..
Nov 22 21:57:54  what patches do we apply atm?
Nov 22 21:58:05  JaMa, i c.
Nov 22 21:58:18  mrmoku, what patches do we apply atm?
Nov 22 21:58:20  (to e-wm)
Nov 22 21:58:36  TAsn: http://shr.pastebin.com/m2c6bb2db
Nov 22 21:59:05  TAsn: there is the set of treviños patches
Nov 22 21:59:14  I'm not sure if they hit upstream too
Nov 22 21:59:31  mrmoku, just check
Nov 22 21:59:39 * mrmoku is svn noob
Nov 22 21:59:42  anyhow, except from illume-flow there's only one patch I made
Nov 22 21:59:54  and that is not upstream ;)
Nov 22 21:59:55  mrmoku, I have no idea how to do it as well. ;|
Nov 22 22:00:07  mrmoku, yep.
Nov 22 22:00:09  will just retry without illume-flow for now
Nov 22 22:03:32  ahh... treviños patches have maxrev anyway :)
Nov 22 22:08:25  mrmoku, forwarded the mail about elm callbacks from e-devels
Nov 22 22:09:04  TAsn: good
Nov 22 22:15:42  good read. thanks TAsn 
Nov 22 22:15:50  interesting
Nov 22 22:16:08  this will allow shr-settings to get rid of some black magic
Nov 22 22:16:46  spaetz__, thank mrmoku for telling me to search e-devel's mails for pyelm stuff :P
Nov 22 22:16:53  spaetz__, you'll have to do it in a branch though
Nov 22 22:16:59  hehe
Nov 22 22:17:02  as we haven't yet bumped to a sufficient efl :|
Nov 22 22:17:08  TAsn: why? aren't we bumping efl already?
Nov 22 22:17:14  ahh, ok
Nov 22 22:17:18  spaetz__, will break shr-settings
Nov 22 22:17:19  ...
Nov 22 22:17:37  but is that efl callback will segfault fix in?
Nov 22 22:17:44 * spaetz__ checks new EFL rev
Nov 22 22:18:08  shr-settings is the only pyelm app we use, right? (except for shr-today which is being rewritten in c anyway)
Nov 22 22:18:13  maybe shr-launcher?
Nov 22 22:18:16  (never used it, no idea)
Nov 22 22:18:22  intone?
Nov 22 22:18:28  mokonnect! :(
Nov 22 22:18:35  will have to tell f4t
Nov 22 22:19:09  mokonnect is also python
Nov 22 22:19:15  I think elm rather than gtk
Nov 22 22:19:21  spaetz, I know
Nov 22 22:19:25  that's what I said. ^
Nov 22 22:19:31   mokonnect! :(
Nov 22 22:19:31   will have to tell f4t
Nov 22 22:19:39  yep
Nov 22 22:20:03  sent him an email.
Nov 22 22:20:10  mrmoku, if mokonnect is on autorev
Nov 22 22:20:14  Or rather than investing time in shr-settings I could resurrect my vala shr-config again :)
Nov 22 22:20:19  please drop the autorev and fixate the rev to 111
Nov 22 22:20:20  TAsn: it's not.
Nov 22 22:20:25  spaetz__, good.
Nov 22 22:20:33  spaetz__, please do.
Nov 22 22:20:37  we have very little on autorev now
Nov 22 22:20:44  spaetz__, Though I must warn you.
Nov 22 22:20:58  shr-wizard also uses shr-setting's modules
Nov 22 22:21:09  which means that won't solve it entirely :P
Nov 22 22:21:15  we really have to fix shr-settings.
Nov 22 22:21:20  yep, I skimmed through the logs, seeing that.
Nov 22 22:21:34  happy to have an shr-wizard
Nov 22 22:21:54  will rewrite that in vala as soon as my shr-config is finished ;)
Nov 22 22:22:00  spaetz__, thanks dos. he pwns. :P
Nov 22 22:22:05  (sometimes)
Nov 22 22:22:07  he really does
Nov 22 22:22:07  ;)
Nov 22 22:22:14  whenver he doesn't suck :)
Nov 22 22:23:35  spaetz__, exactly what the "(sometimes)" meant :P
Nov 22 22:24:17  I'm off to bed.
Nov 22 22:24:22  I'm not ready at all for my test.
Nov 22 22:24:29  lets just hope I'll Ace it. ;P
Nov 22 22:24:32  ciao
Nov 22 22:24:36  and gl for me.
Nov 22 22:24:36  TAsn: good luck
Nov 22 22:24:43  :)
Nov 22 22:24:47  ok, night
Nov 22 22:24:57  mrmoku, I need it :P
Nov 22 22:24:57  good luck
Nov 22 22:25:03  thanks.
Nov 22 22:26:52 * JaMa has gpe-icons merger for postinst :)
Nov 22 22:31:19  mrmoku: BTW; just putting all other e keyboards in onepackage sounds also better to me
Nov 22 22:32:55  spaetz__: yup
Nov 22 22:33:20  spaetz__: but to the thing that we're interested in... no response :|
Nov 22 22:34:33  spaetz__: btw... FYI I turned off BB_GIT_CLONE_FOR_SRCREV on the buildhost
Nov 22 22:36:04  anyway... me is off to bed
Nov 22 22:36:12  gnight
Nov 22 22:37:17  me too, night
Nov 22 22:37:37  mrmoku|away: ok turning off BB thingie. makes sense
Nov 22 22:59:01  wtf? Running task 1040 of 10309
Nov 22 22:59:11  what has been added to the feed?
Nov 22 22:59:35  wow i can restart x without entering pin, nice
Nov 22 23:03:24  fredrin: this sounds rather correct, as gsm service isn't depending on X basically
Nov 22 23:03:55  that's how it should be imho 
Nov 22 23:04:28  yep
Nov 22 23:04:47  DocScrutinizer-8: how's the sound sanitaizing going? 
Nov 22 23:06:03  heh, I don't manage to get any real work done as a solofighter for quite several months now. So probably will take another few days ;-)
Nov 22 23:09:17  not help :( 
Nov 22 23:09:54  what's the problem? 
Nov 22 23:14:52  freesmartphone.org: 03mickey 07cornucopia * r7378029e90ac 10/fsotimed/ (configure.ac src/plugins/sync_time/plugin.vala): fsotimed: call adjtime(2) upon time report
Nov 22 23:19:33  freesmartphone.org: 03mickey 07cornucopia * rce379e24feb7 10/ (15 files in 14 dirs): all: move daemons into @prefix@/sbin/
Nov 22 23:58:48  can't start 'settings' app, maybe becaue this elf update? 
Nov 22 23:59:36  fail
Nov 23 00:27:43  mickey|zzZZzz: just out of curiosity: how does the pre export the controlls for its vibrator?
Nov 23 00:28:23  iirc not at all through a class device
Nov 23 00:28:28  probably some ioctls :/
Nov 23 00:28:40  don't have it handy right now though
Nov 23 00:28:46  alphaone or shoragan may know
Nov 23 00:30:25  both hopefully sleeping right now though ;)
Nov 23 00:30:31  same as i should...
Nov 23 00:31:32  larsc: what? 
Nov 23 00:31:34  btw., don't think i'm not massively grateful about you adopting the kernel, it's just that i have always a hard time adjusting userland 
Nov 23 00:31:57  especially when some people are probably still on .24
Nov 23 00:32:05  mickey|zzZZzz: it's fast
Nov 23 00:32:06  gets a bit messy with all the cases
Nov 23 00:32:29  good
Nov 23 00:32:33  nah, more of an clean up of the apps 
Nov 23 00:33:16  fredrin: clean up? you mean dropping support for older kernels?
Nov 23 00:34:09  mja, but at least the app devs could read thrugh their code and make sure it works with .31 kernel 
Nov 23 00:34:32  and get included in shr by default
Nov 23 00:34:50  or in the feed with thos bb receipes things 
Nov 23 00:34:57  or just use a library that abstracts away all this
Nov 23 00:35:11  mickey|zzZZzz: you have to see my dilemma. It's either trying to support a horrible broken api or breaking userspace. I choose the second.
Nov 23 00:35:36  larsc: i don't quite understand what was so horrible with the previous name, but it's your call anyways
Nov 23 00:35:51  mickey|zzZZzz: And this is not really about the vibrators sysfs name. its really really easy to create a symlink to the old location
Nov 23 00:35:57  stupid to hardcode sysfs paths to programs in the current situation
Nov 23 00:36:30  the broken api are the -pm devices
Nov 23 00:36:37  ah
Nov 23 00:36:43  in fact i considered them quite handy 
Nov 23 00:36:47  from userspace point of view
Nov 23 00:36:54  at last some form of peripheral control standard
Nov 23 00:37:04  i'm scared what comes now...
Nov 23 00:37:11  kek
Nov 23 00:37:30  what's that?
Nov 23 00:38:16  the root of the problem is that obviously linux does not support any class device for peripheral power control
Nov 23 00:38:34  or does it now?
Nov 23 00:38:38  and are you moving to that?
Nov 23 00:40:28  well, at least symlinks are considered mandatory (by me at least ;-P). See how long modules.conf and conf.modules coexisted
Nov 23 00:40:59  no it does not. and yes thats the main problem. and no don't feel like i have the proper knowledge to create a api to solve that
Nov 23 00:41:29  ok, but you still want to take what we know and love away :D
Nov 23 00:41:50  sort of.
Nov 23 00:41:56  then i sincerely hope you have something better to substitute them with
Nov 23 00:42:11  bluetooth and wlan can be controlled by rfkill
Nov 23 00:42:48  uh oh. to my knowledge rfkill is only concerned with RF, not with powering up/down the whole device
Nov 23 00:43:03  although it may make sense to combine that
Nov 23 00:43:21  mickey|zzZZzz: I had that discussion with PaulFertser already.
Nov 23 00:43:36  yep
Nov 23 00:43:36  mickey|zzZZzz: And imo it's better to be implicit here
Nov 23 00:43:54  I was about to suggest to as Paul
Nov 23 00:44:05  asK
Nov 23 00:44:12  ok, gps and gsm?
Nov 23 00:44:13  turning of the radio will also turn of power, cause there is no sense in keeping it runinng
Nov 23 00:44:50  hmm, are you sure?
Nov 23 00:45:04  gsm probably has to stay and it will be renamed back to neo1973-...
Nov 23 00:45:13 * DocScrutinizer-8 muses about rx-only modes
Nov 23 00:45:34  DocScrutinizer-8: our hardware doesn't support it
Nov 23 00:45:53  k, maybe *our* hw doesn't
Nov 23 00:46:51  though GSM (ok, OT) actually does, well kinda
Nov 23 00:47:09  gsm is a completely different beast anyways
Nov 23 00:47:12  gsm is no controlled through rfkill
Nov 23 00:47:20  as RF control goes through at commands in the first place
Nov 23 00:47:30  gps is an interesting corner case
Nov 23 00:47:43  as this has some complex patterns
Nov 23 00:47:49  involving suspend/resume etc.
Nov 23 00:48:10  absolutely
Nov 23 00:48:13  oh well
Nov 23 00:48:15  just keep us posted
Nov 23 00:48:18  i would like to move gps controll completly to userspace
Nov 23 00:49:16  cause it is basically just turning regulators on and off
Nov 23 00:49:28  and there is already an sysfs api for that
Nov 23 00:49:31  hmm, seems tricky for device suspend
Nov 23 00:50:19  muhahahahahah
Nov 23 00:50:22  this is awesome
Nov 23 00:50:22  afaik userspace can register suspend/resume events
Nov 23 00:50:36  if u like openstreetmaps... u'll like the new elementary map widget
Nov 23 00:50:44  s/register/register for/
Nov 23 00:50:44  larsc meant: afaik userspace can register for suspend/resume events
Nov 23 00:50:46  it is OSM... in a scrollable, zoomable widget
Nov 23 00:51:01  larsc: yep
Nov 23 00:51:35  hi raster  :-)
Nov 23 00:51:45  DocScrutinizer-8: yoyo
Nov 23 00:51:52  werner inspired me to go this way
Nov 23 00:52:44  man. this is totally awesome
Nov 23 00:53:26 * mwester_ suggests that perhaps before such changes are made, the one making those changes develop a prototype user-space that can be used by the various distros to validate that in fact doing all this in userspace will work correctly in all corner cases.
Nov 23 00:56:54  quite reasonable suggestion. I pondered all the time about suspend/resume delays and struggle of last days to get a resume time <1sec
Nov 23 00:57:12  :D You know, if we simply have the kernel export the GPIO pins and timers, then we could export EVERYTHING to userspace, and there would never be any reason to have arguments about naming again!
Nov 23 00:57:43  not that the slow resume was related. just shows we want fast resume
Nov 23 00:58:43  mwester_: the kernel exports the gpio pins already
Nov 23 00:59:05  hehehe lol
Nov 23 00:59:34  Cool! Then let's just remove ALL those pesky silly sysfs things that clutter up the tidiness of the kernel, and push it all into userspace! :D
Nov 23 00:59:39  so lets move to a userland-kernel then
Nov 23 00:59:52  We'll need an RT scheduler
Nov 23 01:00:36  and a watchdog
Nov 23 01:01:49  mwester_: sched_setscheduler(getpid(), SCHED_FIFO, ¶m);
Nov 23 01:02:11  wait, isn't that what hird is all about? ;-D
Nov 23 01:02:35  Cool.  So tell me again why implemented LEDs and GSM and power management in-kernel? 
Nov 23 01:03:32  mwester: most of that indeed doesn't belong there :)
Nov 23 01:03:54  I thought so.
Nov 23 01:04:09  Let's re-architect Linux for the gta0x devices!
Nov 23 01:04:27  for sure forced shutdown of LED in kernel is wrong
Nov 23 01:04:36  during suspend
Nov 23 01:04:43  mwester: particularly the LEDs don't really make sense. it's two perfectly unrelated subsystems glued together by in-kernel policy.
Nov 23 01:05:10  aye
Nov 23 01:05:21  kernel shoudl govern what is HAS to
Nov 23 01:05:26  not what it can
Nov 23 01:06:34  and that's basically power management (core voltage etc) first instance, as userland won't ever do this while cpu core borked
Nov 23 01:07:05  as long as userspace si involved in suspend/resuume and any power management decisions - controsl should probably be put there and kernel do the minimum needed for bootup to have everything in a sane state so userspace can make the decisions from there
Nov 23 01:07:53  hmm, guess that's what I intended to say
Nov 23 01:08:14  in the bad old days of ATM, there was a beautiful example: the signaling (q.2931 plus a vaguely tcp-ish transport layer protocol). fore systems, the market leader, bravely implemented it in the solaris kernel. i did it in user space on linux. we had great fun watching those suns crash.
Nov 23 01:09:16  XD
Nov 23 01:12:26  otoh I don't feel exactly happy with userland having to deal with hw diff details more than absolutely needed
Nov 23 01:13:23  DocScrutinizer: if you don't already have a good abstraction, it's the lesser evil, even kernel design issues aside
Nov 23 01:13:26  it depends whihc bits of userland
Nov 23 01:13:40  i.e. fuse is great, while gsm AT@POFF in userland feels wrong to me
Nov 23 01:13:46  i prefer my "userland driver" to be able to segv rather than kernel oops
Nov 23 01:13:51  (if its sane/possible)
Nov 23 01:14:03  DocScrutinizer: otherwise, you get an interface that tries to give you a nice abstraction, and a couple of months later, when you figure out that it needs to handle this or that as well, the interface changes, etc. etc.
Nov 23 01:14:13  theres multiple layers of userland
Nov 23 01:14:21  the lower layers probably need to know about the hw
Nov 23 01:14:24  DocScrutinizer: that's where mickey comes in ;-)
Nov 23 01:14:25  X for instance
Nov 23 01:14:44  even audio - u need to have some abstracted idea of what channels go where on which hw
Nov 23 01:15:05  that's why I emphasized "it *feels* wrong to *me*"
Nov 23 01:15:23  raster: alsa state files :)
Nov 23 01:15:26 * wpwrak giggles
Nov 23 01:15:33  waaaaaaaaaaaaaaaah
Nov 23 01:15:54  but obviously you're right, raster 
Nov 23 01:17:37  wpwrak:  fun fun :)
Nov 23 01:17:47  DocScrutinizer-8:  not always.. but reality is.... userspace already needs to know
Nov 23 01:18:16  so... one way or another... the logic is already there... no extra ugliness is created
Nov 23 01:48:19  Anyone know what's going on with the port of FSO to the Pre?
Nov 23 02:02:35  Does anybody know if the instructions to build SHR at http://shr-project.org/trac/wiki/Building%20SHR are still accurate?
Nov 23 02:08:00  budfive_: they appear to work for me, so far.
Nov 23 02:09:43  rozzin: ok, thanks. Didn't work earlier for me, but I'm trying again. I can work on it, but wanted to know if I was going down the wrong path. thanks.
Nov 23 02:09:55  oh, wait--just as I said that, it bombed-out. hm.
Nov 23 02:10:15  After 9 hours.
Nov 23 02:31:58  well, anyway I really wonder if an app would learn anything important from the gta01 prefix of *::vibrator. And particularly if an app is xupposed to find a gta02::vibrator as well when it knows it's running on a gta01
Nov 23 02:37:29  DocScrutinizer-8: what's that about vibrators and prefixes?
Nov 23 02:37:49  see kernel-ML
Nov 23 02:38:43  "Vibrator redesign" thread?
Nov 23 02:40:39  dunno, mail is other room atm. mails of a few hours ago
Nov 23 02:41:29  Hm. Well, not that thread then.
Nov 23 02:41:53  Must have been "Renaming of devices in 2.6.31".
Nov 23 02:42:03  yep
Nov 23 02:54:41  Is the vibrator-rename going to cause a ripple up through the FSO D-Bus API?
Nov 23 02:59:46  if it is then, fso is badly designed
**** ENDING LOGGING AT Mon Nov 23 02:59:56 2009