**** BEGIN LOGGING AT Mon Mar 09 02:59:57 2009 Mar 09 03:07:47 After flashing to the latest SHR testing, many applications from opkg.org no longer install/run due a missing dependency (gtk+-fastscaling). I saw a thread stating that gtk+-fastscaling has been renamed gtk+. The thread goes on to state "maybe its authors should define something like a "Provides, Replaces, Conflicts: gtk+-fastscaling" as in Debian". Is there a way I can do this manually, or should I wait? Mar 09 03:31:54 bgat, yeah, that was popular today :) Mar 09 04:38:55 hey Mar 09 04:39:27 i was told fso will do the wifi-backend-stuff and'll provide an API Mar 09 04:39:32 is it implemented yet? Mar 09 04:52:41 Hello :D Mar 09 05:20:22 :) Mar 09 06:12:29 freesmartphone.org: 03daniel 07framework * re1c9aaf50586 10/framework/subsystems/ogsmd/gsm/parser.py: (log message trimmed) Mar 09 06:12:29 freesmartphone.org: ogsmd: Make sure we keep valid prefixes over continuation lines Mar 09 06:12:29 freesmartphone.org: This is important when submitting commands with PDU. Mar 09 06:12:29 freesmartphone.org: After receiving the "\r\n> " continuation the parser would reset the Mar 09 06:12:31 freesmartphone.org: valid prefixes so in case of Mar 09 06:12:33 freesmartphone.org: AT+CMGW=23 Mar 09 06:12:35 freesmartphone.org: > Mar 09 06:16:24 PaulFert`, johnsu01: Should be fixed now Mar 09 06:16:33 At least it fixed the problem for me. Mar 09 06:16:45 Please test and see if it causes any new issues Mar 09 06:53:11 alphaone: Thanks! Btw, should zhone issue AckMessage now (due to SMS api changes)? Looks like it's not implemented. Mar 09 06:53:56 PaulFert`: No since zhone still uses the SIM to store and receive messasges Mar 09 06:59:46 alphaone: i see. So it should be ok. Still something is wrong with handling messages in Zhone. E.g. #331... Mar 09 07:03:02 I closed my own ticket #304 as invalid. Well, it happens to me too :) Mar 09 07:05:14 Hehe :-) Mar 09 07:05:44 Yeah, the zhone issue is (afaik) just zhone being lazy and not rereading the status of the message Mar 09 07:06:24 And actually reread just that one message, not retrieve the whole messagebook as it is doing at the moment... Mar 09 07:06:46 Actually it reread the message here. Mar 09 07:07:06 At least when sending a message Mar 09 07:09:01 But when receiving a message it requests the messagebook once (in which the new message turns up as unread) and then doesn't reread the messagebook or the message after it has been read. Mar 09 07:15:03 alphaone: ok, this last read/unread thing doesn't bother me much. I'll look closer if i see any issues with messages. BTW, what will happen now if i compose a message being out of coverage? Mar 09 07:16:29 Or try to suspend before receiving "SIM is ready" signal or after i write the message but before it's sent? Mar 09 07:17:33 before receiving SIM is ready, I think you will wake up again Mar 09 07:18:05 after you write the message before it's being sent shouldn't be a problem I believe. Mar 09 07:18:21 Composing a message out of coverage should just give you an error. Mar 09 07:19:27 But yeah, I guess these sort of corner cases haven't been tested enough. Mar 09 07:21:09 Looks like that, %CSTAT is enabled once and never disabled, so the modem should wake up the phone. Mar 09 07:23:45 alphaone: (write the message and suspend): so, you think that calypso will wake up the phone after the message changes its status from unsent to sent? To me updating the message list in zhone didn't work even without suspending :) Ok, if you're interested, i'll start to look closer at all the annoyances i'm encountering with messages. Mar 09 07:24:15 alphaone: (composing out of coverage): an error? Why can't it simply store it on SIM as unsent? Mar 09 07:24:34 PaulFert`: A closer look would be appreciated, yes :-) Mar 09 07:24:46 Zhone will do that, yes (storing unsent) Mar 09 07:24:58 However, when sending fails it will not retry Mar 09 07:25:37 PaulFert`: The message should change its status from unsent to sent as soon as the SendStoredMessage call returns Mar 09 07:25:59 Which is the time we reread the Messagebook Mar 09 07:26:44 Works here, but maybe I just have so much messages stored on the SIM that the time in between is large enough. Mar 09 07:28:54 alphaone: ok, from now on i'll try to report properly every related issue :) BTW, would you mind adding a button somewhere to force rereading the messagebook and sending unsent messages? Being a hack, it'd be still better than restarting Zhone, imho. Mar 09 07:30:04 PaulFert`: I wouldn't mind applying a patch for that :-) Mar 09 07:31:50 alphaone: ahem... Ok, let's see if i'll need it enough to overcome my laziness. Mar 09 07:35:34 Yeah, I really don't like touching zhone... Mar 09 07:38:54 alphaone: now i tried sending a message. I didn't see it in the list until it appeared as "sent". I understand that touching Zhone is not fun and it's not supposed to be fully-functional app, otoh i see no other way to use the phone on Debian (and to test frameworkd properly). Mar 09 07:39:50 freesmartphone.org: 03daniel 07specs * rf2021027f3b9 10/ (6 files in 3 dirs): otapi: Update org.freesmartphone.GSM.SMS API Mar 09 07:43:31 morning Mar 09 07:44:48 PaulFert`: Actually from looking at the code it will try to send the message and after it either fails or succeeds it will reread the Messagebook. So the message will either appear as sent if sending succeeds or as unsent if it fails. Mar 09 07:46:52 alphaone: ok, that's not exactly intuitive, but not that bad. Too bad that we have no good way to send unsent messages after disabling the timer... Mar 09 07:47:39 PaulFert`: You could just have a "resend" button in the context menu Mar 09 07:47:55 Which would then just AT+CMGS=index Mar 09 07:48:57 morning Mar 09 07:57:02 spaetz: morning... you have testing installed somewhere? Mar 09 07:57:57 spaetz: if so... could you verify if resume by call works for you? Mar 09 07:58:28 spaetz: phone resumed and vibrated for me... but I had no sound Mar 09 07:58:36 spaetz: and vibration did not stop Mar 09 07:59:58 * mrmoku` moves to office now Mar 09 08:05:25 Mirko-paroli: Hey :) Please don't consider my questions or messages as a personal offence. Please. It's just that you're keeping silence on the issues that i consider to be of great importance... You're the frontman of the Paroli project and people are asking you personally about UI design guidelines, targets etc. Imho it's somehow inappropriate to leave all these questions (excluding my rants) unanswered. Mar 09 08:07:39 http://www.paroli-project.org/about-paroli/ Mar 09 08:07:45 answers any questions about UI design Mar 09 08:07:50 and "guidelines" Mar 09 08:23:14 Mirko-paroli: greate writeup on the ideas behind application design goals and techniques. But nothing about the actual UI you're going to present your users soon (no ui design mockups, no ui guidelines, i don't see anything like that). You saw the complaints on the community ML, don't you understand why those people are complaining? Moreover, you still don't answer a thing about "the ui design department". If they're part of the project, ... Mar 09 08:23:21 ... the community is supposed to know them, don't you agree? Mar 09 08:23:59 http://www.paroli-project.org/running-paroli/launcher/ Mar 09 08:24:04 and click your way through the apps Mar 09 08:24:10 all with "mockups" Mar 09 08:29:27 Mirko-paroli: Looks like the list scrolling will be done by swipes... Are you sure it's a good idea for a resistive touchscreen operated by finger? But you know, the most controversial thing is the "pseudo-fullscreen" (which is technically not fullscreen) mode. Integration aside, is the GUI resizable? If yes, then why not run it under the control of a WM? Why not make it easy switchable? Mar 09 08:30:06 send a patch I'll apply it Mar 09 08:30:21 my day only has so many hours Mar 09 08:30:28 and I have to focus on something Mar 09 08:30:34 right now resizing etc is not my focus Mar 09 08:31:37 Mirko-paroli: fair enough. Can you answer the last question about the "design department" please? Mar 09 08:32:08 everyone who is working on paroli influences the ui and the design Mar 09 08:32:17 so if you or anyone else wants to have a say Mar 09 08:32:29 do it as in any other OS project Mar 09 08:32:31 join it Mar 09 08:32:38 earn credit and then use it Mar 09 08:35:49 Mirko-paroli: to me as the community member it seems like there exist some hidden force that can influence on UI design in non-obvious non-predictable ways. And you don't seem to deny it... Either we talk different languages or i can't formulate questions or you can't answer. :( Mar 09 08:36:50 as I told you last time we talked, questions end in '?' and imho do not contain subtle accusations Mar 09 08:38:16 Mirko-paroli: Does the force that is hidden to the community and can influence ui design decisions ("the OM ui design department") exist? (now is that better?) Mar 09 08:39:16 yes as every other company we also have a ui department, but seen that you know of it it can't be really hidden, right? Mar 09 08:41:42 Mirko-paroli: I've never seen any letter in the public mailing list from them, nor the list of people (with their names and e-mails) involved. I call it "hidden", i think that corresponds to the dictionary definition. Mar 09 08:43:24 ok if you wanna call that hidden my view is blurred in that sense Mar 09 08:43:30 what's your point? Mar 09 08:44:29 Mirko-paroli: every force that has an influence on the project must act in a community-transparent way. Is it the case now? Mar 09 08:45:07 why does every force have to act in a "community-transparent way" Mar 09 08:45:09 ? Mar 09 08:45:29 I will not get my gf to write mails whenever she is mad at me and publicize them Mar 09 08:45:35 but it is a force that influences it Mar 09 08:45:39 one among many Mar 09 08:48:43 Mirko-paroli: fair point. But i guess "the ui design department" doesn't influence you in the same way as your gf... Well, i think that's enough. I've got the answers to my questions. Would you mind if i post the script to the community mailing list? Mar 09 08:52:24 actually not really as I am not certain that you won't misrepresent my answers, sorry to have to say that Mar 09 08:53:14 Mirko-paroli: I'd like to send the transcript unaltered, so every sane person will intepret it the right way. Mar 09 08:53:47 Mirko-paroli: i'll refrain from comments, just the irc session with my and your messages. Mar 09 08:54:20 sounds fair Mar 09 08:54:50 and I respect you for asking Mar 09 08:54:52 thx Mar 09 08:56:40 Mirko-paroli: thank you for your work and sorry if i annoyed you. Mar 09 08:56:57 i don't feel annoyed I just think it is unfair treatment Mar 09 08:57:10 but that does not only mean in regards to you but others also Mar 09 08:57:22 in the whole time of the project there has been 1 - 3 patches Mar 09 08:57:29 and at least 40 complaints Mar 09 08:57:44 and most of them because someone didn't check the code or the website before Mar 09 08:59:40 Mirko-paroli: (unfair treatment) some community members feel that they're treated in an unfair way by OM-the-company. Being part of the company you're doomed to get that back. Mar 09 09:01:00 so because some community members treat me unfairly I should do the same to everyone outside of OM and than that would be just? Mar 09 09:01:14 I should ignory anyone outside? Mar 09 09:01:17 ignor Mar 09 09:01:18 e Mar 09 09:03:25 Mirko-paroli: I think no, you shouldn't. But not being fully open contributes to the community frustration. You don't tell us the names of the design department people, that looks suspicious to be honest. Mar 09 09:03:42 hm? Mar 09 09:03:46 that was not my question Mar 09 09:04:07 you defend "some community members" by saying they feel mistreated by "OM-the-company" Mar 09 09:04:39 I ask, so if I do the same as in "some OM-the-company members" feel mistreated by "the community" Mar 09 09:05:07 Mirko-paroli: i hope you'll do your best to not treat anybody unfairly. Mar 09 09:05:18 I certainly do Mar 09 09:05:25 but again you missed my point Mar 09 09:05:29 why Mar 09 09:05:38 is it ok for "some community members" Mar 09 09:05:55 to treat me unfairly Mar 09 09:06:12 because I am part of "OM-the-comp" Mar 09 09:06:18 but never ever the other way around Mar 09 09:06:24 it is a simple question Mar 09 09:06:28 Mirko-paroli: no, it's not ok. Mar 09 09:07:42 I agree Mar 09 09:07:45 in both ways Mar 09 09:07:50 Mirko-paroli: but for the community members to act fairly, they need to see you as pro-community and not pro-company. Mar 09 09:07:51 and I have said so several times Mar 09 09:07:59 nope Mar 09 09:08:06 treat others as you want to be treated Mar 09 09:08:20 acting fairly is not a question of who you talk to Mar 09 09:08:29 it is a question of who you are Mar 09 09:08:35 a question of character Mar 09 09:08:42 for me it is that simple Mar 09 09:09:26 and anything else is a lame excuse Mar 09 09:10:08 Mirko-paroli: it's sometimes fair to put a criminal to jail. But if he's innocent it's not fair. Some your actions might have been intepreted as "criminal". That's how i see it. Mar 09 09:11:46 Mirko-paroli: ++ Mar 09 09:12:36 $$ Mar 09 09:13:01 got an example? Mar 09 09:13:04 ask and I answer Mar 09 09:14:04 Mirko-paroli: ok, thanks a lot again. CU Mar 09 09:14:14 but for the company members to act fairly, they need to see you as pro-company and not pro-community Mar 09 09:14:25 PaulFert`: does that make any sense to you? Mar 09 09:14:41 lol, open source Mar 09 09:15:04 reverse you accusations and check if you would accept them if someone from OM had voiced them Mar 09 09:15:25 Mirko-paroli: to be honest, no. I'd wish all company members were pro-community. I'll think about it. BBL Mar 09 09:17:48 $ opkg install e-wm-utils Mar 09 09:17:48 An error ocurred, return value: -50512. -> what's wrong? Mar 09 09:28:58 <|Leto|> could someone enlighten me upon where to find the icons on shr? Mar 09 09:30:08 |Leto|: http://lists.projects.openmoko.org/pipermail/shr-devel/2009-March/001149.html Mar 09 09:30:42 <|Leto|> hmm Mar 09 09:30:48 <|Leto|> but i'm on shr-testing ^^ Mar 09 09:31:01 <|Leto|> and i'd like to add an icon for a software Mar 09 09:31:24 <|Leto|> in the .desktop files they are just named for example "openmoko-calculator" Mar 09 09:31:44 <|Leto|> find /usr/ -name openmoko-calculator returned the desktop-file only Mar 09 09:40:53 mrmoku`: yes, I have current testing installed. booting into it now Mar 09 09:41:17 is usb-as-mass-storage supposed to work with shr-testing? Mar 09 09:42:03 i am trying to compile something for openmoko and i have problem with my toolchain, because there are missing paths to STL includes Mar 09 09:43:40 also the requirement of older libtool is very annoying.. Mar 09 09:48:45 |Leto|: app icons? in /usr/share/icons/* Mar 09 09:49:00 definitions for the aps in /usr/share/applications Mar 09 09:50:07 mrmoku`: finished calling myself in shr-testing Mar 09 09:50:10 are you here? Mar 09 09:50:18 <|Leto|> hmm Mar 09 09:50:56 <|Leto|> spaetz, could you tell me how to get it displayed with a coloured background? Mar 09 09:51:10 <|Leto|> it sort of happened automagically with my navit icon Mar 09 09:51:23 <|Leto|> but my agps-ui icon just displays transparency Mar 09 09:51:36 no clue, I think the icons just contain the background color Mar 09 09:51:43 but I am not sure Mar 09 09:57:30 Hey there. Would it be possible to include loudmouth in shr? "http://pastebin.com/m4e40a1bc" I tried to compile it using the toolchain, but it failed with stubs-32.h not found error "http://pastebin.com/m45cb8f47" I'm trying to write a jabber client with elementary. The first steps are done and I really would like to start testing it on Freerunner. Mar 09 09:58:55 The bitbake recipe is taken from "http://git.openmoko.org/git/openmoko.git/packages/loudmouth/loudmouth_1.3.2.bb" just a version upgrade. I can't find a opk though,so it doesn't seem to be build Mar 09 10:02:19 if it doesn't build from the OE .bb file, then it won't build in SHR either Mar 09 10:05:31 I didn't try to build from OE Mar 09 10:05:53 http://cgit.openembedded.net/cgit.cgi?url=openembedded/tree/packages/loudmouth/loudmouth_1.3.2.bb&h=fso/milestone5.5 Mar 09 10:09:02 so it should be available in the SHR build environment with a simple bitbake -c build loudmouth Mar 09 10:10:13 so i have to setup the build environment? As on the toolchain wiki it says that for simple development you should not do that? Mar 09 10:10:31 I'll give it a shot then Mar 09 10:10:50 no clue. I have a build environment set up, so I would use that Mar 09 10:11:15 it might be possible to get it without a full fledged env. Never tried that Mar 09 10:12:24 Ok, thanks for your help Mar 09 11:16:44 spaetz: here now Mar 09 11:33:26 task failed: checksum of flex-2.5.31 failed Mar 09 11:33:39 building of shr-unstable... grr Mar 09 11:44:32 where is the package e-wm-menu defined? Mar 09 11:44:41 I can't find a corresponding .bb file? Mar 09 11:44:43 ! Mar 09 11:47:40 spaetz: its in e-wm_svn.bb as subpackage Mar 09 11:48:14 * spaetz checks Mar 09 12:08:56 could it be that files that are marked as config files are not upgraded? Mar 09 12:09:00 CONFFILES_${PN}-menu = "/etc/xdg/menus/applications.menu" Mar 09 12:09:19 (but then, they should not be removed either) Mar 09 12:09:38 well... opkg should ask you (if you modified it) Mar 09 12:10:09 I get that often with frameworkd.conf... it nicely offers you to overwrite, keep or view a diff Mar 09 12:10:19 right. weird. Mar 09 12:10:52 we should just use that new menu-freesmartphone package... Mar 09 12:11:12 it has the advantage to not have to take care about categories Mar 09 12:11:20 mmh, no clue how it is different Mar 09 12:11:40 does different categories mean, that apps will be sorted not only accrding to alphabet? Mar 09 12:11:56 I tried to adapt the file to show phone apps first, but did not succeed Mar 09 12:12:10 as far as I understand it categories decide wether apps are shown in the first place... Mar 09 12:12:25 don't know if we could tweak ordering of the desktop via a menu file Mar 09 12:12:37 raster claimed yes, but I did not succeed Mar 09 12:12:38 might be interesting too... to have all phone apps first for example Mar 09 12:13:00 hmmm... will try that :-) Mar 09 12:13:03 exactly that. I want all my phone apps first, and not look for the "messages" app somewehre in the middle Mar 09 12:13:24 I think the wsod bug still exists, and it just changed color Mar 09 12:13:29 raster claimed it would be possible via "submenus" but I did not manage Mar 09 12:13:40 I now have a black screen with purple stripe screen of death ;] Mar 09 12:13:44 to use them. neither as submenu, nor did it make any other difference Mar 09 12:13:49 I cann it, bswpssod ;] Mar 09 12:13:51 yeah... I read about submenus when researching yesterday... Mar 09 12:13:58 did you add a layout section too? Mar 09 12:14:09 * mrmoku thinks that would be needed for submenus Mar 09 12:14:18 nope. I got stuck understanding the freedesktop specs and then I got distracted Mar 09 12:14:32 * spaetz is easily distracted Mar 09 12:14:48 * mrmoku knows the distraction problem as well :-) Mar 09 12:36:14 * mrmoku thinks raster is wrong and it does not work ;) Mar 09 12:38:13 spaetz: it somehow works when switching to slider mode... Mar 09 12:38:28 I get two sections then... first the phone apps then other stuff Mar 09 12:38:56 though the apps are not one under the other... have to scroll vertically instead :P Mar 09 12:39:55 Raster is never wrong! Mar 09 12:40:35 Dave: he did bribe you... right? Mar 09 12:41:12 :D Mar 09 12:41:18 I admit to nothing! Mar 09 12:42:05 well... and I might be wrong as well :P Mar 09 12:42:48 mrmoku is never wrong! Mar 09 12:43:21 uuhhh. don't say that... I might prove you wrong :D Mar 09 12:43:22 then the only one who can be wrong is Dave... Mar 09 12:43:31 :O Mar 09 12:43:39 well the thing with Dave is.... he is *always* wrong ;) Mar 09 12:43:43 so Dave's the winner Mar 09 12:44:04 * mrmoku goes to lunch before Dave strikes back :P Mar 09 12:44:26 まさか! Mar 09 12:47:29 pfeew, worked around the dysfunctional SF mirror Mar 09 12:48:03 vertical scrolling? innovative :) Mar 09 12:48:29 would be cool it we manage to get the sections below each other. Mar 09 12:48:39 spotting the phone apps sucks currently Mar 09 13:22:21 * bgat is back Mar 09 13:23:14 BAHH, comiling OE sucks when half of the source hosts are not reachable Mar 09 13:57:57 are you experimenting errors in perl bb? Mar 09 13:58:00 for gta02? Mar 09 14:06:24 ptitjes ? Mar 09 14:07:49 Greetings O Wise One! How's Monday so far? Mar 09 14:11:58 hey :-) Mar 09 14:12:01 fairly good Mar 09 14:12:09 spent a lovely week-end with the new girlfriend Mar 09 14:12:17 :-) Mar 09 14:13:36 * Dave rolls his eyes Mar 09 14:18:29 * spaetz holds hands over ears Mar 09 14:26:40 * mrmoku had a nice weekend with his old wife (and children) :P Mar 09 14:27:22 Ainulindale: nice to hear spring starts in france this year :D Mar 09 14:27:58 Ainulindale: hello Mar 09 14:28:20 hey :-) Mar 09 14:28:29 ptitjes: thursday night, restaurant with angus Mar 09 14:28:30 cheers :) Mar 09 14:28:30 remember that Mar 09 14:28:58 I thought it was on friday ? Mar 09 14:29:23 you told me it was on friday (the 13th) !!!!!!!! Mar 09 14:30:07 mwester: has mdev a chance to work on gta02 without tweaking? Mar 09 14:32:54 ptitjes: I was thoroughly mistaken then Mar 09 14:34:36 that's cheat! Mar 09 14:34:51 ok I'll cancel my rendez-vous Mar 09 14:34:53 I'm in Mar 09 14:35:38 does CL3 memory makes a real difference with CL4 ? Mar 09 14:37:56 also guys please advice: transcend, corsair, patriot, kingston, ... Mar 09 14:38:43 Ainulindale: could you give me your advice ? Mar 09 14:40:08 No idea about memory Mar 09 14:42:19 (Though I usually buy Corsair) Mar 09 14:50:54 alphaone: great, I've pulled in the changes and will test as soon as my dist-upgrade is complete Mar 09 14:51:28 I think I also need to file a new report, as for the last couple of days my network has been identified as Unknown rather than T-Mobile, which worked previously Mar 09 14:51:52 Ainulindale: do you know if one can mix CL3 and CL4 ? Mar 09 14:53:08 no idea :-) Mar 09 14:53:11 I'm not a hardware guy Mar 09 14:58:09 hello all Mar 09 14:58:11 mrmoku: no Mar 09 14:58:57 noticed :-) Mar 09 14:59:06 serial devices are wrong... and audio Mar 09 14:59:06 ptitjes: Corsair here as well; no problems (but then I do not overclock either) Mar 09 14:59:41 Does it even load firmware into anything? (but I don't think the gta0x need firmware loaded for the on-board stuff, just for USB-attached devices) Mar 09 15:00:26 well... at least it boots and touchscreen works... better than nothing Mar 09 15:00:34 ptitjes: you can fix CL3 and CL4 on *most* (but not all) motherboards, but they will use the slower timing on all DRAM. You may have to set the timing in the BIOS yourself, though. Mar 09 15:01:47 mrmoku: it should be fairly easy to get mdev working with the onboard stuff in the same way that udev worked. I tried mdev for SlugOS, and got it working -- but I ended up reverting to udev because mdev could not handle the external USB devices as well as udev. Mar 09 15:02:32 mwester: mhhh... good to know. you saved me some work :-) Mar 09 15:02:50 we will have to stick with udev then as well me thinks Mar 09 15:04:45 Another option is to ship with mdev, supporting the base devices and perhaps a few "standard" external usb devices (such as mass storage devices). And let the user "upgrade" to udev if they need more flexibility, but at the cost of boot time. Mar 09 15:05:22 After all, how many people are going to plug something other than usb memory keys, external disks, and maybe a keyboard into the usb port? Mar 09 15:05:50 Debate! Mar 09 15:06:02 * Dave hides from mwester Mar 09 15:06:24 hmmm... true. Having udev installable as an alternative is an alternative :-) Mar 09 15:06:25 There was that time I... Mar 09 15:06:58 will continue to play then... fixing the few devices should be easy Mar 09 15:08:29 about #346 , do I need to tell you some more details? is possible shr does not tell to my operator my sim storage is free again for new messages? Mar 09 15:09:57 a21_work: don't think so... I had full sim and after deleting some messages new messages came in automatically Mar 09 15:11:18 so, I should try again and try to reproduce? Mar 09 15:13:29 a21_work: so you got some SMS but not all? Mar 09 15:13:40 yes Mar 09 15:14:17 so it's difficult to me to know when effectively I've some queue Mar 09 15:15:02 difficult to debug too :( Mar 09 15:15:18 :S Mar 09 15:15:36 though in any way this is more a frameworkd bug than a shr bug if it is a bug :P Mar 09 15:16:27 I only know for sure when I booted qtopia I got them all Mar 09 15:17:11 for me it's a frameworkd bug too Mar 09 15:23:23 johnsu01: Oh, that sounds like a strange problem. Mar 09 15:33:01 mrmoku: about frameworkd bugs, should I report them in a particular place or it's ok as a ticket here in shr? Mar 09 15:33:52 on freesmartphone trac Mar 09 15:34:26 hello Ainulindale , so I should duplicate #346 on that trac too? Mar 09 15:35:47 If it's framework related yes Mar 09 15:36:52 hey, is any of the shr admins as kind as to build a fennec package and add it to the (testing) repo? Mar 09 15:37:13 Last time I checked it wouldn't compile Mar 09 15:37:47 but there is a package in the angstrom repo? but it has so much dependencies : O Mar 09 15:38:14 * mrmoku tries if fennec compiles in unstable... Mar 09 15:38:35 thanks in advance mrmoku Mar 09 15:39:05 root_at: well... I'm not very optimistic either Mar 09 15:40:18 so then i will begin with downloading all the packages and installing now...i am also not optimistc if it will work, but i want to give it a try Mar 09 15:41:22 anybody knows how s3c2410_serialX maps to ttySACY ??? Mar 09 15:43:06 argh...i thought the packages which fennec depends on are just not in the repo, but they are just too old : / Mar 09 15:44:24 mrmoku: phisically port 0 is GSM, port 1 is GPS, port 2 is serial console. So i guess 1-1. Mar 09 15:44:56 PaulFert`: hmmm... I tried 1-1 (with mdev symlinks)... though it does not boot then Mar 09 15:48:11 PaulFert`: well... again mrmoku prooves as being plain stupid.... probably :P Mar 09 15:49:31 * mrmoku forgot the permissions column in mdev.conf :( Mar 09 15:50:34 PaulFert`: boots again... thanks a lot :-) Mar 09 15:50:41 next thing to fix with mdev is sound... Mar 09 15:50:55 mrmoku: You're welcome :) Does it feel to be really faster? Mar 09 15:51:39 will time it on next boot... did not look closely now Mar 09 15:51:47 and I still have to fix sound Mar 09 15:52:07 aplay complains about unable to open timer Mar 09 15:57:21 PaulFert`: what device nodes are involved for sound? I have controlC0 pcmC0D0c pcmC0D0p pcmC0D1c pcmC0D1p in /dev/snd Mar 09 15:57:29 timer is missing... something else missing? Mar 09 15:58:18 mrmoku: speaking of pure alsa, i think no. Mar 09 15:59:14 PaulFert`: what major/minor is timer? Mar 09 15:59:32 mrmoku: 116, 33 Mar 09 15:59:34 is it 116:33 ? Mar 09 15:59:46 mrmoku: yes Mar 09 15:59:59 ahh.. ok. gets created as /dev/timer... though if I symlink it to /dev/snd/timer I get strange sound :-) Mar 09 16:01:09 mrmoku: http://paste.nerv.fi/61284753.txt Mar 09 16:02:05 * mrmoku compares Mar 09 16:02:58 devices are the same.. apart from timer being a symlink... will correct that and reboot Mar 09 16:05:55 PaulFert`: boot until illume shows icons in 1:15m Mar 09 16:06:02 and I still have to disable the e splash Mar 09 16:06:47 mrmoku: i have nothing to compare with :) But sounds great ;) Mar 09 16:07:26 PaulFert`: the fastest I had was 1:18 with udev (last time I did timings) Mar 09 16:07:34 and the e splash costs a lot Mar 09 16:07:57 and a lot means around 10s Mar 09 16:10:49 PaulFert`: and again me was stupid :-) aplay with ogg... sounds working now Mar 09 16:10:50 mrmoku: i don't understand the necessity to boot extra-fast. You're saving like 13 sec with mdev. Are you going to reboot often? ;) Mar 09 16:11:05 mrmoku: aplay with ogg? Cool idea! :))) Mar 09 16:11:15 PaulFert`: well... yes right now I'm rebooting quite often ;) Mar 09 16:11:21 hope that will change some day though Mar 09 16:12:03 my old phone got rebooted about once or twice a year Mar 09 16:14:07 PaulFert`: so in the end you're right... boot time shouldn't matter that much. Everybody wants to boot fast though... Mar 09 16:14:18 and 13s is only, because udev is caching the nodes Mar 09 16:16:51 well, at the current state, I do boot often too, even without any dev work Mar 09 16:16:59 (battery empty etc) Mar 09 16:19:06 hi Mar 09 16:19:08 without e splash its 1:04 Mar 09 16:19:12 dos1: hi Mar 09 16:19:49 mrmoku: what about using menu-freesmartphone? ;x Mar 09 16:20:13 dos1: spaetz and me have the plan to do something even better :P Mar 09 16:20:22 menu-shr? Mar 09 16:20:24 :D Mar 09 16:20:27 yep :D Mar 09 16:20:28 or what? Mar 09 16:20:29 ;] Mar 09 16:20:33 and order the phone apps first Mar 09 16:20:37 does not work though :( Mar 09 16:21:02 just wanted to say, it doesn't work... Mar 09 16:21:58 using menu-freesmartphone sounds good until then Mar 09 16:22:23 yep :-) Mar 09 16:22:59 10 seconds for a boot splash? Mar 09 16:23:11 FUUCKK! Let's remove that darn thing Mar 09 16:23:56 or is it just deferred initialization of stuff that would have to happen anyway before we see the desktop? Mar 09 16:24:22 don't think so... the e splash is suboptimal on gta02 Mar 09 16:25:07 what splash? Mar 09 16:25:07 exquisite? Mar 09 16:25:48 dos1: no... the e startup screen Mar 09 16:26:04 suboptimal is an exxageration then :) Mar 09 16:26:09 mrmoku, enlightenment cant handle this new versions, i am getting a segfault at bootup and recover doesnt help...did compiling fennec already fail? Mar 09 16:26:11 can we kill it for good? Mar 09 16:26:46 ERROR: '/home/mok/src/other/openmoko/shrbuild/shr-unstable/openembedded/packages/mozilla/fennec_hg.bb' failed Mar 09 16:27:31 reason is: /bin/sh: autoconf2.13: command not found Mar 09 16:27:45 there is midori Mar 09 16:27:52 which is ok Mar 09 16:28:10 spaetz, yeah its ok. but i want flash/java/etc Mar 09 16:28:13 spaetz: will take a look now at e config... how to disable that by default Mar 09 16:28:21 ok, thanks. Mar 09 16:28:28 * spaetz turns of splash on his gta02 Mar 09 16:28:30 mrmoku, thanks for the try Mar 09 16:28:37 np Mar 09 16:30:41 flash, java? Mar 09 16:30:43 spaetz: hehe value "show_splash" int: 0; Mar 09 16:33:11 alphaone: My first test message worked, so it seems to be fixed. Mar 09 16:33:19 good Mar 09 16:33:23 alphaone: Thanks! Mar 09 16:34:03 dos1, yeah so i can play these little flash games, watch things on youtube etc and all the things i cant do now with minimo or dillo Mar 09 16:36:27 johnsu01: Great Mar 09 16:43:53 root_at: but... i don't think Fennec is able to do that... Mar 09 16:43:54 wrong Mar 09 16:44:11 he is ; ) Mar 09 16:44:13 in this point, Fennec can do absolutely the same as Minimo i think :P Mar 09 16:44:20 root_at: at FreeRunner? Mar 09 16:44:26 s/at/on/ Mar 09 16:44:38 if i would get him to work, yes Mar 09 16:44:47 i just would have to install the flash plugin Mar 09 16:44:51 root_at, lol, no. Mar 09 16:44:51 root_at: i have working fennec Mar 09 16:44:58 there is no flash plugin Mar 09 16:45:04 at least I haven't heard about one Mar 09 16:45:07 root_at: and i had minimo with gnash - it was loading Mar 09 16:45:12 but it is incredible slow on FR :P Mar 09 16:45:16 really not usable Mar 09 16:45:18 (gnash) Mar 09 16:45:50 for a while I had the YouTube player in Totem working okay on the FR Mar 09 16:46:05 but i read in a community that someone has working fennec on debian and can watch vids on youtube Mar 09 16:46:06 haven't tried it recently though Mar 09 16:46:16 root_at: ah, on debian Mar 09 16:46:24 on debian everything is possible :P Mar 09 16:47:42 any enlightened one in here knowing the gtk+-fastscale story? Mar 09 16:48:23 no, but apperently fastscaling is gtk+ with a patch Mar 09 16:48:32 but it was dropped Mar 09 16:48:41 so there is no gtk+-fastscaling anymore Mar 09 16:48:57 so all those fastscaling packages should just use gtk+ Mar 09 16:49:00 mrmoku: gtk+-fastscaling isn't necesary on new gtk+ Mar 09 16:49:11 it was necessary for some 2007.2 theme to work Mar 09 16:49:16 on old gtk+ Mar 09 16:49:18 yeah... I know. but people are trying to install packages built against it Mar 09 16:49:22 and it's only one, single, little patch Mar 09 16:49:31 and some old packages are builded with it Mar 09 16:49:41 we shouldn't provide gtk+-fastscaling Mar 09 16:49:45 it's only packages problem Mar 09 16:50:03 upstream packaging problem and I would treat it as such Mar 09 16:50:17 if e.g. opkg provides obsolete dependencies it's their problem Mar 09 16:50:18 dos1: have some link to somewhere explaining that? which we could throw it people complaining? Mar 09 16:51:10 mrmoku: don't know Mar 09 16:51:24 mrmoku: i remember it from IRC and analising bb files Mar 09 16:57:48 going home... Mar 09 17:09:45 mrmoku: has the bsod problem been fixed by the way? Mar 09 17:19:29 Ainulindale: windows bsod was fixed. But RNDIS networking still doesn't work. Mar 09 17:21:26 Thanks :-) Mar 09 17:23:03 Ainulindale: the upstream suspects a bug in s3c udc driver, but this issue lacks love from usb hackers. It was propesed to do a low level capture with a hardware tool, but windows users lack that and other devs are not interested. Mar 09 17:25:16 Ainulindale: there's even a trace that looks meaningful to me made with some windows "spy" tool, i send it to the linux-usb ML bzipped, probably that contributed to the absence of reaction. I think some dedicated windows users can push this farther, doing another trace and sending it to linux-usb, so those who understand RNDIS will tell what s3c udc driver is doing wrong. Mar 09 17:27:34 Ok :-) Mar 09 17:27:37 Thanks for the info Mar 09 17:53:04 Ainulindale: you never would have got such an good, profound answer from mrmoku :P Mar 09 18:03:56 thanks a21 Mar 09 18:14:58 my xglamo patch is applying now! yay! :D Mar 09 18:15:04 compiling and testing ;) Mar 09 18:15:20 (do-patch is reporting about whitespace problems, but it works :x) Mar 09 18:15:34 *sigh* ophonekitd on _current_ SHR testing dies with "undefined symbol: frameworkd_handler_new" Mar 09 18:15:55 Wonka: looks like your upgrade wasn't complete Mar 09 18:16:04 Wonka: new image of after upgrading? Mar 09 18:16:05 ;( Mar 09 18:16:14 dos1: "opkg upgrade" only Mar 09 18:16:24 dos1: with preceeding "opkg update" of course Mar 09 18:16:40 Wonka: from old testing? Mar 09 18:16:46 dos1: from older testing, yes Mar 09 18:16:51 is it even supposed to work? ;P Mar 09 18:17:16 if not: why the _fuck_ must there be breaking changes that require reflashing anywhere? Mar 09 18:17:33 Wonka: do it better :P Mar 09 18:17:54 there was really soooooo many changes Mar 09 18:18:31 dos1: then just make packages of "sooooo many" things so that everything changed gets upgraded? Mar 09 18:18:37 dos1: this even works with kernels! Mar 09 18:18:59 Wonka: but exactly that was done! Mar 09 18:19:13 in a wrong way, obviously... Mar 09 18:19:21 Only on systems where one has the concept of "sub-modules" or other means of isolating changes. Mar 09 18:19:27 who's broken? ophonekitd or frameworkd? Mar 09 18:19:41 Wonka: i don't think so - old shr-testing wasn't supported for long time... Mar 09 18:19:43 With systems built by OE, one does not have the concept of a "stable" subsystem. Mar 09 18:19:55 So any change at all might end up rebuilding libc Mar 09 18:20:04 for no really good reason. Mar 09 18:20:12 It's really a major flaw in OE design., Mar 09 18:20:13 IMO Mar 09 18:20:38 i'll try rebooting, maybe something was just not restarted... Mar 09 18:24:37 mrmoku: should i set PR from 3 to 3.1 (and later maybe 3.2 - second patch, which i think i'm going to add later), or 4 with my patch? Mar 09 18:41:06 dos1: patch against xglamo? Mar 09 18:41:13 mrmoku: yep Mar 09 18:41:18 3.1 Mar 09 18:41:21 ok Mar 09 18:41:28 because then if upstream goes 4 we're not borked :-) Mar 09 18:42:10 i thought about the same, but i saw e-wm in fso/milestone5.5 and PR was bumped from 3 to 4 :x Mar 09 18:43:43 dos1: well... they are our upstream :) Mar 09 18:44:00 (until we are ourselves OE citizen) Mar 09 18:44:25 but there is upstream for fso/milestone5.5 :D Mar 09 18:44:53 make index... heh, building bitbake cache is tooo looong :x Mar 09 18:45:42 dos1: well... no fso/milestone5.5 is just a branch in the upstream :P Mar 09 18:45:49 <|Leto|> sone got a recommendation for getting gps to work on shr? Mar 09 18:45:54 heh Mar 09 18:45:55 ok :D Mar 09 18:45:59 <|Leto|> i get a fix with agps-ui just fine Mar 09 18:46:14 <|Leto|> but tangogps as well as navit don't get anything Mar 09 18:46:21 |Leto|: you can try to remove ogpsd.pickle Mar 09 18:46:28 (you can do it from shr-settings) Mar 09 18:46:29 <|Leto|> what's it for? Mar 09 18:46:36 <|Leto|> yes, i've seen that Mar 09 18:46:43 |Leto|: and don't use agps-ui... it conflicts with frameworkd IIRC Mar 09 18:46:45 |Leto|: to remove agps data, which sometimes can be borked Mar 09 18:46:58 <|Leto|> k Mar 09 18:47:09 <|Leto|> mrmoku, is there a similar app for use with frameworkd? Mar 09 18:47:24 |Leto|: zhone Mar 09 18:47:46 <|Leto|> zhone? i thought that was the principal name for the fso-ui? Mar 09 18:48:08 |Leto|: yes it is, but it has really nice GPS ui Mar 09 18:48:15 (for testing) Mar 09 18:48:24 <|Leto|> i can install it along with shr? Mar 09 18:48:29 <|Leto|> without conflict? Mar 09 18:48:49 <|Leto|> within shr, i mean Mar 09 18:48:58 |Leto|: yes, you'll only have to remove /etc/X11/Xsession.d/80zhone file Mar 09 18:49:06 mrmoku: i don't like menu-freesmartphone - openmoko-keyboard and matchbox-keyboard is listed now i illume :x Mar 09 18:49:07 that's the nice thing with the frameworkd architecture Mar 09 18:49:22 <|Leto|> -.- Mar 09 18:49:23 mrmoku: so menu-shr will be nice :D Mar 09 18:49:29 <|Leto|> what's 80zhone file for? Mar 09 18:49:36 dos1: hehe... yeah. it includes anything that smells like a desktop file ;) Mar 09 18:49:37 <|Leto|> do i need to copy it back when i'm done? Mar 09 18:49:37 |Leto|: for autostarting Zhone Mar 09 18:49:42 <|Leto|> ahhh, k Mar 09 18:49:46 <|Leto|> thanks Mar 09 18:49:54 <|Leto|> i'll get back later :-] Mar 09 18:49:57 |Leto|: you don't want to start it everytime in SHR... Mar 09 18:50:02 <|Leto|> :-D Mar 09 18:50:06 <|Leto|> thought so Mar 09 18:51:17 dos1: yeah... btw. I think the problem with e-wm-menu was (is) a opkg problem... Mar 09 18:51:32 the image works nice... just the upgrade not Mar 09 18:51:51 mrmoku: have you unpacked e-wm-menu package? Mar 09 18:52:26 dos1: after the upgrade I did opkg files e-wm-menu Mar 09 18:52:31 and that was empty... Mar 09 18:52:51 the installed-packages.txt from the testlab clearly says the menu file is there... Mar 09 18:53:12 and in between I installed my homebuilt image to SD and that had the menu file Mar 09 18:53:20 image is ok... just the upgrade borked Mar 09 19:09:03 composite is working :) Mar 09 19:09:18 dos1: compiz next? Mar 09 19:09:33 :D Mar 09 19:10:16 toggles_w: composite is usefull to have almost full-screen transparent keyboard and with redrawing screen on for instance illume animations Mar 09 19:10:47 mrmoku: can you review and add bb files from that site: http://kedrigern.dcs.fmph.uniba.sk/~riso/Openmoko/qwo/ ? Mar 09 19:11:12 qwo and transset Mar 09 19:12:06 dos1: awww ahhhh, puuurrrty Mar 09 19:12:25 * mrmoku wants the rain effect :D Mar 09 19:12:31 :DDD Mar 09 19:13:23 will try to build qwo now Mar 09 19:13:25 mrmoku: come here, go outside. lots of rain-effect ;D Mar 09 19:13:33 * toggles_w can't wait for the spinny cube Mar 09 19:15:52 DocScrutinizer: hehe.... where is here? Mar 09 19:16:11 south-germany Mar 09 19:16:37 had that rain effect too then... even some small 'Hagel'-effect today :P Mar 09 19:16:54 yup Mar 09 19:18:07 * DocScrutinizer thought someone lttered with styropor ;D Mar 09 19:19:34 :-) yep... was just like small styropor balls Mar 09 19:20:31 transparency works :) Mar 09 19:20:45 but it's slow... Mar 09 19:21:35 dos1: what did you expect? Mar 09 19:21:48 DocScrutinizer: no, i wasn't expecting anything Mar 09 19:21:57 err Mar 09 19:22:00 ;) Mar 09 19:22:05 i expected it would be slow :D Mar 09 19:22:23 but not as slow as it would be Mar 09 19:22:25 ;> Mar 09 19:22:53 yup, 90% transparency isn't 90% less work :-> Mar 09 19:22:54 semi-transparent keyboard should be usable :) Mar 09 19:22:58 :DD Mar 09 19:24:12 :o Mar 09 19:41:10 ok, i scapped few useless and few nice usages of composite transparency :D Mar 09 19:43:33 THANKS DOS Mar 09 19:43:38 It's clearly useless and annoying! Mar 09 19:43:48 Dave: yep! i know! :D Mar 09 19:44:06 :P Mar 09 19:44:09 but now there will be many messages on maillist about that :D Mar 09 19:44:23 oh great -_- Mar 09 19:45:26 Dave: but think about it from another side Mar 09 19:46:31 if i wouldn't send it on scap, there won't be top-topic on list about transparency - so maybe another top-topic would be something more annoying ;) Mar 09 19:46:50 hmm... Mar 09 19:47:01 * dos1 needs he needs sleep, cause it's saying non-sense Mar 09 19:47:09 s/it/he/ Mar 09 19:47:11 oh... Mar 09 19:47:13 :/ Mar 09 19:48:26 * dos1 hopes that someone will use composite in more useless way Mar 09 19:48:40 as that qwo transculent keyboard :) Mar 09 19:51:18 dos1: still need to solve point 6 to have that... Mar 09 19:51:26 yep Mar 09 19:51:47 mrmoku: but building qwo works? Mar 09 19:51:51 no :( Mar 09 19:52:00 :( Mar 09 19:52:02 still working on it... Mar 09 19:52:25 ahh... ok. no it built. The one from git does not build... Mar 09 19:52:49 s/no/now/ Mar 09 19:52:59 dos1: night! ;-) Mar 09 19:52:59 damn bot... damn fingers :P Mar 09 19:53:21 ohhh Mar 09 19:53:28 ***dos1 hopes that someone will use composite in more useless way Mar 09 19:53:37 nonsense nonsense nonsense nonsense Mar 09 19:53:42 s/useless/usefull/ :DDD Mar 09 19:53:48 hihihihihihihi Mar 09 19:54:10 dos1: btw.... you as python freak... couldn't you just steal the gps part from zhone and make it a standalone app? Mar 09 19:54:22 ;) Mar 09 19:54:29 mrmoku: i will integrate it with shr-settings ;) Mar 09 19:54:40 that's why i separated GPS category Mar 09 19:54:41 better still :D Mar 09 19:56:20 dos1: how do I configure it to use qwo as keyboard? Mar 09 19:56:42 you need desktop file with "Keyboard" in categories Mar 09 19:56:51 ahh Mar 09 19:56:54 and then Illume wrench -> Keyboard -> Qwo Mar 09 19:57:34 hmm somehow rereading the desktop files does not work always... same reason why the cpu trick does not always work Mar 09 19:57:39 * mrmoku restarts x Mar 09 19:57:41 (but it looks like menu-freesmartphone is broken for doing that :( ) Mar 09 19:57:47 qwo looks interesting Mar 09 19:58:18 dos1: how fast is it? Mar 09 19:59:01 hmm... no... no qwo Mar 09 20:00:06 root@om-gta02 ~ $ DISPLAY=:0 qwo Mar 09 20:00:06 /usr/share/qwo/normal.png : Can't open file Mar 09 20:00:06 /usr/share/qwo/caps.png : Can't open file Mar 09 20:00:06 /usr/share/qwo/extra.png : Can't open file Mar 09 20:00:16 those files are there though... Mar 09 20:02:30 <|Leto|> sheesh Mar 09 20:02:41 <|Leto|> mrmoku, dos1: how can i setup my own feeds? Mar 09 20:03:14 <|Leto|> e.g. i guess i simply need to create packages.gz Mar 09 20:03:18 <|Leto|> how do i do that? Mar 09 20:03:53 |Leto|: bitbake package-index Mar 09 20:03:59 <|Leto|> nice Mar 09 20:04:01 <|Leto|> thanky Mar 09 20:04:22 <|Leto|> and that is sufficient? along with the packages to be dumped on a http-server? Mar 09 20:04:41 holy touchphones batman Mar 09 20:07:14 |Leto|: yep... dump tmp/deploy/glibc/ipk on your server and you have your feed Mar 09 20:07:27 <|Leto|> :) Mar 09 20:07:31 * |Leto| bows Mar 09 20:07:38 <|Leto|> arigaot gozaimashita Mar 09 20:07:44 <|Leto|> *arigato Mar 09 20:10:21 :o Mar 09 20:10:39 TAsn: thanks for answering that mail :P Mar 09 20:10:54 mrmoku, ;] Mar 09 20:11:12 I suppose you agree with me ;] Mar 09 20:11:18 1000% Mar 09 20:11:29 though I did not find your words :-) Mar 09 20:11:50 you did not find my words? what do you mean? Mar 09 20:12:28 well... I wanted to answer that mail too... did not know how to write it without getting unpolite though Mar 09 20:12:35 and in the end did not answer :P Mar 09 20:12:36 mrmoku, ;] Mar 09 20:12:44 I hope I was polite enough. Mar 09 20:12:47 ;\ Mar 09 20:12:48 yep Mar 09 20:13:17 mrmoku: next time send me a link to the mails where impolite answer is desired ;) Mar 09 20:13:18 cool. Mar 09 20:13:47 PaulFert`: :D will do ;) Mar 09 20:16:52 * Sharwin_F- has been compiling frozen-bubble for about 4 hours, hope it finishes before going to sleep... Mar 09 20:17:04 Sharwin_F-, lol. ;] Mar 09 20:17:11 isn't there a pkg in the angstrom feeds? Mar 09 20:17:27 TAsn, hope not, I'll kill myself then Mar 09 20:17:28 xD Mar 09 20:19:06 wel, anyway, I think it takes so long because it's the first time I compile something with bitbake Mar 09 20:19:14 Sharwin_F-, oh Mar 09 20:19:16 next ones will be shorter I hope Mar 09 20:19:16 go to bed Mar 09 20:19:21 hahah Mar 09 20:19:25 4 hrs means you have a lot of time left Mar 09 20:19:35 actually, I think I'll be saying the next thing tomorrow night ;] Mar 09 20:19:37 (j/k) Mar 09 20:19:38 the only thing I'm afraid of... is to have free disk space for it... Mar 09 20:19:42 *next = same Mar 09 20:21:18 I've 2,4 Gb free at the moment Mar 09 20:22:11 having so little free space is not good in general Mar 09 20:22:58 TAsn, yeah, I'll try to redo the partitions when I've time Mar 09 20:23:12 you should. Mar 09 20:34:07 Sharwin_F-: shr build will take over 4Gb just for tmp ;) Mar 09 20:38:29 mwester: thanks! I finally bougth CL4 as I had already :) Mar 09 20:39:08 mwester: to discover in my BIOS settings they were managed in CL5, so I fixed it and it runs so fast now :D Mar 09 20:39:16 mwester: very much thanks Mar 09 20:44:23 mrmoku, can I change tmp dir to point to another partition? Mar 09 20:46:27 ah yes, I think in bitbake.conf Mar 09 20:46:40 TMPDIR = "${TOPDIR}/tmp" Mar 09 21:01:04 Sharwin_F-: yep Mar 09 21:11:02 <|Leto|> mrmoku, i'm sorry to bother you again Mar 09 21:11:13 |Leto|: yep? Mar 09 21:11:13 <|Leto|> bitbake package-index only gives me a few bytes of crap Mar 09 21:11:35 <|Leto|> uhm in the package.gz that is Mar 09 21:12:28 <|Leto|> tmp/deploy/glibc/ipk/Packages Mar 09 21:12:32 <|Leto|> that is empty :-X Mar 09 21:13:41 |Leto|: take a look at the one in ipk/all/ for example Mar 09 21:13:47 just the top level one is empty Mar 09 21:17:11 <|Leto|> great Mar 09 21:17:13 <|Leto|> thanks Mar 09 21:17:16 <|Leto|> that worked! Mar 09 21:41:12 hey, i want to port archlinux onto the freerunner...it thought it would be a good idea to build a live-sdhc-card image and run the install setup from there...possible? Mar 09 21:42:09 live sd image? you can just run it from sd ;], but anyhow just copying everything to flash isn't hard. Mar 09 21:42:10 night. Mar 09 21:43:12 but i cant "mount" the freerunner like a hdd and write on it, so i thought it would be easier with the live image Mar 09 21:43:40 or can i? Mar 09 21:46:14 root_at: going uSD way is definetely easier. Mar 09 21:46:46 root_at: but to write to the internal flash you can boot from NOR u-boot and use DFU (with dfu-util) to write your rootsd image to flash. Mar 09 21:47:18 PaulFert`, thanks for confirming my ideas : ) Mar 09 21:47:29 root_at: you can literally mount the booted freerunner with sshfs too Mar 09 21:47:55 i hope to make laaater a jffs2 image,like all the other images...but this isnt in my head right now Mar 09 21:47:57 s/the/a Mar 09 21:48:34 but i like the idea with the image more...it seems to be more work : D Mar 09 21:48:34 root_at: what's the goal? Why not Debian or Gentoo? Really, packaging e17 for Debian (even in a q&d way) would be cool ;) Mar 09 21:49:03 PaulFert`: e17 seems to change fast Mar 09 21:49:12 PaulFert`: at least that's the impression i got Mar 09 21:49:12 lindi-: yes, and Debian folks don't like that. Mar 09 21:49:19 i really dont like debian, and i cant handle gentoo, i have never used it, but i really love archlinux, and its so flexible and easy Mar 09 21:49:30 PaulFert`: yeah a lot of work to figure out what the interdependencies of each part are Mar 09 21:49:48 root_at: then use arch? Mar 09 21:49:58 lindi-: otoh it doesn't have any effect on the other system. I see no reason why not to produce regular packages of all e17 parts from svn. Really. Mar 09 21:50:12 PaulFert`: pkg-e is doing this Mar 09 21:50:33 lindi-, yeah i want to^^ but at first i need to port arch on the freerunner, there is no archlinux for arm atm Mar 09 21:50:56 root_at: is it x86/amd64-only? Mar 09 21:52:09 lindi-, formal yes(i686 and x86_64), but there is also an ppc version, its not official but i think its already usable Mar 09 21:52:22 root_at: ok then it's a lot of work to port it to armel Mar 09 21:52:57 root_at: hundreds of packages in debian have patches that make them work on armel. upstream often does not want them Mar 09 21:53:14 lindi-: sorry for ignorance, but where all that e packages are? And how often are they upgraded, the last time i tried to use e17 from pkg-fso repo, it wasn't usable due to issues known for like a month. Mar 09 21:53:16 lindi-, sure, all these packages...but at first i just will port the basic things(what a wonder) and then i will hope to find people which help me to build a repo or something...or i will use it just for myself Mar 09 21:53:45 would be sad, but i really want to use it, even if i am alone Mar 09 21:54:08 who* Mar 09 21:55:04 PaulFert`: 'I have never touched the full E17 packaging stuff, but it seems that because of constant API breakages in E libs, most of the time a new update means new names for the library packages, thus that these packages need to go through the NEW Debian queue again. This is not only time consuming for both the Debian maintainer(s) and the Debian FTPmasters, but also annoying and prone to errors.' Mar 09 21:56:15 PaulFert`: e17-data seems to be of version 0.16.999.050+svn20081207-2~fr.1 Mar 09 21:56:35 lindi-: read that. I see no reasons why can't someone setup an automated system to package some 2-old days version of all the components from svn every day or two. Without all official debian stuff. Just to make it useful for the time being. Mar 09 21:57:02 PaulFert`: no, it's 0.16.999.050+svn20090203-1 in experimental Mar 09 21:57:20 PaulFert`: but those would not be debian packages Mar 09 21:57:41 PaulFert`: they might be using the same package format but they wouldn't meet the debian policy Mar 09 21:57:43 lindi-: i think that'd be ok for the most Debian-on-freerunner users. They will mostly work. Mar 09 21:57:59 lindi-: yes, it wouldn't be debian packages, i agree. Mar 09 21:58:21 PaulFert`: i have spent enough time fighting with totally broken packages in the past ;) Mar 09 21:59:21 lindi-: what if these packages were a self-contained set you don't care about that much? Still it'll mostly work. And that will widen SHR userbase btw, which i suppose would be a good thing. Mar 09 21:59:56 PaulFert`: if i were to use them then i'd surely care about it? Mar 09 22:00:55 PaulFert`: and if they are critical phone infrastructure then surely they are not something i'd "not care that much" Mar 09 22:01:02 lindi-: why? I use freerunner but i don't care much about whether it works today for sending sms's or not. The same with e17. Let's suppose it was broken by the last update, well, no problem, start kapula :) Mar 09 22:02:21 PaulFert`: i want it to work :) Mar 09 22:02:41 PaulFert`: how do i start kapula if my window manager does not work? Mar 09 22:03:12 PaulFert`: and how do i know that it does not work? often the phone can just stop receiving calls and appear normal. like http://trac.freesmartphone.org/ticket/373 Mar 09 22:03:16 lindi-: but you'll lose nothing. Right now no debian users have fully working and modern e17 working. If those q&d packages existed, most of the time it could work. Mar 09 22:03:51 PaulFert`: i'd get the real e17 packages slower? Mar 09 22:03:59 lindi-: btw, seen you bugreport, there were a similar issue for me. Mickey promised he'll reset parser state on timeout. Mar 09 22:04:03 PaulFert`: since time was spent on making some automatically built ones? Mar 09 22:04:37 PaulFert`: but anyways, feel free to start autobuilding e17 packages, just make sure people don't think they are official or we'll be buried in odd bug reports :P Mar 09 22:05:15 lindi-: you can start another window manager. As to the making real e17 packages slower -- not the case. e17 will not stabilize enough for the Debian folks to like it for at least half a year. Mar 09 22:05:28 PaulFert`: but only if i know that there is something wrong going on Mar 09 22:06:38 PaulFert`: i'll rather just use a rock-solid window manager that has been stable for years than get pretty graphics :) Mar 09 22:06:43 lindi-: keep the packages from the last update and prepare a script to automatically revert. Shouldn't be that hard. And, after all, using a dev phone, it's reasonable to be tolerant to some "minor" inconveniences. Mar 09 22:07:28 PaulFert`: for things like "mplayer" i would see that auto-built packages of "buggy" source code would be ok Mar 09 22:07:36 lindi-: i'd like to try SHR, that's why i'd like to see modern e17 libs packaged, not because of the fancy WM and keyboard (which seem to be much better than matchbox though). Mar 09 22:07:51 PaulFert`: but for critical phone stuff that is running all the time i want proper packages of proper software ;) Mar 09 22:08:24 lindi-: not all users need that level of stability, you know :) Mar 09 22:08:42 PaulFert`: loosing a call thanks to pretty graphics feels just so stupid :) Mar 09 22:09:09 PaulFert`: dunno, all people i've talked to have complained that they can't switch to freerunner as their main phone due to stability issues Mar 09 22:09:56 PaulFert`: but yes, don't get me wrong. autobuild packages sometimes have their place Mar 09 22:10:04 lindi-: i'm using it as the only phone since november... And you know the stuff i use: andy-tracking kernel, Qi, FSO from git, Zhone from git. Sometimes it doesn't work, so, well, i don't care much :) Mar 09 22:10:10 PaulFert`: for example I used to auto-build packages for elinks and mplayer when i followed their development Mar 09 22:10:35 PaulFert`: i care :) Mar 09 22:11:13 i am out, cu Mar 09 22:11:37 but now good-enough elinks and mplayer are in debian unstable so i don't need to follow those projects in my own repo anymore Mar 09 22:11:56 lindi-: if i cared, i'd get somewhere on old nokia and had it in my bag. Mar 09 22:12:06 PaulFert`: but i care about stable free phone :) Mar 09 22:12:18 PaulFert`: free first, stable then. pretty graphics last :) Mar 09 22:14:22 lindi-: i'd fully agree if only SHR wasn't based on the-latest-and-greatest e17 libs and if there were some easy to use keyboard (as matchbox is a bit tough to use on such a small screen, especially after cutting the nails). Mar 09 22:15:03 PaulFert`: if only illume keyboard was portable to other wms :) Mar 09 22:15:14 PaulFert`: i use xvkbd as you might remember Mar 09 22:15:46 lindi-: i do :) Do you use it without stylus? Mar 09 22:16:15 PaulFert`: nope, fingernails Mar 09 22:16:42 with Xorg it is even more usable since X does not spend so much time redrawing the layout Mar 09 22:17:13 lindi-: do you shape your fingernail somehow? Mar 09 22:17:34 not really Mar 09 22:17:50 PaulFert`: as long as the layout has only few keys it is quite easy to use Mar 09 22:18:28 hmm, i hit http://trac.freesmartphone.org/ticket/373 again Mar 09 22:18:32 lindi-: yes, i'm just too lazy to do a layout i can comfortably use. Because i don't care :) Mar 09 22:19:03 lindi-: btw, beware, looks like Mickey won't accept gsm0710muxd bugs anymore, he prefers fso-abyss now. Mar 09 22:19:21 PaulFert`: does fso-abyss compile with debian stable? Mar 09 22:20:03 i got the impression that it needs patches to the valac compiler Mar 09 22:20:27 lindi-: i haven't tried it myself yet... Mar 09 22:21:27 it looks like 10 seconds just is not enough time Mar 09 22:22:34 ogsmd -> dbus -> gsm0710muxd -> calypso -> gsm0710muxd -> dbus -> ogsmd can take a while if lots of tasks have just woken up from suspend Mar 09 22:24:09 lindi-: AT+CTZU=0 is used during suspend time btw. So, do you think that because of slow resume the initial characters (%CPI etc) in the uart buffer get overwritten? Mar 09 22:25:32 PaulFert`: that should throw a buffer overrun anyway Mar 09 22:26:08 PaulFert`: not sure Mar 09 22:26:21 DocScrutinizer: i know :) Mar 09 22:26:25 PaulFert`: i only see that ogsmd is not waiting long enough for AT+CTZU=0 response Mar 09 22:26:47 PaulFert`: and it shouldn't happen on MOKO11 ;-) Mar 09 22:26:57 DocScrutinizer: i have moko11 Mar 09 22:27:23 DocScrutinizer: how large is the calypso's buffer? Mar 09 22:27:24 lindi-: that's a different story though Mar 09 22:27:37 PaulFert`: nobody knows Mar 09 22:27:59 even Dieter didn't answer questions regarding this topic Mar 09 22:28:02 IIRD Mar 09 22:28:10 s/D/C/ Mar 09 22:28:11 :) Mar 09 22:29:42 PaulFert`: quite some heat on ML those tmes, no? ;D Mar 09 22:31:47 DocScrutinizer: are you talking about Trisoft reply about buzz? Mar 09 22:32:00 PaulFert`: hmm, is it possible that hwclock is causing 4 second jump in time and that causes me to hit the timeout? Mar 09 22:32:11 and someone published a IRC quote XD Mar 09 22:32:27 yeah, action! Mar 09 22:32:34 lol Mar 09 22:32:49 DocScrutinizer: nobody commented on my IRC talk with Mirko though. Looks like nobody's interested anymore. Mar 09 22:32:54 sudo should log to syslog before running the command in question, right? Mar 09 22:33:08 OM actually verifying A&-buzz-rework ATM ;-) Mar 09 22:33:10 so 'Apr 9 17:01:50 ginger authpriv.notice sudo: lindi : TTY=console ; PWD=/home/lindi ; USER=root ; COMMAND=/sbin/hwclock --hctosys' Mar 09 22:33:18 should mean that hwclock was run at :50 Mar 09 22:33:20 lindi-: (timeout) still it shouldn't prevent %CPI from coming in... Mar 09 22:33:56 PaulFert`: maybe some command is disabling that and because of timeout it is not enabled again? Mar 09 22:34:09 who killed bzzbot Mar 09 22:34:11 ??? Mar 09 22:34:13 * lindi- would like to read the calypso source code to understand when %CPI is sent... Mar 09 22:34:44 lindi-: alas those are under heavy NDA Mar 09 22:35:28 but with severe googling you might find some equvalent sources on some rushian warez server Mar 09 22:35:52 or sth like that Mar 09 22:36:41 lindi-: feel free to ask for the help with getting through russian sites ;) Mar 09 22:37:14 yes, syslog logs the timestamp before running the command as expected Mar 09 22:38:34 DocScrutinizer: I liked Trisoft reply much. Mar 09 22:38:46 me too ;))) Mar 09 22:40:54 I remember the times when Steve was simply "steve" and how he got bashed for something :) Mar 09 22:41:16 PaulFert`: are you sure there is no command to disable %CPI that ogsmd might be sending? Mar 09 22:42:41 lindi-: it's AT%CPI=mode. But i don't think ogsmd issues it to disable %CPI ever. Mar 09 22:43:32 I remember the times standing side by side to him at OM apartment balcony in the night, smoking and talking bout buzzfix and OM Mar 09 22:44:28 PaulFert`: c.append( "%CPI=3" ) # call progress indication: enable with call number ID, GSM Cause, and ALS Mar 09 22:44:51 PaulFert`: i should try sending that next time i hit the problem Mar 09 22:45:16 lindi-: sounds sensible Mar 09 22:46:00 lindi-: i haven't taken a look at your debug log yet, can you confirm you didn't see %CPI there too? Mar 09 22:46:03 or even better, ask for current value Mar 09 22:46:25 %CPI? Mar 09 22:48:19 PaulFert`: no CPI after 1900 Mar 09 22:49:07 lindi-: so that looks like it wasn't generated at all or was lost in some buffer... Mar 09 22:49:14 Mar 10 00:48:56 ginger user.debug ogsmd.channel DEBUG : got 17 bytes from: '\r\n%CPI: 0\r\n\r\nOK\r\n' Mar 09 22:49:18 Mar 10 00:48:56 ginger user.warn ogsmd.modems.abstract.channel WARNING UNHANDLED INTERMEDIATE: %CPI: 0 Mar 09 22:49:21 Mar 10 00:48:56 ginger user.warn ogsmd.modems.abstract.channel WARNING UNHANDLED INTERMEDIATE: OK Mar 09 22:49:28 DocScrutinizer: that is bit difficult, since %CPI is always interpreted as unsolicited Mar 09 22:49:40 DocScrutinizer: so you can't send "AT%CPI?" command and read its reply Mar 09 22:49:59 duh??? Mar 09 22:50:07 Yup Mar 09 22:50:10 DocScrutinizer: it will always timeout and interpret it as unhandled intermediate command Mar 09 22:50:37 PaulFert`: is it normal for %CPI? to return zero? ;) Mar 09 22:50:52 PaulFert`: probably i sent it on wrong channel Mar 09 22:51:20 lindi-: yes, for me it does the same on the 4th channel. Mar 09 22:51:32 PaulFert`: any idea how to send it on callchannel? Mar 09 22:52:17 lindi-: no, not that familiar with the codebase. And no python debugger integrated yet. Mar 09 22:53:58 DebugInjectString Mar 09 22:57:50 blah no, that adds input Mar 09 22:58:46 PaulFert`: i don't think it is supported Mar 09 22:59:25 lindi-: it'd be possible with integrated python debugger though... Mar 09 23:00:07 yep Mar 09 23:41:11 hello i want to reflash my phone to the new shr-testing, because i had old-testing and made an opkg update and opkg upgrade na dnothing works, as expected Mar 09 23:41:25 what files do i have to backup to keep my messages and my phonelog and so on? Mar 09 23:45:03 are all messages on the simcard? Mar 10 02:43:45 lorns Mar 10 02:47:45 hello **** ENDING LOGGING AT Tue Mar 10 02:59:57 2009