**** BEGIN LOGGING AT Wed Feb 03 02:59:56 2010 Feb 03 07:38:13 Heinervdm: morning Feb 03 07:38:20 moin Feb 03 07:38:37 Heinervdm: maybe we should offer Lukaz to host the ffalarms repo? Feb 03 07:38:51 (if he would be fine with switching to git ;) Feb 03 07:39:27 I can ask him, if he is fine with that. Feb 03 07:40:04 ok Feb 03 07:45:40 * Weiss considers labelling his development FR in some way, so that he doesn't end up accidentally (a) taking the wrong one out with him, or (b) flashing 2.6.32+drm+lots of bugs+megatron to it Feb 03 07:45:47 *to the good one Feb 03 07:48:33 :P Feb 03 07:50:33 hehe Feb 03 07:57:19 freesmartphone.org: 03morphis 07msmcomm * rf6821c9f5543 10/ (9 files in 2 dirs): libmsmcomm: add commands related to charging Feb 03 07:57:20 freesmartphone.org: 03morphis 07msmcomm * r97ae4783e4cb 10/msmcterm/src/msmcterm.c: msmcterm: disable binding to specific device Feb 03 07:57:23 freesmartphone.org: 03morphis 07msmcomm * rac0aacb729ed 10/msmcterm/src/msmcterm.c: msmcterm: react on some more sim events Feb 03 07:57:23 freesmartphone.org: 03morphis 07msmcomm * rfa70a815a21a 10/libmsmcomm/src/ (message_call.c message_system.c msmcomm.h): libmsmcomm: cleanup a little bit Feb 03 07:57:24 freesmartphone.org: 03morphis 07msmcomm * rd00bb408a5ee 10/msmcommd/src/ (frame.c llc.c): msmcommd: fix buggy decoding of incomming frames Feb 03 07:57:47 shr-u: no more gsm registration since yesterday "opkg upgrade". settings->phone->PIN-Settings: Coudn't connect to FSO. known problem? how to start debugging here? Feb 03 08:04:28 topro: I was able to do test call before pushing, but I can confirm that it's somehow less reliable now :/ Feb 03 08:05:30 JaMa|Gone: it's somehow less reliable <-- cool description of whats going on on my neo ;) Feb 03 08:05:52 fso-boot, sounds nice... Feb 03 09:44:25 mickey|zzZzzz, morning Feb 03 09:58:14 now this is a fun bug Feb 03 09:58:42 e crashses if you leave a freerunner running shr plugged into the wall charger Feb 03 09:59:56 file_md5sum_alloc: Failed to open file /etc/freesmartphone/fsodeviced.conf: No such file or directory Feb 03 09:59:56 not deleting modified conffile /etc/freesmartphone/fsodeviced.conf Feb 03 10:07:31 trs80: how about using bit more stable software? ;) Feb 03 10:08:05 lindi-: well, I was using 2007.2 before, is that stable enough? :P Feb 03 10:08:57 trs80: no, choose icewm for stability Feb 03 10:09:21 lindi-: just opkg install icewm? Feb 03 10:10:40 trs80: unfortunately probably not in shr. sorry for bothering you, i use debian Feb 03 10:11:06 well, I'm happy to use whatever works Feb 03 10:12:36 trs80: then do :) Feb 03 10:12:55 lindi-: (more explicitly) debian works ok as a regular phone? Feb 03 10:14:01 last time I used it was about a year ago, and it wasn't overly functional Feb 03 10:14:38 trs80: what bugs did you report? Feb 03 10:16:31 lindi-: none, I just went back to 2007.2. it was mostly the lack of features in the zhone Feb 03 10:25:50 question: maybe you have the same behaviour as mine: Feb 03 10:25:50 GSM does not work after some time of usage (few hours to few days) I think that /etc/init.d/phonefsod also stops. restarting the phonefsod make the GSM work again. Is this a know issue. I am having the lastest firmware (*11*) and the SHR image from January 15 Feb 03 10:27:20 Does your GSM work reliably? Feb 03 10:27:20 Do you have a daemon monitor to automatically restart phonefsod? In that case how do you deal with entering the PIN ? Feb 03 10:31:01 freesmartphone.org: 03morphis 07msmcomm * r17bf4115c012 10/ (5 files in 2 dirs): libmsmcomm: some fixes Feb 03 10:31:51 my revision :: 'revision': '"GSM: gsm_ac_gp_fd_pu_em_cph_ds_vc_cal35_ri_36_amd8_ts0-Moko11"' Feb 03 10:49:10 JaMa|Gone: ^^^^^^ Feb 03 10:55:38 octavsly: me? Feb 03 11:04:23 Q-Master: what? Feb 03 11:04:48 Q-Master: you can remove that one.. Feb 03 11:05:10 Q-Master: right automatically used is in /etc/freesmartphone/conf/GTA02/fsodeviced.conf Feb 03 11:05:55 Q-Master: more correct: you *should* remove it.. Feb 03 11:07:24 lindi-:had a general question about the GSM reliability. I wonder of other people have this "disconnect from network" problem Feb 03 11:10:03 octavsly: which bug report? Feb 03 11:11:39 lindi-: this is a general remark. The GSM in not reliable. After few hours/few days the GUI still displays as being part of the GSM network but in fact you cannot dial nor receive calls. restartting phonefsod solves the problem . I was just wondering if this is a know problem. Feb 03 11:12:52 octavsly: that report is too vague. it's difficult to say if it is really a gsm bug or if it is a frameworkd bug or UI bug Feb 03 11:13:20 octavsly: look what the GSM chip is sending/receiving Feb 03 11:13:22 lindi-: agree. but from your answer I realize you do not have such a problem Feb 03 11:13:27 lindi-: am I right? Feb 03 11:13:30 octavsly: i have reported my bugs :) Feb 03 11:13:44 octavsly: http://lindi.iki.fi/lindi/cgi-bin/openmoko-bugs.py?q=gsm%0D%0Alindi Feb 03 11:14:00 lindi-: I was looking from very high level.. thanks for th elink I will ook at it Feb 03 11:16:33 JaMa|Gone: crap.... Do I need to reconfigure everything? Feb 03 11:16:34 octavsly: bugs need to be analyzed and then chased down :) Feb 03 11:17:33 lindi-: fully agree... just the time and knowledge is sometime against me... Feb 03 11:46:46 Q-Master: check diff with /etc/freesmartphone/conf/GTA02/fsodeviced.conf and if all changes are intentional then use your /etc/freesmartphone/fsodeviced.conf.. Feb 03 11:50:22 this probably is NOT a GSM bug Feb 03 11:54:21 octavsly: also whenever you restart the modem, you need to enter your pin again - at least that's how SHR/FSO is handling that right now Feb 03 11:56:10 DocScrutinizer: yes I have seen that. for monitoring the phonefsod daemon I might need to remove the pin Feb 03 11:56:16 octavsly: modem hangs can be spotted easily by calling *in* to the phone. If you get ringback signaling then the modem is still alive OTA. If modem has locked up then you get "temporarily not available" Feb 03 11:57:20 DocScrutinizer: I usually try to call 123. If I am not registered the dialer does show-up but the call dialog doesn't Feb 03 11:57:41 if I am registered I get an voice error message Feb 03 11:57:46 mickey|sofa: shouldn't frameworkd / fsogsmd / dunnowhat know how to reconnect without a full modem reset? Feb 03 11:58:32 octavsly: i recommended INBOUND call to check *modem* Feb 03 11:58:46 DocScrutinizer: did that too Feb 03 11:59:13 DocScrutinizer: entered directly in the voicemail. Feb 03 11:59:49 duh, then the modem wasn't registered to network anymore Feb 03 12:00:58 DocScrutinizer: in theory yes, in practice it's not feasible Feb 03 12:01:07 DocScrutinizer: in case the GSM network is lost how is this handeled in the software? Feb 03 12:01:20 we can't relaunch the MUX mode without powercycling Feb 03 12:01:21 either T3212 timer for reregistration expired (this can happen any time, though the timer usualy is several hours), or modem actively deregistered Feb 03 12:01:28 probably due to a bug in the GSM firmware Feb 03 12:01:42 and there's no way to get the ready signals without power cycling Feb 03 12:01:43 DocScrutinizer: I presume there is a cache of the pin to re-register again.... Feb 03 12:01:53 mickey|sofa: I C. eeeek Feb 03 12:02:32 however, fsogsmd will save the PIN and deliver it, should it be necessary to powercycle the modem Feb 03 12:02:57 so in fsogsmd a reset will hopefully fully be hidden Feb 03 12:02:58 oooh, didn't know Feb 03 12:03:51 just a matter of finding time to do the remaining fsogsmd work Feb 03 12:04:05 before we can deploy it in SHR Feb 03 12:04:12 mickey|sofa: so not yet implmented Feb 03 12:04:25 not fully Feb 03 12:04:34 mickey|sofa: OK Feb 03 12:04:35 half-way through Feb 03 12:04:56 octavsly: anyway your problem most certainly is a fso hang for whatever reason Feb 03 12:05:10 mickey|sofa: how do you detect that the modem needs a power-cycle? Feb 03 12:06:02 octavsly: the modem NEVER needs a powercycle ;-) FSO needs to sync the modem by powercycling it Feb 03 12:06:28 if it doesn't answer after a couple of retries, it's likely dead Feb 03 12:06:59 or on vacation... and needs to be kicked Feb 03 12:07:36 mickey|sofa: FSO.... what is that? I know you'll kick me too :-D Feb 03 12:07:50 FSO is an acronym for freesmartphone.org Feb 03 12:07:56 ~fso Feb 03 12:07:57 somebody said fso was the freesmartphone.org software release. It includes the backbone software for the phone hardware (and some other things) and a working dialer, and not much else. Feb 03 12:08:06 hmm, not quite Feb 03 12:08:29 apt: no, fso is the freesmartphone.org mobile devices middleware. Feb 03 12:08:31 mickey|sofa: okay Feb 03 12:08:31 Ok. I am running to lunch. till later Feb 03 12:08:36 ~bon appetit Feb 03 12:08:37 rumour has it, bon appetit is smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Smaklig måltid!. Hyvää ruokahalua. Bo Proveito Feb 03 12:09:19 thankx Feb 03 12:10:10 hackers tear down emission trade in half of europe XP Feb 03 12:18:37 DocScrutinizer: when you meet dieter next time it might be useful to ask him about that Feb 03 12:18:51 (inability to drop out of MUX into AT mode) Feb 03 12:19:35 mickey|sofa: I rarely see or talk to Dieter, but for sure I can ping him wrt that Feb 03 12:20:05 mickey|sofa: is there some some specific reason for dropping out of mux mode? Feb 03 12:20:18 I'm not sure if he is willing to do any further development, after OM kicked him off the payroll Feb 03 12:20:48 mrmoku: Can you have a look at this? http://heinervdm.dyndns.info/shr-dist/FOSDEM-2010.pdf Feb 03 12:23:18 mickey|sofa: I'd need a precise description how to reproduce the error, preferrably with basic means like mickeyterm Feb 03 12:25:30 mickey|sofa: even better: ticketnumber Feb 03 12:25:34 :-) Feb 03 12:26:29 omg how I missed those gsm fw bugs :-S Feb 03 12:27:02 almost was afraid we'll never step to MOKO12 ;-P Feb 03 12:27:56 Feb 03 12:31:40 no way with mickeyterm Feb 03 12:31:44 it's binary commands :) Feb 03 12:32:31 lindi-: it's handy for reconnecting Feb 03 12:32:43 lindi-: the alternative is power cycling, but that always takes longer Feb 03 12:36:02 Heinervdm: apart from a couple of typos, it looks good. Might want to add "FSOSHRCON 2010 (planning)" to the "events" page Feb 03 12:36:13 Heinervdm: interesting presentation. a bit too light in the thechnical part Feb 03 12:36:36 mickey|sofa: typos are my big problem... Feb 03 12:36:39 :) Feb 03 12:36:59 Heinervdm: how much time is allocated? Feb 03 12:37:20 only 30 min Feb 03 12:37:33 technical details need to go in the discussion i'm afraid Feb 03 12:37:37 octavsly: but i don't want to takl for 30 Minutes :) Feb 03 12:37:52 Heinervdm: I can imagine 5 minute for questions Feb 03 12:38:07 mickey|sofa: I need a precise question to ask to Dieter Feb 03 12:38:09 i planned for 15 minutes Feb 03 12:38:23 Heinervdm: if they allocated you 30minutes Feb 03 12:38:33 mickey|sofa: not "hey, what's wrong with dropping out of muxmode?" Feb 03 12:38:35 Heinervdm: you should talk arund 25 Feb 03 12:38:40 octavsly: and i asked for 20 minutes, so 5 minutes of discussion, and then i got 30... Feb 03 12:38:57 DocScrutinizer: it doesn't get more precide than "the modem ignores the MUX DISCONNECT command" Feb 03 12:39:04 Heinervdm: nice slides! btw we're using 1.7.4 xorg in shr-u Feb 03 12:39:06 Heinervdm: then you are lucky. it means they think that you have something to say. So you can put more info Feb 03 12:39:27 Heinervdm: and I can bump it to 1.7.4.902 to show how bleeding edge we are :) Feb 03 12:39:38 :D Feb 03 12:40:10 mickey|sofa: If you can point to a source snipplet using a #define MUX_DISCONNECT and failing to issue the cmd, then we may start the game Feb 03 12:40:56 k, i'll see whether i can cook something up Feb 03 12:41:05 it's not that important anyways Feb 03 12:41:13 just one of the many warts Feb 03 12:41:21 same as not responding to TEST commands Feb 03 12:41:28 and ignoreing PSC (power save) commands Feb 03 12:42:00 instead we are blessed with that bogus deep sleep behaviour :D Feb 03 12:42:11 yeah Feb 03 12:42:50 I myself noticed a number of flaws in AT interpreter Feb 03 12:43:23 ATxy != ATxy=0 just one of them Feb 03 12:43:37 err Feb 03 12:43:42 ATxy != ATxy0 just one of them Feb 03 12:44:14 not to forget the sweet full crash on overloading in mux mode :) Feb 03 12:44:17 for AT

Feb 03 12:45:03 that one was tracked down to flaws in userland RTS/CTS handling, no? Feb 03 12:45:32 and the +CME ERROR 512 loop, if you enable CLIP Feb 03 12:45:37 yes and no Feb 03 12:45:46 lets say i would expect that userland should never be able to crash the modem Feb 03 12:45:51 even if it does not honor flow control Feb 03 12:46:00 ack Feb 03 12:46:58 stack overflow in an embedded system triggered via interface always is a bad sign Feb 03 12:47:05 yep Feb 03 12:48:25 mickey|sofa: reconnecting? Feb 03 12:50:04 FWIW: the *hw* FIFO in calypso seems 64bytes each TX/RX Feb 03 12:52:27 lindi-: hooking up to the modem while it's not being reset Feb 03 12:54:00 mickey|sofa: you mean like restarting muxer without dropping from the gsm network? Feb 03 12:54:25 lindi-: no, rather restarting the controlling process, i.e. fsogsmd Feb 03 12:54:37 and reattaching to a modem with a "dirty" state Feb 03 12:54:52 since you don't know who#s been previously talking to it Feb 03 12:55:07 some channels might have been crashed Feb 03 12:55:27 so sending a MUX DISC, dropping into AT, and relaunching MUX would be helpful Feb 03 12:55:57 so a power cycle is the last resort Feb 03 12:56:27 which -- as you said -- drops us off from the net for a while Feb 03 12:56:55 and takes *ages* Feb 03 12:58:20 ya, the calypso is not the fastest out there Feb 03 12:59:28 well a powercycle takes ages on *all* GSM devices. Maybe even ages+1day on calypso Feb 03 12:59:56 true Feb 03 13:00:19 apparantly the SIM interface is like Glamo ;) Feb 03 13:01:06 * DocScrutinizer expects the most delay coming from scanning all the bands for all visible BTS Feb 03 13:02:18 i found +COPS being quite speedy on some models Feb 03 13:02:48 lol, yeah speedy like launching a preloaded app Feb 03 13:02:50 whereas +CFUN ... takes long everywhere Feb 03 13:07:26 * DocScrutinizer wonders if +CFUN starts negotiations with the network Feb 03 13:07:32 Heinervdm: wow, very nice :) Feb 03 13:08:09 TAsn: ping Feb 03 13:09:47 mrmoku: What's with the redesign? I don't really know what you are exactly doing :) Feb 03 13:11:11 and what are the plans for the future? Feb 03 13:11:20 Heinervdm: right now I'm uniting the contacts show and edit views Feb 03 13:11:33 to be able to edit the fields in one view Feb 03 13:11:33 Ok, last Point "World domination" is missing ;) Feb 03 13:11:38 yeah Feb 03 13:11:46 but that one is obvious ;) Feb 03 13:11:53 maybe we could put buzz to a purpose finally. Creating a GSM TX monitor app :-) Feb 03 13:13:02 version 0.0.1 : route headset mic to sounder/speaker Feb 03 13:13:19 Heinervdm: future plans are to convert all views from the old frame based thing (window.h) to the new view system (ui-utils.h) Feb 03 13:13:29 which will include caching of the views Feb 03 13:13:42 aka the contact list, message list will already be loaded when you open them Feb 03 13:13:57 oh, thought that's already done be phoneuid Feb 03 13:14:07 not for the old screens Feb 03 13:14:17 ok Feb 03 13:14:21 right now dialer, idle-screen (and not yet official quick settings) are cached Feb 03 13:14:49 for the contacts and messages we need to put listening to signals for changes in place Feb 03 13:14:57 and convert them to the new view system Feb 03 13:15:54 Heinervdm: in the far, far future we probably want to make multiple toolkit backends functional Feb 03 13:16:00 * DocScrutinizer wonders how long until swap is mandatory for SHR, and will bring everything to a grinding halt despite the ultrafast caching scheme Feb 03 13:16:11 :P Feb 03 13:16:22 fast halt then though ;) Feb 03 13:17:03 don't worry Feb 03 13:17:12 we gain lots of RAM when we boot with fso-boot Feb 03 13:17:19 no more unnecessary processes hanging around :D Feb 03 13:17:22 and drop frameworkd :P Feb 03 13:17:26 err, ya. Feb 03 13:17:29 almost forgot that Feb 03 13:17:46 need some volunteer for the opimd rewrite though ;) Feb 03 13:18:17 true, but vala should be mature enough to make this a nice experience when it's due Feb 03 13:18:32 FSO-BOOT! Feb 03 13:18:35 Someone started fsopimd but never heard of that again Feb 03 13:18:37 yeah... will need convince dos1 to learn it :) Feb 03 13:18:42 mickey|sofa, mrmoku: what about revamping fso-console-image on shr buildhost for om-gta01 (lite is too big) and small console image for rescure would be usefull, but I'm not sure if we want fso or just console-image .. Feb 03 13:19:09 opimd needs a respec instead of rewrite Feb 03 13:19:24 opimd is a pain. Feb 03 13:19:39 it takes up to 5 minutes until the contacs app actually shows contacts Feb 03 13:19:50 JaMa|Gone: hmm... with fso-console one might have more fso related tools? Feb 03 13:20:09 so spontaneous without thinking too much... Feb 03 13:20:19 hmm... maybe even do a shr-console-image? :P Feb 03 13:20:28 mrmoku: but still cannot do rescure call (unless expirienced with AT cmds.. :) Feb 03 13:20:33 spaetz: that's due to ultrafast sqlite instead of nonoptimized grep Feb 03 13:20:49 just make it fso2-console-image Feb 03 13:20:53 and leave out the python Feb 03 13:20:57 spaetz: how many contacts do you have? Feb 03 13:21:02 that saves a couple of thousand tasks to build Feb 03 13:21:02 mrmoku: I was checking http://build.shr-project.org/shr-unstable/images/om-gta01/shr-lite-eglibc-ipk--20100203-om-gta01-testlab/installed-package-sizes.txt Feb 03 13:21:09 and due to opimd even further "optimizing" speed by caching all the shit Feb 03 13:21:22 mrmoku: and with few less fonts we could make it < 64MB Feb 03 13:21:55 mrmoku: about 20-30 Feb 03 13:22:16 spaetz: that is NOT normal Feb 03 13:22:24 well 5 minutes might be an exaggeration, but it takes at least 3 minutes AFTER the SIM PIN succeeded Feb 03 13:22:41 before that I get an empty contacts screen Feb 03 13:22:47 spaetz: I have around the same and it takes 3 seconds to show the first contact Feb 03 13:23:06 spaetz: ahh, that is different thing though Feb 03 13:23:21 that is the time frameworkd takes to get the SIM ready Feb 03 13:23:24 my contacts are mostly on SIM Feb 03 13:23:32 before that you will only see sqlite contacts Feb 03 13:23:47 mrmoku: but it takes 3 minutes after the time I can make phone calls Feb 03 13:23:47 not much opimd could do about that :/ Feb 03 13:23:59 spaetz: a la: 99.9sec for initializing structures and filling cache, 0.1 sec to SELECT the records (opimd), vs: 0.3sec for loading grep, 0.2 sec for grepping all the needed lines out of a 100k lines file Feb 03 13:24:00 hmm Feb 03 13:24:06 so, agenda for FSOSHRCON Feb 03 13:24:09 so there should be plenty of time to fetch those. Feb 03 13:24:12 1.) integrate fso-boot Feb 03 13:24:16 2.) integrate fsogsmd Feb 03 13:24:24 what else do you want for the sprint? Feb 03 13:24:33 ditching opimd Feb 03 13:24:34 :) Feb 03 13:24:35 something UI-relevant? Feb 03 13:24:47 heh, if we can take a month vacation then.... Feb 03 13:24:50 JaMa|Gone: it has to be > 56MB i think Feb 03 13:24:53 :) Feb 03 13:25:08 maybe adjust the UI to palm pre and/or HTC Dream? Feb 03 13:25:15 Heinervdm: for jffs2 before flashing right? Feb 03 13:25:27 ahh, mickey|sofa you could also write the "How to program in vala for SHR programmers" book during the sprint please :) Feb 03 13:25:35 JaMa|Gone: jffs2 is uncompressed, isn't it? Feb 03 13:25:38 Heinervdm: only vibration instead sounds saves few more MB Feb 03 13:25:42 so it's the same Feb 03 13:25:56 spaetz: i'd love to... get the clone machine ready Feb 03 13:26:10 Heinervdm: jffs2 is always compressed Feb 03 13:26:12 replicator: we need a dozen more mickeys please Feb 03 13:26:27 Heinervdm: one of the reason why it's so slow :) Feb 03 13:26:29 spaetz: :) Feb 03 13:26:33 JaMa|Gone: the gta01 has 64MB, 2 for the kernel, and some other for other things, so there was less then 60MB avaible Feb 03 13:26:52 Heinervdm: I see.. Feb 03 13:27:40 mrmoku: BTW, yesterday I've got a black screen during the phone call. Feb 03 13:27:46 JaMa|Gone: we looked that up the last time we build running images for gta01, but i forgot how much space was avaible Feb 03 13:27:53 Heinervdm: btw could remove kernel from /boot if we know that that image (while creating jffs, ubi) but I'm not sure how to do this in OE Feb 03 13:28:01 For one mickey|sofa it takes 6 months to write "how to program vala". How long will it take for 500 mickey|sofa ? X-P Feb 03 13:28:16 depends on the sofa Feb 03 13:28:23 hehehehe Feb 03 13:28:24 Heinervdm: I guess that whole rootfs is prepared once and then different formats just repacked from it Feb 03 13:28:55 JaMa|Gone: i think so too Feb 03 13:28:55 must be a biiiig sofa then :) Feb 03 13:29:06 yoh Feb 03 13:29:14 we should get IKEA to sponsor the book then. Feb 03 13:29:19 Heinervdm: but there is also some PREPROCESS variable for image types.. so maybe we can put rm -rf /boot there.. Feb 03 13:29:21 lol Feb 03 13:29:28 mickey|sofa, DocScrutinizer: do you know how big rootfs partition of gta01 is? Feb 03 13:29:35 I am sure they want to make a commercial out of it when 500 Mickeys sit on the same sofa, typing Feb 03 13:29:42 50some Feb 03 13:29:46 Heinervdm: 54 Feb 03 13:29:46 we can ask replicator to clone mickey WITH sofa.. Feb 03 13:29:49 i think Feb 03 13:30:05 good point Feb 03 13:30:13 more sofas more fun Feb 03 13:30:23 err... more productivity, i mean Feb 03 13:30:30 :) Feb 03 13:30:35 then IKEA can sell it with integrated neo Feb 03 13:30:47 plural Feb 03 13:30:54 err I mean him Feb 03 13:30:55 500 neo FRs Feb 03 13:31:16 amount of neo users will be doubled every year or so then Feb 03 13:31:26 JaMa|Gone: and i think we can remove some kernel modules, there are some relly big ones Feb 03 13:31:34 SOFA oerreboe now with vibrator build it. buzz-guarantee Feb 03 13:31:43 hehe Feb 03 13:31:48 :) Feb 03 13:32:37 we should launch the fsoshr planning soon, is there any month in summer that would be completely not doable for you guys? Feb 03 13:32:51 and just en passant we get the biggest embedded cluster ;-) Feb 03 13:33:49 DocScrutinizer: :) Sofa meshing Feb 03 13:34:04 mid-end July won't work here for sure, but that's just me and not important Feb 03 13:35:13 expect school to have a major impact on that figure Feb 03 13:35:57 mainly German school of all the bundeslaender Feb 03 13:37:14 * mickey|sofa microwaves chili con carne from yesterday Feb 03 13:38:01 http://www.schulferien.org/Schulferien_nach_Jahren/2010/schulferien_2010.html Feb 03 13:39:21 * DocScrutinizer microwaves another coffee latte mach Feb 03 13:44:51 * Weiss harbinges the doom tract Feb 03 13:46:03 those buttons at the bottom of SHR, what are these for? Feb 03 13:46:12 the first one is the callibration util Feb 03 13:46:21 but can I set my own? Feb 03 13:48:50 g2g Feb 03 13:56:45 * DocScrutinizer bans pieterc for driveby-asking Feb 03 13:59:49 * DocScrutinizer /kick pieterc "come back to ask questions when you got time to read the answers" Feb 03 14:07:47 DocScrutinizer, drive by asking? :D Feb 03 14:09:15 cz_jc: like drive by shooting in the USA Feb 03 14:09:18 What's up with xinput_calibrator? It seems that the values that it sets don't make any sense... Feb 03 14:09:47 DocScrutinizer, what if he didn't plan on going right after he asked but was forced to ? Feb 03 14:10:31 then he might avoid spanking next time I meet him here ;-P Feb 03 14:12:35 who's x-calib master? Feb 03 14:12:49 Weiss: ?? Feb 03 14:13:53 not me :)... AIUI the kernel's calibration numbers came from me (I spent an hour or so fiddling and put the numbers on my site along with the other Xorg stuff to make DRI work, not really intended for proper use but good enough) Feb 03 14:13:58 ChristW: I'm having problems too... noticing mostly when using the keyboard... have to press the keys slightly on the right side Feb 03 14:14:10 not very much off though Feb 03 14:14:21 mrmoku: My presses are all over the place, and apparently mirrored in some way. Feb 03 14:15:24 ChristW: hmm... no. that one I don't have Feb 03 14:15:42 ChristW: did you try to remove the calibration file and recalibrate? Feb 03 14:15:55 Yes, but to no avail... Feb 03 14:16:03 :/ Feb 03 14:16:07 I've no problems with my calibration Feb 03 14:16:18 i've done it yesterday Feb 03 14:16:31 calibration works for me too Feb 03 14:16:59 xrandr? or something like that? Feb 03 14:17:06 $ cat /etc/pointercal.xinput Feb 03 14:17:07 xinput set-int-prop "Touchscreen" "Evdev Axes Swap" 8 1; xinput set-int-prop "Touchscreen" "Evdev Axis Calibration" 32 951 373 517 465 Feb 03 14:17:14 BBIAB Feb 03 14:18:18 evdev axes swap sounds fishy Feb 03 14:22:55 WTF? Feb 03 14:25:33 indeed... I have: xinput set-int-prop "Touchscreen" "Evdev Axis Calibration" 32 74 1003 12 1007; Feb 03 14:25:48 no swap in there... Feb 03 14:26:32 xinput set-int-prop "Touchscreen" "Evdev Axis Calibration" 32 13 1002 4 1017; Feb 03 14:26:38 thats mine Feb 03 14:26:39 "same" here.. xinput set-int-prop "Touchscreen" "Evdev Axis Calibration" 32 11 1012 1 1022; Feb 03 14:27:05 ChristW: your /etc/pointercal.xinput seems to be created *before* kernel calibration Feb 03 14:27:18 ChristW: check Xsession.d/89xCalibration Feb 03 14:27:34 ChristW: xinput-calibrator-once.sh should be last script there.. Feb 03 14:27:41 * DocScrutinizer got 99 points. Alas the washing engine subscription is 100 pts Feb 03 14:28:32 ChristW: on sunday morning it was before kernel calibration and then I moved it.. because it's safer to expect that for first time it was executed with ts already calibrated Feb 03 14:30:45 * topro desperately waiting for an update Feb 03 14:31:02 kinda OT: would anybody know how to disable touchpad on a recent Laptop linux? Feb 03 14:31:07 freesmartphone.org: 03mickey 07cornucopia * r3489dad6ea8e 10/tools/fso-boot/ (src/main.vala vapi/linux.vapi): fso-boot: refactor creating a filesystem into seperate method Feb 03 14:31:09 freesmartphone.org: 03mickey 07cornucopia * rf3816693a528 10/tools/fso-boot/ (src/main.vala vapi/linux.vapi): fso-boot: vapi: complete binding of net/if.h Feb 03 14:31:17 DocScrutinizer: using hal, i think Feb 03 14:31:24 on gnome, there is a gui for this Feb 03 14:31:48 on gnome there's a GUI? cool Feb 03 14:31:58 mind to share the name? Feb 03 14:31:59 but it only disable the touchpad when typing :/ Feb 03 14:32:05 so that's not what you need i fear Feb 03 14:32:11 that's exactly what I need Feb 03 14:32:15 gnome-mouse-properties Feb 03 14:32:29 on KDE it's broken Feb 03 14:32:48 KSynaptics Feb 03 14:33:08 borked by X API change AIUI Feb 03 14:36:16 DocScrutinizer: can't turn it off in BIOS? Feb 03 14:37:02 I probably could. But then I never can use it :-/ Feb 03 14:37:13 * mrmoku hates touchpads :P Feb 03 14:37:58 gnome-mouse-properties fails on KDE to do anything wrt tp Feb 03 14:38:02 :-( Feb 03 14:39:06 another case of "inproved sw to FUBAR" - this time X Feb 03 14:40:09 mickeyl: with debug enabled in fso-abyss (and running it in gdb) shouldn't be something logged everytime frameworkd asks for channel? Feb 03 14:40:48 2010.02.03 15:40:07.995 ogsmd.modem.abstract ERROR could not open channel UNSOL, retrying in 2 seconds Feb 03 14:40:51 2010.02.03 15:40:08.4 ogsmd.modem.abstract DEBUG trying to open channel CALL Feb 03 14:41:22 I see framework retrying UNSOL/CALL/MISC channel for last few hours and still without any log message in fso-abyss Feb 03 14:41:42 try Feb 03 14:41:48 in /etc/abyss.conf Feb 03 14:41:51 [libgsm0710mux] Feb 03 14:41:57 log_level = DEBUG Feb 03 14:41:59 log_to = ... Feb 03 14:42:39 fso-abyss is still cycling in poll loop http://pastebin.ca/1777065 Feb 03 14:44:22 mickeyl: FSO_LOG_LEVEL=DEBUG FSO_LOG_TO=stderr:colors is not supported in fso-abyss? Feb 03 14:45:04 ah, you're using the env override Feb 03 14:45:07 that way it should work Feb 03 14:45:19 is that the latest abyss & libgms0710mux? Feb 03 14:46:12 mickeyl: http://pastebin.ca/1777068 Feb 03 14:46:20 mickeyl: yes Feb 03 14:48:15 mrmoku: you got no idea how much e that thouchpad*I* hat Feb 03 14:48:50 mickeyl: http://pastebin.ca/1777076 Feb 03 14:48:56 JaMa|Gone: fso-abyss and libgms0710mux were adjusted later to the logging infrastructure, perhaps i did something wrong there. please try the "explicit" way in /etc/abyss.conf Feb 03 14:49:08 ah, that's better Feb 03 14:49:11 mickeyl: notice that debug output is printed only AFTER frameworkd is killed Feb 03 14:49:29 mickeyl: tried explicit too.. Feb 03 14:50:02 file buffer? Feb 03 14:50:19 i don't think we explicitly flush() in the logger after each statement Feb 03 14:50:54 in any case, something's wrong there Feb 03 14:51:10 it tries sending mux commands before attempting to move into mux mode Feb 03 14:51:14 mickeyl: even after few yours fso-abyss doesn print single line Feb 03 14:51:18 do you have multiple clients running? Feb 03 14:51:40 mickeyl: and then it prints about 1 failed attempt (while frameworkd did hundrets of attempts) Feb 03 14:52:03 what do you mean by multiple? Feb 03 14:52:03 could be multiple fso-abyss? Feb 03 14:52:12 no killed before running.. Feb 03 14:52:46 hmm, no idea. which image do you see this behaviour on and does it fail all the time? Feb 03 14:52:47