**** BEGIN LOGGING AT Sat Aug 14 02:59:57 2010 Aug 14 07:03:27 ok i give up on my jitterless touchscreen idea. my latest idea 'wait for vsync then slowdown pixclock, then get touchscreen position' works and i have almost jitterless input (+-1 pixel jitter still exist). but... our dear touchscreen is not so simple! and it surely requires filtering for finger operations... Aug 14 07:04:12 so, i think proting and using moko driver to .34 is best idea to get perfectly usable touchscreen. Aug 14 07:07:11 seem the touchscreen is just not built for finger operation, as data recieved is terrible.... and this happens only with finger and only if with light pressure Aug 14 07:14:26 gena2x: hi, btw i have now qtmoko on 2.6.34 kernel with debian testing as rootfs Aug 14 07:14:48 gena2x: maybe if you are interested... Aug 14 07:15:20 radekp: hi, radek. and first, how did you solve touchscreen problem? Aug 14 07:15:35 gena2x: which touchscreen problems? Aug 14 07:15:46 jitter Aug 14 07:15:57 ahh havent solved it :) Aug 14 07:16:10 but it's not so bad IMO Aug 14 07:16:43 it's very, very bad in fact. Aug 14 07:16:49 for muctiple reasons. Aug 14 07:16:56 *multiple Aug 14 07:17:06 i did much testing here. Aug 14 07:18:08 yes i have problems typing on the qtmoko small virtual keyboard Aug 14 07:18:22 jitter is +-10 pixels. and finger things are even worse. Aug 14 07:18:25 but i hope to replace that keyboard with something bigger Aug 14 07:18:50 oh, btw! i recently tried default qtmoko input method (handwriting) Aug 14 07:19:23 never tried before, and in fact were a bit amazed. Aug 14 07:20:30 (about keyboards) Aug 14 07:21:03 handwriting could be quite nice - but i mostly type in terminal so it's not for me ;-) Aug 14 07:21:20 stylus. Aug 14 07:21:41 and what is your idea of keyboard? Aug 14 07:21:49 something new? Aug 14 07:23:51 gena2x: i would like the same layout and functionality as in SHR Aug 14 07:24:10 gena2x: i like the SHR keyboard Aug 14 07:24:21 hm... Aug 14 07:24:47 gena2x: + i had idea: first click zooms and seconds presses the key Aug 14 07:25:16 gena2x: usefull e.g. in bus when it's difficult to hit the small keys Aug 14 07:26:16 gena2x: but i have to write some small test app to check if it's usable idea Aug 14 07:26:43 i didn't tried shr's keyboard recently, but i found it too slow to be usable at all, may be just implemented in some other way it may be good. Aug 14 07:28:56 i plan just to have the same layout Aug 14 07:29:11 it has keys like 2x bigger then in qtmoko Aug 14 07:29:16 i think i saw samething like that you talk somewhere, for finger friendly keyboar. first you select place, than it zoom small area and you move finger left/right/up/down to select particular letter. Aug 14 07:30:28 this is not really like in shr, and i think it is slightly better. Aug 14 07:30:59 as you need 'bigger' moves after initial 'click' to change letter. Aug 14 07:31:37 afair, in shrs keyboard you just get magnified area. Aug 14 07:32:21 but qtmoko has 3 imput methods? Aug 14 07:32:32 in fact 4 Aug 14 07:32:34 i thought second one is finger friendly? Aug 14 07:32:55 hm. something disabled? Aug 14 07:32:56 i think it can be good for SMS Aug 14 07:33:23 but i dont write SMS often - i need good keyboard for terminal Aug 14 07:33:45 besides i really hate prediction Aug 14 07:34:27 i feel like i am using microsoft office ;-) Aug 14 07:34:35 heh. Aug 14 07:35:31 anyway keyboard will be for the release after this one Aug 14 07:35:53 i want to now release experimental qtmoko with 2.6.34 kernel Aug 14 07:36:01 then stable one with 2.6.29 Aug 14 07:36:15 don't remind ms office for me. last time i used it, i had to use that damn new damn interface to help pro users who knows program for 15 years find new button locations. Aug 14 07:36:26 but i wanted to fix as many things with 2.6.34 as possible first Aug 14 07:36:50 can i help? Aug 14 07:37:06 i am now trying to find out problem with high current Aug 14 07:37:08 some particular problems with kernel? Aug 14 07:37:30 btw, do it suspend/resume for you without problems? Aug 14 07:37:43 if you boot 2.6.34 to console and suspend - it takes like 50mA Aug 14 07:37:56 2.6.29 takes just 9mA Aug 14 07:38:19 then in 2.6.34 you turn on modem and the current_now drops in suspen to 20mA Aug 14 07:38:43 hm... Aug 14 07:38:50 something similar was in 2.6.29 Aug 14 07:38:52 http://lists.openmoko.org/pipermail/hardware/2009-February/001032.html Aug 14 07:38:56 i saw your question to paul Aug 14 07:39:29 but i am kernel developer, it's quite hard to fix it for me Aug 14 07:39:45 you just want to try patch? Aug 14 07:41:06 which patch? Aug 14 07:41:19 "BTW (on subject) Andy posted a patch today which is stopping burning of Aug 14 07:41:19 70mA(!! duh) via high-level uart lines when modem powered down. Aug 14 07:41:19 Powering down (and up) modem via sysfs node should be fixed and work same time Aug 14 07:41:19 now thanks to patch of PaulFerster. Aug 14 07:41:19 " Aug 14 07:41:41 those patches no longer apply Aug 14 07:41:59 let me try... Aug 14 07:42:07 Does listeting to RHCP makes me lack "taste" or gay or something? Aug 14 07:42:08 the gsm power management is in 2.6.34 slightly modified Aug 14 07:42:52 RHCP? Aug 14 07:42:59 PaulFertser: Red hot chilly peppers? Aug 14 07:43:06 red hot chili peppers, yes Aug 14 07:43:37 * radekp listens to them sometimes too ;-) Aug 14 07:43:50 eh, eh... i can't hear this anymore :( only classic... Aug 14 07:44:27 gena2x: do you go to concerts of modern academic music? Aug 14 07:47:42 PaulFertser: very very rare, and not really academic i mean more something like classical romances or so. i have relatively high quality audio system at home and too few friends with similar music preferences to go somewhere. Aug 14 07:49:46 radekp: so, back to our topic. Aug 14 07:52:14 heh Aug 14 07:52:54 msleep(100); ... msleep(500); ... msleep(1000); Aug 14 07:58:19 heh, my kernel can't even suspend :) Aug 14 07:59:56 mine can ;-) Aug 14 08:00:05 how do you do it? Aug 14 08:00:15 echo mem > /sys/power/state ? Aug 14 08:00:20 debian-gta02:~# echo mem > /sys/power/state Aug 14 08:00:47 this seem because of my last .config changes, i'll revert em now... Aug 14 08:06:44 but as far as i understood, this patch only affect system then GSM is off? Aug 14 08:07:03 so, if GSM is on it can't be reason of additional power draw? Aug 14 08:08:19 gena2x: yes, that's why i never said i know the reason for your additional power draw in suspend. Aug 14 08:09:14 em. _i_ have no power draw :) as _i_ have no suspend :) Aug 14 08:10:27 gena2x: ah, sorry. Aug 14 08:10:46 gena2x: i mean Q-Master reports an increased power consumption in suspend with 2.6.32. Aug 14 08:11:10 yeah, i understand. Aug 14 08:15:10 need first to find out which subsystem responsible for additional draw. Aug 14 08:19:03 that was my idea too Aug 14 08:19:26 but when it takes 50mA with gsm off it's hard to find anything Aug 14 08:43:28 Hey, RHCP steals a song from a russian indie songwriter: http://dl1.zakachali.com/download/05cf212d8a0261bcc9c0be353988a34a (written in 1994, this recording dates to 2000); RHCP http://www.youtube.com/watch?v=k9xCCjL1RZE (recorded 2005) Aug 14 08:54:19 damn. Aug 14 08:54:57 leds seem missing. Aug 14 08:55:15 yup, leds and vibrator Aug 14 08:56:47 straaange Aug 14 08:56:52 actually they work. Aug 14 08:57:03 with om led blah 63 Aug 14 08:57:13 but not with my BUG(). damn... Aug 14 08:57:39 hm. Aug 14 08:58:01 they not missing actually. Aug 14 08:58:26 but my BUG in wrong location :) Aug 14 09:02:14 what is best lisp implementation for embedded devices (something smaller and possibly easier to cross-compile)? Aug 14 09:02:27 clisp(clisp),gcl(GCL),lisp(CMUCL),scl(SCL),sbcl(SBCL),lisp(ACL),openmcl(OpenMCL),ecl(ECL) Aug 14 09:02:45 * JaMa|GoNe used only sbcl on desktop Aug 14 09:07:15 mickeyl: (adjtime) no real new findings... kernel returns TIME_ERROR and glibc doesn't consider that an error Aug 14 09:07:35 mickeyl: and kernel gives that TIME_ERROR, because time_status = STA_UNSYNC Aug 14 09:07:42 dunno what that actually means Aug 14 09:08:37 mickeyl: maybe we should disable the adjtime until that is sorted out? Aug 14 09:09:17 radekp: it seems to me that my FR is also somewhat power hungry in suspend Aug 14 09:09:38 radekp: but I didn't do any kind of measurement, just saw low battery life Aug 14 09:10:53 radekp: can I see the current w/o measuring it with an ammeter? Aug 14 09:11:02 zub: this known feature of both .32 and .34. see one of latest bug reports Aug 14 09:12:44 zub: https://docs.openmoko.org/trac/ticket/2349 Aug 14 09:12:45 zub: yes, it's possible via sysfs node - called current_now Aug 14 09:13:12 zub: find /sys | grep current_now Aug 14 09:13:31 zub: this command will tell you full path to that node Aug 14 09:13:47 thanks, let me see Aug 14 09:13:52 * zub is bothered by the issue Aug 14 09:14:07 enough to try to help :) Aug 14 09:14:12 see bug report, it contains path and details Aug 14 09:15:25 it's known that 29 was ok and 32 is broken? Aug 14 09:15:56 yes. Aug 14 09:16:02 zub: /sys/devices/platform/s3c2440-i2c/i2c-0/0-0073/hdq.0/hdq/dump Aug 14 09:16:08 zub: http://people.openmoko.org/joerg/battery/bq27k-detail Aug 14 09:16:21 zub: echo "0x1B - 0x1A: `pv 0x1B 0x1A`" " SI Standby Current High - Low Byte 3.57 µV (1) R" Aug 14 09:16:43 SI is current in suspend if I understood DocScrutinizer51 correctly :-) Aug 14 09:17:42 info overflow, let me chew it :) Aug 14 09:20:59 hello, somebody know if is possible install new apps from a Market in android-on-freerunner-cupcake-week-2010-16? Aug 14 09:21:55 the browser says that market://something=?I-want is no aviable.... Aug 14 09:28:44 gnustavo: never saw android discussions on this channel, try #android-on-freerunner Aug 14 09:29:02 mrmoku: a lame question - where does the data come from? Aug 14 09:29:19 I mean, I see it's from /sys/devices/platform/s3c2440-i2c/i2c-0/0-0073/hdq.0/hdq/dump, but this in turn comes from where? Aug 14 09:30:05 gena2x thanks Aug 14 09:30:36 JaMa|GoNe: hi. I rebuild from scratch shr-image and in nearly succeded! http://tinderbox.openembedded.org/packages/692445/ Aug 14 09:38:29 zub: from the bq27k chip Aug 14 09:38:59 that's something in the FR that connects to the battery? Aug 14 09:41:28 ok, I googled Aug 14 09:41:38 zub: battery monitoring thing Aug 14 09:41:51 as far as i understand, according to path, it's cpu's i2c bus, connected to pcf (0-0073), and i don't know what is hdq. Aug 14 09:45:31 gena2x: hdq is one-wire bus to the bq27k inside the battery. Aug 14 09:45:36 gena2x: the kernel bitbangs it. Aug 14 09:46:14 gena2x: at least used to, i can't understand why it's under i2c now. Aug 14 09:46:38 may be because it is connected to pcf? Aug 14 09:46:56 which is connected via i2c? Aug 14 09:47:10 path is from CPU :) Aug 14 09:47:15 gena2x: nope... Aug 14 09:47:26 pardon my ignorance - so bq27k is inside the battery, or in the FR? Aug 14 09:47:50 PaulFertser: hm... let me check chematics... Aug 14 09:48:24 gena2x: http://wiki.openmoko.org/images/1/1e/Close_to_Hardware_%E2%80%93Driver_Implementation.pdf Aug 14 09:48:25 zub: battery afail Aug 14 09:48:36 ok :) Aug 14 09:48:56 gena2x: goes directly to gpd14 Aug 14 09:48:59 hdq is going over that "CC" pin on the battery? Aug 14 09:49:28 zub: exactly Aug 14 09:50:25 then I wonder how can it tell what's "standby" (if that's ~ suspend) and normal current Aug 14 09:51:01 unless it's told by the FR that FR is now going to suspend Aug 14 09:51:33 zub: are you wondering how we can measure consumption while in suspend? Aug 14 09:52:12 PaulFertser: hm.. yes. Aug 14 09:52:13 well, I thought I'd see for mayself and I have no ammeter + even if I had, I'm too clumsy with wires :) Aug 14 09:52:40 PaulFertser: path were misleading :) Aug 14 09:52:41 mrmoku suggested DocScritinizer's bq27k-detail script, which looks interesting Aug 14 09:53:06 but last thing is I don't know how can the bq27k tell what's "suspend current" and what's "normal current" Aug 14 09:53:33 my wild guess battery knows nothing about system state - is it suspended or not. Aug 14 09:54:02 gena2x: same guess here Aug 14 09:54:11 so place where one reads 'current_now' - from registers or sysfs node is not important? Aug 14 09:54:47 how does resume due to low battery work? Aug 14 09:55:07 interrupt i guess Aug 14 09:55:08 (or does it?) Aug 14 09:55:21 lindi-: yes, it should Aug 14 09:55:22 pcf detects low battery and wakes. Aug 14 09:55:37 lindi-: pcf is configured to raise an interrupt, right. Aug 14 09:55:49 current_now < 0 means charging? Aug 14 09:55:58 gena2x: yes, it doesn't matter how you get that current_now Aug 14 09:57:00 PaulFertser: thanks for confirm Aug 14 09:58:02 is it known what part of the fr (gsm, wifi, ...?) is responsible for the drain in suspend? can it be identified? Aug 14 09:58:15 althought I guess it could be a complex mix Aug 14 09:58:26 zub: i guess, you should disable everything Aug 14 09:58:48 zub: and turn on one-by-one. or in reverse direction :) Aug 14 09:59:02 so far, seem nobody knows which subsystem Aug 14 09:59:15 ok, too bad I 1) don't know what all there is 2) don't know how to turn the subsystems off :) Aug 14 09:59:22 can you try the command I added to the bug about this? it causes crash here after a while Aug 14 09:59:33 ok, let me try Aug 14 10:00:44 I guess I should kill FSO daemons so they don't interfere, right? Aug 14 10:01:40 zub: yes or learn how to do it with fso if you prefer that Aug 14 10:02:45 hm.. seem we have mass brain-storm today on issue :) Aug 14 10:03:42 extra consumption clearly makes 2.6.32 unusable so it is important Aug 14 10:04:05 heh, i am using fr without suspend at all. Aug 14 10:04:54 just first this i do then i came to place with usb - connect it to charger and forget :) Aug 14 10:04:58 *thing Aug 14 10:05:27 gena2x: some people are using it as phone and pda :) Aug 14 10:05:56 suspend is essential and cpufreq would help Aug 14 10:06:05 i'm a using on not only as phone freqently, but also as player. Aug 14 10:06:27 player... audio/video player? Aug 14 10:06:28 how can you use it without suspend? Aug 14 10:06:44 audio. screen is too small for video Aug 14 10:06:45 GarthPS: "Aborted" looks like Ctrl+C during mkfs.jffs2 :) Aug 14 10:07:02 lindi-: it can live for a day with screen turned off without problems. Aug 14 10:07:34 lindi-: i can listen music for ~3 hours on qtmoko even with glamo on. Aug 14 10:07:35 gena2x: yes barely Aug 14 10:07:41 heh, GTA02 as an audio player? That's a bit masochist, no? Aug 14 10:08:02 i use it for audio too, not horrible Aug 14 10:08:10 so, it's easy, just connects it then you back. Aug 14 10:08:14 mostly podcasts Aug 14 10:08:26 ThibG: wolfson is not bad at all in fact. Aug 14 10:08:52 lindi-: trying while true; do cat /sys/class/power_supply/battery/current_now >>/var/log/power.log; sleep 1; done and seeing wild things Aug 14 10:08:57 gena2x: but I am often off the grid for longer than a day Aug 14 10:08:58 only terrible. really terrible. completely below anything is default headset. Aug 14 10:09:30 cause you had headset bundled with the GTA02? Aug 14 10:09:51 lindi-: i'm rare. but in such cases i have phone only first day. Aug 14 10:10:45 lindi-: byt it seems I'm getting ~ 43000-52000, once saw 29625 Aug 14 10:11:39 I'm somewhat surprised the numbers differ, both what might be suspend and what's "normal" Aug 14 10:11:51 ThibG: wolfson is quite ok, but one needs a bass fix to listen to the music. Aug 14 10:12:16 hm, if you say so Aug 14 10:12:46 * ThibG never ever used headsets with the GTA02, so, its point of view is really biased Aug 14 10:15:15 zub: i also wonder why the numbers are so much different in 2.6.32 & 2.6.34 kernel Aug 14 10:16:09 * radekp has 18mA sometimes which is still too much compared to 12mA with 2.6.29 kernel Aug 14 10:16:35 ThibG: yes. i am usure about my english but bundle include stereo headset. Aug 14 10:16:39 it should be like 10mA with gsm off and 12mA when GSM is in deep sleep Aug 14 10:17:04 Even less Aug 14 10:17:06 ThibG: and it's terrible. Aug 14 10:17:46 radekp: current_now is in uA? so I should be seeing 10000 to 12000? Aug 14 10:17:50 ThibG: using anything else produces high quality sound. Aug 14 10:18:09 zub: yes Aug 14 10:19:35 ThibG: can say how 'high' is quality exactly, didn't investigate it yet. but for my relatively precise ears, quality is good. Aug 14 10:23:33 lindi-: btw, you can also just turn off phone overnight, etc to save battery. my real problem with that i never sure if fr will resume ok on incoming call, etc. i missed few calls without warnings this way, so i just decided to not suspend - this is only reliable way. Aug 14 10:25:51 btw, recently i noticed interesting 'feature'. Aug 14 10:26:41 once, my call were interrupted, but qtmoko were still working, but in some strange state - can't turn off or something like this. Aug 14 10:27:02 it turned out than battery had almost no charge. Aug 14 10:27:17 ok, I don't remember having one bundled with it, but anyway, even with my very-high-quality audio player, the bundled hearbuds are crap Aug 14 10:27:51 JaMa|GoNe: NO! seriously! I was sleeping!!I could not be me! :) anyway sorry for the noise.. It succeded. I will build qt4 now Aug 14 10:36:58 lindi-: /dev/rtc0 not enabled for wakeup events hm Aug 14 10:38:05 googling shows http://docs.openmoko.org/trac/ticket/2318 Aug 14 10:39:55 zub: ohh... that explains why alarms don't work :-) Aug 14 10:46:25 gena2x: resume on incoming call is reliable for me with old ogsmd Aug 14 10:46:41 zub: normal Aug 14 10:47:24 hm? Aug 14 10:47:53 zub: it still works Aug 14 10:48:08 hmm... resume by alarm? Aug 14 10:48:08 aha, so it should work even with the message Aug 14 10:48:10 not for me Aug 14 10:48:51 mrmoku: I tested yesterday (ffalarms) and it worked Aug 14 10:49:08 zub: I tested half an hour ago... and it did not Aug 14 10:49:10 strange Aug 14 10:49:12 but ingeneral it's unreliable, one sleep it was enough for me :) Aug 14 10:49:16 let me re-test Aug 14 10:50:46 works (I set up alarm, then suspend by quick settings and wait) Aug 14 10:51:44 I did the same... Aug 14 10:51:47 * mrmoku tries again too Aug 14 10:52:10 anyway in general things seem quite unreliable. I was thinking if it was possible to create some automated test to test as many things as possible, and occasionally run it with current shr-u (or other distros ;) Aug 14 10:53:19 zub: that would be interesting... but not easy Aug 14 10:53:32 how to select small kernel font back? Aug 14 10:54:58 my kernel suspend backtrace do not fit screen :( Aug 14 10:57:39 mrmoku: second test of ffalarms also passed (this time I let the FR suspend itself) Aug 14 10:58:19 found it... Aug 14 10:58:30 i mean console font size Aug 14 10:58:44 zub: for me it reliable does not work :P Aug 14 10:59:34 then I'm more lucky :) Aug 14 10:59:46 also I have sw from some 2 days ago Aug 14 10:59:49 I'll try with current Aug 14 11:01:16 morning Aug 14 11:02:42 mrmoku: is errno perhaps still set even after adjtimex returning 0? Aug 14 11:03:27 aha, console font size may be changed on-the-fly... Aug 14 11:04:15 mickey|: hehe, I tried that... it is unchanged Aug 14 11:04:19 bummer Aug 14 11:04:22 I loggered errno before and after Aug 14 11:04:27 so is that a bug in eglibc? Aug 14 11:04:39 no idea Aug 14 11:05:02 when calls succeed, errno is normally not changed Aug 14 11:05:09 mickey|: how fast should that adjtime work to adjust a difference of lets say 30mins? Aug 14 11:05:31 zub: thing is adjtime does not work Aug 14 11:06:19 zub: just try... set the time on your FR half an hour off and restart fsotdld Aug 14 11:06:41 you'll never get the correct time again ;) Aug 14 11:06:54 I something like that Aug 14 11:07:20 mickey|: I wonder if we actually need (or even want) adjtime on a phone Aug 14 11:07:21 anyway, so - adjtime, or adjtimex doesn't work? and wht you care about errno? it fails silently? Aug 14 11:07:32 weöö Aug 14 11:07:34 well Aug 14 11:07:37 we do Aug 14 11:07:42 mickey|: does it work even when suspended before the adjusting is finished Aug 14 11:07:46 ? Aug 14 11:08:01 i must confess i have not the slightest idea how it's being done in kernel Aug 14 11:08:18 it's just that everyone recommends not to use hard time setting until you absolutely have to Aug 14 11:08:46 sure it would be nicer... Aug 14 11:09:00 damn. Aug 14 11:09:12 seem my suspend fails in suspending sound... Aug 14 11:09:25 mrmoku: shall we make it optional for now? Aug 14 11:09:31 [fsotdld.sync_time] Aug 14 11:09:37 adjust = 0 Aug 14 11:09:38 ? Aug 14 11:13:16 mickey|: thought about that too :) Aug 14 11:13:39 zub: yep, it fails silently :/ Aug 14 11:14:43 mickey|: I still want to try an old kernel... Aug 14 11:14:54 ok. btw., adjtime is not supposed to cover cases that exceed +-2145 seconds anyway... Aug 14 11:15:10 (on glbc, that is) Aug 14 11:15:19 yep, if it exceeds that falling back to hard setting works fine Aug 14 11:15:37 hmm, ok. i'll make the call optional thn Aug 14 11:15:38 adjtime returns a nice error on that one Aug 14 11:16:43 mickey|: did you ever try adjtimex? maybe that one works better :P Aug 14 11:17:59 mickey|: looks like adjtimex should give us the TIME_ERROR... Aug 14 11:18:56 * mrmoku will try that after lunch Aug 14 11:18:58 bbiab Aug 14 11:20:55 ~bon appetit Aug 14 11:20:56 i heard bon appetit is smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Smaklig m�ltid!. Hyv�� ruokahalua. Bo Proveito Aug 14 11:28:54 freesmartphone.org: 03mickey 07cornucopia * r5bda10d8665d 10/fsotdld/ (conf/default/fsotdld.conf src/plugins/sync_time/plugin.vala): Aug 14 11:28:54 freesmartphone.org: fsotld: sync_time: make the use of adjtime(2) configurable. New config entry is: Aug 14 11:28:54 freesmartphone.org: [fsotdl.sync_time] Aug 14 11:28:54 freesmartphone.org: try_adjtime_before_settime = false Aug 14 11:31:11 yeah. something really wrong with sound on my .34 Aug 14 11:31:32 not i have my suspend-once system back fully functional. Aug 14 11:31:41 s/not/now/ Aug 14 11:31:41 gena2x meant: now i have my suspend-once system back fully functional. Aug 14 11:32:01 so, current_now? Aug 14 11:44:29 lindi-: rtcwake doesn't work for me Aug 14 11:44:37 mrmoku: can you try rtcwake? Aug 14 11:55:03 zub: really? Aug 14 11:55:06 zub: how does it fail? Aug 14 11:55:12 zub: are you using fso? Aug 14 11:55:59 yes, shr-u Aug 14 11:57:09 it tells me:rtcwake: /dev/rtc0 not enabled for wakeup events Aug 14 11:57:13 and odesn't do anything Aug 14 11:57:22 (it exits after the message and phone stays on) Aug 14 11:57:49 I trid with and w/o fsodeviced (although this probably shouldn't have any effect) Aug 14 11:58:13 interestingly ffalarms, which schedules a wakeup via fso, works for me - and doesn't work for mrmoku Aug 14 12:00:24 zub: that is normal Aug 14 12:00:38 zub: just suspend using pm-suspend after that Aug 14 12:00:52 zub: "rtcwake -m no" is not supposed to suspend Aug 14 12:06:50 lindi-: aaaha, that I didn't get Aug 14 12:06:51 ok Aug 14 12:07:13 lindi-: without -m no it does not suspend either for me Aug 14 12:08:30 mrmoku: yes, that is a feature :) Aug 14 12:08:40 lindi-: ok... waiting for wakeup :) Aug 14 12:08:48 (after manually suspending) Aug 14 12:09:11 loves me, loves me not... wakes up, wakes up not... Aug 14 12:09:12 :) Aug 14 12:09:16 * zub also waits Aug 14 12:09:37 * mrmoku still waiting Aug 14 12:10:02 at least the issue is not undecidable :) Aug 14 12:10:13 even though I didn't time the wait Aug 14 12:10:16 zub: btw. do you have /etc/adjtime? Aug 14 12:10:31 when/if the beast weks up, I'll check Aug 14 12:10:36 :P Aug 14 12:10:42 wakes Aug 14 12:10:49 rtcwake -s 30 Aug 14 12:10:55 guess the 30 seconds are over :/ Aug 14 12:11:00 * mrmoku still waiting Aug 14 12:11:05 I did 60, and I think it also expired Aug 14 12:11:25 maybe I should be more exact and time the wait :) Aug 14 12:11:44 yeah Aug 14 12:11:47 * mrmoku fires tzclock Aug 14 12:12:09 no /etc/adjtime here btw Aug 14 12:12:44 ok Aug 14 12:13:17 no, does not wake up Aug 14 12:13:27 * mrmoku tries an old kernel now Aug 14 12:14:13 * gena2x doing 100th reboot Aug 14 12:15:08 ok, no wakup here either :-( Aug 14 12:15:37 (with 2.6.32 - shr-u current default) Aug 14 12:15:54 so the fact that ffalarms work is even more mysterious Aug 14 12:15:59 I mean: works for me :) Aug 14 12:18:34 yeah... magic Aug 14 12:23:49 opensolaris closed :( Aug 14 12:25:16 * gena2x recommends everyone read news while waiting for FR resume Aug 14 12:27:14 * mrmoku already read that before suspending ;) Aug 14 12:27:37 mickeyl: hmm... old kernel gives TIME_ERROR too Aug 14 12:27:53 zub: does not work with old kernel either Aug 14 12:28:39 http://shr-project.org/trac/ticket/1162 Aug 14 12:28:47 I believe this is resolved already Aug 14 12:29:16 my current_now is always huge. Aug 14 12:29:41 zub: does work again=? Aug 14 12:29:55 323437 before suspend Aug 14 12:30:10 and 156562 after :) Aug 14 12:30:13 mrmoku: huh? Aug 14 12:30:36 how to reproduce that thing? Aug 14 12:31:01 i run script with recording current Aug 14 12:31:12 gena2x: I've seen that - that the current_now after suspend dropped :) and another suspend-resume, again dropped, then againa nd it rose Aug 14 12:31:13 and suspended my beast for a while Aug 14 12:31:31 aha. Aug 14 12:31:36 i have only 1 change :( Aug 14 12:31:45 mrmoku: I didn't get < mrmoku> zub: does work again=? Aug 14 12:31:58 somehow can't match it to any context :) Aug 14 12:31:58 zub: #1162 ... root password Aug 14 12:31:59 *chance Aug 14 12:32:16 mrmoku: I thin dos1 (?) fixed it; anyway it currently works IMO Aug 14 12:32:38 mrmoku, did you try building evopedia quith qtx11? :) Aug 14 12:32:40 so it'd be good to close it, to clean up the tickets a bit Aug 14 12:32:58 pespin: no, sorry. not yet :/ Aug 14 12:32:59 mrmoku: how can I fire up the wizard manually? to try... Aug 14 12:33:12 zub: just rm -rf /home/root/.e Aug 14 12:33:17 ok, I'll try Aug 14 12:33:19 and restart enlightenment Aug 14 12:33:28 mrmoku, np :) Aug 14 12:36:23 mrmoku: confirmed, works Aug 14 12:37:07 will U close it? alternatively, I can click it, if the thing lets me Aug 14 12:37:28 lindi-: afair, you may know where were to find patch to keep kernel logs over reboot? seem i need to fix 2nd suspend first :( Aug 14 12:39:06 gena2x: int bts Aug 14 12:39:28 zub: yeah, will close it Aug 14 12:43:06 mrmoku: ok, i committed the optional adjtimex, it's disabled by default, so you have nothing to do but bumping the revision. I'll take a look at adjtimex when I have a chance Aug 14 12:43:19 mickeyl: good, thanks Aug 14 12:43:22 np Aug 14 12:43:32 mickeyl: btw. I tried the old kernel... and it does not work either Aug 14 12:43:37 giving TIME_ERROR too Aug 14 12:43:52 are we sure it did work before? Aug 14 12:44:15 I thought so, because time setting was always working fine for me... but I don't know when you introduced adjtime Aug 14 12:44:37 lindi-: i have no idea about keywords :( Aug 14 12:45:02 mrmoku: i'm not sure either, tbh. chances are i'll never tested that on the device Aug 14 12:45:20 s/i'll/'have/ Aug 14 12:45:20