IRC log of #maemo for Sunday, 2015-08-30

*** erlehmann has quit IRC00:00
SiceloVajb_: hald-addon-bme? in connection with usb-mode issue?00:12
Vajb_Pali told me that hald-addon-bme is responsible for light00:13
Vajb_and actually i managed to turn off the blue led by issuing "kill hald-addon-bme, stop mce, start mce, stop mce, start mce" , but Pali told me it is not goos to go and adviced to reboot00:15
Vajb_goos=good*00:15
Siceloi guess dsme won't like that stop mce too much00:16
Vajb_so Sicelo it's up to you now. Compile hald-addon-bme with debug symbols and see what prints /tmp/hald-addon-bme.log IF i recall all that correctly.00:17
Vajb_well if you do too much start and stop mce it eventually refuses to start mce for awhile00:18
*** Vajb_ is now known as Vajb00:18
Siceloyes, dsme begins to think something's wrong with mce .. if you keep forcing it, dsme will simply reboot device anyway00:19
Siceloi'll have to find time for compiling .. haven't booted my scratchbox in a long time00:20
VajbSicelo: be sure to ask confirmed instructions from Pali. My explanation can't be trusted that much :)00:24
*** animist has quit IRC00:35
*** animist has joined #maemo00:36
DocScrutinizer05stop mce is safe. kill mce isn't01:07
* DocScrutinizer05 thinks hald-addon-bme shouldn't mess with notifier light01:08
bencohstart/stop/start/stop/start/stop doesn't seem to be safe either01:08
DocScrutinizer05hmm?01:08
bencohI had n900 reboot a few times when doing this01:09
DocScrutinizer05I dunno (to lazy to look up) what `stop mce` or `start nmce` does. Could check in upstart event files. But I'd guess it actually *tells* dsme to stop the mce process, so dsme should be happy with that, no matter how often you do it01:10
bencohunless there is a silly timer somewhere or a race (or both)01:10
Vajbif you do stop-start cycle too many times in certain time period. Mce refuses to start for few minutes.01:11
Vajbbut no forced reboot for me today.01:12
*** Oksana has quit IRC01:14
*** Oksana has joined #maemo01:16
DocScrutinizer05http://paste.opensuse.org/35983988  http://paste.opensuse.org/4352405601:19
DocScrutinizer05if there's a problem with repeatedly doing `start mce` && `stop mce` then that's a bug in dsme01:20
DocScrutinizer05Vajb: how does "refuses to start" manifest?01:24
*** Roth has quit IRC01:24
DocScrutinizer05I mean, what's the error message?01:25
DocScrutinizer05or does upstart return without error msg and yet there's no mce process in process list, after `start mce` ?01:26
DocScrutinizer05IroN900:~# initctl status mce01:29
DocScrutinizer05mce (start) running01:29
DocScrutinizer05http://paste.opensuse.org/4558602001:29
DocScrutinizer05IroN900:~# status mce01:30
DocScrutinizer05mce (start) running01:30
*** Oxyd76 has quit IRC01:33
DocScrutinizer05ooh, and device forgot it's connected to fastcharger, after `stop mce && sleep 5 && start mce` (or my USB starts getting old, seen problems with it before)01:34
DocScrutinizer05DANG! could somebody please fix dsme(tool)? http://paste.opensuse.org/7191077501:38
DocScrutinizer05obviously somebody had a massive hangover, not resetting dsme's internal restart counter on killing a process with -k01:40
DocScrutinizer05(hangover while coding)01:41
DocScrutinizer05*sigh*01:41
bencoh"not resetting dsme's internal restart counter on killing a process with -k" so that's a timer/counter issue ?01:42
DocScrutinizer05that's a bug in dsme01:44
DocScrutinizer05it increments internal restart-counters on each start, but forgets to reset them on killing the same job01:44
bencohmeh :))01:44
DocScrutinizer05formally this is in line with the specs, but it's a highly unexpected and nonsensical behavior01:45
DocScrutinizer05actually the respawn-counter should count process aborts that are NOT initiated by dsme. And it probably should even reset the internal counter on any start command for a process01:48
DocScrutinizer05after all the whole thing is meant to keep processes alive and worst case automatically restart them. Not block user from manually restarting them01:49
DocScrutinizer05Pali: ^^^01:50
DocScrutinizer05~seen freemangordon01:50
infobotfreemangordon <~ivo@46.249.74.23> was last seen on IRC in channel #maemo, 13d 11h 22m 31s ago, saying: 'spongie: nice talk, but I have to leave. still, try to resurrect your n900 with booting rescueos on it :)'.01:50
Palido you mean hald-addon-bme?01:50
DocScrutinizer05no, I meant dsme which is borked01:51
PaliHAL is broken by design... we cannot fix this *addon*s problems01:51
PaliDocScrutinizer05: that paste log is not from dsme but from upstart01:52
DocScrutinizer05(HAL) well, I wondered whether hald-addon-bme is the right process to send notification dbus msgs to mce01:52
DocScrutinizer05yes, I know that01:52
Paliand it is problem in upstart01:52
DocScrutinizer05however http://paste.opensuse.org/3598398801:52
Pali"respawning too fast, stopped"01:52
DocScrutinizer05aha01:52
Palino idea how to fix that maemonized upstart01:53
DocScrutinizer05then upstart is *really* idiotic program01:53
Palithis is maemo upstart01:53
Palinormal ubuntu upstart do not do that01:53
DocScrutinizer05I still guess it's a dsme message01:53
Paliyears ago I tried to fix that problem, but no chance...01:53
Palino this is upstart for sure, I know that!01:53
DocScrutinizer05hmm, I could try. but not now, about to get that beer that's waiting for me out there :-)01:54
Palidsme autorespawn works fine01:54
DocScrutinizer05anyway, when you say you know, I won't bother to check. You know your shit. However it's a pity we can't fix it01:55
Palifix could be possible, but that means to read and understand that old maemonized upstart code01:55
DocScrutinizer05what did they do to maemo upstart?01:55
Paliwhich is not ubuntu upstart01:56
DocScrutinizer05sucks01:56
*** futpib has quit IRC01:56
PaliI do not want to touch init process without 100% knowledge of all changes01:57
Palithis is high risk01:57
Palibroken pid 1 means non working system01:57
Paliwithout way to recover01:57
keriolet's just move to systemd01:58
kerio:>01:58
DocScrutinizer05yes I know :-) However I guess PID1 *is* broken. Look at http://paste.opensuse.org/98152157, says it all already. I feel an urge to bash somebody01:59
Palileave this channel01:59
DocScrutinizer05>>Alternatively if --by-id is given...<< WTF?! where is this listed?02:02
DocScrutinizer05this prolly is >>  -i, --id                    arguments are job ids, instead of names<<, NOT >>--by-id<<, suckers!!02:03
DocScrutinizer05IroN900:~# initctl --version02:05
DocScrutinizer05initctl (upstart 0.3.8)02:05
DocScrutinizer05Copyright (C) 2007 Canonical Ltd.02:05
DocScrutinizer05no word about maemo or Nokia02:05
DocScrutinizer05whatever, bbl. cya02:06
DocScrutinizer05Vajb: you could stop mce once with `stop mce`, then start it with  `/usr/sbin/dsmetool -n -1 -t "/sbin/mce --force-syslog"` and stop with `/usr/sbin/dsmetool -k "/sbin/mce --force-syslog" `02:08
*** zGrr has quit IRC02:22
DocScrutinizer05stuff like >>/usr/bin/tonegend -s cept -b 100 -r 20 -D module-stream-restore.id=x-maemo-key-pressed -I media.role=phone,module-stream-restore.id=sink-input-by-media-role:phone -N media.role=phone,mod<< makes me shudder02:27
*** animist has quit IRC02:35
*** florian has quit IRC02:36
*** animist has joined #maemo02:36
*** xorly has quit IRC02:40
*** jon_y has quit IRC02:40
*** freemangordon has joined #maemo02:49
*** Pali has quit IRC03:00
*** vectis has quit IRC03:23
*** erlehmann has joined #maemo03:36
*** norly has quit IRC03:43
*** Smily has quit IRC03:45
*** Smily has joined #maemo03:45
*** sparetire_ has quit IRC03:50
*** jon_y has joined #maemo03:55
DocScrutinizer05particularly since:04:16
DocScrutinizer05IroN900:~# tonegend --help04:16
DocScrutinizer05usage: tonegend [-h] [-d] [-u username] [-s {cept | ansi | japan}] [-b buflen_in_ms] [-r min_req_time_in_ms] [-i] [-8] [-S] [--tag-dtmf tags] [--tag-indicator tags] [--tag-notif tags] [--volume-dtmf volume] [--volume-indicator volume] [--volume-notif volume]04:16
DocScrutinizer05thankyou for missing -D and -t and -N04:17
DocScrutinizer05s/-t/-I/04:18
infobotDocScrutinizer05 meant: thankyou for missing -D and -I and -N04:18
*** FlameReaper has joined #maemo04:20
*** LauRoman has quit IRC04:20
*** LauRoman has joined #maemo04:21
*** Kilroo has quit IRC04:21
*** Humpelst1lzchen has joined #maemo04:41
*** Humpelstilzchen has quit IRC04:43
*** eMHa__ has joined #maemo04:46
*** eMHa_ has quit IRC04:50
*** LauRoman has quit IRC05:10
MaxdamantusHmm .. what causes the VDQ to reset to 0 shortly before EDV1 goes to 1?05:46
Maxdamantushttps://gist.github.com/Maxdamantus/702809177892347cb77805:48
Maxdamantus(that isn't actually the original bq27200.sh script—I modified it to read the registers from /sys instead of using i2c directly)05:49
*** RedW has quit IRC06:01
*** RedW has joined #maemo06:03
Maxdamantusalso, regarding upstart/systemd/pid1 .. I've just been setting my laptop up without such an init system.06:07
Maxdamantusit just has an initramfs that mounts a filesystem and calls a second init script from there, which starts some gettys, udev, syslog, etc06:09
Maxdamantusand the root filesystem is actually just directly from that initramfs.06:09
*** FlameReaper has quit IRC06:48
*** obsed has joined #maemo07:18
*** githogori has quit IRC07:51
*** githogori has joined #maemo07:57
*** Kilroo has joined #maemo08:00
*** githogori has quit IRC08:05
*** githogori has joined #maemo08:08
*** protem has joined #maemo08:36
*** dhbiker has joined #maemo08:38
*** KotCzarny has quit IRC08:53
*** KotCzarny has joined #maemo08:57
VajbDocScrutinizer05: it says something about mce respawning too fast and that process is halted. Also i read mention about that in maemo wiki. https://wiki.maemo.org/LED_patterns#Low_Level <- dont care about low level, but look below last code box. Line starting "Please note:..."09:54
Vajband by reading your lines I can probably say that you read it already :)09:56
*** chadi_ has joined #maemo10:27
*** chadi has quit IRC10:27
*** chadi_ is now known as chadi10:27
*** chadi has joined #maemo10:27
*** frals_ has joined #maemo10:28
*** frals has quit IRC10:28
*** erlehmann has quit IRC10:29
*** Milhouse has quit IRC10:29
*** fortytwo has quit IRC10:29
*** fortytwo_ has joined #maemo10:30
*** fortytwo_ is now known as fortytwo10:30
*** erlehmann has joined #maemo10:30
*** Milhouse has joined #maemo10:44
*** protem has quit IRC11:13
*** vectis has joined #maemo11:16
*** xorly has joined #maemo11:38
*** Pali has joined #maemo11:56
*** LauRoman has joined #maemo12:18
*** FlameReaper has joined #maemo12:24
*** animist has quit IRC12:35
*** animist has joined #maemo12:36
*** Oksanaa has joined #maemo12:48
*** sparetire_ has joined #maemo12:52
DocScrutinizer05https://wiki.maemo.org/index.php?title=LED_patterns&diff=prev&oldid=27181  :-)13:27
*** ccnnjj has quit IRC13:35
*** lxp has joined #maemo13:36
*** Oksanaa has quit IRC13:47
*** shentey has joined #maemo14:10
*** LauRoman has quit IRC14:17
Ras_OlderHas anyone been able to get Pixie WPS working on N900? I grabbed the git clone but make spits errors to my terminal.. Just wondering if I'm wasting my time trying to get this to working condition :D14:29
KotCzarnythat errors?14:29
Ras_OlderYep just a sec I'll put that to pastebin14:31
Ras_Olderhttp://pastebin.com/7ByQXfD7 thats what I'm getting currently and the pixie wps git clone is from https://github.com/wiire/pixiewps14:32
KotCzarnyhmm14:33
KotCzarnylooks like error in system headers?14:33
KotCzarnyalso, looks like you are trying to compile on device14:33
Ras_Olderyep14:33
KotCzarnyhave you tried scratchbox build?14:33
Ras_Oldernah14:33
KotCzarnymy sb is off at the moment so i cant check14:34
KotCzarnywhere did your /usr/include come from?14:34
KotCzarnyor /usr/include/asm specifically14:34
Ras_Olderpixiewps.c has include for asm/byteorder.h where that comes from14:36
*** LauRoman|Alt has joined #maemo14:36
KotCzarnythats what i've asked you14:36
KotCzarnyyou can try fixing it yourself, or trying different file14:39
*** erlehmann has quit IRC14:40
Ras_OlderYep looking at the source codes right now. Dunno if my programming-fu is strong enough but meh I'll give it a go and see what breaks :'D14:40
*** erlehmann has joined #maemo14:41
KotCzarnymake a backup ;)14:41
Ras_OlderYeaaaah safe to say that if this baby takes a dive because I did a mistake I'm just going to reflash it and start from scratch. It has lots of old stuff and project in it that I've haven't been using in couple years14:42
Ras_Older*mumbles* pointers... why does it have to have pointers...14:48
Ras_OlderI really hated those while at school studying programming :D14:48
KotCzarnythey are kind of shortcuts14:54
KotCzarnyand sometimes you can meet mugger in the back alley14:54
Ras_Older:D true14:55
KotCzarnythey are convenient if used cautiously14:55
Ras_OlderSumma summarum this problem is happening because my gcc is too old on this phone. This is why you mentioned the SB?14:57
KotCzarnyno, because you might have copied wrong files14:58
KotCzarnyand /usr/include/asm/ is system area, so most likely you just copied wrong version14:59
*** futpib has joined #maemo15:00
KotCzarnydont remember which package maintains them, i suspect either glibc, kernel or gcc15:01
Wizzuplinux headers iirc15:02
*** Oxyd76 has joined #maemo15:12
*** obsed has quit IRC15:20
*** Oxyd76 has quit IRC15:26
*** ccnnjj has joined #maemo15:32
*** Oxyd76 has joined #maemo15:37
*** xorly has quit IRC15:40
*** Oxyd76 has quit IRC15:44
DocScrutinizer05Ras_Older: building on device is a pretty poor (or geeky) idea15:46
*** Oxyd76 has joined #maemo15:47
KotCzarnydoc, but i must admit, n900 is one of the very few phones that can produce it's own binaries15:48
DocScrutinizer05yes, it _can_ but it's generally cumbersome to set up a working environment. And once it works, it takes ages15:48
KotCzarnysure, but still15:49
DocScrutinizer05well, N900 is a plain linux device :-)15:49
KotCzarnyand thats why it's so unique15:49
DocScrutinizer05with nasty closed blobs but still15:49
DocScrutinizer05~sb15:51
infobotsomebody said scratchbox was a cross-compiling system that uses binfmt_misc, rpc calls, and an nfs mount to make a cross-build appear to be 100% native, and is found at http://www.scratchbox.org/, hosted by maemo now. Also at http://maemo.merlin1991.at/files/SB15:51
KotCzarnyi wonder if neo900 project would help replacing some of those closed packages15:51
DocScrutinizer05neo900 project supports fptf15:51
DocScrutinizer05~fptf15:51
infobotwell, fptf is the Fremantle Porting Task Force, see http://talk.maemo.org/showthread.php?t=9130815:51
*** trumee_ has joined #maemo15:51
*** trumee has quit IRC15:51
*** trumee_ is now known as trumee15:51
KotCzarnyi meant the other way, kind of backporting some hw specific packages15:52
*** erlehmann has joined #maemo15:52
DocScrutinizer05Neo900 project itself won't engage in sw development (except for generic drivers for hardware that's not (yet) found in mainline kernel)15:52
DocScrutinizer05"we" don't have nagic knowledge of any kind, to do any such backporting or RE of closed blobs15:53
DocScrutinizer05magic even15:53
KotCzarnynagic is a fine word15:54
DocScrutinizer05we just guarantee there will be a driver available for all hardware except where explicitly noted (PVR, peripherals firmware blobs)15:55
DocScrutinizer05a FOSS driver15:55
DocScrutinizer05for PVR there are non-FOSS drivers straight from TI afaik15:56
KotCzarnypersonal video recorder?15:56
DocScrutinizer05possibly violent revenge15:56
DocScrutinizer05the GPU in our SoC15:57
WizzupKotCzarny: PowerVR15:58
DocScrutinizer05http://omappedia.org/wiki/Graphics#SGX_software_components15:58
DocScrutinizer05http://listas.gnu.org.ve/pipermail/powervr-devel/2012-September/000036.html15:58
*** xorly has joined #maemo16:08
keriorip obsd compatibility16:19
*** florian has joined #maemo16:19
*** vectis has quit IRC16:21
*** florian has quit IRC16:24
bencohkrkr16:44
*** florian has joined #maemo16:47
*** ssvb has quit IRC16:51
*** norly has joined #maemo16:56
*** florian has quit IRC17:04
*** futpib has quit IRC17:05
*** futpib has joined #maemo17:05
*** erlehmann has quit IRC17:22
*** futpib has quit IRC17:39
*** trumee has quit IRC17:54
*** trumee has joined #maemo17:56
*** LauRoman|Alt has quit IRC17:56
*** LauRoman|Alt has joined #maemo17:59
*** trumee has quit IRC18:00
*** ssvb has joined #maemo18:02
*** trumee has joined #maemo18:12
*** futpib has joined #maemo18:12
*** animist has quit IRC18:41
*** ccnnjj has quit IRC18:43
DocScrutinizer05hm?18:46
*** stryngs has quit IRC18:51
*** kerio has quit IRC18:57
*** kerio has joined #maemo18:57
*** stryngs has joined #maemo19:04
*** shentey has quit IRC19:05
*** darkschneider has quit IRC19:07
*** animist has joined #maemo19:07
*** darkschneider has joined #maemo19:08
*** shentey has joined #maemo19:10
*** shentey has quit IRC19:17
*** LauRoman has joined #maemo19:21
*** LauRoman has quit IRC19:23
*** LauRoman has joined #maemo19:23
*** LauRoman|Alt has quit IRC19:25
*** LauRoman has quit IRC19:26
*** Oxyd76 has quit IRC19:41
*** florian has joined #maemo20:07
*** florian has quit IRC20:26
*** ccnnjj has joined #maemo20:43
Vajbis anyone using dropn900? And how come fixed version isn't on repos?21:00
Vajbhttp://talk.maemo.org/showpost.php?p=1340431&postcount=44321:03
*** Oksana has quit IRC21:07
*** Oksana has joined #maemo21:10
*** vectis has joined #maemo21:18
*** ccnnjj has quit IRC21:19
Siceloi am using DropN90021:32
Vajbinstalled via those instructions?21:36
Siceloi believe so .. haven't clicked your link yet21:36
*** FReaper has joined #maemo21:36
Vajb:p21:36
Sicelonot in repos because the maintainer request was not responded to21:36
Siceloafaik21:37
*** FlameReaper has quit IRC21:40
Vajbyes i read explanation from last page, but my enter finger here was faster21:40
*** FReaper has quit IRC22:08
*** protem has joined #maemo22:34
*** robink_ is now known as robink22:37
totalizatormy holidays are over, I'm back home and it's not a bit cooler Q__Q22:42
totalizatorbut hey, I heve met a new friend http://imgur.com/a/4AfkI22:43
*** florian has joined #maemo22:49
MaxdamantusPsh. There's a video of the guy who ported Plan 9 to the Raspberry Pi compiling the kernel on-device in a couple of minutes.22:59
KotCzarny:)23:00
KotCzarnyjealous?23:01
MaxdamantusI think he says there it takes about 40 minutes to compile the entire OS.23:01
Wizzupplan9 or 9front23:01
Wizzupbecause 9front already works on the pi23:01
MaxdamantusPlan 9.23:01
WizzupWell, you seem to be certain23:01
MaxdamantusPlan 9 also works, because of this guy.23:01
MaxdamantusI don't think it was 9front.23:02
WizzupDon't see a reason to use plan9 - it is not being developed afaik23:02
Wizzup9front is23:02
Maxdamantusin any case 9front probably also works because of the same code contributed by the same guy.23:02
WizzupMaxdamantus: you're right: https://bitbucket.org/mischief/plan9front/commits/1b61f86d62e94506a938ceb3dc684189#chg-sys/src/9/rpi/READMEhttps://bitbucket.org/mischief/plan9front/commits/1b61f86d62e94506a938ceb3dc684189#chg-sys/src/9/rpi/README23:08
Wizzupbleh, double link23:08
*** dhbiker has quit IRC23:20
*** SpeedEvil has quit IRC23:27
Maxdamantusit doesn't really make sense to port 9front. You can write fundamental code like that for Plan 9 and have the 9front/9atom/whatever distributions copy it in.23:27
*** SpeedEvil has joined #maemo23:28
*** APic has quit IRC23:43
*** norly has quit IRC23:48
*** norly has joined #maemo23:48
*** SpeedEvil has quit IRC23:52
*** SpeedEvil has joined #maemo23:53

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!