| villager | can't seem to get python-conic to work | 00:03 |
|---|---|---|
| woglinde | what is python-conic? | 00:04 |
| villager | python bindings for libconic | 00:04 |
| woglinde | and whats libconic good for? | 00:04 |
| *** lcuk has joined #maemo-devel | 00:04 | |
| villager | connecting to the internet and getting proxy settings | 00:05 |
| woglinde | jo re lcuk | 00:05 |
| woglinde | aeh | 00:05 |
| woglinde | you want bypass icd2? | 00:05 |
| villager | conic is the client interface to icd2, isn't it? | 00:05 |
| lcuk | lol woglinde hi im still in #maemo on the proper computer | 00:05 |
| lcuk | errr laptop | 00:06 |
| woglinde | villager oh hm | 00:06 |
| villager | I hoped I wouldn't have to do the messy icd dbus stuff directly instead of using the friendly conic api | 00:07 |
| villager | says here "Applications requesting network connections do not interact directly with ICd2, but instead use the shared libconic library API." | 00:09 |
| woglinde | ah cool | 00:10 |
| woglinde | hm why is python-conic not working? | 00:11 |
| villager | the python-conic examples don't work... never receive signals | 00:11 |
| woglinde | hm | 00:11 |
| woglinde | you could use dbus-monitor I think to look if it sends somehing | 00:11 |
| villager | doesn't seem to | 00:14 |
| *** woglinde_ has joined #maemo-devel | 00:15 | |
| *** JPohlmann has quit IRC | 00:27 | |
| Noeski | Does anyone know how to get the device id for an n900 in code? | 00:27 |
| SpeedEvil | device ID? | 00:32 |
| SpeedEvil | you mean IMSI? | 00:32 |
| *** woglinde has quit IRC | 00:36 | |
| DocScrutinizer51 | darn. how is LP5521 supposed to controll a RGB 3color signal LED and *concurrently* (via engine3) the kbd-backlight?? o.O | 00:36 |
| DocScrutinizer51 | either that's incorrect info in wiki, or I missed something really basic | 00:37 |
| SpeedEvil | it's 3 channel | 00:37 |
| SpeedEvil | I'm unsure though | 00:37 |
| DocScrutinizer51 | R chan, G chan + B cahn | 00:37 |
| DocScrutinizer51 | where's kbd chan?? | 00:38 |
| Noeski | doesn't every device have a unique ID? | 00:38 |
| SpeedEvil | Noeski: ask user to type in serial number. | 00:38 |
| SpeedEvil | It should be possible to read imsi somehow | 00:38 |
| DocScrutinizer51 | IMEI | 00:39 |
| DocScrutinizer51 | not IMSI | 00:39 |
| SpeedEvil | err - yes | 00:39 |
| SpeedEvil | I was confusing myself. | 00:39 |
| Noeski | ahh okay | 00:40 |
| Noeski | thank you | 00:40 |
| DocScrutinizer51 | well I never seen indicator LED to shine Red.But I definitely seen blue, green, yellow, and *white*. You can't do this with a Wchan/color LED | 00:41 |
| V13 | DocScrutinizer51: That's what I'm working at right now | 00:41 |
| SpeedEvil | /sys/class/mmc_host/mmc0/mmc0:bfa2/serial - perpahs | 00:41 |
| V13 | There are three colors and three channels | 00:41 |
| DocScrutinizer51 | so I guess the indicator LED will *need* THREE channels | 00:41 |
| V13 | the three channels are three programmable "channels" | 00:41 |
| V13 | no | 00:41 |
| SpeedEvil | DocScrutinizer51: however - you can do white if you poke the - I think - battery charger chip | 00:41 |
| V13 | you map leds to channes | 00:41 |
| V13 | but channel 3 is the keyboard | 00:41 |
| V13 | so you have 1 and 2 for the leds | 00:42 |
| SpeedEvil | DocScrutinizer51: or maybe not | 00:42 |
| V13 | there is a bitmap to map leds to channels | 00:42 |
| V13 | it works very well as long as you don't map the same led to both channels | 00:42 |
| SpeedEvil | how many hardware LED channels does it have? | 00:42 |
| V13 | I've added a "Low level" section in the wiki. | 00:42 |
| SpeedEvil | not engines | 00:42 |
| V13 | oh.. | 00:42 |
| DocScrutinizer51 | that's mere driver madness then. the chip doewn't support that nonsense | 00:42 |
| V13 | there are three engines | 00:42 |
| DocScrutinizer51 | SpeedEvil: THREE | 00:43 |
| SpeedEvil | that's what Ithought | 00:43 |
| V13 | what is the difference between "leds", channels and engines ? | 00:43 |
| SpeedEvil | the l3/l4 manual may have info | 00:43 |
| V13 | i only understand "leds" and "engines" | 00:43 |
| DocScrutinizer51 | 3 engines 3 drivers 3 chargepumps 3 pins | 00:43 |
| V13 | hm.. | 00:43 |
| V13 | well.. in /sys there are 6 subdors with leds: preffix | 00:44 |
| SpeedEvil | V13: look at the category n900_hardware - look at the hardware hacking page - dl l3/l4 manual from links at bottom | 00:44 |
| V13 | 3 for rgb leds and 6 for keyboard | 00:44 |
| SpeedEvil | one for vibrator | 00:44 |
| V13 | the vibrator is not in lp5523... no ? | 00:45 |
| DocScrutinizer51 | SpeedEvil: also I seem to have learnt there are 4 (6?) independently controlable LEDs for kbd-bl | 00:45 |
| SpeedEvil | There are I think 4 differnet chips controlling LEDs | 00:45 |
| V13 | i believe there are 6 for keyboard | 00:45 |
| SpeedEvil | DocScrutinizer51: yeah - I posted way back a little 'knight rider' effect in awk for the kbd. | 00:45 |
| SpeedEvil | very silly | 00:45 |
| V13 | so, what are you trying to do ? | 00:45 |
| DocScrutinizer51 | how do THAT then with e.g. the one chan of LP5521 sacrificed for kbd? | 00:46 |
| V13 | ok.. for sure there are 6 channels for keyboard | 00:46 |
| V13 | DocScrutinizer51 go to /sys/class/i2c-adapter/i2c-2/2-0032 | 00:46 |
| SpeedEvil | DocScrutinizer51: there is some wackyness | 00:46 |
| V13 | and do: echo "load" > engine3_mode | 00:46 |
| V13 | then do: cat engine3_leds | 00:47 |
| SpeedEvil | DocScrutinizer51: in that something I wasn't sure of turns off the keyboard LEDs | 00:47 |
| V13 | finally: echo "run" > engine3_mode | 00:47 |
| V13 | to restore it | 00:47 |
| V13 | the engine3_leds is a bitmap | 00:47 |
| *** rlinfati has joined #maemo-devel | 00:47 | |
| V13 | there are 9 bits and it maps the 6 (keyboard) channels (?) to the 3rd engine | 00:47 |
| DocScrutinizer51 | V13: yeah. I don't say you're lying. I just say the *hardware* looks different than the driver interface in sys | 00:48 |
| V13 | hmm | 00:48 |
| V13 | I show a picture of 3 chips | 00:49 |
| V13 | somewhere... don't remember where | 00:51 |
| DocScrutinizer51 | I could imagine they actually use *3* LP5521, 1 for indicator and 2 for kbd = 6pcs white LED. and somehow they really managed to mess up with the driver in sys ("err huh? 3 engines. 3 chips? what now?...") | 00:53 |
| V13 | look at this: http://www.national.com/an/AN/AN-1562.pdf | 00:53 |
| V13 | page 6 | 00:53 |
| V13 | but i believe that they work as one | 00:54 |
| V13 | i.e. you can map any led anywhere | 00:54 |
| V13 | but I may be totally wrong | 00:54 |
| V13 | well.. the engines are common | 00:55 |
| V13 | one engine controls 6 leds | 00:55 |
| V13 | (the keyboard engine) | 00:55 |
| V13 | unless this is another chip, it works as one | 00:55 |
| DocScrutinizer51 | nope | 00:56 |
| DocScrutinizer51 | :D | 00:56 |
| V13 | yes | 00:56 |
| V13 | just tested it :) | 00:56 |
| V13 | do this: | 00:56 |
| DocScrutinizer51 | for sure not (my EE hat on) | 00:56 |
| V13 | echo "load" > engine3_mode ; echo 111111111 > engine3_leds ; echo "run" > engine3_mode | 00:56 |
| V13 | and see the led light-up because the keyboard is lighted | 00:57 |
| V13 | but doit while keyboard is open and lighted | 00:57 |
| DocScrutinizer51 | that,s driver madness once again | 00:57 |
| V13 | because mce seems to reset it | 00:57 |
| V13 | oh.. maybe you're write | 00:57 |
| V13 | one sec | 00:58 |
| DocScrutinizer51 | on that p.6 every chip has a different address set | 00:58 |
| V13 | sorry.. I'm not any good in electr. engineering :) | 00:59 |
| DocScrutinizer51 | you can talk to each of those 3 chips individually | 00:59 |
| V13 | strange | 00:59 |
| V13 | the driver seems to act only on one hardware thing | 00:59 |
| DocScrutinizer51 | seems nokia/maemo had exactly same thoughts as you had ;-D | 01:00 |
| V13 | eh? | 01:00 |
| DocScrutinizer51 | as I mentioned above' they confused the 3 chips with the 3 engines per chip | 01:00 |
| V13 | oh | 01:01 |
| DocScrutinizer51 | just kidding. but really? | 01:01 |
| V13 | well.. the driver exposes 3 engines | 01:01 |
| DocScrutinizer51 | may be I'm even right | 01:01 |
| V13 | and handles 9 leds | 01:01 |
| V13 | that's for sure | 01:01 |
| V13 | and somehow, the order is messed | 01:01 |
| V13 | anyway, we know now how it works | 01:02 |
| DocScrutinizer51 | lemme get a beer and study the datasheet once more. maybe I forgot sth as it's quite some time ago I scrutinized that chip for OM | 01:02 |
| V13 | do you have any thoughts on what are safe values for led_current ? | 01:02 |
| SpeedEvil | observe the brightness of the LEDs | 01:03 |
| SpeedEvil | normally - maximum | 01:03 |
| SpeedEvil | don't go over that | 01:03 |
| V13 | ? | 01:03 |
| V13 | using eye you mean ? | 01:03 |
| SpeedEvil | however - they will generally be setup so they can't be damaged | 01:03 |
| SpeedEvil | yes | 01:03 |
| DocScrutinizer51 | SpeedEvil: muhaha | 01:03 |
| V13 | i've found top values of 20 for red | 01:03 |
| DocScrutinizer51 | yep | 01:03 |
| DocScrutinizer51 | lets hope so | 01:03 |
| V13 | and 8 for b/g | 01:03 |
| V13 | using ff makes the led really bright | 01:03 |
| V13 | you hardly can look at it | 01:04 |
| V13 | 0xff | 01:04 |
| DocScrutinizer51 | don't do that! | 01:04 |
| V13 | i know that leds are damages by heat | 01:04 |
| DocScrutinizer51 | might burn your led | 01:04 |
| V13 | (right?) | 01:04 |
| DocScrutinizer51 | also by mere overcurrent | 01:04 |
| V13 | yes.. that's what i'm conserned for | 01:04 |
| DocScrutinizer51 | wait a hour. I'll get my beer and read the ds | 01:05 |
| V13 | well.. i didn't try full brightness for all of them at the same tim e:) | 01:05 |
| V13 | i need to go to sleep... | 01:05 |
| DocScrutinizer51 | 8mll come back later with better answers | 01:05 |
| V13 | to wake in 5h | 01:05 |
| DocScrutinizer51 | eew | 01:05 |
| V13 | sleep(5*86400); | 01:05 |
| V13 | wrong | 01:06 |
| V13 | sleep(5*3600); | 01:06 |
| V13 | thanks for the info&hints | 01:06 |
| V13 | bb && gn | 01:06 |
| DocScrutinizer51 | anyway if thats mA then 8 and 20 make sense | 01:06 |
| *** V13 has quit IRC | 01:06 | |
| DocScrutinizer51 | gn | 01:07 |
| *** kamui__ has joined #maemo-devel | 01:07 | |
| *** Kalessin has joined #maemo-devel | 01:11 | |
| *** lardman has quit IRC | 01:12 | |
| *** DocScrutinizer51 has quit IRC | 01:14 | |
| *** Docscrutemp has joined #maemo-devel | 01:15 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 01:15 | |
| *** woglinde has joined #maemo-devel | 01:20 | |
| *** kamui__ is now known as S_Kamui | 01:21 | |
| *** shinkamui has quit IRC | 01:22 | |
| *** S_Kamui is now known as shinkamui | 01:23 | |
| *** rlinfati has quit IRC | 01:23 | |
| DocScrutinizer51 | k, now it's pretty clear | 01:24 |
| shinkamui | I want to build a new gui app from scratch with C++, I don't know QT or GTK, which kit is recommended | 01:25 |
| shinkamui | and which is easier to work with | 01:25 |
| *** woglinde_ has quit IRC | 01:37 | |
| *** DocScrutinizer51 has quit IRC | 01:45 | |
| *** Docscrutemp has joined #maemo-devel | 01:45 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 01:45 | |
| DocScrutinizer51 | darn roaming | 01:46 |
| DocScrutinizer51 | SpeedEvil: max driver output is 25mA / chan | 01:46 |
| DocScrutinizer51 | looks rather safe for almost all types of LED | 01:46 |
| *** jebba has joined #maemo-devel | 01:47 | |
| woglinde | jo jebba | 01:47 |
| DocScrutinizer51 | also there's separate setup of brightness (0..100) and LED max I. To me it seems natural the driver coders used a sane value for max I and maybe not even crrated a API to change | 01:48 |
| DocScrutinizer51 | woglinde: o/ :-) | 01:48 |
| jebba | hey woglinde | 01:48 |
| DocScrutinizer51 | jebba: hi | 01:50 |
| SpeedEvil | also - modern LEDs tend to be stupidly bright. | 01:52 |
| SpeedEvil | when used as indicators. | 01:52 |
| SpeedEvil | plus - it's got to be visible in the sun | 01:52 |
| SpeedEvil | oh - he left. | 01:52 |
| SpeedEvil | or she | 01:52 |
| DocScrutinizer51 | SpeedEvil: no problem | 01:55 |
| DocScrutinizer51 | I'm here | 01:55 |
| jebba | hey doc | 01:55 |
| DocScrutinizer51 | yes, modern low current LED may well be designed to operate with 1mA and have a max I of 10 | 01:56 |
| SpeedEvil | I was meaning V13 | 01:56 |
| DocScrutinizer51 | and still same brightness as old crap with 20mA or even much more | 01:56 |
| DocScrutinizer51 | SpeedEvil: me too ;-) | 01:57 |
| SpeedEvil | I haven't seen much stuff with uinder 20mA | 01:57 |
| SpeedEvil | hmm - 10% of LEDs at digikey have a test current of 5mA or under | 01:59 |
| SpeedEvil | I suspect that's not Imax though | 01:59 |
| DocScrutinizer51 | nah, that's kinda 'recommended' | 02:00 |
| DocScrutinizer51 | probably also sweet spot for efficiency | 02:01 |
| SpeedEvil | yeah | 02:01 |
| DocScrutinizer51 | (the days of LEDs performing best when pulsed at 30* Imax are gone long time) | 02:01 |
| SpeedEvil | the first 5 I saw are all at 20mA | 02:02 |
| SpeedEvil | yeah | 02:02 |
| SpeedEvil | 20mA max forward | 02:02 |
| SpeedEvil | though specced at 6 | 02:02 |
| SpeedEvil | 5 | 02:02 |
| SpeedEvil | regrettably so. | 02:02 |
| *** rlinfati has joined #maemo-devel | 02:02 | |
| SpeedEvil | all? modern chemistries are most efficient somewhat under max. | 02:02 |
| DocScrutinizer51 | so that's what Imeant. Most will not die instantly at 25mA | 02:03 |
| SpeedEvil | I hadn't seen the ones explicitly specced at 5mA before. | 02:03 |
| DocScrutinizer51 | the pulsing thing was very specific for a particular type of LED and it's physics | 02:03 |
| DocScrutinizer51 | modern designs obsoleted that mostly | 02:04 |
| SpeedEvil | yes, and has lead to a misconception that it was the eye doing it | 02:04 |
| DocScrutinizer51 | s/it's/its | 02:04 |
| SpeedEvil | On a related matter - the backlight pwm freq is annoyingly low. | 02:04 |
| SpeedEvil | at least at night when reading at low illuminations - for me | 02:05 |
| DocScrutinizer51 | LCD PWM? | 02:05 |
| SpeedEvil | yes | 02:05 |
| DocScrutinizer51 | hmm lemme see.... | 02:05 |
| SpeedEvil | I need to see if that can be tweaked | 02:05 |
| SpeedEvil | not up to reading code ATM | 02:05 |
| DocScrutinizer51 | aah yes. some 150..500 | 02:06 |
| SpeedEvil | It looks something like 40Hz or so | 02:06 |
| DocScrutinizer51 | nah a lil more | 02:06 |
| DocScrutinizer51 | I'd guess | 02:06 |
| SpeedEvil | maybe. | 02:06 |
| SpeedEvil | I read ebooks on it - and find it a little unrestful. | 02:06 |
| DocScrutinizer51 | hmm I'm going to continue reading pdf ds now ;-) | 02:07 |
| DocScrutinizer51 | still not found how to speak to R G and B exactly | 02:08 |
| DocScrutinizer51 | thought I remember E dedicated engines | 02:08 |
| SpeedEvil | it's annoyingly complex. | 02:08 |
| DocScrutinizer51 | 3 | 02:08 |
| SpeedEvil | I mean - yes, it avoids waking the CPU | 02:08 |
| DocScrutinizer51 | just right :) | 02:09 |
| SpeedEvil | I'd be tempted to jam in a little micro. | 02:09 |
| DocScrutinizer51 | actually I fell in love with that chip when I learnt about it | 02:09 |
| SpeedEvil | Though that of course raises its own issues. | 02:09 |
| DocScrutinizer51 | duh. you bet | 02:10 |
| DocScrutinizer51 | SpeedEvil: yep was right. 3 engines. Each separately with "reset, load, run, direct" (0x01 OP_MODE) | 02:14 |
| SpeedEvil | :) | 02:15 |
| DocScrutinizer51 | so obviously the API is kinda braindamaed | 02:15 |
| *** rsalveti has quit IRC | 02:18 | |
| DocScrutinizer51 | SpeedEvil: 138 ret |= lp5521_write(client, LP5521_REG_OP_MODE, 0x15); /* 0001 0101 */ | 02:24 |
| DocScrutinizer51 | waaaah | 02:24 |
| DocScrutinizer51 | they switch all three machines to state "load" simultaneously | 02:24 |
| SpeedEvil | I seemed to spot a hell of a lot of dodgy code in the battery code. | 02:25 |
| *** woglinde has quit IRC | 02:25 | |
| SpeedEvil | byt I'm unsure how much of that is from not having read code for a few years | 02:26 |
| SpeedEvil | there are clear boneheads like if(bat_volt_uV>3600)charge_percent=40; | 02:26 |
| SpeedEvil | but that's not serious | 02:26 |
| DocScrutinizer51 | hmm I still try to understand how TF this driver for 5521 is supposed to work | 02:31 |
| *** VDVsx has quit IRC | 02:31 | |
| DocScrutinizer51 | ok there's definitely one bug I spotted yet: there's funny #ifdef LED_CONNECTED_WRONG which is supposed to handle a hw rev where R and B are swapped. Nevertheless they switch R driver to direct bat powered mode unconditionally. For blue LED connected to Red output that should fail | 02:35 |
| *** VDVsx has joined #maemo-devel | 02:37 | |
| DocScrutinizer51 | hmm that driver seems missing a few features. I think I could fix it without breaking API, even rather easily | 02:57 |
| SpeedEvil | :) | 03:02 |
| *** niekt0_ has joined #maemo-devel | 03:28 | |
| *** niekt0 has quit IRC | 03:29 | |
| *** niekt0_ has quit IRC | 03:36 | |
| *** niekt0 has joined #maemo-devel | 03:36 | |
| *** VDVsx has quit IRC | 04:29 | |
| *** BillK has joined #maemo-devel | 04:33 | |
| *** jebba has quit IRC | 04:38 | |
| *** rlinfati has quit IRC | 04:40 | |
| *** jebba has joined #maemo-devel | 04:40 | |
| *** BillK_ has quit IRC | 04:47 | |
| *** jebba has quit IRC | 05:06 | |
| *** SpeedEvil has quit IRC | 05:06 | |
| *** Hukka has quit IRC | 05:06 | |
| *** sr71 has quit IRC | 05:06 | |
| *** BillK has quit IRC | 05:06 | |
| *** plr_ has quit IRC | 05:06 | |
| *** w00t has quit IRC | 05:06 | |
| *** marant has quit IRC | 05:06 | |
| *** ifreq has quit IRC | 05:06 | |
| *** matnel has quit IRC | 05:06 | |
| *** ruskie has quit IRC | 05:06 | |
| *** BillK has joined #maemo-devel | 05:08 | |
| *** plr_ has joined #maemo-devel | 05:08 | |
| *** w00t has joined #maemo-devel | 05:08 | |
| *** marant has joined #maemo-devel | 05:08 | |
| *** ruskie has joined #maemo-devel | 05:08 | |
| *** matnel has joined #maemo-devel | 05:08 | |
| *** ifreq has joined #maemo-devel | 05:08 | |
| *** jebba has joined #maemo-devel | 05:09 | |
| *** SpeedEvil has joined #maemo-devel | 05:09 | |
| *** Hukka has joined #maemo-devel | 05:09 | |
| *** sr71 has joined #maemo-devel | 05:09 | |
| *** lambchop27 has joined #maemo-devel | 05:26 | |
| lambchop27 | Is there any way to use the Google Android app, Star Map, on the N810? thanks | 05:27 |
| *** lambchop27 has quit IRC | 06:23 | |
| *** DocScrutinizer51 has quit IRC | 07:30 | |
| *** Docscrutemp has joined #maemo-devel | 07:30 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 07:30 | |
| *** sleipnir has joined #maemo-devel | 08:07 | |
| *** sri___ has joined #maemo-devel | 08:16 | |
| *** tekojo has joined #maemo-devel | 08:49 | |
| ruskie | jebba, checked out the latest kernel update on the n900 yet? | 09:08 |
| *** NMZ has joined #maemo-devel | 09:25 | |
| *** rlinfati has joined #maemo-devel | 09:26 | |
| *** kamui__ has joined #maemo-devel | 09:26 | |
| *** schasch has joined #maemo-devel | 09:35 | |
| *** sleipnir has quit IRC | 09:37 | |
| *** shinkamui has quit IRC | 09:45 | |
| *** flux has quit IRC | 09:55 | |
| *** flux has joined #maemo-devel | 10:38 | |
| scoopr | damn. 4xmultisampling + n900 high dpi = looks really nice :P | 10:51 |
| flux | opengl? | 11:02 |
| scoopr | yah | 11:02 |
| flux | getting decent FPS? | 11:04 |
| scoopr | just rotating a single model so far, that's runs fine | 11:04 |
| flux | is there 'sync on vblank'-capability? | 11:05 |
| scoopr | haven't yet looked in to it, but I certainly hope so | 11:06 |
| flux | I wonder if anyone's going to release an n900-demo for asm2010 :) | 11:07 |
| *** kodomo_ is now known as kodomo | 11:08 | |
| *** BabelO_ has joined #maemo-devel | 11:14 | |
| BabelO_ | join #maemo | 11:14 |
| lcuk | flux, doesnt that have to be written in asm | 11:35 |
| *** schasch has quit IRC | 11:37 | |
| *** tbf has joined #maemo-devel | 11:45 | |
| flux | lcuk, vblank syncing/ | 11:47 |
| flux | lcuk, that would be provided probably by the X-server, with the assistance of the display driver | 11:48 |
| lcuk | whats that got to do with asm2010 which you mentioned? | 11:48 |
| * lcuk also doesnt see many rendering glitches :) | 11:48 | |
| lcuk | doh! | 11:49 |
| lcuk | i see now above, previous topic! | 11:49 |
| flux | oh, I meant assembly 2010, a large demo party in Finland, definitely only the <4k ones are written in assembly :) | 11:49 |
| lcuk | with the 4k ones can you link tolarger library? | 11:49 |
| lcuk | or it it entirely 4k max | 11:49 |
| flux | using system-installed libraries is ok, AFAIK | 11:50 |
| * lcuk likes the demoscene | 11:50 | |
| * lcuk does inventive things ;) http://www.youtube.com/watch?v=6Z51-vCa0_Q | 11:50 | |
| lcuk | now, about your vsync thing | 11:50 |
| lcuk | f you are running standalone stuff afaik there is, just look @bounce and stuff for example | 11:50 |
| *** hrw|gone is now known as hrw | 11:52 | |
| hrw | morning | 11:52 |
| *** jpetersen has joined #maemo-devel | 11:59 | |
| lcuk | flux, when is compo | 12:13 |
| flux | some time next summer | 12:17 |
| lcuk | this summer dont you mean? | 12:40 |
| *** schasch has joined #maemo-devel | 12:41 | |
| flux | yes | 12:42 |
| *** sri___ has quit IRC | 12:46 | |
| *** andrewgodwin has quit IRC | 12:51 | |
| *** andrewgodwin has joined #maemo-devel | 12:52 | |
| *** VDVsx has joined #maemo-devel | 13:16 | |
| *** jebba has quit IRC | 13:39 | |
| *** JPohlmann has joined #maemo-devel | 13:54 | |
| *** jebba has joined #maemo-devel | 14:00 | |
| *** kamui__ is now known as shinkamui | 14:17 | |
| *** JPohlmann has quit IRC | 15:31 | |
| *** rsalveti has joined #maemo-devel | 15:40 | |
| *** tekojo has quit IRC | 16:06 | |
| *** jebba has quit IRC | 16:08 | |
| *** jebba has joined #maemo-devel | 16:29 | |
| *** DocScrutinizer51 has quit IRC | 16:35 | |
| *** Docscrutemp has joined #maemo-devel | 16:36 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 16:36 | |
| *** NMZ has quit IRC | 16:42 | |
| niekt0 | hi, is there any special reason, why there is no cryptsetup package in maemo? (not even in extra repositories, as far as I check). Or just no one need it? | 17:38 |
| ruskie | niekt0, nobody needed it yet | 17:41 |
| Noeski | is calling a signal by dbus the same as calling a method? | 17:53 |
| jebba | ruskie: ping | 17:57 |
| jebba | the only change to the kernel was some LZO (compression) stuff. It shouldn't affect fbdev | 17:57 |
| *** pillar has quit IRC | 18:04 | |
| *** jpetersen has quit IRC | 18:10 | |
| *** DocScrutinizer51 has quit IRC | 18:44 | |
| *** Docscrutemp has joined #maemo-devel | 18:44 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 18:44 | |
| *** pH5 has joined #maemo-devel | 18:45 | |
| *** DocScrutinizer51 has quit IRC | 18:51 | |
| *** Docscrutemp has joined #maemo-devel | 18:51 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 18:51 | |
| ruskie | jebba, well supposedly HAM makes use of it | 19:00 |
| ruskie | for higher compression on ubifs | 19:01 |
| jebba | ya, looks like it | 19:01 |
| jebba | not just HAM | 19:01 |
| jebba | well, actually i'm not sure what *all* uses it, but i would assume all new files added to NAND. | 19:01 |
| ruskie | well HAM supposedly has a signal for it | 19:01 |
| ruskie | or something | 19:02 |
| jebba | are you running my kernel? You may be able to update before me. I'm on a wifi link and it's pouring today. ;) My laptop has a good enough antenna, but the n900 doesn't so huge packet loss. | 19:03 |
| jebba | I could probably do some USB networking of some sort i suppose. Just havent set that up yet. | 19:03 |
| ruskie | well atm running the one from the update | 19:03 |
| ruskie | but yeah generally I do run yours | 19:03 |
| ruskie | the one without the logo | 19:04 |
| jebba | ah, well, i think the logo is staying. ;) -jebba8 will have logo | 19:06 |
| jebba | i can't believe you don't want tux! | 19:06 |
| jebba | maybe there's a kernel option that you can feed it so it doesnt show on boot even if it's built in. | 19:06 |
| *** hrw is now known as hrw|gone | 19:07 | |
| ruskie | hmm | 19:07 |
| ruskie | and how exactly am I supposed to feed it this? | 19:07 |
| jebba | not sure, that's excercise for you ;) | 19:08 |
| jebba | gotta be there *some* where | 19:09 |
| ruskie | yeah... in the flasher :) | 19:09 |
| ruskie | you can pass custom kernel cmd line params | 19:09 |
| ruskie | well if you have a deb-src for your kernel can always rebuild it myself | 19:10 |
| jebba | ruskie: ok kernel_2.6.28-20094102.6+0m12_armel.deb up there. Building src now | 19:10 |
| ruskie | and of course hoping of not breaking anything | 19:11 |
| jebba | niekt0: cryptsetup would be great. Would likely need other supporting packages. It is something i would really like to spend a weekend on at some point. Basically, if your fone gets stolen now, you are *screwed*. | 19:12 |
| *** JPohlmann has joined #maemo-devel | 19:16 | |
| *** sleipnir has joined #maemo-devel | 19:16 | |
| *** schasch has quit IRC | 19:20 | |
| jebba | ruskie: ok the source is apt-gettable now too | 19:23 |
| *** tbf is now known as tbf|eating | 19:26 | |
| ruskie | hmm | 19:38 |
| *** tbf|eating is now known as tbf | 19:39 | |
| ruskie | did you regenerate your sources and packages lists? | 19:39 |
| jebba | ya | 19:39 |
| ruskie | I can't seem to be able to get it | 19:39 |
| jebba | i'm getting it right now from a total crap connection (GPRS in the mountains) | 19:39 |
| jebba | you trying to get the source or the binary? | 19:40 |
| ruskie | source | 19:40 |
| jebba | kernel_2.6.28-20094102.6+0m12 | 19:40 |
| ruskie | is it a diff url from the normal deb one? | 19:40 |
| jebba | deb-src | 19:40 |
| ruskie | deb-src http://www.freemoe.org/users/jebba unstable main | 19:41 |
| ruskie | apt-get source kernel_2.6.28-20094102.6+0m12 | 19:41 |
| ruskie | E: Unable to find a source package for kernel_2.6.28-20094102.6+0m12 | 19:41 |
| jebba | deb-src http://www.freemoe.org/users/jebba unstable main | 19:41 |
| ruskie | and I did update | 19:41 |
| jebba | just enable that repo only and `apt-get source kernel` | 19:41 |
| jebba | if you really really can't get it for some reason, you can just grab it manually from here: http://www.freemoe.org/users/jebba/dists/unstable/main/source/kernel_2.6.28-20094102.6+0m12/ | 19:43 |
| ruskie | btw is the kernel package now auto installable along with the backup? | 19:43 |
| jebba | the kernel auto installs via apt-get upgrade, but it does *not* backup your old kernel | 19:44 |
| ruskie | meant the modules | 19:44 |
| ruskie | I guess I'll have to do that manually | 19:44 |
| ruskie | ok | 19:45 |
| ruskie | finally have it | 19:45 |
| ruskie | I take it I can just modify a kernel config manually to disable tux and then dpkg-buildpackage it? | 19:46 |
| jebba | ya, take a look here at my scripts | 19:46 |
| jebba | http://www.freemoe.org/users/jebba/scratchbox/kernel/ | 19:46 |
| ruskie | have a lot of extra there | 19:48 |
| jebba | i dont really like how i did the version number here, I should have dropped back down to 0 but oh well. | 19:48 |
| jebba | ruskie: well for 2.6.32 kernels and git kernels, neither of which work yet. (there is no touchscreen driver for the device after 2.6.29) | 19:48 |
| jebba | plus scripts to build just modules based on stock kernel etc | 19:49 |
| jebba | i should add IPv6 too | 19:49 |
| jebba | ....but since i don't use it myself... | 19:49 |
| ruskie | I don't use it as well | 19:51 |
| ruskie | ok | 19:54 |
| ruskie | apparently modified and now building | 19:55 |
| *** kodomo_ has joined #maemo-devel | 19:55 | |
| ruskie | btw jebba got my PM about https apt transport? | 19:58 |
| jebba | ya | 19:58 |
| jebba | but you need more than just that file | 19:58 |
| jebba | though stskeeps has been able to build one recently for mer that works | 19:58 |
| jebba | (if i read him correctly). | 19:59 |
| jebba | they patched up the apt binary funky too | 19:59 |
| ruskie | hmm the git one doesn't work? | 19:59 |
| jebba | ah didnt try that. Git with apt? | 19:59 |
| ruskie | no the apt from git for maemo | 20:00 |
| jebba | ah | 20:00 |
| ruskie | wget "http://maemo.gitorious.org/maemo-af/apt/archive-tarball/master" | 20:00 |
| ruskie | this one | 20:00 |
| jebba | "The archive is currently being generated, try again later" | 20:00 |
| ruskie | yeah try it after a few seconds | 20:01 |
| ruskie | or use a browser | 20:01 |
| jebba | ok, will keep it in mind. dont need it at the moment | 20:01 |
| jebba | do you know anything about packaging python apps? | 20:02 |
| ruskie | not really | 20:02 |
| jebba | I keep getting "ImportError: No module named setuptools" no matter how i kludge it, when trying to build a deb. python-setuptools is there too... | 20:03 |
| jebba | bbiab | 20:03 |
| ruskie | are you using the right python? | 20:03 |
| ruskie | I had a problem with it using the scratchbox one instead of the one I wanted | 20:03 |
| *** tbf has quit IRC | 20:05 | |
| *** rsalveti has quit IRC | 20:10 | |
| *** rsalveti_ has joined #maemo-devel | 20:10 | |
| *** kodomo has quit IRC | 20:12 | |
| *** tbf has joined #maemo-devel | 20:14 | |
| jebba | python2.5 | 20:17 |
| *** DocScrutinizer51 has quit IRC | 20:21 | |
| *** Docscrutemp has joined #maemo-devel | 20:21 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 20:21 | |
| *** mj_ has joined #maemo-devel | 20:35 | |
| *** mj_ has quit IRC | 20:41 | |
| * ruskie hopes he won't break his n900 with the kernel | 20:48 | |
| jebba | heh | 20:49 |
| jebba | did you just rebuild mine without logos? | 20:49 |
| ruskie | up | 20:49 |
| ruskie | just turned off both logo bits | 20:50 |
| jebba | cool | 20:53 |
| ruskie | still building | 20:53 |
| jebba | i'm still downloading over gprs..... | 20:53 |
| ruskie | though it seems to be nearig the end | 20:53 |
| ruskie | ran out of disk space | 20:53 |
| ruskie | so had to move it over to a nfs share | 20:53 |
| jebba | ii dont know how to do it offhand, but if you don't build the -dbg (debug) package it'll probably build a lot faster | 20:54 |
| jebba | i built it in 13m40s | 20:54 |
| jebba | (the binary). And the source package in 5m35s | 20:55 |
| ruskie | btw planing on making an initramfs eventually? | 21:04 |
| jebba | for what? like a rescue one or something? | 21:05 |
| ruskie | yeah | 21:05 |
| ruskie | something that could get used to fixup a system | 21:05 |
| jebba | boot into Mer ;) | 21:06 |
| ruskie | needs to be before bootmenu :) | 21:06 |
| ruskie | since iirc that just does a pivot_root after the kernel gets to init | 21:06 |
| ruskie | or something along those lines | 21:07 |
| ruskie | an initramfs that waits for a second or two for you to hit a key to enter it would probably be better | 21:07 |
| jebba | ah, i was thinking something that would work with bootmenu | 21:07 |
| ruskie | naahh initramfs directly in the kernel | 21:08 |
| jebba | ah gotcha. hmm. | 21:08 |
| jebba | honestly, it's not on my shortlist unless some day i really need it! | 21:09 |
| ruskie | I'll probably need to look into that... but it would need to be built into the kernel not separate | 21:09 |
| *** guido_g has joined #maemo-devel | 21:11 | |
| *** onen|openBmap has joined #maemo-devel | 21:12 | |
| ruskie | ok built | 21:13 |
| ruskie | hmm time for a break repo for me | 21:15 |
| ruskie | to put stuff that might break things for other people | 21:15 |
| jebba | :) | 21:15 |
| ruskie | custom kernel, custom theme :) | 21:16 |
| ruskie | I so wish nolo could be replaced by uboot | 21:17 |
| ruskie | hmm what basic stuff from a regular would be needed to get into a working fbconsole on the device? | 21:18 |
| jebba | just run the getty on tty3 and chvt 3 and you have a console on the device. X paints over it every once in awhile though | 21:20 |
| jebba | got kernel installed | 21:20 |
| ruskie | erm... how do you run it? | 21:20 |
| jebba | rebooting... | 21:20 |
| jebba | look at the sample getty lines in /etc/inittab, then as root, run one of them. Then to switch to it, run `chvt 3` to get console and `chvt 7` to go back to X. Need to have passwords set to log in (i suggest setting root password to something all alpha or you won't be able to log in!) | 21:23 |
| ruskie | seriously??? | 21:23 |
| ruskie | that sucks | 21:23 |
| jebba | well, the keymap in the console is hosed | 21:23 |
| ruskie | it is possible to build a custom keymap into the kernel | 21:24 |
| ruskie | maybe look into that :) | 21:24 |
| ruskie | ok kernel installed | 21:24 |
| jebba | ok, my kernel worked | 21:24 |
| jebba | got -dirty tag though, i think cuz of some quilting | 21:24 |
| ruskie | what's that tag? | 21:25 |
| ruskie | booting | 21:25 |
| ruskie | booted | 21:26 |
| *** kamui__ has joined #maemo-devel | 21:26 | |
| *** pillar has joined #maemo-devel | 21:27 | |
| ruskie | and running | 21:27 |
| *** rsalveti_ has quit IRC | 21:28 | |
| ruskie | thanks | 21:28 |
| *** shinkamui has quit IRC | 21:30 | |
| *** rsalveti_ has joined #maemo-devel | 21:31 | |
| jebba | ya works here too | 21:32 |
| ruskie | seriously need a boot loader we can use on this thing | 21:32 |
| ruskie | I don't want to nearly have a heart attack each time I update my system | 21:32 |
| jebba | the boot_menu thing is ok | 21:36 |
| jebba | though its not that hard to revert if you screw the kernel as long as you have backed up the modules | 21:36 |
| ruskie | true | 21:37 |
| ruskie | but a proper boot manager would be more relaxing | 21:37 |
| ruskie | I mean it's easy if you can boot a known working kernel | 21:38 |
| jebba | ya, certainly. First couple times i was a bit sketched out. But now that i have a working /lib/modules backup and a zImage I can grab, it's not that big of a deal | 21:39 |
| *** DocScrutinizer51 has quit IRC | 21:43 | |
| *** Docscrutemp has joined #maemo-devel | 21:43 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 21:44 | |
| ruskie | https://garage.maemo.org/projects/sharing-cli/ <-- I hope this gets to a usable state sometime soon :) | 21:44 |
| jebba | ya looks cool | 21:44 |
| *** DocScrutinizer51 has quit IRC | 21:45 | |
| *** Docscrutemp has joined #maemo-devel | 21:45 | |
| *** Docscrutemp is now known as DocScrutinizer51 | 21:45 | |
| jebba | any clue of a good common webstats program? I havent done webstats in like 10 years but i'm curious how many folks are hitting the kernel. | 21:45 |
| ruskie | atm looking through the ovi repository | 21:45 |
| ruskie | hmm | 21:45 |
| ruskie | I just scripted something quickly to grab all the ips and uniq them | 21:45 |
| frals | jebba: awstats is pretty decent | 21:45 |
| jebba | grep shows 73 downloads of the kernel.deb (excluding modules/flasher) in the last 8 days. That surprises me. I thought it was going to be like 5 ;) | 21:46 |
| ruskie | lol | 21:46 |
| jebba | more people are using it than burgerspace from extras ;) | 21:46 |
| * jebba installing awstats and hoping it's a zero config thing | 21:47 | |
| frals | most likely not ;) | 21:47 |
| ruskie | grep "repo.codemages.net" /var/log/freya/lighttpd/current | awk '{print $4}' | sort | uniq <-- this is what I use :) | 21:47 |
| ruskie | 37 unique hits... though some are obvious patterns | 21:48 |
| jebba | ya, i do that sometimes, but some real thing running in cron.. then feed that to `host` or someshit too. | 21:49 |
| ruskie | 62.140.137.XXX 120.156.XXX.YYY 77.112.XXX.YYY 79.102.XXX.YYY 80.95.XXX.YYY 91.33.XXX.YYY 92.XXX.YYY.ZZZ | 21:49 |
| jebba | most of the server i've run for so long we just /dev/null'd the logs. | 21:49 |
| ruskie | lol | 21:49 |
| ruskie | all my logs are in tmpfs | 21:49 |
| jebba | ah looks like awstats will just run :) | 21:49 |
| ruskie | depends on what web server you use and how you have logging setup :) | 21:50 |
| jebba | just stock lenny stuff | 21:51 |
| ruskie | ahh most of that crap is pre-configured on debian | 21:51 |
| ruskie | one of the reasons I don't like debian systems | 21:51 |
| ruskie | hm mI wonder if sharing-service-ftp is Free Software or not | 21:52 |
| ruskie | fun... | 21:54 |
| ruskie | Disruptive Publishers have an ovi package named alpine... which is the same as alpine MUA... grumbl | 21:54 |
| jebba | http://www.freemoe.org/cgi-bin/awstats.pl had to set sitedomain and did access.log* for files processing since logrotate has moved a few already. So at every 10 min it gets updated | 21:55 |
| ruskie | ok disabeling ovi repo that the update installed | 21:55 |
| ruskie | not impressed | 21:56 |
| jebba | heh | 21:56 |
| jebba | i went to their website and it was like wallpaper of half naked women. I thought nokia would at least be tasteful (being finns and all) | 21:56 |
| ruskie | :) | 21:57 |
| ruskie | optimist :) | 21:57 |
| ruskie | and disabled ovi | 21:58 |
| *** hrw|gone is now known as hrw | 21:59 | |
| ruskie | http://talk.maemo.org/showthread.php?t=40098 <-- youch | 22:26 |
| SpeedEvil | jebba: I think most finnish women are naked at least some of the time. | 22:35 |
| jebba | http://www.freemoe.org/cgi-bin/anlgform.cgi well here's some crappy stats. | 22:37 |
| jebba | i still dont have awstats going but a bit too tired to figure it otu | 22:37 |
| jebba | http://www.freemoe.org/cgi-bin/awstats.pl | 22:37 |
| *** juke_ has joined #maemo-devel | 22:44 | |
| juke_ | hello | 22:45 |
| juke_ | how can i display a webpage or a text file in a desktop widget ? | 22:45 |
| *** hrw is now known as hrw|gone | 22:50 | |
| *** woglinde has joined #maemo-devel | 22:56 | |
| *** tbf has quit IRC | 22:57 | |
| *** sleipnir has quit IRC | 23:02 | |
| *** JPohlmann has quit IRC | 23:18 | |
| *** onen|openBmap has quit IRC | 23:32 | |
| *** asj_ has joined #maemo-devel | 23:42 | |
| *** flux has quit IRC | 23:55 | |
| *** juke__ has joined #maemo-devel | 23:59 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!