**** BEGIN LOGGING AT Sun Jul 19 02:59:58 2015 Jul 19 03:41:56 Kotczarny: couldn't resist...put headphones in the port and ..yeah just ran gtkgep on one of my n800's ...yeah I see what you mean...it is quasi-almost-totally-usable just need to tweak things..also..due to the n8x0 screen size... text for the most part... is almost legible.. Jul 19 03:42:12 Frankengep Jul 19 06:11:24 bencoh: do you own a n8x0 ? Jul 19 06:12:20 ah yes pop-up effects-plugin windows when reduced ...do not reduce..they are killed..output doesn't seem impeded though.. Jul 19 06:12:37 endo, they are not killed Jul 19 06:12:42 they are not-hildonized Jul 19 06:12:50 ie. they are there, just not on the list Jul 19 06:13:34 long time ago i wrote a little app to show all the windows on the n8x0 (visible and minimized) Jul 19 06:13:54 just for those non-hildonized experiments Jul 19 06:15:07 http://talk.maemo.org/showthread.php?t=18542 Jul 19 06:16:26 ah..non-hildonized...I could see rather ..hear...that the plugins were working ..it was odd/awkward though to lose the input windows.. Jul 19 06:17:12 hehe, oscp core can also play without being visible Jul 19 06:18:11 excellent little program...I'll be using it for sure. Jul 19 06:21:06 well..tomorrow is a new day of puttering ...lotta fun stuff to do now..thanks to Kot and Android 808 and all the other good folks along the real-time road Jul 19 06:21:19 no worries Jul 19 06:22:59 http://www.gossamer-threads.com/lists/maemo/users/33837 Jul 19 06:23:01 also this Jul 19 06:24:47 Now I ordered a whack of adapters, a couple of nice clamp mounts and grips.. and just gotta wait for them to arrive..pick-up,plug-in,amp-out...soooo much lighter than one of my convertible tab/pc's even...can't wait Jul 19 06:25:07 did you order a battery too? Jul 19 06:25:10 :> Jul 19 06:25:36 you know, one of those powerbanks with usb output Jul 19 06:25:49 so you can power all those piggybacks Jul 19 06:26:23 and then you can even throw powered speaker to the mix Jul 19 06:31:00 hahaha...well the pick-up is miniscule..the amps are power-corded proper amps..so shouldn't be much of a drain...if I want to go it on onboard n8x0 power I do have a couple sets of tiny speakers that run off device...and a couple others that run off usb and one set that run off rechargables....got em all somewhere around the shop Jul 19 06:31:31 mind you, you only have ~100-200mA available Jul 19 06:32:16 so while headphones are feasible, speaker could be draining too much Jul 19 06:34:50 yeah it ain't alot running on device and it prolly will drain at a goodly rate off device..It'll be good to test..but yeah on the norm it'll either be personal at late hours with headphones or ...outting to externally powered proper amps during the day..got a couple fenders and marshalls ...nice and beastly...should do the trick Jul 19 06:36:34 and i think you were going to sleep :P Jul 19 06:37:22 yeah..that was before I got introduced to maemo.org :D Jul 19 06:38:06 Sleepless years later ..now I'm getting less on #maemo Jul 19 06:38:14 he he Jul 19 06:38:49 i stopped having sleep deprivation, it's usuall not worth it, unless you're on the roll Jul 19 06:40:42 This is my brain on maemo...This is my maemo-brain on gtkgep...This is my maemo-brain-on-gtkgep with a harp (instead of a guitar) https://www.youtube.com/watch?v=hH9_y91ihTo Jul 19 06:41:14 Don't do gtkgep kids Jul 19 06:41:46 electric guitars rock Jul 19 06:43:19 Chong is the craziest lectric-funny-mofo there is. I'm continually amused and impressed by him... Jul 19 06:43:28 well Jul 19 06:43:59 some guys have skills Jul 19 06:44:08 https://www.youtube.com/watch?v=w14KBCAFoLE Jul 19 06:45:36 I'll watch that then... I'm hittin' the snort-fort ...need sleep...back with joe by 8am... Jul 19 06:45:47 :) Jul 19 07:25:28 KotCzarny: no I dont Jul 19 07:26:01 pity :/ do you know anyone with n8x0 and some dsp programming knowledge? Jul 19 07:28:34 no freaking clue Jul 19 10:53:38 ~seen jonwil Jul 19 10:53:51 where is infobot?? Jul 19 10:54:21 DocScrutinizer05: we are lost Jul 19 11:22:12 Pali: only a click on reboot-icon :-) Jul 19 11:22:58 DocScrutinizer05: now I have 58 days uptime Jul 19 11:23:01 (on my N900) Jul 19 11:23:19 not YOU, I for infobot :-D Jul 19 11:23:36 this can be good for statistics of maemo/hildon applications Jul 19 11:24:04 do you know how to read how much swap space is using application with specific PID? Jul 19 11:24:41 t900:~# uptime Jul 19 11:24:43 13:51:25 up 416 days, 1:57, load average: 0.00, 0.00, 0.00 Jul 19 11:24:48 ;-) Jul 19 11:25:08 so it is time to find memory eaters :D Jul 19 11:25:27 (swap space) I usually use (h)top for this. There's a way via /proc for sure Jul 19 11:25:30 determinate which daemons consume lot of memory Jul 19 11:25:30 http://stackoverflow.com/questions/479953/how-to-find-out-which-processes-are-swapping-in-linux Jul 19 11:25:43 and so which needs to be fixed Jul 19 11:25:45 scroll to lototux answer Jul 19 11:26:00 t900:~# free Jul 19 11:26:02 total used free shared buffers Jul 19 11:26:03 Mem: 245404 240300 5104 0 604 Jul 19 11:26:05 Swap: 786424 164192 622232 Jul 19 11:26:06 Total: 1031828 404492 627336 Jul 19 11:27:29 infobot: +uptime Jul 19 11:27:29 - Uptime for purl - Jul 19 11:27:29 Now: 4m 48s running infobot 1.5.4 (SVN) -- linux Jul 19 11:27:30 1: 59d 8h 41m 19s running infobot 1.5.4 (SVN) -- linux, ended Sun Nov 14 18:39:57 2010 Jul 19 11:27:30 2: 57d 3h 9m 23s running infobot 1.5.4 (SVN) -- linux, ended Fri Jun 26 20:39:27 2009 Jul 19 11:27:31 3: 36d 20h 47m 14s running infobot 1.5.4 (SVN) -- linux, ended Tue Aug 4 17:38:59 2009 Jul 19 11:27:40 KotCzarny: does not work Jul 19 11:27:58 could be for newer kernel, you get the idea and paths tho Jul 19 11:28:03 KotCzarny: there is no VmSwap line in /proc/PID/status file Jul 19 11:28:20 or some kernel debug compile options Jul 19 11:28:42 we need to read it without rebooting :-) Jul 19 11:28:45 DocScrutinizer05: Glad You are back. You really rock. ☺ Jul 19 11:28:58 ~seen jonwil Jul 19 11:28:59 jonwil <~jonwil@27-33-80-219.tpgi.com.au> was last seen on IRC in channel #maemo, 31d 10h 21m 32s ago, saying: 'particularly given that the train trips I take on a regular basis have areas with low or no 2G signal'. Jul 19 11:29:10 ~seen FatPhil Jul 19 11:29:11 fatphil <~pcarmody@87-119-188-75.tll.elisa.ee> was last seen on IRC in channel #maemo, 234d 20h 8m 40s ago, saying: 'I know I'll get in trouble for asking, but what kernel are you running?'. Jul 19 11:29:20 ~seen FatPhil_ Jul 19 11:29:23 fatphil_ <~pcarmody@87-119-183-129.tll.elisa.ee> was last seen on IRC in channel #maemo-ssu, 668d 14h 41m 8s ago, saying: 'yup, those are the US americans'. Jul 19 11:29:32 :-( Jul 19 11:29:37 APic: ta Jul 19 12:01:42 echo $((`sed -n 's/^Swap:\s*\(\S*\)\s*kB$/\1/p' /proc/$PID/smaps 2>/dev/null | tr '\n' '+' | sed 's/+$/\n/'`)) Jul 19 12:01:53 this is working also on maemo kernel 2.6.28 :-) Jul 19 12:02:00 KotCzarny: ^^^ Jul 19 12:02:45 at least I belive that Swap line in smaps is what I want to use... Jul 19 12:03:15 check if it's true, start new program and see if it's 0 Jul 19 12:04:07 [root@Nokia-N900:~]# ./checkswap Jul 19 12:04:07 0 Jul 19 12:04:09 not worky Jul 19 12:04:33 Linux Nokia-N900 2.6.28.10-power53 #1 PREEMPT Wed Dec 10 13:52:39 UTC 2014 armv7l unknown Jul 19 12:04:46 Mem: 235252 228728 6524 0 4728 Jul 19 12:04:46 Swap: 786424 64828 721596 Jul 19 12:08:02 "Swap" shows how much would-be- Jul 19 12:08:03 anonymous memory is also used, but out on swap. Jul 19 12:08:09 from man 5 proc Jul 19 12:14:18 http://pastebin.com/AEZf1fSN Jul 19 12:15:13 pid=1444; sum=0; for each in `grep Swap /proc/$pid/smaps | sed -n 's/^Swap:\s*\(\S*\)\s*kB$/\1/p'`;\ Jul 19 12:15:13 do sum=$[ $sum + $each ]; done; echo $sum Jul 19 12:15:16 try this one Jul 19 12:16:43 (needs bash tho) Jul 19 12:17:12 or just change [ ] to (( )) Jul 19 12:19:53 http://pastebin.com/tFdYdzRL Jul 19 12:19:58 shell script memuse Jul 19 12:20:11 it works also in maemo busybox sh Jul 19 12:21:07 so hildon-status-menu consume most of RAM Jul 19 12:21:12 needs to be fixed... Jul 19 12:21:24 he he Jul 19 12:21:37 most likely it does more than name suggests Jul 19 12:46:56 http://pastebin.com/raw.php?i=BtNiBVyC Jul 19 12:46:59 those are mine Jul 19 12:47:04 in a bit different format Jul 19 12:47:37 i think i shouldnt use 'Size' field Jul 19 12:50:54 http://pastebin.com/raw.php?i=bQ9Yparb Jul 19 12:50:58 rss and swap Jul 19 12:54:14 what is browserd-s5711-nRTComMessagingServer ? Jul 19 12:54:23 and why is it eating 27M of ram Jul 19 12:57:06 and this is the script i used: http://pastebin.com/raw.php?i=e75RvRvk Jul 19 13:30:45 KotCzarny: browserd is server daemon for gecko html browser renderer Jul 19 13:31:24 and RTComMessagingServer is probably server part of GUI Chats/Messaging application Jul 19 13:31:52 so I think that process load that HTML+CSS stuff and render SMS, IM messages, history, etc Jul 19 14:08:53 phone part apps eat most of the mem Jul 19 14:09:38 rtcom server + msgui + callui == 60 megs Jul 19 14:10:11 maybe i should write replacements in gtk, hum Jul 19 14:10:42 <3 Jul 19 14:10:54 I'm sure it would be snappier Jul 19 14:10:59 hehe Jul 19 14:11:19 i bet even pygtk would be Jul 19 14:11:20 not sure you can though Jul 19 14:11:34 erm no, no py pretty please :D Jul 19 14:11:46 ;) Jul 19 14:12:28 but seriously, bloat, bloat everywhere Jul 19 14:16:32 what is modest? Jul 19 14:17:14 mail reader Jul 19 14:17:22 this one is opensource Jul 19 14:17:33 osso-abook + osso-abook-home applet == 20megs Jul 19 14:17:36 doh Jul 19 14:17:52 i have never used modest Jul 19 14:17:58 it shouldnt be running Jul 19 14:18:08 you can disable it Jul 19 14:18:09 and half of the other stuff Jul 19 14:18:14 easily? Jul 19 14:18:27 yeah, that's a one line conf Jul 19 14:18:32 i believe in opt-in bloat Jul 19 14:18:50 not load all, you has swaps Jul 19 14:19:19 KotCzarny: done in maemo to speed up startup Jul 19 14:19:28 vital for dialer Jul 19 14:19:47 doc: i can understand dialer, but what if one never uses some app? Jul 19 14:19:53 arguable for stuff like messaging Jul 19 14:20:13 java, *office have preloaders too, but at least they ask if they should run Jul 19 14:20:19 (never use) like modest? kill it with fire Jul 19 14:20:27 :) Jul 19 14:21:04 it's a tad tricky since it has persistent rerun schedules in alarmd Jul 19 14:21:05 would be nice to have page in settings to enable/disable apps or their preloaders Jul 19 14:21:30 DocScrutinizer05: who/what ? Jul 19 14:21:37 modest Jul 19 14:21:56 7.5megs for modest alone is quite a chunk Jul 19 14:22:03 hmm ... disabling it in the app/desktop file should be enough Jul 19 14:22:08 looks like it works here at least Jul 19 14:22:18 almost 6% of mem Jul 19 14:22:28 bencoh: prolly not, alarmd reruns it on next scheduled check for new mail Jul 19 14:22:52 DocScrutinizer05: right, I dont have any mail account on it :) Jul 19 14:23:08 so nothing to check Jul 19 14:24:03 dunno, possible it actually acts "smart" and doesn't schedule reruns in alarmd then Jul 19 14:24:29 wiki page for slimming down unused/unneeded apps? Jul 19 14:24:48 there arent that many to kill Jul 19 14:24:58 unless you dont use your phone Jul 19 14:25:06 but it would at least provide hints for users what can be disabled Jul 19 14:25:11 some people use mail, some dont Jul 19 14:25:13 etc Jul 19 14:25:14 right Jul 19 14:25:31 you can check modest, calendar, abook, browserd Jul 19 14:25:36 alarmclient -L|less +'/modest' ;; I guess Jul 19 14:25:45 i think abook is needed for contacts Jul 19 14:25:57 KotCzarny: indeed Jul 19 14:25:58 same for browserd Jul 19 14:26:25 abook is needed for contacts which in turn is needed for dialer Jul 19 14:26:28 if we had replacement for dialer/contacts it would be nice Jul 19 14:26:48 browserd is used in a *lot* of things Jul 19 14:26:52 can one make calls/send receive sms without? Jul 19 14:27:03 err, not really Jul 19 14:27:04 without what ? Jul 19 14:27:21 without dialer I guess Jul 19 14:27:22 ie. can i call some dbus command to make a call? Jul 19 14:27:24 you need a UI anyway, unless you want to type some dbus cli Jul 19 14:27:33 see wiki phone control Jul 19 14:27:58 but question is if it goes through dialer or direct to modem Jul 19 14:28:03 KotCzarny: making a call invokesa dialer to show call status and allow for call teardown Jul 19 14:28:37 that's why i thought about dialer/messaging/contacts replacement Jul 19 14:28:37 try: pnatd; ATD; Jul 19 14:28:57 watch the ending semicolon! Jul 19 14:29:02 doc: and for receiving? Jul 19 14:29:11 pnatd ATA Jul 19 14:29:24 but ... inbound call invokes dialer Jul 19 14:29:30 can it be disabled? Jul 19 14:29:35 and dialer prolly triggers ringtone Jul 19 14:29:40 and just call some other dbus service? Jul 19 14:29:53 everything is possible ;-) Jul 19 14:30:08 doc, when you return from the camp im gonna nag you about it Jul 19 14:30:10 :) Jul 19 14:30:16 hihi Jul 19 14:30:42 it would be swell to have phone/messaging part splitted from other social clunk Jul 19 14:30:48 at least for me Jul 19 14:31:18 dialer and messaging are basically unentangled Jul 19 14:31:58 because seriously: Jul 19 14:31:59 14000 156 /usr/bin/rtcom-messaging-ui (/proc/5710) Jul 19 14:31:59 13852 172 /usr/bin/rtcom-call-ui (/proc/5474) Jul 19 14:32:11 and it's running all the time Jul 19 14:32:13 a (potentially) *huge* resource hog is SQLite eventsdb Jul 19 14:32:24 no such app Jul 19 14:32:34 unless you mean they load it into mem Jul 19 14:32:59 dialer and messaging use SQLite to access eventsdb Jul 19 14:33:19 anyone has a freshly flashed n900 to check how much mem they use Jul 19 14:33:20 and that loads basically the whole database into memory Jul 19 14:34:05 and worst thing: it *semms* Nokia didn't ever bother about shrinking the database by purging old entries Jul 19 14:34:54 you can delete single recors from call history via GUI, afaik Jul 19 14:35:10 btw. Jul 19 14:35:13 you can delete single text messages, via GUI Jul 19 14:35:14 phone history is cleaned Jul 19 14:35:21 oldest entry is 6months ago Jul 19 14:35:53 so at least this part isnt cluttered Jul 19 14:36:59 and you can clean sms 'conversations' per number Jul 19 14:37:10 IroN900:~# ls -l `locate el-v1` Jul 19 14:37:12 -rw-r--r-- 1 user users 1199104 2015-07-18 22:27 /home/user/.rtcom-eventlogger/el-v1.db Jul 19 14:37:13 -rw-r--r-- 1 user users 0 2011-03-06 20:12 /home/user/.rtcom-eventlogger/el-v1.db-journal Jul 19 14:37:15 -rw-r--r-- 1 user users 0 2015-07-18 22:27 /var/lib/sqlite3/_home_user_.rtcom-eventlogger_el-v1.db-journal Jul 19 14:37:46 bah, you can even select multiple conversations at once Jul 19 14:38:05 or you simply delete /home/user/.rtcom-eventlogger/el-v1.db ;-P Jul 19 14:38:13 i don't wanna :P Jul 19 14:38:31 but you have more options in that matter available in the ui Jul 19 14:39:05 anyway, I should. >1MB is a reason for phone to get sluggish Jul 19 14:40:19 -rw------- 1 user users 2225152 Jul 19 13:54 /home/user/.rtcom-eventlogger/el-v1.db Jul 19 14:40:22 :) Jul 19 14:40:48 and I just found a convenient way to make sure there's no 'missed calls/SMS' on my phone: date of /home/user/.rtcom-eventlogger/el-v1.db Jul 19 14:40:52 looks like yours got recreated using wide access rights Jul 19 14:41:39 funny, seems you found a minor flaw in maemo Jul 19 14:41:54 Pali: ^^^ might you want to have a look at this? Jul 19 14:42:06 not that it is something important Jul 19 14:42:14 indeed Jul 19 14:42:16 most people use only user and some root Jul 19 14:42:23 DocScrutinizer05: what exactly? Jul 19 14:42:45 Pali: when deleting /home/user/.rtcom-eventlogger/el-v1.db , check if it's recreated with correct file permissions Jul 19 14:42:56 also, /var/lib/sqlite* stuff is also world readable Jul 19 14:43:20 so somebody create it with incorrect permissions? Jul 19 14:43:27 maybe backupmenu restore fails? Jul 19 14:43:32 no idea Jul 19 14:43:48 doc, maybe you deleted it and messaging ui recreated it Jul 19 14:44:13 I moved to this device via BM restore a maybe 6 month or a year ago. Dunno when the file permissions changed Jul 19 14:44:26 KotCzarny: that was my first idea Jul 19 14:44:40 either that, or BM restore Jul 19 14:45:05 ls -la /home/user/ ? Jul 19 14:45:12 Pali: it's no big thing. Just a point for the "boring weekends" todo list ;-) Jul 19 14:45:18 .dbus should be user readable only Jul 19 14:45:39 .mafw* too Jul 19 14:45:49 you will see if it's the only one Jul 19 14:46:22 drwx------ 3 user users 4096 2000-01-01 01:00 /home/user/.dbus/ Jul 19 14:46:35 2000? :) Jul 19 14:46:44 same here Jul 19 14:46:45 it is normal Jul 19 14:47:07 this is time is set when rtc is not working Jul 19 14:47:25 :nod: Jul 19 14:47:28 our rtc battery is bad Jul 19 14:47:41 we have rtc battery at all? Jul 19 14:47:49 yes, n900 has it Jul 19 14:47:51 sure Jul 19 14:47:57 ~bupbat Jul 19 14:47:58 wouldnt it leak after so many years? Jul 19 14:48:00 it has been said that bupbat is use the capacitive type, LiIon are breaking during 12 months, or http://www.digikey.de/product-detail/de/PAS414HR-VG1/587-2157-1-ND/1959153, or https://hbe-shop.de/Art-2112777-TAIYO-YUDEN-PAS414HR-VA5R-KONDENSATOR006F-33V-STAKED-COIN, or http://talk.maemo.org/showthread.php?t=90864 Jul 19 14:48:10 12 months, lol Jul 19 14:48:54 second link is dead Jul 19 14:49:07 or just site rehashing Jul 19 14:49:09 meh, it's dead every other month Jul 19 14:50:29 it should be user replaceable unit :/ Jul 19 14:50:30 pity Jul 19 14:56:09 it shouldn't break sooner than after 10 years of product life Jul 19 14:56:21 4 years remaining Jul 19 14:56:35 alas that series of batteries was broken-by-design Jul 19 15:02:38 oh Jul 19 15:02:42 i never knew Jul 19 15:03:29 > Funny, I got so used to setting date/time after swapping batteries that I assumed this had always be the case. Jul 19 15:03:44 just use ntpdate Jul 19 15:04:01 and if you do it often, make a shortcut on the desktop Jul 19 15:05:03 I kicked out the dang /etc/X11/Xsession.d/30osso_startup_wizard Jul 19 15:05:21 oh, right, gonna do the same, and start documenting Jul 19 15:05:33 settings->time&date [x]auto-update does the rest Jul 19 15:08:19 ping, hay guys, need a little advice, yesterday I was dancing Yarkhushta, and my N900 dropped from my pocket, and there is a LITTLE TINY scratch at the edge Jul 19 15:08:28 nothing big, but, anyone knows how to remove it? Jul 19 15:08:39 um, dont? Jul 19 15:08:40 or make it... less noteble? Jul 19 15:08:41 lol Jul 19 15:08:52 make the survivor scars be visible! Jul 19 15:09:26 I was thinking to use a, (what do you call that thing girls use for nails?) for the color Jul 19 15:09:28 haha KotCzarny Jul 19 15:09:33 that is also an amazing idae Jul 19 15:09:37 idea* Jul 19 15:09:49 antranigv: you could consider getting a cover for your n900 Jul 19 15:09:57 antranigv: not nail varnish, but a permanent black marker pen Jul 19 15:10:00 you can use black permanent marker Jul 19 15:10:04 yeah Jul 19 15:10:12 DocScrutinizer05: cool idea Jul 19 15:10:16 first!! \o/ Jul 19 15:10:17 Wizzup: a cover? Jul 19 15:10:23 like a case? Jul 19 15:10:25 doc: i'm editing the wiki :P Jul 19 15:10:40 permanent black marker doesnt work on the bezel Jul 19 15:10:47 as far as I can tell Jul 19 15:10:51 depends on the marker, he he Jul 19 15:11:04 maybe ... Jul 19 15:11:05 case Jul 19 15:11:05 yes Jul 19 15:11:08 I have this one: http://www.amazon.com/PDair-Aluminum-Metal-Case-Nokia/dp/B004A7M2KK Jul 19 15:11:33 38usd, wow, almost price of the device Jul 19 15:12:33 KotCzarny: lol Jul 19 15:13:41 http://www.ebay.com/sch/Cell-Phones-Smartphones-/9355/i.html?_from=R40&_sop=15&_nkw=nokia%20n900&rt=nc&LH_BIN=1&_trksid=p2045573.m1684 Jul 19 15:13:44 wow, 800usd Jul 19 15:13:57 vintage! collectible! rare! Jul 19 15:13:59 lol Jul 19 15:14:17 is that new? Jul 19 15:14:21 you guys should better care for your vintages :P Jul 19 15:15:22 the pen did it fine! I mean the marker Jul 19 15:15:30 let it dry tho Jul 19 15:15:34 looks better Jul 19 15:15:39 feels the same Jul 19 15:16:14 this will help me to feel the phone, so I wont hold it in the wrong dirrection Jul 19 15:16:17 direction* Jul 19 15:16:27 O.o Jul 19 15:16:49 * antranigv feels like The DareDevil Jul 19 15:16:51 :P Jul 19 15:16:59 you feel the colors? Jul 19 15:17:00 :P Jul 19 15:17:26 KotCzarny: yes, but the wrong way... I'm also daltonic Jul 19 15:17:37 daltonic? is it right to day that? Jul 19 15:17:46 (that's what we say at least) Jul 19 15:17:51 color blind :P Jul 19 15:17:53 I have problem with the colors Jul 19 15:17:54 yes! Jul 19 15:17:59 KotCzarny: tnx! :D Jul 19 15:18:17 https://en.wikipedia.org/wiki/Color_blindness Jul 19 15:18:51 daltonism is used in english only for deuteranopia Jul 19 15:19:05 (red-green color blindness) Jul 19 15:19:44 no I'm not red... I think the green is my problem Jul 19 15:20:55 anyways, my biggest problem will be not having a N900 Jul 19 15:27:23 doc: /etc/X11/Xsession.d/10hildon_welcome is nokia hands? Jul 19 15:30:47 http://wiki.maemo.org/Slimming_OS Jul 19 15:30:55 antranigv: same thing, cheaper here http://www.pdair.com/index.php/aluminum-metal-case-for-nokia-n900-open-screen-design-black.html Jul 19 15:30:56 initial page, anyone has other tips? Jul 19 15:37:43 Wizzup: tnx! Jul 19 15:37:49 I got it from there Jul 19 15:37:51 v happy with it Jul 19 15:37:58 KotCzarny: ah, the hands will be handy, I hate them (: Jul 19 15:38:00 saved my phone (and also my girlfriend's n900) in many occasions Jul 19 15:39:06 https://wiki.maemo.org/N900_The_Perfect_Setup#Cherry Jul 19 15:39:08 wow Jul 19 15:56:04 that cherry's been popped :D hehehee ...ahhhh! Goooodddd Morrnninnnnggg Topeka! Jul 19 15:56:34 Watched that guitar clip...guys got skillz Jul 19 15:56:53 yup. first minute i was like 'oh, is that all' then he started rocking Jul 19 16:00:26 hm ..yeah pdair has 'em cheaper direct...best hard case I have...have em too for my n810's...got the leather "book" style as well...but Boxwave makes the best leather variant..it's got card slots and slots for credit/debit/id cards as well as a huge pocket the width and length of the case front. Jul 19 16:01:02 yeah..watched his other vids as well...pretty darn good. Jul 19 16:06:47 hmm, anyone know if funambol is needed? Jul 19 16:17:59 * DocScrutinizer05 dispises https://wiki.maemo.org/N900_The_Perfect_Setup - it's quite... subjective. And for cherry, it's more involved that simply uninstalling the package Jul 19 16:18:18 perfect setup would be mainline kernel ;) *cough* Jul 19 16:18:35 mainline kernel AND current libraries Jul 19 16:18:41 indeed, what a hybris to claim you found the perfect setup Jul 19 16:19:15 still, anyone has some tips and tricks of what is safe to remove? Jul 19 16:19:22 sort of resembles crappatch Jul 19 16:19:33 ~crappatch Jul 19 16:19:33 speedpatch is probably http://talk.maemo.org/showthread.php?p=1012405#post1012405 >>first i don't realy understand what does this patch do (that is why it is called miracle patch)<< [/quote original-author-of-speedpatch], or http://talk.maemo.org/showpost.php?p=1328060&postcount=3325 [bug report] Jul 19 16:19:47 setup-that-sucks-less ? :) Jul 19 16:20:18 KotCzarny: I don't care too much about 2what to remove" Jul 19 16:20:32 you do Jul 19 16:20:33 since... it's individual Jul 19 16:20:47 otherwise you wouldnt be looking for hildon-welcome etc Jul 19 16:20:52 ~jrtools Jul 19 16:20:52 hmm... jrtools is http://wiki.maemo.org/User:Joerg_rw/tools Jul 19 16:27:04 what is qtm? Jul 19 16:29:55 context? Jul 19 16:30:05 libqtm* Jul 19 16:30:14 qt-mobile, I guess Jul 19 16:30:22 it's some framework, but is it used by anyone/anything? Jul 19 16:58:46 hmm, is nokiamessaging needed? Jul 19 17:07:54 no, prolly obsolete Jul 19 17:09:10 good. 4.5M of rootfs reclaimed Jul 19 17:09:28 also, trying to remove modest messes badly with apt Jul 19 17:09:30 :) Jul 19 17:13:00 it messes with the metapackage Jul 19 17:13:14 yeah, but things get hairy Jul 19 17:15:14 that's the congenital defect of maemo: Nokia decided to pack a single huge metapackage to deploy stuff they think every device should have. It's quite fsckdup a concept Jul 19 17:15:43 can it be fixed by replacing such package by a dummy one? Jul 19 17:15:53 it was very convenient for them since they didn't need to care about *real* dependencies Jul 19 17:16:33 (dummy) that's extremely tricky Jul 19 17:16:51 but basically it is exactly what CSSU does Jul 19 17:17:00 what is the name of the pkg? Jul 19 17:18:31 mp-fremantle-*-pr Jul 19 17:18:51 ii mp-fremantle-community-pr 21.2011.38-1Tmaemo1.2 Community SSU package, includes latest fixes from Gitorious and the community Jul 19 17:18:52 mp-* Jul 19 17:18:55 i have it installed Jul 19 17:19:44 so it has to be something else Jul 19 17:19:52 huh? Jul 19 17:20:02 it? hast to be? sth else? Jul 19 17:20:20 this is the mETApACKAGE Jul 19 17:20:21 try on yours: apt-get remove modest Jul 19 17:20:28 no I won't Jul 19 17:20:37 since I know modest is in mp Jul 19 17:20:41 (doesnt have to say y, it just calculates deps) Jul 19 17:20:58 so, can we get modest OUT of mp- ? Jul 19 17:21:08 sure, edit it Jul 19 17:21:57 which, what, where Jul 19 17:22:12 ~xy Jul 19 17:22:12 xy is probably The XY problem: You want to do X, but don't know how. You think you can solve it using Y, but don't know how to do that, either. You ask about Y, which is a strange thing to want to do. Just ask about X. http://www.catb.org/~esr/faqs/smart-questions.html#goal Jul 19 17:22:15 by edit you mean 'download, unpack, edit, repack, reinstall' ? Jul 19 17:22:52 ask cssu guys Jul 19 17:23:01 who is cssu guys? Jul 19 17:25:22 apt-cache show mp-fremantle-community-pr Jul 19 17:25:57 mohammad7410 Jul 19 17:29:11 apt-cache show mp-fremantle-community-pr|egrep "(Depends:|Replaces:|Conflicts:)" Jul 19 17:30:04 osso-calculator (>= 2.0-1+0cssu1), osso-calculator-ui (>= 2.0-1+0cssu1) Jul 19 17:30:11 wow, calculator with server-client model? Jul 19 17:31:28 apt-cache show mp-fremantle-community-pr|grep "^Depends:"|head -n 1 Jul 19 17:32:10 yep, they pimped calculator beyond sanity Jul 19 17:32:18 hmm, can mp-fremantle-community-pr be simply removed? Jul 19 17:32:34 in theory yes Jul 19 17:32:40 and in practice? Jul 19 17:32:52 in practice you won't receive any updates then Jul 19 17:33:00 afaik Jul 19 17:33:05 ask cssu folks Jul 19 17:33:08 updates to what? Jul 19 17:33:16 to CSSU :-) Jul 19 17:33:35 but that only if they invent some new package Jul 19 17:34:05 which is a description of the meaning of "update2, yes Jul 19 17:34:25 well, already installed pkgs will update Jul 19 17:34:31 no Jul 19 17:34:36 no? Jul 19 17:34:42 *sigh* Jul 19 17:34:45 meaning not yes Jul 19 17:34:47 isn't that apt job? Jul 19 17:34:53 please ask somebody else, i'm not a packager Jul 19 17:35:08 k, changing topic: this is the thunderstorm heading to me http://www.pogodynka.pl/http/assets/products/aktualna_pogoda/PERUN_DIS.GIF Jul 19 17:35:55 maybe i should turn of some of the devices :> Jul 19 17:44:33 KotCzarny: when nokia (or CSSU) rolls out a new release, they provide an update of the mp, and that in turn updates all the packages vie the "depends:". Maemo doesn't use the usual apt-get upgrade, since that wouldn't know about dependencies between packages, updating one package to new version and leaving another package on an incompatible level. I explained that Nokia decided to not care about proper dependencies in "simple" packages Jul 19 17:44:35 and instead packed everything into a metapackage to ensure everything is consistent. So apt-get upgrade is deprecated Jul 19 17:46:20 doc: all would be nice and dandy if we had few gigs for rootfs Jul 19 17:50:44 KotCzarny: we do..? Jul 19 17:50:56 luke: for / ? Jul 19 17:51:27 hm, maybe not Jul 19 17:51:29 rootfs 227.8M 189.3M 34.3M 85% / Jul 19 17:51:30 :| Jul 19 17:51:32 :) Jul 19 17:51:35 new to maemo? Jul 19 17:53:04 no, old Jul 19 17:53:10 i know, just kidding Jul 19 17:53:52 '//dev/mmcblk0p5 9.8G 8.3G 1.0G 89% /media/gentoo Jul 19 17:53:53 :P Jul 19 17:54:14 * Luke-Jr stabs N900 for having such a slow MicroSD bus Jul 19 17:54:35 osso-notes* osso-sketch* osso-rss* out! 4.2M of rootfs back Jul 19 17:55:04 >>> Installing (55 of 542) virtual/acl-0-r1 Jul 19 17:56:53 one more positive side of slimming is fewer trash-apps in app menu Jul 19 17:58:02 -rw-r--r-- 1 root root 23911920 Feb 19 2010 /usr/lib/locale/locale-archive Jul 19 17:58:04 wow Jul 19 18:22:42 Prolog is used on N900 to decide about audio output levels for speakers and audio out (see e.g. first paragraph of http://wiki.maemo.org/Porting/Audio ). That sounds like a crazy idea, doesn't it? I wish I could meet a guy who decided "hey, using prolog for this would be a great idea" Jul 19 18:22:43 lol Jul 19 18:23:03 lol Jul 19 18:28:03 KotCzarny: what the heck are you doing? osoo-notes for sure should have been optified and thus not relevant for rootfs Jul 19 18:28:25 removing apps i never use Jul 19 18:28:38 just wha? Jul 19 18:28:42 why Jul 19 18:29:01 partly for checking how far i can go, partly because i never use them Jul 19 18:29:10 tzz Jul 19 18:29:44 nobody really interested in this Jul 19 18:29:58 http://talk.maemo.org/showthread.php?t=82641 Jul 19 18:30:00 nobody? Jul 19 18:30:16 *shrug* Jul 19 18:30:36 1001 ways to fuck up your device Jul 19 18:30:51 sure, but reflash is quick and easy Jul 19 18:31:00 not to mention you learn by breaking usually Jul 19 18:31:09 (i mean by fixing what you break) Jul 19 18:31:28 so do that in private, i'd suggest. I for one had my share already Jul 19 18:32:47 i didn't know this irc channel had such rules in place Jul 19 18:33:16 there's no such rule in place. there's only a level of interest I have in certain topics Jul 19 18:33:39 #maemo == I ? Jul 19 18:33:47 I didn't know the channel turned into 2how to brick your N900" Jul 19 18:34:22 always thought 'hacking' was 'doing things that aren't always safe' Jul 19 18:34:39 afk Jul 19 18:49:15 KotCzarny: for that stuff like policies, it really make sense to write rules in such "logical" language Jul 19 18:50:07 it is easier to define policies and action based on know facts in prolog Jul 19 18:50:18 as in other laguages like C Jul 19 18:50:21 pali: nice to know, i have only heard about prolog, never actually seen it Jul 19 18:51:29 swi prolog (used in N900) is doing some kind of SLD resolution to achieve or decide if predicate is truth or not Jul 19 18:51:57 and this is part of language Jul 19 18:52:09 you just need to write rules Jul 19 18:52:23 and prolog do all magic below Jul 19 18:52:49 in other programming laguages you need to write how to solve such problem Jul 19 18:53:05 in prolog you just write "description" of problem (not how to solve it) Jul 19 21:52:51 hmm, does anyone actually tested battery life with and without smartreflex? Jul 19 21:54:03 KotCzarny: yes, it was vi, back then Jul 19 21:55:10 and is it worth? i'm using it for some time, yet i don't see anything noticeable Jul 19 21:55:21 (at least according to battery-eye) Jul 19 21:55:58 on a positive note it's stable, didn't have any unexpected reboots Jul 19 21:56:09 let me try to find the thread Jul 19 22:02:39 hmm, seems the thread has disappeared :( Jul 19 22:03:01 i only see vi_ commenting on smartreflex in this thread http://talk.maemo.org/showthread.php?t=82145 Jul 19 22:03:08 the last post from vi_ on TMO is from 06-16-12 Jul 19 22:03:53 KotCzarny: no, it was a beautiful thread, with graphics, for different frequencies and loads Jul 19 22:04:18 unfortunately it seems to be archived/deleted Jul 19 22:09:06 KotCzarny: anyway, I remember something like 8% lower consumption for (500? 600?) MHz Jul 19 22:09:17 uhum, i set mine to 500 too Jul 19 22:09:40 but anyway SR is highly device dependent, so your mileage may vary Jul 19 22:09:41 and 8% goes well into observable margin Jul 19 22:10:05 would be nice to have second n900 just for testing Jul 19 22:10:24 you can check what is the voltage set by SR and compare it without SR Jul 19 22:10:38 that'll give you an idea Jul 19 22:10:56 about what saving (if any) to expect Jul 19 22:11:06 nah, i believe it's working, but i wanted to know how much in battery time it can gain Jul 19 22:11:41 well, you know such a question has no answer :) . usage pattern, you know Jul 19 22:11:56 that's why second n900 :) Jul 19 22:12:07 this one is my main phone Jul 19 22:13:32 KotCzarny: lock your frequency at 500 (for example), then "cat /sys/power/sr_vdd1_voltage" Jul 19 22:14:04 then compare it with voltage for 500 without SR Jul 19 22:14:17 it is in units of .125V, (iirc) Jul 19 22:15:01 38 Jul 19 22:15:35 hmm, it is 32 here Jul 19 22:15:42 are you sure you have SR enabled? Jul 19 22:16:04 1 in /sys/power/sr_vdd[12]_autocomp ? Jul 19 22:16:10 yep Jul 19 22:16:15 then yes Jul 19 22:16:27 without sr its 39 Jul 19 22:17:11 well, then you have a "bad" device which doesn't gain much from SR Jul 19 22:17:20 and what about 250MHz? Jul 19 22:18:08 mine is at 23 with SR Jul 19 22:18:14 24 with, 25 without Jul 19 22:18:50 well, oscillating between 24 and 25 Jul 19 22:19:25 power53? Jul 19 22:19:33 yeah, not much of a difference Jul 19 22:19:53 Linux Nokia-N900 2.6.28.10-power53 #1 PREEMPT Wed Dec 10 13:52:39 UTC 2014 armv7l unknown Jul 19 22:20:55 KotCzarny: have you ever tried to undervolt? Jul 19 22:21:07 never, apart from enabling sr Jul 19 22:21:42 btw. vdd2 value differs a bit more with/wo sr Jul 19 22:22:01 you;d better not enable SR for vdd2 Jul 19 22:22:08 I wonder what it does when usb is plugged in ... it says vdd1==35, vdd2==37 Jul 19 22:22:26 bencoh: locks the freq at 600 (iirc) Jul 19 22:23:15 freemangordon: vdd2 is bad in a way it can crash the device or silent corruption? Jul 19 22:23:29 500 apparently, but yeah, it might be just that Jul 19 22:23:31 so far i didn't notice anything Jul 19 22:23:42 what's wrong with SR on vdd2 ? Jul 19 22:24:12 bad in a way that SR calibration values are wrong (as are those for vdd1), but I didn;t make a patch to fix them (which I did for vdd1) Jul 19 22:26:13 and I can only guess what can an undervolted bus bring, be it crash or silent corruption Jul 19 22:26:33 print a warning in dmesg Jul 19 22:26:37 just for now Jul 19 22:26:39 :) Jul 19 22:26:42 I think there is Jul 19 22:27:12 nope Jul 19 22:27:24 could be, it was a while I last played with it :) Jul 19 22:28:02 freemangordon: too high or too low ? Jul 19 22:28:02 anyway, so far my device was stable, even with vdd2, but as you've said it's not undervolting much Jul 19 22:28:09 too low Jul 19 22:28:14 oh Jul 19 22:28:49 bencoh: hmm, wait Jul 19 22:29:13 it is a bit more complicated Jul 19 22:29:26 for vdd1 there a 5 efuse values Jul 19 22:29:36 and 3 for vdd2 Jul 19 22:30:11 the last 3 ( for 500, 550 and 600) for vdd1 are same on all devices I've ever checked Jul 19 22:30:27 the same goes for 3rd value for vdd2 Jul 19 22:31:05 /sys/power/efuse_vdd1 has 10 lines here Jul 19 22:31:15 while 1st and second values are too high for all the devices I've checked Jul 19 22:32:02 KotCzarny: yes, thosa are for frequencies 125-805 Jul 19 22:32:04 *those Jul 19 22:33:07 bencoh: so, we have 2 values that are always higher than needed and 3 which are too high or too low for a particular device Jul 19 22:33:19 the same goes for vdd2 Jul 19 22:33:55 we have 2 values that are too high and 1 which could be high or low Jul 19 22:34:25 hmm Jul 19 22:34:34 KotCzarny: your 3rd value for vdd2 is 00aac695, right? Jul 19 22:34:42 yes Jul 19 22:34:58 bencoh: ^^^ you have the same, right? Jul 19 22:35:13 freemangordon: and ... how do we know it's too high or too low ? checking against some doc ? Jul 19 22:35:15 009aaedf 00aafda7 00aac695 Jul 19 22:35:32 bencoh: there is no way you can know Jul 19 22:36:01 those are supposed to be parameters, factory caliibrated for a particular SoC Jul 19 22:36:06 I mean ... how did you check ? Jul 19 22:36:17 I didn;t Jul 19 22:36:30 I took the first pair only Jul 19 22:37:03 "we have 2 values that are too high and 1 which could be high or low" compared to what ? empiric values ? Jul 19 22:37:04 for vdd1 that is. The other values are extrapolated Jul 19 22:37:23 I have 009ac0eb 009a8bb0 00aac695 for vdd2 Jul 19 22:37:30 bencoh: compared to what the device needs to be stable, without consuming unnecessary power Jul 19 22:38:04 yep, again 00aac695 ;) Jul 19 22:38:34 factory failsafe of some kind? Jul 19 22:38:47 no idea what it is, but it is wrong Jul 19 22:40:54 I'll read the omap3 documentation, some day 0:) Jul 19 22:41:17 (not saying this just regarding this specific point ... :) Jul 19 22:42:07 KotCzarny: the idea is - TI measures the device performance at different OPPs, loads and temperatures, then programs eFuses with SoC specific values. Those values are used by the kernel to program SR controller, which in turns controls TWL's voltage regulators. Jul 19 22:42:22 *int turn Jul 19 22:42:34 dammit, too sleepy :) Jul 19 22:42:58 pity i have bad device Jul 19 22:43:05 still, works as it is Jul 19 22:43:28 unproductive smartreflex doesn't consume any overhead energy? Jul 19 22:43:40 yeah, your device seems to be on the bad end of the spectrum Jul 19 22:43:50 afaik no Jul 19 22:44:20 thanks for the info :) Jul 19 22:44:35 going to leave it enabled anyway Jul 19 22:44:54 KotCzarny: keep in mind that even 0.25V down make some difference Jul 19 22:45:29 heh, how about a rescueos based kernel with testing script? :) Jul 19 22:45:42 testing what? Jul 19 22:45:53 you've said something about locking voltages Jul 19 22:46:09 or i'm sleepy too and read more that there was said Jul 19 22:46:18 I guess Jul 19 22:46:19 :) Jul 19 22:46:38 ah, i see, you have asked me about undervolting Jul 19 22:46:57 locking was to give you an idea how does SR perform on your device Jul 19 22:47:31 to compare vdd1 voltage with nd without SR Jul 19 22:47:34 *and Jul 19 22:51:37 also that explains why i didnt notice any battery life improvements Jul 19 22:51:52 :nod: Jul 19 22:52:41 in 125mhz vdd1 is 23 Jul 19 22:54:19 and vdd2 is 30 Jul 19 22:58:42 http://talk.maemo.org/showpost.php?p=1032062&postcount=76 Jul 19 22:58:43 he he Jul 19 22:59:42 so actually sr should make chips more stable Jul 19 22:59:59 assuming it's working right Jul 19 23:00:24 and for some chips it could make battery life shorter Jul 19 23:01:53 well, scratch the last line Jul 19 23:13:22 on a side note: http://processors.wiki.ti.com/index.php/Building_Qt_with_OpenGL_ES_accelerated_by_SGX Jul 19 23:17:13 hmm WindowSystem=libpvrQWSWSEGL.so Jul 19 23:18:21 that's for qws ... and afaik not for qt/x11 Jul 19 23:18:43 uhum Jul 19 23:24:08 harmattan has much more tweakables in this file Jul 19 23:29:28 bencoh, was it you who suggested one line disabling for modest? Jul 19 23:33:02 wow, 18yo said 'grow up' to freemangordon :) Jul 20 02:02:44 No idea if this is too device specific, but is the x-loader binary signed or signed and encypted? **** ENDING LOGGING AT Mon Jul 20 02:59:58 2015