IRC log of #maemo for Thursday, 2014-07-17

Sicelo	what adapter?	00:05
M4rtinK	the original one	00:06
Sicelo	by adapter you mean charger?	00:06
M4rtinK	yeah	00:07
M4rtinK	oh, looks like it booted now	00:07
Sicelo	you might have to blacklist some charger module somewhere.. i never had that problem, but there was a time when certain KP had those issues	00:07
M4rtinK	seems I had to let it charge a bit first	00:08
Sicelo	ah.. your battery was too flat :)	00:08
M4rtinK	well, one would thing this is not a problem when the charger is connected	00:08
M4rtinK	I guess it is not that simple :)	00:08
*** jaykru has joined #maemo		00:10
*** _rd has quit IRC		00:18
ult	mine boots successfully approximately half of the times with usb cable connected. judging by syslog it goes into ACT_DEAD the other half.	00:25
ult	but i have g_nokia inserted from rcS-late and maybe percentage of charge left affects bme decision somehow. KP50	00:26
*** flo_lap has joined #maemo		00:30
*** florian has quit IRC		00:30
*** flo_lap is now known as florian		00:30
sixwheeledbeast	~flatbattrecover	00:42
sixwheeledbeast	~flatbatrecover	00:42
infobot	Remove battery for 1 minute. Insert battery. Plug powered *NOKIA WALLCHARGER* to device. Watch steady amber. Let sit and charge. Do NOT try to boot. After 30 min, you got either a) a booted up N900, b) flashing amber which means you can boot, c) steady amber going off - in this case start over again with ~flatbatrecover. CAVEAT! Only works when ~rootfs is OK (no ~bootloop)!.	00:42
sixwheeledbeast	ah, one t :)	00:42
M4rtinK	sixwheeledbeast: BTW, I think I fixed that logging bug	00:47
*** arcean_ has joined #maemo		00:47
sixwheeledbeast	M4rtinK: great I'll be testing it when it's released. planning on doing a lot of walking/cycling for the next month or so.	00:48
M4rtinK	in related news the routing data repository has been updated: http://talk.maemo.org/showpost.php?p=1433062&postcount=1610	00:48
M4rtinK	check the post for a nice example of a 3000 km route computed on device with the 6 GB Europe-car dataset :)	00:49
M4rtinK	sixwheeledbeast: you will be soon able to - I'm sending it to autobuilder now :)	00:50
*** arcean has quit IRC		00:50
sixwheeledbeast	M4rtinK: :) I need to look at getting offline routing working too, so I will be playing with that also.	00:51
*** jaykru has quit IRC		00:52
M4rtinK	I'd like to eventually make it easier to use	00:52
M4rtinK	mainly regarding downloading the data packs - it should be possible to do it directly from modRana	00:52
M4rtinK	sixwheeledbeast: ok, should be built :)	00:57
sixwheeledbeast	M4rtinK: I agree about making it easier to use.	00:58
sixwheeledbeast	M4rtinK: Great, but don't expect me to go walking at this time of night :P	00:58
M4rtinK	bah :D	00:58
M4rtinK	looking forward to your feedback :)	01:05
M4rtinK	and good night :)	01:05
*** M4rtinK has quit IRC		01:06
*** beford has joined #maemo		01:25
*** DrCode has quit IRC		01:25
*** DrCode has joined #maemo		01:26
*** eMHa has quit IRC		01:28
*** louisdk has quit IRC		01:33
*** eMHa has joined #maemo		01:38
*** florian has quit IRC		01:44
*** sixwheeledbeast has quit IRC		01:56
*** sixwheeledbeast has joined #maemo		01:56
*** sixwheeledbeast has quit IRC		02:03
*** lufu has quit IRC		02:12
*** N-Mi has quit IRC		02:24
*** beford has quit IRC		02:28
*** GeneralAntilles has left #maemo		02:35
*** Ashley` has quit IRC		02:42
*** fk_lx has quit IRC		02:42
*** goldkatze has quit IRC		02:43
*** FIQ has quit IRC		02:45
*** FIQ has joined #maemo		02:45
*** FIQ is now known as Guest16118		02:45
*** N-Mi has joined #maemo		02:53
*** N-Mi has joined #maemo		02:53
*** akeripper has joined #maemo		02:53
*** nox- has quit IRC		02:55
*** ult has quit IRC		03:10
*** Kabouik has quit IRC		03:13
*** N-Mi has quit IRC		03:24
*** arcean_ has quit IRC		03:27
*** Snafu777 has joined #maemo		03:49
*** Snafu777 has quit IRC		03:54
*** Snafu777 has joined #maemo		04:05
*** janemba has quit IRC		04:12
*** janemba has joined #maemo		04:13
*** robbiethe1st has joined #maemo		04:16
*** b1101 has quit IRC		04:19
*** Humpelst1lzchen has quit IRC		04:42
*** Humpelstilzchen has joined #maemo		04:46
*** flynx has quit IRC		04:46
*** silviof2 has joined #maemo		05:01
*** Guest16118 has quit IRC		05:03
*** Guest16118 has joined #maemo		05:03
*** Guest16118 is now known as FIQ		05:03
*** silviof1 has quit IRC		05:04
*** tanty_off has quit IRC		05:16
*** tanty_off has joined #maemo		05:17
*** tanty_off is now known as tanty		05:17
*** maybeArgh has joined #maemo		05:33
*** maybeWTF has quit IRC		05:34
*** jaykru has joined #maemo		05:42
*** lxp1 has joined #maemo		06:01
*** lxp has quit IRC		06:03
*** Agge has quit IRC		06:15
*** janemba has quit IRC		06:20
*** janemba has joined #maemo		06:21
*** infobot has quit IRC		06:27
*** infobot has joined #maemo		06:31
infobot	DocScrutinizer: infobot joined!	06:31
*** ChanServ sets mode: +v infobot		06:31
*** janemba has quit IRC		06:58
*** janemba has joined #maemo		06:59
*** robbiethe1st has quit IRC		07:03
*** aloril_ has quit IRC		07:05
*** protem has joined #maemo		07:06
*** protem has joined #maemo		07:06
*** janemba has quit IRC		07:17
*** aloril_ has joined #maemo		07:18
*** janemba has joined #maemo		07:18
*** at1as has quit IRC		07:33
*** at1as has joined #maemo		07:47
*** janemba has quit IRC		07:54
*** janemba has joined #maemo		07:55
*** Snafu777 has quit IRC		08:05
*** pdz has quit IRC		08:14
*** pdz has joined #maemo		08:14
*** VDVsx has quit IRC		08:17
*** pdz- has joined #maemo		08:18
*** pdz has quit IRC		08:19
*** VDVsx has joined #maemo		08:42
*** Trizt has quit IRC		08:44
*** Trizt has joined #maemo		08:45
*** shentey has joined #maemo		08:46
*** Ashley` has joined #maemo		09:09
Ashley`	~flasher	09:11
infobot	i heard flasher is at http://www.jedge.com/n810/flasher/maemo_flasher-3.5_2.5.2.2.tar.gz (also .exe!), or http://www.chakra-project.org/ccr/packages.php?ID=5027 or generally http://www.google.com/search?q=maemo_flasher-3.5_2.5.2.2.tar.gz. HARMattan(N9): https://aur.archlinux.org/packages/fl/flasher-harmattan/flasher-harmattan.tar.gz, or -- list of filenames/md5sums: http://pastebin.com/sYKdNJSH, or http://galif.eu/nokia/	09:11
*** shentey has quit IRC		09:12
*** rcg has joined #maemo		09:22
*** dhbiker has joined #maemo		09:26
*** povbot_ has joined #maemo		09:30
*** Natch_h has joined #maemo		09:31
*** NIN101_ has joined #maemo		09:31
*** Malinux_ has joined #maemo		09:31
*** lbt_ has joined #maemo		09:32
*** fil has joined #maemo		09:33
*** Scorcere1 has joined #maemo		09:33
*** oniongar1ic has joined #maemo		09:33
*** janemba_ has joined #maemo		09:33
*** pigeon_ has joined #maemo		09:33
*** thedead1440_ has joined #maemo		09:33
*** peterbjo1nx has joined #maemo		09:33
*** script_ has joined #maemo		09:33
*** Natch has quit IRC		09:33
*** Natch_h is now known as Natch		09:33
*** povbot has quit IRC		09:34
*** fil_ has quit IRC		09:34
*** akeripper has quit IRC		09:34
*** fuz_ has quit IRC		09:34
*** japa-fi has quit IRC		09:34
*** Venusaur has quit IRC		09:34
*** thedead1440 has quit IRC		09:34
*** pigeon has quit IRC		09:34
*** drawkula has quit IRC		09:34
*** Milhouse has quit IRC		09:34
*** shamus has quit IRC		09:34
*** script has quit IRC		09:34
*** felipe` has quit IRC		09:34
*** janemba has quit IRC		09:34
*** aloril_ has quit IRC		09:34
*** peterbjornx has quit IRC		09:34
*** Scorcerer has quit IRC		09:34
*** oniongarlic has quit IRC		09:34
*** NIN101 has quit IRC		09:34
*** totalizator has quit IRC		09:34
*** lbt has quit IRC		09:34
*** Malinux has quit IRC		09:34
*** ab has quit IRC		09:34
*** japa-fi_ has joined #maemo		09:34
*** ab_ has joined #maemo		09:34
*** drawkula has joined #maemo		09:34
*** ab_ is now known as ab		09:34
*** aloril_ has joined #maemo		09:34
*** Venusaur has joined #maemo		09:36
*** thedead1440_ has quit IRC		09:36
*** thedead1440 has joined #maemo		09:36
*** totalizator has joined #maemo		09:36
*** shamus has joined #maemo		09:37
*** fuz_ has joined #maemo		09:39
*** japa-fi_ has quit IRC		09:40
*** Milhouse has joined #maemo		09:45
*** japa-fi has joined #maemo		09:45
*** goldkatze has joined #maemo		10:00
*** sixwheeledbeast has joined #maemo		10:02
*** hurrian has quit IRC		10:06
*** jmlich has joined #maemo		10:26
*** janemba_ has quit IRC		10:38
*** janemba has joined #maemo		10:38
Maxdamantus	Hm. So dumping/restoring el-v1.db didn't fix it. Looked like it only saved one message afterwards for that person.	10:50
Maxdamantus	but it saves messages for other contacts.	10:50
Maxdamantus	What sort of corruption is it said to actually be?	10:51
Maxdamantus	corruption because the program is stupid (so produces tables it doesn't understand), or because sqlite3 has bugs, or because the file itself corrupts (from the filesystem)?	10:53
*** RedW has joined #maemo		10:55
*** RedM has quit IRC		10:56
Maxdamantus	I guess another form could be where sqlite3 leaves the database in an inconsistent state if the program exits for some reason .. dunno if the rt-messaging thing does that.	10:56
Maxdamantus	assuming sqlite3 doesn't use some transactional system to prevent that.	10:57
*** ult has joined #maemo		10:58
*** hurrian has joined #maemo		10:59
*** hurrian has quit IRC		10:59
*** hurrian has joined #maemo		10:59
*** Ex-Opesa has quit IRC		11:03
*** silviof2 is now known as silviof		11:05
*** florian_kc has joined #maemo		11:09
*** Ex-Opesa has joined #maemo		11:11
*** ult has left #maemo		11:11
Maxdamantus	Mm .. without replacing the el-v1.db file at all, just killing the process using it (so it restarted), it saved some messages that were resent (by the network—not sure how the protocol works there).	11:11
*** Kabouik has joined #maemo		11:12
Maxdamantus	Just one message.	11:13
*** florian_kc is now known as florian		11:19
Maxdamantus	Could there be a 200-message limit?	11:22
Maxdamantus	The contact has 202 for read_events in GroupCache	11:23
Maxdamantus	and the last field, "CONSTRAINT factor UNIQUE(service_id, group_uid)" is 1, unlike all the others, which are 0.	11:23
Maxdamantus	Maybe it intends to delete earlier messages after 200, which could be triggering some bug.	11:24
*** protem has quit IRC		11:29
Maxdamantus	Don't think so.	11:33
*** KotCzarny has joined #maemo		11:52
KotCzarny	hi	11:52
KotCzarny	my n900's clock is speeding at a rate 2min/week, is there any automatic fix for it that doesn't include ntpd?	11:52
bencoh	I'd love an opinion on that as well	11:53
KotCzarny	i could use cron, but maybe there is an app already done?	11:53
bencoh	a cron for that sounds ugly	11:55
KotCzarny	my first thought was that first i measure how fast it speeds up then fix it once per day	11:55
KotCzarny	and cron is doable, ugly but doable	11:56
kerio	KotCzarny: are you sure the drifting doesn't depend on clock speeds?	11:56
bencoh	clock is speeding, so you'll go back in time everyday	11:57
bencoh	definitely sounds bad :p	11:57
KotCzarny	kerio, im not changing clock speed, at leAst im not using it actively	11:57
kerio	blindly removing two minutes for each week doesn't seem like a good idea	11:57
kerio	you are changing clock speed all the time	11:57
kerio	between 0 and 500	11:57
KotCzarny	kerio, but it's quite consistent that i have to fix the clock manually by 1-2 minutes every 2-3 weeks	11:57
kerio	ok	11:57
KotCzarny	that's why fixing it via cron is better than me having to click it through menus	11:58
KotCzarny	and since it's required periodically...	11:58
KotCzarny	otherwise, is there any tool to get time via gsm network?	11:59
kerio	there should be something like that already in the settings	11:59
kerio	but it depends on your network	11:59
kerio	it doesn't work at all for me	11:59
kerio	(h3g italy)	11:59
KotCzarny	i'm on 'plusgsm', but automatic time never worked for me	12:00
KotCzarny	hrm	12:00
KotCzarny	weird, it worked when i enable it right now	12:00
kerio	you could use gps	12:00
KotCzarny	maybe network got some upgrade	12:00
KotCzarny	gps fix is.. well.. slow?	12:00
KotCzarny	;)	12:00
kerio	but free	12:01
kerio	and it works even in the middle of the sahara desert	12:01
KotCzarny	nothing is free, you pay in battery charge	12:01
bencoh	definitely	12:01
*** mhlavink has quit IRC		12:01
KotCzarny	still, weird as it is, it got time from gsm network now	12:01
bencoh	gsm network clock update doesnt seem to work for me either	12:02
bencoh	(france sfr)	12:02
*** mhlavink has joined #maemo		12:05
KotCzarny	hrm	12:07
KotCzarny	weird there is no tool to get gsm network time/date	12:07
kerio	it's not something you can poll for, i think	12:07
*** jmlich has quit IRC		12:07
KotCzarny	AT+CCLK?	12:07
kerio	i dunno	12:08
KotCzarny	it's not a question	12:08
kerio	well	12:08
KotCzarny	it's the command that modems can use	12:08
kerio	sudo pnatd	12:08
*** janemba has quit IRC		12:12
*** janemba has joined #maemo		12:12
*** jmlich has joined #maemo		12:12
DocScrutinizer05	KotCzarny: there's no cron on maemo	12:13
KotCzarny	doc: that's easily fixable	12:13
DocScrutinizer05	check 'alarmed'	12:13
kerio	*alarmd	12:13
kerio	oh no you mean the addition	12:13
*** arcean has joined #maemo		12:13
KotCzarny	i can even write something like: while [ 1 ]; do fix_clock; sleep 1209600; done	12:14
KotCzarny	and run it once per startup	12:14
*** jmlich has quit IRC		12:16
DocScrutinizer05	AT+CCLK probbaly reads the modem's internal cmos clock, not the network's notion of time	12:16
*** LauRoman has quit IRC		12:17
KotCzarny	could be	12:17
KotCzarny	adjtime - correct the time to synchronize the system clock	12:17
KotCzarny	found it	12:17
KotCzarny	standard linux command to fix time	12:17
DocScrutinizer05	network sends time/date via some sort of broadcast	12:17
*** jmlich has joined #maemo		12:17
DocScrutinizer05	I stronly deprecate use of adjtime	12:18
KotCzarny	what do you propose instead?	12:18
kerio	ntp :)	12:18
KotCzarny	mind you, i don't want to pay for network usage	12:18
DocScrutinizer05	indeed, but first of all check what's your system's CMOS clock in relation to system clock	12:18
KotCzarny	not to mention it would have to enable network, fix time, disable network, and run periodically	12:19
KotCzarny	n900 has cmos clock?	12:19
KotCzarny	since when?	12:19
KotCzarny	:)	12:19
DocScrutinizer05	sure	12:19
KotCzarny	every time i remove battery time is lost	12:19
DocScrutinizer05	it even has power-up on cmos alarm	12:19
*** lbt_ is now known as lbt		12:20
DocScrutinizer05	you can set an alarm in maemo osso-clock and then switch down the device. Will power up in time to make alarm yell	12:20
kerio	KotCzarny: it's because the backup battery is known to fail	12:20
KotCzarny	that i know	12:20
KotCzarny	kerio, then it was failed fresh from factory	12:20
DocScrutinizer05	so how do you think that works without a cmos aka hw clock?	12:20
kerio	KotCzarny: eyup, something like that	12:20
KotCzarny	doc: low power subsystem that's independent from device?	12:21
KotCzarny	but powered by the same battery and connected via gpios ?	12:21
DocScrutinizer05	aka cmos clock ;-)	12:21
KotCzarny	:)	12:21
kerio	it's just the hardware clock	12:21
kerio	as opposed to the software clcok	12:21
kerio	*clokc	12:21
kerio	fuck	12:21
KotCzarny	ok, hwclock should work then?	12:21
kerio	clock	12:21
DocScrutinizer05	and not connected viy gpio but integrated into twl4030	12:21
kerio	yeah	12:21
KotCzarny	kerio: cluck?	12:21
DocScrutinizer05	hwclock should work, yes	12:22
kerio	once you know the correct time, hwclock --sys-to-hw	12:22
kerio	modulo correct option name	12:22
DocScrutinizer05	correct	12:22
DocScrutinizer05	though you don't need that, I guess the maemo settings for time already does it	12:22
KotCzarny	i think it returns same time as system	12:22
KotCzarny	yup	12:23
DocScrutinizer05	man hwclock	12:23
DocScrutinizer05	there's that notorious /etc/adjtime file	12:23
KotCzarny	just checked and time in system changed the same moment as in hwclock -r	12:23
DocScrutinizer05	kernel however has "11 minute mode"	12:24
DocScrutinizer05	again, see man hwclock	12:24
DocScrutinizer05	and man ntpd or somesuch	12:25
KotCzarny	busybox's hwclock is limited	12:25
DocScrutinizer05	make sure you never do a hwclock --sys-to-hw on system shutdown. you only may do this when you just adjusted system time to the correct time	12:26
KotCzarny	i hever did hwclock on n900 before	12:26
KotCzarny	still i have /etc/adjtime	12:27
DocScrutinizer05	when you garble your correction factor in /etc/adjtime file, you will be in for really nasty effects, like time of by centuries on every boot	12:27
KotCzarny	:)	12:27
DocScrutinizer05	garbling that factor is easy when doing hwclock --sys2hc on a system with incorrect system time	12:28
KotCzarny	gotta try hwclock's solution in following month	12:29
KotCzarny	to see if it's viable solution	12:29
DocScrutinizer05	that's why your initscripts MUST NOT use hwclock --sys2hc on shutdown	12:29
DocScrutinizer05	alas most linux distros do exactly that nevertheless	12:29
KotCzarny	do n900's default scripts use hwlock?	12:29
DocScrutinizer05	I dunno	12:29
DocScrutinizer05	prolly not in the deprecated way	12:30
DocScrutinizer05	they will probably use hwclock --hw2sys on boot	12:30
*** thedead1440 has quit IRC		12:31
DocScrutinizer05	or that's all managed by closed blob nokia timed	12:31
*** newbieAlert has joined #maemo		12:33
DocScrutinizer05	err clockd	12:34
*** ult has joined #maemo		12:35
DocScrutinizer05	cat /etc/init.d/hwclock	12:35
DocScrutinizer05	fuuuUUUuuuu...	12:36
DocScrutinizer05	stop()	12:36
KotCzarny	eheheh	12:36
DocScrutinizer05	{	12:36
DocScrutinizer05	if test -x "/sbin/hwclock"; then	12:36
DocScrutinizer05	echo -n 'Synching system time to hardware clock ... '	12:36
DocScrutinizer05	/sbin/hwclock -w	12:36
KotCzarny	gotta disable it too	12:36
KotCzarny	4 years in use and found only now?	12:38
DocScrutinizer05	IroN900:~# hwclock;date	12:38
DocScrutinizer05	Thu Jul 17 11:37:59 2014 0.000000 seconds	12:38
DocScrutinizer05	Do., 17. Jul. 2014 11:37:58 CEST	12:38
DocScrutinizer05	consistently a 1s late systime to cmos clock	12:39
DocScrutinizer05	you could try setting systime via settings and then check if cmos clock got updated too	12:39
KotCzarny	http://213.128.137.28/showthread.php?p=1362546	12:40
DocScrutinizer05	immediately after, 10min after	12:40
KotCzarny	also nice	12:40
DocScrutinizer05	wtf? why IP?	12:40
KotCzarny	dont know, google result	12:40
DocScrutinizer05	never use google results with IP in them	12:40
KotCzarny	if im looking for info i'm now afraid	12:40
DocScrutinizer05	and particularly don't post those links anywhere	12:41
KotCzarny	for software, yes	12:41
*** janemba has quit IRC		12:43
*** janemba has joined #maemo		12:44
*** mhlavink has quit IRC		12:49
DocScrutinizer05	actually, thanks to the broken bupbat, you never should use hwclock --sys2hc with adjusting function of scew factor in /etc/adjtime	12:51
kerio	DocScrutinizer05: why is tmo responding to an ip address?	12:51
kerio	instead of rightfully answering 404?	12:51
DocScrutinizer05	next time you remove battery and adjust time after booting up, you will introduce an adjustment factor of ~14 years per hour	12:52
KotCzarny	doc, i rarely had to remove battery	12:52
KotCzarny	so hwtime could work	12:52
KotCzarny	assuming it gets it's data right	12:52
DocScrutinizer05	then when you shut down your device for 10h and boot it up again, your hw clock will receive an adjustment of 140 years into the future	12:53
KotCzarny	we will see	12:54
kerio	:D	12:54
DocScrutinizer05	now figure you correct those 140 years and hwclock gets to know of it ;-P	12:54
kerio	best phone ever	12:54
kerio	also doubles as a time machine	12:55
DocScrutinizer05	try to find out what the correction factor will be after you turning back system time by 140 years	12:55
KotCzarny	pity ntpdate is in extras-devel still	12:55
KotCzarny	i guess most regular debian packages should be in extras	12:56
KotCzarny	at least those console ones	12:56
DocScrutinizer05	I tell you about correction factor: it involves time between last time you adjusted hwclock and this time you adjust hw clock, and those 140 years difference	12:57
DocScrutinizer05	plus some more math magic	12:57
kerio	KotCzarny: "ntpd -q" does the same thing	12:57
kerio	DocScrutinizer05: can't you just rm /etc/adjtime?	12:58
KotCzarny	hwclock -w creates it	12:58
DocScrutinizer05	not really since it also has TZ	12:58
kerio	no, that's /etc/localtime	12:58
DocScrutinizer05	iirc	12:58
KotCzarny	isnt timezone in /etc/timezone ?	12:58
kerio	it's /etc/localtime that's a symlink to the appropriate zoneinfo file	12:58
DocScrutinizer05	something like that is in rtc/adjfile as well	12:59
kerio	which can be in /etc/zoneinfo, /usr/share/zoneinfo	12:59
KotCzarny	right, zoneinfo	12:59
DocScrutinizer05	I usually end up copying a clean template adjtime file to the real one on each boot	12:59
KotCzarny	erm, /etc/timezone in slackware	12:59
*** aloril_ has quit IRC		12:59
KotCzarny	/etc/TZ in uclibc based things	13:00
KotCzarny	ntpd: not found	13:00
KotCzarny	:P	13:00
DocScrutinizer05	--adjust	13:01
DocScrutinizer05	Add or subtract time from the Hardware Clock to account for systematic drift since the last time the clock was set or adjusted. See discussion below.	13:01
*** mhlavink has joined #maemo		13:01
DocScrutinizer05	as long as you don't use that, everything fine I guess	13:01
DocScrutinizer05	or even always use --noadjfile plus other needed parameters for -utc or whatever	13:03
KotCzarny	btw. hwclock is ran only on boot	13:03
KotCzarny	so it's not the 2min/2weeks b0rking factor	13:04
DocScrutinizer05	hmm?	13:05
KotCzarny	it's not ran via cron or anything	13:05
DocScrutinizer05	cat /etc/init.d/hwclock	13:05
KotCzarny	so it won't fix/change the time on running device	13:05
KotCzarny	unless one would calibrate it and run it periodically	13:06
KotCzarny	which is going back to cron resolution	13:06
DocScrutinizer05	yup, that's why I suggested to check if your RTC is more accurate than sysclock, which quite usually is the case	13:06
KotCzarny	my rtc is synced with system time	13:06
DocScrutinizer05	then you probably have something like "automatic time by GSM" enabled	13:07
KotCzarny	nope	13:07
KotCzarny	could it be that some system package syncs time from rtc periodically?	13:07
DocScrutinizer05	then nothing in system SHALL sync hw clock to sysclock on a regular basis	13:07
KotCzarny	it's your guess or you've checked all the code?	13:08
DocScrutinizer05	from RTC, that may actually be	13:08
DocScrutinizer05	no, that's mere deduction	13:08
DocScrutinizer05	~2119	13:08
infobot	The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119.	13:08
KotCzarny	~rtc	13:09
infobot	hmm... rtc is implemented transparently unless sa1100-rtc is loaded. hwclock unnecessary.	13:09
DocScrutinizer05	actually I think sysclock simply stalls when CPU goes into deep suspend	13:09
KotCzarny	but then rtc should deviate from systime	13:09
KotCzarny	could it be that something syncs time when waking up?	13:09
DocScrutinizer05	which it does for me, by 1s	13:09
DocScrutinizer05	yes, of course	13:10
KotCzarny	1s could be ioport access delay or something	13:10
KotCzarny	as its consistent	13:10
KotCzarny	check in a week to confirm	13:10
DocScrutinizer05	not when the later invoked system time is 1s behind	13:10
DocScrutinizer05	I seem to recall that on a true suspend a linux system reads out RTC, calculates difference to sysclock, and resets the sys clock to RTC+difference on resume	13:12
*** aloril_ has joined #maemo		13:12
DocScrutinizer05	so it's not exactly a sync	13:12
DocScrutinizer05	and for sure no adjustment of RTC	13:12
KotCzarny	s/sync/adjust/	13:12
DocScrutinizer05	but again, that's for true suspend-to-ram etc	13:13
DocScrutinizer05	which maemo isn't supposed to do	13:13
KotCzarny	one would have to check kernel patches for n900	13:13
KotCzarny	n900 uses armel, right?	13:13
DocScrutinizer05	nevertheless maemo / omap3 stops master system clock when CPU goes idle, so obviously sysclock will stall as well	13:13
DocScrutinizer05	armel, right	13:14
DocScrutinizer05	there's aslo an adjust field for sysclock	13:14
DocScrutinizer05	somewhere in kernel	13:14
KotCzarny	that's what hwclock uses probably	13:15
DocScrutinizer05	to make all jiffies and whatnot have better accuracy	13:15
KotCzarny	as adjtime is a linux c call	13:15
DocScrutinizer05	5,2 adjtime yes	13:15
DocScrutinizer05	sth like that	13:15
DocScrutinizer05	but it only operates for drift in master system clockgenerator (28MHz?)	13:16
DocScrutinizer05	so only adjusts sysclock accuracy while sysclock actually running	13:16
DocScrutinizer05	I gather a PM kernel has patches to read out RTC pretty frequently to update sysclock after master-CPU-clock(damn the ambiguity between clock=watch and clock=oscillator) got stopped for a second or an hour	13:18
DocScrutinizer05	KotCzarny: anyway, two facts seem to hold true: nothing in system should adjust the RTC unless you set time manually or from external higher stratum source (e.g. GPS, GSM, ntp)	13:20
KotCzarny	yeah, 'nothing SHOULD'	13:21
KotCzarny	but apparently time drifts noticably	13:21
DocScrutinizer05	and you probably can set up ntp in a way to frequently adjust sysclock to RTC	13:21
DocScrutinizer05	and obviously the /etc/init.d/hwclock stop is BS	13:22
KotCzarny	it only has a chance to b0rk the time on reboot	13:22
DocScrutinizer05	cargo cult coding	13:22
DocScrutinizer05	read what I wrote, about when to expect RTC getting adjusted, and patch /etc/init.d/hwclock stop. Then get a alarmed job that does a hwclock --hctosys once a day	13:24
KotCzarny	offtopic, can i remap eur key to \|	13:24
DocScrutinizer05	~kbd	13:24
infobot	Keytable files and keyboard utilities. URL: ftp://ftp.win.tue.nl/pub/linux/utils/kbd/	13:24
DocScrutinizer05	~leete kbd	13:24
*** mhlavink has quit IRC		13:25
DocScrutinizer05	damn, why can't i recall that factoid name	13:25
KotCzarny	~keyboard	13:25
infobot	methinks keyboard is The main input device on PCs, it consists of all the letters in the alphabet and the numbers 0 through 9. Usually, especially on computer keyboards, there are extra keys like cursor keys and function keys.	13:25
DocScrutinizer05	kerio to the rescue!	13:25
KotCzarny	~remap	13:25
kerio	i wat	13:26
DocScrutinizer05	leete kbd mapping	13:26
kerio	~vikbd	13:26
KotCzarny	btw. infobot is wrong, my keyboard doesn't have all keys from MY alphabet	13:26
kerio	or something	13:26
DocScrutinizer05	sth like that	13:26
kerio	it was vi___'s keyboard layout	13:26
kerio	KotCzarny: your alphabet is wrong, then!	13:26
KotCzarny	localizations hell	13:26
KotCzarny	as usual	13:26
DocScrutinizer05	~listkeys vi kbd	13:27
infobot	Factoid search of 'vi kbd' by key returned no results.	13:27
DocScrutinizer05	~listkeys vi*kbd	13:27
infobot	Factoid search of 'vi*kbd' by key (1): vi-kbd.	13:27
DocScrutinizer05	~vi-kbd	13:27
infobot	rumour has it, vi-kbd is http://wiki.maemo.org/Remapping_keyboard/user_vis_awesome_kbdmapping	13:27
DocScrutinizer05	~#maemo kbd is <reply>see vi-kbd	13:27
infobot	DocScrutinizer05: okay	13:28
DocScrutinizer05	~kbd	13:28
infobot	methinks vi-kbd is http://wiki.maemo.org/Remapping_keyboard/user_vis_awesome_kbdmapping	13:28
DocScrutinizer05	once and for all now	13:28
DocScrutinizer05	;-P	13:28
KotCzarny	you have just removed old info: <infobot> Keytable files and keyboard utilities. URL: ftp://ftp.win.tue.nl/pub/linux/utils/kbd/	13:28
DocScrutinizer05	~_default kbd	13:29
infobot	Keytable files and keyboard utilities. URL: ftp://ftp.win.tue.nl/pub/linux/utils/kbd/	13:29
KotCzarny	hrm	13:29
KotCzarny	i have to recompile kernel to use that map?	13:29
DocScrutinizer05	nope!	13:29
*** clopez has quit IRC		13:29
DocScrutinizer05	:-o	13:29
KotCzarny	good	13:30
KotCzarny	looked like so for a moment	13:30
*** teotwaki has joined #maemo		13:30
teotwaki	'sup	13:30
DocScrutinizer05	>>WARNING! This article assumes you have read the main article on remapping keyboard!<<	13:30
DocScrutinizer05	>>You need to edit this file on your N900: /usr/share/X11/xkb/symbols/nokia_vndr/rx-51<<	13:31
KotCzarny	Do not place the copy in the nokia_vndr folder or it will stop your phone from booting	13:32
KotCzarny	:P	13:32
KotCzarny	this is more important	13:32
KotCzarny	;)	13:32
DocScrutinizer05	the copy	13:32
KotCzarny	yeah, and how editors usually make backups of edited files?	13:33
DocScrutinizer05	you're not suposed to use such editor on system files	13:33
KotCzarny	not true	13:33
*** jon_y has quit IRC		13:33
KotCzarny	vi IS the unix editor for system files	13:33
KotCzarny	though on linux it usually makes no backups	13:34
DocScrutinizer05	and I never seen vi do a filename~ or whatever in hte dir of the editied source by defualt	13:34
KotCzarny	still, i usually make a copy with .orig extension before i start to break things	13:35
*** clopez has joined #maemo		13:35
DocScrutinizer05	tjere's quite a number of dirs in a linux system that are not tolerant to creating bogus files in them	13:35
joga	I do get .file.swp when editing something with vim..	13:35
teotwaki	DocScrutinizer05: DocScrutinizer05 vi creates a .$FILENAME.swp by default	13:36
DocScrutinizer05	that's a pretty bad behavior	13:36
teotwaki	why? You shouldn't use files that begin with . in a glob search.	13:36
DocScrutinizer05	eh??	13:37
DocScrutinizer05	how's that paradigm related to e.g. etc/rc3/* ?	13:37
teotwaki	ls /etc/rc3/*	13:38
teotwaki	doesn't list dotfiles	13:38
DocScrutinizer05	I bet init is pretty happy about that	13:38
teotwaki	init shouldn't load files that begin with a dot	13:38
DocScrutinizer05	shouldn't	13:38
teotwaki	and AFAIK, it doesn't.	13:38
DocScrutinizer05	pff	13:38
DocScrutinizer05	it's generally a poor idea to mess in system dirs with tools you don't completely understand how they work	13:39
*** mhlavink_afk has joined #maemo		13:39
DocScrutinizer05	and I don't think I feel tempted to read the source of whatever kbd mapping related code to know if it will care about /usr/share/X11/xkb/symbols/nokia_vndr/.$FILENAME.swp	13:41
KotCzarny	i thinks its not about filenaming	13:41
KotCzarny	but just scanning files in dir	13:42
DocScrutinizer05	exactly	13:42
teotwaki	if it uses a POSIX wildcard, it won't pick it up.	13:42
KotCzarny	and looking for code blocks that match	13:42
teotwaki	anyway, good chat.	13:45
teotwaki	'later.	13:45
DocScrutinizer05	:-)	13:45
*** zGrr has joined #maemo		13:46
DocScrutinizer05	btw NOKIA_vndr and POSIX? really now? ;-P	13:46
*** N-Mi has joined #maemo		13:46
DocScrutinizer05	anyway thanks for pointing at the general posix rule regarding that. didn't know such rule exists	13:47
zGrr	moin :)	13:48
*** aloril_ has quit IRC		13:50
norayr	hey, people... my second n900's usb eventually broken. can I transfer contacts to n810? (while battery still alive)	13:54
federico3	damn usb connectors	13:54
KotCzarny	noray: bluetooth?	13:54
KotCzarny	wifi?	13:54
KotCzarny	sd card?	13:55
norayr	yes, I can transfer abook.db, but is format compatible?	13:55
norayr	I can scp it.	13:55
KotCzarny	lol, after reboot it shows more battery than before	13:55
norayr	I guess db format is incompatible.	13:56
KotCzarny	export contacts to .vdb ?	13:56
KotCzarny	or .vcard	13:56
KotCzarny	dont remember what is the standard	13:56
norayr	hm, vdb keeps all contacts in one file?	13:57
KotCzarny	what i meant, is there an option in phonebook to export contacts to file?	13:57
KotCzarny	woo-haaah! i have tilde mapped to blue-z and pipe to blue-c	13:58
KotCzarny	now i can type shell commands like a pro!	13:58
KotCzarny	yeah, you can export contacts as .vcard	14:00
KotCzarny	it should be one file	14:00
KotCzarny	but you can also zip it or cat it into one file	14:00
KotCzarny	and import anywhere	14:00
KotCzarny	even on adroid etc	14:00
KotCzarny	unless n810 doesn't know about .vcard	14:01
KotCzarny	then just parse it to the format it knows	14:01
KotCzarny	or look for vcard importer	14:01
*** aloril_ has joined #maemo		14:02
KotCzarny	btw. you should try bluetooth method first though	14:03
*** norayr has quit IRC		14:03
*** norayr has joined #maemo		14:04
norayr	thank you	14:04
KotCzarny	though i cant see to massively send all contacts via bt	14:04
*** norayr has quit IRC		14:04
DocScrutinizer05	that's called sync	14:06
KotCzarny	when i see 'sync' i think that it will delete some data	14:07
DocScrutinizer05	send data, receive data, sync	14:08
*** aloril_ has quit IRC		14:08
DocScrutinizer05	are the tree options to chose from, in "Sync"	14:08
DocScrutinizer05	under settings	14:08
DocScrutinizer05	three*	14:09
KotCzarny	where is this sync option?	14:09
DocScrutinizer05	in settings	14:09
KotCzarny	contacts->settings?	14:09
DocScrutinizer05	no	14:09
KotCzarny	unintuitive!	14:09
DocScrutinizer05	why? how are calendar entries related to contacts?	14:10
KotCzarny	how are calendar entries related to settings?	14:10
DocScrutinizer05	via "Sync"	14:11
DocScrutinizer05	:-P	14:11
KotCzarny	settings relate to app idelogically, not to the data	14:11
KotCzarny	:P	14:11
KotCzarny	that's why it's unintuitive	14:11
KotCzarny	and he clearly meant contacts	14:11
KotCzarny	not calendar entries	14:11
DocScrutinizer05	good enough for me	14:11
DocScrutinizer05	send patches to Nokia!	14:11
DocScrutinizer05	and sorry, there is no option to only sync birthdays of all contacts between 30 and 40 years old	14:13
KotCzarny	bah	14:13
KotCzarny	i never enter birthdays on my phone	14:14
KotCzarny	though i often enter street names in username field	14:14
DocScrutinizer05	ok, we will create a special version for you that doesn't sync birthdays then	14:14
DocScrutinizer05	btw you also could use default backup app to create a set of tarballs that also contain contacts in vcard format	14:16
KotCzarny	it wont be any use for n810 though	14:16
KotCzarny	right?	14:16
DocScrutinizer05	I never tried it	14:16
DocScrutinizer05	it's probably easy enough to extract the plain vcard files from the tarball, to import them on 90% of existing PDA-alike devices	14:18
KotCzarny	but not on the n810	14:19
KotCzarny	as it only allows importing ONE contact at a time	14:19
KotCzarny	:)	14:19
KotCzarny	that's why bt method could be the only viable one	14:19
DocScrutinizer05	hmm, I dunno	14:28
DocScrutinizer05	I never tried it	14:28
DocScrutinizer05	might offer bulk inport as well, either for multiple files or for a concatenated file	14:29
DocScrutinizer05	I never used contacts much on diablo	14:29
teotwaki	My initial n900 still has its usb intact	14:35
teotwaki	no idea how it survived all this time	14:35
KotCzarny	teotwaki: same here	14:35
teotwaki	especially considering it's my girlfriend's main driver	14:36
teotwaki	has been for years now	14:36
KotCzarny	:)	14:36
teotwaki	and she's not the... gentle type when she has to plug it in	14:36
DocScrutinizer51	same here (typing with usb attached is kinda PITA though)	14:36
KotCzarny	btw. debian's armel ntpdate works without a hitch on n900	14:36
KotCzarny	i had to unpack it manually though	14:37
KotCzarny	i hate dependency hell	14:37
zGrr	Linux is a mess.	14:37
DocScrutinizer51	not re!ly	14:38
teotwaki	Linux?	14:38
zGrr	I think so. I think that dependency hell is mostly a linux issue.	14:38
DocScrutinizer51	teotwaki: MXR yours? (sorry I always forget)	14:39
teotwaki	DocScrutinizer51: mxr?	14:39
KotCzarny	zgrr, ehehe, ever seen winsxs on win vista/7 ?	14:39
DocScrutinizer51	ooh then prolly not	14:39
teotwaki	zGrr: you mean the windows or apple way of statically linking everything and their brother is such a "cleaner" way?	14:39
teotwaki	DocScrutinizer51: think we had this discussion already	14:39
teotwaki	DocScrutinizer51: I had maemo-archive.wedrop.it	14:40
KotCzarny	every app on win 7 has it's OWN copy of libraries	14:40
teotwaki	which is dead now	14:40
DocScrutinizer51	timeless: mxr is 'empty'	14:40
zGrr	teotwaki: I don't know Windows, nor Apple.	14:40
KotCzarny	apple is like nokia	14:40
teotwaki	zGrr: so your observation of "Linux is a mess" is based on your knowledge in... what? Exactly?	14:40
KotCzarny	they got freebsd and patched it to hell in their own way	14:40
zGrr	teotwaki: I don't remember having as many problems with FreeBSD.	14:41
teotwaki	Most probably because FreeBSD is still maintained, no?	14:41
zGrr	teotwaki: I work mostly with z/OS, where problem is non existing.	14:41
KotCzarny	well	14:42
teotwaki	isn't z/OS Linux?	14:42
KotCzarny	you mismatch categories	14:42
KotCzarny	bsd is NOT freebsd	14:42
KotCzarny	and linux is not * linux	14:42
KotCzarny	so you are comparing apples to oranges	14:42
teotwaki	ah no, I'm confused with Linux for zServer or something along those lines	14:42
zGrr	teotwaki: I also think that it could be properly addressed in Linux. It's a question of priorities.	14:42
KotCzarny	it's a question of maintainers	14:43
KotCzarny	only	14:43
KotCzarny	:)	14:43
KotCzarny	still, ntpdate binary works	14:43
teotwaki	zGrr: again, you're lobbing "Linux is a mess", as if there is only "One True Way" to manage dependencies. I think Debian's way is good. Fedora/CentOS' isn't bad either. It's when you start having users doing stupid things, such as installing badly packaged software, or software packaged for other versions, adding random repositories, that things go apeshit.	14:44
KotCzarny	it's only packaging that's incompatible	14:44
DocScrutinizer05	http://www.tldp.org/HOWTO/Program-Library-HOWTO/introduction.html	14:44
zGrr	teotwaki: yes, we also have Linux on zServers. Linux on Z is a great example of wonderfuly working virtualization.	14:45
teotwaki	If you run a "clean" Debian install, you'll never encounter dependency issues. Same on CentOS. (Fedora being a bit more bleeding edge might have a few issues sometimes)	14:45
KotCzarny	isn't 'usability' the main point of having soft/hardware?	14:45
KotCzarny	teotwaki: ever tried 'testing' ?	14:45
teotwaki	aye	14:45
teotwaki	never went for sid, though.	14:45
KotCzarny	it borks 100% after a few weeks	14:45
KotCzarny	and constantly mismatches autoremove packages	14:45
* kerio has been running a sid home server for close to three years now		14:45
teotwaki	can't say I encountered that problem, and ran with it for over 3 years as my daily driver	14:45
kerio	literally no problems	14:46
kerio	and i tinkered with it quite a bit, too	14:46
teotwaki	Though I switched to Fedora a few months ago.	14:46
zGrr	teotwaki: yes, you're right. it is a question of a packaging system.	14:46
*** plazmonii has joined #maemo		14:46
KotCzarny	still	14:46
KotCzarny	i like my device working	14:46
*** TheOnlyJoey\|desk has joined #maemo		14:46
KotCzarny	and doing things i want	14:46
TheOnlyJoey\|desk	Ohai	14:46
KotCzarny	even if i break it few times	14:46
kerio	haiyo	14:47
TheOnlyJoey\|desk	using Easy Debian, i am getting a 'chroot: cannot execute su: Exec format error' when trying to boot the chroot image, any clue what could fix that?	14:47
TheOnlyJoey\|desk	on n900 btw	14:47
DocScrutinizer05	maybe ask in #debian ?	14:47
zGrr	teotwaki: one of my colleagues still uses Slackware. He says, he uses Slackware because it's the only Linux distro that doesn't upset him.	14:48
DocScrutinizer05	ooh	14:48
KotCzarny	joey, bad arch?	14:48
* KotCzarny uses slackware		14:48
teotwaki	zGrr: well, whatever "works", really.	14:48
TheOnlyJoey\|desk	KotCzarny, not that i am aware of	14:48
KotCzarny	i have installed 4.0 many years ago	14:48
KotCzarny	still use the same installation	14:48
KotCzarny	just do pkg updates sometimes	14:48
KotCzarny	good luck on debian with that :P	14:49
teotwaki	zGrr: I love how stable Debian is, even on testing. These days, I'm on fedora because I need the newer software, and I'm running hardware that's not quite supported just yet.	14:49
DocScrutinizer05	I never used easydeb so esxcuse my ignorance: boot the chroot image??	14:49
KotCzarny	for hardware you just need custom kernel, right?	14:49
teotwaki	KotCzarny: I guess, couldn't be bothered after a while.	14:50
TheOnlyJoey\|desk	hmm, is the most recent kernel for the n900 2.6.28-omap1 on armv71?	14:50
DocScrutinizer05	basically yes	14:50
KotCzarny	nope	14:51
* KotCzarny has 2.6.28.10-power52		14:51
DocScrutinizer05	that's same version	14:51
KotCzarny	is it?	14:51
DocScrutinizer05	2.6.28	14:51
KotCzarny	.10	14:51
KotCzarny	it could be few heaps of bugs better	14:52
Wizzup	I don't know how easy debian does it, do they use loop mounts?	14:52
TheOnlyJoey\|desk	hmm the image i am trying to boot is armhf, should that be compatible?	14:52
KotCzarny	no	14:52
TheOnlyJoey\|desk	Wizzup, well the mounting goes allright, indeed using loop	14:52
KotCzarny	armel for n900	14:52
KotCzarny	unless you use custom kernel	14:53
TheOnlyJoey\|desk	KotCzarny, ah so thats my problem then -_-	14:53
KotCzarny	and in chroot you deefinitely need armel	14:53
DocScrutinizer05	TheOnlyJoey\|desk: again: BOOT?	14:53
TheOnlyJoey\|desk	DocScrutinizer05, boot?	14:53
DocScrutinizer05	I don't think you boot a chroot	14:53
KotCzarny	:)	14:54
KotCzarny	well	14:54
Wizzup	You don't "boot" a chroot, you execute a program in it with a root changed	14:54
TheOnlyJoey\|desk	meh symantics	14:54
KotCzarny	i can imagine booting a chroot	14:54
KotCzarny	all it does is mounting few shortcuts and starting few services	14:54
DocScrutinizer05	aahyes, you can execute init in a chroot	14:55
TheOnlyJoey\|desk	hmm so it seems my options are using a outdated squeeze or a outdated wheezy	14:55
KotCzarny	easygentoo?	14:56
KotCzarny	:>	14:56
TheOnlyJoey\|desk	KotCzarny, i prefer something working :)	14:56
KotCzarny	could be hardcore though	14:56
KotCzarny	you can always build system from a scratch	14:56
KotCzarny	all you need is few basic packages	14:56
KotCzarny	or just grab some preprepared ramdisk	14:57
KotCzarny	and butcher it	14:57
TheOnlyJoey\|desk	meh going for the 2012 wheezy image then	14:57
*** disco_stu has quit IRC		14:58
*** disco_stu has joined #maemo		14:58
KotCzarny	as for rebuilding, you can run distcc in a scratchbox on some powerfull machine	15:00
kerio	Wizzup: well, you "boot" it by mounting the appropriate stuff and running eventual daemons you want	15:01
DocScrutinizer05	or running init?	15:02
Wizzup	Possibly, I only ever mount /proc /sys /dev and just start bash.	15:02
Wizzup	I don't need X etc from it	15:02
DocScrutinizer05	:nod:	15:02
Wizzup	KotCzarny: I was going to put a gentoo chroot on an sd card	15:02
Wizzup	Since I have ready-built binaries anyway	15:02
Wizzup	(gentoo on sd in the n900, obviously)	15:03
KotCzarny	:)	15:03
* DocScrutinizer05 idly wonders if next kernel update on his PC could use kexec instead of a reboot ;-P		15:04
KotCzarny	i pity that nokia didn't opt for using plain x11 with gestures to add touch functionality (system menus, taskbar etc)	15:04
DocScrutinizer05	but I guess that will nuke all process and filehandle tables	15:04
KotCzarny	yup	15:04
KotCzarny	kexec is starting anew	15:04
KotCzarny	you just skipp bios phase	15:05
DocScrutinizer05	:nod: ;-)	15:05
KotCzarny	k, time synced, we will see how much i have to change time in a week	15:06
*** tanty is now known as tanty_off		15:18
*** Snafu777 has joined #maemo		15:25
*** TheOnlyJoey\|desk has quit IRC		15:47
*** erlehmann has joined #maemo		15:47
*** tanty_off is now known as tanty		15:49
*** b1101 has joined #maemo		15:51
*** erlehmann has quit IRC		15:55
*** jon_y has joined #maemo		16:01
*** DrCode has quit IRC		16:12
*** DrCode has joined #maemo		16:13
*** aloril_ has joined #maemo		16:16
*** newbieAlert has quit IRC		16:19
*** AndChat\|300564 has joined #maemo		16:20
*** Kabouik has quit IRC		16:26
*** japa-fi has quit IRC		16:34
Maxdamantus	init sometimes doesn't want to run in a "chroot"	16:35
teotwaki	DocScrutinizer05: ever tried ksplice?	16:36
KotCzarny	only if it needs to be pid 1	16:36
Maxdamantus	particularly, some init systems behave differently or refuse to work if pid != 1	16:36
teotwaki	or kgraft for that matter	16:36
Maxdamantus	Those supposedly only work for particular changes.	16:39
* Maxdamantus hasn't looked into how they actually work.		16:39
Maxdamantus	but it has to do with applying patches that actually can be applied, to the running kernel.	16:39
KotCzarny	but most of the time init is superfluous on a working system	16:40
KotCzarny	ie. chrooted one	16:40
Maxdamantus	much of init systems seems to be superfluous.	16:40
teotwaki	Maxdamantus: I'm not quite sure about that	16:42
teotwaki	Maxdamantus: I know we've been running ksplice in production for a few years, no matter the kernel change, you just need to spend a bit of (CPU) time diff'ing the running kernel and the new one.	16:43
teotwaki	I just have absolutely no first hand experience, and was wondering from a personal computer standpoint if there was any point or not at all.	16:43
Maxdamantus	teotwaki: but you probably can't just use arbitrary kernels.	16:43
teotwaki	Maxdamantus: define arbitrary kernels?	16:44
Maxdamantus	teotwaki: switch to a kernel specified by something in the mainline tree.	16:44
Maxdamantus	specified by a mainline tree*	16:45
Maxdamantus	At best, you'll be able to apply parts that are flagged as applicable.	16:45
KotCzarny	k, bye folks	16:45
*** KotCzarny has left #maemo		16:46
Maxdamantus	You can't just say, replace the running ext4 driver with a new one.	16:46
Maxdamantus	without remounting the filesystems.	16:46
Maxdamantus	unless something says the changes are applicable.	16:46
*** florian has quit IRC		16:47
Maxdamantus	It seems to be focused on providing security patches.	16:47
Maxdamantus	not switching between kernels.	16:47
Maxdamantus	“An evaluation against Linux kernel security patches from May 2005 to May 2008 found that Ksplice was able to apply fixes for all the 64 significant kernel vulnerabilities discovered in that interval”	16:48
*** eMHa has quit IRC		16:48
teotwaki	Hmm	16:48
teotwaki	"While kGraft is, by choice, limited to replacing whole functions and constants they reference, this does not limit the set of code patches that can be applied significantly. kGraft will offer tools to assist in creating the live patch modules, identifying which functions need to be replaced based on a patch, and creating the patch module source code."	16:48
Maxdamantus	“For patches that do introduce semantic changes to data structures, Ksplice requires a programmer to write a short amount of additional code to help apply the patch”	16:48
Maxdamantus	So it's fairly ad-hoc.	16:49
teotwaki	So it would appear both kgraft and ksplice do impose limitations on what can or can't be patched.	16:49
*** konelix_ has joined #maemo		16:54
*** LauRoman has joined #maemo		16:56
*** LauRoman has quit IRC		16:57
*** eMHa has joined #maemo		16:58
*** AndChat\|300564 has quit IRC		17:15
*** DrCode has quit IRC		17:16
*** VDVsx has quit IRC		17:17
*** LauRoman\|Mobile has joined #maemo		17:17
kerio	teotwaki: you can't ksplice linux and turn it into hurd	17:18
kerio	unless you do it veeeeery gradually, i guess	17:18
timeless	DocScrutinizer51: yeah, i need to look into that	17:19
timeless	soryr	17:19
timeless	been fairly busy	17:20
*** ssvb_ has quit IRC		17:20
*** jmlich has quit IRC		17:20
timeless	consider the investigation started	17:21
*** DrCode has joined #maemo		17:22
*** LauRoman\|Mobile has quit IRC		17:25
*** Snafu777_ has joined #maemo		17:26
*** troulouliou_dev has joined #maemo		17:29
*** Snafu777 has quit IRC		17:31
*** teotwaki has quit IRC		17:31
*** FIQ has quit IRC		17:31
*** VDVsx has joined #maemo		17:31
*** FredrIQ has joined #maemo		17:31
*** FredrIQ is now known as FIQ		17:31
*** FIQ is now known as Guest61367		17:31
*** teotwaki has joined #maemo		17:38
*** SAiF has joined #maemo		17:39
*** fk_lx has joined #maemo		17:41
*** flynx has joined #maemo		17:41
*** troulouliou_dev has quit IRC		17:47
*** RiD has joined #maemo		17:47
*** Malinux_ has quit IRC		17:49
*** Malinux has joined #maemo		17:49
*** RiDD has joined #maemo		17:53
*** konelix has joined #maemo		17:56
*** goldkatze_ has joined #maemo		17:56
*** janemba_ has joined #maemo		17:58
*** janemba_ has joined #maemo		17:58
timeless	DocScrutinizer51: ok, so, the data directory for mxr seems to be missing	17:59
timeless	we'll see about investigating	18:00
*** ab_ has joined #maemo		18:01
*** RiD has quit IRC		18:02
*** VDVsx has quit IRC		18:02
*** konelix_ has quit IRC		18:02
*** janemba has quit IRC		18:02
*** hurrian has quit IRC		18:02
*** ab has quit IRC		18:02
*** ab_ is now known as ab		18:02
*** b1101 has quit IRC		18:02
*** goldkatze has quit IRC		18:02
*** fuz_ has quit IRC		18:02
*** ludens has quit IRC		18:03
*** ssvb has joined #maemo		18:03
*** e2718 has joined #maemo		18:05
*** VDVsx has joined #maemo		18:07
*** b1101 has joined #maemo		18:07
*** b1101 is now known as Guest47992		18:08
*** fuz_ has joined #maemo		18:11
*** ludens has joined #maemo		18:16
kerio	DocScrutinizer05: you're in the council, lol	18:18
*** Guest47992 has quit IRC		18:21
*** e2718 has quit IRC		18:25
*** bluelupo has joined #maemo		18:29
DocScrutinizer51	timeless: thanks	18:39
DocScrutinizer51	kerio: yeah, I'm a jackass, no?	18:40
kerio	you done fucked up now!	18:41
DocScrutinizer51	eh?	18:42
DocScrutinizer51	please rephrase	18:42
DocScrutinizer51	I done wut?	18:42
DocScrutinizer51	I been almost done with maemo yesterday already, when tmo mod htheb - instead of moving thread with 4 mouseclicks according to my request - went to epic length of writing a 100 char post to explin that I'm pestering mods	18:46
DocScrutinizer51	spamming, actually	18:46
DocScrutinizer51	after jalyst explained me in public and PM that it's my duty to ask mods for such thread move	18:48
DocScrutinizer51	maybe it's not too late to hold the record for the shortest council term participation ever?	18:50
DocScrutinizer51	when I hurry I may manage to step down before juiceme managed to post the official election results	18:52
DocScrutinizer51	would that count as a negative timespan?	18:52
kerio	:D	18:54
kerio	juice already sent the mail to the ML	18:54
DocScrutinizer51	dang!	18:55
*** psycho_oreos has quit IRC		19:01
*** Snafu777_ has quit IRC		19:01
*** Snafu777_ has joined #maemo		19:01
*** e2718 has joined #maemo		19:24
*** mhanne_ is now known as mhanne		19:25
*** SAiF1 has joined #maemo		19:36
*** SAiF has quit IRC		19:38
*** SAiF1 has quit IRC		19:44
*** felipe` has joined #maemo		19:52
*** Guest61367 has quit IRC		19:53
*** Guest61367 has joined #maemo		19:53
*** Guest61367 is now known as FIQ		19:53
*** mavhc has quit IRC		20:00
*** mavhc has joined #maemo		20:04
*** mavhc has quit IRC		20:07
*** mavhc has joined #maemo		20:11
*** pdz- has quit IRC		20:11
*** b1101 has joined #maemo		20:12
*** b1101 has joined #maemo		20:12
*** b1101 has quit IRC		20:14
*** pdz has joined #maemo		20:14
*** zGrr has quit IRC		20:36
*** bugzy has joined #maemo		20:49
*** jaykru has quit IRC		20:57
*** rcg has quit IRC		21:01
*** MikaT has quit IRC		21:12
*** MikaT has joined #maemo		21:13
*** totalizator has quit IRC		21:19
*** totalizator has joined #maemo		21:25
*** jaykru has joined #maemo		21:36
*** flynx has quit IRC		21:38
*** louisdk has joined #maemo		21:41
*** mkaindl has joined #maemo		21:42
louisdk	Has it been discussed to make a mobile version of tmo?	21:46
*** eMHa has quit IRC		21:59
*** eMHa has joined #maemo		22:06
sixwheeledbeast	louisdk:http://talk.maemo.org/showthread.php?t=88475	22:10
*** bluelupo has quit IRC		22:17
louisdk	sixwheeledbeast, the good thing about a website optimized for mobile is that it could be used on other platforms like Sailfish. Sadly that app is only for Maemo/MeeGo.	22:18
sixwheeledbeast	louisdk: I agree a mobile skin would be good. Tapatalk was talked about too. TBH in microB TMO works fine.	22:22
louisdk	sixwheeledbeast, I have to zoom a lot on mobile browser, but is doable :)	22:24
*** e2718 has quit IRC		22:27
*** Mike11 has joined #maemo		22:37
*** pdz has quit IRC		22:39
*** pdz has joined #maemo		22:39
*** flynx has joined #maemo		22:42
*** LauRoman has joined #maemo		22:55
*** sq-one has joined #maemo		22:59
*** b1101 has joined #maemo		23:05
*** b1101 has joined #maemo		23:05
*** RiDD has quit IRC		23:17
Snafu777_	Wtf	23:40
Snafu777_	Ok, so i create my own application	23:41
Snafu777_	usr/share/applications/hildon/x.desktop	23:41
*** sq-one has quit IRC		23:41
Snafu777_	/opt/usr/share/pixmaps/x.png	23:41
Snafu777_	For one application it is done as root:root for chown	23:41
Snafu777_	that application works dandy	23:41
Snafu777_	now that i tried this one as root:root, it doesnt wanna f'in launch	23:42
Snafu777_	what gives	23:42
Snafu777_	How can one permission set for all root work just fine, and the other doesnt	23:42
*** pdz has quit IRC		23:46
*** norayr has joined #maemo		23:48
*** pdz has joined #maemo		23:50
*** Snafu777_ has quit IRC		23:55
*** pdz has quit IRC		23:57
*** pdz has joined #maemo		23:57
*** pdz has quit IRC		23:58

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