**** BEGIN LOGGING AT Thu Jan 12 02:59:57 2012 Jan 12 04:14:41 epic fail thread: http://talk.maemo.org/showthread.php?p=1149263 Jan 12 04:14:58 a thread that links to a blog post which in turn links back to another thread on tmo Jan 12 04:14:59 lulz? Jan 12 08:07:31 hello Jan 12 08:43:54 Hi there, see: http://talk.maemo.org/showpost.php?p=1148886&postcount=668. Is the proposition of sponsoring server for few months still actual? Jan 12 09:37:57 merry christmas, happy new year, and jolly easter Jan 12 09:48:39 i have installed vlc on debian chroot, because i needed better sound and graphic equalizer, but it seems vlc cant connect to the output device to play the sound. how can i solve this issue? Jan 12 10:43:36 crashanddie: HEYA! Jan 12 10:45:17 WTF?! whom I need to set on infobot's ignorelist??? Jan 12 10:45:48 * DocScrutinizer51 glares at estel Jan 12 10:46:39 "hi there" what a nice greeting ?-/ Jan 12 10:49:18 haha Jan 12 10:54:54 DocScrutinizer, what's up :) Jan 12 11:32:01 what's the best way of cloning an existing maemo installation to a new n900? Jan 12 11:32:12 my current one is dying Jan 12 11:33:24 quite a few config tweaks and stuff. would be best if I could just flash it in one go, dd style Jan 12 11:36:25 I remember hearing about a backup tool that gets you almost everything Jan 12 11:59:27 BackupMenu Jan 12 13:05:16 as long as u take care to flash/install same kernel Jan 12 15:39:04 is there an app where we can use our old broken-screen n950's as a remote door video intercom ( streaming video from camera over http to webserver running on desktop,and maybe using deskltop mic throuh html5 to send voice to n900 loudspeakers to "talk back" ??) Jan 12 15:39:12 oops i mean n900's Jan 12 15:39:28 the n950 is sitting in front of me, screen unbroken (knock gorrilla glass) Jan 12 15:40:15 npm: :/ Jan 12 15:40:48 does skype do video? Jan 12 15:41:03 hmm yeah, that would do it too. Jan 12 15:41:05 yes it does Jan 12 15:41:35 Also - there are lots of leetle video screens you could plug into the composite port Jan 12 15:41:38 i guess i could setup a separate skype account on it... right now it's my downstairs skype/gmail ringer Jan 12 15:41:40 the n900 Jan 12 15:42:11 SpeedEvil: yeah i plugged my tv in back when it was working reliablu Jan 12 15:42:13 y Jan 12 15:42:42 i'm phasing out tv Jan 12 15:43:38 but yeah i guess an existing voip solution would do it Jan 12 15:43:40 I do not have a working TV at the moment Jan 12 15:44:01 i been playing w/ html5 being served off the handfset to a browser Jan 12 15:44:05 In principle I could find the digibox, or my dttv adaptor, manufacture an ariel... Jan 12 15:44:12 And get the TV out of the attic Jan 12 15:44:19 now that i can stream from gstreamer over the web, i was thinking to stream from the camera Jan 12 15:44:24 but get_iplayer works for me Jan 12 15:45:19 i have two analog and two digital video encoders as pci/pci-e on my desktop Jan 12 15:45:30 http://dx.com/4-3-lcd-monitor-visual-reversing-vehicles-reverse-camera-ntsc-pal-dc12v-67482 Jan 12 15:45:32 I was meaning Jan 12 15:46:37 "Metro will drive the new magic across all of our user experiences. Metro means that 1+1 really does equal 3." Jan 12 15:46:43 hmm that's an interesting side-use, as i wanted a small crt to hook it to as it is currently located in the kitchen w/ loudspeakers Jan 12 15:46:48 -- Steve Ballmer Jan 12 15:47:07 that's the kind of math they use in their pointer arithmetic Jan 12 15:47:35 "Somebody call Eddie Murphy and Arsenio Hall. Nokia Lumia is Coming to America and it's fit for a prince." Jan 12 15:47:53 (whatever he meant by that) Jan 12 15:48:00 the worst part about the lumia is the freakin browser that can't actually view half the websites and is totally html5 incapable Jan 12 15:48:14 npm: Oh, don't remind me Jan 12 15:48:23 when will firefox/fennec be available :-) Jan 12 15:48:34 npm: Have been setting it up yesterday and still feel like an inanimate object has raped me Jan 12 15:48:59 WinCE version of IE...bleah Jan 12 15:49:02 i'm using it as my main phone to save my precious from wear and tear and crackage Jan 12 15:49:24 npm: Just get a Symbian phone, they are more functional and pleasant to use Jan 12 15:49:55 i will at some point, for my qt ambassador project, except i have to take a bunch of screenshots/videos first and i'm in the middle of developing Jan 12 15:50:12 (for http://code.google.com/p/voicetogoog/ ) Jan 12 15:50:19 No USB storage. No video playback. No applications in the store. No decent web browser. No way to configure that prime color square hell of home screen. Jan 12 15:50:39 i figure if i wait a little bit longer the c7 will have belle on it Jan 12 15:50:44 No decent games, for Tentacled's sake! Jan 12 15:50:52 npm: Anna is fine Jan 12 15:51:04 actually the square colors can be configured Jan 12 15:51:12 npm: Shit, old ^3 is fine in comparison Jan 12 15:51:14 i can't remember how because i don't care Jan 12 15:51:25 npm: Yes. you get to select from 16 tear-my-eyes-out colors. Jan 12 15:51:43 npm: But it is a progress: they are now offering the whole EGA palette rather than CGA! Jan 12 15:52:14 there are some useful apps, and it rounds out the n950 functionality nicely Jan 12 15:52:26 Have I forgotten "no ebook reader"? =) Jan 12 15:52:39 which is why i wish there was a vmware solution (like on those new android announces) to swithc between mango and meego Jan 12 15:52:54 best of both worlds Jan 12 15:52:55 npm: What is missing from n950 that this abomination has? Jan 12 15:53:09 plus you could run mango on an 1g ram phone :-) Jan 12 15:53:32 netflix? Jan 12 15:53:43 Why do you need netflix, anyway? Jan 12 15:53:47 There is torrent. Jan 12 15:53:47 various drm'd media distrubutions Jan 12 15:53:54 Again, torrent. Jan 12 15:53:59 well that's what the n950 can do Jan 12 15:54:26 but i didn't have my heart sink when i dropped the lumia this morning... i was like... "that's what you're here for" Jan 12 15:54:47 it is still a very nicely built device Jan 12 15:55:04 and works very well as my daily phone on a phone-only sim Jan 12 15:55:08 but it basically has no platform. you can't call this bluish square menu a platform. Jan 12 15:55:39 can't even develop for it in any language used by normal people! Jan 12 15:57:37 gotta run... Jan 12 17:49:43 NIN101, You've deleted Your page about cryptsetup and truecrypt tricks, so I'm unable to check it - which module isn't included in kp, Serpent? Jan 12 17:50:12 AFAIK AES is compiled in kernel natively, Twofish via module, and Serpent isn't XTS at all? Jan 12 17:50:49 XTS is just a block cipher mode, AES is compiled into the kernel =y, Twofish as a module =M, and Serpent isn't in power kernel. Jan 12 17:51:49 thanks. Hm, I've found Your post in kp thread Jan 12 17:52:02 that for kp47 You've compiled module for Serpent Jan 12 17:52:19 wasn't it included? Jan 12 17:52:33 as I said, Serpent isn't in powerkernel, and never was. Jan 12 17:52:43 ok, I got it, but why? Jan 12 17:52:50 if You've prepared it? Jan 12 17:53:06 prepared? Jan 12 17:53:11 any ratoonale given vs including, or just Your post got bloated by other posts? Jan 12 17:53:19 compiled, nothing prepared, ask Pali to include it. Jan 12 17:53:40 NIN101 - I've compiled modules for the serpent cipher and the block cipher mode XTS Jan 12 17:53:46 ok Jan 12 17:53:48 thanks Jan 12 17:54:12 xts was more important for truecrypt than serpent. Jan 12 17:54:25 yea I know what xts is Jan 12 17:54:33 also it is known to be slower than AES.... Jan 12 17:54:38 true Jan 12 17:54:49 the funny thing is that while benchmarking speed in truecrypt Jan 12 17:54:55 I know... :-) Jan 12 17:54:56 using it's option to benchmark Jan 12 17:55:02 Serpent got higher. Bug? Jan 12 17:55:20 affected by XTS working for aes or twofish? Jan 12 17:55:29 or it's really faster on our device? :P Jan 12 17:56:06 BTW, You may be sure that Your swap encrypted all the time isn't affecting speed in any way, most of situations Jan 12 17:56:18 Lava_Croft: bonus points for gibberish private wuery to ask for -q Jan 12 17:56:33 200 mhz processor have higher flowthrough with AES than our flash storage can handle Jan 12 17:56:57 so, overclocked or not, encryption on the fly doesn't affect our read/write speed at all Jan 12 17:57:07 *unless* device is on 100% CPU usage Jan 12 17:57:32 ~seen Lava_Croft Jan 12 17:57:38 cause encrypting uses CPU, obviously, so fully occupied starts to encrypt/decrypt slower Jan 12 17:57:44 lava_croft is currently on #maemo (2h 14m 40s), last said: 'flasher works fine on windows'. Jan 12 17:58:15 Estel_, doesn't omap have cryptographic HW? Jan 12 17:58:24 omap_aes would be interesting indeed Jan 12 17:58:46 freemangordon, no idea, You tell me? If yes, any access to it? Jan 12 17:58:51 Does it have kernel-support? Jan 12 17:59:18 Hmm, will google these days Jan 12 17:59:29 yea, trying already Jan 12 17:59:43 the problem is that even it if have it, I have no idea how we can use it Jan 12 17:59:49 probably through some kernel module? Jan 12 17:59:57 here my guessworking potential ends :P Jan 12 18:00:16 :D, ok, will check what is there and what not Jan 12 18:00:17 http://maemo.org/community/maemo-developers/n900_aes_and_sha1-md5_hw_acceleration_drivers/ Jan 12 18:00:41 thanks freemangordon ;) Jan 12 18:00:44 DocScrutinizer51: i just go to the source Jan 12 18:00:46 :) Jan 12 18:00:49 thanks Jan 12 18:01:32 NIN101, someone tried those? Jan 12 18:03:18 yes, read the thread ;) Jan 12 18:04:06 would be nice to include it in KP Jan 12 18:04:09 :D yeah, end the conclusion is that those slower that SW Jan 12 18:04:17 lol Jan 12 18:04:18 *those are Jan 12 18:04:27 lovely Jan 12 18:04:29 anyway, I don't care too much about it, it's fast anyway. Jan 12 18:04:45 we always need to care, if it can be faster ;) Jan 12 18:04:54 but, it seems it won't be in this case Jan 12 18:06:05 Well, it is not only about speed, I need to read a bit about what exactly security is there in omap, because if we can securely store keys, it is another kind of beer Jan 12 18:06:23 true Jan 12 18:06:34 also, maybe there is a bug in backporting? Jan 12 18:06:45 there was no conclusion on *why* it's slower Jan 12 18:06:57 could be Jan 12 18:07:25 see NIN101, how much benefits come from simply scp'ing something into -devel. A chain of events ;P Jan 12 18:08:02 BTW, freemangordon, have Your read yesterday's log of CSSU thread? Jan 12 18:08:20 it seem that merlin19* is happy about cooperatiown with X-Fade, sic Jan 12 18:08:34 not that his request was solved Jan 12 18:09:02 he just doesn't seem to be blacklisted by X-Fade as rest of #maemo and X-fade actually responds to him Jan 12 18:09:13 wedll, some people doesn't need much to be happy... Jan 12 18:09:49 anyway, merlin19* doesn't understand why we wqant to move repos out from Nokia's protecting wings, which left me quite... confused? ;) Jan 12 18:09:56 anyway, process it still ongoing Jan 12 18:11:13 Estel_, TBH I would prefer to stay with Nokia if possible, moving away will fragment the community even more Jan 12 18:11:29 Will read logs later Jan 12 18:11:54 so merlin1991 will become CSSU-T maintainer too? Jan 12 18:13:23 Estel_: the thing is nokia is providing the infrastructure and some manpower + the official "maemo" sign on top Jan 12 18:13:40 well wheres MAemo6/harmattan for my n900 then Jan 12 18:13:41 sounds like they aint doing much :P Jan 12 18:13:51 if it aint really broken, do not try to fix it until its time Jan 12 18:13:57 and its not the end of 2012, yet Jan 12 18:14:09 obviously things could be better, but that can be worked one without jumping to a completely new system Jan 12 18:14:40 Lava_Croft: it's also been stated that the cobs / apps.formeego.com stuff might move into maemo.org Jan 12 18:14:59 which in turn would most probably end up in nokia funding beyond end of 2012 Jan 12 18:16:11 merlin1991: yeah, but i talk from current reality, not possibility:) Jan 12 18:16:13 (nod) Jan 12 18:16:20 afaik they only stated for real funding stops at the end of 2012 Jan 12 18:16:28 of course more things are happening in the background Jan 12 18:16:46 Lava_Croft: even that isn't the full reality Jan 12 18:17:00 what was stated is that funding till end of 2012 is there Jan 12 18:17:05 futher stuff has to be decided Jan 12 18:18:43 * DocScrutinizer51 still is amazed about thebig word "funding", given you can get decent root servers for less than 100 bucks per month Jan 12 18:19:17 Lava_Croft, I is better to have "maemo" on top of what we are doing. And really, if apps.meego.com (or whatever the name is) becomes a part of maemo.org, that means funding will be prolonged fot at least 3 more years Jan 12 18:19:21 a single draemhost VPS should handle all the maemo users still left anyways Jan 12 18:19:31 incl gigabit backbone with unlimited traffic Jan 12 18:19:39 yeah Jan 12 18:19:57 Hello ! Jan 12 18:20:15 I would "trade" maemo name for properly contactable repos maintaioner every day Jan 12 18:20:37 being tied with ZNokia ikn *any* way always = Jan 12 18:20:40 That could be solved without going out of maemo.org Jan 12 18:21:00 1) Council without any tools to do anything alone, yet being community representative Jan 12 18:21:12 Lets wait for council meeting, hopefuly some decisions will be made (looking at agenda) Jan 12 18:21:18 2) Nokia replying to everything that their lack manpower due to laid offs Jan 12 18:22:02 3) waiting for every crucial response for at least month, from Nokia Jan 12 18:22:27 4) Nokia paying for service guys for maemo.org + Nokia not controlling it = service guys got us in ass Jan 12 18:22:35 Does someone has ever had problems with the MfE parameters being not saved ? Especially, the parameter which says "I want to sync my calendar"... The checkbox always goes back to uncheck after having synchronized the calendar... Jan 12 18:22:45 I'll stop here, as counting like that is sadening Jan 12 18:22:57 Tofe, never (not using it) Jan 12 18:23:26 Estel_, still there is a chance, lets decide after council meeting Jan 12 18:23:46 of course, I'm not calling for arms and cutlasses ;) Jan 12 18:24:10 And our current problem is only partially with Nokia Jan 12 18:24:22 if Nokia can change way they're supervising support for maemo infrastructure, who knows Jan 12 18:24:24 Tofe: it works fine here, but im not using the feature you describe:< Jan 12 18:24:37 yet, as DocScrutinizer correctly stated, hosting for such small community is cheap Jan 12 18:24:42 yeah Jan 12 18:24:43 I guess MfE settings are stored in the .qmf/qmailstore.db, I could try to look into that... Jan 12 18:25:00 bbl Jan 12 18:25:03 we lose much because deciding people aren't part of community and aren't interested either Jan 12 18:25:12 we gain... 100 bucks for hosting? Jan 12 18:26:04 * merlin1991 wouldn't be able to set up a maemo extras like system with a -devel and stable repo and all the voting and other stuff Jan 12 18:26:29 merlin1991, see "Ask the Council" thread Jan 12 18:26:34 many propositions there Jan 12 18:26:51 btw, thats why it'*s said to mirror infrastructure first Jan 12 18:26:58 my maemo is so frankenstein about 80% of it is from sid. eventualy i''l eliminate the other 20% Jan 12 18:26:59 bleeding edge people would use new one Jan 12 18:27:03 hm Jan 12 18:27:07 i'm trying out tmux Jan 12 18:27:18 and possibly fall back to old if encounter problems, until hammered Jan 12 18:27:35 realisticly someone will show up todo all this, realize 3 months laters that it is work after all and we're back to square 1 Jan 12 18:27:57 IMO it's worth the work Jan 12 18:28:12 well, we can always just sit apatheticaly and die Jan 12 18:28:32 due to developers frustration (see problems with kp and repos, + CSSU testing now) Jan 12 18:28:47 due to Nokia taking plug without preparations from our side Jan 12 18:28:50 etc. Jan 12 18:29:06 Have You seen any project that doesn't require work? Jan 12 18:29:25 it would be sin in any religion and philosophy to waste our community potential. Jan 12 18:29:32 plate: someone should write some script to sid-ify the n900 Jan 12 18:29:33 tmux is interesting Jan 12 18:29:48 Estel_: *if* nokia pulls the plug we can depend on a notice about 3 months earlier Jan 12 18:30:03 yea, so why we can't prepare earlier? Jan 12 18:30:17 in case of relatkions with Nokia going bad, we can switch not waiting for their plug Jan 12 18:30:24 also, 3 months isn't so long Jan 12 18:30:31 for hardening bugs of infrastructure swap Jan 12 18:31:01 also, community is more motivated if *we* take initiative and do something, not because someone pulled the plug above our heads Jan 12 18:31:42 LibreOffice was created because Oracle was bitchy, not because they pulled plug. Yea, it's different scale of userbase etc, but main things are similar Jan 12 18:32:02 except for fact that Oracle was plain hostile, and Nokia is plain unexisting, at least in our universe Jan 12 18:32:10 "ecosystem" ;P Jan 12 18:32:15 as you stated earlier I'm quite happy with the infrastructure, if the situation with the maemo.org maintainer gets better then I see no point at all to move Jan 12 18:32:33 "ecosystem" ;P Jan 12 18:32:43 ask Pali if he's also happy Jan 12 18:32:47 probably like hell. Jan 12 18:33:41 IMO ideas about jumping to Debian infrastructure (they would not notice our resource usage, to be honest - a scale) Jan 12 18:33:46 aren't stupid Jan 12 18:34:02 well, there is no other working alternative to non-FOSS android Jan 12 18:34:12 Mer is still infant. Older, but infant Jan 12 18:34:19 webOS - hahaha! Jan 12 18:35:04 it's just that sticking with semi-working, yet comfortable - due to others (Nokia) doing paper work - things is quite a potential waste Jan 12 18:35:11 anyway, I'm really of now Jan 12 18:55:20 http://www.linuxfordevices.com/c/a/News/Tizen-alpha-released/ Jan 12 18:55:25 someone remember how much time we have to council meeting? Jan 12 18:56:32 can't check myself now - browser port blocked and I don't have time to tunnel it :P Jan 12 19:46:33 ...Dear Tentacled One, may Your sleep be eternal and Your wrath terminal! Please, hunt down all GNOME3 developers and devour them in the most painful way possible! While you are at it, swallow Lennart, just as an appetizer. Thanks. Jan 12 19:47:41 I sense some bitterness in you ;) Jan 12 19:48:03 mgedmin: I have been forced to upgrade from Lucid to Precise Pangolin Jan 12 19:48:23 who forced you? Jan 12 19:48:26 before the tentacled one eats them, let me whip them for a while Jan 12 19:48:31 mgedmin: So, instead of Gnome2, I now have a choice between Unity and GNOME3. Jan 12 19:48:40 mgedmin: QEmu did Jan 12 19:48:41 that must be fun Jan 12 19:48:52 I tried both on oneiric Jan 12 19:48:58 QtSDK forced me to use Unity ;) Jan 12 19:49:19 (when the buggy Qt theme or whatever eats all the memory with huge pixmaps, Mutter crashes. Compiz copes.) Jan 12 19:49:20 mgedmin: About as much fun as choosing between dog shit and cat shit for your breakfast Jan 12 19:49:32 this new year gentoo scared the hell out of me by putting a news item saying that gnome3 was unmasked Jan 12 19:49:47 "would you rather type with your left hand chopped off or your right hand chopped off?" Jan 12 19:49:52 fortunately, it is still not in stable so I'm happy =) Jan 12 19:50:05 To polish things up, i got acquainted with Lumia 800. Jan 12 19:50:09 unity is bearable, just... buggy... Jan 12 19:50:32 mgedmin: both are incredible shit. I have got Xfce now, but it is not fully integrated with Ubuntu Jan 12 20:06:10 in the meanwhile, MS already collecting royalties for 70% of android shipments in the US Jan 12 20:06:45 when will they just drop WP7 (and Nokia) and instead call themselves "The Android Company" Jan 12 20:07:57 What gives them the right to collect royalties regarding 70% of Android base mobile devices within the US? Jan 12 20:08:02 And who made that claim? Jan 12 20:08:40 And what makes that other 30% special enough to not fall under the same jurisdiction? Jan 12 20:08:46 the source is MS itself: http://www.microsoft.com/Presspass/press/2012/jan12/01-12LGPR.mspx Jan 12 20:10:20 basically, since all of the current andridiot manufacturers did deals with the devil in the past (all of them made winmob devices at some stage), the devil is now coming back and grabbing them all by the balls Jan 12 20:10:45 In short, how does this Microsoft LG deal influence Google's Android operating system regarding royalties? Jan 12 20:11:18 As far as I am concerned, LG has shit to say over something Google has created. Jan 12 20:12:01 .oO( ? ) Jan 12 20:13:15 read the press release. it's plain extortion -- but they're trying to do it to everyone Jan 12 20:14:25 I can not really tell anything from this agreement, for all I care this was done voluntarily to maximise profits, if this news is not presented from a neutral source, and not the source being party held within the news article, there is little I can judge over. Jan 12 20:15:08 Receiving coverage over a patent portfolio. Jan 12 20:15:38 Meaning, you can use the patents the company holds, which has its portfolio covering you, to sue other companies in violation of it? Jan 12 20:15:42 Or what does this mean? Jan 12 20:15:53 it means the patent system is broken Jan 12 20:16:09 and lets large companies extort other large companies over trivial patents Jan 12 20:16:10 I need not an explanation on the patent system. Jan 12 20:16:13 I need an explanation on this. Jan 12 20:16:57 To me this just means that Samsung, LG, etc... have made deals with Microsoft, so that they can sue other companies that are in violation of Microsoft's patents, within their own products. Jan 12 20:17:44 noi Jan 12 20:18:00 Then what does it mean? Jan 12 20:18:09 "coverage under Microsoft's patent portfolio for tablets, mobile phones and other consumer devices running the Android or Chrome OS platform" Jan 12 20:18:24 Having made a deal with a company so that your product is covered by their patent portfolio. Jan 12 20:18:25 meaning, licenses for all the patents MS believes are in any way related to any of those Jan 12 20:19:05 That last line of yours is confusing. Jan 12 20:19:20 Still do not get. Jan 12 20:19:57 great because I do not get what's the problem :) Jan 12 20:20:04 I still think it just means you are allowed to use Microsoft patents upon your own devices and sue those who violate within your own products. Jan 12 20:20:13 Well, knowing that. Jan 12 20:20:52 How can you tell Microsoft is receiving royalty rights? The news article does not specify anything besides a cryptic "mutually beneficial agreement". Jan 12 20:21:01 From that, how were you able to tell Microsoft was receiving royalty rights? Jan 12 20:21:30 what other kind of agreement can be done? Jan 12 20:21:36 Many. Jan 12 20:21:45 Patents for patents. Jan 12 20:21:50 crosslicensing Jan 12 20:22:01 Microsoft trying to exhort psychological effect on the market with a certain idea. Jan 12 20:22:05 but what's the difference? Jan 12 20:22:17 Royalty fees = you just cash in money per sold product. Jan 12 20:22:29 Patents for patents = sell patents or give permission to use. Jan 12 20:22:38 the problem as we usually see it is ms collecting anything on trivial stuff Jan 12 20:23:12 payment being in physical money, forced crosslicensing or in human souls is not important Jan 12 20:23:58 Now you are just taking prejudices and general hate towards the patent system, in this deal, to come to extreme conclusions. Jan 12 20:24:07 I do not like the patent system and would just as well like to see it pretty much gone too. Jan 12 20:24:14 But you can not just judge like that. Jan 12 20:24:15 * javispedro gtg Jan 12 20:24:48 All this article says is "these companies can now use Microsoft patents to cover on their own Android based devices, gathered within a mutually beneficial deal". Jan 12 20:28:39 Ehehehe http://gizmodo.com/5875563/these-should-be-the-real-logos-of-nokia-starbucks-playboy-apple-and-dunkin-donuts Jan 12 20:32:45 lol!! Jan 12 20:33:02 mcdiabetes Jan 12 21:11:59 Hello Jan 12 21:13:11 I’m having troubles with my N900: the file copying speed between the SD card and the internal memory is incredibly slow (less than 1 MiB/s), is that normal? Jan 12 21:13:16 ive seen articles citing payments on a per-handsebasis to MS Jan 12 21:13:27 it may be in their SEC filing quarterlies, who knows Jan 12 21:13:31 theres definitely extortion going on :) Jan 12 21:13:44 with teh full backing of the glorious patent system, courtes, etc Jan 12 21:14:29 * plate always has extremely slow flash performance on n900. emmc internal flash. never tried a SD card Jan 12 21:14:35 ive seen pages that claim SD cards might be faster Jan 12 21:14:57 im lucky if i get 500K/second over usb storage, with stalls now and then Jan 12 21:15:47 ok, I guess I’ll have to be patient then… Jan 12 21:17:15 try SD card if you have one Jan 12 21:17:48 also, try 2.6.37 kernel from rescueOS, perhaps it performs better Jan 12 21:18:07 not sure what is at fault, bus bandwidth, mem controllers, old kernels.. Jan 12 21:20:19 plate: eh what? Jan 12 21:20:49 rescueOS kernel won't work with maemo. end of story. Jan 12 21:20:59 i didnt say maaemo Jan 12 21:21:02 boot into rescueOS, do your copying Jan 12 21:21:10 if the kernel performs any better than that ancient 2.6.28 Jan 12 21:21:32 yeah well.... Jan 12 21:21:37 if youre transferring 30gb worth of stuff onto myDocs it might be worth a reboot to do if its 10x teh speed :) Jan 12 21:22:22 i mean the 'stalls are suspect to me. it will stoip writing anything for 5 secconds Jan 12 21:22:28 then write 50 or 100K, then stall again Jan 12 21:22:29 then it will write 1MB/second continuous for 30 seconds. so random Jan 12 21:23:00 what files are those? Jan 12 21:23:12 mp3s Jan 12 21:23:18 theyre not a ton of 4 K files Jan 12 21:23:24 almost all of them are 5MB a piece Jan 12 21:23:30 there's also the tracker indexer/thumbnailer that are busy eating all the CPU extracting id3 tags from your mp3es as they're being written Jan 12 21:59:04 Minos: known effect - cp runs into swap hell where aggressive swapping out of mempages congests the bandwidth of same controller/interface trying to write same data to final destination on same eMMC chip. Jan 12 22:01:26 Minos: check swappolube and the tmo thread(s) about it - maybe a wrapper around cp command may help, when it just temporarily disables swapping (or reduces swappiness) prior to invocation of real cp binary, and restores all /proc and /sys tweaks to 'normal' after cp finished Jan 12 22:02:24 Minos: I've seen N900 freeze completely on trying to cp a few GB of data between uSD and eMMC Jan 12 22:15:21 DocScrutinizer, Minos. Today I accidentaly turned on internal swap that I havent been using for months, then, started my usual combo of applications. Chromium, LibreOffice, and something else. Device immediately started to freeze, regaining control only once for every 15-30seconds or so Jan 12 22:15:42 then I realized, that eMMC swap was turned out together with uSD swap Jan 12 22:16:02 disabling eMMC swap cleared issues almost instantly (well, swap takes a while to get disabled) Jan 12 22:16:24 using swap on eMMC is *really* a bad idea under high I/O from other eMMC partitions Jan 12 22:17:07 Estel_: are YOU responsible for messing with infobot's onjoin? Jan 12 22:17:31 with swap on uSD and correct swapiness settings, cp of many gigabites archive, or extracting ED .lzma archive (2GB) or apt-get dist-upgrade from ED don't even "scratch" device responsivness Jan 12 22:17:34 wtf? Jan 12 22:17:52 (you might deduce from the fact I have to ask at all, as well as the way I ask, that I'm not amused) Jan 12 22:18:09 for the good start, I'm going to admit that I'm responsible for everything You may want from me. now, what happened? Jan 12 22:18:14 i had to mkdir /home/tmp and bindmount /tmp so rubygems wouldnt run out of memory compiling Jan 12 22:18:48 DocScrutinizer, seriously, I've not touched anything Jan 12 22:18:55 uhuh Jan 12 22:18:59 AFAICT Jan 12 22:19:02 tmpfs 1024 72 952 7% /tmp a megabyte? Jan 12 22:19:25 [2012-01-12 09:43:54] Hi there, see: http://talk.maemo.org/showpost.php?p=1148886&postcount=668. Is the proposition of sponsoring server for few months still actual? Jan 12 22:19:39 hey, I used plain onjoin for You Jan 12 22:19:43 like You taught me Jan 12 22:19:47 but only once Jan 12 22:19:56 is it messing? Jan 12 22:20:03 I teach you a quite different lesson in a minute Jan 12 22:20:07 :-( Jan 12 22:20:19 fine, but first tell me what actually happened Jan 12 22:20:24 i.e. what's the issue Jan 12 22:20:32 [2012-01-12 09:43:54] Hi there, see: http://talk.maemo.org/showpost.php?p=1148886&postcount=668. Is the proposition of sponsoring server for few months still actual? Jan 12 22:20:36 IS THE ISSUE Jan 12 22:20:47 wtf shall I think about that crap? Jan 12 22:21:08 lol, ven if it's not amusing. It's spamming You about that? Jan 12 22:21:19 is that onjoin for everyone? Jan 12 22:21:24 is that to me? is that from you? why can't you simply post it directly, when I'm online 24/7 anyway?? Jan 12 22:21:25 hm, sorry for that - I did it exactly as per instructions You linked me to Jan 12 22:21:39 I've posted it, but You were not online ATM Jan 12 22:21:44 no, only for DocScrutinizer Jan 12 22:21:47 only he sees it Jan 12 22:21:56 BS, I'm ALWAYS online Jan 12 22:22:03 actually, once I joked about voicemail on IRC Jan 12 22:22:12 * ShadowJK 'd used memoserv :P Jan 12 22:22:12 BS everybody sees it Jan 12 22:22:15 no, You were not when I used onjoin Jan 12 22:22:38 well, DocScrutinizer once told me about onjoin command and how to use it Jan 12 22:22:47 actually, he used onjoin to tell me about it Jan 12 22:22:47 obviously not Jan 12 22:22:58 then I reported back, that it's talking to me on every join Jan 12 22:23:11 but without response. Eventually, it stopped Jan 12 22:23:16 so I though issue is fixed Jan 12 22:23:22 and then you thought it's a good idea to do same thing to me??? Jan 12 22:23:34 and used it when DocScrutinizer was offline, no idea if by timeout or what. Not in user list :P Jan 12 22:23:54 while *I* at very least started message with "estel_: ..." Jan 12 22:23:57 well, have used it few times already, and no one complained Jan 12 22:24:08 I started it with DocScrutinizer Jan 12 22:24:19 it seems that bot interpreted it as second receiver Jan 12 22:24:22 I'm exactly 90s offline every 24h Jan 12 22:24:25 thus double message Jan 12 22:24:55 well, luck or lack of, depending on point of view. It wasn't ny goal to cause problems. Jan 12 22:24:59 Estel_: nonsense Jan 12 22:25:02 there' Jan 12 22:25:17 anyway, what's the issue, no way to turn it off? Jan 12 22:25:23 ~die Jan 12 22:25:24 * infobot takes two shots to the head and crumples to the ground, lifeless. Jan 12 22:25:30 s only ONE onjoin for each nick, so you nuked any previous onjoun without even checking Jan 12 22:26:21 well, any way to check other's people onjoins? Jan 12 22:26:32 and there's NO WAY to list onjoin records, not even for one nick, for the different channels Jan 12 22:27:10 as there's a _default onjoin for every nick, plus one chan specific one for each chan the nick joins Jan 12 22:27:17 no Jan 12 22:27:21 well, actually, it was You who told me to use it ;) Jan 12 22:27:30 I haven't messed with default onjoin Jan 12 22:27:39 I *never* told you to use it Jan 12 22:27:39 what You see is only for Your eyes, as in Bond Jan 12 22:27:45 I told you it existst Jan 12 22:28:13 jokes aside, TBH i still don't get what's the issue. it spams You every 2 seconds, or what? Jan 12 22:28:31 I'm simply disabling onjoin for this and a number of other chan, I think Jan 12 22:28:47 take Your time, have fun Jan 12 22:28:57 it's a too powerful cmd for users messing with it indiscriminately Jan 12 22:29:13 care to tell us what's exactly this issue? Jan 12 22:29:31 Estel_: It spams the CHANNEL every day, and I don't even know it's caused by me Jan 12 22:29:46 and the worst thing is YOU don't see all those problems Jan 12 22:30:03 ok, got it. it spams it once a day when You join. Jan 12 22:30:17 how would I tell? Jan 12 22:30:23 actually, spams for one liner isn't bst word. Now, how to disable it? Jan 12 22:30:31 erm Jan 12 22:30:34 my client joins after 90s when my DSL reconnects Jan 12 22:30:38 wasn't infobot talking like Jan 12 22:30:45 I'm not even at home at that time Jan 12 22:30:49 DocScrutinizer, Estel_ wasn't You to know: Jan 12 22:31:00 want* Jan 12 22:31:29 infobot: DocScrutinizer, Estel_ want You to know: Jan 12 22:33:03 well, at least it was the case when You used it @ me. Anyway, I'm off, good night ladies and gentlemans Jan 12 22:33:25 I got no fsckng clue how and what onjoin command got activated for me, and I don't know who activeated it, and the one who did has no clue if there's been an older more important onjoin already set that gets reset by him Jan 12 22:34:19 coool shit, he's leaving me with this fucking onjoin, and I have to clean all possible alternatives now Jan 12 22:34:24 Before I go - ShadowJK, maybe You know what's the sane current limit for charging internals of N900, *not*battery? Jan 12 22:34:38 fuck shit, I'm still here, so I can help if it's possible Jan 12 22:34:46 before you go, you DAMN REMOVE that onjoin you set Jan 12 22:34:48 ! Jan 12 22:34:50 just stop shouting around and tell me what to do Jan 12 22:34:52 ok Jan 12 22:34:55 any command? Jan 12 22:35:09 I asked for a command to remove it for past 10 minutes Jan 12 22:35:26 just concluded that You're on berserker mode and wouldn't tell me anyway ;) Jan 12 22:35:33 I don't know, but at 1250 I thought it heated up more than you'd expect from just interpolating 1250/950 Jan 12 22:35:41 I don't know any command as I don't know the command you used to install it - got it now? Jan 12 22:36:06 ~+help onjoin Jan 12 22:36:06  Desc: get/set OnJoin message (needs chan option +OnJoin) Jan 12 22:36:07  Usage: onjoin [#chan|_default] [-] [message] Jan 12 22:36:07 Example: onjoin infobot Hey! It's another infobot! Jan 12 22:36:16 ShadowJK, thanks. Basically the concept is that using dual-cell battery current is divided by two for each cell Jan 12 22:36:40 so, it *should* be safe theoretically, to charge with 1.5 A Jan 12 22:36:48 if chip etc won't get fried Jan 12 22:37:05 (750mA per cell in ,parrallel cell setup) Jan 12 22:37:05 onjoin [#chan|_default] -DocScrutinizer[51] Jan 12 22:37:28 Well it doesn't go higher than 1250, and that's what I'm trying to say, that when I increased the current beyond what bme uses, it heated up more than I would have expected Jan 12 22:37:47 because battery or chip itself? the latter I asume? Jan 12 22:37:48 (not the battery, the internals) Jan 12 22:37:53 yea Jan 12 22:37:59 I was using a 2400mAh battery at that time Jan 12 22:38:29 so anyway no point in risking for 1250-950= 300mA Jan 12 22:38:44 ~+help botmail Jan 12 22:38:44  Desc: Send someone botmail Jan 12 22:38:44  Usage: botmail {for [:] }|stats|check|read Jan 12 22:38:45 Example: botmail for infobot: you rock! Jan 12 22:38:45 Example: botmail stats Jan 12 22:38:45 Example: botmail check Jan 12 22:38:45 Example: botmail read Jan 12 22:39:14 thanks ShadowJK! Jan 12 22:39:19 DocScrutinizer, I just did Jan 12 22:39:20 After I noticed the heating up, I made my script drop to 550mA when temperature rose above 50C Jan 12 22:39:20 even that sucks Jan 12 22:39:29 make it explode!! Jan 12 22:39:32 Eventually I just settled for charging at 950 without temperature throttle Jan 12 22:40:04 I wonder what part gets hot so fast Jan 12 22:40:11 due to 300mA difference Jan 12 22:40:16 DocScrutinizer, I just did Jan 12 22:40:27 ~onjoin DocScrutinizer blablabla Jan 12 22:40:28 ok, Estel_ Jan 12 22:40:36 so maybe Jan 12 22:40:42 ~onjoin -DocScrutinizer blablabla Jan 12 22:40:42 Estel_: ok Jan 12 22:40:50 ~onjoin -DocScrutinizer Jan 12 22:40:50 Estel_: ok Jan 12 22:41:06 Well there's known parasitic/undocumented resistance in the path between charging circuit and battery, and we know that the heat is resistance*current*current (current squared), so.. :P Jan 12 22:41:23 DocScrutinizer, he was removing it? Jan 12 22:41:24 ~onjoin -DocScrutinizer Jan 12 22:41:24 ok, DocScrutinizer Jan 12 22:41:34 Well there's known parasitic/undocumented resistance in the path between charging circuit and battery, and we know that the heat is resistance*current*current (current squared), so.. :P Jan 12 22:41:59 thanks for quoting... Jan 12 22:42:17 DocScrutinizer, care to leave berserker mode? I was actua..y clearing onjoin for You Jan 12 22:42:40 infobot: status Jan 12 22:42:40 Since Tue Jan 3 23:54:57 2012, there have been 11 modifications, 348 questions, 0 dunnos, 0 morons and 200 commands. I have been awake for 8d 22h 47m 36s this session, and currently reference 118348 factoids. I'm using about 35500 kB of memory. With 0 active forks. Process time user/system 2580.67/54.35 child 0/0 Jan 12 22:42:58 ~die Jan 12 22:42:58 * infobot takes two shots to the head and crumples to the ground, lifeless. Jan 12 22:42:59 active fork? what so it can "think" for a while? Jan 12 22:43:38 I suspect it forks shitty perl scripts Jan 12 22:43:55 ~+help lobotomize Jan 12 22:43:56 no help on lobotomize. Use 'help' without arguments. Jan 12 22:44:05 anyway, this whole onjoin mess gave me a nice idea for april's fool day Jan 12 22:44:33 does infobot also have channel operator status? Jan 12 22:44:39 ~+help lobotomy Jan 12 22:44:40 I can be given a lobotomy ([o] is required) if people start to abuse me. To bring me back to life, give me an unlobotomy Jan 12 22:44:41 Estel_, I don't really care what you do with your own N900, but personally I'm not risking it, considering the poor availability of N900s today, and their habbit of dying mysterious deaths Jan 12 22:45:06 ShadowJK, exactly true Jan 12 22:45:07 1250 sounds like overkill Jan 12 22:45:15 that's why I concluded it's not worth the risk Jan 12 22:45:26 poro, not if you're charging a 2500mAh battery Jan 12 22:45:31 diferrence from 950 to 1250 is only 300 mA Jan 12 22:45:33 from the battery's point of view Jan 12 22:45:37 not worth the risk Jan 12 22:45:43 are we calkin gCPU MHz or charge current? Jan 12 22:45:50 well actually I'm charging 3070 mAh one... Jan 12 22:45:52 im talking OC speeds Jan 12 22:46:04 we're talking charging current Jan 12 22:46:46 I don't have a overclock either. It's funny, when people started overclocking, they were like "Well I don't care if it dies in a year, next year nokia will have a N910 or whatever and I'll get that instead!" Jan 12 22:46:50 That's why I asked first knowledgeable one (ShadowJK ;) ) before trying, actually Jan 12 22:46:56 well, N9 is out Jan 12 22:47:03 well, N9 is a joke Jan 12 22:47:04 and, id imagine its more or less as open as n900 was Jan 12 22:47:04 anyway Jan 12 22:47:11 poro: hahahhhhahahahahhahah. Jan 12 22:47:13 just QT instead of GTK Jan 12 22:47:17 poro, it's way more closed, people are still struggling to use it Jan 12 22:47:21 ok Jan 12 22:47:26 I use 500-900mhz limits rock stable Jan 12 22:47:42 ~chanset Jan 12 22:47:52 Personally, I plan to use my N900 for years Jan 12 22:47:54 mgedmin had a postiive N9 comment on LWN's tizen thread Jan 12 22:48:10 and that blogger the debian-mobile guy linked to liked his Jan 12 22:48:22 If I would pau such insane money for new to,y I would probably also comment it positively Jan 12 22:48:25 poro, I haven't followed it closely, but to me it appears it's more difficult to break/root/hack N9 than most android handsets Jan 12 22:48:40 ouch Jan 12 22:48:44 ~aegis Jan 12 22:48:44 http://www.developer.nokia.com/Community/Wiki/Harmattan:Developer_Library/Developing_for_Harmattan/Harmattan_security/Security_guide , or "The purpose of this framework is: ... to make sure that the platform meets the requirements set by third party software that requires a safe execution environment.", or http://en.wikipedia.org/wiki/Trusted_Computing#Criticism, or http://en.qi-hardware.com/w/images/1/10/ME_382_LockedUpTechnology2.gif Jan 12 22:48:49 he also had a N950, so i doubt he paid for it or anything. Jan 12 22:48:55 gotta suck up to hte "people who send me free stuff" Jan 12 22:49:15 http://2.bp.blogspot.com/-fUgbx4RTpGY/TreyKLXw6aI/AAAAAAAAANc/5wvYTlwG5xo/s1600/freemobile1.jpg Jan 12 22:49:27 I'm sure that N9 is ok for people who only want to do, uh, web stuff? Jan 12 22:49:44 for 600 pounds Jan 12 22:50:02 Big Bad Browser for 600pounds... Jan 12 22:50:04 its 3x the price of n900 available NIB on ebay Jan 12 22:50:10 nope Jan 12 22:50:23 N900 in ideal condition can be bought for as low as 150dollars Jan 12 22:50:31 withnout searching, for 200$ Jan 12 22:51:15 * ShadowJK wonders if there's a summary page for the open/closed and flasher status of N9 Jan 12 22:51:22 wow it has trusted-comuting crap Jan 12 22:51:25 nevermind Jan 12 22:52:10 ~+chanset #maemo Jan 12 22:52:11 chan: #maemo (see _default also) Jan 12 22:52:11 OnJoin => 0, _time_added => 1179878166, autojoin => infobot Jan 12 22:54:03 Estel_: sorry for berserkermode :-D. Would you please test the ~onjoin DocScrutinizer blabla again? Jan 12 22:55:03 ShadowJK: ^^^ ? Jan 12 22:55:19 ? Jan 12 22:55:45 ShadowJK: Would you please test "~onjoin DocScrutinizer blabla" Jan 12 22:55:57 ~onjoin DocScrutinizer blabla Jan 12 22:56:07 thanks Jan 12 22:58:44 ~onjoin DocScrutinizer Jan 12 22:59:19 ~onjoin #maemo DocScrutinizer Jan 12 23:04:25 can you program the bot via msg or something? Jan 12 23:04:44 maybe cuz i have notices/privmsgs ignored i never noticed what the problem was (nor do i ever join/part if i can avoid it) Jan 12 23:04:53 but this relentless configuring has gone on about 5 screens Jan 12 23:05:22 ah he does work via msg Jan 12 23:05:30 it's finished and tested now, so chill a bit Jan 12 23:05:33 :-D Jan 12 23:06:09 ~android Jan 12 23:06:10 i heard android is an Open Handset Alliance Project by Google or an alleged Open Source phone software stack that's really not open. http://code.google.com/android/, or if it's sometime actually freed, someone might port it on the neo, but the provided binaries are incompatible Jan 12 23:06:25 ~openmoko Jan 12 23:06:26 extra, extra, read all about it, openmoko is a joint venture of FIC and the open source community creating a complete open source phone platform stack. see http://openmoko.com Jan 12 23:06:35 wow, where do i get on board! Jan 12 23:06:58 and yes, you can also /query infobot, but *I* can't do all the stuff you can, as some of the stuff always works when _I_ do it ;-D Jan 12 23:07:18 ~query Jan 12 23:07:18 talk dirty to me! Preferably, do so after you have typed "/query infobot" which should open a new window/tab/whatever with most irc clients. You can talk to me all you like and don't annoy other people with endless queries. Be aware that the stuff you write is logged, so don't get too 1337 :) Jan 12 23:08:01 poro: get on board of... what? Jan 12 23:08:13 DocScrutinizer, sorry, I've beebmafk. And, no problem Jan 12 23:08:19 openmoko!! Jan 12 23:08:36 well, openmoko website was quite outdated when I last checked, it's nice that it's alive Jan 12 23:08:40 ive seen like, homebrew GTA04v12 runs by german tweakers Jan 12 23:08:54 have they fixed mic-buzz issue yet heh Jan 12 23:09:06 DocScrutinizer, honestly, objectively - does it have chance to be ready for end-users-geeks anytime in upcoming years? Jan 12 23:09:16 poro: http://projects.goldelico.com/p/gta04-main/ Jan 12 23:09:47 poro: no, *I* have fixed buzz ;-D Jan 12 23:09:51 yeah ok, i lack a GTA01/2 to plug it into Jan 12 23:09:52 ~joerg Jan 12 23:09:52 i guess joerg is a HW-developer and engineer of Openmoko, usually known as DocScrutinizer Jan 12 23:09:55 I'm asking as You seems to be knowledgeable about openmoko Jan 12 23:10:22 ~DocScrutinizer Jan 12 23:10:22 [docscrutinizer] jOERG, a HW-developer and engineer of Openmoko Jan 12 23:10:32 ~Estel Jan 12 23:10:37 booo ;) Jan 12 23:11:47 poro: Nikolaus of Goldelico is aware of the problem. Right ATM "we" are searching for ways to build new spare cases Jan 12 23:12:23 OTOH you probably can find a used GTA02 for ~80 bucks Jan 12 23:12:34 maybe 60 Jan 12 23:12:48 oh fuck, one question ok? Jan 12 23:13:06 I'm using encrypted partition without filesystem, that I've already set to mapper Jan 12 23:13:07 lest say Jan 12 23:13:21 /dev/mapper/truecrypt1 Jan 12 23:13:28 now I want to make filesystem on it Jan 12 23:13:48 but, neither Truecrypt automaticaly, or I manually can do it, due to f*ckin error: Jan 12 23:14:03 /dev/mapper/truecrypt1 is mounted; will not make a filesystem here! Jan 12 23:14:12 hah Jan 12 23:14:21 ~tizen Jan 12 23:14:22 how the holy shit can I mkfs a encrypted partition without erasing encryption? Jan 12 23:14:27 https://lwn.net/Articles/474805/ Jan 12 23:14:37 scratching my head about it for 2 hours Jan 12 23:14:44 Estel_: a good question Jan 12 23:15:16 "with a very interesting answer for it. You can have it for mere 12 billions of credits" Jan 12 23:15:28 sorry, can't stop myself from quoting star control 2 Jan 12 23:15:52 the worst thing is that it seems to work for other people Jan 12 23:16:00 wtf I am doing wrong?... Jan 12 23:17:30 umount /dev/mapper/truecrypt1 Jan 12 23:17:45 umount: can't umount /dev/mapper/truecrypt1: Invalid argument Jan 12 23:17:57 Estel_: either: there's a way to override the test in mkfs so it will create a FS on loopmounted devices (seems that shall work since ages, as normal loopmounts of files as partitions is a common practice), or: your mount has to be so much different to a normal mount that it doesn't look like a mount to mkfs anymore. For that topic mount -o loop seems obsolete Jan 12 23:18:11 mkfs.ext4 /dev/mapper/truecrypt1 Jan 12 23:18:14 hm Jan 12 23:18:25 mke2fs 1.41.3.maemo0 (12-Oct-2008) Jan 12 23:18:35 /dev/mapper/truecrypt1 is mounted; will not make a filesystem here! Jan 12 23:18:38 paranoia Jan 12 23:18:53 DocScrutinizer, will check for force parameters on mkfs Jan 12 23:19:43 losetup? Jan 12 23:20:20 DocScrutinizer, verbose mode pls? Jan 12 23:20:31 man losetup Jan 12 23:20:58 >>... For that topic mount -o loop seems obsolete Jan 12 23:21:11 no manual entry for losetup, will browse for it ;) Jan 12 23:21:22 yea , guessed that, thanks... will try Jan 12 23:25:12 it seems that there is no way to override tests in mkfs eh... checking losetup Jan 12 23:25:26 the thing is Jan 12 23:25:32 that I can't mount it normally Jan 12 23:25:43 even via -o loop or losetup Jan 12 23:25:56 or mkfs will override encryption Jan 12 23:26:09 it *has* to be mounted by truecrypt, so decryption on the fly Jan 12 23:26:30 then it should mkfs without problem, no idea why it complain about it "being mounted" Jan 12 23:27:18 truecrypt "mounts" /dev/mmcblk0px to /dev/mapper/truecrypt1 , asking for passwords keyfiles etc Jan 12 23:27:46 truecrypt "mounts" /dev/mmcblk0px to /dev/mapper/truecrypt1 , asking for passwords keyfiles etceing mounted Jan 12 23:27:52 Estel_: man 8 mount|less "+/THE LOOP DEVICE" Jan 12 23:28:13 thanks, will check... Jan 12 23:29:06 sorry for my ignorance, but how to merge it with truecrypt mounting, that is absolutely necessary... Jan 12 23:29:25 it seems to me you fist loopmount your file/partition to /dev/mycryptdev, then format that with e.g mkfs.crypt /dev/mycryptfs Jan 12 23:29:26 or I end with *non* encrypted partition, that was mkfs'ed Jan 12 23:29:55 wait wait. so should I ommit mapper at all? Jan 12 23:30:25 then umount it and mount again from file/partition to /dev/mycryptfs TO /mst/cryptstore, according to man 8 mount|less "+/THE LOOP DEVICE" Jan 12 23:30:50 can anyone here recommend a phone that will work great with linux/maemo products that will also work with the sprint network ? Jan 12 23:31:02 i really wish i could go with the nokia n900 but it will not work with sprint... Jan 12 23:31:22 DocScrutinizer, thanks, it seems sane Jan 12 23:31:31 yet, one problem Jan 12 23:31:47 /dev/truecrypt1 dissapears as soon as truecrypt mount is disabled Jan 12 23:32:00 >> it seems to me you fist loopmount your file/partition to /dev/mycryptdev, then format that with e.g mkfs.crypt /dev/mycryptfs Jan 12 23:32:38 don't have /dev/truecrypt1 to mount my encrypted partition to. Jan 12 23:32:59 or I'm missing something? Jan 12 23:34:59 so does anyone have any recommendations for a phone that will work with sprint that i could install linux on and maemo products? Jan 12 23:35:01 losetup seems to be able to mount a file to /dev/CryptedLoop1, without same time mounting /dev/CryptedLoop1 to /mst/MyCryptedStore Jan 12 23:35:10 i would apprecaite any advise you guys might have... Jan 12 23:35:32 DocScrutinizer, yes, just a second, as I seem to miss the obvious Jan 12 23:35:58 how can I mount a partition or file to /dev/CryptedLoop1 without /dev/cryptedLoop1 Jan 12 23:35:59 mkfs should format a /dev/CryptedLoop1 Jan 12 23:36:03 existing Jan 12 23:36:18 losetup Jan 12 23:36:24 yes, but to cryptedloop1 to exist, it need to be mounted by truecrypt Jan 12 23:36:34 or losetup can "create" /dev/? Jan 12 23:36:42 I'd guess it can Jan 12 23:37:03 it's the purpose of losetup, according to manpage Jan 12 23:37:10 will try it, but what wonders me, is why the fuck it works on desktop linux and, reportedly, on other people's devices without so much hassle Jan 12 23:37:26 Setup loop device: Jan 12 23:37:28 losetup [{-e|-E} encryption] [-o offset] [--sizelimit limit] Jan 12 23:37:29 [-p pfd] [-r] {-f[--show]|loopdev} file Jan 12 23:37:39 NIN101 tried it fwq hoursd ago and he was just able to mkfs.ext4 /de//truecrypt1 so I wondeer wtf Jan 12 23:37:44 ah Jan 12 23:37:48 missunderstood that Jan 12 23:37:52 ok, will try Jan 12 23:38:39 dev* not /de// of course >> (...) he was just able to mkfs.ext4 /dev/truecrypt1 (without hassle) Jan 12 23:41:33 damn, it's beyond my understanding Jan 12 23:41:52 even when I suceed at creating /dev/ Jan 12 23:41:56 it Jan 12 23:42:09 *isn't* properly decrypted without truecrypt Jan 12 23:42:25 and if I truecrypt mount loopback device Jan 12 23:42:28 same story Jan 12 23:42:36 won't mkfs it due to it being mounted Jan 12 23:42:56 losetup can't decrypt it without truecrypt that provides passwords and keyfiles. Jan 12 23:43:26 I never looked into truecrypt - would expect they have a decent wiki page that deals with "how to format a raw file to hold a truecrypted FS" Jan 12 23:43:37 heh, they have Jan 12 23:43:48 mkfs.* /dev/mapper/truecrypt1 Jan 12 23:43:59 it should damn work out of the box Jan 12 23:44:04 that's the thrill. Jan 12 23:44:16 On desktop ikt works. NIN101reported that it also works for him Jan 12 23:44:23 maybe your kernel isn't up-to-date Jan 12 23:44:58 TrueZCrypt can also mkfs it just after creation, automaticaly. And, while trying that, ekfs2 complained about it being mounted Jan 12 23:45:09 well, NIN101 uses same kernel-power Jan 12 23:45:18 it seems, that only my mkfs complain Jan 12 23:45:21 quite obviously encrypted FS is using kernel level FS drivers Jan 12 23:45:31 yes, properly loaded Jan 12 23:45:38 it *is* decrypted properly Jan 12 23:45:46 if i provide wrong password, it complains Jan 12 23:46:00 with good password, it mount it properly as raw partition Jan 12 23:46:13 the only thing that doesn't seem to work as it should is damn mkfs Jan 12 23:46:28 pragmatic solution: copy the already formated "empty" raw file to N900 Jan 12 23:46:32 from PC Jan 12 23:46:34 tried mkfs.ext2, *.ext3, *.ext4 Jan 12 23:46:45 no suggestions for any other phones for linux and maemo products other than n900? Jan 12 23:46:47 can't do, it's a partition, not file Jan 12 23:46:59 file is utterly useless on wear-leveled storage Jan 12 23:47:01 dd then :-D Jan 12 23:47:10 right Jan 12 23:47:14 that's the option :P Jan 12 23:47:39 gh0strider, You wont't use "maemo products' on non Maemo phone Jan 12 23:47:43 especially on android Jan 12 23:47:57 and wtf is that springnet You need to use so badly? Jan 12 23:48:20 Estel sprint network Jan 12 23:48:42 whatever Jan 12 23:48:52 is there any maemo phones that might work with sprint? Jan 12 23:48:53 is it some closed propertiary p2p? Jan 12 23:49:04 ~sprint Jan 12 23:49:04 Sprint is a company that promises you the blue sky yet fails to deliver everytime. Do yourself a favor and don't call 1-800-PIN-DROP. Jan 12 23:49:13 ;) Jan 12 23:49:40 may I ask You why You need sprint network? Jan 12 23:50:19 Estel_: i signed a 2 year contract with them, i really did not want to go with sprint but my girl's sister works for sprint so we signed up with her on a 2 year contract... Jan 12 23:51:01 what a meritocratic reason ;) Jan 12 23:51:03 anyway Jan 12 23:51:03 @ghostrider: better question is "do any maemophones have a CDMA chip" ? Jan 12 23:51:22 AFAIK there *is* solution to use such strange wifi provider in N900 Jan 12 23:51:29 LOL Jan 12 23:51:36 Hurrian: There are two public maemo phones. Jan 12 23:51:40 You're talking about wifi, no CDMA, yep? Jan 12 23:51:42 Hurrian: The n900, and the n9. Jan 12 23:51:46 Hurrian: That's it. Jan 12 23:51:49 N9 isn't maemo Jan 12 23:51:50 giving the money directly to her would've been a better ratio tahn paying for sprint to give her 0.5% provision Jan 12 23:51:55 Estel_: debatably Jan 12 23:52:00 N9 is unknown harmattanish object Jan 12 23:52:13 There are up to two then, Jan 12 23:52:21 sucks for americans to have the best networks be CDMA and not GSM like the rest of the world Jan 12 23:52:26 i know i can order a n900 but i would like to get a phone that i could use and install linux on it and since i am with sprint i need a phone that will work with sprint... Jan 12 23:52:46 gh0strider, I don't know any phone with cdma any way, but americans got some, probably Jan 12 23:53:02 buy a tablet + cdma modem? Jan 12 23:53:07 or even better, network? Jan 12 23:53:09 tfu Jan 12 23:53:12 netbook Jan 12 23:53:21 gh0strider: you can't install maemo on a phone, even if you can instlal linux. Jan 12 23:53:40 or, buy a cdma modem that got linux driversand use it with N900 via hostmode? Jan 12 23:53:58 ya i have been using linux for awhile but i am brand new to this whole mobile linux phone area... Jan 12 23:54:06 otherwise, SpeedEvil is totally true so no joy for You Jan 12 23:54:07 @gh0strider lots of android phones have CDMA versions. if you want, grab the cheapest crapola android phone with wifi tethering Jan 12 23:54:23 ...and tether wifi to N900 Jan 12 23:54:24 i think i am going to have to go with an android phone Jan 12 23:54:27 would also do Jan 12 23:54:33 i think they have some ubuntu android versions out there... Jan 12 23:54:44 no, they have not ;) Jan 12 23:55:00 android != GNU Jan 12 23:55:03 only linux kernel Jan 12 23:55:09 and that is even changing lately Jan 12 23:55:15 ~android Jan 12 23:55:15 it has been said that android is an Open Handset Alliance Project by Google or an alleged Open Source phone software stack that's really not open. http://code.google.com/android/, or if it's sometime actually freed, someone might port it on the neo, but the provided binaries are incompatible Jan 12 23:55:58 Estel_: i think you know what i meant... Jan 12 23:56:03 no need to be so technical Jan 12 23:56:06 to be honest, no Jan 12 23:56:09 ~factinfo android Jan 12 23:56:09 android -- created by TimRiker at Tue May 27 20:54:25 2008 (1325 days); last modified at Fri Jul 11 20:47:02 2008 by mjr!i=mjrauhal@myntti.helsinki.fi; it has been requested 21 times, last by Estel_, 55s ago. Jan 12 23:56:11 if linux hw support wasn't good and bsd driver support wasn't shit, google could port the entire Android stack to *BSD ;) Jan 12 23:56:14 okay if thats how you feel... Jan 12 23:56:18 I don't see any relation between ubuntu and androidp on any level Jan 12 23:56:22 wow Jan 12 23:56:31 a genuine timriker factoid :-D Jan 12 23:56:41 mentioning Neo Jan 12 23:56:47 this channel is harsh... =-P Jan 12 23:56:57 gh0strider, no harsh intended Jan 12 23:57:00 @gh0strider you should see tmo. Jan 12 23:57:10 I *really* *honestly* don't know what You mean by ubuntu android Jan 12 23:57:17 without any irony on layer 2 Jan 12 23:57:22 i think he means a chroot Jan 12 23:57:33 which is ass-backwards, painful and stupid on so many levels Jan 12 23:57:34 ah, maybe Jan 12 23:58:04 well, our ED chroot is working well lately thanks to Sulu and others work. I doubt Android got so qorking chroot Jan 12 23:58:21 i've been trying to bear with unity 2d on an ubuntu chroot on a tab 8.9 - it's laggier than the android ui itself Jan 12 23:58:27 because vnc Jan 12 23:58:33 I will grill TrfueCrypt maintainer for uploading version that fails to work with mkfs on my device Jan 12 23:58:43 wait, the maintainer is me Jan 12 23:58:54 well, I need to think about it for a while. BRB Jan 12 23:59:00 and because android doesn't have a native xserver = big problem Jan 12 23:59:07 if you want to run desktop linux apps Jan 12 23:59:20 * SpeedEvil passes Estel_ some barbecue sauce. Jan 13 00:07:03 honestly. Could any good sould apt-get install truecrypt from -testing or -devel, then, create a 5mb or any size encrypted file container... Jan 13 00:07:19 setting truecrypt to also mkfs it with any filesystem? Jan 13 00:07:25 ot's driving me nuts Jan 13 00:07:27 it's Jan 13 00:07:56 NIN101 told me that it works in his case, I need to filter out if it's general bug or just me Jan 13 00:08:02 and if me, why, why, why! Jan 13 00:08:55 it just goes ok, then, at finish, throw a message that it will not mkfs my new encrypted (and mounted by truecrypt) volume, because it's... mounted Jan 13 00:09:30 * SpeedEvil finds that his wifi net won't connect again. Jan 13 00:09:30 same if I try manually to mkfs it Jan 13 00:09:35 * SpeedEvil stabs icd and reboots. Jan 13 00:10:13 ~ressurect Jan 13 00:11:39 ShadowJK, You know how eMMC wear-*leveling works... Jan 13 00:11:54 Nobody does. Jan 13 00:12:05 Unless they're in the right bit of toshiba. Jan 13 00:12:09 It's a secret. Jan 13 00:12:14 does it respect partition layout, i.e. level only on surface of partition, or throws the f* parts all over flash? Jan 13 00:12:20 lol Jan 13 00:12:36 hate forced wear-leveling Jan 13 00:12:42 The general consensus based on what seems to be occurring based on fragmentary info is that: Jan 13 00:12:49 OS and filesysten should control that Jan 13 00:13:00 The blocks - eraseblocks are 256K or so. You can't erase a finer grain than this. Jan 13 00:13:19 Every 1024 blocks or so are gathered into a wear leveling zone. Jan 13 00:13:26 yea, that's true. Thanks. Any info on respecting partition layout? Jan 13 00:13:29 Of this, 1000 blocks are used, and 24 are spare Jan 13 00:14:01 i.e. dd'ing entire partition, I can assume that wear level was fooled, or will I have nd copy of my encrypted blocks all over a drive? Jan 13 00:14:05 The wear leveling occurs per-zone over the 1000 blocks, until an error over a threshold occurs at which point a spare block is mapped. Jan 13 00:14:33 It's probable that the partition will be mostly respected Jan 13 00:14:50 but where it straddles 'zones' - the edges will blur. Jan 13 00:15:03 The simple reason is that you don't want to have to scan over the whole disk to find a sector. Jan 13 00:15:09 thanks for nice summarizaing Jan 13 00:15:15 if You got more info, go on Jan 13 00:15:22 it's interesting. Jan 13 00:15:36 I'm just b****ing about wear leveling in meantime Jan 13 00:15:54 If you only have to look in a small block table, for 1000 blocks, it considerably speeds things up Jan 13 00:15:59 that's about it alas. Jan 13 00:16:03 well, on microSD cards transparent wear-leveling really does respect partitions? I though it doesn't Jan 13 00:16:12 I would suspect same would apply to our eMMC Jan 13 00:16:23 Why do you think it does respect partitions? Jan 13 00:16:32 I thyink it doesn't Jan 13 00:16:54 as hardware leveling -> partitions -> logical filesystem Jan 13 00:17:10 anyway - on truecrypt - walkthrough needed if you want a tester Jan 13 00:17:16 so, making 2 patitions on sd card and running flash killer on only one of them Jan 13 00:17:27 SpeedEvil, with pleasure Jan 13 00:18:00 on wear-leveling - killing partitioned sd should be faster than non partitioned from same batch Jan 13 00:18:13 i.e. half of sd should die quicker than whole sd Jan 13 00:18:22 yet, it seems to die exactly the same speed Jan 13 00:18:23 possibly, yes. Jan 13 00:18:26 as for truecrypt Jan 13 00:18:38 as for truecrypt yep? Jan 13 00:18:42 now root in xterm Jan 13 00:18:44 and Jan 13 00:18:46 The fundamental problem is there is no insight into the process. Jan 13 00:18:47 'truecrypt' Jan 13 00:18:51 yea Jan 13 00:19:00 fuckin closed&forced wear leveling Jan 13 00:19:12 The above could be true for card A, but false for v1.1 of card A from the same maker Jan 13 00:19:19 on Gui, choose "create volume" Jan 13 00:20:06 then, leave "file container" and hit return or click next Jan 13 00:20:30 go through all options, choose for example to create file container on mydocs and make it 5MB size Jan 13 00:20:32 I see no gUI after apt-get install truecrypt Jan 13 00:20:43 encryption and hash doesnt matter Jan 13 00:20:52 wait Jan 13 00:21:02 'truecrypt' from root doesn't give You a GUI? Jan 13 00:21:20 are You sure that You have extras-testing or -devel and apt-get update recently? Jan 13 00:21:30 there is no deasktop icon Jan 13 00:21:31 yet Jan 13 00:21:44 possily not updated - hangon Jan 13 00:21:47 You start GUI from terminal (well, CLI is also available) Jan 13 00:21:49 ok Jan 13 00:22:00 started truecrypot, and it's a cli Jan 13 00:23:05 updating Jan 13 00:23:44 nice Jan 13 00:24:13 new version makes use of XTD modules from kp, and falls back to -m nokernelcrypto if modules aren't present Jan 13 00:24:16 I assumed apt-get would update, when on reflection, it probably doesn't Jan 13 00:24:21 stock kernel Jan 13 00:24:27 old version had nokernelcrypto hardcoded Jan 13 00:24:31 shouldn't matter Jan 13 00:24:42 it will just fallback to -m nokernelcrypto Jan 13 00:24:56 apt-get never updates when not asked to ;) Jan 13 00:25:03 thanks a_ain for testing it Jan 13 00:25:13 ham is slow Jan 13 00:25:20 FAM here Jan 13 00:28:39 updating a few packages Jan 13 00:28:57 including truecrypt Jan 13 00:33:15 'kernel does not support...' using ... Jan 13 00:36:59 ok Jan 13 00:37:12 formatting aes file on mydocs Jan 13 00:37:15 now, follow my walktheough from backscroll of our irc talk Jan 13 00:37:18 ok Jan 13 00:37:26 successuflly Jan 13 00:37:29 which filesystem? Jan 13 00:37:35 You used? Jan 13 00:37:36 Evening Jan 13 00:37:45 g.night Jan 13 00:37:50 and made FAT fs Jan 13 00:38:01 SpeedEvil, it seems that only my mkfs complains Jan 13 00:38:05 thanks for it... Jan 13 00:38:05 'volume created' Jan 13 00:38:14 Where is the right place to ask Nokia N9 related questions? Jan 13 00:38:18 I got it encrypted properly but halt when mkfsing Jan 13 00:38:28 DrGrov, at Nokia support ;) Jan 13 00:38:34 mkfs.vfat 2.11 (12 Mar 2005) Jan 13 00:38:57 Estel_: I mean here on IRC. Jan 13 00:39:04 my mkfs complain that "is mounted; won't make filesystem there!" Jan 13 00:39:13 odd Jan 13 00:39:17 DrGrov, check for harmattan irc Jan 13 00:39:23 I made a file in /home/user/MyDocs/test Jan 13 00:39:29 Estel_: Thanks, I just read the topic and figured it out. :) Jan 13 00:39:32 hahaha same here Jan 13 00:39:37 yet Jan 13 00:39:42 even if it mount it manually Jan 13 00:39:48 even if it mount it manuallywith --filesystem=none Jan 13 00:39:55 err - 'file' not 'test' Jan 13 00:40:01 so it's mounted, decrypted on the fly when needed, and raw partition Jan 13 00:40:04 or file Jan 13 00:40:15 it says that it's mounted so wont mkfs Jan 13 00:40:20 driving me crazy Jan 13 00:40:24 wacky Jan 13 00:40:39 I would just dd my partition to desktop, mkfs it there, and dd back, but due to wear-leveling... Jan 13 00:40:54 I'll get +ome clones of blocks encrypted with same header Jan 13 00:40:58 err no. Jan 13 00:41:02 = aiding cryproanalysis Jan 13 00:41:12 You cannot tell the block device from a block device Jan 13 00:41:17 _everything_ is hidden Jan 13 00:41:23 yes yes Jan 13 00:41:27 wear leveling is invisible Jan 13 00:41:31 oh - right Jan 13 00:41:38 but, wear level already spread my encrypted partitions chunks all over eMMC Jan 13 00:41:43 yeah Jan 13 00:42:02 are You using busybox-power? Jan 13 00:42:06 no Jan 13 00:42:15 This is a mostly stock system Jan 13 00:42:20 I've screwed with it very little Jan 13 00:43:55 SpeedEvil, could you write here small ojutput of Jan 13 00:44:31 losetup -f Jan 13 00:44:42 /dev/loop0 Jan 13 00:44:52 hm Jan 13 00:44:56 In some forum Jan 13 00:45:05 someone said to mkfs /dev/loop0 Jan 13 00:45:17 now I have no way to check if it was created by truecrypt or not Jan 13 00:45:23 will try to see whete it points Jan 13 00:45:31 don't want to mkfs unknown thing :P Jan 13 00:57:55 SpeedEvil, eureca, I'm mkfs'in *something* Jan 13 00:58:12 after losetup /dev/loop0 /dev/mapper/truecrypt1 Jan 13 00:58:33 my /dev/loop0 disappeared, and /dev/loop2 popped up (!!!) Jan 13 00:59:00 yet, when I used mkfs.ext4 /dev/loop0 - despite it seemed to be non-existing Jan 13 00:59:07 it actually started mkfs'ing Jan 13 00:59:20 I just hope it mkfs'es correct partition ;) Jan 13 01:01:07 " Jan 13 01:01:09 ;0 Jan 13 01:01:15 :) Jan 13 01:02:59 yet, now I can't *unmount* it... Jan 13 01:03:09 truecrypt complain about it being in use Jan 13 01:03:20 and losetup plau silly, stating that Jan 13 01:03:37 losetup -d /dev/loop0 no such ioctl Jan 13 01:03:45 losetup -d /dev/loop2 no such ioctl also Jan 13 01:05:06 SpeedEvil, which block size and fragment size is best for ext on our eMMC? Jan 13 01:05:15 considering physical block size? Jan 13 01:06:04 I guess not 256kb, as it would mean every 100b file would take 1/4 MB from free space ;) Jan 13 01:06:49 rebooting, brb **** ENDING LOGGING AT Fri Jan 13 02:59:57 2012