**** BEGIN LOGGING AT Fri Apr 27 03:00:00 2012 Apr 27 04:23:10 robbiethe1st: err nope Apr 27 04:28:52 TI definitely knows that: >>50K total POH when operating across all OPPs and keeping the time spent at OPP5 to less than 10K POH. Total FaultsInThousand(?) Rate increases from design goal of 50FIT to 70FIT under these accelerated conditions. (for MIL(?) 105°C max devices)<< Apr 27 04:30:27 I.E.when running@600MHz, after 10000 hours 7% devices gone burnout Apr 27 04:31:56 and that's most likely a Gaussian graph, like for incandescent bulb lifetime Apr 27 04:32:08 definitely no linear thing Apr 27 04:54:26 My hildon-desktop suddenly crashed and is refusing to start. Anyone else having similar behaviour with CSSU 3.1? Apr 27 04:55:00 I once had that with PR1.0.1 Apr 27 04:55:16 What did you do to fix it? Apr 27 04:55:29 err, reflash IIRC Apr 27 04:56:06 * int_ua reads about signal 11 Apr 27 04:56:07 since I didn't know all the tricks back when Apr 27 04:57:13 I've edited transitions.ini before it crashed, but I've undone these changes through ssh and it's still crashing Apr 27 05:09:46 DocScrutinizer: I've used Ctrl+Shift+X to start terminal (I wished I knew this yesterday in the train where it started) and replaced transitions.ini with the default one. It seems fixed now, rebooting. Apr 27 05:09:51 well, that's at least unelated to CSSU I'd say. Though thanks to a mini patch I pushed to CSSU you at least have a chance to ssh into device at all, when HD is crashing. In original firmware the device would've entered vootloop now Apr 27 05:09:58 bootloop* Apr 27 05:11:09 DocScrutinizer: You saved me days of time! :) And a lot of other's I think ^_^ Apr 27 05:12:53 DocScrutinizer: FYI I've replaced _all_ numbers in transitions.ini with zeros and commented out all transitions long ago and it was working perfectly Apr 27 05:15:10 * int_ua leaving happily Apr 27 05:19:53 Pali, ping Apr 27 08:00:58 Wicked Lasers is running a promotion whereby if a certain Facebook page is liked 2,000 times, the company will send a team of professional divers to "attach a fully-functional Spyder 3 laser onto a live shark in the wild." Apr 27 08:28:37 RST38h: the carnage could be interesting, but why would thay attach a laser? Apr 27 08:29:35 edheldil: because they make lasers Apr 27 08:29:54 and no, it will be purely cosmetic, as laser does not make it far into the water Apr 27 08:32:34 I mean ... what will be the purpose except of as a warning to people in an underwater research station in a room full of water that there's also a shark there? Apr 27 09:36:56 Pali, ping Apr 27 11:35:57 DocScrutinizer51: I think I will wait for the next maemo phone till I switch... the device seeding program smells like victory over elop to me^^ Apr 27 11:36:06 http://communities-dominate.blogs.com/brands/2012/02/the-truth-the-whole-truth-and-nothing-but-the-burning-platforms-memo-is-1-year-old.html Apr 27 11:36:18 nice blog ;) Apr 27 12:33:20 chem|st: sorry, you lost me somewhat Apr 27 12:33:48 chem|st: which switching? Apr 27 12:48:50 freemangordon, pong Apr 27 13:06:40 Pali, your u-boot code seems to fix the issue with -110 errors. i.e. restoring i2c context after enabling it Apr 27 13:07:20 and the function responsible for garbadge in boot menu is video_invertchar in cfb_console.c Apr 27 13:07:36 nice Apr 27 13:07:36 NFC why byte operation does not work as they should Apr 27 13:07:45 and function which garbage screen - I know it Apr 27 13:08:00 but now one of u-boot developer can reproduce same problem on beagle board Apr 27 13:08:05 do you have a solution (besides disabling SW cursor)? Apr 27 13:08:13 aah, great Apr 27 13:08:26 so I wait, I belive he will fix problem :-) Apr 27 13:09:00 temp solution is to disable cursor Apr 27 13:09:28 great news. On the other hand I see some timeout errors from u-boot when initializing, not sure if they are for internal or external mmc Apr 27 13:09:31 e.g function video_invertchar in cfb_consle.c will do nothing Apr 27 13:09:59 that are problems with omap mmc code Apr 27 13:10:07 yeah, I know that, that is what i did, and i saw 3-4 errors (or warnings) from u-boot Apr 27 13:10:12 also one developer sent patch to mailinglist Apr 27 13:10:50 do you have link? Apr 27 13:11:37 http://patchwork.ozlabs.org/patch/147661/ Apr 27 13:12:43 but now I modified rx51 u-boot code more Apr 27 13:12:55 I will build and publish new version Apr 27 13:13:26 great. you will disable SW corsor until there is a roper fix? Apr 27 13:13:45 by publish you mean in extras-devel? Apr 27 13:15:28 Pali ^^^ Apr 27 13:15:40 yes, I will disable cursor, but I mean that I will push code to gitorious and create some builds Apr 27 13:15:46 nothing to extras-devel yet Apr 27 13:17:22 aah, ok. Are you going to do this today? Because here next 4 days ar non-working and I would like to do some testing, etc. If you are not going to publish that today, mind to share the patches? Apr 27 13:17:24 http://lists.denx.de/pipermail/u-boot/2012-April/123261.html Apr 27 13:18:04 ok, I will rebase my git repo and create clean commits Apr 27 13:18:16 thanks Apr 27 13:20:43 there is one problem with maemo 2.6.28 kernel Apr 27 13:21:02 if I enable usb tty in u-boot maemo kernel does not boot Apr 27 13:21:11 it crash something in musb driver Apr 27 13:21:29 so I temporary comment usb tty support in u-boot Apr 27 13:30:19 Pali, and what am I expected to do? :P Apr 27 13:30:58 that if you want usb tty support for testing u-boot (without maemo kernel) you need to enable it & recompile Apr 27 13:31:28 freemangordon, on u-boot mailinglist are patches for ext4 support. I will try to include it Apr 27 13:32:24 pali, isn't it better first to have stable u-boot and after that to improve it? Apr 27 13:32:54 I have sd card formated with ext4 Apr 27 13:33:14 so for testing on my n900 it will be better Apr 27 13:33:20 AIAU there are lots of people waiting your u-boot to become stable to start using it. Apr 27 13:33:30 myself included :D Apr 27 13:34:17 Pali, btw here http://maemo.merlin1991.at/cssu/develdebs/ is gcc 4.6.2 to be used in SB Apr 27 13:34:44 what has new version fixed? Apr 27 13:35:15 Pali, SB has gcc 4.2.4 Apr 27 13:35:22 and that is 4.6.2 Apr 27 13:35:39 I know, but I thought that gcc is stable Apr 27 13:36:19 and c++0x support I do not need :D Apr 27 13:36:44 well, it is, but some SWs don't compile with 4.2.4 Apr 27 13:38:24 ok Apr 27 13:38:35 and thumb support? Apr 27 13:40:46 yeah, that is fixed Apr 27 13:41:05 in binutils 2.22, which is in the same tar.gz Apr 27 13:52:40 do you think that we can recompile qt libraries with enabled thumb? Apr 27 13:58:44 I've already done that Apr 27 13:58:47 :) Apr 27 13:59:13 but still not sure if thumb is stable on n900 and what is needed to stabilize it if not Apr 27 14:10:52 freemangordon, any updates on .37 for Fremantle? Apr 27 14:25:33 camera on n900 is much better than I've expected! Apr 27 14:26:14 LaoLang_cool: it's a lot less granier than most 5mp cameras out there, except the N95 Apr 27 14:26:29 It could produce high quality and beautiful photos! Apr 27 14:27:27 hurrian: granier? Apr 27 14:27:42 I can't find the word on my dictionary :( Apr 27 14:28:10 it's less grainy on the N900 because of not-as-tightly packed sensor and relatively large lens Apr 27 14:28:27 like, white/grey dots on dark parts of a picture Apr 27 14:30:44 ok, although I don't understand, it's very good! Apr 27 14:31:06 yes, it is! Apr 27 14:31:14 ;p Apr 27 15:08:57 freemangordon, sources are on: https://gitorious.org/u-boot-shr/u-boot/commits/pali Apr 27 15:09:07 now with debian/* files Apr 27 15:16:51 visible changes for users: Apr 27 15:16:51 * atag address for kernel is always 0x80000100, so for reusing atags from nolo is needed command: "setenv atagaddr 0x80000100" Apr 27 15:16:51 * u-boot-update-bootmenu needs option ITEM_REUSEATAGS=1 (instead ITEM_USEATAG) - but works also old option (it show warning) Apr 27 15:16:51 * noloboot renamed to attachboot Apr 27 15:16:52 * if not reusing atags from nolo, u-boot will create memory atag which is needed to boot kernels (now maybe meego/upstream kernel will boot without nolo atags) Apr 27 15:34:25 freemangordon, new u-boot deb packages are here: http://atrey.karlin.mff.cuni.cz/~pali/u-boot/ Apr 27 15:49:17 Pali, thanks Apr 27 15:49:29 I will also update TMO thread Apr 27 15:52:12 great Apr 27 16:11:20 freemangordon, TMO thread updated: http://talk.maemo.org/showthread.php?t=81613 Apr 27 16:11:36 :) Apr 27 16:12:01 Instaling Ubuntu on microsd will try the new u-boot with it Apr 27 17:34:50 freemangordon, I updated bootmenu package (pushed to extras-devel) to hide kernel entries Apr 27 17:46:34 DocScrutinizer51: alternative device... switching to another device might be possible as they revive Qt devel which is not needed for WinMo Apr 27 17:47:06 dang, does my IRC UA have auto-"away" now? Apr 27 17:52:49 DocScrutinizer: as long as it doesn't announce it to channels... Apr 27 17:54:21 nah, chem|st must have mega ESP powers :-D I was just sitting at my PC for <60s and the Elop-webpage was still loading when I got highlighted by him (or, according to latest grammar, her) Apr 27 17:55:19 "How to Tell if You (Or Someone You Know) Are a Twitter Assclown" <== WEB2.0 getting really useful now Apr 27 17:56:01 DocScrutinizer51: special powers that is... Apr 27 17:56:08 i Apr 27 17:57:35 'd love to get an n950 to do testing like I did for m5... but well people who code get devices, people who are good in testing only are in need to buy them... without being available on the market a 'no can do' Apr 27 17:58:15 and for sure I do not want a phone without kbd Apr 27 17:58:43 RST38h: pls elaborate twitter assclown... Apr 27 18:00:07 chem|st: not completely right, seems my N950 originally moved to me due to my exceptionally good/harsh/snappish testing Apr 27 18:01:17 well I helped out sequentially with dbus calling... and so on... I didn't even think of asking! Apr 27 18:01:20 konttori said sth like "I already wondered how to get one of those to you" Apr 27 18:01:20 gtg Apr 27 18:01:28 chem: OMG LOL U KNO JST FOOLIN 'ROUND U SHITE AHAHAHAHA! Apr 27 18:01:42 Meanwhile: http://www.geek.com/articles/mobile/nokia-lumia-800-and-710-jailbroken-thanks-to-firmware-oversight-20120418/ Apr 27 18:02:08 HAHAHA Apr 27 18:02:10 not mine Apr 27 18:02:59 onviously the number of devices with that fsxkd prototype BL or whatever it was are in the hundereds or max 1000s Apr 27 18:03:56 meh, who cares about root of crap? Apr 27 18:05:44 * DocScrutinizer idly wonders if it was in "12 monkeys" where brad pit explained "we in here are the sane people, all out there are the lunatics" Apr 27 18:07:36 yep Apr 27 18:07:58 and yes, all Gilliam films make this impression Apr 27 18:16:56 hi, is anybody working libpcap on maemo right now? Apr 27 18:40:28 HRHRHRRRHARRRHRHARR >>Symbian was the bestselling smartphone on five of the six inhabited continents and Nokia's Maemo was the bestselling smartphone on the uninhabited continent of Antarctica too.<< Apr 27 18:40:42 But it is true! Apr 27 18:40:48 aure Apr 27 18:40:50 sure Apr 27 18:41:11 I just like the sarcasm like sulfuric acid Apr 27 18:42:30 I wish somebody would tie Flop to a chair, give him some strychnine to snort, then recite this whole blog over and over again Apr 27 18:42:35 inspite of i've set sound file name with hildon_notification_set_sound, i still don't get sound with hildon notification, anyone have ideas what the reason can be? Apr 27 18:44:01 and also call notify_notification_set_hint_string(NOTIFY_NOTIFICATION(_hildon_notification), HINT_KEY, HINT_VALUE) has no effect Apr 27 18:44:14 notification sounds get converted to wav and stored somewhere in ~/.* Apr 27 18:44:52 I'd not be surprised to find your abscure cals not handle all that correctly Apr 27 18:44:56 DocScrutinizer, yes, i know, i've set filename to /home/user/.local/share/sounds/Message1.acc.wav Apr 27 18:45:13 sorry, nfc Apr 27 18:45:28 what API is that? Apr 27 18:45:45 Apr 27 18:46:11 so c? Apr 27 18:46:20 qt Apr 27 18:46:25 mhm Apr 27 18:46:31 c++ Apr 27 18:46:40 qt honestly never really worked for such stuff, I'd say Apr 27 18:48:09 showing notification works Apr 27 18:48:19 wow Apr 27 18:48:35 I can do that from cmdline via dbus-send Apr 27 18:48:42 :) Apr 27 18:48:51 i know Apr 27 18:49:19 anyway, nfc realy. Back to that Floptrashingblog Apr 27 20:08:39 * SpeedEvil is highly amused to be running out of space on the tiny internal partition on hi ndroid device that everything seems to be installed by default to. Apr 27 20:08:51 (Gift of a cheap tablet) Apr 27 20:16:31 which tablet? Apr 27 20:16:38 and don't you feel dirty for using an android device? Apr 27 20:20:39 flashliht is currently my favourite app Apr 27 20:21:23 on holiday in germany :) Apr 27 20:28:05 kerio: Wopad I7 Apr 27 20:28:15 kerio: It's main use is an angry birds platform Apr 27 20:28:21 I'm still using my n900 more Apr 27 20:28:32 SpeedEvil: ...why not a computer, for angry birds? :\ Apr 27 20:28:40 and didn't you have a n950 too? Apr 27 20:28:49 Is angry birds available on linux? Apr 27 20:29:03 there's an html5 version :) Apr 27 20:29:03 SpeedEvil: http://chrome.angrybirds.com/ Apr 27 20:29:06 yeah Apr 27 20:29:21 Anywya - it's excellent as a 'while the kettle boils' game Apr 27 20:29:24 not sure about space, though Apr 27 20:29:25 ShadowJK: drop by for a beer! Apr 27 20:29:43 Space is OK, but variable Apr 27 20:29:55 The third episode of space is a brainfuck on some levels. Apr 27 20:30:10 - getting 3 *'s - not actually completing them Apr 27 20:31:08 The couple of walkthroughs I've looked at after doing levels that have taken me ages both revealed there was a _much_ easier way that I missed. Apr 27 20:31:47 anyway, why bother with angry birds when there's nethack? :)= Apr 27 20:31:54 s/=// Apr 27 20:31:56 kerio meant: anyway, why bother with angry birds when there's nethack? :) Apr 27 20:32:18 Nethack isn't quite as suitable as a casual game. Apr 27 20:39:06 SpeedEvil: your the same SpeedEvil from gentoo? :P Apr 27 20:39:21 no Apr 27 20:39:29 I'm not involved in gentoo Apr 27 20:39:38 ohh im sorry :( Apr 27 22:09:38 when I got two .tgz that differ only in a few small files, a few bytes in there, while 99.9% are identical... What will a new compressing them into another .tgz get me? 0% compression or 50%? Apr 27 22:12:06 is there even *any* compressing archiver that would result in any compression on archiving two 100% identical .tgz? Apr 27 22:12:58 (given the original .tgz files are >256bytes) Apr 27 22:17:46 * DocScrutinizer just nuked 6.4GB of backupmenu rootfs/optfs.tar Apr 27 22:20:15 0% Apr 27 22:20:24 Well Apr 27 22:20:27 thought as much Apr 27 22:20:56 A block compressor with a large enough window will compress two identical files of under the block size to one Apr 27 22:21:09 But compression is like encryption Apr 27 22:21:28 even the smartest compression algo won't use a "dictionary" of several MB length - based on the one file Apr 27 22:21:50 one or two bytes different in the input stream (the uncompressed tar) - means the output catastrophically differs Apr 27 22:22:11 ouch Apr 27 22:22:36 well, actually they were .tar Apr 27 22:22:55 but that would complicate my basic question Apr 27 22:27:17 dd if=/dev/urandom of=file1 bs=1M count=1; cp file1 file2; tar -czf archive file1 file2 Apr 27 22:28:13 try it with 100k, and bzip2 -9 Apr 27 22:40:12 :-D Apr 27 22:51:04 * SpeedEvil sighs. Apr 27 22:51:13 I need to do some work on that web-proxy thing. Apr 27 22:51:15 Compressing **** ENDING LOGGING AT Sat Apr 28 03:00:00 2012