**** BEGIN LOGGING AT Sat Sep 01 03:00:02 2012 Sep 01 06:25:44 ZogG_laptop: the point is not circumcision, the point is forcing a useless and **dangerous** medical procedure on a child Sep 01 06:26:04 genital mutilation is definetely not a good thing Sep 01 06:26:42 also note that the ban doesn't prevent circumcision, even at a young age, for valid medical reasons Sep 01 06:28:39 and it doesn't prevent circumcision on adults Sep 01 06:30:18 LaoLang_cool: xbindkeys Sep 01 07:18:06 kerio, thank you, will try Sep 01 07:36:23 is there any place i can buy a genuine N900 battery in UK? Sep 01 08:00:43 trumee: probably a lot of webshops? Sep 01 08:00:51 BL-5J is quite common afaik Sep 01 08:03:11 Lava_Croft, yeah being common is the problem. more likelyhood of a fake Sep 01 08:03:36 trumee: I had trouble finding a fake one in a Dutch webshop Sep 01 08:03:50 'trouble' Sep 01 08:04:30 enter 'bl-5j site:uk' in google Sep 01 08:04:33 without the '' Sep 01 08:05:37 Genuine BL-5J goes for 9euros in the webshop where i bought em Sep 01 08:05:50 no idea how much quid that would be Sep 01 08:14:18 >genuine Sep 01 08:14:19 >9€ Sep 01 08:14:21 LOL Sep 01 08:19:38 when talking about SDs and batteries, if you got them for cheap then they're fake Sep 01 08:19:41 no exception Sep 01 08:21:41 and you assume 9euros = fake? Sep 01 08:22:41 yep Sep 01 08:22:49 Weird assumption Sep 01 08:23:36 9 to 14 euros is the common price to genuine BL-5J's over here:) Sep 01 08:23:40 depending on where you buy it Sep 01 08:24:33 why xkeybinds doesn't work here... Sep 01 08:24:50 I use a default rc file, but the setting doesn't take any effects Sep 01 08:25:13 control+shift+q doesn't do a "xbindkeys_show" Sep 01 11:10:53 9E might be a genuine bl-5j, but it's probably old stock Sep 01 11:27:25 yup Sep 01 11:27:36 and its certainly a genuine one, well genuine 4 Sep 01 11:27:37 :) Sep 01 11:28:01 but there's always people who, through the tubes of the internet, can see if they are genuine Sep 01 11:28:04 * Lava_Croft winks Sep 01 11:29:00 the new bl-5 are 1400-something mAh Sep 01 11:30:23 ShadowJK: and a double scud is almost 3000 :D Sep 01 12:34:27 has anyone ever encountered an error in image-viewer that prevents any kind of image display while claiming that there's "insufficient memory to display the image"? Sep 01 12:34:57 i don't even know how to troubleshoot it Sep 01 12:35:38 do you have swap enabled? Sep 01 12:38:55 yep Sep 01 12:38:58 ...i hope Sep 01 12:39:06 man, if it's *that*... Sep 01 12:40:50 and? is it *that*? Sep 01 12:40:54 idk, it's booting Sep 01 12:41:03 nope, it's not that Sep 01 12:41:58 ... Sep 01 12:42:08 .. fucking . Sep 01 12:42:14 Already restored that backup? Sep 01 12:42:22 no, i didn't even make it Sep 01 12:42:24 found the problem Sep 01 12:42:25 ish Sep 01 12:42:45 i don't... Sep 01 12:43:20 http://www.youtube.com/watch?v=iMXScRvd0Zs Sep 01 12:43:22 this is how i fee Sep 01 12:43:23 l Sep 01 12:43:58 yep, found the problem Sep 01 12:44:12 so... Sep 01 12:44:24 ShadowJK: how big is the default emmc swap? Sep 01 12:44:32 (rhetorical question) Sep 01 12:44:36 (but please answer) Sep 01 12:45:01 76i Sep 01 12:45:04 768M Sep 01 12:45:08 ok Sep 01 12:45:12 iirc Sep 01 12:45:14 how big is my 4gb swap partition? Sep 01 12:45:47 it's 0 if you overflow signed integer? :) Sep 01 12:45:59 ex fucking actly Sep 01 12:46:08 does it work with emmc swap? Sep 01 12:46:11 yep Sep 01 12:46:18 lol, fantastic Sep 01 12:46:23 should i make my swap partitions smaller or bigger, then? Sep 01 12:46:42 what if you have uSD 4g swap at prio 10 and emmc swap at prio 5? Sep 01 12:46:58 ShadowJK: eeh, i found that it still uses the swap at the lower priority Sep 01 12:47:12 just much less so Sep 01 12:47:14 not if it's 0 in use Sep 01 12:47:39 no, i mean, i actually tried this at some point Sep 01 12:47:44 in other words. if you disable/enable emmc when usd is in use Sep 01 12:47:44 together with compcache, too - which is stupid Sep 01 12:48:04 hm Sep 01 12:48:07 i just don't want to, though Sep 01 12:48:07 I always have emmc swap enabled too, a side effect Sep 01 12:48:09 it's stupid Sep 01 12:48:25 and i'd have to adjust my swapswap script Sep 01 12:49:22 well, this is fun Sep 01 12:49:25 and by fun i mean FUCKING STUPID Sep 01 12:50:29 gaah, and this piece of crap isn't even open Sep 01 12:50:45 I'm out Sep 01 12:54:00 ShadowJK: what would you do? Sep 01 12:55:46 my scripts already accidentally leaves emmc enabled :) Sep 01 12:57:13 haha Sep 01 12:57:39 meh, you know what, my script is actually pretty good Sep 01 12:58:06 but i don't want to deal with swap priorities :s Sep 01 13:00:10 maybe i'll make it 5gb? Sep 01 13:00:31 DocScrutinizer05: fukken nokia got me again Sep 01 13:04:28 ShadowJK: do you think 512M would be ok for image-viewer? Sep 01 13:04:47 probably Sep 01 13:04:49 meh, it's actually a pretty bad thing regardless Sep 01 13:05:16 because i could have a very low free swap modulo 2GiB, but still have a lot free Sep 01 13:05:26 this is a serious problem ._. Sep 01 13:07:26 kerio: I think I told you not to use insanely large swap Sep 01 13:07:48 DocScrutinizer05: good morning Sep 01 13:07:56 kerio: maybe you want to read manpages about swap and swap related stuff Sep 01 13:08:01 DocScrutinizer05: did u finish Absolut? Sep 01 13:08:15 not even started with absolut Sep 01 13:08:22 kept that for today ;-D Sep 01 13:08:46 :P Sep 01 13:08:55 * DocScrutinizer05 checks if freezer already managed to freeze Absolut Sep 01 13:09:06 he wont find in manpages that nokia's "is thereenough memory?" heurestics have overflows :) Sep 01 13:09:55 hehe Sep 01 13:10:07 but swap itself has limitations Sep 01 13:10:40 the only thing that complains is the fucking image-viewer Sep 01 13:10:56 quite some time since I investigated all this, but I recall limits both in size of single partition as well as in max number of partitions Sep 01 13:11:54 i'm afraid 2gb will overflow too :c Sep 01 13:11:57 and definitely large swap means huge tables resident in RAM which, at a certain turnaround point, negates the purpose of swap Sep 01 13:13:04 there's a certain fixed ratio between size of swap space and size of stay-resident management tables Sep 01 13:13:15 iirc Sep 01 13:14:17 let's assume this ration is 100:1 (an arbitrary number I picked), then for 2GB swap you'll 'waste' 20MB of RAM Sep 01 13:14:32 s/ration/ratio/ Sep 01 13:14:33 DocScrutinizer05 meant: let's assume this ratio is 100:1 (an arbitrary number I picked), then for 2GB swap you'll 'waste' 20MB of RAM Sep 01 13:15:29 probably you also waste CPU cycles to manage that huge amount of swap Sep 01 13:15:58 so i should just use a 1gb partition? Sep 01 13:16:01 that sounds... small :c Sep 01 13:16:36 and for sure you can't speed up IO bandwidth, so from a certain max size of swap you'll find device frozen in minutes of swap IO Sep 01 13:16:56 urgh Sep 01 13:16:57 surely you mean *used* swap there Sep 01 13:16:59 swap trashing Sep 01 13:17:10 one thing that needs to to be fixed on Linux Sep 01 13:17:13 Ikarus: fun fact: your n900 is *constantly* swap trashing Sep 01 13:17:22 basically nowadays swap sizes >RAM*3 are useless Sep 01 13:17:52 kerio: swap trashing means interactivity goes down the drain due to swap behaving Sep 01 13:18:01 most of the time my N900 doesn't get quite that far Sep 01 13:18:29 back in times where swap was RAM*10 or even *20, it been considered acceptable when system stalls for several seconds when swapping. Main concern been to get jobs done at all Sep 01 13:19:41 and systems been so slow it was basically irrelevant if a task needed 30min to terminate, or 5h (with massive swapping). As said above, main concern: it finishes at all eventually Sep 01 13:20:22 ...wait, wtf Sep 01 13:20:23 now you want "realtime" and this absolutely rules out huge swap sizes Sep 01 13:20:27 usb mass storage exports the whole sd? Sep 01 13:20:45 i really hope the swap partition is somehow locked Sep 01 13:20:48 :s Sep 01 13:21:19 no, afaik MSM exports the first partition if that partition is VFAT Sep 01 13:21:42 well, my os x is definetely aware that the uSD has three partitions Sep 01 13:22:07 hmm, are you sure that's not 3 devices with no partition? Sep 01 13:23:15 two devices, one with a single vfat partition (mydocs), and another with three partitions, vfat and two linux swap Sep 01 13:23:15 I *think* cssu ke-recv got patched to export all partitions on uSD Sep 01 13:23:27 o.O Sep 01 13:23:28 s/all partitions on uSD/the uSD/ Sep 01 13:23:55 oh, btw, i'm running cssu-devel now Sep 01 13:24:13 hmm Sep 01 13:24:33 i'm not sure g_file_storage can export more than two files Sep 01 13:24:44 whatever, there's a script doing that. You may tweak it to meet your needs Sep 01 13:24:57 so if you want to export more partitions, you have to export a whole device with a partition table Sep 01 13:26:01 check /usr/sbin/osso-usb-mass-storage-enable.sh Sep 01 13:26:15 hm, it's not that i think Sep 01 13:26:22 that's callable to export manually something Sep 01 13:27:06 ooh, g_file_storage has a way to export more than two devices Sep 01 13:27:35 anyway, no, that script is called by something else to enable ass rage mode Sep 01 13:28:13 /sbin/modprobe g_file_storage stall=0 luns=2 removable Sep 01 13:28:24 yep Sep 01 13:28:28 luns= Sep 01 13:33:41 I think /usr/sbin/osso-usb-mass-storage-enable.sh will be called by ke-recv, with phy device to export as parameter Sep 01 13:34:03 for now it's built to get called max twice Sep 01 13:34:27 first call exportsa to LUN0, second to LUN1 Sep 01 13:35:34 the idea of using a ludicrously large swap partition isn't to have room for more apps running, device becomes mostly unusable at around 200-400M swap in use anyway. The idea is that it takes longer before 4g gets fragmented than it takes for 1g to get fragmented Sep 01 13:35:34 actually a 3rd call will override the second Sep 01 13:35:47 by redefining LUN1, aiui Sep 01 13:36:57 dunno if it gets called by a events.d script, a init.d script, some helper script called by ke-recv or directly by ke-recv Sep 01 13:39:29 seems /usr/sbin/ke-recv has a reference to /usr/sbin/osso-usb-mass-storage-enable.sh Sep 01 13:40:06 so I assume it's called directly by ke-recv process Sep 01 13:40:58 ask pali if you want to learn more, iirc he patched ke-recv Sep 01 13:42:28 ShadowJK: yeah but DocScrutinizer05 claims that a 4gb fragmented swap partition will take exactly 4gb worth of reads to be swapoffed Sep 01 13:43:27 I never said that Sep 01 13:44:27 I said that due to fragmentation, your real amount of reads will be way more than the 'used' swap suggests Sep 01 13:45:43 if you got one 512byte block used per MB of swap space, your swapoff will need to read in the page containing that 1MB for each block to swap in Sep 01 13:46:48 anyway, you can stop complaining now Sep 01 13:46:49 which, regarding read speed both on interface and on MMC internal controller, creates a significant overhead Sep 01 13:46:52 i've got a 1gb partition Sep 01 13:47:42 and besides, swapon is basically free, and ionice -c 3 nice swapoff should put absolutely no strain whatsoever on the system Sep 01 13:48:05 o.O Sep 01 13:48:33 although i'm not sure ionicing swapoff will help Sep 01 13:48:46 because well, it's a kernel thing Sep 01 13:54:38 exactly Sep 01 13:55:01 swapoff does no IO basically Sep 01 13:58:14 anyway, we need an opensource replacement for image-viewer Sep 01 14:16:07 mhm Sep 01 14:16:32 actually a funny bug Sep 01 14:32:03 the parts needed to fix my N900 have finally arrived Sep 01 14:32:15 need to go to the repair shop next week and get them to sort it :) Sep 01 14:33:09 DocScrutinizer05: or maybe i'll just get jacekowski to edit the binary Sep 01 14:33:16 and crack the free ram check Sep 01 14:35:10 should be pretty simple Sep 01 14:35:43 at least when you can live without that check Sep 01 14:35:49 hell yeah i can Sep 01 14:35:56 well, now i don't know Sep 01 14:36:01 i only have 1gb of swap Sep 01 14:36:07 disabling the check it simple like flipping one bit in a compare condition Sep 01 14:36:17 DocScrutinizer05: huh... not really Sep 01 14:36:20 s/it/is Sep 01 14:36:35 i want the check to succeed both times Sep 01 14:36:42 yes Sep 01 14:36:42 not only when there's "not enough ram" as it thinks Sep 01 14:36:55 wut? Sep 01 14:37:06 which compare would you make there? Sep 01 14:37:20 you want the check to take "not enough ram2 path no matter what the real situation? Sep 01 14:37:22 it would have to be "lesser or greater or equal" Sep 01 14:37:33 DocScrutinizer05: no, i want the check to take the other path :) Sep 01 14:37:44 thought as much Sep 01 14:38:11 regardless of ram Sep 01 14:38:20 so it's more than a bit Sep 01 14:38:21 well, if the compare is "<0" now, you can change it to "=0" Sep 01 14:38:26 it could be a byte or maybe two! Sep 01 14:38:48 DocScrutinizer05: but what if you have exactly as much free ram as the threshold? Sep 01 14:38:54 (modulo 4GiB) Sep 01 14:39:05 of course you always can change the branch destination addr to point directly behind the branch Sep 01 14:39:31 kerio: (exactly as much) then you're fsckd ;-P Sep 01 14:39:39 hell naw Sep 01 14:39:52 just hope the correct branch is the one right below, and change that to a NOP Sep 01 14:40:04 or, if it's not, change it to a JMP Sep 01 14:40:10 yep Sep 01 14:40:15 anyway it's simple Sep 01 14:41:23 other possible fix would be to patch the system function that returns the free ram value Sep 01 14:41:39 could even be done per LD_PRELOAD I think Sep 01 14:42:11 i actually thought about that Sep 01 14:42:29 swapoff on fragmented space is mostly a cpu dominated operation, and then followed by the high command overhead. 4K chunks, only VAX used 512 :) Sep 01 14:42:42 I guess the return value of that function is uint, while it's compared as a signed integer Sep 01 14:42:56 so if MSB is set, it turns negative Sep 01 14:43:34 DocScrutinizer05: 4gb would overflow anyway Sep 01 14:43:37 dang, we've seen that somewhere else last few months ;-P Sep 01 14:43:40 it's 32bit Sep 01 14:48:46 why do i need to enter a CAPTCHA while i am registered in maemo talk when i try to send a new topic? Sep 01 14:49:15 kerio: if the function returns byte, then yes Sep 01 14:49:47 if the function returns maybe words or dwords, then it's 8GB or 16GB Sep 01 14:50:09 also can't set a picture for my profile Sep 01 14:50:12 but I rather think we're running into age old 32bit limitations here Sep 01 14:51:20 Timmy: either you are not registered, or tmo is in pathological state (not really something unheard) Sep 01 14:51:49 DocScrutinizer05: tmo? Sep 01 14:52:06 ~tmo Sep 01 14:52:07 tmo is probably http://en.wikipedia.org/wiki/TMO, or http://talk.maemo.org, or http://de.wikipedia.org/wiki/Terrestrial_Trunked_Radio#TMO. It's *not* T-MO (see ~T-MO) or trolls, morons, oxen. Sep 01 14:52:51 also i am registered and my username is Timmy on the website, and it's a while that i am facing this problem Sep 01 14:53:15 hmm Sep 01 14:53:47 not a anoyoing problem, just reporting that Sep 01 14:53:54 err, anoying Sep 01 14:53:56 yeah, thanks Sep 01 14:54:03 yw Sep 01 14:54:20 I think you should contact Reggie, maintainer of tmo Sep 01 14:54:46 or wait for chem|st to picjk up on this threaqd here Sep 01 14:55:12 maybe chem|st can do something to investigate Sep 01 14:55:34 dunno about which powers he got on tmo Sep 01 14:55:46 ping chem|st Sep 01 14:56:42 chem|st: see ^^^ - tmo behaving mad Sep 01 15:03:37 tbh I dunno if that's maybe normal. Though I don't think so. It could be two unrelated probelms as well Sep 01 15:09:18 DocScrutinizer05: yeah but it reported the wrong size for me, and i had ~4G + 256M Sep 01 15:09:36 well, maybe slightly more than 4g of free ram Sep 01 15:09:49 sure Sep 01 15:10:37 not even 32bit desktop linux works particularly great with more than 4GB RAM (ok, slightly different issue, but still...) Sep 01 15:19:13 DocScrutinizer05: would it be theoretically possible to install a faster cpu and more ram on a n900? Sep 01 15:19:26 yes Sep 01 15:19:37 how feasible would that be? Sep 01 15:19:53 hard to say, depends on a lot of preconditions Sep 01 15:20:39 for the faster, you might find a CPU up to 1GHz I think, for RAM you have to see what you can get Sep 01 15:21:23 then swap the whole chip, gifted SMD reworkers can probably do that with ~50% to 75% yield Sep 01 15:21:43 after some practice Sep 01 15:22:03 maybe 25% chances on the first few devices to undergo that rework Sep 01 15:22:53 I.E. calculate for 10 to 20 devices to practice, and you may find you established a maintainable and sustainable process Sep 01 15:23:18 or not, if some unknown murphy effect strikes Sep 01 15:23:47 DocScrutinizer05: you don't have a job right now, right? Sep 01 15:23:54 i have a business proposal for you! Sep 01 15:23:54 then you just need a source for the better chips, and the chips shouldn't be locked Sep 01 15:24:13 kerio: sorry, I can't afford this Sep 01 15:25:00 if I had some 5..10k EUR to 'waste', I'd maybe consider it Sep 01 15:25:49 break-even point would take me to sell quite a number of those pimped N900 though Sep 01 15:27:22 seems like I'd need at very least 30 preorders, if the rework costs 200EUR Sep 01 15:28:02 or take the risk I won't find 30 customers willing to pay 200EUR for the rework, and lose my investment Sep 01 15:28:05 i'm prepared to invest TENS of euros in this project Sep 01 15:28:17 maybe even two or three tens Sep 01 15:28:35 well, then that's obviously a business model for you, not for me Sep 01 15:29:13 ok fine, i'm prepared to invest 200€ Sep 01 15:29:23 I'm more than happy to help with theory Sep 01 15:29:23 but i want an upgraded n900 at the end Sep 01 16:27:13 jacekowski: ping Sep 01 16:28:12 ShadowJK: ooh, it appears that bq27200 changed something in the calibration data without actually performing a full calibration Sep 01 16:28:24 it went from 1305 mAh to 1303 for the full charge Sep 01 16:28:35 is that legit? Sep 01 16:30:33 i believe so, yes Sep 01 16:31:12 yay Sep 01 16:32:48 * kerio mourns the loss of 2 mAhs Sep 01 16:33:51 :P Sep 01 17:29:46 kerio: aging Sep 01 17:30:11 happens to everyone :) Sep 01 17:30:12 see config in ROM data from bq27k-detail.sh Sep 01 17:33:16 actually bin/bq27k-detail : SD=201 thousandths of percent per day AGELMD=1 Sep 01 17:33:49 hm Sep 01 17:33:57 so it's not actually measured Sep 01 17:33:59 laaaaaaaame Sep 01 17:34:20 means the chip will reduce your LastMeasuredDischarge by 0.202% per day Sep 01 17:34:43 to account for naural aging of cell chemistry Sep 01 17:34:48 natural* Sep 01 17:35:31 of course that's a rough guesstimate method Sep 01 17:36:29 but after 30(?) cycles CI=1 anyway, so you will probably never really see massive impact of cell aging Sep 01 17:36:48 since you should run a learning cycle to reset CI=0 Sep 01 17:37:32 CI? Sep 01 17:37:40 "idk what's going on lol"? Sep 01 17:37:42 CalculationIncorrect Sep 01 17:37:54 =D Sep 01 17:38:04 means "forget all values the chip displays, probably bogus" Sep 01 17:38:31 i'll have to download your script again Sep 01 17:38:40 yw Sep 01 17:39:50 http://maemo.cloud-7.de/maemo5/patches_n_tools/calibrate-bq27k.sh http://maemo.cloud-7.de/maemo5/usr/local/sbin/bq27k-detail2 http://maemo.cloud-7.de/maemo5/usr/local/sbin/bq27200.sh Sep 01 17:41:19 for your convenience Sep 01 17:41:30 just wget them Sep 01 17:41:51 and chmod +x Sep 01 17:43:33 nice script there, calibrate-bq27k.sh .. how about calling `charge21.sh` instead of bme for resuming chaarging? Sep 01 17:46:44 kerio: >> 0 CI Capacity Inaccurate flag. A 1 indicates no valid learning cycle. << Sep 01 17:47:15 Sicelo: you're free to edit ;-) Sep 01 17:47:48 ;) Sep 01 17:54:00 kerio: I invested lots of my precious time to make bq27k-detail2.sh output a tad educational. You should use that opportunity Sep 01 17:54:45 ;-) Sep 01 18:02:41 DocScrutinizer05: k Sep 01 18:03:41 anyone know where the source code is that handles changing the background image in diablo? Sep 01 18:05:11 hi unfortunately I kinda bricked my phone Sep 01 18:05:18 anyone can help me restore it Sep 01 18:05:39 I applied a patch and didnt reboot instead the thing just half ass boot Sep 01 18:05:42 JZA: full reset, coldflash, reflash vanilla+combined Sep 01 18:05:50 oh, you just need the latest then Sep 01 18:05:59 "bricked" is much harder than that Sep 01 18:06:17 load rescueos and then undo the patch Sep 01 18:06:30 dafaq, why coldflash? Sep 01 18:06:35 kerio: well by half ass boot means I cant even see the screen Sep 01 18:06:35 or maybe reboot without watchdogs (via rd flag) Sep 01 18:06:42 DocScrutinizer05: he said bricked Sep 01 18:06:57 he said "OMGitDoesNotWork!" Sep 01 18:07:05 kerio: only my keys get light up, but I dont get any sort of prompt cant get on the wifi so I cant even remote access it Sep 01 18:07:23 rescueos could save you Sep 01 18:07:37 JZA: reflash Sep 01 18:07:41 ~flashing Sep 01 18:07:41 extra, extra, read all about it, maemo-flashing is http://wiki.maemo.org/Updating_the_tablet_firmware Sep 01 18:07:51 otherwise, reflash and then restore the backupmenu backup you surely have done before tinkering with your system, right? Sep 01 18:08:01 right? Sep 01 18:08:03 :-) Sep 01 18:08:08 RIGHT? Sep 01 18:08:19 I do have some backups, not as current as I wish Sep 01 18:08:24 kerio: please! be gentle to new users Sep 01 18:09:23 " so pretty please, with sugar on top, clean the fucking car" Sep 01 18:09:33 JZA: try reflashing (unless you got extremely valuable data in your rootfs - a location you usually don't touch as normal user) Sep 01 18:10:04 ~rescueos Sep 01 18:10:08 meh Sep 01 18:10:11 ...dammit infobot Sep 01 18:10:24 kerio: why confuse new users? Sep 01 18:10:25 JZA: if you had a shell, could you undo the damage? Sep 01 18:10:47 DocScrutinizer05: unix is to be fixed, not reinstalled Sep 01 18:10:47 kerio: if he had a shell, he probably had mentioned it Sep 01 18:11:12 he could have one Sep 01 18:11:15 with rescueos! Sep 01 18:11:30 he could as well be a trolling kernel guru Sep 01 18:12:39 he's asking "OMG, my device is bricked! it doesn't boot anymore" which, taking in account the further explanations, was already incorrect, since it boots but doesn't obviously finish to load hildon Sep 01 18:13:40 kerio: if he had messed around with glibc or initscripts, he probably had mentioned that Sep 01 18:14:33 JZA: reflashing will reset your rootfs, but (if you don't flash eMMC/VANILLA image, sth usually not necessary) it will keep your user data incl messages and call log and all Sep 01 18:14:52 only thing you need to restore from backup are the applications themselves Sep 01 18:15:38 backup restore offers an option to select what to restore. DO NOT restore your user data, it would reset the still living data to the state of your last backup Sep 01 18:16:53 JZA: in any case *always* do a backup before you do any restore. So you can revert and possible damage done by restore, by restoring to the current backup you did Sep 01 18:17:22 no restore without previous immediate backup, never! Sep 01 18:17:53 seen users "restoring" their perfectly sane system to dead too many times Sep 01 18:18:06 s/dead/death/ Sep 01 18:18:06 DocScrutinizer05 meant: seen users "restoring" their perfectly sane system to death too many times Sep 01 18:18:57 JZA: if you run into trouble with simple flashing of rootfs/COMBINED, come back here ask for help Sep 01 18:19:33 do NOT flash eMMC/VANILLA Sep 01 18:19:43 okk Sep 01 18:19:55 :-) Sep 01 18:20:13 DocScrutinizer05: I remember this process I never did it because the USB didnt work for me. It had problems detecting the device. I am using Mandriva Linux Sep 01 18:20:27 I hope this time works better Sep 01 18:20:49 the key is to blacklist cdc_phonet Sep 01 18:21:23 instructions how to accomplish that are in wiki I pointed at Sep 01 18:21:41 I hope they apply to mandriva as well Sep 01 18:30:36 JZA: out of curiosity: what's been that "applied patch"? Sep 01 18:31:09 JZA: what exactly did you do? Sep 01 18:31:37 it was... a module that will give priority to userland process Sep 01 18:32:42 meant to be able make the calling trump other process on the bg Sep 01 18:34:13 ouch Sep 01 18:34:16 DocScrutinizer05: anyway I need to juice up the mobile first Sep 01 18:35:08 JZA: (patch) _all_ that is already in stock maemo Sep 01 18:35:12 by default Sep 01 18:35:58 since Nokia engineers had same brilliant idea, you know ;-) Sep 01 18:38:23 JZA: (juice up) does that mean your battery is low? if so, flashing may fail. You should be extremely careful about battery usage (even while charger plugged in it will still eat battery until you 'fixed' your problem!) Sep 01 18:39:26 JZA: so what can I do? Sep 01 18:40:06 test for battery (and other prerequisits for flashing): press and hold "u", plug in fastcharger, check if NOKIA screen with USB logo upper right is steady. If not, your battery is too weak Sep 01 18:40:44 you'll need to get your battery charged by external charger or other means independant of N900 Sep 01 18:41:12 ooh, this test of course needs to be done on powered-down device Sep 01 18:41:57 I.E. remove battery, reinsert battery, hold 'u', plug in charger Sep 01 18:42:25 after test renove battery again immediately, to make sure it won't deplete Sep 01 18:44:21 if device comes up with dim NOKIA screen but doesn't stay there with USB logo, then it's your battery that's too weak. If device doesn't come up at all, either your fastcharger or your battery or your USB receptacle is defect. OR you need a coldflash to restore NOLO bootloader (rarely ever needed) Sep 01 18:45:10 well this will take a while Sep 01 18:45:33 I won't elaborate now on distinguishing between those different defects, unless you actually got that problem Sep 01 18:46:10 JZA: removing and inserting battery, holding u and plugging in charger will take a while? Sep 01 18:46:15 how so? Sep 01 18:46:32 I can do all that in less than 60s Sep 01 18:46:32 charging the battery will take a while Sep 01 18:46:42 how fast can u do that? Sep 01 18:46:48 aaah, your battery is too weak? Sep 01 18:47:21 yeah Sep 01 18:47:26 and you got an external charger for charging it outside N900? Sep 01 18:47:31 no Sep 01 18:47:41 :-S Sep 01 18:48:05