**** BEGIN LOGGING AT Wed Aug 22 03:00:00 2012 Aug 22 04:37:41 any iOS developers here? Aug 22 04:53:10 Was just curious about maemo users take on iOS dev. I'm starting to make an iOS app, but I use an n900 of course Aug 22 05:23:35 LOL Aug 22 05:23:39 http://www.youtube.com/watch?v=AiVnMazRIII&feature=youtube_gdata_player Aug 22 05:23:47 disney's attack on opensource Aug 22 05:23:55 it's almost like joe camel all over again! Aug 22 10:09:16 Estel_, u awake? Aug 22 10:43:22 Hi. When I plug in my headset from my Nokia N9 into my Thinkpad T61p running Ubuntu 12.04 x64 the sound is pretty low untill I press a button on my headset (where the mic is). Does anyone had a similar problem and does anyone know how I can get louder sound without pressing that button? I had a similar problem with my headset form my Nokia N900. Aug 22 10:43:48 Well louisdk Aug 22 10:44:22 I have the prob that the extra line on the 3,5mm plug is evil Aug 22 10:44:40 I have to plug it in for 3/4, then the sound is good Aug 22 11:02:33 timo it's louis ck :) Aug 22 12:08:54 this is weird... everything seems to connect correctly in my n900 except apt-get update :) Aug 22 12:10:20 proxy? Aug 22 12:23:38 no, I'm connected through a shared connection on a Windows 7 computer. Aug 22 12:23:53 the funny thing is now one of the repositories managed to answer. Aug 22 12:24:22 first I received Network unreachable messages, then I got gzip errors and now it seems it might even work as expected :) Aug 22 12:26:02 at least for some repositories Aug 22 12:26:09 I guess I'll try again at home Aug 22 12:27:20 thanks for answering, anyway :) Aug 22 12:34:24 (ram expansion): starting point: http://www.hkinventory.com/public/OfferInventResult.asp?search=offerinvent&MainCate=&SubCate=&Order=&keywords=&words=&mode=&brand=&datecode=&MinQty=&iSearchCateCrit=&MultiSearch=0&pnums=KAT00F00&btnSubmit=Search&Crit1=0&Membership=0&orderby=0 Aug 22 12:36:28 then visit all the linked pages, to occasionally find detailed data about the product (RAM size, NAND size, pinout?) and maybe even links to datasheets Aug 22 12:37:03 also google for each type listed there Aug 22 12:37:43 after a day or two of work, you know 99.5% what component you would like to get to try the RAM expansion Aug 22 12:38:17 meh, no estel, no HtheB, nobody around who pestered me yesterday Aug 22 12:38:46 wmarone: ^^^ Aug 22 12:47:13 DocScrutinizer05: :) Aug 22 12:50:55 For upgrading the n900? Aug 22 12:51:20 yeah - it's messy. I was unable to find sources for the one part that I found that looked plausible. Aug 22 12:51:33 POP RAM isn't really a commonly sold item through disties. Aug 22 12:59:59 SpeedEvil: do you by any chance recall the part number, or where you found it? Aug 22 13:00:54 http://www.docstoc.com/docs/81716996/Cep-Telefonu-Tamir-Kursu for those who think it's easy to find the right part - this is as well stuff you need to read to gather any available bit of info Aug 22 13:02:10 or you contact some friend EE in TPE or HK who knows a bit about sourcing, and ask him to call Samsung about compatible PoP-chip with larger RAM Aug 22 13:02:14 I vaguely recall I went through the micron part selector Aug 22 13:02:26 and it came up with a couple that looked at least plausible. Aug 22 13:02:35 :nod: Aug 22 13:02:37 - the package seems custom for the OMAP Aug 22 13:02:59 So an POP138 (???) parts have a reasonable chance of working. Aug 22 13:03:21 the approach to use a complete SoC PoP stack stolen from another device seems best approach Aug 22 13:03:32 Indeed. Aug 22 13:03:50 Ideally a stack with a 720MHz rated processor under it :) Aug 22 13:04:12 :nod: Aug 22 13:04:36 Some other devices are lots easier. The nexus 7 RAM looks almost trivial to upgrade. Aug 22 13:04:45 also way "easier" to swap a whole SoC than just the PoP Aug 22 13:05:06 Well - as I understand it, the stacks are not glued together in any way Aug 22 13:05:10 they're just balled together Aug 22 13:05:47 but yes, getting one to melt without the other... Aug 22 13:06:03 if nothing else works, you also could do a piggyback PCB on top of CPU that interfaces to POP138(?) on the bottom and to two or more chips for RAM and NAND on top side Aug 22 13:06:21 :) Aug 22 13:07:51 you could also design and build your own phone Aug 22 13:08:15 http://www.mauve.plus.com/opensourcehw.txt Aug 22 13:08:15 so where are those eager hackers who asked about how to destro... err expand their RAM? Aug 22 13:08:22 It's unfortunately complex. Aug 22 13:08:30 r00t|home: Aug 22 13:09:06 And I want to, but I have no energy. Waaaaay more important stuff is unfortunately getting neglected. Aug 22 13:09:15 I actually already connsidered cloning N900 exactly, but even that is more than difficult, regarding the sourcing situation Aug 22 13:09:38 Paint the roof of the gazebo, replace the sheds walls, fix my bathroom, ... Aug 22 13:09:46 indeed. Aug 22 13:10:10 I keep looking at teeny miniPCI 3G modems, and my crap android tablet. Aug 22 13:10:25 if anybody is eager to go that route, why not start with GTA04? Aug 22 13:10:55 (crap in comprison to my n7, it's OK, but comparing a 800MHz proc to 1.4GHz) Aug 22 13:11:03 Lack of actual hw. Aug 22 13:24:01 Estel_ Aug 22 13:24:02 ping, poke.. or whatever Aug 22 13:24:02 just wake up Aug 22 13:24:02 :D Aug 22 13:24:09 hm? Aug 22 13:24:49 mplayer /home/user/porn.mpeg Aug 22 13:24:55 ops, why you're disturbing me, HtheB? Aug 22 13:25:09 accourding to my nephews, it will work Aug 22 13:25:10 they can swap the chip without problems Aug 22 13:25:10 since there s no silicon next to the chip Aug 22 13:25:40 sounds great, although, from experienc,e I wouldn't be so sure about that. CPU stuck between it and PCB isn't problem? Aug 22 13:25:43 why not MyDocs? your home dir will be full in no time :p Aug 22 13:25:52 nope Aug 22 13:25:52 he, he Aug 22 13:25:55 no prob Aug 22 13:26:16 the problem is: finding the chip Aug 22 13:26:20 read the topic Aug 22 13:26:22 someone replied Aug 22 13:26:30 2gb ram + 4gb flash Aug 22 13:26:32 well, I've heard about people who're able to do things, that seasoned E engineers state as impossible, so, it's not that I don't belive it. Maybe You're lucky and have skilled nephew Aug 22 13:26:46 I'll check it, but 2GB RAM sounds like a no-go for our omap Aug 22 13:27:02 talk.maemo.org/showthread.php?p=1254522#post1254522 Aug 22 13:27:05 http://talk.maemo.org/showthread.php?p=1254522#post1254522 Aug 22 13:27:08 4GB flash pose some problems, too (but not to big, can be overcome by big enough microSD) Aug 22 13:27:18 sure, sure, i'll check it :) Aug 22 13:27:24 thanks for pointers Aug 22 13:35:23 do you think it's in GB or in Gb? Aug 22 13:44:20 Estel_, still watching pr0n? Aug 22 13:44:33 nooope, reading TMO Aug 22 13:44:41 more porn there, daily Aug 22 13:44:52 exactly Aug 22 13:44:53 :p Aug 22 13:47:04 Estel_, what can we do with more ram on the N900? Aug 22 13:47:15 port harmattan? :p Aug 22 13:47:20 everything, You know what RAM is used for Aug 22 13:47:33 well, harmattan programs already run on n900 via meecolay or something like that Aug 22 13:47:47 most likely, chromium and iceweasle is going to start flying in super speed Aug 22 13:48:04 and every other Ed program (which worked well even now, with more RAM, it would literally fly) Aug 22 13:48:13 what friggin topic? Aug 22 13:48:45 DocScrutinizer05, well, it's progressing ;) Aug 22 13:49:04 no, it's a prime number Aug 22 13:49:43 [2012-08-22 15:26:30] 2gb ram + 4gb flash [2012-08-22 15:26:46] I'll check it, but 2GB RAM sounds like a no-go for our omap Aug 22 13:49:50 sort your nomenclature Aug 22 13:50:05 nope, i'll check thread Aug 22 13:50:15 2GB rAM is a no go for our omap no matter of my checking ;) Aug 22 13:50:19 don't cut it out of context. Aug 22 13:50:36 read the topic Aug 22 13:50:44 I'll check it Aug 22 13:50:57 talk.maemo.org/showthread.php?p=1254522#post1254522 Aug 22 13:51:01 http://talk.maemo.org/showthread.php?p=1254522#post1254522 Aug 22 13:53:32 those chips are always bit size, never byte size since your arch could use 8, 16, 32 words Aug 22 14:00:37 so you mean that it's 256 MB (4Gb) RAM and 512 MB (4Gb)? Aug 22 14:01:02 DocScrutinizer05, ^^^ Aug 22 14:01:51 (unless it's consumer retail flash) Aug 22 14:02:31 4Gb / 8 Aug 22 14:02:42 = GB Aug 22 14:02:48 yep Aug 22 14:03:06 thats why I calculated it as 256 MB RAM and 512 MB Flash Aug 22 14:03:25 reread your post Aug 22 14:03:30 unless, as said, it was MB, and someone just mixed Mb with MB when writing about it (haven't checked topic yet) Aug 22 14:03:37 oh god, you guys are actually swapping out mounted RAM now? Aug 22 14:04:03 some guys again ponder it ;-P Aug 22 14:04:11 I thought I've seen crazy, but doing it on a modern SOC tops it Aug 22 14:04:27 I'd suggest to swap complete SoC PoP stack though Aug 22 14:04:39 not as simple as 80s/90s hardhacks Aug 22 14:05:44 actually redoing PoP stacking is pretty hard, since it's not a normal PCB and normal soldering, and no fresh virgin device on one end at least Aug 22 14:06:44 so when can I get a refurbished n900 with 32GB RAM? :) :) Aug 22 14:06:50 swapping the complete SoC (incl PoP stack) is maybe feasible Aug 22 14:07:33 DocScrutinizer05, sure, but finding compatible OMAP in stack isn't harder than finding just OneNAND? Aug 22 14:07:45 not at all Aug 22 14:07:54 get it from palm or whatever Aug 22 14:08:15 pandaboard, pandora... Aug 22 14:08:26 sometimes I forget 32bit addressing is limited to 4GB Aug 22 14:09:22 DocScrutinizer05, sure, but how to make N90 0work with such omap Aug 22 14:09:41 are You sure that closed things aren't fixed for our omap? Wouldn't it need to be *very* similar? Aug 22 14:09:49 well, it has to be same processor of course Aug 22 14:09:53 same chip Aug 22 14:10:26 but, same model numbe,r as ShadowJK mentioned, ranged from 500 mhz overclocked to 600 from star,t to 1GHZ not overclocked Aug 22 14:10:43 9and working on much lower voltage) Aug 22 14:10:46 and the only problem I see might be with TrustZone and cert Aug 22 14:10:55 hm Aug 22 14:12:10 the binning (selection of chips to different speed/voltage steppings) might be neglectable, though you for sure want some "better" chip bin than the one we already got Aug 22 14:12:35 there won't be a problem with underclocking / overvolting Aug 22 14:13:08 at least it's for sure less of a problem than OC&undervolting Aug 22 14:13:17 ;-) Aug 22 14:13:32 ;) Aug 22 14:13:53 get a chip with exactly same label and you're probably fine Aug 22 14:15:02 TI had no motivation to change footprint/pinout of omap3430 series, since that would introduce problems that aren't needed for anything Aug 22 14:15:45 but you should make damn sure you get Nokia xloader running on your new CPU Aug 22 14:15:58 so... what should I do Aug 22 14:16:04 with the new root key in ROMBL Aug 22 14:16:06 talk in English please :p Aug 22 14:16:28 (plain english) Aug 22 14:16:30 :D Aug 22 14:16:57 I'm not sure how the bootoption pins are configured on N900, so whether device even boots up in secure mode or not Aug 22 14:16:59 Estel_, just point me out what I should do :) Aug 22 14:17:24 jacekowski should know about that stuff afaik Aug 22 14:20:06 again with this? :D Aug 22 14:21:51 javispedro: hi! yeah, seems somebody found an iron-artist who claims he could do a BGA swap Aug 22 14:22:32 javispedro: so the old dream of 0.5, 1, even 2 or 4 GB of RAM resurrected Aug 22 14:22:58 meh Aug 22 14:23:00 it's probably as immortal as the desire to OC Aug 22 14:23:26 DocScrutinizer05, well, trhumb was also claimed by many* gurus to be impossible ;) Aug 22 14:23:45 more like "to be useless" Aug 22 14:23:49 also? Aug 22 14:23:51 and Oc works fine without any trustable report of damage, for ~3,5 years... so, it's not exactly best argument against it ;) Aug 22 14:23:52 (and I still claim it :) ) Aug 22 14:24:02 javispedro, You may claim it, but facts seem to prove otherwise Aug 22 14:24:10 swap usage dropped considerably Aug 22 14:24:20 check maemo-developers mailing list Aug 22 14:24:22 and yes, nobody ever said thumb is impossible afaik. It always been "don't think about it!" Aug 22 14:24:30 20 MiB is not "considerably" Aug 22 14:24:40 20 MB out of 256 IS considerably. Aug 22 14:24:48 not to mention that it's more or less beginning Aug 22 14:24:55 add to this performance benefits from new toolchain ;) Aug 22 14:25:12 well, the latter is true. Aug 22 14:25:18 remembe,r we still don't have everything sitting in ram as bin; compiled to thumb Aug 22 14:25:34 and you will never have because a lot of it is closed source. Aug 22 14:25:47 as for RAM swap, as said dozens of time, I don't hold my breath for it, but I definitelly keep thumbs (sic!) for this efforts Aug 22 14:26:15 as too many times I've seen "don't even think about it" or "impossible" from experienced people, that later was proven to be just matter of discipline and effort in achieving it Aug 22 14:26:19 estel, even if I swap the ram with the palm pre plus one Aug 22 14:26:23 well, the coll. term is "crossing fingers" Aug 22 14:26:25 dont i have to reprogram it? Aug 22 14:26:42 HtheB, it's the same omap - exactly same - so such need shouldn't arise Aug 22 14:26:49 Estel_: that is a false position Aug 22 14:26:57 forother, follow DocScrutinizer05's advice to ask jacekowski about trusted zone Aug 22 14:27:04 Estel_: most of what has been said as impossible has been, well, impossible. Aug 22 14:27:09 freemangordon may know something about it to, due to work on thumb Aug 22 14:27:14 HtheB, it's more or less unknown ground Aug 22 14:27:20 no one outside TI know it for sure Aug 22 14:27:26 you won't know until You check it Aug 22 14:27:49 javispedro, sure, I'm talking about things that were said as impossible, because most people wasn't willing to show enough effort to try Aug 22 14:27:53 like thumb ;) Aug 22 14:28:07 so if this is successful, did I made something awesomeness for our community? Aug 22 14:28:08 well, everybody outside TI knows, as soon as you read complete SPRUF98D Aug 22 14:28:23 HtheB, definitelly Aug 22 14:28:32 someone will have to find a N900 they are willing to sacrifice for the cause of identifying if its possible to upgrade any hardware chips in N900 Aug 22 14:28:49 *if* you would manage to make N90 0fully usable with 512 MB or 1GB of RAM, it would be rather "breakthrough" than a "just awesome" Aug 22 14:28:57 i think i have a mainboard laying around Aug 22 14:29:01 at my nephew Aug 22 14:29:04 jonwil, agaik HtheB want to sacrifice his own device Aug 22 14:29:18 indeed, a front page headline in MWKN Aug 22 14:29:21 someone is gonna do the dirty job... so Aug 22 14:29:35 it is not going to be a breakthrough Aug 22 14:29:38 DocScrutinizer05: it might turn out that some wiring is needed so RAM chip to correctly recognized, never search for that in TRM though Aug 22 14:29:41 because as DocScrutinizer05 once put rate of success is very low Aug 22 14:29:45 and possibly a business opportunity for 50..300 reworks Aug 22 14:29:47 HtheB, basically, it *would* be vallaha of hardware side, not lesser than thumb and new compiler on software side Aug 22 14:30:09 javispedro, but rate of success was purely made up Aug 22 14:30:14 no one tried it, so no one know Aug 22 14:30:19 freemangordon: quite possible Aug 22 14:30:23 Estel_: what do you mean by "no one tried"? Aug 22 14:30:32 if HtheB nephew will replace ram in 20 devices, we will have good start for calculating chances Aug 22 14:30:42 javispedro, no one tried, physically, to replace N900 RAM Aug 22 14:30:45 oneNAND Aug 22 14:30:47 Estel_: _everyone_ is _everywhere_ trying to reball stuff Aug 22 14:30:47 to be precise Aug 22 14:30:52 or whole PoP Aug 22 14:30:57 Estel_: I've tried to reball a Palm T|X OneNAND chip Aug 22 14:30:59 Estel_, did you check ur pm? Aug 22 14:31:00 Estel_: to much failure Aug 22 14:31:09 Estel_: So stop making it trivially easy -- it is not. Aug 22 14:31:17 javispedro, OTOH, HtheB's nephew claims to have replaced dozens balls eMMC's Aug 22 14:31:23 heh Aug 22 14:31:24 No wonder Aug 22 14:31:24 on phones Aug 22 14:31:40 eMMC is the easier of the pack because it is more or less standard MMC interface (few pins) Aug 22 14:31:42 they can change it, if there is no silicon on the chip itself Aug 22 14:31:53 HtheB: that's the second to most absurd thing I've heard today. Aug 22 14:31:55 and on the N900, they didnt use silicone Aug 22 14:32:01 I kinda play devil advocate here, as - again - I don't hold my breath - but, it's definitelly worth trying, if someone have means and will to do Aug 22 14:32:33 javispedro, agreed about eMMC Aug 22 14:32:38 I aLSO think that it's going to be harder Aug 22 14:32:45 its some sort of glue that is on the sides of the emmc Aug 22 14:32:48 please note that there's a reason that many of the open source hardware improvements (e.g. GTA04) replace entire PCBs never single chips Aug 22 14:32:52 well, person who is accounted for doing it saw board Aug 22 14:33:15 and rated chances high. I'ts possible, and even likely, that he overrated his chances Aug 22 14:33:26 but it's also possible that he is better skilled in practice, that we are in theoretizing ;P Aug 22 14:33:38 no point in discouraging, before actual attempt Aug 22 14:33:57 of course with all warnings, that fail is most likely fatal for chip, board, or both Aug 22 14:34:44 javispedro, if I would have infra-red soldering station-lab (this one that also ultra-fast pre-heat parts of pcb), I would already "waste" dozens of N900's motherboards Aug 22 14:35:13 when HtheB said, that his nephew want to do it with hot air station, I also rised my brew. But, heck, in fact, what doI know about his practice, methods, skills, etc? Aug 22 14:35:27 Estel_: our most gifted EE at OM tried to fix defective soldering on 10 proto PCB, resoldered S3C64000. Yield: 30% Aug 22 14:35:38 I know, i remember it Aug 22 14:35:52 well, to be sure about N900, there is only one way... :P Aug 22 14:36:20 OMAP footprint is no way different to S3C64000 Aug 22 14:36:26 after all, even failed attempt could produce some ncie documentation Aug 22 14:36:29 nice* Aug 22 14:36:29 it is actually worse... Aug 22 14:36:37 * Estel_ nods @ javispedro Aug 22 14:36:40 but of course the skills may differ, and the PCB footprints as well Aug 22 14:36:53 guys, just tell me what I have to get and which chip i have to swap with... thats all I need to know as a hardmware n00b Aug 22 14:37:08 meh @ practice methods and skills :) Aug 22 14:37:46 OTOH, again, it's known fact, that in some exotic countries (sorry for calling turkey exotic, no offense meant), are people that successfuly master to do absolutely crazy modifications - just because they want to sacrifice time and, some times, devices for training Aug 22 14:37:52 again, not holding my breath, but... Aug 22 14:37:52 somebody claimed palm pre+ has same omap3430 but with 512M RAM Aug 22 14:38:09 DocScrutinizer05, yea, it was robbiethe1st. Isn't it true? Aug 22 14:38:33 I do not think it has the same OneNAND amount at least though Aug 22 14:38:33 HtheB: get that phone, remove SoC carefully there, and replace OMAP3430 of N900 with the one you saved from the other phone Aug 22 14:38:35 javispedro, as for merit, do You know how to find suitable part? No matter if in low or big quantities? Aug 22 14:38:56 HtheB, palm pre plus SoC is good candidate, indeed Aug 22 14:39:04 ok Aug 22 14:39:20 DocScrutinizer05 suggets to replace BOTH processor and OneNAND, aka whole SoC. It's probably easier and have more chance of success than only OneNAND. Aug 22 14:39:23 Estel_: see SpeedEvil's comments Aug 22 14:39:28 in backscroll Aug 22 14:39:29 If Your nephew is absolutely sure, he may try to replace oneNAND only Aug 22 14:39:36 the omap chip, that isnt the samsung one, right? Aug 22 14:39:53 the omap chip is a TI chip Aug 22 14:39:57 DocScrutinizer05, lost backscroll, I was doiong some maintentance on power line today Aug 22 14:40:13 HtheB, Samsung is OneNAND Aug 22 14:40:18 under it, there is Omap Aug 22 14:40:28 both OneNAND and omap are SoC Aug 22 14:40:30 that's called Package on Package Aug 22 14:40:32 system on a chip Aug 22 14:40:33 or PoP Aug 22 14:40:48 PoP, not SoC, true, SoC is somewthing little different Aug 22 14:40:49 anyway Aug 22 14:40:52 Samsung is OneNAND Aug 22 14:40:52 OneNAND is no SoC Aug 22 14:40:56 yes, yes Aug 22 14:41:00 SoC is under OneNAND Aug 22 14:41:04 CPu and DSP, yes? Aug 22 14:41:09 (soC) Aug 22 14:41:36 HtheB, to make it more clear. From top, to bottom (bottom is PCB) Aug 22 14:41:46 and GPIO and x-bridge and whatnot, and cam/LCD interface and and and Aug 22 14:41:49 OneNAND (with Samsung mark) -> OMAP -> PCB Aug 22 14:42:09 DocScrutinizer05, right, System on a Chip = many things bundled into one chip Aug 22 14:42:17 yep Aug 22 14:42:36 HtheB, confused, or is it understandable?;) Aug 22 14:42:46 confused Aug 22 14:42:59 HtheB: it's a sandwich Aug 22 14:43:03 just edit the picture forme with circels :p Aug 22 14:43:04 again. On photo You have seen Aug 22 14:43:11 there will be only one circle Aug 22 14:43:15 samsung = OneNAND Aug 22 14:43:21 Omap is *under* it Aug 22 14:43:26 sandwich, as said Aug 22 14:43:35 the CPU/SoC is *under* the Samsung MEM Aug 22 14:43:37 You don't see it, it's between Samsung (OneNAND) and PCB Aug 22 14:43:54 OneNAND is soldered to OMAP, and OMAP is soldered to PCB Aug 22 14:44:04 thus good luck on desoldering OneNAND only Aug 22 14:44:10 (leaving OMAP) Aug 22 14:44:16 and soldering replacement Aug 22 14:44:17 that' Aug 22 14:44:38 why DocScrutinizer05 suggested replacing *whole* chip, i.e. OMAP+ OneNAND Aug 22 14:44:47 even if you make it to unsolder the onenand PoP from the SoC layer, the balls are messed and resoldering is pretty hard Aug 22 14:44:51 i.e. desoldering it to the PCB, from both N900 and palm pre plus, and replacing it Aug 22 14:45:19 HtheB - generally, tell you nephew to do as pleases him, he may resolder whole chip, or half of it :D Aug 22 14:45:24 if he manages to Aug 22 14:45:27 :D Aug 22 14:45:41 just called him Aug 22 14:45:52 and you don't know if fab, on packaging (soldering) the Soc+OneNAND sandwiches, did even use glue to avoid later desoldering when soldering the package to PCB Aug 22 14:45:56 he had a n900 board laying somewhere Aug 22 14:46:03 so, lets see if he can deattach it Aug 22 14:46:15 yea Aug 22 14:46:23 now I understand what you mewan Aug 22 14:46:24 mean Aug 22 14:46:30 so it might be impossible to unsolder the OneNAND from SoC Aug 22 14:46:34 its just like a b-day cake Aug 22 14:46:37 this, + what DocScrutinizer05 said, make it more feasible to replace whole PoP (whole chip) Aug 22 14:46:40 cream on the inside :p Aug 22 14:46:48 yes Aug 22 14:46:53 PCB is cake, OMAP is cream Aug 22 14:47:12 OneNAND is one cake, PCB is another, OMAP is cream Aug 22 14:47:17 bon apetit Aug 22 14:47:23 ._. Aug 22 14:47:50 what is easier, is to remove both cake and cream (OneNAND + OMAP), and replace it with other cream bundled with bigger cake Aug 22 14:48:11 http://en.wikipedia.org/wiki/Package_on_package Aug 22 14:48:21 theoretically, it's also possible to leave cream, and just replace upper cake with bigger, but it's harder, and may be even impossible (if they used soldering + glue) Aug 22 14:48:30 I'm getting hungry Aug 22 14:49:41 hmmm :/ Aug 22 14:49:45 hah, leant PoPt is the oneNAND and PoPb is the SoC Aug 22 14:50:17 DocScrutinizer05, don't confuse him even more Aug 22 14:50:37 all that is needed for purely hardware'ish attemmpt on replace is to remove whole mess from pcxb and replace it with other mess from palm pre plus Aug 22 14:50:42 nah, it's all easy, wikipedia FTW Aug 22 14:50:46 yep Aug 22 14:50:55 tbh never heard of that terminology :) Aug 22 14:51:13 either way this is not true 3d stacking, the memory controllers are still on the peripherty Aug 22 14:51:15 * DocScrutinizer05 neither Aug 22 14:51:26 but it's pretty clear, and I like it Aug 22 14:51:32 (aka ddr's pins will go to the pcb) Aug 22 14:52:50 can you clone the nokia keys/first stage bootloader etc from the old soc? (if you use a palm pre+ one) Aug 22 14:53:06 >>Any mechanically mating top package can be used. For a low-end phone, a smaller memory configuration may be used on the top package. For a high-end phone, more memory could be used with the same bottom package.[1] This simplifies inventory control by the OEM<< Aug 22 14:53:25 the wonders of JEDEC standards Aug 22 14:53:47 jaska: nope Aug 22 14:53:55 thats what i thought. Aug 22 14:54:10 jaska: that's what I already mentioned as maybe worst problem in this approach Aug 22 14:54:51 see above "make damn sure your NOLO/Xloader will work with the new key in ROMBL" Aug 22 14:55:16 javispedro: hmm? Aug 22 14:57:27 javispedro: http://www.google.it/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&ved=0CG0QFjAG&url=http%3A%2F%2Fe2e.ti.com%2Fcfs-file.ashx%2F__key%2FCommunityServer-Discussions-Components-Files%2F353%2F7416.MT29C4G48MAZAPAKQ_2D00_5-IT_5F00_Micron-168_2D00_Ball-NAND-Flash-and-LPDDR-PoP-_2800_TI-OMAP_2900_-MCP.pdf&ei=iNc0UPGaAuqk4ASu64CABw&usg=AFQjCNEhPq6ICqDMejrYR7F8IJe7NPkIZA&cad=rja Aug 22 14:57:38 DocScrutinizer05: the Wikipedia page drawing is slightly misleading Aug 22 14:58:03 err, is it? Aug 22 14:58:22 is the SoC of smaller size than the PoPt? Aug 22 14:58:48 it is entirelly correct =) Aug 22 14:58:53 but yes, the SoC is smaller Aug 22 14:59:47 I don't think so, since the outer pad "ring" on PCBA is same pitch and 3 rows wide Aug 22 15:00:09 the diagram on sprabb3.pdf is more clear IMHO Aug 22 15:00:22 (it is the same though, I was just confused by wikipedia's one) Aug 22 15:00:44 basically, SoC's package is smaller, it is just pre-soldered on a PCB. Aug 22 15:00:45 eeek sprabb3.pdf ?? Aug 22 15:00:57 sounds like spruf98d Aug 22 15:01:03 DocScrutinizer05: no, goog it. Aug 22 15:02:22 javispedro, got a new build of sowatch for me yet? Aug 22 15:02:27 Fig.1 looaks absolutely identical Aug 22 15:02:31 GeneralAntilles: which version are you using? Aug 22 15:02:38 An old one. Aug 22 15:02:54 GeneralAntilles: http://depot.javispedro.com/metawatch/sowatch/sowatch_0.4.1_armel.deb Aug 22 15:03:31 0.3.1 Aug 22 15:03:36 DocScrutinizer05: you mean Fig2? Aug 22 15:03:54 well, that too Aug 22 15:04:09 (it is the same though, I was just confused by wikipedia's one) =) Aug 22 15:07:58 javispedro, aweeesomeeee Aug 22 15:14:28 nandtester here,is it possible to extract the stuff that was erased by nandtesting from a working device & "import" it into the broken one without soldering? Aug 22 15:16:26 cold flash isn't exactly that? Aug 22 15:16:35 (see maemo's wiki for cold flashing) Aug 22 15:29:46 sambo7: see Estel_^^^ Aug 22 15:30:20 sambo7: how would you "import" on a device that doesn't even boot NOLO? Aug 22 15:31:18 beyond that, find a nandread image of CAL partiton on maemo.cloud-7.de Aug 22 15:31:50 beware, I never dared to test restoring such image via nandwrite Aug 22 15:32:00 javispedro, love the update. Aug 22 15:32:08 o/ gan Aug 22 15:32:16 'Morning Aug 22 15:33:38 I treid the cold flash procedure...the deivce does load the temporary bootloader...but that's all it does...found a version of phoenix, but won't be able to test before friday Aug 22 15:34:37 sambo7: so if coldflash gives you semi-working device (bootable) but things like WLAN locale, WLAN MAC, BT MAC etc still fail due to erased CAL partition, feel free to try "importing" that CAL image to your CAL partition Aug 22 15:36:24 sambo7: talk to jacekowski about coldflashing. It's not exactly trivial, and I think you will need several consecutive steps to recover from your overkill damage Aug 22 15:37:45 device doesn't boot properly-get's stuck after loading the bootloader...jacekowski said that it's maybe to recover with the phoenix sw Aug 22 15:38:00 Pali also knows about coldflashing Aug 22 15:38:03 basically coldflashing should just flash the xloader&NOLO combo to the mtd0 partition, then this standard NOLO should be able to flash the rest like usual Aug 22 15:38:47 I.E. your device isn't supposed to work immediately after coldflashing Aug 22 15:39:06 it's just restored to a state where you can do a normal flashing Aug 22 15:39:11 is the dd_if=mtd1ro.bin the CAL image? Aug 22 15:39:19 when coldflashing, flasher send omap peripheral boot message to bootrom, then flasher send 2nd xloader image and omap bootrom will start that 2nd image Aug 22 15:39:23 one of them, yes Aug 22 15:39:41 there are others, done by different tools Aug 22 15:39:54 2nd image waiting for some nokia x.loader messages on usb. one message is "send nolo" Aug 22 15:40:00 though I seem to recall in the end a diff shown 0 differences Aug 22 15:40:31 flasher will send nolo and and 2nd image will flash both 2nd and nolo to nand bootloader partition and 2nd will reboot device Aug 22 15:41:02 so coldflashing erasing and writing new 2nd+nolo to bootloader nand partition Aug 22 15:41:15 I'm rewriting 0xFFFF code Aug 22 15:41:24 err, yes. If I parsed that correctly Aug 22 15:41:33 it will support more options Aug 22 15:41:47 full support of fiasco images (generating, unpacking) Aug 22 15:41:55 also cold flashing Aug 22 15:41:55 \o/ Aug 22 15:42:17 I have some dumps of Mk II flashing protocol, so maybe this will be implemented too Aug 22 15:42:32 * DocScrutinizer05 waves and heads out for a shower. Not that the mere temperature is as insane as last 2 days Aug 22 15:42:41 ~weather eddn Aug 22 15:42:44 Nuernberg, Germany; (EDDN) 49-30N 011-03E 318M; last updated: 2012.08.22 1520 UTC; Dew Point: 55 F (13 C); Pressure (altimeter): 30.00 in. Hg (1016 hPa); Relative Humidity: 41%; Temperature: 80 F (27 C); Visibility: greater than 7 mile(s); Wind: from the W (280 degrees) at 9 MPH (8 KT) Aug 22 15:42:48 (Mk II is used for flashing eMMC, but support everything except rootfs) Aug 22 15:42:59 Pali: does N900's x-loader or nolo verify any signature? Aug 22 15:43:04 also it support flashing when maemo is running Aug 22 15:43:22 javispedro, only omap bootrom checking for x-loader/2nd signature Aug 22 15:43:26 javispedro: afaik not Aug 22 15:43:27 nolo is unsigned on n900 Aug 22 15:43:38 Pali: and the key for that is on omap's e-fuses? Aug 22 15:43:39 but ROMBL is verifying sig of xloader/NOLO Aug 22 15:44:06 though, that can't be entirely true since you can patch NOLO Aug 22 15:44:20 no idea how device checking signature Aug 22 15:44:27 it is not written in omap trm Aug 22 15:44:38 it allegedly is Aug 22 15:44:43 I know that there's a 4x4byte e-fuse block with a sha1sum Aug 22 15:44:47 I don't know if it's used for anything Aug 22 15:45:09 jacekowski mentioned efuse Aug 22 15:45:17 in TrustZone context Aug 22 15:45:18 (I don't even know how to read it from userspace... but it should be possible) Aug 22 15:45:44 I will write pancake to release new version of 0xFFFF. I fixed support for fiasco and it should be new stable version Aug 22 15:45:49 well, THAT is a topic for our thumbhacker Aug 22 15:45:59 and version where will be coldflash support will be released later... Aug 22 15:46:14 freemangordon: can efuse be read from userland? Aug 22 15:46:16 DocScrutinizer05: more like a deadlock for the above's proposed SoC change unless you use fresh-from-factory SoCs Aug 22 15:46:22 *s/deadlock/roadblock Aug 22 15:46:33 :nod: Aug 22 15:47:50 GeneralAntilles: be sure to mention any problems ;P Aug 22 15:48:03 DocScrutinizer05, I looked on nokia phoenix and it using NOLO interface for flashing Aug 22 15:48:11 also sending 2nd for coldflashing Aug 22 15:48:17 javispedro: maybe we'd need to create our own palmpre-xloader&NOLO image to make that SoC boot Aug 22 15:48:17 javispedro, the watch list probably needs a titlebar of some kind. Aug 22 15:48:42 also using fiasco images (for rx-51) Aug 22 15:48:44 javispedro, enhancement request: Ability to select font sizes for watchlets. Aug 22 15:48:54 DocScrutinizer05: palmpre's stock bootloader, bootie, is open enough (more open than nolo's fwiw) Aug 22 15:49:04 so if phoenix has support for writing to CAL it must use 2nd or NOLO for that Aug 22 15:49:39 yep Aug 22 15:49:54 unless they go JTAG Aug 22 15:50:13 GeneralAntilles: not many choices Aug 22 15:50:19 the problem was that after the device responded to the first command "flasher-3.5 -c -h RX-51:2101 -2 2nd.bin-RX-51\:2101\,2102\,2103 -s secondary.bin-RX-51\:2101\,2102\,2103 -S usb" Aug 22 15:50:39 which would mean they need special fixture for phoenix Aug 22 15:50:47 it didn't respond to "flasher-3.5 -F image.bin --flash-only=nolo -f" because usb connection was lost Aug 22 15:50:47 GeneralAntilles: the font is 12pt, at 10.5pt it is also legible (but not at 10 or 9pt, for example 'i's already look horrible) Aug 22 15:50:58 GeneralAntilles: all of the other fonts are caps-only Aug 22 15:51:09 sambo7, use: flasher-3.5 -h RX-51:2101 -F -c -S usb Aug 22 15:51:19 sambo7: sure, you need to reboot Aug 22 15:51:20 this should cold flash full fiasco image Aug 22 15:51:34 for specified hw rev Aug 22 15:52:01 errr, you even should flash NOLO in first cmd I'd guess Aug 22 15:52:06 Pali without the1st command "flasher-3.5 -c -h RX-51:2101 -2 2nd.bin-RX-51\:2101\,2102\,2103 -s secondary.bin-RX-51\:2101\,2102\,2103 -S usb" ? Aug 22 15:52:08 * javispedro ponders if stock fremantle boots with zero'd config partition Aug 22 15:52:48 that's the 1 mio dollar question Aug 22 15:52:53 javispedro, default maemo kernel - NO Aug 22 15:53:17 nolo read partition table from CAL and pass it via ATAG to kernel Aug 22 15:53:33 maemo kernel does not have hardcoded partition table Aug 22 15:53:36 oh, though it was hardcoded on nolo Aug 22 15:53:42 or kernel :) Aug 22 15:53:43 meego (and upstream) kernel yes Aug 22 15:53:56 javispedro, maybe there is hardcoded in NOLO Aug 22 15:54:07 but in CAL is partition table too Aug 22 15:54:11 :/ Aug 22 15:54:24 maybe it is time to create WIKI page about CAL partition Aug 22 15:54:34 what all and in which format is written in CAL Aug 22 15:54:52 (e.g. where is stored partition table, where wifi mac address, ...) Aug 22 15:54:52 sambo7: if Pali is right (no doubt he is) you don't need any other command Aug 22 15:55:28 sambo7, only that one command: flasher-3.5 h RX-51:2101 -F -c -S usb Aug 22 15:55:28 Will try that later, after back home from sports...already excited Aug 22 15:55:42 also make sure, you have usb 1.x port Aug 22 15:55:47 I have problems with usb 3.0 Aug 22 15:56:03 Pali: please spellcheck Aug 22 15:56:17 flasher-3.5 h RX-51:2101 ? Aug 22 15:56:22 ok,will definitely try that later....having my fingers crossed Aug 22 15:56:30 flasher-3.5 -h RX-51:2101 maybe Aug 22 15:56:37 yes Aug 22 15:56:38 -h Aug 22 15:56:55 javispedro, I'd take a caps-only. Aug 22 15:56:56 flasher-3.5 -h RX-51:2101 -F -c -S usb Aug 22 15:57:25 if I only nandtested with a -k I wouldn't have got into that mess Aug 22 15:57:39 yeah :-S Aug 22 15:57:51 and sending 2nd image working if battery is empty (on charger) too Aug 22 15:58:04 on the bright side, if you recover from this, you're kind of a hero Aug 22 15:58:25 but flashing other parts (kernel, rootfs) from NOLO not working... Aug 22 15:58:55 ok,will be back later guys...need to go & do some sports---thanks a lot Aug 22 15:58:57 if battery is empty after coldflashing usb icon will blink Aug 22 15:59:03 yeah, charged battery is MANDATORY Aug 22 15:59:11 javispedro, actually, I think screen resolution may be a partial answer to Pebble's bigger following. Aug 22 15:59:21 Pali: you amaze me with your knowledge Aug 22 16:00:05 if uboot will be bootable from x-loader/2nd and battery charging will be in u-boot we can charge empty battery with coldflashing... Aug 22 16:00:20 long pending plan Aug 22 16:00:38 does anyone know how to make charge21.sh script use 950mA? Aug 22 16:00:53 vi_: check maemo.cloud-7.de Aug 22 16:00:58 GeneralAntilles: ignoring the entire e-paper crap (which people still believe seemingly), I am indeed intrigued by their higher specs Aug 22 16:01:15 Looks like a monster to wear on the wrist, though. Aug 22 16:01:30 I have some problems with u-boot SPL (something like x-loader, bootloader for u-boot)... but I was able to write something to serial console in n900 qemu Aug 22 16:02:21 DocScrutinizer05: I am looking, cannot see charge21.sh Aug 22 16:02:29 :-S Aug 22 16:02:51 http://maemo.cloud-7.de/maemo5/patches_n_tools/ Aug 22 16:02:54 but when u-boot spl will be working (it is small image) we can replace nolo - but without flashing support... Aug 22 16:02:57 500mA Aug 22 16:03:30 vi_: I haven't tested it, but should just work Aug 22 16:03:42 diff to original to see what I changed Aug 22 16:04:20 btw, can somebody provide some flashing dumps? I have LD_PRELOAD library for flasher-3.5 which print info usefull for RE flashing Aug 22 16:04:40 If somebody has different HW revs (not 2101) it will be usefull Aug 22 16:05:23 in new prepaired version of 0xFFFF only loading, sending and image is missing... Aug 22 16:05:42 vi_: I always thought ShadowJK's original already uses 950mA Aug 22 16:05:44 *flashing Aug 22 16:06:50 brb Aug 22 16:07:09 bbl Aug 22 16:17:07 pali Aug 22 16:17:14 hi Aug 22 16:17:17 Pali Aug 22 16:17:20 hi Aug 22 16:17:33 We try to upgrade our N900 ram Aug 22 16:17:46 u think it will be possible by changing the onenand Aug 22 16:17:55 and the thing that lays under it? :p Aug 22 16:18:48 we were talking about this for some hours now Aug 22 16:19:28 replacing n900 ram? Aug 22 16:19:37 yeah Aug 22 16:19:55 upgrading to 1GB Aug 22 16:20:15 the only thing we want to know is where to get some parts Aug 22 16:20:28 we found out that the palm pre plus uses the same as N900 Aug 22 16:20:34 but it has 512MB ram Aug 22 16:20:52 its at least double more than the N900 Aug 22 16:21:25 so, if we can find one with 1GB, i like to sacrifice my N900 to do the job Aug 22 16:21:39 if that works out, i think many others will follow :)) Aug 22 16:22:58 well, "many" Aug 22 16:23:08 swapping out a PoP isn't exactly easy ;p Aug 22 16:24:02 only reason I would consider it is because we have SMD rework stations here Aug 22 16:26:33 If I want to compile something in scratchbox I should use FREMANTLE_ARMEL? Aug 22 16:27:32 Estel_ Aug 22 16:27:35 what about Motorola Milestone XT800 Aug 22 17:10:11 DocScrutinizer05: maemo-pan doesn't appear to have been updated since OS2008, so I guess I'll look into porting it after reading the tmo topic you also linked if that does point me to something working properly on maemo5 (icd2 plugin; like usb networking) Aug 22 17:11:25 s/does /doesn't / Aug 22 17:11:26 zeq meant: DocScrutinizer05: maemo-pan doesn't appear to have been updated since OS2008, so I guess I'll look into porting it after reading the tmo topic you also linked if that doesn't point me to something working properly on maemo5 (icd2 plugin; like usb networki... Aug 22 17:13:34 maemo-pan for os2008 was about using pan on another phone for internet access, right? Aug 22 17:13:39 zeq: good plan Aug 22 17:14:01 the thread seems up to date Aug 22 17:16:38 DocScrutinizer05: I've only had a quick look at it, it seemed to be about a script to set things up for a PAN server Aug 22 17:17:05 PAN is bidir Aug 22 17:17:11 ShadowJK: yes PAN client Aug 22 17:17:11 like ethernet Aug 22 17:17:23 aiui Aug 22 17:17:28 yes, a 'bnep0' device Aug 22 17:17:34 so a server is just offering DHCP etc Aug 22 17:17:35 as I recall Aug 22 17:18:10 if you want a client, just tear down/kick out the server related parts Aug 22 17:18:20 and add standard IP routing Aug 22 17:18:47 would be nice to have it working with icd integration though Aug 22 17:19:03 prolly feasible Aug 22 17:19:50 it seems it was working for OS2008, but used the "dummy" device as the usb net used to on m5 Aug 22 17:20:13 :nod: Aug 22 17:20:41 dummy still same I'd guess Aug 22 17:20:56 I'll take a look at the usb net code and see if I can integrate it with the OS2008 BTPAN manager Aug 22 17:21:00 whole ICD hasn't changed regarding setip of connections Aug 22 17:21:32 * DocScrutinizer05 waves and runs for shopping Aug 22 17:21:45 see you later DocScrutinizer05 Aug 22 17:22:03 * DocScrutinizer05 lives in a stupid part of the world where every shop is closed after 20:00 Aug 22 17:23:12 I hear people on the other side of the Berliner Mauer can go to 20:15! Imagine that, DocScrutinizer05! Aug 22 17:23:35 DocScrutinizer05: gas station Aug 22 17:23:55 no milk **** BEGIN LOGGING AT Wed Aug 22 17:34:23 2012 Aug 22 17:39:45 meh, samsung abandons bada Aug 22 17:40:14 and will join to tizen? :D Aug 22 17:40:16 more like also abandons tizen Aug 22 17:40:22 :( Aug 22 17:40:26 they are delaying it to 2013 Aug 22 17:40:30 http://www.sammobile.com/2012/08/22/no-more-bada-and-tizen-in-2012/ Aug 22 17:40:54 * javispedro kicks freenode Aug 22 17:48:03 lol @ aap Aug 22 17:49:52 high time for raterman Aug 22 17:49:56 raster* Aug 22 17:50:44 293 users \o/ Aug 22 17:51:01 even without chanserv ;-P Aug 22 17:59:09 I fail to understand how Samsung can't have something to ship by now..? Aug 22 18:00:52 It's almost like they've waited for rasterman to code it all before pulling the plug, if indeed that's what has happened.. Aug 22 18:05:19 what? they pulled the plug? on what? Aug 22 18:06:54 Venemo_N9: ^^^ Aug 22 18:07:53 meh, samsung abandons bada Aug 22 18:08:10 ah Aug 22 18:08:19 and will join to tizen? :D Aug 22 18:08:24 no surprise there Aug 22 18:08:37 possibly Aug 22 18:08:40 javispedro> more like also abandons tizen Aug 22 18:08:41 they didn't really pull the plug, just delayed to 2012 Aug 22 18:08:44 *2013 Aug 22 18:08:50 javispedro> they are delaying it to 2013 Aug 22 18:08:55 http://www.sammobile.com/2012/08/22/no-more-bada-and-tizen-in-2012/ Aug 22 18:09:18 /but IMHO I agree with that tabloid's opinion that this is basically just a warning of an eventual cancellation.../ Aug 22 18:09:19 oops should pay attention :) Aug 22 18:21:04 freemangordon: hi Aug 22 18:22:29 I'm currently preparing a IPCLite source tree, the wiki has been updated with build instructions :) Aug 22 18:52:23 IPCLite patchset: 11429 files updated, 0 files merged, 6352 files removed, 0 files unresolved Aug 22 18:52:31 O_o Aug 22 19:01:40 zeq: WTF? :D:D:D Aug 22 19:02:11 yeah, apparently romaxa has been busy! Aug 22 19:02:48 I've got my development device :D Aug 22 19:02:50 11k files is more than busy Aug 22 19:03:11 zeq: about time ;-) Aug 22 19:05:02 I don't have to worry so much about bricking my main device now :) Aug 22 19:05:58 great Aug 22 19:06:02 I know that feeling, from hostmode days Aug 22 19:06:53 zeq: now enjoy the wonders of BM ;-) Aug 22 19:07:34 don't forget to do a BM backup of virgin device! Aug 22 19:07:59 before you clone your primary phone to new one Aug 22 19:09:52 I'm going to keep it pretty clean. Aug 22 19:10:11 do't forget to pick a proper name echo devel900 >/etc/hostname Aug 22 19:10:13 not planning on cloning Aug 22 19:10:54 that's a very good idea! Aug 22 19:10:59 I'd do it for the shits'n'giggles Aug 22 19:13:41 freemangordon: I wonder if fennec-xul/qt would work better from this tree? Aug 22 19:14:09 don't think soo, after all it is based on m-c Aug 22 19:14:21 s/soo/so/ Aug 22 19:14:35 ~ping Aug 22 19:14:51 DocScrutinizer51: ^^^ Aug 22 19:16:37 freemangordon: there are *lots* of changes though Aug 22 19:17:17 well, try it :P Aug 22 19:21:56 I will, as soon as I fixup the patches... Aug 22 19:43:40 freemangordon: bot? Aug 22 19:43:52 yeah bot Aug 22 19:44:04 thanks Aug 22 19:44:08 yeah Aug 22 19:44:11 np Aug 22 19:47:45 povbot_: seen infobot Aug 22 19:47:46 DocScrutinizer51: infobot was last seen in #maemo 2 hours, 36 minutes, and 19 seconds ago: zeq meant: DocScrutinizer05: maemo-pan doesn't appear to have been updated since OS2008, so I guess I'll look into porting it after reading the tmo topic you also linked if that doesn't point me to something working properly on maemo5 (icd2 plugin; like usb networki... Aug 22 19:50:30 usual complete connectivity loss, nothing I can do but ping Tim (who is already online so probably aware of the issue) Aug 22 19:51:44 well, no chanserv, no infobot Aug 22 19:51:57 blame freenode Aug 22 19:53:28 actually no nickserv, no bot Aug 22 19:54:31 /join #freenode and enjoy the chaos Aug 22 19:57:02 freemangordon: it has lots of fixes/enhancements for the PowerVR EGL implementation Aug 22 20:03:55 aaah Aug 22 20:04:20 now that kinda explains the 11k files Aug 22 20:04:40 zeq1: wow, sounds great Aug 22 20:04:47 are you able to compile it? Aug 22 21:52:38 zeq: ping Aug 22 22:00:19 povbot_: seen infobot Aug 22 22:00:19 DocScrutinizer05: infobot was last seen in #maemo 4 hours, 48 minutes, and 53 seconds ago: zeq meant: DocScrutinizer05: maemo-pan doesn't appear to have been updated since OS2008, so I guess I'll look into porting it after reading the tmo topic you also linked if that doesn't point me to something working properly on maemo5 (icd2 plugin; like usb networki... Aug 22 22:00:29 povbot_: seen ChanServ Aug 22 22:00:29 DocScrutinizer05: I have not seen ChanServ. Aug 22 22:00:42 nickserv is back at least Aug 22 22:00:43 SIC! Aug 22 22:01:01 ChanServ: help Aug 22 22:01:48 at least that works Aug 22 22:02:00 heh Aug 22 22:02:23 heh what? Aug 22 22:02:53 povbot didnt see chanserv altho its here... Aug 22 22:03:00 freenode in trouble Aug 22 22:04:02 yeah there was a message earlier, like some $important box had a powerfailure and the ups was awol... Aug 22 22:04:06 infobot got busted during the process Aug 22 22:04:32 I hate that Aug 22 22:05:19 actually freenode crew should offer an account on some freenode server for infobot Aug 22 22:06:02 she's as much services as nickserv ever gets Aug 22 22:07:01 * DocScrutinizer05 heads over to other chan to check what's infobot status Aug 22 22:17:36 DocScrutinizer05: and you keeping op coz of services? Aug 22 22:18:11 ZogG_laptop: WUT? plz r-prshe Aug 22 22:18:50 doesn't matter, read the log. Aug 22 22:19:09 was taking about freenode services down Aug 22 22:20:13 too complicated Aug 22 23:26:40 DocScrutinizer05, Hostmode for N950/N9 died? Aug 22 23:27:17 I stopped cardiopulmonary support Aug 22 23:27:45 I see. Why so, was it een more complicated than in Maemo 5? :) Aug 22 23:28:36 I dispise HARM and I find no cooperate support for hostmode on that unbearable OS Aug 22 23:29:10 yes, aegis spoiled the game pretty much Aug 22 23:29:34 i just hope jolla wont use it... Aug 22 23:29:54 I'm willing to fight kernel. I'm however *NOT* willing to fight bullshoit that stps me from tweaking kernel Aug 22 23:30:21 DocScrutinizer05, absolutely understandable Aug 22 23:30:26 esp for no good reason at all Aug 22 23:30:41 and You know what? even from user perspective, it's same feeling (fighting Os and it's additions) Aug 22 23:31:03 of course support for N950 would be great, but I feel Your pain Aug 22 23:31:10 I know, I gave up when I realized it's not just me Aug 22 23:31:11 tried normal linux way of modifying some files. Once. Aug 22 23:32:31 DocScrutinizer05, BTw, is it about aegis as in Harmattan, or ARm TrustMyAssComputing as whole? Aug 22 23:32:44 i.e Mer under N950/N9 - maybe it would be more bearable for hostmode? Aug 22 23:33:28 Nemo is loosing dumb'icity and becoming better and better every day. sure, aegis kills much fun of using device with other stack (closed CAL etc), but how much it is related to hostmode too? Aug 22 23:34:29 BTW, someone somewhere is asking why "people who worked on hostmode for harm abandoned it". Won't You mind, if I quote You from IRC here, using literal quotation? Aug 22 23:34:38 I feel it's great sumamry of why ;) Aug 22 23:34:49 s/sumamry/summary/ Aug 22 23:38:47 Estel_: for all I heard, once mer would be a reality, it might host h-e-n without problems Aug 22 23:39:11 nice to hear, but isn't it reality. now? Aug 22 23:39:18 one that is jsut expanding? Aug 22 23:47:33 DocScrutinizer05, ^^^ Aug 22 23:51:56 well, i'm not following, supporting and pushing more than 2...3 distros. Sorry too few hours in a day, and nobody from mer/nemo approached me about it Aug 22 23:52:53 seems hostmode has at least as low priority to those projects as has bme-replacement Aug 22 23:53:58 and, I knowit's silly, I got a lil bitter and bored over the years Aug 22 23:56:26 so if they think they can get hostmode from plain upstream (a rather sane assumption, since sane hw on N9(50)), it's fine with me. If they fail, they have to think about incentive as I'm not interested in HARM, nor in hw-kbd-less devices at large Aug 22 23:57:51 my N9(xx) devices came to me as lead Nokians felt I should receive an sample, no matter what I do with it Aug 22 23:58:38 I feel no obligation to fulfill any contract Aug 22 23:59:31 and as things are right now, HARM is extremely unpleasant and annoying to me, whenever I deal with it Aug 23 00:00:01 mer/nemo has to talk me in yet Aug 23 00:00:30 I git no debt to bring Aug 23 00:00:44 got* Aug 23 00:03:16 FYI: blackhat 2012 usa videos: http://dc.pronto185.com/blackhat/ Aug 23 00:26:20 biggest problem with infobot is that it's poorly written perl mess Aug 23 00:28:34 and it was only written for "sane" irc networks, and with a certain netiquette in mind. Apart from just plain bad network code, freenode's funny antics often get interpreted as "I'm banned", and it doesn't try to engage in any ban evasion Aug 23 00:36:18 aah perl, yes Aug 23 00:36:49 friggin biggest problem though is that I got no direct account to retsart it when it's offline Aug 23 00:37:46 Tim anyway uttered "I wish somebody would help on improving the scheduler" Aug 23 00:38:11 Set it to restart by itself! Aug 23 00:38:19 not that *I* had any idea what the bot scheduler does, or how all that magic works Aug 23 00:38:43 robbiethe1st: I suggested that constantly since 18 months Aug 23 00:39:06 heck, I did that in my PHP bot, so... Aug 23 00:39:48 I also suggested Tim should set up a IRC script, linux account with login: "killall infobot", or even mail filter to do the same - alas to no success so far Aug 23 00:41:11 so no way I can do anything when bot is completely offline, other than mail Tim and ask for "/etc/init.d/infobot force-reload" Aug 23 00:41:58 this might take from 30 min to 4 days for any response Aug 23 00:42:52 anyway it seems due once more to send such mail Aug 23 00:44:01 sometimes I doubt Tim understands what importance and relevance "his" bot has acquired meanwhile Aug 23 00:45:38 I think there are around 5k unique users in ~150 chan that rely on the factoid database Aug 23 00:45:51 even if only eventually and occasionaly Aug 23 00:46:44 could any of you replace ~flashing with the proper URL in 10 seconds? Aug 23 00:47:11 or quote ~flatbatrecover ? Aug 23 00:48:26 I nagged freenode staff about it, but they are still reluctant to appreciate value of the bot, and act accordingly (i.e. offer "official" hosting in services domain) Aug 23 00:52:05 * ShadowJK had a script that ran in a loop in screen Aug 23 00:56:41 then you have script which runs on another box, that sshs in and checks for that screen session periodically Aug 23 00:57:17 as bonus, on first box the script checking infobox sshs to second box to check for screen session too :P Aug 23 00:58:14 As long as both machines don't reboot simultaneously, they'll poke eachother back to life Aug 23 00:58:57 (ghetto way of doing autostart when you dont have root access) Aug 23 01:00:18 @reboot in cron Aug 23 01:00:59 that's the right way of doing it Aug 23 01:12:05 sure thing, but I'd already be happy with Tim creating an account for me on his infobot box, with login shell "/etc/init.d/infobot restart" Aug 23 01:12:49 I could do sam ething any time by sending ~die, but obviously only if friggin bot is logged in Aug 23 01:13:10 under any of its 4 personalities at least Aug 23 01:14:39 and if bot refuses to connect to freenode at all (due to whatever fsckup), the only path to recover is to mail/ping/pester Tim to do "/etc/init.d/infobot restart" manually Aug 23 01:16:27 btw: Aug 23 01:16:31 [2012-08-23 03:16:08] [Notice] -NickServ- Information on apt (account apt): Aug 23 01:16:32 [2012-08-23 03:16:08] [Notice] -NickServ- Registered : Aug 16 21:01:06 1999 (13 years, 1 week, 3 days, 04:15:02 ago) Aug 23 01:16:34 [ Aug 23 01:32:12 och my god, i slept on keyboard few sec ago... time to go sleep ;D Aug 23 01:34:22 nite nite guyz Aug 23 01:37:57 good night! Aug 23 01:37:59 everybody Aug 23 01:42:29 hi,everybody,i got a problem. when i install python2.5-qt4,the terminal return the follow message: Aug 23 02:03:53 yay, down to 266, INCL chanServ Aug 23 02:04:49 (btw that terminal msg is a tad terse ;-D) Aug 23 02:06:22 whsailing: seems your connectivity got flawed and thus some of your posts missing? see http://mg.pov.lt/maemo-irclog/%23maemo.2012-08-23.log.html#t2012-08-23T04:42:29 **** ENDING LOGGING AT Thu Aug 23 02:59:59 2012