**** BEGIN LOGGING AT Mon Oct 20 02:59:57 2008 Oct 20 07:59:37 BobOfDoom: is probable that GSM use some part of RAM or can be it shared with that? Oct 20 08:00:34 I'm still curious, what is connected to Chip select2 Oct 20 08:11:43 well pit could be shared Oct 20 08:11:48 but I doubt it Oct 20 08:12:03 cause that'd make GSM certification... somewhat dubious Oct 20 08:12:10 and it's relly not necessary Oct 20 08:12:15 everything can be done via USB Oct 20 08:12:39 yes, CS2 is odd Oct 20 08:12:41 but Oct 20 08:12:43 maybe Oct 20 08:12:48 there's nothing there Oct 20 08:12:54 and it's just got mappings from 650 Oct 20 08:13:04 I noticed the 680 DAL mentions the ASIC6 Oct 20 08:13:14 could be they've just not removed the code Oct 20 08:15:49 I'm switching with GPIOs like mad but I can't get info, that GSM is found Oct 20 08:16:13 I also tried to compare PalmOS and Linux GPIO status - nothing Oct 20 08:16:47 no change with suspend Oct 20 08:17:02 and rewriting to 2.6.27 is not working at all :b Oct 20 08:28:27 hmm'\ Oct 20 08:29:01 does lsusb normally just show two things? Oct 20 08:29:06 root hub Oct 20 08:29:09 and bluetooth? Oct 20 08:29:37 nope Oct 20 08:29:45 I have to enable it through GPIO42 Oct 20 08:29:48 43 Oct 20 08:29:55 I maybe have something Oct 20 08:30:11 (consisting of changing 4 AFN and 2 levels Oct 20 08:30:12 ) Oct 20 08:30:24 gotta go to work and try on the way Oct 20 09:46:41 BobOfDoom: I found that on switching phone on this changes: Oct 20 09:46:59 86,42 switches low, then high Oct 20 09:47:21 32 switches to afn2 then afn0 Oct 20 09:47:49 29,92,95 switches afn1 then afn0 Oct 20 09:48:03 when I do it in linux Oct 20 09:48:28 I got LCD screwed (probably bad clock) Oct 20 09:48:40 and I sometimes can't use touchscreen anymore Oct 20 09:48:56 and it probably also affects SD/MMC Oct 20 09:51:38 :/ Oct 20 09:52:16 well Oct 20 09:52:24 setting 32 to AF 0 is gonna break MMC Oct 20 09:52:40 since it's the MMC clk pin Oct 20 09:53:10 probably that's just some palmos power saving feature, disables MMC clock when not in use, probably nothing to do with phone Oct 20 09:53:42 29 is either AC97 or SSP depending on if its in or out Oct 20 09:53:44 it is weird since GPIOtrace writes all that time Oct 20 09:54:06 92 is MMC Oct 20 09:54:18 I didn't do it atomically - only using GPIOed-ng Oct 20 09:54:33 95 is either keypad or MMC Oct 20 09:55:09 95 is not used for keypad Oct 20 09:55:23 ok, so it's probably MMC then Oct 20 09:55:28 ok Oct 20 09:55:51 that's is a bit weird Oct 20 09:55:58 but not unreasonable I guess Oct 20 10:01:44 86 is that one affecting LCD Oct 20 10:04:15 does it fade to white? Oct 20 10:04:21 or just go beserk? Oct 20 10:04:34 white Oct 20 10:05:12 when you take it high? Oct 20 10:05:16 hmm Oct 20 10:05:24 I dunno why turning on the phone would do that hten Oct 20 10:05:25 putting low Oct 20 10:05:38 oh Oct 20 10:05:53 afer putting back high again it looks blured and with lines Oct 20 10:05:54 perhaps GPIOtrace recorded waking and utnring on the LCD then? Oct 20 10:06:05 nope Oct 20 10:06:07 hmm Oct 20 10:06:09 weirdness Oct 20 10:06:25 I'll try another dump Oct 20 10:09:47 btw. there _are_ some changes on CS2 Oct 20 10:11:29 owah Oct 20 10:11:33 centros are cheap Oct 20 10:12:05 $200 from the local electronics store Oct 20 10:12:23 that is locked to a prepaid telstra plan though Oct 20 10:13:18 oh, completely different GPIO changes Oct 20 10:14:05 amfg Oct 20 10:14:10 completely Oct 20 10:18:54 ok, lets try 3rd and 4th dump Oct 20 10:30:12 wow Oct 20 10:30:33 now I have two SAME dumps Oct 20 10:30:38 I'll try that Oct 20 10:30:54 XD Oct 20 10:30:58 87,40,39,34 --> 1 Oct 20 10:31:04 34 - afn1 Oct 20 10:31:11 41 - afn2 Oct 20 10:31:13 btw, my 680 was mailed today Oct 20 10:31:19 should get it in a day or two Oct 20 10:31:29 direction 35,34,14 -> 0 Oct 20 10:31:34 BobOfDoom: cool! :D Oct 20 10:31:51 BobOfDoom: it will finally have some progress :b Oct 20 10:32:00 nah Oct 20 10:32:04 you've been doing a great job Oct 20 10:32:14 I won't have much time to work on it for a couple of weeks Oct 20 10:32:25 really in trouble assignment wise at the moment Oct 20 10:32:27 neither will I :b Oct 20 10:32:36 hmmm Oct 20 10:41:34 could you assume right order of that changeS? Oct 20 10:42:58 probably only important GPIO is 34... Oct 20 10:58:01 ok, I've got GPIOs set same way, but still nothing :/ Oct 20 11:00:07 :( Oct 20 11:00:19 have you set UDC port 3 stuff? Oct 20 11:00:30 assuming it's connected via port 3 Oct 20 11:01:23 I haven't found something like UP3OCR Oct 20 11:01:36 lsusb reports it's powered Oct 20 11:02:37 what do you mean you haven't found it? Oct 20 11:02:49 0x4060_0024 Oct 20 11:02:51 = UP3OCR Oct 20 11:03:00 is it not the headers? Oct 20 11:04:51 not in mine :b Oct 20 11:04:53 ok Oct 20 11:06:38 % grep UP.OCR . -R Oct 20 11:06:38 ./mach/pxa27x-udc.h:#define UP2OCR __REG(0x40600020) /* USB Port 2 Output Control register */ Oct 20 11:06:40 ./mach/pxa27x-udc.h:#define UP3OCR __REG(0x40600024) /* USB Port 2 Output Control register */ Oct 20 11:06:50 that's 2.6.27 though Oct 20 11:06:58 and the omment is wrong :p Oct 20 11:07:00 that is probably that difference Oct 20 11:07:07 ok Oct 20 11:07:37 I'd like to move further in 2.6.27 code Oct 20 11:07:50 but I'm stuck with that white screen after boot Oct 20 11:08:07 Maybe it is caused by that GPIO initialization on the beginning Oct 20 11:08:18 I don't know if I would write GPIOs there Oct 20 11:08:28 yes, could be Oct 20 11:08:31 if it will set them low/high Oct 20 11:08:48 I got to white screen Oct 20 11:08:54 yes, it messes with them Oct 20 11:09:01 if you mean the GPIO init defined in the machine file Oct 20 11:09:04 the MFP crap Oct 20 11:09:04 (first it was just screen off) Oct 20 11:09:09 yes Oct 20 11:09:15 I'd comment them out to start with Oct 20 11:12:15 does it restrict that GPIOs won't be used? Oct 20 11:12:39 no Oct 20 11:12:45 fine Oct 20 11:12:51 it doesn't touch any GPIO unless you name it there Oct 20 11:13:24 and yeah, you can call gpio_register or whatever it is with GPIOs that haven't been setup like that Oct 20 11:13:33 it's just init stuff Oct 20 11:13:48 it is good idea, but it does something bad Oct 20 11:14:04 I'd rather comment out many of them... Oct 20 11:14:24 yeah, go for it Oct 20 11:14:43 once you've figured out what most of the GPIOs do you can add them back Oct 20 11:14:52 but until then, take it out, as it's probably wrong Oct 20 11:15:04 :) Oct 20 11:15:14 I only added GPIOs I know Oct 20 11:15:24 but maybe there is more :( Oct 20 11:15:34 also, it changes the levels Oct 20 11:16:15 yup Oct 20 11:16:55 I think it makes them all inputs by default too Oct 20 11:16:59 which is not what you want all the time Oct 20 11:17:00 I was trying to read MFP code and I'm not sure, but MFP_PULL_HIGH is the setting I need Oct 20 11:17:09 ugh Oct 20 11:17:18 bad bad MFP Oct 20 11:18:08 * Sleep_Walker is booting 2.6.21 with UP3OCR Oct 20 11:26:57 with UP3OCR - no change I can see Oct 20 11:32:24 * Sleep_Walker is curious what does GPIO79 do.. that is the GPIO which can sometimes turn red led on Oct 20 12:07:16 * BobOfDoom wonders if it's labeled in the design notes as GPIO_DO_RANDOM_STUFF_TO_CONFUSE_HACKNDEV_GUYS Oct 20 13:08:36 :)) Oct 20 18:00:14 BobOfDoom: progress Oct 20 18:00:48 I've got serie of gpio changes which causes red led on and I cannot switch it off :b Oct 20 18:08:12 I've got strong feelings that it still can be on FFUART :b Oct 20 18:08:26 and USB is for data Oct 20 18:08:50 jeezes christ ... Oct 20 18:09:09 they lost their sanity... Oct 20 18:09:41 I have to confirm it Oct 20 18:10:06 but you can't get 200kb/s through UART, can you? Oct 20 18:11:15 921600Bd Oct 20 18:11:48 and modulation? Oct 20 18:12:20 1 638 400 bits/s Oct 20 18:13:55 about 500kbps is possible at normal circumstances\ Oct 20 18:17:37 kbps = kilobit per second? Oct 20 18:17:46 or kilobyte per second? Oct 20 18:18:37 minicom can't operate 480600 :( Oct 20 18:19:31 sure it can Oct 20 18:19:57 it can up to 921kBd Oct 20 18:21:33 marex: not mine Oct 20 18:26:02 I've got output! Oct 20 18:26:06 ERROR Oct 20 18:26:07 :) Oct 20 18:27:49 lucky thinking about USB - we wouldn't find that :b Oct 20 18:30:25 nice Oct 20 18:30:46 that might be command interface, that might probably run at lower speed Oct 20 18:31:48 marex: well, I have my own BASH terminal :D Oct 20 18:31:59 stty -F /dev/ttyS0 460800 Oct 20 18:32:04 tail -f /dev/ttyS0 & Oct 20 18:32:13 echo -e "ATZ\r" > /dev/ttyS0 Oct 20 18:32:15 :) Oct 20 19:01:45 and I can here music from external amplifier - GPIO27 is the right one Oct 20 19:07:39 and finally even loud! Oct 20 19:09:19 np: Kaltenecker Trio - Ivory Tower - 07 - Graffiti! :D Oct 20 19:11:12 BobOfDoom: what about changing silent switch into USB mode switch? :D Oct 20 19:14:31 np: Gun's & Roses - Baccano! OP Oct 20 19:56:27 marex: http://hackndev.com/node/209#comment-821 Oct 20 22:11:06 BobOfDoom: if BCM2133 is in some weird binary mode - how can I leave that? Oct 20 22:22:43 Sleep_Walker, haha ... have fun :) Oct 20 22:23:03 arent those just crc/cmdlen etc. ? Oct 20 22:23:27 ie. ericsson modems has some informations before command Oct 20 23:35:30 Sleep_Walker: that's MUX mode Oct 20 23:35:35 easiest way is to restart it Oct 20 23:35:45 but there's some command to exit it as well Oct 20 23:35:52 I don't know what it is off the top of my head though Oct 20 23:36:12 (easiest way is to reset the chip I mean, toggle power to it) Oct 20 23:53:54 finally back! Oct 20 23:53:57 BobOfDoom: alive? Oct 21 00:06:55 ooh Oct 21 00:06:58 you weren't here Oct 21 00:06:59 herhe Oct 21 00:07:07 10:35 < BobOfDoom> Sleep_Walker: that's MUX mode Oct 21 00:07:07 10:35 < BobOfDoom> easiest way is to restart it Oct 21 00:07:07 10:35 < BobOfDoom> but there's some command to exit it as well Oct 21 00:07:07 10:35 < BobOfDoom> I don't know what it is off the top of my head though Oct 21 00:07:07 10:36 < BobOfDoom> (easiest way is to reset the chip I mean, toggle power to it) Oct 21 00:07:35 BobOfDoom: I wasn't... I was on the way home Oct 21 00:07:58 yeah, I didn't notice you'd left the channel XD Oct 21 00:07:58 (and I was trying again 2.6.27 - disabled every code and still the same) Oct 21 00:08:03 :( Oct 21 00:08:22 that binary mode is called "GSM 07.10 multiplexing" Oct 21 00:08:35 restart it? Oct 21 00:08:38 there's some code to talk to it here: http://git.freesmartphone.org/?p=gsm0710muxd.git;a=summary Oct 21 00:08:48 Sleep_Walker: toggle power Oct 21 00:08:49 I'm not sure where to switch it off :b Oct 21 00:08:53 ah Oct 21 00:09:09 well that's hard then :p Oct 21 00:09:37 that gsm0710muxd app has some code that's supposed to take it out of MUX mode Oct 21 00:10:08 I'd be completely happy with Oct 21 00:10:09 AT Oct 21 00:10:11 OK Oct 21 00:10:12 ;b Oct 21 00:10:23 it's for GPRS Oct 21 00:10:36 so you can have your commnads in one channel and your GPRS in another Oct 21 00:10:37 IIRC Oct 21 00:11:35 ok, I gotta head home now Oct 21 00:11:49 I've gotta write a distributed filesystem from scratch by tomorrow :( Oct 21 00:11:54 I should have started this assignment earlier :p Oct 21 00:12:04 cheat and use OCFS2 ;D Oct 21 00:12:24 I think they'd probably notice :p Oct 21 00:12:43 damn Oct 21 00:12:52 they're clever, aren't they? Oct 21 00:12:59 actually funny thing about uni, is you will get in heaps of trouble if you steal something and claim you wrote it Oct 21 00:13:29 but if you submit someone else's code and properly attribute and cite it, you'll probably actually get partial marks Oct 21 00:14:04 Sleep_Walker: well clever enough to notice OCFS2 :p Oct 21 00:14:19 hehe Oct 21 00:14:50 I also want to see if my 680 has arrived Oct 21 00:14:58 probably not, but the guy said he sent it via express Oct 21 00:15:47 then I'll be able to make phonecalls again Oct 21 00:32:25 AT Oct 21 00:32:26 OK Oct 21 00:32:28 got it :) Oct 21 00:32:38 minicom succesfully reseted that Oct 21 00:55:25 gotta go sleep as winner :D **** ENDING LOGGING AT Tue Oct 21 02:59:57 2008