| 00:00.20 | cr2 | j0b0: linux-arm had it since ages |
| 00:01.59 | NetRipper | cr2, of course |
| 00:03.25 | cr2 | look for kaiser_fixup |
| 00:03.37 | cr2 | http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=commitdiff;h=4805c6d61bfcc18207b2a3a90861254e05a2c241 |
| 00:04.08 | cr2 | you only nned to change the .size (s) |
| 00:05.04 | j0b0 | cr2 but even with the 1 bank, using more than 96M or so can get your raph100 in a state where it wont run unsigned apps anymore |
| 00:05.12 | j0b0 | and only hard reset cures it |
| 00:05.23 | j0b0 | (which is not _that_ big a deal.. but still) |
| 00:05.29 | cr2 | ??? |
| 00:06.54 | j0b0 | it didnt happen to me personally but it was reported at least 4 or 5 times |
| 00:07.21 | NetRipper | cr2, ah, nice |
| 00:07.39 | NetRipper | j0b0, im not sure if those are related |
| 00:07.42 | cr2 | i just don't get it. there is no DRM crap in our kernel ? |
| 00:08.39 | j0b0 | neither am i but the sispicion is out there :) |
| 00:08.47 | j0b0 | *susp |
| 00:08.55 | cr2 | if its some android problem, replace it with something better ;) |
| 00:09.01 | *** join/#htc-linux konsta (n=asds@host81-154-247-76.range81-154.btcentralplus.com) |
| 00:09.06 | cr2 | why is it not in wince ? |
| 00:10.50 | NetRipper | j0b0, i've been using 0x2* memory bank without issues.. i myself had problems using 0x1* bank, but nobody else seems to have issues with it... (nor do i anymore) |
| 00:10.58 | cr2 | using binary modules is retarded anyway |
| 00:11.23 | NetRipper | i do think we shouldn't use more memory than 92M on the first bank though |
| 00:11.52 | j0b0 | cool. there was one recent report of the unsigned thing |
| 00:11.53 | NetRipper | some devices don't boot with 96M, but do boot with 92M |
| 00:11.53 | cr2 | NetRipper: why ? |
| 00:17.02 | NetRipper | cr2, ^^ is why |
| 00:17.46 | NetRipper | cr2, another question, we know the soft-reset interrupt.. can we trigger it manually? |
| 00:22.40 | *** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix) |
| 00:22.55 | *** join/#htc-linux ionstorm (n=ion@ip68-227-226-5.ph.ph.cox.net) |
| 00:23.41 | j0b0 | NetRipper mine doesnt boot at all off 0x20000000 .. just tried it with 92M and even with 76M |
| 00:24.02 | NetRipper | j0b0, also raph100? |
| 00:24.25 | NetRipper | i think the latest kernel didn't boot for me from 0x2* either.. somewhere along the way it may have stopped |
| 00:24.47 | j0b0 | yep .. just changed back to 0x10000000 and it boots |
| 00:28.04 | *** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net) |
| 00:32.23 | tmzt | cr2: the high-speed uart driver is in 2.6.27 on git.android I think if it's been released |
| 00:32.43 | tmzt | I don't think the user component has been released for the new driver, the old one was from upstream as far as I know |
| 00:33.06 | tmzt | have you found registers on the mddi chips? |
| 00:37.45 | tmzt | GPIOx, or are those on the msm? |
| 00:41.41 | cr2 | NetRipper: i think it's a proc_comm code |
| 00:44.58 | cr2 | good night |
| 00:57.27 | NetRipper | tmzt, what do you mean GPIOx? |
| 00:58.52 | NetRipper | tmzt, msm7x00x has two gpio controllers |
| 00:58.56 | NetRipper | if that's what you mean |
| 01:02.52 | tmzt | it does? |
| 01:05.08 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
| 01:10.08 | tmzt | <PROTECTED> |
| 01:10.19 | tmzt | I'm having trouble finding it and just read it |
| 01:16.15 | *** join/#htc-linux Zinbolic (n=chatzill@84.238.80.215) |
| 01:18.39 | NetRipper | tmzt, what are you trying to find regarding mddi? |
| 01:18.44 | NetRipper | the trace regs? |
| 01:20.17 | *** join/#htc-linux exco (n=exco@e181126244.adsl.alicedsl.de) |
| 01:22.08 | NetRipper | tmzt, these logs are realtime and searchable btw... http://irclog.iclem.net/?chan=htc-linux |
| 01:22.16 | NetRipper | searchable as in, ctrl-f |
| 01:22.16 | NetRipper | ;) |
| 01:24.54 | tmzt | dream_kill: are you trying to write the oemsbl or the whole radio.img from g1? |
| 01:25.39 | NetRipper | he pasted a link with, appearantly, the g1 radio modified to be flashed on security unlocked raphael devices |
| 01:25.53 | NetRipper | nobody dared trying it yet it hink |
| 01:25.53 | NetRipper | ;) |
| 01:26.49 | *** join/#htc-linux zycho_ (n=zycho@a89-182-207-108.net-htp.de) |
| 01:27.30 | tmzt | there is no SDCMUX on g1, only on halibut |
| 01:33.50 | *** join/#htc-linux hollo (n=hollo@3e6b025d.rev.stofanet.dk) |
| 01:57.32 | *** join/#htc-linux ALoGeNo (n=alogeno@unaffiliated/alogeno) |
| 01:59.07 | ALoGeNo | lo |
| 02:01.47 | tmzt | hey, ALoGeNo |
| 02:02.22 | ALoGeNo | hey tmzt !! |
| 02:02.39 | ALoGeNo | how are you? |
| 02:11.31 | tmzt | j0b0: what has registers GPIODATA GPIODIR GPIOSEL, the chip connected to mddi? |
| 02:20.02 | *** part/#htc-linux exco (n=exco@e181126244.adsl.alicedsl.de) |
| 02:21.44 | tmzt | 19:09 < cr2> raphael_gpio for that |
| 02:21.44 | tmzt | 19:10 < cr2> do we have the code to query the toshiba gpio pins over mddi ? |
| 02:21.44 | tmzt | 19:10 < maejrep[w]> no idea what that means ;o |
| 02:21.44 | tmzt | 19:11 < cr2> ok :) |
| 02:36.26 | *** join/#htc-linux Balsat (n=kll@87.72.13.87) |
| 02:40.23 | *** part/#htc-linux Balsat (n=kll@87.72.13.87) |
| 03:50.16 | maejrep | <cr2> but blac100 has only 2 lcds, while raph800 -> 6 <-- 6 LCDs?? wtf |
| 03:50.48 | tmzt | that's how many different lcd's there can be |
| 03:52.52 | maejrep | right :| |
| 03:52.55 | maejrep | that sucks |
| 04:14.20 | *** join/#htc-linux swc|666 (n=carramro@unaffiliated/swc666/x-4934821) |
| 04:19.39 | *** join/#htc-linux Summerspittman (n=secondsu@c-24-99-80-238.hsd1.ga.comcast.net) |
| 04:20.42 | maejrep | cr2: your formula at http://wiki.xda-developers.com/index.php?pagename=MSM_SDIO is not working with my dumped values :x |
| 04:58.50 | tmzt | maejrep: how is the new driver coming? |
| 05:01.14 | maejrep | haven't started yet |
| 05:02.59 | tmzt | well ok, I missed the discussion with cr2 |
| 05:19.29 | maejrep | hmm, I can get the formula to work with my dumped values using: |
| 05:19.32 | maejrep | deriving .. 0001fffd fffe007c |
| 05:19.32 | maejrep | fn( 1, 2, 1, 3, 3, 4, 0) |
| 05:19.32 | maejrep | a0 = 0001fffd |
| 05:19.32 | maejrep | a4 = fffe007c |
| 05:19.58 | maejrep | but that doesn't match any of the params listed there :x |
| 05:20.28 | maejrep | (and that's for me b0/b4 values) |
| 05:21.12 | maejrep | my a0 value is not possible given this formula though |
| 05:24.32 | maejrep | (i say that because it's 0001fff4, so ~0xfff4 == 0xb, which is odd, so shifted right, you'd lose a bit) |
| 05:26.09 | tmzt | what's the formula? |
| 05:26.31 | tmzt | cr2 found this in the assembly code? |
| 05:29.00 | tmzt | 03:18 < maejrep> i tried running the same pcom commands that I saw in the trace, and it froze my phone :p |
| 05:29.03 | maejrep | http://wiki.xda-developers.com/index.php?pagename=MSM_SDIO |
| 05:29.03 | tmzt | 03:19 < maejrep> was trying to turn off and on the lcd |
| 05:29.23 | maejrep | yeah he did |
| 05:29.26 | tmzt | is that because dma was not disabled somewhere? |
| 05:29.38 | maejrep | eh, I have no idea ;x |
| 05:30.21 | maejrep | could be that my a0 value couldn't be derived because wifi was off at the time.. guess that's a possibility, but I don't recall it being different with wifi on |
| 05:30.48 | maejrep | one way to find out :] |
| 05:32.11 | tmzt | so we know what it is but not why? |
| 05:32.27 | maejrep | what what is? clock registers? |
| 05:32.36 | tmzt | yeah |
| 05:32.42 | tmzt | the formula |
| 05:33.13 | maejrep | well, I have what haret shows it being set at currently. and cr2 has the formula from disassembling either the driver or spl |
| 05:33.58 | maejrep | but if 'x' = slot, that would mean that my sdc3 (b0/b4) values would start with "0003", no? |
| 05:34.23 | tmzt | what do yours start with? |
| 05:35.16 | maejrep | hmm, no, a0/a4 don't change when enabling wifi |
| 05:35.26 | maejrep | a86000a0 | 0001fff4 fff50b44 |
| 05:35.31 | maejrep | a86000b0 | 0001fffd fffe007c |
| 05:35.37 | tmzt | sdc1? |
| 05:35.40 | maejrep | I can derive b0 and b4, with some tweaks |
| 05:35.44 | maejrep | yeah, wifi is sdc1 |
| 05:38.26 | maejrep | x,1,0x10,0x08;1,3,1,1 <-- know what the "0x08;1" means ? :x |
| 05:38.48 | maejrep | or mask |
| 05:39.27 | maejrep | arg5=arg5&3 |
| 05:39.27 | maejrep | arg6=arg6&1 |
| 05:39.50 | maejrep | but the formula shows arg5 and arg6 being 3 bits, which should be &7 for each |
| 05:41.57 | maejrep | http://www.privatepaste.com/f40hbM1fgV <-- this works, but its slightly different from the formula on the wiki |
| 05:44.26 | tmzt | sorry, maejrep what haret command do you use to dump a0? |
| 05:45.59 | maejrep | # pdump 0xa86000a0 0x20 |
| 05:46.05 | maejrep | should give you registers for sdc1-4 |
| 05:46.33 | tmzt | why is it shifted >>16 |
| 05:46.43 | maejrep | which one? |
| 05:46.43 | tmzt | 01:27 < maejrep> a86000a0 | 0001fff4 fff50b44 |
| 05:46.43 | tmzt | 01:27 < maejrep> a86000b0 | 0001fffd fffe007c |
| 05:47.02 | tmzt | I mean here, it looks like you dumped a0-a8? |
| 05:47.18 | maejrep | first one is a0 and a4 |
| 05:47.21 | maejrep | second one is b0 and b4 |
| 05:47.24 | tmzt | ok |
| 05:47.31 | maejrep | didn't copy a8,ac or b8,bc |
| 05:47.39 | maejrep | because they're both disabled (same value) |
| 05:47.54 | tmzt | so the CKENs are in here too? |
| 05:48.07 | maejrep | no, that's at a8600000 |
| 05:48.53 | tmzt | but it puts a placeholder value in when disabled? |
| 05:48.59 | maejrep | i guess so |
| 05:49.08 | maejrep | a86000a0 | 0001fff4 fff50b44 0019f3ff f418005c | ....D.......\... |
| 05:49.08 | maejrep | a86000b0 | 0001fffd fffe007c 0019f3ff f418005c | ....|.......\... |
| 05:50.47 | tmzt | a0>>16 |
| 05:51.27 | tmzt | arg1=a0>>16 |
| 05:51.45 | tmzt | what does this mean, arg1 is the first argument to the function and it's being assigned to? |
| 05:51.48 | maejrep | the 1st 2 bytes of a0, basically |
| 05:52.05 | maejrep | no, he's talking about reversing the a0/a4 values into the arguments |
| 05:52.25 | tmzt | a0 is 4 bytes, why am I confusing these things ? |
| 05:52.25 | tmzt | ? |
| 05:52.43 | tmzt | 32bits |
| 05:52.47 | maejrep | yeah |
| 05:52.57 | maejrep | so arg1 is the "0001" part of a0 |
| 05:53.21 | maejrep | then "fff4" is supposed to be: ~(arg3<<1) |
| 05:53.27 | maejrep | which is impossible |
| 05:53.48 | tmzt | why? |
| 05:53.55 | maejrep | because x<<1 can never be odd |
| 05:54.08 | tmzt | oh, that's not |
| 05:54.37 | tmzt | what do you mean odd, where do you get that from? |
| 05:55.00 | maejrep | so: fff4 = ~arg3<<1; ~fff4 = arg3<<1; 0x000b = arg3<<1; 0x000b >> 1 = arg3 |
| 05:55.12 | *** join/#htc-linux techie (n=blarg@ip24-250-216-85.ga.at.cox.net) |
| 05:55.19 | maejrep | but 0x000b (11) >>1 is 5 |
| 05:56.05 | maejrep | and going in the reverse: 5<<1 is 0xa, not 0xb |
| 05:56.18 | maejrep | because you lost a bit when shifting right |
| 05:57.21 | maejrep | in order for (~x) to be odd, x has to be even |
| 05:57.54 | tmzt | ok, but why do you say it can't be odd |
| 05:59.38 | maejrep | i'm saying in order for his formula to work, arg3 << 1 has to be even (because there's no way for any value of x to exist such that x << 1 is odd) |
| 06:00.17 | maejrep | and because of the ~ in the formula, and because arg3<<1 has to be even, that means the resulting a0 must be odd |
| 06:00.39 | maejrep | but mine is not |
| 06:00.50 | tmzt | your arg3<<1? |
| 06:01.27 | maejrep | 1 reg arg1 (16bit) + ~(arg3 << 1) (16bit) <-- you agree that "arg3<<1" will always be even, right? |
| 06:02.45 | maejrep | and in assuming that, you must also accept that ~(any even number) must be odd (because bit-0 in an even number is off, so doing a NOT operation on that number will always turn bit-0 on, which will always make the number odd) |
| 06:03.49 | tmzt | 0010 2 even |
| 06:03.53 | tmzt | no |
| 06:04.14 | maejrep | what? |
| 06:04.18 | tmzt | sorry |
| 06:04.43 | tmzt | 0010 = 2 even |
| 06:04.55 | tmzt | ~0010 |
| 06:05.00 | tmzt | 1101 odd |
| 06:05.14 | maejrep | right |
| 06:05.30 | tmzt | yeah, what I'm not getting here is what arg3<<1 is, what are you saying about arg3 |
| 06:05.44 | tmzt | or, more generally, which way are we going here? |
| 06:05.51 | tmzt | from the function to a0/a4? |
| 06:06.03 | tmzt | or from a0/a4 to the args |
| 06:06.10 | maejrep | I'm trying to derive arg3 from a0 |
| 06:06.17 | maejrep | and it can't be done |
| 06:06.26 | maejrep | at least, not reversibly |
| 06:06.52 | maejrep | that is .. the arg3 I get from reversing his formula, will NOT give me the same a0 value by going back through the formula |
| 06:06.56 | maejrep | and never can |
| 06:07.31 | maejrep | I would get 0001fff5 instead of 0001fff4 |
| 06:07.39 | *** join/#htc-linux soljin (i=45ffcf5d@gateway/web/ajax/mibbit.com/x-ea64c6aac553aace) |
| 06:08.16 | maejrep | and that's because ~fff4 == 0xb which is odd, so shifting that right chops off the "odd" bit |
| 06:08.40 | tmzt | he looked at the spl? |
| 06:08.50 | tmzt | could the driver be different in this respect? |
| 06:08.59 | maejrep | i would guess so :x |
| 06:09.15 | tmzt | and of course, we don't know exactly what the hardware does with these values so we can't know if that bit matters can we? |
| 06:09.39 | maejrep | maybe not |
| 06:09.55 | tmzt | have you been able to trace writes to this? |
| 06:09.57 | maejrep | I wouldn't know how to find that out |
| 06:10.03 | maejrep | i've never seen the value change |
| 06:10.20 | tmzt | it doesn't change when you disable wifi or remove an sd card? |
| 06:10.23 | tmzt | neither value? |
| 06:10.24 | maejrep | a0 - bc are always the same |
| 06:10.26 | maejrep | nope |
| 06:10.40 | tmzt | do you have an sdio wifi card (external)? |
| 06:10.46 | maejrep | i could try writing to a0 and a4 and see what happens if I enable wifi ;x |
| 06:10.56 | maejrep | no I don't |
| 06:11.03 | maejrep | but would we be able to use it? |
| 06:11.07 | tmzt | I don't know if you want to do that, your dealing with clock values |
| 06:11.18 | tmzt | it might use a different negotiated clock |
| 06:11.41 | tmzt | I would think a class4/6 would be different too, but I don't know |
| 06:11.55 | maejrep | well, the linux driver modifies it so wince clearly resets it :P |
| 06:12.06 | tmzt | or spl? |
| 06:12.15 | maejrep | yeah could be |
| 06:16.52 | maejrep | i don't know how he found that in spl .. i can't figure out how to disassemble the spl :p |
| 06:22.25 | *** join/#htc-linux the_sys0p (n=the_sys0@cpe-67-49-210-229.bak.res.rr.com) |
| 06:45.26 | *** join/#htc-linux Moku (n=John@f054220243.adsl.alicedsl.de) |
| 07:03.42 | *** join/#htc-linux timebomb (n=tb@e176098083.adsl.alicedsl.de) |
| 07:53.29 | *** join/#htc-linux norby_ (n=chatzill@AMontsouris-152-1-67-155.w83-202.abo.wanadoo.fr) |
| 08:09.10 | *** join/#htc-linux xianthax (n=xianthax@c-75-69-145-210.hsd1.ma.comcast.net) |
| 08:12.43 | *** join/#htc-linux _workkaze (n=kaze@ABordeaux-152-1-51-238.w82-125.abo.wanadoo.fr) |
| 08:37.22 | *** join/#htc-linux Matze88 (n=larischm@e179199020.adsl.alicedsl.de) |
| 08:39.53 | Matze88 | Hi there, I'm new to htc linux developing and want to try to make some progress on diamond with android 1.0. Is there some place where patches with current progress are available or is nearly everything in current kernel & android git repository? |
| 08:40.12 | tmzt | hello |
| 08:40.22 | tmzt | most diamond stuff is in the thread on xda-dev |
| 08:40.54 | Matze88 | yup i know the thread, did not read it completely, maybe i should start doing so... |
| 08:40.59 | tmzt | most everything should be committed to git.linuxtogo.org kernel.git htc-msm-2.6.25 |
| 08:41.08 | tmzt | read back a few pages at least |
| 08:41.20 | tmzt | patches for what? |
| 08:42.35 | Matze88 | the kernel stuff seems to be clear to me, I'm currently test-compiling it the first time.... Will try it out soon. but the android stuff is a bit confusing cause i read that only some people managed to get it running on diamond but i did not see any patches or changes need to be done to run it |
| 08:42.56 | Matze88 | there are only the pre compiled files, am i right? |
| 08:43.04 | *** join/#htc-linux kaze_ (n=kaze@ABordeaux-158-1-27-51.w90-55.abo.wanadoo.fr) |
| 08:43.45 | tmzt | the dzo stuff is using the emulator build I believe |
| 08:44.00 | tmzt | not sure if there a better image for diam/raph |
| 08:45.59 | Matze88 | well okay i'll try that... thank you |
| 08:47.34 | tmzt | well, if you are going to try a dzo build only use the userspace part, I think you are clear on that but I want to make sure |
| 08:47.52 | tmzt | as in, use a diam zImage and mtype |
| 08:48.01 | tmzt | also, which diamond do you have? |
| 08:48.35 | Matze88 | i have diam300 |
| 08:48.43 | tmzt | uk? |
| 08:48.52 | Matze88 | yep i was sure about that, but thx anyway :-) |
| 08:49.02 | Matze88 | no it is a t-mobile mda compact IV, germany |
| 08:49.03 | tmzt | oh de |
| 08:49.13 | tmzt | it's similar to diam100 though |
| 08:49.42 | maejrep | hmm.. |
| 08:49.59 | Matze88 | maybe i'm wrong, mom |
| 08:50.09 | maejrep | root@htcraphael:/media/mmcblk0p1# ./i2cset -y 0 0x66 0x40 0x0400 w <-- turns lcd off and makes the center/ring LEDs do the "breathe" sequence |
| 08:50.27 | tmzt | hey |
| 08:50.32 | maejrep | root@htcraphael:/media/mmcblk0p1# ./i2cset -y 0 0x66 0x40 0x2000 w <-- resets it, turning the lcd back on |
| 08:51.09 | tmzt | is that what was crashing earlier? |
| 08:51.14 | tmzt | lcd off/on |
| 08:51.32 | maejrep | no, was playing with proc comm to turn it off |
| 08:51.39 | maejrep | this is through KLT i2c |
| 08:51.50 | maejrep | root@htcraphael:/media/mmcblk0p1# ./i2cset -y 0 0x66 0x40 0x0201 w <-- turns on "Home" navi backlight led :) |
| 08:52.03 | tmzt | so this disables only the backlight? can you put a color image on it and check? |
| 08:52.20 | maejrep | sweet: 0x2001 is LCD on & Home backlight |
| 08:52.37 | maejrep | pretty sure its just backlight, yeah |
| 08:53.30 | maejrep | hmm |
| 08:54.07 | maejrep | 0x3000 turns LCD on and makes the center ring do a weird shuffle thing |
| 08:54.40 | maejrep | cool, gonna have to document this ;o |
| 08:55.04 | tmzt | now if you could get leds to follow your finger around the ring |
| 08:55.31 | maejrep | lol |
| 09:00.17 | *** join/#htc-linux cr2 (n=cr2@ip-90-187-221-92.web.vodafone.de) |
| 09:01.06 | *** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 09:03.37 | cr2 | tmzt: should not be difficult to do |
| 09:04.15 | tmzt | yeah |
| 09:04.42 | cr2 | maejrep: i know that the dumped values do not always match the formula, but don't understand it |
| 09:04.45 | tmzt | I read the conversation you had with maejrep and am not clear on how you think the driver should be structured |
| 09:05.17 | tmzt | you found those formulas in the spl? |
| 09:05.40 | tmzt | also, which diamond do you have? maejrep |
| 09:05.44 | cr2 | tmzt: we have an mfd that uses 2 i2c ids, and controls bkl(s), leds, etc. |
| 09:06.10 | cr2 | tmzt: yes, and also in the sdhc |
| 09:06.49 | cr2 | tmzt: i guess these are some pll clock dividers, so different values just mean different SD clock frequency |
| 09:07.50 | tmzt | sorry, that was the wheelmouse |
| 09:08.10 | tmzt | <PROTECTED> |
| 09:08.10 | cr2 | maejrep: you started to use the real i2c ids ? good :) |
| 09:08.24 | tmzt | it keeps pasting maejrep |
| 09:10.50 | maejrep | cr2: with i2cset you have to :p |
| 09:10.57 | dzo | hi all |
| 09:11.08 | maejrep | i'm documenting what the LED bits do on cc,40 |
| 09:11.30 | tmzt | real i2c ids? |
| 09:12.11 | cr2 | hi dzo |
| 09:12.25 | Matze88 | sorry that i have to ask a dumb question: what "environment" do you use on your phone for trying around? a complete android build or just a small busybox or something like that? I'm new to this mobile device thing, it's a bit confusing although i have experience on avr32 linux boards, linksys linux routers and so on.... |
| 09:12.35 | dzo | I found the kaiser problem, needs discontig mem to boot. |
| 09:12.41 | cr2 | dzo: you have fixed the sd clock on kaiser ? |
| 09:13.00 | tmzt | Matze88: well, I use debian but dcordes has some angstrom-console images that should work well |
| 09:13.05 | cr2 | dzo: and you've solved this problem too afaik :) |
| 09:13.37 | Matze88 | tmzt: you have debian on your phone? |
| 09:13.51 | tmzt | kaiser problem? problem with what version? |
| 09:13.55 | tmzt | Matze88: yes, on sd card |
| 09:14.08 | dzo | no, you need to boot from an initrd on the card or sd is not powered up. |
| 09:14.27 | cr2 | tmzt: it was accessing the 0x214 reg in the clock area |
| 09:14.39 | Matze88 | tmzt: oh nice... well okay that is to much customizing for my first try, so ok i'll try to mess around with some console builds now, thanks |
| 09:15.32 | tmzt | so ioremap was failing? not sure I'm following |
| 09:16.10 | dzo | tmzt: there is a region in physical memory that causes AMSS to crash if it's written to. could be mdp-ppp blit buffers or something else. |
| 09:16.33 | maejrep | tmzt: "real" ids as in cc, ce, c4, etc are not the actual IDs. those are the IDs shifted left 1 bit. the real IDs are 0x62, 66, and 67 |
| 09:16.53 | tmzt | ah, how were they working then? |
| 09:17.11 | cr2 | maejrep: and the accelerometer id is 0x18 |
| 09:17.14 | tmzt | oh right, it's 7bit |
| 09:17.21 | cr2 | need to check both cams |
| 09:17.49 | tmzt | fuze only has one, 800 has one? |
| 09:18.01 | cr2 | tmzt: the id is shofted left 1 bit, so the LSB is used as a rw flag |
| 09:18.44 | cr2 | umts phones have 2. for videocalls |
| 09:18.58 | maejrep | yeah 800 has only one |
| 09:18.59 | tmzt | ok, so why did the others work in the driver and in i2cdump? |
| 09:19.12 | maejrep | because I was shifting them :p |
| 09:19.14 | cr2 | others ? |
| 09:19.15 | maejrep | 0xce>>1 |
| 09:19.41 | tmzt | how many does diams have? |
| 09:19.50 | maejrep | and I only did that to visually see in the code that it matches my haret dump of "1ce" for example |
| 09:20.04 | cr2 | tmzt: but it's easier to track the "doubled" ids. and wince code uses the doubled values too |
| 09:20.53 | cr2 | so that's why they are in the wiki |
| 09:24.19 | tmzt | alright, you say there is a mfd driver for a similar chip over spi? |
| 09:24.37 | tmzt | Matze88: how many cameras does your device have? |
| 09:25.21 | Matze88 | i have 2 cameras |
| 09:25.36 | Matze88 | back 3.1 Mpix and front small, 640x480 or so? dont really know but i can check |
| 09:27.15 | cr2 | Matze88: yes |
| 09:27.15 | Matze88 | and i think you are right, my device is diam100 not diam300 but i dont know where to check? in WM system -> device info -> identification -> model number it just says HTC Touch Diamond P3700 |
| 09:27.42 | cr2 | Matze88: under the battery |
| 09:29.14 | Marajin_ | 'lo |
| 09:30.02 | Matze88 | front camera has 480x640 pixel (640 is height) |
| 09:30.41 | Marajin_ | Matze88: you can check which version by removing the battery, the 4 digit id and 3 digit number will be there on all HTCs as far as I know |
| 09:31.42 | tmzt | cr2: you say there is a mfd driver for a similar chip over spi? |
| 09:32.41 | cr2 | tmzt: no. it's just the atmegaX8 has an spi port |
| 09:33.30 | Matze88 | ohh i have a DIAM200 |
| 09:33.31 | cr2 | if we would ever do fw update for psoc/microP ... |
| 09:33.44 | cr2 | hehe |
| 09:33.58 | *** join/#htc-linux Xime (n=xime@bankize.net) |
| 09:34.04 | maejrep | 200? :( |
| 09:34.17 | maejrep | cr2: cc,23 doesn't seem to change anything |
| 09:34.22 | Matze88 | maejrep: yes |
| 09:34.26 | tmzt | so 300 is .de |
| 09:34.50 | Matze88 | maejrep: i would do a photo but i have no other cam than the phone's one |
| 09:35.23 | Marajin_ | Hmm, are any new HTCs due for release or am I stuck going for a raph in april? |
| 09:35.26 | Matze88 | tmzt: my phone is a t-mobile DE replacement after warranty (bought this from ebay, don't know what really happened with the first one) |
| 09:35.27 | maejrep | Matze88: I was just alluding to the fact that they keep making new versions ;x |
| 09:36.04 | tmzt | cr2: over the same pins? |
| 09:36.15 | Marajin_ | maejrep: Pretty sure there's a 100, 110, 120, 130, 200, 210, 300, 400, 500, 600, 700 and 800 |
| 09:36.16 | tmzt | or it has pinmux doesn't it |
| 09:36.31 | Matze88 | maejrep: yes, i did not hear anything about DIAM200 in the past, i'm also really sure that i read about diam300 or diam100 somewhere in my phone software (maybe in SPL?) |
| 09:36.42 | tmzt | Matze88: you say you are familiar with avr, have you worked with atmegas? |
| 09:36.47 | Matze88 | tmzt: yes |
| 09:37.25 | tmzt | there are apparently two in the raphaels (Touch Pro) |
| 09:38.41 | Matze88 | tmzt: i did not really got the thing you are talking about, what has to be done with an at mega? |
| 09:39.32 | tmzt | the chip we are writing a driver for is an atmega8?4? with a custom firmware that serves as the keyboard controller, backlight controller, and some other functions |
| 09:39.43 | tmzt | in the touch pro |
| 09:39.56 | cr2 | tmzt: diam* too |
| 09:40.07 | tmzt | the firmware is loaded by the spl (phone boot loader) at boottime |
| 09:40.09 | cr2 | and bstone |
| 09:40.12 | tmzt | cr2: both? |
| 09:40.32 | Matze88 | tmzt: the firmware cant be programmed every time, because flash only has 10000 write cycles on at megas... |
| 09:40.38 | cr2 | tmzt: you can reflash the fw, but it stays there |
| 09:40.52 | cr2 | no need to upload at boot |
| 09:42.08 | maejrep | http://wiki.xda-developers.com/index.php?pagename=RaphaelMicroP - updated with led behavior |
| 09:42.14 | cr2 | Matze88: reflashing the fw seems to be tricky on raph/diam. |
| 09:42.29 | tmzt | it's not ram-based? |
| 09:42.35 | Matze88 | cr2: u mean at mega fw? |
| 09:42.49 | Matze88 | tmzt: no, avrs (at mega XX, every mega) can not execute code from ram |
| 09:42.56 | cr2 | maejrep: i'll gather the SD clock regs dumped values in wiki |
| 09:43.29 | Matze88 | cr2: u mean diamond also has avr on it? but there is no keyboard, only the 7 hw keys... |
| 09:43.52 | Matze88 | or is the touch sensitive area made with that controler? |
| 09:43.55 | cr2 | Matze88: yes. for me it seems that atmega is the master for these transactions, but i'm not 100% sure. we have more urgent problems right now ;) |
| 09:44.04 | Matze88 | yep i knopw |
| 09:44.24 | Matze88 | if you can provide me with a hex dump of the avr code i can look throuhg the disassembly if that would be helpful |
| 09:44.26 | cr2 | Matze88: there are leds and backlight on diam |
| 09:44.35 | *** join/#htc-linux goxboxlive (n=goxboxli@185.84-48-126.nextgentel.com) |
| 09:45.03 | maejrep | Matze88: navi touch area is handled by the controller |
| 09:45.28 | maejrep | responds with several channels of data |
| 09:45.34 | cr2 | hmm. i think the fw is loaded from NAND. who has the full nand dump here ? :) |
| 09:45.46 | cr2 | maejrep: navi is on psoc |
| 09:45.49 | Matze88 | maejrep: oh that is really nice :-) didnt know any commercial application with atmel avr controllers before :-) |
| 09:46.10 | maejrep | oh, nevermind then :P |
| 09:46.24 | maejrep | thought you were talking about i2c in general :x |
| 09:46.28 | cr2 | ok |
| 09:46.50 | cr2 | Matze88: but the fw may be built-in in the spl. |
| 09:47.07 | cr2 | Matze88: i'm looking at ther things right now. |
| 09:47.07 | Matze88 | cr2: so it has to be in custom spls too? |
| 09:47.33 | *** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net) |
| 09:47.44 | tmzt | isn't the blackstone version known, it was in srec blocks wasn't it? |
| 09:47.58 | cr2 | Matze88: they are checking the checksums & versions somehow. so probably you can't just upload $randomjunk |
| 09:48.08 | *** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net) |
| 09:49.42 | cr2 | maejrep: cc,20,02,00 - lcd backlight ? |
| 09:49.49 | cr2 | maejrep: in the wiki |
| 09:50.05 | cr2 | maejrep: the last value is brightness ? |
| 09:50.31 | cr2 | cc,22,val LCDbrightness wr |
| 09:51.34 | cr2 | cc,20 looks like leds, but the backlight may be a led too :) |
| 09:53.52 | maejrep | i don't know |
| 09:55.22 | cr2 | 20 must be leds |
| 09:55.31 | maejrep | lcd backlight doesn't seem to be changing anymore.. earlier i got it to fade into whatever brightness i gave it, and then it would snap back to what it was |
| 09:55.39 | maejrep | 20 isn't changing anything for me |
| 09:55.54 | cr2 | cc,20, ledmisc |
| 09:55.55 | cr2 | cc,20,02,00 - lcd backlight ? |
| 09:55.55 | cr2 | cc,20,0c,00 - led |
| 09:56.11 | cr2 | lcd backlight looks strange in this company |
| 09:56.27 | maejrep | is there a reset? :P |
| 09:56.33 | maejrep | i may have screwed something up |
| 09:56.34 | maejrep | :) |
| 09:56.55 | cr2 | i think cc,20,02,00 is from lcd powerup sequence |
| 09:57.03 | Matze88 | i have another problem right know: i can't make linux boot... i start haret, let it execute default.txt, it displays progress bar and stops at 100% (no progress in 5 minutes, after that i do soft reset). could this be a zImage issue? I tried different initrds but only my own compiled zImage (current git repo without changes) |
| 09:57.06 | *** join/#htc-linux kaze_ (n=kaze@ABordeaux-158-1-27-51.w90-55.abo.wanadoo.fr) |
| 09:57.13 | maejrep | things like "turnoffmethod" and "ledautomode" - do you know what those are specifically, or just basing it on debug message? |
| 09:57.33 | cr2 | all the data is from ioctls |
| 09:58.12 | cr2 | Matze88: it |
| 09:58.24 | cr2 | s possible to boot without initrd |
| 09:58.41 | cr2 | Matze88: so it must be your zimage |
| 09:59.32 | tmzt | check mtype and check your CONFIG_MACHs in .config |
| 09:59.42 | cr2 | yeah, cc,20,02,0 is from lcd powerup, so discard it as a backlight control |
| 10:00.06 | cr2 | must be some led, syncronized with the LCD |
| 10:00.52 | tmzt | Matze88: check mtype and check your CONFIG_MACHs in .config |
| 10:01.10 | cr2 | tmzt: this maze of mtypes may become a problem ;) |
| 10:01.41 | tmzt | have you found anything reliable in the amss or microp for a version, particularly for keyboard layout? |
| 10:02.20 | cr2 | keyboard layout is not a kernel issue |
| 10:02.32 | tmzt | actually, is there a good reason to not just enable sdc1,2,3 and let enum handle it? |
| 10:02.44 | cr2 | the versions are for the firmware, but they are not critical right now. |
| 10:02.47 | Matze88 | make htcdiamond_defconfig ARCH=arm |
| 10:02.47 | Matze88 | make zImage ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- |
| 10:02.47 | Matze88 | CONFIG_MACH_HTCDIAMOND=y |
| 10:02.47 | Matze88 | mtype is nowhere in .config, in default.txt i wrote mtype=1910 as it should be for diamond... |
| 10:03.06 | tmzt | I mean physical layout but since this is not a matrix that might not be a problem |
| 10:03.16 | cr2 | htc must be crazy to change the i2c regs |
| 10:03.51 | tmzt | raph200 could have another new lcd/controller |
| 10:04.01 | cr2 | tmzt: the matrix gpios are all in the wiki |
| 10:04.31 | cr2 | tmzt: then somebody should dump the spl from sram |
| 10:04.55 | tmzt | cr2: I mean a standard matrix qwerty keyboard where this could be an issue, only that the kernel ksyms wouldn't match keys but that some combiniations are not possible and could cause problems if they were active |
| 10:05.30 | tmzt | Matze88: can you install haret and get an ip connection from a pc? |
| 10:05.57 | Matze88 | tmzt: yep that works |
| 10:07.01 | tmzt | cr2: do you have the instructions? |
| 10:07.32 | cr2 | tmzt: as long as they do not overlap, you can always use loadkeys |
| 10:07.33 | maejrep | didnt the diamond machtype change? |
| 10:07.42 | maejrep | 1910 should be raph |
| 10:07.52 | cr2 | tmzt: pwf myspl 0x0 0x100000 |
| 10:07.57 | Matze88 | but on wiki it says 1910 is for diamond too |
| 10:08.06 | maejrep | that was changed |
| 10:08.11 | Matze88 | oh, to what number? |
| 10:08.19 | maejrep | there's now a diamond machtype |
| 10:08.30 | cr2 | Matze88: there was a time when it was the same kernel |
| 10:08.50 | maejrep | htcdiamond MACH_HTCDIAMOND HTCDIAMOND 1805 |
| 10:09.10 | Matze88 | maejrep: ahh thank you, i'll try |
| 10:09.44 | maejrep | and yeah behavior is consistent with unsupport machtype |
| 10:09.48 | maejrep | +ed |
| 10:10.04 | maejrep | it'll just hang after the progress bar gets to 100% |
| 10:10.04 | Matze88 | it woooorks :-) |
| 10:10.29 | cr2 | maejrep: where is the brightness sensor on i2c ? |
| 10:10.39 | *** join/#htc-linux timebomb (n=tb@213.39.244.82) |
| 10:10.40 | maejrep | hmm, not sure |
| 10:10.51 | maejrep | i don't think it is on i2c actually.. |
| 10:10.55 | maejrep | or its a read channel on cc |
| 10:11.06 | cr2 | cc is i2c |
| 10:11.11 | tmzt | I remember it being an adc for some reason |
| 10:11.26 | cr2 | the sensor itself is on an ADC obviously |
| 10:11.28 | maejrep | haretlog-20090116_010225.log-059337: mmutrace 7803d720: e5853000(str) b2300000 000001cc (00000000) |
| 10:11.28 | maejrep | haretlog-20090116_010225.log-059337: mmutrace 7803d790: e5853000(str) b2300000 00000033 (00000000) |
| 10:11.28 | maejrep | haretlog-20090116_010225.log:059337: mmutrace 7803d878: e5853000(str) b2300000 000001cd (00000000) |
| 10:11.28 | maejrep | haretlog-20090116_010225.log-059337: mmutrace 7803d940: 05852000(str) b2300000 00000200 (00000000) |
| 10:11.28 | maejrep | haretlog-20090116_010225.log-059337: mmutrace 7803d944: e5933000(ldr) b230000c 00000000 (00000000) |
| 10:11.31 | maejrep | haretlog-20090116_010225.log-059337: mmutrace 7803d944: e5933000(ldr) b230000c 00000002 (00000000) |
| 10:11.32 | tmzt | whether it was on msm or someelse I don't remember |
| 10:11.35 | maejrep | that's my guess ;x |
| 10:12.00 | maejrep | microp wiki says cc,33 is "ADCval read" |
| 10:12.02 | cr2 | but ADC is on microP |
| 10:12.09 | cr2 | ok |
| 10:12.09 | tmzt | I thought there where devices where it was a gpio? |
| 10:12.23 | cr2 | maejrep: how many leds do we have ? |
| 10:12.35 | cr2 | tmzt: universal |
| 10:12.42 | Matze88 | mh i don't think it should stop booting (and hang, non-blinking cursor) @ mmc initialisation? |
| 10:12.53 | tmzt | maejrep: did you get the keyboard slide switch working, the gpio_read that was failing? |
| 10:12.54 | cr2 | but on older ipaqs it was an ADC too. |
| 10:12.58 | maejrep | 4 in the center ring itself, but we can't control them individually, just behavior. plus the 4 outer navi button backlights |
| 10:13.10 | maejrep | 2 led's on the keyboard but I can't find out how to control them yet |
| 10:13.25 | maejrep | tmzt: it wasn't failing, it was just returning -1 instead of 1 |
| 10:13.34 | tmzt | so the !! worked? |
| 10:13.37 | maejrep | but the behavior was consistent |
| 10:13.42 | maejrep | yes |
| 10:14.04 | cr2 | maejrep: ok |
| 10:14.15 | tmzt | and you are input_report'ing a level with SW_LID? |
| 10:14.24 | maejrep | no, not currently |
| 10:15.15 | *** join/#htc-linux radem205 (n=aaa@e144118.upc-e.chello.nl) |
| 10:15.36 | maejrep | right now that gpio is claimed in board-raph-keypad, but jobo's point was that -keypad isn't useful by itself anyway, because it needs i2c navi data to tell home/send apart, etc |
| 10:15.38 | cr2 | maejrep: the stylus gpio is known ? |
| 10:15.49 | maejrep | so will probably be refactored |
| 10:15.58 | maejrep | cr2: i thought so? I haven't tried it myself |
| 10:16.13 | cr2 | is it in the wiki ? |
| 10:16.33 | maejrep | don't see it searching for stylus |
| 10:16.39 | maejrep | could be one of the blank descriptions though |
| 10:16.48 | maejrep | sec i'll reboot and check |
| 10:17.14 | *** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net) |
| 10:17.14 | maejrep | cr2: i want to be able to type 'reboot' and get a soft reset |
| 10:17.15 | maejrep | :D |
| 10:17.27 | *** join/#htc-linux Mullins (n=bw@89.204.230.161) |
| 10:17.29 | cr2 | maejrep: what is soft reset ? |
| 10:17.38 | maejrep | reboot :| |
| 10:17.53 | cr2 | maejrep: there is arm11 "sleep" mode + devices suspend |
| 10:17.55 | maejrep | like hitting the reset button with the stylus |
| 10:18.00 | cr2 | and there is full arm9 suspend |
| 10:18.09 | *** join/#htc-linux Tinyboom (n=nahh@108.84-49-166.nextgentel.com) |
| 10:18.15 | maejrep | i just want it to reboot :P |
| 10:18.24 | cr2 | maejrep: it's a gpio event |
| 10:18.26 | tmzt | it seems that arm9 has to do the reboot, rebooting arm11 wouldn't work |
| 10:18.38 | cr2 | yes |
| 10:18.48 | maejrep | yeah all that is not entirely clear to me yet |
| 10:18.58 | maejrep | but i think i get it |
| 10:18.59 | cr2 | but it's a proc_comm + some magic ascii codes ;) |
| 10:19.10 | maejrep | not 776655aa? |
| 10:19.16 | maejrep | (that's trout specific?) |
| 10:19.37 | maejrep | wiki shows in smem: +0xfc08c softreset =0x55aa00ff |
| 10:19.39 | cr2 | maejrep: there are many magic ascii code, they probably need a wiki page too |
| 10:19.47 | maejrep | lol |
| 10:20.06 | cr2 | what is 7766 in ascii ? |
| 10:20.23 | cr2 | don't forget that g1 has different amms |
| 10:20.28 | cr2 | s/amms/amss/ |
| 10:20.44 | cr2 | so only the wince devices are intercompatible |
| 10:22.17 | maejrep | cr2: on raph800 its 1-21 |
| 10:22.20 | maejrep | gpio 37 |
| 10:22.28 | maejrep | which is listed as sd card detect on wiki :) |
| 10:22.40 | cr2 | for raph100 ? |
| 10:22.44 | maejrep | ? |
| 10:23.21 | maejrep | maybe its gpio 39 :P |
| 10:23.25 | cr2 | yes, for raph100 |
| 10:23.34 | cr2 | 37 is sd detect |
| 10:23.41 | cr2 | but not for raph800 |
| 10:23.47 | maejrep | right which on raph800 sd is 38 |
| 10:24.02 | maejrep | 38 moved to 39 |
| 10:24.09 | maejrep | so maybe my 37 is raph100's 39 :P |
| 10:25.13 | cr2 | i remeber that people wee complaining about dmesg spam with trout/halibut keypad driver on raph100 |
| 10:25.25 | tmzt | you mean the reset button is gpio 37? |
| 10:25.34 | maejrep | no |
| 10:25.39 | maejrep | not on raph800 anyway |
| 10:25.46 | maejrep | raph800, gpio 37 is the stylus |
| 10:26.09 | cr2 | i'm talking about the stylus |
| 10:26.29 | tmzt | ok |
| 10:26.34 | cr2 | 36 0x24 3,0 20 I reset button, irq. |
| 10:26.43 | maejrep | marked on wiki |
| 10:26.49 | cr2 | yes |
| 10:26.53 | cr2 | you have 37 ? |
| 10:27.01 | maejrep | i mean i marked 37 as raph800 stylus, on wiki |
| 10:27.07 | cr2 | ok |
| 10:27.24 | cr2 | hmm. where is the stylus on raph100 |
| 10:27.29 | cr2 | i hate windows ;) |
| 10:27.37 | tmzt | so there is a stylus detect but the values have to read over i2c? |
| 10:27.43 | maejrep | no |
| 10:27.45 | cr2 | how do i do a net install on nc10 ? |
| 10:27.48 | maejrep | stylus is not on i2c |
| 10:27.52 | maejrep | its just gpio high/low |
| 10:27.57 | maejrep | kind of like sd card insert status |
| 10:28.03 | maejrep | or clamshell open/close status |
| 10:28.09 | tmzt | why am I confusing navi and res screen again?? |
| 10:28.11 | cr2 | yes, i know |
| 10:28.36 | maejrep | cr2: that was to tmzt ;p |
| 10:29.22 | cr2 | ok |
| 10:31.40 | tmzt | cr2: what did you find about mddi? |
| 10:32.02 | tmzt | the controller chip has gpios that are controlable through commands on the mddi bus? |
| 10:32.47 | cr2 | you can query them to determine the lcd pabel type |
| 10:33.05 | cr2 | it's all documented in the wiki |
| 10:33.11 | cr2 | at least for raph100 |
| 10:33.12 | tmzt | have we verified how many controllers there actually are? |
| 10:33.15 | tmzt | I'll look |
| 10:36.46 | cr2 | maejrep: this is on vogue -> msm_proc_comm(0x115,0x200,0); |
| 10:38.10 | cr2 | ok, vogue uses SDC1 for sd. i'll add that to wiki |
| 10:38.53 | cr2 | tita100 has wifi ? |
| 10:39.36 | tmzt | yeah, pretty sure it does |
| 10:39.46 | tmzt | even the verizon one |
| 10:41.38 | cr2 | 1d is arm9 low speed ?? |
| 10:42.18 | cr2 | why does vogue do it on lcd panel init ?? |
| 10:42.23 | tmzt | another mailbox command? |
| 10:42.24 | cr2 | msm_proc_comm(0x11d,0x1,0); |
| 10:42.50 | tmzt | no, ok |
| 10:43.19 | cr2 | looks weird |
| 10:44.45 | cr2 | yes. the lcd panel is sync'd with the arm9 speed :) |
| 11:08.51 | *** join/#htc-linux Balsat (n=kll@87.72.13.87) |
| 11:24.11 | *** join/#htc-linux kiozen (n=oeichler@p54920847.dip0.t-ipconnect.de) |
| 11:25.00 | *** join/#htc-linux Untouchab1e (n=Untoucha@cm-84.209.89.86.getinternet.no) |
| 11:25.07 | Untouchab1e | Good morning :) |
| 11:35.41 | *** join/#htc-linux kiozen (n=oeichler@p54920847.dip0.t-ipconnect.de) |
| 11:51.08 | *** join/#htc-linux MethoS (n=lem@host-091-096-214-175.ewe-ip-backbone.de) |
| 12:24.51 | *** join/#htc-linux oliwan (n=oliwn@client-80-3-103-90.bsh-bng-012.adsl.virginmedia.net) |
| 12:27.01 | oliwan | Does anyone kno what the latest branch on git repo is for Diam/Raph? |
| 12:32.00 | *** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes) |
| 12:35.22 | *** join/#htc-linux hollo (n=hollo@3e6b025d.rev.stofanet.dk) |
| 12:48.33 | cr2 | dcordes: hi |
| 12:51.28 | dcordes | hello |
| 12:51.41 | *** part/#htc-linux oliwan (n=oliwn@client-80-3-103-90.bsh-bng-012.adsl.virginmedia.net) |
| 12:53.05 | cr2 | dcordes: do you know where is the sd/wifi located on tita100 ? |
| 12:54.14 | cr2 | titan uses the vogue code, but afaik it has wifi ? |
| 12:55.52 | cr2 | dcordes: http://wiki.xda-developers.com/index.php?pagename=MSM_SDIO |
| 12:55.55 | cr2 | at the bottom |
| 12:59.59 | cr2 | maejrep: your a0/a4 values are for wifi, can you dump the "sd" values ? |
| 13:01.15 | cr2 | looking for somebody with g1 and /dev/mem access ... |
| 13:05.22 | dcordes | cr2: I looked at the htc-vogue branch but can't see something that makes msm_sdcc changes specific to titan. don't know where wifi/sd are |
| 13:05.51 | dcordes | dream_kill said he has a full nand dump from the g1 |
| 13:13.00 | *** join/#htc-linux Sti_0239 (n=Where_is@118.141-65-87.adsl-dyn.isp.belgacom.be) |
| 13:14.29 | *** join/#htc-linux cr2 (n=cr2@ip-90-187-221-92.web.vodafone.de) |
| 13:15.41 | *** join/#htc-linux Kevin2 (n=Kevin2@207-237-194-161.c3-0.avec-ubr2.nyr-avec.ny.cable.rcn.com) [NETSPLIT VICTIM] |
| 13:15.41 | *** join/#htc-linux Abracadabra (n=aaabraca@unaffiliated/abracadabra) [NETSPLIT VICTIM] |
| 13:15.41 | *** join/#htc-linux ponicg_ (n=ponicg@ool-182d907a.dyn.optonline.net) [NETSPLIT VICTIM] |
| 13:25.40 | *** join/#htc-linux kiozen (n=oeichler@p54920847.dip0.t-ipconnect.de) |
| 13:43.04 | *** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
| 13:56.42 | cr2 | maejrep: http://forum.xda-developers.com/showthread.php?t=471033 |
| 13:58.54 | *** join/#htc-linux mickey_jaw (n=mike@78-105-123-214.zone3.bethere.co.uk) |
| 14:01.36 | *** join/#htc-linux Zinbolic (n=chatzill@84.238.80.215) |
| 14:21.28 | cr2 | the tiacx driver deserves the openoffice/mozilla award ;) |
| 14:22.31 | cr2 | in the ugliest spaghetti programming code contest. |
| 14:24.12 | cr2 | ok, that's something useful |
| 14:24.16 | cr2 | #define TROUT_IRQ MSM_GPIO_TO_INT(29) |
| 14:26.28 | cr2 | power and reset are known |
| 14:27.17 | cr2 | carddetect ? where is it implemented ... |
| 14:27.48 | cr2 | TIWLAN_MSM7000 must be defined |
| 14:31.14 | *** join/#htc-linux MethoS (n=lem@host-091-096-214-175.ewe-ip-backbone.de) |
| 14:32.56 | *** join/#htc-linux MethoS (n=lem@host-091-096-214-175.ewe-ip-backbone.de) |
| 14:35.10 | *** join/#htc-linux MethoS- (n=lem@host-091-097-241-184.ewe-ip-backbone.de) |
| 15:13.39 | *** join/#htc-linux Zinbolic_ (n=chatzill@84.238.80.215) |
| 15:40.29 | *** join/#htc-linux MethoS (n=lem@host-091-097-241-184.ewe-ip-backbone.de) |
| 15:51.54 | *** join/#htc-linux kiozen (n=oeichler@p54920847.dip0.t-ipconnect.de) |
| 16:20.06 | *** part/#htc-linux Zinbolic_ (n=chatzill@84.238.80.215) |
| 16:20.20 | *** join/#htc-linux Zinbolic_ (n=chatzill@84.238.80.215) |
| 16:23.56 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfd02c.pool.einsundeins.de) |
| 16:49.00 | *** join/#htc-linux radem205 (n=aaa@e144118.upc-e.chello.nl) |
| 16:56.41 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 17:44.35 | lupine_85 | cr2: keyboard didn't work for me last time I tried it. dunno if there's been any progress since then |
| 17:46.01 | lupine_85 | has distractions present |
| 17:49.55 | *** join/#htc-linux chab7 (n=kvirc@212.92.4.114) |
| 17:51.19 | *** join/#htc-linux anarsoul (n=Anarchis@213.184.241.199) |
| 17:54.17 | anarsoul | Hi there, does anybody got pwm-backlight driver working on s3c24xx-based devices? I'm stucked on calculating correct pwm_period_ns and brightness values (I have only appropriate tcnt and tcmp values:)) |
| 17:59.22 | *** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde) |
| 18:02.51 | *** join/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey) |
| 18:03.56 | *** join/#htc-linux exco (n=exco@e181075236.adsl.alicedsl.de) |
| 18:36.22 | *** join/#htc-linux Mullins07 (n=bw@89.204.246.114) |
| 18:37.13 | *** join/#htc-linux stefan_schmidt (n=stefan@p5B0365EB.dip.t-dialin.net) |
| 18:37.29 | *** join/#htc-linux Balsat (n=kll@87.72.13.87) |
| 18:41.31 | *** join/#htc-linux pH5 (n=ph5@e178241133.adsl.alicedsl.de) |
| 18:42.28 | *** join/#htc-linux nato2k (n=templarn@76.250.180.218) |
| 18:59.16 | *** join/#htc-linux nato2k (n=templarn@76.250.182.142) |
| 19:07.14 | *** join/#htc-linux kiozen (n=oeichler@rgnb-5d87d5f2.pool.einsundeins.de) |
| 19:27.39 | *** join/#htc-linux stefan_schmidt_ (n=stefan@p5B036CEE.dip.t-dialin.net) |
| 19:28.28 | *** join/#htc-linux cr2 (n=cr2@ip-77-24-230-20.web.vodafone.de) |
| 19:33.14 | cr2 | it's quiet today |
| 19:33.34 | Mullins | yeah |
| 19:37.36 | cr2 | Mullins: any good ideas ? |
| 19:43.20 | Mullins | cr2: no sorry :) |
| 19:43.50 | zycho | ill have a drink |
| 19:43.57 | cr2 | what phone do you have ? |
| 19:44.03 | Mullins | kaiser |
| 19:44.15 | zycho | diam300 |
| 19:44.25 | cr2 | ok |
| 19:44.56 | cr2 | zycho: not in the sd list ? |
| 19:45.12 | cr2 | zycho: is it a cdma or umts phone ? |
| 19:45.18 | zycho | umts |
| 19:45.24 | zycho | whats the sd list? |
| 19:46.22 | cr2 | http://wiki.xda-developers.com/index.php?pagename=MSM_SDIO |
| 19:46.52 | cr2 | not yet |
| 19:47.11 | zycho | just tell me that to do ;) |
| 19:47.13 | cr2 | can you dump the spl from sram ? |
| 19:47.35 | cr2 | you have haret on the phone ? |
| 19:47.39 | zycho | yes |
| 19:48.27 | cr2 | pwf d300spl 0x0 0x100000 |
| 19:49.44 | zycho | do i have to enter this in console? |
| 19:51.16 | *** join/#htc-linux timebomb (n=tb@c150128.adsl.hansenet.de) |
| 19:51.27 | cr2 | you can add that to default.txt, and then press 'Run' |
| 19:52.00 | zycho | ok ill try it |
| 20:08.47 | *** join/#htc-linux pH5 (n=ph5@e178223139.adsl.alicedsl.de) |
| 20:11.28 | zycho | set mtype 1805 |
| 20:11.28 | zycho | set kernel zImage-kll |
| 20:11.28 | zycho | set initrd initrd.droid.cpio.gz |
| 20:11.28 | zycho | set ramsize 0x6000000 |
| 20:11.29 | zycho | set ramaddr 0x10000000 |
| 20:11.31 | zycho | set cmdline "mem=64M msmsdcc_id=2 msmts_calib=0x7a.0x5e.0x35a.0x37f" |
| 20:11.33 | zycho | boot |
| 20:11.35 | zycho | pwf d300spl 0x0 0x100000 |
| 20:11.37 | zycho | is this right? |
| 20:11.50 | zycho | or shoult i add it to cmdline |
| 20:11.53 | cr2 | comment out the 'boot' |
| 20:12.04 | zycho | ah ok |
| 20:12.13 | cr2 | you don't need to boot |
| 20:14.36 | zycho | http://zown.org/ppl/zycho/d300spl.rar |
| 20:15.36 | cr2 | downloading |
| 20:16.50 | cr2 | got it |
| 20:22.09 | zycho | ok |
| 20:23.47 | cr2 | i'll check the sd options |
| 20:26.35 | zycho | is the spl the bootloader? |
| 20:27.28 | nato2k | this would all be a lot easier if the cursor keys on my phone worked, never let your one year old daughter sneak off with your cell phone! |
| 20:28.07 | zycho | hehe |
| 20:28.51 | *** join/#htc-linux Summerspittman (n=secondsu@c-24-99-80-238.hsd1.ga.comcast.net) |
| 20:30.40 | *** join/#htc-linux xianthax (n=xianthax@c-75-69-145-210.hsd1.ma.comcast.net) |
| 20:59.14 | *** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be) |
| 21:15.00 | cr2 | zycho: wifi on sdc1 and sd on sdc2. as expected |
| 21:15.50 | cr2 | added to the list |
| 21:16.02 | zycho | ok |
| 21:16.26 | zycho | does not say much to me ;) |
| 21:16.39 | cr2 | zycho: does linux boot for you ? |
| 21:16.49 | zycho | yes |
| 21:16.59 | cr2 | ok |
| 21:17.10 | zycho | i am using Android_Diam100-20091501.zip |
| 21:18.17 | cr2 | what is the difference to diam100 ? |
| 21:18.54 | zycho | dont know, diam300 is o2 xda device |
| 21:18.55 | *** join/#htc-linux paulproteus (n=paulprot@2002:db69:2513:0:0:0:0:1) |
| 21:19.22 | cr2 | k |
| 21:19.23 | zycho | other accu cover |
| 21:20.12 | cr2 | i doubt that it has any influence on the linux performance :) |
| 21:20.31 | cr2 | NetRipper: are you alive ? |
| 21:20.33 | zycho | i think you're right |
| 21:21.57 | cr2 | your spl says diam100000 |
| 21:22.35 | *** join/#htc-linux chris10230 (n=chatzill@82-32-35-72.cable.ubr05.aztw.blueyonder.co.uk) |
| 21:24.07 | zycho | bootloader modus says DIAM300 MFG 64M |
| 21:29.51 | cr2 | ok |
| 21:32.49 | cr2 | stored at fc118 |
| 21:34.11 | cr2 | +0xfc118 batt, see diamond |
| 21:34.14 | cr2 | strange |
| 21:35.09 | cr2 | +0xfc118 2b batt discharge |
| 21:35.46 | cr2 | not on your spl, obviously |
| 21:48.48 | *** join/#htc-linux timebomb (n=tb@e176097142.adsl.alicedsl.de) |
| 21:50.44 | *** join/#htc-linux MethoS (n=lem@host-091-096-214-073.ewe-ip-backbone.de) |
| 21:51.06 | *** join/#htc-linux Balsat (n=kll@87.72.13.87) |
| 22:09.00 | *** join/#htc-linux Mullins (n=bw@89.204.193.250) |
| 22:18.38 | *** join/#htc-linux nato2k (n=templarn@76.250.182.142) |
| 22:24.02 | tmzt | anarsoul: you can ask in #openmoko/#openmoko-cdevel but I don't know if they use the pwm |
| 22:25.03 | anarsoul | tmzt: nevermind, I've already figured out how to calculate period and duty :) |
| 22:30.25 | *** join/#htc-linux yoyey (n=yoann@lns-bzn-49f-81-56-185-87.adsl.proxad.net) |
| 22:33.57 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 22:38.31 | *** join/#htc-linux Summerspittman (n=secondsu@c-24-99-93-227.hsd1.ga.comcast.net) |
| 22:42.01 | *** join/#htc-linux nizox (n=none@2a01:e35:8a13:a2b0:21c:c0ff:fe25:ff68) |
| 22:47.55 | *** part/#htc-linux yoyey (n=yoann@lns-bzn-49f-81-56-185-87.adsl.proxad.net) |
| 22:53.02 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) |
| 22:53.50 | *** join/#htc-linux seand (n=sean@adsl-69-232-201-43.dsl.pltn13.pacbell.net) |
| 23:04.34 | cr2 | anybody not sleeping ? |
| 23:04.42 | tmzt | yeah |
| 23:07.07 | *** join/#htc-linux Kevin2 (n=Kevin2@207-237-194-161.c3-0.avec-ubr2.nyr-avec.ny.cable.rcn.com) [NETSPLIT VICTIM] |
| 23:07.07 | *** join/#htc-linux Abracadabra (n=aaabraca@unaffiliated/abracadabra) [NETSPLIT VICTIM] |
| 23:07.07 | *** join/#htc-linux ponicg_ (n=ponicg@ool-182d907a.dyn.optonline.net) [NETSPLIT VICTIM] |
| 23:07.32 | exco | :-) |
| 23:08.59 | BabelO | :) |
| 23:09.26 | BabelO | cr2: still with your umts WINDOWS connection ? |
| 23:15.07 | cr2 | hehe |
| 23:15.17 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 23:17.33 | cr2 | BabelO: i have put the umts card in raph, but i have no idea how to configure it :) |
| 23:17.52 | cr2 | wince is not very user-friendly |
| 23:18.06 | BabelO | cr2: yes not at all |
| 23:19.16 | BabelO | cr2: i still fail with my syren adress :( |
| 23:20.38 | maejrep | <cr2> maejrep: your a0/a4 values are for wifi, can you dump the "sd" values ? <-- the sd values I could get to work with the formula with some corrections (at least what I think are corrections :p) |
| 23:20.48 | BabelO | cr2: right now i m testing my computer, latest nvidia driver froze all -> http://www.phoronix-test-suite.com/?k=features |
| 23:22.48 | maejrep | cr2: but my args came out different from anything of those you listed |
| 23:22.53 | cr2 | maejrep: the formula can't be wrong. i've checked all spls for 7x00 and 7x0xA devices i have |
| 23:23.23 | cr2 | maejrep: probably wince uses different frequences |
| 23:23.35 | cr2 | BabelO: what is syren address ? |
| 23:24.02 | cr2 | BabelO: btw, the tiacx spaghetti code includes omap1650 support |
| 23:24.32 | maejrep | cr2: http://www.privatepaste.com/f40hbM1fgV |
| 23:24.45 | BabelO | cr2: 0xfffc1800 |
| 23:24.52 | cr2 | maejrep: have you seen that vogue switches the arm9 speed on lcd off ? |
| 23:24.58 | BabelO | cr2: ok, i did not check it |
| 23:25.09 | maejrep | I saw you commented on it |
| 23:25.12 | cr2 | BabelO: why do you have a problem with it ? |
| 23:26.07 | BabelO | cr2: i always get kernel page fault because of wrong virtual adress message when i try read or write of thei adress |
| 23:26.14 | cr2 | maejrep: these are for wifi ? |
| 23:26.24 | exco | should a kernel built for Magician from the android.git.kernel.org git (there's a magician_defconfig in arch/arm/configs) boot with e.g. initrd.gz & system.img from Vogue? (it doesn't ;-) ) |
| 23:26.37 | cr2 | BabelO: that's odd |
| 23:26.37 | BabelO | cr2: the strange thing is that my ts driver with the other spi regiter at 0xfffc1000 are working with the same code |
| 23:26.37 | maejrep | cr2: no that's sd |
| 23:27.02 | cr2 | maejrep: a0/a4 are for SDC1 |
| 23:27.16 | maejrep | cr2: the numbers at the bottom, in the C comment, are the actual sd values, dumped from b0 and b4 |
| 23:27.26 | cr2 | maejrep: i've heavily edited the MSM_SDIO page, check it. |
| 23:27.31 | maejrep | cr2: the variable is called a0/a4, but its actually b0/b4 |
| 23:27.32 | cr2 | ok |
| 23:28.35 | cr2 | hmm. may we should gather all register values. it may be interesting if somebody will dump the MSM_CLK area on g1 |
| 23:30.51 | maejrep | cr2: I think the only way this deviates from what you have on the wiki is arg7 & 0xa |
| 23:30.58 | cr2 | maejrep: what are the "vogue" numbers ? i find the comments there a bit strange, because they talk about 400MHz divisors |
| 23:31.09 | maejrep | in the wiki you said "arg7 orr 0xa", which i'm not sure what that was supposed to mean |
| 23:31.22 | pH5 | exco: the magician stuff in there is inherited from mainline. it should boot, but no backlight/usb/touchscreen/bluetooth/etc. I just tried updating to 29-rc2, but neither UDC nor flash work right now :/ |
| 23:31.29 | cr2 | 0xa|arg7 |
| 23:31.51 | cr2 | and i have double tildes ;) my wiki problems ;) |
| 23:31.57 | maejrep | cr2: yeah that doesn't work with the numbers i get from haret, because arg7 has to be 0 in mine |
| 23:32.02 | maejrep | yeah, I figured out the tildes :) |
| 23:32.31 | cr2 | maejrep: can you dump the whole 0x200 area ? |
| 23:32.41 | maejrep | but if I used 0xa|arg7, with arg7 being 0, my b4 value would still come out as fffe0a7c |
| 23:32.47 | maejrep | but I needed it to come out as fffe007c |
| 23:32.57 | maejrep | which area? |
| 23:33.05 | *** join/#htc-linux BelzeBoB (n=banan@apn-77-113-91-203.gprs.plus.pl) |
| 23:33.12 | cr2 | maejrep: you can see from the emdh and i2c clock divisora that they also have the same structure 0xa|something |
| 23:33.20 | cr2 | at the same bit location |
| 23:33.52 | cr2 | maejrep: arg7 is always 1 for the hardcoded table values... |
| 23:34.21 | maejrep | I'm just saying what I had to do to reverse the 7 arguments from my b0/b4 values :) |
| 23:34.30 | cr2 | maejrep: ok, i'll check the sdhc.dll now |
| 23:34.58 | exco | hey, pH5, how are you doing? |
| 23:35.25 | maejrep | the vogue sd numbers are in clock-7x000-alt.c -- the first table that is inside an #if 0 block |
| 23:36.15 | maejrep | those are the ones that lavender adjusted from 400MHz to 528MHz divisors |
| 23:37.03 | cr2 | maejrep: i'd like to check the last one |
| 23:37.35 | cr2 | but it's strange that the 400/528 should play any role |
| 23:37.44 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) |
| 23:37.45 | cr2 | what is the TCX0 frequency ? |
| 23:38.37 | cr2 | i'm trying to judge this config knowing the pxa clock generation and distribution, which is well documented |
| 23:38.49 | exco | pH5: well I think it does boot - but as you say - nothing works |
| 23:39.33 | cr2 | maejrep: i'll also check my "wince dmesg" |
| 23:41.06 | cr2 | uh, how i have overlooked it ;) |
| 23:42.25 | cr2 | there is a register clock dump inside sdhc.dll ;) |
| 23:43.06 | cr2 | Sdc2MdREgs |
| 23:43.13 | cr2 | Sdc2NsRegs |
| 23:43.37 | cr2 | GlblClkEna |
| 23:43.54 | cr2 | RowResetReg |
| 23:44.08 | pH5 | exco: I'm just trying to get back into magician business a little bit |
| 23:44.21 | pH5 | I still have hope that we'll get it running before GSM is phased out ;) |
| 23:44.31 | cr2 | lol |
| 23:44.32 | maejrep | lol |
| 23:45.19 | cr2 | pH5: i hope that the asic3_mmc driver will be updated before UMTS will be phased out :) |
| 23:45.46 | maejrep | cr2: i still can't get to my wince dmesg area |
| 23:45.54 | exco | yeah, I was thinking the other day - if there are still some Magicians left once everything is working in Linux |
| 23:46.00 | maejrep | the range on the wiki is all 0's for me |
| 23:46.11 | cr2 | maejrep: do you have 'dump mmu' output ? |
| 23:46.19 | maejrep | yes, somewhere |
| 23:46.36 | cr2 | check the 1MB mappings at the top of 0x1000000000 |
| 23:47.33 | cr2 | aha |
| 23:47.41 | cr2 | it's a huge remapping function |
| 23:48.04 | maejrep | 80000000 | 10000000 | 1MB section | CB AP=1 T=7 ? |
| 23:48.28 | cr2 | the +0x214 that dzo mentioned is also in the list |
| 23:49.01 | cr2 | maejrep: look behind 0x18000000 |
| 23:49.29 | maejrep | 87400000 | 17400000 | 1MB section | CB AP=1 T=7 |
| 23:49.29 | maejrep | 87500000 | 17500000 | 1MB section | CB AP=1 T=7 |
| 23:49.29 | maejrep | 87600000 | | UNMAPPED | |
| 23:49.34 | *** join/#htc-linux Elive_user36_en (n=Elive_us@user-5af2a541.tcl126.dsl.pol.co.uk) |
| 23:49.41 | maejrep | then it jumps to 20000000 |
| 23:49.53 | cr2 | ok |
| 23:49.58 | cr2 | check 174 and 175 |
| 23:50.33 | maejrep | can those be pdump'ed, or should I use vdump with the virtual address? Or will it matter |
| 23:51.37 | cr2 | hmm, it's an array of offsets... |
| 23:51.38 | maejrep | 87400000 | 17400000 | 1MB section | CB AP=1 T=7 |
| 23:51.38 | maejrep | a7400000 | 17400000 | 1MB section | AP=1 T=7 |
| 23:51.38 | maejrep | 87500000 | 17500000 | 1MB section | CB AP=1 T=7 |
| 23:51.38 | maejrep | 9b200000 | 17500000 | 1MB section | CB AP=1 T=7 |
| 23:51.38 | maejrep | a7500000 | 17500000 | 1MB section | AP=1 T=7 |
| 23:51.39 | maejrep | bb200000 | 17500000 | 1MB section | AP=1 T=7 |
| 23:52.12 | cr2 | protected ?? |
| 23:52.12 | tmzt | exco: those are armv6novfp, won't work on pxa |
| 23:52.41 | *** join/#htc-linux MethoS- (n=lem@dyndsl-085-016-163-107.ewe-ip-backbone.de) |
| 23:52.56 | maejrep | i don't know how to tell ;x |
| 23:54.18 | exco | novfp?, tmzt |
| 23:55.31 | tmzt | Md? Ns? |
| 23:55.40 | tmzt | 19:34 < cr2> Sdc2MdREgs |
| 23:55.40 | tmzt | 19:34 < cr2> Sdc2NsRegs |
| 23:56.10 | cr2 | tmzt: i'm trying to resolve the absolute addresses |
| 23:56.18 | cr2 | at least we have some register names now |
| 23:56.24 | Mullins | pH5: Can do you do me a favour? |
| 23:56.30 | tmzt | I was thinking this could give us a hint how they are used |
| 23:56.35 | cr2 | thank you htc engineers :) |
| 23:58.13 | maejrep | cr2: 174 and 175 both had nothing interesting in them |
| 23:58.30 | cr2 | Enable SD TCX0_LOCK |
| 23:59.10 | maejrep | cr2: better than calling them "a0val" and "a4val" ;) |
| 23:59.17 | cr2 | AT@GSDRS= |
| 23:59.19 | cr2 | yeah |
| 23:59.26 | Mullins | pH5: I have requested to join the mobile linux project on linuxtogo so I can push some changes. If you get a chance will you enable it for me please? |
| 23:59.38 | cr2 | M and N are better names |
| 23:59.45 | maejrep | lol |
| 23:59.50 | tmzt | but do we know what they mean? |