IRC log for #htc-linux on 20080714

00:21.55dcordes_dzo: tried a filesystem on card?
00:23.22dzoif you mean a root fs, no, i've tried mounting fat and ext3 partitions and copying back and forth though and that worked.
00:24.43dcordes_yes, rootfs is what I meant. you get no errors any longer copying?
00:25.16dzono, but only with debug messages enabled. still working on it.
00:31.32*** join/#htc-linux pigeon_ (n=pigeon@218-214-60-232.people.net.au)
00:34.20*** join/#htc-linux ginge (n=baz@cpc6-ward3-0-0-cust579.manc.cable.ntl.com)
00:57.09gingedzo: if(freq=100000) did you mean for this to be equals rather than an assignment?
01:04.06dzooops yes, shouldn't make any difference though.
01:09.28gingeokay. I am knocking this on the head for tonight. I will pick this up later. I still get an oops if I comment out the irq check in msmsdcc_irq, and I get exactly the same output if  leave it uncommented... that is the CMD5 request returns 0 and it returns failure at that point in the card transaction.
01:09.44gingedzo: thanks again. night
01:12.15dcordes_night ginge
01:15.02*** join/#htc-linux pronto (n=pronto@pool-71-179-232-167.bltmmd.fios.verizon.net)
02:19.13*** join/#htc-linux pigeon (n=pigeon@60-241-137-179.static.tpgi.com.au)
02:21.07*** join/#htc-linux miknix (n=kvirc@81.193.93.154)
02:34.21effbiaidcordes_: yes, same as before
02:39.47*** join/#htc-linux tcccp (i=hey@223.66.238.89.arpa-addr.in)
02:51.25*** join/#htc-linux the_sys0p (n=the_sys0@cpe-67-49-214-162.bak.res.rr.com)
02:55.49*** join/#htc-linux [loy] (n=nobody@loy.net.ru)
03:35.19dzops
03:35.21dzols
03:36.27[loy])))
03:53.45*** join/#htc-linux the_sys0p (n=the_sys0@cpe-67-49-214-162.bak.res.rr.com)
04:47.37*** join/#htc-linux miknix (n=kvirc@81.193.93.154)
04:52.11*** join/#htc-linux the_sys0p (n=the_sys0@cpe-67-49-214-162.bak.res.rr.com)
05:02.12*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
05:32.57*** join/#htc-linux rob_w (n=bob@M906b.m.pppool.de)
05:53.10*** join/#htc-linux pleemans (n=peter@dD577D009.access.telenet.be)
05:53.44*** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfa784.pool.einsundeins.de)
06:28.07*** join/#htc-linux MrBIOS- (n=aperez@c-76-102-236-252.hsd1.ca.comcast.net)
06:40.16*** join/#htc-linux pronto (n=pronto@pool-71-179-232-167.bltmmd.fios.verizon.net)
06:48.42*** join/#htc-linux AlGe (n=user@chello080109231226.4.uni-klu.teleweb.at)
07:18.29*** join/#htc-linux premy (n=pr@73.205.76-86.rev.gaoland.net)
07:30.16*** join/#htc-linux AlGe (n=user@chello080109231226.4.uni-klu.teleweb.at)
07:52.34*** join/#htc-linux kiozen (n=oeichler@p54928575.dip0.t-ipconnect.de)
08:05.48*** join/#htc-linux ginge_ (n=baz@80.229.188.31)
08:14.10*** join/#htc-linux Bendy (n=sniveri@bb-81-175-215-228.dsl.phnet.fi)
08:42.45premyginge: hi, just sent you more traces.
08:55.51ginge_got them, thanks
08:56.26ginge_so either we have SD1 and SD2 reversed in our code, or the kaiser differs from the polaris
08:56.45ginge_I will have to run some traces of my own so we can compare notes
08:58.25premyginge_: the traces are only for the clock setting part. I still need to check the MSM_SDC1_PHYS.
08:58.41*** join/#htc-linux pronto (n=pronto@pool-71-179-232-167.bltmmd.fios.verizon.net)
08:59.04ginge_I think the clocks are what we have wrong at the moment. That and IRQ handling. dzo is working on cleaning the IRQ level/edge triggering
09:20.16*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
09:26.37*** join/#htc-linux diogene31_ (n=rj@mur31-2-82-243-122-54.fbx.proxad.net)
09:35.13ginge_premy: I have an older haret, do you have alink for the latest, or can you forward yours please
09:36.43*** join/#htc-linux rob_w (n=bob@M906b.m.pppool.de)
09:36.52ginge_premy: also it looks like you may be right about the wifi/sd card assignments. The clock values I see when in Linux on SDC1 are closer to wifi values than SD based on your traces
09:37.01*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
09:40.55*** join/#htc-linux rob_w|laptop (n=rob@M906b.m.pppool.de)
09:54.03cr2kiozen: n560 charging is more complex than i#ve thought :)
09:54.45kiozencr2: sounds like: you destroy your battery right now
09:54.54kiozen:)
09:56.30cr2some fixes are needed to account for the cpld
09:57.19kiozencr2: ok, when should I update the kernel ;)
09:57.39cr2kiozen: if you have some free time you can add the gpio bitmasks here http://www.handhelds.org/moin/moin.cgi/LooxC550Hardware
09:57.48cr2kiozen: in the evening :)
09:58.30cr2the bitmasks like here -> http://wiki.xda-developers.com/index.php?pagename=AthenaGPIO
09:58.45cr2then it's much more easier to read the binary code.
09:59.45kiozencr2: sounds like you want to make the "bock zum gärtner", ok, let's get a login...
09:59.57cr2lol
10:00.28cr2no, i'm just busy with other things now...
10:01.21kiozencr2: no problem to add this stuff, if I would know what to add precisely
10:01.55cr2<PROTECTED>
10:02.10cr2one more column
10:02.31cr2now it's only decimal gpio number and hex.
10:03.01cr2<PROTECTED>
10:07.31cr2there are about 20 ram locations describing the battery state. need to add them to wiki
10:08.09cr2and 2 more  not yet documented gpios found.
10:08.43kiozencr2:  so you want to have in "GPIO init" the 32 bit mask for GPIO 0 to 119?
10:09.37cr2yes, like it's done fore athena
10:10.42kiozencr2: like this: 11 0b 00000800 out 0 gpio 11 0 1 0 1 1 0
10:11.40cr2yes.
10:11.44lamahttp://www.techcrunch.com/2008/07/13/gphone-may-really-happen-and-ammunition-group-may-be-designing-it/
10:12.20kiozencr2: hm, sounds like a real dump job ....
10:12.35kiozencr2: ok, I think I am qualified for that :) can do...
10:14.19cr2kiozen: select and copy column. or awk/perl
10:14.46kiozenyes, definetly :) ok lunch time...
10:17.43cr2lama: we have enough good phones already.
10:38.42premyginge: http://www.handhelds.org/~koconnor/haret/haret-20080622-mmu4.exe
10:46.44ginge_thanks
10:54.04*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
11:00.59*** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net)
11:07.47*** join/#htc-linux rob_w (n=bob@M906b.m.pppool.de)
11:26.55*** join/#htc-linux rob_w|lappi (n=rob@Mb958.m.pppool.de)
11:38.57Foxdiehi cr2, how goes?
11:41.16ginge_can anyone recommend a good GNU/Linux disassembler for ce arm binaries. I am using ChARMeD at the moment and using windows in vmware is an iritant.
11:43.53*** join/#htc-linux excogitation (n=Miranda@e181113012.adsl.alicedsl.de)
11:44.41premyginge: could it be that we need also need to change writel(readl(MSM_CLK_CTL_BASE)& ~(1<<7),MSM_CLK_CTL_BASE) in set_sdcc_host_clock ?
11:44.48premyginge: cegcc ?
11:44.52*** join/#htc-linux rob_w (n=bob@Mb958.m.pppool.de)
11:45.15premyginge: just changed it to 1<<8, and it does not freeze anymore
11:46.17premycould 1<<7 be for a0/a4 and 1<<8 be for a8/ac ?
11:48.23ginge_premy: an interesting thought, worth looking at
12:00.02premyginge_: clk_enable 20
12:00.15premyclk_get_rate 20:-22
12:00.22premyclk_enable 19
12:00.34ginge_is that CMD0 and CMD8?
12:00.37premyclk_set_rate 19:100000
12:00.52premyclk_get_rate 19:100000
12:01.28premyI'm getting this just after the init_dma message
12:02.23ginge_wait... wouldn't we need to shift by 9 because of the width if the registers?
12:02.41premyAre those messages new ?
12:02.57ginge_no, clk_get_rate worked in my original patch
12:07.36premyginge_: I have new traces which include MSM_CLK_CTL_BASE.
12:07.48premy003170: mmutrace 51609654: e5923000(ldr) 4a307000 2015fc2f (00000000)
12:07.48premy003170: mmutrace 5160965c: e5823000(str) 4a307000 2015fd2f (00000000)
12:08.08premywhich must be 1<<8.
12:08.17ginge_yeah, looks like it
12:09.04premymy previous traces included wifi stuff, so that could explain why there was some 1<<7 in them.
12:09.24ginge_did you read dzo's interpretation of how the clock lines work?
12:09.43premypremy: no, where is that ?
12:09.53premys/premy/ginge
12:10.20ginge_dzo's site in the diff file there is comments
12:13.06kiozencr2:  done
12:44.33*** join/#htc-linux rob_w|laptop (n=rob@Mbeb1.m.pppool.de)
12:46.02*** join/#htc-linux Robwoerle (n=bob@Mbeb1.m.pppool.de)
12:55.22kiozencr2: how can I control the LEDs?
12:55.45*** join/#htc-linux zoolooc (n=lucian@nrbg-4dbfa784.pool.einsundeins.de)
12:56.38ginge_premy: with
12:57.10ginge_premy: with
12:57.14ginge_wtf?
12:57.24ginge_premy: with left shit 8 I get to the same place as you
12:57.35ginge_s/shit/shift/
12:58.15premyginge: is that any further than yesterday, though ?
12:59.02ginge_premy: difficult to say. yesterday I was writing what we presume to be wifi registers, so I guess now we are using the right registers, yes.
12:59.17ginge_premy: it doesn't lock at least
12:59.43ginge_now to extract these clock lines from the sd dll
13:05.00Marajinallo allo allo
13:07.17ginge_alreet
13:07.22ginge_hows the hangover?
13:07.42MarajinI don't get hangovers =)
13:08.08ginge_lucky bugger
13:08.14Marajin's me =)
13:19.34Marajinginge_: So what'd I miss in my drunken haze? :)
13:20.05ginge_not much. Premy and I are still trying to figure out where our SD clocks are
13:21.00Marajindidn't dzo work out a way to reverse engineer/disassemble the clocks from the .dlls?
13:23.02ginge_it's not really tricky to disasseble, its interpreting and following the resulting assembly code thats the pita
13:23.30Marajinwell yeah
13:23.33ginge_also dzo only has 1 sd host driver,  we have 2
13:24.01Marajintruuue but wouldn't we be able to get sd read/write support just via the methods he has?
13:24.35ginge_yeah, assuming we can compensate for all the shifting and bitwise ops ont he clocks
13:25.00ginge_shoudn't be too  hard now premy has some traces of that area
13:25.20Marajinginge_: Well here's hopin' :)
13:25.49ginge_yeah, I am struggling to find time to play at the moment, so here's hopin' Premy has time :)
13:26.07Marajinwell it's not end of the world mate :p
13:26.56ginge_still, it bugs me that we are so close and I have to waste half of my life sleeping ;)
13:27.11Marajinhah :P
13:27.23premyginge_: my daughter just woke up, actually, so I'll have to go. Will catch up later.
13:27.27Marajinyou don't but you may considerably shorten your life by going the alternate route ;)
13:27.35ginge_premy: ok, laters
13:27.57*** part/#htc-linux premy (n=pr@73.205.76-86.rev.gaoland.net)
13:28.21*** join/#htc-linux LunohoD_ (n=alex@e180072203.adsl.alicedsl.de)
13:28.38ginge_Marajin: we need to reverse engineer that Chinese dude who doesn't need sleep :)
13:29.33Marajinginge_: I didn't sleep much up to being ill :) I did 40/8 schedule
13:29.42Marajinand that was fine by m
13:29.42Marajine
13:30.01Marajin..sweet, I got my ADSL stable even on a 3db SNR
13:30.04ginge_I generally scape by with 5-6 hours sleep a night, but that scedule insane
13:30.05Marajin3 days since resync
13:30.31Marajin40/8 was fine for me. I could do 20/4 but I would simply have to get up after the 4
13:31.17ginge_I tried out polyphasic sleeping a few years ago while working from home... that really screws with your oerception of what day it is
13:31.39MarajinI have no perception of what day it is
13:31.46MarajinI really don't know
13:31.57ginge_what yea is this? who's the president
13:32.03Marajinhah :P
13:32.25MarajinTBH I did used to answer questions like "what day is it today?" with "summer, I think"
13:32.47MarajinNow I answer them by just repeating the question to my kaiser
13:32.51Marajinhurray voice command :P
13:33.58ginge_wow, how'd you manage to make any of that stuff work? It takes about 20 minutes for the screen to refresh nevermind preocess voice
13:34.17MarajinO.o
13:34.21Marajindude it just works
13:34.29ginge_opening pocket word takes about 3 minutes to close, rotate the screen takes about 2 minutes etc
13:34.37MarajinI have a wma of me talking to it at some point
13:34.41Marajinit's fairly responsive
13:35.52Marajinyour stock rom must dick with it
13:36.22ginge_probably. Its the main reason I want rid :)
13:36.41Marajinginge_: Why don't you flash it then? warranty concern?
13:36.50ginge_Marajin: laziness
13:36.57MarajinDude it takes like 5 minutes
13:36.59ginge_you can only polish a turd so much
13:37.39Marajinyou can make voice command responsive and the screen rotate down to a second or two
13:38.18Marajinand that's WITH a slider noise and even from standby
13:39.43ginge_I am no stranger to new roms, I upgraded all of my other HTC devices to better roms, but refer to turd comment above for the results ;)
13:40.00ginge_every time there were nasty regressions
13:40.27Marajinwell yeah but you can turn the kaiser from arse gravy for the worst kind to merely a mild constipation
13:40.37ginge_rofl
13:41.09Marajin(hurray for stephen fry and QI)
13:44.13*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
13:44.40Marajinginge_: But really I find the kaiser less annoying with the custom roms and voice command always worked great for me
13:46.21ginge_crazy mofo. A friend of mine is writing a wrapper for windows ce dll files that load into a Linux kernel.
13:47.05Marajin...
13:47.05Marajinhah
13:47.13Marajinthat'd make our lives easier ;)
13:47.29Marajinbut we'll probably have working linux code before he finishes no? :)
13:48.07Marajinthough I'm left wondering how many devices actually WORK as intended under linux yet
13:48.10ginge_probably :)
13:52.13zooloochi there
13:52.55ginge_'lo
13:52.59zooloocginge_: so this friend is writing something like wine?
13:54.24ginge_more like the wrapper ntfs.sys layer that was done a while ago
13:54.33zooloocI was asking a similar question in the #wine channel once, if that is possible, they asked back, "what for?" or similar, and argued they have far too much more important issues than providing wince support
13:54.35ginge_he is using reactos code in the same way
13:54.46dcordeshi
13:55.00ginge_wine is more userland, this is kernel related
13:55.06zooloocoh yeah, reactos borrowed some wine code, afaik
13:55.24zooloocho dcordes
13:55.25Marajinginge_: win4lin ? was't that kernel level?
13:55.34Marajinginge_: An what about lindows or whatever it'scalled now
13:55.44ginge_found it captive.. http://www.jankratochvil.net/project/captive/
13:56.25ginge_those projects are all about emulating the win32 api, this is about wrapping those kernel level device drivers into a real kernel driver in a horrible nasty non-gpl way
13:56.26zooloocI'd be interested in having the ability to still run some wince apps (not available in linux) when someday linux could replace wince on my pda phone
13:56.35zoolooclike navigation software
13:56.53ginge_iirc you can run some with some guys crazy patch set, but I saw that years ago
13:56.57Marajinginge_: There's a linux thin for wrapping network drivers I think, they use it for odd wifi cards and stuff
13:57.06ginge_ndiswrapper
13:57.12Marajinthat's it
13:57.18ginge_thats more like what this guy is doing
13:57.51zooloocwasn't captive the only way to get proper,reliable ntfs write support until recently, by using some windows dll/sys files?
13:57.52Marajinyeah
13:58.36zooloocoh, the link says so, too :P
13:59.34*** join/#htc-linux rob_w|lappi (n=rob@Ma54b.m.pppool.de)
14:02.25ginge_has anyone ever tried loading linux through qemu on a severely bare ce image?
14:02.34ginge_something like colinux
14:06.39zooloocginge_: don't know, but that too sounds like an interesting aproach
14:12.17ginge_zoolooc: it would probably run like shit
14:25.48*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
14:41.45*** join/#htc-linux kiozen (n=oeichler@rgnb-5d86614e.pool.einsundeins.de)
14:55.05*** part/#htc-linux zoolooc (n=lucian@nrbg-4dbfa784.pool.einsundeins.de)
15:06.18ginge_Marajin: you ever go to the Linux conferences up here?
15:12.16Marajinginge_: Nope?
15:16.24*** join/#htc-linux Othello_ (i=Magorium@gateway/tor/x-c16742e7657aa355)
15:25.19*** join/#htc-linux Othello__ (i=Magorium@gateway/tor/x-31e0aa9f650ccabb)
15:30.25kiozenBabelO: beat me, I am too dump to change the application's color in qtopia
15:30.40kiozenBabelO: every file I touch has no effect
15:30.53BabelOkiozen: where do you changes values ?
15:31.26kiozenBabelO: I have tried the following:
15:31.58kiozenBabelO:  /opt/Qtopia/default/Trolltech/qpe.conf
15:32.44BabelOkiozen: etc/themes
15:33.01miknixhey BabelO
15:33.02BabelOand themes you selected
15:33.13kiozenand this one /opt/Qtopia/etc/colors/Qtopia.scheme
15:33.21BabelOkiozen: then there is a decorationrc file with a conf
15:34.13BabelOhi miknix
15:34.37kiozenBabelO: yes, but nothing will change e.g. the black dialog background in M
15:36.31BabelOkiozen: black dialog background is set by you in code if you remember :)
15:37.10miknixBabelO: can you give me a hand with mmc reset on the wizard? I spent all day yesterday resenting it and I failed :(
15:37.15kiozenBabelO: I just do setAutoFillBackground(), no color
15:37.52BabelOkiozen: it change for me when you do that :) before it was transparent
15:38.11BabelOmiknix: i only do my code from nk.exe deasm
15:38.58miknixBabelO: these are dumps when inserting and removing the SD card. all GPIOS and registers being set in your board are being traced
15:39.08miknixBabelO: http://pastebin.com/d13a33d2c
15:39.16kiozenBabelO: well yes, that is what setAutoFillBackground() is supposed to do, but where does it define the background color?
15:39.46miknixBabelO: this is my reset code sofar: http://pastebin.com/d243fe586
15:40.51miknixBabelO: notice the OMAP730_IO_CONF2: C2(10 11)=0 C3(14 15)=0 C4(18 19)=0, it chooses GPIOS 25-30
15:41.19miknixBabelO: and then it puts then into reset, GPIOS GPIO1-IL: IL25(25)=0 IL26(26)=0 IL27(27)=0 IL28(28)=0 IL29(29)=0 IL30(30)=0
15:41.32BabelOmiknix: be carefull, in devices.c file there is already a mmc pin mux setting, not for gpio, but ...
15:42.21miknixBabelO: I know.. but looking at the dumps, he first resets the register and then it configures the mux
15:42.30miknixs/he/it/
15:44.12miknixBabelO: but ok, Yesterday I also dumped my rom. I'm going to try IDA on nk.exe then and see what I can come with
15:44.22*** join/#htc-linux matsuu (n=matsuu@eb56.opt2.point.ne.jp)
15:44.31ginge_cya tomorrow peeps. I hope to come back with some kaiser sd card driver insights ;)
15:44.33BabelOmiknix: yes should be easier this way :)
15:44.51miknixerr.. unless you don't know much about asm :)
15:45.36BabelOkiozen: http://sse2.net/gp/ look here it is a theme made by jott, maybe you can find the difference
15:45.54BabelOkiozen: ans file to change
15:50.12*** part/#htc-linux ginge_ (n=baz@80.229.188.31)
15:51.43*** join/#htc-linux the_sys0p (n=the_sys0@cpe-67-49-214-162.bak.res.rr.com)
15:52.23miknixBabelO: what's nk.exe? the sd card manager or is it something more?
15:54.00BabelOmiknix: it is wince kernel
15:54.08miknixoh
15:54.15BabelOsame in windows xp :) you never see buggy screen ?
15:54.33miknixno windows here :)
15:55.36exco_is there a way to build win executables from haret and txt and kernel?
15:57.08miknixBabelO: there is a OMAP850_.DLL here
15:59.34BabelOexco: yes it is a make linload command line
16:01.09BabelOexco: look in tools/make-bootbundle.py from haret sources
16:01.14exconice ... will save me from start open file browser change to sd haret load ... now if I could get around that initial setup (calibration and the other stuff)
16:02.18*** join/#htc-linux MrBIOS_ (n=aperez@c-76-102-236-252.hsd1.ca.comcast.net)
16:05.59*** join/#htc-linux pH5 (n=ph5@e178210181.adsl.alicedsl.de)
16:09.00kiozencr2: looks like the usb cable does not charge the battery in linux
16:10.00kiozencr2: if that is true the device operates for ~6 h with a fully charged battery and moderate lcd usage
16:10.38kiozencr2: that is not too bad :) but maybe we get it better...
16:11.41pH5kiozen: sounds about right. can you trace how the charger ic is controlled in wince? on bq24022 devices like magician or blueangel it was just a gpio to enable charging.
16:12.09kiozenpH5: cr2 is working on it :)
16:13.18pH5nice. are you using the pxa27x-voltage-max158xx driver already to scale down the core voltage?
16:13.18exco_goes for his weekly TopGear fix
16:13.22pH5heh
16:15.14kiozenpH5: you know the string of the kernel switch for that?
16:15.46kiozen# CONFIG_PXA27x_VOLTAGE is not set
16:16.22pH5kiozen: that one and PXA27x_VOLTAGE_MAX158xx
16:17.00kiozenpH5: hm, it's not in the defconfig ...
16:18.22pH5kiozen: one depends on the other, so it doesn't appear if _VOLTAGE is disabled
16:18.35kiozenpH5: you think it would work on the Loox n560 ?
16:18.59pH5kiozen: if the looks is a standard pxa27x design, it should
16:19.25pH5loox*
16:19.40kiozenpH5: ok i try to compile it into the kernel
16:19.56kiozenpH5: how can I test it?
16:20.38pH5kiozen: enable cpufreq, change the frequency. if it doesn't hang, chances are it works
16:20.52pH5unfortunately the pmic can only set the voltage, not return the current value.
16:21.21pH5so you'd have to measure current draw from the battery, core voltage or something like that to see whether it really works.
16:21.53kiozenpH5: ok for that cr2 has to fix the battery monitor
16:22.16pH5what battery monitor chip is n560 using?
16:24.43*** join/#htc-linux danyduck (n=danyduck@adsl-ull-104-165.49-151.net24.it)
16:25.46kiozenpH5: ads7846
16:26.09pH5the main battery? so there's no coulomb counter on this device?
16:26.45kiozenpH5: yes, looks like, however wince has some information about the current
16:27.06kiozenbut I thing cr2 can tell you more
16:27.18pH5kiozen: right, if there's something else, cr2 will find out :)
16:27.29kiozenlol
16:27.39pH5downloads the magician service manual and starts taking apart the broken device
16:27.56kiozenok compiled that pxa27x_voltage_max158xx thing
16:34.35kiozenpH5: that make the kernel crashing :)
16:36.13pH5kiozen: hm, that doesn't sound like success
16:36.41kiozenwell let's decide on point even
16:39.02BabelOhi pH5
16:39.14BabelOpH5: did you see that exco_ find the jtag for magician ?
16:39.27*** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
16:39.59pH5~lart hooks all over the place
16:39.59aptgets a hotmal account and SPAMs hooks all over the place
16:40.25pH5BabelO: yes, I'm just en tour d'opening the broken one up
16:40.38pH5but I'm still in need of a jtag debug board
16:47.54pH5ok, the openmoko guitar pick officially works on magician, too
16:51.32*** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes)
16:53.29cr2pH5: n560 does not use max15xx, it switches 3 pxa gpios on frequency change
16:54.38cr2kiozen: does the AC charge battery in linux ? i've 2 more battery related pxa gpios and one cpld gpio not used by linux now.
16:55.49miknixBabelO: I found the mmc reset section on nk.exe
16:55.55*** join/#htc-linux Bendy (n=sniveri@bb-81-175-215-228.dsl.phnet.fi)
16:56.34miknixBabelO: do you want to have a look at it?
16:58.53pH5cr2: oh, ok. do high-res naked pictures exist?
16:59.03pH5it's amazing how much they work with adhesive tape inside those devices
17:01.06BabelOmiknix: just check it with my code :)
17:01.08cr2pH5: it's not my device, so i#ve not opened it :) it looks line the c550 people opened the device, but did not make pics
17:01.25cr2pH5: http://www.handhelds.org/moin/moin.cgi/LooxC550Hardware
17:01.57*** join/#htc-linux Marex (n=vasum7am@auryn.karlin.mff.cuni.cz)
17:02.18miknixBabelO: the procedure is somewhat different from yours
17:10.46*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
17:23.08excopH5, isn't it possible to push your changes upstream for OpenMoko?
17:24.04excoOM is intended to work on multiple platforms, isn't it?
17:24.10pH5exco: what changes? I'm working on the kernel, not on openmoko.
17:24.19pH5yes, it basically is.
17:24.58exconut you also modified some parts to get the dialer... to work
17:24.58*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
17:25.17exco*but
17:25.31pH5that dialer doesn't exist anymore
17:25.39pH5even the replacement is abandoned by om already
17:26.54pH5there was the 2007 image, the 2008 image, now the ASU(qtopia) and the FSO(e) things. the 2007 image is what I got to work, the 2008 image is what ships one the freerunners, but they don't work on it anymore afaict.
17:27.14excoso one needs to adjust every new version of certain apps for certain platforms?
17:27.44pH5exco: until they start doing the middleware right.
17:28.15pH5the freesmartphone.org parts in the fso image could be it, but I'm not sure how everything will develop.
17:29.24excothat sounds awfully timeconsuming and most work is ... for nothing?
17:29.54pH5not if you do it right
17:30.10pH5as I said, I didn't really work on integrating with openmoko userspace yet.
17:30.22pH5(or any other userspace image, for that matter)
17:30.45excowhat experience does one need to do that?
17:32.18excobtw sound works with the May image + opkg update (+paying attention)
17:32.43pH5a bit of aptitude for programming languages (for FSO knowledge of python would be good, I guess, for qtopia c++), and knowledge about what AT commands the magician needs/supports (it's all in the xda-developers wiki).
17:33.13pH5:) great
17:34.53pH5exco: now that I opened it up I fear I'm too clumsy to solder the jtag connections. first of all, I'll need a smaller iron :)
17:35.19excosmd one would be best
17:36.58excoI once bound a needle with copper wire to a soldering bolt and changed the M35i to blue
17:37.09exco:-)
17:37.22AstainHellbringdamn that sounds scary
17:38.05excowell I was frightened ... but then again I don't have too much of a steady hand
17:45.00BsDaEmRis there a way to convert a key.pem and a crt.pem into a combined der file?
17:45.42BsDaEmRoops wsrong chan
17:58.12excohow is the top bar called in OM (and what if there's just a grey/blue line saying registering...)
17:59.07*** join/#htc-linux TimRiker (n=timr@72-58-213-49.area1.spcsdns.net)
18:06.37pH5exco: that's not the openmoko bar, that's the matchbox window manager's title bar. somehow the theme isn't set up correctly.
18:08.09excoI'll look into that
18:12.28excopH5 did you ever patch the error messages to fit to qvga
18:13.38*** join/#htc-linux Marex (n=vasum7am@auryn.karlin.mff.cuni.cz)
18:19.20*** join/#htc-linux ewon (n=ewan@skynet.skynet.ie)
18:19.55*** join/#htc-linux premy (n=pr@73.205.76-86.rev.gaoland.net)
18:20.37*** join/#htc-linux zwox (n=zwox@unaffiliated/zwox)
18:24.24*** join/#htc-linux JohnnyK (n=johnnyk@nfx-nat-229.pilsfree.net)
18:24.28*** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfa784.pool.einsundeins.de)
18:33.04pH5exco: no, I don't even know if the sizes are hardcoded in libnotify or if it can be changed with the gtk theme.
18:34.23excoonce I hab
18:34.54excohave 2 hands again I'll google like crazy all those things...
18:42.06*** join/#htc-linux apt (i=ibot@pdpc/supporter/active/TimRiker/bot/apt)
18:42.06*** topic/#htc-linux is HTC Linux Channel: Find logs at http://apt.rikers.org/%23htc-linux/ | please check http://handhelds.org/moin/moin.cgi/HTC_2dPhones | http://wiki.xda-developers.com/index.php?pagename=Xanadux | <cr2> let's define a common setup.
18:44.41*** join/#htc-linux |miknix| (n=kvirc@81.193.87.168)
18:44.57*** join/#htc-linux OpnSrc (n=harisha1@12.47.48.5)
19:13.35kiozencr2: would opening the device still help? I will rug it up anyway... thus opening it is not too evil
19:16.06*** part/#htc-linux OpnSrc (n=harisha1@12.47.48.5)
19:18.40kiozencr2: don't think ac does load it in linux, think I remember when the battery was drained nothing helped but using the ac and wince, but I will check the next time..
19:19.45kiozenhm, think my tools aren't appropiate for thoses micro trox screws :/
19:21.57pH5kiozen: small screwdrivers sometimes do the trick.
19:23.15kiozenpH5: I tried already, mine are too worn out :) but I have to phone andress-industries the next days. maybe the have pictures they would share. The open these units on a dayly base.
19:23.44kiozenis hocked by his syntax..
19:23.49kiozengrr
19:24.30pH5heh, I like that line of thinking though
19:27.56*** join/#htc-linux tsdogs (n=tsdogs@62.123.180.130)
19:29.09*** join/#htc-linux OpnSrc (n=harisha1@12.47.48.5)
19:30.24*** part/#htc-linux OpnSrc (n=harisha1@12.47.48.5)
19:31.11*** join/#htc-linux OpnSrc1 (n=harisha1@12.47.48.5)
19:45.04cr2kiozen: added 3 gpios to wiki
19:45.25cr2kiozen: the blue key led is on the cpld.
19:45.48kiozenbah, only draws current..
19:46.21cr2kiozen: i think the others (nk?) are on the cpld too. nleddrv.dll doesn#t have any hardware data.
19:47.10cr2kiozen: you may try to trace this cpld bank, while using the 'NLED*' command. i think you#d see something for the right led.
19:47.19excowhere can I find the Magician rndis inf?
19:47.54kiozencr2: ok, the ones on the top are more important
19:47.56cr2kiozen: another interesting thing to trace is the cpld/gpios at the moment when the battery is charged in full
19:48.27cr2kiozen: i'm trying to document the MCR things now for resume.
19:49.07kiozenok, have to charge the device
19:49.10cr2because i#ve tracked all the battery stuff for now, but don#t have a clear picture what to patch in the kernel yet.
19:50.03kiozencr2:  btw didn't you say leds are in userspace already?
19:54.49cr2kiozen: the blue kbd led. but then i gave up on this idea.
19:54.59kiozenok :)
19:55.51kiozenI need those to show that the deivce is recording, and if it is chargeing
19:56.14kiozenstarted the track recording today.
19:56.46cr2i think they are enabled by cpld gpios. and these can be easily exported to userspace.
19:57.28kiozen<PROTECTED>
19:57.40cr2ok
19:59.42kiozencr2: is it possible to switch off gps from user space?
20:00.01kiozencr2: and bluetooth
20:00.16kiozencr2: in normal operation only one is needed
20:01.00cr2kiozen: the power is hooked to open()
20:01.19cr2kiozen: i thought it's a good idea.
20:01.50cr2pH5: what do you think ? it's maybe the only reason why i'm not using 2.6.26
20:02.21kiozencr2: for the sirfIII for sure, but what about bt, sounds like I have to unload bt stuff if I do not use it, right?
20:02.51cr2kiozen: once you call close() the power is switched off.
20:03.08cr2kiozen: there is no difference between handling gps and bt now.
20:03.40kiozencr2: but if I start bt on runlevel, something will open bt to scan for devices, right?
20:05.26cr2hciattach
20:05.35cr2first
20:05.46cr2then the daemons can scan.
20:06.21cr2it must be handled for the universal & others already. somewhere in these bt start script.
20:07.38kiozenyes, have to look later, device is chargeing...
20:09.32cr2ok
20:11.04pH5cr2: until somebody comes with a better idea, continuing to use open/close is the best we can do.
20:11.07pH5ah, the merge window opened
20:12.31cr2:)
20:12.55cr2pH5: kiozen wants to have more explicit control of the power.
20:13.43cr2pH5: imho, it's easy to modify magician's code to make n560 work, but i think he will lose gps and bt.
20:14.13kiozenwell open close is fine as long as no dump deamon keeps the device open for nothing
20:16.35pH5kiozen: for bt it makes sense to power it off once hciattach is closed, because it would just uselessly draw power otherwise. for gps it's maybe a bit different because you might want to power it to keep a lock.
20:16.44cr2kiozen: for bt it may be the case.
20:17.15*** join/#htc-linux zwox (n=zwox@unaffiliated/zwox)
20:17.48kiozenthink I do a "bluetooth stop" while using the gps.
20:18.04kiozenthe user won't change it all the time anyway.
20:19.06kiozencr2: btw did you find out how wince can measure the current?
20:19.50pH5cr2: there's a patch for the serial power hooks, too. it just has to be adapted.
20:20.13cr2kiozen: i see it switching 2 channels. vbat & temp1 probably. the rest is not so obvious.
20:20.24pH5hm, and git.linuxtogo is down
20:20.36cr2pH5: need a working magician tree...
20:20.41pH5cr2: do you think it's measuring current as voltage drop over some probing resistor?
20:22.23cr2yes
20:22.37cr2as kiozen says there is no other way.
20:24.18*** join/#htc-linux Bendy (n=sniveri@bb-81-175-215-228.dsl.phnet.fi)
20:24.23pH5interesting. hopefully the charging logic isn't done in software.
20:26.14pH5cr2: you'll get a working magician tree.
20:26.42pH5actually, I'll regularly push my local work branch to git.linuxtogo.org
20:30.08*** join/#htc-linux pgupta (n=pgupta@93-96-26-19.zone4.bethere.co.uk)
20:31.05pguptahi guys, i'm trying to locate a linux version to run on my htc artemis ...
20:31.20pguptaxda-developers.com has a reference to http://linuxtogo.org/~htcpxa/htcartemis/qtopia/
20:31.30pguptabut that page does not open :-(
20:31.40pguptadoes anyone know of a mirror?
20:31.57kiozenthat charging take too long, cr2 I do the trace at  work, tomorrow. bbl
20:33.22pguptaanyone?
20:33.58pguptaany pointers on how I can search for linux for my htc device?
20:34.06*** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes)
20:34.15pguptathe main linuxtogo web-page is not accessible :-)
20:34.17pgupta:(
20:34.28pguptahttp://linuxtogo.org/
20:34.31pgupta'elp
20:36.22dcordesoh no
20:37.35pguptahi dcordes .. my problem is i'm trying to get linux to run on my htc artemis, but the only URL (http://linuxtogo.org/~htcpxa/htcartemis/qtopia/) is inaccessible :-(
20:38.14dcordespgupta: try again later. it doesn't take long normally
20:39.10pguptaoh, ok. thanks dcordes.
20:39.16*** part/#htc-linux premy (n=pr@73.205.76-86.rev.gaoland.net)
20:39.26cr2pH5: ok. time to port athena and n560 to newer kernel version
20:52.18Marexcr2: and write asoc ;)
21:00.16*** join/#htc-linux zwox (n=zwox@unaffiliated/zwox)
21:02.35cr2pH5: hehe. sd-to-flash
21:04.38*** join/#htc-linux hagisbasheruk (n=hagisbas@cpc4-clyd1-0-0-cust357.renf.cable.ntl.com)
21:07.50*** join/#htc-linux pronto (n=pronto@pool-71-179-232-167.bltmmd.fios.verizon.net)
21:12.20*** join/#htc-linux OpnSrc1 (n=harisha1@12.47.48.5)
21:28.55*** join/#htc-linux ginge__ (n=ginge__@cpc6-ward3-0-0-cust1008.manc.cable.ntl.com)
21:30.35*** part/#htc-linux ginge__ (n=ginge__@cpc6-ward3-0-0-cust1008.manc.cable.ntl.com)
21:31.53cr2http://cmptv.net/TearDownTV/index.php?videoid=14
21:33.11*** join/#htc-linux premy (n=pr@73.205.76-86.rev.gaoland.net)
21:35.02*** join/#htc-linux gerard (n=gerard@s55932332.adsl.wanadoo.nl)
21:36.21cr2lol
21:36.24cr2patented space-filling technology
21:37.35dcordes*click
21:39.05dcordesthe built in fm transmitter sounds useful
21:40.27cr2kiozen: fyi http://cegt201.bradley.edu/projects/proj2008/gps/Deliverables/Functional Description.pdf
21:40.28dcordeshe has got 2 gigabits of nand?
21:40.36cr24Gb
21:40.54cr2mine has 4
21:41.01*** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde)
21:41.45dcordescr2: where does garmin provide the kernel source?
21:42.08*** join/#htc-linux dzo (n=dzo@121.98.128.127)
21:42.12cr2developer.garmin.com
21:42.16cr2http://www.handhelds.org/moin/moin.cgi/GoPalP4225
21:42.57cr2http://developer.garmin.com/linux/
21:43.54dcordesthanks
21:44.28cr2oh
21:44.58cr2ALC5610
21:45.49cr2ac97+touch. hmm. not wm9712...
21:49.13cr2Marex: new chip for you :)
22:02.14dcordescr2: do you have any idea what could cause niki to not boot kaiser kernel (any longer) ?
22:02.47dcordesfor effbiai it freezes in haret. already tried a kernel with smd disabled
22:15.16cr2no
22:20.51BabelOcr2: zaprudnia in radio morse :)
22:23.14cr2BabelO: ?
22:23.38cr2BabelO: i have not found the TMC in the medion device ;)
22:24.13BabelOcr2: just looking at ham radio CW band, and heard someone from zaprudnia.
22:24.27BabelOcr2: it support TMC ?  you have radio ?
22:28.42cr2yes, in wince.
22:29.11cr2uses an sd-to-flash chip to access flash as an sd chip. weird.
22:29.18BabelOcr2: on artemis no tmc under wince :( you have same chipset than me or it is a TDAxxxx
22:29.49cr2BabelO: i have not found the chip.
22:29.55cr2btw http://cegt201.bradley.edu/projects/proj2008/gps
22:30.19cr2it uses something like that. and mostly the same hardware
22:31.03*** join/#htc-linux AstainZZZZZZ (n=Administ@unaffiliated/astainhellbring)
22:31.43BabelOcr2: nice project
22:34.15BabelOcr2: i m looking for a DGPS signal to receive ...
22:35.33*** join/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes)
22:37.24premyginge: ping
22:41.21cr2BabelO: strange.
22:41.33cr2don't you have a sea port close to you ?
22:41.41Marajinpremy: I think Ginge is out with the missis
22:42.42BabelOcr2: http://www.effective-solutions.co.uk/beacons.html
22:45.46BabelOcr2: or this one for you http://www.wsv.de/fvt/dgps/index.html
22:48.56cr2BabelO: yes, i need a receive beacon too :)
22:49.21BabelOlol, and a VLF antenna :)
22:49.39BabelOi have the software decoder
22:50.22cr2why VLF ?
22:50.27cr2ah 300KHz
22:50.31BabelOyes
22:50.50BabelOyou can make a receiver from an old radio maybe
22:51.18cr2i have a sony icf somewhere
22:51.24BabelObut power is not the same, 200 W vs 1Kw
22:51.32cr2but don#t think it has a digital out.
22:52.16cr2afair it covers 180K - 30M
22:53.06BabelOah, good , sound card output is enough
22:53.45cr2no low frequencies ?
22:54.32cr2i think this one -> http://www.amazon.de/Sony-ICF-SW-7600-Weltempfänger-silber/dp/B00005JY3Q
22:54.52*** join/#htc-linux zwox (i=zwox@unaffiliated/zwox)
22:56.57cr2yes, 150K-30M with 1kHz step
22:57.12BabelOcr2: so maybe you can receive signal
22:58.26cr2how do i decode it ?
22:59.57BabelOcr2: this one do it http://f6cte.free.fr/index_anglais.htm
23:00.10BabelOcr2: there is also multimode
23:03.06cr2BabelO: ok.
23:07.42cr2good night
23:08.09BabelOgood night
23:22.49*** part/#htc-linux premy (n=pr@73.205.76-86.rev.gaoland.net)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.