IRC log for #htc-linux on 20071208

00:01.31*** join/#htc-linux Zoolooc (n=fredsiba@p54952555.dip0.t-ipconnect.de)
00:04.51*** join/#htc-linux echowip (n=bochet@dialup-4.245.118.236.Dial1.SanJose1.Level3.net)
00:32.30*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
00:38.51*** join/#htc-linux ltxda (n=ltxda@c-98-196-125-205.hsd1.tx.comcast.net)
01:46.51dcordes_cr2_: still up?
01:57.40*** join/#htc-linux Stamrogh (n=starman@c83-254-132-174.bredband.comhem.se)
02:00.56*** join/#htc-linux TimRiker (n=timr@rikers.org)
02:16.58Kevin2Hi
02:21.08dcordes_hey Kevin2
02:27.17dcordes_Kevin2: http://pastebin.ca/808967 could you take a look at this? It's watch gpio output from keyboard
02:30.12dcordes_I hit all keyboard keys there
02:30.17dcordes_is it any useful?
02:37.59Kevin2dcordes_: You need to get a newer version of haretconsole - try http://www.handhelds.org/~koconnor/haret/haretconsole-20071207.tar.gz
02:38.29dcordes_Kevin2: ok thanks
02:38.36dcordes_also I think I was watching the wrong gpios there
02:38.55dcordes_cr2_: told me I had to replace 9 with b
02:38.57Kevin2Also, you're going to need to put together a document that maps GPIOs to their functions.  This means taking careful notes and then putting the notes together on a wiki page.
02:39.13dcordes_yes the keyboard thing is a first start
02:39.30Kevin2dcordes_: No cr2_ was wrong.  Don't mess with the registers.  Just run "watch gpios x" and record what the functions do.
02:39.47dcordes_ah ok :)
02:51.59dcordes_Kevin2: new haretconsole installed. Am I good to go with the last kaiser revision you gave me?
02:52.13dcordes_#4 or so from 12-1
02:55.16Kevin2dcordes_: Yes.
03:05.07dcordes_Kevin2: I don't get spamming gpios with the console version you gave me
03:05.19dcordes_is it supposed to be that way?
03:11.52dcordes_Kevin2: ibits were still active because I did not restart haret. I was assuming a new console session would overrun the old "settings"
03:13.26Kevin2dcordes_: No, ibit settings stick around until you restart haret.
03:13.37Kevin2You should now see names instead of the "?".
03:15.38dcordes_Kevin2: no ? in the watch gpios outputs
03:15.56dcordes_ibit the spamming ones again now?
03:17.05Kevin2dcordes_:  I've been writing some haret docs.  You might find http://www.handhelds.org/moin/moin.cgi/HaRET_20Documentation useful.
03:22.42dcordes_Kevin2: my watch gpios x output does not show anything like 002.269 GPIOS   GPGDAT: GPG3(195)=0
03:22.55dcordes_but rather like this 018.514 GPIOS  92b00814(401 402)=1fe3f09
03:25.15dcordes_Kevin2: http://pastebin.ca/809004 oney keyboard keystroke
03:25.23dcordes_one
03:26.00dcordes_could you give me an example on how the refering Kaiser_GPIO wiki entry would look like?
03:28.19Kevin2dcordes_: Hrmm.  Can you add "import regs_msm" to the file haretconsole/memalias.py right after the line that says "import regs_omap"?
03:28.32dcordes_second
03:29.38Kevin2http://wiki.xda-developers.com/index.php?pagename=HermesGPIO is an example wiki
03:29.58dcordes_ok done the deed
03:30.13dcordes_Kevin2: I already took a look on the hermes gpio page
03:30.22dcordes_but I don't understand the parallels to my output
03:30.24Kevin2Now, restart your console and rerun your "watch gpios"
03:30.30dcordes_also restart haret?
03:31.10Kevin2No, you don't need to restart haret.
03:32.03dcordes_http://pastebin.ca/809007
03:33.11dcordes_I actually did nothing there :)
03:33.22dcordes_http://pastebin.ca/809010
03:33.41dcordes_this is the same as http://pastebin.ca/809004 with the import kaiser line now
03:35.18dcordes_what's all the in out stuffs?
03:37.43Kevin2dcordes_: Look at haretconsole/regs_msm.py
03:38.15Kevin2It's just our guess of what the register names are.
03:38.38Kevin2The google android source might have better register names.
03:38.45dcordes_ah ok cr2_ told me he looked up register stuff in the msm tree. Was wondering what he meant
03:48.12dcordes_Kevin2: http://wiki.xda-developers.com/index.php?pagename=Kaiser_GPIO
03:52.44Kevin2dcordes_: Yep - that is what you want to do.  (Note, you don't need to report =0/=1 -- only 0/1 is possible, and the value should be obvious based on the description.)
03:53.25Kevin2Also, there is no reason to show the number in parenthesis -- for example (278) -- this number is internal to haret and has no meaning outside of haret.
03:53.49dcordes_lol what's left then?
03:54.43Kevin2Map all GPIOs and IRQs.  Find out what every key button press does.  Find out what the touch screen does, etc.  See the reference to the docs I posted above.
03:56.45dcordes_ok will try my very best
03:58.08dcordes_Kevin2: just look at the voice button blob
03:58.34dcordes_http://wiki.xda-developers.com/index.php?pagename=Kaiser_GPIO
03:58.56dcordes_I think it is because the microphone is accessed when I press it
03:59.07dcordes_it calls a voice dial function on my rom
03:59.20dcordes_same effect with photo button
03:59.47dcordes_is there a way to have the phone in a mode where it does nothing on input but let haret give me the gpio activity?
04:04.02Kevin2dcordes_: Putting a trace of what happens during each event is not going to help anyone.  You need to figure out what each gpio does and then post a table of "gpio : description".
04:04.39Kevin2As for the phone - you can try putting the device into "flight mode" so the phone doesn't do anything.  You can also try to figure out what sound does, document those gpios, and then ignore them.
04:04.45dcordes_Kevin2: that's why I was asking how to disable functions which are called when I press such button so I have the raw button press
04:06.19dcordes_flight mode is not a good thing since my connection is wifi
04:11.51dcordes_ah key lock is a place to start :D
04:23.46dcordes_Kevin2: hm in keylock mode i also get so much output on button press
04:23.53dcordes_or in some cases nothing at all
04:29.01dcordes_Kevin2: How can I make a new wiki page?
04:29.58Kevin2dcordes_: You can just modify the URL so that it goes to the page that you want to create.
04:33.57dcordes_Kevin2: when I move around the touchscreen there is IRQ19 activity. After I stop moving it stopps at IRQ16
04:34.05dcordes_what am I supposed to put in the map now?
04:35.21Kevin2dcordes_: You're going to have to figure it out.  I don't have the device, so I can't tell you what it should say.
04:35.32dcordes_I mean what's the normal behaviour with other htcs?
04:35.43dcordes_do you also get two irq values while using the touchscreen?
04:36.09Kevin2It varies.  On Apache I get an irq on touch only (nothing on release).
04:36.40Kevin2On Hermes I get irqs repeatedly while touching the screen.  Irqs stop when I lift the pen.
04:36.45dcordes_so I can assume 16 is the release and 19 is the touch?
04:36.57dcordes_Kevin2: I get the hermes behaviour here
04:36.58Kevin2That would be odd, but not impossible.
04:37.16dcordes_It is printing wild while moving around with the pen
04:37.32dcordes_varying between 16 and 19 w/o releases between
04:37.54dcordes_I just hold down and the output is like 16 16 16 16 19 19 16 19 16 just random
04:38.12Kevin2Okay
04:38.39dcordes_ah wait
04:38.48Kevin2The touchscreen sensing may be separate from the main msm chip.  So, you might be seeing a pendown irq, followed by a whole series of external chip "have more data" irqs.
04:39.04dcordes_when I try to tap it calmly, as in I don't move at all, just tap, it is only printing 16
04:39.13Kevin2You might also be seeing a "screen redraw" irq.
04:40.14dcordes_I will put it as 16= tap 19=move for now
04:41.45dcordes_Kevin2: oops there is also 16=19=1 activity when I close/open the slide
04:42.02dcordes_so propably all I see when I use TS is display related (redraw) ?
04:42.22Kevin2possibly
04:44.17dcordes_I also see 16 19
04:44.25dcordes_when a cursor is blinking or so
04:45.12Kevin2Okay - that is a good indication it is purely screen related.
04:45.33dcordes_How can I name this?
04:45.41dcordes_28=keyboard related
04:49.00dcordes_http://wiki.xda-developers.com/index.php?pagename=Kaiser_IRQ
04:53.44dcordes_Kevin2: how do I cancel a trace (in case I set a huge time value)=
04:53.47dcordes_?
04:59.31Kevin2dcordes_: Ughh.  You can't.  :-(
05:02.30dcordes_Kevin2: look is not that great? http://wiki.xda-developers.com/index.php?pagename=Kaiser_IRQ
05:02.38dcordes_how do I make it a neat table?
05:07.48Kevin2edit the htc hermes table and look at the page source.
05:08.22dcordes_ok
05:10.35*** join/#htc-linux thinko (i=jdoe6alp@smaug.rackdragon.com)
05:10.55dcordes_Kevin2: does not work for me can you take a look?
05:13.01rakeem_Hey dcordes, Kevin...
05:13.12dcordes_hi rakeem
05:13.31dcordes_what's up in taiwan? :)
05:14.48rakeem_Any idea what I might try to stop my BA thinking that every keypress is stuck?  Everything seems fine, but local use is impossible as a result...
05:16.10rakeem_...or do you know how I can download sources for 2.6.16 ?
05:17.10*** join/#htc-linux jgw (n=jgw@unaffiliated/jgw)
05:21.01rakeem_I'm in Hong Kong actually, a minor difference 'cos we're all gonna be China sooner or later....
05:22.48dcordesrakeem_: ah sorry. Where in hk are you? kowloon?
05:23.22rakeem_Hong Kong side, Quarry Bay...  You been here  before?
05:23.51dcordesno only detailed the area in school once
05:27.33Kevin2dcordes: Select/unselect "old style formatting".  Or, read up on the formatting by hitting the TextFormattingRules link
05:30.31dcordesKevin2: I get a table now with old markup, but the columns don't fit
05:30.39dcordeshttp://wiki.xda-developers.com/index.php?pagename=Kaiser_IRQ look
05:40.22dcordeshttp://wiki.xda-developers.com/index.php?pagename=Kaiser_IRQ
05:40.41dcordessorry this wasn't supposed to go here
07:44.34dcordesKevin2: still around?
07:52.28*** join/#htc-linux echowip (n=bochet@dialup-4.246.141.77.Dial1.SanJose1.Level3.net)
09:00.16*** join/#htc-linux the_sys0p` (i=the_sys0@gateway/tor/x-5a5a2ddc4ffebeeb)
09:15.45*** join/#htc-linux rob_w (n=bob@M83fd.m.pppool.de)
09:21.15*** part/#htc-linux NAiL (n=repvik@nslu2-linux/pdpc.active.NAiL)
09:33.21dcordescr2_: you said you told me the GPIO for wifi and something else. Can you put them http://wiki.xda-developers.com/index.php?pagename=Kaiser_GPIO here?
10:02.52*** join/#htc-linux Zoolooc (n=fredsiba@p54951E11.dip0.t-ipconnect.de)
10:34.59*** join/#htc-linux rob_w (n=bob@M83fd.m.pppool.de)
11:20.42*** join/#htc-linux pH5 (n=ph5@p5485D3DA.dip.t-dialin.net)
11:22.34*** join/#htc-linux Zoolooc (n=fredsiba@p54951752.dip0.t-ipconnect.de)
11:36.14*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
11:45.20*** part/#htc-linux Keizer (n=keizer@c-69-139-42-47.hsd1.fl.comcast.net)
12:03.23*** join/#htc-linux Marex-notebook (n=marex@gwfm10-3-250.802.cz)
12:05.48*** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
12:36.09*** join/#htc-linux Marex-notebook (n=marex@gwfm10-3-250.802.cz)
12:48.42jeansebhello , where are last  blueangel kernel and images ?
12:49.03BabelOhello jeanseb
12:49.31BabelOjeanseb: kernel is at hh.org and images at http://linuxtogo.org/~htcpxa
12:49.41BabelOi ve started some new build
12:49.54jeansebthanks
12:50.56jeansebon my MPV M2000 the one i've found was send panic when hit the arrow betwenred and green  phone keys
12:52.58BabelOjeanseb: you can try OPIE II image and gpe too, should works, gpe is quite older now
12:53.18jeansebgpe is gnome based right ?
12:53.54BabelOthere is both, but  gpephone is not stable enough, i build it before summer
12:55.29jeansebok OPIE is qt based .
12:55.39BabelOyes
12:56.11BabelOnext step is to build latest openmoko :)
12:56.11jeansebso i'd rather use GPE. I don't use GPRS but WIFI phone
12:56.40jeansebmy next step is to make linux run on the phone
12:56.41jeanseblol
12:56.48jeansebconsol is OK
12:57.08jeansebold build was having white screen and panic
12:57.34BabelOjeanseb: ok,  GPRS works with a ppp script and wifi works perfect with pcmcia control ;), but i advice you tobuild your own image
12:57.39jeansebwill try new one and may be help for building GPIE Image u pto date after
12:57.54BabelOjeanseb: from where is your spv m2000 france ?
12:58.09jeansebqtek 9090 package by orange provider
12:58.21BabelOok lie me :)
12:58.24*** join/#htc-linux Marex-notebook (n=marex@gwfm10-3-250.802.cz)
12:58.26jeansebhttp://www.handhelds.org/moin/moin.cgi/BlueAngel
12:58.27BabelOs/lie/like/
12:58.43BabelOjeanseb: yes it is web page i change ;)
12:58.50jeansebBabelO, pourquoi on cause anglais ?
12:59.01jeanseben plus t chez free
12:59.05jeansebcomme moi
12:59.09BabelOjeanseb: parce que ici c'est anglais only
12:59.10BabelOlol
12:59.14BabelOen pv
12:59.31jeansebor englisj as u want
13:16.11*** join/#htc-linux resin (i=resin@124.125.212.98)
13:17.13resinHello, not a linux problem though... But have some trouble with my QTEK pocket pc(rom), could anyone help?
13:17.40resin<PROTECTED>
13:17.40resin<resin> I had a some prob before when i had tried to upgrade Phorpeth IVVV, but i used HELMI rom to repair and overcome the problem...
13:18.06resinI had a some prob before when i had tried to upgrade Phorpeth IVVV, but i used HELMI rom to repair and overcome the problem...
13:18.40resinbut this time i am not able to recover it... can anyone try to help?
13:18.43resinthanks.
13:38.26BabelOresin: afaik it is only a linux channel :(
13:45.35BabelOpH5: around ?
13:54.38resinyeah, babel0, i know but are there any other channels for support?
13:54.39resinthanks..
13:55.41BabelOresin: i think only xda forum
13:56.53BabelOresin: look here http://wiki.xda-developers.com/ or http://forum.xda-developers.com/
13:56.53BabelOno other idea
14:01.07resin:) already checked it, nvm, thanks.
14:08.04pH5BabelO: now
14:08.11BabelOyes :)
14:08.29BabelOpH5: i ll try to build new OE image for BA this afternoon
14:08.30Kevin2Hi
14:08.46BabelOwhere is kdrvice-imageon or kdrive-w32200 ?
14:08.51BabelOhi Kevin2
14:09.02pH5hi Kevin2
14:09.16pH5BabelO: great. kdrive imageon is part of xserver-kdrive.
14:09.52pH5but 1.4 doesn't have imageon yet, I think. only w100.
14:09.56pH5bbl
14:09.56BabelOpH5: ok because i don't found the bb file
14:10.03BabelOok
14:10.03Kevin2BabelO - are you merging the hh.org tree with mainline?  If so, please join the mailing list at http://projects.linuxtogo.org/mailman/listinfo/mobile-linux-discuss
14:10.28*** join/#htc-linux Foxdie (i=Foxdie@77-98-48-193.cable.ubr05.harb.blueyonder.co.uk)
14:10.38BabelOKevin2: i have started some part only :) adding BA to it, and adding driver one by one
14:12.03BabelOKevin2: i ve started qgit build :)
14:13.48BabelOKevin2: subscription done
14:15.30Kevin2I've written some haret docs at http://www.handhelds.org/moin/moin.cgi/HaRET_20Documentation
14:26.13BabelObokkmarked, i always forgot some command :)
14:34.41resinIs there anyways i could erase the whole content of my QTEK pu10 and make it brand new??? Even removing the covers would do!!!!
14:35.06resinlike a permanant reset without partitions.
14:42.59BabelOKevin2: you need a nice welcome page ;)
14:43.49*** join/#htc-linux kiozen (n=oeichler@rgnb-4db03ddd.pool.einsundeins.de)
14:56.21Kevin2BabelO: ?
14:57.01BabelOKevin2: yes, project page just contain index.php
14:59.29Kevin2BabelO: Are you volunteering to make one?  :-)
15:00.15BabelOKevin2: lol i m not a good designer, i use joomla actually to do this http://linuxbox2.free.fr/qtapp
15:00.58Kevin2I was thinking of enabling the wiki and pointing the main page to the wiki.
15:01.41BabelOyes, good idea
15:05.02BabelOKevin2: where is your how to make a exe boot for oe images ?
15:15.21Kevin2BabelO: The old link is at http://www.handhelds.org/~koconnor/initramfs-installer/README -- I can't seem to view files from svn at linuxtogo.  The new site is http://projects.linuxtogo.org/projects/distrotestdrive/
15:16.03BabelOKevin2: thx
15:16.17BabelOKevin2: i use that for git work http://digilander.libero.it/mcostalba/
15:17.35Kevin2Hrmm - this link seems to work: http://projects.linuxtogo.org/plugins/scmsvn/viewcvs.php/trunk/README?annotate=2&root=distrotestdrive
15:17.50BabelOok
15:22.10*** join/#htc-linux goxboxlive (n=goxboxli@81.80-202-132.nextgentel.com)
15:30.37cr2_hi
15:31.30cr2_Kevin2: i've read the haret doc wiki page. it's great, but i guess we need more examples
15:31.34BabelOhello
15:32.07cr2_dcordes: you should convert the haret gpio numbers into msm gpio numbers.
15:32.09cr2_hi BabelO
15:32.44cr2_Kevin2: is it possible to do this conversion in a "portable" way ?
15:34.29Kevin2cr2_: Yes - just modify haretconsole/regs_msm.py -- last time this file was modified, we didn't know the "real" gpio names.
15:36.27cr2_Kevin2: ok, i'll have a look. there are 105 gpios on msm7x00 in 5 "banks", so it's nice to have the wince sequential number instead of the bit position.
15:37.12cr2_i'll go to buy some food now, but then will come back and try to test the ATI SD driver :)
15:37.14Kevin2cr2_: 21 gpios per bank?
15:37.15cr2_bbl
15:37.23cr2_Kevin2: no, it's more tricky.
15:37.42Kevin2Soubs like s3c
15:37.56Kevin2s/Soubs/Sounds/
15:38.52cr2_[Sa Sep 15 2007] [16:03:38] <cr2_>      Output_Level0 (0......15) --- Addr=0x%x, Val=0x%x
15:38.53cr2_[Sa Sep 15 2007] [16:03:38] <cr2_>      Output_Level1 (16....42) --- Addr=0x%x, Val=0x%x
15:38.53cr2_[Sa Sep 15 2007] [16:03:38] <cr2_>      Output_Level2 (43....67) --- Addr=0x%x, Val=0x%x
15:38.53cr2_[Sa Sep 15 2007] [16:03:38] <cr2_>      Output_Level3 (68....94) --- Addr=0x%x, Val=0x%x
15:38.53cr2_[Sa Sep 15 2007] [16:03:38] <cr2_>      Output_Level4 (95..105) --- Addr=0x%x, Val=0x%x
15:39.08cr2_these are the gpio numbers.
15:39.42cr2_42-16+1=27
15:40.02cr2_67-43+1=25
15:40.18cr2_94-68+1=27
15:40.28cr2_105-95+1=11
15:40.36cr2_15-0+1=16
15:41.20cr2_not a big surprise, if samsung produces these cpus.
15:41.45cr2_bbl
17:59.26*** join/#htc-linux apt (i=ibot@pdpc/supporter/active/TimRiker/bot/apt)
17:59.26*** 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.
17:59.35Pryanoh no
18:00.33Pryanwithout keyboard the proyect is bad, because i buy athena for his keyboard
18:00.35cr2_Kevin2: the s3c24xx spi ? i think pH5 has written some pxa2xx-spi code for the tsc2046 touchscreen driver
18:01.53cr2_Pryan: it's possible to write a quick and dirty driver, but it will not move you forward a lot.
18:03.04cr2_the spi -connected devices are already a horrible mess in the kernel, because everybody is using his own NIH method to access it.
18:03.43Pryanahm
18:03.45Pryanuf
18:04.41Kevin2cr2_: Wouldn't it be better to map msm gpios via android source then via wince logs?
18:05.47cr2_Kevin2: i think wince is better, because android (1) hides some registers, (2) it's easier to interpret the wince code disasm.
18:05.50*** join/#htc-linux lieter_ (n=lieter@a62-251-100-248.adsl.xs4all.nl)
18:06.23cr2_i'll search for the 6 "alt" gpio functions in the logs.
18:06.32cr2_they are for the SD/wfif
18:06.35Kevin2cr2_: BTW, have you seen the hh.org ported git tree ( http://git.linuxtogo.org ).  I also started a mailing list at: http://projects.linuxtogo.org/mailman/listinfo/mobile-linux-discuss
18:07.27*** join/#htc-linux Zoolooc (n=fredsiba@p54952609.dip0.t-ipconnect.de)
18:07.30cr2_Kevin2: i'll subscribe. but i'm surprised that the changes are so huge. there is a 2.6.23 htc-universal patch for 2.6.23
18:07.34Kevin2It's not clear from the output you posted where the 16 bits in Output_Level0 are - are they the lower 16 bits, every other bit, is each gpio 2 bits, etc.
18:07.59cr2_i think the lower, like it's on the S3C24xx
18:08.46cr2_the 2.6.23 universal patch has asic3 driver, acx100 driver, asic3-leds and pda_power (afair)
18:09.34dcordescr2_: yo
18:09.43dcordeswhat about that conversion thing you asked me to do?
18:09.54Kevin2Well, s3c uses gpioA/B/C/D names - not gpio1-105 names.
18:09.59cr2_dcordes: we are trying to sort it out right now.
18:10.03dcordesah ok
18:10.36dcordescr2_: did you notice duplicates in my irq listing? I was wondering whether this is normal or not
18:11.13cr2_Kevin2: ok, let's check it directly. how do you calculate the haret gpio number on msm ?
18:11.54cr2_dcordes: we don't know anything about irq on msm, so these are just some guesses. the msm7x00 gpios are known much better.
18:12.15cr2_i'm grepping for the msm SD gpios.
18:13.34Pryani have found an asic3 driver for kernel
18:13.41Pryanit's usefull ?
18:14.09cr2_Pryan: athena is not using asic3
18:14.36Pryanok
18:14.48Kevin2cr2_: The number in parenthesis is the haret id - it's just (32 * pos) + bitpos -- I don't think we want to mess with that - instead just change the names as they are listed in haretconsole/regs_msm.py
18:15.09Pryanwhat do you need and i will search it in google ?
18:23.47cr2_Kevin2: ok.
18:24.07cr2_Kevin2: i just need a way to compare these 2 numbers.
18:28.49Kevin2cr2_: Which two numbers?
18:29.17cr2_the 0-105 number, and what you see in the haret output.
18:34.48dcordescr2_: do you need output?
18:37.45Kevin2cr2_: Change regs_msm.py to something like: 0xa9200800: ("out0", [(i, "go%d" % i) for i in range(16)])
18:37.45Kevin20xa9200804: ("out2", [(i-16, "go%d" % i) for i in range(16,43)])
18:37.45Kevin2etc.
18:38.28Kevin2bbl
18:42.24paulproteusBabelO, Ce n'est pas grave si vous parlez francais (-:
18:52.37*** join/#htc-linux rob_w (n=bob@Mb071.m.pppool.de)
18:54.34cr2_paulproteus: the problem is that others will want to write/read in german/russian/spanish/etc. do you know all these languages ?
18:54.42paulproteuscr2_, I'm mostly joking. (-:
18:56.35Stamroghit's the "mostly" that is the scary part
19:03.37cr2_Kevin2: are you the "Mobile-linux-discuss" moderator ? :)
19:04.09BabelOcr2_: do you get git repo ?
19:04.28cr2_BabelO: the new one ?
19:04.54BabelO~x en fr paulproteus : yes and we can use babelfish translator :)
19:05.18BabelOcr2_: yes, i ve cloned it on my harddrive two days ago and started some merge
19:05.41paulproteusHah hah, BabelO. (-:
19:16.38*** join/#htc-linux Zoolooc (n=fredsiba@p54952609.dip0.t-ipconnect.de)
19:24.44cr2_~x en de paulproteus : yes and we can use babelfish translator :)
19:24.52cr2_lol
19:25.45cr2_i still can't believe that the machine translation is so bad ;-)
19:26.10cr2_~x en es paulproteus : yes and we can use babelfish translator :)
19:26.29cr2_~x en ru paulproteus : yes and we can use babelfish translator :)
19:26.31paulproteuscr2_, There's no competition.
19:26.39paulproteusWe've been using Babelfish since like 1995.
19:26.50cr2_that one is even more funny.
19:26.51paulproteusbabelfish.altavista.digital.com!
19:27.22ImCoKeMaNgoogle translate is good sometimes too
19:27.45cr2_paulproteus: still the translation quality didn't change since that time.
19:32.29cr2_btw, the japanese -> english translation of ati functions was not bad.
19:33.03cr2_http://www.sigmarion.net/go/bbs/board.php?bo_table=pds_dev&wr_id=10&page=
19:49.00paulproteusWell that's good to know, at least. (-:
19:54.32dcordescr2_: you managed to get the register conversion in haretconsole done?
19:57.15*** join/#htc-linux Babel (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
20:03.35*** join/#htc-linux echowip (n=bochet@dialup-4.246.126.210.Dial1.SanJose1.Level3.net)
20:28.03dcordespH5: you around?
20:40.03dcordesI wonder if it would be easy to make a small driver for the vibration alert
20:45.41pH5dcordes: hi
20:49.10dcordeshey
20:50.49dcordespH5: did you see my gpio / irq research?
20:51.30cr2_dcordes: i've not found the SD gpios in the log.
20:51.56pH5dcordes: no. is it in the wiki?
20:52.12dcordespH5: http://wiki.xda-developers.com/index.php?pagename=Kaiser_IRQ / GPIO
20:53.18cr2_dcordes: the IRQ stuff is a complete mystery :)
20:53.19dcordescr2_: I will copy haret to internal rom and try getting sd gpios
20:53.44dcordescr2_: what do you mean by mystery?
20:53.44pH5dcordes: ah, very good. are those IRQ or GPIO numbers?
20:54.14dcordesin irq it is irq numbers and in gpio gpio number I hope :D
20:54.52pH5ok, good :)
20:55.01cr2_dcordes: the gpio numbers should be decoded back to the register/bit pair.
20:55.24dcordescr2_: the conversion thing you referred to earlier?
20:55.43cr2_194 - Jog wheel Rotate Up
20:55.46cr2_yes.
20:56.01cr2_194%32 and 194/32 ?
20:56.50dcordescr2_: I was uncertain there which number is assigned to what exact action
20:56.57cr2_awk '{print int($1/32),int($1%32)}'
20:56.59cr2_194
20:57.01cr2_6 2
20:57.07dcordeso_O
20:57.49cr2_so it's the third gpio in "bank" 5
20:58.18cr2_the zero bank has 16 gpios [0..15]
20:58.19dcordesshall I put this on the wiki?
20:58.30cr2_i'll reformat the table.
20:58.34dcordescool thanks
20:58.43dcordesit's not even a table yet ^^
20:59.10cr2_the first bank has 17 gpios [16..42]
20:59.21cr2_and so on. look in today's logs.
20:59.35dcordesI've seen it all
20:59.37dcordeslots of math to me
21:00.18dcordescr2_: did you guys update haretconsole yet with the new info?
21:00.27dcordesso I get the right values directly?P
21:01.39dcordessince when is htc-linux.org linked to the linuxtogo htc images site?
21:02.15cr2_it's linked on a wish of goxboxlive.
21:02.43cr2_actually it's true for magician and universal. to some extent for BA
21:03.39cr2_when i'll make the ATI SD run, i'll upload something for athena there. probably for the sable too. and qtopia4 for hermes.
21:03.53dcordescool. Did not ellisway register this domain while we had some talks about htc-linux.org as a central place of information and discussion for htc-linux development?
21:04.43cr2_dcordes: it's a community effort. if you don't do anything about it, nothing moves forward :)
21:05.12dcordesmy mistake was to stop doing anything after I noticed mediawiki is complicated
21:12.03dcordesI read in the new haret documentation one needs all IRQs. To be honest I have no idea how to go on about them
21:13.28cr2_me too. the only reliable source of information is the SPL and nk.exe
21:17.55dcordescr2_: do you have the spl there? I just managed to extract it.
21:19.15cr2_you can dump it from SRAM
21:19.56dcordesI have it on my desktop.
21:20.26dcordesin several *.nb files and bootslpash in windows bitmap ^^
21:29.33cr2_dcordes: i see only 3 gpios in your list. and 2 of them have different functions, which may be just the matrix keyboard.
21:30.40dcordescr2_: so those buttons are probably connected to CPLD?
21:33.24cr2_no.
21:33.36cr2_the CPLD is a different story.
21:33.47cr2_here is my first version: http://wiki.xda-developers.com/index.php?pagename=Kaiser_GPIO
21:35.06cr2_so check other gpios, and fill the "description" field.
21:35.09dcordescr2_: ok that looks pretty different now
21:35.36dcordescr2_: I'm busy tracing IRQ right now and noticed irq 6 has many different functions
21:35.39cr2_bank is in'N'
21:35.41dcordesis that normal behaviour?
21:35.45cr2_yes.
21:36.01cr2_you'd have inN-M
21:36.11cr2_where N is the bank, and M is the bit.
21:37.05dcordesI mean IRQ not GPIO
21:37.16dcordeshttp://wiki.xda-developers.com/index.php?pagename=Kaiser_IRQ like 6 here. it has so many functions
21:41.29*** join/#htc-linux Robwoerle (n=bob@M8d50.m.pppool.de)
21:45.05cr2_dcordes: updated version http://wiki.xda-developers.com/index.php?pagename=Kaiser_GPIO
21:47.14cr2_found the gpios :)
21:47.32cr2_[Do Nov 22 2007] [01:06:17] <cr2>       0x33 - 0x38 gpios are sd related.
21:47.32cr2_[Do Nov 22 2007] [01:06:46] <cr2>       0x3e - 0x43
21:48.02dcordesgreat
21:48.19dcordesI will try to find the ones that belong to the button below the jog wheel
21:48.35cr2_6 pins.
21:49.19cr2_both gpios fit into bank2. good.
21:51.17cr2_http://wiki.xda-developers.com/index.php?pagename=Kaiser_GPIO
21:52.04cr2_dcordes: what is "modem" ?
21:52.29cr2_don't forget about the CPLD gpios too.
21:54.04*** join/#htc-linux dcordes_ (n=dcordes_@unaffiliated/dcordes)
21:54.54dcordes_cr2_, sorry my connection died
21:55.05dcordes_got your messages in the log
21:55.47dcordes_cr2_, you mean modem related 1 2 in kaiser irq?
21:57.59dcordes_1 and 2 are active when I activate the phone function in connection manager
21:58.20cr2_ok.
22:00.40dcordes_I just retried that it's a sure thing
22:00.59dcordes_when I enabled the phone function 1 and 2 start flowing, when I stop the activity stops
22:01.42dcordes_6 seems to indicate connection of external buses
22:02.00dcordes_cr2_, do you have an idea what I can try? Am a bit clueless right now
22:02.10dcordes_I get no irq on touchscreen
22:06.10cr2_ts irq may be on the cpld.
22:06.58cr2_touchscreen is controller over I2C. check TSC2003 driver in the kernel source.
22:07.42dcordes_cr2_, how can I investigate the irqs on cpld again?
22:10.08dcordes_cr2_, I just tried to checkout wifi GPIO
22:10.29dcordes_unfortunately I have no usbnet to my host so I tried putty but there I have no console script
22:13.39cr2_putty is ssh ?
22:14.00cr2_try telnet, and use the right address.
22:14.17dcordes_that's no problem- I had the connection
22:14.29dcordes_pocketputty supports telnet
22:15.10dcordes_with my normal haret-wifi-desktop connection I told haret to listen on a certain port and connected via telnet to localhost port with pocketputty
22:19.33dcordes_cr2_, building synce
22:39.11dcordes_cr2_, can you help me with synce?
22:40.24Zooloochello therer
22:41.14Zooloocdcordes: I managed to connect my kaiser via usb with synce, but only with the ipaq kenrel module, unfortunately not the rndis driver
22:41.46dcordes_Zoolooc, well I don't care. All I need is some usb networking
22:41.58Zooloocdcordes: also, I managed to delete all my contacts when trying to run some python script I was'nt supposed to even try
22:42.13dcordes_ooops
22:42.17dcordes_did you backup?
22:42.21Zooloocguess you want it for the haret telnet session
22:42.37Zooloocyes, I've synced them back at the office
22:43.42dcordes_Zoolooc, yes right. I need to find out irq/gpio of wifi so I need some connection other then wifi to console host
22:44.00Zooloocand the synce folks found out it might happen with WM6 every time partnerships are deleted (I had to delete to create a new one with synce, as WM6 only supports 2 partnerships
22:45.59Zooloocdcordes: alright, you might be luckier with rndis-host than my, you could try it out. you can still use ipaq if it does not work. so these 2 are kernel modules
22:46.32dcordes_I installed synce and I'm now at the connect step
22:46.36Zooloocfor rndis-host, rather use the SVN version of rndis-lite from the synce project, than patching your kernel
22:46.54dcordes_no that's what I don't want, patching the kernel.
22:46.58Zooloocyou installed it from binaries?
22:47.07dcordes_no compiled as descriped here
22:47.15dcordes_http://www.synce.org/index.php/Building_SynCE_with_Windows_Mobile_2005_support_from_Subversion
22:47.43Zooloocok, for ipaq you might have to build the respective kernel module (but at least it's standard included there)
22:48.14dcordes_I will do the ipaq thing. Don't want no experiments but results.
22:48.35Zooloocok, then remember you also must switch your kaiser into serial mode
22:48.42dcordes_wait first things first
22:49.08dcordes_what am I supposed now after building and instlaling synce if I would like to do it the ipaq driver way?
22:49.31Zooloocyou need the ipaq kernel module
22:49.59Zooloocif your system has udev support, just when connecting usb, you should already see something in dmesg
22:50.12Zooloocbut only if you swithed WM6 into serial mode
22:50.12dcordes_ok modulel oaded
22:50.22Zooloocwhich was what I wanted to tell you about
22:50.46dcordes_ok how to switch kaiser into serial mode?
22:51.13Zooloocin the Settings, under "Verbindungen"/ USB to PC just uncheck "Erweiterte Netzwerkfunktiionen aktivieren"
22:51.59Zooloocwhich is supposed to be checked if you'd want to connect it via USB in nework mode with rndis-host
22:53.37dcordes_I want usbnet
22:53.42dcordes_for the telnet
22:54.02Zooloocyou get it via ppp in serial mode with the ipaq
22:54.14dcordes_good
22:54.20dcordes_I will attempt to connect now
22:54.40ZooloocI was referring to network mode as opposed to serial mode actually ppp
22:55.07Zooloocif you have udev you can automate the connection
22:55.57dcordes_remember I only have synce and ipaq driver
22:56.04ZooloocI also built FUR which can mount the whole PDA filesystem under linux, R/W
22:56.05dcordes_that's everything required?
22:56.19dcordes_sounds neat
22:56.32Zooloocfor the network connection to haret, yes
22:57.08dcordes_ok
22:57.10dcordes_connected...
22:57.18Zooloocdoes a new ttyUSB device show up when connecting?
22:57.31dcordes_[115897.074053] usb 7-4.1: PocketPC PDA converter now attached to ttyUSB0
22:57.41Zooloocok
22:58.19Zooloocnow my script triggered by udev runs "synce-serial-config /dev/ttyUSB1 192.168.178.55:192.168.178.57 fritz.box"
22:59.09dcordes_my desktop where kaiser is hooked up is a normal dhclient
22:59.18Zooloocwhich are the local and remote IPs given to the ppp link, and the DNS server, everything is optional, but the remote IP is what you need to run telnet on, when attempting to connect to haret
22:59.51Zooloocyes, I think the dns thing would not work
23:00.33Zooloocafter running that just run "synce-serial-start"
23:00.51Zooloocwhen you want to disconnect you should run "synce-serial-abort", that's all
23:01.15Zooloocthe first too commands can be called in a script by an udev rule
23:01.58Zooloocunfortunately my kaiser is configured to go into standby very sooon, which interrupts the connection, too. but anyway, when you work on it, you can keep it awake
23:03.07dcordes_my router running dhcpd is 192.168.0.1 and my desktop 192.168.0.3
23:03.14dcordes_how would the command look like in my case?
23:03.43Zoolooc2 completely new IPs
23:04.34Zooloocbecause you will see when running "ifconfig" after connecting, that a new ppp0 network interface defined by those new IPs will appear on your desktop machine
23:05.22Zooloocfirst one is the local, then the remote to which you will connect with telnet
23:06.19Zoolooctry "synce-serial-config /dev/ttyUSB0 192.168.0.4:192.168.0.5"
23:06.56Zooloocand after running synce-serial -start you should be able to telnet to 192.168.0.5 when haret is waiting on the other side
23:07.01dcordes_hmmm I have no synce-serial-config lol
23:07.57Zooloocwait a minute, I just "asked" my gentoo which package has that :-)
23:08.11Zooloochmm, there's a package synce-serial
23:08.38Zooloocdcordes: I got that from SVN few weeks ago
23:12.56dcordes_oops
23:14.19Zooloocdcordes_: so how are you getting along?
23:15.59dcordes_http://pastebin.ca/809508
23:16.21Zooloocdcordes_ after setting up the connection, you might want to automate it, I can give you my udev rule and this little 2-lines script, you might also want to get FUR
23:16.44Zooloocouch, I forgot about that
23:16.45*** join/#htc-linux echowip (n=bochet@dialup-4.246.141.252.Dial1.SanJose1.Level3.net)
23:17.18Zooloocdcordes_: it's already running all the time as a service for me
23:17.59dcordes_seen the pastebin?
23:18.07Zooloocdcordes_: do you have the odccm binary? The best would be if it also installed a service script for your distribution
23:18.20dcordes_odccm?
23:18.49Zooloocdcordes_: yes, that's one of the programs which provide this dccm service needed by synce
23:19.27dcordes_can't find it
23:19.49Zooloocdcordes_: then it's yet another synce package you need to get
23:20.15Zooloocbut then it should be sufficient
23:21.13dcordes_can you gentoo search it again?
23:21.45Zooloocit's called that way here
23:21.55dcordes_odccm?
23:21.56Zooloocsynce-odccm
23:22.17Zooloocthe synce guys also have vdccm, but that's said to be deprecated
23:22.31dcordes_no synce-odccm here
23:22.35dcordes_ubuntu
23:22.57Zooloocyou got your whole synce stuff from the unbuntu repositories?
23:23.08dcordes_but synce-dccm
23:23.18dcordes_will be the one
23:23.30Zooloocshould be, yeah
23:23.37dcordes_synce-serial-start cannot find the dccm process.
23:23.37dcordes_Without dccm your PPP connection will soon terminate!
23:23.38dcordes_again
23:23.56Zooloocand this time you ran synce-dccm?
23:24.04dcordes_lol no
23:24.06dcordes_must be the beer
23:24.39Zooloochehe
23:24.41dcordes_there is no such binary in my $PATH although it installed fine
23:25.01Zooloocmaybe it's called jsut dccm
23:25.10dcordes_yep
23:25.10dcordes_ran it
23:25.27Zooloocit might have forked into background
23:25.40Zooloocnow you can run that configure line and then start
23:25.41dcordes_synce-serial-start cannot find the dccm process.
23:26.10Zooloocdoes ubunto also know about other *dccm incarnations?
23:26.24dcordes_ah I tried to run it as #
23:26.58Zoolooclook for it's service script and put it in your services to start
23:27.35dcordes_finally
23:27.37dcordes_synce-serial-start is now waiting for your device to connect
23:27.50dcordes_I get no new interfaces up-
23:28.23Zooloocwas your kaiser connected via USB, configured in serial mode and "alive"?
23:28.42dcordes_kaiser is alive
23:28.48dcordes_it is connected via usb
23:28.50Zooloocyou should also see something on the kaiser display when running synce-serial-start
23:28.57dcordes_and I disabled the erweiterte stuff
23:29.00dcordes_as you discribed
23:29.56Zooloocand the ipaq kernel module loaded and created that /dev/ttyUSB0 device?
23:30.14Zooloocand did you run the config line?
23:30.44Zooloocmaybe you just need odccm
23:30.45dcordes_device was there
23:30.52Zooloocwas?
23:30.58dcordes_I'm rebooting
23:31.00Zooloocor still is?
23:32.04dcordes_ok
23:32.05dcordes_up again
23:32.41dcordes_ok
23:32.46dcordes_the interface is there
23:33.17dcordes_I was trying to read with cat /dev/ttyUSB0
23:33.28dcordes_it said CLIENT
23:33.35dcordes_and then on kaiser Verbinden mit 'USB popped up
23:36.23Zooloocdcordes_: that's it, then, you should now be able to telnet to that IP
23:36.34Zooloocshown as remote in the ppp0 interface
23:36.42dcordes_there is still no interface-
23:37.06dcordes_synce-serial-config /dev/ttyUSB0 192.168.131.102:192.168.131.201
23:37.17Zooloocstrange, waht about odccm now, does it ruin?
23:37.20dcordes_synce-serial-start
23:37.22Zooloocrun?
23:37.34dcordes_#synce-serial-start did not ask for dccm
23:37.44dcordes_synce-serial-start is now waiting for your device to connect
23:37.48dcordes_#no activity on kaiser
23:37.58dcordes_no interface
23:38.08Zooloocvery strange, maybe you really need odccm
23:38.20dcordes_I have it
23:38.23dcordes_and it runs
23:38.56dcordes_I would rather like to try the other non-ipaq method
23:38.59dcordes_maybe it works for me
23:39.39ZooloocI think that one too needs odccm
23:39.56Zooloocbut for me, that one fails sooner, already at the kernel part
23:39.59dcordes_no problem I have odccm
23:42.29dcordes_synce is now waiting ..
23:42.34dcordes_but kaiser does nothing
23:42.42dcordes_odccm is on
23:45.17Zooloocstrange, for me it just connects as soon as pysically connecting the usb cable (well, I told you, odccm servioce is running all the time, and I'm calling the synce-serial-config blabla and the serial-start by an udev rule
23:48.22dcordes_synce-serial-start detected that a SynCE serial connection was already
23:48.22dcordes_started with PID 31436.
23:56.17cr2_how do i block the fscking flash spam in firefox ?

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