IRC log for #htc-linux on 20070625

00:05.41*** join/#htc-linux Hawk||- (n=Hawk@pD9E08185.dip0.t-ipconnect.de)
01:23.50*** join/#htc-linux silven-mobile (n=zmc@69.89.36.38)
03:10.39*** join/#htc-linux thinko (n=jdoe6alp@smaug.rackdragon.com)
04:12.25*** join/#htc-linux dion (n=dion@inhex.net)
05:18.30*** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com)
05:30.32*** join/#htc-linux sadeness_ (n=LamersIn@nat1.rlan.ru)
06:03.37*** join/#htc-linux rob_w|mis (n=rob_w@p549BA1AB.dip0.t-ipconnect.de)
06:03.45*** join/#htc-linux psokolovsky__ (n=psokolov@82.193.98.14)
06:20.23*** join/#htc-linux _law_ (n=law@mx2.stiftadmont.at)
06:25.29*** join/#htc-linux pleemans (n=peter@leemanpe.ugent.be)
06:39.22*** join/#htc-linux pleemans (n=peter@leemanpe.ugent.be)
06:39.23*** join/#htc-linux rob_w|mis (n=rob_w@p549BA1AB.dip0.t-ipconnect.de)
08:40.49*** join/#htc-linux _law_ (n=law@mx2.stiftadmont.at)
08:47.08*** join/#htc-linux pH5 (n=ph5@p5485C2ED.dip.t-dialin.net)
08:55.08*** join/#htc-linux psokolovsky (n=psokolov@82.193.98.14)
09:14.26*** join/#htc-linux psokolovsky (n=psokolov@82.193.98.14)
09:16.25*** join/#htc-linux bd2 (n=cbou@85.21.88.2)
09:34.40goxboxlivegood morning everyone
09:37.27Foxdiegreetings
09:39.46*** join/#htc-linux BabelO (n=Miranda@faim1.itesoft.com)
09:40.04BabelOgood morning
09:40.33BabelOgoxboxlive: ok, so i try mediaplayer, and same than you, no sound playing
09:40.54BabelOi ve same issue before with some mp3, did you try other format ?
09:43.41Foxdiegoxboxlive: are you or anyone else interested in purchasing a damaged Universal very cheap? there is a problem with the motherboard, I think it needs a solder reflow, you can usually get it working by putting it at a funny angle.
09:48.23*** join/#htc-linux rob_w|mis (n=rob_w@p549BA1AB.dip0.t-ipconnect.de)
09:51.30*** join/#htc-linux rob_w|mis (n=rob_w@p549BA1AB.dip0.t-ipconnect.de)
09:52.25*** join/#htc-linux KimuSan^ (n=kimusan@ip205.tankegang.dk)
10:03.15*** join/#htc-linux TBBle (n=tbble@2001:388:f000:0:0:0:0:2d)
10:15.11cr2Foxdie: the most interesting part of the universal is LCD :)
10:15.13*** join/#htc-linux psokolovsky_ (n=psokolov@82.193.98.14)
10:16.03Foxdieof course
10:16.07Foxdiethe lcd is fine
10:16.22Foxdieits just the driver chip or another chip on the motherboard is making intermittent connection
10:16.29cr2if it's working, the rest is not very important.
10:16.34Foxdiemovement or shock makes it crash
10:16.53cr2i've disassembled the universal to look at the chips.
10:17.01Foxdieas have I, I can't find any damage
10:17.13Foxdiebut I did trace the fault to near the SD card slot
10:17.31cr2the LCD is connected by a funny cable, which you can't damage, only the connector on the motherboard.
10:17.48cr2SD ?
10:17.49Foxdieso while I have the chance, if you need any information from motherboard ie. photos / chip numbers then let me know
10:18.09cr2i think we know them all.
10:18.11cr2hmm.
10:18.29cr2probably the i2c RTC chip. if it's on the motherboard.
10:18.34Foxdieaye, the problem with it is there's a loose connection someplace (maybe a split solder connection), when it breaks the connection the PDA crashes
10:18.56cr2ok.
10:19.08Foxdieit happened when I tried to fit a long life battery kit, the new battery cover didn't fit well, I had to force it to close it
10:19.23Foxdieand I think by doing so I added stress to the board causing a track to break or something
10:19.53cr2well, i can buy it, but what i'm going to do with so much devices ? :)
10:20.08Foxdieuse it as a doorstop :D
10:20.11cr2i'd like to trace the USB host path.
10:20.20Foxdieactually no, I was thinking maybe use it for testing flash
10:20.28Foxdieor tracing usb host ;)
10:20.41Foxdieif you can get it to the right angle, it will work fine
10:20.47Foxdiebut any movement will make it crash :(
10:20.59cr2hmm. yes, since we know now where is JTAG reflashing is not that dangerous.
10:21.17*** join/#htc-linux psokolovsky_ (n=psokolov@82.193.98.14)
10:21.26Foxdiecool :)
10:21.36Foxdieshould I bother to buy a new one?
10:21.57cr2see my comment about hermes/athena.
10:22.20Foxdiewhere are they?
10:22.26cr2looking.
10:22.31Foxdie;)
10:22.49cr2[22:02] <cr2> if you want a phone you go hermes, if you want a mini-notebook you go athena.
10:23.58FoxdieI wanted a mini notebook
10:24.02FoxdieI don't make many phone calls on it
10:24.10Foxdiehowever, athenas specs are not impressive
10:24.15cr2i dislike the non-accelerated LCD on universal.
10:24.39cr2well, then you don't need to bother with non-x86.
10:25.08Foxdiewasn't the spec of athena identical to universal except slightly faster cpu?
10:25.19cr2no.
10:25.53Foxdiewhat is better then?
10:26.02Foxdiegraphics acceleration?
10:26.26cr2it has ATI chip + sirf3 gps + s-video/video/vga out, USB host and 128MB RAM. and hsdpa.
10:26.41FoxdieO.O
10:26.47cr2ATI is accelerated.
10:26.59FoxdieOoh the dirty little tease
10:27.03cr2and 8GB microdrive.
10:27.12Foxdiewhen the first specs for it came out it wasn't much better than universal
10:27.15Foxdiebut now I might get one :)
10:27.36cr2microdrive is on the IDE controller.
10:27.54Foxdienice
10:28.00Foxdiewhats the linux support for it like right now? ^_^
10:28.01cr2once i'll write the driver, it will be a great device.
10:28.16cr2i can point you to a screenshot :)
10:28.46cr2Kevin2 has written the CPLD driver and the touchscreen/battery monitor driver.
10:29.01cr2gps is trivial to support.
10:29.12cr2i have traced most of the gpios.
10:29.42cr2the tricky part may be the wifi, because it does something over sdio.
10:30.08cr2but pxa sdio is supported now. the driver is ugly, but it works.
10:30.42cr2the miniSD card is like on hermes (on ATI chip, so not supported).
10:30.54Foxdiehmm, they seem to sell for at least £500 here in the UK..
10:31.18cr2the sound is also on ATI, no RE work has been done here, but it's the same on hermes.
10:31.47cr2yes, but it's a mighty device.
10:32.25FoxdieHmm, I could renew my phone contract and get a blackberry, then sell that on ebay for £300, then where do I get the remaining £200 from?
10:32.33Foxdiehow much do you think my damaged Universal is worth?
10:43.23goxboxliveBabelO, no i havent tried any other formats, maybe i should :-)
10:44.06goxboxliveBabelO, was it with your own build OpieII or was it with mine you tried mediaplyer?
10:48.50*** join/#htc-linux dion (n=dion@inhex.net)
10:52.21goxboxliveBabelO, i am going to try out some ogg and wav files now, both music and kde system sounds
10:54.26goxboxlivehmm ogg is not supported
10:55.01goxboxliveand not wav either, hmm strange
10:55.41goxboxliveMaybe i just needs some more codecs
10:56.06goxboxliveBut what is the name of the codecs for mediaplyer in OPie(1)
11:10.29*** join/#htc-linux DTMonterrey (n=evandro@cornelius.estig.ipb.pt)
11:11.38goxboxliveNo, non of them will play
11:25.34*** join/#htc-linux Tjikkun (n=tjikkun@82-204-54-115.dsl.bbeyond.nl)
11:34.01goxboxliveShouldnt there have been a menu with system sounds or something in OpieII
12:13.46*** join/#htc-linux cr2 (n=konversa@crpl6.physik.uni-wuppertal.de)
12:16.19BabelOgoxboxlive : : i ve tried with your opie image ( miinefailed always with dbmigrate)
12:16.34BabelOgoxboxlive : you can look at plugin in /opt/Qtopia/lib
12:37.29*** join/#htc-linux alsuren (n=alsuren@host86-136-136-96.range86-136.btcentralplus.com)
12:56.13*** join/#htc-linux __law__ (n=law@mail.stiftadmont.at)
13:17.59*** join/#htc-linux ellisway (n=ellis@83.167.181.177)
13:35.15pH5hm. I have patched pxafb to use the frame buffer in SRAM instead of allocating it via dma_alloc_writecombine.
13:35.27pH5but now X doesn't start because I can't mmap /dev/fb0 anymore
13:35.27*** part/#htc-linux alsuren (n=alsuren@host86-136-136-96.range86-136.btcentralplus.com)
14:04.01*** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com)
14:20.39*** join/#htc-linux non-sense (n=Shuntu@unaffiliated/non-sense)
14:23.19*** join/#htc-linux rob_w (n=bob@M9417.m.pppool.de)
14:23.33non-sensehello to all
14:23.50non-senseim kinda new with linux os, i just purchased my first htc phone
14:24.02non-senseim curious, what some of you think the advantages are with using linux
14:43.40pH5non-sense: which phone? linux for most htc phones is in "kernel hackers welcome, everyone else - good luck" state
14:56.23*** join/#htc-linux zmc (n=zmc@adsl-72-50-84-108.prtc.net)
15:03.35*** join/#htc-linux ellisway (n=ellis@83.167.181.177)
15:06.53non-senseis pH5
15:06.56non-senseic*
15:11.18*** join/#htc-linux silven (n=zmc@adsl-72-50-84-108.prtc.net)
15:20.21*** join/#htc-linux BabelO (n=Miranda@faim1.itesoft.com)
15:20.50BabelOcr2 : what is status of HTC Trinity ? same as hermes ?
15:29.47psokolovsky_BabelO: Hi, just a ping, what's on acx_cs front?
15:29.52pH5BabelO: seems to be similar hardware, but do we have a trinity owner?
15:31.05BabelOpsokolovsky_ : acx_cs is loaded and binding to correct socket, but i force value and don't read it from card :( that's why i m unhappy with it
15:31.21BabelOpsokolovsky_ : i send you my code on the evening
15:31.26psokolovsky_BabelO: can you post whatt you have to k-d?
15:31.29BabelOpH5 : yes maybe me
15:31.30psokolovsky_ok
15:33.21BabelOpH5 : thing i m not sure is embedded GPS into trinity it is not a SIRFIII
15:35.59BabelOpsokolovsky_ : good thing is that i test compilation with and w/o cs.c it does not break anything
15:36.04*** part/#htc-linux silven (n=zmc@adsl-72-50-84-108.prtc.net)
15:36.47psokolovsky_ok
15:38.55*** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com)
15:42.44*** join/#htc-linux non-sene (n=Shuntu@CPE001310ece823-CM0012257058c6.cpe.net.cable.rogers.com)
15:45.46*** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com)
16:08.26*** join/#htc-linux dion (n=dion@inhex.net)
16:20.47cr2BabelO: i think it's the qualcomm gps. 6 or 8 channel tech, so it's not very sensitive.
16:21.41cr2the only person with trinity was KimuSan^ but the is silent for a long time.
16:22.40BabelOcr2: : ok so maybe better is arthemis :)
16:23.52BabelOcr2: you say 6 or 8 channel so it is capacity to receive 8 gps sat at same time, what is a correct sensitive channel number ? i don't know enought about this
16:24.33BabelOok but arthemis is OMAP 850 based :(
16:28.39cr2BabelO: the real outcome is that the sensitivity is much lower.. it's a really long discussion, not just 1 sat per channel.
16:30.24cr2omap850 phone is very difficult. i guess keyboard, SD and ts can be fixed.
16:34.02BabelOcr2 : so it is like radio car, something like 0.8uV is better :)
16:34.46non-sense[Slashdot] Mono Coders Hack Linux Silverlight in 21 Days () - http://rss.slashdot.org/~r/Slashdot/slashdot/~3/127800458/article.pl
16:41.45cr2microsoft paid coders ?
16:49.14non-sensei suppose
17:01.12KimuSan^cr2: I have been too busy to do anything about it.
17:01.35KimuSan^but if anyone does any hacks, I will be happy to test
17:07.07*** join/#htc-linux ellisway (n=ellis@83.167.181.177)
17:07.34non-sensei have the new HTC s720, and hav no idea what to do first :S
17:07.40non-senseany inssight? send em away
17:10.37cr2non-sense: what is s720 ?
17:12.28non-senseSmartphone
17:12.36non-senseits not even on the htc website yet
17:12.49cr2the kaiser ?
17:13.01non-sense¿kasier?
17:14.18cr2http://www.pdadb.net/index.php?m=specs&id=759#
17:14.40non-sensethere she is
17:15.00cr2ok. can you run haret ?
17:15.15non-sensei dont know what that is
17:15.19non-sensethis is my first smartphone
17:15.32non-senseim just irc-related, and i seen some channels for it
17:15.38cr2there are 2 wince programs you can run
17:15.39non-senseso i decided to see what i can pickup
17:15.49non-senseorly
17:16.00non-sensewithin the phone?
17:16.01cr2the one is called 'haret' and another one 'testwm5'
17:16.22cr2yes, just normal wince programs .exe
17:16.23non-sensehm, i kno im runnign wm6 standard
17:16.27non-senseif that helps
17:16.32cr2it's ok.
17:16.44cr2does it have wifi ?
17:16.52non-sensei dont think so no
17:17.07cr2Wireless LAN:   802.11b, 802.11g
17:17.09non-senseyes.
17:17.12cr2says the list.
17:17.12non-sense:S
17:17.14non-senseyea
17:17.16non-sensejust seen it
17:17.28non-sensei just remember reading on some reviews sites, that it didnt have wifi
17:17.37non-sensealso, said it was wm5, which isnt tru either :S
17:17.37cr2Expansion Slots:   SDIO, microSD, TransFlash
17:17.45non-sensePimped phone
17:17.50cr2doesn't matter. there is not much difference.
17:17.58non-sensefigures
17:18.45cr2ok, the easiest thing you can do is to get haret.exe, put it on the microSD and run it.
17:19.33non-sensewhere would i find this
17:19.39non-sensepdsdb?
17:19.41non-sensepdadb
17:19.51cr2http://handhelds.org/~koconnor/haret
17:20.10cr2haret-20070624.exe
17:20.50non-senseand just pu ton the phone, via usb
17:21.05cr2put it on the card. create an empty file 'earlyharetlog.txt' in the same directory. run it.
17:21.06cr2yes.
17:21.40non-sensehavnt purchased an MicroSD yet, have to do that today
17:22.01cr2the file haretlog.txt will be  created. post it.
17:22.23cr2ok, you can just upload it on the phone.
17:23.13non-senseok
17:23.19non-senseight
17:23.32non-senseso when i plug in usb, i just opend phone thru my comp
17:23.42non-sensemy pic, my ringt, templates, are visible
17:23.49non-sensei just draggd n dropped in there
17:24.10cr2ok.
17:24.54non-sensewhen i run from phone, i get ms error
17:25.09cr2which one ?
17:25.26non-sensethis haret-20070624
17:26.00cr2i mean 'ms error'
17:26.46non-sensedoesnt say..
17:26.51non-senseWe're Sorry....
17:26.57non-sensea problem has occred with $app
17:27.00cr2so it does not start ?
17:27.07non-senseplease tell ms about this, at no ost to you
17:27.08non-senseno
17:27.13non-senseunless im doing something wrong
17:27.15cr2ok.
17:27.24cr2everything can happen.
17:27.43cr2can you create the earlyharetlog.txt file ?
17:27.57cr2in the same directory and try again ?
17:27.59non-sensehow would i go about doing that
17:28.28cr2create it on the desktop and copy with activesync.
17:28.40non-senselol.
17:28.42non-senseright :S
17:28.44non-sensesrry again lmao
17:31.39non-sensehttp://137pm-surf.com/haretlog.txt
17:31.47cr2ok.
17:33.02cr2good.
17:33.26cr2the cpu is unknown, so it may be confused a bit.
17:33.52non-senseic
17:34.07non-sensehowd u kno
17:34.12cr2but the program is running, and that's good.
17:34.32non-sensehehe
17:34.36non-senseim confuseddd
17:34.40non-senseits running ?
17:35.02cr2you can see in this file the log of its actions.
17:35.12cr2the non-gui part is running.
17:35.21cr2and reports some useful info.
17:35.43cr2hmm. it can actually report much more useful info even at that stage.
17:36.06non-senseyea
17:36.09cr2i must buy some food now, otherwise it'll be too late ;-)
17:36.15non-senseallright
17:36.20non-sensethanks fer yer help
17:36.21cr2will you stay here ?
17:36.23non-senseyep
17:36.29non-sensei'll idle, until i have a uber phone
17:36.31cr2i'll be back in an hour.
17:36.36cr2bbl.
17:36.38non-sensethis is all new to me, and i dont like feeling like a noob.
17:36.39non-senseok
17:36.40non-sensetty then!
17:36.41non-sensepz
17:55.12*** join/#htc-linux GSV (i=GSV@c-71-202-155-188.hsd1.ca.comcast.net)
18:06.50*** join/#htc-linux robster (n=robster@light.bluelinux.co.uk)
18:07.32robsterright
18:07.55robsterso i'm trying to find a kernel source tree for the HTC alpine
18:08.01robsterand my google juice is just not there
18:10.50*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
18:11.15psokolovsky_robster: alpine is in hh.org cvs
18:15.08robsterta
18:27.11BabelOgoxboxlive: power button is working for me in opieII
18:27.57goxboxliveBabelO, for me too with HTC Universal, but not with Magician. But with Universal it suspend the damn thing not shuttinng it off.
18:29.27BabelOi don t understand? it suspend universal right ? it use apm -s
18:30.15goxboxliveyes suspends universal, nothing on magician
18:30.22goxboxliveDid you boot your own image now?
18:30.46BabelOno i can't, i ve to try svn up maybe ;)
18:31.07goxboxliveBabelO, i builded e new image (rev89) overnight, and it starts
18:31.45goxboxliveBut it is unstabel as a vulcano. 4.2.1 worked a lot better
18:32.32cr2non-sense: i've uploaded a modified haret version to http://jornada820.sf.net/files/haret     the filename is haret-libra.exe
18:34.32BabelOdid your image contain battery patch ?
18:35.19*** join/#htc-linux non-sene (n=Shuntu@CPE001310ece823-CM0012257058c6.cpe.net.cable.rogers.com)
18:36.53goxboxliveBabelO, a new patch?
18:37.22BabelOgoxboxlive: yes for new localization of ds2760
18:37.28goxboxliveBabelO, yes
18:38.08goxboxliveWhy do we have to change VisualCharge -> Charge ?
18:39.03BabelOgoxboxlive: don't know , need to look at other device to see how they manage battery
18:39.45BabelOi think we report value at correct place
18:40.02goxboxliveOk
18:41.48cr2hmm. _start is not defined in wince ?
18:43.06cr2non-sene: got a message about haret-libra ?
18:45.07BabelOgoxboxlive: my svn up failed  :( you have to make clean qtopia even with depot_older
18:45.40*** join/#htc-linux sinmsinm (n=alex@1.Red-83-43-229.dynamicIP.rima-tde.net)
18:45.45*** part/#htc-linux sinmsinm (n=alex@1.Red-83-43-229.dynamicIP.rima-tde.net)
18:47.51goxboxliveBabelO, I am shadow building so i am able to svn up. But sometimesit coemplaines about existing file, so i have to delete it manually and then svn up again.
18:53.26BabelOgoxboxlive: ok so same thing happen for me and it is makefile :(
18:53.52goxboxliveok
18:55.09*** part/#htc-linux rmoravcik (n=rmoravci@r2i34.net.upc.cz)
19:03.32*** join/#htc-linux tsdogs (n=tsdogs@62.123.180.130)
19:05.59*** join/#htc-linux Fossi (n=Fossi@e176120175.adsl.alicedsl.de)
19:10.49BabelOgoxboxlive: ok revision 89 finished, lot of change ;)
19:16.36*** join/#htc-linux the_sys0p` (n=the_sys0@cpe-66-75-196-173.bak.res.rr.com)
19:17.23goxboxliveBabelO, yes, but not enough, i guess we can expect more to come since it is so unstable.
19:20.55BabelOgoxboxlive: yesturday i try w/o smart and VNC and works well
19:21.28cr2goxboxlive: Q: How can I make a call with this device? There is no earpiece set? A: No, use the speaker phone, or set the volume to low and you can place it near your ear. Or use bluetooth headset.
19:24.44cr2Q: Does Windows Media Player eat up to 60Mb RAM? Any Solutions? A: At this time, you can use Pocket Music to replace Windows Media Player.
19:25.21BabelOcr2: i prefer gsplayer for music and tcpmp for video ;)
19:25.44BabelOis it a FAQ ?
19:26.42cr2yes. for wince on athena :)
19:26.56cr2and it's impossible to switch the BL off :)
19:27.09BabelOcr2: really ?
19:27.09cr2in linux it's not a problem. gpio+pxa pwm.
19:27.26cr2yes.
19:27.54cr2i'm also interested why it's impossible to use the headset.
19:27.58BabelOcr2: maybe it is to prevent some stupid use of device :)
19:28.15cr2if ak4641 switches audio, and it is.
19:28.33cr2maybe they did not wire it.
19:28.37BabelOcr2: you say impossible to use headset for playing music ?
19:28.43cr2but we can check :)
19:28.47cr2no. to call.
19:29.11cr2linux gives the ability to check everything.
19:29.13BabelOah ok, it is not a2dp compatible ?
19:29.43cr2why ?
19:30.18cr2the i2s sound goes through ATI like on hermes. will be a pita.
19:30.24BabelOi ve tested a a2dp headset with BA and it have mic embedded ;)
19:30.42cr2it#s the same chip on athena.
19:45.17goxboxliveBabelO, Well i have tried out both, but it does not seem to help for me. I just tried it out once again with Magician and it segfault when i hit the applications meulist.
19:45.49goxboxlivecr2, what kind of device are yo talking about?
19:46.54cr2athena
19:47.40goxboxlivei c
19:49.48*** join/#htc-linux bd2 (n=cbou@89.113.85.104)
19:52.42goxboxliveBabelO, have you noticed that with the Qtopia theme the bluetooth icon is on when bluetooth is on .
19:54.32BabelOgoxboxlive: yes i see it yesturday with your image, and also blue color ;) nice
19:54.41goxboxliveyes
19:55.10goxboxliveIs that only defined in a the theme?
19:55.55BabelOon sony ericsson bluetooth icon change if device is visible or not ;)
19:56.02goxboxlivei know
19:56.06BabelOgoxboxlive: yes it is in theme folder
19:56.14goxboxliveok
19:56.47goxboxliveBabelO, did gnuphone compile for you ?
19:56.51BabelOgoxboxlive: yes, it is with svg
19:57.10goxboxlivesvg what?
19:57.15BabelOgoxboxlive: not at this state actually, compiling pim
19:57.40BabelOsvg is for bluetooth icon an white bar where it is displayed,
19:57.57goxboxliveOk, mine didnt compile, do you think it's not compiling because of 4.2.2 version and not 4.2.1 ?
19:58.02goxboxliveahh ok
19:58.21goxboxliveI hate this:
19:58.22goxboxliveAudioVolumeManager::registerHandler "Media" "QPE/MediaServer/MediaVolumeControl"
19:58.23goxboxliveCannot get semaphore /tmp/qtopia-0/sharedCache 't'
19:58.23goxboxliveError 2 No such file or directory
19:58.23goxboxliveSegmentation fault
19:58.23goxboxliveQAbstractSocket::waitForReadyRead() is not allowed in UnconnectedState
19:58.24goxboxliveQAbstractSocket::waitForReadyRead() is not allowed in UnconnectedState
19:58.26goxboxliveQAbstractSocket::waitForReadyRead() is not allowed in UnconnectedState
19:59.02BabelOdon't try since two days now building it if qtopia finish before midnight ;)
19:59.34BabelOgoxboxlive: start  qtopia with strace and you see why it end and also what is this No such file or directory
20:00.06goxboxliveok
20:01.46*** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
20:04.29goxboxliveBabelO, strace doesnt give anything actually. Sometimes it just segfault when entering a menu.
20:04.51goxboxliveI'' try universal with version 89
20:05.00BabelOok
20:16.07goxboxliveThis is not correct:
20:16.17goxboxlivehtcKbdDriverPlugin:create()
20:16.17goxboxlive***Loaded htc keypad plugin!
20:16.17goxboxliveOpened /dev/input/event3 as keypad input
20:16.17goxboxlive***Loaded htc buttons plugin!
20:16.17goxboxliveSet device /dev/input/event2 to handle GPIO
20:16.18goxboxliveSet device /dev/input/event1 to handle ASIC3
20:16.30goxboxlivekbd driver should be at event3
20:17.26BabelOgoxboxlive: it is on universal ?
20:17.42pH5goxboxlive: uh-oh, are you still hardcoding event paths?
20:18.01BabelOpH5: not hardcoded it is param for opieII
20:18.02goxboxlivecr2, yes
20:18.22pH5you should let udev handle that or at least scan all event devices for the input device name
20:18.30goxboxlivepH5, oh no, not me. Hardcododing is not allowed i have been told :-D
20:18.46pH5the /dev/input/event order is not fixed as soon as you start loading input drivers as modules in a different order.
20:19.37BabelOpH5: what is bast method to detect device ? do you have little code to do this ? :)
20:20.09BabelOs/bast/best/
20:22.03BabelOpH5: i see that gpe do it very well ;)
20:22.06pH5BabelO: not sure, so far I only have the touchscreen which is handled by udev. The keyboard still goes through the virtual terminal layer.
20:22.06pH5ioctl (fd, EVIOCGNAME (sizeof(name)), name);
20:22.12pH5can be used to read an input device's name.
20:23.23pH5selecting an input device based on that name would be better than depending on the path. but if you can make opieII not care what input device it gets it's data from, that would be even better.
20:23.46pH5I still have to play with Xserver's event input device support.
20:27.06BabelOpH5: ok, i say that because gpe use my pad button, but no calendar or contact
20:27.46pH5ah right, ba has the keys split between asic3 and pxa like on hx4700?
20:28.21BabelOpH5: yes, pxa, asic3 and also tsc2200
20:29.22BabelOtsdogs say that we can use tty0 to handle button, i am agrre with that but not sure itf it is best solution , what do you think about this ?
20:31.23BabelOusing tty0 is using keyboard output to console 0 ?
20:31.37pH5evdev would be better. aiui you can only get 7bit keycodes through tty.
20:32.19BabelOpH5: ah ok that's a good reason :)
20:33.15pH5that's why we have the input_pda.h hack in the kernel. there are quite some useful keycodes in input.h already for the calendar/contact keys that we can't use just because userspace doesn't support 8bit codes properly.
20:33.39pH5so it will be very good to finally fix userspace :)
20:34.54BabelOpH5: i like your point of view, another question, what about keyboard evdev ? do i need to map all keycode or just calling a standard kenrel function that return me right value depending on language settings ?
20:42.21cr2pH5: why does tty support only 7bit ?
20:44.48goxboxliveBabelO, there are still alot of missing files: http://pastebin.no/2192
20:47.04BabelOtry copying icon folder from source if you have it ?
20:47.19BabelOyou hav ealready done that i think :)
20:51.07goxboxliveBabelO, i havent had any segfault yet with Universal (rev89) (using opie theme and no VNC)
20:51.36BabelOgoxboxlive: ok
20:52.37BabelOgood then
20:53.17goxboxliveBabelO, But the keyboarddriver is buggy. When hitting the letter f it gives me 3. I have ssen that the case 21 is duplicate with 3 in the source. Have to look furter at it
20:55.13pH5actually, I'm not really sure if it is tty itself or maybe the X tty driver. but I know that 8bit keycodes don't make it through that layer here.
20:58.42goxboxliveBabelO, ok, the hex numbers where mixed. The number 3 had case 21 as F had. 3 shall have case 4 and F case 21
21:00.00BabelOwait i ll look
21:08.29*** join/#htc-linux tsdogs (n=tsdogs@62.123.180.130)
21:08.57BabelOgoxboxlive: yes it is an error
21:10.03goxboxliveBabelO, i'll go trough it and try to fix it, there are sveral things that is not correct.
21:10.24goxboxlivetsdogs, is there a way to have kbdtest to print the hexa decimal value?
21:10.57tsdogsgoxboxlive: you'd need to change the source code :)
21:11.19goxboxlivetsdogs, is it you who have made it?
21:11.45goxboxlivedo you have the source? Is it easy to do it or do i have to edit a lot?
21:12.09tsdogsgoxboxlive: pretty easy and yes I made for some keyboard test.
21:12.25tsdogsI need to get the source, which is on my laptop. wait a couple of mins
21:12.32goxboxlivethx
21:16.09cr2tsdogs: hi
21:18.53tsdogscr2: hi
21:19.27tsdogsgoxboxlive: http://84.18.144.44/tsdogs/tools/kbdtest/
21:20.12goxboxlivetsdogs, thx
21:20.28cr2tsdogs: i've patched the mainline roadmap to support the point layers, and teached buildmap to use the postgis database.
21:20.47cr2so you can use the same DB with the mainline code.
21:21.03cr2you'll only miss the routing and the editor.
21:21.20tsdogscr2: cool
21:21.39tsdogswe could merge it.
21:21.56cr2yes.
21:22.01tsdogsgoxboxlive: you only need to replace %d in printf to %x to print hex value
21:22.27goxboxlivetsdogs, ok, thx
21:22.47cr2but i've hit some bugs in the code, and your nice rendering additions to roadmap_canvas are not available too.
21:23.33tsdogscr2:  the nice additions are esily added. :)
21:23.47tsdogsthe bugs must be fixed
21:25.34cr2the style is not written from ~/.roadmap/ , for example.
21:25.50cr2s/written/read/
21:26.06tsdogscr2: I thought there was some code for sking (even in mainline)
21:26.31cr2i get some cryptic message is i add places.
21:26.59cr2the nice thing is that i can use 'rdmxchange'
21:27.05tsdogss/sking/skins/
21:27.10tsdogs:)
21:27.34cr2i've already learned a lot about the intrnal structure
21:27.52cr2and can draw the 'squares' and polygons directly.
21:27.59cr2it's very interesting.
21:28.24tsdogscr2: good, but what do you mean about draw squared directly?
21:28.25cr2now i need to dump the lines in qgis-friendly format.
21:29.07cr2you have points, polylines and polygons.
21:29.17cr2all of them are built of points.
21:29.24cr2with long/lat pairs.
21:29.31tsdogsok
21:29.36cr2buildmap gathers all points.
21:29.52cr2and finds the bounding box.
21:30.04cr2it's called global/square
21:30.39cr2this bounding box is split by the square grid,
21:31.10cr2so the points in one square have <16bit relative coordinates.
21:31.20cr2it's some kind of a tiling process.
21:32.09cr2for a 'point' you know a tile number ('square') and the 'local' <16bit coordinates.
21:32.25cr2for a tiled raster you have all tiles in the file,
21:32.35tsdogsok
21:32.45cr2but here only the tiles which actually have points are stored in rdm.
21:33.16cr2then there are 'dictionaries' which store the strings.
21:33.33cr2and they are linked to the points or lines.
21:34.01cr2the lines also use points, (from , to) and the intermediate segments.
21:34.29tsdogsok, do plygons have strings related?
21:34.59cr2yes, they are called 'landmarks'
21:35.28tsdogsok
21:35.44cr2each polygon described by its border line segments (pairs of points) has an associated landmark.
21:36.12cr2funny terminology, but it's used by the US TIGER public domain database.
21:36.21tsdogs:)
21:37.56tsdogscr2: what's missing from this format?
21:40.10cr2i have some problem with saving the 'places' database.
21:40.40cr2and the 65536 limit must go away.
21:40.51cr2there is no limit on 'lines'
21:40.59cr2but there is one on polygons.
21:41.22cr2maybe the polygon storage format should be changed a bit.
21:42.00tsdogsok
21:42.14cr2as far as i understand it, the polygons are 'global' objects, not related to the 'squares'.
21:42.40cr2maybe it's a good idea to cut the polygons by the squares grid.
21:42.59tsdogsas they are not routable probably.
21:43.02cr2you will add more artifical 'points' to the DB.
21:43.28cr2but then each 'square' can be rendered independently from others.
21:43.44cr2polygons are not routed :)
21:44.07cr2the routing info/data is completely separated.
21:44.18tsdogsho yes, right
21:44.25cr2and is not used in the mainline code now.
21:45.54tsdogsok, but mainline should have the code to handle plugins, so it could be merged.
21:46.16cr2i can commit my changes to CVS.
21:46.37cr2so you can add the canvas goodies too :)
21:46.51tsdogscr2:  sure, so I'll have a look at it, and start migrating the styles.
21:46.57tsdogs:)
21:47.08cr2btw, i've tried to crosscompile for qtopia.
21:47.13cr2and hit 2 problems.
21:47.29cr2you are using statusbar
21:47.33tsdogsfiledialog, you told be
21:47.35cr2which is not there.
21:47.38tsdogsok
21:47.40cr2and the qfiledialog.
21:47.41cr2yes.
21:47.45tsdogsok
21:47.56tsdogsdon't think this are major problems. :)
21:48.07cr2i think statusbar can be replaced by a popup or a tooltip.
21:48.24tsdogspopup is bad, let's go for toolfips.
21:48.27cr2but it is stupid not to have the qfiledialog.
21:48.33tsdogss/fips/tips
21:48.40tsdogs:)
21:48.53cr2i think we should compile a custom qtopia.
21:49.04cr2but it's not a new idea.
21:49.13tsdogsthere was code in qt to build one. I think we could have some sort of define there.
21:49.27cr2i've created a -xplatfrom htcphone
21:49.39cr2to use arm-linux-gnueabi-
21:50.08cr2apart from some silly code, qtopia compiles cleanly.
21:50.49tsdogsgood, at least all this changes in QT4 made it compatible with QPE :)
22:12.00cr2tsdogs: how do i create a db under my name, not postgres ?
22:12.16*** join/#htc-linux dimoniet (n=mal@81.203.43.57.dyn.user.ono.com)
22:13.42tsdogsu need to grant the privileges
22:14.30cr2as postgres ?
22:15.10cr2i always work as postgres to save all these troubles :)
22:15.43tsdogs:) ok, let me find some info
22:17.18tsdogsthis should be pretty explanatory: http://www.cyberciti.biz/faq/howto-add-postgresql-user-account/
22:18.18cr2thanks.
22:18.53tsdogscr2: this one looks more complete. http://www.davidpashley.com/articles/postgresql-user-administration.html
22:19.47tsdogsI think U especially need this: ALTER USER alice CREATEDB;
22:20.53cr2ok.
22:23.50cr2hmm. permission denied.
22:24.32tsdogshmm, strange.
22:25.38cr2ok, but i can create my own new  database.
22:25.45cr2so it's not a problem.
22:25.52tsdogs:) ok.
22:26.55cr2now i only need to switch to amd64 :)
22:27.23tsdogsshould not be a prob. :)
22:27.32cr2yeah.
22:27.51cr2but we should think about fixing the 64bit warnings where necessary.
22:28.27tsdogsok
22:28.31cr2btw, what do you think about label 'priority' ?
22:28.41cr2in the style.
22:29.12tsdogsI think it's pretty good, so depending on the zoom, most usefull names come up. sure
22:29.32cr2so big roads  and selected features have a priority in the label cache.
22:29.45tsdogsactually this something that the garming ~[ try to do
22:29.55cr2this cache does not work reliably anyway.
22:30.06cr2ok.
22:30.41cr2i've also an idea to differentiate between the rotated and non-rotated labels.
22:31.23cr2the apriori 'zero-length' lines, aka points should not have a rotation.
22:31.51cr2so if you'd like to label the point, just call draw_text() without rotation.
22:31.58tsdogsok
22:32.26cr2rotation for a point does not make sense anyway.
22:32.35cr2don't know about the polygons.
22:32.47tsdogsfor plygons it does.
22:32.52cr2simple logic says it applies for them too,
22:33.06tsdogsthough would not know how to calculate the rotation.
22:33.17cr2because they are labeled as 'landmarks' which are points.
22:33.44tsdogsor maybe label wrapping is better for polygons.
22:33.48cr2othewise you need to analyze how much of the polygon is visible
22:33.58cr2and it canbe quite costly on a pda.
22:34.22tsdogsyep.
22:35.00cr2this landmark can be generated as a centroid for the polygon.
22:35.54cr2so it should be located somewhere in the middle :)
22:35.54tsdogsyes.
22:36.23cr2so, should i commit my changes ?
22:36.38cr2the part with the places is quite hacky.
22:36.41tsdogsbtw, I have some strange blue lines on lakes, which interconnect each other on the map. you think it's a map, roadmap or qt problem?
22:37.06cr2you need to comment out a line so it builds the map.
22:37.21tsdogscr2: if it's finished "at least sort of" yes,
22:37.23cr2do you know the lake style name ?
22:37.37tsdogs? where
22:37.45cr2tsdogs: if you comment out the build*places line, it will work.
22:37.51tsdogsok
22:37.54cr2in your map.
22:38.15cr2change the lake color to white.
22:38.43tsdogsfor the pen? so pen shold not be used then.
22:38.46cr2you can also check the segments, 'borders' and 'water' with qgis.
22:39.16tsdogsok
22:39.19BabelOgoxboxlive: always here ?
22:39.30cr2i'll probably write a tool from the rdmxchange code to dump *.rdm back into .mif
22:39.35goxboxliveBabelO, :-)
22:40.04tsdogscr2: so no need for rdminfo anymore ?
22:40.11cr2tsdogs: so you can effectively verify  the *.rdm with qgis.
22:40.22BabelOgoxboxlive: always same issue with mine image :( can't boot... black screen
22:40.38cr2tsdogs: it will be the rdminfo. at least for the mainline branch.
22:40.50tsdogscr2: ok
22:41.40goxboxliveBabelO, hmm that strange, well allways the same with mine, no sound. Together we are chaos :-)
22:42.03tsdogsBabelO: what compiler u using?
22:42.11tsdogsOE?
22:42.23BabelOgoxboxlive: yes, and i m worry because with first release i can show something and have sound :(
22:42.35BabelOtsdogs: i try with both
22:43.08tsdogsBabelO: ok, I need to build one myself.
22:43.55BabelOok :(
22:43.59tsdogsgoxboxlive: so we can make more chaos :)
22:44.16goxboxlive:-)
22:44.28cr2we need a main config.
22:44.43cr2so each of us can do a build
22:44.47goxboxliveI am cleaning up the keyboard driver. Have some issue with CapsLoack and Shift.
22:45.39cr2BabelO: what are you doing with dbus, alsa and bt ?
22:45.53cr2can i just point the to hte OE staging dir ?
22:46.05cr2it worked very well with tslib
22:46.34cr2only the host uic/moc/rcc were confused.
22:48.20BabelOcr2: yes for dbus i point to bootstrap imagebecause need binary, alsa is broken with helix, so need to add it by hand in helix make file and point also to a valid libasound.so :) bt is ok if you point directly to bluez-libs directory
22:51.33cr2BabelO: hmm. what should be on the dbus configure line ? and bt
22:52.31cr2-xplatform htcphone
22:52.40cr2-dbus -dbuspath /home/fcr/armlinux/opie/usr
22:52.58cr2-bluetooth
22:53.07cr2-sound-system alsa
22:53.08BabelOcr2: dbus is something like that : -dbus -dbuspath /home/fcr/armlinux/opie/usr where  /home/fcr/armlinux/opie opie is my sdcard image
22:53.34cr2-extra-qtopiacore-config "-I ~/stuff/tmp/staging/arm-angstrom-linux-gnueabi/include"
22:53.49cr2why the image ?
22:54.08cr2-extra-qtopiacore-config "-L ~/stuff/tmp/staging/arm-angstrom-linux-gnueabi/lib"
22:54.43cr2it says alsa FAIL, bt FAIL
22:54.53cr2dbus FAIL
22:55.02BabelOcr2: because it look for libdbus-1.so under /usr/lib
22:55.33cr2$PATH/usr/lib ?
22:55.38BabelOcr2: for bluetooth -I/local/OE/tmp/work/armv5te-angstrom-linux-gnueabi/bluez-libs-3.11-r0/bluez-libs-3.11/include
22:55.50BabelOcr2: no $PATH/usr
22:56.01BabelOit found lib
22:56.16cr2BabelO: the includes must be in staging too. otherwise OE will not work.
22:56.29tsdogscr2: I think we should do a rootfs with all the needed devel packages, staging is too much split to be used.
22:56.36cr2$DBUSPATH/usr
22:56.45BabelOcr2: yes ;)
22:57.12BabelOand alsa is quite similar -I/local/OE/tmp/work/armv5te-angstrom-linux-gnueabi/alsa-lib-1.0.13-r3/alsa-lib-1.0.13/include
22:57.19cr2tsdogs: but staging is used by OE itself. how is it possible to compile qtopia in oe ?
22:57.33BabelOfor include and you need libasound.so for arm somewhere in library path
22:57.42tsdogscr2: you'd need a bb file to do this correctly I guess
22:57.47cr2you never know when these paths change alsa-lib-1.0.13-r3/alsa-lib-1.0.13/i
22:58.23cr2i hoped OE copies such packages into staging/
22:58.31cr2because it makes sense :)
22:59.04cr2othewise you need some bizarre package version lookup logic.
23:00.54cr2tsdogs: pg says 'only superuser can create procedural language'
23:01.14cr2no permission for language 'C'
23:01.16cr2LOL
23:01.28tsdogscr2: yep, you'd need superuser perms for this.
23:02.22cr2now you know why i do everything under postgres :)
23:02.28tsdogsyou either grant superuser privs to your user, or need to use postgres user for this taks.
23:02.30tsdogs:=
23:02.32tsdogs:)
23:02.55cr2how do i grant all  privs ?
23:03.07tsdogshmm wait
23:05.23cr2BabelO: i have include/alsa and lib/libasound.so in $STAGING/
23:05.33cr2so i don't see a problem here.
23:06.16tsdogsgrant dba to alice
23:06.18BabelOcr2: hummm include/alsa is not good i think because i see some #include <alsa/asound.h>
23:06.57BabelOcr2: ah sorry i don't understand ;) you have a big include and lib directory
23:07.21*** join/#htc-linux non-sense (n=Shuntu@unaffiliated/non-sense)
23:07.44cr2BabelO: i wanted to say that $STAGING/include should be enough.
23:08.04BabelOcr2: yes , ok
23:08.24cr2what about DBUS ?
23:08.43cr2what is it looking for in -dbuspath ?
23:09.04BabelOit is looking for lib/libdbus-1.so
23:09.32BabelOand also dbus in bin/ ( not sure about that)
23:10.32cr2why bin/ ?
23:10.47cr2looks silly for crosscompiling.
23:11.16cr2it's like linking uic with arm libdl.so ;-)
23:11.22BabelOnot sure i know i need to point level - 1 for libdbus-1.so
23:13.30cr2the bin/ is empty in staging. but the directory is there.
23:16.14cr2libdbus-1.so is in $STAGING/lib
23:16.48tsdogscr2: ALTER USER alice SUPERUSER
23:17.08cr2BabelO: so -dbugspath is $STAGING
23:17.16cr2tsdogs: thanks.
23:17.44BabelOcr2: humm, lidbus-1.so is under $STAGING/lib ?
23:18.17cr2tsdogs: great! now it works.
23:18.21cr2BabelO: yes.
23:18.50cr2<PROTECTED>
23:19.02BabelOcr2: good ;) so not at right place on my old image
23:20.26cr2BabelO: otherwise the things would be really complex.
23:20.51BabelOcr2: does it detect dbus + bt now ?
23:20.52cr2what abouut bt ?
23:21.01cr2have not tried yet.
23:21.13cr2i want to iron out the right config.
23:21.44cr2i have libbluetooth.so in lib/
23:21.46BabelOalsa is same as dbus, you need libasound.so somewhere
23:22.07BabelOcr2: so it is good for bt do you have bt header in include ?
23:22.17cr2yes, and include/bluetooth
23:22.38BabelOcr2: top, i think it is good for a test now ;)
23:22.42cr2libasound is in /lib
23:24.02cr2BabelO: should i remove everything from build ?
23:24.14cr2what was the command once again.
23:24.17BabelOcr2: no
23:24.43BabelOif you start configure again in build directory it clean all
23:24.56cr2ok.
23:25.05cr2how do i start configure ?
23:25.18BabelOcr2: you have depot_cache now ?
23:25.31cr2i want a clean setup, not some hacked stuff.
23:25.40cr2rm -rf *
23:25.44cr2in buuld now.
23:26.18cr2i needed to export something.
23:26.39cr2goxboxlive pointed me to some not quite correct wiki
23:26.46BabelOxport QTOPIA_DEPOT_PATH=/local/devel/opie/trunk
23:26.49BabelOexport QTOPIA_DEPOT_PATH=/local/devel/opie/trunk
23:26.51cr2where opie/trunk was missing.
23:26.58cr2ok.
23:27.35cr2and then ?
23:27.43BabelOcr2: yes point directly to trunk, then go to your build directory somewhere else, and start $QTOPIA_DEPOT_PATH/configure -device htcphone
23:28.08cr2ok.
23:28.13BabelOand it populate your build direcotry
23:28.46cr2hmm. i think i need PATH=~/stuff/tmp/cross/bin/:$PATH too
23:28.55BabelOyes ;)
23:29.34cr2PATH=~/stuff/tmp/cross/bin/:$PATH $QTOPIA_DEPOT_PATH/configure -device htcphone
23:29.46cr2looks good :)
23:30.42cr2hmm. why does it recompile qtopia for i386 ???
23:31.14cr2why can't it use my /usr/local/Trolltech/Qt-4.3.0/* instead ?
23:31.35tsdogsgood night.
23:31.35cr2sounds like a big waste.
23:31.43cr2goood night.
23:32.03BabelOcr2: i think they want to have a clean environment not dependant from host machine, and they need moc and uic to create cpp from uic file ;)
23:32.16BabelOgood night tsdogs
23:32.37BabelOcr2: does it fail for alsa and bt now ?
23:33.00cr2<PROTECTED>
23:33.00cr2<PROTECTED>
23:33.00cr2<PROTECTED>
23:33.19cr2The following default/detected values have been used:
23:33.40cr2does not say anything about dbus
23:34.13BabelObad :(
23:35.19cr2'-bluetooth' 'no'
23:35.22cr2hmm.
23:35.24BabelOcr2:  do you add -extra-qtopiacore-config "-I...."
23:35.31cr2should  i add 'yes' ?
23:35.31BabelOor just -I ?
23:35.35cr2yes.
23:35.37cr2....
23:35.48cr2i'm using your configure file.
23:36.34BabelOok it is looking for -extra-qtopiacore-config to pass -I and -L to qt base
23:36.48BabelOand also standard -I -L
23:36.59cr2standard can't be used.
23:37.14cr2because then they are applied to the host system.
23:37.18cr2for some reason.
23:37.20tsdogscr2:  you have -bluetooth ?
23:37.26cr2and everything is fcked.
23:37.29cr2yes.
23:37.30tsdogscr2: also there is a folder config.tests
23:37.53BabelOtsdogs: yes good idea, i debug with it my compile yesturday ;)
23:37.53tsdogsu can check there where the error is
23:38.59cr2'-bluetooth' 'yes'
23:39.00cr2'no'
23:39.04cr2hmm.
23:39.24cr2btw, can i disable infrared ?
23:40.52BabelOcr2: infrared not setup at the moment
23:42.06*** join/#htc-linux cyrill62 (n=user@ble59-5-82-233-205-36.fbx.proxad.net)
23:42.29*** join/#htc-linux cyrill62 (n=user@ble59-5-82-233-205-36.fbx.proxad.net)
23:43.22cr2tsdogs: config.tstes has only a makefile
23:43.58tsdogscr2: yes run make in the test folder which is failing, bt in OE build links to libusb afair
23:45.38cr2Could not find mkspecs for your QMAKESPEC after trying:
23:46.07cr2it tries for the host.
23:46.37cr2ok, it's late , i'll just start the make.
23:47.04tsdogs:) yep good night for real now
23:47.19cr2good night.
23:47.21BabelOgood night
23:47.39*** join/#htc-linux cyrill62 (n=user@ble59-5-82-233-205-36.fbx.proxad.net)
23:48.09GSVany news for hermes?! :|

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