IRC log for #htc-linux on 20080112

00:31.44MichouX|NAhey
00:32.43MichouX|NAI have a problem to install angstrom ipk on my blueangel.... there is someting strange on angstrom feed, it looks like empty ! :
00:32.46MichouX|NAhttp://www.angstrom-distribution.org/unstable/feed/armv5te/base
00:33.00MichouX|NAis there a mirror somewhere ?
00:35.51MichouX|NAhoho ! there is a new angstrom feed ! : http://www.angstrom-distribution.org/feeds/2008/ipk/glibc/armv5te/base
00:42.02MichouX|NAIt's 1h30 old
00:53.37*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
01:16.23*** join/#htc-linux dcordes (n=dcordes_@unaffiliated/dcordes)
01:31.14*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
03:04.15*** join/#htc-linux bernt_ (n=bernt@dslb-084-060-124-122.pools.arcor-ip.net)
03:55.27*** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-b520b089079e637d)
04:04.03*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
04:04.15*** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfbb20.pool.einsundeins.de)
04:22.37*** join/#htc-linux Kevin2 (n=Kevin@207-237-52-122.c3-0.avec-ubr12.nyr-avec.ny.cable.rcn.com)
04:51.55*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
06:25.38*** join/#htc-linux goxboxlive (n=jrs@195.159.97.196)
06:27.28goxboxlivegood morning
07:47.52*** join/#htc-linux hlbot (n=adm@iclem.net)
09:20.43*** join/#htc-linux pH5 (n=ph5@p5485E3D2.dip.t-dialin.net)
09:27.40*** join/#htc-linux rakeem (n=rakeem@n058152243210.netvigator.com)
09:44.16*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
09:49.16*** join/#htc-linux kiozen (n=kiozen@rgnb-4db1dbdf.pool.einsundeins.de)
10:02.41paulproteusMorning.
10:16.11pH5good morning
10:16.49goxboxlivehi pH5
10:18.20goxboxlivepaulproteus: i am upgrading my radio to version 1.18 now. Just to check if it is compatible with our drivers
10:18.30paulproteusgoxboxlive, Okay
10:51.15*** join/#htc-linux rob_w (n=bob@M894f.m.pppool.de)
11:11.05*** join/#htc-linux ymerejt (n=jerry@i02v-87-89-217-253.d4.club-internet.fr)
11:11.45ymerejtcr2, goxboxlive, pH5, tsdogs, HI
11:12.52goxboxlivehi ymerejt
11:13.08goxboxliveymerejt: did you see my screenies from yesterday?
11:13.25ymerejtno where are they ? (on handheld Universal ?)
11:13.38goxboxlivehttp://htc-pxa.linuxtogo.org/~goxboxlive/htcuniversal/screenies/
11:15.40ymerejtgixboxlive, COOL I want these applications !
11:16.58ymerejts/gixboxlive/goxboxlive
11:18.48goxboxliveymerejt: it's tsdogs work, it is suplied in his branch, otherwise i can upload my opt folder i made yesterday.
11:23.18goxboxliveymerejt: i am uploading the opt folder to you now. Also aply the patch in the root of your rootfs to have imlife and roadmap working.
11:23.27ymerejtgoxboxlive, no it's ok but the map application isn't on his branch
11:23.34goxboxliveIt's just to extract it oto the root
11:24.04ymerejt???????
11:24.58goxboxliveI know, it isent in your either. It's something i am bullding seperate and aplying only the binaries when doing make install
11:25.28goxboxliveWell anyway, it's uploaded :
11:25.31ymerejtgoxboxlive, and this application is located ?
11:25.31goxboxlivehttp://htc-pxa.linuxtogo.org/~goxboxlive/htcuniversal/tmp/
11:25.59goxboxliveThe roadmap? In the xanadux sources and at sourceforge.
11:26.07goxboxlivethere are two versions of it
11:26.45goxboxliveit is included in my uploaded opt folder. But you have to extract the patch file too.
11:28.38ymerejtgoxboxlive, why don't push the source ready for compile on the git repository ?
11:30.11goxboxliveymerejt: because i dont know how to do it. Roadmap fom the xanadux sources is tsdogs work. And if that where doable i guess he would have done it himslef in his branch.
11:33.20ymerejtgoxboxlive, everything is doable ;)
11:34.15*** join/#htc-linux aLeSD (n=alex@79.144.175.194)
11:34.29aLeSDhi all
11:35.02aLeSDI don't know if I have in the right place ... but I'm tring to run haret on my smartphone and I'm in trouble
11:35.13aLeSDsomeone know something about that stuff ?
11:35.25goxboxliveymerejt: i know, but i havent had so much time for this now. I am still at work :-) , i have a lot to do on work this days, so i have no time to figure out such new stuff :-)
11:36.55ymerejtgoxboxlive, ok
11:45.27*** join/#htc-linux BabelO (n=Fabrice@unaffiliated/babelo)
11:51.35*** join/#htc-linux kiozen_ (n=kiozen@rgnb-4db1dbdf.pool.einsundeins.de)
11:59.46BabelOhi
11:59.46BabelOaLeSD: what is your smarphone ?
12:00.38aLeSDBabelO: HP ipaq 514
12:01.10BabelOaLeSD: ah ah, cr2 just speak about it yesturday. it is an omap850 cpu ?
12:01.22aLeSDit is
12:01.43aLeSDI can't run haret on it ....
12:01.57BabelOaLeSD: you have to create an empty earlyharetlog.txt file at same place than haret
12:02.06BabelObtw what version of haret did you use ?
12:02.27aLeSD0.5.0 0.5.1 CVS
12:03.39aLeSDmaybe I have unlock the registruies of winMO
12:03.46BabelOaLeSD: ok, can you made test before ? and http://pastebin.ca the result of your earlyharetlog.txt ?
12:06.11aLeSDok
12:06.14aLeSDwait
12:07.42aLeSDBabelO: there's  nothing inside
12:08.10BabelOaLeSD: oups, maybe it is the wrong file name
12:08.11aLeSDI have a haretlog.txt
12:08.20aLeSDI past that
12:08.37aLeSDwow it's a log
12:08.38aLeSD:)
12:09.15aLeSDhttp://pastebin.ca/851930
12:10.32aLeSDI think it can't detect the cpu
12:10.40BabelOyes it is that
12:10.49BabelOaLeSD: you compile haret from sources ?
12:11.15aLeSDyes
12:11.45BabelOaLeSD: good, then you can fake the cpu detect until Kevin2 find a way to fix it
12:12.01BabelOaLeSD: it seem hp don't use same naming in wince than htc
12:13.45aLeSDBabelO: how ?
12:14.02BabelOaLeSD: in haret sources, if you look in src/arch-omap.cpp
12:14.16aLeSDLooking at arch Generic TI OMAP
12:14.16aLeSDException on arch Generic TI OMAP detect
12:14.17BabelOaLeSD: you have a funciton called MachineOMAP850
12:14.30aLeSDHaRET(4)# print "MMU L1 descriptor table address is %08x" MMU
12:14.30aLeSDTerminating haret due to unhandled exception (pc=00017640)
12:15.32aLeSDBabelO: I'm in
12:15.48BabelOaLeSD: then in this function you have CPUInfo[0] which correcpond to SmartPhone in your case
12:15.53BabelOi have to check this
12:16.02aLeSDit is
12:17.30BabelOaLeSD: oups no
12:17.32BabelOi m wrong
12:17.45aLeSDI mean I have an exception
12:17.56aLeSDI don't know if it's a name problem
12:18.09pH5would adding something like this to machlist.txt help? http://en.pastebin.ca/851938
12:18.40BabelOpH5: good that what i m looking for ;)
12:18.50aLeSD:)
12:18.51aLeSDtring
12:20.10BabelOshould work like this :)
12:20.24BabelOpH5: any luck botting kernel on BA ?
12:21.34pH5BabelO: couldn't try yet, my n810 arrived ;)
12:22.00BabelOpH5: oh, you got a n810 ? nice gadget
12:22.22BabelOpH5: you want to participate to maemo devel ?
12:24.06BabelOpH5: there is someone at tapioca-voip who is maintainer on this project
12:29.37pH5BabelO: yes, going to. I wanted to play with integrating maemo and openmoko stuff a little.
12:29.55BabelOok
12:30.18BabelOpH5: i ve an idea why ba don't boot with 2.6.23 kernel, i m trying it now
12:30.51pH5BabelO: good. do you have any news about where the init section mismatches come from?
12:31.14aLeSDBabelO, pH5 ... ok
12:31.22aLeSDnow the problem is :
12:31.25BabelOpH5: no i did not check more
12:31.28aLeSDInitializing for machine 'hp514'
12:31.28aLeSDHaRET(1)# addlist irqs P2V(0xfffecb00) 0x40601c2 32 0
12:31.28aLeSDTerminating haret due to unhandled exception (pc=00017680)
12:31.41aLeSDit's the same of before
12:32.02aLeSDwhe strange thing is that a guy has the same device and can boot it
12:32.41pH5I don't understand where this comes from. I moved pxa2xx_udc_gpio's probe function into __init but the udc warning is still there. and where the asic3 warning comes from I don't even pretend to understand.
12:34.11pH5aLeSD: exactly the same version, or is maybe something wrong with your toolchain?
12:34.34BabelOpH5: maybe it come from the integration of psokolosky gpio driver to have asic3 + pxa gpio in common ?
12:34.34BabelOpH5: i m doing the change and try it
12:35.13aLeSDpH5 ... asking for the version ... but I have the same error on the harlet binar too
12:36.19pH5aLeSD: strange. I don't know how it works, because I've never seen it, but isn't the smartphone wince edition somehow restricted - did you properly authorize the haret binary?
12:37.31aLeSDpH5: I'm sorry but I don't understand the "did you properly authorize the haret binary"
12:43.49aLeSDmaybe my mobile is simply blocken in same way
12:43.58aLeSDI have to ulock the registries
12:47.18*** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net)
12:48.02diogene31Morning everybody.
12:49.04aLeSDI have the log of the other device
12:49.17aLeSDI have to unlock registry it's sur
12:49.30aLeSDis it safe from a vmware windows ?
12:50.23BabelOaLeSD: not sure it works well, i ve tried to flash my device from vmware, and it never works
12:50.59aLeSDok ... I will find a win machine
12:52.21ymerejtgood morning diogene31
12:53.04pH5aLeSD: yeah, something like that. It was only a guess because I remembered something I read some time, but I don't know what I'm talking about here :)
12:53.06pH5hi diogene31
12:56.43goxboxliveBabelO: there?
12:56.58BabelOgoxboxlive: yes
12:57.08goxboxliveHave you seen what tsdogs have made?
12:57.16diogene31ymeretj: The solution about git and kernel: disable in .config CONFIG_LOCALVERSION_AUTO
12:57.23BabelOgoxboxlive: yes i see your post ;
12:57.36BabelOgoxboxlive: that s good he use model / view ;)
12:57.47goxboxliveNow, we only need the terminal. And YES you need that on a linuxphone .-)
12:58.03goxboxliveBabelO: have you still problem configure?
12:58.08diogene31pH5: Remember my blocking msleep. The solution was in the MCI driver, which had a task scheduled while MCI was not yet up. With a little delay, resume works like a charm :)
12:59.35BabelOgoxboxlive: yes, still same problem, i have to reuse ymerejt work, MichouX|NA is working on it too
13:01.19goxboxliveBabelO: I checked out the source today for building a imagre for my NEO and it build just fine.
13:01.32goxboxliveEven with my local addons like imlife, roadmap and so :-)
13:02.38BabelOgoxboxlive: ok
13:07.43*** join/#htc-linux Marex-notebook (n=marex@vasut.kolej.mff.cuni.cz)
13:07.52BabelOgoxboxlive: your latest build is with gstreamer ? or cruxus ?
13:09.21goxboxliveBabelO: now with cruxus. Gstreamer didnt work that well.
13:09.59goxboxlivethough i miss mp3 support iwith cruxus. Is there a way we can add libm,ad to Qtopia somehow?
13:15.20BabelOgoxboxlive: did you try the screenshot app ? should work now
13:15.44*** join/#htc-linux cods (n=cods@rsbac/developer/cods)
13:17.06goxboxliveBabelO: yes i tested, and i had the same bug as with the one from 4.2.x source. It might be that i configure with DEPTH, 16,32 . I have just builded a image for my neo without that setting so i wll test it when i get how from work.
13:27.50goxboxliveymerejt: still here?
13:28.00ymerejtgoxboxlive, yes
13:28.14goxboxliveHave done some progress since yesterday :-)
13:28.26goxboxlivenot personaly, i mean coding for UNI :-)
13:28.41ymerejtwhat kind of progress ?
13:28.56goxboxliveHave you started to look at a terminal for it?
13:29.06goxboxlivein example
13:29.34goxboxliveif you have, check out with tsdog in case he is also trying to make something.
13:30.01goxboxliveBut i guess he is working further on his filemanager to expand the functions in it
13:30.41ymerejtgoxboxlive, there is too much problem with the uni so I prefer to work on bugs before add applications
13:30.58ymerejtgoxboxlive, at this time I'm working on the airplane mode feature
13:31.37BabelOgoxboxlive: what i need to build with cruxus ?
13:34.02goxboxliveymerejt: ok :-)
13:34.11goxboxliveBabelO: just the qtopipa source
13:34.50pH5diogene31: the pxamci? or do you have a proper controller.
13:39.02BabelOgoxboxlive: why i have these error ? WARNING: Failure to find: qlock.cpp
13:39.13BabelOand some others
13:40.05goxboxliveBabelO: i have the same error, but it configure anyway
13:40.30goxboxlivepastebin and let me see the error
13:41.40BabelOgoxboxlive: http://en.pastebin.ca/852005
13:44.25goxboxliveBabelO: that look familiar, but it configure anyway.
13:52.40BabelOgoxboxlive: yes compiling now, my toolchain was too old ;)
13:53.29goxboxlive:-)
13:56.52goxboxlivewell enough for now. Going home. bbl
13:57.02BabelOme too bbl
14:32.48*** join/#htc-linux vekexasia (n=gdf@host164-5-dynamic.0-87-r.retail.telecomitalia.it)
14:33.06vekexasiaHi... can i help someone with htc p3600?
14:33.14vekexasiai wanna help developers to set it up
14:34.27*** join/#htc-linux AlGe (n=alge@chello080109231226.4.uni-klu.teleweb.at)
14:34.33vekexasiahei?
14:38.47*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
14:44.17*** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net)
14:50.58aLeSDjust a question ... if I port the kernel on my device ... after the use of android is quite possible , isn't it  =
14:51.00aLeSD?
14:51.32vekexasiais android working?
14:56.52*** join/#htc-linux goxboxlive (n=goxboxli@81.80-202-132.nextgentel.com)
15:06.32goxboxliveymerejt: testing latest changes now.
15:09.19BabelOaLeSD: we can now for the moment, you have to wait official annouce in March
15:09.32BabelOs/we ca,/we can not/
15:09.40BabelOs/we can/we can not/
15:09.52BabelOgrhh to much error
15:10.46*** join/#htc-linux MichouX|NA (n=michoux@mir31-3-82-234-54-233.fbx.proxad.net)
15:15.43*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
15:16.03ymerejtgoxboxlive, so let me know if some problems happen
15:16.22goxboxlivei will
15:18.27goxboxliveymerejt: ljp told me that dynamic rotation will be added to the source soon.
15:20.25goxboxliveymerejt: i dont know if you remember, but the voicemail number is still not stored.
15:20.28*** join/#htc-linux __a (n=albertsk@79.138.173.59.bredband.tre.se)
15:20.49AlGeBabelo: hi,
15:20.49AlGeany news regarding the BA kbd?
15:20.49__ahello
15:22.03BabelOAlGe: oups, i forget to test my change :=
15:22.10vekexasiasomeone can tell me something for htc p3600?
15:23.47BabelOvekexasia: only thing i know, it is similar to htc hermes
15:24.00BabelOvekexasia: and developpment is still in progress
15:24.09BabelOi mean very similar ;)
15:25.37AlGeBabelo: now I have mapped all keys (gpiodev2 patch) below 128 and can use them with gpe vie keylaunchrc, thanks again
15:27.36BabelOAlGe: for key up, i ve seen a patch somewhere
15:27.58BabelOAlGe: does wifi works well ?
15:28.06*** join/#htc-linux psokolovsky (n=psokolov@nat-users.is.ua)
15:28.15AlGeyes, with some restrictions
15:28.30psokolovskyHi!
15:28.36__ahello everyone! could someone give me some pointers on to how to get the current
15:28.40__astuff running
15:28.45psokolovskypH5: So, where's that imagemagic-native recipe?
15:28.53BabelO__a: on what device ?
15:29.12BabelOhi psokolovsky
15:29.27*** join/#htc-linux jeanseb (n=jeanseb@gazypan.dyndns.org)
15:29.30BabelO__a: check here http://wiki.xda-developers.com/
15:29.34__athe tytn! i got the three files in a folder, but it wont start successfully
15:29.40pH5hey psokolovsky, does not exist
15:29.47psokolovskynice ;-)
15:29.59__ai get /bin/sh: cant access tty, job control turned off
15:31.07AlGeBabelo: I have a wifi setup with several APs doing WDS, so I need some sort of roaming/reconnecting. This does not work flawlessly with the acx driver. my other devices are all capable of this.
15:31.11__aBabelO: im reading the thread on xda already, thnx
15:33.20vekexasiaBabelO can i help with p3600?
15:33.38BabelOAlGe: phone part is working well ? sms also ?
15:34.06__ashould i rename zImage-20070709 to zImage??
15:34.19psokolovskypH5: maybe you know other small cmdline util to resize images?
15:35.25BabelO__a: depend of your default .txt file
15:35.33__aoh
15:35.52__aBabelO: of course, thank you
15:36.44pH5psokolovsky: nope. why do you want to rescale images during build time?
15:36.47BabelOvekexasia: Kevin2 and cr2 are working on it
15:37.10psokolovskypH5: weekend project. porting a game to qvga ;-)
15:37.11Kevin2Hi
15:37.17psokolovskyHi!
15:37.21pH5hi Kevin2
15:37.23BabelOhi Kevin2
15:37.24ymerejtgoxboxlive, very good news (dynamic rotation)
15:37.35goxboxliveyes
15:37.41BabelOKevin2: did you see hp 514 hretlog ?
15:37.45Kevin2HI all.
15:38.00pH5psokolovsky: maybe we have to add imagemagick-native after all. but I didn't do it on purpose for the openmoko qvga icon theme because that is an ugly hack.
15:38.01Kevin2BabelO - the one from a couple of hours ago?
15:38.06ymerejtgoxboxlive, voice mail number : it's in my bug list but with a very low priority ;)
15:38.08BabelOKevin2: yes
15:38.20psokolovskypH5: ok, trying to build that
15:38.21pH5yet there is still no upstream qvga support..
15:38.39goxboxliveymerejt: ok
15:38.41Kevin2BabelO: Yeah.  Wince isn't letting the phone get "Kernel mode" status.
15:39.16Kevin2See line 141 and 145 of the log.
15:39.34psokolovskypH5: also, do you think we'd be able to use xrandr to get virtual VGA screen on X?
15:40.02pH5psokolovsky: on qvga devices?
15:40.03Kevin2aLeSD: The last time this happened - the phone needed to be unlocked in order to run haret.
15:40.08psokolovskypH5: yep
15:42.30pH5psokolovsky: sounds like madness. I like it :)
15:43.25psokolovskypH5: what you mean madness? wince offers that (via 3rd party apps). do you say that X is behind that?
15:45.12pH5psokolovsky: are you talking about downscaling 50% or panning in a larger virtual framebuffer?
15:45.36psokolovskypH5: about both, but downscaling is more useful of course ;-)
15:46.53Kevin2vekexasia: What is p3600?  Is that Hermes?
15:48.04pH5I don't think we have that right now, at least not in the upstream kdrive sources. But there are probably patches floating around, and it shouldn't be too hard to implement. The OLPC guys already did some more complicated post-processing in Xephyr to emulate their screen (http://www.manucornet.net/pub/olpc/xephyr_swizzle.diff)
15:48.42BabelOgoxboxlive: what do you use an empty qconfig.h ? it is no more neeeded
15:49.46psokolovskypH5: well, yep, I'm mostly interested if xrandr supports such a thing on protocol/paradigm level. But it seems, no, so it's really "mode-setting", not "resize" extension ;-\
15:49.48Kevin2vekexasia: Still around?
15:49.58BabelOKevin2: it is trinity ( hermes without keyboard)
15:51.00Kevin2I haven't checked the logs from the last week, but as far as I know, we still need an SD driver for ATI chip.  cr2 was working on that.
15:51.58goxboxliveBabelO: if i skip it i dont get the mousepointer. It's good to have when dealing with maps and with vnc
15:52.06vekexasiaBabel0
15:52.19vekexasiakevin2 i'm here
15:52.24BabelOgoxboxlive: in VNC you can ask back a mouse pointer
15:52.27vekexasiaKevin2 trinity
15:52.53BabelOgoxboxlive: and i think there is a special class in qtopia to activate mouse pointer on demand
15:53.05goxboxliveBabelO: yes, but what about with maps. And also if you connect a BT mouse to it (wich i have done) it great to have.
15:53.18vekexasiaKevin2, Are you working on hermes and trinity or only hermes?
15:53.48Kevin2vekexasia: I only have Hermes.  cr2 has both Trinity and Hermes.  Though, according to cr2 the phones are very similar.
15:54.01vekexasiayeah
15:54.10vekexasiai could try to set up linux on trinity
15:54.13vekexasiacan i?
15:54.57vekexasiaHOw many bricks possibilitys ? :P
15:55.08pH5psokolovsky: xrandr is only a protocol to handle the resizing of the virtual screen and location/orientation of the output viewports given their available modes. which modes are supported by the output device and how they are handled is up to the implementation. so scaling is orthogonal to xrandr as I understand.
15:55.08Kevin2Yeah, I think the hermes kernel might boot on Trinity.  I put some early kernels up at: http://www.handhelds.org/~koconnor/HTCHermes/
15:55.28*** join/#htc-linux ImCoKeMaN (n=imcokema@pool-71-173-158-85.hrbgpa.fios.verizon.net)
15:55.43vekexasiaKevin2
15:55.53vekexasiawhat is the best one of thoose?
15:56.04Kevin2http://www.handhelds.org/~koconnor/HTCHermes/herm-linux-20070930.exe
15:56.12vekexasiaok
15:56.18Kevin2It's a bit dated, later today I'll upload a new one.
15:56.29psokolovskypH5: ok
15:56.37vekexasiaAnd ... there is some problems?
15:56.40vekexasiasorry
15:56.48vekexasiaI just launch the exe?
15:56.56ImCoKeMaNso i haven't been in here for a while is android something that is interesting to people here?
15:57.15Kevin2vekexasia: Yes, just launch the exe.  The main problem is that we can't access the SD card or any permanent storage.
15:57.24vekexasiaso
15:57.36vekexasiai'll have a working phone?
15:57.45vekexasiawithout sd?
15:58.02Kevin2vekexasia: No.  We should be able to get the phone software to run - but SD must come first.
15:58.20vekexasiakevin2
15:58.29vekexasiawhen i launch the exe what does i will se?
15:58.44Kevin2If it works, you should see the linux kernel boot.
15:59.05vekexasiaok
15:59.13vekexasiaand when it booted
15:59.14vekexasia?
15:59.43vekexasiaherm-linux.exe is not a valid win32 application
16:00.17Kevin2vekexasia: You need to copy the .exe to the phone and run it under wince.
16:00.26vekexasiaahn ok
16:00.32*** join/#htc-linux marmotta (n=skodde@85-18-136-66.fastres.net)
16:00.33vekexasiai'm going
16:01.14Kevin2vekexasia: If it boots, you should be able to telnet to the phone over USB.  However, there isn't that much you can do with it yet.
16:01.38vekexasiaok
16:01.51vekexasiajumping to kernel
16:02.09vekexasiait is freezed
16:02.38Kevin2vekexasia: Okay.  We need to wait for cr2 then.
16:03.01vekexasiaok
16:03.19vekexasiais it a problem if i'm running wm6?
16:03.42*** join/#htc-linux __a (n=albertsk@78.156.197.244.bredband.tre.se)
16:03.55Kevin2vekexasia: I don't think so.
16:04.11vekexasiaim going to relaunche the kernel
16:05.03*** join/#htc-linux dcordes (n=dcordes_@unaffiliated/dcordes)
16:05.09vekexasianothing
16:05.16vekexasiait freezes at jumping to kernel
16:05.47vekexasiakevin2...
16:06.00vekexasiaonly relnet?
16:06.00__ahi again!
16:06.02vekexasiatelnet?
16:06.48__aim trying to get usbnet working on my mackbook, but i cant get it working :/
16:06.49Kevin2vekexasia: The .exe only supports telnet at this time.  We need the SD card to really run things.
16:06.55__ausing this: http://www.dsitri.de/wiki.php?page=AJZaurusUSB&referer=AJZaurusUSB
16:07.07vekexasiaok
16:07.27vekexasiathen we will have something grafical?
16:07.41__acould i get some pointers please?
16:07.58dcordeshi Kevin2, others
16:08.00Kevin2Yes - once we get SD working - then getting gpe, openmoko, or qtopia should be a snap.
16:08.16vekexasiaandroid?
16:08.33vekexasiahi dcordes
16:08.35dcordesvekexasia, any userland you want I guess
16:08.37vekexasiaplease dont sighs
16:08.42dcordesalso android hello gooogle
16:08.52vekexasiai luv google :P
16:09.05vekexasiakevin2 could you please write in pvt?
16:09.08ImCoKeMaNseems like android should be a help for the titan
16:09.22dcordeslol
16:09.34dcordeswhy in hell should it hel you with kernel development?
16:09.49dcordesgit.android.com but that's about it
16:09.52ImCoKeMaNwasn't their kernel for a msm7500?
16:09.57dcordesno
16:10.06dcordesmsm7200a but very similar though- almost the same
16:10.10Kevin2Hi dcordes.
16:10.14dcordesandroid is rather the userland
16:10.32ImCoKeMaNso closer to kaiser
16:11.14dcordeshoping it won't depend on correct dma functionallity
16:12.41vekexasiakevin2?
16:13.19Kevin2vekexasia: Are you getting my direct IMs?
16:13.28vekexasiayes
16:13.43vekexasiai'm trying to write you on query
16:14.04ImCoKeMaNoh swetland is: Linux Kernel Lead, Android Project, Google
16:14.51__asweetness! got it up and running :D
16:15.37dcordesImCoKeMaN, I didn't know any titan people were hanging out here
16:15.38Kevin2vekexasia: If you're not seeing my private messages, I don't think we can chat privately.
16:15.54dcordesImCoKeMaN, did you coke with IRQs/GPIOs yet?
16:16.03ImCoKeMaNyeah i haven't been back for a while, but i helped cr2 get some titan info earlier
16:16.18ImCoKeMaNhaven't done much of anything there
16:16.50dcordesthat is the memory map, right?
16:17.04dcordesget latest haret/haretconsole and lookup some irqs
16:17.25ImCoKeMaNi just started getting haretconsole and then dropped off the face of htc-linux irc
16:17.25dcordeshttp://www.handhelds.org/moin/moin.cgi/HaRET_20Documentation
16:17.35dcordeshow dare you
16:17.37vekexasiakevin2 try it now
16:17.54ImCoKeMaNwas in texas 3 months had lots of free time, now back to real life again
16:17.56dcordesbest upgrade both console and haret.exe
16:18.13dcordesreal life== reverse engineer titan
16:18.48ImCoKeMaNhmm i didn't know that
16:19.03ImCoKeMaNbout bout a vogue?
16:19.11ImCoKeMaNerrr, how bout
16:20.20dcordesa vogue?
16:21.47dcordesbbl
16:24.36ImCoKeMaNyeha its a sprint touch
16:38.04AlGeBabelo: BA phone works, but I have to ensure everytime I test it that "output mux" is set to "mixer", will have to look into the mixer initialisation of gomunicator (which currently does not work on BA) to add the proper alsa call.
16:38.54AlGesometimes sound quality (speaker in BA) is so bad that you only hear some cracks, sometimes it was good.
16:39.51AlGei have to recheck if this is reproducable: good sound when calling out from BA, bad when calling into BA.
16:40.32BabelOAlGe: maybe pH5 have an idea about this issue
16:40.43BabelOAlGe: maybe a wrong clock setting
16:41.10AlGecould it also be related to wrong codec selection? if this is selected at all?
16:42.56BabelOAlGe: no there is only uda1380, but did you check rate selection ?
16:43.12BabelO32Khz for example
16:43.41AlGeas a next step I'll try to add a volume-widget to gomunicator which sets phone volume via AT+CLVL...
16:44.13AlGei'v seen same hz in alsamixer but didnt change ... lets try
16:45.18AlGebtw. I still get random wakeups evne with gomunicator closed and gsmd shut down in suspend scripts
16:46.30BabelOAlGe: when you go to suspend, you need to set at+creg=0 or something like
16:46.55AlGethis can be missing, you are right.
16:46.56BabelOAlGe: because unsolicited event like network level change can wakeup the device
16:47.20BabelOand don't forgetr to set it back ;)
16:47.36AlGebut will a get a call in suspend mode with this setting?
16:48.21AlGegomunicator even has some suspend function, where it cloeses the opend ttyS0 device.
16:48.41BabelOAlGe: yes creg is not related to this you should continue to receive the ring event
16:48.46AlGemaybe i have to add creg=0 there
16:50.12BabelOAlGe: it is not 0 it is CREG=1
16:52.06AlGegomunicator does closePhone() on suspend ... which does hangupPhone() ... which only does ATH0 if phone is not idle.
16:52.20AlGeso I add at+CREG=1 there
17:00.44AlGeIf I grep the gomunicator source, I only find CREG=1 requests, so are you sure I have to set it to 1 because nothing else is done yet?
17:03.10*** join/#htc-linux lieter (n=lieter@a62-251-100-248.adsl.xs4all.nl)
17:05.22Kevin2diogene31: Hi.  Did you ever figure out why you were not able to trace?
17:07.17AlGeBabelo: btw: maybe i didnt understand this correct until now: there is no sense in running gsmd and gomunicator in parallel?
17:07.17AlGethe only thing that I need from gsmd is that it initializes the n_tihtc line discipline. am i correct? (my gomunicator handles /dev/ttyS0 raw without embedding command in 0x2.. frames)
17:09.29AlGebut I could add this line discipline initialization similar to the tihtc_test.c example to gomunicator?
17:09.29AlGehaving this initialized helps me then running minicom for testing, and maybe chat initialisation for pppd.
17:10.26Kevin2diogene31: If not, I thought of another thing to try - run "loadfunc coredll VirtualCopy", then "set insn 0xaddr", and then "wi 10" -- I think this will trace all calls to VirtualCopy.
17:15.44*** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-24bdbd402c8f86f8)
17:33.36BabelOAlGe: yes you just need the correct line discipline setting , do you want the code ?
17:37.58AlGeyes, but I could also just read the sample code :-)
17:39.24BabelOAlGe: yep
17:39.44BabelOhey qtopia just finish to build ;)
17:40.06*** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
17:42.17diogene31Kevin2: Hi, I will try in a few minutes, good idea.
17:43.14diogene31Kevin2: And no, I did not advance, I had to disassemble the rom bootstrap to make the suspend/resume work under linux. Sorry.
17:47.13*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
17:55.23BabelOdoes someone have the link to g3.c driver ?
17:59.03*** join/#htc-linux SmallR2003 (n=SmallR20@79-74-6-165.dynamic.dsl.as9105.com)
18:12.02diogene31Kevin2: Did the test. No call to VirtualCopy. This really drives me mad. It's illogical what happens on the MIO.
18:13.37Kevin2diogene31: Okay, thanks for testing.  I'm keeping notes, so I'll put together a wiki page on all the different things we tried.
18:16.46*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
18:22.31bernt<PROTECTED>
18:23.55berntI've got my HTC Universal :-) -- now I can use my BA for normal work and the universal for linux!
18:24.41berntThere was a page describing how to set up a bootloader .. does anybody remember it?
18:25.06BabelObernt: hi, a boot loader for what ?
18:26.08berntthis was a little bit like grub or lilo. It was possible to change to a root device and so on.
18:26.44berntand I think it could also boot windows ... not sure ...
18:31.44diogene31Kevin2: Excellent, I love documentation :)
18:40.55AlGeBabelo: I've converted the test_tihtc program to a little utility that can be used like this
18:40.55AlGe./settihtc /dev/ttyS0 atcmd
18:40.55AlGeI think I put this in my init scripts, so I dont need to add it to any program like minicom, gomunicator etc.
18:40.55AlGesource uploaded at http://www.algepop.net/users/alge/angstrom-ba/settihtc.c
18:42.27BabelOAlGe: yes i ve same somewhere on my harddrive, but not used, because i integrate in qtopia ;)
18:42.35paulproteusMorning, all.
18:43.03BabelOAlGe: good tool , i download it ;)
18:47.36goxboxliveBabelO: can that be used in Qtopia as the mux instead if the one we are using now?
18:48.06BabelOgoxboxlive: sure, it is just a settings
18:49.50goxboxliveBabelO: have you tested it?
18:50.19BabelOno i just finish to recompile a new kernel and qtopia
18:50.46goxboxliveOk, maybe the modem will work now with that file as mux
18:51.19BabelOgoxboxlive: no
18:51.21goxboxliveok
18:51.32goxboxliveBabelO: have you install kde4 yet?
18:51.59BabelOgoxboxlive: it is the same that what i ve done
18:51.59BabelOalready made this test before
18:52.20goxboxliveok
18:54.47BabelOgoxboxlive: yes i ve it on my desktop, but only an old version
18:55.24goxboxliveBabelO: ok, i tested it once but' thats a while ago. I am think of install it soon. Looks good
18:57.18AlGeBabelo: are you sure with +CREG=1, not 0? because the original gomunicator never sets CREG to other values than 1.
18:58.11BabelOAlGe: ah, then maybe it is 0, need to look at AT GSM documentation
19:11.42BabelOAlGe: can you post me your dmesg output ? on pastebin.ca
19:12.34AlGein what Ba state? after boot?
19:12.52BabelOAlGe: yes result of dmesg after boot
19:14.42AlGewell, will do a fresh reboot, but wait some minutes, I'll just finish to add "gettihtc" feature to sett
19:14.53AlGes/sett/setihtc/
19:15.11AlGeand im currently compiling on the ba
19:15.16BabelOAlGe: thanks a lot, my kernel freeze just after pcmcia
19:16.36*** join/#htc-linux Pryan (n=pryan@84.125.94.235.dyn.user.ono.com)
19:30.12*** join/#htc-linux aLeSD (n=alex@79.144.175.194)
19:30.15aLeSDhi all
19:30.29aLeSDI have my smarphone free in the end and I can run haret
19:30.38AlGeBabelo: compiling finished, now you can symlink: ln -sf settihtc gettihtc and:
19:30.38AlGeroot@htcblueangel:~$ ./gettihtc /dev/ttyS0
19:30.38AlGeline discipline for /dev/ttyS0: TIHTC
19:30.38AlGetihtc channel: atcmd
19:30.38AlGewill reboot now.
19:30.40aLeSDBabelO
19:32.23aLeSDI have a problem with the screen : I can see only points
19:32.31cr2hi
19:32.42aLeSDis there the possibility to pass to the kernel something ?
19:32.43cr2a lot to read today :)
19:32.57BabelOhi cr2 :)
19:33.02cr2hi BabelO
19:33.18BabelOaLeSD: which kernel did you try to boot ?
19:33.31aLeSDlinwizard kernel
19:34.00aLeSDsomething works ... I mean if I connect and deconnect usb something happens in the screen
19:34.08aLeSDthe problem is that I can't read
19:34.09aLeSD:)
19:34.13BabelOaLeSD: you need some tweak
19:34.30aLeSDBabelO: what do u mean ?
19:34.43BabelOaLeSD: need to found fb adress, gpio for lcd and some other thing
19:34.46aLeSDbut I'm ahppy
19:35.20aLeSDBabelO: so it's not a problem of resolution etc etc
19:35.34aLeSDBabelO: do u know how could I do ?
19:36.18BabelOaLeSD: with haret
19:36.35BabelOaLeSD: haret print fb adress on start
19:36.47BabelOis it 0x2000000 for you ?
19:37.37aLeSDVideo buffer at 49400100 sx=176 sy=220 mx=44 my=36
19:37.37aLeSDVideo Phys FB=20000100 Fonts=11891064
19:37.55aLeSDHaRET(1)# set MTYPE 6666
19:37.55aLeSDHaRET(2)# set KERNEL "zImage"
19:37.55aLeSDHaRET(3)# set INITRD "initrd.bz2"
19:37.55aLeSDHaRET(4)# set CMDLINE "root=/dev/ram0 ramdisk_size=20504 init=/linuxrc"
19:37.55aLeSDHaRET(5)# bootlinux
19:38.06aLeSDthe last is my defaul.txt
19:38.17BabelO<PROTECTED>
19:39.29aLeSDBabelO: ?
19:39.46aLeSDthat is the start of my FB device no ?
19:40.07BabelOaLeSD: yes, i just check what was my buffer adress
19:41.09aLeSDBabelO ... in that point linux isn't loaded ... I mean after loading initrd ... the screen change and I can't read
19:41.26aLeSDI'm asking ... I mean
19:42.08AlGeBabelo: I have to reboot once more, this time my dmesg is flooded with " asic3_irq_demux: interrupt processing overrun" lines (this happens from time to time, i can not see a negative effect of it btw)
19:47.47BabelOaLeSD: i have same fb adress, you have lot of work to do
19:48.26aLeSDBabelO: I have to compile the kernel or only to work with haret ?
19:48.51aLeSDBabelO: I have time ... :)
19:49.29BabelOaLeSD: both wizard kernel is a good start, but you need haret to found  gpio and other thing
19:49.58aLeSDBabelO: have you some seconds to explain me  some stuff ?
19:50.10BabelOAlGe: asic3_irq_demux: interrupt is not a problem else use dmesg -s 128
19:54.13BabelOaLeSD: what thing ?
19:54.28cr2BabelO: which g3.c are you looking for ?
19:54.38dcordeshi cr2
19:54.42cr2aLeSD: do you sign haret ?
19:54.53BabelOcr2: want to try on my ba
19:55.00aLeSDcr2: no I unlock registy
19:55.10cr2dcordes: hi. is there something in the msm git ?
19:55.30cr2aLeSD: so you allow nonsigned binaries to run ?
19:55.41aLeSDBabelO: well ... haret do more than load the kernel isn't it ?
19:55.43dcordescr2, depends on when you last took a look: 3 days ago
19:55.46aLeSDcr2 yes
19:55.54dcordesusb smd rpc
19:56.23cr2BabelO: the typhoon code in the universal CVS subdir. but it needs a minor change
19:56.24dcordesI'm looking like two times every day :)
19:56.32cr2dcordes: ok :)
19:57.28cr2vekexasia: you can't run the hermes-linux.exe on trinity directly, because trinity has a different mach_type. That's why it hangs.
19:58.05aLeSDok ... someone could explain me what haret do more than load the kernel ?
19:58.21BabelOcr2: ok got it, docg3.c
19:58.22dcordesaLeSD, are you the guy I suggested to try and boot linwizard?
19:58.23aLeSDI mean I'm a nooby in these things ... :|
19:58.38cr2BabelO: can you test it on BA ?
19:58.42aLeSDdcordes: it's my firt day here
19:58.55BabelOcr2: sure, once i found why  my ba don't want to boot ;)
19:59.03cr2hehe.
19:59.25cr2BabelO: there are some quirks about docg3.c
19:59.37vekexasiacr2
19:59.49vekexasiacoould you boot the kernel on p3600?
19:59.50aLeSDok ... in my opinion haret configure all the memory and pass the address (the fb too) to the linux kernel
20:00.41aLeSDok ... the screen works good until
20:01.18aLeSDSutting down hardware is displayed ... then there is the ......................... linux bar to load
20:01.59aLeSDand then all green points for the screen ... the line shutting down hardware doesn't disappear on the top
20:02.01cr2vekexasia: i don't have trinity myself
20:03.09aLeSDthe question is . the 'first' problem for the fb is in haret or in the kernel ?
20:05.29vekexasiacr2 noone is developing that?
20:08.38cr2vekexasia: it's relatively easy to do, but does not bring anything useful.
20:12.06dcordesaLeSD, what's youd device?
20:12.21aLeSDdcordes: HP ipaq 514
20:12.56dcordesdoes it have a htc name?
20:13.44aLeSDcore=ARM926T name=OMAP850 cat= vend=Texas Instruments Inc
20:13.57aLeSDI don't undestand htc name
20:14.16cr2aLeSD: you should dump the SPL.
20:20.11aLeSDcr2: ehm SPL ?
20:20.53dcordessecond stage bootloader tehe
20:21.34aLeSDsoo it's haret stuff
20:21.46aLeSDok
20:22.45dcordesaLeSD, http://www.handhelds.org/moin/moin.cgi/HaRET_20Documentation look at this brand new haret doku
20:23.13aLeSDok dcodes thanks
20:23.28AlGeBabelo: currently I have the asic3 irq overruns in every boot, so I loose the very first seconds in dmesg.
20:23.29AlGebut anyway, here it is:
20:23.29AlGehttp://www.algepop.net/users/alge/angstrom-ba/dmesg-boot.txt
20:23.51*** join/#htc-linux goxboxlive (n=goxboxli@81.80-202-132.nextgentel.com)
20:26.15cr2AlGe: which "real" irq overruns ?
20:27.01AlGelots of these lines:
20:27.01AlGe<3>[   11.230000] asic3_irq_demux: interrupt processing overrun
20:27.11AlGewhat si you mean by "real"?
20:27.58AlGes/si/do/
20:27.58BabelOAlGe: ok it miss information i need, on my old kenrle i remove printk of asic3_irq_demux
20:28.32AlGebut simetimes it doesnt happen, i'll retry some times.
20:28.50AlGeor can I just enable syslog writing to file?
20:29.42AlGecan you give me a hint where it is in the kernel source (asic3_irq_demux)?
20:31.14aLeSDdcordes: to connect to haret to have a console I need a wifi connection in WMobile ... by that connection to my router can I connect to the haret console ?
20:31.32dcordessure
20:31.43dcordeshtc and host only have to be in the same network
20:32.02aLeSDdcordes: ok ... I have the first task
20:32.04dcordesthen you hit listen in haret.exe and in the haretconsole folder run ./console ip.of.htc
20:32.09goxboxlivecr2: i have added the icons now.
20:32.18goxboxliveThey where not that cool :-)
20:32.55BabelOAlGe: kernel26/drivers/mfd/asic3_base.c
20:32.56cr2AlGe: the irq_demux is in asic3_base.
20:33.05dcordeshow does on dump the whole memory again?
20:33.12dcordesI mean create a memory map
20:33.16BabelOcr2: lol, you are not speed enought ;)
20:33.38AlGefound it earlier myself :-)
20:33.38cr2AlGe: the demux code should find the actual=real irq from the multiplexed one.
20:33.42cr2lol
20:33.51dcordeslol BabelO you not fast enough either
20:34.09cr2goxboxlive: i agree, but they are better than the b&w letters.
20:34.44dcordesaLeSD, what's your timezone?
20:34.53goxboxlivecr2: But still there are two letters.
20:35.06goxboxliveG and L or something, hold on i check
20:35.31goxboxlivecr2: G and H
20:36.02goxboxlivecr2: Have you seen tsdogs latest work?
20:38.54cr2goxboxlive: yes, it looks very nice :)
20:38.57dcordesaLeSD, have to go now sorry. I can help you tomorrow
20:38.59dcordesbye
20:39.15cr2goxboxlive: G and H ? then you are missing some icons
20:39.16aLeSDdcordes thanks a lot
20:39.18aLeSDbye
20:40.09goxboxlivecr2: ok
20:40.36goxboxlivecr2: you can test it, i have uploaded the new opt folder with tsdogs work in it.
20:40.49goxboxlivealso the roadmap patch ith the iocns in it.
20:41.21goxboxlivei copied every single png file from the roadmap source so i dont understand what i can be missing
20:43.09AlGecr2: I'll printk the irq in question with
20:43.09AlGeif (iter >= MAX_ASIC_ISR_LOOPS)
20:43.09AlGe<PROTECTED>
20:44.15cr2goxboxlive: can you post the link ?
20:44.24cr2goxboxlive: i'm looking in the source.
20:44.48goxboxlivecr2: of course: http://www.linuxtogo.org/~goxboxlive/htcuniversal/tmp/
20:45.08goxboxlivecr2: while you look in the source, take a look at the cam driver meanwhile :-D
20:45.13cr2goxboxlive: i have 12 rm_*.png files
20:46.02goxboxlivecr2: http://pastebin.no/3083
20:46.29goxboxlivecr2: is it posible to make garmin map or tomtom map work with roadmap?
20:47.00cr2garmin map - yes, tomtom no.
20:47.05goxboxliveI need a map with more small roads in oslo
20:47.22goxboxliveSo if i get a scnadinavia map, i can concert it some how?
20:47.37goxboxlives/concert/convert/
20:47.41cr2goxboxlive: if you get the metroguide/cityselect.
20:48.12goxboxlivecr2: for what devices? pockepc?
20:48.16cr2goxboxlive: the Oslo map you have now was converted from the OSM tile for garmin
20:48.36cr2goxboxlive: no, it's a CD/DVD for garmin devices.
20:48.42cr2i'll post you some links.
20:50.52goxboxlivecr2: will this one work?
20:50.54goxboxlivehttp://www8.garmin.com/cartography/mapSource/citynavntEurope.jsp
20:51.50cr2goxboxlive: i've converted your map from the whole world archive at http://wiki.openstreetmap.org/index.php/User:Computerteddy
20:52.22cr2goxboxlive: no. NT is bad (==format is not known)
20:52.37cr2goxboxlive: you need an older version (non NT)
20:52.41goxboxliveok
20:52.46goxboxlivei look further
20:53.35cr2http://www8.garmin.com/cartography/mapSource/citynavEuro.jsp ?
20:54.01cr2goxboxlive: you'd talk to kiozen. he is our garmin expert :)
20:54.38goxboxliveok
20:55.29cr2hehe. they are not cheap ;)
20:55.57goxboxlive:-)
20:56.23goxboxlivei wonder if it is just enough with an update cd/dvd
20:56.40cr2metroguide is cheaper http://www8.garmin.com/cartography/mapSource/MGEuro.jsp
20:56.50AlGeBabelo: I found an older dmesg (before I applied your gpiodev2 patch):
20:56.50AlGehttp://www.algepop.net/users/alge/angstrom-ba/dmesg-boot-old.txt
20:57.46AlGeore maybe it was already with the patch, dont remember
20:57.57cr2AlGe: pcmcia: request for exclusive IRQ could not be fulfilled. ??
20:58.15BabelOAlGe: thanks a lot, i check ;)
20:58.27goxboxlivethoose are just for major cities
20:58.36cr2i also do not like "acx copy_from_slavemem: warning!  destination not word-aligned!"
20:58.57BabelOcr2: it is normal, i ve not used shared irq for pcmcia driver
20:59.19cr2goxboxlive: do you still have an old file i've given you ?
20:59.33AlGebut it somehow works?
20:59.35cr2goxboxlive: maybe i'd compare OSM to that old file.
20:59.56goxboxlivecr2: yes i think i have it. The one i tested on my computer
21:00.20cr2goxboxlive: let me know if it has the small streets you need.
21:00.42cr2goxboxlive: if you drive through these streets , it's a good idea to upload the gpx tracks to OSM
21:01.33goxboxlivecr2: is gpx tracks the record option in gpsconsole?
21:01.39cr2BabelO: "ds2760-battery ds2760-battery.0: not enough power to charge" <- USB ?
21:02.24BabelOcr2: yes always same issue, we need to tel desktop to send 500mA mode
21:02.41cr2ok.
21:02.45cr2wtf is that ?
21:02.47cr2Alignment trap: keylaunch (1550) PC=0x402272d8 Instr=0xe5102004 Address=0x0001abd5 FSR 0x013
21:03.08BabelOcr2: gpe stuff i think
21:03.23cr2ok.
21:03.49cr2AlGe: check your apps before complaining about the kernel :)
21:07.04AlGecr2: yes USB
21:07.49cr2goxboxlive: don't think so.
21:09.12cr2goxboxlive: i've downloaded 2 files. what's next ?
21:09.34AlGealignment traps: yes, there are lots of them. I wonder why, shoudlnt the cross compilers (built by oe) generate correct code? but anyway, I got used to alignment traps in my early unix days when compiling stuff under OSF1 on my alpha :-)
21:10.32cr2AlGe: you can force alignment traps by weird pointer operation.
21:10.54AlGeBabelo: whats the "500ma" mode? can I set it to this mode when on USB and then it charges?
21:11.02cr2alignment traps badly hurt performance on alpha.
21:11.29BabelOAlGe: don't know how we do that, maybe google can help, or looking at other driver
21:11.30cr2AlGe: it's a deficiency in the pxa*_udc driver.
21:11.42BabelOcr2: what is g3 adress .
21:12.26cr2BabelO: http://wiki.xda-developers.com/index.php?pagename=HTC_DiskOnChip
21:12.57cr2the problem is that you have 8bit access.
21:13.13cr2and the driver does 32bit accesses.
21:13.32cr2the htc IPL works a bit differently, so we need some testing here.
21:13.53cr2we can also check the tffs.o :)
21:16.08BabelOcr2: ok how do we proceed ?
21:16.19cr2BabelO: compile it, and try to load.
21:16.23cr2first.
21:16.30BabelOcr2: already done ;)
21:16.36BabelOsince a while
21:16.42cr2try dd
21:16.46BabelOcr2: but i did ot change phys addr
21:16.49BabelOrecompiling
21:17.05cr2the base addr is a modules parameter.
21:18.37BabelOcr2: :( now i see this parameter
21:18.38*** join/#htc-linux aLeSD (n=alex@79.144.175.194)
21:18.54BabelOalready changed in default sources
21:20.56aLeSDI have an haret console!
21:21.03aLeSDunbelivable
21:21.33BabelO<3>[  159.040000] DoC_G3 ioremap failed
21:21.33BabelO<4>[  200.050000] Doc_G3: chip id=1000
21:21.39BabelOcr2:
21:21.46aLeSDis it possible to broke my device in haret console ?
21:22.02cr2BabelO: cat /proc/iomem ?
21:22.22cr2aLeSD: yes, but it never happened so far ;-)
21:23.01BabelOuh :(
21:23.05aLeSDcr2: do u know what I'm searching for ?
21:23.12BabelOcr2: after second load
21:23.15BabelO<3>[  159.040000] DoC_G3 ioremap failed
21:23.15BabelO<4>[  200.050000] Doc_G3: chip id=1000
21:23.15BabelO<4>[  267.510000] Doc_G3: chip id=200
21:23.15BabelO<4>[  267.510000] Starting flash stuff
21:23.15BabelO<4>[  267.520000] end
21:23.33aLeSDcause I don't know
21:23.49cr2aLeSD: dump the first megabyte of RAM
21:24.17cr2BabelO: do you have this address mapped already ? it's the only reason why it can fail.
21:24.20BabelOcr2: http://pastebin.ca/852479
21:25.37cr2BabelO: strange.
21:25.42aLeSD[V|P]DUMP <addr> <size> is this the command cr2 ?
21:25.56aLeSD[V|P] what does it mean ?
21:26.05aLeSDops
21:26.07aLeSDsorry
21:26.10cr2aLeSD: pwf paqspl 0x10000000 0x100000
21:26.36cr2V is virtual, P is physical.
21:26.47cr2DUMP is the screen hexdump.
21:26.49aLeSDwhy your commadn is different ?
21:27.18cr2pwf is PhysicalWriteFile
21:27.21aLeSDcr2: your command is to dump in a file ?
21:27.36cr2pwf name paddr size
21:28.26aLeSDmmm
21:28.42aLeSDI run it ... but I don't find paqspl
21:28.46aLeSDah
21:28.52aLeSDmaybe in the host
21:28.53aLeSD:)
21:29.31aLeSDmmm ...no
21:29.45cr2goxboxlive: G is 'gps position', H is 'hold'
21:30.00cr2aLeSD: it should be in the same directory as haret.-
21:30.02aLeSDyes
21:30.06aLeSDI found it
21:31.34aLeSDcr2: why I started from the 0x10000000 memory address ?
21:32.36aLeSDI mean for the first MB ...  pwf paqspl 0x0 0x100000 , no ?
21:33.19cr2aLeSD: the RAM starts at 0x10000000 on omap850
21:35.18aLeSDcr2: do u know where I can find more information on omap850 ?
21:36.43cr2aLeSD: it's not documented, so only from the kernel source.
21:36.58BabelOAlGe: still around ?
21:37.05cr2and here http://wiki.xda-developers.com/index.php?pagename=OMAP850_Registers
21:37.43cr2BabelO: btw, it seems that some "empty" regs in this list are the sim access.
21:38.01aLeSDwow
21:40.34aLeSDcr2: now the same question ... what I'm searching ... I don't know the answer
21:40.58aLeSDI mean I have a device that boot linux but the fb doesn't work
21:41.13AlGeBabelo: http://www.algepop.net/users/alge/angstrom-ba/dmesg-new.txt
21:41.13aLeSDand I have the first MB of its memory
21:42.48AlGeBabelo: this time there were no irq overruns, good for you, but I was waiting fir the irq number now to be printed out :-(
21:43.02cr2aLeSD: if you run 'strings paqspl', do you see something interesting ?
21:43.37*** join/#htc-linux jeanseb (n=jeanseb@gazypan.dyndns.org)
21:44.10BabelOAlGe: thanks, i need to apply patch i send to you
21:44.43cr2AlGe: why do you have hwuart twice ?
21:45.18aLeSDcr2:  a lots of errors ... win sucks
21:45.22cr2and with irq=0. afaik it's fixed in the newer kernels (irq=7)
21:45.30aLeSDNULL pointer exception and other stuff
21:45.59cr2aLeSD: where ?
21:46.37aLeSDAccess violation - no RTTI data!
21:46.38aLeSDAttempted a typeid of NULL pointer!
21:46.38aLeSDBad dynamic_cast!
21:46.46aLeSDin the memory
21:47.12aLeSDcr2: maybe u don't understand my position ... I don't know what I'm searching for
21:47.37aLeSDit's the first time I try something like that
21:47.47AlGeBabelo: at the bottom of my last dmesg you can see what happens if acx sees another of my WDS connected APs having all the same ESSID
21:48.04cr2aLeSD: ok, it seems like a haret problem.
21:48.34AlGecr2: I edited the dmesg and pasted two parts together, maybe I missed it somewhat, let me check
21:48.55aLeSDcr2: did I find a problem ?
21:48.57aLeSDwow
21:49.33aLeSDI was thinking that I was searching for memory address of the fb
21:49.54AlGecr2: yes, I dupliceted this line manually, sorry,
21:50.59aLeSDcr2 I'm using the cvs version of haret
21:51.33BabelOAlGe: ok, i can check with my access point
21:56.47*** join/#htc-linux TimRiker (n=timr@rikers.org)
22:03.48*** join/#htc-linux I (n=Jeff@ool-435452d6.dyn.optonline.net)
22:05.13*** join/#htc-linux legodude522 (n=Jeff@ool-435452d6.dyn.optonline.net)
22:07.32AlGeBabelo: now after reboot my settihtc tool doesnt work, that means it only worked in a session where gsmd was runnning at least once before, so it does something more?
22:08.29BabelOAlGe: it work for me here
22:09.18AlGeyes, reinstalled and started gsmd and it works. and now I can set/unset with my tool again.
22:09.44AlGeso I'm missing some little bit which gsmd does and I not
22:10.43BabelOAlGe: it is strnage your tool work for me without gsmd
22:12.02BabelOpH5: are you here ?
22:12.03AlGeyou have qtopia on it? maybe some other tool does what gsmd does which is needed as preparation
22:12.46BabelOAlGe: how look your at command return ? mine always haf a blank line before OK return
22:14.35AlGeIt depends how many apps are talking to ttyS0, sometimes minicom doesnt see saome output chars when gomunicator runs in parallel, but a new line is there anyway
22:14.59BabelOok like me...
22:15.34BabelOthis blank line seem to broke qtopia, or something else maybe
22:15.43BabelOMichouX|NA: around ?
22:16.35AlGeBabelo: tried it again: if i rmmod n_tihtc it starts again, settihtc doesnt help: it seems to work (I stepped through it with gdb) but it has no effect.
22:20.36BabelOgoxboxlive: i don't understgand your backlight powermanagement
22:27.49goxboxlivecr2: extract them to the root of your sd card. But either delete or remove your existing /opt folder.
22:27.56goxboxliveBabelO: my ??
22:28.17BabelOAlGe: you right, i use htcuniversal plugins and i see that polugin multiplexer always contain not needed line discipline
22:28.27AlGeBabelo: I was wrong: no newline in minicom:
22:28.27AlGeAT S7=45 S0=0 L1 V1 X4 &c1 E1 Q0
22:28.27AlGeOK
22:28.27AlGeATZ
22:28.27AlGeOK
22:28.33goxboxliveBabelO: i guess i dont understand it my self
22:28.49BabelOAlGe: what is your Radio ROM ? i want same
22:29.17BabelOgoxboxlive: sorry i forgot my question ;)
22:29.41goxboxlive:-)
22:29.44AlGeBabelo: this is the latest german tmobile 2003 rom, dont know the number, will look on next reboot
22:30.06BabelOAlGe: ok
22:30.23BabelOAlGe: i hope my issue come from that
22:31.25AlGewill reboot now anyway, I want to see the overrun msg again with my new kernel
22:31.44BabelOAlGe: do you keep dump_buffer in n_tihtc ? or did you just unset DEBUG
22:32.16AlGeit dumps anything into dmesg,if you mean this
22:33.44BabelOAlGe: ok, i just remove debug define
22:33.46AlGeI only unset DEBUG, yes, to get rid of the poll messges
22:36.21BabelOAlGe: i have to try your work now, to see if we have same working modem :)
22:38.23AlGeI'm still wondering about my nonworking settihtc, do you still have the old source (without te gettihtc stuff) maybe I messed something up?
22:39.49BabelOi have a 2Gb sd card, so i can keep both gpe and qtopia
22:40.15*** join/#htc-linux hollo_ (n=hollo@3e6b025d.rev.stofanet.dk)
22:40.30*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
22:41.30BabelOAlGe: yes i ve old source, but as i say there is always my old code in qtopia
22:43.01AlGeCould please mail it back to me? I want to diff.
22:43.02AlGesorry for abusing you as version control system :-)
22:50.09BabelOAlGe: let me check
22:51.16BabelOAlGe: uh , no i m not a good cvs ;) i ve this in sources if(strncmp(basename(pname),"gettihtc",8)==0)
22:51.21*** join/#htc-linux TeringTu1y (n=maarten@195-241-125-243.ip.telfort.nl)
22:52.09AlGewell, doesnt matter, I dont believe that it was different in that aspect.
22:53.24BabelOAlGe: all is setup correctly in your image from 21h55 12/01/08?
22:54.31AlGeYou should extract the alge-gpe.tar on top of it (it also cntains settihtc binary), and it still contains libgsmd ipkg, so remove it once booted
22:55.10AlGethis reboot i should see the asic3 irq number ....
22:55.40AlGe<3>[   31.330000] asic3_irq_demux: interrupt processing overrun: irq=26
22:56.46BabelOAlGe: you do %d ?
22:56.51AlGeroot@htcblueangel:~$ dmesg | grep asic3_irq_demux:| wc -l
22:56.51AlGe<PROTECTED>
22:56.59AlGe%d?
22:57.08BabelOAlGe: in the printk
22:57.47AlGeyes, see silence-asic3-demux.patch at algepop.net
22:59.04AlGeagain settihtc doesnt work, and ./tihtc_test fails too (I started from this sample)
23:04.57BabelOAlGe: did you try openmoko ?
23:06.18AlGenot yet, I tried gpephone-image, but was not so happy with it, maybe it is completely misconfigured for the BA, but nothing worked.
23:07.12BabelOAlGe: i ve made an image which work with ba for gpephone, but i change lot of thing in sources
23:07.27BabelOAlge : thing related to at command
23:08.26AlGeok, but for me even the touch screen did not work as I expected it. Or is GPE^2 targeted to touchscreenless machines?
23:09.43AlGeI would feel comfortable with gpe + some working phone application (thatswhy my gomunicatorpatches),
23:10.19BabelOAlGe: you can try openmoko-dialer + qvga patch in gpe
23:10.50AlGebtw. this time the asic3_irq_demux: msgs doesnt stop
23:11.25BabelOAlGe: last time , we notice that it stop when kp driver is loaded
23:12.27AlGei'll press some keys (yes I left the ba after boot untouched until now)
23:12.40AlGeand it seems to stop
23:12.56AlGeand i got a stuck key :-)
23:14.50BabelOgrhh, i can't set two password same
23:14.55BabelOwhy password needed
23:15.11AlGeuse the kbd
23:17.20cr2hehe. i'm reflashing uni back to wince5.
23:17.43AlGeBabelo: you missed the gpe-alge.tar file http://www.algepop.net/users/alge/angstrom-ba/gpe-alge.tar
23:18.18BabelOAlGe: no i download the Angstrom... same hour
23:18.41cr2BabelO: how do you load the docg3 module ?
23:19.15BabelOcr2: modprove docg3 doc_config_location=0x4....
23:19.58BabelOAlGe: are they not the same ?
23:20.23AlGeyou need both
23:20.45AlGeuntar the angstgrom and then add/replace some files from the smell gpe-alge.tar
23:20.56AlGes/smell/small/
23:20.58BabelOAlGe: ahhh ok
23:21.51BabelOi really need to fix touchscreen
23:21.56AlGesorry for not making a single file, but I would to automate this a little bit more
23:22.26AlGeYes touchscreen is not far from beeing unusable
23:23.05*** join/#htc-linux LunohoD (n=alex@e180066094.adsl.alicedsl.de)
23:28.52AlGeI rebooted again, and this time I noticed that the battery level was near to zero (remember my USB cable), maybe this is a reason for my strange n_tihtc behaviuor?
23:29.11AlGeor for the irq 26 overrun?
23:30.03BabelOcr2: what is the irq26 on ba ? i m not sure i understand how to calculate it
23:30.59BabelOcr2: we run with gpio2 patch
23:30.59*** join/#htc-linux the_sys0p` (i=the_sys0@gateway/tor/x-c35d82c119c0a083)
23:30.59AlGeI had the overruns also without gpio2 patch
23:31.50AlGebut I dont know the irq number for the old cases, but it happaned at the same stage imediately after booting, I think there was no difference
23:31.55BabelOAlGe: yes, i say that only for calcul ;) because with this patch, it simplify code, you can call asic3 or pxa gpio in the same way
23:32.16AlGeok
23:35.54cr2BabelO: i don't know about gpio2
23:44.02BabelOAlGe: Media player not working ?
23:44.17BabelOAlGe: i think you need cardctl on device
23:48.35AlGeyes, crashing, I already tried to debug, somewhere in the audio code where it selcts its alsa output
23:49.06BabelOi think i ve used xmms before ;)
23:49.15AlGewhat device?
23:49.22AlGe(cardctl)
23:50.01BabelOcardctl is the old tool to manage pcmcia
23:50.05AlGebut mpg123 is there
23:50.11BabelOso you can set on / off from gpe
23:50.53AlGeyes, I need pccardctl for the acx driver (see wlan0 settings in gpe-alge/etc/network/interfaces)
23:51.29BabelOAlGe: but if you use old tool cardctl, then in gpe->settings->pcmcia you vcan control wifi
23:51.45AlGeahh, now I understand
23:53.07BabelOAlGe: you are french ?
23:53.34AlGeno austrian
23:53.41BabelOok, i see my keyboard layout is ok
23:53.47BabelObut i understand why now ;)
23:55.35AlGewhat bb file will build me th old cardctl?
23:56.40AlGepcmcia-cs?
23:57.34BabelOyes pcmcia cardbus ;)
23:58.13AlGeyes, built, will upload it to my feed.
23:59.31AlGeif you use my alge*feed.conf files, especially http://www.algepop.net/users/alge/angstrom-ba/alge-htcblueangel-feed.conf you can install it in 1 minute

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