IRC log for #htc-linux on 20070524

05:20.44*** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
05:36.01*** join/#htc-linux Tjikkun (n=tjikkun@82-204-54-115.dsl.bbeyond.nl)
05:40.20*** join/#htc-linux akke (n=akke@88.82.38.195)
05:46.53*** join/#htc-linux _law_ (n=law@mail.stiftadmont.at)
06:24.23*** join/#htc-linux goxboxlive (n=goxboxli@195.159.97.196)
06:24.31goxboxlivemorning
06:31.35*** join/#htc-linux pleemans (n=peter@d5152D19B.access.telenet.be)
06:36.42*** join/#htc-linux sadeness_ (n=LamersIn@nat2.rlan.ru)
06:58.17Foxdiemorning goxboxlive, how goes?
07:10.16*** join/#htc-linux rmoravcik (n=rmoravci@adsl-d181.84-47-56.t-com.sk)
07:14.33*** join/#htc-linux psokolovsky (n=psokolov@82.193.98.15)
07:28.27*** join/#htc-linux TripleDES (n=sergio@109.pool85-50-159.dynamic.orange.es)
07:35.02goxboxliveFoxdie, just fine. U 2 i guess
07:50.39FoxdieI'm good too yeah
08:02.25*** join/#htc-linux WizMaui (n=WizMaui@62.112.90.250)
08:06.04*** join/#htc-linux DTMonterrey (n=evandro@cornelius.estig.ipb.pt)
08:49.45*** join/#htc-linux BabelO (n=Miranda@faim1.itesoft.com)
09:05.07*** join/#htc-linux rob_w|tablet (n=bob@gw.suellberg-hamburg.de)
09:20.12*** join/#htc-linux ellisway (n=ellis@host-87-74-241-174.bulldogdsl.com)
09:43.50*** join/#htc-linux tsdogs (n=twostupi@84.18.144.39)
10:00.27goxboxliveBabelO: Any idea? http://pastebin.ca/507024
10:04.11*** part/#htc-linux DTMonterrey (n=evandro@cornelius.estig.ipb.pt)
10:08.35*** join/#htc-linux DTMonterrey (n=evandro@cornelius.estig.ipb.pt)
10:08.35BabelOgoxboxlive : you are compiling for computer ?
10:10.15goxboxliveno
10:10.30goxboxlivei just added the cayenne in to helix
10:10.47goxboxliveIt's actually not helix wich failes, it is mediaserver
10:12.04goxboxlivei see the target=i46, but shoudlnt this be automatic in opie2 to use arm
10:12.27BabelOgoxboxlive: no
10:13.22BabelOi use this command -helix -helix-system-id linux-2.2-libc6-xscale-cross-gcc32
10:15.30goxboxliveBabelO: I have that in my configure file
10:18.11BabelOhmm
10:20.22BabelOcan you try this find src/3rdparty/libraries/helix/src/build/umakecf -name "linux*.cf" -printf "%f\n" | sed "s/\(.*\).cf/\1/"
10:30.06goxboxlivearm-linux-g++ -pipe -W -Wreturn-type -fno-exceptions -mcpu=xscale -mtune=xscale  -
10:30.32goxboxliveBabelOP: Yes i will, just trying to rebuild helix.
10:32.26goxboxlivemy paste here was just to show you that helix is beeing build with arm-linux
10:51.25goxboxliveBabelO: There arent any problem building Helix. It finish, but the problem is mediaserver in Qtopia:
10:52.15goxboxliveBabelO: Here is the output
10:54.06goxboxlivehttp://pastebin.no/1737
10:54.08BabelOon universal you can use this one linux-2.2-libc6-iwmmxt-xscale, it is optimized
10:54.50goxboxliveOk, but then i have to reconfigure right?
10:56.48BabelOgoxboxlive : : but i can't use your image then :(
10:58.30BabelOyou error in media server is always about aomic ?
10:58.31goxboxlivehmmm, i am trying to copy the configured .cf file to umake now
10:58.37goxboxlivehold on i am building it now
11:24.49*** join/#htc-linux BabelO (n=Miranda@faim1.itesoft.com)
11:29.50elliswayhi all
11:34.37cr2hi
11:35.08cr2BabelO: is the "iwmmxt" binary compatible to "normal" ?
11:35.37cr2iwmmxt is used by a780, and afaik there were some problems with that.
11:35.39BabelOcr2 : hi, i don't think so
11:36.06BabelOcr2 : as you say to me there is special intruction set
11:36.24cr2i thinks yes, but not on all cpus.
11:45.33goxboxlivehi cr2: What about the cam driver? Make one and i'll test it. I cant test with my A780 because of the usb bug
11:59.34BabelOcr2: there is big discuss about ts-adc-debounce, i have same issue than H5000 ;)
12:10.59BabelOok i think i ve to wait few hours to have a correct adc patch ;)
12:13.26cr2BabelO: that's why universal stays with the old (and working) driver so far. the ts-adc code needs to stabilize first.
12:14.26BabelOi like to test all new thing i found ;) lol
12:31.32goxboxliveBabelO: Is there only in mkspek that bluaangel-g++ is? Could i easaly change that to htcphone-g++?
12:31.33goxboxlivein htcphone?
12:31.34BabelOgoxboxlive : : yes not a problem :)
12:31.35*** join/#htc-linux WizMaui (n=WizMaui@62.112.90.250)
12:31.47goxboxliveok, so it's not defines anywhere else
13:03.11goxboxlivehi
13:03.25goxboxlivesorry
13:40.29*** join/#htc-linux ashkan (n=ashkan@unaffiliated/askywalker/x-0000002)
13:54.15*** join/#htc-linux progeny (n=progeny@89.129.43.118)
14:03.33*** part/#htc-linux ashkan (n=ashkan@unaffiliated/askywalker/x-0000002)
14:07.11*** join/#htc-linux rob_w|tablet (n=bob@gw.suellberg-hamburg.de)
14:15.45*** join/#htc-linux rob_w (n=bob@gw.suellberg-hamburg.de)
14:25.30*** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com)
15:45.59*** join/#htc-linux eway (n=ellis@host-87-74-241-174.bulldogdsl.com)
16:43.58*** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
16:48.35*** join/#htc-linux TripleDES (n=sergio@82.pool85-50-237.dynamic.orange.es)
17:22.09*** join/#htc-linux jeanseb (n=jeanseb@gazypan.dyndns.org)
17:32.29*** join/#htc-linux babelo (n=babelo@lun34-2-82-238-28-28.fbx.proxad.net)
17:33.12babelohi
17:36.15*** join/#htc-linux babelo (n=babelo@lun34-2-82-238-28-28.fbx.proxad.net)
17:36.49jeansebhi
17:39.04tsdogsjeanseb: hi
17:39.14tsdogsbabelo: hi
17:39.40babelohi tsdogs
17:39.54babelohi jeanseb
17:40.07tsdogsbabelo: trolltech accepted roadmap project :)
17:40.15babelomissing some nice option to gpeirc
17:40.25tsdogs:)
17:40.32babelohi see yesturday
17:44.10*** join/#htc-linux babelo (n=root@lun34-2-82-238-28-28.fbx.proxad.net)
17:45.41*** join/#htc-linux ellisway (n=ellis@host-87-74-241-174.bulldogdsl.com)
17:45.49babelook better with gaim
17:47.53goxboxlivetsdogs: cool that they did. Lets hope you get greenphone then
17:48.03goxboxlivebabelo: You shuld join too.
17:48.16goxboxliveyou have both htcphone and gnuphone
17:48.49tsdogsYou need more space though :) Just finished downloading the SDK 4.2.2 iso :( still have to install but wil take something like 3GB
17:49.07babelook i have to fill web page
17:49.26babelohmm new harddrive saturday so
17:49.27tsdogsBy the way, I almost finished original porting, but I'd need some testing ...
17:50.56babelogood, i have a nice nvidia link today for `3d
17:51.40babelowait i ll check
17:51.52*** join/#htc-linux pH5 (n=ph5@18-175.3-85.cust.bluewin.ch)
17:51.54tsdogsdoes BA have an nvidia ? :)
17:52.05babelono lol
17:52.59*** part/#htc-linux babelo (n=root@lun34-2-82-238-28-28.fbx.proxad.net)
17:59.42*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
18:01.49*** join/#htc-linux scorpio16v (n=scorpio1@ACB25167.ipt.aol.com)
18:03.06BabelOtsdogs: http://www.nzone.com/object/nzone_humanhead_home.html
18:03.13BabelOreal 3d
18:04.15tsdogs:)
18:04.47*** join/#htc-linux RoEn_PC (n=roen@p54a64ae2.dip.t-dialin.net)
18:20.53tsdogsBBL.
18:28.10*** join/#htc-linux psokolovsky_ (n=psokolov@82.193.98.15)
18:32.36*** join/#htc-linux psokolovsky (n=psokolov@82.193.98.15)
18:32.45*** join/#htc-linux _law_ (n=_law_@213.173.86.202)
18:34.17*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
18:35.29goxboxlivei hate helix/mediaserver
18:43.37*** part/#htc-linux scorpio16v (n=scorpio1@ACB25167.ipt.aol.com)
19:03.19*** join/#htc-linux WizMaui (n=WizMaui@62.112.90.250)
19:03.37*** part/#htc-linux WizMaui (n=WizMaui@62.112.90.250)
19:35.02cr2goxboxlive: how do i  compile the a780 kernel ?
19:35.31Kevin2Hi cr2
19:35.43cr2hi Kevin2
19:36.01cr2Kevin2: i'm looking at applying your patch
19:36.45Kevin2cr2: Thanks.
19:37.08goxboxlivecr2: I dont know. I downloaded a precompiled one
19:37.11cr2Kevin2: your clock setting on hermes is 163xxxxx
19:37.27cr2goxboxlive: it's oe something.
19:37.38Kevin2cr2: I've pulled that out of my local tree.
19:37.53goxboxliveok, just change the machine from htcuniversal to a780 and use angstrom as distro
19:37.54cr2ok.
19:38.16cr2goxboxlive: is it opie or moko ?
19:38.26goxboxlivecr2: you change that in conf/local.conf
19:38.40goxboxliveyou better use x11 or something i guess
19:39.00goxboxliveor just a console image, thats the quickest way
19:39.21goxboxlivewhy dont you just download the latest kernel and try it out?
19:39.40cr2goxboxlive: there is no latest kernel
19:39.45cr2it's all in oe.
19:40.06goxboxlivewell i guess someone at #openezx has it somewhere
19:42.47cr2Kevin2: updated.
19:45.07*** join/#htc-linux hlbot873 (n=adm@iclem.net)
19:46.06cr2BabelO: have you changed the blueangel-g++ to htcphone-g++ already ?
19:53.27*** join/#htc-linux hlbot (n=adm@iclem.net)
19:58.39cr2goxboxlive: i don't see the camera patch there.
19:59.05goxboxliveok
19:59.09goxboxlivehold on
20:00.18Kevin2cr2: Do we know if hermes has a 1wire bus, or how it monitors the battery?
20:10.21BabelOcr2: not changed, do you think there is something specific to do ?
20:10.49BabelOKevin2: does your ADC working well ?
20:11.49Kevin2BabelO: Yes.  I just converted over to the generic code - no new problems so far.
20:14.14BabelOKevin2: i have some noise, when i use adc-battery and ts at same time, a read to battery trigger move to touchscreen :(
20:14.30BabelOand sometimes touch pointer move over the screen
20:16.09Kevin2BabelO: Which device is this?
20:16.45BabelOKevin2: it is on BA, with TSC2200 which is MFD
20:16.56Kevin2MFD?
20:17.23BabelOSoC
20:17.47BabelOit have an ADC converter an keyboard driver also
20:18.17BabelO, i ve no pen_gpio for it only irq
20:18.55BabelOit provide a touscreen bit in one register to now pointer state
20:20.11KmarcWhat is mean mfd?
20:20.34BabelOMulti functionnal Device, replacement for SoC
20:20.46BabelOi think it is that now
20:20.46Kmarcok
20:24.59Kevin2BabelO: I think you need to make sure keyboard driver and adc driver don't bump heads.
20:28.22cr2Kevin2: the battery is served by some monster chip :)
20:28.39cr2over i2c. it includes LED control and some on.
20:28.55cr2it#s all in the hh.org wiki.
20:29.31Kevin2cr2: Hermes boots with latest kernel, but there is no screen.  Did the dummy fb driver move somewhere?
20:30.10cr2hm.
20:30.20cr2i'll have a look.
20:30.27cr2TI TPS65021 - battery contrl.
20:30.53cr2Kevin2: did you include LCD ?
20:31.18cr2i've changed LCD_DEVICE -> LCD_CLASS_DEVICE recently.
20:34.17Kevin2Hrmm.   Looks like I have something as a module..
20:35.18*** join/#htc-linux tsdogs (n=twostupi@62.123.180.130)
20:35.50cr2Kevin2: there is a TPS65011 driver in the kernel already.
20:36.32Kevin2cr2: That chip is a voltage regulator; not a battery sensor.
20:37.30cr2then the battery sensor is on the ADC.
20:37.40cr2hm. the touchscreen too ?
20:38.22cr2tsdogs: congratulations with the roadmap
20:38.26*** join/#htc-linux TripleDE1 (n=sergio@133.pool85-50-145.dynamic.orange.es)
20:38.41tsdogscr2: it was your idea :)
20:38.52cr2tsdogs: how far are you now ? there are some other things to do.
20:39.34tsdogsI think I finished the QT4 port, now need to integrate into qtopia SDK and fix eventual bugs.
20:39.40cr2imho we can really do something useful here with a real enduser impact.
20:40.08cr2ok, can you send me a patch ? i'll look at other issues.
20:40.16Kevin2cr2: Hermes boots now.  (I do have an irq patch in my kernel - not sure if it is required.)
20:40.23cr2probably we will just create some small fork.
20:40.31tsdogsOnce I have the original working I'll submit it somewhere for other to test.
20:40.46tsdogsThen I have many ideas on how to improve the interface.
20:40.47cr2Kevin2: with standard defconfig ?
20:41.00cr2tsdogs: is it allowed by the rules ?
20:41.19Kevin2cr2: I'm not sure - I'm using my own config.
20:41.25cr2tsdogs: you are using the main CVS branch ?
20:41.30tsdogscr2: what do you mean ?
20:41.33tsdogsyes.
20:41.48tsdogsother people working on it?
20:42.47cr2ah, ok. "prior to submission"
20:43.11cr2"must not have been posted on another Web site prior to submission"
20:43.24cr2http://trolltech.com/developer/tt/rules
20:43.40cr2Kevin2: post a patch, we don't want 1000 forks ;)
20:44.50cr2tsdogs: the route  planner is not in the CVS branch roadmap_editor, and it has a bit different API.
20:45.03cr2s/ not //
20:45.14tsdogswas reading the whole.
20:45.25tsdogsok then I'll have to test it by myself :(
20:46.18tsdogsI beleived that this one was for editing routes.
20:46.41cr2not only.
20:46.41tsdogsOn this one I have not done anything right now...
20:47.18cr2the difference is not big.
20:47.48tsdogsOnce the main roadmap is in, I'll start that one.
20:47.55tsdogsWill freeze and test.
20:48.12tsdogsIf ok I'll go on developing better interface.
20:48.23cr2anyway, you can send me the the patch. i'm mainly concerned about some additional line styles and font selecton API which is missing.
20:49.04cr2roadmap_editor uses postgis for the road database, which is very cool.
20:49.23tsdogsok. Does not matter if it breaks QT3 ?
20:49.39cr2create a qt4 directory.
20:49.55tsdogsthat was my intention.
20:49.59cr2there are gtk, gtk2, qt and win32 (xp+wince) already.
20:50.09cr2gtk is broken.
20:50.12tsdogsYou will use QT4 ?
20:50.17cr2yes.
20:50.34cr2i can compile the wince version with mingw32ce
20:50.45tsdogsok, then I'll prepare the patch for you to see the state.
20:50.48cr2thanks to Kevin's haret makefile :)
20:50.55tsdogs:)
20:51.24cr2and gtk2 works too.
20:51.34cr2that's the way to world domination :)
20:52.14tsdogsThe main part has been done very well, I have some dubts about the QT implementation though
20:55.07Kevin2cr2: Can you apply the ad7877 patch I sent to kernel-discuss earlier today?
20:56.36cr2it's the one-line "completion" patch ?
20:56.59Kevin2Yeah.
20:57.10cr2ok.
20:57.20Kevin2INIT_COMPLETION( -> init_completion(&
20:58.46cr2done.
20:59.05Kevin2Thanks.
20:59.35cr2Kevin2: what about the phone on apache ?
21:00.06cr2do you really need weird $HTC strings ? the rest looks like standard AT.
21:00.37Kevin2Yeah.  It would probably work if we do the init in wince.
21:00.59Kevin2I think disassembling the wince phone app might help too.  But I'm reluctant to go down that path.
21:01.12cr2and since you have ak4641 it should not be the problem.
21:01.30cr2where do i get the rilgsm.dll ? or is it rilcdma.dll ?
21:01.44Kevin2I'm a little concerned that if the phone roams from one base station to another that it may send those weird htc_ things again.
21:01.46cr2ak4641 is also used on hermes
21:02.50cr2Kevin2: we need to decouple the headphone insert from the "sound PCM"  part in the ak4641 i2c driver.
21:03.03Kevin2There is a ril.dll and a rilphone.dll on Apache.
21:03.31cr2i have done some work on the i2c control there, and can write the full set of mixer controls.
21:03.50cr2rilphone.dll sounds interesting.
21:04.03cr2ril.dll is usually the generic mid-lev api.
21:05.57Kevin2Unfortunately, I can't copy or open those files directly.
21:06.12cr2testwm5 ?
21:06.36cr2it's not perfect, but is still better than nothing.
21:08.12cr2tsdogs: i have copied the day/night styles from the copilot route planner on a780, and i know one other program that can be reused for such styles (normal ascii)
21:08.29tsdogsok
21:08.34tsdogsperfect
21:08.47cr2tsdogs: with a good preferences/styles setup the program looks very good.
21:08.57tsdogsI think so too
21:09.16tsdogsI was of the Idea to do something like TomTom or so
21:09.48cr2the main problem is that there is no way to select the canvas font size, name and colour. it#s all hardcoded
21:10.15tsdogswe can simply add a few call backs.
21:10.21cr2roadmap_editor supports the transparent png icons a la tomtom.
21:10.25tsdogsand read configuration.
21:10.31cr2i'm looking for the screenshots.
21:10.35Kevin2cr2: How did you find names of variables and functions in the dlls?  Did they have some debugging data in them, or did you back track from error messages?
21:10.36tsdogsalso configuration locations are a bit confusing
21:11.19cr2here: http://www.freemap.co.il/roadmap.html
21:12.18cr2and here http://www.freemap.co.i
21:12.21cr2http://www.freemap.co.il
21:12.23cr2sorry.
21:12.52*** part/#htc-linux DTMonterrey (n=evandro@cornelius.estig.ipb.pt)
21:13.11cr2even comes with the xp/wince exe/cab
21:13.16tsdogscr2: cool
21:13.24cr2but only in hebrew afaik.
21:13.44cr2so i've compiled it with gcc and english interface.
21:14.01cr2though it support full utf8 and even right-to-left text.
21:14.24tsdogsbbiaf
21:14.48cr2yeah, it's a very cool program nobody knows about.
21:18.16tsdogsyes.
21:18.41cr2Kevin2: i'll try to check the NAND parameters we set. can you post the full dmesg ?
21:20.12Kevin2http://pastebin.ca/508079
21:21.58*** join/#htc-linux the_sys0p (n=the_sys0@cpe-66-75-196-173.bak.res.rr.com)
21:22.08*** join/#htc-linux the_sys0p (n=the_sys0@cpe-66-75-196-173.bak.res.rr.com)
21:25.25Kevin2I'll be back later.
21:26.20*** join/#htc-linux the_sys0p (n=the_sys0@cpe-66-75-196-173.bak.res.rr.com)
21:31.25*** join/#htc-linux the_sys0p (n=the_sys0@cpe-66-75-196-173.bak.res.rr.com)
21:34.24*** join/#htc-linux Fossi (n=Fossi@e176108051.adsl.alicedsl.de)
21:48.20*** part/#htc-linux rmoravcik (n=rmoravci@adsl-d181.84-47-56.t-com.sk)
21:49.12*** join/#htc-linux Foxdie (i=Foxdie@77-98-48-193.cable.ubr05.harb.blueyonder.co.uk)
22:01.19tsdogscr2: need your e-mail.
22:01.59cr2your is at gmail.com ?
22:02.11tsdogsyes
22:02.24cr2ok.
22:04.50cr2User unknown
22:05.19tsdogsho, sorry it's at gmail but not as tsdogs.
22:05.28tsdogscan send through sf though
22:05.37cr2ok.
22:08.11cr2hehe. 550-This @users.sourceforge.net account doesn't exist or isn't currently
22:08.54tsdogssee in private chat
22:09.09tsdogsHave to check sf though
22:09.16cr2sent.
22:10.51tsdogsok
22:10.59tsdogssent back
22:11.53cr2ok. will take some minutes.
22:11.59tsdogs:)
22:14.18tsdogscr2: you thing that doing some double buffering would improve fb performance ?
22:16.22cr2on universal ?
22:16.40cr2you can use SRAM, but it's only 256K
22:16.46cr2PXA SRAM
22:17.14cr2wince has some hack, so the fb uses SRAM+DRAM.
22:17.21cr2SRAM for the upper part.
22:17.28tsdogsha.
22:17.47cr2there was a guy that claimed to have written such linux driver, but then he disappeared without a trace.
22:18.18tsdogsand having an hidden canvas to draw on and then do an update to the original canvas?
22:19.00tsdogsthis is especially for 3D
22:19.05cr2yes, just allocate a pixmap, and do bitblt() of how it is called in qt.
22:19.31tsdogsyep. but i think it's already done like that. I have to check
22:19.37cr2but then you need the SRAM driver.
22:19.56cr2and the pixmap should not exceed 256K in size, otherwise it will be too complex.
22:20.11tsdogsok DRAM ain't that fast
22:20.22cr2SRAM driver i just ioremap() actually.
22:20.36BabelOgoxboxlive: aby success with helix ?
22:20.56BabelOs/aby/any/
22:20.59goxboxlivenope, it hates me
22:21.13tsdogscr2: ok
22:21.14goxboxlivei tried with imxxwmt
22:21.22goxboxlivebut no way
22:21.42BabelOgoxboxlive: ok :( try without alsa
22:21.53goxboxliveWhats the nake of the fiel with QPE_VERSION failing?
22:21.59BabelOgoxboxlive: do you know how can i mask gpe cursor ?
22:22.14goxboxliveno
22:22.18tsdogscr2: roadmap_editor is building for you? I have errors.
22:22.26cr2qt ?
22:22.31tsdogseven with gtk2
22:22.43cr2gtk2 and wince compiles.
22:22.49BabelOgoxboxlive: need to ask ljp for this, it is also at helix project
22:22.51tsdogshmm.
22:22.53cr2pastebin the error
22:23.07cr2make DESKTOP=GTK2 SHAPEFILES=NO
22:23.12tsdogsok
22:23.38tsdogsroadmap_messagebox.c:74: error: ‘modal’ undeclared (first use in this function)
22:23.41tsdogs??
22:23.53goxboxliveBabelO: Yes, he has helped me, but he doesnt know what causes it. I have copied the helix source also like he told me, but no luck.
22:25.26BabelOonly difference with me ans tsdogs is we have together qt3 on host :(
22:28.32cr2tsdogs:    gtk_window_set_title (GTK_WINDOW(dialog),  roadmap_start_get_title(title));
22:29.24cr2that's  my roadmap_messagebox.c:74
22:31.56tsdogscvs version?
22:32.29cr2yes. cvs up -dfP
22:33.06cr2make DESKTOP=GTK2 SHAPEFILES=NO
22:33.23tsdogssaw it before.
22:33.48cr2CVS/Repository
22:34.03cr2roadmap_editor/src
22:34.07tsdogsyes
22:34.36cr2compliles and works.
22:34.45cr2just have recompiled again.
22:35.03cr2qt is broken because of a slightly changed api.
22:35.23tsdogsok. but gtk ain't working here ... :(
22:36.14tsdogsIt's missing some includes ... roadmap_canvas_agg.cpp:35:34: error: agg_rendering_buffer.h: No such file or directory
22:36.48*** join/#htc-linux the_sys0p (n=the_sys0@cpe-66-75-196-173.bak.res.rr.com)
22:37.48cr2hm. wait, i'll make a cvs diff
22:37.50goxboxlivehelix-client-qtopia-nodist.pf.helixconf
22:37.55goxboxlivesorry
22:38.31goxboxliveqtopianamespace.cpp needs editing the QPE_VERSION
22:39.15cr2tsdogs: can you get and compile agg 2.5.0 locally ?
22:39.29cr2it's not needed for qt, btw.
22:39.36tsdogsI can try to install it. ok
22:40.06cr2i have only minor local patches.
22:40.08cr2-I/usr/include/pgsql
22:40.16cr2-   switch ((int) data) {
22:40.16cr2+   switch ((long) data) {
22:40.27cr2that's all.
22:40.32cr2because i have amd64
22:41.52tsdogsme too
22:42.37tsdogswhy pgsql ? does it use db?
22:43.06cr2for the map creation. it uses postgis.
22:43.14tsdogsok
22:43.18cr2not for the gui part.
22:43.34tsdogslike popt
22:43.46cr2src/buildmap_postgres.*
22:44.22cr2it can't be even compared with the bizarre tiger format ;)
22:44.54tsdogs:) It's really bizarre
22:46.09tsdogserror: cast from ‘void*’ to ‘int’ loses precisio
22:46.12tsdogsn
22:46.35cr2yeah, the code is not 64bit clean.
22:46.40tsdogsdo I need agg-2.5 ? I have 2.4
22:46.51cr2i think 2.4 should work too.
22:46.59tsdogsshould I use m32
22:47.11cr2routing lib does not work with 64bit
22:47.18tsdogsok
22:47.23cr2because of some stupid assumptions.
22:47.39goxboxlivecr2: florian is going to make us a image with cam for A780
22:47.43cr2but the program should compile
22:48.26cr2goxboxlive: it's better if he will merge the pxa-camera and ov95xx.c drivers into some kernel tree.
22:48.40cr2tsdogs: with -m32 you'll need the m32 qt4.
22:49.35cr2patch arrived :)
22:50.08tsdogsfinally.
22:50.14tsdogsgraylisting?
22:51.14cr2yes.
22:51.20tsdogswhat's libdgl ?
22:51.23cr2is the #ifdef QWS really needed now ?
22:51.41cr2make -f Makefile.alone
22:51.51tsdogsnot anymore I think, but have not tested with qtopia yet
22:51.58cr2it's the buggy not 64bit clean DirectedGraphLibrary
22:53.18tsdogsgosh qt errors were less...
23:00.16tsdogscr2 did it build?
23:00.48cr2what ? the patch ?
23:02.44tsdogsyes
23:04.07tsdogsmultiple definition of `traffic_preferences'
23:05.13cr2will try now.
23:05.31tsdogsare they still working on roadmap ?
23:06.51cr2main or _editor ?
23:06.54tsdogsboth
23:07.28cr2the _editor is tested only on wince and xp recently.
23:07.44cr2hm. why does it compile and work for me ?
23:07.59tsdogswhat do you mean ?
23:08.12tsdogsyou have qt4 in /usr ?
23:08.25cr2i mean gtk2.
23:08.39cr2no, my qt is in /usr/local/TrollTech/4.2.2
23:08.42tsdogsthere are a few minor bugs I think
23:08.43tsdogsok
23:09.10cr2ok, patched cleanly.
23:09.17cr2what is the command line ?
23:09.26tsdogsmake DESKTOP=QT4
23:09.34tsdogsyou need to set OPIEDIR also
23:09.40tsdogshave a look at config.mk
23:10.00cr2i compile for amd64 first.
23:10.04tsdogsok builded finally
23:10.07tsdogsok.
23:10.16tsdogsHave not tryed cross yet.
23:10.32cr2ok, the qt4 paths are wrong.
23:11.13cr2make DESKTOP=QT4 QTDIR=/usr/local/Trolltech/Qt-4.2.2
23:11.41cr2make[1]: /usr/local/Trolltech/Qt-4.2.2/bin/moc-qt4: Command not found
23:12.00tsdogschange it in qt4/Makefile
23:12.07tsdogsas I have qt3 and qt4
23:12.34cr2usr/lib64/gcc/x86_64-suse-linux/4.1.0/../../../../x86_64-suse-linux/bin/ld: cannot find -lQtCore
23:12.48tsdogsso roadmap_editor should be the good one and not roadmap
23:13.15tsdogsI think you should also check the LDFLAGS then
23:13.26tsdogsthat's why configure is better...
23:13.45tsdogsand LIBS
23:13.49cr2ok, it's lib, not lib64 here.
23:14.00tsdogsok
23:14.24cr2hm. the binary map format is incompatible.
23:14.37cr2i need the main CVS formatted maps.
23:14.45cr2it's evil ;)
23:15.32tsdogsyes, I can't use the ones I built
23:15.35cr2can you put your usdir.rdm and some map for download ?
23:15.41cr2??
23:15.48tsdogsyes
23:16.09cr2i have the maps with routing info for roadmap_editor, they are a bit different.
23:16.33cr2well, i can get then from roadmap.digitalomaha.net
23:17.32tsdogsNow I understand why they did not work with roadmap :/
23:17.43tsdogsI have the 2006 though now :)
23:17.45cr2hehe. i think they wil not work too :)
23:17.54cr2i need to build my own.
23:18.04tsdogshttp://84.18.144.44/tsdogs/maps.tar.bz2
23:18.09cr2ok. thanks.
23:18.14tsdogsit's only index and SF map
23:19.48cr2position ?
23:20.08cr2these are just minor packaging and convenience issues ;)
23:20.40tsdogs-122396962,37793920
23:21.01cr2hehe. utf8 is broken in your code :)
23:21.08cr2lat long ?
23:21.21tsdogsthere were some Latin1
23:21.27cr2Find->position
23:22.01BabelOand some screen shot please ;)
23:22.03cr2there are some usability changes that are a must for roadmap.
23:22.31cr2like show me _some_ avaiable map
23:23.07tsdogsThat's what I thought too, (was more like explore map)
23:23.11cr2so you don't just stare at the blank canvas, and have no idea what to do, and which maps are available.
23:23.55tsdogsdialogs are implemented very nasty
23:24.15cr2what are the golden gate bridge coordinates ?
23:24.29cr2they are not shown on the canvas too.
23:24.54cr2i'd also like to have the coordinate grid (optional checkbox)
23:25.22tsdogsdo qtroadmap --location=-122396962,37793920
23:25.32cr2ok.
23:26.28cr2-122, 37 :)
23:27.05cr2s## roadmap_option.c, line 439: illegal option --location
23:27.21tsdogssorry --position
23:27.47tsdogsno it's location
23:27.51cr2wow. rotated strings.
23:28.08cr2i've entered Find->Location
23:28.13tsdogsI suppose they are working
23:28.24cr2the coordinates are 10^6 degrees
23:28.36cr2this is not optimal, but nice.
23:28.56cr22^31 may be better. like garmin is doing it.
23:29.18cr2but it's too far-fetched.
23:29.25tsdogs:)
23:29.49tsdogscr2: should we improve roadmap, or go on with roadmap_editor ?
23:30.07cr2no 3D, because i didn't install it.
23:30.17tsdogshm?
23:30.22cr2i suggest to go on with roadmap_editor.
23:30.25tsdogsyou can disable 3D?
23:30.31tsdogsok
23:30.37cr2no icons, and such things.
23:30.52tsdogshoo.
23:30.53tsdogsok
23:31.17cr2roadmap_editor has some deficiences,  but we can easily override them.
23:31.37cr2it has fixed limited number of layers, for example.
23:32.00tsdogsso better view.
23:32.28cr2i suggest to create a fork, and then post the patches to the roadmap mailing list.
23:32.57tsdogsfor roadmap or editor?
23:33.05tsdogsafter the contest :)
23:33.08cr2because otherwise we would not be able to move anywhere.
23:33.31cr2based on the editor.
23:34.27tsdogsok
23:35.20cr2this is of course evil
23:35.24cr2<PROTECTED>
23:35.24cr2<PROTECTED>
23:35.34tsdogsenough for today. I'll check the editor next days.
23:35.35cr2should be DejaVu
23:35.40tsdogs:)
23:35.47tsdogsshould be in config
23:35.57cr2ok. great achievement anyway:) thanks!
23:35.59cr2yes.
23:36.16tsdogsgood night all
23:36.20cr2good night
23:36.23pofcr2: i published the trinity SPL exploit
23:36.40pofsadly it can be exploited on hermes :(
23:37.22pofhttp://forum.xda-developers.com/showthread.php?t=308691
23:37.34pofhttp://forum.xda-developers.com/showthread.php?t=308707
23:38.27cr2pof: that's fun :)
23:38.48cr2pof: is it possible to sign the binaries like haret ?
23:39.06cr2with openssl or something like that, i.e. no m$ tools.
23:39.16pofouch..
23:39.35pofi was going to say yes, with M$ developer certificates
23:39.52pofbut i don't know how to sign wince binaries with open source tools
23:40.03cr2ms cert is ok, but the tools.
23:40.04cr2ok.
23:40.28cr2btw, does haret work on wince6 ?
23:40.39pofyes :)
23:40.48cr2what is actually new in wince6 ?
23:41.05cr2well, then there is still no security whatever in wince :D
23:41.29pofmore speed, less hangs / reboots
23:41.41cr22GB address space ?
23:41.47cr2per process ?
23:41.54cr2or still 32MB
23:42.05pofno idea :-?
23:42.34cr2ok, but the zero security is still appaling.
23:42.42pofyes
23:43.02pofalmost everything you know from wm5 applies to wm6
23:43.17cr2ok.
23:43.28cr2if haret can run, it's not an OS :)
23:43.34poflol
23:44.25pofhave u been doing anything new on hermes ?
23:46.18cr2yes, i'm looking at NAND now
23:46.32cr2and have documented the ATI SD registers.
23:46.42cr2it applies to athena too.
23:47.25cr2are there any free icons, backgrounds etc available ?
23:47.43cr2i think we will soon need to customize the qtopia.
23:48.02cr2btw, what did come out from the iphone icons story ?
23:48.37pofi didn't follow that story
23:49.11pofi think apple contacted xdadev, to remove all the iphone skins from the forum
23:49.33pofso we did, but people republished it in other sites... apple can't control that :P
23:50.23pofas for free icons you can have a look here: http://del.icio.us/tag/icons
23:52.32cr2ok, thanks. it's a sensible issue, we can't just release warez / pictures with unclear ownership on linux :)
23:53.06cr2but we also can't control what other people will use :P
23:54.06pofoh while doing the exploit i thought of a technique i wanted to tell you
23:54.41pofyou can get a checksum of each individual byte of ram from bootloader
23:54.54pofso this can be used to produce a ram dump :)
23:55.23cr2LOL.
23:55.42pofbtw, you can dump flash this way too, it's a bit different
23:56.05pofWhen you request a checksum of flash, checksum command copies requsted portion to a certain offset in ram
23:56.20pofand it can be dumped from ram then. 1-byte checksums from flash directly are still slower
23:56.35cr2ok.
23:56.59cr2unless they will encrypth NAND it's not a big issue so far.
23:57.42pofyes
23:57.53pofi will implement this thechinques in HTCflasher :)
23:58.10pofbtw, have u tried version 2.0beta ?
23:58.27pofi implemented 'rbmc' NAND dump, but it doesn't work as expected
23:58.45pofonly 'cu' is able to dump right with rbmc
23:59.56pofi don't know what the problem is, i think i'm not using the right termios settings for USB port

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