IRC log for #htc-linux on 20080420

00:21.16*** join/#htc-linux mistadman (n=mistadma@adsl-250-234-215.msy.bellsouth.net)
00:24.39*** join/#htc-linux BabelO (n=Fabrice@unaffiliated/babelo)
00:50.51*** join/#htc-linux Dinde (n=kayser@sur-internet.net)
01:06.47*** join/#htc-linux mistadman (n=mistadma@adsl-250-234-215.msy.bellsouth.net)
01:23.50*** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-961afa7fd0bd53b7)
01:40.01*** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-7ba3313f45c3ce0a)
02:03.22*** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-b820b95ef2a484c7)
02:09.30*** join/#htc-linux ltxda0 (n=ltxda@unaffiliated/ltxda)
02:18.15Pryanwenas
02:19.09*** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda)
03:03.41*** join/#htc-linux TimRiker (n=timr@rikers.org)
03:37.37*** join/#htc-linux mistadman (n=mistadma@adsl-250-234-215.msy.bellsouth.net)
05:16.56*** join/#htc-linux bigdissaved (n=david@quandary.cross-lfs.org)
05:19.34bigdissavedHey, My HTC Apache is messed up, touchscreen has a spider crack on it. I got a new phone and I want to build Linux for it. I am going to build Embedded CLFS for it, I am wondering if when I get it working, is there anything that I can do to help further the project for it?
05:20.14bigdissavedI am going to try to remove the/replace the touchcreen, I can take some indepth pics of it them
05:20.30bigdissaveds@then@@
05:48.45*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
06:54.49*** join/#htc-linux hlbot (n=adm@iclem.net)
07:13.06*** join/#htc-linux tsdogs (n=tsdogs@62.123.180.130)
07:19.51*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
07:33.31*** join/#htc-linux goxboxlive (n=goxboxli@208.84-48-176.nextgentel.com)
07:43.38*** join/#htc-linux rob_w (n=bob@X16de.x.pppool.de)
07:53.51*** join/#htc-linux tuff (n=phil@tuffnatty.dialup.corbina.ru)
08:40.11*** join/#htc-linux Robwoerle (n=bob@X09c2.x.pppool.de)
08:50.24*** join/#htc-linux rob__w (n=bob@X0c00.x.pppool.de)
09:03.34*** join/#htc-linux pH5 (n=ph5@p5485EAB4.dip.t-dialin.net)
09:07.38*** join/#htc-linux rob__w (n=bob@X0cef.x.pppool.de)
09:18.45*** join/#htc-linux toi (n=pleemans@d5152D3B4.access.telenet.be)
09:37.49*** join/#htc-linux pH5_ (n=ph5@p5485C069.dip.t-dialin.net)
10:26.39BabelOapt: what is QLandkarte ?
10:26.40aptACTION is QLandkarte to understand recursion, you must first understand recursion
10:26.49BabelOapt: what is QLandkarteGT ?
10:26.50aptACTION is QLandkarteGT to understand recursion, you must first understand recursion
10:26.54BabelOgrhh
10:26.56BabelOhi
10:27.18BabelOpH5_: i try to boot artemis from spl directly
10:30.00pH5_hi BabelO, how?
10:30.16pH5_on hermes I can't upload any 'unsigned' code to ram.
10:31.10pH5_tries to rebase pasic3+ds1wm onto clocklib
10:31.15BabelOpH5_: it seem that guy from linwizard are able to boot using EOL Bootloader
10:31.38BabelOpH5_: with diagnostic card
10:31.54pH5_BabelO: do you have a working recipe for creating a diagnostic card?
10:32.03pH5_when I tried it, my cards were never recognised
10:32.31BabelOpH5_: at the moment i use the image from tornado, but the signature is wrong, i m looking for the good one
10:34.03pH5_BabelO: it would be terrific if you could get that to work. because I still haven't found a possibility to get out of the hermes' bootloader here.
10:34.35BabelOpH5_: and can be good for herald too ;)
10:34.49pH5_sorry, s/hermes/herald/
10:34.51BabelOpH5_: http://pastebin.ca/990680
10:34.58pH5_I don't have hermes
10:35.15BabelOactually i can't to it because my device is CID locked
10:35.32BabelOso i need to found these magic values
10:35.35BabelOany idea ?
10:39.12BabelOpH5_: funny there is also blueangel and co in the perl file... so we can boot from diagnostic card directly
10:39.29pH5_BabelO: the keys are usually at the very end of the SPL
10:39.41BabelOthese keys are usually found near the end of the copy-data section
10:39.43BabelOyes ;)
10:39.55pH5_three keys, 8 bytes each, 32bit aligned
10:39.58BabelOpH5_: i ve the bootloader of artemis here
10:43.07BabelOpH5_: is it that ? http://pastebin.ca/990686
10:44.15pH5_heh, those are exactly the same three keys that I have written down for herald..
10:45.05pH5_yup, same keys in the herald SPL
10:45.31BabelOok so i can use those one ? which order from top to bottom ?
10:46.23pH5_I think so, I'm not sure.
10:47.45pH5_meaning, the keys are from top to bottom C1AED2..., but I don't know which of the three corresponds to which seclevel
10:53.41BabelOok
10:53.42*** join/#htc-linux aLeSD (n=alex@227.Red-83-44-153.dynamicIP.rima-tde.net)
10:53.55BabelOpH5_: now i have to understand how to use this tool
10:54.15*** join/#htc-linux Marex (n=marex@vasut.kolej.mff.cuni.cz)
10:55.37dcordes_moin
10:55.42BabelO~ping pof
10:55.42aptpong pof
10:55.44BabelOhi dcordes_
10:55.46pH5_hi dcordes_
10:56.10dcordes_where's angstrom toolchain when you need it :(
10:57.23dcordes_pH5_, I tried to build the kaiser machine consoel-image with linux-rp as provider virtual/kernel. that also ended up in the 2.6.21 error
11:02.13BabelOpH5_: where can i found XdaDevelopers/NbfUtils.pm
11:06.56*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
11:07.58pH5_BabelO: http://www.xs4all.nl/~itsme/projects/perl/
11:11.25Marexcr2, hi
11:11.43Marexcr2, looxn560_pm.c line 88 and 90 ;)
11:11.49Marexlooks pretty suspicious
11:23.52BabelOpH5_: thanks, now i recompile eol
11:24.49dcordes_SanMehat, ping
11:58.39*** join/#htc-linux aLeSD (n=alex@227.Red-83-44-153.dynamicIP.rima-tde.net)
12:06.21*** join/#htc-linux marbalon (n=balon@aefb79.neoplus.adsl.tpnet.pl)
12:06.48*** join/#htc-linux Oktal (i=mat@cpc3-cmbg8-0-0-cust1018.cmbg.cable.ntl.com)
12:30.19cr2hi
12:30.21cr2Marex: ?
12:31.28Marexcr2, there are two addresses that are the same
12:32.29Marex(or maybe my KT is a bit outdated)
12:32.43cr2i think i've already fixed that.
12:33.42Marexok, sorry then
12:34.14Marexcr2, Im now trying to get pxa_ll_suspend work on palmld
12:34.24Marexlooks like the bootloader does funny stuff
12:34.49Marexit jumps on 0xa1200000 if it's not been suspended and to 0xa0200000 if it was
12:34.59Marexat least that's what I got from disassembly
12:36.44cr20xa0200000 is ok.
12:36.55cr2do you have the SPL at 0xa1200000
12:37.17Marexso that's what it is
12:37.29MarexI was disassembling just the IPL
12:37.34cr2what is the "resume switcht" ?
12:37.48Marexwhat's that ?
12:38.15Marexsorry, this is my first disassembling of a bootloader ;)
12:39.17cr2RCSR bit 0x4 ?
12:41.20Marexyea, when RCSR is 0x4, it jumps to 0xa0200000
12:41.27Marexelse it jumps to 0xa1200000
12:41.41Marexgot this info from qemu :-)
12:43.24Marexah ... got somewhere
12:43.34MarexPM WRITE [0x0000001c] 0x00000000
12:43.34MarexPM WRITE [0x00000018] 0xdf12fe1b
12:43.34Marexpxa2xx_clkpwr_write: CPU frequency change attempt
12:43.34Marexpxa2xx_clkpwr_write: machine entered Standby mode
12:43.48cr2hehe. try n560 ipl in the emu
12:44.03cr2Marex: it's the same on most other devices.
12:44.23Marexcr2, yea :) but palm was something special every time ;)
12:44.38cr2i mean n560 is special
12:44.51MarexI looked in the bootloader of n560 too
12:45.00Marexthere are just some more checks
12:45.37cr2yes, but it's a completely different behaviour
12:45.57cr2you never get through to 0xa0200000 jump
12:46.27cr2the mdoc-g4 code is also very strange.
12:47.31cr2the registers are at +0x201000 and not at 0x1000 like in all other g4 ipls i've seen
12:47.46Marexoh damn
12:47.55Marexnow when I reset the system, it gets into infinite loop
12:48.09Marexcr2, hehe :)
12:48.34cr2can you check the 0x2c4 addr ?
12:48.52cr2it's ipl 2.14.G4 here
12:50.19*** join/#htc-linux dcordes (n=dcordes@dslb-084-062-043-235.pools.arcor-ip.net)
12:50.48Marexnothing there
12:50.59Marexbtw palmld has the ipl in flash, not in DoC
12:51.10Marexit has 512kb flash and a microdrive, you know ;)
12:51.32Marexpalmtt5's ipl looks pretty similar though, and that is DoC
12:52.38cr2ok.
12:52.39Pryanwenas
12:52.48cr2Marex: check the n560 ipl
12:53.13cr2if it's the same version, and what is at +0x2c4
12:53.13Marexcr2, will do later
12:53.16cr2ok
12:53.25Marexbtw it definitelly isnt the same version
12:53.43cr2not 2.14.G4 ?
12:53.49Marex0xd9 0x61 0x02 0x21
12:53.54Marexpalmld doesnt have G4
12:54.05cr2n560
12:54.41Marex0x02 0x40 0xa0 0xe3
12:54.48Marexat 0x2c4
12:59.26dcordesmarbalon: did you include the function for keymap change now into the kernel driver?
13:01.41cr2e3 is good, looks like ldr
13:01.50cr2i'll check
13:03.23Marexmov r4, #2
13:05.01cr22c4:   e59f13a4        ldr     r1, [pc, #932]  ; 0x670
13:05.07cr2hmm.
13:05.22cr2send me the ipl
13:10.56Marexsent
13:12.48marbalondcordes: working on it...
13:16.50*** join/#htc-linux sigma_1234 (n=sigma_12@196.38.218.25)
13:17.14sigma_1234is anyone here?
13:18.22dcordessigma_1234: no
13:19.11sigma_1234lol
13:20.01sigma_1234does any linux run on the htc oxygen s310? im sick of windows
13:20.28cr2Marex: thanks
13:20.34Marexnp
13:20.42Marexyou can tell me what to do with palmld ;D
13:20.53*** join/#htc-linux intheboat (n=intheboa@50.205-247-81.adsl-dyn.isp.belgacom.be)
13:24.39*** join/#htc-linux toi_ (n=pleemans@d5152D3B4.access.telenet.be)
13:33.44dcordesoh I just wanted to point him to linwizard kernel
13:33.54*** join/#htc-linux aLeSD (n=alex@227.Red-83-44-153.dynamicIP.rima-tde.net)
13:40.30*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
13:45.47*** join/#htc-linux patoche (n=root@lns-bzn-28-82-250-131-104.adsl.proxad.net)
13:55.29Marexcr2, maybe I can try writing some gpio switch in asm and put it at 0xa0200000 to see if it really jumps there
14:07.03*** join/#htc-linux AlGe (n=alge@chello080109231226.4.uni-klu.teleweb.at)
14:15.51*** join/#htc-linux mistadman (n=mistadma@adsl-250-197-71.msy.bellsouth.net)
14:21.46mistadmanHey, cr2. Can you explain something to me. Yesterday, you said the HTC Athena's keyboard was connected to a gpio port. Or did I not understand correctly!?!?
14:34.58*** join/#htc-linux aLeSD (n=alex@227.Red-83-44-153.dynamicIP.rima-tde.net)
14:53.39*** join/#htc-linux LunohoD_ (n=alex@e180077231.adsl.alicedsl.de)
15:12.11*** join/#htc-linux paulproteus_ (n=paulprot@wide-rose.makesad.us)
15:25.40cr2Marex: you have 2.14.G4 too
15:25.50cr2mistadman: SSP2 spi port
15:26.26Marexcr2, ok :)
15:26.47cr2Marex: hm. but there is a 4byte shift. i need to check
15:27.07cr2the code i have at 2c4 you have at 2c0
15:27.31cr2but it's the same +0x201000 offset ;)
15:27.46Marex:D
15:28.27cr2and on bit 4 (sleep) set , you jump there.
15:29.03cr2then it checks for G4 sig, the "BIPO" sig, and so on.
15:29.36cr2and in the end is ldr, pc   0xa0100000
15:29.50cr2which is difficult.
15:30.41Marexmaybe there is nop somewhere ?
15:30.42cr2maybe we should write a raw mdoc loader from this g4 code.
15:30.59cr2or better modify the docg3.c
15:31.23cr2the SPL is at 0xa0100000
15:31.26MarexIm nearly half-way done with linear algebra ;)
15:31.46cr2and this IPL part looks like an SPL loader
15:31.49cr2ok :)
15:34.55cr2Marex: the other also includes the "BIPO" check
15:35.22cr2and the g4-style stuff
15:35.55cr2btw, with +0x201000 too.
15:36.16cr2maybe it's some RAM in g4 ?
15:36.51Marexjeez :)
15:39.24cr2hm. palm is clever. the stack pointer is set into the pxa SRAM
15:39.39*** join/#htc-linux CVirus (n=GoD@82.201.178.187)
15:40.24Marexyea, they probably dont need ram at all
15:41.03Marex*for the boot process
15:48.30*** join/#htc-linux patoche (n=root@lns-bzn-28-82-250-131-104.adsl.proxad.net)
16:05.53*** join/#htc-linux kiozen (n=kiozen@rgnb-4db1fdd2.pool.einsundeins.de)
16:06.13kiozenapt: what is QLandkarte GT
16:06.14aptACTION is QLandkarte GT to understand recursion, you must first understand recursion
16:06.33kiozenok, just thought..
16:08.35mistadmanThank! cr2
16:24.25*** join/#htc-linux ChanServ (ChanServ@services.)
16:24.25*** mode/#htc-linux [+o ChanServ] by irc.freenode.net
16:45.23*** join/#htc-linux patoche (n=root@lns-bzn-28-82-250-131-104.adsl.proxad.net)
17:26.04*** join/#htc-linux mistadman (n=mistadma@adsl-250-197-71.msy.bellsouth.net)
17:31.49*** join/#htc-linux marbalon (n=balon@aefx144.neoplus.adsl.tpnet.pl)
17:57.02*** join/#htc-linux Marex (n=marex@vasut.kolej.mff.cuni.cz)
18:05.49*** join/#htc-linux patoche (n=root@lns-bzn-24-82-64-178-37.adsl.proxad.net)
18:32.35Marexcr2, have you found anything about palmld ?
18:32.42Marexabout the ipl that is ...
19:23.51*** join/#htc-linux marbalon (n=balon@aefx144.neoplus.adsl.tpnet.pl)
19:49.23*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
20:00.32*** join/#htc-linux ltxda0 (n=ltxda@unaffiliated/ltxda)
20:00.36*** join/#htc-linux pikapika (n=pikapika@mar75-8-88-164-227-147.fbx.proxad.net)
20:00.48pikapikahi
20:03.35BabelOhi pikapika
20:05.43dcordes_hey BabelO
20:05.58dcordesBabelO, do you have BA android?
20:06.24BabelOdcordes: i just run it a little , but no touchscreen, just playing around with keys
20:06.45dcordesis it faster than qtopia/gpe?
20:07.00BabelOno
20:07.14*** join/#htc-linux CVirus (n=GoD@82.201.178.187)
20:10.05AlGeBabelo, dcordes: hi
20:10.17BabelOhi AlGe
20:10.37dcordeshi AlGe
20:10.54BabelOAlGe: i ve added my messenegr app to image, now it works with, next time i add mp3 support
20:10.57AlGeanything, I missed during the last 2 or so weeks?
20:12.18AlGeBabelo: I tried your last image, I still have the strange behaviour of hangs on second boot ...
20:12.41BabelOAlGe: ok strange
20:12.45dcordesBabelO, mp3 will be cool
20:12.51dcordesdid you fix the alsamixer stuff?
20:13.57AlGeI didnt anything the last weeks with my BA ...
20:14.50BabelOdcordes: yes i add asound.state
20:15.11BabelObut working two days ago on  blueangel sound record
20:15.56AlGeBabelo: outgoing calls with your qtopia image work, incoming calls are of bad sound quality. Same thing as with gpe after setting sound parameters ...
20:16.14dcordesBabelO, that's cool. will you also put that in image?
20:16.32BabelOdcordes: no record does not work
20:16.51*** join/#htc-linux Asmooh (n=asmooh5@asmooh.limburgonline.nu)
20:16.55dcordesBabelO, with which problem do you work?
20:17.01dcordess/problem/program/
20:17.05BabelOdcordes: sound recording
20:17.07dcordesI mean how do you realize the recording
20:17.14dcordesis there a gnu program?
20:17.19BabelOwith "cat"
20:17.22BabelOor arecord
20:17.22dcordesok
20:18.01AlGeBabelo: sending an SMS also killed qtopia for me ... but I had no usb connection available so I couldnt debug further.
20:18.02dcordesdo you get the audio from TTYs0?
20:18.02Asmoohhmm haret 0.5 doesnt fit on my screen ;/
20:18.24dcordesAsmooh, you have smaller than qvga?
20:19.08Asmooheh hy have a SPV C500.. HTC typhoon
20:19.21BabelOAlGe: yes SMS send broke the qtopia thread which wait for ttySX data
20:19.26dcordesah one of the "smartphone" thingies?
20:19.55Asmoohyea
20:20.01Asmoohsmall ones
20:21.25Asmoohi copyed 20060625.img to my storage card using actve sync, is it possible thats the reason it doesnt find the image?
20:21.34AlGeAsmooh: I had similar problems on my HTC S310 (oxygen). I think the solution could be to write a script which starts automatically wieth haret start (startup.txt) and setup the usb connection there?
20:21.48Asmoohhmm
20:22.41dcordesAsmooh, http://vivien.chappelier.free.fr/typhoon/wince.html
20:22.51Asmoohtxx
20:23.31dcordeshope that helps
20:23.34dcordesdoes haret run at all?
20:23.52Asmoohyea it does
20:25.27AlGeBabelo: has the reason for the slow reacting qterminal been found (or more specifically for all these alignment traps which slow it down)?
20:26.07Asmoohi copyed the lines to default.txt but still doesnt fit on the screen
20:26.30BabelOAlGe: yes tsdogs found the reason, it was a cast of a structure into 32 bit word
20:26.32AlGeAsmooh: use startup.txt, then it will autorun
20:26.46BabelOAlGe: i upload the new in next image too
20:26.48AlGeAsmooh: then there is no need to fit on screen
20:27.43Asmoohwell AlGe i thought it would boot the .img but it doenst even find the .img
20:29.58AlGeAsmooh: you may have to modify the filenames for KERNEL and INITRD to fit? Regardless what you fill in default.txt haret will never fit the screen.
20:30.57Asmoohhmm okay
20:31.16AlGeAsmooh: and if I remember correctly you cannot interact without a touchscreen with the haret buttons, so you are left alone with writing all commands to soem autoexecuted file.
20:34.34AlGeBabelo: nice, but the font problem in qterminal is also still there to some extent (cursor and input position shifted) ...
20:50.27cr2hi
20:50.36BabelOhi cr2
20:50.58cr2BabelO: it's damn complicated to write in C what can be easily done in awk ;)
20:51.22*** join/#htc-linux dcordes_ (n=snoopdog@unaffiliated/dcordes)
20:51.24cr2NULL pointers and uninitialized stuff everywhere :)
20:51.32dcordes_Asmooh, "Place commands in a file called "startup.txt". When HaRET starts it will automatically run these commands." from http://handhelds.org/moin/moin.cgi/HaRET_20Documentation
20:51.55cr2*** glibc detected *** ./ssplit7: munmap_chunk(): invalid pointer: 0x00000000006cf2d0 ***
20:52.18cr2dcordes_: why not compile the linload.exe ?
20:52.46dcordes_cr2, I'm not quite aware of linload
20:52.51BabelOcr2: yes i understand, C and pointer, hard way is the string compare and co
20:53.10dcordes_cr2, what does it do what haret doesn't?
20:53.12AlGecr2: I recently discovered the power of valgrind when writing a small C program with lots of mallocs/frees :-)
20:53.53cr2AlGe: yes, 90% of function calls are calloc/free
20:54.22AlGethen give valgrind a try and let it count your errors :-)
20:54.25cr2i've already found 2 bugs, fixed them, rewritten everything to use strtok(), but still segfault somewhere else :(
20:54.41cr2valgrind segfaults too
20:54.52AlGeoops
20:55.15cr2yes, porting awk code to C is evil :)
21:08.23dcordes_bbl
21:08.48*** join/#htc-linux JEEB (n=kanakana@a88-112-94-153.elisa-laajakaista.fi)
21:33.29*** join/#htc-linux LunohoD_ (n=alex@e180068136.adsl.alicedsl.de)
22:01.03cr2lol
22:01.07cr2--16794-- VALGRIND INTERNAL ERROR: Valgrind received a signal 11 (SIGSEGV) - exiting
22:01.07cr2--16794-- si_code=1;  Faulting address: 0xF65C557A;  sp: 0x62311E44
22:01.07cr2valgrind: the 'impossible' happened:
22:01.07cr2<PROTECTED>
22:03.05*** join/#htc-linux mistadman (n=mistadma@32.129.57.81)
22:29.11*** join/#htc-linux patoche (n=root@lns-bzn-24-82-64-178-37.adsl.proxad.net)
22:39.54cr2valgrind is designed only for c++ programmers
23:01.29*** part/#htc-linux Asmooh (n=asmooh5@asmooh.limburgonline.nu)
23:26.30BabelOcr2: lol, the 'impossible' happenned:
23:26.40BabelOthis is a good tool, just miss the blue screen :)
23:29.16cr2but it does not find any errors
23:29.26cr2just segfaults itself
23:29.45cr2i've fixed all the bugs looking at the source, and printf()
23:29.58cr2so it's mainly useless :)
23:33.52BabelOlol
23:33.58BabelOdid you try electicfence ?
23:34.12BabelOcr2:
23:35.50cr2no, i need to install it first.
23:35.51BabelOsorry ElectricFence
23:36.11BabelOok but just usefull for malloc and non threaded program
23:38.31BabelOok i ve a CID unlocker for artemis, i can try  the eol boot tomorrow :)
23:43.35cr2good night
23:47.48BabelOgood night

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