irclog2html for #htc-linux on 20061206

00:03.16*** join/#htc-linux pof (n=pof@62.57.1.173) [NETSPLIT VICTIM]
00:04.49*** join/#htc-linux scatterp (i=scatterp@66.252.10.253) [NETSPLIT VICTIM]
00:04.53*** join/#htc-linux cr2 (n=konversa@crpl22.physik.uni-wuppertal.de) [NETSPLIT VICTIM]
00:05.42*** join/#htc-linux skodde (n=skodde@85-18-136-66.fastres.net) [NETSPLIT VICTIM]
00:05.43*** join/#htc-linux Kmarc (i=kari@83.145.242.1) [NETSPLIT VICTIM]
00:06.36*** join/#htc-linux mndctrl_ (n=mind@2.81-167-1.customer.lyse.net) [NETSPLIT VICTIM]
00:07.16*** join/#htc-linux anYc (i=mario@hadince17.hadiko.uni-karlsruhe.de) [NETSPLIT VICTIM]
00:07.25*** join/#htc-linux toi (n=pleemans@d5152D3B4.access.telenet.be) [NETSPLIT VICTIM]
01:03.06*** join/#htc-linux Ralith (n=ralith@soggy202.drizzle.com)
01:14.04*** join/#htc-linux ralith_ (n=ralith@soggy202.drizzle.com)
01:25.35lkclo fer goodness sake.  i got the same thing, BabelOued.
01:25.37lkclwtf??
01:25.48lkclnuts, he's gone.
01:30.12lkclhow odd.
01:32.22*** join/#htc-linux Ralith (n=ralith@soggy202.drizzle.com)
01:34.31*** join/#htc-linux Ralith (n=ralith@soggy202.drizzle.com)
01:35.26lkclargh tsc2200.h updated file got lost
01:37.25*** join/#htc-linux Ralith (n=ralith@soggy202.drizzle.com)
01:39.11*** join/#htc-linux psokolovsky_ (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com)
01:46.58*** join/#htc-linux g3gg0__ (n=g3gg0@ppp-62-245-162-240.dynamic.mnet-online.de)
01:51.37psokolovsky_lkcl, where it was lost?
01:51.47psokolovsky_lkcl, do you work on .19?
01:51.51lkcldunno - probably by me.  old version creeped in
01:51.57lkclpsoko: i work in cvs main
01:52.19lkclan old version of tsc2200.h crept in (well, older as in from last week)
01:53.50lkclpsokolovsky: i am having fun, trying to get blueangel up-and-running.  i talked to pH5 and to liam, i have pH5's asoc magician code, and i have to merge asoc 0.12.4 to get _that_ to work...
01:54.39psokolovsky_lkcl, please spend some time to convert sable's drivers to be proper platform_driver. did you have resume working on sable?
01:55.21lkclno, i never got resume working - i've stopped on sable, for now, until cr2 gets his device up-and-running.
01:55.31lkclthen i will re-focus on sable.
01:56.03psokolovsky_lkcl, ok. but guys, please convert all your driver to be platform, so we all know that .19 is good for all us.
01:56.10lkclok.
01:56.18lkclany good examples to follow?
01:56.25psokolovsky_quick grep shows you have lots unconverted == resume won't work for those devices
01:56.39psokolovsky_lkcl, it was discussed in k-d ;-)
01:56.45lkcloh ok :) ha ha
01:56.47psokolovsky_i'll drop you link, but later...
01:56.55lkclno it's ok i find it
01:57.14lkcli got time.
03:45.54*** join/#htc-linux bd2 (n=ufo@89.113.84.151)
03:59.29*** join/#htc-linux Kevin2 (n=Kevin@207-237-52-30.c3-0.avec-ubr12.nyr-avec.ny.cable.rcn.com)
04:07.04Kevin2psokolovsky: I'm here now.  Otherwise - feel free to email me.
04:08.09psokolovsky_Kevin2, Hi! So, do you have any plans/ideas on at least putting different haret snapshot builds online at easily available place?
04:09.23Kevin2psokolovsky_: I'd like to, but I don't have a hosting facility to publish them at.
04:10.01psokolovsky_Kevin2, well, i can handle that, just would like to be sure I don't step on your feet here.
04:10.41psokolovsky_Kevin2, and - you have a webspace as hh.org account holder, http://hh.org/~username ;-)
04:10.41Kevin2Not at all.  If you're going to push an exe, let's tag it first.
04:11.01Kevin2psokolovsky_: How do I upload to that space?
04:11.22psokolovsky_Kevin2, yes, I noticed that there was no tags for some time, since 0.4.4. if there's something to tag, please do
04:12.01psokolovsky_Kevin2, ssh to hh.org, and anything under ~/public_html/ will show at that link (subject to perms, of course)
04:12.15Kevin2Well, nothing major - just a bunch of minor improvements, but it is always good to be able to back track to a built exe that is distributed.
04:12.17lkclpsoko: who's the best person to review an asoc 0.12.4 hh.org cvs patch?
04:12.19psokolovsky_so, you can use scp once dir structure is set up
04:12.37lkclthe last asoc update was... 0.12? 0.11?
04:13.12psokolovsky_lkcl, please make sure Spyro knows of your plans to commit it. He commit asoc originally, and I guess, the only stakeholder for it besides you.
04:13.21lkclpsoko: ok, ta.
04:13.39lkclok he's off at the mo.
04:13.39psokolovsky_lkcl, if Spyro wrote good commit msgs, we'd know what version it was ;-)
04:13.44lkclha ha
04:14.01lkclwell pH5 is the other person who needs it.
04:14.07psokolovsky_just forward him your mail personally, I guess. and wait a day ;-)
04:14.16lkclhe's working on magician, with liam
04:14.17lkclokie :)
04:14.21psokolovsky_nice!
04:14.46Kevin2psokolovsky_: Wow, I didn't even know I had shell access.  :-)
04:15.13psokolovsky_Kevin2, is you have CVS over SSH, you have shell, unless it's forcibly closed ;-)
04:15.50Kevin2Is there anything I should know about proper use/restrictions?  I assume it is okay if I put haret binaries up there.
04:16.15psokolovsky_Kevin2, so, my plan was to upload 0.4.2, 0.4.3, 0.4.4 and let people test it, untill we'll be able to spend more time on QA and roll out "official" release
04:16.59Kevin2I think we should just push the latest release.  There were some bugs fixed from the previous releases.
04:17.21psokolovsky_Kevin2, my assumption is just the same. I never heard/read anything like that, just saw other people using it. So, I assume if it's directly related to your hh.org work, and not bigger than few MBs, it's ok.
04:18.05psokolovsky_Kevin2, yep, bugs fixes, and regressions made ;-) remember, ~0.4.2 worked on h3900, latest - doesn't ;-)
04:18.40Kevin2I didn't know latest didn't work on h3900 -- what's wrong?
04:19.48psokolovsky_Kevin2, ah, damn, not latest ;-). 0.4.3+cvs. I told you couple of times - it doesn't boot linux, just drops of to wince. I actually still didn't test latest ;-(
04:21.16Kevin2psokolovsky_: Well, prior to my rewrite of the linux boot code, I really didn't change anything in the linux boot code.  I can say that the previous code was pretty fragile (I think changing code locations could cause destabalization).  I'd be curious to see if the latest code works okay.
04:21.42psokolovsky_Kevin2, I'll check once .19 kernel works well enough, I swear ;-)
04:22.25psokolovsky_Kevin2, so, of course feel free to do it the way you think it's good if you'll handle publishing snapshots
04:22.54psokolovsky_if we'll get reports of specific regressions for some devices with latest, we can provide older snashots then
04:24.42bd2psokolovsky_, hey!
04:25.11psokolovsky_bd2, good morning ;-D
04:25.39bd2heh, yeah.. lately I'm indeed sleeping at night, and wakes up early. feels great ;-)
04:25.49psokolovsky_cool!
04:39.22Ralithanyone here dev'ing for/on the Universal?
04:39.38Ralithmore importantly, anyone know about the status of the DoC drivers?
04:40.41Kevin2psokolovsky_: I've tagged and upload haret to http://www.handhelds.org/~koconnor/haret-0.4.5.exe
04:41.11Kevin2Ralith: There are several developers with the Universal - it is probably the most supported device.  There is no DoC support currently.
04:42.06psokolovsky_Kevin2, nice, thanks! maybe worth to create a subdir for haret just now, in case later you'll want to post kernels, patches, etc. there...
04:43.35Kevin2psokolovsky_: Good point.  I've moved the file.
04:46.26RalithKevin2, on http://www.handhelds.org/moin/moin.cgi/UniversalStatus I see Internal Flash  TBD  TBD  Should be ready, if someone dears to try feel free to contact us at htc-pxa@handhelds.org
04:46.51Ralithand I'm wondering exactly what status it's at, and what the risks are in 'daring to try'
04:48.33Kevin2Ralith: I've heard rumors that some people on another project have done some work, but I don't know anyone who has tried to get it to work on a PDA.  You might want to ask "cr2" if he knows the status.
04:49.17RalithKevin2, handhelds.org is 'another project'?
04:49.30Ralithit says right there it's ready to be tested O.o
04:49.53Ralithalso, on xdadevelopers I've seen someone comment that there was working barebones code, but implied it needed work
04:53.57Kevin2Ralith: To the best of my knowledge, it is ready for someone to try coding it.  But I could be wrong - I haven't really looked at that area yet.
04:54.17Ralithas in, the information exists but it's yet to be implemented?
04:55.11Kevin2M-Sys published docs on the chip - I have the pdf on my harddrive now.  I haven't seen anyone with a driver yet.  (But as you noted, I've heard "rumors" that someone has some bare bones stuff working.)
04:56.19Kevin2Since we can run everything off of SD/Mini-SD, it hasn't been a priority to get the DoC working yet.  Getting the phone working, suspend/resume, etc. have had a higher priority with many of the developers.
04:56.50Ralithyeah, that makes sense
04:59.08Ralithit's good to hear that the docs are published, though
04:59.30Ralithdo you have a general idea how well development in the rest is moving along?
05:11.03Kevin2I'd say progress is definitely being made.  There is even initial support for making phone calls on Universal.
05:11.32psokolovsky_Kevin2, I guess you know, that G2 (32mb) support is in CVS now, and h6300 said they had G3 support or something...
05:12.53RalithKevin2, yeah, I saw that on the wiki. I'm planning on getting one soon and very eager to have it functional under linux.
05:13.04Ralithsounds like it won't be too long :)
05:15.28Kevin2psokolovsky_: No, I didn't know that.
05:16.12psokolovsky_Kevin2, h4000 reads its flash pretty well. and I'm waiting for some madhead to make/test write ;-)
05:16.20Kevin2psokolovsky_: Oops - I forgot to modify the makefile to report version 0.4.5 -- I've fixed up cvs and pushed that exe again.
05:16.52psokolovsky_Kevin2, ok ;-) just in case, cvs tag has -F option to move tag ;-)
05:17.13Kevin2Yeah, I just used "-d" and then reapplied it.
05:17.30Kevin2(After creating a "Makefile" tag, "help" tag, etc -- but I deleted those too.)  :-)
05:25.48*** join/#htc-linux asylume1 (n=insanity@196.211.28.91)
05:26.35*** part/#htc-linux asylume1 (n=insanity@196.211.28.91)
06:18.16*** join/#htc-linux jgw (n=jgw@24.22.48.124)
06:18.35*** part/#htc-linux jgw (n=jgw@24.22.48.124)
06:19.50*** join/#htc-linux jgw (n=jgw@24.22.48.124)
06:20.06*** part/#htc-linux jgw (n=jgw@24.22.48.124)
06:24.30*** join/#htc-linux jgw (n=jgw@c-24-22-48-124.hsd1.mn.comcast.net)
06:27.18*** join/#htc-linux jgw (n=jgw@c-24-22-48-124.hsd1.mn.comcast.net)
06:27.49*** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk)
06:49.10*** part/#htc-linux jgw (n=jgw@c-24-22-48-124.hsd1.mn.comcast.net)
06:58.00*** join/#htc-linux rob_w (n=rob_w@p549B8D52.dip0.t-ipconnect.de)
08:06.30*** join/#htc-linux Fossi (n=cmewes@195.177.48.115)
08:17.36*** join/#htc-linux goxboxlive (n=goxboxli@gw.hjelcowi.no)
09:08.30*** join/#htc-linux booba (n=booba@AAnnecy-256-1-57-131.w90-10.abo.wanadoo.fr)
09:23.34*** join/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com)
10:16.49*** join/#htc-linux TripleDES (n=sergio@89.129.143.241)
10:23.00*** join/#htc-linux rob_w (n=rob_w@p549B8D52.dip0.t-ipconnect.de)
10:50.14*** join/#htc-linux g3gg0_ (n=g3gg0@ppp-82-135-83-49.dynamic.mnet-online.de)
11:27.37*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
11:34.24*** join/#htc-linux ellisway (n=ellis@host-87-74-241-174.bulldogdsl.com)
11:43.47goxboxlivecr2: Are you around?
12:15.06babelouedlkcl : you want some more files to compile ? sound is working ?
12:18.16*** join/#htc-linux WizMaui (n=WizMaui@62.112.90.231)
12:37.57cr2goxboxlive: does the kernel resume now ?
12:39.17goxboxlivecr2: No
12:39.30goxboxliveThe keyboard doesnt work either after boot
12:39.59goxboxliveIt suspends (via ssh) but doesnt wake up with the power button
12:40.21goxboxliveBut i can see that it is working via cat /dev/input/event1
12:45.12*** join/#htc-linux frankie (n=nnnnnnfr@lugbari/people/frankie)
12:48.00cr2hmm. ok.
12:48.02goxboxliveBut cat /dev/input/event3 is dead (keybaord)
12:48.17cr2there must be some other bug.
12:48.37goxboxlivewell, the keybaord worked before your changes yesterday.
12:51.51cr2i know :)
13:07.48*** join/#htc-linux psokolovsky (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com)
13:09.12*** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk)
14:06.20lkclbabeloued: sure.  no not yet, i have debug on gotta test.  alsa device not registering.
14:06.21lkclbbl
14:06.38babelouedlkcl : hi
14:07.16babelouedi send you correct header file for tsc2200.h
14:10.59lkclyou verified what i committed just today?  i found the newer version.
14:11.53babelouedah ok, ok i m connecting at my home network and trying it
14:12.05babelouedany success with sound this night :) ?
14:25.45babelouedlkcl : it is good now :)
14:28.48babelouedlkcl : i can't add you at linkedin i m not registered :(
14:30.13lkclbabeloued: oops
14:30.28lkcli will see.  asoc looks horrendously complex.
14:32.00lkclbabeloued: what's your email address?
14:36.09babelouedlkcl : you got my message ?
14:39.50lkclyep
14:43.32cr2lkcl: you have some H4000 remnants in sable* is it intentional ?
14:56.17babelouedlkcl :thanks
15:00.00*** part/#htc-linux goxboxlive (n=goxboxli@gw.hjelcowi.no)
15:01.06*** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk)
15:03.54*** join/#htc-linux LunohoD_ (n=alex@e180078159.adsl.alicedsl.de)
15:04.38*** join/#htc-linux ccureau (n=chris@cl-64.atl-01.us.sixxs.net)
15:07.08ccureauAny suggestions for trying to discover what the gpios in the Wizard actually do?
15:08.25*** join/#htc-linux [Sk] (n=sk@dag94-1-82-66-35-175.fbx.proxad.net)
15:08.47cr2if your haret supports direct memory access and you know where are the registers, then yes.
15:09.22cr2i'm not familiar with omap850, so can't tell more.
15:11.36ccureauI'm not really familiar with it either...they're pretty secretive devices
15:13.04ccureauI have a haret dump...I'm just not sure where to go from there
15:18.48cr2but there is some working code for omap730
15:19.17ccureauAre those the tornadoes?
15:20.36cr2yes.
15:22.07ccureauRead about those
15:22.11cr2you can also get a demo of ida pro.
15:22.25ccureauida pro?
15:22.30cr2and check the wince binaries.
15:22.40cr2disassembler.
15:22.47ccureauah :)
15:23.34ccureauI'm assuming it runs in windows?
15:24.55ccureauwell, maybe it'll run with wine
15:29.42cr2it works in wine
15:30.00cr2otherwise i'll not mention it :)
15:33.05ccureau:D
15:33.28ccureauSo, really, all I'd need to do is find the program that does the function I'm interested in and trace through
15:33.35ccureauI can give that a shot
15:34.28ccureauI wonder if WinCE will allow you to access the bare hardware
15:35.26LunohoD_there is also ida pro for linux
15:35.57ccureauI didn't see it as an evaluation download
15:37.04LunohoD_i have only cracked version
15:37.18ccureaucontribute? :D
15:38.14LunohoD_what do you mean?
15:39.05ccureauAsking to share the cracked version...but I can find it elsewhere.
16:04.11*** join/#htc-linux JTRipper (n=jtripper@p54B1D6A1.dip.t-dialin.net)
16:32.21*** join/#htc-linux pH5 (n=ph5@e178231180.adsl.alicedsl.de)
17:04.34*** join/#htc-linux asylumed (n=insanity@196.211.28.91)
17:24.11*** join/#htc-linux booba_ (n=booba@AAnnecy-256-1-109-101.w90-9.abo.wanadoo.fr)
17:40.06*** part/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com)
17:42.35*** join/#htc-linux rob_w (n=bob@p85.212.155.144.tisdip.tiscali.de)
18:10.45*** join/#htc-linux BabelOued (n=Fabrice@82.238.28.28)
18:21.58lkclLunohoD: don't use cracked software around here.  it is _vital_ that you do not use it to do reverse-engineering.  you can be sued by manufacturers - and they will win, not because of the case of copyright but because you had unlicensed or illegal software.
18:22.26lkclso please - DO NOT use any cracked software.  the license for ida pro is only $400 - the same as a phone.
18:26.41*** join/#htc-linux asylumed (n=insanity@196.211.28.91)
18:27.01LunohoD_lkcl: i agree with you
18:27.20ccureaulkcl: isn't it as likely that they will sue you for reverse-engineering in the first place?
18:28.32LunohoD_but i won't pay $400 for the software i do not use
18:35.54*** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com)
18:35.57pierroxhi
18:36.10BabelOuedhi pierrox
18:36.21BabelOuedhi goxboxlive
18:36.35goxboxlivehi
18:36.52pierroxhi BabelOued
18:37.08BabelOuedgoxboxlive: you use opie from cvs to builduniversal image ?
18:37.40pierroxlkcl: did you find the time to commit some parts of the g500 patch ?
18:38.05goxboxliveBabelOued: Yes i am building from cvs. Otherwise i will miss some of the Universal spesified stuff. They where not suplied in the 1.2.2 version.
18:38.52BabelOuedgoxboxlive: ok. backlight command in opie works for universal ?
18:39.41goxboxliveDo you mean adjsuting the backlight?
18:39.54BabelOuedoh... yes
18:40.03goxboxliveYes it works, but instead of pxa-fb we use corgi-bl
18:40.16BabelOuedme too
18:40.41BabelOuedi see /sys/class/backlight/corgi-bl but opie can't adjust
18:41.46rmoravcikpierrox: hi
18:42.00pierroxrmoravcik: hi
18:42.45pierroxrmoravcik: lkcl said me that you are working on s3c24xx
18:43.00rmoravcikpierrox: if you want rtp has new version of his patches http://www.rtp-net.org/ipaq/patches/2.6.19/
18:43.19pierroxrmoravcik: yes i had a look at it today
18:43.23rmoravcikpierrox: yes, i'm working on rx3715 (s3c2440)
18:44.10pierroxrmoravcik: i took several things from rx3715, it was really helpfull :-)
18:44.39rmoravcikpierrox: i saw, you make some changes in touchscreen driver
18:45.02pierroxrmoravcik: yes, but it is not really good. i had some problems with the original version
18:45.50rmoravcikpierrox: did you try discuss this problem with rtp?
18:45.51pierroxrmoravcik: i have made some cleaning (indentation + #ifdef) http://www.pierrox.net/G500/20061205/linux-2.6.19-phe-g500.diff
18:46.27pierroxrmoravcik: yes but at this time my mail did not reached it (I may have a wrong email ?)
18:46.42pierroxreached him, sorry (french)
18:49.06rmoravcikpierrox: if you want thare is a universal gpio keyboard driver in hh.org cvs, you can use it. it has support for s3c24xx
18:49.26goxboxlive<PROTECTED>
18:49.43pierroxrmoravcik: yes, i use it :-) but some buttons are not on gpios (?)
18:50.08pierroxmoravcik: did you have succes on PM on the s3c2440 ? it seems that existing s3c2410 code is not far from s3c2440, but does not resume at early stage for me
18:50.13rmoravcikpierrox: so where are connected?
18:50.34pierroxrmoravcik: don't know yet, there is a single interrupt for several buttons
18:50.55rmoravcikpierrox: yes, we finish it yesterday
18:51.15pierroxrmoravcik: really ? great !
18:51.58BabelOuedgoxboxlive: try include input.h :)
18:52.09rmoravcikpierrox: http://trinity.fluff.org/ben/linux-26/2619/
18:57.54*** join/#htc-linux florian (n=fuchs@84-245-183-76.hbg.bpool.celox.de)
18:58.15pierroxrmoravcik: wonderfull ! there are not a lot of diffs
18:58.21florianre
18:58.27pierroxre
18:59.21rmoravcikpierrox: i will look to your patch today, and i think we can start commiting with rtp's last udc patch
19:00.44pierroxrmoravcik: most of things included in this patch comes from rx3715 and h1940, except the modified ts driver and mci driver
19:01.34pierroxrmoravcik: s3c2440 could be merged in s3c2410mci, but my version is not clean : I have an unknown crc error on write
19:01.56rmoravcikok, but i want commit only rtp's last patch, for example udc driver in your 2.6.18 patch was little bit old
19:02.11pierroxrmoravcik: yes, that is true
19:02.20pierroxat least 2 months old
19:02.45rmoravcikpierrox: maybe you can try send mail to ben dooks, he is a expert to s3c24xx
19:02.52pH5hey, are you interested in writing a s3c24xx patch to support the common gpio api (see http://lkml.org/lkml/2006/11/30/331)?
19:03.33pH5this is what I hacked together, but I don't have any samsung device: http://en.pastebin.ca/269653
19:04.13pierroxpH5: not me, i am not qualified enough for that
19:08.14rmoravcikph5: i think it looks good
19:14.36goxboxlive<BabelOued> yes that's right, now it compiled
19:14.38goxboxlivethx
19:15.26goxboxliveBUt it didnt help our keybaord working
19:19.38cr2goxboxlive: i'll test the touchscreen patch now, and commit the corrections.
19:20.26cr2pierrox rmoravcik: should we create device subdirectories for s3c24xx as it is done on pxa ?
19:20.29goxboxlivecr2: Touchscreenpatch? What's wrong with our TS? It has never worked as good as now.
19:21.08cr2goxboxlive: it should be more generic. i can reuse it for sable without rewriting.
19:22.12cr2pH5: can we try to operate the tsc2046 in 16bit or 32bit mode ? the docs say something about 16bit mode, but will the 32bit mode too slow ?
19:22.36goxboxlivecr2: ok
19:23.00pierroxcr2: imho it should be preferable not to duplicate too much files, i know that i don't help for that however
19:25.10lkclcr2: the subdirectories is happening already, sound/soc/s3c244x etc.
19:25.59cr2lkcl: in arch-s3c2410
19:26.08lkclgoboxlive: do a diff between the htcsable_ts2.c and htcuniversal_ts2.c and you will see that the only locations which are different are the GPIO pins and maybe the SSCR1 instead of SSCR2
19:26.26lkclcr2: oh nuts.  it should be called arch-s3c24xx
19:26.35goxboxlivelkcl : ok
19:26.36pH532 bit mode? I thought it was 24bit
19:26.53rmoravcikcr2: in the future there will be probably more s3c24xx devices, so maybe it is good idea
19:26.55cr2it 24bit now, so we can't use pxa2xx-spi
19:27.08pierroxlkcl: sound/soc/s3c244x ? where ? i want it !
19:27.15cr2lol
19:27.28lkclhe he - give me a chance to clean up the asoc-0.12.4 patches ok? :)
19:27.38pierrox:-)
19:27.41lkcli need to talk with Spyro to make sure he is happy with it.
19:27.49lkclplus, you will need to write a joining-module
19:29.48pierroxlkcl: as soon as i get pm working i will switch on sound
19:30.45rmoravcikpierrox: i have done sound on my rx3715 with alsa s3c2440-iis, but not with asoc yet
19:31.27*** join/#htc-linux asylume1 (n=insanity@196.211.116.2)
19:32.44pierroxrmoravcik: fluff is working on sound too ?
19:33.52rmoravcikpierrox: on asoc implementation
19:34.04pH5cr2: hm. the tsc2046 has 8bit and 12bit modes.
19:34.34pierroxrmoravcik: and the other parts are device dependant ?
19:35.11rmoravcikpierrox: yes
19:36.06*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
19:36.26cr2pH5: the ADC. i'm talking about spi. you send the command shifted <<16 now, and receive the 16bit sample data.
19:36.55rmoravcikpierrox: http://morgan.fbi.cz/tmp/kernel26-2.6.19-hh4.patch my patch for rx3715 with sound support
19:37.14cr2i suggest to try 2 options. set spi to 16bit, send <<8 , recv 16 bit. there may be some timing issues according to the manual.
19:37.40pierroxrmoravcik: thanks, i will have a look at it
19:37.55cr2or set spi to 32bit, send <<24, recv >>16. should work, but may be slow (?)
19:38.33pierroxi have enough patches to spent the whole night on my device, great :-)
19:39.22rmoravcikpierrox: but kernel with sound sometimes crash http://pastebin.ca/269717
19:39.42cr2goxboxlive: the keyboard is broken by my (braindead?) platform_device patch. the "new" touchscreen works. i need the opie with .19-hh2 modules ;)
19:40.15goxboxlivehmm, i'l see if i have any, hold on
19:40.37pierroxrmoravcik: nice ! but i don't fear to reboot, i have made that at least a billion times on my g500 :-)
19:41.04rmoravcikpierrox: me too
19:42.02goxboxlivecr2: I have one, shall i upload it to gnulinux.biz?
19:42.09pierroxrmoravcik: the reset button is the most used button on theses toys
19:42.26rmoravcikpierrox: :)
19:44.42pH5cr2: so in either case, we are using the slow 24-clock per conversion mode, right? first the cmd byte is sent, and then two bytes (containing the 12 bit result) are read afterwards
19:45.11cr2goxboxlive: i'd like to debug the keyboard first.
19:45.28cr2pH5: we waste the SPI bandwidth.
19:46.20goxboxlivecr2: Well, anyway it is upped. But, just so you know, hh4 is now tagged.
19:46.28cr2the commands are 8bit, and the samples are 12 (need 16bit) 16 bit spi will be optimal. 32bit spi will be 2 times slower. 24bit spi as now is 1.5 times slower.
19:46.46cr2hh4 ?
19:46.54pH52.6.19-hh4
19:47.04cr2why do i have .12 ? :)
19:47.12cr2-hh2 , sorry.
19:47.44goxboxlivewell, hh4 where tagged today, doint ask me about hh3. Havent seen it.
19:54.55*** join/#htc-linux disturbedsaint (n=simon@212-127-222-63.cable.quicknet.nl)
20:00.35goxboxlivecr2: While you are updatings some of the universal drivers, be sure to add input.h to htcuniversal.c. It is missing and failes compiling without it.
20:00.54asylumedcr2: hi
20:01.14cr2goxboxlive: yes, otherwise i could not compile anything.
20:01.22cr2hi asylumed
20:02.06asylumedcr2: just looking at the tsc2200 stuff that got comitted...
20:02.09asylumedcr2: do you think the tsc2200_keys and tsc2200_ts stuff should go into the input/keyboard and touchscreen directories?
20:04.25cr2asylumed: if they will use pxa2xx-spi, then yes.
20:05.32asylumedcr2: k
20:07.51cr2asylumed: tsc2200 uses 16bit spi ?
20:11.36asylumedcr2: yes
20:13.08cr2asylumed: look at magician.c how pH5 uses pxa2xx-spi
20:14.45cr2goxboxlive: i have reverted the keyboard patch, testing the driver now. it may make resume unhappy, but is still better than the broken keyboard.
20:15.20goxboxliveyes  :-)
20:15.25asylumedcr2: why is it commented out though?
20:17.42cr2asylumed: because pxa2xx-spi does not support 24bit transfers.
20:20.34cr2goxboxlive: why opie switches to tty2 at start, and not to the first free tty ? maybe that explains psplash fiasco ?
20:21.43goxboxliveyes, that might be the reason. Is there way to let opie use the first free tty?
20:22.18goxboxlive<PROTECTED>
20:22.30goxboxlivecould that be the reasom?
20:22.42goxboxlives/reasom/reason/
20:23.22goxboxlive<PROTECTED>
20:25.57cr2yes. since you have enabled many virtual consoles in /etc/inittab, you should change it too.
20:26.35*** join/#htc-linux asylumed (n=insanity@196.211.116.2)
20:26.51*** join/#htc-linux WizMaui (n=WizMaui@62.112.90.231)
20:27.38goxboxliveBut if i remove that, i will only have one console, and that's not good. It's great to have this i mean thow.
20:30.06pierrox|awgood night !
20:31.07cr2goxboxlive: why do you need to set the terminal there ? X usually takes the first available, why not qpe ?
20:33.14goxboxliveWell i have not set the terminal there. That's opie default as far as i know. But in my initab i have tty1-4 and would still like to have that. But do you mean if i just remove the "2" from 'opie' it will solve the psplash issue?
20:35.21cr2it may.
20:35.30cr2keyboard driver updated.
20:36.16cr2hmm. but universal still does not want to resume.
20:37.04cr2the only remaining driver that hx4700 does not have is pxa27x-ohci
20:37.29goxboxliveno, i know now why psplsh use fuc*es upopie- If i pu terminal 3 instead of termional 2 in opie i guess it will work. I tried to remove "2" now, and the shell is showing as same as psplash . So i guess psplash is using tty2 and therefor Opie should use tty3
20:37.53rob_wcr2,  do you have any ideas on how to setup a 3d env , for kdrive ?
20:39.58cr2rob_w: it requires writing a device driver inside Xserver. not something that i'd like to do :)
20:40.47cr2you are opening a very big can of worms.
20:42.27rob_waiii
20:43.06rob_wi just dont how chip companys dont ship samples of how they would like their hardware used
20:43.26rob_wits a pitty that one needs to "explore" all those things on his own
20:49.41*** join/#htc-linux asylumed (n=insanity@196.211.28.91)
20:51.08*** join/#htc-linux WizMaui_ (n=WizMaui@62.112.90.231)
21:00.39goxboxlivewell you have that right, it doesnt want to resume. But the keyboard works now. But i thought you also fixed the evdev stuff that where hardcoded, but it doesnt seems like it since udev dont create a touchscreen0.
21:06.28lkclgoboxlive: i heard from psokolovsky yesterday, and he says that there's a message on kernel-discuss describing the resume code redesign, as part of the platform_devices infrastructure
21:06.38pH5goxboxlive: compare /sys/class/input/input?/modalias with the expression in the  udev rule
21:07.07lkclwe can expect resume failures (like i do with the sable) until we update all of the htc stuff to use the more recent code, like everyone else is.
21:09.12goxboxlivelkcl: ok, i thuoght cr2 tried that out yesterday wich broke our keyborad.
21:09.33goxboxlivepH5: ok,i'll take a look
21:09.54cr2goxboxlive: i have converted the keyboard to platform_driver.
21:10.30goxboxliveok, then we are up 2 date, right?
21:11.43cr2i think yes, but something is still broken.
21:11.58cr2try to remove IR and OHCI.
21:12.16cr2from the kernel config, and resume.
21:12.26pH5cr2: OHCI doesn't have proper suspend/resume support, but it never stopped sus/resume from working on the magician
21:12.50pH5and IR was fixed by paul today or yesterday
21:14.28cr2pH5: i don't have "visible" non-platform_device drivers now.
21:15.11cr2pH5: i have abstracted the htcuniversal_ts2.c to the most extent i could. have a look.
21:15.41cr2but it still needs pxa2xx-spi as a parent.
21:29.09pH5cr2: are you still using char/sa1100-rtc ?
21:31.10goxboxlivecr2: No, removing irda and OCHI didnt help that either.
21:32.05psokolovskycr2, and was pxa2xx-spi fixed?
21:32.29pH5wasn't pxa2xx-spi a platform driver from the beginning?
21:33.10cr2pH5: sa1100-rtc ? i think yes.
21:33.29cr2i wish there was pxa27x-rtc
21:33.32pH5the char one is a device_driver :-/
21:33.37pH5cr2: write it!
21:33.47cr2audio first :)
21:34.11cr2goxboxlive: try to remove sa1100-rtc
21:34.16cr2GPIOB10 S 0 EXT_IRQ 3302911
21:34.17cr2GPIOA3 S 1 EARPHONE_PWR 3302926
21:34.23pH5cr2: :-)
21:34.44cr2hmm. what does htc mean under ext_irq ?
21:34.55cr2i think it is available on all htc phones.
21:35.13psokolovskypH5,  pxa2xx-spi was *not* platform_driver yesterday.
21:35.21cr2it's some signal from BB that leads to big audio switching.
21:35.36cr2and it's not handled now at all.
21:35.47pH5hey lkcl, thanks for updating soc
21:35.49psokolovskypH5, the char one is deprecated, noone should use it now, until further invetsigation is taken place ;-I
21:36.04pH5psokolovsky: pxa2xx-spi is a  platform_driver in my vanilla 2.6.19 tree at least
21:36.39pH5and in 2.6.18, too
21:36.45psokolovskypH5, oops, yeah
21:37.00psokolovskypH5, sorry, did grep on "platform_bus_type" ;-)
21:37.43pH5yeah, I made that mistake, too, at first :)
21:37.52psokolovskyæ-)
21:37.55psokolovsky;-)
21:38.49psokolovskypH5, so any quick ideas what to do with rtc stuff?
21:38.51lkclpH5: no problem
21:40.03goxboxlivecr2: removing sa1100-rtc made it resume. But keyborad is still broken
21:40.05pH5psokolovsky: is there an issue with drivers/rtc/rtc-sa1100.c?
21:41.08psokolovskypH5, yes, it's incorrect per RTC framework. it's actually a copy of char/ one
21:41.37psokolovskypH5, my plan would be: resurrect mainline version of drivers/rtc/rtc-sa1100.c, and ask everyone use it
21:42.33psokolovskypH5, then fix char/ version just in case, and then let people test rtc/. then, ask Todd what changes and why he did to char/, then, port any changes to rtc/.
21:42.45goxboxlivecr2: and also usb0 is not resuming either.
21:42.57pH5psokolovsky: yes, let's do that.
21:43.36lkclpsokolovsky: would rtc being broken explain why any use of msleep in any resume function, on htcsable, results in resume hanging?
21:43.56psokolovskypH5, caveat: someone will need to test that RTC will work after all that ;-). I don't have time for that now, so that's why I didn't post RFC for that on k-d
21:44.27psokolovskylkcl, I doubt that. rtc is rtc, optional device.
21:44.31lkclok.
21:44.44lkcli feel slightly clueless...
21:45.27psokolovskylkcl, me too, that's why I discuss it here, and would like other people check it before making any big changes
21:46.01goxboxlivecr2: i think maybe usb0 is resuming. There might be a setup i am using on my laptop. Beacuse the universal detects when i replugg the USB cable, and if i do a ifup usb0 on my laptop ssh in ti universal works again.
21:46.02pH5lkcl: I'm more than slightly clueless here. what clock does msleep use? perhaps something went wrong in the early clock configuration?
21:46.26lkclmsleep uses schedule_timeout
21:46.35*** join/#htc-linux ellisway (n=ellis@host-87-74-241-174.bulldogdsl.com)
21:47.25pH5lkcl: could you see where exactly it hangs?
21:47.34lkclwhich uses setup_timer() __mod_timer() schedule() and...
21:48.00goxboxlivecr2: well i had to replugg the usb cable to bring up a new ssh session.
21:48.27lkclpH5: well, i suppose i could add in a static variable indicating whether suspend has occurred, then use the 'vibrate' trick again - but after spending several days trying to find it i got pretty tired of digging into the issue
21:48.42lkclso i've given it a rest for a bit
21:50.02pH5I see. At least this way you've got time to clean up asoc in cvs :)
21:50.15psokolovskyGentlemen, I aslo woul like to remind that we have kernel status page, http://www.handhelds.org/moin/moin.cgi/KernelStatus
21:50.30psokolovskyIf found and confirmed issue, please add it there
21:50.46psokolovskyif later it is resolved, please remove/mark it as such ;-)
21:51.46cr2psokolovsky: good idea.
21:51.53pH5as always
21:51.58cr2:)
21:53.17cr2hmm. whi raises this ext_irq ...
21:53.23cr2s/whi/why/
21:53.50cr2pH5: do you have ext_irq ?
22:00.01pH5ext_irq as in irq from the cpld chip?
22:01.02pH5the cpld irqs trigger on SD/earphone/AC plugin/removal
22:03.16cr2hmm. so your ext_irq is the cpld mux ? i think it is SYSINTR ?
22:04.09cr2ok. i'll try 32bit spi.
22:04.40pH5I don't know the terms ext_irq and SYSINTR.
22:06.17cr2strings nk.exe
22:09.52pH5"CPLD: SYSINTR_EARPHONE" is the only occurence of any of this, so it seems to be SYSINTR, whatever that means.
22:10.10cr2ok.
22:10.23cr2preparing to blow the ts :)
22:10.54pH5cr2: good luck
22:12.00cr2hmm. does not work.
22:13.15cr2number_of_entries_in_fifo = ((ssrval >> 12) & 0xf) + 1;
22:13.25cr2this one is non-obvious for me.
22:19.32cr2hmm. i get samples from /dev/input/event0
22:21.15cr21300 -25000, -16000 10000
22:21.26cr2bounding box. why is that ?
22:23.07cr2ah. i need to scale them down.
22:23.40pH5number_of_entries_in_fifo = ((ssrval >> 12) & 0xf) + 1 is the RFL value, straight out of the manual
22:24.39pH5g'night
22:24.47cr2ok. thanks. ts_calibrate died :)
22:24.51cr2good night.
22:49.33disturbedsaintcr2: Mind if I ask a question?
22:50.20cr2yes.
22:50.28cr2:)
22:50.29disturbedsaintohw
22:50.31disturbedsaint:/
22:50.41disturbedsaint:P
22:50.41cr2ask.
22:51.02disturbedsaintHow many people are actively developing Xanadux/linux-ports for the HTC-devices?
22:51.29cr2you can see all them here.
22:51.37disturbedsaintok
22:51.45disturbedsaintthat's way more than I thought
22:52.04ccureauthat was different
22:52.37*** join/#htc-linux disturbedsaint (n=simon@212-127-222-63.cable.quicknet.nl)
22:53.00disturbedsaintaccidentally shutdown  gaim :/
22:53.16cr2well, 1-2 persons per device.
22:53.24disturbedsaintok
22:53.52disturbedsaintadds up to a lot of people :)
22:54.14cr2yes, many chips are shared between the devices.
22:54.56disturbedsaintindead, should make the process easier
22:55.36disturbedsaintbtw concerning that, why is it so that a lot of drivers are (hard)coded to only work on a certain device?
22:55.36disturbedsaintie the driver doesn't work on another device using the same hardware?
22:56.22cr2everybody tries to make it work on his own device first.
22:56.22ccureaudepends
22:56.53cr2and htc changes the gpio/irq/port randomly between the devices. don't know why.
22:57.00ccureaufor instance, same driver would probably work for touchscreen on tornado and wizard, but gpio on wizard is different
22:57.14ccureaukeyboard driver is the same on both
22:57.17disturbedsaintok
22:57.25disturbedsaintI see
22:57.32Ralithcr2, to make life hard on reverse-engineering efforts? :P
22:57.59disturbedsaintat least not to make it easier :P
22:58.14cr2no. to make reusing wince drivers hard .
22:58.18Ralithcr2, what's the status of Universal DoC drivers?
22:59.17cr2somebody should cleanup the c code by lkcl, and check if it can read the bootloader through /dev/mem
22:59.26cr2this one is relatively easy.
22:59.57Ralithso it's pretty much all but done?
23:00.02cr2disturbedsaint: this one is the best example http://wiki.xda-developers.com/index.php?pagename=HTC_Serial
23:00.46cr2Ralith: if you'll compare to DoC G2 , the driver for G3 is non-existent.
23:01.20Ralithcr2, er, which is the Universal on?
23:01.26cr2G3
23:01.46Ralithabout how much effort is it likely to take to get it working?
23:02.04Ralithalso, by chance, is there any help a relatively inexperienced C coder such as I might be able to offer, be it there or elsewhere
23:02.20cr2the "free" docs are incomplete, so it's hard.
23:02.58cr2everything with - here http://wiki.xda-developers.com/index.php?pagename=HTC_DiskOnChip is not documented.
23:03.24cr2but used in the code.
23:05.53cr2the hardware part is complex reverse-engineering.
23:06.44disturbedsaintcr2: I see, the implementation/connect of the part is just different per device
23:06.45cr2the software part (SAFTL) requires some serious knowledge of certain fields in the mathematics
23:07.00cr2but i think we will not implement SAFTL.
23:07.36Ralithwhat's lost by not implementing that?
23:07.50cr2wince compatibility
23:08.07cr2your can't mount wince  partition then.
23:08.14cr2if you'll have one :)
23:08.16Ralithbut you can overwrite it?
23:08.17ccureauugh
23:08.23cr2yes.
23:08.47Ralithwhat /is/ SAFTL, then? Some kind of security?
23:08.50cr2it's you.
23:09.08Ralith?
23:09.10cr2no. SequentialAccess translation layer.
23:09.30RalithI'm just not understanding how you can read/write but not be able to read/write the existing wince data
23:09.37cr2flash translation layer. on SD/CF it is done in hardware
23:09.55cr2linux does it with jffs2
23:10.17Ralithah, so it's just a different method.
23:10.29Ralithkinda like a different filesystem but lower level, correct?
23:10.37cr2the wince data is stored on a tricky "filesystem"
23:10.40cr2yes.
23:10.46Ralithgot it, thanks :0
23:10.48Ralith:)*
23:11.09ccureauif the filesystem work is done in hardware, though, doesn't it seem that there would be a higher level interface to the filesystem?
23:11.12ccureauhooks, at least?
23:11.24cr2the G2 uses INFTL.
23:11.32*** join/#htc-linux lkcl (n=lkcl@535411CB.cable.casema.nl)
23:11.42cr2SD/CF ? no, you still work on a block layer.
23:12.15cr2m-sys hardcoded DOS fat on top of it ;)
23:12.26Ralith:P
23:12.33Ralithyay for outdated filesystems
23:12.53ccureauthe DoC uses that method as well?
23:12.58cr2wince/dos boot compatibility.
23:13.52cr2there are a lot of dumb outdated concepts implemented in DoC software.
23:14.08cr24 partitions, for example. and fat.
23:14.13Ralith-_-
23:14.29ccureaubackwards compatibility bites again
23:14.33Ralithoh, by the way, does the Universal have any sort of media acceleration?
23:14.41cr2and it's closed source on top. also on linux.
23:14.53cr2no, it's plain pxafb.
23:15.11cr2for video.
23:15.59Ralithalright.
23:18.21cr2good night.
23:18.26Ralithnight
23:18.28Raliththanks for the info
23:24.35disturbedsaintthx indeed
23:25.07disturbedsaintbtw just like Ralith I'd like to help as much as I can though I'm not really any good @ coding :)
23:27.47disturbedsaintsomehow thats seems quiet logical :P
23:28.38ccureau:D
23:31.34Ralith:P
23:31.45RalithI have but one thing to say: yay for ebay.
23:32.17ccureauoff until tomorrow...
23:32.36ccureauand I must say: very informative channel...even for Wizard owners. :D
23:33.59WizMaui_yehaw Wizard ;)
23:34.12disturbedsaintwizzard = too big ;)
23:34.22WizMaui_nah
23:34.27WizMaui_gn8 everyone ;)
23:34.42disturbedsaintbye
23:39.50*** join/#htc-linux florian (n=fuchs@84-245-183-76.hbg.bpool.celox.de)
23:45.33disturbedsaintim off too
23:45.35disturbedsaintbye!
23:45.37*** part/#htc-linux disturbedsaint (n=simon@212-127-222-63.cable.quicknet.nl)

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.