irclog2html for #htc-linux on 20061213

00:00.47goxboxliveok
00:02.25cr2so i must hunt down 2 irqs: BT and this external.
00:03.20cr2i guess it's the BT/BB that sets them, but how does it know...
00:04.15cr2the AT command send routine (ffuart_tx) needs a minor change. check_BB_ready() before sending an AT command.
00:04.15goxboxliveOk, now i know about the CMUSS . I got rid of it after chaning the AT+CNMI like it was in my debug log and not as coded in the source of gomunicator.
00:04.38goxboxliveok
00:04.42cr2the code is already there, but needs a separate function.
00:05.22lkclok i give up on blueangel / uda1380 sound.
00:05.26cr2you didn't apparently saw a problem so far, but it's bettery to follow wince in this respect.
00:05.45goxboxliveyes i know. It is a little separated now, but because of the difrrents on each device i think it would be bether to have one AT config file for each phone.
00:05.57lkclgonna try sable again, work on updating the suspend/resume code and platform device stuff, like psoko said i should (from a message on kd@hh)
00:06.05cr2lkcl: it's tough stuff. the SSP uses 16bit MSB with external clock, looks like I2S to me.
00:06.20lkclachh.
00:06.23cr2lkcl: it should not be difficult.
00:06.27lkclptooey.
00:06.32goxboxlive:-)
00:06.48cr2i've already removed pt_regs where is saw them in the sable code.
00:06.55lkcli just want to achieve something and then i will feel better.
00:07.00lkclok cool.
00:07.19cr2lkcl: btw, try to reuse the htcuniversal_ts2.c code for a more abstract tsc2046 driver.
00:07.26lkcloh good idea.
00:08.00cr2i should split it into the generic and universal-specific part.
00:08.01lkcldid you rewrite that already?
00:08.07lkcloh ok not yet then
00:08.16cr2mostly, have a look.
00:08.19lkclcool
00:08.33lkclcvs update in progress...
00:08.40cr2well, otherwise you need to convert your current driver to platform_device
00:09.27cr2i've tried the 16bit and 32bit spi, it works to some extent, but there are bugs there.
00:09.54lkclsorry - what's this for?
00:10.04cr2so i gave up for now, and the driver is pxa-specific. it will be necessary to hack it a bit for hermes.
00:10.24lkcltried 16bit and 32bit spi for what - tsc2046, yes?
00:10.33cr2i wanted to attach it to pxa2xx-spi instead of messing with pxa spi directly.
00:10.36cr2yes.
00:10.38lkclo right
00:10.55cr2but pxa2xx-spi does not support 24bit mode.
00:11.02lkclahh joy!
00:11.42cr2i'm sure it can be done, but probably not now.
00:11.48lkclwell it's probably easier to hack pxa2xx-spi to support 24-bit
00:12.06cr2there are some sample wince drivers which use 8bit spi for tsc2046!
00:12.11cr2maybe.
00:12.31lkclohh.  oh ok.
00:13.41cr2but they are going into big detail, because the ts samples are 12bit, and you can't transfer them in one bus transaction.
00:30.13Kevin2Hi
00:30.23lkclarg i can't find the charger for the sable, in the trip back from holland oops
00:30.26lkclhi kevin2
00:31.42Kevin2cr2: Did you see that there is a RTC8564 driver in the new rtc class part of the .19 kernel?
00:33.14lkcloo - that's the same chip as used in the sable.
00:36.00psokolovsky_lkcl, you should have been disable offending CONFIG_ options instead of patching mainline driver in an adhoc way ;-)
00:36.13lkclo oops...
00:36.23lkclwhat CONFIG_ options?
00:36.42psokolovsky_lkcl, which breaks pxafb.c compile
00:36.54lkcl*thinks*...
00:37.01lkclbut for me, it's broken.
00:37.14lkclpxafb.c doesn't compile:
00:37.35psokolovsky_lkcl, if you fix it, you should submit patch upstream ;-)
00:37.35lkcldrivers/video/pxafb.c: In function 'pxafb_parse_options':
00:37.35lkcldrivers/video/pxafb.c:1227: error: 'struct pxafb_mach_info' has no member named 'bpp'
00:37.35lkcldrivers/video/pxafb.c:1234: error: 'struct pxafb_mach_info' has no member named 'pixclock'
00:37.35lkcldrivers/video/pxafb.c:1235: error: 'struct pxafb_mach_info' has no member named 'pixclock'
00:37.36lkcldrivers/video/pxafb.c:1237: error: 'struct pxafb_mach_info' has no member named 'left_margin'
00:37.36lkcldrivers/video/pxafb.c:1238: error: 'struct pxafb_mach_info' has no member named 'left_margin'
00:37.38lkcldrivers/video/pxafb.c:1240: error: 'struct pxafb_mach_info' has no member named 'right_margin'
00:37.40lkcldrivers/video/pxafb.c:1241: error: 'struct pxafb_mach_info' has no member named 'right_margin'
00:37.42lkcldrivers/video/pxafb.c:1243: error: 'struct pxafb_mach_info' has no member named 'upper_margin'
00:37.44lkcldrivers/video/pxafb.c:1244: error: 'struct pxafb_mach_info' has no member named 'upper_margin'
00:37.46lkcldrivers/video/pxafb.c:1246: error: 'struct pxafb_mach_info' has no member named 'lower_margin'
00:37.48lkcldrivers/video/pxafb.c:1247: error: 'struct pxafb_mach_info' has no member named 'lower_margin'
00:37.53lkcldrivers/video/pxafb.c:1249: error: 'struct pxafb_mach_info' has no member named 'hsync_len'
00:37.53lkcldrivers/video/pxafb.c:1250: error: 'struct pxafb_mach_info' has no member named 'hsync_len'
00:37.54lkcldrivers/video/pxafb.c:1252: error: 'struct pxafb_mach_info' has no member named 'vsync_len'
00:37.58lkcldrivers/video/pxafb.c:1253: error: 'struct pxafb_mach_info' has no member named 'vsync_len'
00:38.00lkcldrivers/video/pxafb.c:1257: error: 'struct pxafb_mach_info' has no member named 'sync'
00:38.02lkcldrivers/video/pxafb.c:1260: error: 'struct pxafb_mach_info' has no member named 'sync'
00:38.04lkcldrivers/video/pxafb.c:1265: error: 'struct pxafb_mach_info' has no member named 'sync'
00:38.06lkcldrivers/video/pxafb.c:1268: error: 'struct pxafb_mach_info' has no member named 'sync'
00:38.08lkcloops.  sorry.
00:38.10lkclautomatic reaction.
00:38.23psokolovsky_and the author of those changes was let know, and prepared fix and will submit it himself
00:38.30lkclack.  understood.
00:38.42psokolovsky_and I didn't apply his aptch, becasuse for 2 weeks .19 is live, noone reported it ;-)
00:38.47psokolovsky_well, nevermind ;-)
00:38.56lkclsorry.  where's upstream?  what's the procedure?
00:38.57psokolovsky_just fyi ;-)
00:39.42psokolovsky_lkcl, generally you should lookup in main git tree who did broken change, and let them no, cc: main Linux list (and k-d of course)
00:39.55lkclok cool.
00:40.01psokolovsky_it's RP who did that this time ;-)
00:40.10cr2Kevin2: yes, i have already enabled it. but it does not shouw up in /sys
00:40.53cr2Kevin2: and there are 2 RTC drivers now. sa1100-rtc and pcf8563.
00:46.02*** join/#htc-linux florian (n=fuchs@84-245-180-141.bpool.celox.de)
00:46.53*** join/#htc-linux parmaster (i=par@dipole.idlepattern.com)
00:54.12cr2Kevin2: please add the examples for the new haret trace/irq commands to hh.org wiki.
00:54.36cr2i always forget their syntax.
00:57.55goxboxlivecr2: I made a screenshoot of gpe. It is starting to have a nice interfcae now: http://handhelds.org/scap/port.23886.png
00:58.41cr2ok, fonts finally look good.
00:58.58goxboxlive:-)
00:59.02cr2why is the gomunicator icon so interpolated.
00:59.25Kevin2cr2: The kernel rtc8563 code doesn't seem to have the epson i2c address in it.
00:59.39cr2Kevin2: ?
00:59.48cr20x51 is hardcoded.
01:01.08cr2hmm. you are right.
01:01.21cr2is it a module parameter ?
01:01.26goxboxlivecr2: Becuse i just found a pics that actually is a applett icon
01:01.55Kevin2I think it may be possible to have sysfs override the parameter.
01:06.29cr2goxboxlive: your screenshot looks like a kde/gnome mixup. both are bloatware.
01:07.24goxboxlive:-) i know. it is a theme i installed. The original GPE theme looks no good i think.
01:07.45cr2i'm biased and dislike gnome. i want a more traditional ui.
01:08.29cr2opie gui i ok for me, but it's just a personal taste. maybe because i have zaurus and a780.
01:08.33goxboxliveme to, i have allways used kde
01:10.03cr2i think trolltech sees the light, and will release the phone edition RSN.
01:10.46cr2i'm also curious how openmoko will look like.
01:12.23Ralithyeah
01:12.26Ralithbe cool if we could port /that/
01:12.47cr2we need to know their kernel api.
01:12.58goxboxliveYes i like opie bether than gpe, but i am using gpe now, because it is more usefull, since keyboard is working as i should, and of course gomunicator.
01:13.02cr2you don't need to port the gui itself
01:13.17Raliththat's what I meant :P
01:13.20Ralithport their whole thing
01:14.02cr2goxboxlive: once i'll finish with the hardcore stuff (sound) i'll do the gui work.
01:14.10goxboxlive:-)
01:14.41cr2it's a shame that opie is not so user-friendly as it should be.
01:14.47cr2on universal.
01:15.13cr2minor annoying glitches that spoil the whole user impression.
01:15.23goxboxliveBut what do you think of Qtopia. When TT release the code, will they also release some other applications we can use, or do people have to start making new ones form scratch?
01:16.36cr2if they will release the phone code, we need to hack on the gsm/umts library.
01:16.39ljp_gpl release will be everything except DRM, SXE, and telephony libs
01:17.12cr2lkcl is keeping low with his greenphone stuff, but he could clear some issues with the cam, for example.
01:17.54cr2but i think sound is more important right now.
01:18.21goxboxliveljp: So when it is released we will be able to use it as opie, is that what you are saying?
01:18.54goxboxliveYes, we have to get sound working, then resume fully working, then we can start o look at cam and that other stuff.
01:19.21cr2if we will get the mic working for the phone it will be a real breakthrough.
01:19.47goxboxliveyes, that will be a progress.
01:19.57cr2it's great that i have most completely debugged the BT, but it can wait too.
01:20.00ljpyes, i have had it working on zaurus for a few weeks now. its will become opieII
01:20.17ljpbut the project might be hosted at linuxtogo.org
01:20.34cr2ljp: good. we are really waiting for it.
01:20.41ljp:) me too
01:20.51goxboxliveOk, sound good. Will it be available in OE?
01:21.52cr2goxboxlive: as long as it compiles cleanly i'll be happy :)
01:22.39goxboxliveWell i have tried to compile Opie without OE, but it is hard, it failes over and over again. I gave up since OE is working so well.
01:23.00cr2yes, that's why i say it :)
01:23.33goxboxliveI guess Trolltech is comming to release it before Openmoko.
01:23.56ljpthis has been planned for months :)
01:24.23goxboxliveljp: Are you allowed to show us some screenshoots of Opie2?
01:24.28ljpi havent been working with oe, since I can never get a full and complete build
01:24.58ljpoe has a stability issue
01:24.58cr2having a good and well documented c++ gui infrastructure is a real advantage.
01:25.48goxboxliveljp: Yes oe.dev might be unstable sometimes, but not the stable branches.
01:26.26ljpi cant get other branches built either
01:26.41goxboxliveok
01:26.50ljpi keep trying, but i dont have time to muck with oe
01:26.54ljp:)
01:26.59*** join/#htc-linux psokolovsky_ (n=psokolov@237.usernat.ip.net.ua)
01:28.03cr2oe is certainly not for somebody who want just to compile a stable version.
01:28.03goxboxlivepsokolovsky_ Opie2 is on it's way, are you going to maintain it in OE if it will be available there?
01:28.07cr2with a horrible python fail backtraces it scares me :)
01:29.31goxboxlivecr2: I dont think we should make a lot of work with Opie now, since there will be a allmost complete version soon.
01:30.11cr2i'm concentrating mainly on the kernel api anyway.
01:30.16goxboxliveWe only have to get a phone application. SInce it not will be available in Opie2.
01:30.22cr2and nobody will write odevice_htc.cpp for us anyway.
01:30.30cr2http://www.pbase.com/isabel95/image/27922860
01:30.35cr2opie2.jpg :D
01:30.59goxboxlive:-)
01:31.42ljpit will look like qtopia 4.2 until we redo the icons, or at least color them differently
01:31.54cr2Opie is 4 weeks old.
01:31.56ljpor come up with our own theme
01:32.19cr2i'll probably rip out the icons from a780.
01:32.31cr2so the gui is consistent between my phones.
01:33.25goxboxlivewell it's getting really late now. g'night
01:33.33cr2good night.
01:36.24Kevin2cr2: good night.
02:08.12*** join/#htc-linux LunohoD_ (n=alex@e180122238.adsl.alicedsl.de)
05:06.15*** join/#htc-linux cr2 (n=konversa@crpl22.physik.uni-wuppertal.de)
07:27.35*** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com)
07:40.24*** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk)
07:47.10*** join/#htc-linux cr2 (n=konversa@crpl22.physik.uni-wuppertal.de)
08:02.17*** join/#htc-linux rob_w (n=rob_w@p549BA6CB.dip0.t-ipconnect.de)
08:07.33*** join/#htc-linux Fossi (n=cmewes@195.177.48.115)
08:13.01lamikrgoxboxlive: AT+CMGS=18 means that the phone has received SMS message to storage index number 18
08:15.11lamikrgoxboxlive: But the current gomunicator code expects that there would come the content of the SMS message in a separate line afterwards.
08:17.15goxboxlivehmm ok
08:17.53lamikrgoxboxlive: One thing you could try is to change the value of property "phoneConfig.wait_ack_before_sending_sms_pdu" to opposite for your phone. That's control's the workflow how the SMS is send. We found out long time ago with CP15, that blueangel and h6300 needed a little bit different steps in there. And therefore we added there this config option.
08:17.54goxboxliveI thought that  AT+CMGS was the send command, and that  AT+CMGR was the receive command
08:18.16goxboxliveI have allready tried that.
08:19.10goxboxliveBut, i am not sure about the sms format. I mean here in norway we use "text" and not "PDU". What is the default in source?
08:19.23goxboxliveIsent it default as pdu?
08:21.39lamikrgoxboxlive: Hmm, yes, you are right from the AT+CMGS command. Sorry, I mixed that in my mind.
08:23.19lamikrWell, the gomunicator encodes the messages to PDU format before sending.
08:23.48lamikrCan you connect to your phone module with minicom?
08:24.23goxboxlivelamikr: No, we havent got so far. FOr now we are using 'cu'
08:24.32goxboxliveAnd i have been able to send sms via cu
08:25.31goxboxliveBut i am veru close to get it working with gomunicator. Because with send it dos everything right beside it hangs on the AT+SMGS command.
08:25.37lamikrOk.
08:26.01goxboxliveAnd with receiving it detects that here is something incomming, but it detect it as a phonecall and not sms.
08:28.31goxboxliveI think also there is a choice in the gomunicator source to set the sms to textmode with changing the AT+CMGF= to 1, and not 0 as default.
08:29.00lamikrFor me, the sms sending failed also in the beginning. We found out that h6300 neeed to receive  response from the GSM module after we sent the "AT+CMGS=<id>" command before we really could send the PDU.
08:29.34lamikrBut with blueangel, the pdu needed to be within the same command with AT+CMGS
08:30.07lamikrAnother difference was the size of the message sent: blueangel wanted: curSize= strlen(pduParam)/2 - 1;
08:30.18lamikrh6300 wanted: curSize= strlen(pduParam) / 2;
08:30.31goxboxliveOk, but is this up2date in the source. Because it seems like the h6300 has "wai for ackn. pdu" set to False.
08:30.35lamikror the sending would hang
08:31.27goxboxliveok, i can recall those line from the source, but wich one is default?
08:32.37goxboxliveWhen i figure something out i'll send you my cvs diff.
08:32.46lamikrhmm, sorry it must then be in a opposite way so that the ba required the message to be splitted.
08:33.24lamikrtry first change the curSize calculation to different one.
08:34.02goxboxliveok i will.
08:38.47lamikrWhen you receive SMS, do you receive CNMI command?
08:39.26lamikrI meant: CMTI
08:40.29lamikrcommand: "+CMTI: "SM", 5"
08:42.19lamikrshould mean that the SMS has been received in index 5. Gomunicator should catch that message and send another command to request to read it from this index without altering it's state to unread (So that it can splash the message briefly) by using command: AT+WUSS=1;+CMGR=%d
08:44.31lamikrI believe that the problem with the received SMS messages is that the code will not go to following if in the end for you...
08:44.33lamikrelse if ((((atCmdLineParam[0] >= '0') && (atCmdLineParam[0] <= '9')) ||
08:44.34lamikr<PROTECTED>
08:44.35lamikr<PROTECTED>
08:44.37lamikr<PROTECTED>
08:45.32lamikrIt has been done a little bit nasty way, but I did not wanted to change it too much because I did not have blueangel for testing.
08:46.00cr2hi guys
08:46.14lamikrcr2: Hi
08:46.29cr2we have extensive logs, is it possible to decode PDU in awk/perl ?
08:47.19cr2lamikr: when i send sms to myself, will it always leave the phone ?
08:47.29lamikrI remember seeing some perl modules for SMS encoding/decoding but I have not tested them myself.
08:47.41cr2ok.
08:47.57lamikrcr2: I think it will always go via sms center
08:48.03cr2good.
08:48.56cr2the current code certainly does not support utf-16 ?
08:49.16goxboxliveNo i dont see this: command: "+CMTI: "SM", 5" .It spams some nasty messages by not finding audio such stuff, and the it breaks and shows it as a phonecall in the gui.
08:49.19goxboxlivehi cr2.
08:49.36cr2hi goxboxlive
08:50.04cr2i remember a wince sms bug that forced utf-16 if you have used ' symbol :)
08:50.24cr2on universal.
08:50.37lamikrcr2: That may be really true. I also believe it may have problems with 7bit message format.
08:50.48cr2doubling the sms length ;)
08:51.23cr2why is 7bit a problem.
08:51.42lamikrThere is one windows utility for playing with PDU, pduspy if I remember correctly which support really many "aspects" of SMS messages.
08:52.33cr2are there any docs for the sim toolkit encoded strings ? is it also pdu ?
08:58.25cr2lamikr: is it really needed to develop the sms stuff over and over again ? do you know the gnokii current status ?
09:06.02lamikrcr2: I think I checked once the SMS code in gnokii and it was not very good. Propably it supported only a "text" messages, not pdu at all.
09:10.26cr2ok, i'm looking at it now, and there is even some hexencoded ucs2 support (utf16).
09:11.14cr2<PROTECTED>
09:11.18cr2<PROTECTED>
09:11.18cr2<PROTECTED>
09:11.18cr2<PROTECTED>
09:11.18cr2<PROTECTED>
09:11.19cr2<PROTECTED>
09:11.19cr2<PROTECTED>
09:12.24cr2http://cvs.savannah.nongnu.org/viewcvs/gnokii/common/phones/atgen.c?rev=1.135&root=gnokii&view=markup
09:13.09lamikrhmm, yes. I also downloaded the sources now. It may be that I have overlooked this one. But I am sure, I found out couple off SMS apps which were able to send and receive sms but used "text format" only.
09:13.59cr2ok.
09:14.17lamikrIt seems gnokii claims to support also logos and ringingtones in sms messages.
09:14.55lamikrI am looking at the common/gsm-encoding.c
09:15.00cr2is is something special, or only a convenience feature ?
09:15.40lamikrlogos?
09:16.55cr2yes.
09:17.25lamikrnot sure yet, check for example a function gn_ringtone_pack() from the gsm-encoding.c which is used for creating a "sms" containing ringing tone.
09:17.58cr2i think it's important for the nokia/whatever handsets, but not for a PDA phone.
09:18.45lamikror gn_ringtone_unpack
09:18.51cr2ok.
09:18.55cr2<PROTECTED>
09:18.55cr2<PROTECTED>
09:18.55cr2<PROTECTED>
09:18.55cr2<PROTECTED>
09:19.14cr2does not look like very failsafe code :)
09:20.46lamikryes... For me, the current sms sending and receiving code in gomunicator works, but for some better app or for the gomunicator itself in the long term, it would probaly make sense to librarize the sms part.
09:21.13*** join/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com)
09:21.28*** join/#htc-linux lkthomas (n=lkthomas@202.85.228.3)
09:21.31lkthomashey all
09:21.36lamikrhi
09:21.44lkthomasI have a question
09:22.01lkthomasdo I have to keep my XDA II blutooth on incase I want to listen phone call via BT headset ?
09:22.08lamikrcr2: I do not know, is the sms part of the code librarizez in this gnokii either or would it require us to make lot of ripping.
09:22.18lkthomasis that possible for instant pairup when a call is coming up ?
09:24.10babelouedhi
09:31.28babelouedlamikr : what do you think about gsmlib ?
10:13.29*** join/#htc-linux TripleDES (n=sergio@89.129.175.52)
10:37.21*** join/#htc-linux rob_w (n=rob_w@p549BA6CB.dip0.t-ipconnect.de)
12:12.01*** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk)
12:22.38cr2lkthomas: it is a difficult question. audio routing is the most murky part of the phone.
12:23.06cr2xda2 (himalaya) uses ASIC5 for bluetooth ?
12:23.56cr2lamikr: i think it is a library. at least looking at the code, it is easy to make a library if it's not the case.
12:36.14babelouedcr2 : i have a question about wifi
13:10.43*** part/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com)
13:51.08*** join/#htc-linux psokolovsky_ (n=psokolov@237.usernat.ip.net.ua)
13:53.56cr2babeloued: ?
14:20.16lkclheya folks: anyone want a universal for $EUR 99 which had a failed (official) ROM upgrade?
14:21.29lkclprobably have to do a JTAG job on it - but hey, that's no hardship :)
14:39.11*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
14:43.06cr2lkcl: failed g3 bootloader ?
14:43.15lkclprobably.  i just asked.
14:44.21cr2imho, it's worth the money.
15:10.09*** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com)
16:09.17*** join/#htc-linux RoEn_PC (n=roen@p54A65D8D.dip.t-dialin.net)
16:31.23*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
16:32.46*** join/#htc-linux JTRipper86 (n=jtripper@p54B1D6C0.dip.t-dialin.net)
16:34.05BabelOre
16:36.09cr2BabelO: what did you want to know about wifi ?
16:36.22BabelOhi cr2
16:36.28cr2btw, why don't you pull all 3 power gpios ?
16:36.38cr2the irq _is_ gpio17
16:36.54BabelOi see in some log thaht a lot of people have same problem, error issue_cmd
16:36.56cr275 is wifi-related, but it's not an irq.
16:37.06BabelOoh, in cvs this is a partial comit
16:37.25BabelObecause other power is setup with new pcmcia module
16:37.34cr2ok.
16:37.38BabelOif you want i send you a new file :)
16:37.56cr2no time for it :)
16:38.11BabelOok
16:38.29BabelOi see in mem.c a VENDOR_ID define
16:38.41cr2is there are "real" pcmcia acx100 driver working ?
16:38.50cr2hx4700 is not "real" .
16:38.56BabelOit is used for an offset
16:39.00BabelOah ok
16:39.23cr2VENDOR_ID is ok.
16:39.28cr20x900 ?
16:39.58cr2do you have a "bt test" command in the bootloader ?
16:40.12BabelO#define ACX_VENDOR_ID 0x8402
16:40.39BabelOsorry this on is hx4700 0x8400104c
16:41.06BabelOhow can i see if there is one ?
16:44.06cr2you need to write a command in the command register and read the reply.
16:44.25cr2should be doable with haret.
16:44.31BabelOok
16:46.42*** join/#htc-linux asylumed (n=insanity@196.211.116.2)
16:50.06BabelOcr2: this one work, but there is some test in c file ... http://linuxbox2.free.fr/blueangel/2.6.19-hh/htcblueangel_acx.c
16:50.17BabelOdo you have command to send in haret ?
16:51.17kopshello,
16:51.17kopsis someone working on htc tornado smartphones ?
16:52.18cr2BabelO: i'll look at the code. the command? it's a 16-bit register write, i forgot the numbers.
16:52.30BabelOok
16:52.47*** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk)
16:52.50cr2kops: it's omap, the main progress is on pxa. with s3c24xx coming.
16:58.35cr2goxboxlive: are you here ?
16:59.10goxboxlivecr2: yes im here.
16:59.57*** join/#htc-linux asylumed (n=insanity@vc-196-207-41-253.3g.vodacom.co.za)
17:01.28cr2goxboxlive: i've copied i2c/ak4641.c to htcuniversal_ak4641.c and modified it.
17:01.37cr2would you like to test ?
17:02.00goxboxliveof course i want to test.
17:03.36cr2i'll commit the changes to CVS. we will need to dramatically extend this driver.
17:03.54cr2for BT and other audio switching.
17:04.19goxboxliveok, but are there any chanches that the speakers will work now?
17:05.04*** join/#htc-linux drno (n=drno@eki.student.utwente.nl)
17:06.13cr2done.
17:06.16cr2i think yes.
17:06.58goxboxlivecr2:Should i only use the htcuniversal_defconfig? Or is there something i need to change (like diable the old sounddriver)
17:07.58goxboxliveforget it, i saw now that you ghave updated it
17:09.59cr2i test the changes i make :)
17:10.07drnoHi all, Im trying to run linux on a HP6915 and found this IRC channel on xda-developers.com so I hope I can contribute/get some usefull information here on the kernel basics
17:10.09cr2the kernel should compile.
17:10.28cr2hi drno
17:10.55cr2the sable kernel should work with minor changes, but will not resume (afaik).
17:11.04goxboxlivecr2: COuld you change this to module instead of using it in the aImage
17:11.20cr2goxboxlive: it's module.
17:11.24drnoIve read about that, but I just started so Ill look into it
17:11.37goxboxliveNo, it is in the zImage
17:11.38cr2drno: try to compile it now.
17:12.27cr2goxboxlive: i have used "select" like the mainline kernel. you may have the modules sync probelms otherwise.
17:12.58cr2drno: i have the device, will help with debugging, compiling, but mainly interested in the gps traces.
17:13.32cr2the hardware is very close to the universal, so it should be relatively easy to support.
17:13.40drnoah, sure Ill try to be as helpfull as I can be, although Im only used to the normal linux maintanance stuff
17:14.02drnoI have read the Sable progress page
17:14.17cr2ok. try to compile the kernel.
17:14.38cr2the progress is good :)
17:14.42drnowill do
17:17.27goxboxlivehmm i dont have any problem using it as a module. OE.DEV is setting it upas a module, that's why i want it.
17:17.31goxboxlivecr2: Still here?
17:18.20cr2yes.
17:18.42cr2BabelO: you turn a lot of gpios, but are they really belong to the wifi ??
17:18.50goxboxliveYYYYYYEEEEIIIHAAAAAAAAA we have the main speakers working. Also when i plug in the headphones, it is swithcing ovet to headphones, and stereospeakers is getteing shut off. But....
17:19.33goxboxliveWhen i again unplug the headphones the sound doesnt swithc over to main speakers (stereo).
17:20.14goxboxliveANd when i then again tries to replug the headphones are dead. But...
17:20.46BabelOcr2: gpio related to pcmcia0 slot are needed ( line before pxa_gpio_mode ) other GPSR and GPCR are tests from me
17:20.58goxboxlivewhen i stop the music, and start it again the speakers works again, also headphones, and the rest is as i explained.
17:22.47cr2goxboxlive: the switching machinery is so primitive now. it was tuned for hx4700 only, as you can see.
17:23.11goxboxliveYEs, great work cr2 as allways :-)
17:23.50cr2we need multiple mixer channels, BT, and so on. eventually we'll replace the main i2c/ak4641.c
17:24.00goxboxliveThe mic is also working, but with a lot of distortion.
17:24.13cr2i'm glad that you are happy :)
17:24.20goxboxlive:-)
17:24.44cr2the equalizer and some other params are missing too ;)
17:25.07cr2try to play with the mic attenuator settings.
17:26.04kopsthanks cr2, i think i must find more information about this to ask more questions ! In fact i thought i should  work for this project and in the same time for my studies.
17:26.04kopsI found this  http://vivien.chappelier.free.fr/typhoon/index.html and I will try it to make me a better idea of the work which is made (sorry for my english !)
17:26.33NK`cr2: do you know if the bootloader for the spv c600 is also working for the spv c100 ?
17:26.34goxboxliveThe mic on the headset is also working, thow it is alot of distortion. If i did a phonecall, noone would recognize what i am saying.
17:27.06*** join/#htc-linux asylumed (n=insanity@196.211.116.2)
17:27.23*** join/#htc-linux rob_w (n=bob@p85.212.180.235.tisdip.tiscali.de)
17:34.01cr2goxboxlive: it's certainly an attenuator problem
17:34.43goxboxliveok
17:34.43cr2and for the phone, we are missing some really minor thing.
17:35.43goxboxliveok, i am about to try it out now:
17:36.06cr2we don't respond to this "external irq" that is always happening when the mic is in use.
17:36.27cr2maybe it's some conflict between the ak4641 and the phone over the mic.
17:37.32cr2if the phone pulls this irq line, and i guess it is, how does it know that it's time to do it ?
17:37.51cr2this is the major question now.
17:38.14cr2NK`: no idea.
17:38.30NK`okay :)
17:39.25cr2omap850 is a big blackbox. maybe because as lkcl says it does a lot of gsm stuff on chip, and you can make a lot of nasty things with it.
17:46.38cr2goxboxlive: i think that the audio switching is a task for some userspace daemon, that will also send the "AT@AUDIOSET" commands.
17:46.57goxboxliverighjt.
17:47.43cr2hmm. but how do we debug the mic issue.
17:47.51goxboxliveIt seems like it doesnt switch speakers when headset are plugged/unplugged.
17:48.51goxboxliveWhen i start a mp3 file with headphones the sound works in the headphones. And when i start without headphones the sound works out of the main speakers (as they should)
17:49.14cr2<PROTECTED>
17:49.29cr2this is the switch irq handler.
17:49.35cr2maybe too simple.
17:49.39goxboxliveBut when i unplugg or replugg the sound stops, and doesnt change over to in example headset when it is plugged. And visa versa
17:50.02goxboxliveWell i t detects the changes , but it doesnt activate the speakers
17:51.33cr2we may hack around this issue right now, but it's not a solution to the phone switcher and BT sound.
17:52.31goxboxliveok i c. Yes it would be great to have it working.
17:54.42cr2static struct snd_ak4641 ak = {
17:54.42cr2<PROTECTED>
17:54.42cr2<PROTECTED>
17:54.42cr2<PROTECTED>
17:54.43cr2<PROTECTED>
17:54.54cr2these are the only controls for now.
17:56.04goxboxliveok
17:58.03psokolovsky_Hi!
17:58.13psokolovsky_lkcl, asoc you upgraded to was 0.12.4?
18:01.09goxboxliveccr2: You are right about the phone/sound. It is still the same, and we need some at+ commands to control it
18:04.32goxboxlivecr2: <3>ALSA sound/arm/pxa2xx-i2sound.c:275: Cannot set I2S clock rate 44100
18:04.32goxboxlive<7>ALSA sound/core/oss/pcm_oss.c:940: HW_PARAMS failed: -22
18:30.35cr2goxboxlive: ACTL(mic_gain, "Mic Capture Gain")
18:30.52cr2have you tried to play with this parameter ?
18:31.13goxboxliveno, but where do i set it?
18:31.23cr2ACTL_BOOL(mic_boost, "Mic Boost (+20dB)")
18:31.30cr2alsaconf
18:31.39cr2sorry.
18:31.42cr2alsamixer
18:31.54goxboxlivei have tried with micboost, but not without. I thought i should try without now
18:32.37goxboxliveWhat is the shell command to record?
18:36.48cr2i'm not really a sound guy.
18:37.07cr2sox ?
18:37.14goxboxliveit something like 'cat  /dev/mic >> test.wav ?
18:40.25cr2try it :)
18:42.52goxboxliveHmm with opie-recorder, the sound is bether without micboost. There are still distortion (like searchin on the AM band) but, now i can hear my voice, and i can interpret what i am saying
18:45.46goxboxlivecr2: We also have a /dev/audio now and also /dev/dsp
18:46.46goxboxlive:-)
18:46.51goxboxliveHHiiihaaaaa
18:47.43goxboxliveThe records are perfect from terminal. It is Opie who makes it sound like crap :  cat /dev/dsp >> test.wav
18:47.52goxboxlivesorry i meant
18:48.18goxboxliveno that was how i recorded. I will try with headphones now.
18:49.24cr2is opie-recorder an alsa-capable program ?
18:49.36goxboxlivemaybe it is not
18:55.28*** join/#htc-linux Ralith (n=ralith@soggy202.drizzle.com)
18:58.39goxboxlivecr2: It seemslike the mic on the headpohones is not working. It uses the mic on device instead. And the record is alot bether if you switch off mic boost.
18:59.39cr2ok.
18:59.40goxboxliveIn alsamixer there is two adjust bars for the mic, but they are the same. When changing one of them, and exit and open alsmixer agaiun they have the same level as the one i adjusted last.
19:00.33goxboxliveBut, really. The record is so perfect, its so load and clear. I'm allmost getting in love with my own voice :-)
19:01.09cr2:)
19:01.28cr2the mic control is probably buggy.
19:02.33cr2now we need to trace who triggers the external irq.
19:03.34goxboxliveok
19:04.58goxboxliveDo you mean switching between headphones and main speakers?
19:13.41asylumeddoes wireless work on the h4000 now?
19:14.25goxboxlive<asylumed> ask psokolovsky_
19:14.35asylumedthx
19:14.40psokolovsky_nope
19:14.43goxboxlive:-)
19:14.48psokolovsky_no connection, at least
19:16.45asylumedpsokolovsky: we seem to be at the same point with the blueangel... firmware upload, but no action...
19:17.40psokolovsky_asylumed, you tried rmoravchik's suggestion?
19:17.54asylumedpsokolovsky: though one thing i am curious about: in the dmesg on most devices it says "ManufacturerID:05 Len:07 Data:TI Test" but on the h4000 and blueangel we get "ManufacturerID:05 Len:07 Data:TI Tes" with the T missing...
19:18.10asylumedpsokolovsky: that data is read from the card right?
19:18.21asylumedpsokolovsky: which suggestion?
19:18.50psokolovsky_asylumed, see k-d
19:19.01asylumedk
19:29.45*** join/#htc-linux Ralith (n=ralith@saunderssbs.saundersinstruments.com)
19:32.51*** join/#htc-linux Ralith (n=ralith@saunderssbs.saundersinstruments.com)
19:38.01goxboxlivecr2: The sounddriver is also resuming after suspend. It starts where it left playing.
19:40.46Ralith:D
19:48.02cr2goxboxlive: found the purpose of gpio99 :)
19:48.26goxboxlivewhat does it do?
19:50.57cr2small camera reset on powerup.
19:51.31cr2GPIO 92   10000000   charge? usb?   I1   FE  
19:51.44cr2this one is a bit mysterious.
19:51.56goxboxliveok
19:53.16cr2btw, what flaslight-related cam modes are available ?
19:53.33cr2i see 4.
19:54.03cr2on, off, on with something on, off with something off.
19:54.06*** join/#htc-linux bd2 (n=ufo@89.113.84.151)
19:54.09cr2what is something ? :)
19:54.16goxboxliveit is the flash
19:54.37cr2timer ?
19:55.19goxboxliveWhen flash is on it has a low brightness, when taking picture it brighten the light like a flash. So the light has two modes
19:56.07cr2ok.
19:56.31goxboxliveoractually three. OFF/ON/FLASH
19:56.33cr2sp gpioa10 is flashlight, and gpioa9 is extra power ?
19:56.43goxboxliveyes that might be
19:57.29*** join/#htc-linux WizMaui (n=WizMaui@62.112.90.231)
19:57.30cr2then we have 1 more "led", very bright flash.
19:57.37goxboxliveyes
19:57.40cr2ok.
19:58.04cr2need to test it though.
19:58.15cr2a9 and a10.
19:58.28ljpFYI: opie based on qtopia 4.2 will be hosted at linuxtogo.org
19:58.44goxboxlivewhen?
19:58.47goxboxliveNow?
20:03.48cr2GPIOB4   0x0010   BT related irq
20:03.51cr2GPIOB10   0x0400   ext ?
20:03.59cr2i need to hunt these two.
20:04.15cr2GPIOC15   0x8000   batt. charge related
20:04.30cr2this one is probably for charging in the suspended state.
20:04.46cr2GPIO 92   10000000   charge? usb?   I1   FE
20:04.52cr2and this strange irq.
20:08.07cr2the keyboard does nor resume, right ?
20:08.17cr2s/nor/not/
20:09.40goxboxliveright
20:24.53*** join/#htc-linux FossiDDR (n=Fossi@e176096124.adsl.alicedsl.de)
20:49.43*** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
20:50.59lamikrcr2: Slowly progressing discussion, but yeah if the gnokii has good sms library encoders and decoders, then we should switch to it. It would be nice if gomunicator could handle incoming 7bit, utf8 and utf16 messages and also messages longer than 160 char. But how about outgoing messages, that's unclear for me. Should the selection of used encoder be an gomunicator option that user can change...
20:51.00lamikr...or could the gomunicator detect automatically from the characters that it will need special encoding...
20:57.40Raliththe latter makes more sense
20:59.10cr2saves you some money :)
21:00.17BabelOuedcr2 i keep only three power pin + reset in htcblueangel_acx.... and always same result
21:02.30ljpgoboxlive: Real Soon Now (tm)
21:09.20goxboxlive:-)
21:17.02drnook, Ive finaly managed to get the crosscompiler to work without errors, but what kernel options am I supposed to use for the Sabre? Its alot and some of them dont have a short description
21:17.41cr2BabelOued: there are 2 registers: +4 and +8
21:18.06cr2drno: make htcsable_defconfig
21:18.19drnohave you got a list of those somewhere online cr2?
21:18.20drnook, Ill try to find that
21:18.51cr2you don't need to find it.
21:19.27drnoI think I do,..
21:21.20cr2BabelOued: 0xAB000000 is the base address
21:21.20cr2OVL    R5, 0xAB000004
21:21.20cr2MOV     R0, #0x24
21:21.20cr2STRH    R0, [R5]
21:21.21cr2MOVL    R0, 0x1388
21:21.21cr2BL      usleep
21:21.21cr2LDR     R0, =aTiTw1100bSilic ; (default value: 0x1030505)
21:21.21cr2MOVL    R4, 0xAB000008
21:21.21cr2LDR     R1, [R4]
21:21.22cr2BL      printf
21:21.22cr2MOVL    R0, 0x1388
21:21.22cr2MOV     R3, #0x900
21:21.22cr2STRH    R3, [R5]
21:21.26cr2BL      usleep
21:21.26cr2LDR     R0, =aTiTw1100bDevic ; (default value: 0x8400104C)
21:21.27cr2LDR     R1, [R4]
21:21.27cr2BL      printf
21:21.27cr2OVL is MOVL
21:25.35cr2BabelOued: add this code to your wifi init function, after the powerup and reset.
21:27.09NK`:)
21:34.03*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde)
21:38.09*** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net)
21:38.27BabelOsorry internet problem
21:38.32BabelOthanks cr2
21:38.50BabelOwhere do you find all this code ?
21:43.40NK`I was wondering too
22:49.01*** join/#htc-linux _alge (n=alge@chello080109231226.4.uni-klu.teleweb.at)
22:49.23*** join/#htc-linux ralith_ (n=ralith@soggy202.drizzle.com)
22:58.10*** join/#htc-linux g3gg0_ (n=g3gg0@ppp-82-135-83-37.dynamic.mnet-online.de)
23:10.58goxboxliveNow cr2: Have you gather some more information this evening?
23:14.27*** part/#htc-linux _alge (n=alge@chello080109231226.4.uni-klu.teleweb.at)
23:22.50BabelOcr2: is this ok ?
23:22.52BabelOu16 *addr=((u16 *)(iobase+0x04));
23:23.03BabelO*addr = 0x24;
23:23.12BabelOmdelay(5);
23:23.21BabelO*addr = 0x900;
23:23.31BabelOmdelay(5);
23:25.44*** join/#htc-linux rejon (n=rejon@c-67-174-202-239.hsd1.ca.comcast.net)
23:25.55*** part/#htc-linux rejon (n=rejon@c-67-174-202-239.hsd1.ca.comcast.net)

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.