| 00:08.33 | *** join/#htc-linux Speedy2 (n=Javier_6@cpe-70-95-210-2.san.res.rr.com) |
| 00:13.17 | *** join/#htc-linux TripleDES (n=sergio@84.77.161.251) |
| 01:39.31 | lkcl | cr2: o bollocks. oh well. |
| 01:40.14 | lkcl | cr2: i still have a copy. |
| 01:45.17 | cr2 | lkcl: i'm going to bed. |
| 01:45.21 | cr2 | good night. |
| 01:45.26 | lkcl | night dude |
| 02:23.44 | lkcl | ahh... in the ak4641.c code there is headphone detect, speaker detect, etc. and also there is code where, if the hp or spk are detected, i2c commands are sent to the ak4641. |
| 04:08.32 | *** join/#htc-linux dashbox (n=whatever@h460bd66d.area7.spcsdns.net) |
| 04:08.41 | dashbox | hey is anyone in here? |
| 04:08.42 | dashbox | ?? |
| 04:26.59 | dashbox | humm |
| 04:52.57 | *** join/#htc-linux g3gg0_ (n=g3gg0@ppp-82-135-88-194.dynamic.mnet-online.de) |
| 07:12.13 | *** join/#htc-linux LunohoD_ (n=alex@e180116219.adsl.alicedsl.de) |
| 07:19.16 | *** join/#htc-linux alex-begin (n=345rt@84.232.212.62) |
| 08:28.08 | lkcl | no nobody here dude :) |
| 08:28.17 | lkcl | oh hes gone anyway |
| 08:28.21 | lkcl | ha ha |
| 08:35.21 | *** join/#htc-linux Fossi (n=cmewes@195.177.48.115) |
| 08:41.30 | pierrox | hi |
| 08:41.46 | pierrox | lkcl: i just send you an email about haret |
| 08:42.10 | pierrox | lkcl: where is supposed to be the "right" version of haret ? |
| 08:42.18 | lkcl | there are many, pierrox. |
| 08:42.39 | pierrox | there are too many, efforts are lost |
| 08:44.04 | pierrox | lkcl: you know how sirf binary looks ? |
| 08:45.53 | cr2 | pierrox: Kevin2 tries to change that. |
| 08:46.28 | cr2 | sirf binary? look at the gpsd source. |
| 08:46.28 | pierrox | cr2: i see that there enormous changes in the hh.org repository |
| 08:46.44 | cr2 | enormous ? |
| 08:47.22 | pierrox | cr2: since 2 months i see that there is a lot of changes |
| 08:47.23 | cr2 | yes. i see them too. |
| 08:48.01 | pierrox | cr2: switching from msvc-arm to cegcc is not the smallest change |
| 08:48.42 | cr2 | pierrox: in the "main" haret ? i've done it long ago. |
| 08:49.23 | Fossi | why's that an either or thing? |
| 08:50.28 | pierrox | cr2: no more than 2 months, it is short for me because i am not fast :-) |
| 08:50.33 | cr2 | running ms tool in wine is a too harsh requirement |
| 08:50.50 | cr2 | at least for me |
| 08:51.57 | Fossi | i understand, but can't both be supported? |
| 08:52.05 | pierrox | cr2: certainly but i must admit that i begin to have some doubt about cegcc reliability (maybe only my version though) |
| 08:52.26 | Fossi | or does one have such big advantages that it's a clear choice? |
| 08:52.44 | cr2 | pierrox: if you'll dump the internal cpu stuff, things will go faster. |
| 08:54.07 | pierrox | cr2: which things will go faster ? you mean the values of registers ? |
| 08:54.31 | cr2 | Fossi: it's a clear choice, imho. i'm running gcc-compiled haret for almost a year. if it's good enough for vlc, the it is good for such simple programs like haret. |
| 08:55.17 | cr2 | pierrox: yes, the dump functions in haret. look how long did it take to do that for the universal, and how fast we worked on sable. |
| 08:56.44 | cr2 | Fossi: and it's much easier to do (and control) many things with gcc. inline asm, linking and so on. |
| 08:57.26 | Fossi | i'm not talking against supporting gcc |
| 08:57.45 | Fossi | but the discussion made me wonder why there can't be support for both |
| 08:58.33 | Fossi | i mean there are loads of crosscompiler workspace setup toolchains. |
| 08:58.43 | Fossi | maybe it's not worth the effort... |
| 08:58.47 | Fossi | *shrug* |
| 08:59.23 | pierrox | cr2: i agree with this, but for example i was writting some code to watch gpios on s3c24xx but it seems to have been already done elsewhere |
| 09:00.37 | pierrox | cr2: i wonder if i work on a stable or moving ground |
| 09:01.09 | pierrox | cr2: you told me about pxa dump function, but it is not in the hh.org haret version |
| 09:02.53 | cr2 | Fossi: i don't know how far is Kevin with the main haret, maybe it's possible to use a ms tool .there are many, and they are all incompatible. i would have dropped it completely, to save the maintenance effort. |
| 09:02.54 | pierrox | about the compiler, personnally i would prefer to keep 2 separate makefiles for the moment, one for gcc, and one for msvc-arm |
| 09:03.50 | cr2 | pierrox: all the pxa functions are in the haret-gnu version in xanadux CVS. |
| 09:04.51 | cr2 | pierrox: to be honest, i don't see _any_ benefit of msvc over gcc. |
| 09:05.54 | lkcl | pierrox: no i don't know what SiRF looks like but apparently people have been successful in partially decoding it. gpsd i believe supports it |
| 09:06.17 | pierrox | cr2: i can't access the xanadux CVS as shown in the HaRET wiki page : anoncvs and no paswword does not work. i had a look at the sf.net page and browsing the CVS source tree tell me that sources have not been modified since 4 weeks |
| 09:06.24 | cr2 | lkcl: afaik, it is even possible to reflash a sirf device in linux. |
| 09:06.31 | lkcl | oooo |
| 09:07.09 | lkcl | http://www.die.net/doc/linux/man/man1/sirfmon.1.html |
| 09:07.11 | pierrox | cr2: i hate msvc tools but it seems that my cegcc compiler produces bad code unfortunately |
| 09:07.44 | cr2 | pierrox: i'll try myself an anonymous checkout. it's good when the sources are stable for a month :) |
| 09:07.47 | pierrox | cr2: yes, and i have even the datasheet of the flash chip for the gps :-) |
| 09:08.55 | pierrox | i was able to read some characters from the uart, they seems to be from the gps but as they are binary i was wondering, and also i am not sure of stty parameters |
| 09:08.58 | cr2 | pierrox: it's actually more interesting to dump the flash. they have come to an idea to charge for the sirf3 firmware updates, and have crippled the firmware compared to sirf2! |
| 09:09.38 | cr2 | pierrox: use sirfmon |
| 09:09.55 | cr2 | hmm. you don't have linux running yet ;) |
| 09:10.50 | pierrox | cr2: not on my pda unfortunately |
| 09:10.54 | lkcl | oh btw cr2: looked up details on the ak4641: you remember i mentioned that i2c traffic yesterday when doing those GPIO phone watches? |
| 09:11.16 | lkcl | well the ak4641 has i2c commands it accepts, to switch headphone/speaker audio routing. |
| 09:11.30 | lkcl | which you probably knew anyway. |
| 09:11.41 | lkcl | but it explains why there was so little gpio going on. |
| 09:12.18 | lkcl | also i looked last night at the asoc sourcecode/patch - i really don't get it. |
| 09:12.53 | lkcl | and - the annoying thing is that the HTC Hermes _also_ uses the ak4641 so that's at least _four_ phones with that soundchip and it wasn't included in the damn asoc rewrite |
| 09:15.18 | lkcl | how about.... suspend/resume? that'd keep me busy for a while ha ha |
| 09:15.33 | lkcl | cr2 what was that about lcd, anyway? |
| 09:16.10 | lkcl | damnit this _always_ happens when i do linux kernel hacking: |
| 09:16.13 | lkcl | [35] Stopped vi dis.py (wd: ~/hw6915) |
| 09:16.13 | lkcl | [36] Stopped vi acxwi.txt (wd: ~/hw6915) |
| 09:16.13 | lkcl | [37] Stopped vi sound/arm/htcsable_audio.c |
| 09:16.13 | lkcl | [38]- Stopped vi sound/arm/htcuniversal_audio.c |
| 09:16.13 | lkcl | [39]+ Stopped vi sound/soc/codecs/ak4535.c |
| 09:16.14 | lkcl | ha ha |
| 09:19.21 | pierrox | i had a stupid idea this moring : it should be possible to find where a dll is mapped into memory (brute force search for example), and then change the code into memory to insert some branching to another function -> this way we could do some sort of "breakpoint" in system dll code ? |
| 09:19.37 | pierrox | it would require some special memory mapping and other tricks |
| 09:19.53 | pierrox | but has anyone already worked on such a thing ? |
| 09:22.11 | pierrox | i have a phone call, shit |
| 09:23.58 | lkcl | well speak to Kevin2 because you can always do some sort of suspend on memory watching, with wi. |
| 09:24.41 | cr2 | lkcl: have you dumped the IPL ? |
| 09:24.59 | lkcl | IPL? yes? i think you told me about that one on... 20th oct or so |
| 09:25.23 | lkcl | maybe - no don't think so. |
| 09:25.26 | cr2 | for lcd trace 'pwrlcd 1' and 'pwrlcd 4'. i should have made more userfriendly command names ;) |
| 09:25.44 | lkcl | well as long as they're in HELP, it's ok |
| 09:25.49 | lkcl | ... and they're... not :) |
| 09:26.02 | cr2 | lkcl: i think yes. imo, suspend will be easy. i'll look. |
| 09:26.25 | lkcl | just a straight pd of that page, right? |
| 09:26.32 | lkcl | how much of it? |
| 09:26.38 | cr2 | pd ipl 0x0 0x800 |
| 09:27.10 | cr2 | we should create the mtd mapping for it in linux :) |
| 09:27.33 | cr2 | it's the better way. |
| 09:28.22 | lkcl | ack. |
| 09:28.30 | cr2 | look at the blueangel_rom.c |
| 09:28.51 | lkcl | okie. |
| 09:28.53 | cr2 | it maps 32MB pxa flash. you need only 0x800 bytes :) |
| 09:30.19 | lkcl | ahh right |
| 09:31.54 | lkcl | oh no don't tell me - let me guess: you want me to create an mtd device from this, so i can get to that area under linux?? |
| 09:32.22 | cr2 | lkcl: there a 6 known (to me) devices with ak4641 |
| 09:32.24 | cr2 | yes. |
| 09:33.06 | lkcl | ok it's worth emailing to kd@hh.org then. what's the other one? hx4700, hermes, universal, hw6915.. |
| 09:33.57 | cr2 | beetles and apache. |
| 09:34.35 | cr2 | 4 in "active" production. |
| 09:34.38 | lkcl | ok. ta. |
| 09:42.11 | lkcl | which of beetles or apache is in active production? |
| 09:42.35 | cr2 | apache |
| 09:42.50 | cr2 | in the US. Kevin has one. |
| 09:42.55 | lkcl | yaay. ok |
| 09:43.00 | cr2 | it's a CDMA phone. |
| 09:43.52 | cr2 | beetles is an old sable with normal SD slot and no wifi. the rest is the same. |
| 09:44.10 | cr2 | SD+miniSD. |
| 09:44.20 | cr2 | SD on asic3 and miniSD on pxa. |
| 09:44.37 | lkcl | cool. |
| 09:44.43 | cr2 | it even booted linux from miniSD |
| 09:45.07 | cr2 | a cheaper device to hack on the gps :) |
| 09:45.30 | cr2 | NickBee had it before. |
| 09:45.36 | lkcl | ta. |
| 10:01.46 | pierrox|aw | the pxa haret version uses intel specific features for the wi command and associated :-( can't work on s3c24xx, snif |
| 10:44.17 | *** join/#htc-linux dullard (n=jim@adsl-static-1-30.uklinux.net) |
| 10:53.50 | cr2 | pierrox|aw: it was written only recently. all the 'wgpio'-type stuff and the hardware related commands is portable to any cpu. if i had a samsung cpu, and read the manual, i've added it too. |
| 10:54.51 | cr2 | i have 'wgpio' and 'dump gpio', 'wreg' and so on running on strongarm wince 2.11 (!) |
| 11:09.06 | *** join/#htc-linux scatterp- (n=zsirc@212.32.67.196) |
| 12:42.59 | pierrox | cr2: yes, because wgpio and wreg are polling functions, but wi is using coprocessor values that are implementation defined. these specific features (software debug) are what i was looking for, but unfortunately they are not available on samsung |
| 12:53.12 | cr2 | :( |
| 13:04.09 | lkcl | can anyone think of a reason why wi would not pick up the acx memory address being modified/accessed? i wanted to trace memory-access to 0xab000000 and... well... nothing happens! could it be due to dma? |
| 13:08.33 | *** part/#htc-linux vitb (n=vitb@85.21.88.2) |
| 13:09.30 | pierrox | maybe this has something to do with software debug register and exception handlers in irq.cpp ? perhaps additional configuration of p14 is needed |
| 13:10.27 | Speedy2 | re |
| 13:10.57 | pierrox | re |
| 13:16.08 | cr2 | lkcl: you can run wi on the dma registers |
| 13:16.20 | lkcl | ah cool |
| 13:16.35 | lkcl | allo Speedy2 |
| 13:17.28 | lkcl | btw pof was going to have a look at the CEOS.nbf for sable, for me, see if he can get stuff out of it cos it's in a different format from one's he's seen before |
| 13:41.09 | *** join/#htc-linux pH5 (n=ph5@p54866378.dip.t-dialin.net) |
| 13:42.56 | *** join/#htc-linux prpplague (n=billybob@69.73.209.14) |
| 13:43.13 | prpplague | ho ho ho, merry freakin monday |
| 13:46.28 | prpplague | i don't see any info on the wiki pages about jtag for the hermes |
| 13:46.37 | prpplague | has anyone got that identified? |
| 13:57.03 | prpplague | anyone done a board disassembly? |
| 14:05.22 | pH5 | prpplague: not that I know of, but then again I'm probably not the best informed person on that topic. |
| 14:06.09 | *** join/#htc-linux JTRipper_ (n=jtripper@p54B1D5E5.dip.t-dialin.net) |
| 14:07.08 | prpplague | pH5: ahh ok |
| 14:07.37 | prpplague | there are plenty of potential pins on the board, just surprised no one has done more hardware research |
| 14:11.07 | pH5 | because I don't have one and my brother wouldn't let me take his apart |
| 14:12.39 | pH5 | prpplague: are you going to jumpstart the hermes port now? |
| 14:13.21 | prpplague | pH5: not sure yet, we have a new s3c24xx starting soon, so i might grab some s3c24xx based phones to do some prototyping and code testing |
| 14:17.56 | pH5 | Ah, ok. I haven't seen any Hermes activity besides what can be found on the wiki and in the forums, which is a pity. I think it is a really nice device. |
| 14:36.15 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.179) |
| 15:28.00 | *** join/#htc-linux alex-begin (n=345rt@84.232.212.62) |
| 15:43.02 | *** join/#htc-linux pH5 (n=ph5@e178217016.adsl.alicedsl.de) |
| 15:43.09 | prpplague | pH5: re |
| 15:55.05 | cr2 | prpplague: i think himalaya is the only htc phone, where jtag is identified. |
| 15:55.50 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 15:56.41 | prpplague | cr2: ahh |
| 16:04.07 | prpplague | cr2: should be that hard to identify after removing all the componnents |
| 16:07.30 | *** join/#htc-linux psokolovsky (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com) |
| 16:09.59 | *** join/#htc-linux pleemans (n=peter@d51A5E421.access.telenet.be) |
| 16:10.53 | cr2 | prpplague: if you have a budget for it ;) |
| 16:15.51 | prpplague | cr2: yea |
| 16:31.11 | goxboxlive | hi |
| 16:31.16 | pH5 | hi goxboxlive |
| 16:31.41 | cr2 | hi |
| 16:32.10 | cr2 | pH5: did you move forward with asic3 owm ? |
| 16:32.13 | goxboxlive | i orderd a new battery today. The sevice senter wanted me to ship in universal and the battery. Well, that would have been takeing allmost a month. |
| 16:32.33 | goxboxlive | hmm seems like the handhelds cvs is down. |
| 16:32.35 | cr2 | crazy. |
| 16:41.33 | *** join/#htc-linux mndctrl_ (n=mind@2.81-167-1.customer.lyse.net) |
| 16:43.20 | goxboxlive | cr2: did you try out the latest sounddriver |
| 16:49.03 | cr2 | goxboxlive: which one ? |
| 16:49.30 | goxboxlive | the one you updated yesterday :-) |
| 16:49.36 | cr2 | no. |
| 16:49.55 | cr2 | you are the main testing person :) |
| 16:50.32 | goxboxlive | :-) |
| 16:50.41 | goxboxlive | Did you see my post at xda? |
| 16:51.01 | goxboxlive | We have get rid of one failure message |
| 16:51.19 | goxboxlive | There got to be something else right? |
| 16:56.58 | cr2 | goxboxlive: i've read your post right now. |
| 16:57.11 | goxboxlive | good |
| 16:57.22 | cr2 | ALSA sound/core/oss/pcm_oss.c:940: HW_PARAMS failed: -22 |
| 16:57.31 | cr2 | i think this one comes as a result of |
| 16:57.48 | cr2 | i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 16:58.03 | cr2 | is it an error message ? or just a debug message ? |
| 16:58.15 | cr2 | what is 22 in errno.h ? |
| 16:59.17 | cr2 | #define EINVAL 22 /* Invalid argument */ |
| 16:59.40 | goxboxlive | i dont know, but as you saw, we got rid of the other message |
| 17:00.01 | cr2 | <PROTECTED> |
| 17:00.01 | cr2 | <PROTECTED> |
| 17:00.01 | cr2 | <PROTECTED> |
| 17:00.01 | cr2 | <PROTECTED> |
| 17:00.46 | cr2 | this one shows that the command was sent to 0x12 (ak4641) |
| 17:00.49 | cr2 | i2c_adapter i2c-0: master_xfer[0] W, addr=0x12, len=2 |
| 17:01.10 | cr2 | have you enabled all i2 debugging messages ? |
| 17:01.19 | cr2 | s/i2 /i2c / |
| 17:07.43 | goxboxlive | I havent added anything debug other than wahts in the htcuniversal_defconfig |
| 17:08.39 | goxboxlive | I can try to do that, but it seems like the hh.org anoncvs is down at the moment, so i dont have any kernels to compile |
| 17:20.25 | cr2 | ok. |
| 17:21.45 | *** join/#htc-linux asylumed (n=insanity@196.211.116.2) |
| 17:22.38 | goxboxlive | cr2 could you try to login to hh.org, not as a user but as a anoncvs |
| 17:22.52 | goxboxlive | bash$ export CVSROOT=:pserver:anoncvs@anoncvs.handhelds.org:/cvs |
| 17:22.52 | goxboxlive | bash$ cvs login |
| 17:22.52 | goxboxlive | password: anoncvs |
| 17:24.11 | cr2 | no. it hangs. |
| 17:30.03 | cr2 | hmm. wince sends ak4641 i2c commands, that linux ak4641 does not even know ;) |
| 17:30.25 | cr2 | they are not implemented. |
| 17:30.39 | cr2 | must look in the datasheet. |
| 17:32.04 | goxboxlive | hmm ok, sounds intresting |
| 17:32.45 | cr2 | it is a solution for our problem, but anyway. |
| 17:40.09 | goxboxlive | cr2: Should i enable this in the kernel: Intel PXA2XX I2C Slave comms support (I2C_PXA_SLAVE) |
| 17:41.02 | cr2 | no. |
| 17:41.21 | cr2 | mic switch and mic power are done by commands, not gpio. |
| 17:41.54 | goxboxlive | ok |
| 17:42.26 | goxboxlive | what about this then: I2C/SMBus Test Stub (I2C_STUB) |
| 17:43.15 | goxboxlive | If not i have all the debug messaging active. |
| 17:43.59 | cr2 | no. |
| 17:46.23 | cr2 | hmm. looks like a bug in wince. |
| 17:47.24 | goxboxlive | realy |
| 17:48.25 | cr2 | it's a _feature_, but i don't like it this way. |
| 17:49.41 | cr2 | we can delete the mic power function, it's done over i2c on universal. |
| 17:50.24 | goxboxlive | ok |
| 17:55.40 | cr2 | equalizer is not implemented. also used by wince. |
| 17:59.22 | cr2 | interesting. there is an asic3 gpio that tells you that mic is on, although it is controlled by i2c command :) |
| 18:11.36 | *** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 18:14.17 | goxboxlive | have to reboot |
| 18:16.54 | cr2 | BabelOued: can you trace the camera on blueangel ? |
| 18:20.03 | *** join/#htc-linux asylume2 (n=insanity@196.211.116.2) |
| 18:20.59 | cr2 | asylumed: hi |
| 18:21.23 | asylumed | cr2: hey |
| 18:21.43 | cr2 | can you check the camera i2c id ? |
| 18:22.03 | asylumed | sure... how though? |
| 18:22.14 | cr2 | i think i have also found a bug in the sound driver. |
| 18:22.36 | cr2 | set trace 0xae301688 |
| 18:22.40 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 18:22.46 | cr2 | and then switch the camera on/off |
| 18:23.00 | cr2 | post the log to pastebin. |
| 18:23.50 | cr2 | i have added the PWM macros/definitions to ipaq-asic3.h, you should fix the ba backlight driver. |
| 18:24.25 | asylumed | k, will have a look a little later at that... |
| 18:27.39 | asylumed | cr2: during wi? |
| 18:28.05 | asylumed | cr2: it spits out alot... |
| 18:28.21 | asylumed | cr2: after i ignore many irq's |
| 18:28.34 | cr2 | asylumed: mask 7,10, |
| 18:28.39 | cr2 | asic3 |
| 18:28.40 | cr2 | hmm |
| 18:29.02 | cr2 | 3,11 |
| 18:29.12 | asylumed | 00ba5e03: debug 010b20c8: e5c20000(strb) 000000a4 ae301688 |
| 18:29.15 | asylumed | 00ba5edf: debug 010b21c8: e5c13000(strb) 000000c0 ae301688 |
| 18:29.21 | cr2 | ok. |
| 18:29.31 | cr2 | 0xa4/2=0x52 |
| 18:29.54 | asylumed | ? |
| 18:29.56 | cr2 | http://wiki.xda-developers.com/index.php?pagename=HTC_I2C_IDs |
| 18:30.04 | cr2 | question solved. |
| 18:30.32 | asylumed | kewl |
| 18:30.37 | cr2 | you have adcm2650 cam |
| 18:30.49 | cr2 | but imho, there are models with a different cam |
| 18:31.31 | BabelOued | cr2: hi |
| 18:31.48 | BabelOued | cr2: i m working on acx_cs.c |
| 18:32.16 | cr2 | BabelOued: the hx4700_pcmcia is buggy |
| 18:32.35 | cr2 | it does not handle the asic3 clocks properly. |
| 18:32.45 | *** join/#htc-linux mreimer (n=mreimer@bilbo.vpop.net) |
| 18:32.45 | asylumed | cr2: what is a 2408W btw? |
| 18:32.51 | cr2 | i have fixed 1 bug (indirectly), but that's not all. |
| 18:33.03 | cr2 | asylumed: wifi eeprom. |
| 18:33.23 | cr2 | eth addr & other settings. |
| 18:39.09 | pH5 | cr2: the first byte in the i2c stream is the address*2? |
| 18:39.20 | pH5 | then the magician has the camera at 0x30 |
| 18:40.23 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 18:48.48 | *** join/#htc-linux RoEn_PC (n=roen@p54A6615B.dip.t-dialin.net) |
| 19:10.21 | alex-begin | cr2: is the HARET Reverse eng tool designed specially for HTC devices or works on any kind of ARM device? |
| 19:16.34 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 19:33.18 | cr2 | pH5: it's 2*id for write, 2*id+1 for read |
| 19:33.47 | cr2 | alex-begin: for many arm devices, maybe for ,ost |
| 19:34.44 | cr2 | pH5: welcome to the club: OV9650 |
| 19:35.01 | goxboxlive | are cr2: you figureing something out with the sound driver |
| 19:36.01 | cr2 | goxboxlive: how can we test the pxa DMA ? |
| 19:38.39 | goxboxlive | i dont know |
| 19:39.05 | cr2 | the lcd uses its own separate dma channels |
| 19:39.21 | cr2 | udc ? |
| 19:40.05 | cr2 | camera is a good candidate, but we don't have the tested and working drivers. |
| 19:41.59 | goxboxlive | I have the latest kernel running now, copied the cource from my desktop pc. |
| 19:42.14 | goxboxlive | Is there any test you want me to do while i have it running? |
| 19:43.05 | cr2 | no. at least not yet. |
| 19:45.59 | goxboxlive | ok |
| 19:51.00 | *** join/#htc-linux doepiedoe (n=doepiedo@84-53-96-51.fiber.unet.nl) |
| 20:02.17 | *** join/#htc-linux RoEn_PC (n=roen@p54A6615B.dip.t-dialin.net) |
| 20:04.04 | pH5 | How is the backlight of the phone keys supposed to be hooked up? I currently have a magician:keys led in the led driver. |
| 20:05.45 | cr2 | pH5: nobody knows, because there are no free software phones. |
| 20:06.27 | pH5 | cr2: I also wonder about the vibra. Should I put it as force-feedback device on the keypad? ;) |
| 20:06.52 | cr2 | we have all events on /dev/input/event1 now, and it is hardcoded in opie. ugly ;) |
| 20:07.13 | cr2 | i think it's a pure userspace issue.- |
| 20:07.35 | cr2 | which program will use/manipulate it ? |
| 20:08.16 | pH5 | cr2: the vibra motor or the keypad illumination? |
| 20:08.52 | cr2 | keypad is tied to the lcd backlight in wince. |
| 20:08.59 | pH5 | first one should be used by phone and calendar(alarm) software, the latter could blink for alarms, but otherwise we could just connect it to the backlight as in wince |
| 20:09.02 | pH5 | ah, yes :) |
| 20:09.09 | cr2 | you mean, the phone keypad ? |
| 20:09.38 | cr2 | i have 3 "backlights": keyboard, phone keys, lcd |
| 20:10.00 | pH5 | yes, only the lift/hangup keys are illuminated here. |
| 20:10.01 | pH5 | does the universal also have one of those light sensors? |
| 20:10.03 | cr2 | well, we need a general solution. |
| 20:10.08 | cr2 | yes. |
| 20:10.42 | pH5 | do you have a hwmon driver for it? |
| 20:10.55 | cr2 | it's a button. no. |
| 20:11.09 | pH5 | oh. just a gpio line? |
| 20:11.27 | cr2 | yes, you can't measure w/m^2 with it :) |
| 20:12.06 | pH5 | :) schade |
| 20:12.13 | cr2 | it's not a photometer ;) |
| 20:13.02 | cr2 | next you would like to have a spectral resolution, measure in near infrared ... |
| 20:13.16 | cr2 | i'd enjoy something like that too :D |
| 20:13.45 | cr2 | or in UV. |
| 20:14.38 | pH5 | ok, I see we're expecting too much from htc :) |
| 20:16.19 | cr2 | why don't you have a working phone ? even lkcl has it :) |
| 20:16.42 | goxboxlive | cr2: We had the light sensor working in .12 . Why is it disabled now? COuldnt we link it to the keyboard light so it will switch on/off? |
| 20:16.49 | pH5 | http://en.pastebin.ca/253443 |
| 20:17.08 | pH5 | I just hooked it up to the pxa_serial_funcs of the btuart - what could I do next? |
| 20:18.06 | pH5 | is the light sensor on a pxa gpio, or on asic3? there is one on the hx4700, too, but I never noticed any code to handle it. |
| 20:18.29 | cr2 | pH5: do you understand the phone protocol ? |
| 20:18.34 | cr2 | asic3 gpio. |
| 20:18.50 | *** join/#htc-linux RoEn_PC (n=roen@p54A6615B.dip.t-dialin.net) |
| 20:19.00 | pH5 | cr2: \x02\x16AT-COMMAND\r\x02 |
| 20:19.11 | cr2 | there was an opie handler for light sensor long ago. |
| 20:19.12 | pH5 | with that I can start to reproduce the commands I traced with haret |
| 20:19.22 | cr2 | ok. |
| 20:19.45 | pH5 | but I don't have sound yet.. |
| 20:20.20 | cr2 | goxboxlive: there is no such "input" in linux/input.h ;) |
| 20:20.32 | goxboxlive | ok |
| 20:20.36 | doepiedoe | http://the.taoofmac.com/space/AT%20Commands |
| 20:20.48 | cr2 | there is LID, and ROTATE. that's all. |
| 20:21.30 | cr2 | doepidoe: http://wiki.xda-developers.com/index.php?pagename=HTC_GSM |
| 20:21.46 | doepiedoe | :) |
| 20:22.11 | cr2 | it seems zaurus has only these 2 "inputs". |
| 20:22.13 | doepiedoe | i want to run linux on my universal, but SD seems to be defect :S |
| 20:22.23 | cr2 | SD ? |
| 20:22.27 | doepiedoe | yeah SD slot |
| 20:22.36 | cr2 | in wince ? |
| 20:22.38 | doepiedoe | yeah |
| 20:22.56 | doepiedoe | can't boot linux without rootfs |
| 20:23.01 | doepiedoe | you can boot |
| 20:23.01 | doepiedoe | ;) |
| 20:23.02 | cr2 | we don't have a doc g3 driver yet. |
| 20:23.14 | doepiedoe | but you can't do much |
| 20:23.22 | pH5 | cr2: do your phones survive suspend? (open gsm ctrl -> suspend -> resume -> does it still work without reset?) |
| 20:23.37 | cr2 | you can flash linux. |
| 20:23.47 | doepiedoe | is there a bootloader already for universal? |
| 20:23.51 | cr2 | pH5: ask goxboxlive, i'm not a linux user myself :) |
| 20:24.15 | cr2 | doepiedoe: it is possible to write one. |
| 20:24.21 | doepiedoe | lol |
| 20:24.23 | doepiedoe | indeed ;) |
| 20:24.32 | cr2 | but i'll be a bit reluctant to be the first to flash it. |
| 20:24.39 | doepiedoe | :P |
| 20:24.43 | doepiedoe | i'll flash mine ;) |
| 20:24.49 | cr2 | no, really. hx4700 has a linux bootloader. |
| 20:25.02 | cr2 | we only need to adjust the gpios. |
| 20:25.06 | doepiedoe | aha |
| 20:25.16 | cr2 | and hx4700 flashes it. |
| 20:25.16 | doepiedoe | sounds nice |
| 20:25.31 | cr2 | the problem is that we need a g3 driver. |
| 20:26.11 | cr2 | the first step will be to write a "readonly" raw block mtd device driver |
| 20:26.34 | cr2 | it sounds awfully complex, but most of the research is already done. |
| 20:26.57 | doepiedoe | some day i will help this community ;) |
| 20:27.28 | doepiedoe | g3 is for which part>? |
| 20:27.29 | cr2 | http://wiki.xda-developers.com/index.php?pagename=HTC_DiskOnChip |
| 20:27.33 | doepiedoe | ahhh |
| 20:27.34 | doepiedoe | ;) |
| 20:27.38 | doepiedoe | like on the onld mda |
| 20:27.49 | doepiedoe | yeah, then i don't need SD anymore |
| 20:28.08 | doepiedoe | so the doc contains the bootloader? |
| 20:28.12 | cr2 | yes. |
| 20:28.25 | cr2 | it contains everything. |
| 20:28.34 | doepiedoe | so can't it be done by jtag? |
| 20:28.40 | cr2 | so it's better to test the g3 driver first on blueangel. |
| 20:28.45 | doepiedoe | ahhh |
| 20:28.51 | doepiedoe | get it |
| 20:29.08 | cr2 | you know the pins and want to solder and don't care about the warranty ? |
| 20:29.15 | cr2 | well, you maybe :) |
| 20:29.32 | doepiedoe | would you like to write linux support for flash and then flash the bootloader? |
| 20:29.35 | lkcl | BabelOued: you're dealing with acx, right? |
| 20:29.37 | lkcl | allo darlins |
| 20:29.40 | cr2 | blueangel has the bootloader and wince in pxa flash |
| 20:29.53 | doepiedoe | pxa of ba has flash? |
| 20:29.54 | cr2 | so the trashed g3 is not a problem. |
| 20:30.08 | cr2 | yes, it's 263 |
| 20:30.09 | doepiedoe | or flash connected to it... |
| 20:30.12 | doepiedoe | nice |
| 20:30.15 | cr2 | 32MN |
| 20:30.19 | cr2 | 32MB |
| 20:30.20 | doepiedoe | i had one to once ;) |
| 20:30.25 | doepiedoe | but broke the display.... |
| 20:30.37 | cr2 | me too :( |
| 20:30.39 | lkcl | doepiedoe i got a blueangel, himalaya, executive, hp6915 - i could do say the blueangel plus i know someone who smashed their screen so i could do that one. |
| 20:30.40 | doepiedoe | :S |
| 20:30.47 | cr2 | but i have replaced it myself. |
| 20:30.48 | doepiedoe | ; |
| 20:31.11 | lkcl | damn things are glass |
| 20:31.33 | cr2 | ba is a nice phone. |
| 20:31.36 | doepiedoe | mine were h3900,h5450,h5550,h2210,xda1,mda1,mda2,h6340,blueangel,universal,hw6515 |
| 20:31.38 | lkcl | so doepiedo, you can do g3 right? |
| 20:31.42 | doepiedoe | i miss a couple :P |
| 20:31.47 | doepiedoe | g3? |
| 20:31.53 | doepiedoe | 3g |
| 20:31.55 | cr2 | we only need to understand the ati vidcap register layout. |
| 20:31.59 | lkcl | cool! oh you have a htc beetles! |
| 20:32.07 | doepiedoe | there was a guy on the main channel |
| 20:32.08 | doepiedoe | noone |
| 20:32.13 | doepiedoe | who wrote an ati driver |
| 20:32.18 | cr2 | and finally find the wifi there. |
| 20:32.26 | doepiedoe | for h740 or something |
| 20:32.38 | cr2 | with video capture ? |
| 20:33.04 | lkcl | g3 - DoC m-systems |
| 20:33.14 | cr2 | the chip has even more stuff packed in, but it's not used by htc |
| 20:33.50 | doepiedoe | http://www.mnementh.co.uk/eseries/ |
| 20:33.58 | doepiedoe | dont know, but think it was ati |
| 20:34.03 | doepiedoe | not the same chip |
| 20:34.04 | cr2 | asylumed: what kernel version are you using ? |
| 20:34.07 | doepiedoe | but this guy was good ;) |
| 20:34.16 | asylumed | 2.6.18-hh0 |
| 20:34.28 | doepiedoe | ATI 215W100AFA12 FA51571 024100 Taiwan |
| 20:34.41 | cr2 | it's more simple chip. |
| 20:34.49 | doepiedoe | ok |
| 20:34.59 | cr2 | and the w100 driver was available on zaurus |
| 20:35.06 | doepiedoe | aha |
| 20:35.21 | doepiedoe | i'll have to go ;) will be back soon, nice talking to you :) |
| 20:36.13 | pH5 | hey lkcl, where is your code to talk to your phone again? |
| 20:36.22 | lkcl | que? which one, dude? |
| 20:36.47 | pH5 | you played with the sable's phone chipset lately, right? |
| 20:38.23 | pH5 | lkcl: can you do anything against the pxa serial ports being shut down during suspend? |
| 20:38.28 | cr2 | pH5: http://hands.com/~lkcl/hp6915/ |
| 20:38.42 | pH5 | I haven't figured out yet how to not reset the GSM chip with a suspend/resume cycle |
| 20:38.43 | pH5 | lkcl: thanks! |
| 20:39.16 | lkcl | pH5: you give me a little more credit than i deserve by asking me that question: i haven't a clue :) |
| 20:39.18 | cr2 | pH5: how do you suspend the CPLD ? |
| 20:39.59 | pH5 | cr2: not at all, it comes back as I left it. |
| 20:40.16 | cr2 | on blueangel the GSM alert gpio is on pxa, but on universal it is on asic3. |
| 20:40.24 | cr2 | ok. |
| 20:41.58 | pH5 | Perhaps I can do something more to save some energy, but the cpld controls the power to the gsm chip, so I think it can't be completely shut down anyway. |
| 20:43.23 | *** join/#htc-linux RoEn_PC (n=roen@p54A6615B.dip.t-dialin.net) |
| 20:43.31 | lkcl | ph5: oh - yes, i did play with the gsm on the hw5915, for a bit. |
| 20:44.02 | lkcl | enough to work out the protocol - and then, knowing that i could come back to it any time, left it at that. |
| 20:44.08 | cr2 | lkcl: do you have a raw 0x800 dump ? i forgot how to convert hexdump to binary. |
| 20:44.16 | lkcl | sure. hehe |
| 20:44.19 | lkcl | mmm...... |
| 20:44.37 | lkcl | i did - i just haven't copied it off the memorycard yet.. |
| 20:45.03 | cr2 | ok, do it, then we'll probably have a working cpu suspend for you. |
| 20:45.54 | lkcl | http://hands.com/~lkcl/hp6915/pd.0x0.0x800 |
| 20:46.03 | lkcl | but a hexdump doesn't quite look right o well |
| 20:46.25 | lkcl | and i'm sure it hasn't taken enough... |
| 20:46.30 | *** join/#htc-linux g3gg0__ (n=g3gg0@ppp-82-135-10-33.dynamic.mnet-online.de) |
| 20:46.38 | lkcl | or too much. |
| 20:47.25 | lkcl | looks very odd. gonna do it again |
| 20:47.32 | cr2 | the first instruction is awy, but it's not a problem. |
| 20:47.38 | cr2 | 0x37312e47 |
| 20:47.46 | cr2 | this is your "magic" |
| 20:48.44 | cr2 | i had one for Kevin's apache, and for my universal. need to ask goxboxlive about his value. hopefully it's the same number for a given device type. |
| 20:49.16 | cr2 | do you have an mtd driver for it ? :) |
| 20:49.44 | lkcl | oh, oops - gimme a few seconds minutes or something ha ha |
| 20:54.19 | lkcl | hmmm... i have to define CONFIG_MTD_MAP_BANK_WIDTH_N |
| 20:55.09 | cr2 | also for the readonly device ? |
| 20:55.58 | cr2 | lkcl: i have found your dump. but write the mtd driver anyway :) it can be reused on other phones. |
| 20:56.03 | lkcl | ack |
| 20:56.09 | goxboxlive | this was fun: http://www.gsm-modem.de/gsm-modem-faq.html |
| 20:57.00 | goxboxlive | WHat value? |
| 21:02.09 | lkcl | well it helped to put _some_ devices in... |
| 21:02.50 | cr2 | goxboxlive: the second 4bytes of IPL. |
| 21:03.12 | cr2 | pd 0x4 4 |
| 21:03.14 | cr2 | in haret |
| 21:03.50 | goxboxlive | do i have to do that now? I am trying to send a sms |
| 21:04.01 | cr2 | no. |
| 21:04.21 | cr2 | by hand ? :) |
| 21:09.14 | goxboxlive | yes with 'cu' |
| 21:09.14 | cr2 | cool. |
| 21:10.11 | pH5 | goxboxlive: :D |
| 21:13.31 | cr2 | lkcl: you can just grab the suspend driver from the universal. |
| 21:13.49 | lkcl | i think that's where i started o no it was the hx4700 one |
| 21:13.54 | cr2 | i wonder if we need 10 clones of it ;) |
| 21:14.14 | cr2 | hx4700 is really weird. like blueangel. |
| 21:16.25 | cr2 | lkcl: we need a little brainstorming to get the raw g3/g4 minidriver. |
| 21:16.34 | lkcl | okie. sure |
| 21:17.06 | cr2 | the idb is in paq69/ |
| 21:17.15 | *** join/#htc-linux alex-begin (n=345rt@84.232.212.62) |
| 21:17.26 | cr2 | i was already trying to convert asm to c. |
| 21:17.36 | goxboxlive | yyiiihhhaaa i have send a sms |
| 21:17.46 | cr2 | to yourself ? |
| 21:18.12 | goxboxlive | no to my girfriends phone |
| 21:18.19 | goxboxlive | I can try to send to my self to |
| 21:18.43 | lkcl | ok can do gimme a moment - asm to c is... fuuun. can you do a print/dump or point me in the direction of what option i'm supposed to get an asm dump to text file? i forgot... |
| 21:18.46 | cr2 | i'm curious if you'll get it in cu :) |
| 21:19.30 | lkcl | well done goboxlive |
| 21:19.33 | cr2 | lkcl: should i save the lst file for you ? |
| 21:19.38 | *** join/#htc-linux Scott_Waye (n=scott@213-152-36-33.dsl.eclipse.net.uk) |
| 21:19.41 | lkcl | anywhere doesn't matter |
| 21:19.53 | lkcl | /home/cr2/paq69 is as good a place as any |
| 21:22.20 | lkcl | ok now i am the proud buyer (not owner...) of an htc hermes. hooray! |
| 21:23.01 | cr2 | crazy. |
| 21:23.03 | lkcl | ok i gotta rest for a bit before tackling asm to c. get food, get drink etc. |
| 21:23.04 | pH5 | wow, this seems to be a good day for hermes development |
| 21:23.06 | lkcl | yep. |
| 21:23.30 | lkcl | well it's one with a samsung processor and i wanted to try that one, see if it's any easier than this frickin about. |
| 21:23.38 | mreimer | pH5: why, what happened? (I'm trying to get hermes 2.5 working on an i.MX device.) |
| 21:23.40 | lkcl | all these htc proprietary chips are pissing me off |
| 21:23.58 | cr2 | pH5: a proper s3* haret is needed anyway. |
| 21:24.09 | lkcl | and i suspect they are annoying htc as well :) |
| 21:24.18 | cr2 | lkcl: which ? |
| 21:24.26 | lkcl | hello htc, nice to see you're watching the irc logs! |
| 21:24.33 | lkcl | do say hello one day, ok? |
| 21:25.20 | pH5 | mreimer: htc hermes, not hermes wlan |
| 21:25.31 | cr2 | lkcl: i've saved the hw6900_ipl.lst file, but let me put my own commented file for universal. |
| 21:25.34 | mreimer | pH5: ah, ok |
| 21:26.11 | cr2 | lkcl: hermes has the 54mbit successor to acx100 ? good luck :) |
| 21:27.12 | cr2 | lkcl: the relevant code starts at loc_25C: |
| 21:28.03 | pH5 | also it has one of the popular ati imageon chipsets |
| 21:29.48 | cr2 | pH5: accel video ? the fb address is in SRAM. |
| 21:30.14 | cr2 | and video/audio capture. will take some time to get right. |
| 21:30.34 | cr2 | lkcl: ipldump_jar.lst3 |
| 21:39.28 | lkcl | ta cr2 |
| 21:39.45 | lkcl | ha ha cr2. all i really want is a phone :) |
| 21:42.08 | lkcl | one i can say is mine, all mine mwahahaha |
| 21:43.09 | pH5 | good night |
| 21:44.07 | lkcl | seg000:0000025C STR R2, [R3] |
| 21:44.07 | lkcl | seg000:00000260 LDR R3, MDMRS |
| 21:44.16 | lkcl | starting from there? |
| 21:44.17 | cr2 | lkcl: the label |
| 21:44.35 | lkcl | you mean loc_23C: |
| 21:44.37 | lkcl | ? |
| 21:45.43 | cr2 | take better this file ipldump_jar.lst3 |
| 21:45.48 | cr2 | in the root dir |
| 21:46.00 | lkcl | i did. |
| 21:46.24 | cr2 | loc_2B0 |
| 21:46.44 | lkcl | okie. |
| 21:47.12 | cr2 | keep this page open |
| 21:47.14 | cr2 | http://wiki.xda-developers.com/index.php?pagename=HTC_DiskOnChip |
| 21:48.28 | cr2 | we need to understand exactly which blocks are read there, so we may try to read the others. |
| 21:51.29 | lkcl | k |
| 21:51.37 | lkcl | what does MVN R2, R2 mean? |
| 21:53.36 | cr2 | move negate |
| 21:54.01 | cr2 | R2=~R2 |
| 21:54.01 | lkcl | Move Not Rd = NOT Op2 |
| 21:54.08 | lkcl | full bit-reverse? |
| 21:54.20 | lkcl | yep it is |
| 21:54.30 | cr2 | yes. to check chipid_1 (G3=0xfdff(~200), G4=0xfbff(~400)) |
| 21:54.59 | cr2 | BNE ret0 |
| 21:55.03 | *** part/#htc-linux Scott_Waye (n=scott@213-152-36-33.dsl.eclipse.net.uk) |
| 21:55.06 | cr2 | means not found. |
| 21:55.10 | lkcl | o bugger. arm processors are pipelined, right? |
| 21:55.36 | cr2 | yes, but this code is straitforward. |
| 21:56.10 | goxboxlive | hmm what does this mean then: +CNMI: 0,0,0,0,0 |
| 21:56.10 | goxboxlive | +CNMI: (0,1,2),(0,1,2,3),(0,2),(0,2),(0,1) |
| 21:56.57 | cr2 | http://wiki.xda-developers.com/index.php?pagename=HTC_GSM |
| 21:57.20 | cr2 | Display new incoming SMS |
| 22:01.23 | goxboxlive | when i try to recive a sms with at+cmgr=1 i get this failure: +CMS ERROR: 321 |
| 22:05.32 | cr2 | hmm. one more 16bit reg. |
| 22:06.38 | cr2 | put into wiki. |
| 22:07.03 | cr2 | goxboxlive: can you read it with gomunicator ? |
| 22:07.20 | goxboxlive | no i dont think so. I am running opie |
| 22:07.25 | lkcl | ok i'm getting _really_ confused. |
| 22:07.39 | goxboxlive | Is there a way to get gommunicator running with Opie? |
| 22:07.48 | lkcl | LDRH R1, [R2] is equivalent to |
| 22:08.04 | lkcl | r1 = *((short*)r2) |
| 22:08.06 | lkcl | right? |
| 22:09.06 | lkcl | ack about the extra reg, cr2 |
| 22:09.20 | goxboxlive | cr2: am i using UMTS ? +BANDSEL: 2,121 |
| 22:12.20 | goxboxlive | +BANDSEL: 2,121 |
| 22:12.21 | goxboxlive | cu: Got termination signal |
| 22:12.21 | goxboxlive | Disconnected. |
| 22:12.25 | cr2 | lkcl: yes |
| 22:12.42 | lkcl | ack |
| 22:12.43 | cr2 | goxboxlive: 121 ? |
| 22:12.53 | goxboxlive | i dont know what that is |
| 22:13.00 | goxboxlive | it came up when i use the bandsel |
| 22:13.03 | cr2 | bandsel=2 -> UMTS, 1 -> GSM |
| 22:13.13 | goxboxlive | Then i am using UMTS |
| 22:13.31 | cr2 | btw, do you can /dev/ttyUSB0 ? |
| 22:13.36 | lkcl | cr2 device id select and device ID select are 0xa and 0x1a. why both? |
| 22:13.43 | cr2 | s/can/have/ |
| 22:14.23 | cr2 | lkcl: are they documented ? |
| 22:14.24 | goxboxlive | no |
| 22:14.32 | lkcl | both are documented as device id select |
| 22:14.39 | lkcl | one is capitalised :) |
| 22:14.47 | cr2 | hmm. don't remember. |
| 22:15.06 | cr2 | goxboxlive: but you have usbhost loaded ? |
| 22:15.13 | cr2 | mount usbfs |
| 22:15.17 | goxboxlive | no |
| 22:15.19 | cr2 | look in /sys/ |
| 22:15.37 | goxboxlive | no usbfs there |
| 22:15.51 | cr2 | well, we can force creation of /dev/ttyUSB0 and /dev/ttyUSB1 |
| 22:15.57 | cr2 | with right modules |
| 22:16.15 | lkcl | cr2: a byte is being stored in 0x3 and the doc page says it's a short. this a problem? |
| 22:16.16 | goxboxlive | why do i want usbfs? |
| 22:16.17 | cr2 | then you are supposed to see something on /dev/ttyUSB0 |
| 22:16.24 | goxboxlive | ok |
| 22:16.43 | cr2 | lkcl: don't think so, but it's dumb. |
| 22:16.45 | lkcl | ok |
| 22:17.28 | cr2 | i can look into the docs. |
| 22:18.21 | doepiedoe | can someone tell me what asic3 is? |
| 22:18.30 | doepiedoe | an extender for pxa proc? |
| 22:19.27 | cr2 | doepiedoe: a companion chip. it's traditional since sa1100 to have such device. |
| 22:19.42 | doepiedoe | ahh ok |
| 22:19.44 | cr2 | for other useful functions. |
| 22:20.07 | cr2 | like usb host, gpio, SD , etc. |
| 22:20.30 | doepiedoe | so gpio goes via asic3? |
| 22:20.46 | lkcl | it's to complement the fact that there's not enough gpio on a pxa to do phone functionality. it has another 64 bits of gpio, it has another SPI bus - lots and LOTS of stuff. |
| 22:21.01 | *** join/#htc-linux alex-begin (n=345rt@84.232.212.62) |
| 22:21.16 | alex-begin | anyone here started to hack the HTC METEOR |
| 22:21.18 | alex-begin | ? |
| 22:21.34 | *** join/#htc-linux lamikr|lap (n=chatzill@aragorn.kortex.jyu.fi) |
| 22:22.02 | alex-begin | has got a Q U A L C O M M MSM6275 chipset solution |
| 22:22.05 | doepiedoe | yeah i did, and now it's dead :) |
| 22:22.20 | alex-begin | need some docs from qualcom site |
| 22:22.32 | alex-begin | seems you need some special code or something |
| 22:22.48 | alex-begin | i sent some emails and they neverreplied |
| 22:23.20 | doepiedoe | (HTC MTeoR) has the same Processor/ Chipsets Qualcomm MSM6275 as the HERMES |
| 22:23.35 | alex-begin | ooo |
| 22:24.03 | doepiedoe | it seems to be only radio chip |
| 22:24.15 | doepiedoe | no |
| 22:24.17 | doepiedoe | its cpu |
| 22:24.40 | doepiedoe | samsung 300mhz :S |
| 22:24.42 | alex-begin | are there any manuals about the MSM6275 and RF and TX controllers used on these devices? |
| 22:24.46 | alex-begin | i mean detailed ones |
| 22:24.54 | *** join/#htc-linux florian__ (n=fuchs@84.245.179.44) |
| 22:24.56 | alex-begin | not HARET stuff |
| 22:25.10 | doepiedoe | like this? http://www.sierrawireless.com/resources/documents/products/MC8755_65%20Datasheet%20rev%201.2.pdf |
| 22:25.39 | *** join/#htc-linux florian__ (n=fuchs@84.245.179.44) |
| 22:26.30 | alex-begin | nope |
| 22:26.42 | doepiedoe | it seems there's just a linux driver available for the chip... |
| 22:26.57 | alex-begin | doepiode : why did your phone die? |
| 22:26.59 | doepiedoe | from pcidevs.txt: D000CQualcomm MSM6275 UMTS chip |
| 22:29.16 | florian | re |
| 22:29.27 | doepiedoe | crappy |
| 22:29.51 | alex-begin | what happened |
| 22:29.52 | alex-begin | ? |
| 22:29.53 | cr2 | universal has MSM6250 |
| 22:30.40 | cr2 | and you talk on the serial port. there is also an usb port, but it's for the umts, and we don't know what to do with it. |
| 22:31.10 | cr2 | the rest is not your business :) it runs some RT unix. |
| 22:32.14 | *** join/#htc-linux ljp (n=lpotter@203.94.178.46) |
| 22:33.24 | doepiedoe | i just drove over my toe |
| 22:33.39 | lkcl | well done, doepiedoe! |
| 22:34.21 | cr2 | lkcl: 0x1a wiki info is wrong, i'll correct it. |
| 22:34.28 | lkcl | ok hooray |
| 22:34.33 | alex-begin | all my friends with htc devices only had very minor problems with their hardware |
| 22:36.44 | lkcl | well until linux is on them i ain't using them, so i only have one problem: wince is on them, |
| 22:36.58 | lkcl | it's a pretty big problem. |
| 22:37.09 | lkcl | the up-side is: all my htc devices are in pretty good condition :) |
| 22:37.25 | alex-begin | heard about this : RTR6250, RFR6250??? |
| 22:37.39 | cr2 | done. |
| 22:38.08 | doepiedoe | anyone knows if theres a key combination to flash the wince image to a SD card on the Universal? |
| 22:40.31 | cr2 | lkcl: the 0x800-0x1000 is the "flash window". you select it in some way, and it's mapped. i was curious why 0x800 is so special in the code. |
| 22:40.46 | cr2 | mapped to the real data. |
| 22:40.51 | lkcl | ooo |
| 22:41.23 | lkcl | what's BL? |
| 22:41.30 | cr2 | and the 0x101a is the read register if i understand correctly. |
| 22:41.30 | lkcl | branch long? |
| 22:41.34 | lkcl | ok |
| 22:41.58 | cr2 | branch & link. 'call' in fortran & basic :) |
| 22:43.13 | cr2 | 0x800 it's 2K chunk ? |
| 22:43.33 | lkcl | ah cool |
| 22:43.39 | lkcl | wtf _this_ ORR R0, R0, R2,LSL#16 |
| 22:43.52 | lkcl | r0 |= (r2 >> 16) |
| 22:43.53 | lkcl | ? |
| 22:44.02 | cr2 | yes. |
| 22:44.10 | lkcl | not r2 << 16 ? |
| 22:44.16 | cr2 | left |
| 22:44.19 | cr2 | << |
| 22:44.24 | lkcl | :) |
| 22:53.30 | lkcl | so far: http://hands.com/~lkcl/hp6915/g3.c |
| 22:53.36 | lkcl | gonna take a break. v. intense. |
| 22:53.55 | cr2 | lkcl: the manual has some primitive sample code, i'll try to rip it out. |
| 23:01.41 | *** join/#htc-linux g3gg0_ (n=g3gg0@ppp-82-135-10-33.dynamic.mnet-online.de) |
| 23:05.38 | *** join/#htc-linux Kmarc (i=kari@markos.biz) [NETSPLIT VICTIM] |
| 23:05.50 | *** join/#htc-linux g3gg0__ (n=g3gg0@ppp-82-135-10-33.dynamic.mnet-online.de) [NETSPLIT VICTIM] |
| 23:05.50 | *** join/#htc-linux mreimer (n=mreimer@bilbo.vpop.net) [NETSPLIT VICTIM] |
| 23:05.50 | *** join/#htc-linux TeringTuby (n=maarten@189-66-dsl.ipact.nl) [NETSPLIT VICTIM] |
| 23:07.45 | cr2 | lkcl: great job. |
| 23:08.16 | cr2 | there are some mistakes. R1=0x1000. always. |
| 23:08.21 | cr2 | *((char*)R1 + 0x3E) |
| 23:08.46 | cr2 | this means -- some_delay(); |