| 00:00.05 | cr2 | well :) i will really enjoy to compile haret with arm-wince-pe-gcc or arm-mingwce or how it is called. |
| 00:01.35 | lkcl | oh you're just sick. |
| 00:01.47 | cr2 | :))) lol. |
| 00:02.05 | lkcl | hmmm... i might be able to.... find out more by booting using the telnet thing... |
| 00:02.11 | lkcl | and manually typing in the commands... |
| 00:02.12 | lkcl | hmmm. |
| 00:02.54 | cr2 | afaik there is a media player for wince to be compiled with migwce. videolan ? |
| 00:03.15 | cr2 | i'm not the only one who is sick :) |
| 00:03.18 | lkcl | good grief. |
| 00:03.43 | lkcl | am just going to try something a bit mad... |
| 00:03.59 | cr2 | i did an interactive haret boot |
| 00:04.09 | cr2 | but the debug info is minimal |
| 00:04.30 | lkcl | better than nothing, and i have the source, so can add more |
| 00:04.55 | cr2 | ok. |
| 00:05.24 | cr2 | debuggin the asm part still will require working IR ;) |
| 00:06.31 | cr2 | take the hx4700 haret source, it is tailored to pxa27x a bit. |
| 00:06.43 | lkcl | waah! disconnected! |
| 00:07.48 | cr2 | how do you telnet ? usb is not working on linux, only wifi. |
| 00:07.53 | lkcl | boot.... |
| 00:07.55 | lkcl | yep - wifi :) |
| 00:08.06 | lkcl | okay, v. funny, i get "goodbye cruel world..." |
| 00:08.42 | cr2 | the most interesting things happen after that line. |
| 00:08.49 | lkcl | reset AC97 with a write to 0x4050000C???? |
| 00:08.52 | lkcl | bugger _that_! |
| 00:09.11 | lkcl | okay, where's the hx4700 haret source, then? |
| 00:09.12 | cr2 | there is no ac97 here. |
| 00:09.20 | cr2 | hmm. wait. |
| 00:09.47 | cr2 | <PROTECTED> |
| 00:11.01 | cr2 | universal has i2s with the same chipset as hx4700 |
| 00:11.40 | lkcl | ta. |
| 00:13.57 | lkcl | okay... got it. |
| 00:15.24 | lkcl | okay. the AC97 reset is there because it's a PXA27x peripheral. |
| 00:16.12 | cr2 | dma ? |
| 00:16.46 | lkcl | dunno |
| 00:16.48 | lkcl | <PROTECTED> |
| 00:16.59 | lkcl | but someone on the xda-developers.com wiki can't spell. |
| 00:17.48 | cr2 | what have i written ? :) |
| 00:19.04 | lkcl | periferals not peripherals |
| 00:19.25 | cr2 | :)) bad,bad,bad... |
| 00:22.26 | lkcl | this being the C21st, spelling is not optional :) "eeuuw we should drug-down our language because it's common usage" yeahright. nooo, we need to _up_ the standards not down them... |
| 00:24.24 | cr2 | yeah. silly me. |
| 00:25.13 | lkcl | hmmm.... |
| 00:25.25 | lkcl | wine c:\\msvc-arm\\bin\\link.exe -nologo -base:0x10000 -stack:0x10000,0x1000 -entry:WinMainCRTStartup -align:4096 -machine:ARM -subsystem:WindowsCE,4.21 -libpath:c:\\msvc-arm\\lib -out:out/haret.exe out/haret.obj out/haret-res.obj out/output.obj out/irq.obj out/util.obj out/s-util.obj out/memory.obj out/gpio.obj out/uart.obj out/video.obj out/asmstuff.obj out/irqchain.obj out/getsetcp.obj out/script.obj out/network.obj out/cpu.obj out/s-cp |
| 00:25.25 | lkcl | Warning: the specified System directory L"c:\\windows\\system32" is not accessible. |
| 00:25.26 | lkcl | fixme:msvcrt:_setmbcp Unreal codepages (e.g. 1252) not implemented |
| 00:25.26 | lkcl | fixme:msvcrt:MSVCRT_setlocale :Codepage only locale not implemented |
| 00:25.28 | lkcl | <PROTECTED> |
| 00:25.37 | lkcl | spot the deliberate mistake? |
| 00:25.38 | lkcl | help! |
| 00:27.42 | cr2 | remove the -base -stack -align stuff. these are default settings. |
| 00:27.46 | lkcl | damn command-line's too long! |
| 00:28.17 | lkcl | yep. |
| 00:28.19 | lkcl | hurrah. |
| 00:28.32 | lkcl | sd card... |
| 00:29.01 | lkcl | cp.. |
| 00:29.02 | lkcl | umount... |
| 00:30.08 | lkcl | run!!! go onnnnn... |
| 00:30.15 | lkcl | damn it. |
| 00:30.17 | lkcl | hmmm.... |
| 00:30.28 | cr2 | what do you get? |
| 00:30.32 | lkcl | i think i should disable the logo stuff: it's not coming up on the screen. |
| 00:30.51 | lkcl | but i still got "goodbye cruel world" on the debug output last time. |
| 00:30.52 | lkcl | so. |
| 00:31.07 | lkcl | bye bye logo stuff cos i think haret.exe is writing to completely the wrong place. |
| 00:31.15 | lkcl | thereby causing a crash. |
| 00:31.32 | cr2 | i have LDFLAGS = -nologo -entry:WinMainCRTStartup -machine:ARM -subsystem:WindowsCE |
| 00:31.56 | cr2 | wrong place ? |
| 00:34.10 | lkcl | yep that's cool. |
| 00:34.28 | lkcl | i have WindowCE4,21 |
| 00:34.47 | lkcl | and CXXFLAGS.DEF = -DARM -DPXA27X=1 -DUNICODE -DUNDER_CE=0x0421 -D_WIN32_WCE=0x0421 -DNDEBUG |
| 00:35.04 | lkcl | when actually it's wince 5.0 this is for. |
| 00:36.01 | cr2 | doesn't matter. your compiler is for the 4.21 |
| 00:36.32 | lkcl | ack. |
| 00:36.35 | cr2 | i don 't think you run the 5.0 compiler ;) |
| 00:37.44 | cr2 | that exists only as... hmm. you know. beta version. |
| 00:39.16 | lkcl | whoops :) |
| 00:39.52 | lkcl | okay, logo stuff commented out... |
| 00:41.25 | lkcl | load |
| 00:41.35 | lkcl | physical initrd address: a0408000 |
| 00:41.49 | lkcl | Preloader physical/virtual address: a2577000 |
| 00:41.53 | lkcl | hmmm... |
| 00:41.55 | lkcl | whoops :) |
| 00:42.06 | lkcl | oops-a-daisy :) |
| 00:42.37 | lkcl | restart _again_... |
| 00:43.14 | cr2 | silly comment: Preloader physical/virtual address. it is physical addrees here |
| 00:44.58 | lkcl | oops forgot to comment out the UART stuff that's in there. |
| 00:45.24 | lkcl | no good trying to write to an h4700 UART that might not be there :) |
| 00:45.54 | cr2 | it is there. it is IR |
| 00:46.09 | lkcl | well it's going, anyway :) |
| 00:46.18 | cr2 | :) |
| 00:46.42 | lkcl | goodbye cruel world... |
| 00:46.49 | lkcl | better, but no banana... |
| 00:47.41 | cr2 | after cli() you can't printf anymore... |
| 00:48.01 | cr2 | IR would be great, if it will work. |
| 00:48.36 | lkcl | hmmm..... goodbye __try, methinks... |
| 00:49.37 | cr2 | maybe i 'll put UART_puts( "Goodbye\n" ); in an endless loop |
| 00:49.55 | lkcl | good thinking. |
| 00:50.01 | lkcl | worth a shot. |
| 00:50.26 | lkcl | hmmm. i'm sure i have irda on this laptop.. |
| 00:50.34 | cr2 | do you have sir on the notebook ? |
| 00:50.47 | lkcl | sir? |
| 00:50.55 | cr2 | SIR and FIR |
| 00:51.19 | cr2 | SIR is slow IR, it is usually a transmitter on uart. |
| 00:51.38 | lkcl | i have _no_ idea :) |
| 00:52.00 | cr2 | i think HTC fakes FIR on the universal :) |
| 00:52.24 | cr2 | hmm. what serial ports do you have on the notebook ? |
| 00:53.11 | lkcl | none - a USB and i have, somewhere, a pl2303 (usb->serial) |
| 00:53.26 | lkcl | haven't used it for a while so it's in a box in the garage somewhere.... |
| 00:53.48 | cr2 | do you have IR ? |
| 00:54.21 | lkcl | yeh, apparently - at the front. |
| 00:54.29 | lkcl | never tried it, though. |
| 00:54.49 | lkcl | another load... |
| 00:55.39 | cr2 | pH5 told me yesterday that he has FIR only. |
| 00:55.39 | lkcl | reset AC97 OOPS. |
| 00:55.39 | lkcl | that's what stops it... |
| 00:55.39 | lkcl | okay, that's gotta go, too :) |
| 00:55.39 | lkcl | ahh, i love this stuff :) |
| 00:55.39 | cr2 | wtf is ac97, if i2s is used ? |
| 00:55.53 | lkcl | sdcard out... |
| 00:56.12 | lkcl | yeh, precisely: there's a write to 0x4050000 with a value 0xC |
| 00:56.31 | lkcl | which is buggering everything up (it's a pxa27x peripheral, but exactly _which_ peripheral... :) |
| 00:56.54 | lkcl | sdcard in |
| 00:56.55 | cr2 | afaik you can use either ac97 or i2s on pxa. |
| 00:57.19 | lkcl | loading kernel... |
| 00:57.28 | lkcl | about to do kernel mode... |
| 00:57.34 | lkcl | oops! |
| 00:57.46 | lkcl | SetKMode(TRUE) failed. |
| 00:57.52 | lkcl | possibly... |
| 00:58.06 | lkcl | it's very close to the cli() |
| 00:58.06 | cr2 | this ac97 is coming from axim, which actually uses ac97. |
| 00:58.12 | cr2 | check it earlier. |
| 00:58.36 | cr2 | the return value of SetKMode |
| 00:59.26 | lkcl | ack. |
| 00:59.45 | lkcl | btw what's ICMR? |
| 00:59.52 | lkcl | (a pxa parameter) |
| 00:59.54 | lkcl | any clue? |
| 01:00.03 | cr2 | interrupt controller something register |
| 01:00.22 | lkcl | ahh. |
| 01:00.34 | lkcl | ICMR is being set to 0 in the h4700 code. |
| 01:00.56 | cr2 | maybe it _is_ cli() ? |
| 01:01.22 | cr2 | no cli is inline asm |
| 01:03.08 | cr2 | ICMR is mask register. but there is also ICMR2 on pxa27x |
| 01:04.17 | cr2 | 0 masks all interrupts |
| 01:04.30 | lkcl | hmmm.... the handhelds.org haret cvs appears to be more up-to-date than the h4700 stuff. |
| 01:06.26 | cr2 | but no pxa27x suport. it's all 2 year old code actually. "up-to-date" is the samsung s3 support added. |
| 01:07.01 | lkcl | urk. |
| 01:07.46 | cr2 | ICMR2 is for the camera. only 1 bit |
| 01:09.08 | lkcl | okay, the GPIO_MAX stuff is what the h4700 does - extends it. |
| 01:09.22 | lkcl | so it just outputs more GPIO in the "dump" command. |
| 01:09.23 | lkcl | that's all. |
| 01:09.35 | cr2 | ok. |
| 01:11.23 | lkcl | card again... |
| 01:14.00 | lkcl | loading... |
| 01:14.09 | lkcl | about to setup load... URK. |
| 01:14.43 | lkcl | hmmm.... |
| 01:14.47 | lkcl | let's get drastic :) |
| 01:14.54 | cr2 | setkmode ? |
| 01:15.23 | lkcl | probably... |
| 01:15.53 | cr2 | Output() the value |
| 01:16.20 | lkcl | no - SetKMode wasn't it. |
| 01:16.25 | lkcl | lemme just... |
| 01:17.41 | lkcl | ok |
| 01:18.11 | lkcl | load kernel... |
| 01:18.11 | lkcl | KMode(old): 1 |
| 01:18.17 | lkcl | splat. |
| 01:18.31 | cr2 | call it twice |
| 01:19.32 | lkcl | ack. |
| 01:20.14 | lkcl | sleep time soon, dude... |
| 01:20.29 | cr2 | :) |
| 01:20.41 | lkcl | 01:20 GMT... |
| 01:21.07 | cr2 | uart.c: 40100000, but my pxa270 manual says 40700000 ??? |
| 01:22.04 | lkcl | hmmm... |
| 01:22.29 | lkcl | oops. |
| 01:22.48 | cr2 | FUART, it is correct. but we need STÃœART, and hx4700 uses stuart. bug ??? |
| 01:22.57 | lkcl | yep. |
| 01:23.36 | lkcl | my brain is turning to mush. |
| 01:23.42 | lkcl | KMode(oldagain): 1 |
| 01:23.43 | cr2 | 2:20 MET |
| 01:23.54 | lkcl | splat. |
| 01:24.56 | lkcl | hmmm... |
| 01:25.00 | cr2 | TRUE if the previous mode was kernel mode |
| 01:25.37 | cr2 | BOOL SetKMode |
| 01:25.41 | lkcl | oh dear. oh well :) |
| 01:26.15 | lkcl | anything in your manual about pxa270 MMUs? |
| 01:26.41 | lkcl | MMU level 1 section masks? |
| 01:26.54 | cr2 | its generic thing. |
| 01:27.52 | cr2 | where is your preloaderPA ? |
| 01:28.01 | cr2 | at a2* ? |
| 01:28.15 | lkcl | preloaderPA? huh? |
| 01:28.28 | lkcl | sorry - nounderstand. |
| 01:28.55 | lkcl | oh. |
| 01:28.56 | lkcl | right. |
| 01:28.57 | cr2 | it a variable |
| 01:29.02 | lkcl | lemme see - print out ... |
| 01:29.04 | lkcl | oh. yes. |
| 01:29.10 | lkcl | 0xa258e000 |
| 01:29.12 | cr2 | there is also a "mmu"pointer |
| 01:29.17 | lkcl | initrd address is a0408000 |
| 01:29.55 | lkcl | will just check... |
| 01:30.27 | lkcl | boot again... 2secs... |
| 01:30.40 | lkcl | zzz... |
| 01:30.47 | lkcl | wakey wakey silly machine... |
| 01:30.53 | cr2 | hmm. the initrd is at the bottom, and the preloader above 32MB. i don 't uderstand his logic completely... |
| 01:30.58 | lkcl | ah. |
| 01:31.05 | lkcl | *shrug* :) |
| 01:32.27 | lkcl | loading... |
| 01:32.35 | lkcl | mmu address: 00690000 |
| 01:32.41 | lkcl | oh! black screen :) |
| 01:32.50 | lkcl | no - it's blue. |
| 01:33.01 | lkcl | preloader addr a2565000 |
| 01:33.10 | lkcl | physical kernel address: a0008000 |
| 01:33.29 | lkcl | physical initrd address: a0408000 |
| 01:33.44 | cr2 | looks reasonable. as it should be. |
| 01:34.13 | cr2 | which kernel are you using? cvs ? |
| 01:34.15 | lkcl | nice blue screen... |
| 01:34.17 | lkcl | yeh. |
| 01:34.24 | cr2 | :))) |
| 01:34.36 | lkcl | the one that cp15 gave oxi. or the code, anyway. |
| 01:34.49 | lkcl | hmmmm.... |
| 01:35.00 | cr2 | try my cvs. |
| 01:35.26 | lkcl | the un0 cvs tag is what i'm using. |
| 01:35.30 | lkcl | not ba0. |
| 01:35.51 | cr2 | at least i set the correct lccr* values |
| 01:36.09 | cr2 | asic3 setup is not very clean. |
| 01:36.26 | lkcl | in htcuniversal_core.c etc? |
| 01:36.35 | cr2 | un0 has all my patches. |
| 01:36.36 | cr2 | yes. |
| 01:36.55 | lkcl | well... then i might as well try it :) |
| 01:37.01 | cr2 | i think it was htcuniversal.c |
| 01:37.19 | lkcl | yep. |
| 01:38.52 | cr2 | hmm. i want a running IR console... LCD setup can be resolved later. |
| 01:39.32 | lkcl | yep.... |
| 01:40.39 | cr2 | i have set the "real" direction and alt values for asic3. irq is also correct. hmm bus_shift... cp15 has some ifdefs... |
| 01:41.49 | lkcl | okay - tried the other kernel (htcuniversal) |
| 01:42.01 | lkcl | i don't get this. |
| 01:42.08 | cr2 | the lcd power gpios are fcked. so the blue screen is not a surprise. |
| 01:42.10 | cr2 | ?? |
| 01:42.12 | lkcl | how come it worked on oxi's universal, and not this one? |
| 01:42.22 | lkcl | okay. |
| 01:42.24 | cr2 | what ? |
| 01:42.49 | lkcl | oxi got a black screen, linux boot info and even a kernel panic. |
| 01:43.04 | cr2 | un0 has the code i have writtten out of my ,,, |
| 01:43.11 | cr2 | :) |
| 01:43.15 | lkcl | :) |
| 01:44.09 | lkcl | try a boot from thingy... remote... |
| 01:44.10 | lkcl | nope. |
| 01:44.19 | lkcl | kernel a0008000 |
| 01:44.24 | lkcl | preloader a259c000 |
| 01:44.30 | lkcl | initrd a0408000 |
| 01:44.33 | lkcl | mmu 0069000 |
| 01:45.53 | lkcl | what's the DACR? |
| 01:46.30 | cr2 | hmm. pwm ? i ĺl look in the manual |
| 01:47.04 | lkcl | there's some code commented out with that set to 0xffffffff |
| 01:47.51 | lkcl | bootagain... |
| 01:47.56 | lkcl | oops. |
| 01:48.00 | lkcl | okay: i give up. |
| 01:48.31 | cr2 | no DACR on pxa27x. it was a Digital-AnalogControlRegister on strongarm |
| 01:48.52 | lkcl | okay - just to check: i'm compiling -march=armv5te -mtune=xscale? |
| 01:48.56 | lkcl | for the kernel? |
| 01:49.05 | cr2 | ok. i ll go to bed too. at least the uart bug is found. |
| 01:49.06 | cr2 | yes. |
| 01:49.11 | lkcl | ack. |
| 01:49.22 | lkcl | i'll look up uart stuff tomorrow on my laptop |
| 01:49.30 | cr2 | ok. |
| 01:49.34 | lkcl | we'll get there - eventually. |
| 01:49.48 | lkcl | maaany months from now, this will be a distant arse-kicking memory... |
| 01:49.49 | cr2 | yes. it 's a queston of pride. |
| 01:49.57 | lkcl | i have none, so that's okay. |
| 01:50.02 | lkcl | :) |
| 01:50.05 | cr2 | :) |
| 01:50.18 | lkcl | love and life, peace and little blue flowers. |
| 01:50.20 | lkcl | night dude. |
| 01:50.26 | cr2 | good night |
| 01:50.39 | lkcl | oh - and say hi to cp15: he left for berlin yesterday. |
| 08:19.19 | *** join/#htc-blueangel rob_w-cgi (n=txuser@dd1010.kasserver.com) |
| 08:19.19 | *** mode/#htc-blueangel [+o rob_w-cgi] by ChanServ |
| 10:48.46 | *** join/#htc-blueangel MC-JetLi (n=jet@ppp-69-222-183-92.dsl.chcgil.ameritech.net) |
| 11:42.34 | *** join/#htc-blueangel lkcl (n=lkcl@bb-87-82-4-146.ukonline.co.uk) |
| 11:47.19 | lkcl | mornin |
| 11:58.54 | cr2 | good morning lkcl |
| 12:00.48 | lkcl | mornin cr2. |
| 12:00.50 | cr2 | when my universal hangs, it is transmitting IR. i get the 0xffff stream from od -v -x /dev/ttyS1 on my notebook. not sure about the exact stty params. |
| 12:01.06 | lkcl | oo! |
| 12:01.38 | cr2 | have found the CONFIG_DEBUG_PXA_STUART define, but don't know how to use it properly |
| 12:01.54 | lkcl | i on the other hand am not getting anything into my irda connection from wince. |
| 12:02.14 | cr2 | how is you irda configured ? |
| 12:02.20 | lkcl | just thought i'd say hello before getting some breakfast, btw - back in 10mins ... oh, my irda? |
| 12:02.22 | lkcl | let's see. |
| 12:02.32 | lkcl | on ttyS1. |
| 12:02.44 | lkcl | <PROTECTED> |
| 12:02.54 | cr2 | good. have a breakfast now :)( |
| 12:03.39 | lkcl | i tried to connit to connect using the irport module - i get rport_open(), can't get iobase of 0x2f8 |
| 12:04.03 | lkcl | which i don't understand. |
| 12:04.10 | lkcl | so i'm using irtty_sir instead |
| 12:04.11 | cr2 | kill all ir* modules and stop irda. you need raw serial connection |
| 12:04.40 | lkcl | and irport, and irnet? |
| 12:04.46 | cr2 | nothing. |
| 12:05.06 | cr2 | universal transmits raw without any protocols |
| 12:05.09 | lkcl | ack. |
| 12:05.22 | cr2 | cu -l /dev/ttyS1 -s 115200 |
| 12:05.46 | lkcl | what debian package would have cu? |
| 12:05.51 | lkcl | or can i use minicom? |
| 12:05.51 | cr2 | uucp |
| 12:05.54 | cr2 | yes. |
| 12:06.04 | cr2 | but set it to 115200 |
| 12:06.36 | lkcl | ack |
| 12:06.52 | cr2 | first you should get the Goodbye message from haret. |
| 12:07.11 | lkcl | just trying to beam a contact.... |
| 12:07.18 | lkcl | to see if i get anything... |
| 12:07.43 | lkcl | nope |
| 12:07.47 | cr2 | then stty speed 115200 and use od |
| 12:07.56 | lkcl | ack |
| 12:07.58 | cr2 | of hexdump |
| 12:08.57 | cr2 | it worked here |
| 12:10.29 | lkcl | kIrDAd apparently is using /dev/ttyS1 |
| 12:10.42 | lkcl | hmmm.... |
| 12:10.57 | cr2 | then you have ir modules loaded lsmod | grep ir |
| 12:11.06 | lkcl | yep. |
| 12:11.14 | lkcl | bye bye :) |
| 12:11.36 | lkcl | _still_ in use... |
| 12:11.45 | lkcl | darn. |
| 12:11.52 | lkcl | okay, reboot time |
| 12:12.09 | cr2 | permissions of /dev/ttyS1 ? |
| 12:12.24 | lkcl | running as root - root dialout crw-rw----- |
| 12:12.31 | lkcl | back in a mo - need food anyway. |
| 12:12.44 | cr2 | cu wants you to be in the uucp group |
| 12:54.54 | rob_w-cgi | hey cr2 could you provide me with your subversion checkout of bitbake ... to grab as tar somewhere ?? |
| 13:00.08 | rob_w-cgi | i dont have it here in my office and our firewall is blocking subversion or anything else then http or ftp |
| 13:01.36 | cr2 | do you have a email ? |
| 13:02.10 | cr2 | it"s 215k |
| 13:04.51 | *** join/#htc-blueangel lkcl (n=lkcl@bb-87-82-4-146.ukonline.co.uk) |
| 13:05.23 | lkcl | hello again |
| 13:07.02 | cr2 | hi. |
| 13:07.18 | rob_w-cgi | robert@linuxdevelopment.de |
| 13:09.23 | cr2 | sent |
| 13:09.57 | rob_w-cgi | thanks so much |
| 13:10.16 | lkcl | ok i managed to get my nokia 6610i to talk to the universal. |
| 13:10.45 | lkcl | that makes my thingy - acer - the failure point. oops. |
| 13:11.49 | cr2 | hmm. does the linux irdadump see your universal ? |
| 13:12.32 | lkcl | yes. |
| 13:12.37 | cr2 | what acer is it ? |
| 13:12.50 | lkcl | travelmate c112. |
| 13:13.02 | lkcl | badged as a c110 but it's actually a c112. |
| 13:13.20 | lkcl | yes - i have the installation report listed on tuxmobil.org :) |
| 13:14.19 | cr2 | i had problems with aspire 1501, but it was amd64. it had both SIR and FIR. with some tricks SIR was working. |
| 13:15.23 | lkcl | okay. |
| 13:16.09 | lkcl | modprobe nsc-ircc: |
| 13:16.10 | cr2 | ok. |
| 13:16.11 | lkcl | nsc-ircc, Found chip at base=0x02e |
| 13:16.12 | lkcl | nsc-ircc, driver loaded (Dag Brattli) |
| 13:16.12 | lkcl | nsc_ircc_open(), can't get iobase of 0x2f8 |
| 13:16.50 | lkcl | that's a request_region() call which is failing in nsc_ircc.c |
| 13:17.32 | lkcl | hmmm... debian kernels have serial built-in.. |
| 13:17.49 | cr2 | aspire had nsc also. on LPC bus which is a hack on amd64 |
| 13:18.05 | lkcl | interesting that it actually detects it. |
| 13:18.16 | lkcl | i wonder if i can... |
| 13:18.30 | lkcl | i wonder if i can move it to 0x3e8 (ttyS2 or 3) and then try again. |
| 13:18.33 | lkcl | back in a mo :) |
| 13:19.04 | cr2 | it"a a software problem, you can write to the address range anyway. |
| 13:19.44 | cr2 | can you enable/disable/remap the port in BIOS ? |
| 13:26.03 | cr2 | tablet pc ??? 1 x Fast infrared port |
| 13:26.24 | cr2 | sez acer |
| 13:27.57 | cr2 | lol. Acer PCs use genuine Microsoft® Windows®. |
| 13:28.07 | *** join/#htc-blueangel lkcl (n=lkcl@bb-87-82-4-146.ukonline.co.uk) |
| 13:29.20 | lkcl | heya again. |
| 13:29.42 | lkcl | okay - debian linux kernels have serial support built-in to the kernel instead of as a module. |
| 13:29.58 | cr2 | BIOS ? |
| 13:30.04 | lkcl | so that's what's taking up the 0x2f8 resources so the nsc-ircc can't get at it. |
| 13:30.21 | lkcl | i have 4 choices in the bios: 0x2f8 0x2fe 0x3f8 0x3fe |
| 13:30.49 | lkcl | all of which are auto-detected by the thingy. kernel. at boot-time. |
| 13:30.50 | cr2 | do you have a normal serial ? |
| 13:30.58 | lkcl | yeh. |
| 13:31.26 | lkcl | i wonder: would it be sufficient to "confuse" things by setting ttyS0-3 at some ridiculous non-existant serial range? :) |
| 13:31.27 | cr2 | hmm. can one disable the serial with setserial ? |
| 13:32.08 | cr2 | afaik it was possible to reassign the irq and base |
| 13:32.13 | lkcl | yes. |
| 13:33.29 | lkcl | well - make zImage on its way. |
| 13:33.40 | cr2 | ok. nsc can both SIR and FIR. |
| 13:33.40 | lkcl | with the serial support as modules. |
| 13:33.47 | lkcl | let's see what happens. |
| 13:34.29 | lkcl | this will be really funny if it works. debugging linux over an IRDA port. |
| 13:34.55 | lkcl | cr2: you might want to see if you can recompile the kernel without a framebuffer built in |
| 13:35.01 | cr2 | :)) it will be funny if we can teach haret to use serial console :) |
| 13:35.21 | lkcl | and to specify console=/dev/ttyS0 on the kernel boot params |
| 13:35.23 | lkcl | yeh :) |
| 13:35.49 | cr2 | console=/dev/ttyS2 |
| 13:35.50 | lkcl | well that h4700 haret already had some debug info sent through the thing. |
| 13:35.54 | lkcl | yeh. that's the one. |
| 13:36.23 | rob_w-cgi | crap installing oe on our NAS box doenst work .. tststs |
| 13:36.24 | cr2 | they sent it to /dev/ttyS0, so it was not working. |
| 13:36.37 | lkcl | will be back in a bit - just letting this kernel compile... |
| 13:37.10 | lkcl | heya rob. back later. oh - cp15 should be along to berlin about now. |
| 13:37.27 | cr2 | rob_w: are you a hacker ? :) |
| 13:38.25 | rob_w-cgi | well i need targets to work for and having oe on a DUAL CPU NAS box would be nice |
| 13:38.45 | cr2 | ok. |
| 13:39.31 | rob_w-cgi | but that nas box as a crappy "probiatory" linux os running ,, where i found it compatible with some slackware binary`s .. so now i am stuck in building boost for monotone |
| 13:40.08 | cr2 | suse had the rpm |
| 13:40.19 | rob_w-cgi | hmm |
| 13:40.26 | rob_w-cgi | did it ? |
| 13:41.05 | cr2 | yes. compiling on amd64 is always a minefield. dumb 32bit programmers. |
| 13:42.31 | cr2 | but it is better now, not like in the old alpha days. |
| 13:46.12 | cr2 | we are spamming the gsm with debug output using hx4700 haret... |
| 13:46.41 | rob_w-cgi | hehe |
| 13:51.47 | cr2 | lkcl: compiling without framebuffer. |
| 13:52.38 | cr2 | don't like such warnings: arch/arm/mach-pxa/generic.c:256: warning: "struct i2c_pxa_platform_data" declared inside parameter list |
| 13:53.25 | rob_w-cgi | crap seems my includes screwed up .. it messes on including linux/limits.h |
| 13:54.15 | cr2 | rob_w: what's wrong with including limits.h ? |
| 13:56.37 | rob_w-cgi | it says on building boost : no include path in which to find limits.h |
| 13:57.03 | rob_w-cgi | i copyied limits.h out of the gcc-lib dirs into /usr/include/ and /usr/include/linux |
| 14:00.23 | cr2 | do you have suse ? 9.3 or 10.0 ? i can send you the rpms |
| 14:01.55 | rob_w-cgi | i dont use suse on that box .. as i said that is a network attached storage server running some fance linux( u dont wanna know-linux) and there i want to bitbake |
| 14:02.21 | cr2 | ok. kernel is compiled without the framebuffer. |
| 14:02.46 | cr2 | ok. |
| 14:04.16 | rob_w-cgi | ok now i got further ... |
| 14:04.25 | *** join/#htc-blueangel Frantique (n=attila@dsl51B79749.pool.t-online.hu) |
| 14:04.37 | Frantique | hello all |
| 14:05.06 | cr2 | hi |
| 14:05.34 | Frantique | is anyone who can help me with my BlueAngel? |
| 14:07.15 | cr2 | for very specific things ask rob_w |
| 14:08.44 | Frantique | ok... I have downloaded the files from http://www.handhelds.org/moin/moin.cgi/BlueAngel.. I would like to switch to Linux through... and.. I have no ideea what to do with the downloaded files... |
| 14:09.01 | Frantique | it says: |
| 14:09.03 | Frantique | STEP 2. Format SD-Card |
| 14:09.03 | Frantique | Become root and create 2 partitions on the SD-Card |
| 14:09.15 | cr2 | do you have a cardreader ? |
| 14:09.15 | Frantique | how to do that? |
| 14:09.24 | Frantique | I have not... |
| 14:09.28 | rob_w-cgi | ok how do i set a -I in a export so all gcc calls will include those dirs ? |
| 14:09.31 | Frantique | I have just my BA... |
| 14:10.18 | cr2 | rob_w: edit the Makefile of configure |
| 14:10.34 | Frantique | now, actually I can see my MDA from my Linux desktop. |
| 14:11.06 | cr2 | Frantique: you can't install linux without a cardreader now. |
| 14:11.19 | Frantique | oh, I understand... |
| 14:11.22 | Frantique | :( |
| 14:11.33 | cr2 | if rob_w will create a SD image file, then it will be possible. |
| 14:12.02 | Frantique | rob_w, can you do that? |
| 14:12.34 | rob_w-cgi | hmm |
| 14:13.17 | cr2 | Frantique: what is the size of your sd card ? |
| 14:13.30 | Frantique | 250 Megs |
| 14:14.11 | cr2 | hmm. |
| 14:14.52 | Frantique | nore specific 256 Mb, mini SD.. |
| 14:14.55 | cr2 | actually it is not a very bright idea to download a 256MB image... |
| 14:15.42 | Frantique | is no other way to do it? |
| 14:15.49 | cr2 | is it possible to fdisk a loop device ? |
| 14:16.35 | Frantique | well... the man is not telling this straight... |
| 14:16.36 | cr2 | should be possible, but i have never tried :) |
| 14:16.40 | Frantique | I have never tried.. |
| 14:16.50 | Frantique | :D |
| 14:17.09 | cr2 | create a 256MB file with dd from /dev/zero and try :) |
| 14:17.26 | cr2 | can be 1MB for testing purposes |
| 14:17.59 | lkcl | Frantique : you will definitely need to buy an SD cardreader - they cost £6 if you order them online, or from a local computer store they will be typically £15 to £20. |
| 14:18.29 | cr2 | lkcl: it is not user friendly. |
| 14:18.43 | lkcl | unless anyone knows of a wince program that can write direct to thingy. SD-cards. |
| 14:18.59 | Frantique | aha.. actually I have some firends who can help me with that... I think that I will try with them... |
| 14:19.01 | cr2 | forget the loop. you mount over loop. fdisk the file. |
| 14:19.28 | cr2 | lkcl: psdwrite in the itsutils |
| 14:19.46 | cr2 | there is also pdocwrite |
| 14:19.53 | lkcl | you really _do_ need to follow the instructions, exactly on the wiki - setting up a fat partition and an ext3. oh! cr2 - there _is_ an sdcard writer program? excellent ;) |
| 14:19.59 | lkcl | for wince, that is |
| 14:20.12 | Frantique | ok, guys, thank you for your help.... |
| 14:20.25 | Frantique | :) |
| 14:20.29 | cr2 | lkcl: they work over activesync |
| 14:20.34 | lkcl | ah _ha_. |
| 14:21.08 | cr2 | when llange appears again we can ask him about synce version :) |
| 14:21.10 | lkcl | ok - just to say i'm off out for a few hours - back later. gotta go get some acupuncture (whoo-hoo) |
| 14:21.35 | cr2 | good luck. |
| 14:22.09 | Frantique | bye all... and happy new year... |
| 14:22.20 | cr2 | thanks. bye. |
| 14:22.25 | *** part/#htc-blueangel Frantique (n=attila@dsl51B79749.pool.t-online.hu) |
| 14:24.20 | rob_w-cgi | well .. thats unfortunate but right now people do need to have either linux knowledge or at least a linux box available |
| 14:27.26 | cr2 | if we can "cat" the parts together, then it will be possible to install from windows, too. it's just running the console over IR is much more geeky than writing the cab installer. |
| 14:29.42 | cr2 | he had a linux box, but no cardreader. it is normal, i also don 't have my own cardreader, i use old zaurus :) |
| 14:57.24 | rob_w-cgi | ah boost is compiling now |
| 15:24.05 | rob_w-cgi | oh it died again |
| 15:58.44 | *** part/#htc-blueangel rob_w-cgi (n=txuser@dd1010.kasserver.com) |
| 16:21.30 | lkcl | card readers are £6 from ebuyer.com!!! |
| 16:38.07 | *** join/#htc-blueangel lkcl (n=lkcl@bb-87-82-4-146.ukonline.co.uk) |
| 16:43.39 | lkcl | hurrah. set up serial_core as a module. apparently i shouldn't have needed to do that: setserial /dev/ttyS1 uart none irq 0 port 0 also does the job, oh well. |
| 16:43.58 | lkcl | http://tuxmobil.org/Infrared-HOWTO/infrared-howto-s-configuration.html |
| 16:47.20 | lkcl | so if i run irattach irda0 none i can get wince to beam contacts across. |
| 16:47.52 | lkcl | but i don't know how to just use hexdump or cu because there's nothing to attach _to_ |
| 16:48.22 | lkcl | if i modprobe ircomm_tty then /dev/ircomm0-31 are there, but nothing is seen on a cu with the beam thing (irattach _not_ running)... |
| 16:48.34 | lkcl | hmmm i might try removing nsc-ircc and using irport instead |
| 16:50.31 | cr2 | remove all ir* modules and talk to /dev/ttyS1 |
| 16:52.24 | lkcl | there isn't a /dev/ttyS1 :) |
| 16:52.59 | lkcl | i'm using udev. it got removed when i did rmmod 8250; rmmod serial_core |
| 16:53.46 | cr2 | hmm. what does bios actually do ? |
| 16:54.39 | lkcl | bios allows you to put the FIR onto 0x2f8,0x3f8 etc. |
| 16:54.54 | cr2 | it says FIR ? |
| 16:55.54 | lkcl | nsc-ircc, Found dongle: Sharp RY5HD01 |
| 16:55.54 | lkcl | yes. |
| 16:55.54 | lkcl | it's all quite odd, really. |
| 16:55.54 | cr2 | damn consumers. |
| 16:55.58 | lkcl | *snort*. |
| 16:56.29 | cr2 | i have looked in the source, there are control registers named after SIR and FIR. |
| 16:56.34 | lkcl | oddly enough, talking to wince (beaming) is now fine. irdadump happily shows conversations going on. |
| 16:58.12 | cr2 | hmm. is the stuart used on universal (gpio46,47) ? |
| 16:58.25 | lkcl | clueless. |
| 16:58.30 | cr2 | wgpio |
| 16:59.40 | lkcl | nsc-ircc is registered as a "net device". |
| 16:59.46 | lkcl | register_netdev |
| 17:00.20 | cr2 | we need to find out where nsc reading its data. |
| 17:00.35 | cr2 | irda on wince has also a ndis interface. |
| 17:00.44 | lkcl | surprise. |
| 17:02.05 | cr2 | if it is stuart on universal in your case (in mine it is) then we have raw serial access on one side. should be looking for the data source on the notebook side |
| 17:04.41 | cr2 | self->qos.baud_rate.bits = IR_9600|IR_19200|IR_38400|IR_57600|IR_115200|IR_576000|IR_1152000 |(IR_4000000 << 8) |
| 17:04.57 | cr2 | the last one is real FIR. |
| 17:05.14 | cr2 | the universal is set to use IR_115200 |
| 17:05.42 | lkcl | ack. |
| 17:06.46 | lkcl | hmmm... setserial -a /dev/ircomm0 not doing much... |
| 17:07.47 | cr2 | hmm. ircomm is on top. how do we get to the raw data... |
| 17:08.27 | lkcl | ifconfig irda0 .... :) |
| 17:08.33 | lkcl | ARGH! :) |
| 17:09.07 | lkcl | okay - i going to make some tea. |
| 17:11.25 | cr2 | nsc_ircc_change_dongle_speed for Sharp RY5HD01 is not doing anything |
| 17:15.50 | cr2 | hehe. learning so much about lowlevel irda. |
| 17:19.24 | cr2 | the main question: can we pick out the raw transmitted data from nsc* driver of do we need to trim it to a simple serial driver, like cp15 did with DPRAM. |
| 17:21.36 | cr2 | maybe you should enable the IRDA_DEBUG and look at the output. |
| 17:29.50 | cr2 | static void nsc_ircc_pio_receive(struct nsc_ircc_cb *self) |
| 17:29.53 | cr2 | <PROTECTED> |
| 17:30.06 | cr2 | <PROTECTED> |
| 17:30.22 | cr2 | <PROTECTED> |
| 17:30.52 | cr2 | looks like normal serial port. |
| 17:33.06 | cr2 | <PROTECTED> |
| 17:33.53 | cr2 | catn 't you just attach the standard serial driver at the address and irq set in bios ? |
| 17:47.21 | cr2 | lkcl: i'll be back in an hour. |
| 17:58.41 | *** join/#htc-blueangel rob_w (n=bob@p85.212.193.215.tisdip.tiscali.de) |
| 17:58.41 | *** mode/#htc-blueangel [+o rob_w] by ChanServ |
| 18:16.25 | *** join/#htc-blueangel apt (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 18:16.25 | *** topic/#htc-blueangel is HTC BlueAngel is still under development ! Feel free to ask and check also -> http://www.handhelds.org/moin/moin.cgi/BlueAngel | http://wiki.xda-developers.com/index.php?pagename=BlueangelResearch | http://gnulinux.biz/files/blueangel/sd/linux/ channel chat logs at http://ibot.rikers.org/%23htc-blueangel/ **upcoming gpe release is in org.openembedded.oz354fam083 dev tree !!!!*** |
| 18:19.07 | *** join/#htc-blueangel apt (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 18:19.07 | *** topic/#htc-blueangel is HTC BlueAngel is still under development ! Feel free to ask and check also -> http://www.handhelds.org/moin/moin.cgi/BlueAngel | http://wiki.xda-developers.com/index.php?pagename=BlueangelResearch | http://gnulinux.biz/files/blueangel/sd/linux/ channel chat logs at http://ibot.rikers.org/%23htc-blueangel/ **upcoming gpe release is in org.openembedded.oz354fam083 dev tree !!!!*** |
| 18:43.43 | *** join/#htc-blueangel pH5 (n=ph5@p5485DEA2.dip.t-dialin.net) |
| 18:43.52 | pH5 | hi |
| 18:45.51 | rob_w | welcome pH5 |
| 18:46.25 | rob_w | u are one which is mentioned on openembedded.org |
| 18:47.03 | pH5 | hi rob_w |
| 18:47.39 | pH5 | I am mentioned on oe.org? |
| 18:49.54 | rob_w | hmm |
| 18:50.01 | rob_w | i dont find it anymore ?!? |
| 18:53.10 | rob_w | well on some site , lets say handhelds related, i found your nickname :-) |
| 18:56.31 | pH5 | heh. that's quite possible. :) |
| 18:59.54 | pH5 | cr2: I tried to copy the code from hx4700.c, but I don't get the magician to talk IR :-/ |
| 19:00.18 | pH5 | cr2: the gpio that controls the ir transceiver seems to be #83 here |
| 19:09.42 | *** join/#htc-blueangel JuNuKN (n=JuNuKN@Md7a8.m.pppool.de) |
| 19:09.59 | cr2 | hi pH5 |
| 19:10.31 | pH5 | hi cr2 |
| 19:10.32 | cr2 | we have found yesterday that you need to change the UART address in uart.c |
| 19:10.57 | pH5 | yay. did you already succeed with the universal? |
| 19:11.21 | cr2 | no. i was pushing lkcl :) |
| 19:11.44 | pH5 | :) |
| 19:11.49 | cr2 | it's more fun to do something together |
| 19:11.51 | JuNuKN | cr2: Hello, I read your comment about my com problem. But it was Connection was not set to receive all. Have anybody else tested the serial output of kernel and haret under windows? |
| 19:13.04 | cr2 | JuNuKN: can you test irda on himalaya ? |
| 19:13.08 | JuNuKN | ~seen cp15 |
| 19:13.12 | apt | cp15 <i=tkejci@p549A4246.dip0.t-ipconnect.de> was last seen on IRC in channel #htc-blueangel, 7d 19h 52m 43s ago, saying: 'So bye'. |
| 19:13.16 | pH5 | hi JuNuKN |
| 19:13.29 | JuNuKN | pH5: Hello |
| 19:16.08 | *** join/#htc-blueangel greentux (n=m@195.227.105.180) |
| 19:16.53 | JuNuKN | What are the exact parameters when you receive serial data from Haret and Kernel ... especially for the following parameters: |
| 19:16.53 | rob_w | JuNuKN, that where his words :-( |
| 19:17.39 | JuNuKN | Bits per second / Num Databits / Parity / Num Stop Bits / Flow control |
| 19:17.40 | rob_w | welcome greentux |
| 19:18.51 | cr2 | i don't know what speed llange has hardcoded there. probably 115200/8/none/1/none |
| 19:18.59 | JuNuKN | rob_w: The words till the ccc is over, - I hope and believe! |
| 19:20.07 | JuNuKN | llange reported in the wiki, that the speed must be given by the commandline argumet, but the rest of parameters is undefined! So what are you using under linux ? |
| 19:20.23 | JuNuKN | I am not so familar with linux to test it with that |
| 19:20.34 | cr2 | rob_w: i 'll ask him. a lot of work is not blueangel specific at all |
| 19:20.56 | rob_w | eh |
| 19:21.30 | cr2 | JuNuKN: i don 't have serial port on universal only IR :) haret for hx4700 has 115200 hardcoded |
| 19:22.57 | JuNuKN | cr2: Differneces to your recomandation under Windows is Hardware Flow control "on" ... Perhaps I have to switch this to none also with my next test ... after next backup-. |
| 19:24.10 | cr2 | PortDCB.BaudRate = 115200 in his patch. |
| 19:24.23 | JuNuKN | Options for Flow control under windows are: Hardware / xOn- xOff/ None .... but I think none I never tested :p |
| 19:24.36 | cr2 | do you have a full rs232 cable ? |
| 19:24.38 | *** join/#htc-blueangel MC-JetLi (n=jet@ppp-69-222-183-92.dsl.chcgil.ameritech.net) |
| 19:24.39 | cr2 | I will set none |
| 19:25.02 | JuNuKN | I have my Seriell SyncCable . |
| 19:26.18 | cr2 | xon/xoff was really used only by DEC, for hardware both sides must support rts/cts dtr/dsr, and you these should be connected. so just set none |
| 19:26.40 | JuNuKN | If this is really the reason, - I will Notice this Seriell Parameters in Big Letters into the Wiki (for us Win users, which believe in default parameters ) ;o) |
| 19:27.30 | pH5 | cr2: what is USB PUEN on the universal? |
| 19:27.38 | JuNuKN | Probably I will test in the next hour or so .... thanks so far. |
| 19:28.40 | cr2 | pH5: i don t know. you need to disassemble the udc driver, or at least look at its data section, which gpios are referenced. |
| 19:29.53 | *** join/#htc-blueangel JuNuKN (n=JuNuKN@Md7a8.m.pppool.de) |
| 19:31.05 | *** join/#htc-blueangel JuNuKN (n=JuNuKN@Md7a8.m.pppool.de) |
| 19:34.17 | JuNuKN | . |
| 19:40.33 | *** join/#htc-blueangel JuNuKN (n=JuNuKN@Md7a8.m.pppool.de) |
| 19:44.32 | cr2 | pH5: do you see the Goodbye. message from haret ? |
| 19:48.27 | pH5 | cr2: I see nothing via IR |
| 19:49.31 | cr2 | hmm. i can't compile haret. |
| 19:51.28 | cr2 | ok. let 's start at the beginning. do you see the magician with irdadump ? |
| 19:53.03 | pH5 | http://www.handhelds.org/moin/moin.cgi/HaRET, "Compiling with Wine" worked for me |
| 19:54.26 | cr2 | it worked here too (with older wine versions), but not with 0.9. maybe i will recompile wine with the latest version. |
| 19:54.28 | pH5 | i see it under wince with irdadump and i can copy files. if i put the magician next to my minicom-on-hx4700, i can see the garbage that is caused by irda discovery on the screen |
| 19:54.52 | pH5 | strange. i just used debian sid's wine 0.9 |
| 19:56.35 | cr2 | i'd really like to compile with gcc, but not now :) |
| 19:58.51 | cr2 | ok. it compiles on amd64, but don't want to link. |
| 19:59.20 | cr2 | fatal error LNK1181: cannot open input file '" |
| 20:25.14 | *** join/#htc-blueangel greentux_alt (n=m@195.227.105.180) |
| 21:00.29 | JuNuKN | cr2: Your tip was pure gold !!!! I am logged in the Himalaya via seriell cable ... further instructions ... no output on device itself! |
| 21:03.42 | lkcl | hey, did i miss anything? hey - JuNuKN - you got a himalaya login. excellent! |
| 21:05.16 | lkcl | cr2: about using the standard serial driver to connect to 0x2f8, irq3 - no: bizarrely, it doesn't work. |
| 21:05.16 | JuNuKN | ... Not a big thing for you linux boys ... I downloaded the kernel and Haret from Wiki and now was for the first time able to get the serial output to my Windows Hyperterminal for further tests |
| 21:05.26 | lkcl | cool. |
| 21:05.56 | lkcl | well it is a big deal for me to see, because i worked really hard on that code, nearly 2 years ago, and had to stop. |
| 21:07.16 | JuNuKN | I read ;o) And llange has proceed ... but now no console output on the Device itself only a penguin in the upper left corner ... |
| 21:09.24 | JuNuKN | ... Hmmm ? And now the backlight is switched on automatically !? And someoutput in the seriell PC console about FB_BLANK_UNBLANK 0 FB_BLANK_Normal 1 |
| 21:09.42 | JuNuKN | and others 1 |
| 21:11.16 | JuNuKN | lkcl and If I can believe in the wiki things, llange was also able to use gpe on the device ... |
| 21:11.59 | JuNuKN | Whats now with you!? Why don't your proceed? |
| 21:12.41 | lkcl | que? |
| 21:12.55 | JuNuKN | I read that you still own a Himalaya ,-if I not fail with it |
| 21:12.59 | lkcl | oh - i have to get my himalaya back - i lent it to someone. |
| 21:13.04 | lkcl | i will have it back within a week. |
| 21:13.23 | cr2 | evil haret don't want to link here. |
| 21:13.32 | lkcl | i wasn't going to keep it because i didn't know what to do about getting sound working. |
| 21:13.46 | lkcl | i just need to change some settings in my bios - brb... |
| 21:14.14 | cr2 | there are two separate problems. enable serial debug and make IR working. |
| 21:14.29 | JuNuKN | Great ... By the way Himalaya works (under WinCE) great with the Spectec WLAN-Card over SDIO ,,,, and I am now able to make some assistence tests with Haret for Analysing things like this... |
| 21:18.26 | cr2 | i need IR, because there is no serial cable here. the same on magician... |
| 21:19.26 | cr2 | i need a working compilation of haret. |
| 21:22.47 | cr2 | JuNuKN: are you running getty on the serial port ? compile your kernel without the LCD support |
| 21:24.35 | JuNuKN | I let Hyperterminal run on the serial port of the PC and the Haret-Asic3-Com Version es well as the 2.3.12 kernel out of Wiki from llange, ... it is the first step ;o) |
| 21:25.46 | cr2 | ok. since we are fighting with IR, can you also test IR ? can you compile haret ? |
| 21:26.15 | cr2 | tracing himalaya asic3 is also good. |
| 21:27.14 | JuNuKN | I have never tested to compile haret, because until now I have no need .... but I have windows and can install the MS Compiler if really needed |
| 21:27.33 | JuNuKN | Tracing of Himalaya needs your advices ... |
| 21:28.11 | JuNuKN | I can now access the SD card, while connected via seriell port |
| 21:29.44 | JuNuKN | And this buissy loop doesn't block the device!!!! The Start button is in the upper left corner instead of the down corner :: i THINK THATS THE TRICK |
| 21:30.02 | JuNuKN | Sorry num lock |
| 21:32.11 | cr2 | it eats the resources. llange puts out very verbose strings. i have reduced the output and run into other problems. anyway. look for the gpios that are not known compared to blueangel |
| 21:33.56 | cr2 | trying to link with arm-wince-pe-ld |
| 21:34.29 | cr2 | src/terminal.cpp: undefined reference to `??_EharetTerminal@@UAAPAXI@Z' |
| 21:34.45 | JuNuKN | OK, I have to study and understand the wiki of this part first ... and need to setup a toolchain to put DoC and sd-card driver into ramdisk-image !? |
| 21:36.48 | cr2 | setup == tar xfvz |
| 21:37.22 | *** join/#htc-blueangel MC-JetLi (n=jet@ppp-69-222-183-92.dsl.chcgil.ameritech.net) |
| 21:37.41 | greentux_alt | hi rob_w ... i am usually on the oe channel... |
| 21:38.05 | JuNuKN | Regarding Haret: I think it should be modified to allow a selection via checkbox for the seriel ports. Enabling only this Lines, which are known for the Mtype ,,, |
| 21:38.45 | cr2 | well, wrtite a patch :) |
| 21:38.55 | JuNuKN | cr2: Was it for me? You speak mystics to me ,-linux starter |
| 21:39.03 | rob_w | greentux_alt, nice |
| 21:39.09 | cr2 | wince patch |
| 21:39.19 | rob_w | how do i get syslogd to dump all in a file ?!? |
| 21:39.48 | cr2 | man syslogd |
| 21:47.49 | *** join/#htc-blueangel apt (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 21:47.49 | *** topic/#htc-blueangel is HTC BlueAngel is still under development ! Feel free to ask and check also -> http://www.handhelds.org/moin/moin.cgi/BlueAngel | http://wiki.xda-developers.com/index.php?pagename=BlueangelResearch | http://gnulinux.biz/files/blueangel/sd/linux/ channel chat logs at http://ibot.rikers.org/%23htc-blueangel/ **upcoming gpe release is in org.openembedded.oz354fam083 dev tree !!!!*** |
| 21:48.21 | JuNuKN | Hopw to proceed now? cr2 told me a few days ago, taht the DoC Driver for the internal flash may still work! I want to test it, but don't know how... |
| 21:49.25 | rob_w | it is in the cvs JuNuKN |
| 21:49.51 | cr2 | get the toolchain ; untar ; get the kernel from cvs; make menuconfig; compile; test |
| 21:50.37 | rob_w | i think the kernel on gnulinux.biz already holds that module |
| 21:55.35 | rob_w | NOTE: package gpe-image-1.0: completed |
| 21:55.35 | rob_w | <PROTECTED> |
| 21:55.39 | JuNuKN | cr2 , - ok thats an advice for now ... I will test in the next days. rob_w: OK, it may be in the CVS, but I am not able at the moment to compile the stuff from there, by problems of my Linux -PC. Therefore I hoped, taht somebody else can compile and upload the new stuff ... but I see ,- I need an envirenment! Maybe the menuconfig is se |
| 21:55.39 | JuNuKN | lfexplaining, - we will see |
| 21:56.23 | rob_w | JuNuKN, get the kernel and modules from the gnulinux.biz link in the topic ! |
| 21:56.45 | rob_w | so that initrd holds these compiled modules |
| 21:57.56 | JuNuKN | But the stuff is compiled for Blueangel and may be for different hardware adresses! ? |
| 21:58.28 | rob_w | oh oops you are talking himalaya , right |
| 21:58.32 | rob_w | sorry forgot |
| 21:59.08 | cr2 | i don't think the addresses matter. there are no blueangel ifdefs there afaik. |
| 21:59.46 | cr2 | ah. what is the kernel version of himalaya kernel ? |
| 21:59.55 | rob_w | i think they do , cp15 did not work on himalaya when making the rom modules |
| 22:00.43 | cr2 | it's not the pxa rom. m-sys. BA and universal have a different DoC. |
| 22:02.02 | cr2 | there is no driver for BA/universal DoC (yet). |
| 22:02.09 | JuNuKN | I used 2.6.12 and also the same Version for initrd-2.6.12.gz like described in the wiki http://wiki.xda-developers.com/index.php?pagename=HimalayaLinuxBooting |
| 22:02.34 | cr2 | and what is the gnulinux.biz version ? |
| 22:02.45 | rob_w | that are blueangel builds |
| 22:02.55 | rob_w | on current cvs afaik |
| 22:03.47 | cr2 | we need the modules. well. then he must forceload the modules. it seems easier to recompile the kernel. |
| 22:03.59 | JuNuKN | gnulinux.biz Version is also 2.6.12, but stored in a subdirectory Bluanagel ! |
| 22:04.25 | cr2 | rob_w: why not compile the hima simultaneously with ba ? |
| 22:04.41 | rob_w | hang on ... |
| 22:08.05 | cr2 | i think we want a 100% api and binary compatibility between our 4 htc phones. |
| 22:09.48 | rob_w | binary you are , api .. hmm there is now real one ,, but but ba and him already share most sources or drivers |
| 22:10.52 | cr2 | hima and ba kernel stuff should be better separated. tsc2200 and mtd moved to ~/drivers/* |
| 22:11.29 | cr2 | universal has asic3 and m-systems g3 flash like on BA. |
| 22:12.04 | cr2 | gomunicator and other gsm goodies like fax should work on any phone. |
| 22:14.20 | cr2 | api. i mean device nodes, controls. |
| 22:15.43 | cr2 | well. i want a common gpe image and opie image for all 4 phones. |
| 22:16.22 | rob_w | well that is not necessary .. it is still doable to have tagged images for all the 4 |
| 22:17.09 | rob_w | by the time all is fine inside oe you just switch between the different MACHINE`s |
| 22:17.44 | rob_w | i am testing fix so that the xanadux.bb file will build the MACHINE dependant defconfig |
| 22:17.52 | cr2 | if we forget about wifi, they are equal from the userspace view. |
| 22:18.31 | cr2 | more or less. magician doesn 't have DoC. |
| 22:19.04 | rob_w | pH5, if i would have a little change to a bb file , what is the most wanted format ? diff ? or complete file ? |
| 22:19.58 | pH5 | rob_w: for changes unified diffs are very popular |
| 22:21.37 | rob_w | what diff parameters do they use most likely ? |
| 22:21.38 | pH5 | I always use -Nurd |
| 22:21.38 | rob_w | ok great |
| 22:21.48 | rob_w | i will file a bug and then drop you a note :-) |
| 22:22.06 | pH5 | rob_w: great |
| 22:25.10 | rob_w | hmm i never did make diff to a totallly new file .. can i deliver the complete file then ? |
| 22:26.10 | pH5 | rob_w: of course |
| 22:27.29 | rob_w | hmm himalaya_leds is broken .. |
| 22:27.42 | rob_w | but at least the bb files work .. |
| 22:29.27 | cr2 | comment out the *PHONEL*. it was discussed yesterday. |
| 22:31.28 | JuNuKN | Nobody else than me has a himalaya ? Console and Touchscreen seems also been missing in the kernel and initrd ... |
| 22:32.22 | cr2 | isn't touchscreen the same as in ba ? |
| 22:33.57 | JuNuKN | Someboy interested in my Log-file of the boot process? |
| 22:36.36 | cr2 | is it different from the xda-dev wiki ? |
| 22:37.01 | rob_w | pH5, http://bugs.openembedded.org/show_bug.cgi?id=559 |
| 22:37.11 | JuNuKN | Yes alittle bit it doesnt inform about asic3 stuff .... I didn't understand |
| 22:38.24 | cr2 | arm-wince-pe-ld doesn't really like c++: it breaks here... class haretNetworkTerminal : public haretTerminal undefined reference to `??_EharetNetworkTerminal@@UAAPAXI@Z' |
| 22:41.45 | cr2 | rob_w: why is you blueangel.conf == husky.conf ? |
| 22:41.49 | pH5 | can we unify the ba and himalaya kernel defconfig to build a single kernel for both of them? |
| 22:42.15 | cr2 | pH5: i don 't think so. |
| 22:43.26 | rob_w | pH5, yeah maybe keep the bug behind for now ,.. i am thinking of something |
| 22:44.38 | rob_w | but many gpio`s are different |
| 22:46.03 | rob_w | cr2 it isnt similar to husky !! ? |
| 22:48.19 | cr2 | i get the husky.conf if i click on a "clone of blueangel.conf" link. |
| 22:51.56 | rob_w | cr2 where ? |
| 22:52.11 | cr2 | on your show_bug 559 link |
| 22:53.07 | rob_w | crap |
| 22:55.30 | pH5 | I think that if we want a single image for both phones, we'll need to build a kernel+modules that works on both of them and it would be good to have a single machine configuration file for both, too. |
| 22:56.49 | cr2 | pH5: i gave up. can you put while(1) at UART_puts( "Goodbye\n" ); in linboot.cpp and correct the FUART in uart.cpp to #define FUART 0x40700000 and compile haret ? |
| 22:57.25 | pH5 | cr2: starting from xanadux cvs? |
| 22:57.37 | cr2 | no. the hx4700. |
| 22:57.58 | cr2 | but they are the same. |
| 22:58.44 | pH5 | I wish somebody would port hx4700 changes to hh.org cvs :) |
| 22:58.51 | pH5 | cr2: I'll try |
| 22:59.11 | rob_w | double crap .. now i change the patch twice |
| 22:59.27 | rob_w | someone will understand |
| 22:59.36 | cr2 | pH5: thanks. |
| 22:59.53 | pH5 | :) |
| 23:00.11 | lkcl | JuNuKN : no, touchscreen isn't missing. |
| 23:00.25 | rob_w | pH5, bringing them together is a nice task .. and even worth it |
| 23:00.26 | lkcl | i coded it up a _long_ time ago. himalaya_ts3 |
| 23:00.36 | lkcl | and stuff. |
| 23:00.57 | rob_w | lkcl, it is not loaded inside the initrd yet i guess ;-) |
| 23:00.59 | lkcl | i don't even know if it compiles - this was 2.6.3 |
| 23:02.34 | JuNuKN | lkcl: are you interested in my actual boot-logs of kernel 2.6.12-hh from llange (wiki) ? |
| 23:03.50 | rob_w | i wonder if the dudes at HTC are able to honour this at one point |
| 23:04.03 | cr2 | rob_w: it is called himalaya.conf now, but inside it says Sharp SL-C760 Use MACHINE = "c7x0". is it right ? |
| 23:04.47 | rob_w | ?? |
| 23:04.49 | lkcl | i'm still replying to the email i promised i'd do but yes, point me at a URL. |
| 23:06.03 | rob_w | cr2, that original file is still "the same" i added "comments to that bug " which hold the actual file content .. as i said :" ..double crap" sorry |
| 23:06.11 | lkcl | cr2: i'd suggest you start with the blueangel.conf and cut/paste that. |
| 23:06.37 | cr2 | lkcl: will you open your hima some time to look for the audio chip ? |
| 23:06.44 | lkcl | you will need to create your own .bb file - xanadux-himalaya-2.6.bb |
| 23:06.49 | cr2 | i have your *.conf |
| 23:07.05 | lkcl | cr2: hopefully i won't have to: a query on the i2c bus should give enough clues. |
| 23:07.20 | cr2 | and if it is L3 ? |
| 23:07.24 | lkcl | but yes, if it doesn't - i don't have anything to lose by cracking it open. |
| 23:07.38 | rob_w | lkcl, i submitted a "crappy" bug to let the himalaya.conf use the xanadux-ba one |
| 23:07.59 | rob_w | lkcl, one small step into the goal of re-unite all HTC devices |
| 23:09.21 | pH5 | rob_w: s/all HTC devices/all HTC smartphones/ |
| 23:09.28 | pH5 | ;) |
| 23:09.47 | rob_w | yeah ,, that is a start |
| 23:11.10 | rob_w | well if i get the mood i will with that at the tsc2200 code and friends .. moving that parts one dir up and together better reflects the usage of the tsc2200 ba + hima |
| 23:11.12 | lkcl | interesting thought. i did try to redo the asic3 and gpio functions so that they were functions not macros. |
| 23:11.21 | rob_w | i am sleeping already .. cu |
| 23:11.28 | lkcl | k rob |
| 23:11.39 | cr2 | alpine and hw6515 are the next. if somebody will trace the gpio |
| 23:12.04 | cr2 | lkcl: there is a "unified" asic3 driver |
| 23:12.25 | cr2 | but not asic3_mmc, unfortunately. |
| 23:14.31 | JuNuKN | regarding opening the Himalaya ... there are a lot of hires images of the hardware details http://forum.xda-developers.com/viewtopic.php?t=5002 |
| 23:16.11 | cr2 | there is a description here http://wiki.xda-developers.com/index.php?pagename=HimalayaHardwareOverview but do you see the sound ? |
| 23:17.50 | pH5 | cr2: http://userpage.fu-berlin.de/~zabel/haret/haret-cr2.exe (and .tar.bz2) |
| 23:18.30 | cr2 | there is an unidentified chip near the cradle connector (the second unidentified one is TSC2200) |
| 23:18.47 | cr2 | pH5: i'll try it right now. |
| 23:19.55 | lkcl | infineon hyb25L256160AC-75 SVV31439 0328 |
| 23:20.56 | lkcl | m-systems DoC plus MD3831-D32-V3-X |
| 23:21.17 | cr2 | on another side: 8x4 = 32 bits plcc or how it is called |
| 23:21.54 | lkcl | F C35AB LCXH16245 not sure about the 1 |
| 23:22.55 | cr2 | here at NW from the cradle http://forum.xda-developers.com/files/xda-pictures_024.jpg |
| 23:22.59 | lkcl | HTC 30H damn it's been scribbled on. |
| 23:24.46 | lkcl | AZU 37W Z186 7 2 could be a Z Z could be a 2 |
| 23:25.23 | cr2 | hmm. |
| 23:26.24 | JuNuKN | If someone will write a haret script to read out i2c Info of Soundchips, I can try them ... |
| 23:26.59 | lkcl | HTC AIC2 MAX4949.1 very unclear |
| 23:28.14 | lkcl | 18 2 6401 ADCM2650 17 45 1 - near the camera. |
| 23:28.49 | lkcl | 0327 CCEA 371 |
| 23:29.24 | lkcl | [obscured]4133 I-BM 0332CC EAIF |
| 23:29.29 | cr2 | ADCM is the camera, even with the linux driver |
| 23:29.39 | lkcl | [obscured] 0328FC EACS |
| 23:29.43 | cr2 | AIC2 is also on BA - clock generator |
| 23:30.19 | lkcl | TWL 3014C 38ACKVJH |
| 23:30.55 | cr2 | UDA1341 is SSOT 28 pin - not like what we seen, but UDA1380 is 32pin!!! |
| 23:31.05 | lkcl | D751774GPH 37A8EVW $4 |
| 23:31.58 | cr2 | lkcl: the sound can't be inside the GSM module. it must be there where i have said :) |
| 23:33.01 | cr2 | looks like UDA1380, and it makes sense. the sound gpio config is the same as blueangel, which has UDA1380. |
| 23:33.37 | lkcl | JOJOCO2TLO 5324AB0606 823A113A very obscured - by a large purple "H" |
| 23:33.50 | lkcl | whatwhatwhat? i'm on a roll, reading numbers on these photos. |
| 23:33.59 | lkcl | uda 1380? where,where? |
| 23:34.40 | cr2 | i think here: xda-pictures 024.jpg. see above. |
| 23:35.04 | lkcl | heyyy, i spotted the tsc2200 yaaay the chip that took EIGHT WEEKS to work out cos friggin TI couldn't be bothered to get their application notes right _and_ their teams were really rude, too, when i pointed out the flaws in their app note. |
| 23:35.21 | cr2 | :)) |
| 23:36.12 | lkcl | what - the uda 1380 is the one with the scribble on it??? |
| 23:37.09 | cr2 | http://www.semiconductors.philips.com/pip/UDA1380.html |
| 23:37.22 | lkcl | 123456789 x 4 equals 36 oh actually it's 8 x 4 equals 32 pins. |
| 23:38.29 | lkcl | oh that's a better picture |
| 23:39.25 | lkcl | 1010C0ZTL0 5324A80606 ??23A113J - the one with the purple H |
| 23:40.02 | *** join/#htc-blueangel MC-JetLi (n=jet@ppp-69-222-183-92.dsl.chcgil.ameritech.net) |
| 23:41.20 | cr2 | i think sound is SE from TSC2200, 32pin like UDA1380, with the scribble |
| 23:42.59 | lkcl | it's definitely phillips symbol - 049.jpg - i can make out 380N, perhaps. a 65??? or maybe 55??? |
| 23:43.12 | cr2 | purple H is gsm flash ROM |
| 23:44.14 | lkcl | RF3110 0316 PO32MLX |
| 23:44.33 | cr2 | yes!!! 049 that's it !! it"s UDA1380HN |
| 23:44.51 | cr2 | just recycle the blueangel driver :) |
| 23:45.59 | lkcl | lannd of hoope and glooooooorrreeeeee mooother of the FREEEEEEEE |
| 23:46.07 | lkcl | *bang* |
| 23:46.10 | cr2 | :))))))))))))))) solved. |
| 23:46.50 | JuNuKN | I thing image http://forum.xda-developers.com/files/xda-pictures_049.jpg is the best picture of this chip -.- |
| 23:47.06 | cr2 | yes. |
| 23:47.33 | lkcl | much joy and celebration. |
| 23:47.41 | cr2 | it has some init problems on BA, but works. |
| 23:52.44 | cr2 | i have edited the wiki. |
| 23:54.59 | lkcl | yeh - damn TI didn't mention that SSPI requires exactly the same number of reads as writes. |
| 23:55.16 | lkcl | their diagrams had three writes followed by two reads. |
| 23:55.30 | lkcl | or two writes, two reads, two writes and one read. |
| 23:56.05 | lkcl | which is genuinely stupid. their answer was "well of _course_ you should know that you need to read one more word and throw it away, because SSPI requires you to match the number of reads and writes". |
| 23:56.14 | lkcl | fucking idiots. |
| 23:56.18 | JuNuKN | And I have updated the Handhels Wiki http://handhelds.org/moin/moin.cgi/Himalaya |
| 23:56.30 | cr2 | i wish we had a spi capture card:) |
| 23:57.16 | cr2 | now we need to find out who is controlling the touchpad on universal. |
| 23:58.18 | lkcl | que? |
| 23:58.18 | cr2 | s/touchpad/touchscreen/ |
| 23:58.35 | lkcl | what you mean - who is controlling the touchscreen on universal? universal, universal... oh :) |
| 23:59.05 | cr2 | it is also some spi device. probably from TI :) |
