| 03:43.41 | *** join/#htc-blueangel geers__ (n=DevzMan@c51008F31.inet.catch.no) |
| 05:16.42 | *** join/#htc-blueangel angom_h (n=angom@red-corp-200.79.154.31.telnor.net) |
| 07:01.11 | *** join/#htc-blueangel ChanServ (ChanServ@services.) |
| 07:01.11 | *** mode/#htc-blueangel [+o ChanServ] by irc.freenode.net |
| 07:18.53 | *** join/#htc-blueangel awelux (n=awelux__@dslb-084-058-159-197.pools.arcor-ip.net) |
| 10:35.20 | mndctrl | crap.. I'm stuck at some library including issues... My library consists of a bunch of -fPIC .o files that all have .h files... But the library itself don't, so how do I link to it? |
| 10:35.39 | mndctrl | remember that this is inside the source... I've not got a .o file for the app yet... |
| 10:36.59 | mndctrl | I've tried including with: #include <libname>, #include<libname.h>, #include <libname/objfilename.h>, but I get no such file or dir EVERY TIME:"=ยค)/!"=)(&% argh... |
| 10:37.34 | mndctrl | tried putting the library in the library path, in the dir of the source files, everything, but no... |
| 10:37.50 | mndctrl | do I need a libname.h file? is yes, what to include in it? |
| 11:38.48 | cr2 | ??? |
| 11:40.57 | cr2 | what's wrong with creating an mylib.a or mylib.so from *.o ? |
| 12:24.15 | mndctrl | I don't have a *.o yet... creating it complains about stuff thats inside of my library... :/ |
| 12:25.07 | mndctrl | gcc -c appname.c complains about some incompatible types in assignment, so I guess I need to include something, but what? |
| 12:25.34 | mndctrl | I don't have a headerfile for my library, just for all the files thats in the library... |
| 12:40.57 | *** join/#htc-blueangel ekkobelli (n=michi@p54A3DCF2.dip.t-dialin.net) |
| 12:44.45 | *** part/#htc-blueangel ekkobelli (n=michi@p54A3DCF2.dip.t-dialin.net) |
| 12:49.28 | *** join/#htc-blueangel oppie32 (i=_oppie_@f130056.upc-f.chello.nl) |
| 12:50.11 | oppie32 | Can I ask someone here for help on the bootloader menu? |
| 13:01.27 | cr2 | which types ? are these some structures you have defined in the *.h files ? |
| 13:02.31 | mndctrl | cr2: yes... function that fails: thisStructure = getStructureFromLibrary(); |
| 13:03.42 | mndctrl | the struct is defined in appname.h and libname->objname.h |
| 13:04.09 | mndctrl | objname.h is ofcourse the headerfil of one of the object files I put inside the library... |
| 13:10.10 | mndctrl | I think my problem is that my library has no headerfile...!? So I can't call it...?!? |
| 13:11.18 | mndctrl | I'm confused ;) |
| 13:20.32 | mndctrl | all howto's I read about this stuff either use the same name for their lib as their app, or they compile with -c, my when I compile with -c it fails :( |
| 13:25.43 | cr2 | headerfiles are not a requirement for anything, but if you use a structure the compiler needs a definition, because it will replace the members with offsets. |
| 13:26.36 | mndctrl | cr2: I've defined the structure in appname.c, but I guess it just can't see what getStructureFromLibrary() returns... |
| 13:27.19 | cr2 | then you should cast getStructureFromLibrary() if it was declared as void. |
| 13:27.19 | mndctrl | is there a way to get around this? |
| 13:27.44 | cr2 | i don't know your code, so it's difficult to give a precise advice. |
| 13:28.01 | cr2 | can you post the 10 lines where it breaks ? |
| 13:28.05 | mndctrl | I'll try casting... 2 sec... |
| 13:29.56 | mndctrl | casting sounded like a solid solution, but now I get: error: conversion to non-scalar type requested |
| 13:30.09 | mndctrl | what? thats an error I haven't seen before... ;) |
| 13:30.18 | cr2 | post this line |
| 13:30.50 | mndctrl | thisPhone = (phoneOptions)getPhoneStatus(); |
| 13:31.07 | mndctrl | phoneOptions is the struct.. |
| 13:31.17 | cr2 | how is thisPhone declared ? |
| 13:31.26 | mndctrl | phoneOptions thisPhone; |
| 13:31.47 | cr2 | what is the return value of getPhoneStatus() |
| 13:32.19 | mndctrl | phoneOptions... |
| 13:32.45 | mndctrl | phoneOptions getPhoneStatus(void); |
| 13:33.02 | mndctrl | thats from the .h file of one of the objectfiles in the library... |
| 13:33.52 | cr2 | ok. it is allowed in c, but is not the greatest practice. there is even a warning in gcc :) |
| 13:33.56 | cr2 | put this line |
| 13:34.02 | cr2 | phoneOptions getPhoneStatus(void) |
| 13:34.11 | cr2 | at the beginning of your c file. |
| 13:34.18 | cr2 | with ; at the end. |
| 13:34.56 | mndctrl | hm... ok... I'll give it a go ;) still need the cast? |
| 13:35.09 | cr2 | not anymore. |
| 13:35.36 | cr2 | and a phoneOptions definition before this line. |
| 13:36.15 | cr2 | struct {dsewd eeojcncncencenc; cfec eede; edede eded;} phoneOptions; |
| 13:36.22 | cr2 | phoneOptions getPhoneStatus(void); |
| 13:36.42 | cr2 | phoneOptions thisPhone; |
| 13:36.53 | cr2 | thisPhone = getPhoneStatus(); |
| 13:36.57 | cr2 | should work. |
| 13:36.57 | mndctrl | it compiled with the struct def in the .h file... |
| 13:37.09 | mndctrl | do I need to put it in the .c file? |
| 13:37.12 | cr2 | then include this .h file |
| 13:37.19 | mndctrl | that I've done ;) |
| 13:37.25 | mndctrl | it compiles fine now... |
| 13:37.44 | mndctrl | I'll give it a go for bitbake now ;) tnx... |
| 13:37.45 | cr2 | great :) |
| 13:42.40 | mndctrl | cr2: tnx dude ;) you saved my day there man... hehe... I've been struggeling with thats for hours ;) |
| 14:04.56 | mndctrl | in my local.conf for bitbake I use DISTRO="familiar", I seem to require a DISTRO_VERSION now... do you know what it should be set to? |
| 16:12.05 | *** join/#htc-blueangel jeanseb (n=jeanseb@lns-bzn-47f-62-147-133-94.adsl.proxad.net) |
| 18:28.36 | *** join/#htc-blueangel oppie32 (i=_oppie_@f130056.upc-f.chello.nl) |
| 18:29.05 | oppie32 | anybody here who knows something about the debugger? |
| 18:50.58 | *** join/#htc-blueangel rob_w (n=bob@p213.54.206.219.tisdip.tiscali.de) |
| 18:50.58 | *** mode/#htc-blueangel [+o rob_w] by ChanServ |
| 18:55.35 | oppie32 | I flashed another rom (WM5) from xda-developers to my device but in the debugger screen I needed to change two options to yes. But I couldn't exit with the mail button on my device |
| 19:04.58 | *** join/#htc-blueangel BabelOued (n=BabelOue@lun34-2-82-238-28-28.fbx.proxad.net) |
| 20:34.18 | *** join/#htc-blueangel ramond (n=ramond@i-83-67-203-22.freedom2surf.net) |
| 20:37.00 | ramond | Hi does linux run off the SD card on the blueangel? |
| 20:37.09 | ramond | I dont need to flash it? |
| 20:38.09 | awelux | you don't need to flash it. |
| 20:38.20 | awelux | You can start from ram. SD card. maybe. |
| 20:38.45 | awelux | If you put haret, the kernel and a ramdisk on the sd card it will work too. |
| 20:39.01 | awelux | But booting linux will always be equivalent to a hard reset. |
| 20:39.46 | ramond | thats ok |
| 20:58.00 | *** join/#htc-blueangel immolo (i=mivaqvt@cpc3-cosh1-4-0-cust151.cos2.cable.ntl.com) |
| 20:58.03 | *** join/#htc-blueangel jeanseb (n=jeanseb@lns-bzn-47f-62-147-133-94.adsl.proxad.net) |
| 20:58.20 | cr2 | hi awelux |
| 20:58.26 | awelux | hi |
| 20:58.36 | jeanseb | hi |
| 20:58.41 | awelux | I've got quite a bit further |
| 20:58.54 | cr2 | i have extended the haret asic3 support, and debugged a big chunk of the battery monitor. |
| 20:59.19 | cr2 | so no i2c control, only w1. |
| 20:59.20 | awelux | nice |
| 20:59.43 | awelux | Currently the kernel is trying to create a new page table. |
| 21:00.02 | awelux | It's creating one to switch on the mmu |
| 21:00.28 | awelux | and later another that's compatible to internal data structures. |
| 21:00.54 | cr2 | so why it does not work ? |
| 21:00.54 | awelux | But I can't find where this new page table get's activated. |
| 21:01.23 | awelux | There is a flush_cache_all() |
| 21:01.33 | awelux | And it dies. |
| 21:01.56 | awelux | That indicates that the pagetable got set already. |
| 21:02.12 | awelux | Probably using the previous pagetable location. |
| 21:02.39 | awelux | But I can't find where this is done. |
| 21:02.56 | awelux | I want to add the screen mapping somehow |
| 21:03.13 | cr2 | i have no idea too. |
| 21:04.09 | awelux | hmm. |
| 21:04.23 | awelux | brb. Might work already ;) |
| 21:05.06 | cr2 | try to boot the kernel :) |
| 21:05.22 | awelux | That won't work. |
| 21:05.41 | awelux | There is to much I need to setup before I get output from the kernel. |
| 21:06.30 | awelux | I hate configuring kernels. It's to generic for most uses. |
| 21:08.56 | awelux | Apparently the kernel wants a taglist of useable memory locations on booting. |
| 21:09.17 | awelux | scratch that |
| 21:09.25 | awelux | just half true |
| 21:11.23 | cr2 | i need to eat something |
| 21:12.23 | awelux | I'm hungry too. Cant decide between some wine and asia noodle snacks though. |
| 21:31.19 | *** part/#htc-blueangel ramond (n=ramond@i-83-67-203-22.freedom2surf.net) |
| 21:59.43 | lamikr | awelux: How is your boot, have you been able to fix the mysterious crashes during the boot? |
| 22:00.00 | awelux | yes. |
| 22:00.53 | awelux | That doesn't mean that it boots though |
| 22:00.54 | awelux | turned out I made a (lots of) mistake(s) when I added my debugging code. |
| 22:00.57 | awelux | I wrote a wrong pagetable |
| 22:01.34 | awelux | when I accessed unmapped memory the cpu would crash a few dozen instructions later. |
| 22:03.05 | awelux | I think I would come much further without the debugging code |
| 22:03.16 | awelux | but then I wouldn't know how much further I would get ;) |
| 22:03.54 | awelux | <- happy. |
| 22:05.08 | awelux | I don't know where the c variable base address off the pagetable got set but it's set to the address of the current page table |
| 22:05.12 | lamikr | awelux: I know the feeling, Congratulations :-) |
| 22:05.30 | awelux | hehe thanks. |
| 22:05.57 | awelux | I doubt I will run out of these feelings soon ;) |
| 22:14.44 | awelux | I found the npoint where my FB mapping get's cleared. |
| 22:15.24 | awelux | The good thing: It should be easy to add it there and get a new restart point. |
| 22:16.10 | awelux | So I could try again with an unmodified kernel, add the fb mapping and get feedback. |
| 22:25.33 | awelux | wow |
| 22:26.54 | awelux | I'm past console_init |
| 22:27.26 | awelux | Have to add a few new debug points. |
| 22:27.26 | lamikr | With the unmodified one? Nice. |
| 22:27.44 | awelux | Well Yes and no. |
| 22:28.00 | awelux | Yes. I would get to this point with the unmodified kernel as well |
| 22:28.14 | awelux | No, I'm running a modified one ;) |
| 22:41.08 | awelux | /* Do the rest non-__init'ed, we're now alive */ |
| 22:41.08 | awelux | rest_init(); |
| 22:41.47 | awelux | alive, alive. I always wanted to say that. |
| 22:43.07 | awelux | I'm a bit puzzeled though by a call to static void __init htcuniversal_init( void ) |
| 22:43.25 | awelux | which seam to happen after I reach this point. |
| 22:44.24 | awelux | But when rest_init is reached all __init functions should be invalidated and the memory is available for other processes. |
| 22:57.41 | awelux | looks like one of the compiled modules doesn't work. |
| 22:57.51 | awelux | *compiled in modules |