| 05:24.49 | *** join/#htc-blueangel TelePuzik (n=UNIX@80.81.39.150) |
| 08:46.05 | *** join/#htc-blueangel cp15 (i=mfclum@p549A1008.dip0.t-ipconnect.de) |
| 08:46.05 | *** mode/#htc-blueangel [+o cp15] by ChanServ |
| 08:46.12 | cp15 | Good morning |
| 08:54.29 | *** join/#htc-blueangel rob_w-cgi (n=txuser@dd1010.kasserver.com) |
| 08:54.30 | *** mode/#htc-blueangel [+o rob_w-cgi] by ChanServ |
| 08:56.04 | *** join/#htc-blueangel rob_w-cgi-2 (n=txuser@dd1010.kasserver.com) |
| 08:56.31 | rob_w-cgi-2 | ?? |
| 08:57.29 | *** mode/#htc-blueangel [+o rob_w] by ChanServ |
| 08:57.47 | *** kick/#htc-blueangel [rob_w-cgi!n=txuser@dd1010.kasserver.com] by rob_w (rob_w) |
| 08:57.58 | rob_w | good morning |
| 09:17.46 | cp15 | Good morning rob |
| 09:18.34 | cp15 | rob: What is that for a www/irc gateway you are using? |
| 09:29.35 | rob_w | cp15 i am reading the tsc2200 and slowly see its features a logic behind it .. very interessting area |
| 09:51.56 | *** join/#htc-blueangel cp15_ (n=549a1008@www.it-schaller.de) |
| 09:52.04 | cp15_ | Test |
| 09:52.12 | cp15 | Hi cp15_ |
| 09:52.16 | cp15_ | Hi cp15 |
| 09:52.22 | cp15 | It works |
| 09:52.25 | cp15_ | Yes |
| 13:32.38 | *** join/#htc-blueangel pH5 (n=ph5@p5486665C.dip.t-dialin.net) |
| 13:32.54 | pH5 | hi |
| 13:33.21 | magnet | hi |
| 13:34.45 | pH5 | cp15: I compiled the kernel but I don't even see the first few black pixels. |
| 13:35.38 | pH5 | cp15: haret's last message is "Goodbye cruel world ..." and after the reset all wince data is still in ram. |
| 13:51.22 | cp15 | Ok, this means the kernel did probably not run... Hmm... |
| 13:51.38 | cp15 | Which haret did you use? |
| 13:52.57 | cp15 | And: Did the penguin get red eyes? |
| 13:54.15 | cp15 | And: Are you able to recompile haret? |
| 13:54.40 | *** join/#htc-blueangel bwegh (n=bwegh@p54A3383C.dip0.t-ipconnect.de) |
| 13:55.00 | bwegh | morning all |
| 13:55.31 | cp15 | morning |
| 13:57.42 | pH5 | morning bwegh |
| 13:57.52 | pH5 | cp15: yes, red eyes and all |
| 13:58.17 | pH5 | haret version was 0.3.6 |
| 14:00.46 | cp15 | Are you able to recompile haret? |
| 14:04.13 | pH5 | cp15: i guess i have to. haret is in handhelds.org cvs? |
| 14:06.21 | cp15 | I think you can use the ba haret (the only modification is the sound setup which shouln't do any harm) |
| 14:06.30 | cp15 | Check it out with cvs -z3 -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/xanadux co -ko -r haret-ba haret |
| 14:06.48 | *** join/#htc-blueangel rob_w-cgi (n=txuser@dd1010.kasserver.com) |
| 14:06.48 | *** mode/#htc-blueangel [+o rob_w-cgi] by ChanServ |
| 14:10.16 | cp15 | There are instructions in http://www.handhelds.org/hypermail/haret/current/0057.html how to compile haret under wine |
| 14:15.51 | bwegh | is there a possibility to access the rom/radio-rom/extended-rom/storage areas using linux? |
| 14:17.36 | bwegh | cp15, what would you think about dividing gomunicator into eg 3 parts: Phone-app, sms-app and gsmd, Phone and sms communicate with gsmd using unix-sockets, rob told me about dbus but i think that it should not be used for the phone to gsm communication so the phone-app has to implement the services for dbus, like the sms-app has to do, too |
| 14:18.17 | cp15 | rom: Yes, load blueangel_rom and you have read and, be careful, probably also write access |
| 14:18.57 | cp15 | radio-rom: No, I have to look at the bootloader how it does it, most likely it is entering the bootloader in the gsm radio |
| 14:19.27 | cp15 | extended-rom: Not yet, my first try to use the existing drivers crashed the device |
| 14:20.23 | cp15 | Accessing radio rom is probably not very useful now, I will maybe look into extended rom when the suspend code works |
| 14:20.54 | cp15 | Then it should be possible to install linux on the extended rom |
| 14:21.16 | cp15 | Next step would be replacing windows in the rom and write a simple bootloader |
| 14:22.16 | bwegh | how did you reactivate the device? |
| 14:23.34 | cp15 | Reactivate? |
| 14:25.14 | bwegh | well you crashed the device - did you just reinstall windows2003se on it or what did you do? |
| 14:26.02 | cp15 | I did a hard reset, the touchscreen calibration, the cut & paste exercise and then booted linux again |
| 14:26.22 | bwegh | i see |
| 14:26.52 | bwegh | i will get an original Setup for Qtek9090 ;) |
| 14:27.43 | bwegh | did you read my thought some lines above about gomunicator, cp15 ? |
| 14:28.19 | cp15 | You mean yesterday? Yes, I have seen it in the irc log |
| 14:28.42 | bwegh | what do you think about it? |
| 14:29.18 | cp15 | I would keep gomunicator a single application for now. It makes it much easier to get "off the ground", try out new things and fix bugs |
| 14:29.39 | cp15 | But for later, I think it is a good idea |
| 14:29.48 | bwegh | ok |
| 14:30.24 | cp15 | I would even remove sms completely and integrate it as a special mailbox to an email client |
| 14:30.55 | bwegh | like in windows |
| 14:31.02 | bwegh | yes - i like it that way, too |
| 14:31.32 | bwegh | but then you need a daemon |
| 14:31.42 | cp15 | Unfortunately there is no lean email client yet for gpe |
| 14:32.16 | bwegh | thats bad |
| 14:32.29 | cp15 | I tried sylpheed but it is much too large and not designed for small resultions |
| 14:34.01 | bwegh | sylpheed is cool, but it realy is large |
| 14:34.31 | cp15 | Also we need integration with gpe-contacts, which sylpheed does not provide |
| 14:38.06 | bwegh | isnt there an gpe project for an e-mail client? |
| 14:39.48 | cp15 | Yes, gpe-mail, but I can't find any useable sources. The stuff in the handhelds gpe cvs is very incomplete (only 2 .c files) |
| 14:42.54 | bwegh | ouch |
| 14:43.10 | bwegh | btw: which ide do you use? |
| 14:44.42 | cp15 | ide? |
| 14:44.54 | cp15 | bash |
| 14:45.08 | cp15 | It contains all features I need :-) |
| 14:45.14 | cp15 | And even more |
| 14:45.53 | bwegh | no vi/emacs? |
| 14:46.40 | cp15 | Of course I use a editor: vi |
| 14:53.41 | rob_w-cgi | hehe |
| 15:02.22 | pH5 | cp15: i compiled ba haret and the only difference i see (hear) is a short noise before not booting linux |
| 15:03.53 | cp15 | Ok, of course the haret didn't make a difference, but now we can modify it |
| 15:04.03 | pH5 | oh, and it says "detected cpu family unknown" in the beginning |
| 15:04.24 | cp15 | Hmm... But the magician has a pxa cpu? |
| 15:04.53 | pH5 | pxa272, should be mostly the same as 270 except for internal flash |
| 15:05.38 | pH5 | and haret works on pxa270 (ipaq hx4700) |
| 15:07.43 | cp15 | Ok, Blueangel has a pxa263, but I don't think this makes a big difference |
| 15:11.00 | cp15 | Insert a line * ((long *) (0x88500020)=0xffffffff; |
| 15:11.12 | cp15 | just before linux_start in linboot.cpp |
| 15:11.39 | cp15 | This should give you 2 white pixels in the left upper corner if it gets to this point |
| 15:26.15 | pH5 | cp15: i see a white and a black pixel in the first two positions and then the lcd fades to white. after this, a hard reset is needed. |
| 15:27.49 | cp15 | Does the lcd fade slowly? |
| 15:27.56 | cp15 | > 1 sec? |
| 15:28.01 | pH5 | yes. |
| 15:28.13 | cp15 | Ok, then the power probably got removed |
| 15:28.38 | pH5 | just by writing 0xffffffff to that address? |
| 15:28.44 | cp15 | Not likely |
| 15:30.15 | cp15 | Are you sure it was only one white pixel, and there was a black one next to it? |
| 15:30.37 | pH5 | i am, but i'll try again. just a moment... |
| 15:30.53 | cp15 | That sounds a bit strange |
| 15:31.27 | rob_w-cgi | pH5 : are we talking Blueangel and what kernel do you try to boot , maybe you miss the correct framebuffer driver |
| 15:31.33 | bwegh | rob_w-cgi, i changed the gomunicator side - the link was kind of broken - fixed it |
| 15:31.42 | rob_w-cgi | oh ? |
| 15:31.49 | pH5 | rob_w-cgi: htc magician |
| 15:32.04 | pH5 | cp15: now i see two white pixels. and the display does not fade |
| 15:32.38 | cp15 | Maybe a temporarily glitch... Sometimes haret seems not to work |
| 15:32.42 | pH5 | also, soft reset is possible |
| 15:33.04 | rob_w-cgi | <PROTECTED> |
| 15:33.25 | cp15 | No, its setup is much too complicated for the first tries |
| 15:33.49 | pH5 | the magician doesn't have a dedicated graphics chipset |
| 15:33.54 | bwegh | rob_w-cgi, changed the link to linuxdevelopment to "http://www.linuxdevelopment.org/projects/projects.html" instead of (afair) "http://www.linuxdevelopment.org/projects/" |
| 15:34.09 | rob_w-cgi | right .. .thanks for tha t |
| 15:34.16 | bwegh | you are welcome |
| 15:34.31 | rob_w-cgi | i am loosing touch these days with all these things |
| 15:34.39 | pH5 | cp15: this means.. start_linux is called, but we never reach head.S? |
| 15:34.46 | cp15 | Seems so |
| 15:34.55 | cp15 | Lets check the decompression code |
| 15:35.45 | bwegh | well if you dont mind that i fix things i find i will go on, rob ;) |
| 15:36.27 | rob_w-cgi | heck no , feel free .. i just so my site is also broken to that .. |
| 15:37.09 | bwegh | hehe ... toooo many ideas ..... tooooo less time ;) |
| 15:37.49 | cp15 | I think we need some webspace in the near future to host our images |
| 15:38.13 | bwegh | cp15, i got some |
| 15:38.53 | cp15 | Ok, when immolo appears, contact him please, so he can upload his gpe blueangel image |
| 15:39.56 | cp15 | pH5: Go to arch/arm/boot/compressed/misc.c |
| 15:40.25 | cp15 | There is a function decompress_kernel |
| 15:41.27 | pH5 | ok... what is the vram address here? |
| 15:42.23 | bwegh | *sing* i am the one and only .... *sing* |
| 15:42.44 | cp15 | Add a * ((long *) (0x5c000110)=0xffffffff; before arch_decomp_setup(); |
| 15:43.01 | cp15 | This should give you 2 white pixels in the middle of the first line |
| 15:43.15 | cp15 | Hope you remember the meaning of all the pixels ;-) |
| 15:44.17 | pH5 | cp15: so far, it's manageable :) |
| 15:44.34 | cp15 | Ok, change it to =0x0, it gives a better contrast to the haret image |
| 15:50.15 | pH5 | compiling... |
| 15:52.58 | pH5 | "tshak" no black pixels at 0x5c000110 :( |
| 15:55.06 | cp15 | Then we have to go deeper into haret |
| 15:55.28 | pH5 | i'm most curious |
| 15:58.01 | cp15 | Ok, first we have to setup the mmu to have a 1:1 mapping for the video memory |
| 15:58.40 | cp15 | Look for mmu [preloaderPa >> 20]... |
| 15:59.20 | pH5 | that's just before our inserted line |
| 15:59.31 | cp15 | and add a mmu [0x5c000000 >> 20] = (0x5c000000 & MMU_L1_SECTION_MASK) | MMU_L1_SECTION | MMU_L1_AP_MASK; |
| 15:59.52 | cp15 | after that |
| 16:00.17 | cp15 | I hope this address range is normally not used |
| 16:00.33 | cp15 | since else the translation will be cached |
| 16:01.53 | cp15 | Then add a * ((long *) (0x5c000078)=0 before the "2 white pixels" line |
| 16:02.55 | cp15 | This should give you 2 black pixels in the first line about 1/4th of the width when the mmu 1:1 mapping worked |
| 16:03.57 | pH5 | cp15: what do you mean by not used? in the pxa270 manual it says the 5c00... range is for the on chip 256k ram |
| 16:04.56 | cp15 | No, the virtual address |
| 16:05.33 | pH5 | ah, ok. you can see how this is all new for me :) |
| 16:05.40 | cp15 | With activated mmu you have to differ between physical address (that address the hardware responds to) and the virtual address (the address the program talks to) |
| 16:06.43 | cp15 | With the mmu command above, we map the physical address of the pxa ram to the virtual address, so it doesn't matter if you are using virtual or physical address |
| 16:06.44 | pH5 | so at this point both virtual addresses 0x5c000020 and 0x88500020 point to the same physical address? |
| 16:06.56 | cp15 | Exactly |
| 16:07.34 | cp15 | But there might be a problem if windows used virtual address 0x5c0xxxxxx recently, since the translation gets cached |
| 16:07.39 | cp15 | But we will find out |
| 16:08.19 | cp15 | Next, we go to wince/asmstuff.asm |
| 16:08.42 | pH5 | ok, i have the black pixels and asmstuff.asm is already opened in my editor :) |
| 16:08.58 | cp15 | At line 179 there should be a CPWAIT |
| 16:09.41 | pH5 | nope |
| 16:09.42 | pH5 | 179 is empty here |
| 16:09.59 | pH5 | after saving machine type and tag list address |
| 16:10.52 | cp15 | Oh, sorry, I looked in the wrong version |
| 16:11.25 | cp15 | Line 150 should be correct |
| 16:11.38 | cp15 | Add the following code after line 150 |
| 16:11.57 | cp15 | mov r11, #0x5c000000 |
| 16:12.25 | cp15 | orr r11, r11, #0x5c |
| 16:12.42 | cp15 | str r0, [r11] |
| 16:13.08 | cp15 | This should give you 2 black pixels at 1/8th of the with when turning of mmu worked |
| 16:13.19 | cp15 | Ok, now compile it and give it a try |
| 16:18.16 | *** join/#htc-blueangel immolo (n=immolo@cpc1-cosh1-4-0-cust126.cos2.cable.ntl.com) |
| 16:18.26 | pH5 | hi immolo |
| 16:19.28 | pH5 | cp15: i'm not sure if i see this correctly, but at least one of the pixels is reddish |
| 16:20.15 | pH5 | yup, second pixel is red |
| 16:21.51 | cp15 | Red? |
| 16:22.30 | cp15 | There should be only black and white ones |
| 16:22.40 | pH5 | as last time, i'll repeat |
| 16:22.45 | cp15 | Ok, maybe a mistake. What pixes do you get now? |
| 16:22.53 | cp15 | pixels |
| 16:24.47 | pH5 | the second pixel is bright red |
| 16:25.07 | pH5 | but i see two black pixels in the center now! |
| 16:26.05 | pH5 | (the ones before arch_decomp_setup) |
| 16:26.44 | cp15 | Hmm... I think it is impossible that this pixels appear without the ones at 1/8 and 1/4 |
| 16:27.37 | pH5 | well, i have them |
| 16:27.44 | pH5 | only that the 2nd 1/8 pixel is red |
| 16:28.07 | cp15 | Ok, maybe there is a offset somewhere |
| 16:28.45 | cp15 | But since we get the decompress pixel now, the deactivation of the mmu probably didn't work |
| 16:32.48 | cp15 | BTW: Did you use startup.txt with or without initrd? |
| 16:33.05 | pH5 | w/o |
| 16:33.10 | cp15 | Ok |
| 16:35.58 | *** join/#htc-blueangel immolo (n=immolo@cpc1-cosh1-4-0-cust126.cos2.cable.ntl.com) |
| 16:37.46 | cp15 | Two things: First we disable compress writing something to a uart |
| 16:37.48 | *** join/#htc-blueangel oho (n=phm4@myrtle.ukc.ac.uk) |
| 16:38.12 | oho | Hello |
| 16:38.17 | cp15 | Goto asm-arm/arch-pxa/uncompress.h |
| 16:38.44 | cp15 | Remove the body of the putc function |
| 16:38.52 | cp15 | Hello oho |
| 16:39.36 | cp15 | Then goto decompress_kernel again |
| 16:39.50 | cp15 | After arch_decomp_setup, add the following code: |
| 16:39.52 | cp15 | { |
| 16:39.55 | cp15 | int i; |
| 16:40.25 | oho | cp15: I have a bit more time now after my trip and will look a bit into the gomunicator code |
| 16:40.47 | cp15 | for (i = 0x5c000020 ; i < 0x5c000020+240*2*160 ; i+=4) |
| 16:41.01 | cp15 | *((unsigned long *)i)=0xffffffff; |
| 16:41.02 | cp15 | } |
| 16:41.11 | cp15 | This should make the top half of the screen white |
| 16:41.42 | cp15 | oho: great, it is still a hmm.. bit buggy |
| 16:41.58 | pH5 | compiling... |
| 16:42.17 | oho | cp15: which gpe image and kernel are you using it with ? |
| 16:42.42 | oho | cp15: also are you using the verstion from cvs ? |
| 16:43.50 | cp15 | gpe-image-h3900-20050814160204.rootfs.tar.bz2 with heavy modifications |
| 16:44.30 | cp15 | But there are some instructions on the handhelds wiki that tell you how to install it and what to modify |
| 16:44.38 | oho | cp15: hmmm could you by any chance send me your working image ? so I don' t have to start mucking about ? |
| 16:45.22 | cp15 | No, it is 100 MB big, contains my private phone numbers, pin and a lot of crap |
| 16:45.37 | cp15 | immolo: Is your image ready? |
| 16:47.12 | oho | cp15: fair enought - I will create my own one |
| 16:47.41 | pH5 | cp15: nope, doesn't seem to return from arch_decomp_setup |
| 16:48.03 | cp15 | Hmm... immolos blueangel gpe image is nearly complete |
| 16:48.33 | cp15 | You could use his, but unfortunately he is not here |
| 16:48.37 | oho | cp15: ah ok, prop. wait for that |
| 16:48.50 | cp15 | pH5: I will look |
| 16:50.52 | cp15 | Hmm... Impossible... arch_decomp_setup is a no-op on pxa |
| 16:51.11 | pH5 | argh |
| 16:52.06 | cp15 | Maybe the center pixel is an error. Comment its code in decompress_kernel and see whether the pixel disappears |
| 16:54.00 | cp15 | Yeah, I know, it is cumbersome, but it too took me days to get the kernel running |
| 16:54.56 | immolo | cp15- my daugter has been taking to hosptial so I haven't had time to add the changes |
| 16:56.16 | cp15 | Oops... Something serious? |
| 16:56.52 | immolo | quite, she has my cold which for a two month old baby isn't good |
| 16:57.49 | pH5 | immolo: hope she gets well soon. |
| 16:58.02 | cp15 | Sorry to hear that |
| 16:58.20 | oho | same here |
| 16:58.26 | pH5 | cp15: the black 1/2 pixels are gone now. perhaps the concept of loops or variables was too much |
| 16:59.23 | immolo | well I'll be in hosiptal most of the week but I'll see about some ba work to keep my mind busy when I'm home |
| 16:59.45 | immolo | thanks for the support anyway :P |
| 17:00.13 | cp15 | How about you? Did your cold go away? |
| 17:00.34 | immolo | naw, I've still got it |
| 17:00.54 | immolo | I was suppose to see the doctor today |
| 17:02.02 | cp15 | You really should before it stays forever |
| 17:02.31 | immolo | heh I'll have togo tomorrow now |
| 17:03.36 | cp15 | oho: I will try to update the latest gpe blueangel image and send it to you |
| 17:04.10 | cp15 | pH5: I will look again at the code |
| 17:05.20 | pH5 | I'll be back in 1/2 hour |
| 17:06.21 | oho | cp15: thank you |
| 17:07.57 | cp15 | oho: Do you have a sd card reader for linux? |
| 17:09.14 | oho | sort off, I have a h3900 and a ready partitioned sd card |
| 17:09.58 | cp15 | Ok, have you already put haret, initrd, zImage and startup.txt on the first partition as described in the wiki? |
| 17:10.13 | oho | yes |
| 17:10.37 | cp15 | Have you added boot_mmc=y to startup.txt? |
| 17:11.04 | oho | yes |
| 17:11.30 | cp15 | Ok, than just a few minutes |
| 17:11.42 | oho | ok - thank you |
| 17:15.43 | cp15 | Here it comes |
| 17:16.50 | oho | thank you - I will give that a go right away |
| 17:20.31 | cp15 | There may be some things you have to fix, but you should be abe to login at least |
| 17:31.24 | cp15 | oho: If you encounter any problems, please tell me |
| 17:32.14 | oho | cp15: thank you - just have to sort a few problems with the size of the second partition out |
| 17:35.18 | pH5 | re |
| 17:37.23 | cp15 | pH5: Can you extend decompress_kernel with a few pixels by yourself to see where the problem is? |
| 17:39.26 | pH5 | cp15: i'll try |
| 17:41.21 | cp15 | Ok, if you get stuck, tell me and we will see |
| 17:41.29 | pH5 | cp15: thanks |
| 17:55.45 | oho | cp15: can I put "set CMDLINE "root=/dev/mmcblk0p2 noinitrd cachepolicy=writeback" into the startup text to make it boot from sd card ? or is this what the "boot_mmc=y" is good for |
| 17:59.02 | cp15 | No, you can't |
| 17:59.13 | cp15 | The kernel doesn't contain the sd drivers |
| 17:59.54 | oho | ah ok - have you ever thought about adapting the h2200 kernel which has the drivers compiled in ? |
| 17:59.57 | cp15 | the boot_mmc is a hint for linuxrc to load the sd drivers, mount the second partition do a pivot_root into the filesystem on this partition and execute init |
| 18:00.34 | oho | ah ok |
| 18:00.37 | cp15 | Currently I want as little as possible compiled into the kernel, since this makes development more flexible |
| 18:00.57 | oho | true |
| 18:02.52 | pH5 | cp15: i added some points to decompress_kernel and the last one displayed was right before makecrc() |
| 18:03.15 | pH5 | cp15: but then i tried to add pixels to makecrc in inflate.c and now it doesn't show any kernel pixels at all |
| 18:05.20 | cp15 | Hmm... Have you tried more than one time? |
| 18:05.29 | pH5 | twice |
| 18:06.53 | cp15 | Hmm... What does that mean? |
| 18:07.28 | pH5 | also it was strange that your fillscreen loop didn't work |
| 18:16.27 | cp15 | In makecrc there is also nothing critical |
| 18:16.47 | pH5 | there has to be some error before that |
| 18:17.22 | cp15 | Maybe there is something corrupting the memory |
| 18:25.23 | *** join/#htc-blueangel JuNuKN (n=JuNuKN@K8b47.k.pppool.de) |
| 18:39.45 | cp15 | There may be some active dma |
| 18:41.03 | cp15 | Try to make a delay loop of about 1 second before linux_start |
| 18:43.15 | pH5 | I went back to a clean kernel and added just the decompress_kernel 1/2 pixel, but I don't see it. The second 1/8 pixel is still red. |
| 18:59.01 | cp15 | Have you tried the delay loop? |
| 18:59.27 | pH5 | how do i do it? |
| 18:59.39 | pH5 | this: "time = GetTickCount(); while ((GetTickCount()-time) < 1000) {}" doesn't seem to work |
| 19:00.01 | cp15 | Did it hang? |
| 19:00.16 | pH5 | yes. no pixels |
| 19:00.21 | cp15 | Good |
| 19:00.33 | pH5 | good? |
| 19:00.36 | cp15 | It would be a problem if it worked |
| 19:00.46 | pH5 | why is that? |
| 19:01.13 | pH5 | ah, wince should be dead by this time? |
| 19:01.14 | cp15 | try something like "for (j=0 ; j < 1000 ; j++) { for (i=0 ; i < 10000 ; i++); }" |
| 19:01.32 | cp15 | Because if GetTickCount gets updated, interrupts are still active |
| 19:01.39 | pH5 | ok... |
| 19:09.03 | *** join/#htc-blueangel JuNuKN (n=JuNuKN@K98a0.k.pppool.de) |
| 19:10.46 | pH5 | cp15: you're right! now i see the kernel pixel again. time to add some more... |
| 19:13.40 | cp15 | Not quite sure... But if it helps... |
| 19:17.44 | pH5 | now the last pixel painted is right before gunzip() |
| 19:20.06 | cp15 | Maybe the crc fails and there is still something wrong |
| 19:20.25 | cp15 | Have you already tried using the uncompressed Image instead of zImage? |
| 19:20.46 | pH5 | of course not |
| 19:21.05 | pH5 | but why would that work? |
| 19:21.49 | cp15 | Because the zImage fails even if there are just a few bytes corrupt |
| 19:22.04 | cp15 | Image will get further |
| 19:22.41 | pH5 | but it doesn't even return from gunzip() now, and before adding the delay loop it already hung in makecrc() |
| 19:22.52 | cp15 | I remember some problems with ipaq because of dma |
| 19:23.10 | cp15 | This might also be related to a different layout |
| 19:23.50 | cp15 | Lets try the Image, if that fails, we can continue debugging zImage |
| 19:26.50 | pH5 | just a moment, i had a screen-fading-white on the first try... |
| 19:31.45 | cp15 | oho: Any progress on your side? |
| 19:32.50 | pH5 | and again. |
| 19:33.12 | pH5 | with the uncompressed image i just see haret's signals and then the lcd power goes off |
| 19:33.18 | oho | the h3900 is very slow on writing to SD, but it has finished now. I also just finished re-compiling my kernel, as I didn't have usb network support. I will re-boot and let you know how it is going... see you in a minute.... |
| 19:34.05 | cp15 | Hmmm... We had this error on the zImage one time too... Maybe it is initialization-related |
| 19:36.04 | pH5 | i am clueless |
| 19:36.35 | cp15 | I not yet... edit init/main.c |
| 19:36.40 | cp15 | goto start_kernel |
| 19:36.55 | cp15 | place a for(;;); before the lock_kernel(); |
| 19:38.11 | cp15 | If the initialisation disables the LCD power, it will then stay on and the kernel will hang |
| 19:38.58 | cp15 | BTW: Where are you from? |
| 19:39.20 | pH5 | i'm in berlin,germany |
| 19:39.30 | *** join/#htc-blueangel oho (n=phm4@myrtle.ukc.ac.uk) |
| 19:39.48 | cp15 | Oh, are you at the CCC? |
| 19:41.42 | pH5 | no, i didn't even know that it is this month |
| 19:42.57 | cp15 | It is next month, see https://events.ccc.de/congress/2005/wiki/22C3_Public_Wiki |
| 19:43.12 | cp15 | would have been a good possibility to meet in real life |
| 19:53.40 | *** join/#htc-blueangel blackhat (n=blackhat@c-24-62-188-126.hsd1.ma.comcast.net) |
| 19:55.15 | pH5 | I won't be here over christmas, so I might be too late for this. |
| 19:55.36 | cp15 | Too bad |
| 19:56.23 | pH5 | cp15: where are you from? leaving berlin on 20051230? |
| 19:56.38 | cp15 | Harald Welte will give a talk about linux on motorola. I will probably met him next weekend and convince him to also tell something about linux on ppc devices |
| 19:56.41 | pH5 | ah, btw. your for loop is still running |
| 19:56.47 | pH5 | so no power-off |
| 19:57.06 | cp15 | Do you see the black pixel lines also? |
| 19:57.44 | pH5 | gnn. forgot to add them. |
| 19:57.48 | pH5 | i'll retry |
| 19:57.51 | *** join/#htc-blueangel rob_w (n=bob@p85.212.165.214.tisdip.tiscali.de) |
| 19:57.51 | *** mode/#htc-blueangel [+o rob_w] by ChanServ |
| 19:58.08 | cp15 | Also try to move the for(;;); a few lines down |
| 19:58.30 | cp15 | I will leave berlin 31st december morning |
| 20:02.38 | pH5 | i have to correct myself, the black lines code was in the kernel, but i can't see them. |
| 20:03.22 | cp15 | Hmm... So it maybe crashed before the for(;;); |
| 20:03.40 | cp15 | It gets a bit difficult now |
| 20:04.29 | cp15 | Might also be a error in the "draw the black lines" code |
| 20:05.16 | pH5 | i'm still trying to wrap my head around what is called when. |
| 20:05.32 | cp15 | Try placing the for(;;); before setup_arch, and if the LCD doesn't disappear, place it after setup_arch |
| 20:05.34 | pH5 | the very first thing in the uncompressed image should be the entry point in head.S, right? |
| 20:05.52 | pH5 | ok, i'll do that |
| 20:06.07 | cp15 | I guess the LCD problem occurs somewhere in setup_arch |
| 20:08.42 | cp15 | Yes, ENTRY(stext) in arch/arm/kernel/head.S is called first on uncompressed Images |
| 20:10.02 | *** join/#htc-blueangel etienn1 (n=etienne@ghorg.leberjuquey.net) |
| 20:10.11 | etienn1 | hi |
| 20:10.57 | pH5 | hi |
| 20:12.43 | pH5 | cp15: for(;;); before setup_arch hangs as you expect. moving... |
| 20:12.43 | etienn1 | just to know |
| 20:12.43 | etienn1 | is it possible to use the blue angel as phone underlinux ? |
| 20:13.00 | cp15 | Yes, still a bit alpha maybe, but it works. |
| 20:13.14 | cp15 | You have to invest some time and patience to install it |
| 20:13.36 | etienn1 | no problem to install it |
| 20:14.07 | etienn1 | can i use it or should have abnother phone as spair ;) |
| 20:15.08 | cp15 | Well, I use it since 2 months now and didn't need a spare phone, but your mileage may vary |
| 20:17.21 | etienn1 | great great great |
| 20:20.38 | rob_w | hmm , cp15 i though you had a spare phone ,.. i never was able to erally " use " it in that sense yet |
| 20:20.56 | etienn1 | i have to buy a sd card reader |
| 20:21.17 | cp15 | It is possible without a card reader, but more difficult |
| 20:21.50 | pH5 | cp15: for(;;); after setup_arch and it still hangs, but now i have to do a hard reset |
| 20:22.25 | etienn1 | how it can be possible without reader ? |
| 20:24.25 | cp15 | Download ramdisk, zImage, startup.txt and haret via windows, start haret, setup networking on your host, log in, repartition your sd card, copy the remaining files via scp from your host and unpack the image |
| 20:24.39 | etienn1 | ok |
| 20:24.48 | cp15 | pH5: Try to move it further down until the screen goes white again |
| 20:24.49 | etienn1 | i will buy a reader tomorrow ;) |
| 20:25.01 | cp15 | It will make the life easier |
| 20:25.13 | pH5 | cp15: that will take some time, probably :) |
| 20:25.22 | cp15 | I will be away now watching a movie |
| 20:25.35 | pH5 | have fun |
| 20:47.15 | oho | I got quite far, all the way to the claibration screen, but now it hangs - also I can' t connect via the usb network. but i will go home now. see you tomorrow and thank you very much for your help so far. bye |
| 20:48.39 | rob_w | cu oho |
| 21:12.50 | *** join/#htc-blueangel ferrett1968 (n=ferrett1@adsl-184-145.37-151.net24.it) |
| 21:20.09 | pH5 | the fading happens somewhere in rest_init |
| 21:25.18 | rob_w | somewhatre i read about a fading to white , i think |
| 21:43.53 | pH5 | good night everyone |
| 22:19.33 | *** join/#htc-blueangel _cronos_ (n=cronos@ron34-1-82-224-149-102.fbx.proxad.net) |
| 22:19.47 | _cronos_ | hie all |
| 22:19.49 | _cronos_ | ... |
| 22:20.03 | _cronos_ | i have a probleme on the final step |
| 22:20.52 | _cronos_ | when i reboot ma ba after step 20 on http://www.handhelds.org/moin/moin.cgi/BlueAngel tutorial, I can t connect ton my ba .. |
| 22:20.59 | _cronos_ | do u have a solution plz ? |
| 22:24.05 | immolo | _cronos_- looking at the guide I don't see a problem with it |
| 22:24.11 | immolo | is the ba in the cradle? |
| 22:24.18 | _cronos_ | yep |
| 22:25.07 | immolo | no idea, I haven't used the guide myself only read over it |
| 22:25.19 | _cronos_ | k |
| 22:26.24 | immolo | anyway let's apply cp15 changes to my distro and see what it gives me |
| 22:27.48 | _cronos_ | i have this msg : init is telling me that id "TO" is respawning to fast |
| 22:28.15 | immolo | bah, haven't seen that messages in awhile |
| 22:28.42 | immolo | can't remember how I fixed it |
| 22:28.57 | immolo | ask me something I know :P |
| 22:35.44 | _cronos_ | lol |
| 22:35.55 | *** join/#htc-blueangel magnet_ (n=magnet@vsg94-1-82-234-118-20.fbx.proxad.net) |
| 22:35.59 | _cronos_ | i will trie somthing i come back later |
| 22:36.01 | _cronos_ | :p |
| 23:00.00 | cp15 | re |
| 23:00.16 | immolo | hey cp15 |
| 23:00.29 | cp15 | Hi immolo... Still up? |
| 23:00.39 | immolo | just got back |
| 23:00.49 | *** join/#htc-blueangel Funklord (n=cow@c-cbd572d5.014-46-73746f28.cust.bredbandsbolaget.se) |
| 23:01.08 | immolo | I'm about to test the distro on my ba :D |
| 23:01.20 | cp15 | Yes? Cool |
| 23:01.42 | immolo | I can't get gomicator to build though |
| 23:01.54 | immolo | you got an ipk around just to test it? |
| 23:02.04 | cp15 | What error messages do you get? |
| 23:02.22 | immolo | something about alsa not being set |
| 23:02.25 | cp15 | Yes, but it has some inconvenient dependencies |
| 23:02.43 | immolo | the deps built fine |
| 23:03.00 | cp15 | Maybe the .bb has a problem |
| 23:03.13 | immolo | maybe, I'm using rob's one |
| 23:03.24 | immolo | the one he submitted to oe |
| 23:04.18 | bwegh | n8 |
| 23:04.24 | cp15 | You are right, there is a gomunicator bb |
| 23:04.48 | cp15 | I try to build it |
| 23:04.52 | immolo | cool |
| 23:05.07 | immolo | I'll go for a quick smoke and make a coffee while you build |
| 23:09.26 | immolo | have you seen the bounty for the wireless drivers of the ba? |
| 23:09.38 | cp15 | Yes, but I don't like it |
| 23:11.08 | immolo | not great but at least it will get some interest on getting the card working |
| 23:11.11 | cp15 | I have two nintendo ds, and there is also a bounty for a open source wireless driver, much higher than the ipaq one |
| 23:11.50 | immolo | the bounty says nothing about the drivers being open sourced on the ipaq |
| 23:11.57 | cp15 | The driver is essential complete, but the guy doesn't want to release it since he wants to write his own tcp ip stack |
| 23:12.20 | immolo | for the ds or ba? |
| 23:12.24 | cp15 | For the ds |
| 23:12.57 | cp15 | With more working together than working alone the driver would have been completed much earlier |
| 23:13.22 | immolo | speaking of writing the tcp ip stack, what happened to the rewrite of the kernel's? |
| 23:13.46 | cp15 | But the bounty created several groups and some individuals which were researching on their own on the project |
| 23:14.04 | cp15 | ten times the effort for a single solution |
| 23:14.19 | immolo | I've never looked at it like that |
| 23:14.27 | cp15 | rewrite of the kernels ip stack? |
| 23:14.40 | immolo | yeah the guys at red hat were going to do it |
| 23:14.51 | cp15 | what is wrong with it? |
| 23:15.17 | *** join/#htc-blueangel _cronos_ (n=cronos@ron34-1-82-224-149-102.fbx.proxad.net) |
| 23:15.27 | immolo | can't remember fully but I think he said it was a little dated and could be wrote in less code and run faster |
| 23:15.49 | cp15 | less code and faster is always ok |
| 23:16.05 | _cronos_ | hie |
| 23:16.05 | immolo | yeah seeing as the kernel is over 3 million lines now |
| 23:16.13 | cp15 | Let me check |
| 23:16.22 | _cronos_ | cp15 i have a pb :( |
| 23:16.27 | _cronos_ | for change :) |
| 23:16.53 | _cronos_ | after booting i have : ID to responding too fast |
| 23:17.18 | cp15 | Are you sure it booted from sd card? |
| 23:17.28 | _cronos_ | yesy |
| 23:17.59 | cp15 | Ah, it is probably the serial getty |
| 23:18.09 | _cronos_ | i can t connect in ssh |
| 23:18.17 | cp15 | Hmm... That is indeed a problem |
| 23:19.01 | cp15 | Do you get a message about dropbear when booting? |
| 23:19.22 | _cronos_ | yes |
| 23:19.39 | cp15 | Ah... Sorry, I think I know |
| 23:19.47 | _cronos_ | ok |
| 23:20.20 | cp15 | Did you add usb0 to the auto line in /etc/network/interfaces? |
| 23:22.23 | _cronos_ | yes |
| 23:22.40 | cp15 | Then I don't know :-( |
| 23:23.14 | _cronos_ | lol |
| 23:23.15 | _cronos_ | np |
| 23:23.22 | _cronos_ | i have purapse an idea |
| 23:23.28 | _cronos_ | tk u :) |
| 23:23.49 | cp15 | Any strange error messages you get? |
| 23:24.02 | _cronos_ | no only this one |
| 23:24.09 | _cronos_ | and no ssh bexose no usb0 ... |
| 23:24.52 | cp15 | Do you have a card reader? |
| 23:25.02 | immolo | cp15- did it build? |
| 23:25.25 | _cronos_ | yes |
| 23:25.32 | cp15 | NOTE: package gomunicator-0.1.3+cvs-20051115-r0: task do_package: started |
| 23:25.32 | cp15 | NOTE: Couldn't find shared library provider for libasound.so.2 |
| 23:25.37 | cp15 | But I got a ipkg |
| 23:25.46 | immolo | hmm let me try |
| 23:25.51 | cp15 | cronos: Do you want to test the newest image? |
| 23:26.12 | immolo | I was about to suggest that :P |
| 23:26.42 | _cronos_ | why not :0 |
| 23:26.57 | cp15 | I have modified the things I said you in your image. Not the cleanest way, but for now ok I hope |
| 23:27.10 | _cronos_ | ok great |
| 23:27.25 | cp15 | Ahh... Clicked wrong |
| 23:44.02 | cp15 | cronos? Can you do dcc? |
| 23:44.14 | _cronos_ | normaly yes ... |
| 23:44.28 | cp15 | Seems not to work |
| 23:44.33 | _cronos_ | k |
| 23:44.33 | _cronos_ | ... |
| 23:44.53 | _cronos_ | so |
| 23:45.09 | _cronos_ | i will go to sleep i am tired |
| 23:45.18 | _cronos_ | and tk u for u re help |
| 23:45.26 | *** part/#htc-blueangel _cronos_ (n=cronos@ron34-1-82-224-149-102.fbx.proxad.net) |
| 23:45.58 | immolo | hey cp15 strange question I know but you tried windows vista yet? |
| 23:46.09 | cp15 | Why should I? |
| 23:46.19 | immolo | not really |
| 23:46.29 | cp15 | Have you? |
| 23:46.37 | immolo | a while back |
| 23:47.05 | cp15 | And? |
| 23:47.13 | immolo | it reminded me of xp with osx stuck on top |
| 23:48.13 | immolo | I just need to fix a pc and I can't install linux on this one as I don't have the time to teach |
| 23:49.27 | immolo | but seeings vista is around the corner I'm thinking of loading that on to save issues of going back up north later to update |
| 23:50.06 | immolo | maybe I should disrespect my laptop and give it a quick test |
| 23:51.03 | immolo | ok my error for for gom |
| 23:51.07 | immolo | | checking how to run the C preprocessor... arm-linux-gcc -E |
| 23:51.07 | immolo | | checking for egrep... grep -E |
| 23:51.07 | immolo | | checking for ANSI C header files... yes |
| 23:51.07 | immolo | | checking for pkg-config... /home/immolo/oe/org.openembedded.oz354fam083/tmp/staging/x86_64-linux/bin/pkg-config |
| 23:51.07 | immolo | | checking for gtk+-2.0 >= 2.0.0 esound sqlite libgpewidget dbus-glib-1 alsa... Package alsa was not found in the pkg-config search path. |
| 23:51.10 | immolo | | Perhaps you should add the directory containing `alsa.pc' |
| 23:51.12 | immolo | | to the PKG_CONFIG_PATH environment variable |
| 23:51.13 | cp15 | Ah, install linux and openoffice, nobody will notice the difference ;-) |
| 23:51.16 | immolo | | No package 'alsa' found |
| 23:51.18 | immolo | | |
| 23:51.20 | immolo | | configure: error: Library requirements (gtk+-2.0 >= 2.0.0 esound sqlite libgpewidget dbus-glib-1 alsa) not met; consider adjusting the PKG_CONFIG_PATH environment variable if your libraries are in a nonstandard prefix so pkg-config can find them. |
| 23:51.24 | immolo | | FATAL: oe_runconf failed |
| 23:51.26 | immolo | NOTE: Task failed: /home/immolo/oe/org.openembedded.oz354fam083/tmp/work/gomunicator-0.0cvs20051105-r2/temp/log.do_configure.4746 |
| 23:51.31 | immolo | NOTE: package gomunicator-0.0cvs20051105-r2: task do_configure: failed |
| 23:51.34 | immolo | ERROR: TaskFailed event exception, aborting |
| 23:51.36 | immolo | NOTE: package gomunicator-0.0cvs20051105: failed |
| 23:51.37 | immolo | ERROR: Build of gomunicator failed |
| 23:51.42 | immolo | cp15- you wouldn't beilve the number of computers I have done that to :P |
| 23:52.55 | immolo | I don't know maybe kde and some crossover office could get the computer running nice |
| 23:53.45 | cp15 | What do you have in packages/alsa? |
| 23:55.22 | immolo | alsa-driver alsa-lib_1.0.8.bb alsa-tools_1.0.8.bb files |
| 23:55.22 | immolo | alsa-driver_0.9.6-hh4c.bb alsa-oss_1.0.8.bb alsa-utils_1.0.8.bb |
| 23:56.34 | cp15 | Try bitbaking alsa-lib |
| 23:56.42 | cp15 | And then gomunicator again |
| 23:56.51 | immolo | trying now |