irclog2html for #htc-blueangel on 20051217

00:16.49cp15Good night, will go to bed now
01:52.26pH5good night
01:52.27*** part/#htc-blueangel pH5 (n=ph5@e178248116.adsl.alicedsl.de)
03:30.48*** join/#htc-blueangel oxi (n=oxi@212.106.207.206)
03:30.49oxihi
08:02.38*** join/#htc-blueangel rob_w (n=bob@p85.212.144.106.tisdip.tiscali.de)
08:02.38*** mode/#htc-blueangel [+o rob_w] by ChanServ
09:36.03*** join/#htc-blueangel eme (n=bosty@BSN-165-118-228.dial-up.dsl.siol.net)
09:36.11emehi all
09:36.36emeanybody here to know something abt linux on BlueAngel?
09:37.06emeI have installed it and have some questions abt.
09:54.52*** join/#htc-blueangel cp15 (i=ryzdfw@p549A4317.dip0.t-ipconnect.de)
09:54.52*** mode/#htc-blueangel [+o cp15] by ChanServ
09:55.05cp15Good morning
10:00.57emegm
10:02.10emecp15 you have compiled kernel and other things for BlueAngel?
10:03.14cp15Take a look at http://www.handhelds.org/moin/moin.cgi/BlueAngel Section Installation Instructions
10:03.25cp15At least if you have a card reader connected to a linux box
10:03.44emetks
10:03.52emei have already installed it
10:04.00emebut have question or two
10:04.49emein shor: should telephone ring or not? and i have in logs df: /dev/.static/dev: Premision denied
10:04.57emeshould I run it like user or root
10:05.20emean also big problem with usbnet on kernel2.6 on pc side
10:06.34emeand cpu is no 100%  load, suspend, power off not working. Is thih normal?
10:11.07cp151. The telephone should ring, but you have to install a ringtone. See http://www.handhelds.org/moin/moin.cgi/Gomunicator
10:11.36cp152. I have also seen the df error messages but haven't looked into it further
10:12.06cp153. It should work with a non privileged user. If not, there is a bug somewhere
10:12.19cp154. What is your usbnet problem?
10:12.48cp155. What do you mean with cpu is no 100% load?
10:13.03emecpu is on 100% load
10:13.18cp156. Real suspend is in the works. Stay tuned
10:13.19emeapp gpe-conf tak 94% of cpu
10:13.36cp15What have you configured before?
10:13.43emenothing
10:13.48emejoust starting gomunicatro
10:14.28emewhen i boot on
10:14.30cp15Can you install strace and take a look what gpe-conf is doing?
10:14.32emecpu is normal
10:15.04emewhen i start anny app in settings then cpu go up to 100% and stay there
10:15.25emecan check v ps ax
10:15.27emewith ps ax
10:15.30emesecond
10:15.39cp15If gomunicator takes 100% CPU I know the reason. It has a bug which triggers when the ringtone ends
10:15.52cp15The ps is somewhat limited :-(
10:16.06emething you are right
10:17.08emewill put ringtone on sdcard and try again.
10:17.58emeis this right location for ringtone /usr/share/gomunicator/
10:18.10emewhat ist the filename of the ringtone please?
10:18.50cp15From the wiki (added a few minutes ago): Copy output.raw as /usr/share/gomunicator/R&C_theme.raw to your device
10:18.50cp15ToDo list
10:19.06cp15Copied a bit too much
10:19.55emewhat abt power off?
10:20.01emewhere can I check this?
10:20.10emeand suspend?
10:24.32cp15There is no power off
10:24.52cp15Suspend partly works. Try apm.disabled --suspend
10:25.10cp15But the power consumption in suspend mode is still to high
10:25.15rob_wi am almost there to compile gomunicator directly .. only some linker things need to be sorted out ..
10:25.50cp15Since it is no real suspend in the current image, just a "turn off the display and put the cpu in slow idle mode"
10:27.31*** join/#htc-blueangel Funklord (n=cow@c-cbd572d5.014-46-73746f28.cust.bredbandsbolaget.se) [NETSPLIT VICTIM]
10:27.31*** join/#htc-blueangel immolo (n=immolo@cpc1-cosh1-4-0-cust126.cos2.cable.ntl.com) [NETSPLIT VICTIM]
10:27.40emeok, thanks. will go now to find ringtone on wiki.
10:43.37*** join/#htc-blueangel Funklord (n=cow@c-cbd572d5.014-46-73746f28.cust.bredbandsbolaget.se) [NETSPLIT VICTIM]
11:00.28*** join/#htc-blueangel pH5 (n=ph5@e178220245.adsl.alicedsl.de)
11:00.50pH5morning all
11:13.06cp15Good morning pH5
11:13.11cp15I have read your bot mail
11:13.20pH5hey cp15.
11:13.23cp15The dma may be the cause for the memory corruption
11:13.44pH5i've changed generic -> magician and removed the OFFSET in vsfb, still the same results
11:13.51pH5(only the display is aligned now)
11:15.28cp15Have you also tried to change the 0x20 to 0x00 in the address?
11:16.34pH5not yet
11:16.36cp15Ok, when you are in windows, dump the following addresses to get information about dma 6 channel:
11:17.42cp15pd 0x40000018 4 (control register)
11:18.36emeCopy output.raw as /usr/share/gomunicator/R&C_theme.raw to your device , then I get C_theme.raw command not found. I know & maybe mistake or what???
11:18.37cp15pd 0x40000260 4 (descriptor register)
11:18.56cp15pd 0x40000264 4 (source register)
11:19.00emeand sorry for interuption
11:19.37cp15pd 0x40000268 4 (target register)
11:19.44cp15pd 0x4000026c 4 (command register)
11:20.12cp15eme: You have to set the filename in " or ' quotes, since the & is a special shell character
11:20.34emehmmm, tks
11:22.31emeit rings :)
11:36.00*** join/#htc-blueangel cp15 (i=ryzdfw@p549A4317.dip0.t-ipconnect.de) [NETSPLIT VICTIM]
11:36.00*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de) [NETSPLIT VICTIM]
11:36.00*** join/#htc-blueangel oxi (n=oxi@212.106.207.206) [NETSPLIT VICTIM]
11:36.00*** join/#htc-blueangel immolo (n=immolo@cpc1-cosh1-4-0-cust126.cos2.cable.ntl.com) [NETSPLIT VICTIM]
11:36.00*** join/#htc-blueangel pH5 (n=ph5@e178220245.adsl.alicedsl.de)
11:36.00*** join/#htc-blueangel rob_w (n=bob@p85.212.144.106.tisdip.tiscali.de)
11:36.01*** mode/#htc-blueangel [+oo cp15 rob_w] by irc.freenode.net
11:36.06*** join/#htc-blueangel Funklord (n=cow@c-cbd572d5.014-46-73746f28.cust.bredbandsbolaget.se) [NETSPLIT VICTIM]
11:40.34*** join/#htc-blueangel ChanServ (ChanServ@services.)
11:40.34*** mode/#htc-blueangel [+o ChanServ] by irc.freenode.net
11:45.52*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de) [NETSPLIT VICTIM]
11:45.52*** join/#htc-blueangel immolo (n=immolo@cpc1-cosh1-4-0-cust126.cos2.cable.ntl.com) [NETSPLIT VICTIM]
11:45.54*** join/#htc-blueangel pH5 (n=ph5@85.178.220.245)
11:47.18*** join/#htc-blueangel oxi (n=oxi@212.106.207.206) [NETSPLIT VICTIM]
11:47.52*** join/#htc-blueangel cp15 (i=yuslpi@p549A4317.dip0.t-ipconnect.de) [NETSPLIT VICTIM]
11:47.52*** mode/#htc-blueangel [+o cp15] by ChanServ
11:57.25*** join/#htc-blueangel rob_w (n=bob@p85.212.144.106.tisdip.tiscali.de) [NETSPLIT VICTIM]
11:57.25*** mode/#htc-blueangel [+o rob_w] by irc.freenode.net
11:57.53*** join/#htc-blueangel oxi (n=oxi@212.106.207.206)
11:57.53*** join/#htc-blueangel pH5 (n=ph5@85.178.220.245)
11:57.53*** join/#htc-blueangel immolo (n=immolo@cpc1-cosh1-4-0-cust126.cos2.cable.ntl.com) [NETSPLIT VICTIM]
11:57.53*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de) [NETSPLIT VICTIM]
12:04.10*** join/#htc-blueangel apt (i=ibot@pdpc/supporter/active/TimRiker/bot/apt)
12:04.10*** 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/ * http://www.userfriendly.org/ *
12:07.01*** join/#htc-blueangel rob_w (n=bob@p85.212.144.106.tisdip.tiscali.de)
12:07.01*** mode/#htc-blueangel [+o rob_w] by ChanServ
12:10.19pH5cp15: control register is 0x00000008
12:10.31pH5cp15: descriptor register is 0xa0031020
12:10.44pH5cp15: source register is 0xa0028460
12:13.47cp15And target?
12:13.50*** join/#htc-blueangel pH1 (n=ph5@e178236159.adsl.alicedsl.de)
12:16.32cp15Hmm... The control register says the dma is disabled
12:16.47cp15But descriptor and source are valid ram addresses
12:16.55cp15Can you look for the target address?
12:19.37pH1that's 0x40000268?
12:20.40cp15yes
12:20.45cp15and command also
12:20.52cp150x4000026c
12:21.06pH1target is 0x41000010
12:21.39pH1and command is 0x90229390
12:22.33cp15target is ssp1
12:24.55cp15Which is the touchscreen controller according to cr2
12:25.43cp15But since it transfers from ram to ssp1 it is probably not the culprit
12:25.52pH1also CKEN(0x41300004) is 01515ee9
12:26.10pH1isn't  the ssp1 clock disabled here?
12:29.37*** join/#htc-blueangel _cronos_ (n=cronos@ron34-1-82-224-149-102.fbx.proxad.net)
12:29.38cp15Seems so
12:32.33cp15But there might be another problem
12:33.04cp15The dma loads descriptors from ram, which gets overwritten by linux. So it might run without control
12:38.12pH5can't we disable all dma channels before jumping into the kernel?
12:38.35cp15It is done already, as I have seen
12:38.48cp15But it disables only 16 channels, the pxa27 has 32 channels
12:39.33cp15I am just checking if it is enough to replace the 16 by 32
12:39.50pH5hm. but 6 should be one of the first 16 channels, shouldn't it?
12:40.49cp15Yeah, but 6 is probably not the problematic one
12:40.59pH5ah, ok
12:41.36pH5this low level stuff is like black magic to me
12:43.43cp15Ok, goto src/wince/cpu-pxa.cpp in haret
12:44.15cp15in function pxaResetDMA replace 16 with 32 and 40 with 63
12:45.58cp15This leaves some map registers out (Memory Stick, SSP3, Quick Capture Interface, DREQ2), but hopefully the disabled channels will be enough
12:46.51pH5done. that's  all?
12:47.02cp15I hope... Just give it a try
12:47.27pH5i hope as the magician has no memory stick and the camera is disabled, this shouldn't matter
12:48.21cp15The DREQ2 might be problematic... It is for dma from/to external (not in the pxa integrated) devices
12:48.41pH5cp15: what kind of devices?
12:49.24cp15Maybe the wireless lan interface to transfer network data... But I don't really know
12:49.53pH5it has no wlan
12:50.03pH5maybe we're lucky
12:52.38cp15We can disable the map registers later, but it requires more than just changing a number since the registers are not continous
12:53.08cp15BTW: Are you planning to go to the CCC congress between christmas and new year?
12:53.36cp15There is a talk about linux on mobile phones, and I am taking part in this one
12:54.31pH5cp15: unfortunately, I'll be out of town until 20060101
12:55.26cp15Oh no, a new year
12:55.35cp15Wasn't there just a new one?
12:58.45pH5the zImage still hangs
12:59.04cp15:-(
12:59.17pH5no wonder time goes by so fast if it takes so long just to get to a proper zImage kernel panic
12:59.52*** join/#htc-blueangel lkcl (n=lkcl@bb-87-82-26-80.ukonline.co.uk)
13:00.18lkclwotcha
13:00.21lkcl~botmail read
13:00.43cp15Hi lkcl, do you have any idea what might be the problem if Image boots but zImage does not?
13:00.46lkclhi oxi
13:00.53lkcleek, no.
13:01.09lkclthat sounds very odd, cp15
13:01.34cp15ph5: I would just continue with the Image. Maybe we get some more hints to the problem later
13:01.58cp15I assume that the zImage fails because of some memory corruption
13:02.15cp15But even with all dma channels disabled it just doesn't work
13:02.26lkclwith the himalaya, booting took several attempts.
13:02.38lkclsometimes they worked, sometimes they didn't.
13:03.21cp15On the BA it works with a 20:1 chance, I would say
13:05.16lkclbtw, cp15: thought i'd look at ahioem.asm again.  you got any clue as to what spi_out_data would do?
13:05.32lkclnothing to do with the screen, at all, would it?  guess?
13:06.12cp15No, not exactly. My best guess is that the toppoly LCD contains some additional logic which is programmed via the spi
13:06.41lkclhmmm... oh dear.
13:07.04lkclin some ways i think we'd be better off looking at the bootloader assembly instead.
13:07.33cp15The bootloader is a bit harder to understand than a single dll
13:07.48cp15But I have an emulator where I can run it and see what it does to the gpios
13:08.07lkclooo.
13:08.08pH5hi lkcl
13:08.12lkclheya ph5
13:08.21pH5cp15: i guess i'll have to do as you propose :)
13:09.54cp15Maybe you could try the BA ramdisk and see if you get a crc error when the kernel unpacks it
13:10.38cp15It might be also quite possible that there is a bug with the memory layout and the decompression stage fails
13:11.22cp15lkcl: Do you have any idea why the image compiled by you is over 2 times the size of the last image?
13:13.48lkclprobably because i compiled in debug info / specified debug info in the bitbake options somewhere.
13:13.53lkclblame immolo :)
13:14.43cp15Have you specified DEBUG_BUILD = "1" or INHIBIT_PACKAGE_STRIP = "1" in your local.conf?
13:14.47lkclwant me to switch it all off and run it again?
13:14.58lkclah, yes - that's right.  local.conf.  yep!
13:15.12cp15Ok, then remove both options
13:15.15lkclack.
13:15.27cp15I also have some requests for new modules to be integrated:
13:15.35lkclshoot.
13:16.07lkclbtw that damn keyboard module still don't get thingied.
13:16.32cp15With autoload: blueangel_suspend blueangel_leds blueangel_dpram blueangel_battery
13:16.36lkcloh - what does LDRH do, again?
13:16.45lkcli've got suspend in already...
13:16.50cp15Without autoload ledscore battery nfs
13:17.07lkcland leds
13:17.12cp15Load register halfword
13:17.19cp15halfword=16 Bit
13:17.26lkclin the top or bottom half?
13:17.46cp15There is only a bottom half on arm. Show me the complete instruction
13:17.57lkclLDRH R0, [R2], #2
13:18.38lkclR2 = 0x90002000
13:18.42lkclwhich is a bit odd.
13:18.55cp15In C: r0=*((short *) (r2+2))
13:19.05lkclta.
13:20.17cp15I will take a look into the 0x6 suspend code next week. This week I can't because I need a working phone
13:20.26lkcl:)
13:20.36lkclack.
13:21.24lkclokay.  so.  module_autoload: suspend, leds, dpram, batter.
13:21.25cp15BTW: I have found your resume. Quite impressive
13:21.31lkcl*snort* :)
13:21.42lkcloh.  that reminds me.  linkedin.com.  very useful.
13:22.10lkcla lot of free software developers are on it and creating projects as "companies".  e.g. "Debian" as a non-profit "company".
13:22.46lkclsend me an email address i'd be delighted to create a link to you (and rob when he wakes up)
13:23.10lkclwhat does "without autoload ledscore battery nfs" mean?
13:23.24pH5cp15: where do I find the ba ramdisk?
13:23.39pH5the BlueangelKernel wiki links seem to be broken
13:24.56cp15http://gnulinux.biz/files/blueangel/sd/linux/
13:25.09pH5cp15: thanks
13:26.46lkclwhat's CMP r1, r0; BCC loc_10001288 again, cp15?
13:27.13cp15lkcl: Add it to the BA_MODULES_kernel26 but not to modules_autoload
13:27.31lkclack.  understood.
13:28.02pH5is there a ba 2.4 kernel?
13:28.22lkclyou _really_ don't want to be using 2.4, ph5.
13:28.29pH5if not, why not just drop the _kernel26 override?
13:28.40lkclyou will get _absolutely_ no help from russell, or anybody.
13:28.49lkclyeh, good point: why _are_ we doing that? :)
13:29.01pH5copy&paste from h3900, i guess
13:29.12lkcli like cut/paste
13:29.27pH5:)
13:32.33lkcloff we go again...
13:33.11cp15if (r1 < r0) goto loc_10001288 (I hope)
13:35.46cp15Ah, please also do a fresh kernel checkout, in the old one battery doesn't work
13:38.11pH5cp15: is there a way to read the voltage of the internal nimh backup battery on ba?
13:38.31lkcli deleted the ../sources/linux26-xan....
13:38.40lkcland everything, really :)
13:38.47cp15Probably, since windows can do it. But I have yet to find out how :-)
13:39.47pH5cp15: it seems to me that for the ipaqs also nobody knows how to do it :-/
13:41.03cp15On the BA it is probably done via the dpram asic, but the protocol is very complicated
13:43.02cp15lkcl: Any idea why zblueangel_kp does not work?
13:49.18cp15Doesn't usb work`
13:49.20cp15Doesn't usb work?
13:52.07pH5hm. i just tried with the initrd and now even the Image doesn't boot
13:52.20pH5cp15: how to I transfer files to a wince device over usb?
13:52.42pH5until now i didn't think the time to learn how to do it wasn't worth it
13:55.09cp15Install synce then you can do a pcp localfile :remotefile
13:55.58*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
13:56.23cp15But OTOH: The ipaq-Module from synce makes some trouble when switching to linux. So if BT is ok for you...
14:00.16pH5it is, once I've saved that initrd into wince storage
14:22.19*** join/#htc-blueangel iclem (n=clement@iclem.net)
14:28.04*** join/#htc-blueangel rob_w (n=bob@p85.212.144.106.tisdip.tiscali.de)
14:28.04*** mode/#htc-blueangel [+o rob_w] by ChanServ
14:37.38cp15Ok, I have to leave now for doing some customer work and a birthday party
14:37.40cp15Bye
14:37.45pH5bye cp15
14:39.04pH5RAMDISK: Compressed image found at block 0
14:39.04pH5Kernel panic - not syncing: VFS: Unable to mount root  fs on unknown-block(1,0)
14:41.56cp15Have you specified root=/dev/ram0 init=/linuxrc ramdisk_size=14336 keepinitrd
14:42.11cp15At least it doesn't seem to get an crc error
14:42.56pH5cp15: yes
14:47.57lkclheya cp15: i'm off out until this evening.  a new build will be done in a couple hours, i'll upload it this evening.
15:15.29*** join/#htc-blueangel eme (n=bosty@BSN-165-118-228.dial-up.dsl.siol.net)
15:23.46pH5cp15: compiling in ext2 support did the trick :)
15:47.24emecp15: good evening. do you have a minute?
16:07.14*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
16:50.52*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
16:54.21*** join/#htc-blueangel eme (n=bosty@86.61.4.70)
16:58.40emegood evening. I have installed from hh-wiki on blueangel, beside no bt and wifi, I do not get usbnet working on host pc. no such device. Yes I do modprobe usbnet.
16:59.22emekernel on host side is  2.6.14-2-k7 os Debian testing/unstable
17:00.57pH5hi eme
17:01.24pH5did you load the cdc_subset driver or something equivalent?
17:01.55emeon host side?
17:01.55pH5yes
17:01.55emeno will do it now
17:02.14pH5on 2.6.14 usbnet is only the framework
17:03.27emeit is the same...no device
17:03.53pH5hm. do see something in the host kernel log when you plug in the cable?
17:04.17emenothing, maybe different usb id or something
17:06.10emelsusb also gave nothing
17:06.43emedo I have to do anything on ba side?
17:07.09pH5can you see if pxa2xx_udc and g_ether modules are loaded?
17:07.21emeusbnet on ba is up
17:08.24pH5that's strange. then it should at least show up in lsusb imho.
17:09.03emei know ... :(
17:09.06*** join/#htc-blueangel JuNuKN (n=JuNuKN@Kce9e.k.pppool.de)
17:15.31ememouse, card reader, camera, everythings popup in kernel.log joust linux on ba does not. I
17:23.27pH5eme: I just have the same problem with my magician. pxa27x_udc loaded, g_ether loaded, but it doesn't show up on the host.
17:23.53pH5I only see a "udc: USB reset" message on the magician's console whenever I plug in the cable
17:37.07emeBus 001 Device 007: ID 049f:505a Compaq Computer Corp. SA-11x0 based Linux Devic
17:37.10emewill se now
17:39.09emefinaly.....root@blueangel:~#
17:39.20pH5eme: what did you change?
17:39.26emenothing,
17:39.31pH5:)
17:39.34emereset it again
17:39.40emeand put it into craddle
17:39.44emeboot wince
17:40.00emeinsert cf card boot linux
17:40.23emething ba must stay in creaddle when booting
17:41.38emeduring boot ba was recognized in syslog
17:44.48oxihi
17:44.59pH5hi oxi
17:45.40*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
18:01.40*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
18:46.02immolohow on earth do you compile in debug stuff with oe :P
18:48.25pH5immolo: set DEBUG_BUILD="1" and INHIBIT_PACKAGE_STRIP="1" in your local configuration file.
18:54.54*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
19:02.02*** join/#htc-blueangel rob_w (n=bob@p85.212.144.106.tisdip.tiscali.de)
19:02.03*** mode/#htc-blueangel [+o rob_w] by ChanServ
19:06.12immolopH5, oh just wondering how lkcl managed to get it
21:06.13*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
21:16.19oxionce the mmc/sd module is loaded, where under /dev do you access the device?
21:17.16pH5mmcblkp0 if you don't use devfs
21:17.29oxithanks
21:24.14*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
21:46.07oxiwhere can i find info on asic3? what it is, how to implement for my device etc.
21:47.05oxido xanadux devs have any irc channel?
21:47.56oxieveryone off already? :)
21:48.38immolo:O a sparcbook on ebay I'm in heaven
21:50.26immolobut then I could also buy a pretty ibook
22:14.49immoloor I could just buy an arm based sub notebook and port gentoo to it, so much choice so little money
23:35.34*** join/#htc-blueangel bhima (n=gf2e@i13pc168.ilkd.uni-karlsruhe.de)
23:49.33oxigentoo rocks

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.