**** BEGIN LOGGING AT Fri May 08 02:59:57 2009
May 08 04:16:52 <slaxxer> where is the dark keyboard theme for shr?
May 08 04:17:30 <slaxxer> white letters an black background
May 08 07:44:58 <jz1> Hi all, I am having a problem when making qemu following these instructions... http://wiki.openmoko.org/wiki/Automatic_emulation_in_Ubuntu
May 08 07:45:28 <jz1> in particular the file...    openmoko/trunk/src/host/qemu-neo1973/openmoko/download.sh
May 08 07:47:19 <jz1> When I "make qemu", output is...
May 08 07:47:19 <jz1> ( cd images && ../openmoko/trunk/src/host/qemu-neo1973/openmoko/download.sh )
May 08 07:47:19 <jz1>    Retrieving available builds list...
May 08 07:47:19 <jz1>     Trying alternative sources
May 08 07:47:19 <jz1>     Kernel is... not found
May 08 07:47:20 <jz1> make: *** [download-images] Error 255
May 08 07:48:00 <jz1> is this the right place to seek help?
May 08 07:48:30 <SpeedEvil> what are you trying to build
May 08 07:48:31 <SpeedEvil> hangon
May 08 07:48:52 <SpeedEvil> yes, in principle - but 9AM is a bit of a quiet time - the US has gone mostly to sleep
May 08 07:49:03 <jz1> ah ok
May 08 07:49:40 <jz1> are you familiar with what  I asked about above?
May 08 07:50:26 <jz1> maybe wholistics aside, the problem I think lies here...
May 08 07:50:32 <jz1> echo "    "Retrieving available builds list...
May 08 07:50:32 <jz1> backup_download_dir="http://downloads.openmoko.org/repository/Om2008.8/"
May 08 07:50:32 <jz1> echo download_dir $download_dir
May 08 07:50:32 <jz1> echo backupdirectory $backup_download_dir
May 08 07:50:32 <jz1> files=`lynx -dump $download_dir | grep http | sed "s,[0-9 \t\.]*$download_dir\([a-zA-Z0-9_\.-]\)[ \t]*,\1,"`
May 08 07:50:33 <jz1> dev_files=`lynx -dump $dev_download_dir | grep http | sed "s,[0-9 \t\.]*$dev_download_dir\([a-zA-Z0-9_\.-]\)[ \t]*,\1,"`
May 08 07:50:36 <jz1> if [[ "${files}" == "" ]]; then
May 08 07:50:37 <SpeedEvil> Sorry - not 
May 08 07:50:38 <jz1> 	echo "    "Trying alternative sources
May 08 07:50:40 <jz1> 	download_dir="$backup_download_dir"
May 08 07:50:42 <jz1> 	dev_download_dir="$backup_download_dir"
May 08 07:50:44 <jz1> 	files=`lynx -dump $download_dir | grep http | sed "s,[0-9 \t\.]*$download_dir\([a-zA-Z0-9_\.-]\)[ \t]*,\1,"`
May 08 07:50:47 <jz1> 	dev_files="$files"
May 08 07:50:49 <jz1> fi
May 08 07:50:50 <SpeedEvil> 2008.8 is AIUI deprecated
May 08 07:51:00 <jz1> ah, that could be it
May 08 07:51:06 <jz1> whats a good link?
May 08 07:51:33 <SpeedEvil> The wiki isn't as up-to-date as one might hope
May 08 07:51:57 <jz1> the script originally just used daily, and 2007, before I put the 2008.8 in
May 08 07:52:41 <SpeedEvil> Sorry - I haven't tried to use qemu in a couple of years - got hardware, then I've not kept up with the recent changes - due to time issues
May 08 07:52:54 <SpeedEvil> have you got qemu up, or do you need to build it?
May 08 07:52:57 <jz1> where do you get your images?
May 08 07:53:14 <jz1> well I think qemu is good now, its just when it tries to get a kernel
May 08 07:53:22 <jz1> i believe its just a broken link
May 08 07:53:44 <jz1> http://downloads.openmoko.org/daily/
May 08 07:53:44 <jz1> doesn't exist anymore, right?
May 08 07:54:10 <jz1> whats the equivalent to what this was?
May 08 07:55:00 <SpeedEvil> http://wiki.openmoko.org/wiki/Distributions
May 08 07:55:04 <SpeedEvil> is probably a good link
May 08 07:55:12 <SpeedEvil> not that that has images
May 08 07:55:18 <SpeedEvil> just links to 
May 08 07:55:50 <jz1> http://downloads.openmoko.org/distro/  ?
May 08 07:56:25 <Heinervdm> Someone knows how to convert tangogps .log to .gpx ?
May 08 07:57:07 <alphaone> gpsbabel will do it
May 08 07:57:10 <jz1> SpeedEvil: thanks for your help, I'll try a few different places
May 08 07:57:16 <SpeedEvil> jz1: good luck
May 08 07:57:25 <SpeedEvil> gpsbabel will do everything!
May 08 07:57:26 <Heinervdm> alphaone: wich -i ?
May 08 07:57:42 <alphaone> Heinervdm: It was more complicated than that :-)
May 08 07:57:49 <Heinervdm> :(
May 08 07:57:49 <mirko-paroli> jz1: what are you looking for?
May 08 07:57:54 <mirko-paroli> om2008.12 ?
May 08 07:57:59 <SpeedEvil> IIRC there is a way to tell tangogps to log in NMEA formatt, which I did
May 08 07:58:14 <jz1> mirko-paroli: sure, that might work
May 08 07:58:23 <mirko-paroli> jz1: http://downloads.openmoko.org/distro/releases/Om2008.12/
May 08 07:58:26 <SpeedEvil> Heinervdm: IIRC you need to set it up as a whitespace seperated format, and tell it what the fields mean
May 08 07:58:32 <jz1> its where the qemu script searches for a kernel
May 08 07:59:01 <Heinervdm> SpeedEvil: ok, will try
May 08 07:59:29 <alphaone> google for http://wiki.openmoko.org/wiki/TangoGPS#Using_tangoGPS_for_OpenStreetMap.org
May 08 07:59:51 <alphaone> Well, there is a way to tell gpsbabel as well...
May 08 08:00:00 <alphaone> That link uses a skript
May 08 08:00:35 <alphaone> http://wiki.openmoko.org/wiki/TangoGPS#Using_the_TangoGPS_logs
May 08 08:00:39 <alphaone> Ah, there we go
May 08 08:01:08 <Heinervdm> alphaone: thats great, it works :)
May 08 08:01:13 <Heinervdm> thx
May 08 08:01:15 <alphaone> np
May 08 08:02:15 <jz1> basically the line in the script is now effectively
May 08 08:02:15 <jz1> files=`lynx -dump "http://downloads.openmoko.org/distro/releases/Om2008.12/" | grep http | sed "s,[0-9 \t\.]*$download_dir\([a-zA-Z0-9_\.-]\)[ \t]*,\1,"`
May 08 08:05:53 <Hopscotch> good morning
May 08 08:08:34 <jz1> goodmorning hopscotch
May 08 08:43:13 <khiraly1> re
May 08 09:04:05 <methril|work> morning
May 08 09:26:35 <CIA-53> freesmartphone.org: 03mickey 07cornucopia * rcceb71486b85 10/libfsotransport/ (3 files in 2 dirs): 
May 08 09:26:35 <CIA-53> freesmartphone.org: enable Transport.GetDelegate, now that http://bugzilla.gnome.org/show_bug.cgi?id=578412 is fixed
May 08 09:26:35 <CIA-53> freesmartphone.org: bump version
May 08 09:55:58 <badcloud> finalllly got debian running!
May 08 09:56:51 <badcloud> zhone doesn't run though.... even after chmod 775
May 08 10:35:08 <skamster> hello there.. i had just a little question about pygtk.. i want to make the label, which show's the text to use the full width.. but it doesn't work
May 08 10:35:35 <skamster> when i use wrap, it use just the half width
May 08 10:35:50 <skamster> when i didn't use wrap, it goes over the display
May 08 10:35:55 <skamster> it's about this code: http://death-head.ch/openmoko/sms-reader/sms-read.py
May 08 10:36:54 <skamster> the strange thing is, i use also buttons in the (v)box
May 08 10:37:14 <skamster> these use the full width of the display, like i want
May 08 10:37:20 <skamster> just the label didn't
May 08 10:38:18 <skamster> could someone help?
May 08 10:40:34 <thefloe_> hi, after last update on shr-testing, enlightenment seg faults at startup... anyone having this behavior too?
May 08 10:50:56 <messerting> Hi folks, I've just installed SHR-testing. Looks very nice!
May 08 10:51:20 <messerting> I've also managed to get online through usb by following the instructions on the wiki
May 08 10:51:41 <messerting> Now, I'd like to use wifi, but I just can't find where to start Mofi...
May 08 10:52:01 <messerting> it is installed and listed by opkg, but where is the icon for it?
May 08 10:56:18 <messerting> oh wait, "opkg list" lists not only installed packages... installing now...
May 08 10:56:47 <ndnihil> opkg lists available package
May 08 10:56:53 <ndnihil> opkg list_installed lists installed packages
May 08 10:57:23 <messerting> ndnihil: thanks, I'm a bit of newb to this opkg system :)
May 08 10:57:34 <messerting> got it installed now - and found the icon
May 08 10:57:38 <messerting> will test it
May 08 10:59:37 <messerting> wow! seems to work
May 08 11:00:47 <messerting> yes. cool. thanks
May 08 11:20:08 <CIA-53> freesmartphone.org: 03mickey 07specs * rad742f282140 10/odevice/org.freesmartphone.Device.Audio.xml.in: org.freesmartphone.Device.Audio: add fso:enumeration org.freesmartphone.Device.SoundState
May 08 11:23:20 <CIA-53> freesmartphone.org: 03mickey 07specs * r74cfda1b6684 10/odevice/org.freesmartphone.Device.Input.xml.in: org.freesmartphone.Device.Input: add fso:enumeration org.freesmartphone.Device.InputState
May 08 11:27:05 <or4n9e> could anyone tell me how to prioritize repostories in /etc/opkg.conf?
May 08 11:27:26 <or4n9e> simple listing by priority doesn't help unfortunately
May 08 11:37:54 <CIA-53> freesmartphone.org: 03mickey 07specs * rbe5ff63a6efe 10/odevice/org.freesmartphone.Device.Audio.xml.in: org.freesmartphone.Device.Audio: add fso:errordomain
May 08 11:38:57 <CIA-53> freesmartphone.org: 03mickey 07specs * r76734e317ee1 10/otapi/org.freesmartphone.GSM.Call.xml.in: org.freesmartpone.Call: change error domain name
May 08 11:41:15 <CIA-53> freesmartphone.org: 03mickey 07cornucopia * rbfded99e5581 10/libfsoframework/ (6 files in 2 dirs): fsoframework: add skeleton for KObjectNotifier
May 08 11:41:15 <CIA-53> freesmartphone.org: 03mickey 07cornucopia * r850fee979cac 10/libfsoframework/vapi/linux26.vapi: fsoframework: linux26.vapi: include sys/ioctl.h here, since it's Linux specific
May 08 12:06:44 <CIA-53> freesmartphone.org: 03mickey 07framework * r71c1645a8160 10/framework/subsystems/odeviced/idlenotifier.py: odeviced: idlenotifier: do not reset all config timeouts when re(enabling) one
May 08 12:21:50 <PBeck> ahoi
May 08 12:29:34 <CIA-53> freesmartphone.org: 03mickey 07cornucopia * r4359aca02353 10/libfsoframework/ (configure.ac fsoframework/logger.vala vapi/posixextra.vapi): fsoframework: depend on vala 0.7.3 which has our syslog patches applied, remove from posixextra
May 08 12:29:36 <CIA-53> freesmartphone.org: 03mickey 07cornucopia * r9d11c512778e 10/libfsoframework/vapi/linux26.vapi: fsoframework: linux26.vapi: add netlink message constants
May 08 13:41:45 <thefloe> Hi, where can I find enlightenment elementary documentation?
May 08 14:34:02 <CIA-53> freesmartphone.org: 03mickey 07cornucopia * r69d6bc60f622 10/libfsoframework/ (6 files in 3 dirs): fsoframework: add enough bindings to get messages from the kobject netlink dispatcher
May 08 15:46:36 <Blu3> hmm, odd.  it just took ~4 minutes to delete a text message
May 08 15:46:54 <SpeedEvil> Set autorepeat faster then. 
May 08 15:47:04 <SpeedEvil> It has to hold down the delete key for ages on long SMSs.
May 08 15:49:03 <max_posedon> you can select text of message and press "delete" I think
May 08 15:52:37 <khiraly1> re
May 08 15:52:41 <khiraly1> mirko-paroli: are you here?
May 08 15:53:22 <khiraly1> mirko-paroli: today I tried to use the phone the whole day. (I missed 4 phone call in total, and one person couldnt reach me at all ;-|)
May 08 15:53:54 <khiraly1> The general impression, if you use paroli phone calling it will more and more likely to fail
May 08 15:55:10 <khiraly1> The easiest to trigger, if you launch the Tele application, you initiate a call, and talk with him. After you hangup. But you dont close the Tele application, in this state if you get calling from others, three or four is even enough to get fail paroli
May 08 15:55:12 <PaulFertser> khiraly1: likely than what?
May 08 15:55:31 <PaulFertser> khiraly1: ah, more and more, likelihood increases with uptime, i see.
May 08 15:55:49 <khiraly1> PaulFertser: If you just launch paroli, and dont use it, it will work fine
May 08 15:56:01 <khiraly1> PaulFertser: nope. Its not triggered over time, but over usage
May 08 15:56:16 <khiraly1> if you dont use paroli at the evening you are reachable
May 08 15:56:38 <khiraly1> but if you use paroli, its likelihood that within 2 hours nobody can you reachn
May 08 15:56:44 <PaulFertser> khiraly1: Have you seen anything suspicious in frameworkd log?
May 08 15:56:57 <khiraly1> PaulFertser: I have only paroli logs saved
May 08 15:57:07 <khiraly1> I saved many different problem.
May 08 15:57:14 <Blu3> no, it took four minutes to delete a single message after selecting it and touching delete
May 08 15:57:24 <Blu3> did it twice now
May 08 15:57:38 <khiraly1> But dont know where is frameworkd log. So If you tell me, I will save it too
May 08 15:57:52 <Blu3> it's in /var/log/
May 08 15:58:19 <PaulFertser> khiraly1: distro-specific, most probably in your syslog.
May 08 15:58:26 <khiraly1> om2009
May 08 15:58:28 <khiraly1> unstable
May 08 15:58:35 <khiraly1> with some tweaking from mirko
May 08 15:59:11 <PaulFertser> khiraly1: i meant it's up to you to find out where the log goes, it's fairly easy.
May 08 15:59:14 <khiraly1> the biggest usability issue, that as paroli gets unusable its more likelihood, that the power button cant initiate suspend anymore
May 08 15:59:53 <khiraly1> But the phone will suspend if you dont use it. Just it is impossible to wake up (as the power button does not work)
May 08 15:59:54 <PaulFertser> khiraly1: what about calling suspend method of frameworkd directly?
May 08 16:00:04 <khiraly1> I hitted that problem at least 4 time today
May 08 16:00:37 <PaulFertser> khiraly1: doesn't look anything like paroli-specific :( unable to resume looks like kernel problem (and yes, i see it ocassionally too)...
May 08 16:00:40 <khiraly1> PaulFertser: Is paroli who controls the power button, or frameworkd
May 08 16:00:49 <khiraly1> ?
May 08 16:01:10 <PaulFertser> khiraly1: Unable to suspend can be due to various reasons, one of which is ogsmd unable to disable unsolicited messages from GSM.
May 08 16:01:19 <khiraly1> PaulFertser: Im pretty sure that it could resume *if* the power button would work as before
May 08 16:01:33 <PaulFertser> khiraly1: (power button control) it depends on your rules.yaml in frameworkd.
May 08 16:01:42 <khiraly1> ok. Im checking it
May 08 16:01:44 <khiraly1> just a sec
May 08 16:02:12 <khiraly1> the second biggest usability issue, that if I resume via the power button, it resumes fine, but the screen stays blanked
May 08 16:02:22 <khiraly1> so there is no feedback if I resumed correctly or not
May 08 16:02:25 <PaulFertser> khiraly1: resume by power button is pure low-level stuff, only bootloader, kernel involved. After that you get at least USB (and usable system if X driver is ok).
May 08 16:02:32 <khiraly1> I need to tap on the screen to see paroli
May 08 16:02:40 <khiraly1> and often I tap on buttons what I dont want
May 08 16:02:59 <khiraly1> PaulFertser: and initiate the suspend?
May 08 16:03:53 <khiraly1> I think resume is fine. The problem that the power button just does not work anymore at a given time. It happens randomly. But if happened already, it will never "cure" 
May 08 16:03:55 <PaulFertser> khiraly1: it's initiated basically by "echo mem > /sys/power/state" but for it to be reasonable (i.e. you don't want to resume because calypso wants to tell you gsm connection status) first some unsols should be disabled.
May 08 16:04:16 <khiraly1> and the event of the power button?
May 08 16:04:18 <khiraly1> what handles it?
May 08 16:05:11 <PaulFertser> khiraly1: it can be handled by anything, including oeventsd from framework.
May 08 16:05:16 <khiraly1> PaulFertser: oh and it goes suspend when usb connected and I dont touch the phone. Rather irritating
May 08 16:05:18 <PaulFertser> khiraly1: so, it's distro-specific.
May 08 16:05:54 <PaulFertser> khiraly1: but resume on power button is pure low-level stuff.
May 08 16:06:11 <khiraly1> I thought there is a connection between them
May 08 16:06:43 <khiraly1> so if initiating suspend does not work, resume will not work either (when the system suspended by self)
May 08 16:06:49 <PaulFertser> khiraly1: no
May 08 16:06:54 <PaulFertser> khiraly1: ne relation whatsoever.
May 08 16:07:27 <khiraly1> rather strange, because it looks like
May 08 16:07:46 <khiraly1> oh, and after resume why the screen stays blank? (if blanked before suspend)
May 08 16:08:10 <khiraly1> so If I dont touch the phone. 1. the screen blanks (black) 2. the system suspends
May 08 16:08:23 <khiraly1> I resume the system via power button. But the screen stays black
May 08 16:08:43 <khiraly1> I need to tap on the screen to see something. So no visual feedback when the resume was fine
May 08 16:09:05 <PaulFertser> khiraly1: probably power button event is eaten by the kernel driver (i don't remember if it's still the case). Or it's just not used as a source of unblanking event.
May 08 16:09:57 <khiraly1> ohh. I have var/log/frameworkd.log
May 08 16:10:07 <khiraly1> I will save it with paroli in the future
May 08 16:12:10 <khiraly1> ok. I have right now broken paroli. Cant call anyone
May 08 16:12:28 <PaulFertser> khiraly1: the best way to see if the phone resumed or not is to plug in USB. On plugging USB it must try to resume anyway. And if you see usb device on your host and can ssh into, well, then it resumed. Otherwise -> kernel (likely) / bootloader (unlikely) problem.
May 08 16:12:58 <khiraly1> PaulFertser: but it goes to suspend when connected to the computer
May 08 16:13:05 <PaulFertser> khiraly1: replug usb then
May 08 16:13:06 <khiraly1> if I dont touch the phone regularly
May 08 16:13:11 <khiraly1> hmm. Ok
May 08 16:13:37 <lindi-> PaulFertser: is there some way to have the phone not resume when you plug usb? that'd be very useful
May 08 16:13:56 <khiraly1> lindi-: why it would be usefull?
May 08 16:14:35 <PaulFertser> lindi-: it was intentional, to make battery charge always controllable, iirc.
May 08 16:14:57 <lindi-> PaulFertser: ok :(
May 08 16:15:17 <PaulFertser> lindi-: not sure if it can be workarounded. Can you share your specific use-case?
May 08 16:15:24 <khiraly1> hmm. Looks like paroli problem right now. There are plenty of errors in the log
May 08 16:15:53 <khiraly1> and in frameworkd.log there are no call events. So looks like paroli failed to send the event
May 08 16:16:20 <lindi-> PaulFertser: well, i don't have any automatic resume so it's tedious for me to manually tell it to suspend again
May 08 16:16:26 <lindi->  s/resume/suspend/
May 08 16:16:43 <khiraly1> what is strange, that when I cant call, I have many calling screen in the illume taskbar. But only one paroli process id (ps -e |grep paroli)
May 08 16:16:55 <PaulFertser> lindi-: i guess kernel should be able to notify userspace and you can catch that and force to suspend again.
May 08 16:16:55 <lindi-> PaulFertser: won't power button and usb plug-in both cause 'EINT09_PMU'?
May 08 16:17:22 <lindi-> PaulFertser: if I could differentiate between those I could automate this
May 08 16:17:47 <khiraly1> lindi-: but when connected what is the benefit to suspend the phone? You have external power source, no need to spare with it
May 08 16:18:02 <lindi-> khiraly1: charges faster
May 08 16:18:16 <lindi-> khiraly1: the external power source can be my external usb li-ion battery
May 08 16:18:40 <lindi-> khiraly1: which I need to spare. and think of the environment!11 :)
May 08 16:18:46 <PaulFertser> lindi-: you can, why not, there's a PMU resume reason available iirc.
May 08 16:18:53 <lindi-> aha
May 08 16:19:06 <lindi-> PaulFertser: but not in sysfs yet?
May 08 16:19:54 <PaulFertser> lindi-: /sys/class/i2c-adapter/i2c-0/0-0073/resume_reason looks like what you need. 
May 08 16:19:57 <khiraly1> Im writing a letter to mirko. This paroli problem have lot of info in the paroli.log, and have the same symptoms when happens
May 08 16:20:16 <lindi-> PaulFertser: says '0002000000' hmm
May 08 16:20:24 <khiraly1> 0000000000
May 08 16:20:29 <khiraly1> ^^in my case
May 08 16:20:33 <PaulFertser> lindi-: i assume you need help decoding? ;)
May 08 16:20:40 <lindi-> no i have the docs
May 08 16:20:46 <lindi-> and I can do trial and error :)
May 08 16:20:58 <khiraly1> there are only three possibilities;)
May 08 16:21:02 <khiraly1> usb, button and nothing
May 08 16:21:10 <lindi-> khiraly1: low battery
May 08 16:23:51 <PaulFertser> lindi-: http://pastebin.com/m50d76558
May 08 16:24:11 <lindi-> seems I hit WSOD while experimenting with this
May 08 16:28:52 <PaulFertser> lindi-: confirmed, resume on usb gives 0400000000, as expected from reading sources.
May 08 16:29:44 <lindi-> PaulFertser: do you have any idea if Qi should be possible to build on debian? (without using toolchain binary from openmoko)
May 08 16:30:04 <PaulFertser> lindi-: And you can look at include/linux/mfd/pcf50633/core.h to know values for other possible resume reasons.
May 08 16:30:12 <lindi-> PaulFertser: I modified config.mk to use arm-linux-gnueabi- but now it is looking for /usr/lib/gcc/arm-linux-gnueabi/4.3.2/include/asm/byteorder.h which I don't have
May 08 16:30:27 <PaulFertser> lindi-: It is possible. Let me see my config
May 08 16:31:09 <lindi-> PaulFertser: $ apt-file search /byteorder.h | nc paste.dyndns.org 1234
May 08 16:31:10 <lindi-> http://rafb.net/p/2gdbuG38.html
May 08 16:32:07 <PaulFertser> lindi-: it's from the kernel headers. You can use make headers_install linux makefile target to install it anywhere.
May 08 16:32:45 <PaulFertser> lindi-: just specify the path with INSTALL_HDR_PATH
May 08 16:32:55 <lindi-> PaulFertser: it needs to be for the exact kernel version I am using?
May 08 16:33:08 <PaulFertser> lindi-: nope, any reasonable arm will do.
May 08 16:33:20 <lindi-> ok but it needs to be arm
May 08 16:33:39 <lindi-> maybe i just build Qi on arm to make it simpler :)
May 08 16:33:44 <PaulFertser> lindi-: don't pretend you don't have configured arm kernel sources.
May 08 16:34:04 <lindi-> PaulFertser: I do but a package can't depend on /home/lindi :)
May 08 16:34:46 <PaulFertser> lindi-: how is this stuff usually solved for other programs that need kernel headers?
May 08 16:35:06 <lindi-> PaulFertser: there's a package for kernel headers
May 08 16:35:14 <lindi-> PaulFertser: and I Build-Depend on that
May 08 16:36:02 <PaulFertser> lindi-: well, if it's for _target_ kernel headers why is there any problem?
May 08 16:37:00 <lindi-> PaulFertser: no problem, i just first thought i'd cross-compile it
May 08 16:37:18 <PaulFertser> lindi-: are you making a proper Qi package?
May 08 16:37:32 <lindi-> PaulFertser: probably not proper but something that works :)
May 08 16:37:48 <lindi-> PaulFertser: btw, couldn't /usr/include/asm/byteorder.h from linux-libc-dev be used?
May 08 16:38:19 <PaulFertser> lindi-: not sure, no idea :-/
May 08 16:38:44 <lindi-> I wish programs built with ./configure && make :)
May 08 16:40:28 <lindi-> PaulFertser: there's probably some trick needed to have the makefile print the commands it is running?
May 08 16:42:05 <PaulFertser> lindi-: kernel makefile or any makefile?
May 08 16:42:38 <lindi-> PaulFertser: Qi makefile
May 08 16:42:48 <khiraly1> ok
May 08 16:43:06 <PaulFertser> lindi-: remove @ from the commands?
May 08 16:43:21 <khiraly1> Mail sended. 10 different reason fails paroli. I attached paroli.log for all of them.
May 08 16:43:22 <khiraly1> ;)
May 08 16:43:38 <lindi-> PaulFertser: ok but something like V=1 in linux?
May 08 16:44:55 <PaulFertser> lindi-: you can just echo all the commands without executing with "-n" switch
May 08 16:45:21 <lindi-> PaulFertser: I'd just want the package builds logs to contain something meaningful
May 08 16:45:34 <lindi-> PaulFertser: so that if somebody has a problem they can compare against that
May 08 16:45:51 <lindi-> PaulFertser: now if the package patches Makefile it'll conflict with upstream changes very easily
May 08 16:46:27 <PaulFertser> lindi-: If somebody really wants to build a bootloader, i wouldn't bother too much on making it end-user friendly.
May 08 16:46:43 <PaulFertser> lindi-: and imho it's already dev-friendly enough.
May 08 16:47:23 <lindi-> PaulFertser: is it in some distro already? ;)
May 08 16:47:34 <lindi-> PaulFertser: anyways, I hit http://rafb.net/p/G5tCwA48.html
May 08 16:48:03 <PaulFertser> lindi-: what, Qi? No, not even in OE, i'm afraid.
May 08 16:49:07 <PaulFertser> lindi-: not good... But you already know what to do, just start it with -v switch or even strace it...
May 08 16:50:47 <PaulFertser> I guess i know why you talk about it. You hit WSOD and decided to try to reproduce it with Qi as it anecdotically solves the issue. But you didn't find a proper package and got distracted by preparing a package :)))
May 08 16:50:48 <lindi-> PaulFertser: -L/usr/lib/gcc/arm-linux-gnueabi/4.3/`ls /usr/lib/gcc/arm-linux-gnueabi/4.3`  is the problem. the directory has more than one file
May 08 16:51:09 <lindi-> PaulFertser: well, i'd hate to run software I can't modify easily :)
May 08 16:51:56 <PaulFertser> lindi-: (`ls...) yeah, looks like a bug
May 08 16:52:05 <lindi-> If I can't "apt-get source foo" on any debian system then the I need much more motivation to start making a small change
May 08 17:04:44 <lindi-> PaulFertser: seems it built. now i'll just sudo cat /dev/mtd1 > /home/lindi/mtd1.uboot.backup and then sudo nandwrite /dev/mtd1 image/qi-s3c2442-master_9ef7754b8243457c.udfu?
May 08 17:18:20 <PaulFertser> lindi-: no!
May 08 17:18:28 <PaulFertser> lindi-: flasherase_all first
May 08 17:19:02 <PaulFertser> lindi-: and i can't say for sure if simple cat will reliably backup the old version.
May 08 17:20:47 <DocScrutinizer> lindi-: I'd use dd anyway
May 08 17:21:23 <lindi-> DocScrutinizer: who's dd different?
May 08 17:22:01 <DocScrutinizer> dunno, never bothered to find out ;-)
May 08 17:22:35 <DocScrutinizer> think cat is mangling some ascii ctrl
May 08 17:23:30 <PaulFertser> lindi-: there's also nanddump ;)
May 08 17:24:19 <lindi-> DocScrutinizer: gnu cat is binary clean
May 08 17:29:41 <CIA-53> freesmartphone.org: 03sudharsh 07framework * r7e04d48f6fab 10/framework/subsystems/otimed/otimed.py: otimed: Fix a couple of tiny typos
May 08 17:31:55 <lindi-> PaulFertser: hmm, which option causes the kernel not print messages on bootup? loglevel=4?
May 08 17:32:08 <PaulFertser> lindi-: yes
May 08 17:32:31 <PaulFertser> lindi-: but if you use /boot/append-GTA02 with loglevel=8 you'll still be able to see it.
May 08 17:32:46 <lindi-> PaulFertser: does that affect even "echo foo" from init scripts?
May 08 17:33:01 <PaulFertser> lindi-: no, that's the different issue
May 08 17:33:03 <lindi-> I thought it was only for kernel messages
May 08 17:33:15 <lindi-> PaulFertser: any idea what supresses those?
May 08 17:34:17 <PaulFertser> lindi-: probably console=tty0 should be the last one to be used by default.
May 08 17:35:12 <lindi-> PaulFertser: aha, their order matters?
May 08 17:35:35 <lindi-> I was just running diff against old and new and was stupid enough to ignore order :)
May 08 17:36:28 <PaulFertser> lindi-: looks like it matters, not sure.
May 08 17:39:02 <lindi-> PaulFertser: so it seems
May 08 17:40:53 <lindi-> ok, i can still caused WSOD and total lockup (not even debug board can halt the cpu anymore) with Qi if I try to use gdb to debug X :P
May 08 17:40:53 <selke> hi, anyone know how to set an other keyboard on shr? i want to use qwo as keyboard...
May 08 17:41:13 <lindi-> (I was suspecting it had nothing to do with boot loader but had to check)
May 08 17:42:23 <PaulFertser> lindi-: well, it seems it's due to the NWAIT thing :(
May 08 17:42:32 <lindi-> PaulFertser: DocScrutinizer: you wouldn't know anybody whom I should specifically ask about this debug board issue? I sent mail to kernel list a few days ago already
May 08 17:42:38 <lindi-> PaulFertser: NWAIT?
May 08 17:43:20 <lindi-> that's some pin of jtag according to google
May 08 17:44:29 <PaulFertser> lindi-: no, that's a line connected connected to glamo by which glamo can really lock the processor.
May 08 17:45:02 <lindi-> PaulFertser: argh
May 08 17:45:09 <lindi-> no way :(
May 08 17:49:16 <lindi-> PaulFertser: if I set a breakpoint to GlamoRestoreHW function of xf86-video-glamo and issue "chvt 1" then gdb stops at the breakpoint and prints nice backtrace but after a few seconds everything stops
May 08 17:49:55 <lindi-> PaulFertser: which sounds scary. if userland (xf86-video-glamo is userland) is too slow the whole CPU will lockup
May 08 17:50:16 <PaulFertser> lindi-: glamo is scary
May 08 17:51:54 <PaulFertser> lindi-: KMS will (semi-)solve that
May 08 17:53:13 <lindi-> PaulFertser: it still won't make placing breakpoints with debug board very safe  but that isn't safe always anyway
May 08 17:55:48 <CIA-53> freesmartphone.org: 03seba.dos1 07framework * r88583fb8a39f 10/framework/subsystems/opimd/pimb_sim_contacts_fso.py: opimd: SIM_Contacts_FSO: don't block whole frameworkd when deleting or updating SIM contact
May 08 17:55:48 <CIA-53> freesmartphone.org: 03seba.dos1 07framework * r3ca073dcfbfd 10/framework/subsystems/opimd/pimd_contacts.py: opimd: Contacts: call upd_contact only once for every used backend
May 08 17:56:15 <lindi-> PaulFertser: ok, in that light KMS sounds very very important
May 08 17:56:47 <PaulFertser> lindi-: try to convince Weiss then ;)
May 08 17:56:58 <PaulFertser> And larsc :)
May 08 20:29:12 <CIA-53> freesmartphone.org: 03seba.dos1 07framework * r668dde7d8104 10/framework/subsystems/opimd/pimb_sim_contacts_fso.py: opimd: SIM_Contacts_FSO: fix Update() and Delete()
May 08 20:49:01 <CIA-53> freesmartphone.org: 03mickey 07cornucopia * r5f3d4a73f2de 10/libfsoframework/ (3 files in 2 dirs): fsoframework: first version of kobjectnotifier now works
May 08 21:21:01 <CIA-53> freesmartphone.org: 03mickey 07specs * r150e31fbc841 10/common/org.freesmartphone.xml.in: add org.freesmartphone.xml.in for common enumerations and errors
May 08 21:25:52 <CIA-53> freesmartphone.org: 03mickey 07specs * rb595bb2e72be 10/ (Makefile common/Makefile common/org.freesmartphone.xml.in): specify common dbus errors in the org.freesmartphone domain
May 08 21:44:33 <lloyd_fedora> neotool give an error "No DFU capable USB device found" after flashing ??
May 08 22:22:14 <DocScrutinizer> http://people.openmoko.org/joerg/bootsplash/
May 09 00:35:11 <Blu3> current shr unst kernel is horking and going dicks up
May 09 00:54:49 <Blu3> it strikes me that the older kernels also did this. when i'm running gprs/pidgin, there must be a race condition causing an oops.  during a suspend/resume event and intermittent but frequent use, it's oopsing
May 09 01:23:01 <slaxxer> i cant get a ring   but the sound works for a phone
May 09 01:23:15 <slaxxer> vibrator works also
**** ENDING LOGGING AT Sat May 09 02:59:57 2009