**** BEGIN LOGGING AT Sat Jun 13 02:59:57 2009
Jun 13 03:00:02 <gnutoo> hi, is there anyone awake?
Jun 13 03:00:33 <gnutoo> what are the good asound.conf and state files to make voip work?
Jun 13 03:18:17 <Toaster`> DocScrutinizer: re: your earlier... Backup is there, but the required config files are missing
Jun 13 03:18:58 <DocScrutinizer> Toaster`: aah, ok
Jun 13 03:19:01 <Toaster`> DocScrutinizer: re: your earlier... shr_backup is there, but the required config files are missing, so it fails to load
Jun 13 03:19:29 <Toaster`> arg, enter got in there unwantedly
Jun 13 03:19:33 <gnutoo> hi,what states+asound.conf files do I need for sip?
Jun 13 03:20:10 <Toaster`> the last change I commited adds a check for the config files and presents an obvious error message
Jun 13 03:21:30 <Toaster`> Hopefully the config files will be added to the ipk soon
Jun 13 03:21:45 <DocScrutinizer> gnutoo: in /usr/share/shr... there is a voip-handset.state
Jun 13 03:23:28 <gnutoo> DocScrutinizer, I hear no sound
Jun 13 03:24:41 <DocScrutinizer> gnutoo: hmm, I never tested nor checked that statefile. Sorry
Jun 13 03:24:51 <gnutoo> DocScrutinizer, and  http://wiki.openmoko.org/wiki/Linphone  's only work for earing the sound not when you talk
Jun 13 03:25:05 <gnutoo> I'm stuck with state files
Jun 13 03:25:16 <gnutoo> ok 
Jun 13 03:25:23 <gnutoo> thanks anyway
Jun 13 03:26:18 <gnutoo> I'd like to make voip work with this setup: linphonec->openvpn-client->GPRS->openvpn-server->asterisk-server
Jun 13 03:27:12 <Toaster`> ok, I'm off.  Watching the hockey game took a lot out of me.  Night all
Jun 13 03:28:00 <gnutoo> does someone have an idea on where I should search 
Jun 13 03:28:01 <gnutoo> ?
Jun 13 04:43:36 <Gnutoo> I'll update SHR
Jun 13 04:43:55 <Gnutoo> becuase I have issues with GPRS
Jun 13 06:44:44 <PaulFertser> raster: hey :)
Jun 13 06:44:54 <PaulFertser> raster: haven't seen you for ages ;)
Jun 13 06:46:04 <raster> PaulFertser: yo
Jun 13 06:46:08 <raster> i'vebeen travelling
Jun 13 06:46:12 <raster> a lot of the time netless
Jun 13 06:46:21 <raster> i'm in singsapore atm almost back home
Jun 13 06:47:03 <PaulFertser> raster: home sweet home :) Nice :)
Jun 13 06:47:27 <raster> yeah.i'm looking forward to it. been away for 6 weeks. i need clean undies.
Jun 13 06:47:29 <PaulFertser> raster: i wanted to talk to you about battery gadget once again. Sorry, but i'm still not sure it does exactly the right thing.
Jun 13 06:47:35 <PaulFertser> LOL
Jun 13 06:48:34 <PaulFertser> raster: i imagine a device with main and backup battery, and backup battery is 1/10th of capacity of main. I'd expect battery gadget to show percentage of the whole capacity available, not an average of two batteries.
Jun 13 06:49:00 <raster> that is exaclyt what it does
Jun 13 06:49:05 <raster> well what batget does
Jun 13 06:49:21 <PaulFertser> raster: and in case capacity is unavailable it should be considered zero. So the power_supply that is USB but hal misrepresents it as a battery shouldn't affect anything as it has no capacity?
Jun 13 06:49:29 <raster> and also the hal api *IF* hal exposes the capacity and current charge levels (as opposed to a percentage only)
Jun 13 06:49:33 <PaulFertser> And by capacity i mean energy, not percentage.
Jun 13 06:50:29 <raster> it's hal's problem it exposes it as a battery although it blatantly is not
Jun 13 06:50:54 <PaulFertser> Ok, but that battery has no capacity. Therefore it should be ignored in calculations, no?
Jun 13 06:51:45 <raster> the code basically adds up all capacities and all the charges
Jun 13 06:51:49 <raster> thaTS WHAT THE BATGET CODE DOES
Jun 13 06:52:25 <PaulFertser> I saw the code
Jun 13 06:52:32 <raster> oops baps
Jun 13 06:52:33 <raster> caps
Jun 13 06:52:55 <raster> the hal code tho adds up the percentages
Jun 13 06:55:44 <PaulFertser> raster: i'm talking about how you calculate percentage to show if gadget is using hal. 
Jun 13 06:56:13 <PaulFertser> raster: hal doesn't expose anything for that usb "battery". So it should be ignored in calculations anyway.
Jun 13 06:57:32 <DieMumie1> Weiss: hm I am looking at the WAIT_RENDERING ioctl now. I think a combination of idle interrupt and an evil read pointer heuristic - hopefully never required - to wake up clients in a wait_queue will work without any busy waiting. 
**** ENDING LOGGING AT Sat Jun 13 07:19:27 2009

**** BEGIN LOGGING AT Sat Jun 13 07:20:00 2009
Jun 13 07:47:05 <khiraly11> doh
**** ENDING LOGGING AT Sat Jun 13 08:24:55 2009

**** BEGIN LOGGING AT Sat Jun 13 08:26:00 2009
Jun 13 08:26:01 <Frank1> how can i compile a single .c file from a bigger build to check if it compiles at all without starting bitbake clean and build always for the whole thing?
Jun 13 08:26:48 <PaulFertser> Frank1: probably "bitbake -c devshell packetname"?
**** ENDING LOGGING AT Sat Jun 13 08:27:09 2009

**** BEGIN LOGGING AT Sat Jun 13 08:27:40 2009
Jun 13 08:29:35 <Frank1> @paulfertser,packetname is the name of the recipe file?
Jun 13 08:29:57 <PaulFertser> Frank1: yes
Jun 13 08:30:36 <Frank1> i trying to avoid this, do not always want to do the whole e-wm when i just want to check the syntax of my file
Jun 13 08:30:43 <PaulFertser> Frank1: there's also an OE configuration variable that determines which terminal will be started with the devshell. You might want to tweak that (or not if gnome-terminal is available).
Jun 13 08:31:06 <PaulFertser> Frank1: make will make only the stuff you changed.
Jun 13 08:31:13 <PaulFertser> Frank1: and in fact you can do even without the devshell.
Jun 13 08:31:38 <PaulFertser> Frank1: just remove the compile stamp and do bitbake -c compile again. It will recompile only what was changed.
Jun 13 08:31:55 <Frank1> where can i remove the compile stamp?
Jun 13 08:32:03 <PaulFertser> Frank1: rm it :)
Jun 13 08:32:08 <PaulFertser> Frank1: tmp/stamps/...
Jun 13 08:32:24 <PaulFertser> You'll find it, don't worry ;)
Jun 13 08:35:44 <Frank1> yes i found it, currently checking wether it does like i want it
Jun 13 08:39:05 <Ainulindale> Good morning guys
Jun 13 08:41:17 <F4t> anyone here can help me with ethernet plugin usage with connman via dbus?
Jun 13 08:49:21 <Gnutoo> hi, I've the following error: "No buffer space available (code=105)" in openvpn ...I tried to do like in the FAQ of openvpn but it didn't solve the problem...my connection is the following: linphonec->openvpn->gprs->...<-pppoe<-openvpn<-asterisk...does someone knows well ppp things such as MTU and similar? --mtu-test returned a value superior to 1500....the problem occur only when I get a sip call....
Jun 13 08:49:32 <Gnutoo> and that I answer
Jun 13 09:01:43 <Ainulindale> mickey|DepecheMo: Depeche Mode is gay!
Jun 13 09:06:15 <max_posedon> Gnutoo, I suggest test with mtr
Jun 13 09:06:28 <max_posedon> I don't think that 1500 correct value
Jun 13 09:07:02 <onen|openBmap> any debian FR user here?
Jun 13 09:07:18 <Gnutoo> max_posedon, ok thanks a lot...what is MTR ?
Jun 13 09:07:49 <max_posedon> kinda traceroute
Jun 13 09:08:09 <Gnutoo> ah ok my traceroot thanks a lot
Jun 13 09:08:30 <max_posedon> no-no, mtr tests packets
Jun 13 09:08:44 <max_posedon> it will show you, if some router split your packet
Jun 13 09:08:51 <max_posedon> (e.g. mtu too big)
Jun 13 09:10:10 <Gnutoo> ok thanks a lot
Jun 13 09:23:47 <bumbl> hmm strange
Jun 13 09:23:54 <bumbl> yesterday the moko worked
Jun 13 09:23:55 <bumbl> today
Jun 13 09:24:07 <bumbl> with the same settup i get a nsod on resume
Jun 13 09:25:52 <mrmoku> mokos work only on even days correctly ;)
Jun 13 09:26:02 <bumbl> ah then
Jun 13 09:26:11 <mrmoku> tomorrow all will be fine again:P
Jun 13 09:26:20 <bumbl> hehe
Jun 13 09:26:49 <bumbl> well
Jun 13 09:32:54 <Frank1> somebody knows where the e_mod_gad_gsm.c is build, i thought it would be build with the e-wm recipe and according to log it builds "all" in illume subfolder where also the e_mod_gad_gsm.c is located, but it runs smoothly and i made some compiler errors in that file so it should exit with problems :(
Jun 13 09:35:49 <bumbl> mrmoku: as it seems the moko wants to drive me crazy ;9
Jun 13 09:37:08 <khiraly11> o yeah
Jun 13 09:37:41 <mrmoku> bumbl: yeah... that's the plan... create an open-source device which makes all open-source geeks commit suicide
Jun 13 09:37:48 <mrmoku> guess who sponsored that ;)
Jun 13 09:38:09 <Frank1> no no, this would be much to predictable
Jun 13 09:40:05 <bumbl> Frank1: aliens have sent this device to the earth in order to force all the open source enthusiasts into madness and destroy the gsm network with the power of the buzz
Jun 13 09:43:34 <Gnutoo> max_posedon, mtr segfaults
Jun 13 09:43:55 <Gnutoo> max_posedon, so I'll see that later ...thanks a lot
Jun 13 09:44:01 <max_posedon> hm
Jun 13 09:44:15 <max_posedon> you can use neo as nat for your pc, and start mtr from PC
Jun 13 09:44:23 <Gnutoo> ok
Jun 13 09:44:28 <Frank1> somebody any suggestions to my questions up there with the e_mod_gad_gsm.c?
Jun 13 09:44:40 <Gnutoo> or maybe worse...
Jun 13 09:44:50 <Gnutoo> maybe that's not mtr that segfaults
Jun 13 09:45:00 <Gnutoo> becuase the shell is blocked after the sefault
Jun 13 09:45:20 <Gnutoo> maybe it's bash,ash or something like that
Jun 13 10:12:36 <Frank1> i got it running, compiling is failing, but this is fine
Jun 13 10:20:00 <Zorkman> stereo
Jun 13 10:20:04 <Zorkman> oeps :)
Jun 13 10:22:08 <Zorkman> Ainulindale: Zorkman: ask Ainulindale which of both is supposed to be loaded by intone ATM. either /usr/share/openmoko/scenarios/headset.state OR /usr/share/shr/scenarii/headset.state
Jun 13 10:31:45 <Zorkman> mm seems like i don't have a /usr/share/openmoko
Jun 13 10:39:11 <CIA-45> SHR: 03seba.dos1 07shr-settings * rdaf54999d3b7 10/ (data/po/shr-settings.pot shr_settings_modules/shr_splash.py): Revert "[splash] s/Themes/Theme/ on hoversel label"
Jun 13 10:40:19 <ptitjes> Ainulindale: ping
Jun 13 10:50:28 <DieMumie1> Weiss: ping..
Jun 13 10:51:12 <DieMumie1> Weiss: do you want the WAIT_RENDERING ioctl to wait for a certain refcount value? .. like 1 .. or until the gem object gets destroyed?
Jun 13 10:52:16 <Zorkman> btw1: shr people, on the latest unstable my "shutting down-splash screen" is distorted by the turning wheel of "shutting the phone down" when shutting down using illume top bar
Jun 13 10:52:41 <Zorkman> btw2: it would be better if we could specify a BOOTsplash and a SHUTDOWNsplas (two different images)
Jun 13 10:54:44 <Ainulindale> Zorkman: that's possible
Jun 13 10:54:50 <Ainulindale> and already the case
Jun 13 10:54:54 <Ainulindale> although the images are the same
Jun 13 10:56:27 <Zorkman> perfect
Jun 13 10:59:18 <ptitjes> Ainulindale: so what's up ?
Jun 13 11:00:26 <Weiss> DieMumie1: good point... '1' i guess.. it should wait until all rendering involving that object has completed. the owner of the object should hold the last reference
Jun 13 11:01:39 <Weiss> DieMumie1: maybe look at how some of the other drivers do it. but i think quite a lot of them just idle the CQ
Jun 13 11:09:52 <DieMumie1> Weiss: wait for idle? or forceh the CQ becoming idle?
Jun 13 11:09:54 <DieMumie1> -h
Jun 13 11:10:45 <DieMumie1> ah and why do you think that you need the wait-ioctl for text rendering? Glamo has alpha blending and 4bit-alpha font rendering..
Jun 13 11:10:56 <DieMumie1> on 565 display the latter should be sufficient..
Jun 13 11:11:42 <mrmoku> ptitjes: you wrong... me right ;)
Jun 13 11:11:59 <ptitjes> hummm ??
Jun 13 11:12:12 <mrmoku> wait for the mail to hit your inbox :P
Jun 13 11:12:13 <ptitjes> mrmoku: but what are you talking about ?? :)
Jun 13 11:12:43 <ptitjes> mrmoku: extensibility
Jun 13 11:12:49 <ptitjes> that is as simple as that
Jun 13 11:12:56 <ptitjes> no code change in ophonekitd
Jun 13 11:13:05 <ptitjes> just configuration change to support a new pim domain
Jun 13 11:13:19 * mrmoku won't discuss it here ;)
Jun 13 11:13:26 <ptitjes> :)
Jun 13 11:13:28 <khiraly11> ptitjes: yeah, looks like paroli does not use threading at the moment
Jun 13 11:13:29 <mrmoku> ptitjes: waiting for your mail :P
Jun 13 11:13:48 <khiraly11> but I'm definietly not the right person to implement proper threading in paroli
Jun 13 11:14:21 <ptitjes> khiraly11: oh so you are Laszlo ?
Jun 13 11:14:32 <khiraly11> oops
Jun 13 11:14:40 <khiraly11> shit my incognito
Jun 13 11:14:51 <ptitjes> héhéhé
Jun 13 11:16:36 <khiraly11> hmm
Jun 13 11:16:55 <bumbl> well the paroli distribution needs some love though
Jun 13 11:17:17 <ptitjes> that won't be me :D
Jun 13 11:17:43 <ptitjes> python is sh.....
Jun 13 11:18:03 <ptitjes> khiraly11: but I can tell you that way:
Jun 13 11:20:27 <ptitjes> Thread.create((data) => { //do some stuff return null;}, null, true));
Jun 13 11:20:38 <Deubeuliou> hi everybody
Jun 13 11:20:55 <Deubeuliou> hello Didier
Jun 13 11:20:59 <ptitjes> hi David
Jun 13 11:21:21 <ptitjes> khiraly11: move to Vala :D
Jun 13 11:21:35 <khiraly11> ptitjes: Vala does not solve the problem
Jun 13 11:21:44 <khiraly11> the same exists for it too
Jun 13 11:21:51 <khiraly11> (creating thread)
Jun 13 11:22:04 <khiraly11> ptitjes: the problem is, that we should put all the gui messing stuff in its own separate thread
Jun 13 11:22:16 <khiraly11> which needs some throurough thinking
Jun 13 11:22:16 <ptitjes> khiraly11: nah
Jun 13 11:22:38 <ptitjes> khiraly11: I already told that 
Jun 13 11:22:45 <khiraly11> yepp
Jun 13 11:22:55 <khiraly11> and Im not the right person to mess with threads
Jun 13 11:22:59 <ptitjes> every long operation has to be run in its own thread
Jun 13 11:23:00 <DocScrutinizer> ptitjes: rewrite paroli in eerrrr lisp? prolog?
Jun 13 11:23:04 <ptitjes> it is easy
Jun 13 11:23:15 <khiraly11> DocScrutinizer: every program should be written in Fortran!
Jun 13 11:23:25 <DocScrutinizer> COBOL
Jun 13 11:24:13 <khiraly11> ptitjes: not as easy as it looks in that case. Because there is no *one single* operation which takes time, but launching some 10-12 separate things, and each of them takes a little time ...
Jun 13 11:24:36 <DocScrutinizer> I never understood why we stopped using batch and basic ;-)
Jun 13 11:24:37 <khiraly11> and the end the gui is not responsible. Which was not obvious until animation
Jun 13 11:24:52 <khiraly11> s/the end/at the end/
Jun 13 11:24:52 <bzzbot> khiraly11 meant: and at the end the gui is not responsible. Which was not obvious until animation
Jun 13 11:24:55 <ptitjes> khiraly11: the problem with people doing python is that they think as python is easy to write, they don't have to think along standard software design
Jun 13 11:25:09 <khiraly11> ptitjes: hehh
Jun 13 11:25:15 <ptitjes> khiraly11: Vala is making programming simple and efficient, generates fast code
Jun 13 11:25:34 <ptitjes> khiraly11: and APIs are as simple as doing python
Jun 13 11:25:53 <ptitjes> using threads in vala is as easy as not using them :)
Jun 13 11:26:11 <DocScrutinizer> ptitjes: (py) couldn't agree more
Jun 13 11:26:16 <khiraly11> its not visual basic to apply some prejudiction
Jun 13 11:26:18 <ptitjes> :)
Jun 13 11:26:18 <Weiss> DieMumie1: text was just an example of what might go on... if _everything_ is accelerated, then synchronisation is hardly needed at all. but that's not an achievable goal even for modern hardware..
Jun 13 11:26:32 <khiraly11> ptitjes: so vala is like java;)
Jun 13 11:26:39 <Weiss> DieMumie1: yep, i mean they wait for the CQ to be completely idle. i think we can do one step better..
Jun 13 11:26:45 <ptitjes> khiraly11: yep with the power of C
Jun 13 11:27:15 <khiraly11> I hate java, and the assumtion, that everybody should use threads
Jun 13 11:27:19 <khiraly11> at least 50 threads
Jun 13 11:27:20 <khiraly11> ;)
Jun 13 11:27:34 <ptitjes> khiraly11: who said you have to use 50 threads
Jun 13 11:27:50 <ptitjes> khiraly11: if you are implementing a web server, then you'll do a pool of worker threads
Jun 13 11:27:56 <ptitjes> whatever the language you use
Jun 13 11:28:08 <ptitjes> you just can't do a GUI app with one single thread
Jun 13 11:28:14 <ptitjes> this is just dumb
Jun 13 11:28:17 <khiraly11> ptitjes: I like this way more. Do not use threads until it is absolutely necessary.
Jun 13 11:28:26 <khiraly11> and after do the right research, and implement properly
Jun 13 11:28:34 <ptitjes> khiraly11: but this is absolutely necessary for gui apps
Jun 13 11:28:37 <khiraly11> not the other way around (as in java world)
Jun 13 11:28:46 <khiraly11> ptitjes: not always;)
Jun 13 11:28:48 <ptitjes> khiraly11: GUI apps are not scripts
Jun 13 11:29:13 <ptitjes> people doing python just starts by making scripts (which is IMHO the only purpose of python BTW)
Jun 13 11:29:46 <ptitjes> then when they start making GUI apps they don't look further understanding they need to do like every people implementing GUI apss in any language
Jun 13 11:29:52 <ptitjes> use threads
Jun 13 11:30:02 <ptitjes> khiraly11: what is the problem
Jun 13 11:30:15 <DieMumie1> if you say to a groups of developers "use threads"
Jun 13 11:30:17 <ptitjes> threads are just function that runs in another || context
Jun 13 11:30:30 <DieMumie1> you get software the needs 40 threads..
Jun 13 11:30:35 <DieMumie1> and that means 40 stacks
Jun 13 11:31:06 <DieMumie1> and 40 different synchronisation ideas
Jun 13 11:31:09 <DocScrutinizer> so what? keep your stacks small then
Jun 13 11:31:18 <khiraly11> ptitjes: I implemented a software in past using python and gtk. It controlled remotely a crane. I didnt use thread, I managed to avoid it till the end
Jun 13 11:31:19 <khiraly11> ;)
Jun 13 11:31:24 <ptitjes> DieMumie1: I don't know the developers you work with, but I would hate to people around just thinking no threads or an infinity
Jun 13 11:31:28 <DieMumie1> or implement something like the kernel bh .. 
Jun 13 11:31:30 <ptitjes> world is not black or white
Jun 13 11:31:35 <DieMumie1> task qeueues.. 
Jun 13 11:31:47 <DieMumie1> and you can live with one or two threads
Jun 13 11:31:48 <khiraly11> so if you implement a simple gui app, you do not need threads
Jun 13 11:31:48 <ptitjes> DieMumie1: aka worker threads
Jun 13 11:32:01 <khiraly11> just when you do lot of processing, or networking background
Jun 13 11:32:28 <khiraly11> like in dbus, which is networking;)
Jun 13 11:32:29 <ptitjes> khiraly11: but you GUI rendering was blocked by the operations running in you callback
Jun 13 11:32:33 <DieMumie1> bah 
Jun 13 11:32:34 <ptitjes> khiraly11: you can't avoid it
Jun 13 11:32:39 <ptitjes> don't look further
Jun 13 11:32:43 <ptitjes> you NEED threads
Jun 13 11:32:44 <DieMumie1> even for dbus you do not need a separate thread
Jun 13 11:32:50 <DieMumie1> select
Jun 13 11:32:53 <khiraly11> ptitjes: doh. You can. Just need to stay below 100ms. Nobody will notice
Jun 13 11:32:55 <ptitjes> DieMumie1: that is wrong
Jun 13 11:33:28 <ptitjes> DieMumie1: DBus callbacks are run through an event fired from the main loop
Jun 13 11:33:42 <ptitjes> if you do long operations in your DBus callback then you need threads
Jun 13 11:33:58 <ptitjes> else you will block any further concurent dbus signaling
Jun 13 11:34:03 <khiraly11> anyway I understand, that we need threads in that very case. Im not fighting against
Jun 13 11:34:09 <DieMumie1> hm for a seti@home dbus service?
Jun 13 11:34:26 <khiraly11> (although cleverly injecting some code in e's loop, we could avoid it. Just dont know e much)
Jun 13 11:34:27 <ptitjes> BTW this is why framework does so badly handle SMS and call handling at the same time
Jun 13 11:34:53 <ptitjes> DieMumie1: oh yes for sure
Jun 13 11:35:07 <ptitjes> DieMumie1: you have to give the handle to another thread to make the processing
Jun 13 11:35:29 <ptitjes> DieMumie1: whatever the processing is that takes more time than a few miliseconds
Jun 13 11:35:48 <ptitjes> basic software design here guys
Jun 13 11:36:47 <ptitjes> why do you think than proper GUI frameworks (like Eclipse) have API support like:
Jun 13 11:36:59 <ptitjes> Plateform.runAsync(Runnable)
Jun 13 11:36:59 <DieMumie1> o_O
Jun 13 11:37:21 <ptitjes> and UIPlateform.busyCursorWhile(Runnable)
Jun 13 11:37:54 <DocScrutinizer> ptitjes: system_architecture seems to be the most demanding part of sw-development ;-)
Jun 13 11:38:07 <DieMumie1> ... whatever ..  "< ptitjes> use threads" is just the root of an evil.. use #cores+1 thread and live with it..
Jun 13 11:38:17 <ptitjes> the latter just does run the runnable in its own thread, concurrently launch a timeout, run the UI animation for the cursor
Jun 13 11:38:42 <ptitjes> and when the timeout occurs changes to a non-modal progress display of the operation
Jun 13 11:38:47 <ptitjes> all this requires threads
Jun 13 11:39:02 <DieMumiee> i just have to kick developers regular;y for opening up new threads for no reason but laziness
Jun 13 11:39:34 <DieMumiee> or you get used to an async style of programming
Jun 13 11:39:46 <ptitjes> DieMumiee: sorry didn't get your sentence
Jun 13 11:40:11 <DieMumiee> which one?
Jun 13 11:40:23 <ptitjes> Plateform.runAsync() is just giving handle to a worker thread
Jun 13 11:40:36 <ptitjes> thus you would normaly have the main + 1 thread
Jun 13 11:40:38 <ptitjes> for sure
Jun 13 11:41:01 <ptitjes> DieMumie1>
Jun 13 11:41:01 <ptitjes> ... whatever ..  "< ptitjes> use threads" is just the root of an evil.. use #cores+1 thread and live with it..
Jun 13 11:41:11 <DocScrutinizer> DieMumiee: "which one?s is a silly question for "i missed asome xxx"
Jun 13 11:42:40 <DocScrutinizer> ok, but I seem not to get the meaning of "I didn't get" ;o) nevermind
Jun 13 11:42:42 <SHR-WugBranler> tracfeed: Ticket #516 (settings-phone - select GSM operators doesn't check error) updated <http://trac.shr-project.org/trac/ticket/516#comment:2>
Jun 13 11:43:17 <DieMumiee> ptitjes: I wanted to say that the statement "use threads" is not a good idea. I then wanted to say that one should try to restrict on number of cores plus one thread
Jun 13 11:43:27 <DocScrutinizer> still waiting for coffee to kick in
Jun 13 11:43:54 <ptitjes> DieMumiee: "use threads" does not mean "use threads in a bad way"
Jun 13 11:43:55 <DieMumiee> ptitjes: i believe we are saying the same thing 
Jun 13 11:44:03 <ptitjes> I do think so too
Jun 13 11:44:22 <DieMumiee> and i wanted to add that with X11 it is also possible to have only one thread
Jun 13 11:44:22 <ptitjes> "use threads" does just "you can't avoid it" :)
Jun 13 11:44:42 <DieMumiee> given that all your libraries provide a reactor style socket 
Jun 13 11:44:57 <ptitjes> DieMumiee: and sockets do the fork
Jun 13 11:45:00 <DieMumiee> then only need one tight select loops..
Jun 13 11:45:03 <DieMumiee> -s
Jun 13 11:45:06 <DieMumiee> +you
Jun 13 11:45:09 <ptitjes> DieMumiee: thus you use threads without even knowing
Jun 13 11:45:36 <ptitjes> which IMHO is not very safe
Jun 13 11:45:46 <DieMumiee> o_O
Jun 13 11:45:48 <ptitjes> (using threads without even knowing)
Jun 13 11:46:03 <DieMumiee> so using sockets is not safe?
Jun 13 11:46:31 <DieMumiee> Ah i see where you misunderstood me .. 
Jun 13 11:46:45 <ptitjes> DieMumiee: didn't say that. just want to point that one must know what happen in the background to use them efficiently
Jun 13 11:46:59 <DocScrutinizer> yeahyeah. If I had to service DMA IO of 7 harddrives, I for sure wouldn't go with less threads than 7, even on a singlecore system
Jun 13 11:47:14 <DieMumiee> libraries like the postgres client library or X11 or directfb or .. provide access to the underlying socket 
Jun 13 11:47:16 <SHR-WugBranler> tracfeed: Ticket #516 (settings-phone - select GSM operators doesn't check error) updated <http://trac.shr-project.org/trac/ticket/516#comment:3>
Jun 13 11:47:40 <DieMumiee> then the user of the library can stick them all into one call of select and wait for any event to happen
Jun 13 11:48:10 <DieMumiee> and thats not using threads without knowing..
Jun 13 11:48:35 <DocScrutinizer> but you need to build your own arbiter
Jun 13 11:49:06 <DocScrutinizer> which is already there in kernel
Jun 13 11:49:13 <ptitjes> and wait/awake
Jun 13 11:49:18 <ptitjes> and do locking appropriately
Jun 13 11:49:27 <DocScrutinizer> nasty
Jun 13 11:50:00 <ptitjes> or correct lock-fre programming
Jun 13 11:50:26 <DocScrutinizer> and still you got problem of reentrance
Jun 13 11:51:03 <ptitjes> stasks
Jun 13 11:51:08 <ptitjes> stacks
Jun 13 11:51:17 <DocScrutinizer> i.e. higher prio wait expires you need to stop whatever you are doing
Jun 13 11:51:46 <DocScrutinizer> how to handle THAT with a central wait dispatcher
Jun 13 11:51:47 <DocScrutinizer> ?
Jun 13 11:52:10 <ptitjes> DocScrutinizer: this is must be handled on case by case
Jun 13 11:52:19 <DieMumiee> DocScrutinizer: with interrupts? 
Jun 13 11:52:23 <ptitjes> sometime re-design just can remove rentrancy
Jun 13 11:52:32 <khiraly11> hmm
Jun 13 11:54:09 <DocScrutinizer> interrupt *is* a new thread. and you mustn't interrupt centrally dispatched workers by this very dispatcher. this is reentrance
Jun 13 11:56:26 <DieMumiee> well i just carry my daughter around and now i need coffee.. this is one of theses occaisons where just cannot drop here .. i have to finish carrying her around..
Jun 13 11:56:30 <DieMumiee> -e
Jun 13 11:56:33 <CIA-45> SHR: 03seba.dos1 07shr-settings * r1e0d2bb71900 10/shr_settings_modules/shr_gsm.py: [GSM] Operators: more error handling
Jun 13 11:56:43 <DieMumiee> ok now i get some coffee 
Jun 13 11:57:20 <DocScrutinizer> dos1: wow, seems operators settings can use a lot of error handling ;-P
Jun 13 11:58:05 <khiraly11> I need to research what is the best for python threading
Jun 13 11:58:29 <khiraly11> would be nice to find an example application what use threading and e already
Jun 13 11:58:56 <ptitjes> DocScrutinizer: except for full preemption
Jun 13 11:59:14 <ptitjes> DocScrutinizer: just like do RT sources
Jun 13 11:59:34 <dos1> DocScrutinizer: maybe i should write "more work on error handling" ;D
Jun 13 11:59:55 <DieMumiee> i lost my daugther ..
Jun 13 12:00:01 <DieMumiee> she was in the other room 
Jun 13 12:00:13 <DieMumiee> too bad that i cannot fork .. 
Jun 13 12:00:38 <DocScrutinizer> there's rumour raster *can* fork ;-)
Jun 13 12:00:48 <ptitjes> héhé
Jun 13 12:01:03 <DocScrutinizer> ptitjes: I'm always thinking RT, it seems
Jun 13 12:01:14 <ptitjes> :)
Jun 13 12:01:19 <DocScrutinizer> ptitjes: as it never hurts, even on non RT systems
Jun 13 12:01:25 <ptitjes> sure
Jun 13 12:01:40 <ptitjes> I do think the same since I programmed with Jack
Jun 13 12:02:14 <ptitjes> and also making some MidiBox C programming helped to rather enhance my RT skills
Jun 13 12:03:00 <DocScrutinizer> ptitjes: most devel don't even know definition of RT, what simply is "a *defined* maximum response time to any event". Nothing about speed at all
Jun 13 12:03:47 <DocScrutinizer> ptitjes: it's simply about not blocking tasks by other tasks
Jun 13 12:04:05 <DocScrutinizer> ptitjes: which directly leads to proper usage of threads
Jun 13 12:04:32 <DocScrutinizer> ptitjes: and IRQs
Jun 13 12:04:33 <ptitjes> DocScrutinizer: yep but you only discover that when you have more hardware response constraints
Jun 13 12:04:43 <DocScrutinizer> sure
Jun 13 12:05:07 <DocScrutinizer> I started programming on a 1MHz Z80 with 4k of RAM
Jun 13 12:05:17 <ptitjes> héhé
Jun 13 12:05:27 <ptitjes> TI4 for me
Jun 13 12:05:43 <DocScrutinizer> and I implemented a tiny-basic on this hw platform ;-)
Jun 13 12:05:50 <ptitjes> wow nice
Jun 13 12:07:15 <DocScrutinizer> also a light-stylus to get CRT-position (yes, this system even had a screenbuffer (48*24char))
Jun 13 12:07:30 <DocScrutinizer> charset in ROM
Jun 13 12:12:11 <DocScrutinizer> on a system like this, it's quite natural you don't want to block IRQ or run into reentrance issues - esp if cpu needs to do all the some80-key kbd polling as well
Jun 13 12:13:15 <khiraly11> http://bethesignal.org/wp-content/uploads/2009/06/css-is-awesome-700x375.jpg
Jun 13 12:13:20 <khiraly11> hehe
Jun 13 12:13:21 <khiraly11> ;)
Jun 13 12:14:38 <ptitjes> hum what is CSS ?
Jun 13 12:14:39 <ptitjes> ;)
Jun 13 12:15:17 <DocScrutinizer> corruptes source sensor? ;)
Jun 13 12:15:31 <dos1> khiraly11: hehe :D
Jun 13 12:15:44 <ptitjes> isn't it for quiche eaters ?
Jun 13 12:16:04 <ptitjes> :D
Jun 13 12:16:09 <dos1> in polish "kaskadowe arkusze stylów" ;)
Jun 13 12:17:26 <bumbl> khiraly11: nice
Jun 13 12:19:03 <DocScrutinizer> (kbd polling) aah I remeber: *incl debouncing* ;-D
Jun 13 12:22:09 <dos1> DocScrutinizer: do you have ability to test some mdbus commands while in roaming?
Jun 13 12:23:13 <khiraly11> dos1: could you help me to test mickey's modification of bug #435?
Jun 13 12:23:18 <khiraly11> cant understand all of them
Jun 13 12:23:46 <dos1> khiraly11: he disabled ousaged from frameworkd and was using fsousaged from cornucopia (FSO 2.0)
Jun 13 12:24:00 <bumbl> dos1: sounds similar to the english phrase
Jun 13 12:24:26 <DocScrutinizer> dos1: I mustn't roam here in D usually, but for the rest of your question: yes, why not
Jun 13 12:24:50 <khiraly11> yepp, just I suppose I need to compile fsousaged
Jun 13 12:25:05 <dos1> bumbl: yep, only sheet -> arkusz doesn't sound similar ;)
Jun 13 12:25:45 <bumbl> bah those litte details don't matter ;p
Jun 13 12:25:55 <CIA-45> SHR: 03seba.dos1 07shr-settings * r7cbc77abeaae 10/shr_settings_modules/shr_gsm.py: [GSM] Operators: add "Automatic" option
Jun 13 12:26:11 <dos1> DocScrutinizer: mdbus -s org.freesmartphone.ogsmd /org/freesmartphone/GSM/Device org.freesmartphone.GSM.Network.GetStatus
Jun 13 12:26:14 <dos1> while in roaming
Jun 13 12:26:51 <DocScrutinizer> dos1: actually this might be an interesting testcase for you. Though t-kom is marked available, I usually get a reject when I try to roam to them
Jun 13 12:27:22 <dos1> hmm
Jun 13 12:27:58 <dos1> well, i can connect to only one provider here, all others are forbidden
Jun 13 12:28:41 <dos1> so i have limited testing ability :x
Jun 13 12:28:50 <DocScrutinizer> O2 and T-Kom have a indirect roaming agreement (via swisscom) to fill in the white gaps in O" network (t-kom is better). But they forbid whereever you don't absolutely need to do this
Jun 13 12:29:26 <DocScrutinizer> dos1: same here (except when I move to a location O2 has no coverage)
Jun 13 12:29:57 <DocScrutinizer> lemme ssh to FR, mompl
Jun 13 12:30:01 <dos1> DocScrutinizer: here Play has roaming agreement with Plus
Jun 13 12:30:24 <dos1> DocScrutinizer: i'm using Plus, but it's in only one side - Play users can connect to Plus where Play don't have coverage
Jun 13 12:31:02 <DocScrutinizer> that's exactly same as here with O2 and TKom
Jun 13 12:31:24 <DocScrutinizer> so you still need the mdbus output? ;-)
Jun 13 12:33:12 <dos1> DocScrutinizer: i need, but maybe i could find some Play SIM to test roaming
Jun 13 12:33:22 <dos1> DocScrutinizer: Plus users can't connect to Play network
Jun 13 12:33:28 <dos1> DocScrutinizer: and i'm Plus user
Jun 13 12:35:39 <DocScrutinizer> Using **pending_return in dbus_connection_send_with_reply_setup() without pending_setup is deprecated and strongly discouraged
Jun 13 12:35:43 <DocScrutinizer> hmmm :-/
Jun 13 12:36:10 <dos1> DocScrutinizer: that's due to that patch to new dbus
Jun 13 12:36:12 <khiraly11> DocScrutinizer: what program does return this?
Jun 13 12:36:39 <khiraly11> dos1: paroli has the same error. It is a warning directly from dbus?
Jun 13 12:37:18 <dos1> DocScrutinizer, khiraly11: that's warning from dbus due to bindings, which should be patched
Jun 13 12:37:47 <DocScrutinizer> khiraly11: mdbus
Jun 13 12:38:17 <khiraly11> ok. so nothing I should worry in paroli code;L)
Jun 13 12:50:20 <SHR-WugBranler> tracfeed: Ticket #511 (Position of Delete button) updated <http://trac.shr-project.org/trac/ticket/511#comment:1>
Jun 13 12:51:43 <Sup3rkiddo> hi all
Jun 13 12:51:45 <SHR-WugBranler> tracfeed: Ticket #514 (Feedback *needed* when sending an sms fails) updated <http://trac.shr-project.org/trac/ticket/514#comment:3>
Jun 13 12:54:21 <SHR-WugBranler> tracfeed: Ticket #517 (opkg lock file in wrong place) closed <http://trac.shr-project.org/trac/ticket/517#comment:1>
Jun 13 12:55:42 <CIA-45> SHR: 03seba.dos1 07shr-settings * rc2e778463796 10/data/po/shr-settings.pot: [data] update pot file
Jun 13 12:56:29 <ptitjes> mrmoku: this is steril discussion
Jun 13 12:56:44 <mrmoku> clean room discussion?
Jun 13 12:56:48 <mrmoku> well...
Jun 13 12:56:49 <ptitjes> mrmoku: I did not say we would have to, I say we don't have to
Jun 13 12:56:52 <ptitjes> but anyway
Jun 13 12:56:58 <ptitjes> I just said let's move on
Jun 13 12:57:03 <ptitjes> and don't talk about that yet
Jun 13 12:57:04 <DocScrutinizer> dos1: what's the story about green=left, red=right now?
Jun 13 12:57:10 <mrmoku> ptitjes: yep, read that after sending my mail :)
Jun 13 12:57:11 <ptitjes> mrmoku: but you did not answer about that
Jun 13 12:57:18 <ptitjes> frrrrrrrrr
Jun 13 12:57:29 <ptitjes> I want we setup something 
Jun 13 12:57:32 * mrmoku moved on ;)
Jun 13 12:57:36 <ptitjes> even if we make it evolve later
Jun 13 12:57:46 <ptitjes> we can't do the perfect stuff at once
Jun 13 12:57:48 <ptitjes> anyway
Jun 13 12:58:33 <DocScrutinizer> dos1: i.e: swap accept/reject, ok/cancel, etc
Jun 13 12:59:28 <ptitjes> same for you btw dos1 
Jun 13 12:59:31 <ptitjes> :)
Jun 13 12:59:41 <dos1> DocScrutinizer: isn't it done correctly now?
Jun 13 13:00:25 <DocScrutinizer> in dialer I still have OK aka accept right side
Jun 13 13:00:34 <CIA-45> SHR: 03seba.dos1 07shr-settings * rdb4319668e6f 10/ (3 files in 2 dirs): [backup][currentprofile][data] fix some gettext issues
Jun 13 13:01:26 <DocScrutinizer> dos1: aka "Call"
Jun 13 13:01:37 <DocScrutinizer> dos1: "Close" left side
Jun 13 13:02:44 <DocScrutinizer> dos1: active call I got "Release" left side
Jun 13 13:03:15 <DocScrutinizer> dos1: all this needs to be fixed to follow "green=left" pragma
Jun 13 13:05:39 <DocScrutinizer> dos1: and while you are about to edit dialer anyway - how about including function to open "recently dialed numbers" list to redial, when hitting "Call" on empty number field
Jun 13 13:09:20 <DocScrutinizer> well, have to buy some decent filets now. Gonna bbq this eve :-))
Jun 13 13:42:05 <CIA-45> SHR: 03seba.dos1 07shr-settings * r8f04c1e80828 10/data/po/pl/ (shr-settings.mo shr-settings.po): [data] po: update polish translation
Jun 13 13:42:44 <max_posedon> mrmoku, Hi!
Jun 13 13:42:54 <max_posedon> to requests about openssh
Jun 13 13:43:13 <max_posedon> 1. opkg install openssh should create /var/run/sshd
Jun 13 13:43:40 <max_posedon> 2. opkg install openssh should also *start* sshd
Jun 13 13:44:14 <TAsn> dos1, here?
Jun 13 13:44:37 <dos1> TAsn: there :D
Jun 13 13:44:40 <TAsn> :)
Jun 13 13:44:57 <TAsn> dos1, F4t just told me connman requires a plugin for dhcp
Jun 13 13:45:04 <dos1> yep
Jun 13 13:45:05 <TAsn> maybe that's why wifiman is broken ;)
Jun 13 13:45:12 <dos1> i think i have that plugin installed
Jun 13 13:45:19 <TAsn> I don't.
Jun 13 13:45:55 <TAsn> nvm, I do.
Jun 13 13:46:31 <TAsn> dos1, if it's a plugin
Jun 13 13:46:39 <TAsn> this probably means you have to explicitly ask for it
Jun 13 13:46:41 <TAsn> doesn't it?
Jun 13 13:46:44 <TAsn> do you?
Jun 13 13:47:33 <dos1> TAsn: well, wifi is plugin too
Jun 13 13:47:42 <dos1> and bluetooth, and ethernet, and loopback...
Jun 13 13:47:48 <dos1> and few others
Jun 13 13:47:51 <TAsn> I know
Jun 13 13:47:59 <TAsn> but you ask for them in dbus
Jun 13 13:48:02 <dos1> no
Jun 13 13:48:13 <TAsn> then how do you call connman?
Jun 13 13:48:28 <dos1> listdevices
Jun 13 13:48:36 <dos1> then i check all devices if they have type=wifi
Jun 13 13:49:47 <mrmoku> max_posedon, what about the config? the default config won't let you in ;)
Jun 13 13:50:18 <max_posedon> mrmoku, why? I just did this commands, and all is fine
Jun 13 13:50:38 <Gnutoo> max_posedon, hi again...how do I use mytraceroute?
Jun 13 13:50:50 <max_posedon> Gnutoo, mtr google.com
Jun 13 13:50:54 <mrmoku> max_posedon, because it does not allow empty passwords... so either you have to install a key... set a password or change the config
Jun 13 13:51:05 <DocScrutinizer> max_posedon: I don't think an install should always start the service automatically
Jun 13 13:51:06 <max_posedon> Gnutoo, and it isn't mytraceroute)
Jun 13 13:51:16 <Gnutoo> max_posedon, I've used mtr 10.0.0.1 and I didn't saw any info on MTU
Jun 13 13:51:25 <Gnutoo> max_posedon, ok
Jun 13 13:51:31 <max_posedon> Gnutoo, it will show you info about loss packets
Jun 13 13:51:59 <Gnutoo> max_posedon, ok it is variable 
Jun 13 13:52:20 <max_posedon> mrmoku, many people setupped password before, I use ssh-key-auth and have persistant home on microSD
Jun 13 13:52:42 <max_posedon> DocScrutinizer, sshd adds automatically to start during boot
Jun 13 13:53:02 <max_posedon> why it shouldn't start after install? you want ask users to reboot?!
Jun 13 13:53:12 <mrmoku> max_posedon, yep... but not all. And default config won't let you in again then :P
Jun 13 13:53:22 <DocScrutinizer> max_posedon: that's ok to create the correct init.d files and links to a set of runlevels
Jun 13 13:53:48 <DocScrutinizer> max_posedon: but you don't want the service to be started immediately after installation
Jun 13 13:54:02 <max_posedon> why?
Jun 13 13:54:21 <max_posedon> mrmoku, anyway, mkdir /var/run/sshd still important
Jun 13 13:54:51 <DocScrutinizer> especially ssh for example, user might want to continue his sessions instead of killing them by starting an alternative service
Jun 13 13:54:59 <max_posedon> DocScrutinizer, ha!
Jun 13 13:55:21 <max_posedon> right now you *can't* install openssh via ssh
Jun 13 13:55:41 <max_posedon> you have to remove dropbear with -force-depends
Jun 13 13:55:47 <max_posedon> mkdir /var/run/sshd
Jun 13 13:55:54 <max_posedon> install openssh
Jun 13 13:56:01 <max_posedon> start it manually 
Jun 13 13:56:04 <mrmoku> max_posedon, well... you can ;)
Jun 13 13:56:06 <max_posedon> it's a pita!
Jun 13 13:56:23 <mrmoku> what does it need /var/run/sshd for? the pid?
Jun 13 13:56:39 <DocScrutinizer> yup
Jun 13 13:56:51 <max_posedon> I'm not sure, but during start sshd want it
Jun 13 13:56:54 <DocScrutinizer> afaik
Jun 13 13:57:05 <max_posedon> for pid it still use /var/run/sshd.pid
Jun 13 13:57:07 <mrmoku> because it starts fine for me without manually creating something
Jun 13 13:57:46 <max_posedon> for me don't, tested 5min ago
Jun 13 13:57:50 <DocScrutinizer> -rw-r--r-- 1 root root 5 May 23 20:37 /var/run/sshd.init.pid
Jun 13 13:57:55 <mrmoku> hmm
Jun 13 13:58:01 <DocScrutinizer> on susue
Jun 13 13:58:08 <DocScrutinizer> OpenSUSE
Jun 13 13:58:13 * mrmoku goes upstairs in search for the neo...
Jun 13 13:58:59 <max_posedon> do you have /var/run/sshd with dropbear?
Jun 13 14:00:00 <max_posedon> oh... I think I know my problem
Jun 13 14:00:10 <max_posedon> I did, /etc/init.d/sshd restart
Jun 13 14:00:31 <max_posedon> with /etc/init.d/sshd start all works with missing that directory..
Jun 13 14:01:37 <DocScrutinizer> dos1: actually on inbound call it seems to me I just clicked left side to "Accept" a minute ago. But then the "close" butten appeared same place as "Accept", which is *really* BAD
Jun 13 14:01:53 <Ainulindale> *yawn*
Jun 13 14:03:06 <DocScrutinizer> dos1: so some screens seem to follow "left=green" pragma, bit not all of them, which is even worse then having all OK buttons at right side instead on left where they belong
Jun 13 14:03:20 <DocScrutinizer> Ainulindale: mooorning :-)
Jun 13 14:03:40 <Ainulindale> DocScrutinizer: I'm just waking up from a nap!
Jun 13 14:04:19 <DocScrutinizer> I'm just late to get meat for bbq. So for real now...
Jun 13 14:04:22 <mrmoku> dos1, (#514) what does frameworkd offer for that?
Jun 13 14:04:24 <DocScrutinizer> cya fellas
Jun 13 14:05:05 <troy__> I have successfully compiled kernel 2.6.24 for openmoko. It works if filesystem is in flash. What kernel config options do I need to select to have it boot with filesystem in microSD card?
Jun 13 14:05:25 <max_posedon> 2.6.24 is *very* outdated
Jun 13 14:05:51 <troy__> Its the only one I can get to reliably dial PPPoE
Jun 13 14:10:37 <dos1> Ainulindale: go and fix -dociswatchingyou theme!
Jun 13 14:10:58 <Ainulindale> No time now I'm leaving soon
Jun 13 14:11:23 <dos1> Ainulindale: DocScrutinizer complained, that you used that pic without permission
Jun 13 14:11:41 <dos1> Ainulindale: he said that he agrees for using it, if only it will have "DocScrutinizer checked it" label
Jun 13 14:11:53 <Ainulindale> He didn't complain at first, he asked later to add a copyright, and I will when I'll have time
Jun 13 14:13:49 <dos1> Ainulindale: huh, he said about that label even before you commited that theme, and he used statement "i agree (...) if and only if (...) DocScrutinizer checked it" :P
Jun 13 14:14:11 <dos1> mrmoku: it just returns dbus error i think
Jun 13 14:14:15 <Ainulindale> No he didn't, not to me that's for sure.
Jun 13 14:14:38 <Ainulindale> And anyway, again, I have no time now, and I'm off.
Jun 13 14:15:24 <dos1> Ainulindale: you're wrong
Jun 13 14:15:26 <dos1> Jun 12 20:59:15 <DocScrutinizer> dos1: Ainulindale: [2009-06-11 21:29:57] <DocScrutinizer> Ainulindale: I authorize you to use it for SHR splashscreen, if and only if you superimpose a "DocScrutinizer checked it" writing ;-)
Jun 13 14:15:32 <bumbl> hell is that doc?
Jun 13 14:15:55 <Ainulindale> dos1: I commited it before.
Jun 13 14:16:27 <dos1> Ainulindale: you're again wrong
Jun 13 14:16:30 <Gnutoo> max_posedon, it gives nearly no packet loss at the end
Jun 13 14:16:32 <dos1> you commited it 	 Fri, 12 Jun 2009 15:03:40 +0000 (17:03 +0200)
Jun 13 14:16:39 <dos1> and Doc said that day before
Jun 13 14:16:44 <dos1> [2009-06-11 21:29:57]
Jun 13 14:16:45 <Ainulindale> Check your hours.
Jun 13 14:16:57 <dos1> Check your days.
Jun 13 14:16:59 <dos1> ;)
Jun 13 14:17:09 <Ainulindale> I did, I grepped, and I have no such thing.
Jun 13 14:17:12 <Ainulindale> And as said before
Jun 13 14:17:26 <Ainulindale> I don't care, I have no time, and I will fix it anyways
Jun 13 14:17:28 <Ainulindale> So why insisting?
Jun 13 14:17:52 <dos1> http://logs.nslu2-linux.org/livelogs/openmoko-cdevel/openmoko-cdevel.20090612.txt
Jun 13 14:18:03 <Gnutoo> max_posedon, what should I do now?
Jun 13 14:18:04 <max_posedon> Gnutoo, so, mtu correct... sorry, have no other ideas
Jun 13 14:18:09 <dos1> Ainulindale: not insisting. i just like probing, that people are wrong :>
Jun 13 14:18:13 <dos1> s/probing/proving/
Jun 13 14:18:14 <bzzbot> dos1 meant: Ainulindale: not insisting. i just like proving, that people are wrong :>
Jun 13 14:18:28 <Gnutoo> max_posedon, openvpn wtith its mtu detection also said that it was correct
Jun 13 14:18:34 <Ainulindale> I'm not wrong, I don't have that in my logs, hence I can't have seen that any possible way
Jun 13 14:18:44 <dos1> Ainulindale: uh?
Jun 13 14:18:46 <dos1> Jun 11 19:29:58 <DocScrutinizer> Ainulindale: I authorize you to use it for SHR splashscreen, if and only if you superimpose a "DocScrutinizer checked it" writing ;-)
Jun 13 14:18:48 <dos1> Jun 11 19:30:07 <Ainulindale> Okay, I will.
Jun 13 14:18:53 <dos1> who wrote "Okay, I will." ?
Jun 13 14:18:54 <dos1> ;>
Jun 13 14:19:12 <Ainulindale> What's the fuck? Why is this not in my logs?!
Jun 13 14:19:14 <Gnutoo> max_posedon, I'll check iptables and tcpmss
Jun 13 14:20:12 <dos1> http://logs.nslu2-linux.org/livelogs/openmoko-cdevel/openmoko-cdevel.20090612.txt knows everything
Jun 13 14:20:13 <Ainulindale> Anyways as told before (several times), I'm off, I have to go to pick up my gal else she'll spank me
Jun 13 14:20:17 <dos1> ok ok ;)
Jun 13 14:21:06 <bumbl> dos1: you proved the unprovable - Ainulindale was wrong
Jun 13 14:22:45 <dos1> ha! ;D
Jun 13 14:25:49 <Ainulindale> bumbl: I wasn't wrong, I don 't have that in my logs
Jun 13 14:26:07 <Ainulindale> And guess what, irssi gz just one week, I think I fiddled too much with my settings
Jun 13 14:26:15 <bumbl> ;)
Jun 13 14:26:26 <Ainulindale> Anyways, off to pick the gal up else I'll really have difficulties to sit
Jun 13 14:26:34 <Ainulindale> (because of the spanking)
Jun 13 14:34:22 <dos1> blah blah, you were :P
Jun 13 14:34:25 <dos1> ;D
Jun 13 14:41:51 <CIA-45> SHR: 03vanous 07shr-settings * r138fd5fd75f4 10/data/po/cs/ (shr-settings.mo shr-settings.po): [data] po: add czech translation
Jun 13 14:43:14 <CIA-45> SHR: 03seba.dos1 07shr-settings * rb819da540088 10/setup.py: [setup.py] update to include czech translation
Jun 13 14:44:17 <CIA-45> SHR: 03vanous 07shr * r7b8438a1a427 10/libframeworkd-phonegui-efl/configure.ac: libframeworkd-phonegui-efl: add czech translation
Jun 13 14:47:25 <CIA-45> SHR: 03vanous 07shr * r0156f2c65454 10/libframeworkd-phonegui-efl/po/cs.po: libframeworkd-phonegui-efl: add czech translation
Jun 13 14:47:27 <CIA-45> SHR: 03seba.dos1 07shr * r38e7cd102057 10/libframeworkd-phonegui-efl/po/pl.po: libframeworkd-phonegui-efl: update polish translation
Jun 13 14:49:33 <dos1> well, i did mess... as always ;x
Jun 13 14:54:11 <Kensan_> alphaone: hi there
Jun 13 14:54:59 <mrmoku> hmm... looks like OE doesn't like my new gcc :(
Jun 13 14:56:20 <CIA-45> SHR: 03seba.dos1 07libframeworkd-phonegui-efl2 * re8d8c47e0131 10/po/pl.po: Update polish translation
Jun 13 15:00:03 * mrmoku would not have thought OEandYourDistro already has an entry for Fedora 11 :P
Jun 13 15:00:29 <doxy2> hi
Jun 13 15:01:53 <doxy2> does the wiki documentation on how to access the touch screen? 
Jun 13 15:01:57 <doxy2> as the accelerometer does
Jun 13 15:02:15 <doxy2> freerunner's, that is
Jun 13 15:05:00 <TAsn> doxy2, try /dev/input/touchscreen0
Jun 13 15:05:20 <TAsn> (I hope that's what you meant)
Jun 13 15:14:17 <TAsn> wow
Jun 13 15:14:21 <TAsn> just installed mokomaze
Jun 13 15:14:31 <TAsn> it got amazing improvements!
Jun 13 15:14:37 <doxy2> TAsn: yeah. but is there info somewhere how the data is formatted?
Jun 13 15:14:52 <TAsn> doxy2, drivers specs?
Jun 13 15:14:57 <TAsn> google for them
Jun 13 15:15:01 <TAsn> there's a page on the wiki
Jun 13 15:15:08 <TAsn> concerning the hardware on the moko
Jun 13 15:15:17 <TAsn> that's how you can easily figure out the model of the ts
Jun 13 15:16:31 <doxy2> TAsn: yeah, i know i can get that info from kernel src, but the point was that the infos not available in wiki :)
Jun 13 15:17:02 <TAsn> you are more than welcomed to add it :)
Jun 13 15:17:15 <doxy2> TAsn: heh. maybe i will
Jun 13 15:17:26 <TAsn> though it's probably a standard interface that's well documented and can be found everywhere :)
Jun 13 15:24:11 <TAsn> I wonder if there's a way to make e17 do a cool transition or at least blank while rotating the screen
Jun 13 15:24:19 <TAsn> because the current handling is ugly.
Jun 13 15:30:45 <max_posedon> mrmoku, why we do export DIPLAY=localhost:0 by default?
Jun 13 15:31:43 <dos1> max_posedon: for running apps from ssh session
Jun 13 15:32:09 <max_posedon> yes, but usually I wan't run them on my laptop, not on neo
Jun 13 15:32:14 <max_posedon> is I'm alone/
Jun 13 15:32:27 <dos1> max_posedon: so just export different value ;P
Jun 13 15:32:51 <max_posedon> I did, e.g. just comment, just start thinking about such strange default for me
Jun 13 15:34:18 <max_posedon> oh... smth wrong for me.
Jun 13 15:34:33 <max_posedon> Warning: untrusted X11 forwarding setup failed: xauth key data not generated
Jun 13 15:34:44 <max_posedon> ssh -X root@neo (openssh)
Jun 13 15:34:49 <max_posedon> what I forgot to do?
Jun 13 15:35:11 <max_posedon> X11Forwarding yes in sshd_config
Jun 13 15:37:57 <Gnutoo> max_posedon, maybe ssh -Y ? 
Jun 13 15:38:16 <max_posedon> yes, -Y works, but -X doesn't
Jun 13 15:38:22 <max_posedon> and I little dislike this
Jun 13 15:38:24 <Gnutoo> ok
Jun 13 16:17:35 <Sup3rkiddo> mickey|DepecheMo, 
Jun 13 16:44:59 <rebelThor> hi guys, any clues on why the latest SHR image and kernel cause a kernel panic on booting? ("Kernel panic - not syncing: VFS: Unable to mound root fs on unkown-block(31,6)" ?
Jun 13 16:45:22 <rebelThor> *shr unstable image from 11th of july
Jun 13 16:46:28 <PaulFertser> rebelThor: what loglevel are you running with? Are you sure you can't get any clues on LCD?
Jun 13 16:48:08 <rebelThor> PaulFertser: no clue on the loglevel :( I do get some other messages like: "INFO: RCU detected CPU - stall (t=4294963810/62000 jiffies)
Jun 13 16:48:42 <PaulFertser> rebelThor: what bootloader are you using?
Jun 13 16:48:53 <Kensan_> rebelThor: what fstype is your root partition? Is it supported by the kernel/initrd?
Jun 13 16:50:10 <rebelThor> PaulFertser: i don't really recall which is it, but could you point me to one that's good for SHR pretty please?
Jun 13 16:50:29 <PaulFertser> Kensan_: i think .29 can output enough information if it can't mount root to understand the reason (and it wasn't like that before, it's a recent mainline improvment).
Jun 13 16:50:37 <rebelThor> Kensan_: i have no clue unfortunately :(
Jun 13 16:50:42 <PaulFertser> rebelThor: I'm biased towards Qi, of course.
Jun 13 16:51:26 <rebelThor> PaulFertser: cool, I'll flash Qi then and see how that goes, let me see if I can find a good version
Jun 13 16:51:37 <PaulFertser> rebelThor: before flashing Qi you need to read the wiki page.
Jun 13 16:51:46 <Kensan_> PaulFertser: that was just a "I'm feeling lucky" kind of reply ;) I have no clue what kernel is currently shipped with shr
Jun 13 16:52:01 <PaulFertser> rebelThor: the good Qi version is at downloads.openmoko.org in experimental directory.
Jun 13 16:52:11 <PaulFertser> Kensan_: unstable should always build git HEAD>
Jun 13 16:52:23 <rebelThor> PaulFertser: thanks, reading the wiki page, and then trying to flash it
Jun 13 16:52:43 <Kensan_> PaulFertser: so what's the progress lately? I've been out of the loop for quite a while...
Jun 13 16:52:58 <PaulFertser> Kensan_: #1024 fixed!
Jun 13 16:53:05 <Kensan_> PaulFertser: your kidding?
Jun 13 16:53:10 <PaulFertser> Kensan_: nope
Jun 13 16:53:13 <PaulFertser> For real
Jun 13 16:53:16 <Kensan_> PaulFertser: wow
Jun 13 16:53:22 <PaulFertser> Dieter is a magician
Jun 13 16:53:34 <Kensan_> PaulFertser: well he's the man ;)
Jun 13 16:53:39 <PaulFertser> Kensan_: Qi for gta01 got SDHC support.
Jun 13 16:53:40 <mrmoku> PaulFertser, do you know the results of mickey|DepecheMo's power meassurements with that fix?
Jun 13 16:53:54 <Kensan_> PaulFertser: oh that's a nice addition.
Jun 13 16:54:02 <Kensan_> PaulFertser: who did it?
Jun 13 16:54:06 <PaulFertser> mrmoku: alphaone did #1024 measurements, quite impressive. 140hrs suspend time.
Jun 13 16:54:35 * mrmoku wonders if we should do a FSOSHRUDCON every 6 months... just to get that fixed sooner ;)
Jun 13 16:54:43 <PaulFertser> Kensan_: I borrowed josch's gta01 on F9C and added SDHC. 
Jun 13 16:55:06 <Kensan_> PaulFertser, ah nice.
Jun 13 16:55:20 <Kensan_> PaulFertser: hm ticket 1024 does not seem to reflect the current state
Jun 13 16:55:22 <PaulFertser> Kensan_: also some folks are motivated to do power consumption fixes on gta01.
Jun 13 16:55:29 <Kensan_> PaulFertser: gta01?
Jun 13 16:55:31 <PaulFertser> Kensan_: fix it :)
Jun 13 16:55:49 <PaulFertser> Kensan_: yes, gta01 is so power-hungry and it seems that at least part of the problems are fixable.
Jun 13 16:55:53 <Kensan_> PaulFertser: heh, no but thanks for the offer ;)
Jun 13 16:56:07 <PaulFertser> Kensan_: i think mrmoku will update you on SHR status. ;)
Jun 13 16:56:08 <Kensan_> PaulFertser: so there's still a pretty hardcore gta01 group...
Jun 13 16:57:11 <Kensan_> mwester has left to other endeavours?
Jun 13 16:57:12 <PaulFertser> Kensan_: not group exactly. But after several kernel fixes kernel git HEAD is finally usable on gta01, so we can continue hacking.
Jun 13 16:57:25 <PaulFertser> Kensan_: no, he's still around, just not much free time on his hands.
Jun 13 16:57:32 <PaulFertser> Kensan_: Dave is still absent :-/
Jun 13 16:57:44 <Kensan_> PaulFertser: who is actually doing the kernel work these days?
Jun 13 16:58:03 <PaulFertser> Kensan_: Hm, read the commit log ;)
Jun 13 16:58:15 <Kensan_> I know the feeling... haven't had time to check in on any of the projects I followed lately...
Jun 13 16:58:22 <DocScrutinizer> Kensan_: #1024 is a hw fix
Jun 13 16:58:55 <Kensan_> DocScrutinizer: hey there doc!
Jun 13 16:58:56 <Kensan_> :)
Jun 13 16:58:58 <DocScrutinizer> some EE messed up circuitry/layout another time :-/
Jun 13 16:58:59 <PaulFertser> I'll add the status update to the ticket right now.
Jun 13 16:59:12 <rebelThor> latest version of Qi prints something about "Magic bitmask 0x1985 not found at 0x..." :(
Jun 13 16:59:17 <rebelThor> over and over*
Jun 13 17:00:17 <DocScrutinizer> Kensan_: though, as usual, we can tell "that's not corect" but we can not prove correcting it would fix #1024
Jun 13 17:00:32 <PaulFertser> rebelThor: on serial console?
Jun 13 17:00:42 <rebelThor> nope, on the FreeRunner's screen
Jun 13 17:00:50 <Kensan_> DocScrutinizer: oh I see... :/
Jun 13 17:00:56 <PaulFertser> rebelThor: then you're wrong.
Jun 13 17:01:03 <PaulFertser> rebelThor: Qi never prints anything on screen.
Jun 13 17:01:03 <Kensan_> DocScrutinizer: did you meet up with raster?
Jun 13 17:01:11 <dos1> rebelThor: how Qi can print something on FR screen?
Jun 13 17:01:15 <DocScrutinizer> Kensan_: anyway we found a botch that seems to cure the damage done
Jun 13 17:01:18 <Kensan_> DocScrutinizer: he's somewhere in DE I believe
Jun 13 17:01:22 <PaulFertser> It can't, rebelThor didn't read the wiki page :P
Jun 13 17:01:42 <DocScrutinizer> Kensan_: (raster) well we met at F9N
Jun 13 17:02:00 <rebelThor> I did, I know it can't print anything, but that's what it's written on the screen lol. Maybe it's not using Qi then
Jun 13 17:02:04 <DocScrutinizer> (which seesm to be months ago...)
Jun 13 17:02:05 <Kensan_> f9N? what and when was that?
Jun 13 17:02:19 <wpwrak> dos1: would actually be nice if it could :)
Jun 13 17:02:30 <mrmoku> (raster is almost back home again)
Jun 13 17:02:37 <Kensan_> wpwrak: Oh hi there. Still around I see :)
Jun 13 17:03:12 <Kensan_> mrmoku I was supposed to meet him this tuesday but couldn't unfortunately
Jun 13 17:03:19 <wpwrak> Kensan: always :-)
Jun 13 17:04:24 <Kensan_> wpwrak: well now that openmoko is officially an all-community project, there's no other way *heh*
Jun 13 17:05:06 <dos1> oh, fine
Jun 13 17:05:26 <dos1> i'm going to open top cover of my FR first time
Jun 13 17:05:37 <dos1> and i can't remove one screw ;x
Jun 13 17:06:03 <DocScrutinizer> dos1: bah.  why?
Jun 13 17:06:18 <dos1> DocScrutinizer: dunno. it's almost out...
Jun 13 17:06:19 <sicu> I'm just about to reflash my bootloader, should i use the Qi from andy's folder, or should i take the one from testing?
Jun 13 17:06:57 <DocScrutinizer> no issue with silly "seal" (the paper patch)?
Jun 13 17:07:55 <DocScrutinizer> dos1: you don't need to *remove* the screws. Unscrew them, that's ok
Jun 13 17:08:05 <dos1> DocScrutinizer: ok
Jun 13 17:08:11 <dos1> DocScrutinizer: so... what's next? :D
Jun 13 17:08:12 <sicu> nm, i'll try testing
Jun 13 17:08:43 <dos1> i must check if i have something like guitar pick here...
Jun 13 17:08:45 <DocScrutinizer> dos1: use your fingernail to slide around the cover lid
Jun 13 17:09:26 <DocScrutinizer> start at aux button. take care not to press button too hard during this!
Jun 13 17:10:18 * rebelThor feels so stupid right now
Jun 13 17:10:23 <dos1> well, bottom part is almost openned
Jun 13 17:10:25 <DocScrutinizer> dos1: gently pulling lid up at aux will enable you to slide your nail around
Jun 13 17:10:33 * rebelThor for flashing the tar.gz :'(
Jun 13 17:10:39 <PaulFertser> sicu: take the one from experimental at downloads.openmoko.org
Jun 13 17:11:17 <dos1> DocScrutinizer: from AUX to up, or bottom?
Jun 13 17:11:19 <PaulFertser> dos1: or use your favourite guitar pick ;)
Jun 13 17:11:35 <PaulFertser> I usually start from the bottom itself.
Jun 13 17:11:42 <DocScrutinizer> I usually go to bottom, then all the way around
Jun 13 17:11:46 <PaulFertser> (probably that's what is written on the wiki)
Jun 13 17:11:54 <sicu> PaulFertser: ok, thanks
Jun 13 17:12:11 <dos1> PaulFertser: my pick is in different place :(
Jun 13 17:12:29 <DocScrutinizer> bah, fingernail works much better
Jun 13 17:12:45 <DocScrutinizer> even muuuch more safe
Jun 13 17:12:54 <DocScrutinizer> for FR ;-D
Jun 13 17:13:06 <lindi-> with external usb battery: if i lower screen brightness to 127 udev will start consuming 10% cpu time
Jun 13 17:13:07 <PaulFertser> I use fingernail usually, yes. But using guitar pick is cool, you know.
Jun 13 17:13:39 <dos1> that sounds are normal? :o
Jun 13 17:14:09 <PaulFertser> dos1: it depends. There're clips that click, so probably yes.
Jun 13 17:14:19 <PaulFertser> dos1: what are you doing that for?
Jun 13 17:14:42 <PaulFertser> Kensan_: #1024 updated.
Jun 13 17:15:24 <lindi-> (will file better report when i get home)
Jun 13 17:15:26 <Kensan_> PaulFertser: awesome thanks
Jun 13 17:15:43 <PaulFertser> lindi-: you're always finding amazing bugs :)
Jun 13 17:16:53 <czaks> hi
Jun 13 17:16:58 <Kensan_> wow 143 hours
Jun 13 17:17:02 <czaks> one question about hardware:
Jun 13 17:17:08 <Kensan_> who would have thought that was possible
Jun 13 17:17:27 <czaks> i've broken the glass from the display, is it relatively cheap to replace it?
Jun 13 17:17:38 <czaks> the display is left untouched
Jun 13 17:17:43 <lindi-> PaulFertser: in a forest and the extra batteries are runnig out :)
Jun 13 17:22:22 <PaulFertser> Who's using splashscreen here? Can anyone confirm that kernel messages are distorting it? 
Jun 13 17:23:27 <dos1> how easy is neo touchscreen to get/replace?
Jun 13 17:23:50 <PaulFertser> dos1: have you broken yoru TS?
Jun 13 17:24:07 <dos1> PaulFertser: no, but i have awfull scratches
Jun 13 17:24:28 <dos1> i was too lazy to apply some screen protector
Jun 13 17:24:29 <PaulFertser> dos1: i used screen protector (bought in nearby cell phone shop) from the day 1.
Jun 13 17:24:41 <dos1> :(
Jun 13 17:24:49 <dos1> sometimes i don't like my laziness :P
Jun 13 17:24:49 <PaulFertser> dos1: czaks: i think DocScrutinizer knows the answer
Jun 13 17:25:11 <czaks> PaulFertser, thanks
Jun 13 17:26:47 <dos1> DocScrutinizer: ?
Jun 13 17:27:37 <bumbl_> ~seen mirko-paroli
Jun 13 17:27:37 <bzzbot> mirko-paroli <n=mirko@91-64-217-87-dynip.superkabel.de> was last seen on IRC in channel #openmoko-cdevel, 16d 20h 5m 32s ago, saying: 'alphaone: any news on the suspend problem in fso ?'.
Jun 13 17:28:29 <czaks> i've bought my neo with an ugly scratch, but it's not the real problem
Jun 13 17:29:29 <PaulFertser> czaks: iirc LCM is semi-hard to replace, the question is how to get a new one. Probably one of the European distributors have some.
Jun 13 17:30:04 <czaks> PaulFertser, is it possible to put here some generic plastic or glass?
Jun 13 17:30:16 <PaulFertser> czaks: it's glued to the PCB so one has to warm the whole device to some 60C and then very slowly unglue the LCM.
Jun 13 17:30:23 <PaulFertser> czaks: resistive touchscreen.
Jun 13 17:30:54 <czaks> i see
Jun 13 17:30:55 <PaulFertser> czaks: you need to actually apply pressure to the touchscreen, it's not capacitive like iphones.
Jun 13 17:31:48 <PaulFertser> czaks: what i don't know is whether replacing only TS (instead of the whole LCM) possible. I guess not.
Jun 13 17:32:55 <czaks> PaulFertser, do you mean by LCM this movable plastic, or the glass inside?
Jun 13 17:33:20 <PaulFertser> czaks: by LCM i mean the whole module, LCD+TS
Jun 13 17:33:34 <PaulFertser> Quite possibly i'm wrong.
Jun 13 17:33:39 <czaks> i see
Jun 13 17:39:31 <PaulFertser> Any GTA01 user here? Please pastebin me s3c24xx-gpio output along with kernel version.
Jun 13 18:00:53 <rtp> PaulFertser: http://pastebin.ca/1459127
Jun 13 18:01:22 <PaulFertser> rtp: thanks!
Jun 13 18:10:54 <rtp> PaulFertser: btw, do you know if putting moko11 on gta01 is a good idea/will work ?
Jun 13 18:11:33 <PaulFertser> rtp: very good idea
Jun 13 18:11:43 <PaulFertser> rtp: i'd say it's mandatory if you plan to use gsm.
Jun 13 18:12:40 <rtp> ok. so I'll try it tonight
Jun 13 18:15:40 <rtp> PaulFertser: do you know where to ask for a kernel config change ? gta01 doesn't have oabi compat which probably explain why gllin doesn't start :/
Jun 13 18:16:02 <PaulFertser> rtp: yep
Jun 13 18:16:45 <PaulFertser> rtp: i wonder why you don't use that tricky special linker that was written specifically to run gllin. Doesn't it work reliably?
Jun 13 18:17:28 <rtp> PaulFertser: I wanted to test the gps on lastest shr image. took the first gllin ipk I found 
Jun 13 18:17:38 <PaulFertser> rtp: i understand.
Jun 13 18:17:55 <rtp> If you have link, I'll try that
Jun 13 18:18:31 <PaulFertser> rtp: first do "zgrep OABI /proc/config.gz" please.
Jun 13 18:19:09 <rtp> already did. it's disabled
Jun 13 18:20:02 <PaulFertser> rtp: CONFIG_OABI_COMPAT=y in OM repo gta01 defconfig. So ask SHR folks.
Jun 13 18:20:08 <PaulFertser> mrmoku: ?
Jun 13 18:20:27 <rtp> :/
Jun 13 18:21:55 <mrmobil> we use the moredrivers one
Jun 13 18:22:32 <mrmobil> the one from kernel source
Jun 13 18:24:35 <PaulFertser> mrmoku: are you sure?
Jun 13 18:25:10 <Defiant> the one labeled 2.6.28 but uname says 2.6.29?
Jun 13 18:25:19 <PaulFertser> dos1: yes, that's ok.
Jun 13 18:25:24 <PaulFertser> Defiant: ^^^
Jun 13 18:26:12 <Sharwin_F> does anyone have problems when receiving sms with default profile on last unstable?  it keeps ringing forever, even if I close the GUI showing it
Jun 13 18:30:26 <TAsn> Sharwin_F, closing the gui isn't supposed to affect it ;)
Jun 13 18:31:49 <Sharwin_F> TAsn, In my opinion it should... if you close the GUI is because you have read it or at least you are aware of it, so no need to continue ringing... and still this doesn't tell me why it keeps ringing forever :P
Jun 13 18:32:00 <mrmobil> PaulFertser: checked that the other day
Jun 13 18:32:13 <dos1> Sharwin_F: what ringtone do you have? :o
Jun 13 18:32:14 <TAsn> Sharwin_F, I just explained it from the technical pov ;)
Jun 13 18:32:16 <PaulFertser> rtp: ?
Jun 13 18:32:38 <mrmobil> we're talking about the new kernel, right?
Jun 13 18:32:41 <PaulFertser> Yes
Jun 13 18:32:41 <TAsn> Sharwin_F, btw, no idea.
Jun 13 18:32:42 <TAsn> ;)
Jun 13 18:32:58 <Sharwin_F> TAsn, ok, so it does, but it shouldn't, do you agree? :P
Jun 13 18:33:05 <Sharwin_F> dos1, default one
Jun 13 18:33:12 <TAsn> Sharwin_F, I don't think this is a wanted behavior, no.
Jun 13 18:33:13 <rtp> PaulFertser: ?
Jun 13 18:33:29 <TAsn> especially if it's the default one, which is far from pretty :)
Jun 13 18:33:32 <dos1> Sharwin_F: how it can keep ringing forever?
Jun 13 18:33:43 <dos1> Sharwin_F: it has only ~6 seconds! :x
Jun 13 18:33:43 * mrmobil will recheck, when back at pc
Jun 13 18:33:52 <TAsn> dos1, loop?
Jun 13 18:34:04 <dos1> TAsn: it isn't looped
Jun 13 18:34:09 <TAsn> Sharwin_F, is it possible you changed it to loop?
Jun 13 18:34:14 <TAsn> shr-settings, please verify.
Jun 13 18:34:15 <PaulFertser> rtp: you need to discuss SHR kernel config issue with mrmobil :)
Jun 13 18:34:37 <TAsn> (is there such an option?)
Jun 13 18:34:42 <Sharwin_F> dos1, I remember setting the ring loop on, but on Vibrate , and I'm using default one now, so it should be everything there should be set as shr-default
Jun 13 18:34:51 <Sharwin_F> TAsn, ^
Jun 13 18:35:01 <TAsn> Sharwin_F, try setting that off
Jun 13 18:35:03 <TAsn> if that's the case
Jun 13 18:35:05 <TAsn> open a bug
Jun 13 18:35:06 <TAsn> :)
Jun 13 18:35:15 * mrmobil moves his ass to his computer
Jun 13 18:35:34 <Sharwin_F> dos1, it rings, then silent for about 4 seconds, then rings another time, and then repeats forever :P
Jun 13 18:35:56 <dos1> Sharwin_F: maybe you changed it accidentaly :P check it
Jun 13 18:35:58 <rtp> PaulFertser: ahh, sure.... well, I can also open a bug and blame Ainulindale for uploading this kernel :)
Jun 13 18:36:00 <dos1> in shr-settings
Jun 13 18:36:36 <PaulFertser> rtp: Ainulindale is not guilty, it's me who persuaded SHR folks to use the latest for gta01 so that the remaining regressions would be found faster ;)
Jun 13 18:36:55 <Sharwin_F> dos1, in default: Ring loop on, Message loop off
Jun 13 18:37:02 <mrmoku> rtp, blaming Ainulindale is certainly the first step ;)
Jun 13 18:37:11 <mrmoku> rest will come :P
Jun 13 18:37:17 <TAsn> dos1, how can he verify it on the frameworks side?
Jun 13 18:37:23 <Sharwin_F> Message length=7
Jun 13 18:37:24 <rtp> PaulFertser: about regression and troubles, I found some 
Jun 13 18:37:27 <TAsn> do you just change the profile files?
Jun 13 18:37:32 <PaulFertser> rtp: please share
Jun 13 18:37:55 <dos1> TAsn: changes in profile files should be equal to changes in shr-settings
Jun 13 18:37:56 <dos1> :P
Jun 13 18:38:05 <TAsn> dos1, duh.
Jun 13 18:38:12 <rtp> mrmoku: Ainulindale is fine to be blamed as long as there's a bug report.. that's not even funny :)
Jun 13 18:38:12 <TAsn> I meant, does shr-settings change the profile files?
Jun 13 18:39:32 <rtp> PaulFertser: I plan to send a mail to shr-devel as reply to Ainulindale. i'll put them here but the main grip is phone stay in suspend when receiving a call :(
Jun 13 18:39:50 <PaulFertser> rtp: upgrade you firmware first.
Jun 13 18:40:06 <rtp> PaulFertser: this may affect the wake up ?
Jun 13 18:40:09 <PaulFertser> rtp: sure
Jun 13 18:40:11 <Sharwin_F> dos1, so it's correct according to shr-settings, no Message loop... a bug then¿
Jun 13 18:40:26 <dos1> Sharwin_F: checks logs
Jun 13 18:40:33 <rtp> PaulFertser: damn... I was thinking this firmware issue was bv3 only :(
Jun 13 18:40:47 <PaulFertser> rtp: moko11 is there for a reason
Jun 13 18:40:50 <Sharwin_F> dos1, hmm logs are deleted when you restart the neo?
Jun 13 18:40:57 <dos1> yep
Jun 13 18:40:59 <mrmoku> rtp, so... the recipe modifies the default config a little
Jun 13 18:41:01 <PaulFertser> rtp: does your phone never wakes up on incoming call?
Jun 13 18:41:09 <mrmoku>                 sed -i -e s,CONFIG_S3C_LOWLEVEL_UART_PORT=2,CONFIG_S3C_LOWLEVEL_UART_PORT=0, ${WORKDIR}/defconfig
Jun 13 18:41:10 <PaulFertser> mrmoku: heheh
Jun 13 18:41:12 <mrmoku>                 sed -i -e s,CONFIG_DEBUG_S3C_UART=2,CONFIG_DEBUG_S3C_UART=0, ${WORKDIR}/defconfig
Jun 13 18:41:14 <rtp> PaulFertser: moko11 is there for rtscts iirc
Jun 13 18:41:26 <mrmoku> hidden in linux-openmoko.inc... that's why I oversaw it
Jun 13 18:41:28 <Sharwin_F> then no log files.... it's really anoying to have the message tone playing forever... I had to reboot :P
Jun 13 18:41:33 <mrmoku> but it does not turn off that ABI thing
Jun 13 18:41:39 <PaulFertser> rtp: anyway, please upgrade first before investigating inability to resume.
Jun 13 18:41:40 <rtp> PaulFertser: does not wake up on incoming call :(
Jun 13 18:41:48 <rtp> PaulFertser: ok
Jun 13 18:41:50 <PaulFertser> rtp: what HW revision is that?
Jun 13 18:42:29 <rtp> PaulFertser: "Revision: 24110240" is that what you need ?
Jun 13 18:42:38 <PaulFertser> rtp: no, i mean bv3? bv4?
Jun 13 18:42:57 <rtp> mrmoku: it must be disabled somewhere :)
Jun 13 18:43:10 <rtp> PaulFertser: bv4. I didn't try (yet) on my bv3
Jun 13 18:43:18 * mrmoku checks buildhost
Jun 13 18:43:52 <PaulFertser> rtp: bv3 has a hardware issue for sure. Probably early bv4 too... Say, if you suspend and resume (by PWR), is modem functional after that?
Jun 13 18:44:26 <rtp> PaulFertser: didn't try.. will try after firmware update
Jun 13 18:44:49 <PaulFertser> rtp: to fix it on bv3 you'll probably need to desolder one transistor.
Jun 13 18:45:27 <rtp> ouch...
Jun 13 18:45:53 <PaulFertser> rtp: a glitch on SoC lines resets the modem (iirc). So reset line should be cut somehow :)
Jun 13 18:46:35 <rtp> yeah, it's a matter of reset but I didn't remember how to solve that
Jun 13 18:47:57 <PaulFertser> rtp: remove Q1001  
Jun 13 18:48:27 <mrmoku> somebody please hit me with a clue-stick? where to find the kernel config in ipk's?
Jun 13 18:49:05 <rtp> PaulFertser: I don't have suited tools :(
Jun 13 18:49:24 <PaulFertser> rtp: no cellphone repair shop nearby?
Jun 13 18:50:30 <rtp> probably but that's not a big problem. if I can use my bv4 while I send my gta02 to germany, it'll be fine
Jun 13 18:51:21 <PaulFertser> rtp: if your bv4 still has that nasty bug you won't be able to fully use it obviously.
Jun 13 18:51:44 <PaulFertser> rtp: is your bv4 pre-MP or MP? What color is the board?
Jun 13 18:51:59 <rtp> red iirc /o\
Jun 13 18:53:26 <PaulFertser> rtp: http://wiki.openmoko.org/wiki/One_transistor_too_many
Jun 13 18:54:25 <PaulFertser> Heh, i've seen this page before :))) But i found Q1001 on schematics myself ;)
Jun 13 18:56:02 <rtp> :(
Jun 13 18:57:06 <PaulFertser> rtp: cell phone repair technician. Just disassemble your device, carefully take out the can (i can teach you how to do it) and find that transistor. Go to the shop and you'll have it removed in a matter of minutes.
Jun 13 18:57:24 <Q-Master> Hi All!
Jun 13 18:58:43 <Q-Master> PaulFertser: have you already made a deep-sleep fix?
Jun 13 18:59:18 <PaulFertser> Q-Master: alphaone did it for me at F9C.
Jun 13 18:59:45 <Q-Master> dos1: SHR become unusable after viewing and then calling contacts from OPIMD.
Jun 13 19:00:00 <Q-Master> PaulFertser: lucky you. 8) I need to catch my friend for that
Jun 13 19:00:07 <PaulFertser> Q-Master: it seems that moscow chipdip has a part that fits.
Jun 13 19:00:14 <rtp> PaulFertser: ok. will try :)
Jun 13 19:00:32 <PaulFertser> rtp: do you need any clarifications on dissassembly/dealing with can?
Jun 13 19:00:41 <dos1> Q-Master: well, we have now many bugs in FSO, which causes unusability of it ;p
Jun 13 19:00:46 <Q-Master> PaulFertser: just a condensator. it's easy to find, but not so easy to solder
Jun 13 19:01:11 <PaulFertser> Q-Master: not that easy. Must be 0805 low-ESR 22uF
Jun 13 19:01:25 <mrmoku> PaulFertser, rtp: ok, there is some special handling for CONFIG_OABI_COMPAT in linux.inc
Jun 13 19:01:33 <Q-Master> dos1: crap. I need an PIM even for contacts. 8(
Jun 13 19:01:36 <PaulFertser> rtp: "The can should be lifted very carefully. Use a pin-pointed knife for that and try to lift a little in every point you can reach going in circle. After several rounds the can will be easily dismounted. It can take about 10 minutes, please be patient."
Jun 13 19:02:08 <rtp> PaulFertser: more infos is better than not enough info, so yes, if you have more infos, please share :)
Jun 13 19:03:00 <dos1> Q-Master: are you sure it's opimd issue? :P
Jun 13 19:03:40 <Q-Master> dos1: it might be a lfb-efl2 issue
Jun 13 19:03:58 <mrmoku> looks like it gets turned off for angstrom
Jun 13 19:04:00 <mrmoku> conf/distro/include/angstrom.inc:ARM_KEEP_OABI = "0"
Jun 13 19:04:14 <mrmoku> and specifically turned on for some openmoko kernels
Jun 13 19:04:25 <mrmoku> conf/distro/include/angstrom.inc:ARM_KEEP_OABI_pn-linux-openmoko-2.6.28_om-gta01 = "1"
Jun 13 19:04:45 <mrmoku> PaulFertser, is that OABI thing something we want in any case?
Jun 13 19:04:59 <PaulFertser> mrmoku: for gta01 unfortunately yes
Jun 13 19:05:02 <mrmoku> if so we should stuff ARM_KEEP_OABI=1 in our shr.conf
Jun 13 19:05:05 <mrmoku> only gta01?
Jun 13 19:05:18 <PaulFertser> mrmoku: because gta01 users have to use proprietary crap to use gps :(
Jun 13 19:05:21 <mrmoku> brb... have to give the goodnight for pablo
Jun 13 19:09:18 <mrmoku> hehe... wifi is flakey for my neo up here :P
Jun 13 19:11:20 <PaulFertser> rtp: can you do "xset dpms force off" (the screen should blank here) and then pastebin me GPIO info?
Jun 13 19:13:59 <rtp> currently trying to update the firmware. will do right after
Jun 13 19:14:07 <rtp> what are you looking for ? :)
Jun 13 19:14:46 <PaulFertser> rtp: current leaks
Jun 13 19:16:11 <rtp> well, imho some leaks need to be fixed on hard level :)
Jun 13 19:16:34 <PaulFertser> Hm, a guy from the USA says he's never received a number with leading plus from GSM network :-O 
Jun 13 19:16:52 <PaulFertser> rtp: do you have anything particular in mind?
Jun 13 19:17:22 <rtp> no but this would not be surprising :)
Jun 13 19:18:31 * mrmoku wonders if that USA guy got confused between his mobile and his cordless ;)
Jun 13 19:18:51 <PaulFertser> mrmoku: he says he don't use landline for 10 years.
Jun 13 19:19:32 <mrmoku> rtp, building a new kernel for you :)
Jun 13 19:21:28 <rtp> mrmoku: thx :)
Jun 13 19:23:12 <rtp> +CGMR: "GSM: gsm_ac_gp_fd_pu_em_cph_ds_vc_cal35_ri_36_amd8_ts0-Moko11"... looks like the firmware update was successfull :)
Jun 13 19:23:21 <PaulFertser> rtp: yay
Jun 13 19:23:47 <bumbl_> PaulFertser: isn't that against the gsm specification?
Jun 13 19:24:33 <PaulFertser> bumbl_: i told him that i suspect just that. Probably he sees some mangled number already, but he says it's in all SHR apps.
Jun 13 19:26:07 <rtp> PaulFertser: wow. first time I'm seeing this. dpms force off failed :(
Jun 13 19:28:38 <mrmoku> juri_, hey, long time no see - guess your gta01 is working :)
Jun 13 19:29:06 <PaulFertser> rtp: any dmesg output?
Jun 13 19:29:43 <juri_> mrmoku: its been working. isn't right now, but thats cause i filled the FS up with messages. time to back it up, and upgrade!
Jun 13 19:29:55 <mrmoku> :)
Jun 13 19:30:27 <rtp> PaulFertser: no. it was X not being happy but the screen went off by itsel http://pastebin.ca/1459197
Jun 13 19:30:41 <rtp> err itself
Jun 13 19:32:02 <PaulFertser> rtp: ok, so now i have gpio with the screen off, will think ...
Jun 13 19:35:20 <pb__> mrmoku: I think it'd be easy enough to wrap the gps stuff to work on an eabi kernel.  I talked with someone about this last year, maybe Ainulindale.
Jun 13 19:35:33 <pb__> that'd save having to enable all the oabi compatibility stuff with the attendant slowdown
Jun 13 19:36:06 <mrmoku> pb__, what would be needed to do that? and who will do it ? ;)
Jun 13 19:36:18 <rtp> PaulFertser: screen off may not appear on gpio. there's a lcpwren bit in the lcd registers to power down the lcd. Also, it doesn't handle handle backlight. bl driver need some care to do that iirc
Jun 13 19:36:45 <pb__> mrmoku: basically you'd just need to build a hybrid c library that exposed the oabi interface to gpsd (or whatever the binary thing is, I forget) but used the eabi syscalls to talk to the kernel.
Jun 13 19:36:57 <PaulFertser> rtp: i was reading the code before i asked so yes, i know about S3C2410_LCDCON1_ENVID :)
Jun 13 19:37:46 <PaulFertser> rtp: so, did you see the backlight was still on after "dpms force off"?
Jun 13 19:38:30 <rtp> PaulFertser: even if I didn't play with it since monthes ago, any question about kernel fb driver, please ask :)
Jun 13 19:38:37 <rtp> PaulFertser: I didn't check sorry :(
Jun 13 19:39:52 <PaulFertser> rtp: if bl is still on it'd explain why gta01 is so power-hungry ;)
Jun 13 19:42:22 <rtp> the bl on the 01 is using a timer of the 2410 right ?
Jun 13 19:42:51 <rtp> if so, iirc, it can be seen through the gpio values
Jun 13 19:43:18 <PaulFertser> rtp: yes, you're right.
Jun 13 19:43:25 <PaulFertser> :)
Jun 13 19:45:14 <rtp> yeah, with new firmware, it wakes up on incoming calls :)
Jun 13 19:45:19 <PaulFertser> rtp: both times it was F0 (primary function, current output 0). If it's wdt then it doesn't tell us anything.
Jun 13 19:46:52 <PaulFertser> rtp: btw, i spent some time trying to understand why my GTA02 sometimes didn't wake up on incoming calls. I proved it was a firmware bug. After that hardware flow control question was raised again and Dieter fixed both bugs and released moko11.
Jun 13 19:47:36 <PaulFertser> rtp: but it was only on first suspend. And since then there's a kernel workaround so moko10 should be able to wake up as well.
Jun 13 19:47:52 <rtp> so  moko11 doesn't fix only flow control ?
Jun 13 19:48:02 <PaulFertser> There's a changelog
Jun 13 19:48:20 <PaulFertser> And it's written there the same i just told you.
Jun 13 19:48:28 <rtp> who reads changelog except kernel changelog ? :)
Jun 13 19:51:46 <PaulFertser> rtp: of course it wasn't me alone who tried to understand why moko10 can't always resume the device. Joerg and Werner helped a lot.
Jun 13 19:53:27 <PaulFertser> rtp: and please don't forget to report kernel regressions to either openmoko-kernel ML or at least to me (and no, i don't read SHR ML).
Jun 13 19:55:52 <rtp> PaulFertser: will do, even if I can't tell they're regression or not. last time I played with my gta01 was loooong ago
Jun 13 19:59:09 <mrmoku> rtp, ok, new kernel is in the feed... if it does not work as expected... ping me
Jun 13 20:01:22 <rtp> mrmoku: thanks
Jun 13 20:15:18 <rtp> mrmoku: looks like om-gta01/Packages.gz not updated for new kernel. opkg upgrade does nothing. issue on my side ?
Jun 13 20:15:59 <mrmoku> hmm...
Jun 13 20:16:05 <mrmoku> i built the index...
Jun 13 20:16:10 <mrmoku> what version have you installed?
Jun 13 20:16:41 <mrmoku> did you opkg update?
Jun 13 20:17:26 <rtp> sure, I did opkg update before :)
Jun 13 20:18:14 <mrmoku> hmm... will bump PR for kernel then... and rebuild
Jun 13 20:18:54 <rtp> the new one is 2.6.29-oe10+gitr119819... isn't that enough ?
Jun 13 20:19:02 <rtp> (I've 2.6.29-oe10+gitr119818)
Jun 13 20:20:01 <mrmoku> hmm that should be enough, yeah
Jun 13 20:20:16 <Q-Master> mrmoku: there's no binutils in SHR unstable
Jun 13 20:20:18 <mrmoku> ahh... ok 
Jun 13 20:20:21 * mrmoku is stupid
Jun 13 20:20:25 <mrmoku> rtp, one minute
Jun 13 20:21:03 <rtp> :)
Jun 13 20:21:58 <mrmoku> Q-Master, http://build.shr-project.org/shr-unstable/ipk/armv4t/binutils_2.18-r4_armv4t.ipk
Jun 13 20:22:02 <mrmoku> there is :)
Jun 13 20:22:20 <Q-Master> hm.
Jun 13 20:22:21 <mrmoku> rtp, before building the index I switched config back to gta02
Jun 13 20:22:31 <mrmoku> it does not build the index for gta01 then :P
Jun 13 20:23:20 <Q-Master> heh.
Jun 13 20:23:55 <Q-Master> mrmoku: and jamvm and classes? 8)
Jun 13 20:24:30 <mrmoku> Q-Master, classpath does not build
Jun 13 20:24:42 <mrmoku> so no java until that is solved...
Jun 13 20:24:55 <rtp> mrmoku: everyone is making stupid mystakes :)
Jun 13 20:31:05 <max_posedon> any chanse that remoko will work some day in SHR-distro?
Jun 13 20:31:10 <mrmoku> rtp, ok, retry :)
Jun 13 20:31:22 <mrmoku> max_posedon, only if somebody is going to fix it :)
Jun 13 20:31:24 <max_posedon> right now it looks like it didn't installs all files
Jun 13 20:31:39 <mrmoku> max_posedon, is it fixed for bluez4?
Jun 13 20:31:45 <dos1> mrmoku: add fsoraw to feed
Jun 13 20:31:49 <max_posedon> I don't know status
Jun 13 20:31:54 <max_posedon> might be not
Jun 13 20:32:13 <bumbl> max_posedon: as it seems valerio has no time/interest in fixing it
Jun 13 20:32:30 <rtp> mrmoku: opkg upgrade is now doing something. thanks :)
Jun 13 20:32:41 <mrmoku> rtp, fine :)
Jun 13 20:32:45 <max_posedon> pure, very interesting project
Jun 13 20:34:08 <bumbl> remoko rocked
Jun 13 20:34:34 <mrmoku> if remoko is that a cool thing (me never tried it)... somebody should just fix it ;)
Jun 13 20:35:51 <bumbl> mrmoku: when you have to do presentations every now and then
Jun 13 20:35:57 <bumbl> remoko is really nice
Jun 13 20:36:18 <bumbl> although i have to say the apple iwork iphone app is nicer
Jun 13 20:36:30 <playya> port it :D
Jun 13 20:36:41 <bumbl> it just works with iwork
Jun 13 20:36:43 <bumbl> ;(
Jun 13 20:37:04 <bumbl> + is closed source -> reimplement it
Jun 13 20:37:11 <mrmoku> so it's not nice... it's borgish
Jun 13 20:37:19 <mrmoku> trying to assimilate you :P
Jun 13 20:37:50 <bumbl> functionality > openness
Jun 13 20:37:59 <bumbl> at least for me
Jun 13 20:38:29 <mrmoku> to a certain degree... yes :)
Jun 13 20:38:31 <playya> if its not buggy
Jun 13 20:38:48 * mrmoku can just talk because he is not doing presentations... and has no iphone :D
Jun 13 20:38:53 <playya> i hate my n95 because of the mp3 player. and i can not fix it
Jun 13 20:39:15 <playya> if you remove a file from the storage it's still in the DB
Jun 13 20:39:24 <bumbl> playya: buggy = functionality killer
Jun 13 20:40:54 <playya> the n95 is a nice device, but the software is closed and no chance to use sth different
Jun 13 20:41:14 <VDVsx> max_posedon, I'm porting remoko for bluez4, is almost done
Jun 13 20:41:19 <mrmoku> dos1, hmm... I thought it made it even into the lite image
Jun 13 20:41:25 <mrmoku> apparently not so :(
Jun 13 20:41:32 <bumbl> ah varerio has time and interest
Jun 13 20:41:43 <dos1> mrmoku: it was in oemerge feed
Jun 13 20:41:51 <dos1> mrmoku: but it isn't in unstable
Jun 13 20:41:56 <gnutoo> hello,how does ppp works manually(with picocom)?
Jun 13 20:41:59 <mrmoku> dos1, yeah...manually built then
Jun 13 20:42:06 <max_posedon> VDVsx, wow, really nice!)
Jun 13 20:42:06 <mrmoku> I thought I added it to task-shr-minimal
Jun 13 20:42:21 <max_posedon> bumbl, you should take your words back)
Jun 13 20:42:38 <gnutoo> max_posedon, hi  I didn't succeed with the vpn...
Jun 13 20:42:42 <bumbl> sry for the insulting words
Jun 13 20:43:16 <max_posedon> gnutoo, still? hm.. remind me what is your problem
Jun 13 20:43:29 <VDVsx> bumbl, I don't have disk space and a OM environment to compile some files, but I already got some help on that :)
Jun 13 20:43:48 <gnutoo> max_posedon, when I use  udp over udp openvpn it makes the vpn exits...but that is only triggered with gprs
Jun 13 20:44:02 <bumbl> hehe same problem here - i have to fix my laptop
Jun 13 20:44:05 <gnutoo> max_posedon, I've tried MTU,fragments etc...
Jun 13 20:44:39 <gnutoo> max_posedon, I've also tried UDP over  TCP...but when I get a sip call the latenci increase to 30s
Jun 13 20:45:20 <gnutoo> max_posedon, I'd ilke to have encrypted calls with the openmoko
Jun 13 20:45:45 <gnutoo> max_posedon, or any other free phone
Jun 13 20:46:53 <max_posedon> ok, so here can be 2 different issues
Jun 13 20:47:26 <max_posedon> in first could you post openvpn.log
Jun 13 20:47:39 <max_posedon> with loglevel 3 for beginning
Jun 13 20:47:48 <gnutoo> I've loglevel 4
Jun 13 20:47:51 <max_posedon> fine.
Jun 13 20:48:10 <gnutoo> in which scenario do I post it?
Jun 13 20:48:12 <max_posedon> also will be nice to have server's log
Jun 13 20:48:17 <gnutoo> the UDP over TCP?
Jun 13 20:48:22 <max_posedon> udp over udp
Jun 13 20:48:28 <gnutoo> ok thanks
Jun 13 20:48:57 <gnutoo> the server is mine so I can access the logs easily
Jun 13 20:49:01 <max_posedon> year ago I did some openvpn developing, so I'm familar with it, but it can takes time
Jun 13 20:49:07 <gnutoo> wow
Jun 13 20:49:10 <gnutoo> thanks a lot
Jun 13 20:51:02 <phobie> dbus.SystemBus() shows "Using **pending_return in dbus_connection_send_with_reply_setup() without pending_setup is deprecated and strongly discouraged". How to fix that?
Jun 13 20:54:00 <mrmoku> oops... russian bip server is gone :P
Jun 13 20:55:01 <F4t> i also want to know the answer to phobie's question :)
Jun 13 21:02:08 <gnutoo> max_posedon, basically here what I do: I connect the vpn server, then I can ping the vpn server tun interface,access to its' web pages etc...but...when I receive a sip call or call someone with SIP it destroys the GPRS connection and the tunnel
Jun 13 21:02:16 <gnutoo> I'll post the log soon
Jun 13 21:02:41 <max_posedon> ditroing gprs very bad
Jun 13 21:03:12 <max_posedon> also your neo may be after this become completly unusable
Jun 13 21:04:17 <gnutoo> max_posedon, as destroying gprs I mean that the gprs connection is gone and that I have to reboot
Jun 13 21:04:27 <gnutoo> max_posedon, the message is write UDPv4 []: No buffer space available (code=105)
Jun 13 21:04:36 <max_posedon> stop-stop
Jun 13 21:04:40 <gnutoo> max_posedon, but doing as in the FAQ didn't change nothing
Jun 13 21:04:46 <max_posedon> gprs connection gone
Jun 13 21:04:53 <max_posedon> and you can dial somebody, yes?
Jun 13 21:05:02 <max_posedon> (with regular dialer)
Jun 13 21:05:20 <gnutoo> I didn't try because the card only does internet access but I'll try to see if I get sms
Jun 13 21:05:31 <max_posedon> plz try
Jun 13 21:05:34 <gnutoo> ok
Jun 13 21:05:37 <max_posedon> because I have same issuy
Jun 13 21:05:37 <gnutoo> thanks
Jun 13 21:05:52 <max_posedon> you can try reproduce it just with wget-ing smth big
Jun 13 21:05:58 <max_posedon> (without any vpn)
Jun 13 21:06:02 <gnutoo> ah ok
Jun 13 21:06:06 <max_posedon> after ~1Mb neo dead for me
Jun 13 21:06:54 <gnutoo> ouch...
Jun 13 21:11:35 <gnutoo> max_posedon, vpn doesn't work anymore...it starts re-trying each time and I've got connection refused...but I've also got some connection that work: Jun 13 22:09:23 router openvpn[27483]: 80.125.172.58:6169 Re-using SSL/TLS context
Jun 13 21:12:19 <gnutoo> tun0 is  not created on the openmoko but tun module is present
Jun 13 21:12:33 <gnutoo> tun0 is present on the server
Jun 13 21:13:33 <max_posedon> sorry, I want full logs, or at least more lines)
Jun 13 21:14:00 <max_posedon> usually it happens, when client can't ping server anymore
Jun 13 21:14:10 <max_posedon> or smth wrong with sertificates
Jun 13 21:14:19 <max_posedon> also, check time and date)
Jun 13 21:14:46 <gnutoo> it worked before  so I don't think it's the certificates
Jun 13 21:15:36 <max_posedon> may be.. I'll be lucky see both logs
Jun 13 21:16:29 <gnutoo> I'll try a system for getting the logs while not connecting the neo trough the network
Jun 13 21:16:42 <gnutoo> becuase I don't have serial
Jun 13 21:21:17 <gnutoo> server config
Jun 13 21:21:18 <gnutoo> http://pastebin.com/m484beaab
Jun 13 21:21:43 <gnutoo> rebooting the openmoko again...maybe it's because of dbus?
Jun 13 21:22:16 <max_posedon> I'm now completly sure, that your problem isn't openvpn
Jun 13 21:22:24 <gnutoo> ah
Jun 13 21:22:25 <max_posedon> plz try do wget test
Jun 13 21:22:28 <gnutoo> ok
Jun 13 21:22:46 <max_posedon> I have similar issue - many traffic -> neo dead
Jun 13 21:22:53 <max_posedon> but for regular irc rock stable
Jun 13 21:22:57 <gnutoo> but I've already transfered half an ogg with ssh over openvpn
Jun 13 21:23:00 <gnutoo> but ssh is tcp
Jun 13 21:23:10 <gnutoo> but I'll try
Jun 13 21:23:23 <max_posedon> no, not over vpn, just regular download
Jun 13 21:23:26 <gnutoo> ok
Jun 13 21:23:36 <gnutoo> I'll do a wget of an ipk
Jun 13 21:23:52 <max_posedon> problem is that its hard to reproduce sometimes
Jun 13 21:23:55 <gnutoo> wesnoth-data would be huge
Jun 13 21:29:00 <gnutoo> max_posedon, it slow but seems to work
Jun 13 21:29:11 <max_posedon> oh... bad
Jun 13 21:29:22 <gnutoo> I'll wait until 500k
Jun 13 21:30:18 <max_posedon> if it works, we will try reproduce it with your scenario
Jun 13 21:30:34 <max_posedon> gprs dead really isn't openvpn releated
Jun 13 21:30:47 <gnutoo> ok
Jun 13 21:32:06 <gnutoo> I've incomming plaintext error on TLS object
Jun 13 21:32:13 <gnutoo> I'll send the logs
Jun 13 21:32:15 <gnutoo> and the config
Jun 13 21:32:46 <gnutoo> server log: http://pastebin.com/m186829ac
Jun 13 21:38:00 <gnutoo> magbeat, openmoko log: http://pastebin.ca/1459313
Jun 13 21:39:29 <gnutoo> max_posedon, and here's the openmoko conf:  http://pastebin.ca/1459317
Jun 13 21:39:48 <max_posedon> I mainly interested in logs
Jun 13 21:40:12 <gnutoo> ok
Jun 13 21:40:14 <max_posedon> (logs fine for me)
Jun 13 21:45:49 <gnutoo> max_posedon, I've packets like this: 22:44:05.257920 IP host248-84-dynamic.51-79-r.retail.telecomitalia.it.16793 > 10.162.56.6.10501: UDP, length 44
Jun 13 21:46:32 <khiraly11> er
Jun 13 21:46:41 <DocScrutinizer2> hehe, IRC deadlock, when trying to log in with ghosting own nick, then changing nick to that ghosted nick. Works fine for reconnects with one client. Using this on two clients/connections starts to pingpong though ;-D
Jun 13 21:47:24 <gnutoo> max_posedon, so what should I do now?
Jun 13 21:47:39 <max_posedon> try make sip call
Jun 13 21:47:46 <max_posedon> and wait for a gprs dead
Jun 13 21:47:47 <DocScrutinizer2> so, my appologies for spaming the channels(s)
Jun 13 21:49:07 <gnutoo> max_posedon, but if I don't even have  the  tun link working...how do I do?
Jun 13 21:49:22 <max_posedon> DocScrutinizer2, btw, did you switched to xchat
Jun 13 21:49:22 <gnutoo> it worked  before but don't work anymore
Jun 13 21:49:30 <max_posedon> gnutoo, ah.. ok, so plz
Jun 13 21:49:40 <max_posedon> POST openvpn.logs
Jun 13 21:49:58 <gnutoo> I've posted them...
Jun 13 21:50:10 <max_posedon> ah.. sorry)
Jun 13 21:50:16 <gnutoo> np
Jun 13 21:50:28 <gnutoo> openmoko log: http://pastebin.ca/1459313
Jun 13 21:50:36 <gnutoo> server log: http://pastebin.com/m186829ac
Jun 13 21:50:36 <DocScrutinizer2> SIP thru VPN over GPRS sounds *rather* kinky anyway
Jun 13 21:50:45 <gnutoo> yes
Jun 13 21:50:56 <gnutoo> over wifi it works well
Jun 13 21:50:58 <max_posedon> gnutoo, ha!
Jun 13 21:51:00 * DocScrutinizer2 back to bbq
Jun 13 21:51:02 <gnutoo> (wifi+openvpn)
Jun 13 21:51:05 <max_posedon> I know your problem!)
Jun 13 21:51:09 <gnutoo> yes?
Jun 13 21:51:15 <max_posedon> Sat Jan  1
Jun 13 21:51:21 <max_posedon> see your date
Jun 13 21:51:23 <gnutoo> ah ouch...the openmoko?
Jun 13 21:51:30 <max_posedon> yes
Jun 13 21:51:34 <gnutoo> they must be syced...
Jun 13 21:52:17 <max_posedon> sorry, correct dates is important
Jun 13 21:52:30 <max_posedon> I see Sat Jan 1 2000
Jun 13 21:52:41 <gnutoo> yes I know but i didn't spot it
Jun 13 21:53:25 <max_posedon> I had this 3 days ago)
Jun 13 21:53:35 <max_posedon> wrong date -> mind break for a 1 hour)
Jun 13 21:54:16 <gnutoo> ah maybe I get it
Jun 13 21:54:23 <gnutoo> wrong date from the GSM network maybe
Jun 13 21:54:36 <max_posedon> GSM network can provide time?!
Jun 13 21:54:52 <gnutoo> maybe
Jun 13 21:54:59 <gnutoo> GPS can
Jun 13 21:55:03 <max_posedon> I thought it can't only TZ, time from GPS
Jun 13 21:55:16 <gnutoo> if  remember well only some of them do it but I'm not shure
Jun 13 21:55:38 <juri_> ok, back to life.
Jun 13 21:55:41 <gnutoo> mmm...connection broke
Jun 13 21:55:47 <max_posedon> logs?)
Jun 13 21:55:49 <gnutoo> and date is ok
Jun 13 21:55:51 <gnutoo> now
Jun 13 21:55:56 <max_posedon> both plz
Jun 13 21:56:03 <max_posedon> np, we will fix it)
Jun 13 21:56:06 <gnutoo> ok
Jun 13 22:00:01 <gnutoo> http://pastebin.ca/1459338 for the openvpn log
Jun 13 22:00:42 <max_posedon> ping gnutoo.homelinux.org
Jun 13 22:00:46 <gnutoo> http://pastebin.ca/1459340 for the logread
Jun 13 22:00:51 <gnutoo> both from the openmoko
Jun 13 22:01:32 <gnutoo> from where?
Jun 13 22:01:38 <gnutoo> from my desktop it works fine
Jun 13 22:01:47 <max_posedon> sorry, can't understand now
Jun 13 22:01:54 <max_posedon> what is time on your neo?
Jun 13 22:02:18 <max_posedon> for me, it looks like your gprs connection just doesn't started properly
Jun 13 22:03:04 <max_posedon> non-working pings, you know, it different issue
Jun 13 22:03:25 <gnutoo> http://pastebin.ca/1459342 log of the server
Jun 13 22:03:34 <gnutoo> ok 
Jun 13 22:03:57 <gnutoo> sun jun 14 00:03:45 cest 2009
Jun 13 22:04:07 <gnutoo> strange...
Jun 13 22:04:21 <gnutoo> I'll restart the openmoko
Jun 13 22:11:27 <gnutoo> it works again
Jun 13 22:11:29 <gnutoo> thanks a lot
Jun 13 22:12:25 <gnutoo> strange a ppp connection with 1500 as mtu...is it normal?
Jun 13 22:14:56 <gnutoo> max_posedon, when the gprs doesn't work becuase of the sip over vpn...I don't have access to my messages
Jun 13 22:15:01 <gnutoo> I'll try mickeyterm
Jun 13 22:16:27 <gnutoo> mmm dbus reject messages
Jun 13 22:17:28 <max_posedon> so, I know this problem
Jun 13 22:17:35 <max_posedon> e.g. have exactly same
Jun 13 22:17:38 <gnutoo> ok
Jun 13 22:17:59 <max_posedon> but completly don't know where it comes from and how to fix(
Jun 13 22:17:59 <gnutoo> did you bugreport ? 
Jun 13 22:18:22 <max_posedon> somebody else did it, I don't rember nick 
Jun 13 22:18:34 <gnutoo> do you  have the bug number?
Jun 13 22:18:35 <max_posedon> and don't know bug number...
Jun 13 22:18:39 <gnutoo> ah ok
Jun 13 22:18:55 <gnutoo> do you think a MTU of 1500 is normal for a ppp connection?
Jun 13 22:20:41 <max_posedon> I tried to change to 1000, without help
Jun 13 22:20:47 <gnutoo> ah ok
Jun 13 22:22:08 <max_posedon> gnutoo, I think http://trac.freesmartphone.org/ticket/384 this is bug what you have
Jun 13 22:22:20 <gnutoo> thanks a lot
Jun 13 22:26:15 <gnutoo> modem seems gone...
Jun 13 22:26:22 <gnutoo> it needs a restart
Jun 13 22:26:29 <gnutoo> I checked with picocom
Jun 13 22:34:26 <[Rui]> hi, anyone with experience on libxml2 sax interface? I've got a 1st attempt doubt
Jun 13 22:34:40 <[Rui]> why's on topic? making an identica client for om :)
Jun 13 22:35:03 <[Rui]> well, generically µblogging, but focusing on identica for starters.
Jun 13 22:36:31 <[Rui]> xmlSAXUserParseMemory return 0 but none of the callback functions seem to get called.
Jun 13 22:37:00 <max_posedon> do you really want sax interface?
Jun 13 22:37:07 <max_posedon> why not DOM?
Jun 13 22:37:38 <[Rui]> max_posedon: two reasons: 1) dom eats lots of memory, 2) sax is the way to go for many repeated kinds of elements
Jun 13 22:38:18 <max_posedon> 1) not so much I think, kinda 10x from xml size
Jun 13 22:38:39 <max_posedon> 2) ok, I haven't expirence with sax anyway
Jun 13 22:38:49 <max_posedon> I used libxml2 reader interface
Jun 13 22:38:57 <[Rui]> rms@roque:~/svn/elmdentica$ wc -c z == 33119 z
Jun 13 22:39:16 <[Rui]> so 10x is like 330KB
Jun 13 22:39:20 <max_posedon> yep
Jun 13 22:39:28 <[Rui]> so my point feels nicely justified :)
Jun 13 22:39:55 <max_posedon> hm.. ok, as you wish
Jun 13 22:40:08 <[Rui]> max_posedon: you don't have much memory to waste on the free runner
Jun 13 22:40:29 <[Rui]> max_posedon: but thanks anyway :)
Jun 13 22:40:34 <max_posedon> 300KB isn't stuff about what I should think, with runned python
Jun 13 22:41:07 <max_posedon> and you won't keep DOM in memory all time, just for a few seconds
Jun 13 22:41:31 <max_posedon> ok, this is your choise anyway.
Jun 13 22:43:17 <[Rui]> I don't think I'm doing it wrong, but the callbacks don't seem to be getting called.
Jun 13 22:43:31 <[Rui]> right now all they do is printf("startDocument") et all
Jun 13 22:50:47 <[Rui]> max_posedon: aha! the example I was following had it wrong, of course it couldn't be a static variable :D phwe
Jun 13 22:50:59 <[Rui]> now it seems to work
Jun 13 23:33:16 <khiraly11> http://khiraly.googlepages.com/Screenshot-2.png
Jun 13 23:52:22 <DocScrutinizer2> ftw is wrong with freenode today?
Jun 13 23:53:55 <DocScrutinizer2> wtf even
Jun 14 00:24:44 <Sargun> Ping
Jun 14 00:36:52 <budfive> anybody know how much current the freerunner should consume if it's on (not asleep), but none of the peripherals are active (screen off, gsm off, wifi off, etc)?
Jun 14 00:49:27 <Toaster`> Greetings all
Jun 14 01:01:01 <playya> Sargun, maybe you should send your ping to someone
Jun 14 01:02:56 <Sargun> multicast ping it was.
Jun 14 01:03:20 <Sargun> How do I send this message in dbus:
Jun 14 01:03:20 <Sargun> <Sargun> SystemBus, busname: com.palm.messagingrouter, object_path: /sms, interface: org.json, method: sendsms, contents: {"body":"lolbamaishappy","priority":"normal","","deliveryReportOptions":"none","callbackNumber":""}
Jun 14 01:03:23 <Sargun> from only the commandline
Jun 14 01:04:18 <playya> do you have dbus-send
Jun 14 01:04:25 <playya> or even better mdbus?
Jun 14 01:05:51 <Sargun> I have dbus-send
Jun 14 01:06:08 <playya> hmm. the tuple might be a problem
Jun 14 01:06:09 <playya> 1 s
Jun 14 01:06:16 <playya> do you have python?
Jun 14 01:06:26 <playya> and a palm pre?
Jun 14 01:07:01 <Sargun> I have a palm pre, how did you guess
Jun 14 01:07:05 <Sargun> and no python
Jun 14 01:07:09 <Sargun> the content is a string.
Jun 14 01:07:16 <Sargun> I should have indicated that
Jun 14 01:07:18 <playya> ok
Jun 14 01:08:07 <Sargun> I was the guy who wrote the palm pre rooter for windows.
Jun 14 01:09:00 <playya> dbus-send --system --print-reply /sms org.json.sendsms 'dict:string:string:"body":"lolbamaishappy","priority":"normal","","deliveryReportOptions":"none","callbackNumber":""'
Jun 14 01:09:38 <playya> oh
Jun 14 01:09:46 <playya> busname missing ;9
Jun 14 01:10:08 <playya> dbus-send --system --print-reply --dest= /com.palm.messagingrouter sms org.json.sendsms 'dict:string:string:"body":"lolbamaishappy","priority":"normal","","deliveryReportOptions":"none","callbackNumber":""'
Jun 14 01:10:26 <playya> hmm. my berentzen is empty
Jun 14 01:10:43 <playya> and nobody wants to drink wodka with me :(
Jun 14 01:10:59 <playya> Sargun, any errors?
Jun 14 01:11:37 <playya> do you know #webos-internals 
Jun 14 01:13:30 <Sargun> Yeah
Jun 14 01:13:34 <Sargun> I was one of the founders.
Jun 14 01:13:39 <Sargun> I would drink vodka with you
Jun 14 01:14:10 <Sargun> f 
Jun 14 01:15:00 <Sargun> dbus-send --system --print-reply --dest= /com.palm.messagingrouter sms org.json.sendsms string:'{"body":"Foopdayoop","priority":"normal","address":"x
Jun 14 01:15:00 <Sargun> bmodder@gmail.com","deliveryReportOptions":"none","callbackNumber":""}'
Jun 14 01:15:40 <Sargun> Must use org.mydomain.Interface.Method notation, no dot in "sms"
Jun 14 01:17:34 <khiraly11> Sargun: do palm pre use X11?
Jun 14 01:17:43 <Sargun> no, framebuffer
Jun 14 01:17:55 <khiraly11> pure framebuffer?
Jun 14 01:18:18 <khiraly11> so no some kind of windowing? nothing?
Jun 14 01:18:39 <Sargun> yup
Jun 14 01:18:48 <Sargun> well, they wrote their own window manager
Jun 14 01:18:51 <khiraly11> Im playing with the idea how hard it could be installing linux and x11, and having shr on it
Jun 14 01:19:03 <khiraly11> and be able to use the broadband chip
Jun 14 01:19:07 <Sargun> khiraly11, join #webos-internals and we'll talk about it
**** ENDING LOGGING AT Sun Jun 14 02:59:57 2009