**** BEGIN LOGGING AT Fri Apr 17 02:59:57 2009
Apr 17 03:59:57 * mwester is incredibly annoyed with nytowl for commiting a change to milestone5.5 with his name name on it that doesn't look anything at all like what was submitted originally, and is in fact not useful at all. 
Apr 17 04:01:08 <DocScrutinizer> hmm
Apr 17 04:02:47 <mwester> Ainulindale:  we're going to stick with patching the 2.628 kernel recipe; I'll fix the version number a bit so that people know its actually a 2.6.29 kernel.
Apr 17 04:17:53 <mwester> Ainulindale:  we also need to "de-autorev" the kernels; upgrades will fail unless we do this (git version number are hashes, not monotonically increasing values).
Apr 17 04:18:32 <mwester> We'll have to manually bump up and adjust the "-oe1" value each time (which I'm going to change to a two-digit numeric as well).
Apr 17 04:25:46 <raster> mwester:  make it 4 digits
Apr 17 04:25:49 <raster> allow for... space
Apr 17 04:25:51 <raster> :)
Apr 17 04:45:43 <pundiramit> hi.. can anyone tell me (or point to me any reference doc) how does openmoko s/w stack uses SMedia Glamo 3D accelerator? or how can an application make use of 3D accelerator?
Apr 17 04:49:46 <CIA-75> SHR: 03mwester 07shr-overlay * rbda200d9cf04 10/openembedded/recipes/shr/frameworkd-config-shr_git.bb: Merge branch 'master' of git@shr.bearstech.com:shr-overlay
Apr 17 04:49:47 <CIA-75> SHR: 03mwester 07shr-overlay * r234e465cd9b0 10/patches/kernel/0001-kernel-remove-patches-and-adjust-config-name.patch: shr-autorev: nail-down the gta02 kernel; also update kernel patch to
Apr 17 05:25:28 <CIA-75> SHR: 03mwester 07shr-overlay * rf66889fcc95e 10/patches/fastboot/ (3 files): SHR patches: fastboot - updated netbase, opkg, and fstab patches so they
Apr 17 05:29:52 <CIA-75> SHR: 03mwester 07shr-overlay * r459f6875645d 10/openembedded/recipes/shr/ (initscripts-shr/g_ether.sh initscripts-shr_git.bb): initscript-shr: g_ether - suppress stderr output; avoid errors being printed
Apr 17 05:35:36 <ptitjes> Ainulindale: ping
Apr 17 05:38:46 <CIA-75> SHR: 03mwester 07shr-overlay * r1a5f07f09ae4 10/openembedded/recipes/images/shr-image.inc: shr-image.inc - fix Exec line for htop
Apr 17 05:41:08 <CIA-75> SHR: 03mwester 07shr-overlay * rfb3de279fc18 10/openembedded/conf/distro/include/shr-autorev.inc: shr-autorev.inc: Fix the gta02 kernel to a specific git hash, add
Apr 17 06:07:08 <PaulFertser> pundiramit: There was some preliminary work on DRM module. Nothing usable yet. Moreover, glamo 3D capabilities are very limited, basically usable only on QVGA with other restrictions.
Apr 17 06:14:25 <ptitjes> PaulFertser: really ?
Apr 17 06:15:00 <PaulFertser> ptitjes: which part of my statement are you questioning?
Apr 17 06:15:18 <ptitjes> PaulFertser: "glamo 3D capabilities are very limited, basically usable only on QVGA with other restrictions"
Apr 17 06:15:20 <PaulFertser> http://git.openmoko.org/?p=xf86-video-glamo.git;a=shortlog;h=refs/heads/dri-aware
Apr 17 06:15:27 <PaulFertser> ptitjes: really
Apr 17 06:15:41 <ptitjes> PaulFertser: does this support texture rendering ?
Apr 17 06:16:04 <PaulFertser> ptitjes: texture size is limited by 512x512 iirc
Apr 17 06:16:42 <PaulFertser> ptitjes: or even 256x256
Apr 17 06:16:47 <ptitjes> PaulFertser: but many texture at the same time ?
Apr 17 06:16:59 <ptitjes> (in the same scene)
Apr 17 06:17:34 <ptitjes> PaulFertser: my question in fact is do you think compositing is possible
Apr 17 06:19:21 <ptitjes> PaulFertser: or making GUIs with Clutter ?
Apr 17 06:19:23 <PaulFertser> ptitjes: i don't know. Raster discussed it to death so many times that i start to wonder how he can remain sane after it. Using glamo 3D engine for VGA desktop is not possible anyway. Some tricks can be done to display a 2D area and a smaller 3D area but it requires too much effort for little (or no) gain because of the other problems. I'm in no way expert in this stuff, but i guess if you search the ML archives, you'll find plenty of info ab
Apr 17 06:19:30 <PaulFertser> ... Glamo limitations. 
Apr 17 06:19:55 <ptitjes> raster can't be sane either ways
Apr 17 06:19:57 <ptitjes> :)
Apr 17 06:20:36 <ptitjes> arf my dreams go away...
Apr 17 06:20:53 <PaulFertser> ptitjes: clutter was addressed by raster the day before yesterday. Bottomline: evas is better, supports more engines; opengl es sucks; glamo was designed for QVGA. 
Apr 17 06:21:22 <ptitjes> arf
Apr 17 06:22:16 <ptitjes> too bad
Apr 17 06:22:32 <ptitjes> PaulFertser: you and raster ruined my day
Apr 17 06:22:34 <ptitjes> pffff
Apr 17 06:22:35 <ptitjes> :D
Apr 17 06:23:08 <ptitjes> gtg
Apr 17 06:23:12 <PaulFertser> ptitjes: i thought everybody knew that glamo gives no advantages over dumb framebuffer.
Apr 17 06:23:18 <ptitjes> PaulFertser: have a nice day
Apr 17 06:23:25 <PaulFertser> ptitjes: have a nice day too :)
Apr 17 06:23:42 <ptitjes> PaulFertser: but I never aware of nothing (this is known)
Apr 17 06:23:43 <ptitjes> :)
Apr 17 06:23:49 <ptitjes> bye
Apr 17 06:29:15 <PaulFertser> pundiramit: DRM parts of glamo driver: http://git.openmoko.org/?p=kernel.git;a=shortlog;h=refs/heads/drm-tracking
Apr 17 06:29:33 <PaulFertser> pundiramit: mesa: http://git.bitwiz.org.uk/?p=mesa.git;a=shortlog
Apr 17 06:30:10 <DocScrutinizer> http://www.google.com/search?q=site%3Alists.openmoko.org+glamo+raster
Apr 17 06:31:46 * DocScrutinizer looks for nice glasses and hearing-aid for ptitjes
Apr 17 06:32:16 <PaulFertser> DocScrutinizer: :)
Apr 17 06:32:36 <DocScrutinizer> to help him out of his state of unawareness
Apr 17 06:33:22 <DocScrutinizer> ~glamo is http://www.google.com/search?q=site%3Alists.openmoko.org+glamo+raster
Apr 17 06:33:22 <bzzbot> DocScrutinizer: okay
Apr 17 06:33:25 <raster> PaulFertser: but useless without the rest of the libGL... glue
Apr 17 06:33:34 <raster> ie just giving access to the video reigster maps is easy
Apr 17 06:33:44 <raster> its filling them with the right data at the right time...
Apr 17 06:34:02 <PaulFertser> raster: Thomas is working on mesa part too, it seems
Apr 17 06:34:30 <raster> and even if it gets done... my money is on it being horribly disappointing
Apr 17 06:34:38 <raster> ie it may finally work
Apr 17 06:34:44 <raster> for basic levels of "work"
Apr 17 06:34:49 <raster> and then perform like total crud
Apr 17 06:34:59 <raster> not to mentionj apps will need to code around the glamo limit specially
Apr 17 06:35:15 <DocScrutinizer> hehe, hardware-decelleration
Apr 17 06:35:28 <raster> you'll try and use bigger textures
Apr 17 06:35:31 <raster> as that is NORMAL
Apr 17 06:35:47 <raster> last time gl hw has 256x256 limi was in the voodoo days
Apr 17 06:35:58 <raster> and evrey dev is happy as larry that that has gone away
Apr 17 06:36:06 <raster> working around that is ugly and painfulk
Apr 17 06:36:10 <raster> u need to tyurn everything into meshes
Apr 17 06:36:28 <DocScrutinizer> *shudder*
Apr 17 06:36:31 <raster> (multiple 256x256 textures over a grid of quads)
Apr 17 06:36:50 <raster> which screws with scaling as u now have seams every 256 source pixels
Apr 17 06:36:59 <raster> as u cant interpolate between 2 textures...
Apr 17 06:37:08 <spaetz>  morning
Apr 17 06:37:09 <raster> unless.. u now go to 255x255 and u duplicate edge pixels....
Apr 17 06:37:11 <DocScrutinizer> uahaaaaha
Apr 17 06:37:25 <PaulFertser> raster: aren't you tired to explain it all again and again? ;)
Apr 17 06:37:28 <raster> now if u cant do this - u have to reduce all textures to 256x256 or lower
Apr 17 06:37:32 <DocScrutinizer> sounds like BIG fun
Apr 17 06:37:39 <raster> whihc cuts qquality and adds yet another texture upload stage
Apr 17 06:37:52 <raster> it's totally not fun
Apr 17 06:38:05 <raster> and every dev iis toyally happy modern 3d has limits that are more usable
Apr 17 06:38:09 <raster> like 2048x2048
Apr 17 06:38:11 <raster> or higher
Apr 17 06:38:27 <raster> (6410 is 4096x4096 max tex size)
Apr 17 06:38:41 <DocScrutinizer> nah, there's no sane dev left over to appreciate the difference ;-)
Apr 17 06:38:59 <DocScrutinizer> all those who could tell gone nuts
Apr 17 06:39:30 <spaetz> shouldn't 256x256 pixel be enough for everybody?
Apr 17 06:39:44 <DocScrutinizer> raster: would you like to argue against that? ;-)
Apr 17 06:39:45 <spaetz> ahh no, that was somebody else saying something simila
Apr 17 06:39:47 <spaetz> r
Apr 17 06:40:00 <raster> HAHAHHAHAHA
Apr 17 06:40:07 <gurugentoo> yay Bill
Apr 17 06:40:22 <raster> i was working on evas using 256x256 texture meshes back in 1999 and 2000
Apr 17 06:40:28 <raster> it was a pita
Apr 17 06:40:32 <raster> the results were sub-par
Apr 17 06:40:44 <raster> i am happy to not have to care about it ever again
Apr 17 06:41:02 <PaulFertser> raster: on an unrelated note, i found a cool book about computer architecture in general and it uses MIPS assembly to explain low-level stuff. Too bad there's no russian translation. David A. Patterson, John L. Hennesy Computer Organization and Design. The Hardware/Software
Apr 17 06:41:07 <PaulFertser> Interface
Apr 17 06:42:02 <raster> PaulFertser:  mips is a good start
Apr 17 06:42:16 <raster> though unfortunately in industry its close to dead
Apr 17 06:42:24 <PaulFertser> And there exists a MIPS virtual machine called SPIM which has special syscalls for input/output. Exactly what is needed for educational purposes.
Apr 17 06:42:46 <spaetz> Ainulindale: morning. Are you fine with me renaming openmoko*3 components to shr-* ?
Apr 17 06:42:59 <spaetz> at least I can't break git branches this way
Apr 17 06:44:15 <max_posedon> hm... `3D` neo limited to qvga, actually it isn't problem for games, but can we hope for at least little bit faster 2D?
Apr 17 06:44:49 <max_posedon> for example for fast/looking nice scrolling
Apr 17 06:45:51 <PaulFertser> max_posedon: something can be done by application designers to avoid slow paths, but it requires much work for little gain. iirc.
Apr 17 06:50:31 <max_posedon> in short, video part brokes myth that with neo, you can easy using you desktop apps without rewriting them
Apr 17 06:51:00 <raster> PaulFertser: oooh reallyu! cool! spim sounds like XMA - thats what my school used to introduce asm to sw enigneers. u has special "print" asm calls - totally not like a real machine
Apr 17 06:51:07 <raster> but good as u had a quick way to get debug out of it
Apr 17 06:51:20 <raster> also it had an "input" command to get input (only stdin/out)
Apr 17 06:51:21 <max_posedon> good neo app should use elementery, avoid slow paths, no opengl
Apr 17 06:51:51 <raster> max_posedon:  qvga is a problem. tried to run qvga lately?
Apr 17 06:51:55 <raster> it doesnt display
Apr 17 06:51:58 <raster> all garbled up
Apr 17 06:52:35 <PaulFertser> raster: yep, syscall list is here: http://en.wikipedia.org/wiki/SPIM 
Apr 17 06:54:11 <spaetz> mmh git plugin for shr trac would be nice
Apr 17 06:54:35 <spaetz> refer to commits when closing bugs etc
Apr 17 06:54:39 <PaulFertser> According to https://docs.openmoko.org/trac/ticket/2263 QVGA works but some bug left.
Apr 17 06:54:50 <max_posedon> I tried, it doesn't work now, but worked in October, I think this is just bug, that will be fixed, sooner or later. But I speak in general.
Apr 17 06:54:51 <spaetz> according to user feedback it all garbles up
Apr 17 06:54:54 <raster> PaulFertser: actually try it... :)
Apr 17 06:55:09 <spaetz> it's all garbled up...
Apr 17 06:55:20 <spaetz> or whatever, my grammar sucks before first coffee)
Apr 17 06:56:14 <PaulFertser> raster: i don't run xorg yet :( and yes, just tried it, doesn't work with xglamo. But who cares about xglamo.
Apr 17 06:56:30 <raster> xorg is the same code as xglamo
Apr 17 06:56:36 <raster> the same code is being ported over
Apr 17 06:56:41 <raster> and without xglamo. u get 0 accel
Apr 17 06:56:49 <raster> a dum fb with fbdev
Apr 17 06:56:54 <raster> which is even more attrocious
Apr 17 06:56:59 <raster> tho most efl apps wont much care
Apr 17 06:57:05 <raster> they'll be about the same
Apr 17 06:57:26 <PaulFertser> raster: i think xorg glamo driver does almost everything xglamo did. 
Apr 17 06:58:06 <raster> its gettign there
Apr 17 06:58:11 <raster> but it will inherit all the same things
Apr 17 06:58:16 <raster> anyway
Apr 17 06:58:23 <PaulFertser> At least it's maintained unlike xglamo.
Apr 17 06:58:25 <raster> if qvga worked.. i'd consdier using it by default
Apr 17 06:58:53 <PaulFertser> Even with FR's screen that is supposed to look blocky (?) in QVGA?
Apr 17 07:06:58 <schasch> -
Apr 17 07:07:31 <PaulFertser> raster: just tried QVGA, it works.
Apr 17 07:08:06 <max_posedon> good time forget about xglamo
Apr 17 07:10:00 <PaulFertser> TS-calibration is wrong though, but it works. With xglamo even. echo qvga-normal > /sys/bus/spi/devices/spi2.0/state
Apr 17 07:10:20 <PaulFertser> xglamo just uses an old sysfs path and nobody wish to touch it.
Apr 17 07:12:38 <PaulFertser> TS-calibration is not wrong, just remained 480x640.
Apr 17 07:12:45 <PaulFertser> And man, it looks ugly!
Apr 17 07:13:15 <PaulFertser> (comparing to ultra-crisp 285 DPI)
Apr 17 07:14:07 <PaulFertser> max_posedon: so, QVGA works, don't complain.
Apr 17 07:19:50 <PaulFertser> raster: This leads me to a conclusion (given that QVGA worked before and works now) that it's so uncool that nobody likes it.
Apr 17 07:26:14 <max_posedon> PaulFertser, I disagree
Apr 17 07:26:40 <max_posedon> many mails sent to OM-community about `how to run games`
Apr 17 07:32:52 <PaulFertser> max_posedon: why don't they use QVGA then?
Apr 17 07:33:14 <max_posedon> they use qvga
Apr 17 07:33:30 <max_posedon> and adds some hacks to running scripts
Apr 17 07:33:43 <max_posedon> that echo, etc.
Apr 17 07:33:55 <PaulFertser> max_posedon: if it was the case xglamo wouldn't be broken for more than 4 months.
Apr 17 07:35:43 <PaulFertser> max_posedon: it's so easy: just craft a patch and ask somebody from SHR to include it in the overlay. Nobody did it -> nobody needs it.
Apr 17 07:36:34 <DocScrutinizer> max_posedon: echo??? ECHO??!!
Apr 17 07:37:08 <DocScrutinizer> don't say this word! >:-(
Apr 17 07:37:15 <DocScrutinizer> ;)
Apr 17 07:38:00 * DocScrutinizer reverts to reading calypso audio-dsp AEC sources
Apr 17 07:38:36 <DocScrutinizer> well "sources"
Apr 17 07:56:08 <Azog> PaulFertser: how do you switch to qvga? xrandr -display :0 -s 240x320?
Apr 17 09:11:38 <Ainulindale> Hey there
Apr 17 09:18:24 <Ainulindale> mickey|zzZZzz, alphaone?
Apr 17 09:18:32 <Ainulindale> I experienced the "no suspend" bug
Apr 17 09:18:40 <alphaone> morning Ainulind1le 
Apr 17 09:20:02 <Ainulindale> Morning, sorry :-)
Apr 17 09:20:22 <Ainulindale> What I did, to check
Apr 17 09:20:28 <Ainulindale> Plugged in
Apr 17 09:20:29 <Ainulindale> root@om-gta02 ~ $ mdbus -s org.freesmartphone.ousaged /org/freesmartphone/Usage org.freesmartphone.Usage.GetResourceUsers "CPU
Apr 17 09:20:32 <Ainulindale> "
Apr 17 09:20:34 <Ainulindale> [':1.9']
Apr 17 09:20:39 <Ainulindale> (it's expected as it'll request the CPU)
Apr 17 09:20:42 <Ainulindale> unplugged
Apr 17 09:20:43 <Ainulindale> the same
Apr 17 09:20:53 <Ainulindale> logs are showing that the CPU resource was disabled
Apr 17 09:21:42 <Ainulindale> it seems to be a "ghost" CPU requester on the framework side
Apr 17 09:25:03 <Ainulindale> alphaone: does it mean something to you?
Apr 17 09:25:32 <Ainulindale> alphaone: I don't recall anything weird apart from restarting the framework while plugged in
Apr 17 09:26:16 <alphaone> Ainulindale: Hmm, what process is behind :1.?
Apr 17 09:26:18 <alphaone> 9
Apr 17 09:26:19 <Ainulindale> Plugging it in resetted the things
Apr 17 09:26:25 <Ainulindale> alphaone: framework
Apr 17 09:26:29 <schasch> Hello, can I switch EABI to OABI in OE-bitbake/mokomakefile for gta01 or gta02?
Apr 17 09:26:31 <alphaone> strange
Apr 17 09:26:50 <Ainulindale> yeah
Apr 17 09:26:54 <Ainulindale> I'm scping the log
Apr 17 09:27:25 <Ainulindale> Wooh, 120k, nifty
Apr 17 09:27:45 <Ainulindale> Heh
Apr 17 09:27:47 <Ainulindale> alphaone: 2009.04.16 23:50:36.750 ogpsd                WARNING  (could not write to '/sys/bus/platform/devices/neo1973-pm-gps.0/pwron': [Errno 2] No such file or directory: '/sys/bus/platform/devices/neo1973-pm-gps.0/pwron'
Apr 17 09:27:51 <Ainulindale> Is this intended?
Apr 17 09:28:32 <alphaone> Ainulindale: Yes, we're writing to two paths..
Apr 17 09:28:48 <alphaone> Some kernel version changed them
Apr 17 09:29:01 <Ainulindale> 2009.04.16 23:51:06.261 dbus.proxies         ERROR    Introspect error on org.freesmartphone.odeviced:/org/freesmartphone/Device/LED/gta02_power_orange: dbus.exceptions.DBusException: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freesmartphone.odeviced was not provided by any .service file <= that too then?
Apr 17 09:29:43 <Ainulindale> alphaone: 2009.04.16 23:51:13.880 ogpsd                WARNING  Attempted to send something while serial is not open
Apr 17 09:29:49 <Ainulindale> and again 2009.04.16 23:51:14.11 ogpsd                WARNING  (could not write to '/sys/bus/platform/devices/neo1973-pm-gps.0/pwron': [Errno 2] No such file or directory: '/sys/bus/platform/devices/neo1973-pm-gps.0/pwron'
Apr 17 09:30:07 <alphaone> I believe that's due to odeviced not offering these paths yet
Apr 17 09:30:56 <Ainulindale> hmmmm alphaone 
Apr 17 09:30:57 <Ainulindale> I was wondering
Apr 17 09:31:02 <Ainulindale> Could that problem be due to a race?
Apr 17 09:31:03 <alphaone> Yes, mickey|zzZZzz insists on disabling GPS when we start frameworkd for some reason I still don't understand
Apr 17 09:31:04 <Ainulindale> You try to call
Apr 17 09:31:06 <Ainulindale> It requests CPU
Apr 17 09:31:11 <Ainulindale> Call is failing before CPU is requested
Apr 17 09:31:23 <Ainulindale> Signal to release CPU is sent before the request
Apr 17 09:31:33 <alphaone> Hmm
Apr 17 09:31:42 <Ainulindale> Do you want my log?
Apr 17 09:32:04 <alphaone> Could be, but CPU state transitions really happen almost instantanious
Apr 17 09:32:16 <alphaone> So I guess it's highly unlikely
Apr 17 09:32:59 <billk> ERROR: '/home/moko/SHR/shr-testing/shr/openembedded/recipes/shr/shr-contacts_git.bb' failed
Apr 17 09:33:00 <billk> make: *** No rule to make target `install'.  Stop.
Apr 17 09:33:02 <Ainulindale> Well yes, but then I can't explain why when I woke up it wouldn't suspend
Apr 17 09:33:10 <billk>  building shr-testing
Apr 17 09:33:16 <Ainulindale> Hmmmm although it could also be ophonekitd failing
Apr 17 09:33:50 <Ainulindale> billk: checking
Apr 17 09:34:09 <billk> Ainulindale: tkx
Apr 17 09:34:22 <Ainulindale> by the way alphaone 
Apr 17 09:34:28 <Ainulindale> I came up with a use case for my idea
Apr 17 09:34:34 <Ainulindale> the one about releasing resource from another owner
Apr 17 09:34:43 <Ainulindale> or releasing resources for a specific process
Apr 17 09:34:46 <Ainulindale> I posted that on the ML
Apr 17 09:34:49 <Ainulindale> (shr-devel)
Apr 17 09:35:06 <alphaone> Ainulindale: Okay, need to sift through my inbox again..
Apr 17 09:35:57 <Ainulindale> alphaone: subject is fsoraw something
Apr 17 09:36:04 <Ainulindale> alphaone: by the way I confirm, I have both logs ophonekitd/frameworkd
Apr 17 09:36:14 <Ainulindale> ophonekitd doesn't receive idle status suspend
Apr 17 09:36:30 <alphaone> okay
Apr 17 09:38:27 <Ainulindale> YES!
Apr 17 09:38:30 <Ainulindale> Found the issue 
Apr 17 09:38:32 <Ainulindale> I was right I think
Apr 17 09:39:06 <Ainulindale> (god bless logs)
Apr 17 09:39:54 <Ainulindale> http://pastebin.archlinux.fr/344279 alphaone 
Apr 17 09:39:59 <Ainulindale> I had no signal
Apr 17 09:40:01 <Ainulindale> But tried to call
Apr 17 09:40:22 <Ainulindale> I had #
Apr 17 09:40:22 <Ainulindale> 2009.04.17 10:27:30.349 ogsmd.device         INFO     org.freesmartphone.GSM.Call.CallStatus: 1 outgoing {'status': 'outgoing', 'peer': '0668513945', 'line': 0, 'direction': 'outgoing'}
Apr 17 09:40:27 <Ainulindale> #
Apr 17 09:40:29 <Ainulindale> 2009.04.17 10:27:30.400 ogsmd.device         INFO     org.freesmartphone.GSM.Call.CallStatus: 1 release {'status': 'release'}
Apr 17 09:40:32 <Ainulindale> And oeventsd reacted later
Apr 17 09:41:06 <Ainulindale> hey max_posedon !
Apr 17 09:41:11 <Ainulindale> Wanna talk about your visa issue?
Apr 17 09:41:29 <Ainulindale> (I'll be right back in ten minutes, please describe to me everything you need to know or need to have in query)
Apr 17 09:41:34 <max_posedon> Hi! no already
Apr 17 09:41:38 <alphaone> Okay, so oeventsd ist (again) the culprit here...
Apr 17 09:42:06 <max_posedon> embassy agree give me visa without "verpflichtungserklaerung")
Apr 17 09:43:04 <max_posedon> Ainulind1le, so, I just need invitation from LinuxHotel, and already contacted them for it.
Apr 17 09:44:11 <alphaone> I would guess the oeventsd call to Request/ReleaseResource should block.
Apr 17 09:44:25 <alphaone> But then everything would become much slower again..
Apr 17 09:46:38 <Ainulindale> alphaone: well I don't think so
Apr 17 09:46:46 <Ainulindale> alphaone: I think you should "queue" certain types of events
Apr 17 09:47:39 <alphaone> Ainulindale: Maybe have a rule that waits for CPU resource to be enabled until it releases it
Apr 17 09:47:54 <alphaone> But you can't do that in oeventsd currently
Apr 17 09:48:12 <Ainulindale> yes
Apr 17 09:48:25 <alphaone> I still like my idea of using a state machine syntax
Apr 17 09:48:40 <alphaone> Would make most of these things easier to describe
Apr 17 09:57:13 <alphaone> Ainulindale: I'm again caugh in a redirection loop on the shr trac :-(
Apr 17 09:59:06 <Ainulindale> for login?
Apr 17 10:01:20 <alphaone> yes
Apr 17 10:01:44 <alphaone> Ainulindale: "Redirect Loop
Apr 17 10:01:44 <alphaone> Firefox has detected that the server is redirecting the request for this address in a way that will never complete."
Apr 17 10:01:52 <Ainulindale> Well if I delete then recreate your account it'll work, it worked for methril|fix_part 
Apr 17 10:01:52 <alphaone> For http://trac.shr-project.org/trac/prefs/account
Apr 17 10:01:57 <Ainulindale> Just there,
Apr 17 10:01:57 <Ainulindale> ?
Apr 17 10:02:24 <alphaone> No matter where I go I always get redirected there
Apr 17 10:02:38 <alphaone> And from ther I probably get redirected there again
Apr 17 10:03:03 <Ainulindale> ok then I'll delete your account and you'll recreate it
Apr 17 10:03:05 <Ainulindale> is that good for you?
Apr 17 10:03:17 <alphaone> okay
Apr 17 10:04:04 <Ainulindale> which account is it?
Apr 17 10:05:18 <alphaone> daniel
Apr 17 10:05:22 <alphaone> should be it
Apr 17 10:05:30 <alphaone> daniel@totalueberwachung.de it the mail
Apr 17 10:05:35 <alphaone> ah wait
Apr 17 10:05:52 <alphaone> I meant my account name is Ainulindale :-)
Apr 17 10:05:57 <Ainulindale> pfff =)
Apr 17 10:06:00 <Ainulindale> deleted "daniel"
Apr 17 10:06:41 <alphaone> WTF?
Apr 17 10:06:50 <alphaone> *Now* I'm logged in as daniel
Apr 17 10:07:05 <Ainulindale> Ah you have another account
Apr 17 10:07:06 <Ainulindale> dwillman
Apr 17 10:07:08 <alphaone> Were there two accounts named daniel?
Apr 17 10:07:15 <Ainulindale> Deleting it as well
Apr 17 10:07:19 <alphaone> "# logged in as daniel
Apr 17 10:07:19 <alphaone> # Logout"
Apr 17 10:07:22 <Ainulindale> (with the same mail)
Apr 17 10:07:29 <Ainulindale> Done.
Apr 17 10:07:34 <Ainulindale> Please delete your cache
Apr 17 10:07:40 <alphaone> still, now I'm logged in as daniel
Apr 17 10:07:50 <Ainulindale> alphaone: http session and such I guess
Apr 17 10:07:54 <Ainulindale> exit your browser
Apr 17 10:08:10 <alphaone> looks like it
Apr 17 10:08:12 <alphaone> fun :-)
Apr 17 10:08:15 <edje2> hi, i did upgrade to frameworkd-9999 and now i sem to miss a file:
Apr 17 10:08:17 <edje2> dbus.proxies:Introspect error on org.freesmartphone.DBus:/org/freesmartphone/DBus: dbus.exceptions.DBusException: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freesmartphone.DBus was not provided by any .service files
Apr 17 10:08:19 <edje2> Does anyone know where that is supposed to come from?
Apr 17 10:10:07 <alphaone> Ainulindale: It's working now, thanks
Apr 17 10:10:20 <alphaone> Damn, now what did I actually want to do there?
Apr 17 10:13:58 <Ainulindale> billk_: works here
Apr 17 10:16:00 <edje2> also i get:
Apr 17 10:16:02 <edje2> DBus Exception. dbus-hlid not installed
Apr 17 10:16:04 <edje2> Where should that come from?
Apr 17 10:17:41 <Ainulindale> alphaone: 2009.04.17 12:16:06.707 ogspd.gpsdevice      INFO     enabling
Apr 17 10:17:44 <Ainulindale> 2009.04.17 12:16:06.743 ogpsd                WARNING  (could not write to '/sys/bus/platform/devices/neo1973-pm-gps.0/pwron': [Errno 2] No such file or directory: '/sys/bus/platform/devices/neo1973-pm-gps.0/pwron')
Apr 17 10:18:03 <alphaone> Ainulindale: Yes
Apr 17 10:18:19 <Ainulindale> not working then
Apr 17 10:18:20 <alphaone> As the write to the other path succeeds
Apr 17 10:18:34 <Ainulindale> well I see no NMEA output
Apr 17 10:19:16 <alphaone> Hmm, what path do you have?
Apr 17 10:20:17 <Ainulindale> power_on
Apr 17 10:23:04 <alphaone> what kernel are you using?
Apr 17 10:23:20 <Ainulindale> hmmm g_ether seems to be failing a lot here
Apr 17 10:23:21 <alphaone> Is power_on reporting 1 or 0?
Apr 17 10:24:03 <Ainulindale> 0
Apr 17 10:24:16 <alphaone> In ogpsd/om.py:
Apr 17 10:24:19 <alphaone> DEVICE_POWER_PATH_OLD = "/sys/bus/platform/devices/neo1973-pm-gps.0/pwron"
Apr 17 10:24:19 <alphaone> DEVICE_POWER_PATH_NEW = "/sys/bus/platform/devices/neo1973-pm-gps.0/power_on"
Apr 17 10:24:21 <Ainulindale> Linux om-gta02 2.6.29-rc3 #1 PREEMPT Thu Apr 16 17:58:30 CEST 2009 armv4tl unknown
Apr 17 10:24:29 <Ainulindale> Might that be due to conf?
Apr 17 10:24:55 <alphaone>     def initializeDevice( self ):
Apr 17 10:24:55 <alphaone>         helpers.writeToFile( DEVICE_POWER_PATH_OLD, "1" )
Apr 17 10:24:55 <alphaone>         helpers.writeToFile( DEVICE_POWER_PATH_NEW, "1" )
Apr 17 10:24:56 <Ainulindale> Ok launching tangogps I have 1
Apr 17 10:25:30 <alphaone> Don't know (due to conf)
Apr 17 10:25:43 <alphaone> Should be a patch some time ago iirc
Apr 17 10:25:45 <Ainulindale> well I have a power_on = 1 anyway
Apr 17 10:25:50 <alphaone> okay
Apr 17 10:25:58 <alphaone> But still no go?
Apr 17 10:26:13 <alphaone> telnet localhost gpsd
Apr 17 10:26:38 <alphaone> then type r<enter>
Apr 17 10:26:50 <alphaone> works here
Apr 17 10:27:31 <Ainulindale> i don't understand why but
Apr 17 10:27:37 <Ainulindale> as soon as framework dims the screen
Apr 17 10:27:39 <Ainulindale> my ssh session dies
Apr 17 10:27:56 <Ainulindale> and then I can't reconnect without plugging out/in
Apr 17 10:30:03 <alphaone> heh
Apr 17 10:30:08 <alphaone> that is strange
Apr 17 10:30:15 <alphaone> Doesn't happen here
Apr 17 10:31:00 <Azog> is the guy from braunschweig, who does the buzz-fixing, here?
Apr 17 10:31:10 <Ainulindale> It's not a guy it's a girl
Apr 17 10:31:17 <Ainulindale> Daniela Willman
Apr 17 10:31:37 <Azog> http://lists.openmoko.org/pipermail/community/2009-April/044807.html says daniel willmann
Apr 17 10:32:04 <Ainulindale> Yeah she signs that way in order to avoid to be importuned by overly active sexually frustrated geeks
Apr 17 10:32:38 <alphaone> Ainulindale: You're an ass :-P
Apr 17 10:32:51 <Ainulindale> Why is that? Because I told your secret?
Apr 17 10:32:54 <alphaone> Azog: I'm that girl
Apr 17 10:33:13 <Azog> alphaone: ah, hi
Apr 17 10:33:32 <Azog> i'm still a bit confused...
Apr 17 10:34:23 <alphaone> Ainulindale: I swear if someone tries to hit on me now I'll hit you at FSOSHRUDCON - hard.
Apr 17 10:34:26 <alphaone> Azog: Why
Apr 17 10:35:08 <Ainulindale> :-)
Apr 17 10:35:30 <Azog> alphaone: i have two freerunners (well a friend has the second), we plan to send them by mail to you for buzz-fixing; would that be ok?
Apr 17 10:37:54 <admiral0> Ainulindale:ping
Apr 17 10:38:11 <alphaone> Azog: I'm working on streamlining the mail in process, but I have very limited time at the moment..
Apr 17 10:38:31 <alphaone> Azog: But yes, that would be okay
Apr 17 10:39:20 <Ainulindale> admiral0: pong
Apr 17 10:39:49 <admiral0> Ainulindale: i've got an interesting lecture for making SHR fasterhttp://www.microxwin.com/
Apr 17 10:40:19 <Azog> alphaone: that's very cool; are you often here, so that i can contact you because of the address? i think in half a week, we decide, what we want to do
Apr 17 10:40:27 <admiral0> i'm at work now and i've seen it in a page, i wanted to share with SHR community
Apr 17 10:40:46 <admiral0> i'll test it when i'll have some time
Apr 17 10:41:57 <Ainulindale> Well check for the license
Apr 17 10:42:02 <Ainulindale> Seems not that obvious
Apr 17 10:42:03 <Ainulindale> Semi BSD
Apr 17 10:44:15 <alphaone> Azog: Yeah, I hang around.
Apr 17 10:44:26 <alphaone> Azog: But better contact me at buzz@totalueberwachung.de
Apr 17 10:44:38 <Azog> alphaone: ok, thanks
Apr 17 10:44:40 <alphaone> np
Apr 17 10:56:08 <CIA-75> freesmartphone.org: 03mickey 07libgsm0710mux * r9a6069992625 10/ (ChangeLog gsm0710mux/channel.vala gsm0710mux/transport.vala): 
Apr 17 10:56:08 <CIA-75> freesmartphone.org: Prohibit reding (hence transferring more input) for a virtual channel, if the modem sets
Apr 17 10:56:08 <CIA-75> freesmartphone.org: the FC bit for this line.
Apr 17 10:57:43 <billk_> Ainulindale: will keep looking
Apr 17 10:57:55 <billk_> Ainulindale: tkx
Apr 17 10:58:11 <von_fritz> Makefile needs to be adjusted for shr-testing/conf/local.conf packages -> recipes
Apr 17 11:13:20 <Ainulindale> von_fritz: already did
Apr 17 11:15:37 <CIA-75> freesmartphone.org: 03mickey 07libgsm0710 * r89e9e7be8ca2 10/src/Makefile.am: fix typo in Makefile.am
Apr 17 11:15:39 <CIA-75> freesmartphone.org: 03mickey 07libgsm0710 * r50079c58e6f4 10/ (AUTHORS MAINTAINERS): 
Apr 17 11:15:39 <CIA-75> freesmartphone.org: Add required MAINTAINERS file. The tarball for this release is now at
Apr 17 11:15:39 <CIA-75> freesmartphone.org: http://www.freesmartphone.org/sources/
Apr 17 11:17:46 <von_fritz> Ainulindale  echo "BBFILES += \"\$${TOPDIR}/shr/openembedded/packages/*/*.bb\"" >> shr-testing/conf/local.conf ; \
Apr 17 11:18:41 <von_fritz> line 172 of Makefile
Apr 17 11:20:52 <Ainulindale> von_fritz: nope
Apr 17 11:21:14 <Ainulindale> Ah, yes, for testing
Apr 17 11:21:21 <Ainulindale> I switched only stable
Apr 17 11:21:23 <Ainulindale> Err
Apr 17 11:21:24 <Ainulindale> Unstable
Apr 17 11:21:28 <Ainulindale> And forgot to switch testing
Apr 17 11:21:29 <Ainulindale> Thanks
Apr 17 11:22:08 <von_fritz> nop :)
Apr 17 11:22:51 <Ainulindale> Fixed.
Apr 17 11:23:02 <CIA-75> SHR: 03ainulindale 07shr-makefile * r8b8cdba8bce2 10/Makefile: Switched testing to recipes too.
Apr 17 11:41:34 <Ainulindale> mickeyl: http://pastebin.archlinux.fr/344279
Apr 17 11:41:42 <Ainulindale> Check line 14
Apr 17 11:41:50 <Ainulindale> I tried to call while there was no carrier
Apr 17 11:41:56 <Ainulindale> it released immediately
Apr 17 11:42:05 <Ainulindale> meanwhile, oeventsd picked up the signal and requested CPU
Apr 17 11:42:14 <Ainulindale> CPU stayed requested
Apr 17 11:43:41 <mickeyl> oh, nasty
Apr 17 11:43:45 <Ainulindale> yep
Apr 17 11:43:52 <Ainulindale> mickeyl: I was able to counter check with ophonekitd
Apr 17 11:43:52 <max_posedon> [SHR-testing, bug] remoko doesn't work
Apr 17 11:43:58 <max_posedon> ImportError: No module named remoko.remoko_server
Apr 17 11:44:03 <mickeyl> can you upload that log to a bug? sounds like a blocker for 5.5
Apr 17 11:44:10 <Ainulindale> mickeyl: It won't receive suspend signal
Apr 17 11:44:17 <Ainulindale> mickeyl: isn't there any bug related to that?
Apr 17 11:44:41 <mickeyl> not that i know offhand
Apr 17 11:47:54 <Ainulindale> Creating it
Apr 17 11:48:41 <Ainulindale> http://trac.freesmartphone.org/ticket/404
Apr 17 11:48:46 <Ainulindale> (heh, 404)
Apr 17 11:50:56 <Ainulindale> mickeyl: and I answered to your mail on t he fsoraw topic
Apr 17 11:51:03 <Ainulindale> I'd be interested in you reading it
Apr 17 11:51:11 <Ainulindale> Because I utterly disagree with you :-)
Apr 17 11:52:04 <Ainulindale> spaetz: hey
Apr 17 11:52:17 <Ainulindale> you totally misunderstood the purpose of ophonekitd.usage
Apr 17 11:53:49 <spaetz> Ainulindale: hey. might be
Apr 17 11:53:55 <spaetz> lunch is ready
Apr 17 11:54:00 <spaetz> I'll be back in 30 min
Apr 17 11:54:08 <spaetz> BTW, bitbake recipe: RREPLACES
Apr 17 11:54:27 <spaetz> is probably what we want for the openmoko-*3 -> shr-* move
Apr 17 11:54:35 <spaetz> away for lunch
Apr 17 11:54:51 <Ainulindale> No
Apr 17 11:54:58 <Ainulindale> We don't need that as we have no reference for that anymore
Apr 17 11:56:30 <Ainulindale> so, mickeyl 
Apr 17 11:56:34 <Ainulindale> your thoughts about that bug?
Apr 17 11:56:48 <mickeyl> dunno
Apr 17 11:56:51 <mickeyl> need to look into it
Apr 17 11:57:07 <Ainulindale> Ah and libeflvala doesn't build
Apr 17 11:57:12 <mickeyl> it does here
Apr 17 11:57:14 <Ainulindale> QA problems with .debug packages
Apr 17 11:57:15 <mickeyl> how does it bail out?
Apr 17 11:57:26 <mickeyl> ah, that should be simple to fix
Apr 17 11:57:33 <mickeyl> let me check in .dev
Apr 17 11:58:47 <Ainulindale>  ERROR: QA Issue: non debug package contains .debug directory: libeflvala-examples path /work/armv4t-angstrom-linux-gnueabi/libeflvala-0.0.0.0+gitr43+676c901eb57f653231fb4c9eeabc2ed3d1b0c072-r0/install/libeflvala-examples/usr/bin/.debug/elementary-vapi
Apr 17 11:59:09 <Ainulindale> Yeah but I didn't look into that knowing it's not mine :-)
Apr 17 11:59:10 <Ainulindale> :w 4
Apr 17 12:00:27 <DocScrutinizer> mickeyl: how could I check for a (possible) bug in newest Device/audio/alsa-scenario.vala ? Donno where to look in git
Apr 17 12:00:27 <mickeyl> Ainulindale: i don't think i want to pursue the architecture discussion further via email. this is something that's better discussed life
Apr 17 12:00:56 <Ainulindale> mickeyl: no problem with me
Apr 17 12:01:04 <Ainulindale> But I hope this use case looks pertinent enough to you :-)
Apr 17 12:01:12 <mickeyl> DocScrutinizer: you mean ./openmoko-gsoc2008/fsod/src/subsystems/Device/audio/alsa-scenario.vala ?
Apr 17 12:01:17 <mickeyl> this is Sup3rkiddo's work
Apr 17 12:01:22 <DocScrutinizer> yup
Apr 17 12:01:26 <DocScrutinizer> line 150
Apr 17 12:01:39 <DocScrutinizer> alsacrrtl
Apr 17 12:01:55 <DocScrutinizer> s/rr//
Apr 17 12:01:55 <bzzbot> DocScrutinizer meant: alsactl
Apr 17 12:02:18 <mickeyl> good find
Apr 17 12:03:00 <cyberalex> hi
Apr 17 12:03:12 <CIA-75> freesmartphone.org: 03mickey 07openmoko-gsoc2008 * rbe461a8c713c 10/fsod/src/subsystems/Device/audio/alsa-scenario.vala: fix alsactl invocation bug. Spotted by DocScrutinizer -- thanks!
Apr 17 12:03:22 <cyberalex> how to add elementary support in openmoko toolchain?
Apr 17 12:22:48 <Ainulindale> mickeyl: tell me when it's ok for libeflvala recipe please :-)
Apr 17 12:23:00 <Ainulindale> mickeyl: and by the way
Apr 17 12:23:22 <Ainulindale> Couldn't we use Superkiddo's work with fsod and current fso python subsystems?
Apr 17 12:26:06 <spaetz> back
Apr 17 12:26:49 <spaetz> how did Iget ophonekitd.usage wong? Its one advantage is that it survives dbus connection aborts. At least this is how mrmoku explained it to me IIRC
Apr 17 12:27:22 <mickeyl> i don't know how current fsod is wrt. API specs, but feel free to have a go. it doesn't support audio playing though and suffers from the same Vala showstoppers as cornucopia atm, namely dbus error marshalling and async.
Apr 17 12:28:49 <Ainulindale> spaetz: well mrmoku was wrong
Apr 17 12:29:06 <Ainulindale> spaetz: this is merely a side effect of what its use is
Apr 17 12:29:19 <Ainulindale> Its use is to allow the developer to request or release resource on ophonekitd behalf
Apr 17 12:29:30 <Ainulindale> That is, being able to release a resource in ophonekitd and not on a global system level
Apr 17 12:29:38 <Ainulindale> The example in my mail is pretty clear
Apr 17 12:29:44 <Ainulindale> (thread fsoraw)
Apr 17 12:30:04 <spaetz> like not wanting to use resourcepolicy because someone else might want to use GSM
Apr 17 12:30:04 <Ainulindale> mickeyl: I thought fsod was able to use framework python susbsytems
Apr 17 12:30:08 <spaetz> yeah, read that
Apr 17 12:30:13 <Ainulindale> spaetz: So?
Apr 17 12:30:24 <mickeyl> Ainulindale: i don't know the status of that
Apr 17 12:30:39 <spaetz> still, if tangogps wants to prevent suspend, it should use the FSO RequestResource CPU call and not ophonekitd
Apr 17 12:31:00 <spaetz> there is no advantage of going through ophonekitd in this case
Apr 17 12:33:01 <Ainulindale> mickeyl: IIRC I helped him with that
Apr 17 12:33:06 <Ainulindale> spaetz: That, of course
Apr 17 12:33:27 <Ainulindale> spaetz: But ophonekitd dbus APIs have NEVER been designed to do that, let us be clear :-)
Apr 17 12:33:37 <Ainulindale> They are mainly there to handle resources ophonekitd needs
Apr 17 12:33:49 <Ainulindale> No need of an additonal layer for other software no interacting with ophonekitd
Apr 17 12:34:08 <Ainulindale> If this API didn't exist, there would be no way for settings to release GSM for ophonekitd
Apr 17 12:34:16 <Ainulindale> Settings would have to release GSM system wide
Apr 17 12:36:11 <spaetz> agreed
Apr 17 12:36:18 <spaetz> that I can understand
Apr 17 12:36:53 <spaetz> but afaik the whole wrapper discussion was about how to use mdbus calls to prevent suspend
Apr 17 12:37:06 <spaetz> so going more towards the example of tangogps
Apr 17 12:37:18 <spaetz> I agree with your example however
Apr 17 12:37:37 <spaetz> that it's a good way to make ophonekitd request/release the GSM resource
Apr 17 12:37:48 <Ainulindale> Well my point is that this solution is a workaround
Apr 17 12:37:53 <Ainulindale> Because it shouldn't be done that way
Apr 17 12:37:57 <Ainulindale> frameworkd should offer some way to do that
Apr 17 12:38:04 <Ainulindale> I had several propositions towards that goal
Apr 17 12:38:24 <Ainulindale> 1) dbus handshake to confirm that process X has the rights to release resource Y in its stead
Apr 17 12:38:27 <Ainulindale> Or
Apr 17 12:38:34 <Ainulindale> 2) conf files with priorities about owners
Apr 17 12:41:42 <spaetz> both sounds reasonable
Apr 17 12:41:56 <spaetz> hey, why do we start the ssh server by default on the FR?
Apr 17 12:44:54 <Ainulindale> because you want to be able to use usb 
Apr 17 12:47:09 <pundiramit> i'm trying to setup an openmoko build environment on my machine for last 6 hrs but with no luck..
Apr 17 12:47:16 <pundiramit> nothing seems to be working for me
Apr 17 12:47:54 <pundiramit> instructions @ http://wiki.openmoko.org/wiki/OpenEmbedded are not much of help
Apr 17 12:48:31 <pundiramit> can anyone point me to right build instructions i.e how to setup openembedded and openmoko build env?
Apr 17 12:48:36 <pundiramit> please
Apr 17 12:59:33 <spaetz> I use the SHR makefile and that worked fine. pundiramit 
Apr 17 12:59:42 <spaetz> no clue about openmoko setup
Apr 17 13:04:04 <pundiramit> i assume its not a recommended one but can SHR be used as a stable system?
Apr 17 13:04:44 <pundiramit> plus development platform for a gtk based app
Apr 17 13:05:57 <Ainulindale> nytowl?
Apr 17 13:06:00 <Ainulindale> :W 4
Apr 17 13:06:02 <Ainulindale> grmbl
Apr 17 13:11:32 <spaetz> I use it as my daily phone
Apr 17 13:12:02 <spaetz> darn bluez stuff, how do I get the FR to search for BT headsets?
Apr 17 13:12:26 <spaetz> mdbus -s org.bluez /org/bluez/1383/hci0 org.bluez.Adapter.ListDevices always returns an empty array
Apr 17 13:12:52 <spaetz> and GetProperties says 'Discovering': False,
Apr 17 13:25:33 <max_posedon> [SHR-testing, bug] startscreen doesn't displaying tangogps, gpe-scetchbook, htop
Apr 17 13:25:41 <von_fritz> spaetz hcitool scan
Apr 17 13:25:54 <max_posedon> (after installing doesn't display gpe-calendar also)
Apr 17 13:32:36 <PaulFertser> Azog: Hey :)
Apr 17 13:32:47 <PaulFertser> Azog: yes, xrandr + proper echo to jbt driver.
Apr 17 13:36:35 <spaetz> k, hcitool scan shows the headscan. progress
Apr 17 13:37:00 <spaetz> gst-launch -v alsasrc device=bluetooth ! audioconvert ! audioresample ! alsasink device=bluetooth sync=false
Apr 17 13:37:08 <spaetz> blarg, mispasted, sorry
Apr 17 13:39:04 <CIA-75> SHR: 03mwester 07shr-overlay * rf66889fcc95e 10/patches/fastboot/ (3 files): SHR patches: fastboot - updated netbase, opkg, and fstab patches so they
Apr 17 13:39:04 <CIA-75> SHR: 03mwester 07shr-overlay * r1a5f07f09ae4 10/openembedded/recipes/images/shr-image.inc: shr-image.inc - fix Exec line for htop
Apr 17 13:39:06 <CIA-75> SHR: 03mwester 07shr-overlay * rbda200d9cf04 10/openembedded/recipes/shr/frameworkd-config-shr_git.bb: Merge branch 'master' of git@shr.bearstech.com:shr-overlay
Apr 17 13:39:06 <CIA-75> SHR: 03mwester 07shr-overlay * r459f6875645d 10/openembedded/recipes/shr/ (initscripts-shr/g_ether.sh initscripts-shr_git.bb): initscript-shr: g_ether - suppress stderr output; avoid errors being printed
Apr 17 13:39:09 <CIA-75> SHR: 03mwester 07shr-overlay * r234e465cd9b0 10/patches/kernel/0001-kernel-remove-patches-and-adjust-config-name.patch: shr-autorev: nail-down the gta02 kernel; also update kernel patch to
Apr 17 13:39:15 <CIA-75> SHR: 03ainulindale 07shr-overlay * rb098cbc4f670 10/openembedded/recipes/shr/frameworkd-config-shr_git.bb: frameworkd-config-shr: Eventually fixed the directory issue. I withdraw my comments on dos1, even though he's mean
Apr 17 13:39:18 <CIA-75> SHR: 03ainulindale 07shr-overlay * r7b3dd4cb2b51 10/openembedded/recipes/shr/frameworkd-config-shr_git.bb: frameworkd-config-shr: Fixed southern redneck shouting SRCREV about racoon and stuff.
Apr 17 13:39:21 <CIA-75> SHR: 03ainulindale 07shr-overlay * r22fb8b31f0ce 10/openembedded/recipes/shr/frameworkd-config-shr_git.bb: frameworkd-config-shr: Fix -f by -d. This is dos1 fault for not warning
Apr 17 13:39:26 <CIA-75> SHR: 03mwester 07shr-overlay * rfb3de279fc18 10/openembedded/conf/distro/include/shr-autorev.inc: shr-autorev.inc: Fix the gta02 kernel to a specific git hash, add
Apr 17 13:40:10 <Ainulindale> mwester: you forgot to push to testing
Apr 17 13:42:15 <CIA-75> freesmartphone.org: 03mickey 07libgsm0710mux * r114044266a09 10/ (ChangeLog MAINTAINERS Makefile.am configure.ac): release as 0.3.3
Apr 17 13:43:44 <billk_> spaetz: I use mdbus -s org.shr.ophonekitd.Usage /org/shr/ophonekitd/Usage org.shr.ophonekitd.Usage.RequestResource 'Bluetooth'
Apr 17 13:47:25 <spaetz> billk_: ahh, ok
Apr 17 13:48:13 <billk_> Then stop/start /init.d/bluetooth - if paired headest works fine (a2dp" profile
Apr 17 13:48:51 <CIA-75> freesmartphone.org: 03mickey 07fso-abyss * r0c713f42f078 10/ (AUTHORS ChangeLog MAINTAINERS Makefile.am configure.ac): release as 0.3.3
Apr 17 13:52:03 <Weiss> is the new "Testing" safe to upgrade to from the old testing at the moment? - is opkg upgrade sensible, or is a reflash required?
Apr 17 13:56:14 <Ainulindale> Weiss: nope
Apr 17 13:56:15 <Ainulindale> reflash
Apr 17 13:57:15 <Weiss> ok. but if i grab the latest shr-lite-image-om-gta02.jffs2 right now, it won't be in some odd not-quite-merged-yet status?
Apr 17 13:58:59 <Weiss> (i.e. the current images are expected to vaguely work at the moment?)
Apr 17 14:02:11 <mwester> Ainulindale:  :p  I didn't forget to push to testing -- I didn't test!
Apr 17 14:02:22 * mwester did a "drive-by commit" last night
Apr 17 14:03:37 <mwester> Ainulindale:  we can now begin to use "_shr" overrides and place shr-specific versions of files in subdirs named "shr".
Apr 17 14:04:07 <mwester> But the problem remains: how do we get an SHR change to appear on the milestone5.5 branch in OE, once you merge into OE?
Apr 17 14:08:14 <PaulFertser> spaetz: does your headset support A2Dp
Apr 17 14:15:57 <PaulFertser> spaetz: because if it doesn't, there's no sense in trying to play audio through it (that is, SCO over USB won't work without special tweaks).
Apr 17 14:19:49 <mwester> Weiss:  the current images work rather well, actually -- flash one and try it!
Apr 17 14:21:57 <Weiss> mwester: trying now. someone seems to have stolen the kernel images from http://build.shr-project.org/shr-testing/images/om-gta02/ though (i guess the autobuilder is in progress or something?)
Apr 17 14:22:20 <Weiss> or maybe i should try shr-unstable, if they're temporarily equivalent?
Apr 17 14:23:11 <CM_work> mwester: Any chance of that kernel working on a GTA01? :)
Apr 17 14:23:42 <mwester> Weiss: I think you would  prefer the shr-unstable images, assuming they work -- and nobody has flogged me since I commited the updates 9 hours ago so I suspect they are good.
Apr 17 14:24:04 <mwester> CM_work:  the 2.6.24 kernels for the GTA01 work reasonably well on the gta01 right now.
Apr 17 14:24:13 <mwester> The 2.6.29 kernels do not.
Apr 17 14:57:32 <spaetz> PaulFertser: 2.0 +EDR
Apr 17 14:57:32 <spaetz> HFP
Apr 17 14:57:33 <spaetz> HSP
Apr 17 15:01:17 <spaetz> PaulFertser: Sony Ericsson HBH-PV703 
Apr 17 15:04:37 <spaetz> googling doesn't seem to mention A2Dp for it.
Apr 17 15:28:53 <Ainulind1le> mwester: nah, everyone should fetch testing
Apr 17 15:28:57 <Ainulind1le> I didn't update unstable for three days
Apr 17 15:50:59 <Ainulind1le> Well anyway, updating unstable
Apr 17 15:51:02 <Ainulind1le> mwester: by the way
Apr 17 15:51:05 <Ainulind1le> I had a weird issue
Apr 17 15:51:09 <Ainulind1le> Wondering if you could be the culprit
Apr 17 15:51:18 <Ainulind1le> When the screen is dimmed
Apr 17 15:51:22 <Ainulind1le> usb drops
Apr 17 15:54:25 <PaulFertser> spaetz: so don't try to playback any sound through it as well as record.
Apr 17 15:54:43 <PaulFertser> spaetz: try it for gsm calls like described on the wiki (the latest method, requires new fso)
Apr 17 16:00:00 <spaetz> PaulFertser: simple-agent craps out for me with:Creating device failed: org.bluez.Error.AuthenticationFailed: Authentication Failed
Apr 17 16:00:14 <spaetz> I don't get the opportunity to enter the pin 0000
Apr 17 16:00:35 <Ainulind1le> spaetz: you should ask Okhin 
Apr 17 16:00:40 <Ainulind1le> He's working on a bluez thingy
Apr 17 16:00:48 <spaetz> ok, thanks
Apr 17 16:01:50 <spaetz> hcitool scan wouldn't show anything anymore (I actually suspect the darn headset paired with my laptop although I turned bluetooth off there)
Apr 17 16:03:17 <spaetz> mmh, need to turn off laptop and restart the whole sheband
Apr 17 16:03:21 <spaetz> shebang
Apr 17 16:03:26 <spaetz> bbiab
Apr 17 16:08:35 <Gromgull> Hi SHR people - if I download hte testing image NOW, is it the old unstable? And have hte opkg-repositories been switched?
Apr 17 16:08:49 <Ainulind1le> yes
Apr 17 16:09:52 <Weiss> however, there still seems to be no kernel image in http://build.shr-project.org/shr-testing/images/om-gta02/
Apr 17 16:10:40 <Ainulind1le> Hmmm that's weird it was there yesterday
Apr 17 16:10:50 <tilman> it was there 4 hours ago :)
Apr 17 16:11:26 <Ainulind1le> Rebuilding it then
Apr 17 16:11:44 <Ainulind1le> but still, weird
Apr 17 16:11:53 <Ainulind1le> I bet that's mwester's or dos1's fault
Apr 17 16:11:58 <Ainulind1le> Or even TAsn's
Apr 17 16:12:05 <dos1> Or Dave
Apr 17 16:12:06 <dos1> 's
Apr 17 16:12:15 <Ainulind1le> Now that you say that
Apr 17 16:12:18 <TAsn> lol ;[
Apr 17 16:12:23 <Ainulind1le> Yeah I think it was Dave's fault
Apr 17 16:12:26 <TAsn> it's dave's fault.
Apr 17 16:12:36 <dos1> Of course it's Dave's fault
Apr 17 16:12:39 <dos1> ;D
Apr 17 16:12:46 <TAsn> dave is the destroyer of all mokos
Apr 17 16:12:51 <Weiss> the uImage that was there a bit before was only about 342 bytes long
Apr 17 16:13:01 <Ainulind1le> well Weiss I downloaded it yesterday
Apr 17 16:13:05 <Ainulind1le> So it's really weird
Apr 17 16:13:13 <TAsn> cya
Apr 17 16:13:22 <TAsn> I'm going back to study.
Apr 17 16:16:31 <Zorkman> is there a big difference between the current testing and the unstable one?
Apr 17 16:16:47 <dos1> Zorkman: i don't think there is any difference ;)
Apr 17 16:17:34 <Ainulind1le> Zorkman: they are equal
Apr 17 16:17:43 <Ainulind1le> except unstable isn't up to date
Apr 17 16:17:49 <Zorkman> will there be updates for the testing, or will it be frozen?
Apr 17 16:18:00 <Ainulind1le> There will be bug updates, but frozen in terms of development
Apr 17 16:18:10 <Ainulind1le> But we'll break unstable I think
Apr 17 16:18:13 <Zorkman> and dos1: are you planning to integrate a force fast charge option in shr settings?
Apr 17 16:18:23 <dos1> Zorkman: yes, i'm planning
Apr 17 16:18:29 <Zorkman> perfectly
Apr 17 16:18:40 <CIA-75> SHR: 03mwester 07shr-overlay * ra34da99b8c27 10/openembedded/recipes/images/shr-image.inc: shr-image.inc: Fix (temporary; real fix is upstream) to make udev stop complaining
Apr 17 16:18:48 <mwester> There.  Now they are no longer equal.
Apr 17 16:18:52 <Zorkman> but i guess that will not be upgradable in shr testing?
Apr 17 16:18:59 <Zorkman> hehe
Apr 17 16:19:28 <mwester> opkg upgrade != "a reasonable thing to do with SHR at present, either stable or testing"
Apr 17 16:19:46 <Ainulind1le> Nope, too many name changes.
Apr 17 16:19:51 <Ainulind1le> (for OE integration)
Apr 17 16:20:05 <Ainulind1le> Rebuilding the kernel for testing guys
Apr 17 16:21:33 * mwester says a bad word.
Apr 17 16:21:42 <lindi-> Ainulind1le: shouldn't name changes be changed?
Apr 17 16:21:47 <lindi-> Ainulind1le: shouldn't name changes be handled?
Apr 17 16:21:51 <lindi-> by opkg?
Apr 17 16:22:00 <Ainulind1le> What do you mean?
Apr 17 16:32:23 <PaulFertser> spaetz: Yes, you need to set your headset to the pairing mode.
Apr 17 16:32:46 <PaulFertser> spaetz: Read you headset instructions.
Apr 17 16:37:12 <Gromgull> the one-liner for updating opkg feed files on the shr-blog is broken...
Apr 17 16:37:14 <Gromgull> sed -e 's/unstable/testing/g' /etc/opkg/*
Apr 17 16:37:17 <Gromgull> will just print the new files
Apr 17 16:37:22 <Ainulind1le> sed -i
Apr 17 16:37:27 <Ainulind1le> Did I forgot the -i?
Apr 17 16:37:33 <Gromgull> yes :) 
Apr 17 16:52:04 <mwester> Ainulindale:  ERROR: Error in executing: /r/shr2/shr-unstable/openembedded/recipes/freesmartphone/frameworkd_git.bb
Apr 17 16:52:08 <mwester> On gta01 builds
Apr 17 16:52:21 <mwester> Applying patch otimed.patch
Apr 17 16:56:18 <Ainulind1le> weird
Apr 17 16:56:18 <Ainulind1le> here it went well
Apr 17 16:57:49 <mwester> diff openembedded/recipes/freesmartphone/frameworkd/om-gta0*/otimed.patch
Apr 17 16:58:23 <Ainulind1le> no diff
Apr 17 16:58:42 <mwester> make update-shr-unstable
Apr 17 16:59:57 <Ainulind1le> Sir yes sir
Apr 17 17:00:29 <mwester> so when will the ISP get your internet connection at home?  That must be painfully slow...
Apr 17 17:01:04 <Ainulind1le> 50k/s
Apr 17 17:01:18 <mwester> ouch.
Apr 17 17:01:26 <Ainulind1le> yeah...
Apr 17 17:02:26 <dos1> Ainulind1le: i had 60KB/s for years
Apr 17 17:02:35 <Ainulind1le> mwester: still no diff
Apr 17 17:04:12 <dos1> Ainulind1le: now i have 120KB/s
Apr 17 17:04:34 <Ainulind1le> well I was used to have 1,5M down, 200K up
Apr 17 17:04:39 <Ainulind1le> so you see the change
Apr 17 17:04:50 <Ainulind1le> (bytes)
Apr 17 17:05:23 <mwester> http://cgit.openembedded.net/cgit.cgi?url=openembedded/tree/recipes/freesmartphone/frameworkd/om-gta01/otimed.patch&h=fso/milestone5.5
Apr 17 17:05:23 <mwester> Note the path in the patch - it's very "not correct".
Apr 17 17:05:23 <mwester> That's the version I'm getting pulled down in my env.
Apr 17 17:05:23 <mwester> Here's the commit:
Apr 17 17:05:23 <mwester> http://cgit.openembedded.net/cgit.cgi?url=openembedded/commit/&id=e4f800531b524dc4d40f0c07aa8e111acdc157f7
Apr 17 17:06:14 <dos1> Ainulind1le: ~2 years ago i had 32KB/s :P
Apr 17 17:06:37 <dos1> for long time
Apr 17 17:06:38 <dos1> and before that 16KB/s
Apr 17 17:07:01 <dos1> those all on ADSL in Poland
Apr 17 17:07:14 <Ainulind1le> mwester: so?
Apr 17 17:07:16 <Ainulind1le> dos1: meh =)
Apr 17 17:07:46 <dos1> 640KB/s is the fastest option here in Polish Telecom for individual customest...
Apr 17 17:07:51 <mwester> Ainulind1le:  er... yeah... well... I just thought you might like to know why the gta01 builds are failing on for everyone.  That's all.
Apr 17 17:07:54 <dos1> s/customert/customers/
Apr 17 17:08:04 <Ainulind1le> mwester: well here it didn't! :-)
Apr 17 17:08:11 <Ainulind1le> mwester: That's why I don't get you
Apr 17 17:08:31 <Ainulind1le> I'll retry later now I have stuff to do :-)
Apr 17 17:08:48 <dos1> and now I have 120KB/s for only one month :P
Apr 17 17:08:53 <dos1> s/have/have had/
Apr 17 17:08:54 <bzzbot> dos1 meant: and now I have had 120KB/s for only one month :P
Apr 17 17:08:54 <mwester> Ok, well it did on the buildhost. but no big deal, we'll wait until somebody important complains! :p
Apr 17 17:09:13 <Ainulind1le> which buildhost?
Apr 17 17:09:19 <mwester> bearstech
Apr 17 17:09:22 <Ainulind1le> ?
Apr 17 17:09:25 <Ainulind1le> you're using it?
Apr 17 17:09:28 <mwester> That's why there was not kernel - the build failed.
Apr 17 17:09:34 <Ainulind1le> Errr
Apr 17 17:09:37 <Ainulind1le> I rebuilt the kernel
Apr 17 17:09:48 <Ainulind1le> And it was for GTA02
Apr 17 17:09:57 <Ainulind1le> Do you have access to the buildhost?
Apr 17 17:10:07 <mwester> no, and don't want it either!
Apr 17 17:10:16 <mwester> I just want to build locally.
Apr 17 17:10:19 <mwester> And I can't.
Apr 17 17:10:41 <mwester> Because, apparently, my ISP is changing patches and commits on the fly just for me, that nobody else can see.
Apr 17 17:11:08 <Ainulind1le> well let me try to rebuild the framework ok?
Apr 17 17:11:14 <mwester> No
Apr 17 17:11:31 <mwester> Much less effort to just diff the two patches.
Apr 17 17:11:39 <mwester> The gta01 and gta02 versions should be identical.
Apr 17 17:11:50 <Ainulind1le> well they are
Apr 17 17:11:59 <Ainulind1le> hence my "no diff"
Apr 17 17:12:02 <mwester> I just showed you the commit where they are not.
Apr 17 17:12:24 <dos1> why there are all of locales installing on opkg upgrade?
Apr 17 17:12:45 <Ainulind1le> Huh?
Apr 17 17:12:52 <Ainulind1le> They shouldn't
Apr 17 17:13:05 <dos1> but they are
Apr 17 17:13:06 <dos1> ;D
Apr 17 17:13:13 <Ainulind1le> Well listen I don't have time for that I truly have to leave
Apr 17 17:13:22 <Ainulind1le> so please check that for me will you? :-)
Apr 17 17:13:58 <dos1> if only i can find it
Apr 17 17:13:59 <dos1> ;)
Apr 17 17:15:01 <admiral0> hi
Apr 17 17:31:53 <bumbl> hi mirko-paroli 
Apr 17 17:33:04 <mirko-paroli> hi bumbl
Apr 17 17:33:11 <mirko-paroli> how's it happening?
Apr 17 17:57:56 <bumbl> dos1: ping
Apr 17 17:58:25 <dos1> bumbl: pong
Apr 17 18:00:06 <bumbl> dos1: could it be that the accels are off by default?
Apr 17 18:00:20 <dos1> bumbl: i don't think so
Apr 17 18:00:26 <dos1> why are you thinking that?
Apr 17 18:00:35 <dos1> err
Apr 17 18:00:40 <dos1> they are off by default
Apr 17 18:00:54 <bumbl> dos1: because moko eightball does not work
Apr 17 18:00:57 <dos1> but if only you want to read from those
Apr 17 18:01:02 <bumbl> how to enable?
Apr 17 18:01:06 <dos1> they are turned on automatically
Apr 17 18:01:22 <dos1> bumbl: change moko eightball to use new accels api :P
Apr 17 18:01:32 <bumbl> ah so it changed?
Apr 17 18:01:39 <dos1> yep
Apr 17 18:01:44 <dos1> to better
Apr 17 18:02:37 <bumbl> ok
Apr 17 18:02:41 <bumbl> i will change it
Apr 17 18:02:47 <bumbl> as moko eightball is a must have
Apr 17 18:03:24 <bumbl> anyway get to go
Apr 17 18:11:02 * PaulFertser wonders if his mail to smartphone-userland with FSO patch to use the new usb_mode path was lost somewhere
Apr 17 18:11:52 <dos1> PaulFertser: can't you commit it?
Apr 17 18:12:10 <PaulFertser> dos1: sure, i'm no dev, have no commit access anywhere :)
Apr 17 18:12:55 <dos1> PaulFerster: i have commit access to FSO
Apr 17 18:13:17 <dos1> hmm
Apr 17 18:14:04 <dos1> i can't see why patch like that shouldn't be commited directly
Apr 17 18:14:22 <dos1> maybe someone will spank me now
Apr 17 18:14:41 <dos1> ;)
Apr 17 18:16:30 <dos1> (i've just got FSO commit access from mickey|lunch, cause he said that if i have access to SHR i should have to FSO too; so i'm not FSO guy of course, and my opinion isn't very valuable ;) )
Apr 17 18:18:08 <PaulFertser> dos1: you might want to try it on your device first ;)
Apr 17 18:18:32 <dos1> PaulFertser: can you send it to me?
Apr 17 18:18:52 <dos1> i must register to freesmartphone maillist ;x
Apr 17 18:24:24 <PaulFertser> dos1: http://projects.linuxtogo.org/pipermail/smartphones-userland/2009-April/001430.html
Apr 17 18:26:44 <dos1> PaulFertser: i really don't know, what can be said to prove that it shoudn't be commited
Apr 17 18:26:44 <dos1> so
Apr 17 18:27:17 <dos1> i will take a risk and commit it
Apr 17 18:27:24 <dos1> (after test ofc)
Apr 17 18:32:00 <TAsn> dos1, I need eve!
Apr 17 18:32:06 <TAsn> man, played with minimo today
Apr 17 18:32:09 <TAsn> so slow!
Apr 17 19:44:23 <lindi-> PaulFertser: I managed to reproduce "calypso runs out of memory" bug and captured its error output with AVR :)
Apr 17 19:45:49 <PaulFertser> lindi-: nice :)
Apr 17 19:47:40 <PaulFertser> lindi-: though i'm afraid of little help. Nice thing is that it doesn't seem to be reproducible without multiplexing and there's a hope that fso-abyss won't trigger that.
Apr 17 19:49:02 <lindi-> PaulFertser: just made a voip (SIP) call with 3g usb stick and the sound quality is quite nice. a finnish operator offers free calls to their mobile phones with that service so it'll make testing things very easy for me :)
Apr 17 19:49:15 <CIA-75> freesmartphone.org: 03seba.dos1 07framework * r280074afeaa1 10/framework/subsystems/odeviced/powercontrol_neo.py: 
Apr 17 19:49:15 <CIA-75> freesmartphone.org: odeviced: fix USB Host mode with latest kernels. Patch by Paul Fertser.
Apr 17 19:49:15 <CIA-75> freesmartphone.org: Signed-off-by: Paul Fertser <fercerpav at gmail.com>
Apr 17 19:50:55 <lindi-> PaulFertser: does the above mean that odeviced won't work with older kernels ^
Apr 17 19:50:58 <PaulFertser> lindi-: what sip client are you using on freerunner?
Apr 17 19:51:04 <PaulFertser> lindi-: no, it doesn't
Apr 17 19:51:05 <lindi-> PaulFertser: linphone
Apr 17 19:51:15 <PaulFertser> lindi-: which version?
Apr 17 19:51:33 <CIA-75> SHR: 03seba.dos1 07shr-settings * r709fc633f123 10/shr_settings_modules/shr_clock.py: [clock] use fso api instead of calling directly date. Patch by Michael Scherer (misc)
Apr 17 19:51:38 <lindi-> PaulFertser: 3.0.0-3 from debian unstable
Apr 17 19:51:41 <PaulFertser> lindi-: how do you cope with echo?
Apr 17 19:51:52 <lindi-> PaulFertser: usb headset
Apr 17 19:52:00 <dos1> lindi-: if it will don't work on older kernels i wouldn't commit that. and FSO guys would be very angry on me then :P
Apr 17 19:52:08 <PaulFertser> lindi-: Cheater :)
Apr 17 19:52:36 <lindi-> PaulFertser: linphone 2.1.1-1+b1 on my debian stable laptop made quite machine-sounding noise though so there has been improvement
Apr 17 19:52:39 <dos1> PaulFertser: usbmode in shr-settings works with your patch, thanks :)
Apr 17 19:52:59 <lindi-> dos1: did you test that it worked with older ones? ;)
Apr 17 19:53:16 <dos1> lindi-: it can't don't work on older ones, if it worked before
Apr 17 19:53:18 <dos1> look at code
Apr 17 19:53:19 <dos1> :P
Apr 17 19:53:34 <PaulFertser> dos1: thank you for giving me a hand here :)
Apr 17 19:53:56 <dos1> (what does "giving a hand" mean? :D)
Apr 17 19:54:47 <PaulFertser> dos1: http://idioms.thefreedictionary.com/give+a+hand
Apr 17 19:55:10 <dos1> PaulFertser: thanks. And no problem :)
Apr 17 19:55:22 <lindi-> PaulFertser: i need to figure out how I can attach that 3g stick more permanently to the phone
Apr 17 19:55:35 <lindi-> PaulFertser: the usb connector sticks out in a very irrating way :(
Apr 17 19:55:54 <lindi-> are there L shaped usb cables that would not stick out that much?
Apr 17 19:55:54 <PaulFertser> lindi-: there're testpoints inside ;)
Apr 17 19:57:04 <lindi-> PaulFertser: i'm not sure if I want to solder anything ;)
Apr 17 19:57:21 <lindi-> (at least anything inside the case where I could break something)
Apr 17 19:57:35 <PaulFertser> lindi-: i know. But i see no other way except using some cool angle mini-usb connector.
Apr 17 19:58:24 <lindi-> such a connector would solve it nicely
Apr 17 19:59:21 <DocScrutinizer> there are L-shaped ones
Apr 17 20:02:54 <ingoa> hello, i have a question to the new testing image
Apr 17 20:03:39 <mirko-paroli> ingoa: shr?
Apr 17 20:03:43 <ingoa> i can't connect to the phone whit "ifconfig usb0..."
Apr 17 20:04:06 <tilman> did you reboot once after flashing the image?
Apr 17 20:04:12 <ingoa> mickey|lunch: yes
Apr 17 20:04:33 <ingoa> tilman: yes, two times ;)
Apr 17 20:04:58 <PaulFertser> ingoa: depmod -a ?
Apr 17 20:05:26 <DocScrutinizer> lindi-: you checked wiki/specialized_usb_cables ?
Apr 17 20:07:31 <lindi-> mickey|lunch: should fso-abyss work on debian's valac or are patches still needed?
Apr 17 20:07:47 <lindi-> DocScrutinizer: a while ago but has there been updates?
Apr 17 20:07:56 <lindi-> s/has/have/
Apr 17 20:07:56 <bzzbot> lindi- meant: DocScrutinizer: a while ago but have there been updates?
Apr 17 20:08:26 <ingoa> PaulFertser: depmod give my nothing back. 
Apr 17 20:08:33 <PaulFertser> ingoa: it shouldn.
Apr 17 20:08:37 <DocScrutinizer> yup at end of page there's alink to a L-cable 5pole
Apr 17 20:08:58 <PaulFertser> ingoa: it should just make everything needed to allow module loading.
Apr 17 20:09:03 <DocScrutinizer> lindi-: someone in US
Apr 17 20:12:49 <dos1> ok, have to go now
Apr 17 20:12:53 <dos1> good night
Apr 17 20:13:10 <PaulFertser> dos1: night :)
Apr 17 20:18:49 <DocScrutinizer> lindi-: https://www.argentdata.com/catalog/product_info.php?products_id=111
Apr 17 20:20:12 <PaulFertser> DocScrutinizer: Shielded?
Apr 17 20:21:05 <DocScrutinizer> PaulFertser: seems no
Apr 17 20:21:22 <PaulFertser> Anyway, the best cable out there. Angled jack and 5 wires make an excellent opportunity to build nice Y-cable.
Apr 17 20:21:53 <DocScrutinizer> also with 90° L-jacks, there's always 4 ways you could build them (if the plug itself isn't rotaion symmetric)
Apr 17 20:22:38 <DocScrutinizer> this one will point downwards so you can't place FR on a desk anymore
Apr 17 20:23:11 <DocScrutinizer> (except if you drill a hole in the table :-D )
Apr 17 20:23:11 <PaulFertser> DocScrutinizer: 4 ways? Free your mind from orthodoxy!
Apr 17 20:23:40 <DocScrutinizer> yeah I know there are virtually infinite number of angles
Apr 17 20:23:52 <PaulFertser> :)
Apr 17 20:24:19 <mwester> No, the argent data cable is not shielded.  The plastic insulation is very thick, though -- will that do for you?   ;-)
Apr 17 20:24:26 <TAsn> http://wiki.openmoko.org/wiki/Kernel-developers-are-not-polite
Apr 17 20:24:30 <TAsn> lol?
Apr 17 20:24:59 <max_posedon> truth!)
Apr 17 20:25:01 <mwester> I would try this one (I use this on my N800, it works very nicely!)  http://www.electronicproductonline.com/catalog/product_info.php?products_id=2043
Apr 17 20:25:51 <mwester> (somebody should put that on the wiki if it isn't already there)
Apr 17 20:25:57 <PaulFertser> Lacks the ID pin.
Apr 17 20:26:33 <mwester> Clarify
Apr 17 20:26:49 <DocScrutinizer> break_alsa_settings() ? WTF, this is planned for MS6.0
Apr 17 20:27:15 <DocScrutinizer> PaulFertser: mwester: 4pole ?
Apr 17 20:27:16 <PaulFertser> mwester: when you build a Y-cable, it makes sense to connect ID pin of mini-usb jack to the mini-usb receptable you're going to plug a wallcharger in to force 1A charge mode.
Apr 17 20:27:27 <mwester> Why not?  The kernel folks have been implementing new versions of break_kernel_apis() every few weeks?
Apr 17 20:27:32 <DocScrutinizer> NOOOPE!
Apr 17 20:27:57 <DocScrutinizer> PaulFertser: it makes sense to solder a 47k there
Apr 17 20:27:57 <mwester> PaulFertser:  so you want a cable, then.
Apr 17 20:28:01 <TAsn> mwester, You are the angriest person I have ever seen! ;]
Apr 17 20:28:18 <PaulFertser> DocScrutinizer: what if i want to plug it to a PC? ;)
Apr 17 20:28:31 <DocScrutinizer> PaulFertser: so you can use arbitrary usb-powersupplies, not only the OM-wallcharger
Apr 17 20:28:38 <mwester> The N800 adaptor is for USB devices, not external power!
Apr 17 20:28:39 <PaulFertser> DocScrutinizer: :)
Apr 17 20:28:55 <DocScrutinizer> PaulFertser: so what?
Apr 17 20:28:56 <PaulFertser> mwester: and we're talking about a perfect cable, that includes building Y-cable ;)
Apr 17 20:29:13 <PaulFertser> DocScrutinizer: mwester's adapter is very nice, but not that suitable for a Y-cable.
Apr 17 20:29:21 <mwester> TAsn:  why do you say that?  Because I have spent 2 years, perhaps, trying to correct Om's kernel development approach?
Apr 17 20:29:37 <DocScrutinizer> PaulFertser: that's not my point
Apr 17 20:29:49 <DocScrutinizer> PaulFertser: 47k is
Apr 17 20:29:50 <TAsn> mwester, are you by any chance british and 50 y/o? ;]
Apr 17 20:30:17 <PaulFertser> DocScrutinizer: do you think that i'd rather use a non-OM charger than a PC? 
Apr 17 20:30:45 <mwester> No, I am American (Dutch, originally) and almost 50 y/o -- I am a curmudgeon. :p
Apr 17 20:32:01 <TAsn> mwester, exactly what I was suggesting ;]
Apr 17 20:32:15 <DocScrutinizer> PaulFertser: I don't exactly see the issue when using a PC (except your y-cable reasonably should have a female USB-A receptacle on the other end anyway)
Apr 17 20:32:54 <PaulFertser> DocScrutinizer: wouldn't FR try to sink 1A from a PC if i have a 42k hardwired?
Apr 17 20:32:58 <PaulFertser> 47 even
Apr 17 20:33:08 <DocScrutinizer> probably yes
Apr 17 20:33:20 <PaulFertser> _That_ is my point ;)
Apr 17 20:33:57 <DocScrutinizer> depends on userland managing this. There's no real point as 47k also should switch to HOSTMODE ;-)
Apr 17 20:34:08 <PaulFertser> Anyway, this Y-cable discussion doesn't mkae much sense. ;) Everybody builds it the way he likes :)
Apr 17 20:34:19 <DocScrutinizer> sure
Apr 17 20:35:07 <DocScrutinizer> and I consider building it with a switch to adapt for different usecases
Apr 17 20:35:10 <mwester> TAsn:  To be serious, my job is field tech support, so I am paid to deal with very difficult people, and to be very political, and polite, and accomodating, while at the same time figuring out how to solve a problem.  It is a very challenging job, and for my hobby I do not wish to have to do that as well -- I can say the truth here.
Apr 17 20:35:38 <TAsn> mwester, I'm kidding ;]
Apr 17 20:35:44 <lindi-> PaulFertser: except those who are too lazy to build one :)
Apr 17 20:35:50 <DocScrutinizer> 15k on/off; ID=[0|47k|NC]
Apr 17 20:36:00 <mwester> I know, but behind all joking is a grain of truth... :)
Apr 17 20:36:00 <lindi-> PaulFertser: i'd love to be able to buy one as an openmoko accessory ;)
Apr 17 20:36:08 <PaulFertser> mwester: tough job, wish you patience.
Apr 17 20:36:29 <TAsn> mwester, yeah, the truth is I think your are a curmudgeon, though I'm joking when I complain about it.
Apr 17 20:36:42 <DocScrutinizer> now we get to know mwester's true nature ;-D
Apr 17 20:36:43 <PaulFertser> DocScrutinizer: btw, do you think 30k will be ok? I mean i want to be able to use the same switchless cable both for charger-powered and FR-powered cases.
Apr 17 20:36:47 <mwester> PaulFertser:  it comes with great rewards, solving a problem in a lab was never so much fun as solving a _real_ problem for a customer.  :)
Apr 17 20:36:51 <TAsn> You kicking OM in the balls all the time serves my interests 
Apr 17 20:37:09 <DocScrutinizer> lindi-: how much would you pay?
Apr 17 20:37:17 <lindi-> DocScrutinizer: good question, 10-20 EUR?
Apr 17 20:37:33 <TAsn> Anyhow, I'm off to bed. From now on my patches will include a blog post, like in the wiki entry.
Apr 17 20:37:33 <mwester> PaulFertser:  30K ?  Isn't in the ambiguous range betwen 47K and 15K?
Apr 17 20:37:34 <TAsn> night.
Apr 17 20:37:37 <DocScrutinizer> 20EUR might be the break-even
Apr 17 20:37:43 <PaulFertser> mwester: i mean for pulldown.
Apr 17 20:37:46 <mwester> Ah
Apr 17 20:37:55 <DocScrutinizer> for me building a small series
Apr 17 20:38:25 <mwester> PaulFertser:  curiousity question for you...  how hard is it for you to order something like a cable or an adaptor from a US or Canadian company?
Apr 17 20:38:27 <DocScrutinizer> PaulFertser: I'd use 18k
Apr 17 20:38:41 <DocScrutinizer> PaulFertser: a wild guess
Apr 17 20:38:50 <DocScrutinizer> maybe 22k
Apr 17 20:39:00 <lindi-> DocScrutinizer: i could ask around how many others would be intersted here in finland
Apr 17 20:39:09 <PaulFertser> mwester: Not that hard. Probably expensive. And have to use some services to do paypal payment.
Apr 17 20:39:19 <mwester> You can't do paypal?
Apr 17 20:39:50 <PaulFertser> DocScrutinizer: why do you think that 8k pulldown (when both are active) won't be too strong? Why do you prefer strong pulldowns to weak?
Apr 17 20:40:24 <DocScrutinizer> PaulFertser: gut feeling
Apr 17 20:40:36 <DocScrutinizer> PaulFertser: or call it experience
Apr 17 20:40:46 * mwester applies ohms law...
Apr 17 20:41:12 <PaulFertser> mwester: not in a decent way. I can get visa classic and transfer money from visa to paypal accounts. Not the other way. Not doable without a visa/mastercard card or some questionable services.
Apr 17 20:41:39 <PaulFertser> mwester: That said, many advanced users buy on eBay from time to time. I've no such experience.
Apr 17 20:42:05 * mwester knows very little about such things in Europe, other than some dealing with nl, uk, it
Apr 17 20:42:38 <PaulFertser> At least EU doesn't have to face paypal's inappropriate policies.
Apr 17 20:42:40 <DocScrutinizer> generaly speaking it's a PITA
Apr 17 20:42:58 <DocScrutinizer> to order from outside EU
Apr 17 20:43:04 <mwester> Ok, I just ask because it seems you have access to a lot of high-tech stuff, and I'm wondering how folks get it if it is so expensive to get across the border, and if one needs visas just to go to .de (for example).
Apr 17 20:43:54 <PaulFertser> mwester: what kind of high-tech stuff are you talking about? We can't even buy a new N810 in Moscow.
Apr 17 20:44:24 <mwester> You have access to smt soldering gear, it seems -- something I have found difficult to locate here.
Apr 17 20:44:32 <DocScrutinizer> lol, there's no new ones anymore. Not just in moscow, raher nowhere
Apr 17 20:45:12 <mwester> Motorola has lots of it, but I somehow think that showing up at their cellular dev. center with a Neo or FR to have a buzzfix done will not be effective.  Humorous, but not effective.
Apr 17 20:45:15 <PaulFertser> mwester: I just work at a company developing and producing electronic devices :) And imo some occassional smt soldering can be done with a regular soldering iron.
Apr 17 20:45:22 <DocScrutinizer> And my spare N810 is waiting for pickup at he store for a week now :-/
Apr 17 20:46:19 <mwester> Well, I'm going to have to give the buzz work a try.
Apr 17 20:46:19 <PaulFertser> mwester: and about smt soldering on streets, well, we have a cheap labour cost, that's why even reballing BGA in cell phones is cost-effective. SMT soldering stations can be easily bought in Moscow.
Apr 17 20:46:51 <DocScrutinizer> mwester: sorry to hear this ;-)
Apr 17 20:46:58 <mwester> hehe!
Apr 17 20:47:02 <mwester> I need to locate the parts.
Apr 17 20:47:18 <mwester> "Radio Shack" doesn't have them... :p
Apr 17 20:47:22 <DocScrutinizer> hmm, here's plenty of them next to me
Apr 17 20:47:46 <DocScrutinizer> oh noes!! radio-shack! doesn't have!?!?! ;D
Apr 17 20:48:17 <mwester> Yeah, I was kinda hoping the Om would send a shipment to one of the North American Om employees or distributors, who would do the transactions on this side of the ocean. 
Apr 17 20:48:30 <DocScrutinizer> honestly, our "radio shack" as well doesn't have them
Apr 17 20:48:44 <mwester> There was Michael (somebody) in california, iirc.
Apr 17 20:48:56 <DocScrutinizer> michael shiloh?
Apr 17 20:48:59 <PaulFertser> mwester: take into account that some people joke about moscow being a different country (cheap internet, plenty of parts and devices, job opportunities, etc).
Apr 17 20:49:40 <DocScrutinizer> getting-shot opportunities
Apr 17 20:50:01 <PaulFertser> Nah, i guess it's not even comparable to Bronx.
Apr 17 20:50:50 <DocScrutinizer> nah in Bronx they don't need KGB to kill one another ;-)
Apr 17 20:51:13 <PaulFertser> (we don't have many illegal arms here, but yes, violence on streets is common in some places :-( )
Apr 17 20:51:14 <mwester> It's "self-serve" in the Bronx, and on the south-side of Chicago.
Apr 17 20:52:29 <mwester> Yeah, it was Michael Shiloh.  Haven't heard from him in a loong time, I guess he's also gone.
Apr 17 20:52:46 <DocScrutinizer> Think so
Apr 17 20:53:04 <PaulFertser> He confirmed that on IRC several months ago.
Apr 17 20:53:18 <DocScrutinizer> but steve should still live in US as well
Apr 17 20:55:09 <DocScrutinizer> mwester: if you're willing to do the sendout business yourself I'm sure OM would send a few 100 parts to you for free
Apr 17 20:56:15 <mwester> Not sure how many folks are even interested on this side of the pond, to be honest.  We should probably enquire... and maybe set something up.
Apr 17 20:57:36 <mwester> SHR is pretty close to making a working device out of this thing, the buzzfix is probably the biggest issue remaining for many folks.
Apr 17 20:57:38 <DocScrutinizer> mwester: there's no need to know how many. Just ask OM for 200 and send out on request
Apr 17 20:58:10 <DocScrutinizer> I bet Ainulindale will love to hear this
Apr 17 20:58:47 <mwester> oops - must run out for a bit -- back shorty!
Apr 17 21:07:01 <wurp2|working> mwester: Don't call me shorty!
Apr 17 21:07:11 <wurp2|working> My height is roughly within the norm
Apr 17 21:15:05 <lindi-> DocScrutinizer: maybe you should invent some freerunner docking station product that'd have Y-cable, tiny hub and small space for usb devices? ;)
Apr 17 21:15:35 <DocScrutinizer> ME problems
Apr 17 21:15:57 <lindi-> DocScrutinizer: ME?
Apr 17 21:16:06 <DocScrutinizer> mech eng
Apr 17 21:16:42 <lindi-> ok yeah, attachting it to the phone reliably won't be easy
Apr 17 21:17:09 <lindi-> http://www.usbfirewire.com/ucablecam5b.html seems to have lots of weird angled usb cables
Apr 17 21:18:46 <DocScrutinizer> lindi-: add it to the wiki then!
Apr 17 21:19:11 <DocScrutinizer> "parts"
Apr 17 22:10:25 <lindi-> PaulFertser: you wouldn't happen to know a nice way to add timestamps to traffic captured from serial port?
Apr 17 22:11:24 <PaulFertser> lindi-: yes, i wouldn't :( i guess you'll end up with some ad-hoc scheme.
Apr 17 22:12:52 <PaulFertser> lindi-: like have a special start-char after which a timestamp is expected and escape that start-char everywhere else...
Apr 17 22:12:55 <lindi-> ok i'll ad-hoc
Apr 17 22:13:30 <lindi-> Since it seems I can nicely reproduce #1024 here
Apr 17 22:17:04 <budfive> lindi-: are you working on 1024?
Apr 17 22:19:02 <lindi-> PaulFertser: script -c '(echo -n ws;read) | socat - /dev/ttyACM0 | cat -A | while read l; do echo "`date` $l"; done' calypso.log
Apr 17 22:20:06 <lindi-> budfive: i'd love to help
Apr 17 22:20:25 <budfive> lindi-: i talked to dieter about it for a while, and sent him some logs
Apr 17 22:21:12 <lindi-> budfive: yep you were faster :)
Apr 17 22:21:20 <budfive> lindi-: He was saying that there are big problems with my 32khz oscillator and that my gta02 recamps as soon as it tries to deep sleep.
Apr 17 22:21:37 <budfive> lindi-: so in my case at least the 32khz oscillator is directly responsible
Apr 17 22:21:47 <lindi-> budfive: what was the conclusion on how that could be fixed?
Apr 17 22:21:48 <budfive> lindi-: i'm sending him my phone so that he has something that reliably shows 1024
Apr 17 22:22:26 <budfive> lindi-: he sent me a custom firmware to run the oscillator with more current, but it didn't noticeably change anything
Apr 17 22:22:40 <budfive> lindi-: i'm probalby the only one with moko12b firmware in my calypso. :)
Apr 17 22:22:59 <lindi-> budfive: what does that fix?
Apr 17 22:23:06 <lindi-> ah
Apr 17 22:23:21 <budfive> lindi-: it was his firmware to allow the oscillator current to be programmable
Apr 17 22:24:50 <budfive> lindi-: looks like alphaone was playing with it too. Look at http://totalueberwachung.de/~alphaone/1024/
Apr 17 22:25:41 <lindi-> budfive: hmm, dieter should really put some comments to the bug tracker too
Apr 17 22:26:35 <budfive> lindi-: i guess. it seems like he's the only one who can really work on the bug. Nobody else can even interpret the logs, i think
Apr 17 22:27:36 <budfive> lindi-: if you have smt soldering equipment and a signal generator, you can try your own 32khz signal and see if that fixes it (i don't have a signal generator)
Apr 17 22:27:56 <budfive> lindi-: from the pictures it looks like alphaone tried that, though :)
Apr 17 22:28:24 <lindi-> budfive: i don't but my collegue at work has
Apr 17 22:28:40 <lindi-> budfive: but I'm not allowed to use gsm at work which kind of makes this difficult
Apr 17 22:30:02 <lindi-> budfive: ok and alphaone did not succeed?
Apr 17 22:30:34 <budfive> lindi-: no idea. didn't talk to him. just randomly stumbled on the pictures. :)
Apr 17 22:34:37 <lindi-> alphaone: any comments? ;)
Apr 17 22:36:05 <PaulFertser> lindi-: i thought you need to add timestamps on the avr itself.
Apr 17 22:36:36 <lindi-> PaulFertser: yes later
Apr 17 22:36:45 <lindi-> PaulFertser: now I'm just streaming at this state
Apr 17 22:37:05 <PaulFertser> lindi-: great you found a way to do it.
Apr 17 22:37:14 <lindi-> i need to make it more robust before making it log to flash
Apr 17 22:37:29 <lindi-> and for that i need to check the timing with oscilloscope that i only have at work
Apr 17 22:38:29 <lindi-> PaulFertser: yep I have the AVR generate an interrupt on falling edge. then i wrote interrupt handler in assembler that reads 1 + 8 + 1 bits without any loops
Apr 17 22:39:11 <lindi-> PaulFertser: it then returns from the interrupt handler as fast as possible so that the C code has 0.5 bits (33 cycles) to execute before the next character triggers an interrupt
Apr 17 22:39:55 <alphaone> Hehe
Apr 17 22:40:24 <alphaone> lindi-, budfive: The results were inconclusive since I didn't have a trace cable at that time.
Apr 17 22:40:34 <alphaone> I need to retry this soon
Apr 17 22:40:38 <alphaone> Got a cable now
Apr 17 22:40:59 <alphaone> So I can really see if the Calypso even goes to deep sleep
Apr 17 22:41:39 <budfive> alphaone: are you talking to dieter? he'd love to know anything you can tell him
Apr 17 22:42:13 <alphaone> budfive: Yes, he knows, we're talking
Apr 17 22:42:38 <alphaone> Didn't post to any list yet since my "results" are just confusing at best...
Apr 17 22:43:12 <budfive> alphaone: did you try to replace the mystery resistor? you have the equipment for it, righ?
Apr 17 22:43:51 <alphaone> budfive: That was the first thing I did
Apr 17 23:35:05 <DocScrutinizer> budfive: bragger ;-) moko12b, pfff
Apr 17 23:39:08 <DocScrutinizer> it's really surprising you say your osc is definitely borked. As this isn't the standard issue we see when looking into #1024. So your case might even be a red herring on efforts to find true nature of #1024
Apr 17 23:39:26 <billk_> Still trying to build shr-testing - getting "METADATA_BRANCH   = "fso/milestone5"
Apr 17 23:39:54 <billk_> at the start and error: pathspec 'fso/milestone5.5' did not match any file(s) known to git.
Apr 17 23:39:59 <billk_> in make update.
Apr 17 23:40:14 <billk_> something I need to change?
Apr 17 23:40:36 <mwester> billk_:  sounds like you best just start over.
Apr 17 23:40:49 <billk_> rm -rf * ?
Apr 17 23:40:51 <budfive> DocScrutinizer: well, how is it non-standard?
Apr 17 23:41:01 <budfive> DocScrutinizer: the symptoms are 1024 to the letter
Apr 17 23:41:02 <mwester> billk_:  keep your "downloads" directory.
Apr 17 23:41:11 <billk_> tkx :(
Apr 17 23:41:48 <budfive> DocScrutinizer: it's certainly possible there are multiple causes of 1024 and i'm in the minority of those, but do we have evidence of that?
Apr 17 23:41:49 <DocScrutinizer> all other tests were non-conclusive wrt to osc so far. It even seems more likely osc isn't the issue at all
Apr 17 23:42:12 <budfive> DocScrutinizer: dieter looked at other people's logs and saw other causes?
Apr 17 23:42:39 <DocScrutinizer> budfive: you can't tell the cause from the logs
Apr 17 23:43:16 <mwester> All of the different triggers may just be that - triggers to some deeper problem that we do not yet understand.
Apr 17 23:43:54 <budfive> DocScrutinizer: i haven't attempted to parse the logs myself. All I'm saying here comes from Dieter
Apr 17 23:43:57 <mwester> It might even be something as simple as a race condition in the software in the Calypso, and anything that affects the timing can change the symptoms.
Apr 17 23:44:09 <DocScrutinizer> but e.g. alphaone did a nice series of tests and they look quite decent for stability of osc (well some of them, the ones that weren't spilled by humm introduced by probe)
Apr 17 23:44:18 <mwester> But I'm sure glad that people are still working on this.
Apr 17 23:44:32 <DocScrutinizer> mwester: exactly (triggers)
Apr 17 23:45:16 <DocScrutinizer> mwester: (race) even a quite possible cause
Apr 17 23:45:18 <budfive> DocScrutinizer: Dieter was saying the 32khz oscillator needs something on the order of 1ppm precision. When I scoped the output of mine, it looked fine, but I didn't check how precise it was. Most likely not that precise
Apr 17 23:46:04 <DocScrutinizer> nah, that's a misperceptiuon as osc is "tuned" so absolute precision doesn't really matter
Apr 17 23:46:27 <mwester> Stability would be the issue then, though.
Apr 17 23:46:33 <DocScrutinizer> right
Apr 17 23:46:52 <DocScrutinizer> we found no instabilities so far
Apr 17 23:47:08 <budfive> DocScrutinizer: Either way, I'm sending him my phone. Hopefully it'll help
Apr 17 23:47:14 <mwester> I have 4 devices here -- all of them "bounce", and they do it all the time, regardless of temperature or other conditions (battery, etc).
Apr 17 23:47:17 <DocScrutinizer> though it's quite clear #1024 can be trggered by disturbing the osc
Apr 17 23:47:54 <mwester> nod
Apr 17 23:48:52 <DocScrutinizer> the only reasonable thing to do at present state of investigations is to see whether we can STOP #1024 by applying a rocksolid clock. If *not* the osc is off the list
Apr 17 23:49:21 <budfive> DocScrutinizer: alphaone was about to do that i believe
Apr 17 23:49:30 * mwester would think we should pull on the other strings -- what other factors trigger the bouncing?
Apr 17 23:49:42 <DocScrutinizer> yup, and he got no conclusive results on that either
Apr 17 23:49:51 <mwester> Yep.
Apr 17 23:49:54 <DocScrutinizer> mwester: fw bugs
Apr 17 23:50:01 <mwester> I think so.
Apr 17 23:50:16 <DocScrutinizer> I suspect the "tuning"
Apr 17 23:51:09 <budfive> mwester: does your phone retain registration in deep sleep at all, or does it bounce immediately when it enters deep sleep (that's what mine does)
Apr 17 23:51:15 <DocScrutinizer> FW is checking how long (in clock cycles) one timeslice of GSM really is. that would eliminate problems with absolute precision of osc
Apr 17 23:51:19 <mwester> So there is some hope that we could get a fix, or if not a fix, perhaps an in-calypso workaround that would let it stay asleep.  And that's probably the biggest remaining power issue we have left in the device.
Apr 17 23:51:42 <mwester> budfive:  retains registration for up to 20 - 35 seconds.
Apr 17 23:51:45 <mwester> Then wake up.
Apr 17 23:52:02 <DocScrutinizer> mwester: sleep=2 is a punishment of ~2..4mA
Apr 17 23:52:10 <DocScrutinizer> not THAT bad
Apr 17 23:52:19 <mwester> It is when the phone is suspended.
Apr 17 23:53:10 <DocScrutinizer> yup, so what. It's as well if the phone is up and running
Apr 17 23:53:44 <DocScrutinizer> mwester: it takes about 30sec for calypso to enter deepsleep after last "action" anyway
Apr 17 23:53:47 <mwester> User expectations is the difference.
Apr 17 23:54:20 <DocScrutinizer> hmm, so maybe moko12 disables sleep=4, and we sell it as "fixed" ;-D
Apr 17 23:54:40 <mwester> If I'm using it to play mp3's I don't expect it to last long on the battery, a few ma extra makes no difference.  But if it's in my pocket, asleep all day, I want it to last as long as possible, and the extra ma will decrease that time measurably.
Apr 17 23:55:16 <DocScrutinizer> everything is "meassurable"
Apr 17 23:55:27 * mwester rolls his eyes.
Apr 17 23:56:13 <DocScrutinizer> atm we get a standby of ~70h (that's what mickey says for his phones, and those all sufer from #1024)
Apr 17 23:56:21 <DocScrutinizer> do the math yourself
Apr 17 23:57:42 <DocScrutinizer> or, for your convenience, we *might* see a diff like "now it's 90h" with sleep=4
Apr 17 23:58:09 <DocScrutinizer> probably less
Apr 18 00:00:10 <DocScrutinizer> anyway, sure it's quite desirable to fix #1024
Apr 18 00:00:26 <DocScrutinizer> or at least understand it's nature
Apr 18 00:00:40 <DocScrutinizer> for now we're still far from that
Apr 18 00:02:43 <DocScrutinizer> for continiuing speculations about cause: if there's a deviation introduced into the osc32+tuning system - either by osc instability or even by a mishap of BTS - this might mess up the tuning part in FW to a point where it never recovers
Apr 18 00:03:27 <DocScrutinizer> after all sleep=4 is a "young" feature of calypso FW
Apr 18 00:05:51 <DocScrutinizer> dieter is mentioning there's an errorflag set and never reset when #1024 happens. Same flag appears occasionally in normal operation, but is reset on next "cycle" without #1024 present
Apr 18 00:06:36 <DocScrutinizer> so maybe tuning algo is somehow locking under certain circumstances
Apr 18 00:07:12 <DocScrutinizer> overflow of a tunig-offset register or dunno what
Apr 18 00:07:22 <DocScrutinizer> we don't have source of this part
Apr 18 00:08:37 <DocScrutinizer> alphaone replaced the mystic 220k by 100k and it didn't show the "bingo!!" effect. So this also is off the list
Apr 18 00:10:01 <DocScrutinizer> precision of xtal 10ppm to 20..30ppm only means deviation of actual tuning of particular part from specified freq. It doesn't tell anything about stability
Apr 18 00:10:14 <DocScrutinizer> so this is off the list as well
Apr 18 00:11:32 <DocScrutinizer> well. maybe not completely. A xtal "far off" the specified freq might more likely drive tuning algo to some limit where this bugger chokes
Apr 18 00:12:09 <DocScrutinizer> *might*
Apr 18 00:12:48 <DocScrutinizer> or it's sth totally different nce again (see GPS and sdcard clock)
Apr 18 00:14:56 <DocScrutinizer> mwester: honestly the gprs-lockup due to memleak is scaring me muuuch more
Apr 18 00:18:38 <budfive> DocScrutinizer: that issue (gprs) doesn't show up in muxer-less operation (at least not as readily). So maybe the calypso responds badly to the muxer ignoring the flow control
Apr 18 00:19:03 <DocScrutinizer> maybe
Apr 18 00:19:08 <budfive> DocScrutinizer: trying to break gprs with abyss is on my list for tonight
Apr 18 00:19:30 <DocScrutinizer> but Dieter said there's a memleak. And it didn't sound he was in doubt of that
Apr 18 00:20:17 <budfive> DocScrutinizer: true. there is a bug. i'm hopefull we can avoid tickling it :)
Apr 18 00:20:35 <DocScrutinizer> so no mater if it's mux operation or sth else, Dieter seems determined to find that memleak and he seems convinced it's cause of the problem
Apr 18 00:21:24 <budfive> DocScrutinizer: so is he actually reverse-engineering the firmware? 
Apr 18 00:21:47 <budfive> DocScrutinizer: if he's trying to find a memory leak in something for which he doesn't have sources...
Apr 18 00:21:51 <DocScrutinizer> and he said it's difficult to virtually impossible to find this bug, as there's a lot of dynamically allocated space for IPC, and we don't have the sources for that
Apr 18 00:22:04 <DocScrutinizer> exactly
Apr 18 00:22:22 <DocScrutinizer> almost Dieter's words
Apr 18 00:23:43 <DocScrutinizer> so THIS is a *really* *nasty* bug. No workaround (yet?). No way to recover - except powercycling modem. And a real showstopper, not a small impact on standby time
Apr 18 00:26:43 <DocScrutinizer> budfive: anyway your help is greatly appreciated. Just to mention it :)
Apr 18 00:42:02 <zing> i just did a opkg upgrade to the latest in shr-testing on my gta02 and lost all icons and no usb net access.
Apr 18 00:42:19 <zing> have an illume bar on top though :)
Apr 18 00:43:01 <penguinator> happened to me too, i just reflashed
Apr 18 00:43:26 <zing> penguinator, good to know i'm not the only one...  i will reflash also.  thanks.
Apr 18 00:43:47 <penguinator> someone might have a fix, but i didn't bother to find it
Apr 18 00:44:03 <Sharwin_F> zing, about the usb, look if on your PC you have eth1 and no usb0
Apr 18 00:45:23 <zing> Sharwin_F, nope, neither came up... my dmesg  is silent also about anything connecting by usb
Apr 18 01:03:52 <DocScrutinizer> as I get it, testing seen major upgrade yesterday. So maybe it's just a good idea to reflash, and opkg upgrade just breaking things
Apr 18 01:05:42 <DocScrutinizer> zing: I flashed http://build.shr-project.org/shr-testing/images/om-gta02/shr-image-om-gta02.jffs2 yesterday and it worked quite nice.
Apr 18 01:08:31 <DocScrutinizer> [2009-04-17 18:19:30] <mwester> opkg upgrade != "a reasonable thing to do with SHR at present, either stable or testing"
Apr 18 01:09:27 <DocScrutinizer> [2009-04-17 18:16:31] <Zorkman> is there a big difference between the current testing and the unstable one?
Apr 18 01:09:29 <DocScrutinizer> [2009-04-17 18:16:48] <dos1> Zorkman: i don't think there is any difference ;)
Apr 18 01:09:31 <DocScrutinizer> [2009-04-17 18:17:35] <Ainulind1le> Zorkman: they are equal
Apr 18 01:10:55 <DocScrutinizer> [2009-04-17 15:52:04] <Weiss> is the new "Testing" safe to upgrade to from the old testing at the moment? - is opkg upgrade sensible, or is a reflash required?
Apr 18 01:10:56 <DocScrutinizer> [2009-04-17 15:56:15] <Ainulindale> Weiss: nope
Apr 18 01:10:58 <DocScrutinizer> [2009-04-17 15:56:16] <Ainulindale> reflash
Apr 18 01:12:31 <DocScrutinizer> zing: penguinator: HTH
Apr 18 01:33:23 <zing> DocScrutinizer, yes thankyou, i was afk for awhile there... i will definitely reflash
Apr 18 01:33:45 <DocScrutinizer> enjoy! :-)
**** ENDING LOGGING AT Sat Apr 18 02:59:57 2009