**** BEGIN LOGGING AT Tue May 05 02:59:57 2009
May 05 05:14:20 <Ainulindale> Meh, sleeping is impossible today.
May 05 05:34:35 <CIA-75> SHR: 03mok 07shr-overlay * r58c23bec6943 10/openembedded/ (2 files in 2 dirs): add recipe for intone music player
May 05 05:34:46 <Ainulindale> Oh, a mrmoku|a` 
May 05 05:37:19 <mrmoku|a`> Ainulindale: hey
May 05 05:37:52 <mrmoku> damn bullshit...
May 05 05:37:57 <Ainulindale> What?
May 05 05:38:00 <mrmoku> not even the lite image finished to build
May 05 05:38:26 <mrmoku> /home/shr/shr-unstable/tmp/work/armv4t-angstrom-linux-gnueabi/dbus-hlid-0.9.0+gitre82bb41c8f7b41b85db918166db6682962853713-r0/git/src/obj.c:874: error: too many arguments to function 'server_ListBusNames'
May 05 05:38:41 <Ainulindale> meh
May 05 05:39:16 <mrmoku> who needs it anyway?
May 05 05:39:23 <Ainulindale> =)
May 05 05:42:27 <mrmoku> so... part of the vala stuff does not build with old vala... I update vala... and now part of the vala stuff does not build with new vala
May 05 05:42:30 <mrmoku> crap ! :(
May 05 05:43:19 <Ainulindale> that's ptitjes' fault :-)
May 05 05:43:54 <mrmoku> like allways ;)
May 05 05:44:10 <mrmoku> ptitjes: ping... come here... want to spank you :P
May 05 05:44:46 * mrmoku is wondering though... because ptitjes did not even commit in that project ;D
May 05 05:44:46 <ptitjes> mhhhhh
May 05 05:45:24 <mrmoku> ptitjes: can you fix that?
May 05 05:46:03 <ptitjes> yeah! sure: you have too many arguments to function 'server_ListBusNames'
May 05 05:46:08 <ptitjes> ok good night
May 05 05:46:10 <Ainulindale> haha
May 05 05:46:11 <Ainulindale> dumbass
May 05 05:46:15 <mrmoku> hmm...
May 05 05:46:29 <mrmoku> how comes you think that :P
May 05 05:47:20 <ptitjes> hello btw
May 05 05:47:36 <tobii_> good morning
May 05 05:47:44 <mrmoku> hmm... maybe I will spank myself even...
May 05 05:47:46 <mrmoku> lets see
May 05 05:48:27 <ptitjes> mrmoku: sorry I don't know about dbus-hlid yet
May 05 05:48:45 <Ainulindale> ptitjes: holidays for me this week, available 24/7
May 05 05:48:51 <mrmoku> me neither... but I found out that it is *not* set to autorev
May 05 05:49:05 <ptitjes> I work on saturday
May 05 05:49:06 * mrmoku starts spanking himself
May 05 05:49:07 <Ainulindale> ptitjes: I also wrote a mail on shr-devel waiting for your opinion
May 05 05:49:22 <mrmoku> Ainulindale: what are holidays? some french invention?
May 05 05:49:43 <ptitjes> Ainulindale: I started an answer yesterday at 12h but then felt asleep
May 05 05:49:44 <ptitjes> :(
May 05 05:49:52 <Ainulindale> well I didn't sleep (yet)
May 05 05:50:01 <ptitjes> I've seen that :)
May 05 05:50:10 <Ainulindale> ptitjes: oh and I started working on some NAS config which might be of interest to you
May 05 05:50:28 <ptitjes> Ainulindale: Mainly, I'll add Profiles
May 05 05:50:39 <ptitjes> s/ll/d/
May 05 05:50:39 <bzzbot> ptitjes meant: Ainulindale: Mainly, I'd add Profiles
May 05 05:50:54 <Ainulindale> Profiles would be in "settings"
May 05 05:51:14 <Ainulindale> think "macro" functional contexts
May 05 05:51:20 <Ainulindale> even though we can subdivide them
May 05 05:51:25 <ptitjes> Ainulindale: I would like application to request a profile change
May 05 05:51:38 <Ainulindale> ptitjes: so?
May 05 05:51:44 <ptitjes> Ainulindale: a screen to appear and ask user for the change
May 05 05:51:49 <Ainulindale> so?
May 05 05:52:06 <ptitjes> Settings might be more and more detailed
May 05 05:52:15 <Ainulindale> So?
May 05 05:52:25 <Ainulindale> How isn't it coherent with what I just said?
May 05 05:52:43 <mrmoku> ptitjes: what Ainulindale means is... settings has some EditProfiles() method and some SelectProfile() method...
May 05 05:52:56 <ptitjes> Though I would also like applications to be able to request the Settings to appear with a settings filter (a la Eclipse)
May 05 05:53:09 <Ainulindale> I'd say more something along the lines of org.shr.ophonekitd.Settings.Profiles.MethodToKickYourAss()
May 05 05:53:17 <Ainulindale> Settings is macro
May 05 05:53:20 <Ainulindale> Profiles is micro
May 05 05:53:33 <Ainulindale> See what I mean ptitjes?
May 05 05:53:40 <ptitjes> Ainulindale: ok but then must be detailed
May 05 05:53:51 <Ainulindale> ptitjes: that's why I wrote functional contexts
May 05 05:53:56 <Ainulindale> Not functional "functions" :-)
May 05 05:53:59 <ptitjes> Ainulindale: like I added a comment after "Dialer" "Do you include Call Management there ?"
May 05 05:54:11 <Ainulindale> Send your reply then we'll talk
May 05 05:54:30 <ptitjes> Wait I'm trying to remember what I wanted to write...
May 05 05:54:51 <Ainulindale> Dialer is all that is related to call, "Dialer" is just a handler here, not a final name
May 05 05:54:57 <ptitjes> Ainulindale: did you see what I did for Vala and DBus ?
May 05 05:55:12 <Ainulindale> no
May 05 05:55:18 <Ainulindale> apart from vala-dbus-binding-tool
May 05 05:55:42 <ptitjes> We now have integer enumerations serialized as string over dbus
May 05 05:55:48 <Ainulindale> yeah I know
May 05 05:55:48 <ptitjes> (it is part of the language)
May 05 05:55:52 <ptitjes> :D
May 05 05:56:13 <ptitjes> Also proper error support is here but I didn't sent a mail yet
May 05 05:57:52 <ptitjes> Ah yes
May 05 05:57:55 <ptitjes> Location
May 05 05:58:09 <Ainulindale> send a mail then we'll talk
May 05 05:58:16 <ptitjes> yep
May 05 05:58:16 <Ainulindale> although I think "Location" is too specific :-p
May 05 05:58:18 * ptitjes is writing
May 05 05:58:32 <Ainulindale> (just to piss you off)
May 05 05:58:36 <ptitjes> Ainulindale: I know you will say it is not phone-related but I don't aggree
May 05 05:58:41 <ptitjes> :)
May 05 05:59:16 <Ainulindale> I won't say that
May 05 05:59:20 <Ainulindale> It's obvious
May 05 05:59:22 <Ainulindale> it's not phone related
May 05 05:59:27 <Ainulindale> but it's ophonekitd related
May 05 05:59:32 <Ainulindale> The thing is to find the right context
May 05 05:59:42 <Ainulindale> If there's no other, then let us think about an appropriate one
May 05 05:59:49 * mrmoku was thinking about Location too... but then was not sure in how far that will be in frameworkd...
May 05 06:00:13 <Ainulindale> mrmoku: throw ideas even if they sound stupid then we'll have material
May 05 06:00:17 <Ainulindale> the intent is to be able to talk
May 05 06:00:34 <Ainulindale> (I'm currently wondering the amount of cafeine in my blood)
May 05 06:00:39 <mrmoku> well... I'm the only one that started throwing :P
May 05 06:00:48 <mrmoku> (and will continue to do so)
May 05 06:00:54 <Ainulindale> don't undermine my contribution :-)
May 05 06:01:08 <mrmoku> Ainulindale: :)
May 05 06:01:26 * mrmoku stops spanking himself, because dbus-hlid autorev still does not build
May 05 06:01:29 <mrmoku> mickey|bbq: ping
May 05 06:01:50 <alphaone> Ainulindale: btw since the buzz fix party I'm using shr as well :-)
May 05 06:02:03 <alphaone> It's quite nice I must say.
May 05 06:02:22 <Ainulindale> You just started?
May 05 06:02:24 <Ainulindale> I hate you more than ever
May 05 06:02:30 <alphaone> Ainulindale: Heh
May 05 06:02:35 <Ainulindale> But good to hear anyway
May 05 06:02:41 <Ainulindale> alphaone: still waiting for your contribution on shr-devel
May 05 06:02:44 <Ainulindale> Please reply
May 05 06:02:47 <Ainulindale> I'd like your opinion as well
May 05 06:02:55 <Ainulindale> As you're bounded to help us building that
May 05 06:02:55 <alphaone> Well, been using zhone/fso image until then
May 05 06:03:01 <Ainulindale> zhone sucks :-)
May 05 06:03:09 <alphaone> Yeah, I need to think about that.
May 05 06:03:15 <alphaone> Just read a minute ago
May 05 06:03:16 <Ainulindale> SHR will rule as soon as we'll do our dbus thing
May 05 06:03:19 <Ainulindale> I'm damn serious
May 05 06:03:29 <Ainulindale> UI preloading/caching is a must have
May 05 06:03:31 <alphaone> Still recovering from the buzz fix party :-)
May 05 06:03:34 <Ainulindale> modularity is a must must have for me
May 05 06:03:59 <alphaone> Which resulted in me cleaning up my soldering corner afterwards
May 05 06:04:28 <alphaone> yeah, if you have modularity and are fast that's awesome.
May 05 06:04:36 <alphaone> zhone is....okay.
May 05 06:04:51 <alphaone> What I'm missing is the GPS view zhone offers
May 05 06:05:02 <alphaone> But shr will get that one way or the other :-)
May 05 06:05:15 <Ainulindale> well there's agpsui
May 05 06:05:18 <Ainulindale> or tangogps
May 05 06:05:24 <alphaone> Maybe we can continue zhone simply as debug tool
May 05 06:05:26 <Ainulindale> and you can always launch zhone on SHR
May 05 06:05:47 <Ainulindale> alphaone: well I think the good plan would be to base FSO on top of SHR, IMHO
May 05 06:05:51 <alphaone> Ainulindale: agpsui used the gps port directly last I looked
May 05 06:05:54 <Ainulindale> or the other way around
May 05 06:06:06 <Ainulindale> we're both struggling with stupid issues
May 05 06:06:21 <alphaone> And tgps doesn't tell me which SVs got almanac, ephemeris, ....
May 05 06:06:28 <Ainulindale> yeah
May 05 06:06:34 <alphaone> Yeah
May 05 06:06:41 <alphaone> Btw, is shr in oe now?
May 05 06:06:52 <Ainulindale> waiting for my commit
May 05 06:06:58 <alphaone> Haven't looked since I'm limited to ms5.5 branch at the moment
May 05 06:07:05 <alphaone> or rather limiting myself
May 05 06:07:09 <Ainulindale> I'll do the final review when it'll build in unstable
May 05 06:07:15 <alphaone> great
May 05 06:07:41 <Ainulindale> that is this week hopefully
May 05 06:07:44 <Ainulindale> we normalized everything
May 05 06:07:49 <Ainulindale> we have mickey|bbq's approval
May 05 06:07:52 <Ainulindale> nytowl's too
May 05 06:09:48 <alphaone> Ainulindale: I still have to understand and/or talk you out of doing the resource duplication stuff you're doing in ophonekitd
May 05 06:09:54 <alphaone> cool
May 05 06:10:00 <alphaone> you'll merge where then?
May 05 06:10:19 <alphaone> org.oe.dev and fight with Koen to no end?
May 05 06:10:36 <Ainulindale> nope
May 05 06:10:39 <Ainulindale> fso ms5.5
May 05 06:10:46 <Ainulindale> alphaone: And I still stand my ground
May 05 06:10:52 <alphaone> okay, good
May 05 06:10:54 <Ainulindale> This usage interface is needed
May 05 06:11:08 <Ainulindale> Use case, again
May 05 06:11:11 <alphaone> Ainulindale: Well, if it's needed we need to implement it elsewhere
May 05 06:11:13 <Ainulindale> debug
May 05 06:11:19 <Ainulindale> zhone + ophonekitd
May 05 06:11:24 <Ainulindale> I want to suspend ophonekitd call handling
May 05 06:12:22 <Ainulindale> I disable the GSM in ophonekitd
May 05 06:12:22 <Ainulindale> And not system wide
May 05 06:12:22 <Ainulindale> How can I do that currently?
May 05 06:12:22 <alphaone> Ainulindale: As I said understand and/or
May 05 06:12:22 <Ainulindale> Yeah :-)
May 05 06:12:22 <Ainulindale> That's why I repeated one of my use cases
May 05 06:12:22 <Ainulindale> I still think one should be able to suspend a specific resource owner 
May 05 06:12:35 <alphaone> I guess I'll best understand it when we talk face to face
May 05 06:12:41 <Ainulindale> I'm not sure about that
May 05 06:12:43 <Ainulindale> I'm fuzzy :-)
May 05 06:12:58 <alphaone> Well, that's usually when the resource owner calls ReleaseResource
May 05 06:13:18 <Ainulindale> How can you do that on an external point of view?
May 05 06:13:19 <alphaone> But you will need to tell ophonekitd that somehow...
May 05 06:13:24 <Ainulindale> Voila.
May 05 06:13:37 <alphaone> I will need to learn more about ophonekitd I guess
May 05 06:13:39 <Ainulindale> It'd be doable if settings where within ophonekitd
May 05 06:13:47 <Ainulindale> that's not the case currently
May 05 06:13:51 <Ainulindale> And even if it was
May 05 06:13:56 <Ainulindale> I'd do it on top of dbus
May 05 06:14:01 <Ainulindale> (to be coherent)
May 05 06:14:54 <alphaone> Ah, okay....
May 05 06:15:24 <Ainulindale> If I want to have some "control panel" app anyway
May 05 06:15:24 <alphaone> I thought settings would talk to the fso API directly
May 05 06:15:40 <Ainulindale> Different layers
May 05 06:15:50 <Ainulindale> GSM on a system level or GSM on the SHR level
May 05 06:15:54 <Ainulindale> (i.e. ophonekitd)
May 05 06:16:04 <Ainulindale> ophonekitd is a subsystem by itself
May 05 06:16:37 <alphaone> But - in theory instead of enabling GSM in ophonekitd you could set the ResourcePolicy to enabled, right?
May 05 06:16:48 <alphaone> And instead of releasing that again set it back to auto
May 05 06:17:20 <Ainulindale> we could do that, yes, and that's an interesting alternative we should talk about
May 05 06:17:38 <Ainulindale> but that still means I have to be able to release in ophonekitd stead
May 05 06:17:45 <Ainulindale> As I may have another user :-)
May 05 06:18:00 <raster> pffft
May 05 06:18:09 <raster> Ainulindale: YO
May 05 06:18:12 <Ainulindale> raster: you're the pfft!
May 05 06:18:38 <alphaone> If you do then that user will have requested the resource
May 05 06:18:38 <Ainulindale> You're an uber pfft!
May 05 06:18:38 <alphaone> so auto wont affect anything
May 05 06:18:38 <Ainulindale> alphaone: my point was
May 05 06:18:42 <Ainulindale> If I want to disable GSM specifically in ophonekitd
May 05 06:18:45 <Ainulindale> I can't use resource policy for that
May 05 06:18:47 <raster> Ainulindale: only for you baby.
May 05 06:18:51 <Ainulindale> I have to act on ophonekitd
May 05 06:19:15 <Ainulindale> alphaone: let us say I have GPS handling in ophonekitd for call geolocalization
May 05 06:19:24 <Ainulindale> but that is power consuming
May 05 06:19:28 <Ainulindale> And I also have navit
May 05 06:19:31 <Ainulindale> I see low batteries
May 05 06:19:35 <methril|work> morning
May 05 06:19:38 <Ainulindale> I want to disable GPS in ophonekitd but not in navit
May 05 06:19:54 <Ainulindale> Because I'm on the road and I can't drive without GPS :-)
May 05 06:20:02 <Ainulindale> How would you do that?
May 05 06:20:26 <alphaone> Well, you'll always have to be able to tell ophonekitd what to do and what not.
May 05 06:20:44 <alphaone> Ideally you'd tell it not to do call geolocalization anymore
May 05 06:20:56 <alphaone> And it would know it doesn't need GPS
May 05 06:21:08 <Ainulindale> yep and this way I'd act on ophonekitd
May 05 06:21:19 <Ainulindale> this way I'd need usage
May 05 06:21:30 <Ainulindale> Let us say I didn't install a settings library for ophonekitd
May 05 06:21:35 <Ainulindale> Because I don't want to
May 05 06:21:37 <Ainulindale> How can I do that?
May 05 06:21:45 <ptitjes> sent
May 05 06:21:58 <Ainulindale> Apart from not being an annoying bastard that is
May 05 06:22:14 <Ainulindale> s/an annoying bastard/a raster/
May 05 06:22:15 <bzzbot> Ainulindale meant: Apart from not being a raster that is
May 05 06:22:55 * raster sticks a rhinoceros up Ainulindale's butt-hole
May 05 06:22:56 <alphaone> Well, If you don't need settings you deserve to drain your battery :-)
May 05 06:23:05 <Ainulindale> Oh yeah baby hit me one more time :-)
May 05 06:23:09 <alphaone> raster: You'll need something bigger
May 05 06:23:15 <Ainulindale> alphaone: nah I may have ologicd plus rules
May 05 06:23:24 <Ainulindale> Or a rule processing engine different altogether from ophonekitd
May 05 06:23:28 <raster> alphaone: true. that butt-hole is a bit loose
May 05 06:23:32 <Ainulindale> it might even be an oeventsd rule
May 05 06:23:47 <Ainulindale> alphaone: use case
May 05 06:23:47 <alphaone> Bah!
May 05 06:23:49 <Ainulindale> low battery
May 05 06:23:53 <Ainulindale> oeventsd => dbus method
May 05 06:23:56 <alphaone> go away with oeventsd
May 05 06:23:58 <Ainulindale> "release resource ophonekitd"
May 05 06:24:12 <Ainulindale> ophonekitd receives a "can I release for process X (description)"
May 05 06:24:16 <Ainulindale> ophonekitd sends ACK/NACK
May 05 06:24:32 <Ainulindale> So a specific software can ship with a set of allowed "masters"
May 05 06:24:44 <Ainulindale> that's a possible solution
May 05 06:24:58 <alphaone> Okay, that could be an interesting API
May 05 06:25:07 <Ainulindale> I described that a month ago :-)
May 05 06:25:36 <alphaone> Well, I ignored it a month ago :-)
May 05 06:26:05 <Ainulindale> I know
May 05 06:26:20 <Ainulindale> That's because you consider me as a rast^W^Wan annoying bastard :-)
May 05 06:26:41 <alphaone> ;-)
May 05 06:26:44 <Ainulindale> But I'm not sure this is the right solution please keep that in mind
May 05 06:26:49 <Ainulindale> I'm just looking for the good one
May 05 06:26:56 <Ainulindale> I still think there's a need for such a thing
May 05 06:27:15 <alphaone> No, to be honest I'm still concentrating on the lower levels and haven't thought too much about the problems on the consumer side of our API
May 05 06:27:21 <alphaone> Or multi consumer side
May 05 06:27:33 <Ainulindale> multi is the key factor here
May 05 06:27:43 <alphaone> yeah
May 05 06:28:00 <alphaone> Plus you want to launch applications on events
May 05 06:28:12 <Ainulindale> not exactly
May 05 06:28:14 <alphaone> We have always had the main app running (zhone)
May 05 06:28:18 <Ainulindale> we want to do stuff on events
May 05 06:28:24 <alphaone> yeah
May 05 06:28:41 <alphaone> Ideally we'd do something like android does
May 05 06:28:49 <alphaone> have a sort of events dispatcher
May 05 06:28:50 <Ainulindale> Being crappy?
May 05 06:28:55 <Ainulindale> alphaone: check for ologicd
May 05 06:28:57 <Ainulindale> ptitjes went there
May 05 06:29:13 <alphaone> Ainulindale: I'm very interested in ologicd
May 05 06:29:26 <alphaone> I hope we can kick oeventsd
May 05 06:29:32 <Ainulindale> So do I
May 05 06:29:47 <alphaone> Need to talk with ptitjes at S9N
May 05 06:29:54 <ptitjes> alphaone: I'll do that talk then
May 05 06:29:55 <alphaone> F9N
May 05 06:29:59 <Ainulindale> F9N you mean
May 05 06:30:03 <Ainulindale> :-)
May 05 06:30:08 <alphaone> yeah :-)
May 05 06:30:12 <ptitjes> gtg guys
May 05 06:30:13 <ptitjes> c
May 05 06:30:14 <ptitjes> u
May 05 06:30:17 <Ainulindale> l
May 05 06:30:18 <alphaone> ptitjes: That would be great
May 05 06:30:19 <alphaone> cu
May 05 06:35:37 <Ainulindale> alphaone: waiting for comments... :-)
May 05 06:36:43 * mrmoku has to move into the office now... no holiday today :(
May 05 06:39:55 <alphaone> Ainulindale: ?
May 05 06:39:59 <alphaone> On shr-devel?
May 05 06:46:01 <gustavold> I've upgraded some packages from shr-testing repo, now I'm getting segfault in the call elementary.Window()
May 05 06:46:02 <gustavold> any idea?
May 05 06:47:50 <Ainulindale> alphaone: yep
May 05 07:21:55 <SHR-WugBranler`> tracfeed: Ticket #200 ([Illume-keyboard] The numeric keyboard should only contain numbers) updated <http://trac.shr-project.org/trac/ticket/200#comment:6>
May 05 07:38:13 * jc IMEI or S/N, what would you like to provide for GSM.Buzz - Quality Enhancement?
May 05 07:49:18 <alexxy> Ainulindale: mrmoku|away : ping 
May 05 07:49:38 <alexxy> Ainulindale: mrmoku|away :  what the state of rebuilding shr-unstable?
May 05 09:11:15 <Mirv> does anyone happen to know where/how that Om distro's "lock screen" aux button handling was done?
May 05 09:11:48 <Mirv> I'd like to have something similar on Debian, with the addition of disabling touch while locked, just wondering on which level it'd be good to do
May 05 09:12:01 <Mirv> or if someone has implemented such thing already, I'd like to know
May 05 09:15:07 <mrmoku> alexxy: it stopped at dbus-hlid :( looks like it has some problem with the newer vala
May 05 09:15:26 <mrmoku> Ainulindale: didn't ptitjes say he replied?
May 05 09:16:48 <TAsn> Mirv, the key was bound in illume
May 05 09:17:01 <TAsn> cya.
May 05 09:18:19 <Mirv> TAsn: ok.
May 05 09:18:34 <Mirv> I think I'll map the 177 to some key and then use matchbox (which I use) to launch... something
May 05 09:18:51 <Mirv> is there a neat way to disable/enable touch events?
May 05 09:21:56 <TAsn> ok, I came back for a sec
May 05 09:22:05 <TAsn> doing what you suggested is a good idea
May 05 09:22:08 <TAsn> and as for touch events
May 05 09:22:40 <TAsn> one way is to write a fullscreen gtk app that will act as the screen saver, and therefore will get all the touch events
May 05 09:22:45 <Heinervdm> mrmoku: so we have to wait again 10 h for new images?
May 05 09:22:48 <TAsn> and that is a blocker.
May 05 09:22:54 <TAsn> cya, now for good.
May 05 09:23:00 <Mirv> TAsn: good idea, thanks again
May 05 09:23:06 <TAsn> sure thing.
May 05 09:30:03 <mrmoku> Heinervdm: depends on whoever fixes dbus-hlid
May 05 10:10:50 <mrmoku> mickey|bbq: ping
May 05 10:11:07 <mrmoku> mickey|bbq: dbus-hlid does not build with vala 0.7.1... possible?
May 05 10:12:42 <mrmoku> /home/shr/shr-unstable/tmp/work/armv4t-angstrom-linux-gnueabi/dbus-hlid-0.9.0+gitr29+65207a9f64ce33e7ba79cd88b55cc4ca9df05430-r0/git/src/obj.c:957: error: too many arguments to function 'server_ListBusNames'
May 05 10:15:52 <mrmoku> any vala expert around who could take a look?
May 05 10:43:48 <CIA-75> freesmartphone.org: 03mickey 07framework * rdb3d3ccfa4d7 10/ (ChangeLog framework/patterns/loophole.py): 
May 05 10:43:48 <CIA-75> freesmartphone.org: patterns: add loophole object to prepare for post-mortem/live interactive debugging
May 05 10:43:48 <CIA-75> freesmartphone.org: next step: hand out references to interesting objects to network interpreter
May 05 10:48:14 <mickey|bbq> mrmoku|away: you need to take that up with ptitjes, the dbus-experimental stuff is uhm... experimental
May 05 10:48:35 <mickey|bbq> it compiles fine with the master branch
May 05 10:51:41 <mrmoku> mickey|bbq: ahh, ok :(
May 05 10:52:03 <mickey|bbq> i can't get anything dbus to compile with his latest branch
May 05 10:52:27 <mrmoku> grr... and he isn't even here to let me spank him until he fixes it :P
May 05 10:52:28 <mickey|bbq> better go back to mickey/0.7/posix or an earlier revision for now
May 05 10:53:03 <emmentaler> hello mickey
May 05 10:53:04 <mrmoku> ok, will try mickey/0.7/posix then
May 05 11:01:16 <Zorkman> ~_~ hi bzzbot
May 05 11:01:26 <Zorkman> ~seen Dave
May 05 11:01:26 <bzzbot> dave <n=dave@pool-138-89-1-87.mad.east.verizon.net> was last seen on IRC in channel #openmoko-cdevel, 18d 21h 40m 31s ago, saying: 'dos1, I thought opimd was just a legend!? :o'.
May 05 11:03:01 <max_posedon> mickeyl, hi! could you say few words about gsm0710muxd status?
May 05 11:03:23 <max_posedon> few days ago, all gentoo users get problem with deleting -9.3.3. sources from server)
May 05 11:04:43 <max_posedon> so, 1. could you return it back? 2. what version users should use now? (I maintain gentoo openmoko)
May 05 11:07:46 <mickeyl> ?
May 05 11:08:01 <mickeyl> oh
May 05 11:08:07 <mickeyl> yes, i can bring it back
May 05 11:08:45 <max_posedon> and if it outdated, what about 2nd question?
May 05 11:09:05 <mickeyl> until further notice, gsm0710muxd is the thing to use
May 05 11:09:20 <mickeyl> the successor will not be introduced before next milestone
May 05 11:09:37 <mickeyl> it's alive and kicking, but we don't want to introduce too many variables
May 05 11:09:40 <max_posedon> hm... gentoo... we like test upstream too)
May 05 11:09:56 <mickeyl> well
May 05 11:09:57 <mickeyl> if you want
May 05 11:10:00 <mickeyl> use fso-abyss
May 05 11:10:05 <mickeyl> it rocks
May 05 11:10:07 <mickeyl> :)
May 05 11:10:18 <max_posedon> hm.. master? or some specific commit?
May 05 11:10:27 <mickeyl> master should be fine
May 05 11:12:02 <mickeyl> it needs libgsm0710, libgsm0710mux, libfsotransport
May 05 11:12:03 <mickeyl> so you will need to provide recipes for these as well
May 05 11:12:04 <Ainulindale> mickeyl!
May 05 11:12:04 <mickeyl> although i did source releases iirc
May 05 11:12:04 <Ainulindale> Activity required! Activity required!
May 05 11:12:05 <mickeyl> no chance
May 05 11:12:05 <Ainulindale> Please post on shr-devel at once or the universe will implode! Cucumber error!
May 05 11:12:05 <mickeyl> stuck in deep problem
May 05 11:12:05 <Ainulindale> Damn :-)
May 05 11:12:05 <Ainulindale> How's life apart from that? :-)
May 05 11:12:05 <alphaone> Ah, get a new universe
May 05 11:12:06 <Ainulindale> alphaone: can I borrow yours?
May 05 11:12:15 <alphaone> Ainulindale: Sure, just don't drop it
May 05 11:12:16 <mickeyl> other that these bugs, i'm feeling well, thanks
May 05 11:13:04 <raster> mickeyl: boo
May 05 11:16:07 <max_posedon> mickeyl, libgsm0710, libgsm0710mux, libfsotransport how they depends on each other? or from system companents. or I can find this info in sources?
May 05 11:16:07 <mickeyl> yo raster
May 05 11:16:18 <mickeyl> max_posedon: see configure.ac
May 05 11:16:36 <max_posedon> thanks
May 05 11:16:36 <mickeyl> max_posedon: libgsm0710mux depends on libgsm0710mux
May 05 11:17:09 <max_posedon> sorry, latest message... strange)
May 05 11:17:18 <raster> mickeyl: just curious. what is your opinion of android on a non-political level?
May 05 11:18:17 <mickeyl> raster: never had much time to look into it. what i do know is that the application framework services are pretty well designed.
May 05 11:18:48 <raster> mickeyl: hmm - havent gotten that far
May 05 11:18:49 <mickeyl> the lower levels and the VM is not my thing
May 05 11:18:52 <raster> but the core os sucks balls
May 05 11:18:55 <mickeyl> yep
May 05 11:19:05 <raster> its a half-pretence at somthing unix-like
May 05 11:19:12 <raster> but fails absymally in most departments
May 05 11:19:13 <mickeyl> yeah. Android/Linux
May 05 11:19:37 <raster> hell the shell's error for "command not found" is actually "find: permission denied"
May 05 11:19:43 <raster> (if u try use find)
May 05 11:20:05 <raster> some bizarre partion/image format instead of a simple yaffs/ubifs/jfs2 or whatever image
May 05 11:20:14 <raster> open isfar from what i would call it
May 05 11:20:28 <max_posedon> oh, btw, shr-devs, what about buildin unifs support? 
May 05 11:20:31 <raster> i spent 2 days screwing ith this htc dream to finally get a shell on it and root
May 05 11:20:52 <raster> my opinion of android has actally gone down rather than up after screwing with it
May 05 11:28:06 <CIA-75> freesmartphone.org: 03mickey 07framework * r973a972b5603 10/framework/ (frameworkd patterns/loophole.py): inject controller object into loophole; from there we should be able to gather all interesting references
May 05 11:28:47 <mickeyl> raster: *nod* pretty depressing given the industry's interest :/
May 05 11:28:53 <mickeyl> mirko-paroli: ping
May 05 11:28:59 <mickeyl> mirko-paroli: next bunch of commits is for you
May 05 11:29:02 <mirko-paroli> mickeyl: pong
May 05 11:29:06 <mirko-paroli> oh
May 05 11:29:07 <mirko-paroli> nice :)
May 05 11:29:12 <mirko-paroli> what am I getting?
May 05 11:29:14 <mirko-paroli> ah, edje?
May 05 11:29:17 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * rf7b0ec1d5feb 10/examples/ (4 files in 2 dirs): examples: add ecore and edje example skeletons
May 05 11:29:18 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * r0442064b5cc8 10/ (6 files in 5 dirs): ecore.vapi: add enough canvas to show a window
May 05 11:29:19 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * r07ae7827c555 10/ (7 files in 4 dirs): examples: bind more stuff to make edje example application work
May 05 11:29:20 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * r423b108ce012 10/ (examples/edje/edje.vala vapi/edje.vapi): example: connecting a signal works
May 05 11:29:22 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * r4f7b5583b53e 10/ (4 files in 2 dirs): edje.vapi: add more callbacks
May 05 11:29:25 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * r451bec04b042 10/ (examples/edje/edje.vala vapi/edje.vapi): edje.vapi: add missing functions
May 05 11:29:28 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * r39992c3f13a7 10/vapi/evas.vapi: evas.vapi: bind more of evas
May 05 11:29:37 <mirko-paroli> mickeyl: yay!!!
May 05 11:29:38 <mirko-paroli> thx
May 05 11:29:53 <mickeyl> np, had to distract myself while being buried in some bug
May 05 11:30:01 <mirko-paroli> now I only need to get vala to build and I am set to go ;)
May 05 11:30:21 <raster> mickeyl: yeah. to be honest i would have been much more accepting of it if they did a qtopia-like thing where they uses a regular unix/linux baseos and just redid the ui and app layer
May 05 11:30:23 <Ainulindale> mirko-paroli: there's a patch for fso_action on shr-user
May 05 11:30:33 <raster> the rest of the changs were just totally unnecessary
May 05 11:30:39 <mickeyl> right
May 05 11:30:55 <raster> odd too
May 05 11:30:55 <Ainulindale> errr
May 05 11:30:57 <raster> they use bluez
May 05 11:31:02 <Ainulindale> s/mirko-paroli/mickeyl/
May 05 11:31:02 <raster> and have dbus-dameon JUST for bluez
May 05 11:31:11 <raster> hen they bridge that over to their own different subsstems
May 05 11:31:30 <mirko-paroli> Ainulindale: I would have ignored you anyway :P
May 05 11:31:42 <mickeyl> Ainulindale: thanks, I'm only subscribed to -devel. will check the archives
May 05 11:32:06 <Ainulindale> http://lists.shr-project.org/pipermail/shr-user/2009-May/000080.html
May 05 11:32:09 <mickeyl> thanks
May 05 11:32:22 <raster> i'm going to see if i can't get a new userspace on android
May 05 11:32:28 <raster> so re-use the kernel work/port and drivers
May 05 11:32:38 <raster> and "fix" userspace to be something that pops out of oe
May 05 11:32:59 <mickeyl> s/android/g1/ ?
May 05 11:33:08 <mickeyl> sounds pretty cool
May 05 11:33:25 <mickeyl> the WM HTCs look like a dead-end :/
May 05 11:33:41 <mickeyl> i don't think the port will ever be finished enough
May 05 11:34:02 <mickeyl> which is sad, given that i shelled out 500 EUR for one of those
May 05 11:34:05 <mickeyl> oh well
May 05 11:34:05 <mrmoku> mickeyl: kernel side? that's sad :(
May 05 11:34:46 <mickeyl> i might be exaggarating, but judging from the level of understanding "we" have of all the hardware, my hopes are very low
May 05 11:34:53 <raster> mickeyl: well as thats the only available android thing atm... yes
May 05 11:34:54 <raster> :)
May 05 11:35:17 <raster> to me it seems like the fastest track to get an open os
May 05 11:35:28 <raster> though rooting/hacking the g1 is a royal pain
May 05 11:35:28 <mickeyl> you have the dev-g1 or the real one?
May 05 11:35:32 <mickeyl> ah, the real one, then
May 05 11:35:39 <raster> it took me 2 days and i've had to use non-free software/services along the way
May 05 11:35:46 <mickeyl> fun...
May 05 11:35:51 <raster> well i actually got the singapore/au htc dream
May 05 11:35:52 <mickeyl> not
May 05 11:35:55 <raster> different os build to us/uk
May 05 11:35:59 <raster> even to europe
May 05 11:36:08 <raster> it was totally locked down
May 05 11:36:37 <raster> needed to use a closedapp to generate a magic sd-card (goldcard) image that put a specially generated hader ont he bootblock based off he sdcard serial id
May 05 11:36:57 * mickeyl shivers
May 05 11:36:58 <raster> and that would then make the bootloader alow a reflash of an older version .hdb image
May 05 11:37:12 <raster> (which as best i know no one knows how to make one and the format is not fuly known)
May 05 11:37:21 <CIA-75> SHR: 03mok 07shr-overlay * r45d5e54d4365 10/openembedded/conf/distro/include/shr-autorev-unstable.inc: autorev: switch vala-native to mickey/0.7/posix branch
May 05 11:37:23 <raster> so u find a "buggy"older hdb (rc29 or rc7)
May 05 11:37:28 <raster> use the magic sd to flash
May 05 11:37:31 <CIA-75> SHR: 03mok 07shr-overlay * re31fa1916487 10/openembedded/recipes/intone/intone_svn.bb: intone: add missing RDEPENDS
May 05 11:37:42 <raster> now u have a buggy adroid with the root-console-listening-to-keystrokes bug
May 05 11:38:05 <raster> and then you can begin to do things... one you learn enough about android's userspace... which is totally non-linux-like
May 05 11:38:22 <raster> i suspect may people would not do this or have the patience
May 05 11:38:33 <raster> some say the iphone is easier to jailbreak/hack than android
May 05 11:38:42 <raster> even with the console/root bug its more work
May 05 11:39:52 <raster> anyway - i think there is a good chance of a decent userspace. my next goal is to make a custom moot image as the rootfs and init and initrc files are on a read-only image so u need to re-gnerate it
May 05 11:40:21 <raster> if i can nicely replace init... then i can begin to banch off into a better userspace inside some chroot
May 05 11:42:39 <mickeyl> sounds cool, but...
May 05 11:43:06 <mickeyl> ... i'm afraid that'll keep you busy for some months with stuff where you're not exactly an expert in. sounds a bit like a wasted talent to me
May 05 11:43:47 <CIA-75> freesmartphone.org: 03mok 07libframeworkd-glib * r78454ef2ea30 10/src/opimd/frameworkd-glib-opimd-contacts.c: 
May 05 11:43:47 <CIA-75> freesmartphone.org: opimd: fix ContactQuery.GetMultipleResults
May 05 11:43:47 <CIA-75> freesmartphone.org: Signed-off-by: Klaus Kurzmann <mok@fluxnetz.de>
May 05 11:43:48 <CIA-75> freesmartphone.org: 03mok 07libframeworkd-glib * rfbaa066826bf 10/src/opimd/ (2 files): 
May 05 11:43:48 <CIA-75> freesmartphone.org: opimd: implement MessageQuery.GetMultipleResults
May 05 11:43:50 <CIA-75> freesmartphone.org: Signed-off-by: Klaus Kurzmann <mok@fluxnetz.de>
May 05 11:44:39 <CIA-75> freesmartphone.org: 03mickey 07framework * r9fb573a66540 10/framework/patterns/loophole.py: loophole: deal with varying line endings
May 05 11:45:10 <mickeyl> but then again... if no one else is interested in doing that... who else but you :)
May 05 11:45:25 <mickeyl> by that logic i might start writing kernel drivers for the HTC Touch Pro... oh well
May 05 11:45:29 * mickeyl sighs
May 05 11:46:33 <raster> hahaha
May 05 11:46:35 <raster> true
May 05 11:46:48 <raster> but... show me a 3g phone i can run real linux on easily...
May 05 11:46:55 <raster> right now - i have none of those
May 05 11:47:02 <raster> this is as close as i can find
May 05 11:47:13 <max_posedon> HTC Wizard?
May 05 11:47:41 <max_posedon> as I know few people work on poring fso to it.
May 05 11:47:56 <raster> max_posedon: 3g.
May 05 11:48:01 <max_posedon> ok, it isn't 3g, just cdma
May 05 11:48:13 <raster> cdma is also useless as i live in a gsm world
May 05 11:48:23 <raster> but it is gsm up to edge
May 05 11:48:27 <raster> also a tad ugly
May 05 11:48:48 <raster> i also washoping for reasonable speed cpu and decent ram
May 05 11:49:47 <raster> g1 is probably as good as it gets - kernel at least is already ported
May 05 11:49:54 <raster> drivers are done
May 05 11:50:05 <raster> (from the kernel side)
May 05 11:50:09 <mrmoku> Ainulindale: if ptitjes pops up... tell him to fix his crap
May 05 11:50:32 <mrmoku> Ainulindale: and that I switched vala to mickey/0.7/posix branch (because his crap does not work) :P
May 05 11:50:42 * mrmoku has to go now
May 05 12:04:21 <Ainulindale> which crap?
May 05 12:06:27 <mickeyl> not crap
May 05 12:06:33 <mickeyl> exciting stuff actually
May 05 12:06:40 <mickeyl> though a bit buggy yet
May 05 12:06:49 <mickeyl> (dbus error marshalling)
May 05 12:07:54 <devilshawk> hello! any SHR devs around? which package provides libecore.so in unstable now? it seems like there is no ecore package but only a ecore-dbg and ecore-dev pkg in the feed?
May 05 12:48:28 <Ainulindale> mickeyl: will you be able to reply today? :-)
May 05 12:48:34 <Ainulindale> dos1: I'd like your opinion on my dbusification mail
May 05 12:54:59 <Ainulindale> dos1: I'd like your opinion on my dbusification mail (was I saying)
May 05 13:04:37 <dos1> Ainulindale: i agree with everything, i think we should start to work ;>
May 05 13:04:55 <dos1> to define methods and signals
May 05 13:05:20 <Ainulindale> well don't you have any suggestion?
May 05 13:05:23 <Mirv> which would be the easiest method not to brighten up display when the screen is touched?
May 05 13:06:01 <Mirv> can an application supress it (if the application receives the events) or does it need to be set somewhere else?
May 05 13:12:28 <dos1> Ainulindale: at the moment no, idea is very good i think. but of course i will have suggestions about methods and signals ;)
May 05 13:13:21 <Ainulindale> then please draft what you'd like
May 05 13:15:01 <Ainulindale> dos1: I'm interested in reactions, ideas, even if they're stupid
May 05 13:15:08 <Ainulindale> It might help us
May 05 13:16:11 <dos1> i would like listing available phoneguis like backends in opimd
May 05 13:16:20 <Ainulindale> that's what I was planning to do
May 05 13:16:25 <Ainulindale> even told that in my mail
May 05 13:22:45 <Heinervdm> Is it possible to write a program that pops up like the illume keyboard within illume?
May 05 13:22:58 <raster> yes
May 05 13:23:03 <raster> oh wait
May 05 13:23:07 <raster> u dont want to replace the kbd?
May 05 13:23:12 <Heinervdm> no
May 05 13:23:15 <raster> no
May 05 13:23:22 <raster> illume has no layout policy for that
May 05 13:23:24 <raster> just for the kbd
May 05 13:23:33 <raster> i intend to add it at some point
May 05 13:23:58 <Heinervdm> i tought that this were a good thing for the notifier app
May 05 13:24:30 <Heinervdm> s/were/would be/
May 05 13:24:33 <bzzbot> Heinervdm meant: i tought that this would be a good thing for the notifier app
May 05 13:24:35 <raster> yes
May 05 13:24:41 <raster> but it doesnt have such a thing currently
May 05 13:24:46 <Heinervdm> ok
May 05 13:24:59 <dos1> Heinervdm: i have the same idea some time ago ;)
May 05 13:25:08 <Heinervdm> dos1: ;)
May 05 13:26:00 <Heinervdm> dos1: did you do sth to the messages domain?
May 05 13:27:24 <dos1> no...
May 05 13:28:04 <Heinervdm> isn't it just copy&paste? then i will do it
May 05 13:29:00 <shoragan> Ainulindale, where is the new shr devel list?
May 05 13:29:10 <Ainulindale> lists.shr-project.org
May 05 13:29:23 <Ainulindale> Ah and you'd be interested in the talk too shoragan :-)
May 05 13:29:39 <shoragan> there is only -user
May 05 13:31:33 <max_posedon> Ainulindale, whome I should ask to get ubifs support in default shr kernel?
May 05 13:31:45 <max_posedon> will be nice, if it will be `in release`
May 05 13:31:52 <Ainulindale> hmmm I guess mwester should be able enough :-)
May 05 13:32:00 <Ainulindale> shoragan: nah
May 05 13:32:11 <Ainulindale> shoragan: http://lists.shr-project.org/mailman/listinfo/shr-devel
May 05 13:32:16 <max_posedon> mwester, your opinion?
May 05 13:32:47 <shoragan> ah, i thought with 'you'd be interested in the talk too shoragan' you meant a different list
May 05 13:32:59 <Ainulindale> nah :-)
May 05 13:33:10 <Ainulindale> http://lists.shr-project.org/pipermail/shr-devel/2009-May/000004.html
May 05 13:33:13 <Ainulindale> this one specifically
May 05 13:33:57 <dos1> Heinervdm: there should be some changes i think
May 05 13:34:28 <dos1> Heinervdm: today i want to work on general backends support (configuration via dbus)
May 05 13:34:48 <Heinervdm> ok
May 05 13:34:52 <dos1> Heinervdm: after that merging contacts, updating with new fields and then messages :)
May 05 13:35:53 <shoragan> dos1, did you test opimd with your changes?
May 05 13:36:08 <dos1> shoragan: yes, why are you asking?
May 05 13:36:11 <shoragan> all your dbus calls seem to be synchronous
May 05 13:36:52 <shoragan> or do you access the ogsmd objects directly?
May 05 13:37:18 <alexxy> Ainulindale: mrmoku|away : whats curent status of rebuilding??
May 05 13:37:43 <dos1> shoragan: before my changes there was synchronous calls too
May 05 13:37:59 <CIA-75> freesmartphone.org: 03mickey 07framework * rf0a0e72952b0 10/framework/resource.py: resource.py: get rid of callback factory and rewrite for improved clarity. This does NOT fix #416 though.
May 05 13:38:25 <shoragan> hmm, it shouldn't work this way
May 05 13:38:31 <dos1> shoragan: i will change it to async after getting everything to work
May 05 13:38:32 <shoragan> i'll need to take a deeper look
May 05 13:38:52 <mickeyl> shoragan: i just commented on #416, it looks like a dbus problem to me atm.
May 05 13:38:56 <mickeyl> will need to add some debugging to the daemon
May 05 13:39:49 <shoragan> mickeyl, what is you dbus conf to see the replies?
May 05 13:40:00 <mickeyl> i can't see replies
May 05 13:40:04 <shoragan> oh
May 05 13:40:06 <shoragan> :/
May 05 13:40:23 <mickeyl> the dbus guys know it's broken, but they don't know and/or don't care to fix it
May 05 13:40:41 <shoragan> oh
May 05 13:40:48 <alphaone> that sucks
May 05 13:40:58 <shoragan> where did you talk to them?
May 05 13:41:03 <mickeyl> i didn't
May 05 13:41:07 <mickeyl> check some mails
May 05 13:41:09 <mickeyl> lindi did
May 05 13:41:12 <mickeyl> months ago
May 05 13:41:12 <alphaone> Which means bustle wont work
May 05 13:41:24 <alphaone> Not for profiling anyway
May 05 13:41:44 <shoragan> i'll be back later
May 05 13:41:45 <mickeyl> well, not saying that we can't fix it... but it's going to be a dbus hacking session
May 05 13:41:59 <alphaone> mickeyl: When does your ICE leave? :-)
May 05 13:42:38 <mickeyl> hehe
May 05 13:42:56 <mickeyl> lets first see where and how dbus is swalloing our methods
May 05 13:43:04 <mickeyl> lets take care about the replies later
May 05 13:43:13 <alphaone> s/where/if and/
May 05 13:43:18 <alphaone> where
May 05 13:43:28 <mickeyl> right
May 05 13:43:35 <mickeyl> although i'm really suspecting dbus
May 05 13:43:37 <mickeyl> our code looks good
May 05 13:43:51 <alphaone> But it always only happens in that usecase
May 05 13:43:56 <mickeyl> yes
May 05 13:44:01 <alphaone> So what does dbus care about that?
May 05 13:44:02 <mickeyl> that may hint that we hit a limit
May 05 13:44:11 <mickeyl> like i've written in #416
May 05 13:44:27 <alphaone> looking
May 05 13:44:31 <mickeyl> the now synchronized asyncqueue means we have more outstanding calls than before
May 05 13:44:42 <mickeyl> which could be the reason why we didn't see this happening before
May 05 13:46:00 <tmzt> there's nothing really special about a dbus reply right? it's just a message in the opposite direction?
May 05 13:46:53 <alphaone> oh!
May 05 13:46:54 <mickeyl> no
May 05 13:46:55 <alphaone> wait
May 05 13:47:11 <mickeyl> calls and replies have sequence numbers
May 05 13:47:28 <mickeyl> yes?
May 05 13:47:31 <alphaone> that could be it
May 05 13:47:47 <alphaone> frameworkd all is one user to dbus, right
May 05 13:47:51 <mickeyl> correct
May 05 13:47:54 <mickeyl> one process
May 05 13:47:55 <alphaone> And many many things happen at that time
May 05 13:47:59 <mickeyl> correct
May 05 13:48:07 <mickeyl> lots of outstanding calls
May 05 13:48:17 <alphaone> signals are sent about idle state, display brightness called, ...
May 05 13:48:22 <alphaone> Can you see how many?
May 05 13:48:28 <mickeyl> i wouldn't know how
May 05 13:48:30 <alphaone> DBus has limits for that
May 05 13:48:38 <alphaone> gimme a minute
May 05 13:48:41 <mickeyl> oki
May 05 13:48:47 <mickeyl> i know about the limit directive
May 05 13:48:51 <mickeyl> but it's not used in our config
May 05 13:49:01 <mickeyl> so i guessed it's unlimited
May 05 13:49:08 <tmzt> I mean for security purposes, the configuration has to allow the user to send the reply to the path? but I guess that's not the issue here
May 05 13:49:10 <mickeyl> although from the timeouts incident we should know better
May 05 13:49:37 <playya> to the git pros: i created a local branch and want to push it to git.fso. this is what happens: http://pastebin.ca/1413252
May 05 13:49:48 <mickeyl> tmzt: yes, usually eavesdrop can be enabled, it must be a problem in dbus, since the calls are shown nicely
May 05 13:50:02 <playya> any ideas how to get the code pushed?
May 05 13:50:23 <alphaone> I think there are defaults
May 05 13:50:42 <mickeyl> playya: try git push origin <yourbranch>:<remotebranch>
May 05 13:51:13 <playya> Everything up-to-date
May 05 13:51:16 <mickeyl> doesn't look like a local problem though
May 05 13:51:40 <alphaone> http://www.nabble.com/upping-default-limits-td8654329.html
May 05 13:51:56 <playya> hrhr. CIA-75 only looks on master ?
May 05 13:52:08 <mickeyl> playya: no, it should show everything
May 05 13:52:13 <playya> it seems that some push already worked
May 05 13:52:17 <mickeyl> the revs are not online
May 05 13:52:25 <mickeyl> i can't see them in gitweb
May 05 13:52:27 <mickeyl> can you?
May 05 13:52:34 <playya> http://git.freesmartphone.org/?p=fso-monitord.git;a=shortlog;h=refs/heads/testing/dbus-hlid
May 05 13:53:08 <mickeyl> hmm?
May 05 13:53:10 <mickeyl> what now
May 05 13:53:16 <mickeyl> fso-monitor or dbus-hlid?
May 05 13:53:43 <playya> this branch is for testing fso-monitord using dbus-hlid
May 05 13:53:50 <mickeyl> ah
May 05 13:54:11 <mickeyl> ok, revs are there
May 05 13:54:15 <mickeyl> although we have two
May 05 13:54:20 <mickeyl> testing/dbus-hlid and dbus-hlid
May 05 13:54:25 <alphaone> http://cgit.freedesktop.org/dbus/dbus/tree/bus/config-parser.c#n403
May 05 13:54:26 <mickeyl> are these the same?
May 05 13:54:35 <playya> i didn't want to push this to master
May 05 13:54:39 <mickeyl> alphaone: uh oh!
May 05 13:54:49 <playya> yes. i should delete dbus-hlid
May 05 13:54:54 <alphaone> Could we hit max_incomplete_connections?
May 05 13:55:07 <mickeyl> alphaone: yes, that's what i would think
May 05 13:55:14 <alphaone> I'll try
May 05 13:55:19 <mickeyl> cool
May 05 13:56:22 <playya> the problem with dbus-hlid is that i don't know the daemon behind the service
May 05 13:56:40 <playya> so I won't be able to restart it 
May 05 13:56:51 <mickeyl> you can gather the pid
May 05 13:57:00 <mickeyl> for any busname iirc
May 05 13:57:28 <playya> i saw a get user or sth similar
May 05 13:57:32 <alphaone> hmm, actually that seems to be the number of dbus connections
May 05 13:57:56 <alphaone> does oeventsd always try to open a new connection?
May 05 13:58:01 <mickeyl> yes
May 05 13:58:04 <mickeyl> oooh
May 05 13:58:07 <mickeyl> YES!
May 05 13:58:13 <mickeyl> maybe they leak
May 05 13:58:17 <mickeyl> look here this comment:
May 05 13:58:29 <mickeyl>     #FIXME: Add Cache for dbus objects
May 05 13:58:30 <mickeyl> :D
May 05 13:58:33 <alphaone> HAH
May 05 13:58:38 <alphaone> okay, I'll try
May 05 13:58:42 <alphaone> then laugh
May 05 13:59:06 <mickeyl> this slows down operation btw.
May 05 13:59:12 <mickeyl> we need to do this cache asap
May 05 13:59:37 <Ainulindale> DIE oeventsd DIE
May 05 13:59:38 <alphaone> heh
May 05 13:59:43 <alphaone> Where is it?
May 05 13:59:49 <mickeyl> actions.py
May 05 13:59:53 <alphaone> Oh, why did I even ask....
May 05 13:59:59 <alphaone> so it's in oeventsd
May 05 14:00:01 <Mirv> well, I ended up setting brightness to 0 manually. now I have pretty much funtional AUX -> lock screen on Debian, which is what I sought for :)
May 05 14:00:36 <alphaone> Mirv: Then you sill accept touchscreen clicks though...
May 05 14:02:09 <alphaone> okay, it worked this time
May 05 14:02:16 <mickeyl> !
May 05 14:02:22 <alphaone> Trying if the old limit also causes problems
May 05 14:02:29 <alphaone> Since I have a differnt image now
May 05 14:04:47 <alphaone> mickeyl: can you test?
May 05 14:05:02 <mickeyl> what did you do now?
May 05 14:05:58 <mickeyl> playya: 
May 05 14:06:05 <mickeyl> dbus will tell you
May 05 14:06:07 <mickeyl> mickey@andromeda:~$ mdbus -s org.freedesktop.DBus / org.freedesktop.DBus.GetConnectionUnixProcessID org.freesmartphone.DBus
May 05 14:06:07 <mickeyl> 22390
May 05 14:06:21 <alphaone> mickeyl: I flashed shr last week
May 05 14:06:25 <playya> hmm. ok
May 05 14:06:39 <alphaone> And not sure this is a valid test setup for our problem
May 05 14:07:04 <mickeyl> ok, i'll try something different first
May 05 14:07:08 <mickeyl> i'll leave the limit
May 05 14:07:13 <mickeyl> but add a cahce
May 05 14:07:21 <mickeyl> lets see whether this changes anything
May 05 14:07:40 <alphaone> Okay, cache is good anyway
May 05 14:07:44 <mickeyl> *nod*
May 05 14:07:51 <alphaone> We should test the other thing as well, though
May 05 14:08:02 <alphaone> To make sure we understand the problem
May 05 14:09:07 <alphaone> oeventsd is insane at places
May 05 14:11:58 <Mirv> alphaone: that's not a problem since what I did was a fullscreen GTK application that absorbs all the touches ()http://users.tkk.fi/~tajyrink/moko/neolockscreen.png
May 05 14:12:15 <alphaone> ah, okay then
May 05 14:12:21 <Mirv> and a script that either launches it or kills it
May 05 14:12:42 <mickeyl> cache done. testing...
May 05 14:12:46 <alphaone> okay
May 05 14:17:33 <Azog> alphaone: was the weekend very stressful fixing all those freerunners? 
May 05 14:18:56 <alphaone> Azog: It wasn't particularly relaxing :-)
May 05 14:19:01 <alphaone> But was fun
May 05 14:19:53 <Azog> too bad, our freerunners didn't arrive in time (but we packed a few candies in the package)
May 05 14:19:54 <DocScrutinizer> how may broken devices ? ;-)
May 05 14:20:49 <mickeyl> hmm, no that didn't change anything
May 05 14:21:06 <mickeyl> one more test, then i play with the hardcoded limits
May 05 14:21:25 <alphaone> Just tried max_incomplete_connections
May 05 14:21:27 <alphaone> didn't work
May 05 14:21:50 <DocScrutinizer> hehe - pushing limits :-)
May 05 14:21:55 <mickeyl> so we need to add more debug to the dbus daemon
May 05 14:22:29 <sylar91> hi
May 05 14:22:48 <DocScrutinizer> mickeyl: flood it with gdb breakpoints ;D
May 05 14:22:56 <alphaone> trying max_connections_per_user
May 05 14:23:14 <sylar91> anyone has successfully run omnewrotate with backlight support on shr? :/
May 05 14:23:48 <DocScrutinizer> hmm seems we did this communication before
May 05 14:23:59 <DocScrutinizer> what is "backlite support"?
May 05 14:25:11 <sylar91> backlight ;) it is an omnewritate mode by which while rotating the display backlight is turned off
May 05 14:25:37 <DocScrutinizer> oh, very sophisticated feature
May 05 14:25:51 <alphaone> mickeyl: nope...
May 05 14:26:35 <mickeyl> fun
May 05 14:26:41 <alphaone> nope
May 05 14:26:46 <DocScrutinizer> lol
May 05 14:27:37 <sylar91> :/ it is only nice because it hides xrandr 'bad'(from the gui pov)  effect
May 05 14:29:55 <DocScrutinizer> sylar91: reducing light intensity to 0 is somewhat tricky for LCM. Might lock up in hw
May 05 14:30:51 <DocScrutinizer> the effect would be light doesn't come up again. Dunno what's your " :-/ " tho
May 05 14:31:01 <sylar91> it makes the same thing shr makes with autodimming eheh
May 05 14:36:14 <mickeyl> interesting
May 05 14:36:19 <mickeyl> with the cache it fails a bit later
May 05 14:36:30 <mickeyl> this really looks like one of the limits
May 05 14:37:25 <alphaone> mickeyl: I better reboot
May 05 14:37:38 <alphaone> Don't know whether SIGHUP works for the limits...
May 05 14:38:08 <alphaone> and restarting the dbus system daemon is....well, interesting
May 05 14:38:14 <alphaone> It fails later?
May 05 14:38:17 <alphaone> how later?
May 05 14:38:31 <alphaone> What do you see more compared to before?
May 05 14:39:29 <mickeyl> the ReleaseResource call gets processed
May 05 14:39:37 <SHR-WugBranler`> tracfeed: Ticket #424 (Boot/Reboot dialogue only 1 pixel wide) updated <http://trac.shr-project.org/trac/ticket/424#comment:1>
May 05 14:39:41 <mickeyl> the answer doesn't make it through though
May 05 14:40:00 <mickeyl> the SetScenario() then gets lost
May 05 14:49:12 <Zorkman> DocScrutinizer: any news on the clue for #1024 (yeah i'm just buggin you ;-))
May 05 15:00:20 <DocScrutinizer> Zorkman: we know how to stop #1024 in one particular device with one particular nasty ugly hw fix. We neither know *why* it works, nor *if* it works for other devices or if we can find a not so ugly way to do it
May 05 15:01:14 <Zorkman> thanks; is the current implementation of the nasty and ugly fix harder to do than the hw buzz fix?
May 05 15:01:15 <SHR-WugBranler`> tracfeed: Ticket #185 (Problems mounting uSD card) updated <http://trac.shr-project.org/trac/ticket/185#comment:9>
May 05 15:01:44 <DocScrutinizer> yup, buzzfix is a snap compared to that
May 05 15:02:01 <Zorkman> ah well, we can't have it all :)
May 05 15:03:59 <DocScrutinizer> I emphasize on we don't even know *why* it works for now. So stay patient... :-)
May 05 15:04:10 <Zorkman> i will :)
May 05 15:04:17 <Zorkman> i'm reallyu gratefull for the updates you're giving here
May 05 15:04:47 <Zorkman> we won't hold it against anyone if it won't work in the end
May 05 15:05:19 <DocScrutinizer> night be another layout issue, analog to uSD+GPS problem
May 05 15:05:33 <DocScrutinizer> might even
May 05 15:06:04 <DocScrutinizer> might be reflow flaws issue
May 05 15:06:18 <DocScrutinizer> we have no good story so far
May 05 15:06:51 <Weiss> DocScrutinizer: anecdotal evidence which might help: i was affected by #1024 on my first FR, but my second FR (after a replacement due to a faulty WLAN module) doesn't seem to have it
May 05 15:07:33 <Weiss> same SIM, network and location (but slightly more recent software)
May 05 15:07:54 <DocScrutinizer> that's one of the points - there have to be different classes of FR, and we don't know the diffs yet
May 05 15:08:52 <DocScrutinizer> that's why I prefer the reflow theory. alas we couldn't find a proof yet
May 05 15:10:13 <DocScrutinizer> and in the end we're possibly talking about a few dozen of reworks actually done if ever we find out how to do a decent rework
May 05 15:11:02 <DocScrutinizer> as rework might be nasty or nearly impossible to implement
May 05 15:11:24 <DocScrutinizer> and no funding on all this research... :-/
May 05 15:11:59 <Zorkman> DocScrutinizer: sorry, had to reboot, are you still talkinga bout 1024? is there no funding?
May 05 15:12:10 <DocScrutinizer> yup
May 05 15:12:34 <Zorkman> ?? I thought it was OM's main priority to get it usable as a phone and i thought that #1024 was also mentionned
May 05 15:13:05 <Weiss> Zorkman: http://pastie.org/468782 (the bit you missed)
May 05 15:13:14 <DocScrutinizer> I'm no regular employee of OM anymore, and Dieter's expense limit is exhausted long ago
May 05 15:13:47 <Zorkman> I thought that Dieter had enough funds (not really sure why i thought that...)
May 05 15:14:02 <Zorkman> thanks Weiss
May 05 15:14:06 <DocScrutinizer> Zorkman: there's very few actual complaints wrt #1024
May 05 15:14:59 <Weiss> the deep sleep prevention stuff works around it sufficiently?
May 05 15:15:16 <DocScrutinizer> sleep=2 works
May 05 15:15:32 <Zorkman> I had it... but now with that deep  sleep prevention i don't get it anymore, but the FR burns more power in suspend....
May 05 15:15:42 <DocScrutinizer> reliably stops all #1024 
May 05 15:15:52 <Weiss> Zorkman: enough power drain to actually notice, though?
May 05 15:16:07 <DocScrutinizer> Zorkman: that's correct
May 05 15:16:11 <Zorkman> DocScrutinizer will comment on that one
May 05 15:16:22 <Zorkman> isn't it like 20-30mA more at least?
May 05 15:17:05 <alphaone> mickeyl: I'm out of ideas again...
May 05 15:17:15 <alphaone> Tested the limits that looked sort of low.
May 05 15:17:37 <DocScrutinizer> from my tests it seems like ~8mA for sleep2 vs ~14mA for sleep2, w/o recamping issues
May 05 15:17:46 <mickeyl> k
May 05 15:17:51 <Zorkman> ah it's less than i thought
May 05 15:17:54 <DocScrutinizer> that's for modem only
May 05 15:18:27 <DocScrutinizer> add a few mA for system in suspend on both values
May 05 15:18:28 <alphaone> DocScrutinizer: Mind testing how much it is for the whole system suspended with sleep=2 and sleep=4?
May 05 15:18:49 <DocScrutinizer> occasionally I'll do
May 05 15:19:06 <alphaone> So we'd have about 1.5 times the standby time we do now?
May 05 15:19:45 <alphaone> That would already be pretty sweet..
May 05 15:19:51 <DocScrutinizer> I'd guess we see more sth like 1.2 
May 05 15:20:25 <alphaone> hmm, k
May 05 15:21:53 <Zorkman> still a nice improvement
May 05 15:22:21 <alphaone> true
May 05 15:22:35 <alphaone> The question is is it worth the effort?
May 05 15:22:46 <alphaone> Was it worth the effort we already had?
May 05 15:23:08 <DocScrutinizer> btw this test is pretty easy, as bat discharges for modem only, when hoocked up to usb and charging actually stops. whole system running from usb, modem running from bat (=coulomb counter)
May 05 15:23:34 <alphaone> ah, nice
May 05 15:23:51 <alphaone> So that's why charging restarts periodically
May 05 15:24:05 <DocScrutinizer> easily done by everyone, with sHR
May 05 15:24:17 <DocScrutinizer> exactly
May 05 15:24:19 <alphaone> true
May 05 15:25:27 <DocScrutinizer> charging should start automatically (by PMU bat management) on a bat vltage equiv ~80%
May 05 15:25:39 <DocScrutinizer> maybe 85%
May 05 15:25:54 <DocScrutinizer> or 75
May 05 15:25:58 <DocScrutinizer> depends
May 05 15:26:38 <DocScrutinizer> but that's exactly what is considered best practice for LiIon bat management
May 05 15:29:10 <DocScrutinizer> we should NOT restart bat charge periodically every few minutes. We're lucky kernel states it's doing exactly this but obviously is "fixed by bug" as charging isn't actually reenabled
May 05 15:31:10 <DocScrutinizer> hehe - "fixed by bug" - I'll register (C) for this term
May 05 15:33:12 <DocScrutinizer> alphaone: I know I still didn't report on all this in #300 :-/ sorry for that, sometimes I hate the formal wording I need to use for ticket comments
May 05 15:50:36 <BluesLee> Sharwin_F: you use xboard?
May 05 15:54:16 <Sharwin_F> Blu3, used to, not at the moment thought :P
May 05 15:54:29 <BluesLee> Sharwin_F: is it fullscreen?
May 05 15:54:57 <BluesLee> Sharwin_F: or just a quarter or so, tried it months ago ...
May 05 15:55:04 <Sharwin_F> BluesLee, don't know if there's an option, there's no need to play it fullscreen
May 05 15:55:30 <mickeyl> alphaone: for the records, reverting resource.py back to the version from ms5.1 does not change the problem
May 05 15:55:38 <mickeyl> i will now revert the idlenotifier back and see what happens then
May 05 15:55:41 <BluesLee> Sharwin_F: the old versions used only the upper left corner afair
May 05 15:55:50 <DocScrutinizer> hmmpf... " like ~8mA for sleep2^H4 vs ~14mA for sleep2,"
May 05 15:55:59 <alphaone> mickeyl: Whoa, okay...
May 05 15:56:09 <alphaone> Fun stuff
May 05 15:56:13 <alphaone> git-bisect
May 05 15:56:39 <BluesLee> how can i change the fact that incoming sms are wrongly dated? time=realtime-2hours ?
May 05 15:56:39 <Sharwin_F> Blu3, oh, but no fullscreen I mean you can see the shelf at the top
May 05 15:57:00 <alphaone> BluesLee: How is that?
May 05 15:57:09 <Sharwin_F> BluesLee, it takes athe full screen down the shelf, it looks good
May 05 15:57:20 <BluesLee> Sharwin_F: thats fine!
May 05 15:57:20 <alphaone> BluesLee: Can you paste the whole content of the timestamp?
May 05 15:58:11 <BluesLee> alphaone: mrmoku said that he would fix it but the latest tetsing broke suspend/resume so that i am back with th eproblem
May 05 15:59:10 <Sharwin_F> BluesLee, sure, it works fine :)
May 05 15:59:11 <BluesLee> alphaone: the timezone for gsm is not localtime
May 05 15:59:35 <BluesLee> where do i set timezone for gsm apps?
May 05 16:05:46 <DocScrutinizer> for the way: anybody any concept how to signal missed calls in a convenient way?
May 05 16:05:52 <DocScrutinizer> bye
May 05 16:07:28 <mirko-paroli> DocScrutinizer: with a number on the homescreen?
May 05 16:07:55 <DocScrutinizer> hehe, for paroli I bet there is a decent plan how to do that
May 05 16:08:16 <mirko-paroli> it is already implemented ;)
May 05 16:08:38 <DocScrutinizer> anyway I wonder how to notice the call I missed while leaving SHR terminal or mokomaze for a minute
May 05 16:09:12 <DocScrutinizer> nevertheless, need to eat now
May 05 16:09:14 <DocScrutinizer> bye
May 05 16:09:32 <Heinervdm> DocScrutinizer: with the notifier App, but in a different design, but for this raster has to implement a new profile into illume :)
May 05 16:10:58 <DocScrutinizer> gimmegimmegimme, I'm close to swapping phones back to klofisch due to this issue
May 05 16:12:38 <DocScrutinizer> i'm fine with everyting from an illumebar icon to flashing led to ringtone starting over again every 5 min. just the way it's no it's unbearable
May 05 16:13:03 <DocScrutinizer> it's noW
May 05 16:14:05 <mickeyl> ok
May 05 16:14:06 <SHR-WugBranler`> tracfeed: Ticket #424 (Boot/Reboot dialogue only 1 pixel wide) closed <http://trac.shr-project.org/trac/ticket/424#comment:2>
May 05 16:14:24 <mickeyl> my deepest suspicions...
May 05 16:14:41 <mickeyl> resource.py + idlenotifier.py from ms5.1 => OK
May 05 16:14:49 <mickeyl> updating resource again
May 05 16:16:07 <DocScrutinizer> errr idlenotifier? might this mean the issue FR isn't suspending again after resume - until I touch the screen - might be caused by this as well?
May 05 16:16:22 <mickeyl> if my suspicion is correct, i'll throw an axe over to Braunschweig :D
May 05 16:16:36 <mickeyl> let me do finish this experiment before drawing conclusions though 
May 05 16:21:41 <BluesLee> alphaone: what do you need exactly for the timezone issue?
May 05 16:22:49 <alphaone> mickeyl: Make sure you miss me, though :-)
May 05 16:23:25 <alphaone> BluesLee: could you start cli-framework and  issue a gsmsim.RetrieveMessage(<no>) for the message in question?
May 05 16:24:03 <alphaone> you can just gsmsim.RetrieveMessagebook("all") and look for the message in question
May 05 16:24:06 <BluesLee> alphaone: ehmm ... i am a dummy user;-)
May 05 16:24:48 <alphaone> BluesLee: Okay, so let's start by telling me which distro you use?
May 05 16:24:58 <BluesLee> shr testing
May 05 16:25:12 <alphaone> Can you ssh into the device?
May 05 16:25:17 <BluesLee> sure
May 05 16:25:25 <dos1> DocScrutinizer, mickeyl: that issue with autosuspend after resume is because frameworkd doesn't set busy state after resume
May 05 16:25:34 <dos1> mickeyl: can you look on it?
May 05 16:25:54 <mickeyl> not now
May 05 16:26:07 <mickeyl> right now i'm hunting something crazy
May 05 16:26:32 <dos1> mickeyl: new idlenotifier fixes issues with touchscreen being ignored by frameworkd... what the problem now?
May 05 16:26:37 <dos1> s/what/what's/
May 05 16:26:37 <BluesLee> alphaone: 'timestamp': 'Mon May  4 09:38:00 2009 +0200'
May 05 16:26:37 <bzzbot> dos1 meant: mickeyl: new idlenotifier fixes issues with touchscreen being ignored by frameworkd... what's the problem now?
May 05 16:26:39 <alphaone> BluesLee: please do so and run cli-framework
May 05 16:26:42 <alphaone> ah, okay
May 05 16:26:56 <BluesLee> google is my friend
May 05 16:27:03 <alphaone> BluesLee: And what is the time the messageing app displays?
May 05 16:27:28 <BluesLee> 7:38
May 05 16:27:33 <alphaone> And what is the time that should be displayed?
May 05 16:27:41 <BluesLee> 9:38
May 05 16:27:44 <alphaone> 9:38?
May 05 16:27:46 <alphaone> okay
May 05 16:27:49 <alphaone> Ainulindale: !!!!
May 05 16:27:51 <alphaone> BUUUUUG!
May 05 16:27:54 <alphaone> ;-)
May 05 16:28:23 <Ainulindale> What?
May 05 16:28:29 <Ainulindale> Not my fault
May 05 16:28:31 <mickeyl> ok
May 05 16:28:32 <Ainulindale> That's dos1's fault
May 05 16:28:39 <mickeyl> new resource.py + idlenotifier.py from ms5.1 => OK
May 05 16:28:43 <mickeyl> ~shoot shoragan
May 05 16:28:59 * mickeyl mutates into a green killer monster
May 05 16:29:25 <dos1> alphaone, Ainulindale: mrmoku was working on it...
May 05 16:29:34 <Ainulindale> Nah don't deny it it's your fault
May 05 16:29:34 <alphaone> ah, okay
May 05 16:29:46 <alphaone> Ainulindale: He didn't
May 05 16:30:20 <BluesLee> i will write my own sms app, starting now
May 05 16:30:22 <alphaone> mickeyl: So still what exactly is the problem?
May 05 16:30:29 <alphaone> Hehe
May 05 16:30:40 <BluesLee> just a joke ...
May 05 16:32:37 <alphaone> BluesLee: See, I laughed :-)
May 05 16:33:18 <BluesLee> one moment please
May 05 16:33:35 <BluesLee> there is something about a 'status-report-indicator' 
May 05 16:33:53 <BluesLee> does it work? and why is it not implemented yet?
May 05 16:33:59 <alphaone> yes
May 05 16:34:00 <alphaone> Huh?
May 05 16:34:05 <alphaone> define implemented
May 05 16:34:07 <BluesLee> in shr
May 05 16:34:15 <BluesLee> can i activate it there?
May 05 16:34:19 <dos1> BluesLee: let's implement it
May 05 16:34:20 <dos1> ;)
May 05 16:34:34 <BluesLee> i have to go:-)
May 05 16:35:04 <alphaone> lol
May 05 16:35:25 <BluesLee> dos1: on the neo thats a must have feature as dummy users as me dont trust their neos
May 05 16:36:23 <BluesLee> cu later
May 05 16:36:55 <Zorkman_> the butthole of openmoko: http://threads.moss-pultz.com/wp-content/uploads/2009/05/76199412-500x480.jpg
May 05 16:36:56 <mwester> mickeyl:  green?   I've never thought of you as green...  more a greyish-purple shade, tinged with orange highlights on the razor-sharp claws.
May 05 16:38:05 <mwester> max_posedon:  ubifs?  User's aren't confused enough with u-boot, Qi, NAND and NOR flash? :D
May 05 16:38:32 <max_posedon> ubifs can be great, if it works
May 05 16:38:43 <mwester> If it works.
May 05 16:38:50 <max_posedon> and one my friend-insider in nokia says that it works
May 05 16:39:02 <mwester> It's being used on the Sheevaplug by some.
May 05 16:39:07 <max_posedon> so let us[users] test
May 05 16:39:20 <max_posedon> just small and interesting option I think
May 05 16:39:27 <mwester> So its worth letting dev-types (not users) start working with it, I think.
May 05 16:39:42 <mwester> It certainly would be faster than jffs2 at boot.
May 05 16:40:10 <mwester> And I think there's even support for generating a ubifs image in OE.
May 05 16:40:53 <max_posedon> so, whould this option will be enabled? jffs2->ubifs converstion is little task, if compare it to kernel recompilation.
May 05 16:41:17 <mwester> But enabling it in the kernel is easy; I left an empty patch in the SHR repo just so we can patch the defconfig to suit SHR needs. :)   I'll get that into unstable very soon.
May 05 16:41:39 <mwester> gta02 only?
May 05 16:43:07 <mwester> One caveat: if it causes the kernel to grow larger than 2MBytes, Ainulindale will be unhappy -- so we'll have to see if that is an issue.
May 05 16:43:55 <max_posedon> I hope it won't
May 05 16:45:07 * DocMobilizer curses  this kernel-2M annoyance
May 05 16:47:08 <mwester> Yeah.  A bit short-sighted, I think...  allocating 4MB for the kernel, but only reading 2MB causes more problems than the small boot-time speedup it provides.
May 05 16:47:47 <CIA-75> freesmartphone.org: 03seba.dos1 07framework * rdb2b61b54ee7 10/framework/subsystems/opimd/backend_manager.py: opimd: add some useful dbus methods regarding backends
May 05 16:48:37 <CIA-75> freesmartphone.org: 03mickey 07framework * r3d97e5f3a85a 10/framework/patterns/dbuscache.py: patterns: add a dbus cache for object connections
May 05 16:48:39 <CIA-75> freesmartphone.org: 03mickey 07framework * r9c31977a2185 10/framework/subsystems/oeventsd/ (action.py fso_actions.py leds_actions.py oevents.py): oeventsd: use dbus cache and misc. cleanups
May 05 16:49:32 <dos1> please take a look on "OpenMoko and LCD TV" topic on community maillist :DDD
May 05 16:49:39 <dos1> interesting behaviour :x
May 05 16:50:31 <max_posedon> I think it is some kind of strange bluetooth spam
May 05 16:51:47 <DocMobilizer> btw wrt to joerg's-screenlock: raster confirmed we can have a msgbox at least hovering above dialer saying 'press aux to unlock' (alpha would need more work). there's a big issue tho to avoid ts-clicks resetting X idle-suspend timer
May 05 16:52:41 <mickeyl> fwiw, i could easily make the first click being swallowed
May 05 16:52:54 <mickeyl> or any clicks until the unlock has been performed
May 05 16:52:54 <DocMobilizer> which is mandatory for a decent screenlock
May 05 16:52:54 <Wonka> i wrote a small screenlocker that just opens the TS event device in exclusive mode...
May 05 16:53:58 <mickeyl> ptitjes: ping
May 05 16:54:04 <ptitjes> mickeyl: hello
May 05 16:54:18 <mickeyl> ptitjes: heya! got a moment? having some problems with dbus-experimental
May 05 16:54:29 <ptitjes> yeah just got home
May 05 16:54:32 <ptitjes> :)
May 05 16:54:37 <ptitjes> you're lucky
May 05 16:54:43 <Wonka> just open it and then do:
May 05 16:54:44 <Wonka>         fcntl.ioctl(fd, 0x40044590, ts_locked)
May 05 16:54:45 <DocMobilizer> well the goal is to have fr autosusp even when hammering the lockscreen
May 05 16:54:59 <Wonka> with ts_locked = 1 -> ts locked, and 0 -> unlocked
May 05 16:55:40 <ptitjes> mickeyl: what problems ?
May 05 16:55:52 <mickeyl> ptitjes: ok, first problem is a regression with software that does not use dbus error specifications:
May 05 16:56:07 <mickeyl> dbus-hlid does not compile with pending-upstream-move
May 05 16:56:13 <mickeyl> where as it compiles with master
May 05 16:59:06 <ptitjes> mickeyl: arf
May 05 16:59:31 <ptitjes> mickeyl: any valac output ?
May 05 17:00:05 <mickeyl> http://pastebin.ca/1413410
May 05 17:00:13 <mickeyl> compiles with valac, breaks in C code
May 05 17:00:24 <mickeyl> it may be illegal Vala code
May 05 17:00:30 <DocMobilizer> ptitjes: meeow
May 05 17:00:31 <mickeyl> we define throws DBus.Error
May 05 17:00:35 <mickeyl> but do not throw any
May 05 17:01:04 <ptitjes> oups criticals
May 05 17:04:17 <ptitjes> wait this is dynamic dbus
May 05 17:04:29 <ptitjes> mickeyl: I think I did not touch the dynamic part
May 05 17:06:10 <mickeyl> well, me neither
May 05 17:06:19 <mickeyl> still it works with master
May 05 17:06:40 <mickeyl> and mickey/0.7/posix
May 05 17:20:00 <Ainulindale> Oh, a bug introduced by ptitjes
May 05 17:20:05 <Ainulindale> We got leverage
May 05 17:20:37 <ptitjes> yes
May 05 17:20:43 <ptitjes> and that may not be the last one
May 05 17:20:46 <ptitjes> ;)
May 05 17:20:50 <DocScrutinizer> tztztz
May 05 17:20:52 * dos1 thinks Ainulindale is the biggest bug in SHR ;)
May 05 17:20:55 <dos1> ~seen Dave
May 05 17:20:56 <bzzbot> dave <n=dave@pool-138-89-1-87.mad.east.verizon.net> was last seen on IRC in channel #openmoko-cdevel, 19d 4h 1s ago, saying: 'dos1, I thought opimd was just a legend!? :o'.
May 05 17:21:09 <dos1> because Dave is gone :(
May 05 17:21:12 <DocScrutinizer> masterbug
May 05 17:21:26 <DocScrutinizer> ;D
May 05 17:28:35 <CIA-75> freesmartphone.org: 03mok 07libframeworkd-glib * rabfe3711bbfc 10/src/opimd/ (4 files in 2 dirs): 
May 05 17:28:35 <CIA-75> freesmartphone.org: opimd: fix MessageQuery interface to use proxy instead of query path
May 05 17:28:35 <CIA-75> freesmartphone.org: Signed-off-by: Klaus Kurzmann <mok@fluxnetz.de>
May 05 17:33:42 <mrmoku> alexxy: this time the build stoped at libframeworkd-glib... my fault :(
May 05 17:34:04 <mrmoku> fixed libframeworkd-glib and restarted the build
May 05 17:34:30 <mrmoku> NOTE: Running task 6932 of 6981
May 05 17:34:34 <mrmoku> (of shr-lite-image)
May 05 17:36:49 <DocScrutinizer> mrmoku: you really should think about some automation ;)
May 05 17:37:17 <Ainulindale> We do have automation
May 05 17:37:19 <Ainulindale> We just disabled it
May 05 17:42:34 <lindi-> DocScrutinizer: please see http://docs.openmoko.org/trac/ticket/2282
May 05 17:43:12 <DocScrutinizer> *sigh* Aye mon capitan
May 05 17:43:41 <CIA-75> freesmartphone.org: 03daniel 07framework * r6a12a0e3a15c 10/framework/subsystems/ogsmd/modems/abstract/mediator.py: 
May 05 17:43:41 <CIA-75> freesmartphone.org: ogsmd: Handle the case where AT+CLIR? only returns OK
May 05 17:43:41 <CIA-75> freesmartphone.org: o.f.GSM.Network.GetCallingIdentification now returns "unknown".
May 05 17:47:29 <DocScrutinizer> lindi-: hmmpf, everything lower than 6dBm difference is considered negligable in non-hifi-audio. Amyway feel free to ramp up control#4 (iirc) accordingly
May 05 17:47:53 <ptitjes> mickeyl: hum this seems like a clash in between two of my error management patches, in between the server impl. and client impl.
May 05 17:48:28 <CIA-75> freesmartphone.org: 03Frederik.Sdun 07dbus-hlid * r3eb39ba4e2ec 10fso-monitord/src/ (monitor.vala phone.vala preferences.vala): added some missing const for interfaces
May 05 17:48:29 <CIA-75> freesmartphone.org: 03Frederik.Sdun 07dbus-hlid * r47e2846be079 10fso-monitord/src/monitor.vala: fixed typo for HLID_BUSNAME
May 05 17:48:32 <ptitjes> mickeyl: because you method is a dbus served methods that also calls some other dbus method
May 05 17:48:41 <lindi-> DocScrutinizer: well, the issue is that people can hear the difference. maybe my voip provider is doing automatic gain control and that reduces the margin?
May 05 17:48:43 <mickeyl> can you fix that?
May 05 17:48:54 <DocScrutinizer> lindi-: actually I consider this result a proof of decent quality of FR with A7 statefile
May 05 17:49:05 <ptitjes> mickeyl: yeah sure
May 05 17:49:10 <lindi-> DocScrutinizer: but people still complain they can't hear what I am saying :)
May 05 17:49:37 <DocScrutinizer> surely not because of 4dB missing volume level
May 05 17:49:38 <ptitjes> mickeyl: but I need to make a test project for this
May 05 17:49:46 <mickeyl> ok
May 05 17:49:52 <mickeyl> we'll come to the 2nd bug later :)
May 05 17:50:03 <mickeyl> (which might be a feature...)
May 05 17:50:18 <ptitjes> mickeyl: just out of curiosity why do you implement a service that is the exact replica of another service
May 05 17:50:44 <mickeyl> dbus-hlid ?
May 05 17:50:46 <mickeyl> not a replica
May 05 17:50:50 <mickeyl> a missing feature in dbus
May 05 17:50:55 <mickeyl> read the readme
May 05 17:51:28 <ptitjes> mickeyl: I must tell you some other things. I think you have to declare all dbus methods to throw DBus.Error
May 05 17:51:36 <DocScrutinizer> lindi-: if people say they can't hear you you should produce a record to actually document this particular issue.
May 05 17:52:02 <ptitjes> mickeyl: that declaration enables to have dbus technical exceptions to propagate
May 05 17:52:06 <lindi-> DocScrutinizer: the problem is that many consumer phones can't record calls :(
May 05 17:52:11 <lindi-> DocScrutinizer: but yes, i am trying
May 05 17:52:17 <ptitjes> mickeyl: about the readme I read it
May 05 17:52:35 <ptitjes> mickeyl: but the code of ListBusNames is explicit
May 05 17:53:17 <lindi-> DocScrutinizer: btw, if you view the spectrograms of the calls side by side you see interesting differences in how they handle silence suppression
May 05 17:53:18 <DocScrutinizer> lindi-: if we're suspecting a bug in the dark corner, we mustn't start to search under the lantern just because it's nicely bright there
May 05 17:53:21 <mickeyl> ah well, use git blame, that's not my code :)
May 05 17:53:38 <mickeyl> apparantly playya did not know that there's already a function for that
May 05 17:54:30 <mickeyl> (declare all dbus methods to throw dbus.error) yes, i'll do this once i can play with the error propagation
May 05 17:54:40 <playya> a function to get the pid of the busname?
May 05 17:54:56 <mickeyl> playya: no, a list of all available busses
May 05 17:55:02 <mickeyl> err
May 05 17:55:04 <mickeyl> busnames
May 05 17:55:32 <mickeyl> or was it me?
May 05 17:55:36 <mickeyl> i don't remember
May 05 17:55:41 <playya> you use it in dbus-hlid. its org.freedesktop.BusNames
May 05 17:56:02 <mickeyl> ah right
May 05 17:56:05 <mickeyl> dunno why i export it
May 05 17:56:08 <mickeyl> should be an internal method
May 05 17:56:12 <mickeyl> since it's used by the other functions
May 05 17:56:20 <playya> can we add a GetMethodsByObject?
May 05 17:56:41 <playya> and it might be nice to have dbus-hlid on session bus
May 05 17:56:57 <mickeyl> ptitjes: error marshalling will work on both static and dynamic dbus i hope?
May 05 17:57:03 <ptitjes> mickeyl: you export it because you did not do any interface
May 05 17:57:17 <mickeyl> playya: possible
May 05 17:57:21 <mickeyl> dbus-hlid is quite old
May 05 17:57:29 <mickeyl> my first vala project w/ dbus
May 05 17:57:50 <mickeyl> playya: what is the semantic of GetMethodsByObject? All methods regardless of interface? What do you need that for?
May 05 17:58:15 <playya> but i think its ok to export ListBusNames because you won't need a second object in your app
May 05 17:58:35 <mickeyl> playya: we need to hook it on both system and session bus then. We do not have sane session busses though on most OE-based distros
May 05 17:58:58 <playya> but i'm using it on my PC :)
May 05 17:59:41 <mickeyl> good
May 05 17:59:44 <ptitjes> playya: but you traverse two times more the dbus wire
May 05 18:00:05 <playya> GetMethodsByObjects might be useful if you want to visualize the dbus for testing, or sth else
May 05 18:02:05 <mickeyl> no problem with using this, just make it not bail out if there's no session bus
May 05 18:03:13 <mickeyl> Might use GetInterfaceForObject( s: interface ), interface supporting wildcard
May 05 18:09:03 <CIA-75> freesmartphone.org: 03 07framework * r31b557ddfe6a 10/framework/subsystems/oeventsd/fso_actions.py: oeventsd: use subprocess.call to get rid of zombies (when using xset)
May 05 18:13:06 <ptitjes> mickeyl: the problem comes from the fact you do not declare the throws clause
May 05 18:13:17 <mickeyl> ok
May 05 18:13:27 <ptitjes> mickeyl: that should fire a better message for sure
May 05 18:13:30 <mickeyl> right
May 05 18:13:52 <ptitjes> mickeyl: but I don't know whether or not we should allow dbus methods that do not throw 
May 05 18:13:58 <mickeyl> so Vala should warn, if you call methods from methods were you do claim you throw something but the methods in between do not throw
May 05 18:14:00 <mickeyl> ?
May 05 18:14:05 <mickeyl> we better ask Jürg here
May 05 18:14:11 <ptitjes> nah
May 05 18:16:26 <DocScrutinizer> hmm, 10uF ceram-C with actual 7.6uF :-(
May 05 18:16:51 <DocScrutinizer> Zorkman_: that's been for you ;-)
May 05 18:17:25 <ptitjes> mickeyl: wait that I clear the problem
May 05 18:17:40 <ptitjes> mickeyl: but I need to tweak the code
May 05 18:25:47 <lindi-> DocScrutinizer: ok, managed to get a recording from a nokia phone
May 05 18:25:52 <lindi-> DocScrutinizer: it shows clear difference
May 05 18:27:22 <DocScrutinizer> lindi-: good
May 05 18:28:18 <DocScrutinizer> mickey|tv: enjoy :-)
May 05 18:28:32 <mickey|tv> thanks. need a break after 20h bug hunting
May 05 18:29:12 <DocScrutinizer> sure :)
May 05 18:30:35 <DocScrutinizer> mickey|tv: you'll love to hear we seem to have traced down #1024 with help of your device
May 05 18:30:57 <DocScrutinizer> well, mainly Dieter of course
May 05 18:35:45 <lindi-> DocScrutinizer: control #4 is already at maximum value in gsmhandset-a7.state
May 05 18:36:10 <lindi-> DocScrutinizer: maybe you mean #48?
May 05 18:37:15 <DocScrutinizer> lindi-: sorry I meant #5. see http://wiki.openmoko.org/wiki/Neo_Freerunner_audio_subsystem **please**
May 05 18:37:39 <lindi-> DocScrutinizer: ok, i'll investigate on that
May 05 18:38:28 <DocScrutinizer> lindi-: nevertheless you might want to check #48 for maximum value as well 
May 05 18:39:50 <DocScrutinizer> lindi-: I'm not even sure you used the right one of the both alsa.state files from #2121
May 05 18:40:00 <Zorkman_> DocScrutinizer: what do you mean, that 1024 is caused by a faulty C?
May 05 18:40:31 <DocScrutinizer> this C is +22/-82% (!!!)
May 05 18:40:43 <Zorkman_> wtf??
May 05 18:40:47 <Zorkman_> -82%
May 05 18:40:53 <DocScrutinizer> lol, yep
May 05 18:41:05 <DocScrutinizer> http://en.wikipedia.org/wiki/EIA_Class_2_dielectric
May 05 18:41:20 <DocScrutinizer> Y5V
May 05 18:41:23 <Zorkman_> LOL
May 05 18:41:33 <lindi-> DocScrutinizer: I used the one named gsmhandset-a7.state
May 05 18:41:50 <Zorkman_> those C's must be handrolled by monkeys
May 05 18:41:51 <DocScrutinizer> that's the right one by Wendy and me
May 05 18:42:00 <alphaone> DocScrutinizer: Ah, wait
May 05 18:42:14 <alphaone> lindi-: can you give me the link?
May 05 18:42:30 <alphaone> DocScrutinizer: There was one thing that can't be right iirc
May 05 18:42:33 <alphaone> Just remembered
May 05 18:42:41 <Zorkman_> Or it might just be darn diffucult to accomplish such randomness in you production of capacitators :)
May 05 18:42:52 <lindi-> alphaone: to http://docs.openmoko.org/trac/ticket/2282 ?
May 05 18:42:59 <DocScrutinizer> lindi-: NOPE its the WRONG one
May 05 18:43:17 <DocScrutinizer> alphaone: http://docs.openmoko.org/trac/ticket/2121
May 05 18:43:41 <DocScrutinizer> lindi-: http://docs.openmoko.org/trac/attachment/ticket/2121/gsmhandset.state.new
May 05 18:43:54 <lindi-> DocScrutinizer: huh
May 05 18:43:57 <alphaone> DocScrutinizer: Yeah, because http://docs.openmoko.org/trac/attachment/ticket/2121/gsmhandset-a7.state#L452
May 05 18:44:06 <alphaone> Volume 0, WTF
May 05 18:44:47 <DocScrutinizer> exactly. 
May 05 18:45:17 <DocScrutinizer> Mat mixed up echo and sensitivity issue once again, thus spoiling everything again
May 05 18:46:42 <alphaone> DocScrutinizer: Care to clarify in the ticket`
May 05 18:46:43 <lindi-> DocScrutinizer: you should clearly state this or distros will pick the wrong one
May 05 18:46:51 <alphaone> So noone is fooled by this?
May 05 18:47:13 <DocScrutinizer> I clearly stated it in #2121
May 05 18:47:31 <DocScrutinizer> I clearly stated it in my post to [community]
May 05 18:48:22 <DocScrutinizer> I'm clearly stating here:
May 05 18:48:43 <lindi-> I just clearly misread it :)
May 05 18:48:45 <DocScrutinizer> http://docs.openmoko.org/trac/attachment/ticket/2121/gsmhandset.state.new IS THE ONE *REAL* STATEFILE
May 05 18:48:50 <DocScrutinizer> http://docs.openmoko.org/trac/attachment/ticket/2121/gsmhandset.state.new IS THE ONE *REAL* STATEFILE
May 05 18:48:56 <lindi-> reruns tests...
May 05 18:49:00 <DocScrutinizer> one more time for the rest 
May 05 18:49:02 <DocScrutinizer> http://docs.openmoko.org/trac/attachment/ticket/2121/gsmhandset.state.new IS THE ONE *REAL* STATEFILE
May 05 18:49:32 <DocScrutinizer> FOR *ALL* GTA02
May 05 18:51:51 <alphaone> https://docs.openmoko.org/trac/ticket/2121#comment:3
May 05 18:52:05 <DocScrutinizer> and *if* you don't like the results from THE ONE *REAL* STATEFILE, please REFER to http://wiki.openmoko.org/wiki/Neo_Freerunner_audio_subsystem
May 05 18:52:21 <alphaone> unfortunately I can't just obsolete the old attachment
May 05 18:53:57 <DocScrutinizer> alphaone: :-)) thanks anyway
May 05 19:00:19 <alexxy> mrmoku: Ainulindale: what the state of rebuilding?
May 05 19:00:34 <mrmoku> alexxy: hehe... let me check
May 05 19:00:48 <mrmoku> Running task 7434 of 7468
May 05 19:00:52 <mrmoku> of fat image
May 05 19:00:53 <alexxy> i want my phone to function again
May 05 19:01:09 <alexxy> so it will finish soon
May 05 19:04:00 <Sharwin_F> mrmoku, did you finally build intone? :P
May 05 19:05:40 <CIA-75> freesmartphone.org: 03seba.dos1 07framework * raaed277dec2e 10/framework/subsystems/opimd/backend_manager.py: opimd: again few useful methods - Enable, Disable, SetAsDefault
May 05 19:05:46 <mrmoku> alexxy: ok... I put the lite image up... just for you :-)
May 05 19:05:49 <dos1> mrmoku: rebuild! :D
May 05 19:05:58 <mrmoku> (and anybody else wanting it)
May 05 19:06:11 <mrmoku> the feed is not yet finished though...
May 05 19:06:21 <mrmoku> Sharwin_F: yeah... built it... on my local host ;)
May 05 19:06:35 <Sharwin_F> mrmoku, better than nothing haha
May 05 19:06:47 <mrmoku> dos1: if SetAsDefault is persistent? yeah ;)
May 05 19:06:56 <dos1> mrmoku: yes, it is
May 05 19:07:09 <mrmoku> Sharwin_F: will build it when building the feed has finished
May 05 19:07:18 <Sharwin_F> ok :)
May 05 19:07:19 <mrmoku> it has one problem though... the icons don't work
May 05 19:07:31 <mrmoku> (play/fwd/rev)
May 05 19:07:31 <dos1> Enable and Disable are so much persistent
May 05 19:07:37 <Sharwin_F> so the image being builded now has all he problems from the last one solved?
May 05 19:07:40 <dos1> when you call Disable, it's disabled in config
May 05 19:07:51 <mrmoku> Sharwin_F: and lots of new ones ;)
May 05 19:07:54 <dos1> but contacts from that backend are still visible on that one session
May 05 19:08:07 <Sharwin_F> mrmoku, lol
May 05 19:08:20 <dos1> mrmoku: previous commit had GetDefaultBackend, GetEnabled, GetProperties
May 05 19:08:46 <mrmoku> dos1: and I found some segfaults... guess we're going forward :D
May 05 19:09:14 <mrmoku> dos1: started to implement the message list
May 05 19:09:15 <Sharwin_F> mrmoku, buttons are working on the one I installed ysterday or so from opkg.org
May 05 19:09:25 <mrmoku> Sharwin_F: yep... I know :(
May 05 19:09:49 <Sharwin_F> maybe the problem is on git version?
May 05 19:10:02 <Sharwin_F> and the one at opkg is older
May 05 19:10:27 <dos1> mrmoku: i implemented that methods for some shr-settings "storage" module in future ;)
May 05 19:12:16 <mrmoku> Sharwin_F: hmm... I cannot find any icons in the source...
May 05 19:15:06 <mrmoku> dos1: (contacts still visible) for the dbus session... or frameworkd lifetime?
May 05 19:15:14 <mrmoku> the second would be wrong me thinks
May 05 19:15:18 <dos1> mrmoku: frameworkd lifetime
May 05 19:15:27 <dos1> mrmoku: i know. it's my TODO ;)
May 05 19:15:31 <mrmoku> ok :P
May 05 19:16:08 <dos1> Enable doesn't change anything in contact list too
May 05 19:16:31 <dos1> you need to call Init, which you can call even with backend disabled
May 05 19:17:31 <dos1> only InitAllEntries cares about disabled state of backend
May 05 19:17:37 <dos1> and initing on startup of course
May 05 19:20:52 <mrmoku> dos1: guess you will put that into shr-settings?
May 05 19:21:23 <mrmoku> dos1: ahh, and another thing... is there a way to know from which backend a contact comes from?
May 05 19:21:33 <mrmoku> Think of showing a nice SIM icon for SIM contacts and so on...
May 05 19:22:11 <dos1> mrmoku: there is, but i think not via dbus
May 05 19:22:25 <dos1> something like _used_backends
May 05 19:26:21 <mrmoku> dos1: what do you think? could that be part of a contacts fields?
May 05 19:26:49 <mrmoku> or would an extra dbus call to get that info more apropriate?
May 05 19:26:53 <dos1> mrmoku: not field, dbus call
May 05 19:27:01 <dos1> one contact can have data from different backends
May 05 19:27:26 <dos1> err... it will be able to have, as merging is not implemented yet :D
May 05 19:27:56 <mrmoku> Contact.GetBackends with a(s) as result?
May 05 19:28:26 <mrmoku> anyway... not yet important :P
May 05 19:32:09 <dos1> mrmoku: yes, exactly
May 05 19:35:57 <CIA-75> freesmartphone.org: 03seba.dos1 07framework * r8d528f66b765 10/framework/subsystems/opimd/backend_manager.py: opimd: implement GetBackends()
May 05 19:36:11 <dos1> mrmoku: it's not THAT GetBackends
May 05 19:36:18 <dos1> that will be GetUsedBackends ;)
May 05 19:43:22 <alexxy> dos1: mrmoku : Ainulindale: why long sms messages are splited in messages app?
May 05 19:43:44 <max_posedon> hm... long messages sending works now?)
May 05 19:43:56 <mrmoku> alexxy: because they are splitted by frameworkd... and we don't combine them
May 05 19:44:28 <max_posedon> hm.. if speak abount incoming, why?
May 05 19:44:34 <alexxy> also sending long messages dont works at all 
May 05 19:44:35 <alexxy> =)
May 05 19:44:51 <mrmoku> alexxy: no? that should work... I thought
May 05 19:45:03 <alexxy> realy?! 
May 05 19:45:07 <alexxy> they dont work
May 05 19:45:15 <max_posedon> yep, confirm
May 05 19:45:26 <mrmoku> alphaone|gone: ping :)
May 05 19:45:50 <max_posedon> I use some shr-testing updated before shr was broke
May 05 19:46:23 <mrmoku> max_posedon: I never tried... because I'm no SMS writer ;)
May 05 19:46:33 <mrmoku> but I thought sending would work...
May 05 19:46:59 <alexxy> mrmoku: no
May 05 19:47:04 <alexxy> it doesnt work
May 05 19:47:15 <mrmoku> yeah... believe you :P
May 05 19:47:27 <alexxy> also it'll be good if i'll recive long sms as one sms not as its parts
May 05 19:47:59 <mrmoku> well... frameworkd receives it as parts... because long ones are split in several SMS
May 05 19:48:01 <max_posedon> as for me, I don't even dream about it)
May 05 19:48:15 <mrmoku> and it is an open discussion who should combine them... 
May 05 19:48:33 <mrmoku> though... clearly they have to be combined... totally agree on that one
May 05 19:48:49 <max_posedon> FSO high level API or shr
May 05 19:49:03 <mrmoku> exactly ;)
May 05 19:49:05 <max_posedon> fso high level api doesn't present now
May 05 19:49:11 <mrmoku> dos1: does opimd combine messages? :P
May 05 19:49:27 <dos1> mrmoku: don't know. Messages in opimd is still mystery for me
May 05 19:49:37 <dos1> i will discover it in few days i think
May 05 19:49:50 <mrmoku> where is your pim.py test thingie?
May 05 19:49:58 <mrmoku> have lots of split SMS on my SIM ;)
May 05 19:50:52 <dos1> mrmoku:  http://openmoko.opendevice.org/~dos/pim.py
May 05 19:51:14 <mrmoku> thanks
May 05 19:51:47 <Heinervdm> split messages can only occur on sim. Am i right?
May 05 19:52:14 <Heinervdm> so the sim backend have to do it
May 05 19:52:35 <dos1> Heinervdm: no. messages are splited when sending
May 05 19:53:05 <max_posedon> but what will be, if message splitted incorrectly?!
May 05 19:53:08 <Heinervdm> dos1: but all messages recived by gsm were saved on sim first?
May 05 19:53:31 <dos1> Heinervdm: that's one of things i will have to figure out ;D
May 05 19:53:52 <thrashold> Is elementary/elementary-python from 2009-03-13 too old to develop with? I found ebuilds for these for Gentoo and some of the examples I found are not working.
May 05 19:55:37 <max_posedon> thrashold, I think that version should be fine
May 05 19:55:45 <max_posedon> may be examples outdated
May 05 19:59:32 <thrashold> I guess the problem is with the examples. The __set__ for win.destroy accepts a tuple according to the example, and a callable with my version of elementary-python. Thanks. :)
May 05 20:00:34 <alexxy> dos1: mrmoku: so what the state of sending long sms? =)
May 05 20:00:47 <dos1> alexxy: why asking me? :D
May 05 20:01:15 <alexxy> =)
May 05 20:01:30 <alexxy> btw i still cannot get verify email from shr-track]
May 05 20:02:15 <alexxy> Ainulindale: ^^
May 05 20:05:06 <alexxy> mrmoku: also every time after reflash i get this stupid error
May 05 20:05:08 <alexxy> Collected errors:
May 05 20:05:08 <alexxy>  * Failed to download http://build.shr-project.org/shr-unstable/ipk//armv4/Packages.gz, error 404
May 05 20:05:20 <alexxy> =)
May 05 20:05:37 <alexxy> why not simply remove this feed from config by default
May 05 20:06:36 <max_posedon> as I remember official answer, nobody knows where it come from
May 05 20:07:15 <mrmoku> alexxy: it gets created automatically by OE... because theoretically there could be packages for armv4 arch
May 05 20:07:48 <alexxy> omg
May 05 20:08:13 <alexxy> then create empty arm4 feed =)
May 05 20:09:00 * Weiss plays tracker modules via mikmod on his FR
May 05 20:09:11 <DocScrutinizer> alexxy: good point
May 05 20:09:54 <wurp2|working> Ainulindale talked about doing just that, alexxy
May 05 20:10:20 <wurp2|working> He just hasn't done it yet.  Or maybe he was telling someone else to; it was kind of ambiguously phrased
May 05 20:10:25 <wurp2|working> and he's kind of bossy
May 05 20:12:56 <mrmoku> DocScrutinizer: read the mail about the statefile... aren't the statefiles from om repo?
May 05 20:17:07 <DocScrutinizer> which mail?
May 05 20:17:27 <DocScrutinizer> my mail?
May 05 20:18:31 <DocScrutinizer> mrmoku: I dunno where distros get their statefiles from. But I'm rather sure even OM got it wrong all the time
May 05 20:18:35 <DocScrutinizer> :-(
May 05 20:18:56 <mrmoku> not even now?
May 05 20:19:00 <mrmoku> still wrong?
May 05 20:19:18 <DocScrutinizer> I'm not checking OM repos
May 05 20:19:38 <max_posedon> )
May 05 20:19:43 <DocScrutinizer> I pointed nytowl to it
May 05 20:20:19 <DocScrutinizer> he told me it's on his todo-list "to look into the statefiles"
May 05 20:20:26 <nytowl> yup DocScrutinizer has been on my case already :)
May 05 20:20:56 <wurp2|working> nytowl: Is there a good microphone recording statefile hanging around out there?
May 05 20:21:23 <wurp2|working> Dictator doesn't work on neo1973 because it doesn't know what state file to use.  I hope to find a good one & send them a patch
May 05 20:22:20 <nytowl> you should be able to check the differences btween the 2 stae file on the freerunner and then modify it for the 1973
May 05 20:22:31 <DocScrutinizer> yup
May 05 20:22:57 <DocScrutinizer> diff between 1973 and FR is tiny
May 05 20:23:39 <nytowl> IIRC it's just the names and control  numbers don't match up 100%
May 05 20:24:02 <mrmoku> Ainulindale: is it normal that webkit-gtk takes endless and three days to build?
May 05 20:24:10 <mrmoku> looks like it is hanging....
May 05 20:24:16 <TAsn> wurp2|working, there's http://wiki.openmoko.org/wiki/Call_Recorder (which I wrote) as well, please send me the 1973 patch when you get it.
May 05 20:24:42 <mrmoku> TAsn: btw... is there some bb for your recorder?
May 05 20:25:20 <nytowl> i was just about to ask the same thing mrmoku 
May 05 20:25:21 <TAsn> mrmoku, nope ;[
May 05 20:25:24 <wurp2|working> TAsn: OK, will do
May 05 20:25:27 <TAsn> though it's just autotools
May 05 20:25:27 <TAsn> ;\
May 05 20:25:31 <TAsn> or wait, there is.
May 05 20:25:43 <mrmoku> nytowl: so you have the same problem?
May 05 20:25:55 <TAsn> mrmoku, 
May 05 20:25:56 <wurp2|working> nytowl: OK, I'll do that
May 05 20:26:06 <TAsn> sec, will give you a link
May 05 20:26:30 <mrmoku> nytowl: ahh... you were talking about the recorder :P
May 05 20:26:37 <TAsn> I can't assure the quality of it, though it's working here
May 05 20:26:43 <TAsn> that's what I used in order to build the packages.
May 05 20:27:09 <TAsn> http://projects.openmoko.org/plugins/scmsvn/viewcvs.php/*checkout*/trunk/callrec.bb?rev=4&root=callrec
May 05 20:27:17 <Heinervdm> mrmoku: i'm building it too, and it took al long time for me
May 05 20:27:21 <TAsn> feel free to modify it and send me a patch if I did something bad.
May 05 20:27:35 <mrmoku> Heinervdm: ok... will try to be patient then :(
May 05 20:28:15 <Heinervdm> mrmoku: so dbus-hlid is fixed?
May 05 20:28:17 <TAsn> I forgot I wrote that bb. ;]
May 05 20:28:44 <TAsn> anyhow, good night
May 05 20:28:50 <TAsn> mrmoku, let me know if you have any issues with it.
May 05 20:29:02 <TAsn> (with the app in general and the bb in specific)
May 05 20:30:27 <mrmoku> nytowl: will you add it to fso/ms5.5? Makes no sense that I stuff it in our overlay... if you will add it too :P
May 05 20:30:44 <mrmoku> Heinervdm: the problem was the vala version... it builds with the correct one
May 05 20:31:03 <nytowl> mrmoku, sure it looks like it needs a couple of minor tweaks
May 05 20:31:05 <Heinervdm> mrmoku: ah, you changed to mickey 0.7
May 05 20:31:34 <mrmoku> Heinervdm: yep
May 05 20:32:19 <mrmoku> ahh... webkit-gtk finally finished :D
May 05 20:32:49 <mrmoku> TAsn: thanks and good night
May 05 20:35:00 <spaetz> hey FSO people: Subject: [Shr-User] [PATCH] for fso_actions.py shouldn't produce zombies any   longer
May 05 20:35:18 <spaetz> this should probably applied
May 05 20:35:37 <spaetz> those xset zombies are quite annoying
May 05 20:37:14 <playya> afair there was a commit by alphaone. but i can't remember the branch
May 05 20:37:39 <SHR-WugBranler`> tracfeed: Ticket #185 (Problems mounting uSD card) closed <http://trac.shr-project.org/trac/ticket/185#comment:10>
May 05 20:38:17 <BluesLee> i need a good texteditor on the neo, which one should i use?
May 05 20:38:32 <mrmoku> spaetz: 31b557ddfe6a00f246d60fbb1388ab6be9826845
May 05 20:38:46 <Sharwin_F> BluesLee, vi? emacs? the fight starts xD
May 05 20:38:47 <mrmoku>     oeventsd: use subprocess.call to get rid of zombies (when using xset)
May 05 20:39:00 <BluesLee> Sharwin_F: no ... on the neo not on the desktop
May 05 20:39:06 <spaetz> ahh, thanks
May 05 20:39:22 <BluesLee> Sharwin_F: thanx for the b&w gtk theme!
May 05 20:39:27 <spaetz> mrmoku: btw, I am in contact with the mokomaze maintainer
May 05 20:39:38 <BluesLee> Sharwin_F: i mean thanx for the link
May 05 20:39:48 <mrmoku> spaetz: ahh... and?
May 05 20:39:52 <spaetz> and he told me a few things on how to package 0.5.1
May 05 20:40:02 <TAsn> BluesLee, how did you change the illume keyboard's height in order to make the big keyboard?
May 05 20:40:13 <spaetz> so I'll be working on that. no idea about libode though
May 05 20:40:56 <mrmoku> spaetz: ask him what the minimal ode version is he needs... maybe we could just tweak the configure check for it... to live with the older ode
May 05 20:41:00 <Sharwin_F> BluesLee, hmm I suppouse you got it from the txt file I showed you yesterday? :P
May 05 20:41:06 <BluesLee> TAsn: the second line should do the job kbd 130 76
May 05 20:41:12 <BluesLee> Sharwin_F: yes
May 05 20:41:24 <TAsn> BluesLee, ok, I feel like an idiot.
May 05 20:41:28 <TAsn> thanks and gn. ;]
May 05 20:41:31 <BluesLee> Sharwin_F: its not perfect, for instance in tagogps black on black 
May 05 20:41:40 <BluesLee> TAsn: no problem
May 05 20:41:42 <Sharwin_F> BluesLee, was you the one who had a script for installing certain things after reflashing?
May 05 20:41:49 <TAsn> forgot the structure of the keyboard file, should have probably looked there before asking. ;]
May 05 20:41:52 <spaetz> mrmoku: yeah, i''ll check that with him
May 05 20:41:53 <TAsn> cya.
May 05 20:41:59 <Sharwin_F> would be great to have it and modify with my preferences :D
May 05 20:42:22 <BluesLee> Sharwin_F: i wrote that it is a script but only some opkg commands, but a scripts should be a problem
May 05 20:42:35 <BluesLee> no good editor in the feed?
May 05 20:42:43 <spaetz> nano :)
May 05 20:42:53 <mrmoku> ouch
May 05 20:42:54 <BluesLee> what about enotes or defora or whatever
May 05 20:43:21 <Sharwin_F> BluesLee, dos 1 told yesterday he programs with the fr, he might know a good one
May 05 20:43:24 <mrmoku> Sharwin_F: how's your abiword going?
May 05 20:43:28 <BluesLee> a basic one would be okay, but no vi on the neo
May 05 20:43:51 <mrmoku> vim rules! (not the busybox one though...)
May 05 20:43:56 <Sharwin_F> mrmoku, stopped playing with it atm, I'm on exams hehe
May 05 20:44:08 <dos1|neo> i'm using nano and illume terminal keyboard ;D
May 05 20:44:13 <mrmoku> Sharwin_F: k :)
May 05 20:44:22 <dos1|neo> but that linewrapping is horrible
May 05 20:44:32 <spaetz> you can turn that off
May 05 20:44:39 <spaetz> just one setting in /etc/nanorc
May 05 20:44:42 <BluesLee> the illume terminal kbd is unusable even with a stylus;-)
May 05 20:44:57 <wurp2|working> dos1|neo: You badly need a bt kbd!  And a release on which it works ;-)
May 05 20:44:58 <dos1|neo> spaetz: oh, i will look at it, thanks :)
May 05 20:45:08 <wurp2|working> BluesLee: I can type on it with my finger, albeit slowly
May 05 20:45:22 <spaetz> set nowrap or similar
May 05 20:45:54 <BluesLee> who wrote the sms apps?
May 05 20:46:23 <BluesLee> it should be easy to do a quick note app
May 05 20:46:25 <spaetz> ~seen quickdev
May 05 20:46:25 <bzzbot> quickdev <n=quickdev@p5B054A61.dip.t-dialin.net> was last seen on IRC in channel #openmoko-cdevel, 69d 6h 11m 5s ago, saying: 'and for you?'.
May 05 20:46:37 <spaetz> this guy
May 05 20:47:37 <BluesLee> ohhh
May 05 20:47:46 <BluesLee> why he isnt here anymore?
May 05 20:48:26 <dos1|neo> BluesLee: ask him
May 05 20:49:04 <CIA-75> SHR: 03mok 07shr-overlay * r5ef5206eb03a 10/openembedded/recipes/tasks/task-shr-feed.bb: feed: remove zenity, add lf-phonegui-efl2 and intone
May 05 20:53:43 <spaetz> yay
May 05 20:53:50 <spaetz> intone
May 05 20:53:55 <spaetz> and nighht
May 05 20:53:57 <mrmoku> without icons though ;)
May 05 20:53:59 <mrmoku> night
May 05 20:58:31 <Heinervdm> can i set PARALLEL_MAKE to a value bigger than 4?
May 05 20:58:39 <ptitjes> mickey|tv: ok I found the fixes
May 05 20:59:35 <ptitjes> mickey|tv: the criticals and the C errors are from two different problems but both related to the fact you don't have throws DBus.Error clauses
May 05 20:59:38 <mrmoku> Heinervdm: yeah... let your cores glow ;)
May 05 21:00:01 <Heinervdm> mrmoku: i will test on 2x Xeon Quad Core
May 05 21:00:18 <ptitjes> mickey|tv: I will fix the dbus-experimental and pending-upstream-move branches (by amending the commits)
May 05 21:00:20 <mrmoku> so... 8 might be appropriate
May 05 21:00:27 <Heinervdm> mrmoku: ok
May 05 21:00:44 <ptitjes> mickey|tv: in the mean time you can add throws clauses, I think that will do
May 05 21:04:51 <ptitjes> mickey|tv: and the more I think it... there is chance I report an error there for the missing throws DBus.Error
May 05 21:14:40 <dos1|neo> Heinervdm, mrmoku: 2x4 cores? i would say 16 ;p
May 05 21:14:53 <dos1|neo> 2xnumber of corec
May 05 21:14:56 <dos1|neo> 2x4=8
May 05 21:15:02 <dos1|neo> 8x2=16
May 05 21:15:04 <dos1|neo> ;)
May 05 21:15:11 <dos1|neo> cores*
May 05 21:15:28 <Heinervdm> dos1|neo: now it's running with 8, but i think the first run isn't much faster because the internetconnection of this mashine is only 2 MBit
May 05 21:16:09 <Heinervdm> it needs more time for donwloading then for building ;)
May 05 21:17:11 <dos1|neo> Heinervdm: can you give me some of your cores? ;D
May 05 21:17:30 * mrmoku offers 4Mbit for 2 cores ;)
May 05 21:17:34 <Heinervdm> it's not really my mashine ;)
May 05 21:17:55 <Heinervdm> it's the dev server of the company i'm working for
May 05 21:18:22 * dos1|neo offers... working opimd ;D
May 05 21:20:01 <CIA-75> freesmartphone.org: 03mickey 07framework * r072434b0ac58 10/framework/subsystems/oeventsd/action.py: oeventsd: fix regression introduced by cleanup
May 05 21:20:33 <playya> shr guys: do you have a styleguide?
May 05 21:25:57 <m0nt0> hola
May 05 21:32:28 <CIA-75> freesmartphone.org: 03mickey 07libeflvala * r83a50a86bb44 10/vapi/eina.vapi: eina.vapi: replace deprecated # syntax with owned keyword
May 05 21:33:26 * mrmoku is off to bed... gnight all
May 05 21:35:46 <lindi-> DocScrutinizer: now the volume level is ok but people complain about odd distortion :)
May 05 21:36:22 <playya> is this a correct regex to detect a telephonenumber: ^[+0]?0?[:digit:]{3,31}$
May 05 21:40:25 <pwerken_> are there country codes that start with a zero ?
May 05 21:41:11 <playya> which country codes?
May 05 21:41:14 <pwerken_> your regex allows +0somenr
May 05 21:41:15 <playya> oh wait
May 05 21:41:41 <Heinervdm> whats with 00somenr ?
May 05 21:42:06 <playya> "^{+,0,00}?[:digit:]{3,31}$"
May 05 21:42:20 <tmzt> do gsm specs say anything?
May 05 21:50:48 <Weiss> would've thought you'd want some kind of international phone network spec for that, rather than GSM.. (?)
May 05 21:58:28 <dos1|neo> oh
May 05 21:58:46 <dos1|neo> i think i have fix for incorrect timestamps
May 05 21:59:49 <dos1|neo> mrmoku|away: you applied your timezone patch only for message list, not for message view
May 05 22:01:18 <DocScrutinizer> lindi-: if distortion happens I recommend reducing #48 by one step, increasing #5 accordingly to get same level at far end, and see if distortion is away 
May 05 22:02:49 <DocScrutinizer> lindi-: think of #48 like the record level dial for a tape recorder, and #5 the playback volume dial
May 05 22:04:34 <DocScrutinizer> lindi-: if you set #48 very odd, you either get distortion one way, or noise the other way. You need to find correct setting for #49 close to clipping (distortion), then adjust actual level to your preferences with #5
May 05 22:05:05 <DocScrutinizer> s/#49/#48/
May 05 22:05:06 <bzzbot> DocScrutinizer meant: lindi-: if you set #48 very odd, you either get distortion one way, or noise the other way. You need to find correct setting for #48 close to clipping (distortion), then adjust actual level to your preferences with #5
May 05 22:06:38 <lindi-> DocScrutinizer: it seems to be non-deterministic
May 05 22:07:03 <lindi-> DocScrutinizer: it can be heard only on some calls, and during those calls it persists during the whole call
May 05 22:07:33 <ptitjes> mickey|tv: what were the other problems
May 05 22:07:34 <ptitjes> ?
May 05 22:07:54 <DocScrutinizer> ny recommendation is: call to a quite place far end, set #5 to a rather low level (~10..15 below usual), then adjust #48 to get low but clear sound (no distortion) when *shouting* at the device. Then level up #5 to get good volume level when speaking gently to device
May 05 22:09:14 <DocScrutinizer> try to level up #48 bat much as possible in procedure mentioned above
May 05 22:09:18 <mickey|tv> ptitjes: http://rafb.net/p/J38Wuc83.html
May 05 22:09:35 <mickey|tv> ptitjes: even after trying to add [DBus name = "foo.bar"] to the errordomain clauses
May 05 22:09:49 <mickey|tv> ptitjes: do you have a proper example for how to use all the new goodies?
May 05 22:10:08 <mickey|tv> (Note that I'm only using dynamic dbus though)
May 05 22:10:27 <ptitjes> mickey|tv: yes my test case locally :)
May 05 22:10:43 <ptitjes> mickey|tv: that is [DBus (name = "")]
May 05 22:11:07 <lindi-> DocScrutinizer: still i don't understand where the non-determinism comes from
May 05 22:11:13 <ptitjes> mickey|tv: but it seems the same missing throws clause also
May 05 22:11:30 <mickey|tv> ok
May 05 22:11:36 <DocScrutinizer> lindi-: btw - is distortion a worse problem than people not hearing you?
May 05 22:11:43 <ptitjes> mickey|tv: yes there is an example in dbus-glib1.vapi
May 05 22:11:49 <ptitjes> I'm stupid I had forgot this one
May 05 22:12:09 <ptitjes> mickey|tv: because DBus.Error is handled in the same way
May 05 22:12:11 <lindi-> DocScrutinizer: two out of two people say yes
May 05 22:12:18 <mickey|tv> k
May 05 22:12:23 <DocScrutinizer> lindi-: there's plenty of sources for non-determinism in GSM
May 05 22:13:33 <DocScrutinizer> lindi-: (2 of 2) That would bother me, honestly ;-) If people prefer to not hear you at all, rather than get some distortion
May 05 22:13:49 <lindi-> DocScrutinizer: remember voice distortion caused by AT%CBHZ=1 ?
May 05 22:13:51 <DocScrutinizer> XD
May 05 22:14:03 <DocScrutinizer> yup, cloudy
May 05 22:14:09 * mickey|tv still never got any real proof
May 05 22:14:21 <lindi-> I've got recorded calls now :)
May 05 22:14:35 <ptitjes> mickey|tv: apply this http://pastebin.com/m6410fd4e on your vala
May 05 22:14:43 <lindi-> DocScrutinizer: so I'm somewhat suspecting something similar
May 05 22:14:46 <DocScrutinizer> lindi-: I closed your ticket for that behalf
May 05 22:14:48 <ptitjes> mickey|tv: you'll get proper reporting of missing error clauses
May 05 22:15:01 <ptitjes> mickey|tv: this will remove a lot of criticals
May 05 22:15:02 <lindi-> DocScrutinizer: the one about -a7.state?
May 05 22:15:11 <mickey|tv> ok
May 05 22:15:16 <ptitjes> mickey|tv: yet I hope to not have to clone all the projects ;)
May 05 22:15:41 <ptitjes> mickey|tv: however with this patch dbus-hlid builds correctly
May 05 22:16:17 <ptitjes> (I mean it reports the errors and fail)
May 05 22:16:35 <ptitjes> (if we can say it is correct ;) )
May 05 22:16:35 <mickey|tv> hmm, against which is that?
May 05 22:16:45 <mickey|tv> fails to apply here
May 05 22:16:56 <ptitjes> that is a git diff
May 05 22:17:59 <mickey|tv> i don't have the rev you base on
May 05 22:18:09 <ptitjes> mickey|tv: nah really ?
May 05 22:18:31 <DocScrutinizer> lindi-: http://docs.openmoko.org/trac/ticket/2282#comment:1
May 05 22:18:37 <ptitjes> mickey|tv: 8429deec54d9ec22b178a29656a7ab85e2f9fe5c
May 05 22:18:49 <mickey|tv> right
May 05 22:18:51 <mickey|tv> have this here
May 05 22:18:53 <mickey|tv> doesn#t apply
May 05 22:18:58 <mickey|tv> perhaps broken in pastebin
May 05 22:19:03 <ptitjes> arf
May 05 22:19:24 <lindi-> DocScrutinizer: yep that is all fine
May 05 22:20:28 <DocScrutinizer> lindi-: so publish your new results and we'll see how to tackle this
May 05 22:20:59 <DocScrutinizer> lindi-: either change topic of #2282 or open a new one
May 05 22:21:01 <mickey|tv> tabs have been converted to spaces
May 05 22:21:05 <mickey|tv> that's why it doesn't apply
May 05 22:21:13 <mickey|tv> use a better pastein :)
May 05 22:21:47 <DocScrutinizer> muahaha, *bad* for python
May 05 22:21:51 <lindi-> DocScrutinizer: yep i'll open a new one
May 05 22:21:59 <lindi-> DocScrutinizer: the non-determinism scares
May 05 22:22:00 <lindi-> DocScrutinizer: the non-determinism scares me
May 05 22:22:04 <mickey|tv> s/python/patch/
May 05 22:22:18 <DocScrutinizer> lindi-: me too, sometimes 
May 05 22:22:19 <lindi-> DocScrutinizer: i'm thinking about setting up automated daily 60-second test calls that get recorded :)
May 05 22:22:21 <DocScrutinizer> :-o
May 05 22:22:37 <lindi-> since my voip provider gives me 3000 minutes free per month
May 05 22:22:47 <DocScrutinizer> duh
May 05 22:22:55 <DocScrutinizer> which one???
May 05 22:22:57 <lindi-> (if I call their GSM numbers of course)
May 05 22:23:04 <DocScrutinizer> aaaah
May 05 22:23:10 <DocScrutinizer> so nebermind
May 05 22:23:21 <DocScrutinizer> never even
May 05 22:23:37 <ptitjes> mickey|tv: arf even the valadoc's one does convert to spaces...
May 05 22:24:16 <mickey|tv> that sucks
May 05 22:24:18 <DocScrutinizer> hehe
May 05 22:24:47 <DocScrutinizer> I kinda hate python for this. obsolete since COBOL
May 05 22:25:00 <mickey|tv> ??
May 05 22:25:04 <mickey|tv> s/python/patch/
May 05 22:25:19 <mickey|tv> the patch binary breaks when you convert spaces to tabs
May 05 22:25:20 <DocScrutinizer> col7 == comment mark
May 05 22:25:47 <DocScrutinizer> col3 == additional comment space in regular lines
May 05 22:25:48 <lindi-> DocScrutinizer: is there any docs on what happens inside calypso vrt. audio and how to control it?
May 05 22:25:54 <DocScrutinizer> col73 even
May 05 22:26:26 <DocScrutinizer> mickey|tv: format-syntax lang are *evil*
May 05 22:26:42 <mickey|tv> we agree to disagree here
May 05 22:26:51 <DocScrutinizer> honestly not even linebreak should make any diff
May 05 22:27:21 <DocScrutinizer> lindi-: sparse and poor
May 05 22:27:23 <DocScrutinizer> :-/
May 05 22:28:35 <DocScrutinizer> mickey|tv: no intetion to argue. U know: real hackers can code FORTRAN in *any* language ;)
May 05 22:29:31 <mickey|tv> yo
May 05 22:29:48 <DocScrutinizer> http://www.phttp://www.pbm.com/~lindahl/real.programmers.htmlbm.com/~lindahl/real.programmers.html
May 05 22:29:59 <mickey|tv> ptitjes: i'll be back tomorrow, need to catch some sleep now. g'night
May 05 22:31:45 <DocScrutinizer> lindi-: nevertheless you should contact alphaone|gone for the "nondeterministic" topic, it seems there's something going on that needs investigation
May 05 22:32:17 <dent> What's the state of shr-unstable now? should I stick to testing, or is unstable ok? (I lost track again... trying to catch up.)
May 05 22:32:55 <DocScrutinizer> lindi-: alphaone|gone thought it's related to moko-FW version, but this theory seems to not pan out as well
May 05 22:33:40 <lindi-> DocScrutinizer: hmm
May 05 22:33:49 <DocScrutinizer> neverheless *something* obviously is odd there
May 05 22:35:03 <DocScrutinizer> maybe calypso ALC is locking up on extreme conditions or sth. No bright idea what might be the root cause, but its too much complaints same general direction to simply ignore them
May 05 22:35:11 <lindi-> DocScrutinizer: check http://iki.fi/lindi/openmoko/call-quality/recordings/
May 05 22:35:24 <lindi-> DocScrutinizer: openmoko-correct-state2.wav shows the problem, openmoko-correct-state3.wav and openmoko-correct-state4.wav don't
May 05 22:35:50 <DocScrutinizer> lindi-: check your permissions ;-)
May 05 22:36:30 <DocScrutinizer> bbiab
May 05 22:36:59 <lindi-> DocScrutinizer: should work now
May 05 22:38:06 <lindi-> DocScrutinizer: how about restarting wolfson in the middle of the call just to make sure it is not causing this?
May 05 22:38:47 <DocScrutinizer> WTF?! sounds like backwards
May 05 22:39:00 <lindi-> DocScrutinizer: do we know that wolfson is innocent? ;)
May 05 22:39:19 <DocScrutinizer> wolfson isn't mangeling audio either way whatsoever
May 05 22:39:30 <DocScrutinizer> yup
May 05 22:39:34 <DocScrutinizer> we know
May 05 22:39:46 <DocScrutinizer> if you set #63 to mic2
May 05 22:39:47 <lindi-> DocScrutinizer: but i thought those gain controls in alsa were part of wolfson?
May 05 22:40:00 <DocScrutinizer> they are
May 05 22:40:19 <DocScrutinizer> but that can't cause the mess I heard 10sec ago ;-)
May 05 22:41:19 <DocScrutinizer> I'm not trained to listen to finish(?) but THAT sounds RATHER STRAAAANGE
May 05 22:42:12 <DocScrutinizer> never heard such a mess of echo back and forth and background noise even with my worst RTP-SIP-commections
May 05 22:42:34 <lindi-> DocScrutinizer: hmm, was there really echo?
May 05 22:42:47 <lindi-> DocScrutinizer: mostly just everything I say gets clipped
May 05 22:43:03 <DocScrutinizer> half of the sound feels like playback backwards echo to me
May 05 22:43:31 <lindi-> DocScrutinizer: are those other recordings ok by your standards?
May 05 22:44:07 <DocScrutinizer> yeah I noticed some clipping. but thats not really bad and just indicates we got reserve in amplification and you should probably go down on #48 from 3 to 2
May 05 22:44:36 <DocScrutinizer> (other) was so buffled I hadn't even a test on them
May 05 22:45:16 <lindi-> DocScrutinizer: hmm, what do you mean?
May 05 22:45:39 <lindi-> DocScrutinizer: (re the other recordings)
May 05 22:45:41 <DocScrutinizer> lindi-: (3 to 2) though you should check first if it gets better on adjusting #5
May 05 22:46:01 <DocScrutinizer> a minute please
May 05 22:46:19 <lindi-> i'll be away for 45 minutes or so now
May 05 22:49:00 <DocScrutinizer> lindi-: ok, it's pretty clear now those recordings have a strong slow dynamic compression+noisegate which makes the start of a word ramp up in volume - this leads to other non-finish feeling like it's backwards
May 05 22:49:18 <DocScrutinizer> the second recording also has a little clipping
May 05 22:49:44 <DocScrutinizer> you should do as adviced: go down #48 1 step, readjust #5
May 05 22:50:12 <DocScrutinizer> and you should try other AT%Nxxxx values, urgently!
May 05 22:52:56 <DocScrutinizer> I never was aware ther could be such a bad compression/noisegate (such a slow attack time) anywhere in our device. I wonder if it might be introduced by some other ghost cooking the audio in amoniak
May 05 22:54:22 <DocScrutinizer> to realy rely on these recordings we had to verify all the involved components to ensure what is Freerunner and what is recording equipment
May 05 22:55:39 <DocScrutinizer> anyway, I'm rather scared now, as I never before heard such effects on OM devices
May 05 22:57:53 <DocScrutinizer> a standard "one two, one two, popp! eeeek, pop, one two three" probably would help a lot to take out oddity of finish language which nobody except lindi- knows how it should sound in original.
May 05 22:58:29 <DocScrutinizer> an plain recording done by laptop microphone and e.g. krecord would also help a lot, to compare
May 05 23:02:04 <DocScrutinizer> basics of noisegate: low volumes get even lower, high volumes are boosted. the ramp up and ramp down is a timeconstant that's much too long in the samples I heard, also the noisegate effect is too strong (too much boost for loud words, low words a too much attenuated)
May 05 23:04:04 <DocScrutinizer> now I suggest to c&p all this mumbling of mine of he last 30min to some appropriate place in wiki or where-ever, and I'll go to watch TV for a while
May 05 23:05:39 <tmzt> how much of this is caused by using an anolg link from baseband to wm codec?
May 05 23:05:48 <tmzt> analog
May 05 23:06:05 <DocScrutinizer> there's no codec in wm
May 05 23:06:31 <DocScrutinizer> wm8753 is a mere audio mixer for calypso
May 05 23:06:47 <tmzt> it doesn't have one or your not using it in this design?
May 05 23:07:12 <DocScrutinizer> we are not using it, and it doesn't have a GSM codec
May 05 23:07:23 <DocScrutinizer> it has a PCM codec
May 05 23:07:26 <tmzt> pcm codec
May 05 23:07:28 <DocScrutinizer> aka ADC
May 05 23:07:31 <tmzt> right
May 05 23:07:49 <tmzt> many phones use ssp or other pcm link
May 05 23:07:53 <DocScrutinizer> it's not used for gsmhandset.state
May 05 23:08:02 <tmzt> between baseband and codec
May 05 23:08:05 <DocScrutinizer> we use analog audio
May 05 23:08:23 <DocScrutinizer> our baseband does PCM as well
May 05 23:08:36 <DocScrutinizer> see schematics
May 05 23:08:39 <tmzt> which is what I'm asking, is this a major reason for these problems?
May 05 23:08:47 <DocScrutinizer> nope
May 05 23:08:56 <tmzt> ok
May 05 23:09:56 <DocScrutinizer> calypso is the major problem (heard this before? ;D )
May 05 23:10:40 <DocScrutinizer> ahh sorry. calypso == baseband
May 05 23:11:59 <DocScrutinizer> (for the 119 who might wonder what I'm talking about)
May 05 23:12:06 <tmzt> how can volume in bp affect pcm though?
May 05 23:12:22 <DocScrutinizer> err, please rephrase
May 05 23:12:31 <DocScrutinizer> elaborate
May 05 23:12:42 <tmzt> the AT commands for setting volume
May 05 23:12:47 <DocScrutinizer> yup
May 05 23:13:09 <tmzt> since part of the solution seems to set that to yhe highest setting
May 05 23:13:09 <DocScrutinizer> these are for playback (earpiece) mainly
May 05 23:13:41 <DocScrutinizer> yup, as we got too low volume on earpiece sometimes aka on some devices
May 05 23:14:18 <tmzt> earpiece isn't connected directly to bp though?
May 05 23:14:29 <DocScrutinizer> so we need decent level on WM8753 *input* so we can juggle with mixer settings
May 05 23:14:46 <DocScrutinizer> nope, is thru WM8753 as well as mic
May 05 23:15:58 <DocScrutinizer> http://wiki.openmoko.org/wiki/Neo_1973_audio_subsystem#using_phone-internal_microphone_and_speaker
May 05 23:16:06 <DocScrutinizer> tho this is incorrect
May 05 23:16:50 <tmzt> I'm asking because the gta03 list said mc71 would work the same way, with analog rather than pcm routing
May 05 23:17:03 <DocScrutinizer> this is correct one
May 05 23:17:06 <DocScrutinizer> http://people.openmoko.org/joerg/ALSA/doc/WM8753_control_diag_gsmhandset_mic_std.png
May 05 23:17:27 <DocScrutinizer> mc75i, yup correct
May 05 23:17:55 <DocScrutinizer> on the two prototypes actually existing it's exactly the same as in GTA02
May 05 23:18:30 <DocScrutinizer> as we used WM8753 as well, and WM8753 has no decent usecase for this
May 05 23:19:06 <DocScrutinizer> we need the PCM I2S for MCP (SoC, CPU whatever U name it), and BT
May 05 23:19:36 <tmzt> network mode would be to complicated?
May 05 23:20:25 <DocScrutinizer> see http://wiki.openmoko.org/wiki/Neo_1973_audio_subsystem#using_Bluetooth_headset_with_GSM
May 05 23:20:41 <DocScrutinizer> what is "network mode" ?
May 05 23:25:26 <tmzt> more than two devices on ssp like busses
May 05 23:32:02 <DocScrutinizer> what is ssp?
May 05 23:38:00 <tmzt> synchronous serial port I think
May 05 23:38:11 <lindi-> DocScrutinizer: ok the problem is that I can't reproduce the worst file (*2.wav) reliably
May 05 23:38:23 <tmzt> not to familiar with sasung socs though
May 05 23:39:55 <DocScrutinizer> lindi-: first adjust statefile to ge really good sound w/o clipping under standard circumstances. Then we try to spot the "non-deterministic issue"
May 05 23:40:17 <lindi-> DocScrutinizer: but i don't think it is clipping normally, 3 and 4 sound just ok for me :)
May 05 23:40:23 <DocScrutinizer> also please notice my comment above regading AT%Nxxxx  and noisegate
May 05 23:40:41 <lindi-> DocScrutinizer: AT%N0187 set yes, is there some way to configure noisegate?
May 05 23:40:45 <DocScrutinizer> lindi-: 3 & 4 have clipping as well
May 05 23:40:50 <tmzt> is the 'clipping' analog or digital?
May 05 23:40:51 <lindi-> really?
May 05 23:41:15 <DocScrutinizer> lindi-: AFAIKT
May 05 23:41:53 <lindi-> ok, at least we agree that 2 sounds very odd compared to 3&4
May 05 23:42:02 <DocScrutinizer> or this in quatization noise if you happen to record these with 8bit e.g.
May 05 23:42:28 <DocScrutinizer> yup, 2 sounds extremely odd and scary
May 05 23:42:30 <lindi-> DocScrutinizer: my friend used some odd symbian program to do the recording on the nokia phone
May 05 23:42:42 <DocScrutinizer> duh
May 05 23:42:44 <DocScrutinizer> :-/
May 05 23:43:36 <DocScrutinizer> please try meddling the AT%N value. It's controlling noisegate and AEC concurrently
May 05 23:43:55 <lindi-> hrm. very many variables indeed
May 05 23:44:50 <DocScrutinizer> though I would not believe this is actual quality you encounter on cellphone. It sounds like a bad compression algo introduced by this nokia recorder app
May 05 23:46:01 <DocScrutinizer> (many vars) that'S the crux of the bisquit: reducing vars in test setup until you *know for sure* what you get in your results
May 05 23:46:07 <lindi-> DocScrutinizer: well, did you compare this to nokia1.wav?
May 05 23:46:17 <DocScrutinizer> not yet
May 05 23:46:31 <lindi-> since that is nokia->nokia call
May 05 23:46:34 <lindi-> for reference
May 05 23:47:29 <DocScrutinizer> same scary compander mismatch issue on nokia1.wav
May 05 23:47:57 <lindi-> "good", then we can blame the recorder app
May 05 23:48:13 <lindi-> what's compander?
May 05 23:48:21 <DocScrutinizer> yup, and alas all the recording are hmmm... void?
May 05 23:48:41 <DocScrutinizer> compressor-expander audio processor. e.g DBX
May 05 23:48:48 <DocScrutinizer> or HiCOM
May 05 23:48:52 <lindi-> ok
May 05 23:49:14 <lindi-> DocScrutinizer: I should put a real mic near the speaker of the nokia phone then?
May 05 23:49:21 <DocScrutinizer> all those compress on start of channel and expand on end
May 05 23:50:00 <DocScrutinizer> you should use your SIP provider's voicebox. great quality, extremely convenient
May 05 23:50:29 <DocScrutinizer> can send inbound recorded calls to you via email
May 05 23:50:32 <lindi-> DocScrutinizer: except that that way it costs. I can only make voip->gsm calls for free :)
May 05 23:51:13 <DocScrutinizer> not my problem. You asked for a good test setup, not for a cheap useles one ;-)
May 05 23:51:13 <tmzt> can your sip client record?
May 05 23:51:27 <DocScrutinizer> my sip provider does
May 05 23:51:48 <lindi-> tmzt: yep, i did that too, see bug #2282
May 05 23:52:03 <DocScrutinizer> my sip client does as well, if you use the ALSA-file setup I developed for twinkle
May 05 23:52:04 <tmzt> url?
May 05 23:52:12 <lindi-> tmzt: http://docs.openmoko.org/trac/ticket/2282
May 05 23:52:24 <lindi-> DocScrutinizer: i think my provider has voicemail but no support for sending those via email
May 05 23:52:33 <lindi-> DocScrutinizer: however, asterisk at work can do that
May 05 23:52:49 <DocScrutinizer> yup, for sure
May 05 23:53:02 <lindi-> DocScrutinizer: I'd capture the actual RTP packets and decode the audio from those :)
May 05 23:53:20 <DocScrutinizer> there's no use in testbeds which test the FE more than the FR
May 05 23:53:38 <lindi-> FE?
May 05 23:53:46 <DocScrutinizer> np, Wireshark can do this
May 05 23:53:52 <DocScrutinizer> Far End
May 05 23:54:07 <lindi-> ok
May 05 23:54:43 <lindi-> would be also interesting to know if gsm compression codec is uncompressed and then compressed again
May 05 23:54:56 <DocScrutinizer> tho wireshark knows .au only. np for alay
May 05 23:55:00 <lindi-> I mean, is the compression end-to-end in this voip->gsm case
May 05 23:55:01 <DocScrutinizer> aplay
May 05 23:55:52 <DocScrutinizer> basically GSM codec is similar to MP3
May 05 23:56:07 <DocScrutinizer> there's no such thing like a dedicated compression
May 05 23:56:25 <lindi-> what do you mean dedicated compression?
May 05 23:56:26 <tmzt> send them to vlc might work
May 05 23:56:31 <DocScrutinizer> it depends on GSM->RTP gateway
May 05 23:56:36 <playya> any of the shr guys still alive?
May 05 23:56:59 <lindi-> DocScrutinizer: the same operator runs both so at least they could do it
May 05 23:57:40 <DocScrutinizer> lindi-: sorry I ge a little tired ow. maybe I'm ot that receptive anymore
May 05 23:57:57 <lindi-> no problem
May 05 23:58:45 <DocScrutinizer> anyway GSM raw data is an abstract coding of audio
May 05 23:59:27 <DocScrutinizer> it's impossible to record in hiuman readable form without employing a GSM-decoder
May 05 23:59:49 <tmzt> are there sip providers who pass it through unaltered?
May 05 23:59:53 * mwester-laptop checks pricing on a 10 uF capacitor, one that is reasonably temperature-insensitive...
May 05 23:59:55 <DocScrutinizer> so the concept of compression is kinda void in GSM raw data
May 06 00:00:59 <DocScrutinizer> mwester-laptop: check for pricing of a decent 22uF X5R
May 06 00:01:08 <mwester-laptop> 22?
May 06 00:01:11 <DocScrutinizer> 0805
May 06 00:01:34 <quatrox> audio?
May 06 00:01:35 <mwester-laptop> big enough that I might have a chance to solder it.
May 06 00:01:52 <DocScrutinizer> well if 7.6 is too low, and 100 fixes #1024 ;-)
May 06 00:02:16 <mwester-laptop> It's just a filter then?
May 06 00:02:30 <DocScrutinizer> V-DBB buffer C
May 06 00:02:41 <mwester-laptop> That's it?  Wow!
May 06 00:02:43 <DocScrutinizer> V-DBB aka V-PLL
May 06 00:02:48 <mwester-laptop> heh!
May 06 00:03:08 <DocScrutinizer> C1009
May 06 00:03:24 <mwester-laptop> So one might not even have to remove C1009, if one can solder on top of it?
May 06 00:03:34 <DocScrutinizer> probably
May 06 00:03:56 <DocScrutinizer> don't ask me questions. we're at the beginning still
May 06 00:04:28 <DocScrutinizer> welcome to investigate and come up with results and news tho
May 06 00:05:00 <mwester-laptop> ok.  :)  I have a gta01 with the problem open, and I'm willing to attempt some soldering
May 06 00:06:13 <DocScrutinizer> cool :)))
May 06 00:07:46 <DocScrutinizer> mwester-laptop: please post to me and dieter via PM
May 06 00:08:23 <DocScrutinizer> we don't want to start a stampede on OM (or even inside OM ;)
May 06 00:08:27 <Dr_>  
May 06 00:08:49 <lindi-> tmzt: http://en.wikipedia.org/wiki/Enhanced_full_rate has an interesting detail :)
May 06 00:10:05 <DocScrutinizer> mwester-laptop: and we like to make some credit out of this unfunded research ;-))
May 06 00:10:43 <DocScrutinizer> maybe OM likes to start another project if they see Dieter is really great and comes up with results
May 06 00:16:17 <tmzt> lindi-: interesting
May 06 00:16:25 <DocScrutinizer> lindi-: intersting detail on EFR?
May 06 00:17:04 <tmzt> but I'm talking about a sip gateway that passes through gsm codec data
May 06 00:17:31 <tmzt> maybe asterik would work??
May 06 00:18:32 * DocScrutinizer feels like "wo took my brain"
May 06 00:18:33 <playya> tmzt, you should realize that nearly all providers block SIP protocols
May 06 00:19:00 <DocScrutinizer> playya: which providers?
May 06 00:19:04 <playya> i already thought about ZRTP on FR
May 06 00:19:25 <DocScrutinizer> playya: use twinkle. it has ZRTP
May 06 00:19:39 <DocScrutinizer> mrmoku|away: what's with twinkle bb?
May 06 00:19:40 <playya> afair every in germany
May 06 00:19:50 <tmzt> playya: not talking about ip from the fr
May 06 00:19:52 <DocScrutinizer> DSL, GPRS?
May 06 00:20:11 <tmzt> I mean treating sip like a trunk
May 06 00:20:25 <playya> t-online even wants to block it if you have a hotspot flat
May 06 00:20:47 <playya> you want to send raw data?
May 06 00:20:58 <DocScrutinizer> I actually did a decent SIP call via N810, M800-klofisch and UMTS-HSDPA flatrate on o2-loop
May 06 00:21:54 <playya> i want to try it over GPRS
May 06 00:22:19 <DocScrutinizer> use a low-bw codec and it might work
May 06 00:22:42 <playya> + encryption
May 06 00:22:50 <playya> which might kill my FR
May 06 00:23:11 <DocScrutinizer> btw blocking SIP isn't that easy. You need either staefull packet nspection, or you block port 5060
May 06 00:23:30 <DocScrutinizer> SIP can work on other ports than 5060
May 06 00:23:33 <playya> afair they are blocking port 5060
May 06 00:23:48 <playya> but most servers use 5060
May 06 00:23:55 <DocScrutinizer> sure
May 06 00:24:13 <playya> it was in the media because of the skype port fpr iphone
May 06 00:24:14 <DocScrutinizer> bur your proxy doesn't have to
May 06 00:25:06 <DocScrutinizer> some providers might allow other ports as well (5062 e.g.)
May 06 00:25:25 <DocScrutinizer> client side is free choice anyway
May 06 00:25:53 <playya> but you can filter the most recent port
May 06 00:26:11 <DocScrutinizer> hm g'n8
May 06 00:26:12 <playya> maybe you can try to setup a ssh tunnel to a server
May 06 00:26:37 <playya> good night
May 06 02:17:45 <DocScrutinizer> hmm, after an uptime of ~3d, hwcloch is (still?) 11sec ahead of DCF77, but systime is 2:50 behind of hwclock :-/
May 06 02:18:40 <DocScrutinizer> as usual systime is much less accurate than RTC
**** ENDING LOGGING AT Wed May 06 02:59:57 2009