**** BEGIN LOGGING AT Tue Jan 25 02:59:57 2011
Jan 25 04:32:09 <DocScrutinizer51> hmm, I wonder how you'd repartition NAND to make NOLO partition a little bit larger
Jan 25 05:32:00 <DocScrutinizer> cat /proc/mtd; 
Jan 25 05:32:03 <DocScrutinizer> IroN900:/home/user# ls -l /initrd/
Jan 25 05:32:05 <DocScrutinizer> insgesamt 0
Jan 25 05:32:36 <DocScrutinizer> to me it seems we got a rather comfortably sized free partition
Jan 25 05:32:43 <DocScrutinizer> will check this
Jan 25 06:04:43 <DocScrutinizer> http://mg.pov.lt/maemo-irclog/%23maemo.2011-01-25.log.html#t2011-01-25T07:33:42
Jan 25 06:05:43 <DocScrutinizer> when gnutoo resurrects, somebody may point him to chanlog and this short § above
Jan 25 06:41:45 <DocScrutinizer> PaulFertser: ping
Jan 25 06:50:09 <PaulFertser> DocScrutinizer: pong
Jan 25 06:50:22 <DocScrutinizer> morning :-)
Jan 25 06:50:29 <DocScrutinizer> PaulFertser: http://mg.pov.lt/maemo-irclog/%23maemo.2011-01-25.log.html#t2011-01-25T07:33:42
Jan 25 06:51:20 <PaulFertser> DocScrutinizer: yep, so this link, will pass to GNUtoo, haven't read yet.
Jan 25 06:54:31 <DocScrutinizer> jus was interested in your beating me up, for flaws or glitches in my logic
Jan 25 07:15:55 <PaulFertser> DocScrutinizer: (change 4 bytes) if only it was in RAM, yes. But you need a bit more: actually load code from nand doing ECC and skipping bad blocks etc.
Jan 25 07:16:43 <DocScrutinizer> umm, aiui you can change the bytes in NOLO img file
Jan 25 07:16:54 <PaulFertser> DocScrutinizer: it would be easier to write some specialised bootstrap-loader to the NOLO partition that would in turn load u-boot from the initrd partition.
Jan 25 07:17:07 <PaulFertser> DocScrutinizer: you can, but it won't help load u-boot from nand.
Jan 25 07:17:19 <DocScrutinizer> aaah, I see
Jan 25 07:17:42 <DocScrutinizer> bad block in initrd uBoot -yes, that's the first killer I heard so far
Jan 25 07:18:19 <PaulFertser> DocScrutinizer: and i mean, why you need NOLO anyway? I do not remember but probably u-boot can be split into two parts... Or just use a very minimal u-boot version in the nolo partition to bootstrap real u-boot.
Jan 25 07:18:32 <DocScrutinizer> check for magic and skip block if not found. Magic = starting 4 bytes of uBoot
Jan 25 07:19:00 <DocScrutinizer> Uboot is too large, and can't flash
Jan 25 07:21:26 <PaulFertser> DocScrutinizer: tbh, i'm rather sick from all this nokia shit, seeing hurdles that wouldn't be there if some devs/managers weren't too stupid is depressing :/ 
Jan 25 07:22:10 <DocScrutinizer> a lot is NDA shit as usual. E.G. NOLO has gfx drivers (c) TI afaik
Jan 25 07:22:55 <DocScrutinizer> milestone is waaaay more depressing ;-D
Jan 25 07:22:55 <PaulFertser> DocScrutinizer: ok, another "plan": why not to get rid of the stupid mtd layout and just use whatever space it takes for the u-boot partition?
Jan 25 07:23:48 <PaulFertser> DocScrutinizer: xloader doesn't need to know about that most probably, as it'd be enough to load first X bytes of u-boot, and those X bytes will fetch the remainder from NAND.
Jan 25 07:23:58 <DocScrutinizer> [2011-01-25 06:51:40] * DocScrutinizer wonders anyway WhereTF the NAND "partition table" might be - whether this is a mtdparts.h that simply is #include'd to everything that deals with NAND, like flasher, libcal/sysinfo, kernel...
Jan 25 07:26:04 <PaulFertser> DocScrutinizer: if we do not care about flasher and sysinfo, then we can pass whatever partition table to the kernel from u-boot.
Jan 25 07:27:27 <DocScrutinizer> if you don't care about flasher, so how would you flash any NAND partition?
Jan 25 07:27:48 <PaulFertser> DocScrutinizer: ah, btw, flasher/sysupgrade uses kernel interface to get partitions, so it's not hardcoded.
Jan 25 07:28:03 <DocScrutinizer> :-D
Jan 25 07:28:04 <PaulFertser> DocScrutinizer: (flash) from inside a running u-boot or kernel.
Jan 25 07:28:18 <DocScrutinizer> hehe, bootstrap problem
Jan 25 07:29:17 <PaulFertser> Not really, once you start good u-boot with console anyway (e.g. by embedding it in the kernel partition like it's currently done) you can basically do anything you want.
Jan 25 07:29:37 <PaulFertser> gtg, see you in an hour.
Jan 25 07:29:51 <DocScrutinizer> cya, and thanks for sparring
Jan 25 07:43:54 <JaMa|Zzzz> moin
Jan 25 07:48:15 <CIA-77> freesmartphone.org: 03morphis 07msmcomm * r38ef95c49915 10/ (8 files in 6 dirs): libmsmcomm: add command message to retrieve call forward status from sim
Jan 25 07:48:17 <CIA-77> freesmartphone.org: 03morphis 07msmcomm * r6e3bec6d726e 10/libmsmcomm/msmcomm/simmessage.vala: libmsmcomm: remove debug printf statement
Jan 25 07:53:44 <morphis> JaMa: moin
Jan 25 08:04:49 <PaulFertser> DocScrutinizer: back
Jan 25 08:05:22 <PaulFertser> DocScrutinizer: i understand you do not want to loose all compatibility with maemo. Ok, but what makes you do not like the u-boot-in-the-kernel approach?
Jan 25 08:05:27 <PaulFertser> Used by PK.
Jan 25 08:06:16 <DocScrutinizer> the dependencies between kernel and uboot
Jan 25 08:06:49 <DocScrutinizer> also it seems even this approach has too few space to get a proper fullsized uBoot
Jan 25 08:07:52 <DocScrutinizer> and quite obviously it will block usual system updates of whatever system you are running from that kernel
Jan 25 08:08:28 <DocScrutinizer> every kernel update will break uBoot-kernel-hybrid
Jan 25 08:09:02 <PaulFertser> DocScrutinizer: well, when every new kernel is bundled with u-boot anyway (as titan does), why bother?
Jan 25 08:09:20 <DocScrutinizer> actually titan doesn't
Jan 25 08:10:26 <DocScrutinizer> and still the available free space is too low for a proper uBoot, in kernel partition, afaik
Jan 25 08:12:43 <DocScrutinizer> most recent PK is by mohammadAG
Jan 25 08:12:53 <DocScrutinizer> PK+uBoot pkg
Jan 25 08:14:15 <PaulFertser> Ok...
Jan 25 08:14:30 <DocScrutinizer> FSCK, my neighbour and landlord and friend thinks it's funny to bash off the plaster from the walls next room
Jan 25 08:14:53 <DocScrutinizer> >X-(
Jan 25 08:19:59 <DocScrutinizer> ali1234 (uBoot-kernel-hybrid developer) just claimed sth along the line of flashing a new kernel won't break anything, the device wipes uBoot and reverts to original OOTB behaviour, that's a feature, not a flaw/problem/bug
Jan 25 08:20:24 <DocScrutinizer> he nevertheless prefers my mtd4 suggestion, as it's "cleaner" :-P
Jan 25 08:20:50 <CIA-77> freesmartphone.org: 03morphis 07msmcomm * r4e2c0c8b5627 10/libmsmcomm/ (9 files in 5 dirs): libmsmcomm: add first messages for the sound subsystem
Jan 25 08:20:51 <CIA-77> freesmartphone.org: 03morphis 07msmcomm * r20c3cf9df408 10/libmsmcomm/TODO: libmsmcomm: add list of messages they are known left to implement
Jan 25 08:23:17 <CIA-77> freesmartphone.org: 03morphis 07msmcomm * rfc705af7c153 10/msmcomm-specs/src/ (Makefile.am sound.vala): msmcomm-specs: add dbus api specification for sound subsystem
Jan 25 08:24:26 <DocScrutinizer> so, the funny statistics are: #maemo: <RobbieThe1st> That would be -damn- cool || #meego-arm: <Stskeeps> DocScrutinizer: well, i'm not saying you can't do it, but considering how much trouble we had already with uboot, it's not a direction i'd go in personally :) ||  #openmoko-cdevel:  <PaulFertser> DocScrutinizer: tbh, i'm rather sick from all this nokia shit, seeing hurdles that wouldn't be there if some devs/managers weren't 
Jan 25 08:24:28 <DocScrutinizer> too stupid is depressing :/ 
Jan 25 08:35:59 <PaulFertser> DocScrutinizer: i see... what if you use a full-featured u-boot in kernel partition and fetch the kernel from the big internal storage?
Jan 25 08:36:18 <PaulFertser> (not criticizing your idea, but pondering about pros/cons of alternatives)
Jan 25 08:53:50 <DocScrutinizer> a nice alternative
Jan 25 08:54:40 <DocScrutinizer> but won't comply with simple reflashing
Jan 25 08:56:32 <DocScrutinizer> also it won't empower GNUtoo to replace NOLO by uBoot, and aiui that's what he is planning to do
Jan 25 09:01:03 <PaulFertser> DocScrutinizer: heh, why reflash when you can copy? ;)
Jan 25 09:01:53 <DocScrutinizer> copy a new maemo version? where from?
Jan 25 09:02:42 <PaulFertser> DocScrutinizer: ah, you're talking about stock maemo compatibility again. Isn't it obvious pr1.3 was the last anyway?
Jan 25 09:12:16 <Azog> DocScrutinizer: hi, i know tina and tobi (from willich), can i query you?
Jan 25 09:14:56 <DocScrutinizer> Azog: go ahead
Jan 25 10:31:48 <mrmoku> GNUtoo|laptop: 07:56 < DocScrutinizer> 07:04:23> http://mg.pov.lt/maemo-irclog/%23maemo.2011-01-25.log.html#t2011-01-25T07:33:42
Jan 25 10:31:51 <mrmoku> 07:56 < DocScrutinizer> 07:05:22> when gnutoo resurrects, somebody may point him to chanlog and this short § above
Jan 25 10:32:51 <mrmoku> GNUtoo|laptop: 07:56 < DocScrutinizer> 06:31:43> IroN900:/home/user# ls -l /initrd/
Jan 25 10:32:54 <mrmoku> 07:56 < DocScrutinizer> 06:31:44> insgesamt 0
Jan 25 10:33:06 <mrmoku> 07:56 < DocScrutinizer> 06:32:16> to me it seems we got a rather comfortably sized free partition
Jan 25 10:35:31 <GNUtoo|laptop> ok
Jan 25 10:39:52 <GNUtoo|laptop> mrmoku, what do you think of getting uboot or barebox on nokia n900?
Jan 25 10:41:42 <mrmoku> GNUtoo|laptop: I don't know barebox... but real u-boot on n900 would be great indeed
Jan 25 10:42:04 <GNUtoo|laptop> I don't think I'll port real uboot, but barebox = uboot v2 instead
Jan 25 10:42:13 <GNUtoo|laptop> real uboot is hard to work with
Jan 25 10:42:25 <GNUtoo|laptop> but it has some advantages
Jan 25 10:42:30 <GNUtoo|laptop> such as ttyACM0 drivers
Jan 25 10:42:38 <GNUtoo|laptop> so I'm still undecided
Jan 25 10:42:43 <GNUtoo|laptop> barebox is easier
Jan 25 10:42:53 <GNUtoo|laptop> but more work is needed
Jan 25 10:43:02 <GNUtoo|laptop> you need some driver for usb serial
Jan 25 10:43:07 <GNUtoo|laptop> only imx has them
Jan 25 10:44:37 <GNUtoo|laptop> note that doing what I do or replacing the bootloader is dangerous
Jan 25 10:44:55 <GNUtoo|laptop> even if there is coldflash 
Jan 25 10:45:01 <misc> "do not do it at home, kids"
Jan 25 10:45:01 <GNUtoo|laptop> this scenario is problematic:
Jan 25 10:45:11 <GNUtoo|laptop> if you flash a bootloader
Jan 25 10:45:20 <GNUtoo|laptop> and you have battery discharged....
Jan 25 10:47:00 <floyd2> GNUtoo|laptop: what do you mean with 'real uboot'
Jan 25 10:48:19 <GNUtoo|laptop> I mean that:
Jan 25 10:49:09 <GNUtoo|laptop> bootrom_of_omap_of_nokia900->xload-signed(for now until we use the atack vector to replace that)->u-boot
Jan 25 10:49:19 <GNUtoo|laptop> instead of:
Jan 25 10:49:34 <GNUtoo|laptop> bootrom_of_omap_of_nokia900->xload-signed(for now until we use the atack vector to replace that)->nolo->kernel_or_uboot_chainloaded
Jan 25 10:50:21 <floyd2> ah, ok
Jan 25 10:51:07 <GNUtoo|laptop> if battery is discharged with our custom bootloader, we have no way to recover but buy another battery, coldboot the previous bootloader trough bootrom
Jan 25 10:51:14 <GNUtoo|laptop> so I think I'll buy a second battery
Jan 25 12:42:52 <CIA-77> freesmartphone.org: 03morphis 07cornucopia * rfe3f0feb9bbd 10/fsogsmd/src/plugins/modem_qualcomm_palm/ (5 files): fsogsmd: modem_qualcomm_palm: initial work for network register/unregister mediators
Jan 25 12:42:55 <CIA-77> freesmartphone.org: 03morphis 07cornucopia * rbadbd88a7a04 10/fsogsmd/conf/palm_pre/fsogsmd.conf: fsogsmd: palm_pre conf: default loglevel to DEBUG
Jan 25 12:43:36 <CIA-77> freesmartphone.org: 03morphis 07msmcomm * rc50d40121f52 10/msmcommd/conf/msmcommd.conf: msmcommd: conf: default loglevel to DEBUG
Jan 25 12:50:27 <CIA-77> freesmartphone.org: 03morphis 07cornucopia * r29b43a7d6e29 10/libfsobasics/fsobasics/kernel.vala: libfsobasics: kernel module loader: first set value then check it ...
Jan 25 13:12:48 <morphis> JaMa: ping
Jan 25 13:17:16 <JaMa> pong
Jan 25 13:26:25 <morphis> JaMa: you still have the issues with not working e17 too?
Jan 25 13:26:39 <morphis> as for me changing to newer SRCREV does not fix the problem
Jan 25 13:26:48 <JaMa> no and I didn't have it..
Jan 25 13:27:31 <JaMa> I have even newer SRCREV in my branch again
Jan 25 13:27:37 <JaMa> but this time only rev is changed
Jan 25 13:33:08 <morphis> jepp I changed my SRCREV to the newest but is still there
Jan 25 13:34:16 <morphis> seems to have to debug that in detail ...
Jan 25 13:34:35 <morphis> s/seems to/seem like I/ 
Jan 25 13:37:08 <JaMa> did I send you link for that e-mail on OE ML?
Jan 25 13:37:26 <JaMa> that guy has maybe same problem, but haven't replied yet
Jan 25 13:47:34 <morphis> I saw the mail
Jan 25 13:47:39 <morphis> mail I should talk to him
Jan 25 13:59:02 <morphis> yeah! first opkg upgrade on palmpre over WLAN is done!
Jan 25 14:03:01 <mickey|office> congrats
Jan 25 14:03:41 <pespin> morphis, great! now only remains downloading all youtube via torrent :P
Jan 25 14:03:53 <morphis> hehe
Jan 25 14:06:38 <morphis> mickey|office: what about conmann integration, you already though about it?
Jan 25 14:08:47 <mickey|office> morphis: just briefly, we need to hook with fsogsmd and the resource concept into it
Jan 25 14:09:13 <mickey|office> or add a plugin that serves as connman agent
Jan 25 14:09:19 <mickey|office> offering the high level API we want to clients
Jan 25 14:09:50 <morphis> so we just tell conmann, we are here and let it handle our api to activate PPP for example?
Jan 25 14:09:58 <morphis> or do we wrap conmann?
Jan 25 14:10:34 <mickey|office> i don't know which is better
Jan 25 14:10:43 <mickey|office> i need more experience with the connman API to judge
Jan 25 14:11:13 <mickey|office> both ways are possible, with connman being the "master" and us being connman plugins or with us being the "master" having a connman agent.
Jan 25 14:11:35 <mickey|office> in the end what i'd love is an API as simple as
Jan 25 14:11:53 <mickey|office> SetOnlinePreferences( "local", "wifi", "bt", "GSM" )
Jan 25 14:12:08 <mickey|office> Start
Jan 25 14:12:09 <mickey|office> Stop
Jan 25 14:12:23 <morphis> jepp
Jan 25 14:12:37 <morphis> as far as I know ofono did a plugin for connammn
Jan 25 14:15:11 <mickey|office> we also need to update the FSO plugin for bluez
Jan 25 14:15:22 <mickey|office> the one Felix Huber wrote
Jan 25 14:15:33 <mickey|office> it's still referring to fso1
Jan 25 14:16:44 <mickey|office> but one step after another... first ophoned here
Jan 25 14:17:20 <morphis> ok
Jan 25 14:17:41 <morphis> is the bluez plugin upstream?
Jan 25 14:18:18 <mickey|office> not. it has been submitted, but the bluez guys were so finnicky with whitespace issues and stuff that after trying for some time, Prof. Huber just gave up :/
Jan 25 14:18:27 <morphis> ok
Jan 25 14:18:45 <morphis> and where is it now?
Jan 25 14:18:48 <mickey|office> OE
Jan 25 14:19:01 <mickey|office> there's no svn for it afaik
Jan 25 14:19:06 <mickey|office> scm, fwiw
Jan 25 14:19:16 <mickey|office> i can ask him whether we have the last version
Jan 25 14:19:26 <mickey|office> he has BT with the FR and his car running 100% solid
Jan 25 14:20:02 <morphis> http://git.openembedded.net/cgit.cgi/openembedded/tree/recipes/bluez/files/obexd-add-fso-support.patch ?
Jan 25 14:20:22 <mickey|office> yes, that looks like it
Jan 25 14:21:13 <mickey|office> hmm, not sure whether that's all
Jan 25 14:21:26 <morphis> ok
Jan 25 14:21:42 <morphis> bluetooth would be the next thing I want for palmpre
Jan 25 14:21:46 <mickey|office> good
Jan 25 14:21:54 <morphis> but first we need running ppp connection
Jan 25 14:21:58 <mickey|office> ok
Jan 25 14:22:02 <mickey|office> that should be very easy
Jan 25 14:22:20 <mickey|office> i need to find some time to install the dual boot...
Jan 25 14:23:16 <morphis> hm /org/freesmartphone/Device/PowerControl does not show up ...
Jan 25 14:23:22 <morphis> mickey|office: its great :)
Jan 25 14:23:52 <mickey|office> what do you have to do for BT power?
Jan 25 14:24:04 <mickey|office> rfkill? sysfs dance? or something else?
Jan 25 14:25:18 <morphis> attach the hsuart to bluez
Jan 25 14:25:33 <morphis> it's the same type of hsuart used for the modem
Jan 25 14:25:50 <morphis> there is already hci_attach to attach serial bluetooth drivers
Jan 25 14:26:12 <morphis> but we need to rewrite it to support the hsuart as it is completly different to normal uarts
Jan 25 14:26:22 <morphis> so in theory it should be simple
Jan 25 14:28:40 <mickey|office> hmm, k
Jan 25 14:30:06 <morphis> but first ppp
Jan 25 14:33:00 <mickey|office> I'd expect this to work more or less out of the box when the correct dataPort settings are given
Jan 25 14:34:00 <morphis> what I have to do to get it running with fsogsmd?
Jan 25 14:36:42 <mickey|office> configure the data port
Jan 25 14:36:45 <mickey|office> in the config
Jan 25 14:37:03 <morphis> data_access?
Jan 25 14:37:08 <mickey|office> yes
Jan 25 14:37:22 <mickey|office> the only problem might be the SetCredentials call
Jan 25 14:37:33 <mickey|office> i'm not sure where this tries to write to
Jan 25 14:37:47 <mickey|office> as this is the first modem with binary/AT parallel
Jan 25 14:38:00 <mickey|office> if in doubt, register an own mediator 
Jan 25 14:38:11 <mickey|office> but ideally we'd just use the At mediators for that
Jan 25 14:40:04 <morphis> 2000-01-01T00:33:38.478424Z [CRITICAL] fsogsmd : GLib <fsotransport>: fso_framework_abstract_command_queue_enqueueCommand: assertion `self != NULL' failed
Jan 25 14:40:16 <morphis> hm there seems to be some AT mediator involved ..
Jan 25 14:40:36 <morphis> after SetCredentials
Jan 25 14:40:44 <morphis> so you were right :)
Jan 25 14:41:33 <mickey|office> yep
Jan 25 14:41:40 <mickey|office> so first we need to open a 2nd channel
Jan 25 14:41:44 <mickey|office> (in the medium class)
Jan 25 14:42:07 <mickey|office> then we need to return this channel when the corresponding v250 and PDP calls are being requested
Jan 25 14:42:12 <mickey|office> (in channelForCommand or so)
Jan 25 14:42:20 <mickey|office> s/medium/modem/
Jan 25 14:42:35 <morphis> ah right there comes something back in my mind :)
Jan 25 14:43:49 <morphis> but first I have to do something for my exam tomorrow ...
Jan 25 14:48:15 <HeinervdmWork> morphis: that patch was my version for pbap in obex felix huber is working on hfp in bluez
Jan 25 14:48:37 <morphis> HeinervdmWork: ah ok
Jan 25 14:48:59 <morphis> HeinervdmWork: seems to have some knowledge in dealing with bluez :)
Jan 25 14:49:04 <morphis> s/to/you/
Jan 25 14:49:04 <apt> morphis meant: HeinervdmWork: seems you have some knowledge in dealing with bluez :)
Jan 25 14:49:11 <HeinervdmWork> not really :)
Jan 25 14:49:14 <morphis> hm ok
Jan 25 14:49:27 <HeinervdmWork> i tried to implement it but that version didn't work
Jan 25 14:49:27 <morphis> as I am searching someone to get it running on the palmpre
Jan 25 14:49:31 <morphis> ok
Jan 25 14:49:53 <HeinervdmWork> the problem is, that i have not much to test bluetooth
Jan 25 14:50:33 <HeinervdmWork> implementing pbap without a device that can talk pbap was to hard ;)
Jan 25 14:50:52 <HeinervdmWork> i just have a bluetooth headset and a bluethooth dongle for my pc
Jan 25 14:52:02 <HeinervdmWork> morphis: do you have a startingpoint for bluetooth on the pre? or is it from scratch
Jan 25 14:52:42 <morphis> I know whats to do :)
Jan 25 14:52:52 <HeinervdmWork> ok :)
Jan 25 14:53:04 <morphis> should be not too much effort to get it running (in theory)
Jan 25 14:53:28 <morphis> should be enough when bluez is running and we have initial bluetooth running
Jan 25 14:53:36 <morphis> everything else will come
Jan 25 14:53:46 <morphis> so you want to try it?
Jan 25 14:54:25 <HeinervdmWork> i'm short in time, so i don't know if i can handle this
Jan 25 14:54:56 <morphis> ok, first step should be simple
Jan 25 14:55:17 <morphis> palm uses a hsuart for communication with the bluetooth chip
Jan 25 14:55:27 <morphis> there is the hci_attach utlitity 
Jan 25 14:55:37 <morphis> it's std. bluetooth utlity on linux
Jan 25 14:55:47 <morphis> but we can't use it as is
Jan 25 14:56:00 <morphis> as it uses std. terminal commands to open the serial connection
Jan 25 14:56:05 <morphis> and we have a hsuart
Jan 25 14:56:16 <morphis> (the same typ of hsuart is used for the modem)
Jan 25 14:56:24 <morphis> /dev/btuart is the devnode
Jan 25 14:56:52 <morphis> so we need to rewrite hci_attach for our needs
Jan 25 14:57:12 <morphis> thats the first step
Jan 25 14:57:21 <HeinervdmWork> ok
Jan 25 14:57:26 <morphis> than there could be some sysfs nodes involved
Jan 25 14:59:36 <morphis> look at the hci_attach documentation and code how they are dealing with uarts used for bluetooth communication
Jan 25 14:59:40 <GNUtoo|laptop> hi morphis, porting uboot has some caveats:
Jan 25 14:59:53 <GNUtoo|laptop> first xload is signed
Jan 25 15:00:03 <morphis> really?
Jan 25 15:00:07 <GNUtoo|laptop> basically bootrom->xload is signed
Jan 25 15:00:11 <GNUtoo|laptop> but xload->nolo is not
Jan 25 15:00:14 <morphis> and it needs to be?
Jan 25 15:00:19 <GNUtoo|laptop> I'm not sure
Jan 25 15:00:27 <GNUtoo|laptop> doc pointed to some wiki
Jan 25 15:00:34 <GNUtoo|laptop> where they explain a bit how it works
Jan 25 15:00:45 <GNUtoo|laptop> but If I understood well....
Jan 25 15:00:50 <GNUtoo|laptop> it's not broken yet
Jan 25 15:00:50 <morphis> I thought xload does not need to be signed
Jan 25 15:00:58 <GNUtoo|laptop> on n900 it does
Jan 25 15:01:03 <GNUtoo|laptop> so or you crack that
Jan 25 15:01:08 <morphis> as bootie on palmpre is not signed as far as I know
Jan 25 15:01:11 <GNUtoo|laptop> or you don't replace it
Jan 25 15:01:12 <morphis> hm
Jan 25 15:01:35 <GNUtoo|laptop> then you need a proprietary flasher to "coldboot" something
Jan 25 15:01:57 <GNUtoo|laptop> coldboot is talking to the bootrom to load something
Jan 25 15:01:58 <morphis> aha
Jan 25 15:02:18 <GNUtoo|laptop> so basically the goal for now is to replace nolo by barebox
Jan 25 15:02:22 <GNUtoo|laptop> because there is one more caveat
Jan 25 15:02:25 <GNUtoo|laptop> uboot is too big
Jan 25 15:02:29 <GNUtoo|laptop> and flashing fails
Jan 25 15:02:33 <GNUtoo|laptop> because of that
Jan 25 15:03:44 <HeinervdmWork> morphis: it seems that there are different backends for hciattach in bluez, so one can just add one for the pre
Jan 25 15:03:46 <morphis> barebox?
Jan 25 15:04:06 <morphis> HeinervdmWork: jepp you can do that if the hsuart fits in
Jan 25 15:04:11 <morphis> if not, rewrite it
Jan 25 15:04:25 <morphis> when I took a short look it did not fit in very well
Jan 25 15:06:35 <morphis> GNUtoo|laptop: the picture on http://barebox.org/ is nice :)
Jan 25 15:07:09 <GNUtoo|laptop> the important things are:
Jan 25 15:07:12 <GNUtoo|laptop> *it's very small
Jan 25 15:07:17 <GNUtoo|laptop> so it fits in the 100k
Jan 25 15:07:20 <GNUtoo|laptop> busybox doesn't
Jan 25 15:07:28 <GNUtoo|laptop> busybox is about 200k
Jan 25 15:07:40 <GNUtoo|laptop> and very hard to shrink
Jan 25 15:07:51 <GNUtoo|laptop> and it's easier to work with
Jan 25 15:07:53 <GNUtoo|laptop> kernel style
Jan 25 15:08:10 <GNUtoo|laptop> the bad news is the USB support for omap
Jan 25 15:09:04 <HeinervdmWork> morphis: i will look at it when i'm at home
Jan 25 15:09:11 <morphis> HeinervdmWork: great
Jan 25 15:09:27 <morphis> GNUtoo|laptop: barebox does not have usb support?
Jan 25 15:09:36 <GNUtoo|laptop> it does
Jan 25 15:09:43 <GNUtoo|laptop> but no serial gadget for omap
Jan 25 15:10:40 <morphis> hm
Jan 25 15:10:44 <morphis> not good
Jan 25 15:15:14 <mickey|office> barebox design guidelines sound good
Jan 25 15:16:09 <morphis> GNUtoo|laptop: where do you see the serial gadget is not for omap?
Jan 25 15:17:23 <GNUtoo|laptop> prompt: Arc OTG device core    dep: <choice> && ARCH_IMX
Jan 25 15:17:40 <GNUtoo|laptop> which implements usb peripheral controller
Jan 25 15:19:21 <morphis> whats with USB_GADGET_SERIAL?
Jan 25 15:19:39 <GNUtoo|laptop> when I add it
Jan 25 15:19:41 <GNUtoo|laptop> it does that:
Jan 25 15:20:19 <GNUtoo|laptop> twl4030.c:(.text.usb_composite_register+0x50): undefined reference to `usb_gadget_register_driver'
Jan 25 15:20:33 <GNUtoo|laptop> twl4030.c:(.text.do_mycdev+0x34): undefined reference to `fsl_udc_irq'
Jan 25 15:20:33 <morphis> hm
Jan 25 15:20:38 <GNUtoo|laptop> twl4030.c:(.text.serial_putc+0x68): undefined reference to `fsl_udc_irq'
Jan 25 15:20:41 <GNUtoo|laptop> so....
Jan 25 15:21:08 <morphis> hm
Jan 25 15:21:20 <morphis> do you already have a board file for the n900?
Jan 25 15:21:27 <GNUtoo|laptop> drivers/usb/gadget/fsl_udc.c:int usb_gadget_register_driver(struct usb_gadget_driver *driver)
Jan 25 15:21:43 <GNUtoo|laptop> yes and no, it's nearly unchanged from beagleboard
Jan 25 15:22:42 <GNUtoo|laptop> obj-$(CONFIG_USB_GADGET_DRIVER_ARC) += fsl_udc.o
Jan 25 15:22:52 <GNUtoo|laptop> and DRIVER_ARC depend on imx
Jan 25 15:22:54 <morphis> hm
Jan 25 15:23:02 <morphis> maybe we should ask at the ML
Jan 25 15:25:27 <GNUtoo|laptop> so I remove imx
Jan 25 15:25:33 <GNUtoo|laptop> and it complain about that:
Jan 25 15:25:34 <GNUtoo|laptop> fsl_udc_irq
Jan 25 15:25:45 <GNUtoo|laptop> twl4030.c:(.text.do_mycdev+0x34): undefined reference to `fsl_udc_irq'
Jan 25 15:25:50 <GNUtoo|laptop> which is part of the kernel
Jan 25 15:25:53 <morphis> but it complained about already before you removed the imx part
Jan 25 15:26:00 <GNUtoo|laptop> they are able to import kenrel stuff
Jan 25 15:26:02 <GNUtoo|laptop> which is good
Jan 25 15:26:02 <morphis> you mean the linux kerenL?
Jan 25 15:26:05 <GNUtoo|laptop> yes
Jan 25 15:26:15 <morphis> so at build time they import kernel stuff?
Jan 25 15:26:24 <GNUtoo|laptop> no
Jan 25 15:26:28 <GNUtoo|laptop> at development time
Jan 25 15:26:34 <GNUtoo|laptop> they copy stuff from the kernel
Jan 25 15:26:37 <GNUtoo|laptop> and adapt it
Jan 25 15:28:12 <morphis> ah ok
Jan 25 16:04:18 <PaulFertser> lindi-: meh, couldn't find any considerable amount of snow around here that's dense enough to build an igloo the traditional way. Will try to use some casing to compact the snow for construction.
Jan 25 16:08:00 <CIA-77> freesmartphone.org: 03morphis 07cornucopia * r147d159cc289 10/fsodeviced/src/plugins/backlight_omappanel/plugin.vala: 
Jan 25 16:08:00 <CIA-77> freesmartphone.org: fsodeviced: backlight_omappanel: correct state handling for set_backlight_power
Jan 25 16:08:00 <CIA-77> freesmartphone.org: Signed-off-by: Simon Busch <morphis@gravedo.de>
Jan 25 17:57:21 <CIA-77> freesmartphone.org: 03mickey 07cornucopia * r959d7be7ab62 10/fsodeviced/src/plugins/ (kernel_input/plugin.vala router_alsa/plugin.vala): fsodeviced: fix compiling with gee 0.7
Jan 25 18:06:46 <CIA-77> freesmartphone.org: 03mickey 07libgisi * rb6064d0b9050 10/.gitignore: add .gitignore
Jan 25 18:06:47 <CIA-77> freesmartphone.org: 03mickey 07libgisi * rd0e15adb8a14 10/configure.ac: this project needs C99, so check for it
Jan 25 18:31:07 <playya_> return (GIsiModem *)(void *)(uintptr_t)index; : MMMMM MULTICAST
Jan 25 18:31:47 <mrmoku> heh
Jan 25 18:32:03 <mrmoku> fun :
Jan 25 18:32:04 <mrmoku> P
Jan 25 18:36:48 <mrmoku> gnutoo: nice patch on lkml: [PATCH] bq27x00_battery
Jan 25 18:36:57 <mrmoku> his patch add support for properties POWER_SUPPLY_PROP_CHARGE_NOW,
Jan 25 18:36:57 <mrmoku> POWER_SUPPLY_PROP_CHARGE_FULL, POWER_SUPPLY_PROP_ENERGY_NOW,
Jan 25 18:36:58 <mrmoku> POWER_SUPPLY_PROP_ONLINE in module bq27x00_battery
Jan 25 18:50:50 <gnutoo> mrmoku, ok
Jan 25 19:21:07 <PaulFertser> You might be interested in a highly cricital towards FSF essay about the history of the modern software by Rob Landley: http://landley.net/notes.html#19-07-2010 i have to admit he has a point.
Jan 25 19:21:42 <[Rui]> PaulFertser: without reading it, I doubt he has a point :)
Jan 25 19:21:47 <[Rui]> but I'm loading the page
Jan 25 19:22:42 <PaulFertser> [Rui]: well... i'm a free software lover myself, and i have high respect for RMS and FSF, and some disrespect for the so-called open-source. But it made me think...
Jan 25 19:22:48 <[Rui]> ok, he looks like a fully fledged troll
Jan 25 19:23:27 <PaulFertser> [Rui]: he is clearly biased but still his opinion seems interesting to me...
Jan 25 19:23:28 <[Rui]> if you start your reasoning based upon the absurd, you can derive any fact (that's logic)
Jan 25 19:23:45 <[Rui]> I'd take a truckload of salt before believing anything he says
Jan 25 19:23:58 <[Rui]> if one can spot lies so early in the text, I can't imagine the rest
Jan 25 19:24:13 <PaulFertser> [Rui]: what's exactly a lie?
Jan 25 19:24:14 <[Rui]> «In 1998 he switched to trying to claim credit for Linux, ignoring not just Linus's work but the contributions of Tanenbaum and comp.os.minix community, the Berkeley developers, MIT's project Athena, and many others.»
Jan 25 19:24:42 <PaulFertser> [Rui]: agreed, that doesn't sound correct at all. Literally speaking.
Jan 25 19:24:52 <[Rui]> 2nd paragraph
Jan 25 19:25:27 <[Rui]> «Stallman's own Gnu/Hurd project failed to produce a usable clone»
Jan 25 19:25:33 <PaulFertser> [Rui]: but look at it another way: it's true everybody said "linux, linux". FSF clearly wanted a wider recognition. So they started to insist they have a share.
Jan 25 19:26:25 <[Rui]> PaulFertser: well, anyone saying "Linux" as the whole, should say kernel32.dll rather than Windows, Mach rather than MacOS, never use the word "Red Hat", Fedora, Debian, etc... 
Jan 25 19:26:32 <[Rui]> not if they want to be coherent :)
Jan 25 19:27:38 <[Rui]> so if you built a project aimed at bettering mankind, and then some kids added a small (albeit important) piece and started calling the wholy by that piece, wouldn't you insist for recognition?
Jan 25 19:27:52 <PaulFertser> [Rui]: if you think i didn't read GNU/Linux FAQ at gnu.org, you're wrong. I read it several times (and even translated half of it to russian).
Jan 25 19:28:25 <[Rui]> I don't think you didn't :) don't worry. Just commenting. All this tlak is, though, a bit offtopic here :)
Jan 25 19:28:33 <[Rui]> s/tlak/talk/
Jan 25 19:28:33 <apt> [Rui] meant: I don't think you didn't :) don't worry. Just commenting. All this talk is, though, a bit offtopic here :)
Jan 25 19:29:32 <[Rui]> «To this day, the FSF requires a physical copyright assignment statement be filed (on paper) with the FSF before accepting patches from a new developer»
Jan 25 19:29:47 <[Rui]> no shit, sherlock. how do you want a contract to be legal and valid? :)
Jan 25 19:33:38 <[Rui]> flame bait all around :)
Jan 25 19:35:28 <PaulFertser> [Rui]: yeah.
Jan 25 19:35:39 <PaulFertser> [Rui]: but it doesn't feel to me that it's all totally void.
Jan 25 19:36:01 <[Rui]> you only need to distort here and there
Jan 25 19:36:12 <[Rui]> add a few facts in the middle so you seem legit
Jan 25 19:55:57 <mrmoku> PaulFertser: interesting read
Jan 25 20:02:48 <PaulFertser> I guess i know need to buy an FSF membership to compensate the damage :)
Jan 25 21:18:23 <CIA-77> freesmartphone.org: 03felix.huber 07zhone * r6c53cde43c33 10/src/zhone: add page turning for profiles and page wrap
Jan 26 02:01:11 <DocScrutinizer> funnny there's no post since 9h
Jan 26 02:04:07 <DocScrutinizer> interesting PoV related to FOSS
Jan 26 02:06:56 <DocScrutinizer> btw it probably is Tannenbaum (with double-N), and I actively participated in that process
Jan 26 02:10:22 <DocScrutinizer> arguing about nomenclature and ""history"" is moot anyway
**** ENDING LOGGING AT Wed Jan 26 02:59:57 2011