**** BEGIN LOGGING AT Fri Jun 08 02:59:59 2012
Jun 08 07:59:42 <prahal> gasp ... the marvell sd8686 + libertas driver does not support EAP as far as running wpa_supplicant shows
Jun 08 07:59:52 <prahal> ie running on desktop then on phone
Jun 08 08:00:04 <prahal> works only on desktop
Jun 08 08:00:22 <prahal> I love the do everything in firmware
Jun 08 08:00:51 <prahal> especially when the firmware is not featurefull 
Jun 08 08:07:11 <prahal> though my setup is the most complex one can imagine ... it is so easy to setup nowadays that it might well become widespread
Jun 08 08:07:40 <prahal> openvswitch(vlan) + hostapd + freeradius
Jun 08 08:18:43 <MMlosh> one more lame question:  what makes a package appear in Core?  what should I do if I need something else  (speaking of netstat, espeak / flite).  is it pulling from unstable or not?
Jun 08 08:27:44 <CIA-76> freesmartphone.org: 03morphis 07libfso-glib * r4cec5139e6a5 10/configure.ac: Keep up with fso-specs version
Jun 08 08:27:44 <CIA-76> freesmartphone.org: 03morphis 07specs * rad9c36467a25 10/org.freesmartphone.GSM/org.freesmartphone.GSM.Call.xml.in: org.freesmartphone.GSM.Call: reorder content; methods come first then signals
Jun 08 08:27:44 <CIA-76> freesmartphone.org: 03morphis 07specs * ra161400b7f19 10/org.freesmartphone.GSM/org.freesmartphone.GSM.Call.xml.in: org.freesmartphone.GSM.Call: rework transfer method and add new method name for call deflection
Jun 08 08:27:44 <CIA-76> freesmartphone.org: 03morphis 07specs * r3b9e0f367e90 10/configure.ac: Bump version
Jun 08 08:38:28 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * rd9fa4a4344c0 10/fsogsmd/src/ (lib/mediator.vala plugins/dbus_service/gsm_call_service.vala): fsogsmd: dbus_service: adjust for recent API changes
Jun 08 08:38:29 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * rc10d9e627c15 10/fsogsmd/src/lib/ (at/atcall.vala at/atcallmediators.vala call.vala): fsogsmd: lib: at: implement call transfer API
Jun 08 08:51:46 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * rd3626a1167dc 10/fsogsmd/src/lib/at/atmediatorhelpers.vala: fsogsmd: lib: at: activate call transfer mediator
Jun 08 08:52:55 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * r1257533eaaae 10/fsogsmd/src/lib/at/atmediatorhelpers.vala: fsogsmd: lib: at: activate call transfer mediator
Jun 08 09:39:18 <prahal> me revert what I told about the marvell fw until I fully test a new clue
Jun 08 09:39:52 <prahal> ie bug in the kernel : http://code.google.com/p/chromium-os/issues/detail?id=15981
Jun 08 10:48:54 <prahal> well the whole code around that fix went to /dev/null ... no more auth tracking in the kernel
Jun 08 10:54:54 <prahal> thus back to the drawing board ... the wifi driver is too slow , there is not enough entropy on the hostap side or wpa_supplicant runs too slow on the phone and need a higher authentication timeout  ... or all of those at the same time
Jun 08 16:41:56 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * rfc6a34713b51 10/fsogsmd/ (src/lib/at/atcommands.vala tests/testatcommand.vala): fsogsmd: lib: at: add implementation for +CTFR AT command
Jun 08 16:41:57 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * re18752af5138 10/fsogsmd/src/lib/ (4 files in 2 dirs): fsogsmd: lib: at: implement call deflection API
Jun 08 16:41:57 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * r4fdef7f6e2bb 10/fsogsmd/src/plugins/modem_samsung/callhandler.vala: fsogsmd: modem_samsung: adjust for recent call handler API changes
Jun 08 16:41:58 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * r845bb7929ff2 10/fsogsmd/src/ (4 files in 4 dirs): fsogsmd: lib: move common call handler functionality in the abstract base class
Jun 08 16:41:58 <CIA-76> freesmartphone.org: 03morphis 07cornucopia * r1c632e324778 10/fsogsmd/src/plugins/modem_nokia_isi/callhandler.vala: fsogsmd: modem_nokia_isi: fixup call handler after API changes
Jun 08 16:46:38 <GNUtoo> ah nice thanks
Jun 08 17:51:53 <GNUtoo> morphis, hi
Jun 08 17:52:03 <GNUtoo> I want to add a dep of giounix
Jun 08 17:52:10 <GNUtoo> to fsoaudiod
Jun 08 17:52:15 <GNUtoo> so I did that:
Jun 08 17:52:31 <GNUtoo> --pkg gio-unix-2.0
Jun 08 17:52:35 <GNUtoo> in the Makefile.am
Jun 08 17:52:41 <GNUtoo> but still I've got:
Jun 08 17:52:46 <GNUtoo> cmthandler.c:28:34: fatal error: gio/gunixinputstream.h: No such file or directory
Jun 08 17:52:53 <GNUtoo> tough pkgconfig says:
Jun 08 17:53:51 <GNUtoo> pkg-config  --libs  --cflags gio-unix-2.0 => --pthread  -I/home/gnutoo/embedded/oe/oe-core/shr/oetmps/build/sysroots/om-gta04/usr/include/gio-unix-2.0/ [...]
Jun 08 17:57:36 <GNUtoo> ah right htere are stuff below
Jun 08 17:57:39 <GNUtoo> (split screen)
Jun 08 17:57:51 <GNUtoo>         $(GLIB_CFLAGS) \
Jun 08 17:57:53 <GNUtoo> etc...
Jun 08 18:01:14 <morphis> GNUtoo: yes you need to adjust cflags as well
Jun 08 18:01:22 <GNUtoo> ok
Jun 08 18:01:27 <GNUtoo> what's the name of the .m4?
Jun 08 18:01:38 <GNUtoo> like PKG_CHECK_MODULES( ?
Jun 08 18:02:31 <morphis> yes
Jun 08 18:02:37 <morphis> there should be already one for glib
Jun 08 18:03:09 <GNUtoo> is it GIO ?
Jun 08 18:03:19 <GNUtoo> GIO_UNIX?
Jun 08 18:05:33 <GNUtoo> I mean I've read the .m4 and I didn't find the answer
Jun 08 18:05:43 <GNUtoo> for instance there is
Jun 08 18:05:44 <GNUtoo> PKG_CHECK_MODULES(GEE,
Jun 08 18:05:46 <GNUtoo> for GEE
Jun 08 18:05:52 <GNUtoo> but for gio-unix?
Jun 08 18:06:19 <morphis> which m4 do you read?
Jun 08 18:06:23 <morphis> it's all in configure.ac
Jun 08 18:07:01 <GNUtoo> /home/gnutoo/embedded/oe/oe-core/shr/oetmps/build/sysroots/nokia900/usr/share/aclocal/glib-2.0.m4
Jun 08 18:07:43 <morphis> GNUtoo: something like http://pastie.org/4051842
Jun 08 18:08:02 <morphis> that should be all
Jun 08 18:08:21 <GNUtoo> so no stuff like http://pastie.org/private/nyk4c3xffr6hjubux98bwa
Jun 08 18:08:42 <morphis> it's the same
Jun 08 18:08:47 <GNUtoo> ah ok
Jun 08 18:08:50 <morphis> but it's glib so add it to GLIB_CFLAGS
Jun 08 18:09:58 <GNUtoo> thanks a lot
Jun 08 18:11:20 <morphis> no problem
Jun 08 19:13:18 <GNUtoo> morphis, can you look in #vala?
Jun 08 19:22:47 * DocScrutinizer05 curses KDE4
Jun 08 19:23:07 <DocScrutinizer05> if only for this dolphin abomination
Jun 08 19:24:17 <DocScrutinizer05> I mean, what's wrong with those guys, to replace feature-rich mature software like konqueror by crippled POC/POS like dolphin - and then call that progress?!
Jun 08 19:25:38 <DocScrutinizer05> kde4 windows manager constantly fails to refresh parts of taskbar
Jun 08 19:26:09 <DocScrutinizer05> so I have to move mouse cursur over it to make it refresh
Jun 08 19:26:52 <DocScrutinizer05> (I already suspected Xorg, but then why would it only fail for windowsbar and taskbar)
Jun 08 19:27:29 <DocScrutinizer05> and then the THEME, OMG. I think I bought a Lenovo with B&W LCD
Jun 08 19:27:56 <DocScrutinizer05> actually I have to search for anything not gray in gray
Jun 08 19:28:33 <DocScrutinizer05> plasma - PFFFFF!
Jun 08 19:29:53 <DocScrutinizer05> Kontact gone
Jun 08 19:30:05 <DocScrutinizer05> KTNX
Jun 08 19:31:59 * DocScrutinizer05 hopes for KDE6 maybe reaching same maturity as KDE3 had, sometime in the future, maybe 2023
Jun 08 19:32:53 <DocScrutinizer05> ~lart qt4
Jun 08 19:32:53 * apt teaches qt4 that M$ Access is a database. No, really, a database. A real live multi-user... well, ok, not multi-user, but a database. Yeah, that sounds right.
Jun 08 19:33:00 <Alex[sp3dev]> i would like that there is a special utility to configure settings but you cannot change settings from apps. so that you can set it up once and for all and the setup won't break because you click in the wrong place. kde3 at least had kiosk mode
Jun 08 19:33:53 <DocScrutinizer05> kde4 is a toy for kids
Jun 08 19:34:21 <Alex[sp3dev]> kde4 is a fractal of abstractions
Jun 08 19:34:25 <DocScrutinizer05> hotkey for ding? not working. Great!
Jun 08 19:35:07 <Alex[sp3dev]> maybe you have ideas what to write a rant about? i need to populate my hateblog now that i have some time
Jun 08 19:38:07 <DocScrutinizer05> window hotkeys neither rise the window, nor start the app when not started (basically they do nuttin), nor get they stored persistently so are vanished after next session start
Jun 08 19:38:19 <Alex[sp3dev]> how the hell is gcc inline asm working.. my brain hurts
Jun 08 19:39:15 <DocScrutinizer05> so obviously the whole hotkey feature is completely broken or fake in KDE4
Jun 08 19:39:23 <DocScrutinizer05> which basically kills me
Jun 08 19:39:34 <Alex[sp3dev]> i feel i can't go back to floating wm after using wmii and i3
Jun 08 19:40:59 <DocScrutinizer05> I used to use ding with hotkey, and whenever I felt like I highlighted a word somewhere and hit ding's hotkey to get the translation. Now this needs some mouse action to switch to ding window 'manually'
Jun 08 19:41:46 <DocScrutinizer05> I already forgot all the maximize/minimize/whatnot-else hotkeys I got used to on KDE3
Jun 08 19:42:02 <DocScrutinizer05> FSCK U KDE4
Jun 08 19:43:19 <DocScrutinizer05> well, I shouldn't get that upset. Should've known my hopes for kde4 would work _better_ on a clean fresh install rather than a fsckdup upgrade from kde3... were silly
Jun 08 19:43:23 <PaulFertser> Alex[sp3dev]++
Jun 08 19:43:40 <Alex[sp3dev]> PaulFertser: tried i3? it behaves nicely with multiple monitors
Jun 08 19:44:05 <PaulFertser> Alex[sp3dev]: i'm using ion3 right atm but on all other machines i'm gradually switching to XMonad.
Jun 08 19:45:48 <PaulFertser> Also, i have only one monitor everywhere.
Jun 08 19:46:51 <PaulFertser> DocScrutinizer05: gnome users do not seem to be feeling any better than you, so that looks like some plague for all the desktop environment users.
Jun 08 19:54:37 <morphis> JaMa: ping
Jun 08 19:56:18 <DocScrutinizer05> PaulFertser: indeed it's a plague in whole computer/IT scene: make stuff ""simpler"" and ever more like a disgusting blend of the worst parts of macOS and Windoze
Jun 08 19:56:38 <DocScrutinizer05> I wonder when we get tiles in KDE, and nothing but tiles
Jun 08 20:01:02 <DocScrutinizer05> while until maybe 5 years ago linux development been driven by professionals that wanted to do something cute in their free time and knew from their jobs all the bright parts and many downsides of what existed outside linux world, and were not interested in stupidly duplicating either - nowadays it seems 90+% of development is done by windows/mac kids that simply think "with a bit of work I can make this linux the *cheaper* 
Jun 08 20:01:04 <DocScrutinizer05> alternative/replacement to what I love so much for an OS right now"
Jun 08 20:04:22 <DocScrutinizer05> and those kids *will* bring tiles to gnome and kde, as otherwise you might spot on first glance this is *NOT* uebercool windows7 - I guess that's all they are afraid of, that somebody would make them blush on commenting about this crappy linux doesn't have acmetiles/whatever like a real OS
Jun 08 20:05:15 <Alex[sp3dev]> btw I installed windows 8 on a vm today (needed it for omapflash). the UI is total disaster. I wonder how long it takes before ubuntu guys copy that
Jun 08 20:05:44 <DocScrutinizer05> not long, see my last post
Jun 08 20:07:01 <DocScrutinizer05> shuttleworth has the idiotic idea to imitate rather than differentiate
Jun 08 20:08:04 <DocScrutinizer05> and all the other distros follow - no f'ing clue why
Jun 08 20:09:48 <DocScrutinizer05> probably because all the main core app devel teams are looking at canonical stuff way too much, so when KDE4 comes with tiles tomorrow, basically not much any distro can do about it
Jun 08 20:12:14 <DocScrutinizer05> well, opensuse re-installed support for KDE3 with their recent release
Jun 08 20:12:26 <DocScrutinizer05> there must be a reason for that
Jun 08 21:08:26 <morphis> GNUtoo: I think we should switch back to this channel for further discussion
Jun 08 21:09:40 <morphis> hm gta04.org is still down
Jun 08 21:10:04 <GNUtoo> morphis, ok
Jun 08 21:10:24 <GNUtoo> http://www.pastie.org/private/yfthwm153ivyj872soo32a
Jun 08 21:10:32 <GNUtoo> that's a new failure
Jun 08 21:13:04 <GNUtoo> morphis, or maybe I push newer vala in a branch
Jun 08 21:13:13 <GNUtoo> I use it only for fsoaudiod
Jun 08 21:13:16 <GNUtoo> to see if it works
Jun 08 21:13:20 <GNUtoo> and we fix right after?
Jun 08 21:14:37 <morphis> sounds good to me
Jun 08 21:15:19 <GNUtoo> ok
Jun 08 21:15:20 <morphis> GNUtoo: maybe skip fsotdld
Jun 08 21:15:24 <GNUtoo> I'll also fix some null stuff
Jun 08 21:15:30 <GNUtoo> because the newer failure were in vapi
Jun 08 21:15:42 <GNUtoo> the ones only for devshell
Jun 08 21:17:50 <GNUtoo> ah the vapi seem generated automatically
Jun 08 21:17:58 <GNUtoo> maybe it would be better if you did the fix
Jun 08 21:18:08 <GNUtoo> because I don't feel very confortable with that stuff
Jun 08 21:18:26 <GNUtoo> I'll push my fix tough
Jun 08 21:18:31 <GNUtoo> since it doesn't touch the vapis
Jun 08 21:19:38 <CIA-76> freesmartphone.org: 03GNUtoo 07cornucopia * ra2fadf980add 10/libfsobasics/fsobasics/threading.vala: 
Jun 08 21:19:38 <CIA-76> freesmartphone.org: fsobasics: fix threading compilation with vala 0.16.0
Jun 08 21:19:38 <CIA-76> freesmartphone.org: Without that fix we have:
Jun 08 21:19:38 <CIA-76> freesmartphone.org:  threading.vala:77.53-77.55: error: Cannot convert from `int' to `void*'
Jun 08 21:19:38 <CIA-76> freesmartphone.org:  void* param = 0x0 )
Jun 08 21:19:38 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Jun 08 21:20:58 <GNUtoo> morphis, how to change the vapis?
Jun 08 21:21:03 <GNUtoo> it's generated with fso-specs?
Jun 08 21:21:14 <morphis> why?
Jun 08 21:21:37 <GNUtoo> because some contains the exact same error:
Jun 08 21:21:40 <GNUtoo> param = 0x0
Jun 08 21:21:52 <GNUtoo> since I pushed I'll rebuild and see
Jun 08 21:23:56 <morphis> ok
Jun 08 21:27:36 <Alex[sp3dev]> haha. NEXUS PWNED finally. got uboot running. hope the next week @ work i'll make it boot instead of SBL, not instead of zImage
Jun 08 21:28:26 <Alex[sp3dev]> http://piccy.info/view3/3119067/8156d1868fac2739cd27bd08a7660e0f/orig/ seems useless without external microsd though
Jun 08 21:30:22 <GNUtoo> nice
Jun 08 21:31:35 <Alex[sp3dev]> the good thing is that xloader doesn't seem to check signature and omapflash can revive the brick from usb. so i think it won't be hard to flash it to emmc
Jun 08 21:32:26 <GNUtoo> wow!!!!
Jun 08 21:33:05 <GNUtoo> ah but xloader is signed tough
Jun 08 21:33:30 <Alex[sp3dev]> but at least we can disassemble xloader and verify it has no malware. it's not that large to disassemble it
Jun 08 21:33:36 <GNUtoo> ok nice!!!!
Jun 08 21:41:39 <morphis> I am off
Jun 08 21:41:41 <morphis> gn8
Jun 08 21:53:41 <DocScrutinizer05> GNUtoo: afaik on OMAP the xloader needs to be signed, or ROMBOOT will not execute it
Jun 08 21:53:57 <GNUtoo> really?
Jun 08 21:54:04 <GNUtoo> what about beagleboard, gta04 etc...
Jun 08 21:54:09 <GNUtoo> they got a dummy signature
Jun 08 21:54:19 <GNUtoo> with gpsign or something like that
Jun 08 21:54:21 <DocScrutinizer05> on xloader?
Jun 08 21:54:24 <GNUtoo> yes
Jun 08 21:54:26 <GNUtoo> MLO
Jun 08 21:54:34 <DocScrutinizer05> hmm
Jun 08 21:54:42 <GNUtoo> it's signed but you can load your modified versions
Jun 08 21:55:18 <DocScrutinizer05> might depend on hw-selected bootmode
Jun 08 21:55:22 <GNUtoo> indeed
Jun 08 21:56:08 <DocScrutinizer05> I've been told on N900 we can't replace xloader, as it needs signature or won't get executed
Jun 08 21:56:46 <GNUtoo> that's true
Jun 08 21:56:51 <DocScrutinizer05> the guy telling me might be wrong
Jun 08 21:56:52 <GNUtoo> I think there is a secure mode
Jun 08 21:57:00 <GNUtoo> it's like fuses or something like that
Jun 08 21:57:04 <DocScrutinizer05> yes, there's HS
Jun 08 21:57:18 <GNUtoo> basically it depedn on the device
Jun 08 21:57:32 <GNUtoo> some SOC are in secure mode
Jun 08 21:57:36 <GNUtoo> some other aren't
Jun 08 21:57:36 <DocScrutinizer05> but on N900 it's not used
Jun 08 21:57:42 <GNUtoo> what's HS?
Jun 08 21:57:49 <DocScrutinizer05> High Security
Jun 08 21:57:56 <GNUtoo> ah if it's not used it's nice
Jun 08 21:58:14 <GNUtoo> maybe the guy got afraid by some text on the serial console?
Jun 08 21:58:22 <GNUtoo> like from nolo or something that said signed
Jun 08 21:58:25 <DocScrutinizer05> nope, probably not
Jun 08 21:58:28 <GNUtoo> ok
Jun 08 21:58:44 <DocScrutinizer05> jacekowski isn't afraid of anything ;-D
Jun 08 21:58:56 <GNUtoo> ok
Jun 08 21:59:17 <GNUtoo> so how did he concolude that it was impossible to replace?
Jun 08 21:59:37 <GNUtoo> do we have a way to usbload something trough a free uttility(not flasher)
Jun 08 21:59:46 <DocScrutinizer05> he connected serial comsole to N900 testpoints (actually he found out which are the testpoints), and fried his UART during the enterprise
Jun 08 22:00:05 <GNUtoo> like by modifying the omap SOC loading uttilities
Jun 08 22:00:06 <GNUtoo> ok
Jun 08 22:00:07 <DocScrutinizer05> he claimed it's written in spruf98
Jun 08 22:00:27 <DocScrutinizer05> omap3430 TRM
Jun 08 22:01:21 <DocScrutinizer05> well, 0xffff should be able to do that (load sth thru usb), afaik
Jun 08 22:02:23 <DocScrutinizer05> nokia flasher will also upload rather arbitrary images afaik
Jun 08 22:02:37 <DocScrutinizer05> for coldflashing
Jun 08 22:04:02 <GNUtoo> DocScrutinizer05, no I mean coldflashing
Jun 08 22:04:20 <GNUtoo> I wanted something free to talk to the bootrom
Jun 08 22:04:22 <GNUtoo> like pusb
Jun 08 22:04:26 <GNUtoo> or pserial
Jun 08 22:04:32 <GNUtoo> only that they don't work on n900
Jun 08 22:04:39 <GNUtoo> (pusb, pserial not tried)
Jun 08 22:05:58 <DocScrutinizer05> I think ROMBOOT format is razther simple and completely documented in spruf98
Jun 08 22:06:24 <DocScrutinizer05> something like kermit or such
Jun 08 22:07:19 <DocScrutinizer05> the basic protocol is (who might have guessed) probably DFU
Jun 08 22:07:42 <GNUtoo> DocScrutinizer05, ok but the secure mode is the same or not?
Jun 08 22:07:46 <DocScrutinizer05> which is specified by USB group
Jun 08 22:07:51 <GNUtoo> because pusb does work I think
Jun 08 22:08:07 <GNUtoo> but not on n900....even by changing the USB ID
Jun 08 22:08:10 <Alex[sp3dev]> DocScrutinizer05: xloader needs to be signed on HS. we can use samsung xloader because it doesn't check uboot validity (although i need to recheck that)
Jun 08 22:08:11 <DocScrutinizer05> secure mode doesn't kick in on this level
Jun 08 22:08:27 <DocScrutinizer05> it kicks in after loading xloader/MLO to RAM
Jun 08 22:09:15 <DocScrutinizer05> well, N900 xloader also doesn't check
Jun 08 22:09:35 <Alex[sp3dev]> so what? are there checks after xloader?
Jun 08 22:09:45 <DocScrutinizer05> nope
Jun 08 22:10:03 <DocScrutinizer05> HS not used on N900
Jun 08 22:10:53 <Alex[sp3dev]> i have an HS device (galaxy nexus). but SBL (uboot) doesn't have any signature so I think I can shove any binary up the mmc. will try when I'll be at work PC which has the tools
Jun 08 22:15:43 <DocScrutinizer05> :nod:
Jun 08 22:17:23 <DocScrutinizer05> first simple foolproof test: change the string "uBoot" in uBoot to "UbOOT" (or any other insignificant change of any arbitrary byte that doesn't break things) - then see if that tainted uBoot still works
Jun 08 22:17:35 <Alex[sp3dev]> sure, thought of that but was busy to check
Jun 08 22:18:27 <Alex[sp3dev]> you see, i need to reboot to windows to use omapflash. dunno if i can boot an HS device via usb in linux, but my main target is getting uboot running, not rewriting all omap software
Jun 08 22:18:28 <DocScrutinizer05> after all theoretically MLO could have a MD5sum of uBoot hardcoded, to check
Jun 08 22:19:45 <DocScrutinizer05> this way uBoot wouldn't need any signature but still you can't change a byte
Jun 08 22:20:09 <Alex[sp3dev]> how is MLO verified after all? where are OEM keys stored? is there some TI master key which would allow to boot a signed binary on any OEM device? 
Jun 08 22:20:33 <DocScrutinizer05> all good questions
Jun 08 22:20:40 <DocScrutinizer05> not entirely clear
Jun 08 22:20:52 <DocScrutinizer05> except for the "where are they stored" part
Jun 08 22:21:03 <DocScrutinizer05> which I think is EFUSE
Jun 08 22:21:10 <DocScrutinizer05> whatever that means
Jun 08 22:21:29 <DocScrutinizer05> some OTP area in ROM
Jun 08 22:22:46 <Alex[sp3dev]> good god seems there's no md5 hardcoded. but i'm better off checking by modding SBL since I've already verified omapflash can unbrick after flashing crap
Jun 08 22:23:13 <DocScrutinizer05> and I think MLO actually needs a real mature signature, with a public key where the private key is in EFUSE (or swap public and private to your liking)
Jun 08 22:24:42 <Alex[sp3dev]> yes that would make sense. let's just hope samsung xloader will accept any crap
Jun 08 22:27:14 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * r6ba420d53d28 10cornucopia/fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/ (cmthandler.vala plugin.vala): (log message trimmed)
Jun 08 22:27:14 <CIA-76> freesmartphone.org: fsoaudiod: rework the n900 forwarder
Jun 08 22:27:14 <CIA-76> freesmartphone.org: Fow now only writting the incomming audio to a file works.
Jun 08 22:27:14 <CIA-76> freesmartphone.org: But it works well(buffer underruns well handled etc...).
Jun 08 22:27:14 <CIA-76> freesmartphone.org: The code has been imported from the modified cmtspeechtest
Jun 08 22:27:14 <CIA-76> freesmartphone.org:  uttility which worked well.
Jun 08 22:27:15 <CIA-76> freesmartphone.org: The fsoaudiod forwarder didn't work well even without alsa.
Jun 08 22:27:15 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * r8c0e217e8480 10cornucopia/fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/cmthandler.vala: 
Jun 08 22:27:16 <CIA-76> freesmartphone.org: fsoaudiod: gsmvoice_alsa_cmtspeechdata: fix the runRecordThread variable atomicity.
Jun 08 22:27:16 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Jun 08 22:27:17 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * r09eb99e0ab84 10cornucopia/fsoaudiod/ (3 files in 2 dirs): 
Jun 08 22:27:17 <CIA-76> freesmartphone.org: fsoaudiod: n900 forwarder: switch to async
Jun 08 22:27:18 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Jun 08 22:27:20 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * r5c70d0fb3ea9 10cornucopia/fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/cmthandler.vala: 
Jun 08 22:27:20 <CIA-76> freesmartphone.org: fsoaudiod: nokia n900 forwarder : add new function from #vala
Jun 08 22:27:20 <CIA-76> freesmartphone.org: Thanks juergbi for read_from_modem_and_write_to_file on #vala on GIMPNet IRC servers:
Jun 08 22:27:33 <GNUtoo> note that it's in a branch
Jun 08 22:28:29 <CIA-76> freesmartphone.org: fsoaudiod: n900 forwarder: privatize some functions
Jun 08 22:28:29 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Jun 08 22:28:29 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * ra0c857cac985 10cornucopia/fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/ (cmthandler.vala plugin.vala): 
Jun 08 22:28:29 <CIA-76> freesmartphone.org: fsoaudiod: n900 forwarder: make it more object oriented.
Jun 08 22:28:29 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Jun 08 22:30:30 <GNUtoo> Basically yesterday I tried the forwarder again and I found that the xrun/buffer-underrun/EPIPE were not due to alsa
Jun 08 22:30:43 <GNUtoo> but rather to how the modem lib was used
Jun 08 22:34:23 <DocScrutinizer05> sure thing
Jun 08 22:35:04 <DocScrutinizer05> I told you there's still a flaw probably in the way you (don't) check/sync-to timing msgs of modem
Jun 08 22:37:00 <DocScrutinizer05> I just completely fail to understand why Nokia decided modem was clock slave and not master, for recording
Jun 08 22:38:38 <GNUtoo> ok
Jun 08 22:38:46 <DocScrutinizer05> anyway there's afaik messages telling the CPU about if last 5ms segment was early or late, and you need to tune to that
Jun 08 22:39:03 <GNUtoo> CMTSPEECH: nokiamodem_backend: read bytes AB:63:00:41, control channel message (TIMING_CONFIG_NTF).
Jun 08 22:39:07 <GNUtoo> I've lots of stuff like that
Jun 08 22:39:11 <GNUtoo> any idea on what they means?
Jun 08 22:39:24 <DocScrutinizer05> that are the timing messages :-)
Jun 08 22:39:33 <GNUtoo> yes I bet that
Jun 08 22:39:39 <GNUtoo> but what do they indicate
Jun 08 22:39:51 <GNUtoo> do they for instance indicate how much I'm drifting?
Jun 08 22:39:57 <GNUtoo> or should I reply to them
Jun 08 22:40:02 <DocScrutinizer05> sorry, I don't know the bits in these msgs by name
Jun 08 22:40:13 <GNUtoo> ok
Jun 08 22:40:15 <GNUtoo> neither do I
Jun 08 22:40:32 <GNUtoo> I should look at libcmtspeech then
Jun 08 22:40:38 <DocScrutinizer05> no, you proabably don't reply, you just adjust your drift
Jun 08 22:40:49 <GNUtoo> ok
Jun 08 22:41:03 <GNUtoo> but for instance for EPIPE the lib ajust his drift....
Jun 08 22:41:35 <GNUtoo> apart a drrr it seem better now
Jun 08 22:41:54 <GNUtoo> ah still some EPIPE eareable
Jun 08 22:48:29 <GNUtoo>                 } else if (cmtevent.state == CMTSPEECH_STATE_ACTIVE_DLUL &&cmtevent.msg_type == CMTSPEECH_TIMING_CONFIG_NTF) { update_uplink_frame_timing(u, &cmtevent);pa_log_debug("updated UL timing params");
Jun 08 22:48:39 <GNUtoo> that's for uplink it seems
Jun 08 22:48:44 <GNUtoo> and I don't use uplink yet
Jun 08 22:54:59 <GNUtoo> oh
Jun 08 22:55:01 <GNUtoo> CMTSPEECH: nokiamodem_backend: DL buffer overrun (mmaphw 127, hw 124, appl 110, slot 1, count 3, hwdelay 17).
Jun 08 22:55:06 <GNUtoo> it's not underrun
Jun 08 22:55:09 <GNUtoo> but overrun
Jun 08 22:56:30 <GNUtoo> http://www.pastie.org/private/lvaeuophzdml3wvngscgg
Jun 08 22:58:08 <GNUtoo> ah ok it's an EPIPE still
Jun 08 22:58:30 <GNUtoo> http://www.pastie.org/private/lovlym6zffeys8siba6wg
Jun 08 22:58:44 <GNUtoo> hmmm
Jun 08 22:59:00 <GNUtoo> let me look if the async stuff is realtime
Jun 09 00:18:59 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * r5dc8d73b2333 10cornucopia/fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/cmthandler.vala: 
Jun 09 00:18:59 <CIA-76> freesmartphone.org: fsoaudiod: n900 forwarder: cleanup: remove comented code.
Jun 09 00:18:59 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Jun 09 00:19:00 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * r965838eb72e8 10cornucopia/fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/cmthandler.vala: 
Jun 09 00:19:00 <CIA-76> freesmartphone.org: fsoaudiod: n900 forwarder: reorder some functions to make the critical section executed before.
Jun 09 00:19:00 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
Jun 09 00:19:01 <CIA-76> freesmartphone.org: 03GNUtoo 07gnutoo/n900-better-forwarder * r249eb97adbd6 10cornucopia/fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/cmthandler.vala: 
Jun 09 00:19:01 <CIA-76> freesmartphone.org: fsoaudiod: n900 forwarder: move file creation/deletion at beginning of call
Jun 09 00:19:02 <CIA-76> freesmartphone.org: The file creation/deletion were moved in the beginning of the call instead
Jun 09 00:19:02 <CIA-76> freesmartphone.org:  of at the start of the deamon in order to have a new file for each call,
Jun 09 00:19:03 <CIA-76> freesmartphone.org:  because else it appends at the end of the file after the previous call...
Jun 09 00:19:03 <CIA-76> freesmartphone.org: Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org>
**** ENDING LOGGING AT Sat Jun 09 02:59:59 2012