**** BEGIN LOGGING AT Fri Dec 09 02:59:58 2011
Dec 09 03:20:06 <karl333> lol, bitbake... I don't get it. last year I simply changed some recipes and it worked for me, now I can't even get started. also, you really should take a look at your build-shr page because it seems.... incomplete. 
Dec 09 06:24:39 <GNUtoo> JaMa|Off, hi it seem that toolchain broke crespo+nokia900 kernels
Dec 09 06:24:51 <GNUtoo> they don't boot anymore
Dec 09 06:34:09 <CIA-82> SHR: 03GNUtoo 07meta-smartphone * r32329e2cfbf2 10/meta-nokia/recipes-core/base-files/ (base-files/nokia900/fstab base-files_3.0.14.bbappend): meta-nokia: override /etc/fstab for mounting the internal 32GB
Dec 09 06:34:19 <CIA-82> SHR: 03GNUtoo 07meta-smartphone * ra07a02bcc5dc 10/meta-nokia/recipes-kernel/linux/linux-nokia900-meego_git.bb: meta-nokia: linux-nokia900-meego: replace PR by a comment telling where to bump PR
Dec 09 06:35:40 <mrmoku> GNUtoo: really? shit :/
Dec 09 06:36:05 <GNUtoo> mrmoku, yes I spent the whole day of yesterday trying to understand why it was not booting
Dec 09 06:36:17 <GNUtoo> first crespo => I abandoned
Dec 09 06:36:38 <GNUtoo> then nokia900 => I reverted my changes this morning and it can be only the toolchain
Dec 09 06:39:15 <GNUtoo> what should we do?
Dec 09 06:40:35 <mrmoku> GNUtoo: does the buildhost already have a n900 kernel built with the same toolchain?
Dec 09 06:40:44 <mrmoku> just to exclude some build madness
Dec 09 06:41:08 <GNUtoo> I don't think so
Dec 09 06:41:15 <GNUtoo> I didn't push my defconfig changes
Dec 09 06:41:24 <mrmoku> ok, we should here what JaMa|Off says
Dec 09 06:41:28 <mrmoku> s/here/hear/
Dec 09 06:41:29 <apt> mrmoku meant: ok, we should hear what JaMa|Off says
Dec 09 06:41:29 <GNUtoo> so the kernel didn't get rebuild on buildhost
Dec 09 06:41:38 <GNUtoo> but I need to push some changes
Dec 09 06:41:42 <GNUtoo> HIDP is lacking
Dec 09 06:41:49 <mrmoku> ok
Dec 09 06:41:50 <GNUtoo> and I need that to try to fix bluetooth
Dec 09 06:42:23 <mrmoku> GNUtoo: want me to try if you kernel boots for me?
Dec 09 06:42:25 <GNUtoo> I mean I can scan etc...but I'm not sure bluetooth works because I cannot really test without that module
Dec 09 06:42:27 <mrmoku> to exclue it's the boot env
Dec 09 06:42:32 <mrmoku> +d
Dec 09 06:42:36 <GNUtoo> if you want
Dec 09 06:42:39 <GNUtoo> just bump PR
Dec 09 06:42:44 <GNUtoo> or rebuild the kenrel
Dec 09 06:42:59 <GNUtoo> bitbake -c clean virtual/kernel; bitbake shr-lite-image
Dec 09 06:43:00 <mrmoku> ok, will take some time as I have not built for n900 yet
Dec 09 06:43:03 <GNUtoo> and unpack taht and see
Dec 09 06:43:05 <GNUtoo> ok
Dec 09 06:43:07 <mrmoku> but wanted to do that today anyway
Dec 09 06:43:11 <GNUtoo> ok
Dec 09 06:44:12 <GNUtoo> so so far crespo+n900 kenrel cannot be rebuilt because of that runtime issue
Dec 09 06:44:25 <GNUtoo> I've serial console on crespo
Dec 09 06:44:30 <GNUtoo> maybe I should try there
Dec 09 08:03:50 <JaMa|Off> GNUtoo: could you please try kernel image from here http://norman-schleicher.de/jenkins/job/shr-core-n900/lastSuccessfulBuild/artifact/shr-core/tmp-eglibc/deploy/images/nokia900/ ?
Dec 09 08:04:27 <JaMa|Off> GNUtoo: it was rebuilt from scratch with newer gcc, but also with cloog/ppl enabled 
Dec 09 09:00:54 <GNUtoo> JaMa|Wrk, ok thanks will try
Dec 09 09:07:05 <GNUtoo> JaMa|Wrk, it's the same: it blocks at boot
Dec 09 09:07:13 <GNUtoo> Starting kernel ...
Dec 09 09:07:16 <GNUtoo> and then nothing
Dec 09 09:08:54 <JaMa|Wrk> can you report it to khem?
Dec 09 09:09:15 <JaMa|Wrk> it's probably one of linaro patches added to meta-oe (as one was only moved from meta-oe to oe-core)
Dec 09 09:09:40 <JaMa|Wrk> so one of this http://git.openembedded.org/meta-openembedded/commit/?id=2378ee8f21741abd23e434260a88c62cb0e151f1
Dec 09 09:10:20 <PaulFertser> FYI OpenWrt uses the linaro toolchain for both MIPS and ARM targets.
Dec 09 09:10:29 <GNUtoo> ok
Dec 09 09:11:39 <PaulFertser> But 4.6 is not the default there yet, they stick to 4.5 for now.
Dec 09 10:02:56 * [Rui] waves
Dec 09 10:04:15 <JaMa|Wrk> hi
Dec 09 10:04:54 <[Rui]> JaMa|Wrk: hey! know if current shr-core is not totally unusable? I have an shr-core from about a month ago...
Dec 09 10:06:38 <JaMa|Wrk> [Rui]: check http://www.shr-project.org/trac/wiki/Stabilizing
Dec 09 10:07:25 <[Rui]> JaMa|Wrk: I saw your email the other day :)
Dec 09 10:08:27 <[Rui]> for some reason this image degraded into not even loading the framework, self degradation, I wonder if there are sd card issues...
Dec 09 10:08:36 <JaMa|Wrk> so better read the reports as people are less biased then me :)
Dec 09 10:08:42 <[Rui]> hehe
Dec 09 10:22:58 <[Rui]> this is not good
Dec 09 10:23:11 <[Rui]> after a few suspend/resume cycles it went off and doesn't turn back on.
Dec 09 10:23:26 <[Rui]> thinking it could be battery, I plugged the wall charger, but...
Dec 09 10:23:40 <[Rui]> still looks dead
Dec 09 10:41:11 <mrmoku> mickey_office: heh, sad robot is in the dynamic most popular songs list of jamendo :-)
Dec 09 10:41:46 <[Rui]> I'm a sad, sad robot...
Dec 09 10:41:59 <mrmoku> :P
Dec 09 10:55:43 <[Rui]> shit, mid-way during the e17 wizard I seemed to go out of battery again. I'll just have to let it rest for a while longer.
Dec 09 10:56:31 <mrmoku> [Rui]: the wizard you can exit anyway... no keyboard
Dec 09 10:56:37 <mrmoku> does not make much sense right now
Dec 09 11:12:15 <[Rui]> mrmoku: yeah, but I just doesn't even power on, so I guess the battery ran out again
Dec 09 11:13:10 <mrmoku> ok
Dec 09 11:13:25 <[Rui]> now that I've let it have some more time chargin, let's hope it goes better
Dec 09 11:15:38 <mickey_office> mrmoku: hey, excellent news :)
Dec 09 11:15:50 <mickey_office> guess my promotion tour for it worked ;)
Dec 09 11:15:57 <[Rui]> mickey_office: what do you have to do with pornophonique_
Dec 09 11:15:58 <[Rui]> ?
Dec 09 11:16:05 <mickey_office> nothing, i just like the guys' music
Dec 09 11:16:10 <[Rui]> ah :)
Dec 09 11:16:17 <mickey_office> so i did some advertising for them
Dec 09 11:16:18 <mickey_office> just for fun
Dec 09 11:16:20 <[Rui]> hehe
Dec 09 11:16:21 <mickey_office> in my talks
Dec 09 11:17:00 <[Rui]> two years ago they came to a web programming closed event organized by one of our biggest isp/portal
Dec 09 11:17:29 <[Rui]> they do a kind of Yahoo Hack Days called Sapo Codebits, and they usually invite a band to play
Dec 09 11:17:38 <[Rui]> and Pornophonique went there. they were very cool
Dec 09 11:19:20 <[Rui]> it's very weird/funny to see a guy on a guitar and another one with a Nintendo Gameboy both grooving a lot ;)
Dec 09 11:19:45 <[Rui]> mrmoku: I managed to boot, it went into a black screen again
Dec 09 11:20:09 <[Rui]> mrmoku: first the shr-logo boot splash, then it went away, then it looks like it's off
Dec 09 11:20:44 <[Rui]> it's not suspended (or at least it doesn't resume when pressing the power  button).
Dec 09 11:27:26 <jake42> [Rui]: i had the same problem
Dec 09 11:27:39 <[Rui]> it's happening every time
Dec 09 11:27:58 <jake42> did you untar the image with --numeric-owner?
Dec 09 11:28:01 <[Rui]> I'll download the lite just to see if there's a difference
Dec 09 11:28:13 <[Rui]> I untarred it with sudo tar xzvpf
Dec 09 11:28:34 <jake42> using xzpf --numeric-owner fixed the problem for me
Dec 09 11:29:13 <jake42> which image are you trying?
Dec 09 11:30:58 <[Rui]> jake42: that could only be due to a coincidence
Dec 09 11:33:21 <jake42> [Rui]: i flashed 004 first with xzf and then with xzpf --numeric-owner. so maybe it was the p option
Dec 09 11:33:58 <[Rui]> p is preserve permissions, which I shouldn't need anyways as I'm running it as root
Dec 09 11:34:38 <jake42> had the same problem with 003 and 002. but maybe several problems are entangled 
Dec 09 11:35:45 <mrmoku> [Rui]: you have to use --numeric-owner because otherwise it will use maybe wrong uids from your host
Dec 09 11:36:22 <mrmoku> which leads to dbus-exec-helper thing having wrong ownership
Dec 09 11:37:04 <mrmoku> which leads to fsodeviced not being able to dbus-launch fsousaged
Dec 09 11:37:20 <mrmoku> which leads to backlight being off :)
Dec 09 11:37:28 <[Rui]> mrmoku: when I tar something as root on one machine and untar it on another with different uids, I see the ids as they were different than in the other machine
Dec 09 11:37:43 <[Rui]> maybe there is something different now?
Dec 09 11:37:54 <mrmoku> dunno
Dec 09 11:38:04 <[Rui]> I mostly saw this in "old" centos 4 and 5 servers...
Dec 09 11:38:14 <[Rui]> I suppose my tar is much more recent :)
Dec 09 11:38:15 <JaMa|Wrk> [Rui]: you need to preserve the numbers which corresponds to numbers in untared passwd
Dec 09 11:38:41 <JaMa|Wrk> [Rui]: not to numbers used in passwd on system where you're untaring it
Dec 09 11:39:23 <mrmoku> [Rui]: all fine when it does not find the user in your host's passwd... problem is when it finds a wrong one
Dec 09 11:39:33 <mrmoku> leading to a wron uid
Dec 09 11:39:34 <mrmoku> +g
Dec 09 11:40:22 <[Rui]> let's try
Dec 09 11:46:28 <[Rui]> oh, definitely much better now.
Dec 09 11:46:36 <[Rui]> unwind_backtrace bla bla bla
Dec 09 11:50:34 <JaMa|Wrk> [Rui]: which image?
Dec 09 11:50:45 <[Rui]> full
Dec 09 11:50:58 <[Rui]> but I'm untarring again in case something bad happened 
Dec 09 11:51:09 <JaMa|Wrk> full from where
Dec 09 11:51:14 <[Rui]> and I have to append glamo.sd_max_clk=1000000
Dec 09 11:51:22 <[Rui]> http://build.shr-project.org/shr-core/images/om-gta02/
Dec 09 11:55:01 <[Rui]> seems to be booting, that first looooooong boot, though :)
Dec 09 11:55:25 <[Rui]> but the first boot attempt just stopped that after that old_state=off message
Dec 09 11:58:24 <[Rui]> ouch, everything is so small
Dec 09 11:58:34 <[Rui]> the default size on the e wizzard needs to be changed
Dec 09 12:00:02 <[Rui]> how do I fix this without restarting X after nuking .e ? is it the scale section in config?
Dec 09 12:00:48 <JaMa|Wrk> yes it's somewhere in menu
Dec 09 12:01:41 <[Rui]> almost impossible to click anything, though, I'll just nuke it
Dec 09 12:02:01 <[Rui]> oh...
Dec 09 12:02:09 <[Rui]> no network device on my side
Dec 09 12:03:45 <PaulFertser> (everything's too small) does it mean the X server is incorrectly determining physical dimensions/dpi?
Dec 09 12:04:00 <PaulFertser> Because E used to work just fine based on the actual dpi data iirc.
Dec 09 12:04:20 <[Rui]> PaulFertser: I think it's the scaling that's totally wrong
Dec 09 12:04:42 <[Rui]> I'm powering it off, for some reason I can't even give commands in the terminal so I need to mount the device
Dec 09 12:05:06 <PaulFertser> Default scaling should be just fine...
Dec 09 12:05:23 <[Rui]> PaulFertser: the default is about half as small as it should be :)
Dec 09 12:05:30 <[Rui]> the default should be changed probably to 2.0
Dec 09 12:08:28 <[Rui]> confirmed. sizing 2.0 is what brings back the normal size for stuff
Dec 09 12:08:50 <JaMa|Wrk> PaulFertser: e wizard added new page to select scaling and default value is not right for fr
Dec 09 12:10:45 <PaulFertser> JaMa|Wrk: E takes real dpi into account, so i can't really imagine how the default can be twice less than it should be unless there's a bug somewher.
Dec 09 12:13:14 <JaMa|Wrk> my point is that it's not caused by change in xserver, but this wizard page
Dec 09 12:13:44 <JaMa|Wrk> and this page is "new"
Dec 09 12:13:59 * pabs3 wonders if this blog post is relevant http://www.fooishbar.org/blog/tech/x/dpi-2011-12-07-04-35.html
Dec 09 12:33:53 <PaulFertser> Damn, why not simply try to "guess" the real size of the display based on the available EDID and let the user an easy way to override it? 
Dec 09 12:47:40 <[Rui]> EDID is not trustworthy
Dec 09 12:48:19 <[Rui]> but the preview areas should be larger, so the user could know it's not scaled down but real size that he's seeing
Dec 09 12:48:42 <[Rui]> I think that what X does is have a whitelist of trustworthy edid screens?
Dec 09 12:48:46 <[Rui]> innit?
Dec 09 12:49:03 <[Rui]> anyways, that option is probably a bit unnecessary, but...
Dec 09 12:49:10 <[Rui]> I don't know the reasoning behind it.
Dec 09 12:49:13 <[Rui]> gotta go
Dec 09 12:57:51 <PaulFertser> http://lists.fedoraproject.org/pipermail/devel/2011-October/157846.html is what looks correct to me.
Dec 09 13:19:04 * PaulFertser can't understand why some developers care so fucking much about trying to not make user participate in configuration of his own equipment.
Dec 09 13:21:07 <PaulFertser> I mean they keep inventing all kinds of automagic stupid tricks just to avoid asking the user about anything.
Dec 09 13:21:17 <JaMa|Wrk> are you still talking about e17 wizard?
Dec 09 13:21:42 <PaulFertser> No, about that EDID/DPI discussion on the fedora mailing list.
Dec 09 13:21:42 <JaMa|Wrk> because they invited user to participate in configuration by clicking on right size which suits him the best
Dec 09 13:21:47 <JaMa|Wrk> ok
Dec 09 13:21:52 <jonwil> better than going the other way and having too many options
Dec 09 13:22:12 <PaulFertser> I prefer options.
Dec 09 13:22:58 <misc> usually, that's options because default choice are not good, and I think if the option or autodetection was perfect, people would not care 
Dec 09 13:36:19 <PaulFertser> Too often the autodetection turns out to be highly annoying for various reasons, mostly because developers assume asking user is bad. But in the end they make me work much more to revert the effects of improper automagic.
Dec 09 14:50:33 <mrmoku> gah, android-rpc is still broken :/
Dec 09 14:50:35 <mrmoku> | main.vala:102.22-102.37: error: dynamic methods are not supported for `DBus.Object?'
Dec 09 14:50:38 <mrmoku> |         uint reply = bus.request_name("org.freesmartphone.ousaged", (uint) 0);
Dec 09 14:50:41 <mrmoku> |                      ^^^^^^^^^^^^^^^^
Dec 09 15:42:49 <angelox|laptop> good morning
Dec 09 16:09:28 <mrmoku> JaMa|Wrk: hmm.... 005 images dir is empty?
Dec 09 16:09:38 * mrmoku wanted to test 005 on gta02
Dec 09 16:09:53 <Slyon> hi
Dec 09 16:12:13 <mrmoku> hi Slyon 
Dec 09 16:12:32 <Slyon> hey mrmoku
Dec 09 16:15:40 <JaMa|Wrk> mrmoku: not every staging will get images build 
Dec 09 16:15:58 <mrmoku> JaMa|Wrk: ahh, ok. So I take the previous image and update
Dec 09 16:16:04 <JaMa|Wrk> mrmoku: but you can install 004 change feed to latest (you should always change them with staging image) and upgrade
Dec 09 16:16:27 <JaMa|Wrk> yup, exactly :)
Dec 09 16:16:37 <JaMa|Wrk> image build is taking so long on buildhost... that's why
Dec 09 16:16:48 <mrmoku> Then you can upgrade with opkg or reflash to newer image (if images are available in staging area too)
Dec 09 16:16:52 <mrmoku> you even wrote it :)
Dec 09 16:17:05 <mrmoku> sure, fine
Dec 09 16:17:24 <JaMa|Wrk> after cornucopia bump I'll start 006 image builds
Dec 09 16:17:31 <JaMa|Wrk> but it's still working on qt upgrade..
Dec 09 16:18:04 <JaMa|Wrk> should I add fsosystemd.service to cornucopia?
Dec 09 16:18:14 <mrmoku> ok, got a local n900 build to test and try the gisi patches PaulFertser provided
Dec 09 16:23:01 <mrmoku> Slyon: does the display work for you with current rc4 ?
Dec 09 16:23:07 * JaMa|Wrk finishing image to benchmark how much time we can save by disabling sstate packages..
Dec 09 16:24:43 <Slyon> mrmoku, no. i'm not even using 3.2 i'm still at hw-validation
Dec 09 16:25:01 <Slyon> 3.2 didn't even compile for me. I just reported the bug @gta04-owners ML
Dec 09 16:27:05 <mrmoku> Slyon: yeah, read that mail
Dec 09 16:27:14 <mrmoku> I'm always doing a fresh config
Dec 09 16:27:40 <mrmoku> I turned on SmartReflex this time... to see what's happening
Dec 09 16:27:44 <mrmoku> it boots fine
Dec 09 16:27:53 <mrmoku> dunno how to measure power-savings now though
Dec 09 16:28:05 <Slyon> i tried the 3.2 kernel once and it didn't boot due to some strange MMC error
Dec 09 16:28:10 <mrmoku> and for me the display is on at boot... and after a while turns off
Dec 09 16:28:17 <mrmoku> and then it's black forever
Dec 09 16:28:21 <mrmoku> whatever I do
Dec 09 16:28:38 <mrmoku> the xset -dpms does not help as our X is without DPMS
Dec 09 16:28:50 <mrmoku> JaMa|Wrk: I can confirm that n900 kernel does not boot
Dec 09 16:29:27 <Slyon> hmm strange... i've no idea what causes this
Dec 09 16:30:07 <JaMa|Wrk> mrmoku: do you know if you have built kernel after binutils upgrade to 2.22 and before gcc upgrade?
Dec 09 16:30:29 <mrmoku> JaMa|Wrk: hmm... when was gcc upgrade?
Dec 09 16:30:31 <JaMa|Wrk> mrmoku: I've talked with koen today and he said he has a lot of issues like this caused by binutils-2.22 (not gcc change)
Dec 09 16:30:39 <mrmoku> ok
Dec 09 16:30:43 <JaMa|Wrk> mrmoku: yesterday or day before
Dec 09 16:31:02 <mrmoku> then I probably have both
Dec 09 16:35:09 * JaMa|Wrk bbl
Dec 09 16:38:07 <CIA-82> SHR: 03lukasmaerdian 07meta-smartphone * rd0d35a2b3d20 10/meta-shr/recipes-shr/3rdparty/sensmon_git.bb: meta-shr: import sensmon from oe-classic
Dec 09 16:38:17 <CIA-82> SHR: 03lukasmaerdian 07meta-smartphone * r449d64515d31 10/meta-shr/recipes-shr/3rdparty/valacompass_git.bb: meta-shr: import valacompass from oe-classic
Dec 09 16:38:17 <CIA-82> SHR: 03lukasmaerdian 07meta-smartphone * r2e30a4bea278 10/meta-shr/recipes-shr/3rdparty/libsensmon_git.bb: meta-shr: import libsensmon from oe-classic
Dec 09 16:51:47 <dos1> http://build.shr-project.org/tests/site/
Dec 09 16:51:56 <dos1> file listing is now powered by ajax :)
Dec 09 16:52:14 <dos1> (please test it)
Dec 09 16:52:50 <dos1> i'll add some image as loading placeholder (animated spinner or smth)
Dec 09 17:11:32 <mrmoku> dos1: one thing... if java script is turned off file listing is on top
Dec 09 17:11:38 <mrmoku> is that changeable?
Dec 09 17:15:17 <dos1> mrmoku: yup
Dec 09 17:15:48 <mrmoku> good, because one would have to scroll to see the nice about and downloads :)
Dec 09 17:16:04 <dos1> downloads won't work without javascript
Dec 09 17:16:21 * mrmoku tries what happens
Dec 09 17:16:35 <dos1> but ofc i can make some placeholder
Dec 09 17:17:12 <mrmoku> dunno if all folks would be happy to have to use javascript for downloading...
Dec 09 17:17:23 <dos1> mrmoku: they don't have
Dec 09 17:17:27 <mrmoku> ahh?
Dec 09 17:17:37 <dos1> mrmoku: if javascript is turned off, then they can navigate just like they're doing it now
Dec 09 17:18:28 <dos1> that's why i put file listing on top, btw
Dec 09 17:18:34 <mrmoku> they can still click the Downloads button link
Dec 09 17:18:35 <mrmoku> hmm
Dec 09 17:18:36 <mrmoku> ok
Dec 09 17:18:40 <dos1> mrmoku: oh
Dec 09 17:18:40 <mrmoku> then leave it as is
Dec 09 17:18:50 <mrmoku> maybe disable or hide the button
Dec 09 17:18:54 <dos1> mrmoku: i forgot to hide it :)
Dec 09 17:18:57 <mrmoku> ok :)
Dec 09 17:19:11 <mrmoku> then just hide it and leave the rest :)
Dec 09 17:19:18 <dos1> ok
Dec 09 17:21:43 * PaulFertser despises JS and "fancy" webpages. Semantic markup ftw.
Dec 09 17:22:27 * mrmoku dinner
Dec 09 17:22:28 <mrmoku> bbl
Dec 09 17:36:11 <dos1> PaulFertser: i tried to leave markup as simple as it could be and to make all "fancinnes" as additional thing in JS file
Dec 09 17:36:47 <dos1> PaulFertser: so HTML file without JS and CSS will still be at least as usable, as buildhost site is now
Dec 09 17:37:53 <dos1> and i think i managed to do so
Dec 09 17:38:03 <PaulFertser> dos1: but you've spent time on "useless fanciness" neverthereless.
Dec 09 17:40:00 <dos1> PaulFertser: i can include it in portfolio and show it as a preview of my JS knowledge, so that time is not wasted for me :)
Dec 09 17:42:12 <PaulFertser> dos1: you know who has JS knowledge? ;)
Dec 09 17:42:24 <PaulFertser> dos1: Fabrice Bellard who wrote a fast x86 emulator in JS.
Dec 09 17:42:44 <dos1> yeah
Dec 09 17:42:58 <dos1> that's knowledge written by big K :D
Dec 09 17:50:28 <dos1> but well, i didn't mean "showing knowledge" as showing it to you, or somebody else in community, or some other computer geek - in this way, it's really "useless fanciness" and i'm fully aware of it
Dec 09 17:52:25 <dos1> i just think looking for some paid job would be nice, especially if i want to buy gta04 ;]
Dec 09 17:59:44 <PaulFertser> dos1: heh, i wasn't intending to downplay your work :) Are you really willing to be a web-developer btw?
Dec 09 18:02:10 <dos1> PaulFertser: i think it would be good choice until i finish studies, as it shouldn't be so hard to do both together
Dec 09 18:02:23 <PaulFertser> dos1: i think considering your experience you can find a paid job in embedded developing.
Dec 09 18:02:34 <dos1> PaulFertser: but later? meeeh... too boring in long run :)
Dec 09 18:03:11 <dos1> embedde developing, yeah, that's also what i want to study in second stage
Dec 09 18:03:52 <dos1> +d
Dec 09 18:15:20 <PaulFertser> Do you folks now there's going to be a 30k people street demonstration in Moscow tomorrow?
Dec 09 18:24:51 <dos1> i don't follow news so closely, but i know that there's not so nice now in Moscow
Dec 09 18:24:57 <dos1> PaulFertser: it's about elections, isn't it?
Dec 09 18:25:11 <PaulFertser> dos1: of course. "elections" even.
Dec 09 18:29:11 <dos1> i'm glad more and more people see that something is wrong... and i hope there won't be any victims of any kind :x
Dec 09 18:30:30 <PaulFertser> We'll see...
Dec 09 18:38:33 <GNUtoo> hi JaMa 
Dec 09 18:41:21 <JaMa> hi
Dec 09 18:42:14 <GNUtoo> khem told me that :
Dec 09 18:42:22 <GNUtoo> <khem> GNUtoo: remove these http://pastebin.com/sz1aas6Q
Dec 09 18:42:23 <GNUtoo> <khem> and then rebuild your gcc 
Dec 09 18:42:23 <GNUtoo> <khem> and kernel
Dec 09 18:42:23 <GNUtoo> <khem> and see if it boots
Dec 09 18:42:33 <GNUtoo> and it booted
Dec 09 18:42:44 <GNUtoo> basically reverting the last change make phones boot again
Dec 09 18:43:05 <JaMa> last change? file://linaro/gcc-4.6-linaro-r106841.patch ?
Dec 09 18:43:15 <GNUtoo> no
Dec 09 18:43:34 <GNUtoo> the thing you pointed me at
Dec 09 18:43:41 <GNUtoo> http://git.openembedded.org/meta-openembedded/commit/?id=2378ee8f21741abd23e434260a88c62cb0e151f1
Dec 09 18:43:58 <JaMa> ok.. can you bisect it a bit more and solve it with khem?
Dec 09 18:44:27 <GNUtoo> It took a long time to rebuild the toolchain
Dec 09 18:44:33 <GNUtoo> so kehm is gone now
Dec 09 18:44:37 <GNUtoo> *khem
Dec 09 18:44:47 <JaMa> I think if you point to exact linaro patch causing it, khem will fix it or at least report it back to linaro people
Dec 09 18:44:49 <GNUtoo> and he doesn't even know that it works
Dec 09 18:44:55 <GNUtoo> ok
Dec 09 18:45:09 <GNUtoo> but without pointers from khme it will be very very long
Dec 09 18:45:33 <JaMa> I know I had to do this twice already :/
Dec 09 18:45:33 <GNUtoo> there are 22 patches
Dec 09 18:45:45 <GNUtoo> even with bisect it would be long
Dec 09 18:46:03 <JaMa> and sometimes those patches depend on each other 
Dec 09 18:46:08 <JaMa> so you cannot just remove some
Dec 09 18:46:12 <GNUtoo> indeed
Dec 09 18:46:20 <GNUtoo> that's why I need khem's help
Dec 09 18:46:57 <JaMa> but for start you try to rebuild just gcc-cross (keep intermediate and initial)
Dec 09 18:47:19 <JaMa> if it's enough to get bootable image then you can rebuild only gcc-cross after every change
Dec 09 18:47:32 <GNUtoo> ok
Dec 09 18:47:39 <GNUtoo> I'll test something on n900 and be back
Dec 09 18:47:54 <JaMa> me is far from image.. NOTE: Running task 279 of 2335
Dec 09 18:48:11 <GNUtoo> ah ok
Dec 09 19:07:06 <Slyon> WOW! HP is going to OpenSource WebOS
Dec 09 19:11:23 <pespin> Slyon, link? :)
Dec 09 19:12:36 <Slyon> pespin, http://www.hp.com/hpinfo/newsroom/press/2011/111209xa.html?mtxs=rss-corp-news
Dec 09 19:12:42 <pespin> thanks
Dec 09 19:12:45 <Slyon> yw
Dec 09 19:13:11 <Slyon> that's awesome... probably we can adopt it to FSO
Dec 09 19:28:15 <GNUtoo> Slyon, nice
Dec 09 19:28:31 <GNUtoo> btw GRRR n900 still doesn't boot with toolchain changes reverted
Dec 09 19:28:38 <GNUtoo> maybe I did something wrong
Dec 09 19:28:46 <GNUtoo> I'll clean toolchain for n900 too
Dec 09 19:28:54 <GNUtoo> I think they have separate staging area
Dec 09 19:28:59 <GNUtoo> (the machines)
Dec 09 19:37:11 <GNUtoo> Slyon, well let's see what part they will release as free software
Dec 09 19:37:18 <GNUtoo> I guess not the 3d userspace driver
Dec 09 19:37:30 <GNUtoo> and I guess the whole webos depends on it
Dec 09 19:37:46 <GNUtoo> I also wonder about the older devices
Dec 09 19:37:57 <GNUtoo> for instance modem code for the palm-pres
Dec 09 19:38:28 <Slyon> GNUtoo, i don't think it depends on 3d drivers. as it's mostly a HTML/JS UI
Dec 09 19:39:16 <GNUtoo> what about the cards-like interface
Dec 09 19:39:35 <GNUtoo> that seem to depend on compositing
Dec 09 19:41:01 <Slyon> probably... but i don't need a 3d driver to enable compositing on the GTA04 as well
Dec 09 19:41:17 <GNUtoo> indeed
Dec 09 19:41:48 <Slyon> i guess the powerVR 3d driver is a problem, which HP isn't able to solve
Dec 09 19:41:52 <GNUtoo> like for instance elfe which already does have a card-like interface
Dec 09 19:41:57 <Slyon> right
Dec 09 19:42:09 <GNUtoo> so if we don't use webos's interface
Dec 09 19:42:12 <GNUtoo> what's left?
Dec 09 19:42:17 <GNUtoo> maybe the modem parts
Dec 09 19:42:22 <GNUtoo> gps
Dec 09 19:42:23 <GNUtoo> ?
Dec 09 19:43:13 <Slyon> yeah, we could probably fully support the Palm Pre Plus/2 devices
Dec 09 19:43:26 <GNUtoo> not totally fully because of bugs
Dec 09 19:43:26 <Slyon> if they open the modem protocol
Dec 09 19:43:28 <GNUtoo> but way better
Dec 09 19:43:41 <GNUtoo> for instance there seem to be a NEON bug
Dec 09 19:43:46 <GNUtoo> with the kenrel or hardware
Dec 09 19:43:56 <GNUtoo> that prevent running certain kind of apps
Dec 09 19:44:27 <Slyon> hmm
Dec 09 19:44:43 <GNUtoo> or force people to disable neon in all apps
Dec 09 19:44:50 <GNUtoo> like for instance like for tegra
Dec 09 19:45:04 <GNUtoo> (I don't remember which tegra version tough)
Dec 09 19:46:01 <Slyon> yes, that's a problem
Dec 09 19:46:25 <GNUtoo> but well disabling neon will only impact speed
Dec 09 19:46:32 <GNUtoo> I guess that can be handled in oe
Dec 09 19:57:09 <mrmoku> dos1: not that it would be a great thing... but Downloads flicker once
Dec 09 19:58:11 <dos1> mrmoku: you mean, when selected item fades out?
Dec 09 20:00:16 <mrmoku> dos1: now when initially selecting downloads it shows the 'Please select image type thing'
Dec 09 20:00:21 <mrmoku> which then disappears
Dec 09 20:00:33 <mrmoku> and then reappears
Dec 09 20:00:54 <dos1> hmm, strange
Dec 09 20:00:57 <dos1> mrmoku: which browser?
Dec 09 20:01:24 <mrmoku> dos1: firefox 
Dec 09 20:01:37 <mrmoku> 8.0.1
Dec 09 20:02:03 * mrmoku tries chromium
Dec 09 20:02:05 <dos1> ok
Dec 09 20:02:13 <TAsn> don't!
Dec 09 20:02:15 <mrmoku> dos1: does not happen with chromium
Dec 09 20:02:19 <dos1> on chromium here it works flawlessly, checking firefox now
Dec 09 20:02:21 <TAsn> people should stop using chromium, in firefox we trust!
Dec 09 20:02:40 <dos1> right
Dec 09 20:02:54 <dos1> flicker happens here too
Dec 09 20:02:58 <dos1> ok, i'll look at it
Dec 09 20:03:02 <mrmoku> maybe it's just excited to meet a brand new firefox... and flickers for happiness :P
Dec 09 20:03:45 <mrmoku> dos1: other than that... I'm happy... we can install it :)
Dec 09 20:04:08 <dos1> mrmoku: still selecting image number for shr-core-staging is missing :D
Dec 09 20:04:19 <mrmoku> ahh ok
Dec 09 20:04:20 <mrmoku> yeah
Dec 09 20:04:28 <mrmoku> there's just latest
Dec 09 20:09:43 <dos1> mrmoku: flicker fixed :)
Dec 09 20:10:01 <dos1> mrmoku: also found bug - keyboard arrows didn't work on firefox - also fixed :)
Dec 09 20:10:08 <mrmoku> yo :D
Dec 09 20:10:57 <mrmoku> dos1: nice... left right... how about up down to select the image type ? :P
Dec 09 20:11:16 <dos1> hmm
Dec 09 20:11:32 <dos1> maybe later :D
Dec 09 20:12:04 <dos1> but nice idea, i'll try
Dec 09 21:20:20 <dos1> mrmoku: shr-core-staging done
Dec 09 21:20:22 <dos1> :)
Dec 09 21:26:51 <GNUtoo> JaMa, I don't understand anymore a thing, n900 still doesn't boot with theses patch reverted
Dec 09 21:26:58 <GNUtoo> I'm about to throw the towel
Dec 09 22:01:01 <JaMa> GNUtoo: but you said it booted after reverting last gcc change or no?
Dec 09 22:01:18 <GNUtoo> crespo yes
Dec 09 22:01:22 <JaMa> GNUtoo: I can build few variants tomorrow from work, but I won't be able to test them there
Dec 09 22:01:31 <GNUtoo> ok thanks
Dec 09 22:01:32 <dos1> http://build.shr-project.org/
Dec 09 22:01:36 <dos1> time to test!
Dec 09 22:04:00 <Slyon> dos1, awesome!
Dec 09 22:04:12 <JaMa> dos1: can you make Files align left in File Listing?
Dec 09 22:04:37 <dos1> JaMa: yup, sec
Dec 09 22:04:39 <JaMa> dos1: ie http://build.shr-project.org/#fshr-core%2Fimages%2Fpalmpre%2F is harder to see difference between uImages now
Dec 09 22:05:16 <JaMa> dos1: or even don't limit the height when in File Listing mode?
Dec 09 22:08:11 <nschle85> dos1: can you left align please in file listing mode ? 
Dec 09 22:09:22 <dos1> JaMa, nschle85: check now
Dec 09 22:10:16 <nschle85> dos1: yup now its much nicer :-)
Dec 09 22:10:48 <dos1> ok :)
Dec 09 22:11:16 <JaMa> dos1: much better, thanks!
Dec 09 22:12:14 <dos1|N900> if you have some other request or if you see some bug, please ping me ;)
Dec 09 22:12:15 <JaMa> dos1: do you parse some config for crespo -> Nexus S etc?
Dec 09 22:12:24 <JaMa> dos1: or will something break when we add new image?
Dec 09 22:13:19 <dos1|N900> JaMa: nothing will break, it just won't be displayed until added to wizard
Dec 09 22:14:27 <dos1|N900> but i can do some config later - when i'll have some time i'll make it checking for image existence on-the-fly and automatically hide images which are not there
Dec 09 22:14:42 <nschle85> dos1: i think there is a typo in the description of the shr variants SHR full / chosen -> choosen
Dec 09 22:15:32 <dos1|N900> nschle85: thanks, fixed :)
Dec 09 22:17:34 <JaMa> dos1|N900: ah, ok that's fine with me for now as users can switch to file listing if it's missing
Dec 09 22:18:27 <JaMa> GNUtoo: can you try to revert binutils upgrade to 2.22 if you're not use it was caused by gcc upgrade?
Dec 09 22:18:48 <dos1|N900> JaMa: when you'll add some new image, ping me immediatelly ;)
Dec 09 22:21:56 <mrmoku> ~hail dos1
Dec 09 22:21:58 * apt bows down to dos1 and chants, "I'M NOT WORTHY!!"
Dec 09 22:22:06 <mrmoku> great thing
Dec 09 22:22:57 <mrmoku> now we need to style our distro to hold what the web is promissing :P
Dec 09 22:23:12 <mrmoku> anyway good work dos1
Dec 09 22:23:20 * mrmoku off to bed
Dec 09 22:23:25 <mrmoku> gnight all
Dec 09 22:24:04 <dos1|N900> good night ;)
Dec 09 22:25:34 <nschle85> mrmoku: good night
**** ENDING LOGGING AT Sat Dec 10 02:59:56 2011