**** BEGIN LOGGING AT Thu Jun 03 02:59:57 2010
Jun 03 04:20:37 <JaMa|Zzzz> max_posedon: and loosing touchscreen (much more often than with 2.6.32).. gdrm-2.6.34 is available too, but issues are the same
Jun 03 05:41:52 <gena2x> guys, you may build your kernel with optimizations for size instead of speed and your touchscreen bug will disappear. i did this for .29
Jun 03 05:43:04 <gena2x> this touchscreen bug (then touchscreen stop sending events) were main reason which stopped my optimization effort in February 
Jun 03 05:43:14 <gena2x> i spend much time trying to fix it
Jun 03 05:43:31 <gena2x> but didn't found the solution.
Jun 03 05:43:58 <gena2x> seem some race, but i'm unsure
Jun 03 05:44:25 <gena2x> this bug is perfectly reproducible
Jun 03 05:45:20 <gena2x> only funny thing i found while investigation is that our touchscreen sometimes start sending absolutely correct events (no need to filter)
Jun 03 05:45:59 <gena2x> this happens then screen is blanked for some period and perfectly reproducible too
Jun 03 05:46:35 <gena2x> have nice day ;)
Jun 03 05:47:04 <undrwater> gena2x: can you describe the bug in more detail?
Jun 03 05:47:33 <gena2x> touchscreen stop sending events.
Jun 03 05:47:50 <gena2x> after some time
Jun 03 05:48:08 <gena2x> i got this after very short time perion
Jun 03 05:48:14 <gena2x> *period
Jun 03 05:49:59 <undrwater> thank you
Jun 03 05:50:43 <gena2x> or you meant some internals description?
Jun 03 05:55:19 <gena2x> touchscreen driver (without filters) is a timer and interrupt handler. timer is for sending event data once in N ticks, interrupt for handling hardware data. but it has (from my POV) some very nasty details, like up/down states and fact your must check state after interrupt, etc
Jun 03 05:56:42 <gena2x> oh, may be even 2 interrupts, i forgot it.
Jun 03 05:57:00 <gena2x> one updown second for data coversion
Jun 03 06:02:16 <gena2x> for example, we may recieve down interrupt, but somehow we never sure stylus is down, even if we are waiting for up, instead we are checking after each conversion is it up?
Jun 03 06:03:02 <gena2x> and overwise somehow it is not working
Jun 03 06:03:09 <gena2x> and so on
Jun 03 06:50:58 <JaMa> gena2x: can you post your findings on that ticket?
Jun 03 06:51:20 <gena2x> add is in progress
Jun 03 06:51:50 <JaMa> add? the ticket already exists
Jun 03 06:51:50 <gena2x> i just trying to find logs, and if i'll success i can write real cause also...
Jun 03 06:51:56 <JaMa> or add comment? :)
Jun 03 06:52:07 <gena2x> exactly
Jun 03 06:52:11 <JaMa> gena2x: great! thanks, highly appreciated
Jun 03 06:52:34 <gena2x> just possible workaround...
Jun 03 06:53:00 <gena2x> and also problem may be actually different, but look very same
Jun 03 07:46:32 <JaMa> gena2x: btw CC_OPTIMIZE_FOR_SIZE is enabled for 2.6.34, do you mean some other specific option?
Jun 03 07:51:27 <gena2x> no.
Jun 03 07:51:37 <gena2x> :(
Jun 03 07:52:07 <gena2x> so, may be without it this just easier to reproduce
Jun 03 07:54:29 <gena2x> i posted all my findings we a bit more internals
Jun 03 07:56:20 <gena2x> so only workaround remain is reloading touchscreen modules
Jun 03 07:56:31 <gena2x> which is not so easy in case of X...
Jun 03 07:59:03 <JaMa> what is .34-backported?
Jun 03 08:00:12 <gena2x> original touchscreen driver (were in linux-next in february), brought back to .29
Jun 03 08:00:21 <gena2x> this were my first try
Jun 03 08:01:26 <gena2x> i assume now (without checking, and according to larsc report (where he told we need user space filter), that this driver is now in .34
Jun 03 08:01:27 <JaMa> can you try this on 2.6.34?
Jun 03 08:01:41 <JaMa> yes it is
Jun 03 08:02:05 <gena2x> sorry, no solution, only description of investigation.
Jun 03 08:02:11 <gena2x> what do you want to try?
Jun 03 08:02:14 <JaMa> not sure if exactly the same or if there were some adjustments after february
Jun 03 08:02:48 <gena2x> i think this really don't matter for this problem
Jun 03 08:03:05 <JaMa> gena2x: if you can test it on 2.6.34 as it is in git.openmoko.org now
Jun 03 08:03:43 <gena2x> test is it same issue? or try to fix it once again?
Jun 03 08:03:47 <JaMa> as we would like to move from .29 as soon as possible (all major issues are fixed)
Jun 03 08:04:06 <gena2x> nice to hear.
Jun 03 08:04:13 <JaMa> test the same issue but not on .29 (with backported ts)
Jun 03 08:04:16 <Q-Master> have anyone seen panicking last time?
Jun 03 08:04:36 <JaMa> because I've never seen it on .29 and with .32 only few times, but with .34 I get it really soon
Jun 03 08:05:08 <JaMa> gena2x: so maybe with .34 the problem is more exposed and you will find the real cause there (the race cond or whatever)..
Jun 03 08:05:11 <gena2x> are you using same compiler?
Jun 03 08:05:38 <gena2x> i had really no problems to reproduce.
Jun 03 08:05:38 <JaMa> yes for http://build.shr-project.org/shr-kms/ builds (gcc-4.4.4)
Jun 03 08:05:55 <JaMa> local builds are now with gcc-4.5
Jun 03 08:06:28 <gena2x> i didn't understand from your post is the compiler for .32 and for .34 differs
Jun 03 08:06:50 <JaMa> no .29, .32. .34 downloaded from shr area all built with gcc-4.4.4
Jun 03 08:07:18 <gena2x> so, yes IMO this may be reason ok more frequent bugs.
Jun 03 08:07:27 <gena2x> *of
Jun 03 08:07:36 <JaMa> gcc version?
Jun 03 08:07:56 <gena2x> all with same
Jun 03 08:08:10 <gena2x> stop.
Jun 03 08:08:23 <JaMa> all are built with same gcc, but lost ts happens the most with .34, then [Rui] reported it from .32 too and I've never seen it on .29 (where is also -Os not -O2..)
Jun 03 08:09:12 <gena2x> i definetly had lost ts with -O2 on 29.
Jun 03 08:09:35 <gena2x> also, original repost of lindi- seem about .29 too
Jun 03 08:09:42 <gena2x> but with -Os
Jun 03 08:10:21 <gena2x> ok, mb something else were changed in .34.
Jun 03 08:11:00 <gena2x> now, rethinking this issue i understood that i didn't took in accaunt that everything works with -Os
Jun 03 08:11:39 <gena2x> oh, i tried so many ideas that time, spend may be a week on it
Jun 03 08:12:16 <gena2x> i can try to investigate this again
Jun 03 08:12:26 <gena2x> this is what you want?
Jun 03 08:13:12 <gena2x> as a minimum i can say for sure that happens with hardware and is it same issue
Jun 03 08:16:38 <gena2x> not exactly 'didn't took in accaunt' but just have ideas about adding some delays...
Jun 03 08:26:38 <gena2x> JaMa: which sources to build? from git?
Jun 03 08:30:41 <JaMa> gena2x: http://git.openmoko.org/?p=kernel.git;a=shortlog;h=refs/heads/om-gta02-2.6.34
Jun 03 08:30:57 <gena2x> k
Jun 03 08:31:50 <gena2x> need add some debug info and build separate touchscreen driver...
Jun 03 08:32:02 <gena2x> i mean not info but printk
Jun 03 08:32:10 <JaMa> gena2x: yes I meant that would be great if you can test it again with 2.6.34 and maybe you will find something new about this issue, or some other issue or just great tip.. 
Jun 03 08:32:24 <gena2x> k
Jun 03 08:32:46 <JaMa> gena2x: as 2.6.29 will be left behind soon it's not worth to spend much time on it
Jun 03 08:34:00 <JaMa> gena2x: you can start from provideded defconfig if you want to be as close to SHR builds as possible (I have also patch adding more debug - while keeping ts issue still there)
Jun 03 08:34:01 <gena2x> i think 90% of bugs are the same, so 90% of .29 effort will be useful for any future kernel.
Jun 03 08:34:44 <JaMa> at least you don't have to rebase patch for .34 if you find the fix :)
Jun 03 08:35:06 <gena2x> ok, i'll keep posting progress here
Jun 03 08:36:13 <JaMa> or that trac (as not all read IRC here or logs later) so lets keep all usefull information in more public trac please
Jun 03 08:36:44 <gena2x> k
Jun 03 08:39:49 <JaMa> gena2x: here is defconfig I'm using now (with TS as module and some debug) http://gitorious.org/~jama/angstrom/jama-shr-experimental/trees/org.openembedded.dev/recipes/linux/linux-openmoko-2.6.34/om-gta02
Jun 03 08:40:35 <gena2x> ok, kernel debug facilities are in fact unneeded
Jun 03 08:40:43 <gena2x> printks are enought
Jun 03 08:40:55 <gena2x> in this case
Jun 03 08:42:38 <gena2x> is it easy to reproduce for you?
Jun 03 08:46:25 <gena2x> if not, and for example once per day, i can give a patch with some printks to be sure if cause if same.
Jun 03 08:47:31 <JaMa> gena2x: with 2.6.34 I had it few times in less then an hour after boot
Jun 03 08:48:27 <gena2x> k
Jun 03 08:49:11 <JaMa> gena2x: but I'm usually away from my neo, so touching screen is difficult to test for me :/ and also I'm testing gcc-4.5 now so my images can be different and only today I had first image which actually can boot :)
Jun 03 08:49:34 <gena2x> congrats :)
Jun 03 08:51:17 <JaMa> hehe, all thanks goes to khem and gcc devs...
Jun 03 09:07:11 <gena2x> JaMa: got FTBFS without debugging stuff :)
Jun 03 09:08:50 <gena2x> i mean 'with debugging stuff turned off, i got FTBFS'
Jun 03 09:21:57 <gena2x> JaMa: bug: SAMSUNG_PM_DEBUG should depend on DEBUG_LL (need printascii)...
Jun 03 09:23:25 <JaMa> yes seen that too and had to enable it manually
Jun 03 09:24:24 <JaMa> but probably this should be probably reported only upstream in vanila
Jun 03 09:29:37 <gena2x> ok, my first boot of .34 on freerunner, fingers crossed...
Jun 03 09:29:53 <gena2x> init ok
Jun 03 09:30:13 <gena2x> someone configured bigger font :)
Jun 03 09:30:51 <gena2x> ssh ok
Jun 03 09:31:26 <gena2x> even X ok.
Jun 03 09:31:30 <gena2x> good
Jun 03 09:33:39 <[Rui]> gena2x: what Linux are you running? uImage-2.6.34-oe2+gitr1+dd1225cc08c3375bf80289ac1965c724881b149a-r0-om-gta02.bin ?
Jun 03 09:34:26 <gena2x> [Rui]: self-build om-gta02-2.6.34 branch on debian.
Jun 03 09:34:42 <[Rui]> gena2x: oh... I'm really craving a little bit more stability in 2.6.3x.... :(
Jun 03 09:34:44 <gena2x> gcc 4.3
Jun 03 09:35:25 <gena2x> i'm in love with .29 in fact. what do you miss in .29?
Jun 03 09:35:44 <gena2x>  /devices/virtual/input/inputX?
Jun 03 09:36:58 <[Rui]> gena2x: speed :)
Jun 03 09:37:38 <gena2x> [Rui]: compile .29 without debugging stuff and it will be faster that .32/.34 and still without bugs
Jun 03 09:37:51 <gena2x> almost :)
Jun 03 09:37:59 <[Rui]> gena2x: I heard not, that it had some serious issues with stability, as well
Jun 03 09:38:19 <[Rui]> gena2x: anyways, 29 is not worth maintaining
Jun 03 09:38:19 <gena2x> [Rui]: rly? it's default on debian and qtmoko
Jun 03 09:38:38 <[Rui]> gena2x: it was some time ago...
Jun 03 09:39:13 <gena2x> it's unchanged in debian since january
Jun 03 09:39:14 <[Rui]> JaMa: how can I self compile 2.6.29-rc3 with a different optimization in my build environment?
Jun 03 09:39:47 <JaMa> [Rui]: easiest way is to move defconfig from kernel tree to OE tree
Jun 03 09:40:19 <JaMa> [Rui]: then -c menuconfig -b kernel_recipe, adjust and copy workdir/om-gta02*/linux-openmoko*/git/.config over defconfig in OE
Jun 03 09:40:30 <[Rui]> JaMa: although not a newbie compiling Linux, a newbie compiling a custom Linux for OE :) maybe you jumped a step there? :)
Jun 03 09:40:52 <[Rui]> what did you mean with "move defconfig from kernel tree to OE tree" ?
Jun 03 09:41:09 <gena2x> git checkout andy-tacking, menuconfig to exclude all suboptions in kernel hacking/kernel debugging, build and have fun
Jun 03 09:41:30 <gena2x> use your favorite .config as base
Jun 03 09:41:42 <JaMa> [Rui]: see my gitorious repo :)
Jun 03 09:41:52 <[Rui]> besides, I'd rather have some non packaged uImage and modules so I can have the current Linux in parallel for alternate boot
Jun 03 09:42:14 <gena2x> easyest way to get yout current config is zcat /proc/config.gz
Jun 03 09:42:38 <gena2x> *your
Jun 03 09:42:49 <[Rui]> shouldn't I use -c devshell or something to compile?
Jun 03 09:42:58 <gena2x> whoot?
Jun 03 09:43:00 <gena2x> no.
Jun 03 09:43:05 <gena2x> usual compile
Jun 03 09:43:31 <gena2x> make V=1 -j$PARALLEL O=$1 ARCH=arm CONFIG_DEBUG_SECTION_MISMATCH=y EXTRAVERSION=$VERSION
Jun 03 09:43:38 <gena2x> something like this
Jun 03 09:43:51 <gena2x> ops, CONFIG_DEBUG_SECTION_MISMATCH=y
Jun 03 09:43:56 <gena2x> sure not need
Jun 03 09:44:35 <gena2x> or i don't know, this is in my script inherided from some ooolllder version
Jun 03 09:45:40 <gena2x> oh, you discussing shr way :)
Jun 03 09:46:42 <JaMa> OE way :)
Jun 03 09:47:59 <JaMa> [Rui]: http://gitorious.org/~jama/angstrom/jama-shr-experimental/commit/48e0545c1797eaf70d734c8e8bb8352205eb9e50 found this one?
Jun 03 09:48:19 <gena2x> hm, i like that .34 it feels somehow faster even than .29
Jun 03 09:48:20 <JaMa> [Rui]: actually you can call -c menuconfig first
Jun 03 09:48:27 <gena2x> and from .32 i tried
Jun 03 09:49:02 <JaMa> then copy git/.config as om-gta02/defconfig-oe in recipes/linux/linux-openmoko-2.6.29
Jun 03 09:49:13 <JaMa> [Rui]: and then adjust the recipe as that commit
Jun 03 09:49:32 <JaMa> [Rui]: and call -c clean -b recipe; bitbake -b recipe to rebuild it with this defconfig
Jun 03 09:53:34 <[Rui]> JaMa: I'm with a slight headache, which makes me a whole lot dumber, but I think I'm missing something.
Jun 03 09:54:04 <[Rui]> JaMa: you mean I should just use that defconfig into 2.6.29?
Jun 03 09:55:14 <[Rui]> JaMa: or you mean to tell me to just clone your Linux tree and try it out, cross compiling directly?
Jun 03 09:58:13 <JaMa> [Rui]: no and now, your build envirnoment means working OE for building shr images?
Jun 03 09:58:32 <JaMa> [Rui]: or just cross compiler?
Jun 03 09:59:31 <[Rui]> JaMa: it has always failed for one reason or another, probably a lot of bad luck with the make update times
Jun 03 10:00:02 <[Rui]> JaMa: I usually just bother into getting enough libraries compiled in order to compile omnewrotate or elmdentica :|
Jun 03 10:00:18 <JaMa> hmm now I'm confused.. do you want to use OE or not for custom kernel? :)
Jun 03 10:00:21 <[Rui]> let me see where it's breaking
Jun 03 10:00:55 <[Rui]> JaMa: I want to try to see if I can be of some help, but I need to understand what I need to do to compile
Jun 03 10:01:39 <[Rui]> whether it's through OE environment or direct cross compile, I don't care much :)
Jun 03 10:02:00 <[Rui]> whatever's easier to get the same kernel
Jun 03 10:02:02 <JaMa> [Rui]: ok do yo need step-by-step? :)
Jun 03 10:02:14 <[Rui]> JaMa: is it written somewhere?
Jun 03 10:02:24 <gena2x> [Rui]: forget about OE
Jun 03 10:02:27 <JaMa> if you didn't ugprade for long, please do so now
Jun 03 10:02:35 <gena2x> [Rui]: just build kernel :)
Jun 03 10:02:55 <JaMa> rm -rf shr-unstable/tmp
Jun 03 10:02:59 <gena2x> [Rui]: i posted instruction above
Jun 03 10:03:02 <JaMa> bitbake virtual/kernel
Jun 03 10:03:14 <[Rui]> JaMa: I upgraded about a week ago
Jun 03 10:03:29 <JaMa> that's all, it will build std kernel used in shr now
Jun 03 10:03:49 <JaMa> bitbake -c menuconfig virtual/kernel
Jun 03 10:03:50 <[Rui]> JaMa: is that too long? because removing the tmp will mean a whole lot of wait time
Jun 03 10:04:05 <[Rui]> JaMa: but that will make packages, which will replace the current kernel
Jun 03 10:04:10 <JaMa> [Rui]: it depends when you rebuilt from scrats for last time
Jun 03 10:04:27 <[Rui]> from scratch, about two weeks ago
Jun 03 10:04:35 <[Rui]> but it kept failing
Jun 03 10:04:36 <JaMa> [Rui]: after staging changes it's better from scratch
Jun 03 10:05:17 <JaMa> [Rui]: if you show me the error I can give you solution :)
Jun 03 10:06:01 <[Rui]> JaMa: it's always some package or another, but since it takes a lot of time, I usually leave it doing stuff during the night and only notice it in the morning
Jun 03 10:06:14 <[Rui]> let's see where it stops now...
Jun 03 10:06:30 <[Rui]> gena2x: OE has it's issues, but it also allows very clean compilation and packages...
Jun 03 10:07:23 <JaMa> [Rui]: I did rebuild from scratch about 20 times over last 14 (due to gcc) and I would say that OE is in pretty good shape.. at least for building kernel (which is using only minimum of libs + cross compiler)
Jun 03 10:07:32 <gena2x> [Rui]: sure, for packages with dependences. but kernel can be built separately without any harm.
Jun 03 10:08:42 <gena2x> JaMa: my first WSOD in .34 :)
Jun 03 10:09:03 <[Rui]> JaMa: good shape for a fatty! ;) I wish it was a bit leaner!
Jun 03 10:09:12 <[Rui]> not fatty like me :)
Jun 03 10:10:12 <gena2x> last time i tried to build shr........ i had to change my /bin/sh to /bin/bash!
Jun 03 10:10:21 <gena2x> and this were not enough...
Jun 03 10:11:04 <[Rui]> gena2x: I don't think that's a problem in Fedora
Jun 03 10:15:19 <JaMa> gena2x: on ubuntu? known issue with /bin/sh -> /bin/dash
Jun 03 10:15:43 <gena2x> debian
Jun 03 10:15:59 <JaMa> also using dash?
Jun 03 10:16:36 <gena2x> yes. but i am unsure what about current debians
Jun 03 10:16:49 <[Rui]> JaMa: http://pastebin.com/hjBvBbRj
Jun 03 10:16:53 <gena2x> but original idea were in debian
Jun 03 10:17:04 <gena2x> (i think)
Jun 03 10:17:22 <gena2x> and significant effort were put to make whole system compatible
Jun 03 10:19:11 <JaMa> [Rui]: bitbake -c clean -b /media/devdisk/openmoko/shr-unstable/openembedded/recipes/matchbox2/matchbox-panel-2_svn.bb; bitbake -b /media/devdisk/openmoko/shr-unstable/openembedded/recipes/matchbox2/matchbox-panel-2_svn.bb
Jun 03 10:19:26 <JaMa> [Rui]: and something in between, mmt
Jun 03 10:21:08 <JaMa> [Rui]: http://www.mail-archive.com/shr-devel@lists.shr-project.org/msg01025.html
Jun 03 10:22:28 <JaMa> [Rui]: remove svn/svn.o-hand.com/repos/matchbox/trunk/matchbox-panel-2/ in your downloads directory
Jun 03 10:23:08 <[Rui]> JaMa: weird, could I be missing some shr-devel mails? I don't remember that one!
Jun 03 10:23:29 <JaMa> it's over 6 months old.. :)
Jun 03 10:24:18 <JaMa> and I have to say: the error is quite self-explanatory IMHO and it's not so OE specific...
Jun 03 10:25:05 <[Rui]> JaMa: self explanatory, perhaps... but still left me baffled, why didn't it merely update?
Jun 03 10:25:45 <gena2x> JaMa: hmm, seems sd io in .34 is much faster for some reason...
Jun 03 10:25:53 <JaMa> svn just refuse to add directory if it already exists
Jun 03 10:26:05 <JaMa> and if you ie svn update to newer revision
Jun 03 10:26:29 <JaMa> then downgrade to older (whould be interesting to fine why - maybe some really obsolete recipe)
Jun 03 10:27:07 <JaMa> and then want to svn update back (to currenty prefered svn revision) then it fails because of that "downgrade-leftover-directory-applets"
Jun 03 10:27:17 <[Rui]> well, removing matchbox-panel* from downloads didn't solve it :( I can... rebuild from scratch again, but *sigh*
Jun 03 10:27:24 <JaMa> no
Jun 03 10:27:36 <JaMa> did you remove that dir I asked?
Jun 03 10:27:59 <[Rui]> yes
Jun 03 10:28:11 <[Rui]> no downloads/tmp/matchbox but downloads/matchbox
Jun 03 10:28:27 <JaMa> then show me output of -c clean + build
Jun 03 10:28:32 <JaMa> 12:22:18 < JaMa> [Rui]: remove svn/svn.o-hand.com/repos/matchbox/trunk/matchbox-panel-2/ in your downloads directory
Jun 03 10:28:36 <JaMa> this one?
Jun 03 10:29:40 <[Rui]> JaMa: fuck me :( I looked at your email, and didn't see the other one.
Jun 03 10:30:56 <[Rui]> a fscking holiday which I could use to do some nice stuff and here I am with a migraine making me so much dumber
Jun 03 10:35:16 <gena2x> JaMa: compiled both -Os and -O2 - they work fine
Jun 03 10:35:28 <gena2x> JaMa: thinking...
Jun 03 10:35:38 <gena2x> JaMa: i mean ts works find
Jun 03 10:35:43 <gena2x> *fine
Jun 03 10:37:47 <[Rui]> gena2x: the WS doesn't fall on my definition of fine :)
Jun 03 10:38:02 <[Rui]> gena2x: specially because it'll hit you 99% of time you get a call
Jun 03 10:38:40 <gena2x> [Rui] "WS"?
Jun 03 10:39:03 <[Rui]> gena2x: white screen (but not oD -- of Death)
Jun 03 10:39:09 <gena2x> [Rui]: oh, shorter version of WSOD :)
Jun 03 10:39:48 <gena2x> I thought it never were really of D
Jun 03 10:40:03 <gena2x> my WS's are always without D :)
Jun 03 10:40:15 <[Rui]> gena2x: there were times where not even ping would respond
Jun 03 10:40:20 <Weiss> there original WSoD was definitely "of D"
Jun 03 10:40:30 <Weiss> when Glamo accidentally resets itself and hardlocks the memory bus
Jun 03 10:40:42 <Weiss> these days it's just my screwing up the GPU/LCM programming
Jun 03 10:40:59 <[Rui]> 2591 of 5800 (last time it broke compilation it stopped just under half the packages
Jun 03 10:42:19 <gena2x> oh... so direction is good.
Jun 03 10:42:49 <gena2x> and we as brave om users have -2 letters
Jun 03 10:44:48 <gena2x> can't reproduce.
Jun 03 10:44:59 <gena2x> seem i have to go for .29 for the issue
Jun 03 10:48:22 <JaMa> [Rui]: that's for elmdentica or something, right? because for kernel it's only about 1000 bitbake tasks
Jun 03 10:49:08 <[Rui]> JaMa: no... for shr-lite-image
Jun 03 10:49:14 <[Rui]> bb -c build shr-lite-image
Jun 03 10:52:00 <[Rui]> well, going out... and lunch.. bbl
Jun 03 10:52:55 <JaMa> [Rui]: ah ok.. -c build is not needed btw
Jun 03 10:53:02 <[Rui]> oh
Jun 03 10:53:17 <JaMa> bb -k shr-lite-image is better
Jun 03 10:53:41 <JaMa> will build as much as possible and skip failed + stuff depending on failed
Jun 03 10:53:43 <gena2x> aha, finally reproduced it ;)
Jun 03 10:54:10 <[Rui]> JaMa: I see why
Jun 03 10:54:24 <gena2x> 15 minutes of touching...
Jun 03 10:54:43 <[Rui]> gena2x: hey, this isn't a pr0n channel ;)
Jun 03 10:54:54 <gena2x> :)
Jun 03 10:55:12 <[Rui]> ah... I so remember the geek joke about why the mars lander crashed...
Jun 03 10:55:24 <gena2x> [Rui]: hey, fr is not a woman!
Jun 03 10:55:24 <[Rui]> telnet marslander
Jun 03 10:55:41 <[Rui]> motd warning about lack of space, asking people to clear some junk
Jun 03 10:55:48 <[Rui]> # rm -rf / pr0n
Jun 03 10:56:03 <[Rui]> ^C ^D ^C ^C ^Z....
Jun 03 10:56:23 <gena2x> :)
Jun 03 10:56:58 <[Rui]> gtg
Jun 03 11:06:49 <gena2x> hm, module reloading not helps
Jun 03 11:09:21 <gena2x> oh yes, damn dummy old driver i watched few months ago.
Jun 03 21:26:45 <gena2x> kernel dev who wrote following in driver "From tests, it seems that it is unlikely to get a pen-up event during the conversion process which means we can ignore any pen-up events with less than the requisite count done" should be fired.
Jun 03 21:27:48 <gena2x> and change his profession to "tester".
Jun 03 21:28:29 <[Rui]> gena2x: ouch, what was the consequence?
Jun 03 21:29:11 <gena2x> no matter. this just means he don't know what he is doing.
Jun 03 21:29:22 <gena2x> so he did 'tests'
Jun 03 21:30:56 <gena2x> but i may be wrong :)
Jun 03 21:34:50 <gena2x> so i have to be fired because of unreasonable charge...
Jun 03 21:36:50 <[Rui]> gena2x: wtf?
Jun 03 21:37:49 <gena2x> [Rui]: ok, just i'm talking a bit too much :)
Jun 03 21:38:16 <[Rui]> gena2x: well, I'm sorry for you if you are being fired :(
Jun 03 21:39:06 <gena2x> [Rui]: may english is far from being good, i just wanted to say that may be i am not right in my rant to that kernel dev
Jun 03 21:39:48 <[Rui]> gena2x: ah, ok, no harm done, I misread you anyway, and had understood you said "change" rather than "charge" :)
Jun 03 21:40:23 <gena2x> [Rui]: k
Jun 03 21:45:14 <gena2x> [Rui]: while i am touching it, how do you like the .29 without debugging?
Jun 03 21:46:05 <[Rui]> sure, got an uImage and modules.tgz somewhere?
Jun 03 21:46:29 <gena2x> [Rui]: oh, i hope you already builded it :)
Jun 03 21:47:01 <[Rui]> gena2x: no, I hardly understood what was needed.
Jun 03 21:48:13 <gena2x> http://www.bsdmn.com/openmoko/kernel/optim-stage2/
Jun 03 21:48:27 <gena2x> do not forget to extract modules
Jun 03 21:48:59 <[Rui]> gena2x: and carefully as not to break the current ones in the phone :)
Jun 03 21:49:35 <gena2x> it has separate version
Jun 03 21:49:48 <gena2x> so separate dir in /lib/modules
Jun 03 21:49:56 <gena2x> oh, wait
Jun 03 21:50:18 <[Rui]> gena2x: I wait, what's up?
Jun 03 21:50:25 <gena2x> your shr may have kms or something like this
Jun 03 21:50:37 <[Rui]> gena2x: I think not.
Jun 03 21:50:45 <gena2x> if not it'll be ok
Jun 03 21:51:02 <[Rui]> gena2x: if it isn't, I'll just mount the sd card in my laptop and change the uImage link
Jun 03 21:51:24 <gena2x> yeah. i am always using this way
Jun 03 21:51:28 <[Rui]> it worries me more that you have packaged the firmware as well :)
Jun 03 21:51:53 <gena2x> oh, remove it if you wish
Jun 03 21:52:01 <gena2x> i don't think fw is changed
Jun 03 21:52:21 <[Rui]> yeah
Jun 03 21:53:02 <[Rui]> any append tokens needed?
Jun 03 21:53:09 <gena2x> nothing
Jun 03 21:53:21 <[Rui]> ok
Jun 03 21:53:28 <[Rui]> going for reboot now...
Jun 03 21:55:58 <[Rui]> gena2x: one big problem: not being recognized as an ethernet device
Jun 03 21:56:14 <gena2x> hm.
Jun 03 21:56:36 <gena2x> try to replug
Jun 03 21:56:46 <gena2x> should work in fact
Jun 03 21:57:10 <[Rui]> eth1: unregister 'cdc_ether' usb-0000:00:13.1-1, CDC Ethernet Device
Jun 03 21:57:10 <[Rui]> usb 3-1: new full speed USB device using ohci_hcd and address 70
Jun 03 21:57:11 <[Rui]> hub 3-0:1.0: unable to enumerate USB device on port 1
Jun 03 21:57:11 <[Rui]> usb 3-1: new full speed USB device using ohci_hcd and address 71
Jun 03 21:57:12 <gena2x> the not recongsed is ofter sign of no modules available
Jun 03 21:57:46 <gena2x> can you run term on neo and check is lsmod is empty?
Jun 03 21:58:32 <[Rui]> gena2x: no, because I haven't the screen calibrated on boot
Jun 03 21:59:14 <gena2x> i can't connect calibration of absence/presence of modules
Jun 03 21:59:46 <gena2x> oh, or you mean your touchscreen is not working too?
Jun 03 22:00:05 <gena2x> s/of/to/
Jun 03 22:02:40 <[Rui]> gena2x: yeah, it seems the update to pointercal-xinput overwrote my link
Jun 03 22:02:45 <[Rui]> booting, now...
Jun 03 22:03:04 <gena2x> :)
Jun 03 22:03:38 <[Rui]> still no eth device on my laptop
Jun 03 22:03:52 <[Rui]> lsmod is empty
Jun 03 22:04:11 <gena2x> have you donloaded modules?
Jun 03 22:04:16 <[Rui]> yes
Jun 03 22:04:38 <gena2x> put to / and sayed tar xzvf <name> ?
Jun 03 22:04:48 <[Rui]> In lib/modules/FUBAR, FUBAR must be a perfect match to the version string of the kernel, no?
Jun 03 22:05:16 <gena2x> /lib/modules/`uname -r`
Jun 03 22:05:50 <[Rui]> well, it points to the right place...
Jun 03 22:05:58 <gena2x> hm...
Jun 03 22:06:09 <gena2x> try to load some module with modprobe
Jun 03 22:06:50 <gena2x> for example modprobe cdrom
Jun 03 22:06:59 <ThibG> maybe run depmod?
Jun 03 22:07:17 <gena2x> may be, but should not be :)
Jun 03 22:07:22 <[Rui]> ThibG: shouldn't update-modules do the trick?
Jun 03 22:07:42 <gena2x> wait, try to load some module
Jun 03 22:07:52 <gena2x> with modprobe
Jun 03 22:07:54 <ThibG> didn't really follow what you were doing :)
Jun 03 22:09:26 <[Rui]> well, depmod -a appears to have worked better than update-modules, so rebooting now.
Jun 03 22:09:42 <gena2x> oh, no need to reboot in fact
Jun 03 22:10:00 <[Rui]> gena2x: more of a clean process
Jun 03 22:10:02 <gena2x> strange, i thought this build should provide dependences
Jun 03 22:10:26 <gena2x> it has modules.dep in archive
Jun 03 22:11:01 <gena2x> also i thought i works not only for me but also for other people, so it's strange that it is not working
Jun 03 22:11:22 <[Rui]> modules now loaded
Jun 03 22:11:27 <gena2x> nice
Jun 03 22:11:33 <gena2x> ssh is ok?
Jun 03 22:11:36 <[Rui]> also with the proper devices.
Jun 03 22:11:38 <[Rui]> gena2x: yeah
Jun 03 22:11:42 <gena2x> good
Jun 03 22:11:56 <[Rui]> gena2x: doesn't seem as fast as the 2.6.3x Linuxes I tried
Jun 03 22:12:09 <gena2x> but it's much more stable
Jun 03 22:12:36 <gena2x> yes. a bit slower, but much faster than default .29
Jun 03 22:12:43 <[Rui]> gena2x: yeah
Jun 03 22:13:09 <gena2x> i feel now that .34 is faster, but i felt that .32 is slower
Jun 03 22:13:39 <[Rui]> yay, dragging is no longer slow as hell
Jun 03 22:13:47 <[Rui]> net operations are faster as well
Jun 03 22:14:04 <gena2x> but all this can be checked only with lmbench or remain only as feeling
Jun 03 22:14:14 <[Rui]> gena2x: thanks, I think I'll be using your Linux while a more stable 2.6.3x doesn't come up
Jun 03 22:14:30 <[Rui]> gena2x: it's definitely noticeable
Jun 03 22:14:42 <gena2x> yeah, wrote whole poem on this topic in mailing list in january
Jun 03 22:14:44 <[Rui]> gena2x: no psychological effect here
Jun 03 22:15:31 <gena2x> but unfortunately for shr users shr devs decided to go for .32 and do not use kernel like this
Jun 03 22:16:27 <[Rui]> gena2x: the devs do very well to do so
Jun 03 22:16:52 <[Rui]> gena2x: and 2.6.29-rc3 is still the kernel in shr-u
Jun 03 22:17:01 <[Rui]> 2.6.32 doesn't yet give enough stability
Jun 03 22:17:13 <gena2x> k. have fun, no matter for me, .32 is good goal too.
Jun 03 22:18:39 <[Rui]> gena2x: while 2.6.3x doesn't get good enough, I'd add my vote for your kernel config (unless some major issue comes up that I don't yet know about...)
Jun 03 22:18:48 <[Rui]> gena2x: any WS?
Jun 03 22:19:36 <gena2x> ws probaility is higher that in debug kernel, but only in similar conditions - cold temperature and resume.
Jun 03 22:20:17 <[Rui]> gena2x: on resume is bad due to received calls
Jun 03 22:20:46 <gena2x> i think, touchsreen should work
Jun 03 22:21:03 <gena2x> and properly learned location of 'answer' button may help :)
Jun 03 22:21:20 <gena2x> also no debug should use less power.
Jun 03 22:22:35 <gena2x> and about votes - this topic now covered with dust no need to speak about it, lets go for .32/.34 and just have fun if you like nodebug kernel
Jun 03 22:23:33 <gena2x> where were huge flame in fact. i do not want to even recall it.
Jun 03 22:33:11 <[Rui]> gena2x: hehe
Jun 03 22:33:24 <[Rui]> gena2x: it's never a good idea to stay fixed with such an old kernel
Jun 03 22:33:35 <supermagnum> hi, do anuone here have experience with xinput_calibrator ?
Jun 03 22:34:48 <[Rui]> supermagnum: just a tiny bit, if I can be of help...
Jun 03 22:35:10 <supermagnum> i cant get it to detect my screen
Jun 03 22:35:31 <supermagnum> Error: No calibratable devices found.
Jun 03 22:37:28 <supermagnum> i have compiled it from source and are trying to run the commando from /src
Jun 03 22:37:58 <[Rui]> you should run the calibrator...
Jun 03 22:38:15 <[Rui]> supermagnum: xinput_calibrator_once.sh
Jun 03 22:38:26 <supermagnum> ah, i will try that
Jun 03 22:38:33 <[Rui]> do it with an ssh connection (easier)
Jun 03 22:38:46 <[Rui]> because you'll have to place some content in /etc/pointercal.xinput
Jun 03 22:40:16 <supermagnum> huh.. thats weird.. the script you mentioned are not in xinput folder
Jun 03 22:40:53 <supermagnum> and i downloaded the source form svn
Jun 03 22:41:02 <j_14> Hello. How can I disable the touchscreen on my FR so that I can put it on my pocket without turning the backlight on?
Jun 03 22:41:22 <gena2x> j_14: depends on your distro
Jun 03 22:41:41 <gena2x> j_14: in qtmoko, you have to press aux button to lock your screen
Jun 03 22:42:53 <[Rui]> supermagnum: sorry, it's in $PATH on shr-unstalbe
Jun 03 22:42:57 <j_14> gena2x: I'm using debian.  I was thinking of disabling the touchscreen in Xorg, but didn't find any info on that. 
Jun 03 22:43:22 <supermagnum> and where o i get that ?
Jun 03 22:43:37 <[Rui]> supermagnum: what are you using, shr-unstable?
Jun 03 22:43:40 <[Rui]> or something else?
Jun 03 22:43:56 <[Rui]> if shr-unstable and it's updated, it's already there.
Jun 03 22:44:08 <supermagnum> i am using : http://github.com/tias/xinput_calibrator
Jun 03 22:44:09 <gena2x> j_14: oh, debian, nice. but i dont know. never tried where to lock screen. i can try to ask in #openmoko-debian
Jun 03 22:44:42 <gena2x> s/i can try to ask/you can try to ask/
Jun 03 22:44:43 <apt> gena2x meant: j_14: oh, debian, nice. but i dont know. never tried where to lock screen. you can try to ask in #openmoko-debian
Jun 03 22:45:31 <[Rui]> supermagnum: what is the GNU/Linux you're running in your Freerunner?
Jun 03 22:45:42 <supermagnum> be aware that i am trying to get it to work on a laptop
Jun 03 22:46:03 <supermagnum> a panasonic cf-29
Jun 03 22:46:28 <[Rui]> supermagnum: then I'm afraid I'll not be of much help
Jun 03 22:46:43 <supermagnum> :(
Jun 03 22:48:02 <[Rui]> supermagnum: sent you in priv the script in shr, perhaps it'll help you
Jun 03 22:48:04 <[Rui]> now I'm off to bed. good night all
Jun 03 22:48:20 <gena2x> gn
Jun 03 22:49:25 <undrwater> gena2x: about to try your kernel as well
Jun 03 22:49:59 <gena2x> this matters only on shr. qtmoko and debian already using fast kernel
Jun 03 22:50:07 <[Rui]> undrwater: I found out that in order to test the kernel you need to use it for a bit of time
Jun 03 22:50:30 <undrwater> [Rui]: didn't mean to wake you ;)
Jun 03 22:50:47 <undrwater> <--using shr-u
Jun 03 22:50:55 <[Rui]> so I'll not rush to evaluate :) for now gena2x's 2.6.29 compiled version is working fine, but let's see how it goes :)
Jun 03 22:51:01 <[Rui]> good night!
Jun 03 22:51:19 <undrwater> not using the phone full time right now...
Jun 03 22:51:24 <undrwater> part time with treo
Jun 03 22:52:01 <gena2x> i'm curently have a pleasure to do a bit dev for it
Jun 03 22:52:16 <gena2x> right now touchscreen issue.
Jun 03 22:52:41 <undrwater> what will be the benefit of fixing that bug?
Jun 03 22:52:43 <gena2x> also want to provide proper fix for bluetooth.
Jun 03 22:53:16 <gena2x> this is inportant bug to make .32, .34 and .29 with -O2 work.
Jun 03 22:53:48 <gena2x> fix for bluetooth is conceptually ready and working, but need some development to put to to share
Jun 03 22:53:55 <gena2x> s/share/shape/
Jun 03 22:53:56 <apt> gena2x meant: fix for bluetooth is conceptually ready and working, but need some development to put to to shape
Jun 03 22:54:39 <undrwater> what do these fixes improve?  touchscreen and bluetooth?
Jun 03 22:55:24 <gena2x> for touchscreen issue i still have no fix as i am talking too much today.
Jun 03 22:55:39 <gena2x> bluetooth allows using bluetooth headsets without any pain
Jun 03 22:56:01 <gena2x> at least in qtmoko.
Jun 03 22:56:20 <gena2x> shr have own fix, but i think it a bit less then mine
Jun 03 22:58:41 <undrwater> gena2x: are you talking UI to access bt headset?
Jun 03 22:59:06 <gena2x> ui is already working in qtmoko for example
Jun 03 23:00:14 <undrwater> so you're talking about register / deregister ease?
Jun 03 23:00:24 <undrwater> going from bt to handset, etc?
Jun 03 23:00:33 <gena2x> mine idea of fix based partially on investigations of other people allows to hear sound while in gsm calls
Jun 03 23:00:43 <undrwater> OH!
Jun 03 23:00:45 <undrwater> yay
Jun 03 23:00:57 <undrwater> that's my problem now
Jun 03 23:01:12 <gena2x> oh, so you can check my solution is have time?
Jun 03 23:01:42 <gena2x> i even prepared instruction how to check :)
Jun 03 23:01:58 <undrwater> OK...i can try...how much time do you think it could take?
Jun 03 23:02:07 <gena2x> few minutes
Jun 03 23:02:12 <gena2x> mb 15
Jun 03 23:02:23 <gena2x> depending on how fast you type
Jun 03 23:02:27 <undrwater> sounds good...let me move my SIM 
Jun 03 23:02:39 <undrwater> <-slow touch typist
Jun 03 23:02:51 <gena2x> which distribution you using?
Jun 03 23:03:00 <undrwater> shr-u
Jun 03 23:03:24 <gena2x> if shr, the fix may interfere with shr's one, but it should be easy to check
Jun 03 23:03:43 <undrwater> not a problem...not depending on this phone for daily use
Jun 03 23:03:54 <gena2x> k, i'll prepare link to tarball...
Jun 03 23:04:17 <undrwater> are we using part of the wiki's instructions?
Jun 03 23:04:27 <gena2x> no need
Jun 03 23:04:35 <gena2x> all instructions inside
Jun 03 23:04:46 <undrwater> cool
Jun 03 23:04:54 <gena2x> first check if your headset works in principe
Jun 03 23:05:01 <gena2x> then we can check it for shr
Jun 03 23:05:27 <gena2x> instruction is for checking if headset works for openmoko in pricipe
Jun 03 23:05:52 <undrwater> recipe?
Jun 03 23:06:02 <gena2x> and at the end you sould hear your voice from bt headset in om speaker and vise versa
Jun 03 23:06:12 <undrwater> oohhh...cool :)
Jun 03 23:06:13 <gena2x> oh, mb wrong word, wait...
Jun 03 23:06:42 <gena2x> i mean 'in principle'
Jun 03 23:07:00 <gena2x> if this will work, you'll have no problem in GSM
Jun 03 23:09:42 <undrwater> you've got the link yet?
Jun 03 23:09:53 <gena2x> copying to web server
Jun 03 23:10:10 <undrwater> ok
Jun 03 23:11:41 <gena2x> http://www.bsdmn.com/openmoko/bttestideacheck.tgz
Jun 03 23:11:48 <gena2x> instruction inside
Jun 03 23:11:58 <undrwater> ok...wgetting now
Jun 03 23:12:14 <gena2x> but i am unsure how all this applies to SHR
Jun 03 23:12:34 <gena2x> may be SHR people already have working sound
Jun 03 23:13:11 <undrwater> we'll see :)
Jun 03 23:13:15 <gena2x> may be not.
Jun 03 23:13:26 <undrwater> oh...should i untar to /?
Jun 03 23:13:35 <gena2x> anywhere you want
Jun 03 23:13:41 <gena2x> no matter
Jun 03 23:39:02 <undrwater> gena2x: it's going to take a while...shr doesn't use the same files and such
Jun 03 23:45:54 * Blu3 is away: on the road to tampa/ybor city, orlando/gay days@disney
Jun 04 00:15:38 <gena2x> undrwater: oh, i've gone to sleep...
**** ENDING LOGGING AT Fri Jun 04 02:59:57 2010