**** BEGIN LOGGING AT Wed Jul 07 02:59:56 2010
Jul 07 03:00:34 <sjsharksfan> preset reset
Jul 07 03:02:11 <Abyssul> Not what Im looking for
Jul 07 03:02:55 <sjsharksfan> oh
Jul 07 03:03:02 <sjsharksfan> sorry
Jul 07 06:36:55 <swisstomcat> morning
Jul 07 06:55:28 <swisstomcat_pre> yay
Jul 07 06:56:05 <swisstomcat_pre> good job
Jul 07 06:56:05 <swisstomcat_pre> wirc on 1.4.5
Jul 07 09:55:07 <bhuey> uNiXpSyChO: when the uberkernel going to be updated ?
Jul 07 09:55:21 <bhuey> is it now the 'stable' kernel ?
Jul 07 10:26:06 <swisstomcat> rehi
Jul 07 10:47:12 <noradtux_> hi
Jul 07 10:50:17 <noradtux> I was wondering if the cpu of the pre is driven with different voltages depending on speed. I looked at the overclicking-patch of the UberKernel and to me it looks like 600MHz and 720MHz use the same voltage. Is that correct?
Jul 07 10:51:02 <noradtux> that would mean that powerconsumption/heat should be about the same for 600 and 720 ...
Jul 07 10:51:36 <ka6sox> noradtux, yes, there are different voltages for different frequencies.
Jul 07 10:52:59 <noradtux> ka6sox: yes, thx, that's what I found out aswell. The question is, if I read the source correctly and 600 and 720 MHz use the same voltage.
Jul 07 10:53:18 <swisstomcat> hi
Jul 07 10:53:29 <swisstomcat> i'm at 720mhz
Jul 07 10:54:55 <noradtux> swisstomcat: i am currently trying that aswell. have been running at 600 up to now
Jul 07 10:55:15 <swisstomcat> i'm running screenstate 250/720 ... no problems so far
Jul 07 10:55:31 <swisstomcat> battery seems fine
Jul 07 10:56:26 <noradtux> swisstomcat:  my personal rule of thumb for "real" computers usually is that it's okay as long as I do not need to increase the voltage. And 800MHz already has quite an increase in voltage.
Jul 07 10:56:47 <noradtux> anyone tried under-volting?
Jul 07 10:56:48 <ka6sox> norad tux what is the voltage value, for 600 and 720?
Jul 07 10:57:17 <ka6sox> noradtux, I'm pretty sure that both warthog and uNiXpSyChO have tried undervolting.
Jul 07 10:58:10 <swisstomcat> probably a good rule, noradtux .. i've given up overclocking/overvolting .. remember it well from the early days tho :)
Jul 07 10:58:24 <noradtux> 1.20V (OPP3), 1.27V (OPP4), 1.35 (OPP5), 1.35 (OPP6), 1.42 (OPP7)
Jul 07 10:58:35 <noradtux> OPP7 would be 800MHz
Jul 07 10:59:02 <noradtux> 0.975V (OPP1), 1.050V (OPP2), 1.20V (OPP3), 1.27V (OPP4), 1.35 (OPP5), 1.35 (OPP6), 1.42 (OPP7)
Jul 07 10:59:30 <swisstomcat> does going under 500mhz lower the voltage?
Jul 07 10:59:37 <noradtux> yes
Jul 07 11:00:10 <noradtux> but with webos 1.4.5 and the current uberkernel the min frequency seems to be locked to 500MHz again :/
Jul 07 11:00:10 <swisstomcat> good to know
Jul 07 11:00:21 <swisstomcat> really?
Jul 07 11:00:32 <swisstomcat> i set it to 250mhz (screenstate)
Jul 07 11:01:15 <noradtux> me too, but when I checked yesterday it would stick to 500 with locked screen
Jul 07 11:01:22 <swisstomcat> have to check then
Jul 07 11:04:09 <swisstomcat> doctoring my other pre to 1.4.5 ATM
Jul 07 11:04:23 <noradtux> seems to be fixed in git, but no new packages yet
Jul 07 11:05:14 <noradtux> yeah, doctoring is the only way to get a working pre currently, they fsck'ed up modifying the fstab to make /media/internal writable by PDK-apps
Jul 07 11:06:03 <swisstomcat> unless you fiddle with fstab
Jul 07 11:06:59 <noradtux> sure
Jul 07 11:07:01 <noradtux>  I was thinking of the averasge user ;)
Jul 07 11:07:43 <swisstomcat> yeah, no chance there
Jul 07 11:08:16 <swisstomcat> they might work on 1.4.5.1 for the rest of the world before they release it
Jul 07 11:10:13 <noradtux> yeah. I mean, ti was nice having the release early, but it breaks navit. I spent half the weekend to figure out why it was crashing ;) Just because it _expects_ to be able to write into it's data-directory ...
Jul 07 11:10:43 <swisstomcat> oh no
Jul 07 11:14:45 <noradtux> but the good thing is that webos now brings all the necessary libs/APIs, so we can now work on getting it into preware :)
Jul 07 11:14:58 <swisstomcat> very nice
Jul 07 11:15:08 <swisstomcat> i've tested navit too, looks good
Jul 07 11:15:19 <swisstomcat> even into the catalog?
Jul 07 11:15:27 <noradtux> maybe
Jul 07 11:16:56 <noradtux> one problem is that webos doesn't ship the files needed for gettext-style localization. So we need to put some files onto the rootfs and I do not think that this is allowed for apps in the catalog ;)
Jul 07 11:17:03 <swisstomcat> i wish i could help but i got not farther than installing the PDK :)
Jul 07 11:21:49 <noradtux> you will be the first victim for testing the packages when they are done ;)
Jul 07 11:23:22 <swisstomcat> i can do that
Jul 07 12:41:30 <SimpsonTP> my pre updated to 1.4.5 and now the majority of patches doesn't work anymore
Jul 07 12:41:38 <SimpsonTP> is that expected behaviour ?
Jul 07 12:43:27 <rwhitby> have you read the PreCentral thread on that?
Jul 07 12:44:17 <swisstomcat> http://forums.precentral.net/webos-patches/253860-attention-patches-must-removed-prior-ota-update.html
Jul 07 12:46:36 <rwhitby> there's another which lists the ones that have been ported
Jul 07 12:47:37 <rwhitby> 280 or so out of 335
Jul 07 12:47:52 <SimpsonTP> okay
Jul 07 12:47:57 <SimpsonTP> and when its too late ?
Jul 07 12:48:21 <rwhitby> that's also in that thread
Jul 07 12:53:16 <SimpsonTP> rwhitby, hmm on which page ?
Jul 07 12:55:38 <swisstomcat> probably http://forums.precentral.net/2539700-post13.html?
Jul 07 12:58:49 <rwhitby> yup
Jul 07 12:59:25 <SimpsonTP> k
Jul 07 12:59:29 <SimpsonTP> lets ee
Jul 07 12:59:33 <noradtux> about govnah: does it still fill up your memory or does it use some kind of round-robin store for the graphs?
Jul 07 12:59:47 <rwhitby> noradtux: only stores last 300 samples now
Jul 07 13:00:01 <noradtux> good :)
Jul 07 13:00:02 <swisstomcat> 0.5.0 is the latest
Jul 07 13:00:13 <noradtux> yep,running here
Jul 07 13:00:31 <swisstomcat> interested in a german version of govnah?
Jul 07 13:00:49 <noradtux> not really, I run my whole webos on english ;)
Jul 07 13:01:53 <noradtux> except navit, german street names spoken with english pronounciation are really hard to understand :D
Jul 07 13:04:06 <sebastianha> If I do a complete reset with safe erase on my palm will this delete preware and patches, too?
Jul 07 13:04:20 <rwhitby> preware yes, patches no.
Jul 07 13:04:27 <sebastianha> mmh, ok
Jul 07 13:04:29 <rwhitby> you need to do a full erase and webos doctor
Jul 07 13:05:00 <sebastianha> can I somehow see if there are any patches? I bought a used palm and want to check
Jul 07 13:05:51 <sebastianha> or should I take the latest doctor?
Jul 07 13:06:33 <swisstomcat> lastest doctor is probably the best bet
Jul 07 13:07:01 <sebastianha> ok
Jul 07 13:07:26 <noradtux> just doctor it and enjoy beeing sure that you have a clean system
Jul 07 13:10:57 <rwhitby> for a used device, always full reset and doctor
Jul 07 13:12:23 <sebastianha> ok
Jul 07 13:26:11 <SimpsonTP> okay, I don't get it or it doesn't work
Jul 07 13:26:44 <SimpsonTP> some patches install fine but won't work
Jul 07 13:29:18 <swisstomcat> what did you do?
Jul 07 13:38:31 <SimpsonTP> I installed 1.4.5 without taking in account the stuf mentioned in the mail
Jul 07 13:38:45 <SimpsonTP> forum I mean
Jul 07 13:40:44 <swisstomcat> what did you do after reading in the forum?
Jul 07 13:40:50 <SimpsonTP> then I tried to do steps in post 13
Jul 07 13:41:09 <SimpsonTP> which seem to work, but installing the patches doesn't make them work
Jul 07 13:42:53 <swisstomcat> so you ran EPR?
Jul 07 13:43:12 <SimpsonTP> Emergency Patch Removal
Jul 07 13:43:29 <swisstomcat> yes
Jul 07 13:43:50 <SimpsonTP> I do that by selecting install ? or is there a actual invocation place ?
Jul 07 13:45:19 <swisstomcat> you start it from preware (the EPR)
Jul 07 13:46:07 <SimpsonTP> okay, i'm lost, I look it up as a package ?
Jul 07 13:46:45 <swisstomcat> start preware
Jul 07 13:46:51 <SimpsonTP> ack
Jul 07 13:48:11 <swisstomcat> you can start typing to search
Jul 07 13:48:11 <swisstomcat> emergency
Jul 07 13:48:40 <SimpsonTP> k
Jul 07 13:51:02 <SimpsonTP> I only I don't see Emergency Patch Removal there, only Recovery
Jul 07 14:19:04 <noradtux> you guys don't happen to know where palm stores the EAS profile-data, especially the policy it gets from the EAS server?
Jul 07 14:27:06 <SimpsonTP> swisstomcat, and then....
Jul 07 14:27:19 <swisstomcat> install
Jul 07 14:28:10 <SimpsonTP> EPR stands for recovery or removal ?
Jul 07 14:28:31 <swisstomcat> recovery
Jul 07 14:29:16 <SimpsonTP> okay, and afterwards ?
Jul 07 14:29:44 <swisstomcat> what does it say after installing?
Jul 07 14:29:51 <swisstomcat> it has been a while since i ran it
Jul 07 14:29:53 <SimpsonTP> phone reboot
Jul 07 14:29:58 <SimpsonTP> running that now
Jul 07 14:29:58 <swisstomcat> ok
Jul 07 14:31:27 <swisstomcat> bbl
Jul 07 17:06:39 <mikePre> curious if anyone else is seeing this or if anyone knows why I am seeing an update for 1.4.1.1 from webOS on my Pre (which is running 1.4.1.1), I am running latest UberKernel...
Jul 07 17:10:19 <egaudet_work> have you installed anything outside of preware in the past
Jul 07 17:11:42 <mikePre> there is not much I havent installed... :-).....I am also up for a new something or other
Jul 07 17:15:25 <egaudet_work> well something installed outside of the feeds may have installed some ipk to / which would give you the update bug
Jul 07 17:15:36 <mikePre> i found a shell script that toggles the update daemon so that I dont get the update messages, and I just toggled it back on today to see if 1.4.5 was around...and it popped up with 1.4.1.1 again
Jul 07 17:17:50 <sjsharksfan> I've seen that happen with others around update time
Jul 07 17:18:50 <mikePre> i only have 1 file in / and that is the md5sums file
Jul 07 17:21:52 <egaudet_work> I meant if something installed with ipk install  using /, it has been known to cause the repeated update issue in the past
Jul 07 17:24:38 <mikePre> at one point I couldnt get the pdf viewer app to show up, and I think I pacckaged it up and tried an ipk install, tht sound like something that could cause that?  and how would one remedy it?  just wondering when 1.4.5 comes out, if it will know to go to 1.4.5 and not try to go back to 1.4.1.1
Jul 07 17:26:26 <egaudet_work> any ipk install could cause this
Jul 07 17:27:13 <egaudet_work> Doctor would remedy it
Jul 07 17:40:48 <mikePre> yeah, im done with DR for now, unless I have to....I will wait for 1.4.5 and see how that goes
Jul 07 17:40:48 <mikePre> thanks
Jul 07 17:50:30 * torchie refresh
Jul 07 17:52:29 <SimpsonTP> egaudet_work, I stil can't get the vkb to work :(
Jul 07 17:53:47 <egaudet_work> what's the symptoms
Jul 07 17:56:42 <SimpsonTP> well first nothing, then I did the 1.4.5 update and then it showed me a dialog that it couldn't find stuff and I needed to reinstall it
Jul 07 17:56:55 <SimpsonTP> now I reinstalled it, and it doesn't do anything anymore :)
Jul 07 17:57:12 <egaudet_work> what did you reinstall
Jul 07 17:57:26 <SimpsonTP> the 2 packages, vkb + the themes
Jul 07 17:57:40 <egaudet_work> and when you double tap nothing comes up?
Jul 07 17:58:00 <SimpsonTP> nope
Jul 07 17:58:08 <egaudet_work> has it worked in the past?
Jul 07 17:58:21 <SimpsonTP> nope, we talked before, you were busy back then
Jul 07 17:58:56 <egaudet_work> is this German Pre+ ?
Jul 07 17:59:14 <SimpsonTP> yup
Jul 07 18:00:25 <SimpsonTP> known issues ?
Jul 07 18:00:53 <egaudet_work> no but my gut feeling is I'm missing a locale or something
Jul 07 18:01:10 <SimpsonTP> the phone is setup english
Jul 07 18:01:18 <egaudet_work> can you novaterm in?
Jul 07 18:01:38 <SimpsonTP> uhm not now
Jul 07 18:01:42 <SimpsonTP> forgot my cable
Jul 07 18:01:55 <egaudet_work> and do ls /usr/palm/frameworks/mojo/submissions/347/resources
Jul 07 18:02:03 <egaudet_work> terminal app?
Jul 07 18:04:41 <SimpsonTP> egaudet_work, did that
Jul 07 18:05:22 <egaudet_work> what do you have for en_ *
Jul 07 18:05:59 <SimpsonTP> ca,cn,gb,ie,mx,pl,us
Jul 07 18:06:24 <egaudet_work> and de_ ?
Jul 07 18:06:33 <SimpsonTP> de,ch
Jul 07 18:06:44 <SimpsonTP> but i'm using the phone in netherlands
Jul 07 18:06:58 <egaudet_work> what do you have for nl_ or _nl
Jul 07 18:07:17 <SimpsonTP> just nl_nl
Jul 07 18:09:50 <jhopixi> how often do people see google redirecting to the wrong language? for some reason, the goog keeps sending me to the french site
Jul 07 18:10:35 <SimpsonTP> browser settings
Jul 07 18:14:47 <egaudet_work> SimpsonTP: type the following in terminal
Jul 07 18:15:00 <egaudet_work> cat /usr/lib/luna/customization/locale.txt
Jul 07 18:16:10 <SimpsonTP> cat not found...
Jul 07 18:16:15 <SimpsonTP> great
Jul 07 18:16:20 <SimpsonTP> need a package for that
Jul 07 18:17:33 <egaudet_work> no
Jul 07 18:17:36 <egaudet_work>  /bin/cat
Jul 07 18:20:23 <SimpsonTP> okay, what do you wanna know ?
Jul 07 18:22:10 <egaudet_work> the different languageCode and countryCodes
Jul 07 18:22:33 <egaudet_work> example I have for laungageName English, languageCode en, countryCode us
Jul 07 18:23:00 <SimpsonTP> here its de
Jul 07 18:23:10 <PuffTheMagic> we are gonna need to localize wirc
Jul 07 18:23:38 <egaudet_work> SimpsonTP: so languageCode en, countryCode de
Jul 07 18:24:22 <SimpsonTP> hmm no de de
Jul 07 18:24:36 <egaudet_work> for languageName English?
Jul 07 18:24:48 <SimpsonTP> anyway I need to stop now, need to save batery :(
Jul 07 18:24:53 <SimpsonTP> forgot cable
Jul 07 18:25:28 <egaudet_work> well if it's en_de it makes sense why it's not working and I can add that localization to the patch
Jul 07 18:25:54 <SimpsonTP> okay
Jul 07 18:26:00 <SimpsonTP> I'll let you know !
Jul 07 18:31:40 <PuffTheMagic> someone should make a solar backplate
Jul 07 18:32:19 <egaudet_work> s/someone/PuffTheMagic/
Jul 07 18:32:25 <oil> so you can set your phone down screen-down in the sun?
Jul 07 18:32:33 <PuffTheMagic> yeah
Jul 07 18:32:43 <oil> sounds like a good way to scratch the shit out of it
Jul 07 18:33:09 <PuffTheMagic> not if u have a shell for it
Jul 07 18:33:50 <egaudet_work> solar shell
Jul 07 18:39:28 <torchie> just make the whole device of solar panels
Jul 07 18:39:51 <torchie> and use the kinetic energy provided by typing on the keyboard
Jul 07 18:40:06 <torchie> the more you use the phone, the more it charges!
Jul 07 18:41:47 <PuffTheMagic> what about piezo electric from typing
Jul 07 18:45:32 <NachoB> egaudet_work: updated KeyBoss, Sym working, wallpaper and ringtone still reverting to default
Jul 07 18:47:36 <torchie> can you get electricity from a capactiive screen
Jul 07 18:47:40 <jhopixi> make the lcd solar underneath.. would be cool if possible
Jul 07 18:47:48 <torchie> like suck the energy out of the guy touching it
Jul 07 18:48:08 <torchie> I don' t get why they don't have solar panels on the faces of these things like calculators
Jul 07 18:50:09 <PuffTheMagic> netbook here
Jul 07 19:27:09 <noradtux_wirc> hape: navit seems to work, including speech :-)
Jul 07 19:33:11 <Robi_> the slide mechanism is probably the best way to engineer a manual charging system
Jul 07 20:57:16 <swisstomcat> evening
Jul 07 22:32:25 <bpadalino|work> another 15 minutes to putz around
Jul 07 22:32:34 <halfhalo_class> until?
Jul 07 22:32:45 <bpadalino|work> test is done running and i can kick off another one before heading home
Jul 07 22:33:11 <bpadalino|work> silly data generator takes a long time to run
Jul 07 22:33:17 <halfhalo_class> ewww
Jul 07 22:33:19 <halfhalo_class> data
Jul 07 22:33:27 <bpadalino|work> i know, right ?
Jul 07 22:33:32 <ka6sox-work> bpadalino, I had to find a bug in some logic that had 84 terms today, sucked.
Jul 07 22:33:34 <bpadalino|work> who needs that stuff
Jul 07 22:33:42 <bpadalino|work> 84 terms? that is ridiculous
Jul 07 22:33:51 <bpadalino|work> what was the bug in the end ?
Jul 07 22:34:22 <ka6sox-work> inverted input
Jul 07 22:34:39 <bpadalino|work> funky
Jul 07 22:34:50 <ka6sox-work> ya, I got my signs mixed up
Jul 07 22:35:06 <bpadalino|work> i sometimes wonder how much energy is wasted swapping bytes
Jul 07 22:35:29 <ka6sox-work> oh because of LE and NBO?
Jul 07 22:35:53 <bpadalino|work> given all the data and all the computers and everything going on .. that, and i wonder how much stuff ends up being converted from two's compliment to offset-binary
Jul 07 22:36:08 <ka6sox-work> ewww
Jul 07 23:00:41 <rwhitby> morning
Jul 07 23:01:30 <swishy> morning
Jul 07 23:14:59 <beelzabub> morning
Jul 07 23:38:11 <oc80z> whats good
Jul 07 23:39:32 <rwhitby> oc80z: 1.4.5 is pretty good, once you fix /etc/fstab :)
Jul 07 23:40:06 <oc80z> whats swapping?
Jul 07 23:40:50 <oc80z> (did germany say 1.4.5 fixed the cards issue?)
Jul 07 23:41:04 <ark3qqq> I've discovered that my pre doesn't like switching clock frequencies too often. It's happy on screenstate, but tends to freeze on On-Demand or Conservative.
Jul 07 23:41:47 <oc80z> wassa deal with fstab?
Jul 07 23:41:51 <rwhitby> that's quite common
Jul 07 23:42:16 <rwhitby> oc80z: missing umask=0000 on /media/internal for OTA vs doctored
Jul 07 23:42:39 <oc80z> oh damn.
Jul 07 23:42:52 <ark3qqq> rwhitby: common re the Pre freezing? (i.e. were you answering me?)
Jul 07 23:42:58 <rwhitby> oc80z: combined with PDK jails, and some games don't work.
Jul 07 23:43:00 <rwhitby> ark3qqq: yes
Jul 07 23:44:15 <ark3qqq> rwhitby: Gotcha. Yeah, it's too bad. I was getting crazy-good battery life with on-demand scaling from 250-1050 and Warthog, whereas I'm only getting quite-good battery life with screenstate 500-1050.
Jul 07 23:44:32 <ark3qqq> Warthog is pretty awesome regardless.
Jul 07 23:48:36 <bhuey> 1.4.5 is out on Sprint now ?
Jul 07 23:48:59 <oil> that rumor said its to come out today
Jul 07 23:49:22 <bhuey> what about the rumor that it's going to fix the TMC problem ?
Jul 07 23:49:24 * oc80z hittin update update update
Jul 07 23:49:56 <rwhitby> bhuey: I run Uber on mine, so don't know what the unmodified 1.4.5 OS does for TMX
Jul 07 23:49:56 <ark3qqq> What kernel(s) does the uinput module work under?
Jul 07 23:49:59 <rwhitby> TMC
Jul 07 23:50:03 <rwhitby> ark3qqq: any
Jul 07 23:50:44 <bhuey> rwhitby: uberkernel works well, it's almost like having a new device
Jul 07 23:50:57 <ark3qqq> rwhitby: Not a kernel module, then? So that's not my problem... More reading to do.
Jul 07 23:51:02 <egaudet> hmm how else can I clean up if I can't catch SIGTERM?
Jul 07 23:51:34 <beelzabub> why can't you catch sigterm?
Jul 07 23:51:43 <egaudet> Luna seemingly doesn't like me to
Jul 07 23:51:58 <beelzabub> you mean for a PDK app?
Jul 07 23:52:00 <egaudet> when I catch it in my service, upon reboot the wallpaper is reset
Jul 07 23:52:17 <beelzabub> what is your service doing?
Jul 07 23:52:50 <rwhitby> egaudet: I reckon there is a LunaSysMgr termination routine which saves sqlite databases back to disk, and that's not being rung
Jul 07 23:52:53 <rwhitby> run
Jul 07 23:53:30 <rwhitby> that would also explain people loosing settings when we do a tellbootie to reboot the device on kernel installs
Jul 07 23:53:42 <egaudet> beelzabub, creating a uinput device and piping the keypad input device into it along with other stuff
Jul 07 23:54:11 <egaudet> rwhitby, that makes sense.  I dont' have enough experience with signals, can I catch it to clean up and still let it pass on to luna
Jul 07 23:54:25 <rwhitby> egaudet: me neither
Jul 07 23:54:36 <ark3qqq> I hope Luna flushes an app's database when you toss the app; I always toss all apps before hitting the reboot key in Preware.
Jul 07 23:54:37 <beelzabub> still not clear on exactly what you're trying to do
Jul 07 23:54:39 <beelzabub> what's uinput?
Jul 07 23:54:51 <egaudet> userspace input module
Jul 07 23:54:59 <beelzabub> not familiar with what it is
Jul 07 23:55:12 <beelzabub> i imagine something like fuse but for input devices?
Jul 07 23:55:21 <ark3qqq> egaudet: You can remove your signal handler and then signal again.
Jul 07 23:55:32 <beelzabub> anyways, don't see the point of it since hidd handles all the input
Jul 07 23:55:35 <ark3qqq> egaudet: Where by "remove" I mean "replace with original."
Jul 07 23:55:39 <egaudet> it allows userspace to create/write to input devices.  Which I'm using to re-link keypad0 to so that hidd reads my created input device for keypad events
Jul 07 23:55:49 <beelzabub> ahh
Jul 07 23:56:18 <beelzabub> so how does that have anything to do with lunasysmgr?
Jul 07 23:56:27 <egaudet> all that stuff works, but as soon as I sigaction(SIGTERM), reboot resets wallpaper, so I need to not do that but still clean up
Jul 07 23:57:25 <egaudet> Also the service writes to the keypad input device to set repeat rate stuff, which I really want to reset back to default when the service dies.  That's the main reason for catching the signal
Jul 07 23:57:49 <beelzabub> sure
Jul 07 23:57:57 <beelzabub> i'm still failing to see what any of that has to do with the wallpaper
Jul 07 23:58:06 <beelzabub> if it's a trully standalone service
Jul 07 23:58:37 <egaudet> beelzabub, the wallpaper doesnt' have anything to do with it per se.  Just a side effect of rebooting while the service is running
Jul 07 23:58:46 <beelzabub> don't see how
Jul 07 23:58:53 <beelzabub> unless you're doing something else as well
Jul 07 23:59:05 <beelzabub> like interacting with the wallpaper service
Jul 07 23:59:45 <egaudet> I'm not doing anything with the wallpaper service or anything outside of input stuff
Jul 08 00:00:01 <beelzabub> i think you're hitting upon some other side-effect though
Jul 08 00:00:03 <beelzabub> unrelated to the signal
Jul 08 00:00:17 <egaudet> also if I don't register any service methods, I don't get the same behavior.  But it's very repeatable to take away the sigaction(SIGTERM) and reboot with wallpaper in tact
Jul 08 00:00:28 <beelzabub> (if it's trully a standalone service that's not using PDK or anything like that)
Jul 08 00:00:40 <egaudet> Luna service
Jul 08 00:03:02 <rwhitby> beelzabub: I think it's a case that we're somehow stopping write-back of sqlite preferences data that is cached somewhere.
Jul 08 00:08:56 <egaudet> I don't understand, if my app is running as it's own process, shouldn't catching the signal not impact other processes like LunaSysMgr
Jul 08 00:09:00 <ark3qqq> Interesting. Maybe Warthog _is_ my problem with uinput et al. Someone posted "(oh and I noticed keyboss doesnt recognize the uinput module when I have the warthog kernel installed... Installed Uber Kernel and it worked fine)"
Jul 08 00:09:49 <egaudet> ark3qqq, were you getting uinput message from keyboss?
Jul 08 00:09:52 <ark3qqq> egaudet: After cleanup, does your process exit(0)?
Jul 08 00:10:11 <ark3qqq> egaudet: No, though I'm not sure what that would look like.
Jul 08 00:10:38 <egaudet> so what issue were you running into with uinput?
Jul 08 00:10:40 <rwhitby> ark3qqq: ah, yeah.  all bets are always off with Warthog - it's a developer test-bed, not an end-user kernel.
Jul 08 00:11:07 <egaudet> no my signal handler does not exit(0)
Jul 08 00:11:51 <ark3qqq> egaudet: Heh, I was trying to avoid wasting your time... I have vkb, uinput, and keyboss installed, but I don't have working backspace or space bar from the vkb in, e.g., messaging.
Jul 08 00:12:26 <egaudet> no working backspace in messaging
Jul 08 00:12:34 <egaudet> does your keyboard stop functioning altogether?
Jul 08 00:13:00 <ark3qqq> Hard keyboard seems to work fine. vkb letters etc. work, but no backspace or space bar.
Jul 08 00:13:45 <egaudet> that's odd
Jul 08 00:13:47 <ark3qqq> egaudet: Since I'm running Warthog, don't bother worrying about it. :)
Jul 08 00:14:07 <egaudet> well, it's warthog today but it could have future impact too
Jul 08 00:14:47 <egaudet> well if you go back to warthog, I'd like to know if the uinput module can be inserted cleanly
Jul 08 00:14:56 <ark3qqq> egaudet: My point about exit(0) earlier was that if your process is getting sent a SIGTERM, luna might be expecting a particular return code from it.
Jul 08 00:15:48 <ark3qqq> err, go back? I'm staying with Warthog due to awesome speed and battery life; the vkb is lower priority.
Jul 08 00:16:37 <egaudet> cool, well if you get a chance novaterm in and try to insmod the uinput module for me
Jul 08 00:16:49 <ark3qqq> Oh, easily done with Terminal. One sec.
Jul 08 00:19:15 <egaudet> ark3qqq, exit(0) seems to work btw
Jul 08 00:19:23 <ark3qqq> Can't find insmod. modprobe uinput gives FATAL: Module uinput not found.
Jul 08 00:19:35 <ark3qqq> Not what I expected...
Jul 08 00:20:02 <egaudet> /sbin/insmod not there?
Jul 08 00:20:18 <egaudet> modprobe won't find it because it's not in standard modules area
Jul 08 00:20:48 <egaudet> try /sbin/insmod /media/cryptofs/apps/usr/palm/applications/org.webosinternals.uinput/uinput.ko
Jul 08 00:20:51 <ark3qqq> Okay, insmod is there. Where is
Jul 08 00:24:08 <egaudet> I guess we should lsmod to see if it's in there already
Jul 08 00:24:09 <ark3qqq> Now that's odd. I can't tab past org.webosinternals.u -- you sure about that path?
Jul 08 00:24:28 <egaudet> oh right it's org.webosinternals.modules.uinput sorry
Jul 08 00:25:33 <ark3qqq> Yep, it's already loaded.
Jul 08 00:25:49 <ark3qqq> (the path is quite a bit deeper than that, but tab found it rather quickly)
Jul 08 00:26:28 <ark3qqq> Anyhow: uinput size=10368 used_by=1 with no further info.
Jul 08 00:27:05 <egaudet> lsmod shows it?
Jul 08 00:27:27 <ark3qqq> Yes, that's the content of the uinput line from lsmod.
Jul 08 00:28:19 <egaudet> ls -l /dev/input
Jul 08 00:29:43 <ark3qqq> That's a bit long to convey without pasting. I guess I should novaterm in.
Jul 08 00:29:46 <beelzabub> sorry - had to do something else
Jul 08 00:30:02 <beelzabub> rwhitby: as egaudget pointed out, it's it's own process
Jul 08 00:30:13 <beelzabub> it shouldn't impact anything else like writing out the sqlite database
Jul 08 00:30:15 <ark3qqq> I notice the uinput line does not have g+r though; the other stuff does.
Jul 08 00:30:38 <egaudet> beelzabub, and exit(0) in the signal handler "fixes" it too
Jul 08 00:32:29 <egaudet> ark3qqq, what does keypad0 point to
Jul 08 00:32:44 <ark3qqq> egaudet: to event6
Jul 08 00:32:59 <egaudet> tail -f /var/log/messages |grep keyboss
Jul 08 00:33:58 <ark3qqq> Yep, there are a bunch of them; didn't need the grep.
Jul 08 00:34:35 <ark3qqq> I see event type 1, code 28, value 1 and a correspending 0, 0, 0 at the end.
Jul 08 00:36:18 <egaudet> when you hit backspace what do you see
Jul 08 00:40:08 <ark3qqq> egaudet: Sorry, a bit clumsy with copy-paste. http://paste.pocoo.org/show/234817/
Jul 08 00:46:11 <egaudet> luna-send -n 1 palm://org.webosinternals.keyboss/emulateKey '{"code":14,"keydown":true}'
Jul 08 00:46:19 <egaudet> luna-send -n 1 palm://org.webosinternals.keyboss/emulateKey '{"code":14,"keydown":false}'
Jul 08 00:46:57 <ark3qqq> Suggestion on what I might want open when I do this? Maybe messaging with a few letters already typed?
Jul 08 00:47:06 <egaudet> yeah
Jul 08 00:47:11 <egaudet> memos or messaging
Jul 08 00:47:14 <egaudet> type a few letters first
Jul 08 00:47:32 <egaudet> make sure the tail -f line is going in another window
Jul 08 00:49:06 <ark3qqq> http://paste.pocoo.org/show/234819/
Jul 08 00:49:32 <ark3qqq> First few lines are me pressing "a" a bunch of times. Last few lines are me running the sends twice. No letters were deleted, by the way.
Jul 08 00:55:32 <egaudet> ark3qqq, look for version 0.1.2 of keyboss in feed, it has a little more logging
Jul 08 00:57:23 <ark3qqq> Just updated feeds; it hasn't shown up for me yet.
Jul 08 00:57:32 <egaudet> should soon
Jul 08 01:03:03 <oc80z> ummm 1.4.5 please.
Jul 08 01:03:49 <ark3qqq> Yay, update available.
Jul 08 01:03:57 <ark3qqq> Ack. update to keyboss available.
Jul 08 01:04:01 <ark3qqq> (not 1.4.5)
Jul 08 01:05:28 <ark3qqq> Err, this doesn't sound good: 2010-07-08T01:03:55.571411Z [4114] palm-webos-device user.notice ApplicationInstallerUtility: install: tried to run /med
Jul 08 01:05:28 <ark3qqq> ia/cryptofs/apps/.scripts/org.webosinternals.keyboss/pmPostInstall.script as post-install script - result = 1
Jul 08 01:06:01 <ark3qqq> Oops, sorry about the scroll. Anyway, Preware reported success.
Jul 08 01:08:53 <ark3qqq> egaudet: Anyhow, it all just works now. The luna-sends cause the character to be deleted. More importantly, the vkb now has working backspace and space bar.
Jul 08 01:09:39 <egaudet> interesting
Jul 08 01:09:47 <ark3qqq> egaudet: You fixed it. Hopefully you have some idea why... I'd find it unsatisfying not to know why it suddenly started working were I you.
Jul 08 01:09:57 <egaudet> sorry to disappoint lol
Jul 08 01:10:25 <egaudet> maybe hidd didn't get restarted properly the last time but this time it did
Jul 08 01:10:46 <ark3qqq> I have a working vkb and I get to keep warthog. No disappointment here. Hopefully I was helpful.
Jul 08 01:10:59 <ark3qqq> Or, in other words: Thanks!
Jul 08 01:14:39 <ark3qqq> Anyway, I'm out. I'll ask about novaterm crashing with every keypress next time.
Jul 08 01:18:34 <synackfin1> anyone here?
Jul 08 01:20:37 <synackfin1> hmm
Jul 08 01:43:54 <oc80z> wow
Jul 08 01:58:23 <oc80z> i thought the changelog for 1.4.5 would be not what i am reading..
Jul 08 02:14:04 <bpadalino> too incremental ?
Jul 08 02:14:38 <zsoc> like always, there's a lot snuck in not in the changelog
Jul 08 02:16:18 <oil> small updates monthly/bi-monthly > medium updates yearly
Jul 08 02:20:23 <halfhalo> uggggg
Jul 08 02:20:32 <halfhalo> I hate this xml parser now
Jul 08 02:22:29 * oil thought it was "easy peasy"?
Jul 08 02:22:46 <halfhalo> it no work with new version of node.js
Jul 08 02:22:53 <oil> lol
Jul 08 02:23:15 <halfhalo> Now Im stuck doing it the hard way
Jul 08 02:23:31 <halfhalo> since there is also no getElementsByTagName, etc
Jul 08 02:23:52 <halfhalo> All there are are emmiters
Jul 08 02:38:09 <rwhitby> So, who can set up http://github.com/mape/node-wargames for us for #webos-internals ?
Jul 08 02:41:30 <Tibfib> whoa
Jul 08 02:42:41 <Tibfib> tell halfhalo to
Jul 08 02:43:40 <halfhalo> psha
Jul 08 02:43:59 <halfhalo> thats like 3 lines of typing right there
Jul 08 02:44:51 <Tibfib> halfhalo: do it do it do it
Jul 08 02:45:16 <halfhalo> iamiamiam
Jul 08 02:45:22 <Tibfib> goodgoodgood :)
Jul 08 02:45:29 <halfhalo> i gotta set up a vm type thing for it though
Jul 08 02:46:55 <zsoc> raster: ping
Jul 08 02:53:33 <halfhalo> :/
Jul 08 02:53:39 <halfhalo> its being difficult
Jul 08 02:55:04 <halfhalo> It SHOULD be running… but… its not
Jul 08 02:55:36 <raster> zsoc: boop
Jul 08 02:59:08 <zsoc> raster: what the hell is this galaxy s garbage? it's all.. androidy :/
Jul 08 02:59:37 <raster> it is
Jul 08 02:59:41 <raster> don't look at me
**** ENDING LOGGING AT Thu Jul 08 02:59:57 2010