**** BEGIN LOGGING AT Mon Aug 29 02:59:57 2011 Aug 29 05:18:00 lets chat here Aug 29 05:18:15 did you not want it on the wiki yet? Aug 29 05:18:36 no no...thats good. Aug 29 05:18:39 kk Aug 29 05:18:46 we should publish what we find. Aug 29 05:18:55 I'm torn about putting the novacom on Angstrom instructions on there though Aug 29 05:19:13 I see using novacom as more of a "stop gap" until we get networking up and running Aug 29 05:19:23 like adb? Aug 29 05:19:35 what do you mean? Aug 29 05:19:47 novacom == adb for android Aug 29 05:19:50 yeah Aug 29 05:20:02 novacom == adb, just different platforms Aug 29 05:20:08 but I have novacom working in Angstrom Aug 29 05:20:25 what we need to do is work with Jack87 to document all this. Aug 29 05:20:32 he likes pushing buttons :D Aug 29 05:20:42 yep Aug 29 05:20:55 vol_up recovery Aug 29 05:21:07 vol_dn bootloader Aug 29 05:21:16 pwr power Aug 29 05:21:20 home alt OS? Aug 29 05:21:25 yes Aug 29 05:21:34 power+home is how you hard reset Aug 29 05:21:48 lets not use vol_dn for all os... too much work, have to get the timing right, etc Aug 29 05:21:54 so home would be a much better choice Aug 29 05:22:04 rwhitby, I have had to do that one several times, lol Aug 29 05:22:05 turn it on then release power? Aug 29 05:22:23 before I got novacom working, that was the only way to get out of angstrom short of letting the battery die, lol Aug 29 05:22:27 yeah Aug 29 05:22:37 hold home and power, then once it starts booting just let go of power Aug 29 05:22:50 keep holding home until you see kprints output Aug 29 05:23:25 oh, it looks like that screen corruption is just because of the screen sleeping after all Aug 29 05:23:36 tapped the home button and the screen is all fixed Aug 29 05:23:54 rwhitby, do you know of a good way to base64 decode something in bash? Aug 29 05:24:04 or of a small utility that does it? Aug 29 05:24:10 nope Aug 29 05:24:13 damn... Aug 29 05:24:37 how hard would it be for you to write something tiny in C that would just take input, base64 decode it, give it as output Aug 29 05:25:15 to where I could cat /dev/tokens/WIFICAL | decoder > /tmp/ath6k/token_bdata.SD32.bin Aug 29 05:25:22 static or with lib support? Aug 29 05:25:28 just static Aug 29 05:25:32 preferably Aug 29 05:25:36 so it would be portable Aug 29 05:26:32 I could probably use bootie config to do it... Aug 29 05:26:37 just add an extra option in there Aug 29 05:27:20 I'd rather not mod anything in bootie Aug 29 05:27:29 I'm not talking about modding bootie Aug 29 05:27:33 just reading the tokens Aug 29 05:27:38 kk Aug 29 05:27:41 right now I am using the Palm tokens binary Aug 29 05:27:49 I don't want to use anything proprietary if I can help it Aug 29 05:27:52 ya, can't use that in the future Aug 29 05:28:20 for testing and dev, it's fine... but when it releases to the wild... I want it to be all FOSS if at all possible Aug 29 05:28:21 is it strictly base64 encoded? Aug 29 05:28:28 the WIFICAL token is Aug 29 05:28:53 I took the contents of that token and base64 decoded it... the result MD5 matches the file in /tmp/ath6k/ Aug 29 05:28:58 I'll be jacques could do it in about 10minutes Aug 29 05:29:10 jacques, PING! Aug 29 05:29:20 Fringe. Aug 29 05:29:26 oh... that's right Aug 29 05:31:20 dang...I find myself dumping more and more stuff on the TP and just working directly on it. Aug 29 05:32:36 John Lee Hooker - Waiting in the Wings Aug 29 05:32:44 haha Aug 29 05:38:27 the BT and other things are easy to spot...that Atheros is not easy to find in /sys Aug 29 05:40:57 ah HA! Aug 29 05:41:01 the coreutils package Aug 29 05:41:08 it has a base64 encoding and decoding program Aug 29 05:42:11 okay looks like I need to head for the source and find AR6000 or ar6k or atheros or whatever they call it. Aug 29 05:42:31 I think that may be buily as part of the kernel source Aug 29 05:42:39 and it is definitely patched by palm Aug 29 05:42:46 busybox has a uudecode applet (not builtin on webos version though) Aug 29 05:44:12 doesn't look to be there in the Angstrom one either Aug 29 05:44:45 I will probably just write a C program based on bootie-config to mount the tokens and with a flag do the base64 decode of the WIFICAL token Aug 29 05:44:59 I would still love to know what is in that token though Aug 29 05:45:14 I know it is base64 encoded binary data... but what is hiding in that binary data... Aug 29 05:45:30 apparently uuencoding and base64 are similar but not identical. the busybox uuencode applet can be told to do base64 encoding, which leads me to believe the uudecode applet would decode base64 Aug 29 05:46:10 yeah, but that doesn't help if the uudecode applet isn't there Aug 29 05:46:13 which it isn't Aug 29 05:46:17 in webOS or Angstrom Aug 29 05:46:20 so we need one.... Aug 29 05:46:25 jacques, is that easy? Aug 29 05:46:36 the coreutils package has a base64 encoding and decoding program in it Aug 29 05:46:40 I already said that Aug 29 05:46:53 then we are golden right? Aug 29 05:46:57 I can just cat the token, pipe to base64 -d and output to wherever I need Aug 29 05:47:00 not completely Aug 29 05:47:10 I need to have something to get the data from the token in the first place Aug 29 05:47:21 right now I am using the palm binary, I want to use something else Aug 29 05:47:29 likely something based off of bootie-config Aug 29 05:47:47 basically make my own FOSS version of the palm tokens binary Aug 29 05:50:10 where exactly is the token data? Aug 29 05:50:19 in mmcblk0 Aug 29 05:50:22 it isn't in a partition Aug 29 05:50:26 it isn't in a filesystem Aug 29 05:50:30 in a raw partition? if so, you could just dd it Aug 29 05:50:50 read the offset? Aug 29 05:50:56 is there is one Aug 29 05:53:51 cryptk, see this? "Token calibration file already exists. Skipping token data conversion." Aug 29 05:53:56 yes Aug 29 05:54:05 that is if the file in /tmp/ath6k already exists Aug 29 05:54:13 the problem is that on initial boot it doesn't Aug 29 05:54:33 so then it would be created... by taking the WIFICAL token, base64 decoding it to that file Aug 29 05:55:02 and just DDing it out isn't that simple... Aug 29 05:55:11 you need to know the start position of the token and the length of it Aug 29 05:55:22 which needs to be parsed out of the token header data Aug 29 05:55:36 if you have that information then you can just dd it out... but not until you have the offset and length Aug 29 05:55:58 dd if=mkblk0 offset=XXXX bytes=yyyy of=/tmp/ath6k then base64decode Aug 29 05:56:06 yeah Aug 29 05:56:13 but you need to get that offset and the lenght Aug 29 05:56:34 and if I am going to have a C program to get that info for me... then it is just 2 more lines to make that same program do the decode Aug 29 05:56:40 mmcblk0 is the whole device - which partition is it in? Aug 29 05:56:57 beginning of device Aug 29 05:57:18 with an offset Aug 29 05:57:24 actually I think it may be in mmcblk0p1 Aug 29 05:57:30 even better Aug 29 05:57:32 but it is a raw partition Aug 29 05:57:35 no filesystem Aug 29 05:57:40 and therefore it isn't mountable Aug 29 05:57:47 the dd is the only way then. Aug 29 05:57:52 no it isn't Aug 29 05:57:58 easiest...sorry Aug 29 05:58:00 we already have a tool to do 99% of this Aug 29 05:58:03 that is what I have been saying Aug 29 05:58:04 http://git.webos-internals.org/tools/bootie-config/ Aug 29 05:58:13 that has 99% of what we need already Aug 29 05:58:23 just because it is in C doesn't mean we can't use it Aug 29 05:58:25 so what do you need...the offset and length? Aug 29 05:58:33 yes Aug 29 05:58:40 okay here is what I'd do Aug 29 05:58:42 I need the offset and length of the WIFICAL token Aug 29 05:58:47 I can get that from bootie-config Aug 29 05:58:48 reencode it to base64 Aug 29 05:58:57 then do a hex match of the pattern. Aug 29 05:59:11 ok, and what good does that do me Aug 29 05:59:23 it's offset and length can change from device to device and before/after a doctor Aug 29 05:59:28 it won't be in the same place every time Aug 29 05:59:43 so please please please just listen to what I am saying here Aug 29 05:59:49 understood. Aug 29 05:59:52 however Aug 29 05:59:52 bootie-config does 99% of the job already Aug 29 06:00:02 if thats a cal then it shouldnt change right? Aug 29 06:00:09 it has the logic to read the header info and get the offset and the length Aug 29 06:00:22 does it change? Aug 29 06:00:26 from that I can literally add about 2 lines and just have it spit out the file we need Aug 29 06:00:32 ka6sox, I would suspect it does Aug 29 06:00:39 I have no clue what is stored in that cal Aug 29 06:01:03 the atheros driver is FOSS right? Aug 29 06:01:07 wish I had od on the tp Aug 29 06:01:07 but I have a feeling it is a conglomeration of all of the other wifi related tokens and a little extra Aug 29 06:01:21 or does it have binary' blobs? Aug 29 06:01:21 ka6sox, I don't think so - not the one on the tp Aug 29 06:01:25 yes the atheros driver is FOSS... the TP is using a patched version Aug 29 06:01:39 that patched version may or may not be FOSS Aug 29 06:01:44 many atheros devices have a proprietary and FOSS driver Aug 29 06:01:52 kk Aug 29 06:02:33 that is why I am saying it would be easiest to jsut add the two lines to bootie-config that we need, strip out what we don't... then we have the beginnings of an FOSS PmWiFiService Aug 29 06:02:40 license: GPL and additional rights Aug 29 06:02:54 fine Aug 29 06:02:59 do it however you want Aug 29 06:03:03 I'm taking a break, bbiaf Aug 29 06:03:11 calm down Aug 29 06:03:17 i asked a question Aug 29 06:03:18 was just correcting myself Aug 29 06:03:21 yeah Aug 29 06:03:22 :D Aug 29 06:03:32 we aren't going a different direction Aug 29 06:03:40 I am back to finding the power pin. Aug 29 06:03:42 I gotta get sleep anyway. 1am and I have work in the morning Aug 29 06:03:49 yup Aug 29 06:05:15 I don't understand the bootie-config part Aug 29 06:05:41 I just did a strings on first MiB of /dev/mmcblk0p1 and it looks like a FAT partition Aug 29 06:06:06 I am used to having a /proc/mtd file which tells which blocks are in which partition Aug 29 06:07:37 thats what I'm used to as well. Aug 29 06:08:16 http://pastebin.com/KXatYT5z is how they are extracting the PBL Aug 29 06:14:08 sigh. alarm clock on tp is pretty broken Aug 29 06:27:14 Tide does a good job of source browsing. Aug 29 06:27:28 the filepicker needs help though. Aug 29 07:22:19 ok, now to hack up bootie-config to get what I need Aug 29 07:24:05 jacques, it's mmcblk0p12 my bad Aug 29 07:24:09 that's there the tokens are at Aug 29 08:31:43 Hi Jack! Aug 29 08:34:06 ka6sox, hello! Aug 29 08:34:20 i was about to head to bed but had to say hi before i take off Aug 29 08:34:39 ya, I should be in bed too.... Aug 29 08:34:44 tired of readin source Aug 29 08:34:54 ka6sox, by the way i doctored my TP last night and set up 4 gig partition for chroot Aug 29 08:34:58 is that large enough? Aug 29 08:35:00 yup Aug 29 08:35:34 what do i need to do to install ubuntu chroot? (is it simple as installing package from preware?) Aug 29 08:35:45 I don't know if I'll ever get time to finish working on Tailor Aug 29 08:35:47 :( Aug 29 08:36:00 (or I might be done already) Aug 29 08:36:46 Jack87: UbuntuChroot wiki pae Aug 29 08:36:47 done = good? Aug 29 08:36:47 page Aug 29 08:36:57 * Jack87 is hunting for it Aug 29 08:37:14 * Jack87 prolly should not look at wiki or he wont sleep tonight Aug 29 08:37:46 I have to remember the command to make my TP not turn off the BL while I'm working on source. Aug 29 08:38:57 BL? back light? Aug 29 08:39:46 backlight Aug 29 08:39:50 * Jack87 found wiki.. oh boy here it starts Aug 29 08:40:06 Jack87, go to bed, do NOT read that page, Aug 29 08:40:12 turn it off now. Aug 29 08:40:32 when do you think we can do some poking around? Aug 29 08:41:07 tuesday night the soonest but latest thursday night :) so real soon Aug 29 08:41:18 okay great Aug 29 08:41:23 we have to find a number of things Aug 29 08:41:30 in sysfs. Aug 29 08:41:42 so we can use them on other things. Aug 29 08:41:48 like turning on and off the wifi Aug 29 08:42:37 ya that would be really handy Aug 29 08:43:30 * Jack87 needs to come up with key combinations to bring up command buttons and over lays to go with em Aug 29 08:44:24 all missing keys/commands we can create an over lay that comes up with combination of the TP buttons pressed. kinda like we did alternatives keyboard and overlay for quake Aug 29 08:56:31 fun stuff....okay I'm done reading source tonight Aug 29 08:56:58 I would have thought it was in arch/arm/arm-msm but I guess I'm missingit. Aug 29 09:35:01 http://git.cryptkcoding.com/Angstrom-TP/WIFICAL-pull/ Aug 29 09:35:03 there ya go Aug 29 09:35:18 it's a binary, when you run it, it just spits out the contents of the WIFICAL token Aug 29 09:35:46 eventually I will make it take a file destination as an argument and it will just kick you out the base64 decoded file that we need Aug 29 09:41:36 cryptk, is this git new? Aug 29 09:42:17 that's my personal git Aug 29 09:45:31 * Jack87 has never seen it before Aug 29 09:47:42 yep, usually isn't much in there Aug 29 09:49:14 what are some of the known "issues" that are being worked through for xserver on TP? Aug 29 09:50:11 not necessarily worked thought but have been brought to attention? Aug 29 10:10:53 well, the omapfb driver for X is the only one in my angstrom rootfs and it doesn't work with the TP Aug 29 10:10:59 that is the primary issue Aug 29 10:11:18 that is the point at which the X init stops according to it's logs Aug 29 14:46:31 dtzWill: do you have a list of things that needs to be done with X? Aug 29 14:48:41 elpollodiablo1: nope, not formally. probably should. Aug 29 14:51:00 alright well I planning on getting into working on something simple just to get familiar with it. and it also might be good for other to be like ok that is on the radar I dont need to ask. Aug 29 15:50:18 dtzWill: is the esc issue a BT keyboard issue only? Aug 29 15:52:29 elpollodiablo1: I think it's eveywhere, the ESC key isn't even reported to the Xserver by the SDL libraries :( Aug 29 15:52:49 (doesn't work with a virtual keyboard either in X) Aug 29 15:54:42 NuttyBunny: ah, I thought it worked in a vkb inside chroot. maybe I was mistaken. thanks I will try and see. Aug 29 15:56:59 would we beable to pass another key to the Xserver and have that interperate that as esc? Aug 29 15:57:44 dtzWill or NuttyBunny for VKB so we dont have to worry about the low level capture? Aug 29 15:58:11 NuttyBunny: not working with vkb is strange, i can look into that. Aug 29 15:58:42 i assume you mean xvkb (or similar, just calling them all xvkb to differentiate from webos's vkb) Aug 29 16:00:07 dtzWill: The vkb crashes with shift as well :) (xvkbd, that's the one I'm testing with) Aug 29 16:06:23 lol of course it does Aug 29 16:10:08 I know, I assumed that you were thinking on a way to fix it ;) Aug 29 17:00:22 oh #$%!#%# Aug 29 17:00:23 lol Aug 29 17:00:33 after debugging more than i wanted to Aug 29 17:00:57 i'm either doing something wrong, or some pieces of code in the x server don't like 24bits Aug 29 17:01:23 i changed to 32 in the same way i set 24 Aug 29 17:01:34 and those random crashes (lxterminal, etc) go away Aug 29 17:01:34 -.- Aug 29 17:03:16 isnt it suggeted that PDK use 32bit images? Aug 29 17:03:50 elpollodiablo1: unrelated, and that's if you're using their SDL wrapper (it'll convert it to 32bit anyway or someting, don't remember) Aug 29 17:03:54 well i think it's unrelated anyway Aug 29 17:04:14 the crashes are /not/ in my code or related to my rendering logic Aug 29 17:04:34 I was wondering about how you were doing the vkb with out the sdl wrapper when I was looking through the code before Aug 29 17:04:53 well i'm using SDL Aug 29 17:04:58 and using sdl+gles for rendering Aug 29 17:05:03 and vkb is done with PDL_Setkeyboardstate(1) Aug 29 17:05:26 dont you have to have pdl.h? Aug 29 17:05:41 or did I just miss where you did that. Aug 29 17:05:42 elpollodiablo1: nah, i mean it'd be nice Aug 29 17:05:51 but i just define the prototypes myself at the top of sdl.c Aug 29 17:06:18 oh alright still learning some of this stuff Aug 29 17:07:26 apparently x uses 32bit pixmaps for 24bpp buffers... hrrmm Aug 29 17:09:33 o_O this is lame lol Aug 29 17:09:54 fine, i'll just cave and use 32bit Aug 29 17:10:06 yay wasted memory bandwidth Aug 29 17:10:33 that was my question and you asnwered it Aug 29 17:54:27 anyone here have a pixi (that they use? xD) Aug 29 17:54:55 wish my stupid pixi would turn on, just feel bad shipping binaries i haven't even verified run xD Aug 29 18:03:18 mkay, well pushed it. we'll see what happens XD Aug 29 18:03:31 dtzWill: Pixi, 2.1, will it work for you? Aug 29 18:03:43 NuttyBunny: yep, should be fine. Aug 29 18:03:52 just iirc there's something about pixi supporting different GPU formats Aug 29 18:04:01 texture formats w.r.t gpu goodness Aug 29 18:04:13 anyway i'm worried latest version won't work at all on pixi devices Aug 29 18:04:29 so I should install.. xecutah, xserver, xterm or you need the ubuntu chroot as well? Aug 29 18:04:52 NuttyBunny: so when v0.8.2 of xserver comes out, give that a go if you would. I'll ping you when it hits the feeds if that's useful Aug 29 18:04:59 NuttyBunny: xecutah/xserver/xterm is sufficient Aug 29 18:05:10 xterm optional xD, just worried it won't even start Aug 29 18:06:33 ok, lemme find a working battery, this is dead :) Aug 29 18:06:36 oh wow that was fast Aug 29 18:06:40 (already in feeds) Aug 29 18:06:57 guess i hit timing just right (ticks every 15min, right?) Aug 29 18:07:46 cool! (no idea on the ticks) :) Aug 29 18:09:07 argh, both spares are dead... you'll have to wait a few till it charges enough to boot Aug 29 18:11:08 NuttyBunny: okay np at all Aug 29 18:11:15 NuttyBunny: actually looks like someone reported pixi success Aug 29 18:11:39 but *tyvm* and i might bug you later when i'm potentially breaking things on the pixi ;) Aug 29 18:11:48 what an honor, i know Aug 29 18:11:53 ...kidding aside thanks man ^.^ Aug 29 18:11:55 hehehe Aug 29 18:12:09 whenever you need it, if I'm here I can help :) Aug 29 18:13:30 very much appreciated, thanks :) Aug 29 18:52:29 dtzWill, is it possible to open a special overlay with specilaty keys using a combination of physical keys? Aug 29 18:52:53 by physical keys i mean physical buttons Aug 29 18:53:04 such as vol/pwr Aug 29 19:01:27 Jack87: possible? yeah :). just steal some space from X, and put out overlay on top of it Aug 29 19:01:43 although i supose since we're using GLES atm we could always make it a transparent overlay and have little/no performance hit Aug 29 19:01:49 opaciy controlled via tweaks :P Aug 29 19:02:29 rwhitby/those who might know: can I do a luna-send from a pdk application? Aug 29 19:02:45 i guess via as unjailed oot process i can do the hell i want Aug 29 19:02:52 just looking to remove the luna-send tweaks delay on startup Aug 29 19:02:55 nbd, but just a thought. Aug 29 19:05:35 * dtzWill apt-get install's gnome and kde Aug 29 19:21:12 dtzWill, indeed! Aug 29 19:21:36 dtzWill, we can perhaps do something similar to quake Aug 29 19:22:08 or something similar to what we did with quake Aug 29 19:22:44 but i wonder if we can just make a call to make the overlay come up.. not nessiairly always there. kinda like the alternative keyboard controls quake had Aug 29 19:22:49 indeed, only that was a bit of a kludge, we can do this much better Aug 29 19:22:57 :) Aug 29 19:23:04 and yeah being able to add/remove it would be nice Aug 29 19:23:11 maybe hijack the keyboard dismiss button Aug 29 19:23:12 quake was your first! or first of sdk pretty much Aug 29 19:23:14 to toggle the overlay Aug 29 19:23:25 yeah, and it should have been done using the gpu to do the blending Aug 29 19:23:25 dtzWill, that sounds like an idea :) Aug 29 19:23:31 as it doesnt even do anything right now Aug 29 19:23:53 however how will we make it comforh using physical keyboard? Aug 29 19:23:59 cntrl + a key? Aug 29 19:24:17 actually.. physical palm keyboard has a keyboard/dismiss button on it Aug 29 19:25:30 so it does, lol Aug 29 19:25:36 didn't realize that hehehe Aug 29 19:25:58 :) Aug 29 19:30:10 dtzWill, what are some important keys we are missing?? Aug 29 19:30:39 alt/super/ins/del/home/end/pgup/pgdwn/esc Aug 29 19:34:27 dtzWill, anything i can do to help? Aug 29 19:35:00 dtzWill, oh any way to get right click? (its so hard without meta tap :( ) Aug 29 19:54:14 Jack87: escape is big one Aug 29 19:54:25 pageup/pagedown too, but i have the hp bt keyboard which doesn't hvae those anyway Aug 29 19:54:26 :P Aug 29 19:54:34 control/alt/arrow keys for non-bt csae Aug 29 19:54:42 maybe you tap the dismiss button and it cycles between overlays Aug 29 19:54:43 hehe Aug 29 19:54:48 ya Aug 29 19:54:59 (including none) Aug 29 19:55:06 i was thinking cntrl + up/dwn for pg up/dwn Aug 29 19:55:13 and left/right for home/end Aug 29 19:55:36 hehe now you are thinking Aug 29 19:55:53 multiple overlays = multiple key options = macros! Aug 29 19:56:01 who remembers using macros :) Aug 29 19:56:17 it be great for on screen tapping for a series of common commands Aug 29 20:32:36 Xserver is acting a fool on my pre Aug 29 20:33:47 it loads and is just a balck screen untill I go to card view. then it shows the normal Xserver look. Aug 29 20:34:24 then go back to it and click and the screen turns black again Aug 29 20:41:54 oh by the way I just updated to the Xserver 0.8.3 Aug 29 20:54:00 elpollodiablo1: sigh, okay so it's not just that guy on p|c Aug 29 20:54:00 damn. Aug 29 20:54:03 but but but Aug 29 20:54:04 :( Aug 29 20:55:22 elpollodiablo1: anything interesting in /tmp/xserver.log? Aug 29 21:14:21 that black in full-screen, Ok in cardview thing is like the problem we had in the old days Aug 29 21:14:41 linuxjacques: indeed, and it was b/c we initialized and rendered on different threads. Aug 29 21:14:45 which turned out to be a threading issue IIRC (according to dtzWil) Aug 29 21:14:56 atm i'm thinking it's because of uninitialized alpha values Aug 29 21:14:56 but Aug 29 21:14:56 :/ Aug 29 21:15:06 yeah that could be it too Aug 29 21:15:12 at one point I thought it was that Aug 29 21:15:15 maybe i can get away with just initializing them once myself Aug 29 21:15:20 why not? Aug 29 21:15:35 if i have to do it per-update (sanitize the alpha channel) that'd be lame Aug 29 21:15:38 did this start after you changed 24bit -> 32bit ? Aug 29 21:15:41 oh i could probbaly just hardcode it in the shader Aug 29 21:15:41 lol Aug 29 21:15:43 linuxjacques: yep Aug 29 21:16:08 interesting :-( Aug 29 21:18:46 maybe something lower-level was zeroing the alpha channel for you before Aug 29 21:19:12 meaning, it would not necessarily get slower if you do it yourself now (but I haven't looked at the code in forever) Aug 29 21:19:44 dtzWill: you are using a shader? Aug 29 21:22:59 dtzWill: nothing in /tmp/xserver.log which seem kind of strange. Aug 29 21:25:46 dtzWill: hopefully we can get this out of testing feed pretty soon so these kind of issue do not cause mass riots anymore. Aug 29 21:27:41 elpollodiablo1: indeed. Aug 29 21:27:55 linuxjacques: i am, so yeah i think icould just do it there. Aug 29 21:28:00 hmm. node 0.5 on my TP? I THINK YES! Aug 29 21:29:00 mostly because it builds faster on that than my laptop. Aug 29 21:31:33 anone with a pre who has the x issue, can you test http://ompldr.org/vYTQ0MQ ? Aug 29 21:31:37 elpollodiablo1: if yuo have a second :) Aug 29 21:31:38 bbl Aug 29 21:34:56 dtzWill: what would be the easiest way to dl that straight to my pre? Aug 29 21:39:07 is there a dahnldr app? Aug 29 21:40:44 I am heading home now I will try when I get there. ~45min Aug 29 22:11:26 elpollodiablo1: great, ty Aug 29 22:11:47 elpollodiablo1: wget http://ompldr.org/vYTQ0MQ -O /media/cryptofs/apps/usr/palm/applications/org.webosinternals.xserver/bin/Xsdl Aug 29 22:13:41 dtzWill: if he can't run the xserver, how will get it? ;) Aug 29 22:13:47 (he) Aug 29 22:16:11 lol :(. novaterm :( Aug 29 22:16:18 or bluetooth keyboard and no typos Aug 29 22:21:16 I prefer ssh Aug 29 22:22:40 linuxjacques: yeah me too hehe Aug 29 23:32:23 dtzWill: I was using wget earlier but it just saved the file as vYTQOMQ so I wasnt sure if that was the actual file or just a redirect page. and I had another terminal app on my pre so it is all good. Aug 29 23:33:42 elpollodiablo1: kk, well it's the replacement Xsdl so just copy it over the other and we'll see what happens Aug 29 23:33:52 (I can upload the original xsdl later if that's useful) Aug 29 23:33:55 no change Aug 29 23:34:03 still the black screen issue Aug 29 23:34:17 okay Aug 29 23:34:55 elpollodiablo1: did 0.8.2 work for you? (did you try it earlier?) Aug 29 23:35:33 I dont think I ever had 0.8.2 when did you update to that? Aug 29 23:36:17 no worries. pushed them back to back this morning. Aug 29 23:36:28 the last one I had was I think 0.7.4 the tweaks keyboard one. Aug 29 23:38:05 elpollodiablo1: would you try http://ompldr.org/vYTQ1aA ? Aug 29 23:38:11 same thing, won't work, but with debug logging enabled Aug 29 23:38:12 wait Aug 29 23:38:12 nvm Aug 29 23:38:14 1 sec Aug 29 23:38:14 lol Aug 29 23:38:31 lol ok just let me know Aug 29 23:38:43 (i have separate general debugging and GL-debugging, enabled just teh general one in thta build) Aug 29 23:40:33 elpollodiablo1: http://ompldr.org/vYTQ1ag if you would Aug 29 23:40:50 run it for a bit (10 seconds or so should be fine) and shoot me /tmp/xserver.log ? Aug 29 23:43:29 want me to click or go to card view or anything? Aug 29 23:49:52 does seem like anything is being written to /tmp/xserver/log Aug 29 23:51:03 elpollodiablo1: doesn't? Aug 29 23:51:10 err /tmp/xserver.log, not /tmp/xserver/log Aug 29 23:51:18 well then that's an issue o_O Aug 29 23:51:18 lol Aug 29 23:51:20 it is 0 bytes Aug 29 23:51:24 lol! Aug 29 23:51:24 well kay Aug 29 23:51:30 I put . I just cant type some time Aug 29 23:52:03 although it says it was modified recently Aug 30 00:06:00 nah it's probably getting stuck before anything else happens Aug 30 00:06:01 strange Aug 30 00:06:36 will have to debug with my pre later, after i revive it hehe Aug 30 00:24:09 ok just let me know if you need me to do anything else Aug 30 00:35:51 elpollodiablo1: tyvm, will do. don't think i'll have time to poke at it tonight sorry :( Aug 30 00:36:02 try 0.8.2 or earlier if you want a working x server again :( Aug 30 00:36:22 fwiw @ all, i've had no issues using debian's dbus Aug 30 00:36:25 just...started it Aug 30 00:36:25 lol Aug 30 00:36:36 so kde worked easy peasy Aug 30 00:36:40 gnome i haven't tried yet Aug 30 00:36:43 but just FYI Aug 30 00:37:59 yeah, that was one driver for getting debian in quickly Aug 30 00:38:27 shame synergy in these guys is a PITA lol Aug 30 00:38:46 synergy in debian has a known issue, and grabbed one from testing and i get all caps (not sure if that's our xserver's fault or what) Aug 30 00:39:07 otherwise my tp chilling on its ts is a nice little third monitor for misc stuff like IRC ^.^^^^^^^ Aug 30 00:39:20 which it...already is, but i hafta switch keyboards (the horror!) Aug 30 00:48:31 would we not say that the BT keyboard is "fully" (as well as we can) supported an the redmine ticket is satisfied? Aug 30 00:49:53 ka6sox: hmm, i suppose so! Aug 30 00:49:55 \o/ Aug 30 00:50:30 scoutcamper, you can close that one...(or can I?) Aug 30 00:51:01 ka6sox, im not an admin on that redmine Aug 30 00:51:07 ka6sox, you and dtzWill can Aug 30 00:51:20 dtzWill, want the honors? you did it. Aug 30 00:51:22 yep Aug 30 00:51:37 not if you already have it open Aug 30 00:51:37 O:) Aug 30 00:52:56 not open Aug 30 00:53:02 you are in charge.... Aug 30 00:56:19 no idea where this even is, i don't see such a bug/issue Aug 30 00:56:19 lol Aug 30 00:56:21 * dtzWill redmine fail Aug 30 00:56:22 :( Aug 30 00:56:59 iirc its #259 Aug 30 00:57:35 http://redmine.webos-internals.org/issues/527 Aug 30 00:57:38 that onee Aug 30 00:58:32 oh well we don't autodetect yet Aug 30 00:58:37 which would be nice Aug 30 00:58:45 and if i cared i can't imagine would be super hard Aug 30 00:59:06 surely /something/ is different when bt is attached, even if we don't have useful tools to list attached devcies Aug 30 00:59:16 (and it's a lot easier if we do hehe) Aug 30 01:03:19 since tweaks is a set one time and forget...teh keyboard shouldnt' change Aug 30 01:03:25 so autodetect is useless. Aug 30 01:03:43 unless we take it out of tweaks. Aug 30 01:03:43 heh yeah. until we allow dynamic resizing :) Aug 30 01:03:50 right Aug 30 01:03:54 so I'd say close i8t Aug 30 01:04:06 until xrandr works then its moot Aug 30 01:13:54 haha closed in most unnecessarily verbose way possible O:) Aug 30 01:14:21 perfect :D Aug 30 01:21:22 Preware visual annoyance: the status text is too close to the big spinner on the installing/loading/etc screens. Does anyone else agree? Aug 30 01:31:23 Xuzz: device-specific spacing of the update/install screens is an awesome thing to fix. Aug 30 01:31:32 rwhitby: that's on all devices Aug 30 01:31:44 and I mean like 4 pixels here Aug 30 01:32:18 oh, sure. I have no design sense, so check with oil Aug 30 01:32:39 make the change, post screenshot showing difference, done :) Aug 30 01:34:27 okay :p **** ENDING LOGGING AT Tue Aug 30 02:15:47 2011 **** BEGIN LOGGING AT Tue Aug 30 02:16:14 2011 Aug 30 02:16:31 rwhitby: I'm not sure if you saw "booto"; he was in #webos-internals, and he said he's looking for a better signal/noise ratio for a WebOS channel. Aug 30 02:16:56 I see the topic say to invite other developers, but I'm going to wait on that until someone else says it's okay. Aug 30 02:17:17 booto is good to invite here in private **** ENDING LOGGING AT Tue Aug 30 02:18:15 2011 **** BEGIN LOGGING AT Tue Aug 30 02:20:15 2011 Aug 30 02:27:59 booto: welcome Aug 30 02:29:18 thanks Aug 30 02:30:11 was looking for a place with a better snr than internals :) Aug 30 02:31:23 booto: welcome, and understood. it's why this channel exists :) Aug 30 02:39:43 anyone who asks end-user questions in here gets booted, unless they have contributed stuff already ;) Aug 30 02:44:04 rwhitby: looks like the deps for that usbip userland stuff are libsysfs + glib-2.0 ... I built libsysfs from sysfsutils, and grabbed the glib-2.0 from the device itself Aug 30 02:44:33 are they in the WIDK ? Aug 30 02:45:39 operation replace-on-device-service-launcher-with-a-faster-method is a go. this plus using a widk built node replacement should net a significant decrease in service launching time. Aug 30 02:46:40 operation halfhalo is making us jealous is a succsess Aug 30 02:47:27 halfhalo++. i knew you were poking at replacing node, but what're you doing with the service launching? that's excellent :D Aug 30 02:48:56 dtzWill: I'm betting its not going to work since I'm using a node version thats waaaayyyy different (0.5.5 vs the shipping 0.2.3), but the pure replacement of 0.2.3 with a widk 0.2.3 makes a big difference in itself. like, widk is twice as fast. Aug 30 02:50:40 halfhalo: that's promising--however, i'm concerned--do we have any ideas *why* their version would be so much slower? Aug 30 02:50:59 surely our code isn't faster just b/c it was built with the WIDK ^.^ Aug 30 02:51:01 maybe theu turned on every single debugging feature and compiled with -O0 Aug 30 02:52:53 it has to do with how its built Aug 30 02:55:17 lol okay -.-. "they got different (slower) binary from the same source" -> "they built it differently". no wai ;) Aug 30 02:55:57 anyway anyway, nevermind me, i guess i'm just curious now. good news, halfhalo :D Aug 30 02:56:02 halfhalo, any idea if thier node uses both cores? im sure ours does Aug 30 02:57:21 the only thing different is how its built. the code is the same, so if one uses both than the other does. except for the fact node is single threaded (except with my forky. heh) Aug 30 02:58:06 halfhalo: wonderful, thanks. i'm sure you know what you're doing :) Aug 30 02:59:22 * scoutcamper trusts halfhalo Aug 30 02:59:40 but i think .5 had issues when i tried it Aug 30 02:59:54 it does **** ENDING LOGGING AT Tue Aug 30 02:59:57 2011