**** BEGIN LOGGING AT Thu Nov 25 02:59:58 2010 Nov 25 03:00:07 Hahaha Yes, it is a good joke :) Nov 25 03:00:53 but it would be nice to be able to edit it :) more customizability is what Homebrewing is all about :) Nov 25 03:01:31 rick-home, Shaikh http://www.webos-internals.org/wiki/Change_just_type Nov 25 03:02:36 Awesome! Nov 25 03:02:50 Thats exactly what I was looking for haha Nov 25 03:02:58 that is thanks to oil for discovering and me for documenting :D Nov 25 03:04:49 Yes! Thank you guys! Nov 25 03:05:01 good luck making it configurable Nov 25 03:05:19 i couldn't find the js responsible for just type in the card view Nov 25 03:05:22 i suspect its not in js Nov 25 03:05:25 oil, i just provided it for those of us who want to make our own :D Nov 25 03:05:33 thanks to you :D Nov 25 03:05:34 was only able to do it by changing the localization file Nov 25 03:07:33 oil, i still like mine, it says "Dont push my buttons" Nov 25 03:08:21 heehee Nov 25 03:10:16 destinal: so for this hidd stuff, you just want to be able to monitor the socket information and figure out what's being passed along ? Nov 25 03:11:03 bpadalino: we need to figure out what is sent on both sockets as well as what to do with the plugintable function call Nov 25 03:11:41 I got a bit of info yesterday on that call Nov 25 03:11:42 sec Nov 25 03:11:44 sure Nov 25 03:13:29 PuffTheMagic, Ping! Nov 25 03:16:01 bpadalino: from what it looked like in IDA, PluginTable returns pointers to: Nov 25 03:16:03 Init, Exit, Suspend, Resume, Poll Nov 25 03:16:14 rick-home, do the instructions on the http://www.webos-internals.org/wiki/Change_just_type page make sense? Nov 25 03:16:16 and PluginTable is the export Nov 25 03:16:33 or anyone else? Nov 25 03:16:53 PluginTable .. okie dokie Nov 25 03:17:41 I'd have to guess it returns a struct of 5 pointers since I was unfortunately looking at the ARM code Nov 25 03:17:48 let me try the x86 which is a little more my speed Nov 25 03:17:52 ok Nov 25 03:18:09 i am bootig up the emu again .. so i can play around Nov 25 03:18:59 so for that test we get pretty far... luna gets ready to draw... and then Nov 25 03:19:01 bpadalino: if you want to make a quick wrapper that dloads a plugin and calls that function with a declaration that it returns a pointer to a struct of 5 pointers, that would help Nov 25 03:19:02 QGLShader::compile: "Compile failed. Nov 25 03:19:03 " Nov 25 03:19:03 Vertex shader for simpleShaderProg (MainVertexShader & PositionOnlyVertexShader) failed to compile Nov 25 03:19:03 QGLShader::link: "Link Error: Vertex shader is missing. Nov 25 03:19:30 also for shader blitShaderProg Nov 25 03:20:00 egaudet, is there a way to see the webos 2.0 boot up on like rwhitby or oil s pre 2 to see where we are in terms of hurdles to cross? Nov 25 03:20:36 i would but i dont have my cable with me :( Nov 25 03:20:43 destinal: i am trying to stop hidd in the middle of execution and print out some info.. Nov 25 03:22:06 scoutcamper: you don't make any sense.. Nov 25 03:22:16 scoutcamper, not sure what else I could gain from a pre2 with regards to this Nov 25 03:22:33 egaudet, you are freezing at a boot up process right? Nov 25 03:22:54 it's crashing after luna boots and tries to draw Nov 25 03:22:55 egaudet: I guess it may be interesting to try palm's kernel and ddk on a pre2 just to see if it works there Nov 25 03:23:09 like whether palm gave us something that doesn't match what they use Nov 25 03:23:14 destinal, yeah that I want to see for sure Nov 25 03:23:33 we have dspbridge driver too Nov 25 03:23:50 what else would we need to go to a newer kernel? Nov 25 03:25:06 egaudet, just for a "how far away are we" could you get a boot-up log from a Pre 2 to see what else is left after it draws luna, what else does the kernel do before webos loads to the ui and launcher etc. Nov 25 03:26:36 does that make sense or am i insane? Nov 25 03:27:42 egaudet: by a newer kernel what do you mean.. Nov 25 03:28:02 2.6.36 Nov 25 03:28:30 I don't know why but Nov 25 03:28:36 egaudet: hmm, don't know Nov 25 03:45:53 * oil can get whatever log is needed Nov 25 03:54:56 bpadalino: anything interesting? Nov 25 03:55:47 nothing yet .. just frustration Nov 25 03:58:54 * halfhalo will donate a box of cookies to the first person to do whatever it is you are trying to do to work Nov 25 03:59:05 lol Nov 25 03:59:47 hey oil Nov 25 03:59:51 hey Nov 25 03:59:54 * halfhalo is not kidding Nov 25 04:01:05 * ka6sox looks thru 3714 pages of doco to find 1 number :P Nov 25 04:01:42 hah Nov 25 04:01:43 sure your probably busy with getting thins working in 2.0, bu how goes your xbox app? Nov 25 04:01:52 things* Nov 25 04:01:54 havent touched it since i left for devdays Nov 25 04:01:59 halfhalo, would that include diagnosing my christmas display? bcuz i just did that :D Nov 25 04:02:07 no Nov 25 04:02:13 :( Nov 25 04:02:37 i do christmas lights to music, i am trying to write a remote-access app for the program we use Nov 25 04:04:08 oh, you're one of "those" people Nov 25 04:04:39 The type that put up the lights in june? Nov 25 04:04:48 and put them to music Nov 25 04:04:55 so their neighbors get no sleep with all the flashing lights Nov 25 04:05:07 HAH Nov 25 04:05:20 break out the bb gun! Nov 25 04:05:55 well, they just gotta sneak over and take one bulb Nov 25 04:05:58 then the whole thing goes dark Nov 25 04:05:59 hahaha Nov 25 04:06:08 not the new strands... :( Nov 25 04:07:11 they work even when there are missing bulbs... Nov 25 04:07:21 BUT you take out the fuse? GOLDEN Nov 25 04:08:09 holycrap90dollarsfor8GBram Nov 25 04:08:15 halfhalo, oil i do displays and populate strands bulb by bulb, and i dont use GFI's, so if it rains, thats my "security system" Nov 25 04:08:18 :D Nov 25 04:08:40 * halfhalo has wire cutters Nov 25 04:09:28 * scoutcamper has a soldering iron, and isnt afraid to use it :D Nov 25 04:09:49 * halfhalo has crowbar Nov 25 04:10:19 ah, you could just wait him out Nov 25 04:10:30 his electric bill will give him a heart attack Nov 25 04:10:36 * scoutcamper had someone cut his lights last year, so he put about 6 mile sof fishing line up as a trip system, you would have had to cut every piece of wire to get through :D Nov 25 04:11:07 * halfhalo has bolt cutters as well as a chainsaw Nov 25 04:11:14 electric chainsaw as well, so quite Nov 25 04:11:27 oil, you would be suprised, if you think about it, lights to music, the lights are off 99.9% of the time, the electric bill is about the same as if i were using the ac Nov 25 04:11:45 * oil would plug his lights into scoutcampers house Nov 25 04:11:59 halfhalo, i have a police station that loves my display Nov 25 04:12:14 so? Nov 25 04:12:17 :p Nov 25 04:12:25 vandal! Nov 25 04:12:39 tell me something I don't know... Nov 25 04:14:24 the police love my display, so i have them station a gaurd 24/7 :D Nov 25 04:14:39 tranq gun Nov 25 04:14:39 * scoutcamper has police arrest halfhalo :D Nov 25 04:14:45 tazer Nov 25 04:15:07 * oil would pay to see halfhalo get tazed Nov 25 04:15:11 * halfhalo is immune from the ravenges of a tazer due to his... incidents with his desktop Nov 25 04:15:18 lol Nov 25 04:15:21 * halfhalo would pay to see that as well actually Nov 25 04:15:51 its settled then Nov 25 04:15:51 i have 220 ac electrical cords with no ends, ie liove 220 wires that can be un-taped asily and appliaed to an assailent :D Nov 25 04:15:55 record it, and sell tickets Nov 25 04:16:08 * halfhalo has a squirt gun. Nov 25 04:16:28 * scoutcamper thinks a squirt gun does nothing to 220Volt AC Nov 25 04:16:28 and/or a hose Nov 25 04:16:37 hose, maybe Nov 25 04:16:51 * scoutcamper also has shotgun :d Nov 25 04:17:18 * halfhalo has a reckless disregard for human life Nov 25 04:17:19 WIN Nov 25 04:17:46 * scoutcamper gives halfhalo the win on that Nov 25 04:18:19 and, i didnt put my lights up till haloween, and finished them today :D Nov 25 04:19:19 hah Nov 25 04:19:25 with 8 people helping every day i worked Nov 25 04:19:25 * halfhalo goes back to his node stuff Nov 25 04:19:44 * scoutcamper goes back to programming his app Nov 25 04:20:08 is anyone here familiar with visual basic? specifically visual basic and com ports? Nov 25 04:20:21 if you need a good laugh http://sendables.jibjab.com/view/OuMxyVbkzA2FM0DX Nov 25 04:20:53 destinal: so if you were to write a plugin, you think Init which returns a pointer to the table for Suspend, Resume, Exit, Poll ? Nov 25 04:21:01 Quad, LOL! Nov 25 04:21:27 I almost fell out fo my chair Nov 25 04:22:01 that needs front page precentral :D Nov 25 04:22:35 lol Nov 25 04:22:59 heehee Nov 25 04:25:06 bpadalino: no, I think PluginTable() returns a pointer to a struct of pointers for callback to Init() Suspend() Resume() Exit() Poll() Nov 25 04:25:54 bpadalino: but we should test out this theory by making our own binary that dynamically loads a plugin and calls its symbol PluginTable and see if we get reasonable pointers for the five Nov 25 04:26:15 that's what you want to do? hrm .. ok .. Nov 25 04:26:31 i was going to suggest trying to write a silly plugin Nov 25 04:26:42 but i guess being able to load it and ask for stuff is good too Nov 25 04:27:04 either way would work Nov 25 04:27:30 i saw there was some work done on the wiki already Nov 25 04:27:46 hidTest.c there Nov 25 04:28:07 yeah, hidTest calls hidd the userland way, we could see if hidd will in turn read from us Nov 25 04:28:31 how do you want to do it ? Nov 25 04:29:10 bpadalino: I'm not that familiar with writing shared libs, I was hoping you might be :) Nov 25 04:29:26 ha, never .. lets figure it out Nov 25 04:30:47 ok lemme grab emulator Nov 25 04:34:53 bpadalino: there's an example of dynamic load on the web somewhere I found that was good Nov 25 04:35:38 bpadalino: http://www.ibm.com/developerworks/library/l-shobj/ Nov 25 04:35:40 easy enough Nov 25 04:36:32 hmm those are linked though, sec Nov 25 04:38:28 bpadalino: ah, there -- http://tldp.org/HOWTO/Program-Library-HOWTO/dl-libraries.html Nov 25 04:38:44 so we need dlopen to load the lib and dlsym to map the symbol Nov 25 04:38:52 you could just link it Nov 25 04:39:27 egaudet: could, hidd doesn't, but we could.. Nov 25 04:39:42 for now dlopen works Nov 25 04:40:07 yeah well hidd has to open multiple plugins, for your case linking is quickest/easiest unless you already coded it Nov 25 04:41:23 true, I haven't yet .. I guess I'll try that Nov 25 04:41:28 hrm .. Nov 25 04:42:10 i used this: http://tldp.org/HOWTO/Program-Library-HOWTO/shared-libraries.html - section 3.4 .. to compile a shared library .. it would be interesting to get it to load and call Init() Nov 25 04:42:46 bpadalino: but Init's not the exported symbol, we need to call PluginTable if I read the libraries right Nov 25 04:42:54 right Nov 25 04:43:06 so get PluginTable figured out so it can call Init() appropriately .. Nov 25 04:46:04 alright well I'm giving up for a couple of days Nov 25 04:47:01 lol Nov 25 04:47:29 okie dokie Nov 25 05:00:28 destinal: so when taking a look at libhidkeypad.so .. Nov 25 05:00:38 it looks like the PluginTable is as follows .. Nov 25 05:01:23 http://pastie.textmate.org/private/nd5zyy6jdn7yry10jb6kvg Nov 25 05:01:35 so GetEventCallBack, Init, Exit, Suspend, Resume and Poll .. Nov 25 05:01:46 let me take a gander at another plugin .. Nov 25 05:01:50 one with more pointers Nov 25 05:02:36 oh nevemrind .. they're all the same Nov 25 05:03:52 ah so I missed GetEventCallback previously Nov 25 05:05:39 so i don't know what each of those return .. Nov 25 05:05:40 if anything .. Nov 25 05:07:40 i have no idea what GetEventCallback does .. but its super tiny Nov 25 05:09:46 return yet more callbacks? not sure why if I was right and we already have 6 of which it is one Nov 25 05:12:07 any idea what __i686.get_pc_thunk.cx is ? Nov 25 05:12:38 http://pastie.textmate.org/private/jmg3iipbnblm9b2mht1gxa Nov 25 05:16:53 "Return values are usually returned in EAX as specified by the ABI, but other values of interest may be returned implic- itly in other registers, such as the pointer implicitly returned in register EBX by the get pc thunk bx function." Nov 25 05:17:39 obscure reference in http://dspace.sunyconnect.suny.edu/bitstream/1951/44615/1/105522902.sbu.pdf Nov 25 05:18:43 bpadalino: more http://zation99.spaces.live.com/blog/cns!FBB24DF09D1B595A!1754.entry Nov 25 05:19:27 yeah i found a definition .. Nov 25 05:19:36 looks like it just puts the current program counter into cx Nov 25 05:19:57 so stuff is returned in eax Nov 25 05:20:11 "x86 doesn't have PC-relative addressing modes so x86 compilers have to use some tricks to copy the value of %eip into a general purpose register,which is necessary for finding the address of the GOT (Global Offset Table)" Nov 25 05:21:54 it looks like it's doing a table lookup .. Nov 25 05:22:05 but i can't figure out the table or what it's passing back Nov 25 05:25:06 bpadalino: the aforementioned global offset table? Nov 25 05:27:04 Data references from position-independent code are usually made indirectly, through global offset tables (GOTs), which store the addresses of all accessed global variables. Nov 25 05:27:17 ^wikipedia position independent code :) Nov 25 05:27:28 hey folks, kind of have a big problem here Nov 25 05:27:44 yes Nov 25 05:27:47 it points to the got! Nov 25 05:28:04 decided to repartition the pre and it crashed in the midst of resize2fs Nov 25 05:28:18 fprimex: heh, doctor? Nov 25 05:28:23 webosdr is refusing to work Nov 25 05:28:23 interesting .. and those addresses end up being: pLsHandle, gHidAvrcpInfo (for the avrcp plugin) and pMainLoop Nov 25 05:28:34 destinal: tired, no dice :/ Nov 25 05:28:36 fprimex: which partition were you resizing? Nov 25 05:28:48 I was resizing var Nov 25 05:29:09 and I had commented media/internal from fstab for the recreation part Nov 25 05:29:16 so... yeah Nov 25 05:29:29 do you need any data off the device? Nov 25 05:29:37 is there a 'repartiton nuke' function? Nov 25 05:29:37 nope Nov 25 05:29:41 wipe away Nov 25 05:29:50 dd some zeros into the LVM partition Nov 25 05:29:59 then doctor Nov 25 05:30:20 how would I go about the dd? quickinstall linux console? Nov 25 05:30:29 rwhitby: yeah, I've done that succesfully before too Nov 25 05:30:38 fprimex: recovery mode memboot Nov 25 05:30:48 fprimex: you need to boot the installer image Nov 25 05:30:52 quickinstall is not a recovery tool Nov 25 05:30:54 destinal: looks like the GetEventCallback returns some Info thinger .. Nov 25 05:31:00 rwhitby: not familiar with that Nov 25 05:31:25 fprimex: http://www.webos-internals.org/wiki/Last_Resort_Emergency_BootLoader_Recovery Nov 25 05:31:31 skip down to "alternative recovery" Nov 25 05:32:09 ok, I will have a look Nov 25 05:32:12 do the first few steps and stop before the Next, mount all disks steps Nov 25 05:32:53 bpadalino: info thinger eh:? Nov 25 05:33:27 info thinger .. Nov 25 05:33:33 size is apparently 0x18 as well .. Nov 25 05:33:37 same as PluginTable .. Nov 25 05:33:54 so maybe it's just a way to return the same structure that is built up ? Nov 25 05:34:58 possible.. hmmm Nov 25 05:35:02 it's in .bss instead of .data Nov 25 05:35:28 so maybe it's assigned after it's loaded or something Nov 25 05:37:29 bpadalino: did you look at export symbols, want to make sure I was reading it right that there's only the one Nov 25 05:37:45 i didn't .. Nov 25 05:37:55 where do i look at the exported symbols ? Nov 25 05:38:23 I was using IDA but I remember there's a way with the standard object tools Nov 25 05:38:30 nm seems to list things .. Nov 25 05:38:32 just not what that way is Nov 25 05:38:33 but ineed to figure out the codes Nov 25 05:39:47 what was the method you thought was exported ? Nov 25 05:41:44 ah yes - PluginTable was the only thing exported it looks like when i do nm -D Nov 25 05:42:32 ok, cool Nov 25 05:43:03 so PluginTable is exported and has a method to return the info which i believe is also the PluginTable .. again ? Nov 25 05:43:17 it does look like it's just those methods .. function pointers .. Nov 25 05:43:22 not sure of the arguments or anything .. Nov 25 05:43:44 yeah, with any luck most of them don't have any but we can disassemble them and figure that out easily enough Nov 25 05:43:56 yeah Nov 25 05:44:20 also what they return Nov 25 05:44:37 the only other thing i know is to compile a shared lib .. Nov 25 05:45:04 $ gcc -fPIC -shared -Wl,-soname,libhidsilly -o libhidsilly.so libhidsilly.c .. assuming libhidsilly is what you were compiling .. Nov 25 05:46:07 want to make one that just does printfs for each with void returns and no params and see what happens? :) Nov 25 05:46:37 obviously we'll fail for the sockets part but we can see if we can get loaded and called Nov 25 05:54:48 bpadalino: perhaps we can just export the struct directly, I don't even really know if it's supposed to be a function or how we tell Nov 25 05:55:01 I suppose seeing if there's code there would do it Nov 25 05:55:04 it's just data .. Nov 25 05:55:06 not a function Nov 25 05:55:12 it's a structure of pointers .. Nov 25 05:55:14 ok, cool, so it's just a struct of the pointers Nov 25 05:55:54 yeah, let me get something going .. maybe you can work on it more .. i have to go to bed soon Nov 25 06:12:22 ok, I have a root prompt through novaterm Nov 25 06:12:49 rwhitby: which device should I dd and does it matter how much? Nov 25 06:12:57 one second Nov 25 06:13:48 rwhitby: p3 and a few megs would probably be overkill? but double check me :) Nov 25 06:14:30 dd if=/dev/zero of=/dev/mmcblk0p3 bs=1024k count=1 Nov 25 06:14:34 * fprimex always does count=1 bs=512 for MBRs Nov 25 06:14:43 k Nov 25 06:14:46 you don't want to overwrite this MBR Nov 25 06:14:54 no sir, I do not Nov 25 06:14:57 :) Nov 25 06:15:05 p3 is the big LVM partition Nov 25 06:15:08 fprimex: yeah p0 has tokens in it that if you don't have backed up, you're not getting back Nov 25 06:15:11 stay far away Nov 25 06:15:17 err p1 Nov 25 06:15:27 gotcha Nov 25 06:15:40 I should back those up at some point, probably Nov 25 06:16:12 hmm - I should make save/restore back up that partition Nov 25 06:16:13 yeah I usually just dd the first 10 megs of /dev/mmcblk0 off to somewhere, just make sure you don't fat finger if/of ;) Nov 25 06:16:52 you can't easily do it now because you'd need to mount a filesystem to store it in usually Nov 25 06:17:04 mmkie dokie, zeros in, zeros out Nov 25 06:17:08 webosdr again? Nov 25 06:17:08 destinal: you can stream it off using novacom and tar Nov 25 06:17:31 * rwhitby goes to add that to the metadoctor backup target Nov 25 06:17:37 fprimex: yep, normal doctor Nov 25 06:17:51 k Nov 25 06:17:55 rwhitby: good point, I rarely use more than get put or tty Nov 25 06:18:46 rwhitby: I say 10m, I think geist once said 8m, but basically nvram area plus bootie for ease Nov 25 06:19:47 if you wanted in save/restore you could just copy off tokens, the rest technically can be rebuilt, we just don't have good docs currently on how Nov 25 06:20:01 I imagine trenchcoat would do the job Nov 25 06:20:24 does trenchcoat read the tokens? Nov 25 06:20:55 it doesn't read in a human readable way afaik, you can use /sbin/tokens to get output. Nov 25 06:21:48 I suspect that the in house version of the tokens binary can edit based on the help text but it looks like the release version has that removed Nov 25 06:21:57 bash-3.2$ novacom -w run file://bin/dd -- if=/dev/mmcblk0p1 > nvram.bin Nov 25 06:21:57 8192+0 records in Nov 25 06:21:57 8192+0 records out Nov 25 06:24:06 nice, are we sure there's nothing before start of p1 that we can't afford to lose? Nov 25 06:24:26 8 megs sounds like exactly what geist said so that makes sense. Nov 25 06:25:18 destinal: I believe trenchcoat will recreate the MBR Nov 25 06:26:16 so, theoretically fdisk, restore p1, amd doctor. even on a totally blank disk that should work with omap3 usb boot Nov 25 06:26:26 s/amd/and/ Nov 25 06:26:28 destinal meant: so, theoretically fdisk, restore p1, and doctor. even on a totally blank disk that should work with omap3 usb boot Nov 25 06:26:49 destinal, why does the s/ stuff work for you and not me? Nov 25 06:26:57 ~botsnack Nov 25 06:26:58 thanks, destinal Nov 25 06:27:04 scoutcamper: because infobot likes me Nov 25 06:27:10 ~lart infobot Nov 25 06:27:10 * infobot chops ka6sox-away in half with a free Solaris 7 CD Nov 25 06:27:11 ~s/work/works Nov 25 06:27:46 it's not so hard to do if you're friends with the bot Nov 25 06:27:50 s/o/0/g Nov 25 06:27:50 destinal meant: it's n0t s0 hard t0 d0 if y0u're friends with the b0t Nov 25 06:27:55 destinal, do i have to have voice or something?(voice is the + sign or something Nov 25 06:28:09 destinal, how do i friend infobot? Nov 25 06:28:20 scoutcamper: it can only be the last line you type and you have to properly terminate your expression Nov 25 06:28:37 s/ion/ion./ Nov 25 06:28:37 destinal meant: scoutcamper: it can only be the last line you type and you have to properly terminate your expression. Nov 25 06:28:42 note the trailing slash Nov 25 06:29:03 s/infobot/infobot1/ Nov 25 06:29:04 scoutcamper meant: destinal, how do i friend infobot1? Nov 25 06:29:05 * ka6sox learned something tonight Nov 25 06:29:10 ooh Nov 25 06:29:15 cool :D Nov 25 06:30:43 ~botsnack Nov 25 06:30:44 scoutcamper: :) Nov 25 06:30:58 ~lart infobot Nov 25 06:30:58 * infobot whacks scoutcamper upside the head Nov 25 06:31:11 ~botsnack Nov 25 06:31:12 :), halfhalo Nov 25 06:31:28 ~oil Nov 25 06:31:28 extra, extra, read all about it, oil is the breaker of wirc Nov 25 06:31:37 stupid...needs fixing Nov 25 06:31:50 lol Nov 25 06:31:55 ~halfhalo Nov 25 06:31:55 i guess halfhalo is a web programmer specializing in ruby on rails and node.js server side applications, and also is dabbling in sysadmin work as well. He is also insane in the membrane, and drives way to fast on the streets. Nov 25 06:32:10 ~infobot no oil is the best! Nov 25 06:32:10 okay, ka6sox Nov 25 06:32:10 pretty easy to tell who set their own response, eh? Nov 25 06:32:39 !halfhalo1 Nov 25 06:32:44 ~halfhalo1 Nov 25 06:32:44 halfhalo1 is, like, a maniac killer Nov 25 06:33:08 * ka6sox is afraid to see what infobot thinks of him. Nov 25 06:33:10 I... possibly Nov 25 06:33:28 ~ka6sox Nov 25 06:33:29 it has been said that ka6sox is Chief Server Infrastructure Architect and Sys-Admin Team Lead(aka BOFH), Work Time:High Power RF, VHDL/Verilog Design, Play Time: Dabble with Kernel development. Nov 25 06:33:30 ~halfhalo is also four-eights-halo. Nov 25 06:33:30 okay, destinal Nov 25 06:33:52 grr missed a t there somewhere Nov 25 06:34:30 or h. or whatever. coffee or sleep needed Nov 25 06:34:43 * ka6sox votes for sleep Nov 25 06:34:51 if i remember correctly, puffys response is funny Nov 25 06:34:53 destinal, both, or mtn dew IV Nov 25 06:35:01 ~PuffTheMagic Nov 25 06:35:01 well, puffthemagic is a dragon, or by the river, or out of puff, or the terminus ENFORCER.. RAAAGGE Puff the magic but don't choke on it. ;) Nov 25 06:35:17 oh my Nov 25 06:35:20 lol Nov 25 06:35:20 morning! Nov 25 06:35:30 morgan Nov 25 06:35:54 ~ut5 Nov 25 06:36:01 ~ugt Nov 25 06:36:01 i heard ugt is Universal Greeting Time. Created in #mipslinux, it is a rule that states that whenever somebody enters an IRC channel it is always morning, and it is always late when the person leaves. The local time of any other people in the channel, including the greeter, is irrelevant. http://www.total-knowledge.com/~ilya/mips/ugt.html Nov 25 06:36:22 oi, still got 'We were unable to reset your phone' :( Nov 25 06:36:40 the funny thing is I think PuffTheMagic used to change nicks a lot before webos-internals and PuffTheMagic is just the one he used when he started here. Since then though every time he's tried to change it everyone gets confused or makes fun of him :) Nov 25 06:36:52 lol Nov 25 06:36:56 yeah Nov 25 06:37:25 now hes stuck Nov 25 06:37:55 unless he goes /nick Nov 25 06:39:13 I occasionally think of switching to my real name except this nick is kind of comfortable Nov 25 06:39:54 lol Nov 25 06:40:02 * oil feels weird using his real nameon the interwebs Nov 25 06:40:26 I've always used mine, since the 1980's. Nov 25 06:40:32 For most things I use my First name on the interwebs, but for most stuff that I don't use that I use halfhalo Nov 25 06:41:07 most stuffs in google related to halfhalo are me Nov 25 06:41:22 my internet name is impossible to search for, lol Nov 25 06:41:38 * halfhalo takes that as a challenge Nov 25 06:41:40 oil, what about your twitter? Nov 25 06:41:47 oil was taken Nov 25 06:41:53 so i wasn't going to signup Nov 25 06:41:58 but swisstomcat cheated Nov 25 06:41:59 oil: lol, like having a nick of "the" or "and" or "google" :) Nov 25 06:42:29 oil, how did he cheat? Nov 25 06:42:38 asked for my email under falze pretenses Nov 25 06:42:50 and the next email i got was "thanks for signing up for twitter!" Nov 25 06:42:52 Those are the best pretenses Nov 25 06:43:16 oil, but now you have one :D Nov 25 06:43:19 yeah Nov 25 06:43:28 and 100 suckers Nov 25 06:43:31 i mean, followers Nov 25 06:43:49 NOT ANYMORE! Nov 25 06:45:00 maybe i should post about what i eat more often they'll go away Nov 25 06:45:17 o\ everyone Nov 25 06:45:28 o\ to you too Nov 25 06:45:32 oil, just say you eat oil :D Nov 25 06:45:35 :P Nov 25 06:45:47 and only post links to oil in saudi arabia Nov 25 06:47:03 oh, now 101, great Nov 25 06:47:10 post about BP propaganda Nov 25 06:47:19 "we're sorry" Nov 25 06:47:22 "sorry..." Nov 25 06:47:32 "so sorry about the birds" Nov 25 06:47:42 "apologising about the fish" Nov 25 06:47:43 * oil stops with the southpark reference nobody got Nov 25 06:47:50 you could go on for days Nov 25 06:47:57 "It was the french! We swear!" Nov 25 06:48:04 oil, just talk about dead fish Nov 25 06:48:08 dead fish 1 Nov 25 06:48:10 dead fish 2 Nov 25 06:48:12 dead fish 3 Nov 25 06:48:13 etc Nov 25 06:48:19 ill post pictures of animals covered in chocholate! Nov 25 06:48:24 :D Nov 25 06:48:30 http://www.youtube.com/watch?v=tCL1svBhVgk Nov 25 06:48:48 oil, to cariboo, oil tastes like chocolate milk, thus we can drill in alaska Nov 25 06:48:52 ~palm Nov 25 06:48:52 palm 1z 0wned by d4 ip4q (ph33r) Nov 25 06:49:07 lmao Nov 25 06:52:51 ~hp Nov 25 06:52:52 i guess hp is The best company known to man. or SkyNet of the future. Nov 25 06:52:59 ~skynet Nov 25 06:52:59 somebody said skynet was HP based on http://www.webos-internals.org/wiki/Application:SkyNet Nov 25 06:53:09 lol Nov 25 06:53:16 * scoutcamper set those :D Nov 25 06:53:56 ~skynet Nov 25 06:54:02 lol Nov 25 06:54:08 :( Nov 25 06:55:28 rwhitby: is it me or is virtualbox a lot better than vmware fusion on OSX? Nov 25 06:55:32 performance wise Nov 25 06:55:59 destinal: dunno Nov 25 06:56:12 rwhitby: do you use both or only vbox? Nov 25 06:56:30 destinal: I use vbox for emulators and vmware for linux Nov 25 06:56:39 same here Nov 25 06:56:56 virtualbox networking kept dropping NFS on me. vmware networking has been rock solid Nov 25 06:57:38 routed or bridged? Nov 25 06:57:46 NAT Nov 25 06:58:00 guest NFS mounting from the host Nov 25 06:59:55 Morning! ;) Nov 25 07:03:45 'Nite! Nov 25 07:04:20 i'm going to be an uncle!!! :D Nov 25 07:05:54 what, right now? Nov 25 07:06:07 If so, pretty sudden! Nov 25 07:06:45 instant uncle? Nov 25 07:07:10 like popcorn? Nov 25 07:07:31 lol Nov 25 07:07:31 http://cargocollective.com/4thamendment Nov 25 07:08:12 if its not cleared up by the next time i fly, im definitely going to order one :) Nov 25 07:08:58 bpadalino: so, I know you're sleeping, but we have a working hidd plugin (with nothing but printf's) but it runs! Nov 25 07:09:01 http://pastie.textmate.org/1324933\ Nov 25 07:09:04 http://pastie.textmate.org/1324933 Nov 25 07:10:47 I suppose linking to the generic init, suspend, resume functions would be the next step and then actually making something readable from it Nov 25 07:14:39 I suspect to actually add additional keybaords we'd have to wrap the current plugin in another plugin that returns its events and our own though by the fact that we don't get polled and names probably have to be unique Nov 25 07:23:14 bpadalino: ok, updated on wiki now that it works. http://www.webos-internals.org/wiki/Hidd Nov 25 07:23:23 * destinal must go pass out now Nov 25 07:23:29 night night Nov 25 07:24:22 oil: nite Nov 25 07:24:56 * oil is about to finish off another box of kleenex ;( Nov 25 07:25:09 oil: you know, you could have said it at any time, we definitely should have done that plugin a year ago :) Nov 25 07:25:15 * destinal-sleep really goes away now Nov 25 07:25:25 what plugin? Nov 25 07:25:42 * oil is lost Nov 25 07:25:42 oil: sample hidd plugin that does nothing but does load and gets called Nov 25 07:26:03 * oil doesn't believe it was his idea :) Nov 25 07:26:24 no, but I mean, we had a wiki article a year ago and this was only a few hours of work Nov 25 07:26:38 oh, lol Nov 25 07:27:41 oh well, definitely exciting at any rate Nov 25 07:28:00 lol, later Nov 25 07:33:37 morning Nov 25 07:44:18 swisstomcat: Morning, did your Navit test on the Pre 2 worked? Nov 25 07:46:17 morning hape Nov 25 07:46:21 it doesn't seem to work Nov 25 07:46:27 it doesn't display the map Nov 25 07:46:31 the rest seems to work Nov 25 07:46:38 morning swisstomcat Nov 25 07:46:54 morning ice8lue Nov 25 07:48:45 did u think about the pre? Nov 25 07:49:00 swisstomcat: What is the value behind R: in the right top line? If you have a good GPS signal it should be 6 with no GPS data at all it will be 0.0000 Nov 25 07:49:11 let me check Nov 25 07:49:35 0.0 Nov 25 07:49:47 R:0,000000m Nov 25 07:49:50 So no GPS signal Nov 25 07:50:07 let me start bfg maps Nov 25 07:50:23 if you go into the menu ans selcet action=> city. Do it find a city and can yopu select show in map Nov 25 07:50:41 I expect this to work. So map is fine. Nov 25 07:50:41 yeah, that worked .. even streets .. so the map data seems to work Nov 25 07:50:48 ok Nov 25 07:51:21 Next one there is a navit.log in the folder where you put the map. Can you pastbin me the contend Nov 25 07:51:36 k Nov 25 07:52:23 http://pastebin.com/S66GvgwT Nov 25 07:57:03 maybe a problem getting the gps data Nov 25 07:57:03 vehicle_webos:vehicle_webos_open:PDL_ServiceCallWithCallback failed Nov 25 07:57:13 jup exactliy that Nov 25 07:57:18 aha Nov 25 07:57:22 which call is that? Nov 25 07:57:34 We had that on some other Pres to but never was able to track down the problem. Nov 25 07:58:00 what call are you using? pdl? Nov 25 07:58:10 I will forword the information to noradtux_ who is doling the C coding Nov 25 07:58:26 allright, he can ask me if he needs help or testing Nov 25 08:00:08 http://git.webos-internals.org/?p=preware/cross-compile.git;a=blob;f=packages/apps/navit/patches/navit__vehicle__webos__vehicle_webos.c Nov 25 08:00:24 I expect that the call in line 120 fails for whatevere resion Nov 25 08:00:38 s/resion/reason/ Nov 25 08:00:38 hape meant: I expect that the call in line 120 fails for whatevere reason Nov 25 08:01:40 i see Nov 25 08:01:50 yeah, that seems like a standard call Nov 25 08:02:02 are there other service calls in the code? Nov 25 08:02:26 or can we get the value of the error? Nov 25 08:02:46 to see if it's a generic service call problem or a problem with just this call Nov 25 08:03:34 We do not have any other service calls. I expect that noradtux has to add some debug code and send you a special version or you compile your own test version Nov 25 08:03:55 i see .. i have no cross-compiling setup tho Nov 25 08:04:02 i only do mojo stuff Nov 25 08:04:25 ok, I send the info to noradtux. I expect he will contact you here Nov 25 08:07:10 swisstomcat: Thanks for the testing. Good to see that it is running on WebOS2 without bigger problems Nov 25 08:09:00 hape: sure Nov 25 08:11:57 Hey guys Nov 25 08:12:33 I have a problem, my preware gets stuck at "completed" with the progress circle spinning Nov 25 08:12:53 I doctored it and it worked fine... until i restored it using save/restore Nov 25 08:13:18 and now its doing the stuck on completed again Nov 25 08:13:57 aww man everyone is asleep! hehe I am gonna go post this on precentral now :) Goodnight! Nov 25 10:51:41 hey Nov 25 11:26:44 Hi, I had problems with my browser yesterday, I've come to realize that not only were my bookmarks deleted somehow, it won't let me add more bookmarks or keep a history of were I've been??? Nov 25 13:44:29 Greetings everyone :) Nov 25 13:50:14 * NuttyBunny looks around and sees everyone asleep, drooling on they keyboards... hope they have some kleenex handy for the mess :D Nov 25 13:55:13 hey NuttyBunny Nov 25 14:01:36 scoutcamper: we found out the quit message issue Nov 25 14:03:57 quit:false? :D Nov 25 14:08:55 ya Nov 25 14:13:03 Cool puffthemagic, I have another wirc issue tho, whenever I close it by swiping up the main card, it auto swipes the othercards, but if I try to run it again it displays the splash andnever runs (times out eventually and disappears)... has this been repo Nov 25 14:13:04 rted? Nov 25 14:13:44 yaa Nov 25 14:13:47 webos bug Nov 25 14:13:53 dont swipe the main card first Nov 25 14:13:58 or use jstop to kill the main app Nov 25 14:14:15 us jstop to kill the left over processes i mean Nov 25 14:16:37 Ok, I will try it, because not even a java services restart work (didn't try Luna restart, only full restart, and that works) :D Nov 25 14:17:41 destinal-sleep: saw your message about the hidd plugin .. Nov 25 14:17:43 neat .. Nov 25 14:18:28 now to just understand how to communicate between those sockets and make something useful .. Nov 25 14:19:12 NuttyBunny: just find the wirc process in jstop and swipe it u will get an option to kill it Nov 25 14:21:23 let's see Nov 25 14:25:42 Excellent!! So this is the reason Palm said in the PDK docs that the cards shouldn't close themselves? Nov 25 14:26:46 Another one!! I have the option for joining my favorite channels (#webos and #webos-internals of course), but it doesn't join automatically Nov 25 14:29:36 no Nov 25 14:29:39 this is just a bug Nov 25 14:29:49 there is an option for that Nov 25 14:30:06 joining channels Nov 25 14:30:59 Ok, no prob, I just tap the menu - freenode - favoites - channel :D Thanks! Nov 25 14:31:18 is your nick registered? Nov 25 14:33:46 yep, and I send the msg to nickserv, I found an option for running commands after connect :) Nov 25 14:34:10 can I use that for joining the channels also? Nov 25 14:38:37 no Nov 25 14:38:50 Damn Telcel :D Nov 25 14:38:57 nightburn: dont use on-connect commands to identify to freeenode Nov 25 14:38:58 remove that line Nov 25 14:39:13 put your ns password as your server password for freenode Nov 25 14:39:22 it will identify u automiacally Nov 25 14:39:23 and Nov 25 14:39:30 it will do it before u join any channels Nov 25 14:39:46 which is important if u have a cloak Nov 25 14:39:49 and dont want people sing your IP Nov 25 14:39:52 seeing Nov 25 14:39:58 ah, great, let me do it :) Nov 25 14:51:51 puffthemagic: It still doesn't join automatically, I've tried using #webos-internals and webos-internals as the channel name and nothing :( Nov 25 14:52:06 It does authorize my nick tho :D Nov 25 14:54:04 are u using the favorite channel list or perform on connect Nov 25 14:54:26 because the fav chan list does not autojoin Nov 25 14:54:31 its like a bookmark i beleive Nov 25 14:55:09 but it would be an easy feature to add Nov 25 15:14:37 puffthemagic: The option in the advanced server prefs says"auto open favorites", so I thought it would join the channels :D Nov 25 15:19:33 Has anyone successfully doctored a pre2 with a Sprint Comm Board? Nov 25 15:21:29 bpadalino: hey Nov 25 15:22:09 morning Nov 25 15:22:27 happy thxgivn. Nov 25 15:23:07 yep, for those of us in the US. Nov 25 15:23:10 ka6sox: morning Nov 25 15:23:25 hiya...need cofffee. Nov 25 15:24:19 Davide-NYC: rwhitby did it at devdays in NYC, though he used a customized doctor Nov 25 15:25:01 destinal: hi Nov 25 15:25:16 bpadalino: so, nice job on the plugin, did you see wiki? Nov 25 15:25:39 yeah i saw that you were able to load it and it called init Nov 25 15:25:57 very cool stuff Nov 25 15:26:09 now to figure out how the command and event sockets work .. Nov 25 15:26:27 bpadalino: well, first we should link libhidd and call the generic init, suspend, resume, I think Nov 25 15:26:33 okie dokie Nov 25 15:26:35 go for it Nov 25 15:26:36 since that's what most of the plugins seem to do Nov 25 15:26:45 but yeah, those sockets... Nov 25 15:27:20 maybe disassemble the Init() of one of those other plugins and see if it opens a socket in there .. Nov 25 15:27:35 maybe the socket name is passed in during init Nov 25 15:27:57 bpadalino: I think the sockets may be between hidd and luna, I'm not sure the plugins use them Nov 25 15:28:05 ah Nov 25 15:28:07 gotcha Nov 25 15:28:21 well, i need to drive for a bit .. i'll be back later tonight .. Nov 25 15:28:33 but yeah let me disasm libhid too to see what the generic funcs do Nov 25 15:28:46 good luck and hopefully you can get something that simulates something cool Nov 25 15:28:47 :) Nov 25 15:28:51 there's so little code in the keypad driver that they seem really simple Nov 25 15:29:02 yeah, i doubt they're doing a lot Nov 25 15:32:35 destinal: what caused my nick switching habbits to come up last night? Nov 25 15:32:59 PuffTheMagic: your infobot name definition had me thinking about it Nov 25 15:33:10 ~PuffTheMagic Nov 25 15:33:10 extra, extra, read all about it, puffthemagic is a dragon, or by the river, or out of puff, or the terminus ENFORCER.. RAAAGGE Puff the magic but don't choke on it. ;) Nov 25 15:34:01 * PuffTheMagic wonders why my name in infobot came up Nov 25 15:34:17 look back up the logs Nov 25 15:34:36 PuffTheMagic: people were playing with infobot, don't remember who now Nov 25 15:34:38 i did that, didnt see a cause Nov 25 15:34:56 ~destinal Nov 25 15:34:57 i guess destinal is just this guy, you know? Nov 25 15:35:19 who the hell sets these things Nov 25 15:35:19 probably because people were checking out different things that infobot knew about different people? Nov 25 15:35:52 i should make a wirc add on app that just queries infobot for people Nov 25 15:36:23 PuffTheMagic: btw terminus apparently broken in 2.0 but I don't think we can test it without a 2.0 device Nov 25 15:36:52 maybe some things missing from the jails? Nov 25 15:36:56 probably Nov 25 15:37:04 how do u know its broken on 2.0 Nov 25 15:37:06 who tried it Nov 25 15:37:19 got a report in #webos yesterday Nov 25 15:37:33 oh Nov 25 15:37:37 [#WEBOS] destinal: terminus crashed phone. I just tried launching it again. Same thing. Nov 25 15:37:39 people are dying for a terminal Nov 25 15:37:39 [#WEBOS] destinal: just tried a 3rd time. This time I got a black screen and solid square cursor in upper left hand corner. But it is unresponsive to any typing. Nov 25 15:37:44 i get a email a week about terminus Nov 25 15:41:43 hello Nov 25 15:42:10 wondering if someone could help me set up networking on an ubuntu rootfs chrooted on pre Nov 25 15:42:20 I set it up following the instructions here: https://wiki.ubuntu.com/ARM/RootfsFromScratch Nov 25 15:42:28 but I can't get the networking set up Nov 25 15:43:19 is /dev bind mounted to your chroot? Nov 25 15:44:32 I was woundering if the sr71 blackbird kernel if I have it set to 500/1200 will it be on 1200 if my screen is off but have a app open like pandora or wIRC? Nov 25 15:45:43 daltonj82, since the governor only cares about he screenstate its @ 500 afaik Nov 25 15:45:55 s/he/the/ Nov 25 15:45:57 ka6sox-away meant: daltonj82, since tthe governor only cares about he screenstate its @ 500 afaik Nov 25 15:46:11 ok thanks Nov 25 15:46:50 PuffTheMagic: I mounted it the same way I mounted the debian rootfs, the way described in the webos internals wiki Nov 25 15:47:17 that didnt answer my question Nov 25 15:47:38 mount --bind /dev /media/cf/dev Nov 25 15:48:02 did u copy over /etc/resolv.conf? Nov 25 15:48:12 no Nov 25 15:48:33 rsanchez, can you ping out of the chroot? Nov 25 15:48:38 an IP address. Nov 25 15:48:38 wont be able to use dns if u dont Nov 25 15:48:50 alright I'll do that Nov 25 15:48:57 couldn't ping anything Nov 25 15:49:11 including an IP address (not a name) Nov 25 15:49:46 yep Nov 25 15:50:51 okay so you have a more fundamental problem than DNS Nov 25 15:51:16 which distro of ubuntu-arm? Nov 25 15:51:22 lucid? later? Nov 25 15:51:43 lucid Nov 25 15:52:47 * ka6sox-away tries to remember if ubuntu-arm uses udev. Nov 25 15:53:52 looks like copying resolv.conf did it, I can ping google's ip address Nov 25 15:54:02 kewl. Nov 25 15:54:12 interesting that it needed to do the lookup. Nov 25 15:54:23 usually its just slow. Nov 25 15:54:29 pinging google.com works too Nov 25 15:54:53 thanks PuffTheMagic, ka6sox Nov 25 15:57:19 dns on the pre is wonky Nov 25 15:57:34 if its not working right the device comes to a hault Nov 25 15:57:43 i have this issue when trying to mess with freetether Nov 25 16:03:55 ugh...requiring DNS to work before *anything* works... Nov 25 16:04:06 DNS will be the DEATH of the internet. Nov 25 16:06:02 when I was doing networking jobs, dns was the first thing i always checked when there was an issue ;) Nov 25 16:37:14 ok, still having trouble since my partition resizing crash Nov 25 16:37:31 webos doctor dies at 82% with unable to restore Nov 25 16:38:36 ouch Nov 25 16:39:17 have you looked at the log of the java as its going by? Nov 25 16:39:26 and seen what its dying on? Nov 25 16:40:16 nope Nov 25 16:40:24 what log am I looking for? Nov 25 16:40:39 what platform are you doctoring from? Nov 25 16:41:39 mac Nov 25 16:42:14 are you using terminal to java -jar webosdoctor* ? Nov 25 16:43:30 nope, double clicking the jar Nov 25 16:43:40 I will try it again from the terminal so I can see the output Nov 25 16:43:43 go to terminal. Nov 25 16:43:48 ya, that will help Nov 25 16:44:38 fprimex: 82 is really far Nov 25 16:44:55 yep. Nov 25 16:45:21 like, I wonder if it would boot, type far Nov 25 16:45:46 yeah, I've had it come on and start going through the motions Nov 25 16:45:56 but there's still some problem with the partitions Nov 25 16:46:09 I've considered rebuilding them myself and then doing a reset on the handset Nov 25 16:46:20 but I have a turkey dinner to get to ;) Nov 25 16:46:31 early Nov 25 16:46:38 (or youhave to drive) Nov 25 16:47:07 6 course italian meal Nov 25 16:47:16 probably going to be there for lunch through dinner Nov 25 16:47:38 what? and you aren't starting @ 9pm? Nov 25 16:48:33 startin' around 1 I think Nov 25 16:48:45 for Italians thats like...early Nov 25 16:49:25 he's a transplant from new jersey Nov 25 16:49:51 I'm from NC, and we're in FL, so I really don't know what a right time would be :) Nov 25 16:50:43 fprimex, I would think that trenchcoat would see that the partitions are hozed and just nuke/pave Nov 25 16:50:50 if that were the issue. Nov 25 16:51:12 and if it got to 82 it did that. I wonder if there could be something wrong with p2 Nov 25 16:51:17 which is /boot Nov 25 16:51:22 I've dd'ed zeros onto the lvm device as suggested by rwhitby and also gone in and lvremoved them when that didn't work Nov 25 16:51:33 errr?! Nov 25 16:51:41 * fprimex just saw 90 and 95% fly by Nov 25 16:51:50 he he... Nov 25 16:51:58 mebbe this time is the ticket? Nov 25 16:52:08 I swear this is about the 5th or 6th time I've run it :| Nov 25 16:52:46 fprimex: did it actually ever bomb out / give terminal failure before or just stall at 80?\ Nov 25 16:52:58 bombed out with unable to restore at 82 Nov 25 16:53:14 we scared it by doing the command line Nov 25 16:54:04 ah, I don't think the % at the terminal and % in the GUI are the same, maybe Nov 25 16:54:13 no, they're not Nov 25 16:54:30 yeah, it's at 64% in the GUI - says "Reconnecting to modem" in terminal Nov 25 16:54:51 # Around 72% the carrier apps are being installed Nov 25 16:54:53 # Around 82% the ROM is being verified Nov 25 16:54:54 from the wiki Nov 25 16:55:43 fprimex: I wondered if rwhitby's dd was too little, I usually dd about 10 megs Nov 25 16:55:52 which is probably overkill but it worked for me Nov 25 16:56:07 ya, hozing the partiton should have been enough Nov 25 16:56:13 I will try that next depending on what we have to go on after this Nov 25 16:56:18 and trenchcoat would cry UNCLE and nuke/pave Nov 25 16:56:42 fprimex: at least this time hopefully you'll see the actual error Nov 25 16:57:26 fprimex: this is an official palm doctor, not meta? Nov 25 16:57:55 destinal: yeah, downloaded via the wiki links Nov 25 16:58:44 when does carrier check happen? Nov 25 16:58:52 INFO: ROM Verifyer: ./usr/share/systemsounds/shuffling_01-ondemand.pcm: FAILED Nov 25 16:59:02 corrupt ROM file? Nov 25 16:59:21 or bad flash Nov 25 16:59:34 WARNING: Flash Failure Nov 25 17:00:10 WARNING: flashing failed, move to failed card Nov 25 17:00:22 hardware failure? Nov 25 17:00:27 fprimex: probably try downloading the doctor again and md5sum both copies just to be sure Nov 25 17:00:38 fprimex: see if your other was corrupted somehow Nov 25 17:00:39 ya, thats what I"d do Nov 25 17:00:50 also try dding 10 megs over p3, what can it hurt Nov 25 17:00:55 k, I will try, but I've also downloaded it about 3 times Nov 25 17:01:01 and check palm's site for the md5sum Nov 25 17:01:08 yeah, will do that too Nov 25 17:01:09 ok Nov 25 17:01:47 ka6sox-away: yeah I was thinking about it and we really need our own tool. doctor is great but a tool with some advanced recovery options would be nice Nov 25 17:02:13 plus one written in easily readable open source python or something we don't have to decompile and reverse :) Nov 25 17:03:14 would need to work natively on all platforms...so python might be good. Nov 25 17:03:26 but you still need novacom drivers Nov 25 17:03:32 is the "md5 fingerprint" in java security cert supposed to be the md5 sum? Nov 25 17:03:43 if so, then the official one I got from palm differs from the earlier download Nov 25 17:03:57 ka6sox-away: well we could integrate novacom drivers within the tool using libusb with pyusb Nov 25 17:04:06 but yeah we'd need to reimplement Nov 25 17:04:59 * ka6sox-away thinks about the windows users and pyusb drivers Nov 25 17:05:13 ka6sox-away: they tell me libusb will work under windows Nov 25 17:05:40 okay thats good then. Nov 25 17:06:19 though a novacom binary would be nice to make in C.. I need to work on my C (and my python now it seems) :) Nov 25 17:06:24 a set of utils that would allow recovery from corrupted flash would help reduce return rates. Nov 25 17:06:43 bpadalino did most of a novacom binary way back when, that could be enhanced Nov 25 17:06:50 still libusb tho Nov 25 17:14:26 wirc + stacks = amazing Nov 25 17:18:26 fprimex, I don't know about java md5 stuff...but I would expect that palm should be able to provide the md5sums of the doctors (if we don't have that one the wiki) Nov 25 17:18:34 s/one/on/ Nov 25 17:18:34 ka6sox-away meant: fprimex, I don't know about java md5 stuff...but I would expect that palm should be able to provide the md5sums of the doctors (if we don't have that on the wiki) Nov 25 17:22:05 destinal: what is the lvm device again? Nov 25 17:22:49 n/m got it Nov 25 17:23:35 fprimex, good because he pinged out Nov 25 17:26:33 ok, dd'ed 10MB to that... try again I guess Nov 25 17:27:54 again I'd use terminal so you can watch whats happening Nov 25 17:28:30 I'm trying on the windows machine now, but I have java set to open the console on app start Nov 25 17:28:41 this is the one downloaded from palm Nov 25 17:49:45 bleh, 10MB to the lvm partition & redownloading from palm still isn't getting it :/ Nov 25 17:51:41 failing @ the same place? Nov 25 17:51:46 complete log: http://pastebin.com/XuAYNfP5 Nov 25 17:51:48 yep Nov 25 17:51:53 flashing failure Nov 25 17:55:03 yep, thats UGLY. Nov 25 17:55:24 hmm :/ Nov 25 17:55:29 phone still boots though Nov 25 17:55:34 pre? pre+? Nov 25 17:55:54 pre+ Nov 25 17:56:15 I would say hardware flash failure Nov 25 17:56:58 I can't remember if bootie can do an fsck. Nov 25 17:58:00 I see that different files are failing the check now Nov 25 17:58:27 guess it's time to see if I can get a replacement Nov 25 17:59:56 looks like the /usr partition is failing Nov 25 18:02:49 well, thanks for everything ka6sox-away; I'm gonna go eat some turkey and play some donkey kong. I'll worry about this later Nov 25 18:15:22 destinal: what type of advanced recovery options Nov 25 18:17:47 PuffTheMagic: things like nuking partitions or rebuilding the fat fs when people get corruption, this is the third person I've seen come in in about a week who needed to boot installer image and dd stuff Nov 25 18:18:05 backup and restore tokens would be nice also Nov 25 18:18:11 or edit Nov 25 18:18:13 we havea token tool Nov 25 18:18:22 egaudet and I made it Nov 25 18:18:24 ;) Nov 25 18:18:37 i need to get my kexec bootloader running this weekend Nov 25 18:18:42 PuffTheMagic: oh really? is it in git? Nov 25 18:18:46 that would help with all the frankenpre shit Nov 25 18:18:54 destinal: ya its called bootie-config Nov 25 18:18:57 nice Nov 25 18:19:36 yeah mainly I'd like to see a lot more advanced capability preferably in a python based gui or something, so you can sa Nov 25 18:19:49 say, back up tokens, change them around, recalibrate the accelerometer, whatever Nov 25 18:19:52 well that aint gonna happen Nov 25 18:19:57 not python based at least Nov 25 18:20:07 PC side I mean Nov 25 18:20:11 pygtk and/or pyqt are a pain to get on Macos Nov 25 18:20:22 it would be fine if it were just a windows/linux tool Nov 25 18:20:31 making people use darwinports to get pygtk would suck Nov 25 18:20:37 hmm Nov 25 18:20:46 Calibre seems to do find on mac Nov 25 18:20:46 i guess we could make a app bundle Nov 25 18:20:49 that is like Nov 25 18:20:51 30mb Nov 25 18:20:53 :) Nov 25 18:20:55 and inlucde it all Nov 25 18:21:24 destinal: py2app works really nice Nov 25 18:21:46 so maybe users wont need to download anything Nov 25 18:21:51 if we make a univeral binary Nov 25 18:21:53 yeah bundle up pyqt into an app bundle like calibre does, should be fine Nov 25 18:22:15 except i would pygtk Nov 25 18:22:25 im afraid of qt Nov 25 18:22:27 * halfhalo would use ruby and titanium Nov 25 18:22:34 but mayve wrapped in python it owuldnt matter Nov 25 18:22:47 heh, with qt being in webos now we should probably get more familiar with it Nov 25 18:23:02 its a language in itself Nov 25 18:23:08 ontop of C++ Nov 25 18:23:14 which is gross in itself Nov 25 18:23:50 sure but does it matter what you use if it's got python bindings? Nov 25 18:23:57 idk Nov 25 18:23:59 it might Nov 25 18:24:16 really depends on who wrote the bindings Nov 25 18:24:17 for example Nov 25 18:24:26 there are like 15 lisp bindings for qt Nov 25 18:24:29 and they are all different Nov 25 18:24:32 some very lispy Nov 25 18:24:37 some basically C like Nov 25 18:24:45 some use smoke Nov 25 18:24:46 some dont Nov 25 18:26:29 PuffTheMagic: i'd like to make a full fledged package manager, too, for the webos ipkg lib. it would be nice to be able to do some of this frankendoctorish stuff without having to redoctor, just say, use version X of this package and version Y of that Nov 25 18:26:59 I was mainly thinking on-pc with the rest of this because then it would work to fix things before ever booting Nov 25 18:27:49 package manager? Nov 25 18:27:51 would be nice if it could make requests to palm's update service using your pre's tokens too so you can choose what version you want from there, download and install Nov 25 18:27:53 in python Nov 25 18:27:54 u mean Nov 25 18:27:55 one stop shopping Nov 25 18:27:57 install emerge? Nov 25 18:28:02 and portage Nov 25 18:28:03 ? Nov 25 18:28:04 no, talking an ipk frontent Nov 25 18:28:08 frontend Nov 25 18:28:19 like preware? Nov 25 18:28:30 * PuffTheMagic is so confused Nov 25 18:28:32 except for the _palm_ library Nov 25 18:28:43 what is the _palm_ library Nov 25 18:28:44 the OS itself Nov 25 18:28:49 not apps Nov 25 18:28:58 idk what u mean Nov 25 18:29:05 there are two ipkg libs Nov 25 18:29:12 one for webos itself, one for apps Nov 25 18:29:26 webos's one is /usr/lib/ipkg Nov 25 18:29:28 IIRC Nov 25 18:29:59 the feed that is used for OTA? Nov 25 18:30:23 yeah that's the primary thing it's used for Nov 25 18:30:48 this way you could do OTA updates without using palm's updater for one Nov 25 18:30:50 i dont see how that and frankendoctors are related Nov 25 18:31:21 nor do I see why you would want to not use their updater Nov 25 18:31:43 PuffTheMagic: there are times when you need more flexibility Nov 25 18:32:10 install a specific version of a specific package, or just update a specific package, and being able to do it from installer image would be handy Nov 25 18:33:05 admittedly the circumstances are fairly rare where would would need it but I've run into a few times where I wish I'd had one Nov 25 18:33:49 s/would/you/ Nov 25 18:33:50 destinal meant: admittedly the circumstances are fairly rare where you would need it but I've run into a few times where I wish I'd had one Nov 25 18:34:57 it would be cool to be able to do the same for both ipkg libs. then you could have functionality similar to wosqi to Nov 25 18:34:59 o Nov 25 18:36:59 basically a framework where you could say, ok, just got a new pre3.14159, i'd like to back up its tokens, swap in package X, install preware and turn on dev mode, usbnet, and skip first use and activation, all before I ever boot it Nov 25 18:37:32 without needing to build special doctors Nov 25 18:38:08 oh and change its carrier customizations, etc Nov 25 18:39:21 destinal: im not really understanding how what u are describing is different than what metadoctor does Nov 25 18:39:38 PuffTheMagic: similar goals, different approach Nov 25 18:39:40 or how it could even be done without metadoctor Nov 25 18:40:01 PuffTheMagic: anything can be done by booting the installer image and talking to the device, that's what the doctor does Nov 25 18:40:21 I'm just talking about eliminating the middleman Nov 25 18:40:25 Well I'm kinda annoyed at palm eurostore Nov 25 18:40:43 destinal: seems like a different middle man Nov 25 18:40:53 write, writing our own then Nov 25 18:40:53 not no middleman Nov 25 18:41:01 s/write/right/ Nov 25 18:41:01 destinal meant: right, writing our own then Nov 25 18:41:18 but one with a lot more functionality along the lines of the things we want to do Nov 25 18:41:30 and not something that necessarily HAS to blow away the device and take forever to run Nov 25 18:41:35 i dont really get how what "we want to do" comes in here Nov 25 18:41:42 but dont let me stop u from doing this Nov 25 18:42:14 I want a tool that *can* restore / reinstall but also just twiddle things Nov 25 18:42:16 fast Nov 25 18:42:32 , It'll NEVER work! :D Nov 25 18:42:36 j/k Nov 25 18:43:29 basic tools like lvm management and dd in the payload would take care of a lot. Nov 25 18:43:54 just ignore me Nov 25 18:44:26 not ignoring you...I see feature creep. Nov 25 18:45:00 not from me Nov 25 18:45:06 i dont even understand the purpose of this yet Nov 25 18:45:17 correct..not from you. Nov 25 18:45:51 yeah, it's me, I see a lot of things this could do but I should probably try to keep the unix philosophy in mind Nov 25 18:46:05 featuritis / bloat tends to be the other way Nov 25 18:46:05 PuffTheMagic, is your kexec going to be able to load a ramfs? Nov 25 18:48:02 of course, i plan on making a minimal uImage with a cpio built in that contains the kexec loader, that people can set to their "main" kernel Nov 25 18:48:15 so the kernel would have to have block support and ramfs Nov 25 18:48:24 if it is able to load anything else Nov 25 18:49:27 PuffTheMagic: what's the advantage over just usb booting? Nov 25 18:49:46 give me some context here? Nov 25 18:49:57 So huh.. I ordered a pre touchstone cover... and they charged me for it Nov 25 18:50:01 sent me an order confirm Nov 25 18:50:03 PuffTheMagic: I mean maybe I don' understand what it's supposed to do Nov 25 18:50:04 and a shipping confirm Nov 25 18:50:08 but it never turned up Nov 25 18:50:11 and they refunded the charge Nov 25 18:50:19 but they won't answer their phone Nov 25 18:50:20 literally Nov 25 18:50:21 /ever/ Nov 25 18:50:27 that's kinda irritating Nov 25 18:50:33 marajin_: who's they? Nov 25 18:50:46 and I can't log in to the site cause it just whines that my login details are wrong Nov 25 18:50:52 and offers me no way to reset them Nov 25 18:50:53 destinal: the point is to have real bootloader, so people can switch between different kernels easily Nov 25 18:50:58 destinal: Palm's official webstore Nov 25 18:51:00 or boot different userlands, etc Nov 25 18:51:09 like grub Nov 25 18:51:12 but not grub Nov 25 18:51:26 PuffTheMagic, so it should be easy to get it to load a ramfs then Nov 25 18:51:28 PuffTheMagic: ah, something that can do a graphical menu, maybe navigate with volume buttons or keyboard? Nov 25 18:51:37 ka6sox-away: yes of course Nov 25 18:51:44 destinal: yes Nov 25 18:51:45 well Nov 25 18:51:47 so that solves a bunch of problems Nov 25 18:51:49 its going to use the fb Nov 25 18:51:51 sure Nov 25 18:51:56 but its going to look more like curses Nov 25 18:51:59 than gui Nov 25 18:52:06 with HIDD or not? Nov 25 18:52:09 oh Nov 25 18:52:11 okay Nov 25 18:52:19 the fso guys are doing a more elaborate one with evas/ecore Nov 25 18:52:27 mine already works Nov 25 18:52:31 theirs is a work in progress Nov 25 18:52:34 lets start with something simple Nov 25 18:52:46 and create a ramfs that has the tools we need Nov 25 18:53:02 so is it a chainloader of sorts? a linux kernel that can load in other kernels and ramfs's and start them Nov 25 18:53:13 destinal: ya Nov 25 18:53:20 kexec chainloaded bootloader Nov 25 18:53:35 ok, cool. that's really awesome Nov 25 18:53:55 * ka6sox-away starts making a list of utils to put in the ramfs. Nov 25 18:54:25 finally could do a rescue imagine on the device if you don't have a pc, boot multiple versions of webos, android, whatever Nov 25 18:54:36 s/imagine/image/ Nov 25 18:54:36 destinal meant: finally could do a rescue image on the device if you don't have a pc, boot multiple versions of webos, android, whatever Nov 25 18:54:42 * destinal can't type today Nov 25 18:54:44 destinal: exactly Nov 25 18:55:13 installing kernels would be easier too since they dont have to muck with symlinks Nov 25 18:55:18 or renaming stuff Nov 25 18:55:31 yeah kernel development where only one can be active is annoying Nov 25 18:56:06 not having to usb boot everything or swap links would be really nice. so you say it actually works? Nov 25 18:56:21 destinal: ya i tired getting this working a month or so ago but it didnt work cause kexec dont work on armv7's Nov 25 18:56:23 ?now Nov 25 18:56:26 even in mainline Nov 25 18:56:28 but someone fixed it Nov 25 18:56:35 and the fix is in the fso kernel git repo Nov 25 18:56:40 so i have to pull that out Nov 25 18:57:10 and build it Nov 25 18:57:13 then we are good to go Nov 25 18:57:14 PuffTheMagic: have you looked at it to know whether it's a trivial backport? Nov 25 18:57:27 there is no backporting to do Nov 25 18:57:31 it was only fixed for .26 Nov 25 18:57:40 oh, ok. nice. Nov 25 18:57:42 i dont think they submited it to lkml for upstream Nov 25 18:57:55 there was a similar fix for a different cpu Nov 25 18:58:01 there was a similar fix for a different cpu Nov 25 18:58:03 and they just copied it for v7's Nov 25 19:03:50 .26 is close enough Nov 25 19:03:56 PuffTheMagic: you know what would be awesome but impractical? pxe-style boot over usbnet or wifi Nov 25 19:04:46 not really much point though esp usbnet since you can already boot usb Nov 25 19:05:35 destinal, TI does that already Nov 25 19:05:59 ka6sox-away: yeah, I've booted the omap3 directly with a tool I found source code for, it's liked on the wiki Nov 25 19:06:16 if you mean the omap usb boot Nov 25 19:06:32 yes Nov 25 19:06:51 did you try it with a pre? Nov 25 19:06:59 ka6sox-away: yup Nov 25 19:07:13 ka6sox-away: http://www.webos-internals.org/wiki/Last_Resort_Emergency_BootLoader_Recovery Nov 25 19:07:23 under OMAP boot Procedure Nov 25 19:07:39 kk... Nov 25 19:08:08 so we could create an image that uses OMAP boot+kexec+rootfs as a recovery tool? Nov 25 19:08:23 s/rootfs/ramfs/ Nov 25 19:08:24 ka6sox-away meant: so we could create an image that uses OMAP boot+kexec+ramfs as a recovery tool? Nov 25 19:10:15 ka6sox-away: well currently I've only tried booting bootie with the omap loader, it's the only boot loader we've got that works so far Nov 25 19:10:25 we haven't really tried porting uboot or whatever Nov 25 19:11:20 i tried omap usb boot also Nov 25 19:11:21 it works Nov 25 19:11:39 you could omap usb boot bootie and usb load kexec / ramfs etc Nov 25 19:12:06 we could just try to port another boot loader but that seems like a lot of work Nov 25 19:12:23 why Nov 25 19:12:30 that is way to much work Nov 25 19:12:34 bootie is fine Nov 25 19:12:39 quick question: I saw someone mention that the Pre2 was using the ondemandtcl governor... is there a way to get ondemandtcl on a 1.4.5 Pre? D: Nov 25 19:12:41 kexec-loader is fine Nov 25 19:13:10 destinal: the kernel init's in like 1 second Nov 25 19:13:18 thats not that long really Nov 25 19:13:25 to get from reboot to the loader Nov 25 19:13:36 evening Nov 25 19:13:58 PuffTheMagic, you are assuming that bootie isn't corrupted Nov 25 19:14:13 well that can be repaired with bootie-config Nov 25 19:14:14 ka6sox-away: you can always get bootie from the doctor Nov 25 19:14:18 and the image from the doctor Nov 25 19:14:33 but we can't distribute a tool with it included unfortunately Nov 25 19:14:33 ive already tested this by wiping all my tokens Nov 25 19:14:42 destinal, I thougth we were trying NOT to have to tear apart the doctor? Nov 25 19:14:56 right but Nov 25 19:15:08 wreaking your tokens really means u hosed your device some how Nov 25 19:15:16 so that is not a every day issue Nov 25 19:15:33 PuffTheMagic, I'm not talking about tokens...just bootie being hozed Nov 25 19:15:41 ka6sox-away: 99% of the recovery questions we see in here still have a working bootie so I'm ok with it for now Nov 25 19:16:05 eventually we may want to try our hand at a bootloader but the investment isn't worth the benefit ATM IMO Nov 25 19:16:05 okay Nov 25 19:16:39 I mean...uboot works with an OMAP Nov 25 19:16:59 so its not that difficult Nov 25 19:17:57 uboot/uboot2 work Nov 25 19:17:58 BUT Nov 25 19:18:04 they dont have fb support for our device Nov 25 19:18:09 so that would mean a lot of work Nov 25 19:18:11 ka6sox-away: maybe once I get my beagleboard I'll play with it a bit, it would be nice if the only option wasn't a binary only blob Nov 25 19:18:13 which i dont want to do Nov 25 19:18:20 i've booted uboot Nov 25 19:18:24 its useless with out fb Nov 25 19:18:49 PuffTheMagic: or USB Nov 25 19:18:50 PuffTheMagic, we should look @ what the beagleboard guys are doing. Nov 25 19:19:55 what they beagleboard guys are doing is sorta meaningless Nov 25 19:20:01 ka6sox-away: I wonder if the omap jtag over usb interface extends to giving some kind of simple serial access between boot loader and host Nov 25 19:20:10 we would need to port the sirloin board stuff their setup Nov 25 19:20:34 destinal, dunno. Nov 25 19:20:38 without a full jtag implmentation preferably :) Nov 25 19:21:21 I need to look @ my beagleboard and see what can be done. Nov 25 19:21:49 PuffTheMagic, yes, the FB stuff would be a pain Nov 25 19:22:15 unless we can get emulator code for a PL2303 in uboot. Nov 25 19:23:16 then we would have usb<>serial Nov 25 19:24:06 ka6sox-away: it's tempting to implement novacom ;) Nov 25 19:24:20 we dont need to distribute bootie to fix it Nov 25 19:24:24 just create a backup image with dd Nov 25 19:24:29 that can be restored Nov 25 19:26:02 does dd work on windows7? Nov 25 19:26:14 in cygwin Nov 25 19:26:28 im sure we can make a dd for windows Nov 25 19:26:39 with mingw toolchain ot something Nov 25 19:26:54 this makes too many assumptions..and relys on proprietary code Nov 25 19:26:55 a special version for usb Nov 25 19:27:14 what are we talkling about that is propreitary? Nov 25 19:27:20 bootie Nov 25 19:27:34 i would love to have a opensource bootloader Nov 25 19:27:50 but i dont think its worth the time Nov 25 19:27:52 not my time Nov 25 19:28:11 bootie works great...for what its intended for. Nov 25 19:28:40 what else would u want it to do Nov 25 19:28:42 but what I would like is a solution that we can distribute, from end to end. Nov 25 19:28:43 that is does not do Nov 25 19:29:04 doesn't allow me to distribute it. Nov 25 19:29:36 what would having a distributable bootloader gain us? Nov 25 19:31:04 a solution that doesn't depend on the flash partitions being intact Nov 25 19:31:34 we have like 10000 projects that have a better return that adding fb+sirloin support to uboot2 Nov 25 19:31:44 but if someone has the knowhow to do so then go for it Nov 25 19:33:12 ka6sox-away: dd is done on the device in a booted installer image btw, that we can automate in the other tool I was thinking of if bootie will start well enough to get into recovery Nov 25 19:34:00 destinal: "if bootie will start well enough to get into recovery" Nov 25 19:34:04 thats why. Nov 25 19:34:20 backup bootie! Nov 25 19:34:35 well actually it always should but you need the palm boot.bin from the doctor unfortunately. Nov 25 19:35:02 boot.bin is just fwloader.bin from uboot i bet Nov 25 19:35:20 we can replace bootie with out having fb Nov 25 19:35:21 PuffTheMagic: nah, boot.bin is bootie and geist said it's not based on uboot Nov 25 19:35:23 thats pretty easy Nov 25 19:35:40 destinal: yet it looks 99.9% like uboot, amazing Nov 25 19:35:56 i've heard the claim too Nov 25 19:36:04 PuffTheMagic: he said he used the commands he was familiar with and he knew uboot so, meh, maybe Nov 25 19:36:23 makes sense to me Nov 25 19:36:38 PuffTheMagic, "backup bootie" is NOT a solution when its already corrupt and you need it. Nov 25 19:37:06 ka6sox-away: doctor image has it so it's good enough for now for the user to provide it Nov 25 19:37:50 eventually I'd like to do our own also Nov 25 19:38:08 if nothing else for the learning experience Nov 25 19:38:29 destinal, okay for now. Nov 25 19:39:05 but tearing a doctor apart is something more advanced...expecially for a windows7 user Nov 25 19:39:29 s/expecially/especially/ Nov 25 19:39:29 ka6sox-away meant: but tearing a doctor apart is something more advanced...especially for a windows7 user Nov 25 19:39:46 ka6sox-away: sure thats not an option, but we can get people to back it up before starting stupid shit Nov 25 19:39:49 not so much, winrar, right click, extract, find a tar, right click, extract. we've had them do it and we could automate it with a tool Nov 25 19:39:52 and that will solve most issues Nov 25 19:40:09 PuffTheMagic: true, just back up the start of flash Nov 25 19:40:16 right Nov 25 19:40:18 thats all im saying Nov 25 19:40:50 whats going to happen when the hose our custom bootloader Nov 25 19:40:58 they are still going to need sometthing to start from Nov 25 19:41:01 just as much work Nov 25 19:41:21 I'd like to eventually have more and more pieces of the stack open or replaceable by OSS components but that's not the first one on my list atm :) Nov 25 19:41:25 PuffTheMagic, its not a question about doing stupid sh*t...its a question of things breaking Nov 25 19:41:36 things only break when u do stupid shit Nov 25 19:41:56 PuffTheMagic, FALSE Nov 25 19:42:13 well we've seen catastrophic flash corruption without stupid things afaik but it was due to damage, write fatigue probably, and we couldn't get past it Nov 25 19:42:21 bootie does not get messed up from doing anything besides dd on your whole flash Nov 25 19:42:23 user: k[t has ban? Nov 25 19:42:52 destinal: u sure they didnt do stupid shit, they could be lying Nov 25 19:43:25 dunno, they could be pulling battery while it's on I guess, we're always warned that that's bad Nov 25 19:43:35 though I do it enough without issues so far Nov 25 19:43:36 i do that all the time Nov 25 19:43:40 it shouldnt kill bootie Nov 25 19:44:11 only time my bootie got messed was when i zeroed out my bootie area to see what would happen Nov 25 19:44:14 and lost my tokens Nov 25 19:44:14 PuffTheMagic: I guess if you lose power to your device while doctoring and it's writing bootie, that's one in a million tho Nov 25 19:44:36 destinal: that falls into the stupid shit category Nov 25 19:45:18 at any rate, it's a closed piece of the stack, so on my list to get to eventually, but for now, hidd :) Nov 25 19:46:26 right, I am not saying it needs to be replaced NOW...it should just be on the radar. Nov 25 19:46:50 ka6sox-away: its def on my radar Nov 25 19:46:57 i've gone down this path a few times now Nov 25 19:47:00 for now, your kext + ramfs Nov 25 19:47:09 its beyond solo skills Nov 25 19:47:15 i would join in again if others helped Nov 25 19:47:18 PuffTheMagic, yep... Nov 25 19:47:31 I need to look @ uboot2 again. Nov 25 19:47:39 uboot2 is nicely organized Nov 25 19:47:45 we just need to add our board to its setup Nov 25 19:47:46 very Nov 25 19:47:57 right, that would need done Nov 25 19:48:39 Greetings :) Nov 25 19:48:42 okay I gotta drive 3hrs...bbl Nov 25 19:48:48 good ideas. Nov 25 19:49:09 so u should all switch to using gentoo Nov 25 19:49:20 my 1 toolchain builds apps and kernels Nov 25 19:49:27 cause it has the right binutils/glibc combo Nov 25 19:49:29 and with that I'm outta here :D Nov 25 19:49:31 no need for multiple chains Nov 25 19:49:32 ;) Nov 25 19:49:37 ka6sox-away: latter happy t-day Nov 25 19:49:51 PuffTheMagic, you too! Nov 25 19:51:44 PuffTheMagic: yeah I think I will try gentoo actually Nov 25 19:51:49 I haven't for years Nov 25 19:52:21 im starting my ovelray over Nov 25 19:52:29 though on OSX I can use the PDK too, need to play with that Nov 25 19:52:33 and going to be pushing sutff later tonight Nov 25 19:52:47 for a cleaner setup Nov 25 19:53:05 glibc 2.6.1 and bintuls 2.19 is magic Nov 25 19:53:22 but 2.6.1 isnt in the gentoo overlay any more Nov 25 19:53:29 i had it pull it out from the depths of old cvs history Nov 25 19:53:52 and i had to add a patch to allow binutils 2.20 to build it Nov 25 19:53:59 stupid autotools fail Nov 25 19:54:08 looked for 1.[3-9] Nov 25 19:54:23 or something like that Nov 25 19:54:33 2.[3-9] Nov 25 19:54:41 so 1.19 didnt match that Nov 25 20:37:03 ok I think the silly hidd example is getting closer to proper Nov 25 20:46:15 http://www.webos-internals.org/wiki/Hidd Nov 25 20:46:21 sample hidd plugin is coming along Nov 25 20:55:54 bpadalino: whenever you get back from celebrations or whatever, check it out.. Nov 25 21:03:29 back now .. Nov 25 21:03:31 looking now . Nov 25 21:05:27 destinal: hrm .. so what is the pointer that is passed in ? Nov 25 21:05:33 during the geteventcallback ? Nov 25 21:20:30 hi rlopin Nov 25 22:07:12 bpadalino: I'm not actually sure, but it does appear to be there from what i can see in IDA Nov 25 22:07:33 interesting Nov 25 22:07:59 I'm not really sure what the rest of the bytes are either Nov 25 22:08:33 have you printed out the values ? Nov 25 22:08:39 do they look like addresses ? Nov 25 22:09:03 negative, I spent a little while getting the generic calls for libhid put in there and they *seem* to work so far Nov 25 22:09:19 so at least it doesn't crash! Nov 25 22:09:21 you can see another init event in the hidd verbose output Nov 25 22:09:30 which I assume is from generic Nov 25 22:11:02 interesting Nov 25 22:13:37 bpadalino: let me try printing the contents of the pointer real quick Nov 25 22:13:42 okie dokie .. Nov 25 22:13:47 how do you know the size of it ? Nov 25 22:14:21 bpadalino: IDA guess. maybe not accurate but at least it doesn't crash :) Nov 25 22:14:27 heh Nov 25 22:14:57 i see gPluginPrvInfo in the libhidinputdev.so being 54 bytes .. Nov 25 22:15:23 and for libhidavrcp.so i see gHidAvrcpInfo being 0x8b4 bytes .. Nov 25 22:15:27 0x54 bytes rather before Nov 25 22:15:51 so 0x54 correlates to 21 32-bit integers .. Nov 25 22:16:03 ah Nov 25 22:16:22 but i have no clue where that comes from :( Nov 25 22:38:34 bpadalino: so sizeof(a1) with the callback returns 4, which would make sense if it was index 0 of 22 dwords in the IDA guess -- can that be trusted? Nov 25 22:39:53 seems to be reliably pointing at the same data across invocations also Nov 25 22:39:58 it's just a pointer .. Nov 25 22:40:07 so you'll get a sizeof being 4 no matter what Nov 25 22:40:42 print out the contents going up to 21 ... 32 bits at a time .. Nov 25 22:40:49 i am curious if there are things that may look like sizes .. Nov 25 22:40:52 or just addresses in there Nov 25 22:41:45 howdy folks! Nov 25 22:42:38 Anyone know anything about what to do if the palm:// service stops running? i.e. how to fix it Nov 25 22:43:25 panic .. Nov 25 22:43:27 ajay: run. Nov 25 22:43:39 destinal: which direction? Nov 25 22:43:41 and watch for the zombies Nov 25 22:43:48 uh huh Nov 25 22:43:50 morning Nov 25 22:43:56 morning rwhitby Nov 25 22:43:57 ajay: palm:// isn't a service Nov 25 22:43:59 mornin! Nov 25 22:44:01 palm:// is not a single service Nov 25 22:44:26 there are a number of executables which contribute services to the palm:// bus Nov 25 22:45:57 destinal: get that printout of those values ? Nov 25 22:45:58 rwhitby: Is there documentation online about this? I've been having trouble finding any Nov 25 22:46:36 I don't know why all/most of the associated executables with palm:// (And today pulse) stopped running.. Nov 25 22:47:06 maybe they're taking thanksgiving off .. Nov 25 22:47:48 bpadalino: palm:// stopped working wks ago actually Nov 25 22:47:59 bpadalino: what's the best printf to get hex bytes for them? Nov 25 22:48:03 plus I don't even celebrate thanksgiving Nov 25 22:48:18 %8.8x Nov 25 22:48:37 ajay: well i'm not jewish, but i can understand why some people are walking around on saturday! Nov 25 22:49:19 bpadalino: I think that's a stretch Nov 25 22:49:47 bpadalino: so.. Nov 25 22:49:48 ajay: there is not - it's an internal thing which Palm doesn't document. Nov 25 22:49:50 int GetEventCallBack(int *a1) { Nov 25 22:49:52 very much so Nov 25 22:49:52 for (i=0; i<22; i++) printf("%8.8x", a1[i]); Nov 25 22:49:59 sure Nov 25 22:49:59 or am I wrong? Nov 25 22:50:01 2.0.1 doctor is out: http://palm.cdnetworks.net/rom/pre2/p201r0d11242010/wrep201rod/webosdoctorp102ueuna-wr.jar Nov 25 22:50:08 may want a \n in there Nov 25 22:50:09 for good measure Nov 25 22:51:15 rwhitby: Okay. do you know which binaries are involved? I'd like to get them runnin again Nov 25 22:51:23 bpadalino: http://webos.pastebin.com/mMHUKPDU Nov 25 22:52:17 ajay: there are at least 20 binaries involved Nov 25 22:52:28 look in /usr/share/dbus-1/system-services/ Nov 25 22:53:08 destinal: yeah doesn't make much sense to me .. :( Nov 25 22:53:44 Hmm.. I wonder if I'm looking at the right thing. Nov 25 22:53:57 My problem is I can't click links from emails, phone numbers from the web, and a few other quirks. Nov 25 22:54:08 bpadalino: I rebooted and the result was identical so at least there's that Nov 25 22:54:20 At some point a program complained about palm:// but maybe that's just a piece of what's borked.. Nov 25 22:55:21 well being the same is good .. Nov 25 22:56:26 for being linux, this phone really gets easily (And confusingly/non-auto-fixingly) screwed up when you fill up its partitions Nov 25 22:57:28 ajay: I've had some real production linux systems seriously lose it when / became full Nov 25 22:58:00 destinal: even post-space-freeing and reboot? Nov 25 22:58:15 eh, true. usually not Nov 25 22:58:48 destinal: wish i could help out with this private data stuff .. Nov 25 22:58:51 ajay: did something happen to the database files? Nov 25 22:59:13 are you sure returning 0 is right for GetEventCallBack ? Nov 25 22:59:44 bpadalino: the example I disasm'd did Nov 25 22:59:59 destinal: A bunch of DBs were acting screwy for a minute, but (iirc) all my data came back/apps were acting normal again (post space-freeing) Nov 25 23:00:14 so what's the general opinion of 2.0.x ? Nov 25 23:00:27 Marajin: most of us wish we had it Nov 25 23:00:33 heh Nov 25 23:00:41 no release outside the US or something? Nov 25 23:00:44 or Pre 2 only or what? Nov 25 23:00:45 haha.. i was about to say that, but sure that's not the answer marajin was lookin for Nov 25 23:01:03 Marajin: pre2 only so far Nov 25 23:01:34 bpadalino: you could help out, grab an emulator :) Nov 25 23:01:42 true Nov 25 23:01:49 i have some people over .. Nov 25 23:01:52 i will bust it out soon enough i guess Nov 25 23:01:54 oh well not now Nov 25 23:02:11 sometime when you're not celebrating :) Nov 25 23:02:18 meh Nov 25 23:02:38 speaking of 2.0, was there nothing (other than QuickOffice) folks were able to extract from 2.0? (I'm particularly curious about flash) Nov 25 23:02:49 bpadalino: we're getting ready to go up to my mom's in a few Nov 25 23:03:05 ajay: I tried to get flash to work but it's complicated :) Nov 25 23:03:06 is the pre 2 actually significantly better? Nov 25 23:03:10 she's hacking up some webos hidd plugins too ? Nov 25 23:03:11 nice Nov 25 23:03:17 slightly better camera and twice the storage is all I've heard about it Nov 25 23:04:24 1GHz processor slipped by ? Nov 25 23:04:30 3630 versus 3430 ? Nov 25 23:05:55 bpadalino: anyway I have a feeling we'll get something useful working soon enough which is more than I could have said about hidd before Nov 25 23:06:04 meh Nov 25 23:06:18 okie dokie Nov 25 23:07:05 I'd have liked higher screen resolution over the CPU upgrade :P Nov 25 23:07:06 but ok Nov 25 23:07:08 maybe this weekend we'll get some scriptable input events happening Nov 25 23:07:13 bpadalino: I'm thinking about writing a wrapper for an existing plugin that proxies for it so we can trace more than hidd wants to give us Nov 25 23:07:19 ajay: use the mime table reset utility Nov 25 23:07:21 sounds good Nov 25 23:07:23 hmm.. pulse's issue is trouble creating a directory.. maybe some perms got changed somewhere Nov 25 23:08:08 bpadalino: also it's a good smoke test, if we're screwing anything up it should make the existing plugin fail, when it works reliably I guess we have the interface more or less correct Nov 25 23:08:11 rwhitby: Is that diff from the files checkin util? Nov 25 23:08:15 because I Ran that alread Nov 25 23:08:26 yeah i believe so Nov 25 23:09:30 bpadalino: I have a deadline on this stuff now, I have a beagleboard on the way so once I get a kernel and luna working on it I'll need custom hidd plugins to get any useful input into it :) Nov 25 23:09:32 destinal: did you sing to flash? "Why ya gotta go and make things so complicated!" maybe extracting the browser from 2.0 as well would help? Nov 25 23:09:33 that's not too shabby of a plan anyway Nov 25 23:10:37 ajay: lol, yeah. but it hated the song and the 2.0 browser hung. well the browser hung anyway but I didn't actually sing. Nov 25 23:11:34 destinal: oh yeah you're on it.. :) Nov 25 23:12:23 I tried just the plugin and it didn't render anything, I tried changing config settings in etc and it didn't do anything. I figure there may be db settings needed to turn on flash and they're not there, so I pulled over 2.0 browser but needless to say the infrastructure it expects is much different so it fails Nov 25 23:12:45 huh. intense Nov 25 23:13:04 I could play with it a lot more but I figure my time's better spent getting 2.0 working :) (well and hidd which is important too for other reasons) Nov 25 23:13:12 hidd? Nov 25 23:13:35 hidd is the layer between the physical hardware / kernel modules and luna (the GUI) Nov 25 23:13:51 it's also modular and uses a plugin for each type of device Nov 25 23:14:03 accelerometer, keypad, touchscreen, proximity sensor, etc etc Nov 25 23:16:03 ajay: basically if we want to add custom input devices (physical or virtual) hidd's the place to do it, but it's palm custom proprietary stuff and we don't have documentation or source code from palm, so we have to reverse engineer Nov 25 23:16:25 huh.. wonder if playing there would help the n900 folks play games better.. ;^) Nov 25 23:16:43 destinal: oh! very cool Nov 25 23:17:00 nah SDL also talks to hidd afaik but it presents a standard SDL interface so they have that all abstracted for them Nov 25 23:18:01 destinal: I was thinking for maybe the gesture area (assuming accelorometer/etc work the same) Nov 25 23:19:33 I think they just get events. they've probably had some reverse engineering of there own or studying palm's patches to learn about the custom stuff palm's doing but I think the guy who started the webos games on n900 work is also the guy who did DrNokSnes, so I think it's in pretty capable hands :) Nov 25 23:21:04 really what they need most is fake multitouch. beyond that they could use a way to legally buy webos games without buying pre's of their own but I think the catalog is pretty set on you having a palm profile with a device SN and possibly MEID Nov 25 23:21:56 speaking of I still want n900 skype.. ;^) Nov 25 23:22:20 ajay: I guess we're going to have skype soon anyway so we don't need to fake hildon which could be a huge pain Nov 25 23:23:18 destinal: I think we already had this discussion before, I thought I convinced you that there is a need.. ;^) Nov 25 23:23:31 is the omaplfb module new to 2.0 ? Nov 25 23:24:00 yeah, I like the idea but I don't know much about libx11 or the hildon extensions to it. we may still want it for other apps later but skype is coming with verizon pre2's apparently Nov 25 23:25:09 that's the kicker - it's Verizon Data-only. no wifi, no worldwide Nov 25 23:25:31 ajay: and you really think that problem can't be solved? Nov 25 23:25:53 destinal: are you challenging yourself to solve it? :) Nov 25 23:27:16 ajay: *shrug* - it would be an interesting challenge but i've got several of those more important ATM (IMO) Nov 25 23:27:48 I have a feeling skype will be pretty easy if people want it but I haven't seen it or anything Nov 25 23:29:07 i guess my point was more that there's no reason to poo-poo someone who wants to see it on the device via an alternate method. I agree -- spend your time on the more important projects Nov 25 23:30:09 ajay: oh, yeah, it would be cool, but I bet getting the webos skype working without restriction is way easier than the hildon and x layers, that's all Nov 25 23:30:26 if someone wants to do it it would be awesome for all kinds of reasons Nov 25 23:30:38 I suspect skype will go through a dedicated verizon number, not over wifi. Nov 25 23:31:06 that's one way they could be guaranteed that people can only use it on verizon Nov 25 23:31:18 rwhitby: Oh yeah I think I read that's how it works, it takes minutes, not over data Nov 25 23:31:20 true Nov 25 23:31:21 and there is enough money involved that they will have smart people enforcing that guarantee Nov 25 23:31:55 so I suspect the best you'll get on another carrier is the IM and presence information, but not calls. Nov 25 23:32:13 in that case maybe the n900 binary is the way (though straight SIP would be cool tbh, has anyone come up with a SIP softphone for webos yet? oh right probably because no catalog for hybrid yet) Nov 25 23:34:07 yeah I remember someone had ported a SIP client, but the desc made it sound a little unworkable. that'd be cool.. Nov 25 23:34:46 running asterisk on pre with pulse channels and IAX to a home PBX would be fun too Nov 25 23:35:13 or SIP / IAX to a telco etc Nov 25 23:35:20 (speaking of skype.. damned if I can login to ribbit anymore, grr..) Nov 25 23:36:19 destinal. want me tocreate a page 4 precovery? Nov 25 23:36:19 I wonder if anyone ever tried to get android apps working outside of android.. Nov 25 23:36:37 mm.. asterisk Nov 25 23:39:53 ka6sox-away: eh? Nov 25 23:41:06 ooo permissions for /var were wrong, the plot thickens Nov 25 23:46:31 hrmm Nov 25 23:48:46 running ps I can't see any processes that should be running that aren't. :( Nov 25 23:48:59 :( Nov 26 00:09:53 ajay: re android I think people are trying to set up dalvik based frameworks for ubuntu probably for use on tablets, don't know their status yet Nov 26 00:10:48 since we're talking about android now, is /dev/android_adb_enable generally only readable by root ? Nov 26 00:11:08 destinal: Interesting.. I was just thinking about how palm's removing the jvm from WebOS2, am wondering if that would make it harder to run android apps on the pre.. ;^) Nov 26 00:12:13 ajay: not harder, already really hard, :) palm's jvm was only really intended for service side code anyway. Nov 26 00:13:05 ajay: probably far easier to run android on pre than to run android apps in luna Nov 26 00:13:52 From lookin at the WebOS 2 Dr. does it seem like there's a lot more space taken up by palm stuff? (wondering if removing Classic, Java mighta been related to that) Nov 26 00:14:40 beagleboard Cx runs android and is already really close in specs to pre Nov 26 00:15:08 destinal: what's the plan moving forward with hidd ? Nov 26 00:15:35 destinal: I don't think specs matter at all for Andrizzle, since its Java and I reckon everyones' hardware has different specs Nov 26 00:15:49 kernel modules matter :) Nov 26 00:16:13 The bigger problem is that dalvik stuff you mentioned.. yeah maybe some kernel stuff (hope not that sounds like a PitA! :) Nov 26 00:17:24 bpadalino: the next thing to do I think would be to figure out how the keypad driver stuffs keys in and let's try making our own Nov 26 00:17:32 I'm sure the google would not be happy ! Nov 26 00:17:33 bpadalino: wrapper may help with that or may not be needed Nov 26 00:17:52 ajay: google is happy for everyone to run android on everything, that's their goal Nov 26 00:18:19 destinal: Oh, I'm sure running AndroidOS on the Pre would not be a problem, running Android Apps on WebOS, there'd be the kicker.. ;^) Nov 26 00:18:29 they probably would be happy with that too Nov 26 00:18:57 eh, not worth it Nov 26 00:19:00 bpadalino: we could then try the test harness to read key input from it or go for broke and see whether luna picks em up Nov 26 00:19:01 World domination doesn't come from having competing platforms Nov 26 00:19:39 bpadalino: just have it stuff in an "a" every X polls or whatever Nov 26 00:19:46 * linuxjacques is not sure wold domination is a good thing. Nov 26 00:19:57 * ajay never said it was :) Nov 26 00:19:57 how does the plugin tell hidd it has something to push up ? Nov 26 00:19:57 s/ol/orl/ Nov 26 00:20:23 is there a function pointer in that original set of data that we need to use ? Nov 26 00:22:56 then we can write a kinect plugin for hidd Nov 26 00:23:09 ooo Nov 26 00:23:12 bpadalino: bpadalino hmm, it starts up a reader thread when init is called Nov 26 00:23:19 HidPluginGenericInit is pretty intense .. Nov 26 00:23:20 and the SCALE demo can be webos on a projector controlled by kinect Nov 26 00:23:21 yeah i see that Nov 26 00:23:33 but what does that thread end up doing, know what i mean ? Nov 26 00:23:38 what is the execution like Nov 26 00:27:01 bpadalino: HidReadEventSource(spEventSource... seems to be important Nov 26 00:27:19 then a hash table lookup Nov 26 00:27:37 ah ha! Nov 26 00:27:39 pReportFn Nov 26 00:27:50 that's what a1 is .. that gets passed in Nov 26 00:27:56 so GetEventCallBack .. Nov 26 00:28:05 is a pointer to a function that you call when you report something Nov 26 00:28:13 aha Nov 26 00:28:16 right Nov 26 00:29:02 always the same because the same function code was there :) Nov 26 00:29:49 bpadalino: I'll be back later this evening, have to go do the family thing Nov 26 00:29:53 enjoy Nov 26 00:40:21 rwhitby: Where's that MIME recovery tool? Is that the name? I don't see it under Linux apps Nov 26 00:40:47 testing feed Nov 26 00:41:21 I have a webosinternals/test... Nov 26 00:41:34 s/si/s-i Nov 26 00:42:41 hmm, I'm trying to look at the full URL of the feed but seems like that can't be edited Nov 26 00:42:48 or viewed Nov 26 00:43:36 yeah, someone with mojo skills needs to fix that :) Nov 26 00:44:29 well, I definitely have a testing feed, I imagine its the right one but can't check.. ;^) Nov 26 00:47:27 i'll add another, should I add all or armv7 ? Nov 26 00:48:40 got it Nov 26 00:51:10 * ajay wonders if any would want to port xflux to the pre.. Nov 26 00:54:12 rwhitby: didn't help.. :( Nov 26 01:01:39 ajay: with the permissions stuff you were reporting, I think that device is asking for a full erase and doctor Nov 26 01:01:46 eh, maybe a doctorin it'll be.. (boy that was such a pain last time!) hm, I wonder if I should use the bigger /var doctor Nov 26 01:02:09 ajay: there's also a metadoctor feature to add a completely separate ext3 partition Nov 26 01:02:28 ooo.. me likey ext3! Nov 26 01:03:13 and you can specify the size Nov 26 01:03:18 I never expected to get this intimate with a phone in my life! my.. Nov 26 01:03:43 options,options.. (which makes it hard to make "decisions, decisions") Nov 26 01:04:05 X, home, and maybe some apps, wonder what else extra/bigger partitions could be used for Nov 26 01:04:18 is dtz here.. Nov 26 01:04:46 not here.. was wondering how to best repartition Nov 26 01:04:52 (size-wise) Nov 26 01:06:25 well gonna food, ttfn! Nov 26 01:21:39 wirc ftw :) Nov 26 01:22:39 bpadalino: if you have any more insights, let me know Nov 26 01:22:50 Happy Thanksgiving :) Nov 26 01:23:13 destinal-pre: no insights as of now .. Nov 26 01:24:14 NuttyBunny: happy thanksgiving to you as well Nov 26 01:24:21 destinal-pre: there are some constants i don't quite understand from the decompilation .. Nov 26 01:24:54 And even happier black friday shopping spree ;) hehehe Nov 26 01:26:07 heh, christmas shopping to do but I already got my toy for myself (beagleboard Xm) Nov 26 01:27:20 I want my Pre 2 but Palm hasn't given my coupon yet :'( Nov 26 01:27:32 ups ground time is bound to seem like forever Nov 26 01:28:22 NuttyBunny: haven't heard of anyone getting em yet Nov 26 01:28:24 NuttyBunny: I haven't heard of anyone getting a coupon yet Nov 26 01:29:08 No? bah... the worse thing is that I never got my free Pre Plus either :( Nov 26 01:45:20 NuttyBunny: what kind of stuff are you hacking on? Nov 26 01:46:20 * halfhalo has been getting node stuff working with the shipping node version and has had success with a lot of packages Nov 26 01:47:03 * rwhitby worked out how to stop an old device being wiped when you log into a new device with the same profile. Nov 26 01:47:16 nice Nov 26 01:47:41 halfhalo: hey there's an idea, that would work in the new 20 frankendoctor that lacks luna Nov 26 01:48:10 destinal-pre Right now, not much, as my Pre is my only phone and I can't play with it too much, I'd like to play and help with the frankendoc also Nov 26 01:48:25 I have yet to test them on device, but theoretically they should work (assuming I pacakged em up right) Nov 26 01:50:14 NuttyBunny: I recommend a $150 beagleboard (well once we get it stabilized as a dev platform) Nov 26 01:50:34 I have a webserver and ftp server supposedly working Nov 26 01:51:58 halfhalo: cool. node.js is going to be huge Nov 26 01:52:10 yup! Nov 26 01:53:17 I have to stop procrastinating and do something, I still haven't even updated my single app on the catalog :( Nov 26 01:53:27 NuttyBunny: going to use pre2 as a dev handset or comm board swap and use pre as that (or are you on GSM now) Nov 26 01:54:10 I'm on GSM, so, my Pre 2 will be my main dev, and my Pre - will be my dev Pre :) Nov 26 01:54:25 Pre 2 : Main Device, Pre - Development device Nov 26 01:54:30 dev, dev.. hehehe Nov 26 01:57:17 * halfhalo goes to nom on donuts Nov 26 01:59:27 * rwhitby is considering moving to Pre 2 as main device this weekend Nov 26 02:00:01 how do you like the build quality of the pre2, rwhitby ? Nov 26 02:00:09 is it like a tank Nov 26 02:00:10 h Nov 26 02:00:15 has you had have dropped on front of it yet Nov 26 02:00:26 bpadalino: I've never had a problem with build quality on any of my webOS devices. Nov 26 02:00:48 interesting .. i have a crack developing in my sprint pre :( Nov 26 02:01:23 mine's cracked too, near the charge port Nov 26 02:01:56 the only crack I have on any device is the one which was already there on an old sprint Pre that someone sent me Nov 26 02:02:52 I think the usb crack is due to pulling too hard the little cover, or plugging the cable over the cover (I always pull it a little so it doesn't get caught) :) Almost 12 months and no issues Nov 26 02:03:40 I never had a major issue with either of my pre's either Nov 26 02:03:55 Although the usb cover came off the sprint one Nov 26 02:04:09 I lost a usb charge cover on my first pre and it had dead pixels out of the box Nov 26 02:05:23 I only have the yellow blotches whenever it gets hot, but I think all Pres have those :D Nov 26 02:06:23 mine doesn't Nov 26 02:06:31 * halfhalo has bacon Nov 26 02:11:17 bacon on the pre? Nov 26 02:12:45 he cooks bacon on his Pre Nov 26 02:12:52 on the mirror :D Nov 26 02:13:15 mmmm .... bacon Nov 26 02:16:28 the USB cover coming off the pre isn't a problem Nov 26 02:16:30 it's an improvement Nov 26 02:17:23 * halfhalo finished his bacon and preceded straight to popcorn Nov 26 02:23:12 * NuttyBunny begins cooking a delicious quesadilla with his Pre overclocked at 1.2Ghz ;) Nov 26 02:25:03 the burning cmos adds a unique smokiness Nov 26 02:25:19 yup! Nov 26 02:26:05 hehehe... I remember burning an AMD processor once... sickening smoke! :S Nov 26 02:27:17 Well guys, gotta go, there's a store here with a nice promotion, electronics (TVs and Car Stereos), 48 months, fixed interest, 15% discount, and I get to pay starting on feb :D Nov 26 02:37:11 hey all I got a question about kernels. first why can only some pres run the sr71 blackbird kernel? and secondly what's ur inputs on these kernals what's the best one to run? Nov 26 02:42:51 destinal-pre: i think the report function looks something like this: // void *reportFunction( char *command, 1, 0, &(PluginTable.Init) ) Nov 26 02:42:59 not sure what the 1 and 0 really do .. Nov 26 02:43:12 but i think the bulk of the stuff happens in the char * first argument .. Nov 26 02:46:05 ah ha, i think i get it .. Nov 26 02:46:27 the first 1 there should be command number maybe .. so they can get a count of the number of events happening and have them happen in sequence? Nov 26 02:46:57 i think the char * is a null terminated string .. Nov 26 02:51:16 destinal-away: HidReadEventSource populates the char* that gets passed into the reportFunction **** ENDING LOGGING AT Fri Nov 26 02:59:58 2010