IRC log for #htc-linux on 20130111

00:31.29*** join/#htc-linux stroughtonsmith (~steven@86-42-152-116-dynamic.b-ras1.bbh.dublin.eircom.net)
00:44.19*** join/#htc-linux fakker (fakker@unaffiliated/fakker)
01:25.24*** join/#htc-linux fakker (fakker@cpc30-hitc6-2-0-cust607.9-2.cable.virginmedia.com)
01:25.24*** join/#htc-linux fakker (fakker@unaffiliated/fakker)
01:37.01*** join/#htc-linux ElFinLazz (~elfinlazz@182.215.84.22)
02:10.02*** join/#htc-linux Cotulla (~myfakemai@109.205.253.11)
02:31.27*** join/#htc-linux ccxCZ (~ccxCZ@156.200.broadband11.iol.cz)
06:45.08*** join/#htc-linux kiozen (~kiozen@p578a42db.dip0.t-ipconnect.de)
07:34.30*** join/#htc-linux Bry8Star (~Bry8Star@gateway/tor-sasl/bry8star)
07:49.24*** join/#htc-linux lamikr (lamikr@nat/nokia/x-dgmjkkoslunwxgwj)
07:49.33*** join/#htc-linux eR^zeRa` (~zzeratul@ip-84-42-202-42.net.upcbroadband.cz)
08:05.05*** join/#htc-linux ntg (christer@141.0.137.20)
11:30.30*** join/#htc-linux a-st (~a-st@unaffiliated/a-st)
11:51.41*** join/#htc-linux arif-ali (~arif-ali@81.144.163.60)
11:56.46*** join/#htc-linux arif-ali (~arif-ali@81.144.163.60)
12:02.05*** join/#htc-linux arif-ali (~arif-ali@81.144.163.60)
12:07.33*** join/#htc-linux Cotulla (~myfakemai@109.205.253.11)
12:08.21*** join/#htc-linux kiozen (~kiozen@p578a42db.dip0.t-ipconnect.de)
12:22.39*** join/#htc-linux arif-ali (~arif-ali@81.144.163.60)
12:39.22*** join/#htc-linux Marc (~marc1706@phpbb/modifications/marc1706)
12:46.27*** join/#htc-linux ntg-c (~christer@anon-168-72.vpn.ipredator.se)
13:11.07*** join/#htc-linux helicopter88 (~helicopte@host130-27-dynamic.49-79-r.retail.telecomitalia.it)
13:21.26fakkerhttps://twitter.com/rkkski/status/288956938778189824/photo/1 Cotulla
13:38.26*** join/#htc-linux ntg (christer@141.0.137.20)
13:55.48*** join/#htc-linux lamikr (lamikr@nat/nokia/x-huomnowbotmyyrka)
15:15.11*** join/#htc-linux NEMESIS| (nembnc@bash0rt.euch.kacknubs.de)
15:27.07styblafakker: buahahahahahaha :D
15:28.35styblafakker: I've even managed to translate the tweet. the last word is equal to 'insert', right?
15:37.26*** join/#htc-linux ntg-c (christer@141.0.137.20)
16:21.30*** join/#htc-linux helicopter88_2 (~helicopte@host68-94-dynamic.22-79-r.retail.telecomitalia.it)
16:43.39*** join/#htc-linux arif-ali (~arif-ali@81.144.163.60)
17:20.50*** join/#htc-linux ali1234 (~ajbuxton@headfuzz.co.uk)
17:30.47detuleqcom really has a botched v4l2 implementations don't they
17:45.49*** join/#htc-linux ALoGeNo (~alogeno@unaffiliated/alogeno)
17:50.38*** join/#htc-linux paulk-desktop (~paulk@lib33-1-82-233-88-171.fbx.proxad.net)
17:54.59Cotullawhyy
17:55.33*** join/#htc-linux kiozen (~kiozen@ppp-93-104-81-177.dynamic.mnet-online.de)
17:58.52detulebecause all sensor drivers are tied to their media/video/msm crap
17:59.27detuleand there's duplicate ioctl's all over the place....
18:00.31detulethere should be a way to register the sensors independently and have them talk to their media server via the v4l2 subdev system
18:00.52detuleas it is, I can't take an upstream sensor driver and make it work with their server without modifying it extensively
18:00.57detulei might as well write it from scratch
18:01.00Cotullau know last times I prefer simple systems
18:01.13Cotullawith a lot of hardcoded things. they are much easy to maintain and rewrite from zero
18:01.20Cotullaas well works faster
18:01.24Cotullatakes less space
18:01.42CotullaI remembered how I try to figure
18:01.46Cotullahow msm-video works
18:01.48Cotulla:D
18:01.51Cotullathat was awful
18:03.14*** join/#htc-linux helicopter88_3 (~helicopte@host240-94-dynamic.22-79-r.retail.telecomitalia.it)
18:08.42fakkerslaps Cotulla
18:08.46fakkermotherf-
18:08.59Cotullaheadshot
18:09.05Cotullafakker dies
18:09.10Cotulla:(
18:09.37fakkeryou sadface for me?
18:09.38fakkerawwwwww
18:09.39fakker:D
18:10.02Cotullafakker respawns
18:10.07Cotullaheadshot
18:10.10Cotullafakker dies
18:10.11Cotulla:(
18:10.35fakkeraww
18:10.44fakkeryou too pro Cotulla
18:12.36Cotullaclockpro
18:12.37Cotulla:D
18:12.39fakkerpedik
18:12.46fakker:F
18:15.12Cotullau tell us truth finally
18:17.02MarcI think I have found out how HTC is building kernels: grab any given branch from CAF and port it to your device. afterwards delete random shit so anyone that would like to port that kernel to any similar device will have big problems doing so.
18:17.49Cotullawhy u think so?
18:18.03fakkercuz we know so
18:18.05fakkerpedik
18:18.25styblafakker: sounds like standard "enterprise" process to me ;)
18:18.47fakkerlike KFC?
18:18.57fakker:D
18:19.18MarcCotulla: I'm comparing a htc kernel with one from CAF directly
18:19.26Marcand that's the only explanation that makes sense :P
18:19.26*** join/#htc-linux helicopter88 (~helicopte@host200-116-dynamic.23-79-r.retail.telecomitalia.it)
18:19.35Cotullawhat do u found?
18:19.40fakkerlot of pedik
18:19.43fakkerMEDIC
18:19.45fakkerRTCW
18:19.51fakkerPEDIK
18:19.58styblaWunderbar!
18:20.43MarcI found out that they deleted random files for no obvious reasons
18:21.13styblaMarc: have you considered reasons like GNU GPL?
18:21.39Cotullafor example which files?
18:21.42Marcand files that are in the CAF branches are not covered by the GPL or what do you mean?
18:21.47styblaerrr that was incorrect, but I meant legal reasons.
18:21.48Marcclock-pcom-lookup.c
18:21.58Cotullamaybe it was added later to CAF?
18:22.02Marcpretty much all include/linux files for support of older devices
18:22.08Cotullabefore it was branched
18:22.12styblaMarc: yes, gnu/gpl was incorrect. I meant legal reasons in general.
18:22.13Marctheir kernel is newer ;)
18:22.23Cotullahow u detected that?
18:22.40Marchtc kernel is 3.0.16
18:22.50Marcat least it should be newer
18:22.58Marcand some code is actually newer by looking at the files
18:23.00Cotullaand CAF?
18:23.06Marc3.0.8
18:23.21Cotullainside msm files or inside generic files?
18:23.28Marcthen again, htc tends to randomly add updated files or use old ones
18:23.36Marcboth
18:23.47Cotullacuz it's called development :P
18:24.36Cotullau somehow assumed that all files in CAF are good
18:24.43Marcthey also tend to add newer patches and forget to apply the previous patches to that file ;)
18:24.46Cotullabut actually new changes often not well tested
18:24.52Cotullanew changes to any code
18:24.55MarcI'm not saying that CAF is perfect
18:24.58Cotullaso it's very common to use old instead
18:25.04Cotullaif something not working
18:25.11Cotullaso maybe they reverted
18:25.13Cotullareplaced
18:25.18Cotullasome files to old versions
18:26.55MarcI hope that they at least have an internal version control system :P
18:27.02Cotullawho knows
18:28.30Cotullabut good software can be developed without version control system too
18:28.48Cotullamy hello world was done without such systems and it works pretty well!
18:30.04*** join/#htc-linux helicopter88_2 (~helicopte@host98-116-dynamic.55-79-r.retail.telecomitalia.it)
18:30.14Marcif only one person works on it it's ok to not use a vcs
18:31.17Cotullabut hello world was done by me and my brain!
18:31.22Cotullaso two persons worked under it
18:31.59fakkerpfff
18:32.04fakkerstealing my credits again
18:32.16Cotullaand fakker worked under it too
18:32.19Cotullaso 3 persons
18:32.19fakker^
18:32.24Cotullawithout control system :D :P
18:32.29fakkeri just looked at it
18:32.33fakkerthat's all i needed to do
18:33.14Marcoh you made it write itself?
18:33.16Marcimpressive :P
18:33.52fakkeryep, why you think Cotulla talks to me? he is jealous
18:36.17Marchehe
18:49.25*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
18:58.24*** join/#htc-linux helicopter88 (~helicopte@host8-27-dynamic.49-79-r.retail.telecomitalia.it)
19:06.10ali1234never ceases to amaze me how broken source releases usually are
19:06.20ali1234from vendors i mean
19:07.05ali1234shit often won't compile unless you use the exact same release of RHEL they used, put source in exactly the same path, and have exactly the same user name as the developer
19:07.24ali1234and that's when they don't randomly delete stuff
19:10.00*** join/#htc-linux Bry8Star (~Bry8Star@gateway/tor-sasl/bry8star)
19:34.46*** join/#htc-linux khorben__ (~dont@194.29.236.67)
20:24.38*** join/#htc-linux zeusk_ (6ad095bb@gateway/web/freenode/ip.106.208.149.187)
20:24.45zeusk_hello world
20:25.51Marchellas
20:26.47zeusk_how is it goin marc ?
20:26.55zeusk_htc and caf still eating your brain ?
20:27.00Marcas always
20:27.47zeusk_they should start pushing stuff to mainline faster and test patches before they put stuff in -master
20:36.15MarcCotulla: did you write leo's touch driver (which I'm guessing since the file says so)?
20:39.19Cotullayes
20:39.22Cotullawho else
20:39.35MarcI'm guessing you used the synaptics i2c rmi code while writing it?
20:39.41Cotullanot at all
20:40.11Cotullait ever was not released at that time
20:40.26Cotullahi zerusk
20:40.37Marcjust wondering :P
20:40.49Marcanyway, found duplicated code :P
20:40.53Cotullaand?
20:41.21MarcI'm trying to implement a filter
20:43.01Cotullawhat kind of?
20:43.06Marcand while comparing with the synaptics code I noticed that you seem to be excessively using input_mt_sync() compared to the synaptics code
20:43.32Marcit's a duplicated filter which basically ignores small movements
20:44.21zeusk_hey cotulla
20:44.37Cotulla11
20:45.06zeusk_marc, you can just write your own basic filter, see delta over the last few updates
20:45.21zeusk_or use a static delta, if change is less than delta, ignore the current update
20:45.36Marcthat's basically what I'm adding ;)
20:46.04zeusk_linux is bloated with too much stuff which hardly anyone uses and then people complain that everyone does not use same interface
20:46.14zeusk_their interface is not made with everything in mind ofcourse
20:47.07zeusk_btw i tried using cotulla's ts driver under lk, never got a response from i2c address of TS chip
20:47.16Marcthat's what I'm basically trying to fix: https://www.youtube.com/watch?v=GRrKEDP59XE
20:47.24Cotullawhat kind response?
20:47.30zeusk_probably bad reset/init code for it, i know my gpio driver is foobar so may have to fix that first
20:47.48CotullaMarc yeah irritating in WRT
20:47.52Cotullau hold it
20:47.57Cotullafor context menu
20:48.06Cotullaand it drop menu in wrong location :D
20:48.12Marchehe
20:48.57zeusk_but then how would that filtering work when user is drawing line/swiping (using a static delta)
20:49.19Cotullamaybe if user moves thinger
20:49.26Cotullabut if he press
20:49.30Cotullaand hold
20:49.32Cotullanot
20:49.36zeusk_Hmm
20:50.01MarcI'll play around with the threshold
20:50.13Marcbravo has that filter
20:50.19Marcand it's set to 12, whatever that means :P
20:50.44zeusk_yeah, that should improve it. but i face that issue only when screen is dirty or fingers are sweaty
20:50.45Cotulla12 pixels maybe
20:50.56Marc12 pixels sounds like a lot to me
20:51.15zeusk_i think TS resolution is not equal to screen resolution
20:51.20zeusk_i'm not sure though
20:51.24Cotullayes it is
20:51.35Marcwe should measure in inches :P
20:51.57Cotullananometers only no way
20:52.38Marcin other news: I really wonder how lazy the broadcom coders are: https://github.com/marc1706/desire_kernel_35/blob/master/drivers/net/wireless/bcm4329/wl_iw.c#L5009
20:53.12zeusk_that doesn't beat this: http://imgur.com/TuKIC
20:53.15Cotullawhat?
20:53.26Marcthe useless brackets Cotulla ;)
20:53.44Cotullawhere?
20:53.58MarcLine 5010
20:54.03Cotullawhat's wrong there?
20:54.13zeusk_probably they copy/paste another function
20:54.16Marcwhat's the purpose of those brackets?
20:54.17CotullaI don't see anything
20:54.30Cotulla5010 and 5024?
20:54.36zeusk_cotulla, after else if on line 509, there is another indent and bracket on 510
20:54.46Cotullayes
20:54.48Cotullait's normal
20:54.51Cotullapurpose is simple
20:54.56zeusk_but TIT_HANDLER still wins
20:55.00Cotullakeep some region with own local variables
20:55.04Cotullacuz it's .C file
20:55.14Cotullau can't declare them like in C++ in anyplace
20:55.28Marcanyway, -O3 didn't like it :P
20:55.32Cotullaso to prevent scrolling up to function start, they were placed in local block
20:55.41CotullaI am using the techno also
20:55.45Cotulla*this
20:55.49CotullaGCC problem>
20:55.50Cotulla?>
20:56.00zeusk_o.O what did -O3 say about it ?
20:56.17zeusk_i never had problem, i use it { } all the time for multi line expressions
20:56.53zeusk_infact it is gcc standard iirc for multiline expression
20:56.53Marcnvm, it complained about the for loop right below it
20:56.57Marchttps://github.com/marc1706/desire_kernel_35/commit/61b973265cfa09184be2588aee32fb151e71bd0d
20:57.19Marcout of bounds :P
20:57.30Cotulla?
20:57.42Cotullaif u wanna truth . . .
20:57.48Cotullait's actually nothing
20:57.52Marchttps://github.com/marc1706/desire_kernel_35/commit/61b973265cfa09184be2588aee32fb151e71bd0d#L0L5149
20:57.53Cotullau should read into that code
20:57.59Cotullathat exactly it's doing
20:58.11Marcit didn't like the i+1
20:58.26zeusk_i wonder why do they even use bcopy
20:58.41zeusk_why not just use memcpy
20:58.47Cotullait's define
20:59.10zeusk_i know it is define but some standards are just so stupid
20:59.17CotullaCotullau should read into that code
20:59.17CotullaCotullathat exactly it's doing
20:59.28Cotullathen all ur questions are gone
20:59.29Cotulla:D
20:59.33Marchehe
20:59.37CotullaI am serious
20:59.42Marca bit long for a good night read :P
20:59.43Cotullaat 146%
20:59.47zeusk_i read printf("Hello World");
20:59.50Cotullait won't be good fater that
20:59.55fakker147
21:00.02Cotullafakker wins
21:00.05zeusk_fakker, 156
21:00.06Cotulla:D
21:00.09Cotullazeusk too late
21:00.13Cotullawe already counted
21:00.14fakker147 is top score in snooker
21:00.14zeusk_cockpro wins
21:00.14Cotulla:P
21:00.28fakker180!
21:00.28zeusk_~seen cockpro
21:00.30aptzeusk_: i haven't seen 'cockpro'
21:00.30Cotullahe drunk and he leaves
21:00.30fakkerDARTS
21:00.39zeusk_apt 42
21:00.39aptsomebody said 42 was the answer to life the universe and everything, see also http://en.wikipedia.org/wiki/the_answer_to_life,_the_universe,_and_everything
21:00.41Cotulla~seen cockpoo
21:00.42aptcockpoo <704f2866@gateway/web/freenode/ip.112.79.40.102> was last seen on IRC in channel #htc-linux, 9d 1h 4m 17s ago, saying: 'hi'.
21:00.43fakkerwho is drunk?
21:01.03zeusk_fakker, maybe, we all are ?
21:01.20fakkeri haven't had a drink for awhile
21:01.28fakkerbeing 11 days
21:01.41zeusk_what if someone mixed stuff in your water supply and you don't even know ? :o
21:01.47fakkeri drink my own urine
21:01.54Cotullawtf
21:02.00zeusk_wtf
21:02.24fakkerIs it necessary for me to drink my own urine?
21:02.28zeusk_is your real name by any chance, bear gryll ?
21:02.36fakkerNo, but I do it anyway because it's sterile and I like the taste.
21:02.47fakkeromgwtf
21:02.47fakkerhave you fuckers
21:02.51fakkernot seen Dodgeball?
21:02.57fakkerfucking brick to face
21:02.58zeusk_wtgbbq
21:02.59fakkerbrick
21:03.02fakkerto
21:03.06fakkerface
21:03.08fakkeryou don't even know the 5 D's of Dodgeball
21:03.13fakkershoots himself
21:03.35zeusk_Hmm, i used to think dodgeball was a game all this time
21:03.41fakkeryou've got to learn the five d's of dodgeball: dodge, duck, dip, dive and dodge!
21:04.09fakkerhttp://www.youtube.com/watch?v=18ASBsQfXnw
21:05.42zeusk_LOL
21:06.25zeusk_would you like to take a break from that fine lead based paint
21:06.26zeusk_LOL
21:07.12fakkergo rent the movie
21:07.58MarcI agree with fakker
21:08.22fakkerhe's gone to get it
21:08.23fakkerdamn he works fast
21:08.27Cotullayea
21:09.15fakkeryou too Cotulla
21:09.22fakkerthere's a Russian beauty in the movie
21:09.28fakkeryour kinda lady
21:13.57MarcDownloading with 25 MiB/s ... I love my internet connection :P
21:30.41fakkeri love... lamp
21:30.50fakker:D
21:31.13fakkeri'm hoping you know that one too
21:54.27*** part/#htc-linux Cotulla (~myfakemai@109.205.253.11)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.