IRC log for #htc-linux on 20120804

00:07.07*** join/#htc-linux Bry8Star (~Bry8Star@gateway/tor-sasl/bry8star)
00:13.02*** join/#htc-linux ccxCZ (~ccxCZ@178.238.36.182)
00:17.34*** join/#htc-linux rekoil (~rekoil@c-5fa6e055.04-2-64736c12.cust.bredbandsbolaget.se)
00:36.32*** join/#htc-linux non-sense (~eu@unaffiliated/non-sense)
00:37.17*** part/#htc-linux non-sense (~eu@unaffiliated/non-sense)
03:00.14*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
03:32.21*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
04:02.25*** join/#htc-linux NeoMatrixJR_away (~NeoMatrix@173-18-84-218.client.mchsi.com)
04:31.51*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
04:35.20*** join/#htc-linux Ondalf (~ondalf@cable-roi-fff8dd00-39.dhcp.inet.fi)
04:53.50*** join/#htc-linux LTxda (~anon@unaffiliated/ltxda)
05:12.14*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
05:27.16*** join/#htc-linux mes (~mes@S0106000ce55186df.cc.shawcable.net)
05:33.47*** join/#htc-linux LargePrime (~LargePrim@70-9-18-213.pools.spcsdns.net)
05:42.18*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
05:49.42*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
05:52.47*** join/#htc-linux Ondalf (~ondalf@cable-roi-fff8dd00-39.dhcp.inet.fi)
06:03.55*** join/#htc-linux MacDrunk (~marper@201.165.128.103)
06:05.05*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
06:18.04*** join/#htc-linux MacDrunk (~marper@201.165.128.103)
06:39.14*** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info)
07:00.12*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
07:18.14*** join/#htc-linux NeoMatrixJR_away (~NeoMatrix@173-18-84-218.client.mchsi.com)
07:18.21*** join/#htc-linux LargePrime_ (~LargePrim@108.101.68.152)
07:54.13*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
08:06.55*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
08:31.50*** join/#htc-linux eR^zeRa` (~zzeratul@88.103.98.168)
09:22.57*** join/#htc-linux marc1706 (~Marc@phpbb/modifications/marc1706)
09:40.50*** join/#htc-linux rob_w (~bob@host-188-174-193-83.customer.m-online.net)
09:40.51*** join/#htc-linux rob_w (~bob@unaffiliated/rob-w/x-1112029)
09:48.12*** join/#htc-linux Rajko (~rajkosto@cable-178-149-121-233.dynamic.sbb.rs)
09:49.11*** join/#htc-linux bitrot (~rajkosto@wan.rajkonet.info)
10:01.59*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
10:36.24*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
10:45.01*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
10:49.00*** join/#htc-linux walter79 (~walter79@dslb-088-068-017-149.pools.arcor-ip.net)
10:50.02*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
10:54.58*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
10:56.56*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
11:04.04*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
11:13.42*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
11:30.08*** join/#htc-linux skodde (~skodde@unaffiliated/skodde)
11:52.24*** join/#htc-linux eR^Z` (~zzeratul@88.103.98.168)
11:55.06*** join/#htc-linux zeusk_ (77eb3613@gateway/web/freenode/ip.119.235.54.19)
12:38.42*** join/#htc-linux NeoMatrixJR_away (~NeoMatrix@173-18-84-218.client.mchsi.com)
12:39.46*** join/#htc-linux Cotulla (~myfakemai@nat100-255-205-109.tvoe.tv)
12:58.24*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
13:27.23*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
13:32.54*** join/#htc-linux detule (~detule@unaffiliated/d3tul3)
13:40.36*** join/#htc-linux NeoMatrixJR (~NeoMatrix@173-18-84-218.client.mchsi.com)
13:45.20*** join/#htc-linux NeoMatrixJR_away (~NeoMatrix@173.18.84.218)
14:22.17*** join/#htc-linux AstainHellbring (AstainHell@unaffiliated/astainhellbring)
14:35.12*** join/#htc-linux ElFinLazz (~elfinlazz@182.215.84.22)
14:38.17*** join/#htc-linux GNUtoo (~gnutoo@90.84.144.141)
16:23.21*** join/#htc-linux NeoMatrixJR_away (~NeoMatrix@173-18-84-218.client.mchsi.com)
16:38.05*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
17:00.01*** join/#htc-linux GNUtoo (~gnutoo@90.84.144.85)
17:36.21*** join/#htc-linux mes (~mes@sentry.lazo.ca)
18:41.28marc1706zeusk or zeusk_: are you available right now?
18:41.36zeusk_yep
18:41.49marc1706I just found this commit: https://www.codeaurora.org/gitweb/quic/le/?p=kernel/msm.git;a=commitdiff;h=c9057d8ed06ce9c36f406e5a558a570f2eede02a
18:42.51zeusk_our framebuffer is far from codeaurora, securecrt is working on bringing it up on .32 right now
18:43.22marc1706I know, but is that all that makes triple buffer? O.o
18:44.58zeusk_yes, we were already using double buffer for vsync
18:45.23zeusk_so they should have already put up some frame selection code before switching 2 to 3
18:45.56zeusk_imo android is going in the wrong direction, i don't like triple buffering
18:45.59marc1706ah ok
18:46.17marc1706it makes it silky smooth though (that's from a movie :D )
18:46.54zeusk_marc1706, It is not only triple buffering, but many other optimisations
18:47.28zeusk_and triple buffering increases ram usage, requires too much logic moving around frames and selecting which one to draw on and which one to serve to screen
18:47.49marc1706hm, mdp from 3.0 compiled and then it told me I'm missing register_display_notifier ...
18:47.55zeusk_also it causes input lag since frames are 2~3 frames behind user input
18:48.09marc1706which is why they want vsync ...
18:48.13zeusk_have you checked your board file code ?
18:48.36zeusk_vsync is only for preventing screen tear (screen update while drawing is not complete)
18:48.53zeusk_so they use double buffer and on vsync swap buffers
18:49.27marc1706register_display_notifier is from msm_fb
18:49.35marc1706so the new one probably doesn't have that function anymore ...
18:50.03zeusk_what is it used for/in ?
18:50.15marc1706it's used in htc_battery
18:50.49zeusk_lol, htc battery is big piece of mess
18:50.59marc1706tell me about it
18:52.05zeusk_hmm, it's a google commit for using display notifier
18:52.11zeusk_i think it can be removed
18:52.34zeusk_along with htc_power_policy
18:53.09zeusk_wakelock for device charging animation... facepalm
18:53.23marc1706display_notifier(htc_power_policy, NOTIFY_POWER);
18:53.25marc1706yeah ...
18:53.47marc1706le commenting that crap out ...
18:54.55marc1706I also disabled all mipi files
18:55.39marc1706it's not like the leo would need those xD
18:56.08zeusk_were they even used ? :o
18:57.03zeusk_btw check "batt_notify" too, they're using the functions used to notify msmfb
19:00.39*** join/#htc-linux Alex[sp3dev] (~alexander@178.76.204.60)
19:00.56marc1706yeah, disabled that, too
19:06.50zeusk_should it be disabled ?
19:09.47Cotullalinux way detected
19:09.52Cotullahi marc1706
19:10.37marc1706well if there is no register_display_notifier and the batt_notify is basically using the display_notifier then I think I should disabled it :D
19:10.39marc1706hi Cotulla
19:11.19Alex[sp3dev]marc1706: calling display driver from the battery driver seems kinda strange :)
19:11.37CotullaAlexx, why?
19:11.57Cotullait's a usual linux way?
19:12.05Alex[sp3dev]Cotulla: no, it's usual qualcomm way
19:12.08zeusk_alex, they are doing that for wakelock (charging animation)...
19:12.19Cotullaor qualcomm
19:12.20Alex[sp3dev]zeusk_: fucktards
19:12.23zeusk_also to notify framebuffer for power policy... WTF
19:12.25Cotullaand put definition inside audio headers
19:12.56Alex[sp3dev]this is policy. this should be done in userland anyway. that's what cpufreq and framebuffer ioctl is for. to avoid hardcoding logic in drivers
19:13.11Cotullawhy not
19:13.47Cotullait's not needed to control that from userland
19:13.49Alex[sp3dev]because this makes drivers depend on each other and harder to maintain?
19:13.55Cotullaonly useless code which increase error ration
19:14.00Cotullasize of the system
19:14.05Cotullaas well decrease speed
19:14.08Alex[sp3dev]ok, when you write the code once and throw it away forever, that's ok
19:14.10Cotullaincrease complexity
19:14.28Alex[sp3dev]Cotulla: oh please. speed has never been an issue for the last decade
19:14.34CotullaAndroid?
19:14.35Cotulla:D
19:15.11Cotullaif we have component A with correlates with component B via internal way
19:15.15Cotullait can stay forever
19:15.19Alex[sp3dev]I wonder what was the reason not to make android fast from the beginning. why did it took 5 years?
19:15.32zeusk_dunno what's wrong with google but android is becoming laggy on leo while wp becomes smoother than android ever was..
19:16.32Alex[sp3dev]zeusk_: I guess people at google all have 16-core smartphones with infinite ram
19:16.37Cotullaso connection between battery and display driver is independent until they are in same address [space
19:16.57Cotullaas well such code can be ported to anything fast
19:17.30zeusk_cotulla but using a wakelock for charging animation is fucking stupid
19:17.47Alex[sp3dev]Cotulla: I once saw some 'clever' htc kernel code which just limited the cpu frequency when the screen was off :). This doesn't seem quite alright
19:17.49Cotullaif AC or USB connected - why not
19:17.57Cotullait's only question of charge speed
19:18.04Cotullawith a small difference
19:18.52Cotullawhen screen off user usually listening only music or sync something right?
19:19.29Alex[sp3dev]Cotulla: what if the user is listening to some 10 mbps music? or installing a game that extracts the archive?
19:19.48Cotullayeah in that case it sucks
19:20.09Cotullabut from other point of view, display DMA is off, so memory bandwidth is free.
19:20.17Cotullaand it can access memory faster
19:20.57zeusk_cotulla if the lock is for the fact that battery charger cannot work during suspend, does not have overcharge interrupt then they should say that not "Lock suspend for DOPOD charging animation"
19:20.58Alex[sp3dev]so. modular design allows easier portin of separate drivers. hardcoding policy in drivers makes them harder to maintain but allows to write them faster. so htc/qcom does that
19:22.08zeusk_Alex, is TI better than qualcomm for kernel stuff ?
19:22.32Cotulladisaggree with u. modular design usually assume heavy OS integration
19:22.39Cotullawhich make porting harder
19:22.43marc1706btw, what charging animation? :D
19:22.45marc1706the led?
19:22.52Alex[sp3dev]zeusk_: imho, yes. they have ALSA for audio and most code working in vanilla kernel
19:22.54Cotullano picture on the screen
19:22.56*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
19:23.15Alex[sp3dev]zeusk_: the best is samsung. they even have v4l2 camera in vanilla and DRM driver for hdmi/display
19:24.07Alex[sp3dev]zeusk_: I even had linaro 3.3 kernel running on my s2 and had made most hardware work with only writing the board file. though I failed to debug suspend without jtag
19:24.20Cotullau are not strong
19:24.24Cotullawhat is wrong there?
19:24.31zeusk_strong ? LOL
19:24.38marc1706this kind of reminds me of users complaining about the missing headset icon in ICS ... it makes sense for BT but having a notification icon that you have a wired headset connected is kind of stupid ...
19:25.07marc1706like you wouldn't know that you just connected your headset ...
19:25.09Alex[sp3dev]Cotulla: dunno. it just hanged. I don't know if it hanged on resume or before suspend. Had no time to do it back then.
19:25.22Cotullajust put some counter somewhere
19:25.24Cotullaand look
19:25.27Cotullathe last value
19:25.37Cotullabut I guess there is no reset button
19:25.39Cotullaright?
19:26.44Alex[sp3dev]Cotulla: right. have you seen a pda with a reset button recently?
19:26.55zeusk_marc1706. __htc_power_policy is needed i think, they're using it for vbus_wake_lock
19:27.04zeusk_Alex, yes
19:27.22zeusk_leo :D (though not so recent)
19:27.40Cotullano
19:27.55Cotullaso u may fail on that
19:28.00Cotullaon memory log after reset
19:28.55Alex[sp3dev]Cotulla: I guess I'll have to use UART. all stuff like vibrator/torch is controlled by i2c, not with a simple gpio
19:29.20Alex[sp3dev]well, I'll better use android kernel 3.0 now and do some userland stuff first
19:30.54Cotullaokk
19:31.03marc1706zeusk_: k, left it in
19:31.14marc1706will need to upgrade to jb or ics at least though ...
19:31.24marc1706probably better to upgrad to jb
19:31.37marc1706currently on gingerbread as I was testing something
19:32.27zeusk_i prefer gb on leo too, ics and jb are too laggy
19:32.38marc1706I didn't have issues with ics
19:37.36*** join/#htc-linux Ondalf (~ondalf@cable-roi-fff8dd00-39.dhcp.inet.fi)
19:37.39WilldHm, I have a tablet where the charger is on 12v. Is this error appropriate?
19:37.41Willd<3>[    6.374442] tegra-udc tegra-udc.0: vbus_draw regulator error -19; charging disabled
19:39.53zeusk_no idea what would msmfb do with the battery charge status...
19:40.21Alex[sp3dev]Willd: 19 is ENODEV
19:40.22Willdmsmfb?
19:40.37Alex[sp3dev]Willd: which probably means the regulator supply is not defined
19:40.39zeusk_willd, that was for previous discussion
19:40.45Willdzeusk_: Oh, soz
19:41.18WilldAlex[sp3dev]: I'm wondering if this is was intended..
19:42.14Willd-is*
19:42.47zeusk_willd if it's the official stock kernel, it should be fine
19:43.01WilldHah..
19:43.23WilldYou might not remember all the fuzz about TabCo..
19:43.39WilldI have one of FusionGarages Grid 10
19:43.51WilldMost things are probably not stock :(
19:45.26zeusk_:holymotherofgod:
19:45.30zeusk_Dotcom purchased €375,000 worth of shares of the nearly bankrupt company LetsBuyIt.com and subsequently announced his intention to invest €50 million in the company. Unknown to others, Dotcom did not have the funds available to invest, although the announcement caused the share value of LetsBuyIt.com to jump by nearly 300%. Dotcom sold his shares a few days later for €1,568,000
19:45.40zeusk_this guy is the hero of my life
19:46.13WilldUh, isn't that insider trading?
19:46.22zeusk_yep, he was arrested
19:46.41WilldBut not on that charge, eh?
19:47.26Cotullalol
19:47.28CotullaDotcom troll
19:48.21zeusk_Kim dotcom (megaupload guy)
19:50.16Cotullawhy he did that? :O
19:50.37WilldCotulla: Kind of how he works? :)
19:50.39CotullaGSOD
19:52.37Alex[sp3dev]zeusk_: so he was arrested because he didn't give FBI access to his CP archive?
19:52.57zeusk_CP ?
19:53.15*** join/#htc-linux frankdrey (~AndChat10@c-71-231-113-121.hsd1.wa.comcast.net)
19:53.22frankdreygot the toshiba tg01
19:53.24Alex[sp3dev]zeusk_: cmon, we all know what paid file hostings are for
19:53.31frankdreyNow I need to find that rom for it.
19:53.34frankdreyI have the kernel...
19:53.58frankdreyOr the ramdisk...
19:54.14frankdreyOn g1 they're together, so I'm not used to this :P
19:54.53zeusk_alex, i dunno what is CP archive (which is surely not co-processor)
19:56.51Alex[sp3dev]zeusk_: indeed, a central processor. really, why was megaupload closed? because FBI got jealous and wanted more money?
19:57.32zeusk_because of copyright infringement suit from MAFIAA
19:57.53zeusk_Music And Film industry Association of America
20:15.43zeusk_fail alex, CP.... i don't think that they would arrest him for that, isn't that stuff available all over net anyway ?
20:17.58Alex[sp3dev]zeusk_: as a guy living in russia I would assume he simply didn't share enough of his income with the officials :)
20:18.43zeusk_hmm
20:19.40Cotulla:D
20:20.12Alex[sp3dev]if RIAA wanted, say, X millon dollars, he could probably pay X/10 to the police instead :)
20:21.06zeusk_make that x/100
20:21.46jonprythat is what is wrong with america
20:22.27Alex[sp3dev]jonpry: so, this algorithm would work in america too? xD
20:25.50jonpryno
20:27.26Cotullajonpry, 10000$ for data from ur computer at work, ok?
20:27.55zeusk_LOL
20:28.09Alex[sp3dev]Cotulla: ok, I agree :)
20:28.24Cotullau in moscow unfortunatly
20:28.58Alex[sp3dev]Cotulla: does living in the USA automatically makes your computer have valuable data?
20:29.33Cotullajust checking "jonpry: so, this algorithm would work in america too? xD"
20:30.05zeusk_Cotulla, what would you do anyway with some crap microsoft excel sheets and probably some more corporate crap ?
20:32.23Cotullajonpry seems admin
20:32.35Cotullamaybe his DB has a lot of stuffs
20:33.34Alex[sp3dev]Cotulla: I'll tell you one little secret. large data (DB, sources) are kept on servers. workstations are just terminal clients :)
20:34.15Cotullayeah
20:35.16Alex[sp3dev]so you should ask for "all data from all machines in your local network reachable from your PC".
20:35.29Cotullaright ;)
20:35.41Cotulla"all data from all machines in your local country"
20:35.46Cotullathat's ever better!
20:36.01Alex[sp3dev]one problem is how to keep all that data
20:36.31Cotulla"all data from all machines in your local country on CD disks mailed by mail"
20:37.15zeusk_better use bluray or mail will cost more than bribe :P
20:37.34Alex[sp3dev]http://img13.imageshack.us/img13/6610/tierrahdeg2.jpg
20:38.52Cotullalol yeah
20:40.35zeusk_lol
21:07.25jonprylike the RIAA spends all this energy trying to sue youtube when they play there crap for fractions of pennies on the radio and MTV
21:07.37jonprybut they don't want money
21:07.54jonprythey want to call the shots and sell stuff on the internet when they choose
21:08.38jonpryso everybody else has to get sued and go to jail for watching music videos online because the RIAA can't figure out if they want .00001 cents or .00002
21:09.05*** join/#htc-linux Echo31 (~olivier@lan31-9-88-177-158-48.fbx.proxad.net)
21:09.07Alex[sp3dev]yeah. and if you don't want to buy from them or even listen to music, you're a criminal because you leave them without their undrawn profit
21:10.56zeusk_seriously, how many associations does america even have ? i've even heard they have one for dentists...
21:11.44Alex[sp3dev]zeusk_: if you have perfect teeth that don't hurt, you're depriving dentists from income! you must go to jail
21:12.03zeusk_lol
21:19.26jonpryif the RIAA could just learn to take bribes like normal people
21:21.17Alex[sp3dev]RIAA is just a way to distract people while laws and constitution are changed to allow more control over private life
21:26.28zeusk_SOPA, lamar smith must burn in hell
21:28.04*** join/#htc-linux mastermerlin (~Adium@p4FEE4CD8.dip.t-dialin.net)
21:30.37*** join/#htc-linux walter79 (~walter79@dslb-088-068-017-149.pools.arcor-ip.net)
21:48.24*** join/#htc-linux Cotulla (~myfakemai@nat100-255-205-109.tvoe.tv)
21:48.31Cotullahm
21:48.35Cotullais it possible to disable ARM code
21:48.38Cotullaallow only THUMB2?
21:54.18*** join/#htc-linux leviathan (~quassel@2001:470:26:484:6ef0:49ff:fee6:8dca)
21:59.21zeusk_cotulla, don't think so
21:59.50zeusk_but it is possible to run in thumb-2 only, get a cortex-M0
23:20.51*** join/#htc-linux GNUtoo (~gnutoo@90.84.146.205)
23:22.36*** join/#htc-linux BabelO (~fcr@AMontpellier-553-1-235-86.w86-202.abo.wanadoo.fr)
23:22.36*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
23:50.55*** join/#htc-linux surge (surge@pool-98-118-154-23.bflony.fios.verizon.net)
23:53.38*** join/#htc-linux surgex (surge@pool-98-118-154-23.bflony.fios.verizon.net)

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