IRC log for #htc-linux on 20120418

00:00.25detuleoh no microp
00:03.50jonprypresumably it is similar to rhod microp anyways
00:04.21detuleyeah he would be better off trying to write a whitestone board file or two in 3.x than tracing through the microps in .27
00:04.37detulefigure out the new toolchain?
00:05.27jonpryhaven't tried but i think those changes to msm_tssc.h will help
00:07.02jonpryprobably have to disassemble the object files to really figure it out
00:08.10jonprymaybe we can upgrade tssc manager to a version that doesn't use bitfield structs
00:09.30detulei seem to remember something recently on lkml about bitfields and gcc
00:10.13jonpryit generates psycho code
00:10.28detulei don't even know what the lineage of our tssc_manager is
00:10.31jonpryfortunately we can remove 90% of them without any real work
00:10.41jonpryits an htc contraption
00:10.55detuleoh so if i poke in a kernel of theirs i might find one
00:11.24jonpryit has to be from a resistive touch device otherwise they don't use the msm controller
00:12.48jonpryrewriting it to not use bitfields at all is probably not beyond our technical capabilities
00:13.20rpierce9922h later I'm at 60%
00:13.30detulewifi on or off? :)
00:13.33rpierce99on
00:13.46rpierce99and my number ported so it's rang 3 or 4 times
00:14.02detulei suspect you'll find similar numbers with wifi off whenever you run that experiment
00:14.04detulethough i could be wrong
00:14.42rpierce99losing 40% in 22h is enough testing i can turn wifi on, or charge again and then turn wifi on
00:14.51rpierce99^if
00:15.21detuleprobably best to recharge anew before starting a new test -> going back to winmo to read the battery?
00:15.33rpierce99no this is just in android
00:15.40detulemy android battery indicator is at 73% after 12 hrs
00:15.48detulewith mobile data and what not
00:16.05rpierce99are you running with this pm too?
00:16.38detulesame pm, other changes though that prevent idle power collapse...and no wifi so mine is power collapsing in suspend
00:16.47detuleyour kernel though powercollapses in idle
00:17.07detuleso technically it could be that even though wifi is on you are still powercollapsing
00:17.18rpierce99so to get better battery life we prevent sleeping? seems backwards
00:17.34detuleno better battery life turn off mobile data
00:17.59detulejust booted in winmo to check my battery -> 60% (this is what i mean by lately android giving me accurate readings)
00:18.21detulenot bad 40% after 12-13hrs with mobile data
00:18.39detulethough i don't really trust either indicator with my after market battery
00:25.01detulejonpry, thouse samples_avg the only bitfields it uses?
00:25.36jonprycontrol registers
00:28.51jonprylooking at the disassembly i don't see anything strange
00:29.05jonprybut maybe that is because i built it with the old toolchain
00:36.13detulei don't understand what's the point of assigning a data type to a bitfield
00:36.25detulelike unsigned long foo :16
00:36.30detulewhy the long?
00:36.39*** join/#htc-linux AstainHellbring (AstainHell@unaffiliated/astainhellbring)
00:36.51jonpryi'm not sure
00:37.14jonpryi think it is just for dealing with strong type checking options
00:38.11jonpryhttp://pastebin.com/M02K67bi
00:38.44jonprydiff of the disassemblies
00:39.17rpierce99winmo also says 60%
00:39.58jonprynothing is actually different except for some addresses and the handling of 16 bit bitfields
00:40.34jonpryso i guess ldrh is not equiv to ldr lsr
00:41.56detulethat's quite a diff
00:43.59detule30mins to download the new toolchain
00:44.22jonpryi think ldrh and ldr lsl are the same
00:44.37jonprywhat is stranger is the handling of the lower halfs
00:44.52jonpryas ldr and ldrh are definitely not equiv
00:47.23jonprywhich would explain why x works and y doesn't
00:49.16detulex didn't quite work either
00:49.24detulefrom what i remember
00:49.27detuleit was pretty erratic
00:50.19jonprywell i blame it on this ldr/ldrh thing
00:50.30jonprywhich is actually a gcc bug
00:50.32detulehttps://lkml.org/lkml/2012/2/1/220
00:51.04*** join/#htc-linux |lippa|^ (~lippa@CPE-124-180-129-57.lnse4.lon.bigpond.net.au)
00:52.11ahigerdTECHNICALLY that's not a bug?
00:52.23ahigerdNot that it much helps
00:56.06jonprychanging the bit field struct to shorts causes the return of ldrh
00:56.38detulewith the new toolchain/
00:58.46jonpryyes
00:58.55jonpryrecommended patch: http://pastebin.com/bnYb8LZ5
01:01.06jonprywe should probably tell the man about this one
01:01.14detuleyou boot this?
01:01.21jonpryno
01:01.38detulethe man sounds busy
01:02.13jonprybut this is broke ass
01:05.45jonprythough i imagine they will say something like bitfields are not part of the c specification and thus their behavior is always undefined
01:06.42detulebefore telling you it's our own fault for trying to use proprietary drivers on unsupported hardware
01:10.33*** join/#htc-linux lipp[a] (~lippa@CPE-203-51-167-249.lns11.lon.bigpond.net.au)
01:13.23detulejonpry, you da man
01:13.39jonpryit works?
01:13.44detulesure does
01:14.07jonprynot sure what we are going to do with a new compiler
01:14.18jonpryprogress is progress i guess :p
01:14.32detuleisn't there something here about less overhead due to the alignment trap or what not
01:14.41detulenot sure if we need to disable some config option or something
01:15.11jonpryi think the alignment trap really is broken on armv6
01:15.22jonprybut we don't have anything that needs it that i know of
01:21.38*** join/#htc-linux ahigerd (~ahigerd@173.218.156.102)
01:21.39*** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd)
01:23.16detuleer i guess this was 2011.03 not quite the latest toolchain
01:24.02jonprynot sure how old mine is either. its possible the bug was fixed
01:24.41jonpryor not. i guess captaj has something pretty new
01:26.30detulei'll try it later with the latest, trying with alignment trap disabled now expecting this thing to take off
01:27.13jonpryhyperspace jump?
01:28.25detulenot that i can tell gotta go thanks for squashing this
01:39.18ahigerdRemind me why stock gcc-arm can't build the kernel? You told me before but I forgot
01:41.20jonpryi think it can now
01:41.33*** join/#htc-linux CptAJ (AJ@190.74.22.95)
01:45.52ahigerdAh, nice. detule had mentioned something about alignment problems which is what reminded me
01:49.34*** join/#htc-linux |lippa|^ (~lippa@CPE-121-220-120-207.lnse3.win.bigpond.net.au)
01:50.38jonpryyou were gone when i fixed it
01:51.37ahigerdOh, heh, nice
01:52.11ahigerdMaybe I should try it, since I have stock 4.5.0 gcc arm-elf installed
01:53.06ahigerd(I know it's a little out of date; I haven't needed to update it since this is perfectly fine for building binaries for my job)
01:56.10jonprygive it a whirl
01:56.46ahigerdWhat's the best repo to clone?
01:57.52jonprylinux-msm-rhod
01:57.56ahigerdLink?
01:58.41jonpryhttps://gitorious.org/linux-msm-rhod
01:59.12ahigerdbows
01:59.41ahigerdmake msmconfig, I believe I heard?
02:00.06jonpryi just copy arch/arm/configs/msm_defconfig .config
02:00.15ahigerdAh, msm_defconfig, okay
02:00.18ahigerdThanks
02:00.27jonprythen its make ARCH=arm CROSS_COMPILE=/path/to/gcc zImage
02:01.07ahigerdYeah, I remember that part from building a kernel for my work device
02:02.53ahigerdheadscratches at gitorious
02:03.00ahigerdThat can't be right... gotta be a mistake. GPL3?
02:03.09ahigerdI thought the Linux kernel was GPL2 with no upgrade clause
02:03.26jonpryyeah i dunno [acl] setup the project
02:04.19ahigerd...bah *pokes gitorious with stick* clone failed
02:04.21ahigerdretries
02:05.37ahigerdHuh.
02:05.39ahigerdI can't clone.
02:06.26jonprygit://gitorious.org/linux-msm-rhod/linux-msm-rhod.git ?
02:06.42ahigerdYup
02:07.02ahigerdTrying https just to see if it makes a difference, but I'm getting early EOFs
02:08.44jonprysometimes its just broke
02:10.42ahigerdGood job gitorious. >.>
02:10.59ahigerdWell. https hasn't DIED yet, so... that's a good sign I guess?
02:11.28ahigerdBetter than the 3.5 hours it took to push my work project to Codesion last week x.x
02:12.00jonpryit will take a long time to clone that repo. close to 1gb
02:12.28ahigerdYeah, the .git of that project that took 3.5 hours was only 343MB
02:12.42ahigerdWhy is it that free project hosts have better performance than paid ones?
02:13.08ahigerdis trying to talk the company into switching to Atlassian
02:14.23rpierce99we use the atlassian suite at work
02:14.38ahigerdI use Bitbucket for all of my personal projects
02:15.01ahigerdSo I'm going to hope that Atlassian:Bitbucket !== CollabNet::SourceForge
02:15.06ahigerder
02:15.10ahigerdI totally botched the punctuation there
02:15.11ahigerdBut you get it
02:15.16ahigerd(CollabNet is AWFUL.)
02:16.40rpierce99I dunno, call me a microsoft fanboy but I'll take VS2010 + TFS over pretty much anything
02:17.06ahigerdTFS?
02:17.13rpierce99team foundation server
02:17.28ahigerdI'll give VS2010 props for being a solid IDE, but that doesn't make up for the problems inherent with the REST of developing on Windows.
02:17.54rpierce99true, i've only developed C# in that environment
02:18.08ahigerdBasically all of my work is C++
02:18.36ahigerdYou don't do DSP in C#. ;)
02:19.07ahigerdYou also try to avoid cross-platform development in C# >.>
02:19.51rpierce99yeah my work is primarily a windows shop, although much of the dotcom team is rolling macs these days, but all of the proprietary apps are either web apps or windows
02:20.48ahigerdAFAICT our customers are pretty evenly split between Windows and Mac
02:21.03ahigerdBut then again I'm in the music business
02:21.47rpierce99yeah that does tend to sku things, I'm surprised half of them still run windows
02:22.02ahigerdDon't be
02:22.12rpierce99s/sku/skew/
02:22.24ahigerdHeh, nice bot
02:22.32rpierce99oh you haven't met apt?
02:22.34ahigerdNope
02:22.41rpierce99apt where is ramconsole
02:22.41aptextra, extra, read all about it, ramconsole is pwf dm 0x8e0000 0x20000
02:22.46ahigerdNice.
02:22.59rpierce99apt kill ahigerd
02:22.59aptACTION shoots a hyper-charged anti-graviton gun at ahigerd
02:23.04ahigerd;_;
02:23.21ahigerdapt, who are you?
02:23.22aptahigerd: I think you lost me on that one
02:23.26ahigerdHeh
02:23.31ahigerdLast bot I played with could answer that one
02:23.46ahigerderror: Unable to get pack index https://git.gitorious.org/linux-msm-rhod/linux-msm-rhod.git/objects/pack/pack-98d762584b5475dbd94986aec5f4285d81ed9182.idx
02:23.51ahigerdHmm, nope, just going to ahve to try this again tomorrow
02:24.05jonprywe also have unilinky
02:24.22jonpry[[foo]]
02:24.23unilinkyhttp://htc-linux.org/wiki/index.php?title=foo
02:25.33ahigerdConvenient. Over in #qt we just have !rtfm and some factoids
02:26.01ahigerd(Though !rtfm is nice)
02:26.36rpierce99do you really need a bot for that? lol
02:27.38ahigerdIt's just like unilinky there
02:27.48ahigerd!rtfm QLabel::setCaption
02:27.49ahigerdor osmething
02:28.02ahigerdAnd it makes a link, including an anchor to the desired method
02:28.08rpierce99got it
02:28.30ahigerdAnd for people who only look at their highlights, !tell rpierce99 [rtfm QFoo]
02:28.32rpierce99we should have one that parses the text for obvious stupid questions and sends them to http://www.catb.org/~esr/faqs/smart-questions.html
02:28.36ahigerdXD Yes.
02:29.32ahigerdOr "Can anyone ____?" -> "Yes."
02:29.53rpierce99Can you _____? -> If I wanted to, yes. For you, no.
02:30.01ahigerdEven better.
02:30.33rpierce99apt can you tell me how to install linux on my htc?
02:30.52ahigerdOkay guys, night
02:31.03rpierce99night
02:45.47*** join/#htc-linux jonpry (~jon@adsl-98-85-69-254.mco.bellsouth.net)
03:35.17*** join/#htc-linux detule (~detule@unaffiliated/d3tul3)
03:57.07detulesilly Werror flag
05:12.57*** join/#htc-linux Regulator (~duffman@c-68-47-141-16.hsd1.ar.comcast.net)
05:13.08*** join/#htc-linux Regulator (~duffman@unaffiliated/regulator)
05:38.23*** join/#htc-linux kiozen (~kiozen@ppp-93-104-75-139.dynamic.mnet-online.de)
07:04.19*** join/#htc-linux MacDrunk (~marper@201.165.128.158)
07:10.12*** part/#htc-linux MacDrunk (~marper@201.165.128.158)
07:52.17*** join/#htc-linux gauner1986 (~Miranda@ip-109-91-241-106.unitymediagroup.de)
07:54.24*** join/#htc-linux NeoMatrixJR_away (~NeoMatrix@173-18-84-218.client.mchsi.com)
08:25.48*** join/#htc-linux swc|666 (~goldbond@Aircrack-NG/Friend)
09:55.18*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
10:31.09*** join/#htc-linux zeusk (77eb360e@gateway/web/freenode/ip.119.235.54.14)
10:59.53*** join/#htc-linux khorben_ (~dont@port-83-236-217-161.static.qsc.de)
11:01.04*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
11:08.43*** join/#htc-linux Alex[sp3dev] (~alexander@85.202.228.18)
11:50.22*** join/#htc-linux yuuki_ (~yuuki@dsbg-4d0d2c81.pool.mediaWays.net)
11:53.49*** join/#htc-linux yuuki_ (~yuuki@dsbg-4d0d2c81.pool.mediaWays.net)
11:55.11*** join/#htc-linux _Joey_ (~Yuuki@dsbg-4d0d2c81.pool.mediaWays.net)
12:27.21*** join/#htc-linux ElFinLazz (~elfinlazz@182.215.84.22)
12:32.49*** join/#htc-linux ahigerd (~ahigerd@173.218.156.102)
12:33.07*** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd)
12:51.17zeuskAlex[sp3dev] you there ?
12:51.50*** join/#htc-linux helicopter88 (~helicopte@95.235.115.81)
12:52.58Alex[sp3dev]zeusk: yes
12:55.30zeuskhow is this ? http://pastebin.com/MNL5T7A7
12:55.46zeuski am using it on leo to have multiple fonts at the same time and can switch between them during runtime
12:56.04zeuskgood for writing menu system, which i am doing now
12:56.14*** join/#htc-linux gauner1986 (~Miranda@ip-109-91-241-106.unitymediagroup.de)
12:58.16zeuskit worked :D
13:01.02Alex[sp3dev]zeusk: the code looks ok, but i don't think it is necessary. 8x8 font is readable on x1, and of course on larger displays. i don't know why you need to switch fonts at runtime. as for menu, i had it once (you may look up older revisions) but it's proven useless. fastboot is enough. and btw, look at CA, they have a newer version of lk with some 2d graphics code and rewritten fbcon
13:01.36zeusklol, they just copied that from newos.org/lk.git
13:02.08zeuski thought of using that, but it seems too complex and unecessary
13:03.20*** join/#htc-linux detule (~detule@unaffiliated/d3tul3)
13:10.15zeuskbtw i did that because 8x8 was barely readable on leo and i can't use 8x16 everywhere in the menu, otherwise it would be very small.
13:18.24*** join/#htc-linux gauner1986 (~Miranda@ip-109-91-241-106.unitymediagroup.de)
13:32.37*** join/#htc-linux paulk-desktop (~paulk@lib33-1-82-233-88-171.fbx.proxad.net)
13:34.14*** join/#htc-linux ahigerd (~ahigerd@173.218.156.102)
13:34.14*** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd)
13:36.17*** join/#htc-linux DuperMa (~Duper@46-116-62-170.bb.netvision.net.il)
13:43.53*** join/#htc-linux DuperMan (~Duper@46-116-63-242.bb.netvision.net.il)
13:59.27*** join/#htc-linux rpierce99 (~rpierce99@96-42-107-19.dhcp.stcd.mn.charter.com)
14:03.01*** join/#htc-linux mgross029 (c0234f46@gateway/web/freenode/ip.192.35.79.70)
14:24.48*** join/#htc-linux gauner1986 (~Miranda@ip-109-91-241-106.unitymediagroup.de)
14:30.03ahigerdchecks to see if gitorious is being less stupid today
14:44.29*** join/#htc-linux anver (~miranda@shpd-95-53-191-145.vologda.ru)
14:44.40*** part/#htc-linux anver (~miranda@shpd-95-53-191-145.vologda.ru)
14:46.45*** join/#htc-linux DuperMa (~Duper@46-116-63-242.bb.netvision.net.il)
14:47.05*** join/#htc-linux DuperMan (~Duper@46-116-63-242.bb.netvision.net.il)
14:47.34zeuskLOL, Alex, i found why switching font during runtime won
14:47.54zeuskwon't be good, seems font_width and font_height is not constant
14:48.03zeuskso it's a problem.
14:53.29*** join/#htc-linux whitekidney (~wk@141.0.136.22)
14:56.05*** join/#htc-linux DuperMan (~Duper@46-116-63-242.bb.netvision.net.il)
15:06.11*** join/#htc-linux DuperMan (~Duper@46-116-63-242.bb.netvision.net.il)
15:17.26*** join/#htc-linux DuperMan (~Duper@46-116-63-242.bb.netvision.net.il)
15:18.12*** join/#htc-linux DuperMa (~Duper@46-116-63-242.bb.netvision.net.il)
15:21.17*** join/#htc-linux DuperMan (~Duper@46-116-63-242.bb.netvision.net.il)
15:48.51*** join/#htc-linux DuperMan (~Duper@46-116-63-242.bb.netvision.net.il)
15:50.52*** join/#htc-linux rob_w (~bob@ppp-188-174-41-66.dynamic.mnet-online.de)
15:50.53*** join/#htc-linux rob_w (~bob@unaffiliated/rob-w/x-1112029)
16:12.44*** join/#htc-linux xUmaRix (3c3129b5@gateway/web/freenode/ip.60.49.41.181)
16:18.06*** join/#htc-linux LordDeath (~LordDeath@cable-81-173-164-174.netcologne.de)
16:50.35*** join/#htc-linux whitekidney (~wk@141.0.136.22)
17:02.02*** join/#htc-linux |Jeroen| (~jeroen@d5153E8CC.access.telenet.be)
17:10.06*** join/#htc-linux MethoS- (~clemens@134.102.106.250)
17:13.47*** join/#htc-linux mgross029 (c0234f46@gateway/web/freenode/ip.192.35.79.70)
17:17.24*** join/#htc-linux helicopter88_2 (~helicopte@95.235.115.81)
17:29.03*** join/#htc-linux ahigerd (~ahigerd@173.218.156.102)
17:29.03*** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd)
17:34.34*** join/#htc-linux mes_ (~mes@S0106000ce55186df.cc.shawcable.net)
17:39.56*** join/#htc-linux kiozen (~kiozen@ppp-93-104-75-139.dynamic.mnet-online.de)
17:50.03zeuskwhat will happen if i call clock_enable with invalid id (-1) through proc_comm ?
17:50.26zeuskwill baseband crash or just put some kind of error in the APP_CMD (result)
18:18.22*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
18:33.42ahigerdUm. *stares at compiler* How the heck did I land an OABI ARM compiler O.o
18:33.50ahigerdXD
18:33.52*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
18:34.24ahigerdOh. I didn't. Never mind.
18:36.03*** join/#htc-linux helicopter88_2 (~helicopte@host155-119-dynamic.31-79-r.retail.telecomitalia.it)
18:42.45*** join/#htc-linux helicopter88 (~helicopte@host155-119-dynamic.31-79-r.retail.telecomitalia.it)
18:45.05ahigerdNo, I did.
18:45.47zeusk...
18:46.21ahigerdI thought I was missing a cflag
18:46.36ahigerdBut when I set it, it didn't change anything
18:46.41ahigerdSo I need to get myself a new compiler. XD
18:53.35zeuskI don't usually download toolchains, but when i do i make sure they're outdated ones ;)
18:53.58helicopter88Codesourcery 2010/9 FTW
18:55.23ahigerdThe toolchain I have was one I had installed for something else
18:55.26ahigerdI was just hoping it'd work :P
19:28.42*** join/#htc-linux paulk-desktop (~paulk@lib33-1-82-233-88-171.fbx.proxad.net)
19:29.28*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
20:00.06*** join/#htc-linux swc|666 (~goldbond@Aircrack-NG/Friend)
20:18.29jonprywe can compile with new toolchains now. my phone is currently running such a contraption
20:22.11*** join/#htc-linux ahigerd (~ahigerd@173.218.156.102)
20:22.12*** join/#htc-linux ahigerd (~ahigerd@libqxt/developer/ahigerd)
20:25.51*** join/#htc-linux MN-- (~yaaic@027d6295.bb.sky.com)
20:29.37*** part/#htc-linux MN-- (~yaaic@027d6295.bb.sky.com)
20:47.22*** join/#htc-linux raymonddull (~raymonddu@c-69-245-114-102.hsd1.mi.comcast.net)
21:23.12*** join/#htc-linux rajkosto (~rajkosto@de.iandoh.info)
21:25.05*** join/#htc-linux LordDeath (~LordDeath@cable-81-173-165-254.netcologne.de)
21:49.57*** join/#htc-linux detule (~detule@unaffiliated/d3tul3)
22:07.44detulejonpry, up to 3.3.1
22:07.49detulehad to undo some damage to drivers/usb
22:32.31jonpryi've been running close to 20 hours with data on
22:32.43jonprysays 50%
22:33.54rpierce99my wifi off test is about 8.5 hrs in and android says 84%
22:34.23detulejonpry, what kernel?
22:36.21jonprygood question
22:36.21arrrghhhlol
22:36.39jonpry3.2 with your timer.c and idle pc patches. newer toolchain
23:36.13*** join/#htc-linux NetRippah (~netripper@tikkie.net)
23:51.52*** join/#htc-linux jonpry (~jon@adsl-98-85-69-254.mco.bellsouth.net)

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