IRC log for #htc-linux on 20130307

00:06.47*** join/#htc-linux {simonsimons34} (~SimonSick@50.2.8.91)
00:49.02*** join/#htc-linux DuperMan (~Duper@93-172-11-159.bb.netvision.net.il)
00:56.58*** join/#htc-linux surge (~surge@pool-98-118-154-23.bflony.fios.verizon.net)
01:01.25*** join/#htc-linux detule (~detule@unaffiliated/d3tul3)
01:29.08*** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info)
01:56.57eXtremoXirXes: bruce released cm10 with ion working. I think it's 3.4.10 kernel
02:18.00*** join/#htc-linux {simonsimons34} (~SimonSick@50.2.8.91)
02:28.00*** join/#htc-linux {simonsimons34} (~SimonSick@50.2.8.91)
02:28.08jonpryit looks like success to me. http://pastie.org/6408645
02:31.06*** join/#htc-linux Marc (~quassel@phpbb/modifications/marc1706)
02:35.12*** part/#htc-linux langpoo (~zz@109.205.253.11)
02:55.21*** join/#htc-linux phh (~quassel@137.194.15.151)
03:47.38XirXeseXtremo, He's actually just updating stuff from CAF 3.0.xx.
06:53.17*** join/#htc-linux kiozen (~kiozen@p578a42db.dip0.t-ipconnect.de)
08:15.36*** join/#htc-linux gauner1986 (~Adium@pd95b906c.dip0.t-ipconnect.de)
08:20.11*** join/#htc-linux lamikr (lamikr@nat/nokia/x-dgmkbduvsueiihro)
08:44.19*** part/#htc-linux gauner1986 (~Adium@pd95b906c.dip0.t-ipconnect.de)
09:55.10*** join/#htc-linux paulk-desktop (~paulk@lib33-1-82-233-88-171.fbx.proxad.net)
10:04.08*** join/#htc-linux |lippa|^ (~lippa@CPE-121-219-134-46.lnse2.lon.bigpond.net.au)
10:19.37*** join/#htc-linux lipp[a] (~lippa@101.160.202.217)
10:27.14*** join/#htc-linux Cotulla (~myfakemai@109.205.253.11)
10:39.07*** join/#htc-linux |lippa|^ (~lippa@CPE-124-181-138-10.vic.bigpond.net.au)
11:23.49*** join/#htc-linux detule (~detule@unaffiliated/d3tul3)
11:27.30*** part/#htc-linux Cotulla (~myfakemai@109.205.253.11)
11:38.04*** join/#htc-linux Cotulla (~myfakemai@109.205.253.11)
12:43.41*** join/#htc-linux gauner1986 (~Adium@pd95b906c.dip0.t-ipconnect.de)
12:48.35DuperManfor $LOTION in /dev/skin; do "rm -fr ~/.well && mkdir -p ~/.well/1"; else do "rm -fr ~/.well && mkdir ~/.well/0"; end [done?]
12:51.28*** part/#htc-linux gauner1986 (~Adium@pd95b906c.dip0.t-ipconnect.de)
13:18.56*** join/#htc-linux Bry8Star{T (~Bry8Star-@gateway/tor-sasl/bry8star-2)
13:31.07*** join/#htc-linux helicopter88 (~helicopte@host180-115-dynamic.23-79-r.retail.telecomitalia.it)
13:36.24DuperManI just honestly spammed for the first time in ... ever
13:36.26DuperManfun
13:36.26DuperManwanna bid on my GTX 560 ti 448 LE? auction's over soon and it shouldn't sell for high enough number of moneypoints
13:40.52fakkerlol
13:41.18fakkersets mode: +b DuperMan
13:41.40DuperMan<3
13:42.05DuperManhe who bids well haswell?
14:01.25*** join/#htc-linux stroughtonsmith (~steven@86-42-146-96-dynamic.b-ras1.bbh.dublin.eircom.net)
14:12.05*** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info)
14:15.23*** join/#htc-linux eR^Z` (~zzeratul@ip-94-112-113-32.net.upcbroadband.cz)
14:16.04*** join/#htc-linux XirXes (~xirxes@71-35-193-90.slkc.qwest.net)
14:23.13*** join/#htc-linux fakker (~fakker@cpc30-hitc6-2-0-cust607.9-2.cable.virginmedia.com)
14:23.14*** join/#htc-linux fakker (~fakker@unaffiliated/fakker)
15:00.57*** join/#htc-linux CptAJ (~cptaj@190.204.206.207)
15:16.28*** join/#htc-linux lamikr (lamikr@nat/nokia/x-rnkwejbhjooyqoyv)
15:54.22*** join/#htc-linux fakker (fakker@cpc30-hitc6-2-0-cust607.9-2.cable.virginmedia.com)
15:54.22*** join/#htc-linux fakker (fakker@unaffiliated/fakker)
16:42.57*** join/#htc-linux paulk-desktop (~paulk@lib33-1-82-233-88-171.fbx.proxad.net)
17:06.39*** join/#htc-linux fakker (fakker@cpc30-hitc6-2-0-cust607.9-2.cable.virginmedia.com)
17:06.40*** join/#htc-linux fakker (fakker@unaffiliated/fakker)
17:15.12fakkerhttps://bugbase.adobe.com/index.cfm?event=bug&id=3515154
17:33.53*** join/#htc-linux noobhands (fakker@cpc30-hitc6-2-0-cust607.9-2.cable.virginmedia.com)
17:33.54*** join/#htc-linux noobhands (fakker@unaffiliated/fakker)
17:45.54*** join/#htc-linux arrrghhh_ (~arrrghhh@unaffiliated/arrrghhh)
17:46.26*** join/#htc-linux ElFinLazz (~elfinlazz@182.215.84.22)
17:52.32*** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info)
17:53.04*** join/#htc-linux XirXes (~xirxes@71-35-193-90.slkc.qwest.net)
17:56.41*** join/#htc-linux helicopter88_2 (~helicopte@95.235.115.42)
19:04.39*** join/#htc-linux BabelO (~fcr@2a01:e35:8adc:3150:21e:8cff:fe2d:7d3f)
19:04.39*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
19:54.25*** join/#htc-linux helicopter88 (~helicopte@95.234.115.219)
19:59.34d3tul3hey jonpry
19:59.53jonpryhi d3tul3
20:00.04d3tul3busybox up and going?
20:00.10jonpryseems to be
20:00.30jonprydon't have a real terminal
20:00.30d3tul3nice, glibc?
20:00.35jonpryyeah
20:00.45jonprythe busybox is static though
20:00.52jonpryhaven't tried dynamic
20:01.11jonpryno SMP yet either
20:01.17Cotullad3tul3
20:01.23jonpryand it's flying off the table
20:01.24Cotulla3+3=6
20:01.34jonpryi guess it was running at 27mhz before
20:01.55d3tul327? isn't that thing rated at 400mhz or whatever
20:02.07jonpryit's at 540 now
20:02.36d3tul3nice
20:02.54jonpryyeah it's a bit faster
20:03.07jonpryespecially since each "core" only runs at clk/4 or clk/6
20:03.11d3tul3TI-83?
20:03.55d3tul3hey i have a question if you guys have a minute
20:04.31jonpryyeah
20:04.57d3tul3there's this userland side optimization of the gettimeofday() syscall
20:04.59d3tul3https://github.com/Evervolv/android_bionic/commit/e5bf5b23699a6e0bb69303d029d7b9da7dd60167
20:05.20d3tul3via kernel side user-helper pages
20:05.33d3tul3here's the kernel side code https://github.com/CyanogenMod/android_kernel_samsung_d2/blob/cm-10.1/arch/arm/kernel/entry-armv.S#L806
20:07.00jonpryyeah
20:07.07jonprythat's like armv6 TLS
20:07.12Cotullahey d3tul3 h0w d2y?
20:07.28d3tul3so this userspace implementation does all these reads at addresses ilke (*(int32_t *)0xffff0f54)
20:07.36jonpryright
20:07.37d3tul3that has to be virtual because it's done in userland
20:07.52jonpryit has to be virtual because the MMU is on
20:07.58d3tul3what the hell is it point to in the physical world?
20:08.31jonprysomewhere, maybe in mm/init.c
20:08.39jonprythere is code to map the "high page"
20:08.43jonpryto something
20:09.25d3tul3this https://github.com/CyanogenMod/android_kernel_samsung_d2/blob/cm-10.1/arch/arm/mm/mmu.c#L1018 >
20:09.26d3tul3?
20:10.01jonpryyeah that thing
20:10.17d3tul3so it's like entirely random, not a fixed location
20:10.20jonpryso it maps to vectors_page
20:10.34d3tul3vectors_page is allocated at boot
20:10.57jonprysounds right
20:11.06jonpryjust an allocated piece of physical ram
20:11.23d3tul3alrght, for some reason i thought high vectors had like a 1-1 identity mapping
20:11.37jonpryno
20:13.00d3tul3ok, thanks
20:13.58d3tul3i have to stop thinking that every hex number i see in an assembly file is a physical address :p
20:14.36jonprythe thing is
20:14.56jonpryin modern kernels that page is probably not readable or writable
20:15.03jonpryfrom userland
20:15.17jonpryso it causes a fault
20:15.36jonprythat is handled exactly like a syscall
20:23.01jonprymaybe that is only a problem with mainline kernels
20:26.53d3tul3<PROTECTED>
20:27.04d3tul3they have some sort of a time drift problem on the 3.0 kernel
20:27.38d3tul3i can't imagine that implementation of gtod is that much more efficient than the syscall
20:27.55Cotullayea it sucks
20:32.23jonprygo linaro
20:33.22Cotullano thks
20:33.25d3tul3no not you too
20:34.03jonpryi think linaro invented that patch
20:34.59Cotullabut it's completely stupid
20:35.11CotullaRTC should be accessed only by demand without polling
20:36.04jonpryit isn't rtc time
20:36.21jonpryit updates that shit on every tick with wall clock time
20:36.47jonpryer
20:36.50jonpryupdate_vsyscall_old(&xt, &tk->wall_to_monotonic, tk->clock, tk->mult);
20:37.09CotullaIn Windows CE it was global
20:40.59d3tul3every tick? there has to be better ways of doing that
20:41.40jonpryi'm not sure that is that bad
20:41.51jonprysome sort of clock gets updated every tick
20:42.08jonpryso at worst it's like duplicating a small struct
20:45.28*** join/#htc-linux rajkosto (~rajkosto@wan.rajkonet.info)
21:03.20*** join/#htc-linux gauner1986 (~Adium@hmbg-5f76425a.pool.mediaWays.net)
21:34.05*** join/#htc-linux gauner19861 (~Adium@hmbg-4d068804.pool.mediaWays.net)
21:54.58*** join/#htc-linux ALoGeNo (~alogeno@unaffiliated/alogeno)
22:19.14*** join/#htc-linux ALoGeNo (~alogeno@unaffiliated/alogeno)
22:21.33*** part/#htc-linux Cotulla (~myfakemai@109.205.253.11)
22:26.04*** join/#htc-linux gauner1986 (~Adium@hmbg-4d068804.pool.mediaWays.net)
22:38.16*** part/#htc-linux gauner1986 (~Adium@hmbg-4d068804.pool.mediaWays.net)
22:51.14*** join/#htc-linux ALoGeNo (~alogeno@214.Red-79-158-11.staticIP.rima-tde.net)
22:51.14*** join/#htc-linux ALoGeNo (~alogeno@unaffiliated/alogeno)
23:00.13*** join/#htc-linux kiozen (~kiozen@p578a42db.dip0.t-ipconnect.de)
23:30.22jonpryBrought up 2 CPUs
23:30.34jonprybut it's all a ruse
23:30.45jonprythere is something funny with ASID on different threads
23:30.53jonpryPC in coresys, SSR = 000A0000
23:31.07jonpryif i set any bits in SSR[ASID] it will fault
23:31.21jonprywhich is unrecoverable because EVB is not even reachable
23:31.40jonpryit's like global bit, is global only for thread 0
23:31.47jonprywhich can't possibly be right
23:38.05jonprynvm
23:38.13jonpryits just impossible to debug

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