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) |
01:56.57 | eXtremo | XirXes: 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) |
02:28.08 | jonpry | it 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) |
03:47.38 | XirXes | eXtremo, 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) |
12:48.35 | DuperMan | for $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) |
13:36.24 | DuperMan | I just honestly spammed for the first time in ... ever |
13:36.26 | DuperMan | fun |
13:36.26 | DuperMan | wanna bid on my GTX 560 ti 448 LE? auction's over soon and it shouldn't sell for high enough number of moneypoints |
13:40.52 | fakker | lol |
13:41.18 | fakker | sets mode: +b DuperMan |
13:41.40 | DuperMan | <3 |
13:42.05 | DuperMan | he 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) |
17:15.12 | fakker | https://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) |
19:59.34 | d3tul3 | hey jonpry |
19:59.53 | jonpry | hi d3tul3 |
20:00.04 | d3tul3 | busybox up and going? |
20:00.10 | jonpry | seems to be |
20:00.30 | jonpry | don't have a real terminal |
20:00.30 | d3tul3 | nice, glibc? |
20:00.35 | jonpry | yeah |
20:00.45 | jonpry | the busybox is static though |
20:00.52 | jonpry | haven't tried dynamic |
20:01.11 | jonpry | no SMP yet either |
20:01.17 | Cotulla | d3tul3 |
20:01.23 | jonpry | and it's flying off the table |
20:01.24 | Cotulla | 3+3=6 |
20:01.34 | jonpry | i guess it was running at 27mhz before |
20:01.55 | d3tul3 | 27? isn't that thing rated at 400mhz or whatever |
20:02.07 | jonpry | it's at 540 now |
20:02.36 | d3tul3 | nice |
20:02.54 | jonpry | yeah it's a bit faster |
20:03.07 | jonpry | especially since each "core" only runs at clk/4 or clk/6 |
20:03.11 | d3tul3 | TI-83? |
20:03.55 | d3tul3 | hey i have a question if you guys have a minute |
20:04.31 | jonpry | yeah |
20:04.57 | d3tul3 | there's this userland side optimization of the gettimeofday() syscall |
20:04.59 | d3tul3 | https://github.com/Evervolv/android_bionic/commit/e5bf5b23699a6e0bb69303d029d7b9da7dd60167 |
20:05.20 | d3tul3 | via kernel side user-helper pages |
20:05.33 | d3tul3 | here's the kernel side code https://github.com/CyanogenMod/android_kernel_samsung_d2/blob/cm-10.1/arch/arm/kernel/entry-armv.S#L806 |
20:07.00 | jonpry | yeah |
20:07.07 | jonpry | that's like armv6 TLS |
20:07.12 | Cotulla | hey d3tul3 h0w d2y? |
20:07.28 | d3tul3 | so this userspace implementation does all these reads at addresses ilke (*(int32_t *)0xffff0f54) |
20:07.36 | jonpry | right |
20:07.37 | d3tul3 | that has to be virtual because it's done in userland |
20:07.52 | jonpry | it has to be virtual because the MMU is on |
20:07.58 | d3tul3 | what the hell is it point to in the physical world? |
20:08.31 | jonpry | somewhere, maybe in mm/init.c |
20:08.39 | jonpry | there is code to map the "high page" |
20:08.43 | jonpry | to something |
20:09.25 | d3tul3 | this https://github.com/CyanogenMod/android_kernel_samsung_d2/blob/cm-10.1/arch/arm/mm/mmu.c#L1018 > |
20:09.26 | d3tul3 | ? |
20:10.01 | jonpry | yeah that thing |
20:10.17 | d3tul3 | so it's like entirely random, not a fixed location |
20:10.20 | jonpry | so it maps to vectors_page |
20:10.34 | d3tul3 | vectors_page is allocated at boot |
20:10.57 | jonpry | sounds right |
20:11.06 | jonpry | just an allocated piece of physical ram |
20:11.23 | d3tul3 | alrght, for some reason i thought high vectors had like a 1-1 identity mapping |
20:11.37 | jonpry | no |
20:13.00 | d3tul3 | ok, thanks |
20:13.58 | d3tul3 | i have to stop thinking that every hex number i see in an assembly file is a physical address :p |
20:14.36 | jonpry | the thing is |
20:14.56 | jonpry | in modern kernels that page is probably not readable or writable |
20:15.03 | jonpry | from userland |
20:15.17 | jonpry | so it causes a fault |
20:15.36 | jonpry | that is handled exactly like a syscall |
20:23.01 | jonpry | maybe that is only a problem with mainline kernels |
20:26.53 | d3tul3 | <PROTECTED> |
20:27.04 | d3tul3 | they have some sort of a time drift problem on the 3.0 kernel |
20:27.38 | d3tul3 | i can't imagine that implementation of gtod is that much more efficient than the syscall |
20:27.55 | Cotulla | yea it sucks |
20:32.23 | jonpry | go linaro |
20:33.22 | Cotulla | no thks |
20:33.25 | d3tul3 | no not you too |
20:34.03 | jonpry | i think linaro invented that patch |
20:34.59 | Cotulla | but it's completely stupid |
20:35.11 | Cotulla | RTC should be accessed only by demand without polling |
20:36.04 | jonpry | it isn't rtc time |
20:36.21 | jonpry | it updates that shit on every tick with wall clock time |
20:36.47 | jonpry | er |
20:36.50 | jonpry | update_vsyscall_old(&xt, &tk->wall_to_monotonic, tk->clock, tk->mult); |
20:37.09 | Cotulla | In Windows CE it was global |
20:40.59 | d3tul3 | every tick? there has to be better ways of doing that |
20:41.40 | jonpry | i'm not sure that is that bad |
20:41.51 | jonpry | some sort of clock gets updated every tick |
20:42.08 | jonpry | so 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) |
23:30.22 | jonpry | Brought up 2 CPUs |
23:30.34 | jonpry | but it's all a ruse |
23:30.45 | jonpry | there is something funny with ASID on different threads |
23:30.53 | jonpry | PC in coresys, SSR = 000A0000 |
23:31.07 | jonpry | if i set any bits in SSR[ASID] it will fault |
23:31.21 | jonpry | which is unrecoverable because EVB is not even reachable |
23:31.40 | jonpry | it's like global bit, is global only for thread 0 |
23:31.47 | jonpry | which can't possibly be right |
23:38.05 | jonpry | nvm |
23:38.13 | jonpry | its just impossible to debug |