irclog2html for #htc-linux on 20100419

07:19.40*** join/#htc-linux apt (ibot@rikers.org)
07:19.40*** topic/#htc-linux is Welcome to the http://htc-linux.org project! Please read the Wiki: http://htc-linux.org/wiki || Logs: http://irclog.netripper.com/?chan=htc-linux || This is no android support channel.
07:27.37*** join/#htc-linux NoMercy__ (~NoMercy@p4FDE6A03.dip.t-dialin.net)
07:27.40*** join/#htc-linux MethoS- (~clemens@134.102.106.250)
07:30.55*** join/#htc-linux NoMercy (~NoMercy@p4FDE773A.dip.t-dialin.net)
07:36.58*** join/#htc-linux MethoS- (~clemens@134.102.106.250)
07:40.41*** join/#htc-linux Tigerzen (~Tigerzen@81-66-207-36.rev.numericable.fr)
07:45.07*** join/#htc-linux NoMercy_ (~NoMercy@p4FDE7443.dip.t-dialin.net)
07:52.51*** join/#htc-linux Markinus (~Miranda@gtng-4db04fb9.pool.mediaWays.net)
07:56.05*** join/#htc-linux goxboxlive (~jrs@mail2.hjellnesconsult.no)
07:56.30*** join/#htc-linux tmzt_ (~ircuser@99-157-224-139.lightspeed.bcvloh.sbcglobal.net)
08:06.13*** join/#htc-linux DenPal (~hoper@post1.centr.by)
08:06.25*** part/#htc-linux DenPal (~hoper@post1.centr.by)
08:11.32*** join/#htc-linux kvaster (~kvaster@93.84.112.80)
08:19.42*** join/#htc-linux leviathan (~quassel@etx-public-dock-55-dhcp.ethz.ch)
08:29.14*** join/#htc-linux mickey|office (~Mickey@dialbs-092-079-168-007.static.arcor-ip.net)
08:54.08leviathanmickey|office: [ 1263.763610] wakeup wake lock: SMD_RPCCALL
08:54.22leviathanit seems to come from the battery information
08:54.27leviathanas soon as it is discharging
08:54.31leviathanit starts to flicker
09:18.08mickey|officeok, so we need to shut up that battery
09:20.43IceBoneI need to shut up about buying a desire and go out and buy one.
09:22.54*** join/#htc-linux leviathan__ (~quassel@guest-docking-nat-2-195.ethz.ch)
09:28.02*** join/#htc-linux Captnoord (~Captnoord@145.74.216.27)
09:36.14*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
09:42.34CaptnoordMarkinus: hmmm.... have you tried using the old fb_console from the .27 repo
09:42.48Captnoordalso remove the mdp hacks for the leo
09:42.50MarkinusCaptnoord: yes, exactly the same
09:43.10CaptnoordI still wonder what that timer thingy is
09:43.12MarkinusCaptnoord: but MDP is disabled or?
09:43.24CaptnoordI dono
09:43.25MarkinusCaptnoord: we didn't have a panel file or other MDP stuff.
09:43.34CaptnoordI wonder what is needed
09:43.46Captnoordit somehow feels like things get blocked
09:43.57Captnoordresulting in the timers going slowly
09:44.02MarkinusCaptnoord:  yes, it's very strange
09:44.20Captnoordthe entire .32 repo is done using async irq stuff
09:44.34Captnoordbut I still wonder why it works with the leo
09:44.41Captnoordand not with the 7k stuff
09:45.01MarkinusCaptnoord: yes . .. I don'T understand it too. . .
09:45.11MarkinusCaptnoord: on leo works both, htc fb and MDP FB
09:45.15Captnoordyea
09:45.24Captnoordbtw htc also has a fb console
09:45.40Captnoordits in the mdp directory
09:45.41CaptnoordI guess
09:45.45Captnoorddrivers/video
09:48.26Markinusthis is this one if we didn't use the htc fb console. But it could be that we need the MDP stuff, like panel file etc.
09:48.57CaptnoordI don't actualy know
09:49.01Captnoordwe might check the irc logs
09:49.05MarkinusCaptnoord: our one should work out of the box
09:49.09Captnoordif they mention these things
09:49.12Captnoordyup
09:49.57CaptnoordI really would like to get the uart working....
09:49.58Captnoordhmmm
09:50.12Captnoordor tomorrow
09:50.50Captnoordbesides if I compare the leo and the armv6 config file
09:50.56Captnoordthere are some major differences
09:51.11Captnoordmaybe its something small
09:52.27Captnoordlike we config the amms version
09:52.29Captnoordleo doesn't
09:53.40MarkinusCaptnoord: in the defconfig there isn'T a amss version, but after the kconfig run he activate the first one
09:53.42Markinus6225 or so
09:53.51Captnoordor the htc_fb_console isn't defines on our config file
09:53.55Captnoordwhile it is on the leo's
09:54.11MarkinusCaptnoord: he is defined in our config file
09:54.16Captnoordmmmm
09:54.22Captnoordbleh
09:56.04MarkinusCaptnoord: I will test to remove the whole mdp stuff from the Makefiles . .. so we can be sure this doesn't do problems
09:56.33Captnoordyea
09:56.40Captnoordwe should do that
09:56.47Captnoordturn everything off
09:56.58Captnoordso we can eliminate all the variables
09:57.22MarkinusCaptnoord: right. Much if the mdp stuff is fix in, without config defines
09:57.50*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
09:58.30Captnoordatm I don't have much clue why it doesn't work
09:58.49Captnoordbut I want to try something.....
09:58.52Captnoordconfig related
10:00.04CaptnoordHTC_FB_CONSOLE_BOOT isn't defines
10:00.08Captnoordcan you check that
10:00.44*** join/#htc-linux GNUtoo (~GNUtoo@host21-134-dynamic.25-79-r.retail.telecomitalia.it)
10:01.06MarkinusCaptnoord: the name is a bit wrong
10:01.13Markinusit means ONLY on boot
10:01.43Captnoordhmmmmm
10:01.48MarkinusCaptnoord: if you define it then it's only on boot
10:01.56Captnoordyea I figure...
10:02.10Captnoordits armv6 vs leo atm
10:02.25Captnoordand because we are simply sending the data to the fb
10:02.30Markinusyes, I think so too
10:02.31Captnoordits so low level
10:02.33Captnoordit should work
10:02.41Captnoordso there is something simple that makes it not work
10:03.56MarkinusCaptnoord: I removed MDP from Makefiles, same problem . . so something else is wrong
10:05.51Captnoordcheck FB_MODE_HELPERS
10:06.35MarkinusCONFIG_FB_MODE_HELPERS=y
10:06.58Captnoordon leo its off
10:07.31Markinusok, I test it
10:08.23Captnoordand FB_MSM_LCDL is on.... but I don't have a clue if that would have influence
10:09.31Captnoordhmmmm
10:09.38Captnoordtiming is correct now
10:09.45CaptnoordI didn't disable fb_console
10:10.00Markinusaptnoord: the FB MSM isn't used I think
10:10.10Captnoordpossible
10:11.18CaptnoordI disabled a whole bunch of debug settings
10:14.09CaptnoordI think leak detection is on
10:14.17Captnoordresulting on all mem related things beeing tracked
10:15.04CaptnoordI disabled CONFIG_DEBUG_SLAB_LEAK=y
10:15.35MarkinusCaptnoord: the ramconsole isn'T working too . .
10:15.39MarkinusCaptnoord: we see nothing
10:16.31Captnoordramconsole you need to set the mem offsets
10:16.36Captnoordleo has more mem than us
10:17.23MarkinusCaptnoord: I changed it already to our old values, but nothing . .
10:17.44MarkinusCaptnoord: we forgott something . . .but what?!?!
10:18.09Captnoordmaybe its related to how the board files are handled
10:18.13Captnoordyou have split them
10:18.17Captnoordinto a main board
10:18.19Captnoordand a module
10:18.21Captnoordor
10:18.23Captnoordsystem file
10:18.50Captnoordhmmm I should be studying now
10:18.54Captnoordso I won't test it
10:19.06Captnoordbut scrap the idea... for a test
10:19.12Captnoordand dump everything like it was
10:19.20Captnoordmake*
10:19.35CaptnoordI don't say your idea is wrong
10:19.46Captnoordbut we should first work towards something that works
10:19.54Captnoordand then make it pretty
10:21.53MarkinusCaptnoord: I tried it alrerady to, copied all thinks in one file . .same problem ..  :(
10:21.59Captnoordhmmmmm
10:22.00Captnoordbleh
10:22.16MarkinusCaptnoord: but you'Re right, we should do it easy now . . better for error handling
10:22.24Captnoordthe fb init is beeing called?
10:22.35Markinusyes
10:22.47Captnoordthe htc_fb_console_write is beeing called..... trough the printk's
10:23.20Captnoordnot that it will make any difference btw
10:23.23Captnoordas I tested that
10:24.44phh[10:55:56] <leviathan> mickey|office: [ 1263.763610] wakeup wake lock: SMD_RPCCALL <------------- *wakeup* wake lock, it means that the system won't be fully woken up until this task is finished
10:25.02phhand any way, AFAIK this task is called by the kernel itself, so it's not the reason it wakes up
10:25.08*** join/#htc-linux DenPal (~hoper@post1.centr.by)
10:25.20phhhum I can check that.
10:25.51phhah right. it's A9 who sends that.
10:25.54phhand I don't know when
10:27.19phhwell you can try disabling battery driver then I guess.
10:27.46phhCaptnoord: concerning battery, instead of using wince values, can't we calcaulte them by ourselves ?
10:28.05phhI mean, the model is quite simple, E=U+A*i+B*T
10:28.33phhwe can do many measures, around the same E, with various i/T values
10:28.48Captnoordsure we can
10:28.56CaptnoordI don't have the time todo the measurements
10:28.56phhit's just boring to do, right.
10:29.03Captnoordand....
10:29.09Captnoordonly thing really missing is a filter
10:29.16phhCaptnoord: *for rhod* ? :p
10:29.20Captnoordfor rhod
10:29.32CaptnoordI didn't had time to do the reversing
10:29.37Captnoordafter tomorrow I will
10:29.40Captnoordbit exam
10:29.44Captnoordregarding the exam
10:29.53Captnoordneed to get back to the books
10:29.56Captnoord1 test
10:30.20phhyou lucky
10:30.21phhI have
10:30.21phhhum
10:30.27phhI don't want to count how many tests.
10:30.42Captnoordnah
10:30.43Captnoordhave fun all
10:30.50Captnoordtalk to ya all tomorrow
10:30.54Captnoordor tonight
10:31.11MarkinusCaptnoord: have now all in one file
10:31.14MarkinusCaptnoord: same again
10:31.19MarkinusCaptnoord: bye
10:36.28*** join/#htc-linux M3DLG (~M3DLG@212.183.140.49)
10:38.01*** part/#htc-linux DenPal (~hoper@post1.centr.by)
10:38.03mickey|officephh: I'm quite desperate wrt. that suspend/resume. I even patched rtc_suspend to call msm_sleep_set_max_time with something really huge – still no dice :/
10:38.25phhI don't think that's the problem
10:38.32mickey|officeGNUtoo: leviathan: perhaps we need to either move to the latest upstream kernel or go back to where we came from and bisect
10:39.04mickey|officephh: could it be battery readings as leviathan indicated?
10:39.21phhmickey|office: I don't think that would make it wake up instantly
10:39.36mickey|officek
10:40.09phhand anyway, IIRC, the last log you gave me, showed that it doesn't even try to do the architecture sleep
10:40.43GNUtoomickey|office, I can try other kernels but maybe the problem is a bug in .32
10:40.55GNUtooI'll try that
10:40.59GNUtoobut later
10:41.04GNUtooI've things to do today
10:42.18mickey|officeok
10:42.48mickey|officephh: yeah, i have tried to follow the call stack, but it's very hard without jtag and just printk for someone like me
10:43.03mickey|officephh: who is supposed to call msm_sleep() after all?
10:43.09phhdon't know :p
10:43.13phhlet me check
10:43.14mickey|officehehe, k
10:44.23phhmickey|office: kernel/power/suspend.c
10:44.38phhpm.c calls suspend_set_ops() with his own struct including msm_pm_sleep
10:46.36GNUtoomickey|office, will function_trace work for you?
10:46.53mickey|officewell, not that i think i can provide much value anyways to the debugging
10:47.01mickey|officeit just strikes me as being worthless only watching you guys...
10:47.12mickey|officebut i must confess, my strengths are more in userland ;)
10:49.20*** join/#htc-linux alain2210 (~alain@AMontsouris-151-1-87-26.w90-2.abo.wanadoo.fr)
10:50.02alain2210mickey|office: hello
10:51.45alain2210i'm reading the research about suspend resume
10:52.13alain2210I'm testing it via ssh
10:52.18phh(there are so many frenchs here.)
10:52.29alain2210phh: yes
10:52.56alain2210this morning suspend cannot access msm_sleep
10:53.05alain2210after some times its works....
10:53.48alain2210may be it can access msm_sleep only when battery is charged full
10:54.20phhlol ?
10:54.30phhI'm not saying no, byt why ?
10:54.49alain2210I understand nothing
10:54.55alain2210I observ
10:55.06alain2210and I remeber the openezx case
11:01.02*** join/#htc-linux marex (~marex@vasut.kolej.mff.cuni.cz)
11:02.39alain2210mickey: how do you have a working keyboard on dream ? I only have letters (no alt)
11:05.59mickey|officealain2210: good morning. "working" in the sense of all keys emitting proper input nodes. I did not use X yet
11:06.09mickey|officei guess one needs a special xmodmap
11:08.07alain2210mickey: I have removed X
11:08.17alain2210so now I'm using console
11:08.39alain2210and I still not have all the keys
11:08.50mickey|officeah
11:09.04mickey|officehmm, ok i will take another look when i'm back home
11:09.17mickey|officeNOTE that key composition is not done on input layer
11:09.17alain2210I use an "h2200" image
11:09.19mickey|officebut on the layers above
11:10.32alain2210yes where ?
11:10.57mickey|officein X or whatever you use as display manager
11:11.35mickey|officedon't bother with teaching fbcon how to compose
11:11.42mickey|officefbcon is only temporary anyways
11:11.50mickey|officeit's already gone in my defconfig
11:19.40alain2210yes, but to be able to write echo mem > /sys/power/state without usb and ssh
11:19.48alain2210how do you do this ?
11:21.35mickey|officeyou cheat :)
11:21.37mickey|officelog in
11:21.38mickey|officecd ..
11:21.41mickey|officecd sys
11:21.42mickey|officecd power
11:21.45mickey|officeecho mem > state
11:21.46mickey|office:)
11:22.03mickey|office> is actually available
11:22.10mickey|officeshift-dot or so
11:22.14alain2210but > is alt M on my keyboard
11:22.38mickey|officehmm
11:22.46mickey|officeon low level console it should be shift-dot
11:22.52mickey|officejust try
11:23.13alain2210yes you're right
11:23.46mickey|office:)
11:29.25*** join/#htc-linux Orkspalter (~Orkspalte@p54B58D5B.dip0.t-ipconnect.de)
11:30.53alain2210mickey: did you succeed in powering it off ?
11:31.14GNUtooalain2210, for keyboard we didn't map it yet,for how to suspend via keyboard:
11:31.16GNUtoocd ..
11:31.17GNUtoocd ..
11:31.19GNUtooetc...
11:31.22GNUtoocd sys
11:31.24GNUtoocd power
11:31.27GNUtooecho mem > state
11:31.35alain2210GNUtoo : thanks
11:31.52alain2210I had not discovered >....
11:31.58alain2210no it is good.
11:32.23alain2210and you can shut the phone off ?
11:32.39GNUtooshift + . is > for me
11:32.50GNUtoo. is the key at right of space
11:33.49alain2210yes : i mean : now it is good. mickey has given the answer
11:34.31GNUtoook
11:34.43GNUtooah ok I should read logs
11:34.52GNUtooentirely
11:34.55GNUtoobefore responding
11:35.13GNUtooalain2210, maybe you could map the keyboard
11:35.22GNUtooask mickey|office on how to do that
11:35.34GNUtoothere are keymaps files in oe
11:35.56alain2210GNUtoo: mickey has said it was not useful
11:36.02GNUtooah why?
11:36.37alain2210because console is not used when in X
11:36.43GNUtooyes
11:36.50GNUtooI was more talking about X
11:37.02GNUtoobecause in console or in X it uses the default mapping
11:37.03alain2210ok
11:37.20GNUtooso if I got keyboard mapping in X it would be great
11:37.33alain2210yes
11:37.41alain2210(after suspend.....)
11:38.00mickey|office*nod*
11:38.02GNUtoosuspend is top priority for me,that's why I didn't do X mapping yet
11:38.04mickey|officesuspend/resume is mission critical now
11:38.10alain2210yes
11:38.14mickey|officeit's awful we're hanging already for so long on it
11:38.21GNUtooalain2210, are you good enough to help us with suspend?
11:38.29alain2210sorry no
11:38.35GNUtoook
11:38.36mickey|officeGPS will be very simple once we found out how to toggle power control
11:38.44mickey|office(which itself might be hard)
11:38.46GNUtooso that's why I asked for keyboard mapping
11:38.46alain2210I have worked on that for ezx
11:39.02GNUtooon what?
11:39.12GNUtoosuspend? keyboard? gps?
11:39.28alain2210but with drwyrm
11:39.34alain2210suspend
11:39.48GNUtoowow
11:40.19alain2210yes drwyrm is understanding the kernel very well
11:41.19mickey|officebbl, lunch
11:42.24GNUtoobbl too
11:43.09alain2210what means bbl
11:45.20*** join/#htc-linux GNUtoo|oeee (~GNUtoo@host21-134-dynamic.25-79-r.retail.telecomitalia.it)
11:45.47phhbe back later
11:46.18*** part/#htc-linux Orkspalter (~Orkspalte@p54B58D5B.dip0.t-ipconnect.de)
11:46.54*** join/#htc-linux jason10 (~jason@r3n197.net.upc.cz)
11:47.04jason10Hello
11:47.28alain2210phh : thanks
11:47.45jason10has anybody managed to get tiwlan running in ettercap on G1 (Openmoko/debian or android)?
11:48.54phhjason10: what about wl1251 ?
11:49.14jason10sorry for my noobishness
11:49.41phhjason10: a driver for the same chipset, which is included in upstream kernel
11:49.51phhmainstream even
11:49.55jason10nice will read about it quickly
11:49.59jason10thanks
11:52.00jason10phh, does this mean things like ettercap can run on android within a debian image? I wander if arp poisoning will work on G1
11:52.10phhI don't know.
11:52.20phhbut it's a standard linux driver, using nl80211 stack and such
11:52.24phhso it has more chances to work.
11:52.34jason10ok thanks for the direction
11:56.44*** join/#htc-linux mickey|zzZZzz (~mickey@80.81.242.146)
11:57.29*** join/#htc-linux Modu (~Modu@hubble.bibibox.fr)
11:57.34phhIceBone: thanks.
11:57.35Moduhi
11:57.55phhModu: which bootloader ?
11:58.44ModuI'd like to known how bootloader works, to know what part lead to a bricked device
11:59.02jason10sounds dodgy :P
11:59.23phhModu: just the part when you write over the bootloader NAND
11:59.49Moduis it the SPL / recovery image / boot image ?
12:00.01ModuIn my mind only the SPL is critical
12:00.06phhyup
12:00.19phhModu: you have a tri-color SPL screen too ?
12:00.34ModuYeah, an htc magic
12:01.05ModuSo even if I destroy the recovery image and the boot device, I should be able to repair it with fastboot ?
12:01.05phhyeah I don't know htc android devices
12:01.12phhah I don't know what fastboot is
12:01.18phhif it's from recovery or from spl
12:01.48phhrecovery is a linux partition ?
12:01.53Moduok, any suggestion where I can ask this question
12:02.02jason10xda forum
12:02.09jason10forsure Modu
12:02.23Moduis there any IRC chan for xda ?
12:02.26phh#xda-devs
12:02.30Moduthanks
12:03.17Modugood bye
12:03.19*** part/#htc-linux Modu (~Modu@hubble.bibibox.fr)
12:23.35alain2210phh: now if I do mem>/sys/power/state I seem to get a suspend (black console)
12:23.38alain2210http://pastebin.ca/1867719
12:24.24phhalain2210: no idle 0 msm_sleep.
12:24.29phh=====> didn't sleep.
12:26.06alain2210but
12:26.33alain2210[ 2898.456756] msm_sleep(): vector ea00000e 35392e30 -> e51ff004 10030a28
12:26.40alain2210[ 2898.456756] msm_pm_collapse(): returned 1
12:26.46phhalain2210: that's *idle* sleep
12:27.01alain2210ok
12:27.21alain2210idle sleep is not always working
12:27.28phhit depends on the cpu activity
12:27.38alain2210thanks
12:27.41phhit needs to be idling for more than 20ms iirc
12:28.01alain2210yes I read this value
12:28.37alain2210but why it need this vector line for idle ?
12:28.54phhidling, from a CPU point of view, is exactly a sleep
12:29.12phhit just wakes up on any interrupt, so it doesn't change its behaviour at all
12:29.37phhI mean compared to if it was not sleeping
12:30.04alain2210ok
12:40.13*** join/#htc-linux GNUtoo (~GNUtoo@host21-134-dynamic.25-79-r.retail.telecomitalia.it)
12:42.29*** join/#htc-linux leviathan (~quassel@c-82-192-226-27.customer.ggaweb.ch)
12:47.56*** join/#htc-linux gauner1986 (~gauner198@p5B38668C.dip.t-dialin.net)
12:52.17*** join/#htc-linux br1ck (~br1ck@xdslfd057.osnanet.de)
12:52.59*** join/#htc-linux Benouch (~44ecb89c@gateway/web/freenode/x-bbhqqfefsjmrhodn)
12:59.41*** join/#htc-linux GNUtoo (~GNUtoo@host21-134-dynamic.25-79-r.retail.telecomitalia.it)
13:00.18*** join/#htc-linux Markinus (~Miranda@gtng-4db04fb9.pool.mediaWays.net)
13:03.31*** join/#htc-linux hfctsaot (~heroc@108.108.179.229)
13:03.31*** join/#htc-linux Orkspalter_ (~Orkspalte@p54B58D5B.dip0.t-ipconnect.de)
13:03.52hfctsaotphh u around
13:09.19*** join/#htc-linux Guest86671 (~yadayada@pool-72-84-153-242.slsbmd.east.verizon.net)
13:20.01*** join/#htc-linux Vito89 (~quassel@gw.loccal.net)
14:00.45*** part/#htc-linux Fordyy (~MrFordihu@d90-130-14-1.cust.tele2.se)
14:01.26phhToAsTcfh: yes.
14:03.12gauner1986mhm.. i find studying for the lpi 102 easier than the 101.. :)
14:12.30*** part/#htc-linux Orkspalter (~Orkspalte@p54B58D5B.dip0.t-ipconnect.de)
14:22.53leviathanGNUtoo: ok
14:22.57leviathanfound out the point
14:23.03leviathanfollowing things: one
14:23.14leviathanit seems to be something with the battery driver
14:23.24leviathanthe reason for the wakeups are rpc calls
14:23.35leviathanvery likely from battery usage info
14:24.12leviathantwo points: it stops flickering if you have mplayer running while suspend
14:24.16leviathanthen it really suspends
14:25.43leviathanlol
14:25.50leviathanGNUtoo: its not mplayer
14:25.54leviathanits the usb port itself
14:26.00leviathanheadset, usbgadget
14:26.02leviathan...
14:26.53leviathanthats a really strange bug
14:26.54leviathan...
14:29.28phh[16:25:15] <leviathan> the reason for the wakeups are rpc calls <<<<-------- where do you get that ?!?
14:29.38leviathanphh: dmesg
14:29.56phhleviathan: more precisely ?
14:30.03leviathanmom
14:30.10leviathanI'll bootup most recent kernel
14:30.15leviathanand pastebin the lines
14:32.50GNUtooleviathan, are you shure it really suspend?
14:33.00leviathanyes
14:33.20GNUtoodmesg?
14:38.32*** join/#htc-linux Jaferdark (~Jafterdar@97-87-4-243.dhcp.mdsn.wi.charter.com)
14:42.59leviathanwakeup wake lock: SMD_RPCCALL
14:43.03leviathanuuhm
14:43.04*** join/#htc-linux x-fak (~WinGuru@reactos/tester/x-fak)
14:43.16leviathanonly if its not connected
14:43.19leviathanvery strange
15:02.12GNUtooI'll go soon
15:24.32*** join/#htc-linux majster (~majster@cpc12-aztw25-2-0-cust68.aztw.cable.virginmedia.com)
15:26.42*** join/#htc-linux [acl] (~abel@96.246.167.90)
15:28.19*** join/#htc-linux ToAsTcfh (~toastcfh@24.94.158.107)
15:28.44[acl]phh: ?
15:32.02*** join/#htc-linux luminoso (~lumos@av-217-129-131-108.netvisao.pt)
15:32.39*** join/#htc-linux NoMercy (~NoMercy@p4FDE75A0.dip.t-dialin.net)
15:50.35*** join/#htc-linux toi (~toi@195.177.247.4)
16:00.38*** join/#htc-linux kiozen (~oeichler@rgnb-5d87d48a.pool.mediaWays.net)
16:26.20*** join/#htc-linux GNUtoo|oeee (~GNUtoo@host21-134-dynamic.25-79-r.retail.telecomitalia.it)
16:28.09GNUtoo|oeeehi
16:28.18GNUtoo|oeeeany news about suspend?
16:34.26alain2210GNUtoo|oeee : does suspend ever work ?
16:34.43alain2210I have tried the android os and
16:35.02alain2210echo mem > /sys/power/state does not work
16:36.09phhalain2210: with which kernel ?
16:36.20phhthe problem here is clearly the kernel, not the userland
16:36.35alain2210using the official kernel of android 1.0
16:36.45phhah.
16:36.56phhoh yes they had weird stuff
16:37.03phhit's) not /sys/power/state, it's /sys/power_android/something
16:37.29alain2210but there is also a /sys/power/state
16:38.26phhyes but no.
16:38.31phhtry with /sys/android_power/request_state
16:38.48*** join/#htc-linux Guest86671 (~yadayada@pool-71-255-177-209.slsbmd.east.verizon.net)
16:44.28alain2210phh: but cat /sys/android_power/request_state : wake, no mem
16:47.37chamonixhey guys
16:51.03*** join/#htc-linux Cass (~Cass@cass.demon.co.uk)
16:51.55GNUtoo|oeeealain2210, in cm-mod 2.6.29 it works
16:52.39GNUtoo|oeeestrangely it say something just after suspending console
16:52.42GNUtoo|oeeecould it be VT?
16:53.06GNUtoo|oeeephh, that's with very ld kernel
16:53.15GNUtoo|oeee*old kenrel
16:53.27GNUtoo|oeeewith cm-mod .29 I think it's /sys/power/state
16:53.40*** join/#htc-linux luminoso (~lumos@av-217-129-131-108.netvisao.pt)
16:54.34alain2210GNUtoo|oeee: what is exactly cm-mod ? kernel ? rootfs ?
16:54.44GNUtoo|oeeecyanogen mod
16:54.49alain2210and what command have you used ?
16:55.01GNUtoo|oeeeecho mem > /sys/power/state
16:55.33GNUtoo|oeeeI think the replicant kernel also suspends but I bet it's less recent than cm-mod
16:55.40GNUtoo|oeeeand also it was pre-built
16:56.21alain2210could you precise ?
16:56.54*** join/#htc-linux swc|666 (~infidel20@unaffiliated/swc666/x-4934821)
16:56.58*** join/#htc-linux mickey|zzZZzz (~mickey@openmoko/coreteam/mickey)
16:58.51GNUtoo|oeeeI'll paste a link
16:59.01GNUtoo|oeeelet me power on my desktop computer
17:01.36*** join/#htc-linux GNUtoo (~GNUtoo@host21-134-dynamic.25-79-r.retail.telecomitalia.it)
17:03.05alain2210GNUtoo|oeee: another thing, I have tested with pm_debug core.
17:03.30alain2210and I does not wait the 5 seconds
17:03.37alain2210s/I/it/
17:06.49*** join/#htc-linux GlemSom (~glemsom@0x5da34bca.cpe.ge-1-1-0-1105.sdnqu1.customer.tele.dk)
17:07.09*** join/#htc-linux diiishe-away (~mmm@96.56.222.18)
17:07.36*** join/#htc-linux kvaster (~kvaster@live.bn.by)
17:07.41GNUtooalain2210, http://github.com/cyanogen/cm-kernel
17:07.59*** join/#htc-linux Kawaiius (~yadayada@pool-72-84-156-62.slsbmd.east.verizon.net)
17:10.26alain2210sorry, but with what rootfs ?
17:10.53alain2210and you have said that there was a prebuilt version ?
17:12.02*** join/#htc-linux marex (~marex@vasut.kolej.mff.cuni.cz)
17:12.54GNUtooalain2210, http://www.cyanogenmod.com/
17:13.07GNUtooalain2210, beware it contain proprietary software
17:13.23GNUtoolike proprietary lower level android libs + google sruff
17:13.36GNUtoolike the ril, the libcamera,the libgps etc...
17:13.41*** join/#htc-linux gauner1986 (~gauner198@p5B38668C.dip.t-dialin.net)
17:14.05GNUtooI used that rootfs only for testing and for dumps
17:14.20GNUtoos/testing/testing suspend/
17:14.31GNUtooelse I used replicant
17:14.35GNUtooanother project
17:14.52alain2210yes it have read the site of replicant project
17:14.53GNUtoohttp://trac.osuosl.org/trac/replicant/wiki
17:15.26alain2210but for cyanogenmod.com you use it on a sd card ?
17:15.29alain2210in nand ?
17:16.11*** join/#htc-linux hfctsaot (~heroc@108.108.179.229)
17:17.34GNUtoonand
17:18.04GNUtoobasically I backuped my phone with cm-recovery-1.4 or something like this...let me find the real filename
17:18.15GNUtoothen flashed an old cm-mod
17:18.19GNUtoothen upgraded it
17:18.35GNUtoo(because I didn't have proprietary libs in replicant)
17:18.44alain2210I also have a cm-recovery
17:18.44GNUtooand then saved it
17:18.46GNUtoook
17:21.17phh[18:54:58] <GNUtoo|oeee> phh, that's with very ld kernel <------ he said android 1.0;
17:24.18chamonix#@!#?!!
17:24.53hfctsaotphh did u commit that oc patch somewhere
17:25.47*** join/#htc-linux MethoS- (~clemens@134.102.106.250)
17:27.34*** join/#htc-linux diiishe-away (~mmm@96.56.222.18)
17:28.10chamonix\o/
17:29.36chamonixwho decided 10 set 192.168.20.x to usb0?
17:29.43chamonixs/10/to
17:30.38chamonixphh: u wanna laugh? guess what my router/dns' IP is?
17:32.03GNUtoochamonix, 127.0.0.1?
17:32.07chamonixrofl
17:32.10chamonixalmost
17:32.13chamonix192.168.20.1
17:32.14chamonix:-D
17:32.18GNUtoook
17:32.27GNUtooah I didn't read what's up
17:32.41chamonixand I've been wondering y WLAN netowrking didn't work
17:32.59GNUtooinit 0 && mount the card && nano /media/sdcard/etc/network/interface
17:33.08chamonixobviously usb0 wasn't doing any name resolution
17:33.28chamonixwell, I fixed ini instead
17:33.36chamonixt
17:35.52phhchamonix: .
17:36.15chamonix.?
17:36.18chamonixlol
17:36.30phhit's to say that there is nothing to say :p
17:36.38chamonixthat's how I felt >.<
17:36.47chamonix:P
17:38.39*** join/#htc-linux Markinus (~Miranda@gtng-4db04fb9.pool.mediaWays.net)
17:38.47chamonixphh: I was not able to prove it because I used my phone but I have to try loading my phone using AC power instead of USB and repeat some test..... I have the feeling that at some point in time it stops charging from usb and pretends battery is full even if it's not (locat report BATTERY_STATE_UNKNOWN)
17:39.19phhchamonix: oO
17:39.47chamonixI saw this when running ledeffects because I registered battery broadcasts
17:40.12chamonixwhile charging I get D/WifiService( 1209): ACTION_BATTERY_CHANGED pluggedType: 1
17:40.28chamonixbut after a while it goes to unknown
17:40.41chamonixwhile = when the meter says it's full
17:40.53phhhum
17:41.05chamonixright now I have Status is BATTERY_STATUS_CHARGING
17:45.02chamonixsome1 any seen this?
17:45.03chamonixI/HTC Acoustic( 1920): libhtc_acoustic.so version 1.0.1.1.
17:45.07chamonixE/HTC Acoustic( 1920): can't open /dev/htc-acoustic -1
17:45.10chamonixE/AudioHardwareMSM72XX( 1920): Could not set acoustic parameters to share memory: -1
17:45.35phhyup that's normal
17:45.44chamonixlol
17:45.45chamonixok
17:46.26chamonixhow often do these threads die and restart?
17:46.33phhin a new thread ?
17:46.36phhhum that's more annoying
17:46.49chamonixI assume
17:46.54chamonixW/AudioSystem( 1209): AudioFlinger server died
17:46.59chamonixI/AudioFlinger( 1920): AudioFlinger's thread 0xbad8 ready to run
17:47.02phhreally annoying.
17:49.04*** join/#htc-linux Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
17:50.32chamonixI can confirm, earlier in logcat same happened and it had pid 1701
17:52.46GNUtooleviathan, can I have your wireless-compat recipe again?
17:53.04*** join/#htc-linux JairunCaloth (~dbrowne@c-24-98-17-158.hsd1.ga.comcast.net)
17:54.12Captnoordphh if you check the docs....
17:54.20Captnoordyou find some stuff about rf
17:54.24phhCaptnoord: ?
17:54.25Captnoordon the raph
17:54.28Captnoordand diamond
17:54.33Captnoordwe have 2 rf amplifiers
17:54.41Captnoordthey never get turned off
17:54.41Captnoordor
17:54.43Captnoordset to idle
17:54.50Captnoordthey seem to be set trough vreg's
17:54.56phhtwo ?
17:54.57phhwhy two ?
17:54.59Captnoordyea
17:55.08Captnoordgsm and umts
17:55.20Captnoorddiff freq spectrum
17:55.21Captnoordbla bla
17:55.22Captnoordlong story
17:55.57phhthere 4 gsm bands and 3 umts bands -_-'
17:56.20phhor 4 umts ?
17:56.44Captnoordphh pm
17:56.52phhyes yes i've seen
18:08.49chamonixphh: now I get D/BatteryBroadcastHandler( 1511): Status is BATTERY_STATUS_UNKNOWN
18:10.15chamonixif now I D/Exec.exec( 2044): Executing command echo 1 > /dbgfs/htc_battery/charger_state
18:10.21chamonixD/BatteryBroadcastHandler( 1511): Status is BATTERY_STATUS_CHARGING
18:10.34chamonix<PROTECTED>
18:15.01*** join/#htc-linux zachy (~george@85.193.4.18)
18:19.41chamonixthat may be bad implementation in userland but normally I'm expecting BATTERY_STATUS_FULL
18:20.25*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
18:22.39GNUtoohi, some people have wifi on other phones than dream,How can PSM(power saving mode) be activated?
18:24.23mickeyliwconfig <iface> power <params>
18:25.03mickeylcan also tweak txpower
18:27.20chamonixthere are also a few power params in tiwlan.ini
18:27.55mickeylyou mean the calibration file?
18:28.01mickeyl(wl1251-nvs.bin here)
18:28.25chamonixno, ./init.etc/wifi/tiwlan.ini in rootfs.img
18:28.43mickeylok, android
18:28.43chamonixdefaultPowerLevel = 2           # 0 - ELP # 1 - PD # 2 - AWAKE
18:28.44chamonixPowerSavePowerLevel = 2         # 0 - ELP # 1 - PD # 2 - AWAKE
18:29.19leviathanGNUtoo: sure you can have the recipe
18:29.37GNUtoook
18:29.38GNUtoothanks
18:29.42leviathanGNUtoo: http://ftp.o2s.ch/compat-wireless
18:29.44leviathanlook here
18:29.46leviathan^^
18:30.01GNUtoothanks
18:30.32GNUtoomickeyl, no I mean PSM
18:31.22GNUtoowhat param should I put
18:31.26GNUtoobecause I've an error
18:31.55mickeylaccording to the docs, iwconfig wlan0 power {period N|timeout N|saving N|off}
18:31.59GNUtoo$ iwconfig wlan0 power 200ms Error for wireless request "Set Power Management" (8B2C) :   SET failed on device wlan0 ; Invalid argument.
18:32.17mickeylperiod 200 ?
18:32.31mickeylor timeout 200, depending on what you want
18:32.44GNUtoo<PROTECTED>
18:32.46GNUtoook
18:32.48mickeyl200u
18:33.18leviathanGNUtoo: suspend works, when I have the headset connected ^_^
18:33.43GNUtooleviathan, realy? dmesg?
18:33.46leviathanuhm
18:33.52leviathanbetter: whats not in dmesg
18:33.52GNUtoomickeyl, still no go
18:33.58leviathanit suspends smoothly
18:34.10leviathanlike if it would be connected to usb-ether
18:34.15GNUtoodid phh confirm?
18:34.22leviathanask him
18:34.23leviathan:)
18:34.29leviathanphh: do you confirm?
18:34.35leviathanbtw
18:34.38leviathanafk.smoking
18:34.47GNUtoobecause he's the expert,not me
18:39.21*** join/#htc-linux majster (~majster@cpc12-aztw25-2-0-cust68.aztw.cable.virginmedia.com)
18:39.37leviathanre
18:41.55*** join/#htc-linux leaigor (~laigor@188.134.16.241)
18:45.33mickeylwrt. suspend
18:45.39mickeylhere's the ultimate proof you have to fulfill
18:45.40mickeylhttp://pastebin.com/QjKC4rVX
18:45.41mickeyl:D
18:46.00*** join/#htc-linux toi (~toi@d54C2AA76.access.telenet.be)
18:46.34GNUtoolol
18:46.42GNUtoomickeyl, the driver has PSM
18:46.48GNUtoobut no luck with iwconfig
18:47.02mickeylah, pity, no experience with the new API here
18:47.03GNUtoowl1251 I mean
18:47.12GNUtoolibnl doesn't compile btw
18:47.42GNUtooso no iw
18:47.43mickeyllibnl can be tough, i have an own snapshot for libnl2
18:47.56GNUtoolibnl2 compiles?
18:48.03mickeylmy snapshot, yes.
18:48.07GNUtoook
18:48.10GNUtooand what about iw
18:48.11GNUtooit works?
18:48.29mickeylon our wifi?
18:48.41GNUtoono generally
18:48.43GNUtooit compiles
18:48.45GNUtooand run
18:48.54mickeylsure
18:48.56GNUtooon our wifi it would be great to try it
18:48.57GNUtoook
18:48.57mickeylit's part of fso2-demo-image
18:49.01mickeylso it has to built
18:49.02GNUtoook
18:49.26GNUtoohow do I build it? change the srcrev and that's it?
18:49.35GNUtooI'm under oe.dev + shr distro
18:49.44mickeylbitbake wireless-tools should do it
18:49.46mickeylat least it does here
18:49.54mickeylunless they broke it since last week when i last tried
18:50.00GNUtoowireless-tools? isn't that iwconfig
18:50.01GNUtoo?
18:50.03mickeylyes
18:50.08mickeylthat's what we talking about, no?
18:50.09GNUtooiw != iwconfig
18:50.12mickeyloh
18:50.12GNUtoono
18:50.15mickeylwhat's iw?
18:50.17GNUtooiw is a new tool
18:50.21GNUtooit uses libnl
18:50.22mickeylcool, never seen that, sorry
18:50.37mickeylthought you were abbreviating iwconfig :)
18:51.07phh[20:36:23] <leviathan> phh: do you confirm? <----- confirm what ?
18:51.34leviathanphh: that it suspends well with headset connected ;-)
18:51.37GNUtoohttp://www.linuxwireless.org/en/users/Documentation/iw
18:51.56phhleviathan: hum, I'd say there is a wakelock on H2W ?
18:52.02phhbut without the dmesg I can't say anything.
18:52.18leviathanhmm
18:52.21mickeylyeah, there's wakelock on usb an h2w
18:52.27leviathanhmm
18:52.29mickeylwith these it doesn't actually suspend
18:52.30leviathanperhaö
18:52.34mickeylunless you patched the kernel
18:52.38leviathanohh
18:52.40leviathanokee
18:52.45leviathanthat might it be
18:52.48mickeylcheck the pastebin for how a real suspend looks like
18:52.49leviathanok in this case
18:52.51leviathanI admit
18:52.58leviathansuspending is officially broken
18:53.00leviathanfor me too
18:53.01mickeylaaah
18:53.02mickeylfinally
18:53.03mickeyl*phew*
18:53.06mickeylthat took us a while :D
18:53.07phh:p
18:53.17phhhttp://pastebin.com/QjKC4rVX
18:53.17leviathanIm soooo happy >_<
18:53.18GNUtooindeed
18:53.21phhmickeyl: that won't make me belive it.
18:53.31phhwell, if TIMESTAMP2>>>>>>> TIMESTAMP1
18:53.34phhpossibly.
18:53.45mickeyl:)
18:54.10mickeylone hour of sleep would be something i'd appreciate
18:54.17mickeylif more, be my guest ;)
18:55.21*** join/#htc-linux Markinus (~Markinus@gtng-4db04fb9.pool.mediaWays.net)
18:55.51phhI can get > 24hours sleep ! :p
18:57.17mickeylyeah
18:57.24mickeyli envy your kernel and your hardware
18:57.26mickeylthat's for sure
18:57.59phhlol ?
18:58.02phhyou can use 2.6.27 too
18:59.25mickeylhmm, i could
18:59.32mickeyland perhaps i need to if we can't solve this suspend/resume
18:59.42mickeylbut i don't feel like backporting devtmpfs
18:59.47GNUtoomickeyl, 29 works with suspend/resume
18:59.58mickeyl.29?
19:00.02GNUtoo2.6.29
19:00.06GNUtoofrom cyanogen mod
19:00.19GNUtoo(used only for testing suspend and dumps)
19:00.20mickeylinteresting, so it looks like upstream broke it
19:00.31mickeylperhaps it's fixed by now? ever tried master?
19:00.42GNUtoomickeyl, there are bugreport abuot issues with timer and 2.6.32 under kenrel.org bugzilla
19:02.52CaptnoordMarkinus: awake?
19:02.56mickeylgot a link?
19:03.06GNUtooyes let me look
19:03.59MarkinusCaptnoord: yes
19:04.14Captnoordgot any other idea's
19:04.19mickeylalthough as phh shows us msm_sleep is not being called at all
19:04.23CaptnoordI think the entire console stuff is never been called
19:04.24Captnoord:S
19:04.24mickeylso i doubt whether it's a timer issue
19:04.34gauner1986Markinus: gimme some good news :)
19:04.45GNUtoomickeyl,  https://bugzilla.kernel.org/show_bug.cgi?id=14424
19:04.47MarkinusCaptnoord: ?!? why?
19:04.55GNUtooah ok
19:04.56Captnoordfb_console clear
19:04.59CaptnoordI made it clear white
19:05.02Captnoordshould work
19:05.08Captnoordand its beeing called from the init
19:05.12Captnoordbut not white
19:05.21Captnoordso in my eyes it doesn't work
19:05.25MarkinusCaptnoord: hmm, but why on leo? . . .
19:05.32CaptnoordI don't have a clue
19:05.33Captnoord:S
19:05.34MarkinusCaptnoord: we have same init
19:05.37GNUtoomickeyl, I'll look and trace again
19:05.39CaptnoordI know
19:06.45GNUtoobut not at once
19:07.40CaptnoordMarkinus: ram console also doesn't work
19:07.42MarkinusCaptnoord: a intresting thing is, that if we replace the clock-wince.o with clock.o we didn't have vibration, but this should be independent in this booting phase
19:07.44Captnoordeven if we set it on early
19:07.55MarkinusCaptnoord: yes, same for me
19:08.03Captnoordyup
19:08.11Captnoordhmmmm
19:08.12Captnoordlol
19:08.55MarkinusCaptnoord: I made a very small defconfig, disabled all things, the kernel is 1.8 MB . . but same
19:09.01Captnoordyea
19:09.07CaptnoordI once had the timing correct
19:09.11Captnoordbut didn't save the config
19:09.12Captnoordbleh
19:09.48*** join/#htc-linux ToAsTcfh (~toastcfh@24.94.158.107)
19:10.16Captnoordand the default commandline we use
19:10.19Captnoordsets a console
19:10.22Captnoordconsole name
19:10.25Captnoordwhich we don't use
19:10.29Captnoordas the fb_console
19:10.31Captnoordis named
19:10.33Captnoordfb_con
19:10.35Captnoordor something
19:10.39CaptnoordI dono how that works
19:13.41*** join/#htc-linux hfctsaot (~heroc@108.108.179.229)
19:15.40GNUtoomickeyl, how does msm_sleep looks in dmesg ?
19:16.33mickeylif you enable the pm debug then like that:
19:17.32mickeylmsm_sleep(): mode %d delay %u idle %
19:17.33GNUtooah ok
19:17.33hfctsaotphh; the oc has been reworked ill commit it tonight to my tree. Idk if I've commit it anywhere or not but it will end all the "528mhz" questions
19:17.33GNUtooI saw that already,but idle % was wrong number
19:18.57phhhfctsaot: ok
19:19.13hfctsaotphh have u guys attempted to undervolt 528 by oc pll1
19:19.23phhhfctsaot: undervolt ?
19:19.37phhyou don't need to be on pll1 to undervolt you know ?
19:19.41hfctsaotuse pll1 voltage
19:19.49phhpll are just clocks
19:19.52phhnot power supply
19:19.57phhhfctsaot: just change the VDD_X value
19:20.02phhto undervolt
19:20.05hfctsaotok
19:20.06phhand yes we have undervoltage patches
19:20.09phhand it does nothing.
19:20.15hfctsaot:/
19:20.21MarkinusCaptnoord: I mean I used a clean cmd line for booting . . but I cannot test in the Moment. Had a accident and cannot sit long on desk . . . .
19:20.28Captnoordk
19:20.37CaptnoordMarkinus: its okey
19:20.38Captnoordrelax
19:20.40Captnoorddo your thing
19:20.40phhhfctsaot: rosie looks good, but it's buggy.
19:20.50hfctsaotyeah
19:20.58hfctsaotmy Rom?
19:21.02phhhfctsaot: don't know
19:21.06hfctsaotoh
19:21.08phhI go back to aosp.
19:21.19hfctsaottry porting mine
19:21.28hfctsaotits a release
19:21.33hfctsaot2.1
19:22.08Captnoordhmmm
19:22.17Captnoordphh could we config the usb to be a uart?
19:22.28hfctsaotalso the weather wallpaper crappy u may need to remove. tends to slow some devices down
19:22.30Captnoordso if we plug it in
19:22.31phhCaptnoord: theorically yes
19:22.35Captnoordhmmmm
19:22.44Captnoordhaven't seen any implementations....
19:22.44Captnoordor
19:22.57Captnoorddidn't know where to search
19:23.04phhbtw, did you test the last kernel ?
19:23.08phhif it killed/saved usb ?
19:23.17Captnoordnope didn't test it
19:23.19CaptnoordI will in a moment
19:25.35*** join/#htc-linux bartman (~bart@tau.jukie.net)
19:31.43mickeylanyone knowing which device node implements the AKM8976 ioctls?
19:33.51phh<PROTECTED>
19:34.30phhwait what ?
19:34.33phhyour kernel has a real driver
19:34.48phhI mean a /dev/input one
19:35.16phhI didn't know this exist
19:35.51mickeyl<PROTECTED>
19:36.08mickeylhave to identify the akmd device
19:36.09mickeylnode
19:36.17phh/dev/akmxxxx_dem
19:36.45phh<PROTECTED>
19:36.46*** join/#htc-linux NoMercy (~NoMercy@p4FDE75A0.dip.t-dialin.net)
19:36.47phhaccording to sourcecode.
19:37.02mickeylaah
19:37.03mickeylof course
19:37.08mickeylthanks
19:37.29mickeyldidn't occur to me to look under /dev, searched everything below /sys and /dev/input
19:37.34phhlol
19:38.18mickeylyep, that works
19:42.36phhCaptnoord: I didn't kill usb at least
19:42.48*** join/#htc-linux balans (~BRPOLDB@82-170-215-5.ip.telfort.nl)
19:42.55phh199900 of total memory available from linux.
19:43.10phhI need to find 100kbytes  somewhere.
19:45.10Captnoordre
19:45.13Captnoordjust wanted to test
19:45.14Captnoordgood
19:45.17Captnoordworks?
19:45.23Captnoordbtw
19:45.37Captnoordregarding usb not working on kitchen bootloader
19:45.38phhwell I'm now rebooting without usb to see.
19:45.39Captnoordtinboot
19:45.42Captnoordand stuff
19:45.53Captnoordthe android bootloader has nice stuff regarding it
19:46.37*** join/#htc-linux [acl] (~abel@96.246.167.90)
19:47.04phh[acl]:
19:47.21phhCaptnoord: ok, my usb changes (both init phy and clock) didn't change anything.
19:47.39phh[acl]: what did you do then ?
19:47.51Captnoordphh so they don't work
19:47.53[acl]phh: what you said. Implement those missing functions
19:48.01phh[acl]: ok
19:48.26phh[acl]: just adding functions ? then we can reuse this libhardware_legacy for AOSP stuff too ?
19:49.03[acl]phh: we are testing that now. it should work fine
19:49.14phhwe ? with mrpippy ?
19:49.25[acl]no with klinux
19:49.32phhand the rootfs provided by klinux doesn't work for me
19:50.15[acl]i havent tested his copy yet. I'll do that later today. Im still at work
19:50.37phhok
19:50.41[acl]i had an issue where the wifi tethering app would crash during install this morning so i just gave up.
19:51.04[acl]but there are other who reported success, so im confused myself.
19:51.29*** join/#htc-linux swc|666 (~infidel20@unaffiliated/swc666/x-4934821)
19:51.37phhanyway I hadn't tested it for wifi :p
19:51.44phhit's quite nice, but it's still far from usable
19:52.00phhI added 17MB of memory in last kernel commit, but I don't think that's going to help.
19:52.21[acl]phh: lets hope it does.
19:53.12[acl]Also i didnt know anything about those functions that were being called, so i'm just returning 0s for now. Eventually i may need to write a wrapper to the original hardware_legacy.so
19:53.31phhok
19:53.43phhI think we need to make the wifi init stuff more proper
19:53.58phhlike all driver-specific in external shell script
19:54.06phhso that there is a single libhardware_legacy.so
19:54.10phhand we need a git tree for it.
19:54.13[acl]definitely. also is it worth to post the source? or do you just want the binary?
19:54.22[acl]ok .. so source
19:55.14[acl]Anyways i gotta go. Thanks again for all the help man. later
19:55.49*** join/#htc-linux NoMercy_ (~NoMercy@p4FDE740B.dip.t-dialin.net)
19:56.23*** join/#htc-linux NoMercy__ (~NoMercy@p4FDE6D50.dip.t-dialin.net)
19:59.22*** join/#htc-linux nomercysave (~NoMercy@p4FDE7167.dip.t-dialin.net)
19:59.50phhstinebd: I'm going to hate that, and you too I guess, but is it possible to have our own "xdandroid" patch set ?
20:00.05phhlike the mms threshould on SMS that I hate
20:00.09mickeylhmm, damn compass doesn't want to talk to me
20:00.16stinebdphh: why hate? repo lets us do that easily
20:00.31phhstinebd: I don't want to fork, for stuff that isn't hardware linked
20:00.35phhmickeyl: talk louder
20:00.42mickeylhehe
20:01.05stinebdphh: we can always maintain a repo manifest of "core only" repositories
20:01.14stinebdplus one for our "enhanced" crap
20:01.16nomercysavephh whatever you did to free more ram makes android that much smoother - thanks
20:01.26phhnomercysave: removed ADSP ands CAMERA pmem.
20:01.33phhsince adsp and camera doesn't work, it's not a big loss
20:01.38stinebdphh: i was actually considering doing so for little things like cyanogen's launcher changes
20:01.50phhstinebd: ok cool then
20:01.59phhthen we do a single android release, for ALL android devices ! :p
20:02.31phhandroid-capable
20:02.35phhnot just android.
20:03.36stinebdphh: alternatively we may be able to use the same manifest for all of it and set buildspec.mk switches to change the apps which are built. i have to investigate if that's practical, though.
20:05.25stinebdie, have repo import the normal packages/apps/Mms and our own packages/apps/XDAMms and select which to use through a buildspec.mk define
20:05.26*** join/#htc-linux the_eye_ (~dimitris@ppp-94-69-22-85.home.otenet.gr)
20:05.35phhyes yes I understood
20:05.44phhI think that's the best
20:05.50phhso we can publish "XDAMMS" outside
20:06.03phhmaybe even android's market.
20:07.39phhor we want to go upstream.
20:07.45leviathanok, people
20:07.52leviathannot only the smartphone needs sleep
20:07.53leviathanme too
20:07.58phhleviathan: 'night
20:08.02leviathanI'll bump in tomorrow
20:08.09leviathanwhile lecture, perhaps
20:08.14leviathanor afterwards
20:08.20leviathanlets see it tomorrow
20:08.30leviathanphh: we have btw also some issues with mtd
20:08.38leviathansubpage size reading is not implemented
20:08.41leviathanand thx ^^
20:08.45phhI know
20:08.50phhwe also have them.
20:09.04phhand well, reading is not a big problem
20:09.07phhwriting is another one.
20:09.12leviathando you have some experimental patch around?
20:09.17phhno
20:09.19leviathanI could perhaps use it to extend
20:09.20leviathanhmm
20:09.28leviathanwhats about codeaurora?
20:09.30leviathando they?
20:09.50phhno
20:09.53phhthey only use yaffs2
20:09.58leviathanf...
20:09.59leviathanok
20:10.04phhas you say.
20:10.22leviathanso really NOone is using subpage size reading, so no one wrote a patch...
20:10.24leviathanok
20:10.41phhnot "noone"
20:10.46phh"noone that uses qualcomm stuff"
20:10.56leviathanok
20:11.14leviathanwell thats what I ment
20:11.23leviathanand the reason why I should sleep now
20:11.28leviathanread you tomorrow
20:11.32leviathangn8 everyone
20:11.56phhstinebd: is there any info about official App2SD stuff ?
20:12.34phhI was thinking about doing something clean (one ext2 loopback image per app)
20:12.48phhbut as google people are supposed to do it, it's maybe not worth the effort.
20:13.04*** part/#htc-linux zachy (~george@85.193.4.18)
20:13.07stinebdphh: not even a whisper
20:13.08*** join/#htc-linux JuNu_KN (~chatzilla@kons-5f711eff.pool.mediaWays.net)
20:13.19phhcool.
20:15.22*** join/#htc-linux pipomolo42 (~alex@home.boeglin.org)
20:37.19*** join/#htc-linux NoMercy (~NoMercy@p4FDE6CF1.dip.t-dialin.net)
20:44.27*** join/#htc-linux NoMercy_ (~NoMercy@p4FDE77C9.dip.t-dialin.net)
20:44.37*** join/#htc-linux Zoolooc (~fredsibar@p54957AE5.dip.t-dialin.net)
20:47.47*** join/#htc-linux NoMercy (~NoMercy@p4FDE7961.dip.t-dialin.net)
20:50.13*** join/#htc-linux ImCoKeMaN (~imcokeman@pool-74-99-149-19.hrbgpa.fios.verizon.net)
20:53.26GNUtoomickeyl, you want compas?
20:53.53GNUtoowe have accelerometer+compas driver?
20:54.40GNUtoofor suspend I'll try something
20:54.52GNUtoocm-mod with 2.6.32 android + cm-mod defconfig
20:55.03GNUtooif it doesn't work it's .32
21:05.25*** join/#htc-linux NoMercy_ (~NoMercy@p4FDE73B5.dip.t-dialin.net)
21:15.35ToAsTcfhphh: i see the oc thread has filled with assholes :(
21:16.13stinebdi'm not in there yet ;)
21:17.27*** join/#htc-linux NoMercy (~NoMercy@p4FDE6ADF.dip.t-dialin.net)
21:18.10ToAsTcfhuhg i hate it when u get a bunch of know it alls in one thread
21:18.44ToAsTcfhsome of the dream guys are the biggest dicks of them all
21:20.27stinebdooh 195MB memory now
21:21.03*** join/#htc-linux gauner1986 (~Adium@p5B38668C.dip.t-dialin.net)
21:21.17*** join/#htc-linux NoMercy_ (~NoMercy@p4FDE78FE.dip.t-dialin.net)
21:26.30phhToAsTcfh: that was totally unexpected.
21:26.40phh(yes, I'm being ironic)
21:28.37stinebdwhy the hell am i running a non-JIT build?
21:28.46phhstinebd: lol ?
21:28.57phhyou don't run your own build ?
21:29.03stinebdi do, but i built it without jit for some reason
21:29.07ToAsTcfhphh:?
21:29.08stinebdprobably testing
21:29.11phhmeeeeeeeeeh ?
21:29.16phhI've seen JIT in it -_-'
21:29.25stinebdwell i'm not using a kitchen build
21:29.44phhoh you rebuilt it meanwhile ?
21:29.49stinebdyeah
21:29.52stinebdnot sure why without jit though
21:30.16stinebdof course it's been like a week and a half since i've done anything with it
21:31.03stinebdi should try one of yours
21:31.07stinebdanything fancy in them?
21:31.21phhnot at all
21:31.32phhwell, one has Thai fonts.
21:31.45phhapart from that, there are all made from your kitchen, as is
21:31.46ToAsTcfhphh: for the mem divisor... just change it 4?
21:31.54phhToAsTcfh: from 3 to 4?
21:31.59ToAsTcfhyeah
21:32.10phhyes
21:32.18phhbut I don't think memory is the problem
21:32.28phhi'm not even sure the divisor is for the memory -_-'
21:32.35ToAsTcfh:/
21:34.02ToAsTcfhi cant figure out why i cant get below 610 ocin
21:34.10ToAsTcfhlike 575
21:34.25ToAsTcfhwell i can but its not accurate
21:35.18mickeylGNUtoo: accelerometer + compass would be interesting, yeah, that's why i took a short look. probably we need to trace what the android lib is doing
21:35.22mickeylwhich is closed IIRC
21:35.43GNUtoomickeyl, ok can I pm you?
21:35.48mickeylsure
21:58.00*** join/#htc-linux bkero (~freenode@horace.dionysian-mind.net)
21:58.00*** join/#htc-linux bkero (~freenode@osuosl/staff/bkero)
22:03.41*** join/#htc-linux luminoso (~lumos@av-217-129-131-108.netvisao.pt)
22:21.41*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
22:22.14*** join/#htc-linux ravishi (~stormy@c-98-249-6-188.hsd1.va.comcast.net)
22:29.29*** join/#htc-linux Zeman4323 (~Zeman4323@c-71-239-97-204.hsd1.il.comcast.net)
22:30.19stinebdi think i built without jit because of lowmem diamond
23:06.31*** join/#htc-linux surge (surge@pool-98-118-154-21.bflony.fios.verizon.net)
23:17.28*** join/#htc-linux Bry8Star_ (~Bry8Star@cpe-76-170-247-232.socal.res.rr.com)
23:32.10*** join/#htc-linux ravishi (~stormy@c-98-249-6-188.hsd1.va.comcast.net)
23:49.51*** part/#htc-linux gauner1986 (~Adium@p5B38668C.dip.t-dialin.net)

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with infobot logs, split per channel, etc.