IRC log for #htc-linux on 20111116

00:00.35WisTilt2this is different on .39 but i think a change i made to msm_fb timing in .27 is way way too long here on .39
00:01.08WisTilt2this is set for wake lock timeout at 5secs and should be like 200ms
00:03.50*** join/#htc-linux d3tul3 (~detule@pool-96-234-141-27.bltmmd.east.verizon.net)
00:05.12jonprywakelock timeout just keeps it from passing out
00:05.45WisTilt2yes but in .27 its 1/5 what we have it set at in .39 and im sure this caused this same problem
00:05.56WisTilt2im building it now
00:06.37d3tul3i guess it's old news but stopping the daemon has brought back normal ooperation on the screen for me
00:11.17WisTilt2jonpry: we're still using your looped power on panel also.  you think we still need that?
00:11.55WisTilt2wasn't that to make sure it finally came up back when we had all the failed wake issues?
00:19.04WisTilt2that seems to have fix it.  waking almost immediately every time now.  going to leave it in sleep for 5mins to be sure.
00:20.54jonpryalmost immediately?
00:21.16WisTilt2within 1/2 sec or so like before.  well under 1 sec anyway
00:21.22*** join/#htc-linux gassed (faxed@c-76-30-161-72.hsd1.tx.comcast.net)
00:24.23jonpryoh that sounds good
00:24.56jonprynow we just to to rewrite all of android in fixed point
00:25.04WisTilt2sleeping another 5mins but that last cycle it woke up under 1 sec
00:25.22WisTilt2scbs is still cooking right along too
00:34.56*** join/#htc-linux mitsutaka (~mitsutaka@rt.miraclelinux.com)
00:35.44*** join/#htc-linux Bry8Star_ (~Bry8Star@gateway/tor-sasl/bry8star)
00:41.45*** join/#htc-linux BabelO (~fcr@AMontpellier-553-1-44-52.w90-48.abo.wanadoo.fr)
00:41.45*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
00:45.08WisTilt2Jonpry: an interesting error from scbs->
00:45.10WisTilt2[  735.202056] Freezing of user space  aborted
00:45.10WisTilt2[  735.216094] scbs            S c03405d4     0   175      1 0x00080001
00:45.10WisTilt2[  735.227752] [<c03405d4>] (schedule+0x3a4/0x424) from [<c03407d8>] (preempt_schedule+0x48/0x64)
00:45.10WisTilt2[  735.239379] [<c03407d8>] (preempt_schedule+0x48/0x64) from [<c00405cc>] (scbs_ioctl+0xa0/0x2d4)
00:45.10WisTilt2[  735.251434] [<c00405cc>] (scbs_ioctl+0xa0/0x2d4) from [<c00ccdd4>] (vfs_ioctl+0x24/0x40)
00:45.10WisTilt2[  735.263336] [<c00ccdd4>] (vfs_ioctl+0x24/0x40) from [<c00cd9d4>] (do_vfs_ioctl+0x1b0/0x1d0)
00:45.12WisTilt2[  735.275421] [<c00cd9d4>] (do_vfs_ioctl+0x1b0/0x1d0) from [<c00cda28>] (sys_ioctl+0x34/0x54)
00:45.14WisTilt2[  735.287658] [<c00cda28>] (sys_ioctl+0x34/0x54) from [<c0026ee0>] (ret_fast_syscall+0x0/0x30)
00:45.18WisTilt2[  735.325988]
00:45.20WisTilt2[  735.332214] Restarting tasks ...
00:46.41WisTilt2that happened coming out of sleep btw after a 10min period
00:47.41d3tul3WisTilt2, the issue was msm_fb wakelock?
00:50.58WisTilt2well sort of.  timeout was set to 5 instead of 1 like .27 and that seems to have fixed it.  something else is causing it though so this isn't really the right fix but will work for now.
00:51.59WisTilt2all my wakes are well under 1 sec now no matter how long in sleep. logs show arm11 sleeping and waking so there is a timing issue somewhere that needs a fixin'
00:54.47WisTilt2d3tul3: do you run with autobl on or off?
00:56.29d3tul3well currently it doesn't matter in GB because it's looking for a sysfs hook
00:56.32d3tul3rather than a dbgfs
00:57.02*** join/#htc-linux TheXev (~user@50.sub-174-255-48.myvzw.com)
00:57.12d3tul3anyways, i am trying to make it all proper and what not -> acl has a nice ls driver that i am trying to incorporate but userland is giving me issues, keeps sending ioctl enables regardless what kind of settings i set
00:57.45WisTilt2?  autobl is working fine here with gb
00:58.34d3tul3hm it shouldn't be, can you actually see the screen dimming when you shine led?
00:59.02WisTilt2yep works great. been using it for a week now and just thought one of the emwe images addressed that and fixed it
00:59.20*** join/#htc-linux nappington (~nappingto@pool-108-28-112-137.washdc.fios.verizon.net)
01:00.09d3tul3auto bl should not be working in GB for multiple reasons not only is it looking for a sysfs auto_bl hook but it's also looking for different read out values from the sensor(scale 0-9)
01:00.16WisTilt2screen dims way down when i put finger over sensor and pops right back up with office lights when i remove it
01:00.22d3tul3damnit
01:00.48WisTilt2all i did was checked auto brightness in settings and worked right off
01:01.25WisTilt2this is with the current .39 tree also. only other thing i have is BT stuff on my end but autobl worked well before that
01:01.57d3tul3WisTilt2, does disabling auto_bl work as well?
01:02.39WisTilt2yep. disable then brightness is at the level of the slider up and down
01:03.51d3tul3ok i see in liblights where it's reading the dbgfs hook, at least i am not going crazy thanks for the heads up
01:04.09WisTilt2so userland never got changed to sysfs?
01:04.26d3tul3no it checks for sysfs, if it doesn't exist it defaults to dbgfs
01:04.51WisTilt2ah.  that's not half a bad idea actually until we get it all proper in the kernel
01:05.02d3tul3config_automatic_brightness_available
01:05.05d3tul3sorry
01:05.23d3tul3https://gitorious.org/xdandroid/hardware_msm7k/blobs/gingerbread/liblights/lights.c#line192
01:05.27d3tul3that's what i meant
01:05.58WisTilt2thats why its still working
01:05.58d3tul3ok well, i think abel's driver is just too damn sophisticated for our aosp build, i'll just port our existing code from -klt temporarily until we figure this out
01:06.35d3tul3yeah that's the defaulting back to dbgfs somehow i thought we had move past dbgfs
01:07.11WisTilt2ok after close to 30 times all wakes have happened no more than 1 sec so im calling it temp fixed and going home
01:08.38d3tul3you guys going to push the quick-wake-up config change, it doesn't build right without it correct?
01:09.36WisTilt2dont know if jonpry pushed that.  mine has all the BT stuff in it but you can just manually add it to msm_defconfig.  you need the name?
01:09.57d3tul3no i've been doing it, i am just wondering if the autobuild can work without it
01:10.23WisTilt2yeah wont affect anything unless someone uses the rootfs with the daemon
01:10.57d3tul3i mean i wonder if the autobuild can build the kernel
01:11.30jonpryprobably should pusha config
01:12.42bzod3tul3: was the camera broken in your compiled kernel also?
01:13.08d3tul3bzo, yes
01:13.25bzohmm, strange
01:13.37bzoI don't see any unpushed changes in my tree
01:14.02d3tul3well something i did could have broken it
01:14.13d3tul3though i did not touch any of the camera files scouts honor :)
01:14.30bzoWisTilt2: have you tried your camera since the fix?
01:14.55d3tul3all i've done post your camera commit is return microp-keypad back to something vanilla, not sure how that would interact with the camera
01:15.25bzocamera stuff is pretty isolated, don't think it could easily get affected by something else
01:15.27WisTilt2bzo: since your jpeg fix yes, worked great.  haven't tried it since then
01:15.51bzook, must be some weirdness in d3tul3's tree then
01:16.04d3tul3bzo, i tried the autobuild last night as well
01:16.08WisTilt2hmm, no its broke again just tried it
01:16.33*** join/#htc-linux hardwalker (~hardwalke@122-117-115-146.HINET-IP.hinet.net)
01:16.55WisTilt2[  756.382141] adsp: module JPEGTASK: region (vaddr 40020000 len 524288) clashes with registered region (vaddr 4006c000 paddr 26e13000 len 4096)
01:16.55WisTilt2[  756.382293] adsp: not patching JPEGTASK, lookup (40020000, 49152) failed
01:16.55WisTilt2[  756.382324] module JPEGTASK: verify failed.
01:17.00bzoI haven't pulled since I pushed it, so maybe something did get broke then
01:17.09bzothat was the error I saw before the fix
01:17.58WisTilt2yeah looks the same as before
01:19.53*** join/#htc-linux Bry8Star (~Bry8Star@gateway/tor-sasl/bry8star)
01:19.59*** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com)
01:32.58jonpryso its all good now? wrt wakes?
01:34.27WisTilt2wakes good after 30 cycles.  im testing it without that 10 loop panel on thing to see if we really need it in .39
01:36.12WisTilt2hey jonpry: do they have Royal Jamaican cigars on that island?
01:38.21*** part/#htc-linux nappington (~nappingto@pool-108-28-112-137.washdc.fios.verizon.net)
01:39.35d3tul3WisTilt2, right you are about auto_bl thanks for the heads up i would've been here spinning my wheels for a while longer
01:41.22WisTilt2so when will that be changed to sysfs for good or will it stay like that with both options for awhile?
01:41.48d3tul3i am about to change it to sysfs, dbgfs hook doesn't exist in nand or in .35
01:42.10WisTilt2ill change it on kernel side then.  never did since userland didn't support it:)
01:42.25d3tul3WisTilt2, i've already done it
01:42.27WisTilt2or you changing it on kernel side?
01:42.31WisTilt2ok np
01:42.37d3tul3i've moved it to board-*-led it has a nice spot for it there
01:44.21WisTilt2ok good deal. im heading home before the wife throws dinner in the trash.  catch you guys later.
02:20.54jonpryneed ics build server
02:21.00*** join/#htc-linux furtardo (~mks@nat/yahoo/x-zsjniykxfggbjhrf)
02:30.47*** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com)
02:32.43d3tul3bzo you have an idea what's broken?
02:56.57*** join/#htc-linux TheXev (~user@50.sub-174-255-48.myvzw.com)
03:24.33*** join/#htc-linux Rob2222 (~Miranda@p5DE4B2A3.dip.t-dialin.net)
03:40.18bzod3tul3: no idea, hope it's not one of these weird environmental things
04:07.24*** join/#htc-linux raymonddull (~raymonddu@c-98-209-145-76.hsd1.mi.comcast.net)
04:11.04*** join/#htc-linux AstainHellbring (AstainHell@unaffiliated/astainhellbring)
04:11.51*** join/#htc-linux swc|666 (~gecko@unaffiliated/swc666/x-4934821)
04:15.13*** join/#htc-linux Rob2223 (~Miranda@p4FFF392F.dip.t-dialin.net)
05:00.11*** join/#htc-linux Regulator (~duffman@unaffiliated/regulator)
06:32.37*** join/#htc-linux XeKToReX (~xTx@CPE-124-188-18-211.hdbq1.win.bigpond.net.au)
07:41.02*** join/#htc-linux XeKToReX (~xTx@CPE-124-188-18-211.hdbq1.win.bigpond.net.au)
07:44.44*** join/#htc-linux rob_w (~bob@ppp-188-174-99-85.dynamic.mnet-online.de)
07:51.12*** join/#htc-linux kiozen (~kiozen@p5DDF3AD3.dip.t-dialin.net)
07:56.36*** join/#htc-linux XeKToReX (~xTx@CPE-124-188-18-211.hdbq1.win.bigpond.net.au)
08:07.50*** join/#htc-linux XeKToReX (~xTx@CPE-124-188-18-211.hdbq1.win.bigpond.net.au)
08:15.48*** join/#htc-linux marc1706 (~marc@phpbb/modifications/marc1706)
08:25.42*** join/#htc-linux lamikr (lamikr@nat/nokia/x-mwtvvvmesbnnpfat)
08:33.36*** join/#htc-linux BabelO (~fcr@AMontpellier-553-1-44-52.w90-48.abo.wanadoo.fr)
08:33.36*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
10:11.54*** join/#htc-linux swc|666 (~gecko@unaffiliated/swc666/x-4934821)
10:14.10*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
10:49.25*** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:31b3:f822:77cc:67b5)
11:01.09*** join/#htc-linux XeKToReX (~xTx@115-64-132-224.static.tpgi.com.au)
11:01.17*** part/#htc-linux XeKToReX (~xTx@115-64-132-224.static.tpgi.com.au)
11:15.30*** join/#htc-linux tmzt (~tmzt@adsl-76-244-152-58.dsl.akrnoh.sbcglobal.net)
11:19.13*** join/#htc-linux mitsutaka (~mitsutaka@p2004-ipbf2206marunouchi.tokyo.ocn.ne.jp)
11:21.15*** join/#htc-linux bartman (~bart@2607:f2c0:a000:175:2e0:81ff:fe47:3d01)
11:38.25*** join/#htc-linux NYL (~NYL@92.83.118.224)
11:38.37*** join/#htc-linux rzk (~rzk@95-28-241-71.broadband.corbina.ru)
11:39.00*** join/#htc-linux MethoS- (~clemens@134.102.106.250)
12:06.07*** join/#htc-linux dobrin (~dobrin@85.91.150.26)
12:24.02*** join/#htc-linux kiozen (~kiozen@p5DDF3AD3.dip.t-dialin.net)
12:24.02*** join/#htc-linux Willd (willd@citu-202.citu.kth.se)
12:24.02*** join/#htc-linux tomaw (tom@freenode/staff/tomaw)
12:24.02*** join/#htc-linux Dinde (kayser@sur-internet.net)
12:31.56*** join/#htc-linux marc1706 (~marc@phpbb/modifications/marc1706)
12:49.01*** join/#htc-linux arif-ali (~arif-ali@ip-81-23-53-226.ask4internet.com)
12:59.31*** join/#htc-linux kiozen (~kiozen@p5DDF3AD3.dip.t-dialin.net)
13:09.47*** join/#htc-linux kiozen (~kiozen@p5DDF3AD3.dip.t-dialin.net)
13:17.46*** join/#htc-linux NYL_ (~NYL@92.83.118.224)
13:20.08*** join/#htc-linux mgross029 (c0234f46@gateway/web/freenode/ip.192.35.79.70)
13:48.24*** join/#htc-linux riotz (riotz@unaffiliated/riotz)
14:02.22*** join/#htc-linux DJWillis (djwillis@cpc1-bath5-2-0-cust122.aztw.cable.virginmedia.com)
14:09.44*** join/#htc-linux helicopter88 (~helicopte@host22-119-dynamic.27-79-r.retail.telecomitalia.it)
14:10.38*** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com)
14:15.59*** join/#htc-linux Alex[sp3dev] (~alexander@178.176.24.16)
14:17.41*** join/#htc-linux Cotulla (~myfakemai@nat100-255-205-109.tvoe.tv)
14:17.50Cotullahi
14:19.36Alex[sp3dev]hi
14:20.02Cotullahow are u, Alexx?
14:21.02Alex[sp3dev]Cotulla: fine but a bit busy this week
14:21.54Alex[sp3dev]Cotulla: so i got myself a galaxy s2. i should say it is a real shovel
14:22.21Alex[sp3dev]jONpRY: hi
14:22.36Cotullafuu
14:22.45Cotullau lost in my eyes :D
14:23.05Alex[sp3dev]Cotulla: shoul i've gone for an evo 3d instead?))
14:23.51Alex[sp3dev]i thought about getting a wp7 device. but then again, no official native sdk, no opengl and having to use windows + visual studio is a no go for me
14:26.02Cotulladunno
14:26.10Cotullalol
14:26.24Cotullau think Eclipse + Android plugin nice?
14:26.27Cotullait's crap
14:26.32Cotullalaggy crap
14:28.03*** join/#htc-linux riotz_ (riotz@u.cant.drop.my.firewall.made.with.packet-filter.org)
14:29.07Alex[sp3dev]no, eclipse is not better. anyway, i still need to finish some stuff about hd mini and xperia. but first i need to do some homework ;)
14:29.20Cotullalol
14:29.41Cotullaok
14:30.10Alex[sp3dev]Cotulla: leo is also 4.3 inches, right?
14:30.36Cotullayes
14:31.19*** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com)
14:31.44Alex[sp3dev]Cotulla: do you sometimes find it is like a shovel? but actually turns out to be very good for reading. ok, back to work
14:32.13Cotullano it doesn't look like shovel
14:32.18Cotullabut Samsung Focus - yes
14:32.25Cotullaa plastic shovel
14:32.43Alex[sp3dev]Cotulla: i meant the size, not the design
14:32.48Cotullahm I may go and play to sand
14:32.55Cotullawith Focus
14:32.59Cotulla:D
14:33.18Cotullahm size
14:33.19Cotullahm no
14:33.26Alex[sp3dev]so you have samsung focus?
14:33.30Cotullayes
14:33.50Cotullahttp://cotulla.pp.ru/leo/WP7S/IMAG0425.jpg
14:34.03Alex[sp3dev]someone donated it for unlock? or you're working on some other project?
14:34.10Cotullafor custom roms
14:34.22CotullaLEO has rather big weight and metal, it's why I don't feel it like a shovel
14:35.24Alex[sp3dev]is the stock bootloader there locked? are you loading magldr from wp7 for flashing it for the first time?
14:35.25Cotullaand Focus has good end for shovel: you can grab sand via it very nice
14:35.35Cotullayes, locked
14:35.38CotullaI am flashing MAGLDR
14:35.43Cotullaand then users can flash custom roms via it
14:35.55Cotullato make it more safe
14:36.03Alex[sp3dev]how do you flash it if the bootloader is locked? i mean, via wince flash driver or you load it to ram?
14:36.21Cotullait's from samsung DLMODE
14:36.47Alex[sp3dev]why not use DLMODE for everything? is it unsafe and can trash the bootloader?
14:37.10Cotullayes some kind of, I decide that it's not really safe
14:37.20Alex[sp3dev]ok, i see now
14:37.24Cotullaand I won't try ofcourse
14:37.33*** join/#htc-linux balans2 (~user@82-170-217-205.ip.telfort.nl)
14:37.57Cotullaanyway I need develop now new usb connection system
14:38.58*** join/#htc-linux GNUtoo (~gnutoo@host246-111-dynamic.48-82-r.retail.telecomitalia.it)
14:39.36*** join/#htc-linux balans21 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
14:39.52*** join/#htc-linux rzk (~rzk@95-24-81-229.broadband.corbina.ru)
14:43.51Alex[sp3dev]ok, see you later
14:43.51*** part/#htc-linux Alex[sp3dev] (~alexander@178.176.24.16)
14:44.07Cotullabb
14:46.50Blistais bored!!
14:47.25Cotullaformats Blista brain and push fresh system to it
14:49.11*** join/#htc-linux balans22 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
14:49.34Blista[Now running Android] is still bored, dispite the live wallpaper!
15:10.52*** join/#htc-linux emwe (~mweirauch@cable-86-56-10-252.cust.telecolumbus.net)
15:19.37*** join/#htc-linux LordDeath (~LordDeath@cable-81-173-166-124.netcologne.de)
15:25.26*** join/#htc-linux balans21 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
15:30.17*** join/#htc-linux balans22 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
15:33.52*** join/#htc-linux avinashhm (~avinash-h@192.91.66.189)
15:39.01*** join/#htc-linux balans21 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
15:40.32*** join/#htc-linux balans23 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
15:42.34*** join/#htc-linux balans24 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
15:58.05*** join/#htc-linux mgross029 (~mgross@cpe-107-10-11-218.neo.res.rr.com)
15:59.19*** join/#htc-linux NYL (~NYL@109.97.253.189)
16:01.51*** join/#htc-linux rpierce99 (~rpierce99@96-42-102-103.dhcp.stcd.mn.charter.com)
16:04.27*** join/#htc-linux helicopter88 (~helicopte@host22-119-dynamic.27-79-r.retail.telecomitalia.it)
16:04.34*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
16:25.54*** join/#htc-linux avinashhm (~avinash-h@192.94.92.14)
16:29.23*** join/#htc-linux joplayer (~jonathan@2a01:e34:ef80:50:21d:60ff:feec:b36c)
16:36.54*** join/#htc-linux Markinus (~Adium@212.255.33.119)
16:41.17*** join/#htc-linux balans21 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
16:48.50*** join/#htc-linux Blista (~nel@209.59.220.47)
16:51.47*** join/#htc-linux LordDeath_ (~LordDeath@cable-81-173-166-124.netcologne.de)
16:53.06*** join/#htc-linux NYL_ (~NYL@109.97.253.189)
16:55.46*** join/#htc-linux WisTilt2 (~wisgreg@wireless251.wirelesstcp.net)
16:57.06*** join/#htc-linux Markinus1 (~Adium@212.255.33.119)
16:59.22WisTilt2jonpry: i looked deeper into this slow wake and looks like the problem might be something to do with when scbs sends lots of data to userland that's queued during sleep.  longer sleep, more data and it progressively takes longer to wake until all data is sent.  make any sense?
17:00.50jONpRYits possible. maybe try renicing the daemon. if it helps there is something that can probably be done in the kernel
17:01.19WisTilt2detule/bzo: i think we need to fix pmem in .39 also.  we are getting lots of weird memory overlaps at times that seem to be related to acoustic possibly.
17:02.15WisTilt2jonpry, is the data to userland real low priority or can that be set?  seems like irq's might be disabled during the transfer also.
17:02.25Cotullawhat's up?
17:02.49Cotullahidden meaning jpONRY
17:03.13jONpRYits just an ioctl for every element. should be limited by the processes priority
17:03.51jONpRYno disable of irq, not even a bulk transfer. should be lots of opportunity for the process to yield
17:04.51WisTilt2is each data block transfers as a separate process or is it all one big transfer once it starts?
17:05.31jONpRYone ioctl per measurement
17:07.03WisTilt2ill try setting renice to something like 15 and see if it makes any diff.  i assume its just at the default now
17:07.18jONpRYjust default
17:28.48*** join/#htc-linux balans22 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
17:41.46*** join/#htc-linux jONpRY (~jon@unaffiliated/jonpry)
17:51.00*** join/#htc-linux fpu (~nonperson@abxh72.neoplus.adsl.tpnet.pl)
17:51.11fpuhello
17:51.15*** join/#htc-linux |Jeroen| (~jeroen@d5152B25B.access.telenet.be)
17:52.47Cotullahey
17:53.22fpugiving up
17:53.57fpuscrew usb host, someone screwed up fso packages in debian....no telephony either due to some stupid error
17:56.56*** join/#htc-linux kiozen (~kiozen@ppp-93-104-95-75.dynamic.mnet-online.de)
17:57.48joplayerHi guys, just an offtopic question, has anyone been able to compile ICS with GCC4.6 ?
17:58.20*** join/#htc-linux NYL_ (~NYL@109.97.253.189)
18:00.26jONpRYhi Cotulla. CAPS LOCK IS CRUISE CONTROL FOR COOL
18:01.32jONpRYi'm probably behind the times, but i can't stop lmao
18:01.59CotullaYeAh It'S CoOl
18:02.03fpujoplayer: Why not use gcc included with ndk?
18:03.03jONpRYprobably talking about the pthread build problems
18:03.15jONpRYlots of interesting "solutions"
18:05.31joplayerfpu: I'm not using the ndk but i indeed should
18:06.08joplayerjONpRY, not aware of this problem but maybe
18:06.42jONpRYandroid/ics/external/llvm/lib/Support/Threading.cpp:96: undefined reference to `pthread_create'
18:09.49joplayerOh yeah that is the problem indeed  i didn't remember
18:10.01joplayerso gcc 4.4 could solve the thing ?
18:10.48jONpRYhow does one even manage to build the android with the wrong gcc. isn't that like an a serious engineering effort?
18:11.46WisTilt2jonpry, suspend_devices_and_enter rc was what again? reniced to 15 and still same thing.  after 10mins sleep it took exactly 10secs for screen to wake and looks like scbs_take_measure is what cleared it to come up.
18:11.57jONpRYthe real solution has something to do with branches. but i can't really decode what was said
18:12.11jONpRYrc = resume_console
18:12.31jONpRYend = after resume_console
18:12.41WisTilt2hmm, does the data copying to userland also get stuck in the console buffer possibly?
18:13.15jONpRYwell no. but the printks from get data point etc do
18:14.03jONpRYprobably can remove all printk from scbs.c, doesn't do much anymore
18:14.23jONpRYyou have a quick dmesg i could look at?
18:14.29WisTilt2thats what the problem is then.  same as in .27 where the console buffer got filled up and took forever to dump
18:14.53WisTilt2how much dmesg you want, just from where i started the wake up until the end of data point dumps?
18:15.31WisTilt2this is actually a bug in printk, which i thought was fixed in .38 and later
18:15.36joplayerjONpRY, just wanted to know thx for your help
18:15.41*** join/#htc-linux fpu_ (~nonperson@afcl207.neoplus.adsl.tpnet.pl)
18:16.35WisTilt2http://pastebin.com/vPQGUFt8
18:17.21jONpRYjoplayer, this is one method: http://pastebin.com/raw.php?i=zTsG4k5G
18:18.41jONpRYWisTilt2, need dmesg from during sleep. thats all stuff after
18:18.45joplayerThanks a lot, that will help me for sure
18:20.09WisTilt2http://pastebin.com/BzmEMsAR
18:21.02jONpRYyeah so printks all over. pm.c timer.c and the rc
18:21.07jONpRYer rtc
18:21.22WisTilt2im building with all the logging disabled and i bet that fixes it.  i need to look at that printk console deal again, i know that was a big problem in .27 and scbs doing same thing.
18:22.20jONpRYyou remove the logging from timer rtc pm as well?
18:22.35WisTilt2no will do now.
18:24.00WisTilt2is that kernel/timer.c?
18:25.26jONpRYmach-msm
18:25.43jONpRYits making the msm_timer: update slee
18:26.06WisTilt2ill disable all of it and test.  then we need to wrap those printk's in if debug like the other stuff
18:26.07jONpRYWake up in 5 seconds is from the rtc-msm7x01
18:26.29jONpRYARM11 Waking up, ARM11 Entering sleep has got to go
18:26.41jONpRYmsm_irq stuff
18:26.57WisTilt2i put that in back when we were verifying it was actually sleeping:)
18:31.09*** join/#htc-linux Alex[sp3dev] (~alexander@178.176.24.16)
18:32.04WisTilt2ok think i got them all so building.  back shortly
18:33.20*** join/#htc-linux paulk (~paulk@lib33-1-82-233-88-171.fbx.proxad.net)
18:42.44*** part/#htc-linux Cotulla (~myfakemai@nat100-255-205-109.tvoe.tv)
19:03.51mgross029WisTilt2, just curious did you get bt working on .39 saw something back in the logs but wasn't sure if you got it working
19:07.49WisTilt2mrgross029: working as far as pairing but audio paths aren't working yet.  problem in acoustic so that's the next project.
19:08.13rpierce99mgross029: unless i'm mistaken it requires new BT firmware though
19:08.33WisTilt2have it
19:08.45rpierce99right, just didn't want mgross029 thinking it would "Just work"
19:09.24WisTilt2jonpry: that seems to have fixed it.  sleep for 15mins and waking fast every time.  nice to see dmesg without all the spam.
19:09.59WisTilt2im still going to look at the printk console issue.  should be able to fill the buffer without it slowing down like that
19:12.07mgross029WisTilt2, Ok I was just curious because I rememberd emwe had to adjust bcrm_patchram_plus to work with .35, because his rfcomm was located in a different fs than what was used in .27
19:12.26mgross029but if you are pairing you must have gotten that piece working.
19:13.15rpierce99emwe wrote a lot of stuff over for .35, .39 is mostly copypasta from .27 from what I understand
19:14.48mgross029rpierce99, yep I was just checking
19:15.27rpierce99now that i'm thinking about it, was the new firmware for wifi?
19:15.37rpierce99should have just kept his mouth shut
19:17.24*** join/#htc-linux stroughtonsmith (~steven@86-44-101-179-dynamic.b-ras2.bbh.dublin.eircom.net)
19:19.49*** join/#htc-linux BabelO (~fcr@AMontpellier-553-1-44-52.w90-48.abo.wanadoo.fr)
19:19.49*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
19:20.09mgross029WisTilt2, your latest autobuild is missing zImage again and I keep getting an error when trying to compile on quickwakeup.h :-\
19:20.47WisTilt2the config didn't get pushed with the other changes
19:21.17mgross029Ok...
19:22.23WisTilt2jonpry im going to push these changes with the printk's commented out for now.  we can add them to debug logic later if you want.
19:22.41WisTilt2waking great every time now
19:24.14*** join/#htc-linux rob_w (~bob@ppp-188-174-99-85.dynamic.mnet-online.de)
19:32.03*** join/#htc-linux swc|666 (~gecko@unaffiliated/swc666/x-4934821)
19:32.46WisTilt2ok everything pushed so give autobuild a few mins to get and build it and see if all is well now.
19:33.43detuleWisTilt2, i saw your ping about pmem
19:34.09detulethe pmem spam in dmesg, that's a debugging message that we need to verify if it happens in .27
19:34.14WisTilt2hi detule. yeah im seeing random pmem overlaps happening mainly with acoustic i believe
19:34.17detulecan't see it in .27 because the debugging printk is not in place
19:35.16detuleperhaps a brave soul running .27 can fire up a test kernel for us with the debugging printk in place
19:35.31WisTilt2this might be whats breaking the camera now and then.  camera worked last night after i had made a call if that makes any sense.  fresh boot it breaks.
19:36.26detuleyeah we were looking with bzo last night and it certainly seems to affect camera - bzo booted without acoustic and he was able to use the camera
19:36.27mgross029detule, saw the logs earlier with you and bzo do you lose or have audio problems after using camera?
19:36.39detulelost camera problems without acoustic
19:37.19WisTilt2camera works every time if you remove acoustic?
19:37.52detulethat was where we left it off with bzo last night, perhaps he has more to say after some more testing
19:37.57Alex[sp3dev]i'd say check your kernel. on my 35, both acoustic and camera work fine simultaneously.. so you can just steal it
19:38.30WisTilt2isnt this acoustic mainly from .35 detule? dont know exactly what you had in that merge
19:38.53detulei added almost nothing, only thing was Alex[sp3dev]'s adie semaphore fix
19:39.10detulewhich apparently is not the issue since he's not having problems with it
19:39.30detuleother than that the acoustic in .39 is as you and jonpry had it
19:39.50mgross029I stumbled upon this patch from CM.  When I was searching for errors I was seeing in my dmesg https://github.com/CyanogenMod/cm-kernel/compare/51b75ba46f...793c0c0244
19:40.02Alex[sp3dev]you should probably take kdiff3/meld and compare the files against 35 or my 27 fork with acoustic fixes
19:40.39WisTilt2ill dig into this after lunch. need engineering fluid before i can continue.
19:41.18detuleWisTilt2, to disable acoustic and test camera without it the easiest thing is to comment out the chmod /dev/htc_acoustic_wince line in init.cfg/init.ginger*
19:41.59mgross029I have put it into my local .35 repo to see if it would fix a mic or audio problem I was having.  So far it has seemed to help some with the mic disconnect on first call and no ringer
19:43.28mgross029Not sure if it will help with acoustic and camera problem on .39 but just thought I would throw it out there.
19:44.23WisTilt2detule: ok will try that thanks.  mgross029 my autobuild built it fully after these last commits so you should be good to go now.
19:44.28WisTilt2lunch, bbl
19:45.41mgross029WisTilt2, Thanks already pulled and compiling :p
19:51.47*** part/#htc-linux Alex[sp3dev] (~alexander@178.176.24.16)
20:18.59*** join/#htc-linux balans2 (~Barry@53564629.cm-6-7b.dynamic.ziggo.nl)
20:44.19mgross029WisTilt2, getting error with hciattach -> port --enable_hci could not be opened, error 2 and I don't see anything in the hciconfig
20:44.58mgross029Can't even enable bt on my Rhod500 in .39...  Still some pieces missing I believe.
20:45.25WisTilt2i need a nap after that lunch
20:45.38WisTilt2mgross029: i dont have any BT stuff pushed to the tree yet
20:45.55mgross029Ahh.  That would be why. :p
20:45.57mgross029lol
20:46.03*** join/#htc-linux toastcfh (~toastcfh@41-32.200-68.tampabay.res.rr.com)
20:46.15mgross029WisTilt2, Enjoy your nap
20:46.26*** join/#htc-linux toastcfh (~toastcfh@unaffiliated/toastcfh)
20:46.27WisTilt2jk on the nap but sure sounds good
20:47.13*** join/#htc-linux Markinus (~Adium@212.255.33.119)
20:47.20WisTilt2mgross029: if you want to test BT real quick i'll give you a link to my .39 kernel with it.  curious if you get hci0 to come up on your device.
20:47.45mgross029Sure...
20:49.32WisTilt2you're running GB correct?
20:54.32mgross029Correct with acoustic
21:03.21WisTilt2well something is partially broke now.  Enabling BT never enables on userland side even though hci0 gets created. :-( mgross029 let me know if you're seeing same thing.
21:05.10emwe.35 and .27 with that gb image is fine, just fyi.
21:05.15emwei only got a sleep issue on .35.
21:05.19mgross029WisTilt2, Yeah I am unable to enable hear as well
21:05.35WisTilt2thanks emwe, that was my next question
21:05.45WisTilt2mgross029 are you getting pairing though?
21:05.53emweWisTilt2: i was about to write that after reading the logs from last night. forgot. sorry.
21:06.07mgross029WisTilt2, no can't even enable
21:06.21mgross029Now I cannot connect adb to the phone
21:06.24WisTilt2check to see if hci0 is there
21:06.53mgross029Phone is locked
21:07.22WisTilt2might need to check it in console.  my adb still working but responding very slow now
21:07.22mgross029Its like a SOD with the screen on...  LOL
21:08.25WisTilt2i made the mistake of pulling all the commits and didn't keep a clean BT tree that i had working.  guess i need to revert commits back a week or so in a new local tree.
21:09.27mgross029Trying to do too much at once. :p
21:10.09WisTilt2emwe: which of these is GB userland using for BT?
21:10.42WisTilt2sys/class/bluetooth/hci0
21:10.44*** join/#htc-linux arif-ali_ (~arif-ali@94-192-24-56.zone6.bethere.co.uk)
21:11.02WisTilt2sys/devices/platform/msm_serial_hs.1/tty/ttyHS1/hci0
21:12.12emweyou are asking good questions :P
21:12.14emweone second...
21:12.37WisTilt2i dont have the GB source here or i'd look myself:)
21:17.36mgross029bbl
21:18.25emweWisTilt2: now the question it what does what where :)
21:18.46WisTilt2rats nest?
21:19.51*** join/#htc-linux helicopter88 (~helicopte@host22-119-dynamic.27-79-r.retail.telecomitalia.it)
21:20.08emwesystem/bluetooth/bluedroid/bluetooth.c is where i am looking now
21:22.35*** join/#htc-linux Alex[sp3dev] (~alexander@178.176.24.16)
21:24.21emwewhat was the last error message again WisTilt2?
21:24.43emwefrom logcat
21:25.37WisTilt2i just rebooted phone.  no error, just nothing enabled after bluetooth rfkill on
21:25.50WisTilt2you need a logcat i can boot it back up and get it
21:26.21emwewell, no hurries today
21:26.27emwei need to get to bed a bit earlier today
21:26.39emweyou can mail it, though.
21:26.50emwejust not sure i'll come around to it earlier then in 22hours :)
21:26.53WisTilt2as far as you know though shouldn't it be in sys/class/bluetooth?
21:27.11emwenot sure. i just got it working in kernelland.
21:27.46WisTilt2i'll revert these other commits and get it back to where i had it working.  i'll let you know what i found when i resolve it.  thanks
21:27.54Alex[sp3dev]emwe: when will we roll out acoustic xdandroid?
21:28.17emweAlex[sp3dev]: when stinebds think's it's ready.
21:28.19detuleWisTilt2, any of the recent commits break something BT related?
21:28.27emwei got so much side tracked and abandoned acoustic debugging...
21:28.55emweAlex[sp3dev]: i am sharing my inofficial builds among the peeps. everybody riding on now.
21:29.45Alex[sp3dev]emwe: i see. i'm running CM7 on NAND now with 35 kernel.. Almost flawlessly except for still image capture fails
21:30.23emweAlex[sp3dev]: must i be jealous? ;)
21:30.30WisTilt2detule im thinking so. im reverting my tree locally back to last week where i had it working except for audio path.  probably going to be this pmem problem because im sure seeing a lot of spam about overlapping memory lately.
21:30.35Alex[sp3dev]emwe: yes of course
21:31.45emweAlex[sp3dev]: giving the fact some feature patches are coming from cm we pull in, i whonder how much it's still worth pursuing this "running after" thing.
21:33.20Alex[sp3dev]emwe: well, xdandroid's main difference is the ability to run from sd card. maybe it would make sense to only leave custom rootfs and switch to using cm's system and stuff? for nand builds cm is surely better - it has autobuilder, and, besides, having separate initrd for every device simplifies everything and allows automatic updates
21:34.33Alex[sp3dev]on the other hand, given the project maturity and our devices age, it is best to do the bare minimum needed to implement missing drivers and fix bugs. starting a large project is a waste of time now
21:34.37emweAlex[sp3dev]: i only know xdandroid. i dunno at all how's it done on CM. so i cna't say anything on that.
21:34.46emweyeah :/
21:34.52emwebut i still wan't to rebase on your tree :P
21:35.10emwejust don't wanna start this now when i can waste my time on userland.
21:35.10Alex[sp3dev]emwe: that would make sense now that wl1251 and suspend are working
21:35.24emweyeah, the time...
21:36.44Alex[sp3dev]i plan to fix camera focus and snapshot on kovsky. overall tiny bugs are left and i'm contended with the experience i got. xperia android is usable on a daily basis. although i got myself a new toy for experimenting with opengl es2, neon and stuff like that
21:38.05emweyeah, read that
21:38.30emwealso using the RHOD daily. otherwise it's not worth the effort just tinkering on the device :)
21:38.48emweso i gotta run. gnight everyone!
21:38.53Alex[sp3dev]good night
21:50.17jONpRYWisTilt2, working good eh?
21:50.55WisTilt2oh yeah, scbs flying right along and after numerous long sleeps it wakes quick everytime.
21:51.15WisTilt2scbs is so nice.  battery is happier
21:51.59WisTilt2working on BT again while im in a good mood
21:55.00WisTilt2jonpry: any reason we have I2C_BOARD_INFO("audience_A1010", 0xf4>>1) disabled?
21:55.11detuleaudience_a1010 is noise reduction right?
21:55.14WisTilt2just noticed it is commented out
21:55.27WisTilt2yeah and it should be on i'd think unless we're doing it elsewhere
21:55.34jONpRYa9 takes care of it
21:55.39detulei think  i tried enabling it once, and lost mic
21:56.04jONpRYi extracted its firmware and stuff to make it all work. just to find out it gets loaded all by itself
21:56.15Alex[sp3dev]WisTilt2: i think you should consult my 27 branch with acoustic for a1010
21:56.46WisTilt2Alex[sp3dev] i see you have it enabled
21:57.05Alex[sp3dev]WisTilt2: it was done by JB and he surely knows how stuff works
21:57.52jONpRYwinmo turns it on. but i think they moved the code into amss later or something
21:58.37WisTilt2building it with it enabled to see if it kills mic now.  probably needed now with dual mic working right.
21:59.36*** part/#htc-linux Alex[sp3dev] (~alexander@178.176.24.16)
21:59.57jONpRYafaik a1010 is dual mic
22:00.56jONpRYnobody ever believed in arm9 take over of audio stuffs :)
22:08.18*** join/#htc-linux [acl] (~abel@96.246.167.90)
22:19.32*** join/#htc-linux rajkosto (~rajkosto@2001:470:d76b:da7a:31b3:f822:77cc:67b5)
23:12.39*** join/#htc-linux mitsutaka (~mitsutaka@p2004-ipbf2206marunouchi.tokyo.ocn.ne.jp)
23:12.43*** join/#htc-linux Cass (~Cass@188-220-34-222.zone11.bethere.co.uk)
23:13.52*** join/#htc-linux mgross029_ (~mgross@cpe-107-10-11-218.neo.res.rr.com)
23:14.25*** join/#htc-linux d3tul3 (~detule@pool-96-234-141-27.bltmmd.east.verizon.net)
23:15.37*** join/#htc-linux mgross__ (~mgross@cpe-107-10-11-218.neo.res.rr.com)
23:26.16*** join/#htc-linux Wally__ (~wally@63-226-130-207.mpls.qwest.net)
23:39.49*** part/#htc-linux marc1706 (~marc@phpbb/modifications/marc1706)
23:48.56*** join/#htc-linux vinceweis_ (433575ca@gateway/web/freenode/ip.67.53.117.202)
23:49.07*** join/#htc-linux leviathan (~quassel@2001:470:26:484:6ef0:49ff:fee6:8dca)
23:51.06*** join/#htc-linux riotz_ (riotz@unaffiliated/riotz)
23:52.41*** join/#htc-linux avinashhm (~avinash-h@192.94.92.14)

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