irclog2html for #htc-linux on 20100406

00:02.13gauner1986nothing to do tomorrow NetRipper? :)
00:02.24gauner1986or are you a student, like me
00:02.34NetRippernah i got work
00:02.43NetRipperim off to bed in a min
00:02.47NetRipperalarm goes in 5 hours
00:02.49NetRipperless
00:02.53gauner1986ah
00:02.54gauner1986poor guy
00:02.58NetRipperyeh
00:03.01gauner1986i just have to get up at 11am
00:03.06NetRipperdarn you
00:03.17gauner1986i love my student life
00:03.18gauner1986:P
00:03.30NetRipperyea i loved it too
00:03.44NetRipperbut its nothing more than sentiment to me now
00:03.44NetRipper:P
00:04.28gauner1986at least you got money now
00:04.30gauner1986;)
00:06.51tmzt_mickey|zzZZzz: what are you removing wakelocks for?
00:09.55ToAsTcfhphh: ok well i get this
00:10.34ToAsTcfhphh: <4>[ 3320.702392] msm_cpufreq_target 595200 r 1 (480000-595200) selected 595200
00:11.41ToAsTcfhno errors found in last_kmsg. strange i know. as soon as i set the clock it locks the phone up no matter what i do
00:12.02ToAsTcfhitll hang for a few then reboot
00:19.48ToAsTcfhalso i seen u say something about memory divisors. do u tung that could be the issue?
00:20.07ToAsTcfhbzo: ping
00:20.48bzoyes?
00:21.08ToAsTcfhi seen u and phh talking about memory earlier
00:21.22ToAsTcfhdo u think that could be my issue
00:21.33ToAsTcfhoverclocking
00:22.29ToAsTcfhi set the clock from within android and it just causes the phone to hang and reboot after a minute or so
00:22.43bzoI wouldn't think so. I don't know if you kernel can control that, but even if it did it wouldn't cause a problem for small increases in freq
00:23.28ToAsTcfhyeah i get the same results even when setting it to 528
00:23.55ToAsTcfhactually anything over 480 (max pll1
00:24.44bzoyour phone does not use pll2 for the max freq?
00:25.16ToAsTcfhwhen the clock moves up to pll2 it hangs
00:25.28ToAsTcfhwhen using ur patch
00:26.10bzodoes it normally use pll2?
00:26.16ToAsTcfhyep
00:26.21ToAsTcfhfor 528
00:26.26AstainHellbringahhh feels sooo nice to be home again
00:27.23ToAsTcfhbzo: #endif
00:27.23ToAsTcfh{ 528000, ACPU_PLL_2, 2, 1, 132000, 3, VDD_7, 160000, 0, 5, -1 },
00:27.23ToAsTcfh{ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
00:27.23ToAsTcfh};
00:28.44ToAsTcfhwhat i ended up doin is hardcoding it in acpuclock. then bringing the max in the config down to 480. at that point it boots.
00:29.00ToAsTcfhthen i echo the max after im booted in
00:29.07ToAsTcfhand it hangs
00:29.12ToAsTcfhwith no errors
00:30.00bzosounds like the same as booting with the oc patch
00:30.11ToAsTcfhyeh ut it
00:30.18ToAsTcfhit is
00:30.57ToAsTcfhi did it this way to get a better look at kmsg and last_kmsg
00:31.21ToAsTcfhseeing android is such a pain at retrieving ramconsole
00:32.19ToAsTcfhi seen somewherethat it could be the radio stopping it. but if that were the case then 528 would work fine
00:33.02ToAsTcfhso something must be conflicting man
00:33.14bzonot exactly, the oc patch changes the divider of pll2 to 1 instead of 2
00:33.23bzoso it runs different even at 528
00:33.45ToAsTcfhdifferent how?
00:33.54ToAsTcfhlike what would be changed
00:34.13ToAsTcfhmem dividers and such
00:34.18bzopll2 runs at 1 divider instead of 2 (1.056 ghz)
00:35.09ToAsTcfhi understand that much. but ur saying its not really being down clocked to 528 then?
00:35.38bzostock = pll2 @ 1056myz /2 = 528mhz
00:35.48bzooc = pll2 @ 528mhz/1 = 528mhz
00:36.09gauner1986im off to bed
00:36.13gauner1986good night
00:37.34ToAsTcfhi understand that man. i just dont understand why that would be a big deal if i wasnt running over 528
00:37.52ToAsTcfhi guess other things run of that bus also
00:37.56bzomaybe it doesn't like the 1 divider?
00:38.23ToAsTcfhthat dosnt leave me with much options then huh lol
00:38.54bzoyou could try leaving the divider at 2, though I think you may have done that already once
00:39.12ToAsTcfhno ive only used ur source
00:39.32ToAsTcfhbut how to oc with the divider of 3
00:39.34ToAsTcfh2
00:39.57ToAsTcfh528 is half which makes it the max
00:40.02bzothere should be a comment that says something about changing the divider, comment out that code
00:40.13bzoset the oc freq to twice what you are aiming for
00:41.49ToAsTcfh* Program clock divider */
00:42.38ToAsTcfh* AHB_CLK_DIV */
00:42.38ToAsTcfhclk_div = (readl(A11S_CLK_SEL_ADDR) >> 1) & 0x03;
00:42.38ToAsTcfh/*
00:42.38ToAsTcfh<PROTECTED>
00:42.38ToAsTcfh<PROTECTED>
00:42.39ToAsTcfh<PROTECTED>
00:43.55ToAsTcfhnm
00:46.14ToAsTcfhbzo: does this section need to go also >>> CLK_SEL_SRC1N0 */
00:47.17bzocomment out the 11 lines of code after the comment
00:56.47ToAsTcfh# echo 610000 >/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
00:56.51ToAsTcfh<4>[  185.404388] msm_cpufreq_target 595200 r 1 (480000-595200) selected 595200
00:57.59ToAsTcfhbzo: unless u meant to set it twice the desired speed in acpuclock
00:58.10*** join/#htc-linux blendmaster1024 (~blendmast@unaffiliated/blendmaster1024)
00:58.48ToAsTcfhit did the same thing though either way
00:58.56bzooh, I forgot you have no kernel args, how are you setting the oc freq?
00:59.11ToAsTcfhin acpuclock
00:59.52bzoright, so set the desired freq to 2x what you want (i.e. 1.2ghz will give you 600mhz)
01:01.14ToAsTcfhstatic int oc_freq_khz = 1200000;
01:01.14ToAsTcfhmodule_param_named(oc_freq_khz, oc_freq_khz, int, S_IRUGO | S_IWUSR | S_IWGRP);
01:02.59ToAsTcfhand when echoing just echo 1200000 ?
01:03.06ToAsTcfhor 600000
01:03.30bzotry the higher # for now, you'll have to fix some other stuff if this actually works
01:07.18ToAsTcfhecho 1200000 >/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
01:07.27ToAsTcfhno dice it didnt even set it
01:08.30bzotry setting oc_freq_khz /=2 right after the code you commented out, and echo with 60000
01:12.38ToAsTcfhspeed->a11clk_src_div = (speed->a11clk_src_div + 2) /2 - 1;
01:12.38ToAsTcfha11clk_khz_new = oc_freq_khz / (speed->a11clk_src_div + 1);
01:12.38ToAsTcfh
01:12.40ToAsTcfh?
01:14.17bzoactually, do this instead, right before the code you commented out:
01:14.26bzowritel(oc_freq_khz / 19200, MSM_CLK_CTL_BASE+0x33c);
01:14.32bzochange to
01:14.38bzowritel(oc_freq_khz * 2/ 19200, MSM_CLK_CTL_BASE+0x33c);
01:15.20bzothen don't worry about any doubling, set the oc_freq_khz to the actual freq you want, as well as your echo thing
01:16.56bzobbl
01:24.55Hoochsterevening
01:26.29AstainHellbringevening Hoochster
01:26.49Hoochsterwhatup
01:27.18AstainHellbringjust chillin
01:28.59Hoochsterme as well heh started to weed eat the back yard after work heh nowhere near done lol
01:33.13*** join/#htc-linux br1ck_ (~br1ck@xdslca014.osnanet.de)
01:35.01*** join/#htc-linux toastcfh_ (~chatzilla@24.94.158.107)
01:46.02*** join/#htc-linux Roguefoxx (~roguefoxx@99-28-135-147.lightspeed.cktnmi.sbcglobal.net)
01:51.21*** join/#htc-linux mastermerlin_ (~merlin@pD957FBF9.dip.t-dialin.net)
01:52.00*** join/#htc-linux randomblame (~randombla@c-71-59-184-96.hsd1.wa.comcast.net)
01:56.49*** join/#htc-linux DaveWM (dave@h-96-13-58-197.ip.alltel.net)
01:58.28*** join/#htc-linux BHSPitMonkey (~stephen@unaffiliated/bhspitmonkey)
02:07.39*** join/#htc-linux jamenlang2 (~jamenlang@68-170-60-12.mammothnetworks.com)
02:09.58ToAsTcfhwhoa chatzilla is massive
02:11.51randomblamemaybe that's why they call it chatzilla rather than chat... I got nothin
02:12.36jamenlang2chatsqueak
02:12.46randomblamebang that's it
02:12.58randomblameI need coffee
02:13.07jamenlang2does anyone have an easy way to contact GlemSom?
02:13.20jamenlang2i assume glemsom@anapnea.net
02:15.54ToAsTcfhlol
02:16.28ToAsTcfhi lke it but it looks like its for blind people
02:16.42ToAsTcfhidk could be just me
02:17.53*** join/#htc-linux Kawaiius (~yadayada@pool-72-84-136-77.slsbmd.east.verizon.net)
02:22.33*** join/#htc-linux unitypunk (~unitypunk@241.sub-75-208-33.myvzw.com)
02:27.50*** join/#htc-linux enjalot (~enjalot@c-98-230-50-70.hsd1.fl.comcast.net)
02:39.59*** part/#htc-linux |Jason8| (~JayAte@server2.tiltshellz.org)
02:59.25*** join/#htc-linux the_sys0p (~the_sys0p@116.199.69.35)
03:28.48*** join/#htc-linux Termana (~bradley@123-3-85-159.static.dsl.dodo.com.au)
03:34.57*** join/#htc-linux Kristian21x (~trjmkztdg@pool-173-63-172-239.nwrknj.fios.verizon.net)
03:53.06*** join/#htc-linux swc|666 (~infidel20@unaffiliated/swc666/x-4934821)
03:58.25*** join/#htc-linux randomblame (~randombla@c-71-59-184-96.hsd1.wa.comcast.net)
04:11.33*** join/#htc-linux rashire (~ed1112war@pool-98-114-206-111.phlapa.fios.verizon.net)
04:12.00*** join/#htc-linux ALoGeNo (~314@127.Red-83-63-196.staticIP.rima-tde.net)
04:12.00*** join/#htc-linux ALoGeNo (~314@unaffiliated/alogeno)
04:12.04*** join/#htc-linux luminoso (~lumos@av-217-129-131-108.netvisao.pt)
04:13.52*** join/#htc-linux goxboxlive (~jrs@mail2.hjellnesconsult.no)
04:17.46TermanaAnyone know what I need to add to CONFIG_CMDLINE to get the framebuffer console to activate on the Nexus One?
04:18.29tmzt_yeah, it might be eaiser here
04:18.37tmzt_the problem is fbcon is normally disabled
04:18.49tmzt_so you need to change .config as well, not just the cmdline
04:19.26tmzt_CONFIG_FRAMEBUFFER_CONSOLE=y
04:19.30tmzt_CONFIG_VT=y
04:19.40Termanayeah, thats fine. I noticed that the normal framebuffer console options don't show up in menuconfig in N1 trees for some reason
04:19.45tmzt_I think some people here have already run X on snapdragon
04:20.18TermanaI guess the fact that CONFIG_VT might be disabled, and they only show up if its enabled
04:20.55TermanaSo what exactly do I need to add to CONFIG_CMDLINE to just get a framebuffer console to show up?
04:21.06TermanaOr do I need to add anything at all?
04:21.36tmzt_no, CONFIG_VT doesn't really have anything to do with it
04:21.43tmzt_but you need to activate them for it to be usable
04:22.15Termanatmzt_: Maybe not, but looking through the correct kconfig, the fbcon options won't show unless VT is enabled
04:22.25tmzt_hmm
04:22.26tmzt_maybe
04:22.39tmzt_just activate both of them and build the kernel
04:22.55Termanano modifications to cmdline?
04:23.03tmzt_probably not
04:23.10tmzt_unless you want to rotate it
04:23.37*** join/#htc-linux luminoso_ (~lumos@av-217-129-131-108.netvisao.pt)
04:25.37Termanatmzt_: Thanks :) Building now. The reason I wanted a framebuffer console is because I've rebased cyanogen's kernel to .34-rc2, and its not booting. I don't have a N1 yet (as you know from our conversation the other day), but I'm getting people to test it so hopefully they will tell me what the errors are.
04:25.58tmzt_what are you porting?
04:26.03*** join/#htc-linux luminoso_ (~lumos@av-217-129-131-108.netvisao.pt)
04:26.14*** join/#htc-linux DaveWM (~dave@h-96-13-67-251.ip.alltel.net)
04:32.44Termanatmzt_: What am I porting? Look in my last message :P I'm not porting anything particularly or userspace wise if thats what you mean. I'm just rebasing the kernel to 34-rc2, which was the latest when I started, now its 34-rc3 or 34-rc4
04:33.13tmzt_yeah, what os
04:34.10Termanatmzt_: Android I guess. I'm not coupling the kernel with any userspace, I just assume that everyone thats testing it is using Cyanogen's userspace.
04:34.28*** join/#htc-linux luminoso_ (~lumos@av-217-129-131-108.netvisao.pt)
04:34.32TermanaI'm just providing the kernel for flashing with fastboot
04:36.37Termanatmzt_: Do I need to change the console= part to console=ttyS0 or tty0 or leaving it as ttyMSM should be fine?
04:39.45tmzt_add either of those
04:40.04tmzt_I think that's part of how google restricts console access
04:40.13tmzt_(see the first g1 and our early comments here )
04:42.41Termanatmzt_: So thats a yes to change ttyMSM to tty0?
04:44.17tmzt_yeah
04:56.50tmzt_http://www.linuxfordevices.com/c/a/News/TI-WiLink-60-WL1271/?kc=rss
04:59.08*** join/#htc-linux DaveWM (dave@h-98-134-22-115.ip.alltel.net)
05:01.31*** join/#htc-linux Kawaiius (~yadayada@pool-71-255-177-18.slsbmd.east.verizon.net)
05:17.00*** join/#htc-linux kiozen (~oeichler@rgnb-5d87c5f8.pool.mediaWays.net)
06:01.54*** join/#htc-linux fish1209 (~fish1209@unaffiliated/fish0912)
06:02.58*** join/#htc-linux goxboxlive (~jrs@mail2.hjellnesconsult.no)
06:05.32*** join/#htc-linux moishaspb (~chatzilla@user-12hdsgv.cable.mindspring.com)
06:07.28*** join/#htc-linux hes80 (~Stephan_H@188.104.243.170)
06:18.23*** join/#htc-linux tfried2001 (~tfried200@82.211.188.186)
06:31.13*** join/#htc-linux polyrhythmic (~charles@c-67-171-16-39.hsd1.wa.comcast.net)
06:37.41*** join/#htc-linux Vito_| (~Sysel@gw.loccal.net)
06:39.18*** join/#htc-linux polyrhythmic (~charles@c-67-171-16-39.hsd1.wa.comcast.net)
06:43.56*** join/#htc-linux kiozen (~oeichler@p54920F0B.dip0.t-ipconnect.de)
06:45.37*** part/#htc-linux hes80 (~Stephan_H@188.104.243.170)
06:54.16*** join/#htc-linux DaveWM (dave@h-98-134-43-213.ip.alltel.net)
07:05.03*** join/#htc-linux mickey|office (~Mickey@dialbs-092-079-168-007.static.arcor-ip.net)
07:12.41*** part/#htc-linux tfried2001 (~tfried200@82.211.188.186)
07:14.47*** join/#htc-linux leobaillard (~leobailla@leobaillard.org)
07:18.11*** join/#htc-linux Markinus (~Miranda@gtng-4db049d0.pool.mediaWays.net)
07:23.12*** join/#htc-linux makkonen (~makkonen@cpe-66-69-229-9.austin.res.rr.com)
07:24.54*** join/#htc-linux makkonen (~makkonen@cpe-66-69-229-9.austin.res.rr.com)
07:30.40*** join/#htc-linux GNUtoo (~GNUtoo@host137-158-dynamic.54-79-r.retail.telecomitalia.it)
07:33.50*** join/#htc-linux zachy (~george@ip-85-160-21-202.eurotel.cz)
07:41.47*** join/#htc-linux Captnoord (~Captnoord@145.74.218.14)
07:42.37*** join/#htc-linux luminoso (~lumos@av-217-129-131-108.netvisao.pt)
08:09.58*** join/#htc-linux johnb81 (~john@88.226.97.204)
08:23.39*** join/#htc-linux kvaster (~kvaster@93.84.112.80)
08:30.49*** join/#htc-linux m3dlg (~m3dlg@212.183.140.51)
08:51.57*** join/#htc-linux Vito_| (~Sysel@195.113.241.225)
08:55.45*** join/#htc-linux dcordes (~dccordes@unaffiliated/dcordes)
08:56.26*** join/#htc-linux zenity (~polymod@general-ld-216.t-mobile.co.uk)
09:09.27*** join/#htc-linux DaveWM (dave@h-98-134-12-38.ip.alltel.net)
09:21.50*** part/#htc-linux jumoit (~samuel@121.77.87.145)
09:47.21dcordeshi
09:47.24dcordesMarkinus, what was the exact command you used to pull in+merge ?
09:47.37Markinusdcordes: hi
09:47.39Markinusmoment
09:47.52Markinusgit pull git://android.git.kernel.org/kernel/msm.git  android-msm-2.6.32
09:47.59Markinusin our .32 repo
09:48.13dcordesthen it gives the error
09:48.18dcordesnot ff
09:48.38dcordesdid you put no parameter?
09:49.05Markinusand you are in our branch htc-msm-32 ?
09:49.27Markinusdcordes: only this paras above
09:50.19dcordesMarkinus, ok right.. now I remember. It actually worked but I canceled it because of the low speed..
09:50.22dcordesnevermind
09:51.26GNUtooleviathan, hi
09:51.47GNUtoothanks dcordes in new xf86-video-msm gtk issue seem fixed
09:52.08dcordesGNUtoo, :D awesome!
09:52.20dcordesGNUtoo, can you update OE and add the headers ??
09:52.27GNUtooI'll do that
09:52.30dcordesnice
09:52.56dcordesMarkinus, let's try what NetRipper proposed and make a minimal kernel
09:53.15Markinusdcordes: I did it already
09:53.21Markinusdcordes: 2 mb
09:53.37Markinusdcordes: but same signal problem
09:53.40dcordesok I didn go through the logs yet.. crap
09:53.58Markinusdcordes: but only with a initrd
09:54.25dcordesyou mean like you didn't try with rootfs on SD ? what about the rootdelay=n test ?
09:54.42Markinusdcordes: I disabled mmc too
09:54.45Markinusdcordes: no sd
09:55.00Markinusdcordes: oput rootdelay 10 in cmd
09:55.07dcordesfreezes?
09:55.16Markinusdcordes: if I have no initrd
09:55.21Markinusit waits 10 sekonds
09:55.32Markinusdcordes: then crash because nothing to bott. is ok
09:55.37Markinus*boot
09:55.46dcordesok well it crashes for me always
09:55.49dcordeswith normal kernel
09:55.53GNUtoodcordes, still doesn't xrandr -o 1 tough,I think we'll need to add it,but I was told it was easy
09:56.05Markinusdcordes: if I have a initrd . .(tested more of them) signal problem after freeining init memory
09:56.25Markinusdcordes: ot it's hanging after them, depend of the kernel konfig
09:56.33dcordesGNUtoo, what does #xorg say about -o ?
09:57.41leviathanGNUtoo: hi
09:57.50leviathansry, was away, taking a shower
09:58.01dcordesMarkinus, I have some openmoko rootfs that always goes a bit further
09:58.05leviathandrinking coffee and smoking a cigaret
09:58.08leviathannow I'm back
09:58.16dcordesMarkinus, freezes on udev most of the time
09:58.18GNUtoodcordes, I'll look in the logs
09:58.28leviathanGNUtoo: wuhuu
09:58.29leviathanyeahh
09:58.44dcordesGNUtoo, I mean specific to msm. maybe they can help directly
09:58.46Markinusdcordes: without initrd so signal problems .. strange
09:58.59GNUtoodcordes, I'll look in my logs
09:59.01dcordesMarkinus, is it possibly memory (mapping) related ?
09:59.10dcordesGNUtoo, gotcha
09:59.12Markinusdcordes: maybe . .
09:59.33GNUtooFeb 11 19:57:36 <jbarnes>       GNUtoo: you have to implement the create/allocate/destroy shadow hooks
09:59.35dcordesMarkinus, maybe try with different sized initrds
09:59.45dcordesMarkinus, or we make initrd that testwrites mem
09:59.50Markinusdcordes: i tried smal and big . .allways the same
09:59.50GNUtooFeb 11 20:01:29 <CosmicPenguin> GNUtoo: the MDP blit can do that rotation, so you are at least covered on the HW acceleration front
09:59.51dcordesMarkinus, what about memtest ?
10:00.04Markinuswince is working good . .hmm
10:00.14GNUtooMPD is xf86-video-msm specific
10:00.33dcordesMarkinus, I mean like something that writes to all available ram to exclude mapping problem
10:01.28Markinusdcordes: hmm, yes . .this could be a test . .I'Ve in the moment 88mb. . I will try 64
10:02.38Markinusor 32 . . but maybe yes. . .. but we would need some stuff what du such test
10:02.44Markinus*do
10:02.51Markinus*does )
10:03.14dcordeshttp://www.arm.linux.org.uk/developer/stresstests.php
10:05.45dcordesdead link
10:06.04dcordeshttp://www.arm.linux.org.uk/mailinglists/ where to register there??
10:07.30dcordeshttp://www.arm.linux.org.uk/mailinglists/lists.php
10:11.53*** join/#htc-linux leaigor (~laigor@188.134.16.241)
10:15.32GNUtoobtw here's the patch: http://pastebin.com/zmx4yZbi
10:15.38GNUtooI'll commit in oe soon
10:15.48GNUtoo(compiling)
10:16.15GNUtoo(compiling midori so can't test if the recipe I didn't modify yet is fine)
10:16.33GNUtoo(I built copying the git dir and running the temp script)
10:22.08leviathanGNUtoo: the new xf86-video-msm fixes gtk?
10:22.23leviathanhow has the recipe to look like?
10:26.47*** join/#htc-linux metsie (~metsie@212-127-230-46.cable.quicknet.nl)
10:41.05*** join/#htc-linux balans (~Gebruiker@82-170-215-5.ip.telfort.nl)
10:53.18leviathanGNUtoo: [  331.681701] wl1251_sdio mmc0:0001:1: firmware: requesting wl1251-fw.bin
10:53.28leviathan[  391.688171] wl1251: ERROR could not get firmware: -2
10:54.20phh-E = ENOENT
10:54.31phh-2*
10:56.06phhaka copy the firmware in /lib/firmware/wl1251-fw.bin
10:58.17leviathanphh: don't you think I didn't -.-
10:58.29leviathanroot@htcdream ~ $ ls /lib/firmware/
10:58.29leviathanwl1251-fw.bin   wl1251-nvs.bin
10:58.47phhthen your distrib put firmwares somewhere else than /lib/firmware ?
10:58.47leviathanbut it doesnt load it somehow...
10:58.55leviathanhmm
10:58.58leviathannope?
10:59.15phhudev is started ?
11:01.08leviathanI could try to reboot
11:01.09leviathan...
11:01.19leviathanand then try again...
11:01.43leviathansometimes reboot solves a lot of problems, even if its UNIX
11:01.46leviathan;-)
11:03.52leviathanphh: oh
11:03.55leviathanno
11:04.07leviathanfuck, I know why its not working >_<
11:04.34leviathanon which bus it wants to write the fw
11:04.48phhcould not ***get*** the firmware, not write
11:04.58leviathanand on which bus he tries to read the file simultanously if I have installed the rootfs on sd-card?
11:05.05leviathan>_<
11:05.06phhno it's not the problem
11:05.13phhit's too different busses
11:05.20leviathansure?
11:05.22leviathanok
11:05.24phhtotally sure
11:05.26leviathanthen its strange
11:05.46phhyou're totally sure it's /lib/firmware the correct path for you ?
11:05.56NetRipperMarkinus, i think the problem may be caused by timer.c making use of the smsm_*() functions
11:06.00NetRipperwhich aren't working
11:06.18NetRipper(implemented in smd.c)
11:07.07leviathanyes
11:07.18leviathanphh: http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/firmwares/marvell-sdio-fw_9.70.3-p37.bb
11:07.25leviathandoes it exactly the same way
11:07.26leviathan:)
11:08.17NetRipperMarkinus, if i remove smd from board-htcleo.c initialization, the kernel doesnt seem to hang, but instead it causes an enormous amount of errors from timer.c as all smsm() actions timeout
11:08.39phhleviathan: trace udev then
11:08.50leviathanhow?
11:08.51leviathan:)
11:08.57phhhaha
11:08.58phhno clue.
11:09.29phhoh and iirc there are two different ways to send a firmware file
11:10.23leviathanphh: hmm
11:10.27leviathanI found udevadm
11:10.36phhhum, only used in the sources
11:10.41phhweird, I thought I'd seen two
11:10.46leviathanhttp://pastebin.com/h1ew7gp2
11:11.48phhah no, it's nvs which has two ways
11:12.10phhor only one.
11:12.12phhwhat's this code ?
11:18.30leviathanudevadm
11:18.32leviathan:)
11:18.32*** join/#htc-linux the_sys0p (~the_sys0p@116.199.69.35)
11:18.34leviathanthe trace
11:18.59GNUtooleviathan, for firmware, udev sys nodes meant for firmware don't even appear
11:19.20GNUtoo/sys/class/firmware/foo doesn't appear
11:19.33*** join/#htc-linux Wout (~wout@internet.office.tweakers.net)
11:19.48*** join/#htc-linux gauner1986 (~gauner@ip-145-93-244-217.fontys.nl)
11:19.55Woutphh: little bugreport, charging isn't working with the latest build
11:19.58gauner1986hi
11:20.08phhWout: it is.
11:20.13phhjust as usual, you must be charging when booting
11:20.24WoutI know
11:20.36Woutis it only wall charging, or should charging through usb work too?
11:20.46phhno clue
11:20.50Woutbecause my phone was connected to the pc and charging in winmo
11:20.56Woutthen I booted android
11:21.25leviathanGNUtoo: hmm
11:21.33leviathanperhaps the kernel doesnt know firmware?
11:21.40leviathanand its a configuration issue?
11:21.40phhWout: charging for >10 minutes, and not fully charged ?
11:21.40GNUtooleviathan, it does
11:21.52GNUtooleviathan, look into docs for firmware
11:22.06Woutphh: the pone had about 80% left, I hooked it to usb, was charging in winmo, then booted android
11:22.11Woutnow 5 hours later
11:22.14Woutand am at 40%
11:22.22phhWout: booted, just after you hooked usb ?
11:22.37Woutfew mins later, why?
11:22.40GNUtooDocumentation/firmware_class
11:22.50phhWout: the charging logic needs several minutes to settle
11:22.58Woutah
11:23.36*** join/#htc-linux GeekLad (~GeekLad@adsl-71-254-21.jax.bellsouth.net)
11:40.54phh[ 2157.670623] suspend: enter suspend(2010-04-05 01:54:30.950000000 UTC)
[ 2158.606140] suspend: exit suspend, ret = 0 (2010-04-05 15:06:49.390000000 UTC)

11:40.55phh<3
11:41.03phhchamonix: see that, and be jealous ! :p
11:41.23phh(hint: no sim in a gsm-only phone.)
11:41.59leviathanGNUtoo: xorg now works again
11:42.02leviathanfixed xorg.conf
11:42.13GNUtoook
11:42.19GNUtooit was simple
11:42.21phhleviathan: with xf86-video-msm ?
11:42.23GNUtoo1->2
11:42.27leviathanyes
11:42.31GNUtoo0->1
11:42.32phhI need to try that
11:43.12leviathan:)
11:43.46leviathanGNUtoo: I heard with the newest xf86-video-msm version the gtk issue got fixed??
11:43.50leviathanis that true?
11:44.12leviathanthen we need a recipe for this one
11:44.37GNUtooleviathan, yes it's true
11:44.41GNUtoopatch is here:
11:44.49GNUtoohttp://pastebin.com/zmx4yZbi
11:45.07GNUtooleviathan, I need to finish a compilation before making the recipe
11:46.22leviathanok
11:46.25leviathanvery well
11:48.18ToAsTcfhphh:i still wasnt able to get it working
11:53.19phhI_love_noobs++;
11:53.35phh"why doesn't my earphone speaker work while I added force_headset=2 ? OMFG."
11:54.06phhToAsTcfh: what's the current status ?
11:54.33leviathanmickey|office: heyo
11:54.36leviathana question
11:54.50leviathanwhere gets firmware typically installed in OE?
11:54.54leviathan/lib/firmware
11:54.56leviathanright?
11:55.15leviathanbecause it doesnt recognize the firmware file, when I install it there >_<
11:55.28*** join/#htc-linux Snipey (~home@53524E9B.cable.casema.nl)
11:56.00phhleviathan: is there an image somewhere I could try ?
11:56.08leviathanuhm, yes
11:56.18leviathanhttp://people.ee.ethz.ch/~davidlan/downloads/OE/htcdream/images/htcdream/
11:56.24leviathanI'll build a new one now
11:56.45leviathannope not now
11:57.01leviathanfirst GNUtoo needs to make a new recipe with its patch
11:57.06leviathanthen I will build a new image
11:57.07leviathan:)
11:57.17leviathanso you can try this one
11:57.22phhok
11:57.27leviathangui doesnt start there
11:57.32leviathanbecause xorg.conf is broken
11:57.43phhdon't care about gui :p
11:57.45leviathan/dev/fb needs to be /dev/fb0 in order to start
11:57.49leviathanin xorg.conf
11:57.54leviathanthen it will give you gui
11:58.00leviathanbut gtk is still broken
11:58.09leviathanthats what GNUtoo will fix with his patch
11:59.11phhuser ? password ? ssh on usbnet ?
11:59.56leviathanno password :)
12:00.02leviathanjust enter
12:00.03*** join/#htc-linux julcel (~julcel@2a01:e35:2f3d:a720:224:23ff:fe03:c92f)
12:00.04leviathanuser root
12:00.25leviathanoh
12:00.36leviathanand you need to do ifup usb0 on the phone first
12:00.38leviathan...
12:00.55phhok
12:00.55leviathanthis release is so what of brolen
12:00.57leviathan>_<
12:01.01leviathan*broken
12:01.09phhwell, knowing the problems is enough :p
12:01.27leviathanbtw: why is there no recipe for wxmaxima and maxima in OE?
12:01.41phhmaxima ?
12:01.46leviathanyes
12:01.53leviathanthis clisp thing
12:02.00phhdoesn't help much :p
12:02.17leviathanit would help me  a lot in the lectures ;-P
12:02.33leviathanalso gnuplot would be nice
12:02.34leviathan:)
12:02.48leviathanI'll care about later
12:02.56phhI'd like maple, since that's what I know how to use
12:02.59phhbut on arm... errr
12:03.01phhso I use ssh.
12:03.03leviathanas soon as we have wifi, sound, graphics(+rotation)
12:03.06phhI'll maybe try NX later
12:03.20leviathanhmm
12:03.30leviathanphh: it took a long shot some time ago
12:03.47leviathanuntil I found people who where interested too to get clisp running on ARM
12:03.52leviathanmainly my zaurus
12:03.54leviathan>_<
12:05.13leviathanbut it never made it in mainstream, like I see
12:06.03leviathanI guess I will send in some patches for these math-tools, as soon as we got the basic stuff workin'
12:06.32leviathanok for now
12:07.56phhnow, 256MB to send on my phone at less than 1MB/s
12:07.57phhZZZZzzzzzzz
12:13.16*** join/#htc-linux GNUtoo (~GNUtoo@host137-158-dynamic.54-79-r.retail.telecomitalia.it)
12:15.15phhargh, I forgot I have no keyboard on this phone -_-'
12:19.08phhleviathan: what can I do to make it automatically enable usb0 at boot ?
12:22.45leviathanhmm
12:22.53leviathanphh: first it did
12:23.11*** join/#htc-linux lkcl (~lkcl@nat65.mia.three.co.uk)
12:23.13leviathanbut someone or somewhat broke the concerning init and rc stuff
12:23.26leviathanits a bb-issue I fear
12:23.36leviathanand frankly I don't know exactly ^^"
12:23.45phhok
12:26.08leviathanGNUtoo: it loads the firmware after a while
12:26.15leviathanand even gets an IP
12:26.21leviathanbut... I cant bring it active
12:26.22leviathan...
12:26.31phhoO
12:26.34leviathanifconfig: SIOCSIFFLAGS: Unknown error 132
12:27.10phhhum I don't know this one
12:27.11GNUtoorfkill
12:27.16phhoh right.
12:27.24leviathanhow can we fix this issue?
12:27.37phhecho 1 > /sys/class/rfkill/rfkillX/state ?
12:27.47phhI guess X is 1
12:30.17leviathanhmm
12:30.20leviathanyes
12:30.23leviathannow it hangs again
12:30.30leviathanbecause its requesting firmware
12:30.31leviathan...
12:30.33leviathanhmm
12:30.40phhhaha.
12:30.44leviathansomeone is some what of broken...
12:30.51leviathanifconfig: SIOCSIFFLAGS: No such file or directory
12:31.01leviathan/someone/something/
12:31.20leviathanonly question: WTF?
12:31.25phhhehe.
12:31.38phhI had no such problems on ubuntu 8.04 :p
12:33.10GNUtooleviathan, how did you get it load the firmware?
12:33.23leviathanI didnt
12:33.26leviathanjust timed out
12:33.35GNUtooah dmesg?
12:33.37leviathanthe IP was a random ip
12:33.40leviathan>_<
12:33.51GNUtooavahi?
12:34.00leviathan[ 2838.273406] wl1251: ERROR could not get firmware: -2
12:34.01leviathanyes
12:34.03leviathanseems
12:34.05GNUtoonice
12:34.09GNUtoowow
12:34.13GNUtoohow do you have this error
12:34.23GNUtooI know how to inject firmware manually
12:34.35GNUtoobecause I read the doc
12:34.44GNUtooDocumentation/firmware_class
12:35.25GNUtooleviathan, if it requires a firmware maybe you get futhurer
12:36.54*** join/#htc-linux majster (~majster@cpc12-aztw25-2-0-cust68.aztw.cable.virginmedia.com)
12:37.46GNUtoomoreover there are 2 firmwares
12:37.51GNUtooone is the firmware
12:38.01GNUtooand the other is the nvs calibration files
12:38.23GNUtooleaigor, ^^^
12:38.25GNUtoooops
12:38.32GNUtooleviathan, ^^^
12:38.46GNUtooleaig[or] sorry
12:38.52leviathannp
12:38.54leviathan:)
12:38.56leviathanuhm
12:39.01leviathanfollowing thing
12:39.09leviathanhow do you inject the firmware manually?
12:39.17GNUtooleviathan, like they say in the doc
12:39.22GNUtooDocumentation/firmware_class
12:39.25leviathanok
12:40.40GNUtooleviathan, how did you get it to want to load a firmware?
12:40.49GNUtoobtw I should commit xf86-video-msm
12:42.00phhGNUtoo: just load msm_wifi and ifconfig wlan0 up
12:42.03phhtheorically.
12:42.23GNUtoothat didn't work for me
12:42.28GNUtooifconfig wlanN up hanged
12:42.47phhwell, normally it timesout after 15s saying ENOENT
12:43.32GNUtooit doesn't,it stays blocked
12:43.37GNUtooethernally
12:43.49phh<PROTECTED>
12:43.57phhnothing indmesg ?
12:44.03GNUtooand sdio transfer never completes
12:44.12GNUtooyes sdio command 53 never completes
12:44.35GNUtoobut maybe leviathan got something because it doesn't even asked for the firmware for me
12:44.51leviathanGNUtoo: its because I have mmc debug active
12:44.59leviathanso it tells me everything its doing
12:45.10leviathanand additionally you can monitor with udevadm
12:45.11leviathan:)
12:45.18GNUtoo<leviathan> [ 2838.273406] wl1251: ERROR could not get firmware: -2 is not mmc debugging
12:45.23leviathanyes
12:45.26leviathanexactly
12:45.27leviathan:)
12:45.33leviathanhmm??
12:45.35leviathanoh
12:45.37GNUtoothat's not mmc debugging and doesn't seem userspace
12:45.38leviathannya
12:45.44leviathanhmm
12:45.46leviathanstrange
12:45.54leviathanit appears, if you activate mmc debugging
12:45.54leviathan:)
12:45.56*** join/#htc-linux Orkspalter (~Orkspalte@p54B58CB9.dip0.t-ipconnect.de)
12:46.02GNUtooit didn't for me
12:46.07leviathanhmm, strange
12:46.33GNUtoomaybe pastebin me your defconfig
12:46.42leviathancan you please give me the steps with which you have inserted the firmware?
12:46.43leviathansure
12:47.22leviathanhttp://pastebin.com/jMKv7AH6
12:49.45*** join/#htc-linux m3dlg (~m3dlg@212.183.140.22)
12:49.54GNUtooleviathan, I didn't
12:50.01GNUtoothe nodes for firmware didn't appear
12:50.20leviathanyes
12:50.27leviathanthats exactly my problem too
12:50.56leviathanit says, that it would be created trough request_firmware
12:51.07leviathanperhaps something has changed...
12:51.45GNUtooleviathan, let me check something
12:51.54leviathankk
12:58.04GNUtoo[ 5499.201599] wl1251_sdio mmc0:0001:1: firmware: requesting wl1251-fw.bin
12:58.05GNUtooindeed
13:00.45GNUtoo[ 5667.030639] wl1251: firmware booted (Rev 4.0.4.3.5)
13:00.46GNUtoomanually
13:01.06phhhehe
13:01.07GNUtooyou have to be fast tough
13:01.09phhgreat :p
13:01.16GNUtoowlan0 seem up
13:01.21GNUtoolet me install wireless-tools
13:01.24phhyeah no it should just work
13:01.27phhniw*
13:01.28phhnow*
13:01.37*** join/#htc-linux Vito_| (~Sysel@195.113.241.225)
13:02.58*** join/#htc-linux mickey|zzZZzz (~mickey@80.81.242.146)
13:03.23*** join/#htc-linux k_linux (~klinux@41.214.195.247)
13:05.39leviathanGNUtoo: yes
13:05.42leviathanit works
13:06.06GNUtooyou got it working ?
13:06.07GNUtooalso
13:06.50phhso your udev is broken ..?
13:07.05GNUtoopossible
13:07.14GNUtooit seemed broken at boot
13:07.14phhwhat's not broken ? p
13:07.17GNUtootimeout 3s
13:07.22GNUtooand a long list of path
13:08.04GNUtoophh, the android we try to replace(just kidding,not even true)
13:08.17phh:p
13:16.42phhdon't want no udev ?
13:17.14leviathanhmm
13:17.26leviathanI'll look if the driver does some things
13:17.37leviathanwhich are not appropriate with 2.6.32 anymore
13:18.06mickey|officecould be
13:19.14k_linuxHI there
13:22.20GNUtooleaigor, did it associate
13:22.27*** join/#htc-linux Captnoord (~Captnoord@145.74.218.14)
13:22.31phhk_linux: so everything works now ?
13:22.34GNUtooleviathan, did it associate
13:22.34phhhey Captnoord
13:22.37Captnoordphh yo
13:22.45Captnoorddiamonds panel code works on my raphael
13:22.46GNUtoomickey|office, udev 151 seem broken
13:22.46leviathanGNUtoo: nope
13:22.51GNUtooneither for me
13:22.54Captnoordbesides it has the exact same lcd display
13:22.59Captnoordonly needed to change a bit in the init
13:23.04phhCaptnoord: and you're able tot totally shut off the mddi link ?
13:23.11k_linuxphh: not yet I still have issues running wifi
13:23.23CaptnoordI just copy pasted the entire panel and stuff...
13:23.28k_linuxeven if many users confirm that is working ! :)
13:23.30Captnoorddouble checked the spi stuff
13:23.38Captnoordchanged some small things
13:23.40Captnoordand it works
13:23.48phhCaptnoord: but do you shut off mddi clock ?
13:23.51Captnoordbut only problem is that raphael has more panel types
13:23.53Captnoordhmmmm
13:23.56Captnoordgood questions
13:23.59Captnoord-s
13:23.59mickey|officeGNUtoo: do you know how this process works?
13:24.07phhCaptnoord: ask your battery meter ? :p
13:24.07Captnoordshouldn't that be done in the mddi.c?
13:24.10Captnoordlol
13:24.13mickey|officei'd write a firmware loader userland helper then
13:24.14phhno clue
13:24.15Captnoordat school atm
13:24.17phhok
13:24.19GNUtoomickey|office, what dhcp+authentification?
13:24.20mickey|officewant to avoid udev
13:24.25phhand you don't have your geek stuff at school ? :p
13:24.40mickey|officeGNUtoo: no, firmwareloading via udev
13:24.45Captnoordphh yea we do
13:24.45GNUtoomickey|office, really? can't we just fix udev or workarround?
13:24.51GNUtoomickey|office, there are docs in kenrel:
13:24.53mickey|officewell
13:24.54Captnoordbut I don't take my battery rig around
13:24.56mickey|officei don't want udev on my system
13:25.00mickey|officebut feel free to fix it
13:25.04mickey|office;)
13:25.14GNUtooDocumentation/firmware_class/
13:25.19GNUtoomickey|office, we can keep udev
13:25.21mickey|officeok, i will read that when i'm back home
13:25.28GNUtooit's short
13:25.32GNUtoobut read when you're home
13:25.38GNUtooelse I can paste here
13:25.40GNUtooit's so short
13:26.13phhCaptnoord: btw, my diamond had a 13hours long sleep ! :p
13:26.28GNUtoocd /sys/class/firmware/xxx/
13:26.31mickey|officemy dream always wakes up...
13:26.34phhwithout a SIM card in it to wake it up, it's easy thouogh.
13:26.35mickey|officedoesn#t want to sleep
13:26.37mickey|officeworking bee :)
13:26.39GNUtooecho 1 > loading
13:26.51GNUtoocat ../firmware.fw > data
13:26.55GNUtooecho 0 > loading
13:26.56CaptnoordI commited a codeaurora patch regarding sd card shit
13:27.02Captnoordwakelock crap
13:27.02GNUtoothen re-do for calibration
13:27.03phhCaptnoord: i've seen
13:27.06mickey|officeGNUtoo: awesome. that's simply enough to write an fsodeviced plugin
13:27.10GNUtoobtw leviathan I suspect bad calibration
13:27.11phhbut i've never seen this wake lock causing any problem
13:27.26Captnoordnah its because it wasn't named obviouse
13:27.33Captnoordat least it didn't cause problems
13:27.43Captnoordbut now it doesn't lock unless it really is needed
13:27.43GNUtooApr  6 15:04:57 router hostapd: wlan0: STA 00:1f:df:31:b9:8d IEEE 802.11: authenticated
13:27.43GNUtooApr  6 15:04:57 router hostapd: wlan0: STA 00:1f:df:31:b9:8d IEEE 802.11: associated (aid 3)
13:27.44GNUtooApr  6 15:04:57 router hostapd: wlan0: STA 00:1f:df:31:b9:8d RADIUS: starting accounting session 4BA21E4D-000003A5
13:27.44GNUtooApr  6 15:05:07 router hostapd: wlan0: STA 00:1f:df:31:b9:8d IEEE 802.11: authenticated
13:27.44GNUtooApr  6 15:05:07 router hostapd: wlan0: STA 00:1f:df:31:b9:8d IEEE 802.11: associated (aid 3)
13:27.45GNUtooApr  6 15:05:07 router hostapd: wlan0: STA 00:1f:df:31:b9:8d RADIUS: starting accounting session 4BA21E4D-000003A6
13:27.48GNUtooApr  6 15:05:40 router hostapd: wlan0: STA 00:1f:df:31:b9:8d IEEE 802.11: authenticated
13:27.50GNUtooApr  6 15:05:40 router hostapd: wlan0: STA 00:1f:df:31:b9:8d IEEE 802.11: associated (aid 3)
13:27.52GNUtoooops
13:27.54GNUtooagain
13:27.56GNUtoodidn't took the address
13:27.58GNUtoobut the content
13:28.11GNUtoohttp://pastebin.com/74qHdRnk
13:28.38leviathanhmm
13:28.39leviathancool
13:28.45leviathanseems to come up
13:28.52leviathanso we have the basics
13:28.54Captnoordbut 13 hour long sleep?
13:28.59leviathannow we need to make it run smoothly
13:29.00leviathan:)
13:29.10phhbtw, the MAC address is totally random for the lower 24 bits
13:29.13phhhigher 24bits is nokia
13:29.24phhCaptnoord: bah the phone has no reason to wake up
13:29.29Captnoordtrue....
13:29.45phhit still sounds impressive :p
13:29.52phheven if it's "logical" it doesn't wake up
13:29.59AstainHellbringmorning
13:30.00Captnoord?
13:30.01Captnoord:P
13:30.03Captnoordlolz
13:30.09Captnoordso it is just dead
13:30.09Captnoordhehe
13:30.19phhCaptnoord: it's a game console :p
13:30.25phhwhich wakes up quickly
13:30.26Captnoordhehehe
13:30.31Captnoordcool
13:31.18Captnoordat least with the swapped panel code my phone is finaly getting 'cold'
13:31.34Captnoordcooler than 25 degrease
13:32.47GNUtoophh, i know...I was told that in the linux wireless channel
13:32.51Captnoordlets build with my pref config file
13:32.58phhGNUtoo: ok
13:33.09GNUtooaltough I didn't know how much bits exactly
13:34.22*** join/#htc-linux android-res (~yciduo@80.179.91.44)
13:35.10*** join/#htc-linux ToAsTcfh (~chatzilla@24.94.158.107)
13:51.12GNUtooI think I needed to keep wpa_supplicant on
13:51.25phhGNUtoo: to keep your wpa connection on ? yeah most likely.
13:51.45GNUtooor maybe it was calibration
13:51.53phhToAsTcfh: hum, can you make me remind of what happens when you start at lower frequency, and only then set higher frequency ?
13:51.54GNUtoophh, I've open access point
13:51.54*** join/#htc-linux lkcl (~lkcl@nat66.mia.three.co.uk)
13:55.14*** join/#htc-linux Cablekevin (~Cablekevi@house.speedxs.nl)
13:55.16CablekevinHi!
14:01.15GNUtoonice it works
14:01.56GNUtoolistenning to radio trough wifi
14:03.31phhhuhu
14:03.31CablekevinWith what :) ?
14:03.48GNUtoomplayer,htcdream
14:03.52GNUtoounder SHR
14:03.55GNUtooin command line
14:04.02CablekevinThats awsome :D
14:04.28CablekevinI was checking out the [PROJ] Linux on HTC LEO thread.
14:04.37CablekevinI like the progress.
14:04.56phhCablekevin: leo is stuck :/
14:06.15CablekevinYeah couldn't figure what was happening at the moment so i thought, lets go to the IRC chan.
14:06.21CablekevinWhats the problem at the moment?
14:06.40CablekevinAnd what happened with the other "Android" thread, were there to much trolls or something?
14:07.13phhtoo many trolls.
14:09.17Cablekevinbah.
14:09.26CablekevinBut there was some dutch guy with an website.
14:09.33CablekevinHow to put Android on your HD2.
14:09.34phhnetripper ?
14:09.57CablekevinNo,
14:10.03Cablekevinhld on, might recall it.
14:10.04phhah.
14:11.30CablekevinCrap, cannot recall it.
14:12.37CablekevinBut how come the progress is stuck?
14:39.21chamonixphh: wow!
14:39.28phhchamonix: :p
14:39.33chamonixHeHeHe
14:43.14chamonixphh: r u thinking of using the xmpp client as a reader?
14:43.35phhchamonix: not a reader, but to notify the rss reader that there is something new
14:44.03chamonixkk.... just wanted to be sure because that are the kind of commands I've started working on
14:44.26chamonixany specific reader u want me to look on to?
14:44.40phhhum no
14:44.49*** join/#htc-linux balans (~Gebruiker@82-170-215-5.ip.telfort.nl)
14:44.50phhI think it's going to be a new reader anyway
14:45.00chamonixlol
14:45.46chamonixwell I'm going to see if I find one opensource as I may have to add a braodcast handler to it allowing to pass the feed that needs to be updated
14:46.28phhI'd prefer the whole protocol being on xmpp, so that the whole RSS hasn't to be read/understood/compared actually
14:46.40phhI'm not sure if that would change power consumption a lot
14:47.59*** part/#htc-linux Orkspalter (~Orkspalte@p54B58CB9.dip0.t-ipconnect.de)
14:48.18chamonixI'll look into XMPP Message type to see if there is a cheap way to encode rss-like items
14:49.38chamonixwhat I still don't like about it is that it's broadcasting..... I'd like to see RSS with bookmarks, tags, stars and read-mark
14:51.22*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
15:08.59*** join/#htc-linux luminoso_ (~lumos@av-217-129-131-108.netvisao.pt)
15:09.26*** join/#htc-linux mastermerlin (~merlin@pD957FBF9.dip.t-dialin.net)
15:10.20*** join/#htc-linux luminoso__ (~lumos@av-217-129-131-108.netvisao.pt)
15:10.54*** join/#htc-linux m3dlg (~m3dlg@212.183.140.23)
15:12.27*** join/#htc-linux hfctsaot (~fircuser@173-137-7-95.pools.spcsdns.net)
15:12.53*** join/#htc-linux hfctsaot (~fircuser@173-137-7-95.pools.spcsdns.net)
15:13.20hfctsaot\o/ I'm here
15:15.21*** join/#htc-linux kvaster (~kvaster@live.bn.by)
15:16.25hfctsaotphh u seen I got the same deal. even after bzo attempted to have me change the multiplier back to 2
15:18.08hfctsaotit seems like its the code all together its not liking
15:18.59phhI had some thought about it
15:19.17phhwell, I did a little test for overclocking wince
15:19.22phhI can do the same thing for android
15:19.45phhI just don't know how to do that with dd mmm
15:20.46hfctsaotis it possibly the table missing that's causing the freeze
15:21.20hfctsaotwell lack of use of the table in pll2 I mean
15:21.34*** join/#htc-linux johnb81 (~john@88.226.97.204)
15:22.14phhget http://husson.hd.free.fr/ttttt
15:22.44phhdd if=tttttt of=/dev/mem bs=1 seek=2824864572
15:22.57phhthis should set pll2 to ~ 1200MHz
15:23.14phhhum
15:23.18phhmaybe not, word vs byte
15:23.45hfctsaotI really don't think its a radio or anything stopping it. it just seems to not pike jumping up using that code for some reaaon
15:23.49phhsame url, bs=4
15:24.22hfctsaotbab ill have to do it when I get home. sorry
15:24.35phh....
15:25.04hfctsaotlol sorry man I'm at work
15:25.57hfctsaotthis shits just been bugging the hell out of me
15:27.18hfctsaotso what have u done in this?
15:30.11phhchange pll2 speed
15:32.33hfctsaotusing bzo patch still or no
15:36.11leviathanGNUtoo: do you have a fix for broken firmware loading now?
15:36.41GNUtooleviathan, what kind of fix?
15:37.17leviathanmanually injecting the firmware file for wifi doesnt seem so
15:37.23leviathanuhm practicable
15:38.26GNUtooleviathan, did you try udev 141?
15:38.57hfctsaotphh why so high of a clock on pll2
15:39.03leviathanhmm
15:39.10leviathanGNUtoo: does this work?
15:39.16phhhfctsaot: CPU clock = pll2/2.
15:39.19GNUtooleviathan, also be shure of having the good filenames
15:39.25leviathanok
15:39.39GNUtooleviathan, check bob copeland's page for that
15:39.47GNUtooleviathan, untested for udev 141
15:40.44hfctsaotso I just need to patch my kernel with that file or what
15:43.00*** join/#htc-linux leobaillard (~leobailla@leobaillard.org)
15:43.00*** join/#htc-linux maejrep[w] (~madCoder@68.64.144.221)
15:43.00*** join/#htc-linux radiochickenwax (~user@ip72-201-232-49.ph.ph.cox.net)
15:43.00*** join/#htc-linux phh (~quassel@2a01:e35:2e4b:b2b0:250:8dff:fee1:c793)
15:46.46*** join/#htc-linux balans (~Gebruiker@82-170-215-5.ip.telfort.nl)
15:49.57*** join/#htc-linux Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
15:55.50leviathanGNUtoo: with which version it works?
15:56.19GNUtooleviathan, ?
15:56.31GNUtooleviathan, under SHR we have udev 151
15:56.36GNUtooit seem to be broken
15:56.40GNUtooalso look at boot messages
15:56.43GNUtooit timeouts etc...
15:56.50GNUtoonot only under dream
15:56.54GNUtoobut also under eee7011
15:56.58GNUtoo*701
15:57.14GNUtooso I had the advise of trying udev 141
15:57.19GNUtoonot shure if it'll work
15:57.23GNUtoobut it's worth a try
15:57.33leviathanok
15:57.38leviathanI'll build ude 141
15:57.39leviathan:)
15:59.42*** join/#htc-linux Vito_| (~Sysel@195.113.241.225)
16:00.58ToAsTcfhphh: ok man home for a few
16:01.01leviathanGNUtoo: btw: did you push the new recipe?
16:01.03*** join/#htc-linux balsat (~balsat@87.72.13.34)
16:01.13GNUtooleviathan, not yet, I was working
16:01.14leviathanGNUtoo: for xf86-video-msm
16:01.17leviathanok
16:01.23GNUtooI'll do it now
16:01.26leviathank
16:01.29GNUtoonot push but create first
16:01.37ToAsTcfhphh: so just throw this on the devices and run th command u gave correct
16:01.41ToAsTcfh?
16:01.45phhToAsTcfh: yes
16:02.06leviathanGNUtoo: if you have the new recipe, please tell me, then I can build it into the new rootfs
16:02.16leviathanso, that we have working graphics, sound and wifi then
16:02.17leviathan:)
16:02.22GNUtoook
16:02.29ToAsTcfhok well i guess i gootta flash .27 real quick im using .29
16:02.53GNUtooleviathan, a bit of oe work is needed: also to change xorg.conf and to include xf86-input-tslib
16:03.14ToAsTcfhphh:unless u think it may work with .29 also?
16:03.25leviathanyes
16:03.31leviathanI mean to GNUtoo
16:03.36leviathanGNUtoo: yes
16:03.44phhToAsTcfh: if you have /dev/mem it will work too
16:03.54leviathanGNUtoo: and: we need to get rotation finally :)
16:04.25GNUtoook
16:04.44GNUtooleviathan, also to fix alsa routing and to make bluetooth work
16:04.50leviathanuhhm
16:04.51*** join/#htc-linux hes80 (~Stephan_H@188.104.243.170)
16:04.54leviathanalsa routing is fixed
16:04.56GNUtooleviathan, and to look into suspend
16:04.56ToAsTcfhphh: i dont have /dev/mem on either kernel
16:04.57GNUtoook
16:05.19GNUtooleviathan, by alsa routing I mean something that works with alsamixer fine
16:05.25leviathanaha
16:05.26leviathanokee
16:05.26phhToAsTcfh: you told me you have it ...
16:06.01ToAsTcfhi thought i did. then when i got home i confirmed and told u i didnt
16:06.06ToAsTcfhyesterday
16:06.07GNUtooleviathan, also...keyboard mapping
16:06.35ToAsTcfhbah so nm this wont work :9
16:06.39ToAsTcfh:(
16:07.54ToAsTcfhphh: bah... so no dice huh?
16:08.10phhToAsTcfh: bah you'll have to do that in kernel instead of userland
16:08.26*** join/#htc-linux DarkMasterHalo (~ssoulard@modemcable187.118-80-70.mc.videotron.ca)
16:08.27*** part/#htc-linux DarkMasterHalo (~ssoulard@modemcable187.118-80-70.mc.videotron.ca)
16:08.29*** join/#htc-linux DarkMasterHalo (~ssoulard@modemcable187.118-80-70.mc.videotron.ca)
16:08.43leviathanGNUtoo: yes
16:10.17GNUtoois it dcordes who commited xf86-video-msm?
16:10.40ToAsTcfhphh: so im confused... i need to increase the speed of pll2 to 1200 in acpuclock?
16:10.58phhyup
16:11.15ToAsTcfhi did that lastnight with bzo and no dice
16:11.29phhToAsTcfh: ah, you have a git diff ?
16:11.44phhGNUtoo: btw, did you try playing videos on xf86-video-msm ?
16:11.55phhhow well does it perform ?
16:12.02GNUtoophh, xv makes it crash
16:12.06phhah.
16:12.28GNUtoophh, but videos are fine in 320x240 cpu max speed and no other interuption like light going down and alsa
16:12.49GNUtoophh, maybe it would be better without all the debug stuff in the kernel
16:13.51phhplaying videos with x11 isn't going to be cpu efficient anyway
16:14.06ToAsTcfhphh: he had me change the divider of pll2  then so other stuff to keep it where i could just add 600000 instead 1200000
16:14.17ToAsTcfhwe tried both though
16:14.27phhToAsTcfh: and you always get crash ?
16:14.41ToAsTcfhphh: yeah
16:15.02GNUtoophh, if I would get 480x320  I would be very happy but I'm already happy it can play mpeg4 videos
16:15.04ToAsTcfhit wont even set the clock if i increASE  it to 1200000
16:15.05*** join/#htc-linux android-res (~yciduo@80.179.91.44)
16:15.26phhToAsTcfh: can it boot on a pll1 frequency if pll2's speed is changed ?
16:15.40ToAsTcfhyeah
16:15.56ToAsTcfhbut only if i stick it to not increase to pll2
16:16.09ToAsTcfhie set the max lower
16:16.13phhok
16:16.28phhdid you try commenting the request_pll code ?
16:16.29ToAsTcfhits gotta be the code its self
16:16.35android-resDoes anyone has an idea how can i scan another process memory? no root permissions...
16:16.45phhandroid-res: you can't
16:16.49phhyou need root permission
16:17.06ToAsTcfhphh: or else it would be fine with 528
16:17.42ToAsTcfhbut no matter what i put pll2 at it wll crash as soon as i set the clock
16:17.57ToAsTcfhit sets it in kmsg though
16:17.59phhpossibly the pll request stuff.
16:18.02android-resCan I access /proc/pid/mem somehow? only for read?
16:18.13phhandroid-res: without root access ? no.
16:18.37phhor maybe if it's a process you own
16:18.42android-resI'm trying for ages getting something from the ril...
16:18.53android-resalways hit a brick wall
16:19.07phhandroid-res: get root access and strace it ?
16:19.20android-resI need it on production app..
16:19.32phh??
16:19.36android-resI tried setting up something with the NDK and ptrace..
16:19.37phhwhat do you want to do ?!?
16:19.48phhwithout root access you can't do anything, period
16:19.56android-resI need the channel the phone is currently using
16:19.57MarkinusNetRipper: ping
16:20.03ToAsTcfhphh ok so no way around the pll request stuff then i take it
16:20.04android-resARFCN/UARFCN
16:20.14phhToAsTcfh: yes way around.
16:20.21phhToAsTcfh: request all plls at startup
16:20.23phhand only then change them
16:21.16ToAsTcfhbah i have no way to pull that off. i suck at C
16:21.38phhToAsTcfh: I'll try I guess.
16:21.47android-resI got it on a rooted phone using internal phone functions and a platform key.. But rooted phone is not enough :(
16:21.53ToAsTcfhi agree though. sounded to me like a pll issue from the start though
16:22.21phhandroid-res: uh ?
16:22.29phhyou mean you needed something else than just rooted phone ?
16:22.35android-resyes
16:22.49android-resI need it to work on a regular out of the box phone
16:22.57android-reswith a market apk :(
16:23.18ToAsTcfhphh: u dont think it could be the table defining pll2 as 528 could be the issue do u?
16:23.26android-resIf i could only inject code to the phone process :)
16:23.27phhandroid-res: ls -l /dev/smd0 ?
16:23.35phhToAsTcfh: no
16:23.41ToAsTcfhok
16:23.47phhandroid-res: there are only two ways to inject code in phone process
16:23.50phheither you're root
16:24.00phhor you have the correct uid/gid to access to the process.
16:24.04android-resI know..
16:24.17phhandroid-res: but isn't there some security exploits in current kernels ?
16:24.35android-resYes.. I don't know if wanna go there...
16:24.58android-resprefer something less "intrusive" than priv. escalation..
16:25.16android-reswhat was ur direction with the /smd0?
16:25.18phhand you think you can inject code in a "stranger" 's application without such things ... ?
16:25.33phhif you can have access to /dev/smd0, you can do the stuff by yourself
16:25.57ToAsTcfhandroid-res what device do u have?
16:26.02android-rescrw-r-----    1 radio    radio     254,   0 Apr  6 19:23 /dev/smd0
16:26.04android-res:(
16:26.20phhandroid-res: can't you get radio group privilege through android perm things ?
16:26.21android-resGot N1, ADP1, ADP2
16:26.43android-resWhats android perm things?
16:27.08phhthe things that give you more or less rights if you ask it ... ?
16:27.17android-resno..
16:27.21phhno what ?
16:27.38android-resIt only gives u permissions to get specific intents and such...
16:27.44android-resno UID change..
16:27.49phhand BT radio access
16:27.52phhand network access
16:27.56phhand file system access
16:28.00phhwhich means GID changes.
16:28.20*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
16:28.21phhand you only need radio gid, not uid
16:28.23phhI think.
16:28.29android-resYeah...
16:28.42android-resDo u think i can get that using the permissions?
16:28.49phhdon't know
16:29.09android-resdoes asking for permissions changes ur uid?
16:29.13android-resdidn't knew that...
16:29.16phhnot uid
16:29.17phhjust gid
16:29.24android-ressame for me
16:29.42android-resif i could get into radio my problems are solved
16:30.10android-resbeing in the same gid meaning i could ptrace on the f*er right?
16:30.21phhi'm not sure
16:30.45android-resfrom what i remember (man ptrace - ATTACH part)
16:31.13android-resThanks. I'll check the permissions direction.. sounds good
16:31.36android-resdo u have any good reading meterial about permissions and gid/uid relation
16:31.38android-res?
16:31.43phhnot at all
16:31.56phhwell, a big old grep in android sources
16:32.01phh(I've one running right now)
16:32.08Captnoordre
16:33.04ToAsTcfhphh: well just ping this nick if u make anyprogress on the pll stuff. im logging it so ill get all pings for sure
16:33.11phhok
16:33.39*** part/#htc-linux Termana (~bradley@123-3-85-159.static.dsl.dodo.com.au)
16:33.43ToAsTcfhim back off to work. good luck man. god i just want this shit to work
16:34.06ToAsTcfhim sure u do too. just s ill leave u guys alone :P
16:34.16android-res:)
16:35.35phhandroid-res: got it.
16:35.41phhthere is no radio perm.
16:35.48phhsee etc/platform.xml
16:38.31Captnoordhmm the panel powerdown seems to save a lot of power
16:39.26phhCaptnoord: but I don't seem to have much battery life than chamonix while I have a 1.5Ah on diam100 and he has 1350mAh raph100 ?!?
16:39.31android-resYeah.. I'm looking into it right now...
16:40.01Captnoordi'm fixing my battery power rig so I can measure the current
16:40.04android-resphh: do u have any other idea?
16:40.14phh*android-resno
16:40.19ToAsTcfhandroid-res: what is the uid issue u are having?
16:40.23*** join/#htc-linux jumoit (~samuel@121.77.87.145)
16:40.28ToAsTcfhshared uids?
16:40.42phhToAsTcfh: android-res want's ril's uid
16:40.45phheasy uh ,
16:40.47phh?
16:40.51ToAsTcfhlol
16:41.01android-resToAsTcfh: I'm trying to get the channel that the phone is currently on (ARFCN/UARFCN)
16:41.25android-resthe ril is running under radio
16:41.36ToAsTcfhna i dont think there is a uid for it but im not for sure. packages.xml tells all uids
16:41.41phh[18:29:48] <android-res> I got it on a rooted phone using internal phone functions and a platform key.. But rooted phone is not enough :( <--------- can you explain abit ?
16:41.55*** join/#htc-linux Kawaiius (~yadayada@pool-71-255-179-38.slsbmd.east.verizon.net)
16:41.57phhyou have another apk that asks the ril for the infos you want ?
16:42.26android-resLet's take it step by step
16:42.38android-res1. I've got HTC FTD
16:42.49phhftd ?
16:42.52android-resLittle bit reversing and some log reading
16:42.58*** part/#htc-linux jumoit (~samuel@121.77.87.145)
16:43.02android-resField Test Display
16:43.04phhok
16:43.07ToAsTcfhok
16:43.13phhhtc ftd gives the info you want ?
16:43.23android-resand i found out the internal function that was used to get the data..
16:43.39ToAsTcfhok
16:43.42android-resthe function returns a buffer which was pretty straight foreward to parse
16:44.04android-resso i created an apk that uses that function through java reflection
16:44.11android-resphh: yes
16:44.23phhhtc ftd doesn't need root access, does it ?
16:44.25android-resBut that didn't work
16:44.33ToAsTcfhnow u have shared uid issues with using that app on that shared uid
16:44.48ToAsTcfhresign the apks in the build all of them
16:44.54android-resphh: no, but it runs under the Phone process cause it's also signed with the platform key used to build the phonr
16:45.08android-resBeen there done that..
16:45.11ToAsTcfhgo to packages.xml
16:45.12phhandroid-res: ok
16:45.25ToAsTcfhwhat is the uid for ur app
16:45.50android-resThe intersting part is that the function won't work cause under the hood the funtion issues a ril request that is sent through a socket to the rild
16:45.56Captnoordhmmm I guess I need to charge
16:46.10Captnoord30% battery with my battery rig doesn't allow it to boot
16:46.20android-resToAsTcfh: I've got no problem with running my apk on a rooted phone
16:46.20phhCaptnoord: haha
16:46.42phhandroid-res: and why the hell do you want to publish such an app uh ?
16:46.43android-resI'm trying to do it in a regualr apk
16:47.07Captnoordabout 50 mv on a 0.54 ohm resistor
16:47.08Captnoordlol
16:47.54ToAsTcfhrooted or not the sign keys u used to sign ur apk dont match the keys for that uid
16:47.57android-resI work in a company that developes radio related apps and the only parameter that we need and is not in the standart sdk is this channel thing
16:48.31ToAsTcfhshared uids require all apks on that uid to be signed with the same keys
16:48.40*** join/#htc-linux sxe (~quassel@ip-88-153-21-39.unitymediagroup.de)
16:48.54Captnoordradio as in hw
16:48.59phhCaptnoord: no, ril
16:48.59Captnoordor radio as in software
16:49.03Captnoordyea I know
16:49.04phhjust the ril
16:49.07Captnoordthe hw part
16:49.08ToAsTcfhin order to do that u have to resign every apk in the build including frameworks
16:49.10Captnoordor the sw part
16:49.27android-resToAsTcfh: I'm not thinking about running inside the phone app cause i know it won't happen
16:49.31phhToAsTcfh: use a rooted build is easier.
16:49.48ToAsTcfhoh
16:49.53ToAsTcfhok nm
16:50.05ToAsTcfhur having uid issues
16:50.22ToAsTcfhbecause of the way u signed the apk with test keys
16:50.41ToAsTcfhthe build is signed with htc release keys
16:50.54android-resI'm trying to think on other directions (like memory scanning the rild process/injecting java to the phone process/writing to devices used to connect to baseband/writing to the socket connecting the java and reild/etc...)
16:51.21phhandroid-res: using a kernel security flaw would be way easier.
16:51.40ToAsTcfhim out good luck
16:51.56android-resFrom ur experience how long that those leave?
16:52.08android-reslive?
16:53.34android-resthe kernel security flaws...
16:53.42phhG1 is still with android 1.6.
16:53.46phhthat gives the idea.
16:53.57android-resWhat about OTA?
16:54.00android-resupdates?
16:54.11*** join/#htc-linux Wout (~a@s55927d21.adsl.wanadoo.nl)
16:54.14phhthey do it really not often at all
16:54.29android-reshave they done it for kernel updates?
16:54.39Jafterdark<PROTECTED>
16:55.05Captnoordhero got updated to 2.1
16:55.52android-resToAsTcfh: if i run as i root what is the most "clean way" to run in the phone process?
16:56.37phhandroid-res: trace the ril to know how it gets the infos you want
16:56.50phhif it always know this info, you can dump its memory and search in it
16:57.03phhif not, read/write to smd0 on your own
16:57.41android-resI prefer running in the phone process and invoking the java shit...
16:57.57android-resMore clean and no need to use the ndk..
16:58.13android-resIf i already got root access why no use it..
17:03.06android-resDoes anyone has any good priv. on eclair 2.1?
17:04.05*** join/#htc-linux jumoit (~samuel@121.77.87.145)
17:05.35Captnoordthen
17:05.39Captnoordswap the battery
17:05.41Captnoordwith the rig
17:05.43Captnoord:P
17:05.44Captnoordbleh
17:05.47Captnoordcan't wait
17:06.06phh300mW in sleep
17:06.07phh"oops"
17:06.13Captnoord300mW?
17:06.17Captnoordhmmm
17:06.27Captnoordwe have a 4V battery
17:06.34Captnoord0.3 / 4
17:06.51phh3.6*1.5 ~= 5Wh
17:06.59phhoh right, 300mW is what we actually have I guess.
17:07.22Jafterdarkthis is what i get
17:07.25JafterdarkE/SendTransaction( 1616): java.io.IOException: Cannot establish route for http:/
17:08.03JafterdarkD/ConnectivityService( 1245): stopUsingNetworkFeature for net 0: enableMMS
17:09.01Captnoordlol
17:10.23JafterdarkD/ConnectivityService( 1245): special network already active
17:10.52Captnoordhmmm
17:11.05Captnoordnow I need to know if microp gets suspended before mddi
17:11.08Captnoordas I need it
17:11.09Captnoordlol
17:12.03phhmicrop gets suspended ... ?
17:12.08*** join/#htc-linux GlemSom (~glemsom@0x5da34bca.cpe.ge-1-1-0-1105.sdnqu1.customer.tele.dk)
17:12.10Captnoordi2c gets
17:12.14phhok
17:12.20phhhey GlemSom
17:12.30phhGlemSom: seems you had some fun with your website :p
17:12.32GlemSomHey :)
17:12.44Woutchamonix: hey, I sent you some radio logs via pm on xda
17:13.16phhGlemSom: what's the "stripped" story ?
17:13.20GlemSomphh, Yeah, spent some time trying to make it look better... Though... I HATE doing anything with GUI... :P
17:13.25Jafterdarkcan anyone help if you have time
17:13.29GlemSomphh, Stripped debug symbols
17:13.44GlemSomohh wait, or was it --unneeded
17:13.45GlemSomlet me check
17:13.46phhGlemSom: how ? from what ?
17:13.53GlemSomKernel modules
17:14.03phhuh ? didn't we already do strip before ?
17:14.15GlemSomNo, just the wifi modules
17:14.19GlemSom(in build.sh)
17:14.25GlemSomNot all other kernel modules
17:14.46Captnoordworks
17:15.00Captnoord50mV with spikes
17:15.01phhGlemSom: oh ok
17:15.05GlemSomit was --strip-debug :P
17:15.16phhCaptnoord: I= ?
17:15.18GlemSomphh, It reduces the size with over 50%
17:15.23phhoh 50mB is I.
17:15.24phhmV*
17:15.29phhV = 4V I guess :D
17:15.53phhCaptnoord: so that makes 360mW ?
17:16.04Captnoordphh in suspend we use constantly 20mv with large spikes every 1 to 1.5 seconds
17:16.19Captnoordon raph it is
17:16.27Captnoordevery second*
17:16.30Captnoordspikes
17:16.37phhevery second ?
17:16.40phhhum
17:16.44phhand in windows there is no such thing ?
17:16.52jamenlangGlemSom: did you get my message on XDA?
17:17.12Captnoordcan't boot windows
17:17.19Captnoordbecause I didn't charge it yet
17:17.19Captnoord:P
17:17.29Captnoordbtw its a 0.5 ohm resistor
17:17.32GlemSomjamenlang, Uhhm... doesn't seem so...
17:17.38phhCaptnoord: I know, you already said that.
17:18.03Jafterdarkwhat are you guy's working on
17:18.18jamenlangi think it was just a 'visitor message' not a private one
17:18.21phhJafterdark: Captnoord is playing with battery consumption.
17:18.23Captnoordphh spikes are HUGE
17:18.42Captnoordat least 50 mV on top of the 20mv
17:18.47Captnoord70mV spikes
17:18.52phhCaptnoord: like the radio is sending a keepalive every second.
17:18.54GlemSomjamenlang, Ohh yeah, regarding twitter... I did answer that... didn't I ?
17:18.58CaptnoordI dono
17:18.59jamenlangi dunno
17:19.08phhCaptnoord: well, you'll compare with wimo.
17:19.21leviathanGNUtoo: did you push the recipe now?
17:20.42Captnoordyea
17:20.42CaptnoordI will
17:20.43Captnoordwe do have the same spikes when we are awake
17:20.43Captnoordbut they are less
17:20.43Captnoordhmmm
17:20.44Captnoordhard to say
17:20.52phhCaptnoord: just different scales ? :p
17:21.08Captnoordnah same scale
17:21.15Captnoordthe spikes are always 70mv
17:21.22Captnoordonly when everything is awake
17:21.28Captnoordthe default drain is about 50mv
17:21.31Captnoordso the spike is lower
17:21.34Captnoordonly 20mv
17:21.41phhhow width is the spike ?
17:21.42Captnoordso if you say its the radio
17:21.43phh100us ?
17:22.11Captnoordabout 7x 5 ms
17:22.15Captnoordbrb
17:22.16phhoO
17:22.19phhthat's really long
17:23.18GNUtooleviathan, building it
17:23.22GNUtooI'll eat then push
17:23.26leviathanok
17:23.37GNUtooelse I can pastebin it
17:23.41GNUtooif you want
17:23.48GNUtoobut not shure that it'll work
17:23.51leviathanyeaah
17:23.56leviathansure, just pastebin it
17:24.05leviathanor
17:24.06leviathanno
17:24.06GNUtooah sorry
17:24.08leviathanwait
17:24.08GNUtooson't work
17:24.13leviathanI'll pull it regulary
17:24.15GNUtoobecause of files://
17:24.24leviathanpush it
17:24.32GNUtooI really need to go to eat
17:24.34leviathanthen your signature keeps conserved
17:24.37leviathannp
17:24.42leviathantake your time
17:24.44GNUtooI can't push it before beeing shure it works
17:24.52GNUtooI want to keep my commit access
17:25.01GNUtooalso because it's usefull for us
17:25.03leviathanI've still around 60 years to suffer under my existens I fear
17:25.07leviathanso take your time :)
17:25.34leviathansure
17:26.07Jafterdarkif you need any log just hit me up
17:26.55*** part/#htc-linux DarkMasterHalo (~ssoulard@modemcable187.118-80-70.mc.videotron.ca)
17:27.11leviathan:)
17:27.22*** part/#htc-linux android-res (~yciduo@80.179.91.44)
17:27.35leviathanfirst we need to ensure it works at all ^^
17:28.28Captnoordphh make that 4x 5ms
17:28.32Captnoordon avarage
17:28.34Captnoord20ms
17:28.37phhCaptnoord: still way too much
17:28.42Captnoordyup
17:28.46CaptnoordI wonder if it is wakeup
17:28.49Captnoordand resuspend
17:30.48Captnoordnah on windows boot
17:30.48Captnoordthat doesn't continue
17:30.50Captnoordits about 70mv
17:30.56Captnoordbut thats on init I guess
17:31.13Captnoordto calculate the current
17:31.15phhtell the result when it's sleeping ?
17:31.15Captnoordyou have todo
17:31.18Captnoordyea
17:31.21Captnoordcharging now
17:31.24phhhaha.
17:31.26Captnoordvoltage / resistance
17:31.27Captnoordso
17:31.31Captnoord70mv / 0.5
17:31.35Captnoordis 140 mA
17:31.46phhif it's charging it's kind of wrong.
17:31.47Captnoordmax on bootloader
17:32.24Captnoordwon android its 20mv with large spikes
17:32.30Captnoordwhich we can say is kinda like
17:32.33Captnoord30mv constant
17:32.40Captnoordthat translates into 60mA
17:32.49Captnoord1350 mA/h
17:33.02Jafterdarkis this what your working on W/BatteryStatsImpl( 1245): Couldn't get kernel wake lock stats
17:33.13Captnoordis 22 hours
17:33.23Captnoordbut the max we can get is 13 hours?
17:33.26phhCaptnoord: mA*h not mA/h .
17:33.37phhit often wakes up.
17:33.50phhand battery efficientcy isn't 100% I guess
17:37.57Captnoordphh don't make start all over again
17:38.08Captnoordits mA / 1 hour
17:38.14Captnoordso if I divice by the drain
17:38.18CaptnoordI get the amount of time
17:38.20Captnoordnah
17:38.23CaptnoordI did start again
17:38.53Captnoordor what every
17:39.24phhif it's mA/h, if you devide by the drain, you get 1/hours
17:39.25phhnot hours
17:39.31Captnoordtrue
17:39.33Captnoordyour correct
17:40.16phhcan you disable microp ?
17:40.25phhI don't think it does any vital stuff
17:40.34CaptnoordI can disable it
17:40.39Captnoordsaves more than 10mA
17:40.45phhhuhu
17:40.45Captnoordabout 5mA for the led
17:40.56Captnoordand another 5mA for the uC
17:41.00Captnoordor maybe more
17:41.12phhand you can make it start again after that ? :D
17:41.19Captnoordhehe
17:41.20CaptnoordI hope
17:41.42Captnoordlast time I checked it doesn't really seem to have a suspend function
17:43.05phhthere is nothing there either
17:43.49Captnoordand nothing in the dll...
17:43.52Captnoordnothing direct
17:43.59phhbah you have microp's doc no ?
17:44.10Captnoordnope
17:44.11Captnoordyou?
17:44.19phhhow did you fix your microp ?
17:44.27Captnoordhehe
17:44.29phhwell then you can ask the SPL/OEMSBL
17:44.30Captnoorddiag image
17:44.36phhor diag image.
17:44.39Captnoorddiag rom*
17:44.44Captnoordyou gave me
17:44.50phhreverse it :p
17:45.13CaptnoordI didn't
17:45.16Captnoordjust loaded
17:45.21Captnoordand did test all
17:45.24Captnoordand it reset the microp
17:45.36CaptnoordI can't reverse it because I can't extract it
17:45.46phhah
17:45.56Captnoordbut
17:46.04phhyou're sure there is anything to do btw ?
17:46.05Captnoordthere is interesting code in the other microp
17:46.13Captnoordwhat do you mean?
17:46.15phhvreg down/vreg up isn't enough ?
17:46.19CaptnoordI think it got loads of stuff
17:46.24Captnoordnah the microp disables the lcm
17:46.34Captnoordsaves a couple of mA
17:46.39*** join/#htc-linux mickey|zzZZzz (~mickey@openmoko/coreteam/mickey)
17:48.58CaptnoordI also think channel power is related to the radio
17:49.06Captnoordwhich is also in the microp dll
17:49.30*** join/#htc-linux zachy (~george@85.193.4.18)
17:50.18Captnoordbtw the only real difference I found (panel spi init) is that
17:50.25Captnoord{0x00160004,0x0000A1EF}, is {0x00160004,0x0000A0EF},
17:50.30Captnoordat least on wince
17:50.33phhjust one bit ?
17:50.34phhwow.
17:50.34Captnoorddon't know what it means
17:50.36Captnoordyea
17:50.45Captnoordthere are 4 lcd supported on wince
17:50.59Captnoordand I think its the same as on diamond
17:51.02Captnoordexactly the same
17:51.26CaptnoordI know somebody split the stuff
17:51.35Captnoordbut i'm ready to merge
17:51.41Captnoordhehe
17:51.52CaptnoordI know what phone Wout is gonna get
17:51.53Captnoord:P
17:51.55phhWout: still no keyboard.
17:52.06phhno tv out
17:52.10WoutCaptnoord: nope, because of what phh said
17:52.10Captnoordhehe
17:52.13Captnoordhehe
17:52.24phhseriously, when the TP3 will goes out ?
17:52.30phhor nexus pro
17:52.40phhthe best would be TP3 with hdmi + analog tv out
17:52.45Woutnexus pro prolly around august
17:52.50Wouttp3 will be a smaller tp
17:53.03Woutbest bet would be samsung galaxy pro
17:53.06Woutsomewhere this fall
17:53.11phhna
17:53.13phhI want a htc.
17:53.29Captnoordhackable
17:53.29Captnoord:P
17:53.30phhWout: anyway, is desire way better concerning normal use ?
17:53.38Wouthope htc switches to amoled soon
17:53.48phhI agree it's way better concerning games
17:53.53Woutphh: sense on android really adds heaps of functionality
17:54.04Woutand they've tweaked the hell out of it
17:54.15phhWout: you can have desire's sense on TP2 :p
17:54.30Woutyeah....crippled and slow :P
17:54.46phhah ?
17:54.49phhI havent' tested it yet
17:55.01phhI believe that it could be smooth with proper SD handling
17:55.04Woutthe 2.1 build is way behind
17:55.07phhbut linux totally sucks concerning SD.
17:55.20Woutbut seeing that the hero will get a sense upgrade
17:55.25Woutit should run on our tp2 too
17:56.02WoutI was running a memory killer on the desire, after playing around for a few mins it read: 238mb free
17:56.03Woutlol
17:56.58phhoh and we definitely have some memory leak.
17:57.07phhafter 3 days of continuous use, I had only 70MB free.
17:57.11phhwith all apps killed
17:57.48Captnoordbooted into wince
17:58.39Captnoord50 mV constant with display on normal backlight
17:59.57Captnoordhmmm
18:00.08Captnoordphh want to know what the suspend current is?
18:00.16phhyes
18:00.18phhunmetered ?
18:00.26Captnoordspikes are max 40mV
18:00.35phhin wince ?
18:00.42Captnoordyea
18:00.43Captnoordthe rest
18:00.45Captnoordlets say
18:00.49phha spike every second ?
18:01.01phha 30ms long spike ?
18:01.02Captnoord< 5mV
18:01.08*** join/#htc-linux luminoso (~lumos@av-217-129-131-108.netvisao.pt)
18:01.18phhOK
18:01.28phhyou know what to do now.
18:01.29Captnoordyea
18:01.32Captnoord30ms
18:01.40phh30ms sounds really long
18:01.49Captnoordgot a old scope
18:01.53Captnoordcan't make a proper picture
18:01.54Captnoordbleh
18:01.58phhthat makes 3% of the time mmm
18:01.59leviathanGNUtoo: re?
18:02.15Captnoordphh it doesn't work that way....
18:02.25Captnoordyou need to intergrate to get the correct value
18:02.31CaptnoordI wish I had a digital scope
18:02.43CaptnoordI can try tomorrow at school to get sole 1337 pictures
18:02.44phhCaptnoord: you don't know how to do an electronical integrator ? :p
18:02.51CaptnoordI know
18:02.55Captnoordbut they are not accurate
18:03.00Captnoordand hard to make accurate
18:03.07Captnoordfor our purposes
18:03.15phhoh right
18:03.17Captnoordbut I can just use my voltmeter
18:03.21Captnoordit avarages it a bit
18:03.24phh30ms spikes of 40mV makes 26days.
18:03.29phhof battery life*
18:03.31phhright.
18:03.34phhtime to eat
18:05.01Captnoord2mv max
18:05.09Captnoordincluding the spikes
18:05.24Captnoord0.44 mv constant
18:05.49Captnoordthat is about 1 mA drain
18:07.09*** join/#htc-linux kiozen (~oeichler@rgnb-5d87c5f8.pool.mediaWays.net)
18:08.04Captnoordphh btw that was with the panel code
18:08.08Captnoordnow gonna try without
18:08.27Captnoorddefault kernel
18:12.24Captnoordthe panel code doesn't do shit
18:12.25Captnoordlolz
18:12.27Captnoord20mv
18:12.31Captnoordwith same spilkes
18:13.35Woutany progress on the batt readout?
18:14.01Captnoordyea goes slow
18:14.12Captnoorddon't worry
18:14.15CaptnoordI won't forget you
18:14.26Wouthehe, just curious ;)
18:19.17Captnoordbleh because I copied the diamond panel code the microp stuff has already been done
18:19.51Captnoorda bit hacky
18:20.01Captnoordbut should be working
18:20.55leviathanGNUtoo: re?
18:21.26*** join/#htc-linux marex (~marex@vasut.kolej.mff.cuni.cz)
18:23.40mickeylok, wifi almost works on the dream
18:23.49mickeylthanks all who were involved
18:23.51mickeyl!
18:23.59GNUtooleviathan, back
18:24.32phhCaptnoord: 20mV with panel not off, 2mV with panel off ?
18:24.47Captnoordnah
18:24.56Captnoordwith diamond panel code 20mV
18:25.05Captnoordwith auto suspend
18:25.07Captnoord20 mV
18:25.13phhok
18:25.17Captnoordnow killing microp
18:25.20Captnoordleds
18:25.34phhif killing leds, some people might hate me.
18:25.40Captnoordlol
18:25.45Captnoordnah we can make it optional
18:26.12phhit's already optional
18:26.15Captnoordbut its stupid to turn a led on when you want battery live
18:26.17Captnoordlol
18:26.23leviathanGNUtoo: wb
18:26.24Captnoordk
18:26.38leviathanmickeyl: you're welcome ;)
18:26.50mickeyl:)
18:26.54GNUtoothanks
18:26.55phhCaptnoord: it's an option in chamonix's ledeffects app
18:27.08GNUtoo(to leviathan  and mickeyl)
18:27.14mickeyltwo little issues open, but we can fix them
18:27.18leviathanyes
18:27.19GNUtoothat are?
18:27.21Captnoordk
18:27.32mickeylkernel WARNING which i don't like
18:27.34leviathanGNUtoo: hmm, perhaps the firmware issue?
18:27.39leviathan:)
18:27.44Captnoordhmmm
18:27.45GNUtoothat is one
18:27.51Captnoordwhat else can draw that much power
18:27.53mickeyland interface index somewhat broken
18:27.55phh"udev sucks" is the second issue ? :D
18:27.56leviathanloading it manually is not very practicable
18:27.57Captnoordmaybe we don't sleep deep anough
18:27.57phhah.
18:28.04mickeyludhcpc complains wlan0 does not exist
18:28.06GNUtoomickeyl, ah that's udev rules
18:28.10mickeylnah
18:28.11mickeylno udev here
18:28.15leviathanmhhh
18:28.17GNUtoomickeyl, last part of mac address is random
18:28.20phhmickeyl: fix the mac ?
18:28.25mickeylhmm?
18:28.25phhI have a patch for that somwhere
18:28.28GNUtoomickeyl, so it creates new interface each time
18:28.38mickeyli don't see how that should disturb udhcpc
18:28.39GNUtoowlan0->wlan1->wlan2->wlan3 etc...
18:28.42mickeylit's rather:
18:28.44leviathanI'll look
18:28.47GNUtooah
18:28.56mickeylroot@htcdream:~# udhcpc wlan0
18:28.56mickeyludhcpc: SIOCGIFINDEX: No such device
18:29.04mickeylwhich is somewhat worrying
18:29.06GNUtoomickeyl, udhcpc -i wlan0
18:29.09mickeylbut manually setting works
18:29.12mickeylwah!
18:29.16phh-_-'
18:29.19*** join/#htc-linux bzo (~chatzilla@netblock-68-183-234-14.dslextreme.com)
18:29.22mickeylhahaha
18:29.25mickeyli'm such noob
18:29.27mickeylthanks guys
18:29.30phhI rememebr I did the same thing.
18:29.37mickeylawesome
18:29.43mickeylnow i will polish my firmware loader and commit
18:29.49GNUtootoo much fso hacking and less embedded stuff?
18:29.49leviathanhmm, ok
18:29.51mickeylafter today's TV session
18:29.54GNUtoook
18:29.55mickey|tv:)
18:29.56leviathanlol
18:30.16leviathangood idea
18:30.36leviathanlets see if they have already newest doctor who (eleventh hour) on google video xD
18:31.03phhleviathan: it's damn good.
18:31.20leviathanphh: you've seen it?
18:31.25phhyup
18:31.33leviathanI just know
18:31.41leviathanhe now always says jeronymo
18:31.48leviathannot "molto bene" anymore
18:31.48phhhum
18:31.52phhI heard it only once
18:31.56stinebdinteresting
18:32.02leviathanand the new doctor also uses guns
18:32.07phhand it was a story about a poll.
18:32.09phhguns ... ?
18:32.13leviathanyeah
18:32.14leviathantrailer
18:32.18phhI haven't seen a damn gun in this episode -_-'
18:32.20leviathanonce basically
18:32.21leviathan:)
18:32.28leviathandunno
18:32.33leviathanjust saw it in the trailer
18:32.53*** join/#htc-linux Fordyy (~MrFordihu@d90-130-14-1.cust.tele2.se)
18:32.54leviathanhe only once had a gun in all the movie history
18:32.56stinebdi wonder if bluetooth pairing via confirmation number works in the aosp build
18:32.58leviathanbut never fired one
18:33.05phhstinebd: ah you too ?
18:33.11stinebdphh: yes, pin entry is broken
18:33.13leviathanphh: anyway
18:33.20leviathanI'll watch out for doctor who
18:33.23leviathan^_^
18:33.29leviathanuntil mickey|tv is back
18:33.35stinebdphh: did pin entry work in babijoee's?
18:33.37leviathanbecause he is the one, with the firmwareloade
18:34.02phhstinebd: i'm not sure
18:34.06phhpossibly not.
18:34.47stinebdwell no matter, shouldn't be too hard to get it working
18:35.10phhstinebd: I had such problems on my desktop
18:35.14phhand yes it's hard.
18:35.28stinebdthe problem is in UI
18:35.42phhmaybe not
18:35.45phhcould be dbus problem
18:36.01stinebdwell that would be even easier to fix
18:36.08phhtrue
18:36.19phhbut I've checked logcat
18:36.22phhthere is nothing really explicit.
18:36.39stinebdso have i
18:37.22Captnoordhmm current sensing is a bit of
18:37.28Captnoordshould show 140 mA
18:37.37Captnoordbut shows me 87 mA
18:38.00phhthere is a second hidden battery ! :p
18:38.03phhhum wait
18:38.07phhthat'd do the opposite.
18:38.25CaptnoordI know that there is a second battery
18:38.34Captnoordbackup battery for RTC
18:38.39phhuh ?
18:38.49phhoh yeah I don't lose time.
18:38.50Captnoordin the left cornor of your phone
18:38.59WoutCaptnoord: could it be that some commit broke your clock drift problem?
18:39.04Woutproblem=fix
18:39.23CaptnoordWout I think the clock drift problem is related to the fact that we suspend / wake so often
18:39.32Woutk
18:39.42Captnoorddo you still have the problem?
18:39.47phhCaptnoord: I do
18:39.50Captnoordk
18:39.51phhmy phone ran for three days
18:39.57phhI got ~ 1hour of drift
18:39.59Captnoordk
18:39.59Woutyeah, notice dit today
18:40.01stinebdclearly the solution is pm.sleep_mode=4
18:40.05Woutless than before though
18:40.11phhCaptnoord: I think the problem is that we don't read hardware clock
18:40.19CaptnoordI dono
18:40.21Captnoordpossible
18:41.22phhstinebd: does it even work ?
18:41.30Captnoord.text:1001B400 aRil_pmprocSetT unicode 0, <RIL_PmProc: Set timer to 4s before entering suspend mode>
18:41.33Captnoordhmmm ril suspend
18:41.38phh?!?
18:41.49phhwhich timer is that?
18:41.59CaptnoordRilPhone.dll
18:42.47CaptnoordAT+HTCPDPIDLE
18:43.10phhPDP ~= data connection
18:43.16Captnoordk
18:43.40phhI guess that's to say to antenna that our radio is on only once every 4 seconds ?
18:46.10phhbut I can't see why it would be PDP specific mm
18:46.10Captnoordat least something doesn't get suspended
18:46.43phhI guess it would just change the frequency of the spikes
18:47.01phhI should trace AT commands.
18:47.28Captnoordand that Enable_Level2_Lock dex call?
18:47.36phhyou can test it :p
18:47.50Captnoordhehe
18:47.52Captnoordnot today
18:47.55stinebdworks with confirmation code
18:48.01CaptnoordI will hit the bed early
18:48.17phhstinebd: ?
18:48.47stinebdphh: if both devices can pair via confirmation code, it works. if either one requires pin it doesn't
18:48.59phhstinebd: wow
18:49.01phhthat's weird
18:49.10phhand really annoying.
18:49.25MarkinusCaptnoord: what'S with out vregs? We didn't disable they all . .
18:49.35CaptnoordI dono
18:49.45Captnoordsdcc should't be disabled
18:49.49Captnoordif they are the correct ones
18:49.50Captnoordat least
18:49.57phhCaptnoord: uh ?
18:50.03MarkinusCaptnoord: lcd, wlan, bt . .
18:50.24Captnoordi'm running with a special build of mine
18:50.28Captnoorddiamond panel code
18:50.34Captnoordbut I don't know...
18:50.45Captnoordbecause the wince has a special power down function
18:50.49Captnoordwith 2 spi commands
18:51.11Captnoordand I don't know why it doesn't want to stay suspended
18:51.41CaptnoordI could have tested with radio off
18:52.37*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
18:53.36Captnoordnah i'm tired
18:53.37Captnoordnn all
18:58.20MarkinusNetRipper: I did a create a minimal config for Leo. The kernel is now ca. 2.2 MB. But still same problem, signal oder hangings after freeing init memory
18:58.46MarkinusNetRipper: so it's something very low level
18:59.08NetRipperMarkinus, disable smd in board-htcleo.c
18:59.16NetRipperuncomment that &smd line
18:59.18MarkinusNetRipper:  I did it
18:59.21NetRipperok
18:59.22NetRipperhm
18:59.23MarkinusNetRipper: disabled all stuff
18:59.41MarkinusNetRipper: no other files as board file
18:59.51MarkinusNetRipper:  no dex, etc.
19:00.00NetRipperi got the kernel to spew out error messages for > 60 sec
19:00.27Markinusok
19:00.29NetRippererror messages originating from timer.c, because smd was disabled
19:00.52MarkinusNetRipper: yes .. . this is needed, RPC too . .
19:00.54NetRipperit tries to sync the clock with smem clock
19:00.59NetRipperit timesout all the time
19:01.01MarkinusNetRipper: so I desabled all this stuff
19:01.02NetRippercausing these errors
19:01.23Markinushmm
19:02.09Markinusit makes a difference if we compile with AEABI or not
19:02.26MarkinusNetRipper: without sthe signal message is comming directly after freeing init memory
19:02.37NetRipperwithout what?
19:02.40NetRipperoh
19:02.51MarkinusNetRipper: with  hangs after freeing init memory
19:03.07MarkinusNetRipper: then the error is comming only if we have other stuff in kernel
19:03.26MarkinusNetRipper: but is hnaging
19:04.52MarkinusNetRipper: Here is my config
19:04.52Markinushttp://pastebin.com/JSpfyVMX
19:05.44MarkinusNetRipper: I disabled the whole PM stuff to . .(had to patch some files fot it)
19:09.32NetRippermaybe some of androids drivers are still in place.. like smem stuff
19:09.46NetRipperwrong writes to smem could cause hangs as well
19:10.03NetRipperi'm starting to get out of ideas :p
19:10.14MarkinusNetRipper: right :)
19:10.31Markinusand this wrong signal . .
19:10.49*** join/#htc-linux Vito_| (~Sysel@gw.loccal.net)
19:11.40MarkinusNetRipper:  hmm, right this is the stuff what I not disabled . . the android driver
19:12.39NetRipperthere is much stuff in arch-msm that uses smem
19:15.03MarkinusNetRipper: hmm yes ..
19:15.24leviathanGNUtoo: did you push?
19:15.33GNUtooleviathan, many typos
19:15.37GNUtooso not yet
19:15.41leviathanok
19:15.46GNUtooI usually take a long time for pushing
19:20.39*** join/#htc-linux bombenbach (~bombenbac@77.47.80.54.dynamic.cablesurf.de)
19:20.53*** join/#htc-linux eval- (~eval-@bokchoy.rutgers.edu)
19:21.25bombenbachHi, I'm curious about htc-msm-2.6.29
19:21.33bombenbachdoes it compile for Rhodium?
19:21.50bombenbachI mean does the kernel boot on Rhodium?
19:22.15phhno
19:22.23bombenbachwhy?
19:22.36phhwell, unless you backported it.
19:22.43phhforwardported ?
19:22.43phhhum
19:22.46phhported
19:22.46phh.
19:23.28bombenbachso, adding the msm7200 support from 2.27 probably won't be enough?
19:23.40phh?
19:23.57GNUtoobooting and testing
19:24.25bombenbachI don't quite understand why it doesn't boot. I mean the hardware support
19:24.30bombenbachshould be inside?
19:24.35bombenbachfor msm7200
19:24.54phhbombenbach: ?
19:25.00phhI really don't understand what do you mean
19:25.39bombenbachI can't understand why 2.6.29 won't boot although 2.6.27 does
19:25.57phhbecause the drivers for rhodium hasn't been backported
19:28.27Markinusbombenbach: the drivers aren't in the mainline of the kernel . .
19:28.45bombenbachoh, then it's clear
19:30.11bombenbachIt's just this thread http://tinyurl.com/yg8q9bz which makes me think
19:30.26bombenbachmy bcm4329+iwlist problem might be a kernel-related one
19:31.04bombenbachI've already took that N1 bcm4329 module and compiled it with our 2.6.27 kernel
19:31.12Markinusbombenbach: NEXUS!
19:31.24bombenbachwhich worked but the iwlist problem was still there
19:31.33phhbombenbach: .....
19:31.36phhhow to put that
19:31.39phhhum
19:31.49phhwe are using N1's bcm4329.
19:32.45GNUtooleviathan, should we do another xorg.conf while we're at it?
19:32.48Markinusphh: from .29 or .32?
19:32.55phhMarkinus: .32 iirc
19:33.55bombenbachphh: I see, but if bcm4329 is the same, but iwlist scan works with android kernel 2.6.29
19:33.56phhbombenbach: maybe try with a bcm4325 driver ?
19:34.17bombenbachphh: where can I find it?
19:34.22phhI7500
19:34.41bombenbachlisten on http://android.git.kernel.org/ ?
19:34.46phh?
19:35.04eval-what advantages will .29 and .32 bring?
19:35.15phheval-: tvout & usb otg drivers
19:35.23bombenbachphh: I mean where can I get the module source
19:35.28bombenbach?
19:35.45phhbombenbach: google I7500 kernel source code ?
19:35.53bombenbachphh: yes
19:36.08bombenbachI will
19:36.46Markinusphh: btw: I prepared already a .32 repo . . .we could start with a bit porting . . (cloned Leos one) http://gitorious.org/linux-on-htc-armv6/linux-on-htc-armv6/commits/htc-msm-2.6.32-armv6
19:36.57eval-phh: all can do usb otg?  which can do tvout?
19:37.17eval-Markinus:  i thought stinebd or someone was already playing with .32?
19:37.20Markinusbombenbach: it'S a Samsung . . . ..
19:37.29stinebdwahaha
19:37.30stinebdnot me
19:37.41eval-hmm.  ok maybe i was already thinking of Markinus then :p
19:37.52Markinusha :)
19:37.58stinebdi've got enough crap to do with just aosp
19:38.11eval-stinebd:  yea i'm impressed you took that on.  why?  where did our previous builds come from
19:38.16bombenbachsorry for being that lazy; just in case someone else needs it: http://tinyurl.com/y9tusg6
19:38.27stinebdeval-: dunno
19:38.31stinebdmanup or something
19:38.36leviathanGNUtoo: yes
19:38.44leviathanwe should do another xorg.conf
19:39.14phheval-: tvout are for TPs
19:39.19phhTP1/2
19:39.23phhfor usb otg, no clue.
19:39.30phhit has to be reversed from the SPL
19:39.50phhMarkinus: which tree have you forked from ?
19:40.04bombenbachphh: what's actually the diff between bcm4329 and 4325
19:40.04bombenbachand which bcm is in rhodium?
19:40.10Markinusphh: Leo one, anbd this is forked from Nexus .32
19:40.14phhbombenbach: bc4325 in rhodium
19:40.16phhMarkinus: ok fine
19:40.23eval-phh: they can do tvout on winmo??  some adapter someplace?
19:40.36phhbombenbach: bcm4329=bcm4325 + FM out + wifi N
19:40.38phheval-: .
19:40.46phheval-: it's an advertised feature you know ? :p
19:40.57phhnot a hidden one
19:41.15phheval-: get something like that: http://www.dealextreme.com/details.dx/sku.34011
19:41.16eval-mine just came with crappy headphone adapter
19:41.21phhmine too
19:41.25phhhum
19:41.25phhno
19:41.27phhno headphone adapter.
19:41.50bombenbachphh: well, that sounds quite promising. I hope it will compile.
19:41.56*** part/#htc-linux jumoit (~samuel@121.77.87.145)
19:42.19phhbombenbach: it won't compile as is ...
19:42.22GNUtooleviathan, ok should I push now or prepare xorg.conf
19:42.23GNUtoo?
19:42.43phhbombenbach: ah you mean I7500's driver
19:42.43leviathanuhm
19:42.46leviathanfirst push
19:42.50leviathanthen we prepare xorg
19:42.55leviathanand write in the comment
19:43.01eval-can our arm actually decode 640x480 of some format (raw? mpeg2?) fast enough or at least ~20fps?
19:43.03GNUtoook
19:43.04leviathanthat its allowing the usage of the new driver
19:43.20bombenbachphh: of course, only the module
19:43.31bombenbachbcm4325
19:44.05phheval-: we have a qdsp.
19:44.17phheval-: google for hero's performance maybe
19:44.31*** join/#htc-linux fish1209 (~fish1209@unaffiliated/fish0912)
19:44.42*** part/#htc-linux zachy (~george@85.193.4.18)
19:45.46eval-wow.  this project keeps getting cooler.  when i can grab diamonds for $50 there will be some crazy tech/art projects to do
19:46.30eval-oh, i guess only pros for the video out eh.  hmm
19:46.33phhdiam can't tv out.
19:46.36phhAFAIK
19:46.57makkoneneval-: in winmo they can do mp4 with hardware acceleration -- seems to be well above 20fps. I just transcoded something a couple nights ago, and it looks pretty great on the tp2 screen.
19:47.19makkonenno idea the specifics of it, or if that hardware has drivers in linux, though.
19:47.32phhmakkonen: lol.
19:47.40phhmakkonen: you know I was just asking this on #xda-devs ?
19:47.50makkonensynchronicity!
19:47.57eval-makkonen:  oh ugh yea the decode on linux might fail
19:48.02phhmakkonen: which software uses it ?
19:48.07eval-makkonen:  is there anything like ndiswrapper for video??  ;)
19:48.13phheval-: we have android drivers :p
19:48.24bombenbachjust a question: why do you actually mouse and touchpad support in 2.6.27? Touchscreen works fine without
19:48.26eval-phh:  b/c the google phones do mp4?
19:48.33phheval-: even H264
19:48.38eval-neato
19:48.38bombenbach*do you actually include*
19:48.55bzomakkonen: are you switched over to the tp2 now?
19:49.03phhbombenbach: some possible uinput use
19:49.12makkonenbzo: yeah.
19:49.28bombenbachpss: well it actually confuses Xorg a lot
19:49.32bzouh oh, am I the last cdma diam/raph user here?
19:49.40bombenbachbut nevermind
19:49.53makkonenbzo: you're getting close.
19:50.03eval-phh: how useful has been having a spare diam been?  should i be selfish and take my useless cdma diamond to europe or give it to my brother
19:50.25phheval-: totally useless.
19:50.27phhwell
19:50.36phhoh well you meant for non development puprose ?
19:50.41phheval-: diam has 3D, rhod doesn't :D
19:50.43eval-phh:  no, i mean purely for dev
19:50.47phhand diam has sound
19:50.49bzomakkonen: I've been tempted to get the tp2 myself, but I'm not sure I can deal with the size
19:50.50eval-phh:  or silly projects
19:50.53phhso I use diam as game console + spotify.
19:51.04phhbut for development, it's not really usefull
19:51.08eval-bzo:  it's enormous.  i was shocked.  no more skinny jeans ;)
19:51.24*** join/#htc-linux [acl] (~abel@96.246.167.90)
19:51.47phhbzo: I compared it with a "MP4 player", that used to have an enormous screen
19:51.49eval-phh:  its not annoying to have your actual phone always between OSes?
19:51.59phhI call it.... "lol".
19:52.04phheval-: uh ?
19:52.08phheval-: diam is always on android
19:52.09phhand rhod is
19:52.10phhwait...
19:52.13phhalways on android too.
19:52.28*** join/#htc-linux Orkspalter (~Orkspalte@p54B58CB9.dip0.t-ipconnect.de)
19:52.30eval-phh: really?  i can't do either b/c of battery =/
19:52.38phheval-: even on rhod ?!?
19:52.44eval-phh:  i told you already!
19:52.49phheval-: my diamond with radio off does >1 day
19:52.56eval-phh:  best so far is 12h to use 80%, airplane mode, no wifi, no sim
19:52.57phhmy rhod does ~ 20hours
19:52.58bzoI've been scarred by having an apache for 2 years. I can't go back! ;)
19:52.59makkonenbzo: yeah. even coming from a touch pro, the tp2 feels big. But most of the increase is in the screen, and I was always bugged by how small the tp/diam screen was, so I like it.
19:53.09eval-phh:  my diam500 does 6-8hrs
19:53.40eval-makkonen:  and the screen is so gorgeous!
19:53.54phheval-: well, on winmo the screen doesn't seem really good
19:53.56eval-makkonen:  tho not quite amoled
19:53.57phhbut on android
19:53.59phhit's O_O
19:54.04bzothe apache was such a brick. I guess the tp2 is 2/3 the thickness
19:54.08GNUtooleviathan, can I try to push high priority push also,else it break compilation of xorg under shr
19:54.21makkonenbzo: scarred how? I was scarred by the fact that my apache had a text-to-speech 'YOUR BATTERY IS LOW' voice whenever the battery dropped below 20%, and it wouldn't turn off even if you muted it. god, what a piece.
19:54.34phhmakkonen: lol
19:54.50eval-phh:  i'm still not using android much on my TP2 b/c of speed and battery issues.  it's superfast on winmo, can go 24hrs on 10-15% battery even with wifi (it sleeps real well in winmo i guess)
19:55.08phheval-: wifi is disactivated on sleep in winmo -_-'
19:55.14bzomakkonen: eh? My apache never spoke to me :/
19:55.25phhbzo: jealous ? :p
19:55.34eval-phh:  yea thats what i suspected
19:55.42bzophh: haha yes a little
19:55.57phheval-: for diam500, ok about the battery life
19:55.59phhbut rhod210 ?!?
19:56.10Markinusphh: I put you and captnoord in this project ..
19:56.17phhMarkinus: ok
19:56.27eval-phh:  yeah, i know.  it sucks.  i will add the pm symbols and see wtf is going on
19:56.52eval-phh:  also, i'm pretty sure my diam500 battery is shit.  with push gmail on winmo and some text/voice use i only get 8-12hrs
19:57.04eval-push really drains the fuck out of batteries it seems
19:57.48phheval-: it shouldn't
19:58.02phhon android, it only does 16hours -> 14hours for chamonix iirc
19:58.12makkonenI've been pretty off android since switching to the rhod400, since winmo with sense 2.5 is new and interesting, and android has no sound, which makes it sort of useless. no ringing... not great for a phone.
19:58.14phhon windows mobile... it's another story.
19:58.40phhmakkonen: you're allowed to try sound too ! :p
19:58.45leviathanGNUtoo: hmm
19:59.03leviathanonly push if it breaks nothing ;-)
19:59.19phhI'm starting to think wether we could work with qualcomm actually
19:59.33bzoI must confess I've been mostly on wm lately as well. The latest diam sense2.5 builds are very good
19:59.35phhI don't think anybody ever talked to them about us ... ?
19:59.47phhwell discussed I mean
19:59.47makkonenyeah. haven't dived in to it yet. I wasn't able to do anything with raph800/diam500 sound/mic stuff, so I'm scared.
19:59.59Cablekevin:O
20:00.29eval-phh:  herm.  ok, i will git reset, up the buf_log, add ramconsole and do some actual sim-less rhod210 tests +/- wifi, +/- gsync & report in a few days.
20:00.32phhmakkonen: yeah it takes a lot of time to used to :/
20:00.51eval-bzo:  whose do you use?  i am on OMJ's 21887 (i think?) v1.3.  needs a few fixes but very fast, pretty
20:01.01phheval-: simless, without wifi, you should get days of battery life
20:01.04bzophh: we need to put in a text to speech engine in makkonen's phone to force him to do our will
20:01.10eval-phh:  i tell you, 12-16hrs.
20:01.12phhbzo: .
20:01.23phheval-: ah you said that ?
20:01.28phheval-: well that's possible.
20:01.32GNUtooleviathan, all is machine arch
20:01.34phhI'd expect really more though
20:01.34GNUtooso no risk
20:01.39bzoeval-: omj also, don't remember the specifics the latest sense 2.5/wm 6.5.5
20:01.41phhI'll wait for captnoord's measures
20:01.43eval-phh:  but like i said, the led is flipping green/orange
20:01.52GNUtooalso xorg is default preference -1 but for shr
20:01.53phheval-: 16hours with flipping led ?
20:01.58leviathanGNUtoo: ok
20:02.05leviathanthen make it
20:02.07eval-bzo:  i'm using sense 2.1 and wm6.5 (how can you handle the x replacement??)  is 2.5 worth it?
20:02.13leviathanGNUtoo: make it so ;-)
20:02.19eval-phh:  ok, so the flipping led A LOT is much less than 12-16
20:02.25eval-phh:  the flipping led now and then gives me 12-16
20:02.32eval-phh:  i have yet to experience the green only glory
20:02.40phheval-: which rate ?
20:02.52phheval-: well, simless, there is no reason for it to change ...
20:03.01phhmy diamond can do >12hours continuous sleep simless
20:03.08eval-phh:  i know.  it's weird.
20:03.13eval-phh:  wait, we are talking rhod now
20:03.23eval-phh:  the diam500 i haven't touched since i gave chamonix those logs
20:03.24bzoeval-: eh, haven't tried a lot of builds really. Seems like with the latest 2.5 builds there isn't much of a speed penalty, so why not?
20:03.35phheval-: well, I haven't tried rhod simless.
20:03.44phheval-: but it's just to give the idea.
20:03.50eval-bzo: ok.  i was of the idea that 2.5 was a slowdown.  maybe i'll give it a shot.  no one seems to bugfix their 2.1 builds these days
20:05.01eval-phh: ok.  well, i will do more thorough tests and pipe kmsg, and we'll see!  maybe clean xdandroid & data.img (is the 3/31 worth trying?  or go straight to stinebd's aosp.. is stable?)
20:05.02bzoeval-: it used to be a lot slower, but the latest builds were made to run with our class of device
20:05.10*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
20:05.19phheval-: take stinebd's aosp :p
20:05.36phhI had no stability problem with it
20:05.38eval-phh:  your tongue scares me :p
20:05.42phhbut I used it for... 10 minutes ? :p
20:05.45phhmm no
20:05.47phhlot more
20:05.50phhwell I did my 12hours sleep with it
20:06.35eval-i might as well.  i like having source and it would make sense for myn or others to eventually mod some(eclair?)thing with a repo instead of these unknown mixed mods from N1 or whatnot
20:06.49eval-phh:  no JIT tho right?  that's not quite working?
20:07.02phheval-: stinebd's aosp uses jit
20:07.47leviathanphh: it was sooo cool
20:07.57phhleviathan: mm ?
20:07.59phhah right.
20:08.01leviathanthis scene where he asked the alien who always saved the earth
20:08.05bzodo the native android devices use the jit?
20:08.05phh:p
20:08.13phhleviathan: na, it's "Is this planet protected ?"
20:08.17leviathanyes
20:08.17phhbzo: afaik no
20:08.20eval-bzo:  not (yet) to my knowledge
20:08.20leviathan^^
20:08.24leviathan"run"
20:08.26leviathan:D
20:08.32phhleviathan: na it's way better
20:08.35phh"basically.... run"
20:08.36eval-bzo:  some google devs twittered its coming with some .32 builds?
20:08.54leviathanits soo coool
20:08.59leviathanI'm loving doctor who
20:09.01phhleviathan: you counted the numbers of geronimo ?
20:09.06leviathan:)
20:09.09leviathanonce
20:09.11leviathanokee
20:10.23phhgeronimo, and then, plouf in the swimingpool in the library.
20:10.25leviathanmickey|tv: back?
20:10.33leviathanphh: yeahh
20:10.37leviathanthat was cool ^^
20:11.10phhleviathan: the virus that always went into the clocks was fun too.
20:11.19leviathanyes =)
20:11.35Markinusphh: we should speak above the new structure too . . .what will we change -..
20:12.01phhMarkinus: true
20:12.08Markinusphh: I changed already the dex interface, so it'S in one place
20:12.20phhah ? ok let me checkout it first
20:12.27Markinusphh: ok
20:12.47leviathanphh: were you on fosdem?
20:12.59phhleviathan: no
20:13.02leviathanhmm
20:13.25Markinusphh: (btw you're admin too)
20:13.30phhMarkinus: even admin ?
20:13.31phhwow.
20:13.34leviathanlol
20:13.49leviathan"Oh, I'm admin? didnt know that" :)
20:13.49phhleviathan: well, i'm the only one who has admin access on my gito project :p
20:13.59leviathanahh, ok
20:14.32leviathanis mickey|tv back now?
20:14.48leviathanI'd like to have this firmware loader up and running
20:14.50leviathan... :)
20:15.04leviathanoh and GNUtoo your patch would be nice to be seen in OE too
20:15.21leviathanso that I can put it all together and publish a new image
20:15.29*** join/#htc-linux Wout (~a@s55927d21.adsl.wanadoo.nl)
20:15.31Markinusphh: I created a team . . and put it in this project
20:15.46GNUtooleviathan, very soon
20:15.52leviathanok
20:17.14GNUtooleviathan, I'll test my PR d->e and push
20:17.20GNUtoothat needs reparsing
20:17.28GNUtooor maybe I'll just push
20:17.29phhMarkinus: proc_comm_wince.c ?
20:17.34phhI guess cr2 isn't really happy with that
20:17.35GNUtooit's xf86-video-msm anyway
20:17.41GNUtooand no one apart us uses it
20:18.17Markinusphh: yes
20:18.27Markinusphh: no? why?
20:18.37phhMarkinus: I'd prefer one send_dex_call() function, and one fake proc_comm() function, that uses send_dex_call
20:18.45phhMarkinus: he hates that we call it proc_comm, since it's dex :p
20:18.50eval-phh:  is your diam100 super hot?  my 500 is much hotter in android than my TP2
20:18.56phheval-: oO
20:19.11phheval-: do you feel how cold a stone is ?
20:19.16eval-friends notice it too haha when i show it to them
20:19.17Markinusphh: ahh, ok, yeah . . maybe better
20:19.20phhthat's basically how hot my TD is.
20:19.22eval-phh:  :p  i hate you
20:19.26phheval-: I love you too
20:19.43eval-phh:  if that's true... wtf?  how is 500/100 so diff?
20:19.48phheval-: don't know
20:19.51phheval-: It's maybe just you.
20:20.05eval-phh: i'm just that hot its true
20:20.11phh:p
20:20.13eval-i should get a new battery and see
20:20.20eval-tho i really doubt...
20:20.28eval-anyway, people complain about this on xda-devs i'm not crazy
20:20.39bzoeval-: mine is warm too
20:21.01*** join/#htc-linux Vito_| (~Sysel@gw.loccal.net)
20:21.18makkonenhave you tried turning off the cdma radio? see if that cools it down completely?
20:21.21eval-bzo:  do you think it could be as simple as it's sleep restarting?
20:21.26eval-bzo:  you are diam500?
20:22.00bzoeval-: dunno, yes diam500
20:22.19eval-makkonen:  good question.  ill try airplane for a bit.
20:22.23eval-bzo:  how is your battery life
20:23.00bzomaybe 8 hours?
20:23.06GNUtooleviathan, done,also do that :
20:23.35eval-bzo:  ok, pretty similar.  6-8 here
20:24.14GNUtoohttp://pastebin.com/Mz3jT1q6
20:24.20GNUtoothat is not in oe
20:24.25GNUtoofeel free to submit to shr
20:24.30GNUtoobut tell it comes from me
20:24.45bzophh: could the screens be different on the diam 100/500? maybe the stuff captnoord is looking at is broken on the diam500 as well
20:25.47eval-bzo:  i think the touchscreen's are different... we're epson they're... toshiba?
20:25.55eval-bzo:  what is captnoord looking at that's broken
20:26.12bzopowering down stuff during suspend
20:27.14eval-bzo:  well our suspend is def. all messed up.  i 'feel' as if the screen is particularly an issue with the screen spam in dmesg and the fact i have to tap home several times sometimes to get it to light up again (do you have this issue to?)  my TP2 doesn't seem to
20:27.54bzoyeah, I see that too
20:28.51Markinusphh: the clockfine is in the moment a modified one from the old repo. But there a new one, this would be nice. Maybe like this 7x30 one ..
20:29.01leviathanGNUtoo: oh, no need for this patch
20:29.02Markinus*clockfile
20:29.14leviathanbecause I'm compiling binary-localedata
20:29.19leviathanworks fine ;-)
20:29.53GNUtooleviathan, ah?
20:30.20GNUtooleviathan, what's the difference between localdata and binary-localedata/
20:30.26GNUtoos#/#?#
20:31.29leviathanGNUtoo: for binary localedata you have precompiled binary's I guess
20:31.31leviathananyway
20:31.40leviathanyou need qemu to build them at all
20:31.56GNUtooyes
20:32.01leviathanbecause its buggy for qemu arm6
20:32.01GNUtoothat's what I do
20:32.03GNUtooI use qemu
20:32.07leviathanits disabled by default
20:32.07phhMarkinus: 7x30's is a monster
20:32.09GNUtooand it seem to work
20:32.11GNUtoofor me
20:32.22phhMarkinus: clock-wince is definitely not good
20:32.26phh7x30 is definitely not good either
20:32.26GNUtoonote the seem
20:32.27Markinusphh: yeah . .
20:32.34Markinusphh: hmm
20:32.37GNUtoomaybe it fails somewhere I didn't saw
20:32.44Markinusphh: aso a completly new one
20:32.58GNUtooor maybe it was disabled(I remember that) and now fixed(maybe I missed that)
20:33.02phhMarkinus: we could rely on rpc :p
20:33.39leviathanGNUtoo: You need ENABLE_BINARY_LOCALE_GENERATION_armv6-novfp = "1"
20:33.55leviathanin your local.conf
20:34.01GNUtooI've ENABLE_BINARY_LOCALE_GENERATION
20:34.04GNUtooalways
20:34.09leviathanENABLE_BINARY_LOCALE_GENERATION_armv6-novfp
20:34.12leviathan;)
20:34.16leviathanvery important
20:34.24leviathanexplicitly for this cpu
20:34.25leviathan:)
20:34.26Markinusphh: hmm, you mean 3000000f?
20:34.31phhMarkinus: yes
20:34.46*** join/#htc-linux Bry8Star_ (~Bry8Star@cpe-76-170-247-232.socal.res.rr.com)
20:34.49Markinusphh: but then we have to find out what the enums are . .
20:35.09GNUtooleviathan, ah no not in this local.conf
20:35.15phhMarkinus: set clock, compare before/after ?
20:35.15Markinusphh: but right . . this would be the cleanest way
20:35.24Markinusphh: hmm, right! :)
20:35.33phhMarkinus: we would rely on a "black box", which doesn't look right
20:35.48phhbut seriously, I've worked on clock-wince, I don't want such thing again
20:36.05Markinusphh: you're right
20:36.19GNUtoo-rw-r--r--. 1 embedded embedded    42470 2010-04-02 16:16 eglibc-locale-zh-tw_2.11-r9.4.4_armv6-novfp.ipk
20:36.46Markinusphh: is this interface on dia/blach etc. too?
20:36.51Markinus*black
20:36.57phhMarkinus: yes
20:37.03Markinusgood
20:37.23phhMarkinus: I changed HEAD to the actual branch instead of master btw
20:37.32eval-makkonen:  yeah, even neocore isn't heating me much with radio off
20:37.55Markinusphh: ok
20:37.57makkonenso... that's good. maybe.
20:39.04makkonenWhen I'm on the phone with the raph800 in winmo, or downloading over evdo, it gets uncomfortably hot. I think it's just a high drain radio. for some reason, android isn't turning it off enough, I guess.
20:39.14makkonenI don't know how to dig down into that.
20:41.10phhMarkinus: haha.
20:41.12phhbig problem.
20:41.24phhor not ?
20:41.37phhah no
20:41.38phhnothing.
20:41.44phhit seemed to me qdsp5 was COMP only
20:41.55Markinusphh: ahh, ok, no, is for both
20:42.53Markinusphh: I think we can take this amss detection and switch function, but we should put it direct in the mach-msm dir . .
20:43.08phhyup
20:43.19phhMarkinus: I don't know how amss detection works on !wince devices
20:44.02Markinusphh: do you mean there is one?
20:44.16Markinusphh: not complatly static for this device?
20:44.32phhMarkinus: I mean that I don't know if the field we use to detect AMSS works for android devices
20:45.04Markinusphh: hehe I didn't see tha ! :) sry
20:45.12Markinusright
20:45.35phhMarkinus: haha
20:45.39phhI totally forgot pmic.c
20:45.42phhit has some stuff.
20:45.54Markinusphh: yes . ..
20:46.10GNUtooleviathan, also before booting
20:46.22leviathanyes?
20:46.23phhlike lp_mode
20:46.26phhthat I still don't know what it is.
20:47.02Markinusphh: and take a look to nexus microp file .. . . this is a monster . .
20:47.18phhMarkinus: how do you think I got your bma150 microp driver ?
20:47.25phhit comes from there :p
20:47.25Markinusphh: hehe
20:47.28Markinus:)
20:47.30phhand yes it's a big monster.
20:47.45*** join/#htc-linux julcel (~julcel@2a01:e35:2f3d:a720:224:23ff:fe03:c92f)
20:47.47eval-makkonen:  interesting.  never noticed the heat in winmo.  yea i wonder.  ok well at least that's a place to look
20:48.00GNUtooleviathan, ah no the login issue is not present on shr-image
20:48.11leviathanah, ok
20:48.16GNUtooleviathan, was present only on mickey's fso2 demo image
20:48.22GNUtoobasically pam was missing
20:48.23phhMarkinus: board-htcleo-pm.c ?
20:48.24phhseriously ?
20:48.25GNUtooso login failled
20:48.28GNUtooso rm /bin/login
20:48.33GNUtoocd /bin
20:48.35Markinusphh: it's the om interface
20:48.38GNUtooln -s busybox login
20:48.38Markinus*pm
20:48.40GNUtooand it worked
20:48.43leviathanhmm
20:48.44leviathanstrange
20:48.46Markinus30000060
20:48.51phhMarkinus: I know, but it's not going to be leo specific
20:48.58leviathanperhaps I should clean up the tmp-folder
20:49.05leviathanand let it rebuild over night
20:49.06Markinusphh: hmm, yes . . . right ups
20:49.22phhMarkinus: same goes for mmc
20:49.29phhin spite of what qualcomm guys do.
20:49.55leviathanGNUtoo: what do you mean?
20:50.05Markinusphh: there we can meybe take this one from our repo . ..  mmc?
20:50.13leviathanGNUtoo: should I let it build from scratch over night?
20:50.25Markinusor rename this
20:50.26Markinusright
20:50.29phhMarkinus: rename.
20:50.32GNUtooyes
20:50.34phhMarkinus: yours mmc should be way different from ours
20:50.35phhI guess
20:50.52GNUtooleviathan, was only for fso2-demo-image with only zhone as framebuffer
20:51.07leviathanoh, ok
20:52.22Markinusphh: hmm, I don'T think it'S very different . .. same GPIO nomber, config, etc.
20:52.30phhMarkinus: oh really ?
20:52.37phhok then
20:53.33leviathanGNUtoo: so, then I'll try the new image :)
20:53.44leviathannext issue is the smoother loading of the firmware
20:53.49leviathanthat should be done by mickey|tv
20:54.07GNUtooleviathan, did you try with udev 151 and correct filenames
20:54.14GNUtooalso with udev 141 and correct filenames
20:54.15GNUtoo?
20:54.15leviathanuhm
20:54.26Markinusphh: Leo has like nexus a GPIO for wlan power
20:54.27leviathanI'll try it with 141 now
20:54.38leviathanbecause new image has 141 preinstalled
20:54.39leviathan:)
20:55.09phhMarkinus: wifi stuff can be merged in mmc.c as I did in 2.6.27 I guess
20:55.15phhor possibly another file
20:55.31Markinusphh: this stuff isn'T in:
20:55.31Markinusint trout_wifi_power(int on)
20:56.04Markinusphh: o ne, is there
20:56.11Markinushtcleo_wifi_power
20:56.14Markinusoverseen it
20:56.57*** join/#htc-linux hfctsaot (~fircuser@173-137-7-95.pools.spcsdns.net)
20:58.18phhMarkinus: codeaurora's tree is such a mess
20:59.01Markinusphh: hehe, and so much different trees
20:59.24phhI wish I'd work for them.
21:01.14hfctsaotphh u ever figure
21:01.20phhhfctsaot: ?
21:02.39phhMarkinus: pff I spent so much time translation pm_* calls and it's there -_-'
21:03.06Markinusphh: I din'T thad didn'T know that . .:(
21:03.13phhuh ?
21:03.17Markinusargh :)
21:03.19phhI've only one 000061 cal actually
21:03.26phhcall*
21:03.30*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
21:03.34phhah no
21:03.37MarkinusI didn'T know that you didn't knoiw that . . I know this file . .
21:03.39phhjust weird documentation format
21:03.42phhMarkinus: lol
21:03.46phhMarkinus: I've already been in this file
21:03.53phhI just totally forgot
21:04.07Markinusphh: ahh ok
21:04.19GNUtoomickey|tv, wifi is not bumped in oe?
21:04.29phhpm_secure_mpp_config_digital_output(0x06)(enum, enum, enum, 0x24, 0x58, 0x18); //vreg id, level, control:

21:04.29phhvs
21:04.36phh#define SECURE_MPP_CONFIG_DIGITAL_OUTPUT_PROC 5

21:04.43phhint pmic_secure_mpp_control_digital_output(enum mpp_which which,
        enum mpp_dlogic_level level,
        enum mpp_dlogic_out_ctrl out)
21:04.44phhat least
21:04.46phhI was correct.
21:05.01Markinusyeah :)
21:05.28phhMarkinus: I still can't see what lp_mode does :'(
21:05.31phhjust on/off ?
21:05.37Markinusphh :I decoded some function on Leo , ,here
21:05.38Markinushttp://www.htc-linux.org/wiki/index.php?title=Leo_RPC
21:05.38Markinusmaybe you need somethink
21:05.42ToAsTcfhphh: sorry im home now
21:05.47ToAsTcfhany luck
21:06.20ToAsTcfheh guess not :(
21:06.30Markinusphh: no on/off . . this is 30000060 . . no 61
21:06.44phhlow power on off ? hum.
21:06.55Markinusphh: hmm, maybe
21:07.00phhsounds stupid.
21:07.05Markinusha :)
21:07.05phhwhat means "vreg low power" ?
21:07.11Markinushehehe
21:07.12phhit's either on or off -_-'
21:08.08Markinusthis is pm_mode_controll . .hmm maybe we can configure there some other stuff.  ..  its a cmd enum
21:08.16phhMarkinus: btw
21:08.20phhstruct usb_pwr_sel_switch_args {
        uint32_t cmd;
        uint32_t switch_id;
        uint32_t app_mask;
};
21:08.24phhrpc_pmapp.c
21:08.27phhin some codeaurora's tree.
21:08.35Markinusahhhh
21:08.56phhbranch is eclair
21:09.03phhok that must be android.git.kernel.org
21:09.18ToAsTcfhphh: that dude ever figure out his uid issues?
21:09.24phhToAsTcfh: no
21:09.30ToAsTcfh:/
21:09.34phhToAsTcfh: it'd be a security flaw anyway
21:09.48phhI don't think google guys would have keep an easy one
21:10.10ToAsTcfhwhat esigning?
21:10.20ToAsTcfhresigning
21:10.33phhna, being able to access RIL
21:10.59ToAsTcfhoh yeah onl if ur root i believe
21:11.08ToAsTcfhbut not from an app
21:12.05ToAsTcfhi didnt really understand what h was tryin to do really
21:12.28phhToAsTcfh: accesss to ril -_-'
21:12.40ToAsTcfhbut why
21:12.40Markinusphh: I found it, nice file ..
21:13.01phhToAsTcfh: i'm not sure
21:13.06ToAsTcfhlol
21:13.14bzophh: is it normal that the diam500 is using the htcraphael_mddi_power_client?
21:13.25phhbzo: bwah why not ?
21:13.43bzobecause there is a htcdiamond__mddi_power_client
21:13.44ToAsTcfhbzo pll2/2 never worked
21:13.56phhbzo: "diamond" here means diam100
21:14.02ToAsTcfhsame  issue
21:14.46bzophh: ah, so maybe the fixes captnoord is making in the raph panel stuff will affect the diam500
21:14.53phhbzo: maybe
21:15.02bzoToAsTcfh: yep, seems the same as before
21:15.21phhMarkinus: how can they be so messy ...
21:15.30bzophh: because the main differences seem to be the micropklt_lcd_control stuff in the diam panel
21:15.32phhit's like they have three developers on each feature, working on its own tree
21:16.21ToAsTcfhi wonder if itll overclock rght before i hit it with a large rock
21:17.36*** join/#htc-linux Kawaiius (~yadayada@pool-71-255-183-81.slsbmd.east.verizon.net)
21:17.52phhMarkinus: hum interesting stuff in htc_battery.c  "*  CHARGER_EN (CPLD MISC2 bit[0]) is move to PMIC (MPP_14)."
21:18.03phhmeans wome things are accessible only through rpc, not with gpio
21:18.14phhI thought every control could be done through gpios
21:22.18Markinushmm
21:23.38ToAsTcfhbzo: so its not the divider or the speed so all thats left is conflict in the kernel somewhere i guess
21:25.21phhbzo: did you try removing all the code but changing the pll's speed ?
21:25.40bzoToAsTcfh: it seems that the register to control the clock stuff does not work in native android devices
21:25.41*** join/#htc-linux sxe (~quassel@ip-88-153-21-39.unitymediagroup.de)
21:25.51bzophh: I had ToAsTcfh comment out the multiplier change, so yes
21:26.19bzophh: I didn't have him comment out the acpuclock table adjustments though
21:26.40*** join/#htc-linux seg (~seg@cpc2-leds3-0-0-cust928.leed.cable.ntl.com)
21:26.45phhbzo: maybe try removing everything but the pll2 change ?
21:26.54phhbzo: btw, I rethought about AXI concerns
21:27.09phhbzo: "turbo hw" has a different pll1
21:27.14phhso it's likely that AXI is based on pll1
21:27.16phhnot 2
21:27.38bzoseems logical
21:27.49phhstill we don't know how it's calculated, right.
21:28.07bzoright
21:28.18ToAsTcfhphh: bzo: current acpuclock.c http://pastebin.com/j58c29Mc
21:29.30bzoToAsTcfh: if you want to try phh's suggestion, comment out lines 526-547
21:32.08Markinusphh: the acpuclock-arm11.c seems to be the old one. .  without modifcations
21:32.16phhMarkinus: yup
21:32.22phhwe'll have to backport ours there
21:32.32bzophh: hmm perhaps that would explain why the axi register is the same in turbo/non-turbo devices. They rely purely on the higher clock and not any other different config
21:33.14phhbzo: with fixed /6 divider ?
21:33.19phhI can't believe so
21:34.08ToAsTcfharch/arm/mach-msm/acpuclock.c: In function 'acpuclk_init':
21:34.10ToAsTcfharch/arm/mach-msm/acpuclock.c:485: warning: unused variable 'reg_clkctl'
21:34.11ToAsTcfharch/arm/mach-msm/acpuclock.c:484: warning: unused variable 'a11clk_khz_new'
21:34.16ToAsTcfhjust for fyi
21:34.26bzophh: 960/768*133 = 166
21:34.37ToAsTcfhhero is turbo
21:34.44phhToAsTcfh: don't care :p
21:34.56phhbzo: I know, that's why I said "fixed /6"
21:35.00ToAsTcfh:/
21:35.09bzophh: seems like too much of a coincidence
21:35.21phhbzo: btw, on android devices, they set frequency to 0 when they want low speed
21:36.07bzophh: have we ever tried setting pll1 of a non-turbo device to turbo speed?
21:36.14phhNOPE
21:36.15phhoops
21:36.31bzoI guess that will be an interesting experiment for me to try :)
21:36.31phhbzo: is it worth a try ?
21:36.32phhdefinitely.
21:38.35ToAsTcfhok booting wish me luck :P
21:38.58phhgood crash
21:39.16ToAsTcfhi can only hope for a good crash
21:39.30ToAsTcfhall my crashes suck with no errors
21:41.47*** part/#htc-linux Orkspalter (~Orkspalte@p54B58CB9.dip0.t-ipconnect.de)
21:42.09phhToAsTcfh: and this one ?
21:42.17ToAsTcfhlol i echo 600 and got 528
21:42.31phhyes it's normal
21:42.36ToAsTcfhlol
21:42.44phhif 528 works, it means you're @ 600.
21:42.49phhtheorically
21:42.56ToAsTcfh:/
21:43.06ToAsTcfhok
21:43.15phhif bzo gives you correct indications
21:43.20phhand if you applied them
21:43.28phhToAsTcfh: now try a benchmark app
21:45.30ToAsTcfhhttp://pastebin.com/u2qWyUCQ
21:49.08bzoToAsTcfh: save yourself some work next time. Comment using /* at start of line 526, and */ at end of line 547
21:49.43ToAsTcfhlol sorry
21:50.06ToAsTcfhso pi gave me 13276
21:50.11bzonothing to be sorry about, just trying to pass on some knowledge
21:51.08GNUtoomickey|tv, leviathan firmware loading works with udev
21:51.13GNUtoono patch needed
21:51.19GNUtoomaybe luck don't know
21:51.20mickey|tvcool
21:51.24GNUtooand good filenames
21:51.30GNUtooand also good calibration data
21:51.31mickey|tvstill i almost finished my uploader
21:51.33GNUtoobye
21:51.36mickey|tvcu
21:51.43mickey|tvmy dream is udev free zone
21:51.46mickey|tvand will be
21:51.53GNUtoothat's why I repeated to leviathan to test it
21:51.56GNUtoohe didn't
21:51.58GNUtooso I did it
21:52.00GNUtoobye
21:52.05mickey|tvright
21:52.07mickey|tvg'night, cu
21:52.20stinebdi wonder who's running a mirror of xdandroid.southcape.org
21:53.54ToAsTcfhphh: im not sure i can say this is faster in pi
21:53.59*** join/#htc-linux kvaster (~kvaster@live.bn.by)
21:54.06mickey|tvGNUtoo: btw., backlight fix has been pushed
21:54.14GNUtoook nice
21:54.32mickey|tvnow that wifi works, what do we need to continue on?
21:54.37mickey|tvbluetooth can wait
21:54.40mickey|tvsuspend/resume i think
21:54.42mickey|tvwhat do you think?
21:54.43stinebdjamenlang: ah it's you with the mirror ;)
21:57.02*** join/#htc-linux julcel (~julcel@2a01:e35:2f3d:a720:224:23ff:fe03:c92f)
21:57.12stinebdor perhaps live bookmarks
21:58.51ToAsTcfhbzo it is slower for sure but it also keeps selecting 480 and then selecting 528 constantly
22:01.54ToAsTcfhbzo also i just moved the min down to 245000 and yeah kmsg is switching back and forth between 245 and 528 like rapidly
22:02.04leviathanmickey|tv: so GNUtoo says wifi works?
22:02.09ToAsTcfha few times a second
22:02.13leviathanmickey|tv: anyway graphics works now
22:02.19mickey|tvleviathan: wifi works like a charm now
22:02.25mickey|tvgraphics as well? splendid
22:02.28leviathangtk issue has dissapeared
22:02.30mickey|tvnow we need suspend/resume, if i may say
22:02.42mickey|tvlast i found was that it immediately resumes
22:02.47leviathanmickey|tv: how do you get wifi running like charme?
22:02.48mickey|tvno matter what processes i had run
22:02.51leviathandid I miss something?
22:03.02mickey|tvno, just all the work you guys previously did
22:03.12mickey|tvcompiling stuff into kernel
22:03.18mickey|tvloading the proper modules
22:03.21mickey|tvifconfig up
22:03.25mickey|tvloading firmware
22:03.31leviathanoh
22:03.31mickey|tvah, and rfkill before ifconfig
22:03.34leviathanok
22:03.46leviathanbut we need automatation of this process
22:03.48leviathan:)
22:03.49mickey|tvsure
22:03.54mickey|tvfirmwareloader compiles now
22:04.15mickey|tvthe ugly thing is, we will not be able to distribute images
22:04.25mickey|tvas we have no rights for the firmware redistribution
22:04.38leviathanhmm
22:04.45mickey|tvso people will have to put them somewhere manually
22:04.56leviathanmickey|tv: you can download it from ftp.o2s.ch? ;-)
22:05.06mickey|tvi can
22:05.06leviathanyeah
22:05.13leviathanother possibility
22:05.14mickey|tvbut i dare not to put those things online
22:05.20mickey|tvwith my signature :)
22:05.24mickey|tv[things = images]
22:05.24leviathanyes :)
22:06.04mickey|tvoh wait
22:06.08mickey|tvwe could grab them
22:06.12mickey|tvfrom the filesystem
22:06.16mickey|tvcan we read from nand?
22:08.06leviathanuhm
22:08.09leviathanof course
22:08.18leviathanwe even can build rootimages for nand :)
22:10.18leviathanok, I need to sleep now
22:10.26leviathanbut it sounds like a plan :)
22:10.57leviathangn8
22:11.26leviathanbtw: Ventura is a nice app ;-D
22:11.53leviathanthis+UMTS=awsome
22:12.14leviathanonly question: can it play flash, or will I need the mobile streaming version
22:12.15leviathanbut anway
22:12.19leviathanI need sleep now
22:12.21leviathangn8
22:15.01mickey|tvgnu8
22:15.01mickey|tvcu
22:17.33mickey|tvi guess now that it works, there's no benefit in leaving it to be a module, right?
22:17.39mickey|tv(meaning, we can compile it in the kernel)
22:22.37ToAsTcfhphh: bzo: necore sux bad too
22:23.08ToAsTcfhi know something is wrong here
22:23.46ToAsTcfhits so bad smoke looks like its out of a comic strip
22:25.02ToAsTcfh<4>[ 2573.895935] msm_cpufreq_target 528000 r 1 (245000-528000) selected 528000
22:25.04ToAsTcfh<7>[ 2581.569000] tty_buffer_alloc: cb4a0800
22:25.06ToAsTcfh<7>[ 2581.574371] tty_buffer_free: c1fde800
22:28.08*** join/#htc-linux Cable[SQL] (~Cablekevi@house.speedxs.nl)
22:28.12ToAsTcfh2.6fps omg
22:32.47eval-wow my diam500 is hotter trying to 'sleep' than when i was watching neocore (both airplane mode)
22:35.36ToAsTcfhbzo it seems like its not using the max like it selects it but then goes beack to the min for some reason
22:40.21*** join/#htc-linux bombenbach (~bombenbac@77.47.80.54.dynamic.cablesurf.de)
22:41.33*** join/#htc-linux marex (~marex@vasut.kolej.mff.cuni.cz)
22:42.04makkoneneval-: that's really weird.
22:42.12bombenbachIt seems that bcm4329 has just got updated -> RC209 http://tinyurl.com/ydfe462
22:42.15*** join/#htc-linux antoszka (~antoszka@unaffiliated/antoszka)
22:42.41bombenbachneed to try that before I begin messing with bcm4325 aka b43
22:44.31eval-why are there a slew of people with 0 feedback selling 250-300 nexus on ebay
22:45.36eval-oh: Item location:北京, 北京, China ... probably cheap knockoff?
23:10.01*** join/#htc-linux Wout (~a@s55927d21.adsl.wanadoo.nl)
23:12.54*** join/#htc-linux YellowGTO|Nix (~quassel@c-71-226-239-246.hsd1.pa.comcast.net)
23:13.21*** join/#htc-linux DCDemon (~PSPdemon@c-76-123-109-83.hsd1.fl.comcast.net)
23:26.36*** join/#htc-linux PSPdemon (~PSPdemon@c-76-123-109-83.hsd1.fl.comcast.net)
23:35.28*** join/#htc-linux DCDemon (~PSPdemon@c-76-123-109-83.hsd1.fl.comcast.net)
23:39.31*** join/#htc-linux PSPdemon (~PSPdemon@c-76-123-109-83.hsd1.fl.comcast.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.